{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.999969938373666,
  "eval_steps": 500,
  "global_step": 16632,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "auxiliary_loss_clip": 0.05028445,
      "auxiliary_loss_mlp": 0.02215396,
      "balance_loss_clip": 2.43573999,
      "balance_loss_mlp": 1.76983953,
      "epoch": 6.012325266796934e-05,
      "flos": 24456507091200.0,
      "grad_norm": 55.0007689153118,
      "language_loss": 2.85272503,
      "learning_rate": 0.0,
      "loss": 1.94613922,
      "num_input_tokens_seen": 19155,
      "step": 1,
      "time_per_iteration": 14.6953706741333
    },
    {
      "auxiliary_loss_clip": 0.03379929,
      "auxiliary_loss_mlp": 0.01460768,
      "balance_loss_clip": 1.62770474,
      "balance_loss_mlp": 1.19068694,
      "epoch": 0.00012024650533593868,
      "flos": 20225931246720.0,
      "grad_norm": 36.21369939567931,
      "language_loss": 1.83011317,
      "learning_rate": 4.4628432569317594e-07,
      "loss": 1.87852001,
      "num_input_tokens_seen": 36175,
      "step": 2,
      "time_per_iteration": 2.4755165576934814
    },
    {
      "auxiliary_loss_clip": 0.03319962,
      "auxiliary_loss_mlp": 0.01441938,
      "balance_loss_clip": 1.62580693,
      "balance_loss_mlp": 1.18959546,
      "epoch": 0.000180369758003908,
      "flos": 22309935454080.0,
      "grad_norm": 32.71534780746326,
      "language_loss": 1.57499588,
      "learning_rate": 7.073439208833112e-07,
      "loss": 1.62261486,
      "num_input_tokens_seen": 54870,
      "step": 3,
      "time_per_iteration": 2.429921865463257
    },
    {
      "auxiliary_loss_clip": 0.03362487,
      "auxiliary_loss_mlp": 0.01452239,
      "balance_loss_clip": 1.62410331,
      "balance_loss_mlp": 1.15640903,
      "epoch": 0.00024049301067187735,
      "flos": 22414650577920.0,
      "grad_norm": 51.139974107101715,
      "language_loss": 1.67653668,
      "learning_rate": 8.925686513863519e-07,
      "loss": 1.724684,
      "num_input_tokens_seen": 74575,
      "step": 4,
      "time_per_iteration": 2.5677058696746826
    },
    {
      "auxiliary_loss_clip": 0.03402039,
      "auxiliary_loss_mlp": 0.01504593,
      "balance_loss_clip": 1.6250596,
      "balance_loss_mlp": 1.21677387,
      "epoch": 0.0003006162633398467,
      "flos": 21396978449280.0,
      "grad_norm": 56.19314693935947,
      "language_loss": 1.91152883,
      "learning_rate": 1.0362401141348472e-06,
      "loss": 1.96059501,
      "num_input_tokens_seen": 92580,
      "step": 5,
      "time_per_iteration": 2.7236881256103516
    },
    {
      "auxiliary_loss_clip": 0.03371567,
      "auxiliary_loss_mlp": 0.01516591,
      "balance_loss_clip": 1.61563373,
      "balance_loss_mlp": 1.22171497,
      "epoch": 0.000360739516007816,
      "flos": 21652375127040.0,
      "grad_norm": 33.79752291712156,
      "language_loss": 1.61086583,
      "learning_rate": 1.153628246576487e-06,
      "loss": 1.65974748,
      "num_input_tokens_seen": 109705,
      "step": 6,
      "time_per_iteration": 2.7299628257751465
    },
    {
      "auxiliary_loss_clip": 0.03354501,
      "auxiliary_loss_mlp": 0.01487324,
      "balance_loss_clip": 1.61587119,
      "balance_loss_mlp": 1.20389163,
      "epoch": 0.0004208627686757854,
      "flos": 27159742897920.0,
      "grad_norm": 25.002730852649435,
      "language_loss": 1.53405786,
      "learning_rate": 1.2528784983718962e-06,
      "loss": 1.58247614,
      "num_input_tokens_seen": 129425,
      "step": 7,
      "time_per_iteration": 2.7463953495025635
    },
    {
      "auxiliary_loss_clip": 0.03321391,
      "auxiliary_loss_mlp": 0.01442803,
      "balance_loss_clip": 1.61226654,
      "balance_loss_mlp": 1.16528416,
      "epoch": 0.0004809860213437547,
      "flos": 31319096135040.0,
      "grad_norm": 31.607545338794147,
      "language_loss": 1.43749309,
      "learning_rate": 1.338852977079528e-06,
      "loss": 1.48513496,
      "num_input_tokens_seen": 149210,
      "step": 8,
      "time_per_iteration": 2.7866902351379395
    },
    {
      "auxiliary_loss_clip": 0.03369369,
      "auxiliary_loss_mlp": 0.01498114,
      "balance_loss_clip": 1.61160421,
      "balance_loss_mlp": 1.21372795,
      "epoch": 0.000541109274011724,
      "flos": 32160411463680.0,
      "grad_norm": 71.01780320086927,
      "language_loss": 1.49897873,
      "learning_rate": 1.4146878417666224e-06,
      "loss": 1.54765356,
      "num_input_tokens_seen": 169055,
      "step": 9,
      "time_per_iteration": 2.7751379013061523
    },
    {
      "auxiliary_loss_clip": 0.0330958,
      "auxiliary_loss_mlp": 0.01474859,
      "balance_loss_clip": 1.6153338,
      "balance_loss_mlp": 1.20592284,
      "epoch": 0.0006012325266796934,
      "flos": 18916808163840.0,
      "grad_norm": 23.230106739376225,
      "language_loss": 1.44756222,
      "learning_rate": 1.4825244398280232e-06,
      "loss": 1.49540663,
      "num_input_tokens_seen": 188045,
      "step": 10,
      "time_per_iteration": 2.68770432472229
    },
    {
      "auxiliary_loss_clip": 0.03366141,
      "auxiliary_loss_mlp": 0.01493413,
      "balance_loss_clip": 1.6211977,
      "balance_loss_mlp": 1.21951771,
      "epoch": 0.0006613557793476627,
      "flos": 20774861867520.0,
      "grad_norm": 18.65159540626604,
      "language_loss": 1.45394874,
      "learning_rate": 1.5438901072051983e-06,
      "loss": 1.5025444,
      "num_input_tokens_seen": 207035,
      "step": 11,
      "time_per_iteration": 2.667656421661377
    },
    {
      "auxiliary_loss_clip": 0.03294416,
      "auxiliary_loss_mlp": 0.01451949,
      "balance_loss_clip": 1.60798573,
      "balance_loss_mlp": 1.17557406,
      "epoch": 0.000721479032015632,
      "flos": 16581680997120.0,
      "grad_norm": 18.695296112122374,
      "language_loss": 1.45259953,
      "learning_rate": 1.5999125722696629e-06,
      "loss": 1.50006318,
      "num_input_tokens_seen": 223225,
      "step": 12,
      "time_per_iteration": 2.7024025917053223
    },
    {
      "auxiliary_loss_clip": 0.03336224,
      "auxiliary_loss_mlp": 0.01408365,
      "balance_loss_clip": 1.61989343,
      "balance_loss_mlp": 1.14934683,
      "epoch": 0.0007816022846836014,
      "flos": 23805471144960.0,
      "grad_norm": 14.075216949325474,
      "language_loss": 1.29353356,
      "learning_rate": 1.6514482443788434e-06,
      "loss": 1.34097958,
      "num_input_tokens_seen": 242570,
      "step": 13,
      "time_per_iteration": 2.7084388732910156
    },
    {
      "auxiliary_loss_clip": 0.03290794,
      "auxiliary_loss_mlp": 0.01471898,
      "balance_loss_clip": 1.61285377,
      "balance_loss_mlp": 1.20334339,
      "epoch": 0.0008417255373515708,
      "flos": 19172204841600.0,
      "grad_norm": 5.850888394264952,
      "language_loss": 1.2096591,
      "learning_rate": 1.6991628240650723e-06,
      "loss": 1.25728607,
      "num_input_tokens_seen": 261215,
      "step": 14,
      "time_per_iteration": 2.667006015777588
    },
    {
      "auxiliary_loss_clip": 0.03275878,
      "auxiliary_loss_mlp": 0.0143239,
      "balance_loss_clip": 1.6181798,
      "balance_loss_mlp": 1.16898489,
      "epoch": 0.00090184879001954,
      "flos": 26395564026240.0,
      "grad_norm": 14.910512461050104,
      "language_loss": 1.12998271,
      "learning_rate": 1.7435840350181584e-06,
      "loss": 1.17706537,
      "num_input_tokens_seen": 280035,
      "step": 15,
      "time_per_iteration": 4.154406309127808
    },
    {
      "auxiliary_loss_clip": 0.03241476,
      "auxiliary_loss_mlp": 0.01412075,
      "balance_loss_clip": 1.60297179,
      "balance_loss_mlp": 1.16240263,
      "epoch": 0.0009619720426875094,
      "flos": 24679500785280.0,
      "grad_norm": 4.507189979377554,
      "language_loss": 1.1118933,
      "learning_rate": 1.7851373027727038e-06,
      "loss": 1.15842879,
      "num_input_tokens_seen": 300265,
      "step": 16,
      "time_per_iteration": 4.129991769790649
    },
    {
      "auxiliary_loss_clip": 0.03228652,
      "auxiliary_loss_mlp": 0.01418673,
      "balance_loss_clip": 1.60950506,
      "balance_loss_mlp": 1.17834735,
      "epoch": 0.0010220952953554788,
      "flos": 18624531196800.0,
      "grad_norm": 4.703126954084794,
      "language_loss": 1.12791896,
      "learning_rate": 1.8241705979033208e-06,
      "loss": 1.17439222,
      "num_input_tokens_seen": 317375,
      "step": 17,
      "time_per_iteration": 2.643577814102173
    },
    {
      "auxiliary_loss_clip": 0.0316523,
      "auxiliary_loss_mlp": 0.01381873,
      "balance_loss_clip": 1.60745263,
      "balance_loss_mlp": 1.1499393,
      "epoch": 0.001082218548023448,
      "flos": 26142537646080.0,
      "grad_norm": 4.210276973546229,
      "language_loss": 1.08280706,
      "learning_rate": 1.860972167459798e-06,
      "loss": 1.12827802,
      "num_input_tokens_seen": 337975,
      "step": 18,
      "time_per_iteration": 2.686873435974121
    },
    {
      "auxiliary_loss_clip": 0.03191023,
      "auxiliary_loss_mlp": 0.01399878,
      "balance_loss_clip": 1.60620987,
      "balance_loss_mlp": 1.13456583,
      "epoch": 0.0011423418006914173,
      "flos": 19609776322560.0,
      "grad_norm": 5.6844132181873706,
      "language_loss": 1.02718735,
      "learning_rate": 1.89578346593066e-06,
      "loss": 1.07309651,
      "num_input_tokens_seen": 356635,
      "step": 19,
      "time_per_iteration": 2.646482229232788
    },
    {
      "auxiliary_loss_clip": 0.0313424,
      "auxiliary_loss_mlp": 0.01342309,
      "balance_loss_clip": 1.60809088,
      "balance_loss_mlp": 1.12172449,
      "epoch": 0.0012024650533593868,
      "flos": 17895365107200.0,
      "grad_norm": 3.8666625595729234,
      "language_loss": 1.16783154,
      "learning_rate": 1.928808765521199e-06,
      "loss": 1.21259713,
      "num_input_tokens_seen": 375625,
      "step": 20,
      "time_per_iteration": 2.7233519554138184
    },
    {
      "auxiliary_loss_clip": 0.03122459,
      "auxiliary_loss_mlp": 0.01379383,
      "balance_loss_clip": 1.58902466,
      "balance_loss_mlp": 1.12990153,
      "epoch": 0.001262588306027356,
      "flos": 21252043071360.0,
      "grad_norm": 4.496949259977923,
      "language_loss": 1.06348491,
      "learning_rate": 1.9602224192552076e-06,
      "loss": 1.10850334,
      "num_input_tokens_seen": 394350,
      "step": 21,
      "time_per_iteration": 2.6852407455444336
    },
    {
      "auxiliary_loss_clip": 0.03027258,
      "auxiliary_loss_mlp": 0.01384844,
      "balance_loss_clip": 1.57162118,
      "balance_loss_mlp": 1.14928687,
      "epoch": 0.0013227115586953253,
      "flos": 26104077158400.0,
      "grad_norm": 4.063335269739779,
      "language_loss": 1.06536341,
      "learning_rate": 1.9901744328983746e-06,
      "loss": 1.10948443,
      "num_input_tokens_seen": 413255,
      "step": 22,
      "time_per_iteration": 2.6957693099975586
    },
    {
      "auxiliary_loss_clip": 0.02976226,
      "auxiliary_loss_mlp": 0.01337113,
      "balance_loss_clip": 1.57302022,
      "balance_loss_mlp": 1.12549293,
      "epoch": 0.0013828348113632948,
      "flos": 23951376190080.0,
      "grad_norm": 4.0043070186919945,
      "language_loss": 0.91960108,
      "learning_rate": 2.018794797290208e-06,
      "loss": 0.96273446,
      "num_input_tokens_seen": 433065,
      "step": 23,
      "time_per_iteration": 2.7151856422424316
    },
    {
      "auxiliary_loss_clip": 0.02942104,
      "auxiliary_loss_mlp": 0.01363138,
      "balance_loss_clip": 1.56537771,
      "balance_loss_mlp": 1.14169502,
      "epoch": 0.001442958064031264,
      "flos": 15959851724160.0,
      "grad_norm": 2.8984264488227947,
      "language_loss": 1.08421791,
      "learning_rate": 2.046196897962839e-06,
      "loss": 1.12727034,
      "num_input_tokens_seen": 451175,
      "step": 24,
      "time_per_iteration": 2.647780179977417
    },
    {
      "auxiliary_loss_clip": 0.02831438,
      "auxiliary_loss_mlp": 0.01330443,
      "balance_loss_clip": 1.55820298,
      "balance_loss_mlp": 1.11882234,
      "epoch": 0.0015030813166992333,
      "flos": 18108350801280.0,
      "grad_norm": 4.086800214849867,
      "language_loss": 1.01468349,
      "learning_rate": 2.0724802282696944e-06,
      "loss": 1.05630231,
      "num_input_tokens_seen": 468775,
      "step": 25,
      "time_per_iteration": 2.6759719848632812
    },
    {
      "auxiliary_loss_clip": 0.02825556,
      "auxiliary_loss_mlp": 0.01311203,
      "balance_loss_clip": 1.56096947,
      "balance_loss_mlp": 1.1007272,
      "epoch": 0.0015632045693672028,
      "flos": 22234558763520.0,
      "grad_norm": 2.9939950823146444,
      "language_loss": 1.06775463,
      "learning_rate": 2.0977325700720194e-06,
      "loss": 1.10912228,
      "num_input_tokens_seen": 488530,
      "step": 26,
      "time_per_iteration": 2.669372081756592
    },
    {
      "auxiliary_loss_clip": 0.02768238,
      "auxiliary_loss_mlp": 0.01325317,
      "balance_loss_clip": 1.55126643,
      "balance_loss_mlp": 1.12437797,
      "epoch": 0.001623327822035172,
      "flos": 23991955580160.0,
      "grad_norm": 2.6367761940866243,
      "language_loss": 0.95648253,
      "learning_rate": 2.122031762649933e-06,
      "loss": 0.99741799,
      "num_input_tokens_seen": 510495,
      "step": 27,
      "time_per_iteration": 2.723604917526245
    },
    {
      "auxiliary_loss_clip": 0.02746536,
      "auxiliary_loss_mlp": 0.01312074,
      "balance_loss_clip": 1.55639291,
      "balance_loss_mlp": 1.13011312,
      "epoch": 0.0016834510747031415,
      "flos": 19677647070720.0,
      "grad_norm": 2.51170184201261,
      "language_loss": 1.06443524,
      "learning_rate": 2.1454471497582483e-06,
      "loss": 1.10502148,
      "num_input_tokens_seen": 528605,
      "step": 28,
      "time_per_iteration": 2.710132598876953
    },
    {
      "auxiliary_loss_clip": 0.02710518,
      "auxiliary_loss_mlp": 0.01320961,
      "balance_loss_clip": 1.5401603,
      "balance_loss_mlp": 1.13337302,
      "epoch": 0.0017435743273711108,
      "flos": 20923819568640.0,
      "grad_norm": 2.442918982822478,
      "language_loss": 1.02692401,
      "learning_rate": 2.1680407726407727e-06,
      "loss": 1.06723869,
      "num_input_tokens_seen": 548515,
      "step": 29,
      "time_per_iteration": 2.7579588890075684
    },
    {
      "auxiliary_loss_clip": 0.02700055,
      "auxiliary_loss_mlp": 0.01313743,
      "balance_loss_clip": 1.53621042,
      "balance_loss_mlp": 1.12691784,
      "epoch": 0.00180369758003908,
      "flos": 19528976678400.0,
      "grad_norm": 3.457351653427521,
      "language_loss": 1.19303858,
      "learning_rate": 2.189868360711334e-06,
      "loss": 1.23317659,
      "num_input_tokens_seen": 564025,
      "step": 30,
      "time_per_iteration": 2.7372548580169678
    },
    {
      "auxiliary_loss_clip": 0.02619048,
      "auxiliary_loss_mlp": 0.01338699,
      "balance_loss_clip": 1.52283287,
      "balance_loss_mlp": 1.1572144,
      "epoch": 0.0018638208327070496,
      "flos": 27453169100160.0,
      "grad_norm": 2.9388233477720576,
      "language_loss": 1.02500653,
      "learning_rate": 2.2109801597326265e-06,
      "loss": 1.0645839,
      "num_input_tokens_seen": 583345,
      "step": 31,
      "time_per_iteration": 2.7278308868408203
    },
    {
      "auxiliary_loss_clip": 0.02589437,
      "auxiliary_loss_mlp": 0.01332604,
      "balance_loss_clip": 1.52444232,
      "balance_loss_mlp": 1.15274119,
      "epoch": 0.0019239440853750188,
      "flos": 13589460380160.0,
      "grad_norm": 3.0802985300117496,
      "language_loss": 0.95587057,
      "learning_rate": 2.2314216284658796e-06,
      "loss": 0.99509096,
      "num_input_tokens_seen": 600010,
      "step": 32,
      "time_per_iteration": 2.6885178089141846
    },
    {
      "auxiliary_loss_clip": 0.02570478,
      "auxiliary_loss_mlp": 0.013027,
      "balance_loss_clip": 1.51956296,
      "balance_loss_mlp": 1.13571191,
      "epoch": 0.001984067338042988,
      "flos": 11253866336640.0,
      "grad_norm": 3.076006431182188,
      "language_loss": 0.95174849,
      "learning_rate": 2.2512340280885094e-06,
      "loss": 0.99048024,
      "num_input_tokens_seen": 616295,
      "step": 33,
      "time_per_iteration": 2.630277395248413
    },
    {
      "auxiliary_loss_clip": 0.02427026,
      "auxiliary_loss_mlp": 0.01303763,
      "balance_loss_clip": 1.48692524,
      "balance_loss_mlp": 1.14640713,
      "epoch": 0.0020441905907109576,
      "flos": 22386245898240.0,
      "grad_norm": 2.19908778643991,
      "language_loss": 0.91466326,
      "learning_rate": 2.270454923596497e-06,
      "loss": 0.95197117,
      "num_input_tokens_seen": 637640,
      "step": 34,
      "time_per_iteration": 2.7644619941711426
    },
    {
      "auxiliary_loss_clip": 0.0237835,
      "auxiliary_loss_mlp": 0.01269303,
      "balance_loss_clip": 1.45181537,
      "balance_loss_mlp": 1.11518967,
      "epoch": 0.0021043138433789266,
      "flos": 49778580337920.0,
      "grad_norm": 2.726513935839375,
      "language_loss": 0.76664925,
      "learning_rate": 2.2891186125067434e-06,
      "loss": 0.80312586,
      "num_input_tokens_seen": 659710,
      "step": 35,
      "time_per_iteration": 2.8921875953674316
    },
    {
      "auxiliary_loss_clip": 0.02349098,
      "auxiliary_loss_mlp": 0.01274919,
      "balance_loss_clip": 1.4645474,
      "balance_loss_mlp": 1.13139129,
      "epoch": 0.002164437096046896,
      "flos": 20557961591040.0,
      "grad_norm": 4.391738830652312,
      "language_loss": 0.88711071,
      "learning_rate": 2.307256493152974e-06,
      "loss": 0.92335081,
      "num_input_tokens_seen": 679670,
      "step": 36,
      "time_per_iteration": 2.7051918506622314
    },
    {
      "auxiliary_loss_clip": 0.02291043,
      "auxiliary_loss_mlp": 0.01334067,
      "balance_loss_clip": 1.4514792,
      "balance_loss_mlp": 1.18796396,
      "epoch": 0.0022245603487148656,
      "flos": 26542295084160.0,
      "grad_norm": 3.317240939905439,
      "language_loss": 0.92799407,
      "learning_rate": 2.3248973825097614e-06,
      "loss": 0.9642452,
      "num_input_tokens_seen": 700170,
      "step": 37,
      "time_per_iteration": 2.736189842224121
    },
    {
      "auxiliary_loss_clip": 0.02251115,
      "auxiliary_loss_mlp": 0.01277599,
      "balance_loss_clip": 1.44627595,
      "balance_loss_mlp": 1.15610075,
      "epoch": 0.0022846836013828346,
      "flos": 20338188226560.0,
      "grad_norm": 7.254900237501925,
      "language_loss": 1.04009855,
      "learning_rate": 2.3420677916238357e-06,
      "loss": 1.07538557,
      "num_input_tokens_seen": 718545,
      "step": 38,
      "time_per_iteration": 2.717092275619507
    },
    {
      "auxiliary_loss_clip": 0.02221873,
      "auxiliary_loss_mlp": 0.01258488,
      "balance_loss_clip": 1.44061971,
      "balance_loss_mlp": 1.13555932,
      "epoch": 0.002344806854050804,
      "flos": 26247575992320.0,
      "grad_norm": 2.3980310601176247,
      "language_loss": 0.8540687,
      "learning_rate": 2.358792165262154e-06,
      "loss": 0.88887233,
      "num_input_tokens_seen": 739865,
      "step": 39,
      "time_per_iteration": 2.729814291000366
    },
    {
      "auxiliary_loss_clip": 0.02196512,
      "auxiliary_loss_mlp": 0.01250654,
      "balance_loss_clip": 1.43028092,
      "balance_loss_mlp": 1.12200332,
      "epoch": 0.0024049301067187736,
      "flos": 11801539981440.0,
      "grad_norm": 2.6927796083967332,
      "language_loss": 0.90215278,
      "learning_rate": 2.3750930912143747e-06,
      "loss": 0.93662447,
      "num_input_tokens_seen": 755770,
      "step": 40,
      "time_per_iteration": 2.652189016342163
    },
    {
      "auxiliary_loss_clip": 0.02148083,
      "auxiliary_loss_mlp": 0.01277126,
      "balance_loss_clip": 1.42066431,
      "balance_loss_mlp": 1.16330528,
      "epoch": 0.0024650533593867426,
      "flos": 20631506688000.0,
      "grad_norm": 4.20323719666801,
      "language_loss": 0.93329579,
      "learning_rate": 2.3909914837471044e-06,
      "loss": 0.96754795,
      "num_input_tokens_seen": 773440,
      "step": 41,
      "time_per_iteration": 2.6637909412384033
    },
    {
      "auxiliary_loss_clip": 0.02113187,
      "auxiliary_loss_mlp": 0.01258577,
      "balance_loss_clip": 1.41261578,
      "balance_loss_mlp": 1.15424466,
      "epoch": 0.002525176612054712,
      "flos": 18406122549120.0,
      "grad_norm": 2.842144062106217,
      "language_loss": 0.97722417,
      "learning_rate": 2.4065067449483835e-06,
      "loss": 1.01094186,
      "num_input_tokens_seen": 790455,
      "step": 42,
      "time_per_iteration": 2.6459341049194336
    },
    {
      "auxiliary_loss_clip": 0.02075838,
      "auxiliary_loss_mlp": 0.01304844,
      "balance_loss_clip": 1.41428459,
      "balance_loss_mlp": 1.19660187,
      "epoch": 0.0025852998647226816,
      "flos": 28184023128960.0,
      "grad_norm": 2.4960241772092204,
      "language_loss": 0.97440422,
      "learning_rate": 2.4216569070848724e-06,
      "loss": 1.00821102,
      "num_input_tokens_seen": 810645,
      "step": 43,
      "time_per_iteration": 2.7594943046569824
    },
    {
      "auxiliary_loss_clip": 0.02095906,
      "auxiliary_loss_mlp": 0.01318647,
      "balance_loss_clip": 1.41281855,
      "balance_loss_mlp": 1.20506418,
      "epoch": 0.0026454231173906506,
      "flos": 14283110897280.0,
      "grad_norm": 2.0273665833954135,
      "language_loss": 0.9364934,
      "learning_rate": 2.4364587585915504e-06,
      "loss": 0.97063887,
      "num_input_tokens_seen": 827470,
      "step": 44,
      "time_per_iteration": 2.692615270614624
    },
    {
      "auxiliary_loss_clip": 0.02060208,
      "auxiliary_loss_mlp": 0.01275303,
      "balance_loss_clip": 1.41113913,
      "balance_loss_mlp": 1.17612088,
      "epoch": 0.00270554637005862,
      "flos": 22419211605120.0,
      "grad_norm": 2.361400658571967,
      "language_loss": 0.98666775,
      "learning_rate": 2.450927955901469e-06,
      "loss": 1.02002287,
      "num_input_tokens_seen": 847285,
      "step": 45,
      "time_per_iteration": 2.7195322513580322
    },
    {
      "auxiliary_loss_clip": 0.02033795,
      "auxiliary_loss_mlp": 0.01232111,
      "balance_loss_clip": 1.3965255,
      "balance_loss_mlp": 1.1438005,
      "epoch": 0.0027656696227265896,
      "flos": 23985778440960.0,
      "grad_norm": 2.195102108830285,
      "language_loss": 1.02617097,
      "learning_rate": 2.465079122983384e-06,
      "loss": 1.05883002,
      "num_input_tokens_seen": 867545,
      "step": 46,
      "time_per_iteration": 2.707385540008545
    },
    {
      "auxiliary_loss_clip": 0.01998348,
      "auxiliary_loss_mlp": 0.01274711,
      "balance_loss_clip": 1.38709247,
      "balance_loss_mlp": 1.18301487,
      "epoch": 0.0028257928753945586,
      "flos": 37669503087360.0,
      "grad_norm": 2.7655537814074775,
      "language_loss": 0.88142788,
      "learning_rate": 2.4789259401737868e-06,
      "loss": 0.91415852,
      "num_input_tokens_seen": 889915,
      "step": 47,
      "time_per_iteration": 2.8250505924224854
    },
    {
      "auxiliary_loss_clip": 0.01959817,
      "auxiliary_loss_mlp": 0.01257752,
      "balance_loss_clip": 1.3761456,
      "balance_loss_mlp": 1.17049098,
      "epoch": 0.002885916128062528,
      "flos": 22454547609600.0,
      "grad_norm": 3.1940090818211533,
      "language_loss": 0.87990278,
      "learning_rate": 2.492481223656015e-06,
      "loss": 0.9120785,
      "num_input_tokens_seen": 908975,
      "step": 48,
      "time_per_iteration": 2.6966872215270996
    },
    {
      "auxiliary_loss_clip": 0.01959587,
      "auxiliary_loss_mlp": 0.01244453,
      "balance_loss_clip": 1.36591625,
      "balance_loss_mlp": 1.15137434,
      "epoch": 0.0029460393807304976,
      "flos": 27012796358400.0,
      "grad_norm": 2.2473168101519976,
      "language_loss": 0.89837682,
      "learning_rate": 2.5057569967437924e-06,
      "loss": 0.93041724,
      "num_input_tokens_seen": 929810,
      "step": 49,
      "time_per_iteration": 2.7578954696655273
    },
    {
      "auxiliary_loss_clip": 0.0195209,
      "auxiliary_loss_mlp": 0.01236839,
      "balance_loss_clip": 1.36050057,
      "balance_loss_mlp": 1.15086579,
      "epoch": 0.0030061626333984666,
      "flos": 15851832549120.0,
      "grad_norm": 2.678205539545924,
      "language_loss": 0.91038376,
      "learning_rate": 2.51876455396287e-06,
      "loss": 0.94227308,
      "num_input_tokens_seen": 948650,
      "step": 50,
      "time_per_iteration": 2.7679762840270996
    },
    {
      "auxiliary_loss_clip": 0.01949342,
      "auxiliary_loss_mlp": 0.01201979,
      "balance_loss_clip": 1.36510336,
      "balance_loss_mlp": 1.11948586,
      "epoch": 0.003066285886066436,
      "flos": 31827052316160.0,
      "grad_norm": 6.998995439329058,
      "language_loss": 0.87146574,
      "learning_rate": 2.5315145187866316e-06,
      "loss": 0.90297896,
      "num_input_tokens_seen": 966455,
      "step": 51,
      "time_per_iteration": 2.80306339263916
    },
    {
      "auxiliary_loss_clip": 0.0190609,
      "auxiliary_loss_mlp": 0.01209201,
      "balance_loss_clip": 1.35485411,
      "balance_loss_mlp": 1.12952113,
      "epoch": 0.0031264091387344056,
      "flos": 41427482774400.0,
      "grad_norm": 2.0861152285633393,
      "language_loss": 0.95259207,
      "learning_rate": 2.5440168957651953e-06,
      "loss": 0.98374504,
      "num_input_tokens_seen": 988110,
      "step": 52,
      "time_per_iteration": 2.896541118621826
    },
    {
      "auxiliary_loss_clip": 0.01902539,
      "auxiliary_loss_mlp": 0.01242914,
      "balance_loss_clip": 1.35093045,
      "balance_loss_mlp": 1.16213787,
      "epoch": 0.0031865323914023747,
      "flos": 23440941970560.0,
      "grad_norm": 7.435644268017929,
      "language_loss": 0.92274666,
      "learning_rate": 2.5562811176888872e-06,
      "loss": 0.95420128,
      "num_input_tokens_seen": 1008550,
      "step": 53,
      "time_per_iteration": 2.766141891479492
    },
    {
      "auxiliary_loss_clip": 0.01892641,
      "auxiliary_loss_mlp": 0.01197169,
      "balance_loss_clip": 1.35497916,
      "balance_loss_mlp": 1.11548698,
      "epoch": 0.003246655644070344,
      "flos": 14429195510400.0,
      "grad_norm": 2.450631049511502,
      "language_loss": 0.82807302,
      "learning_rate": 2.5683160883431093e-06,
      "loss": 0.85897124,
      "num_input_tokens_seen": 1026840,
      "step": 54,
      "time_per_iteration": 4.120918273925781
    },
    {
      "auxiliary_loss_clip": 0.01889992,
      "auxiliary_loss_mlp": 0.0120935,
      "balance_loss_clip": 1.34473562,
      "balance_loss_mlp": 1.12957573,
      "epoch": 0.0033067788967383136,
      "flos": 35918247496320.0,
      "grad_norm": 2.5900754745378096,
      "language_loss": 0.81334317,
      "learning_rate": 2.580130221340046e-06,
      "loss": 0.84433663,
      "num_input_tokens_seen": 1048875,
      "step": 55,
      "time_per_iteration": 4.219754695892334
    },
    {
      "auxiliary_loss_clip": 0.01879125,
      "auxiliary_loss_mlp": 0.0120095,
      "balance_loss_clip": 1.33860755,
      "balance_loss_mlp": 1.1210326,
      "epoch": 0.003366902149406283,
      "flos": 22958732862720.0,
      "grad_norm": 2.659004278044306,
      "language_loss": 0.87156689,
      "learning_rate": 2.5917314754514246e-06,
      "loss": 0.90236759,
      "num_input_tokens_seen": 1066435,
      "step": 56,
      "time_per_iteration": 2.8395962715148926
    },
    {
      "auxiliary_loss_clip": 0.01880357,
      "auxiliary_loss_mlp": 0.0116199,
      "balance_loss_clip": 1.33138347,
      "balance_loss_mlp": 1.08779407,
      "epoch": 0.003427025402074252,
      "flos": 26582838560640.0,
      "grad_norm": 1.935221967321181,
      "language_loss": 0.92988372,
      "learning_rate": 2.6031273868139713e-06,
      "loss": 0.96030712,
      "num_input_tokens_seen": 1090330,
      "step": 57,
      "time_per_iteration": 2.7686638832092285
    },
    {
      "auxiliary_loss_clip": 0.01843143,
      "auxiliary_loss_mlp": 0.01215065,
      "balance_loss_clip": 1.33584261,
      "balance_loss_mlp": 1.1426816,
      "epoch": 0.0034871486547422216,
      "flos": 23951196622080.0,
      "grad_norm": 2.15414077833082,
      "language_loss": 0.99566519,
      "learning_rate": 2.614325098333948e-06,
      "loss": 1.02624726,
      "num_input_tokens_seen": 1109840,
      "step": 58,
      "time_per_iteration": 2.8083763122558594
    },
    {
      "auxiliary_loss_clip": 0.01824337,
      "auxiliary_loss_mlp": 0.01196799,
      "balance_loss_clip": 1.32193422,
      "balance_loss_mlp": 1.12532127,
      "epoch": 0.003547271907410191,
      "flos": 21214983214080.0,
      "grad_norm": 2.1868428241543847,
      "language_loss": 0.88220567,
      "learning_rate": 2.625331386578098e-06,
      "loss": 0.91241705,
      "num_input_tokens_seen": 1128415,
      "step": 59,
      "time_per_iteration": 2.7655038833618164
    },
    {
      "auxiliary_loss_clip": 0.01845515,
      "auxiliary_loss_mlp": 0.01158162,
      "balance_loss_clip": 1.3304832,
      "balance_loss_mlp": 1.0851109,
      "epoch": 0.00360739516007816,
      "flos": 16504903676160.0,
      "grad_norm": 2.056067662422279,
      "language_loss": 0.93512869,
      "learning_rate": 2.63615268640451e-06,
      "loss": 0.9651655,
      "num_input_tokens_seen": 1146515,
      "step": 60,
      "time_per_iteration": 2.7158358097076416
    },
    {
      "auxiliary_loss_clip": 0.01825818,
      "auxiliary_loss_mlp": 0.01171191,
      "balance_loss_clip": 1.3151412,
      "balance_loss_mlp": 1.10262179,
      "epoch": 0.0036675184127461296,
      "flos": 19464805031040.0,
      "grad_norm": 2.191723713012442,
      "language_loss": 0.89917547,
      "learning_rate": 2.6467951135575943e-06,
      "loss": 0.92914557,
      "num_input_tokens_seen": 1166330,
      "step": 61,
      "time_per_iteration": 2.739276885986328
    },
    {
      "auxiliary_loss_clip": 0.01809171,
      "auxiliary_loss_mlp": 0.01142422,
      "balance_loss_clip": 1.31125736,
      "balance_loss_mlp": 1.07523572,
      "epoch": 0.003727641665414099,
      "flos": 20957323979520.0,
      "grad_norm": 1.900677919231022,
      "language_loss": 0.88398385,
      "learning_rate": 2.657264485425803e-06,
      "loss": 0.91349977,
      "num_input_tokens_seen": 1186010,
      "step": 62,
      "time_per_iteration": 2.707585573196411
    },
    {
      "auxiliary_loss_clip": 0.01789304,
      "auxiliary_loss_mlp": 0.01161585,
      "balance_loss_clip": 1.30166042,
      "balance_loss_mlp": 1.09153748,
      "epoch": 0.003787764918082068,
      "flos": 18406050721920.0,
      "grad_norm": 2.7560077151168842,
      "language_loss": 0.96264374,
      "learning_rate": 2.6675663401385186e-06,
      "loss": 0.99215263,
      "num_input_tokens_seen": 1204985,
      "step": 63,
      "time_per_iteration": 2.656906843185425
    },
    {
      "auxiliary_loss_clip": 0.01796334,
      "auxiliary_loss_mlp": 0.01171342,
      "balance_loss_clip": 1.3090297,
      "balance_loss_mlp": 1.10444212,
      "epoch": 0.0038478881707500376,
      "flos": 12459243962880.0,
      "grad_norm": 3.893973202840674,
      "language_loss": 0.98791313,
      "learning_rate": 2.677705954159056e-06,
      "loss": 1.01758993,
      "num_input_tokens_seen": 1223545,
      "step": 64,
      "time_per_iteration": 2.6722960472106934
    },
    {
      "auxiliary_loss_clip": 0.01803413,
      "auxiliary_loss_mlp": 0.01148376,
      "balance_loss_clip": 1.30799556,
      "balance_loss_mlp": 1.08066511,
      "epoch": 0.003908011423418007,
      "flos": 13553334276480.0,
      "grad_norm": 2.7879410686899453,
      "language_loss": 0.85275209,
      "learning_rate": 2.6876883585136904e-06,
      "loss": 0.88226998,
      "num_input_tokens_seen": 1241175,
      "step": 65,
      "time_per_iteration": 2.6174910068511963
    },
    {
      "auxiliary_loss_clip": 0.017804,
      "auxiliary_loss_mlp": 0.01155168,
      "balance_loss_clip": 1.29514968,
      "balance_loss_mlp": 1.08750486,
      "epoch": 0.003968134676085976,
      "flos": 18333475292160.0,
      "grad_norm": 1.9571751601986513,
      "language_loss": 0.85182297,
      "learning_rate": 2.697518353781685e-06,
      "loss": 0.88117868,
      "num_input_tokens_seen": 1259315,
      "step": 66,
      "time_per_iteration": 2.658709764480591
    },
    {
      "auxiliary_loss_clip": 0.01783856,
      "auxiliary_loss_mlp": 0.0115392,
      "balance_loss_clip": 1.29331565,
      "balance_loss_mlp": 1.07791221,
      "epoch": 0.004028257928753946,
      "flos": 20485242506880.0,
      "grad_norm": 2.3698059868821684,
      "language_loss": 0.96468216,
      "learning_rate": 2.7072005239581103e-06,
      "loss": 0.99406004,
      "num_input_tokens_seen": 1277055,
      "step": 67,
      "time_per_iteration": 2.641746759414673
    },
    {
      "auxiliary_loss_clip": 0.01754411,
      "auxiliary_loss_mlp": 0.01155409,
      "balance_loss_clip": 1.28637362,
      "balance_loss_mlp": 1.08288217,
      "epoch": 0.004088381181421915,
      "flos": 18843837684480.0,
      "grad_norm": 2.7406767736009443,
      "language_loss": 0.94616473,
      "learning_rate": 2.7167392492896727e-06,
      "loss": 0.97526288,
      "num_input_tokens_seen": 1294355,
      "step": 68,
      "time_per_iteration": 2.64577579498291
    },
    {
      "auxiliary_loss_clip": 0.01749401,
      "auxiliary_loss_mlp": 0.0115652,
      "balance_loss_clip": 1.28411579,
      "balance_loss_mlp": 1.08666384,
      "epoch": 0.004148504434089885,
      "flos": 19427817000960.0,
      "grad_norm": 2.1205839470398553,
      "language_loss": 0.95834875,
      "learning_rate": 2.7261387181735195e-06,
      "loss": 0.98740798,
      "num_input_tokens_seen": 1313525,
      "step": 69,
      "time_per_iteration": 2.646423101425171
    },
    {
      "auxiliary_loss_clip": 0.01742981,
      "auxiliary_loss_mlp": 0.01158017,
      "balance_loss_clip": 1.28688824,
      "balance_loss_mlp": 1.09292889,
      "epoch": 0.004208627686757853,
      "flos": 20811023884800.0,
      "grad_norm": 2.3019682152894108,
      "language_loss": 0.98015571,
      "learning_rate": 2.7354029381999196e-06,
      "loss": 1.00916576,
      "num_input_tokens_seen": 1330505,
      "step": 70,
      "time_per_iteration": 2.632352352142334
    },
    {
      "auxiliary_loss_clip": 0.01749632,
      "auxiliary_loss_mlp": 0.01144484,
      "balance_loss_clip": 1.27708316,
      "balance_loss_mlp": 1.07515192,
      "epoch": 0.004268750939425823,
      "flos": 19098623831040.0,
      "grad_norm": 2.6068441171994383,
      "language_loss": 0.9399153,
      "learning_rate": 2.7445357464116983e-06,
      "loss": 0.96885651,
      "num_input_tokens_seen": 1349615,
      "step": 71,
      "time_per_iteration": 2.668095588684082
    },
    {
      "auxiliary_loss_clip": 0.01823203,
      "auxiliary_loss_mlp": 0.01299011,
      "balance_loss_clip": 1.43772125,
      "balance_loss_mlp": 1.26105523,
      "epoch": 0.004328874192093792,
      "flos": 52439635514880.0,
      "grad_norm": 2.410621039578377,
      "language_loss": 0.65646505,
      "learning_rate": 2.75354081884615e-06,
      "loss": 0.68768716,
      "num_input_tokens_seen": 1410275,
      "step": 72,
      "time_per_iteration": 3.161329507827759
    },
    {
      "auxiliary_loss_clip": 0.01806294,
      "auxiliary_loss_mlp": 0.01279273,
      "balance_loss_clip": 1.43114007,
      "balance_loss_mlp": 1.24131632,
      "epoch": 0.004388997444761762,
      "flos": 66473239564800.0,
      "grad_norm": 2.3489531348975827,
      "language_loss": 0.63693726,
      "learning_rate": 2.7624216794188286e-06,
      "loss": 0.66779292,
      "num_input_tokens_seen": 1473020,
      "step": 73,
      "time_per_iteration": 3.2129199504852295
    },
    {
      "auxiliary_loss_clip": 0.01722247,
      "auxiliary_loss_mlp": 0.01143778,
      "balance_loss_clip": 1.26832461,
      "balance_loss_mlp": 1.0759716,
      "epoch": 0.004449120697429731,
      "flos": 18952970181120.0,
      "grad_norm": 5.099355792569285,
      "language_loss": 0.85942364,
      "learning_rate": 2.771181708202938e-06,
      "loss": 0.88808388,
      "num_input_tokens_seen": 1490385,
      "step": 74,
      "time_per_iteration": 2.6345021724700928
    },
    {
      "auxiliary_loss_clip": 0.01726457,
      "auxiliary_loss_mlp": 0.01162006,
      "balance_loss_clip": 1.26792169,
      "balance_loss_mlp": 1.09281671,
      "epoch": 0.004509243950097701,
      "flos": 21105491581440.0,
      "grad_norm": 2.014284118337651,
      "language_loss": 0.97076827,
      "learning_rate": 2.779824149153005e-06,
      "loss": 0.99965286,
      "num_input_tokens_seen": 1509725,
      "step": 75,
      "time_per_iteration": 2.652470350265503
    },
    {
      "auxiliary_loss_clip": 0.01704382,
      "auxiliary_loss_mlp": 0.01143069,
      "balance_loss_clip": 1.26351404,
      "balance_loss_mlp": 1.07655072,
      "epoch": 0.004569367202765669,
      "flos": 20698730991360.0,
      "grad_norm": 2.3505645788118255,
      "language_loss": 0.87827659,
      "learning_rate": 2.788352117317012e-06,
      "loss": 0.90675104,
      "num_input_tokens_seen": 1527245,
      "step": 76,
      "time_per_iteration": 2.6256396770477295
    },
    {
      "auxiliary_loss_clip": 0.01705014,
      "auxiliary_loss_mlp": 0.01147183,
      "balance_loss_clip": 1.26219273,
      "balance_loss_mlp": 1.07746994,
      "epoch": 0.004629490455433639,
      "flos": 28658474899200.0,
      "grad_norm": 1.8559931700429257,
      "language_loss": 0.91905999,
      "learning_rate": 2.796768605577095e-06,
      "loss": 0.94758201,
      "num_input_tokens_seen": 1548930,
      "step": 77,
      "time_per_iteration": 2.687666893005371
    },
    {
      "auxiliary_loss_clip": 0.01694467,
      "auxiliary_loss_mlp": 0.01168123,
      "balance_loss_clip": 1.26229167,
      "balance_loss_mlp": 1.09764647,
      "epoch": 0.004689613708101608,
      "flos": 11072409805440.0,
      "grad_norm": 2.1652402395719954,
      "language_loss": 0.92441529,
      "learning_rate": 2.80507649095533e-06,
      "loss": 0.95304114,
      "num_input_tokens_seen": 1565695,
      "step": 78,
      "time_per_iteration": 2.58862042427063
    },
    {
      "auxiliary_loss_clip": 0.0169101,
      "auxiliary_loss_mlp": 0.01155126,
      "balance_loss_clip": 1.25769186,
      "balance_loss_mlp": 1.08693886,
      "epoch": 0.004749736960769578,
      "flos": 21799106184960.0,
      "grad_norm": 2.108146663949779,
      "language_loss": 0.82620943,
      "learning_rate": 2.813278540517843e-06,
      "loss": 0.85467076,
      "num_input_tokens_seen": 1582625,
      "step": 79,
      "time_per_iteration": 2.6427085399627686
    },
    {
      "auxiliary_loss_clip": 0.01703252,
      "auxiliary_loss_mlp": 0.01134423,
      "balance_loss_clip": 1.25972986,
      "balance_loss_mlp": 1.06428039,
      "epoch": 0.004809860213437547,
      "flos": 19792597570560.0,
      "grad_norm": 2.1708779551822315,
      "language_loss": 0.91327262,
      "learning_rate": 2.8213774169075505e-06,
      "loss": 0.94164932,
      "num_input_tokens_seen": 1601725,
      "step": 80,
      "time_per_iteration": 2.6145427227020264
    },
    {
      "auxiliary_loss_clip": 0.01672714,
      "auxiliary_loss_mlp": 0.01144933,
      "balance_loss_clip": 1.25285006,
      "balance_loss_mlp": 1.07450438,
      "epoch": 0.004869983466105517,
      "flos": 26574327037440.0,
      "grad_norm": 2.2416963077029606,
      "language_loss": 0.94922227,
      "learning_rate": 2.829375683533245e-06,
      "loss": 0.97739875,
      "num_input_tokens_seen": 1622420,
      "step": 81,
      "time_per_iteration": 2.6512176990509033
    },
    {
      "auxiliary_loss_clip": 0.01686987,
      "auxiliary_loss_mlp": 0.01148845,
      "balance_loss_clip": 1.25672615,
      "balance_loss_mlp": 1.08256507,
      "epoch": 0.004930106718773485,
      "flos": 12823378087680.0,
      "grad_norm": 3.669547251248084,
      "language_loss": 0.9616437,
      "learning_rate": 2.8372758094402803e-06,
      "loss": 0.99000204,
      "num_input_tokens_seen": 1640715,
      "step": 82,
      "time_per_iteration": 2.643275499343872
    },
    {
      "auxiliary_loss_clip": 0.01670098,
      "auxiliary_loss_mlp": 0.01158464,
      "balance_loss_clip": 1.24507713,
      "balance_loss_mlp": 1.08875036,
      "epoch": 0.004990229971441455,
      "flos": 25774919902080.0,
      "grad_norm": 1.9758254064941683,
      "language_loss": 0.86667681,
      "learning_rate": 2.84508017388607e-06,
      "loss": 0.89496243,
      "num_input_tokens_seen": 1662210,
      "step": 83,
      "time_per_iteration": 2.670858144760132
    },
    {
      "auxiliary_loss_clip": 0.01662818,
      "auxiliary_loss_mlp": 0.01157515,
      "balance_loss_clip": 1.24631083,
      "balance_loss_mlp": 1.08756244,
      "epoch": 0.005050353224109424,
      "flos": 17457254922240.0,
      "grad_norm": 2.9897262230439945,
      "language_loss": 0.91794491,
      "learning_rate": 2.852791070641559e-06,
      "loss": 0.94614816,
      "num_input_tokens_seen": 1681070,
      "step": 84,
      "time_per_iteration": 2.6115405559539795
    },
    {
      "auxiliary_loss_clip": 0.01651316,
      "auxiliary_loss_mlp": 0.01176073,
      "balance_loss_clip": 1.35944915,
      "balance_loss_mlp": 1.13964319,
      "epoch": 0.005110476476777394,
      "flos": 69805460367360.0,
      "grad_norm": 1.4054310774635628,
      "language_loss": 0.62552822,
      "learning_rate": 2.8604107120381682e-06,
      "loss": 0.65380204,
      "num_input_tokens_seen": 1747140,
      "step": 85,
      "time_per_iteration": 3.2172696590423584
    },
    {
      "auxiliary_loss_clip": 0.01649871,
      "auxiliary_loss_mlp": 0.01128349,
      "balance_loss_clip": 1.23618102,
      "balance_loss_mlp": 1.05791974,
      "epoch": 0.005170599729445363,
      "flos": 24790105739520.0,
      "grad_norm": 1.7500560553946958,
      "language_loss": 0.9083637,
      "learning_rate": 2.8679412327780482e-06,
      "loss": 0.9361459,
      "num_input_tokens_seen": 1767475,
      "step": 86,
      "time_per_iteration": 2.6689977645874023
    },
    {
      "auxiliary_loss_clip": 0.01653536,
      "auxiliary_loss_mlp": 0.01160042,
      "balance_loss_clip": 1.24235368,
      "balance_loss_mlp": 1.08913648,
      "epoch": 0.005230722982113333,
      "flos": 23258048895360.0,
      "grad_norm": 2.4677866560037742,
      "language_loss": 0.82005864,
      "learning_rate": 2.8753846935240833e-06,
      "loss": 0.84819448,
      "num_input_tokens_seen": 1784980,
      "step": 87,
      "time_per_iteration": 2.670135021209717
    },
    {
      "auxiliary_loss_clip": 0.01643607,
      "auxiliary_loss_mlp": 0.01154,
      "balance_loss_clip": 1.24081039,
      "balance_loss_mlp": 1.0852406,
      "epoch": 0.005290846234781301,
      "flos": 16727909264640.0,
      "grad_norm": 1.961193686878651,
      "language_loss": 0.9573524,
      "learning_rate": 2.8827430842847267e-06,
      "loss": 0.9853285,
      "num_input_tokens_seen": 1803030,
      "step": 88,
      "time_per_iteration": 2.6151418685913086
    },
    {
      "auxiliary_loss_clip": 0.01659738,
      "auxiliary_loss_mlp": 0.01148207,
      "balance_loss_clip": 1.24067593,
      "balance_loss_mlp": 1.08106875,
      "epoch": 0.005350969487449271,
      "flos": 20886077352960.0,
      "grad_norm": 2.347517995338878,
      "language_loss": 0.85947311,
      "learning_rate": 2.8900183276075957e-06,
      "loss": 0.88755256,
      "num_input_tokens_seen": 1822865,
      "step": 89,
      "time_per_iteration": 2.6218857765197754
    },
    {
      "auxiliary_loss_clip": 0.01646946,
      "auxiliary_loss_mlp": 0.0113309,
      "balance_loss_clip": 1.23444152,
      "balance_loss_mlp": 1.06652355,
      "epoch": 0.00541109274011724,
      "flos": 26209977431040.0,
      "grad_norm": 2.241965829183184,
      "language_loss": 0.91555059,
      "learning_rate": 2.8972122815946455e-06,
      "loss": 0.94335091,
      "num_input_tokens_seen": 1842435,
      "step": 90,
      "time_per_iteration": 2.6585636138916016
    },
    {
      "auxiliary_loss_clip": 0.016283,
      "auxiliary_loss_mlp": 0.01133835,
      "balance_loss_clip": 1.23020971,
      "balance_loss_mlp": 1.06564736,
      "epoch": 0.00547121599278521,
      "flos": 21178569801600.0,
      "grad_norm": 2.145498282744817,
      "language_loss": 0.8583855,
      "learning_rate": 2.90432674275074e-06,
      "loss": 0.88600683,
      "num_input_tokens_seen": 1860065,
      "step": 91,
      "time_per_iteration": 2.5992393493652344
    },
    {
      "auxiliary_loss_clip": 0.01626555,
      "auxiliary_loss_mlp": 0.01138118,
      "balance_loss_clip": 1.22574973,
      "balance_loss_mlp": 1.07236195,
      "epoch": 0.005531339245453179,
      "flos": 19718801078400.0,
      "grad_norm": 2.2441876605118694,
      "language_loss": 0.86894584,
      "learning_rate": 2.91136344867656e-06,
      "loss": 0.89659262,
      "num_input_tokens_seen": 1878135,
      "step": 92,
      "time_per_iteration": 2.618533134460449
    },
    {
      "auxiliary_loss_clip": 0.01619025,
      "auxiliary_loss_mlp": 0.0117656,
      "balance_loss_clip": 1.21727514,
      "balance_loss_mlp": 1.10861111,
      "epoch": 0.005591462498121149,
      "flos": 17636089760640.0,
      "grad_norm": 3.0834147923948843,
      "language_loss": 0.91943467,
      "learning_rate": 2.918324080615938e-06,
      "loss": 0.9473905,
      "num_input_tokens_seen": 1894895,
      "step": 93,
      "time_per_iteration": 3.972862720489502
    },
    {
      "auxiliary_loss_clip": 0.01630959,
      "auxiliary_loss_mlp": 0.01151034,
      "balance_loss_clip": 1.22391486,
      "balance_loss_mlp": 1.0800333,
      "epoch": 0.005651585750789117,
      "flos": 20011221699840.0,
      "grad_norm": 2.185108478248704,
      "language_loss": 0.87443423,
      "learning_rate": 2.925210265866963e-06,
      "loss": 0.90225422,
      "num_input_tokens_seen": 1913220,
      "step": 94,
      "time_per_iteration": 4.0016748905181885
    },
    {
      "auxiliary_loss_clip": 0.01567896,
      "auxiliary_loss_mlp": 0.01056951,
      "balance_loss_clip": 1.31671977,
      "balance_loss_mlp": 1.02242851,
      "epoch": 0.005711709003457087,
      "flos": 59812957981440.0,
      "grad_norm": 1.37905093686974,
      "language_loss": 0.6814155,
      "learning_rate": 2.932023580065507e-06,
      "loss": 0.70766395,
      "num_input_tokens_seen": 1970970,
      "step": 95,
      "time_per_iteration": 4.422675848007202
    },
    {
      "auxiliary_loss_clip": 0.0160875,
      "auxiliary_loss_mlp": 0.0115055,
      "balance_loss_clip": 1.21243787,
      "balance_loss_mlp": 1.08384049,
      "epoch": 0.005771832256125056,
      "flos": 15559591495680.0,
      "grad_norm": 2.32620823946676,
      "language_loss": 0.90304142,
      "learning_rate": 2.9387655493491906e-06,
      "loss": 0.93063444,
      "num_input_tokens_seen": 1988930,
      "step": 96,
      "time_per_iteration": 3.933746576309204
    },
    {
      "auxiliary_loss_clip": 0.01600813,
      "auxiliary_loss_mlp": 0.01137477,
      "balance_loss_clip": 1.21507454,
      "balance_loss_mlp": 1.07539272,
      "epoch": 0.005831955508793026,
      "flos": 22528380015360.0,
      "grad_norm": 2.6204599891678457,
      "language_loss": 0.89715993,
      "learning_rate": 2.9454376524092147e-06,
      "loss": 0.92454278,
      "num_input_tokens_seen": 2006285,
      "step": 97,
      "time_per_iteration": 2.60235595703125
    },
    {
      "auxiliary_loss_clip": 0.01587801,
      "auxiliary_loss_mlp": 0.01133789,
      "balance_loss_clip": 1.20708227,
      "balance_loss_mlp": 1.06507659,
      "epoch": 0.005892078761460995,
      "flos": 22049834094720.0,
      "grad_norm": 2.0322866039359617,
      "language_loss": 0.76363957,
      "learning_rate": 2.952041322436969e-06,
      "loss": 0.79085547,
      "num_input_tokens_seen": 2024905,
      "step": 98,
      "time_per_iteration": 2.591341733932495
    },
    {
      "auxiliary_loss_clip": 0.01539643,
      "auxiliary_loss_mlp": 0.01040228,
      "balance_loss_clip": 1.29784811,
      "balance_loss_mlp": 1.00627744,
      "epoch": 0.005952202014128965,
      "flos": 68539143317760.0,
      "grad_norm": 1.664661326294011,
      "language_loss": 0.65456271,
      "learning_rate": 2.9585779489718204e-06,
      "loss": 0.68036139,
      "num_input_tokens_seen": 2086220,
      "step": 99,
      "time_per_iteration": 3.197314739227295
    },
    {
      "auxiliary_loss_clip": 0.0158858,
      "auxiliary_loss_mlp": 0.01141251,
      "balance_loss_clip": 1.20651817,
      "balance_loss_mlp": 1.07096481,
      "epoch": 0.006012325266796933,
      "flos": 22960887678720.0,
      "grad_norm": 1.9390287261590322,
      "language_loss": 0.90803373,
      "learning_rate": 2.9650488796560464e-06,
      "loss": 0.93533206,
      "num_input_tokens_seen": 2103365,
      "step": 100,
      "time_per_iteration": 2.6139118671417236
    },
    {
      "auxiliary_loss_clip": 0.01600248,
      "auxiliary_loss_mlp": 0.01146735,
      "balance_loss_clip": 1.2089324,
      "balance_loss_mlp": 1.08059716,
      "epoch": 0.006072448519464903,
      "flos": 17347942857600.0,
      "grad_norm": 2.0278910381237685,
      "language_loss": 0.90985447,
      "learning_rate": 2.971455421902446e-06,
      "loss": 0.93732429,
      "num_input_tokens_seen": 2121995,
      "step": 101,
      "time_per_iteration": 2.5757923126220703
    },
    {
      "auxiliary_loss_clip": 0.01587865,
      "auxiliary_loss_mlp": 0.0114934,
      "balance_loss_clip": 1.209023,
      "balance_loss_mlp": 1.07929277,
      "epoch": 0.006132571772132872,
      "flos": 24681116897280.0,
      "grad_norm": 2.1615040216640615,
      "language_loss": 0.90539706,
      "learning_rate": 2.9777988444798075e-06,
      "loss": 0.93276918,
      "num_input_tokens_seen": 2141815,
      "step": 102,
      "time_per_iteration": 2.629912853240967
    },
    {
      "auxiliary_loss_clip": 0.01582211,
      "auxiliary_loss_mlp": 0.01131483,
      "balance_loss_clip": 1.20589662,
      "balance_loss_mlp": 1.06696725,
      "epoch": 0.006192695024800842,
      "flos": 21465675210240.0,
      "grad_norm": 3.2030167113942607,
      "language_loss": 0.87929732,
      "learning_rate": 2.9840803790210285e-06,
      "loss": 0.90643418,
      "num_input_tokens_seen": 2161125,
      "step": 103,
      "time_per_iteration": 2.636082887649536
    },
    {
      "auxiliary_loss_clip": 0.01581841,
      "auxiliary_loss_mlp": 0.01136058,
      "balance_loss_clip": 1.20760822,
      "balance_loss_mlp": 1.06977773,
      "epoch": 0.006252818277468811,
      "flos": 17420410546560.0,
      "grad_norm": 1.8735352852882923,
      "language_loss": 0.9355042,
      "learning_rate": 2.990301221458371e-06,
      "loss": 0.96268308,
      "num_input_tokens_seen": 2179510,
      "step": 104,
      "time_per_iteration": 2.6202988624572754
    },
    {
      "auxiliary_loss_clip": 0.01573431,
      "auxiliary_loss_mlp": 0.01146299,
      "balance_loss_clip": 1.1980418,
      "balance_loss_mlp": 1.08240306,
      "epoch": 0.006312941530136781,
      "flos": 19099557584640.0,
      "grad_norm": 2.7313439084549205,
      "language_loss": 0.96352613,
      "learning_rate": 2.9964625333900544e-06,
      "loss": 0.99072349,
      "num_input_tokens_seen": 2197870,
      "step": 105,
      "time_per_iteration": 2.5902717113494873
    },
    {
      "auxiliary_loss_clip": 0.01570953,
      "auxiliary_loss_mlp": 0.01158322,
      "balance_loss_clip": 1.19701767,
      "balance_loss_mlp": 1.08727288,
      "epoch": 0.006373064782804749,
      "flos": 24060831909120.0,
      "grad_norm": 2.553229446543694,
      "language_loss": 0.86808527,
      "learning_rate": 3.002565443382063e-06,
      "loss": 0.89537799,
      "num_input_tokens_seen": 2217495,
      "step": 106,
      "time_per_iteration": 2.647836923599243
    },
    {
      "auxiliary_loss_clip": 0.01555072,
      "auxiliary_loss_mlp": 0.0114002,
      "balance_loss_clip": 1.18401361,
      "balance_loss_mlp": 1.07216597,
      "epoch": 0.006433188035472719,
      "flos": 18332433797760.0,
      "grad_norm": 2.1469546870305622,
      "language_loss": 0.83395767,
      "learning_rate": 3.008611048208843e-06,
      "loss": 0.86090857,
      "num_input_tokens_seen": 2236520,
      "step": 107,
      "time_per_iteration": 2.5631537437438965
    },
    {
      "auxiliary_loss_clip": 0.01482584,
      "auxiliary_loss_mlp": 0.01035622,
      "balance_loss_clip": 1.25975657,
      "balance_loss_mlp": 1.00396001,
      "epoch": 0.006493311288140688,
      "flos": 62562387594240.0,
      "grad_norm": 0.998404320905639,
      "language_loss": 0.64882183,
      "learning_rate": 3.014600414036285e-06,
      "loss": 0.67400384,
      "num_input_tokens_seen": 2300140,
      "step": 108,
      "time_per_iteration": 3.1637301445007324
    },
    {
      "auxiliary_loss_clip": 0.01548455,
      "auxiliary_loss_mlp": 0.0113263,
      "balance_loss_clip": 1.18692231,
      "balance_loss_mlp": 1.06463289,
      "epoch": 0.006553434540808658,
      "flos": 19500141035520.0,
      "grad_norm": 2.053521313218508,
      "language_loss": 0.97542548,
      "learning_rate": 3.0205345775501937e-06,
      "loss": 1.00223637,
      "num_input_tokens_seen": 2317320,
      "step": 109,
      "time_per_iteration": 2.5526187419891357
    },
    {
      "auxiliary_loss_clip": 0.01545556,
      "auxiliary_loss_mlp": 0.01140444,
      "balance_loss_clip": 1.18802834,
      "balance_loss_mlp": 1.07449782,
      "epoch": 0.006613557793476627,
      "flos": 21105132445440.0,
      "grad_norm": 1.630666508662532,
      "language_loss": 0.8404845,
      "learning_rate": 3.0264145470332218e-06,
      "loss": 0.8673445,
      "num_input_tokens_seen": 2337820,
      "step": 110,
      "time_per_iteration": 2.616708993911743
    },
    {
      "auxiliary_loss_clip": 0.01540502,
      "auxiliary_loss_mlp": 0.01147396,
      "balance_loss_clip": 1.18204618,
      "balance_loss_mlp": 1.08125901,
      "epoch": 0.006673681046144597,
      "flos": 26030747543040.0,
      "grad_norm": 2.010077669083485,
      "language_loss": 0.82815057,
      "learning_rate": 3.032241303393073e-06,
      "loss": 0.85502958,
      "num_input_tokens_seen": 2358560,
      "step": 111,
      "time_per_iteration": 2.625277280807495
    },
    {
      "auxiliary_loss_clip": 0.01541746,
      "auxiliary_loss_mlp": 0.01132896,
      "balance_loss_clip": 1.18560529,
      "balance_loss_mlp": 1.06923842,
      "epoch": 0.006733804298812566,
      "flos": 23147767163520.0,
      "grad_norm": 2.3265765920509334,
      "language_loss": 0.93846434,
      "learning_rate": 3.0380158011446e-06,
      "loss": 0.96521074,
      "num_input_tokens_seen": 2379005,
      "step": 112,
      "time_per_iteration": 2.611826181411743
    },
    {
      "auxiliary_loss_clip": 0.01544588,
      "auxiliary_loss_mlp": 0.01139652,
      "balance_loss_clip": 1.18151045,
      "balance_loss_mlp": 1.07608914,
      "epoch": 0.006793927551480535,
      "flos": 11764444210560.0,
      "grad_norm": 2.191456257024638,
      "language_loss": 0.79307699,
      "learning_rate": 3.0437389693482466e-06,
      "loss": 0.81991935,
      "num_input_tokens_seen": 2395610,
      "step": 113,
      "time_per_iteration": 2.5578553676605225
    },
    {
      "auxiliary_loss_clip": 0.01533231,
      "auxiliary_loss_mlp": 0.01133007,
      "balance_loss_clip": 1.1781671,
      "balance_loss_mlp": 1.06748986,
      "epoch": 0.006854050804148504,
      "flos": 19171953446400.0,
      "grad_norm": 2.4098368230777236,
      "language_loss": 0.93183094,
      "learning_rate": 3.0494117125071475e-06,
      "loss": 0.95849335,
      "num_input_tokens_seen": 2415005,
      "step": 114,
      "time_per_iteration": 2.589203357696533
    },
    {
      "auxiliary_loss_clip": 0.01540303,
      "auxiliary_loss_mlp": 0.01136206,
      "balance_loss_clip": 1.17805719,
      "balance_loss_mlp": 1.07645822,
      "epoch": 0.006914174056816474,
      "flos": 21981891519360.0,
      "grad_norm": 2.2584966292907476,
      "language_loss": 0.94365168,
      "learning_rate": 3.055034911425055e-06,
      "loss": 0.97041684,
      "num_input_tokens_seen": 2433965,
      "step": 115,
      "time_per_iteration": 2.646357536315918
    },
    {
      "auxiliary_loss_clip": 0.01533715,
      "auxiliary_loss_mlp": 0.01117057,
      "balance_loss_clip": 1.1748389,
      "balance_loss_mlp": 1.05087209,
      "epoch": 0.006974297309484443,
      "flos": 16289152634880.0,
      "grad_norm": 2.70043417163262,
      "language_loss": 0.81703532,
      "learning_rate": 3.0606094240271244e-06,
      "loss": 0.84354299,
      "num_input_tokens_seen": 2451605,
      "step": 116,
      "time_per_iteration": 2.586735725402832
    },
    {
      "auxiliary_loss_clip": 0.01526877,
      "auxiliary_loss_mlp": 0.01127142,
      "balance_loss_clip": 1.17653263,
      "balance_loss_mlp": 1.06324625,
      "epoch": 0.007034420562152413,
      "flos": 26104005331200.0,
      "grad_norm": 3.307788139369655,
      "language_loss": 0.87972999,
      "learning_rate": 3.0661360861454656e-06,
      "loss": 0.90627015,
      "num_input_tokens_seen": 2472035,
      "step": 117,
      "time_per_iteration": 2.620471477508545
    },
    {
      "auxiliary_loss_clip": 0.01524723,
      "auxiliary_loss_mlp": 0.01145442,
      "balance_loss_clip": 1.17411852,
      "balance_loss_mlp": 1.08030558,
      "epoch": 0.007094543814820382,
      "flos": 14204609723520.0,
      "grad_norm": 2.9924321404075283,
      "language_loss": 0.84531951,
      "learning_rate": 3.071615712271274e-06,
      "loss": 0.87202114,
      "num_input_tokens_seen": 2489285,
      "step": 118,
      "time_per_iteration": 2.5893595218658447
    },
    {
      "auxiliary_loss_clip": 0.01534311,
      "auxiliary_loss_mlp": 0.01162676,
      "balance_loss_clip": 1.17496443,
      "balance_loss_mlp": 1.09849334,
      "epoch": 0.007154667067488351,
      "flos": 14976007228800.0,
      "grad_norm": 2.3342046067825684,
      "language_loss": 0.9922536,
      "learning_rate": 3.0770490962752172e-06,
      "loss": 1.01922345,
      "num_input_tokens_seen": 2506460,
      "step": 119,
      "time_per_iteration": 2.553607225418091
    },
    {
      "auxiliary_loss_clip": 0.01536343,
      "auxiliary_loss_mlp": 0.0112179,
      "balance_loss_clip": 1.17107284,
      "balance_loss_mlp": 1.05970621,
      "epoch": 0.00721479032015632,
      "flos": 20193288762240.0,
      "grad_norm": 15.184582650523994,
      "language_loss": 0.8922199,
      "learning_rate": 3.082437012097686e-06,
      "loss": 0.91880119,
      "num_input_tokens_seen": 2525565,
      "step": 120,
      "time_per_iteration": 2.594038963317871
    },
    {
      "auxiliary_loss_clip": 0.0152173,
      "auxiliary_loss_mlp": 0.01132875,
      "balance_loss_clip": 1.17209888,
      "balance_loss_mlp": 1.06893134,
      "epoch": 0.00727491357282429,
      "flos": 23147228459520.0,
      "grad_norm": 1.8714293252338285,
      "language_loss": 0.93179202,
      "learning_rate": 3.0877802144103967e-06,
      "loss": 0.95833802,
      "num_input_tokens_seen": 2546605,
      "step": 121,
      "time_per_iteration": 2.604733467102051
    },
    {
      "auxiliary_loss_clip": 0.01523365,
      "auxiliary_loss_mlp": 0.01147342,
      "balance_loss_clip": 1.17240238,
      "balance_loss_mlp": 1.08449531,
      "epoch": 0.007335036825492259,
      "flos": 15521669712000.0,
      "grad_norm": 3.459256696633899,
      "language_loss": 0.9007864,
      "learning_rate": 3.09307943925077e-06,
      "loss": 0.92749345,
      "num_input_tokens_seen": 2560730,
      "step": 122,
      "time_per_iteration": 2.574723958969116
    },
    {
      "auxiliary_loss_clip": 0.01518139,
      "auxiliary_loss_mlp": 0.01144373,
      "balance_loss_clip": 1.16636777,
      "balance_loss_mlp": 1.07761562,
      "epoch": 0.007395160078160229,
      "flos": 24243365848320.0,
      "grad_norm": 2.3584371957887527,
      "language_loss": 0.92536861,
      "learning_rate": 3.0983354046304154e-06,
      "loss": 0.9519937,
      "num_input_tokens_seen": 2579550,
      "step": 123,
      "time_per_iteration": 2.6269428730010986
    },
    {
      "auxiliary_loss_clip": 0.01516059,
      "auxiliary_loss_mlp": 0.01128002,
      "balance_loss_clip": 1.1623075,
      "balance_loss_mlp": 1.06648982,
      "epoch": 0.007455283330828198,
      "flos": 31759792099200.0,
      "grad_norm": 2.2815383880508238,
      "language_loss": 0.70950246,
      "learning_rate": 3.103548811118979e-06,
      "loss": 0.73594308,
      "num_input_tokens_seen": 2600390,
      "step": 124,
      "time_per_iteration": 2.6536858081817627
    },
    {
      "auxiliary_loss_clip": 0.01505263,
      "auxiliary_loss_mlp": 0.01123382,
      "balance_loss_clip": 1.16403329,
      "balance_loss_mlp": 1.0602963,
      "epoch": 0.007515406583496167,
      "flos": 26615157822720.0,
      "grad_norm": 2.209027792278896,
      "language_loss": 0.88152587,
      "learning_rate": 3.108720342404542e-06,
      "loss": 0.9078123,
      "num_input_tokens_seen": 2620770,
      "step": 125,
      "time_per_iteration": 2.638362407684326
    },
    {
      "auxiliary_loss_clip": 0.01518682,
      "auxiliary_loss_mlp": 0.01142877,
      "balance_loss_clip": 1.16336763,
      "balance_loss_mlp": 1.08041179,
      "epoch": 0.007575529836164136,
      "flos": 18223696350720.0,
      "grad_norm": 3.2666806178006214,
      "language_loss": 0.8218801,
      "learning_rate": 3.1138506658316945e-06,
      "loss": 0.84849572,
      "num_input_tokens_seen": 2639900,
      "step": 126,
      "time_per_iteration": 2.5847151279449463
    },
    {
      "auxiliary_loss_clip": 0.0151346,
      "auxiliary_loss_mlp": 0.01139835,
      "balance_loss_clip": 1.16303134,
      "balance_loss_mlp": 1.07870495,
      "epoch": 0.007635653088832106,
      "flos": 21580410228480.0,
      "grad_norm": 3.414985913792108,
      "language_loss": 0.67462319,
      "learning_rate": 3.1189404329183404e-06,
      "loss": 0.70115614,
      "num_input_tokens_seen": 2657450,
      "step": 127,
      "time_per_iteration": 2.6407558917999268
    },
    {
      "auxiliary_loss_clip": 0.01501589,
      "auxiliary_loss_mlp": 0.01131695,
      "balance_loss_clip": 1.16464043,
      "balance_loss_mlp": 1.06851387,
      "epoch": 0.007695776341500075,
      "flos": 25375054723200.0,
      "grad_norm": 4.270137040519248,
      "language_loss": 0.88051999,
      "learning_rate": 3.1239902798522317e-06,
      "loss": 0.90685284,
      "num_input_tokens_seen": 2678150,
      "step": 128,
      "time_per_iteration": 2.612355947494507
    },
    {
      "auxiliary_loss_clip": 0.01504377,
      "auxiliary_loss_mlp": 0.01141049,
      "balance_loss_clip": 1.16031742,
      "balance_loss_mlp": 1.0778681,
      "epoch": 0.007755899594168045,
      "flos": 22343906741760.0,
      "grad_norm": 1.8014598610728807,
      "language_loss": 0.84627002,
      "learning_rate": 3.129000827968184e-06,
      "loss": 0.87272429,
      "num_input_tokens_seen": 2698290,
      "step": 129,
      "time_per_iteration": 2.6156222820281982
    },
    {
      "auxiliary_loss_clip": 0.01497321,
      "auxiliary_loss_mlp": 0.01131018,
      "balance_loss_clip": 1.15940428,
      "balance_loss_mlp": 1.06778908,
      "epoch": 0.007816022846836013,
      "flos": 22638230784000.0,
      "grad_norm": 2.192384067395493,
      "language_loss": 0.97199118,
      "learning_rate": 3.133972684206866e-06,
      "loss": 0.99827456,
      "num_input_tokens_seen": 2717630,
      "step": 130,
      "time_per_iteration": 2.6140406131744385
    },
    {
      "auxiliary_loss_clip": 0.01492265,
      "auxiliary_loss_mlp": 0.01132502,
      "balance_loss_clip": 1.15676177,
      "balance_loss_mlp": 1.06884396,
      "epoch": 0.007876146099503984,
      "flos": 18182901479040.0,
      "grad_norm": 2.6004986070079252,
      "language_loss": 0.82402104,
      "learning_rate": 3.138906441556014e-06,
      "loss": 0.85026872,
      "num_input_tokens_seen": 2735835,
      "step": 131,
      "time_per_iteration": 2.5636849403381348
    },
    {
      "auxiliary_loss_clip": 0.01500218,
      "auxiliary_loss_mlp": 0.01127216,
      "balance_loss_clip": 1.15860033,
      "balance_loss_mlp": 1.06613278,
      "epoch": 0.007936269352171952,
      "flos": 27119486730240.0,
      "grad_norm": 3.1391445150548836,
      "language_loss": 0.82872903,
      "learning_rate": 3.143802679474861e-06,
      "loss": 0.85500336,
      "num_input_tokens_seen": 2756335,
      "step": 132,
      "time_per_iteration": 4.080858469009399
    },
    {
      "auxiliary_loss_clip": 0.01491852,
      "auxiliary_loss_mlp": 0.01126826,
      "balance_loss_clip": 1.15346324,
      "balance_loss_mlp": 1.06493282,
      "epoch": 0.007996392604839923,
      "flos": 19026335710080.0,
      "grad_norm": 2.430030618541236,
      "language_loss": 0.9506011,
      "learning_rate": 3.1486619643025565e-06,
      "loss": 0.97678792,
      "num_input_tokens_seen": 2775090,
      "step": 133,
      "time_per_iteration": 3.9750640392303467
    },
    {
      "auxiliary_loss_clip": 0.01487712,
      "auxiliary_loss_mlp": 0.01127585,
      "balance_loss_clip": 1.16184449,
      "balance_loss_mlp": 1.06793308,
      "epoch": 0.008056515857507891,
      "flos": 25484151306240.0,
      "grad_norm": 1.667690260557188,
      "language_loss": 0.73300362,
      "learning_rate": 3.153484849651286e-06,
      "loss": 0.75915658,
      "num_input_tokens_seen": 2795320,
      "step": 134,
      "time_per_iteration": 4.0368804931640625
    },
    {
      "auxiliary_loss_clip": 0.01483885,
      "auxiliary_loss_mlp": 0.01129553,
      "balance_loss_clip": 1.15032005,
      "balance_loss_mlp": 1.06489432,
      "epoch": 0.00811663911017586,
      "flos": 20557566541440.0,
      "grad_norm": 3.2583640340397686,
      "language_loss": 0.88490343,
      "learning_rate": 3.1582718767847806e-06,
      "loss": 0.9110378,
      "num_input_tokens_seen": 2812815,
      "step": 135,
      "time_per_iteration": 2.5706255435943604
    },
    {
      "auxiliary_loss_clip": 0.01487791,
      "auxiliary_loss_mlp": 0.01132749,
      "balance_loss_clip": 1.15467739,
      "balance_loss_mlp": 1.06737447,
      "epoch": 0.00817676236284383,
      "flos": 18799738761600.0,
      "grad_norm": 2.3872646448381247,
      "language_loss": 0.89151216,
      "learning_rate": 3.1630235749828485e-06,
      "loss": 0.91771758,
      "num_input_tokens_seen": 2830445,
      "step": 136,
      "time_per_iteration": 2.5756802558898926
    },
    {
      "auxiliary_loss_clip": 0.01483359,
      "auxiliary_loss_mlp": 0.01107188,
      "balance_loss_clip": 1.14921665,
      "balance_loss_mlp": 1.04748821,
      "epoch": 0.008236885615511799,
      "flos": 23873593288320.0,
      "grad_norm": 2.2425191343364554,
      "language_loss": 0.83894992,
      "learning_rate": 3.1677404618925676e-06,
      "loss": 0.86485541,
      "num_input_tokens_seen": 2846965,
      "step": 137,
      "time_per_iteration": 2.6100833415985107
    },
    {
      "auxiliary_loss_clip": 0.01480652,
      "auxiliary_loss_mlp": 0.01117771,
      "balance_loss_clip": 1.1488812,
      "balance_loss_mlp": 1.05783308,
      "epoch": 0.00829700886817977,
      "flos": 24643626076800.0,
      "grad_norm": 1.7736266978645918,
      "language_loss": 0.90182352,
      "learning_rate": 3.1724230438666953e-06,
      "loss": 0.92780781,
      "num_input_tokens_seen": 2867520,
      "step": 138,
      "time_per_iteration": 2.6025633811950684
    },
    {
      "auxiliary_loss_clip": 0.01469131,
      "auxiliary_loss_mlp": 0.01120798,
      "balance_loss_clip": 1.14758015,
      "balance_loss_mlp": 1.05647254,
      "epoch": 0.008357132120847738,
      "flos": 25262007644160.0,
      "grad_norm": 2.5326850374808063,
      "language_loss": 0.91454554,
      "learning_rate": 3.177071816289865e-06,
      "loss": 0.94044477,
      "num_input_tokens_seen": 2885675,
      "step": 139,
      "time_per_iteration": 2.5911431312561035
    },
    {
      "auxiliary_loss_clip": 0.01486186,
      "auxiliary_loss_mlp": 0.01122771,
      "balance_loss_clip": 1.15409458,
      "balance_loss_mlp": 1.06082976,
      "epoch": 0.008417255373515706,
      "flos": 27344898529920.0,
      "grad_norm": 2.2806446367544484,
      "language_loss": 0.85681903,
      "learning_rate": 3.181687263893095e-06,
      "loss": 0.88290858,
      "num_input_tokens_seen": 2905960,
      "step": 140,
      "time_per_iteration": 2.6238269805908203
    },
    {
      "auxiliary_loss_clip": 0.01472527,
      "auxiliary_loss_mlp": 0.01119801,
      "balance_loss_clip": 1.14867771,
      "balance_loss_mlp": 1.05890918,
      "epoch": 0.008477378626183677,
      "flos": 17639070589440.0,
      "grad_norm": 3.897387513384623,
      "language_loss": 0.84215617,
      "learning_rate": 3.186269861057098e-06,
      "loss": 0.86807942,
      "num_input_tokens_seen": 2922780,
      "step": 141,
      "time_per_iteration": 2.547086715698242
    },
    {
      "auxiliary_loss_clip": 0.01476949,
      "auxiliary_loss_mlp": 0.01134924,
      "balance_loss_clip": 1.14633572,
      "balance_loss_mlp": 1.07365096,
      "epoch": 0.008537501878851645,
      "flos": 13881342297600.0,
      "grad_norm": 2.201710229460226,
      "language_loss": 0.8140502,
      "learning_rate": 3.1908200721048745e-06,
      "loss": 0.84016895,
      "num_input_tokens_seen": 2938765,
      "step": 142,
      "time_per_iteration": 2.5310564041137695
    },
    {
      "auxiliary_loss_clip": 0.01394938,
      "auxiliary_loss_mlp": 0.01033514,
      "balance_loss_clip": 1.20332384,
      "balance_loss_mlp": 1.00261533,
      "epoch": 0.008597625131519616,
      "flos": 71248101281280.0,
      "grad_norm": 1.0481380678861736,
      "language_loss": 0.67008018,
      "learning_rate": 3.195338351584042e-06,
      "loss": 0.69436479,
      "num_input_tokens_seen": 3006665,
      "step": 143,
      "time_per_iteration": 3.2908971309661865
    },
    {
      "auxiliary_loss_clip": 0.01468626,
      "auxiliary_loss_mlp": 0.01123175,
      "balance_loss_clip": 1.14668345,
      "balance_loss_mlp": 1.06228304,
      "epoch": 0.008657748384187584,
      "flos": 17602836744960.0,
      "grad_norm": 2.1555867117648204,
      "language_loss": 0.84040892,
      "learning_rate": 3.1998251445393258e-06,
      "loss": 0.86632693,
      "num_input_tokens_seen": 3024335,
      "step": 144,
      "time_per_iteration": 2.5745999813079834
    },
    {
      "auxiliary_loss_clip": 0.01455917,
      "auxiliary_loss_mlp": 0.01112389,
      "balance_loss_clip": 1.14087963,
      "balance_loss_mlp": 1.04930329,
      "epoch": 0.008717871636855555,
      "flos": 19715317459200.0,
      "grad_norm": 1.8360984748164988,
      "language_loss": 0.88471884,
      "learning_rate": 3.204280886775619e-06,
      "loss": 0.91040194,
      "num_input_tokens_seen": 3043300,
      "step": 145,
      "time_per_iteration": 2.5579302310943604
    },
    {
      "auxiliary_loss_clip": 0.01471664,
      "auxiliary_loss_mlp": 0.0112909,
      "balance_loss_clip": 1.14207482,
      "balance_loss_mlp": 1.06619477,
      "epoch": 0.008777994889523523,
      "flos": 24717422568960.0,
      "grad_norm": 1.7737621531723198,
      "language_loss": 0.85963774,
      "learning_rate": 3.208706005112005e-06,
      "loss": 0.88564527,
      "num_input_tokens_seen": 3064610,
      "step": 146,
      "time_per_iteration": 2.5866341590881348
    },
    {
      "auxiliary_loss_clip": 0.01375054,
      "auxiliary_loss_mlp": 0.01030385,
      "balance_loss_clip": 1.18814135,
      "balance_loss_mlp": 1.00082064,
      "epoch": 0.008838118142191492,
      "flos": 70132067758080.0,
      "grad_norm": 0.8628216619847996,
      "language_loss": 0.60204238,
      "learning_rate": 3.213100917627104e-06,
      "loss": 0.62609679,
      "num_input_tokens_seen": 3130385,
      "step": 147,
      "time_per_iteration": 3.224196672439575
    },
    {
      "auxiliary_loss_clip": 0.01463024,
      "auxiliary_loss_mlp": 0.01122554,
      "balance_loss_clip": 1.14601707,
      "balance_loss_mlp": 1.06466651,
      "epoch": 0.008898241394859462,
      "flos": 20044797937920.0,
      "grad_norm": 2.90727007877629,
      "language_loss": 0.84645188,
      "learning_rate": 3.2174660338961135e-06,
      "loss": 0.87230766,
      "num_input_tokens_seen": 3149760,
      "step": 148,
      "time_per_iteration": 2.5968053340911865
    },
    {
      "auxiliary_loss_clip": 0.01468888,
      "auxiliary_loss_mlp": 0.01144426,
      "balance_loss_clip": 1.14840221,
      "balance_loss_mlp": 1.07991028,
      "epoch": 0.008958364647527431,
      "flos": 10743611685120.0,
      "grad_norm": 2.740979599299946,
      "language_loss": 0.88686371,
      "learning_rate": 3.2218017552198588e-06,
      "loss": 0.91299689,
      "num_input_tokens_seen": 3164500,
      "step": 149,
      "time_per_iteration": 2.5281078815460205
    },
    {
      "auxiliary_loss_clip": 0.01464736,
      "auxiliary_loss_mlp": 0.01109027,
      "balance_loss_clip": 1.14274573,
      "balance_loss_mlp": 1.05090094,
      "epoch": 0.009018487900195401,
      "flos": 29127467802240.0,
      "grad_norm": 2.0048823269355913,
      "language_loss": 0.93001103,
      "learning_rate": 3.226108474846181e-06,
      "loss": 0.95574868,
      "num_input_tokens_seen": 3182455,
      "step": 150,
      "time_per_iteration": 2.6309010982513428
    },
    {
      "auxiliary_loss_clip": 0.01453068,
      "auxiliary_loss_mlp": 0.01110994,
      "balance_loss_clip": 1.13838744,
      "balance_loss_mlp": 1.05467951,
      "epoch": 0.00907861115286337,
      "flos": 32963661354240.0,
      "grad_norm": 1.8804704290952083,
      "language_loss": 0.74305212,
      "learning_rate": 3.2303865781839817e-06,
      "loss": 0.76869273,
      "num_input_tokens_seen": 3203995,
      "step": 151,
      "time_per_iteration": 2.6726725101470947
    },
    {
      "auxiliary_loss_clip": 0.01463953,
      "auxiliary_loss_mlp": 0.01127498,
      "balance_loss_clip": 1.14192545,
      "balance_loss_mlp": 1.06899047,
      "epoch": 0.009138734405531338,
      "flos": 21762441377280.0,
      "grad_norm": 2.8661847016129176,
      "language_loss": 0.88202453,
      "learning_rate": 3.234636443010188e-06,
      "loss": 0.90793908,
      "num_input_tokens_seen": 3222575,
      "step": 152,
      "time_per_iteration": 2.586298704147339
    },
    {
      "auxiliary_loss_clip": 0.01464188,
      "auxiliary_loss_mlp": 0.01121655,
      "balance_loss_clip": 1.14814472,
      "balance_loss_mlp": 1.0626229,
      "epoch": 0.009198857658199309,
      "flos": 20842517134080.0,
      "grad_norm": 4.25413307712974,
      "language_loss": 0.84180242,
      "learning_rate": 3.238858439669943e-06,
      "loss": 0.86766088,
      "num_input_tokens_seen": 3240180,
      "step": 153,
      "time_per_iteration": 2.555903911590576
    },
    {
      "auxiliary_loss_clip": 0.01453985,
      "auxiliary_loss_mlp": 0.01136568,
      "balance_loss_clip": 1.139691,
      "balance_loss_mlp": 1.07600951,
      "epoch": 0.009258980910867277,
      "flos": 24827381078400.0,
      "grad_norm": 1.9241127881808737,
      "language_loss": 0.89898205,
      "learning_rate": 3.2430529312702712e-06,
      "loss": 0.92488766,
      "num_input_tokens_seen": 3259800,
      "step": 154,
      "time_per_iteration": 2.59694766998291
    },
    {
      "auxiliary_loss_clip": 0.01458693,
      "auxiliary_loss_mlp": 0.01157528,
      "balance_loss_clip": 1.14241755,
      "balance_loss_mlp": 1.09842443,
      "epoch": 0.009319104163535248,
      "flos": 28767786963840.0,
      "grad_norm": 2.016152984236956,
      "language_loss": 0.89719486,
      "learning_rate": 3.2472202738674737e-06,
      "loss": 0.92335707,
      "num_input_tokens_seen": 3280400,
      "step": 155,
      "time_per_iteration": 2.6048922538757324
    },
    {
      "auxiliary_loss_clip": 0.01462118,
      "auxiliary_loss_mlp": 0.011184,
      "balance_loss_clip": 1.14079165,
      "balance_loss_mlp": 1.06098843,
      "epoch": 0.009379227416203216,
      "flos": 16582004219520.0,
      "grad_norm": 2.352405376931058,
      "language_loss": 0.8663311,
      "learning_rate": 3.2513608166485063e-06,
      "loss": 0.89213622,
      "num_input_tokens_seen": 3297600,
      "step": 156,
      "time_per_iteration": 2.5507659912109375
    },
    {
      "auxiliary_loss_clip": 0.01459403,
      "auxiliary_loss_mlp": 0.01120798,
      "balance_loss_clip": 1.14356971,
      "balance_loss_mlp": 1.06271887,
      "epoch": 0.009439350668871187,
      "flos": 18329919845760.0,
      "grad_norm": 2.2212045492080152,
      "language_loss": 0.99220395,
      "learning_rate": 3.2554749021065498e-06,
      "loss": 1.01800597,
      "num_input_tokens_seen": 3313635,
      "step": 157,
      "time_per_iteration": 2.513223648071289
    },
    {
      "auxiliary_loss_clip": 0.01443777,
      "auxiliary_loss_mlp": 0.01144576,
      "balance_loss_clip": 1.13854277,
      "balance_loss_mlp": 1.08635402,
      "epoch": 0.009499473921539155,
      "flos": 24349912565760.0,
      "grad_norm": 2.0620233773336776,
      "language_loss": 0.88287854,
      "learning_rate": 3.2595628662110186e-06,
      "loss": 0.9087621,
      "num_input_tokens_seen": 3333735,
      "step": 158,
      "time_per_iteration": 2.622466802597046
    },
    {
      "auxiliary_loss_clip": 0.01450484,
      "auxiliary_loss_mlp": 0.01125572,
      "balance_loss_clip": 1.13807034,
      "balance_loss_mlp": 1.06668293,
      "epoch": 0.009559597174207124,
      "flos": 16399326625920.0,
      "grad_norm": 2.630174686943085,
      "language_loss": 0.86455482,
      "learning_rate": 3.2636250385721982e-06,
      "loss": 0.89031541,
      "num_input_tokens_seen": 3348800,
      "step": 159,
      "time_per_iteration": 2.5305891036987305
    },
    {
      "auxiliary_loss_clip": 0.01439689,
      "auxiliary_loss_mlp": 0.01130469,
      "balance_loss_clip": 1.13249254,
      "balance_loss_mlp": 1.07115078,
      "epoch": 0.009619720426875094,
      "flos": 22856890826880.0,
      "grad_norm": 2.1226552290314915,
      "language_loss": 0.86700493,
      "learning_rate": 3.2676617426007263e-06,
      "loss": 0.89270651,
      "num_input_tokens_seen": 3368595,
      "step": 160,
      "time_per_iteration": 2.589097261428833
    },
    {
      "auxiliary_loss_clip": 0.01446639,
      "auxiliary_loss_mlp": 0.01121242,
      "balance_loss_clip": 1.13815951,
      "balance_loss_mlp": 1.06647718,
      "epoch": 0.009679843679543063,
      "flos": 19135001329920.0,
      "grad_norm": 2.399768866187191,
      "language_loss": 0.91769397,
      "learning_rate": 3.2716732956621042e-06,
      "loss": 0.94337279,
      "num_input_tokens_seen": 3384975,
      "step": 161,
      "time_per_iteration": 2.5633223056793213
    },
    {
      "auxiliary_loss_clip": 0.01451841,
      "auxiliary_loss_mlp": 0.01112201,
      "balance_loss_clip": 1.13853979,
      "balance_loss_mlp": 1.05669701,
      "epoch": 0.009739966932211033,
      "flos": 20302995876480.0,
      "grad_norm": 1.7731282281010101,
      "language_loss": 0.91328567,
      "learning_rate": 3.2756600092264203e-06,
      "loss": 0.9389261,
      "num_input_tokens_seen": 3404755,
      "step": 162,
      "time_per_iteration": 2.596208095550537
    },
    {
      "auxiliary_loss_clip": 0.01334637,
      "auxiliary_loss_mlp": 0.01027357,
      "balance_loss_clip": 1.16269302,
      "balance_loss_mlp": 1.0044688,
      "epoch": 0.009800090184879002,
      "flos": 67034234177280.0,
      "grad_norm": 1.1799250159430006,
      "language_loss": 0.72456527,
      "learning_rate": 3.279622189013474e-06,
      "loss": 0.74818516,
      "num_input_tokens_seen": 3467210,
      "step": 163,
      "time_per_iteration": 3.121882915496826
    },
    {
      "auxiliary_loss_clip": 0.01437328,
      "auxiliary_loss_mlp": 0.01118105,
      "balance_loss_clip": 1.1356113,
      "balance_loss_mlp": 1.06255364,
      "epoch": 0.00986021343754697,
      "flos": 17164690646400.0,
      "grad_norm": 2.0775708426327135,
      "language_loss": 0.84366113,
      "learning_rate": 3.283560135133457e-06,
      "loss": 0.86921549,
      "num_input_tokens_seen": 3483220,
      "step": 164,
      "time_per_iteration": 2.5571401119232178
    },
    {
      "auxiliary_loss_clip": 0.01428237,
      "auxiliary_loss_mlp": 0.01103308,
      "balance_loss_clip": 1.1271925,
      "balance_loss_mlp": 1.04839993,
      "epoch": 0.00992033669021494,
      "flos": 17749424148480.0,
      "grad_norm": 2.209673035564369,
      "language_loss": 0.89136529,
      "learning_rate": 3.2874741422233565e-06,
      "loss": 0.91668069,
      "num_input_tokens_seen": 3501465,
      "step": 165,
      "time_per_iteration": 2.5336015224456787
    },
    {
      "auxiliary_loss_clip": 0.01431874,
      "auxiliary_loss_mlp": 0.01130153,
      "balance_loss_clip": 1.12813318,
      "balance_loss_mlp": 1.07040501,
      "epoch": 0.00998045994288291,
      "flos": 25297164080640.0,
      "grad_norm": 1.684703336109223,
      "language_loss": 0.79867178,
      "learning_rate": 3.2913644995792465e-06,
      "loss": 0.82429206,
      "num_input_tokens_seen": 3520480,
      "step": 166,
      "time_per_iteration": 2.6028335094451904
    },
    {
      "auxiliary_loss_clip": 0.01436563,
      "auxiliary_loss_mlp": 0.01124782,
      "balance_loss_clip": 1.13194895,
      "balance_loss_mlp": 1.06646466,
      "epoch": 0.01004058319555088,
      "flos": 32298954220800.0,
      "grad_norm": 2.2362976367249465,
      "language_loss": 0.91916299,
      "learning_rate": 3.2952314912845914e-06,
      "loss": 0.94477642,
      "num_input_tokens_seen": 3539570,
      "step": 167,
      "time_per_iteration": 2.677703380584717
    },
    {
      "auxiliary_loss_clip": 0.01427267,
      "auxiliary_loss_mlp": 0.01141527,
      "balance_loss_clip": 1.12919569,
      "balance_loss_mlp": 1.08573735,
      "epoch": 0.010100706448218848,
      "flos": 11319941404800.0,
      "grad_norm": 17.821889522094118,
      "language_loss": 0.90689933,
      "learning_rate": 3.299075396334735e-06,
      "loss": 0.93258727,
      "num_input_tokens_seen": 3555465,
      "step": 168,
      "time_per_iteration": 2.54180908203125
    },
    {
      "auxiliary_loss_clip": 0.014222,
      "auxiliary_loss_mlp": 0.01113226,
      "balance_loss_clip": 1.12524581,
      "balance_loss_mlp": 1.05605364,
      "epoch": 0.010160829700886819,
      "flos": 29719491765120.0,
      "grad_norm": 1.7414076639704423,
      "language_loss": 0.87183541,
      "learning_rate": 3.3028964887576868e-06,
      "loss": 0.89718962,
      "num_input_tokens_seen": 3578970,
      "step": 169,
      "time_per_iteration": 2.643301010131836
    },
    {
      "auxiliary_loss_clip": 0.01424114,
      "auxiliary_loss_mlp": 0.0111806,
      "balance_loss_clip": 1.12758684,
      "balance_loss_mlp": 1.06136453,
      "epoch": 0.010220952953554787,
      "flos": 20412343854720.0,
      "grad_norm": 1.849921094585435,
      "language_loss": 0.84461129,
      "learning_rate": 3.306695037731344e-06,
      "loss": 0.87003303,
      "num_input_tokens_seen": 3597275,
      "step": 170,
      "time_per_iteration": 2.5366506576538086
    },
    {
      "auxiliary_loss_clip": 0.01432615,
      "auxiliary_loss_mlp": 0.0113412,
      "balance_loss_clip": 1.12817299,
      "balance_loss_mlp": 1.07694769,
      "epoch": 0.010281076206222756,
      "flos": 31285124847360.0,
      "grad_norm": 1.954262740453898,
      "language_loss": 0.89880979,
      "learning_rate": 3.3104713076972827e-06,
      "loss": 0.9244771,
      "num_input_tokens_seen": 3618905,
      "step": 171,
      "time_per_iteration": 3.988121509552002
    },
    {
      "auxiliary_loss_clip": 0.01427192,
      "auxiliary_loss_mlp": 0.01113776,
      "balance_loss_clip": 1.13071418,
      "balance_loss_mlp": 1.05967844,
      "epoch": 0.010341199458890726,
      "flos": 21982286568960.0,
      "grad_norm": 1.9890784317586754,
      "language_loss": 0.88885248,
      "learning_rate": 3.314225558471224e-06,
      "loss": 0.91426218,
      "num_input_tokens_seen": 3639610,
      "step": 172,
      "time_per_iteration": 5.470140695571899
    },
    {
      "auxiliary_loss_clip": 0.01415364,
      "auxiliary_loss_mlp": 0.0112938,
      "balance_loss_clip": 1.12345064,
      "balance_loss_mlp": 1.07430518,
      "epoch": 0.010401322711558695,
      "flos": 30810529422720.0,
      "grad_norm": 2.4544757027320867,
      "language_loss": 0.81026721,
      "learning_rate": 3.317958045350308e-06,
      "loss": 0.83571458,
      "num_input_tokens_seen": 3664030,
      "step": 173,
      "time_per_iteration": 2.6381349563598633
    },
    {
      "auxiliary_loss_clip": 0.01427926,
      "auxiliary_loss_mlp": 0.01111802,
      "balance_loss_clip": 1.12783968,
      "balance_loss_mlp": 1.05908799,
      "epoch": 0.010461445964226665,
      "flos": 24715124098560.0,
      "grad_norm": 2.3130663912890372,
      "language_loss": 0.82624376,
      "learning_rate": 3.3216690192172596e-06,
      "loss": 0.85164112,
      "num_input_tokens_seen": 3683615,
      "step": 174,
      "time_per_iteration": 3.964766025543213
    },
    {
      "auxiliary_loss_clip": 0.01421404,
      "auxiliary_loss_mlp": 0.01130851,
      "balance_loss_clip": 1.12515438,
      "balance_loss_mlp": 1.07563281,
      "epoch": 0.010521569216894634,
      "flos": 27710361457920.0,
      "grad_norm": 4.53227548004005,
      "language_loss": 0.72889018,
      "learning_rate": 3.325358726641591e-06,
      "loss": 0.75441277,
      "num_input_tokens_seen": 3704540,
      "step": 175,
      "time_per_iteration": 2.6175849437713623
    },
    {
      "auxiliary_loss_clip": 0.01424875,
      "auxiliary_loss_mlp": 0.01134794,
      "balance_loss_clip": 1.12624717,
      "balance_loss_mlp": 1.077335,
      "epoch": 0.010581692469562603,
      "flos": 12458346122880.0,
      "grad_norm": 2.215496166423868,
      "language_loss": 0.98043299,
      "learning_rate": 3.329027409977902e-06,
      "loss": 1.00602973,
      "num_input_tokens_seen": 3721320,
      "step": 176,
      "time_per_iteration": 2.536527156829834
    },
    {
      "auxiliary_loss_clip": 0.01409507,
      "auxiliary_loss_mlp": 0.01136615,
      "balance_loss_clip": 1.12271774,
      "balance_loss_mlp": 1.08325744,
      "epoch": 0.010641815722230573,
      "flos": 19427601519360.0,
      "grad_norm": 2.3793021769723945,
      "language_loss": 0.76715404,
      "learning_rate": 3.3326753074614087e-06,
      "loss": 0.79261529,
      "num_input_tokens_seen": 3739385,
      "step": 177,
      "time_per_iteration": 2.5875117778778076
    },
    {
      "auxiliary_loss_clip": 0.01420815,
      "auxiliary_loss_mlp": 0.01106223,
      "balance_loss_clip": 1.12172461,
      "balance_loss_mlp": 1.05150616,
      "epoch": 0.010701938974898541,
      "flos": 18332577452160.0,
      "grad_norm": 2.5208344481808145,
      "language_loss": 0.7673099,
      "learning_rate": 3.3363026533007716e-06,
      "loss": 0.79258031,
      "num_input_tokens_seen": 3756360,
      "step": 178,
      "time_per_iteration": 2.540008783340454
    },
    {
      "auxiliary_loss_clip": 0.01429671,
      "auxiliary_loss_mlp": 0.01113634,
      "balance_loss_clip": 1.1279,
      "balance_loss_mlp": 1.05660415,
      "epoch": 0.010762062227566512,
      "flos": 19203985399680.0,
      "grad_norm": 5.625168480636054,
      "language_loss": 0.84153199,
      "learning_rate": 3.3399096777683303e-06,
      "loss": 0.866965,
      "num_input_tokens_seen": 3773930,
      "step": 179,
      "time_per_iteration": 2.5780298709869385
    },
    {
      "auxiliary_loss_clip": 0.0141849,
      "auxiliary_loss_mlp": 0.01115483,
      "balance_loss_clip": 1.12092733,
      "balance_loss_mlp": 1.05883479,
      "epoch": 0.01082218548023448,
      "flos": 31425427370880.0,
      "grad_norm": 2.3201176097769336,
      "language_loss": 0.83794487,
      "learning_rate": 3.3434966072878213e-06,
      "loss": 0.86328459,
      "num_input_tokens_seen": 3793630,
      "step": 180,
      "time_per_iteration": 2.653860092163086
    },
    {
      "auxiliary_loss_clip": 0.01420244,
      "auxiliary_loss_mlp": 0.01126542,
      "balance_loss_clip": 1.12370443,
      "balance_loss_mlp": 1.07032311,
      "epoch": 0.01088230873290245,
      "flos": 25046436170880.0,
      "grad_norm": 3.389870122364849,
      "language_loss": 0.77760291,
      "learning_rate": 3.3470636645196674e-06,
      "loss": 0.80307078,
      "num_input_tokens_seen": 3813610,
      "step": 181,
      "time_per_iteration": 2.602083206176758
    },
    {
      "auxiliary_loss_clip": 0.01414328,
      "auxiliary_loss_mlp": 0.01132399,
      "balance_loss_clip": 1.11986363,
      "balance_loss_mlp": 1.07656169,
      "epoch": 0.01094243198557042,
      "flos": 22893411980160.0,
      "grad_norm": 3.2463410828251447,
      "language_loss": 0.76189518,
      "learning_rate": 3.3506110684439156e-06,
      "loss": 0.78736246,
      "num_input_tokens_seen": 3831390,
      "step": 182,
      "time_per_iteration": 2.5717039108276367
    },
    {
      "auxiliary_loss_clip": 0.01413239,
      "auxiliary_loss_mlp": 0.0112979,
      "balance_loss_clip": 1.11999297,
      "balance_loss_mlp": 1.07323682,
      "epoch": 0.011002555238238388,
      "flos": 17165049782400.0,
      "grad_norm": 2.063151557577861,
      "language_loss": 0.87629259,
      "learning_rate": 3.3541390344409054e-06,
      "loss": 0.90172291,
      "num_input_tokens_seen": 3849705,
      "step": 183,
      "time_per_iteration": 2.6329638957977295
    },
    {
      "auxiliary_loss_clip": 0.01416452,
      "auxiliary_loss_mlp": 0.01115786,
      "balance_loss_clip": 1.12317443,
      "balance_loss_mlp": 1.0637629,
      "epoch": 0.011062678490906358,
      "flos": 22310150935680.0,
      "grad_norm": 2.3173840021027616,
      "language_loss": 0.86574775,
      "learning_rate": 3.357647774369736e-06,
      "loss": 0.89107013,
      "num_input_tokens_seen": 3869230,
      "step": 184,
      "time_per_iteration": 2.6276638507843018
    },
    {
      "auxiliary_loss_clip": 0.01413436,
      "auxiliary_loss_mlp": 0.01111929,
      "balance_loss_clip": 1.12426162,
      "balance_loss_mlp": 1.0555191,
      "epoch": 0.011122801743574327,
      "flos": 24388373053440.0,
      "grad_norm": 1.7694754831112893,
      "language_loss": 0.8363868,
      "learning_rate": 3.3611374966446085e-06,
      "loss": 0.86164051,
      "num_input_tokens_seen": 3889735,
      "step": 185,
      "time_per_iteration": 2.5675837993621826
    },
    {
      "auxiliary_loss_clip": 0.01421919,
      "auxiliary_loss_mlp": 0.01112162,
      "balance_loss_clip": 1.12186217,
      "balance_loss_mlp": 1.05322468,
      "epoch": 0.011182924996242297,
      "flos": 18150258994560.0,
      "grad_norm": 8.656498439157938,
      "language_loss": 0.71250904,
      "learning_rate": 3.3646084063091142e-06,
      "loss": 0.73784983,
      "num_input_tokens_seen": 3908855,
      "step": 186,
      "time_per_iteration": 2.55117130279541
    },
    {
      "auxiliary_loss_clip": 0.01417885,
      "auxiliary_loss_mlp": 0.01114629,
      "balance_loss_clip": 1.12164783,
      "balance_loss_mlp": 1.06229615,
      "epoch": 0.011243048248910266,
      "flos": 15486800584320.0,
      "grad_norm": 2.20026710514897,
      "language_loss": 1.02145648,
      "learning_rate": 3.3680607051085194e-06,
      "loss": 1.04678166,
      "num_input_tokens_seen": 3923865,
      "step": 187,
      "time_per_iteration": 2.4884941577911377
    },
    {
      "auxiliary_loss_clip": 0.01408593,
      "auxiliary_loss_mlp": 0.0111436,
      "balance_loss_clip": 1.12207508,
      "balance_loss_mlp": 1.05795002,
      "epoch": 0.011303171501578235,
      "flos": 40916868986880.0,
      "grad_norm": 1.7090552237737895,
      "language_loss": 0.75063562,
      "learning_rate": 3.371494591560139e-06,
      "loss": 0.77586514,
      "num_input_tokens_seen": 3946870,
      "step": 188,
      "time_per_iteration": 2.7186062335968018
    },
    {
      "auxiliary_loss_clip": 0.01309838,
      "auxiliary_loss_mlp": 0.01021905,
      "balance_loss_clip": 1.1507796,
      "balance_loss_mlp": 1.00168729,
      "epoch": 0.011363294754246205,
      "flos": 66302697790080.0,
      "grad_norm": 0.7554724394987258,
      "language_loss": 0.56205845,
      "learning_rate": 3.3749102610218297e-06,
      "loss": 0.58537591,
      "num_input_tokens_seen": 4010005,
      "step": 189,
      "time_per_iteration": 3.196989059448242
    },
    {
      "auxiliary_loss_clip": 0.01405679,
      "auxiliary_loss_mlp": 0.01126205,
      "balance_loss_clip": 1.1169889,
      "balance_loss_mlp": 1.06998563,
      "epoch": 0.011423418006914174,
      "flos": 24900279730560.0,
      "grad_norm": 6.005254765255265,
      "language_loss": 0.94985455,
      "learning_rate": 3.3783079057586833e-06,
      "loss": 0.97517335,
      "num_input_tokens_seen": 4029035,
      "step": 190,
      "time_per_iteration": 2.6015756130218506
    },
    {
      "auxiliary_loss_clip": 0.01406775,
      "auxiliary_loss_mlp": 0.0110634,
      "balance_loss_clip": 1.11924791,
      "balance_loss_mlp": 1.05264854,
      "epoch": 0.011483541259582144,
      "flos": 19791879298560.0,
      "grad_norm": 2.7197161783663604,
      "language_loss": 0.84206676,
      "learning_rate": 3.3816877150079665e-06,
      "loss": 0.86719787,
      "num_input_tokens_seen": 4046995,
      "step": 191,
      "time_per_iteration": 2.5323266983032227
    },
    {
      "auxiliary_loss_clip": 0.01404397,
      "auxiliary_loss_mlp": 0.01125127,
      "balance_loss_clip": 1.11493492,
      "balance_loss_mlp": 1.0720073,
      "epoch": 0.011543664512250112,
      "flos": 26176939896960.0,
      "grad_norm": 2.201907053112553,
      "language_loss": 0.91682249,
      "learning_rate": 3.385049875042367e-06,
      "loss": 0.94211769,
      "num_input_tokens_seen": 4065865,
      "step": 192,
      "time_per_iteration": 2.597071647644043
    },
    {
      "auxiliary_loss_clip": 0.01400303,
      "auxiliary_loss_mlp": 0.01119157,
      "balance_loss_clip": 1.1158154,
      "balance_loss_mlp": 1.0609349,
      "epoch": 0.011603787764918083,
      "flos": 23768985905280.0,
      "grad_norm": 2.1300935308954405,
      "language_loss": 0.86971748,
      "learning_rate": 3.3883945692315938e-06,
      "loss": 0.89491206,
      "num_input_tokens_seen": 4085305,
      "step": 193,
      "time_per_iteration": 2.5387165546417236
    },
    {
      "auxiliary_loss_clip": 0.01403924,
      "auxiliary_loss_mlp": 0.01097898,
      "balance_loss_clip": 1.11476374,
      "balance_loss_mlp": 1.04558897,
      "epoch": 0.011663911017586051,
      "flos": 25954688494080.0,
      "grad_norm": 2.2112664099146864,
      "language_loss": 0.92197573,
      "learning_rate": 3.3917219781023906e-06,
      "loss": 0.94699395,
      "num_input_tokens_seen": 4105185,
      "step": 194,
      "time_per_iteration": 2.621145486831665
    },
    {
      "auxiliary_loss_clip": 0.01407568,
      "auxiliary_loss_mlp": 0.01108546,
      "balance_loss_clip": 1.11843884,
      "balance_loss_mlp": 1.0549016,
      "epoch": 0.01172403427025402,
      "flos": 17895149625600.0,
      "grad_norm": 2.9085011049502643,
      "language_loss": 0.89838111,
      "learning_rate": 3.3950322793970014e-06,
      "loss": 0.9235422,
      "num_input_tokens_seen": 4123160,
      "step": 195,
      "time_per_iteration": 2.6105167865753174
    },
    {
      "auxiliary_loss_clip": 0.0140414,
      "auxiliary_loss_mlp": 0.01122494,
      "balance_loss_clip": 1.11828589,
      "balance_loss_mlp": 1.06665659,
      "epoch": 0.01178415752292199,
      "flos": 17894539094400.0,
      "grad_norm": 3.818867041584348,
      "language_loss": 0.85752439,
      "learning_rate": 3.3983256481301445e-06,
      "loss": 0.88279068,
      "num_input_tokens_seen": 4140425,
      "step": 196,
      "time_per_iteration": 2.5276665687561035
    },
    {
      "auxiliary_loss_clip": 0.01398811,
      "auxiliary_loss_mlp": 0.01108084,
      "balance_loss_clip": 1.11376595,
      "balance_loss_mlp": 1.05334306,
      "epoch": 0.011844280775589959,
      "flos": 22893555634560.0,
      "grad_norm": 4.012936769829807,
      "language_loss": 0.93208826,
      "learning_rate": 3.4016022566445335e-06,
      "loss": 0.95715725,
      "num_input_tokens_seen": 4159555,
      "step": 197,
      "time_per_iteration": 2.5361878871917725
    },
    {
      "auxiliary_loss_clip": 0.01397215,
      "auxiliary_loss_mlp": 0.01112574,
      "balance_loss_clip": 1.11556768,
      "balance_loss_mlp": 1.05873883,
      "epoch": 0.01190440402825793,
      "flos": 26980333441920.0,
      "grad_norm": 2.102881369770901,
      "language_loss": 0.79127729,
      "learning_rate": 3.4048622746649966e-06,
      "loss": 0.81637526,
      "num_input_tokens_seen": 4180480,
      "step": 198,
      "time_per_iteration": 2.634080648422241
    },
    {
      "auxiliary_loss_clip": 0.01395536,
      "auxiliary_loss_mlp": 0.0112367,
      "balance_loss_clip": 1.11708331,
      "balance_loss_mlp": 1.07031214,
      "epoch": 0.011964527280925898,
      "flos": 20521584092160.0,
      "grad_norm": 1.822930891076439,
      "language_loss": 0.88180757,
      "learning_rate": 3.4081058693512278e-06,
      "loss": 0.90699971,
      "num_input_tokens_seen": 4198835,
      "step": 199,
      "time_per_iteration": 2.5609943866729736
    },
    {
      "auxiliary_loss_clip": 0.01406401,
      "auxiliary_loss_mlp": 0.01127057,
      "balance_loss_clip": 1.12024951,
      "balance_loss_mlp": 1.06916928,
      "epoch": 0.012024650533593867,
      "flos": 27745984771200.0,
      "grad_norm": 1.8984927758396204,
      "language_loss": 0.81235689,
      "learning_rate": 3.411333205349222e-06,
      "loss": 0.83769149,
      "num_input_tokens_seen": 4219335,
      "step": 200,
      "time_per_iteration": 2.5740325450897217
    },
    {
      "auxiliary_loss_clip": 0.01404394,
      "auxiliary_loss_mlp": 0.01104464,
      "balance_loss_clip": 1.11645484,
      "balance_loss_mlp": 1.0486263,
      "epoch": 0.012084773786261837,
      "flos": 10452017076480.0,
      "grad_norm": 2.468785306108615,
      "language_loss": 0.87407899,
      "learning_rate": 3.4145444448414217e-06,
      "loss": 0.89916754,
      "num_input_tokens_seen": 4236940,
      "step": 201,
      "time_per_iteration": 2.507051706314087
    },
    {
      "auxiliary_loss_clip": 0.01402561,
      "auxiliary_loss_mlp": 0.01115408,
      "balance_loss_clip": 1.11822748,
      "balance_loss_mlp": 1.05897415,
      "epoch": 0.012144897038929806,
      "flos": 23105751229440.0,
      "grad_norm": 1.6606607098224535,
      "language_loss": 0.84100997,
      "learning_rate": 3.4177397475956223e-06,
      "loss": 0.86618966,
      "num_input_tokens_seen": 4256755,
      "step": 202,
      "time_per_iteration": 2.555718421936035
    },
    {
      "auxiliary_loss_clip": 0.01390737,
      "auxiliary_loss_mlp": 0.01109407,
      "balance_loss_clip": 1.10995722,
      "balance_loss_mlp": 1.05538154,
      "epoch": 0.012205020291597776,
      "flos": 21033203460480.0,
      "grad_norm": 1.7696686502964045,
      "language_loss": 0.9015373,
      "learning_rate": 3.4209192710126685e-06,
      "loss": 0.92653871,
      "num_input_tokens_seen": 4276505,
      "step": 203,
      "time_per_iteration": 2.53181529045105
    },
    {
      "auxiliary_loss_clip": 0.01283381,
      "auxiliary_loss_mlp": 0.0102524,
      "balance_loss_clip": 1.13304305,
      "balance_loss_mlp": 1.00664377,
      "epoch": 0.012265143544265745,
      "flos": 68447785075200.0,
      "grad_norm": 1.014308913471958,
      "language_loss": 0.61245143,
      "learning_rate": 3.4240831701729837e-06,
      "loss": 0.63553762,
      "num_input_tokens_seen": 4330965,
      "step": 204,
      "time_per_iteration": 3.04207444190979
    },
    {
      "auxiliary_loss_clip": 0.01399078,
      "auxiliary_loss_mlp": 0.01112439,
      "balance_loss_clip": 1.1131146,
      "balance_loss_mlp": 1.05912888,
      "epoch": 0.012325266796933715,
      "flos": 17019252478080.0,
      "grad_norm": 3.06258456984693,
      "language_loss": 0.91636145,
      "learning_rate": 3.4272315978819516e-06,
      "loss": 0.94147664,
      "num_input_tokens_seen": 4348200,
      "step": 205,
      "time_per_iteration": 2.5142509937286377
    },
    {
      "auxiliary_loss_clip": 0.01407598,
      "auxiliary_loss_mlp": 0.01123055,
      "balance_loss_clip": 1.11846578,
      "balance_loss_mlp": 1.06788456,
      "epoch": 0.012385390049601683,
      "flos": 20190056538240.0,
      "grad_norm": 2.185024001407723,
      "language_loss": 0.89197052,
      "learning_rate": 3.4303647047142043e-06,
      "loss": 0.9172771,
      "num_input_tokens_seen": 4365460,
      "step": 206,
      "time_per_iteration": 2.536412000656128
    },
    {
      "auxiliary_loss_clip": 0.01397336,
      "auxiliary_loss_mlp": 0.01104995,
      "balance_loss_clip": 1.11256635,
      "balance_loss_mlp": 1.05187559,
      "epoch": 0.012445513302269652,
      "flos": 16253134272000.0,
      "grad_norm": 2.021868357623696,
      "language_loss": 0.95616609,
      "learning_rate": 3.43348263905683e-06,
      "loss": 0.98118943,
      "num_input_tokens_seen": 4383650,
      "step": 207,
      "time_per_iteration": 2.5420873165130615
    },
    {
      "auxiliary_loss_clip": 0.01397126,
      "auxiliary_loss_mlp": 0.0111878,
      "balance_loss_clip": 1.11766446,
      "balance_loss_mlp": 1.06556463,
      "epoch": 0.012505636554937622,
      "flos": 23769380954880.0,
      "grad_norm": 1.8848398223328486,
      "language_loss": 0.76012576,
      "learning_rate": 3.436585547151547e-06,
      "loss": 0.78528482,
      "num_input_tokens_seen": 4403765,
      "step": 208,
      "time_per_iteration": 2.558464765548706
    },
    {
      "auxiliary_loss_clip": 0.01385122,
      "auxiliary_loss_mlp": 0.01113114,
      "balance_loss_clip": 1.11152649,
      "balance_loss_mlp": 1.05994642,
      "epoch": 0.012565759807605591,
      "flos": 30591546157440.0,
      "grad_norm": 2.3871136578092074,
      "language_loss": 0.98527992,
      "learning_rate": 3.4396735731358586e-06,
      "loss": 1.01026237,
      "num_input_tokens_seen": 4421935,
      "step": 209,
      "time_per_iteration": 2.5995798110961914
    },
    {
      "auxiliary_loss_clip": 0.0139087,
      "auxiliary_loss_mlp": 0.01117035,
      "balance_loss_clip": 1.11314762,
      "balance_loss_mlp": 1.06286621,
      "epoch": 0.012625883060273561,
      "flos": 40113511355520.0,
      "grad_norm": 2.6439605937013115,
      "language_loss": 0.85144472,
      "learning_rate": 3.4427468590832302e-06,
      "loss": 0.87652373,
      "num_input_tokens_seen": 4441470,
      "step": 210,
      "time_per_iteration": 2.6678504943847656
    },
    {
      "auxiliary_loss_clip": 0.01388797,
      "auxiliary_loss_mlp": 0.01117875,
      "balance_loss_clip": 1.11192298,
      "balance_loss_mlp": 1.06723487,
      "epoch": 0.01268600631294153,
      "flos": 27089178629760.0,
      "grad_norm": 2.2254476475348524,
      "language_loss": 0.97279471,
      "learning_rate": 3.445805545042314e-06,
      "loss": 0.9978615,
      "num_input_tokens_seen": 4459950,
      "step": 211,
      "time_per_iteration": 5.365017890930176
    },
    {
      "auxiliary_loss_clip": 0.01397705,
      "auxiliary_loss_mlp": 0.0112021,
      "balance_loss_clip": 1.11718869,
      "balance_loss_mlp": 1.0659461,
      "epoch": 0.012746129565609499,
      "flos": 16982767238400.0,
      "grad_norm": 2.134535624308585,
      "language_loss": 0.95090449,
      "learning_rate": 3.448849769075239e-06,
      "loss": 0.97608364,
      "num_input_tokens_seen": 4478390,
      "step": 212,
      "time_per_iteration": 3.974487066268921
    },
    {
      "auxiliary_loss_clip": 0.01385195,
      "auxiliary_loss_mlp": 0.01121101,
      "balance_loss_clip": 1.11384463,
      "balance_loss_mlp": 1.06798172,
      "epoch": 0.012806252818277469,
      "flos": 46533476995200.0,
      "grad_norm": 1.801155671046002,
      "language_loss": 0.7605617,
      "learning_rate": 3.4518796672950093e-06,
      "loss": 0.78562462,
      "num_input_tokens_seen": 4501665,
      "step": 213,
      "time_per_iteration": 4.131496906280518
    },
    {
      "auxiliary_loss_clip": 0.01389486,
      "auxiliary_loss_mlp": 0.01107454,
      "balance_loss_clip": 1.11245823,
      "balance_loss_mlp": 1.0553118,
      "epoch": 0.012866376070945438,
      "flos": 14388616120320.0,
      "grad_norm": 2.3704250774061415,
      "language_loss": 0.86484456,
      "learning_rate": 3.4548953739020187e-06,
      "loss": 0.8898139,
      "num_input_tokens_seen": 4519055,
      "step": 214,
      "time_per_iteration": 2.504061460494995
    },
    {
      "auxiliary_loss_clip": 0.01388202,
      "auxiliary_loss_mlp": 0.01130033,
      "balance_loss_clip": 1.11825395,
      "balance_loss_mlp": 1.07512581,
      "epoch": 0.012926499323613408,
      "flos": 26140813793280.0,
      "grad_norm": 2.1660884496381674,
      "language_loss": 0.7752043,
      "learning_rate": 3.4578970212197196e-06,
      "loss": 0.80038661,
      "num_input_tokens_seen": 4540870,
      "step": 215,
      "time_per_iteration": 2.600130081176758
    },
    {
      "auxiliary_loss_clip": 0.01395256,
      "auxiliary_loss_mlp": 0.01112857,
      "balance_loss_clip": 1.11571765,
      "balance_loss_mlp": 1.06069148,
      "epoch": 0.012986622576281377,
      "flos": 30117202128000.0,
      "grad_norm": 2.174483311634191,
      "language_loss": 0.9040401,
      "learning_rate": 3.460884739729461e-06,
      "loss": 0.92912126,
      "num_input_tokens_seen": 4560395,
      "step": 216,
      "time_per_iteration": 2.5753557682037354
    },
    {
      "auxiliary_loss_clip": 0.01387074,
      "auxiliary_loss_mlp": 0.01107028,
      "balance_loss_clip": 1.10984027,
      "balance_loss_mlp": 1.0543375,
      "epoch": 0.013046745828949347,
      "flos": 13954025468160.0,
      "grad_norm": 3.3866784082882018,
      "language_loss": 0.93752033,
      "learning_rate": 3.463858658104523e-06,
      "loss": 0.96246135,
      "num_input_tokens_seen": 4575785,
      "step": 217,
      "time_per_iteration": 2.508173942565918
    },
    {
      "auxiliary_loss_clip": 0.01382646,
      "auxiliary_loss_mlp": 0.01108176,
      "balance_loss_clip": 1.10974121,
      "balance_loss_mlp": 1.05338728,
      "epoch": 0.013106869081617315,
      "flos": 17347835116800.0,
      "grad_norm": 1.9223427116840066,
      "language_loss": 0.93535614,
      "learning_rate": 3.4668189032433696e-06,
      "loss": 0.96026433,
      "num_input_tokens_seen": 4594985,
      "step": 218,
      "time_per_iteration": 2.5009257793426514
    },
    {
      "auxiliary_loss_clip": 0.01376981,
      "auxiliary_loss_mlp": 0.01106107,
      "balance_loss_clip": 1.10825384,
      "balance_loss_mlp": 1.0551095,
      "epoch": 0.013166992334285284,
      "flos": 25884914325120.0,
      "grad_norm": 1.8694227413307443,
      "language_loss": 0.86191523,
      "learning_rate": 3.46976560030214e-06,
      "loss": 0.88674611,
      "num_input_tokens_seen": 4616125,
      "step": 219,
      "time_per_iteration": 2.5708844661712646
    },
    {
      "auxiliary_loss_clip": 0.01382943,
      "auxiliary_loss_mlp": 0.01103565,
      "balance_loss_clip": 1.11130381,
      "balance_loss_mlp": 1.05223393,
      "epoch": 0.013227115586953254,
      "flos": 31175956437120.0,
      "grad_norm": 1.7734535489206427,
      "language_loss": 0.87669885,
      "learning_rate": 3.4726988727263976e-06,
      "loss": 0.901564,
      "num_input_tokens_seen": 4637795,
      "step": 220,
      "time_per_iteration": 2.6110036373138428
    },
    {
      "auxiliary_loss_clip": 0.01375176,
      "auxiliary_loss_mlp": 0.01105941,
      "balance_loss_clip": 1.1068542,
      "balance_loss_mlp": 1.05859137,
      "epoch": 0.013287238839621223,
      "flos": 20409470766720.0,
      "grad_norm": 2.37770432935208,
      "language_loss": 0.86565876,
      "learning_rate": 3.475618842282164e-06,
      "loss": 0.89046991,
      "num_input_tokens_seen": 4656835,
      "step": 221,
      "time_per_iteration": 2.5673229694366455
    },
    {
      "auxiliary_loss_clip": 0.01379184,
      "auxiliary_loss_mlp": 0.01115692,
      "balance_loss_clip": 1.10567832,
      "balance_loss_mlp": 1.06335878,
      "epoch": 0.013347362092289193,
      "flos": 14137134024960.0,
      "grad_norm": 2.096535039758878,
      "language_loss": 0.92238522,
      "learning_rate": 3.4785256290862486e-06,
      "loss": 0.94733399,
      "num_input_tokens_seen": 4673015,
      "step": 222,
      "time_per_iteration": 2.4693238735198975
    },
    {
      "auxiliary_loss_clip": 0.01376244,
      "auxiliary_loss_mlp": 0.01107215,
      "balance_loss_clip": 1.108006,
      "balance_loss_mlp": 1.051759,
      "epoch": 0.013407485344957162,
      "flos": 21797705554560.0,
      "grad_norm": 2.301863774865024,
      "language_loss": 0.95768869,
      "learning_rate": 3.481419351635897e-06,
      "loss": 0.98252332,
      "num_input_tokens_seen": 4692355,
      "step": 223,
      "time_per_iteration": 2.525021553039551
    },
    {
      "auxiliary_loss_clip": 0.01377515,
      "auxiliary_loss_mlp": 0.01107445,
      "balance_loss_clip": 1.10881639,
      "balance_loss_mlp": 1.05639923,
      "epoch": 0.013467608597625132,
      "flos": 18621622195200.0,
      "grad_norm": 4.021672774997323,
      "language_loss": 0.88157892,
      "learning_rate": 3.484300126837776e-06,
      "loss": 0.90642858,
      "num_input_tokens_seen": 4710080,
      "step": 224,
      "time_per_iteration": 2.4916093349456787
    },
    {
      "auxiliary_loss_clip": 0.01377218,
      "auxiliary_loss_mlp": 0.0110686,
      "balance_loss_clip": 1.10765481,
      "balance_loss_mlp": 1.05118883,
      "epoch": 0.013527731850293101,
      "flos": 18552314903040.0,
      "grad_norm": 1.9915145769366958,
      "language_loss": 0.89413893,
      "learning_rate": 3.487168070036317e-06,
      "loss": 0.9189797,
      "num_input_tokens_seen": 4728980,
      "step": 225,
      "time_per_iteration": 2.5031356811523438
    },
    {
      "auxiliary_loss_clip": 0.01373314,
      "auxiliary_loss_mlp": 0.0112267,
      "balance_loss_clip": 1.10717499,
      "balance_loss_mlp": 1.06921697,
      "epoch": 0.01358785510296107,
      "flos": 19165381257600.0,
      "grad_norm": 3.1512186588840936,
      "language_loss": 0.99168986,
      "learning_rate": 3.4900232950414224e-06,
      "loss": 1.01664972,
      "num_input_tokens_seen": 4747020,
      "step": 226,
      "time_per_iteration": 2.5122225284576416
    },
    {
      "auxiliary_loss_clip": 0.01379799,
      "auxiliary_loss_mlp": 0.0111529,
      "balance_loss_clip": 1.10999382,
      "balance_loss_mlp": 1.06038165,
      "epoch": 0.01364797835562904,
      "flos": 23329941966720.0,
      "grad_norm": 2.1916836262515784,
      "language_loss": 0.91006362,
      "learning_rate": 3.4928659141555727e-06,
      "loss": 0.93501449,
      "num_input_tokens_seen": 4765000,
      "step": 227,
      "time_per_iteration": 2.5267276763916016
    },
    {
      "auxiliary_loss_clip": 0.01262221,
      "auxiliary_loss_mlp": 0.01031758,
      "balance_loss_clip": 1.11911011,
      "balance_loss_mlp": 1.01373386,
      "epoch": 0.013708101608297009,
      "flos": 70993746097920.0,
      "grad_norm": 0.9413837932204581,
      "language_loss": 0.57647407,
      "learning_rate": 3.4956960382003234e-06,
      "loss": 0.59941387,
      "num_input_tokens_seen": 4833210,
      "step": 228,
      "time_per_iteration": 3.2246227264404297
    },
    {
      "auxiliary_loss_clip": 0.01368924,
      "auxiliary_loss_mlp": 0.01114144,
      "balance_loss_clip": 1.10611475,
      "balance_loss_mlp": 1.06417131,
      "epoch": 0.013768224860964979,
      "flos": 16325170997760.0,
      "grad_norm": 2.448532707591361,
      "language_loss": 0.87790722,
      "learning_rate": 3.4985137765422354e-06,
      "loss": 0.90273798,
      "num_input_tokens_seen": 4850120,
      "step": 229,
      "time_per_iteration": 2.520725965499878
    },
    {
      "auxiliary_loss_clip": 0.01378691,
      "auxiliary_loss_mlp": 0.0110128,
      "balance_loss_clip": 1.10664546,
      "balance_loss_mlp": 1.05068755,
      "epoch": 0.013828348113632948,
      "flos": 20193037367040.0,
      "grad_norm": 3.740747102856291,
      "language_loss": 0.8407408,
      "learning_rate": 3.501319237118231e-06,
      "loss": 0.8655405,
      "num_input_tokens_seen": 4866215,
      "step": 230,
      "time_per_iteration": 2.516561985015869
    },
    {
      "auxiliary_loss_clip": 0.01376157,
      "auxiliary_loss_mlp": 0.01117923,
      "balance_loss_clip": 1.1083101,
      "balance_loss_mlp": 1.06799793,
      "epoch": 0.013888471366300916,
      "flos": 20741070147840.0,
      "grad_norm": 1.9950068402642451,
      "language_loss": 0.90644079,
      "learning_rate": 3.5041125264604056e-06,
      "loss": 0.93138158,
      "num_input_tokens_seen": 4885630,
      "step": 231,
      "time_per_iteration": 2.5194008350372314
    },
    {
      "auxiliary_loss_clip": 0.0138042,
      "auxiliary_loss_mlp": 0.01110323,
      "balance_loss_clip": 1.11219323,
      "balance_loss_mlp": 1.05999279,
      "epoch": 0.013948594618968886,
      "flos": 22090628966400.0,
      "grad_norm": 2.180840857458781,
      "language_loss": 0.83808231,
      "learning_rate": 3.5068937497203002e-06,
      "loss": 0.86298978,
      "num_input_tokens_seen": 4905570,
      "step": 232,
      "time_per_iteration": 2.5337069034576416
    },
    {
      "auxiliary_loss_clip": 0.01380607,
      "auxiliary_loss_mlp": 0.01092671,
      "balance_loss_clip": 1.10332799,
      "balance_loss_mlp": 1.04088664,
      "epoch": 0.014008717871636855,
      "flos": 19063108258560.0,
      "grad_norm": 4.047560144586128,
      "language_loss": 0.74348074,
      "learning_rate": 3.509663010692652e-06,
      "loss": 0.76821351,
      "num_input_tokens_seen": 4923535,
      "step": 233,
      "time_per_iteration": 2.509115695953369
    },
    {
      "auxiliary_loss_clip": 0.01385038,
      "auxiliary_loss_mlp": 0.01122754,
      "balance_loss_clip": 1.11052656,
      "balance_loss_mlp": 1.06982541,
      "epoch": 0.014068841124304825,
      "flos": 14530822064640.0,
      "grad_norm": 3.4904575543659124,
      "language_loss": 0.85897446,
      "learning_rate": 3.512420411838642e-06,
      "loss": 0.8840524,
      "num_input_tokens_seen": 4939200,
      "step": 234,
      "time_per_iteration": 2.515504837036133
    },
    {
      "auxiliary_loss_clip": 0.01376332,
      "auxiliary_loss_mlp": 0.01113065,
      "balance_loss_clip": 1.10977292,
      "balance_loss_mlp": 1.06309211,
      "epoch": 0.014128964376972794,
      "flos": 18077396256000.0,
      "grad_norm": 2.4531900933922306,
      "language_loss": 0.89403313,
      "learning_rate": 3.515166054308634e-06,
      "loss": 0.91892713,
      "num_input_tokens_seen": 4956620,
      "step": 235,
      "time_per_iteration": 2.4948620796203613
    },
    {
      "auxiliary_loss_clip": 0.01377438,
      "auxiliary_loss_mlp": 0.01119191,
      "balance_loss_clip": 1.11209154,
      "balance_loss_mlp": 1.06766915,
      "epoch": 0.014189087629640764,
      "flos": 25334331678720.0,
      "grad_norm": 2.1103530005430255,
      "language_loss": 0.85574841,
      "learning_rate": 3.5179000379644498e-06,
      "loss": 0.88071471,
      "num_input_tokens_seen": 4975650,
      "step": 236,
      "time_per_iteration": 2.58736252784729
    },
    {
      "auxiliary_loss_clip": 0.01372902,
      "auxiliary_loss_mlp": 0.01099199,
      "balance_loss_clip": 1.10394585,
      "balance_loss_mlp": 1.0481056,
      "epoch": 0.014249210882308733,
      "flos": 36139744713600.0,
      "grad_norm": 1.794890374651873,
      "language_loss": 0.82350588,
      "learning_rate": 3.520622461401154e-06,
      "loss": 0.8482269,
      "num_input_tokens_seen": 4997415,
      "step": 237,
      "time_per_iteration": 2.653890609741211
    },
    {
      "auxiliary_loss_clip": 0.01372444,
      "auxiliary_loss_mlp": 0.01118588,
      "balance_loss_clip": 1.10811889,
      "balance_loss_mlp": 1.06480098,
      "epoch": 0.014309334134976702,
      "flos": 12932977461120.0,
      "grad_norm": 2.6889564776364963,
      "language_loss": 0.77308869,
      "learning_rate": 3.5233334219683935e-06,
      "loss": 0.79799902,
      "num_input_tokens_seen": 5013905,
      "step": 238,
      "time_per_iteration": 2.5154969692230225
    },
    {
      "auxiliary_loss_clip": 0.01370011,
      "auxiliary_loss_mlp": 0.01115214,
      "balance_loss_clip": 1.1105926,
      "balance_loss_mlp": 1.06724429,
      "epoch": 0.014369457387644672,
      "flos": 20777519473920.0,
      "grad_norm": 1.901824439886258,
      "language_loss": 0.87094796,
      "learning_rate": 3.526033015791284e-06,
      "loss": 0.89580011,
      "num_input_tokens_seen": 5033645,
      "step": 239,
      "time_per_iteration": 2.697768211364746
    },
    {
      "auxiliary_loss_clip": 0.0135388,
      "auxiliary_loss_mlp": 0.0110157,
      "balance_loss_clip": 1.1003207,
      "balance_loss_mlp": 1.05321884,
      "epoch": 0.01442958064031264,
      "flos": 25848536826240.0,
      "grad_norm": 2.1022571165152266,
      "language_loss": 0.92993557,
      "learning_rate": 3.528721337790862e-06,
      "loss": 0.95449007,
      "num_input_tokens_seen": 5052875,
      "step": 240,
      "time_per_iteration": 2.5998799800872803
    },
    {
      "auxiliary_loss_clip": 0.01363125,
      "auxiliary_loss_mlp": 0.01108026,
      "balance_loss_clip": 1.10578895,
      "balance_loss_mlp": 1.06139135,
      "epoch": 0.014489703892980611,
      "flos": 28219718269440.0,
      "grad_norm": 3.7286801621874743,
      "language_loss": 0.8499316,
      "learning_rate": 3.531398481704111e-06,
      "loss": 0.87464309,
      "num_input_tokens_seen": 5075005,
      "step": 241,
      "time_per_iteration": 2.6013524532318115
    },
    {
      "auxiliary_loss_clip": 0.01362522,
      "auxiliary_loss_mlp": 0.01119205,
      "balance_loss_clip": 1.11130524,
      "balance_loss_mlp": 1.06911302,
      "epoch": 0.01454982714564858,
      "flos": 22490925108480.0,
      "grad_norm": 1.9842216956317018,
      "language_loss": 0.88549984,
      "learning_rate": 3.534064540103573e-06,
      "loss": 0.91031706,
      "num_input_tokens_seen": 5091875,
      "step": 242,
      "time_per_iteration": 2.5850167274475098
    },
    {
      "auxiliary_loss_clip": 0.01362224,
      "auxiliary_loss_mlp": 0.01106639,
      "balance_loss_clip": 1.10480428,
      "balance_loss_mlp": 1.05485404,
      "epoch": 0.014609950398316548,
      "flos": 21653201139840.0,
      "grad_norm": 3.55145796641499,
      "language_loss": 0.86753094,
      "learning_rate": 3.536719604416555e-06,
      "loss": 0.89221954,
      "num_input_tokens_seen": 5111290,
      "step": 243,
      "time_per_iteration": 2.5179176330566406
    },
    {
      "auxiliary_loss_clip": 0.01367839,
      "auxiliary_loss_mlp": 0.0110693,
      "balance_loss_clip": 1.1074779,
      "balance_loss_mlp": 1.05619431,
      "epoch": 0.014670073650984519,
      "flos": 21869993675520.0,
      "grad_norm": 1.636914612065287,
      "language_loss": 0.84211922,
      "learning_rate": 3.5393637649439464e-06,
      "loss": 0.86686695,
      "num_input_tokens_seen": 5132265,
      "step": 244,
      "time_per_iteration": 2.562384843826294
    },
    {
      "auxiliary_loss_clip": 0.01374775,
      "auxiliary_loss_mlp": 0.01118863,
      "balance_loss_clip": 1.10892034,
      "balance_loss_mlp": 1.06652975,
      "epoch": 0.014730196903652487,
      "flos": 23183713699200.0,
      "grad_norm": 3.5734834474497483,
      "language_loss": 0.78790694,
      "learning_rate": 3.54199711087864e-06,
      "loss": 0.81284332,
      "num_input_tokens_seen": 5148575,
      "step": 245,
      "time_per_iteration": 2.5267293453216553
    },
    {
      "auxiliary_loss_clip": 0.01372587,
      "auxiliary_loss_mlp": 0.01106462,
      "balance_loss_clip": 1.1042397,
      "balance_loss_mlp": 1.05253208,
      "epoch": 0.014790320156320457,
      "flos": 23222605150080.0,
      "grad_norm": 1.9975971415531737,
      "language_loss": 0.83980095,
      "learning_rate": 3.5446197303235913e-06,
      "loss": 0.86459142,
      "num_input_tokens_seen": 5170415,
      "step": 246,
      "time_per_iteration": 2.5849974155426025
    },
    {
      "auxiliary_loss_clip": 0.0136706,
      "auxiliary_loss_mlp": 0.01101148,
      "balance_loss_clip": 1.10368872,
      "balance_loss_mlp": 1.05103219,
      "epoch": 0.014850443408988426,
      "flos": 15815490963840.0,
      "grad_norm": 2.438868014737639,
      "language_loss": 0.90017498,
      "learning_rate": 3.5472317103095034e-06,
      "loss": 0.92485702,
      "num_input_tokens_seen": 5188565,
      "step": 247,
      "time_per_iteration": 2.50834321975708
    },
    {
      "auxiliary_loss_clip": 0.01365806,
      "auxiliary_loss_mlp": 0.01100704,
      "balance_loss_clip": 1.09931266,
      "balance_loss_mlp": 1.05185175,
      "epoch": 0.014910566661656396,
      "flos": 22781657790720.0,
      "grad_norm": 2.3599414979908655,
      "language_loss": 0.77891719,
      "learning_rate": 3.549833136812155e-06,
      "loss": 0.80358231,
      "num_input_tokens_seen": 5207810,
      "step": 248,
      "time_per_iteration": 2.5498206615448
    },
    {
      "auxiliary_loss_clip": 0.01366736,
      "auxiliary_loss_mlp": 0.0110798,
      "balance_loss_clip": 1.10773385,
      "balance_loss_mlp": 1.0578407,
      "epoch": 0.014970689914324365,
      "flos": 26865023806080.0,
      "grad_norm": 1.9976307824361628,
      "language_loss": 0.83752239,
      "learning_rate": 3.552424094769381e-06,
      "loss": 0.86226952,
      "num_input_tokens_seen": 5226210,
      "step": 249,
      "time_per_iteration": 2.555947780609131
    },
    {
      "auxiliary_loss_clip": 0.01360423,
      "auxiliary_loss_mlp": 0.01106543,
      "balance_loss_clip": 1.1018579,
      "balance_loss_mlp": 1.05850208,
      "epoch": 0.015030813166992334,
      "flos": 13985662371840.0,
      "grad_norm": 2.130508993705231,
      "language_loss": 0.93491417,
      "learning_rate": 3.5550046680977174e-06,
      "loss": 0.95958382,
      "num_input_tokens_seen": 5241660,
      "step": 250,
      "time_per_iteration": 2.4924814701080322
    },
    {
      "auxiliary_loss_clip": 0.0136866,
      "auxiliary_loss_mlp": 0.01113585,
      "balance_loss_clip": 1.10640812,
      "balance_loss_mlp": 1.06261063,
      "epoch": 0.015090936419660304,
      "flos": 24717817618560.0,
      "grad_norm": 2.6071368708473552,
      "language_loss": 0.97118521,
      "learning_rate": 3.5575749397087034e-06,
      "loss": 0.99600768,
      "num_input_tokens_seen": 5261090,
      "step": 251,
      "time_per_iteration": 2.5534534454345703
    },
    {
      "auxiliary_loss_clip": 0.01363485,
      "auxiliary_loss_mlp": 0.01110481,
      "balance_loss_clip": 1.10152173,
      "balance_loss_mlp": 1.06124806,
      "epoch": 0.015151059672328273,
      "flos": 25738793798400.0,
      "grad_norm": 4.671852518599087,
      "language_loss": 0.84020257,
      "learning_rate": 3.5601349915248707e-06,
      "loss": 0.86494225,
      "num_input_tokens_seen": 5279175,
      "step": 252,
      "time_per_iteration": 6.8410022258758545
    },
    {
      "auxiliary_loss_clip": 0.01358484,
      "auxiliary_loss_mlp": 0.01114374,
      "balance_loss_clip": 1.10373652,
      "balance_loss_mlp": 1.06509256,
      "epoch": 0.015211182924996243,
      "flos": 21871214737920.0,
      "grad_norm": 2.954963566843214,
      "language_loss": 0.98369592,
      "learning_rate": 3.5626849044954064e-06,
      "loss": 1.00842452,
      "num_input_tokens_seen": 5296975,
      "step": 253,
      "time_per_iteration": 2.5377111434936523
    },
    {
      "auxiliary_loss_clip": 0.01248912,
      "auxiliary_loss_mlp": 0.01054778,
      "balance_loss_clip": 1.11190462,
      "balance_loss_mlp": 1.03742111,
      "epoch": 0.015271306177664212,
      "flos": 66895080888960.0,
      "grad_norm": 0.8673010973163903,
      "language_loss": 0.55665803,
      "learning_rate": 3.5652247586115167e-06,
      "loss": 0.57969499,
      "num_input_tokens_seen": 5358375,
      "step": 254,
      "time_per_iteration": 3.1389174461364746
    },
    {
      "auxiliary_loss_clip": 0.01362744,
      "auxiliary_loss_mlp": 0.01116424,
      "balance_loss_clip": 1.09876251,
      "balance_loss_mlp": 1.06716669,
      "epoch": 0.01533142943033218,
      "flos": 26834069260800.0,
      "grad_norm": 2.000381231832675,
      "language_loss": 0.89998662,
      "learning_rate": 3.567754632921479e-06,
      "loss": 0.92477834,
      "num_input_tokens_seen": 5377255,
      "step": 255,
      "time_per_iteration": 2.57558274269104
    },
    {
      "auxiliary_loss_clip": 0.01359343,
      "auxiliary_loss_mlp": 0.01130865,
      "balance_loss_clip": 1.101511,
      "balance_loss_mlp": 1.08060622,
      "epoch": 0.01539155268300015,
      "flos": 20813753318400.0,
      "grad_norm": 2.5013527323490248,
      "language_loss": 0.85574663,
      "learning_rate": 3.5702746055454075e-06,
      "loss": 0.88064873,
      "num_input_tokens_seen": 5395320,
      "step": 256,
      "time_per_iteration": 2.55669903755188
    },
    {
      "auxiliary_loss_clip": 0.01365031,
      "auxiliary_loss_mlp": 0.01114993,
      "balance_loss_clip": 1.10173476,
      "balance_loss_mlp": 1.06497252,
      "epoch": 0.01545167593566812,
      "flos": 15961862885760.0,
      "grad_norm": 4.026885389394537,
      "language_loss": 0.71599841,
      "learning_rate": 3.5727847536897254e-06,
      "loss": 0.74079859,
      "num_input_tokens_seen": 5411970,
      "step": 257,
      "time_per_iteration": 2.5195844173431396
    },
    {
      "auxiliary_loss_clip": 0.01359155,
      "auxiliary_loss_mlp": 0.01105034,
      "balance_loss_clip": 1.10190642,
      "balance_loss_mlp": 1.05546713,
      "epoch": 0.01551179918833609,
      "flos": 22601745544320.0,
      "grad_norm": 1.9088300716225362,
      "language_loss": 0.94414544,
      "learning_rate": 3.5752851536613596e-06,
      "loss": 0.96878737,
      "num_input_tokens_seen": 5430245,
      "step": 258,
      "time_per_iteration": 2.5318691730499268
    },
    {
      "auxiliary_loss_clip": 0.0135839,
      "auxiliary_loss_mlp": 0.0111157,
      "balance_loss_clip": 1.09958136,
      "balance_loss_mlp": 1.06302762,
      "epoch": 0.015571922441004058,
      "flos": 22816706486400.0,
      "grad_norm": 2.2241151549438865,
      "language_loss": 0.93147421,
      "learning_rate": 3.577775880881658e-06,
      "loss": 0.9561739,
      "num_input_tokens_seen": 5448905,
      "step": 259,
      "time_per_iteration": 2.537464141845703
    },
    {
      "auxiliary_loss_clip": 0.01351674,
      "auxiliary_loss_mlp": 0.01099412,
      "balance_loss_clip": 1.10240221,
      "balance_loss_mlp": 1.05334973,
      "epoch": 0.015632045693672027,
      "flos": 18947439486720.0,
      "grad_norm": 1.8100825929263045,
      "language_loss": 0.97499704,
      "learning_rate": 3.5802570099000424e-06,
      "loss": 0.99950784,
      "num_input_tokens_seen": 5466405,
      "step": 260,
      "time_per_iteration": 2.5359139442443848
    },
    {
      "auxiliary_loss_clip": 0.01367381,
      "auxiliary_loss_mlp": 0.01118224,
      "balance_loss_clip": 1.10392773,
      "balance_loss_mlp": 1.070207,
      "epoch": 0.015692168946339995,
      "flos": 29971728046080.0,
      "grad_norm": 2.3978499308780394,
      "language_loss": 0.87622184,
      "learning_rate": 3.5827286144073947e-06,
      "loss": 0.90107793,
      "num_input_tokens_seen": 5487055,
      "step": 261,
      "time_per_iteration": 2.595839500427246
    },
    {
      "auxiliary_loss_clip": 0.01357842,
      "auxiliary_loss_mlp": 0.01111158,
      "balance_loss_clip": 1.09972167,
      "balance_loss_mlp": 1.06225848,
      "epoch": 0.015752292199007967,
      "flos": 19392085946880.0,
      "grad_norm": 2.0419986323467567,
      "language_loss": 0.67188448,
      "learning_rate": 3.5851907672491904e-06,
      "loss": 0.69657445,
      "num_input_tokens_seen": 5506600,
      "step": 262,
      "time_per_iteration": 2.529513120651245
    },
    {
      "auxiliary_loss_clip": 0.0135694,
      "auxiliary_loss_mlp": 0.01121272,
      "balance_loss_clip": 1.10174894,
      "balance_loss_mlp": 1.07101297,
      "epoch": 0.015812415451675936,
      "flos": 20339804338560.0,
      "grad_norm": 2.5881735238743597,
      "language_loss": 0.68002111,
      "learning_rate": 3.587643540438383e-06,
      "loss": 0.70480323,
      "num_input_tokens_seen": 5524350,
      "step": 263,
      "time_per_iteration": 2.511111259460449
    },
    {
      "auxiliary_loss_clip": 0.01354858,
      "auxiliary_loss_mlp": 0.01106401,
      "balance_loss_clip": 1.09695625,
      "balance_loss_mlp": 1.05754912,
      "epoch": 0.015872538704343905,
      "flos": 17525412979200.0,
      "grad_norm": 2.8885617583956966,
      "language_loss": 0.853661,
      "learning_rate": 3.590087005168037e-06,
      "loss": 0.87827361,
      "num_input_tokens_seen": 5542145,
      "step": 264,
      "time_per_iteration": 2.5206456184387207
    },
    {
      "auxiliary_loss_clip": 0.01362892,
      "auxiliary_loss_mlp": 0.01097039,
      "balance_loss_clip": 1.102947,
      "balance_loss_mlp": 1.05016589,
      "epoch": 0.015932661957011873,
      "flos": 15260490944640.0,
      "grad_norm": 2.918258430303743,
      "language_loss": 1.04258764,
      "learning_rate": 3.5925212318237344e-06,
      "loss": 1.06718695,
      "num_input_tokens_seen": 5557920,
      "step": 265,
      "time_per_iteration": 2.4944159984588623
    },
    {
      "auxiliary_loss_clip": 0.01364934,
      "auxiliary_loss_mlp": 0.01116949,
      "balance_loss_clip": 1.10501623,
      "balance_loss_mlp": 1.06504536,
      "epoch": 0.015992785209679845,
      "flos": 20302528999680.0,
      "grad_norm": 2.4039413117482957,
      "language_loss": 0.7523104,
      "learning_rate": 3.5949462899957323e-06,
      "loss": 0.77712917,
      "num_input_tokens_seen": 5576290,
      "step": 266,
      "time_per_iteration": 2.53471040725708
    },
    {
      "auxiliary_loss_clip": 0.01351882,
      "auxiliary_loss_mlp": 0.01100797,
      "balance_loss_clip": 1.10244286,
      "balance_loss_mlp": 1.05211234,
      "epoch": 0.016052908462347814,
      "flos": 23362368969600.0,
      "grad_norm": 1.8377807903895496,
      "language_loss": 0.90750849,
      "learning_rate": 3.5973622484909068e-06,
      "loss": 0.93203533,
      "num_input_tokens_seen": 5595205,
      "step": 267,
      "time_per_iteration": 2.5514185428619385
    },
    {
      "auxiliary_loss_clip": 0.01361136,
      "auxiliary_loss_mlp": 0.0111957,
      "balance_loss_clip": 1.10222924,
      "balance_loss_mlp": 1.07224357,
      "epoch": 0.016113031715015783,
      "flos": 21286588976640.0,
      "grad_norm": 2.6644671198576932,
      "language_loss": 0.85729837,
      "learning_rate": 3.599769175344462e-06,
      "loss": 0.88210547,
      "num_input_tokens_seen": 5612645,
      "step": 268,
      "time_per_iteration": 2.535179615020752
    },
    {
      "auxiliary_loss_clip": 0.01354218,
      "auxiliary_loss_mlp": 0.01097601,
      "balance_loss_clip": 1.10415888,
      "balance_loss_mlp": 1.05051351,
      "epoch": 0.01617315496768375,
      "flos": 18914689261440.0,
      "grad_norm": 2.670727820412289,
      "language_loss": 0.88510281,
      "learning_rate": 3.602167137831432e-06,
      "loss": 0.909621,
      "num_input_tokens_seen": 5628345,
      "step": 269,
      "time_per_iteration": 2.504873514175415
    },
    {
      "auxiliary_loss_clip": 0.01358593,
      "auxiliary_loss_mlp": 0.01104483,
      "balance_loss_clip": 1.10061765,
      "balance_loss_mlp": 1.05281782,
      "epoch": 0.01623327822035172,
      "flos": 16546488647040.0,
      "grad_norm": 2.0895170298773316,
      "language_loss": 0.97050202,
      "learning_rate": 3.6045562024779565e-06,
      "loss": 0.9951328,
      "num_input_tokens_seen": 5645940,
      "step": 270,
      "time_per_iteration": 2.5209782123565674
    },
    {
      "auxiliary_loss_clip": 0.0135913,
      "auxiliary_loss_mlp": 0.01118428,
      "balance_loss_clip": 1.10646236,
      "balance_loss_mlp": 1.07024348,
      "epoch": 0.016293401473019692,
      "flos": 23513481486720.0,
      "grad_norm": 2.3759671943872127,
      "language_loss": 0.86464715,
      "learning_rate": 3.606936435072361e-06,
      "loss": 0.88942271,
      "num_input_tokens_seen": 5665690,
      "step": 271,
      "time_per_iteration": 2.5473685264587402
    },
    {
      "auxiliary_loss_clip": 0.01354972,
      "auxiliary_loss_mlp": 0.01107112,
      "balance_loss_clip": 1.09663105,
      "balance_loss_mlp": 1.05926085,
      "epoch": 0.01635352472568766,
      "flos": 29016072748800.0,
      "grad_norm": 2.4129579947204833,
      "language_loss": 0.81222534,
      "learning_rate": 3.609307900676025e-06,
      "loss": 0.83684611,
      "num_input_tokens_seen": 5683190,
      "step": 272,
      "time_per_iteration": 2.584460496902466
    },
    {
      "auxiliary_loss_clip": 0.01349206,
      "auxiliary_loss_mlp": 0.01122272,
      "balance_loss_clip": 1.09948301,
      "balance_loss_mlp": 1.07570875,
      "epoch": 0.01641364797835563,
      "flos": 13370513028480.0,
      "grad_norm": 2.101619129271149,
      "language_loss": 0.81057894,
      "learning_rate": 3.611670663634051e-06,
      "loss": 0.83529371,
      "num_input_tokens_seen": 5699780,
      "step": 273,
      "time_per_iteration": 2.498856782913208
    },
    {
      "auxiliary_loss_clip": 0.01347155,
      "auxiliary_loss_mlp": 0.01106375,
      "balance_loss_clip": 1.09480536,
      "balance_loss_mlp": 1.05857229,
      "epoch": 0.016473771231023598,
      "flos": 18878239935360.0,
      "grad_norm": 2.139345533409641,
      "language_loss": 0.91233766,
      "learning_rate": 3.614024787585744e-06,
      "loss": 0.93687284,
      "num_input_tokens_seen": 5716980,
      "step": 274,
      "time_per_iteration": 2.547649621963501
    },
    {
      "auxiliary_loss_clip": 0.01347435,
      "auxiliary_loss_mlp": 0.01114881,
      "balance_loss_clip": 1.09782517,
      "balance_loss_mlp": 1.06650615,
      "epoch": 0.016533894483691566,
      "flos": 22601637803520.0,
      "grad_norm": 1.759998900929906,
      "language_loss": 0.87899989,
      "learning_rate": 3.6163703354748927e-06,
      "loss": 0.90362298,
      "num_input_tokens_seen": 5737780,
      "step": 275,
      "time_per_iteration": 2.549131393432617
    },
    {
      "auxiliary_loss_clip": 0.01350587,
      "auxiliary_loss_mlp": 0.01104661,
      "balance_loss_clip": 1.099159,
      "balance_loss_mlp": 1.05495048,
      "epoch": 0.01659401773635954,
      "flos": 21507188353920.0,
      "grad_norm": 1.6307218884404493,
      "language_loss": 0.80728763,
      "learning_rate": 3.6187073695598707e-06,
      "loss": 0.8318401,
      "num_input_tokens_seen": 5758330,
      "step": 276,
      "time_per_iteration": 2.548225164413452
    },
    {
      "auxiliary_loss_clip": 0.01340912,
      "auxiliary_loss_mlp": 0.01100361,
      "balance_loss_clip": 1.09751666,
      "balance_loss_mlp": 1.05592024,
      "epoch": 0.016654140989027507,
      "flos": 32850973411200.0,
      "grad_norm": 3.0254175799438774,
      "language_loss": 0.8099457,
      "learning_rate": 3.621035951423551e-06,
      "loss": 0.83435845,
      "num_input_tokens_seen": 5778340,
      "step": 277,
      "time_per_iteration": 2.6365997791290283
    },
    {
      "auxiliary_loss_clip": 0.01339161,
      "auxiliary_loss_mlp": 0.01095437,
      "balance_loss_clip": 1.09108841,
      "balance_loss_mlp": 1.04911256,
      "epoch": 0.016714264241695476,
      "flos": 12306228024960.0,
      "grad_norm": 2.0563842439974733,
      "language_loss": 0.80398297,
      "learning_rate": 3.623356141983041e-06,
      "loss": 0.82832897,
      "num_input_tokens_seen": 5794295,
      "step": 278,
      "time_per_iteration": 2.5450284481048584
    },
    {
      "auxiliary_loss_clip": 0.01344861,
      "auxiliary_loss_mlp": 0.01102195,
      "balance_loss_clip": 1.09682906,
      "balance_loss_mlp": 1.05629945,
      "epoch": 0.016774387494363444,
      "flos": 27123796362240.0,
      "grad_norm": 1.695341178338216,
      "language_loss": 0.90399712,
      "learning_rate": 3.6256680014992486e-06,
      "loss": 0.92846769,
      "num_input_tokens_seen": 5814405,
      "step": 279,
      "time_per_iteration": 2.6079492568969727
    },
    {
      "auxiliary_loss_clip": 0.01349991,
      "auxiliary_loss_mlp": 0.01119104,
      "balance_loss_clip": 1.09726751,
      "balance_loss_mlp": 1.07122993,
      "epoch": 0.016834510747031413,
      "flos": 20191493082240.0,
      "grad_norm": 2.1942574233548098,
      "language_loss": 0.94055426,
      "learning_rate": 3.6279715895862713e-06,
      "loss": 0.96524525,
      "num_input_tokens_seen": 5832795,
      "step": 280,
      "time_per_iteration": 2.5388219356536865
    },
    {
      "auxiliary_loss_clip": 0.01350182,
      "auxiliary_loss_mlp": 0.01112538,
      "balance_loss_clip": 1.09594345,
      "balance_loss_mlp": 1.06454444,
      "epoch": 0.016894633999699385,
      "flos": 27274262434560.0,
      "grad_norm": 1.8832431167776666,
      "language_loss": 0.74161649,
      "learning_rate": 3.6302669652206183e-06,
      "loss": 0.7662437,
      "num_input_tokens_seen": 5855750,
      "step": 281,
      "time_per_iteration": 2.6330034732818604
    },
    {
      "auxiliary_loss_clip": 0.01345181,
      "auxiliary_loss_mlp": 0.01118325,
      "balance_loss_clip": 1.09767163,
      "balance_loss_mlp": 1.07307339,
      "epoch": 0.016954757252367354,
      "flos": 14902964922240.0,
      "grad_norm": 2.479813267575734,
      "language_loss": 0.80187267,
      "learning_rate": 3.632554186750274e-06,
      "loss": 0.82650775,
      "num_input_tokens_seen": 5872610,
      "step": 282,
      "time_per_iteration": 2.4776694774627686
    },
    {
      "auxiliary_loss_clip": 0.01351329,
      "auxiliary_loss_mlp": 0.01122504,
      "balance_loss_clip": 1.0999676,
      "balance_loss_mlp": 1.0753684,
      "epoch": 0.017014880505035322,
      "flos": 21358805270400.0,
      "grad_norm": 1.968930159867162,
      "language_loss": 0.77522027,
      "learning_rate": 3.6348333119035937e-06,
      "loss": 0.79995859,
      "num_input_tokens_seen": 5892985,
      "step": 283,
      "time_per_iteration": 2.5298852920532227
    },
    {
      "auxiliary_loss_clip": 0.01350023,
      "auxiliary_loss_mlp": 0.01094977,
      "balance_loss_clip": 1.10065103,
      "balance_loss_mlp": 1.05024981,
      "epoch": 0.01707500375770329,
      "flos": 35333154858240.0,
      "grad_norm": 3.61162032230288,
      "language_loss": 0.84020209,
      "learning_rate": 3.6371043977980503e-06,
      "loss": 0.8646521,
      "num_input_tokens_seen": 5914060,
      "step": 284,
      "time_per_iteration": 2.6063029766082764
    },
    {
      "auxiliary_loss_clip": 0.01341853,
      "auxiliary_loss_mlp": 0.01106124,
      "balance_loss_clip": 1.09582341,
      "balance_loss_mlp": 1.05808318,
      "epoch": 0.01713512701037126,
      "flos": 23582070506880.0,
      "grad_norm": 2.3823244694630543,
      "language_loss": 0.96769047,
      "learning_rate": 3.639367500948819e-06,
      "loss": 0.99217021,
      "num_input_tokens_seen": 5932860,
      "step": 285,
      "time_per_iteration": 2.5391108989715576
    },
    {
      "auxiliary_loss_clip": 0.01343575,
      "auxiliary_loss_mlp": 0.0109574,
      "balance_loss_clip": 1.09766126,
      "balance_loss_mlp": 1.05151343,
      "epoch": 0.01719525026303923,
      "flos": 27634661544960.0,
      "grad_norm": 2.035805266977711,
      "language_loss": 0.93776071,
      "learning_rate": 3.6416226772772178e-06,
      "loss": 0.96215391,
      "num_input_tokens_seen": 5952725,
      "step": 286,
      "time_per_iteration": 2.5694897174835205
    },
    {
      "auxiliary_loss_clip": 0.01337339,
      "auxiliary_loss_mlp": 0.01089256,
      "balance_loss_clip": 1.09304154,
      "balance_loss_mlp": 1.04355073,
      "epoch": 0.0172553735157072,
      "flos": 26979722910720.0,
      "grad_norm": 1.7015879555808981,
      "language_loss": 0.92440021,
      "learning_rate": 3.643869982119001e-06,
      "loss": 0.94866616,
      "num_input_tokens_seen": 5970560,
      "step": 287,
      "time_per_iteration": 2.587815284729004
    },
    {
      "auxiliary_loss_clip": 0.01339656,
      "auxiliary_loss_mlp": 0.0109326,
      "balance_loss_clip": 1.09215975,
      "balance_loss_mlp": 1.04784083,
      "epoch": 0.01731549676837517,
      "flos": 14056621689600.0,
      "grad_norm": 2.8754417687215152,
      "language_loss": 1.01777673,
      "learning_rate": 3.646109470232502e-06,
      "loss": 1.04210591,
      "num_input_tokens_seen": 5982980,
      "step": 288,
      "time_per_iteration": 2.4412081241607666
    },
    {
      "auxiliary_loss_clip": 0.01233328,
      "auxiliary_loss_mlp": 0.01123814,
      "balance_loss_clip": 1.10167682,
      "balance_loss_mlp": 1.10893714,
      "epoch": 0.017375620021043137,
      "flos": 66510694471680.0,
      "grad_norm": 0.9307221348947577,
      "language_loss": 0.6387279,
      "learning_rate": 3.6483411958066417e-06,
      "loss": 0.66229939,
      "num_input_tokens_seen": 6049445,
      "step": 289,
      "time_per_iteration": 3.2459771633148193
    },
    {
      "auxiliary_loss_clip": 0.01343565,
      "auxiliary_loss_mlp": 0.01113671,
      "balance_loss_clip": 1.09856653,
      "balance_loss_mlp": 1.07023072,
      "epoch": 0.01743574327371111,
      "flos": 15225154940160.0,
      "grad_norm": 2.427925743478862,
      "language_loss": 0.8839581,
      "learning_rate": 3.6505652124687957e-06,
      "loss": 0.90853047,
      "num_input_tokens_seen": 6064150,
      "step": 290,
      "time_per_iteration": 5.3138415813446045
    },
    {
      "auxiliary_loss_clip": 0.01339214,
      "auxiliary_loss_mlp": 0.0109791,
      "balance_loss_clip": 1.09575605,
      "balance_loss_mlp": 1.05330157,
      "epoch": 0.017495866526379078,
      "flos": 25373869574400.0,
      "grad_norm": 1.9230198200190896,
      "language_loss": 0.84609938,
      "learning_rate": 3.6527815732925258e-06,
      "loss": 0.87047058,
      "num_input_tokens_seen": 6083920,
      "step": 291,
      "time_per_iteration": 2.563136339187622
    },
    {
      "auxiliary_loss_clip": 0.01346307,
      "auxiliary_loss_mlp": 0.01112867,
      "balance_loss_clip": 1.10312796,
      "balance_loss_mlp": 1.06456339,
      "epoch": 0.017555989779047047,
      "flos": 26359473836160.0,
      "grad_norm": 1.6011466264553436,
      "language_loss": 0.7267921,
      "learning_rate": 3.6549903308051806e-06,
      "loss": 0.75138378,
      "num_input_tokens_seen": 6105460,
      "step": 292,
      "time_per_iteration": 2.574538230895996
    },
    {
      "auxiliary_loss_clip": 0.01334533,
      "auxiliary_loss_mlp": 0.01108476,
      "balance_loss_clip": 1.09444451,
      "balance_loss_mlp": 1.06327212,
      "epoch": 0.017616113031715015,
      "flos": 22338807010560.0,
      "grad_norm": 2.3082278197228443,
      "language_loss": 0.87150347,
      "learning_rate": 3.6571915369953646e-06,
      "loss": 0.89593351,
      "num_input_tokens_seen": 6122890,
      "step": 293,
      "time_per_iteration": 2.5188400745391846
    },
    {
      "auxiliary_loss_clip": 0.01335353,
      "auxiliary_loss_mlp": 0.01117073,
      "balance_loss_clip": 1.09437716,
      "balance_loss_mlp": 1.07210779,
      "epoch": 0.017676236284382984,
      "flos": 20156911263360.0,
      "grad_norm": 2.58156296475737,
      "language_loss": 0.81073326,
      "learning_rate": 3.6593852433202797e-06,
      "loss": 0.83525747,
      "num_input_tokens_seen": 6142890,
      "step": 294,
      "time_per_iteration": 3.9038243293762207
    },
    {
      "auxiliary_loss_clip": 0.01334855,
      "auxiliary_loss_mlp": 0.01121771,
      "balance_loss_clip": 1.09100759,
      "balance_loss_mlp": 1.0766387,
      "epoch": 0.017736359537050956,
      "flos": 25223331674880.0,
      "grad_norm": 1.8832542294351495,
      "language_loss": 0.83657086,
      "learning_rate": 3.6615715007129453e-06,
      "loss": 0.86113715,
      "num_input_tokens_seen": 6162030,
      "step": 295,
      "time_per_iteration": 2.562040090560913
    },
    {
      "auxiliary_loss_clip": 0.01340007,
      "auxiliary_loss_mlp": 0.01122642,
      "balance_loss_clip": 1.10024905,
      "balance_loss_mlp": 1.07736683,
      "epoch": 0.017796482789718925,
      "flos": 20338798757760.0,
      "grad_norm": 4.3199352828927235,
      "language_loss": 0.84518093,
      "learning_rate": 3.6637503595892897e-06,
      "loss": 0.86980742,
      "num_input_tokens_seen": 6180540,
      "step": 296,
      "time_per_iteration": 2.534372091293335
    },
    {
      "auxiliary_loss_clip": 0.01341111,
      "auxiliary_loss_mlp": 0.01109048,
      "balance_loss_clip": 1.09664202,
      "balance_loss_mlp": 1.06470215,
      "epoch": 0.017856606042386893,
      "flos": 22379206832640.0,
      "grad_norm": 9.47845339023804,
      "language_loss": 0.87818229,
      "learning_rate": 3.665921869855132e-06,
      "loss": 0.90268391,
      "num_input_tokens_seen": 6199425,
      "step": 297,
      "time_per_iteration": 2.5354127883911133
    },
    {
      "auxiliary_loss_clip": 0.01339649,
      "auxiliary_loss_mlp": 0.01107176,
      "balance_loss_clip": 1.09500957,
      "balance_loss_mlp": 1.06306839,
      "epoch": 0.017916729295054862,
      "flos": 20230061310720.0,
      "grad_norm": 2.06483633729234,
      "language_loss": 0.88298631,
      "learning_rate": 3.6680860809130346e-06,
      "loss": 0.90745461,
      "num_input_tokens_seen": 6219170,
      "step": 298,
      "time_per_iteration": 2.5008327960968018
    },
    {
      "auxiliary_loss_clip": 0.01336008,
      "auxiliary_loss_mlp": 0.01130674,
      "balance_loss_clip": 1.0964514,
      "balance_loss_mlp": 1.08444476,
      "epoch": 0.01797685254772283,
      "flos": 19390972625280.0,
      "grad_norm": 2.431403372387156,
      "language_loss": 0.8863672,
      "learning_rate": 3.6702430416690516e-06,
      "loss": 0.91103399,
      "num_input_tokens_seen": 6237930,
      "step": 299,
      "time_per_iteration": 2.5303800106048584
    },
    {
      "auxiliary_loss_clip": 0.01341325,
      "auxiliary_loss_mlp": 0.0111682,
      "balance_loss_clip": 1.09554935,
      "balance_loss_mlp": 1.07087636,
      "epoch": 0.018036975800390802,
      "flos": 24426007528320.0,
      "grad_norm": 3.7544407034872926,
      "language_loss": 0.64941609,
      "learning_rate": 3.672392800539357e-06,
      "loss": 0.67399752,
      "num_input_tokens_seen": 6257170,
      "step": 300,
      "time_per_iteration": 2.5367417335510254
    },
    {
      "auxiliary_loss_clip": 0.01341658,
      "auxiliary_loss_mlp": 0.01116804,
      "balance_loss_clip": 1.09932256,
      "balance_loss_mlp": 1.07140875,
      "epoch": 0.01809709905305877,
      "flos": 15778933896960.0,
      "grad_norm": 2.4864296136859214,
      "language_loss": 0.88327873,
      "learning_rate": 3.6745354054567686e-06,
      "loss": 0.90786338,
      "num_input_tokens_seen": 6274780,
      "step": 301,
      "time_per_iteration": 2.4992129802703857
    },
    {
      "auxiliary_loss_clip": 0.01219983,
      "auxiliary_loss_mlp": 0.01103495,
      "balance_loss_clip": 1.08691525,
      "balance_loss_mlp": 1.08976173,
      "epoch": 0.01815722230572674,
      "flos": 67348382526720.0,
      "grad_norm": 0.838361940170133,
      "language_loss": 0.62254959,
      "learning_rate": 3.676670903877158e-06,
      "loss": 0.64578432,
      "num_input_tokens_seen": 6340435,
      "step": 302,
      "time_per_iteration": 3.241762399673462
    },
    {
      "auxiliary_loss_clip": 0.01330805,
      "auxiliary_loss_mlp": 0.01112582,
      "balance_loss_clip": 1.0919019,
      "balance_loss_mlp": 1.06675768,
      "epoch": 0.01821734555839471,
      "flos": 15485615435520.0,
      "grad_norm": 2.328279772106465,
      "language_loss": 0.89759696,
      "learning_rate": 3.6787993427857567e-06,
      "loss": 0.92203081,
      "num_input_tokens_seen": 6358160,
      "step": 303,
      "time_per_iteration": 2.4968881607055664
    },
    {
      "auxiliary_loss_clip": 0.01337659,
      "auxiliary_loss_mlp": 0.01118862,
      "balance_loss_clip": 1.09677899,
      "balance_loss_mlp": 1.07220364,
      "epoch": 0.018277468811062677,
      "flos": 24097424889600.0,
      "grad_norm": 1.8383419497140816,
      "language_loss": 0.8044219,
      "learning_rate": 3.680920768703364e-06,
      "loss": 0.82898712,
      "num_input_tokens_seen": 6378485,
      "step": 304,
      "time_per_iteration": 2.5360758304595947
    },
    {
      "auxiliary_loss_clip": 0.01331076,
      "auxiliary_loss_mlp": 0.01100502,
      "balance_loss_clip": 1.09878373,
      "balance_loss_mlp": 1.05708599,
      "epoch": 0.01833759206373065,
      "flos": 20959335141120.0,
      "grad_norm": 1.520208305976294,
      "language_loss": 0.82941401,
      "learning_rate": 3.6830352276924415e-06,
      "loss": 0.85372984,
      "num_input_tokens_seen": 6397845,
      "step": 305,
      "time_per_iteration": 2.518463611602783
    },
    {
      "auxiliary_loss_clip": 0.01332569,
      "auxiliary_loss_mlp": 0.01096086,
      "balance_loss_clip": 1.09137988,
      "balance_loss_mlp": 1.05276537,
      "epoch": 0.018397715316398618,
      "flos": 19390757143680.0,
      "grad_norm": 1.7907027535521303,
      "language_loss": 0.90931678,
      "learning_rate": 3.685142765363119e-06,
      "loss": 0.93360329,
      "num_input_tokens_seen": 6416475,
      "step": 306,
      "time_per_iteration": 2.4926939010620117
    },
    {
      "auxiliary_loss_clip": 0.01326175,
      "auxiliary_loss_mlp": 0.01092986,
      "balance_loss_clip": 1.08908272,
      "balance_loss_mlp": 1.04999924,
      "epoch": 0.018457838569066586,
      "flos": 29132531619840.0,
      "grad_norm": 2.6068671595959394,
      "language_loss": 0.86325067,
      "learning_rate": 3.687243426879095e-06,
      "loss": 0.88744229,
      "num_input_tokens_seen": 6437520,
      "step": 307,
      "time_per_iteration": 2.5516645908355713
    },
    {
      "auxiliary_loss_clip": 0.01327815,
      "auxiliary_loss_mlp": 0.01108366,
      "balance_loss_clip": 1.09392428,
      "balance_loss_mlp": 1.06037188,
      "epoch": 0.018517961821734555,
      "flos": 19208654167680.0,
      "grad_norm": 4.175295809441147,
      "language_loss": 0.71828777,
      "learning_rate": 3.6893372569634466e-06,
      "loss": 0.74264956,
      "num_input_tokens_seen": 6455680,
      "step": 308,
      "time_per_iteration": 2.5085878372192383
    },
    {
      "auxiliary_loss_clip": 0.01333285,
      "auxiliary_loss_mlp": 0.01103971,
      "balance_loss_clip": 1.09073353,
      "balance_loss_mlp": 1.06005394,
      "epoch": 0.018578085074402523,
      "flos": 19863018184320.0,
      "grad_norm": 2.231224728646827,
      "language_loss": 0.91654509,
      "learning_rate": 3.6914242999043395e-06,
      "loss": 0.94091773,
      "num_input_tokens_seen": 6474880,
      "step": 309,
      "time_per_iteration": 2.5095369815826416
    },
    {
      "auxiliary_loss_clip": 0.01341962,
      "auxiliary_loss_mlp": 0.01107076,
      "balance_loss_clip": 1.09292269,
      "balance_loss_mlp": 1.06096554,
      "epoch": 0.018638208327070496,
      "flos": 29606947476480.0,
      "grad_norm": 1.8483677204587343,
      "language_loss": 0.72587085,
      "learning_rate": 3.69350459956065e-06,
      "loss": 0.75036126,
      "num_input_tokens_seen": 6495945,
      "step": 310,
      "time_per_iteration": 2.5712952613830566
    },
    {
      "auxiliary_loss_clip": 0.01332616,
      "auxiliary_loss_mlp": 0.01111299,
      "balance_loss_clip": 1.09699655,
      "balance_loss_mlp": 1.06673861,
      "epoch": 0.018698331579738464,
      "flos": 45731555907840.0,
      "grad_norm": 2.1563336998008786,
      "language_loss": 0.74176562,
      "learning_rate": 3.695578199367497e-06,
      "loss": 0.76620477,
      "num_input_tokens_seen": 6519930,
      "step": 311,
      "time_per_iteration": 2.712759017944336
    },
    {
      "auxiliary_loss_clip": 0.01339766,
      "auxiliary_loss_mlp": 0.01116196,
      "balance_loss_clip": 1.09383547,
      "balance_loss_mlp": 1.07249331,
      "epoch": 0.018758454832406433,
      "flos": 20483662308480.0,
      "grad_norm": 2.5542300347140925,
      "language_loss": 0.9172585,
      "learning_rate": 3.6976451423416825e-06,
      "loss": 0.94181812,
      "num_input_tokens_seen": 6535070,
      "step": 312,
      "time_per_iteration": 2.4732766151428223
    },
    {
      "auxiliary_loss_clip": 0.01339407,
      "auxiliary_loss_mlp": 0.01119684,
      "balance_loss_clip": 1.0953604,
      "balance_loss_mlp": 1.07390773,
      "epoch": 0.0188185780850744,
      "flos": 15777784661760.0,
      "grad_norm": 2.161492194659412,
      "language_loss": 0.89793909,
      "learning_rate": 3.699705471087043e-06,
      "loss": 0.92252994,
      "num_input_tokens_seen": 6554135,
      "step": 313,
      "time_per_iteration": 2.510206937789917
    },
    {
      "auxiliary_loss_clip": 0.01341727,
      "auxiliary_loss_mlp": 0.01103901,
      "balance_loss_clip": 1.09423101,
      "balance_loss_mlp": 1.05740905,
      "epoch": 0.018878701337742373,
      "flos": 22455732758400.0,
      "grad_norm": 1.9300042227099097,
      "language_loss": 0.72875816,
      "learning_rate": 3.7017592277997256e-06,
      "loss": 0.75321436,
      "num_input_tokens_seen": 6572275,
      "step": 314,
      "time_per_iteration": 2.526021718978882
    },
    {
      "auxiliary_loss_clip": 0.01329749,
      "auxiliary_loss_mlp": 0.01106551,
      "balance_loss_clip": 1.09181952,
      "balance_loss_mlp": 1.06313467,
      "epoch": 0.018938824590410342,
      "flos": 30993530238720.0,
      "grad_norm": 2.6946247362473437,
      "language_loss": 0.8969605,
      "learning_rate": 3.7038064542733654e-06,
      "loss": 0.92132354,
      "num_input_tokens_seen": 6594520,
      "step": 315,
      "time_per_iteration": 2.5963077545166016
    },
    {
      "auxiliary_loss_clip": 0.01331882,
      "auxiliary_loss_mlp": 0.01096604,
      "balance_loss_clip": 1.09381747,
      "balance_loss_mlp": 1.05237722,
      "epoch": 0.01899894784307831,
      "flos": 23258910821760.0,
      "grad_norm": 3.584483433203057,
      "language_loss": 0.80503726,
      "learning_rate": 3.7058471919041945e-06,
      "loss": 0.8293221,
      "num_input_tokens_seen": 6614245,
      "step": 316,
      "time_per_iteration": 2.5061070919036865
    },
    {
      "auxiliary_loss_clip": 0.01326129,
      "auxiliary_loss_mlp": 0.01094618,
      "balance_loss_clip": 1.09128261,
      "balance_loss_mlp": 1.05172658,
      "epoch": 0.01905907109574628,
      "flos": 17457901367040.0,
      "grad_norm": 2.0691252717151656,
      "language_loss": 0.90178567,
      "learning_rate": 3.7078814816960605e-06,
      "loss": 0.9259932,
      "num_input_tokens_seen": 6632015,
      "step": 317,
      "time_per_iteration": 2.488750696182251
    },
    {
      "auxiliary_loss_clip": 0.01322913,
      "auxiliary_loss_mlp": 0.01097754,
      "balance_loss_clip": 1.08982813,
      "balance_loss_mlp": 1.05278862,
      "epoch": 0.019119194348414248,
      "flos": 14970225139200.0,
      "grad_norm": 2.4249503565227197,
      "language_loss": 0.90965402,
      "learning_rate": 3.709909364265374e-06,
      "loss": 0.93386078,
      "num_input_tokens_seen": 6649015,
      "step": 318,
      "time_per_iteration": 2.468766212463379
    },
    {
      "auxiliary_loss_clip": 0.01325631,
      "auxiliary_loss_mlp": 0.01091806,
      "balance_loss_clip": 1.08999777,
      "balance_loss_mlp": 1.04984403,
      "epoch": 0.01917931760108222,
      "flos": 25482822503040.0,
      "grad_norm": 2.37059658156526,
      "language_loss": 0.93873966,
      "learning_rate": 3.7119308798459706e-06,
      "loss": 0.96291411,
      "num_input_tokens_seen": 6669225,
      "step": 319,
      "time_per_iteration": 2.542235851287842
    },
    {
      "auxiliary_loss_clip": 0.01208177,
      "auxiliary_loss_mlp": 0.01047628,
      "balance_loss_clip": 1.08579254,
      "balance_loss_mlp": 1.03418112,
      "epoch": 0.01923944085375019,
      "flos": 71556967353600.0,
      "grad_norm": 0.9394075574673466,
      "language_loss": 0.59889984,
      "learning_rate": 3.7139460682939026e-06,
      "loss": 0.62145793,
      "num_input_tokens_seen": 6725775,
      "step": 320,
      "time_per_iteration": 3.035402297973633
    },
    {
      "auxiliary_loss_clip": 0.01324748,
      "auxiliary_loss_mlp": 0.01103163,
      "balance_loss_clip": 1.09002709,
      "balance_loss_mlp": 1.06065321,
      "epoch": 0.019299564106418157,
      "flos": 19682495406720.0,
      "grad_norm": 2.6129098674880686,
      "language_loss": 0.89904606,
      "learning_rate": 3.715954969092154e-06,
      "loss": 0.9233253,
      "num_input_tokens_seen": 6744170,
      "step": 321,
      "time_per_iteration": 2.496229887008667
    },
    {
      "auxiliary_loss_clip": 0.01334418,
      "auxiliary_loss_mlp": 0.01118722,
      "balance_loss_clip": 1.09399271,
      "balance_loss_mlp": 1.07387519,
      "epoch": 0.019359687359086126,
      "flos": 24387151991040.0,
      "grad_norm": 2.3717172801585624,
      "language_loss": 0.82467294,
      "learning_rate": 3.7179576213552805e-06,
      "loss": 0.8492043,
      "num_input_tokens_seen": 6764565,
      "step": 322,
      "time_per_iteration": 2.563275098800659
    },
    {
      "auxiliary_loss_clip": 0.0133429,
      "auxiliary_loss_mlp": 0.0109176,
      "balance_loss_clip": 1.09193921,
      "balance_loss_mlp": 1.05008447,
      "epoch": 0.019419810611754094,
      "flos": 23951376190080.0,
      "grad_norm": 2.1752211565318262,
      "language_loss": 0.72402394,
      "learning_rate": 3.719954063833981e-06,
      "loss": 0.74828446,
      "num_input_tokens_seen": 6785310,
      "step": 323,
      "time_per_iteration": 2.523047924041748
    },
    {
      "auxiliary_loss_clip": 0.01322281,
      "auxiliary_loss_mlp": 0.01090866,
      "balance_loss_clip": 1.08683538,
      "balance_loss_mlp": 1.0484277,
      "epoch": 0.019479933864422067,
      "flos": 22160223567360.0,
      "grad_norm": 1.9307618113805254,
      "language_loss": 0.92239487,
      "learning_rate": 3.721944334919596e-06,
      "loss": 0.94652629,
      "num_input_tokens_seen": 6803290,
      "step": 324,
      "time_per_iteration": 2.5241775512695312
    },
    {
      "auxiliary_loss_clip": 0.01330676,
      "auxiliary_loss_mlp": 0.01088987,
      "balance_loss_clip": 1.09341323,
      "balance_loss_mlp": 1.04790807,
      "epoch": 0.019540057117090035,
      "flos": 22236821320320.0,
      "grad_norm": 3.3302843913081985,
      "language_loss": 0.65465653,
      "learning_rate": 3.7239284726485375e-06,
      "loss": 0.67885315,
      "num_input_tokens_seen": 6822570,
      "step": 325,
      "time_per_iteration": 2.4959139823913574
    },
    {
      "auxiliary_loss_clip": 0.01329969,
      "auxiliary_loss_mlp": 0.0110512,
      "balance_loss_clip": 1.09819543,
      "balance_loss_mlp": 1.06182337,
      "epoch": 0.019600180369758004,
      "flos": 23076771932160.0,
      "grad_norm": 13.386541323931564,
      "language_loss": 0.76534313,
      "learning_rate": 3.72590651470665e-06,
      "loss": 0.78969401,
      "num_input_tokens_seen": 6841910,
      "step": 326,
      "time_per_iteration": 2.5278851985931396
    },
    {
      "auxiliary_loss_clip": 0.01322459,
      "auxiliary_loss_mlp": 0.01104874,
      "balance_loss_clip": 1.09264433,
      "balance_loss_mlp": 1.06112385,
      "epoch": 0.019660303622425972,
      "flos": 25410857604480.0,
      "grad_norm": 2.2853184985182184,
      "language_loss": 0.79478103,
      "learning_rate": 3.727878498433505e-06,
      "loss": 0.81905437,
      "num_input_tokens_seen": 6862480,
      "step": 327,
      "time_per_iteration": 2.5820560455322266
    },
    {
      "auxiliary_loss_clip": 0.01331935,
      "auxiliary_loss_mlp": 0.01110933,
      "balance_loss_clip": 1.09449315,
      "balance_loss_mlp": 1.0684706,
      "epoch": 0.01972042687509394,
      "flos": 23657519024640.0,
      "grad_norm": 2.140462945493788,
      "language_loss": 0.80812597,
      "learning_rate": 3.7298444608266328e-06,
      "loss": 0.8325547,
      "num_input_tokens_seen": 6882015,
      "step": 328,
      "time_per_iteration": 2.542212963104248
    },
    {
      "auxiliary_loss_clip": 0.01327113,
      "auxiliary_loss_mlp": 0.01096352,
      "balance_loss_clip": 1.08755779,
      "balance_loss_mlp": 1.05248344,
      "epoch": 0.019780550127761913,
      "flos": 18223480869120.0,
      "grad_norm": 2.2872169302560157,
      "language_loss": 0.93529749,
      "learning_rate": 3.731804438545683e-06,
      "loss": 0.95953214,
      "num_input_tokens_seen": 6899785,
      "step": 329,
      "time_per_iteration": 2.4751827716827393
    },
    {
      "auxiliary_loss_clip": 0.01334545,
      "auxiliary_loss_mlp": 0.01110017,
      "balance_loss_clip": 1.09319758,
      "balance_loss_mlp": 1.06681561,
      "epoch": 0.01984067338042988,
      "flos": 22418780641920.0,
      "grad_norm": 3.3340798863750694,
      "language_loss": 0.74575531,
      "learning_rate": 3.7337584679165324e-06,
      "loss": 0.77020097,
      "num_input_tokens_seen": 6918575,
      "step": 330,
      "time_per_iteration": 3.8920044898986816
    },
    {
      "auxiliary_loss_clip": 0.01331179,
      "auxiliary_loss_mlp": 0.01117037,
      "balance_loss_clip": 1.0919801,
      "balance_loss_mlp": 1.07364511,
      "epoch": 0.01990079663309785,
      "flos": 17055199013760.0,
      "grad_norm": 4.940256493397481,
      "language_loss": 0.93349528,
      "learning_rate": 3.7357065849353186e-06,
      "loss": 0.95797753,
      "num_input_tokens_seen": 6936965,
      "step": 331,
      "time_per_iteration": 3.857365846633911
    },
    {
      "auxiliary_loss_clip": 0.01318373,
      "auxiliary_loss_mlp": 0.01088302,
      "balance_loss_clip": 1.08976138,
      "balance_loss_mlp": 1.04741287,
      "epoch": 0.01996091988576582,
      "flos": 15961791058560.0,
      "grad_norm": 2.1021148078879697,
      "language_loss": 0.92604125,
      "learning_rate": 3.737648825272422e-06,
      "loss": 0.95010793,
      "num_input_tokens_seen": 6953475,
      "step": 332,
      "time_per_iteration": 2.4573404788970947
    },
    {
      "auxiliary_loss_clip": 0.01325031,
      "auxiliary_loss_mlp": 0.01092416,
      "balance_loss_clip": 1.09436011,
      "balance_loss_mlp": 1.04938102,
      "epoch": 0.02002104313843379,
      "flos": 23586451966080.0,
      "grad_norm": 2.9250033335138896,
      "language_loss": 0.75577587,
      "learning_rate": 3.739585224276384e-06,
      "loss": 0.77995026,
      "num_input_tokens_seen": 6971630,
      "step": 333,
      "time_per_iteration": 3.9254531860351562
    },
    {
      "auxiliary_loss_clip": 0.01327376,
      "auxiliary_loss_mlp": 0.01087271,
      "balance_loss_clip": 1.09201992,
      "balance_loss_mlp": 1.04611969,
      "epoch": 0.02008116639110176,
      "flos": 34094883352320.0,
      "grad_norm": 2.1965969866719397,
      "language_loss": 0.78864884,
      "learning_rate": 3.7415158169777673e-06,
      "loss": 0.8127954,
      "num_input_tokens_seen": 6992775,
      "step": 334,
      "time_per_iteration": 2.599663734436035
    },
    {
      "auxiliary_loss_clip": 0.0132656,
      "auxiliary_loss_mlp": 0.01097045,
      "balance_loss_clip": 1.08734179,
      "balance_loss_mlp": 1.05312824,
      "epoch": 0.020141289643769728,
      "flos": 19683716469120.0,
      "grad_norm": 1.8804950161494818,
      "language_loss": 0.83226275,
      "learning_rate": 3.7434406380929575e-06,
      "loss": 0.85649884,
      "num_input_tokens_seen": 7011425,
      "step": 335,
      "time_per_iteration": 2.5641684532165527
    },
    {
      "auxiliary_loss_clip": 0.01322965,
      "auxiliary_loss_mlp": 0.01088119,
      "balance_loss_clip": 1.08998549,
      "balance_loss_mlp": 1.04610991,
      "epoch": 0.020201412896437697,
      "flos": 20740567357440.0,
      "grad_norm": 2.8221465495189912,
      "language_loss": 0.92273647,
      "learning_rate": 3.745359722027911e-06,
      "loss": 0.94684732,
      "num_input_tokens_seen": 7029450,
      "step": 336,
      "time_per_iteration": 2.520214080810547
    },
    {
      "auxiliary_loss_clip": 0.01322602,
      "auxiliary_loss_mlp": 0.01082996,
      "balance_loss_clip": 1.08685946,
      "balance_loss_mlp": 1.04158306,
      "epoch": 0.020261536149105665,
      "flos": 20266510636800.0,
      "grad_norm": 2.3302545987054275,
      "language_loss": 0.88378799,
      "learning_rate": 3.7472731028818428e-06,
      "loss": 0.90784395,
      "num_input_tokens_seen": 7047555,
      "step": 337,
      "time_per_iteration": 2.533212900161743
    },
    {
      "auxiliary_loss_clip": 0.01313623,
      "auxiliary_loss_mlp": 0.01102207,
      "balance_loss_clip": 1.08497429,
      "balance_loss_mlp": 1.05867171,
      "epoch": 0.020321659401773638,
      "flos": 25848752307840.0,
      "grad_norm": 1.5362255420478748,
      "language_loss": 0.89893103,
      "learning_rate": 3.7491808144508626e-06,
      "loss": 0.92308939,
      "num_input_tokens_seen": 7068185,
      "step": 338,
      "time_per_iteration": 2.603816270828247
    },
    {
      "auxiliary_loss_clip": 0.01324362,
      "auxiliary_loss_mlp": 0.01100494,
      "balance_loss_clip": 1.08938348,
      "balance_loss_mlp": 1.05800724,
      "epoch": 0.020381782654441606,
      "flos": 17495033051520.0,
      "grad_norm": 2.3547377437641495,
      "language_loss": 0.85036457,
      "learning_rate": 3.7510828902315576e-06,
      "loss": 0.87461311,
      "num_input_tokens_seen": 7085955,
      "step": 339,
      "time_per_iteration": 2.526341199874878
    },
    {
      "auxiliary_loss_clip": 0.01328845,
      "auxiliary_loss_mlp": 0.01099023,
      "balance_loss_clip": 1.0922718,
      "balance_loss_mlp": 1.05632198,
      "epoch": 0.020441905907109575,
      "flos": 24243940465920.0,
      "grad_norm": 2.8083599186060364,
      "language_loss": 0.88858235,
      "learning_rate": 3.75297936342452e-06,
      "loss": 0.91286105,
      "num_input_tokens_seen": 7106345,
      "step": 340,
      "time_per_iteration": 2.543834686279297
    },
    {
      "auxiliary_loss_clip": 0.01323876,
      "auxiliary_loss_mlp": 0.0108679,
      "balance_loss_clip": 1.08833981,
      "balance_loss_mlp": 1.04251611,
      "epoch": 0.020502029159777543,
      "flos": 22233301787520.0,
      "grad_norm": 4.507233545455789,
      "language_loss": 0.88281071,
      "learning_rate": 3.7548702669378253e-06,
      "loss": 0.90691733,
      "num_input_tokens_seen": 7125070,
      "step": 341,
      "time_per_iteration": 2.5411481857299805
    },
    {
      "auxiliary_loss_clip": 0.01326993,
      "auxiliary_loss_mlp": 0.01102801,
      "balance_loss_clip": 1.08800399,
      "balance_loss_mlp": 1.05990982,
      "epoch": 0.020562152412445512,
      "flos": 23987861429760.0,
      "grad_norm": 2.9408702366544794,
      "language_loss": 0.80213362,
      "learning_rate": 3.756755633390458e-06,
      "loss": 0.82643157,
      "num_input_tokens_seen": 7144675,
      "step": 342,
      "time_per_iteration": 2.560076951980591
    },
    {
      "auxiliary_loss_clip": 0.01315378,
      "auxiliary_loss_mlp": 0.01097228,
      "balance_loss_clip": 1.08657849,
      "balance_loss_mlp": 1.05252409,
      "epoch": 0.020622275665113484,
      "flos": 26975305537920.0,
      "grad_norm": 2.4694771969817886,
      "language_loss": 0.89687574,
      "learning_rate": 3.7586354951156886e-06,
      "loss": 0.92100179,
      "num_input_tokens_seen": 7165505,
      "step": 343,
      "time_per_iteration": 2.5710017681121826
    },
    {
      "auxiliary_loss_clip": 0.01326274,
      "auxiliary_loss_mlp": 0.01096059,
      "balance_loss_clip": 1.09418142,
      "balance_loss_mlp": 1.05555153,
      "epoch": 0.020682398917781453,
      "flos": 22600704049920.0,
      "grad_norm": 1.9578545417081916,
      "language_loss": 0.78210926,
      "learning_rate": 3.7605098841644e-06,
      "loss": 0.80633259,
      "num_input_tokens_seen": 7184605,
      "step": 344,
      "time_per_iteration": 2.5151400566101074
    },
    {
      "auxiliary_loss_clip": 0.01311232,
      "auxiliary_loss_mlp": 0.01102438,
      "balance_loss_clip": 1.08548689,
      "balance_loss_mlp": 1.05868864,
      "epoch": 0.02074252217044942,
      "flos": 15013605790080.0,
      "grad_norm": 1.856941840949793,
      "language_loss": 0.74990153,
      "learning_rate": 3.7623788323083666e-06,
      "loss": 0.77403826,
      "num_input_tokens_seen": 7203065,
      "step": 345,
      "time_per_iteration": 2.4936389923095703
    },
    {
      "auxiliary_loss_clip": 0.01318657,
      "auxiliary_loss_mlp": 0.01102253,
      "balance_loss_clip": 1.09153533,
      "balance_loss_mlp": 1.05962384,
      "epoch": 0.02080264542311739,
      "flos": 25337958952320.0,
      "grad_norm": 2.1365767476921906,
      "language_loss": 0.90456986,
      "learning_rate": 3.7642423710434837e-06,
      "loss": 0.92877901,
      "num_input_tokens_seen": 7222995,
      "step": 346,
      "time_per_iteration": 2.562014579772949
    },
    {
      "auxiliary_loss_clip": 0.01313735,
      "auxiliary_loss_mlp": 0.01093796,
      "balance_loss_clip": 1.08558679,
      "balance_loss_mlp": 1.05452871,
      "epoch": 0.02086276867578536,
      "flos": 24388804016640.0,
      "grad_norm": 1.9901221156972229,
      "language_loss": 0.7887423,
      "learning_rate": 3.7661005315929563e-06,
      "loss": 0.81281763,
      "num_input_tokens_seen": 7244625,
      "step": 347,
      "time_per_iteration": 2.5340518951416016
    },
    {
      "auxiliary_loss_clip": 0.01319291,
      "auxiliary_loss_mlp": 0.01098395,
      "balance_loss_clip": 1.09145069,
      "balance_loss_mlp": 1.05533695,
      "epoch": 0.02092289192845333,
      "flos": 24462205459200.0,
      "grad_norm": 1.8115406080892402,
      "language_loss": 0.71126294,
      "learning_rate": 3.7679533449104354e-06,
      "loss": 0.7354399,
      "num_input_tokens_seen": 7263255,
      "step": 348,
      "time_per_iteration": 2.5632107257843018
    },
    {
      "auxiliary_loss_clip": 0.01321399,
      "auxiliary_loss_mlp": 0.01102718,
      "balance_loss_clip": 1.0878005,
      "balance_loss_mlp": 1.05977845,
      "epoch": 0.0209830151811213,
      "flos": 17451185523840.0,
      "grad_norm": 2.6166965018956314,
      "language_loss": 0.76650506,
      "learning_rate": 3.7698008416831116e-06,
      "loss": 0.79074621,
      "num_input_tokens_seen": 7279275,
      "step": 349,
      "time_per_iteration": 2.469172477722168
    },
    {
      "auxiliary_loss_clip": 0.01306461,
      "auxiliary_loss_mlp": 0.01103875,
      "balance_loss_clip": 1.0866605,
      "balance_loss_mlp": 1.06250906,
      "epoch": 0.021043138433789268,
      "flos": 24573995562240.0,
      "grad_norm": 1.6622499677827813,
      "language_loss": 0.85162234,
      "learning_rate": 3.7716430523347664e-06,
      "loss": 0.87572575,
      "num_input_tokens_seen": 7300180,
      "step": 350,
      "time_per_iteration": 2.5673046112060547
    },
    {
      "auxiliary_loss_clip": 0.01314907,
      "auxiliary_loss_mlp": 0.01090451,
      "balance_loss_clip": 1.09089875,
      "balance_loss_mlp": 1.05115986,
      "epoch": 0.021103261686457236,
      "flos": 24454053072000.0,
      "grad_norm": 2.1384174562302634,
      "language_loss": 0.79781115,
      "learning_rate": 3.773480007028776e-06,
      "loss": 0.82186472,
      "num_input_tokens_seen": 7317430,
      "step": 351,
      "time_per_iteration": 2.4930734634399414
    },
    {
      "auxiliary_loss_clip": 0.0132186,
      "auxiliary_loss_mlp": 0.01104357,
      "balance_loss_clip": 1.09109211,
      "balance_loss_mlp": 1.06144142,
      "epoch": 0.021163384939125205,
      "flos": 14683083816960.0,
      "grad_norm": 2.639562652236807,
      "language_loss": 0.87569571,
      "learning_rate": 3.775311735671078e-06,
      "loss": 0.89995795,
      "num_input_tokens_seen": 7334875,
      "step": 352,
      "time_per_iteration": 2.4874041080474854
    },
    {
      "auxiliary_loss_clip": 0.01314178,
      "auxiliary_loss_mlp": 0.01103263,
      "balance_loss_clip": 1.08946502,
      "balance_loss_mlp": 1.06001389,
      "epoch": 0.021223508191793177,
      "flos": 24493195918080.0,
      "grad_norm": 1.8311512367302993,
      "language_loss": 0.82362443,
      "learning_rate": 3.7771382679130878e-06,
      "loss": 0.84779876,
      "num_input_tokens_seen": 7355185,
      "step": 353,
      "time_per_iteration": 2.551184892654419
    },
    {
      "auxiliary_loss_clip": 0.01311988,
      "auxiliary_loss_mlp": 0.01097338,
      "balance_loss_clip": 1.08867431,
      "balance_loss_mlp": 1.0570451,
      "epoch": 0.021283631444461146,
      "flos": 24126978804480.0,
      "grad_norm": 2.760516330628348,
      "language_loss": 0.81079543,
      "learning_rate": 3.7789596331545845e-06,
      "loss": 0.83488864,
      "num_input_tokens_seen": 7374425,
      "step": 354,
      "time_per_iteration": 2.5483357906341553
    },
    {
      "auxiliary_loss_clip": 0.0131645,
      "auxiliary_loss_mlp": 0.01092506,
      "balance_loss_clip": 1.08512473,
      "balance_loss_mlp": 1.04899466,
      "epoch": 0.021343754697129114,
      "flos": 25192233475200.0,
      "grad_norm": 2.033471943140301,
      "language_loss": 0.81049144,
      "learning_rate": 3.780775860546545e-06,
      "loss": 0.83458096,
      "num_input_tokens_seen": 7394175,
      "step": 355,
      "time_per_iteration": 2.531355142593384
    },
    {
      "auxiliary_loss_clip": 0.01313725,
      "auxiliary_loss_mlp": 0.01090674,
      "balance_loss_clip": 1.08578539,
      "balance_loss_mlp": 1.048998,
      "epoch": 0.021403877949797083,
      "flos": 17274182279040.0,
      "grad_norm": 2.1836311633094274,
      "language_loss": 0.89332211,
      "learning_rate": 3.7825869789939474e-06,
      "loss": 0.91736603,
      "num_input_tokens_seen": 7412645,
      "step": 356,
      "time_per_iteration": 2.487934112548828
    },
    {
      "auxiliary_loss_clip": 0.01310702,
      "auxiliary_loss_mlp": 0.01085635,
      "balance_loss_clip": 1.0877552,
      "balance_loss_mlp": 1.04314899,
      "epoch": 0.021464001202465055,
      "flos": 30917435276160.0,
      "grad_norm": 1.9999570316142263,
      "language_loss": 0.8024286,
      "learning_rate": 3.784393017158528e-06,
      "loss": 0.82639205,
      "num_input_tokens_seen": 7432275,
      "step": 357,
      "time_per_iteration": 2.559035062789917
    },
    {
      "auxiliary_loss_clip": 0.01312653,
      "auxiliary_loss_mlp": 0.01085897,
      "balance_loss_clip": 1.08524656,
      "balance_loss_mlp": 1.04705894,
      "epoch": 0.021524124455133024,
      "flos": 18186385098240.0,
      "grad_norm": 2.3131962209489867,
      "language_loss": 0.76576149,
      "learning_rate": 3.786194003461506e-06,
      "loss": 0.78974694,
      "num_input_tokens_seen": 7450245,
      "step": 358,
      "time_per_iteration": 2.481630325317383
    },
    {
      "auxiliary_loss_clip": 0.01310087,
      "auxiliary_loss_mlp": 0.01091409,
      "balance_loss_clip": 1.08329225,
      "balance_loss_mlp": 1.04837501,
      "epoch": 0.021584247707800992,
      "flos": 13805786039040.0,
      "grad_norm": 2.7019611385348297,
      "language_loss": 0.88065517,
      "learning_rate": 3.787989966086264e-06,
      "loss": 0.90467018,
      "num_input_tokens_seen": 7466845,
      "step": 359,
      "time_per_iteration": 2.4632513523101807
    },
    {
      "auxiliary_loss_clip": 0.01317968,
      "auxiliary_loss_mlp": 0.0108999,
      "balance_loss_clip": 1.08885562,
      "balance_loss_mlp": 1.05041277,
      "epoch": 0.02164437096046896,
      "flos": 23294713703040.0,
      "grad_norm": 2.829075214936934,
      "language_loss": 0.75971377,
      "learning_rate": 3.789780932980997e-06,
      "loss": 0.78379339,
      "num_input_tokens_seen": 7485450,
      "step": 360,
      "time_per_iteration": 2.520082473754883
    },
    {
      "auxiliary_loss_clip": 0.0119617,
      "auxiliary_loss_mlp": 0.0104815,
      "balance_loss_clip": 1.07889986,
      "balance_loss_mlp": 1.03508461,
      "epoch": 0.02170449421313693,
      "flos": 68899578341760.0,
      "grad_norm": 0.9443858592817307,
      "language_loss": 0.64945179,
      "learning_rate": 3.79156693186132e-06,
      "loss": 0.67189497,
      "num_input_tokens_seen": 7553780,
      "step": 361,
      "time_per_iteration": 3.2192747592926025
    },
    {
      "auxiliary_loss_clip": 0.01309203,
      "auxiliary_loss_mlp": 0.01088367,
      "balance_loss_clip": 1.08199942,
      "balance_loss_mlp": 1.04855132,
      "epoch": 0.0217646174658049,
      "flos": 25228539146880.0,
      "grad_norm": 3.0445687010131754,
      "language_loss": 0.78258491,
      "learning_rate": 3.7933479902128433e-06,
      "loss": 0.80656058,
      "num_input_tokens_seen": 7574155,
      "step": 362,
      "time_per_iteration": 2.5585684776306152
    },
    {
      "auxiliary_loss_clip": 0.01311552,
      "auxiliary_loss_mlp": 0.01094549,
      "balance_loss_clip": 1.08443356,
      "balance_loss_mlp": 1.05444729,
      "epoch": 0.02182474071847287,
      "flos": 22893124671360.0,
      "grad_norm": 2.0296218972829916,
      "language_loss": 0.92699635,
      "learning_rate": 3.7951241352937077e-06,
      "loss": 0.95105743,
      "num_input_tokens_seen": 7592320,
      "step": 363,
      "time_per_iteration": 2.4929096698760986
    },
    {
      "auxiliary_loss_clip": 0.01307803,
      "auxiliary_loss_mlp": 0.0110003,
      "balance_loss_clip": 1.08417583,
      "balance_loss_mlp": 1.06114376,
      "epoch": 0.02188486397114084,
      "flos": 23658991482240.0,
      "grad_norm": 2.3473632528503794,
      "language_loss": 0.89641362,
      "learning_rate": 3.7968953941370915e-06,
      "loss": 0.92049193,
      "num_input_tokens_seen": 7611185,
      "step": 364,
      "time_per_iteration": 2.5262603759765625
    },
    {
      "auxiliary_loss_clip": 0.01313328,
      "auxiliary_loss_mlp": 0.01094652,
      "balance_loss_clip": 1.08764839,
      "balance_loss_mlp": 1.05285728,
      "epoch": 0.021944987223808807,
      "flos": 21543637680000.0,
      "grad_norm": 1.9697107324262233,
      "language_loss": 0.79484671,
      "learning_rate": 3.798661793553676e-06,
      "loss": 0.81892651,
      "num_input_tokens_seen": 7631970,
      "step": 365,
      "time_per_iteration": 2.4976956844329834
    },
    {
      "auxiliary_loss_clip": 0.0130722,
      "auxiliary_loss_mlp": 0.01095163,
      "balance_loss_clip": 1.08438301,
      "balance_loss_mlp": 1.05274868,
      "epoch": 0.022005110476476776,
      "flos": 16070887641600.0,
      "grad_norm": 1.8231594100674484,
      "language_loss": 0.84438133,
      "learning_rate": 3.8004233601340808e-06,
      "loss": 0.8684051,
      "num_input_tokens_seen": 7649745,
      "step": 366,
      "time_per_iteration": 2.5000321865081787
    },
    {
      "auxiliary_loss_clip": 0.01313517,
      "auxiliary_loss_mlp": 0.01090237,
      "balance_loss_clip": 1.08552587,
      "balance_loss_mlp": 1.05199432,
      "epoch": 0.022065233729144748,
      "flos": 21433715084160.0,
      "grad_norm": 1.844294205659595,
      "language_loss": 0.86911905,
      "learning_rate": 3.8021801202512694e-06,
      "loss": 0.89315659,
      "num_input_tokens_seen": 7668830,
      "step": 367,
      "time_per_iteration": 2.4984028339385986
    },
    {
      "auxiliary_loss_clip": 0.01314941,
      "auxiliary_loss_mlp": 0.01097101,
      "balance_loss_clip": 1.0840205,
      "balance_loss_mlp": 1.05552065,
      "epoch": 0.022125356981812717,
      "flos": 21543709507200.0,
      "grad_norm": 2.166900845087149,
      "language_loss": 0.84954393,
      "learning_rate": 3.803932100062912e-06,
      "loss": 0.87366438,
      "num_input_tokens_seen": 7687240,
      "step": 368,
      "time_per_iteration": 3.8960788249969482
    },
    {
      "auxiliary_loss_clip": 0.01312568,
      "auxiliary_loss_mlp": 0.01085779,
      "balance_loss_clip": 1.08075666,
      "balance_loss_mlp": 1.04622507,
      "epoch": 0.022185480234480685,
      "flos": 20704153944960.0,
      "grad_norm": 4.5238428897406475,
      "language_loss": 0.75726801,
      "learning_rate": 3.8056793255137264e-06,
      "loss": 0.78125143,
      "num_input_tokens_seen": 7704440,
      "step": 369,
      "time_per_iteration": 3.908716917037964
    },
    {
      "auxiliary_loss_clip": 0.01307742,
      "auxiliary_loss_mlp": 0.01102635,
      "balance_loss_clip": 1.08364892,
      "balance_loss_mlp": 1.0634867,
      "epoch": 0.022245603487148654,
      "flos": 25193203142400.0,
      "grad_norm": 2.1716404296819607,
      "language_loss": 0.8283217,
      "learning_rate": 3.8074218223377844e-06,
      "loss": 0.85242546,
      "num_input_tokens_seen": 7727160,
      "step": 370,
      "time_per_iteration": 3.9636473655700684
    },
    {
      "auxiliary_loss_clip": 0.01306043,
      "auxiliary_loss_mlp": 0.01097403,
      "balance_loss_clip": 1.08258677,
      "balance_loss_mlp": 1.05722928,
      "epoch": 0.022305726739816623,
      "flos": 21395936954880.0,
      "grad_norm": 3.290609403606036,
      "language_loss": 0.81994444,
      "learning_rate": 3.8091596160607834e-06,
      "loss": 0.84397888,
      "num_input_tokens_seen": 7747730,
      "step": 371,
      "time_per_iteration": 2.505772113800049
    },
    {
      "auxiliary_loss_clip": 0.01312886,
      "auxiliary_loss_mlp": 0.010942,
      "balance_loss_clip": 1.08814073,
      "balance_loss_mlp": 1.05288219,
      "epoch": 0.022365849992484595,
      "flos": 22492146170880.0,
      "grad_norm": 2.0477414870826864,
      "language_loss": 0.83611,
      "learning_rate": 3.8108927320022896e-06,
      "loss": 0.86018085,
      "num_input_tokens_seen": 7766765,
      "step": 372,
      "time_per_iteration": 2.5135765075683594
    },
    {
      "auxiliary_loss_clip": 0.01306217,
      "auxiliary_loss_mlp": 0.01091586,
      "balance_loss_clip": 1.08448815,
      "balance_loss_mlp": 1.05112648,
      "epoch": 0.022425973245152563,
      "flos": 17856581397120.0,
      "grad_norm": 4.033468741804688,
      "language_loss": 0.78865242,
      "learning_rate": 3.8126211952779548e-06,
      "loss": 0.81263047,
      "num_input_tokens_seen": 7784010,
      "step": 373,
      "time_per_iteration": 3.9449522495269775
    },
    {
      "auxiliary_loss_clip": 0.01310325,
      "auxiliary_loss_mlp": 0.01089303,
      "balance_loss_clip": 1.08591986,
      "balance_loss_mlp": 1.04769862,
      "epoch": 0.022486096497820532,
      "flos": 15483029656320.0,
      "grad_norm": 2.5650283271208494,
      "language_loss": 0.77905834,
      "learning_rate": 3.8143450308016952e-06,
      "loss": 0.80305457,
      "num_input_tokens_seen": 7801305,
      "step": 374,
      "time_per_iteration": 2.476372718811035
    },
    {
      "auxiliary_loss_clip": 0.01302677,
      "auxiliary_loss_mlp": 0.01078784,
      "balance_loss_clip": 1.07821178,
      "balance_loss_mlp": 1.03739512,
      "epoch": 0.0225462197504885,
      "flos": 27784157950080.0,
      "grad_norm": 1.9737072839295156,
      "language_loss": 0.86098623,
      "learning_rate": 3.8160642632878525e-06,
      "loss": 0.88480085,
      "num_input_tokens_seen": 7823965,
      "step": 375,
      "time_per_iteration": 2.568939208984375
    },
    {
      "auxiliary_loss_clip": 0.01308314,
      "auxiliary_loss_mlp": 0.01100359,
      "balance_loss_clip": 1.08594131,
      "balance_loss_mlp": 1.05839717,
      "epoch": 0.02260634300315647,
      "flos": 19975490645760.0,
      "grad_norm": 2.238733232050167,
      "language_loss": 0.88854146,
      "learning_rate": 3.817778917253314e-06,
      "loss": 0.91262823,
      "num_input_tokens_seen": 7842115,
      "step": 376,
      "time_per_iteration": 2.514871120452881
    },
    {
      "auxiliary_loss_clip": 0.01307735,
      "auxiliary_loss_mlp": 0.01084168,
      "balance_loss_clip": 1.08049881,
      "balance_loss_mlp": 1.04542482,
      "epoch": 0.02266646625582444,
      "flos": 16028189349120.0,
      "grad_norm": 2.1383514500195613,
      "language_loss": 0.74967128,
      "learning_rate": 3.8194890170196155e-06,
      "loss": 0.77359033,
      "num_input_tokens_seen": 7857830,
      "step": 377,
      "time_per_iteration": 2.4893674850463867
    },
    {
      "auxiliary_loss_clip": 0.01299877,
      "auxiliary_loss_mlp": 0.01095816,
      "balance_loss_clip": 1.08405828,
      "balance_loss_mlp": 1.0550468,
      "epoch": 0.02272658950849241,
      "flos": 20404622430720.0,
      "grad_norm": 2.3126811011248107,
      "language_loss": 0.9934442,
      "learning_rate": 3.8211945867150055e-06,
      "loss": 1.01740122,
      "num_input_tokens_seen": 7875840,
      "step": 378,
      "time_per_iteration": 2.535586357116699
    },
    {
      "auxiliary_loss_clip": 0.0118095,
      "auxiliary_loss_mlp": 0.01062227,
      "balance_loss_clip": 1.06609571,
      "balance_loss_mlp": 1.04992473,
      "epoch": 0.02278671276116038,
      "flos": 69847332647040.0,
      "grad_norm": 1.0027553635834887,
      "language_loss": 0.75430381,
      "learning_rate": 3.822895650276492e-06,
      "loss": 0.77673554,
      "num_input_tokens_seen": 7940190,
      "step": 379,
      "time_per_iteration": 3.14353609085083
    },
    {
      "auxiliary_loss_clip": 0.01309714,
      "auxiliary_loss_mlp": 0.01091818,
      "balance_loss_clip": 1.0808115,
      "balance_loss_mlp": 1.05331302,
      "epoch": 0.022846836013828347,
      "flos": 38508771340800.0,
      "grad_norm": 2.177222785802864,
      "language_loss": 0.783696,
      "learning_rate": 3.824592231451859e-06,
      "loss": 0.8077113,
      "num_input_tokens_seen": 7960840,
      "step": 380,
      "time_per_iteration": 2.6313724517822266
    },
    {
      "auxiliary_loss_clip": 0.01303885,
      "auxiliary_loss_mlp": 0.01086314,
      "balance_loss_clip": 1.08352447,
      "balance_loss_mlp": 1.04768968,
      "epoch": 0.02290695926649632,
      "flos": 20959478795520.0,
      "grad_norm": 2.2643992893925593,
      "language_loss": 0.96738076,
      "learning_rate": 3.826284353801652e-06,
      "loss": 0.99128282,
      "num_input_tokens_seen": 7975500,
      "step": 381,
      "time_per_iteration": 2.4949941635131836
    },
    {
      "auxiliary_loss_clip": 0.0131303,
      "auxiliary_loss_mlp": 0.01088939,
      "balance_loss_clip": 1.08519292,
      "balance_loss_mlp": 1.04940903,
      "epoch": 0.022967082519164288,
      "flos": 24022407335040.0,
      "grad_norm": 2.07957927508733,
      "language_loss": 0.88007009,
      "learning_rate": 3.827972040701142e-06,
      "loss": 0.90408981,
      "num_input_tokens_seen": 7993880,
      "step": 382,
      "time_per_iteration": 2.5108087062835693
    },
    {
      "auxiliary_loss_clip": 0.01305117,
      "auxiliary_loss_mlp": 0.01097925,
      "balance_loss_clip": 1.08524728,
      "balance_loss_mlp": 1.05882406,
      "epoch": 0.023027205771832256,
      "flos": 20997149184000.0,
      "grad_norm": 2.308514563186946,
      "language_loss": 0.84561276,
      "learning_rate": 3.829655315342268e-06,
      "loss": 0.86964315,
      "num_input_tokens_seen": 8012730,
      "step": 383,
      "time_per_iteration": 2.514528274536133
    },
    {
      "auxiliary_loss_clip": 0.01303385,
      "auxiliary_loss_mlp": 0.01112403,
      "balance_loss_clip": 1.08614123,
      "balance_loss_mlp": 1.07435107,
      "epoch": 0.023087329024500225,
      "flos": 21360816432000.0,
      "grad_norm": 2.179193204534442,
      "language_loss": 0.83356506,
      "learning_rate": 3.831334200735543e-06,
      "loss": 0.85772288,
      "num_input_tokens_seen": 8031275,
      "step": 384,
      "time_per_iteration": 2.490983009338379
    },
    {
      "auxiliary_loss_clip": 0.01303931,
      "auxiliary_loss_mlp": 0.01092374,
      "balance_loss_clip": 1.08914554,
      "balance_loss_mlp": 1.05608618,
      "epoch": 0.023147452277168194,
      "flos": 21872435800320.0,
      "grad_norm": 1.7668915939961258,
      "language_loss": 0.88927078,
      "learning_rate": 3.8330087197119426e-06,
      "loss": 0.91323388,
      "num_input_tokens_seen": 8051600,
      "step": 385,
      "time_per_iteration": 2.5229649543762207
    },
    {
      "auxiliary_loss_clip": 0.01307087,
      "auxiliary_loss_mlp": 0.01120547,
      "balance_loss_clip": 1.0859611,
      "balance_loss_mlp": 1.08216166,
      "epoch": 0.023207575529836166,
      "flos": 18916700423040.0,
      "grad_norm": 1.7522415908177935,
      "language_loss": 0.69699174,
      "learning_rate": 3.83467889492477e-06,
      "loss": 0.72126812,
      "num_input_tokens_seen": 8070600,
      "step": 386,
      "time_per_iteration": 2.489877700805664
    },
    {
      "auxiliary_loss_clip": 0.01308004,
      "auxiliary_loss_mlp": 0.01092931,
      "balance_loss_clip": 1.08742118,
      "balance_loss_mlp": 1.05521369,
      "epoch": 0.023267698782504134,
      "flos": 25046005207680.0,
      "grad_norm": 2.0080706504945542,
      "language_loss": 0.87942457,
      "learning_rate": 3.836344748851495e-06,
      "loss": 0.90343398,
      "num_input_tokens_seen": 8090680,
      "step": 387,
      "time_per_iteration": 2.5519814491271973
    },
    {
      "auxiliary_loss_clip": 0.01308927,
      "auxiliary_loss_mlp": 0.01085443,
      "balance_loss_clip": 1.08691835,
      "balance_loss_mlp": 1.04591358,
      "epoch": 0.023327822035172103,
      "flos": 28879217930880.0,
      "grad_norm": 2.7454276149442975,
      "language_loss": 0.83656776,
      "learning_rate": 3.838006303795566e-06,
      "loss": 0.86051142,
      "num_input_tokens_seen": 8114610,
      "step": 388,
      "time_per_iteration": 2.5873231887817383
    },
    {
      "auxiliary_loss_clip": 0.01305076,
      "auxiliary_loss_mlp": 0.01088436,
      "balance_loss_clip": 1.08552885,
      "balance_loss_mlp": 1.05219591,
      "epoch": 0.02338794528784007,
      "flos": 27121533805440.0,
      "grad_norm": 6.0999110916679085,
      "language_loss": 0.94020915,
      "learning_rate": 3.839663581888206e-06,
      "loss": 0.96414423,
      "num_input_tokens_seen": 8133975,
      "step": 389,
      "time_per_iteration": 2.556816816329956
    },
    {
      "auxiliary_loss_clip": 0.01298351,
      "auxiliary_loss_mlp": 0.01084319,
      "balance_loss_clip": 1.08533287,
      "balance_loss_mlp": 1.04595757,
      "epoch": 0.02344806854050804,
      "flos": 21322355944320.0,
      "grad_norm": 2.2801711694030815,
      "language_loss": 0.87858617,
      "learning_rate": 3.841316605090178e-06,
      "loss": 0.90241289,
      "num_input_tokens_seen": 8153570,
      "step": 390,
      "time_per_iteration": 2.5291900634765625
    },
    {
      "auxiliary_loss_clip": 0.01303754,
      "auxiliary_loss_mlp": 0.0109218,
      "balance_loss_clip": 1.08749747,
      "balance_loss_mlp": 1.05639374,
      "epoch": 0.023508191793176012,
      "flos": 24789997998720.0,
      "grad_norm": 2.184041115767674,
      "language_loss": 0.89398754,
      "learning_rate": 3.842965395193529e-06,
      "loss": 0.91794682,
      "num_input_tokens_seen": 8170075,
      "step": 391,
      "time_per_iteration": 2.543987274169922
    },
    {
      "auxiliary_loss_clip": 0.0130094,
      "auxiliary_loss_mlp": 0.01076434,
      "balance_loss_clip": 1.08453941,
      "balance_loss_mlp": 1.03974128,
      "epoch": 0.02356831504584398,
      "flos": 25995375624960.0,
      "grad_norm": 2.0509513442390026,
      "language_loss": 0.8607955,
      "learning_rate": 3.84460997382332e-06,
      "loss": 0.88456917,
      "num_input_tokens_seen": 8190420,
      "step": 392,
      "time_per_iteration": 2.5236165523529053
    },
    {
      "auxiliary_loss_clip": 0.0129821,
      "auxiliary_loss_mlp": 0.01089614,
      "balance_loss_clip": 1.08452177,
      "balance_loss_mlp": 1.05227804,
      "epoch": 0.02362843829851195,
      "flos": 19062461813760.0,
      "grad_norm": 2.061757974902649,
      "language_loss": 0.88949513,
      "learning_rate": 3.8462503624393256e-06,
      "loss": 0.91337335,
      "num_input_tokens_seen": 8208790,
      "step": 393,
      "time_per_iteration": 2.495410680770874
    },
    {
      "auxiliary_loss_clip": 0.01308041,
      "auxiliary_loss_mlp": 0.01108463,
      "balance_loss_clip": 1.08971059,
      "balance_loss_mlp": 1.06881428,
      "epoch": 0.023688561551179918,
      "flos": 16071031296000.0,
      "grad_norm": 1.9445143273333805,
      "language_loss": 0.81455553,
      "learning_rate": 3.84788658233771e-06,
      "loss": 0.83872056,
      "num_input_tokens_seen": 8226885,
      "step": 394,
      "time_per_iteration": 2.471144676208496
    },
    {
      "auxiliary_loss_clip": 0.01299191,
      "auxiliary_loss_mlp": 0.01090305,
      "balance_loss_clip": 1.08261597,
      "balance_loss_mlp": 1.05158556,
      "epoch": 0.023748684803847887,
      "flos": 21724375939200.0,
      "grad_norm": 2.1154644933840387,
      "language_loss": 0.85621691,
      "learning_rate": 3.84951865465269e-06,
      "loss": 0.88011181,
      "num_input_tokens_seen": 8246825,
      "step": 395,
      "time_per_iteration": 2.5246951580047607
    },
    {
      "auxiliary_loss_clip": 0.0118279,
      "auxiliary_loss_mlp": 0.01038124,
      "balance_loss_clip": 1.07161582,
      "balance_loss_mlp": 1.02663231,
      "epoch": 0.02380880805651586,
      "flos": 61926192881280.0,
      "grad_norm": 0.9798267886652037,
      "language_loss": 0.63866138,
      "learning_rate": 3.851146600358172e-06,
      "loss": 0.66087043,
      "num_input_tokens_seen": 8302835,
      "step": 396,
      "time_per_iteration": 2.929664373397827
    },
    {
      "auxiliary_loss_clip": 0.01296999,
      "auxiliary_loss_mlp": 0.01072306,
      "balance_loss_clip": 1.08221042,
      "balance_loss_mlp": 1.03537536,
      "epoch": 0.023868931309183827,
      "flos": 20266331068800.0,
      "grad_norm": 2.5270554548033073,
      "language_loss": 0.83809316,
      "learning_rate": 3.852770440269372e-06,
      "loss": 0.86178619,
      "num_input_tokens_seen": 8320745,
      "step": 397,
      "time_per_iteration": 2.5194251537323
    },
    {
      "auxiliary_loss_clip": 0.01301068,
      "auxiliary_loss_mlp": 0.0109072,
      "balance_loss_clip": 1.08536923,
      "balance_loss_mlp": 1.05309761,
      "epoch": 0.023929054561851796,
      "flos": 21139103733120.0,
      "grad_norm": 2.217874653480365,
      "language_loss": 0.84606141,
      "learning_rate": 3.854390195044404e-06,
      "loss": 0.86997926,
      "num_input_tokens_seen": 8339540,
      "step": 398,
      "time_per_iteration": 2.5117952823638916
    },
    {
      "auxiliary_loss_clip": 0.01300739,
      "auxiliary_loss_mlp": 0.01077769,
      "balance_loss_clip": 1.08120346,
      "balance_loss_mlp": 1.0391928,
      "epoch": 0.023989177814519765,
      "flos": 13698521049600.0,
      "grad_norm": 2.6147474600277825,
      "language_loss": 0.86140633,
      "learning_rate": 3.856005885185868e-06,
      "loss": 0.88519132,
      "num_input_tokens_seen": 8354890,
      "step": 399,
      "time_per_iteration": 2.4803245067596436
    },
    {
      "auxiliary_loss_clip": 0.01296234,
      "auxiliary_loss_mlp": 0.01088423,
      "balance_loss_clip": 1.08429503,
      "balance_loss_mlp": 1.05132508,
      "epoch": 0.024049301067187733,
      "flos": 26322018929280.0,
      "grad_norm": 1.9520564483724356,
      "language_loss": 0.86424005,
      "learning_rate": 3.857617531042398e-06,
      "loss": 0.88808668,
      "num_input_tokens_seen": 8375845,
      "step": 400,
      "time_per_iteration": 2.547632932662964
    },
    {
      "auxiliary_loss_clip": 0.01302918,
      "auxiliary_loss_mlp": 0.01082507,
      "balance_loss_clip": 1.08725262,
      "balance_loss_mlp": 1.04552794,
      "epoch": 0.024109424319855705,
      "flos": 24425432910720.0,
      "grad_norm": 5.219954263965941,
      "language_loss": 0.79198027,
      "learning_rate": 3.8592251528102065e-06,
      "loss": 0.81583452,
      "num_input_tokens_seen": 8395240,
      "step": 401,
      "time_per_iteration": 2.5446009635925293
    },
    {
      "auxiliary_loss_clip": 0.01295131,
      "auxiliary_loss_mlp": 0.0108941,
      "balance_loss_clip": 1.08119714,
      "balance_loss_mlp": 1.05283642,
      "epoch": 0.024169547572523674,
      "flos": 29604397610880.0,
      "grad_norm": 2.042010850322047,
      "language_loss": 0.7874651,
      "learning_rate": 3.8608287705345976e-06,
      "loss": 0.81131053,
      "num_input_tokens_seen": 8416950,
      "step": 402,
      "time_per_iteration": 2.5405514240264893
    },
    {
      "auxiliary_loss_clip": 0.01301418,
      "auxiliary_loss_mlp": 0.01081609,
      "balance_loss_clip": 1.08259165,
      "balance_loss_mlp": 1.04305673,
      "epoch": 0.024229670825191642,
      "flos": 22601458235520.0,
      "grad_norm": 2.5081682172187754,
      "language_loss": 0.94641459,
      "learning_rate": 3.86242840411147e-06,
      "loss": 0.97024494,
      "num_input_tokens_seen": 8433660,
      "step": 403,
      "time_per_iteration": 2.4966673851013184
    },
    {
      "auxiliary_loss_clip": 0.01304372,
      "auxiliary_loss_mlp": 0.01091688,
      "balance_loss_clip": 1.08116198,
      "balance_loss_mlp": 1.05220604,
      "epoch": 0.02428979407785961,
      "flos": 18150258994560.0,
      "grad_norm": 2.6112802569762135,
      "language_loss": 0.99814641,
      "learning_rate": 3.864024073288798e-06,
      "loss": 1.02210689,
      "num_input_tokens_seen": 8450180,
      "step": 404,
      "time_per_iteration": 2.466996669769287
    },
    {
      "auxiliary_loss_clip": 0.01304505,
      "auxiliary_loss_mlp": 0.0110411,
      "balance_loss_clip": 1.08484054,
      "balance_loss_mlp": 1.06696403,
      "epoch": 0.024349917330527583,
      "flos": 15304984917120.0,
      "grad_norm": 2.9939423114247674,
      "language_loss": 0.87750989,
      "learning_rate": 3.865615797668091e-06,
      "loss": 0.90159607,
      "num_input_tokens_seen": 8467775,
      "step": 405,
      "time_per_iteration": 2.4818620681762695
    },
    {
      "auxiliary_loss_clip": 0.01311309,
      "auxiliary_loss_mlp": 0.01101068,
      "balance_loss_clip": 1.08778858,
      "balance_loss_mlp": 1.06308818,
      "epoch": 0.024410040583195552,
      "flos": 20773892200320.0,
      "grad_norm": 2.3720614800987323,
      "language_loss": 0.9329437,
      "learning_rate": 3.867203596705844e-06,
      "loss": 0.95706749,
      "num_input_tokens_seen": 8486765,
      "step": 406,
      "time_per_iteration": 2.4722554683685303
    },
    {
      "auxiliary_loss_clip": 0.013037,
      "auxiliary_loss_mlp": 0.01091249,
      "balance_loss_clip": 1.08672905,
      "balance_loss_mlp": 1.05231547,
      "epoch": 0.02447016383586352,
      "flos": 21798854789760.0,
      "grad_norm": 2.039096268801717,
      "language_loss": 0.87003422,
      "learning_rate": 3.86878748971496e-06,
      "loss": 0.89398366,
      "num_input_tokens_seen": 8506515,
      "step": 407,
      "time_per_iteration": 2.50523042678833
    },
    {
      "auxiliary_loss_clip": 0.01300194,
      "auxiliary_loss_mlp": 0.01084337,
      "balance_loss_clip": 1.08679223,
      "balance_loss_mlp": 1.04723918,
      "epoch": 0.02453028708853149,
      "flos": 33948116380800.0,
      "grad_norm": 2.0433415160358543,
      "language_loss": 0.74289078,
      "learning_rate": 3.8703674958661596e-06,
      "loss": 0.76673603,
      "num_input_tokens_seen": 8528035,
      "step": 408,
      "time_per_iteration": 3.954195499420166
    },
    {
      "auxiliary_loss_clip": 0.01305114,
      "auxiliary_loss_mlp": 0.0109265,
      "balance_loss_clip": 1.08610761,
      "balance_loss_mlp": 1.05462217,
      "epoch": 0.024590410341199458,
      "flos": 21793000872960.0,
      "grad_norm": 3.308419862807477,
      "language_loss": 0.92414629,
      "learning_rate": 3.871943634189376e-06,
      "loss": 0.94812393,
      "num_input_tokens_seen": 8546455,
      "step": 409,
      "time_per_iteration": 3.9460253715515137
    },
    {
      "auxiliary_loss_clip": 0.01302177,
      "auxiliary_loss_mlp": 0.01077419,
      "balance_loss_clip": 1.08635283,
      "balance_loss_mlp": 1.04318166,
      "epoch": 0.02465053359386743,
      "flos": 35114782124160.0,
      "grad_norm": 3.5334584229635304,
      "language_loss": 0.82585609,
      "learning_rate": 3.873515923575128e-06,
      "loss": 0.84965205,
      "num_input_tokens_seen": 8568450,
      "step": 410,
      "time_per_iteration": 3.980009078979492
    },
    {
      "auxiliary_loss_clip": 0.01304367,
      "auxiliary_loss_mlp": 0.0109329,
      "balance_loss_clip": 1.08642793,
      "balance_loss_mlp": 1.05616808,
      "epoch": 0.0247106568465354,
      "flos": 27451409333760.0,
      "grad_norm": 2.5289261243078394,
      "language_loss": 0.77953219,
      "learning_rate": 3.875084382775879e-06,
      "loss": 0.80350882,
      "num_input_tokens_seen": 8589340,
      "step": 411,
      "time_per_iteration": 2.5406899452209473
    },
    {
      "auxiliary_loss_clip": 0.01302626,
      "auxiliary_loss_mlp": 0.01099911,
      "balance_loss_clip": 1.08339143,
      "balance_loss_mlp": 1.06166887,
      "epoch": 0.024770780099203367,
      "flos": 20703794808960.0,
      "grad_norm": 2.2736611440976167,
      "language_loss": 0.8649956,
      "learning_rate": 3.87664903040738e-06,
      "loss": 0.88902104,
      "num_input_tokens_seen": 8607150,
      "step": 412,
      "time_per_iteration": 2.4936506748199463
    },
    {
      "auxiliary_loss_clip": 0.01169076,
      "auxiliary_loss_mlp": 0.01061884,
      "balance_loss_clip": 1.06054318,
      "balance_loss_mlp": 1.05091715,
      "epoch": 0.024830903351871336,
      "flos": 69551859369600.0,
      "grad_norm": 0.8498793146915012,
      "language_loss": 0.5854373,
      "learning_rate": 3.878209884949994e-06,
      "loss": 0.6077469,
      "num_input_tokens_seen": 8669865,
      "step": 413,
      "time_per_iteration": 4.616122722625732
    },
    {
      "auxiliary_loss_clip": 0.01294664,
      "auxiliary_loss_mlp": 0.01093039,
      "balance_loss_clip": 1.0805521,
      "balance_loss_mlp": 1.05267429,
      "epoch": 0.024891026604539304,
      "flos": 32270477713920.0,
      "grad_norm": 1.6834506535470901,
      "language_loss": 0.80654544,
      "learning_rate": 3.879766964750006e-06,
      "loss": 0.83042252,
      "num_input_tokens_seen": 8690235,
      "step": 414,
      "time_per_iteration": 2.613189458847046
    },
    {
      "auxiliary_loss_clip": 0.01291478,
      "auxiliary_loss_mlp": 0.01096323,
      "balance_loss_clip": 1.08050799,
      "balance_loss_mlp": 1.05917728,
      "epoch": 0.024951149857207276,
      "flos": 18840282238080.0,
      "grad_norm": 2.3857012662037014,
      "language_loss": 0.80220962,
      "learning_rate": 3.881320288020917e-06,
      "loss": 0.82608765,
      "num_input_tokens_seen": 8706295,
      "step": 415,
      "time_per_iteration": 2.4934582710266113
    },
    {
      "auxiliary_loss_clip": 0.01309313,
      "auxiliary_loss_mlp": 0.01085492,
      "balance_loss_clip": 1.08717442,
      "balance_loss_mlp": 1.04834676,
      "epoch": 0.025011273109875245,
      "flos": 15377201210880.0,
      "grad_norm": 2.9143216112797234,
      "language_loss": 0.96300751,
      "learning_rate": 3.882869872844723e-06,
      "loss": 0.98695552,
      "num_input_tokens_seen": 8724200,
      "step": 416,
      "time_per_iteration": 2.508213520050049
    },
    {
      "auxiliary_loss_clip": 0.0129781,
      "auxiliary_loss_mlp": 0.01077933,
      "balance_loss_clip": 1.08289409,
      "balance_loss_mlp": 1.03883266,
      "epoch": 0.025071396362543213,
      "flos": 18915515274240.0,
      "grad_norm": 1.5542633821098661,
      "language_loss": 0.77392864,
      "learning_rate": 3.884415737173176e-06,
      "loss": 0.7976861,
      "num_input_tokens_seen": 8744170,
      "step": 417,
      "time_per_iteration": 2.537775754928589
    },
    {
      "auxiliary_loss_clip": 0.01294253,
      "auxiliary_loss_mlp": 0.01094293,
      "balance_loss_clip": 1.08534837,
      "balance_loss_mlp": 1.05674255,
      "epoch": 0.025131519615211182,
      "flos": 25337958952320.0,
      "grad_norm": 1.6428277988881896,
      "language_loss": 0.77056283,
      "learning_rate": 3.8859578988290344e-06,
      "loss": 0.79444826,
      "num_input_tokens_seen": 8765120,
      "step": 418,
      "time_per_iteration": 2.568434953689575
    },
    {
      "auxiliary_loss_clip": 0.01302401,
      "auxiliary_loss_mlp": 0.01076298,
      "balance_loss_clip": 1.08733559,
      "balance_loss_mlp": 1.04051173,
      "epoch": 0.02519164286787915,
      "flos": 18953149749120.0,
      "grad_norm": 2.5073441006542896,
      "language_loss": 0.8146444,
      "learning_rate": 3.887496375507294e-06,
      "loss": 0.83843136,
      "num_input_tokens_seen": 8783500,
      "step": 419,
      "time_per_iteration": 2.4727859497070312
    },
    {
      "auxiliary_loss_clip": 0.01296939,
      "auxiliary_loss_mlp": 0.01085368,
      "balance_loss_clip": 1.08577549,
      "balance_loss_mlp": 1.04664934,
      "epoch": 0.025251766120547123,
      "flos": 17421092904960.0,
      "grad_norm": 2.441449583315054,
      "language_loss": 0.73835105,
      "learning_rate": 3.8890311847764065e-06,
      "loss": 0.76217413,
      "num_input_tokens_seen": 8801175,
      "step": 420,
      "time_per_iteration": 2.4857852458953857
    },
    {
      "auxiliary_loss_clip": 0.01296221,
      "auxiliary_loss_mlp": 0.01096019,
      "balance_loss_clip": 1.0814327,
      "balance_loss_mlp": 1.05920732,
      "epoch": 0.02531188937321509,
      "flos": 25045430590080.0,
      "grad_norm": 1.7161945115572104,
      "language_loss": 0.78964794,
      "learning_rate": 3.890562344079484e-06,
      "loss": 0.81357038,
      "num_input_tokens_seen": 8820215,
      "step": 421,
      "time_per_iteration": 2.5336453914642334
    },
    {
      "auxiliary_loss_clip": 0.01294997,
      "auxiliary_loss_mlp": 0.01090051,
      "balance_loss_clip": 1.08439362,
      "balance_loss_mlp": 1.05085516,
      "epoch": 0.02537201262588306,
      "flos": 30592228515840.0,
      "grad_norm": 2.869340971483033,
      "language_loss": 0.81685144,
      "learning_rate": 3.89208987073549e-06,
      "loss": 0.84070194,
      "num_input_tokens_seen": 8839660,
      "step": 422,
      "time_per_iteration": 2.573496103286743
    },
    {
      "auxiliary_loss_clip": 0.01298319,
      "auxiliary_loss_mlp": 0.0108221,
      "balance_loss_clip": 1.08178818,
      "balance_loss_mlp": 1.04699564,
      "epoch": 0.02543213587855103,
      "flos": 26065365275520.0,
      "grad_norm": 1.8121756505453113,
      "language_loss": 0.83335894,
      "learning_rate": 3.893613781940409e-06,
      "loss": 0.8571642,
      "num_input_tokens_seen": 8859280,
      "step": 423,
      "time_per_iteration": 2.52799129486084
    },
    {
      "auxiliary_loss_clip": 0.0129138,
      "auxiliary_loss_mlp": 0.0108122,
      "balance_loss_clip": 1.07955217,
      "balance_loss_mlp": 1.04545772,
      "epoch": 0.025492259131218997,
      "flos": 36022818965760.0,
      "grad_norm": 1.9244656666327609,
      "language_loss": 0.74307859,
      "learning_rate": 3.895134094768415e-06,
      "loss": 0.76680458,
      "num_input_tokens_seen": 8880560,
      "step": 424,
      "time_per_iteration": 2.7121007442474365
    },
    {
      "auxiliary_loss_clip": 0.01300802,
      "auxiliary_loss_mlp": 0.01100311,
      "balance_loss_clip": 1.0853703,
      "balance_loss_mlp": 1.06481028,
      "epoch": 0.02555238238388697,
      "flos": 18588045957120.0,
      "grad_norm": 3.73347394985447,
      "language_loss": 0.83358854,
      "learning_rate": 3.896650826173015e-06,
      "loss": 0.85759968,
      "num_input_tokens_seen": 8899155,
      "step": 425,
      "time_per_iteration": 2.481843948364258
    },
    {
      "auxiliary_loss_clip": 0.01296658,
      "auxiliary_loss_mlp": 0.01088809,
      "balance_loss_clip": 1.07773089,
      "balance_loss_mlp": 1.05028021,
      "epoch": 0.025612505636554938,
      "flos": 24243186280320.0,
      "grad_norm": 5.574313270590668,
      "language_loss": 0.85630953,
      "learning_rate": 3.898163992988186e-06,
      "loss": 0.88016415,
      "num_input_tokens_seen": 8917890,
      "step": 426,
      "time_per_iteration": 2.594860076904297
    },
    {
      "auxiliary_loss_clip": 0.01169802,
      "auxiliary_loss_mlp": 0.01028576,
      "balance_loss_clip": 1.06239998,
      "balance_loss_mlp": 1.01727474,
      "epoch": 0.025672628889222907,
      "flos": 60586941265920.0,
      "grad_norm": 0.902800011789361,
      "language_loss": 0.57298642,
      "learning_rate": 3.899673611929491e-06,
      "loss": 0.59497017,
      "num_input_tokens_seen": 8978260,
      "step": 427,
      "time_per_iteration": 3.1978750228881836
    },
    {
      "auxiliary_loss_clip": 0.01297447,
      "auxiliary_loss_mlp": 0.01093804,
      "balance_loss_clip": 1.08717608,
      "balance_loss_mlp": 1.05827975,
      "epoch": 0.025732752141890875,
      "flos": 19573255169280.0,
      "grad_norm": 2.370365355975273,
      "language_loss": 0.8783105,
      "learning_rate": 3.901179699595194e-06,
      "loss": 0.90222299,
      "num_input_tokens_seen": 8994460,
      "step": 428,
      "time_per_iteration": 2.5779666900634766
    },
    {
      "auxiliary_loss_clip": 0.0129053,
      "auxiliary_loss_mlp": 0.01077985,
      "balance_loss_clip": 1.08059037,
      "balance_loss_mlp": 1.03945637,
      "epoch": 0.025792875394558847,
      "flos": 31284262920960.0,
      "grad_norm": 1.5888378280551276,
      "language_loss": 0.85852361,
      "learning_rate": 3.902682272467353e-06,
      "loss": 0.8822087,
      "num_input_tokens_seen": 9016670,
      "step": 429,
      "time_per_iteration": 2.579270124435425
    },
    {
      "auxiliary_loss_clip": 0.01295644,
      "auxiliary_loss_mlp": 0.01081738,
      "balance_loss_clip": 1.0792129,
      "balance_loss_mlp": 1.04354286,
      "epoch": 0.025852998647226816,
      "flos": 32379610210560.0,
      "grad_norm": 2.306694592278278,
      "language_loss": 0.88351828,
      "learning_rate": 3.904181346912895e-06,
      "loss": 0.90729213,
      "num_input_tokens_seen": 9039720,
      "step": 430,
      "time_per_iteration": 2.5945487022399902
    },
    {
      "auxiliary_loss_clip": 0.01297089,
      "auxiliary_loss_mlp": 0.01081621,
      "balance_loss_clip": 1.08788157,
      "balance_loss_mlp": 1.0467639,
      "epoch": 0.025913121899894784,
      "flos": 20193288762240.0,
      "grad_norm": 1.866764061918992,
      "language_loss": 0.8402797,
      "learning_rate": 3.905676939184698e-06,
      "loss": 0.86406684,
      "num_input_tokens_seen": 9059850,
      "step": 431,
      "time_per_iteration": 2.5026040077209473
    },
    {
      "auxiliary_loss_clip": 0.01293406,
      "auxiliary_loss_mlp": 0.0107492,
      "balance_loss_clip": 1.08274579,
      "balance_loss_mlp": 1.0407784,
      "epoch": 0.025973245152562753,
      "flos": 14720430983040.0,
      "grad_norm": 4.540804590060525,
      "language_loss": 0.86289227,
      "learning_rate": 3.907169065422638e-06,
      "loss": 0.88657558,
      "num_input_tokens_seen": 9077590,
      "step": 432,
      "time_per_iteration": 2.4859097003936768
    },
    {
      "auxiliary_loss_clip": 0.01294266,
      "auxiliary_loss_mlp": 0.01072584,
      "balance_loss_clip": 1.08290887,
      "balance_loss_mlp": 1.03825152,
      "epoch": 0.02603336840523072,
      "flos": 30992991534720.0,
      "grad_norm": 2.1806656000718854,
      "language_loss": 0.75949878,
      "learning_rate": 3.908657741654636e-06,
      "loss": 0.78316724,
      "num_input_tokens_seen": 9099880,
      "step": 433,
      "time_per_iteration": 2.563768148422241
    },
    {
      "auxiliary_loss_clip": 0.01295576,
      "auxiliary_loss_mlp": 0.01093916,
      "balance_loss_clip": 1.08065009,
      "balance_loss_mlp": 1.05469584,
      "epoch": 0.026093491657898694,
      "flos": 17674262939520.0,
      "grad_norm": 1.8972918397434833,
      "language_loss": 0.89419562,
      "learning_rate": 3.910142983797699e-06,
      "loss": 0.91809046,
      "num_input_tokens_seen": 9118620,
      "step": 434,
      "time_per_iteration": 2.4933855533599854
    },
    {
      "auxiliary_loss_clip": 0.01296454,
      "auxiliary_loss_mlp": 0.011044,
      "balance_loss_clip": 1.08591163,
      "balance_loss_mlp": 1.06625271,
      "epoch": 0.026153614910566662,
      "flos": 17857874286720.0,
      "grad_norm": 2.4857059990355808,
      "language_loss": 0.80055976,
      "learning_rate": 3.9116248076589305e-06,
      "loss": 0.82456827,
      "num_input_tokens_seen": 9135655,
      "step": 435,
      "time_per_iteration": 2.472984552383423
    },
    {
      "auxiliary_loss_clip": 0.01291785,
      "auxiliary_loss_mlp": 0.01087401,
      "balance_loss_clip": 1.07963979,
      "balance_loss_mlp": 1.05039859,
      "epoch": 0.02621373816323463,
      "flos": 20011113959040.0,
      "grad_norm": 2.2315123560050645,
      "language_loss": 0.86554694,
      "learning_rate": 3.913103228936546e-06,
      "loss": 0.88933885,
      "num_input_tokens_seen": 9153520,
      "step": 436,
      "time_per_iteration": 2.4957656860351562
    },
    {
      "auxiliary_loss_clip": 0.01295969,
      "auxiliary_loss_mlp": 0.01094134,
      "balance_loss_clip": 1.08391356,
      "balance_loss_mlp": 1.05751312,
      "epoch": 0.0262738614159026,
      "flos": 19281193683840.0,
      "grad_norm": 2.1859195461373577,
      "language_loss": 0.74891704,
      "learning_rate": 3.914578263220868e-06,
      "loss": 0.77281809,
      "num_input_tokens_seen": 9170750,
      "step": 437,
      "time_per_iteration": 2.493433713912964
    },
    {
      "auxiliary_loss_clip": 0.01294013,
      "auxiliary_loss_mlp": 0.01094299,
      "balance_loss_clip": 1.08470142,
      "balance_loss_mlp": 1.05557966,
      "epoch": 0.026333984668570568,
      "flos": 18807208790400.0,
      "grad_norm": 3.1187635584237485,
      "language_loss": 0.91061419,
      "learning_rate": 3.916049925995316e-06,
      "loss": 0.9344973,
      "num_input_tokens_seen": 9188430,
      "step": 438,
      "time_per_iteration": 2.480621099472046
    },
    {
      "auxiliary_loss_clip": 0.01161758,
      "auxiliary_loss_mlp": 0.01022264,
      "balance_loss_clip": 1.05759215,
      "balance_loss_mlp": 1.01248932,
      "epoch": 0.02639410792123854,
      "flos": 64572020691840.0,
      "grad_norm": 0.8679921724154448,
      "language_loss": 0.62627381,
      "learning_rate": 3.917518232637377e-06,
      "loss": 0.64811409,
      "num_input_tokens_seen": 9255835,
      "step": 439,
      "time_per_iteration": 3.195735454559326
    },
    {
      "auxiliary_loss_clip": 0.01302715,
      "auxiliary_loss_mlp": 0.01091455,
      "balance_loss_clip": 1.08799112,
      "balance_loss_mlp": 1.05385661,
      "epoch": 0.02645423117390651,
      "flos": 28473462921600.0,
      "grad_norm": 3.493478589745417,
      "language_loss": 0.75755179,
      "learning_rate": 3.918983198419573e-06,
      "loss": 0.78149343,
      "num_input_tokens_seen": 9276835,
      "step": 440,
      "time_per_iteration": 2.5442724227905273
    },
    {
      "auxiliary_loss_clip": 0.01293268,
      "auxiliary_loss_mlp": 0.01076078,
      "balance_loss_clip": 1.08369565,
      "balance_loss_mlp": 1.03983831,
      "epoch": 0.026514354426574478,
      "flos": 18551237495040.0,
      "grad_norm": 3.0311629252525614,
      "language_loss": 0.83290154,
      "learning_rate": 3.920444838510415e-06,
      "loss": 0.85659492,
      "num_input_tokens_seen": 9295075,
      "step": 441,
      "time_per_iteration": 2.5400731563568115
    },
    {
      "auxiliary_loss_clip": 0.01296894,
      "auxiliary_loss_mlp": 0.01086063,
      "balance_loss_clip": 1.08225048,
      "balance_loss_mlp": 1.04801142,
      "epoch": 0.026574477679242446,
      "flos": 20667812359680.0,
      "grad_norm": 2.214105421451439,
      "language_loss": 0.78743005,
      "learning_rate": 3.92190316797534e-06,
      "loss": 0.81125963,
      "num_input_tokens_seen": 9314205,
      "step": 442,
      "time_per_iteration": 2.514350175857544
    },
    {
      "auxiliary_loss_clip": 0.01159651,
      "auxiliary_loss_mlp": 0.01014663,
      "balance_loss_clip": 1.05674577,
      "balance_loss_mlp": 1.00464976,
      "epoch": 0.026634600931910415,
      "flos": 57956125340160.0,
      "grad_norm": 0.9579676985050465,
      "language_loss": 0.64539683,
      "learning_rate": 3.92335820177765e-06,
      "loss": 0.66714001,
      "num_input_tokens_seen": 9367395,
      "step": 443,
      "time_per_iteration": 2.980947971343994
    },
    {
      "auxiliary_loss_clip": 0.01295985,
      "auxiliary_loss_mlp": 0.01087874,
      "balance_loss_clip": 1.08613324,
      "balance_loss_mlp": 1.0515871,
      "epoch": 0.026694724184578387,
      "flos": 15815131827840.0,
      "grad_norm": 4.077748732564081,
      "language_loss": 0.82599366,
      "learning_rate": 3.924809954779425e-06,
      "loss": 0.84983224,
      "num_input_tokens_seen": 9385185,
      "step": 444,
      "time_per_iteration": 2.4806153774261475
    },
    {
      "auxiliary_loss_clip": 0.01298424,
      "auxiliary_loss_mlp": 0.01087132,
      "balance_loss_clip": 1.08269048,
      "balance_loss_mlp": 1.04769778,
      "epoch": 0.026754847437246355,
      "flos": 23440259612160.0,
      "grad_norm": 2.147313420151961,
      "language_loss": 0.95792741,
      "learning_rate": 3.9262584417424425e-06,
      "loss": 0.98178297,
      "num_input_tokens_seen": 9403225,
      "step": 445,
      "time_per_iteration": 2.5299737453460693
    },
    {
      "auxiliary_loss_clip": 0.01294879,
      "auxiliary_loss_mlp": 0.01094843,
      "balance_loss_clip": 1.08487761,
      "balance_loss_mlp": 1.05562341,
      "epoch": 0.026814970689914324,
      "flos": 17341801632000.0,
      "grad_norm": 2.343351243696989,
      "language_loss": 0.91824687,
      "learning_rate": 3.9277036773290725e-06,
      "loss": 0.9421441,
      "num_input_tokens_seen": 9420540,
      "step": 446,
      "time_per_iteration": 2.499513626098633
    },
    {
      "auxiliary_loss_clip": 0.01289724,
      "auxiliary_loss_mlp": 0.0108143,
      "balance_loss_clip": 1.08277738,
      "balance_loss_mlp": 1.04402161,
      "epoch": 0.026875093942582293,
      "flos": 17894718662400.0,
      "grad_norm": 2.2114211953160776,
      "language_loss": 0.79940355,
      "learning_rate": 3.92914567610317e-06,
      "loss": 0.82311511,
      "num_input_tokens_seen": 9438840,
      "step": 447,
      "time_per_iteration": 2.512451410293579
    },
    {
      "auxiliary_loss_clip": 0.01293725,
      "auxiliary_loss_mlp": 0.01075116,
      "balance_loss_clip": 1.08448386,
      "balance_loss_mlp": 1.04047394,
      "epoch": 0.026935217195250265,
      "flos": 21723980889600.0,
      "grad_norm": 3.0785195002079484,
      "language_loss": 0.8635205,
      "learning_rate": 3.930584452530952e-06,
      "loss": 0.88720894,
      "num_input_tokens_seen": 9457215,
      "step": 448,
      "time_per_iteration": 3.8676347732543945
    },
    {
      "auxiliary_loss_clip": 0.01287436,
      "auxiliary_loss_mlp": 0.01094766,
      "balance_loss_clip": 1.08288908,
      "balance_loss_mlp": 1.06071949,
      "epoch": 0.026995340447918233,
      "flos": 23622685810560.0,
      "grad_norm": 3.6800706122280054,
      "language_loss": 0.88914955,
      "learning_rate": 3.9320200209818755e-06,
      "loss": 0.91297162,
      "num_input_tokens_seen": 9475615,
      "step": 449,
      "time_per_iteration": 3.8611867427825928
    },
    {
      "auxiliary_loss_clip": 0.01297281,
      "auxiliary_loss_mlp": 0.01090528,
      "balance_loss_clip": 1.08358872,
      "balance_loss_mlp": 1.05309629,
      "epoch": 0.027055463700586202,
      "flos": 17931275729280.0,
      "grad_norm": 2.237156602718341,
      "language_loss": 0.80366027,
      "learning_rate": 3.933452395729493e-06,
      "loss": 0.82753831,
      "num_input_tokens_seen": 9493975,
      "step": 450,
      "time_per_iteration": 2.487452983856201
    },
    {
      "auxiliary_loss_clip": 0.01290778,
      "auxiliary_loss_mlp": 0.01080574,
      "balance_loss_clip": 1.0869211,
      "balance_loss_mlp": 1.04390478,
      "epoch": 0.02711558695325417,
      "flos": 25118903859840.0,
      "grad_norm": 1.711941683483557,
      "language_loss": 0.81541884,
      "learning_rate": 3.934881590952304e-06,
      "loss": 0.83913231,
      "num_input_tokens_seen": 9514810,
      "step": 451,
      "time_per_iteration": 3.931917428970337
    },
    {
      "auxiliary_loss_clip": 0.01289713,
      "auxiliary_loss_mlp": 0.01094382,
      "balance_loss_clip": 1.08662915,
      "balance_loss_mlp": 1.05697417,
      "epoch": 0.02717571020592214,
      "flos": 24239559006720.0,
      "grad_norm": 1.6293262699563116,
      "language_loss": 0.76990795,
      "learning_rate": 3.936307620734599e-06,
      "loss": 0.79374886,
      "num_input_tokens_seen": 9533635,
      "step": 452,
      "time_per_iteration": 2.5682435035705566
    },
    {
      "auxiliary_loss_clip": 0.01289312,
      "auxiliary_loss_mlp": 0.01086099,
      "balance_loss_clip": 1.08467674,
      "balance_loss_mlp": 1.049263,
      "epoch": 0.02723583345859011,
      "flos": 25118939773440.0,
      "grad_norm": 1.8070407921933467,
      "language_loss": 0.72996998,
      "learning_rate": 3.937730499067294e-06,
      "loss": 0.7537241,
      "num_input_tokens_seen": 9555420,
      "step": 453,
      "time_per_iteration": 3.963440418243408
    },
    {
      "auxiliary_loss_clip": 0.01284835,
      "auxiliary_loss_mlp": 0.01084017,
      "balance_loss_clip": 1.08236194,
      "balance_loss_mlp": 1.0483973,
      "epoch": 0.02729595671125808,
      "flos": 42741597847680.0,
      "grad_norm": 1.8750299677014342,
      "language_loss": 0.82110214,
      "learning_rate": 3.939150239848748e-06,
      "loss": 0.8447907,
      "num_input_tokens_seen": 9578950,
      "step": 454,
      "time_per_iteration": 2.7149899005889893
    },
    {
      "auxiliary_loss_clip": 0.01289311,
      "auxiliary_loss_mlp": 0.01082397,
      "balance_loss_clip": 1.08481574,
      "balance_loss_mlp": 1.04939997,
      "epoch": 0.02735607996392605,
      "flos": 21430985650560.0,
      "grad_norm": 1.9082207368311865,
      "language_loss": 0.75332922,
      "learning_rate": 3.9405668568855866e-06,
      "loss": 0.77704632,
      "num_input_tokens_seen": 9598160,
      "step": 455,
      "time_per_iteration": 2.5228545665740967
    },
    {
      "auxiliary_loss_clip": 0.01287275,
      "auxiliary_loss_mlp": 0.01094347,
      "balance_loss_clip": 1.08059645,
      "balance_loss_mlp": 1.05953836,
      "epoch": 0.027416203216594017,
      "flos": 20851280052480.0,
      "grad_norm": 2.7086095501431653,
      "language_loss": 0.80716991,
      "learning_rate": 3.941980363893499e-06,
      "loss": 0.83098614,
      "num_input_tokens_seen": 9616010,
      "step": 456,
      "time_per_iteration": 2.516589403152466
    },
    {
      "auxiliary_loss_clip": 0.01285345,
      "auxiliary_loss_mlp": 0.01078621,
      "balance_loss_clip": 1.08194852,
      "balance_loss_mlp": 1.04240477,
      "epoch": 0.027476326469261986,
      "flos": 13224500242560.0,
      "grad_norm": 2.1735788221840195,
      "language_loss": 0.81588703,
      "learning_rate": 3.9433907744980384e-06,
      "loss": 0.83952665,
      "num_input_tokens_seen": 9634000,
      "step": 457,
      "time_per_iteration": 2.5105857849121094
    },
    {
      "auxiliary_loss_clip": 0.01289553,
      "auxiliary_loss_mlp": 0.0108411,
      "balance_loss_clip": 1.08104074,
      "balance_loss_mlp": 1.04896736,
      "epoch": 0.027536449721929958,
      "flos": 24024526237440.0,
      "grad_norm": 2.1549821104524036,
      "language_loss": 0.9370966,
      "learning_rate": 3.944798102235412e-06,
      "loss": 0.96083325,
      "num_input_tokens_seen": 9653455,
      "step": 458,
      "time_per_iteration": 2.5741076469421387
    },
    {
      "auxiliary_loss_clip": 0.01287464,
      "auxiliary_loss_mlp": 0.010925,
      "balance_loss_clip": 1.08236825,
      "balance_loss_mlp": 1.05816746,
      "epoch": 0.027596572974597926,
      "flos": 13006055681280.0,
      "grad_norm": 2.3494062019256106,
      "language_loss": 0.79069924,
      "learning_rate": 3.9462023605532545e-06,
      "loss": 0.8144989,
      "num_input_tokens_seen": 9669650,
      "step": 459,
      "time_per_iteration": 2.497664451599121
    },
    {
      "auxiliary_loss_clip": 0.0129346,
      "auxiliary_loss_mlp": 0.01085722,
      "balance_loss_clip": 1.08861709,
      "balance_loss_mlp": 1.04697895,
      "epoch": 0.027656696227265895,
      "flos": 26143076350080.0,
      "grad_norm": 1.9124675039781192,
      "language_loss": 0.8325392,
      "learning_rate": 3.947603562811407e-06,
      "loss": 0.85633105,
      "num_input_tokens_seen": 9691415,
      "step": 460,
      "time_per_iteration": 2.5582635402679443
    },
    {
      "auxiliary_loss_clip": 0.01151267,
      "auxiliary_loss_mlp": 0.0101601,
      "balance_loss_clip": 1.05213904,
      "balance_loss_mlp": 1.00680721,
      "epoch": 0.027716819479933864,
      "flos": 60697222997760.0,
      "grad_norm": 1.6215983804402434,
      "language_loss": 0.73642015,
      "learning_rate": 3.949001722282675e-06,
      "loss": 0.75809288,
      "num_input_tokens_seen": 9755605,
      "step": 461,
      "time_per_iteration": 3.087291717529297
    },
    {
      "auxiliary_loss_clip": 0.01287992,
      "auxiliary_loss_mlp": 0.01083537,
      "balance_loss_clip": 1.08910763,
      "balance_loss_mlp": 1.05011082,
      "epoch": 0.027776942732601832,
      "flos": 31211938886400.0,
      "grad_norm": 2.7804600514429563,
      "language_loss": 0.8082794,
      "learning_rate": 3.950396852153582e-06,
      "loss": 0.83199471,
      "num_input_tokens_seen": 9776270,
      "step": 462,
      "time_per_iteration": 2.6029269695281982
    },
    {
      "auxiliary_loss_clip": 0.01286385,
      "auxiliary_loss_mlp": 0.01076973,
      "balance_loss_clip": 1.08374095,
      "balance_loss_mlp": 1.04450035,
      "epoch": 0.027837065985269804,
      "flos": 22674644196480.0,
      "grad_norm": 3.08259408141153,
      "language_loss": 0.9058404,
      "learning_rate": 3.951788965525118e-06,
      "loss": 0.929474,
      "num_input_tokens_seen": 9794465,
      "step": 463,
      "time_per_iteration": 2.498464822769165
    },
    {
      "auxiliary_loss_clip": 0.01147477,
      "auxiliary_loss_mlp": 0.01009215,
      "balance_loss_clip": 1.04916048,
      "balance_loss_mlp": 1.00001204,
      "epoch": 0.027897189237937773,
      "flos": 62182487399040.0,
      "grad_norm": 0.8875297763799335,
      "language_loss": 0.59043205,
      "learning_rate": 3.953178075413476e-06,
      "loss": 0.61199898,
      "num_input_tokens_seen": 9849685,
      "step": 464,
      "time_per_iteration": 3.052896022796631
    },
    {
      "auxiliary_loss_clip": 0.01298732,
      "auxiliary_loss_mlp": 0.01099971,
      "balance_loss_clip": 1.08889878,
      "balance_loss_mlp": 1.06413662,
      "epoch": 0.02795731249060574,
      "flos": 24493160004480.0,
      "grad_norm": 3.3047635515512463,
      "language_loss": 0.81387413,
      "learning_rate": 3.954564194750784e-06,
      "loss": 0.83786106,
      "num_input_tokens_seen": 9869505,
      "step": 465,
      "time_per_iteration": 2.5674421787261963
    },
    {
      "auxiliary_loss_clip": 0.01284941,
      "auxiliary_loss_mlp": 0.01086584,
      "balance_loss_clip": 1.08092511,
      "balance_loss_mlp": 1.05105984,
      "epoch": 0.02801743574327371,
      "flos": 23733003456000.0,
      "grad_norm": 3.297863121086697,
      "language_loss": 0.78347957,
      "learning_rate": 3.955947336385828e-06,
      "loss": 0.80719483,
      "num_input_tokens_seen": 9890950,
      "step": 466,
      "time_per_iteration": 2.543804407119751
    },
    {
      "auxiliary_loss_clip": 0.01285657,
      "auxiliary_loss_mlp": 0.01088145,
      "balance_loss_clip": 1.08362174,
      "balance_loss_mlp": 1.05390835,
      "epoch": 0.02807755899594168,
      "flos": 20629100476800.0,
      "grad_norm": 1.7710125411846602,
      "language_loss": 0.87694323,
      "learning_rate": 3.957327513084761e-06,
      "loss": 0.90068126,
      "num_input_tokens_seen": 9911265,
      "step": 467,
      "time_per_iteration": 2.578993320465088
    },
    {
      "auxiliary_loss_clip": 0.01290305,
      "auxiliary_loss_mlp": 0.01103747,
      "balance_loss_clip": 1.08553684,
      "balance_loss_mlp": 1.06707859,
      "epoch": 0.02813768224860965,
      "flos": 19244564789760.0,
      "grad_norm": 2.029708135486315,
      "language_loss": 0.86303067,
      "learning_rate": 3.958704737531818e-06,
      "loss": 0.88697124,
      "num_input_tokens_seen": 9929025,
      "step": 468,
      "time_per_iteration": 2.4949283599853516
    },
    {
      "auxiliary_loss_clip": 0.01286382,
      "auxiliary_loss_mlp": 0.01081463,
      "balance_loss_clip": 1.08134127,
      "balance_loss_mlp": 1.04457951,
      "epoch": 0.02819780550127762,
      "flos": 20813968800000.0,
      "grad_norm": 2.75835791171987,
      "language_loss": 0.91922009,
      "learning_rate": 3.9600790223300065e-06,
      "loss": 0.94289857,
      "num_input_tokens_seen": 9945190,
      "step": 469,
      "time_per_iteration": 2.5109784603118896
    },
    {
      "auxiliary_loss_clip": 0.0128354,
      "auxiliary_loss_mlp": 0.01092779,
      "balance_loss_clip": 1.082798,
      "balance_loss_mlp": 1.0563004,
      "epoch": 0.028257928753945588,
      "flos": 19974125928960.0,
      "grad_norm": 2.6045041952410464,
      "language_loss": 0.81770122,
      "learning_rate": 3.96145038000181e-06,
      "loss": 0.8414644,
      "num_input_tokens_seen": 9962820,
      "step": 470,
      "time_per_iteration": 2.490478515625
    },
    {
      "auxiliary_loss_clip": 0.01286234,
      "auxiliary_loss_mlp": 0.01088199,
      "balance_loss_clip": 1.08035004,
      "balance_loss_mlp": 1.05112529,
      "epoch": 0.028318052006613557,
      "flos": 20484488321280.0,
      "grad_norm": 1.8763918805517323,
      "language_loss": 0.93138921,
      "learning_rate": 3.962818822989861e-06,
      "loss": 0.9551335,
      "num_input_tokens_seen": 9982595,
      "step": 471,
      "time_per_iteration": 2.5220422744750977
    },
    {
      "auxiliary_loss_clip": 0.01280883,
      "auxiliary_loss_mlp": 0.01094213,
      "balance_loss_clip": 1.07921922,
      "balance_loss_mlp": 1.05861735,
      "epoch": 0.02837817525928153,
      "flos": 28514832410880.0,
      "grad_norm": 1.830050496122536,
      "language_loss": 0.76000166,
      "learning_rate": 3.964184363657625e-06,
      "loss": 0.78375256,
      "num_input_tokens_seen": 10004645,
      "step": 472,
      "time_per_iteration": 2.579195976257324
    },
    {
      "auxiliary_loss_clip": 0.01288117,
      "auxiliary_loss_mlp": 0.01078818,
      "balance_loss_clip": 1.07968354,
      "balance_loss_mlp": 1.04448617,
      "epoch": 0.028438298511949497,
      "flos": 18551668458240.0,
      "grad_norm": 1.6841795139999398,
      "language_loss": 0.93579555,
      "learning_rate": 3.965547014290071e-06,
      "loss": 0.95946491,
      "num_input_tokens_seen": 10022555,
      "step": 473,
      "time_per_iteration": 2.5237925052642822
    },
    {
      "auxiliary_loss_clip": 0.01293871,
      "auxiliary_loss_mlp": 0.011125,
      "balance_loss_clip": 1.08397865,
      "balance_loss_mlp": 1.07771468,
      "epoch": 0.028498421764617466,
      "flos": 16910227722240.0,
      "grad_norm": 2.4989010098811377,
      "language_loss": 0.88353378,
      "learning_rate": 3.96690678709433e-06,
      "loss": 0.90759754,
      "num_input_tokens_seen": 10041025,
      "step": 474,
      "time_per_iteration": 2.5154855251312256
    },
    {
      "auxiliary_loss_clip": 0.01284451,
      "auxiliary_loss_mlp": 0.01088887,
      "balance_loss_clip": 1.08153296,
      "balance_loss_mlp": 1.0524807,
      "epoch": 0.028558545017285435,
      "flos": 27778699082880.0,
      "grad_norm": 2.3295711215100265,
      "language_loss": 0.78736937,
      "learning_rate": 3.968263694200355e-06,
      "loss": 0.81110275,
      "num_input_tokens_seen": 10060775,
      "step": 475,
      "time_per_iteration": 2.598573684692383
    },
    {
      "auxiliary_loss_clip": 0.01141951,
      "auxiliary_loss_mlp": 0.01019321,
      "balance_loss_clip": 1.04605865,
      "balance_loss_mlp": 1.01078606,
      "epoch": 0.028618668269953403,
      "flos": 65654367258240.0,
      "grad_norm": 0.9542091715056401,
      "language_loss": 0.66972047,
      "learning_rate": 3.969617747661569e-06,
      "loss": 0.69133323,
      "num_input_tokens_seen": 10120225,
      "step": 476,
      "time_per_iteration": 3.0931899547576904
    },
    {
      "auxiliary_loss_clip": 0.01286595,
      "auxiliary_loss_mlp": 0.01085553,
      "balance_loss_clip": 1.08214438,
      "balance_loss_mlp": 1.04824078,
      "epoch": 0.028678791522621375,
      "flos": 21937074324480.0,
      "grad_norm": 2.3198832276840786,
      "language_loss": 0.83700073,
      "learning_rate": 3.970968959455509e-06,
      "loss": 0.86072218,
      "num_input_tokens_seen": 10137880,
      "step": 477,
      "time_per_iteration": 2.5587594509124756
    },
    {
      "auxiliary_loss_clip": 0.01293006,
      "auxiliary_loss_mlp": 0.01090454,
      "balance_loss_clip": 1.08577788,
      "balance_loss_mlp": 1.05354643,
      "epoch": 0.028738914775289344,
      "flos": 24572128055040.0,
      "grad_norm": 1.9992978548442268,
      "language_loss": 0.81995547,
      "learning_rate": 3.97231734148446e-06,
      "loss": 0.84379011,
      "num_input_tokens_seen": 10156930,
      "step": 478,
      "time_per_iteration": 2.577746868133545
    },
    {
      "auxiliary_loss_clip": 0.01284946,
      "auxiliary_loss_mlp": 0.01083995,
      "balance_loss_clip": 1.08103776,
      "balance_loss_mlp": 1.04787433,
      "epoch": 0.028799038027957313,
      "flos": 23257977068160.0,
      "grad_norm": 1.6753054313428546,
      "language_loss": 0.80957401,
      "learning_rate": 3.973662905576082e-06,
      "loss": 0.8332634,
      "num_input_tokens_seen": 10176295,
      "step": 479,
      "time_per_iteration": 2.546095371246338
    },
    {
      "auxiliary_loss_clip": 0.01281748,
      "auxiliary_loss_mlp": 0.01084051,
      "balance_loss_clip": 1.07977986,
      "balance_loss_mlp": 1.04542732,
      "epoch": 0.02885916128062528,
      "flos": 22164102236160.0,
      "grad_norm": 2.5180577648561915,
      "language_loss": 0.73924029,
      "learning_rate": 3.975005663484038e-06,
      "loss": 0.76289821,
      "num_input_tokens_seen": 10195790,
      "step": 480,
      "time_per_iteration": 2.536846876144409
    },
    {
      "auxiliary_loss_clip": 0.01279903,
      "auxiliary_loss_mlp": 0.01073384,
      "balance_loss_clip": 1.08160353,
      "balance_loss_mlp": 1.04095864,
      "epoch": 0.02891928453329325,
      "flos": 22932842135040.0,
      "grad_norm": 1.8458472054812554,
      "language_loss": 0.87754673,
      "learning_rate": 3.976345626888605e-06,
      "loss": 0.9010796,
      "num_input_tokens_seen": 10218405,
      "step": 481,
      "time_per_iteration": 2.5575757026672363
    },
    {
      "auxiliary_loss_clip": 0.01139935,
      "auxiliary_loss_mlp": 0.01008764,
      "balance_loss_clip": 1.04474068,
      "balance_loss_mlp": 1.00080097,
      "epoch": 0.028979407785961222,
      "flos": 57432941792640.0,
      "grad_norm": 0.8263788188585188,
      "language_loss": 0.66088074,
      "learning_rate": 3.9776828073972864e-06,
      "loss": 0.68236774,
      "num_input_tokens_seen": 10271005,
      "step": 482,
      "time_per_iteration": 2.8484065532684326
    },
    {
      "auxiliary_loss_clip": 0.01296542,
      "auxiliary_loss_mlp": 0.01079778,
      "balance_loss_clip": 1.08520031,
      "balance_loss_mlp": 1.04544604,
      "epoch": 0.02903953103862919,
      "flos": 16722737706240.0,
      "grad_norm": 2.7450364491080927,
      "language_loss": 0.79021853,
      "learning_rate": 3.979017216545415e-06,
      "loss": 0.81398177,
      "num_input_tokens_seen": 10288405,
      "step": 483,
      "time_per_iteration": 2.492056369781494
    },
    {
      "auxiliary_loss_clip": 0.01292591,
      "auxiliary_loss_mlp": 0.01091694,
      "balance_loss_clip": 1.08602524,
      "balance_loss_mlp": 1.05500102,
      "epoch": 0.02909965429129716,
      "flos": 16763640318720.0,
      "grad_norm": 1.9281853500685082,
      "language_loss": 0.75804806,
      "learning_rate": 3.980348865796749e-06,
      "loss": 0.78189087,
      "num_input_tokens_seen": 10306875,
      "step": 484,
      "time_per_iteration": 2.4888498783111572
    },
    {
      "auxiliary_loss_clip": 0.01288126,
      "auxiliary_loss_mlp": 0.01085602,
      "balance_loss_clip": 1.08308113,
      "balance_loss_mlp": 1.05129325,
      "epoch": 0.029159777543965128,
      "flos": 19785343023360.0,
      "grad_norm": 2.1407580525097365,
      "language_loss": 0.83864254,
      "learning_rate": 3.9816777665440615e-06,
      "loss": 0.86237979,
      "num_input_tokens_seen": 10323965,
      "step": 485,
      "time_per_iteration": 2.4940836429595947
    },
    {
      "auxiliary_loss_clip": 0.01294474,
      "auxiliary_loss_mlp": 0.01087138,
      "balance_loss_clip": 1.09030056,
      "balance_loss_mlp": 1.05147076,
      "epoch": 0.029219900796633096,
      "flos": 19642670202240.0,
      "grad_norm": 2.563064328038223,
      "language_loss": 0.84566319,
      "learning_rate": 3.983003930109732e-06,
      "loss": 0.8694793,
      "num_input_tokens_seen": 10342620,
      "step": 486,
      "time_per_iteration": 2.5129122734069824
    },
    {
      "auxiliary_loss_clip": 0.01284369,
      "auxiliary_loss_mlp": 0.01092371,
      "balance_loss_clip": 1.08095777,
      "balance_loss_mlp": 1.05658412,
      "epoch": 0.02928002404930107,
      "flos": 25885704424320.0,
      "grad_norm": 2.054091571761709,
      "language_loss": 0.89366752,
      "learning_rate": 3.984327367746315e-06,
      "loss": 0.91743493,
      "num_input_tokens_seen": 10364610,
      "step": 487,
      "time_per_iteration": 2.540311098098755
    },
    {
      "auxiliary_loss_clip": 0.01289396,
      "auxiliary_loss_mlp": 0.01066399,
      "balance_loss_clip": 1.0853796,
      "balance_loss_mlp": 1.03340197,
      "epoch": 0.029340147301969037,
      "flos": 20660234590080.0,
      "grad_norm": 2.742621271370116,
      "language_loss": 0.88385177,
      "learning_rate": 3.985648090637122e-06,
      "loss": 0.90740979,
      "num_input_tokens_seen": 10380910,
      "step": 488,
      "time_per_iteration": 2.4969539642333984
    },
    {
      "auxiliary_loss_clip": 0.01284013,
      "auxiliary_loss_mlp": 0.01082493,
      "balance_loss_clip": 1.08265543,
      "balance_loss_mlp": 1.04718328,
      "epoch": 0.029400270554637006,
      "flos": 24428018689920.0,
      "grad_norm": 1.7755787439662505,
      "language_loss": 0.88873637,
      "learning_rate": 3.986966109896785e-06,
      "loss": 0.91240144,
      "num_input_tokens_seen": 10400665,
      "step": 489,
      "time_per_iteration": 6.790199041366577
    },
    {
      "auxiliary_loss_clip": 0.01278338,
      "auxiliary_loss_mlp": 0.01076341,
      "balance_loss_clip": 1.07796097,
      "balance_loss_mlp": 1.04110324,
      "epoch": 0.029460393807304974,
      "flos": 20120892900480.0,
      "grad_norm": 1.8706154609460035,
      "language_loss": 0.88531554,
      "learning_rate": 3.988281436571815e-06,
      "loss": 0.90886235,
      "num_input_tokens_seen": 10420150,
      "step": 490,
      "time_per_iteration": 2.512439250946045
    },
    {
      "auxiliary_loss_clip": 0.01284785,
      "auxiliary_loss_mlp": 0.01085391,
      "balance_loss_clip": 1.08060074,
      "balance_loss_mlp": 1.05105901,
      "epoch": 0.029520517059972943,
      "flos": 17675914965120.0,
      "grad_norm": 2.321623302917756,
      "language_loss": 0.91213524,
      "learning_rate": 3.989594081641164e-06,
      "loss": 0.93583703,
      "num_input_tokens_seen": 10438210,
      "step": 491,
      "time_per_iteration": 2.504706859588623
    },
    {
      "auxiliary_loss_clip": 0.01276362,
      "auxiliary_loss_mlp": 0.01072244,
      "balance_loss_clip": 1.08065987,
      "balance_loss_mlp": 1.03929424,
      "epoch": 0.029580640312640915,
      "flos": 18953185662720.0,
      "grad_norm": 2.1340502694181165,
      "language_loss": 0.85581422,
      "learning_rate": 3.9909040560167675e-06,
      "loss": 0.8793003,
      "num_input_tokens_seen": 10455125,
      "step": 492,
      "time_per_iteration": 2.4902396202087402
    },
    {
      "auxiliary_loss_clip": 0.01287623,
      "auxiliary_loss_mlp": 0.01101975,
      "balance_loss_clip": 1.08603764,
      "balance_loss_mlp": 1.0665462,
      "epoch": 0.029640763565308884,
      "flos": 18726121837440.0,
      "grad_norm": 3.462192112926633,
      "language_loss": 0.84109783,
      "learning_rate": 3.992211370544093e-06,
      "loss": 0.86499381,
      "num_input_tokens_seen": 10470990,
      "step": 493,
      "time_per_iteration": 2.4672958850860596
    },
    {
      "auxiliary_loss_clip": 0.01282306,
      "auxiliary_loss_mlp": 0.01075822,
      "balance_loss_clip": 1.0805037,
      "balance_loss_mlp": 1.04246759,
      "epoch": 0.029700886817976852,
      "flos": 20595308757120.0,
      "grad_norm": 2.2472980944724905,
      "language_loss": 0.86534023,
      "learning_rate": 3.99351603600268e-06,
      "loss": 0.88892156,
      "num_input_tokens_seen": 10490685,
      "step": 494,
      "time_per_iteration": 3.902836322784424
    },
    {
      "auxiliary_loss_clip": 0.01287378,
      "auxiliary_loss_mlp": 0.01083918,
      "balance_loss_clip": 1.0839076,
      "balance_loss_mlp": 1.05261397,
      "epoch": 0.02976101007064482,
      "flos": 22236857233920.0,
      "grad_norm": 2.1647534696936215,
      "language_loss": 0.86475515,
      "learning_rate": 3.994818063106668e-06,
      "loss": 0.88846815,
      "num_input_tokens_seen": 10509435,
      "step": 495,
      "time_per_iteration": 2.514432191848755
    },
    {
      "auxiliary_loss_clip": 0.01275081,
      "auxiliary_loss_mlp": 0.01075853,
      "balance_loss_clip": 1.08007073,
      "balance_loss_mlp": 1.04328489,
      "epoch": 0.029821133323312793,
      "flos": 23732644320000.0,
      "grad_norm": 2.059852336325344,
      "language_loss": 0.62328988,
      "learning_rate": 3.99611746250533e-06,
      "loss": 0.64679921,
      "num_input_tokens_seen": 10530050,
      "step": 496,
      "time_per_iteration": 2.5275046825408936
    },
    {
      "auxiliary_loss_clip": 0.01279255,
      "auxiliary_loss_mlp": 0.01086962,
      "balance_loss_clip": 1.08461344,
      "balance_loss_mlp": 1.05448961,
      "epoch": 0.02988125657598076,
      "flos": 22419498913920.0,
      "grad_norm": 2.5208037597506583,
      "language_loss": 0.88890362,
      "learning_rate": 3.997414244783595e-06,
      "loss": 0.91256583,
      "num_input_tokens_seen": 10551370,
      "step": 497,
      "time_per_iteration": 2.5553793907165527
    },
    {
      "auxiliary_loss_clip": 0.01282831,
      "auxiliary_loss_mlp": 0.01079375,
      "balance_loss_clip": 1.08253932,
      "balance_loss_mlp": 1.0456146,
      "epoch": 0.02994137982864873,
      "flos": 13845108453120.0,
      "grad_norm": 12.174860950256292,
      "language_loss": 0.8480438,
      "learning_rate": 3.998708420462557e-06,
      "loss": 0.87166584,
      "num_input_tokens_seen": 10569225,
      "step": 498,
      "time_per_iteration": 2.4612936973571777
    },
    {
      "auxiliary_loss_clip": 0.0127958,
      "auxiliary_loss_mlp": 0.01081799,
      "balance_loss_clip": 1.08111262,
      "balance_loss_mlp": 1.04994583,
      "epoch": 0.0300015030813167,
      "flos": 23908354675200.0,
      "grad_norm": 4.883971593763153,
      "language_loss": 0.78737938,
      "learning_rate": 4e-06,
      "loss": 0.81099319,
      "num_input_tokens_seen": 10586170,
      "step": 499,
      "time_per_iteration": 2.510436534881592
    },
    {
      "auxiliary_loss_clip": 0.01282371,
      "auxiliary_loss_mlp": 0.01084829,
      "balance_loss_clip": 1.0840838,
      "balance_loss_mlp": 1.05197501,
      "epoch": 0.030061626333984667,
      "flos": 22016796560640.0,
      "grad_norm": 2.554895241732742,
      "language_loss": 0.82729942,
      "learning_rate": 3.9999999620799e-06,
      "loss": 0.85097146,
      "num_input_tokens_seen": 10606205,
      "step": 500,
      "time_per_iteration": 2.5825767517089844
    },
    {
      "auxiliary_loss_clip": 0.01273273,
      "auxiliary_loss_mlp": 0.0108432,
      "balance_loss_clip": 1.07897866,
      "balance_loss_mlp": 1.04927194,
      "epoch": 0.03012174958665264,
      "flos": 23039747988480.0,
      "grad_norm": 2.70584552463603,
      "language_loss": 0.88422096,
      "learning_rate": 3.9999998483196e-06,
      "loss": 0.90779686,
      "num_input_tokens_seen": 10625995,
      "step": 501,
      "time_per_iteration": 2.4917614459991455
    },
    {
      "auxiliary_loss_clip": 0.01283542,
      "auxiliary_loss_mlp": 0.01075952,
      "balance_loss_clip": 1.08245063,
      "balance_loss_mlp": 1.04388428,
      "epoch": 0.030181872839320608,
      "flos": 18953257489920.0,
      "grad_norm": 2.5946176285018385,
      "language_loss": 0.87230331,
      "learning_rate": 3.9999996587191065e-06,
      "loss": 0.89589822,
      "num_input_tokens_seen": 10644105,
      "step": 502,
      "time_per_iteration": 2.4917454719543457
    },
    {
      "auxiliary_loss_clip": 0.01277821,
      "auxiliary_loss_mlp": 0.01079999,
      "balance_loss_clip": 1.08339572,
      "balance_loss_mlp": 1.04745483,
      "epoch": 0.030241996091988577,
      "flos": 16728017005440.0,
      "grad_norm": 2.489175126715502,
      "language_loss": 0.84597397,
      "learning_rate": 3.999999393278425e-06,
      "loss": 0.86955214,
      "num_input_tokens_seen": 10661090,
      "step": 503,
      "time_per_iteration": 2.4932639598846436
    },
    {
      "auxiliary_loss_clip": 0.01271936,
      "auxiliary_loss_mlp": 0.01086758,
      "balance_loss_clip": 1.08046758,
      "balance_loss_mlp": 1.05407047,
      "epoch": 0.030302119344656545,
      "flos": 28621271387520.0,
      "grad_norm": 1.5807164771259137,
      "language_loss": 0.88186139,
      "learning_rate": 3.999999051997567e-06,
      "loss": 0.90544832,
      "num_input_tokens_seen": 10682380,
      "step": 504,
      "time_per_iteration": 2.56295108795166
    },
    {
      "auxiliary_loss_clip": 0.01275223,
      "auxiliary_loss_mlp": 0.01091183,
      "balance_loss_clip": 1.08020985,
      "balance_loss_mlp": 1.05842459,
      "epoch": 0.030362242597324514,
      "flos": 15669334523520.0,
      "grad_norm": 2.3838600279076,
      "language_loss": 0.78126585,
      "learning_rate": 3.9999986348765425e-06,
      "loss": 0.80492997,
      "num_input_tokens_seen": 10699925,
      "step": 505,
      "time_per_iteration": 2.480445146560669
    },
    {
      "auxiliary_loss_clip": 0.01140739,
      "auxiliary_loss_mlp": 0.01028196,
      "balance_loss_clip": 1.04706335,
      "balance_loss_mlp": 1.02061462,
      "epoch": 0.030422365849992486,
      "flos": 72125973676800.0,
      "grad_norm": 0.870952608927018,
      "language_loss": 0.5500617,
      "learning_rate": 3.999998141915371e-06,
      "loss": 0.571751,
      "num_input_tokens_seen": 10766525,
      "step": 506,
      "time_per_iteration": 3.2477879524230957
    },
    {
      "auxiliary_loss_clip": 0.01274462,
      "auxiliary_loss_mlp": 0.01084991,
      "balance_loss_clip": 1.07882833,
      "balance_loss_mlp": 1.05254245,
      "epoch": 0.030482489102660455,
      "flos": 19427817000960.0,
      "grad_norm": 2.0466558741268903,
      "language_loss": 0.83180654,
      "learning_rate": 3.999997573114069e-06,
      "loss": 0.85540104,
      "num_input_tokens_seen": 10786725,
      "step": 507,
      "time_per_iteration": 2.5006020069122314
    },
    {
      "auxiliary_loss_clip": 0.01278781,
      "auxiliary_loss_mlp": 0.0107368,
      "balance_loss_clip": 1.08043861,
      "balance_loss_mlp": 1.04156542,
      "epoch": 0.030542612355328423,
      "flos": 20375822701440.0,
      "grad_norm": 2.3023127570760886,
      "language_loss": 0.88539124,
      "learning_rate": 3.999996928472659e-06,
      "loss": 0.90891588,
      "num_input_tokens_seen": 10805390,
      "step": 508,
      "time_per_iteration": 2.491718053817749
    },
    {
      "auxiliary_loss_clip": 0.01281478,
      "auxiliary_loss_mlp": 0.01069494,
      "balance_loss_clip": 1.08113074,
      "balance_loss_mlp": 1.03590083,
      "epoch": 0.030602735607996392,
      "flos": 34677354297600.0,
      "grad_norm": 2.393352617223587,
      "language_loss": 0.71333444,
      "learning_rate": 3.999996207991165e-06,
      "loss": 0.73684418,
      "num_input_tokens_seen": 10828030,
      "step": 509,
      "time_per_iteration": 2.6113429069519043
    },
    {
      "auxiliary_loss_clip": 0.01273273,
      "auxiliary_loss_mlp": 0.01070398,
      "balance_loss_clip": 1.08074403,
      "balance_loss_mlp": 1.03961766,
      "epoch": 0.03066285886066436,
      "flos": 23658668259840.0,
      "grad_norm": 1.9777524453834636,
      "language_loss": 0.82129699,
      "learning_rate": 3.999995411669614e-06,
      "loss": 0.84473372,
      "num_input_tokens_seen": 10845240,
      "step": 510,
      "time_per_iteration": 2.497653007507324
    },
    {
      "auxiliary_loss_clip": 0.01279349,
      "auxiliary_loss_mlp": 0.01078315,
      "balance_loss_clip": 1.08482146,
      "balance_loss_mlp": 1.04567599,
      "epoch": 0.030722982113332332,
      "flos": 23002975440000.0,
      "grad_norm": 2.2365081371497673,
      "language_loss": 0.83666289,
      "learning_rate": 3.999994539508036e-06,
      "loss": 0.86023962,
      "num_input_tokens_seen": 10864325,
      "step": 511,
      "time_per_iteration": 2.5393166542053223
    },
    {
      "auxiliary_loss_clip": 0.01279135,
      "auxiliary_loss_mlp": 0.01079795,
      "balance_loss_clip": 1.07971609,
      "balance_loss_mlp": 1.04834712,
      "epoch": 0.0307831053660003,
      "flos": 24750855152640.0,
      "grad_norm": 2.1820433155217915,
      "language_loss": 0.81997889,
      "learning_rate": 3.9999935915064655e-06,
      "loss": 0.84356821,
      "num_input_tokens_seen": 10883860,
      "step": 512,
      "time_per_iteration": 2.5206189155578613
    },
    {
      "auxiliary_loss_clip": 0.01276475,
      "auxiliary_loss_mlp": 0.01080265,
      "balance_loss_clip": 1.07971728,
      "balance_loss_mlp": 1.04769683,
      "epoch": 0.03084322861866827,
      "flos": 26140885620480.0,
      "grad_norm": 1.8765872012861904,
      "language_loss": 0.86925894,
      "learning_rate": 3.9999925676649374e-06,
      "loss": 0.89282638,
      "num_input_tokens_seen": 10904555,
      "step": 513,
      "time_per_iteration": 2.5361359119415283
    },
    {
      "auxiliary_loss_clip": 0.01284275,
      "auxiliary_loss_mlp": 0.01082116,
      "balance_loss_clip": 1.08333755,
      "balance_loss_mlp": 1.04909563,
      "epoch": 0.03090335187133624,
      "flos": 18771298168320.0,
      "grad_norm": 1.6939573488395416,
      "language_loss": 0.79182673,
      "learning_rate": 3.999991467983491e-06,
      "loss": 0.8154906,
      "num_input_tokens_seen": 10923700,
      "step": 514,
      "time_per_iteration": 2.4916253089904785
    },
    {
      "auxiliary_loss_clip": 0.01275758,
      "auxiliary_loss_mlp": 0.01064672,
      "balance_loss_clip": 1.08312798,
      "balance_loss_mlp": 1.03422558,
      "epoch": 0.030963475124004207,
      "flos": 23221886878080.0,
      "grad_norm": 2.73808374141246,
      "language_loss": 0.7710793,
      "learning_rate": 3.999990292462167e-06,
      "loss": 0.79448354,
      "num_input_tokens_seen": 10942730,
      "step": 515,
      "time_per_iteration": 2.5126349925994873
    },
    {
      "auxiliary_loss_clip": 0.0127282,
      "auxiliary_loss_mlp": 0.01068888,
      "balance_loss_clip": 1.07649148,
      "balance_loss_mlp": 1.03615308,
      "epoch": 0.03102359837667218,
      "flos": 42525595411200.0,
      "grad_norm": 2.000979851494337,
      "language_loss": 0.82705641,
      "learning_rate": 3.999989041101011e-06,
      "loss": 0.85047352,
      "num_input_tokens_seen": 10967120,
      "step": 516,
      "time_per_iteration": 2.6882431507110596
    },
    {
      "auxiliary_loss_clip": 0.01272552,
      "auxiliary_loss_mlp": 0.0107441,
      "balance_loss_clip": 1.07974148,
      "balance_loss_mlp": 1.04117489,
      "epoch": 0.031083721629340148,
      "flos": 21176953689600.0,
      "grad_norm": 2.778314691075019,
      "language_loss": 0.78658086,
      "learning_rate": 3.999987713900071e-06,
      "loss": 0.81005049,
      "num_input_tokens_seen": 10986775,
      "step": 517,
      "time_per_iteration": 2.527066707611084
    },
    {
      "auxiliary_loss_clip": 0.01269635,
      "auxiliary_loss_mlp": 0.0107296,
      "balance_loss_clip": 1.0796026,
      "balance_loss_mlp": 1.04105997,
      "epoch": 0.031143844882008116,
      "flos": 29716187713920.0,
      "grad_norm": 1.984569413089594,
      "language_loss": 0.90648293,
      "learning_rate": 3.999986310859396e-06,
      "loss": 0.92990887,
      "num_input_tokens_seen": 11011360,
      "step": 518,
      "time_per_iteration": 2.561100482940674
    },
    {
      "auxiliary_loss_clip": 0.01282975,
      "auxiliary_loss_mlp": 0.01094649,
      "balance_loss_clip": 1.08737111,
      "balance_loss_mlp": 1.05976832,
      "epoch": 0.031203968134676085,
      "flos": 23112467072640.0,
      "grad_norm": 1.9318406331711715,
      "language_loss": 0.86248553,
      "learning_rate": 3.999984831979039e-06,
      "loss": 0.8862617,
      "num_input_tokens_seen": 11030150,
      "step": 519,
      "time_per_iteration": 2.522749900817871
    },
    {
      "auxiliary_loss_clip": 0.01279583,
      "auxiliary_loss_mlp": 0.010891,
      "balance_loss_clip": 1.07926714,
      "balance_loss_mlp": 1.05750918,
      "epoch": 0.03126409138734405,
      "flos": 20954379064320.0,
      "grad_norm": 2.0505355165339667,
      "language_loss": 0.86720949,
      "learning_rate": 3.999983277259057e-06,
      "loss": 0.89089626,
      "num_input_tokens_seen": 11049145,
      "step": 520,
      "time_per_iteration": 2.499417304992676
    },
    {
      "auxiliary_loss_clip": 0.01281129,
      "auxiliary_loss_mlp": 0.01090228,
      "balance_loss_clip": 1.08172321,
      "balance_loss_mlp": 1.05734992,
      "epoch": 0.031324214640012026,
      "flos": 21650112570240.0,
      "grad_norm": 2.0575898356951234,
      "language_loss": 0.89220858,
      "learning_rate": 3.999981646699509e-06,
      "loss": 0.91592216,
      "num_input_tokens_seen": 11068835,
      "step": 521,
      "time_per_iteration": 2.5391504764556885
    },
    {
      "auxiliary_loss_clip": 0.01273869,
      "auxiliary_loss_mlp": 0.01084906,
      "balance_loss_clip": 1.08069706,
      "balance_loss_mlp": 1.05090737,
      "epoch": 0.03138433789267999,
      "flos": 23441337020160.0,
      "grad_norm": 2.046365236274203,
      "language_loss": 0.71245646,
      "learning_rate": 3.999979940300456e-06,
      "loss": 0.73604417,
      "num_input_tokens_seen": 11088980,
      "step": 522,
      "time_per_iteration": 2.5181522369384766
    },
    {
      "auxiliary_loss_clip": 0.01277435,
      "auxiliary_loss_mlp": 0.01087087,
      "balance_loss_clip": 1.07822466,
      "balance_loss_mlp": 1.05554461,
      "epoch": 0.03144446114534796,
      "flos": 18982164960000.0,
      "grad_norm": 3.235221947515852,
      "language_loss": 0.84783804,
      "learning_rate": 3.999978158061963e-06,
      "loss": 0.87148327,
      "num_input_tokens_seen": 11104300,
      "step": 523,
      "time_per_iteration": 2.472031354904175
    },
    {
      "auxiliary_loss_clip": 0.01282311,
      "auxiliary_loss_mlp": 0.01075782,
      "balance_loss_clip": 1.08011234,
      "balance_loss_mlp": 1.04261756,
      "epoch": 0.031504584398015935,
      "flos": 22637692080000.0,
      "grad_norm": 3.53419866585608,
      "language_loss": 0.90326703,
      "learning_rate": 3.999976299984099e-06,
      "loss": 0.92684793,
      "num_input_tokens_seen": 11123335,
      "step": 524,
      "time_per_iteration": 2.509885549545288
    },
    {
      "auxiliary_loss_clip": 0.01285763,
      "auxiliary_loss_mlp": 0.01084426,
      "balance_loss_clip": 1.08335018,
      "balance_loss_mlp": 1.05066633,
      "epoch": 0.0315647076506839,
      "flos": 25297056339840.0,
      "grad_norm": 3.1148100584455056,
      "language_loss": 0.80058461,
      "learning_rate": 3.999974366066933e-06,
      "loss": 0.82428652,
      "num_input_tokens_seen": 11140880,
      "step": 525,
      "time_per_iteration": 2.519080877304077
    },
    {
      "auxiliary_loss_clip": 0.012756,
      "auxiliary_loss_mlp": 0.01083036,
      "balance_loss_clip": 1.0773108,
      "balance_loss_mlp": 1.05046833,
      "epoch": 0.03162483090335187,
      "flos": 16982839065600.0,
      "grad_norm": 2.0852517719392165,
      "language_loss": 0.80368161,
      "learning_rate": 3.999972356310538e-06,
      "loss": 0.827268,
      "num_input_tokens_seen": 11158710,
      "step": 526,
      "time_per_iteration": 2.4859261512756348
    },
    {
      "auxiliary_loss_clip": 0.01285138,
      "auxiliary_loss_mlp": 0.010677,
      "balance_loss_clip": 1.08423853,
      "balance_loss_mlp": 1.03329587,
      "epoch": 0.03168495415601984,
      "flos": 18734489706240.0,
      "grad_norm": 2.800288459454659,
      "language_loss": 0.81673521,
      "learning_rate": 3.999970270714991e-06,
      "loss": 0.84026361,
      "num_input_tokens_seen": 11177550,
      "step": 527,
      "time_per_iteration": 3.8811967372894287
    },
    {
      "auxiliary_loss_clip": 0.01271994,
      "auxiliary_loss_mlp": 0.01085217,
      "balance_loss_clip": 1.07671165,
      "balance_loss_mlp": 1.05162358,
      "epoch": 0.03174507740868781,
      "flos": 21214875473280.0,
      "grad_norm": 1.9558711426931759,
      "language_loss": 0.93774784,
      "learning_rate": 3.999968109280371e-06,
      "loss": 0.96131992,
      "num_input_tokens_seen": 11196230,
      "step": 528,
      "time_per_iteration": 2.4739744663238525
    },
    {
      "auxiliary_loss_clip": 0.01273306,
      "auxiliary_loss_mlp": 0.01074173,
      "balance_loss_clip": 1.07699466,
      "balance_loss_mlp": 1.04198682,
      "epoch": 0.03180520066135578,
      "flos": 24787663614720.0,
      "grad_norm": 1.87565134130626,
      "language_loss": 0.83922398,
      "learning_rate": 3.99996587200676e-06,
      "loss": 0.86269873,
      "num_input_tokens_seen": 11214935,
      "step": 529,
      "time_per_iteration": 5.339153289794922
    },
    {
      "auxiliary_loss_clip": 0.01277911,
      "auxiliary_loss_mlp": 0.01083384,
      "balance_loss_clip": 1.08498645,
      "balance_loss_mlp": 1.05182922,
      "epoch": 0.03186532391402375,
      "flos": 24864261367680.0,
      "grad_norm": 1.9475299213372474,
      "language_loss": 0.90660334,
      "learning_rate": 3.999963558894243e-06,
      "loss": 0.93021625,
      "num_input_tokens_seen": 11235310,
      "step": 530,
      "time_per_iteration": 2.5239710807800293
    },
    {
      "auxiliary_loss_clip": 0.01272192,
      "auxiliary_loss_mlp": 0.01075399,
      "balance_loss_clip": 1.07449484,
      "balance_loss_mlp": 1.04109061,
      "epoch": 0.03192544716669172,
      "flos": 21215055041280.0,
      "grad_norm": 2.201504003702701,
      "language_loss": 0.76071054,
      "learning_rate": 3.999961169942907e-06,
      "loss": 0.78418648,
      "num_input_tokens_seen": 11254425,
      "step": 531,
      "time_per_iteration": 2.4894371032714844
    },
    {
      "auxiliary_loss_clip": 0.01270717,
      "auxiliary_loss_mlp": 0.01061048,
      "balance_loss_clip": 1.07549882,
      "balance_loss_mlp": 1.02859974,
      "epoch": 0.03198557041935969,
      "flos": 24353216616960.0,
      "grad_norm": 3.073395398223609,
      "language_loss": 0.90312344,
      "learning_rate": 3.999958705152843e-06,
      "loss": 0.92644107,
      "num_input_tokens_seen": 11274595,
      "step": 532,
      "time_per_iteration": 3.9291367530822754
    },
    {
      "auxiliary_loss_clip": 0.01139221,
      "auxiliary_loss_mlp": 0.01013038,
      "balance_loss_clip": 1.04662085,
      "balance_loss_mlp": 1.00555122,
      "epoch": 0.032045693672027656,
      "flos": 61827367587840.0,
      "grad_norm": 0.7398484017489936,
      "language_loss": 0.57966787,
      "learning_rate": 3.9999561645241445e-06,
      "loss": 0.60119051,
      "num_input_tokens_seen": 11336705,
      "step": 533,
      "time_per_iteration": 3.1299264430999756
    },
    {
      "auxiliary_loss_clip": 0.01270256,
      "auxiliary_loss_mlp": 0.0108017,
      "balance_loss_clip": 1.07609022,
      "balance_loss_mlp": 1.04848409,
      "epoch": 0.03210581692469563,
      "flos": 28401174800640.0,
      "grad_norm": 1.984427574189356,
      "language_loss": 0.86596537,
      "learning_rate": 3.999953548056907e-06,
      "loss": 0.88946962,
      "num_input_tokens_seen": 11356820,
      "step": 534,
      "time_per_iteration": 2.548802137374878
    },
    {
      "auxiliary_loss_clip": 0.01270592,
      "auxiliary_loss_mlp": 0.0106526,
      "balance_loss_clip": 1.07795644,
      "balance_loss_mlp": 1.03414667,
      "epoch": 0.03216594017736359,
      "flos": 24717709877760.0,
      "grad_norm": 2.8101864068775853,
      "language_loss": 0.77494359,
      "learning_rate": 3.999950855751232e-06,
      "loss": 0.79830205,
      "num_input_tokens_seen": 11376645,
      "step": 535,
      "time_per_iteration": 2.5143158435821533
    },
    {
      "auxiliary_loss_clip": 0.01273824,
      "auxiliary_loss_mlp": 0.01080524,
      "balance_loss_clip": 1.07955217,
      "balance_loss_mlp": 1.04905248,
      "epoch": 0.032226063430031565,
      "flos": 31175453646720.0,
      "grad_norm": 2.21633001275123,
      "language_loss": 0.80781323,
      "learning_rate": 3.999948087607219e-06,
      "loss": 0.8313567,
      "num_input_tokens_seen": 11397310,
      "step": 536,
      "time_per_iteration": 2.571321964263916
    },
    {
      "auxiliary_loss_clip": 0.0127437,
      "auxiliary_loss_mlp": 0.01073678,
      "balance_loss_clip": 1.07974696,
      "balance_loss_mlp": 1.04039514,
      "epoch": 0.03228618668269954,
      "flos": 32198225506560.0,
      "grad_norm": 3.1319395201691136,
      "language_loss": 0.70305437,
      "learning_rate": 3.999945243624975e-06,
      "loss": 0.72653484,
      "num_input_tokens_seen": 11418475,
      "step": 537,
      "time_per_iteration": 2.5643370151519775
    },
    {
      "auxiliary_loss_clip": 0.01274361,
      "auxiliary_loss_mlp": 0.0107843,
      "balance_loss_clip": 1.08375478,
      "balance_loss_mlp": 1.04705405,
      "epoch": 0.0323463099353675,
      "flos": 22670154996480.0,
      "grad_norm": 2.270865022404991,
      "language_loss": 0.828614,
      "learning_rate": 3.999942323804607e-06,
      "loss": 0.85214186,
      "num_input_tokens_seen": 11436630,
      "step": 538,
      "time_per_iteration": 2.5318655967712402
    },
    {
      "auxiliary_loss_clip": 0.01282063,
      "auxiliary_loss_mlp": 0.01079965,
      "balance_loss_clip": 1.08134007,
      "balance_loss_mlp": 1.04801726,
      "epoch": 0.032406433188035474,
      "flos": 26905172232960.0,
      "grad_norm": 1.7570041660039513,
      "language_loss": 0.79285616,
      "learning_rate": 3.999939328146225e-06,
      "loss": 0.81647646,
      "num_input_tokens_seen": 11457275,
      "step": 539,
      "time_per_iteration": 2.5331685543060303
    },
    {
      "auxiliary_loss_clip": 0.01271826,
      "auxiliary_loss_mlp": 0.01070166,
      "balance_loss_clip": 1.07804561,
      "balance_loss_mlp": 1.03652549,
      "epoch": 0.03246655644070344,
      "flos": 31503928544640.0,
      "grad_norm": 2.1578675588173466,
      "language_loss": 0.77643526,
      "learning_rate": 3.999936256649943e-06,
      "loss": 0.79985517,
      "num_input_tokens_seen": 11476925,
      "step": 540,
      "time_per_iteration": 2.5724873542785645
    },
    {
      "auxiliary_loss_clip": 0.01281682,
      "auxiliary_loss_mlp": 0.01070801,
      "balance_loss_clip": 1.08246756,
      "balance_loss_mlp": 1.03882873,
      "epoch": 0.03252667969337141,
      "flos": 23218331431680.0,
      "grad_norm": 2.097653839066907,
      "language_loss": 0.85517079,
      "learning_rate": 3.999933109315878e-06,
      "loss": 0.87869561,
      "num_input_tokens_seen": 11496830,
      "step": 541,
      "time_per_iteration": 2.495028018951416
    },
    {
      "auxiliary_loss_clip": 0.01269965,
      "auxiliary_loss_mlp": 0.01079159,
      "balance_loss_clip": 1.07954597,
      "balance_loss_mlp": 1.04613841,
      "epoch": 0.032586802946039384,
      "flos": 14757454926720.0,
      "grad_norm": 2.4102764613127583,
      "language_loss": 0.89100009,
      "learning_rate": 3.9999298861441496e-06,
      "loss": 0.91449136,
      "num_input_tokens_seen": 11515605,
      "step": 542,
      "time_per_iteration": 2.490851402282715
    },
    {
      "auxiliary_loss_clip": 0.01274224,
      "auxiliary_loss_mlp": 0.0107854,
      "balance_loss_clip": 1.07985973,
      "balance_loss_mlp": 1.04641259,
      "epoch": 0.03264692619870735,
      "flos": 24280677100800.0,
      "grad_norm": 1.8370824708215936,
      "language_loss": 0.71028399,
      "learning_rate": 3.999926587134879e-06,
      "loss": 0.73381162,
      "num_input_tokens_seen": 11536230,
      "step": 543,
      "time_per_iteration": 2.5286896228790283
    },
    {
      "auxiliary_loss_clip": 0.01268757,
      "auxiliary_loss_mlp": 0.01088585,
      "balance_loss_clip": 1.07183158,
      "balance_loss_mlp": 1.0565412,
      "epoch": 0.03270704945137532,
      "flos": 22893160584960.0,
      "grad_norm": 3.1021649006178613,
      "language_loss": 0.91572207,
      "learning_rate": 3.999923212288192e-06,
      "loss": 0.93929553,
      "num_input_tokens_seen": 11554715,
      "step": 544,
      "time_per_iteration": 2.4921436309814453
    },
    {
      "auxiliary_loss_clip": 0.01274905,
      "auxiliary_loss_mlp": 0.01083245,
      "balance_loss_clip": 1.0798986,
      "balance_loss_mlp": 1.05353785,
      "epoch": 0.032767172704043286,
      "flos": 18041018757120.0,
      "grad_norm": 3.7502744098743688,
      "language_loss": 0.65763068,
      "learning_rate": 3.999919761604216e-06,
      "loss": 0.68121219,
      "num_input_tokens_seen": 11571370,
      "step": 545,
      "time_per_iteration": 2.4578700065612793
    },
    {
      "auxiliary_loss_clip": 0.01271545,
      "auxiliary_loss_mlp": 0.01067537,
      "balance_loss_clip": 1.07631397,
      "balance_loss_mlp": 1.03656638,
      "epoch": 0.03282729595671126,
      "flos": 22528739151360.0,
      "grad_norm": 2.139599197895755,
      "language_loss": 0.91862822,
      "learning_rate": 3.999916235083083e-06,
      "loss": 0.94201905,
      "num_input_tokens_seen": 11588560,
      "step": 546,
      "time_per_iteration": 2.5076115131378174
    },
    {
      "auxiliary_loss_clip": 0.0126959,
      "auxiliary_loss_mlp": 0.01070576,
      "balance_loss_clip": 1.07379556,
      "balance_loss_mlp": 1.03765059,
      "epoch": 0.03288741920937923,
      "flos": 20410620001920.0,
      "grad_norm": 2.3647671834954482,
      "language_loss": 0.81899041,
      "learning_rate": 3.999912632724925e-06,
      "loss": 0.84239209,
      "num_input_tokens_seen": 11605685,
      "step": 547,
      "time_per_iteration": 2.4750301837921143
    },
    {
      "auxiliary_loss_clip": 0.01271216,
      "auxiliary_loss_mlp": 0.01071361,
      "balance_loss_clip": 1.07726562,
      "balance_loss_mlp": 1.03912687,
      "epoch": 0.032947542462047195,
      "flos": 20777986350720.0,
      "grad_norm": 1.76780359875636,
      "language_loss": 0.81039995,
      "learning_rate": 3.999908954529881e-06,
      "loss": 0.83382571,
      "num_input_tokens_seen": 11626290,
      "step": 548,
      "time_per_iteration": 2.5423567295074463
    },
    {
      "auxiliary_loss_clip": 0.01271097,
      "auxiliary_loss_mlp": 0.01078742,
      "balance_loss_clip": 1.07667351,
      "balance_loss_mlp": 1.04491043,
      "epoch": 0.03300766571471517,
      "flos": 19901263190400.0,
      "grad_norm": 3.070341316265579,
      "language_loss": 0.67900735,
      "learning_rate": 3.999905200498087e-06,
      "loss": 0.70250577,
      "num_input_tokens_seen": 11643950,
      "step": 549,
      "time_per_iteration": 2.461216688156128
    },
    {
      "auxiliary_loss_clip": 0.01264902,
      "auxiliary_loss_mlp": 0.01075252,
      "balance_loss_clip": 1.0763402,
      "balance_loss_mlp": 1.04375708,
      "epoch": 0.03306778896738313,
      "flos": 17967760968960.0,
      "grad_norm": 1.955900987754541,
      "language_loss": 0.86177558,
      "learning_rate": 3.999901370629689e-06,
      "loss": 0.88517714,
      "num_input_tokens_seen": 11662560,
      "step": 550,
      "time_per_iteration": 2.47871732711792
    },
    {
      "auxiliary_loss_clip": 0.01274555,
      "auxiliary_loss_mlp": 0.01086557,
      "balance_loss_clip": 1.08146811,
      "balance_loss_mlp": 1.05499077,
      "epoch": 0.033127912220051105,
      "flos": 21653380707840.0,
      "grad_norm": 1.8347107206490618,
      "language_loss": 0.80992532,
      "learning_rate": 3.99989746492483e-06,
      "loss": 0.83353639,
      "num_input_tokens_seen": 11682265,
      "step": 551,
      "time_per_iteration": 2.4783577919006348
    },
    {
      "auxiliary_loss_clip": 0.01279485,
      "auxiliary_loss_mlp": 0.01085097,
      "balance_loss_clip": 1.08024275,
      "balance_loss_mlp": 1.05300593,
      "epoch": 0.03318803547271908,
      "flos": 30188376927360.0,
      "grad_norm": 5.198533295083856,
      "language_loss": 0.8628543,
      "learning_rate": 3.999893483383658e-06,
      "loss": 0.88650012,
      "num_input_tokens_seen": 11699300,
      "step": 552,
      "time_per_iteration": 2.5497093200683594
    },
    {
      "auxiliary_loss_clip": 0.01276707,
      "auxiliary_loss_mlp": 0.01079008,
      "balance_loss_clip": 1.08169341,
      "balance_loss_mlp": 1.04500961,
      "epoch": 0.03324815872538704,
      "flos": 20376038183040.0,
      "grad_norm": 3.35803272880239,
      "language_loss": 0.93045783,
      "learning_rate": 3.999889426006326e-06,
      "loss": 0.95401502,
      "num_input_tokens_seen": 11716955,
      "step": 553,
      "time_per_iteration": 2.481351852416992
    },
    {
      "auxiliary_loss_clip": 0.01271001,
      "auxiliary_loss_mlp": 0.0107448,
      "balance_loss_clip": 1.07799721,
      "balance_loss_mlp": 1.04045725,
      "epoch": 0.033308281978055014,
      "flos": 24494560634880.0,
      "grad_norm": 5.347904269112907,
      "language_loss": 0.7888574,
      "learning_rate": 3.999885292792986e-06,
      "loss": 0.81231219,
      "num_input_tokens_seen": 11736130,
      "step": 554,
      "time_per_iteration": 2.5355350971221924
    },
    {
      "auxiliary_loss_clip": 0.01267049,
      "auxiliary_loss_mlp": 0.01081849,
      "balance_loss_clip": 1.07656705,
      "balance_loss_mlp": 1.04737329,
      "epoch": 0.03336840523072298,
      "flos": 23400326666880.0,
      "grad_norm": 2.502474885874672,
      "language_loss": 0.82027483,
      "learning_rate": 3.999881083743795e-06,
      "loss": 0.84376377,
      "num_input_tokens_seen": 11754425,
      "step": 555,
      "time_per_iteration": 2.480738639831543
    },
    {
      "auxiliary_loss_clip": 0.01272573,
      "auxiliary_loss_mlp": 0.0107962,
      "balance_loss_clip": 1.07746768,
      "balance_loss_mlp": 1.0466466,
      "epoch": 0.03342852848339095,
      "flos": 30550571717760.0,
      "grad_norm": 2.6150130205433646,
      "language_loss": 0.88445431,
      "learning_rate": 3.999876798858914e-06,
      "loss": 0.90797627,
      "num_input_tokens_seen": 11772845,
      "step": 556,
      "time_per_iteration": 2.5686826705932617
    },
    {
      "auxiliary_loss_clip": 0.01270657,
      "auxiliary_loss_mlp": 0.01086897,
      "balance_loss_clip": 1.07758415,
      "balance_loss_mlp": 1.05289841,
      "epoch": 0.03348865173605892,
      "flos": 22893304239360.0,
      "grad_norm": 2.65398686665017,
      "language_loss": 0.83658433,
      "learning_rate": 3.999872438138503e-06,
      "loss": 0.86015987,
      "num_input_tokens_seen": 11792850,
      "step": 557,
      "time_per_iteration": 2.4996397495269775
    },
    {
      "auxiliary_loss_clip": 0.0127623,
      "auxiliary_loss_mlp": 0.01065361,
      "balance_loss_clip": 1.08202505,
      "balance_loss_mlp": 1.03388989,
      "epoch": 0.03354877498872689,
      "flos": 17676022705920.0,
      "grad_norm": 3.071701654278524,
      "language_loss": 0.94121724,
      "learning_rate": 3.999868001582729e-06,
      "loss": 0.96463311,
      "num_input_tokens_seen": 11809670,
      "step": 558,
      "time_per_iteration": 2.469963312149048
    },
    {
      "auxiliary_loss_clip": 0.01267375,
      "auxiliary_loss_mlp": 0.01073513,
      "balance_loss_clip": 1.07459259,
      "balance_loss_mlp": 1.04099238,
      "epoch": 0.03360889824139486,
      "flos": 21652985658240.0,
      "grad_norm": 2.1421963364554015,
      "language_loss": 0.77135956,
      "learning_rate": 3.99986348919176e-06,
      "loss": 0.79476845,
      "num_input_tokens_seen": 11829665,
      "step": 559,
      "time_per_iteration": 2.4862797260284424
    },
    {
      "auxiliary_loss_clip": 0.01270665,
      "auxiliary_loss_mlp": 0.0108025,
      "balance_loss_clip": 1.07785761,
      "balance_loss_mlp": 1.04868329,
      "epoch": 0.033669021494062826,
      "flos": 21795730306560.0,
      "grad_norm": 2.2136735764504754,
      "language_loss": 0.87594247,
      "learning_rate": 3.9998589009657675e-06,
      "loss": 0.89945161,
      "num_input_tokens_seen": 11848190,
      "step": 560,
      "time_per_iteration": 2.562744140625
    },
    {
      "auxiliary_loss_clip": 0.01264831,
      "auxiliary_loss_mlp": 0.01067122,
      "balance_loss_clip": 1.07568228,
      "balance_loss_mlp": 1.03689086,
      "epoch": 0.0337291447467308,
      "flos": 21866222747520.0,
      "grad_norm": 4.9504778131490115,
      "language_loss": 0.81872606,
      "learning_rate": 3.999854236904925e-06,
      "loss": 0.84204555,
      "num_input_tokens_seen": 11864795,
      "step": 561,
      "time_per_iteration": 2.4801275730133057
    },
    {
      "auxiliary_loss_clip": 0.01265414,
      "auxiliary_loss_mlp": 0.0107009,
      "balance_loss_clip": 1.07690668,
      "balance_loss_mlp": 1.03911924,
      "epoch": 0.03378926799939877,
      "flos": 24245951627520.0,
      "grad_norm": 1.7699075076382416,
      "language_loss": 0.82244611,
      "learning_rate": 3.999849497009409e-06,
      "loss": 0.84580117,
      "num_input_tokens_seen": 11885275,
      "step": 562,
      "time_per_iteration": 2.519102096557617
    },
    {
      "auxiliary_loss_clip": 0.01272253,
      "auxiliary_loss_mlp": 0.0108053,
      "balance_loss_clip": 1.0792439,
      "balance_loss_mlp": 1.04862928,
      "epoch": 0.033849391252066735,
      "flos": 16507812677760.0,
      "grad_norm": 2.243640090035936,
      "language_loss": 0.84315574,
      "learning_rate": 3.999844681279401e-06,
      "loss": 0.8666836,
      "num_input_tokens_seen": 11903595,
      "step": 563,
      "time_per_iteration": 2.4492974281311035
    },
    {
      "auxiliary_loss_clip": 0.01270524,
      "auxiliary_loss_mlp": 0.01083307,
      "balance_loss_clip": 1.07820928,
      "balance_loss_mlp": 1.05107248,
      "epoch": 0.03390951450473471,
      "flos": 15669298609920.0,
      "grad_norm": 2.243490051481645,
      "language_loss": 0.94325721,
      "learning_rate": 3.99983978971508e-06,
      "loss": 0.96679556,
      "num_input_tokens_seen": 11917815,
      "step": 564,
      "time_per_iteration": 2.4681451320648193
    },
    {
      "auxiliary_loss_clip": 0.01269156,
      "auxiliary_loss_mlp": 0.0106997,
      "balance_loss_clip": 1.07518101,
      "balance_loss_mlp": 1.03682995,
      "epoch": 0.03396963775740267,
      "flos": 22674787850880.0,
      "grad_norm": 2.478504994341654,
      "language_loss": 0.9405092,
      "learning_rate": 3.999834822316635e-06,
      "loss": 0.96390051,
      "num_input_tokens_seen": 11936305,
      "step": 565,
      "time_per_iteration": 2.4794232845306396
    },
    {
      "auxiliary_loss_clip": 0.01142712,
      "auxiliary_loss_mlp": 0.01039409,
      "balance_loss_clip": 1.05036592,
      "balance_loss_mlp": 1.03192317,
      "epoch": 0.034029761010070644,
      "flos": 64392683063040.0,
      "grad_norm": 1.0848581903627883,
      "language_loss": 0.54903507,
      "learning_rate": 3.9998297790842535e-06,
      "loss": 0.57085627,
      "num_input_tokens_seen": 11998940,
      "step": 566,
      "time_per_iteration": 4.444222450256348
    },
    {
      "auxiliary_loss_clip": 0.01272133,
      "auxiliary_loss_mlp": 0.01074469,
      "balance_loss_clip": 1.0785104,
      "balance_loss_mlp": 1.04037571,
      "epoch": 0.034089884262738616,
      "flos": 25004204755200.0,
      "grad_norm": 3.000057000328792,
      "language_loss": 0.76652658,
      "learning_rate": 3.999824660018126e-06,
      "loss": 0.78999257,
      "num_input_tokens_seen": 12018860,
      "step": 567,
      "time_per_iteration": 2.5354883670806885
    },
    {
      "auxiliary_loss_clip": 0.01265446,
      "auxiliary_loss_mlp": 0.01079722,
      "balance_loss_clip": 1.07800078,
      "balance_loss_mlp": 1.04910874,
      "epoch": 0.03415000751540658,
      "flos": 28439096584320.0,
      "grad_norm": 1.8408513697147326,
      "language_loss": 0.80652207,
      "learning_rate": 3.999819465118447e-06,
      "loss": 0.8299737,
      "num_input_tokens_seen": 12039675,
      "step": 568,
      "time_per_iteration": 3.996645212173462
    },
    {
      "auxiliary_loss_clip": 0.01266546,
      "auxiliary_loss_mlp": 0.0108589,
      "balance_loss_clip": 1.07965386,
      "balance_loss_mlp": 1.05365551,
      "epoch": 0.034210130768074554,
      "flos": 21468727866240.0,
      "grad_norm": 1.9195599722514187,
      "language_loss": 0.86389709,
      "learning_rate": 3.999814194385413e-06,
      "loss": 0.88742137,
      "num_input_tokens_seen": 12057680,
      "step": 569,
      "time_per_iteration": 3.916688919067383
    },
    {
      "auxiliary_loss_clip": 0.01265683,
      "auxiliary_loss_mlp": 0.01075379,
      "balance_loss_clip": 1.07660246,
      "balance_loss_mlp": 1.0441227,
      "epoch": 0.03427025402074252,
      "flos": 18697501676160.0,
      "grad_norm": 1.815386908674507,
      "language_loss": 0.95938987,
      "learning_rate": 3.9998088478192255e-06,
      "loss": 0.98280048,
      "num_input_tokens_seen": 12076135,
      "step": 570,
      "time_per_iteration": 2.4799740314483643
    },
    {
      "auxiliary_loss_clip": 0.01267132,
      "auxiliary_loss_mlp": 0.01076246,
      "balance_loss_clip": 1.07352054,
      "balance_loss_mlp": 1.04127038,
      "epoch": 0.03433037727341049,
      "flos": 20849987162880.0,
      "grad_norm": 2.1779943541162785,
      "language_loss": 0.79948843,
      "learning_rate": 3.9998034254200846e-06,
      "loss": 0.82292223,
      "num_input_tokens_seen": 12094785,
      "step": 571,
      "time_per_iteration": 3.8558449745178223
    },
    {
      "auxiliary_loss_clip": 0.01267811,
      "auxiliary_loss_mlp": 0.01083382,
      "balance_loss_clip": 1.07853591,
      "balance_loss_mlp": 1.05014658,
      "epoch": 0.03439050052607846,
      "flos": 25410282986880.0,
      "grad_norm": 4.571174304564672,
      "language_loss": 0.80325431,
      "learning_rate": 3.999797927188199e-06,
      "loss": 0.82676625,
      "num_input_tokens_seen": 12114590,
      "step": 572,
      "time_per_iteration": 2.5146992206573486
    },
    {
      "auxiliary_loss_clip": 0.01275728,
      "auxiliary_loss_mlp": 0.01072275,
      "balance_loss_clip": 1.08139789,
      "balance_loss_mlp": 1.03987396,
      "epoch": 0.03445062377874643,
      "flos": 17640147997440.0,
      "grad_norm": 1.8272520339256504,
      "language_loss": 0.84589422,
      "learning_rate": 3.999792353123774e-06,
      "loss": 0.86937428,
      "num_input_tokens_seen": 12132390,
      "step": 573,
      "time_per_iteration": 2.4722063541412354
    },
    {
      "auxiliary_loss_clip": 0.01268498,
      "auxiliary_loss_mlp": 0.01066585,
      "balance_loss_clip": 1.07571602,
      "balance_loss_mlp": 1.03625786,
      "epoch": 0.0345107470314144,
      "flos": 16764502245120.0,
      "grad_norm": 2.4119711403306447,
      "language_loss": 0.7641983,
      "learning_rate": 3.999786703227023e-06,
      "loss": 0.78754914,
      "num_input_tokens_seen": 12149035,
      "step": 574,
      "time_per_iteration": 2.478041648864746
    },
    {
      "auxiliary_loss_clip": 0.01268233,
      "auxiliary_loss_mlp": 0.01070726,
      "balance_loss_clip": 1.07792163,
      "balance_loss_mlp": 1.03989804,
      "epoch": 0.03457087028408237,
      "flos": 14684448533760.0,
      "grad_norm": 2.5343917115280123,
      "language_loss": 0.83672482,
      "learning_rate": 3.9997809774981606e-06,
      "loss": 0.86011446,
      "num_input_tokens_seen": 12167530,
      "step": 575,
      "time_per_iteration": 2.473597764968872
    },
    {
      "auxiliary_loss_clip": 0.01263558,
      "auxiliary_loss_mlp": 0.01075363,
      "balance_loss_clip": 1.07851744,
      "balance_loss_mlp": 1.04424894,
      "epoch": 0.03463099353675034,
      "flos": 20011293527040.0,
      "grad_norm": 1.9706343163024558,
      "language_loss": 0.84135616,
      "learning_rate": 3.9997751759374025e-06,
      "loss": 0.86474538,
      "num_input_tokens_seen": 12186340,
      "step": 576,
      "time_per_iteration": 2.4883697032928467
    },
    {
      "auxiliary_loss_clip": 0.01270893,
      "auxiliary_loss_mlp": 0.01071544,
      "balance_loss_clip": 1.08502579,
      "balance_loss_mlp": 1.0416944,
      "epoch": 0.03469111678941831,
      "flos": 25301150490240.0,
      "grad_norm": 2.13412835635318,
      "language_loss": 0.86487353,
      "learning_rate": 3.99976929854497e-06,
      "loss": 0.88829786,
      "num_input_tokens_seen": 12204090,
      "step": 577,
      "time_per_iteration": 2.5000736713409424
    },
    {
      "auxiliary_loss_clip": 0.01265697,
      "auxiliary_loss_mlp": 0.01076331,
      "balance_loss_clip": 1.08083272,
      "balance_loss_mlp": 1.04476428,
      "epoch": 0.034751240042086275,
      "flos": 23259413612160.0,
      "grad_norm": 1.8650738549208734,
      "language_loss": 0.72178662,
      "learning_rate": 3.9997633453210845e-06,
      "loss": 0.74520695,
      "num_input_tokens_seen": 12224850,
      "step": 578,
      "time_per_iteration": 2.5300705432891846
    },
    {
      "auxiliary_loss_clip": 0.01266366,
      "auxiliary_loss_mlp": 0.01082288,
      "balance_loss_clip": 1.07687616,
      "balance_loss_mlp": 1.04945803,
      "epoch": 0.03481136329475425,
      "flos": 23769237300480.0,
      "grad_norm": 2.3731535740735743,
      "language_loss": 0.77602941,
      "learning_rate": 3.999757316265973e-06,
      "loss": 0.79951596,
      "num_input_tokens_seen": 12244935,
      "step": 579,
      "time_per_iteration": 2.4892470836639404
    },
    {
      "auxiliary_loss_clip": 0.01265137,
      "auxiliary_loss_mlp": 0.01080635,
      "balance_loss_clip": 1.07855868,
      "balance_loss_mlp": 1.04816198,
      "epoch": 0.03487148654742222,
      "flos": 20157521794560.0,
      "grad_norm": 1.9065227364371031,
      "language_loss": 0.86615121,
      "learning_rate": 3.999751211379863e-06,
      "loss": 0.88960898,
      "num_input_tokens_seen": 12262140,
      "step": 580,
      "time_per_iteration": 2.496328353881836
    },
    {
      "auxiliary_loss_clip": 0.01270842,
      "auxiliary_loss_mlp": 0.0106775,
      "balance_loss_clip": 1.07863641,
      "balance_loss_mlp": 1.03856754,
      "epoch": 0.034931609800090184,
      "flos": 15669585918720.0,
      "grad_norm": 2.7049339501381997,
      "language_loss": 0.8202002,
      "learning_rate": 3.999745030662987e-06,
      "loss": 0.84358615,
      "num_input_tokens_seen": 12280930,
      "step": 581,
      "time_per_iteration": 2.4669504165649414
    },
    {
      "auxiliary_loss_clip": 0.01267412,
      "auxiliary_loss_mlp": 0.01071462,
      "balance_loss_clip": 1.08008754,
      "balance_loss_mlp": 1.04111099,
      "epoch": 0.034991733052758156,
      "flos": 16362374509440.0,
      "grad_norm": 2.132807967618713,
      "language_loss": 0.77002013,
      "learning_rate": 3.99973877411558e-06,
      "loss": 0.79340887,
      "num_input_tokens_seen": 12299125,
      "step": 582,
      "time_per_iteration": 2.5026347637176514
    },
    {
      "auxiliary_loss_clip": 0.01265817,
      "auxiliary_loss_mlp": 0.01078064,
      "balance_loss_clip": 1.08111489,
      "balance_loss_mlp": 1.04580593,
      "epoch": 0.03505185630542612,
      "flos": 19387309438080.0,
      "grad_norm": 1.9906301773961352,
      "language_loss": 0.87732178,
      "learning_rate": 3.999732441737877e-06,
      "loss": 0.90076059,
      "num_input_tokens_seen": 12316905,
      "step": 583,
      "time_per_iteration": 2.450899600982666
    },
    {
      "auxiliary_loss_clip": 0.01270841,
      "auxiliary_loss_mlp": 0.01087585,
      "balance_loss_clip": 1.07904279,
      "balance_loss_mlp": 1.05628026,
      "epoch": 0.03511197955809409,
      "flos": 21323828401920.0,
      "grad_norm": 2.6634599873608877,
      "language_loss": 0.81370026,
      "learning_rate": 3.99972603353012e-06,
      "loss": 0.83728456,
      "num_input_tokens_seen": 12335070,
      "step": 584,
      "time_per_iteration": 2.5133345127105713
    },
    {
      "auxiliary_loss_clip": 0.01266393,
      "auxiliary_loss_mlp": 0.01070571,
      "balance_loss_clip": 1.0767653,
      "balance_loss_mlp": 1.0396955,
      "epoch": 0.035172102810762065,
      "flos": 14136595320960.0,
      "grad_norm": 3.401273540972964,
      "language_loss": 0.92766976,
      "learning_rate": 3.999719549492551e-06,
      "loss": 0.95103943,
      "num_input_tokens_seen": 12350315,
      "step": 585,
      "time_per_iteration": 2.4437124729156494
    },
    {
      "auxiliary_loss_clip": 0.01267492,
      "auxiliary_loss_mlp": 0.01075471,
      "balance_loss_clip": 1.07925153,
      "balance_loss_mlp": 1.04435754,
      "epoch": 0.03523222606343003,
      "flos": 20296890564480.0,
      "grad_norm": 3.549505062469986,
      "language_loss": 0.87795913,
      "learning_rate": 3.9997129896254165e-06,
      "loss": 0.90138876,
      "num_input_tokens_seen": 12366030,
      "step": 586,
      "time_per_iteration": 2.461998701095581
    },
    {
      "auxiliary_loss_clip": 0.01270725,
      "auxiliary_loss_mlp": 0.01079455,
      "balance_loss_clip": 1.08072937,
      "balance_loss_mlp": 1.04862714,
      "epoch": 0.035292349316098,
      "flos": 20375822701440.0,
      "grad_norm": 1.8672240406682776,
      "language_loss": 0.76943094,
      "learning_rate": 3.999706353928965e-06,
      "loss": 0.79293275,
      "num_input_tokens_seen": 12384895,
      "step": 587,
      "time_per_iteration": 2.461668014526367
    },
    {
      "auxiliary_loss_clip": 0.01272382,
      "auxiliary_loss_mlp": 0.01067356,
      "balance_loss_clip": 1.07962024,
      "balance_loss_mlp": 1.03502691,
      "epoch": 0.03535247256876597,
      "flos": 21468871520640.0,
      "grad_norm": 6.087030229719717,
      "language_loss": 0.78839141,
      "learning_rate": 3.999699642403449e-06,
      "loss": 0.8117888,
      "num_input_tokens_seen": 12404980,
      "step": 588,
      "time_per_iteration": 2.504309892654419
    },
    {
      "auxiliary_loss_clip": 0.01269633,
      "auxiliary_loss_mlp": 0.01078179,
      "balance_loss_clip": 1.07796383,
      "balance_loss_mlp": 1.04418015,
      "epoch": 0.03541259582143394,
      "flos": 23623044946560.0,
      "grad_norm": 2.055504965058517,
      "language_loss": 0.94321144,
      "learning_rate": 3.99969285504912e-06,
      "loss": 0.96668953,
      "num_input_tokens_seen": 12423835,
      "step": 589,
      "time_per_iteration": 2.4861948490142822
    },
    {
      "auxiliary_loss_clip": 0.01271455,
      "auxiliary_loss_mlp": 0.01076565,
      "balance_loss_clip": 1.07956159,
      "balance_loss_mlp": 1.04623866,
      "epoch": 0.03547271907410191,
      "flos": 33726367768320.0,
      "grad_norm": 2.322929290027473,
      "language_loss": 0.83637238,
      "learning_rate": 3.99968599186624e-06,
      "loss": 0.85985255,
      "num_input_tokens_seen": 12443135,
      "step": 590,
      "time_per_iteration": 2.5837550163269043
    },
    {
      "auxiliary_loss_clip": 0.01262852,
      "auxiliary_loss_mlp": 0.01069764,
      "balance_loss_clip": 1.07772732,
      "balance_loss_mlp": 1.04074848,
      "epoch": 0.03553284232676988,
      "flos": 21142695093120.0,
      "grad_norm": 2.880934480499134,
      "language_loss": 0.87028849,
      "learning_rate": 3.999679052855065e-06,
      "loss": 0.89361459,
      "num_input_tokens_seen": 12462895,
      "step": 591,
      "time_per_iteration": 2.471809148788452
    },
    {
      "auxiliary_loss_clip": 0.01266617,
      "auxiliary_loss_mlp": 0.01078025,
      "balance_loss_clip": 1.07488751,
      "balance_loss_mlp": 1.04536152,
      "epoch": 0.03559296557943785,
      "flos": 20046593617920.0,
      "grad_norm": 2.324912267432322,
      "language_loss": 0.82935798,
      "learning_rate": 3.999672038015861e-06,
      "loss": 0.85280442,
      "num_input_tokens_seen": 12481515,
      "step": 592,
      "time_per_iteration": 2.475825548171997
    },
    {
      "auxiliary_loss_clip": 0.01163403,
      "auxiliary_loss_mlp": 0.01042793,
      "balance_loss_clip": 1.06023741,
      "balance_loss_mlp": 1.0369513,
      "epoch": 0.035653088832105814,
      "flos": 60334597244160.0,
      "grad_norm": 0.8946623584723222,
      "language_loss": 0.59819162,
      "learning_rate": 3.999664947348893e-06,
      "loss": 0.62025356,
      "num_input_tokens_seen": 12548220,
      "step": 593,
      "time_per_iteration": 3.1318681240081787
    },
    {
      "auxiliary_loss_clip": 0.01267086,
      "auxiliary_loss_mlp": 0.01071923,
      "balance_loss_clip": 1.08094192,
      "balance_loss_mlp": 1.03983164,
      "epoch": 0.035713212084773786,
      "flos": 20113135562880.0,
      "grad_norm": 2.707095985408309,
      "language_loss": 0.86923981,
      "learning_rate": 3.999657780854429e-06,
      "loss": 0.89262986,
      "num_input_tokens_seen": 12566105,
      "step": 594,
      "time_per_iteration": 2.480548620223999
    },
    {
      "auxiliary_loss_clip": 0.01264633,
      "auxiliary_loss_mlp": 0.01071495,
      "balance_loss_clip": 1.0750066,
      "balance_loss_mlp": 1.04054785,
      "epoch": 0.03577333533744176,
      "flos": 26285785084800.0,
      "grad_norm": 2.022232648847156,
      "language_loss": 0.83355463,
      "learning_rate": 3.999650538532742e-06,
      "loss": 0.85691583,
      "num_input_tokens_seen": 12586680,
      "step": 595,
      "time_per_iteration": 2.5608370304107666
    },
    {
      "auxiliary_loss_clip": 0.01264116,
      "auxiliary_loss_mlp": 0.01077248,
      "balance_loss_clip": 1.07853663,
      "balance_loss_mlp": 1.04570556,
      "epoch": 0.035833458590109724,
      "flos": 10889732211840.0,
      "grad_norm": 3.975459646866357,
      "language_loss": 0.96219558,
      "learning_rate": 3.999643220384106e-06,
      "loss": 0.98560917,
      "num_input_tokens_seen": 12601605,
      "step": 596,
      "time_per_iteration": 2.4495489597320557
    },
    {
      "auxiliary_loss_clip": 0.01266936,
      "auxiliary_loss_mlp": 0.01076897,
      "balance_loss_clip": 1.07881594,
      "balance_loss_mlp": 1.04685605,
      "epoch": 0.035893581842777696,
      "flos": 22090198003200.0,
      "grad_norm": 2.546966433500354,
      "language_loss": 0.8273415,
      "learning_rate": 3.999635826408799e-06,
      "loss": 0.85077983,
      "num_input_tokens_seen": 12620365,
      "step": 597,
      "time_per_iteration": 2.5030415058135986
    },
    {
      "auxiliary_loss_clip": 0.01263799,
      "auxiliary_loss_mlp": 0.01069854,
      "balance_loss_clip": 1.07954168,
      "balance_loss_mlp": 1.03885996,
      "epoch": 0.03595370509544566,
      "flos": 23038347358080.0,
      "grad_norm": 1.778157566472502,
      "language_loss": 0.81453472,
      "learning_rate": 3.999628356607101e-06,
      "loss": 0.83787131,
      "num_input_tokens_seen": 12641140,
      "step": 598,
      "time_per_iteration": 2.554788589477539
    },
    {
      "auxiliary_loss_clip": 0.01259513,
      "auxiliary_loss_mlp": 0.01071988,
      "balance_loss_clip": 1.07961512,
      "balance_loss_mlp": 1.03968191,
      "epoch": 0.03601382834811363,
      "flos": 20777734955520.0,
      "grad_norm": 1.924807245635329,
      "language_loss": 0.81037521,
      "learning_rate": 3.999620810979295e-06,
      "loss": 0.83369023,
      "num_input_tokens_seen": 12661080,
      "step": 599,
      "time_per_iteration": 2.5619709491729736
    },
    {
      "auxiliary_loss_clip": 0.01267438,
      "auxiliary_loss_mlp": 0.01081938,
      "balance_loss_clip": 1.07717586,
      "balance_loss_mlp": 1.05218291,
      "epoch": 0.036073951600781605,
      "flos": 23951627585280.0,
      "grad_norm": 2.4656747586614496,
      "language_loss": 0.85833454,
      "learning_rate": 3.999613189525668e-06,
      "loss": 0.88182831,
      "num_input_tokens_seen": 12678270,
      "step": 600,
      "time_per_iteration": 2.550051689147949
    },
    {
      "auxiliary_loss_clip": 0.01259526,
      "auxiliary_loss_mlp": 0.01085252,
      "balance_loss_clip": 1.07389736,
      "balance_loss_mlp": 1.05494857,
      "epoch": 0.03613407485344957,
      "flos": 18912283050240.0,
      "grad_norm": 2.0754472473818395,
      "language_loss": 0.82199258,
      "learning_rate": 3.999605492246508e-06,
      "loss": 0.84544039,
      "num_input_tokens_seen": 12697295,
      "step": 601,
      "time_per_iteration": 2.4767348766326904
    },
    {
      "auxiliary_loss_clip": 0.01256591,
      "auxiliary_loss_mlp": 0.01060178,
      "balance_loss_clip": 1.07390904,
      "balance_loss_mlp": 1.02966058,
      "epoch": 0.03619419810611754,
      "flos": 23038526926080.0,
      "grad_norm": 2.3440356128618745,
      "language_loss": 0.7542612,
      "learning_rate": 3.999597719142107e-06,
      "loss": 0.77742893,
      "num_input_tokens_seen": 12716165,
      "step": 602,
      "time_per_iteration": 2.471942663192749
    },
    {
      "auxiliary_loss_clip": 0.01255861,
      "auxiliary_loss_mlp": 0.01064639,
      "balance_loss_clip": 1.07274139,
      "balance_loss_mlp": 1.03354931,
      "epoch": 0.03625432135878551,
      "flos": 29457774293760.0,
      "grad_norm": 1.9754909640674911,
      "language_loss": 0.79593658,
      "learning_rate": 3.999589870212761e-06,
      "loss": 0.81914151,
      "num_input_tokens_seen": 12735475,
      "step": 603,
      "time_per_iteration": 2.553173780441284
    },
    {
      "auxiliary_loss_clip": 0.0126047,
      "auxiliary_loss_mlp": 0.01066756,
      "balance_loss_clip": 1.07811737,
      "balance_loss_mlp": 1.03723943,
      "epoch": 0.03631444461145348,
      "flos": 23508525409920.0,
      "grad_norm": 3.384710501025295,
      "language_loss": 0.86839509,
      "learning_rate": 3.9995819454587664e-06,
      "loss": 0.89166737,
      "num_input_tokens_seen": 12754540,
      "step": 604,
      "time_per_iteration": 2.480534791946411
    },
    {
      "auxiliary_loss_clip": 0.01263579,
      "auxiliary_loss_mlp": 0.01071492,
      "balance_loss_clip": 1.07967842,
      "balance_loss_mlp": 1.03890014,
      "epoch": 0.03637456786412145,
      "flos": 16618130323200.0,
      "grad_norm": 2.1549927094794272,
      "language_loss": 0.80670166,
      "learning_rate": 3.999573944880424e-06,
      "loss": 0.83005238,
      "num_input_tokens_seen": 12773050,
      "step": 605,
      "time_per_iteration": 3.824784517288208
    },
    {
      "auxiliary_loss_clip": 0.01261027,
      "auxiliary_loss_mlp": 0.01069907,
      "balance_loss_clip": 1.07714224,
      "balance_loss_mlp": 1.04121375,
      "epoch": 0.03643469111678942,
      "flos": 15851832549120.0,
      "grad_norm": 2.2164480756607836,
      "language_loss": 0.85706699,
      "learning_rate": 3.9995658684780375e-06,
      "loss": 0.88037634,
      "num_input_tokens_seen": 12791240,
      "step": 606,
      "time_per_iteration": 2.45992112159729
    },
    {
      "auxiliary_loss_clip": 0.01265168,
      "auxiliary_loss_mlp": 0.0107834,
      "balance_loss_clip": 1.07771611,
      "balance_loss_mlp": 1.04770291,
      "epoch": 0.03649481436945739,
      "flos": 23620387340160.0,
      "grad_norm": 2.2349045283287383,
      "language_loss": 0.82342446,
      "learning_rate": 3.999557716251912e-06,
      "loss": 0.84685957,
      "num_input_tokens_seen": 12812245,
      "step": 607,
      "time_per_iteration": 3.967804431915283
    },
    {
      "auxiliary_loss_clip": 0.01258655,
      "auxiliary_loss_mlp": 0.01071174,
      "balance_loss_clip": 1.07661784,
      "balance_loss_mlp": 1.04194403,
      "epoch": 0.036554937622125354,
      "flos": 21755581879680.0,
      "grad_norm": 2.1604117318991154,
      "language_loss": 0.8344844,
      "learning_rate": 3.999549488202358e-06,
      "loss": 0.85778272,
      "num_input_tokens_seen": 12831085,
      "step": 608,
      "time_per_iteration": 3.9322891235351562
    },
    {
      "auxiliary_loss_clip": 0.01263897,
      "auxiliary_loss_mlp": 0.01069309,
      "balance_loss_clip": 1.07878816,
      "balance_loss_mlp": 1.03674114,
      "epoch": 0.036615060874793326,
      "flos": 17819772935040.0,
      "grad_norm": 2.1656117325242477,
      "language_loss": 0.81954956,
      "learning_rate": 3.999541184329688e-06,
      "loss": 0.84288162,
      "num_input_tokens_seen": 12849115,
      "step": 609,
      "time_per_iteration": 2.4642422199249268
    },
    {
      "auxiliary_loss_clip": 0.01270212,
      "auxiliary_loss_mlp": 0.01089263,
      "balance_loss_clip": 1.08474493,
      "balance_loss_mlp": 1.05958033,
      "epoch": 0.0366751841274613,
      "flos": 26753808320640.0,
      "grad_norm": 1.9050483483729914,
      "language_loss": 0.79137301,
      "learning_rate": 3.999532804634215e-06,
      "loss": 0.81496775,
      "num_input_tokens_seen": 12868005,
      "step": 610,
      "time_per_iteration": 2.523236036300659
    },
    {
      "auxiliary_loss_clip": 0.01266789,
      "auxiliary_loss_mlp": 0.01087561,
      "balance_loss_clip": 1.07946277,
      "balance_loss_mlp": 1.056638,
      "epoch": 0.03673530738012926,
      "flos": 22196960202240.0,
      "grad_norm": 2.058157630182323,
      "language_loss": 0.87253487,
      "learning_rate": 3.9995243491162575e-06,
      "loss": 0.89607835,
      "num_input_tokens_seen": 12886890,
      "step": 611,
      "time_per_iteration": 2.4754512310028076
    },
    {
      "auxiliary_loss_clip": 0.01262467,
      "auxiliary_loss_mlp": 0.01092681,
      "balance_loss_clip": 1.08016503,
      "balance_loss_mlp": 1.06298554,
      "epoch": 0.036795430632797235,
      "flos": 24681655601280.0,
      "grad_norm": 8.001205117711043,
      "language_loss": 0.72764426,
      "learning_rate": 3.999515817776136e-06,
      "loss": 0.75119567,
      "num_input_tokens_seen": 12906130,
      "step": 612,
      "time_per_iteration": 3.910191535949707
    },
    {
      "auxiliary_loss_clip": 0.01262253,
      "auxiliary_loss_mlp": 0.01076147,
      "balance_loss_clip": 1.07695389,
      "balance_loss_mlp": 1.04493773,
      "epoch": 0.0368555538854652,
      "flos": 17748921358080.0,
      "grad_norm": 3.0170571918431355,
      "language_loss": 0.79205078,
      "learning_rate": 3.999507210614175e-06,
      "loss": 0.81543475,
      "num_input_tokens_seen": 12925260,
      "step": 613,
      "time_per_iteration": 2.4712016582489014
    },
    {
      "auxiliary_loss_clip": 0.01258812,
      "auxiliary_loss_mlp": 0.01080852,
      "balance_loss_clip": 1.07804215,
      "balance_loss_mlp": 1.05052531,
      "epoch": 0.03691567713813317,
      "flos": 20594554571520.0,
      "grad_norm": 2.0334073345437895,
      "language_loss": 0.93529367,
      "learning_rate": 3.9994985276307e-06,
      "loss": 0.95869035,
      "num_input_tokens_seen": 12944590,
      "step": 614,
      "time_per_iteration": 2.481989860534668
    },
    {
      "auxiliary_loss_clip": 0.01265953,
      "auxiliary_loss_mlp": 0.01078354,
      "balance_loss_clip": 1.08072233,
      "balance_loss_mlp": 1.04585779,
      "epoch": 0.036975800390801145,
      "flos": 33650380546560.0,
      "grad_norm": 4.282530987571834,
      "language_loss": 0.72850531,
      "learning_rate": 3.999489768826041e-06,
      "loss": 0.75194836,
      "num_input_tokens_seen": 12964785,
      "step": 615,
      "time_per_iteration": 2.590994119644165
    },
    {
      "auxiliary_loss_clip": 0.01261948,
      "auxiliary_loss_mlp": 0.01073271,
      "balance_loss_clip": 1.07618034,
      "balance_loss_mlp": 1.04387379,
      "epoch": 0.03703592364346911,
      "flos": 28293694329600.0,
      "grad_norm": 2.752167712145806,
      "language_loss": 0.81885445,
      "learning_rate": 3.999480934200528e-06,
      "loss": 0.84220666,
      "num_input_tokens_seen": 12986705,
      "step": 616,
      "time_per_iteration": 2.5316333770751953
    },
    {
      "auxiliary_loss_clip": 0.01261862,
      "auxiliary_loss_mlp": 0.01069787,
      "balance_loss_clip": 1.07772827,
      "balance_loss_mlp": 1.04062891,
      "epoch": 0.03709604689613708,
      "flos": 31504215853440.0,
      "grad_norm": 2.105602933460105,
      "language_loss": 0.68274242,
      "learning_rate": 3.999472023754499e-06,
      "loss": 0.70605892,
      "num_input_tokens_seen": 13010560,
      "step": 617,
      "time_per_iteration": 2.572199821472168
    },
    {
      "auxiliary_loss_clip": 0.01266493,
      "auxiliary_loss_mlp": 0.01070664,
      "balance_loss_clip": 1.08158422,
      "balance_loss_mlp": 1.0387398,
      "epoch": 0.03715617014880505,
      "flos": 19609381272960.0,
      "grad_norm": 2.2786636072509285,
      "language_loss": 0.80186707,
      "learning_rate": 3.99946303748829e-06,
      "loss": 0.82523859,
      "num_input_tokens_seen": 13028935,
      "step": 618,
      "time_per_iteration": 2.4802162647247314
    },
    {
      "auxiliary_loss_clip": 0.01266185,
      "auxiliary_loss_mlp": 0.0107521,
      "balance_loss_clip": 1.0783639,
      "balance_loss_mlp": 1.04328597,
      "epoch": 0.03721629340147302,
      "flos": 15924192497280.0,
      "grad_norm": 3.648389328328535,
      "language_loss": 0.91155064,
      "learning_rate": 3.999453975402242e-06,
      "loss": 0.9349646,
      "num_input_tokens_seen": 13046000,
      "step": 619,
      "time_per_iteration": 2.5026469230651855
    },
    {
      "auxiliary_loss_clip": 0.01261508,
      "auxiliary_loss_mlp": 0.01079631,
      "balance_loss_clip": 1.07998168,
      "balance_loss_mlp": 1.04937613,
      "epoch": 0.03727641665414099,
      "flos": 21104090951040.0,
      "grad_norm": 2.378477328271392,
      "language_loss": 0.94196492,
      "learning_rate": 3.9994448374967e-06,
      "loss": 0.96537626,
      "num_input_tokens_seen": 13062995,
      "step": 620,
      "time_per_iteration": 2.4629948139190674
    },
    {
      "auxiliary_loss_clip": 0.01259356,
      "auxiliary_loss_mlp": 0.01079201,
      "balance_loss_clip": 1.07666337,
      "balance_loss_mlp": 1.04627514,
      "epoch": 0.037336539906808956,
      "flos": 24131683486080.0,
      "grad_norm": 2.275627158655139,
      "language_loss": 0.77419603,
      "learning_rate": 3.999435623772008e-06,
      "loss": 0.79758161,
      "num_input_tokens_seen": 13084120,
      "step": 621,
      "time_per_iteration": 2.5375559329986572
    },
    {
      "auxiliary_loss_clip": 0.01258357,
      "auxiliary_loss_mlp": 0.01066698,
      "balance_loss_clip": 1.07940221,
      "balance_loss_mlp": 1.03498816,
      "epoch": 0.03739666315947693,
      "flos": 22346384780160.0,
      "grad_norm": 3.7693986113650273,
      "language_loss": 0.86629665,
      "learning_rate": 3.999426334228518e-06,
      "loss": 0.88954723,
      "num_input_tokens_seen": 13100035,
      "step": 622,
      "time_per_iteration": 2.475057363510132
    },
    {
      "auxiliary_loss_clip": 0.01260206,
      "auxiliary_loss_mlp": 0.01068153,
      "balance_loss_clip": 1.07839918,
      "balance_loss_mlp": 1.03782654,
      "epoch": 0.0374567864121449,
      "flos": 20449511452800.0,
      "grad_norm": 2.2395120542559743,
      "language_loss": 0.90277684,
      "learning_rate": 3.999416968866581e-06,
      "loss": 0.92606038,
      "num_input_tokens_seen": 13118070,
      "step": 623,
      "time_per_iteration": 2.4850597381591797
    },
    {
      "auxiliary_loss_clip": 0.01264985,
      "auxiliary_loss_mlp": 0.01081644,
      "balance_loss_clip": 1.08187222,
      "balance_loss_mlp": 1.0506258,
      "epoch": 0.037516909664812866,
      "flos": 19208043636480.0,
      "grad_norm": 1.9167099724974273,
      "language_loss": 0.8409313,
      "learning_rate": 3.999407527686551e-06,
      "loss": 0.86439753,
      "num_input_tokens_seen": 13136355,
      "step": 624,
      "time_per_iteration": 2.4656646251678467
    },
    {
      "auxiliary_loss_clip": 0.01262814,
      "auxiliary_loss_mlp": 0.01071085,
      "balance_loss_clip": 1.07678711,
      "balance_loss_mlp": 1.0400908,
      "epoch": 0.03757703291748084,
      "flos": 35005218664320.0,
      "grad_norm": 6.194727931453058,
      "language_loss": 0.66913795,
      "learning_rate": 3.999398010688788e-06,
      "loss": 0.69247699,
      "num_input_tokens_seen": 13155435,
      "step": 625,
      "time_per_iteration": 2.602140426635742
    },
    {
      "auxiliary_loss_clip": 0.01257444,
      "auxiliary_loss_mlp": 0.01070195,
      "balance_loss_clip": 1.0778625,
      "balance_loss_mlp": 1.03741217,
      "epoch": 0.0376371561701488,
      "flos": 25483899911040.0,
      "grad_norm": 3.957133360189669,
      "language_loss": 0.76833189,
      "learning_rate": 3.999388417873652e-06,
      "loss": 0.79160833,
      "num_input_tokens_seen": 13174295,
      "step": 626,
      "time_per_iteration": 2.510777473449707
    },
    {
      "auxiliary_loss_clip": 0.01261254,
      "auxiliary_loss_mlp": 0.01076318,
      "balance_loss_clip": 1.07908964,
      "balance_loss_mlp": 1.04501367,
      "epoch": 0.037697279422816775,
      "flos": 18185630912640.0,
      "grad_norm": 1.9189899510225719,
      "language_loss": 0.8159169,
      "learning_rate": 3.999378749241506e-06,
      "loss": 0.83929259,
      "num_input_tokens_seen": 13192500,
      "step": 627,
      "time_per_iteration": 2.4697675704956055
    },
    {
      "auxiliary_loss_clip": 0.01265322,
      "auxiliary_loss_mlp": 0.01078306,
      "balance_loss_clip": 1.08127236,
      "balance_loss_mlp": 1.04745483,
      "epoch": 0.03775740267548475,
      "flos": 24644272521600.0,
      "grad_norm": 2.057187205426108,
      "language_loss": 0.88823539,
      "learning_rate": 3.999369004792719e-06,
      "loss": 0.91167164,
      "num_input_tokens_seen": 13213470,
      "step": 628,
      "time_per_iteration": 2.5090696811676025
    },
    {
      "auxiliary_loss_clip": 0.01259195,
      "auxiliary_loss_mlp": 0.01073796,
      "balance_loss_clip": 1.07494068,
      "balance_loss_mlp": 1.04368377,
      "epoch": 0.03781752592815271,
      "flos": 21288205088640.0,
      "grad_norm": 2.2501177098618843,
      "language_loss": 0.79715705,
      "learning_rate": 3.999359184527658e-06,
      "loss": 0.82048696,
      "num_input_tokens_seen": 13232365,
      "step": 629,
      "time_per_iteration": 2.4849469661712646
    },
    {
      "auxiliary_loss_clip": 0.01259917,
      "auxiliary_loss_mlp": 0.01066528,
      "balance_loss_clip": 1.07699418,
      "balance_loss_mlp": 1.03695226,
      "epoch": 0.037877649180820684,
      "flos": 22089623385600.0,
      "grad_norm": 1.7642934692815957,
      "language_loss": 0.76698381,
      "learning_rate": 3.999349288446696e-06,
      "loss": 0.79024827,
      "num_input_tokens_seen": 13251920,
      "step": 630,
      "time_per_iteration": 2.4914400577545166
    },
    {
      "auxiliary_loss_clip": 0.01265778,
      "auxiliary_loss_mlp": 0.0107072,
      "balance_loss_clip": 1.07917047,
      "balance_loss_mlp": 1.0404408,
      "epoch": 0.03793777243348865,
      "flos": 14501339976960.0,
      "grad_norm": 2.7240984363816225,
      "language_loss": 0.91351134,
      "learning_rate": 3.99933931655021e-06,
      "loss": 0.9368763,
      "num_input_tokens_seen": 13267440,
      "step": 631,
      "time_per_iteration": 2.4541170597076416
    },
    {
      "auxiliary_loss_clip": 0.01255653,
      "auxiliary_loss_mlp": 0.01083744,
      "balance_loss_clip": 1.0763135,
      "balance_loss_mlp": 1.05060351,
      "epoch": 0.03799789568615662,
      "flos": 21908418249600.0,
      "grad_norm": 1.6180283359924887,
      "language_loss": 0.92133188,
      "learning_rate": 3.999329268838575e-06,
      "loss": 0.94472587,
      "num_input_tokens_seen": 13287850,
      "step": 632,
      "time_per_iteration": 2.5121684074401855
    },
    {
      "auxiliary_loss_clip": 0.01259781,
      "auxiliary_loss_mlp": 0.01062413,
      "balance_loss_clip": 1.07871342,
      "balance_loss_mlp": 1.0330162,
      "epoch": 0.03805801893882459,
      "flos": 24827021942400.0,
      "grad_norm": 2.0148989286941728,
      "language_loss": 0.83056724,
      "learning_rate": 3.999319145312175e-06,
      "loss": 0.85378921,
      "num_input_tokens_seen": 13307760,
      "step": 633,
      "time_per_iteration": 2.5197947025299072
    },
    {
      "auxiliary_loss_clip": 0.01259317,
      "auxiliary_loss_mlp": 0.01067856,
      "balance_loss_clip": 1.07732654,
      "balance_loss_mlp": 1.03760028,
      "epoch": 0.03811814219149256,
      "flos": 30482952364800.0,
      "grad_norm": 1.6227433678314849,
      "language_loss": 0.69665444,
      "learning_rate": 3.999308945971392e-06,
      "loss": 0.71992618,
      "num_input_tokens_seen": 13331230,
      "step": 634,
      "time_per_iteration": 2.55253267288208
    },
    {
      "auxiliary_loss_clip": 0.01192886,
      "auxiliary_loss_mlp": 0.0103806,
      "balance_loss_clip": 1.09763944,
      "balance_loss_mlp": 1.03066874,
      "epoch": 0.03817826544416053,
      "flos": 66992577379200.0,
      "grad_norm": 0.8982167014853317,
      "language_loss": 0.61641932,
      "learning_rate": 3.999298670816614e-06,
      "loss": 0.63872874,
      "num_input_tokens_seen": 13394760,
      "step": 635,
      "time_per_iteration": 3.146446466445923
    },
    {
      "auxiliary_loss_clip": 0.01254231,
      "auxiliary_loss_mlp": 0.01068562,
      "balance_loss_clip": 1.07546675,
      "balance_loss_mlp": 1.0386641,
      "epoch": 0.038238388696828496,
      "flos": 20485350247680.0,
      "grad_norm": 2.340059841025445,
      "language_loss": 0.83915699,
      "learning_rate": 3.9992883198482294e-06,
      "loss": 0.86238492,
      "num_input_tokens_seen": 13412775,
      "step": 636,
      "time_per_iteration": 2.476250648498535
    },
    {
      "auxiliary_loss_clip": 0.01259703,
      "auxiliary_loss_mlp": 0.0109219,
      "balance_loss_clip": 1.07833409,
      "balance_loss_mlp": 1.06164861,
      "epoch": 0.03829851194949647,
      "flos": 17965893461760.0,
      "grad_norm": 3.1027326803037427,
      "language_loss": 0.79340589,
      "learning_rate": 3.999277893066632e-06,
      "loss": 0.81692487,
      "num_input_tokens_seen": 13427835,
      "step": 637,
      "time_per_iteration": 2.456866979598999
    },
    {
      "auxiliary_loss_clip": 0.01256788,
      "auxiliary_loss_mlp": 0.01081746,
      "balance_loss_clip": 1.07472932,
      "balance_loss_mlp": 1.04982221,
      "epoch": 0.03835863520216444,
      "flos": 22456522857600.0,
      "grad_norm": 2.631098707569537,
      "language_loss": 0.84100312,
      "learning_rate": 3.999267390472215e-06,
      "loss": 0.86438853,
      "num_input_tokens_seen": 13447295,
      "step": 638,
      "time_per_iteration": 2.4854836463928223
    },
    {
      "auxiliary_loss_clip": 0.01264593,
      "auxiliary_loss_mlp": 0.01066857,
      "balance_loss_clip": 1.07729816,
      "balance_loss_mlp": 1.0358386,
      "epoch": 0.038418758454832405,
      "flos": 22164425458560.0,
      "grad_norm": 2.2610814988316514,
      "language_loss": 0.70198774,
      "learning_rate": 3.999256812065381e-06,
      "loss": 0.72530222,
      "num_input_tokens_seen": 13468455,
      "step": 639,
      "time_per_iteration": 2.5341312885284424
    },
    {
      "auxiliary_loss_clip": 0.01261178,
      "auxiliary_loss_mlp": 0.01083601,
      "balance_loss_clip": 1.07824373,
      "balance_loss_mlp": 1.05117583,
      "epoch": 0.03847888170750038,
      "flos": 22747435107840.0,
      "grad_norm": 3.161721484167637,
      "language_loss": 0.84965062,
      "learning_rate": 3.999246157846526e-06,
      "loss": 0.87309849,
      "num_input_tokens_seen": 13489085,
      "step": 640,
      "time_per_iteration": 2.479825735092163
    },
    {
      "auxiliary_loss_clip": 0.01263059,
      "auxiliary_loss_mlp": 0.01084914,
      "balance_loss_clip": 1.07908535,
      "balance_loss_mlp": 1.05244112,
      "epoch": 0.03853900496016834,
      "flos": 22711201263360.0,
      "grad_norm": 2.3790389278669632,
      "language_loss": 0.82238561,
      "learning_rate": 3.9992354278160574e-06,
      "loss": 0.84586537,
      "num_input_tokens_seen": 13509120,
      "step": 641,
      "time_per_iteration": 2.5051443576812744
    },
    {
      "auxiliary_loss_clip": 0.01156679,
      "auxiliary_loss_mlp": 0.01008662,
      "balance_loss_clip": 1.06533051,
      "balance_loss_mlp": 1.00179517,
      "epoch": 0.038599128212836314,
      "flos": 70399136355840.0,
      "grad_norm": 0.9278082749220062,
      "language_loss": 0.65519875,
      "learning_rate": 3.999224621974381e-06,
      "loss": 0.67685217,
      "num_input_tokens_seen": 13562005,
      "step": 642,
      "time_per_iteration": 3.0849153995513916
    },
    {
      "auxiliary_loss_clip": 0.01255475,
      "auxiliary_loss_mlp": 0.0106688,
      "balance_loss_clip": 1.07520175,
      "balance_loss_mlp": 1.03738749,
      "epoch": 0.03865925146550429,
      "flos": 23295144666240.0,
      "grad_norm": 2.0768647898304433,
      "language_loss": 0.79402781,
      "learning_rate": 3.999213740321906e-06,
      "loss": 0.81725138,
      "num_input_tokens_seen": 13582185,
      "step": 643,
      "time_per_iteration": 2.5406243801116943
    },
    {
      "auxiliary_loss_clip": 0.01255064,
      "auxiliary_loss_mlp": 0.01079353,
      "balance_loss_clip": 1.07431197,
      "balance_loss_mlp": 1.04965806,
      "epoch": 0.03871937471817225,
      "flos": 21430446946560.0,
      "grad_norm": 1.8254828721768888,
      "language_loss": 0.83234519,
      "learning_rate": 3.999202782859046e-06,
      "loss": 0.85568935,
      "num_input_tokens_seen": 13599555,
      "step": 644,
      "time_per_iteration": 2.5066418647766113
    },
    {
      "auxiliary_loss_clip": 0.01257015,
      "auxiliary_loss_mlp": 0.01072592,
      "balance_loss_clip": 1.0748713,
      "balance_loss_mlp": 1.04054856,
      "epoch": 0.038779497970840224,
      "flos": 34277309550720.0,
      "grad_norm": 2.146945430305243,
      "language_loss": 0.82316136,
      "learning_rate": 3.9991917495862165e-06,
      "loss": 0.84645748,
      "num_input_tokens_seen": 13621160,
      "step": 645,
      "time_per_iteration": 3.9618613719940186
    },
    {
      "auxiliary_loss_clip": 0.01260651,
      "auxiliary_loss_mlp": 0.01071265,
      "balance_loss_clip": 1.07684183,
      "balance_loss_mlp": 1.04043758,
      "epoch": 0.03883962122350819,
      "flos": 22748189293440.0,
      "grad_norm": 2.6479130491560428,
      "language_loss": 0.82056963,
      "learning_rate": 3.9991806405038345e-06,
      "loss": 0.84388882,
      "num_input_tokens_seen": 13641915,
      "step": 646,
      "time_per_iteration": 5.284180641174316
    },
    {
      "auxiliary_loss_clip": 0.01259478,
      "auxiliary_loss_mlp": 0.01083434,
      "balance_loss_clip": 1.0797646,
      "balance_loss_mlp": 1.05196238,
      "epoch": 0.03889974447617616,
      "flos": 21945837242880.0,
      "grad_norm": 3.0451762793030377,
      "language_loss": 0.81926131,
      "learning_rate": 3.999169455612323e-06,
      "loss": 0.84269047,
      "num_input_tokens_seen": 13661410,
      "step": 647,
      "time_per_iteration": 2.4871902465820312
    },
    {
      "auxiliary_loss_clip": 0.01258569,
      "auxiliary_loss_mlp": 0.01068944,
      "balance_loss_clip": 1.07886529,
      "balance_loss_mlp": 1.03952289,
      "epoch": 0.03895986772884413,
      "flos": 31504826384640.0,
      "grad_norm": 2.403922103412966,
      "language_loss": 0.84255946,
      "learning_rate": 3.999158194912106e-06,
      "loss": 0.86583459,
      "num_input_tokens_seen": 13681705,
      "step": 648,
      "time_per_iteration": 2.5516202449798584
    },
    {
      "auxiliary_loss_clip": 0.01256716,
      "auxiliary_loss_mlp": 0.01077625,
      "balance_loss_clip": 1.07713509,
      "balance_loss_mlp": 1.04643965,
      "epoch": 0.0390199909815121,
      "flos": 19901011795200.0,
      "grad_norm": 2.8062318683260097,
      "language_loss": 0.84226859,
      "learning_rate": 3.9991468584036086e-06,
      "loss": 0.86561203,
      "num_input_tokens_seen": 13700400,
      "step": 649,
      "time_per_iteration": 2.4766805171966553
    },
    {
      "auxiliary_loss_clip": 0.01256804,
      "auxiliary_loss_mlp": 0.01072138,
      "balance_loss_clip": 1.07573342,
      "balance_loss_mlp": 1.04047608,
      "epoch": 0.03908011423418007,
      "flos": 21612478095360.0,
      "grad_norm": 1.8952921502191973,
      "language_loss": 0.80048692,
      "learning_rate": 3.999135446087263e-06,
      "loss": 0.82377625,
      "num_input_tokens_seen": 13720145,
      "step": 650,
      "time_per_iteration": 2.4821808338165283
    },
    {
      "auxiliary_loss_clip": 0.01252613,
      "auxiliary_loss_mlp": 0.01075616,
      "balance_loss_clip": 1.07297516,
      "balance_loss_mlp": 1.04424036,
      "epoch": 0.039140237486848035,
      "flos": 18661411486080.0,
      "grad_norm": 2.273368604882989,
      "language_loss": 0.78747356,
      "learning_rate": 3.9991239579635e-06,
      "loss": 0.81075591,
      "num_input_tokens_seen": 13737500,
      "step": 651,
      "time_per_iteration": 2.46699857711792
    },
    {
      "auxiliary_loss_clip": 0.0125271,
      "auxiliary_loss_mlp": 0.0108055,
      "balance_loss_clip": 1.07314301,
      "balance_loss_mlp": 1.0471952,
      "epoch": 0.03920036073951601,
      "flos": 18661124177280.0,
      "grad_norm": 2.3704974938140313,
      "language_loss": 0.87563097,
      "learning_rate": 3.999112394032757e-06,
      "loss": 0.89896357,
      "num_input_tokens_seen": 13754750,
      "step": 652,
      "time_per_iteration": 3.8582844734191895
    },
    {
      "auxiliary_loss_clip": 0.01248335,
      "auxiliary_loss_mlp": 0.01069296,
      "balance_loss_clip": 1.07328165,
      "balance_loss_mlp": 1.0398035,
      "epoch": 0.03926048399218398,
      "flos": 31354468053120.0,
      "grad_norm": 2.3604241463018427,
      "language_loss": 0.79046333,
      "learning_rate": 3.999100754295471e-06,
      "loss": 0.81363964,
      "num_input_tokens_seen": 13771990,
      "step": 653,
      "time_per_iteration": 2.526585578918457
    },
    {
      "auxiliary_loss_clip": 0.01262864,
      "auxiliary_loss_mlp": 0.01073018,
      "balance_loss_clip": 1.07725561,
      "balance_loss_mlp": 1.04185724,
      "epoch": 0.039320607244851945,
      "flos": 29603499770880.0,
      "grad_norm": 2.56684786017853,
      "language_loss": 0.85858661,
      "learning_rate": 3.999089038752085e-06,
      "loss": 0.88194543,
      "num_input_tokens_seen": 13792750,
      "step": 654,
      "time_per_iteration": 2.5386204719543457
    },
    {
      "auxiliary_loss_clip": 0.01132717,
      "auxiliary_loss_mlp": 0.01018074,
      "balance_loss_clip": 1.05075288,
      "balance_loss_mlp": 1.01125526,
      "epoch": 0.03938073049751992,
      "flos": 66534609951360.0,
      "grad_norm": 0.7282546574417021,
      "language_loss": 0.49919501,
      "learning_rate": 3.999077247403041e-06,
      "loss": 0.5207029,
      "num_input_tokens_seen": 13858570,
      "step": 655,
      "time_per_iteration": 3.144606590270996
    },
    {
      "auxiliary_loss_clip": 0.01251074,
      "auxiliary_loss_mlp": 0.01071078,
      "balance_loss_clip": 1.07508349,
      "balance_loss_mlp": 1.04141855,
      "epoch": 0.03944085375018788,
      "flos": 23367827836800.0,
      "grad_norm": 2.077461336000576,
      "language_loss": 0.80552369,
      "learning_rate": 3.9990653802487886e-06,
      "loss": 0.82874525,
      "num_input_tokens_seen": 13876335,
      "step": 656,
      "time_per_iteration": 2.5098366737365723
    },
    {
      "auxiliary_loss_clip": 0.0126201,
      "auxiliary_loss_mlp": 0.01089813,
      "balance_loss_clip": 1.07760072,
      "balance_loss_mlp": 1.05393112,
      "epoch": 0.039500977002855854,
      "flos": 18548292579840.0,
      "grad_norm": 2.455662967782196,
      "language_loss": 0.76272631,
      "learning_rate": 3.999053437289776e-06,
      "loss": 0.78624451,
      "num_input_tokens_seen": 13892640,
      "step": 657,
      "time_per_iteration": 2.4509992599487305
    },
    {
      "auxiliary_loss_clip": 0.0125788,
      "auxiliary_loss_mlp": 0.01071534,
      "balance_loss_clip": 1.07601774,
      "balance_loss_mlp": 1.0404917,
      "epoch": 0.039561100255523826,
      "flos": 25338174433920.0,
      "grad_norm": 1.8797334727118136,
      "language_loss": 0.81405187,
      "learning_rate": 3.999041418526457e-06,
      "loss": 0.83734608,
      "num_input_tokens_seen": 13910085,
      "step": 658,
      "time_per_iteration": 2.53115177154541
    },
    {
      "auxiliary_loss_clip": 0.01251363,
      "auxiliary_loss_mlp": 0.01073824,
      "balance_loss_clip": 1.07394862,
      "balance_loss_mlp": 1.04073143,
      "epoch": 0.03962122350819179,
      "flos": 18219889509120.0,
      "grad_norm": 3.168404237022681,
      "language_loss": 0.9142555,
      "learning_rate": 3.999029323959287e-06,
      "loss": 0.93750733,
      "num_input_tokens_seen": 13928800,
      "step": 659,
      "time_per_iteration": 2.4812357425689697
    },
    {
      "auxiliary_loss_clip": 0.01257327,
      "auxiliary_loss_mlp": 0.01070502,
      "balance_loss_clip": 1.07588959,
      "balance_loss_mlp": 1.03981745,
      "epoch": 0.03968134676085976,
      "flos": 20522230536960.0,
      "grad_norm": 2.2204490384318465,
      "language_loss": 0.79300106,
      "learning_rate": 3.999017153588724e-06,
      "loss": 0.81627929,
      "num_input_tokens_seen": 13948325,
      "step": 660,
      "time_per_iteration": 2.489722728729248
    },
    {
      "auxiliary_loss_clip": 0.01254503,
      "auxiliary_loss_mlp": 0.01071884,
      "balance_loss_clip": 1.07736778,
      "balance_loss_mlp": 1.04017472,
      "epoch": 0.03974147001352773,
      "flos": 22422587483520.0,
      "grad_norm": 1.6167554534474062,
      "language_loss": 0.81583601,
      "learning_rate": 3.999004907415231e-06,
      "loss": 0.83909988,
      "num_input_tokens_seen": 13969090,
      "step": 661,
      "time_per_iteration": 2.509481430053711
    },
    {
      "auxiliary_loss_clip": 0.01126748,
      "auxiliary_loss_mlp": 0.0101096,
      "balance_loss_clip": 1.04644966,
      "balance_loss_mlp": 1.00375926,
      "epoch": 0.0398015932661957,
      "flos": 71128769322240.0,
      "grad_norm": 0.9158705297985051,
      "language_loss": 0.69400656,
      "learning_rate": 3.998992585439272e-06,
      "loss": 0.71538365,
      "num_input_tokens_seen": 14037555,
      "step": 662,
      "time_per_iteration": 3.2361743450164795
    },
    {
      "auxiliary_loss_clip": 0.01258082,
      "auxiliary_loss_mlp": 0.01076716,
      "balance_loss_clip": 1.07840908,
      "balance_loss_mlp": 1.04491115,
      "epoch": 0.03986171651886367,
      "flos": 16800951571200.0,
      "grad_norm": 1.920196942889544,
      "language_loss": 0.82885408,
      "learning_rate": 3.998980187661314e-06,
      "loss": 0.85220212,
      "num_input_tokens_seen": 14055765,
      "step": 663,
      "time_per_iteration": 2.4980053901672363
    },
    {
      "auxiliary_loss_clip": 0.012629,
      "auxiliary_loss_mlp": 0.0106596,
      "balance_loss_clip": 1.079687,
      "balance_loss_mlp": 1.03346324,
      "epoch": 0.03992183977153164,
      "flos": 24535068197760.0,
      "grad_norm": 2.163861249384643,
      "language_loss": 0.86956382,
      "learning_rate": 3.998967714081826e-06,
      "loss": 0.89285243,
      "num_input_tokens_seen": 14074195,
      "step": 664,
      "time_per_iteration": 2.4950802326202393
    },
    {
      "auxiliary_loss_clip": 0.0125038,
      "auxiliary_loss_mlp": 0.01066368,
      "balance_loss_clip": 1.07546103,
      "balance_loss_mlp": 1.03521848,
      "epoch": 0.03998196302419961,
      "flos": 15595897167360.0,
      "grad_norm": 1.8984004696743426,
      "language_loss": 0.84463418,
      "learning_rate": 3.998955164701281e-06,
      "loss": 0.86780167,
      "num_input_tokens_seen": 14090215,
      "step": 665,
      "time_per_iteration": 2.423750400543213
    },
    {
      "auxiliary_loss_clip": 0.01265943,
      "auxiliary_loss_mlp": 0.01084658,
      "balance_loss_clip": 1.08004594,
      "balance_loss_mlp": 1.05130303,
      "epoch": 0.04004208627686758,
      "flos": 25305065072640.0,
      "grad_norm": 1.8826682892978763,
      "language_loss": 0.81620437,
      "learning_rate": 3.998942539520158e-06,
      "loss": 0.83971035,
      "num_input_tokens_seen": 14112150,
      "step": 666,
      "time_per_iteration": 2.5383729934692383
    },
    {
      "auxiliary_loss_clip": 0.01254627,
      "auxiliary_loss_mlp": 0.01073062,
      "balance_loss_clip": 1.07571125,
      "balance_loss_mlp": 1.03956389,
      "epoch": 0.04010220952953555,
      "flos": 23475847011840.0,
      "grad_norm": 1.962142242362111,
      "language_loss": 0.8700698,
      "learning_rate": 3.998929838538932e-06,
      "loss": 0.89334673,
      "num_input_tokens_seen": 14131475,
      "step": 667,
      "time_per_iteration": 2.5061609745025635
    },
    {
      "auxiliary_loss_clip": 0.01255584,
      "auxiliary_loss_mlp": 0.0107014,
      "balance_loss_clip": 1.08016467,
      "balance_loss_mlp": 1.03964663,
      "epoch": 0.04016233278220352,
      "flos": 18617025254400.0,
      "grad_norm": 2.1698844712723084,
      "language_loss": 0.80154771,
      "learning_rate": 3.998917061758087e-06,
      "loss": 0.82480502,
      "num_input_tokens_seen": 14146165,
      "step": 668,
      "time_per_iteration": 2.4839277267456055
    },
    {
      "auxiliary_loss_clip": 0.01126079,
      "auxiliary_loss_mlp": 0.01018097,
      "balance_loss_clip": 1.04595125,
      "balance_loss_mlp": 1.01127839,
      "epoch": 0.040222456034871484,
      "flos": 70906194696960.0,
      "grad_norm": 0.7882309858167069,
      "language_loss": 0.60060382,
      "learning_rate": 3.998904209178107e-06,
      "loss": 0.62204564,
      "num_input_tokens_seen": 14215005,
      "step": 669,
      "time_per_iteration": 3.215690851211548
    },
    {
      "auxiliary_loss_clip": 0.01256024,
      "auxiliary_loss_mlp": 0.01079718,
      "balance_loss_clip": 1.07496119,
      "balance_loss_mlp": 1.04807949,
      "epoch": 0.040282579287539456,
      "flos": 23764712186880.0,
      "grad_norm": 1.8127088321060816,
      "language_loss": 0.86165673,
      "learning_rate": 3.9988912807994785e-06,
      "loss": 0.88501418,
      "num_input_tokens_seen": 14235510,
      "step": 670,
      "time_per_iteration": 2.5203728675842285
    },
    {
      "auxiliary_loss_clip": 0.0125233,
      "auxiliary_loss_mlp": 0.01079248,
      "balance_loss_clip": 1.07622802,
      "balance_loss_mlp": 1.04889703,
      "epoch": 0.04034270254020743,
      "flos": 18478518410880.0,
      "grad_norm": 2.168893746212811,
      "language_loss": 0.75634074,
      "learning_rate": 3.998878276622692e-06,
      "loss": 0.77965653,
      "num_input_tokens_seen": 14254565,
      "step": 671,
      "time_per_iteration": 2.4886200428009033
    },
    {
      "auxiliary_loss_clip": 0.01262422,
      "auxiliary_loss_mlp": 0.01074617,
      "balance_loss_clip": 1.08058035,
      "balance_loss_mlp": 1.04274058,
      "epoch": 0.040402825792875394,
      "flos": 17201858244480.0,
      "grad_norm": 1.9058916643122634,
      "language_loss": 0.9249537,
      "learning_rate": 3.998865196648242e-06,
      "loss": 0.94832402,
      "num_input_tokens_seen": 14271885,
      "step": 672,
      "time_per_iteration": 2.451547145843506
    },
    {
      "auxiliary_loss_clip": 0.012548,
      "auxiliary_loss_mlp": 0.01071366,
      "balance_loss_clip": 1.07658672,
      "balance_loss_mlp": 1.03748655,
      "epoch": 0.040462949045543366,
      "flos": 19172168928000.0,
      "grad_norm": 1.996862517924773,
      "language_loss": 0.90170383,
      "learning_rate": 3.998852040876622e-06,
      "loss": 0.9249655,
      "num_input_tokens_seen": 14289670,
      "step": 673,
      "time_per_iteration": 2.484511137008667
    },
    {
      "auxiliary_loss_clip": 0.01252367,
      "auxiliary_loss_mlp": 0.0108116,
      "balance_loss_clip": 1.07483292,
      "balance_loss_mlp": 1.0481385,
      "epoch": 0.04052307229821133,
      "flos": 24019821555840.0,
      "grad_norm": 1.9896264085857376,
      "language_loss": 0.75430405,
      "learning_rate": 3.998838809308334e-06,
      "loss": 0.77763939,
      "num_input_tokens_seen": 14309285,
      "step": 674,
      "time_per_iteration": 2.5134782791137695
    },
    {
      "auxiliary_loss_clip": 0.01261879,
      "auxiliary_loss_mlp": 0.01066614,
      "balance_loss_clip": 1.07696247,
      "balance_loss_mlp": 1.03423643,
      "epoch": 0.0405831955508793,
      "flos": 16436601964800.0,
      "grad_norm": 3.09904855653083,
      "language_loss": 0.78010499,
      "learning_rate": 3.9988255019438766e-06,
      "loss": 0.80338991,
      "num_input_tokens_seen": 14328300,
      "step": 675,
      "time_per_iteration": 2.496034860610962
    },
    {
      "auxiliary_loss_clip": 0.01255584,
      "auxiliary_loss_mlp": 0.01082005,
      "balance_loss_clip": 1.07585895,
      "balance_loss_mlp": 1.04874527,
      "epoch": 0.040643318803547275,
      "flos": 24279922915200.0,
      "grad_norm": 1.8333487957403605,
      "language_loss": 0.77007121,
      "learning_rate": 3.998812118783757e-06,
      "loss": 0.79344714,
      "num_input_tokens_seen": 14346395,
      "step": 676,
      "time_per_iteration": 2.4976956844329834
    },
    {
      "auxiliary_loss_clip": 0.01260003,
      "auxiliary_loss_mlp": 0.01081987,
      "balance_loss_clip": 1.07815659,
      "balance_loss_mlp": 1.04901397,
      "epoch": 0.04070344205621524,
      "flos": 17712076982400.0,
      "grad_norm": 3.3940211917861802,
      "language_loss": 0.85783064,
      "learning_rate": 3.9987986598284804e-06,
      "loss": 0.88125062,
      "num_input_tokens_seen": 14364605,
      "step": 677,
      "time_per_iteration": 2.469655990600586
    },
    {
      "auxiliary_loss_clip": 0.0125407,
      "auxiliary_loss_mlp": 0.01069729,
      "balance_loss_clip": 1.07640111,
      "balance_loss_mlp": 1.03711295,
      "epoch": 0.04076356530888321,
      "flos": 26177658168960.0,
      "grad_norm": 1.8070210356236929,
      "language_loss": 0.76470745,
      "learning_rate": 3.998785125078559e-06,
      "loss": 0.78794551,
      "num_input_tokens_seen": 14385265,
      "step": 678,
      "time_per_iteration": 2.5258257389068604
    },
    {
      "auxiliary_loss_clip": 0.01255311,
      "auxiliary_loss_mlp": 0.01073823,
      "balance_loss_clip": 1.07597566,
      "balance_loss_mlp": 1.04242349,
      "epoch": 0.04082368856155118,
      "flos": 35773455772800.0,
      "grad_norm": 1.846638510811123,
      "language_loss": 0.8250283,
      "learning_rate": 3.998771514534505e-06,
      "loss": 0.84831965,
      "num_input_tokens_seen": 14406090,
      "step": 679,
      "time_per_iteration": 2.6074466705322266
    },
    {
      "auxiliary_loss_clip": 0.01259219,
      "auxiliary_loss_mlp": 0.01066007,
      "balance_loss_clip": 1.08097255,
      "balance_loss_mlp": 1.03477407,
      "epoch": 0.04088381181421915,
      "flos": 28146640049280.0,
      "grad_norm": 1.9566088282602192,
      "language_loss": 0.76550275,
      "learning_rate": 3.998757828196835e-06,
      "loss": 0.78875506,
      "num_input_tokens_seen": 14425130,
      "step": 680,
      "time_per_iteration": 2.5173959732055664
    },
    {
      "auxiliary_loss_clip": 0.01258014,
      "auxiliary_loss_mlp": 0.01074278,
      "balance_loss_clip": 1.07527781,
      "balance_loss_mlp": 1.03975499,
      "epoch": 0.04094393506688712,
      "flos": 27597673514880.0,
      "grad_norm": 1.793116836579398,
      "language_loss": 0.83212912,
      "learning_rate": 3.9987440660660685e-06,
      "loss": 0.855452,
      "num_input_tokens_seen": 14447355,
      "step": 681,
      "time_per_iteration": 2.533306121826172
    },
    {
      "auxiliary_loss_clip": 0.01256628,
      "auxiliary_loss_mlp": 0.01070229,
      "balance_loss_clip": 1.07486796,
      "balance_loss_mlp": 1.03763771,
      "epoch": 0.04100405831955509,
      "flos": 23112036109440.0,
      "grad_norm": 1.923237819898852,
      "language_loss": 0.71372402,
      "learning_rate": 3.998730228142726e-06,
      "loss": 0.7369926,
      "num_input_tokens_seen": 14466790,
      "step": 682,
      "time_per_iteration": 2.469986915588379
    },
    {
      "auxiliary_loss_clip": 0.01254337,
      "auxiliary_loss_mlp": 0.01078318,
      "balance_loss_clip": 1.07603168,
      "balance_loss_mlp": 1.0472523,
      "epoch": 0.04106418157222306,
      "flos": 20156731695360.0,
      "grad_norm": 1.7890478386380715,
      "language_loss": 0.72652054,
      "learning_rate": 3.998716314427333e-06,
      "loss": 0.74984711,
      "num_input_tokens_seen": 14485195,
      "step": 683,
      "time_per_iteration": 3.851506471633911
    },
    {
      "auxiliary_loss_clip": 0.0125478,
      "auxiliary_loss_mlp": 0.01076279,
      "balance_loss_clip": 1.08191979,
      "balance_loss_mlp": 1.04571402,
      "epoch": 0.041124304824891024,
      "flos": 17420697855360.0,
      "grad_norm": 2.5919534416718935,
      "language_loss": 0.8154422,
      "learning_rate": 3.998702324920417e-06,
      "loss": 0.83875281,
      "num_input_tokens_seen": 14503370,
      "step": 684,
      "time_per_iteration": 2.450343608856201
    },
    {
      "auxiliary_loss_clip": 0.01257497,
      "auxiliary_loss_mlp": 0.01070754,
      "balance_loss_clip": 1.08056235,
      "balance_loss_mlp": 1.03744721,
      "epoch": 0.041184428077558996,
      "flos": 25780163287680.0,
      "grad_norm": 1.4567538840531897,
      "language_loss": 0.90762269,
      "learning_rate": 3.9986882596225085e-06,
      "loss": 0.93090522,
      "num_input_tokens_seen": 14526415,
      "step": 685,
      "time_per_iteration": 2.5699713230133057
    },
    {
      "auxiliary_loss_clip": 0.01256911,
      "auxiliary_loss_mlp": 0.01075932,
      "balance_loss_clip": 1.07822669,
      "balance_loss_mlp": 1.04441333,
      "epoch": 0.04124455133022697,
      "flos": 22964766347520.0,
      "grad_norm": 2.3615197925156055,
      "language_loss": 0.88268769,
      "learning_rate": 3.998674118534141e-06,
      "loss": 0.90601611,
      "num_input_tokens_seen": 14546595,
      "step": 686,
      "time_per_iteration": 3.9234211444854736
    },
    {
      "auxiliary_loss_clip": 0.0125873,
      "auxiliary_loss_mlp": 0.01076943,
      "balance_loss_clip": 1.076828,
      "balance_loss_mlp": 1.04530478,
      "epoch": 0.04130467458289493,
      "flos": 21289067015040.0,
      "grad_norm": 1.8318210298611695,
      "language_loss": 0.71688837,
      "learning_rate": 3.998659901655851e-06,
      "loss": 0.74024504,
      "num_input_tokens_seen": 14566590,
      "step": 687,
      "time_per_iteration": 3.942305088043213
    },
    {
      "auxiliary_loss_clip": 0.01255608,
      "auxiliary_loss_mlp": 0.01072926,
      "balance_loss_clip": 1.08176732,
      "balance_loss_mlp": 1.04331398,
      "epoch": 0.041364797835562905,
      "flos": 19974233669760.0,
      "grad_norm": 1.7497910595364712,
      "language_loss": 0.86064255,
      "learning_rate": 3.998645608988177e-06,
      "loss": 0.88392788,
      "num_input_tokens_seen": 14585965,
      "step": 688,
      "time_per_iteration": 2.4797966480255127
    },
    {
      "auxiliary_loss_clip": 0.01254938,
      "auxiliary_loss_mlp": 0.01081282,
      "balance_loss_clip": 1.07974577,
      "balance_loss_mlp": 1.04992962,
      "epoch": 0.04142492108823087,
      "flos": 21906227520000.0,
      "grad_norm": 2.771328456523103,
      "language_loss": 0.83210683,
      "learning_rate": 3.998631240531661e-06,
      "loss": 0.85546899,
      "num_input_tokens_seen": 14606015,
      "step": 689,
      "time_per_iteration": 2.489409923553467
    },
    {
      "auxiliary_loss_clip": 0.01252353,
      "auxiliary_loss_mlp": 0.01080227,
      "balance_loss_clip": 1.07592487,
      "balance_loss_mlp": 1.04935205,
      "epoch": 0.04148504434089884,
      "flos": 27639617621760.0,
      "grad_norm": 1.8939316836370124,
      "language_loss": 0.68170244,
      "learning_rate": 3.998616796286848e-06,
      "loss": 0.70502818,
      "num_input_tokens_seen": 14629955,
      "step": 690,
      "time_per_iteration": 3.948004961013794
    },
    {
      "auxiliary_loss_clip": 0.01252039,
      "auxiliary_loss_mlp": 0.01076404,
      "balance_loss_clip": 1.07605553,
      "balance_loss_mlp": 1.04524314,
      "epoch": 0.041545167593566815,
      "flos": 20518387781760.0,
      "grad_norm": 1.7162383688328091,
      "language_loss": 0.74996936,
      "learning_rate": 3.998602276254286e-06,
      "loss": 0.7732538,
      "num_input_tokens_seen": 14648000,
      "step": 691,
      "time_per_iteration": 2.4792983531951904
    },
    {
      "auxiliary_loss_clip": 0.01251074,
      "auxiliary_loss_mlp": 0.0108103,
      "balance_loss_clip": 1.07716656,
      "balance_loss_mlp": 1.04889083,
      "epoch": 0.04160529084623478,
      "flos": 11868907939200.0,
      "grad_norm": 1.9914756311841812,
      "language_loss": 0.84468967,
      "learning_rate": 3.998587680434526e-06,
      "loss": 0.86801076,
      "num_input_tokens_seen": 14662235,
      "step": 692,
      "time_per_iteration": 2.4478113651275635
    },
    {
      "auxiliary_loss_clip": 0.01256471,
      "auxiliary_loss_mlp": 0.01077471,
      "balance_loss_clip": 1.07570076,
      "balance_loss_mlp": 1.04409242,
      "epoch": 0.04166541409890275,
      "flos": 14828306503680.0,
      "grad_norm": 2.159535987280252,
      "language_loss": 0.89016283,
      "learning_rate": 3.99857300882812e-06,
      "loss": 0.91350228,
      "num_input_tokens_seen": 14676065,
      "step": 693,
      "time_per_iteration": 2.4439892768859863
    },
    {
      "auxiliary_loss_clip": 0.01259553,
      "auxiliary_loss_mlp": 0.01066019,
      "balance_loss_clip": 1.08082128,
      "balance_loss_mlp": 1.03531075,
      "epoch": 0.04172553735157072,
      "flos": 25808137004160.0,
      "grad_norm": 2.8487276133935975,
      "language_loss": 0.81734371,
      "learning_rate": 3.998558261435626e-06,
      "loss": 0.84059942,
      "num_input_tokens_seen": 14694955,
      "step": 694,
      "time_per_iteration": 2.5195086002349854
    },
    {
      "auxiliary_loss_clip": 0.01254915,
      "auxiliary_loss_mlp": 0.01072142,
      "balance_loss_clip": 1.07507145,
      "balance_loss_mlp": 1.04117191,
      "epoch": 0.04178566060423869,
      "flos": 24279815174400.0,
      "grad_norm": 2.3671278321000675,
      "language_loss": 0.83598286,
      "learning_rate": 3.9985434382576015e-06,
      "loss": 0.85925347,
      "num_input_tokens_seen": 14715510,
      "step": 695,
      "time_per_iteration": 2.5144309997558594
    },
    {
      "auxiliary_loss_clip": 0.01255687,
      "auxiliary_loss_mlp": 0.01080167,
      "balance_loss_clip": 1.07851648,
      "balance_loss_mlp": 1.04769409,
      "epoch": 0.04184578385690666,
      "flos": 18222008411520.0,
      "grad_norm": 2.322678800478601,
      "language_loss": 0.84757161,
      "learning_rate": 3.99852853929461e-06,
      "loss": 0.87093008,
      "num_input_tokens_seen": 14731755,
      "step": 696,
      "time_per_iteration": 2.4426229000091553
    },
    {
      "auxiliary_loss_clip": 0.01252212,
      "auxiliary_loss_mlp": 0.01085219,
      "balance_loss_clip": 1.07542408,
      "balance_loss_mlp": 1.05243671,
      "epoch": 0.041905907109574626,
      "flos": 22776342577920.0,
      "grad_norm": 2.0853579128612427,
      "language_loss": 0.93085659,
      "learning_rate": 3.998513564547216e-06,
      "loss": 0.9542309,
      "num_input_tokens_seen": 14750810,
      "step": 697,
      "time_per_iteration": 2.538133144378662
    },
    {
      "auxiliary_loss_clip": 0.01248941,
      "auxiliary_loss_mlp": 0.01076087,
      "balance_loss_clip": 1.07568896,
      "balance_loss_mlp": 1.04573655,
      "epoch": 0.0419660303622426,
      "flos": 20156947176960.0,
      "grad_norm": 3.234299987591336,
      "language_loss": 0.83806789,
      "learning_rate": 3.998498514015987e-06,
      "loss": 0.86131817,
      "num_input_tokens_seen": 14768435,
      "step": 698,
      "time_per_iteration": 2.4838712215423584
    },
    {
      "auxiliary_loss_clip": 0.01252099,
      "auxiliary_loss_mlp": 0.01096734,
      "balance_loss_clip": 1.07491469,
      "balance_loss_mlp": 1.06333137,
      "epoch": 0.042026153614910564,
      "flos": 23076376882560.0,
      "grad_norm": 2.583247213264913,
      "language_loss": 0.91226494,
      "learning_rate": 3.998483387701495e-06,
      "loss": 0.93575329,
      "num_input_tokens_seen": 14786690,
      "step": 699,
      "time_per_iteration": 2.5066189765930176
    },
    {
      "auxiliary_loss_clip": 0.01129553,
      "auxiliary_loss_mlp": 0.01009266,
      "balance_loss_clip": 1.05193877,
      "balance_loss_mlp": 1.00390112,
      "epoch": 0.042086276867578536,
      "flos": 64495243370880.0,
      "grad_norm": 0.9052507285585839,
      "language_loss": 0.67849785,
      "learning_rate": 3.998468185604312e-06,
      "loss": 0.69988608,
      "num_input_tokens_seen": 14853840,
      "step": 700,
      "time_per_iteration": 3.123793840408325
    },
    {
      "auxiliary_loss_clip": 0.01259235,
      "auxiliary_loss_mlp": 0.01083059,
      "balance_loss_clip": 1.07865882,
      "balance_loss_mlp": 1.04951322,
      "epoch": 0.04214640012024651,
      "flos": 15487016065920.0,
      "grad_norm": 2.987524473886307,
      "language_loss": 0.89117229,
      "learning_rate": 3.998452907725016e-06,
      "loss": 0.91459525,
      "num_input_tokens_seen": 14869580,
      "step": 701,
      "time_per_iteration": 2.49013352394104
    },
    {
      "auxiliary_loss_clip": 0.01255575,
      "auxiliary_loss_mlp": 0.01079551,
      "balance_loss_clip": 1.08155131,
      "balance_loss_mlp": 1.04753149,
      "epoch": 0.04220652337291447,
      "flos": 23877040993920.0,
      "grad_norm": 3.997151742114414,
      "language_loss": 0.67208856,
      "learning_rate": 3.998437554064184e-06,
      "loss": 0.69543982,
      "num_input_tokens_seen": 14891065,
      "step": 702,
      "time_per_iteration": 2.5157768726348877
    },
    {
      "auxiliary_loss_clip": 0.01122044,
      "auxiliary_loss_mlp": 0.01004701,
      "balance_loss_clip": 1.04455519,
      "balance_loss_mlp": 0.99945587,
      "epoch": 0.042266646625582445,
      "flos": 63795451628160.0,
      "grad_norm": 0.8445266931710613,
      "language_loss": 0.60818982,
      "learning_rate": 3.9984221246224006e-06,
      "loss": 0.62945735,
      "num_input_tokens_seen": 14954815,
      "step": 703,
      "time_per_iteration": 3.1593618392944336
    },
    {
      "auxiliary_loss_clip": 0.01119477,
      "auxiliary_loss_mlp": 0.01005413,
      "balance_loss_clip": 1.04243517,
      "balance_loss_mlp": 1.00016809,
      "epoch": 0.04232676987825041,
      "flos": 50018863345920.0,
      "grad_norm": 1.0184104379284098,
      "language_loss": 0.57682896,
      "learning_rate": 3.9984066194002494e-06,
      "loss": 0.59807789,
      "num_input_tokens_seen": 15003050,
      "step": 704,
      "time_per_iteration": 2.9713449478149414
    },
    {
      "auxiliary_loss_clip": 0.01255192,
      "auxiliary_loss_mlp": 0.01074428,
      "balance_loss_clip": 1.07865191,
      "balance_loss_mlp": 1.04212236,
      "epoch": 0.04238689313091838,
      "flos": 21616105368960.0,
      "grad_norm": 2.2978268656013348,
      "language_loss": 0.87722951,
      "learning_rate": 3.998391038398319e-06,
      "loss": 0.90052569,
      "num_input_tokens_seen": 15021990,
      "step": 705,
      "time_per_iteration": 2.504945993423462
    },
    {
      "auxiliary_loss_clip": 0.01243634,
      "auxiliary_loss_mlp": 0.01070954,
      "balance_loss_clip": 1.07341313,
      "balance_loss_mlp": 1.04127133,
      "epoch": 0.042447016383586354,
      "flos": 19135109070720.0,
      "grad_norm": 1.7895424711153658,
      "language_loss": 0.71079034,
      "learning_rate": 3.998375381617201e-06,
      "loss": 0.73393619,
      "num_input_tokens_seen": 15040700,
      "step": 706,
      "time_per_iteration": 2.4908573627471924
    },
    {
      "auxiliary_loss_clip": 0.01247668,
      "auxiliary_loss_mlp": 0.01068946,
      "balance_loss_clip": 1.07594621,
      "balance_loss_mlp": 1.03621149,
      "epoch": 0.04250713963625432,
      "flos": 24426007528320.0,
      "grad_norm": 2.607427980261203,
      "language_loss": 0.93483216,
      "learning_rate": 3.9983596490574875e-06,
      "loss": 0.95799834,
      "num_input_tokens_seen": 15056725,
      "step": 707,
      "time_per_iteration": 2.5100858211517334
    },
    {
      "auxiliary_loss_clip": 0.01253305,
      "auxiliary_loss_mlp": 0.01070148,
      "balance_loss_clip": 1.07454944,
      "balance_loss_mlp": 1.03819966,
      "epoch": 0.04256726288892229,
      "flos": 30367391333760.0,
      "grad_norm": 4.925244098397407,
      "language_loss": 0.81191194,
      "learning_rate": 3.998343840719776e-06,
      "loss": 0.83514655,
      "num_input_tokens_seen": 15077550,
      "step": 708,
      "time_per_iteration": 2.5856199264526367
    },
    {
      "auxiliary_loss_clip": 0.01257358,
      "auxiliary_loss_mlp": 0.01080113,
      "balance_loss_clip": 1.07808971,
      "balance_loss_mlp": 1.04668701,
      "epoch": 0.04262738614159026,
      "flos": 16362661818240.0,
      "grad_norm": 2.347093696339523,
      "language_loss": 0.82389688,
      "learning_rate": 3.998327956604666e-06,
      "loss": 0.84727162,
      "num_input_tokens_seen": 15094955,
      "step": 709,
      "time_per_iteration": 2.4977023601531982
    },
    {
      "auxiliary_loss_clip": 0.01261804,
      "auxiliary_loss_mlp": 0.01070312,
      "balance_loss_clip": 1.08144331,
      "balance_loss_mlp": 1.03838801,
      "epoch": 0.04268750939425823,
      "flos": 20412379768320.0,
      "grad_norm": 2.666747884559168,
      "language_loss": 0.85512519,
      "learning_rate": 3.99831199671276e-06,
      "loss": 0.87844628,
      "num_input_tokens_seen": 15113395,
      "step": 710,
      "time_per_iteration": 2.490229845046997
    },
    {
      "auxiliary_loss_clip": 0.0125992,
      "auxiliary_loss_mlp": 0.01071554,
      "balance_loss_clip": 1.08292031,
      "balance_loss_mlp": 1.04008269,
      "epoch": 0.0427476326469262,
      "flos": 20302959962880.0,
      "grad_norm": 2.414943052683915,
      "language_loss": 0.84779704,
      "learning_rate": 3.998295961044662e-06,
      "loss": 0.87111175,
      "num_input_tokens_seen": 15132920,
      "step": 711,
      "time_per_iteration": 2.475520133972168
    },
    {
      "auxiliary_loss_clip": 0.01248932,
      "auxiliary_loss_mlp": 0.01067966,
      "balance_loss_clip": 1.07379985,
      "balance_loss_mlp": 1.03647137,
      "epoch": 0.042807755899594166,
      "flos": 21650794928640.0,
      "grad_norm": 1.654643123845041,
      "language_loss": 0.85393977,
      "learning_rate": 3.9982798496009804e-06,
      "loss": 0.87710875,
      "num_input_tokens_seen": 15153115,
      "step": 712,
      "time_per_iteration": 2.5120770931243896
    },
    {
      "auxiliary_loss_clip": 0.01256493,
      "auxiliary_loss_mlp": 0.01073196,
      "balance_loss_clip": 1.07418346,
      "balance_loss_mlp": 1.04255915,
      "epoch": 0.04286787915226214,
      "flos": 21435007973760.0,
      "grad_norm": 2.51691310507806,
      "language_loss": 0.90784538,
      "learning_rate": 3.998263662382328e-06,
      "loss": 0.93114227,
      "num_input_tokens_seen": 15172770,
      "step": 713,
      "time_per_iteration": 2.4832911491394043
    },
    {
      "auxiliary_loss_clip": 0.01121091,
      "auxiliary_loss_mlp": 0.01023259,
      "balance_loss_clip": 1.04465437,
      "balance_loss_mlp": 1.0183239,
      "epoch": 0.04292800240493011,
      "flos": 66397970615040.0,
      "grad_norm": 0.8811728709994043,
      "language_loss": 0.63713288,
      "learning_rate": 3.9982473993893165e-06,
      "loss": 0.65857637,
      "num_input_tokens_seen": 15240055,
      "step": 714,
      "time_per_iteration": 3.2018778324127197
    },
    {
      "auxiliary_loss_clip": 0.01251504,
      "auxiliary_loss_mlp": 0.0108492,
      "balance_loss_clip": 1.07920837,
      "balance_loss_mlp": 1.05392528,
      "epoch": 0.042988125657598075,
      "flos": 31650264552960.0,
      "grad_norm": 1.838967079899082,
      "language_loss": 0.74468529,
      "learning_rate": 3.998231060622563e-06,
      "loss": 0.76804948,
      "num_input_tokens_seen": 15261585,
      "step": 715,
      "time_per_iteration": 2.571183204650879
    },
    {
      "auxiliary_loss_clip": 0.01254444,
      "auxiliary_loss_mlp": 0.01075104,
      "balance_loss_clip": 1.07951045,
      "balance_loss_mlp": 1.04232144,
      "epoch": 0.04304824891026605,
      "flos": 33248468292480.0,
      "grad_norm": 2.255241924457185,
      "language_loss": 0.72790945,
      "learning_rate": 3.998214646082688e-06,
      "loss": 0.75120491,
      "num_input_tokens_seen": 15281160,
      "step": 716,
      "time_per_iteration": 2.6089766025543213
    },
    {
      "auxiliary_loss_clip": 0.01110715,
      "auxiliary_loss_mlp": 0.01006259,
      "balance_loss_clip": 1.03539634,
      "balance_loss_mlp": 1.00139511,
      "epoch": 0.04310837216293401,
      "flos": 64064782782720.0,
      "grad_norm": 1.36413784072223,
      "language_loss": 0.65578079,
      "learning_rate": 3.998198155770314e-06,
      "loss": 0.67695051,
      "num_input_tokens_seen": 15344505,
      "step": 717,
      "time_per_iteration": 3.1340420246124268
    },
    {
      "auxiliary_loss_clip": 0.01110468,
      "auxiliary_loss_mlp": 0.01006442,
      "balance_loss_clip": 1.03563023,
      "balance_loss_mlp": 1.00155497,
      "epoch": 0.043168495415601985,
      "flos": 61343757849600.0,
      "grad_norm": 0.9831549069726905,
      "language_loss": 0.5879755,
      "learning_rate": 3.998181589686065e-06,
      "loss": 0.60914469,
      "num_input_tokens_seen": 15404050,
      "step": 718,
      "time_per_iteration": 2.9251046180725098
    },
    {
      "auxiliary_loss_clip": 0.0125152,
      "auxiliary_loss_mlp": 0.01074686,
      "balance_loss_clip": 1.07893729,
      "balance_loss_mlp": 1.04130697,
      "epoch": 0.04322861866826996,
      "flos": 20704261685760.0,
      "grad_norm": 1.8847359734447025,
      "language_loss": 0.91464567,
      "learning_rate": 3.99816494783057e-06,
      "loss": 0.93790781,
      "num_input_tokens_seen": 15424190,
      "step": 719,
      "time_per_iteration": 2.480520725250244
    },
    {
      "auxiliary_loss_clip": 0.01247591,
      "auxiliary_loss_mlp": 0.01071594,
      "balance_loss_clip": 1.07297122,
      "balance_loss_mlp": 1.04157722,
      "epoch": 0.04328874192093792,
      "flos": 30373352991360.0,
      "grad_norm": 1.6197938620912151,
      "language_loss": 0.66681176,
      "learning_rate": 3.99814823020446e-06,
      "loss": 0.69000363,
      "num_input_tokens_seen": 15446500,
      "step": 720,
      "time_per_iteration": 2.5536956787109375
    },
    {
      "auxiliary_loss_clip": 0.01247406,
      "auxiliary_loss_mlp": 0.01075771,
      "balance_loss_clip": 1.07514811,
      "balance_loss_mlp": 1.04399025,
      "epoch": 0.043348865173605894,
      "flos": 21944795748480.0,
      "grad_norm": 1.9383275550222532,
      "language_loss": 0.77480465,
      "learning_rate": 3.9981314368083684e-06,
      "loss": 0.79803646,
      "num_input_tokens_seen": 15465830,
      "step": 721,
      "time_per_iteration": 2.487926721572876
    },
    {
      "auxiliary_loss_clip": 0.01252422,
      "auxiliary_loss_mlp": 0.01086979,
      "balance_loss_clip": 1.07756495,
      "balance_loss_mlp": 1.0560087,
      "epoch": 0.04340898842627386,
      "flos": 15264225959040.0,
      "grad_norm": 2.511747489707896,
      "language_loss": 0.88315809,
      "learning_rate": 3.998114567642933e-06,
      "loss": 0.90655208,
      "num_input_tokens_seen": 15479985,
      "step": 722,
      "time_per_iteration": 2.4527525901794434
    },
    {
      "auxiliary_loss_clip": 0.01256942,
      "auxiliary_loss_mlp": 0.01076551,
      "balance_loss_clip": 1.08006072,
      "balance_loss_mlp": 1.04653418,
      "epoch": 0.04346911167894183,
      "flos": 27965434913280.0,
      "grad_norm": 1.8231900395995235,
      "language_loss": 0.84438741,
      "learning_rate": 3.998097622708792e-06,
      "loss": 0.86772239,
      "num_input_tokens_seen": 15501545,
      "step": 723,
      "time_per_iteration": 3.894731283187866
    },
    {
      "auxiliary_loss_clip": 0.01260079,
      "auxiliary_loss_mlp": 0.01079058,
      "balance_loss_clip": 1.0832659,
      "balance_loss_mlp": 1.04796767,
      "epoch": 0.0435292349316098,
      "flos": 29242202820480.0,
      "grad_norm": 2.045688703605771,
      "language_loss": 0.82843268,
      "learning_rate": 3.99808060200659e-06,
      "loss": 0.85182405,
      "num_input_tokens_seen": 15521725,
      "step": 724,
      "time_per_iteration": 2.553359031677246
    },
    {
      "auxiliary_loss_clip": 0.01252676,
      "auxiliary_loss_mlp": 0.01086063,
      "balance_loss_clip": 1.0775435,
      "balance_loss_mlp": 1.05359006,
      "epoch": 0.04358935818427777,
      "flos": 20558356640640.0,
      "grad_norm": 1.988237117483734,
      "language_loss": 0.79552901,
      "learning_rate": 3.998063505536971e-06,
      "loss": 0.81891632,
      "num_input_tokens_seen": 15540910,
      "step": 725,
      "time_per_iteration": 3.8466293811798096
    },
    {
      "auxiliary_loss_clip": 0.01261587,
      "auxiliary_loss_mlp": 0.01072773,
      "balance_loss_clip": 1.07915306,
      "balance_loss_mlp": 1.04049134,
      "epoch": 0.04364948143694574,
      "flos": 14464926564480.0,
      "grad_norm": 2.091345909136279,
      "language_loss": 0.87197322,
      "learning_rate": 3.998046333300584e-06,
      "loss": 0.89531684,
      "num_input_tokens_seen": 15558640,
      "step": 726,
      "time_per_iteration": 3.849003791809082
    },
    {
      "auxiliary_loss_clip": 0.01126927,
      "auxiliary_loss_mlp": 0.010231,
      "balance_loss_clip": 1.05230141,
      "balance_loss_mlp": 1.01828372,
      "epoch": 0.043709604689613706,
      "flos": 50067268922880.0,
      "grad_norm": 0.9173065387859451,
      "language_loss": 0.55872881,
      "learning_rate": 3.998029085298079e-06,
      "loss": 0.58022904,
      "num_input_tokens_seen": 15612975,
      "step": 727,
      "time_per_iteration": 3.1951780319213867
    },
    {
      "auxiliary_loss_clip": 0.01253438,
      "auxiliary_loss_mlp": 0.01077901,
      "balance_loss_clip": 1.07788634,
      "balance_loss_mlp": 1.04626238,
      "epoch": 0.04376972794228168,
      "flos": 13991588115840.0,
      "grad_norm": 2.02170613627145,
      "language_loss": 0.82037175,
      "learning_rate": 3.998011761530112e-06,
      "loss": 0.84368515,
      "num_input_tokens_seen": 15631070,
      "step": 728,
      "time_per_iteration": 2.482454776763916
    },
    {
      "auxiliary_loss_clip": 0.01247585,
      "auxiliary_loss_mlp": 0.01070672,
      "balance_loss_clip": 1.07657981,
      "balance_loss_mlp": 1.04048872,
      "epoch": 0.04382985119494965,
      "flos": 22009901149440.0,
      "grad_norm": 2.15447269369695,
      "language_loss": 0.76986587,
      "learning_rate": 3.997994361997338e-06,
      "loss": 0.79304838,
      "num_input_tokens_seen": 15647825,
      "step": 729,
      "time_per_iteration": 3.9494788646698
    },
    {
      "auxiliary_loss_clip": 0.01255064,
      "auxiliary_loss_mlp": 0.01079642,
      "balance_loss_clip": 1.07632136,
      "balance_loss_mlp": 1.04876661,
      "epoch": 0.043889974447617615,
      "flos": 24206521472640.0,
      "grad_norm": 1.9931661930854947,
      "language_loss": 0.94973999,
      "learning_rate": 3.997976886700417e-06,
      "loss": 0.97308707,
      "num_input_tokens_seen": 15668260,
      "step": 730,
      "time_per_iteration": 2.5052812099456787
    },
    {
      "auxiliary_loss_clip": 0.01249149,
      "auxiliary_loss_mlp": 0.01072001,
      "balance_loss_clip": 1.07240653,
      "balance_loss_mlp": 1.03897977,
      "epoch": 0.04395009770028559,
      "flos": 17274541415040.0,
      "grad_norm": 3.497273825391156,
      "language_loss": 0.88617849,
      "learning_rate": 3.997959335640013e-06,
      "loss": 0.90938997,
      "num_input_tokens_seen": 15685630,
      "step": 731,
      "time_per_iteration": 2.4540822505950928
    },
    {
      "auxiliary_loss_clip": 0.01249975,
      "auxiliary_loss_mlp": 0.01070735,
      "balance_loss_clip": 1.07549679,
      "balance_loss_mlp": 1.04187393,
      "epoch": 0.04401022095295355,
      "flos": 12310286261760.0,
      "grad_norm": 3.172461984134687,
      "language_loss": 0.88943565,
      "learning_rate": 3.997941708816791e-06,
      "loss": 0.91264272,
      "num_input_tokens_seen": 15698645,
      "step": 732,
      "time_per_iteration": 2.443603277206421
    },
    {
      "auxiliary_loss_clip": 0.01251924,
      "auxiliary_loss_mlp": 0.01078384,
      "balance_loss_clip": 1.07617939,
      "balance_loss_mlp": 1.04731798,
      "epoch": 0.044070344205621524,
      "flos": 20959658363520.0,
      "grad_norm": 2.2864011189564546,
      "language_loss": 0.85841179,
      "learning_rate": 3.997924006231419e-06,
      "loss": 0.88171488,
      "num_input_tokens_seen": 15716775,
      "step": 733,
      "time_per_iteration": 2.4763846397399902
    },
    {
      "auxiliary_loss_clip": 0.01255541,
      "auxiliary_loss_mlp": 0.01087581,
      "balance_loss_clip": 1.07699037,
      "balance_loss_mlp": 1.05377281,
      "epoch": 0.044130467458289496,
      "flos": 13845288021120.0,
      "grad_norm": 2.4033747210251843,
      "language_loss": 0.91343981,
      "learning_rate": 3.9979062278845685e-06,
      "loss": 0.93687105,
      "num_input_tokens_seen": 15733320,
      "step": 734,
      "time_per_iteration": 2.479966402053833
    },
    {
      "auxiliary_loss_clip": 0.01247769,
      "auxiliary_loss_mlp": 0.01067713,
      "balance_loss_clip": 1.07781398,
      "balance_loss_mlp": 1.03811288,
      "epoch": 0.04419059071095746,
      "flos": 28655063107200.0,
      "grad_norm": 2.4223649138158003,
      "language_loss": 0.78063184,
      "learning_rate": 3.9978883737769125e-06,
      "loss": 0.80378664,
      "num_input_tokens_seen": 15752705,
      "step": 735,
      "time_per_iteration": 2.5412886142730713
    },
    {
      "auxiliary_loss_clip": 0.01243273,
      "auxiliary_loss_mlp": 0.01067476,
      "balance_loss_clip": 1.07247901,
      "balance_loss_mlp": 1.03785276,
      "epoch": 0.04425071396362543,
      "flos": 28183304856960.0,
      "grad_norm": 2.4100771569239217,
      "language_loss": 0.88533014,
      "learning_rate": 3.9978704439091305e-06,
      "loss": 0.90843761,
      "num_input_tokens_seen": 15772800,
      "step": 736,
      "time_per_iteration": 2.5225746631622314
    },
    {
      "auxiliary_loss_clip": 0.01246785,
      "auxiliary_loss_mlp": 0.01080797,
      "balance_loss_clip": 1.07856536,
      "balance_loss_mlp": 1.05097103,
      "epoch": 0.0443108372162934,
      "flos": 23658452778240.0,
      "grad_norm": 3.9013177773359704,
      "language_loss": 0.84129977,
      "learning_rate": 3.997852438281901e-06,
      "loss": 0.86457562,
      "num_input_tokens_seen": 15793665,
      "step": 737,
      "time_per_iteration": 2.5188817977905273
    },
    {
      "auxiliary_loss_clip": 0.01252148,
      "auxiliary_loss_mlp": 0.01075241,
      "balance_loss_clip": 1.07787871,
      "balance_loss_mlp": 1.04183817,
      "epoch": 0.04437096046896137,
      "flos": 33979861025280.0,
      "grad_norm": 2.422845689764848,
      "language_loss": 0.84817713,
      "learning_rate": 3.997834356895906e-06,
      "loss": 0.87145102,
      "num_input_tokens_seen": 15813175,
      "step": 738,
      "time_per_iteration": 2.577906608581543
    },
    {
      "auxiliary_loss_clip": 0.01115685,
      "auxiliary_loss_mlp": 0.01022955,
      "balance_loss_clip": 1.0399214,
      "balance_loss_mlp": 1.0186398,
      "epoch": 0.04443108372162934,
      "flos": 67397506375680.0,
      "grad_norm": 0.8785747309996341,
      "language_loss": 0.59174645,
      "learning_rate": 3.9978161997518324e-06,
      "loss": 0.61313283,
      "num_input_tokens_seen": 15872050,
      "step": 739,
      "time_per_iteration": 3.050217628479004
    },
    {
      "auxiliary_loss_clip": 0.01249794,
      "auxiliary_loss_mlp": 0.01072794,
      "balance_loss_clip": 1.07896817,
      "balance_loss_mlp": 1.04344428,
      "epoch": 0.04449120697429731,
      "flos": 29752672953600.0,
      "grad_norm": 2.144007634602174,
      "language_loss": 0.91327178,
      "learning_rate": 3.997797966850369e-06,
      "loss": 0.93649769,
      "num_input_tokens_seen": 15891085,
      "step": 740,
      "time_per_iteration": 2.5430283546447754
    },
    {
      "auxiliary_loss_clip": 0.01254131,
      "auxiliary_loss_mlp": 0.01071002,
      "balance_loss_clip": 1.08104539,
      "balance_loss_mlp": 1.04218912,
      "epoch": 0.04455133022696528,
      "flos": 36502119072000.0,
      "grad_norm": 2.2230877648367002,
      "language_loss": 0.72081542,
      "learning_rate": 3.997779658192205e-06,
      "loss": 0.74406677,
      "num_input_tokens_seen": 15914225,
      "step": 741,
      "time_per_iteration": 2.61557674407959
    },
    {
      "auxiliary_loss_clip": 0.01243308,
      "auxiliary_loss_mlp": 0.01080602,
      "balance_loss_clip": 1.0737747,
      "balance_loss_mlp": 1.05108547,
      "epoch": 0.044611453479633245,
      "flos": 28803661672320.0,
      "grad_norm": 1.7620407795085722,
      "language_loss": 0.8850745,
      "learning_rate": 3.997761273778037e-06,
      "loss": 0.90831363,
      "num_input_tokens_seen": 15934540,
      "step": 742,
      "time_per_iteration": 2.523158073425293
    },
    {
      "auxiliary_loss_clip": 0.0124367,
      "auxiliary_loss_mlp": 0.01060847,
      "balance_loss_clip": 1.07474065,
      "balance_loss_mlp": 1.03094912,
      "epoch": 0.04467157673230122,
      "flos": 20010970304640.0,
      "grad_norm": 1.8629622203053944,
      "language_loss": 0.8397727,
      "learning_rate": 3.997742813608561e-06,
      "loss": 0.86281782,
      "num_input_tokens_seen": 15952560,
      "step": 743,
      "time_per_iteration": 2.475792169570923
    },
    {
      "auxiliary_loss_clip": 0.01250785,
      "auxiliary_loss_mlp": 0.01070553,
      "balance_loss_clip": 1.0769738,
      "balance_loss_mlp": 1.04115546,
      "epoch": 0.04473169998496919,
      "flos": 18004964480640.0,
      "grad_norm": 4.313392535547785,
      "language_loss": 0.79953408,
      "learning_rate": 3.997724277684479e-06,
      "loss": 0.82274747,
      "num_input_tokens_seen": 15970620,
      "step": 744,
      "time_per_iteration": 2.4453978538513184
    },
    {
      "auxiliary_loss_clip": 0.0124343,
      "auxiliary_loss_mlp": 0.01070831,
      "balance_loss_clip": 1.07467628,
      "balance_loss_mlp": 1.04162514,
      "epoch": 0.044791823237637154,
      "flos": 20631722169600.0,
      "grad_norm": 1.8845812783584215,
      "language_loss": 0.8536855,
      "learning_rate": 3.99770566600649e-06,
      "loss": 0.87682807,
      "num_input_tokens_seen": 15987325,
      "step": 745,
      "time_per_iteration": 2.478398323059082
    },
    {
      "auxiliary_loss_clip": 0.01243613,
      "auxiliary_loss_mlp": 0.01068784,
      "balance_loss_clip": 1.07413411,
      "balance_loss_mlp": 1.03874326,
      "epoch": 0.04485194649030513,
      "flos": 31176171918720.0,
      "grad_norm": 1.6666171424001954,
      "language_loss": 0.68985176,
      "learning_rate": 3.997686978575302e-06,
      "loss": 0.71297574,
      "num_input_tokens_seen": 16008310,
      "step": 746,
      "time_per_iteration": 2.550433874130249
    },
    {
      "auxiliary_loss_clip": 0.01253217,
      "auxiliary_loss_mlp": 0.01081538,
      "balance_loss_clip": 1.08141601,
      "balance_loss_mlp": 1.05059147,
      "epoch": 0.04491206974297309,
      "flos": 26143291831680.0,
      "grad_norm": 5.231731256899493,
      "language_loss": 0.6895206,
      "learning_rate": 3.997668215391625e-06,
      "loss": 0.71286809,
      "num_input_tokens_seen": 16029620,
      "step": 747,
      "time_per_iteration": 2.5539510250091553
    },
    {
      "auxiliary_loss_clip": 0.01250931,
      "auxiliary_loss_mlp": 0.01083399,
      "balance_loss_clip": 1.07707548,
      "balance_loss_mlp": 1.05225003,
      "epoch": 0.044972192995641064,
      "flos": 20667668705280.0,
      "grad_norm": 1.9367419301393973,
      "language_loss": 0.66610992,
      "learning_rate": 3.997649376456168e-06,
      "loss": 0.68945324,
      "num_input_tokens_seen": 16049065,
      "step": 748,
      "time_per_iteration": 2.4767324924468994
    },
    {
      "auxiliary_loss_clip": 0.01254079,
      "auxiliary_loss_mlp": 0.01082655,
      "balance_loss_clip": 1.0833205,
      "balance_loss_mlp": 1.05297232,
      "epoch": 0.045032316248309036,
      "flos": 16106834177280.0,
      "grad_norm": 2.524246522525193,
      "language_loss": 0.76673996,
      "learning_rate": 3.997630461769647e-06,
      "loss": 0.79010731,
      "num_input_tokens_seen": 16066765,
      "step": 749,
      "time_per_iteration": 2.4764151573181152
    },
    {
      "auxiliary_loss_clip": 0.01251115,
      "auxiliary_loss_mlp": 0.01080557,
      "balance_loss_clip": 1.07920742,
      "balance_loss_mlp": 1.05139875,
      "epoch": 0.045092439500977,
      "flos": 17858843953920.0,
      "grad_norm": 2.2951621087839453,
      "language_loss": 0.88605535,
      "learning_rate": 3.997611471332778e-06,
      "loss": 0.90937209,
      "num_input_tokens_seen": 16085980,
      "step": 750,
      "time_per_iteration": 2.441727638244629
    },
    {
      "auxiliary_loss_clip": 0.01248424,
      "auxiliary_loss_mlp": 0.01077257,
      "balance_loss_clip": 1.07408965,
      "balance_loss_mlp": 1.0438782,
      "epoch": 0.04515256275364497,
      "flos": 24462815990400.0,
      "grad_norm": 1.7534666436342763,
      "language_loss": 0.74666762,
      "learning_rate": 3.9975924051462825e-06,
      "loss": 0.7699244,
      "num_input_tokens_seen": 16106260,
      "step": 751,
      "time_per_iteration": 2.5410470962524414
    },
    {
      "auxiliary_loss_clip": 0.01243635,
      "auxiliary_loss_mlp": 0.01079546,
      "balance_loss_clip": 1.07410443,
      "balance_loss_mlp": 1.05074525,
      "epoch": 0.04521268600631294,
      "flos": 20916385453440.0,
      "grad_norm": 2.295971916288258,
      "language_loss": 0.69418418,
      "learning_rate": 3.997573263210883e-06,
      "loss": 0.71741599,
      "num_input_tokens_seen": 16123475,
      "step": 752,
      "time_per_iteration": 2.479640483856201
    },
    {
      "auxiliary_loss_clip": 0.0124532,
      "auxiliary_loss_mlp": 0.01065509,
      "balance_loss_clip": 1.07489073,
      "balance_loss_mlp": 1.03668404,
      "epoch": 0.04527280925898091,
      "flos": 13371374954880.0,
      "grad_norm": 2.436510867077347,
      "language_loss": 0.9194591,
      "learning_rate": 3.997554045527305e-06,
      "loss": 0.94256735,
      "num_input_tokens_seen": 16138335,
      "step": 753,
      "time_per_iteration": 2.48738169670105
    },
    {
      "auxiliary_loss_clip": 0.01249109,
      "auxiliary_loss_mlp": 0.01084318,
      "balance_loss_clip": 1.07751989,
      "balance_loss_mlp": 1.05508828,
      "epoch": 0.04533293251164888,
      "flos": 23254565276160.0,
      "grad_norm": 2.2098293179081305,
      "language_loss": 0.91399074,
      "learning_rate": 3.997534752096277e-06,
      "loss": 0.937325,
      "num_input_tokens_seen": 16157110,
      "step": 754,
      "time_per_iteration": 2.49340558052063
    },
    {
      "auxiliary_loss_clip": 0.01239653,
      "auxiliary_loss_mlp": 0.01078328,
      "balance_loss_clip": 1.07571316,
      "balance_loss_mlp": 1.04742885,
      "epoch": 0.04539305576431685,
      "flos": 12422004537600.0,
      "grad_norm": 2.2623222092758892,
      "language_loss": 0.78551352,
      "learning_rate": 3.997515382918531e-06,
      "loss": 0.80869335,
      "num_input_tokens_seen": 16174155,
      "step": 755,
      "time_per_iteration": 2.477518320083618
    },
    {
      "auxiliary_loss_clip": 0.01251429,
      "auxiliary_loss_mlp": 0.0107947,
      "balance_loss_clip": 1.07907486,
      "balance_loss_mlp": 1.05004895,
      "epoch": 0.04545317901698482,
      "flos": 16070995382400.0,
      "grad_norm": 1.9951826329934674,
      "language_loss": 0.78643048,
      "learning_rate": 3.9974959379948015e-06,
      "loss": 0.80973947,
      "num_input_tokens_seen": 16192240,
      "step": 756,
      "time_per_iteration": 2.458444833755493
    },
    {
      "auxiliary_loss_clip": 0.01114259,
      "auxiliary_loss_mlp": 0.01039793,
      "balance_loss_clip": 1.04198599,
      "balance_loss_mlp": 1.03504896,
      "epoch": 0.045513302269652785,
      "flos": 66396139021440.0,
      "grad_norm": 0.8212813490298987,
      "language_loss": 0.62675792,
      "learning_rate": 3.997476417325827e-06,
      "loss": 0.64829844,
      "num_input_tokens_seen": 16255775,
      "step": 757,
      "time_per_iteration": 3.1493735313415527
    },
    {
      "auxiliary_loss_clip": 0.01245854,
      "auxiliary_loss_mlp": 0.01070045,
      "balance_loss_clip": 1.0769403,
      "balance_loss_mlp": 1.04174519,
      "epoch": 0.04557342552232076,
      "flos": 21471169991040.0,
      "grad_norm": 1.44699575784267,
      "language_loss": 0.84337306,
      "learning_rate": 3.997456820912346e-06,
      "loss": 0.86653209,
      "num_input_tokens_seen": 16277015,
      "step": 758,
      "time_per_iteration": 2.503798007965088
    },
    {
      "auxiliary_loss_clip": 0.01240061,
      "auxiliary_loss_mlp": 0.01067799,
      "balance_loss_clip": 1.07182825,
      "balance_loss_mlp": 1.04024935,
      "epoch": 0.04563354877498873,
      "flos": 23732680233600.0,
      "grad_norm": 1.884062487466806,
      "language_loss": 0.88095313,
      "learning_rate": 3.997437148755101e-06,
      "loss": 0.90403175,
      "num_input_tokens_seen": 16296005,
      "step": 759,
      "time_per_iteration": 2.508530855178833
    },
    {
      "auxiliary_loss_clip": 0.01250368,
      "auxiliary_loss_mlp": 0.01078588,
      "balance_loss_clip": 1.07985139,
      "balance_loss_mlp": 1.04801106,
      "epoch": 0.045693672027656694,
      "flos": 25735741142400.0,
      "grad_norm": 1.9368538333189138,
      "language_loss": 0.73655021,
      "learning_rate": 3.9974174008548405e-06,
      "loss": 0.75983977,
      "num_input_tokens_seen": 16315300,
      "step": 760,
      "time_per_iteration": 2.5314486026763916
    },
    {
      "auxiliary_loss_clip": 0.01251546,
      "auxiliary_loss_mlp": 0.01074986,
      "balance_loss_clip": 1.0838654,
      "balance_loss_mlp": 1.0469594,
      "epoch": 0.045753795280324666,
      "flos": 19719016560000.0,
      "grad_norm": 1.9591846113694338,
      "language_loss": 0.82625586,
      "learning_rate": 3.9973975772123105e-06,
      "loss": 0.84952116,
      "num_input_tokens_seen": 16333820,
      "step": 761,
      "time_per_iteration": 2.489737033843994
    },
    {
      "auxiliary_loss_clip": 0.01244902,
      "auxiliary_loss_mlp": 0.01075711,
      "balance_loss_clip": 1.07635701,
      "balance_loss_mlp": 1.04683852,
      "epoch": 0.04581391853299264,
      "flos": 23255786338560.0,
      "grad_norm": 1.8086329031035762,
      "language_loss": 0.79603302,
      "learning_rate": 3.997377677828266e-06,
      "loss": 0.81923914,
      "num_input_tokens_seen": 16355290,
      "step": 762,
      "time_per_iteration": 2.529679775238037
    },
    {
      "auxiliary_loss_clip": 0.01107278,
      "auxiliary_loss_mlp": 0.01013927,
      "balance_loss_clip": 1.03486323,
      "balance_loss_mlp": 1.00932586,
      "epoch": 0.0458740417856606,
      "flos": 64231155601920.0,
      "grad_norm": 1.040188661578853,
      "language_loss": 0.58747685,
      "learning_rate": 3.9973577027034585e-06,
      "loss": 0.60868889,
      "num_input_tokens_seen": 16415995,
      "step": 763,
      "time_per_iteration": 4.526393175125122
    },
    {
      "auxiliary_loss_clip": 0.01247215,
      "auxiliary_loss_mlp": 0.01074598,
      "balance_loss_clip": 1.07709277,
      "balance_loss_mlp": 1.04574943,
      "epoch": 0.045934165038328575,
      "flos": 20770121272320.0,
      "grad_norm": 2.375432560898797,
      "language_loss": 0.87829518,
      "learning_rate": 3.9973376518386475e-06,
      "loss": 0.90151328,
      "num_input_tokens_seen": 16433120,
      "step": 764,
      "time_per_iteration": 3.922713279724121
    },
    {
      "auxiliary_loss_clip": 0.01248823,
      "auxiliary_loss_mlp": 0.01079525,
      "balance_loss_clip": 1.07897139,
      "balance_loss_mlp": 1.050879,
      "epoch": 0.04599428829099654,
      "flos": 30262891691520.0,
      "grad_norm": 2.1029605367212705,
      "language_loss": 0.85938191,
      "learning_rate": 3.997317525234592e-06,
      "loss": 0.8826654,
      "num_input_tokens_seen": 16453360,
      "step": 765,
      "time_per_iteration": 2.58488392829895
    },
    {
      "auxiliary_loss_clip": 0.01251573,
      "auxiliary_loss_mlp": 0.01072392,
      "balance_loss_clip": 1.07816458,
      "balance_loss_mlp": 1.04087281,
      "epoch": 0.04605441154366451,
      "flos": 23038921975680.0,
      "grad_norm": 3.967786082390371,
      "language_loss": 0.88215971,
      "learning_rate": 3.997297322892056e-06,
      "loss": 0.90539932,
      "num_input_tokens_seen": 16471160,
      "step": 766,
      "time_per_iteration": 3.8616089820861816
    },
    {
      "auxiliary_loss_clip": 0.01246871,
      "auxiliary_loss_mlp": 0.01076691,
      "balance_loss_clip": 1.07713246,
      "balance_loss_mlp": 1.04738927,
      "epoch": 0.046114534796332485,
      "flos": 22017407091840.0,
      "grad_norm": 2.4426135146880097,
      "language_loss": 0.83775514,
      "learning_rate": 3.997277044811806e-06,
      "loss": 0.86099076,
      "num_input_tokens_seen": 16488940,
      "step": 767,
      "time_per_iteration": 2.5143752098083496
    },
    {
      "auxiliary_loss_clip": 0.01247751,
      "auxiliary_loss_mlp": 0.01063066,
      "balance_loss_clip": 1.0803442,
      "balance_loss_mlp": 1.03327525,
      "epoch": 0.04617465804900045,
      "flos": 29862380067840.0,
      "grad_norm": 2.665062574564417,
      "language_loss": 0.87119257,
      "learning_rate": 3.99725669099461e-06,
      "loss": 0.89430076,
      "num_input_tokens_seen": 16509505,
      "step": 768,
      "time_per_iteration": 2.5956757068634033
    },
    {
      "auxiliary_loss_clip": 0.01242505,
      "auxiliary_loss_mlp": 0.01069907,
      "balance_loss_clip": 1.07306993,
      "balance_loss_mlp": 1.04260862,
      "epoch": 0.04623478130166842,
      "flos": 25630056351360.0,
      "grad_norm": 2.3409045144873186,
      "language_loss": 0.74993503,
      "learning_rate": 3.9972362614412395e-06,
      "loss": 0.77305919,
      "num_input_tokens_seen": 16528840,
      "step": 769,
      "time_per_iteration": 3.953342914581299
    },
    {
      "auxiliary_loss_clip": 0.01241863,
      "auxiliary_loss_mlp": 0.01068419,
      "balance_loss_clip": 1.07704031,
      "balance_loss_mlp": 1.04214561,
      "epoch": 0.04629490455433639,
      "flos": 20449080489600.0,
      "grad_norm": 3.942513204832819,
      "language_loss": 0.86191607,
      "learning_rate": 3.997215756152471e-06,
      "loss": 0.88501894,
      "num_input_tokens_seen": 16548335,
      "step": 770,
      "time_per_iteration": 2.489542245864868
    },
    {
      "auxiliary_loss_clip": 0.01250659,
      "auxiliary_loss_mlp": 0.01070067,
      "balance_loss_clip": 1.07556844,
      "balance_loss_mlp": 1.04126596,
      "epoch": 0.04635502780700436,
      "flos": 23148736830720.0,
      "grad_norm": 2.090621220760043,
      "language_loss": 0.86928719,
      "learning_rate": 3.99719517512908e-06,
      "loss": 0.89249444,
      "num_input_tokens_seen": 16567725,
      "step": 771,
      "time_per_iteration": 2.477332830429077
    },
    {
      "auxiliary_loss_clip": 0.01248303,
      "auxiliary_loss_mlp": 0.01080548,
      "balance_loss_clip": 1.07343459,
      "balance_loss_mlp": 1.05045962,
      "epoch": 0.04641515105967233,
      "flos": 23292020183040.0,
      "grad_norm": 3.480609359772438,
      "language_loss": 0.83834875,
      "learning_rate": 3.997174518371848e-06,
      "loss": 0.86163729,
      "num_input_tokens_seen": 16588175,
      "step": 772,
      "time_per_iteration": 2.558382987976074
    },
    {
      "auxiliary_loss_clip": 0.01246311,
      "auxiliary_loss_mlp": 0.01066675,
      "balance_loss_clip": 1.07794011,
      "balance_loss_mlp": 1.03984118,
      "epoch": 0.046475274312340296,
      "flos": 25115204759040.0,
      "grad_norm": 1.8278369870749573,
      "language_loss": 0.73804945,
      "learning_rate": 3.997153785881557e-06,
      "loss": 0.76117933,
      "num_input_tokens_seen": 16607735,
      "step": 773,
      "time_per_iteration": 2.511962413787842
    },
    {
      "auxiliary_loss_clip": 0.01242208,
      "auxiliary_loss_mlp": 0.01069037,
      "balance_loss_clip": 1.0767622,
      "balance_loss_mlp": 1.03909159,
      "epoch": 0.04653539756500827,
      "flos": 25264916645760.0,
      "grad_norm": 2.159746583844111,
      "language_loss": 0.78493178,
      "learning_rate": 3.997132977658996e-06,
      "loss": 0.8080442,
      "num_input_tokens_seen": 16627225,
      "step": 774,
      "time_per_iteration": 2.5290379524230957
    },
    {
      "auxiliary_loss_clip": 0.01238986,
      "auxiliary_loss_mlp": 0.01068362,
      "balance_loss_clip": 1.07366967,
      "balance_loss_mlp": 1.04169452,
      "epoch": 0.046595520817676234,
      "flos": 35404150089600.0,
      "grad_norm": 2.6148932968465695,
      "language_loss": 0.73463553,
      "learning_rate": 3.997112093704952e-06,
      "loss": 0.75770903,
      "num_input_tokens_seen": 16647785,
      "step": 775,
      "time_per_iteration": 2.601954221725464
    },
    {
      "auxiliary_loss_clip": 0.01243921,
      "auxiliary_loss_mlp": 0.01060676,
      "balance_loss_clip": 1.07473838,
      "balance_loss_mlp": 1.03225648,
      "epoch": 0.046655644070344206,
      "flos": 18112516778880.0,
      "grad_norm": 1.8466338060532805,
      "language_loss": 0.77191961,
      "learning_rate": 3.997091134020217e-06,
      "loss": 0.79496562,
      "num_input_tokens_seen": 16667555,
      "step": 776,
      "time_per_iteration": 2.4902849197387695
    },
    {
      "auxiliary_loss_clip": 0.01238718,
      "auxiliary_loss_mlp": 0.01070349,
      "balance_loss_clip": 1.07383513,
      "balance_loss_mlp": 1.04366958,
      "epoch": 0.04671576732301218,
      "flos": 29205286617600.0,
      "grad_norm": 2.093483963289726,
      "language_loss": 0.71109039,
      "learning_rate": 3.997070098605585e-06,
      "loss": 0.73418105,
      "num_input_tokens_seen": 16686875,
      "step": 777,
      "time_per_iteration": 2.647341728210449
    },
    {
      "auxiliary_loss_clip": 0.01243771,
      "auxiliary_loss_mlp": 0.01079351,
      "balance_loss_clip": 1.07673287,
      "balance_loss_mlp": 1.05002534,
      "epoch": 0.04677589057568014,
      "flos": 30478319510400.0,
      "grad_norm": 1.9743188650380956,
      "language_loss": 0.76996791,
      "learning_rate": 3.997048987461856e-06,
      "loss": 0.79319906,
      "num_input_tokens_seen": 16706420,
      "step": 778,
      "time_per_iteration": 2.646841526031494
    },
    {
      "auxiliary_loss_clip": 0.01241011,
      "auxiliary_loss_mlp": 0.0106939,
      "balance_loss_clip": 1.07557082,
      "balance_loss_mlp": 1.04001689,
      "epoch": 0.046836013828348115,
      "flos": 20557674282240.0,
      "grad_norm": 2.005052451336899,
      "language_loss": 0.79054034,
      "learning_rate": 3.997027800589829e-06,
      "loss": 0.81364441,
      "num_input_tokens_seen": 16726390,
      "step": 779,
      "time_per_iteration": 2.565671443939209
    },
    {
      "auxiliary_loss_clip": 0.01231719,
      "auxiliary_loss_mlp": 0.01067857,
      "balance_loss_clip": 1.07208145,
      "balance_loss_mlp": 1.04123771,
      "epoch": 0.04689613708101608,
      "flos": 25447378757760.0,
      "grad_norm": 1.6658308374977617,
      "language_loss": 0.77407861,
      "learning_rate": 3.997006537990308e-06,
      "loss": 0.79707432,
      "num_input_tokens_seen": 16748965,
      "step": 780,
      "time_per_iteration": 2.643284320831299
    },
    {
      "auxiliary_loss_clip": 0.01238833,
      "auxiliary_loss_mlp": 0.0107527,
      "balance_loss_clip": 1.0750674,
      "balance_loss_mlp": 1.04891324,
      "epoch": 0.04695626033368405,
      "flos": 23001395241600.0,
      "grad_norm": 1.7802463794733479,
      "language_loss": 0.76502752,
      "learning_rate": 3.996985199664099e-06,
      "loss": 0.78816861,
      "num_input_tokens_seen": 16768620,
      "step": 781,
      "time_per_iteration": 2.5727956295013428
    },
    {
      "auxiliary_loss_clip": 0.01249443,
      "auxiliary_loss_mlp": 0.01073532,
      "balance_loss_clip": 1.07888877,
      "balance_loss_mlp": 1.04477882,
      "epoch": 0.047016383586352024,
      "flos": 29133357632640.0,
      "grad_norm": 2.67820962447853,
      "language_loss": 0.73549056,
      "learning_rate": 3.99696378561201e-06,
      "loss": 0.75872028,
      "num_input_tokens_seen": 16789755,
      "step": 782,
      "time_per_iteration": 2.6457245349884033
    },
    {
      "auxiliary_loss_clip": 0.01242569,
      "auxiliary_loss_mlp": 0.01069719,
      "balance_loss_clip": 1.07818317,
      "balance_loss_mlp": 1.04343307,
      "epoch": 0.04707650683901999,
      "flos": 14976330451200.0,
      "grad_norm": 2.5240838310153864,
      "language_loss": 0.80276,
      "learning_rate": 3.996942295834855e-06,
      "loss": 0.82588285,
      "num_input_tokens_seen": 16807585,
      "step": 783,
      "time_per_iteration": 2.5471854209899902
    },
    {
      "auxiliary_loss_clip": 0.01235665,
      "auxiliary_loss_mlp": 0.01064256,
      "balance_loss_clip": 1.07603633,
      "balance_loss_mlp": 1.03834033,
      "epoch": 0.04713663009168796,
      "flos": 21651118151040.0,
      "grad_norm": 2.326109942048762,
      "language_loss": 0.81375659,
      "learning_rate": 3.996920730333448e-06,
      "loss": 0.83675581,
      "num_input_tokens_seen": 16827220,
      "step": 784,
      "time_per_iteration": 2.561842679977417
    },
    {
      "auxiliary_loss_clip": 0.01241403,
      "auxiliary_loss_mlp": 0.01068954,
      "balance_loss_clip": 1.07363915,
      "balance_loss_mlp": 1.04285896,
      "epoch": 0.04719675334435593,
      "flos": 21325408600320.0,
      "grad_norm": 3.200522746687193,
      "language_loss": 0.80228728,
      "learning_rate": 3.996899089108607e-06,
      "loss": 0.82539082,
      "num_input_tokens_seen": 16846230,
      "step": 785,
      "time_per_iteration": 2.4844813346862793
    },
    {
      "auxiliary_loss_clip": 0.01242893,
      "auxiliary_loss_mlp": 0.01063234,
      "balance_loss_clip": 1.07928646,
      "balance_loss_mlp": 1.0381999,
      "epoch": 0.0472568765970239,
      "flos": 17931383470080.0,
      "grad_norm": 2.186008494275165,
      "language_loss": 0.89448529,
      "learning_rate": 3.996877372161152e-06,
      "loss": 0.91754657,
      "num_input_tokens_seen": 16865325,
      "step": 786,
      "time_per_iteration": 2.4741830825805664
    },
    {
      "auxiliary_loss_clip": 0.01241003,
      "auxiliary_loss_mlp": 0.01068337,
      "balance_loss_clip": 1.06833756,
      "balance_loss_mlp": 1.03915501,
      "epoch": 0.04731699984969187,
      "flos": 18077324428800.0,
      "grad_norm": 2.2738306717570125,
      "language_loss": 0.76539093,
      "learning_rate": 3.9968555794919065e-06,
      "loss": 0.78848433,
      "num_input_tokens_seen": 16882930,
      "step": 787,
      "time_per_iteration": 2.448213577270508
    },
    {
      "auxiliary_loss_clip": 0.01250633,
      "auxiliary_loss_mlp": 0.01068563,
      "balance_loss_clip": 1.0820123,
      "balance_loss_mlp": 1.0407393,
      "epoch": 0.047377123102359836,
      "flos": 23185078416000.0,
      "grad_norm": 2.7142084372379203,
      "language_loss": 0.81182063,
      "learning_rate": 3.996833711101698e-06,
      "loss": 0.83501261,
      "num_input_tokens_seen": 16900710,
      "step": 788,
      "time_per_iteration": 2.5112040042877197
    },
    {
      "auxiliary_loss_clip": 0.01242586,
      "auxiliary_loss_mlp": 0.01078506,
      "balance_loss_clip": 1.07943511,
      "balance_loss_mlp": 1.04953814,
      "epoch": 0.04743724635502781,
      "flos": 22747794243840.0,
      "grad_norm": 1.877253285300506,
      "language_loss": 0.84504247,
      "learning_rate": 3.996811766991355e-06,
      "loss": 0.86825335,
      "num_input_tokens_seen": 16919210,
      "step": 789,
      "time_per_iteration": 2.568422317504883
    },
    {
      "auxiliary_loss_clip": 0.0124699,
      "auxiliary_loss_mlp": 0.01071354,
      "balance_loss_clip": 1.07941151,
      "balance_loss_mlp": 1.04455578,
      "epoch": 0.04749736960769577,
      "flos": 17238702620160.0,
      "grad_norm": 2.3772834910979395,
      "language_loss": 0.81960428,
      "learning_rate": 3.996789747161709e-06,
      "loss": 0.8427878,
      "num_input_tokens_seen": 16937125,
      "step": 790,
      "time_per_iteration": 2.533909559249878
    },
    {
      "auxiliary_loss_clip": 0.01240714,
      "auxiliary_loss_mlp": 0.01064129,
      "balance_loss_clip": 1.07476413,
      "balance_loss_mlp": 1.03594792,
      "epoch": 0.047557492860363745,
      "flos": 40479261592320.0,
      "grad_norm": 2.057979567772816,
      "language_loss": 0.88179904,
      "learning_rate": 3.996767651613597e-06,
      "loss": 0.9048475,
      "num_input_tokens_seen": 16958610,
      "step": 791,
      "time_per_iteration": 2.70927357673645
    },
    {
      "auxiliary_loss_clip": 0.01242579,
      "auxiliary_loss_mlp": 0.01063953,
      "balance_loss_clip": 1.07735348,
      "balance_loss_mlp": 1.0357604,
      "epoch": 0.04761761611303172,
      "flos": 18698004466560.0,
      "grad_norm": 2.6759705848844186,
      "language_loss": 0.90337658,
      "learning_rate": 3.996745480347854e-06,
      "loss": 0.92644191,
      "num_input_tokens_seen": 16977300,
      "step": 792,
      "time_per_iteration": 2.5280330181121826
    },
    {
      "auxiliary_loss_clip": 0.01244433,
      "auxiliary_loss_mlp": 0.0107635,
      "balance_loss_clip": 1.07612813,
      "balance_loss_mlp": 1.04950356,
      "epoch": 0.04767773936569968,
      "flos": 20921987975040.0,
      "grad_norm": 1.979640043812563,
      "language_loss": 0.73490667,
      "learning_rate": 3.996723233365324e-06,
      "loss": 0.75811452,
      "num_input_tokens_seen": 16994950,
      "step": 793,
      "time_per_iteration": 2.4967477321624756
    },
    {
      "auxiliary_loss_clip": 0.01251229,
      "auxiliary_loss_mlp": 0.01067985,
      "balance_loss_clip": 1.08046937,
      "balance_loss_mlp": 1.03958917,
      "epoch": 0.047737862618367655,
      "flos": 23732680233600.0,
      "grad_norm": 1.9734411627117254,
      "language_loss": 0.86496353,
      "learning_rate": 3.996700910666847e-06,
      "loss": 0.88815564,
      "num_input_tokens_seen": 17014760,
      "step": 794,
      "time_per_iteration": 2.541639804840088
    },
    {
      "auxiliary_loss_clip": 0.01245251,
      "auxiliary_loss_mlp": 0.01078193,
      "balance_loss_clip": 1.07593322,
      "balance_loss_mlp": 1.04908252,
      "epoch": 0.04779798587103562,
      "flos": 23695764030720.0,
      "grad_norm": 35.03331845758941,
      "language_loss": 0.70094812,
      "learning_rate": 3.996678512253272e-06,
      "loss": 0.72418255,
      "num_input_tokens_seen": 17032715,
      "step": 795,
      "time_per_iteration": 2.5224030017852783
    },
    {
      "auxiliary_loss_clip": 0.01244542,
      "auxiliary_loss_mlp": 0.01076427,
      "balance_loss_clip": 1.07734275,
      "balance_loss_mlp": 1.04797149,
      "epoch": 0.04785810912370359,
      "flos": 23183641872000.0,
      "grad_norm": 1.7880897926234682,
      "language_loss": 0.80784398,
      "learning_rate": 3.996656038125449e-06,
      "loss": 0.83105361,
      "num_input_tokens_seen": 17052215,
      "step": 796,
      "time_per_iteration": 2.4957098960876465
    },
    {
      "auxiliary_loss_clip": 0.01245762,
      "auxiliary_loss_mlp": 0.01063612,
      "balance_loss_clip": 1.07751513,
      "balance_loss_mlp": 1.03538287,
      "epoch": 0.047918232376371564,
      "flos": 18040623707520.0,
      "grad_norm": 3.0904967361896576,
      "language_loss": 0.81703985,
      "learning_rate": 3.996633488284228e-06,
      "loss": 0.84013361,
      "num_input_tokens_seen": 17069225,
      "step": 797,
      "time_per_iteration": 2.451547861099243
    },
    {
      "auxiliary_loss_clip": 0.01116969,
      "auxiliary_loss_mlp": 0.01009572,
      "balance_loss_clip": 1.0456568,
      "balance_loss_mlp": 1.0050658,
      "epoch": 0.04797835562903953,
      "flos": 62442588758400.0,
      "grad_norm": 0.9161351180304066,
      "language_loss": 0.64460701,
      "learning_rate": 3.996610862730465e-06,
      "loss": 0.6658724,
      "num_input_tokens_seen": 17126680,
      "step": 798,
      "time_per_iteration": 3.0242950916290283
    },
    {
      "auxiliary_loss_clip": 0.01248157,
      "auxiliary_loss_mlp": 0.01071013,
      "balance_loss_clip": 1.07410622,
      "balance_loss_mlp": 1.0438807,
      "epoch": 0.0480384788817075,
      "flos": 21507296094720.0,
      "grad_norm": 1.851269406555387,
      "language_loss": 0.91196877,
      "learning_rate": 3.996588161465018e-06,
      "loss": 0.93516046,
      "num_input_tokens_seen": 17144835,
      "step": 799,
      "time_per_iteration": 2.5169241428375244
    },
    {
      "auxiliary_loss_clip": 0.01245512,
      "auxiliary_loss_mlp": 0.01075066,
      "balance_loss_clip": 1.07970643,
      "balance_loss_mlp": 1.04621708,
      "epoch": 0.048098602134375466,
      "flos": 21726710323200.0,
      "grad_norm": 2.2827408794087605,
      "language_loss": 0.86596709,
      "learning_rate": 3.996565384488748e-06,
      "loss": 0.88917285,
      "num_input_tokens_seen": 17165030,
      "step": 800,
      "time_per_iteration": 2.4946484565734863
    },
    {
      "auxiliary_loss_clip": 0.01246489,
      "auxiliary_loss_mlp": 0.01075286,
      "balance_loss_clip": 1.07660556,
      "balance_loss_mlp": 1.04871416,
      "epoch": 0.04815872538704344,
      "flos": 22931082368640.0,
      "grad_norm": 3.458195410231451,
      "language_loss": 0.83910966,
      "learning_rate": 3.996542531802518e-06,
      "loss": 0.86232746,
      "num_input_tokens_seen": 17184895,
      "step": 801,
      "time_per_iteration": 2.5180957317352295
    },
    {
      "auxiliary_loss_clip": 0.012468,
      "auxiliary_loss_mlp": 0.01072515,
      "balance_loss_clip": 1.07785225,
      "balance_loss_mlp": 1.04501331,
      "epoch": 0.04821884863971141,
      "flos": 43174716042240.0,
      "grad_norm": 2.3504046974825106,
      "language_loss": 0.79904783,
      "learning_rate": 3.996519603407196e-06,
      "loss": 0.82224095,
      "num_input_tokens_seen": 17208225,
      "step": 802,
      "time_per_iteration": 2.664400815963745
    },
    {
      "auxiliary_loss_clip": 0.01244724,
      "auxiliary_loss_mlp": 0.01069605,
      "balance_loss_clip": 1.07769346,
      "balance_loss_mlp": 1.04253221,
      "epoch": 0.048278971892379376,
      "flos": 18620006083200.0,
      "grad_norm": 1.9975675206562173,
      "language_loss": 0.86513817,
      "learning_rate": 3.996496599303649e-06,
      "loss": 0.88828146,
      "num_input_tokens_seen": 17226305,
      "step": 803,
      "time_per_iteration": 3.847806215286255
    },
    {
      "auxiliary_loss_clip": 0.01239669,
      "auxiliary_loss_mlp": 0.01064092,
      "balance_loss_clip": 1.07638907,
      "balance_loss_mlp": 1.03712749,
      "epoch": 0.04833909514504735,
      "flos": 20230061310720.0,
      "grad_norm": 3.3957759857752925,
      "language_loss": 0.85617793,
      "learning_rate": 3.996473519492753e-06,
      "loss": 0.8792156,
      "num_input_tokens_seen": 17244545,
      "step": 804,
      "time_per_iteration": 3.9010961055755615
    },
    {
      "auxiliary_loss_clip": 0.01243587,
      "auxiliary_loss_mlp": 0.01073093,
      "balance_loss_clip": 1.07667589,
      "balance_loss_mlp": 1.04588962,
      "epoch": 0.04839921839771532,
      "flos": 24645170361600.0,
      "grad_norm": 3.9632404868175786,
      "language_loss": 0.86168587,
      "learning_rate": 3.99645036397538e-06,
      "loss": 0.88485265,
      "num_input_tokens_seen": 17265730,
      "step": 805,
      "time_per_iteration": 2.493422508239746
    },
    {
      "auxiliary_loss_clip": 0.01237,
      "auxiliary_loss_mlp": 0.01068786,
      "balance_loss_clip": 1.07276845,
      "balance_loss_mlp": 1.04251266,
      "epoch": 0.048459341650383285,
      "flos": 24827452905600.0,
      "grad_norm": 2.0428492173876864,
      "language_loss": 0.68074137,
      "learning_rate": 3.9964271327524085e-06,
      "loss": 0.70379931,
      "num_input_tokens_seen": 17284820,
      "step": 806,
      "time_per_iteration": 2.5187482833862305
    },
    {
      "auxiliary_loss_clip": 0.01238216,
      "auxiliary_loss_mlp": 0.01062313,
      "balance_loss_clip": 1.07449675,
      "balance_loss_mlp": 1.03463221,
      "epoch": 0.04851946490305126,
      "flos": 22163204396160.0,
      "grad_norm": 1.9891827639438235,
      "language_loss": 0.77223712,
      "learning_rate": 3.9964038258247214e-06,
      "loss": 0.79524237,
      "num_input_tokens_seen": 17305085,
      "step": 807,
      "time_per_iteration": 2.5002756118774414
    },
    {
      "auxiliary_loss_clip": 0.01233753,
      "auxiliary_loss_mlp": 0.01073023,
      "balance_loss_clip": 1.07018268,
      "balance_loss_mlp": 1.04611778,
      "epoch": 0.04857958815571922,
      "flos": 19792022952960.0,
      "grad_norm": 2.372314994282698,
      "language_loss": 0.86724633,
      "learning_rate": 3.9963804431932005e-06,
      "loss": 0.89031404,
      "num_input_tokens_seen": 17322715,
      "step": 808,
      "time_per_iteration": 3.8439013957977295
    },
    {
      "auxiliary_loss_clip": 0.01244054,
      "auxiliary_loss_mlp": 0.01069863,
      "balance_loss_clip": 1.07447898,
      "balance_loss_mlp": 1.04224205,
      "epoch": 0.048639711408387194,
      "flos": 18697968552960.0,
      "grad_norm": 1.9519769957507593,
      "language_loss": 0.89985967,
      "learning_rate": 3.996356984858732e-06,
      "loss": 0.92299885,
      "num_input_tokens_seen": 17341455,
      "step": 809,
      "time_per_iteration": 2.5018386840820312
    },
    {
      "auxiliary_loss_clip": 0.01243507,
      "auxiliary_loss_mlp": 0.01073199,
      "balance_loss_clip": 1.07807541,
      "balance_loss_mlp": 1.04556584,
      "epoch": 0.048699834661055166,
      "flos": 24863507182080.0,
      "grad_norm": 2.0426929709655415,
      "language_loss": 0.84630668,
      "learning_rate": 3.996333450822208e-06,
      "loss": 0.8694737,
      "num_input_tokens_seen": 17360765,
      "step": 810,
      "time_per_iteration": 3.883864641189575
    },
    {
      "auxiliary_loss_clip": 0.01247299,
      "auxiliary_loss_mlp": 0.01069101,
      "balance_loss_clip": 1.07698536,
      "balance_loss_mlp": 1.04195702,
      "epoch": 0.04875995791372313,
      "flos": 20704010290560.0,
      "grad_norm": 1.6850177970647835,
      "language_loss": 0.80617166,
      "learning_rate": 3.99630984108452e-06,
      "loss": 0.82933569,
      "num_input_tokens_seen": 17380625,
      "step": 811,
      "time_per_iteration": 2.5062718391418457
    },
    {
      "auxiliary_loss_clip": 0.01234437,
      "auxiliary_loss_mlp": 0.01075688,
      "balance_loss_clip": 1.07247007,
      "balance_loss_mlp": 1.04890203,
      "epoch": 0.048820081166391104,
      "flos": 18588297352320.0,
      "grad_norm": 1.8283325809864828,
      "language_loss": 0.74756324,
      "learning_rate": 3.9962861556465615e-06,
      "loss": 0.77066457,
      "num_input_tokens_seen": 17399355,
      "step": 812,
      "time_per_iteration": 2.5213494300842285
    },
    {
      "auxiliary_loss_clip": 0.01239913,
      "auxiliary_loss_mlp": 0.0108147,
      "balance_loss_clip": 1.07801151,
      "balance_loss_mlp": 1.05531573,
      "epoch": 0.04888020441905907,
      "flos": 22707322594560.0,
      "grad_norm": 2.133409844960452,
      "language_loss": 0.90314323,
      "learning_rate": 3.996262394509233e-06,
      "loss": 0.92635709,
      "num_input_tokens_seen": 17418240,
      "step": 813,
      "time_per_iteration": 2.5120720863342285
    },
    {
      "auxiliary_loss_clip": 0.01235485,
      "auxiliary_loss_mlp": 0.01060716,
      "balance_loss_clip": 1.07429349,
      "balance_loss_mlp": 1.03528905,
      "epoch": 0.04894032767172704,
      "flos": 22784351310720.0,
      "grad_norm": 2.1912719187944285,
      "language_loss": 0.75002581,
      "learning_rate": 3.9962385576734335e-06,
      "loss": 0.77298784,
      "num_input_tokens_seen": 17436250,
      "step": 814,
      "time_per_iteration": 2.479597568511963
    },
    {
      "auxiliary_loss_clip": 0.01238339,
      "auxiliary_loss_mlp": 0.01073285,
      "balance_loss_clip": 1.07385266,
      "balance_loss_mlp": 1.04597437,
      "epoch": 0.04900045092439501,
      "flos": 25516147345920.0,
      "grad_norm": 1.9812739330943436,
      "language_loss": 0.83839464,
      "learning_rate": 3.9962146451400675e-06,
      "loss": 0.86151087,
      "num_input_tokens_seen": 17455750,
      "step": 815,
      "time_per_iteration": 2.545210599899292
    },
    {
      "auxiliary_loss_clip": 0.01242332,
      "auxiliary_loss_mlp": 0.01066371,
      "balance_loss_clip": 1.07592249,
      "balance_loss_mlp": 1.03902447,
      "epoch": 0.04906057417706298,
      "flos": 25958136199680.0,
      "grad_norm": 2.3654767481053676,
      "language_loss": 0.90904683,
      "learning_rate": 3.996190656910043e-06,
      "loss": 0.93213379,
      "num_input_tokens_seen": 17474995,
      "step": 816,
      "time_per_iteration": 2.5946240425109863
    },
    {
      "auxiliary_loss_clip": 0.01243095,
      "auxiliary_loss_mlp": 0.01063479,
      "balance_loss_clip": 1.07603812,
      "balance_loss_mlp": 1.0364306,
      "epoch": 0.04912069742973095,
      "flos": 18624638937600.0,
      "grad_norm": 2.4563360718106364,
      "language_loss": 0.80206913,
      "learning_rate": 3.996166592984268e-06,
      "loss": 0.82513487,
      "num_input_tokens_seen": 17493395,
      "step": 817,
      "time_per_iteration": 2.542229413986206
    },
    {
      "auxiliary_loss_clip": 0.0123988,
      "auxiliary_loss_mlp": 0.01076566,
      "balance_loss_clip": 1.07686591,
      "balance_loss_mlp": 1.05004203,
      "epoch": 0.049180820682398915,
      "flos": 23699786353920.0,
      "grad_norm": 1.8170935039504736,
      "language_loss": 0.84975958,
      "learning_rate": 3.996142453363656e-06,
      "loss": 0.87292403,
      "num_input_tokens_seen": 17514565,
      "step": 818,
      "time_per_iteration": 2.508615255355835
    },
    {
      "auxiliary_loss_clip": 0.01245111,
      "auxiliary_loss_mlp": 0.01068236,
      "balance_loss_clip": 1.07409656,
      "balance_loss_mlp": 1.04022145,
      "epoch": 0.04924094393506689,
      "flos": 22420396753920.0,
      "grad_norm": 3.2968567874784522,
      "language_loss": 0.75655162,
      "learning_rate": 3.996118238049124e-06,
      "loss": 0.77968508,
      "num_input_tokens_seen": 17534590,
      "step": 819,
      "time_per_iteration": 2.5291829109191895
    },
    {
      "auxiliary_loss_clip": 0.01241697,
      "auxiliary_loss_mlp": 0.01062928,
      "balance_loss_clip": 1.07796013,
      "balance_loss_mlp": 1.03834665,
      "epoch": 0.04930106718773486,
      "flos": 15738246766080.0,
      "grad_norm": 2.2047996342106155,
      "language_loss": 0.85105658,
      "learning_rate": 3.996093947041586e-06,
      "loss": 0.87410283,
      "num_input_tokens_seen": 17551900,
      "step": 820,
      "time_per_iteration": 2.456725597381592
    },
    {
      "auxiliary_loss_clip": 0.01240709,
      "auxiliary_loss_mlp": 0.01065732,
      "balance_loss_clip": 1.07420337,
      "balance_loss_mlp": 1.03892159,
      "epoch": 0.049361190440402825,
      "flos": 26250628648320.0,
      "grad_norm": 1.9313174443054077,
      "language_loss": 0.90514338,
      "learning_rate": 3.996069580341966e-06,
      "loss": 0.92820781,
      "num_input_tokens_seen": 17571485,
      "step": 821,
      "time_per_iteration": 2.552177667617798
    },
    {
      "auxiliary_loss_clip": 0.01237725,
      "auxiliary_loss_mlp": 0.01080742,
      "balance_loss_clip": 1.07364774,
      "balance_loss_mlp": 1.05432534,
      "epoch": 0.0494213136930708,
      "flos": 21252366293760.0,
      "grad_norm": 1.9457102468323255,
      "language_loss": 0.89642507,
      "learning_rate": 3.996045137951188e-06,
      "loss": 0.91960973,
      "num_input_tokens_seen": 17591410,
      "step": 822,
      "time_per_iteration": 2.4803619384765625
    },
    {
      "auxiliary_loss_clip": 0.0124025,
      "auxiliary_loss_mlp": 0.01063365,
      "balance_loss_clip": 1.07704973,
      "balance_loss_mlp": 1.03444457,
      "epoch": 0.04948143694573876,
      "flos": 27965506740480.0,
      "grad_norm": 1.6721570864429145,
      "language_loss": 0.67191553,
      "learning_rate": 3.996020619870178e-06,
      "loss": 0.69495165,
      "num_input_tokens_seen": 17612010,
      "step": 823,
      "time_per_iteration": 2.5521817207336426
    },
    {
      "auxiliary_loss_clip": 0.01113993,
      "auxiliary_loss_mlp": 0.01005757,
      "balance_loss_clip": 1.04316378,
      "balance_loss_mlp": 1.00120342,
      "epoch": 0.049541560198406734,
      "flos": 66180995533440.0,
      "grad_norm": 1.3597094069857931,
      "language_loss": 0.62263799,
      "learning_rate": 3.995996026099866e-06,
      "loss": 0.64383548,
      "num_input_tokens_seen": 17673430,
      "step": 824,
      "time_per_iteration": 3.1534554958343506
    },
    {
      "auxiliary_loss_clip": 0.01242674,
      "auxiliary_loss_mlp": 0.01072878,
      "balance_loss_clip": 1.07462335,
      "balance_loss_mlp": 1.04458928,
      "epoch": 0.049601683451074706,
      "flos": 22892693708160.0,
      "grad_norm": 1.8949680508776576,
      "language_loss": 0.90363204,
      "learning_rate": 3.995971356641185e-06,
      "loss": 0.92678756,
      "num_input_tokens_seen": 17689545,
      "step": 825,
      "time_per_iteration": 2.4887239933013916
    },
    {
      "auxiliary_loss_clip": 0.01240682,
      "auxiliary_loss_mlp": 0.01068896,
      "balance_loss_clip": 1.07500732,
      "balance_loss_mlp": 1.04081035,
      "epoch": 0.04966180670374267,
      "flos": 21433643256960.0,
      "grad_norm": 3.4454821184335183,
      "language_loss": 0.66797441,
      "learning_rate": 3.9959466114950695e-06,
      "loss": 0.6910702,
      "num_input_tokens_seen": 17705965,
      "step": 826,
      "time_per_iteration": 2.474905490875244
    },
    {
      "auxiliary_loss_clip": 0.0124177,
      "auxiliary_loss_mlp": 0.01067724,
      "balance_loss_clip": 1.07563317,
      "balance_loss_mlp": 1.04032993,
      "epoch": 0.04972192995641064,
      "flos": 23107367341440.0,
      "grad_norm": 2.6053337035041992,
      "language_loss": 0.78375435,
      "learning_rate": 3.995921790662459e-06,
      "loss": 0.8068493,
      "num_input_tokens_seen": 17724580,
      "step": 827,
      "time_per_iteration": 2.5347914695739746
    },
    {
      "auxiliary_loss_clip": 0.01242948,
      "auxiliary_loss_mlp": 0.01079771,
      "balance_loss_clip": 1.07559633,
      "balance_loss_mlp": 1.05137515,
      "epoch": 0.04978205320907861,
      "flos": 40406147458560.0,
      "grad_norm": 2.9075729392404845,
      "language_loss": 0.78656793,
      "learning_rate": 3.995896894144294e-06,
      "loss": 0.80979514,
      "num_input_tokens_seen": 17747755,
      "step": 828,
      "time_per_iteration": 2.669144868850708
    },
    {
      "auxiliary_loss_clip": 0.01234163,
      "auxiliary_loss_mlp": 0.01059369,
      "balance_loss_clip": 1.07293701,
      "balance_loss_mlp": 1.03341699,
      "epoch": 0.04984217646174658,
      "flos": 25228539146880.0,
      "grad_norm": 1.801032769760919,
      "language_loss": 0.83437985,
      "learning_rate": 3.995871921941519e-06,
      "loss": 0.85731518,
      "num_input_tokens_seen": 17768550,
      "step": 829,
      "time_per_iteration": 2.541841745376587
    },
    {
      "auxiliary_loss_clip": 0.01238873,
      "auxiliary_loss_mlp": 0.01082721,
      "balance_loss_clip": 1.07341337,
      "balance_loss_mlp": 1.05266857,
      "epoch": 0.04990229971441455,
      "flos": 15959636242560.0,
      "grad_norm": 2.379943050940154,
      "language_loss": 0.75488675,
      "learning_rate": 3.99584687405508e-06,
      "loss": 0.77810264,
      "num_input_tokens_seen": 17786080,
      "step": 830,
      "time_per_iteration": 2.5370733737945557
    },
    {
      "auxiliary_loss_clip": 0.01240261,
      "auxiliary_loss_mlp": 0.0107173,
      "balance_loss_clip": 1.07410598,
      "balance_loss_mlp": 1.04359674,
      "epoch": 0.04996242296708252,
      "flos": 18405116968320.0,
      "grad_norm": 1.986546773729812,
      "language_loss": 0.79863107,
      "learning_rate": 3.995821750485929e-06,
      "loss": 0.821751,
      "num_input_tokens_seen": 17803635,
      "step": 831,
      "time_per_iteration": 2.4857072830200195
    },
    {
      "auxiliary_loss_clip": 0.01196039,
      "auxiliary_loss_mlp": 0.01076753,
      "balance_loss_clip": 1.08537984,
      "balance_loss_mlp": 1.05020475,
      "epoch": 0.05002254621975049,
      "flos": 17858053854720.0,
      "grad_norm": 2.2083579232609147,
      "language_loss": 0.91245604,
      "learning_rate": 3.995796551235016e-06,
      "loss": 0.93518388,
      "num_input_tokens_seen": 17822190,
      "step": 832,
      "time_per_iteration": 2.6722090244293213
    },
    {
      "auxiliary_loss_clip": 0.01208011,
      "auxiliary_loss_mlp": 0.01079697,
      "balance_loss_clip": 1.072855,
      "balance_loss_mlp": 1.05417418,
      "epoch": 0.050082669472418455,
      "flos": 45660273367680.0,
      "grad_norm": 1.9986415376887188,
      "language_loss": 0.83324468,
      "learning_rate": 3.9957712763032974e-06,
      "loss": 0.85612178,
      "num_input_tokens_seen": 17846915,
      "step": 833,
      "time_per_iteration": 2.7380354404449463
    },
    {
      "auxiliary_loss_clip": 0.01212716,
      "auxiliary_loss_mlp": 0.01062455,
      "balance_loss_clip": 1.0718925,
      "balance_loss_mlp": 1.03475058,
      "epoch": 0.05014279272508643,
      "flos": 37962067363200.0,
      "grad_norm": 2.1953223908704627,
      "language_loss": 0.8219378,
      "learning_rate": 3.995745925691733e-06,
      "loss": 0.84468949,
      "num_input_tokens_seen": 17867270,
      "step": 834,
      "time_per_iteration": 2.7344141006469727
    },
    {
      "auxiliary_loss_clip": 0.01229672,
      "auxiliary_loss_mlp": 0.01065577,
      "balance_loss_clip": 1.07608032,
      "balance_loss_mlp": 1.03726506,
      "epoch": 0.0502029159777544,
      "flos": 20996179516800.0,
      "grad_norm": 2.126276246460614,
      "language_loss": 0.91635203,
      "learning_rate": 3.995720499401282e-06,
      "loss": 0.93930453,
      "num_input_tokens_seen": 17884880,
      "step": 835,
      "time_per_iteration": 2.5124807357788086
    },
    {
      "auxiliary_loss_clip": 0.01241332,
      "auxiliary_loss_mlp": 0.01066821,
      "balance_loss_clip": 1.07261693,
      "balance_loss_mlp": 1.03865206,
      "epoch": 0.050263039230422364,
      "flos": 15888066393600.0,
      "grad_norm": 1.9954982039182585,
      "language_loss": 0.76064563,
      "learning_rate": 3.995694997432911e-06,
      "loss": 0.78372711,
      "num_input_tokens_seen": 17903695,
      "step": 836,
      "time_per_iteration": 2.4958484172821045
    },
    {
      "auxiliary_loss_clip": 0.01223456,
      "auxiliary_loss_mlp": 0.01071503,
      "balance_loss_clip": 1.07376409,
      "balance_loss_mlp": 1.04534841,
      "epoch": 0.050323162483090336,
      "flos": 23732752060800.0,
      "grad_norm": 2.1861736002081855,
      "language_loss": 0.83395815,
      "learning_rate": 3.9956694197875855e-06,
      "loss": 0.85690773,
      "num_input_tokens_seen": 17920745,
      "step": 837,
      "time_per_iteration": 2.535252332687378
    },
    {
      "auxiliary_loss_clip": 0.01210587,
      "auxiliary_loss_mlp": 0.00827203,
      "balance_loss_clip": 1.07563329,
      "balance_loss_mlp": 1.07636249,
      "epoch": 0.0503832857357583,
      "flos": 20266223328000.0,
      "grad_norm": 3.1550277814872705,
      "language_loss": 0.73076892,
      "learning_rate": 3.995643766466275e-06,
      "loss": 0.75114679,
      "num_input_tokens_seen": 17938220,
      "step": 838,
      "time_per_iteration": 2.5712361335754395
    },
    {
      "auxiliary_loss_clip": 0.01203015,
      "auxiliary_loss_mlp": 0.01073303,
      "balance_loss_clip": 1.06845963,
      "balance_loss_mlp": 1.04625463,
      "epoch": 0.05044340898842627,
      "flos": 17785011548160.0,
      "grad_norm": 1.7153880318213135,
      "language_loss": 0.83041525,
      "learning_rate": 3.995618037469953e-06,
      "loss": 0.85317838,
      "num_input_tokens_seen": 17957325,
      "step": 839,
      "time_per_iteration": 2.510937452316284
    },
    {
      "auxiliary_loss_clip": 0.01234295,
      "auxiliary_loss_mlp": 0.01070547,
      "balance_loss_clip": 1.07371879,
      "balance_loss_mlp": 1.04411781,
      "epoch": 0.050503532241094246,
      "flos": 22966526113920.0,
      "grad_norm": 2.0404168296533074,
      "language_loss": 0.85622895,
      "learning_rate": 3.995592232799595e-06,
      "loss": 0.87927735,
      "num_input_tokens_seen": 17975875,
      "step": 840,
      "time_per_iteration": 2.508007764816284
    },
    {
      "auxiliary_loss_clip": 0.01215873,
      "auxiliary_loss_mlp": 0.01062745,
      "balance_loss_clip": 1.08593535,
      "balance_loss_mlp": 1.03384829,
      "epoch": 0.05056365549376221,
      "flos": 22776989022720.0,
      "grad_norm": 2.9750553662566777,
      "language_loss": 0.94513535,
      "learning_rate": 3.99556635245618e-06,
      "loss": 0.9679215,
      "num_input_tokens_seen": 17994340,
      "step": 841,
      "time_per_iteration": 2.5414319038391113
    },
    {
      "auxiliary_loss_clip": 0.01240704,
      "auxiliary_loss_mlp": 0.01077376,
      "balance_loss_clip": 1.07565808,
      "balance_loss_mlp": 1.0493381,
      "epoch": 0.05062377874643018,
      "flos": 30916968399360.0,
      "grad_norm": 2.523214183916328,
      "language_loss": 0.77575576,
      "learning_rate": 3.995540396440688e-06,
      "loss": 0.79893655,
      "num_input_tokens_seen": 18015260,
      "step": 842,
      "time_per_iteration": 3.9567034244537354
    },
    {
      "auxiliary_loss_clip": 0.01227883,
      "auxiliary_loss_mlp": 0.01072986,
      "balance_loss_clip": 1.07589078,
      "balance_loss_mlp": 1.04580653,
      "epoch": 0.05068390199909815,
      "flos": 19647159402240.0,
      "grad_norm": 2.0162273758978047,
      "language_loss": 0.78336877,
      "learning_rate": 3.995514364754105e-06,
      "loss": 0.80637741,
      "num_input_tokens_seen": 18033960,
      "step": 843,
      "time_per_iteration": 3.8950881958007812
    },
    {
      "auxiliary_loss_clip": 0.01228292,
      "auxiliary_loss_mlp": 0.01066261,
      "balance_loss_clip": 1.07687461,
      "balance_loss_mlp": 1.04030919,
      "epoch": 0.05074402525176612,
      "flos": 37962103276800.0,
      "grad_norm": 1.792246861780705,
      "language_loss": 0.83359987,
      "learning_rate": 3.995488257397417e-06,
      "loss": 0.85654533,
      "num_input_tokens_seen": 18056700,
      "step": 844,
      "time_per_iteration": 2.66231107711792
    },
    {
      "auxiliary_loss_clip": 0.01226355,
      "auxiliary_loss_mlp": 0.0106437,
      "balance_loss_clip": 1.07186198,
      "balance_loss_mlp": 1.03758371,
      "epoch": 0.05080414850443409,
      "flos": 22054610603520.0,
      "grad_norm": 2.155546461550287,
      "language_loss": 0.76555032,
      "learning_rate": 3.995462074371614e-06,
      "loss": 0.78845751,
      "num_input_tokens_seen": 18075815,
      "step": 845,
      "time_per_iteration": 2.578859567642212
    },
    {
      "auxiliary_loss_clip": 0.01217867,
      "auxiliary_loss_mlp": 0.01072954,
      "balance_loss_clip": 1.06965065,
      "balance_loss_mlp": 1.04511893,
      "epoch": 0.05086427175710206,
      "flos": 20225787592320.0,
      "grad_norm": 1.867128996955264,
      "language_loss": 0.87666845,
      "learning_rate": 3.99543581567769e-06,
      "loss": 0.89957666,
      "num_input_tokens_seen": 18095095,
      "step": 846,
      "time_per_iteration": 2.514352321624756
    },
    {
      "auxiliary_loss_clip": 0.01218399,
      "auxiliary_loss_mlp": 0.01070723,
      "balance_loss_clip": 1.07218754,
      "balance_loss_mlp": 1.04388928,
      "epoch": 0.05092439500977003,
      "flos": 15159223526400.0,
      "grad_norm": 1.8377556550718166,
      "language_loss": 0.87795925,
      "learning_rate": 3.9954094813166394e-06,
      "loss": 0.90085053,
      "num_input_tokens_seen": 18112675,
      "step": 847,
      "time_per_iteration": 3.9573960304260254
    },
    {
      "auxiliary_loss_clip": 0.01184846,
      "auxiliary_loss_mlp": 0.01068434,
      "balance_loss_clip": 1.07143998,
      "balance_loss_mlp": 1.04092073,
      "epoch": 0.050984518262437994,
      "flos": 22055149307520.0,
      "grad_norm": 2.628172156180215,
      "language_loss": 0.82061779,
      "learning_rate": 3.995383071289462e-06,
      "loss": 0.84315062,
      "num_input_tokens_seen": 18130745,
      "step": 848,
      "time_per_iteration": 2.6481239795684814
    },
    {
      "auxiliary_loss_clip": 0.01237024,
      "auxiliary_loss_mlp": 0.01078139,
      "balance_loss_clip": 1.07553422,
      "balance_loss_mlp": 1.05095935,
      "epoch": 0.05104464151510597,
      "flos": 30225329043840.0,
      "grad_norm": 2.6271318484109183,
      "language_loss": 0.87401921,
      "learning_rate": 3.995356585597158e-06,
      "loss": 0.8971709,
      "num_input_tokens_seen": 18152410,
      "step": 849,
      "time_per_iteration": 3.9041121006011963
    },
    {
      "auxiliary_loss_clip": 0.01232527,
      "auxiliary_loss_mlp": 0.01065631,
      "balance_loss_clip": 1.07164991,
      "balance_loss_mlp": 1.03917861,
      "epoch": 0.05110476476777394,
      "flos": 18332900674560.0,
      "grad_norm": 2.2562986068911015,
      "language_loss": 0.83392179,
      "learning_rate": 3.995330024240732e-06,
      "loss": 0.85690331,
      "num_input_tokens_seen": 18170870,
      "step": 850,
      "time_per_iteration": 2.4967684745788574
    },
    {
      "auxiliary_loss_clip": 0.01222735,
      "auxiliary_loss_mlp": 0.01064767,
      "balance_loss_clip": 1.07124782,
      "balance_loss_mlp": 1.03839827,
      "epoch": 0.051164888020441904,
      "flos": 37998732170880.0,
      "grad_norm": 2.8399936564895736,
      "language_loss": 0.6495626,
      "learning_rate": 3.995303387221192e-06,
      "loss": 0.67243755,
      "num_input_tokens_seen": 18191555,
      "step": 851,
      "time_per_iteration": 2.6590030193328857
    },
    {
      "auxiliary_loss_clip": 0.01221832,
      "auxiliary_loss_mlp": 0.01079729,
      "balance_loss_clip": 1.07160592,
      "balance_loss_mlp": 1.05023658,
      "epoch": 0.051225011273109876,
      "flos": 23038634666880.0,
      "grad_norm": 2.2558170343702675,
      "language_loss": 0.83235192,
      "learning_rate": 3.995276674539547e-06,
      "loss": 0.85536754,
      "num_input_tokens_seen": 18208620,
      "step": 852,
      "time_per_iteration": 2.5351204872131348
    },
    {
      "auxiliary_loss_clip": 0.01209264,
      "auxiliary_loss_mlp": 0.01076837,
      "balance_loss_clip": 1.07325673,
      "balance_loss_mlp": 1.04915655,
      "epoch": 0.05128513452577785,
      "flos": 18259822454400.0,
      "grad_norm": 1.9365288850717792,
      "language_loss": 0.80546242,
      "learning_rate": 3.995249886196811e-06,
      "loss": 0.82832342,
      "num_input_tokens_seen": 18226370,
      "step": 853,
      "time_per_iteration": 2.5126612186431885
    },
    {
      "auxiliary_loss_clip": 0.01234795,
      "auxiliary_loss_mlp": 0.01072971,
      "balance_loss_clip": 1.07312644,
      "balance_loss_mlp": 1.04480219,
      "epoch": 0.05134525777844581,
      "flos": 27198957571200.0,
      "grad_norm": 1.9945722649075286,
      "language_loss": 0.75480604,
      "learning_rate": 3.995223022193999e-06,
      "loss": 0.77788365,
      "num_input_tokens_seen": 18247075,
      "step": 854,
      "time_per_iteration": 2.521136999130249
    },
    {
      "auxiliary_loss_clip": 0.01214391,
      "auxiliary_loss_mlp": 0.01071637,
      "balance_loss_clip": 1.07310426,
      "balance_loss_mlp": 1.04350328,
      "epoch": 0.051405381031113785,
      "flos": 28362247436160.0,
      "grad_norm": 2.1119381876064898,
      "language_loss": 0.81608504,
      "learning_rate": 3.99519608253213e-06,
      "loss": 0.83894527,
      "num_input_tokens_seen": 18265680,
      "step": 855,
      "time_per_iteration": 2.5770468711853027
    },
    {
      "auxiliary_loss_clip": 0.01084305,
      "auxiliary_loss_mlp": 0.00839029,
      "balance_loss_clip": 1.04575634,
      "balance_loss_mlp": 1.15224361,
      "epoch": 0.05146550428378175,
      "flos": 65618169327360.0,
      "grad_norm": 1.0042976504068604,
      "language_loss": 0.65651131,
      "learning_rate": 3.995169067212227e-06,
      "loss": 0.67574465,
      "num_input_tokens_seen": 18327015,
      "step": 856,
      "time_per_iteration": 3.1591265201568604
    },
    {
      "auxiliary_loss_clip": 0.01204665,
      "auxiliary_loss_mlp": 0.01060097,
      "balance_loss_clip": 1.07042229,
      "balance_loss_mlp": 1.03271437,
      "epoch": 0.05152562753644972,
      "flos": 22054861998720.0,
      "grad_norm": 2.2446012315865076,
      "language_loss": 0.76849931,
      "learning_rate": 3.9951419762353116e-06,
      "loss": 0.79114693,
      "num_input_tokens_seen": 18345235,
      "step": 857,
      "time_per_iteration": 2.6026453971862793
    },
    {
      "auxiliary_loss_clip": 0.01195276,
      "auxiliary_loss_mlp": 0.01060537,
      "balance_loss_clip": 1.06812525,
      "balance_loss_mlp": 1.03252316,
      "epoch": 0.051585750789117694,
      "flos": 18509544783360.0,
      "grad_norm": 2.376093041208949,
      "language_loss": 0.89038754,
      "learning_rate": 3.995114809602412e-06,
      "loss": 0.91294569,
      "num_input_tokens_seen": 18362350,
      "step": 858,
      "time_per_iteration": 2.5573222637176514
    },
    {
      "auxiliary_loss_clip": 0.01207491,
      "auxiliary_loss_mlp": 0.01060131,
      "balance_loss_clip": 1.07195282,
      "balance_loss_mlp": 1.03298724,
      "epoch": 0.05164587404178566,
      "flos": 23730238108800.0,
      "grad_norm": 1.7732660427176947,
      "language_loss": 0.74813759,
      "learning_rate": 3.9950875673145605e-06,
      "loss": 0.77081382,
      "num_input_tokens_seen": 18383390,
      "step": 859,
      "time_per_iteration": 2.6116816997528076
    },
    {
      "auxiliary_loss_clip": 0.01200518,
      "auxiliary_loss_mlp": 0.01078996,
      "balance_loss_clip": 1.07151818,
      "balance_loss_mlp": 1.04926562,
      "epoch": 0.05170599729445363,
      "flos": 16252882876800.0,
      "grad_norm": 2.464442423931956,
      "language_loss": 0.91136885,
      "learning_rate": 3.995060249372788e-06,
      "loss": 0.93416399,
      "num_input_tokens_seen": 18399220,
      "step": 860,
      "time_per_iteration": 2.5147926807403564
    },
    {
      "auxiliary_loss_clip": 0.01236353,
      "auxiliary_loss_mlp": 0.01063292,
      "balance_loss_clip": 1.07511687,
      "balance_loss_mlp": 1.03722048,
      "epoch": 0.0517661205471216,
      "flos": 23985922095360.0,
      "grad_norm": 1.888055019229181,
      "language_loss": 0.82224113,
      "learning_rate": 3.99503285577813e-06,
      "loss": 0.84523761,
      "num_input_tokens_seen": 18419005,
      "step": 861,
      "time_per_iteration": 2.5715174674987793
    },
    {
      "auxiliary_loss_clip": 0.0120857,
      "auxiliary_loss_mlp": 0.01060628,
      "balance_loss_clip": 1.07130527,
      "balance_loss_mlp": 1.03410411,
      "epoch": 0.05182624379978957,
      "flos": 29277718392960.0,
      "grad_norm": 1.8381376592938383,
      "language_loss": 0.78692919,
      "learning_rate": 3.995005386531627e-06,
      "loss": 0.80962121,
      "num_input_tokens_seen": 18440550,
      "step": 862,
      "time_per_iteration": 2.6206045150756836
    },
    {
      "auxiliary_loss_clip": 0.01200965,
      "auxiliary_loss_mlp": 0.01073814,
      "balance_loss_clip": 1.06939781,
      "balance_loss_mlp": 1.04799294,
      "epoch": 0.05188636705245754,
      "flos": 24170826332160.0,
      "grad_norm": 2.043088376184506,
      "language_loss": 0.89069796,
      "learning_rate": 3.9949778416343195e-06,
      "loss": 0.91344577,
      "num_input_tokens_seen": 18461950,
      "step": 863,
      "time_per_iteration": 2.6068830490112305
    },
    {
      "auxiliary_loss_clip": 0.0121325,
      "auxiliary_loss_mlp": 0.01067477,
      "balance_loss_clip": 1.07461524,
      "balance_loss_mlp": 1.03905749,
      "epoch": 0.051946490305125506,
      "flos": 26760703731840.0,
      "grad_norm": 3.125802667825891,
      "language_loss": 0.76103586,
      "learning_rate": 3.9949502210872525e-06,
      "loss": 0.78384316,
      "num_input_tokens_seen": 18480555,
      "step": 864,
      "time_per_iteration": 2.5935940742492676
    },
    {
      "auxiliary_loss_clip": 0.01185308,
      "auxiliary_loss_mlp": 0.0106913,
      "balance_loss_clip": 1.05829477,
      "balance_loss_mlp": 1.0409255,
      "epoch": 0.05200661355779348,
      "flos": 21502519585920.0,
      "grad_norm": 2.0739711194129105,
      "language_loss": 0.78992319,
      "learning_rate": 3.994922524891474e-06,
      "loss": 0.81246763,
      "num_input_tokens_seen": 18499645,
      "step": 865,
      "time_per_iteration": 2.559828996658325
    },
    {
      "auxiliary_loss_clip": 0.01220478,
      "auxiliary_loss_mlp": 0.01067321,
      "balance_loss_clip": 1.07081628,
      "balance_loss_mlp": 1.04053497,
      "epoch": 0.05206673681046144,
      "flos": 18114492026880.0,
      "grad_norm": 2.8354941581406194,
      "language_loss": 0.85845405,
      "learning_rate": 3.994894753048032e-06,
      "loss": 0.88133204,
      "num_input_tokens_seen": 18516810,
      "step": 866,
      "time_per_iteration": 2.4799506664276123
    },
    {
      "auxiliary_loss_clip": 0.01198853,
      "auxiliary_loss_mlp": 0.01064054,
      "balance_loss_clip": 1.0777576,
      "balance_loss_mlp": 1.03785205,
      "epoch": 0.052126860063129415,
      "flos": 17524191916800.0,
      "grad_norm": 2.5149075774074072,
      "language_loss": 0.87631929,
      "learning_rate": 3.9948669055579815e-06,
      "loss": 0.89894837,
      "num_input_tokens_seen": 18532510,
      "step": 867,
      "time_per_iteration": 2.5681934356689453
    },
    {
      "auxiliary_loss_clip": 0.01168289,
      "auxiliary_loss_mlp": 0.01070839,
      "balance_loss_clip": 1.06075227,
      "balance_loss_mlp": 1.04626989,
      "epoch": 0.05218698331579739,
      "flos": 32598054771840.0,
      "grad_norm": 1.4800476005793581,
      "language_loss": 0.63637638,
      "learning_rate": 3.9948389824223785e-06,
      "loss": 0.65876764,
      "num_input_tokens_seen": 18557380,
      "step": 868,
      "time_per_iteration": 2.6763460636138916
    },
    {
      "auxiliary_loss_clip": 0.01235884,
      "auxiliary_loss_mlp": 0.01067624,
      "balance_loss_clip": 1.073071,
      "balance_loss_mlp": 1.03888226,
      "epoch": 0.05224710656846535,
      "flos": 22127293774080.0,
      "grad_norm": 1.9793633351175597,
      "language_loss": 0.83025193,
      "learning_rate": 3.994810983642281e-06,
      "loss": 0.85328698,
      "num_input_tokens_seen": 18575720,
      "step": 869,
      "time_per_iteration": 2.515263557434082
    },
    {
      "auxiliary_loss_clip": 0.01223619,
      "auxiliary_loss_mlp": 0.01058445,
      "balance_loss_clip": 1.07171845,
      "balance_loss_mlp": 1.03206372,
      "epoch": 0.052307229821133325,
      "flos": 11145092976000.0,
      "grad_norm": 1.9993669762815234,
      "language_loss": 0.87700975,
      "learning_rate": 3.994782909218751e-06,
      "loss": 0.8998304,
      "num_input_tokens_seen": 18592185,
      "step": 870,
      "time_per_iteration": 2.4927549362182617
    },
    {
      "auxiliary_loss_clip": 0.01235293,
      "auxiliary_loss_mlp": 0.01065714,
      "balance_loss_clip": 1.07340336,
      "balance_loss_mlp": 1.03980994,
      "epoch": 0.05236735307380129,
      "flos": 19128070005120.0,
      "grad_norm": 2.099647956412021,
      "language_loss": 0.81003368,
      "learning_rate": 3.994754759152854e-06,
      "loss": 0.83304381,
      "num_input_tokens_seen": 18609560,
      "step": 871,
      "time_per_iteration": 2.5126662254333496
    },
    {
      "auxiliary_loss_clip": 0.01205697,
      "auxiliary_loss_mlp": 0.0106062,
      "balance_loss_clip": 1.07587111,
      "balance_loss_mlp": 1.03582406,
      "epoch": 0.05242747632646926,
      "flos": 20960663944320.0,
      "grad_norm": 1.9524730775383052,
      "language_loss": 0.81130838,
      "learning_rate": 3.994726533445656e-06,
      "loss": 0.8339715,
      "num_input_tokens_seen": 18629405,
      "step": 872,
      "time_per_iteration": 2.560051918029785
    },
    {
      "auxiliary_loss_clip": 0.01096864,
      "auxiliary_loss_mlp": 0.01030318,
      "balance_loss_clip": 1.05232286,
      "balance_loss_mlp": 1.02519214,
      "epoch": 0.052487599579137234,
      "flos": 65020542842880.0,
      "grad_norm": 0.9019198494168122,
      "language_loss": 0.61693221,
      "learning_rate": 3.9946982320982274e-06,
      "loss": 0.63820398,
      "num_input_tokens_seen": 18681480,
      "step": 873,
      "time_per_iteration": 3.06003475189209
    },
    {
      "auxiliary_loss_clip": 0.01206362,
      "auxiliary_loss_mlp": 0.01055572,
      "balance_loss_clip": 1.07213938,
      "balance_loss_mlp": 1.03020477,
      "epoch": 0.0525477228318052,
      "flos": 23288859786240.0,
      "grad_norm": 1.93122624115511,
      "language_loss": 0.88771755,
      "learning_rate": 3.994669855111643e-06,
      "loss": 0.91033685,
      "num_input_tokens_seen": 18700390,
      "step": 874,
      "time_per_iteration": 2.576326608657837
    },
    {
      "auxiliary_loss_clip": 0.01205856,
      "auxiliary_loss_mlp": 0.01063366,
      "balance_loss_clip": 1.07055449,
      "balance_loss_mlp": 1.03733051,
      "epoch": 0.05260784608447317,
      "flos": 32230221546240.0,
      "grad_norm": 1.883844190560903,
      "language_loss": 0.74723351,
      "learning_rate": 3.994641402486977e-06,
      "loss": 0.76992565,
      "num_input_tokens_seen": 18721280,
      "step": 875,
      "time_per_iteration": 2.64450740814209
    },
    {
      "auxiliary_loss_clip": 0.01214965,
      "auxiliary_loss_mlp": 0.01056066,
      "balance_loss_clip": 1.07321906,
      "balance_loss_mlp": 1.02922034,
      "epoch": 0.052667969337141136,
      "flos": 24463211040000.0,
      "grad_norm": 1.724593397416063,
      "language_loss": 0.93172318,
      "learning_rate": 3.99461287422531e-06,
      "loss": 0.95443344,
      "num_input_tokens_seen": 18741545,
      "step": 876,
      "time_per_iteration": 2.552605390548706
    },
    {
      "auxiliary_loss_clip": 0.01126476,
      "auxiliary_loss_mlp": 0.01030353,
      "balance_loss_clip": 1.04974389,
      "balance_loss_mlp": 1.02594244,
      "epoch": 0.05272809258980911,
      "flos": 57784329567360.0,
      "grad_norm": 0.8173137274975907,
      "language_loss": 0.62917572,
      "learning_rate": 3.994584270327722e-06,
      "loss": 0.65074402,
      "num_input_tokens_seen": 18801400,
      "step": 877,
      "time_per_iteration": 3.1108858585357666
    },
    {
      "auxiliary_loss_clip": 0.01210986,
      "auxiliary_loss_mlp": 0.01072648,
      "balance_loss_clip": 1.07201231,
      "balance_loss_mlp": 1.04484797,
      "epoch": 0.05278821584247708,
      "flos": 17420805596160.0,
      "grad_norm": 2.1294198440006737,
      "language_loss": 0.85420358,
      "learning_rate": 3.994555590795299e-06,
      "loss": 0.87703985,
      "num_input_tokens_seen": 18819670,
      "step": 878,
      "time_per_iteration": 2.573084592819214
    },
    {
      "auxiliary_loss_clip": 0.01232923,
      "auxiliary_loss_mlp": 0.01063167,
      "balance_loss_clip": 1.07050157,
      "balance_loss_mlp": 1.03752494,
      "epoch": 0.052848339095145046,
      "flos": 26137258346880.0,
      "grad_norm": 1.9502374454425166,
      "language_loss": 0.83035666,
      "learning_rate": 3.9945268356291275e-06,
      "loss": 0.8533175,
      "num_input_tokens_seen": 18840580,
      "step": 879,
      "time_per_iteration": 2.522031545639038
    },
    {
      "auxiliary_loss_clip": 0.01200795,
      "auxiliary_loss_mlp": 0.0106825,
      "balance_loss_clip": 1.07149363,
      "balance_loss_mlp": 1.04122543,
      "epoch": 0.05290846234781302,
      "flos": 16472081623680.0,
      "grad_norm": 2.071213064451731,
      "language_loss": 0.84218007,
      "learning_rate": 3.9944980048302985e-06,
      "loss": 0.86487055,
      "num_input_tokens_seen": 18859295,
      "step": 880,
      "time_per_iteration": 3.934206247329712
    },
    {
      "auxiliary_loss_clip": 0.01192925,
      "auxiliary_loss_mlp": 0.0106315,
      "balance_loss_clip": 1.07080853,
      "balance_loss_mlp": 1.03706717,
      "epoch": 0.05296858560048098,
      "flos": 19865173000320.0,
      "grad_norm": 2.1498725176777214,
      "language_loss": 0.87145591,
      "learning_rate": 3.994469098399906e-06,
      "loss": 0.89401662,
      "num_input_tokens_seen": 18877485,
      "step": 881,
      "time_per_iteration": 2.5754125118255615
    },
    {
      "auxiliary_loss_clip": 0.01218282,
      "auxiliary_loss_mlp": 0.01067732,
      "balance_loss_clip": 1.070889,
      "balance_loss_mlp": 1.04021811,
      "epoch": 0.053028708853148955,
      "flos": 24388588535040.0,
      "grad_norm": 1.9092364171747176,
      "language_loss": 0.87847406,
      "learning_rate": 3.994440116339046e-06,
      "loss": 0.90133423,
      "num_input_tokens_seen": 18898275,
      "step": 882,
      "time_per_iteration": 3.947329044342041
    },
    {
      "auxiliary_loss_clip": 0.01233567,
      "auxiliary_loss_mlp": 0.01056917,
      "balance_loss_clip": 1.07226801,
      "balance_loss_mlp": 1.02988076,
      "epoch": 0.05308883210581693,
      "flos": 36393166143360.0,
      "grad_norm": 3.021160918032195,
      "language_loss": 0.69564617,
      "learning_rate": 3.994411058648816e-06,
      "loss": 0.71855098,
      "num_input_tokens_seen": 18920665,
      "step": 883,
      "time_per_iteration": 2.6199071407318115
    },
    {
      "auxiliary_loss_clip": 0.01170083,
      "auxiliary_loss_mlp": 0.01066639,
      "balance_loss_clip": 1.06000412,
      "balance_loss_mlp": 1.04105711,
      "epoch": 0.05314895535848489,
      "flos": 22855095146880.0,
      "grad_norm": 1.815307914756506,
      "language_loss": 0.75968301,
      "learning_rate": 3.994381925330319e-06,
      "loss": 0.78205025,
      "num_input_tokens_seen": 18939835,
      "step": 884,
      "time_per_iteration": 2.5956761837005615
    },
    {
      "auxiliary_loss_clip": 0.01182458,
      "auxiliary_loss_mlp": 0.01069316,
      "balance_loss_clip": 1.07529938,
      "balance_loss_mlp": 1.04336381,
      "epoch": 0.053209078611152864,
      "flos": 12860330204160.0,
      "grad_norm": 2.0766826139352164,
      "language_loss": 0.85653013,
      "learning_rate": 3.994352716384659e-06,
      "loss": 0.87904787,
      "num_input_tokens_seen": 18958405,
      "step": 885,
      "time_per_iteration": 2.6230082511901855
    },
    {
      "auxiliary_loss_clip": 0.01185331,
      "auxiliary_loss_mlp": 0.01069384,
      "balance_loss_clip": 1.05735052,
      "balance_loss_mlp": 1.04251468,
      "epoch": 0.05326920186382083,
      "flos": 12164596698240.0,
      "grad_norm": 3.4819633081118546,
      "language_loss": 0.85972977,
      "learning_rate": 3.994323431812945e-06,
      "loss": 0.88227689,
      "num_input_tokens_seen": 18975445,
      "step": 886,
      "time_per_iteration": 3.9707014560699463
    },
    {
      "auxiliary_loss_clip": 0.01187475,
      "auxiliary_loss_mlp": 0.01065515,
      "balance_loss_clip": 1.07113957,
      "balance_loss_mlp": 1.03824019,
      "epoch": 0.0533293251164888,
      "flos": 22704485420160.0,
      "grad_norm": 2.1228156076950615,
      "language_loss": 0.89277673,
      "learning_rate": 3.994294071616286e-06,
      "loss": 0.91530657,
      "num_input_tokens_seen": 18991930,
      "step": 887,
      "time_per_iteration": 2.5956554412841797
    },
    {
      "auxiliary_loss_clip": 0.01156326,
      "auxiliary_loss_mlp": 0.01075968,
      "balance_loss_clip": 1.06784678,
      "balance_loss_mlp": 1.04721451,
      "epoch": 0.053389448369156774,
      "flos": 26940939200640.0,
      "grad_norm": 1.98232819460545,
      "language_loss": 0.75045687,
      "learning_rate": 3.994264635795796e-06,
      "loss": 0.77277982,
      "num_input_tokens_seen": 19009790,
      "step": 888,
      "time_per_iteration": 4.084397554397583
    },
    {
      "auxiliary_loss_clip": 0.0117827,
      "auxiliary_loss_mlp": 0.01075751,
      "balance_loss_clip": 1.07159567,
      "balance_loss_mlp": 1.04752207,
      "epoch": 0.05344957162182474,
      "flos": 25556331686400.0,
      "grad_norm": 1.9056828682042726,
      "language_loss": 0.88298452,
      "learning_rate": 3.994235124352592e-06,
      "loss": 0.90552473,
      "num_input_tokens_seen": 19030170,
      "step": 889,
      "time_per_iteration": 2.637068033218384
    },
    {
      "auxiliary_loss_clip": 0.01227286,
      "auxiliary_loss_mlp": 0.01047467,
      "balance_loss_clip": 1.07140684,
      "balance_loss_mlp": 1.02270699,
      "epoch": 0.05350969487449271,
      "flos": 19719591177600.0,
      "grad_norm": 1.7807413465836133,
      "language_loss": 0.89102387,
      "learning_rate": 3.994205537287791e-06,
      "loss": 0.91377139,
      "num_input_tokens_seen": 19048075,
      "step": 890,
      "time_per_iteration": 2.497800588607788
    },
    {
      "auxiliary_loss_clip": 0.01206129,
      "auxiliary_loss_mlp": 0.01065784,
      "balance_loss_clip": 1.0695436,
      "balance_loss_mlp": 1.04133403,
      "epoch": 0.053569818127160676,
      "flos": 27016351804800.0,
      "grad_norm": 2.4296235003731823,
      "language_loss": 0.93274426,
      "learning_rate": 3.994175874602517e-06,
      "loss": 0.95546335,
      "num_input_tokens_seen": 19067465,
      "step": 891,
      "time_per_iteration": 2.5771398544311523
    },
    {
      "auxiliary_loss_clip": 0.01205061,
      "auxiliary_loss_mlp": 0.01065856,
      "balance_loss_clip": 1.07170367,
      "balance_loss_mlp": 1.03747201,
      "epoch": 0.05362994137982865,
      "flos": 13188338225280.0,
      "grad_norm": 2.1834680020691946,
      "language_loss": 0.71693456,
      "learning_rate": 3.994146136297893e-06,
      "loss": 0.73964375,
      "num_input_tokens_seen": 19085505,
      "step": 892,
      "time_per_iteration": 2.536660671234131
    },
    {
      "auxiliary_loss_clip": 0.012045,
      "auxiliary_loss_mlp": 0.0112842,
      "balance_loss_clip": 1.06974065,
      "balance_loss_mlp": 1.6300416,
      "epoch": 0.05369006463249662,
      "flos": 28658008022400.0,
      "grad_norm": 1.9794291813656961,
      "language_loss": 0.8236239,
      "learning_rate": 3.994116322375049e-06,
      "loss": 0.84695309,
      "num_input_tokens_seen": 19104360,
      "step": 893,
      "time_per_iteration": 2.602376937866211
    },
    {
      "auxiliary_loss_clip": 0.01205823,
      "auxiliary_loss_mlp": 0.01067895,
      "balance_loss_clip": 1.07102883,
      "balance_loss_mlp": 1.04189563,
      "epoch": 0.053750187885164585,
      "flos": 28913153304960.0,
      "grad_norm": 1.9212043746376017,
      "language_loss": 0.81746852,
      "learning_rate": 3.994086432835114e-06,
      "loss": 0.84020567,
      "num_input_tokens_seen": 19124680,
      "step": 894,
      "time_per_iteration": 2.631514310836792
    },
    {
      "auxiliary_loss_clip": 0.0121284,
      "auxiliary_loss_mlp": 0.01062131,
      "balance_loss_clip": 1.06932735,
      "balance_loss_mlp": 1.03633404,
      "epoch": 0.05381031113783256,
      "flos": 15158828476800.0,
      "grad_norm": 2.7376740373639628,
      "language_loss": 0.75101012,
      "learning_rate": 3.994056467679221e-06,
      "loss": 0.77375978,
      "num_input_tokens_seen": 19142895,
      "step": 895,
      "time_per_iteration": 2.5075786113739014
    },
    {
      "auxiliary_loss_clip": 0.01213128,
      "auxiliary_loss_mlp": 0.01061693,
      "balance_loss_clip": 1.07725394,
      "balance_loss_mlp": 1.03589654,
      "epoch": 0.05387043439050053,
      "flos": 21835232288640.0,
      "grad_norm": 2.111736705586089,
      "language_loss": 0.86780262,
      "learning_rate": 3.9940264269085065e-06,
      "loss": 0.89055085,
      "num_input_tokens_seen": 19163125,
      "step": 896,
      "time_per_iteration": 2.603149890899658
    },
    {
      "auxiliary_loss_clip": 0.01233952,
      "auxiliary_loss_mlp": 0.01068852,
      "balance_loss_clip": 1.07206035,
      "balance_loss_mlp": 1.52363825,
      "epoch": 0.053930557643168495,
      "flos": 17310308382720.0,
      "grad_norm": 2.223050599621852,
      "language_loss": 0.87642455,
      "learning_rate": 3.9939963105241115e-06,
      "loss": 0.89945269,
      "num_input_tokens_seen": 19179385,
      "step": 897,
      "time_per_iteration": 2.4829087257385254
    },
    {
      "auxiliary_loss_clip": 0.01213993,
      "auxiliary_loss_mlp": 0.01063299,
      "balance_loss_clip": 1.07268429,
      "balance_loss_mlp": 1.03578568,
      "epoch": 0.05399068089583647,
      "flos": 17348481561600.0,
      "grad_norm": 1.7753305616385673,
      "language_loss": 0.90397179,
      "learning_rate": 3.993966118527175e-06,
      "loss": 0.9267447,
      "num_input_tokens_seen": 19198725,
      "step": 898,
      "time_per_iteration": 2.5142648220062256
    },
    {
      "auxiliary_loss_clip": 0.01212827,
      "auxiliary_loss_mlp": 0.01080085,
      "balance_loss_clip": 1.07265902,
      "balance_loss_mlp": 1.054479,
      "epoch": 0.05405080414850443,
      "flos": 17486952491520.0,
      "grad_norm": 3.0103010763471603,
      "language_loss": 0.92109048,
      "learning_rate": 3.993935850918845e-06,
      "loss": 0.94401962,
      "num_input_tokens_seen": 19212380,
      "step": 899,
      "time_per_iteration": 2.5852553844451904
    },
    {
      "auxiliary_loss_clip": 0.01197865,
      "auxiliary_loss_mlp": 0.01072378,
      "balance_loss_clip": 1.06803334,
      "balance_loss_mlp": 1.04436338,
      "epoch": 0.054110927401172404,
      "flos": 24496787278080.0,
      "grad_norm": 2.1574361704972804,
      "language_loss": 0.75842959,
      "learning_rate": 3.9939055077002665e-06,
      "loss": 0.78113198,
      "num_input_tokens_seen": 19232235,
      "step": 900,
      "time_per_iteration": 2.583963394165039
    },
    {
      "auxiliary_loss_clip": 0.01221599,
      "auxiliary_loss_mlp": 0.0105628,
      "balance_loss_clip": 1.06833267,
      "balance_loss_mlp": 1.03127027,
      "epoch": 0.054171050653840376,
      "flos": 22930040874240.0,
      "grad_norm": 2.1765975908191035,
      "language_loss": 0.73895329,
      "learning_rate": 3.993875088872592e-06,
      "loss": 0.7617321,
      "num_input_tokens_seen": 19251460,
      "step": 901,
      "time_per_iteration": 2.5631070137023926
    },
    {
      "auxiliary_loss_clip": 0.01189995,
      "auxiliary_loss_mlp": 0.01065611,
      "balance_loss_clip": 1.0692575,
      "balance_loss_mlp": 1.04074383,
      "epoch": 0.05423117390650834,
      "flos": 12933192942720.0,
      "grad_norm": 2.7497067550315157,
      "language_loss": 0.84882116,
      "learning_rate": 3.9938445944369745e-06,
      "loss": 0.87137723,
      "num_input_tokens_seen": 19269060,
      "step": 902,
      "time_per_iteration": 2.5651910305023193
    },
    {
      "auxiliary_loss_clip": 0.01179484,
      "auxiliary_loss_mlp": 0.01065627,
      "balance_loss_clip": 1.07582521,
      "balance_loss_mlp": 1.03875756,
      "epoch": 0.05429129715917631,
      "flos": 19901335017600.0,
      "grad_norm": 1.7756390456982243,
      "language_loss": 0.86547661,
      "learning_rate": 3.993814024394569e-06,
      "loss": 0.88792765,
      "num_input_tokens_seen": 19288620,
      "step": 903,
      "time_per_iteration": 2.6966285705566406
    },
    {
      "auxiliary_loss_clip": 0.01218427,
      "auxiliary_loss_mlp": 0.01058187,
      "balance_loss_clip": 1.07121587,
      "balance_loss_mlp": 1.03333211,
      "epoch": 0.05435142041184428,
      "flos": 16908611610240.0,
      "grad_norm": 2.248947337203656,
      "language_loss": 0.74946427,
      "learning_rate": 3.993783378746537e-06,
      "loss": 0.77223045,
      "num_input_tokens_seen": 19306615,
      "step": 904,
      "time_per_iteration": 2.534850597381592
    },
    {
      "auxiliary_loss_clip": 0.01216298,
      "auxiliary_loss_mlp": 0.01071087,
      "balance_loss_clip": 1.06919765,
      "balance_loss_mlp": 1.04627907,
      "epoch": 0.05441154366451225,
      "flos": 23948323534080.0,
      "grad_norm": 2.217768241182223,
      "language_loss": 0.85702819,
      "learning_rate": 3.993752657494039e-06,
      "loss": 0.87990201,
      "num_input_tokens_seen": 19321680,
      "step": 905,
      "time_per_iteration": 2.557027816772461
    },
    {
      "auxiliary_loss_clip": 0.01207649,
      "auxiliary_loss_mlp": 0.01070205,
      "balance_loss_clip": 1.07417011,
      "balance_loss_mlp": 1.04515886,
      "epoch": 0.05447166691718022,
      "flos": 19975382904960.0,
      "grad_norm": 1.8010172629073642,
      "language_loss": 0.74334836,
      "learning_rate": 3.993721860638241e-06,
      "loss": 0.76612693,
      "num_input_tokens_seen": 19339760,
      "step": 906,
      "time_per_iteration": 2.5589680671691895
    },
    {
      "auxiliary_loss_clip": 0.01203725,
      "auxiliary_loss_mlp": 0.01064354,
      "balance_loss_clip": 1.07185113,
      "balance_loss_mlp": 1.03877187,
      "epoch": 0.05453179016984819,
      "flos": 24936513575040.0,
      "grad_norm": 2.0601340770347467,
      "language_loss": 0.87160665,
      "learning_rate": 3.993690988180309e-06,
      "loss": 0.89428747,
      "num_input_tokens_seen": 19359585,
      "step": 907,
      "time_per_iteration": 2.6042590141296387
    },
    {
      "auxiliary_loss_clip": 0.01215902,
      "auxiliary_loss_mlp": 0.01072223,
      "balance_loss_clip": 1.07247162,
      "balance_loss_mlp": 1.04599679,
      "epoch": 0.05459191342251616,
      "flos": 18115102558080.0,
      "grad_norm": 1.9147250809076852,
      "language_loss": 0.87107456,
      "learning_rate": 3.9936600401214165e-06,
      "loss": 0.89395589,
      "num_input_tokens_seen": 19378590,
      "step": 908,
      "time_per_iteration": 2.5289340019226074
    },
    {
      "auxiliary_loss_clip": 0.0120781,
      "auxiliary_loss_mlp": 0.01067651,
      "balance_loss_clip": 1.06887841,
      "balance_loss_mlp": 1.04075694,
      "epoch": 0.054652036675184125,
      "flos": 19208295031680.0,
      "grad_norm": 2.5870235586430548,
      "language_loss": 0.89760447,
      "learning_rate": 3.9936290164627345e-06,
      "loss": 0.92035908,
      "num_input_tokens_seen": 19397910,
      "step": 909,
      "time_per_iteration": 2.586700916290283
    },
    {
      "auxiliary_loss_clip": 0.01209808,
      "auxiliary_loss_mlp": 0.01074114,
      "balance_loss_clip": 1.07129347,
      "balance_loss_mlp": 1.04730344,
      "epoch": 0.0547121599278521,
      "flos": 16325745615360.0,
      "grad_norm": 2.618369487881624,
      "language_loss": 0.70765096,
      "learning_rate": 3.99359791720544e-06,
      "loss": 0.73049021,
      "num_input_tokens_seen": 19415950,
      "step": 910,
      "time_per_iteration": 2.53511905670166
    },
    {
      "auxiliary_loss_clip": 0.01200062,
      "auxiliary_loss_mlp": 0.01053318,
      "balance_loss_clip": 1.06863356,
      "balance_loss_mlp": 1.02862978,
      "epoch": 0.05477228318052007,
      "flos": 20339014239360.0,
      "grad_norm": 1.8868595352342914,
      "language_loss": 0.83306777,
      "learning_rate": 3.993566742350714e-06,
      "loss": 0.85560155,
      "num_input_tokens_seen": 19435275,
      "step": 911,
      "time_per_iteration": 2.587833881378174
    },
    {
      "auxiliary_loss_clip": 0.01200872,
      "auxiliary_loss_mlp": 0.01070358,
      "balance_loss_clip": 1.06841397,
      "balance_loss_mlp": 1.04303503,
      "epoch": 0.054832406433188034,
      "flos": 21973092687360.0,
      "grad_norm": 2.4222115748246744,
      "language_loss": 0.75998247,
      "learning_rate": 3.993535491899736e-06,
      "loss": 0.78269482,
      "num_input_tokens_seen": 19452090,
      "step": 912,
      "time_per_iteration": 2.5228497982025146
    },
    {
      "auxiliary_loss_clip": 0.01198495,
      "auxiliary_loss_mlp": 0.01050651,
      "balance_loss_clip": 1.06942177,
      "balance_loss_mlp": 1.02553356,
      "epoch": 0.054892529685856006,
      "flos": 16398931576320.0,
      "grad_norm": 2.250017035664635,
      "language_loss": 0.82952851,
      "learning_rate": 3.993504165853694e-06,
      "loss": 0.85202003,
      "num_input_tokens_seen": 19470865,
      "step": 913,
      "time_per_iteration": 2.5697543621063232
    },
    {
      "auxiliary_loss_clip": 0.01212633,
      "auxiliary_loss_mlp": 0.01058776,
      "balance_loss_clip": 1.08327341,
      "balance_loss_mlp": 1.03445721,
      "epoch": 0.05495265293852397,
      "flos": 23912341084800.0,
      "grad_norm": 2.2020157072855473,
      "language_loss": 0.83744383,
      "learning_rate": 3.993472764213772e-06,
      "loss": 0.86015797,
      "num_input_tokens_seen": 19492145,
      "step": 914,
      "time_per_iteration": 2.5590310096740723
    },
    {
      "auxiliary_loss_clip": 0.01220928,
      "auxiliary_loss_mlp": 0.00848193,
      "balance_loss_clip": 1.07384896,
      "balance_loss_mlp": 1.1141088,
      "epoch": 0.055012776191191944,
      "flos": 23586954756480.0,
      "grad_norm": 3.3950793702037205,
      "language_loss": 0.90389013,
      "learning_rate": 3.9934412869811655e-06,
      "loss": 0.92458141,
      "num_input_tokens_seen": 19511015,
      "step": 915,
      "time_per_iteration": 2.596266984939575
    },
    {
      "auxiliary_loss_clip": 0.01213258,
      "auxiliary_loss_mlp": 0.01055884,
      "balance_loss_clip": 1.07529211,
      "balance_loss_mlp": 1.03152943,
      "epoch": 0.055072899443859916,
      "flos": 17528501548800.0,
      "grad_norm": 1.9940370684154272,
      "language_loss": 0.897946,
      "learning_rate": 3.993409734157064e-06,
      "loss": 0.92063737,
      "num_input_tokens_seen": 19529040,
      "step": 916,
      "time_per_iteration": 2.4945449829101562
    },
    {
      "auxiliary_loss_clip": 0.01193556,
      "auxiliary_loss_mlp": 0.01062865,
      "balance_loss_clip": 1.08301973,
      "balance_loss_mlp": 1.03829622,
      "epoch": 0.05513302269652788,
      "flos": 21687172427520.0,
      "grad_norm": 2.4024600693564215,
      "language_loss": 0.80377972,
      "learning_rate": 3.993378105742666e-06,
      "loss": 0.82634395,
      "num_input_tokens_seen": 19549540,
      "step": 917,
      "time_per_iteration": 2.6394553184509277
    },
    {
      "auxiliary_loss_clip": 0.01153868,
      "auxiliary_loss_mlp": 0.01061776,
      "balance_loss_clip": 1.0677619,
      "balance_loss_mlp": 1.03607464,
      "epoch": 0.05519314594919585,
      "flos": 21613340021760.0,
      "grad_norm": 1.7691600682540194,
      "language_loss": 0.79320288,
      "learning_rate": 3.9933464017391705e-06,
      "loss": 0.81535935,
      "num_input_tokens_seen": 19567570,
      "step": 918,
      "time_per_iteration": 2.6425232887268066
    },
    {
      "auxiliary_loss_clip": 0.01211273,
      "auxiliary_loss_mlp": 0.01056598,
      "balance_loss_clip": 1.07026434,
      "balance_loss_mlp": 1.03180277,
      "epoch": 0.05525326920186382,
      "flos": 21798567480960.0,
      "grad_norm": 2.2911785852361013,
      "language_loss": 0.89679003,
      "learning_rate": 3.99331462214778e-06,
      "loss": 0.91946876,
      "num_input_tokens_seen": 19585330,
      "step": 919,
      "time_per_iteration": 2.5472095012664795
    },
    {
      "auxiliary_loss_clip": 0.01225496,
      "auxiliary_loss_mlp": 0.01072748,
      "balance_loss_clip": 1.07110202,
      "balance_loss_mlp": 1.04705799,
      "epoch": 0.05531339245453179,
      "flos": 28439635288320.0,
      "grad_norm": 2.7296299240980435,
      "language_loss": 0.87570059,
      "learning_rate": 3.993282766969699e-06,
      "loss": 0.89868307,
      "num_input_tokens_seen": 19604970,
      "step": 920,
      "time_per_iteration": 5.368548154830933
    },
    {
      "auxiliary_loss_clip": 0.01202036,
      "auxiliary_loss_mlp": 0.01058711,
      "balance_loss_clip": 1.07139444,
      "balance_loss_mlp": 1.03392744,
      "epoch": 0.05537351570719976,
      "flos": 37375143131520.0,
      "grad_norm": 1.9243711142471018,
      "language_loss": 0.65750051,
      "learning_rate": 3.993250836206136e-06,
      "loss": 0.68010795,
      "num_input_tokens_seen": 19626235,
      "step": 921,
      "time_per_iteration": 2.6874630451202393
    },
    {
      "auxiliary_loss_clip": 0.01223011,
      "auxiliary_loss_mlp": 0.01067371,
      "balance_loss_clip": 1.07324648,
      "balance_loss_mlp": 1.03917766,
      "epoch": 0.05543363895986773,
      "flos": 20084479488000.0,
      "grad_norm": 2.584578921048857,
      "language_loss": 0.72042388,
      "learning_rate": 3.993218829858301e-06,
      "loss": 0.74332768,
      "num_input_tokens_seen": 19644305,
      "step": 922,
      "time_per_iteration": 2.543826103210449
    },
    {
      "auxiliary_loss_clip": 0.01208275,
      "auxiliary_loss_mlp": 0.01067381,
      "balance_loss_clip": 1.08123207,
      "balance_loss_mlp": 1.04109502,
      "epoch": 0.0554937622125357,
      "flos": 24533200690560.0,
      "grad_norm": 4.640836319641821,
      "language_loss": 0.82212853,
      "learning_rate": 3.993186747927408e-06,
      "loss": 0.84488511,
      "num_input_tokens_seen": 19662130,
      "step": 923,
      "time_per_iteration": 2.5858380794525146
    },
    {
      "auxiliary_loss_clip": 0.01215078,
      "auxiliary_loss_mlp": 0.01068384,
      "balance_loss_clip": 1.06863916,
      "balance_loss_mlp": 1.04315984,
      "epoch": 0.055553885465203665,
      "flos": 14320063013760.0,
      "grad_norm": 2.2588118448663743,
      "language_loss": 0.78730309,
      "learning_rate": 3.993154590414675e-06,
      "loss": 0.81013775,
      "num_input_tokens_seen": 19680715,
      "step": 924,
      "time_per_iteration": 2.5336220264434814
    },
    {
      "auxiliary_loss_clip": 0.01179954,
      "auxiliary_loss_mlp": 0.01058713,
      "balance_loss_clip": 1.07005489,
      "balance_loss_mlp": 1.03252292,
      "epoch": 0.05561400871787164,
      "flos": 27381132374400.0,
      "grad_norm": 2.6713120625843105,
      "language_loss": 1.02390254,
      "learning_rate": 3.993122357321319e-06,
      "loss": 1.04628932,
      "num_input_tokens_seen": 19700535,
      "step": 925,
      "time_per_iteration": 4.010425090789795
    },
    {
      "auxiliary_loss_clip": 0.01163645,
      "auxiliary_loss_mlp": 0.01054933,
      "balance_loss_clip": 1.05221725,
      "balance_loss_mlp": 1.02862406,
      "epoch": 0.05567413197053961,
      "flos": 23221096778880.0,
      "grad_norm": 2.200777159822035,
      "language_loss": 0.8130753,
      "learning_rate": 3.993090048648564e-06,
      "loss": 0.83526105,
      "num_input_tokens_seen": 19718825,
      "step": 926,
      "time_per_iteration": 4.102551698684692
    },
    {
      "auxiliary_loss_clip": 0.01222813,
      "auxiliary_loss_mlp": 0.01069178,
      "balance_loss_clip": 1.0737493,
      "balance_loss_mlp": 1.04271317,
      "epoch": 0.055734255223207574,
      "flos": 25264952559360.0,
      "grad_norm": 2.964969896398472,
      "language_loss": 0.73120177,
      "learning_rate": 3.993057664397634e-06,
      "loss": 0.75412166,
      "num_input_tokens_seen": 19739080,
      "step": 927,
      "time_per_iteration": 2.527183771133423
    },
    {
      "auxiliary_loss_clip": 0.01130848,
      "auxiliary_loss_mlp": 0.01007181,
      "balance_loss_clip": 1.05468082,
      "balance_loss_mlp": 1.00277066,
      "epoch": 0.055794378475875546,
      "flos": 66503116702080.0,
      "grad_norm": 0.7820258878125067,
      "language_loss": 0.59870601,
      "learning_rate": 3.9930252045697585e-06,
      "loss": 0.62008631,
      "num_input_tokens_seen": 19802960,
      "step": 928,
      "time_per_iteration": 3.1208837032318115
    },
    {
      "auxiliary_loss_clip": 0.01217137,
      "auxiliary_loss_mlp": 0.01063333,
      "balance_loss_clip": 1.07297695,
      "balance_loss_mlp": 1.03786993,
      "epoch": 0.05585450172854351,
      "flos": 25337635729920.0,
      "grad_norm": 2.657287173913943,
      "language_loss": 0.95007539,
      "learning_rate": 3.992992669166168e-06,
      "loss": 0.97288007,
      "num_input_tokens_seen": 19822765,
      "step": 929,
      "time_per_iteration": 2.550182819366455
    },
    {
      "auxiliary_loss_clip": 0.01187307,
      "auxiliary_loss_mlp": 0.01068717,
      "balance_loss_clip": 1.06842649,
      "balance_loss_mlp": 1.0409174,
      "epoch": 0.05591462498121148,
      "flos": 33911738881920.0,
      "grad_norm": 2.1842478256239244,
      "language_loss": 0.7206881,
      "learning_rate": 3.992960058188094e-06,
      "loss": 0.74324834,
      "num_input_tokens_seen": 19843590,
      "step": 930,
      "time_per_iteration": 2.704300880432129
    },
    {
      "auxiliary_loss_clip": 0.01205205,
      "auxiliary_loss_mlp": 0.0106243,
      "balance_loss_clip": 1.07465911,
      "balance_loss_mlp": 1.03601325,
      "epoch": 0.055974748233879455,
      "flos": 17930880679680.0,
      "grad_norm": 2.992742214691456,
      "language_loss": 0.85260862,
      "learning_rate": 3.992927371636776e-06,
      "loss": 0.87528497,
      "num_input_tokens_seen": 19860230,
      "step": 931,
      "time_per_iteration": 2.5343477725982666
    },
    {
      "auxiliary_loss_clip": 0.01237587,
      "auxiliary_loss_mlp": 0.00823184,
      "balance_loss_clip": 1.10563719,
      "balance_loss_mlp": 1.06617951,
      "epoch": 0.05603487148654742,
      "flos": 24021976371840.0,
      "grad_norm": 2.37893649935355,
      "language_loss": 0.83495522,
      "learning_rate": 3.9928946095134525e-06,
      "loss": 0.85556287,
      "num_input_tokens_seen": 19880795,
      "step": 932,
      "time_per_iteration": 2.5562260150909424
    },
    {
      "auxiliary_loss_clip": 0.0121563,
      "auxiliary_loss_mlp": 0.01070444,
      "balance_loss_clip": 1.07321918,
      "balance_loss_mlp": 1.04335928,
      "epoch": 0.05609499473921539,
      "flos": 17307758517120.0,
      "grad_norm": 5.889604115011838,
      "language_loss": 0.74117213,
      "learning_rate": 3.992861771819365e-06,
      "loss": 0.76403284,
      "num_input_tokens_seen": 19897960,
      "step": 933,
      "time_per_iteration": 2.498112678527832
    },
    {
      "auxiliary_loss_clip": 0.0117536,
      "auxiliary_loss_mlp": 0.01069474,
      "balance_loss_clip": 1.07733071,
      "balance_loss_mlp": 1.04281878,
      "epoch": 0.05615511799188336,
      "flos": 20994742972800.0,
      "grad_norm": 2.331500448204897,
      "language_loss": 0.86854899,
      "learning_rate": 3.99282885855576e-06,
      "loss": 0.89099741,
      "num_input_tokens_seen": 19913315,
      "step": 934,
      "time_per_iteration": 2.609607458114624
    },
    {
      "auxiliary_loss_clip": 0.01178463,
      "auxiliary_loss_mlp": 0.01069507,
      "balance_loss_clip": 1.07047081,
      "balance_loss_mlp": 1.04447246,
      "epoch": 0.05621524124455133,
      "flos": 17273535834240.0,
      "grad_norm": 2.6067818713856035,
      "language_loss": 0.8048414,
      "learning_rate": 3.992795869723885e-06,
      "loss": 0.82732105,
      "num_input_tokens_seen": 19928790,
      "step": 935,
      "time_per_iteration": 2.630005121231079
    },
    {
      "auxiliary_loss_clip": 0.01121719,
      "auxiliary_loss_mlp": 0.01009546,
      "balance_loss_clip": 1.04589105,
      "balance_loss_mlp": 1.00532579,
      "epoch": 0.0562753644972193,
      "flos": 58719370458240.0,
      "grad_norm": 0.8268353941187083,
      "language_loss": 0.69203669,
      "learning_rate": 3.99276280532499e-06,
      "loss": 0.71334934,
      "num_input_tokens_seen": 19988785,
      "step": 936,
      "time_per_iteration": 2.99532413482666
    },
    {
      "auxiliary_loss_clip": 0.01229328,
      "auxiliary_loss_mlp": 0.01061983,
      "balance_loss_clip": 1.07230377,
      "balance_loss_mlp": 1.03730655,
      "epoch": 0.05633548774988727,
      "flos": 17457039440640.0,
      "grad_norm": 2.1979195646578087,
      "language_loss": 0.75785667,
      "learning_rate": 3.992729665360331e-06,
      "loss": 0.78076977,
      "num_input_tokens_seen": 20007685,
      "step": 937,
      "time_per_iteration": 2.4748282432556152
    },
    {
      "auxiliary_loss_clip": 0.01115146,
      "auxiliary_loss_mlp": 0.01004898,
      "balance_loss_clip": 1.05102253,
      "balance_loss_mlp": 1.00079715,
      "epoch": 0.05639561100255524,
      "flos": 70654928083200.0,
      "grad_norm": 0.8610421630523332,
      "language_loss": 0.64420635,
      "learning_rate": 3.992696449831162e-06,
      "loss": 0.66540682,
      "num_input_tokens_seen": 20072750,
      "step": 938,
      "time_per_iteration": 3.0760161876678467
    },
    {
      "auxiliary_loss_clip": 0.01189525,
      "auxiliary_loss_mlp": 0.01063866,
      "balance_loss_clip": 1.07609892,
      "balance_loss_mlp": 1.0372225,
      "epoch": 0.056455734255223204,
      "flos": 20485996692480.0,
      "grad_norm": 2.909173545331531,
      "language_loss": 0.79749429,
      "learning_rate": 3.992663158738745e-06,
      "loss": 0.82002819,
      "num_input_tokens_seen": 20089070,
      "step": 939,
      "time_per_iteration": 2.5867676734924316
    },
    {
      "auxiliary_loss_clip": 0.01194965,
      "auxiliary_loss_mlp": 0.0106925,
      "balance_loss_clip": 1.07582033,
      "balance_loss_mlp": 1.04378676,
      "epoch": 0.056515857507891176,
      "flos": 22053569109120.0,
      "grad_norm": 1.7871310028209755,
      "language_loss": 0.74238747,
      "learning_rate": 3.992629792084341e-06,
      "loss": 0.76502961,
      "num_input_tokens_seen": 20108790,
      "step": 940,
      "time_per_iteration": 2.5813028812408447
    },
    {
      "auxiliary_loss_clip": 0.01214268,
      "auxiliary_loss_mlp": 0.01063601,
      "balance_loss_clip": 1.07232594,
      "balance_loss_mlp": 1.03754222,
      "epoch": 0.05657598076055915,
      "flos": 24025316336640.0,
      "grad_norm": 2.4397014242942725,
      "language_loss": 0.70547098,
      "learning_rate": 3.992596349869216e-06,
      "loss": 0.72824967,
      "num_input_tokens_seen": 20128455,
      "step": 941,
      "time_per_iteration": 2.5666110515594482
    },
    {
      "auxiliary_loss_clip": 0.01154887,
      "auxiliary_loss_mlp": 0.01064707,
      "balance_loss_clip": 1.07694435,
      "balance_loss_mlp": 1.03849268,
      "epoch": 0.05663610401322711,
      "flos": 20480609652480.0,
      "grad_norm": 2.120997432076242,
      "language_loss": 0.80879784,
      "learning_rate": 3.992562832094637e-06,
      "loss": 0.83099383,
      "num_input_tokens_seen": 20145775,
      "step": 942,
      "time_per_iteration": 2.615006923675537
    },
    {
      "auxiliary_loss_clip": 0.01198012,
      "auxiliary_loss_mlp": 0.01070202,
      "balance_loss_clip": 1.06569827,
      "balance_loss_mlp": 1.04354692,
      "epoch": 0.056696227265895086,
      "flos": 21069042255360.0,
      "grad_norm": 2.1205185095260695,
      "language_loss": 0.88050014,
      "learning_rate": 3.9925292387618755e-06,
      "loss": 0.90318227,
      "num_input_tokens_seen": 20164315,
      "step": 943,
      "time_per_iteration": 2.537804126739502
    },
    {
      "auxiliary_loss_clip": 0.01211865,
      "auxiliary_loss_mlp": 0.01058704,
      "balance_loss_clip": 1.07266855,
      "balance_loss_mlp": 1.03454065,
      "epoch": 0.05675635051856306,
      "flos": 17821317219840.0,
      "grad_norm": 2.3912290263554676,
      "language_loss": 0.75273067,
      "learning_rate": 3.992495569872206e-06,
      "loss": 0.77543634,
      "num_input_tokens_seen": 20182760,
      "step": 944,
      "time_per_iteration": 2.5158066749572754
    },
    {
      "auxiliary_loss_clip": 0.01214209,
      "auxiliary_loss_mlp": 0.01061553,
      "balance_loss_clip": 1.06859994,
      "balance_loss_mlp": 1.03779483,
      "epoch": 0.05681647377123102,
      "flos": 23114945111040.0,
      "grad_norm": 1.8489148838888723,
      "language_loss": 0.79560876,
      "learning_rate": 3.992461825426906e-06,
      "loss": 0.81836641,
      "num_input_tokens_seen": 20203830,
      "step": 945,
      "time_per_iteration": 2.5499508380889893
    },
    {
      "auxiliary_loss_clip": 0.01208364,
      "auxiliary_loss_mlp": 0.01057077,
      "balance_loss_clip": 1.06919873,
      "balance_loss_mlp": 1.03236508,
      "epoch": 0.056876597023898995,
      "flos": 16070528505600.0,
      "grad_norm": 4.081894037189643,
      "language_loss": 0.82539171,
      "learning_rate": 3.992428005427252e-06,
      "loss": 0.84804618,
      "num_input_tokens_seen": 20220365,
      "step": 946,
      "time_per_iteration": 2.492144823074341
    },
    {
      "auxiliary_loss_clip": 0.01227495,
      "auxiliary_loss_mlp": 0.01055883,
      "balance_loss_clip": 1.06973875,
      "balance_loss_mlp": 1.03013396,
      "epoch": 0.05693672027656696,
      "flos": 16835641130880.0,
      "grad_norm": 1.8767368385453804,
      "language_loss": 0.78982449,
      "learning_rate": 3.992394109874529e-06,
      "loss": 0.81265831,
      "num_input_tokens_seen": 20238640,
      "step": 947,
      "time_per_iteration": 2.480041980743408
    },
    {
      "auxiliary_loss_clip": 0.01187302,
      "auxiliary_loss_mlp": 0.01067591,
      "balance_loss_clip": 1.07060218,
      "balance_loss_mlp": 1.04150772,
      "epoch": 0.05699684352923493,
      "flos": 21389113370880.0,
      "grad_norm": 4.744162497354199,
      "language_loss": 0.85213029,
      "learning_rate": 3.9923601387700225e-06,
      "loss": 0.87467921,
      "num_input_tokens_seen": 20251025,
      "step": 948,
      "time_per_iteration": 2.522787570953369
    },
    {
      "auxiliary_loss_clip": 0.01225527,
      "auxiliary_loss_mlp": 0.01063168,
      "balance_loss_clip": 1.07010472,
      "balance_loss_mlp": 1.03602374,
      "epoch": 0.057056966781902904,
      "flos": 15560309767680.0,
      "grad_norm": 2.4978031255856035,
      "language_loss": 0.87184274,
      "learning_rate": 3.992326092115019e-06,
      "loss": 0.89472973,
      "num_input_tokens_seen": 20269775,
      "step": 949,
      "time_per_iteration": 2.484736680984497
    },
    {
      "auxiliary_loss_clip": 0.01206128,
      "auxiliary_loss_mlp": 0.01065205,
      "balance_loss_clip": 1.06985974,
      "balance_loss_mlp": 1.04169714,
      "epoch": 0.05711709003457087,
      "flos": 19937856170880.0,
      "grad_norm": 2.1560561914792413,
      "language_loss": 0.78873062,
      "learning_rate": 3.992291969910811e-06,
      "loss": 0.81144392,
      "num_input_tokens_seen": 20287715,
      "step": 950,
      "time_per_iteration": 2.4868617057800293
    },
    {
      "auxiliary_loss_clip": 0.01194313,
      "auxiliary_loss_mlp": 0.01066897,
      "balance_loss_clip": 1.06735468,
      "balance_loss_mlp": 1.04267406,
      "epoch": 0.05717721328723884,
      "flos": 30332701774080.0,
      "grad_norm": 1.9471889257276236,
      "language_loss": 0.82429546,
      "learning_rate": 3.992257772158691e-06,
      "loss": 0.84690756,
      "num_input_tokens_seen": 20307070,
      "step": 951,
      "time_per_iteration": 2.6099483966827393
    },
    {
      "auxiliary_loss_clip": 0.01178076,
      "auxiliary_loss_mlp": 0.0105885,
      "balance_loss_clip": 1.06287479,
      "balance_loss_mlp": 1.03256392,
      "epoch": 0.05723733653990681,
      "flos": 23654358627840.0,
      "grad_norm": 2.4020693099392014,
      "language_loss": 0.86390215,
      "learning_rate": 3.992223498859958e-06,
      "loss": 0.88627142,
      "num_input_tokens_seen": 20324945,
      "step": 952,
      "time_per_iteration": 2.553435802459717
    },
    {
      "auxiliary_loss_clip": 0.01204458,
      "auxiliary_loss_mlp": 0.01061521,
      "balance_loss_clip": 1.06802976,
      "balance_loss_mlp": 1.03349531,
      "epoch": 0.05729745979257478,
      "flos": 22055759838720.0,
      "grad_norm": 2.2137311881131576,
      "language_loss": 0.79160559,
      "learning_rate": 3.9921891500159084e-06,
      "loss": 0.81426537,
      "num_input_tokens_seen": 20346135,
      "step": 953,
      "time_per_iteration": 2.5988008975982666
    },
    {
      "auxiliary_loss_clip": 0.01198522,
      "auxiliary_loss_mlp": 0.01063322,
      "balance_loss_clip": 1.06947112,
      "balance_loss_mlp": 1.03789473,
      "epoch": 0.05735758304524275,
      "flos": 19604353368960.0,
      "grad_norm": 1.9838836779139906,
      "language_loss": 0.86833405,
      "learning_rate": 3.992154725627848e-06,
      "loss": 0.89095247,
      "num_input_tokens_seen": 20364450,
      "step": 954,
      "time_per_iteration": 2.5365943908691406
    },
    {
      "auxiliary_loss_clip": 0.01219935,
      "auxiliary_loss_mlp": 0.01062331,
      "balance_loss_clip": 1.07086492,
      "balance_loss_mlp": 1.03745222,
      "epoch": 0.057417706297910716,
      "flos": 19099018880640.0,
      "grad_norm": 2.5886432241836483,
      "language_loss": 0.88173354,
      "learning_rate": 3.9921202256970804e-06,
      "loss": 0.90455616,
      "num_input_tokens_seen": 20383500,
      "step": 955,
      "time_per_iteration": 2.514401912689209
    },
    {
      "auxiliary_loss_clip": 0.01196086,
      "auxiliary_loss_mlp": 0.01061478,
      "balance_loss_clip": 1.07444859,
      "balance_loss_mlp": 1.03609824,
      "epoch": 0.05747782955057869,
      "flos": 16654507822080.0,
      "grad_norm": 2.292065801486436,
      "language_loss": 0.89249617,
      "learning_rate": 3.992085650224914e-06,
      "loss": 0.91507185,
      "num_input_tokens_seen": 20400295,
      "step": 956,
      "time_per_iteration": 2.5152671337127686
    },
    {
      "auxiliary_loss_clip": 0.01177877,
      "auxiliary_loss_mlp": 0.01054682,
      "balance_loss_clip": 1.06748939,
      "balance_loss_mlp": 1.0295881,
      "epoch": 0.05753795280324665,
      "flos": 14502058248960.0,
      "grad_norm": 2.7306624775191684,
      "language_loss": 0.75743902,
      "learning_rate": 3.99205099921266e-06,
      "loss": 0.77976459,
      "num_input_tokens_seen": 20419085,
      "step": 957,
      "time_per_iteration": 2.583271026611328
    },
    {
      "auxiliary_loss_clip": 0.0118643,
      "auxiliary_loss_mlp": 0.01066111,
      "balance_loss_clip": 1.06577969,
      "balance_loss_mlp": 1.03917027,
      "epoch": 0.057598076055914625,
      "flos": 18076318848000.0,
      "grad_norm": 2.2018324367838487,
      "language_loss": 0.79883564,
      "learning_rate": 3.992016272661633e-06,
      "loss": 0.82136106,
      "num_input_tokens_seen": 20437465,
      "step": 958,
      "time_per_iteration": 3.997917413711548
    },
    {
      "auxiliary_loss_clip": 0.01195151,
      "auxiliary_loss_mlp": 0.01057337,
      "balance_loss_clip": 1.06888437,
      "balance_loss_mlp": 1.03424597,
      "epoch": 0.0576581993085826,
      "flos": 22124600254080.0,
      "grad_norm": 2.576731120965464,
      "language_loss": 0.88333404,
      "learning_rate": 3.99198147057315e-06,
      "loss": 0.90585899,
      "num_input_tokens_seen": 20456235,
      "step": 959,
      "time_per_iteration": 4.038728952407837
    },
    {
      "auxiliary_loss_clip": 0.01181518,
      "auxiliary_loss_mlp": 0.01061272,
      "balance_loss_clip": 1.06792104,
      "balance_loss_mlp": 1.03691745,
      "epoch": 0.05771832256125056,
      "flos": 33181746779520.0,
      "grad_norm": 1.9840624198998669,
      "language_loss": 0.78611934,
      "learning_rate": 3.991946592948529e-06,
      "loss": 0.80854726,
      "num_input_tokens_seen": 20476825,
      "step": 960,
      "time_per_iteration": 2.6978962421417236
    },
    {
      "auxiliary_loss_clip": 0.01147103,
      "auxiliary_loss_mlp": 0.0106759,
      "balance_loss_clip": 1.06781864,
      "balance_loss_mlp": 1.04155433,
      "epoch": 0.057778445813918534,
      "flos": 24170143973760.0,
      "grad_norm": 2.9027025976053795,
      "language_loss": 0.93228149,
      "learning_rate": 3.991911639789094e-06,
      "loss": 0.95442843,
      "num_input_tokens_seen": 20496965,
      "step": 961,
      "time_per_iteration": 2.6799280643463135
    },
    {
      "auxiliary_loss_clip": 0.01190444,
      "auxiliary_loss_mlp": 0.01065825,
      "balance_loss_clip": 1.06575835,
      "balance_loss_mlp": 1.03943253,
      "epoch": 0.0578385690665865,
      "flos": 29643037666560.0,
      "grad_norm": 1.9922293158632958,
      "language_loss": 0.67869675,
      "learning_rate": 3.991876611096169e-06,
      "loss": 0.70125949,
      "num_input_tokens_seen": 20518035,
      "step": 962,
      "time_per_iteration": 2.6245076656341553
    },
    {
      "auxiliary_loss_clip": 0.0115926,
      "auxiliary_loss_mlp": 0.01065757,
      "balance_loss_clip": 1.05397344,
      "balance_loss_mlp": 1.04215336,
      "epoch": 0.05789869231925447,
      "flos": 20885430908160.0,
      "grad_norm": 2.9175803394960362,
      "language_loss": 0.885342,
      "learning_rate": 3.991841506871084e-06,
      "loss": 0.90759218,
      "num_input_tokens_seen": 20534740,
      "step": 963,
      "time_per_iteration": 2.563469171524048
    },
    {
      "auxiliary_loss_clip": 0.01194357,
      "auxiliary_loss_mlp": 0.01057861,
      "balance_loss_clip": 1.07305384,
      "balance_loss_mlp": 1.03317273,
      "epoch": 0.057958815571922444,
      "flos": 26031106679040.0,
      "grad_norm": 2.858711691247023,
      "language_loss": 0.84866524,
      "learning_rate": 3.99180632711517e-06,
      "loss": 0.87118739,
      "num_input_tokens_seen": 20553485,
      "step": 964,
      "time_per_iteration": 5.440260648727417
    },
    {
      "auxiliary_loss_clip": 0.01201285,
      "auxiliary_loss_mlp": 0.01067832,
      "balance_loss_clip": 1.07195854,
      "balance_loss_mlp": 1.04263127,
      "epoch": 0.05801893882459041,
      "flos": 18077683564800.0,
      "grad_norm": 6.0364997886788965,
      "language_loss": 0.77651596,
      "learning_rate": 3.99177107182976e-06,
      "loss": 0.79920709,
      "num_input_tokens_seen": 20572155,
      "step": 965,
      "time_per_iteration": 2.607905864715576
    },
    {
      "auxiliary_loss_clip": 0.01229002,
      "auxiliary_loss_mlp": 0.0106954,
      "balance_loss_clip": 1.1747005,
      "balance_loss_mlp": 1.04432702,
      "epoch": 0.05807906207725838,
      "flos": 17748885444480.0,
      "grad_norm": 2.005634553913885,
      "language_loss": 0.81018603,
      "learning_rate": 3.99173574101619e-06,
      "loss": 0.83317143,
      "num_input_tokens_seen": 20590395,
      "step": 966,
      "time_per_iteration": 2.6351230144500732
    },
    {
      "auxiliary_loss_clip": 0.01208459,
      "auxiliary_loss_mlp": 0.01060168,
      "balance_loss_clip": 1.06918192,
      "balance_loss_mlp": 1.03759027,
      "epoch": 0.058139185329926346,
      "flos": 18040372312320.0,
      "grad_norm": 1.8928077223393354,
      "language_loss": 0.76280046,
      "learning_rate": 3.9917003346758035e-06,
      "loss": 0.78548682,
      "num_input_tokens_seen": 20608435,
      "step": 967,
      "time_per_iteration": 2.5184640884399414
    },
    {
      "auxiliary_loss_clip": 0.01121109,
      "auxiliary_loss_mlp": 0.01090119,
      "balance_loss_clip": 1.05865812,
      "balance_loss_mlp": 1.08535051,
      "epoch": 0.05819930858259432,
      "flos": 62363297485440.0,
      "grad_norm": 0.8115790733219859,
      "language_loss": 0.57345188,
      "learning_rate": 3.991664852809939e-06,
      "loss": 0.59556425,
      "num_input_tokens_seen": 20668575,
      "step": 968,
      "time_per_iteration": 3.04945969581604
    },
    {
      "auxiliary_loss_clip": 0.0120065,
      "auxiliary_loss_mlp": 0.01055805,
      "balance_loss_clip": 1.07094705,
      "balance_loss_mlp": 1.03063989,
      "epoch": 0.05825943183526229,
      "flos": 19135360465920.0,
      "grad_norm": 2.1014976469854396,
      "language_loss": 0.82199919,
      "learning_rate": 3.991629295419945e-06,
      "loss": 0.84456372,
      "num_input_tokens_seen": 20687355,
      "step": 969,
      "time_per_iteration": 2.5850179195404053
    },
    {
      "auxiliary_loss_clip": 0.01216189,
      "auxiliary_loss_mlp": 0.00808659,
      "balance_loss_clip": 1.07261205,
      "balance_loss_mlp": 1.04277611,
      "epoch": 0.058319555087930255,
      "flos": 29022465369600.0,
      "grad_norm": 2.347194270274993,
      "language_loss": 0.77391684,
      "learning_rate": 3.991593662507167e-06,
      "loss": 0.79416531,
      "num_input_tokens_seen": 20705710,
      "step": 970,
      "time_per_iteration": 2.5915369987487793
    },
    {
      "auxiliary_loss_clip": 0.01193036,
      "auxiliary_loss_mlp": 0.01060076,
      "balance_loss_clip": 1.06819022,
      "balance_loss_mlp": 1.03426743,
      "epoch": 0.05837967834059823,
      "flos": 18879999701760.0,
      "grad_norm": 2.6318922983924216,
      "language_loss": 0.92249507,
      "learning_rate": 3.991557954072958e-06,
      "loss": 0.94502622,
      "num_input_tokens_seen": 20722405,
      "step": 971,
      "time_per_iteration": 2.585118055343628
    },
    {
      "auxiliary_loss_clip": 0.0119261,
      "auxiliary_loss_mlp": 0.0105587,
      "balance_loss_clip": 1.0672096,
      "balance_loss_mlp": 1.03218293,
      "epoch": 0.05843980159326619,
      "flos": 25703062744320.0,
      "grad_norm": 1.842121426587099,
      "language_loss": 0.85853374,
      "learning_rate": 3.991522170118673e-06,
      "loss": 0.88101852,
      "num_input_tokens_seen": 20741480,
      "step": 972,
      "time_per_iteration": 2.590047597885132
    },
    {
      "auxiliary_loss_clip": 0.01183313,
      "auxiliary_loss_mlp": 0.01062928,
      "balance_loss_clip": 1.07254517,
      "balance_loss_mlp": 1.03999245,
      "epoch": 0.058499924845934165,
      "flos": 25552129795200.0,
      "grad_norm": 2.0283360681900664,
      "language_loss": 0.87479758,
      "learning_rate": 3.991486310645667e-06,
      "loss": 0.89726001,
      "num_input_tokens_seen": 20759685,
      "step": 973,
      "time_per_iteration": 2.658601760864258
    },
    {
      "auxiliary_loss_clip": 0.01217088,
      "auxiliary_loss_mlp": 0.00811407,
      "balance_loss_clip": 1.07249939,
      "balance_loss_mlp": 1.04588127,
      "epoch": 0.05856004809860214,
      "flos": 16436171001600.0,
      "grad_norm": 2.116011697756705,
      "language_loss": 0.7512033,
      "learning_rate": 3.991450375655301e-06,
      "loss": 0.77148825,
      "num_input_tokens_seen": 20778180,
      "step": 974,
      "time_per_iteration": 2.5226144790649414
    },
    {
      "auxiliary_loss_clip": 0.01211687,
      "auxiliary_loss_mlp": 0.00806364,
      "balance_loss_clip": 1.07348013,
      "balance_loss_mlp": 1.03797174,
      "epoch": 0.0586201713512701,
      "flos": 39458824116480.0,
      "grad_norm": 7.1136996950571305,
      "language_loss": 0.76723361,
      "learning_rate": 3.991414365148936e-06,
      "loss": 0.78741407,
      "num_input_tokens_seen": 20802705,
      "step": 975,
      "time_per_iteration": 2.6912994384765625
    },
    {
      "auxiliary_loss_clip": 0.01226424,
      "auxiliary_loss_mlp": 0.01065896,
      "balance_loss_clip": 1.07146168,
      "balance_loss_mlp": 1.04193461,
      "epoch": 0.058680294603938074,
      "flos": 23365170230400.0,
      "grad_norm": 2.7474774104515616,
      "language_loss": 0.77054274,
      "learning_rate": 3.99137827912794e-06,
      "loss": 0.79346591,
      "num_input_tokens_seen": 20822540,
      "step": 976,
      "time_per_iteration": 2.510050058364868
    },
    {
      "auxiliary_loss_clip": 0.01188474,
      "auxiliary_loss_mlp": 0.01078487,
      "balance_loss_clip": 1.06816506,
      "balance_loss_mlp": 1.05233264,
      "epoch": 0.05874041785660604,
      "flos": 32232017226240.0,
      "grad_norm": 2.0060845212821086,
      "language_loss": 0.87296009,
      "learning_rate": 3.991342117593679e-06,
      "loss": 0.89562976,
      "num_input_tokens_seen": 20844175,
      "step": 977,
      "time_per_iteration": 2.6715548038482666
    },
    {
      "auxiliary_loss_clip": 0.01195426,
      "auxiliary_loss_mlp": 0.0106558,
      "balance_loss_clip": 1.07209563,
      "balance_loss_mlp": 1.04041481,
      "epoch": 0.05880054110927401,
      "flos": 22310043194880.0,
      "grad_norm": 1.5162413927943732,
      "language_loss": 0.792575,
      "learning_rate": 3.991305880547527e-06,
      "loss": 0.81518501,
      "num_input_tokens_seen": 20864730,
      "step": 978,
      "time_per_iteration": 2.577559471130371
    },
    {
      "auxiliary_loss_clip": 0.01207888,
      "auxiliary_loss_mlp": 0.01076532,
      "balance_loss_clip": 1.17905569,
      "balance_loss_mlp": 1.05031765,
      "epoch": 0.05886066436194198,
      "flos": 27380450016000.0,
      "grad_norm": 2.0463819775454537,
      "language_loss": 0.80693382,
      "learning_rate": 3.991269567990855e-06,
      "loss": 0.82977808,
      "num_input_tokens_seen": 20885200,
      "step": 979,
      "time_per_iteration": 3.206923007965088
    },
    {
      "auxiliary_loss_clip": 0.01099098,
      "auxiliary_loss_mlp": 0.0103787,
      "balance_loss_clip": 1.04910827,
      "balance_loss_mlp": 1.03343523,
      "epoch": 0.05892078761460995,
      "flos": 59584493525760.0,
      "grad_norm": 0.9469908296516281,
      "language_loss": 0.59043694,
      "learning_rate": 3.9912331799250415e-06,
      "loss": 0.61180663,
      "num_input_tokens_seen": 20940325,
      "step": 980,
      "time_per_iteration": 3.4336395263671875
    },
    {
      "auxiliary_loss_clip": 0.01219429,
      "auxiliary_loss_mlp": 0.01066015,
      "balance_loss_clip": 1.0707798,
      "balance_loss_mlp": 1.04025424,
      "epoch": 0.05898091086727792,
      "flos": 15414081500160.0,
      "grad_norm": 2.282963099956916,
      "language_loss": 0.86809731,
      "learning_rate": 3.9911967163514665e-06,
      "loss": 0.89095175,
      "num_input_tokens_seen": 20958220,
      "step": 981,
      "time_per_iteration": 2.4974305629730225
    },
    {
      "auxiliary_loss_clip": 0.0120085,
      "auxiliary_loss_mlp": 0.01055677,
      "balance_loss_clip": 1.07026434,
      "balance_loss_mlp": 1.03235948,
      "epoch": 0.059041034119945886,
      "flos": 23655328295040.0,
      "grad_norm": 2.457992680729703,
      "language_loss": 0.79750216,
      "learning_rate": 3.991160177271513e-06,
      "loss": 0.82006741,
      "num_input_tokens_seen": 20978920,
      "step": 982,
      "time_per_iteration": 2.5989937782287598
    },
    {
      "auxiliary_loss_clip": 0.01261767,
      "auxiliary_loss_mlp": 0.01061672,
      "balance_loss_clip": 1.17790926,
      "balance_loss_mlp": 1.03754449,
      "epoch": 0.05910115737261386,
      "flos": 24754087376640.0,
      "grad_norm": 2.2154336956594216,
      "language_loss": 0.84149545,
      "learning_rate": 3.9911235626865654e-06,
      "loss": 0.86472988,
      "num_input_tokens_seen": 20999490,
      "step": 983,
      "time_per_iteration": 2.592073917388916
    },
    {
      "auxiliary_loss_clip": 0.01206933,
      "auxiliary_loss_mlp": 0.01060914,
      "balance_loss_clip": 1.0672071,
      "balance_loss_mlp": 1.03745413,
      "epoch": 0.05916128062528183,
      "flos": 11728749070080.0,
      "grad_norm": 1.9993185416718362,
      "language_loss": 0.8466599,
      "learning_rate": 3.9910868725980125e-06,
      "loss": 0.86933845,
      "num_input_tokens_seen": 21017865,
      "step": 984,
      "time_per_iteration": 2.499333620071411
    },
    {
      "auxiliary_loss_clip": 0.01206577,
      "auxiliary_loss_mlp": 0.01054064,
      "balance_loss_clip": 1.07903004,
      "balance_loss_mlp": 1.03094912,
      "epoch": 0.059221403877949795,
      "flos": 21902995296000.0,
      "grad_norm": 4.879450703343181,
      "language_loss": 0.77561241,
      "learning_rate": 3.9910501070072465e-06,
      "loss": 0.79821885,
      "num_input_tokens_seen": 21035900,
      "step": 985,
      "time_per_iteration": 2.539445161819458
    },
    {
      "auxiliary_loss_clip": 0.01155959,
      "auxiliary_loss_mlp": 0.0106455,
      "balance_loss_clip": 1.06829143,
      "balance_loss_mlp": 1.04099405,
      "epoch": 0.05928152713061777,
      "flos": 20514580940160.0,
      "grad_norm": 2.052150881378456,
      "language_loss": 0.9066211,
      "learning_rate": 3.991013265915661e-06,
      "loss": 0.92882621,
      "num_input_tokens_seen": 21053235,
      "step": 986,
      "time_per_iteration": 2.58915638923645
    },
    {
      "auxiliary_loss_clip": 0.01210589,
      "auxiliary_loss_mlp": 0.0106358,
      "balance_loss_clip": 1.06605864,
      "balance_loss_mlp": 1.03712702,
      "epoch": 0.05934165038328574,
      "flos": 24495135252480.0,
      "grad_norm": 2.0737832676270753,
      "language_loss": 0.75812083,
      "learning_rate": 3.9909763493246525e-06,
      "loss": 0.78086257,
      "num_input_tokens_seen": 21073090,
      "step": 987,
      "time_per_iteration": 2.56268572807312
    },
    {
      "auxiliary_loss_clip": 0.01210099,
      "auxiliary_loss_mlp": 0.01055015,
      "balance_loss_clip": 1.06846976,
      "balance_loss_mlp": 1.03127992,
      "epoch": 0.059401773635953704,
      "flos": 38728041914880.0,
      "grad_norm": 2.2575136837372436,
      "language_loss": 0.71988893,
      "learning_rate": 3.990939357235621e-06,
      "loss": 0.74254012,
      "num_input_tokens_seen": 21094895,
      "step": 988,
      "time_per_iteration": 2.659303665161133
    },
    {
      "auxiliary_loss_clip": 0.01080044,
      "auxiliary_loss_mlp": 0.01043521,
      "balance_loss_clip": 1.04851151,
      "balance_loss_mlp": 1.03863323,
      "epoch": 0.059461896888621676,
      "flos": 58023565125120.0,
      "grad_norm": 0.9643695972733062,
      "language_loss": 0.71181428,
      "learning_rate": 3.99090228964997e-06,
      "loss": 0.73304999,
      "num_input_tokens_seen": 21147555,
      "step": 989,
      "time_per_iteration": 3.0398290157318115
    },
    {
      "auxiliary_loss_clip": 0.01185191,
      "auxiliary_loss_mlp": 0.01068021,
      "balance_loss_clip": 1.06462073,
      "balance_loss_mlp": 1.04100776,
      "epoch": 0.05952202014128964,
      "flos": 22127760650880.0,
      "grad_norm": 3.0693769626661345,
      "language_loss": 0.78354836,
      "learning_rate": 3.990865146569105e-06,
      "loss": 0.80608046,
      "num_input_tokens_seen": 21167845,
      "step": 990,
      "time_per_iteration": 2.628450870513916
    },
    {
      "auxiliary_loss_clip": 0.01205954,
      "auxiliary_loss_mlp": 0.0105944,
      "balance_loss_clip": 1.07024026,
      "balance_loss_mlp": 1.03458452,
      "epoch": 0.059582143393957614,
      "flos": 20445776438400.0,
      "grad_norm": 2.1111790395980563,
      "language_loss": 0.86311287,
      "learning_rate": 3.990827927994434e-06,
      "loss": 0.8857668,
      "num_input_tokens_seen": 21185085,
      "step": 991,
      "time_per_iteration": 2.530669927597046
    },
    {
      "auxiliary_loss_clip": 0.0122565,
      "auxiliary_loss_mlp": 0.0106215,
      "balance_loss_clip": 1.06936622,
      "balance_loss_mlp": 1.03766441,
      "epoch": 0.059642266646625586,
      "flos": 20594877793920.0,
      "grad_norm": 1.933745670090089,
      "language_loss": 0.77218014,
      "learning_rate": 3.9907906339273674e-06,
      "loss": 0.79505813,
      "num_input_tokens_seen": 21204230,
      "step": 992,
      "time_per_iteration": 2.483783483505249
    },
    {
      "auxiliary_loss_clip": 0.01155944,
      "auxiliary_loss_mlp": 0.01060697,
      "balance_loss_clip": 1.06729412,
      "balance_loss_mlp": 1.03755796,
      "epoch": 0.05970238989929355,
      "flos": 19352655792000.0,
      "grad_norm": 2.960335431540363,
      "language_loss": 0.75085944,
      "learning_rate": 3.9907532643693215e-06,
      "loss": 0.77302587,
      "num_input_tokens_seen": 21222655,
      "step": 993,
      "time_per_iteration": 2.6115572452545166
    },
    {
      "auxiliary_loss_clip": 0.01174158,
      "auxiliary_loss_mlp": 0.01077981,
      "balance_loss_clip": 1.07896256,
      "balance_loss_mlp": 1.05098021,
      "epoch": 0.05976251315196152,
      "flos": 30264040926720.0,
      "grad_norm": 1.9686551783607777,
      "language_loss": 0.78871083,
      "learning_rate": 3.990715819321712e-06,
      "loss": 0.81123221,
      "num_input_tokens_seen": 21242310,
      "step": 994,
      "time_per_iteration": 2.6749072074890137
    },
    {
      "auxiliary_loss_clip": 0.01224067,
      "auxiliary_loss_mlp": 0.01076125,
      "balance_loss_clip": 1.0720942,
      "balance_loss_mlp": 1.05193758,
      "epoch": 0.05982263640462949,
      "flos": 23185150243200.0,
      "grad_norm": 2.506593804803191,
      "language_loss": 0.80195391,
      "learning_rate": 3.99067829878596e-06,
      "loss": 0.82495588,
      "num_input_tokens_seen": 21261410,
      "step": 995,
      "time_per_iteration": 2.481417179107666
    },
    {
      "auxiliary_loss_clip": 0.01177044,
      "auxiliary_loss_mlp": 0.01063555,
      "balance_loss_clip": 1.06830716,
      "balance_loss_mlp": 1.03812742,
      "epoch": 0.05988275965729746,
      "flos": 27850879463040.0,
      "grad_norm": 1.9668162029316845,
      "language_loss": 0.86790198,
      "learning_rate": 3.990640702763487e-06,
      "loss": 0.89030802,
      "num_input_tokens_seen": 21280080,
      "step": 996,
      "time_per_iteration": 2.625706672668457
    },
    {
      "auxiliary_loss_clip": 0.01189764,
      "auxiliary_loss_mlp": 0.01072742,
      "balance_loss_clip": 1.0757277,
      "balance_loss_mlp": 1.04431021,
      "epoch": 0.05994288290996543,
      "flos": 24680003575680.0,
      "grad_norm": 3.0767806845943197,
      "language_loss": 0.88084173,
      "learning_rate": 3.990603031255718e-06,
      "loss": 0.90346682,
      "num_input_tokens_seen": 21296765,
      "step": 997,
      "time_per_iteration": 3.9969847202301025
    },
    {
      "auxiliary_loss_clip": 0.01157596,
      "auxiliary_loss_mlp": 0.01031384,
      "balance_loss_clip": 1.15298545,
      "balance_loss_mlp": 1.02647305,
      "epoch": 0.0600030061626334,
      "flos": 69929568835200.0,
      "grad_norm": 1.02057964995631,
      "language_loss": 0.75460589,
      "learning_rate": 3.990565284264083e-06,
      "loss": 0.7764957,
      "num_input_tokens_seen": 21363345,
      "step": 998,
      "time_per_iteration": 4.631951808929443
    },
    {
      "auxiliary_loss_clip": 0.01183423,
      "auxiliary_loss_mlp": 0.01062806,
      "balance_loss_clip": 1.07864499,
      "balance_loss_mlp": 1.03776038,
      "epoch": 0.06006312941530137,
      "flos": 26540140268160.0,
      "grad_norm": 1.8757349560470702,
      "language_loss": 0.75966203,
      "learning_rate": 3.990527461790013e-06,
      "loss": 0.78212428,
      "num_input_tokens_seen": 21385290,
      "step": 999,
      "time_per_iteration": 2.6595571041107178
    },
    {
      "auxiliary_loss_clip": 0.01204512,
      "auxiliary_loss_mlp": 0.01056483,
      "balance_loss_clip": 1.0655539,
      "balance_loss_mlp": 1.03171158,
      "epoch": 0.060123252667969335,
      "flos": 27344000689920.0,
      "grad_norm": 1.7916408498364047,
      "language_loss": 0.82484126,
      "learning_rate": 3.990489563834943e-06,
      "loss": 0.84745121,
      "num_input_tokens_seen": 21407625,
      "step": 1000,
      "time_per_iteration": 2.5711843967437744
    },
    {
      "auxiliary_loss_clip": 0.0119475,
      "auxiliary_loss_mlp": 0.01070673,
      "balance_loss_clip": 1.07067955,
      "balance_loss_mlp": 1.0454843,
      "epoch": 0.06018337592063731,
      "flos": 27016710940800.0,
      "grad_norm": 2.171629907366366,
      "language_loss": 0.86091167,
      "learning_rate": 3.990451590400309e-06,
      "loss": 0.8835659,
      "num_input_tokens_seen": 21426835,
      "step": 1001,
      "time_per_iteration": 2.630709409713745
    },
    {
      "auxiliary_loss_clip": 0.01197385,
      "auxiliary_loss_mlp": 0.01055398,
      "balance_loss_clip": 1.06689107,
      "balance_loss_mlp": 1.03180695,
      "epoch": 0.06024349917330528,
      "flos": 25592960580480.0,
      "grad_norm": 2.1489276183072756,
      "language_loss": 0.73945892,
      "learning_rate": 3.990413541487551e-06,
      "loss": 0.76198673,
      "num_input_tokens_seen": 21444920,
      "step": 1002,
      "time_per_iteration": 4.0176098346710205
    },
    {
      "auxiliary_loss_clip": 0.01221868,
      "auxiliary_loss_mlp": 0.01058949,
      "balance_loss_clip": 1.06987667,
      "balance_loss_mlp": 1.03488088,
      "epoch": 0.060303622425973244,
      "flos": 26133271937280.0,
      "grad_norm": 2.778439871108027,
      "language_loss": 0.75676692,
      "learning_rate": 3.990375417098112e-06,
      "loss": 0.77957511,
      "num_input_tokens_seen": 21463555,
      "step": 1003,
      "time_per_iteration": 2.58474063873291
    },
    {
      "auxiliary_loss_clip": 0.01200914,
      "auxiliary_loss_mlp": 0.0105964,
      "balance_loss_clip": 1.07945132,
      "balance_loss_mlp": 1.03578663,
      "epoch": 0.060363745678641216,
      "flos": 20377187418240.0,
      "grad_norm": 2.026264775996035,
      "language_loss": 0.70297629,
      "learning_rate": 3.990337217233437e-06,
      "loss": 0.72558177,
      "num_input_tokens_seen": 21481990,
      "step": 1004,
      "time_per_iteration": 2.5648927688598633
    },
    {
      "auxiliary_loss_clip": 0.01219348,
      "auxiliary_loss_mlp": 0.01070984,
      "balance_loss_clip": 1.07260704,
      "balance_loss_mlp": 1.04626,
      "epoch": 0.06042386893130918,
      "flos": 17749172753280.0,
      "grad_norm": 2.633538805133491,
      "language_loss": 0.83320796,
      "learning_rate": 3.990298941894976e-06,
      "loss": 0.85611123,
      "num_input_tokens_seen": 21500385,
      "step": 1005,
      "time_per_iteration": 2.4894700050354004
    },
    {
      "auxiliary_loss_clip": 0.01107081,
      "auxiliary_loss_mlp": 0.01010858,
      "balance_loss_clip": 1.04905033,
      "balance_loss_mlp": 1.00666142,
      "epoch": 0.06048399218397715,
      "flos": 68538496872960.0,
      "grad_norm": 0.89587268647693,
      "language_loss": 0.59039563,
      "learning_rate": 3.9902605910841794e-06,
      "loss": 0.61157501,
      "num_input_tokens_seen": 21561040,
      "step": 1006,
      "time_per_iteration": 3.185408592224121
    },
    {
      "auxiliary_loss_clip": 0.01190072,
      "auxiliary_loss_mlp": 0.01054852,
      "balance_loss_clip": 1.06259263,
      "balance_loss_mlp": 1.03024673,
      "epoch": 0.060544115436645125,
      "flos": 23258515772160.0,
      "grad_norm": 2.387725534704207,
      "language_loss": 0.7446155,
      "learning_rate": 3.990222164802503e-06,
      "loss": 0.76706469,
      "num_input_tokens_seen": 21580655,
      "step": 1007,
      "time_per_iteration": 2.550104856491089
    },
    {
      "auxiliary_loss_clip": 0.01199285,
      "auxiliary_loss_mlp": 0.01062672,
      "balance_loss_clip": 1.06600547,
      "balance_loss_mlp": 1.03832912,
      "epoch": 0.06060423868931309,
      "flos": 23878441624320.0,
      "grad_norm": 1.9477385052047491,
      "language_loss": 0.80524254,
      "learning_rate": 3.9901836630514006e-06,
      "loss": 0.82786202,
      "num_input_tokens_seen": 21599650,
      "step": 1008,
      "time_per_iteration": 2.576491594314575
    },
    {
      "auxiliary_loss_clip": 0.01243083,
      "auxiliary_loss_mlp": 0.01057452,
      "balance_loss_clip": 1.17467427,
      "balance_loss_mlp": 1.03330016,
      "epoch": 0.06066436194198106,
      "flos": 18728061171840.0,
      "grad_norm": 1.9072104984405902,
      "language_loss": 0.78198016,
      "learning_rate": 3.990145085832335e-06,
      "loss": 0.80498552,
      "num_input_tokens_seen": 21617550,
      "step": 1009,
      "time_per_iteration": 2.567054271697998
    },
    {
      "auxiliary_loss_clip": 0.01207874,
      "auxiliary_loss_mlp": 0.01058349,
      "balance_loss_clip": 1.08072305,
      "balance_loss_mlp": 1.03492451,
      "epoch": 0.06072448519464903,
      "flos": 24640465680000.0,
      "grad_norm": 1.7897389105736101,
      "language_loss": 0.92961478,
      "learning_rate": 3.990106433146769e-06,
      "loss": 0.952277,
      "num_input_tokens_seen": 21635865,
      "step": 1010,
      "time_per_iteration": 2.549360513687134
    },
    {
      "auxiliary_loss_clip": 0.0116607,
      "auxiliary_loss_mlp": 0.00826994,
      "balance_loss_clip": 1.07665706,
      "balance_loss_mlp": 1.07715666,
      "epoch": 0.060784608447317,
      "flos": 17378825575680.0,
      "grad_norm": 2.20160227278177,
      "language_loss": 0.72036451,
      "learning_rate": 3.9900677049961665e-06,
      "loss": 0.74029517,
      "num_input_tokens_seen": 21653945,
      "step": 1011,
      "time_per_iteration": 2.6423544883728027
    },
    {
      "auxiliary_loss_clip": 0.01191718,
      "auxiliary_loss_mlp": 0.01068049,
      "balance_loss_clip": 1.05802798,
      "balance_loss_mlp": 1.04100049,
      "epoch": 0.06084473169998497,
      "flos": 23692208584320.0,
      "grad_norm": 5.982863501147838,
      "language_loss": 0.87598157,
      "learning_rate": 3.990028901381999e-06,
      "loss": 0.89857924,
      "num_input_tokens_seen": 21671230,
      "step": 1012,
      "time_per_iteration": 2.534186840057373
    },
    {
      "auxiliary_loss_clip": 0.01191472,
      "auxiliary_loss_mlp": 0.01064868,
      "balance_loss_clip": 1.06487715,
      "balance_loss_mlp": 1.04120481,
      "epoch": 0.06090485495265294,
      "flos": 23546339452800.0,
      "grad_norm": 2.020234721219651,
      "language_loss": 0.76824385,
      "learning_rate": 3.989990022305734e-06,
      "loss": 0.79080725,
      "num_input_tokens_seen": 21691155,
      "step": 1013,
      "time_per_iteration": 2.562086343765259
    },
    {
      "auxiliary_loss_clip": 0.01211172,
      "auxiliary_loss_mlp": 0.00804544,
      "balance_loss_clip": 1.07038319,
      "balance_loss_mlp": 1.03521681,
      "epoch": 0.06096497820532091,
      "flos": 20339301548160.0,
      "grad_norm": 2.7763004057192417,
      "language_loss": 0.8597998,
      "learning_rate": 3.98995106776885e-06,
      "loss": 0.87995696,
      "num_input_tokens_seen": 21707405,
      "step": 1014,
      "time_per_iteration": 2.5221683979034424
    },
    {
      "auxiliary_loss_clip": 0.01219479,
      "auxiliary_loss_mlp": 0.01064693,
      "balance_loss_clip": 1.06985402,
      "balance_loss_mlp": 1.03822851,
      "epoch": 0.061025101457988874,
      "flos": 26939035779840.0,
      "grad_norm": 2.3211893848740623,
      "language_loss": 0.73360002,
      "learning_rate": 3.98991203777282e-06,
      "loss": 0.75644171,
      "num_input_tokens_seen": 21728090,
      "step": 1015,
      "time_per_iteration": 2.5620009899139404
    },
    {
      "auxiliary_loss_clip": 0.01190099,
      "auxiliary_loss_mlp": 0.01059671,
      "balance_loss_clip": 1.06643105,
      "balance_loss_mlp": 1.03550756,
      "epoch": 0.061085224710656846,
      "flos": 25375054723200.0,
      "grad_norm": 1.7278419064041828,
      "language_loss": 0.79242778,
      "learning_rate": 3.9898729323191275e-06,
      "loss": 0.81492543,
      "num_input_tokens_seen": 21747950,
      "step": 1016,
      "time_per_iteration": 2.5964720249176025
    },
    {
      "auxiliary_loss_clip": 0.01176857,
      "auxiliary_loss_mlp": 0.0105824,
      "balance_loss_clip": 1.06588602,
      "balance_loss_mlp": 1.0341959,
      "epoch": 0.06114534796332482,
      "flos": 24824759385600.0,
      "grad_norm": 2.2925296645510995,
      "language_loss": 0.75948983,
      "learning_rate": 3.989833751409254e-06,
      "loss": 0.7818408,
      "num_input_tokens_seen": 21767900,
      "step": 1017,
      "time_per_iteration": 2.631725788116455
    },
    {
      "auxiliary_loss_clip": 0.01191442,
      "auxiliary_loss_mlp": 0.01076628,
      "balance_loss_clip": 1.0714395,
      "balance_loss_mlp": 1.05117667,
      "epoch": 0.061205471215992784,
      "flos": 20631434860800.0,
      "grad_norm": 2.283524948878038,
      "language_loss": 0.85985792,
      "learning_rate": 3.989794495044685e-06,
      "loss": 0.88253856,
      "num_input_tokens_seen": 21787375,
      "step": 1018,
      "time_per_iteration": 2.572498083114624
    },
    {
      "auxiliary_loss_clip": 0.01239632,
      "auxiliary_loss_mlp": 0.01072175,
      "balance_loss_clip": 1.17450309,
      "balance_loss_mlp": 1.04565132,
      "epoch": 0.061265594468660756,
      "flos": 16508351381760.0,
      "grad_norm": 3.1387369909267524,
      "language_loss": 0.77724612,
      "learning_rate": 3.989755163226909e-06,
      "loss": 0.8003642,
      "num_input_tokens_seen": 21806275,
      "step": 1019,
      "time_per_iteration": 2.559656858444214
    },
    {
      "auxiliary_loss_clip": 0.01168654,
      "auxiliary_loss_mlp": 0.0105951,
      "balance_loss_clip": 1.0626719,
      "balance_loss_mlp": 1.0347141,
      "epoch": 0.06132571772132872,
      "flos": 26246211275520.0,
      "grad_norm": 1.9015181061021196,
      "language_loss": 0.84547591,
      "learning_rate": 3.989715755957418e-06,
      "loss": 0.86775762,
      "num_input_tokens_seen": 21826430,
      "step": 1020,
      "time_per_iteration": 2.66471004486084
    },
    {
      "auxiliary_loss_clip": 0.01201459,
      "auxiliary_loss_mlp": 0.01059094,
      "balance_loss_clip": 1.06926239,
      "balance_loss_mlp": 1.03174746,
      "epoch": 0.06138584097399669,
      "flos": 37414788768000.0,
      "grad_norm": 1.9328704699728074,
      "language_loss": 0.79373932,
      "learning_rate": 3.989676273237705e-06,
      "loss": 0.81634492,
      "num_input_tokens_seen": 21847800,
      "step": 1021,
      "time_per_iteration": 2.701507806777954
    },
    {
      "auxiliary_loss_clip": 0.01193468,
      "auxiliary_loss_mlp": 0.01058916,
      "balance_loss_clip": 1.06506681,
      "balance_loss_mlp": 1.03645647,
      "epoch": 0.061445964226664665,
      "flos": 17420661941760.0,
      "grad_norm": 2.202858582003627,
      "language_loss": 0.87798876,
      "learning_rate": 3.9896367150692705e-06,
      "loss": 0.90051258,
      "num_input_tokens_seen": 21863385,
      "step": 1022,
      "time_per_iteration": 2.520646810531616
    },
    {
      "auxiliary_loss_clip": 0.01256946,
      "auxiliary_loss_mlp": 0.01059836,
      "balance_loss_clip": 1.17252076,
      "balance_loss_mlp": 1.03574407,
      "epoch": 0.06150608747933263,
      "flos": 22600021691520.0,
      "grad_norm": 2.0719514898472906,
      "language_loss": 0.83114821,
      "learning_rate": 3.989597081453611e-06,
      "loss": 0.85431606,
      "num_input_tokens_seen": 21881880,
      "step": 1023,
      "time_per_iteration": 2.5900027751922607
    },
    {
      "auxiliary_loss_clip": 0.01120008,
      "auxiliary_loss_mlp": 0.01015397,
      "balance_loss_clip": 1.0508641,
      "balance_loss_mlp": 1.01139164,
      "epoch": 0.0615662107320006,
      "flos": 56741482005120.0,
      "grad_norm": 0.8969573229655349,
      "language_loss": 0.65092325,
      "learning_rate": 3.989557372392231e-06,
      "loss": 0.67227727,
      "num_input_tokens_seen": 21940550,
      "step": 1024,
      "time_per_iteration": 3.1315290927886963
    },
    {
      "auxiliary_loss_clip": 0.01161621,
      "auxiliary_loss_mlp": 0.01066837,
      "balance_loss_clip": 1.05619431,
      "balance_loss_mlp": 1.03986037,
      "epoch": 0.06162633398466857,
      "flos": 22564793427840.0,
      "grad_norm": 2.047035854799912,
      "language_loss": 0.88249445,
      "learning_rate": 3.989517587886636e-06,
      "loss": 0.90477902,
      "num_input_tokens_seen": 21958390,
      "step": 1025,
      "time_per_iteration": 2.5561535358428955
    },
    {
      "auxiliary_loss_clip": 0.01196872,
      "auxiliary_loss_mlp": 0.01057575,
      "balance_loss_clip": 1.06642127,
      "balance_loss_mlp": 1.03455544,
      "epoch": 0.06168645723733654,
      "flos": 25593104234880.0,
      "grad_norm": 6.10241395841338,
      "language_loss": 0.84530538,
      "learning_rate": 3.989477727938335e-06,
      "loss": 0.86784983,
      "num_input_tokens_seen": 21978625,
      "step": 1026,
      "time_per_iteration": 2.571310520172119
    },
    {
      "auxiliary_loss_clip": 0.01171062,
      "auxiliary_loss_mlp": 0.01067544,
      "balance_loss_clip": 1.07499468,
      "balance_loss_mlp": 1.04230785,
      "epoch": 0.06174658049000451,
      "flos": 15997917162240.0,
      "grad_norm": 1.8992643429412435,
      "language_loss": 0.82404929,
      "learning_rate": 3.989437792548839e-06,
      "loss": 0.84643531,
      "num_input_tokens_seen": 21996035,
      "step": 1027,
      "time_per_iteration": 2.578012466430664
    },
    {
      "auxiliary_loss_clip": 0.01167124,
      "auxiliary_loss_mlp": 0.01057718,
      "balance_loss_clip": 1.0709635,
      "balance_loss_mlp": 1.03316092,
      "epoch": 0.06180670374267248,
      "flos": 11285970117120.0,
      "grad_norm": 2.5755062592196,
      "language_loss": 0.84035152,
      "learning_rate": 3.989397781719663e-06,
      "loss": 0.86259997,
      "num_input_tokens_seen": 22011625,
      "step": 1028,
      "time_per_iteration": 2.5923354625701904
    },
    {
      "auxiliary_loss_clip": 0.01087213,
      "auxiliary_loss_mlp": 0.01006276,
      "balance_loss_clip": 1.04401517,
      "balance_loss_mlp": 1.002056,
      "epoch": 0.06186682699534045,
      "flos": 65130142216320.0,
      "grad_norm": 0.9386424605360532,
      "language_loss": 0.60446727,
      "learning_rate": 3.989357695452323e-06,
      "loss": 0.62540221,
      "num_input_tokens_seen": 22066035,
      "step": 1029,
      "time_per_iteration": 2.943061590194702
    },
    {
      "auxiliary_loss_clip": 0.01183189,
      "auxiliary_loss_mlp": 0.01063623,
      "balance_loss_clip": 1.06694174,
      "balance_loss_mlp": 1.03749275,
      "epoch": 0.061926950248008414,
      "flos": 21105742976640.0,
      "grad_norm": 2.416411101689549,
      "language_loss": 0.82904816,
      "learning_rate": 3.98931753374834e-06,
      "loss": 0.85151625,
      "num_input_tokens_seen": 22085015,
      "step": 1030,
      "time_per_iteration": 2.6117117404937744
    },
    {
      "auxiliary_loss_clip": 0.01224888,
      "auxiliary_loss_mlp": 0.01067935,
      "balance_loss_clip": 1.07139075,
      "balance_loss_mlp": 1.04329407,
      "epoch": 0.061987073500676386,
      "flos": 17748454481280.0,
      "grad_norm": 4.110198106305974,
      "language_loss": 0.79837704,
      "learning_rate": 3.989277296609237e-06,
      "loss": 0.82130527,
      "num_input_tokens_seen": 22102775,
      "step": 1031,
      "time_per_iteration": 2.470902442932129
    },
    {
      "auxiliary_loss_clip": 0.01193966,
      "auxiliary_loss_mlp": 0.01076398,
      "balance_loss_clip": 1.07010484,
      "balance_loss_mlp": 1.05001688,
      "epoch": 0.06204719675334436,
      "flos": 21836237869440.0,
      "grad_norm": 1.7198605640623903,
      "language_loss": 0.77513051,
      "learning_rate": 3.98923698403654e-06,
      "loss": 0.79783416,
      "num_input_tokens_seen": 22121680,
      "step": 1032,
      "time_per_iteration": 2.5531489849090576
    },
    {
      "auxiliary_loss_clip": 0.01202595,
      "auxiliary_loss_mlp": 0.01069016,
      "balance_loss_clip": 1.06348825,
      "balance_loss_mlp": 1.04395819,
      "epoch": 0.06210732000601232,
      "flos": 19353697286400.0,
      "grad_norm": 2.8219982169132094,
      "language_loss": 0.89283907,
      "learning_rate": 3.989196596031776e-06,
      "loss": 0.91555524,
      "num_input_tokens_seen": 22138155,
      "step": 1033,
      "time_per_iteration": 2.508654832839966
    },
    {
      "auxiliary_loss_clip": 0.01212642,
      "auxiliary_loss_mlp": 0.01058993,
      "balance_loss_clip": 1.06752968,
      "balance_loss_mlp": 1.03493631,
      "epoch": 0.062167443258680295,
      "flos": 24749382695040.0,
      "grad_norm": 2.7005893466601063,
      "language_loss": 0.85055178,
      "learning_rate": 3.989156132596479e-06,
      "loss": 0.87326813,
      "num_input_tokens_seen": 22157420,
      "step": 1034,
      "time_per_iteration": 2.545006513595581
    },
    {
      "auxiliary_loss_clip": 0.01188412,
      "auxiliary_loss_mlp": 0.01058543,
      "balance_loss_clip": 1.06625164,
      "balance_loss_mlp": 1.03371203,
      "epoch": 0.06222756651134827,
      "flos": 34458478773120.0,
      "grad_norm": 2.0278922021140775,
      "language_loss": 0.80742383,
      "learning_rate": 3.989115593732182e-06,
      "loss": 0.82989335,
      "num_input_tokens_seen": 22178620,
      "step": 1035,
      "time_per_iteration": 4.0297064781188965
    },
    {
      "auxiliary_loss_clip": 0.01168223,
      "auxiliary_loss_mlp": 0.01069993,
      "balance_loss_clip": 1.07188118,
      "balance_loss_mlp": 1.04281306,
      "epoch": 0.06228768976401623,
      "flos": 25666469763840.0,
      "grad_norm": 3.3885385627443125,
      "language_loss": 0.78227878,
      "learning_rate": 3.989074979440421e-06,
      "loss": 0.80466092,
      "num_input_tokens_seen": 22197125,
      "step": 1036,
      "time_per_iteration": 4.044435024261475
    },
    {
      "auxiliary_loss_clip": 0.01202217,
      "auxiliary_loss_mlp": 0.01066537,
      "balance_loss_clip": 1.0693171,
      "balance_loss_mlp": 1.04237366,
      "epoch": 0.062347813016684205,
      "flos": 25295619795840.0,
      "grad_norm": 1.794658920648856,
      "language_loss": 0.86704051,
      "learning_rate": 3.989034289722739e-06,
      "loss": 0.88972807,
      "num_input_tokens_seen": 22217575,
      "step": 1037,
      "time_per_iteration": 2.5914485454559326
    },
    {
      "auxiliary_loss_clip": 0.01204047,
      "auxiliary_loss_mlp": 0.01056904,
      "balance_loss_clip": 1.06681371,
      "balance_loss_mlp": 1.03102326,
      "epoch": 0.06240793626935217,
      "flos": 26907039740160.0,
      "grad_norm": 3.3720356162683984,
      "language_loss": 0.80837935,
      "learning_rate": 3.988993524580676e-06,
      "loss": 0.83098882,
      "num_input_tokens_seen": 22236840,
      "step": 1038,
      "time_per_iteration": 2.6048426628112793
    },
    {
      "auxiliary_loss_clip": 0.01145765,
      "auxiliary_loss_mlp": 0.0107656,
      "balance_loss_clip": 1.05513287,
      "balance_loss_mlp": 1.04828358,
      "epoch": 0.06246805952202014,
      "flos": 21615782146560.0,
      "grad_norm": 3.965360659677508,
      "language_loss": 0.85608822,
      "learning_rate": 3.98895268401578e-06,
      "loss": 0.87831151,
      "num_input_tokens_seen": 22256465,
      "step": 1039,
      "time_per_iteration": 2.67110276222229
    },
    {
      "auxiliary_loss_clip": 0.01198135,
      "auxiliary_loss_mlp": 0.01068964,
      "balance_loss_clip": 1.06665421,
      "balance_loss_mlp": 1.04356086,
      "epoch": 0.0625281827746881,
      "flos": 19311896833920.0,
      "grad_norm": 1.8227440915158006,
      "language_loss": 0.8106916,
      "learning_rate": 3.9889117680296e-06,
      "loss": 0.83336258,
      "num_input_tokens_seen": 22274025,
      "step": 1040,
      "time_per_iteration": 2.6068663597106934
    },
    {
      "auxiliary_loss_clip": 0.0122329,
      "auxiliary_loss_mlp": 0.01065123,
      "balance_loss_clip": 1.07256353,
      "balance_loss_mlp": 1.03946877,
      "epoch": 0.06258830602735609,
      "flos": 27745769289600.0,
      "grad_norm": 2.818621148230795,
      "language_loss": 0.69767642,
      "learning_rate": 3.988870776623685e-06,
      "loss": 0.72056055,
      "num_input_tokens_seen": 22292245,
      "step": 1041,
      "time_per_iteration": 3.932661533355713
    },
    {
      "auxiliary_loss_clip": 0.0122029,
      "auxiliary_loss_mlp": 0.01054424,
      "balance_loss_clip": 1.06687617,
      "balance_loss_mlp": 1.02885389,
      "epoch": 0.06264842928002405,
      "flos": 23222605150080.0,
      "grad_norm": 1.9892516945938106,
      "language_loss": 0.81552386,
      "learning_rate": 3.9888297097995905e-06,
      "loss": 0.83827102,
      "num_input_tokens_seen": 22311455,
      "step": 1042,
      "time_per_iteration": 3.93043851852417
    },
    {
      "auxiliary_loss_clip": 0.01218153,
      "auxiliary_loss_mlp": 0.01050677,
      "balance_loss_clip": 1.06817603,
      "balance_loss_mlp": 1.02702618,
      "epoch": 0.06270855253269202,
      "flos": 38399495189760.0,
      "grad_norm": 1.7645122934528015,
      "language_loss": 0.76377797,
      "learning_rate": 3.988788567558874e-06,
      "loss": 0.78646624,
      "num_input_tokens_seen": 22333750,
      "step": 1043,
      "time_per_iteration": 2.64691424369812
    },
    {
      "auxiliary_loss_clip": 0.01197614,
      "auxiliary_loss_mlp": 0.0106415,
      "balance_loss_clip": 1.06762719,
      "balance_loss_mlp": 1.03991449,
      "epoch": 0.06276867578535998,
      "flos": 22453542028800.0,
      "grad_norm": 2.007423210682907,
      "language_loss": 0.92504358,
      "learning_rate": 3.988747349903097e-06,
      "loss": 0.94766128,
      "num_input_tokens_seen": 22351940,
      "step": 1044,
      "time_per_iteration": 2.5594093799591064
    },
    {
      "auxiliary_loss_clip": 0.0119986,
      "auxiliary_loss_mlp": 0.01071861,
      "balance_loss_clip": 1.06418705,
      "balance_loss_mlp": 1.04686236,
      "epoch": 0.06282879903802796,
      "flos": 22930435923840.0,
      "grad_norm": 1.8799293467685745,
      "language_loss": 0.86093235,
      "learning_rate": 3.988706056833821e-06,
      "loss": 0.88364947,
      "num_input_tokens_seen": 22372085,
      "step": 1045,
      "time_per_iteration": 2.565340995788574
    },
    {
      "auxiliary_loss_clip": 0.01188895,
      "auxiliary_loss_mlp": 0.01062511,
      "balance_loss_clip": 1.0657934,
      "balance_loss_mlp": 1.03789473,
      "epoch": 0.06288892229069593,
      "flos": 34819237019520.0,
      "grad_norm": 2.2263567792633947,
      "language_loss": 0.78175974,
      "learning_rate": 3.9886646883526125e-06,
      "loss": 0.80427372,
      "num_input_tokens_seen": 22392020,
      "step": 1046,
      "time_per_iteration": 2.6921207904815674
    },
    {
      "auxiliary_loss_clip": 0.01203741,
      "auxiliary_loss_mlp": 0.0107082,
      "balance_loss_clip": 1.06820488,
      "balance_loss_mlp": 1.04714513,
      "epoch": 0.06294904554336389,
      "flos": 19427134642560.0,
      "grad_norm": 4.060171943681887,
      "language_loss": 0.77613354,
      "learning_rate": 3.988623244461039e-06,
      "loss": 0.79887915,
      "num_input_tokens_seen": 22411180,
      "step": 1047,
      "time_per_iteration": 2.5352025032043457
    },
    {
      "auxiliary_loss_clip": 0.01209618,
      "auxiliary_loss_mlp": 0.01060305,
      "balance_loss_clip": 1.07002759,
      "balance_loss_mlp": 1.0356046,
      "epoch": 0.06300916879603187,
      "flos": 40661867358720.0,
      "grad_norm": 4.920300748383546,
      "language_loss": 0.77172208,
      "learning_rate": 3.988581725160672e-06,
      "loss": 0.79442132,
      "num_input_tokens_seen": 22435105,
      "step": 1048,
      "time_per_iteration": 2.684413194656372
    },
    {
      "auxiliary_loss_clip": 0.0119128,
      "auxiliary_loss_mlp": 0.010663,
      "balance_loss_clip": 1.06998539,
      "balance_loss_mlp": 1.04125404,
      "epoch": 0.06306929204869983,
      "flos": 23804142341760.0,
      "grad_norm": 3.550967429240787,
      "language_loss": 0.77603805,
      "learning_rate": 3.988540130453087e-06,
      "loss": 0.79861379,
      "num_input_tokens_seen": 22452710,
      "step": 1049,
      "time_per_iteration": 2.5848257541656494
    },
    {
      "auxiliary_loss_clip": 0.01202964,
      "auxiliary_loss_mlp": 0.0105678,
      "balance_loss_clip": 1.06777334,
      "balance_loss_mlp": 1.03208041,
      "epoch": 0.0631294153013678,
      "flos": 18915802583040.0,
      "grad_norm": 3.607397545677041,
      "language_loss": 0.82718027,
      "learning_rate": 3.988498460339862e-06,
      "loss": 0.8497777,
      "num_input_tokens_seen": 22470175,
      "step": 1050,
      "time_per_iteration": 2.523500442504883
    },
    {
      "auxiliary_loss_clip": 0.01220408,
      "auxiliary_loss_mlp": 0.01063028,
      "balance_loss_clip": 1.07249928,
      "balance_loss_mlp": 1.03889978,
      "epoch": 0.06318953855403578,
      "flos": 24280174310400.0,
      "grad_norm": 2.6112443242256607,
      "language_loss": 0.77305663,
      "learning_rate": 3.988456714822575e-06,
      "loss": 0.79589105,
      "num_input_tokens_seen": 22490020,
      "step": 1051,
      "time_per_iteration": 2.5461220741271973
    },
    {
      "auxiliary_loss_clip": 0.01197116,
      "auxiliary_loss_mlp": 0.01068972,
      "balance_loss_clip": 1.071787,
      "balance_loss_mlp": 1.04398596,
      "epoch": 0.06324966180670374,
      "flos": 22528918719360.0,
      "grad_norm": 2.2949919161433647,
      "language_loss": 0.80194324,
      "learning_rate": 3.98841489390281e-06,
      "loss": 0.82460415,
      "num_input_tokens_seen": 22509685,
      "step": 1052,
      "time_per_iteration": 2.584480047225952
    },
    {
      "auxiliary_loss_clip": 0.01222693,
      "auxiliary_loss_mlp": 0.01058389,
      "balance_loss_clip": 1.07188749,
      "balance_loss_mlp": 1.03405833,
      "epoch": 0.06330978505937171,
      "flos": 15778107884160.0,
      "grad_norm": 3.913642491201759,
      "language_loss": 0.77650714,
      "learning_rate": 3.988372997582155e-06,
      "loss": 0.79931796,
      "num_input_tokens_seen": 22527905,
      "step": 1053,
      "time_per_iteration": 2.5122177600860596
    },
    {
      "auxiliary_loss_clip": 0.01199802,
      "auxiliary_loss_mlp": 0.00837031,
      "balance_loss_clip": 1.06940889,
      "balance_loss_mlp": 1.09944177,
      "epoch": 0.06336990831203967,
      "flos": 21471098163840.0,
      "grad_norm": 2.511052655569475,
      "language_loss": 0.84855127,
      "learning_rate": 3.988331025862195e-06,
      "loss": 0.86891961,
      "num_input_tokens_seen": 22546335,
      "step": 1054,
      "time_per_iteration": 2.6102490425109863
    },
    {
      "auxiliary_loss_clip": 0.0116997,
      "auxiliary_loss_mlp": 0.01063845,
      "balance_loss_clip": 1.05432224,
      "balance_loss_mlp": 1.03877497,
      "epoch": 0.06343003156470765,
      "flos": 18478877546880.0,
      "grad_norm": 2.0776474324760117,
      "language_loss": 0.85550886,
      "learning_rate": 3.9882889787445225e-06,
      "loss": 0.87784696,
      "num_input_tokens_seen": 22563885,
      "step": 1055,
      "time_per_iteration": 2.5325145721435547
    },
    {
      "auxiliary_loss_clip": 0.01161531,
      "auxiliary_loss_mlp": 0.01064731,
      "balance_loss_clip": 1.05427635,
      "balance_loss_mlp": 1.03997159,
      "epoch": 0.06349015481737562,
      "flos": 25154886309120.0,
      "grad_norm": 2.4658233574034476,
      "language_loss": 0.80740762,
      "learning_rate": 3.988246856230734e-06,
      "loss": 0.82967019,
      "num_input_tokens_seen": 22583035,
      "step": 1056,
      "time_per_iteration": 2.623706579208374
    },
    {
      "auxiliary_loss_clip": 0.01154617,
      "auxiliary_loss_mlp": 0.01059581,
      "balance_loss_clip": 1.06178474,
      "balance_loss_mlp": 1.032866,
      "epoch": 0.06355027807004358,
      "flos": 26871775562880.0,
      "grad_norm": 4.222766734071776,
      "language_loss": 0.81305873,
      "learning_rate": 3.988204658322426e-06,
      "loss": 0.83520067,
      "num_input_tokens_seen": 22605055,
      "step": 1057,
      "time_per_iteration": 2.8047425746917725
    },
    {
      "auxiliary_loss_clip": 0.01196525,
      "auxiliary_loss_mlp": 0.01061224,
      "balance_loss_clip": 1.15195322,
      "balance_loss_mlp": 1.03828776,
      "epoch": 0.06361040132271156,
      "flos": 21396691140480.0,
      "grad_norm": 3.6279079620434347,
      "language_loss": 0.83423674,
      "learning_rate": 3.988162385021196e-06,
      "loss": 0.85681427,
      "num_input_tokens_seen": 22623760,
      "step": 1058,
      "time_per_iteration": 2.8255083560943604
    },
    {
      "auxiliary_loss_clip": 0.01188552,
      "auxiliary_loss_mlp": 0.01063326,
      "balance_loss_clip": 1.06800032,
      "balance_loss_mlp": 1.03687358,
      "epoch": 0.06367052457537953,
      "flos": 25733765894400.0,
      "grad_norm": 2.4874884186023927,
      "language_loss": 0.87744606,
      "learning_rate": 3.988120036328651e-06,
      "loss": 0.89996487,
      "num_input_tokens_seen": 22643000,
      "step": 1059,
      "time_per_iteration": 2.6268696784973145
    },
    {
      "auxiliary_loss_clip": 0.01165884,
      "auxiliary_loss_mlp": 0.01064609,
      "balance_loss_clip": 1.0592078,
      "balance_loss_mlp": 1.03875279,
      "epoch": 0.0637306478280475,
      "flos": 17631420992640.0,
      "grad_norm": 2.2562275321471947,
      "language_loss": 0.91517717,
      "learning_rate": 3.988077612246394e-06,
      "loss": 0.93748206,
      "num_input_tokens_seen": 22660460,
      "step": 1060,
      "time_per_iteration": 2.599611520767212
    },
    {
      "auxiliary_loss_clip": 0.01190239,
      "auxiliary_loss_mlp": 0.01064788,
      "balance_loss_clip": 1.07670641,
      "balance_loss_mlp": 1.03888416,
      "epoch": 0.06379077108071547,
      "flos": 13662610427520.0,
      "grad_norm": 2.008884687205767,
      "language_loss": 0.86711103,
      "learning_rate": 3.988035112776035e-06,
      "loss": 0.88966131,
      "num_input_tokens_seen": 22679270,
      "step": 1061,
      "time_per_iteration": 2.5593855381011963
    },
    {
      "auxiliary_loss_clip": 0.01199466,
      "auxiliary_loss_mlp": 0.01060984,
      "balance_loss_clip": 1.06607914,
      "balance_loss_mlp": 1.03459072,
      "epoch": 0.06385089433338344,
      "flos": 28478849961600.0,
      "grad_norm": 2.1716157622158554,
      "language_loss": 0.77480567,
      "learning_rate": 3.987992537919185e-06,
      "loss": 0.79741013,
      "num_input_tokens_seen": 22699330,
      "step": 1062,
      "time_per_iteration": 2.634864091873169
    },
    {
      "auxiliary_loss_clip": 0.01181947,
      "auxiliary_loss_mlp": 0.01062999,
      "balance_loss_clip": 1.0666213,
      "balance_loss_mlp": 1.03919315,
      "epoch": 0.0639110175860514,
      "flos": 24311057028480.0,
      "grad_norm": 2.361444996122634,
      "language_loss": 0.86546463,
      "learning_rate": 3.987949887677459e-06,
      "loss": 0.88791406,
      "num_input_tokens_seen": 22717945,
      "step": 1063,
      "time_per_iteration": 2.621023654937744
    },
    {
      "auxiliary_loss_clip": 0.0121612,
      "auxiliary_loss_mlp": 0.01059545,
      "balance_loss_clip": 1.06560588,
      "balance_loss_mlp": 1.0348444,
      "epoch": 0.06397114083871938,
      "flos": 22090772620800.0,
      "grad_norm": 2.099732942199714,
      "language_loss": 0.80368745,
      "learning_rate": 3.9879071620524744e-06,
      "loss": 0.82644415,
      "num_input_tokens_seen": 22736790,
      "step": 1064,
      "time_per_iteration": 2.497345209121704
    },
    {
      "auxiliary_loss_clip": 0.0120656,
      "auxiliary_loss_mlp": 0.01066726,
      "balance_loss_clip": 1.0665164,
      "balance_loss_mlp": 1.04077411,
      "epoch": 0.06403126409138735,
      "flos": 19572824206080.0,
      "grad_norm": 2.3856589810131514,
      "language_loss": 0.84151387,
      "learning_rate": 3.987864361045851e-06,
      "loss": 0.86424673,
      "num_input_tokens_seen": 22754745,
      "step": 1065,
      "time_per_iteration": 2.5229530334472656
    },
    {
      "auxiliary_loss_clip": 0.01175216,
      "auxiliary_loss_mlp": 0.01054766,
      "balance_loss_clip": 1.06935275,
      "balance_loss_mlp": 1.0317347,
      "epoch": 0.06409138734405531,
      "flos": 40807413267840.0,
      "grad_norm": 3.4399674580425756,
      "language_loss": 0.68509996,
      "learning_rate": 3.987821484659211e-06,
      "loss": 0.70739979,
      "num_input_tokens_seen": 22776780,
      "step": 1066,
      "time_per_iteration": 2.7499170303344727
    },
    {
      "auxiliary_loss_clip": 0.01219783,
      "auxiliary_loss_mlp": 0.01072257,
      "balance_loss_clip": 1.07190156,
      "balance_loss_mlp": 1.04640031,
      "epoch": 0.06415151059672328,
      "flos": 20441610460800.0,
      "grad_norm": 3.88002229464114,
      "language_loss": 0.90236521,
      "learning_rate": 3.987778532894181e-06,
      "loss": 0.92528564,
      "num_input_tokens_seen": 22793915,
      "step": 1067,
      "time_per_iteration": 2.5137240886688232
    },
    {
      "auxiliary_loss_clip": 0.0119281,
      "auxiliary_loss_mlp": 0.01064272,
      "balance_loss_clip": 1.06861818,
      "balance_loss_mlp": 1.0413835,
      "epoch": 0.06421163384939126,
      "flos": 18072045129600.0,
      "grad_norm": 2.2452089177674397,
      "language_loss": 0.83321106,
      "learning_rate": 3.987735505752391e-06,
      "loss": 0.85578185,
      "num_input_tokens_seen": 22812670,
      "step": 1068,
      "time_per_iteration": 2.559892177581787
    },
    {
      "auxiliary_loss_clip": 0.01190157,
      "auxiliary_loss_mlp": 0.01060091,
      "balance_loss_clip": 1.0733515,
      "balance_loss_mlp": 1.0370003,
      "epoch": 0.06427175710205922,
      "flos": 25119442563840.0,
      "grad_norm": 3.286705454194583,
      "language_loss": 0.89712274,
      "learning_rate": 3.987692403235471e-06,
      "loss": 0.91962522,
      "num_input_tokens_seen": 22832440,
      "step": 1069,
      "time_per_iteration": 2.6122753620147705
    },
    {
      "auxiliary_loss_clip": 0.01246881,
      "auxiliary_loss_mlp": 0.01075869,
      "balance_loss_clip": 1.16698861,
      "balance_loss_mlp": 1.0506686,
      "epoch": 0.06433188035472719,
      "flos": 17380549428480.0,
      "grad_norm": 2.882622116157054,
      "language_loss": 0.95839447,
      "learning_rate": 3.987649225345056e-06,
      "loss": 0.98162192,
      "num_input_tokens_seen": 22845495,
      "step": 1070,
      "time_per_iteration": 2.5294783115386963
    },
    {
      "auxiliary_loss_clip": 0.01148763,
      "auxiliary_loss_mlp": 0.01055262,
      "balance_loss_clip": 1.06830287,
      "balance_loss_mlp": 1.02982247,
      "epoch": 0.06439200360739517,
      "flos": 23546267625600.0,
      "grad_norm": 1.8550227522505478,
      "language_loss": 0.88202214,
      "learning_rate": 3.987605972082782e-06,
      "loss": 0.90406233,
      "num_input_tokens_seen": 22865390,
      "step": 1071,
      "time_per_iteration": 2.6722991466522217
    },
    {
      "auxiliary_loss_clip": 0.01160253,
      "auxiliary_loss_mlp": 0.01052592,
      "balance_loss_clip": 1.06161022,
      "balance_loss_mlp": 1.02858365,
      "epoch": 0.06445212686006313,
      "flos": 21979772616960.0,
      "grad_norm": 1.7050556204732423,
      "language_loss": 0.75824261,
      "learning_rate": 3.987562643450292e-06,
      "loss": 0.78037107,
      "num_input_tokens_seen": 22885495,
      "step": 1072,
      "time_per_iteration": 2.589935064315796
    },
    {
      "auxiliary_loss_clip": 0.01178975,
      "auxiliary_loss_mlp": 0.01064652,
      "balance_loss_clip": 1.06774855,
      "balance_loss_mlp": 1.03840184,
      "epoch": 0.0645122501127311,
      "flos": 25921291824000.0,
      "grad_norm": 1.9482781333518797,
      "language_loss": 0.80642629,
      "learning_rate": 3.987519239449226e-06,
      "loss": 0.82886261,
      "num_input_tokens_seen": 22904845,
      "step": 1073,
      "time_per_iteration": 2.6098790168762207
    },
    {
      "auxiliary_loss_clip": 0.01197722,
      "auxiliary_loss_mlp": 0.01056891,
      "balance_loss_clip": 1.06617081,
      "balance_loss_mlp": 1.03389502,
      "epoch": 0.06457237336539907,
      "flos": 25626034028160.0,
      "grad_norm": 1.751247334258551,
      "language_loss": 0.80249214,
      "learning_rate": 3.987475760081233e-06,
      "loss": 0.82503831,
      "num_input_tokens_seen": 22925940,
      "step": 1074,
      "time_per_iteration": 3.993777275085449
    },
    {
      "auxiliary_loss_clip": 0.01177586,
      "auxiliary_loss_mlp": 0.01063461,
      "balance_loss_clip": 1.06528497,
      "balance_loss_mlp": 1.03914261,
      "epoch": 0.06463249661806704,
      "flos": 19463979018240.0,
      "grad_norm": 2.156404738549543,
      "language_loss": 0.79318714,
      "learning_rate": 3.987432205347958e-06,
      "loss": 0.81559765,
      "num_input_tokens_seen": 22944375,
      "step": 1075,
      "time_per_iteration": 3.9172990322113037
    },
    {
      "auxiliary_loss_clip": 0.01167003,
      "auxiliary_loss_mlp": 0.01061363,
      "balance_loss_clip": 1.05438328,
      "balance_loss_mlp": 1.03814137,
      "epoch": 0.064692619870735,
      "flos": 24498044254080.0,
      "grad_norm": 2.467688460909924,
      "language_loss": 0.8743701,
      "learning_rate": 3.987388575251055e-06,
      "loss": 0.89665377,
      "num_input_tokens_seen": 22959145,
      "step": 1076,
      "time_per_iteration": 2.571411609649658
    },
    {
      "auxiliary_loss_clip": 0.01196198,
      "auxiliary_loss_mlp": 0.01055286,
      "balance_loss_clip": 1.07145369,
      "balance_loss_mlp": 1.03234971,
      "epoch": 0.06475274312340297,
      "flos": 17018677860480.0,
      "grad_norm": 1.9575562840587992,
      "language_loss": 0.81004441,
      "learning_rate": 3.98734486979218e-06,
      "loss": 0.83255935,
      "num_input_tokens_seen": 22978100,
      "step": 1077,
      "time_per_iteration": 2.522670030593872
    },
    {
      "auxiliary_loss_clip": 0.01200342,
      "auxiliary_loss_mlp": 0.01063584,
      "balance_loss_clip": 1.06635761,
      "balance_loss_mlp": 1.03888345,
      "epoch": 0.06481286637607095,
      "flos": 24572379450240.0,
      "grad_norm": 2.674507698973867,
      "language_loss": 0.91557676,
      "learning_rate": 3.987301088972986e-06,
      "loss": 0.93821603,
      "num_input_tokens_seen": 22997285,
      "step": 1078,
      "time_per_iteration": 2.5893747806549072
    },
    {
      "auxiliary_loss_clip": 0.01225489,
      "auxiliary_loss_mlp": 0.0106041,
      "balance_loss_clip": 1.07109666,
      "balance_loss_mlp": 1.03641284,
      "epoch": 0.06487298962873891,
      "flos": 21105635235840.0,
      "grad_norm": 2.7377853121534392,
      "language_loss": 0.79086232,
      "learning_rate": 3.987257232795137e-06,
      "loss": 0.8137213,
      "num_input_tokens_seen": 23016285,
      "step": 1079,
      "time_per_iteration": 2.487137794494629
    },
    {
      "auxiliary_loss_clip": 0.01148927,
      "auxiliary_loss_mlp": 0.01059818,
      "balance_loss_clip": 1.05222917,
      "balance_loss_mlp": 1.03619123,
      "epoch": 0.06493311288140688,
      "flos": 24608182331520.0,
      "grad_norm": 2.3035097649758867,
      "language_loss": 0.69664478,
      "learning_rate": 3.987213301260294e-06,
      "loss": 0.71873224,
      "num_input_tokens_seen": 23036420,
      "step": 1080,
      "time_per_iteration": 4.006195783615112
    },
    {
      "auxiliary_loss_clip": 0.01174828,
      "auxiliary_loss_mlp": 0.0106691,
      "balance_loss_clip": 1.06694889,
      "balance_loss_mlp": 1.04072022,
      "epoch": 0.06499323613407486,
      "flos": 25337994865920.0,
      "grad_norm": 1.8816511155085698,
      "language_loss": 0.7250495,
      "learning_rate": 3.987169294370123e-06,
      "loss": 0.74746692,
      "num_input_tokens_seen": 23056945,
      "step": 1081,
      "time_per_iteration": 4.078016996383667
    },
    {
      "auxiliary_loss_clip": 0.01191914,
      "auxiliary_loss_mlp": 0.01060806,
      "balance_loss_clip": 1.15285718,
      "balance_loss_mlp": 1.03577244,
      "epoch": 0.06505335938674282,
      "flos": 20375714960640.0,
      "grad_norm": 2.976536428341487,
      "language_loss": 0.84134829,
      "learning_rate": 3.987125212126294e-06,
      "loss": 0.86387551,
      "num_input_tokens_seen": 23074940,
      "step": 1082,
      "time_per_iteration": 2.611605405807495
    },
    {
      "auxiliary_loss_clip": 0.01207984,
      "auxiliary_loss_mlp": 0.01059528,
      "balance_loss_clip": 1.07319331,
      "balance_loss_mlp": 1.03489923,
      "epoch": 0.06511348263941079,
      "flos": 25337923038720.0,
      "grad_norm": 2.3317014200784376,
      "language_loss": 0.82401091,
      "learning_rate": 3.987081054530478e-06,
      "loss": 0.84668601,
      "num_input_tokens_seen": 23093420,
      "step": 1083,
      "time_per_iteration": 2.5405869483947754
    },
    {
      "auxiliary_loss_clip": 0.01174195,
      "auxiliary_loss_mlp": 0.01060445,
      "balance_loss_clip": 1.07356143,
      "balance_loss_mlp": 1.03513706,
      "epoch": 0.06517360589207877,
      "flos": 20332801186560.0,
      "grad_norm": 2.9325057626298663,
      "language_loss": 0.79011863,
      "learning_rate": 3.987036821584348e-06,
      "loss": 0.81246501,
      "num_input_tokens_seen": 23111550,
      "step": 1084,
      "time_per_iteration": 2.633962392807007
    },
    {
      "auxiliary_loss_clip": 0.01176586,
      "auxiliary_loss_mlp": 0.01061196,
      "balance_loss_clip": 1.06325483,
      "balance_loss_mlp": 1.03722298,
      "epoch": 0.06523372914474673,
      "flos": 31681650061440.0,
      "grad_norm": 1.9807720031509701,
      "language_loss": 0.66130853,
      "learning_rate": 3.986992513289584e-06,
      "loss": 0.68368638,
      "num_input_tokens_seen": 23130335,
      "step": 1085,
      "time_per_iteration": 2.623504638671875
    },
    {
      "auxiliary_loss_clip": 0.0117506,
      "auxiliary_loss_mlp": 0.0106282,
      "balance_loss_clip": 1.06515741,
      "balance_loss_mlp": 1.03928792,
      "epoch": 0.0652938523974147,
      "flos": 20778165918720.0,
      "grad_norm": 2.093824201519614,
      "language_loss": 0.76590407,
      "learning_rate": 3.9869481296478645e-06,
      "loss": 0.78828281,
      "num_input_tokens_seen": 23152380,
      "step": 1086,
      "time_per_iteration": 2.5879526138305664
    },
    {
      "auxiliary_loss_clip": 0.01199623,
      "auxiliary_loss_mlp": 0.01055929,
      "balance_loss_clip": 1.06513119,
      "balance_loss_mlp": 1.03124118,
      "epoch": 0.06535397565008266,
      "flos": 16690993061760.0,
      "grad_norm": 2.3264679911228514,
      "language_loss": 0.85129821,
      "learning_rate": 3.986903670660872e-06,
      "loss": 0.87385368,
      "num_input_tokens_seen": 23171630,
      "step": 1087,
      "time_per_iteration": 2.511742353439331
    },
    {
      "auxiliary_loss_clip": 0.01188605,
      "auxiliary_loss_mlp": 0.01057494,
      "balance_loss_clip": 1.06811059,
      "balance_loss_mlp": 1.03389096,
      "epoch": 0.06541409890275064,
      "flos": 26868220116480.0,
      "grad_norm": 1.9627707746288217,
      "language_loss": 0.77970618,
      "learning_rate": 3.9868591363302945e-06,
      "loss": 0.80216712,
      "num_input_tokens_seen": 23192520,
      "step": 1088,
      "time_per_iteration": 2.616218328475952
    },
    {
      "auxiliary_loss_clip": 0.01195006,
      "auxiliary_loss_mlp": 0.01062739,
      "balance_loss_clip": 1.06818581,
      "balance_loss_mlp": 1.03980327,
      "epoch": 0.06547422215541861,
      "flos": 20521620005760.0,
      "grad_norm": 2.4399243989732837,
      "language_loss": 0.71390957,
      "learning_rate": 3.9868145266578186e-06,
      "loss": 0.73648703,
      "num_input_tokens_seen": 23210710,
      "step": 1089,
      "time_per_iteration": 2.5650622844696045
    },
    {
      "auxiliary_loss_clip": 0.0117592,
      "auxiliary_loss_mlp": 0.00807978,
      "balance_loss_clip": 1.06647468,
      "balance_loss_mlp": 1.04776788,
      "epoch": 0.06553434540808657,
      "flos": 22016616992640.0,
      "grad_norm": 2.000081489051539,
      "language_loss": 0.85323596,
      "learning_rate": 3.9867698416451366e-06,
      "loss": 0.87307489,
      "num_input_tokens_seen": 23230305,
      "step": 1090,
      "time_per_iteration": 2.5791780948638916
    },
    {
      "auxiliary_loss_clip": 0.01214988,
      "auxiliary_loss_mlp": 0.01053428,
      "balance_loss_clip": 1.06873822,
      "balance_loss_mlp": 1.03012216,
      "epoch": 0.06559446866075455,
      "flos": 24608649208320.0,
      "grad_norm": 1.692122282020335,
      "language_loss": 0.71675372,
      "learning_rate": 3.9867250812939434e-06,
      "loss": 0.73943788,
      "num_input_tokens_seen": 23249015,
      "step": 1091,
      "time_per_iteration": 2.524648427963257
    },
    {
      "auxiliary_loss_clip": 0.0112718,
      "auxiliary_loss_mlp": 0.01059641,
      "balance_loss_clip": 1.06338203,
      "balance_loss_mlp": 1.03488088,
      "epoch": 0.06565459191342252,
      "flos": 24274679529600.0,
      "grad_norm": 2.378314415057797,
      "language_loss": 0.82636607,
      "learning_rate": 3.986680245605936e-06,
      "loss": 0.84823424,
      "num_input_tokens_seen": 23265105,
      "step": 1092,
      "time_per_iteration": 2.8152034282684326
    },
    {
      "auxiliary_loss_clip": 0.01217887,
      "auxiliary_loss_mlp": 0.01059727,
      "balance_loss_clip": 1.06653273,
      "balance_loss_mlp": 1.03435969,
      "epoch": 0.06571471516609048,
      "flos": 24787124910720.0,
      "grad_norm": 1.9884923854961465,
      "language_loss": 0.7141096,
      "learning_rate": 3.986635334582814e-06,
      "loss": 0.73688567,
      "num_input_tokens_seen": 23283950,
      "step": 1093,
      "time_per_iteration": 2.648366928100586
    },
    {
      "auxiliary_loss_clip": 0.01190877,
      "auxiliary_loss_mlp": 0.01062309,
      "balance_loss_clip": 1.06941783,
      "balance_loss_mlp": 1.03684616,
      "epoch": 0.06577483841875846,
      "flos": 26214071581440.0,
      "grad_norm": 1.629856260676979,
      "language_loss": 0.87775266,
      "learning_rate": 3.986590348226282e-06,
      "loss": 0.90028453,
      "num_input_tokens_seen": 23305005,
      "step": 1094,
      "time_per_iteration": 2.611783504486084
    },
    {
      "auxiliary_loss_clip": 0.01196522,
      "auxiliary_loss_mlp": 0.01065223,
      "balance_loss_clip": 1.07000995,
      "balance_loss_mlp": 1.03853226,
      "epoch": 0.06583496167142643,
      "flos": 25080802508160.0,
      "grad_norm": 1.4986106042245428,
      "language_loss": 0.81367749,
      "learning_rate": 3.986545286538044e-06,
      "loss": 0.83629489,
      "num_input_tokens_seen": 23323220,
      "step": 1095,
      "time_per_iteration": 2.6105403900146484
    },
    {
      "auxiliary_loss_clip": 0.01171226,
      "auxiliary_loss_mlp": 0.01057199,
      "balance_loss_clip": 1.0698247,
      "balance_loss_mlp": 1.03503776,
      "epoch": 0.06589508492409439,
      "flos": 25629804956160.0,
      "grad_norm": 1.9463469050724667,
      "language_loss": 0.70004129,
      "learning_rate": 3.986500149519811e-06,
      "loss": 0.7223255,
      "num_input_tokens_seen": 23342235,
      "step": 1096,
      "time_per_iteration": 2.6628947257995605
    },
    {
      "auxiliary_loss_clip": 0.01206153,
      "auxiliary_loss_mlp": 0.01072204,
      "balance_loss_clip": 1.07087755,
      "balance_loss_mlp": 1.04739714,
      "epoch": 0.06595520817676236,
      "flos": 23621249266560.0,
      "grad_norm": 1.95808333764464,
      "language_loss": 0.77587497,
      "learning_rate": 3.986454937173292e-06,
      "loss": 0.79865849,
      "num_input_tokens_seen": 23363680,
      "step": 1097,
      "time_per_iteration": 2.5664379596710205
    },
    {
      "auxiliary_loss_clip": 0.0121717,
      "auxiliary_loss_mlp": 0.01063793,
      "balance_loss_clip": 1.06849146,
      "balance_loss_mlp": 1.04049933,
      "epoch": 0.06601533142943034,
      "flos": 33801708545280.0,
      "grad_norm": 1.9999713666033208,
      "language_loss": 0.78775114,
      "learning_rate": 3.986409649500203e-06,
      "loss": 0.81056082,
      "num_input_tokens_seen": 23385590,
      "step": 1098,
      "time_per_iteration": 2.6263554096221924
    },
    {
      "auxiliary_loss_clip": 0.01254995,
      "auxiliary_loss_mlp": 0.01072959,
      "balance_loss_clip": 1.16318274,
      "balance_loss_mlp": 1.04735315,
      "epoch": 0.0660754546820983,
      "flos": 20259184262400.0,
      "grad_norm": 2.8455302558143747,
      "language_loss": 0.81899405,
      "learning_rate": 3.986364286502261e-06,
      "loss": 0.84227365,
      "num_input_tokens_seen": 23402945,
      "step": 1099,
      "time_per_iteration": 2.5424156188964844
    },
    {
      "auxiliary_loss_clip": 0.01183987,
      "auxiliary_loss_mlp": 0.01055729,
      "balance_loss_clip": 1.06312323,
      "balance_loss_mlp": 1.03059983,
      "epoch": 0.06613557793476627,
      "flos": 19354164163200.0,
      "grad_norm": 1.940237122032918,
      "language_loss": 0.82582057,
      "learning_rate": 3.986318848181186e-06,
      "loss": 0.84821767,
      "num_input_tokens_seen": 23421410,
      "step": 1100,
      "time_per_iteration": 2.5277209281921387
    },
    {
      "auxiliary_loss_clip": 0.01192002,
      "auxiliary_loss_mlp": 0.01060321,
      "balance_loss_clip": 1.07192683,
      "balance_loss_mlp": 1.03614509,
      "epoch": 0.06619570118743424,
      "flos": 13772568936960.0,
      "grad_norm": 2.3698108715995128,
      "language_loss": 0.73673934,
      "learning_rate": 3.986273334538702e-06,
      "loss": 0.75926256,
      "num_input_tokens_seen": 23438870,
      "step": 1101,
      "time_per_iteration": 2.545546770095825
    },
    {
      "auxiliary_loss_clip": 0.01201792,
      "auxiliary_loss_mlp": 0.01062904,
      "balance_loss_clip": 1.06455195,
      "balance_loss_mlp": 1.03851366,
      "epoch": 0.06625582444010221,
      "flos": 17857874286720.0,
      "grad_norm": 4.564811369163246,
      "language_loss": 0.86143339,
      "learning_rate": 3.986227745576533e-06,
      "loss": 0.88408029,
      "num_input_tokens_seen": 23456975,
      "step": 1102,
      "time_per_iteration": 2.5051393508911133
    },
    {
      "auxiliary_loss_clip": 0.01189575,
      "auxiliary_loss_mlp": 0.0105959,
      "balance_loss_clip": 1.06624293,
      "balance_loss_mlp": 1.03508103,
      "epoch": 0.06631594769277017,
      "flos": 11838707579520.0,
      "grad_norm": 2.3838972733759025,
      "language_loss": 0.81932425,
      "learning_rate": 3.98618208129641e-06,
      "loss": 0.84181589,
      "num_input_tokens_seen": 23473440,
      "step": 1103,
      "time_per_iteration": 2.5489401817321777
    },
    {
      "auxiliary_loss_clip": 0.01204955,
      "auxiliary_loss_mlp": 0.00821134,
      "balance_loss_clip": 1.07047033,
      "balance_loss_mlp": 1.07102799,
      "epoch": 0.06637607094543815,
      "flos": 19793351756160.0,
      "grad_norm": 1.8494668546361632,
      "language_loss": 0.82212621,
      "learning_rate": 3.986136341700063e-06,
      "loss": 0.84238708,
      "num_input_tokens_seen": 23493880,
      "step": 1104,
      "time_per_iteration": 2.569284677505493
    },
    {
      "auxiliary_loss_clip": 0.0117171,
      "auxiliary_loss_mlp": 0.01050948,
      "balance_loss_clip": 1.06535995,
      "balance_loss_mlp": 1.02618861,
      "epoch": 0.06643619419810612,
      "flos": 25485659677440.0,
      "grad_norm": 1.674453414127921,
      "language_loss": 0.80616009,
      "learning_rate": 3.986090526789227e-06,
      "loss": 0.82838666,
      "num_input_tokens_seen": 23514920,
      "step": 1105,
      "time_per_iteration": 2.650836229324341
    },
    {
      "auxiliary_loss_clip": 0.01185983,
      "auxiliary_loss_mlp": 0.01061431,
      "balance_loss_clip": 1.07025373,
      "balance_loss_mlp": 1.03833985,
      "epoch": 0.06649631745077408,
      "flos": 16946533393920.0,
      "grad_norm": 2.0041114872029078,
      "language_loss": 0.96829402,
      "learning_rate": 3.986044636565639e-06,
      "loss": 0.99076807,
      "num_input_tokens_seen": 23531635,
      "step": 1106,
      "time_per_iteration": 2.545681953430176
    },
    {
      "auxiliary_loss_clip": 0.0120748,
      "auxiliary_loss_mlp": 0.01060144,
      "balance_loss_clip": 1.06974435,
      "balance_loss_mlp": 1.03524113,
      "epoch": 0.06655644070344206,
      "flos": 17858592558720.0,
      "grad_norm": 2.077196130817546,
      "language_loss": 0.82625329,
      "learning_rate": 3.985998671031039e-06,
      "loss": 0.84892952,
      "num_input_tokens_seen": 23551020,
      "step": 1107,
      "time_per_iteration": 2.55208683013916
    },
    {
      "auxiliary_loss_clip": 0.01147928,
      "auxiliary_loss_mlp": 0.01014782,
      "balance_loss_clip": 1.13096881,
      "balance_loss_mlp": 1.01056218,
      "epoch": 0.06661656395611003,
      "flos": 61419350021760.0,
      "grad_norm": 0.7992800992168448,
      "language_loss": 0.56762016,
      "learning_rate": 3.9859526301871705e-06,
      "loss": 0.58924723,
      "num_input_tokens_seen": 23610675,
      "step": 1108,
      "time_per_iteration": 3.0996246337890625
    },
    {
      "auxiliary_loss_clip": 0.01190544,
      "auxiliary_loss_mlp": 0.01060948,
      "balance_loss_clip": 1.06359744,
      "balance_loss_mlp": 1.03536594,
      "epoch": 0.066676687208778,
      "flos": 20662856282880.0,
      "grad_norm": 2.792563791525604,
      "language_loss": 0.72755802,
      "learning_rate": 3.9859065140357795e-06,
      "loss": 0.75007296,
      "num_input_tokens_seen": 23628710,
      "step": 1109,
      "time_per_iteration": 2.570469379425049
    },
    {
      "auxiliary_loss_clip": 0.01159147,
      "auxiliary_loss_mlp": 0.01064347,
      "balance_loss_clip": 1.0659076,
      "balance_loss_mlp": 1.03905106,
      "epoch": 0.06673681046144596,
      "flos": 20923280864640.0,
      "grad_norm": 2.1865432226292354,
      "language_loss": 0.78031182,
      "learning_rate": 3.985860322578614e-06,
      "loss": 0.80254674,
      "num_input_tokens_seen": 23649160,
      "step": 1110,
      "time_per_iteration": 2.6239068508148193
    },
    {
      "auxiliary_loss_clip": 0.01163757,
      "auxiliary_loss_mlp": 0.01056316,
      "balance_loss_clip": 1.06802034,
      "balance_loss_mlp": 1.03258193,
      "epoch": 0.06679693371411394,
      "flos": 31065818359680.0,
      "grad_norm": 2.0186322653832978,
      "language_loss": 0.71698946,
      "learning_rate": 3.985814055817427e-06,
      "loss": 0.73919022,
      "num_input_tokens_seen": 23671995,
      "step": 1111,
      "time_per_iteration": 2.7007100582122803
    },
    {
      "auxiliary_loss_clip": 0.01177219,
      "auxiliary_loss_mlp": 0.01069141,
      "balance_loss_clip": 1.06778455,
      "balance_loss_mlp": 1.04513264,
      "epoch": 0.0668570569667819,
      "flos": 21726135705600.0,
      "grad_norm": 2.174539858357906,
      "language_loss": 0.78634948,
      "learning_rate": 3.985767713753971e-06,
      "loss": 0.8088131,
      "num_input_tokens_seen": 23690705,
      "step": 1112,
      "time_per_iteration": 2.665090560913086
    },
    {
      "auxiliary_loss_clip": 0.01150751,
      "auxiliary_loss_mlp": 0.0106445,
      "balance_loss_clip": 1.05291378,
      "balance_loss_mlp": 1.0402627,
      "epoch": 0.06691718021944987,
      "flos": 22747255539840.0,
      "grad_norm": 2.400443235003305,
      "language_loss": 0.79269671,
      "learning_rate": 3.985721296390005e-06,
      "loss": 0.81484878,
      "num_input_tokens_seen": 23709990,
      "step": 1113,
      "time_per_iteration": 4.017433166503906
    },
    {
      "auxiliary_loss_clip": 0.01157597,
      "auxiliary_loss_mlp": 0.01061791,
      "balance_loss_clip": 1.06595159,
      "balance_loss_mlp": 1.03847361,
      "epoch": 0.06697730347211785,
      "flos": 16545626720640.0,
      "grad_norm": 1.7761953581480479,
      "language_loss": 0.82650596,
      "learning_rate": 3.985674803727289e-06,
      "loss": 0.84869981,
      "num_input_tokens_seen": 23728485,
      "step": 1114,
      "time_per_iteration": 2.66770601272583
    },
    {
      "auxiliary_loss_clip": 0.01069168,
      "auxiliary_loss_mlp": 0.01016101,
      "balance_loss_clip": 1.0449748,
      "balance_loss_mlp": 1.01166618,
      "epoch": 0.06703742672478581,
      "flos": 59782326658560.0,
      "grad_norm": 0.8409293515507801,
      "language_loss": 0.58100414,
      "learning_rate": 3.985628235767584e-06,
      "loss": 0.60185683,
      "num_input_tokens_seen": 23786650,
      "step": 1115,
      "time_per_iteration": 4.631579637527466
    },
    {
      "auxiliary_loss_clip": 0.01187378,
      "auxiliary_loss_mlp": 0.01064851,
      "balance_loss_clip": 1.07080984,
      "balance_loss_mlp": 1.03887546,
      "epoch": 0.06709754997745378,
      "flos": 16800197385600.0,
      "grad_norm": 2.8642662220929167,
      "language_loss": 0.91588414,
      "learning_rate": 3.985581592512658e-06,
      "loss": 0.93840653,
      "num_input_tokens_seen": 23802555,
      "step": 1116,
      "time_per_iteration": 2.5949795246124268
    },
    {
      "auxiliary_loss_clip": 0.01177749,
      "auxiliary_loss_mlp": 0.00816988,
      "balance_loss_clip": 1.07026303,
      "balance_loss_mlp": 1.06006992,
      "epoch": 0.06715767323012176,
      "flos": 22123917895680.0,
      "grad_norm": 1.8080041763185188,
      "language_loss": 0.87371922,
      "learning_rate": 3.985534873964279e-06,
      "loss": 0.89366663,
      "num_input_tokens_seen": 23822945,
      "step": 1117,
      "time_per_iteration": 2.703951358795166
    },
    {
      "auxiliary_loss_clip": 0.01093954,
      "auxiliary_loss_mlp": 0.01005615,
      "balance_loss_clip": 1.03779864,
      "balance_loss_mlp": 1.00132322,
      "epoch": 0.06721779648278972,
      "flos": 66618100137600.0,
      "grad_norm": 0.8827841041776882,
      "language_loss": 0.59844798,
      "learning_rate": 3.985488080124218e-06,
      "loss": 0.61944371,
      "num_input_tokens_seen": 23874075,
      "step": 1118,
      "time_per_iteration": 4.475669860839844
    },
    {
      "auxiliary_loss_clip": 0.01186515,
      "auxiliary_loss_mlp": 0.01056698,
      "balance_loss_clip": 1.0621165,
      "balance_loss_mlp": 1.03265321,
      "epoch": 0.06727791973545769,
      "flos": 22382474970240.0,
      "grad_norm": 3.4774748039627883,
      "language_loss": 0.8342942,
      "learning_rate": 3.985441210994251e-06,
      "loss": 0.85672629,
      "num_input_tokens_seen": 23889720,
      "step": 1119,
      "time_per_iteration": 2.535940170288086
    },
    {
      "auxiliary_loss_clip": 0.01190778,
      "auxiliary_loss_mlp": 0.01057783,
      "balance_loss_clip": 1.06962109,
      "balance_loss_mlp": 1.03512156,
      "epoch": 0.06733804298812565,
      "flos": 24280210224000.0,
      "grad_norm": 2.211687750310762,
      "language_loss": 0.84851825,
      "learning_rate": 3.9853942665761545e-06,
      "loss": 0.87100393,
      "num_input_tokens_seen": 23909385,
      "step": 1120,
      "time_per_iteration": 4.017683267593384
    },
    {
      "auxiliary_loss_clip": 0.01220746,
      "auxiliary_loss_mlp": 0.01066115,
      "balance_loss_clip": 1.07265472,
      "balance_loss_mlp": 1.04162955,
      "epoch": 0.06739816624079363,
      "flos": 15918230839680.0,
      "grad_norm": 1.9199342718461165,
      "language_loss": 0.78822112,
      "learning_rate": 3.985347246871708e-06,
      "loss": 0.81108975,
      "num_input_tokens_seen": 23926830,
      "step": 1121,
      "time_per_iteration": 2.4743101596832275
    },
    {
      "auxiliary_loss_clip": 0.01089512,
      "auxiliary_loss_mlp": 0.01007833,
      "balance_loss_clip": 1.0395987,
      "balance_loss_mlp": 1.00409031,
      "epoch": 0.0674582894934616,
      "flos": 71398567353600.0,
      "grad_norm": 0.7557446687332641,
      "language_loss": 0.58347619,
      "learning_rate": 3.985300151882694e-06,
      "loss": 0.60444963,
      "num_input_tokens_seen": 23992640,
      "step": 1122,
      "time_per_iteration": 3.2536373138427734
    },
    {
      "auxiliary_loss_clip": 0.01158287,
      "auxiliary_loss_mlp": 0.01064791,
      "balance_loss_clip": 1.06976271,
      "balance_loss_mlp": 1.04041219,
      "epoch": 0.06751841274612956,
      "flos": 25264952559360.0,
      "grad_norm": 1.9169359993617905,
      "language_loss": 0.71557915,
      "learning_rate": 3.985252981610901e-06,
      "loss": 0.73780996,
      "num_input_tokens_seen": 24011135,
      "step": 1123,
      "time_per_iteration": 2.642385721206665
    },
    {
      "auxiliary_loss_clip": 0.01155322,
      "auxiliary_loss_mlp": 0.01062979,
      "balance_loss_clip": 1.06660748,
      "balance_loss_mlp": 1.03478563,
      "epoch": 0.06757853599879754,
      "flos": 23802741711360.0,
      "grad_norm": 1.8112590089396507,
      "language_loss": 0.78821993,
      "learning_rate": 3.985205736058114e-06,
      "loss": 0.81040293,
      "num_input_tokens_seen": 24030695,
      "step": 1124,
      "time_per_iteration": 2.625953435897827
    },
    {
      "auxiliary_loss_clip": 0.01199398,
      "auxiliary_loss_mlp": 0.01053047,
      "balance_loss_clip": 1.06761551,
      "balance_loss_mlp": 1.03032601,
      "epoch": 0.0676386592514655,
      "flos": 21033742164480.0,
      "grad_norm": 2.3483394470155887,
      "language_loss": 0.71561515,
      "learning_rate": 3.985158415226128e-06,
      "loss": 0.73813963,
      "num_input_tokens_seen": 24050680,
      "step": 1125,
      "time_per_iteration": 2.5361976623535156
    },
    {
      "auxiliary_loss_clip": 0.01175773,
      "auxiliary_loss_mlp": 0.01071265,
      "balance_loss_clip": 1.07004786,
      "balance_loss_mlp": 1.04571807,
      "epoch": 0.06769878250413347,
      "flos": 25556331686400.0,
      "grad_norm": 2.7506463679843507,
      "language_loss": 0.81051481,
      "learning_rate": 3.985111019116736e-06,
      "loss": 0.83298516,
      "num_input_tokens_seen": 24067205,
      "step": 1126,
      "time_per_iteration": 2.596774101257324
    },
    {
      "auxiliary_loss_clip": 0.01081984,
      "auxiliary_loss_mlp": 0.01006116,
      "balance_loss_clip": 1.03780389,
      "balance_loss_mlp": 1.00213432,
      "epoch": 0.06775890575680145,
      "flos": 70655251305600.0,
      "grad_norm": 0.7974349759856977,
      "language_loss": 0.59766507,
      "learning_rate": 3.985063547731735e-06,
      "loss": 0.61854607,
      "num_input_tokens_seen": 24131320,
      "step": 1127,
      "time_per_iteration": 3.1434102058410645
    },
    {
      "auxiliary_loss_clip": 0.01212644,
      "auxiliary_loss_mlp": 0.01054407,
      "balance_loss_clip": 1.06904054,
      "balance_loss_mlp": 1.03068447,
      "epoch": 0.06781902900946941,
      "flos": 24235500769920.0,
      "grad_norm": 2.1091829590563007,
      "language_loss": 0.81434685,
      "learning_rate": 3.985016001072925e-06,
      "loss": 0.8370173,
      "num_input_tokens_seen": 24149930,
      "step": 1128,
      "time_per_iteration": 2.5007717609405518
    },
    {
      "auxiliary_loss_clip": 0.01171031,
      "auxiliary_loss_mlp": 0.01049269,
      "balance_loss_clip": 1.07150316,
      "balance_loss_mlp": 1.02345979,
      "epoch": 0.06787915226213738,
      "flos": 22417523665920.0,
      "grad_norm": 2.0353338945221733,
      "language_loss": 0.75527853,
      "learning_rate": 3.984968379142109e-06,
      "loss": 0.77748156,
      "num_input_tokens_seen": 24169590,
      "step": 1129,
      "time_per_iteration": 2.5951085090637207
    },
    {
      "auxiliary_loss_clip": 0.0113329,
      "auxiliary_loss_mlp": 0.01060341,
      "balance_loss_clip": 1.06761754,
      "balance_loss_mlp": 1.03530741,
      "epoch": 0.06793927551480534,
      "flos": 37706922080640.0,
      "grad_norm": 1.9157595062371864,
      "language_loss": 0.72491419,
      "learning_rate": 3.984920681941094e-06,
      "loss": 0.74685049,
      "num_input_tokens_seen": 24189965,
      "step": 1130,
      "time_per_iteration": 2.8679919242858887
    },
    {
      "auxiliary_loss_clip": 0.01166104,
      "auxiliary_loss_mlp": 0.01059441,
      "balance_loss_clip": 1.06505978,
      "balance_loss_mlp": 1.03489542,
      "epoch": 0.06799939876747332,
      "flos": 20631398947200.0,
      "grad_norm": 2.607843312225482,
      "language_loss": 0.80783767,
      "learning_rate": 3.984872909471688e-06,
      "loss": 0.83009315,
      "num_input_tokens_seen": 24208045,
      "step": 1131,
      "time_per_iteration": 2.7554352283477783
    },
    {
      "auxiliary_loss_clip": 0.01200847,
      "auxiliary_loss_mlp": 0.01066078,
      "balance_loss_clip": 1.06567776,
      "balance_loss_mlp": 1.04184222,
      "epoch": 0.06805952202014129,
      "flos": 14864755829760.0,
      "grad_norm": 2.09101309082411,
      "language_loss": 0.80399406,
      "learning_rate": 3.984825061735701e-06,
      "loss": 0.82666326,
      "num_input_tokens_seen": 24223805,
      "step": 1132,
      "time_per_iteration": 2.486359119415283
    },
    {
      "auxiliary_loss_clip": 0.01184655,
      "auxiliary_loss_mlp": 0.01063268,
      "balance_loss_clip": 1.07271254,
      "balance_loss_mlp": 1.03897333,
      "epoch": 0.06811964527280925,
      "flos": 48909434947200.0,
      "grad_norm": 1.5192230891698713,
      "language_loss": 0.63448668,
      "learning_rate": 3.9847771387349495e-06,
      "loss": 0.65696585,
      "num_input_tokens_seen": 24249475,
      "step": 1133,
      "time_per_iteration": 2.8064792156219482
    },
    {
      "auxiliary_loss_clip": 0.01125387,
      "auxiliary_loss_mlp": 0.0105997,
      "balance_loss_clip": 1.04968143,
      "balance_loss_mlp": 1.03256381,
      "epoch": 0.06817976852547723,
      "flos": 15377273038080.0,
      "grad_norm": 2.003233265821166,
      "language_loss": 0.7454046,
      "learning_rate": 3.9847291404712506e-06,
      "loss": 0.76725817,
      "num_input_tokens_seen": 24267980,
      "step": 1134,
      "time_per_iteration": 2.624812364578247
    },
    {
      "auxiliary_loss_clip": 0.01179735,
      "auxiliary_loss_mlp": 0.00799421,
      "balance_loss_clip": 1.06837308,
      "balance_loss_mlp": 1.02759528,
      "epoch": 0.0682398917781452,
      "flos": 20155690200960.0,
      "grad_norm": 2.1688591520864264,
      "language_loss": 0.87023634,
      "learning_rate": 3.984681066946423e-06,
      "loss": 0.89002788,
      "num_input_tokens_seen": 24286805,
      "step": 1135,
      "time_per_iteration": 2.585642099380493
    },
    {
      "auxiliary_loss_clip": 0.01183329,
      "auxiliary_loss_mlp": 0.00796706,
      "balance_loss_clip": 1.06317592,
      "balance_loss_mlp": 1.02166605,
      "epoch": 0.06830001503081316,
      "flos": 23440618748160.0,
      "grad_norm": 2.6761458311863247,
      "language_loss": 0.78343618,
      "learning_rate": 3.984632918162291e-06,
      "loss": 0.80323648,
      "num_input_tokens_seen": 24305855,
      "step": 1136,
      "time_per_iteration": 2.566164493560791
    },
    {
      "auxiliary_loss_clip": 0.01193156,
      "auxiliary_loss_mlp": 0.01069832,
      "balance_loss_clip": 1.06738663,
      "balance_loss_mlp": 1.04467845,
      "epoch": 0.06836013828348114,
      "flos": 34349813153280.0,
      "grad_norm": 2.340356846484295,
      "language_loss": 0.84292316,
      "learning_rate": 3.984584694120679e-06,
      "loss": 0.86555302,
      "num_input_tokens_seen": 24326535,
      "step": 1137,
      "time_per_iteration": 2.6787359714508057
    },
    {
      "auxiliary_loss_clip": 0.01165026,
      "auxiliary_loss_mlp": 0.01058841,
      "balance_loss_clip": 1.06723642,
      "balance_loss_mlp": 1.03470111,
      "epoch": 0.06842026153614911,
      "flos": 23148844571520.0,
      "grad_norm": 2.0964565607321455,
      "language_loss": 0.78649384,
      "learning_rate": 3.984536394823418e-06,
      "loss": 0.80873251,
      "num_input_tokens_seen": 24345810,
      "step": 1138,
      "time_per_iteration": 2.6176974773406982
    },
    {
      "auxiliary_loss_clip": 0.01213634,
      "auxiliary_loss_mlp": 0.01056506,
      "balance_loss_clip": 1.06674552,
      "balance_loss_mlp": 1.03179431,
      "epoch": 0.06848038478881707,
      "flos": 24608972430720.0,
      "grad_norm": 3.13388214743193,
      "language_loss": 0.85622025,
      "learning_rate": 3.984488020272336e-06,
      "loss": 0.87892163,
      "num_input_tokens_seen": 24366095,
      "step": 1139,
      "time_per_iteration": 2.508000135421753
    },
    {
      "auxiliary_loss_clip": 0.01163463,
      "auxiliary_loss_mlp": 0.01066501,
      "balance_loss_clip": 1.06544924,
      "balance_loss_mlp": 1.04006052,
      "epoch": 0.06854050804148504,
      "flos": 40880994278400.0,
      "grad_norm": 1.7204646785224804,
      "language_loss": 0.74888623,
      "learning_rate": 3.984439570469271e-06,
      "loss": 0.77118576,
      "num_input_tokens_seen": 24388665,
      "step": 1140,
      "time_per_iteration": 2.7564125061035156
    },
    {
      "auxiliary_loss_clip": 0.0118914,
      "auxiliary_loss_mlp": 0.00800851,
      "balance_loss_clip": 1.06498921,
      "balance_loss_mlp": 1.03154659,
      "epoch": 0.06860063129415302,
      "flos": 31686354743040.0,
      "grad_norm": 2.45191140560547,
      "language_loss": 0.68056405,
      "learning_rate": 3.9843910454160574e-06,
      "loss": 0.70046401,
      "num_input_tokens_seen": 24407705,
      "step": 1141,
      "time_per_iteration": 2.640537977218628
    },
    {
      "auxiliary_loss_clip": 0.01204433,
      "auxiliary_loss_mlp": 0.01066097,
      "balance_loss_clip": 1.06606078,
      "balance_loss_mlp": 1.04024053,
      "epoch": 0.06866075454682098,
      "flos": 26542007775360.0,
      "grad_norm": 1.9656488939580798,
      "language_loss": 0.79275274,
      "learning_rate": 3.984342445114538e-06,
      "loss": 0.81545806,
      "num_input_tokens_seen": 24428390,
      "step": 1142,
      "time_per_iteration": 2.570099353790283
    },
    {
      "auxiliary_loss_clip": 0.01189556,
      "auxiliary_loss_mlp": 0.01060456,
      "balance_loss_clip": 1.0652585,
      "balance_loss_mlp": 1.03663826,
      "epoch": 0.06872087779948895,
      "flos": 29789768724480.0,
      "grad_norm": 2.125905014405525,
      "language_loss": 0.69020647,
      "learning_rate": 3.984293769566553e-06,
      "loss": 0.71270663,
      "num_input_tokens_seen": 24450810,
      "step": 1143,
      "time_per_iteration": 2.620107650756836
    },
    {
      "auxiliary_loss_clip": 0.0117239,
      "auxiliary_loss_mlp": 0.01059341,
      "balance_loss_clip": 1.05389881,
      "balance_loss_mlp": 1.03746605,
      "epoch": 0.06878100105215693,
      "flos": 26941118768640.0,
      "grad_norm": 1.6917533243342024,
      "language_loss": 0.74627882,
      "learning_rate": 3.98424501877395e-06,
      "loss": 0.76859611,
      "num_input_tokens_seen": 24469965,
      "step": 1144,
      "time_per_iteration": 2.562654495239258
    },
    {
      "auxiliary_loss_clip": 0.0119792,
      "auxiliary_loss_mlp": 0.01061626,
      "balance_loss_clip": 1.0696485,
      "balance_loss_mlp": 1.03625786,
      "epoch": 0.06884112430482489,
      "flos": 10670748946560.0,
      "grad_norm": 2.984439777550312,
      "language_loss": 0.92093068,
      "learning_rate": 3.984196192738577e-06,
      "loss": 0.94352621,
      "num_input_tokens_seen": 24486370,
      "step": 1145,
      "time_per_iteration": 2.471287250518799
    },
    {
      "auxiliary_loss_clip": 0.01214304,
      "auxiliary_loss_mlp": 0.01066132,
      "balance_loss_clip": 1.0649333,
      "balance_loss_mlp": 1.04107475,
      "epoch": 0.06890124755749286,
      "flos": 20193647898240.0,
      "grad_norm": 2.4757197178980532,
      "language_loss": 0.8224296,
      "learning_rate": 3.984147291462285e-06,
      "loss": 0.84523398,
      "num_input_tokens_seen": 24503780,
      "step": 1146,
      "time_per_iteration": 2.5051491260528564
    },
    {
      "auxiliary_loss_clip": 0.01209052,
      "auxiliary_loss_mlp": 0.01063667,
      "balance_loss_clip": 1.06731439,
      "balance_loss_mlp": 1.04087377,
      "epoch": 0.06896137081016084,
      "flos": 20449224144000.0,
      "grad_norm": 2.5202236995326737,
      "language_loss": 0.85243917,
      "learning_rate": 3.98409831494693e-06,
      "loss": 0.8751663,
      "num_input_tokens_seen": 24522320,
      "step": 1147,
      "time_per_iteration": 2.4951117038726807
    },
    {
      "auxiliary_loss_clip": 0.01162497,
      "auxiliary_loss_mlp": 0.01064921,
      "balance_loss_clip": 1.06436813,
      "balance_loss_mlp": 1.04087651,
      "epoch": 0.0690214940628288,
      "flos": 18368703555840.0,
      "grad_norm": 2.3113134841473237,
      "language_loss": 0.8587389,
      "learning_rate": 3.984049263194367e-06,
      "loss": 0.8810131,
      "num_input_tokens_seen": 24540445,
      "step": 1148,
      "time_per_iteration": 2.592421531677246
    },
    {
      "auxiliary_loss_clip": 0.01180267,
      "auxiliary_loss_mlp": 0.01064534,
      "balance_loss_clip": 1.06626916,
      "balance_loss_mlp": 1.03891587,
      "epoch": 0.06908161731549677,
      "flos": 20558033418240.0,
      "grad_norm": 2.3341387100873874,
      "language_loss": 0.69488913,
      "learning_rate": 3.9840001362064575e-06,
      "loss": 0.71733713,
      "num_input_tokens_seen": 24557105,
      "step": 1149,
      "time_per_iteration": 2.554088830947876
    },
    {
      "auxiliary_loss_clip": 0.01214463,
      "auxiliary_loss_mlp": 0.01053825,
      "balance_loss_clip": 1.06594753,
      "balance_loss_mlp": 1.02907753,
      "epoch": 0.06914174056816474,
      "flos": 27563666313600.0,
      "grad_norm": 1.8541888159764939,
      "language_loss": 0.83762074,
      "learning_rate": 3.983950933985064e-06,
      "loss": 0.86030358,
      "num_input_tokens_seen": 24578240,
      "step": 1150,
      "time_per_iteration": 2.526684045791626
    },
    {
      "auxiliary_loss_clip": 0.0118867,
      "auxiliary_loss_mlp": 0.0105958,
      "balance_loss_clip": 1.06788731,
      "balance_loss_mlp": 1.03521347,
      "epoch": 0.06920186382083271,
      "flos": 15304015249920.0,
      "grad_norm": 4.312248316102627,
      "language_loss": 0.81371701,
      "learning_rate": 3.983901656532052e-06,
      "loss": 0.83619946,
      "num_input_tokens_seen": 24593585,
      "step": 1151,
      "time_per_iteration": 3.8767220973968506
    },
    {
      "auxiliary_loss_clip": 0.01210657,
      "auxiliary_loss_mlp": 0.01061027,
      "balance_loss_clip": 1.06765771,
      "balance_loss_mlp": 1.03767347,
      "epoch": 0.06926198707350067,
      "flos": 25191227894400.0,
      "grad_norm": 1.8404893434885063,
      "language_loss": 0.85450852,
      "learning_rate": 3.983852303849291e-06,
      "loss": 0.8772254,
      "num_input_tokens_seen": 24613110,
      "step": 1152,
      "time_per_iteration": 2.513389825820923
    },
    {
      "auxiliary_loss_clip": 0.01194226,
      "auxiliary_loss_mlp": 0.01058253,
      "balance_loss_clip": 1.06703782,
      "balance_loss_mlp": 1.03559148,
      "epoch": 0.06932211032616864,
      "flos": 13256137146240.0,
      "grad_norm": 2.442297144011295,
      "language_loss": 0.90685719,
      "learning_rate": 3.983802875938651e-06,
      "loss": 0.92938197,
      "num_input_tokens_seen": 24628795,
      "step": 1153,
      "time_per_iteration": 3.843212842941284
    },
    {
      "auxiliary_loss_clip": 0.0116571,
      "auxiliary_loss_mlp": 0.01055147,
      "balance_loss_clip": 1.05471826,
      "balance_loss_mlp": 1.03109074,
      "epoch": 0.06938223357883662,
      "flos": 24827381078400.0,
      "grad_norm": 2.5465244461108454,
      "language_loss": 0.81399202,
      "learning_rate": 3.983753372802008e-06,
      "loss": 0.83620059,
      "num_input_tokens_seen": 24645480,
      "step": 1154,
      "time_per_iteration": 2.5371267795562744
    },
    {
      "auxiliary_loss_clip": 0.01191813,
      "auxiliary_loss_mlp": 0.01059657,
      "balance_loss_clip": 1.07322764,
      "balance_loss_mlp": 1.03625607,
      "epoch": 0.06944235683150458,
      "flos": 27267977554560.0,
      "grad_norm": 1.964007526828095,
      "language_loss": 0.75378811,
      "learning_rate": 3.983703794441237e-06,
      "loss": 0.77630281,
      "num_input_tokens_seen": 24664630,
      "step": 1155,
      "time_per_iteration": 2.61171293258667
    },
    {
      "auxiliary_loss_clip": 0.0118704,
      "auxiliary_loss_mlp": 0.00793952,
      "balance_loss_clip": 1.06438863,
      "balance_loss_mlp": 1.01722181,
      "epoch": 0.06950248008417255,
      "flos": 25808065176960.0,
      "grad_norm": 1.920563873994528,
      "language_loss": 0.70790964,
      "learning_rate": 3.98365414085822e-06,
      "loss": 0.72771955,
      "num_input_tokens_seen": 24684210,
      "step": 1156,
      "time_per_iteration": 2.5296638011932373
    },
    {
      "auxiliary_loss_clip": 0.01231957,
      "auxiliary_loss_mlp": 0.00798384,
      "balance_loss_clip": 1.15258408,
      "balance_loss_mlp": 1.02437449,
      "epoch": 0.06956260333684053,
      "flos": 22271546793600.0,
      "grad_norm": 1.9913972250886258,
      "language_loss": 0.7471891,
      "learning_rate": 3.98360441205484e-06,
      "loss": 0.76749253,
      "num_input_tokens_seen": 24702490,
      "step": 1157,
      "time_per_iteration": 3.9423065185546875
    },
    {
      "auxiliary_loss_clip": 0.01182633,
      "auxiliary_loss_mlp": 0.01056211,
      "balance_loss_clip": 1.06340563,
      "balance_loss_mlp": 1.03170133,
      "epoch": 0.0696227265895085,
      "flos": 29681390413440.0,
      "grad_norm": 1.7397935482058873,
      "language_loss": 0.71791697,
      "learning_rate": 3.983554608032982e-06,
      "loss": 0.74030542,
      "num_input_tokens_seen": 24724340,
      "step": 1158,
      "time_per_iteration": 2.6307575702667236
    },
    {
      "auxiliary_loss_clip": 0.01213456,
      "auxiliary_loss_mlp": 0.01060266,
      "balance_loss_clip": 1.06703401,
      "balance_loss_mlp": 1.0359236,
      "epoch": 0.06968284984217646,
      "flos": 25523545547520.0,
      "grad_norm": 2.089709290236777,
      "language_loss": 0.79619133,
      "learning_rate": 3.983504728794533e-06,
      "loss": 0.81892854,
      "num_input_tokens_seen": 24745550,
      "step": 1159,
      "time_per_iteration": 3.899813652038574
    },
    {
      "auxiliary_loss_clip": 0.01214749,
      "auxiliary_loss_mlp": 0.01059208,
      "balance_loss_clip": 1.06830204,
      "balance_loss_mlp": 1.03302956,
      "epoch": 0.06974297309484444,
      "flos": 20698192287360.0,
      "grad_norm": 11.298384426823219,
      "language_loss": 0.80917019,
      "learning_rate": 3.983454774341387e-06,
      "loss": 0.83190978,
      "num_input_tokens_seen": 24762575,
      "step": 1160,
      "time_per_iteration": 2.4611637592315674
    },
    {
      "auxiliary_loss_clip": 0.01244288,
      "auxiliary_loss_mlp": 0.01055671,
      "balance_loss_clip": 1.14829993,
      "balance_loss_mlp": 1.03106594,
      "epoch": 0.0698030963475124,
      "flos": 26505199313280.0,
      "grad_norm": 1.7873573907411655,
      "language_loss": 0.75794333,
      "learning_rate": 3.983404744675437e-06,
      "loss": 0.78094292,
      "num_input_tokens_seen": 24782605,
      "step": 1161,
      "time_per_iteration": 2.5486936569213867
    },
    {
      "auxiliary_loss_clip": 0.01182771,
      "auxiliary_loss_mlp": 0.01062082,
      "balance_loss_clip": 1.06590068,
      "balance_loss_mlp": 1.03692842,
      "epoch": 0.06986321960018037,
      "flos": 23040430346880.0,
      "grad_norm": 2.2738955936392897,
      "language_loss": 0.82700789,
      "learning_rate": 3.9833546397985794e-06,
      "loss": 0.84945643,
      "num_input_tokens_seen": 24802910,
      "step": 1162,
      "time_per_iteration": 2.5651886463165283
    },
    {
      "auxiliary_loss_clip": 0.01188518,
      "auxiliary_loss_mlp": 0.01062209,
      "balance_loss_clip": 1.066378,
      "balance_loss_mlp": 1.03635275,
      "epoch": 0.06992334285284833,
      "flos": 28584822061440.0,
      "grad_norm": 2.1809663904259793,
      "language_loss": 0.79496068,
      "learning_rate": 3.983304459712716e-06,
      "loss": 0.81746793,
      "num_input_tokens_seen": 24823305,
      "step": 1163,
      "time_per_iteration": 2.575932502746582
    },
    {
      "auxiliary_loss_clip": 0.01195685,
      "auxiliary_loss_mlp": 0.01060575,
      "balance_loss_clip": 1.06514144,
      "balance_loss_mlp": 1.03495646,
      "epoch": 0.06998346610551631,
      "flos": 20595344670720.0,
      "grad_norm": 2.0650118353054636,
      "language_loss": 0.78857553,
      "learning_rate": 3.983254204419749e-06,
      "loss": 0.81113809,
      "num_input_tokens_seen": 24842155,
      "step": 1164,
      "time_per_iteration": 2.5194385051727295
    },
    {
      "auxiliary_loss_clip": 0.01154248,
      "auxiliary_loss_mlp": 0.01074027,
      "balance_loss_clip": 1.06568289,
      "balance_loss_mlp": 1.04647803,
      "epoch": 0.07004358935818428,
      "flos": 22528810978560.0,
      "grad_norm": 1.6865995448939972,
      "language_loss": 0.73009568,
      "learning_rate": 3.983203873921583e-06,
      "loss": 0.75237846,
      "num_input_tokens_seen": 24862080,
      "step": 1165,
      "time_per_iteration": 2.6458218097686768
    },
    {
      "auxiliary_loss_clip": 0.01186724,
      "auxiliary_loss_mlp": 0.01056736,
      "balance_loss_clip": 1.06491661,
      "balance_loss_mlp": 1.03281081,
      "epoch": 0.07010371261085224,
      "flos": 28949997680640.0,
      "grad_norm": 2.451425667487029,
      "language_loss": 0.80968428,
      "learning_rate": 3.983153468220128e-06,
      "loss": 0.83211887,
      "num_input_tokens_seen": 24886165,
      "step": 1166,
      "time_per_iteration": 2.6297972202301025
    },
    {
      "auxiliary_loss_clip": 0.01183632,
      "auxiliary_loss_mlp": 0.01047168,
      "balance_loss_clip": 1.0699507,
      "balance_loss_mlp": 1.02225327,
      "epoch": 0.07016383586352022,
      "flos": 23659171050240.0,
      "grad_norm": 2.408447230800185,
      "language_loss": 0.84560871,
      "learning_rate": 3.983102987317295e-06,
      "loss": 0.8679167,
      "num_input_tokens_seen": 24905775,
      "step": 1167,
      "time_per_iteration": 2.6032495498657227
    },
    {
      "auxiliary_loss_clip": 0.01200387,
      "auxiliary_loss_mlp": 0.01055876,
      "balance_loss_clip": 1.06852269,
      "balance_loss_mlp": 1.03098464,
      "epoch": 0.07022395911618819,
      "flos": 19792130693760.0,
      "grad_norm": 2.2498279048232885,
      "language_loss": 0.89519179,
      "learning_rate": 3.983052431214997e-06,
      "loss": 0.91775441,
      "num_input_tokens_seen": 24924295,
      "step": 1168,
      "time_per_iteration": 2.5281834602355957
    },
    {
      "auxiliary_loss_clip": 0.01193339,
      "auxiliary_loss_mlp": 0.01065073,
      "balance_loss_clip": 1.06564808,
      "balance_loss_mlp": 1.03724909,
      "epoch": 0.07028408236885615,
      "flos": 21689147675520.0,
      "grad_norm": 2.0912175778985413,
      "language_loss": 0.88953197,
      "learning_rate": 3.983001799915153e-06,
      "loss": 0.91211617,
      "num_input_tokens_seen": 24943210,
      "step": 1169,
      "time_per_iteration": 2.5864615440368652
    },
    {
      "auxiliary_loss_clip": 0.01215195,
      "auxiliary_loss_mlp": 0.01065213,
      "balance_loss_clip": 1.06798065,
      "balance_loss_mlp": 1.03945184,
      "epoch": 0.07034420562152413,
      "flos": 25630271832960.0,
      "grad_norm": 2.379615026133279,
      "language_loss": 0.83910972,
      "learning_rate": 3.982951093419681e-06,
      "loss": 0.8619138,
      "num_input_tokens_seen": 24960360,
      "step": 1170,
      "time_per_iteration": 2.5139658451080322
    },
    {
      "auxiliary_loss_clip": 0.01230706,
      "auxiliary_loss_mlp": 0.0081491,
      "balance_loss_clip": 1.14877701,
      "balance_loss_mlp": 1.05708098,
      "epoch": 0.0704043288741921,
      "flos": 20810449267200.0,
      "grad_norm": 2.0276202940663968,
      "language_loss": 0.75698322,
      "learning_rate": 3.982900311730506e-06,
      "loss": 0.77743936,
      "num_input_tokens_seen": 24978290,
      "step": 1171,
      "time_per_iteration": 2.604780435562134
    },
    {
      "auxiliary_loss_clip": 0.01183345,
      "auxiliary_loss_mlp": 0.01058747,
      "balance_loss_clip": 1.06594253,
      "balance_loss_mlp": 1.03451133,
      "epoch": 0.07046445212686006,
      "flos": 25593176062080.0,
      "grad_norm": 1.9942595762859727,
      "language_loss": 0.89049661,
      "learning_rate": 3.9828494548495514e-06,
      "loss": 0.91291749,
      "num_input_tokens_seen": 24997055,
      "step": 1172,
      "time_per_iteration": 2.5661892890930176
    },
    {
      "auxiliary_loss_clip": 0.01193993,
      "auxiliary_loss_mlp": 0.01054256,
      "balance_loss_clip": 1.06297743,
      "balance_loss_mlp": 1.02863765,
      "epoch": 0.07052457537952803,
      "flos": 25556978131200.0,
      "grad_norm": 4.853086079042627,
      "language_loss": 0.82086641,
      "learning_rate": 3.982798522778748e-06,
      "loss": 0.84334892,
      "num_input_tokens_seen": 25017490,
      "step": 1173,
      "time_per_iteration": 2.6214759349823
    },
    {
      "auxiliary_loss_clip": 0.01204165,
      "auxiliary_loss_mlp": 0.01058064,
      "balance_loss_clip": 1.06553257,
      "balance_loss_mlp": 1.03211236,
      "epoch": 0.070584698632196,
      "flos": 17968515154560.0,
      "grad_norm": 2.118853023502844,
      "language_loss": 0.82527643,
      "learning_rate": 3.9827475155200245e-06,
      "loss": 0.84789872,
      "num_input_tokens_seen": 25035660,
      "step": 1174,
      "time_per_iteration": 2.4745283126831055
    },
    {
      "auxiliary_loss_clip": 0.01186692,
      "auxiliary_loss_mlp": 0.01058767,
      "balance_loss_clip": 1.06431007,
      "balance_loss_mlp": 1.0335784,
      "epoch": 0.07064482188486397,
      "flos": 25370888745600.0,
      "grad_norm": 1.7783316230677921,
      "language_loss": 0.8485356,
      "learning_rate": 3.982696433075317e-06,
      "loss": 0.87099028,
      "num_input_tokens_seen": 25054785,
      "step": 1175,
      "time_per_iteration": 2.603943109512329
    },
    {
      "auxiliary_loss_clip": 0.01203466,
      "auxiliary_loss_mlp": 0.01069554,
      "balance_loss_clip": 1.06915951,
      "balance_loss_mlp": 1.04498458,
      "epoch": 0.07070494513753194,
      "flos": 24899848767360.0,
      "grad_norm": 1.7956910741854903,
      "language_loss": 0.83424342,
      "learning_rate": 3.982645275446563e-06,
      "loss": 0.85697365,
      "num_input_tokens_seen": 25075180,
      "step": 1176,
      "time_per_iteration": 2.562504529953003
    },
    {
      "auxiliary_loss_clip": 0.01149565,
      "auxiliary_loss_mlp": 0.01070728,
      "balance_loss_clip": 1.05205953,
      "balance_loss_mlp": 1.04398918,
      "epoch": 0.07076506839019991,
      "flos": 22338447874560.0,
      "grad_norm": 2.503694822698019,
      "language_loss": 0.74494392,
      "learning_rate": 3.982594042635701e-06,
      "loss": 0.76714683,
      "num_input_tokens_seen": 25093035,
      "step": 1177,
      "time_per_iteration": 2.5684187412261963
    },
    {
      "auxiliary_loss_clip": 0.01193759,
      "auxiliary_loss_mlp": 0.01066199,
      "balance_loss_clip": 1.06666112,
      "balance_loss_mlp": 1.04003274,
      "epoch": 0.07082519164286788,
      "flos": 18660800954880.0,
      "grad_norm": 1.9208109525459056,
      "language_loss": 0.85823941,
      "learning_rate": 3.982542734644673e-06,
      "loss": 0.88083899,
      "num_input_tokens_seen": 25112520,
      "step": 1178,
      "time_per_iteration": 2.542872428894043
    },
    {
      "auxiliary_loss_clip": 0.01083283,
      "auxiliary_loss_mlp": 0.01008745,
      "balance_loss_clip": 1.03682506,
      "balance_loss_mlp": 1.00409603,
      "epoch": 0.07088531489553584,
      "flos": 63654107610240.0,
      "grad_norm": 0.91247882605911,
      "language_loss": 0.63208199,
      "learning_rate": 3.982491351475427e-06,
      "loss": 0.65300226,
      "num_input_tokens_seen": 25177760,
      "step": 1179,
      "time_per_iteration": 3.2518677711486816
    },
    {
      "auxiliary_loss_clip": 0.01211694,
      "auxiliary_loss_mlp": 0.01065832,
      "balance_loss_clip": 1.06961322,
      "balance_loss_mlp": 1.04131114,
      "epoch": 0.07094543814820382,
      "flos": 21572688804480.0,
      "grad_norm": 5.037731513842757,
      "language_loss": 0.83243799,
      "learning_rate": 3.98243989312991e-06,
      "loss": 0.85521317,
      "num_input_tokens_seen": 25195260,
      "step": 1180,
      "time_per_iteration": 2.525660991668701
    },
    {
      "auxiliary_loss_clip": 0.01184167,
      "auxiliary_loss_mlp": 0.01063706,
      "balance_loss_clip": 1.06460118,
      "balance_loss_mlp": 1.038064,
      "epoch": 0.07100556140087179,
      "flos": 22089946608000.0,
      "grad_norm": 2.696060999376245,
      "language_loss": 0.88489223,
      "learning_rate": 3.982388359610074e-06,
      "loss": 0.90737098,
      "num_input_tokens_seen": 25212740,
      "step": 1181,
      "time_per_iteration": 2.545790672302246
    },
    {
      "auxiliary_loss_clip": 0.01189952,
      "auxiliary_loss_mlp": 0.01069703,
      "balance_loss_clip": 1.07089353,
      "balance_loss_mlp": 1.04452562,
      "epoch": 0.07106568465353975,
      "flos": 47922286400640.0,
      "grad_norm": 2.6002387288313242,
      "language_loss": 0.83758378,
      "learning_rate": 3.9823367509178725e-06,
      "loss": 0.86018038,
      "num_input_tokens_seen": 25236420,
      "step": 1182,
      "time_per_iteration": 2.766767740249634
    },
    {
      "auxiliary_loss_clip": 0.01204165,
      "auxiliary_loss_mlp": 0.01059387,
      "balance_loss_clip": 1.07266378,
      "balance_loss_mlp": 1.03262496,
      "epoch": 0.07112580790620772,
      "flos": 23440798316160.0,
      "grad_norm": 3.1273364546181304,
      "language_loss": 0.79313695,
      "learning_rate": 3.982285067055262e-06,
      "loss": 0.81577241,
      "num_input_tokens_seen": 25255120,
      "step": 1183,
      "time_per_iteration": 2.5836963653564453
    },
    {
      "auxiliary_loss_clip": 0.01216311,
      "auxiliary_loss_mlp": 0.01064705,
      "balance_loss_clip": 1.0660007,
      "balance_loss_mlp": 1.0386939,
      "epoch": 0.0711859311588757,
      "flos": 31868888682240.0,
      "grad_norm": 2.185604274808392,
      "language_loss": 0.79295337,
      "learning_rate": 3.982233308024204e-06,
      "loss": 0.81576347,
      "num_input_tokens_seen": 25275150,
      "step": 1184,
      "time_per_iteration": 2.5465188026428223
    },
    {
      "auxiliary_loss_clip": 0.01160314,
      "auxiliary_loss_mlp": 0.01067199,
      "balance_loss_clip": 1.06793237,
      "balance_loss_mlp": 1.04209352,
      "epoch": 0.07124605441154366,
      "flos": 19610315026560.0,
      "grad_norm": 2.2792059595597585,
      "language_loss": 0.76754522,
      "learning_rate": 3.98218147382666e-06,
      "loss": 0.78982031,
      "num_input_tokens_seen": 25293680,
      "step": 1185,
      "time_per_iteration": 2.615647554397583
    },
    {
      "auxiliary_loss_clip": 0.01214563,
      "auxiliary_loss_mlp": 0.01067227,
      "balance_loss_clip": 1.06807399,
      "balance_loss_mlp": 1.04143071,
      "epoch": 0.07130617766421163,
      "flos": 14684448533760.0,
      "grad_norm": 2.997435203268458,
      "language_loss": 0.65342814,
      "learning_rate": 3.982129564464596e-06,
      "loss": 0.67624605,
      "num_input_tokens_seen": 25310050,
      "step": 1186,
      "time_per_iteration": 2.4411468505859375
    },
    {
      "auxiliary_loss_clip": 0.01201222,
      "auxiliary_loss_mlp": 0.01058876,
      "balance_loss_clip": 1.07055187,
      "balance_loss_mlp": 1.03328156,
      "epoch": 0.07136630091687961,
      "flos": 26067915141120.0,
      "grad_norm": 2.181661753030015,
      "language_loss": 0.70020473,
      "learning_rate": 3.98207757993998e-06,
      "loss": 0.72280574,
      "num_input_tokens_seen": 25331020,
      "step": 1187,
      "time_per_iteration": 2.5722804069519043
    },
    {
      "auxiliary_loss_clip": 0.01164949,
      "auxiliary_loss_mlp": 0.0105858,
      "balance_loss_clip": 1.06971073,
      "balance_loss_mlp": 1.03479755,
      "epoch": 0.07142642416954757,
      "flos": 15669190869120.0,
      "grad_norm": 2.759748590360436,
      "language_loss": 0.79341507,
      "learning_rate": 3.9820255202547845e-06,
      "loss": 0.81565034,
      "num_input_tokens_seen": 25347875,
      "step": 1188,
      "time_per_iteration": 2.5465219020843506
    },
    {
      "auxiliary_loss_clip": 0.01215599,
      "auxiliary_loss_mlp": 0.01062855,
      "balance_loss_clip": 1.07120919,
      "balance_loss_mlp": 1.03721356,
      "epoch": 0.07148654742221554,
      "flos": 19755322231680.0,
      "grad_norm": 1.902467531249299,
      "language_loss": 0.84652185,
      "learning_rate": 3.981973385410981e-06,
      "loss": 0.86930645,
      "num_input_tokens_seen": 25366715,
      "step": 1189,
      "time_per_iteration": 2.499077320098877
    },
    {
      "auxiliary_loss_clip": 0.0117672,
      "auxiliary_loss_mlp": 0.00799794,
      "balance_loss_clip": 1.06718588,
      "balance_loss_mlp": 1.02726746,
      "epoch": 0.07154667067488352,
      "flos": 23471824688640.0,
      "grad_norm": 2.7426786993965138,
      "language_loss": 0.76727188,
      "learning_rate": 3.9819211754105494e-06,
      "loss": 0.78703701,
      "num_input_tokens_seen": 25385450,
      "step": 1190,
      "time_per_iteration": 3.9924471378326416
    },
    {
      "auxiliary_loss_clip": 0.01215541,
      "auxiliary_loss_mlp": 0.01073321,
      "balance_loss_clip": 1.0683856,
      "balance_loss_mlp": 1.04587901,
      "epoch": 0.07160679392755148,
      "flos": 18332936588160.0,
      "grad_norm": 2.0581370917741344,
      "language_loss": 0.75650471,
      "learning_rate": 3.981868890255468e-06,
      "loss": 0.77939332,
      "num_input_tokens_seen": 25403940,
      "step": 1191,
      "time_per_iteration": 2.4657485485076904
    },
    {
      "auxiliary_loss_clip": 0.01174098,
      "auxiliary_loss_mlp": 0.01057043,
      "balance_loss_clip": 1.06450653,
      "balance_loss_mlp": 1.02950549,
      "epoch": 0.07166691718021945,
      "flos": 17747017937280.0,
      "grad_norm": 2.4257656857668546,
      "language_loss": 0.73760068,
      "learning_rate": 3.981816529947719e-06,
      "loss": 0.75991213,
      "num_input_tokens_seen": 25420410,
      "step": 1192,
      "time_per_iteration": 3.9080147743225098
    },
    {
      "auxiliary_loss_clip": 0.0120921,
      "auxiliary_loss_mlp": 0.01054245,
      "balance_loss_clip": 1.06358385,
      "balance_loss_mlp": 1.0299263,
      "epoch": 0.07172704043288743,
      "flos": 22451925916800.0,
      "grad_norm": 2.432473034795723,
      "language_loss": 0.78139824,
      "learning_rate": 3.9817640944892896e-06,
      "loss": 0.8040328,
      "num_input_tokens_seen": 25439415,
      "step": 1193,
      "time_per_iteration": 2.478437662124634
    },
    {
      "auxiliary_loss_clip": 0.01193303,
      "auxiliary_loss_mlp": 0.01058219,
      "balance_loss_clip": 1.07088089,
      "balance_loss_mlp": 1.03238606,
      "epoch": 0.07178716368555539,
      "flos": 23222210100480.0,
      "grad_norm": 1.9912876299860613,
      "language_loss": 0.85382998,
      "learning_rate": 3.981711583882166e-06,
      "loss": 0.87634522,
      "num_input_tokens_seen": 25458715,
      "step": 1194,
      "time_per_iteration": 2.5872788429260254
    },
    {
      "auxiliary_loss_clip": 0.0119767,
      "auxiliary_loss_mlp": 0.01066727,
      "balance_loss_clip": 1.0713203,
      "balance_loss_mlp": 1.04137158,
      "epoch": 0.07184728693822336,
      "flos": 25150828072320.0,
      "grad_norm": 2.49217318708141,
      "language_loss": 0.81845641,
      "learning_rate": 3.981658998128341e-06,
      "loss": 0.84110039,
      "num_input_tokens_seen": 25477985,
      "step": 1195,
      "time_per_iteration": 3.937192678451538
    },
    {
      "auxiliary_loss_clip": 0.01176478,
      "auxiliary_loss_mlp": 0.01057096,
      "balance_loss_clip": 1.06785274,
      "balance_loss_mlp": 1.03293252,
      "epoch": 0.07190741019089132,
      "flos": 22711237176960.0,
      "grad_norm": 1.937818594780431,
      "language_loss": 0.79854977,
      "learning_rate": 3.981606337229808e-06,
      "loss": 0.82088548,
      "num_input_tokens_seen": 25497110,
      "step": 1196,
      "time_per_iteration": 2.607423782348633
    },
    {
      "auxiliary_loss_clip": 0.01184528,
      "auxiliary_loss_mlp": 0.00797301,
      "balance_loss_clip": 1.06800556,
      "balance_loss_mlp": 1.01831508,
      "epoch": 0.0719675334435593,
      "flos": 29349791032320.0,
      "grad_norm": 2.787793865224825,
      "language_loss": 0.70888603,
      "learning_rate": 3.9815536011885655e-06,
      "loss": 0.72870427,
      "num_input_tokens_seen": 25516555,
      "step": 1197,
      "time_per_iteration": 2.641493082046509
    },
    {
      "auxiliary_loss_clip": 0.01162825,
      "auxiliary_loss_mlp": 0.01056886,
      "balance_loss_clip": 1.0651269,
      "balance_loss_mlp": 1.03154182,
      "epoch": 0.07202765669622727,
      "flos": 17639788861440.0,
      "grad_norm": 2.0350242378403762,
      "language_loss": 0.86015046,
      "learning_rate": 3.98150079000661e-06,
      "loss": 0.88234758,
      "num_input_tokens_seen": 25533895,
      "step": 1198,
      "time_per_iteration": 4.077868461608887
    },
    {
      "auxiliary_loss_clip": 0.01167777,
      "auxiliary_loss_mlp": 0.01065492,
      "balance_loss_clip": 1.0653677,
      "balance_loss_mlp": 1.04007626,
      "epoch": 0.07208777994889523,
      "flos": 21434038306560.0,
      "grad_norm": 2.231543778523985,
      "language_loss": 0.83816004,
      "learning_rate": 3.981447903685947e-06,
      "loss": 0.86049277,
      "num_input_tokens_seen": 25554195,
      "step": 1199,
      "time_per_iteration": 2.6173348426818848
    },
    {
      "auxiliary_loss_clip": 0.01218952,
      "auxiliary_loss_mlp": 0.01057267,
      "balance_loss_clip": 1.07312489,
      "balance_loss_mlp": 1.03307986,
      "epoch": 0.07214790320156321,
      "flos": 26940867373440.0,
      "grad_norm": 2.1877017054245136,
      "language_loss": 0.76951146,
      "learning_rate": 3.981394942228581e-06,
      "loss": 0.79227364,
      "num_input_tokens_seen": 25574155,
      "step": 1200,
      "time_per_iteration": 2.555222511291504
    },
    {
      "auxiliary_loss_clip": 0.01194527,
      "auxiliary_loss_mlp": 0.01070903,
      "balance_loss_clip": 1.06888199,
      "balance_loss_mlp": 1.04510629,
      "epoch": 0.07220802645423118,
      "flos": 23879949995520.0,
      "grad_norm": 2.398704174967709,
      "language_loss": 0.82372916,
      "learning_rate": 3.98134190563652e-06,
      "loss": 0.84638345,
      "num_input_tokens_seen": 25592735,
      "step": 1201,
      "time_per_iteration": 2.540445327758789
    },
    {
      "auxiliary_loss_clip": 0.01202862,
      "auxiliary_loss_mlp": 0.01060337,
      "balance_loss_clip": 1.06654203,
      "balance_loss_mlp": 1.03307366,
      "epoch": 0.07226814970689914,
      "flos": 19243631036160.0,
      "grad_norm": 2.4844168720150637,
      "language_loss": 0.68989384,
      "learning_rate": 3.981288793911775e-06,
      "loss": 0.71252578,
      "num_input_tokens_seen": 25611510,
      "step": 1202,
      "time_per_iteration": 2.5117533206939697
    },
    {
      "auxiliary_loss_clip": 0.01190827,
      "auxiliary_loss_mlp": 0.00797477,
      "balance_loss_clip": 1.07042778,
      "balance_loss_mlp": 1.02247047,
      "epoch": 0.07232827295956712,
      "flos": 19172025273600.0,
      "grad_norm": 2.1738820401864185,
      "language_loss": 0.87562704,
      "learning_rate": 3.98123560705636e-06,
      "loss": 0.89551008,
      "num_input_tokens_seen": 25629560,
      "step": 1203,
      "time_per_iteration": 2.5440328121185303
    },
    {
      "auxiliary_loss_clip": 0.01165123,
      "auxiliary_loss_mlp": 0.01069527,
      "balance_loss_clip": 1.06198359,
      "balance_loss_mlp": 1.0437181,
      "epoch": 0.07238839621223508,
      "flos": 17639752947840.0,
      "grad_norm": 1.9823673486944118,
      "language_loss": 0.78466356,
      "learning_rate": 3.981182345072293e-06,
      "loss": 0.80701005,
      "num_input_tokens_seen": 25648330,
      "step": 1204,
      "time_per_iteration": 2.6112170219421387
    },
    {
      "auxiliary_loss_clip": 0.01200281,
      "auxiliary_loss_mlp": 0.0107023,
      "balance_loss_clip": 1.06807721,
      "balance_loss_mlp": 1.0449934,
      "epoch": 0.07244851946490305,
      "flos": 28292401440000.0,
      "grad_norm": 1.8275814698892219,
      "language_loss": 0.8218677,
      "learning_rate": 3.981129007961593e-06,
      "loss": 0.84457278,
      "num_input_tokens_seen": 25669470,
      "step": 1205,
      "time_per_iteration": 2.573889970779419
    },
    {
      "auxiliary_loss_clip": 0.01187683,
      "auxiliary_loss_mlp": 0.0079391,
      "balance_loss_clip": 1.06963205,
      "balance_loss_mlp": 1.01633668,
      "epoch": 0.07250864271757101,
      "flos": 22564829341440.0,
      "grad_norm": 2.3330723865503264,
      "language_loss": 0.76684058,
      "learning_rate": 3.981075595726283e-06,
      "loss": 0.7866565,
      "num_input_tokens_seen": 25690470,
      "step": 1206,
      "time_per_iteration": 2.5994017124176025
    },
    {
      "auxiliary_loss_clip": 0.01195918,
      "auxiliary_loss_mlp": 0.0106011,
      "balance_loss_clip": 1.0697,
      "balance_loss_mlp": 1.03450418,
      "epoch": 0.072568765970239,
      "flos": 21762405463680.0,
      "grad_norm": 1.9244264359581478,
      "language_loss": 0.77764928,
      "learning_rate": 3.981022108368387e-06,
      "loss": 0.80020952,
      "num_input_tokens_seen": 25709205,
      "step": 1207,
      "time_per_iteration": 2.506878137588501
    },
    {
      "auxiliary_loss_clip": 0.01194729,
      "auxiliary_loss_mlp": 0.01055464,
      "balance_loss_clip": 1.06556308,
      "balance_loss_mlp": 1.03160977,
      "epoch": 0.07262888922290696,
      "flos": 25519702792320.0,
      "grad_norm": 3.543086582320235,
      "language_loss": 0.7955451,
      "learning_rate": 3.9809685458899345e-06,
      "loss": 0.81804699,
      "num_input_tokens_seen": 25728485,
      "step": 1208,
      "time_per_iteration": 2.5797102451324463
    },
    {
      "auxiliary_loss_clip": 0.01193765,
      "auxiliary_loss_mlp": 0.0105732,
      "balance_loss_clip": 1.06537342,
      "balance_loss_mlp": 1.03382349,
      "epoch": 0.07268901247557492,
      "flos": 21246548290560.0,
      "grad_norm": 2.5006084242123596,
      "language_loss": 0.7888763,
      "learning_rate": 3.980914908292955e-06,
      "loss": 0.81138718,
      "num_input_tokens_seen": 25747730,
      "step": 1209,
      "time_per_iteration": 2.5232326984405518
    },
    {
      "auxiliary_loss_clip": 0.01199913,
      "auxiliary_loss_mlp": 0.01063797,
      "balance_loss_clip": 1.06651545,
      "balance_loss_mlp": 1.03964496,
      "epoch": 0.0727491357282429,
      "flos": 25479302970240.0,
      "grad_norm": 2.357422485467626,
      "language_loss": 0.81396508,
      "learning_rate": 3.980861195579486e-06,
      "loss": 0.83660215,
      "num_input_tokens_seen": 25768050,
      "step": 1210,
      "time_per_iteration": 2.5782530307769775
    },
    {
      "auxiliary_loss_clip": 0.01185982,
      "auxiliary_loss_mlp": 0.01068143,
      "balance_loss_clip": 1.06947231,
      "balance_loss_mlp": 1.04267967,
      "epoch": 0.07280925898091087,
      "flos": 24462169545600.0,
      "grad_norm": 2.7616279220963627,
      "language_loss": 0.84537047,
      "learning_rate": 3.98080740775156e-06,
      "loss": 0.8679117,
      "num_input_tokens_seen": 25787985,
      "step": 1211,
      "time_per_iteration": 2.5680668354034424
    },
    {
      "auxiliary_loss_clip": 0.01173146,
      "auxiliary_loss_mlp": 0.01055217,
      "balance_loss_clip": 1.06400347,
      "balance_loss_mlp": 1.03091002,
      "epoch": 0.07286938223357883,
      "flos": 18288191220480.0,
      "grad_norm": 2.7774349724346505,
      "language_loss": 0.90878206,
      "learning_rate": 3.98075354481122e-06,
      "loss": 0.93106568,
      "num_input_tokens_seen": 25803620,
      "step": 1212,
      "time_per_iteration": 2.5289597511291504
    },
    {
      "auxiliary_loss_clip": 0.01212581,
      "auxiliary_loss_mlp": 0.01053667,
      "balance_loss_clip": 1.06973803,
      "balance_loss_mlp": 1.02957487,
      "epoch": 0.07292950548624681,
      "flos": 21214803646080.0,
      "grad_norm": 1.8046083526528385,
      "language_loss": 0.72789836,
      "learning_rate": 3.9806996067605055e-06,
      "loss": 0.75056082,
      "num_input_tokens_seen": 25823315,
      "step": 1213,
      "time_per_iteration": 2.4908602237701416
    },
    {
      "auxiliary_loss_clip": 0.01162268,
      "auxiliary_loss_mlp": 0.01056773,
      "balance_loss_clip": 1.06477261,
      "balance_loss_mlp": 1.0320375,
      "epoch": 0.07298962873891478,
      "flos": 24642009964800.0,
      "grad_norm": 1.9692764154646545,
      "language_loss": 0.84317774,
      "learning_rate": 3.980645593601465e-06,
      "loss": 0.86536813,
      "num_input_tokens_seen": 25842605,
      "step": 1214,
      "time_per_iteration": 2.6685433387756348
    },
    {
      "auxiliary_loss_clip": 0.01216763,
      "auxiliary_loss_mlp": 0.01059925,
      "balance_loss_clip": 1.06965017,
      "balance_loss_mlp": 1.03437805,
      "epoch": 0.07304975199158274,
      "flos": 27052765217280.0,
      "grad_norm": 2.1411961938525756,
      "language_loss": 0.84169424,
      "learning_rate": 3.980591505336144e-06,
      "loss": 0.86446112,
      "num_input_tokens_seen": 25863030,
      "step": 1215,
      "time_per_iteration": 2.529304265975952
    },
    {
      "auxiliary_loss_clip": 0.01199013,
      "auxiliary_loss_mlp": 0.01061761,
      "balance_loss_clip": 1.1310364,
      "balance_loss_mlp": 1.03638148,
      "epoch": 0.07310987524425071,
      "flos": 33549544091520.0,
      "grad_norm": 3.5015031401045085,
      "language_loss": 0.81053972,
      "learning_rate": 3.980537341966595e-06,
      "loss": 0.83314741,
      "num_input_tokens_seen": 25888015,
      "step": 1216,
      "time_per_iteration": 2.708005428314209
    },
    {
      "auxiliary_loss_clip": 0.01167847,
      "auxiliary_loss_mlp": 0.01057722,
      "balance_loss_clip": 1.05554605,
      "balance_loss_mlp": 1.03402328,
      "epoch": 0.07316999849691869,
      "flos": 28110944908800.0,
      "grad_norm": 2.473932988038879,
      "language_loss": 0.76125169,
      "learning_rate": 3.980483103494872e-06,
      "loss": 0.78350741,
      "num_input_tokens_seen": 25908660,
      "step": 1217,
      "time_per_iteration": 2.588402509689331
    },
    {
      "auxiliary_loss_clip": 0.01176315,
      "auxiliary_loss_mlp": 0.01061517,
      "balance_loss_clip": 1.06679177,
      "balance_loss_mlp": 1.03876019,
      "epoch": 0.07323012174958665,
      "flos": 14392602529920.0,
      "grad_norm": 2.1724715232538485,
      "language_loss": 0.8690449,
      "learning_rate": 3.98042878992303e-06,
      "loss": 0.89142323,
      "num_input_tokens_seen": 25927215,
      "step": 1218,
      "time_per_iteration": 2.593597888946533
    },
    {
      "auxiliary_loss_clip": 0.01200673,
      "auxiliary_loss_mlp": 0.01065401,
      "balance_loss_clip": 1.06702137,
      "balance_loss_mlp": 1.04122555,
      "epoch": 0.07329024500225462,
      "flos": 21616428591360.0,
      "grad_norm": 2.1426433930107502,
      "language_loss": 0.86631531,
      "learning_rate": 3.9803744012531305e-06,
      "loss": 0.88897604,
      "num_input_tokens_seen": 25945500,
      "step": 1219,
      "time_per_iteration": 2.52473783493042
    },
    {
      "auxiliary_loss_clip": 0.0121,
      "auxiliary_loss_mlp": 0.0105692,
      "balance_loss_clip": 1.06746864,
      "balance_loss_mlp": 1.03410363,
      "epoch": 0.0733503682549226,
      "flos": 13224141106560.0,
      "grad_norm": 3.305342997540625,
      "language_loss": 0.8493489,
      "learning_rate": 3.980319937487235e-06,
      "loss": 0.87201804,
      "num_input_tokens_seen": 25963105,
      "step": 1220,
      "time_per_iteration": 2.46592378616333
    },
    {
      "auxiliary_loss_clip": 0.01165873,
      "auxiliary_loss_mlp": 0.01061011,
      "balance_loss_clip": 1.05964828,
      "balance_loss_mlp": 1.03679991,
      "epoch": 0.07341049150759056,
      "flos": 20886975192960.0,
      "grad_norm": 3.8939684076398953,
      "language_loss": 0.7698487,
      "learning_rate": 3.98026539862741e-06,
      "loss": 0.79211754,
      "num_input_tokens_seen": 25981690,
      "step": 1221,
      "time_per_iteration": 2.571197748184204
    },
    {
      "auxiliary_loss_clip": 0.01169132,
      "auxiliary_loss_mlp": 0.01062984,
      "balance_loss_clip": 1.06757903,
      "balance_loss_mlp": 1.03905916,
      "epoch": 0.07347061476025853,
      "flos": 15413614623360.0,
      "grad_norm": 1.8133769186156972,
      "language_loss": 0.9182173,
      "learning_rate": 3.980210784675722e-06,
      "loss": 0.94053853,
      "num_input_tokens_seen": 25999890,
      "step": 1222,
      "time_per_iteration": 2.628995895385742
    },
    {
      "auxiliary_loss_clip": 0.01146531,
      "auxiliary_loss_mlp": 0.01057682,
      "balance_loss_clip": 1.06302798,
      "balance_loss_mlp": 1.0343051,
      "epoch": 0.0735307380129265,
      "flos": 11108859131520.0,
      "grad_norm": 2.462969248388403,
      "language_loss": 0.905487,
      "learning_rate": 3.980156095634242e-06,
      "loss": 0.9275291,
      "num_input_tokens_seen": 26016445,
      "step": 1223,
      "time_per_iteration": 2.6526036262512207
    },
    {
      "auxiliary_loss_clip": 0.01212245,
      "auxiliary_loss_mlp": 0.01070113,
      "balance_loss_clip": 1.06858075,
      "balance_loss_mlp": 1.04604506,
      "epoch": 0.07359086126559447,
      "flos": 23732392924800.0,
      "grad_norm": 1.9685783344553158,
      "language_loss": 0.82043421,
      "learning_rate": 3.980101331505045e-06,
      "loss": 0.84325778,
      "num_input_tokens_seen": 26036080,
      "step": 1224,
      "time_per_iteration": 2.5358564853668213
    },
    {
      "auxiliary_loss_clip": 0.01207861,
      "auxiliary_loss_mlp": 0.01058427,
      "balance_loss_clip": 1.06556463,
      "balance_loss_mlp": 1.03249931,
      "epoch": 0.07365098451826244,
      "flos": 20993270515200.0,
      "grad_norm": 2.512858573433487,
      "language_loss": 0.83136499,
      "learning_rate": 3.9800464922902076e-06,
      "loss": 0.85402787,
      "num_input_tokens_seen": 26055805,
      "step": 1225,
      "time_per_iteration": 2.538318395614624
    },
    {
      "auxiliary_loss_clip": 0.01176354,
      "auxiliary_loss_mlp": 0.01058111,
      "balance_loss_clip": 1.0642333,
      "balance_loss_mlp": 1.03337502,
      "epoch": 0.0737111077709304,
      "flos": 19933582452480.0,
      "grad_norm": 1.8229889990168386,
      "language_loss": 0.90514988,
      "learning_rate": 3.979991577991808e-06,
      "loss": 0.92749459,
      "num_input_tokens_seen": 26073905,
      "step": 1226,
      "time_per_iteration": 2.5840744972229004
    },
    {
      "auxiliary_loss_clip": 0.01218461,
      "auxiliary_loss_mlp": 0.01051263,
      "balance_loss_clip": 1.06613278,
      "balance_loss_mlp": 1.0262413,
      "epoch": 0.07377123102359838,
      "flos": 16581537342720.0,
      "grad_norm": 2.8602489682009664,
      "language_loss": 0.76921117,
      "learning_rate": 3.97993658861193e-06,
      "loss": 0.79190838,
      "num_input_tokens_seen": 26091700,
      "step": 1227,
      "time_per_iteration": 2.491011381149292
    },
    {
      "auxiliary_loss_clip": 0.01194869,
      "auxiliary_loss_mlp": 0.01052756,
      "balance_loss_clip": 1.06692696,
      "balance_loss_mlp": 1.02868795,
      "epoch": 0.07383135427626634,
      "flos": 28328563457280.0,
      "grad_norm": 1.4652449490589148,
      "language_loss": 0.85645741,
      "learning_rate": 3.9798815241526575e-06,
      "loss": 0.87893367,
      "num_input_tokens_seen": 26114105,
      "step": 1228,
      "time_per_iteration": 3.9786975383758545
    },
    {
      "auxiliary_loss_clip": 0.01195063,
      "auxiliary_loss_mlp": 0.01054569,
      "balance_loss_clip": 1.06334829,
      "balance_loss_mlp": 1.03103662,
      "epoch": 0.07389147752893431,
      "flos": 20047168235520.0,
      "grad_norm": 2.193714897879391,
      "language_loss": 0.79796851,
      "learning_rate": 3.97982638461608e-06,
      "loss": 0.82046485,
      "num_input_tokens_seen": 26131165,
      "step": 1229,
      "time_per_iteration": 2.5239813327789307
    },
    {
      "auxiliary_loss_clip": 0.01198655,
      "auxiliary_loss_mlp": 0.00803293,
      "balance_loss_clip": 1.06507528,
      "balance_loss_mlp": 1.03532076,
      "epoch": 0.07395160078160229,
      "flos": 18114132890880.0,
      "grad_norm": 2.0434423696237207,
      "language_loss": 0.78048635,
      "learning_rate": 3.979771170004287e-06,
      "loss": 0.80050582,
      "num_input_tokens_seen": 26150040,
      "step": 1230,
      "time_per_iteration": 3.881772756576538
    },
    {
      "auxiliary_loss_clip": 0.0120977,
      "auxiliary_loss_mlp": 0.0105056,
      "balance_loss_clip": 1.06776762,
      "balance_loss_mlp": 1.02575254,
      "epoch": 0.07401172403427025,
      "flos": 23586918842880.0,
      "grad_norm": 1.970803162220986,
      "language_loss": 0.81341505,
      "learning_rate": 3.979715880319372e-06,
      "loss": 0.83601832,
      "num_input_tokens_seen": 26169380,
      "step": 1231,
      "time_per_iteration": 2.4915425777435303
    },
    {
      "auxiliary_loss_clip": 0.01225826,
      "auxiliary_loss_mlp": 0.01067904,
      "balance_loss_clip": 1.13030195,
      "balance_loss_mlp": 1.04322755,
      "epoch": 0.07407184728693822,
      "flos": 26359904799360.0,
      "grad_norm": 2.1041716610920305,
      "language_loss": 0.95045739,
      "learning_rate": 3.979660515563434e-06,
      "loss": 0.97339469,
      "num_input_tokens_seen": 26189420,
      "step": 1232,
      "time_per_iteration": 2.5925018787384033
    },
    {
      "auxiliary_loss_clip": 0.01193464,
      "auxiliary_loss_mlp": 0.01065024,
      "balance_loss_clip": 1.06739712,
      "balance_loss_mlp": 1.04175484,
      "epoch": 0.0741319705396062,
      "flos": 22200443821440.0,
      "grad_norm": 1.7772434679745015,
      "language_loss": 0.80863416,
      "learning_rate": 3.979605075738569e-06,
      "loss": 0.83121908,
      "num_input_tokens_seen": 26209300,
      "step": 1233,
      "time_per_iteration": 2.5042660236358643
    },
    {
      "auxiliary_loss_clip": 0.01213845,
      "auxiliary_loss_mlp": 0.01063909,
      "balance_loss_clip": 1.065925,
      "balance_loss_mlp": 1.03662229,
      "epoch": 0.07419209379227416,
      "flos": 39200482523520.0,
      "grad_norm": 2.401168986623431,
      "language_loss": 0.70857382,
      "learning_rate": 3.979549560846883e-06,
      "loss": 0.73135138,
      "num_input_tokens_seen": 26228110,
      "step": 1234,
      "time_per_iteration": 4.008370637893677
    },
    {
      "auxiliary_loss_clip": 0.01179001,
      "auxiliary_loss_mlp": 0.01072177,
      "balance_loss_clip": 1.06553078,
      "balance_loss_mlp": 1.04536641,
      "epoch": 0.07425221704494213,
      "flos": 22781657790720.0,
      "grad_norm": 1.9245137227351239,
      "language_loss": 0.77483892,
      "learning_rate": 3.979493970890478e-06,
      "loss": 0.79735065,
      "num_input_tokens_seen": 26247020,
      "step": 1235,
      "time_per_iteration": 2.544532299041748
    },
    {
      "auxiliary_loss_clip": 0.01206293,
      "auxiliary_loss_mlp": 0.01055388,
      "balance_loss_clip": 1.0651567,
      "balance_loss_mlp": 1.03159392,
      "epoch": 0.0743123402976101,
      "flos": 22272983337600.0,
      "grad_norm": 1.9890696111345716,
      "language_loss": 0.83020198,
      "learning_rate": 3.979438305871464e-06,
      "loss": 0.85281885,
      "num_input_tokens_seen": 26265750,
      "step": 1236,
      "time_per_iteration": 2.4901630878448486
    },
    {
      "auxiliary_loss_clip": 0.01160047,
      "auxiliary_loss_mlp": 0.00794154,
      "balance_loss_clip": 1.06171668,
      "balance_loss_mlp": 1.01765919,
      "epoch": 0.07437246355027807,
      "flos": 29315029645440.0,
      "grad_norm": 1.7067638454671585,
      "language_loss": 0.75762439,
      "learning_rate": 3.979382565791951e-06,
      "loss": 0.77716649,
      "num_input_tokens_seen": 26287905,
      "step": 1237,
      "time_per_iteration": 4.071123838424683
    },
    {
      "auxiliary_loss_clip": 0.01145521,
      "auxiliary_loss_mlp": 0.00793981,
      "balance_loss_clip": 1.06203604,
      "balance_loss_mlp": 1.01531601,
      "epoch": 0.07443258680294604,
      "flos": 31944732249600.0,
      "grad_norm": 1.6931481053439517,
      "language_loss": 0.77491069,
      "learning_rate": 3.979326750654053e-06,
      "loss": 0.79430574,
      "num_input_tokens_seen": 26311795,
      "step": 1238,
      "time_per_iteration": 2.696892261505127
    },
    {
      "auxiliary_loss_clip": 0.01182517,
      "auxiliary_loss_mlp": 0.01059301,
      "balance_loss_clip": 1.06205595,
      "balance_loss_mlp": 1.03489912,
      "epoch": 0.074492710055614,
      "flos": 22675290641280.0,
      "grad_norm": 1.9605511041995878,
      "language_loss": 0.86468959,
      "learning_rate": 3.9792708604598854e-06,
      "loss": 0.88710779,
      "num_input_tokens_seen": 26330330,
      "step": 1239,
      "time_per_iteration": 2.5642828941345215
    },
    {
      "auxiliary_loss_clip": 0.01169271,
      "auxiliary_loss_mlp": 0.01052527,
      "balance_loss_clip": 1.06347513,
      "balance_loss_mlp": 1.02701616,
      "epoch": 0.07455283330828198,
      "flos": 21284901037440.0,
      "grad_norm": 2.0159037462177753,
      "language_loss": 0.8854124,
      "learning_rate": 3.979214895211569e-06,
      "loss": 0.90763044,
      "num_input_tokens_seen": 26348865,
      "step": 1240,
      "time_per_iteration": 2.5696630477905273
    },
    {
      "auxiliary_loss_clip": 0.01179907,
      "auxiliary_loss_mlp": 0.01063872,
      "balance_loss_clip": 1.06437361,
      "balance_loss_mlp": 1.03743172,
      "epoch": 0.07461295656094995,
      "flos": 24388408967040.0,
      "grad_norm": 1.7217903520696973,
      "language_loss": 0.8870734,
      "learning_rate": 3.979158854911225e-06,
      "loss": 0.90951121,
      "num_input_tokens_seen": 26368210,
      "step": 1241,
      "time_per_iteration": 2.591825008392334
    },
    {
      "auxiliary_loss_clip": 0.01074126,
      "auxiliary_loss_mlp": 0.01003846,
      "balance_loss_clip": 1.03139472,
      "balance_loss_mlp": 0.99962628,
      "epoch": 0.07467307981361791,
      "flos": 62109660574080.0,
      "grad_norm": 0.8892070443971893,
      "language_loss": 0.63118911,
      "learning_rate": 3.979102739560979e-06,
      "loss": 0.65196884,
      "num_input_tokens_seen": 26424890,
      "step": 1242,
      "time_per_iteration": 3.1899678707122803
    },
    {
      "auxiliary_loss_clip": 0.01170987,
      "auxiliary_loss_mlp": 0.01065727,
      "balance_loss_clip": 1.06281996,
      "balance_loss_mlp": 1.03673518,
      "epoch": 0.07473320306628589,
      "flos": 24863148046080.0,
      "grad_norm": 2.209222373428118,
      "language_loss": 0.62898684,
      "learning_rate": 3.9790465491629595e-06,
      "loss": 0.65135396,
      "num_input_tokens_seen": 26446405,
      "step": 1243,
      "time_per_iteration": 2.6175901889801025
    },
    {
      "auxiliary_loss_clip": 0.01193865,
      "auxiliary_loss_mlp": 0.01058556,
      "balance_loss_clip": 1.06227231,
      "balance_loss_mlp": 1.03329515,
      "epoch": 0.07479332631895386,
      "flos": 24897442556160.0,
      "grad_norm": 1.8121215080235986,
      "language_loss": 0.76420856,
      "learning_rate": 3.978990283719296e-06,
      "loss": 0.78673279,
      "num_input_tokens_seen": 26466070,
      "step": 1244,
      "time_per_iteration": 2.573676347732544
    },
    {
      "auxiliary_loss_clip": 0.01186545,
      "auxiliary_loss_mlp": 0.00790778,
      "balance_loss_clip": 1.0632453,
      "balance_loss_mlp": 1.01155066,
      "epoch": 0.07485344957162182,
      "flos": 17815247821440.0,
      "grad_norm": 3.2470050273282554,
      "language_loss": 0.68727005,
      "learning_rate": 3.978933943232123e-06,
      "loss": 0.70704323,
      "num_input_tokens_seen": 26479350,
      "step": 1245,
      "time_per_iteration": 2.501337766647339
    },
    {
      "auxiliary_loss_clip": 0.01206211,
      "auxiliary_loss_mlp": 0.01058363,
      "balance_loss_clip": 1.06255531,
      "balance_loss_mlp": 1.03323388,
      "epoch": 0.0749135728242898,
      "flos": 25010202326400.0,
      "grad_norm": 1.8220562544163381,
      "language_loss": 0.88734066,
      "learning_rate": 3.978877527703576e-06,
      "loss": 0.90998638,
      "num_input_tokens_seen": 26498255,
      "step": 1246,
      "time_per_iteration": 2.528078556060791
    },
    {
      "auxiliary_loss_clip": 0.01215589,
      "auxiliary_loss_mlp": 0.01075493,
      "balance_loss_clip": 1.06375861,
      "balance_loss_mlp": 1.04848003,
      "epoch": 0.07497369607695777,
      "flos": 17822071405440.0,
      "grad_norm": 4.4635089513940915,
      "language_loss": 0.88105309,
      "learning_rate": 3.9788210371357945e-06,
      "loss": 0.90396392,
      "num_input_tokens_seen": 26515375,
      "step": 1247,
      "time_per_iteration": 2.4396097660064697
    },
    {
      "auxiliary_loss_clip": 0.01184817,
      "auxiliary_loss_mlp": 0.01066776,
      "balance_loss_clip": 1.06071091,
      "balance_loss_mlp": 1.04151547,
      "epoch": 0.07503381932962573,
      "flos": 15121086261120.0,
      "grad_norm": 2.281722735345676,
      "language_loss": 0.64747941,
      "learning_rate": 3.978764471530921e-06,
      "loss": 0.66999531,
      "num_input_tokens_seen": 26533595,
      "step": 1248,
      "time_per_iteration": 2.520212411880493
    },
    {
      "auxiliary_loss_clip": 0.01189207,
      "auxiliary_loss_mlp": 0.00792377,
      "balance_loss_clip": 1.06806946,
      "balance_loss_mlp": 1.01520908,
      "epoch": 0.0750939425822937,
      "flos": 12816734071680.0,
      "grad_norm": 2.1356445220973868,
      "language_loss": 0.74403358,
      "learning_rate": 3.978707830891102e-06,
      "loss": 0.76384938,
      "num_input_tokens_seen": 26549405,
      "step": 1249,
      "time_per_iteration": 2.4669113159179688
    },
    {
      "auxiliary_loss_clip": 0.01171327,
      "auxiliary_loss_mlp": 0.010716,
      "balance_loss_clip": 1.06289768,
      "balance_loss_mlp": 1.04577899,
      "epoch": 0.07515406583496168,
      "flos": 24206844695040.0,
      "grad_norm": 4.311755211924707,
      "language_loss": 0.81867629,
      "learning_rate": 3.978651115218482e-06,
      "loss": 0.84110558,
      "num_input_tokens_seen": 26567200,
      "step": 1250,
      "time_per_iteration": 2.566483736038208
    },
    {
      "auxiliary_loss_clip": 0.01150385,
      "auxiliary_loss_mlp": 0.0106958,
      "balance_loss_clip": 1.05922806,
      "balance_loss_mlp": 1.04499888,
      "epoch": 0.07521418908762964,
      "flos": 26688164215680.0,
      "grad_norm": 2.7253448028688187,
      "language_loss": 0.6666007,
      "learning_rate": 3.978594324515215e-06,
      "loss": 0.68880033,
      "num_input_tokens_seen": 26586190,
      "step": 1251,
      "time_per_iteration": 2.633263111114502
    },
    {
      "auxiliary_loss_clip": 0.01058067,
      "auxiliary_loss_mlp": 0.01012627,
      "balance_loss_clip": 1.02996564,
      "balance_loss_mlp": 1.00871706,
      "epoch": 0.0752743123402976,
      "flos": 59095140589440.0,
      "grad_norm": 0.8924714101643109,
      "language_loss": 0.7034409,
      "learning_rate": 3.9785374587834515e-06,
      "loss": 0.72414792,
      "num_input_tokens_seen": 26650710,
      "step": 1252,
      "time_per_iteration": 3.2008719444274902
    },
    {
      "auxiliary_loss_clip": 0.01204259,
      "auxiliary_loss_mlp": 0.01064079,
      "balance_loss_clip": 1.06131053,
      "balance_loss_mlp": 1.03995109,
      "epoch": 0.07533443559296558,
      "flos": 23477032160640.0,
      "grad_norm": 2.2476254377378786,
      "language_loss": 0.79695463,
      "learning_rate": 3.97848051802535e-06,
      "loss": 0.81963801,
      "num_input_tokens_seen": 26669000,
      "step": 1253,
      "time_per_iteration": 2.4874281883239746
    },
    {
      "auxiliary_loss_clip": 0.01169721,
      "auxiliary_loss_mlp": 0.01061326,
      "balance_loss_clip": 1.06683898,
      "balance_loss_mlp": 1.03697181,
      "epoch": 0.07539455884563355,
      "flos": 20879110114560.0,
      "grad_norm": 5.690615884276562,
      "language_loss": 0.93358594,
      "learning_rate": 3.978423502243069e-06,
      "loss": 0.95589644,
      "num_input_tokens_seen": 26683075,
      "step": 1254,
      "time_per_iteration": 2.565964698791504
    },
    {
      "auxiliary_loss_clip": 0.01173593,
      "auxiliary_loss_mlp": 0.01058288,
      "balance_loss_clip": 1.06195998,
      "balance_loss_mlp": 1.03450584,
      "epoch": 0.07545468209830151,
      "flos": 27672906551040.0,
      "grad_norm": 2.1966610481206414,
      "language_loss": 0.88229012,
      "learning_rate": 3.97836641143877e-06,
      "loss": 0.90460891,
      "num_input_tokens_seen": 26701875,
      "step": 1255,
      "time_per_iteration": 2.6177263259887695
    },
    {
      "auxiliary_loss_clip": 0.01202587,
      "auxiliary_loss_mlp": 0.01063878,
      "balance_loss_clip": 1.06171119,
      "balance_loss_mlp": 1.03871322,
      "epoch": 0.0755148053509695,
      "flos": 14136990370560.0,
      "grad_norm": 1.8715698154690987,
      "language_loss": 0.79628718,
      "learning_rate": 3.978309245614618e-06,
      "loss": 0.81895185,
      "num_input_tokens_seen": 26719050,
      "step": 1256,
      "time_per_iteration": 2.4832043647766113
    },
    {
      "auxiliary_loss_clip": 0.01065496,
      "auxiliary_loss_mlp": 0.01005113,
      "balance_loss_clip": 1.03181362,
      "balance_loss_mlp": 1.00134647,
      "epoch": 0.07557492860363746,
      "flos": 58235257929600.0,
      "grad_norm": 0.7721144620458134,
      "language_loss": 0.57998478,
      "learning_rate": 3.9782520047727825e-06,
      "loss": 0.6006909,
      "num_input_tokens_seen": 26780650,
      "step": 1257,
      "time_per_iteration": 3.2268896102905273
    },
    {
      "auxiliary_loss_clip": 0.01153431,
      "auxiliary_loss_mlp": 0.01062086,
      "balance_loss_clip": 1.06216741,
      "balance_loss_mlp": 1.03781509,
      "epoch": 0.07563505185630542,
      "flos": 24644380262400.0,
      "grad_norm": 2.1376427709870556,
      "language_loss": 0.89918262,
      "learning_rate": 3.978194688915432e-06,
      "loss": 0.92133778,
      "num_input_tokens_seen": 26798725,
      "step": 1258,
      "time_per_iteration": 2.636545181274414
    },
    {
      "auxiliary_loss_clip": 0.01169214,
      "auxiliary_loss_mlp": 0.01057191,
      "balance_loss_clip": 1.06371605,
      "balance_loss_mlp": 1.03251505,
      "epoch": 0.07569517510897339,
      "flos": 15522998515200.0,
      "grad_norm": 1.8561083712582425,
      "language_loss": 0.81233096,
      "learning_rate": 3.978137298044741e-06,
      "loss": 0.83459508,
      "num_input_tokens_seen": 26817005,
      "step": 1259,
      "time_per_iteration": 2.5479156970977783
    },
    {
      "auxiliary_loss_clip": 0.01194896,
      "auxiliary_loss_mlp": 0.01056252,
      "balance_loss_clip": 1.06431675,
      "balance_loss_mlp": 1.03300643,
      "epoch": 0.07575529836164137,
      "flos": 22928532503040.0,
      "grad_norm": 1.9950165252349916,
      "language_loss": 0.75836444,
      "learning_rate": 3.978079832162885e-06,
      "loss": 0.78087592,
      "num_input_tokens_seen": 26836655,
      "step": 1260,
      "time_per_iteration": 2.5715529918670654
    },
    {
      "auxiliary_loss_clip": 0.011937,
      "auxiliary_loss_mlp": 0.01065433,
      "balance_loss_clip": 1.11412454,
      "balance_loss_mlp": 1.04054201,
      "epoch": 0.07581542161430933,
      "flos": 19500428344320.0,
      "grad_norm": 2.168828247531321,
      "language_loss": 0.84623969,
      "learning_rate": 3.978022291272044e-06,
      "loss": 0.86883104,
      "num_input_tokens_seen": 26854925,
      "step": 1261,
      "time_per_iteration": 2.5466198921203613
    },
    {
      "auxiliary_loss_clip": 0.01209735,
      "auxiliary_loss_mlp": 0.01061643,
      "balance_loss_clip": 1.06570005,
      "balance_loss_mlp": 1.03845716,
      "epoch": 0.0758755448669773,
      "flos": 24973465691520.0,
      "grad_norm": 2.592489498431488,
      "language_loss": 0.82536894,
      "learning_rate": 3.977964675374399e-06,
      "loss": 0.84808272,
      "num_input_tokens_seen": 26876170,
      "step": 1262,
      "time_per_iteration": 2.5429248809814453
    },
    {
      "auxiliary_loss_clip": 0.0120218,
      "auxiliary_loss_mlp": 0.01059533,
      "balance_loss_clip": 1.05982685,
      "balance_loss_mlp": 1.03477371,
      "epoch": 0.07593566811964528,
      "flos": 22747973811840.0,
      "grad_norm": 2.9610122487166897,
      "language_loss": 0.82413363,
      "learning_rate": 3.977906984472136e-06,
      "loss": 0.84675074,
      "num_input_tokens_seen": 26895005,
      "step": 1263,
      "time_per_iteration": 2.4843833446502686
    },
    {
      "auxiliary_loss_clip": 0.01159708,
      "auxiliary_loss_mlp": 0.01059023,
      "balance_loss_clip": 1.06179988,
      "balance_loss_mlp": 1.03550327,
      "epoch": 0.07599579137231324,
      "flos": 23112395245440.0,
      "grad_norm": 2.979874282031962,
      "language_loss": 0.76107848,
      "learning_rate": 3.977849218567442e-06,
      "loss": 0.78326577,
      "num_input_tokens_seen": 26913930,
      "step": 1264,
      "time_per_iteration": 2.649636745452881
    },
    {
      "auxiliary_loss_clip": 0.01210203,
      "auxiliary_loss_mlp": 0.01058491,
      "balance_loss_clip": 1.11172533,
      "balance_loss_mlp": 1.03489995,
      "epoch": 0.07605591462498121,
      "flos": 14502058248960.0,
      "grad_norm": 3.877292587258297,
      "language_loss": 0.80940992,
      "learning_rate": 3.977791377662507e-06,
      "loss": 0.83209687,
      "num_input_tokens_seen": 26931485,
      "step": 1265,
      "time_per_iteration": 2.5221712589263916
    },
    {
      "auxiliary_loss_clip": 0.01146886,
      "auxiliary_loss_mlp": 0.01062752,
      "balance_loss_clip": 1.06094623,
      "balance_loss_mlp": 1.03643048,
      "epoch": 0.07611603787764919,
      "flos": 23514199758720.0,
      "grad_norm": 2.3510964044771936,
      "language_loss": 0.65532875,
      "learning_rate": 3.977733461759524e-06,
      "loss": 0.67742515,
      "num_input_tokens_seen": 26951670,
      "step": 1266,
      "time_per_iteration": 2.6282331943511963
    },
    {
      "auxiliary_loss_clip": 0.0116,
      "auxiliary_loss_mlp": 0.01061693,
      "balance_loss_clip": 1.05770111,
      "balance_loss_mlp": 1.03739786,
      "epoch": 0.07617616113031715,
      "flos": 21507188353920.0,
      "grad_norm": 2.241382006311566,
      "language_loss": 0.79583824,
      "learning_rate": 3.977675470860691e-06,
      "loss": 0.81805515,
      "num_input_tokens_seen": 26970335,
      "step": 1267,
      "time_per_iteration": 3.945256233215332
    },
    {
      "auxiliary_loss_clip": 0.0117997,
      "auxiliary_loss_mlp": 0.01052217,
      "balance_loss_clip": 1.06036615,
      "balance_loss_mlp": 1.02961493,
      "epoch": 0.07623628438298512,
      "flos": 14573161221120.0,
      "grad_norm": 4.251490918519658,
      "language_loss": 0.73491597,
      "learning_rate": 3.977617404968205e-06,
      "loss": 0.75723779,
      "num_input_tokens_seen": 26986025,
      "step": 1268,
      "time_per_iteration": 2.5150680541992188
    },
    {
      "auxiliary_loss_clip": 0.01189119,
      "auxiliary_loss_mlp": 0.01052798,
      "balance_loss_clip": 1.06004155,
      "balance_loss_mlp": 1.02901566,
      "epoch": 0.07629640763565308,
      "flos": 14720395069440.0,
      "grad_norm": 2.9854896167729645,
      "language_loss": 0.82232034,
      "learning_rate": 3.977559264084269e-06,
      "loss": 0.8447395,
      "num_input_tokens_seen": 27004045,
      "step": 1269,
      "time_per_iteration": 3.8626296520233154
    },
    {
      "auxiliary_loss_clip": 0.01193153,
      "auxiliary_loss_mlp": 0.01060527,
      "balance_loss_clip": 1.06224036,
      "balance_loss_mlp": 1.03611302,
      "epoch": 0.07635653088832106,
      "flos": 14902929008640.0,
      "grad_norm": 2.3788313749918792,
      "language_loss": 0.88725567,
      "learning_rate": 3.977501048211088e-06,
      "loss": 0.90979242,
      "num_input_tokens_seen": 27022070,
      "step": 1270,
      "time_per_iteration": 2.4969308376312256
    },
    {
      "auxiliary_loss_clip": 0.01190557,
      "auxiliary_loss_mlp": 0.01058427,
      "balance_loss_clip": 1.06118488,
      "balance_loss_mlp": 1.03422761,
      "epoch": 0.07641665414098903,
      "flos": 26651571235200.0,
      "grad_norm": 4.034692484542531,
      "language_loss": 0.71129233,
      "learning_rate": 3.977442757350869e-06,
      "loss": 0.73378217,
      "num_input_tokens_seen": 27041755,
      "step": 1271,
      "time_per_iteration": 2.5552656650543213
    },
    {
      "auxiliary_loss_clip": 0.0115789,
      "auxiliary_loss_mlp": 0.01067158,
      "balance_loss_clip": 1.0614779,
      "balance_loss_mlp": 1.04348326,
      "epoch": 0.07647677739365699,
      "flos": 25192808092800.0,
      "grad_norm": 1.6975590448899378,
      "language_loss": 0.82622957,
      "learning_rate": 3.977384391505823e-06,
      "loss": 0.84847999,
      "num_input_tokens_seen": 27061540,
      "step": 1272,
      "time_per_iteration": 2.6253504753112793
    },
    {
      "auxiliary_loss_clip": 0.01173707,
      "auxiliary_loss_mlp": 0.00829344,
      "balance_loss_clip": 1.05758429,
      "balance_loss_mlp": 1.08582973,
      "epoch": 0.07653690064632497,
      "flos": 20558141159040.0,
      "grad_norm": 1.667615180357115,
      "language_loss": 0.80246925,
      "learning_rate": 3.977325950678162e-06,
      "loss": 0.82249975,
      "num_input_tokens_seen": 27081395,
      "step": 1273,
      "time_per_iteration": 3.9171383380889893
    },
    {
      "auxiliary_loss_clip": 0.01179059,
      "auxiliary_loss_mlp": 0.0106415,
      "balance_loss_clip": 1.06146085,
      "balance_loss_mlp": 1.03921187,
      "epoch": 0.07659702389899294,
      "flos": 22269320150400.0,
      "grad_norm": 1.9677418411818408,
      "language_loss": 0.81379402,
      "learning_rate": 3.977267434870103e-06,
      "loss": 0.83622611,
      "num_input_tokens_seen": 27101175,
      "step": 1274,
      "time_per_iteration": 2.5613925457000732
    },
    {
      "auxiliary_loss_clip": 0.01178902,
      "auxiliary_loss_mlp": 0.01068299,
      "balance_loss_clip": 1.05823231,
      "balance_loss_mlp": 1.04262137,
      "epoch": 0.0766571471516609,
      "flos": 32636120209920.0,
      "grad_norm": 2.0095454542086224,
      "language_loss": 0.73129481,
      "learning_rate": 3.977208844083865e-06,
      "loss": 0.75376683,
      "num_input_tokens_seen": 27124505,
      "step": 1275,
      "time_per_iteration": 4.009122848510742
    },
    {
      "auxiliary_loss_clip": 0.01204654,
      "auxiliary_loss_mlp": 0.01063474,
      "balance_loss_clip": 1.06163502,
      "balance_loss_mlp": 1.03755808,
      "epoch": 0.07671727040432888,
      "flos": 15267386355840.0,
      "grad_norm": 2.3013042357854943,
      "language_loss": 0.79700208,
      "learning_rate": 3.9771501783216685e-06,
      "loss": 0.81968337,
      "num_input_tokens_seen": 27140960,
      "step": 1276,
      "time_per_iteration": 2.446730613708496
    },
    {
      "auxiliary_loss_clip": 0.01217226,
      "auxiliary_loss_mlp": 0.01054173,
      "balance_loss_clip": 1.10234904,
      "balance_loss_mlp": 1.030164,
      "epoch": 0.07677739365699685,
      "flos": 28184094956160.0,
      "grad_norm": 2.4586467129555527,
      "language_loss": 0.59358799,
      "learning_rate": 3.97709143758574e-06,
      "loss": 0.61630201,
      "num_input_tokens_seen": 27160985,
      "step": 1277,
      "time_per_iteration": 2.580177068710327
    },
    {
      "auxiliary_loss_clip": 0.01196268,
      "auxiliary_loss_mlp": 0.01055851,
      "balance_loss_clip": 1.06126285,
      "balance_loss_mlp": 1.03200912,
      "epoch": 0.07683751690966481,
      "flos": 18296128126080.0,
      "grad_norm": 2.6395948800344025,
      "language_loss": 0.74656469,
      "learning_rate": 3.977032621878305e-06,
      "loss": 0.76908588,
      "num_input_tokens_seen": 27178390,
      "step": 1278,
      "time_per_iteration": 2.4842662811279297
    },
    {
      "auxiliary_loss_clip": 0.01157661,
      "auxiliary_loss_mlp": 0.01060497,
      "balance_loss_clip": 1.0576911,
      "balance_loss_mlp": 1.03645205,
      "epoch": 0.07689764016233278,
      "flos": 21981101420160.0,
      "grad_norm": 3.045077566881005,
      "language_loss": 0.88511616,
      "learning_rate": 3.976973731201596e-06,
      "loss": 0.90729773,
      "num_input_tokens_seen": 27197505,
      "step": 1279,
      "time_per_iteration": 2.550074815750122
    },
    {
      "auxiliary_loss_clip": 0.01164891,
      "auxiliary_loss_mlp": 0.01058741,
      "balance_loss_clip": 1.05634952,
      "balance_loss_mlp": 1.03422022,
      "epoch": 0.07695776341500075,
      "flos": 22235995307520.0,
      "grad_norm": 2.5726771041571905,
      "language_loss": 0.82817745,
      "learning_rate": 3.976914765557845e-06,
      "loss": 0.8504138,
      "num_input_tokens_seen": 27214260,
      "step": 1280,
      "time_per_iteration": 2.529118776321411
    },
    {
      "auxiliary_loss_clip": 0.01187206,
      "auxiliary_loss_mlp": 0.01061051,
      "balance_loss_clip": 1.05984366,
      "balance_loss_mlp": 1.03695917,
      "epoch": 0.07701788666766872,
      "flos": 16143750380160.0,
      "grad_norm": 2.0807081573788127,
      "language_loss": 0.76081622,
      "learning_rate": 3.9768557249492875e-06,
      "loss": 0.78329879,
      "num_input_tokens_seen": 27232525,
      "step": 1281,
      "time_per_iteration": 2.4998531341552734
    },
    {
      "auxiliary_loss_clip": 0.01165001,
      "auxiliary_loss_mlp": 0.01058488,
      "balance_loss_clip": 1.05557799,
      "balance_loss_mlp": 1.03405023,
      "epoch": 0.07707800992033668,
      "flos": 19463045264640.0,
      "grad_norm": 1.941160384270207,
      "language_loss": 0.75216877,
      "learning_rate": 3.9767966093781634e-06,
      "loss": 0.77440369,
      "num_input_tokens_seen": 27249800,
      "step": 1282,
      "time_per_iteration": 2.5537612438201904
    },
    {
      "auxiliary_loss_clip": 0.01199182,
      "auxiliary_loss_mlp": 0.01068291,
      "balance_loss_clip": 1.0577662,
      "balance_loss_mlp": 1.04347134,
      "epoch": 0.07713813317300466,
      "flos": 18990281433600.0,
      "grad_norm": 2.2407635074516348,
      "language_loss": 0.84083015,
      "learning_rate": 3.976737418846713e-06,
      "loss": 0.86350489,
      "num_input_tokens_seen": 27268895,
      "step": 1283,
      "time_per_iteration": 2.4931952953338623
    },
    {
      "auxiliary_loss_clip": 0.01187636,
      "auxiliary_loss_mlp": 0.01065511,
      "balance_loss_clip": 1.05854023,
      "balance_loss_mlp": 1.03903425,
      "epoch": 0.07719825642567263,
      "flos": 18113953322880.0,
      "grad_norm": 2.476431088103227,
      "language_loss": 0.75026894,
      "learning_rate": 3.976678153357181e-06,
      "loss": 0.77280033,
      "num_input_tokens_seen": 27288180,
      "step": 1284,
      "time_per_iteration": 2.4868905544281006
    },
    {
      "auxiliary_loss_clip": 0.01171935,
      "auxiliary_loss_mlp": 0.01072715,
      "balance_loss_clip": 1.0558964,
      "balance_loss_mlp": 1.04899263,
      "epoch": 0.0772583796783406,
      "flos": 42194426993280.0,
      "grad_norm": 1.7664813412507139,
      "language_loss": 0.76049483,
      "learning_rate": 3.976618812911817e-06,
      "loss": 0.78294134,
      "num_input_tokens_seen": 27311815,
      "step": 1285,
      "time_per_iteration": 2.707158088684082
    },
    {
      "auxiliary_loss_clip": 0.0120348,
      "auxiliary_loss_mlp": 0.01065498,
      "balance_loss_clip": 1.06172025,
      "balance_loss_mlp": 1.04243112,
      "epoch": 0.07731850293100857,
      "flos": 24753692327040.0,
      "grad_norm": 2.0635069572162545,
      "language_loss": 0.83756602,
      "learning_rate": 3.9765593975128685e-06,
      "loss": 0.86025578,
      "num_input_tokens_seen": 27331890,
      "step": 1286,
      "time_per_iteration": 2.501821517944336
    },
    {
      "auxiliary_loss_clip": 0.01169881,
      "auxiliary_loss_mlp": 0.0106127,
      "balance_loss_clip": 1.05504346,
      "balance_loss_mlp": 1.03682005,
      "epoch": 0.07737862618367654,
      "flos": 17565884628480.0,
      "grad_norm": 2.6480422458879445,
      "language_loss": 0.76859426,
      "learning_rate": 3.97649990716259e-06,
      "loss": 0.79090571,
      "num_input_tokens_seen": 27348320,
      "step": 1287,
      "time_per_iteration": 2.4943323135375977
    },
    {
      "auxiliary_loss_clip": 0.01170238,
      "auxiliary_loss_mlp": 0.0105872,
      "balance_loss_clip": 1.05258465,
      "balance_loss_mlp": 1.03543818,
      "epoch": 0.0774387494363445,
      "flos": 25627147349760.0,
      "grad_norm": 1.6881683759824808,
      "language_loss": 0.84483165,
      "learning_rate": 3.976440341863237e-06,
      "loss": 0.86712122,
      "num_input_tokens_seen": 27367670,
      "step": 1288,
      "time_per_iteration": 2.563023805618286
    },
    {
      "auxiliary_loss_clip": 0.01197817,
      "auxiliary_loss_mlp": 0.01059949,
      "balance_loss_clip": 1.05485392,
      "balance_loss_mlp": 1.03707314,
      "epoch": 0.07749887268901248,
      "flos": 12239865648000.0,
      "grad_norm": 2.164736729580838,
      "language_loss": 0.85310096,
      "learning_rate": 3.976380701617068e-06,
      "loss": 0.8756786,
      "num_input_tokens_seen": 27385485,
      "step": 1289,
      "time_per_iteration": 2.4442203044891357
    },
    {
      "auxiliary_loss_clip": 0.0119558,
      "auxiliary_loss_mlp": 0.01049891,
      "balance_loss_clip": 1.0550034,
      "balance_loss_mlp": 1.02671695,
      "epoch": 0.07755899594168045,
      "flos": 25081736261760.0,
      "grad_norm": 1.9355346363803563,
      "language_loss": 0.85395646,
      "learning_rate": 3.976320986426344e-06,
      "loss": 0.8764112,
      "num_input_tokens_seen": 27405110,
      "step": 1290,
      "time_per_iteration": 2.491056442260742
    },
    {
      "auxiliary_loss_clip": 0.01165431,
      "auxiliary_loss_mlp": 0.01063664,
      "balance_loss_clip": 1.0583775,
      "balance_loss_mlp": 1.03876114,
      "epoch": 0.07761911919434841,
      "flos": 14246410176000.0,
      "grad_norm": 2.048544371838898,
      "language_loss": 0.90912133,
      "learning_rate": 3.9762611962933315e-06,
      "loss": 0.93141234,
      "num_input_tokens_seen": 27422855,
      "step": 1291,
      "time_per_iteration": 2.5296120643615723
    },
    {
      "auxiliary_loss_clip": 0.01063723,
      "auxiliary_loss_mlp": 0.01011043,
      "balance_loss_clip": 1.02685416,
      "balance_loss_mlp": 1.00627446,
      "epoch": 0.07767924244701638,
      "flos": 67237202954880.0,
      "grad_norm": 0.8874479904709917,
      "language_loss": 0.65095544,
      "learning_rate": 3.9762013312202955e-06,
      "loss": 0.6717031,
      "num_input_tokens_seen": 27487190,
      "step": 1292,
      "time_per_iteration": 3.230682134628296
    },
    {
      "auxiliary_loss_clip": 0.01185415,
      "auxiliary_loss_mlp": 0.01058336,
      "balance_loss_clip": 1.05759096,
      "balance_loss_mlp": 1.03503036,
      "epoch": 0.07773936569968436,
      "flos": 28550635292160.0,
      "grad_norm": 1.8783728786898657,
      "language_loss": 0.87705308,
      "learning_rate": 3.9761413912095075e-06,
      "loss": 0.89949059,
      "num_input_tokens_seen": 27510465,
      "step": 1293,
      "time_per_iteration": 2.599426507949829
    },
    {
      "auxiliary_loss_clip": 0.01116893,
      "auxiliary_loss_mlp": 0.01079327,
      "balance_loss_clip": 1.05588436,
      "balance_loss_mlp": 1.05012131,
      "epoch": 0.07779948895235232,
      "flos": 27490264871040.0,
      "grad_norm": 2.103355026491555,
      "language_loss": 0.85120755,
      "learning_rate": 3.976081376263239e-06,
      "loss": 0.87316972,
      "num_input_tokens_seen": 27528645,
      "step": 1294,
      "time_per_iteration": 2.6664233207702637
    },
    {
      "auxiliary_loss_clip": 0.01152312,
      "auxiliary_loss_mlp": 0.01058779,
      "balance_loss_clip": 1.05731416,
      "balance_loss_mlp": 1.03369713,
      "epoch": 0.07785961220502029,
      "flos": 18223301301120.0,
      "grad_norm": 2.5106051730394525,
      "language_loss": 0.79053247,
      "learning_rate": 3.976021286383768e-06,
      "loss": 0.81264341,
      "num_input_tokens_seen": 27546165,
      "step": 1295,
      "time_per_iteration": 2.5529909133911133
    },
    {
      "auxiliary_loss_clip": 0.01145645,
      "auxiliary_loss_mlp": 0.01059502,
      "balance_loss_clip": 1.052472,
      "balance_loss_mlp": 1.03509998,
      "epoch": 0.07791973545768827,
      "flos": 24608218245120.0,
      "grad_norm": 2.4770748893685397,
      "language_loss": 0.87928391,
      "learning_rate": 3.975961121573371e-06,
      "loss": 0.90133536,
      "num_input_tokens_seen": 27566520,
      "step": 1296,
      "time_per_iteration": 2.6050217151641846
    },
    {
      "auxiliary_loss_clip": 0.01199516,
      "auxiliary_loss_mlp": 0.01063421,
      "balance_loss_clip": 1.0574789,
      "balance_loss_mlp": 1.03891194,
      "epoch": 0.07797985871035623,
      "flos": 14282069402880.0,
      "grad_norm": 2.9589175995729646,
      "language_loss": 0.96104366,
      "learning_rate": 3.9759008818343305e-06,
      "loss": 0.9836731,
      "num_input_tokens_seen": 27581960,
      "step": 1297,
      "time_per_iteration": 2.442082405090332
    },
    {
      "auxiliary_loss_clip": 0.01169086,
      "auxiliary_loss_mlp": 0.01059244,
      "balance_loss_clip": 1.05804729,
      "balance_loss_mlp": 1.0358429,
      "epoch": 0.0780399819630242,
      "flos": 26610453141120.0,
      "grad_norm": 2.2838807321652306,
      "language_loss": 0.75959408,
      "learning_rate": 3.97584056716893e-06,
      "loss": 0.78187734,
      "num_input_tokens_seen": 27601415,
      "step": 1298,
      "time_per_iteration": 2.6006946563720703
    },
    {
      "auxiliary_loss_clip": 0.01139115,
      "auxiliary_loss_mlp": 0.00819846,
      "balance_loss_clip": 1.05488229,
      "balance_loss_mlp": 1.0644865,
      "epoch": 0.07810010521569218,
      "flos": 21834514016640.0,
      "grad_norm": 4.0639267411131135,
      "language_loss": 0.80770528,
      "learning_rate": 3.9757801775794575e-06,
      "loss": 0.82729489,
      "num_input_tokens_seen": 27621490,
      "step": 1299,
      "time_per_iteration": 2.6070175170898438
    },
    {
      "auxiliary_loss_clip": 0.01153521,
      "auxiliary_loss_mlp": 0.0106363,
      "balance_loss_clip": 1.05366075,
      "balance_loss_mlp": 1.03909695,
      "epoch": 0.07816022846836014,
      "flos": 25081233471360.0,
      "grad_norm": 2.08973995769173,
      "language_loss": 0.86592889,
      "learning_rate": 3.975719713068202e-06,
      "loss": 0.88810039,
      "num_input_tokens_seen": 27640600,
      "step": 1300,
      "time_per_iteration": 2.607984781265259
    },
    {
      "auxiliary_loss_clip": 0.01195339,
      "auxiliary_loss_mlp": 0.01055136,
      "balance_loss_clip": 1.05512214,
      "balance_loss_mlp": 1.03043616,
      "epoch": 0.0782203517210281,
      "flos": 40917515431680.0,
      "grad_norm": 2.133961906184151,
      "language_loss": 0.72056824,
      "learning_rate": 3.975659173637458e-06,
      "loss": 0.74307299,
      "num_input_tokens_seen": 27663070,
      "step": 1301,
      "time_per_iteration": 2.6316583156585693
    },
    {
      "auxiliary_loss_clip": 0.01188846,
      "auxiliary_loss_mlp": 0.01068177,
      "balance_loss_clip": 1.05700922,
      "balance_loss_mlp": 1.04425216,
      "epoch": 0.07828047497369607,
      "flos": 41172014269440.0,
      "grad_norm": 1.8635475486364876,
      "language_loss": 0.70935726,
      "learning_rate": 3.97559855928952e-06,
      "loss": 0.73192751,
      "num_input_tokens_seen": 27686425,
      "step": 1302,
      "time_per_iteration": 2.694078207015991
    },
    {
      "auxiliary_loss_clip": 0.01157026,
      "auxiliary_loss_mlp": 0.00807689,
      "balance_loss_clip": 1.05854809,
      "balance_loss_mlp": 1.04317915,
      "epoch": 0.07834059822636405,
      "flos": 23508130360320.0,
      "grad_norm": 2.1171932592363976,
      "language_loss": 0.82126296,
      "learning_rate": 3.9755378700266864e-06,
      "loss": 0.8409102,
      "num_input_tokens_seen": 27704900,
      "step": 1303,
      "time_per_iteration": 2.5692477226257324
    },
    {
      "auxiliary_loss_clip": 0.0117742,
      "auxiliary_loss_mlp": 0.01064393,
      "balance_loss_clip": 1.05251122,
      "balance_loss_mlp": 1.0399313,
      "epoch": 0.07840072147903202,
      "flos": 20193899293440.0,
      "grad_norm": 1.6571278494302157,
      "language_loss": 0.75208318,
      "learning_rate": 3.9754771058512585e-06,
      "loss": 0.7745012,
      "num_input_tokens_seen": 27724890,
      "step": 1304,
      "time_per_iteration": 2.5045342445373535
    },
    {
      "auxiliary_loss_clip": 0.0119823,
      "auxiliary_loss_mlp": 0.0106105,
      "balance_loss_clip": 1.0592494,
      "balance_loss_mlp": 1.03702903,
      "epoch": 0.07846084473169998,
      "flos": 21360816432000.0,
      "grad_norm": 1.975064385580126,
      "language_loss": 0.76174819,
      "learning_rate": 3.975416266765542e-06,
      "loss": 0.78434104,
      "num_input_tokens_seen": 27743115,
      "step": 1305,
      "time_per_iteration": 2.4738821983337402
    },
    {
      "auxiliary_loss_clip": 0.01133405,
      "auxiliary_loss_mlp": 0.01066026,
      "balance_loss_clip": 1.05631757,
      "balance_loss_mlp": 1.04167151,
      "epoch": 0.07852096798436796,
      "flos": 25410965345280.0,
      "grad_norm": 1.7003105257272415,
      "language_loss": 0.84975249,
      "learning_rate": 3.975355352771841e-06,
      "loss": 0.87174678,
      "num_input_tokens_seen": 27763570,
      "step": 1306,
      "time_per_iteration": 4.060633182525635
    },
    {
      "auxiliary_loss_clip": 0.01186259,
      "auxiliary_loss_mlp": 0.01045235,
      "balance_loss_clip": 1.05862713,
      "balance_loss_mlp": 1.02304983,
      "epoch": 0.07858109123703592,
      "flos": 24571481610240.0,
      "grad_norm": 2.4065625511026596,
      "language_loss": 0.90150112,
      "learning_rate": 3.975294363872468e-06,
      "loss": 0.92381608,
      "num_input_tokens_seen": 27780030,
      "step": 1307,
      "time_per_iteration": 3.8742942810058594
    },
    {
      "auxiliary_loss_clip": 0.0114008,
      "auxiliary_loss_mlp": 0.01058473,
      "balance_loss_clip": 1.0535022,
      "balance_loss_mlp": 1.03386796,
      "epoch": 0.07864121448970389,
      "flos": 20698874645760.0,
      "grad_norm": 1.8778080999027962,
      "language_loss": 0.83412665,
      "learning_rate": 3.975233300069735e-06,
      "loss": 0.85611212,
      "num_input_tokens_seen": 27796225,
      "step": 1308,
      "time_per_iteration": 2.5883257389068604
    },
    {
      "auxiliary_loss_clip": 0.01170854,
      "auxiliary_loss_mlp": 0.0105575,
      "balance_loss_clip": 1.09223533,
      "balance_loss_mlp": 1.03251672,
      "epoch": 0.07870133774237187,
      "flos": 22966526113920.0,
      "grad_norm": 1.479312861942881,
      "language_loss": 0.77480745,
      "learning_rate": 3.975172161365958e-06,
      "loss": 0.79707348,
      "num_input_tokens_seen": 27815975,
      "step": 1309,
      "time_per_iteration": 2.5950517654418945
    },
    {
      "auxiliary_loss_clip": 0.01191854,
      "auxiliary_loss_mlp": 0.01069343,
      "balance_loss_clip": 1.05746651,
      "balance_loss_mlp": 1.04341531,
      "epoch": 0.07876146099503983,
      "flos": 18842832103680.0,
      "grad_norm": 1.9267887761688451,
      "language_loss": 0.80338287,
      "learning_rate": 3.975110947763453e-06,
      "loss": 0.82599473,
      "num_input_tokens_seen": 27832255,
      "step": 1310,
      "time_per_iteration": 2.488211154937744
    },
    {
      "auxiliary_loss_clip": 0.0116691,
      "auxiliary_loss_mlp": 0.0080258,
      "balance_loss_clip": 1.05800986,
      "balance_loss_mlp": 1.03434253,
      "epoch": 0.0788215842477078,
      "flos": 23805794367360.0,
      "grad_norm": 2.1714460809035345,
      "language_loss": 0.73188102,
      "learning_rate": 3.9750496592645435e-06,
      "loss": 0.75157589,
      "num_input_tokens_seen": 27852180,
      "step": 1311,
      "time_per_iteration": 3.9456276893615723
    },
    {
      "auxiliary_loss_clip": 0.01182738,
      "auxiliary_loss_mlp": 0.01076606,
      "balance_loss_clip": 1.05830264,
      "balance_loss_mlp": 1.05182242,
      "epoch": 0.07888170750037576,
      "flos": 21579907438080.0,
      "grad_norm": 2.0226517736299297,
      "language_loss": 0.85988206,
      "learning_rate": 3.974988295871553e-06,
      "loss": 0.8824755,
      "num_input_tokens_seen": 27871435,
      "step": 1312,
      "time_per_iteration": 2.546945333480835
    },
    {
      "auxiliary_loss_clip": 0.01169489,
      "auxiliary_loss_mlp": 0.0106227,
      "balance_loss_clip": 1.05543721,
      "balance_loss_mlp": 1.03998983,
      "epoch": 0.07894183075304374,
      "flos": 19864849777920.0,
      "grad_norm": 1.9637712943324916,
      "language_loss": 0.82224369,
      "learning_rate": 3.9749268575868085e-06,
      "loss": 0.84456128,
      "num_input_tokens_seen": 27890625,
      "step": 1313,
      "time_per_iteration": 2.5260608196258545
    },
    {
      "auxiliary_loss_clip": 0.01177841,
      "auxiliary_loss_mlp": 0.00802716,
      "balance_loss_clip": 1.05751276,
      "balance_loss_mlp": 1.03046298,
      "epoch": 0.07900195400571171,
      "flos": 16143463071360.0,
      "grad_norm": 2.9558166374315182,
      "language_loss": 0.73633945,
      "learning_rate": 3.97486534441264e-06,
      "loss": 0.75614506,
      "num_input_tokens_seen": 27906530,
      "step": 1314,
      "time_per_iteration": 3.8893513679504395
    },
    {
      "auxiliary_loss_clip": 0.01173597,
      "auxiliary_loss_mlp": 0.00800991,
      "balance_loss_clip": 1.09289122,
      "balance_loss_mlp": 1.02948976,
      "epoch": 0.07906207725837967,
      "flos": 23730417676800.0,
      "grad_norm": 1.632907592563888,
      "language_loss": 0.79647273,
      "learning_rate": 3.974803756351379e-06,
      "loss": 0.81621861,
      "num_input_tokens_seen": 27926725,
      "step": 1315,
      "time_per_iteration": 2.6121950149536133
    },
    {
      "auxiliary_loss_clip": 0.01182759,
      "auxiliary_loss_mlp": 0.01064599,
      "balance_loss_clip": 1.05315781,
      "balance_loss_mlp": 1.03952909,
      "epoch": 0.07912220051104765,
      "flos": 24315905364480.0,
      "grad_norm": 1.9863534812643666,
      "language_loss": 0.73712748,
      "learning_rate": 3.974742093405362e-06,
      "loss": 0.75960106,
      "num_input_tokens_seen": 27947875,
      "step": 1316,
      "time_per_iteration": 2.5542471408843994
    },
    {
      "auxiliary_loss_clip": 0.01158585,
      "auxiliary_loss_mlp": 0.01064344,
      "balance_loss_clip": 1.05928206,
      "balance_loss_mlp": 1.0398823,
      "epoch": 0.07918232376371562,
      "flos": 18880035615360.0,
      "grad_norm": 2.987674638518574,
      "language_loss": 0.65224314,
      "learning_rate": 3.974680355576927e-06,
      "loss": 0.67447245,
      "num_input_tokens_seen": 27965040,
      "step": 1317,
      "time_per_iteration": 2.5529956817626953
    },
    {
      "auxiliary_loss_clip": 0.01190753,
      "auxiliary_loss_mlp": 0.01064959,
      "balance_loss_clip": 1.09238696,
      "balance_loss_mlp": 1.04027045,
      "epoch": 0.07924244701638358,
      "flos": 27376284038400.0,
      "grad_norm": 2.10885541310027,
      "language_loss": 0.73085082,
      "learning_rate": 3.974618542868415e-06,
      "loss": 0.75340796,
      "num_input_tokens_seen": 27985330,
      "step": 1318,
      "time_per_iteration": 2.6214659214019775
    },
    {
      "auxiliary_loss_clip": 0.0113609,
      "auxiliary_loss_mlp": 0.01059206,
      "balance_loss_clip": 1.05669296,
      "balance_loss_mlp": 1.03666377,
      "epoch": 0.07930257026905156,
      "flos": 25120340403840.0,
      "grad_norm": 1.6435995105208747,
      "language_loss": 0.90417397,
      "learning_rate": 3.97455665528217e-06,
      "loss": 0.92612696,
      "num_input_tokens_seen": 28007615,
      "step": 1319,
      "time_per_iteration": 2.6516027450561523
    },
    {
      "auxiliary_loss_clip": 0.01172799,
      "auxiliary_loss_mlp": 0.01054952,
      "balance_loss_clip": 1.05552268,
      "balance_loss_mlp": 1.03052628,
      "epoch": 0.07936269352171953,
      "flos": 21834478103040.0,
      "grad_norm": 2.718758878503946,
      "language_loss": 0.80164897,
      "learning_rate": 3.974494692820539e-06,
      "loss": 0.82392645,
      "num_input_tokens_seen": 28027765,
      "step": 1320,
      "time_per_iteration": 2.5317859649658203
    },
    {
      "auxiliary_loss_clip": 0.01173778,
      "auxiliary_loss_mlp": 0.01058758,
      "balance_loss_clip": 1.05645776,
      "balance_loss_mlp": 1.03578663,
      "epoch": 0.07942281677438749,
      "flos": 16939889377920.0,
      "grad_norm": 2.6861214567157754,
      "language_loss": 0.69065177,
      "learning_rate": 3.974432655485872e-06,
      "loss": 0.71297717,
      "num_input_tokens_seen": 28044225,
      "step": 1321,
      "time_per_iteration": 2.5478408336639404
    },
    {
      "auxiliary_loss_clip": 0.01179854,
      "auxiliary_loss_mlp": 0.01056748,
      "balance_loss_clip": 1.05625021,
      "balance_loss_mlp": 1.0327034,
      "epoch": 0.07948294002705546,
      "flos": 18986941468800.0,
      "grad_norm": 2.1474158004917916,
      "language_loss": 0.83983111,
      "learning_rate": 3.9743705432805195e-06,
      "loss": 0.86219716,
      "num_input_tokens_seen": 28062915,
      "step": 1322,
      "time_per_iteration": 2.52219295501709
    },
    {
      "auxiliary_loss_clip": 0.01194876,
      "auxiliary_loss_mlp": 0.01054199,
      "balance_loss_clip": 1.05403364,
      "balance_loss_mlp": 1.03108406,
      "epoch": 0.07954306327972344,
      "flos": 21653452535040.0,
      "grad_norm": 2.1241999569277774,
      "language_loss": 0.90565312,
      "learning_rate": 3.974308356206838e-06,
      "loss": 0.92814386,
      "num_input_tokens_seen": 28082175,
      "step": 1323,
      "time_per_iteration": 2.4719998836517334
    },
    {
      "auxiliary_loss_clip": 0.01154877,
      "auxiliary_loss_mlp": 0.01056666,
      "balance_loss_clip": 1.05411124,
      "balance_loss_mlp": 1.03344417,
      "epoch": 0.0796031865323914,
      "flos": 23220270766080.0,
      "grad_norm": 2.536179061922817,
      "language_loss": 0.82231033,
      "learning_rate": 3.974246094267187e-06,
      "loss": 0.84442574,
      "num_input_tokens_seen": 28102645,
      "step": 1324,
      "time_per_iteration": 2.5817573070526123
    },
    {
      "auxiliary_loss_clip": 0.01172159,
      "auxiliary_loss_mlp": 0.01051899,
      "balance_loss_clip": 1.05541742,
      "balance_loss_mlp": 1.02710342,
      "epoch": 0.07966330978505937,
      "flos": 23294534135040.0,
      "grad_norm": 6.823954890189742,
      "language_loss": 0.78952956,
      "learning_rate": 3.974183757463925e-06,
      "loss": 0.81177008,
      "num_input_tokens_seen": 28122805,
      "step": 1325,
      "time_per_iteration": 2.546828031539917
    },
    {
      "auxiliary_loss_clip": 0.01118436,
      "auxiliary_loss_mlp": 0.00824185,
      "balance_loss_clip": 1.05249095,
      "balance_loss_mlp": 1.07054067,
      "epoch": 0.07972343303772735,
      "flos": 18363783392640.0,
      "grad_norm": 2.0881667375617563,
      "language_loss": 0.88152897,
      "learning_rate": 3.974121345799418e-06,
      "loss": 0.90095514,
      "num_input_tokens_seen": 28140530,
      "step": 1326,
      "time_per_iteration": 2.612086057662964
    },
    {
      "auxiliary_loss_clip": 0.01191167,
      "auxiliary_loss_mlp": 0.01050751,
      "balance_loss_clip": 1.05429244,
      "balance_loss_mlp": 1.0256927,
      "epoch": 0.07978355629039531,
      "flos": 21762513204480.0,
      "grad_norm": 2.0700527780954165,
      "language_loss": 0.8311559,
      "learning_rate": 3.974058859276032e-06,
      "loss": 0.85357511,
      "num_input_tokens_seen": 28159640,
      "step": 1327,
      "time_per_iteration": 2.4839165210723877
    },
    {
      "auxiliary_loss_clip": 0.01198057,
      "auxiliary_loss_mlp": 0.01052118,
      "balance_loss_clip": 1.05804348,
      "balance_loss_mlp": 1.02777576,
      "epoch": 0.07984367954306328,
      "flos": 18551309322240.0,
      "grad_norm": 2.649129925990115,
      "language_loss": 0.79176372,
      "learning_rate": 3.9739962978961354e-06,
      "loss": 0.81426549,
      "num_input_tokens_seen": 28177050,
      "step": 1328,
      "time_per_iteration": 2.4634907245635986
    },
    {
      "auxiliary_loss_clip": 0.01190816,
      "auxiliary_loss_mlp": 0.01056077,
      "balance_loss_clip": 1.05813503,
      "balance_loss_mlp": 1.03064942,
      "epoch": 0.07990380279573125,
      "flos": 16904050583040.0,
      "grad_norm": 3.35893583109153,
      "language_loss": 0.73646677,
      "learning_rate": 3.973933661662101e-06,
      "loss": 0.75893569,
      "num_input_tokens_seen": 28193245,
      "step": 1329,
      "time_per_iteration": 2.4774179458618164
    },
    {
      "auxiliary_loss_clip": 0.01163311,
      "auxiliary_loss_mlp": 0.01062977,
      "balance_loss_clip": 1.05622911,
      "balance_loss_mlp": 1.03932631,
      "epoch": 0.07996392604839922,
      "flos": 24098358643200.0,
      "grad_norm": 2.3118984942611016,
      "language_loss": 0.81353331,
      "learning_rate": 3.973870950576305e-06,
      "loss": 0.83579618,
      "num_input_tokens_seen": 28213570,
      "step": 1330,
      "time_per_iteration": 2.5662713050842285
    },
    {
      "auxiliary_loss_clip": 0.0119718,
      "auxiliary_loss_mlp": 0.0079722,
      "balance_loss_clip": 1.05741024,
      "balance_loss_mlp": 1.02525437,
      "epoch": 0.08002404930106718,
      "flos": 14278729438080.0,
      "grad_norm": 1.9860645284044605,
      "language_loss": 0.88573027,
      "learning_rate": 3.9738081646411255e-06,
      "loss": 0.90567428,
      "num_input_tokens_seen": 28229980,
      "step": 1331,
      "time_per_iteration": 2.431732177734375
    },
    {
      "auxiliary_loss_clip": 0.01190779,
      "auxiliary_loss_mlp": 0.00795673,
      "balance_loss_clip": 1.05585694,
      "balance_loss_mlp": 1.02053428,
      "epoch": 0.08008417255373516,
      "flos": 40406219285760.0,
      "grad_norm": 3.154731185758471,
      "language_loss": 0.7350865,
      "learning_rate": 3.973745303858942e-06,
      "loss": 0.754951,
      "num_input_tokens_seen": 28253840,
      "step": 1332,
      "time_per_iteration": 2.67429518699646
    },
    {
      "auxiliary_loss_clip": 0.01195874,
      "auxiliary_loss_mlp": 0.01053874,
      "balance_loss_clip": 1.09148037,
      "balance_loss_mlp": 1.03067589,
      "epoch": 0.08014429580640313,
      "flos": 18478913460480.0,
      "grad_norm": 2.040073806978841,
      "language_loss": 0.82675743,
      "learning_rate": 3.973682368232138e-06,
      "loss": 0.84925497,
      "num_input_tokens_seen": 28271675,
      "step": 1333,
      "time_per_iteration": 2.516737222671509
    },
    {
      "auxiliary_loss_clip": 0.01174358,
      "auxiliary_loss_mlp": 0.01056122,
      "balance_loss_clip": 1.08929944,
      "balance_loss_mlp": 1.03237569,
      "epoch": 0.0802044190590711,
      "flos": 22053461368320.0,
      "grad_norm": 2.4278991120773297,
      "language_loss": 0.75136638,
      "learning_rate": 3.9736193577631015e-06,
      "loss": 0.77367121,
      "num_input_tokens_seen": 28291850,
      "step": 1334,
      "time_per_iteration": 2.6159162521362305
    },
    {
      "auxiliary_loss_clip": 0.01165321,
      "auxiliary_loss_mlp": 0.01057155,
      "balance_loss_clip": 1.05587149,
      "balance_loss_mlp": 1.03381395,
      "epoch": 0.08026454231173906,
      "flos": 24572128055040.0,
      "grad_norm": 1.8897166213671848,
      "language_loss": 0.80116844,
      "learning_rate": 3.973556272454221e-06,
      "loss": 0.82339317,
      "num_input_tokens_seen": 28310780,
      "step": 1335,
      "time_per_iteration": 2.591738224029541
    },
    {
      "auxiliary_loss_clip": 0.0105923,
      "auxiliary_loss_mlp": 0.01020418,
      "balance_loss_clip": 1.03156483,
      "balance_loss_mlp": 1.01681745,
      "epoch": 0.08032466556440704,
      "flos": 52581841459200.0,
      "grad_norm": 0.7394834998864181,
      "language_loss": 0.56052023,
      "learning_rate": 3.973493112307889e-06,
      "loss": 0.58131665,
      "num_input_tokens_seen": 28369985,
      "step": 1336,
      "time_per_iteration": 3.218465805053711
    },
    {
      "auxiliary_loss_clip": 0.0117439,
      "auxiliary_loss_mlp": 0.01059226,
      "balance_loss_clip": 1.05666602,
      "balance_loss_mlp": 1.03623092,
      "epoch": 0.080384788817075,
      "flos": 23842602829440.0,
      "grad_norm": 2.2992934074015645,
      "language_loss": 0.67511618,
      "learning_rate": 3.9734298773265005e-06,
      "loss": 0.69745237,
      "num_input_tokens_seen": 28388670,
      "step": 1337,
      "time_per_iteration": 2.5586206912994385
    },
    {
      "auxiliary_loss_clip": 0.01165354,
      "auxiliary_loss_mlp": 0.01072423,
      "balance_loss_clip": 1.05749238,
      "balance_loss_mlp": 1.04685235,
      "epoch": 0.08044491206974297,
      "flos": 25300719527040.0,
      "grad_norm": 1.8725330679223322,
      "language_loss": 0.86730832,
      "learning_rate": 3.973366567512453e-06,
      "loss": 0.88968611,
      "num_input_tokens_seen": 28411845,
      "step": 1338,
      "time_per_iteration": 2.639244794845581
    },
    {
      "auxiliary_loss_clip": 0.01135332,
      "auxiliary_loss_mlp": 0.01072382,
      "balance_loss_clip": 1.05084527,
      "balance_loss_mlp": 1.04496396,
      "epoch": 0.08050503532241095,
      "flos": 22376549226240.0,
      "grad_norm": 2.5213288721145526,
      "language_loss": 0.87351358,
      "learning_rate": 3.973303182868147e-06,
      "loss": 0.89559072,
      "num_input_tokens_seen": 28427875,
      "step": 1339,
      "time_per_iteration": 2.601747512817383
    },
    {
      "auxiliary_loss_clip": 0.01181457,
      "auxiliary_loss_mlp": 0.01051846,
      "balance_loss_clip": 1.05542326,
      "balance_loss_mlp": 1.0293988,
      "epoch": 0.08056515857507891,
      "flos": 18369421827840.0,
      "grad_norm": 2.347119697257138,
      "language_loss": 0.89481509,
      "learning_rate": 3.973239723395988e-06,
      "loss": 0.91714811,
      "num_input_tokens_seen": 28446615,
      "step": 1340,
      "time_per_iteration": 2.5073328018188477
    },
    {
      "auxiliary_loss_clip": 0.01082345,
      "auxiliary_loss_mlp": 0.01008388,
      "balance_loss_clip": 1.03208733,
      "balance_loss_mlp": 1.00443065,
      "epoch": 0.08062528182774688,
      "flos": 51348130980480.0,
      "grad_norm": 0.8780791031969445,
      "language_loss": 0.64887929,
      "learning_rate": 3.97317618909838e-06,
      "loss": 0.66978663,
      "num_input_tokens_seen": 28505290,
      "step": 1341,
      "time_per_iteration": 3.041718006134033
    },
    {
      "auxiliary_loss_clip": 0.01191823,
      "auxiliary_loss_mlp": 0.01058982,
      "balance_loss_clip": 1.05658054,
      "balance_loss_mlp": 1.03353107,
      "epoch": 0.08068540508041486,
      "flos": 17599712261760.0,
      "grad_norm": 1.95042885043803,
      "language_loss": 0.89343381,
      "learning_rate": 3.973112579977733e-06,
      "loss": 0.91594189,
      "num_input_tokens_seen": 28522735,
      "step": 1342,
      "time_per_iteration": 2.530014753341675
    },
    {
      "auxiliary_loss_clip": 0.01174859,
      "auxiliary_loss_mlp": 0.01062768,
      "balance_loss_clip": 1.06562257,
      "balance_loss_mlp": 1.03754377,
      "epoch": 0.08074552833308282,
      "flos": 10561185486720.0,
      "grad_norm": 2.561275773477413,
      "language_loss": 0.762501,
      "learning_rate": 3.973048896036459e-06,
      "loss": 0.7848773,
      "num_input_tokens_seen": 28539460,
      "step": 1343,
      "time_per_iteration": 2.54150128364563
    },
    {
      "auxiliary_loss_clip": 0.01097797,
      "auxiliary_loss_mlp": 0.01009369,
      "balance_loss_clip": 1.07620847,
      "balance_loss_mlp": 1.00524461,
      "epoch": 0.08080565158575079,
      "flos": 60840254954880.0,
      "grad_norm": 0.795330008709181,
      "language_loss": 0.57427239,
      "learning_rate": 3.972985137276974e-06,
      "loss": 0.59534407,
      "num_input_tokens_seen": 28599855,
      "step": 1344,
      "time_per_iteration": 3.0688464641571045
    },
    {
      "auxiliary_loss_clip": 0.01152022,
      "auxiliary_loss_mlp": 0.01060404,
      "balance_loss_clip": 1.05450237,
      "balance_loss_mlp": 1.03587091,
      "epoch": 0.08086577483841875,
      "flos": 18332361970560.0,
      "grad_norm": 2.434786142918157,
      "language_loss": 0.86915791,
      "learning_rate": 3.972921303701695e-06,
      "loss": 0.89128214,
      "num_input_tokens_seen": 28617585,
      "step": 1345,
      "time_per_iteration": 3.9129714965820312
    },
    {
      "auxiliary_loss_clip": 0.01196309,
      "auxiliary_loss_mlp": 0.01059386,
      "balance_loss_clip": 1.0585413,
      "balance_loss_mlp": 1.03696275,
      "epoch": 0.08092589809108673,
      "flos": 21543601766400.0,
      "grad_norm": 1.7988402900729554,
      "language_loss": 0.87749505,
      "learning_rate": 3.972857395313042e-06,
      "loss": 0.90005207,
      "num_input_tokens_seen": 28636355,
      "step": 1346,
      "time_per_iteration": 3.8620026111602783
    },
    {
      "auxiliary_loss_clip": 0.011804,
      "auxiliary_loss_mlp": 0.01053055,
      "balance_loss_clip": 1.05572617,
      "balance_loss_mlp": 1.02974975,
      "epoch": 0.0809860213437547,
      "flos": 22128012046080.0,
      "grad_norm": 1.5713892466306822,
      "language_loss": 0.92796767,
      "learning_rate": 3.972793412113439e-06,
      "loss": 0.9503023,
      "num_input_tokens_seen": 28656260,
      "step": 1347,
      "time_per_iteration": 2.5278053283691406
    },
    {
      "auxiliary_loss_clip": 0.01204254,
      "auxiliary_loss_mlp": 0.0105952,
      "balance_loss_clip": 1.09220338,
      "balance_loss_mlp": 1.03436649,
      "epoch": 0.08104614459642266,
      "flos": 21725489260800.0,
      "grad_norm": 1.6864334237843444,
      "language_loss": 0.8950876,
      "learning_rate": 3.972729354105312e-06,
      "loss": 0.91772532,
      "num_input_tokens_seen": 28675865,
      "step": 1348,
      "time_per_iteration": 2.501950740814209
    },
    {
      "auxiliary_loss_clip": 0.01152564,
      "auxiliary_loss_mlp": 0.01063474,
      "balance_loss_clip": 1.09079063,
      "balance_loss_mlp": 1.0403831,
      "epoch": 0.08110626784909064,
      "flos": 23951878980480.0,
      "grad_norm": 3.4467167325069368,
      "language_loss": 0.76749754,
      "learning_rate": 3.97266522129109e-06,
      "loss": 0.78965795,
      "num_input_tokens_seen": 28696255,
      "step": 1349,
      "time_per_iteration": 2.610936164855957
    },
    {
      "auxiliary_loss_clip": 0.0119493,
      "auxiliary_loss_mlp": 0.01058324,
      "balance_loss_clip": 1.05528498,
      "balance_loss_mlp": 1.03397,
      "epoch": 0.0811663911017586,
      "flos": 19025689265280.0,
      "grad_norm": 1.8989390577797505,
      "language_loss": 0.88429129,
      "learning_rate": 3.972601013673205e-06,
      "loss": 0.90682381,
      "num_input_tokens_seen": 28713905,
      "step": 1350,
      "time_per_iteration": 3.824913740158081
    },
    {
      "auxiliary_loss_clip": 0.01178826,
      "auxiliary_loss_mlp": 0.00858069,
      "balance_loss_clip": 1.08871794,
      "balance_loss_mlp": 1.14298868,
      "epoch": 0.08122651435442657,
      "flos": 15341290588800.0,
      "grad_norm": 2.1761253749468312,
      "language_loss": 0.82542384,
      "learning_rate": 3.972536731254092e-06,
      "loss": 0.84579277,
      "num_input_tokens_seen": 28732075,
      "step": 1351,
      "time_per_iteration": 2.5552327632904053
    },
    {
      "auxiliary_loss_clip": 0.01191306,
      "auxiliary_loss_mlp": 0.01050706,
      "balance_loss_clip": 1.05221069,
      "balance_loss_mlp": 1.02551651,
      "epoch": 0.08128663760709455,
      "flos": 23221563655680.0,
      "grad_norm": 1.961345574255941,
      "language_loss": 0.75411856,
      "learning_rate": 3.972472374036189e-06,
      "loss": 0.77653861,
      "num_input_tokens_seen": 28751150,
      "step": 1352,
      "time_per_iteration": 3.8863532543182373
    },
    {
      "auxiliary_loss_clip": 0.01186822,
      "auxiliary_loss_mlp": 0.00810722,
      "balance_loss_clip": 1.05904818,
      "balance_loss_mlp": 1.04813504,
      "epoch": 0.08134676085976252,
      "flos": 22965628273920.0,
      "grad_norm": 1.8251132912884378,
      "language_loss": 0.82826072,
      "learning_rate": 3.972407942021935e-06,
      "loss": 0.84823614,
      "num_input_tokens_seen": 28773360,
      "step": 1353,
      "time_per_iteration": 2.5281124114990234
    },
    {
      "auxiliary_loss_clip": 0.01072565,
      "auxiliary_loss_mlp": 0.01009286,
      "balance_loss_clip": 1.02626729,
      "balance_loss_mlp": 1.00544798,
      "epoch": 0.08140688411243048,
      "flos": 64322115816960.0,
      "grad_norm": 0.8549421491455913,
      "language_loss": 0.59737825,
      "learning_rate": 3.972343435213775e-06,
      "loss": 0.61819685,
      "num_input_tokens_seen": 28833390,
      "step": 1354,
      "time_per_iteration": 3.1323704719543457
    },
    {
      "auxiliary_loss_clip": 0.01149119,
      "auxiliary_loss_mlp": 0.01057509,
      "balance_loss_clip": 1.05428398,
      "balance_loss_mlp": 1.03501415,
      "epoch": 0.08146700736509845,
      "flos": 22491858862080.0,
      "grad_norm": 2.1241352541284133,
      "language_loss": 0.82738227,
      "learning_rate": 3.972278853614154e-06,
      "loss": 0.84944856,
      "num_input_tokens_seen": 28852430,
      "step": 1355,
      "time_per_iteration": 2.562041759490967
    },
    {
      "auxiliary_loss_clip": 0.01203307,
      "auxiliary_loss_mlp": 0.010584,
      "balance_loss_clip": 1.09103155,
      "balance_loss_mlp": 1.03371167,
      "epoch": 0.08152713061776642,
      "flos": 20447823513600.0,
      "grad_norm": 1.9701699135637438,
      "language_loss": 0.71023792,
      "learning_rate": 3.972214197225521e-06,
      "loss": 0.73285496,
      "num_input_tokens_seen": 28870685,
      "step": 1356,
      "time_per_iteration": 2.5126326084136963
    },
    {
      "auxiliary_loss_clip": 0.01186326,
      "auxiliary_loss_mlp": 0.01057462,
      "balance_loss_clip": 1.05405152,
      "balance_loss_mlp": 1.03401375,
      "epoch": 0.08158725387043439,
      "flos": 23550218121600.0,
      "grad_norm": 2.730456872983281,
      "language_loss": 0.70089281,
      "learning_rate": 3.972149466050329e-06,
      "loss": 0.72333068,
      "num_input_tokens_seen": 28889860,
      "step": 1357,
      "time_per_iteration": 2.4966936111450195
    },
    {
      "auxiliary_loss_clip": 0.01182222,
      "auxiliary_loss_mlp": 0.0105655,
      "balance_loss_clip": 1.05668271,
      "balance_loss_mlp": 1.03310156,
      "epoch": 0.08164737712310235,
      "flos": 22017335264640.0,
      "grad_norm": 2.652534457599916,
      "language_loss": 0.8441186,
      "learning_rate": 3.97208466009103e-06,
      "loss": 0.86650634,
      "num_input_tokens_seen": 28905865,
      "step": 1358,
      "time_per_iteration": 2.517043352127075
    },
    {
      "auxiliary_loss_clip": 0.01171836,
      "auxiliary_loss_mlp": 0.01063684,
      "balance_loss_clip": 1.0558331,
      "balance_loss_mlp": 1.03782797,
      "epoch": 0.08170750037577033,
      "flos": 23367827836800.0,
      "grad_norm": 2.120111389558974,
      "language_loss": 1.02303839,
      "learning_rate": 3.972019779350084e-06,
      "loss": 1.04539359,
      "num_input_tokens_seen": 28925250,
      "step": 1359,
      "time_per_iteration": 2.55159330368042
    },
    {
      "auxiliary_loss_clip": 0.01127729,
      "auxiliary_loss_mlp": 0.0105833,
      "balance_loss_clip": 1.05472505,
      "balance_loss_mlp": 1.03404701,
      "epoch": 0.0817676236284383,
      "flos": 28397978490240.0,
      "grad_norm": 2.4215780227663326,
      "language_loss": 0.84065056,
      "learning_rate": 3.971954823829951e-06,
      "loss": 0.86251116,
      "num_input_tokens_seen": 28943445,
      "step": 1360,
      "time_per_iteration": 2.690230131149292
    },
    {
      "auxiliary_loss_clip": 0.01194892,
      "auxiliary_loss_mlp": 0.01068033,
      "balance_loss_clip": 1.05528533,
      "balance_loss_mlp": 1.04410791,
      "epoch": 0.08182774688110626,
      "flos": 19208905562880.0,
      "grad_norm": 2.49016972531053,
      "language_loss": 0.71722198,
      "learning_rate": 3.971889793533093e-06,
      "loss": 0.73985118,
      "num_input_tokens_seen": 28962695,
      "step": 1361,
      "time_per_iteration": 2.4689581394195557
    },
    {
      "auxiliary_loss_clip": 0.0115907,
      "auxiliary_loss_mlp": 0.01057113,
      "balance_loss_clip": 1.04698396,
      "balance_loss_mlp": 1.03216243,
      "epoch": 0.08188787013377424,
      "flos": 22784099915520.0,
      "grad_norm": 2.200134914475846,
      "language_loss": 0.76544893,
      "learning_rate": 3.971824688461976e-06,
      "loss": 0.78761077,
      "num_input_tokens_seen": 28982120,
      "step": 1362,
      "time_per_iteration": 2.5434932708740234
    },
    {
      "auxiliary_loss_clip": 0.01194065,
      "auxiliary_loss_mlp": 0.01055553,
      "balance_loss_clip": 1.05716801,
      "balance_loss_mlp": 1.03339171,
      "epoch": 0.08194799338644221,
      "flos": 16468095214080.0,
      "grad_norm": 2.5264349234260406,
      "language_loss": 0.72381639,
      "learning_rate": 3.971759508619069e-06,
      "loss": 0.7463125,
      "num_input_tokens_seen": 28998100,
      "step": 1363,
      "time_per_iteration": 2.438025712966919
    },
    {
      "auxiliary_loss_clip": 0.0119535,
      "auxiliary_loss_mlp": 0.01066038,
      "balance_loss_clip": 1.05911791,
      "balance_loss_mlp": 1.04053903,
      "epoch": 0.08200811663911017,
      "flos": 23913633974400.0,
      "grad_norm": 2.9670142336255867,
      "language_loss": 0.77221179,
      "learning_rate": 3.971694254006844e-06,
      "loss": 0.79482567,
      "num_input_tokens_seen": 29017095,
      "step": 1364,
      "time_per_iteration": 2.5069632530212402
    },
    {
      "auxiliary_loss_clip": 0.01146967,
      "auxiliary_loss_mlp": 0.01066932,
      "balance_loss_clip": 1.07619596,
      "balance_loss_mlp": 1.04002666,
      "epoch": 0.08206823989177814,
      "flos": 17896550256000.0,
      "grad_norm": 1.6891767437743876,
      "language_loss": 0.82133698,
      "learning_rate": 3.971628924627776e-06,
      "loss": 0.84347594,
      "num_input_tokens_seen": 29037240,
      "step": 1365,
      "time_per_iteration": 2.6055777072906494
    },
    {
      "auxiliary_loss_clip": 0.01184743,
      "auxiliary_loss_mlp": 0.01059947,
      "balance_loss_clip": 1.05738413,
      "balance_loss_mlp": 1.03635502,
      "epoch": 0.08212836314444612,
      "flos": 22088186841600.0,
      "grad_norm": 1.7937527562322797,
      "language_loss": 0.82316637,
      "learning_rate": 3.97156352048434e-06,
      "loss": 0.84561336,
      "num_input_tokens_seen": 29056250,
      "step": 1366,
      "time_per_iteration": 2.517164707183838
    },
    {
      "auxiliary_loss_clip": 0.01151298,
      "auxiliary_loss_mlp": 0.01076251,
      "balance_loss_clip": 1.05310857,
      "balance_loss_mlp": 1.05193257,
      "epoch": 0.08218848639711408,
      "flos": 17597485618560.0,
      "grad_norm": 2.2678581268030182,
      "language_loss": 0.81732595,
      "learning_rate": 3.97149804157902e-06,
      "loss": 0.83960152,
      "num_input_tokens_seen": 29073380,
      "step": 1367,
      "time_per_iteration": 2.5454294681549072
    },
    {
      "auxiliary_loss_clip": 0.0119738,
      "auxiliary_loss_mlp": 0.01062053,
      "balance_loss_clip": 1.0557487,
      "balance_loss_mlp": 1.03858113,
      "epoch": 0.08224860964978205,
      "flos": 17857838373120.0,
      "grad_norm": 2.267629418302392,
      "language_loss": 0.83387166,
      "learning_rate": 3.9714324879142946e-06,
      "loss": 0.856466,
      "num_input_tokens_seen": 29091330,
      "step": 1368,
      "time_per_iteration": 2.453807830810547
    },
    {
      "auxiliary_loss_clip": 0.01149537,
      "auxiliary_loss_mlp": 0.0104851,
      "balance_loss_clip": 1.05284393,
      "balance_loss_mlp": 1.02639651,
      "epoch": 0.08230873290245003,
      "flos": 25227533566080.0,
      "grad_norm": 1.672830543746323,
      "language_loss": 0.81268024,
      "learning_rate": 3.971366859492653e-06,
      "loss": 0.83466071,
      "num_input_tokens_seen": 29110375,
      "step": 1369,
      "time_per_iteration": 2.5987918376922607
    },
    {
      "auxiliary_loss_clip": 0.01134819,
      "auxiliary_loss_mlp": 0.00897687,
      "balance_loss_clip": 1.05420494,
      "balance_loss_mlp": 1.20826483,
      "epoch": 0.08236885615511799,
      "flos": 31759935753600.0,
      "grad_norm": 2.3569416676976394,
      "language_loss": 0.75210464,
      "learning_rate": 3.971301156316582e-06,
      "loss": 0.77242965,
      "num_input_tokens_seen": 29129395,
      "step": 1370,
      "time_per_iteration": 2.675539493560791
    },
    {
      "auxiliary_loss_clip": 0.01145145,
      "auxiliary_loss_mlp": 0.01061663,
      "balance_loss_clip": 1.05532598,
      "balance_loss_mlp": 1.03771377,
      "epoch": 0.08242897940778596,
      "flos": 23185832601600.0,
      "grad_norm": 1.7746662832576605,
      "language_loss": 0.74430811,
      "learning_rate": 3.971235378388573e-06,
      "loss": 0.76637614,
      "num_input_tokens_seen": 29148650,
      "step": 1371,
      "time_per_iteration": 2.606079339981079
    },
    {
      "auxiliary_loss_clip": 0.01093728,
      "auxiliary_loss_mlp": 0.01061006,
      "balance_loss_clip": 1.05149138,
      "balance_loss_mlp": 1.0356741,
      "epoch": 0.08248910266045394,
      "flos": 34491480393600.0,
      "grad_norm": 2.191756877851688,
      "language_loss": 0.70751309,
      "learning_rate": 3.971169525711122e-06,
      "loss": 0.72906041,
      "num_input_tokens_seen": 29170785,
      "step": 1372,
      "time_per_iteration": 3.0113046169281006
    },
    {
      "auxiliary_loss_clip": 0.01164975,
      "auxiliary_loss_mlp": 0.0105387,
      "balance_loss_clip": 1.0682075,
      "balance_loss_mlp": 1.02953911,
      "epoch": 0.0825492259131219,
      "flos": 13436228960640.0,
      "grad_norm": 2.477101707785931,
      "language_loss": 0.88175493,
      "learning_rate": 3.9711035982867246e-06,
      "loss": 0.90394342,
      "num_input_tokens_seen": 29185210,
      "step": 1373,
      "time_per_iteration": 2.701233386993408
    },
    {
      "auxiliary_loss_clip": 0.01146135,
      "auxiliary_loss_mlp": 0.01061176,
      "balance_loss_clip": 1.05686069,
      "balance_loss_mlp": 1.03772759,
      "epoch": 0.08260934916578987,
      "flos": 25812446636160.0,
      "grad_norm": 1.7556608681570234,
      "language_loss": 0.82056987,
      "learning_rate": 3.971037596117882e-06,
      "loss": 0.84264302,
      "num_input_tokens_seen": 29205210,
      "step": 1374,
      "time_per_iteration": 2.609203577041626
    },
    {
      "auxiliary_loss_clip": 0.0104044,
      "auxiliary_loss_mlp": 0.01022793,
      "balance_loss_clip": 1.02812564,
      "balance_loss_mlp": 1.01902616,
      "epoch": 0.08266947241845783,
      "flos": 63460009491840.0,
      "grad_norm": 0.8224806421073857,
      "language_loss": 0.60635924,
      "learning_rate": 3.970971519207095e-06,
      "loss": 0.62699157,
      "num_input_tokens_seen": 29265350,
      "step": 1375,
      "time_per_iteration": 3.2022194862365723
    },
    {
      "auxiliary_loss_clip": 0.01058865,
      "auxiliary_loss_mlp": 0.01009147,
      "balance_loss_clip": 1.0207082,
      "balance_loss_mlp": 1.00588059,
      "epoch": 0.08272959567112581,
      "flos": 69993704568960.0,
      "grad_norm": 0.9075195824136949,
      "language_loss": 0.62189412,
      "learning_rate": 3.970905367556871e-06,
      "loss": 0.64257431,
      "num_input_tokens_seen": 29321475,
      "step": 1376,
      "time_per_iteration": 3.061965227127075
    },
    {
      "auxiliary_loss_clip": 0.011643,
      "auxiliary_loss_mlp": 0.01069168,
      "balance_loss_clip": 1.07208121,
      "balance_loss_mlp": 1.0456841,
      "epoch": 0.08278971892379378,
      "flos": 20413205781120.0,
      "grad_norm": 1.8100377799768825,
      "language_loss": 0.82557929,
      "learning_rate": 3.970839141169718e-06,
      "loss": 0.84791398,
      "num_input_tokens_seen": 29341405,
      "step": 1377,
      "time_per_iteration": 2.646111488342285
    },
    {
      "auxiliary_loss_clip": 0.01167409,
      "auxiliary_loss_mlp": 0.01055555,
      "balance_loss_clip": 1.05490303,
      "balance_loss_mlp": 1.03203547,
      "epoch": 0.08284984217646174,
      "flos": 26250233598720.0,
      "grad_norm": 1.7518277713017192,
      "language_loss": 0.84448099,
      "learning_rate": 3.970772840048147e-06,
      "loss": 0.86671066,
      "num_input_tokens_seen": 29361955,
      "step": 1378,
      "time_per_iteration": 2.589545249938965
    },
    {
      "auxiliary_loss_clip": 0.01179713,
      "auxiliary_loss_mlp": 0.01066904,
      "balance_loss_clip": 1.05354333,
      "balance_loss_mlp": 1.04240668,
      "epoch": 0.08290996542912972,
      "flos": 27194683852800.0,
      "grad_norm": 3.0575904561770804,
      "language_loss": 0.87582791,
      "learning_rate": 3.970706464194672e-06,
      "loss": 0.89829403,
      "num_input_tokens_seen": 29382395,
      "step": 1379,
      "time_per_iteration": 2.603670358657837
    },
    {
      "auxiliary_loss_clip": 0.01161965,
      "auxiliary_loss_mlp": 0.01062357,
      "balance_loss_clip": 1.0676018,
      "balance_loss_mlp": 1.03919435,
      "epoch": 0.08297008868179769,
      "flos": 38618191146240.0,
      "grad_norm": 2.1479643117290066,
      "language_loss": 0.78515524,
      "learning_rate": 3.970640013611812e-06,
      "loss": 0.80739844,
      "num_input_tokens_seen": 29404460,
      "step": 1380,
      "time_per_iteration": 2.7013583183288574
    },
    {
      "auxiliary_loss_clip": 0.01177288,
      "auxiliary_loss_mlp": 0.01057459,
      "balance_loss_clip": 1.05616283,
      "balance_loss_mlp": 1.03279459,
      "epoch": 0.08303021193446565,
      "flos": 19974736460160.0,
      "grad_norm": 2.938776720082317,
      "language_loss": 0.86105061,
      "learning_rate": 3.970573488302083e-06,
      "loss": 0.88339806,
      "num_input_tokens_seen": 29422675,
      "step": 1381,
      "time_per_iteration": 2.5222864151000977
    },
    {
      "auxiliary_loss_clip": 0.01190585,
      "auxiliary_loss_mlp": 0.00890074,
      "balance_loss_clip": 1.0583744,
      "balance_loss_mlp": 1.19936597,
      "epoch": 0.08309033518713363,
      "flos": 13662646341120.0,
      "grad_norm": 3.7238317189119847,
      "language_loss": 0.88423169,
      "learning_rate": 3.970506888268011e-06,
      "loss": 0.90503824,
      "num_input_tokens_seen": 29439840,
      "step": 1382,
      "time_per_iteration": 2.5308749675750732
    },
    {
      "auxiliary_loss_clip": 0.01154684,
      "auxiliary_loss_mlp": 0.01059978,
      "balance_loss_clip": 1.05543566,
      "balance_loss_mlp": 1.03745937,
      "epoch": 0.0831504584398016,
      "flos": 17968551068160.0,
      "grad_norm": 2.8361848316785485,
      "language_loss": 0.77227545,
      "learning_rate": 3.970440213512121e-06,
      "loss": 0.79442203,
      "num_input_tokens_seen": 29457360,
      "step": 1383,
      "time_per_iteration": 3.9899027347564697
    },
    {
      "auxiliary_loss_clip": 0.011873,
      "auxiliary_loss_mlp": 0.01058517,
      "balance_loss_clip": 1.05763626,
      "balance_loss_mlp": 1.03459144,
      "epoch": 0.08321058169246956,
      "flos": 22601386408320.0,
      "grad_norm": 1.9280332702507494,
      "language_loss": 0.82792896,
      "learning_rate": 3.97037346403694e-06,
      "loss": 0.85038716,
      "num_input_tokens_seen": 29477040,
      "step": 1384,
      "time_per_iteration": 2.5434463024139404
    },
    {
      "auxiliary_loss_clip": 0.0114428,
      "auxiliary_loss_mlp": 0.01057706,
      "balance_loss_clip": 1.05584311,
      "balance_loss_mlp": 1.02971518,
      "epoch": 0.08327070494513754,
      "flos": 22850426378880.0,
      "grad_norm": 2.445617375163999,
      "language_loss": 0.84979594,
      "learning_rate": 3.970306639845e-06,
      "loss": 0.8718158,
      "num_input_tokens_seen": 29492010,
      "step": 1385,
      "time_per_iteration": 3.997011184692383
    },
    {
      "auxiliary_loss_clip": 0.01152675,
      "auxiliary_loss_mlp": 0.01067989,
      "balance_loss_clip": 1.05599785,
      "balance_loss_mlp": 1.0429548,
      "epoch": 0.0833308281978055,
      "flos": 22782986593920.0,
      "grad_norm": 3.22368202664472,
      "language_loss": 0.68799365,
      "learning_rate": 3.970239740938835e-06,
      "loss": 0.71020031,
      "num_input_tokens_seen": 29511850,
      "step": 1386,
      "time_per_iteration": 2.5908398628234863
    },
    {
      "auxiliary_loss_clip": 0.01170583,
      "auxiliary_loss_mlp": 0.01056343,
      "balance_loss_clip": 1.05424988,
      "balance_loss_mlp": 1.03229821,
      "epoch": 0.08339095145047347,
      "flos": 20812604083200.0,
      "grad_norm": 1.7340584476387966,
      "language_loss": 0.82052439,
      "learning_rate": 3.97017276732098e-06,
      "loss": 0.84279364,
      "num_input_tokens_seen": 29531415,
      "step": 1387,
      "time_per_iteration": 2.5233919620513916
    },
    {
      "auxiliary_loss_clip": 0.01171544,
      "auxiliary_loss_mlp": 0.01068417,
      "balance_loss_clip": 1.05527532,
      "balance_loss_mlp": 1.04350209,
      "epoch": 0.08345107470314143,
      "flos": 18515326872960.0,
      "grad_norm": 2.881857166179083,
      "language_loss": 0.77649295,
      "learning_rate": 3.970105718993978e-06,
      "loss": 0.79889262,
      "num_input_tokens_seen": 29549525,
      "step": 1388,
      "time_per_iteration": 2.6173527240753174
    },
    {
      "auxiliary_loss_clip": 0.0112539,
      "auxiliary_loss_mlp": 0.01065496,
      "balance_loss_clip": 1.05334783,
      "balance_loss_mlp": 1.04036713,
      "epoch": 0.08351119795580941,
      "flos": 18807567926400.0,
      "grad_norm": 2.1308519509124317,
      "language_loss": 0.7875731,
      "learning_rate": 3.970038595960369e-06,
      "loss": 0.80948198,
      "num_input_tokens_seen": 29568705,
      "step": 1389,
      "time_per_iteration": 4.030388593673706
    },
    {
      "auxiliary_loss_clip": 0.01169976,
      "auxiliary_loss_mlp": 0.0105954,
      "balance_loss_clip": 1.05550814,
      "balance_loss_mlp": 1.03548348,
      "epoch": 0.08357132120847738,
      "flos": 18441817689600.0,
      "grad_norm": 2.2344612518923666,
      "language_loss": 0.87197471,
      "learning_rate": 3.969971398222699e-06,
      "loss": 0.89426982,
      "num_input_tokens_seen": 29585855,
      "step": 1390,
      "time_per_iteration": 2.4936447143554688
    },
    {
      "auxiliary_loss_clip": 0.01157074,
      "auxiliary_loss_mlp": 0.01067847,
      "balance_loss_clip": 1.05366492,
      "balance_loss_mlp": 1.04129887,
      "epoch": 0.08363144446114534,
      "flos": 25922333318400.0,
      "grad_norm": 1.7341200012645293,
      "language_loss": 0.86993837,
      "learning_rate": 3.969904125783517e-06,
      "loss": 0.89218765,
      "num_input_tokens_seen": 29607280,
      "step": 1391,
      "time_per_iteration": 3.943129062652588
    },
    {
      "auxiliary_loss_clip": 0.01152783,
      "auxiliary_loss_mlp": 0.01070827,
      "balance_loss_clip": 1.05673289,
      "balance_loss_mlp": 1.04697299,
      "epoch": 0.08369156771381332,
      "flos": 18041306065920.0,
      "grad_norm": 2.5061505314481667,
      "language_loss": 0.87487209,
      "learning_rate": 3.969836778645371e-06,
      "loss": 0.89710814,
      "num_input_tokens_seen": 29624130,
      "step": 1392,
      "time_per_iteration": 2.5995640754699707
    },
    {
      "auxiliary_loss_clip": 0.01180615,
      "auxiliary_loss_mlp": 0.01061944,
      "balance_loss_clip": 1.05436158,
      "balance_loss_mlp": 1.03818607,
      "epoch": 0.08375169096648129,
      "flos": 22675111073280.0,
      "grad_norm": 4.552697192328562,
      "language_loss": 0.79898894,
      "learning_rate": 3.969769356810819e-06,
      "loss": 0.82141447,
      "num_input_tokens_seen": 29643210,
      "step": 1393,
      "time_per_iteration": 2.5235164165496826
    },
    {
      "auxiliary_loss_clip": 0.01195648,
      "auxiliary_loss_mlp": 0.01056736,
      "balance_loss_clip": 1.05996811,
      "balance_loss_mlp": 1.03364515,
      "epoch": 0.08381181421914925,
      "flos": 26103215232000.0,
      "grad_norm": 1.7724223673837824,
      "language_loss": 0.84837413,
      "learning_rate": 3.969701860282415e-06,
      "loss": 0.87089789,
      "num_input_tokens_seen": 29663920,
      "step": 1394,
      "time_per_iteration": 2.5576984882354736
    },
    {
      "auxiliary_loss_clip": 0.01143659,
      "auxiliary_loss_mlp": 0.01056912,
      "balance_loss_clip": 1.05838418,
      "balance_loss_mlp": 1.03335643,
      "epoch": 0.08387193747181723,
      "flos": 20629782835200.0,
      "grad_norm": 2.204175162367542,
      "language_loss": 0.83022642,
      "learning_rate": 3.969634289062719e-06,
      "loss": 0.8522321,
      "num_input_tokens_seen": 29683825,
      "step": 1395,
      "time_per_iteration": 2.6070992946624756
    },
    {
      "auxiliary_loss_clip": 0.01186113,
      "auxiliary_loss_mlp": 0.00872722,
      "balance_loss_clip": 1.05989683,
      "balance_loss_mlp": 1.16321957,
      "epoch": 0.0839320607244852,
      "flos": 13443196199040.0,
      "grad_norm": 3.9227238215097096,
      "language_loss": 0.8299917,
      "learning_rate": 3.969566643154293e-06,
      "loss": 0.8505801,
      "num_input_tokens_seen": 29698775,
      "step": 1396,
      "time_per_iteration": 2.504648447036743
    },
    {
      "auxiliary_loss_clip": 0.01183941,
      "auxiliary_loss_mlp": 0.01062161,
      "balance_loss_clip": 1.06043482,
      "balance_loss_mlp": 1.03632832,
      "epoch": 0.08399218397715316,
      "flos": 23477247642240.0,
      "grad_norm": 2.161229699558928,
      "language_loss": 0.76433504,
      "learning_rate": 3.969498922559703e-06,
      "loss": 0.78679609,
      "num_input_tokens_seen": 29719430,
      "step": 1397,
      "time_per_iteration": 2.5364434719085693
    },
    {
      "auxiliary_loss_clip": 0.0115222,
      "auxiliary_loss_mlp": 0.01051812,
      "balance_loss_clip": 1.05606222,
      "balance_loss_mlp": 1.02684915,
      "epoch": 0.08405230722982113,
      "flos": 25920717206400.0,
      "grad_norm": 2.079525483133427,
      "language_loss": 0.78191149,
      "learning_rate": 3.969431127281516e-06,
      "loss": 0.80395186,
      "num_input_tokens_seen": 29739685,
      "step": 1398,
      "time_per_iteration": 2.612788677215576
    },
    {
      "auxiliary_loss_clip": 0.01191011,
      "auxiliary_loss_mlp": 0.01055553,
      "balance_loss_clip": 1.05782962,
      "balance_loss_mlp": 1.03184235,
      "epoch": 0.0841124304824891,
      "flos": 17967437746560.0,
      "grad_norm": 2.517325156913226,
      "language_loss": 0.95005888,
      "learning_rate": 3.969363257322304e-06,
      "loss": 0.97252452,
      "num_input_tokens_seen": 29756165,
      "step": 1399,
      "time_per_iteration": 2.446211814880371
    },
    {
      "auxiliary_loss_clip": 0.01173696,
      "auxiliary_loss_mlp": 0.01065278,
      "balance_loss_clip": 1.05642569,
      "balance_loss_mlp": 1.03859913,
      "epoch": 0.08417255373515707,
      "flos": 25629661301760.0,
      "grad_norm": 2.3651670784164684,
      "language_loss": 0.81925595,
      "learning_rate": 3.96929531268464e-06,
      "loss": 0.8416456,
      "num_input_tokens_seen": 29776425,
      "step": 1400,
      "time_per_iteration": 2.5432238578796387
    },
    {
      "auxiliary_loss_clip": 0.01169142,
      "auxiliary_loss_mlp": 0.01058607,
      "balance_loss_clip": 1.05840206,
      "balance_loss_mlp": 1.03462243,
      "epoch": 0.08423267698782504,
      "flos": 26249730808320.0,
      "grad_norm": 2.1079399510985963,
      "language_loss": 0.86965173,
      "learning_rate": 3.969227293371099e-06,
      "loss": 0.89192921,
      "num_input_tokens_seen": 29796440,
      "step": 1401,
      "time_per_iteration": 2.589743137359619
    },
    {
      "auxiliary_loss_clip": 0.01194684,
      "auxiliary_loss_mlp": 0.01061837,
      "balance_loss_clip": 1.05581963,
      "balance_loss_mlp": 1.03607583,
      "epoch": 0.08429280024049302,
      "flos": 20119707751680.0,
      "grad_norm": 2.162950752363109,
      "language_loss": 0.87466592,
      "learning_rate": 3.969159199384263e-06,
      "loss": 0.8972311,
      "num_input_tokens_seen": 29814755,
      "step": 1402,
      "time_per_iteration": 2.462029218673706
    },
    {
      "auxiliary_loss_clip": 0.01153605,
      "auxiliary_loss_mlp": 0.00846838,
      "balance_loss_clip": 1.06595874,
      "balance_loss_mlp": 1.11910236,
      "epoch": 0.08435292349316098,
      "flos": 42924526836480.0,
      "grad_norm": 2.345988116567741,
      "language_loss": 0.88876134,
      "learning_rate": 3.9690910307267125e-06,
      "loss": 0.90876579,
      "num_input_tokens_seen": 29834785,
      "step": 1403,
      "time_per_iteration": 2.7763559818267822
    },
    {
      "auxiliary_loss_clip": 0.01165005,
      "auxiliary_loss_mlp": 0.01057184,
      "balance_loss_clip": 1.05639803,
      "balance_loss_mlp": 1.03236449,
      "epoch": 0.08441304674582895,
      "flos": 22857285876480.0,
      "grad_norm": 1.882429202391723,
      "language_loss": 0.80095994,
      "learning_rate": 3.969022787401033e-06,
      "loss": 0.82318187,
      "num_input_tokens_seen": 29854695,
      "step": 1404,
      "time_per_iteration": 2.545522689819336
    },
    {
      "auxiliary_loss_clip": 0.01178434,
      "auxiliary_loss_mlp": 0.01068557,
      "balance_loss_clip": 1.05901933,
      "balance_loss_mlp": 1.04379702,
      "epoch": 0.08447316999849692,
      "flos": 18697501676160.0,
      "grad_norm": 2.188668829162983,
      "language_loss": 0.83544254,
      "learning_rate": 3.968954469409811e-06,
      "loss": 0.85791242,
      "num_input_tokens_seen": 29872180,
      "step": 1405,
      "time_per_iteration": 2.511488914489746
    },
    {
      "auxiliary_loss_clip": 0.01178522,
      "auxiliary_loss_mlp": 0.01056872,
      "balance_loss_clip": 1.05285501,
      "balance_loss_mlp": 1.03363848,
      "epoch": 0.08453329325116489,
      "flos": 25483971738240.0,
      "grad_norm": 1.5831089315003686,
      "language_loss": 0.79751539,
      "learning_rate": 3.968886076755639e-06,
      "loss": 0.81986934,
      "num_input_tokens_seen": 29893205,
      "step": 1406,
      "time_per_iteration": 2.5482656955718994
    },
    {
      "auxiliary_loss_clip": 0.01178643,
      "auxiliary_loss_mlp": 0.0106595,
      "balance_loss_clip": 1.06923532,
      "balance_loss_mlp": 1.04226279,
      "epoch": 0.08459341650383286,
      "flos": 20920048640640.0,
      "grad_norm": 1.8019563381331154,
      "language_loss": 0.7957387,
      "learning_rate": 3.96881760944111e-06,
      "loss": 0.81818467,
      "num_input_tokens_seen": 29911970,
      "step": 1407,
      "time_per_iteration": 2.542900323867798
    },
    {
      "auxiliary_loss_clip": 0.01182629,
      "auxiliary_loss_mlp": 0.01052793,
      "balance_loss_clip": 1.05551589,
      "balance_loss_mlp": 1.02984524,
      "epoch": 0.08465353975650082,
      "flos": 13043079624960.0,
      "grad_norm": 2.1497825349822746,
      "language_loss": 0.91968518,
      "learning_rate": 3.968749067468819e-06,
      "loss": 0.94203937,
      "num_input_tokens_seen": 29929925,
      "step": 1408,
      "time_per_iteration": 2.5068957805633545
    },
    {
      "auxiliary_loss_clip": 0.01069168,
      "auxiliary_loss_mlp": 0.0101737,
      "balance_loss_clip": 1.02955818,
      "balance_loss_mlp": 1.01386535,
      "epoch": 0.0847136630091688,
      "flos": 60877422552960.0,
      "grad_norm": 0.88392969356341,
      "language_loss": 0.61851841,
      "learning_rate": 3.968680450841368e-06,
      "loss": 0.63938379,
      "num_input_tokens_seen": 29985950,
      "step": 1409,
      "time_per_iteration": 3.221773147583008
    },
    {
      "auxiliary_loss_clip": 0.01185434,
      "auxiliary_loss_mlp": 0.01057048,
      "balance_loss_clip": 1.05449247,
      "balance_loss_mlp": 1.03417218,
      "epoch": 0.08477378626183676,
      "flos": 22046530043520.0,
      "grad_norm": 2.0555788572955818,
      "language_loss": 0.86712021,
      "learning_rate": 3.968611759561355e-06,
      "loss": 0.88954502,
      "num_input_tokens_seen": 30004330,
      "step": 1410,
      "time_per_iteration": 2.483055591583252
    },
    {
      "auxiliary_loss_clip": 0.01176704,
      "auxiliary_loss_mlp": 0.01052888,
      "balance_loss_clip": 1.05510259,
      "balance_loss_mlp": 1.02799749,
      "epoch": 0.08483390951450473,
      "flos": 16690059308160.0,
      "grad_norm": 2.7087830253917407,
      "language_loss": 0.73945272,
      "learning_rate": 3.968542993631388e-06,
      "loss": 0.76174873,
      "num_input_tokens_seen": 30022555,
      "step": 1411,
      "time_per_iteration": 2.5148556232452393
    },
    {
      "auxiliary_loss_clip": 0.01083948,
      "auxiliary_loss_mlp": 0.01005256,
      "balance_loss_clip": 1.02542758,
      "balance_loss_mlp": 1.00163221,
      "epoch": 0.08489403276717271,
      "flos": 51584640082560.0,
      "grad_norm": 0.9120626682348201,
      "language_loss": 0.56755084,
      "learning_rate": 3.968474153054073e-06,
      "loss": 0.58844286,
      "num_input_tokens_seen": 30077220,
      "step": 1412,
      "time_per_iteration": 3.0098018646240234
    },
    {
      "auxiliary_loss_clip": 0.01155241,
      "auxiliary_loss_mlp": 0.01074551,
      "balance_loss_clip": 1.05215609,
      "balance_loss_mlp": 1.04878998,
      "epoch": 0.08495415601984067,
      "flos": 17092330698240.0,
      "grad_norm": 2.171377085053349,
      "language_loss": 0.89338124,
      "learning_rate": 3.96840523783202e-06,
      "loss": 0.9156791,
      "num_input_tokens_seen": 30094600,
      "step": 1413,
      "time_per_iteration": 2.5200605392456055
    },
    {
      "auxiliary_loss_clip": 0.01163606,
      "auxiliary_loss_mlp": 0.01053293,
      "balance_loss_clip": 1.05668688,
      "balance_loss_mlp": 1.02868867,
      "epoch": 0.08501427927250864,
      "flos": 23148413608320.0,
      "grad_norm": 2.1201041957451294,
      "language_loss": 0.87991083,
      "learning_rate": 3.968336247967844e-06,
      "loss": 0.90207976,
      "num_input_tokens_seen": 30114475,
      "step": 1414,
      "time_per_iteration": 2.552064895629883
    },
    {
      "auxiliary_loss_clip": 0.01167031,
      "auxiliary_loss_mlp": 0.01065573,
      "balance_loss_clip": 1.056283,
      "balance_loss_mlp": 1.04337621,
      "epoch": 0.08507440252517662,
      "flos": 19063467394560.0,
      "grad_norm": 1.907492519910922,
      "language_loss": 0.77443445,
      "learning_rate": 3.96826718346416e-06,
      "loss": 0.79676056,
      "num_input_tokens_seen": 30133350,
      "step": 1415,
      "time_per_iteration": 2.536937952041626
    },
    {
      "auxiliary_loss_clip": 0.01177887,
      "auxiliary_loss_mlp": 0.0106011,
      "balance_loss_clip": 1.0552938,
      "balance_loss_mlp": 1.03866386,
      "epoch": 0.08513452577784458,
      "flos": 60182296600320.0,
      "grad_norm": 2.330268511497652,
      "language_loss": 0.70902288,
      "learning_rate": 3.968198044323587e-06,
      "loss": 0.73140287,
      "num_input_tokens_seen": 30159005,
      "step": 1416,
      "time_per_iteration": 2.8506417274475098
    },
    {
      "auxiliary_loss_clip": 0.01172443,
      "auxiliary_loss_mlp": 0.01065794,
      "balance_loss_clip": 1.05826163,
      "balance_loss_mlp": 1.04030681,
      "epoch": 0.08519464903051255,
      "flos": 27308485117440.0,
      "grad_norm": 2.161262581385308,
      "language_loss": 0.75072491,
      "learning_rate": 3.968128830548748e-06,
      "loss": 0.77310729,
      "num_input_tokens_seen": 30179450,
      "step": 1417,
      "time_per_iteration": 2.6399874687194824
    },
    {
      "auxiliary_loss_clip": 0.01163884,
      "auxiliary_loss_mlp": 0.01053818,
      "balance_loss_clip": 1.05194819,
      "balance_loss_mlp": 1.03029823,
      "epoch": 0.08525477228318051,
      "flos": 20266438809600.0,
      "grad_norm": 3.7367646176802873,
      "language_loss": 0.82233644,
      "learning_rate": 3.968059542142265e-06,
      "loss": 0.84451348,
      "num_input_tokens_seen": 30197235,
      "step": 1418,
      "time_per_iteration": 2.520747423171997
    },
    {
      "auxiliary_loss_clip": 0.01049132,
      "auxiliary_loss_mlp": 0.0102198,
      "balance_loss_clip": 1.02819371,
      "balance_loss_mlp": 1.0182128,
      "epoch": 0.08531489553584849,
      "flos": 67615017183360.0,
      "grad_norm": 0.8681441192398974,
      "language_loss": 0.5662781,
      "learning_rate": 3.9679901791067685e-06,
      "loss": 0.58698922,
      "num_input_tokens_seen": 30257410,
      "step": 1419,
      "time_per_iteration": 3.1313247680664062
    },
    {
      "auxiliary_loss_clip": 0.01187534,
      "auxiliary_loss_mlp": 0.0105756,
      "balance_loss_clip": 1.05239344,
      "balance_loss_mlp": 1.03390884,
      "epoch": 0.08537501878851646,
      "flos": 27526965592320.0,
      "grad_norm": 2.289767681583507,
      "language_loss": 0.70208764,
      "learning_rate": 3.967920741444886e-06,
      "loss": 0.72453862,
      "num_input_tokens_seen": 30277865,
      "step": 1420,
      "time_per_iteration": 2.517216444015503
    },
    {
      "auxiliary_loss_clip": 0.01150014,
      "auxiliary_loss_mlp": 0.01050475,
      "balance_loss_clip": 1.05045009,
      "balance_loss_mlp": 1.02665687,
      "epoch": 0.08543514204118442,
      "flos": 22784243569920.0,
      "grad_norm": 1.5635443028992941,
      "language_loss": 0.87954116,
      "learning_rate": 3.967851229159252e-06,
      "loss": 0.901546,
      "num_input_tokens_seen": 30298545,
      "step": 1421,
      "time_per_iteration": 4.0657830238342285
    },
    {
      "auxiliary_loss_clip": 0.01083126,
      "auxiliary_loss_mlp": 0.01007651,
      "balance_loss_clip": 1.02647924,
      "balance_loss_mlp": 1.00395584,
      "epoch": 0.0854952652938524,
      "flos": 60990721027200.0,
      "grad_norm": 0.7913341862450514,
      "language_loss": 0.63474226,
      "learning_rate": 3.967781642252502e-06,
      "loss": 0.65565002,
      "num_input_tokens_seen": 30361725,
      "step": 1422,
      "time_per_iteration": 3.0729610919952393
    },
    {
      "auxiliary_loss_clip": 0.01145952,
      "auxiliary_loss_mlp": 0.0105375,
      "balance_loss_clip": 1.05371153,
      "balance_loss_mlp": 1.03071904,
      "epoch": 0.08555538854652037,
      "flos": 28038046256640.0,
      "grad_norm": 2.816338005345657,
      "language_loss": 0.82965112,
      "learning_rate": 3.967711980727276e-06,
      "loss": 0.85164815,
      "num_input_tokens_seen": 30382180,
      "step": 1423,
      "time_per_iteration": 4.000679969787598
    },
    {
      "auxiliary_loss_clip": 0.01154985,
      "auxiliary_loss_mlp": 0.01062477,
      "balance_loss_clip": 1.0565834,
      "balance_loss_mlp": 1.03837311,
      "epoch": 0.08561551179918833,
      "flos": 23509279595520.0,
      "grad_norm": 1.6839336331805932,
      "language_loss": 0.75055611,
      "learning_rate": 3.967642244586213e-06,
      "loss": 0.77273077,
      "num_input_tokens_seen": 30402980,
      "step": 1424,
      "time_per_iteration": 2.6163113117218018
    },
    {
      "auxiliary_loss_clip": 0.0113492,
      "auxiliary_loss_mlp": 0.01054978,
      "balance_loss_clip": 1.05402708,
      "balance_loss_mlp": 1.0321734,
      "epoch": 0.08567563505185631,
      "flos": 17926930183680.0,
      "grad_norm": 2.0534475492613002,
      "language_loss": 0.75968826,
      "learning_rate": 3.96757243383196e-06,
      "loss": 0.78158724,
      "num_input_tokens_seen": 30420800,
      "step": 1425,
      "time_per_iteration": 2.5937554836273193
    },
    {
      "auxiliary_loss_clip": 0.01187342,
      "auxiliary_loss_mlp": 0.01048808,
      "balance_loss_clip": 1.05553544,
      "balance_loss_mlp": 1.02633739,
      "epoch": 0.08573575830452428,
      "flos": 19719519350400.0,
      "grad_norm": 2.144556785493343,
      "language_loss": 0.93508077,
      "learning_rate": 3.9675025484671624e-06,
      "loss": 0.95744228,
      "num_input_tokens_seen": 30439620,
      "step": 1426,
      "time_per_iteration": 2.482816219329834
    },
    {
      "auxiliary_loss_clip": 0.01140865,
      "auxiliary_loss_mlp": 0.0106059,
      "balance_loss_clip": 1.05449331,
      "balance_loss_mlp": 1.03492403,
      "epoch": 0.08579588155719224,
      "flos": 17931563038080.0,
      "grad_norm": 2.688516678536283,
      "language_loss": 0.75635034,
      "learning_rate": 3.967432588494471e-06,
      "loss": 0.7783649,
      "num_input_tokens_seen": 30457300,
      "step": 1427,
      "time_per_iteration": 2.62381649017334
    },
    {
      "auxiliary_loss_clip": 0.01186955,
      "auxiliary_loss_mlp": 0.01054606,
      "balance_loss_clip": 1.05530071,
      "balance_loss_mlp": 1.03229034,
      "epoch": 0.08585600480986022,
      "flos": 16033324993920.0,
      "grad_norm": 3.0212352255102766,
      "language_loss": 0.82156587,
      "learning_rate": 3.96736255391654e-06,
      "loss": 0.84398156,
      "num_input_tokens_seen": 30471580,
      "step": 1428,
      "time_per_iteration": 3.8691062927246094
    },
    {
      "auxiliary_loss_clip": 0.01173836,
      "auxiliary_loss_mlp": 0.01059837,
      "balance_loss_clip": 1.05977106,
      "balance_loss_mlp": 1.03594768,
      "epoch": 0.08591612806252819,
      "flos": 28657433404800.0,
      "grad_norm": 2.2942504496676923,
      "language_loss": 0.79669249,
      "learning_rate": 3.967292444736023e-06,
      "loss": 0.81902921,
      "num_input_tokens_seen": 30492720,
      "step": 1429,
      "time_per_iteration": 2.5814480781555176
    },
    {
      "auxiliary_loss_clip": 0.01165681,
      "auxiliary_loss_mlp": 0.01061517,
      "balance_loss_clip": 1.05703187,
      "balance_loss_mlp": 1.03847396,
      "epoch": 0.08597625131519615,
      "flos": 20959119659520.0,
      "grad_norm": 2.1062116865199294,
      "language_loss": 0.8821522,
      "learning_rate": 3.967222260955578e-06,
      "loss": 0.90442419,
      "num_input_tokens_seen": 30509535,
      "step": 1430,
      "time_per_iteration": 3.900022506713867
    },
    {
      "auxiliary_loss_clip": 0.01141174,
      "auxiliary_loss_mlp": 0.01070572,
      "balance_loss_clip": 1.05605721,
      "balance_loss_mlp": 1.04744518,
      "epoch": 0.08603637456786412,
      "flos": 23256360956160.0,
      "grad_norm": 1.648138255737804,
      "language_loss": 0.82049704,
      "learning_rate": 3.96715200257787e-06,
      "loss": 0.84261447,
      "num_input_tokens_seen": 30529490,
      "step": 1431,
      "time_per_iteration": 2.582465171813965
    },
    {
      "auxiliary_loss_clip": 0.01148417,
      "auxiliary_loss_mlp": 0.01056838,
      "balance_loss_clip": 1.05611229,
      "balance_loss_mlp": 1.03285336,
      "epoch": 0.0860964978205321,
      "flos": 28694170039680.0,
      "grad_norm": 1.5753224400325532,
      "language_loss": 0.7782678,
      "learning_rate": 3.967081669605559e-06,
      "loss": 0.80032033,
      "num_input_tokens_seen": 30550205,
      "step": 1432,
      "time_per_iteration": 2.6299450397491455
    },
    {
      "auxiliary_loss_clip": 0.01164173,
      "auxiliary_loss_mlp": 0.0105869,
      "balance_loss_clip": 1.05182111,
      "balance_loss_mlp": 1.03462112,
      "epoch": 0.08615662107320006,
      "flos": 19318397195520.0,
      "grad_norm": 2.122380661239903,
      "language_loss": 0.72660923,
      "learning_rate": 3.967011262041315e-06,
      "loss": 0.74883783,
      "num_input_tokens_seen": 30568830,
      "step": 1433,
      "time_per_iteration": 2.554478406906128
    },
    {
      "auxiliary_loss_clip": 0.01149844,
      "auxiliary_loss_mlp": 0.00800064,
      "balance_loss_clip": 1.05399513,
      "balance_loss_mlp": 1.03091598,
      "epoch": 0.08621674432586802,
      "flos": 15851688894720.0,
      "grad_norm": 2.7851188848530937,
      "language_loss": 0.85714734,
      "learning_rate": 3.9669407798878065e-06,
      "loss": 0.87664646,
      "num_input_tokens_seen": 30585730,
      "step": 1434,
      "time_per_iteration": 2.5908522605895996
    },
    {
      "auxiliary_loss_clip": 0.01167269,
      "auxiliary_loss_mlp": 0.01054149,
      "balance_loss_clip": 1.05405867,
      "balance_loss_mlp": 1.03080797,
      "epoch": 0.086276867578536,
      "flos": 14100648785280.0,
      "grad_norm": 2.148692344025134,
      "language_loss": 0.78694725,
      "learning_rate": 3.966870223147707e-06,
      "loss": 0.80916142,
      "num_input_tokens_seen": 30603180,
      "step": 1435,
      "time_per_iteration": 2.5363612174987793
    },
    {
      "auxiliary_loss_clip": 0.01074257,
      "auxiliary_loss_mlp": 0.0100768,
      "balance_loss_clip": 1.05984211,
      "balance_loss_mlp": 1.00405622,
      "epoch": 0.08633699083120397,
      "flos": 70184857772160.0,
      "grad_norm": 0.8873535031849531,
      "language_loss": 0.57923287,
      "learning_rate": 3.96679959182369e-06,
      "loss": 0.60005224,
      "num_input_tokens_seen": 30668895,
      "step": 1436,
      "time_per_iteration": 3.2980265617370605
    },
    {
      "auxiliary_loss_clip": 0.01172773,
      "auxiliary_loss_mlp": 0.01052721,
      "balance_loss_clip": 1.05625916,
      "balance_loss_mlp": 1.02721071,
      "epoch": 0.08639711408387193,
      "flos": 30298874140800.0,
      "grad_norm": 2.3282229442786573,
      "language_loss": 0.69479793,
      "learning_rate": 3.966728885918437e-06,
      "loss": 0.71705294,
      "num_input_tokens_seen": 30688955,
      "step": 1437,
      "time_per_iteration": 2.596550941467285
    },
    {
      "auxiliary_loss_clip": 0.01119571,
      "auxiliary_loss_mlp": 0.0105307,
      "balance_loss_clip": 1.05314267,
      "balance_loss_mlp": 1.03040862,
      "epoch": 0.08645723733653991,
      "flos": 20297680663680.0,
      "grad_norm": 2.026902945362301,
      "language_loss": 0.723836,
      "learning_rate": 3.966658105434627e-06,
      "loss": 0.74556243,
      "num_input_tokens_seen": 30706095,
      "step": 1438,
      "time_per_iteration": 2.6453452110290527
    },
    {
      "auxiliary_loss_clip": 0.011752,
      "auxiliary_loss_mlp": 0.0104966,
      "balance_loss_clip": 1.05709982,
      "balance_loss_mlp": 1.02629542,
      "epoch": 0.08651736058920788,
      "flos": 32890583134080.0,
      "grad_norm": 1.6839047953935795,
      "language_loss": 0.64292943,
      "learning_rate": 3.966587250374945e-06,
      "loss": 0.66517806,
      "num_input_tokens_seen": 30729025,
      "step": 1439,
      "time_per_iteration": 2.64375638961792
    },
    {
      "auxiliary_loss_clip": 0.0115247,
      "auxiliary_loss_mlp": 0.01055511,
      "balance_loss_clip": 1.05702353,
      "balance_loss_mlp": 1.03138304,
      "epoch": 0.08657748384187584,
      "flos": 22637368857600.0,
      "grad_norm": 1.9747424452714684,
      "language_loss": 0.8745411,
      "learning_rate": 3.966516320742077e-06,
      "loss": 0.89662087,
      "num_input_tokens_seen": 30746155,
      "step": 1440,
      "time_per_iteration": 2.5738162994384766
    },
    {
      "auxiliary_loss_clip": 0.01153302,
      "auxiliary_loss_mlp": 0.00793454,
      "balance_loss_clip": 1.0578903,
      "balance_loss_mlp": 1.01519179,
      "epoch": 0.08663760709454381,
      "flos": 23658380951040.0,
      "grad_norm": 2.3783300870478095,
      "language_loss": 0.83693051,
      "learning_rate": 3.9664453165387124e-06,
      "loss": 0.85639805,
      "num_input_tokens_seen": 30761410,
      "step": 1441,
      "time_per_iteration": 2.6063992977142334
    },
    {
      "auxiliary_loss_clip": 0.01085319,
      "auxiliary_loss_mlp": 0.0101356,
      "balance_loss_clip": 1.03012192,
      "balance_loss_mlp": 1.01041341,
      "epoch": 0.08669773034721179,
      "flos": 62686564911360.0,
      "grad_norm": 0.8530046741951038,
      "language_loss": 0.60457605,
      "learning_rate": 3.966374237767545e-06,
      "loss": 0.62556487,
      "num_input_tokens_seen": 30823010,
      "step": 1442,
      "time_per_iteration": 3.19811749458313
    },
    {
      "auxiliary_loss_clip": 0.0116451,
      "auxiliary_loss_mlp": 0.01051623,
      "balance_loss_clip": 1.05652547,
      "balance_loss_mlp": 1.02887809,
      "epoch": 0.08675785359987975,
      "flos": 20667489137280.0,
      "grad_norm": 2.4859857258956715,
      "language_loss": 0.79005504,
      "learning_rate": 3.96630308443127e-06,
      "loss": 0.8122164,
      "num_input_tokens_seen": 30841980,
      "step": 1443,
      "time_per_iteration": 2.5527591705322266
    },
    {
      "auxiliary_loss_clip": 0.01178351,
      "auxiliary_loss_mlp": 0.01049782,
      "balance_loss_clip": 1.05524683,
      "balance_loss_mlp": 1.02682245,
      "epoch": 0.08681797685254772,
      "flos": 26941118768640.0,
      "grad_norm": 1.551168130742296,
      "language_loss": 0.82787728,
      "learning_rate": 3.966231856532584e-06,
      "loss": 0.85015857,
      "num_input_tokens_seen": 30863280,
      "step": 1444,
      "time_per_iteration": 2.5685172080993652
    },
    {
      "auxiliary_loss_clip": 0.01193986,
      "auxiliary_loss_mlp": 0.0105288,
      "balance_loss_clip": 1.05847955,
      "balance_loss_mlp": 1.02998042,
      "epoch": 0.0868781001052157,
      "flos": 17712831168000.0,
      "grad_norm": 1.9912234202628565,
      "language_loss": 0.86965901,
      "learning_rate": 3.966160554074189e-06,
      "loss": 0.89212763,
      "num_input_tokens_seen": 30881710,
      "step": 1445,
      "time_per_iteration": 2.491346836090088
    },
    {
      "auxiliary_loss_clip": 0.01179781,
      "auxiliary_loss_mlp": 0.01054057,
      "balance_loss_clip": 1.06129634,
      "balance_loss_mlp": 1.03263497,
      "epoch": 0.08693822335788366,
      "flos": 19896522595200.0,
      "grad_norm": 1.9861155721774153,
      "language_loss": 0.82124597,
      "learning_rate": 3.96608917705879e-06,
      "loss": 0.84358436,
      "num_input_tokens_seen": 30900225,
      "step": 1446,
      "time_per_iteration": 2.5155255794525146
    },
    {
      "auxiliary_loss_clip": 0.01069396,
      "auxiliary_loss_mlp": 0.01004954,
      "balance_loss_clip": 1.03010166,
      "balance_loss_mlp": 1.00180721,
      "epoch": 0.08699834661055163,
      "flos": 67023747406080.0,
      "grad_norm": 0.723766240555001,
      "language_loss": 0.54819649,
      "learning_rate": 3.966017725489091e-06,
      "loss": 0.56894004,
      "num_input_tokens_seen": 30959580,
      "step": 1447,
      "time_per_iteration": 3.1502277851104736
    },
    {
      "auxiliary_loss_clip": 0.01150335,
      "auxiliary_loss_mlp": 0.01054432,
      "balance_loss_clip": 1.05708385,
      "balance_loss_mlp": 1.03229451,
      "epoch": 0.0870584698632196,
      "flos": 13480507451520.0,
      "grad_norm": 2.992249319629708,
      "language_loss": 0.8437885,
      "learning_rate": 3.965946199367804e-06,
      "loss": 0.86583614,
      "num_input_tokens_seen": 30976775,
      "step": 1448,
      "time_per_iteration": 2.551645040512085
    },
    {
      "auxiliary_loss_clip": 0.0119344,
      "auxiliary_loss_mlp": 0.01054923,
      "balance_loss_clip": 1.05936146,
      "balance_loss_mlp": 1.03276229,
      "epoch": 0.08711859311588757,
      "flos": 16107013745280.0,
      "grad_norm": 4.210181490187109,
      "language_loss": 0.80519062,
      "learning_rate": 3.965874598697638e-06,
      "loss": 0.82767427,
      "num_input_tokens_seen": 30990495,
      "step": 1449,
      "time_per_iteration": 2.451266050338745
    },
    {
      "auxiliary_loss_clip": 0.01137404,
      "auxiliary_loss_mlp": 0.01050557,
      "balance_loss_clip": 1.0571816,
      "balance_loss_mlp": 1.02809834,
      "epoch": 0.08717871636855554,
      "flos": 38472357928320.0,
      "grad_norm": 1.9595189644924056,
      "language_loss": 0.71343768,
      "learning_rate": 3.965802923481313e-06,
      "loss": 0.73531723,
      "num_input_tokens_seen": 31014080,
      "step": 1450,
      "time_per_iteration": 2.7288060188293457
    },
    {
      "auxiliary_loss_clip": 0.0112499,
      "auxiliary_loss_mlp": 0.01057577,
      "balance_loss_clip": 1.05714262,
      "balance_loss_mlp": 1.0351181,
      "epoch": 0.0872388396212235,
      "flos": 17600574188160.0,
      "grad_norm": 2.1332848560021285,
      "language_loss": 0.83552158,
      "learning_rate": 3.965731173721542e-06,
      "loss": 0.85734719,
      "num_input_tokens_seen": 31031210,
      "step": 1451,
      "time_per_iteration": 2.5917136669158936
    },
    {
      "auxiliary_loss_clip": 0.01136661,
      "auxiliary_loss_mlp": 0.00793719,
      "balance_loss_clip": 1.05699229,
      "balance_loss_mlp": 1.02087259,
      "epoch": 0.08729896287389148,
      "flos": 25259385951360.0,
      "grad_norm": 2.0310730835785735,
      "language_loss": 0.74943852,
      "learning_rate": 3.965659349421049e-06,
      "loss": 0.76874232,
      "num_input_tokens_seen": 31049710,
      "step": 1452,
      "time_per_iteration": 2.6472387313842773
    },
    {
      "auxiliary_loss_clip": 0.0115545,
      "auxiliary_loss_mlp": 0.01056369,
      "balance_loss_clip": 1.055058,
      "balance_loss_mlp": 1.03185964,
      "epoch": 0.08735908612655945,
      "flos": 15632454234240.0,
      "grad_norm": 3.8617338582513687,
      "language_loss": 0.80128908,
      "learning_rate": 3.965587450582556e-06,
      "loss": 0.82340729,
      "num_input_tokens_seen": 31066160,
      "step": 1453,
      "time_per_iteration": 2.5252108573913574
    },
    {
      "auxiliary_loss_clip": 0.01163223,
      "auxiliary_loss_mlp": 0.01055695,
      "balance_loss_clip": 1.05562103,
      "balance_loss_mlp": 1.03249693,
      "epoch": 0.08741920937922741,
      "flos": 20339660684160.0,
      "grad_norm": 2.0144094220252486,
      "language_loss": 0.71426308,
      "learning_rate": 3.96551547720879e-06,
      "loss": 0.73645222,
      "num_input_tokens_seen": 31085270,
      "step": 1454,
      "time_per_iteration": 2.5762574672698975
    },
    {
      "auxiliary_loss_clip": 0.0107413,
      "auxiliary_loss_mlp": 0.01007969,
      "balance_loss_clip": 1.02957368,
      "balance_loss_mlp": 1.00465477,
      "epoch": 0.08747933263189539,
      "flos": 62819795433600.0,
      "grad_norm": 0.7765726661350218,
      "language_loss": 0.58604121,
      "learning_rate": 3.96544342930248e-06,
      "loss": 0.60686219,
      "num_input_tokens_seen": 31148445,
      "step": 1455,
      "time_per_iteration": 3.138570785522461
    },
    {
      "auxiliary_loss_clip": 0.01189513,
      "auxiliary_loss_mlp": 0.01055131,
      "balance_loss_clip": 1.05740499,
      "balance_loss_mlp": 1.03243303,
      "epoch": 0.08753945588456336,
      "flos": 33035877648000.0,
      "grad_norm": 2.031102427098728,
      "language_loss": 0.775038,
      "learning_rate": 3.965371306866359e-06,
      "loss": 0.7974844,
      "num_input_tokens_seen": 31168770,
      "step": 1456,
      "time_per_iteration": 2.6047418117523193
    },
    {
      "auxiliary_loss_clip": 0.01127646,
      "auxiliary_loss_mlp": 0.01051387,
      "balance_loss_clip": 1.05144501,
      "balance_loss_mlp": 1.02870154,
      "epoch": 0.08759957913723132,
      "flos": 35547182046720.0,
      "grad_norm": 2.430167652290147,
      "language_loss": 0.72266495,
      "learning_rate": 3.96529910990316e-06,
      "loss": 0.74445522,
      "num_input_tokens_seen": 31189270,
      "step": 1457,
      "time_per_iteration": 2.7274932861328125
    },
    {
      "auxiliary_loss_clip": 0.01171924,
      "auxiliary_loss_mlp": 0.01046058,
      "balance_loss_clip": 1.05369461,
      "balance_loss_mlp": 1.02438569,
      "epoch": 0.0876597023898993,
      "flos": 23911120022400.0,
      "grad_norm": 1.6206134596293675,
      "language_loss": 0.86420918,
      "learning_rate": 3.965226838415622e-06,
      "loss": 0.88638896,
      "num_input_tokens_seen": 31210385,
      "step": 1458,
      "time_per_iteration": 2.5647056102752686
    },
    {
      "auxiliary_loss_clip": 0.01167594,
      "auxiliary_loss_mlp": 0.01054817,
      "balance_loss_clip": 1.05929649,
      "balance_loss_mlp": 1.03252482,
      "epoch": 0.08771982564256726,
      "flos": 18114025150080.0,
      "grad_norm": 2.0684349009193315,
      "language_loss": 0.80577344,
      "learning_rate": 3.965154492406486e-06,
      "loss": 0.82799757,
      "num_input_tokens_seen": 31229745,
      "step": 1459,
      "time_per_iteration": 2.534670829772949
    },
    {
      "auxiliary_loss_clip": 0.0112896,
      "auxiliary_loss_mlp": 0.01052615,
      "balance_loss_clip": 1.0552963,
      "balance_loss_mlp": 1.02889204,
      "epoch": 0.08777994889523523,
      "flos": 17712005155200.0,
      "grad_norm": 2.353208371814136,
      "language_loss": 0.84149063,
      "learning_rate": 3.9650820718784945e-06,
      "loss": 0.8633064,
      "num_input_tokens_seen": 31248280,
      "step": 1460,
      "time_per_iteration": 4.252511024475098
    },
    {
      "auxiliary_loss_clip": 0.01173025,
      "auxiliary_loss_mlp": 0.01052468,
      "balance_loss_clip": 1.05379105,
      "balance_loss_mlp": 1.03117681,
      "epoch": 0.0878400721479032,
      "flos": 12819930382080.0,
      "grad_norm": 4.8754693487306255,
      "language_loss": 0.80502784,
      "learning_rate": 3.965009576834394e-06,
      "loss": 0.82728279,
      "num_input_tokens_seen": 31262190,
      "step": 1461,
      "time_per_iteration": 2.5258021354675293
    },
    {
      "auxiliary_loss_clip": 0.01185506,
      "auxiliary_loss_mlp": 0.01058105,
      "balance_loss_clip": 1.08982849,
      "balance_loss_mlp": 1.03608704,
      "epoch": 0.08790019540057117,
      "flos": 26392690938240.0,
      "grad_norm": 1.7226847751673817,
      "language_loss": 0.76390898,
      "learning_rate": 3.964937007276932e-06,
      "loss": 0.78634512,
      "num_input_tokens_seen": 31283690,
      "step": 1462,
      "time_per_iteration": 4.035697937011719
    },
    {
      "auxiliary_loss_clip": 0.01172076,
      "auxiliary_loss_mlp": 0.01057392,
      "balance_loss_clip": 1.05871022,
      "balance_loss_mlp": 1.03374052,
      "epoch": 0.08796031865323914,
      "flos": 19134031662720.0,
      "grad_norm": 2.0945109168234106,
      "language_loss": 0.7455709,
      "learning_rate": 3.9648643632088634e-06,
      "loss": 0.7678656,
      "num_input_tokens_seen": 31302505,
      "step": 1463,
      "time_per_iteration": 2.545161485671997
    },
    {
      "auxiliary_loss_clip": 0.01185757,
      "auxiliary_loss_mlp": 0.01057302,
      "balance_loss_clip": 1.05653811,
      "balance_loss_mlp": 1.03329337,
      "epoch": 0.0880204419059071,
      "flos": 26064287867520.0,
      "grad_norm": 1.9194947950212204,
      "language_loss": 0.83017516,
      "learning_rate": 3.964791644632941e-06,
      "loss": 0.85260576,
      "num_input_tokens_seen": 31323070,
      "step": 1464,
      "time_per_iteration": 2.565160036087036
    },
    {
      "auxiliary_loss_clip": 0.01167899,
      "auxiliary_loss_mlp": 0.01063378,
      "balance_loss_clip": 1.05491424,
      "balance_loss_mlp": 1.04096639,
      "epoch": 0.08808056515857508,
      "flos": 22377842115840.0,
      "grad_norm": 2.05675057302645,
      "language_loss": 0.78117168,
      "learning_rate": 3.964718851551923e-06,
      "loss": 0.80348444,
      "num_input_tokens_seen": 31341880,
      "step": 1465,
      "time_per_iteration": 2.5701022148132324
    },
    {
      "auxiliary_loss_clip": 0.0119436,
      "auxiliary_loss_mlp": 0.01060358,
      "balance_loss_clip": 1.05879271,
      "balance_loss_mlp": 1.03918672,
      "epoch": 0.08814068841124305,
      "flos": 23185293897600.0,
      "grad_norm": 2.1301398677264185,
      "language_loss": 0.85075903,
      "learning_rate": 3.9646459839685675e-06,
      "loss": 0.87330616,
      "num_input_tokens_seen": 31361995,
      "step": 1466,
      "time_per_iteration": 3.976207733154297
    },
    {
      "auxiliary_loss_clip": 0.01119677,
      "auxiliary_loss_mlp": 0.0079117,
      "balance_loss_clip": 1.05345154,
      "balance_loss_mlp": 1.01382589,
      "epoch": 0.08820081166391101,
      "flos": 25155281358720.0,
      "grad_norm": 2.1786291209668596,
      "language_loss": 0.83956236,
      "learning_rate": 3.964573041885641e-06,
      "loss": 0.85867083,
      "num_input_tokens_seen": 31381515,
      "step": 1467,
      "time_per_iteration": 2.6584904193878174
    },
    {
      "auxiliary_loss_clip": 0.01173276,
      "auxiliary_loss_mlp": 0.01058601,
      "balance_loss_clip": 1.05626035,
      "balance_loss_mlp": 1.03536677,
      "epoch": 0.08826093491657899,
      "flos": 22231685675520.0,
      "grad_norm": 1.793268306223792,
      "language_loss": 0.75704187,
      "learning_rate": 3.964500025305907e-06,
      "loss": 0.77936065,
      "num_input_tokens_seen": 31400345,
      "step": 1468,
      "time_per_iteration": 2.52948260307312
    },
    {
      "auxiliary_loss_clip": 0.01173741,
      "auxiliary_loss_mlp": 0.01058029,
      "balance_loss_clip": 1.05695343,
      "balance_loss_mlp": 1.03713155,
      "epoch": 0.08832105816924696,
      "flos": 22126826897280.0,
      "grad_norm": 1.8559897663152325,
      "language_loss": 0.8039223,
      "learning_rate": 3.9644269342321355e-06,
      "loss": 0.82623994,
      "num_input_tokens_seen": 31419620,
      "step": 1469,
      "time_per_iteration": 3.997688055038452
    },
    {
      "auxiliary_loss_clip": 0.01192303,
      "auxiliary_loss_mlp": 0.01060648,
      "balance_loss_clip": 1.0578227,
      "balance_loss_mlp": 1.03809404,
      "epoch": 0.08838118142191492,
      "flos": 17566495159680.0,
      "grad_norm": 2.1566482021590314,
      "language_loss": 0.77848309,
      "learning_rate": 3.9643537686670974e-06,
      "loss": 0.80101264,
      "num_input_tokens_seen": 31437970,
      "step": 1470,
      "time_per_iteration": 2.4564836025238037
    },
    {
      "auxiliary_loss_clip": 0.01184202,
      "auxiliary_loss_mlp": 0.01066918,
      "balance_loss_clip": 1.05512452,
      "balance_loss_mlp": 1.04397058,
      "epoch": 0.0884413046745829,
      "flos": 20777196251520.0,
      "grad_norm": 1.8652659549643773,
      "language_loss": 0.84152907,
      "learning_rate": 3.964280528613569e-06,
      "loss": 0.86404026,
      "num_input_tokens_seen": 31457040,
      "step": 1471,
      "time_per_iteration": 2.5191969871520996
    },
    {
      "auxiliary_loss_clip": 0.01160087,
      "auxiliary_loss_mlp": 0.01057659,
      "balance_loss_clip": 1.06921208,
      "balance_loss_mlp": 1.0378468,
      "epoch": 0.08850142792725087,
      "flos": 22125462180480.0,
      "grad_norm": 1.8061843909292565,
      "language_loss": 0.83565956,
      "learning_rate": 3.964207214074324e-06,
      "loss": 0.85783696,
      "num_input_tokens_seen": 31477520,
      "step": 1472,
      "time_per_iteration": 2.609419107437134
    },
    {
      "auxiliary_loss_clip": 0.01167856,
      "auxiliary_loss_mlp": 0.0105972,
      "balance_loss_clip": 1.05550039,
      "balance_loss_mlp": 1.03707051,
      "epoch": 0.08856155117991883,
      "flos": 22418744728320.0,
      "grad_norm": 2.3285326427946127,
      "language_loss": 0.82838386,
      "learning_rate": 3.964133825052146e-06,
      "loss": 0.85065961,
      "num_input_tokens_seen": 31495575,
      "step": 1473,
      "time_per_iteration": 2.5579936504364014
    },
    {
      "auxiliary_loss_clip": 0.01122504,
      "auxiliary_loss_mlp": 0.01060368,
      "balance_loss_clip": 1.05263925,
      "balance_loss_mlp": 1.03950691,
      "epoch": 0.0886216744325868,
      "flos": 29937002572800.0,
      "grad_norm": 1.6012922988826765,
      "language_loss": 0.78777742,
      "learning_rate": 3.964060361549816e-06,
      "loss": 0.80960613,
      "num_input_tokens_seen": 31520020,
      "step": 1474,
      "time_per_iteration": 2.721236228942871
    },
    {
      "auxiliary_loss_clip": 0.01152612,
      "auxiliary_loss_mlp": 0.01068983,
      "balance_loss_clip": 1.07052612,
      "balance_loss_mlp": 1.04319835,
      "epoch": 0.08868179768525478,
      "flos": 23982833525760.0,
      "grad_norm": 1.8004191272671695,
      "language_loss": 0.78462052,
      "learning_rate": 3.963986823570121e-06,
      "loss": 0.80683649,
      "num_input_tokens_seen": 31539265,
      "step": 1475,
      "time_per_iteration": 2.60508394241333
    },
    {
      "auxiliary_loss_clip": 0.01187234,
      "auxiliary_loss_mlp": 0.01050283,
      "balance_loss_clip": 1.05627513,
      "balance_loss_mlp": 1.0279547,
      "epoch": 0.08874192093792274,
      "flos": 43177553216640.0,
      "grad_norm": 1.6384832172061847,
      "language_loss": 0.7396518,
      "learning_rate": 3.963913211115848e-06,
      "loss": 0.76202691,
      "num_input_tokens_seen": 31563425,
      "step": 1476,
      "time_per_iteration": 2.6666409969329834
    },
    {
      "auxiliary_loss_clip": 0.01165178,
      "auxiliary_loss_mlp": 0.01057523,
      "balance_loss_clip": 1.0550977,
      "balance_loss_mlp": 1.03436053,
      "epoch": 0.0888020441905907,
      "flos": 32852445868800.0,
      "grad_norm": 1.6086390125562438,
      "language_loss": 0.74552554,
      "learning_rate": 3.9638395241897895e-06,
      "loss": 0.76775259,
      "num_input_tokens_seen": 31584525,
      "step": 1477,
      "time_per_iteration": 2.653693675994873
    },
    {
      "auxiliary_loss_clip": 0.01184569,
      "auxiliary_loss_mlp": 0.0105138,
      "balance_loss_clip": 1.05460894,
      "balance_loss_mlp": 1.02784801,
      "epoch": 0.08886216744325869,
      "flos": 23149347361920.0,
      "grad_norm": 1.8862392302905042,
      "language_loss": 0.86755311,
      "learning_rate": 3.963765762794739e-06,
      "loss": 0.88991261,
      "num_input_tokens_seen": 31603325,
      "step": 1478,
      "time_per_iteration": 2.481678009033203
    },
    {
      "auxiliary_loss_clip": 0.01172388,
      "auxiliary_loss_mlp": 0.01055656,
      "balance_loss_clip": 1.0547328,
      "balance_loss_mlp": 1.03428185,
      "epoch": 0.08892229069592665,
      "flos": 23331593992320.0,
      "grad_norm": 1.6796737396519537,
      "language_loss": 0.77599698,
      "learning_rate": 3.963691926933495e-06,
      "loss": 0.79827738,
      "num_input_tokens_seen": 31624820,
      "step": 1479,
      "time_per_iteration": 2.5577874183654785
    },
    {
      "auxiliary_loss_clip": 0.01169502,
      "auxiliary_loss_mlp": 0.01051735,
      "balance_loss_clip": 1.06883597,
      "balance_loss_mlp": 1.02834582,
      "epoch": 0.08898241394859462,
      "flos": 26213784272640.0,
      "grad_norm": 2.6005372299476988,
      "language_loss": 0.7806316,
      "learning_rate": 3.9636180166088555e-06,
      "loss": 0.80284399,
      "num_input_tokens_seen": 31646080,
      "step": 1480,
      "time_per_iteration": 2.5667178630828857
    },
    {
      "auxiliary_loss_clip": 0.01176094,
      "auxiliary_loss_mlp": 0.01061731,
      "balance_loss_clip": 1.05491555,
      "balance_loss_mlp": 1.03747165,
      "epoch": 0.0890425372012626,
      "flos": 23550613171200.0,
      "grad_norm": 1.7230803744806364,
      "language_loss": 0.66436011,
      "learning_rate": 3.963544031823624e-06,
      "loss": 0.68673837,
      "num_input_tokens_seen": 31665770,
      "step": 1481,
      "time_per_iteration": 2.5299625396728516
    },
    {
      "auxiliary_loss_clip": 0.01137769,
      "auxiliary_loss_mlp": 0.01052177,
      "balance_loss_clip": 1.05413735,
      "balance_loss_mlp": 1.03038597,
      "epoch": 0.08910266045393056,
      "flos": 23002795872000.0,
      "grad_norm": 1.8883646877639868,
      "language_loss": 0.96223301,
      "learning_rate": 3.9634699725806065e-06,
      "loss": 0.98413253,
      "num_input_tokens_seen": 31683805,
      "step": 1482,
      "time_per_iteration": 2.600769519805908
    },
    {
      "auxiliary_loss_clip": 0.01152973,
      "auxiliary_loss_mlp": 0.01061518,
      "balance_loss_clip": 1.05664361,
      "balance_loss_mlp": 1.03823686,
      "epoch": 0.08916278370659853,
      "flos": 31936508035200.0,
      "grad_norm": 2.965354733436817,
      "language_loss": 0.7824477,
      "learning_rate": 3.96339583888261e-06,
      "loss": 0.80459261,
      "num_input_tokens_seen": 31704630,
      "step": 1483,
      "time_per_iteration": 2.6356382369995117
    },
    {
      "auxiliary_loss_clip": 0.01167338,
      "auxiliary_loss_mlp": 0.0107889,
      "balance_loss_clip": 1.05649185,
      "balance_loss_mlp": 1.0557636,
      "epoch": 0.08922290695926649,
      "flos": 17530404969600.0,
      "grad_norm": 2.775592602752793,
      "language_loss": 0.85460865,
      "learning_rate": 3.963321630732448e-06,
      "loss": 0.87707096,
      "num_input_tokens_seen": 31723255,
      "step": 1484,
      "time_per_iteration": 2.521491527557373
    },
    {
      "auxiliary_loss_clip": 0.01195185,
      "auxiliary_loss_mlp": 0.01061755,
      "balance_loss_clip": 1.06053257,
      "balance_loss_mlp": 1.03822339,
      "epoch": 0.08928303021193447,
      "flos": 32125075459200.0,
      "grad_norm": 1.7776800892137086,
      "language_loss": 0.80499876,
      "learning_rate": 3.963247348132932e-06,
      "loss": 0.82756811,
      "num_input_tokens_seen": 31747045,
      "step": 1485,
      "time_per_iteration": 2.5922036170959473
    },
    {
      "auxiliary_loss_clip": 0.01172634,
      "auxiliary_loss_mlp": 0.01061864,
      "balance_loss_clip": 1.05730295,
      "balance_loss_mlp": 1.0383085,
      "epoch": 0.08934315346460243,
      "flos": 22125210785280.0,
      "grad_norm": 2.52907113862076,
      "language_loss": 0.82853043,
      "learning_rate": 3.96317299108688e-06,
      "loss": 0.85087538,
      "num_input_tokens_seen": 31766615,
      "step": 1486,
      "time_per_iteration": 2.5227699279785156
    },
    {
      "auxiliary_loss_clip": 0.01144919,
      "auxiliary_loss_mlp": 0.0106884,
      "balance_loss_clip": 1.05871999,
      "balance_loss_mlp": 1.04597604,
      "epoch": 0.0894032767172704,
      "flos": 22565583527040.0,
      "grad_norm": 3.645222599128206,
      "language_loss": 0.76581693,
      "learning_rate": 3.963098559597111e-06,
      "loss": 0.78795457,
      "num_input_tokens_seen": 31785855,
      "step": 1487,
      "time_per_iteration": 2.585028886795044
    },
    {
      "auxiliary_loss_clip": 0.01157187,
      "auxiliary_loss_mlp": 0.01062588,
      "balance_loss_clip": 1.05447996,
      "balance_loss_mlp": 1.0380547,
      "epoch": 0.08946339996993838,
      "flos": 20193396503040.0,
      "grad_norm": 2.6502370406749445,
      "language_loss": 0.83103651,
      "learning_rate": 3.963024053666449e-06,
      "loss": 0.85323423,
      "num_input_tokens_seen": 31804210,
      "step": 1488,
      "time_per_iteration": 2.5531671047210693
    },
    {
      "auxiliary_loss_clip": 0.01172286,
      "auxiliary_loss_mlp": 0.01049049,
      "balance_loss_clip": 1.0566386,
      "balance_loss_mlp": 1.02693582,
      "epoch": 0.08952352322260634,
      "flos": 48360181104000.0,
      "grad_norm": 1.7693566292890348,
      "language_loss": 0.71657169,
      "learning_rate": 3.962949473297718e-06,
      "loss": 0.73878503,
      "num_input_tokens_seen": 31826150,
      "step": 1489,
      "time_per_iteration": 2.748378276824951
    },
    {
      "auxiliary_loss_clip": 0.01149268,
      "auxiliary_loss_mlp": 0.0105354,
      "balance_loss_clip": 1.05275655,
      "balance_loss_mlp": 1.03054488,
      "epoch": 0.08958364647527431,
      "flos": 31793081028480.0,
      "grad_norm": 2.0582958044499793,
      "language_loss": 0.89333296,
      "learning_rate": 3.962874818493745e-06,
      "loss": 0.91536105,
      "num_input_tokens_seen": 31848060,
      "step": 1490,
      "time_per_iteration": 2.7027411460876465
    },
    {
      "auxiliary_loss_clip": 0.01184481,
      "auxiliary_loss_mlp": 0.01065339,
      "balance_loss_clip": 1.05729842,
      "balance_loss_mlp": 1.04239166,
      "epoch": 0.08964376972794229,
      "flos": 23368186972800.0,
      "grad_norm": 2.5118489442322662,
      "language_loss": 0.74041879,
      "learning_rate": 3.9628000892573635e-06,
      "loss": 0.76291704,
      "num_input_tokens_seen": 31870040,
      "step": 1491,
      "time_per_iteration": 2.5703282356262207
    },
    {
      "auxiliary_loss_clip": 0.0118967,
      "auxiliary_loss_mlp": 0.00806942,
      "balance_loss_clip": 1.05985188,
      "balance_loss_mlp": 1.04633868,
      "epoch": 0.08970389298061025,
      "flos": 23294785530240.0,
      "grad_norm": 1.8254582310338925,
      "language_loss": 0.76950085,
      "learning_rate": 3.9627252855914055e-06,
      "loss": 0.78946698,
      "num_input_tokens_seen": 31890400,
      "step": 1492,
      "time_per_iteration": 2.514705181121826
    },
    {
      "auxiliary_loss_clip": 0.01186607,
      "auxiliary_loss_mlp": 0.0105474,
      "balance_loss_clip": 1.05934525,
      "balance_loss_mlp": 1.03261447,
      "epoch": 0.08976401623327822,
      "flos": 33761703772800.0,
      "grad_norm": 2.106506929682526,
      "language_loss": 0.71034497,
      "learning_rate": 3.962650407498707e-06,
      "loss": 0.7327584,
      "num_input_tokens_seen": 31913435,
      "step": 1493,
      "time_per_iteration": 2.6029624938964844
    },
    {
      "auxiliary_loss_clip": 0.01188673,
      "auxiliary_loss_mlp": 0.01056902,
      "balance_loss_clip": 1.05794764,
      "balance_loss_mlp": 1.03375113,
      "epoch": 0.08982413948594618,
      "flos": 23911335504000.0,
      "grad_norm": 2.076088876937707,
      "language_loss": 0.86863351,
      "learning_rate": 3.962575454982109e-06,
      "loss": 0.8910892,
      "num_input_tokens_seen": 31932435,
      "step": 1494,
      "time_per_iteration": 2.508139133453369
    },
    {
      "auxiliary_loss_clip": 0.01086697,
      "auxiliary_loss_mlp": 0.01063276,
      "balance_loss_clip": 1.05280256,
      "balance_loss_mlp": 1.03914821,
      "epoch": 0.08988426273861416,
      "flos": 16837544551680.0,
      "grad_norm": 1.8630122950744596,
      "language_loss": 0.82844341,
      "learning_rate": 3.962500428044454e-06,
      "loss": 0.84994316,
      "num_input_tokens_seen": 31950125,
      "step": 1495,
      "time_per_iteration": 2.683941602706909
    },
    {
      "auxiliary_loss_clip": 0.01170157,
      "auxiliary_loss_mlp": 0.01054693,
      "balance_loss_clip": 1.05918872,
      "balance_loss_mlp": 1.0321033,
      "epoch": 0.08994438599128213,
      "flos": 14793365548800.0,
      "grad_norm": 6.602765023709375,
      "language_loss": 0.701684,
      "learning_rate": 3.962425326688585e-06,
      "loss": 0.7239325,
      "num_input_tokens_seen": 31968050,
      "step": 1496,
      "time_per_iteration": 2.5291266441345215
    },
    {
      "auxiliary_loss_clip": 0.01162834,
      "auxiliary_loss_mlp": 0.01047548,
      "balance_loss_clip": 1.05468535,
      "balance_loss_mlp": 1.02693689,
      "epoch": 0.09000450924395009,
      "flos": 17384320356480.0,
      "grad_norm": 1.5963090340107997,
      "language_loss": 0.79752511,
      "learning_rate": 3.962350150917351e-06,
      "loss": 0.81962895,
      "num_input_tokens_seen": 31985675,
      "step": 1497,
      "time_per_iteration": 2.5172297954559326
    },
    {
      "auxiliary_loss_clip": 0.01127397,
      "auxiliary_loss_mlp": 0.0105524,
      "balance_loss_clip": 1.05693519,
      "balance_loss_mlp": 1.03193474,
      "epoch": 0.09006463249661807,
      "flos": 24280317964800.0,
      "grad_norm": 2.270428470659929,
      "language_loss": 0.82687432,
      "learning_rate": 3.9622749007336035e-06,
      "loss": 0.84870064,
      "num_input_tokens_seen": 32005180,
      "step": 1498,
      "time_per_iteration": 4.057194471359253
    },
    {
      "auxiliary_loss_clip": 0.01168836,
      "auxiliary_loss_mlp": 0.01060404,
      "balance_loss_clip": 1.05768895,
      "balance_loss_mlp": 1.03855276,
      "epoch": 0.09012475574928604,
      "flos": 13661928069120.0,
      "grad_norm": 4.0742591383826285,
      "language_loss": 0.79041547,
      "learning_rate": 3.962199576140195e-06,
      "loss": 0.8127079,
      "num_input_tokens_seen": 32022970,
      "step": 1499,
      "time_per_iteration": 2.527783155441284
    },
    {
      "auxiliary_loss_clip": 0.01161343,
      "auxiliary_loss_mlp": 0.00787912,
      "balance_loss_clip": 1.05504107,
      "balance_loss_mlp": 1.00948477,
      "epoch": 0.090184879001954,
      "flos": 23327751237120.0,
      "grad_norm": 1.591349316446943,
      "language_loss": 0.92961919,
      "learning_rate": 3.962124177139981e-06,
      "loss": 0.9491117,
      "num_input_tokens_seen": 32043055,
      "step": 1500,
      "time_per_iteration": 2.5245378017425537
    },
    {
      "auxiliary_loss_clip": 0.01148352,
      "auxiliary_loss_mlp": 0.01056684,
      "balance_loss_clip": 1.05602717,
      "balance_loss_mlp": 1.03280616,
      "epoch": 0.09024500225462198,
      "flos": 23002688131200.0,
      "grad_norm": 3.585268945874999,
      "language_loss": 0.74299836,
      "learning_rate": 3.962048703735822e-06,
      "loss": 0.76504868,
      "num_input_tokens_seen": 32061900,
      "step": 1501,
      "time_per_iteration": 3.9402801990509033
    },
    {
      "auxiliary_loss_clip": 0.01066696,
      "auxiliary_loss_mlp": 0.01016742,
      "balance_loss_clip": 1.04115081,
      "balance_loss_mlp": 1.01292706,
      "epoch": 0.09030512550728995,
      "flos": 62189203242240.0,
      "grad_norm": 0.7346539458841886,
      "language_loss": 0.58344895,
      "learning_rate": 3.96197315593058e-06,
      "loss": 0.60428333,
      "num_input_tokens_seen": 32122745,
      "step": 1502,
      "time_per_iteration": 3.178743839263916
    },
    {
      "auxiliary_loss_clip": 0.01148019,
      "auxiliary_loss_mlp": 0.01070823,
      "balance_loss_clip": 1.05187154,
      "balance_loss_mlp": 1.04710054,
      "epoch": 0.09036524875995791,
      "flos": 38800689171840.0,
      "grad_norm": 2.3388638137498536,
      "language_loss": 0.69822526,
      "learning_rate": 3.961897533727119e-06,
      "loss": 0.72041368,
      "num_input_tokens_seen": 32145125,
      "step": 1503,
      "time_per_iteration": 2.7136096954345703
    },
    {
      "auxiliary_loss_clip": 0.01136416,
      "auxiliary_loss_mlp": 0.01058915,
      "balance_loss_clip": 1.05259335,
      "balance_loss_mlp": 1.0368253,
      "epoch": 0.09042537201262588,
      "flos": 21690081429120.0,
      "grad_norm": 2.178340786075428,
      "language_loss": 0.86266947,
      "learning_rate": 3.961821837128306e-06,
      "loss": 0.88462281,
      "num_input_tokens_seen": 32166255,
      "step": 1504,
      "time_per_iteration": 2.636479139328003
    },
    {
      "auxiliary_loss_clip": 0.01145905,
      "auxiliary_loss_mlp": 0.01069652,
      "balance_loss_clip": 1.05633998,
      "balance_loss_mlp": 1.04401016,
      "epoch": 0.09048549526529386,
      "flos": 22267021680000.0,
      "grad_norm": 2.3572676148247154,
      "language_loss": 0.72550428,
      "learning_rate": 3.961746066137014e-06,
      "loss": 0.7476598,
      "num_input_tokens_seen": 32184010,
      "step": 1505,
      "time_per_iteration": 3.9850544929504395
    },
    {
      "auxiliary_loss_clip": 0.01134673,
      "auxiliary_loss_mlp": 0.01052933,
      "balance_loss_clip": 1.05248451,
      "balance_loss_mlp": 1.02993798,
      "epoch": 0.09054561851796182,
      "flos": 14610939350400.0,
      "grad_norm": 2.166874227117722,
      "language_loss": 0.80677038,
      "learning_rate": 3.961670220756114e-06,
      "loss": 0.82864648,
      "num_input_tokens_seen": 32201635,
      "step": 1506,
      "time_per_iteration": 2.5869858264923096
    },
    {
      "auxiliary_loss_clip": 0.01139233,
      "auxiliary_loss_mlp": 0.01052137,
      "balance_loss_clip": 1.05140781,
      "balance_loss_mlp": 1.0306797,
      "epoch": 0.09060574177062979,
      "flos": 27636169916160.0,
      "grad_norm": 1.9244742593211628,
      "language_loss": 0.76189566,
      "learning_rate": 3.961594300988482e-06,
      "loss": 0.78380936,
      "num_input_tokens_seen": 32221940,
      "step": 1507,
      "time_per_iteration": 4.141975402832031
    },
    {
      "auxiliary_loss_clip": 0.01058796,
      "auxiliary_loss_mlp": 0.01010573,
      "balance_loss_clip": 1.02549386,
      "balance_loss_mlp": 1.00661564,
      "epoch": 0.09066586502329776,
      "flos": 66085797513600.0,
      "grad_norm": 0.7301178857652109,
      "language_loss": 0.57632238,
      "learning_rate": 3.961518306836998e-06,
      "loss": 0.5970161,
      "num_input_tokens_seen": 32276495,
      "step": 1508,
      "time_per_iteration": 2.9964821338653564
    },
    {
      "auxiliary_loss_clip": 0.01161904,
      "auxiliary_loss_mlp": 0.01055329,
      "balance_loss_clip": 1.05364466,
      "balance_loss_mlp": 1.03254771,
      "epoch": 0.09072598827596573,
      "flos": 18916449027840.0,
      "grad_norm": 1.946227083421863,
      "language_loss": 0.85385394,
      "learning_rate": 3.961442238304543e-06,
      "loss": 0.87602627,
      "num_input_tokens_seen": 32294130,
      "step": 1509,
      "time_per_iteration": 2.5556492805480957
    },
    {
      "auxiliary_loss_clip": 0.01164025,
      "auxiliary_loss_mlp": 0.01064585,
      "balance_loss_clip": 1.0559442,
      "balance_loss_mlp": 1.04141045,
      "epoch": 0.0907861115286337,
      "flos": 24821742643200.0,
      "grad_norm": 2.2457622079531983,
      "language_loss": 0.8450948,
      "learning_rate": 3.961366095394002e-06,
      "loss": 0.86738086,
      "num_input_tokens_seen": 32313555,
      "step": 1510,
      "time_per_iteration": 2.6287920475006104
    },
    {
      "auxiliary_loss_clip": 0.01152068,
      "auxiliary_loss_mlp": 0.01059978,
      "balance_loss_clip": 1.05525887,
      "balance_loss_mlp": 1.03723311,
      "epoch": 0.09084623478130167,
      "flos": 21652842003840.0,
      "grad_norm": 2.172176150358652,
      "language_loss": 0.85411179,
      "learning_rate": 3.961289878108262e-06,
      "loss": 0.87623221,
      "num_input_tokens_seen": 32331430,
      "step": 1511,
      "time_per_iteration": 2.5501222610473633
    },
    {
      "auxiliary_loss_clip": 0.01144599,
      "auxiliary_loss_mlp": 0.01050694,
      "balance_loss_clip": 1.05410194,
      "balance_loss_mlp": 1.02878332,
      "epoch": 0.09090635803396964,
      "flos": 27639258485760.0,
      "grad_norm": 1.4427513685903741,
      "language_loss": 0.84912413,
      "learning_rate": 3.9612135864502135e-06,
      "loss": 0.87107706,
      "num_input_tokens_seen": 32353705,
      "step": 1512,
      "time_per_iteration": 2.6442928314208984
    },
    {
      "auxiliary_loss_clip": 0.01147996,
      "auxiliary_loss_mlp": 0.01048243,
      "balance_loss_clip": 1.05238926,
      "balance_loss_mlp": 1.02721453,
      "epoch": 0.0909664812866376,
      "flos": 17669127294720.0,
      "grad_norm": 2.3280006614011888,
      "language_loss": 0.87142682,
      "learning_rate": 3.961137220422749e-06,
      "loss": 0.89338923,
      "num_input_tokens_seen": 32370520,
      "step": 1513,
      "time_per_iteration": 2.563004732131958
    },
    {
      "auxiliary_loss_clip": 0.01168633,
      "auxiliary_loss_mlp": 0.01052458,
      "balance_loss_clip": 1.0570339,
      "balance_loss_mlp": 1.03153706,
      "epoch": 0.09102660453930557,
      "flos": 23951448017280.0,
      "grad_norm": 2.1163989995864743,
      "language_loss": 0.86534721,
      "learning_rate": 3.961060780028764e-06,
      "loss": 0.8875581,
      "num_input_tokens_seen": 32389105,
      "step": 1514,
      "time_per_iteration": 2.5247397422790527
    },
    {
      "auxiliary_loss_clip": 0.0112907,
      "auxiliary_loss_mlp": 0.01054426,
      "balance_loss_clip": 1.05622387,
      "balance_loss_mlp": 1.03426814,
      "epoch": 0.09108672779197355,
      "flos": 25812949426560.0,
      "grad_norm": 2.0012730821916427,
      "language_loss": 0.89925855,
      "learning_rate": 3.960984265271159e-06,
      "loss": 0.92109352,
      "num_input_tokens_seen": 32408065,
      "step": 1515,
      "time_per_iteration": 2.656680107116699
    },
    {
      "auxiliary_loss_clip": 0.01158434,
      "auxiliary_loss_mlp": 0.01052408,
      "balance_loss_clip": 1.05478299,
      "balance_loss_mlp": 1.03035426,
      "epoch": 0.09114685104464151,
      "flos": 29639482220160.0,
      "grad_norm": 2.124135782522118,
      "language_loss": 0.85358018,
      "learning_rate": 3.9609076761528335e-06,
      "loss": 0.87568855,
      "num_input_tokens_seen": 32427225,
      "step": 1516,
      "time_per_iteration": 2.5991122722625732
    },
    {
      "auxiliary_loss_clip": 0.01159755,
      "auxiliary_loss_mlp": 0.01055221,
      "balance_loss_clip": 1.05800748,
      "balance_loss_mlp": 1.03240478,
      "epoch": 0.09120697429730948,
      "flos": 33729635905920.0,
      "grad_norm": 2.05099865610027,
      "language_loss": 0.81417572,
      "learning_rate": 3.960831012676692e-06,
      "loss": 0.83632553,
      "num_input_tokens_seen": 32450510,
      "step": 1517,
      "time_per_iteration": 2.652472496032715
    },
    {
      "auxiliary_loss_clip": 0.01176377,
      "auxiliary_loss_mlp": 0.01069365,
      "balance_loss_clip": 1.05681431,
      "balance_loss_mlp": 1.04675126,
      "epoch": 0.09126709754997746,
      "flos": 18401381953920.0,
      "grad_norm": 1.650330387910903,
      "language_loss": 0.77734596,
      "learning_rate": 3.960754274845642e-06,
      "loss": 0.79980338,
      "num_input_tokens_seen": 32468425,
      "step": 1518,
      "time_per_iteration": 2.508869171142578
    },
    {
      "auxiliary_loss_clip": 0.01163094,
      "auxiliary_loss_mlp": 0.01061503,
      "balance_loss_clip": 1.05306602,
      "balance_loss_mlp": 1.03971124,
      "epoch": 0.09132722080264542,
      "flos": 22091957769600.0,
      "grad_norm": 1.8746944278411497,
      "language_loss": 0.86216724,
      "learning_rate": 3.960677462662594e-06,
      "loss": 0.88441324,
      "num_input_tokens_seen": 32487510,
      "step": 1519,
      "time_per_iteration": 2.5309321880340576
    },
    {
      "auxiliary_loss_clip": 0.01159858,
      "auxiliary_loss_mlp": 0.01054698,
      "balance_loss_clip": 1.05261195,
      "balance_loss_mlp": 1.03148818,
      "epoch": 0.09138734405531339,
      "flos": 21033131633280.0,
      "grad_norm": 2.0977056898079063,
      "language_loss": 0.73361552,
      "learning_rate": 3.96060057613046e-06,
      "loss": 0.75576109,
      "num_input_tokens_seen": 32507250,
      "step": 1520,
      "time_per_iteration": 2.5505049228668213
    },
    {
      "auxiliary_loss_clip": 0.01157324,
      "auxiliary_loss_mlp": 0.01062146,
      "balance_loss_clip": 1.05614948,
      "balance_loss_mlp": 1.03978252,
      "epoch": 0.09144746730798137,
      "flos": 20083940784000.0,
      "grad_norm": 2.5836459409947605,
      "language_loss": 0.84871328,
      "learning_rate": 3.960523615252156e-06,
      "loss": 0.87090796,
      "num_input_tokens_seen": 32526045,
      "step": 1521,
      "time_per_iteration": 2.531982898712158
    },
    {
      "auxiliary_loss_clip": 0.01116375,
      "auxiliary_loss_mlp": 0.01067953,
      "balance_loss_clip": 1.05180883,
      "balance_loss_mlp": 1.04421878,
      "epoch": 0.09150759056064933,
      "flos": 22778210085120.0,
      "grad_norm": 1.8460065580217138,
      "language_loss": 0.8434211,
      "learning_rate": 3.960446580030599e-06,
      "loss": 0.8652643,
      "num_input_tokens_seen": 32546575,
      "step": 1522,
      "time_per_iteration": 2.6617746353149414
    },
    {
      "auxiliary_loss_clip": 0.0117907,
      "auxiliary_loss_mlp": 0.01062049,
      "balance_loss_clip": 1.05365157,
      "balance_loss_mlp": 1.03992414,
      "epoch": 0.0915677138133173,
      "flos": 27564205017600.0,
      "grad_norm": 2.0010648389972676,
      "language_loss": 0.8102994,
      "learning_rate": 3.960369470468711e-06,
      "loss": 0.83271062,
      "num_input_tokens_seen": 32568795,
      "step": 1523,
      "time_per_iteration": 2.5278143882751465
    },
    {
      "auxiliary_loss_clip": 0.01157007,
      "auxiliary_loss_mlp": 0.00795537,
      "balance_loss_clip": 1.05854332,
      "balance_loss_mlp": 1.02340651,
      "epoch": 0.09162783706598528,
      "flos": 17674765729920.0,
      "grad_norm": 3.247803748906381,
      "language_loss": 0.74556911,
      "learning_rate": 3.960292286569418e-06,
      "loss": 0.76509464,
      "num_input_tokens_seen": 32587010,
      "step": 1524,
      "time_per_iteration": 2.533238649368286
    },
    {
      "auxiliary_loss_clip": 0.0113401,
      "auxiliary_loss_mlp": 0.01056572,
      "balance_loss_clip": 1.0523926,
      "balance_loss_mlp": 1.03416085,
      "epoch": 0.09168796031865324,
      "flos": 18478195188480.0,
      "grad_norm": 2.0255726076191825,
      "language_loss": 0.86106497,
      "learning_rate": 3.960215028335644e-06,
      "loss": 0.88297081,
      "num_input_tokens_seen": 32602375,
      "step": 1525,
      "time_per_iteration": 2.5458011627197266
    },
    {
      "auxiliary_loss_clip": 0.01160855,
      "auxiliary_loss_mlp": 0.01045536,
      "balance_loss_clip": 1.05560803,
      "balance_loss_mlp": 1.02368498,
      "epoch": 0.0917480835713212,
      "flos": 29387605075200.0,
      "grad_norm": 2.2998286185292156,
      "language_loss": 0.74329484,
      "learning_rate": 3.96013769577032e-06,
      "loss": 0.76535869,
      "num_input_tokens_seen": 32621460,
      "step": 1526,
      "time_per_iteration": 2.6242287158966064
    },
    {
      "auxiliary_loss_clip": 0.0118423,
      "auxiliary_loss_mlp": 0.01058365,
      "balance_loss_clip": 1.05610669,
      "balance_loss_mlp": 1.03589368,
      "epoch": 0.09180820682398917,
      "flos": 19829262378240.0,
      "grad_norm": 1.926553207756451,
      "language_loss": 0.77380705,
      "learning_rate": 3.960060288876378e-06,
      "loss": 0.79623306,
      "num_input_tokens_seen": 32640440,
      "step": 1527,
      "time_per_iteration": 2.4857473373413086
    },
    {
      "auxiliary_loss_clip": 0.0117049,
      "auxiliary_loss_mlp": 0.01052835,
      "balance_loss_clip": 1.05101466,
      "balance_loss_mlp": 1.02985168,
      "epoch": 0.09186833007665715,
      "flos": 23841848643840.0,
      "grad_norm": 1.7369661645416155,
      "language_loss": 0.78218365,
      "learning_rate": 3.959982807656753e-06,
      "loss": 0.80441689,
      "num_input_tokens_seen": 32660020,
      "step": 1528,
      "time_per_iteration": 2.5360777378082275
    },
    {
      "auxiliary_loss_clip": 0.01143211,
      "auxiliary_loss_mlp": 0.01050171,
      "balance_loss_clip": 1.0521872,
      "balance_loss_mlp": 1.02886832,
      "epoch": 0.09192845332932512,
      "flos": 12932726065920.0,
      "grad_norm": 2.8216388968307333,
      "language_loss": 0.76876384,
      "learning_rate": 3.959905252114384e-06,
      "loss": 0.79069763,
      "num_input_tokens_seen": 32678170,
      "step": 1529,
      "time_per_iteration": 2.5686495304107666
    },
    {
      "auxiliary_loss_clip": 0.01186314,
      "auxiliary_loss_mlp": 0.0079079,
      "balance_loss_clip": 1.05413091,
      "balance_loss_mlp": 1.01402617,
      "epoch": 0.09198857658199308,
      "flos": 24568177559040.0,
      "grad_norm": 2.078637330458293,
      "language_loss": 0.82868236,
      "learning_rate": 3.959827622252211e-06,
      "loss": 0.8484534,
      "num_input_tokens_seen": 32697540,
      "step": 1530,
      "time_per_iteration": 2.528794527053833
    },
    {
      "auxiliary_loss_clip": 0.01133386,
      "auxiliary_loss_mlp": 0.01058511,
      "balance_loss_clip": 1.06042874,
      "balance_loss_mlp": 1.03611135,
      "epoch": 0.09204869983466106,
      "flos": 20266941600000.0,
      "grad_norm": 2.7867869229389717,
      "language_loss": 0.83862007,
      "learning_rate": 3.959749918073179e-06,
      "loss": 0.86053896,
      "num_input_tokens_seen": 32716805,
      "step": 1531,
      "time_per_iteration": 2.5940892696380615
    },
    {
      "auxiliary_loss_clip": 0.0113647,
      "auxiliary_loss_mlp": 0.01047935,
      "balance_loss_clip": 1.05203843,
      "balance_loss_mlp": 1.02551198,
      "epoch": 0.09210882308732903,
      "flos": 20885646389760.0,
      "grad_norm": 2.40925776766462,
      "language_loss": 0.81330812,
      "learning_rate": 3.959672139580233e-06,
      "loss": 0.83515215,
      "num_input_tokens_seen": 32736385,
      "step": 1532,
      "time_per_iteration": 2.577235221862793
    },
    {
      "auxiliary_loss_clip": 0.01161101,
      "auxiliary_loss_mlp": 0.01054051,
      "balance_loss_clip": 1.05990183,
      "balance_loss_mlp": 1.03099608,
      "epoch": 0.09216894633999699,
      "flos": 30956326727040.0,
      "grad_norm": 2.1129730529221784,
      "language_loss": 0.83712405,
      "learning_rate": 3.9595942867763235e-06,
      "loss": 0.85927558,
      "num_input_tokens_seen": 32757140,
      "step": 1533,
      "time_per_iteration": 2.6085715293884277
    },
    {
      "auxiliary_loss_clip": 0.01144966,
      "auxiliary_loss_mlp": 0.0104919,
      "balance_loss_clip": 1.05487084,
      "balance_loss_mlp": 1.02674329,
      "epoch": 0.09222906959266497,
      "flos": 13151565676800.0,
      "grad_norm": 2.0316249191300093,
      "language_loss": 0.89692444,
      "learning_rate": 3.959516359664402e-06,
      "loss": 0.91886592,
      "num_input_tokens_seen": 32774860,
      "step": 1534,
      "time_per_iteration": 2.560976028442383
    },
    {
      "auxiliary_loss_clip": 0.01150648,
      "auxiliary_loss_mlp": 0.01060939,
      "balance_loss_clip": 1.05135679,
      "balance_loss_mlp": 1.03617966,
      "epoch": 0.09228919284533293,
      "flos": 25994477784960.0,
      "grad_norm": 2.275219924248099,
      "language_loss": 0.75629532,
      "learning_rate": 3.959438358247424e-06,
      "loss": 0.77841121,
      "num_input_tokens_seen": 32795250,
      "step": 1535,
      "time_per_iteration": 2.590768337249756
    },
    {
      "auxiliary_loss_clip": 0.01165007,
      "auxiliary_loss_mlp": 0.01045691,
      "balance_loss_clip": 1.04877067,
      "balance_loss_mlp": 1.02467477,
      "epoch": 0.0923493160980009,
      "flos": 18660800954880.0,
      "grad_norm": 1.8460175491267252,
      "language_loss": 0.81477356,
      "learning_rate": 3.959360282528346e-06,
      "loss": 0.83688056,
      "num_input_tokens_seen": 32813805,
      "step": 1536,
      "time_per_iteration": 2.5068020820617676
    },
    {
      "auxiliary_loss_clip": 0.01178701,
      "auxiliary_loss_mlp": 0.01050358,
      "balance_loss_clip": 1.05230856,
      "balance_loss_mlp": 1.02973461,
      "epoch": 0.09240943935066886,
      "flos": 21140576190720.0,
      "grad_norm": 2.195315411546666,
      "language_loss": 0.89235657,
      "learning_rate": 3.959282132510131e-06,
      "loss": 0.91464716,
      "num_input_tokens_seen": 32830960,
      "step": 1537,
      "time_per_iteration": 4.124530553817749
    },
    {
      "auxiliary_loss_clip": 0.01156569,
      "auxiliary_loss_mlp": 0.01060364,
      "balance_loss_clip": 1.05115664,
      "balance_loss_mlp": 1.03720188,
      "epoch": 0.09246956260333684,
      "flos": 20592435669120.0,
      "grad_norm": 2.419909202146827,
      "language_loss": 0.80911404,
      "learning_rate": 3.959203908195741e-06,
      "loss": 0.83128333,
      "num_input_tokens_seen": 32848275,
      "step": 1538,
      "time_per_iteration": 2.555480480194092
    },
    {
      "auxiliary_loss_clip": 0.01057762,
      "auxiliary_loss_mlp": 0.01017607,
      "balance_loss_clip": 1.0289197,
      "balance_loss_mlp": 1.0137924,
      "epoch": 0.09252968585600481,
      "flos": 67558710614400.0,
      "grad_norm": 0.7326619234038901,
      "language_loss": 0.5741455,
      "learning_rate": 3.959125609588142e-06,
      "loss": 0.59489912,
      "num_input_tokens_seen": 32917730,
      "step": 1539,
      "time_per_iteration": 4.670522212982178
    },
    {
      "auxiliary_loss_clip": 0.01154331,
      "auxiliary_loss_mlp": 0.01049804,
      "balance_loss_clip": 1.05379295,
      "balance_loss_mlp": 1.02755964,
      "epoch": 0.09258980910867277,
      "flos": 17383853479680.0,
      "grad_norm": 2.5244353851334598,
      "language_loss": 0.6706208,
      "learning_rate": 3.959047236690304e-06,
      "loss": 0.69266206,
      "num_input_tokens_seen": 32934910,
      "step": 1540,
      "time_per_iteration": 2.497007131576538
    },
    {
      "auxiliary_loss_clip": 0.01143622,
      "auxiliary_loss_mlp": 0.01046373,
      "balance_loss_clip": 1.05054176,
      "balance_loss_mlp": 1.02350879,
      "epoch": 0.09264993236134075,
      "flos": 19865927185920.0,
      "grad_norm": 1.745557933283031,
      "language_loss": 0.83886689,
      "learning_rate": 3.958968789505198e-06,
      "loss": 0.86076683,
      "num_input_tokens_seen": 32953840,
      "step": 1541,
      "time_per_iteration": 2.575521230697632
    },
    {
      "auxiliary_loss_clip": 0.01078577,
      "auxiliary_loss_mlp": 0.01014772,
      "balance_loss_clip": 1.02356029,
      "balance_loss_mlp": 1.01069498,
      "epoch": 0.09271005561400872,
      "flos": 62284401262080.0,
      "grad_norm": 0.8887921319015558,
      "language_loss": 0.61884367,
      "learning_rate": 3.9588902680358e-06,
      "loss": 0.63977718,
      "num_input_tokens_seen": 33011410,
      "step": 1542,
      "time_per_iteration": 3.0866987705230713
    },
    {
      "auxiliary_loss_clip": 0.0116073,
      "auxiliary_loss_mlp": 0.01054241,
      "balance_loss_clip": 1.05572999,
      "balance_loss_mlp": 1.03290248,
      "epoch": 0.09277017886667668,
      "flos": 23329870139520.0,
      "grad_norm": 1.6421952065219911,
      "language_loss": 0.82431793,
      "learning_rate": 3.958811672285086e-06,
      "loss": 0.84646761,
      "num_input_tokens_seen": 33031675,
      "step": 1543,
      "time_per_iteration": 3.969069719314575
    },
    {
      "auxiliary_loss_clip": 0.01134819,
      "auxiliary_loss_mlp": 0.01059313,
      "balance_loss_clip": 1.05058241,
      "balance_loss_mlp": 1.03715229,
      "epoch": 0.09283030211934466,
      "flos": 54745169875200.0,
      "grad_norm": 1.8297290651731999,
      "language_loss": 0.7179746,
      "learning_rate": 3.958733002256038e-06,
      "loss": 0.73991591,
      "num_input_tokens_seen": 33056355,
      "step": 1544,
      "time_per_iteration": 2.905487537384033
    },
    {
      "auxiliary_loss_clip": 0.01164396,
      "auxiliary_loss_mlp": 0.01053977,
      "balance_loss_clip": 1.0526042,
      "balance_loss_mlp": 1.02913392,
      "epoch": 0.09289042537201263,
      "flos": 30334784762880.0,
      "grad_norm": 1.691556555425474,
      "language_loss": 0.77237624,
      "learning_rate": 3.958654257951637e-06,
      "loss": 0.79456002,
      "num_input_tokens_seen": 33079520,
      "step": 1545,
      "time_per_iteration": 2.594414710998535
    },
    {
      "auxiliary_loss_clip": 0.01138274,
      "auxiliary_loss_mlp": 0.0104952,
      "balance_loss_clip": 1.05402517,
      "balance_loss_mlp": 1.02818108,
      "epoch": 0.09295054862468059,
      "flos": 17746838369280.0,
      "grad_norm": 2.959016047578255,
      "language_loss": 0.74520642,
      "learning_rate": 3.9585754393748706e-06,
      "loss": 0.76708436,
      "num_input_tokens_seen": 33096135,
      "step": 1546,
      "time_per_iteration": 3.9474639892578125
    },
    {
      "auxiliary_loss_clip": 0.01161308,
      "auxiliary_loss_mlp": 0.01052578,
      "balance_loss_clip": 1.05402732,
      "balance_loss_mlp": 1.02946365,
      "epoch": 0.09301067187734856,
      "flos": 23658021815040.0,
      "grad_norm": 2.026381835327108,
      "language_loss": 0.84034383,
      "learning_rate": 3.9584965465287275e-06,
      "loss": 0.86248273,
      "num_input_tokens_seen": 33115245,
      "step": 1547,
      "time_per_iteration": 2.5713911056518555
    },
    {
      "auxiliary_loss_clip": 0.0114248,
      "auxiliary_loss_mlp": 0.0105182,
      "balance_loss_clip": 1.04959869,
      "balance_loss_mlp": 1.02985024,
      "epoch": 0.09307079513001654,
      "flos": 27527719777920.0,
      "grad_norm": 2.0871849131709936,
      "language_loss": 0.67568362,
      "learning_rate": 3.958417579416199e-06,
      "loss": 0.69762659,
      "num_input_tokens_seen": 33136640,
      "step": 1548,
      "time_per_iteration": 2.629547357559204
    },
    {
      "auxiliary_loss_clip": 0.01122767,
      "auxiliary_loss_mlp": 0.01057982,
      "balance_loss_clip": 1.05126691,
      "balance_loss_mlp": 1.03545189,
      "epoch": 0.0931309183826845,
      "flos": 20627340710400.0,
      "grad_norm": 1.9046972450781892,
      "language_loss": 0.83350712,
      "learning_rate": 3.9583385380402795e-06,
      "loss": 0.85531461,
      "num_input_tokens_seen": 33155060,
      "step": 1549,
      "time_per_iteration": 2.6733181476593018
    },
    {
      "auxiliary_loss_clip": 0.01173461,
      "auxiliary_loss_mlp": 0.01046868,
      "balance_loss_clip": 1.0569911,
      "balance_loss_mlp": 1.02550602,
      "epoch": 0.09319104163535247,
      "flos": 29020921084800.0,
      "grad_norm": 1.7418675037786757,
      "language_loss": 0.75808239,
      "learning_rate": 3.958259422403966e-06,
      "loss": 0.78028572,
      "num_input_tokens_seen": 33175420,
      "step": 1550,
      "time_per_iteration": 2.5949177742004395
    },
    {
      "auxiliary_loss_clip": 0.01150137,
      "auxiliary_loss_mlp": 0.01071568,
      "balance_loss_clip": 1.05701232,
      "balance_loss_mlp": 1.04674852,
      "epoch": 0.09325116488802045,
      "flos": 25301545539840.0,
      "grad_norm": 2.2901676843496963,
      "language_loss": 0.83363521,
      "learning_rate": 3.95818023251026e-06,
      "loss": 0.85585225,
      "num_input_tokens_seen": 33194120,
      "step": 1551,
      "time_per_iteration": 2.6354622840881348
    },
    {
      "auxiliary_loss_clip": 0.01070022,
      "auxiliary_loss_mlp": 0.01095191,
      "balance_loss_clip": 1.02603698,
      "balance_loss_mlp": 1.61678183,
      "epoch": 0.09331128814068841,
      "flos": 61536203942400.0,
      "grad_norm": 0.87965634133191,
      "language_loss": 0.61862975,
      "learning_rate": 3.958100968362163e-06,
      "loss": 0.64028192,
      "num_input_tokens_seen": 33261080,
      "step": 1552,
      "time_per_iteration": 3.2560019493103027
    },
    {
      "auxiliary_loss_clip": 0.01067926,
      "auxiliary_loss_mlp": 0.01033418,
      "balance_loss_clip": 1.02976322,
      "balance_loss_mlp": 1.02867341,
      "epoch": 0.09337141139335638,
      "flos": 53293700171520.0,
      "grad_norm": 0.8516254911756155,
      "language_loss": 0.58916771,
      "learning_rate": 3.958021629962681e-06,
      "loss": 0.61018121,
      "num_input_tokens_seen": 33330235,
      "step": 1553,
      "time_per_iteration": 3.25569748878479
    },
    {
      "auxiliary_loss_clip": 0.0115257,
      "auxiliary_loss_mlp": 0.01057298,
      "balance_loss_clip": 1.06284237,
      "balance_loss_mlp": 1.03425479,
      "epoch": 0.09343153464602436,
      "flos": 23476852592640.0,
      "grad_norm": 1.72815007581565,
      "language_loss": 0.87467062,
      "learning_rate": 3.957942217314823e-06,
      "loss": 0.89676929,
      "num_input_tokens_seen": 33349035,
      "step": 1554,
      "time_per_iteration": 2.5997087955474854
    },
    {
      "auxiliary_loss_clip": 0.01146938,
      "auxiliary_loss_mlp": 0.01056765,
      "balance_loss_clip": 1.0507946,
      "balance_loss_mlp": 1.03406811,
      "epoch": 0.09349165789869232,
      "flos": 19353481804800.0,
      "grad_norm": 1.8801212045448559,
      "language_loss": 0.81200826,
      "learning_rate": 3.957862730421599e-06,
      "loss": 0.83404529,
      "num_input_tokens_seen": 33368060,
      "step": 1555,
      "time_per_iteration": 2.5348825454711914
    },
    {
      "auxiliary_loss_clip": 0.01061985,
      "auxiliary_loss_mlp": 0.0104314,
      "balance_loss_clip": 1.02095282,
      "balance_loss_mlp": 1.03977823,
      "epoch": 0.09355178115136029,
      "flos": 67502580635520.0,
      "grad_norm": 0.997381150092209,
      "language_loss": 0.59629101,
      "learning_rate": 3.957783169286024e-06,
      "loss": 0.61734229,
      "num_input_tokens_seen": 33430825,
      "step": 1556,
      "time_per_iteration": 3.1305603981018066
    },
    {
      "auxiliary_loss_clip": 0.01167913,
      "auxiliary_loss_mlp": 0.01065249,
      "balance_loss_clip": 1.05246568,
      "balance_loss_mlp": 1.04391003,
      "epoch": 0.09361190440402825,
      "flos": 37341638720640.0,
      "grad_norm": 1.8507075355396525,
      "language_loss": 0.84420449,
      "learning_rate": 3.9577035339111155e-06,
      "loss": 0.86653608,
      "num_input_tokens_seen": 33454855,
      "step": 1557,
      "time_per_iteration": 2.6349875926971436
    },
    {
      "auxiliary_loss_clip": 0.01106851,
      "auxiliary_loss_mlp": 0.01075589,
      "balance_loss_clip": 1.04926777,
      "balance_loss_mlp": 1.0516156,
      "epoch": 0.09367202765669623,
      "flos": 24899705112960.0,
      "grad_norm": 1.9344503149162864,
      "language_loss": 0.78060937,
      "learning_rate": 3.957623824299893e-06,
      "loss": 0.80243379,
      "num_input_tokens_seen": 33476000,
      "step": 1558,
      "time_per_iteration": 2.71237850189209
    },
    {
      "auxiliary_loss_clip": 0.0115758,
      "auxiliary_loss_mlp": 0.01055428,
      "balance_loss_clip": 1.0535996,
      "balance_loss_mlp": 1.03336191,
      "epoch": 0.0937321509093642,
      "flos": 15705568368000.0,
      "grad_norm": 1.9910814817631814,
      "language_loss": 0.79729378,
      "learning_rate": 3.957544040455379e-06,
      "loss": 0.81942379,
      "num_input_tokens_seen": 33493845,
      "step": 1559,
      "time_per_iteration": 2.519101619720459
    },
    {
      "auxiliary_loss_clip": 0.0113735,
      "auxiliary_loss_mlp": 0.01067095,
      "balance_loss_clip": 1.05141127,
      "balance_loss_mlp": 1.0458281,
      "epoch": 0.09379227416203216,
      "flos": 20483698222080.0,
      "grad_norm": 2.2360972993240122,
      "language_loss": 0.76485443,
      "learning_rate": 3.957464182380599e-06,
      "loss": 0.78689885,
      "num_input_tokens_seen": 33510850,
      "step": 1560,
      "time_per_iteration": 2.5743136405944824
    },
    {
      "auxiliary_loss_clip": 0.01135336,
      "auxiliary_loss_mlp": 0.01066319,
      "balance_loss_clip": 1.05189443,
      "balance_loss_mlp": 1.04390752,
      "epoch": 0.09385239741470014,
      "flos": 24352498344960.0,
      "grad_norm": 1.8218549194015512,
      "language_loss": 0.81066704,
      "learning_rate": 3.95738425007858e-06,
      "loss": 0.83268356,
      "num_input_tokens_seen": 33530430,
      "step": 1561,
      "time_per_iteration": 2.6112961769104004
    },
    {
      "auxiliary_loss_clip": 0.01169797,
      "auxiliary_loss_mlp": 0.01055214,
      "balance_loss_clip": 1.05206347,
      "balance_loss_mlp": 1.03343451,
      "epoch": 0.0939125206673681,
      "flos": 33291489807360.0,
      "grad_norm": 2.5651116649916244,
      "language_loss": 0.61222744,
      "learning_rate": 3.957304243552354e-06,
      "loss": 0.6344775,
      "num_input_tokens_seen": 33551975,
      "step": 1562,
      "time_per_iteration": 2.626699447631836
    },
    {
      "auxiliary_loss_clip": 0.01161295,
      "auxiliary_loss_mlp": 0.01063531,
      "balance_loss_clip": 1.05953634,
      "balance_loss_mlp": 1.04275286,
      "epoch": 0.09397264392003607,
      "flos": 19244923925760.0,
      "grad_norm": 2.3201285129485134,
      "language_loss": 0.85000223,
      "learning_rate": 3.957224162804956e-06,
      "loss": 0.87225056,
      "num_input_tokens_seen": 33569850,
      "step": 1563,
      "time_per_iteration": 2.537867546081543
    },
    {
      "auxiliary_loss_clip": 0.01160679,
      "auxiliary_loss_mlp": 0.01056693,
      "balance_loss_clip": 1.05530453,
      "balance_loss_mlp": 1.03580725,
      "epoch": 0.09403276717270405,
      "flos": 19317930318720.0,
      "grad_norm": 2.264358492601823,
      "language_loss": 0.76249576,
      "learning_rate": 3.9571440078394205e-06,
      "loss": 0.78466946,
      "num_input_tokens_seen": 33590510,
      "step": 1564,
      "time_per_iteration": 2.569422721862793
    },
    {
      "auxiliary_loss_clip": 0.01155982,
      "auxiliary_loss_mlp": 0.01055514,
      "balance_loss_clip": 1.06261706,
      "balance_loss_mlp": 1.03435445,
      "epoch": 0.09409289042537201,
      "flos": 23583471137280.0,
      "grad_norm": 2.2121213470529786,
      "language_loss": 0.80234188,
      "learning_rate": 3.9570637786587895e-06,
      "loss": 0.82445693,
      "num_input_tokens_seen": 33608810,
      "step": 1565,
      "time_per_iteration": 2.555333375930786
    },
    {
      "auxiliary_loss_clip": 0.01151983,
      "auxiliary_loss_mlp": 0.0106829,
      "balance_loss_clip": 1.05090725,
      "balance_loss_mlp": 1.04748797,
      "epoch": 0.09415301367803998,
      "flos": 20078446003200.0,
      "grad_norm": 2.1695145049123563,
      "language_loss": 0.75791264,
      "learning_rate": 3.956983475266103e-06,
      "loss": 0.78011543,
      "num_input_tokens_seen": 33627265,
      "step": 1566,
      "time_per_iteration": 2.5790772438049316
    },
    {
      "auxiliary_loss_clip": 0.01143436,
      "auxiliary_loss_mlp": 0.01078395,
      "balance_loss_clip": 1.05343819,
      "balance_loss_mlp": 1.5408802,
      "epoch": 0.09421313693070796,
      "flos": 21062075016960.0,
      "grad_norm": 2.219322228617997,
      "language_loss": 0.78245521,
      "learning_rate": 3.956903097664407e-06,
      "loss": 0.80467349,
      "num_input_tokens_seen": 33644810,
      "step": 1567,
      "time_per_iteration": 2.593858242034912
    },
    {
      "auxiliary_loss_clip": 0.01155362,
      "auxiliary_loss_mlp": 0.01054355,
      "balance_loss_clip": 1.05386257,
      "balance_loss_mlp": 1.03393412,
      "epoch": 0.09427326018337592,
      "flos": 24316156759680.0,
      "grad_norm": 2.064782258190575,
      "language_loss": 0.82442731,
      "learning_rate": 3.956822645856749e-06,
      "loss": 0.84652448,
      "num_input_tokens_seen": 33665665,
      "step": 1568,
      "time_per_iteration": 2.5692715644836426
    },
    {
      "auxiliary_loss_clip": 0.01186399,
      "auxiliary_loss_mlp": 0.01050723,
      "balance_loss_clip": 1.05698466,
      "balance_loss_mlp": 1.02789426,
      "epoch": 0.09433338343604389,
      "flos": 20263888944000.0,
      "grad_norm": 2.2988660845124333,
      "language_loss": 0.76480532,
      "learning_rate": 3.9567421198461814e-06,
      "loss": 0.78717649,
      "num_input_tokens_seen": 33684760,
      "step": 1569,
      "time_per_iteration": 2.489922523498535
    },
    {
      "auxiliary_loss_clip": 0.01118612,
      "auxiliary_loss_mlp": 0.0105991,
      "balance_loss_clip": 1.05112243,
      "balance_loss_mlp": 1.03591323,
      "epoch": 0.09439350668871185,
      "flos": 12742973493120.0,
      "grad_norm": 2.219714831850327,
      "language_loss": 0.85613203,
      "learning_rate": 3.956661519635756e-06,
      "loss": 0.87791729,
      "num_input_tokens_seen": 33700750,
      "step": 1570,
      "time_per_iteration": 2.5774660110473633
    },
    {
      "auxiliary_loss_clip": 0.01122979,
      "auxiliary_loss_mlp": 0.0105643,
      "balance_loss_clip": 1.05312991,
      "balance_loss_mlp": 1.0323379,
      "epoch": 0.09445362994137983,
      "flos": 25962266263680.0,
      "grad_norm": 1.5804812266916313,
      "language_loss": 0.76523054,
      "learning_rate": 3.95658084522853e-06,
      "loss": 0.78702462,
      "num_input_tokens_seen": 33724430,
      "step": 1571,
      "time_per_iteration": 2.6757442951202393
    },
    {
      "auxiliary_loss_clip": 0.01133502,
      "auxiliary_loss_mlp": 0.01053381,
      "balance_loss_clip": 1.04924881,
      "balance_loss_mlp": 1.03151822,
      "epoch": 0.0945137531940478,
      "flos": 19715353372800.0,
      "grad_norm": 1.6541776610975596,
      "language_loss": 0.79374641,
      "learning_rate": 3.956500096627561e-06,
      "loss": 0.81561524,
      "num_input_tokens_seen": 33743455,
      "step": 1572,
      "time_per_iteration": 2.600696563720703
    },
    {
      "auxiliary_loss_clip": 0.01143772,
      "auxiliary_loss_mlp": 0.01064729,
      "balance_loss_clip": 1.05482936,
      "balance_loss_mlp": 1.04086351,
      "epoch": 0.09457387644671576,
      "flos": 23617047375360.0,
      "grad_norm": 1.7919013296474864,
      "language_loss": 0.87676793,
      "learning_rate": 3.956419273835913e-06,
      "loss": 0.89885294,
      "num_input_tokens_seen": 33763435,
      "step": 1573,
      "time_per_iteration": 2.6169488430023193
    },
    {
      "auxiliary_loss_clip": 0.01157401,
      "auxiliary_loss_mlp": 0.01060854,
      "balance_loss_clip": 1.05487692,
      "balance_loss_mlp": 1.03679776,
      "epoch": 0.09463399969938374,
      "flos": 26907291135360.0,
      "grad_norm": 4.115764256561912,
      "language_loss": 0.81898856,
      "learning_rate": 3.95633837685665e-06,
      "loss": 0.84117115,
      "num_input_tokens_seen": 33784325,
      "step": 1574,
      "time_per_iteration": 2.5944738388061523
    },
    {
      "auxiliary_loss_clip": 0.01163494,
      "auxiliary_loss_mlp": 0.0105581,
      "balance_loss_clip": 1.0651145,
      "balance_loss_mlp": 1.03431666,
      "epoch": 0.0946941229520517,
      "flos": 23659566099840.0,
      "grad_norm": 1.7698475406768694,
      "language_loss": 0.81076109,
      "learning_rate": 3.95625740569284e-06,
      "loss": 0.83295411,
      "num_input_tokens_seen": 33802510,
      "step": 1575,
      "time_per_iteration": 4.0059428215026855
    },
    {
      "auxiliary_loss_clip": 0.01179561,
      "auxiliary_loss_mlp": 0.01067018,
      "balance_loss_clip": 1.05460763,
      "balance_loss_mlp": 1.044595,
      "epoch": 0.09475424620471967,
      "flos": 24134053783680.0,
      "grad_norm": 2.139997361908705,
      "language_loss": 0.86727715,
      "learning_rate": 3.956176360347553e-06,
      "loss": 0.88974291,
      "num_input_tokens_seen": 33819980,
      "step": 1576,
      "time_per_iteration": 2.4934327602386475
    },
    {
      "auxiliary_loss_clip": 0.01066079,
      "auxiliary_loss_mlp": 0.01040781,
      "balance_loss_clip": 1.03184152,
      "balance_loss_mlp": 1.03753877,
      "epoch": 0.09481436945738765,
      "flos": 68426168065920.0,
      "grad_norm": 0.9983005684582642,
      "language_loss": 0.65914381,
      "learning_rate": 3.956095240823862e-06,
      "loss": 0.68021238,
      "num_input_tokens_seen": 33878925,
      "step": 1577,
      "time_per_iteration": 3.12056040763855
    },
    {
      "auxiliary_loss_clip": 0.01146206,
      "auxiliary_loss_mlp": 0.01048159,
      "balance_loss_clip": 1.05278289,
      "balance_loss_mlp": 1.02748787,
      "epoch": 0.09487449271005562,
      "flos": 16654076858880.0,
      "grad_norm": 2.06515791280812,
      "language_loss": 0.79197598,
      "learning_rate": 3.956014047124844e-06,
      "loss": 0.81391954,
      "num_input_tokens_seen": 33897600,
      "step": 1578,
      "time_per_iteration": 4.015286445617676
    },
    {
      "auxiliary_loss_clip": 0.01181216,
      "auxiliary_loss_mlp": 0.01059665,
      "balance_loss_clip": 1.0543437,
      "balance_loss_mlp": 1.03744447,
      "epoch": 0.09493461596272358,
      "flos": 24275685110400.0,
      "grad_norm": 1.8451357841332325,
      "language_loss": 0.77985489,
      "learning_rate": 3.955932779253578e-06,
      "loss": 0.80226368,
      "num_input_tokens_seen": 33917365,
      "step": 1579,
      "time_per_iteration": 2.513169288635254
    },
    {
      "auxiliary_loss_clip": 0.01129192,
      "auxiliary_loss_mlp": 0.01068118,
      "balance_loss_clip": 1.05336118,
      "balance_loss_mlp": 1.04277372,
      "epoch": 0.09499473921539155,
      "flos": 21870173243520.0,
      "grad_norm": 1.8742281381555126,
      "language_loss": 0.7313627,
      "learning_rate": 3.955851437213144e-06,
      "loss": 0.75333583,
      "num_input_tokens_seen": 33936680,
      "step": 1580,
      "time_per_iteration": 2.59196138381958
    },
    {
      "auxiliary_loss_clip": 0.01153607,
      "auxiliary_loss_mlp": 0.01059287,
      "balance_loss_clip": 1.05490541,
      "balance_loss_mlp": 1.03790092,
      "epoch": 0.09505486246805953,
      "flos": 33547137880320.0,
      "grad_norm": 1.764348424770447,
      "language_loss": 0.7758491,
      "learning_rate": 3.955770021006627e-06,
      "loss": 0.79797804,
      "num_input_tokens_seen": 33960685,
      "step": 1581,
      "time_per_iteration": 2.6468141078948975
    },
    {
      "auxiliary_loss_clip": 0.0114883,
      "auxiliary_loss_mlp": 0.0106053,
      "balance_loss_clip": 1.06200695,
      "balance_loss_mlp": 1.03909636,
      "epoch": 0.09511498572072749,
      "flos": 21215342350080.0,
      "grad_norm": 1.939717258271165,
      "language_loss": 0.87108123,
      "learning_rate": 3.955688530637116e-06,
      "loss": 0.89317483,
      "num_input_tokens_seen": 33980015,
      "step": 1582,
      "time_per_iteration": 3.943749189376831
    },
    {
      "auxiliary_loss_clip": 0.01171473,
      "auxiliary_loss_mlp": 0.01066541,
      "balance_loss_clip": 1.05443192,
      "balance_loss_mlp": 1.04309297,
      "epoch": 0.09517510897339546,
      "flos": 14611262572800.0,
      "grad_norm": 3.958624038837923,
      "language_loss": 0.66848898,
      "learning_rate": 3.955606966107699e-06,
      "loss": 0.69086909,
      "num_input_tokens_seen": 33997705,
      "step": 1583,
      "time_per_iteration": 2.528214931488037
    },
    {
      "auxiliary_loss_clip": 0.01175074,
      "auxiliary_loss_mlp": 0.01059559,
      "balance_loss_clip": 1.05801415,
      "balance_loss_mlp": 1.03595519,
      "epoch": 0.09523523222606343,
      "flos": 27817339138560.0,
      "grad_norm": 1.8156715554857168,
      "language_loss": 0.70310569,
      "learning_rate": 3.95552532742147e-06,
      "loss": 0.72545207,
      "num_input_tokens_seen": 34017465,
      "step": 1584,
      "time_per_iteration": 3.9374709129333496
    },
    {
      "auxiliary_loss_clip": 0.01136259,
      "auxiliary_loss_mlp": 0.01056116,
      "balance_loss_clip": 1.05417848,
      "balance_loss_mlp": 1.03556383,
      "epoch": 0.0952953554787314,
      "flos": 20706272847360.0,
      "grad_norm": 1.4653412588862806,
      "language_loss": 0.8085202,
      "learning_rate": 3.955443614581525e-06,
      "loss": 0.83044392,
      "num_input_tokens_seen": 34038550,
      "step": 1585,
      "time_per_iteration": 2.658764362335205
    },
    {
      "auxiliary_loss_clip": 0.01156534,
      "auxiliary_loss_mlp": 0.0105136,
      "balance_loss_clip": 1.05471265,
      "balance_loss_mlp": 1.02823377,
      "epoch": 0.09535547873139937,
      "flos": 24787627701120.0,
      "grad_norm": 1.888139747344243,
      "language_loss": 0.71915269,
      "learning_rate": 3.955361827590961e-06,
      "loss": 0.74123162,
      "num_input_tokens_seen": 34058665,
      "step": 1586,
      "time_per_iteration": 2.5734524726867676
    },
    {
      "auxiliary_loss_clip": 0.01045144,
      "auxiliary_loss_mlp": 0.01010412,
      "balance_loss_clip": 1.02752435,
      "balance_loss_mlp": 1.00674021,
      "epoch": 0.09541560198406734,
      "flos": 71912194905600.0,
      "grad_norm": 0.8153150383920831,
      "language_loss": 0.55521101,
      "learning_rate": 3.955279966452883e-06,
      "loss": 0.5757665,
      "num_input_tokens_seen": 34109655,
      "step": 1587,
      "time_per_iteration": 2.9945757389068604
    },
    {
      "auxiliary_loss_clip": 0.01133424,
      "auxiliary_loss_mlp": 0.01059992,
      "balance_loss_clip": 1.05920863,
      "balance_loss_mlp": 1.03659153,
      "epoch": 0.09547572523673531,
      "flos": 28982604251520.0,
      "grad_norm": 1.82403793794127,
      "language_loss": 0.81030267,
      "learning_rate": 3.955198031170391e-06,
      "loss": 0.83223689,
      "num_input_tokens_seen": 34131115,
      "step": 1588,
      "time_per_iteration": 2.650310754776001
    },
    {
      "auxiliary_loss_clip": 0.0113642,
      "auxiliary_loss_mlp": 0.01057681,
      "balance_loss_clip": 1.05823517,
      "balance_loss_mlp": 1.03552008,
      "epoch": 0.09553584848940327,
      "flos": 24133910129280.0,
      "grad_norm": 1.494204970556393,
      "language_loss": 0.81835914,
      "learning_rate": 3.955116021746594e-06,
      "loss": 0.8403002,
      "num_input_tokens_seen": 34151925,
      "step": 1589,
      "time_per_iteration": 2.6201870441436768
    },
    {
      "auxiliary_loss_clip": 0.01120598,
      "auxiliary_loss_mlp": 0.01180891,
      "balance_loss_clip": 1.05364001,
      "balance_loss_mlp": 1.72993553,
      "epoch": 0.09559597174207124,
      "flos": 42851376789120.0,
      "grad_norm": 2.2299908302934957,
      "language_loss": 0.64793944,
      "learning_rate": 3.955033938184601e-06,
      "loss": 0.67095435,
      "num_input_tokens_seen": 34175395,
      "step": 1590,
      "time_per_iteration": 2.8149635791778564
    },
    {
      "auxiliary_loss_clip": 0.01141017,
      "auxiliary_loss_mlp": 0.01054142,
      "balance_loss_clip": 1.05025387,
      "balance_loss_mlp": 1.03231502,
      "epoch": 0.09565609499473922,
      "flos": 32670845683200.0,
      "grad_norm": 1.7344772996572568,
      "language_loss": 0.83031487,
      "learning_rate": 3.954951780487526e-06,
      "loss": 0.85226643,
      "num_input_tokens_seen": 34197760,
      "step": 1591,
      "time_per_iteration": 2.6802310943603516
    },
    {
      "auxiliary_loss_clip": 0.01161069,
      "auxiliary_loss_mlp": 0.0106088,
      "balance_loss_clip": 1.053231,
      "balance_loss_mlp": 1.03838539,
      "epoch": 0.09571621824740718,
      "flos": 18478410670080.0,
      "grad_norm": 3.0702287123028094,
      "language_loss": 0.74089086,
      "learning_rate": 3.9548695486584835e-06,
      "loss": 0.76311034,
      "num_input_tokens_seen": 34215330,
      "step": 1592,
      "time_per_iteration": 2.53536319732666
    },
    {
      "auxiliary_loss_clip": 0.01165021,
      "auxiliary_loss_mlp": 0.01054975,
      "balance_loss_clip": 1.04969668,
      "balance_loss_mlp": 1.03280163,
      "epoch": 0.09577634150007515,
      "flos": 29387497334400.0,
      "grad_norm": 1.8843232081617205,
      "language_loss": 0.7426821,
      "learning_rate": 3.954787242700592e-06,
      "loss": 0.76488203,
      "num_input_tokens_seen": 34237745,
      "step": 1593,
      "time_per_iteration": 2.5781474113464355
    },
    {
      "auxiliary_loss_clip": 0.01171833,
      "auxiliary_loss_mlp": 0.01063316,
      "balance_loss_clip": 1.05449164,
      "balance_loss_mlp": 1.04221547,
      "epoch": 0.09583646475274313,
      "flos": 22747830157440.0,
      "grad_norm": 1.8146648384153161,
      "language_loss": 0.69971657,
      "learning_rate": 3.954704862616971e-06,
      "loss": 0.72206807,
      "num_input_tokens_seen": 34256565,
      "step": 1594,
      "time_per_iteration": 2.5351474285125732
    },
    {
      "auxiliary_loss_clip": 0.01169788,
      "auxiliary_loss_mlp": 0.01054699,
      "balance_loss_clip": 1.05502379,
      "balance_loss_mlp": 1.03364682,
      "epoch": 0.0958965880054111,
      "flos": 23218367345280.0,
      "grad_norm": 2.287325745684747,
      "language_loss": 0.8221311,
      "learning_rate": 3.954622408410747e-06,
      "loss": 0.84437597,
      "num_input_tokens_seen": 34275970,
      "step": 1595,
      "time_per_iteration": 2.5167031288146973
    },
    {
      "auxiliary_loss_clip": 0.01151809,
      "auxiliary_loss_mlp": 0.01053354,
      "balance_loss_clip": 1.05359948,
      "balance_loss_mlp": 1.03007281,
      "epoch": 0.09595671125807906,
      "flos": 21324438933120.0,
      "grad_norm": 1.9629308948334019,
      "language_loss": 0.8493253,
      "learning_rate": 3.954539880085045e-06,
      "loss": 0.87137687,
      "num_input_tokens_seen": 34295490,
      "step": 1596,
      "time_per_iteration": 2.5645833015441895
    },
    {
      "auxiliary_loss_clip": 0.01165085,
      "auxiliary_loss_mlp": 0.01059292,
      "balance_loss_clip": 1.05610669,
      "balance_loss_mlp": 1.03638041,
      "epoch": 0.09601683451074704,
      "flos": 39603472185600.0,
      "grad_norm": 1.9832210066634568,
      "language_loss": 0.69044495,
      "learning_rate": 3.9544572776429945e-06,
      "loss": 0.71268868,
      "num_input_tokens_seen": 34319990,
      "step": 1597,
      "time_per_iteration": 2.695521593093872
    },
    {
      "auxiliary_loss_clip": 0.01165789,
      "auxiliary_loss_mlp": 0.01074746,
      "balance_loss_clip": 1.05419123,
      "balance_loss_mlp": 1.54365134,
      "epoch": 0.096076957763415,
      "flos": 23732716147200.0,
      "grad_norm": 1.9023679787666776,
      "language_loss": 0.7486822,
      "learning_rate": 3.954374601087729e-06,
      "loss": 0.77108759,
      "num_input_tokens_seen": 34339225,
      "step": 1598,
      "time_per_iteration": 2.5535457134246826
    },
    {
      "auxiliary_loss_clip": 0.01175142,
      "auxiliary_loss_mlp": 0.01061279,
      "balance_loss_clip": 1.05916417,
      "balance_loss_mlp": 1.03793812,
      "epoch": 0.09613708101608297,
      "flos": 34678108483200.0,
      "grad_norm": 1.6904192005081171,
      "language_loss": 0.68824667,
      "learning_rate": 3.954291850422382e-06,
      "loss": 0.71061087,
      "num_input_tokens_seen": 34361020,
      "step": 1599,
      "time_per_iteration": 2.614008665084839
    },
    {
      "auxiliary_loss_clip": 0.01149529,
      "auxiliary_loss_mlp": 0.01061083,
      "balance_loss_clip": 1.05227804,
      "balance_loss_mlp": 1.03987575,
      "epoch": 0.09619720426875093,
      "flos": 20740028653440.0,
      "grad_norm": 2.1423247440638526,
      "language_loss": 0.83919674,
      "learning_rate": 3.954209025650093e-06,
      "loss": 0.86130285,
      "num_input_tokens_seen": 34378630,
      "step": 1600,
      "time_per_iteration": 2.6057565212249756
    },
    {
      "auxiliary_loss_clip": 0.01148865,
      "auxiliary_loss_mlp": 0.01056708,
      "balance_loss_clip": 1.05405331,
      "balance_loss_mlp": 1.03501201,
      "epoch": 0.09625732752141891,
      "flos": 13042720488960.0,
      "grad_norm": 2.1783865083936593,
      "language_loss": 0.80323339,
      "learning_rate": 3.954126126774001e-06,
      "loss": 0.82528913,
      "num_input_tokens_seen": 34397110,
      "step": 1601,
      "time_per_iteration": 2.532188653945923
    },
    {
      "auxiliary_loss_clip": 0.01174786,
      "auxiliary_loss_mlp": 0.01056368,
      "balance_loss_clip": 1.05403721,
      "balance_loss_mlp": 1.03412318,
      "epoch": 0.09631745077408688,
      "flos": 22273629782400.0,
      "grad_norm": 2.158997287225043,
      "language_loss": 0.82530141,
      "learning_rate": 3.954043153797251e-06,
      "loss": 0.84761298,
      "num_input_tokens_seen": 34414165,
      "step": 1602,
      "time_per_iteration": 2.5283467769622803
    },
    {
      "auxiliary_loss_clip": 0.01137055,
      "auxiliary_loss_mlp": 0.01059697,
      "balance_loss_clip": 1.05501747,
      "balance_loss_mlp": 1.0367136,
      "epoch": 0.09637757402675484,
      "flos": 24754266944640.0,
      "grad_norm": 6.315069477407284,
      "language_loss": 0.62665462,
      "learning_rate": 3.953960106722989e-06,
      "loss": 0.6486221,
      "num_input_tokens_seen": 34434445,
      "step": 1603,
      "time_per_iteration": 2.602215528488159
    },
    {
      "auxiliary_loss_clip": 0.01186595,
      "auxiliary_loss_mlp": 0.01054838,
      "balance_loss_clip": 1.0571804,
      "balance_loss_mlp": 1.03071058,
      "epoch": 0.09643769727942282,
      "flos": 22525758322560.0,
      "grad_norm": 2.116229904256095,
      "language_loss": 0.71335554,
      "learning_rate": 3.953876985554364e-06,
      "loss": 0.73576987,
      "num_input_tokens_seen": 34453095,
      "step": 1604,
      "time_per_iteration": 2.5427417755126953
    },
    {
      "auxiliary_loss_clip": 0.01171401,
      "auxiliary_loss_mlp": 0.01060388,
      "balance_loss_clip": 1.05571413,
      "balance_loss_mlp": 1.03951478,
      "epoch": 0.09649782053209079,
      "flos": 30921026636160.0,
      "grad_norm": 2.0722582744854856,
      "language_loss": 0.80024838,
      "learning_rate": 3.953793790294527e-06,
      "loss": 0.82256627,
      "num_input_tokens_seen": 34473680,
      "step": 1605,
      "time_per_iteration": 2.5889570713043213
    },
    {
      "auxiliary_loss_clip": 0.0115893,
      "auxiliary_loss_mlp": 0.01048274,
      "balance_loss_clip": 1.05538893,
      "balance_loss_mlp": 1.02639949,
      "epoch": 0.09655794378475875,
      "flos": 25337635729920.0,
      "grad_norm": 2.23233270344022,
      "language_loss": 0.74762869,
      "learning_rate": 3.953710520946634e-06,
      "loss": 0.76970071,
      "num_input_tokens_seen": 34492610,
      "step": 1606,
      "time_per_iteration": 2.5858347415924072
    },
    {
      "auxiliary_loss_clip": 0.011657,
      "auxiliary_loss_mlp": 0.0105088,
      "balance_loss_clip": 1.05459559,
      "balance_loss_mlp": 1.03000677,
      "epoch": 0.09661806703742673,
      "flos": 22346061557760.0,
      "grad_norm": 2.191371122763678,
      "language_loss": 0.75749904,
      "learning_rate": 3.953627177513843e-06,
      "loss": 0.77966481,
      "num_input_tokens_seen": 34511855,
      "step": 1607,
      "time_per_iteration": 2.4936752319335938
    },
    {
      "auxiliary_loss_clip": 0.01137694,
      "auxiliary_loss_mlp": 0.01047739,
      "balance_loss_clip": 1.05233049,
      "balance_loss_mlp": 1.02626967,
      "epoch": 0.0966781902900947,
      "flos": 17457578144640.0,
      "grad_norm": 2.0697676439345405,
      "language_loss": 0.86718303,
      "learning_rate": 3.953543759999312e-06,
      "loss": 0.88903737,
      "num_input_tokens_seen": 34528905,
      "step": 1608,
      "time_per_iteration": 2.5565075874328613
    },
    {
      "auxiliary_loss_clip": 0.01123314,
      "auxiliary_loss_mlp": 0.01062524,
      "balance_loss_clip": 1.05569077,
      "balance_loss_mlp": 1.03856277,
      "epoch": 0.09673831354276266,
      "flos": 36903995412480.0,
      "grad_norm": 2.420247121031451,
      "language_loss": 0.71316564,
      "learning_rate": 3.953460268406207e-06,
      "loss": 0.73502409,
      "num_input_tokens_seen": 34548480,
      "step": 1609,
      "time_per_iteration": 2.7789041996002197
    },
    {
      "auxiliary_loss_clip": 0.01145896,
      "auxiliary_loss_mlp": 0.01055898,
      "balance_loss_clip": 1.05397773,
      "balance_loss_mlp": 1.03476238,
      "epoch": 0.09679843679543064,
      "flos": 20701388597760.0,
      "grad_norm": 2.333330129938826,
      "language_loss": 0.84379005,
      "learning_rate": 3.953376702737693e-06,
      "loss": 0.86580789,
      "num_input_tokens_seen": 34565410,
      "step": 1610,
      "time_per_iteration": 2.5723299980163574
    },
    {
      "auxiliary_loss_clip": 0.01155982,
      "auxiliary_loss_mlp": 0.01053808,
      "balance_loss_clip": 1.05507839,
      "balance_loss_mlp": 1.03132546,
      "epoch": 0.0968585600480986,
      "flos": 23514415240320.0,
      "grad_norm": 3.1224342164307877,
      "language_loss": 0.6686992,
      "learning_rate": 3.953293062996939e-06,
      "loss": 0.69079709,
      "num_input_tokens_seen": 34584840,
      "step": 1611,
      "time_per_iteration": 2.566049337387085
    },
    {
      "auxiliary_loss_clip": 0.01124173,
      "auxiliary_loss_mlp": 0.01052381,
      "balance_loss_clip": 1.05195558,
      "balance_loss_mlp": 1.0310303,
      "epoch": 0.09691868330076657,
      "flos": 20121072468480.0,
      "grad_norm": 2.2670617070606705,
      "language_loss": 0.81127125,
      "learning_rate": 3.953209349187115e-06,
      "loss": 0.83303678,
      "num_input_tokens_seen": 34603360,
      "step": 1612,
      "time_per_iteration": 2.6863596439361572
    },
    {
      "auxiliary_loss_clip": 0.0117788,
      "auxiliary_loss_mlp": 0.01066505,
      "balance_loss_clip": 1.05754912,
      "balance_loss_mlp": 1.04468942,
      "epoch": 0.09697880655343454,
      "flos": 16544692967040.0,
      "grad_norm": 2.1255997082134566,
      "language_loss": 0.81140232,
      "learning_rate": 3.953125561311398e-06,
      "loss": 0.83384615,
      "num_input_tokens_seen": 34620760,
      "step": 1613,
      "time_per_iteration": 3.894010543823242
    },
    {
      "auxiliary_loss_clip": 0.01148982,
      "auxiliary_loss_mlp": 0.01054779,
      "balance_loss_clip": 1.0593195,
      "balance_loss_mlp": 1.03185511,
      "epoch": 0.09703892980610251,
      "flos": 26104184899200.0,
      "grad_norm": 1.9118217132781186,
      "language_loss": 0.84393263,
      "learning_rate": 3.953041699372964e-06,
      "loss": 0.86597025,
      "num_input_tokens_seen": 34640695,
      "step": 1614,
      "time_per_iteration": 2.607609510421753
    },
    {
      "auxiliary_loss_clip": 0.01070613,
      "auxiliary_loss_mlp": 0.02214933,
      "balance_loss_clip": 1.02780175,
      "balance_loss_mlp": 3.5985992,
      "epoch": 0.09709905305877048,
      "flos": 60443622000000.0,
      "grad_norm": 0.7249257101515483,
      "language_loss": 0.54638219,
      "learning_rate": 3.952957763374992e-06,
      "loss": 0.57923758,
      "num_input_tokens_seen": 34702395,
      "step": 1615,
      "time_per_iteration": 3.0793280601501465
    },
    {
      "auxiliary_loss_clip": 0.01030188,
      "auxiliary_loss_mlp": 0.01012207,
      "balance_loss_clip": 1.02454853,
      "balance_loss_mlp": 1.00896442,
      "epoch": 0.09715917631143844,
      "flos": 57639932893440.0,
      "grad_norm": 0.7640146911246476,
      "language_loss": 0.58249009,
      "learning_rate": 3.952873753320666e-06,
      "loss": 0.60291409,
      "num_input_tokens_seen": 34768910,
      "step": 1616,
      "time_per_iteration": 4.721745491027832
    },
    {
      "auxiliary_loss_clip": 0.01151384,
      "auxiliary_loss_mlp": 0.01057428,
      "balance_loss_clip": 1.05397773,
      "balance_loss_mlp": 1.03403914,
      "epoch": 0.09721929956410642,
      "flos": 20558212986240.0,
      "grad_norm": 1.7186798916040393,
      "language_loss": 0.68864793,
      "learning_rate": 3.952789669213172e-06,
      "loss": 0.7107361,
      "num_input_tokens_seen": 34787680,
      "step": 1617,
      "time_per_iteration": 2.5400750637054443
    },
    {
      "auxiliary_loss_clip": 0.0114929,
      "auxiliary_loss_mlp": 0.01056019,
      "balance_loss_clip": 1.05281448,
      "balance_loss_mlp": 1.03066313,
      "epoch": 0.09727942281677439,
      "flos": 27344359825920.0,
      "grad_norm": 1.727278229135788,
      "language_loss": 0.80796432,
      "learning_rate": 3.952705511055698e-06,
      "loss": 0.83001745,
      "num_input_tokens_seen": 34808330,
      "step": 1618,
      "time_per_iteration": 2.623757839202881
    },
    {
      "auxiliary_loss_clip": 0.0115949,
      "auxiliary_loss_mlp": 0.01050074,
      "balance_loss_clip": 1.05306888,
      "balance_loss_mlp": 1.02928352,
      "epoch": 0.09733954606944235,
      "flos": 24900028335360.0,
      "grad_norm": 2.1935990150563245,
      "language_loss": 0.93155861,
      "learning_rate": 3.952621278851435e-06,
      "loss": 0.95365417,
      "num_input_tokens_seen": 34830020,
      "step": 1619,
      "time_per_iteration": 2.5793087482452393
    },
    {
      "auxiliary_loss_clip": 0.01169499,
      "auxiliary_loss_mlp": 0.01052788,
      "balance_loss_clip": 1.05496585,
      "balance_loss_mlp": 1.03147376,
      "epoch": 0.09739966932211033,
      "flos": 31503928544640.0,
      "grad_norm": 1.8676591550611403,
      "language_loss": 0.88680565,
      "learning_rate": 3.9525369726035784e-06,
      "loss": 0.90902853,
      "num_input_tokens_seen": 34850330,
      "step": 1620,
      "time_per_iteration": 4.057949066162109
    },
    {
      "auxiliary_loss_clip": 0.01150341,
      "auxiliary_loss_mlp": 0.01058617,
      "balance_loss_clip": 1.05212879,
      "balance_loss_mlp": 1.03391731,
      "epoch": 0.0974597925747783,
      "flos": 23878764846720.0,
      "grad_norm": 2.191948545924734,
      "language_loss": 0.77567935,
      "learning_rate": 3.952452592315324e-06,
      "loss": 0.79776895,
      "num_input_tokens_seen": 34871640,
      "step": 1621,
      "time_per_iteration": 2.5788774490356445
    },
    {
      "auxiliary_loss_clip": 0.01128641,
      "auxiliary_loss_mlp": 0.01065113,
      "balance_loss_clip": 1.04922485,
      "balance_loss_mlp": 1.03981721,
      "epoch": 0.09751991582744626,
      "flos": 17019575700480.0,
      "grad_norm": 1.9612774912704087,
      "language_loss": 0.77619147,
      "learning_rate": 3.952368137989871e-06,
      "loss": 0.79812902,
      "num_input_tokens_seen": 34888100,
      "step": 1622,
      "time_per_iteration": 2.541710138320923
    },
    {
      "auxiliary_loss_clip": 0.01147763,
      "auxiliary_loss_mlp": 0.01059617,
      "balance_loss_clip": 1.05404973,
      "balance_loss_mlp": 1.03686035,
      "epoch": 0.09758003908011423,
      "flos": 28402826826240.0,
      "grad_norm": 3.179270059110057,
      "language_loss": 0.85670602,
      "learning_rate": 3.9522836096304225e-06,
      "loss": 0.87877989,
      "num_input_tokens_seen": 34910485,
      "step": 1623,
      "time_per_iteration": 3.9655256271362305
    },
    {
      "auxiliary_loss_clip": 0.01170271,
      "auxiliary_loss_mlp": 0.01060101,
      "balance_loss_clip": 1.05371916,
      "balance_loss_mlp": 1.03765368,
      "epoch": 0.09764016233278221,
      "flos": 18144297336960.0,
      "grad_norm": 2.4996693519326283,
      "language_loss": 0.80522782,
      "learning_rate": 3.952199007240184e-06,
      "loss": 0.82753158,
      "num_input_tokens_seen": 34928615,
      "step": 1624,
      "time_per_iteration": 2.488090991973877
    },
    {
      "auxiliary_loss_clip": 0.01168107,
      "auxiliary_loss_mlp": 0.01047627,
      "balance_loss_clip": 1.04990554,
      "balance_loss_mlp": 1.02650356,
      "epoch": 0.09770028558545017,
      "flos": 15265842071040.0,
      "grad_norm": 2.738018976605643,
      "language_loss": 0.85630202,
      "learning_rate": 3.952114330822364e-06,
      "loss": 0.87845939,
      "num_input_tokens_seen": 34946045,
      "step": 1625,
      "time_per_iteration": 2.4655022621154785
    },
    {
      "auxiliary_loss_clip": 0.01173414,
      "auxiliary_loss_mlp": 0.01052354,
      "balance_loss_clip": 1.05291343,
      "balance_loss_mlp": 1.0308249,
      "epoch": 0.09776040883811814,
      "flos": 23472435219840.0,
      "grad_norm": 2.660910616132785,
      "language_loss": 0.85332584,
      "learning_rate": 3.952029580380172e-06,
      "loss": 0.87558353,
      "num_input_tokens_seen": 34962865,
      "step": 1626,
      "time_per_iteration": 2.5363078117370605
    },
    {
      "auxiliary_loss_clip": 0.01162048,
      "auxiliary_loss_mlp": 0.01501221,
      "balance_loss_clip": 1.05466771,
      "balance_loss_mlp": 2.35174513,
      "epoch": 0.09782053209078612,
      "flos": 24499480798080.0,
      "grad_norm": 2.009905442177808,
      "language_loss": 0.83585185,
      "learning_rate": 3.9519447559168234e-06,
      "loss": 0.86248451,
      "num_input_tokens_seen": 34983505,
      "step": 1627,
      "time_per_iteration": 2.539125919342041
    },
    {
      "auxiliary_loss_clip": 0.01162899,
      "auxiliary_loss_mlp": 0.01051954,
      "balance_loss_clip": 1.05166316,
      "balance_loss_mlp": 1.03059149,
      "epoch": 0.09788065534345408,
      "flos": 21580158833280.0,
      "grad_norm": 2.6464265337645347,
      "language_loss": 0.8433156,
      "learning_rate": 3.951859857435534e-06,
      "loss": 0.86546409,
      "num_input_tokens_seen": 35001825,
      "step": 1628,
      "time_per_iteration": 2.5112733840942383
    },
    {
      "auxiliary_loss_clip": 0.01165165,
      "auxiliary_loss_mlp": 0.01053007,
      "balance_loss_clip": 1.05013406,
      "balance_loss_mlp": 1.03120339,
      "epoch": 0.09794077859612205,
      "flos": 23842459175040.0,
      "grad_norm": 1.8655836334929567,
      "language_loss": 0.75802571,
      "learning_rate": 3.951774884939523e-06,
      "loss": 0.7802074,
      "num_input_tokens_seen": 35023075,
      "step": 1629,
      "time_per_iteration": 2.515502691268921
    },
    {
      "auxiliary_loss_clip": 0.01125973,
      "auxiliary_loss_mlp": 0.01055712,
      "balance_loss_clip": 1.05820298,
      "balance_loss_mlp": 1.03258514,
      "epoch": 0.09800090184879003,
      "flos": 23659889322240.0,
      "grad_norm": 1.7160384589155977,
      "language_loss": 0.78478312,
      "learning_rate": 3.951689838432013e-06,
      "loss": 0.80659997,
      "num_input_tokens_seen": 35043480,
      "step": 1630,
      "time_per_iteration": 2.6961770057678223
    },
    {
      "auxiliary_loss_clip": 0.01162611,
      "auxiliary_loss_mlp": 0.01051676,
      "balance_loss_clip": 1.05648804,
      "balance_loss_mlp": 1.02873993,
      "epoch": 0.09806102510145799,
      "flos": 17055773631360.0,
      "grad_norm": 1.8993806658781305,
      "language_loss": 0.86484474,
      "learning_rate": 3.951604717916228e-06,
      "loss": 0.88698757,
      "num_input_tokens_seen": 35061490,
      "step": 1631,
      "time_per_iteration": 2.4997169971466064
    },
    {
      "auxiliary_loss_clip": 0.01162967,
      "auxiliary_loss_mlp": 0.01055001,
      "balance_loss_clip": 1.05410099,
      "balance_loss_mlp": 1.03421044,
      "epoch": 0.09812114835412596,
      "flos": 23878477537920.0,
      "grad_norm": 5.373839608041589,
      "language_loss": 0.83362824,
      "learning_rate": 3.9515195233953975e-06,
      "loss": 0.85580796,
      "num_input_tokens_seen": 35079670,
      "step": 1632,
      "time_per_iteration": 2.569882392883301
    },
    {
      "auxiliary_loss_clip": 0.01144978,
      "auxiliary_loss_mlp": 0.01055043,
      "balance_loss_clip": 1.05601931,
      "balance_loss_mlp": 1.03475344,
      "epoch": 0.09818127160679392,
      "flos": 20595488325120.0,
      "grad_norm": 1.7080570668151474,
      "language_loss": 0.78590512,
      "learning_rate": 3.951434254872751e-06,
      "loss": 0.80790532,
      "num_input_tokens_seen": 35099205,
      "step": 1633,
      "time_per_iteration": 2.575016736984253
    },
    {
      "auxiliary_loss_clip": 0.01162457,
      "auxiliary_loss_mlp": 0.01056445,
      "balance_loss_clip": 1.04965949,
      "balance_loss_mlp": 1.03429639,
      "epoch": 0.0982413948594619,
      "flos": 15487339288320.0,
      "grad_norm": 2.024162239931351,
      "language_loss": 0.72985095,
      "learning_rate": 3.951348912351521e-06,
      "loss": 0.75203991,
      "num_input_tokens_seen": 35115270,
      "step": 1634,
      "time_per_iteration": 2.4762630462646484
    },
    {
      "auxiliary_loss_clip": 0.0115223,
      "auxiliary_loss_mlp": 0.01058461,
      "balance_loss_clip": 1.05130434,
      "balance_loss_mlp": 1.0361805,
      "epoch": 0.09830151811212987,
      "flos": 24207958016640.0,
      "grad_norm": 2.581790637599988,
      "language_loss": 0.73380458,
      "learning_rate": 3.951263495834947e-06,
      "loss": 0.75591147,
      "num_input_tokens_seen": 35134065,
      "step": 1635,
      "time_per_iteration": 2.5592756271362305
    },
    {
      "auxiliary_loss_clip": 0.01150834,
      "auxiliary_loss_mlp": 0.01065018,
      "balance_loss_clip": 1.05477858,
      "balance_loss_mlp": 1.04034221,
      "epoch": 0.09836164136479783,
      "flos": 20594590485120.0,
      "grad_norm": 1.8301700929490785,
      "language_loss": 0.77896667,
      "learning_rate": 3.951178005326264e-06,
      "loss": 0.80112517,
      "num_input_tokens_seen": 35154870,
      "step": 1636,
      "time_per_iteration": 2.6120121479034424
    },
    {
      "auxiliary_loss_clip": 0.01159528,
      "auxiliary_loss_mlp": 0.01056802,
      "balance_loss_clip": 1.05581856,
      "balance_loss_mlp": 1.03565395,
      "epoch": 0.09842176461746581,
      "flos": 19934157070080.0,
      "grad_norm": 1.92287549066641,
      "language_loss": 0.69749123,
      "learning_rate": 3.951092440828715e-06,
      "loss": 0.71965456,
      "num_input_tokens_seen": 35171850,
      "step": 1637,
      "time_per_iteration": 2.5223934650421143
    },
    {
      "auxiliary_loss_clip": 0.01181545,
      "auxiliary_loss_mlp": 0.01054405,
      "balance_loss_clip": 1.05375409,
      "balance_loss_mlp": 1.03284025,
      "epoch": 0.09848188787013377,
      "flos": 21214659991680.0,
      "grad_norm": 2.756839637545218,
      "language_loss": 0.77671051,
      "learning_rate": 3.951006802345545e-06,
      "loss": 0.79907012,
      "num_input_tokens_seen": 35188795,
      "step": 1638,
      "time_per_iteration": 2.496187210083008
    },
    {
      "auxiliary_loss_clip": 0.0113475,
      "auxiliary_loss_mlp": 0.01047383,
      "balance_loss_clip": 1.05406177,
      "balance_loss_mlp": 1.02644968,
      "epoch": 0.09854201112280174,
      "flos": 30154226071680.0,
      "grad_norm": 1.5656149296689226,
      "language_loss": 0.72798473,
      "learning_rate": 3.950921089880003e-06,
      "loss": 0.74980605,
      "num_input_tokens_seen": 35212100,
      "step": 1639,
      "time_per_iteration": 2.6436333656311035
    },
    {
      "auxiliary_loss_clip": 0.01166791,
      "auxiliary_loss_mlp": 0.01048613,
      "balance_loss_clip": 1.05338788,
      "balance_loss_mlp": 1.02711964,
      "epoch": 0.09860213437546972,
      "flos": 21795730306560.0,
      "grad_norm": 1.9801089851135547,
      "language_loss": 0.88515592,
      "learning_rate": 3.950835303435337e-06,
      "loss": 0.90731001,
      "num_input_tokens_seen": 35230390,
      "step": 1640,
      "time_per_iteration": 2.5419862270355225
    },
    {
      "auxiliary_loss_clip": 0.01174093,
      "auxiliary_loss_mlp": 0.01044783,
      "balance_loss_clip": 1.05824256,
      "balance_loss_mlp": 1.02444577,
      "epoch": 0.09866225762813768,
      "flos": 21835555511040.0,
      "grad_norm": 1.9201905273997841,
      "language_loss": 0.80891514,
      "learning_rate": 3.950749443014801e-06,
      "loss": 0.83110392,
      "num_input_tokens_seen": 35250405,
      "step": 1641,
      "time_per_iteration": 2.4970133304595947
    },
    {
      "auxiliary_loss_clip": 0.01166419,
      "auxiliary_loss_mlp": 0.01061047,
      "balance_loss_clip": 1.05172825,
      "balance_loss_mlp": 1.03870702,
      "epoch": 0.09872238088080565,
      "flos": 17599855916160.0,
      "grad_norm": 2.6159265104882685,
      "language_loss": 0.86033541,
      "learning_rate": 3.95066350862165e-06,
      "loss": 0.88261008,
      "num_input_tokens_seen": 35262820,
      "step": 1642,
      "time_per_iteration": 2.484950542449951
    },
    {
      "auxiliary_loss_clip": 0.01142692,
      "auxiliary_loss_mlp": 0.01054525,
      "balance_loss_clip": 1.05469298,
      "balance_loss_mlp": 1.03369951,
      "epoch": 0.09878250413347361,
      "flos": 27636134002560.0,
      "grad_norm": 1.7801416034561532,
      "language_loss": 0.80831039,
      "learning_rate": 3.950577500259144e-06,
      "loss": 0.83028257,
      "num_input_tokens_seen": 35284490,
      "step": 1643,
      "time_per_iteration": 2.605327844619751
    },
    {
      "auxiliary_loss_clip": 0.01171143,
      "auxiliary_loss_mlp": 0.01073007,
      "balance_loss_clip": 1.05660951,
      "balance_loss_mlp": 1.05150211,
      "epoch": 0.0988426273861416,
      "flos": 16544728880640.0,
      "grad_norm": 1.9317255457843905,
      "language_loss": 0.82200927,
      "learning_rate": 3.950491417930543e-06,
      "loss": 0.84445071,
      "num_input_tokens_seen": 35302815,
      "step": 1644,
      "time_per_iteration": 2.498096227645874
    },
    {
      "auxiliary_loss_clip": 0.0115706,
      "auxiliary_loss_mlp": 0.01107175,
      "balance_loss_clip": 1.05251122,
      "balance_loss_mlp": 1.59641171,
      "epoch": 0.09890275063880956,
      "flos": 21215270522880.0,
      "grad_norm": 1.8638243837790351,
      "language_loss": 0.68059838,
      "learning_rate": 3.9504052616391124e-06,
      "loss": 0.70324075,
      "num_input_tokens_seen": 35321175,
      "step": 1645,
      "time_per_iteration": 2.523047924041748
    },
    {
      "auxiliary_loss_clip": 0.01056882,
      "auxiliary_loss_mlp": 0.01014486,
      "balance_loss_clip": 1.02419293,
      "balance_loss_mlp": 1.01117182,
      "epoch": 0.09896287389147752,
      "flos": 59379372910080.0,
      "grad_norm": 0.855389406388609,
      "language_loss": 0.60874653,
      "learning_rate": 3.950319031388119e-06,
      "loss": 0.62946022,
      "num_input_tokens_seen": 35381740,
      "step": 1646,
      "time_per_iteration": 3.0704548358917236
    },
    {
      "auxiliary_loss_clip": 0.01143906,
      "auxiliary_loss_mlp": 0.01055024,
      "balance_loss_clip": 1.05667746,
      "balance_loss_mlp": 1.03257728,
      "epoch": 0.0990229971441455,
      "flos": 29642678530560.0,
      "grad_norm": 2.3027520038816327,
      "language_loss": 0.73075736,
      "learning_rate": 3.950232727180833e-06,
      "loss": 0.7527467,
      "num_input_tokens_seen": 35403760,
      "step": 1647,
      "time_per_iteration": 2.628739833831787
    },
    {
      "auxiliary_loss_clip": 0.01153838,
      "auxiliary_loss_mlp": 0.01060916,
      "balance_loss_clip": 1.05322778,
      "balance_loss_mlp": 1.04082918,
      "epoch": 0.09908312039681347,
      "flos": 21834873152640.0,
      "grad_norm": 2.051447786257446,
      "language_loss": 0.84169978,
      "learning_rate": 3.950146349020525e-06,
      "loss": 0.86384732,
      "num_input_tokens_seen": 35424050,
      "step": 1648,
      "time_per_iteration": 2.5864462852478027
    },
    {
      "auxiliary_loss_clip": 0.0106881,
      "auxiliary_loss_mlp": 0.01005105,
      "balance_loss_clip": 1.03026688,
      "balance_loss_mlp": 1.00176704,
      "epoch": 0.09914324364948143,
      "flos": 57564304807680.0,
      "grad_norm": 0.726670937568241,
      "language_loss": 0.55669916,
      "learning_rate": 3.950059896910473e-06,
      "loss": 0.57743835,
      "num_input_tokens_seen": 35481690,
      "step": 1649,
      "time_per_iteration": 3.0419468879699707
    },
    {
      "auxiliary_loss_clip": 0.01165643,
      "auxiliary_loss_mlp": 0.01046315,
      "balance_loss_clip": 1.05141282,
      "balance_loss_mlp": 1.02547681,
      "epoch": 0.09920336690214941,
      "flos": 34123934476800.0,
      "grad_norm": 2.623191987484333,
      "language_loss": 0.89675653,
      "learning_rate": 3.949973370853954e-06,
      "loss": 0.91887611,
      "num_input_tokens_seen": 35498635,
      "step": 1650,
      "time_per_iteration": 2.588137149810791
    },
    {
      "auxiliary_loss_clip": 0.01036032,
      "auxiliary_loss_mlp": 0.0144039,
      "balance_loss_clip": 1.03135657,
      "balance_loss_mlp": 2.24221325,
      "epoch": 0.09926349015481738,
      "flos": 71216428464000.0,
      "grad_norm": 0.9358428949393979,
      "language_loss": 0.63712549,
      "learning_rate": 3.94988677085425e-06,
      "loss": 0.66188967,
      "num_input_tokens_seen": 35565720,
      "step": 1651,
      "time_per_iteration": 3.332594394683838
    },
    {
      "auxiliary_loss_clip": 0.0116121,
      "auxiliary_loss_mlp": 0.01059745,
      "balance_loss_clip": 1.05166888,
      "balance_loss_mlp": 1.03765523,
      "epoch": 0.09932361340748534,
      "flos": 23148700917120.0,
      "grad_norm": 2.1162862688045454,
      "language_loss": 0.87910157,
      "learning_rate": 3.949800096914643e-06,
      "loss": 0.90131104,
      "num_input_tokens_seen": 35586000,
      "step": 1652,
      "time_per_iteration": 3.9281647205352783
    },
    {
      "auxiliary_loss_clip": 0.01159769,
      "auxiliary_loss_mlp": 0.01055081,
      "balance_loss_clip": 1.05604231,
      "balance_loss_mlp": 1.03423131,
      "epoch": 0.09938373666015332,
      "flos": 19828651847040.0,
      "grad_norm": 1.9661125049244077,
      "language_loss": 0.82249516,
      "learning_rate": 3.949713349038422e-06,
      "loss": 0.84464371,
      "num_input_tokens_seen": 35604355,
      "step": 1653,
      "time_per_iteration": 2.53544282913208
    },
    {
      "auxiliary_loss_clip": 0.01167622,
      "auxiliary_loss_mlp": 0.01308306,
      "balance_loss_clip": 1.05234385,
      "balance_loss_mlp": 1.98688579,
      "epoch": 0.09944385991282129,
      "flos": 22090664880000.0,
      "grad_norm": 2.0037517690098094,
      "language_loss": 0.7959438,
      "learning_rate": 3.949626527228875e-06,
      "loss": 0.82070303,
      "num_input_tokens_seen": 35625495,
      "step": 1654,
      "time_per_iteration": 2.5702474117279053
    },
    {
      "auxiliary_loss_clip": 0.01181917,
      "auxiliary_loss_mlp": 0.01060069,
      "balance_loss_clip": 1.05903685,
      "balance_loss_mlp": 1.03956532,
      "epoch": 0.09950398316548925,
      "flos": 19828867328640.0,
      "grad_norm": 2.0621416000831823,
      "language_loss": 0.80815011,
      "learning_rate": 3.949539631489295e-06,
      "loss": 0.83057004,
      "num_input_tokens_seen": 35645030,
      "step": 1655,
      "time_per_iteration": 3.868032932281494
    },
    {
      "auxiliary_loss_clip": 0.01175547,
      "auxiliary_loss_mlp": 0.01057891,
      "balance_loss_clip": 1.05241048,
      "balance_loss_mlp": 1.03714871,
      "epoch": 0.09956410641815722,
      "flos": 25003701964800.0,
      "grad_norm": 1.8608580094479896,
      "language_loss": 0.80454743,
      "learning_rate": 3.9494526618229765e-06,
      "loss": 0.82688177,
      "num_input_tokens_seen": 35664305,
      "step": 1656,
      "time_per_iteration": 2.51297664642334
    },
    {
      "auxiliary_loss_clip": 0.01166463,
      "auxiliary_loss_mlp": 0.01057007,
      "balance_loss_clip": 1.05555654,
      "balance_loss_mlp": 1.03581166,
      "epoch": 0.0996242296708252,
      "flos": 19317714837120.0,
      "grad_norm": 1.770071720689501,
      "language_loss": 0.89150345,
      "learning_rate": 3.949365618233217e-06,
      "loss": 0.91373825,
      "num_input_tokens_seen": 35684060,
      "step": 1657,
      "time_per_iteration": 2.4929065704345703
    },
    {
      "auxiliary_loss_clip": 0.01160839,
      "auxiliary_loss_mlp": 0.01058143,
      "balance_loss_clip": 1.05551374,
      "balance_loss_mlp": 1.03595853,
      "epoch": 0.09968435292349316,
      "flos": 21871609787520.0,
      "grad_norm": 2.4592786444855115,
      "language_loss": 0.84841943,
      "learning_rate": 3.9492785007233195e-06,
      "loss": 0.87060928,
      "num_input_tokens_seen": 35703250,
      "step": 1658,
      "time_per_iteration": 2.545592784881592
    },
    {
      "auxiliary_loss_clip": 0.01075334,
      "auxiliary_loss_mlp": 0.01011324,
      "balance_loss_clip": 1.02364349,
      "balance_loss_mlp": 1.00805748,
      "epoch": 0.09974447617616113,
      "flos": 65384533313280.0,
      "grad_norm": 0.9011613202575146,
      "language_loss": 0.60811186,
      "learning_rate": 3.949191309296585e-06,
      "loss": 0.62897837,
      "num_input_tokens_seen": 35762165,
      "step": 1659,
      "time_per_iteration": 4.476860523223877
    },
    {
      "auxiliary_loss_clip": 0.01152187,
      "auxiliary_loss_mlp": 0.01056904,
      "balance_loss_clip": 1.05162883,
      "balance_loss_mlp": 1.03521955,
      "epoch": 0.0998045994288291,
      "flos": 23659817495040.0,
      "grad_norm": 1.8827729914234468,
      "language_loss": 0.85034382,
      "learning_rate": 3.949104043956321e-06,
      "loss": 0.87243474,
      "num_input_tokens_seen": 35781520,
      "step": 1660,
      "time_per_iteration": 2.546563148498535
    },
    {
      "auxiliary_loss_clip": 0.01148782,
      "auxiliary_loss_mlp": 0.01057632,
      "balance_loss_clip": 1.05674124,
      "balance_loss_mlp": 1.03440952,
      "epoch": 0.09986472268149707,
      "flos": 19609704495360.0,
      "grad_norm": 1.9969897590365777,
      "language_loss": 0.79797053,
      "learning_rate": 3.949016704705836e-06,
      "loss": 0.82003462,
      "num_input_tokens_seen": 35799565,
      "step": 1661,
      "time_per_iteration": 2.5231924057006836
    },
    {
      "auxiliary_loss_clip": 0.0116542,
      "auxiliary_loss_mlp": 0.01054105,
      "balance_loss_clip": 1.05111098,
      "balance_loss_mlp": 1.03186059,
      "epoch": 0.09992484593416504,
      "flos": 26213317395840.0,
      "grad_norm": 2.1468896675873674,
      "language_loss": 0.8371197,
      "learning_rate": 3.948929291548443e-06,
      "loss": 0.85931492,
      "num_input_tokens_seen": 35821085,
      "step": 1662,
      "time_per_iteration": 3.9578418731689453
    },
    {
      "auxiliary_loss_clip": 0.01155283,
      "auxiliary_loss_mlp": 0.01064081,
      "balance_loss_clip": 1.0523982,
      "balance_loss_mlp": 1.04015541,
      "epoch": 0.09998496918683301,
      "flos": 17493632421120.0,
      "grad_norm": 2.3259479624722097,
      "language_loss": 0.89115012,
      "learning_rate": 3.9488418044874546e-06,
      "loss": 0.91334379,
      "num_input_tokens_seen": 35839840,
      "step": 1663,
      "time_per_iteration": 2.5194926261901855
    },
    {
      "auxiliary_loss_clip": 0.01175404,
      "auxiliary_loss_mlp": 0.01052199,
      "balance_loss_clip": 1.05684125,
      "balance_loss_mlp": 1.02998996,
      "epoch": 0.10004509243950098,
      "flos": 22784925928320.0,
      "grad_norm": 1.743401963197557,
      "language_loss": 0.70255482,
      "learning_rate": 3.948754243526191e-06,
      "loss": 0.72483081,
      "num_input_tokens_seen": 35861545,
      "step": 1664,
      "time_per_iteration": 2.5678064823150635
    },
    {
      "auxiliary_loss_clip": 0.01141779,
      "auxiliary_loss_mlp": 0.01049941,
      "balance_loss_clip": 1.05589592,
      "balance_loss_mlp": 1.02834034,
      "epoch": 0.10010521569216894,
      "flos": 16253385667200.0,
      "grad_norm": 2.263127252197607,
      "language_loss": 0.78512859,
      "learning_rate": 3.94866660866797e-06,
      "loss": 0.80704582,
      "num_input_tokens_seen": 35878295,
      "step": 1665,
      "time_per_iteration": 2.544363021850586
    },
    {
      "auxiliary_loss_clip": 0.01171428,
      "auxiliary_loss_mlp": 0.01063369,
      "balance_loss_clip": 1.06061125,
      "balance_loss_mlp": 1.04150605,
      "epoch": 0.10016533894483691,
      "flos": 23402589223680.0,
      "grad_norm": 2.65350261847189,
      "language_loss": 0.69971788,
      "learning_rate": 3.9485788999161165e-06,
      "loss": 0.72206593,
      "num_input_tokens_seen": 35898990,
      "step": 1666,
      "time_per_iteration": 2.5455663204193115
    },
    {
      "auxiliary_loss_clip": 0.01112771,
      "auxiliary_loss_mlp": 0.01063019,
      "balance_loss_clip": 1.05683374,
      "balance_loss_mlp": 1.03947532,
      "epoch": 0.10022546219750489,
      "flos": 19354164163200.0,
      "grad_norm": 1.9353468030300898,
      "language_loss": 0.78491497,
      "learning_rate": 3.948491117273956e-06,
      "loss": 0.80667293,
      "num_input_tokens_seen": 35916225,
      "step": 1667,
      "time_per_iteration": 2.6437623500823975
    },
    {
      "auxiliary_loss_clip": 0.01151876,
      "auxiliary_loss_mlp": 0.01055557,
      "balance_loss_clip": 1.05457985,
      "balance_loss_mlp": 1.03071368,
      "epoch": 0.10028558545017285,
      "flos": 27085766837760.0,
      "grad_norm": 2.3563942944329352,
      "language_loss": 0.76787591,
      "learning_rate": 3.948403260744817e-06,
      "loss": 0.78995025,
      "num_input_tokens_seen": 35934630,
      "step": 1668,
      "time_per_iteration": 2.592874050140381
    },
    {
      "auxiliary_loss_clip": 0.01182209,
      "auxiliary_loss_mlp": 0.01053109,
      "balance_loss_clip": 1.05670881,
      "balance_loss_mlp": 1.0311383,
      "epoch": 0.10034570870284082,
      "flos": 25847136195840.0,
      "grad_norm": 2.7386390687248694,
      "language_loss": 0.77872705,
      "learning_rate": 3.948315330332031e-06,
      "loss": 0.80108023,
      "num_input_tokens_seen": 35953855,
      "step": 1669,
      "time_per_iteration": 2.513627767562866
    },
    {
      "auxiliary_loss_clip": 0.01190107,
      "auxiliary_loss_mlp": 0.01057438,
      "balance_loss_clip": 1.06029034,
      "balance_loss_mlp": 1.03491926,
      "epoch": 0.1004058319555088,
      "flos": 26249587153920.0,
      "grad_norm": 2.0687039450440823,
      "language_loss": 0.85414052,
      "learning_rate": 3.948227326038933e-06,
      "loss": 0.876616,
      "num_input_tokens_seen": 35974555,
      "step": 1670,
      "time_per_iteration": 2.51472806930542
    },
    {
      "auxiliary_loss_clip": 0.01177051,
      "auxiliary_loss_mlp": 0.010486,
      "balance_loss_clip": 1.05490303,
      "balance_loss_mlp": 1.02771449,
      "epoch": 0.10046595520817676,
      "flos": 25374480105600.0,
      "grad_norm": 1.5443826482261862,
      "language_loss": 0.76658237,
      "learning_rate": 3.9481392478688586e-06,
      "loss": 0.7888388,
      "num_input_tokens_seen": 35996830,
      "step": 1671,
      "time_per_iteration": 2.51322603225708
    },
    {
      "auxiliary_loss_clip": 0.01065699,
      "auxiliary_loss_mlp": 0.01003461,
      "balance_loss_clip": 1.02352548,
      "balance_loss_mlp": 1.0006355,
      "epoch": 0.10052607846084473,
      "flos": 67461821677440.0,
      "grad_norm": 0.7823527675825388,
      "language_loss": 0.60696256,
      "learning_rate": 3.948051095825149e-06,
      "loss": 0.62765419,
      "num_input_tokens_seen": 36054465,
      "step": 1672,
      "time_per_iteration": 3.0984489917755127
    },
    {
      "auxiliary_loss_clip": 0.01141992,
      "auxiliary_loss_mlp": 0.01057888,
      "balance_loss_clip": 1.05360401,
      "balance_loss_mlp": 1.03542852,
      "epoch": 0.10058620171351271,
      "flos": 21360493209600.0,
      "grad_norm": 2.653941530334597,
      "language_loss": 0.76910186,
      "learning_rate": 3.947962869911147e-06,
      "loss": 0.79110068,
      "num_input_tokens_seen": 36073480,
      "step": 1673,
      "time_per_iteration": 2.54616117477417
    },
    {
      "auxiliary_loss_clip": 0.01133007,
      "auxiliary_loss_mlp": 0.01057777,
      "balance_loss_clip": 1.05265832,
      "balance_loss_mlp": 1.03467369,
      "epoch": 0.10064632496618067,
      "flos": 16800125558400.0,
      "grad_norm": 2.075012489307534,
      "language_loss": 0.73531806,
      "learning_rate": 3.947874570130197e-06,
      "loss": 0.75722593,
      "num_input_tokens_seen": 36091830,
      "step": 1674,
      "time_per_iteration": 2.5548598766326904
    },
    {
      "auxiliary_loss_clip": 0.01171794,
      "auxiliary_loss_mlp": 0.01780695,
      "balance_loss_clip": 1.05477774,
      "balance_loss_mlp": 2.85056043,
      "epoch": 0.10070644821884864,
      "flos": 23624445576960.0,
      "grad_norm": 1.947592461691345,
      "language_loss": 0.79441464,
      "learning_rate": 3.947786196485649e-06,
      "loss": 0.82393956,
      "num_input_tokens_seen": 36111400,
      "step": 1675,
      "time_per_iteration": 2.533487319946289
    },
    {
      "auxiliary_loss_clip": 0.01181756,
      "auxiliary_loss_mlp": 0.01066175,
      "balance_loss_clip": 1.05722833,
      "balance_loss_mlp": 1.0460403,
      "epoch": 0.1007665714715166,
      "flos": 24462564595200.0,
      "grad_norm": 2.374048784936178,
      "language_loss": 0.81413126,
      "learning_rate": 3.947697748980853e-06,
      "loss": 0.83661056,
      "num_input_tokens_seen": 36129345,
      "step": 1676,
      "time_per_iteration": 2.4949231147766113
    },
    {
      "auxiliary_loss_clip": 0.01173757,
      "auxiliary_loss_mlp": 0.01058342,
      "balance_loss_clip": 1.05773973,
      "balance_loss_mlp": 1.03689623,
      "epoch": 0.10082669472418458,
      "flos": 16799119977600.0,
      "grad_norm": 2.1661904412477426,
      "language_loss": 0.86184305,
      "learning_rate": 3.947609227619163e-06,
      "loss": 0.88416409,
      "num_input_tokens_seen": 36146255,
      "step": 1677,
      "time_per_iteration": 2.4750161170959473
    },
    {
      "auxiliary_loss_clip": 0.0116078,
      "auxiliary_loss_mlp": 0.01050818,
      "balance_loss_clip": 1.05268204,
      "balance_loss_mlp": 1.02905035,
      "epoch": 0.10088681797685255,
      "flos": 13553513844480.0,
      "grad_norm": 1.8735701724344236,
      "language_loss": 0.86215031,
      "learning_rate": 3.947520632403936e-06,
      "loss": 0.88426638,
      "num_input_tokens_seen": 36164050,
      "step": 1678,
      "time_per_iteration": 2.5260531902313232
    },
    {
      "auxiliary_loss_clip": 0.01161823,
      "auxiliary_loss_mlp": 0.0105341,
      "balance_loss_clip": 1.0594461,
      "balance_loss_mlp": 1.03174949,
      "epoch": 0.10094694122952051,
      "flos": 25265706744960.0,
      "grad_norm": 2.2348908032406563,
      "language_loss": 0.89885199,
      "learning_rate": 3.947431963338532e-06,
      "loss": 0.92100441,
      "num_input_tokens_seen": 36183530,
      "step": 1679,
      "time_per_iteration": 2.578390121459961
    },
    {
      "auxiliary_loss_clip": 0.01077165,
      "auxiliary_loss_mlp": 0.01005154,
      "balance_loss_clip": 1.02589488,
      "balance_loss_mlp": 1.00214982,
      "epoch": 0.10100706448218849,
      "flos": 69854299885440.0,
      "grad_norm": 0.7750616707489534,
      "language_loss": 0.52971166,
      "learning_rate": 3.947343220426312e-06,
      "loss": 0.55053484,
      "num_input_tokens_seen": 36248550,
      "step": 1680,
      "time_per_iteration": 3.1289310455322266
    },
    {
      "auxiliary_loss_clip": 0.01181018,
      "auxiliary_loss_mlp": 0.01197711,
      "balance_loss_clip": 1.05625939,
      "balance_loss_mlp": 1.75721645,
      "epoch": 0.10106718773485646,
      "flos": 20007163463040.0,
      "grad_norm": 1.8552620588372701,
      "language_loss": 0.76806682,
      "learning_rate": 3.947254403670641e-06,
      "loss": 0.79185408,
      "num_input_tokens_seen": 36266065,
      "step": 1681,
      "time_per_iteration": 2.461486577987671
    },
    {
      "auxiliary_loss_clip": 0.01155376,
      "auxiliary_loss_mlp": 0.01058062,
      "balance_loss_clip": 1.0553968,
      "balance_loss_mlp": 1.03257513,
      "epoch": 0.10112731098752442,
      "flos": 13479825093120.0,
      "grad_norm": 3.287273216525816,
      "language_loss": 0.94134235,
      "learning_rate": 3.947165513074889e-06,
      "loss": 0.96347672,
      "num_input_tokens_seen": 36280960,
      "step": 1682,
      "time_per_iteration": 2.4972054958343506
    },
    {
      "auxiliary_loss_clip": 0.01171807,
      "auxiliary_loss_mlp": 0.01052214,
      "balance_loss_clip": 1.05372882,
      "balance_loss_mlp": 1.03089952,
      "epoch": 0.1011874342401924,
      "flos": 18515901490560.0,
      "grad_norm": 1.8762755942465152,
      "language_loss": 0.87784034,
      "learning_rate": 3.947076548642425e-06,
      "loss": 0.90008062,
      "num_input_tokens_seen": 36299010,
      "step": 1683,
      "time_per_iteration": 2.457242965698242
    },
    {
      "auxiliary_loss_clip": 0.01135138,
      "auxiliary_loss_mlp": 0.01061408,
      "balance_loss_clip": 1.05900049,
      "balance_loss_mlp": 1.03917587,
      "epoch": 0.10124755749286037,
      "flos": 20702861055360.0,
      "grad_norm": 3.204158047910409,
      "language_loss": 0.74946457,
      "learning_rate": 3.946987510376624e-06,
      "loss": 0.77143002,
      "num_input_tokens_seen": 36318400,
      "step": 1684,
      "time_per_iteration": 2.593762159347534
    },
    {
      "auxiliary_loss_clip": 0.01057541,
      "auxiliary_loss_mlp": 0.01006962,
      "balance_loss_clip": 1.02733672,
      "balance_loss_mlp": 1.00348115,
      "epoch": 0.10130768074552833,
      "flos": 56109456247680.0,
      "grad_norm": 0.756496410994953,
      "language_loss": 0.61084735,
      "learning_rate": 3.9468983982808615e-06,
      "loss": 0.63149238,
      "num_input_tokens_seen": 36381815,
      "step": 1685,
      "time_per_iteration": 3.2083797454833984
    },
    {
      "auxiliary_loss_clip": 0.01157197,
      "auxiliary_loss_mlp": 0.01055481,
      "balance_loss_clip": 1.05331862,
      "balance_loss_mlp": 1.03340316,
      "epoch": 0.1013678039981963,
      "flos": 33402346156800.0,
      "grad_norm": 2.9622982536318663,
      "language_loss": 0.61394286,
      "learning_rate": 3.946809212358516e-06,
      "loss": 0.63606966,
      "num_input_tokens_seen": 36404320,
      "step": 1686,
      "time_per_iteration": 2.629000186920166
    },
    {
      "auxiliary_loss_clip": 0.01147346,
      "auxiliary_loss_mlp": 0.01057377,
      "balance_loss_clip": 1.06210196,
      "balance_loss_mlp": 1.03479874,
      "epoch": 0.10142792725086427,
      "flos": 31905338008320.0,
      "grad_norm": 2.4381170151817018,
      "language_loss": 0.81129277,
      "learning_rate": 3.946719952612972e-06,
      "loss": 0.83333999,
      "num_input_tokens_seen": 36427510,
      "step": 1687,
      "time_per_iteration": 2.6599340438842773
    },
    {
      "auxiliary_loss_clip": 0.01176007,
      "auxiliary_loss_mlp": 0.01050807,
      "balance_loss_clip": 1.05758405,
      "balance_loss_mlp": 1.02928948,
      "epoch": 0.10148805050353224,
      "flos": 28475905046400.0,
      "grad_norm": 2.0939914167331377,
      "language_loss": 0.71902454,
      "learning_rate": 3.94663061904761e-06,
      "loss": 0.74129272,
      "num_input_tokens_seen": 36448230,
      "step": 1688,
      "time_per_iteration": 2.5608816146850586
    },
    {
      "auxiliary_loss_clip": 0.01154131,
      "auxiliary_loss_mlp": 0.01059129,
      "balance_loss_clip": 1.05744731,
      "balance_loss_mlp": 1.03770745,
      "epoch": 0.1015481737562002,
      "flos": 25148888737920.0,
      "grad_norm": 2.0769117973926807,
      "language_loss": 0.86740959,
      "learning_rate": 3.94654121166582e-06,
      "loss": 0.88954222,
      "num_input_tokens_seen": 36464395,
      "step": 1689,
      "time_per_iteration": 2.5547194480895996
    },
    {
      "auxiliary_loss_clip": 0.01168087,
      "auxiliary_loss_mlp": 0.01053719,
      "balance_loss_clip": 1.05032992,
      "balance_loss_mlp": 1.03397799,
      "epoch": 0.10160829700886818,
      "flos": 30882781630080.0,
      "grad_norm": 1.8718334423079868,
      "language_loss": 0.8813554,
      "learning_rate": 3.946451730470993e-06,
      "loss": 0.90357345,
      "num_input_tokens_seen": 36486475,
      "step": 1690,
      "time_per_iteration": 2.5791540145874023
    },
    {
      "auxiliary_loss_clip": 0.01156331,
      "auxiliary_loss_mlp": 0.01056095,
      "balance_loss_clip": 1.05317354,
      "balance_loss_mlp": 1.0339818,
      "epoch": 0.10166842026153615,
      "flos": 20412020632320.0,
      "grad_norm": 1.848734048949071,
      "language_loss": 0.83459806,
      "learning_rate": 3.946362175466521e-06,
      "loss": 0.85672235,
      "num_input_tokens_seen": 36505310,
      "step": 1691,
      "time_per_iteration": 2.5394649505615234
    },
    {
      "auxiliary_loss_clip": 0.01159971,
      "auxiliary_loss_mlp": 0.01051899,
      "balance_loss_clip": 1.05292833,
      "balance_loss_mlp": 1.03040516,
      "epoch": 0.10172854351420411,
      "flos": 33476968661760.0,
      "grad_norm": 1.5554040917278282,
      "language_loss": 0.66454363,
      "learning_rate": 3.946272546655801e-06,
      "loss": 0.68666232,
      "num_input_tokens_seen": 36529820,
      "step": 1692,
      "time_per_iteration": 4.077995300292969
    },
    {
      "auxiliary_loss_clip": 0.01146332,
      "auxiliary_loss_mlp": 0.01072514,
      "balance_loss_clip": 1.05594361,
      "balance_loss_mlp": 1.0504241,
      "epoch": 0.1017886667668721,
      "flos": 23550325862400.0,
      "grad_norm": 1.6763555841319027,
      "language_loss": 0.75622815,
      "learning_rate": 3.94618284404223e-06,
      "loss": 0.77841663,
      "num_input_tokens_seen": 36549000,
      "step": 1693,
      "time_per_iteration": 2.6031036376953125
    },
    {
      "auxiliary_loss_clip": 0.01136766,
      "auxiliary_loss_mlp": 0.01054501,
      "balance_loss_clip": 1.05954564,
      "balance_loss_mlp": 1.03129101,
      "epoch": 0.10184879001954006,
      "flos": 23296078419840.0,
      "grad_norm": 1.7318061046272604,
      "language_loss": 0.87409329,
      "learning_rate": 3.9460930676292105e-06,
      "loss": 0.89600599,
      "num_input_tokens_seen": 36567515,
      "step": 1694,
      "time_per_iteration": 3.9995791912078857
    },
    {
      "auxiliary_loss_clip": 0.01130341,
      "auxiliary_loss_mlp": 0.01056828,
      "balance_loss_clip": 1.05114627,
      "balance_loss_mlp": 1.03284311,
      "epoch": 0.10190891327220802,
      "flos": 18333116156160.0,
      "grad_norm": 1.919646286561325,
      "language_loss": 0.7953403,
      "learning_rate": 3.946003217420147e-06,
      "loss": 0.81721199,
      "num_input_tokens_seen": 36586190,
      "step": 1695,
      "time_per_iteration": 2.56355619430542
    },
    {
      "auxiliary_loss_clip": 0.01128123,
      "auxiliary_loss_mlp": 0.01057347,
      "balance_loss_clip": 1.0502497,
      "balance_loss_mlp": 1.03376698,
      "epoch": 0.10196903652487599,
      "flos": 26465374108800.0,
      "grad_norm": 1.7253950315677058,
      "language_loss": 0.86351335,
      "learning_rate": 3.945913293418447e-06,
      "loss": 0.88536799,
      "num_input_tokens_seen": 36607495,
      "step": 1696,
      "time_per_iteration": 2.6476330757141113
    },
    {
      "auxiliary_loss_clip": 0.01162962,
      "auxiliary_loss_mlp": 0.01052422,
      "balance_loss_clip": 1.05220771,
      "balance_loss_mlp": 1.03126228,
      "epoch": 0.10202915977754397,
      "flos": 21869526798720.0,
      "grad_norm": 2.7416056884323914,
      "language_loss": 0.82340646,
      "learning_rate": 3.945823295627519e-06,
      "loss": 0.84556031,
      "num_input_tokens_seen": 36628555,
      "step": 1697,
      "time_per_iteration": 3.9938836097717285
    },
    {
      "auxiliary_loss_clip": 0.01181714,
      "auxiliary_loss_mlp": 0.01048634,
      "balance_loss_clip": 1.05460072,
      "balance_loss_mlp": 1.02700925,
      "epoch": 0.10208928303021193,
      "flos": 22309755886080.0,
      "grad_norm": 2.372492895130597,
      "language_loss": 0.81229579,
      "learning_rate": 3.9457332240507775e-06,
      "loss": 0.83459926,
      "num_input_tokens_seen": 36646250,
      "step": 1698,
      "time_per_iteration": 2.474506139755249
    },
    {
      "auxiliary_loss_clip": 0.01148845,
      "auxiliary_loss_mlp": 0.01047703,
      "balance_loss_clip": 1.05510545,
      "balance_loss_mlp": 1.02686489,
      "epoch": 0.1021494062828799,
      "flos": 22125569921280.0,
      "grad_norm": 2.7362126731869414,
      "language_loss": 0.75769895,
      "learning_rate": 3.945643078691637e-06,
      "loss": 0.7796644,
      "num_input_tokens_seen": 36666675,
      "step": 1699,
      "time_per_iteration": 2.5797290802001953
    },
    {
      "auxiliary_loss_clip": 0.01154056,
      "auxiliary_loss_mlp": 0.01048007,
      "balance_loss_clip": 1.05329156,
      "balance_loss_mlp": 1.02691865,
      "epoch": 0.10220952953554788,
      "flos": 19646728439040.0,
      "grad_norm": 1.8431023863595613,
      "language_loss": 0.80172378,
      "learning_rate": 3.945552859553516e-06,
      "loss": 0.82374442,
      "num_input_tokens_seen": 36685225,
      "step": 1700,
      "time_per_iteration": 2.5174407958984375
    },
    {
      "auxiliary_loss_clip": 0.0116711,
      "auxiliary_loss_mlp": 0.01045643,
      "balance_loss_clip": 1.05290627,
      "balance_loss_mlp": 1.02494812,
      "epoch": 0.10226965278821584,
      "flos": 29787290686080.0,
      "grad_norm": 1.9666881978190862,
      "language_loss": 0.77117586,
      "learning_rate": 3.945462566639836e-06,
      "loss": 0.79330337,
      "num_input_tokens_seen": 36705985,
      "step": 1701,
      "time_per_iteration": 3.937105178833008
    },
    {
      "auxiliary_loss_clip": 0.01173961,
      "auxiliary_loss_mlp": 0.01050267,
      "balance_loss_clip": 1.05487573,
      "balance_loss_mlp": 1.02832007,
      "epoch": 0.10232977604088381,
      "flos": 27016818681600.0,
      "grad_norm": 2.2635970448667817,
      "language_loss": 0.78121394,
      "learning_rate": 3.945372199954019e-06,
      "loss": 0.80345619,
      "num_input_tokens_seen": 36725815,
      "step": 1702,
      "time_per_iteration": 2.541058301925659
    },
    {
      "auxiliary_loss_clip": 0.01153432,
      "auxiliary_loss_mlp": 0.01051298,
      "balance_loss_clip": 1.05406308,
      "balance_loss_mlp": 1.03135419,
      "epoch": 0.10238989929355179,
      "flos": 20777519473920.0,
      "grad_norm": 2.205698764785944,
      "language_loss": 0.9453299,
      "learning_rate": 3.945281759499494e-06,
      "loss": 0.96737719,
      "num_input_tokens_seen": 36742345,
      "step": 1703,
      "time_per_iteration": 2.531846523284912
    },
    {
      "auxiliary_loss_clip": 0.01038489,
      "auxiliary_loss_mlp": 0.01016016,
      "balance_loss_clip": 1.04027462,
      "balance_loss_mlp": 1.01220179,
      "epoch": 0.10245002254621975,
      "flos": 57698322451200.0,
      "grad_norm": 0.8703133893021714,
      "language_loss": 0.54981375,
      "learning_rate": 3.94519124527969e-06,
      "loss": 0.57035875,
      "num_input_tokens_seen": 36798775,
      "step": 1704,
      "time_per_iteration": 3.1068997383117676
    },
    {
      "auxiliary_loss_clip": 0.01180324,
      "auxiliary_loss_mlp": 0.01048115,
      "balance_loss_clip": 1.0553019,
      "balance_loss_mlp": 1.02650213,
      "epoch": 0.10251014579888772,
      "flos": 16800125558400.0,
      "grad_norm": 2.4984952322017975,
      "language_loss": 0.83908957,
      "learning_rate": 3.945100657298039e-06,
      "loss": 0.86137396,
      "num_input_tokens_seen": 36816295,
      "step": 1705,
      "time_per_iteration": 2.4519641399383545
    },
    {
      "auxiliary_loss_clip": 0.0105298,
      "auxiliary_loss_mlp": 0.01019988,
      "balance_loss_clip": 1.03344798,
      "balance_loss_mlp": 1.01691234,
      "epoch": 0.1025702690515557,
      "flos": 68565500922240.0,
      "grad_norm": 0.772230297834778,
      "language_loss": 0.60364008,
      "learning_rate": 3.9450099955579765e-06,
      "loss": 0.62436974,
      "num_input_tokens_seen": 36882030,
      "step": 1706,
      "time_per_iteration": 3.1839938163757324
    },
    {
      "auxiliary_loss_clip": 0.0114433,
      "auxiliary_loss_mlp": 0.01048457,
      "balance_loss_clip": 1.04970741,
      "balance_loss_mlp": 1.02668953,
      "epoch": 0.10263039230422366,
      "flos": 14866623336960.0,
      "grad_norm": 2.546464925469809,
      "language_loss": 0.8606329,
      "learning_rate": 3.94491926006294e-06,
      "loss": 0.88256073,
      "num_input_tokens_seen": 36899245,
      "step": 1707,
      "time_per_iteration": 2.5396816730499268
    },
    {
      "auxiliary_loss_clip": 0.01164831,
      "auxiliary_loss_mlp": 0.01047735,
      "balance_loss_clip": 1.05649161,
      "balance_loss_mlp": 1.02788687,
      "epoch": 0.10269051555689163,
      "flos": 25337599816320.0,
      "grad_norm": 1.5883761278519797,
      "language_loss": 0.73039073,
      "learning_rate": 3.944828450816369e-06,
      "loss": 0.75251639,
      "num_input_tokens_seen": 36920950,
      "step": 1708,
      "time_per_iteration": 2.5603139400482178
    },
    {
      "auxiliary_loss_clip": 0.0115009,
      "auxiliary_loss_mlp": 0.0156235,
      "balance_loss_clip": 1.05417132,
      "balance_loss_mlp": 2.42871428,
      "epoch": 0.10275063880955959,
      "flos": 21068826773760.0,
      "grad_norm": 1.7575007822836726,
      "language_loss": 0.91337562,
      "learning_rate": 3.944737567821709e-06,
      "loss": 0.94050002,
      "num_input_tokens_seen": 36938900,
      "step": 1709,
      "time_per_iteration": 2.537498950958252
    },
    {
      "auxiliary_loss_clip": 0.01129839,
      "auxiliary_loss_mlp": 0.01054313,
      "balance_loss_clip": 1.0622344,
      "balance_loss_mlp": 1.03278422,
      "epoch": 0.10281076206222757,
      "flos": 30366780802560.0,
      "grad_norm": 2.2155538612842536,
      "language_loss": 0.88064063,
      "learning_rate": 3.944646611082406e-06,
      "loss": 0.90248215,
      "num_input_tokens_seen": 36957010,
      "step": 1710,
      "time_per_iteration": 2.6491658687591553
    },
    {
      "auxiliary_loss_clip": 0.01162923,
      "auxiliary_loss_mlp": 0.01052097,
      "balance_loss_clip": 1.05081046,
      "balance_loss_mlp": 1.0315814,
      "epoch": 0.10287088531489554,
      "flos": 22418313765120.0,
      "grad_norm": 2.040467019812419,
      "language_loss": 0.79383427,
      "learning_rate": 3.944555580601908e-06,
      "loss": 0.81598449,
      "num_input_tokens_seen": 36977690,
      "step": 1711,
      "time_per_iteration": 2.511686325073242
    },
    {
      "auxiliary_loss_clip": 0.0114834,
      "auxiliary_loss_mlp": 0.01054245,
      "balance_loss_clip": 1.05249047,
      "balance_loss_mlp": 1.03247786,
      "epoch": 0.1029310085675635,
      "flos": 25115994858240.0,
      "grad_norm": 2.1474448764818286,
      "language_loss": 0.73989224,
      "learning_rate": 3.944464476383668e-06,
      "loss": 0.76191813,
      "num_input_tokens_seen": 36997300,
      "step": 1712,
      "time_per_iteration": 2.6086416244506836
    },
    {
      "auxiliary_loss_clip": 0.01127561,
      "auxiliary_loss_mlp": 0.01063333,
      "balance_loss_clip": 1.0524081,
      "balance_loss_mlp": 1.04149342,
      "epoch": 0.10299113182023148,
      "flos": 19865639877120.0,
      "grad_norm": 1.7238740079466595,
      "language_loss": 0.87099469,
      "learning_rate": 3.94437329843114e-06,
      "loss": 0.89290357,
      "num_input_tokens_seen": 37016110,
      "step": 1713,
      "time_per_iteration": 2.5488805770874023
    },
    {
      "auxiliary_loss_clip": 0.01162137,
      "auxiliary_loss_mlp": 0.01059579,
      "balance_loss_clip": 1.05320048,
      "balance_loss_mlp": 1.03990972,
      "epoch": 0.10305125507289944,
      "flos": 20447608032000.0,
      "grad_norm": 1.6759586785187683,
      "language_loss": 0.72233427,
      "learning_rate": 3.944282046747782e-06,
      "loss": 0.74455142,
      "num_input_tokens_seen": 37036405,
      "step": 1714,
      "time_per_iteration": 2.540839910507202
    },
    {
      "auxiliary_loss_clip": 0.01167751,
      "auxiliary_loss_mlp": 0.01058349,
      "balance_loss_clip": 1.05415404,
      "balance_loss_mlp": 1.03647387,
      "epoch": 0.10311137832556741,
      "flos": 26250772302720.0,
      "grad_norm": 2.093736037244032,
      "language_loss": 0.90964705,
      "learning_rate": 3.944190721337053e-06,
      "loss": 0.93190807,
      "num_input_tokens_seen": 37057580,
      "step": 1715,
      "time_per_iteration": 2.558284044265747
    },
    {
      "auxiliary_loss_clip": 0.01162621,
      "auxiliary_loss_mlp": 0.01062693,
      "balance_loss_clip": 1.05007958,
      "balance_loss_mlp": 1.04184294,
      "epoch": 0.10317150157823539,
      "flos": 35298932175360.0,
      "grad_norm": 1.875706327962939,
      "language_loss": 0.75722075,
      "learning_rate": 3.944099322202418e-06,
      "loss": 0.77947384,
      "num_input_tokens_seen": 37079120,
      "step": 1716,
      "time_per_iteration": 2.6406195163726807
    },
    {
      "auxiliary_loss_clip": 0.01158983,
      "auxiliary_loss_mlp": 0.01072647,
      "balance_loss_clip": 1.05451608,
      "balance_loss_mlp": 1.05019951,
      "epoch": 0.10323162483090335,
      "flos": 25739943033600.0,
      "grad_norm": 2.1749093302522073,
      "language_loss": 0.85292661,
      "learning_rate": 3.944007849347342e-06,
      "loss": 0.87524295,
      "num_input_tokens_seen": 37099710,
      "step": 1717,
      "time_per_iteration": 2.5754611492156982
    },
    {
      "auxiliary_loss_clip": 0.01127489,
      "auxiliary_loss_mlp": 0.01070926,
      "balance_loss_clip": 1.05767846,
      "balance_loss_mlp": 1.0489434,
      "epoch": 0.10329174808357132,
      "flos": 16289870906880.0,
      "grad_norm": 2.108945013379632,
      "language_loss": 0.8266778,
      "learning_rate": 3.943916302775292e-06,
      "loss": 0.8486619,
      "num_input_tokens_seen": 37117775,
      "step": 1718,
      "time_per_iteration": 2.543630599975586
    },
    {
      "auxiliary_loss_clip": 0.01164306,
      "auxiliary_loss_mlp": 0.01054119,
      "balance_loss_clip": 1.05519748,
      "balance_loss_mlp": 1.03279221,
      "epoch": 0.10335187133623928,
      "flos": 36687166963200.0,
      "grad_norm": 2.2999681639479226,
      "language_loss": 0.73060048,
      "learning_rate": 3.943824682489742e-06,
      "loss": 0.75278473,
      "num_input_tokens_seen": 37140280,
      "step": 1719,
      "time_per_iteration": 2.6379857063293457
    },
    {
      "auxiliary_loss_clip": 0.01161706,
      "auxiliary_loss_mlp": 0.01051204,
      "balance_loss_clip": 1.05447817,
      "balance_loss_mlp": 1.03131938,
      "epoch": 0.10341199458890726,
      "flos": 14975648092800.0,
      "grad_norm": 2.1092099486940343,
      "language_loss": 0.9252882,
      "learning_rate": 3.9437329884941665e-06,
      "loss": 0.94741726,
      "num_input_tokens_seen": 37158350,
      "step": 1720,
      "time_per_iteration": 2.492255926132202
    },
    {
      "auxiliary_loss_clip": 0.01134385,
      "auxiliary_loss_mlp": 0.0105438,
      "balance_loss_clip": 1.05007017,
      "balance_loss_mlp": 1.03324366,
      "epoch": 0.10347211784157523,
      "flos": 21031587348480.0,
      "grad_norm": 1.732835571101408,
      "language_loss": 0.79152822,
      "learning_rate": 3.943641220792039e-06,
      "loss": 0.81341588,
      "num_input_tokens_seen": 37177120,
      "step": 1721,
      "time_per_iteration": 2.5552775859832764
    },
    {
      "auxiliary_loss_clip": 0.01126297,
      "auxiliary_loss_mlp": 0.01060872,
      "balance_loss_clip": 1.05196166,
      "balance_loss_mlp": 1.03586149,
      "epoch": 0.1035322410942432,
      "flos": 19792094780160.0,
      "grad_norm": 2.733970394153954,
      "language_loss": 0.81011301,
      "learning_rate": 3.9435493793868434e-06,
      "loss": 0.8319847,
      "num_input_tokens_seen": 37195895,
      "step": 1722,
      "time_per_iteration": 2.5907089710235596
    },
    {
      "auxiliary_loss_clip": 0.01059375,
      "auxiliary_loss_mlp": 0.01078303,
      "balance_loss_clip": 1.02992845,
      "balance_loss_mlp": 1.07540655,
      "epoch": 0.10359236434691117,
      "flos": 52698874947840.0,
      "grad_norm": 0.9461026808466435,
      "language_loss": 0.67147028,
      "learning_rate": 3.943457464282059e-06,
      "loss": 0.69284701,
      "num_input_tokens_seen": 37247270,
      "step": 1723,
      "time_per_iteration": 2.9297189712524414
    },
    {
      "auxiliary_loss_clip": 0.01166948,
      "auxiliary_loss_mlp": 0.01055123,
      "balance_loss_clip": 1.05324006,
      "balance_loss_mlp": 1.03505993,
      "epoch": 0.10365248759957914,
      "flos": 18405404277120.0,
      "grad_norm": 3.213318586721877,
      "language_loss": 0.78115261,
      "learning_rate": 3.9433654754811745e-06,
      "loss": 0.80337334,
      "num_input_tokens_seen": 37265595,
      "step": 1724,
      "time_per_iteration": 2.4845407009124756
    },
    {
      "auxiliary_loss_clip": 0.01145895,
      "auxiliary_loss_mlp": 0.01056077,
      "balance_loss_clip": 1.05374432,
      "balance_loss_mlp": 1.0349772,
      "epoch": 0.1037126108522471,
      "flos": 47553555335040.0,
      "grad_norm": 1.8318637742336048,
      "language_loss": 0.7496568,
      "learning_rate": 3.943273412987676e-06,
      "loss": 0.77167654,
      "num_input_tokens_seen": 37286660,
      "step": 1725,
      "time_per_iteration": 2.8051393032073975
    },
    {
      "auxiliary_loss_clip": 0.01132911,
      "auxiliary_loss_mlp": 0.01056396,
      "balance_loss_clip": 1.05178928,
      "balance_loss_mlp": 1.03480732,
      "epoch": 0.10377273410491508,
      "flos": 22816670572800.0,
      "grad_norm": 2.0853435869543926,
      "language_loss": 0.74944401,
      "learning_rate": 3.943181276805054e-06,
      "loss": 0.77133703,
      "num_input_tokens_seen": 37304915,
      "step": 1726,
      "time_per_iteration": 2.5633206367492676
    },
    {
      "auxiliary_loss_clip": 0.01150464,
      "auxiliary_loss_mlp": 0.01052084,
      "balance_loss_clip": 1.05729389,
      "balance_loss_mlp": 1.03067422,
      "epoch": 0.10383285735758305,
      "flos": 26138694890880.0,
      "grad_norm": 11.845370406244118,
      "language_loss": 0.73643637,
      "learning_rate": 3.9430890669368035e-06,
      "loss": 0.75846183,
      "num_input_tokens_seen": 37325265,
      "step": 1727,
      "time_per_iteration": 2.5860443115234375
    },
    {
      "auxiliary_loss_clip": 0.01157211,
      "auxiliary_loss_mlp": 0.01056291,
      "balance_loss_clip": 1.05561399,
      "balance_loss_mlp": 1.03517938,
      "epoch": 0.10389298061025101,
      "flos": 17091791994240.0,
      "grad_norm": 2.259312824890131,
      "language_loss": 0.84974486,
      "learning_rate": 3.942996783386422e-06,
      "loss": 0.87187982,
      "num_input_tokens_seen": 37341650,
      "step": 1728,
      "time_per_iteration": 2.4807839393615723
    },
    {
      "auxiliary_loss_clip": 0.01153481,
      "auxiliary_loss_mlp": 0.01050793,
      "balance_loss_clip": 1.0540278,
      "balance_loss_mlp": 1.02988398,
      "epoch": 0.10395310386291898,
      "flos": 20776513893120.0,
      "grad_norm": 2.5216398377819798,
      "language_loss": 0.70835495,
      "learning_rate": 3.942904426157406e-06,
      "loss": 0.7303977,
      "num_input_tokens_seen": 37360270,
      "step": 1729,
      "time_per_iteration": 2.5593297481536865
    },
    {
      "auxiliary_loss_clip": 0.01155551,
      "auxiliary_loss_mlp": 0.01053764,
      "balance_loss_clip": 1.05540073,
      "balance_loss_mlp": 1.03103077,
      "epoch": 0.10401322711558696,
      "flos": 12820540913280.0,
      "grad_norm": 2.6513950258452765,
      "language_loss": 0.81322199,
      "learning_rate": 3.9428119952532605e-06,
      "loss": 0.83531517,
      "num_input_tokens_seen": 37375225,
      "step": 1730,
      "time_per_iteration": 3.9438223838806152
    },
    {
      "auxiliary_loss_clip": 0.01084529,
      "auxiliary_loss_mlp": 0.01047079,
      "balance_loss_clip": 1.05317879,
      "balance_loss_mlp": 1.02678919,
      "epoch": 0.10407335036825492,
      "flos": 23184683366400.0,
      "grad_norm": 1.7240102171951683,
      "language_loss": 0.75962621,
      "learning_rate": 3.942719490677489e-06,
      "loss": 0.78094232,
      "num_input_tokens_seen": 37395165,
      "step": 1731,
      "time_per_iteration": 2.814145803451538
    },
    {
      "auxiliary_loss_clip": 0.01124171,
      "auxiliary_loss_mlp": 0.01048416,
      "balance_loss_clip": 1.04956865,
      "balance_loss_mlp": 1.02874613,
      "epoch": 0.10413347362092289,
      "flos": 26104184899200.0,
      "grad_norm": 1.918544005759015,
      "language_loss": 0.8289305,
      "learning_rate": 3.9426269124336e-06,
      "loss": 0.85065639,
      "num_input_tokens_seen": 37414845,
      "step": 1732,
      "time_per_iteration": 2.947763681411743
    },
    {
      "auxiliary_loss_clip": 0.01135663,
      "auxiliary_loss_mlp": 0.01049166,
      "balance_loss_clip": 1.05447412,
      "balance_loss_mlp": 1.02954388,
      "epoch": 0.10419359687359087,
      "flos": 12641059630080.0,
      "grad_norm": 2.2252035222939455,
      "language_loss": 0.83000064,
      "learning_rate": 3.942534260525104e-06,
      "loss": 0.8518489,
      "num_input_tokens_seen": 37432490,
      "step": 1733,
      "time_per_iteration": 3.9466803073883057
    },
    {
      "auxiliary_loss_clip": 0.01149287,
      "auxiliary_loss_mlp": 0.0105177,
      "balance_loss_clip": 1.05274796,
      "balance_loss_mlp": 1.0317663,
      "epoch": 0.10425372012625883,
      "flos": 12125094716160.0,
      "grad_norm": 2.165286297854378,
      "language_loss": 0.76285768,
      "learning_rate": 3.942441534955514e-06,
      "loss": 0.78486824,
      "num_input_tokens_seen": 37449435,
      "step": 1734,
      "time_per_iteration": 2.5231964588165283
    },
    {
      "auxiliary_loss_clip": 0.01132421,
      "auxiliary_loss_mlp": 0.01045411,
      "balance_loss_clip": 1.04897571,
      "balance_loss_mlp": 1.02575386,
      "epoch": 0.1043138433789268,
      "flos": 25337563902720.0,
      "grad_norm": 2.168417789257319,
      "language_loss": 0.74906653,
      "learning_rate": 3.9423487357283465e-06,
      "loss": 0.77084494,
      "num_input_tokens_seen": 37469105,
      "step": 1735,
      "time_per_iteration": 2.5897128582000732
    },
    {
      "auxiliary_loss_clip": 0.01160858,
      "auxiliary_loss_mlp": 0.01051041,
      "balance_loss_clip": 1.0521853,
      "balance_loss_mlp": 1.03064489,
      "epoch": 0.10437396663159478,
      "flos": 29167149352320.0,
      "grad_norm": 1.8636968946175627,
      "language_loss": 0.78428417,
      "learning_rate": 3.94225586284712e-06,
      "loss": 0.80640316,
      "num_input_tokens_seen": 37490540,
      "step": 1736,
      "time_per_iteration": 3.9105682373046875
    },
    {
      "auxiliary_loss_clip": 0.01161866,
      "auxiliary_loss_mlp": 0.01057834,
      "balance_loss_clip": 1.05446315,
      "balance_loss_mlp": 1.03789067,
      "epoch": 0.10443408988426274,
      "flos": 25080946162560.0,
      "grad_norm": 1.9149999599365735,
      "language_loss": 0.70723796,
      "learning_rate": 3.942162916315356e-06,
      "loss": 0.72943497,
      "num_input_tokens_seen": 37511905,
      "step": 1737,
      "time_per_iteration": 2.5707595348358154
    },
    {
      "auxiliary_loss_clip": 0.01146523,
      "auxiliary_loss_mlp": 0.01052125,
      "balance_loss_clip": 1.04708982,
      "balance_loss_mlp": 1.02889144,
      "epoch": 0.1044942131369307,
      "flos": 26759662237440.0,
      "grad_norm": 2.2712310612509525,
      "language_loss": 0.81423521,
      "learning_rate": 3.942069896136581e-06,
      "loss": 0.83622169,
      "num_input_tokens_seen": 37533635,
      "step": 1738,
      "time_per_iteration": 2.5924885272979736
    },
    {
      "auxiliary_loss_clip": 0.0117531,
      "auxiliary_loss_mlp": 0.01057467,
      "balance_loss_clip": 1.05036664,
      "balance_loss_mlp": 1.03518713,
      "epoch": 0.10455433638959867,
      "flos": 18442571875200.0,
      "grad_norm": 1.8630843064450895,
      "language_loss": 0.74904883,
      "learning_rate": 3.9419768023143196e-06,
      "loss": 0.77137661,
      "num_input_tokens_seen": 37552035,
      "step": 1739,
      "time_per_iteration": 3.8633999824523926
    },
    {
      "auxiliary_loss_clip": 0.01143867,
      "auxiliary_loss_mlp": 0.01052484,
      "balance_loss_clip": 1.05389833,
      "balance_loss_mlp": 1.03219473,
      "epoch": 0.10461445964226665,
      "flos": 23218977876480.0,
      "grad_norm": 1.6212283588393184,
      "language_loss": 0.7723701,
      "learning_rate": 3.941883634852104e-06,
      "loss": 0.79433364,
      "num_input_tokens_seen": 37571540,
      "step": 1740,
      "time_per_iteration": 2.565011978149414
    },
    {
      "auxiliary_loss_clip": 0.01146818,
      "auxiliary_loss_mlp": 0.01049176,
      "balance_loss_clip": 1.05365014,
      "balance_loss_mlp": 1.02881551,
      "epoch": 0.10467458289493461,
      "flos": 24345243797760.0,
      "grad_norm": 2.407472764127049,
      "language_loss": 0.8584013,
      "learning_rate": 3.941790393753467e-06,
      "loss": 0.8803612,
      "num_input_tokens_seen": 37588265,
      "step": 1741,
      "time_per_iteration": 2.5485949516296387
    },
    {
      "auxiliary_loss_clip": 0.01151023,
      "auxiliary_loss_mlp": 0.01053775,
      "balance_loss_clip": 1.05233812,
      "balance_loss_mlp": 1.03290153,
      "epoch": 0.10473470614760258,
      "flos": 21287953693440.0,
      "grad_norm": 2.623997973886344,
      "language_loss": 0.75444317,
      "learning_rate": 3.941697079021942e-06,
      "loss": 0.77649117,
      "num_input_tokens_seen": 37606860,
      "step": 1742,
      "time_per_iteration": 2.538252353668213
    },
    {
      "auxiliary_loss_clip": 0.01125131,
      "auxiliary_loss_mlp": 0.01059896,
      "balance_loss_clip": 1.05563211,
      "balance_loss_mlp": 1.04016685,
      "epoch": 0.10479482940027056,
      "flos": 21687208341120.0,
      "grad_norm": 2.987555389720698,
      "language_loss": 0.87398732,
      "learning_rate": 3.94160369066107e-06,
      "loss": 0.89583766,
      "num_input_tokens_seen": 37625210,
      "step": 1743,
      "time_per_iteration": 2.640143632888794
    },
    {
      "auxiliary_loss_clip": 0.01132163,
      "auxiliary_loss_mlp": 0.01048536,
      "balance_loss_clip": 1.05117619,
      "balance_loss_mlp": 1.02751994,
      "epoch": 0.10485495265293852,
      "flos": 21573694385280.0,
      "grad_norm": 2.065369371468023,
      "language_loss": 0.75548315,
      "learning_rate": 3.941510228674391e-06,
      "loss": 0.77729023,
      "num_input_tokens_seen": 37644110,
      "step": 1744,
      "time_per_iteration": 2.587905168533325
    },
    {
      "auxiliary_loss_clip": 0.01165242,
      "auxiliary_loss_mlp": 0.01052565,
      "balance_loss_clip": 1.05707669,
      "balance_loss_mlp": 1.0333004,
      "epoch": 0.10491507590560649,
      "flos": 37961923708800.0,
      "grad_norm": 2.0435678240331194,
      "language_loss": 0.79312658,
      "learning_rate": 3.941416693065451e-06,
      "loss": 0.81530464,
      "num_input_tokens_seen": 37665800,
      "step": 1745,
      "time_per_iteration": 2.649773359298706
    },
    {
      "auxiliary_loss_clip": 0.0117486,
      "auxiliary_loss_mlp": 0.01062079,
      "balance_loss_clip": 1.05321646,
      "balance_loss_mlp": 1.04193258,
      "epoch": 0.10497519915827447,
      "flos": 26396282298240.0,
      "grad_norm": 2.170149219786544,
      "language_loss": 0.82769597,
      "learning_rate": 3.941323083837794e-06,
      "loss": 0.85006535,
      "num_input_tokens_seen": 37685095,
      "step": 1746,
      "time_per_iteration": 2.5292367935180664
    },
    {
      "auxiliary_loss_clip": 0.01156022,
      "auxiliary_loss_mlp": 0.0106003,
      "balance_loss_clip": 1.05622184,
      "balance_loss_mlp": 1.04053915,
      "epoch": 0.10503532241094243,
      "flos": 40662190581120.0,
      "grad_norm": 1.7223830177455677,
      "language_loss": 0.70099699,
      "learning_rate": 3.941229400994971e-06,
      "loss": 0.72315753,
      "num_input_tokens_seen": 37707445,
      "step": 1747,
      "time_per_iteration": 2.711099863052368
    },
    {
      "auxiliary_loss_clip": 0.01152918,
      "auxiliary_loss_mlp": 0.01061526,
      "balance_loss_clip": 1.05539751,
      "balance_loss_mlp": 1.04097414,
      "epoch": 0.1050954456636104,
      "flos": 29789409588480.0,
      "grad_norm": 2.636072705669235,
      "language_loss": 0.84105939,
      "learning_rate": 3.941135644540535e-06,
      "loss": 0.86320388,
      "num_input_tokens_seen": 37728325,
      "step": 1748,
      "time_per_iteration": 2.6599011421203613
    },
    {
      "auxiliary_loss_clip": 0.01171076,
      "auxiliary_loss_mlp": 0.0105079,
      "balance_loss_clip": 1.0515753,
      "balance_loss_mlp": 1.0298568,
      "epoch": 0.10515556891627838,
      "flos": 23948754497280.0,
      "grad_norm": 1.7328280465115748,
      "language_loss": 0.7181977,
      "learning_rate": 3.941041814478041e-06,
      "loss": 0.74041641,
      "num_input_tokens_seen": 37748910,
      "step": 1749,
      "time_per_iteration": 2.4981110095977783
    },
    {
      "auxiliary_loss_clip": 0.01155722,
      "auxiliary_loss_mlp": 0.0106304,
      "balance_loss_clip": 1.05379081,
      "balance_loss_mlp": 1.04201162,
      "epoch": 0.10521569216894634,
      "flos": 18259606972800.0,
      "grad_norm": 2.1780280474511593,
      "language_loss": 0.8222928,
      "learning_rate": 3.940947910811047e-06,
      "loss": 0.8444804,
      "num_input_tokens_seen": 37765745,
      "step": 1750,
      "time_per_iteration": 2.5448145866394043
    },
    {
      "auxiliary_loss_clip": 0.01155562,
      "auxiliary_loss_mlp": 0.01061524,
      "balance_loss_clip": 1.06383872,
      "balance_loss_mlp": 1.04104376,
      "epoch": 0.10527581542161431,
      "flos": 15630909949440.0,
      "grad_norm": 2.775513187735577,
      "language_loss": 0.92179334,
      "learning_rate": 3.940853933543114e-06,
      "loss": 0.94396418,
      "num_input_tokens_seen": 37780520,
      "step": 1751,
      "time_per_iteration": 2.5375444889068604
    },
    {
      "auxiliary_loss_clip": 0.0116326,
      "auxiliary_loss_mlp": 0.01049581,
      "balance_loss_clip": 1.05527735,
      "balance_loss_mlp": 1.02929175,
      "epoch": 0.10533593867428227,
      "flos": 18296559089280.0,
      "grad_norm": 1.9873912552102726,
      "language_loss": 0.79567432,
      "learning_rate": 3.940759882677805e-06,
      "loss": 0.81780273,
      "num_input_tokens_seen": 37799515,
      "step": 1752,
      "time_per_iteration": 2.510979413986206
    },
    {
      "auxiliary_loss_clip": 0.01113114,
      "auxiliary_loss_mlp": 0.01068994,
      "balance_loss_clip": 1.0522809,
      "balance_loss_mlp": 1.04649925,
      "epoch": 0.10539606192695025,
      "flos": 29023219555200.0,
      "grad_norm": 1.8895087992437076,
      "language_loss": 0.75686044,
      "learning_rate": 3.940665758218686e-06,
      "loss": 0.77868152,
      "num_input_tokens_seen": 37818695,
      "step": 1753,
      "time_per_iteration": 2.6459739208221436
    },
    {
      "auxiliary_loss_clip": 0.01134791,
      "auxiliary_loss_mlp": 0.0105846,
      "balance_loss_clip": 1.05337143,
      "balance_loss_mlp": 1.03646576,
      "epoch": 0.10545618517961822,
      "flos": 19969313506560.0,
      "grad_norm": 3.033317272160769,
      "language_loss": 0.83883905,
      "learning_rate": 3.940571560169328e-06,
      "loss": 0.86077154,
      "num_input_tokens_seen": 37837860,
      "step": 1754,
      "time_per_iteration": 2.579953193664551
    },
    {
      "auxiliary_loss_clip": 0.0113494,
      "auxiliary_loss_mlp": 0.01057352,
      "balance_loss_clip": 1.05794764,
      "balance_loss_mlp": 1.03541732,
      "epoch": 0.10551630843228618,
      "flos": 16143427157760.0,
      "grad_norm": 2.775360992915459,
      "language_loss": 0.68719,
      "learning_rate": 3.940477288533302e-06,
      "loss": 0.70911288,
      "num_input_tokens_seen": 37856260,
      "step": 1755,
      "time_per_iteration": 2.584129571914673
    },
    {
      "auxiliary_loss_clip": 0.0115885,
      "auxiliary_loss_mlp": 0.01067593,
      "balance_loss_clip": 1.05414724,
      "balance_loss_mlp": 1.04613495,
      "epoch": 0.10557643168495416,
      "flos": 23440115957760.0,
      "grad_norm": 3.9945891738566868,
      "language_loss": 0.76852429,
      "learning_rate": 3.940382943314182e-06,
      "loss": 0.79078877,
      "num_input_tokens_seen": 37876960,
      "step": 1756,
      "time_per_iteration": 2.5326125621795654
    },
    {
      "auxiliary_loss_clip": 0.01178649,
      "auxiliary_loss_mlp": 0.01068982,
      "balance_loss_clip": 1.05550528,
      "balance_loss_mlp": 1.04847801,
      "epoch": 0.10563655493762213,
      "flos": 21799034357760.0,
      "grad_norm": 1.764921852785999,
      "language_loss": 0.80173922,
      "learning_rate": 3.940288524515547e-06,
      "loss": 0.82421553,
      "num_input_tokens_seen": 37897070,
      "step": 1757,
      "time_per_iteration": 2.476378917694092
    },
    {
      "auxiliary_loss_clip": 0.01146942,
      "auxiliary_loss_mlp": 0.01060747,
      "balance_loss_clip": 1.05320334,
      "balance_loss_mlp": 1.04004073,
      "epoch": 0.10569667819029009,
      "flos": 53800863275520.0,
      "grad_norm": 1.6585433724236767,
      "language_loss": 0.78746974,
      "learning_rate": 3.940194032140976e-06,
      "loss": 0.80954671,
      "num_input_tokens_seen": 37923635,
      "step": 1758,
      "time_per_iteration": 2.8608648777008057
    },
    {
      "auxiliary_loss_clip": 0.01162769,
      "auxiliary_loss_mlp": 0.01053312,
      "balance_loss_clip": 1.0574826,
      "balance_loss_mlp": 1.03198564,
      "epoch": 0.10575680144295807,
      "flos": 22925515760640.0,
      "grad_norm": 1.8110435378713954,
      "language_loss": 0.91936111,
      "learning_rate": 3.940099466194054e-06,
      "loss": 0.94152188,
      "num_input_tokens_seen": 37942650,
      "step": 1759,
      "time_per_iteration": 2.544886589050293
    },
    {
      "auxiliary_loss_clip": 0.01154576,
      "auxiliary_loss_mlp": 0.01060638,
      "balance_loss_clip": 1.05604362,
      "balance_loss_mlp": 1.03797674,
      "epoch": 0.10581692469562604,
      "flos": 14136667148160.0,
      "grad_norm": 3.9262514160582693,
      "language_loss": 0.77181423,
      "learning_rate": 3.940004826678365e-06,
      "loss": 0.79396641,
      "num_input_tokens_seen": 37960660,
      "step": 1760,
      "time_per_iteration": 2.530640125274658
    },
    {
      "auxiliary_loss_clip": 0.01158429,
      "auxiliary_loss_mlp": 0.01059902,
      "balance_loss_clip": 1.05340552,
      "balance_loss_mlp": 1.03718102,
      "epoch": 0.105877047948294,
      "flos": 25958674903680.0,
      "grad_norm": 2.326696521662957,
      "language_loss": 0.88962048,
      "learning_rate": 3.939910113597498e-06,
      "loss": 0.91180384,
      "num_input_tokens_seen": 37978625,
      "step": 1761,
      "time_per_iteration": 2.5215654373168945
    },
    {
      "auxiliary_loss_clip": 0.01108339,
      "auxiliary_loss_mlp": 0.0105114,
      "balance_loss_clip": 1.0517664,
      "balance_loss_mlp": 1.45875061,
      "epoch": 0.10593717120096197,
      "flos": 30664768032000.0,
      "grad_norm": 2.0151983851540787,
      "language_loss": 0.78292489,
      "learning_rate": 3.9398153269550464e-06,
      "loss": 0.80451965,
      "num_input_tokens_seen": 38000005,
      "step": 1762,
      "time_per_iteration": 2.696902275085449
    },
    {
      "auxiliary_loss_clip": 0.01068063,
      "auxiliary_loss_mlp": 0.01010323,
      "balance_loss_clip": 1.04267836,
      "balance_loss_mlp": 1.00703323,
      "epoch": 0.10599729445362994,
      "flos": 66436682497920.0,
      "grad_norm": 1.3483662875667615,
      "language_loss": 0.60488391,
      "learning_rate": 3.939720466754602e-06,
      "loss": 0.62566769,
      "num_input_tokens_seen": 38066165,
      "step": 1763,
      "time_per_iteration": 3.306469202041626
    },
    {
      "auxiliary_loss_clip": 0.0115544,
      "auxiliary_loss_mlp": 0.0104598,
      "balance_loss_clip": 1.05570102,
      "balance_loss_mlp": 1.02528536,
      "epoch": 0.10605741770629791,
      "flos": 23948179879680.0,
      "grad_norm": 1.7054940994568681,
      "language_loss": 0.80317086,
      "learning_rate": 3.939625532999763e-06,
      "loss": 0.82518506,
      "num_input_tokens_seen": 38086150,
      "step": 1764,
      "time_per_iteration": 2.552823781967163
    },
    {
      "auxiliary_loss_clip": 0.01136143,
      "auxiliary_loss_mlp": 0.01060363,
      "balance_loss_clip": 1.0511446,
      "balance_loss_mlp": 1.03663993,
      "epoch": 0.10611754095896588,
      "flos": 19387524919680.0,
      "grad_norm": 1.6537239392430785,
      "language_loss": 0.80056667,
      "learning_rate": 3.9395305256941314e-06,
      "loss": 0.8225317,
      "num_input_tokens_seen": 38104205,
      "step": 1765,
      "time_per_iteration": 2.5708858966827393
    },
    {
      "auxiliary_loss_clip": 0.01159118,
      "auxiliary_loss_mlp": 0.01053861,
      "balance_loss_clip": 1.05231571,
      "balance_loss_mlp": 1.03228474,
      "epoch": 0.10617766421163385,
      "flos": 22237755073920.0,
      "grad_norm": 1.984565709032615,
      "language_loss": 0.76713896,
      "learning_rate": 3.939435444841306e-06,
      "loss": 0.78926879,
      "num_input_tokens_seen": 38122005,
      "step": 1766,
      "time_per_iteration": 2.503915548324585
    },
    {
      "auxiliary_loss_clip": 0.01176376,
      "auxiliary_loss_mlp": 0.01058708,
      "balance_loss_clip": 1.05312979,
      "balance_loss_mlp": 1.03639162,
      "epoch": 0.10623778746430182,
      "flos": 28404407024640.0,
      "grad_norm": 2.088663084327038,
      "language_loss": 0.77353531,
      "learning_rate": 3.939340290444895e-06,
      "loss": 0.79588616,
      "num_input_tokens_seen": 38143365,
      "step": 1767,
      "time_per_iteration": 2.5584492683410645
    },
    {
      "auxiliary_loss_clip": 0.01003662,
      "auxiliary_loss_mlp": 0.01009336,
      "balance_loss_clip": 1.03118181,
      "balance_loss_mlp": 1.00604594,
      "epoch": 0.10629791071696978,
      "flos": 64234639221120.0,
      "grad_norm": 0.6808700566882854,
      "language_loss": 0.57865024,
      "learning_rate": 3.939245062508506e-06,
      "loss": 0.59878027,
      "num_input_tokens_seen": 38210035,
      "step": 1768,
      "time_per_iteration": 4.984228134155273
    },
    {
      "auxiliary_loss_clip": 0.01138102,
      "auxiliary_loss_mlp": 0.01041145,
      "balance_loss_clip": 1.05171156,
      "balance_loss_mlp": 1.02105808,
      "epoch": 0.10635803396963776,
      "flos": 22747578762240.0,
      "grad_norm": 2.0203134394279396,
      "language_loss": 0.8646363,
      "learning_rate": 3.939149761035749e-06,
      "loss": 0.88642883,
      "num_input_tokens_seen": 38231230,
      "step": 1769,
      "time_per_iteration": 3.202700614929199
    },
    {
      "auxiliary_loss_clip": 0.01135906,
      "auxiliary_loss_mlp": 0.00974813,
      "balance_loss_clip": 1.05130911,
      "balance_loss_mlp": 1.31261635,
      "epoch": 0.10641815722230573,
      "flos": 31395586147200.0,
      "grad_norm": 1.818069600040015,
      "language_loss": 0.61748302,
      "learning_rate": 3.9390543860302395e-06,
      "loss": 0.63859022,
      "num_input_tokens_seen": 38253890,
      "step": 1770,
      "time_per_iteration": 2.7175233364105225
    },
    {
      "auxiliary_loss_clip": 0.01075149,
      "auxiliary_loss_mlp": 0.01004877,
      "balance_loss_clip": 1.03438854,
      "balance_loss_mlp": 1.00177717,
      "epoch": 0.1064782804749737,
      "flos": 58552527784320.0,
      "grad_norm": 0.8822162008919421,
      "language_loss": 0.57055509,
      "learning_rate": 3.9389589374955925e-06,
      "loss": 0.59135532,
      "num_input_tokens_seen": 38304290,
      "step": 1771,
      "time_per_iteration": 3.024432897567749
    },
    {
      "auxiliary_loss_clip": 0.01142526,
      "auxiliary_loss_mlp": 0.01058026,
      "balance_loss_clip": 1.05510736,
      "balance_loss_mlp": 1.03724825,
      "epoch": 0.10653840372764166,
      "flos": 23987825516160.0,
      "grad_norm": 1.8842535519936046,
      "language_loss": 0.88483286,
      "learning_rate": 3.938863415435429e-06,
      "loss": 0.90683842,
      "num_input_tokens_seen": 38324725,
      "step": 1772,
      "time_per_iteration": 4.080848455429077
    },
    {
      "auxiliary_loss_clip": 0.01180133,
      "auxiliary_loss_mlp": 0.01055365,
      "balance_loss_clip": 1.05284476,
      "balance_loss_mlp": 1.03238189,
      "epoch": 0.10659852698030964,
      "flos": 18294655668480.0,
      "grad_norm": 2.4944421194177697,
      "language_loss": 0.7587527,
      "learning_rate": 3.93876781985337e-06,
      "loss": 0.78110766,
      "num_input_tokens_seen": 38340735,
      "step": 1773,
      "time_per_iteration": 2.4679439067840576
    },
    {
      "auxiliary_loss_clip": 0.01125118,
      "auxiliary_loss_mlp": 0.01062614,
      "balance_loss_clip": 1.0542767,
      "balance_loss_mlp": 1.03858113,
      "epoch": 0.1066586502329776,
      "flos": 32160591031680.0,
      "grad_norm": 2.1824496372538023,
      "language_loss": 0.82977128,
      "learning_rate": 3.938672150753041e-06,
      "loss": 0.85164857,
      "num_input_tokens_seen": 38361315,
      "step": 1774,
      "time_per_iteration": 4.039280414581299
    },
    {
      "auxiliary_loss_clip": 0.01149524,
      "auxiliary_loss_mlp": 0.00953471,
      "balance_loss_clip": 1.05328441,
      "balance_loss_mlp": 1.28024817,
      "epoch": 0.10671877348564557,
      "flos": 17785155202560.0,
      "grad_norm": 2.8293273150900915,
      "language_loss": 0.76345521,
      "learning_rate": 3.9385764081380704e-06,
      "loss": 0.78448522,
      "num_input_tokens_seen": 38377425,
      "step": 1775,
      "time_per_iteration": 2.5170490741729736
    },
    {
      "auxiliary_loss_clip": 0.01080124,
      "auxiliary_loss_mlp": 0.01008592,
      "balance_loss_clip": 1.0308938,
      "balance_loss_mlp": 1.00549257,
      "epoch": 0.10677889673831355,
      "flos": 63510177813120.0,
      "grad_norm": 0.8322799037707013,
      "language_loss": 0.57461029,
      "learning_rate": 3.9384805920120876e-06,
      "loss": 0.59549749,
      "num_input_tokens_seen": 38440275,
      "step": 1776,
      "time_per_iteration": 3.1108837127685547
    },
    {
      "auxiliary_loss_clip": 0.01147282,
      "auxiliary_loss_mlp": 0.01066076,
      "balance_loss_clip": 1.0519371,
      "balance_loss_mlp": 1.04153109,
      "epoch": 0.10683901999098151,
      "flos": 22017694400640.0,
      "grad_norm": 1.8658450700407851,
      "language_loss": 0.83312041,
      "learning_rate": 3.938384702378727e-06,
      "loss": 0.85525399,
      "num_input_tokens_seen": 38461820,
      "step": 1777,
      "time_per_iteration": 4.040153741836548
    },
    {
      "auxiliary_loss_clip": 0.01114431,
      "auxiliary_loss_mlp": 0.0093014,
      "balance_loss_clip": 1.05465508,
      "balance_loss_mlp": 1.23516345,
      "epoch": 0.10689914324364948,
      "flos": 25042952551680.0,
      "grad_norm": 1.8128607428067909,
      "language_loss": 0.87356329,
      "learning_rate": 3.938288739241625e-06,
      "loss": 0.89400899,
      "num_input_tokens_seen": 38482235,
      "step": 1778,
      "time_per_iteration": 2.647101879119873
    },
    {
      "auxiliary_loss_clip": 0.01134409,
      "auxiliary_loss_mlp": 0.00932054,
      "balance_loss_clip": 1.05395174,
      "balance_loss_mlp": 1.24022281,
      "epoch": 0.10695926649631746,
      "flos": 16435129507200.0,
      "grad_norm": 1.9887797655935426,
      "language_loss": 0.84043854,
      "learning_rate": 3.938192702604417e-06,
      "loss": 0.86110318,
      "num_input_tokens_seen": 38500690,
      "step": 1779,
      "time_per_iteration": 2.6101789474487305
    },
    {
      "auxiliary_loss_clip": 0.01138533,
      "auxiliary_loss_mlp": 0.00925509,
      "balance_loss_clip": 1.05352974,
      "balance_loss_mlp": 1.22474432,
      "epoch": 0.10701938974898542,
      "flos": 16979211792000.0,
      "grad_norm": 2.002449799199395,
      "language_loss": 0.67376244,
      "learning_rate": 3.9380965924707495e-06,
      "loss": 0.69440281,
      "num_input_tokens_seen": 38518405,
      "step": 1780,
      "time_per_iteration": 2.549173593521118
    },
    {
      "auxiliary_loss_clip": 0.01162453,
      "auxiliary_loss_mlp": 0.0104931,
      "balance_loss_clip": 1.05594707,
      "balance_loss_mlp": 1.02832913,
      "epoch": 0.10707951300165339,
      "flos": 15888102307200.0,
      "grad_norm": 2.8166631777892346,
      "language_loss": 0.91754705,
      "learning_rate": 3.938000408844265e-06,
      "loss": 0.93966466,
      "num_input_tokens_seen": 38535060,
      "step": 1781,
      "time_per_iteration": 2.4879205226898193
    },
    {
      "auxiliary_loss_clip": 0.01125462,
      "auxiliary_loss_mlp": 0.01054765,
      "balance_loss_clip": 1.05047202,
      "balance_loss_mlp": 1.03427243,
      "epoch": 0.10713963625432135,
      "flos": 14247164361600.0,
      "grad_norm": 1.936880045850461,
      "language_loss": 0.79316783,
      "learning_rate": 3.9379041517286105e-06,
      "loss": 0.81497014,
      "num_input_tokens_seen": 38552855,
      "step": 1782,
      "time_per_iteration": 2.5683634281158447
    },
    {
      "auxiliary_loss_clip": 0.01156334,
      "auxiliary_loss_mlp": 0.01052677,
      "balance_loss_clip": 1.05509377,
      "balance_loss_mlp": 1.03154135,
      "epoch": 0.10719975950698933,
      "flos": 16756780821120.0,
      "grad_norm": 2.103230605396249,
      "language_loss": 0.78852177,
      "learning_rate": 3.937807821127436e-06,
      "loss": 0.81061196,
      "num_input_tokens_seen": 38570075,
      "step": 1783,
      "time_per_iteration": 2.5359129905700684
    },
    {
      "auxiliary_loss_clip": 0.01157104,
      "auxiliary_loss_mlp": 0.01052217,
      "balance_loss_clip": 1.05217326,
      "balance_loss_mlp": 1.03077102,
      "epoch": 0.1072598827596573,
      "flos": 22710626645760.0,
      "grad_norm": 2.014188304373719,
      "language_loss": 0.86534739,
      "learning_rate": 3.937711417044395e-06,
      "loss": 0.88744056,
      "num_input_tokens_seen": 38587970,
      "step": 1784,
      "time_per_iteration": 2.548200845718384
    },
    {
      "auxiliary_loss_clip": 0.01147063,
      "auxiliary_loss_mlp": 0.01054039,
      "balance_loss_clip": 1.05259371,
      "balance_loss_mlp": 1.03253376,
      "epoch": 0.10732000601232526,
      "flos": 23258264376960.0,
      "grad_norm": 2.3463612640585962,
      "language_loss": 1.00922799,
      "learning_rate": 3.937614939483143e-06,
      "loss": 1.03123903,
      "num_input_tokens_seen": 38605840,
      "step": 1785,
      "time_per_iteration": 2.5721778869628906
    },
    {
      "auxiliary_loss_clip": 0.01164347,
      "auxiliary_loss_mlp": 0.0105859,
      "balance_loss_clip": 1.05583596,
      "balance_loss_mlp": 1.03771615,
      "epoch": 0.10738012926499324,
      "flos": 24207060176640.0,
      "grad_norm": 1.4212215929391596,
      "language_loss": 0.84679282,
      "learning_rate": 3.937518388447339e-06,
      "loss": 0.86902225,
      "num_input_tokens_seen": 38627070,
      "step": 1786,
      "time_per_iteration": 2.542224168777466
    },
    {
      "auxiliary_loss_clip": 0.0117655,
      "auxiliary_loss_mlp": 0.01054945,
      "balance_loss_clip": 1.05225635,
      "balance_loss_mlp": 1.03240252,
      "epoch": 0.1074402525176612,
      "flos": 20923065383040.0,
      "grad_norm": 1.9138394717432452,
      "language_loss": 0.78682321,
      "learning_rate": 3.937421763940642e-06,
      "loss": 0.80913818,
      "num_input_tokens_seen": 38645840,
      "step": 1787,
      "time_per_iteration": 2.4811363220214844
    },
    {
      "auxiliary_loss_clip": 0.01167691,
      "auxiliary_loss_mlp": 0.01045219,
      "balance_loss_clip": 1.05416906,
      "balance_loss_mlp": 1.02411902,
      "epoch": 0.10750037577032917,
      "flos": 16946928443520.0,
      "grad_norm": 2.2144798946887603,
      "language_loss": 0.82744288,
      "learning_rate": 3.937325065966719e-06,
      "loss": 0.84957194,
      "num_input_tokens_seen": 38664770,
      "step": 1788,
      "time_per_iteration": 2.486591339111328
    },
    {
      "auxiliary_loss_clip": 0.01174604,
      "auxiliary_loss_mlp": 0.0105649,
      "balance_loss_clip": 1.05280089,
      "balance_loss_mlp": 1.03646326,
      "epoch": 0.10756049902299715,
      "flos": 20266546550400.0,
      "grad_norm": 1.8472713126561915,
      "language_loss": 0.77797037,
      "learning_rate": 3.9372282945292335e-06,
      "loss": 0.80028129,
      "num_input_tokens_seen": 38683865,
      "step": 1789,
      "time_per_iteration": 2.4798715114593506
    },
    {
      "auxiliary_loss_clip": 0.01177027,
      "auxiliary_loss_mlp": 0.01057915,
      "balance_loss_clip": 1.05381703,
      "balance_loss_mlp": 1.03476453,
      "epoch": 0.10762062227566511,
      "flos": 23586523793280.0,
      "grad_norm": 3.184103868463773,
      "language_loss": 0.74557352,
      "learning_rate": 3.937131449631859e-06,
      "loss": 0.767923,
      "num_input_tokens_seen": 38702485,
      "step": 1790,
      "time_per_iteration": 2.511636257171631
    },
    {
      "auxiliary_loss_clip": 0.01169726,
      "auxiliary_loss_mlp": 0.00938638,
      "balance_loss_clip": 1.05489826,
      "balance_loss_mlp": 1.24680865,
      "epoch": 0.10768074552833308,
      "flos": 24310626065280.0,
      "grad_norm": 2.4968458708852888,
      "language_loss": 0.787543,
      "learning_rate": 3.9370345312782645e-06,
      "loss": 0.80862665,
      "num_input_tokens_seen": 38722475,
      "step": 1791,
      "time_per_iteration": 2.568056344985962
    },
    {
      "auxiliary_loss_clip": 0.01133596,
      "auxiliary_loss_mlp": 0.01063447,
      "balance_loss_clip": 1.05383611,
      "balance_loss_mlp": 1.04086852,
      "epoch": 0.10774086878100106,
      "flos": 25299965341440.0,
      "grad_norm": 2.200417121974988,
      "language_loss": 0.70931971,
      "learning_rate": 3.936937539472126e-06,
      "loss": 0.73129016,
      "num_input_tokens_seen": 38743285,
      "step": 1792,
      "time_per_iteration": 2.600085735321045
    },
    {
      "auxiliary_loss_clip": 0.01142262,
      "auxiliary_loss_mlp": 0.01045445,
      "balance_loss_clip": 1.04931021,
      "balance_loss_mlp": 1.02299786,
      "epoch": 0.10780099203366902,
      "flos": 22054035985920.0,
      "grad_norm": 2.1000180608698455,
      "language_loss": 0.76260114,
      "learning_rate": 3.9368404742171236e-06,
      "loss": 0.78447825,
      "num_input_tokens_seen": 38763035,
      "step": 1793,
      "time_per_iteration": 2.572072744369507
    },
    {
      "auxiliary_loss_clip": 0.01116791,
      "auxiliary_loss_mlp": 0.01063843,
      "balance_loss_clip": 1.05621743,
      "balance_loss_mlp": 1.04195666,
      "epoch": 0.10786111528633699,
      "flos": 22747471021440.0,
      "grad_norm": 1.6480251302865685,
      "language_loss": 0.84855539,
      "learning_rate": 3.936743335516936e-06,
      "loss": 0.8703618,
      "num_input_tokens_seen": 38784900,
      "step": 1794,
      "time_per_iteration": 2.609758138656616
    },
    {
      "auxiliary_loss_clip": 0.0111417,
      "auxiliary_loss_mlp": 0.01050221,
      "balance_loss_clip": 1.04893255,
      "balance_loss_mlp": 1.0270226,
      "epoch": 0.10792123853900495,
      "flos": 20851064570880.0,
      "grad_norm": 1.7379188196477193,
      "language_loss": 0.74893898,
      "learning_rate": 3.936646123375246e-06,
      "loss": 0.77058291,
      "num_input_tokens_seen": 38804695,
      "step": 1795,
      "time_per_iteration": 2.638037919998169
    },
    {
      "auxiliary_loss_clip": 0.01120903,
      "auxiliary_loss_mlp": 0.01055481,
      "balance_loss_clip": 1.04834831,
      "balance_loss_mlp": 1.03340387,
      "epoch": 0.10798136179167293,
      "flos": 17748705876480.0,
      "grad_norm": 2.871285409630518,
      "language_loss": 0.816194,
      "learning_rate": 3.936548837795741e-06,
      "loss": 0.83795786,
      "num_input_tokens_seen": 38822395,
      "step": 1796,
      "time_per_iteration": 2.5850188732147217
    },
    {
      "auxiliary_loss_clip": 0.01141756,
      "auxiliary_loss_mlp": 0.01079207,
      "balance_loss_clip": 1.05691266,
      "balance_loss_mlp": 1.05407786,
      "epoch": 0.1080414850443409,
      "flos": 13589639948160.0,
      "grad_norm": 3.419364175750862,
      "language_loss": 0.73908269,
      "learning_rate": 3.936451478782111e-06,
      "loss": 0.76129234,
      "num_input_tokens_seen": 38839865,
      "step": 1797,
      "time_per_iteration": 2.557633638381958
    },
    {
      "auxiliary_loss_clip": 0.01154181,
      "auxiliary_loss_mlp": 0.01048792,
      "balance_loss_clip": 1.04969335,
      "balance_loss_mlp": 1.02856231,
      "epoch": 0.10810160829700886,
      "flos": 16253421580800.0,
      "grad_norm": 2.0939263054375337,
      "language_loss": 0.81682545,
      "learning_rate": 3.936354046338046e-06,
      "loss": 0.83885515,
      "num_input_tokens_seen": 38857300,
      "step": 1798,
      "time_per_iteration": 2.513463020324707
    },
    {
      "auxiliary_loss_clip": 0.01138673,
      "auxiliary_loss_mlp": 0.01051201,
      "balance_loss_clip": 1.05210626,
      "balance_loss_mlp": 1.02864647,
      "epoch": 0.10816173154967684,
      "flos": 15158002464000.0,
      "grad_norm": 3.030944472805612,
      "language_loss": 0.85908282,
      "learning_rate": 3.936256540467242e-06,
      "loss": 0.88098145,
      "num_input_tokens_seen": 38874960,
      "step": 1799,
      "time_per_iteration": 2.5537757873535156
    },
    {
      "auxiliary_loss_clip": 0.01150948,
      "auxiliary_loss_mlp": 0.01060112,
      "balance_loss_clip": 1.0533489,
      "balance_loss_mlp": 1.03945303,
      "epoch": 0.10822185480234481,
      "flos": 17785334770560.0,
      "grad_norm": 2.288696189098336,
      "language_loss": 0.77842802,
      "learning_rate": 3.9361589611733955e-06,
      "loss": 0.80053866,
      "num_input_tokens_seen": 38893610,
      "step": 1800,
      "time_per_iteration": 2.525588274002075
    },
    {
      "auxiliary_loss_clip": 0.01174462,
      "auxiliary_loss_mlp": 0.01047272,
      "balance_loss_clip": 1.0539434,
      "balance_loss_mlp": 1.02729225,
      "epoch": 0.10828197805501277,
      "flos": 25556654908800.0,
      "grad_norm": 1.8514088340532386,
      "language_loss": 0.73017943,
      "learning_rate": 3.9360613084602075e-06,
      "loss": 0.7523967,
      "num_input_tokens_seen": 38913485,
      "step": 1801,
      "time_per_iteration": 2.525867462158203
    },
    {
      "auxiliary_loss_clip": 0.0118406,
      "auxiliary_loss_mlp": 0.01048449,
      "balance_loss_clip": 1.05555606,
      "balance_loss_mlp": 1.02819586,
      "epoch": 0.10834210130768075,
      "flos": 28984435845120.0,
      "grad_norm": 2.010276646442789,
      "language_loss": 0.6619364,
      "learning_rate": 3.935963582331381e-06,
      "loss": 0.6842615,
      "num_input_tokens_seen": 38935650,
      "step": 1802,
      "time_per_iteration": 2.537238121032715
    },
    {
      "auxiliary_loss_clip": 0.01155606,
      "auxiliary_loss_mlp": 0.01060518,
      "balance_loss_clip": 1.05577672,
      "balance_loss_mlp": 1.03880954,
      "epoch": 0.10840222456034872,
      "flos": 20264212166400.0,
      "grad_norm": 1.8289527022952365,
      "language_loss": 0.81348848,
      "learning_rate": 3.935865782790621e-06,
      "loss": 0.83564973,
      "num_input_tokens_seen": 38954130,
      "step": 1803,
      "time_per_iteration": 2.522123098373413
    },
    {
      "auxiliary_loss_clip": 0.01157875,
      "auxiliary_loss_mlp": 0.0105661,
      "balance_loss_clip": 1.05198944,
      "balance_loss_mlp": 1.03433013,
      "epoch": 0.10846234781301668,
      "flos": 19863054097920.0,
      "grad_norm": 1.7210735345813057,
      "language_loss": 0.91151452,
      "learning_rate": 3.9357679098416365e-06,
      "loss": 0.93365937,
      "num_input_tokens_seen": 38972905,
      "step": 1804,
      "time_per_iteration": 2.494198799133301
    },
    {
      "auxiliary_loss_clip": 0.01127104,
      "auxiliary_loss_mlp": 0.01052659,
      "balance_loss_clip": 1.04971588,
      "balance_loss_mlp": 1.03033113,
      "epoch": 0.10852247106568465,
      "flos": 26469037296000.0,
      "grad_norm": 2.0311915269725773,
      "language_loss": 0.76216924,
      "learning_rate": 3.935669963488139e-06,
      "loss": 0.7839669,
      "num_input_tokens_seen": 38993255,
      "step": 1805,
      "time_per_iteration": 2.656996488571167
    },
    {
      "auxiliary_loss_clip": 0.01146295,
      "auxiliary_loss_mlp": 0.01046556,
      "balance_loss_clip": 1.0540185,
      "balance_loss_mlp": 1.02648091,
      "epoch": 0.10858259431835263,
      "flos": 30081506987520.0,
      "grad_norm": 1.8972499773626876,
      "language_loss": 0.86001384,
      "learning_rate": 3.935571943733843e-06,
      "loss": 0.88194239,
      "num_input_tokens_seen": 39012610,
      "step": 1806,
      "time_per_iteration": 2.5956003665924072
    },
    {
      "auxiliary_loss_clip": 0.01165673,
      "auxiliary_loss_mlp": 0.00864615,
      "balance_loss_clip": 1.05128169,
      "balance_loss_mlp": 1.0969882,
      "epoch": 0.10864271757102059,
      "flos": 19063180085760.0,
      "grad_norm": 2.1070937393849083,
      "language_loss": 0.81135738,
      "learning_rate": 3.9354738505824635e-06,
      "loss": 0.83166021,
      "num_input_tokens_seen": 39030120,
      "step": 1807,
      "time_per_iteration": 3.880457878112793
    },
    {
      "auxiliary_loss_clip": 0.01135429,
      "auxiliary_loss_mlp": 0.01050362,
      "balance_loss_clip": 1.05027163,
      "balance_loss_mlp": 1.03102648,
      "epoch": 0.10870284082368856,
      "flos": 24715052271360.0,
      "grad_norm": 1.8669969049392539,
      "language_loss": 0.79192108,
      "learning_rate": 3.9353756840377225e-06,
      "loss": 0.813779,
      "num_input_tokens_seen": 39049875,
      "step": 1808,
      "time_per_iteration": 2.599863290786743
    },
    {
      "auxiliary_loss_clip": 0.01152373,
      "auxiliary_loss_mlp": 0.01052007,
      "balance_loss_clip": 1.05400634,
      "balance_loss_mlp": 1.03146768,
      "epoch": 0.10876296407635654,
      "flos": 20627663932800.0,
      "grad_norm": 1.7839947534501197,
      "language_loss": 0.78929681,
      "learning_rate": 3.935277444103342e-06,
      "loss": 0.81134057,
      "num_input_tokens_seen": 39068935,
      "step": 1809,
      "time_per_iteration": 2.5424091815948486
    },
    {
      "auxiliary_loss_clip": 0.01175535,
      "auxiliary_loss_mlp": 0.01051487,
      "balance_loss_clip": 1.05220628,
      "balance_loss_mlp": 1.03074503,
      "epoch": 0.1088230873290245,
      "flos": 21579835610880.0,
      "grad_norm": 2.5145839146064843,
      "language_loss": 0.85074115,
      "learning_rate": 3.935179130783046e-06,
      "loss": 0.87301135,
      "num_input_tokens_seen": 39087370,
      "step": 1810,
      "time_per_iteration": 3.8687565326690674
    },
    {
      "auxiliary_loss_clip": 0.01127238,
      "auxiliary_loss_mlp": 0.01054796,
      "balance_loss_clip": 1.05304861,
      "balance_loss_mlp": 1.03158581,
      "epoch": 0.10888321058169247,
      "flos": 26469037296000.0,
      "grad_norm": 1.8160795807416954,
      "language_loss": 0.63268286,
      "learning_rate": 3.935080744080564e-06,
      "loss": 0.65450323,
      "num_input_tokens_seen": 39106635,
      "step": 1811,
      "time_per_iteration": 2.623990058898926
    },
    {
      "auxiliary_loss_clip": 0.01150654,
      "auxiliary_loss_mlp": 0.01045606,
      "balance_loss_clip": 1.05331945,
      "balance_loss_mlp": 1.02497077,
      "epoch": 0.10894333383436045,
      "flos": 25848608653440.0,
      "grad_norm": 2.0305890106447784,
      "language_loss": 0.74137306,
      "learning_rate": 3.934982283999626e-06,
      "loss": 0.7633357,
      "num_input_tokens_seen": 39126335,
      "step": 1812,
      "time_per_iteration": 2.580094814300537
    },
    {
      "auxiliary_loss_clip": 0.0114117,
      "auxiliary_loss_mlp": 0.0104843,
      "balance_loss_clip": 1.0527339,
      "balance_loss_mlp": 1.02694893,
      "epoch": 0.10900345708702841,
      "flos": 19537093152000.0,
      "grad_norm": 2.0903664712275773,
      "language_loss": 0.72594631,
      "learning_rate": 3.934883750543966e-06,
      "loss": 0.74784231,
      "num_input_tokens_seen": 39144820,
      "step": 1813,
      "time_per_iteration": 3.9049789905548096
    },
    {
      "auxiliary_loss_clip": 0.0114043,
      "auxiliary_loss_mlp": 0.01047362,
      "balance_loss_clip": 1.05502391,
      "balance_loss_mlp": 1.02654862,
      "epoch": 0.10906358033969638,
      "flos": 23623296341760.0,
      "grad_norm": 1.8458425012945472,
      "language_loss": 0.82677799,
      "learning_rate": 3.93478514371732e-06,
      "loss": 0.84865588,
      "num_input_tokens_seen": 39165945,
      "step": 1814,
      "time_per_iteration": 2.5530643463134766
    },
    {
      "auxiliary_loss_clip": 0.01139016,
      "auxiliary_loss_mlp": 0.0105091,
      "balance_loss_clip": 1.05373335,
      "balance_loss_mlp": 1.03052509,
      "epoch": 0.10912370359236434,
      "flos": 21214731818880.0,
      "grad_norm": 2.4636313301600823,
      "language_loss": 0.83993071,
      "learning_rate": 3.934686463523429e-06,
      "loss": 0.86183,
      "num_input_tokens_seen": 39183520,
      "step": 1815,
      "time_per_iteration": 2.587502956390381
    },
    {
      "auxiliary_loss_clip": 0.01142148,
      "auxiliary_loss_mlp": 0.01050636,
      "balance_loss_clip": 1.05219221,
      "balance_loss_mlp": 1.02851069,
      "epoch": 0.10918382684503232,
      "flos": 13553190622080.0,
      "grad_norm": 2.5815783525169733,
      "language_loss": 0.71825099,
      "learning_rate": 3.9345877099660315e-06,
      "loss": 0.74017882,
      "num_input_tokens_seen": 39201190,
      "step": 1816,
      "time_per_iteration": 2.5558204650878906
    },
    {
      "auxiliary_loss_clip": 0.01163107,
      "auxiliary_loss_mlp": 0.01057687,
      "balance_loss_clip": 1.05236435,
      "balance_loss_mlp": 1.0360986,
      "epoch": 0.10924395009770028,
      "flos": 27964321591680.0,
      "grad_norm": 3.6250761178373274,
      "language_loss": 0.72785282,
      "learning_rate": 3.9344888830488744e-06,
      "loss": 0.75006074,
      "num_input_tokens_seen": 39221210,
      "step": 1817,
      "time_per_iteration": 3.9970643520355225
    },
    {
      "auxiliary_loss_clip": 0.0111492,
      "auxiliary_loss_mlp": 0.01049322,
      "balance_loss_clip": 1.05173159,
      "balance_loss_mlp": 1.0277456,
      "epoch": 0.10930407335036825,
      "flos": 25593750679680.0,
      "grad_norm": 1.571266706376158,
      "language_loss": 0.67180085,
      "learning_rate": 3.934389982775706e-06,
      "loss": 0.6934433,
      "num_input_tokens_seen": 39242025,
      "step": 1818,
      "time_per_iteration": 2.7400951385498047
    },
    {
      "auxiliary_loss_clip": 0.01151881,
      "auxiliary_loss_mlp": 0.01058172,
      "balance_loss_clip": 1.05681348,
      "balance_loss_mlp": 1.03649926,
      "epoch": 0.10936419660303623,
      "flos": 18406194376320.0,
      "grad_norm": 2.20821593452066,
      "language_loss": 0.73447675,
      "learning_rate": 3.934291009150275e-06,
      "loss": 0.75657725,
      "num_input_tokens_seen": 39259870,
      "step": 1819,
      "time_per_iteration": 2.6866824626922607
    },
    {
      "auxiliary_loss_clip": 0.01141015,
      "auxiliary_loss_mlp": 0.00914971,
      "balance_loss_clip": 1.05416584,
      "balance_loss_mlp": 1.19809675,
      "epoch": 0.1094243198557042,
      "flos": 23840052963840.0,
      "grad_norm": 2.6512341078368706,
      "language_loss": 0.74268186,
      "learning_rate": 3.934191962176335e-06,
      "loss": 0.76324171,
      "num_input_tokens_seen": 39278500,
      "step": 1820,
      "time_per_iteration": 2.5806283950805664
    },
    {
      "auxiliary_loss_clip": 0.01177392,
      "auxiliary_loss_mlp": 0.01054841,
      "balance_loss_clip": 1.05644917,
      "balance_loss_mlp": 1.03300142,
      "epoch": 0.10948444310837216,
      "flos": 14643940970880.0,
      "grad_norm": 2.736113311684918,
      "language_loss": 0.82928252,
      "learning_rate": 3.934092841857642e-06,
      "loss": 0.85160482,
      "num_input_tokens_seen": 39294800,
      "step": 1821,
      "time_per_iteration": 2.451608419418335
    },
    {
      "auxiliary_loss_clip": 0.01146086,
      "auxiliary_loss_mlp": 0.01051777,
      "balance_loss_clip": 1.05366707,
      "balance_loss_mlp": 1.03183317,
      "epoch": 0.10954456636104014,
      "flos": 27818811596160.0,
      "grad_norm": 2.9186865837401084,
      "language_loss": 0.75937796,
      "learning_rate": 3.933993648197955e-06,
      "loss": 0.78135657,
      "num_input_tokens_seen": 39314625,
      "step": 1822,
      "time_per_iteration": 2.6391220092773438
    },
    {
      "auxiliary_loss_clip": 0.01141527,
      "auxiliary_loss_mlp": 0.01050522,
      "balance_loss_clip": 1.04971313,
      "balance_loss_mlp": 1.0306859,
      "epoch": 0.1096046896137081,
      "flos": 33620934372480.0,
      "grad_norm": 1.7272316106548256,
      "language_loss": 0.80167818,
      "learning_rate": 3.933894381201034e-06,
      "loss": 0.82359874,
      "num_input_tokens_seen": 39336465,
      "step": 1823,
      "time_per_iteration": 2.645444393157959
    },
    {
      "auxiliary_loss_clip": 0.01148588,
      "auxiliary_loss_mlp": 0.01045773,
      "balance_loss_clip": 1.05550957,
      "balance_loss_mlp": 1.02567434,
      "epoch": 0.10966481286637607,
      "flos": 26980010219520.0,
      "grad_norm": 1.4515807839594805,
      "language_loss": 0.79629862,
      "learning_rate": 3.933795040870645e-06,
      "loss": 0.81824225,
      "num_input_tokens_seen": 39357930,
      "step": 1824,
      "time_per_iteration": 2.5845563411712646
    },
    {
      "auxiliary_loss_clip": 0.01140257,
      "auxiliary_loss_mlp": 0.01058076,
      "balance_loss_clip": 1.05330133,
      "balance_loss_mlp": 1.03807282,
      "epoch": 0.10972493611904403,
      "flos": 23036551678080.0,
      "grad_norm": 1.887021416441878,
      "language_loss": 0.878443,
      "learning_rate": 3.933695627210554e-06,
      "loss": 0.90042633,
      "num_input_tokens_seen": 39376380,
      "step": 1825,
      "time_per_iteration": 2.547545909881592
    },
    {
      "auxiliary_loss_clip": 0.01126402,
      "auxiliary_loss_mlp": 0.01055808,
      "balance_loss_clip": 1.04952133,
      "balance_loss_mlp": 1.03562605,
      "epoch": 0.10978505937171201,
      "flos": 38104632443520.0,
      "grad_norm": 1.82050709045123,
      "language_loss": 0.76530933,
      "learning_rate": 3.933596140224532e-06,
      "loss": 0.78713137,
      "num_input_tokens_seen": 39399935,
      "step": 1826,
      "time_per_iteration": 2.747860908508301
    },
    {
      "auxiliary_loss_clip": 0.01084699,
      "auxiliary_loss_mlp": 0.01045711,
      "balance_loss_clip": 1.04171455,
      "balance_loss_mlp": 1.04203939,
      "epoch": 0.10984518262437998,
      "flos": 59849694616320.0,
      "grad_norm": 0.8373417760297183,
      "language_loss": 0.54928392,
      "learning_rate": 3.93349657991635e-06,
      "loss": 0.57058799,
      "num_input_tokens_seen": 39460685,
      "step": 1827,
      "time_per_iteration": 3.135824203491211
    },
    {
      "auxiliary_loss_clip": 0.01096369,
      "auxiliary_loss_mlp": 0.01024397,
      "balance_loss_clip": 1.04514062,
      "balance_loss_mlp": 1.02060568,
      "epoch": 0.10990530587704794,
      "flos": 66719837410560.0,
      "grad_norm": 0.7343750495826956,
      "language_loss": 0.55379772,
      "learning_rate": 3.933396946289784e-06,
      "loss": 0.57500529,
      "num_input_tokens_seen": 39524765,
      "step": 1828,
      "time_per_iteration": 3.1334316730499268
    },
    {
      "auxiliary_loss_clip": 0.01154213,
      "auxiliary_loss_mlp": 0.01051168,
      "balance_loss_clip": 1.05446637,
      "balance_loss_mlp": 1.0300082,
      "epoch": 0.10996542912971592,
      "flos": 25447199189760.0,
      "grad_norm": 2.383963141557744,
      "language_loss": 0.840554,
      "learning_rate": 3.933297239348612e-06,
      "loss": 0.86260784,
      "num_input_tokens_seen": 39543640,
      "step": 1829,
      "time_per_iteration": 2.577261447906494
    },
    {
      "auxiliary_loss_clip": 0.01124583,
      "auxiliary_loss_mlp": 0.0104636,
      "balance_loss_clip": 1.05349362,
      "balance_loss_mlp": 1.02508163,
      "epoch": 0.11002555238238389,
      "flos": 44018186186880.0,
      "grad_norm": 1.968336757415494,
      "language_loss": 0.88815951,
      "learning_rate": 3.933197459096614e-06,
      "loss": 0.90986896,
      "num_input_tokens_seen": 39567525,
      "step": 1830,
      "time_per_iteration": 2.8070120811462402
    },
    {
      "auxiliary_loss_clip": 0.0106435,
      "auxiliary_loss_mlp": 0.01008486,
      "balance_loss_clip": 1.03776455,
      "balance_loss_mlp": 1.00474286,
      "epoch": 0.11008567563505185,
      "flos": 54065133590400.0,
      "grad_norm": 0.6900487421878563,
      "language_loss": 0.55497819,
      "learning_rate": 3.9330976055375756e-06,
      "loss": 0.57570654,
      "num_input_tokens_seen": 39628470,
      "step": 1831,
      "time_per_iteration": 3.13627290725708
    },
    {
      "auxiliary_loss_clip": 0.0115311,
      "auxiliary_loss_mlp": 0.01065887,
      "balance_loss_clip": 1.05626249,
      "balance_loss_mlp": 1.04369032,
      "epoch": 0.11014579888771983,
      "flos": 24243150366720.0,
      "grad_norm": 2.3406376724747484,
      "language_loss": 0.90758264,
      "learning_rate": 3.932997678675282e-06,
      "loss": 0.92977256,
      "num_input_tokens_seen": 39646670,
      "step": 1832,
      "time_per_iteration": 2.5892183780670166
    },
    {
      "auxiliary_loss_clip": 0.01088983,
      "auxiliary_loss_mlp": 0.01015858,
      "balance_loss_clip": 1.04798102,
      "balance_loss_mlp": 1.01197195,
      "epoch": 0.1102059221403878,
      "flos": 57743965658880.0,
      "grad_norm": 0.7161625606429809,
      "language_loss": 0.59867024,
      "learning_rate": 3.932897678513523e-06,
      "loss": 0.61971867,
      "num_input_tokens_seen": 39712915,
      "step": 1833,
      "time_per_iteration": 3.1634912490844727
    },
    {
      "auxiliary_loss_clip": 0.01163809,
      "auxiliary_loss_mlp": 0.01049993,
      "balance_loss_clip": 1.05105364,
      "balance_loss_mlp": 1.02854776,
      "epoch": 0.11026604539305576,
      "flos": 16795923667200.0,
      "grad_norm": 2.595673732252315,
      "language_loss": 0.80191374,
      "learning_rate": 3.93279760505609e-06,
      "loss": 0.8240518,
      "num_input_tokens_seen": 39730650,
      "step": 1834,
      "time_per_iteration": 2.5062639713287354
    },
    {
      "auxiliary_loss_clip": 0.01134198,
      "auxiliary_loss_mlp": 0.01059757,
      "balance_loss_clip": 1.05248809,
      "balance_loss_mlp": 1.03659463,
      "epoch": 0.11032616864572373,
      "flos": 23988076911360.0,
      "grad_norm": 2.432153527348238,
      "language_loss": 0.90856171,
      "learning_rate": 3.932697458306779e-06,
      "loss": 0.93050128,
      "num_input_tokens_seen": 39751065,
      "step": 1835,
      "time_per_iteration": 2.6446104049682617
    },
    {
      "auxiliary_loss_clip": 0.01127064,
      "auxiliary_loss_mlp": 0.01062591,
      "balance_loss_clip": 1.05323362,
      "balance_loss_mlp": 1.03858209,
      "epoch": 0.1103862918983917,
      "flos": 19683141851520.0,
      "grad_norm": 2.2182426790260976,
      "language_loss": 0.63614547,
      "learning_rate": 3.932597238269386e-06,
      "loss": 0.65804201,
      "num_input_tokens_seen": 39769245,
      "step": 1836,
      "time_per_iteration": 2.57279372215271
    },
    {
      "auxiliary_loss_clip": 0.01135004,
      "auxiliary_loss_mlp": 0.01054659,
      "balance_loss_clip": 1.05108941,
      "balance_loss_mlp": 1.03420293,
      "epoch": 0.11044641515105967,
      "flos": 32160878340480.0,
      "grad_norm": 2.1327394323874866,
      "language_loss": 0.72571188,
      "learning_rate": 3.932496944947711e-06,
      "loss": 0.74760854,
      "num_input_tokens_seen": 39790830,
      "step": 1837,
      "time_per_iteration": 2.700165271759033
    },
    {
      "auxiliary_loss_clip": 0.01159941,
      "auxiliary_loss_mlp": 0.01060546,
      "balance_loss_clip": 1.05262589,
      "balance_loss_mlp": 1.0398159,
      "epoch": 0.11050653840372764,
      "flos": 16689233295360.0,
      "grad_norm": 2.294818244377463,
      "language_loss": 0.78771389,
      "learning_rate": 3.93239657834556e-06,
      "loss": 0.80991876,
      "num_input_tokens_seen": 39809475,
      "step": 1838,
      "time_per_iteration": 2.5129714012145996
    },
    {
      "auxiliary_loss_clip": 0.01149173,
      "auxiliary_loss_mlp": 0.01062505,
      "balance_loss_clip": 1.05377603,
      "balance_loss_mlp": 1.04201281,
      "epoch": 0.11056666165639562,
      "flos": 21208877902080.0,
      "grad_norm": 2.2882681140596763,
      "language_loss": 0.71975917,
      "learning_rate": 3.932296138466736e-06,
      "loss": 0.74187589,
      "num_input_tokens_seen": 39826355,
      "step": 1839,
      "time_per_iteration": 2.5251216888427734
    },
    {
      "auxiliary_loss_clip": 0.01181104,
      "auxiliary_loss_mlp": 0.0089095,
      "balance_loss_clip": 1.05805528,
      "balance_loss_mlp": 1.14669549,
      "epoch": 0.11062678490906358,
      "flos": 19165488998400.0,
      "grad_norm": 2.1813690172982936,
      "language_loss": 0.7860378,
      "learning_rate": 3.93219562531505e-06,
      "loss": 0.80675828,
      "num_input_tokens_seen": 39845335,
      "step": 1840,
      "time_per_iteration": 2.4960312843322754
    },
    {
      "auxiliary_loss_clip": 0.01152629,
      "auxiliary_loss_mlp": 0.01051731,
      "balance_loss_clip": 1.05122828,
      "balance_loss_mlp": 1.02953422,
      "epoch": 0.11068690816173155,
      "flos": 24895287740160.0,
      "grad_norm": 1.6216683001745351,
      "language_loss": 0.88125634,
      "learning_rate": 3.932095038894311e-06,
      "loss": 0.90329987,
      "num_input_tokens_seen": 39865065,
      "step": 1841,
      "time_per_iteration": 2.554553747177124
    },
    {
      "auxiliary_loss_clip": 0.0112823,
      "auxiliary_loss_mlp": 0.01054659,
      "balance_loss_clip": 1.05335593,
      "balance_loss_mlp": 1.03348708,
      "epoch": 0.11074703141439952,
      "flos": 16472368932480.0,
      "grad_norm": 1.9926821948323934,
      "language_loss": 0.90484089,
      "learning_rate": 3.931994379208334e-06,
      "loss": 0.92666978,
      "num_input_tokens_seen": 39882780,
      "step": 1842,
      "time_per_iteration": 2.540863513946533
    },
    {
      "auxiliary_loss_clip": 0.01149516,
      "auxiliary_loss_mlp": 0.01051408,
      "balance_loss_clip": 1.04979372,
      "balance_loss_mlp": 1.03201246,
      "epoch": 0.11080715466706749,
      "flos": 19172420323200.0,
      "grad_norm": 2.477460914049156,
      "language_loss": 0.86271918,
      "learning_rate": 3.931893646260937e-06,
      "loss": 0.88472843,
      "num_input_tokens_seen": 39900295,
      "step": 1843,
      "time_per_iteration": 2.529961585998535
    },
    {
      "auxiliary_loss_clip": 0.01118983,
      "auxiliary_loss_mlp": 0.00865283,
      "balance_loss_clip": 1.05424786,
      "balance_loss_mlp": 1.09868824,
      "epoch": 0.11086727791973545,
      "flos": 27704687109120.0,
      "grad_norm": 1.5962430450048317,
      "language_loss": 0.74646568,
      "learning_rate": 3.931792840055941e-06,
      "loss": 0.76630831,
      "num_input_tokens_seen": 39922075,
      "step": 1844,
      "time_per_iteration": 2.7359042167663574
    },
    {
      "auxiliary_loss_clip": 0.01173984,
      "auxiliary_loss_mlp": 0.01051315,
      "balance_loss_clip": 1.05293655,
      "balance_loss_mlp": 1.0289638,
      "epoch": 0.11092740117240343,
      "flos": 18514967736960.0,
      "grad_norm": 1.9911394909815205,
      "language_loss": 0.75474167,
      "learning_rate": 3.931691960597165e-06,
      "loss": 0.77699471,
      "num_input_tokens_seen": 39940115,
      "step": 1845,
      "time_per_iteration": 3.963844060897827
    },
    {
      "auxiliary_loss_clip": 0.01146522,
      "auxiliary_loss_mlp": 0.01053341,
      "balance_loss_clip": 1.05106926,
      "balance_loss_mlp": 1.03247929,
      "epoch": 0.1109875244250714,
      "flos": 20522446018560.0,
      "grad_norm": 1.5547304070784076,
      "language_loss": 0.76111203,
      "learning_rate": 3.9315910078884375e-06,
      "loss": 0.78311068,
      "num_input_tokens_seen": 39959920,
      "step": 1846,
      "time_per_iteration": 2.553382158279419
    },
    {
      "auxiliary_loss_clip": 0.0116525,
      "auxiliary_loss_mlp": 0.01051166,
      "balance_loss_clip": 1.05264819,
      "balance_loss_mlp": 1.03034043,
      "epoch": 0.11104764767773936,
      "flos": 14098601710080.0,
      "grad_norm": 2.6973114196975447,
      "language_loss": 0.86378586,
      "learning_rate": 3.931489981933584e-06,
      "loss": 0.88594997,
      "num_input_tokens_seen": 39974755,
      "step": 1847,
      "time_per_iteration": 2.4833598136901855
    },
    {
      "auxiliary_loss_clip": 0.01172377,
      "auxiliary_loss_mlp": 0.0104702,
      "balance_loss_clip": 1.05012369,
      "balance_loss_mlp": 1.0250622,
      "epoch": 0.11110777093040733,
      "flos": 20594518657920.0,
      "grad_norm": 2.0812836104282977,
      "language_loss": 0.7696563,
      "learning_rate": 3.931388882736438e-06,
      "loss": 0.79185033,
      "num_input_tokens_seen": 39993355,
      "step": 1848,
      "time_per_iteration": 2.5071895122528076
    },
    {
      "auxiliary_loss_clip": 0.01165956,
      "auxiliary_loss_mlp": 0.0105228,
      "balance_loss_clip": 1.05775547,
      "balance_loss_mlp": 1.03183603,
      "epoch": 0.11116789418307531,
      "flos": 21870065502720.0,
      "grad_norm": 1.8656214800392978,
      "language_loss": 0.77769387,
      "learning_rate": 3.931287710300832e-06,
      "loss": 0.79987621,
      "num_input_tokens_seen": 40012410,
      "step": 1849,
      "time_per_iteration": 3.884786605834961
    },
    {
      "auxiliary_loss_clip": 0.01128372,
      "auxiliary_loss_mlp": 0.00861065,
      "balance_loss_clip": 1.04705453,
      "balance_loss_mlp": 1.09269142,
      "epoch": 0.11122801743574327,
      "flos": 15523106256000.0,
      "grad_norm": 2.9254362634663096,
      "language_loss": 0.71905291,
      "learning_rate": 3.931186464630601e-06,
      "loss": 0.73894727,
      "num_input_tokens_seen": 40029315,
      "step": 1850,
      "time_per_iteration": 2.515484571456909
    },
    {
      "auxiliary_loss_clip": 0.01164432,
      "auxiliary_loss_mlp": 0.01051492,
      "balance_loss_clip": 1.05442584,
      "balance_loss_mlp": 1.02961683,
      "epoch": 0.11128814068841124,
      "flos": 14392279307520.0,
      "grad_norm": 2.4179567592518967,
      "language_loss": 0.81686932,
      "learning_rate": 3.931085145729588e-06,
      "loss": 0.83902854,
      "num_input_tokens_seen": 40045765,
      "step": 1851,
      "time_per_iteration": 3.8807077407836914
    },
    {
      "auxiliary_loss_clip": 0.01157004,
      "auxiliary_loss_mlp": 0.01052031,
      "balance_loss_clip": 1.055107,
      "balance_loss_mlp": 1.03157496,
      "epoch": 0.11134826394107922,
      "flos": 16653933204480.0,
      "grad_norm": 2.4260299337590587,
      "language_loss": 0.8814429,
      "learning_rate": 3.930983753601631e-06,
      "loss": 0.90353322,
      "num_input_tokens_seen": 40061660,
      "step": 1852,
      "time_per_iteration": 2.471973180770874
    },
    {
      "auxiliary_loss_clip": 0.01161991,
      "auxiliary_loss_mlp": 0.01054442,
      "balance_loss_clip": 1.05421484,
      "balance_loss_mlp": 1.03228092,
      "epoch": 0.11140838719374718,
      "flos": 16690993061760.0,
      "grad_norm": 2.027169406133809,
      "language_loss": 0.7195285,
      "learning_rate": 3.930882288250578e-06,
      "loss": 0.74169284,
      "num_input_tokens_seen": 40080180,
      "step": 1853,
      "time_per_iteration": 2.525233268737793
    },
    {
      "auxiliary_loss_clip": 0.01075803,
      "auxiliary_loss_mlp": 0.01007345,
      "balance_loss_clip": 1.02697611,
      "balance_loss_mlp": 1.00329149,
      "epoch": 0.11146851044641515,
      "flos": 60976355587200.0,
      "grad_norm": 0.7731089647395152,
      "language_loss": 0.5361495,
      "learning_rate": 3.930780749680273e-06,
      "loss": 0.55698097,
      "num_input_tokens_seen": 40138910,
      "step": 1854,
      "time_per_iteration": 3.053013563156128
    },
    {
      "auxiliary_loss_clip": 0.01153558,
      "auxiliary_loss_mlp": 0.01050058,
      "balance_loss_clip": 1.05138135,
      "balance_loss_mlp": 1.02771819,
      "epoch": 0.11152863369908313,
      "flos": 22193835719040.0,
      "grad_norm": 2.2827665931082293,
      "language_loss": 0.84803677,
      "learning_rate": 3.9306791378945705e-06,
      "loss": 0.87007296,
      "num_input_tokens_seen": 40157745,
      "step": 1855,
      "time_per_iteration": 4.079181432723999
    },
    {
      "auxiliary_loss_clip": 0.01143661,
      "auxiliary_loss_mlp": 0.01057637,
      "balance_loss_clip": 1.05315256,
      "balance_loss_mlp": 1.03714514,
      "epoch": 0.11158875695175109,
      "flos": 19537524115200.0,
      "grad_norm": 2.914195552531238,
      "language_loss": 0.81766975,
      "learning_rate": 3.9305774528973205e-06,
      "loss": 0.8396827,
      "num_input_tokens_seen": 40175375,
      "step": 1856,
      "time_per_iteration": 2.5623080730438232
    },
    {
      "auxiliary_loss_clip": 0.01159363,
      "auxiliary_loss_mlp": 0.01043203,
      "balance_loss_clip": 1.05191338,
      "balance_loss_mlp": 1.02155435,
      "epoch": 0.11164888020441906,
      "flos": 25442709989760.0,
      "grad_norm": 1.6935128136133162,
      "language_loss": 0.82834935,
      "learning_rate": 3.93047569469238e-06,
      "loss": 0.85037494,
      "num_input_tokens_seen": 40195715,
      "step": 1857,
      "time_per_iteration": 2.5342061519622803
    },
    {
      "auxiliary_loss_clip": 0.01137239,
      "auxiliary_loss_mlp": 0.01045785,
      "balance_loss_clip": 1.04906797,
      "balance_loss_mlp": 1.02573371,
      "epoch": 0.11170900345708702,
      "flos": 15632741543040.0,
      "grad_norm": 2.439290762670628,
      "language_loss": 0.82834053,
      "learning_rate": 3.930373863283608e-06,
      "loss": 0.85017073,
      "num_input_tokens_seen": 40213975,
      "step": 1858,
      "time_per_iteration": 2.54972243309021
    },
    {
      "auxiliary_loss_clip": 0.01131711,
      "auxiliary_loss_mlp": 0.01062186,
      "balance_loss_clip": 1.04906988,
      "balance_loss_mlp": 1.03896368,
      "epoch": 0.111769126709755,
      "flos": 23039424766080.0,
      "grad_norm": 2.012051316304491,
      "language_loss": 0.91546327,
      "learning_rate": 3.930271958674866e-06,
      "loss": 0.93740225,
      "num_input_tokens_seen": 40233905,
      "step": 1859,
      "time_per_iteration": 2.5711865425109863
    },
    {
      "auxiliary_loss_clip": 0.01162583,
      "auxiliary_loss_mlp": 0.01048627,
      "balance_loss_clip": 1.05240035,
      "balance_loss_mlp": 1.02724123,
      "epoch": 0.11182924996242297,
      "flos": 20850705434880.0,
      "grad_norm": 2.245113773054152,
      "language_loss": 0.81779045,
      "learning_rate": 3.930169980870018e-06,
      "loss": 0.83990252,
      "num_input_tokens_seen": 40252810,
      "step": 1860,
      "time_per_iteration": 2.5366621017456055
    },
    {
      "auxiliary_loss_clip": 0.01143181,
      "auxiliary_loss_mlp": 0.01056441,
      "balance_loss_clip": 1.04900789,
      "balance_loss_mlp": 1.0349834,
      "epoch": 0.11188937321509093,
      "flos": 17455315587840.0,
      "grad_norm": 3.406932317504339,
      "language_loss": 0.74914628,
      "learning_rate": 3.930067929872931e-06,
      "loss": 0.77114248,
      "num_input_tokens_seen": 40272000,
      "step": 1861,
      "time_per_iteration": 2.5295350551605225
    },
    {
      "auxiliary_loss_clip": 0.01169574,
      "auxiliary_loss_mlp": 0.01045156,
      "balance_loss_clip": 1.05190647,
      "balance_loss_mlp": 1.02593958,
      "epoch": 0.11194949646775891,
      "flos": 24095916518400.0,
      "grad_norm": 1.9079875536027526,
      "language_loss": 0.88699341,
      "learning_rate": 3.929965805687474e-06,
      "loss": 0.90914059,
      "num_input_tokens_seen": 40290660,
      "step": 1862,
      "time_per_iteration": 2.507788896560669
    },
    {
      "auxiliary_loss_clip": 0.01157024,
      "auxiliary_loss_mlp": 0.01056418,
      "balance_loss_clip": 1.05598664,
      "balance_loss_mlp": 1.03522301,
      "epoch": 0.11200961972042688,
      "flos": 25153880728320.0,
      "grad_norm": 2.2867914056401686,
      "language_loss": 0.86597919,
      "learning_rate": 3.92986360831752e-06,
      "loss": 0.88811362,
      "num_input_tokens_seen": 40307820,
      "step": 1863,
      "time_per_iteration": 2.551884889602661
    },
    {
      "auxiliary_loss_clip": 0.01150367,
      "auxiliary_loss_mlp": 0.01048621,
      "balance_loss_clip": 1.04892993,
      "balance_loss_mlp": 1.02543449,
      "epoch": 0.11206974297309484,
      "flos": 21288312829440.0,
      "grad_norm": 1.8327121190940634,
      "language_loss": 0.64318842,
      "learning_rate": 3.929761337766945e-06,
      "loss": 0.6651783,
      "num_input_tokens_seen": 40327430,
      "step": 1864,
      "time_per_iteration": 2.554988145828247
    },
    {
      "auxiliary_loss_clip": 0.01108925,
      "auxiliary_loss_mlp": 0.01048382,
      "balance_loss_clip": 1.05502105,
      "balance_loss_mlp": 1.02853322,
      "epoch": 0.11212986622576282,
      "flos": 18915982151040.0,
      "grad_norm": 2.04272452832174,
      "language_loss": 0.74396062,
      "learning_rate": 3.929658994039627e-06,
      "loss": 0.76553363,
      "num_input_tokens_seen": 40344545,
      "step": 1865,
      "time_per_iteration": 2.604609966278076
    },
    {
      "auxiliary_loss_clip": 0.01108673,
      "auxiliary_loss_mlp": 0.01055251,
      "balance_loss_clip": 1.05095124,
      "balance_loss_mlp": 1.031636,
      "epoch": 0.11218998947843078,
      "flos": 22054754257920.0,
      "grad_norm": 3.0519852168631494,
      "language_loss": 0.8436622,
      "learning_rate": 3.929556577139446e-06,
      "loss": 0.86530149,
      "num_input_tokens_seen": 40362300,
      "step": 1866,
      "time_per_iteration": 2.667299747467041
    },
    {
      "auxiliary_loss_clip": 0.01092703,
      "auxiliary_loss_mlp": 0.00858728,
      "balance_loss_clip": 1.04663968,
      "balance_loss_mlp": 1.08910871,
      "epoch": 0.11225011273109875,
      "flos": 24571697091840.0,
      "grad_norm": 1.5651177131079566,
      "language_loss": 0.81154943,
      "learning_rate": 3.929454087070286e-06,
      "loss": 0.83106375,
      "num_input_tokens_seen": 40384720,
      "step": 1867,
      "time_per_iteration": 2.7144594192504883
    },
    {
      "auxiliary_loss_clip": 0.01173465,
      "auxiliary_loss_mlp": 0.01050659,
      "balance_loss_clip": 1.05332088,
      "balance_loss_mlp": 1.03057218,
      "epoch": 0.11231023598376672,
      "flos": 28438665621120.0,
      "grad_norm": 2.3372424844108455,
      "language_loss": 0.87082124,
      "learning_rate": 3.929351523836035e-06,
      "loss": 0.89306247,
      "num_input_tokens_seen": 40404000,
      "step": 1868,
      "time_per_iteration": 2.554851531982422
    },
    {
      "auxiliary_loss_clip": 0.0115419,
      "auxiliary_loss_mlp": 0.00882469,
      "balance_loss_clip": 1.05532956,
      "balance_loss_mlp": 1.13492608,
      "epoch": 0.1123703592364347,
      "flos": 14426466076800.0,
      "grad_norm": 2.364277927329223,
      "language_loss": 0.67975354,
      "learning_rate": 3.9292488874405795e-06,
      "loss": 0.70012009,
      "num_input_tokens_seen": 40418665,
      "step": 1869,
      "time_per_iteration": 2.5024349689483643
    },
    {
      "auxiliary_loss_clip": 0.0113011,
      "auxiliary_loss_mlp": 0.01051404,
      "balance_loss_clip": 1.05061996,
      "balance_loss_mlp": 1.02979124,
      "epoch": 0.11243048248910266,
      "flos": 22236282616320.0,
      "grad_norm": 1.9816918831467798,
      "language_loss": 0.77294546,
      "learning_rate": 3.929146177887814e-06,
      "loss": 0.79476058,
      "num_input_tokens_seen": 40437870,
      "step": 1870,
      "time_per_iteration": 2.5979137420654297
    },
    {
      "auxiliary_loss_clip": 0.01123907,
      "auxiliary_loss_mlp": 0.0104765,
      "balance_loss_clip": 1.05093932,
      "balance_loss_mlp": 1.02595365,
      "epoch": 0.11249060574177062,
      "flos": 18584167288320.0,
      "grad_norm": 1.9142429927003874,
      "language_loss": 0.7633574,
      "learning_rate": 3.929043395181631e-06,
      "loss": 0.78507304,
      "num_input_tokens_seen": 40455570,
      "step": 1871,
      "time_per_iteration": 2.595543622970581
    },
    {
      "auxiliary_loss_clip": 0.01106128,
      "auxiliary_loss_mlp": 0.01043939,
      "balance_loss_clip": 1.05405915,
      "balance_loss_mlp": 1.02391183,
      "epoch": 0.1125507289944386,
      "flos": 22856567604480.0,
      "grad_norm": 3.6369397529343153,
      "language_loss": 0.82348675,
      "learning_rate": 3.928940539325929e-06,
      "loss": 0.84498739,
      "num_input_tokens_seen": 40473600,
      "step": 1872,
      "time_per_iteration": 2.631897449493408
    },
    {
      "auxiliary_loss_clip": 0.01174336,
      "auxiliary_loss_mlp": 0.01053023,
      "balance_loss_clip": 1.05394673,
      "balance_loss_mlp": 1.03328228,
      "epoch": 0.11261085224710657,
      "flos": 19676390094720.0,
      "grad_norm": 2.459916987864868,
      "language_loss": 0.83033681,
      "learning_rate": 3.9288376103246095e-06,
      "loss": 0.85261041,
      "num_input_tokens_seen": 40490025,
      "step": 1873,
      "time_per_iteration": 2.4860427379608154
    },
    {
      "auxiliary_loss_clip": 0.0113921,
      "auxiliary_loss_mlp": 0.01048948,
      "balance_loss_clip": 1.05363822,
      "balance_loss_mlp": 1.02732348,
      "epoch": 0.11267097549977453,
      "flos": 26063246373120.0,
      "grad_norm": 2.383281029800944,
      "language_loss": 0.92532277,
      "learning_rate": 3.928734608181575e-06,
      "loss": 0.94720435,
      "num_input_tokens_seen": 40511580,
      "step": 1874,
      "time_per_iteration": 2.643134117126465
    },
    {
      "auxiliary_loss_clip": 0.01139009,
      "auxiliary_loss_mlp": 0.0105937,
      "balance_loss_clip": 1.05332923,
      "balance_loss_mlp": 1.03960478,
      "epoch": 0.11273109875244251,
      "flos": 21068036674560.0,
      "grad_norm": 1.4774661587835527,
      "language_loss": 0.75254059,
      "learning_rate": 3.928631532900729e-06,
      "loss": 0.77452439,
      "num_input_tokens_seen": 40530155,
      "step": 1875,
      "time_per_iteration": 2.581851005554199
    },
    {
      "auxiliary_loss_clip": 0.01161745,
      "auxiliary_loss_mlp": 0.01052616,
      "balance_loss_clip": 1.05576694,
      "balance_loss_mlp": 1.03328085,
      "epoch": 0.11279122200511048,
      "flos": 27088999061760.0,
      "grad_norm": 1.950447294627082,
      "language_loss": 0.71855521,
      "learning_rate": 3.928528384485984e-06,
      "loss": 0.74069881,
      "num_input_tokens_seen": 40549500,
      "step": 1876,
      "time_per_iteration": 2.5400876998901367
    },
    {
      "auxiliary_loss_clip": 0.01145523,
      "auxiliary_loss_mlp": 0.01047697,
      "balance_loss_clip": 1.0550108,
      "balance_loss_mlp": 1.02706146,
      "epoch": 0.11285134525777844,
      "flos": 20187901722240.0,
      "grad_norm": 1.911808775601769,
      "language_loss": 0.76934922,
      "learning_rate": 3.9284251629412475e-06,
      "loss": 0.79128146,
      "num_input_tokens_seen": 40567475,
      "step": 1877,
      "time_per_iteration": 2.561160087585449
    },
    {
      "auxiliary_loss_clip": 0.01163647,
      "auxiliary_loss_mlp": 0.01051709,
      "balance_loss_clip": 1.05351222,
      "balance_loss_mlp": 1.03002453,
      "epoch": 0.11291146851044641,
      "flos": 12458453863680.0,
      "grad_norm": 2.2675324319310435,
      "language_loss": 0.87862074,
      "learning_rate": 3.928321868270436e-06,
      "loss": 0.90077436,
      "num_input_tokens_seen": 40583280,
      "step": 1878,
      "time_per_iteration": 2.4787580966949463
    },
    {
      "auxiliary_loss_clip": 0.01138673,
      "auxiliary_loss_mlp": 0.01045297,
      "balance_loss_clip": 1.05512655,
      "balance_loss_mlp": 1.02479339,
      "epoch": 0.11297159176311439,
      "flos": 23842315520640.0,
      "grad_norm": 2.621209764456612,
      "language_loss": 0.81212544,
      "learning_rate": 3.928218500477466e-06,
      "loss": 0.83396518,
      "num_input_tokens_seen": 40603080,
      "step": 1879,
      "time_per_iteration": 2.629356861114502
    },
    {
      "auxiliary_loss_clip": 0.0115032,
      "auxiliary_loss_mlp": 0.01056682,
      "balance_loss_clip": 1.05136013,
      "balance_loss_mlp": 1.03527188,
      "epoch": 0.11303171501578235,
      "flos": 29930538124800.0,
      "grad_norm": 4.077688631626403,
      "language_loss": 0.70445269,
      "learning_rate": 3.928115059566259e-06,
      "loss": 0.72652268,
      "num_input_tokens_seen": 40623255,
      "step": 1880,
      "time_per_iteration": 2.6035544872283936
    },
    {
      "auxiliary_loss_clip": 0.01178225,
      "auxiliary_loss_mlp": 0.01044918,
      "balance_loss_clip": 1.1001091,
      "balance_loss_mlp": 1.02425909,
      "epoch": 0.11309183826845032,
      "flos": 16180558842240.0,
      "grad_norm": 2.116783011241823,
      "language_loss": 0.72416842,
      "learning_rate": 3.928011545540734e-06,
      "loss": 0.74639988,
      "num_input_tokens_seen": 40641570,
      "step": 1881,
      "time_per_iteration": 2.548734664916992
    },
    {
      "auxiliary_loss_clip": 0.01145082,
      "auxiliary_loss_mlp": 0.0085694,
      "balance_loss_clip": 1.0570507,
      "balance_loss_mlp": 1.09093809,
      "epoch": 0.1131519615211183,
      "flos": 12020702814720.0,
      "grad_norm": 2.9043577092074893,
      "language_loss": 0.74206108,
      "learning_rate": 3.927907958404819e-06,
      "loss": 0.76208133,
      "num_input_tokens_seen": 40658775,
      "step": 1882,
      "time_per_iteration": 2.591176748275757
    },
    {
      "auxiliary_loss_clip": 0.01173598,
      "auxiliary_loss_mlp": 0.01051127,
      "balance_loss_clip": 1.05602241,
      "balance_loss_mlp": 1.03014612,
      "epoch": 0.11321208477378626,
      "flos": 26250125857920.0,
      "grad_norm": 2.469989626265307,
      "language_loss": 0.79471856,
      "learning_rate": 3.92780429816244e-06,
      "loss": 0.81696582,
      "num_input_tokens_seen": 40679555,
      "step": 1883,
      "time_per_iteration": 2.5392062664031982
    },
    {
      "auxiliary_loss_clip": 0.01134745,
      "auxiliary_loss_mlp": 0.01052752,
      "balance_loss_clip": 1.05141366,
      "balance_loss_mlp": 1.03119886,
      "epoch": 0.11327220802645423,
      "flos": 13626376583040.0,
      "grad_norm": 2.1043459935630917,
      "language_loss": 0.77605367,
      "learning_rate": 3.927700564817529e-06,
      "loss": 0.79792869,
      "num_input_tokens_seen": 40697295,
      "step": 1884,
      "time_per_iteration": 3.937434196472168
    },
    {
      "auxiliary_loss_clip": 0.01083078,
      "auxiliary_loss_mlp": 0.01004033,
      "balance_loss_clip": 1.04387927,
      "balance_loss_mlp": 1.00026608,
      "epoch": 0.1133323312791222,
      "flos": 57191802814080.0,
      "grad_norm": 0.7919419566409772,
      "language_loss": 0.55193293,
      "learning_rate": 3.927596758374019e-06,
      "loss": 0.57280409,
      "num_input_tokens_seen": 40758095,
      "step": 1885,
      "time_per_iteration": 3.042633295059204
    },
    {
      "auxiliary_loss_clip": 0.01089269,
      "auxiliary_loss_mlp": 0.01044867,
      "balance_loss_clip": 1.05306768,
      "balance_loss_mlp": 1.02509034,
      "epoch": 0.11339245453179017,
      "flos": 24351708245760.0,
      "grad_norm": 2.38087563256885,
      "language_loss": 0.90391439,
      "learning_rate": 3.927492878835848e-06,
      "loss": 0.92525578,
      "num_input_tokens_seen": 40777140,
      "step": 1886,
      "time_per_iteration": 2.8427810668945312
    },
    {
      "auxiliary_loss_clip": 0.01114154,
      "auxiliary_loss_mlp": 0.010516,
      "balance_loss_clip": 1.05278146,
      "balance_loss_mlp": 1.03156137,
      "epoch": 0.11345257778445814,
      "flos": 22670693700480.0,
      "grad_norm": 2.40273735281062,
      "language_loss": 0.84933096,
      "learning_rate": 3.927388926206953e-06,
      "loss": 0.87098849,
      "num_input_tokens_seen": 40797505,
      "step": 1887,
      "time_per_iteration": 4.256206750869751
    },
    {
      "auxiliary_loss_clip": 0.01138291,
      "auxiliary_loss_mlp": 0.01051314,
      "balance_loss_clip": 1.05652392,
      "balance_loss_mlp": 1.03200173,
      "epoch": 0.11351270103712612,
      "flos": 20988242611200.0,
      "grad_norm": 3.468927976040216,
      "language_loss": 0.76596081,
      "learning_rate": 3.927284900491277e-06,
      "loss": 0.78785688,
      "num_input_tokens_seen": 40812970,
      "step": 1888,
      "time_per_iteration": 2.5871238708496094
    },
    {
      "auxiliary_loss_clip": 0.01133271,
      "auxiliary_loss_mlp": 0.01053928,
      "balance_loss_clip": 1.05932438,
      "balance_loss_mlp": 1.03245866,
      "epoch": 0.11357282428979408,
      "flos": 37347923600640.0,
      "grad_norm": 1.9096095660155081,
      "language_loss": 0.68174821,
      "learning_rate": 3.927180801692764e-06,
      "loss": 0.7036202,
      "num_input_tokens_seen": 40837745,
      "step": 1889,
      "time_per_iteration": 2.814681053161621
    },
    {
      "auxiliary_loss_clip": 0.01173829,
      "auxiliary_loss_mlp": 0.01042354,
      "balance_loss_clip": 1.0563519,
      "balance_loss_mlp": 1.02282763,
      "epoch": 0.11363294754246205,
      "flos": 21757018423680.0,
      "grad_norm": 2.0413913939254233,
      "language_loss": 0.84025353,
      "learning_rate": 3.927076629815362e-06,
      "loss": 0.86241543,
      "num_input_tokens_seen": 40856490,
      "step": 1890,
      "time_per_iteration": 3.8938915729522705
    },
    {
      "auxiliary_loss_clip": 0.0115655,
      "auxiliary_loss_mlp": 0.01051713,
      "balance_loss_clip": 1.07090926,
      "balance_loss_mlp": 1.03135228,
      "epoch": 0.11369307079513001,
      "flos": 22601637803520.0,
      "grad_norm": 2.658255470296402,
      "language_loss": 0.6513992,
      "learning_rate": 3.926972384863022e-06,
      "loss": 0.67348182,
      "num_input_tokens_seen": 40874070,
      "step": 1891,
      "time_per_iteration": 2.561370611190796
    },
    {
      "auxiliary_loss_clip": 0.0113965,
      "auxiliary_loss_mlp": 0.01045898,
      "balance_loss_clip": 1.05160582,
      "balance_loss_mlp": 1.02583528,
      "epoch": 0.11375319404779799,
      "flos": 21944257044480.0,
      "grad_norm": 2.0573653359879183,
      "language_loss": 0.8829596,
      "learning_rate": 3.9268680668396956e-06,
      "loss": 0.9048152,
      "num_input_tokens_seen": 40892425,
      "step": 1892,
      "time_per_iteration": 2.5908050537109375
    },
    {
      "auxiliary_loss_clip": 0.01116635,
      "auxiliary_loss_mlp": 0.01066349,
      "balance_loss_clip": 1.05676436,
      "balance_loss_mlp": 1.04349673,
      "epoch": 0.11381331730046595,
      "flos": 26395456285440.0,
      "grad_norm": 2.591574952980665,
      "language_loss": 0.72671586,
      "learning_rate": 3.926763675749339e-06,
      "loss": 0.74854571,
      "num_input_tokens_seen": 40912190,
      "step": 1893,
      "time_per_iteration": 4.0842084884643555
    },
    {
      "auxiliary_loss_clip": 0.01169124,
      "auxiliary_loss_mlp": 0.01055297,
      "balance_loss_clip": 1.05184102,
      "balance_loss_mlp": 1.03377938,
      "epoch": 0.11387344055313392,
      "flos": 23804716959360.0,
      "grad_norm": 2.3567219504341854,
      "language_loss": 0.79576576,
      "learning_rate": 3.92665921159591e-06,
      "loss": 0.81800997,
      "num_input_tokens_seen": 40928395,
      "step": 1894,
      "time_per_iteration": 2.5028445720672607
    },
    {
      "auxiliary_loss_clip": 0.0114787,
      "auxiliary_loss_mlp": 0.01055345,
      "balance_loss_clip": 1.0567795,
      "balance_loss_mlp": 1.034567,
      "epoch": 0.1139335638058019,
      "flos": 34522865902080.0,
      "grad_norm": 4.042974892059261,
      "language_loss": 0.79527926,
      "learning_rate": 3.926554674383371e-06,
      "loss": 0.81731141,
      "num_input_tokens_seen": 40946555,
      "step": 1895,
      "time_per_iteration": 2.6137096881866455
    },
    {
      "auxiliary_loss_clip": 0.01084048,
      "auxiliary_loss_mlp": 0.01016831,
      "balance_loss_clip": 1.03435159,
      "balance_loss_mlp": 1.01318324,
      "epoch": 0.11399368705846986,
      "flos": 70587811520640.0,
      "grad_norm": 0.801062419408952,
      "language_loss": 0.6330744,
      "learning_rate": 3.926450064115686e-06,
      "loss": 0.65408319,
      "num_input_tokens_seen": 41004910,
      "step": 1896,
      "time_per_iteration": 3.180043935775757
    },
    {
      "auxiliary_loss_clip": 0.01142867,
      "auxiliary_loss_mlp": 0.01049838,
      "balance_loss_clip": 1.05571103,
      "balance_loss_mlp": 1.02814233,
      "epoch": 0.11405381031113783,
      "flos": 21324259365120.0,
      "grad_norm": 1.8667886133517009,
      "language_loss": 0.85227072,
      "learning_rate": 3.926345380796821e-06,
      "loss": 0.87419772,
      "num_input_tokens_seen": 41026385,
      "step": 1897,
      "time_per_iteration": 2.56990122795105
    },
    {
      "auxiliary_loss_clip": 0.011744,
      "auxiliary_loss_mlp": 0.00874748,
      "balance_loss_clip": 1.05525458,
      "balance_loss_mlp": 1.12527442,
      "epoch": 0.11411393356380581,
      "flos": 19719627091200.0,
      "grad_norm": 2.3031617954161625,
      "language_loss": 0.7974872,
      "learning_rate": 3.9262406244307465e-06,
      "loss": 0.81797868,
      "num_input_tokens_seen": 41045315,
      "step": 1898,
      "time_per_iteration": 2.5475571155548096
    },
    {
      "auxiliary_loss_clip": 0.01112383,
      "auxiliary_loss_mlp": 0.01054104,
      "balance_loss_clip": 1.04847193,
      "balance_loss_mlp": 1.03119195,
      "epoch": 0.11417405681647377,
      "flos": 17530440883200.0,
      "grad_norm": 2.2355279805599024,
      "language_loss": 0.73234856,
      "learning_rate": 3.926135795021435e-06,
      "loss": 0.75401342,
      "num_input_tokens_seen": 41063390,
      "step": 1899,
      "time_per_iteration": 2.574889659881592
    },
    {
      "auxiliary_loss_clip": 0.0104308,
      "auxiliary_loss_mlp": 0.0100227,
      "balance_loss_clip": 1.03377366,
      "balance_loss_mlp": 0.99824089,
      "epoch": 0.11423418006914174,
      "flos": 59674666619520.0,
      "grad_norm": 0.912580245704262,
      "language_loss": 0.63439256,
      "learning_rate": 3.92603089257286e-06,
      "loss": 0.65484607,
      "num_input_tokens_seen": 41124180,
      "step": 1900,
      "time_per_iteration": 3.156296968460083
    },
    {
      "auxiliary_loss_clip": 0.01105932,
      "auxiliary_loss_mlp": 0.01049422,
      "balance_loss_clip": 1.04873073,
      "balance_loss_mlp": 1.02877545,
      "epoch": 0.1142943033218097,
      "flos": 22963114321920.0,
      "grad_norm": 4.850799468382208,
      "language_loss": 0.78096414,
      "learning_rate": 3.925925917089001e-06,
      "loss": 0.80251771,
      "num_input_tokens_seen": 41143485,
      "step": 1901,
      "time_per_iteration": 2.6657254695892334
    },
    {
      "auxiliary_loss_clip": 0.01164516,
      "auxiliary_loss_mlp": 0.01048674,
      "balance_loss_clip": 1.05489802,
      "balance_loss_mlp": 1.02842045,
      "epoch": 0.11435442657447768,
      "flos": 18256267008000.0,
      "grad_norm": 2.198859597093439,
      "language_loss": 0.8380692,
      "learning_rate": 3.925820868573839e-06,
      "loss": 0.86020112,
      "num_input_tokens_seen": 41161695,
      "step": 1902,
      "time_per_iteration": 2.513610363006592
    },
    {
      "auxiliary_loss_clip": 0.01152342,
      "auxiliary_loss_mlp": 0.01050447,
      "balance_loss_clip": 1.0504775,
      "balance_loss_mlp": 1.02735567,
      "epoch": 0.11441454982714565,
      "flos": 24061191045120.0,
      "grad_norm": 2.729317549808627,
      "language_loss": 0.77497834,
      "learning_rate": 3.925715747031356e-06,
      "loss": 0.79700619,
      "num_input_tokens_seen": 41181715,
      "step": 1903,
      "time_per_iteration": 2.578197479248047
    },
    {
      "auxiliary_loss_clip": 0.01140772,
      "auxiliary_loss_mlp": 0.01037269,
      "balance_loss_clip": 1.05385864,
      "balance_loss_mlp": 1.0190537,
      "epoch": 0.11447467307981361,
      "flos": 25337707557120.0,
      "grad_norm": 2.1266560457897996,
      "language_loss": 0.75657326,
      "learning_rate": 3.925610552465539e-06,
      "loss": 0.77835369,
      "num_input_tokens_seen": 41201770,
      "step": 1904,
      "time_per_iteration": 2.5906918048858643
    },
    {
      "auxiliary_loss_clip": 0.01142985,
      "auxiliary_loss_mlp": 0.0104608,
      "balance_loss_clip": 1.05162358,
      "balance_loss_mlp": 1.0240736,
      "epoch": 0.11453479633248159,
      "flos": 21726063878400.0,
      "grad_norm": 2.371010695521114,
      "language_loss": 0.92593598,
      "learning_rate": 3.9255052848803764e-06,
      "loss": 0.9478265,
      "num_input_tokens_seen": 41220590,
      "step": 1905,
      "time_per_iteration": 2.5492703914642334
    },
    {
      "auxiliary_loss_clip": 0.01159109,
      "auxiliary_loss_mlp": 0.01044333,
      "balance_loss_clip": 1.05255497,
      "balance_loss_mlp": 1.02248168,
      "epoch": 0.11459491958514956,
      "flos": 12969714096000.0,
      "grad_norm": 5.679498144773597,
      "language_loss": 0.77582419,
      "learning_rate": 3.925399944279861e-06,
      "loss": 0.79785854,
      "num_input_tokens_seen": 41237250,
      "step": 1906,
      "time_per_iteration": 2.516207456588745
    },
    {
      "auxiliary_loss_clip": 0.01171906,
      "auxiliary_loss_mlp": 0.01048771,
      "balance_loss_clip": 1.0538615,
      "balance_loss_mlp": 1.02813601,
      "epoch": 0.11465504283781752,
      "flos": 22711273090560.0,
      "grad_norm": 2.7011501943811544,
      "language_loss": 0.82136858,
      "learning_rate": 3.925294530667986e-06,
      "loss": 0.84357536,
      "num_input_tokens_seen": 41256680,
      "step": 1907,
      "time_per_iteration": 2.4873287677764893
    },
    {
      "auxiliary_loss_clip": 0.01129454,
      "auxiliary_loss_mlp": 0.01061923,
      "balance_loss_clip": 1.05093229,
      "balance_loss_mlp": 1.04037035,
      "epoch": 0.1147151660904855,
      "flos": 23398387332480.0,
      "grad_norm": 4.277214450906316,
      "language_loss": 0.84924662,
      "learning_rate": 3.92518904404875e-06,
      "loss": 0.87116033,
      "num_input_tokens_seen": 41270955,
      "step": 1908,
      "time_per_iteration": 2.549497127532959
    },
    {
      "auxiliary_loss_clip": 0.01030513,
      "auxiliary_loss_mlp": 0.01011539,
      "balance_loss_clip": 1.02806044,
      "balance_loss_mlp": 1.0073185,
      "epoch": 0.11477528934315347,
      "flos": 63011843498880.0,
      "grad_norm": 0.9354675897516539,
      "language_loss": 0.60990548,
      "learning_rate": 3.925083484426153e-06,
      "loss": 0.63032597,
      "num_input_tokens_seen": 41319180,
      "step": 1909,
      "time_per_iteration": 2.9537484645843506
    },
    {
      "auxiliary_loss_clip": 0.0117608,
      "auxiliary_loss_mlp": 0.01046285,
      "balance_loss_clip": 1.05737257,
      "balance_loss_mlp": 1.02649641,
      "epoch": 0.11483541259582143,
      "flos": 16325601960960.0,
      "grad_norm": 2.5681538095874097,
      "language_loss": 0.78672588,
      "learning_rate": 3.924977851804197e-06,
      "loss": 0.80894947,
      "num_input_tokens_seen": 41337480,
      "step": 1910,
      "time_per_iteration": 2.4439966678619385
    },
    {
      "auxiliary_loss_clip": 0.01153218,
      "auxiliary_loss_mlp": 0.0104491,
      "balance_loss_clip": 1.05934525,
      "balance_loss_mlp": 1.02503753,
      "epoch": 0.1148955358484894,
      "flos": 21580410228480.0,
      "grad_norm": 2.80498823930451,
      "language_loss": 0.767308,
      "learning_rate": 3.9248721461868875e-06,
      "loss": 0.78928924,
      "num_input_tokens_seen": 41354650,
      "step": 1911,
      "time_per_iteration": 2.5395491123199463
    },
    {
      "auxiliary_loss_clip": 0.01145384,
      "auxiliary_loss_mlp": 0.01047131,
      "balance_loss_clip": 1.04978228,
      "balance_loss_mlp": 1.02663922,
      "epoch": 0.11495565910115738,
      "flos": 27673696650240.0,
      "grad_norm": 2.4006628447791902,
      "language_loss": 0.79143667,
      "learning_rate": 3.9247663675782336e-06,
      "loss": 0.81336188,
      "num_input_tokens_seen": 41376935,
      "step": 1912,
      "time_per_iteration": 2.62797474861145
    },
    {
      "auxiliary_loss_clip": 0.01169692,
      "auxiliary_loss_mlp": 0.00864808,
      "balance_loss_clip": 1.05306554,
      "balance_loss_mlp": 1.11206329,
      "epoch": 0.11501578235382534,
      "flos": 20632368614400.0,
      "grad_norm": 2.247680567380297,
      "language_loss": 0.77998257,
      "learning_rate": 3.924660515982246e-06,
      "loss": 0.8003276,
      "num_input_tokens_seen": 41396105,
      "step": 1913,
      "time_per_iteration": 2.503399610519409
    },
    {
      "auxiliary_loss_clip": 0.01157971,
      "auxiliary_loss_mlp": 0.01052931,
      "balance_loss_clip": 1.05105877,
      "balance_loss_mlp": 1.03178322,
      "epoch": 0.1150759056064933,
      "flos": 19829046896640.0,
      "grad_norm": 2.257356375028276,
      "language_loss": 0.70086765,
      "learning_rate": 3.924554591402939e-06,
      "loss": 0.72297668,
      "num_input_tokens_seen": 41415600,
      "step": 1914,
      "time_per_iteration": 2.498991012573242
    },
    {
      "auxiliary_loss_clip": 0.01028196,
      "auxiliary_loss_mlp": 0.01000571,
      "balance_loss_clip": 1.05073166,
      "balance_loss_mlp": 0.99766272,
      "epoch": 0.11513602885916129,
      "flos": 70045776311040.0,
      "grad_norm": 0.7653795952186894,
      "language_loss": 0.61031091,
      "learning_rate": 3.92444859384433e-06,
      "loss": 0.6305986,
      "num_input_tokens_seen": 41478760,
      "step": 1915,
      "time_per_iteration": 3.4412107467651367
    },
    {
      "auxiliary_loss_clip": 0.01155862,
      "auxiliary_loss_mlp": 0.01047809,
      "balance_loss_clip": 1.05486155,
      "balance_loss_mlp": 1.02667284,
      "epoch": 0.11519615211182925,
      "flos": 15741730385280.0,
      "grad_norm": 2.4105403381936426,
      "language_loss": 0.93077338,
      "learning_rate": 3.924342523310436e-06,
      "loss": 0.95281017,
      "num_input_tokens_seen": 41495720,
      "step": 1916,
      "time_per_iteration": 2.6327672004699707
    },
    {
      "auxiliary_loss_clip": 0.01154159,
      "auxiliary_loss_mlp": 0.01057648,
      "balance_loss_clip": 1.05567956,
      "balance_loss_mlp": 1.03518939,
      "epoch": 0.11525627536449722,
      "flos": 20667632791680.0,
      "grad_norm": 1.9511683175163712,
      "language_loss": 0.72493082,
      "learning_rate": 3.9242363798052806e-06,
      "loss": 0.74704885,
      "num_input_tokens_seen": 41513585,
      "step": 1917,
      "time_per_iteration": 2.552410364151001
    },
    {
      "auxiliary_loss_clip": 0.01131495,
      "auxiliary_loss_mlp": 0.01043259,
      "balance_loss_clip": 1.053792,
      "balance_loss_mlp": 1.02239728,
      "epoch": 0.1153163986171652,
      "flos": 20303283185280.0,
      "grad_norm": 2.931427155215243,
      "language_loss": 0.74340689,
      "learning_rate": 3.92413016333289e-06,
      "loss": 0.76515436,
      "num_input_tokens_seen": 41533390,
      "step": 1918,
      "time_per_iteration": 2.611922025680542
    },
    {
      "auxiliary_loss_clip": 0.01137571,
      "auxiliary_loss_mlp": 0.01043909,
      "balance_loss_clip": 1.05085087,
      "balance_loss_mlp": 1.02397752,
      "epoch": 0.11537652186983316,
      "flos": 17639321984640.0,
      "grad_norm": 3.127649392904145,
      "language_loss": 0.86862481,
      "learning_rate": 3.92402387389729e-06,
      "loss": 0.89043963,
      "num_input_tokens_seen": 41551015,
      "step": 1919,
      "time_per_iteration": 2.5762908458709717
    },
    {
      "auxiliary_loss_clip": 0.01127315,
      "auxiliary_loss_mlp": 0.01058071,
      "balance_loss_clip": 1.04911613,
      "balance_loss_mlp": 1.03427708,
      "epoch": 0.11543664512250112,
      "flos": 21069401391360.0,
      "grad_norm": 1.9583244932234056,
      "language_loss": 0.86496717,
      "learning_rate": 3.923917511502512e-06,
      "loss": 0.88682103,
      "num_input_tokens_seen": 41568055,
      "step": 1920,
      "time_per_iteration": 2.5835633277893066
    },
    {
      "auxiliary_loss_clip": 0.0115644,
      "auxiliary_loss_mlp": 0.01050555,
      "balance_loss_clip": 1.05324733,
      "balance_loss_mlp": 1.03040838,
      "epoch": 0.11549676837516909,
      "flos": 22747542848640.0,
      "grad_norm": 2.058182846956204,
      "language_loss": 0.79124546,
      "learning_rate": 3.923811076152589e-06,
      "loss": 0.81331539,
      "num_input_tokens_seen": 41587435,
      "step": 1921,
      "time_per_iteration": 2.5367937088012695
    },
    {
      "auxiliary_loss_clip": 0.01164213,
      "auxiliary_loss_mlp": 0.01062121,
      "balance_loss_clip": 1.05262995,
      "balance_loss_mlp": 1.04042482,
      "epoch": 0.11555689162783707,
      "flos": 19168972617600.0,
      "grad_norm": 1.9085449911309864,
      "language_loss": 0.78953028,
      "learning_rate": 3.923704567851557e-06,
      "loss": 0.81179357,
      "num_input_tokens_seen": 41604975,
      "step": 1922,
      "time_per_iteration": 3.9209723472595215
    },
    {
      "auxiliary_loss_clip": 0.01091897,
      "auxiliary_loss_mlp": 0.01058493,
      "balance_loss_clip": 1.05144727,
      "balance_loss_mlp": 1.03787029,
      "epoch": 0.11561701488050503,
      "flos": 24572056227840.0,
      "grad_norm": 2.567780799768338,
      "language_loss": 0.84174967,
      "learning_rate": 3.923597986603456e-06,
      "loss": 0.86325359,
      "num_input_tokens_seen": 41626155,
      "step": 1923,
      "time_per_iteration": 2.8341729640960693
    },
    {
      "auxiliary_loss_clip": 0.01163888,
      "auxiliary_loss_mlp": 0.01054135,
      "balance_loss_clip": 1.05603969,
      "balance_loss_mlp": 1.03288031,
      "epoch": 0.115677138133173,
      "flos": 17092546179840.0,
      "grad_norm": 2.5796379672093064,
      "language_loss": 0.81236786,
      "learning_rate": 3.9234913324123264e-06,
      "loss": 0.83454806,
      "num_input_tokens_seen": 41644805,
      "step": 1924,
      "time_per_iteration": 2.8505258560180664
    },
    {
      "auxiliary_loss_clip": 0.01061979,
      "auxiliary_loss_mlp": 0.01047805,
      "balance_loss_clip": 1.03572011,
      "balance_loss_mlp": 1.0438714,
      "epoch": 0.11573726138584098,
      "flos": 62703875266560.0,
      "grad_norm": 0.8470400578636588,
      "language_loss": 0.61230803,
      "learning_rate": 3.923384605282212e-06,
      "loss": 0.63340586,
      "num_input_tokens_seen": 41709345,
      "step": 1925,
      "time_per_iteration": 4.646650791168213
    },
    {
      "auxiliary_loss_clip": 0.01150377,
      "auxiliary_loss_mlp": 0.01074973,
      "balance_loss_clip": 1.05193686,
      "balance_loss_mlp": 1.05210876,
      "epoch": 0.11579738463850894,
      "flos": 22601135013120.0,
      "grad_norm": 1.7840139497305811,
      "language_loss": 0.75308484,
      "learning_rate": 3.923277805217161e-06,
      "loss": 0.77533835,
      "num_input_tokens_seen": 41730210,
      "step": 1926,
      "time_per_iteration": 2.5421175956726074
    },
    {
      "auxiliary_loss_clip": 0.01109141,
      "auxiliary_loss_mlp": 0.0086136,
      "balance_loss_clip": 1.04434645,
      "balance_loss_mlp": 1.1034379,
      "epoch": 0.11585750789117691,
      "flos": 21726135705600.0,
      "grad_norm": 2.572507425613488,
      "language_loss": 0.72459483,
      "learning_rate": 3.923170932221222e-06,
      "loss": 0.74429977,
      "num_input_tokens_seen": 41750270,
      "step": 1927,
      "time_per_iteration": 2.6635801792144775
    },
    {
      "auxiliary_loss_clip": 0.01134854,
      "auxiliary_loss_mlp": 0.01049437,
      "balance_loss_clip": 1.05348957,
      "balance_loss_mlp": 1.0272646,
      "epoch": 0.11591763114384489,
      "flos": 26287544851200.0,
      "grad_norm": 2.4232941566432555,
      "language_loss": 0.86663133,
      "learning_rate": 3.92306398629845e-06,
      "loss": 0.88847423,
      "num_input_tokens_seen": 41772975,
      "step": 1928,
      "time_per_iteration": 4.031736373901367
    },
    {
      "auxiliary_loss_clip": 0.01125197,
      "auxiliary_loss_mlp": 0.01053817,
      "balance_loss_clip": 1.0527184,
      "balance_loss_mlp": 1.03235936,
      "epoch": 0.11597775439651285,
      "flos": 23000461488000.0,
      "grad_norm": 1.7195344669927148,
      "language_loss": 0.77452219,
      "learning_rate": 3.922956967452898e-06,
      "loss": 0.79631233,
      "num_input_tokens_seen": 41791765,
      "step": 1929,
      "time_per_iteration": 2.582120656967163
    },
    {
      "auxiliary_loss_clip": 0.01169503,
      "auxiliary_loss_mlp": 0.01052053,
      "balance_loss_clip": 1.05431819,
      "balance_loss_mlp": 1.03296709,
      "epoch": 0.11603787764918082,
      "flos": 31941715507200.0,
      "grad_norm": 1.7594767426171465,
      "language_loss": 0.77109575,
      "learning_rate": 3.922849875688626e-06,
      "loss": 0.7933113,
      "num_input_tokens_seen": 41815615,
      "step": 1930,
      "time_per_iteration": 2.5957276821136475
    },
    {
      "auxiliary_loss_clip": 0.01138317,
      "auxiliary_loss_mlp": 0.01050441,
      "balance_loss_clip": 1.04940009,
      "balance_loss_mlp": 1.02898276,
      "epoch": 0.1160980009018488,
      "flos": 22271654534400.0,
      "grad_norm": 2.0128243354079305,
      "language_loss": 0.72119689,
      "learning_rate": 3.922742711009693e-06,
      "loss": 0.74308443,
      "num_input_tokens_seen": 41834810,
      "step": 1931,
      "time_per_iteration": 2.5467703342437744
    },
    {
      "auxiliary_loss_clip": 0.01143522,
      "auxiliary_loss_mlp": 0.01059991,
      "balance_loss_clip": 1.05389714,
      "balance_loss_mlp": 1.03664982,
      "epoch": 0.11615812415451676,
      "flos": 22783633038720.0,
      "grad_norm": 1.84162368696264,
      "language_loss": 0.82275778,
      "learning_rate": 3.922635473420164e-06,
      "loss": 0.84479284,
      "num_input_tokens_seen": 41854975,
      "step": 1932,
      "time_per_iteration": 3.967134952545166
    },
    {
      "auxiliary_loss_clip": 0.01054724,
      "auxiliary_loss_mlp": 0.01010379,
      "balance_loss_clip": 1.053774,
      "balance_loss_mlp": 1.00603962,
      "epoch": 0.11621824740718473,
      "flos": 67146096107520.0,
      "grad_norm": 0.7682730043955269,
      "language_loss": 0.61015415,
      "learning_rate": 3.922528162924105e-06,
      "loss": 0.63080513,
      "num_input_tokens_seen": 41911105,
      "step": 1933,
      "time_per_iteration": 3.0733373165130615
    },
    {
      "auxiliary_loss_clip": 0.01114106,
      "auxiliary_loss_mlp": 0.00842354,
      "balance_loss_clip": 1.06359649,
      "balance_loss_mlp": 1.0669601,
      "epoch": 0.11627837065985269,
      "flos": 20375930442240.0,
      "grad_norm": 2.54754193513171,
      "language_loss": 0.8592447,
      "learning_rate": 3.922420779525586e-06,
      "loss": 0.87880927,
      "num_input_tokens_seen": 41931750,
      "step": 1934,
      "time_per_iteration": 2.6656596660614014
    },
    {
      "auxiliary_loss_clip": 0.01115496,
      "auxiliary_loss_mlp": 0.01059382,
      "balance_loss_clip": 1.05196762,
      "balance_loss_mlp": 1.03592157,
      "epoch": 0.11633849391252067,
      "flos": 21725812483200.0,
      "grad_norm": 2.3958511557226387,
      "language_loss": 0.65897548,
      "learning_rate": 3.9223133232286776e-06,
      "loss": 0.68072426,
      "num_input_tokens_seen": 41949400,
      "step": 1935,
      "time_per_iteration": 2.6300792694091797
    },
    {
      "auxiliary_loss_clip": 0.01175022,
      "auxiliary_loss_mlp": 0.01050036,
      "balance_loss_clip": 1.05524111,
      "balance_loss_mlp": 1.03047371,
      "epoch": 0.11639861716518864,
      "flos": 18805341283200.0,
      "grad_norm": 2.504014383210827,
      "language_loss": 0.75834501,
      "learning_rate": 3.922205794037456e-06,
      "loss": 0.78059554,
      "num_input_tokens_seen": 41968100,
      "step": 1936,
      "time_per_iteration": 2.4842000007629395
    },
    {
      "auxiliary_loss_clip": 0.01173349,
      "auxiliary_loss_mlp": 0.01049694,
      "balance_loss_clip": 1.05387664,
      "balance_loss_mlp": 1.02773523,
      "epoch": 0.1164587404178566,
      "flos": 21214983214080.0,
      "grad_norm": 2.328378404355905,
      "language_loss": 0.8405962,
      "learning_rate": 3.922098191955998e-06,
      "loss": 0.86282659,
      "num_input_tokens_seen": 41986375,
      "step": 1937,
      "time_per_iteration": 2.502117395401001
    },
    {
      "auxiliary_loss_clip": 0.01145243,
      "auxiliary_loss_mlp": 0.0104761,
      "balance_loss_clip": 1.05093718,
      "balance_loss_mlp": 1.02647436,
      "epoch": 0.11651886367052458,
      "flos": 27818632028160.0,
      "grad_norm": 4.9746264553029516,
      "language_loss": 0.7628144,
      "learning_rate": 3.921990516988384e-06,
      "loss": 0.78474295,
      "num_input_tokens_seen": 42006055,
      "step": 1938,
      "time_per_iteration": 2.609222650527954
    },
    {
      "auxiliary_loss_clip": 0.01175088,
      "auxiliary_loss_mlp": 0.0105754,
      "balance_loss_clip": 1.05450058,
      "balance_loss_mlp": 1.0365355,
      "epoch": 0.11657898692319255,
      "flos": 22889569224960.0,
      "grad_norm": 1.7591232114281288,
      "language_loss": 0.79424649,
      "learning_rate": 3.921882769138696e-06,
      "loss": 0.81657279,
      "num_input_tokens_seen": 42024995,
      "step": 1939,
      "time_per_iteration": 2.546771287918091
    },
    {
      "auxiliary_loss_clip": 0.01143275,
      "auxiliary_loss_mlp": 0.01053818,
      "balance_loss_clip": 1.05222821,
      "balance_loss_mlp": 1.03195524,
      "epoch": 0.11663911017586051,
      "flos": 24315905364480.0,
      "grad_norm": 2.7337634309664693,
      "language_loss": 0.86572385,
      "learning_rate": 3.9217749484110215e-06,
      "loss": 0.88769484,
      "num_input_tokens_seen": 42042640,
      "step": 1940,
      "time_per_iteration": 2.587029218673706
    },
    {
      "auxiliary_loss_clip": 0.0114771,
      "auxiliary_loss_mlp": 0.01054296,
      "balance_loss_clip": 1.05461681,
      "balance_loss_mlp": 1.03410244,
      "epoch": 0.11669923342852849,
      "flos": 42340152470400.0,
      "grad_norm": 1.4395700694582279,
      "language_loss": 0.75498748,
      "learning_rate": 3.921667054809449e-06,
      "loss": 0.77700752,
      "num_input_tokens_seen": 42067005,
      "step": 1941,
      "time_per_iteration": 2.756819725036621
    },
    {
      "auxiliary_loss_clip": 0.01142764,
      "auxiliary_loss_mlp": 0.00857947,
      "balance_loss_clip": 1.05036712,
      "balance_loss_mlp": 1.09793282,
      "epoch": 0.11675935668119646,
      "flos": 14642288945280.0,
      "grad_norm": 2.238744590874912,
      "language_loss": 0.89193738,
      "learning_rate": 3.921559088338068e-06,
      "loss": 0.91194445,
      "num_input_tokens_seen": 42082295,
      "step": 1942,
      "time_per_iteration": 2.533463716506958
    },
    {
      "auxiliary_loss_clip": 0.01157602,
      "auxiliary_loss_mlp": 0.0104903,
      "balance_loss_clip": 1.05344713,
      "balance_loss_mlp": 1.02959871,
      "epoch": 0.11681947993386442,
      "flos": 35116470063360.0,
      "grad_norm": 1.7985649532044958,
      "language_loss": 0.68047285,
      "learning_rate": 3.921451049000975e-06,
      "loss": 0.70253915,
      "num_input_tokens_seen": 42105295,
      "step": 1943,
      "time_per_iteration": 2.6642913818359375
    },
    {
      "auxiliary_loss_clip": 0.0114482,
      "auxiliary_loss_mlp": 0.01047674,
      "balance_loss_clip": 1.05336821,
      "balance_loss_mlp": 1.02700353,
      "epoch": 0.11687960318653239,
      "flos": 38983259024640.0,
      "grad_norm": 1.8204696420808466,
      "language_loss": 0.6912185,
      "learning_rate": 3.921342936802265e-06,
      "loss": 0.71314341,
      "num_input_tokens_seen": 42125520,
      "step": 1944,
      "time_per_iteration": 2.677407741546631
    },
    {
      "auxiliary_loss_clip": 0.01152352,
      "auxiliary_loss_mlp": 0.01042135,
      "balance_loss_clip": 1.0514226,
      "balance_loss_mlp": 1.02260852,
      "epoch": 0.11693972643920036,
      "flos": 25994980575360.0,
      "grad_norm": 1.587655881321349,
      "language_loss": 0.82824177,
      "learning_rate": 3.921234751746038e-06,
      "loss": 0.85018665,
      "num_input_tokens_seen": 42146335,
      "step": 1945,
      "time_per_iteration": 2.554234743118286
    },
    {
      "auxiliary_loss_clip": 0.01131029,
      "auxiliary_loss_mlp": 0.01048336,
      "balance_loss_clip": 1.04711962,
      "balance_loss_mlp": 1.02786767,
      "epoch": 0.11699984969186833,
      "flos": 27272107618560.0,
      "grad_norm": 2.803313029658502,
      "language_loss": 0.75780559,
      "learning_rate": 3.9211264938363975e-06,
      "loss": 0.77959919,
      "num_input_tokens_seen": 42165320,
      "step": 1946,
      "time_per_iteration": 2.5951461791992188
    },
    {
      "auxiliary_loss_clip": 0.01126981,
      "auxiliary_loss_mlp": 0.01042634,
      "balance_loss_clip": 1.04801786,
      "balance_loss_mlp": 1.02347732,
      "epoch": 0.1170599729445363,
      "flos": 15267853232640.0,
      "grad_norm": 1.9327939025602758,
      "language_loss": 0.69170189,
      "learning_rate": 3.921018163077448e-06,
      "loss": 0.7133981,
      "num_input_tokens_seen": 42182955,
      "step": 1947,
      "time_per_iteration": 2.575732469558716
    },
    {
      "auxiliary_loss_clip": 0.01143114,
      "auxiliary_loss_mlp": 0.01055993,
      "balance_loss_clip": 1.05300927,
      "balance_loss_mlp": 1.03553629,
      "epoch": 0.11712009619720427,
      "flos": 17164439251200.0,
      "grad_norm": 1.8421452006453982,
      "language_loss": 0.85038567,
      "learning_rate": 3.920909759473295e-06,
      "loss": 0.87237668,
      "num_input_tokens_seen": 42200760,
      "step": 1948,
      "time_per_iteration": 2.525824785232544
    },
    {
      "auxiliary_loss_clip": 0.01071888,
      "auxiliary_loss_mlp": 0.01327662,
      "balance_loss_clip": 1.04258156,
      "balance_loss_mlp": 2.0077157,
      "epoch": 0.11718021944987224,
      "flos": 70940991997440.0,
      "grad_norm": 0.8626807356051814,
      "language_loss": 0.65186858,
      "learning_rate": 3.920801283028054e-06,
      "loss": 0.6758641,
      "num_input_tokens_seen": 42265745,
      "step": 1949,
      "time_per_iteration": 3.1817641258239746
    },
    {
      "auxiliary_loss_clip": 0.0114883,
      "auxiliary_loss_mlp": 0.01051341,
      "balance_loss_clip": 1.05234671,
      "balance_loss_mlp": 1.03182614,
      "epoch": 0.1172403427025402,
      "flos": 27453456408960.0,
      "grad_norm": 1.6086578534386813,
      "language_loss": 0.72137278,
      "learning_rate": 3.920692733745835e-06,
      "loss": 0.74337447,
      "num_input_tokens_seen": 42286245,
      "step": 1950,
      "time_per_iteration": 2.5708179473876953
    },
    {
      "auxiliary_loss_clip": 0.01161787,
      "auxiliary_loss_mlp": 0.01055764,
      "balance_loss_clip": 1.05380809,
      "balance_loss_mlp": 1.03573656,
      "epoch": 0.11730046595520818,
      "flos": 15668723992320.0,
      "grad_norm": 2.6186429736162853,
      "language_loss": 0.76524949,
      "learning_rate": 3.920584111630755e-06,
      "loss": 0.78742504,
      "num_input_tokens_seen": 42302710,
      "step": 1951,
      "time_per_iteration": 2.497272253036499
    },
    {
      "auxiliary_loss_clip": 0.01118101,
      "auxiliary_loss_mlp": 0.01061959,
      "balance_loss_clip": 1.05036366,
      "balance_loss_mlp": 1.04194379,
      "epoch": 0.11736058920787615,
      "flos": 25630164092160.0,
      "grad_norm": 1.9848867179370924,
      "language_loss": 0.76375544,
      "learning_rate": 3.9204754166869325e-06,
      "loss": 0.78555608,
      "num_input_tokens_seen": 42324115,
      "step": 1952,
      "time_per_iteration": 2.6761531829833984
    },
    {
      "auxiliary_loss_clip": 0.01117236,
      "auxiliary_loss_mlp": 0.01067686,
      "balance_loss_clip": 1.0459168,
      "balance_loss_mlp": 1.04728913,
      "epoch": 0.11742071246054411,
      "flos": 21434289701760.0,
      "grad_norm": 2.2331394545968597,
      "language_loss": 0.72275239,
      "learning_rate": 3.920366648918491e-06,
      "loss": 0.74460161,
      "num_input_tokens_seen": 42342505,
      "step": 1953,
      "time_per_iteration": 2.58205246925354
    },
    {
      "auxiliary_loss_clip": 0.011388,
      "auxiliary_loss_mlp": 0.00965308,
      "balance_loss_clip": 1.05053067,
      "balance_loss_mlp": 1.29889035,
      "epoch": 0.11748083571321208,
      "flos": 15997845335040.0,
      "grad_norm": 2.7332661403031526,
      "language_loss": 0.79257196,
      "learning_rate": 3.920257808329552e-06,
      "loss": 0.81361306,
      "num_input_tokens_seen": 42360525,
      "step": 1954,
      "time_per_iteration": 2.546494722366333
    },
    {
      "auxiliary_loss_clip": 0.01112452,
      "auxiliary_loss_mlp": 0.01064413,
      "balance_loss_clip": 1.05890346,
      "balance_loss_mlp": 1.04305041,
      "epoch": 0.11754095896588006,
      "flos": 16180056051840.0,
      "grad_norm": 2.267735624052953,
      "language_loss": 0.85795456,
      "learning_rate": 3.920148894924246e-06,
      "loss": 0.87972319,
      "num_input_tokens_seen": 42377045,
      "step": 1955,
      "time_per_iteration": 2.6169769763946533
    },
    {
      "auxiliary_loss_clip": 0.01152582,
      "auxiliary_loss_mlp": 0.00955751,
      "balance_loss_clip": 1.04714417,
      "balance_loss_mlp": 1.28870428,
      "epoch": 0.11760108221854802,
      "flos": 13261596013440.0,
      "grad_norm": 2.53502835899001,
      "language_loss": 0.77986956,
      "learning_rate": 3.920039908706701e-06,
      "loss": 0.80095291,
      "num_input_tokens_seen": 42393960,
      "step": 1956,
      "time_per_iteration": 2.538344383239746
    },
    {
      "auxiliary_loss_clip": 0.01147018,
      "auxiliary_loss_mlp": 0.0106205,
      "balance_loss_clip": 1.05163908,
      "balance_loss_mlp": 1.04208207,
      "epoch": 0.11766120547121599,
      "flos": 24498439303680.0,
      "grad_norm": 2.0007919168406443,
      "language_loss": 0.80456066,
      "learning_rate": 3.91993084968105e-06,
      "loss": 0.82665128,
      "num_input_tokens_seen": 42413160,
      "step": 1957,
      "time_per_iteration": 2.5484495162963867
    },
    {
      "auxiliary_loss_clip": 0.01155837,
      "auxiliary_loss_mlp": 0.01056486,
      "balance_loss_clip": 1.05143189,
      "balance_loss_mlp": 1.03692412,
      "epoch": 0.11772132872388397,
      "flos": 17784005967360.0,
      "grad_norm": 2.253676816155962,
      "language_loss": 0.78356004,
      "learning_rate": 3.919821717851428e-06,
      "loss": 0.8056832,
      "num_input_tokens_seen": 42432590,
      "step": 1958,
      "time_per_iteration": 2.5432913303375244
    },
    {
      "auxiliary_loss_clip": 0.01137817,
      "auxiliary_loss_mlp": 0.01054987,
      "balance_loss_clip": 1.05194342,
      "balance_loss_mlp": 1.03391123,
      "epoch": 0.11778145197655193,
      "flos": 13217030213760.0,
      "grad_norm": 1.8892732562270742,
      "language_loss": 0.76747322,
      "learning_rate": 3.919712513221976e-06,
      "loss": 0.78940117,
      "num_input_tokens_seen": 42450135,
      "step": 1959,
      "time_per_iteration": 2.5276200771331787
    },
    {
      "auxiliary_loss_clip": 0.01144867,
      "auxiliary_loss_mlp": 0.01060032,
      "balance_loss_clip": 1.04876947,
      "balance_loss_mlp": 1.04018402,
      "epoch": 0.1178415752292199,
      "flos": 20230204965120.0,
      "grad_norm": 2.295346807349742,
      "language_loss": 0.70227647,
      "learning_rate": 3.919603235796832e-06,
      "loss": 0.72432542,
      "num_input_tokens_seen": 42470050,
      "step": 1960,
      "time_per_iteration": 2.6025986671447754
    },
    {
      "auxiliary_loss_clip": 0.01146344,
      "auxiliary_loss_mlp": 0.01059941,
      "balance_loss_clip": 1.05103159,
      "balance_loss_mlp": 1.03963923,
      "epoch": 0.11790169848188788,
      "flos": 13040134709760.0,
      "grad_norm": 3.5303481738363094,
      "language_loss": 0.81307012,
      "learning_rate": 3.9194938855801406e-06,
      "loss": 0.83513296,
      "num_input_tokens_seen": 42484335,
      "step": 1961,
      "time_per_iteration": 4.004866361618042
    },
    {
      "auxiliary_loss_clip": 0.01148194,
      "auxiliary_loss_mlp": 0.00887008,
      "balance_loss_clip": 1.04900384,
      "balance_loss_mlp": 1.15622234,
      "epoch": 0.11796182173455584,
      "flos": 22265728790400.0,
      "grad_norm": 1.882794442844709,
      "language_loss": 0.9247185,
      "learning_rate": 3.919384462576049e-06,
      "loss": 0.94507056,
      "num_input_tokens_seen": 42502720,
      "step": 1962,
      "time_per_iteration": 2.529153823852539
    },
    {
      "auxiliary_loss_clip": 0.01130924,
      "auxiliary_loss_mlp": 0.01057578,
      "balance_loss_clip": 1.05040145,
      "balance_loss_mlp": 1.03803921,
      "epoch": 0.1180219449872238,
      "flos": 10635017892480.0,
      "grad_norm": 2.160887962706216,
      "language_loss": 0.87861317,
      "learning_rate": 3.919274966788707e-06,
      "loss": 0.90049815,
      "num_input_tokens_seen": 42519460,
      "step": 1963,
      "time_per_iteration": 2.582219362258911
    },
    {
      "auxiliary_loss_clip": 0.01143996,
      "auxiliary_loss_mlp": 0.00862324,
      "balance_loss_clip": 1.04984224,
      "balance_loss_mlp": 1.11297774,
      "epoch": 0.11808206823989177,
      "flos": 20923532259840.0,
      "grad_norm": 1.8964746199548845,
      "language_loss": 0.83975017,
      "learning_rate": 3.919165398222265e-06,
      "loss": 0.85981333,
      "num_input_tokens_seen": 42539420,
      "step": 1964,
      "time_per_iteration": 3.9737255573272705
    },
    {
      "auxiliary_loss_clip": 0.01115898,
      "auxiliary_loss_mlp": 0.01064274,
      "balance_loss_clip": 1.05253851,
      "balance_loss_mlp": 1.04433072,
      "epoch": 0.11814219149255975,
      "flos": 20777770869120.0,
      "grad_norm": 1.8318993761861306,
      "language_loss": 0.83347541,
      "learning_rate": 3.919055756880879e-06,
      "loss": 0.85527706,
      "num_input_tokens_seen": 42558225,
      "step": 1965,
      "time_per_iteration": 2.6023194789886475
    },
    {
      "auxiliary_loss_clip": 0.01164829,
      "auxiliary_loss_mlp": 0.01053371,
      "balance_loss_clip": 1.04963541,
      "balance_loss_mlp": 1.03345108,
      "epoch": 0.11820231474522772,
      "flos": 48759938542080.0,
      "grad_norm": 1.6328952453432575,
      "language_loss": 0.74504232,
      "learning_rate": 3.918946042768707e-06,
      "loss": 0.76722431,
      "num_input_tokens_seen": 42580790,
      "step": 1966,
      "time_per_iteration": 2.704552412033081
    },
    {
      "auxiliary_loss_clip": 0.0114806,
      "auxiliary_loss_mlp": 0.01054517,
      "balance_loss_clip": 1.05682302,
      "balance_loss_mlp": 1.03465712,
      "epoch": 0.11826243799789568,
      "flos": 16690598012160.0,
      "grad_norm": 5.638078656269434,
      "language_loss": 0.7306813,
      "learning_rate": 3.918836255889908e-06,
      "loss": 0.75270706,
      "num_input_tokens_seen": 42597355,
      "step": 1967,
      "time_per_iteration": 3.896916389465332
    },
    {
      "auxiliary_loss_clip": 0.01150651,
      "auxiliary_loss_mlp": 0.01054334,
      "balance_loss_clip": 1.05035746,
      "balance_loss_mlp": 1.03461695,
      "epoch": 0.11832256125056366,
      "flos": 16909868586240.0,
      "grad_norm": 2.2987294741986934,
      "language_loss": 0.88442409,
      "learning_rate": 3.9187263962486456e-06,
      "loss": 0.90647399,
      "num_input_tokens_seen": 42616060,
      "step": 1968,
      "time_per_iteration": 2.516496181488037
    },
    {
      "auxiliary_loss_clip": 0.01148844,
      "auxiliary_loss_mlp": 0.01049827,
      "balance_loss_clip": 1.05349922,
      "balance_loss_mlp": 1.0296452,
      "epoch": 0.11838268450323162,
      "flos": 22820405587200.0,
      "grad_norm": 1.8660465561385662,
      "language_loss": 0.66790843,
      "learning_rate": 3.918616463849087e-06,
      "loss": 0.68989515,
      "num_input_tokens_seen": 42636285,
      "step": 1969,
      "time_per_iteration": 2.551422595977783
    },
    {
      "auxiliary_loss_clip": 0.01118416,
      "auxiliary_loss_mlp": 0.01053008,
      "balance_loss_clip": 1.04925656,
      "balance_loss_mlp": 1.03170514,
      "epoch": 0.11844280775589959,
      "flos": 33545844990720.0,
      "grad_norm": 2.119612902792752,
      "language_loss": 0.80773586,
      "learning_rate": 3.918506458695399e-06,
      "loss": 0.82945007,
      "num_input_tokens_seen": 42658320,
      "step": 1970,
      "time_per_iteration": 2.6891331672668457
    },
    {
      "auxiliary_loss_clip": 0.01077612,
      "auxiliary_loss_mlp": 0.01005543,
      "balance_loss_clip": 1.03888416,
      "balance_loss_mlp": 1.00264633,
      "epoch": 0.11850293100856757,
      "flos": 66350998604160.0,
      "grad_norm": 0.7981344684973836,
      "language_loss": 0.66166365,
      "learning_rate": 3.918396380791754e-06,
      "loss": 0.68249524,
      "num_input_tokens_seen": 42721500,
      "step": 1971,
      "time_per_iteration": 4.485416889190674
    },
    {
      "auxiliary_loss_clip": 0.0113963,
      "auxiliary_loss_mlp": 0.01048706,
      "balance_loss_clip": 1.05109584,
      "balance_loss_mlp": 1.02910781,
      "epoch": 0.11856305426123553,
      "flos": 24681045070080.0,
      "grad_norm": 1.9768675251710752,
      "language_loss": 0.79809463,
      "learning_rate": 3.918286230142327e-06,
      "loss": 0.819978,
      "num_input_tokens_seen": 42739825,
      "step": 1972,
      "time_per_iteration": 2.5836904048919678
    },
    {
      "auxiliary_loss_clip": 0.01121804,
      "auxiliary_loss_mlp": 0.00929307,
      "balance_loss_clip": 1.05085468,
      "balance_loss_mlp": 1.22933614,
      "epoch": 0.1186231775139035,
      "flos": 24280102483200.0,
      "grad_norm": 2.4421955258219517,
      "language_loss": 0.72831452,
      "learning_rate": 3.918176006751292e-06,
      "loss": 0.74882561,
      "num_input_tokens_seen": 42758695,
      "step": 1973,
      "time_per_iteration": 2.616840362548828
    },
    {
      "auxiliary_loss_clip": 0.01126376,
      "auxiliary_loss_mlp": 0.01045729,
      "balance_loss_clip": 1.05222845,
      "balance_loss_mlp": 1.02566659,
      "epoch": 0.11868330076657148,
      "flos": 21757413473280.0,
      "grad_norm": 2.46406893503361,
      "language_loss": 0.72072142,
      "learning_rate": 3.918065710622832e-06,
      "loss": 0.74244249,
      "num_input_tokens_seen": 42778510,
      "step": 1974,
      "time_per_iteration": 2.6020305156707764
    },
    {
      "auxiliary_loss_clip": 0.01125692,
      "auxiliary_loss_mlp": 0.01038947,
      "balance_loss_clip": 1.06100917,
      "balance_loss_mlp": 1.01928985,
      "epoch": 0.11874342401923944,
      "flos": 17193274894080.0,
      "grad_norm": 2.754204847086295,
      "language_loss": 0.78464663,
      "learning_rate": 3.917955341761128e-06,
      "loss": 0.80629301,
      "num_input_tokens_seen": 42793995,
      "step": 1975,
      "time_per_iteration": 2.599945068359375
    },
    {
      "auxiliary_loss_clip": 0.01116691,
      "auxiliary_loss_mlp": 0.01051801,
      "balance_loss_clip": 1.0528934,
      "balance_loss_mlp": 1.03281081,
      "epoch": 0.11880354727190741,
      "flos": 15229572312960.0,
      "grad_norm": 2.430186128702226,
      "language_loss": 0.75286448,
      "learning_rate": 3.917844900170364e-06,
      "loss": 0.77454937,
      "num_input_tokens_seen": 42809000,
      "step": 1976,
      "time_per_iteration": 2.5963006019592285
    },
    {
      "auxiliary_loss_clip": 0.0115275,
      "auxiliary_loss_mlp": 0.01046629,
      "balance_loss_clip": 1.05091858,
      "balance_loss_mlp": 1.02759171,
      "epoch": 0.11886367052457537,
      "flos": 27309706179840.0,
      "grad_norm": 1.5094173278239351,
      "language_loss": 0.75177848,
      "learning_rate": 3.91773438585473e-06,
      "loss": 0.7737723,
      "num_input_tokens_seen": 42831585,
      "step": 1977,
      "time_per_iteration": 2.591916561126709
    },
    {
      "auxiliary_loss_clip": 0.01165803,
      "auxiliary_loss_mlp": 0.01052229,
      "balance_loss_clip": 1.0515132,
      "balance_loss_mlp": 1.03295279,
      "epoch": 0.11892379377724335,
      "flos": 21798280172160.0,
      "grad_norm": 2.1774149384139125,
      "language_loss": 0.73740005,
      "learning_rate": 3.9176237988184165e-06,
      "loss": 0.75958037,
      "num_input_tokens_seen": 42848420,
      "step": 1978,
      "time_per_iteration": 2.4662814140319824
    },
    {
      "auxiliary_loss_clip": 0.01128392,
      "auxiliary_loss_mlp": 0.0105097,
      "balance_loss_clip": 1.05591011,
      "balance_loss_mlp": 1.03145552,
      "epoch": 0.11898391702991132,
      "flos": 13991013498240.0,
      "grad_norm": 1.7395930735324432,
      "language_loss": 0.73378623,
      "learning_rate": 3.917513139065616e-06,
      "loss": 0.75557983,
      "num_input_tokens_seen": 42866645,
      "step": 1979,
      "time_per_iteration": 2.5810251235961914
    },
    {
      "auxiliary_loss_clip": 0.01122549,
      "auxiliary_loss_mlp": 0.01046205,
      "balance_loss_clip": 1.05192518,
      "balance_loss_mlp": 1.02744186,
      "epoch": 0.11904404028257928,
      "flos": 32234567091840.0,
      "grad_norm": 2.2129578082359354,
      "language_loss": 0.98480397,
      "learning_rate": 3.917402406600525e-06,
      "loss": 1.00649142,
      "num_input_tokens_seen": 42888515,
      "step": 1980,
      "time_per_iteration": 2.6522841453552246
    },
    {
      "auxiliary_loss_clip": 0.01143052,
      "auxiliary_loss_mlp": 0.01051718,
      "balance_loss_clip": 1.05186379,
      "balance_loss_mlp": 1.03128517,
      "epoch": 0.11910416353524726,
      "flos": 23586272398080.0,
      "grad_norm": 2.0845675367680827,
      "language_loss": 0.85961068,
      "learning_rate": 3.917291601427342e-06,
      "loss": 0.8815583,
      "num_input_tokens_seen": 42909035,
      "step": 1981,
      "time_per_iteration": 2.569117784500122
    },
    {
      "auxiliary_loss_clip": 0.01143128,
      "auxiliary_loss_mlp": 0.01059039,
      "balance_loss_clip": 1.0519383,
      "balance_loss_mlp": 1.03826094,
      "epoch": 0.11916428678791523,
      "flos": 25333038789120.0,
      "grad_norm": 1.9283616913701151,
      "language_loss": 0.85265642,
      "learning_rate": 3.91718072355027e-06,
      "loss": 0.87467808,
      "num_input_tokens_seen": 42927555,
      "step": 1982,
      "time_per_iteration": 2.5797674655914307
    },
    {
      "auxiliary_loss_clip": 0.01133814,
      "auxiliary_loss_mlp": 0.01046581,
      "balance_loss_clip": 1.05236936,
      "balance_loss_mlp": 1.02756715,
      "epoch": 0.11922441004058319,
      "flos": 19788431592960.0,
      "grad_norm": 2.437038735480931,
      "language_loss": 0.84974194,
      "learning_rate": 3.917069772973513e-06,
      "loss": 0.87154585,
      "num_input_tokens_seen": 42945300,
      "step": 1983,
      "time_per_iteration": 2.5384044647216797
    },
    {
      "auxiliary_loss_clip": 0.01115105,
      "auxiliary_loss_mlp": 0.01049909,
      "balance_loss_clip": 1.05551076,
      "balance_loss_mlp": 1.03040636,
      "epoch": 0.11928453329325117,
      "flos": 21536347219200.0,
      "grad_norm": 3.5431422392311105,
      "language_loss": 0.77307051,
      "learning_rate": 3.916958749701277e-06,
      "loss": 0.79472065,
      "num_input_tokens_seen": 42961295,
      "step": 1984,
      "time_per_iteration": 2.5942463874816895
    },
    {
      "auxiliary_loss_clip": 0.01147611,
      "auxiliary_loss_mlp": 0.0104995,
      "balance_loss_clip": 1.05121756,
      "balance_loss_mlp": 1.03123462,
      "epoch": 0.11934465654591914,
      "flos": 20815010294400.0,
      "grad_norm": 1.8131537365760881,
      "language_loss": 0.83009195,
      "learning_rate": 3.9168476537377745e-06,
      "loss": 0.85206747,
      "num_input_tokens_seen": 42980330,
      "step": 1985,
      "time_per_iteration": 2.5186705589294434
    },
    {
      "auxiliary_loss_clip": 0.0114665,
      "auxiliary_loss_mlp": 0.01042697,
      "balance_loss_clip": 1.05732918,
      "balance_loss_mlp": 1.02324188,
      "epoch": 0.1194047797985871,
      "flos": 19060486565760.0,
      "grad_norm": 2.162651561195976,
      "language_loss": 0.74525714,
      "learning_rate": 3.916736485087216e-06,
      "loss": 0.76715064,
      "num_input_tokens_seen": 42996125,
      "step": 1986,
      "time_per_iteration": 2.501781940460205
    },
    {
      "auxiliary_loss_clip": 0.01133479,
      "auxiliary_loss_mlp": 0.01040915,
      "balance_loss_clip": 1.04895854,
      "balance_loss_mlp": 1.0223304,
      "epoch": 0.11946490305125507,
      "flos": 27190805184000.0,
      "grad_norm": 1.9351143957200532,
      "language_loss": 0.72062194,
      "learning_rate": 3.916625243753819e-06,
      "loss": 0.74236584,
      "num_input_tokens_seen": 43014180,
      "step": 1987,
      "time_per_iteration": 2.5986461639404297
    },
    {
      "auxiliary_loss_clip": 0.01142069,
      "auxiliary_loss_mlp": 0.01047694,
      "balance_loss_clip": 1.04907751,
      "balance_loss_mlp": 1.02751207,
      "epoch": 0.11952502630392305,
      "flos": 21140791672320.0,
      "grad_norm": 2.6499638490411925,
      "language_loss": 0.71660423,
      "learning_rate": 3.916513929741799e-06,
      "loss": 0.73850179,
      "num_input_tokens_seen": 43032120,
      "step": 1988,
      "time_per_iteration": 2.5340828895568848
    },
    {
      "auxiliary_loss_clip": 0.0115255,
      "auxiliary_loss_mlp": 0.01052999,
      "balance_loss_clip": 1.05189526,
      "balance_loss_mlp": 1.03189933,
      "epoch": 0.11958514955659101,
      "flos": 22124241118080.0,
      "grad_norm": 1.8207622465953963,
      "language_loss": 0.80927891,
      "learning_rate": 3.91640254305538e-06,
      "loss": 0.83133435,
      "num_input_tokens_seen": 43052215,
      "step": 1989,
      "time_per_iteration": 2.548616886138916
    },
    {
      "auxiliary_loss_clip": 0.01123967,
      "auxiliary_loss_mlp": 0.01055039,
      "balance_loss_clip": 1.04762232,
      "balance_loss_mlp": 1.03345013,
      "epoch": 0.11964527280925898,
      "flos": 17421452040960.0,
      "grad_norm": 2.6812235165936897,
      "language_loss": 0.75735188,
      "learning_rate": 3.916291083698784e-06,
      "loss": 0.77914196,
      "num_input_tokens_seen": 43069720,
      "step": 1990,
      "time_per_iteration": 2.5588481426239014
    },
    {
      "auxiliary_loss_clip": 0.01070776,
      "auxiliary_loss_mlp": 0.01011814,
      "balance_loss_clip": 1.04346037,
      "balance_loss_mlp": 1.00857151,
      "epoch": 0.11970539606192696,
      "flos": 70679741402880.0,
      "grad_norm": 0.85324638721916,
      "language_loss": 0.55302262,
      "learning_rate": 3.916179551676238e-06,
      "loss": 0.57384849,
      "num_input_tokens_seen": 43123130,
      "step": 1991,
      "time_per_iteration": 3.1593527793884277
    },
    {
      "auxiliary_loss_clip": 0.01127255,
      "auxiliary_loss_mlp": 0.01049894,
      "balance_loss_clip": 1.05571568,
      "balance_loss_mlp": 1.03102255,
      "epoch": 0.11976551931459492,
      "flos": 21215019127680.0,
      "grad_norm": 3.788502171727554,
      "language_loss": 0.78429937,
      "learning_rate": 3.916067946991971e-06,
      "loss": 0.80607086,
      "num_input_tokens_seen": 43140015,
      "step": 1992,
      "time_per_iteration": 2.6007132530212402
    },
    {
      "auxiliary_loss_clip": 0.01166055,
      "auxiliary_loss_mlp": 0.01049861,
      "balance_loss_clip": 1.05031168,
      "balance_loss_mlp": 1.03012013,
      "epoch": 0.11982564256726289,
      "flos": 25989306226560.0,
      "grad_norm": 2.592766573836646,
      "language_loss": 0.7918514,
      "learning_rate": 3.915956269650216e-06,
      "loss": 0.8140105,
      "num_input_tokens_seen": 43160105,
      "step": 1993,
      "time_per_iteration": 2.5283327102661133
    },
    {
      "auxiliary_loss_clip": 0.01117372,
      "auxiliary_loss_mlp": 0.01047431,
      "balance_loss_clip": 1.04796672,
      "balance_loss_mlp": 1.02809572,
      "epoch": 0.11988576581993086,
      "flos": 21650866755840.0,
      "grad_norm": 3.2616393357395785,
      "language_loss": 0.82154298,
      "learning_rate": 3.915844519655208e-06,
      "loss": 0.84319097,
      "num_input_tokens_seen": 43179835,
      "step": 1994,
      "time_per_iteration": 2.604741096496582
    },
    {
      "auxiliary_loss_clip": 0.0114549,
      "auxiliary_loss_mlp": 0.01045819,
      "balance_loss_clip": 1.05759418,
      "balance_loss_mlp": 1.0280447,
      "epoch": 0.11994588907259883,
      "flos": 17857407409920.0,
      "grad_norm": 2.5427180354856422,
      "language_loss": 0.89008641,
      "learning_rate": 3.915732697011183e-06,
      "loss": 0.91199952,
      "num_input_tokens_seen": 43197210,
      "step": 1995,
      "time_per_iteration": 2.523365020751953
    },
    {
      "auxiliary_loss_clip": 0.01139889,
      "auxiliary_loss_mlp": 0.0104831,
      "balance_loss_clip": 1.0534637,
      "balance_loss_mlp": 1.02873611,
      "epoch": 0.1200060123252668,
      "flos": 24462744163200.0,
      "grad_norm": 1.9398591662631763,
      "language_loss": 0.74066138,
      "learning_rate": 3.9156208017223825e-06,
      "loss": 0.76254338,
      "num_input_tokens_seen": 43215050,
      "step": 1996,
      "time_per_iteration": 2.5914840698242188
    },
    {
      "auxiliary_loss_clip": 0.01126063,
      "auxiliary_loss_mlp": 0.0104736,
      "balance_loss_clip": 1.04856801,
      "balance_loss_mlp": 1.02749944,
      "epoch": 0.12006613557793476,
      "flos": 18732191235840.0,
      "grad_norm": 2.0618777042747203,
      "language_loss": 0.8791343,
      "learning_rate": 3.915508833793048e-06,
      "loss": 0.90086854,
      "num_input_tokens_seen": 43233900,
      "step": 1997,
      "time_per_iteration": 2.5554914474487305
    },
    {
      "auxiliary_loss_clip": 0.01155183,
      "auxiliary_loss_mlp": 0.00886712,
      "balance_loss_clip": 1.05134785,
      "balance_loss_mlp": 1.15757585,
      "epoch": 0.12012625883060274,
      "flos": 22267739952000.0,
      "grad_norm": 2.136047879464062,
      "language_loss": 0.79063749,
      "learning_rate": 3.915396793227428e-06,
      "loss": 0.81105644,
      "num_input_tokens_seen": 43252105,
      "step": 1998,
      "time_per_iteration": 2.533999443054199
    },
    {
      "auxiliary_loss_clip": 0.01152608,
      "auxiliary_loss_mlp": 0.00899942,
      "balance_loss_clip": 1.05016279,
      "balance_loss_mlp": 1.18143344,
      "epoch": 0.1201863820832707,
      "flos": 21758885930880.0,
      "grad_norm": 2.0798024135985878,
      "language_loss": 0.73231286,
      "learning_rate": 3.915284680029769e-06,
      "loss": 0.75283831,
      "num_input_tokens_seen": 43270315,
      "step": 1999,
      "time_per_iteration": 2.528099775314331
    },
    {
      "auxiliary_loss_clip": 0.01165019,
      "auxiliary_loss_mlp": 0.01058324,
      "balance_loss_clip": 1.05215883,
      "balance_loss_mlp": 1.03915489,
      "epoch": 0.12024650533593867,
      "flos": 21907987286400.0,
      "grad_norm": 2.5027874956362854,
      "language_loss": 0.75097722,
      "learning_rate": 3.915172494204323e-06,
      "loss": 0.7732107,
      "num_input_tokens_seen": 43289935,
      "step": 2000,
      "time_per_iteration": 3.8933932781219482
    },
    {
      "auxiliary_loss_clip": 0.01141305,
      "auxiliary_loss_mlp": 0.01053046,
      "balance_loss_clip": 1.052248,
      "balance_loss_mlp": 1.03347158,
      "epoch": 0.12030662858860665,
      "flos": 21689219502720.0,
      "grad_norm": 1.6104929415315563,
      "language_loss": 0.84737599,
      "learning_rate": 3.915060235755344e-06,
      "loss": 0.8693195,
      "num_input_tokens_seen": 43309325,
      "step": 2001,
      "time_per_iteration": 2.555734157562256
    },
    {
      "auxiliary_loss_clip": 0.01139069,
      "auxiliary_loss_mlp": 0.01053195,
      "balance_loss_clip": 1.0482688,
      "balance_loss_mlp": 1.03513527,
      "epoch": 0.12036675184127461,
      "flos": 12933228856320.0,
      "grad_norm": 2.380684526008672,
      "language_loss": 0.73744106,
      "learning_rate": 3.91494790468709e-06,
      "loss": 0.75936371,
      "num_input_tokens_seen": 43327010,
      "step": 2002,
      "time_per_iteration": 3.9716906547546387
    },
    {
      "auxiliary_loss_clip": 0.01122159,
      "auxiliary_loss_mlp": 0.01059501,
      "balance_loss_clip": 1.05417776,
      "balance_loss_mlp": 1.03844833,
      "epoch": 0.12042687509394258,
      "flos": 20851028657280.0,
      "grad_norm": 2.207588136758955,
      "language_loss": 0.77753758,
      "learning_rate": 3.9148355010038185e-06,
      "loss": 0.7993542,
      "num_input_tokens_seen": 43345650,
      "step": 2003,
      "time_per_iteration": 2.5920002460479736
    },
    {
      "auxiliary_loss_clip": 0.01148759,
      "auxiliary_loss_mlp": 0.01058923,
      "balance_loss_clip": 1.04940081,
      "balance_loss_mlp": 1.0393374,
      "epoch": 0.12048699834661056,
      "flos": 23878513451520.0,
      "grad_norm": 1.5901874930960935,
      "language_loss": 0.72107381,
      "learning_rate": 3.914723024709793e-06,
      "loss": 0.74315059,
      "num_input_tokens_seen": 43365555,
      "step": 2004,
      "time_per_iteration": 2.527144432067871
    },
    {
      "auxiliary_loss_clip": 0.01146303,
      "auxiliary_loss_mlp": 0.01062034,
      "balance_loss_clip": 1.05354643,
      "balance_loss_mlp": 1.04150629,
      "epoch": 0.12054712159927852,
      "flos": 19756363726080.0,
      "grad_norm": 2.0999419591616433,
      "language_loss": 0.78534287,
      "learning_rate": 3.914610475809279e-06,
      "loss": 0.80742621,
      "num_input_tokens_seen": 43384990,
      "step": 2005,
      "time_per_iteration": 3.9148948192596436
    },
    {
      "auxiliary_loss_clip": 0.01104684,
      "auxiliary_loss_mlp": 0.01848327,
      "balance_loss_clip": 1.04250169,
      "balance_loss_mlp": 2.9925499,
      "epoch": 0.12060724485194649,
      "flos": 51672763123200.0,
      "grad_norm": 0.9968664815838961,
      "language_loss": 0.58042544,
      "learning_rate": 3.914497854306543e-06,
      "loss": 0.60995555,
      "num_input_tokens_seen": 43436335,
      "step": 2006,
      "time_per_iteration": 2.8950796127319336
    },
    {
      "auxiliary_loss_clip": 0.01145724,
      "auxiliary_loss_mlp": 0.01052454,
      "balance_loss_clip": 1.05312788,
      "balance_loss_mlp": 1.03388143,
      "epoch": 0.12066736810461445,
      "flos": 18990425088000.0,
      "grad_norm": 2.569168696998359,
      "language_loss": 0.76575094,
      "learning_rate": 3.9143851602058575e-06,
      "loss": 0.78773272,
      "num_input_tokens_seen": 43456495,
      "step": 2007,
      "time_per_iteration": 2.550290822982788
    },
    {
      "auxiliary_loss_clip": 0.01135056,
      "auxiliary_loss_mlp": 0.01059835,
      "balance_loss_clip": 1.05314851,
      "balance_loss_mlp": 1.04001057,
      "epoch": 0.12072749135728243,
      "flos": 16471973882880.0,
      "grad_norm": 2.6940490896643396,
      "language_loss": 0.82868421,
      "learning_rate": 3.914272393511494e-06,
      "loss": 0.85063308,
      "num_input_tokens_seen": 43473085,
      "step": 2008,
      "time_per_iteration": 2.554098606109619
    },
    {
      "auxiliary_loss_clip": 0.01164579,
      "auxiliary_loss_mlp": 0.01052384,
      "balance_loss_clip": 1.05145442,
      "balance_loss_mlp": 1.03292906,
      "epoch": 0.1207876146099504,
      "flos": 18077108947200.0,
      "grad_norm": 2.2226199483690934,
      "language_loss": 0.83845901,
      "learning_rate": 3.91415955422773e-06,
      "loss": 0.86062872,
      "num_input_tokens_seen": 43491135,
      "step": 2009,
      "time_per_iteration": 2.4581258296966553
    },
    {
      "auxiliary_loss_clip": 0.01166548,
      "auxiliary_loss_mlp": 0.01053268,
      "balance_loss_clip": 1.05425644,
      "balance_loss_mlp": 1.03217983,
      "epoch": 0.12084773786261836,
      "flos": 21871573873920.0,
      "grad_norm": 1.7177718484415285,
      "language_loss": 0.84070957,
      "learning_rate": 3.914046642358844e-06,
      "loss": 0.86290777,
      "num_input_tokens_seen": 43510440,
      "step": 2010,
      "time_per_iteration": 3.87064528465271
    },
    {
      "auxiliary_loss_clip": 0.01138612,
      "auxiliary_loss_mlp": 0.01029668,
      "balance_loss_clip": 1.05504799,
      "balance_loss_mlp": 1.41387534,
      "epoch": 0.12090786111528634,
      "flos": 18333044328960.0,
      "grad_norm": 1.7532307548796595,
      "language_loss": 0.83881837,
      "learning_rate": 3.9139336579091174e-06,
      "loss": 0.86050117,
      "num_input_tokens_seen": 43530145,
      "step": 2011,
      "time_per_iteration": 2.5405757427215576
    },
    {
      "auxiliary_loss_clip": 0.01137946,
      "auxiliary_loss_mlp": 0.01056826,
      "balance_loss_clip": 1.05030715,
      "balance_loss_mlp": 1.03748977,
      "epoch": 0.1209679843679543,
      "flos": 21105850717440.0,
      "grad_norm": 2.2144079040968156,
      "language_loss": 0.95837605,
      "learning_rate": 3.913820600882834e-06,
      "loss": 0.98032367,
      "num_input_tokens_seen": 43549315,
      "step": 2012,
      "time_per_iteration": 2.6060707569122314
    },
    {
      "auxiliary_loss_clip": 0.01135881,
      "auxiliary_loss_mlp": 0.01046057,
      "balance_loss_clip": 1.05440927,
      "balance_loss_mlp": 1.0259819,
      "epoch": 0.12102810762062227,
      "flos": 29241053585280.0,
      "grad_norm": 2.1417363623006276,
      "language_loss": 0.80387414,
      "learning_rate": 3.913707471284283e-06,
      "loss": 0.82569349,
      "num_input_tokens_seen": 43569240,
      "step": 2013,
      "time_per_iteration": 2.6297709941864014
    },
    {
      "auxiliary_loss_clip": 0.0111938,
      "auxiliary_loss_mlp": 0.0104825,
      "balance_loss_clip": 1.05389166,
      "balance_loss_mlp": 1.02757943,
      "epoch": 0.12108823087329025,
      "flos": 17930701111680.0,
      "grad_norm": 4.844003660159937,
      "language_loss": 0.77253908,
      "learning_rate": 3.9135942691177515e-06,
      "loss": 0.79421532,
      "num_input_tokens_seen": 43587710,
      "step": 2014,
      "time_per_iteration": 2.6008076667785645
    },
    {
      "auxiliary_loss_clip": 0.01155933,
      "auxiliary_loss_mlp": 0.01043335,
      "balance_loss_clip": 1.05310178,
      "balance_loss_mlp": 1.02368963,
      "epoch": 0.12114835412595822,
      "flos": 22091850028800.0,
      "grad_norm": 5.295026242165726,
      "language_loss": 0.87172776,
      "learning_rate": 3.913480994387535e-06,
      "loss": 0.89372045,
      "num_input_tokens_seen": 43606000,
      "step": 2015,
      "time_per_iteration": 2.546238660812378
    },
    {
      "auxiliary_loss_clip": 0.011604,
      "auxiliary_loss_mlp": 0.01044948,
      "balance_loss_clip": 1.05028927,
      "balance_loss_mlp": 1.02610123,
      "epoch": 0.12120847737862618,
      "flos": 20412343854720.0,
      "grad_norm": 3.7349479964449728,
      "language_loss": 0.69643474,
      "learning_rate": 3.913367647097926e-06,
      "loss": 0.71848822,
      "num_input_tokens_seen": 43624815,
      "step": 2016,
      "time_per_iteration": 2.4829280376434326
    },
    {
      "auxiliary_loss_clip": 0.01146932,
      "auxiliary_loss_mlp": 0.01039566,
      "balance_loss_clip": 1.05671573,
      "balance_loss_mlp": 1.01837039,
      "epoch": 0.12126860063129415,
      "flos": 22309037614080.0,
      "grad_norm": 2.6979832366887955,
      "language_loss": 0.8057881,
      "learning_rate": 3.913254227253225e-06,
      "loss": 0.82765305,
      "num_input_tokens_seen": 43643960,
      "step": 2017,
      "time_per_iteration": 2.5871660709381104
    },
    {
      "auxiliary_loss_clip": 0.01149195,
      "auxiliary_loss_mlp": 0.01041434,
      "balance_loss_clip": 1.05301881,
      "balance_loss_mlp": 1.0208106,
      "epoch": 0.12132872388396213,
      "flos": 13699275235200.0,
      "grad_norm": 2.3934647816737575,
      "language_loss": 0.69396234,
      "learning_rate": 3.913140734857731e-06,
      "loss": 0.71586859,
      "num_input_tokens_seen": 43662650,
      "step": 2018,
      "time_per_iteration": 2.500706195831299
    },
    {
      "auxiliary_loss_clip": 0.01135758,
      "auxiliary_loss_mlp": 0.01047637,
      "balance_loss_clip": 1.06123447,
      "balance_loss_mlp": 1.02858734,
      "epoch": 0.12138884713663009,
      "flos": 26466954307200.0,
      "grad_norm": 1.7490491637023857,
      "language_loss": 0.72828591,
      "learning_rate": 3.91302716991575e-06,
      "loss": 0.75011981,
      "num_input_tokens_seen": 43684205,
      "step": 2019,
      "time_per_iteration": 2.651951313018799
    },
    {
      "auxiliary_loss_clip": 0.01105969,
      "auxiliary_loss_mlp": 0.01051963,
      "balance_loss_clip": 1.06014538,
      "balance_loss_mlp": 1.03184021,
      "epoch": 0.12144897038929806,
      "flos": 26141603892480.0,
      "grad_norm": 1.90653374969136,
      "language_loss": 0.91982889,
      "learning_rate": 3.912913532431586e-06,
      "loss": 0.94140828,
      "num_input_tokens_seen": 43706320,
      "step": 2020,
      "time_per_iteration": 2.6866672039031982
    },
    {
      "auxiliary_loss_clip": 0.01139428,
      "auxiliary_loss_mlp": 0.01047248,
      "balance_loss_clip": 1.05560911,
      "balance_loss_mlp": 1.02759027,
      "epoch": 0.12150909364196603,
      "flos": 24717530309760.0,
      "grad_norm": 2.203851537745432,
      "language_loss": 0.77801257,
      "learning_rate": 3.912799822409549e-06,
      "loss": 0.79987931,
      "num_input_tokens_seen": 43724805,
      "step": 2021,
      "time_per_iteration": 2.5806519985198975
    },
    {
      "auxiliary_loss_clip": 0.01165932,
      "auxiliary_loss_mlp": 0.01045934,
      "balance_loss_clip": 1.05442858,
      "balance_loss_mlp": 1.02689672,
      "epoch": 0.121569216894634,
      "flos": 25186990089600.0,
      "grad_norm": 2.7776771253090553,
      "language_loss": 0.80865026,
      "learning_rate": 3.912686039853952e-06,
      "loss": 0.83076894,
      "num_input_tokens_seen": 43742320,
      "step": 2022,
      "time_per_iteration": 2.4969534873962402
    },
    {
      "auxiliary_loss_clip": 0.01142376,
      "auxiliary_loss_mlp": 0.01058877,
      "balance_loss_clip": 1.05785167,
      "balance_loss_mlp": 1.0372045,
      "epoch": 0.12162934014730196,
      "flos": 13444094039040.0,
      "grad_norm": 1.944793233016679,
      "language_loss": 0.8533293,
      "learning_rate": 3.912572184769108e-06,
      "loss": 0.87534183,
      "num_input_tokens_seen": 43760665,
      "step": 2023,
      "time_per_iteration": 2.5711355209350586
    },
    {
      "auxiliary_loss_clip": 0.0113488,
      "auxiliary_loss_mlp": 0.01049064,
      "balance_loss_clip": 1.05145597,
      "balance_loss_mlp": 1.02866697,
      "epoch": 0.12168946339996994,
      "flos": 16946138344320.0,
      "grad_norm": 2.199542277006388,
      "language_loss": 0.85357249,
      "learning_rate": 3.912458257159335e-06,
      "loss": 0.87541187,
      "num_input_tokens_seen": 43779020,
      "step": 2024,
      "time_per_iteration": 2.5757813453674316
    },
    {
      "auxiliary_loss_clip": 0.01164089,
      "auxiliary_loss_mlp": 0.01047669,
      "balance_loss_clip": 1.05038691,
      "balance_loss_mlp": 1.02835703,
      "epoch": 0.12174958665263791,
      "flos": 29821585196160.0,
      "grad_norm": 2.5427062198664836,
      "language_loss": 0.72016573,
      "learning_rate": 3.912344257028954e-06,
      "loss": 0.74228334,
      "num_input_tokens_seen": 43798850,
      "step": 2025,
      "time_per_iteration": 2.5489659309387207
    },
    {
      "auxiliary_loss_clip": 0.0114262,
      "auxiliary_loss_mlp": 0.01044947,
      "balance_loss_clip": 1.05232847,
      "balance_loss_mlp": 1.02551651,
      "epoch": 0.12180970990530587,
      "flos": 24641902224000.0,
      "grad_norm": 1.6221642400268326,
      "language_loss": 0.76204312,
      "learning_rate": 3.912230184382286e-06,
      "loss": 0.78391886,
      "num_input_tokens_seen": 43820130,
      "step": 2026,
      "time_per_iteration": 2.6011173725128174
    },
    {
      "auxiliary_loss_clip": 0.01143951,
      "auxiliary_loss_mlp": 0.01040691,
      "balance_loss_clip": 1.0517664,
      "balance_loss_mlp": 1.02110493,
      "epoch": 0.12186983315797385,
      "flos": 20521691832960.0,
      "grad_norm": 2.5369581138655373,
      "language_loss": 0.89190024,
      "learning_rate": 3.912116039223659e-06,
      "loss": 0.91374665,
      "num_input_tokens_seen": 43838485,
      "step": 2027,
      "time_per_iteration": 2.546464204788208
    },
    {
      "auxiliary_loss_clip": 0.01139269,
      "auxiliary_loss_mlp": 0.01043713,
      "balance_loss_clip": 1.05189896,
      "balance_loss_mlp": 1.02559352,
      "epoch": 0.12192995641064182,
      "flos": 27818344719360.0,
      "grad_norm": 1.9321298905541324,
      "language_loss": 0.75699538,
      "learning_rate": 3.912001821557399e-06,
      "loss": 0.77882516,
      "num_input_tokens_seen": 43859080,
      "step": 2028,
      "time_per_iteration": 2.5826075077056885
    },
    {
      "auxiliary_loss_clip": 0.01122344,
      "auxiliary_loss_mlp": 0.01050271,
      "balance_loss_clip": 1.05485964,
      "balance_loss_mlp": 1.03056598,
      "epoch": 0.12199007966330978,
      "flos": 22017119783040.0,
      "grad_norm": 2.5361943394995876,
      "language_loss": 0.76769638,
      "learning_rate": 3.911887531387839e-06,
      "loss": 0.78942251,
      "num_input_tokens_seen": 43879030,
      "step": 2029,
      "time_per_iteration": 2.6352481842041016
    },
    {
      "auxiliary_loss_clip": 0.01153738,
      "auxiliary_loss_mlp": 0.0104387,
      "balance_loss_clip": 1.0537082,
      "balance_loss_mlp": 1.02462959,
      "epoch": 0.12205020291597775,
      "flos": 23295216493440.0,
      "grad_norm": 2.509319266169487,
      "language_loss": 0.79294682,
      "learning_rate": 3.911773168719313e-06,
      "loss": 0.81492287,
      "num_input_tokens_seen": 43898505,
      "step": 2030,
      "time_per_iteration": 2.530721426010132
    },
    {
      "auxiliary_loss_clip": 0.01164905,
      "auxiliary_loss_mlp": 0.01041792,
      "balance_loss_clip": 1.05478275,
      "balance_loss_mlp": 1.02155006,
      "epoch": 0.12211032616864573,
      "flos": 26031609469440.0,
      "grad_norm": 6.679680701791425,
      "language_loss": 0.74412334,
      "learning_rate": 3.911658733556155e-06,
      "loss": 0.76619029,
      "num_input_tokens_seen": 43917945,
      "step": 2031,
      "time_per_iteration": 2.535252571105957
    },
    {
      "auxiliary_loss_clip": 0.01171382,
      "auxiliary_loss_mlp": 0.01040097,
      "balance_loss_clip": 1.05961561,
      "balance_loss_mlp": 1.02220356,
      "epoch": 0.12217044942131369,
      "flos": 20410943224320.0,
      "grad_norm": 1.8191351815347432,
      "language_loss": 0.75424182,
      "learning_rate": 3.911544225902707e-06,
      "loss": 0.77635658,
      "num_input_tokens_seen": 43937385,
      "step": 2032,
      "time_per_iteration": 2.4712228775024414
    },
    {
      "auxiliary_loss_clip": 0.01152064,
      "auxiliary_loss_mlp": 0.01039382,
      "balance_loss_clip": 1.05475378,
      "balance_loss_mlp": 1.02122688,
      "epoch": 0.12223057267398166,
      "flos": 22857142222080.0,
      "grad_norm": 1.8251130328847562,
      "language_loss": 0.89156616,
      "learning_rate": 3.911429645763311e-06,
      "loss": 0.91348064,
      "num_input_tokens_seen": 43958130,
      "step": 2033,
      "time_per_iteration": 2.6003806591033936
    },
    {
      "auxiliary_loss_clip": 0.01152469,
      "auxiliary_loss_mlp": 0.01043322,
      "balance_loss_clip": 1.05758834,
      "balance_loss_mlp": 1.02474928,
      "epoch": 0.12229069592664964,
      "flos": 20047563285120.0,
      "grad_norm": 2.124365429327312,
      "language_loss": 0.65382642,
      "learning_rate": 3.911314993142311e-06,
      "loss": 0.67578429,
      "num_input_tokens_seen": 43976800,
      "step": 2034,
      "time_per_iteration": 2.5123090744018555
    },
    {
      "auxiliary_loss_clip": 0.01145604,
      "auxiliary_loss_mlp": 0.01046437,
      "balance_loss_clip": 1.05338168,
      "balance_loss_mlp": 1.02614737,
      "epoch": 0.1223508191793176,
      "flos": 22274240313600.0,
      "grad_norm": 1.6292373445009074,
      "language_loss": 0.76602161,
      "learning_rate": 3.911200268044055e-06,
      "loss": 0.78794199,
      "num_input_tokens_seen": 43996620,
      "step": 2035,
      "time_per_iteration": 2.55924391746521
    },
    {
      "auxiliary_loss_clip": 0.01169022,
      "auxiliary_loss_mlp": 0.01042946,
      "balance_loss_clip": 1.05481398,
      "balance_loss_mlp": 1.02310944,
      "epoch": 0.12241094243198557,
      "flos": 21285978445440.0,
      "grad_norm": 1.9166182489866401,
      "language_loss": 0.71420723,
      "learning_rate": 3.911085470472892e-06,
      "loss": 0.73632693,
      "num_input_tokens_seen": 44016175,
      "step": 2036,
      "time_per_iteration": 2.469261646270752
    },
    {
      "auxiliary_loss_clip": 0.0113895,
      "auxiliary_loss_mlp": 0.01049968,
      "balance_loss_clip": 1.05719376,
      "balance_loss_mlp": 1.029881,
      "epoch": 0.12247106568465355,
      "flos": 17382381022080.0,
      "grad_norm": 1.6165526719515169,
      "language_loss": 0.83160383,
      "learning_rate": 3.910970600433178e-06,
      "loss": 0.85349298,
      "num_input_tokens_seen": 44035060,
      "step": 2037,
      "time_per_iteration": 2.556770086288452
    },
    {
      "auxiliary_loss_clip": 0.01148078,
      "auxiliary_loss_mlp": 0.01050501,
      "balance_loss_clip": 1.05506635,
      "balance_loss_mlp": 1.02925777,
      "epoch": 0.12253118893732151,
      "flos": 27045438842880.0,
      "grad_norm": 3.314865382894265,
      "language_loss": 0.80046469,
      "learning_rate": 3.910855657929267e-06,
      "loss": 0.82245046,
      "num_input_tokens_seen": 44053330,
      "step": 2038,
      "time_per_iteration": 2.574760913848877
    },
    {
      "auxiliary_loss_clip": 0.0109981,
      "auxiliary_loss_mlp": 0.02297184,
      "balance_loss_clip": 1.04865754,
      "balance_loss_mlp": 3.76881576,
      "epoch": 0.12259131218998948,
      "flos": 53861518368000.0,
      "grad_norm": 0.8298302125395872,
      "language_loss": 0.58661234,
      "learning_rate": 3.910740642965518e-06,
      "loss": 0.62058234,
      "num_input_tokens_seen": 44107575,
      "step": 2039,
      "time_per_iteration": 4.402273178100586
    },
    {
      "auxiliary_loss_clip": 0.01122483,
      "auxiliary_loss_mlp": 0.01060125,
      "balance_loss_clip": 1.05077159,
      "balance_loss_mlp": 1.03739202,
      "epoch": 0.12265143544265744,
      "flos": 17891917401600.0,
      "grad_norm": 2.028054528905554,
      "language_loss": 0.8052513,
      "learning_rate": 3.910625555546292e-06,
      "loss": 0.82707739,
      "num_input_tokens_seen": 44126075,
      "step": 2040,
      "time_per_iteration": 2.565009355545044
    },
    {
      "auxiliary_loss_clip": 0.01140837,
      "auxiliary_loss_mlp": 0.01049336,
      "balance_loss_clip": 1.05085492,
      "balance_loss_mlp": 1.02972603,
      "epoch": 0.12271155869532542,
      "flos": 21799932197760.0,
      "grad_norm": 2.1998711227052645,
      "language_loss": 0.82817101,
      "learning_rate": 3.910510395675953e-06,
      "loss": 0.8500728,
      "num_input_tokens_seen": 44145605,
      "step": 2041,
      "time_per_iteration": 3.923802137374878
    },
    {
      "auxiliary_loss_clip": 0.01132261,
      "auxiliary_loss_mlp": 0.01049405,
      "balance_loss_clip": 1.05308533,
      "balance_loss_mlp": 1.02789927,
      "epoch": 0.12277168194799339,
      "flos": 19828759587840.0,
      "grad_norm": 1.602312885762978,
      "language_loss": 0.67034459,
      "learning_rate": 3.9103951633588694e-06,
      "loss": 0.69216132,
      "num_input_tokens_seen": 44164770,
      "step": 2042,
      "time_per_iteration": 2.540523052215576
    },
    {
      "auxiliary_loss_clip": 0.01132894,
      "auxiliary_loss_mlp": 0.01046926,
      "balance_loss_clip": 1.0547266,
      "balance_loss_mlp": 1.02716136,
      "epoch": 0.12283180520066135,
      "flos": 23221024951680.0,
      "grad_norm": 3.151065764697575,
      "language_loss": 0.81705689,
      "learning_rate": 3.910279858599409e-06,
      "loss": 0.83885503,
      "num_input_tokens_seen": 44184025,
      "step": 2043,
      "time_per_iteration": 3.9773476123809814
    },
    {
      "auxiliary_loss_clip": 0.01148156,
      "auxiliary_loss_mlp": 0.01045878,
      "balance_loss_clip": 1.05534959,
      "balance_loss_mlp": 1.02523088,
      "epoch": 0.12289192845332933,
      "flos": 18588476920320.0,
      "grad_norm": 2.2910606504174034,
      "language_loss": 0.80318296,
      "learning_rate": 3.910164481401946e-06,
      "loss": 0.82512325,
      "num_input_tokens_seen": 44202950,
      "step": 2044,
      "time_per_iteration": 2.516981363296509
    },
    {
      "auxiliary_loss_clip": 0.01115025,
      "auxiliary_loss_mlp": 0.01048872,
      "balance_loss_clip": 1.05652118,
      "balance_loss_mlp": 1.02928591,
      "epoch": 0.1229520517059973,
      "flos": 25769532862080.0,
      "grad_norm": 1.8823251832660108,
      "language_loss": 0.78177607,
      "learning_rate": 3.910049031770853e-06,
      "loss": 0.80341506,
      "num_input_tokens_seen": 44221115,
      "step": 2045,
      "time_per_iteration": 2.672640085220337
    },
    {
      "auxiliary_loss_clip": 0.01160149,
      "auxiliary_loss_mlp": 0.01056657,
      "balance_loss_clip": 1.05975103,
      "balance_loss_mlp": 1.03611755,
      "epoch": 0.12301217495866526,
      "flos": 20887154760960.0,
      "grad_norm": 2.838984674348378,
      "language_loss": 0.67325813,
      "learning_rate": 3.90993350971051e-06,
      "loss": 0.69542623,
      "num_input_tokens_seen": 44240575,
      "step": 2046,
      "time_per_iteration": 2.52974796295166
    },
    {
      "auxiliary_loss_clip": 0.01171241,
      "auxiliary_loss_mlp": 0.01050857,
      "balance_loss_clip": 1.06041861,
      "balance_loss_mlp": 1.03186655,
      "epoch": 0.12307229821133324,
      "flos": 22378811783040.0,
      "grad_norm": 2.3305262084138203,
      "language_loss": 0.72482026,
      "learning_rate": 3.909817915225297e-06,
      "loss": 0.74704123,
      "num_input_tokens_seen": 44257145,
      "step": 2047,
      "time_per_iteration": 2.5194969177246094
    },
    {
      "auxiliary_loss_clip": 0.01152372,
      "auxiliary_loss_mlp": 0.0105406,
      "balance_loss_clip": 1.05485559,
      "balance_loss_mlp": 1.03276896,
      "epoch": 0.1231324214640012,
      "flos": 23367396873600.0,
      "grad_norm": 2.084616334700215,
      "language_loss": 0.76772887,
      "learning_rate": 3.909702248319597e-06,
      "loss": 0.78979313,
      "num_input_tokens_seen": 44278035,
      "step": 2048,
      "time_per_iteration": 3.958188056945801
    },
    {
      "auxiliary_loss_clip": 0.01141879,
      "auxiliary_loss_mlp": 0.01041745,
      "balance_loss_clip": 1.05402565,
      "balance_loss_mlp": 1.02445328,
      "epoch": 0.12319254471666917,
      "flos": 23767154311680.0,
      "grad_norm": 2.1918369811742435,
      "language_loss": 0.8533715,
      "learning_rate": 3.909586508997797e-06,
      "loss": 0.87520778,
      "num_input_tokens_seen": 44296980,
      "step": 2049,
      "time_per_iteration": 2.571837902069092
    },
    {
      "auxiliary_loss_clip": 0.01118971,
      "auxiliary_loss_mlp": 0.01048179,
      "balance_loss_clip": 1.05910885,
      "balance_loss_mlp": 1.02890253,
      "epoch": 0.12325266796933713,
      "flos": 23550146294400.0,
      "grad_norm": 1.7493574162631031,
      "language_loss": 0.75288624,
      "learning_rate": 3.909470697264285e-06,
      "loss": 0.77455777,
      "num_input_tokens_seen": 44318005,
      "step": 2050,
      "time_per_iteration": 2.675042152404785
    },
    {
      "auxiliary_loss_clip": 0.01132009,
      "auxiliary_loss_mlp": 0.01044045,
      "balance_loss_clip": 1.05339479,
      "balance_loss_mlp": 1.02435136,
      "epoch": 0.12331279122200511,
      "flos": 24423996366720.0,
      "grad_norm": 2.68311152473532,
      "language_loss": 0.80793393,
      "learning_rate": 3.909354813123452e-06,
      "loss": 0.82969439,
      "num_input_tokens_seen": 44335260,
      "step": 2051,
      "time_per_iteration": 2.602168083190918
    },
    {
      "auxiliary_loss_clip": 0.01169126,
      "auxiliary_loss_mlp": 0.010546,
      "balance_loss_clip": 1.0594542,
      "balance_loss_mlp": 1.4473834,
      "epoch": 0.12337291447467308,
      "flos": 25484294960640.0,
      "grad_norm": 2.175772641455349,
      "language_loss": 0.80037475,
      "learning_rate": 3.909238856579693e-06,
      "loss": 0.82261199,
      "num_input_tokens_seen": 44355315,
      "step": 2052,
      "time_per_iteration": 2.545330762863159
    },
    {
      "auxiliary_loss_clip": 0.01157949,
      "auxiliary_loss_mlp": 0.01049516,
      "balance_loss_clip": 1.0543642,
      "balance_loss_mlp": 1.02904773,
      "epoch": 0.12343303772734104,
      "flos": 23550002640000.0,
      "grad_norm": 2.51064261979418,
      "language_loss": 0.73530626,
      "learning_rate": 3.909122827637406e-06,
      "loss": 0.75738096,
      "num_input_tokens_seen": 44373020,
      "step": 2053,
      "time_per_iteration": 2.5315303802490234
    },
    {
      "auxiliary_loss_clip": 0.01168074,
      "auxiliary_loss_mlp": 0.01048336,
      "balance_loss_clip": 1.05406356,
      "balance_loss_mlp": 1.43379021,
      "epoch": 0.12349316098000902,
      "flos": 47557074867840.0,
      "grad_norm": 1.546647524278449,
      "language_loss": 0.7417475,
      "learning_rate": 3.909006726300991e-06,
      "loss": 0.7639116,
      "num_input_tokens_seen": 44397525,
      "step": 2054,
      "time_per_iteration": 2.733426809310913
    },
    {
      "auxiliary_loss_clip": 0.01142511,
      "auxiliary_loss_mlp": 0.01038066,
      "balance_loss_clip": 1.05202544,
      "balance_loss_mlp": 1.02039886,
      "epoch": 0.12355328423267699,
      "flos": 25045969294080.0,
      "grad_norm": 2.2338720709342867,
      "language_loss": 0.84679157,
      "learning_rate": 3.908890552574849e-06,
      "loss": 0.86859727,
      "num_input_tokens_seen": 44415890,
      "step": 2055,
      "time_per_iteration": 2.553285837173462
    },
    {
      "auxiliary_loss_clip": 0.01127021,
      "auxiliary_loss_mlp": 0.01044305,
      "balance_loss_clip": 1.05633116,
      "balance_loss_mlp": 1.02625632,
      "epoch": 0.12361340748534495,
      "flos": 27709140395520.0,
      "grad_norm": 2.199578139863847,
      "language_loss": 0.77617639,
      "learning_rate": 3.908774306463384e-06,
      "loss": 0.79788959,
      "num_input_tokens_seen": 44436625,
      "step": 2056,
      "time_per_iteration": 2.6780505180358887
    },
    {
      "auxiliary_loss_clip": 0.01154444,
      "auxiliary_loss_mlp": 0.01046755,
      "balance_loss_clip": 1.05323362,
      "balance_loss_mlp": 1.02750301,
      "epoch": 0.12367353073801293,
      "flos": 26140598311680.0,
      "grad_norm": 2.1076683935805205,
      "language_loss": 0.82466829,
      "learning_rate": 3.908657987971009e-06,
      "loss": 0.84668034,
      "num_input_tokens_seen": 44455265,
      "step": 2057,
      "time_per_iteration": 2.5323843955993652
    },
    {
      "auxiliary_loss_clip": 0.01142672,
      "auxiliary_loss_mlp": 0.01047011,
      "balance_loss_clip": 1.05445826,
      "balance_loss_mlp": 1.02718616,
      "epoch": 0.1237336539906809,
      "flos": 25156035544320.0,
      "grad_norm": 2.671671913417777,
      "language_loss": 0.78128791,
      "learning_rate": 3.90854159710213e-06,
      "loss": 0.80318475,
      "num_input_tokens_seen": 44475815,
      "step": 2058,
      "time_per_iteration": 2.5779802799224854
    },
    {
      "auxiliary_loss_clip": 0.01136224,
      "auxiliary_loss_mlp": 0.01050054,
      "balance_loss_clip": 1.05464125,
      "balance_loss_mlp": 1.02920473,
      "epoch": 0.12379377724334886,
      "flos": 15304589867520.0,
      "grad_norm": 2.3898816353296444,
      "language_loss": 0.83420825,
      "learning_rate": 3.9084251338611624e-06,
      "loss": 0.85607105,
      "num_input_tokens_seen": 44494045,
      "step": 2059,
      "time_per_iteration": 2.513643264770508
    },
    {
      "auxiliary_loss_clip": 0.01132335,
      "auxiliary_loss_mlp": 0.01050996,
      "balance_loss_clip": 1.05538714,
      "balance_loss_mlp": 1.0299077,
      "epoch": 0.12385390049601683,
      "flos": 21316717509120.0,
      "grad_norm": 2.336095641790558,
      "language_loss": 0.8126924,
      "learning_rate": 3.908308598252523e-06,
      "loss": 0.8345257,
      "num_input_tokens_seen": 44509120,
      "step": 2060,
      "time_per_iteration": 2.5674915313720703
    },
    {
      "auxiliary_loss_clip": 0.01148402,
      "auxiliary_loss_mlp": 0.01046752,
      "balance_loss_clip": 1.05321133,
      "balance_loss_mlp": 1.02646303,
      "epoch": 0.1239140237486848,
      "flos": 15116309752320.0,
      "grad_norm": 2.821716728813867,
      "language_loss": 0.86156619,
      "learning_rate": 3.9081919902806306e-06,
      "loss": 0.88351774,
      "num_input_tokens_seen": 44525780,
      "step": 2061,
      "time_per_iteration": 2.5107247829437256
    },
    {
      "auxiliary_loss_clip": 0.01151831,
      "auxiliary_loss_mlp": 0.01041773,
      "balance_loss_clip": 1.05409372,
      "balance_loss_mlp": 1.02334356,
      "epoch": 0.12397414700135277,
      "flos": 21976791788160.0,
      "grad_norm": 1.9692461060430528,
      "language_loss": 0.85043037,
      "learning_rate": 3.908075309949906e-06,
      "loss": 0.87236643,
      "num_input_tokens_seen": 44543125,
      "step": 2062,
      "time_per_iteration": 2.505032539367676
    },
    {
      "auxiliary_loss_clip": 0.01119588,
      "auxiliary_loss_mlp": 0.01051273,
      "balance_loss_clip": 1.05261898,
      "balance_loss_mlp": 1.03072178,
      "epoch": 0.12403427025402074,
      "flos": 13400892956160.0,
      "grad_norm": 1.8902696648397572,
      "language_loss": 0.78696734,
      "learning_rate": 3.907958557264774e-06,
      "loss": 0.80867594,
      "num_input_tokens_seen": 44560275,
      "step": 2063,
      "time_per_iteration": 2.5467782020568848
    },
    {
      "auxiliary_loss_clip": 0.01119857,
      "auxiliary_loss_mlp": 0.01047845,
      "balance_loss_clip": 1.05567157,
      "balance_loss_mlp": 1.02728176,
      "epoch": 0.12409439350668872,
      "flos": 15304374385920.0,
      "grad_norm": 2.231831592999023,
      "language_loss": 0.79203665,
      "learning_rate": 3.907841732229663e-06,
      "loss": 0.81371373,
      "num_input_tokens_seen": 44577640,
      "step": 2064,
      "time_per_iteration": 2.6105082035064697
    },
    {
      "auxiliary_loss_clip": 0.01141314,
      "auxiliary_loss_mlp": 0.01047336,
      "balance_loss_clip": 1.05447388,
      "balance_loss_mlp": 1.02870321,
      "epoch": 0.12415451675935668,
      "flos": 25009376313600.0,
      "grad_norm": 2.3362225499568727,
      "language_loss": 0.92701232,
      "learning_rate": 3.907724834849002e-06,
      "loss": 0.94889879,
      "num_input_tokens_seen": 44594860,
      "step": 2065,
      "time_per_iteration": 2.561166286468506
    },
    {
      "auxiliary_loss_clip": 0.01145169,
      "auxiliary_loss_mlp": 0.01044162,
      "balance_loss_clip": 1.05176854,
      "balance_loss_mlp": 1.0242182,
      "epoch": 0.12421464001202465,
      "flos": 23659673840640.0,
      "grad_norm": 1.8006732441293682,
      "language_loss": 0.80863726,
      "learning_rate": 3.907607865127225e-06,
      "loss": 0.83053052,
      "num_input_tokens_seen": 44614780,
      "step": 2066,
      "time_per_iteration": 2.5636234283447266
    },
    {
      "auxiliary_loss_clip": 0.01058435,
      "auxiliary_loss_mlp": 0.01012389,
      "balance_loss_clip": 1.04296637,
      "balance_loss_mlp": 1.009027,
      "epoch": 0.12427476326469263,
      "flos": 65732904345600.0,
      "grad_norm": 0.8683677350965583,
      "language_loss": 0.63308573,
      "learning_rate": 3.907490823068766e-06,
      "loss": 0.65379393,
      "num_input_tokens_seen": 44671240,
      "step": 2067,
      "time_per_iteration": 3.1505072116851807
    },
    {
      "auxiliary_loss_clip": 0.01112822,
      "auxiliary_loss_mlp": 0.01055078,
      "balance_loss_clip": 1.04787898,
      "balance_loss_mlp": 1.03382301,
      "epoch": 0.12433488651736059,
      "flos": 24535427333760.0,
      "grad_norm": 2.5337523285753782,
      "language_loss": 0.93499005,
      "learning_rate": 3.907373708678063e-06,
      "loss": 0.95666903,
      "num_input_tokens_seen": 44691050,
      "step": 2068,
      "time_per_iteration": 2.630073308944702
    },
    {
      "auxiliary_loss_clip": 0.01158964,
      "auxiliary_loss_mlp": 0.01045415,
      "balance_loss_clip": 1.05563402,
      "balance_loss_mlp": 1.02755725,
      "epoch": 0.12439500977002856,
      "flos": 21031659175680.0,
      "grad_norm": 1.997201416403602,
      "language_loss": 0.80782008,
      "learning_rate": 3.9072565219595596e-06,
      "loss": 0.82986391,
      "num_input_tokens_seen": 44709850,
      "step": 2069,
      "time_per_iteration": 2.542318344116211
    },
    {
      "auxiliary_loss_clip": 0.01115218,
      "auxiliary_loss_mlp": 0.01051681,
      "balance_loss_clip": 1.05384183,
      "balance_loss_mlp": 1.03059244,
      "epoch": 0.12445513302269653,
      "flos": 26830621555200.0,
      "grad_norm": 1.704206229940813,
      "language_loss": 0.77611697,
      "learning_rate": 3.907139262917696e-06,
      "loss": 0.797786,
      "num_input_tokens_seen": 44731475,
      "step": 2070,
      "time_per_iteration": 2.6355955600738525
    },
    {
      "auxiliary_loss_clip": 0.01152462,
      "auxiliary_loss_mlp": 0.01045458,
      "balance_loss_clip": 1.05397284,
      "balance_loss_mlp": 1.02576458,
      "epoch": 0.1245152562753645,
      "flos": 18368919037440.0,
      "grad_norm": 2.317248202191959,
      "language_loss": 0.80648828,
      "learning_rate": 3.907021931556922e-06,
      "loss": 0.82846749,
      "num_input_tokens_seen": 44749685,
      "step": 2071,
      "time_per_iteration": 2.5311942100524902
    },
    {
      "auxiliary_loss_clip": 0.01154299,
      "auxiliary_loss_mlp": 0.01049257,
      "balance_loss_clip": 1.05731249,
      "balance_loss_mlp": 1.028705,
      "epoch": 0.12457537952803246,
      "flos": 33107986200960.0,
      "grad_norm": 1.8675846256044368,
      "language_loss": 0.78215098,
      "learning_rate": 3.906904527881684e-06,
      "loss": 0.80418658,
      "num_input_tokens_seen": 44772165,
      "step": 2072,
      "time_per_iteration": 2.643401861190796
    },
    {
      "auxiliary_loss_clip": 0.01145701,
      "auxiliary_loss_mlp": 0.01049578,
      "balance_loss_clip": 1.06064439,
      "balance_loss_mlp": 1.03096914,
      "epoch": 0.12463550278070043,
      "flos": 22270217990400.0,
      "grad_norm": 1.8229890460224134,
      "language_loss": 0.75211215,
      "learning_rate": 3.9067870518964355e-06,
      "loss": 0.7740649,
      "num_input_tokens_seen": 44790580,
      "step": 2073,
      "time_per_iteration": 2.563952684402466
    },
    {
      "auxiliary_loss_clip": 0.01095515,
      "auxiliary_loss_mlp": 0.01048243,
      "balance_loss_clip": 1.04759598,
      "balance_loss_mlp": 1.02779913,
      "epoch": 0.12469562603336841,
      "flos": 14679025580160.0,
      "grad_norm": 2.0871559588633,
      "language_loss": 0.90724468,
      "learning_rate": 3.906669503605631e-06,
      "loss": 0.92868221,
      "num_input_tokens_seen": 44806730,
      "step": 2074,
      "time_per_iteration": 2.5923633575439453
    },
    {
      "auxiliary_loss_clip": 0.01102493,
      "auxiliary_loss_mlp": 0.01052313,
      "balance_loss_clip": 1.05211163,
      "balance_loss_mlp": 1.0304023,
      "epoch": 0.12475574928603637,
      "flos": 24644775312000.0,
      "grad_norm": 2.5231633831015308,
      "language_loss": 0.84147346,
      "learning_rate": 3.906551883013728e-06,
      "loss": 0.86302149,
      "num_input_tokens_seen": 44825550,
      "step": 2075,
      "time_per_iteration": 2.6978607177734375
    },
    {
      "auxiliary_loss_clip": 0.0110991,
      "auxiliary_loss_mlp": 0.01055844,
      "balance_loss_clip": 1.05163026,
      "balance_loss_mlp": 1.03436279,
      "epoch": 0.12481587253870434,
      "flos": 21762980081280.0,
      "grad_norm": 2.1216772921760656,
      "language_loss": 0.73464394,
      "learning_rate": 3.9064341901251865e-06,
      "loss": 0.7563014,
      "num_input_tokens_seen": 44844155,
      "step": 2076,
      "time_per_iteration": 4.018374681472778
    },
    {
      "auxiliary_loss_clip": 0.01100131,
      "auxiliary_loss_mlp": 0.01043761,
      "balance_loss_clip": 1.05178523,
      "balance_loss_mlp": 1.02491426,
      "epoch": 0.12487599579137232,
      "flos": 21432529935360.0,
      "grad_norm": 2.0500246983375154,
      "language_loss": 0.75818098,
      "learning_rate": 3.906316424944469e-06,
      "loss": 0.77961993,
      "num_input_tokens_seen": 44863780,
      "step": 2077,
      "time_per_iteration": 2.6102442741394043
    },
    {
      "auxiliary_loss_clip": 0.01151145,
      "auxiliary_loss_mlp": 0.01049329,
      "balance_loss_clip": 1.05195582,
      "balance_loss_mlp": 1.0290637,
      "epoch": 0.12493611904404028,
      "flos": 16107624276480.0,
      "grad_norm": 2.0416758803695014,
      "language_loss": 0.82537973,
      "learning_rate": 3.906198587476043e-06,
      "loss": 0.84738445,
      "num_input_tokens_seen": 44881480,
      "step": 2078,
      "time_per_iteration": 2.4960215091705322
    },
    {
      "auxiliary_loss_clip": 0.01140174,
      "auxiliary_loss_mlp": 0.01052223,
      "balance_loss_clip": 1.05389214,
      "balance_loss_mlp": 1.0323745,
      "epoch": 0.12499624229670825,
      "flos": 21580266574080.0,
      "grad_norm": 1.827059966121622,
      "language_loss": 0.75101686,
      "learning_rate": 3.906080677724374e-06,
      "loss": 0.77294081,
      "num_input_tokens_seen": 44900390,
      "step": 2079,
      "time_per_iteration": 3.9292311668395996
    },
    {
      "auxiliary_loss_clip": 0.01163984,
      "auxiliary_loss_mlp": 0.01053664,
      "balance_loss_clip": 1.0576272,
      "balance_loss_mlp": 1.03357697,
      "epoch": 0.1250563655493762,
      "flos": 25699040421120.0,
      "grad_norm": 2.1393901123622223,
      "language_loss": 0.83601749,
      "learning_rate": 3.905962695693935e-06,
      "loss": 0.85819393,
      "num_input_tokens_seen": 44920375,
      "step": 2080,
      "time_per_iteration": 2.548893451690674
    },
    {
      "auxiliary_loss_clip": 0.01148899,
      "auxiliary_loss_mlp": 0.01053071,
      "balance_loss_clip": 1.05153704,
      "balance_loss_mlp": 1.0335927,
      "epoch": 0.12511648880204418,
      "flos": 16909509450240.0,
      "grad_norm": 2.3155418447759217,
      "language_loss": 0.85098481,
      "learning_rate": 3.9058446413892e-06,
      "loss": 0.8730045,
      "num_input_tokens_seen": 44938415,
      "step": 2081,
      "time_per_iteration": 2.471673011779785
    },
    {
      "auxiliary_loss_clip": 0.01147152,
      "auxiliary_loss_mlp": 0.01042632,
      "balance_loss_clip": 1.04788697,
      "balance_loss_mlp": 1.02361798,
      "epoch": 0.12517661205471217,
      "flos": 17567500740480.0,
      "grad_norm": 8.934326014745363,
      "language_loss": 0.77032465,
      "learning_rate": 3.905726514814646e-06,
      "loss": 0.7922225,
      "num_input_tokens_seen": 44957135,
      "step": 2082,
      "time_per_iteration": 3.9454596042633057
    },
    {
      "auxiliary_loss_clip": 0.01156564,
      "auxiliary_loss_mlp": 0.01052557,
      "balance_loss_clip": 1.0634383,
      "balance_loss_mlp": 1.03146935,
      "epoch": 0.12523673530738014,
      "flos": 16033791870720.0,
      "grad_norm": 6.908561753227237,
      "language_loss": 0.79476643,
      "learning_rate": 3.9056083159747495e-06,
      "loss": 0.81685764,
      "num_input_tokens_seen": 44974480,
      "step": 2083,
      "time_per_iteration": 2.5277655124664307
    },
    {
      "auxiliary_loss_clip": 0.01140811,
      "auxiliary_loss_mlp": 0.01049152,
      "balance_loss_clip": 1.05097342,
      "balance_loss_mlp": 1.02715778,
      "epoch": 0.1252968585600481,
      "flos": 18807747494400.0,
      "grad_norm": 2.379155325094044,
      "language_loss": 0.90332973,
      "learning_rate": 3.9054900448739966e-06,
      "loss": 0.92522931,
      "num_input_tokens_seen": 44990310,
      "step": 2084,
      "time_per_iteration": 2.501002788543701
    },
    {
      "auxiliary_loss_clip": 0.01127345,
      "auxiliary_loss_mlp": 0.0104866,
      "balance_loss_clip": 1.05328441,
      "balance_loss_mlp": 1.02882373,
      "epoch": 0.12535698181271607,
      "flos": 27271568914560.0,
      "grad_norm": 2.2239083366700547,
      "language_loss": 0.80655241,
      "learning_rate": 3.905371701516869e-06,
      "loss": 0.8283124,
      "num_input_tokens_seen": 45010720,
      "step": 2085,
      "time_per_iteration": 2.635072946548462
    },
    {
      "auxiliary_loss_clip": 0.0116349,
      "auxiliary_loss_mlp": 0.01042309,
      "balance_loss_clip": 1.05333424,
      "balance_loss_mlp": 1.02234137,
      "epoch": 0.12541710506538403,
      "flos": 22054107813120.0,
      "grad_norm": 3.7634063220440583,
      "language_loss": 0.88145858,
      "learning_rate": 3.905253285907856e-06,
      "loss": 0.90351653,
      "num_input_tokens_seen": 45030360,
      "step": 2086,
      "time_per_iteration": 3.827904462814331
    },
    {
      "auxiliary_loss_clip": 0.01130891,
      "auxiliary_loss_mlp": 0.01045604,
      "balance_loss_clip": 1.05413651,
      "balance_loss_mlp": 1.02667332,
      "epoch": 0.125477228318052,
      "flos": 12603173760000.0,
      "grad_norm": 2.1954657711089705,
      "language_loss": 0.87132895,
      "learning_rate": 3.905134798051447e-06,
      "loss": 0.89309394,
      "num_input_tokens_seen": 45045085,
      "step": 2087,
      "time_per_iteration": 2.5272021293640137
    },
    {
      "auxiliary_loss_clip": 0.01144125,
      "auxiliary_loss_mlp": 0.01053559,
      "balance_loss_clip": 1.05715442,
      "balance_loss_mlp": 1.03230381,
      "epoch": 0.12553735157071996,
      "flos": 23878549365120.0,
      "grad_norm": 1.9239123910592149,
      "language_loss": 0.73555183,
      "learning_rate": 3.905016237952136e-06,
      "loss": 0.75752866,
      "num_input_tokens_seen": 45065145,
      "step": 2088,
      "time_per_iteration": 2.5579521656036377
    },
    {
      "auxiliary_loss_clip": 0.0107455,
      "auxiliary_loss_mlp": 0.01016813,
      "balance_loss_clip": 1.03255272,
      "balance_loss_mlp": 1.01384473,
      "epoch": 0.12559747482338796,
      "flos": 69920841830400.0,
      "grad_norm": 0.7661677034651387,
      "language_loss": 0.61760056,
      "learning_rate": 3.904897605614418e-06,
      "loss": 0.63851422,
      "num_input_tokens_seen": 45126230,
      "step": 2089,
      "time_per_iteration": 3.0992448329925537
    },
    {
      "auxiliary_loss_clip": 0.01143473,
      "auxiliary_loss_mlp": 0.01054248,
      "balance_loss_clip": 1.05491865,
      "balance_loss_mlp": 1.03385139,
      "epoch": 0.12565759807605592,
      "flos": 24279563779200.0,
      "grad_norm": 2.581448376273402,
      "language_loss": 0.77511358,
      "learning_rate": 3.904778901042793e-06,
      "loss": 0.79709077,
      "num_input_tokens_seen": 45145545,
      "step": 2090,
      "time_per_iteration": 2.572330951690674
    },
    {
      "auxiliary_loss_clip": 0.01063986,
      "auxiliary_loss_mlp": 0.01010217,
      "balance_loss_clip": 1.04171848,
      "balance_loss_mlp": 1.00732005,
      "epoch": 0.12571772132872389,
      "flos": 56451180286080.0,
      "grad_norm": 0.7671400189645787,
      "language_loss": 0.5931133,
      "learning_rate": 3.90466012424176e-06,
      "loss": 0.6138553,
      "num_input_tokens_seen": 45206845,
      "step": 2091,
      "time_per_iteration": 3.042355537414551
    },
    {
      "auxiliary_loss_clip": 0.01151657,
      "auxiliary_loss_mlp": 0.01045887,
      "balance_loss_clip": 1.05338073,
      "balance_loss_mlp": 1.026968,
      "epoch": 0.12577784458139185,
      "flos": 41245846675200.0,
      "grad_norm": 1.9242185867673192,
      "language_loss": 0.63356638,
      "learning_rate": 3.904541275215825e-06,
      "loss": 0.6555419,
      "num_input_tokens_seen": 45228495,
      "step": 2092,
      "time_per_iteration": 2.6864523887634277
    },
    {
      "auxiliary_loss_clip": 0.01146589,
      "auxiliary_loss_mlp": 0.01064023,
      "balance_loss_clip": 1.05683959,
      "balance_loss_mlp": 1.04236293,
      "epoch": 0.12583796783405982,
      "flos": 19755501799680.0,
      "grad_norm": 1.9437579895585686,
      "language_loss": 0.80578882,
      "learning_rate": 3.904422353969493e-06,
      "loss": 0.82789499,
      "num_input_tokens_seen": 45245720,
      "step": 2093,
      "time_per_iteration": 2.5314667224884033
    },
    {
      "auxiliary_loss_clip": 0.01143077,
      "auxiliary_loss_mlp": 0.01056379,
      "balance_loss_clip": 1.05196929,
      "balance_loss_mlp": 1.03622079,
      "epoch": 0.12589809108672778,
      "flos": 22602104680320.0,
      "grad_norm": 2.048947932805511,
      "language_loss": 0.75677121,
      "learning_rate": 3.904303360507276e-06,
      "loss": 0.7787658,
      "num_input_tokens_seen": 45265650,
      "step": 2094,
      "time_per_iteration": 2.5267324447631836
    },
    {
      "auxiliary_loss_clip": 0.01114022,
      "auxiliary_loss_mlp": 0.01053012,
      "balance_loss_clip": 1.05050802,
      "balance_loss_mlp": 1.0335927,
      "epoch": 0.12595821433939577,
      "flos": 45222845541120.0,
      "grad_norm": 1.7120973124373875,
      "language_loss": 0.77101839,
      "learning_rate": 3.9041842948336835e-06,
      "loss": 0.79268873,
      "num_input_tokens_seen": 45287790,
      "step": 2095,
      "time_per_iteration": 2.7804951667785645
    },
    {
      "auxiliary_loss_clip": 0.01142785,
      "auxiliary_loss_mlp": 0.01053303,
      "balance_loss_clip": 1.05342913,
      "balance_loss_mlp": 1.03291798,
      "epoch": 0.12601833759206374,
      "flos": 14319811618560.0,
      "grad_norm": 2.6369227180684423,
      "language_loss": 0.83098644,
      "learning_rate": 3.904065156953232e-06,
      "loss": 0.85294729,
      "num_input_tokens_seen": 45305720,
      "step": 2096,
      "time_per_iteration": 2.546971321105957
    },
    {
      "auxiliary_loss_clip": 0.01156374,
      "auxiliary_loss_mlp": 0.01057355,
      "balance_loss_clip": 1.05493915,
      "balance_loss_mlp": 1.03801894,
      "epoch": 0.1260784608447317,
      "flos": 21288241002240.0,
      "grad_norm": 1.7969668076072398,
      "language_loss": 0.75599492,
      "learning_rate": 3.903945946870439e-06,
      "loss": 0.7781322,
      "num_input_tokens_seen": 45325290,
      "step": 2097,
      "time_per_iteration": 2.5111801624298096
    },
    {
      "auxiliary_loss_clip": 0.01147395,
      "auxiliary_loss_mlp": 0.01060255,
      "balance_loss_clip": 1.05237889,
      "balance_loss_mlp": 1.04209983,
      "epoch": 0.12613858409739967,
      "flos": 26251311006720.0,
      "grad_norm": 2.1113063430883336,
      "language_loss": 0.87022901,
      "learning_rate": 3.9038266645898246e-06,
      "loss": 0.89230549,
      "num_input_tokens_seen": 45344465,
      "step": 2098,
      "time_per_iteration": 2.557436466217041
    },
    {
      "auxiliary_loss_clip": 0.01117843,
      "auxiliary_loss_mlp": 0.01059941,
      "balance_loss_clip": 1.05579495,
      "balance_loss_mlp": 1.037637,
      "epoch": 0.12619870735006763,
      "flos": 21579979265280.0,
      "grad_norm": 6.748844620920046,
      "language_loss": 0.69801199,
      "learning_rate": 3.903707310115912e-06,
      "loss": 0.7197898,
      "num_input_tokens_seen": 45362465,
      "step": 2099,
      "time_per_iteration": 2.606476068496704
    },
    {
      "auxiliary_loss_clip": 0.01140503,
      "auxiliary_loss_mlp": 0.01064493,
      "balance_loss_clip": 1.04974389,
      "balance_loss_mlp": 1.04280901,
      "epoch": 0.1262588306027356,
      "flos": 23367037737600.0,
      "grad_norm": 2.0597270704323707,
      "language_loss": 0.82081306,
      "learning_rate": 3.903587883453228e-06,
      "loss": 0.84286302,
      "num_input_tokens_seen": 45382700,
      "step": 2100,
      "time_per_iteration": 2.576624631881714
    },
    {
      "auxiliary_loss_clip": 0.01147738,
      "auxiliary_loss_mlp": 0.01055058,
      "balance_loss_clip": 1.05217898,
      "balance_loss_mlp": 1.03435111,
      "epoch": 0.12631895385540357,
      "flos": 23949185460480.0,
      "grad_norm": 2.1689774318871433,
      "language_loss": 0.8025766,
      "learning_rate": 3.903468384606302e-06,
      "loss": 0.82460451,
      "num_input_tokens_seen": 45401005,
      "step": 2101,
      "time_per_iteration": 2.559868335723877
    },
    {
      "auxiliary_loss_clip": 0.01088573,
      "auxiliary_loss_mlp": 0.01064528,
      "balance_loss_clip": 1.03378069,
      "balance_loss_mlp": 1.06076145,
      "epoch": 0.12637907710807156,
      "flos": 70282138780800.0,
      "grad_norm": 0.7147294253138414,
      "language_loss": 0.57111919,
      "learning_rate": 3.903348813579662e-06,
      "loss": 0.59265018,
      "num_input_tokens_seen": 45466555,
      "step": 2102,
      "time_per_iteration": 3.131472587585449
    },
    {
      "auxiliary_loss_clip": 0.01135201,
      "auxiliary_loss_mlp": 0.01051772,
      "balance_loss_clip": 1.0519712,
      "balance_loss_mlp": 1.03170872,
      "epoch": 0.12643920036073952,
      "flos": 18915084311040.0,
      "grad_norm": 3.129208852872319,
      "language_loss": 0.93597353,
      "learning_rate": 3.903229170377845e-06,
      "loss": 0.9578433,
      "num_input_tokens_seen": 45485165,
      "step": 2103,
      "time_per_iteration": 2.564546585083008
    },
    {
      "auxiliary_loss_clip": 0.01141654,
      "auxiliary_loss_mlp": 0.01037378,
      "balance_loss_clip": 1.05018306,
      "balance_loss_mlp": 1.0186857,
      "epoch": 0.1264993236134075,
      "flos": 27782470010880.0,
      "grad_norm": 2.071818606101232,
      "language_loss": 0.78215814,
      "learning_rate": 3.903109455005387e-06,
      "loss": 0.80394852,
      "num_input_tokens_seen": 45504630,
      "step": 2104,
      "time_per_iteration": 2.5594379901885986
    },
    {
      "auxiliary_loss_clip": 0.01130369,
      "auxiliary_loss_mlp": 0.01049038,
      "balance_loss_clip": 1.05392241,
      "balance_loss_mlp": 1.03062057,
      "epoch": 0.12655944686607545,
      "flos": 24754697907840.0,
      "grad_norm": 1.708487066566394,
      "language_loss": 0.8126899,
      "learning_rate": 3.902989667466828e-06,
      "loss": 0.83448398,
      "num_input_tokens_seen": 45524885,
      "step": 2105,
      "time_per_iteration": 2.617420196533203
    },
    {
      "auxiliary_loss_clip": 0.01163684,
      "auxiliary_loss_mlp": 0.01050858,
      "balance_loss_clip": 1.05561745,
      "balance_loss_mlp": 1.02982986,
      "epoch": 0.12661957011874342,
      "flos": 24133048202880.0,
      "grad_norm": 3.349103037679372,
      "language_loss": 0.83813316,
      "learning_rate": 3.90286980776671e-06,
      "loss": 0.86027855,
      "num_input_tokens_seen": 45545000,
      "step": 2106,
      "time_per_iteration": 2.5255205631256104
    },
    {
      "auxiliary_loss_clip": 0.01122078,
      "auxiliary_loss_mlp": 0.01043973,
      "balance_loss_clip": 1.05415726,
      "balance_loss_mlp": 1.02427924,
      "epoch": 0.12667969337141138,
      "flos": 24569614103040.0,
      "grad_norm": 1.820627967007029,
      "language_loss": 0.73757148,
      "learning_rate": 3.902749875909578e-06,
      "loss": 0.75923198,
      "num_input_tokens_seen": 45564210,
      "step": 2107,
      "time_per_iteration": 2.6430203914642334
    },
    {
      "auxiliary_loss_clip": 0.01165685,
      "auxiliary_loss_mlp": 0.01050436,
      "balance_loss_clip": 1.05534863,
      "balance_loss_mlp": 1.03132713,
      "epoch": 0.12673981662407935,
      "flos": 22961677777920.0,
      "grad_norm": 1.9955672243692584,
      "language_loss": 0.79420471,
      "learning_rate": 3.90262987189998e-06,
      "loss": 0.81636596,
      "num_input_tokens_seen": 45583030,
      "step": 2108,
      "time_per_iteration": 2.507863759994507
    },
    {
      "auxiliary_loss_clip": 0.01165743,
      "auxiliary_loss_mlp": 0.01043587,
      "balance_loss_clip": 1.05282593,
      "balance_loss_mlp": 1.02350008,
      "epoch": 0.12679993987674734,
      "flos": 17274864637440.0,
      "grad_norm": 1.9866891798789112,
      "language_loss": 0.7635386,
      "learning_rate": 3.902509795742467e-06,
      "loss": 0.7856319,
      "num_input_tokens_seen": 45602265,
      "step": 2109,
      "time_per_iteration": 2.475743055343628
    },
    {
      "auxiliary_loss_clip": 0.01111647,
      "auxiliary_loss_mlp": 0.01058203,
      "balance_loss_clip": 1.0493989,
      "balance_loss_mlp": 1.03779411,
      "epoch": 0.1268600631294153,
      "flos": 17275080119040.0,
      "grad_norm": 2.221441469484212,
      "language_loss": 0.8272413,
      "learning_rate": 3.902389647441592e-06,
      "loss": 0.84893978,
      "num_input_tokens_seen": 45620595,
      "step": 2110,
      "time_per_iteration": 2.555478811264038
    },
    {
      "auxiliary_loss_clip": 0.01143084,
      "auxiliary_loss_mlp": 0.00979418,
      "balance_loss_clip": 1.05402446,
      "balance_loss_mlp": 1.29373991,
      "epoch": 0.12692018638208327,
      "flos": 24061047390720.0,
      "grad_norm": 2.450585125732985,
      "language_loss": 0.78596592,
      "learning_rate": 3.90226942700191e-06,
      "loss": 0.80719095,
      "num_input_tokens_seen": 45641140,
      "step": 2111,
      "time_per_iteration": 2.5817017555236816
    },
    {
      "auxiliary_loss_clip": 0.01128189,
      "auxiliary_loss_mlp": 0.01066807,
      "balance_loss_clip": 1.05406606,
      "balance_loss_mlp": 1.04406166,
      "epoch": 0.12698030963475124,
      "flos": 31831900652160.0,
      "grad_norm": 2.462588267953572,
      "language_loss": 0.77281225,
      "learning_rate": 3.902149134427982e-06,
      "loss": 0.79476213,
      "num_input_tokens_seen": 45662315,
      "step": 2112,
      "time_per_iteration": 2.681741237640381
    },
    {
      "auxiliary_loss_clip": 0.01124859,
      "auxiliary_loss_mlp": 0.01051951,
      "balance_loss_clip": 1.05219257,
      "balance_loss_mlp": 1.0308392,
      "epoch": 0.1270404328874192,
      "flos": 25187744275200.0,
      "grad_norm": 2.4462191951567522,
      "language_loss": 0.85403985,
      "learning_rate": 3.902028769724367e-06,
      "loss": 0.87580794,
      "num_input_tokens_seen": 45680335,
      "step": 2113,
      "time_per_iteration": 2.5981204509735107
    },
    {
      "auxiliary_loss_clip": 0.01128877,
      "auxiliary_loss_mlp": 0.01048358,
      "balance_loss_clip": 1.04941344,
      "balance_loss_mlp": 1.02766299,
      "epoch": 0.12710055614008717,
      "flos": 15997342544640.0,
      "grad_norm": 2.369586098920705,
      "language_loss": 0.7448976,
      "learning_rate": 3.9019083328956315e-06,
      "loss": 0.76666999,
      "num_input_tokens_seen": 45696240,
      "step": 2114,
      "time_per_iteration": 2.5360043048858643
    },
    {
      "auxiliary_loss_clip": 0.01152893,
      "auxiliary_loss_mlp": 0.01055848,
      "balance_loss_clip": 1.05621231,
      "balance_loss_mlp": 1.03475988,
      "epoch": 0.12716067939275516,
      "flos": 15085642515840.0,
      "grad_norm": 2.821696915438223,
      "language_loss": 0.83184218,
      "learning_rate": 3.901787823946341e-06,
      "loss": 0.85392964,
      "num_input_tokens_seen": 45713695,
      "step": 2115,
      "time_per_iteration": 3.8670804500579834
    },
    {
      "auxiliary_loss_clip": 0.01146418,
      "auxiliary_loss_mlp": 0.01049393,
      "balance_loss_clip": 1.0546757,
      "balance_loss_mlp": 1.02921057,
      "epoch": 0.12722080264542313,
      "flos": 28366736636160.0,
      "grad_norm": 2.0976600125182436,
      "language_loss": 0.86779702,
      "learning_rate": 3.901667242881065e-06,
      "loss": 0.88975513,
      "num_input_tokens_seen": 45736655,
      "step": 2116,
      "time_per_iteration": 2.61495304107666
    },
    {
      "auxiliary_loss_clip": 0.01132067,
      "auxiliary_loss_mlp": 0.00969611,
      "balance_loss_clip": 1.04917526,
      "balance_loss_mlp": 1.27845955,
      "epoch": 0.1272809258980911,
      "flos": 32379897519360.0,
      "grad_norm": 2.0081224224122627,
      "language_loss": 0.7042737,
      "learning_rate": 3.9015465897043775e-06,
      "loss": 0.72529048,
      "num_input_tokens_seen": 45758195,
      "step": 2117,
      "time_per_iteration": 2.6310253143310547
    },
    {
      "auxiliary_loss_clip": 0.01125715,
      "auxiliary_loss_mlp": 0.01052655,
      "balance_loss_clip": 1.05090117,
      "balance_loss_mlp": 1.02967143,
      "epoch": 0.12734104915075906,
      "flos": 16034402401920.0,
      "grad_norm": 2.2573735597765436,
      "language_loss": 0.86778629,
      "learning_rate": 3.901425864420852e-06,
      "loss": 0.88957,
      "num_input_tokens_seen": 45774280,
      "step": 2118,
      "time_per_iteration": 3.945040702819824
    },
    {
      "auxiliary_loss_clip": 0.01150453,
      "auxiliary_loss_mlp": 0.01046701,
      "balance_loss_clip": 1.05088794,
      "balance_loss_mlp": 1.02735376,
      "epoch": 0.12740117240342702,
      "flos": 18260325244800.0,
      "grad_norm": 1.892355164566308,
      "language_loss": 0.87145126,
      "learning_rate": 3.901305067035068e-06,
      "loss": 0.89342284,
      "num_input_tokens_seen": 45792760,
      "step": 2119,
      "time_per_iteration": 2.5300886631011963
    },
    {
      "auxiliary_loss_clip": 0.01144184,
      "auxiliary_loss_mlp": 0.00928415,
      "balance_loss_clip": 1.0526706,
      "balance_loss_mlp": 1.20534086,
      "epoch": 0.127461295656095,
      "flos": 12121790664960.0,
      "grad_norm": 2.161859134602549,
      "language_loss": 0.87498701,
      "learning_rate": 3.901184197551605e-06,
      "loss": 0.89571297,
      "num_input_tokens_seen": 45804300,
      "step": 2120,
      "time_per_iteration": 2.469304084777832
    },
    {
      "auxiliary_loss_clip": 0.01163227,
      "auxiliary_loss_mlp": 0.0104043,
      "balance_loss_clip": 1.05259347,
      "balance_loss_mlp": 1.0211184,
      "epoch": 0.12752141890876295,
      "flos": 23149095966720.0,
      "grad_norm": 3.0201181968507833,
      "language_loss": 0.75503194,
      "learning_rate": 3.901063255975046e-06,
      "loss": 0.7770685,
      "num_input_tokens_seen": 45823780,
      "step": 2121,
      "time_per_iteration": 3.8766956329345703
    },
    {
      "auxiliary_loss_clip": 0.01110492,
      "auxiliary_loss_mlp": 0.01046142,
      "balance_loss_clip": 1.04793835,
      "balance_loss_mlp": 1.02593589,
      "epoch": 0.12758154216143094,
      "flos": 21615997628160.0,
      "grad_norm": 10.249906193211675,
      "language_loss": 0.83058739,
      "learning_rate": 3.900942242309978e-06,
      "loss": 0.85215372,
      "num_input_tokens_seen": 45840495,
      "step": 2122,
      "time_per_iteration": 2.577893018722534
    },
    {
      "auxiliary_loss_clip": 0.01147898,
      "auxiliary_loss_mlp": 0.0105129,
      "balance_loss_clip": 1.05724728,
      "balance_loss_mlp": 1.03122711,
      "epoch": 0.1276416654140989,
      "flos": 15924874855680.0,
      "grad_norm": 1.8756384973293219,
      "language_loss": 0.78799856,
      "learning_rate": 3.90082115656099e-06,
      "loss": 0.80999047,
      "num_input_tokens_seen": 45857735,
      "step": 2123,
      "time_per_iteration": 2.5404915809631348
    },
    {
      "auxiliary_loss_clip": 0.01166016,
      "auxiliary_loss_mlp": 0.01049498,
      "balance_loss_clip": 1.05500889,
      "balance_loss_mlp": 1.02918482,
      "epoch": 0.12770178866676687,
      "flos": 22382690451840.0,
      "grad_norm": 2.8535545380220277,
      "language_loss": 0.7921719,
      "learning_rate": 3.900699998732673e-06,
      "loss": 0.814327,
      "num_input_tokens_seen": 45876485,
      "step": 2124,
      "time_per_iteration": 2.477965831756592
    },
    {
      "auxiliary_loss_clip": 0.01157771,
      "auxiliary_loss_mlp": 0.00915889,
      "balance_loss_clip": 1.05397081,
      "balance_loss_mlp": 1.19300056,
      "epoch": 0.12776191191943484,
      "flos": 21652482867840.0,
      "grad_norm": 1.9454626683393996,
      "language_loss": 0.75339675,
      "learning_rate": 3.900578768829623e-06,
      "loss": 0.77413332,
      "num_input_tokens_seen": 45894645,
      "step": 2125,
      "time_per_iteration": 3.912152051925659
    },
    {
      "auxiliary_loss_clip": 0.01157484,
      "auxiliary_loss_mlp": 0.00910421,
      "balance_loss_clip": 1.05601883,
      "balance_loss_mlp": 1.19101906,
      "epoch": 0.1278220351721028,
      "flos": 25735561574400.0,
      "grad_norm": 2.299275066377786,
      "language_loss": 0.77680981,
      "learning_rate": 3.900457466856434e-06,
      "loss": 0.79748881,
      "num_input_tokens_seen": 45913755,
      "step": 2126,
      "time_per_iteration": 2.548546075820923
    },
    {
      "auxiliary_loss_clip": 0.01124445,
      "auxiliary_loss_mlp": 0.01045847,
      "balance_loss_clip": 1.05738878,
      "balance_loss_mlp": 1.02672625,
      "epoch": 0.12788215842477077,
      "flos": 41243224982400.0,
      "grad_norm": 1.5030248867320313,
      "language_loss": 0.69325984,
      "learning_rate": 3.9003360928177085e-06,
      "loss": 0.71496272,
      "num_input_tokens_seen": 45936095,
      "step": 2127,
      "time_per_iteration": 2.7434451580047607
    },
    {
      "auxiliary_loss_clip": 0.01044959,
      "auxiliary_loss_mlp": 0.01600361,
      "balance_loss_clip": 1.03847456,
      "balance_loss_mlp": 2.52254343,
      "epoch": 0.12794228167743876,
      "flos": 70877430881280.0,
      "grad_norm": 0.9633337272223885,
      "language_loss": 0.62765646,
      "learning_rate": 3.900214646718047e-06,
      "loss": 0.6541096,
      "num_input_tokens_seen": 46004655,
      "step": 2128,
      "time_per_iteration": 3.2330756187438965
    },
    {
      "auxiliary_loss_clip": 0.01144488,
      "auxiliary_loss_mlp": 0.01045499,
      "balance_loss_clip": 1.05185652,
      "balance_loss_mlp": 1.0245899,
      "epoch": 0.12800240493010673,
      "flos": 16289727252480.0,
      "grad_norm": 4.406400778939859,
      "language_loss": 0.77194238,
      "learning_rate": 3.900093128562056e-06,
      "loss": 0.79384226,
      "num_input_tokens_seen": 46023610,
      "step": 2129,
      "time_per_iteration": 2.49585223197937
    },
    {
      "auxiliary_loss_clip": 0.01120201,
      "auxiliary_loss_mlp": 0.01057829,
      "balance_loss_clip": 1.05194116,
      "balance_loss_mlp": 1.03579938,
      "epoch": 0.1280625281827747,
      "flos": 20631542601600.0,
      "grad_norm": 2.321244752293836,
      "language_loss": 0.79784781,
      "learning_rate": 3.899971538354343e-06,
      "loss": 0.81962812,
      "num_input_tokens_seen": 46041725,
      "step": 2130,
      "time_per_iteration": 2.6086061000823975
    },
    {
      "auxiliary_loss_clip": 0.01141912,
      "auxiliary_loss_mlp": 0.01054821,
      "balance_loss_clip": 1.05541718,
      "balance_loss_mlp": 1.03368485,
      "epoch": 0.12812265143544266,
      "flos": 22638230784000.0,
      "grad_norm": 2.0333279291806465,
      "language_loss": 0.71327096,
      "learning_rate": 3.899849876099518e-06,
      "loss": 0.73523831,
      "num_input_tokens_seen": 46061095,
      "step": 2131,
      "time_per_iteration": 2.5541210174560547
    },
    {
      "auxiliary_loss_clip": 0.01101008,
      "auxiliary_loss_mlp": 0.01049103,
      "balance_loss_clip": 1.05238068,
      "balance_loss_mlp": 1.02876592,
      "epoch": 0.12818277468811062,
      "flos": 34714701463680.0,
      "grad_norm": 2.3382552888801964,
      "language_loss": 0.72897524,
      "learning_rate": 3.899728141802197e-06,
      "loss": 0.75047636,
      "num_input_tokens_seen": 46082670,
      "step": 2132,
      "time_per_iteration": 2.7614617347717285
    },
    {
      "auxiliary_loss_clip": 0.0110523,
      "auxiliary_loss_mlp": 0.01054977,
      "balance_loss_clip": 1.04953039,
      "balance_loss_mlp": 1.03329325,
      "epoch": 0.1282428979407786,
      "flos": 23112107936640.0,
      "grad_norm": 1.9642762745602185,
      "language_loss": 0.82118356,
      "learning_rate": 3.8996063354669935e-06,
      "loss": 0.84278566,
      "num_input_tokens_seen": 46102410,
      "step": 2133,
      "time_per_iteration": 2.616182804107666
    },
    {
      "auxiliary_loss_clip": 0.01158863,
      "auxiliary_loss_mlp": 0.0105102,
      "balance_loss_clip": 1.05171287,
      "balance_loss_mlp": 1.03054023,
      "epoch": 0.12830302119344655,
      "flos": 20886508316160.0,
      "grad_norm": 2.7736510827666137,
      "language_loss": 0.80198789,
      "learning_rate": 3.899484457098528e-06,
      "loss": 0.82408673,
      "num_input_tokens_seen": 46121145,
      "step": 2134,
      "time_per_iteration": 2.5145421028137207
    },
    {
      "auxiliary_loss_clip": 0.01155995,
      "auxiliary_loss_mlp": 0.01049326,
      "balance_loss_clip": 1.05894172,
      "balance_loss_mlp": 1.0292275,
      "epoch": 0.12836314444611455,
      "flos": 21397768548480.0,
      "grad_norm": 2.6031746819427704,
      "language_loss": 0.82921124,
      "learning_rate": 3.899362506701421e-06,
      "loss": 0.85126442,
      "num_input_tokens_seen": 46140740,
      "step": 2135,
      "time_per_iteration": 2.521883010864258
    },
    {
      "auxiliary_loss_clip": 0.01140498,
      "auxiliary_loss_mlp": 0.01051685,
      "balance_loss_clip": 1.05532789,
      "balance_loss_mlp": 1.03119266,
      "epoch": 0.1284232676987825,
      "flos": 13662466773120.0,
      "grad_norm": 2.5755676571254673,
      "language_loss": 0.77400082,
      "learning_rate": 3.899240484280298e-06,
      "loss": 0.79592264,
      "num_input_tokens_seen": 46156805,
      "step": 2136,
      "time_per_iteration": 2.53132963180542
    },
    {
      "auxiliary_loss_clip": 0.01039973,
      "auxiliary_loss_mlp": 0.0103465,
      "balance_loss_clip": 1.03507328,
      "balance_loss_mlp": 1.03131187,
      "epoch": 0.12848339095145048,
      "flos": 59994737735040.0,
      "grad_norm": 0.9062860582387595,
      "language_loss": 0.5918833,
      "learning_rate": 3.899118389839785e-06,
      "loss": 0.61262953,
      "num_input_tokens_seen": 46222085,
      "step": 2137,
      "time_per_iteration": 3.331674098968506
    },
    {
      "auxiliary_loss_clip": 0.01152497,
      "auxiliary_loss_mlp": 0.01049406,
      "balance_loss_clip": 1.05377579,
      "balance_loss_mlp": 1.02968824,
      "epoch": 0.12854351420411844,
      "flos": 13881378211200.0,
      "grad_norm": 2.855591330032333,
      "language_loss": 0.82489932,
      "learning_rate": 3.898996223384512e-06,
      "loss": 0.84691828,
      "num_input_tokens_seen": 46239970,
      "step": 2138,
      "time_per_iteration": 2.522308111190796
    },
    {
      "auxiliary_loss_clip": 0.01155505,
      "auxiliary_loss_mlp": 0.01049386,
      "balance_loss_clip": 1.05531025,
      "balance_loss_mlp": 1.02796435,
      "epoch": 0.1286036374567864,
      "flos": 22637943475200.0,
      "grad_norm": 3.7844505449857095,
      "language_loss": 0.78723609,
      "learning_rate": 3.898873984919113e-06,
      "loss": 0.80928504,
      "num_input_tokens_seen": 46257740,
      "step": 2139,
      "time_per_iteration": 2.539017677307129
    },
    {
      "auxiliary_loss_clip": 0.0112953,
      "auxiliary_loss_mlp": 0.01046656,
      "balance_loss_clip": 1.04851472,
      "balance_loss_mlp": 1.0262835,
      "epoch": 0.12866376070945437,
      "flos": 16324775948160.0,
      "grad_norm": 1.9667935507730232,
      "language_loss": 0.84905934,
      "learning_rate": 3.8987516744482215e-06,
      "loss": 0.87082118,
      "num_input_tokens_seen": 46275445,
      "step": 2140,
      "time_per_iteration": 2.609879493713379
    },
    {
      "auxiliary_loss_clip": 0.01141051,
      "auxiliary_loss_mlp": 0.01045196,
      "balance_loss_clip": 1.05094457,
      "balance_loss_mlp": 1.02605081,
      "epoch": 0.12872388396212234,
      "flos": 11874546374400.0,
      "grad_norm": 3.6328093676822437,
      "language_loss": 0.86175317,
      "learning_rate": 3.898629291976476e-06,
      "loss": 0.88361561,
      "num_input_tokens_seen": 46291710,
      "step": 2141,
      "time_per_iteration": 2.5380964279174805
    },
    {
      "auxiliary_loss_clip": 0.011425,
      "auxiliary_loss_mlp": 0.01042437,
      "balance_loss_clip": 1.05265546,
      "balance_loss_mlp": 1.02241015,
      "epoch": 0.12878400721479033,
      "flos": 28366700722560.0,
      "grad_norm": 6.198603096869942,
      "language_loss": 0.68386787,
      "learning_rate": 3.898506837508518e-06,
      "loss": 0.70571727,
      "num_input_tokens_seen": 46311335,
      "step": 2142,
      "time_per_iteration": 2.606330156326294
    },
    {
      "auxiliary_loss_clip": 0.01158568,
      "auxiliary_loss_mlp": 0.01208883,
      "balance_loss_clip": 1.05319333,
      "balance_loss_mlp": 1.73464215,
      "epoch": 0.1288441304674583,
      "flos": 25885632597120.0,
      "grad_norm": 2.1008484213093506,
      "language_loss": 0.8318544,
      "learning_rate": 3.89838431104899e-06,
      "loss": 0.85552895,
      "num_input_tokens_seen": 46330985,
      "step": 2143,
      "time_per_iteration": 2.539196252822876
    },
    {
      "auxiliary_loss_clip": 0.01170197,
      "auxiliary_loss_mlp": 0.01221131,
      "balance_loss_clip": 1.05712664,
      "balance_loss_mlp": 1.75766504,
      "epoch": 0.12890425372012626,
      "flos": 20813789232000.0,
      "grad_norm": 1.614771935305401,
      "language_loss": 0.81665587,
      "learning_rate": 3.898261712602539e-06,
      "loss": 0.84056914,
      "num_input_tokens_seen": 46351295,
      "step": 2144,
      "time_per_iteration": 2.4903626441955566
    },
    {
      "auxiliary_loss_clip": 0.01128498,
      "auxiliary_loss_mlp": 0.01046578,
      "balance_loss_clip": 1.04690099,
      "balance_loss_mlp": 1.02538228,
      "epoch": 0.12896437697279423,
      "flos": 22565870835840.0,
      "grad_norm": 1.9373065418741835,
      "language_loss": 0.7809763,
      "learning_rate": 3.898139042173813e-06,
      "loss": 0.8027271,
      "num_input_tokens_seen": 46368600,
      "step": 2145,
      "time_per_iteration": 2.537078619003296
    },
    {
      "auxiliary_loss_clip": 0.01163695,
      "auxiliary_loss_mlp": 0.01044335,
      "balance_loss_clip": 1.05103672,
      "balance_loss_mlp": 1.02461755,
      "epoch": 0.1290245002254622,
      "flos": 17493776075520.0,
      "grad_norm": 2.342339000658999,
      "language_loss": 0.82205039,
      "learning_rate": 3.898016299767465e-06,
      "loss": 0.84413075,
      "num_input_tokens_seen": 46387370,
      "step": 2146,
      "time_per_iteration": 2.4538824558258057
    },
    {
      "auxiliary_loss_clip": 0.01137615,
      "auxiliary_loss_mlp": 0.01044856,
      "balance_loss_clip": 1.05349159,
      "balance_loss_mlp": 1.02401853,
      "epoch": 0.12908462347813016,
      "flos": 36315957859200.0,
      "grad_norm": 2.4995566866270744,
      "language_loss": 0.7163707,
      "learning_rate": 3.897893485388149e-06,
      "loss": 0.73819542,
      "num_input_tokens_seen": 46409570,
      "step": 2147,
      "time_per_iteration": 2.6480953693389893
    },
    {
      "auxiliary_loss_clip": 0.01139449,
      "auxiliary_loss_mlp": 0.01047212,
      "balance_loss_clip": 1.05029869,
      "balance_loss_mlp": 1.02857935,
      "epoch": 0.12914474673079815,
      "flos": 22528703237760.0,
      "grad_norm": 3.214006976296634,
      "language_loss": 0.71835649,
      "learning_rate": 3.897770599040521e-06,
      "loss": 0.74022305,
      "num_input_tokens_seen": 46429320,
      "step": 2148,
      "time_per_iteration": 2.5493922233581543
    },
    {
      "auxiliary_loss_clip": 0.01163156,
      "auxiliary_loss_mlp": 0.01040048,
      "balance_loss_clip": 1.05428028,
      "balance_loss_mlp": 1.02163053,
      "epoch": 0.12920486998346611,
      "flos": 21471888263040.0,
      "grad_norm": 2.007625228251199,
      "language_loss": 0.79210961,
      "learning_rate": 3.897647640729242e-06,
      "loss": 0.81414163,
      "num_input_tokens_seen": 46450155,
      "step": 2149,
      "time_per_iteration": 2.472214460372925
    },
    {
      "auxiliary_loss_clip": 0.0115457,
      "auxiliary_loss_mlp": 0.01039553,
      "balance_loss_clip": 1.05263031,
      "balance_loss_mlp": 1.01915646,
      "epoch": 0.12926499323613408,
      "flos": 27308556944640.0,
      "grad_norm": 28.094723201985126,
      "language_loss": 0.75932825,
      "learning_rate": 3.897524610458975e-06,
      "loss": 0.78126955,
      "num_input_tokens_seen": 46470280,
      "step": 2150,
      "time_per_iteration": 2.583061456680298
    },
    {
      "auxiliary_loss_clip": 0.01147277,
      "auxiliary_loss_mlp": 0.01044536,
      "balance_loss_clip": 1.05100036,
      "balance_loss_mlp": 1.02525949,
      "epoch": 0.12932511648880204,
      "flos": 22091131756800.0,
      "grad_norm": 3.501252316424131,
      "language_loss": 0.70738649,
      "learning_rate": 3.8974015082343835e-06,
      "loss": 0.72930467,
      "num_input_tokens_seen": 46487605,
      "step": 2151,
      "time_per_iteration": 2.495633125305176
    },
    {
      "auxiliary_loss_clip": 0.01165053,
      "auxiliary_loss_mlp": 0.01042354,
      "balance_loss_clip": 1.05645084,
      "balance_loss_mlp": 1.02379274,
      "epoch": 0.12938523974147,
      "flos": 20302780394880.0,
      "grad_norm": 2.0911879068712906,
      "language_loss": 0.83986878,
      "learning_rate": 3.897278334060137e-06,
      "loss": 0.86194283,
      "num_input_tokens_seen": 46505100,
      "step": 2152,
      "time_per_iteration": 2.4576117992401123
    },
    {
      "auxiliary_loss_clip": 0.01155071,
      "auxiliary_loss_mlp": 0.01055423,
      "balance_loss_clip": 1.052526,
      "balance_loss_mlp": 1.03599191,
      "epoch": 0.12944536299413797,
      "flos": 19499961467520.0,
      "grad_norm": 2.32682917579046,
      "language_loss": 0.78860021,
      "learning_rate": 3.897155087940906e-06,
      "loss": 0.81070507,
      "num_input_tokens_seen": 46524020,
      "step": 2153,
      "time_per_iteration": 2.4761552810668945
    },
    {
      "auxiliary_loss_clip": 0.0112209,
      "auxiliary_loss_mlp": 0.01049512,
      "balance_loss_clip": 1.0587008,
      "balance_loss_mlp": 1.43294477,
      "epoch": 0.12950548624680594,
      "flos": 27707919333120.0,
      "grad_norm": 1.7247830739546506,
      "language_loss": 0.80130327,
      "learning_rate": 3.897031769881364e-06,
      "loss": 0.82301927,
      "num_input_tokens_seen": 46544640,
      "step": 2154,
      "time_per_iteration": 4.008934497833252
    },
    {
      "auxiliary_loss_clip": 0.01148052,
      "auxiliary_loss_mlp": 0.01048444,
      "balance_loss_clip": 1.05391407,
      "balance_loss_mlp": 1.02896512,
      "epoch": 0.12956560949947393,
      "flos": 17565740974080.0,
      "grad_norm": 1.905674736202788,
      "language_loss": 0.83265448,
      "learning_rate": 3.896908379886188e-06,
      "loss": 0.8546195,
      "num_input_tokens_seen": 46561395,
      "step": 2155,
      "time_per_iteration": 2.4858200550079346
    },
    {
      "auxiliary_loss_clip": 0.01157406,
      "auxiliary_loss_mlp": 0.01048891,
      "balance_loss_clip": 1.05346072,
      "balance_loss_mlp": 1.02953172,
      "epoch": 0.1296257327521419,
      "flos": 20740711011840.0,
      "grad_norm": 2.6411165690162273,
      "language_loss": 0.75885141,
      "learning_rate": 3.896784917960055e-06,
      "loss": 0.78091443,
      "num_input_tokens_seen": 46579395,
      "step": 2156,
      "time_per_iteration": 4.0071587562561035
    },
    {
      "auxiliary_loss_clip": 0.01108867,
      "auxiliary_loss_mlp": 0.01047997,
      "balance_loss_clip": 1.05753994,
      "balance_loss_mlp": 1.02856588,
      "epoch": 0.12968585600480986,
      "flos": 16395735265920.0,
      "grad_norm": 2.2329684924901665,
      "language_loss": 0.86579466,
      "learning_rate": 3.896661384107648e-06,
      "loss": 0.88736331,
      "num_input_tokens_seen": 46597090,
      "step": 2157,
      "time_per_iteration": 2.623391628265381
    },
    {
      "auxiliary_loss_clip": 0.01163959,
      "auxiliary_loss_mlp": 0.01053978,
      "balance_loss_clip": 1.05051041,
      "balance_loss_mlp": 1.03375971,
      "epoch": 0.12974597925747783,
      "flos": 28329533124480.0,
      "grad_norm": 2.6681095633731333,
      "language_loss": 0.80936694,
      "learning_rate": 3.896537778333651e-06,
      "loss": 0.83154631,
      "num_input_tokens_seen": 46617355,
      "step": 2158,
      "time_per_iteration": 2.556306838989258
    },
    {
      "auxiliary_loss_clip": 0.01168612,
      "auxiliary_loss_mlp": 0.0105725,
      "balance_loss_clip": 1.05498242,
      "balance_loss_mlp": 1.03769982,
      "epoch": 0.1298061025101458,
      "flos": 9683025782400.0,
      "grad_norm": 2.548516663119676,
      "language_loss": 0.74692291,
      "learning_rate": 3.896414100642752e-06,
      "loss": 0.76918149,
      "num_input_tokens_seen": 46633130,
      "step": 2159,
      "time_per_iteration": 3.8341352939605713
    },
    {
      "auxiliary_loss_clip": 0.01120635,
      "auxiliary_loss_mlp": 0.0105665,
      "balance_loss_clip": 1.04890037,
      "balance_loss_mlp": 1.03551412,
      "epoch": 0.12986622576281376,
      "flos": 27709535445120.0,
      "grad_norm": 1.9742773416805233,
      "language_loss": 0.82187343,
      "learning_rate": 3.89629035103964e-06,
      "loss": 0.84364629,
      "num_input_tokens_seen": 46650575,
      "step": 2160,
      "time_per_iteration": 2.619985342025757
    },
    {
      "auxiliary_loss_clip": 0.01149181,
      "auxiliary_loss_mlp": 0.01042913,
      "balance_loss_clip": 1.05828536,
      "balance_loss_mlp": 1.02329135,
      "epoch": 0.12992634901548175,
      "flos": 18802719590400.0,
      "grad_norm": 1.6981534412638162,
      "language_loss": 0.8222059,
      "learning_rate": 3.896166529529008e-06,
      "loss": 0.84412682,
      "num_input_tokens_seen": 46668780,
      "step": 2161,
      "time_per_iteration": 2.506188154220581
    },
    {
      "auxiliary_loss_clip": 0.01143413,
      "auxiliary_loss_mlp": 0.01049762,
      "balance_loss_clip": 1.05141544,
      "balance_loss_mlp": 1.02998543,
      "epoch": 0.12998647226814972,
      "flos": 29127575543040.0,
      "grad_norm": 2.0756361319277556,
      "language_loss": 0.82257283,
      "learning_rate": 3.896042636115551e-06,
      "loss": 0.84450459,
      "num_input_tokens_seen": 46687550,
      "step": 2162,
      "time_per_iteration": 2.5854687690734863
    },
    {
      "auxiliary_loss_clip": 0.01128018,
      "auxiliary_loss_mlp": 0.01048263,
      "balance_loss_clip": 1.04941368,
      "balance_loss_mlp": 1.02836716,
      "epoch": 0.13004659552081768,
      "flos": 19573686132480.0,
      "grad_norm": 2.7510951109062467,
      "language_loss": 0.72996664,
      "learning_rate": 3.895918670803968e-06,
      "loss": 0.75172949,
      "num_input_tokens_seen": 46706730,
      "step": 2163,
      "time_per_iteration": 2.556450366973877
    },
    {
      "auxiliary_loss_clip": 0.01168068,
      "auxiliary_loss_mlp": 0.01027341,
      "balance_loss_clip": 1.05443156,
      "balance_loss_mlp": 1.38424635,
      "epoch": 0.13010671877348565,
      "flos": 22490709626880.0,
      "grad_norm": 2.66227958513746,
      "language_loss": 0.81849277,
      "learning_rate": 3.895794633598958e-06,
      "loss": 0.84044689,
      "num_input_tokens_seen": 46724250,
      "step": 2164,
      "time_per_iteration": 3.8764946460723877
    },
    {
      "auxiliary_loss_clip": 0.01117577,
      "auxiliary_loss_mlp": 0.01039249,
      "balance_loss_clip": 1.04914641,
      "balance_loss_mlp": 1.01976979,
      "epoch": 0.1301668420261536,
      "flos": 23878226142720.0,
      "grad_norm": 2.306127741083391,
      "language_loss": 0.72178757,
      "learning_rate": 3.8956705245052256e-06,
      "loss": 0.74335581,
      "num_input_tokens_seen": 46744105,
      "step": 2165,
      "time_per_iteration": 2.6277711391448975
    },
    {
      "auxiliary_loss_clip": 0.01111682,
      "auxiliary_loss_mlp": 0.01040979,
      "balance_loss_clip": 1.05376935,
      "balance_loss_mlp": 1.02005792,
      "epoch": 0.13022696527882158,
      "flos": 23150065633920.0,
      "grad_norm": 1.8911191195027592,
      "language_loss": 0.75086975,
      "learning_rate": 3.8955463435274765e-06,
      "loss": 0.77239633,
      "num_input_tokens_seen": 46764250,
      "step": 2166,
      "time_per_iteration": 2.7056262493133545
    },
    {
      "auxiliary_loss_clip": 0.01162649,
      "auxiliary_loss_mlp": 0.01044523,
      "balance_loss_clip": 1.05048418,
      "balance_loss_mlp": 1.02506781,
      "epoch": 0.13028708853148954,
      "flos": 26908548111360.0,
      "grad_norm": 1.8671885225749185,
      "language_loss": 0.83024746,
      "learning_rate": 3.895422090670421e-06,
      "loss": 0.85231912,
      "num_input_tokens_seen": 46786865,
      "step": 2167,
      "time_per_iteration": 2.566591501235962
    },
    {
      "auxiliary_loss_clip": 0.01108371,
      "auxiliary_loss_mlp": 0.01057247,
      "balance_loss_clip": 1.04981565,
      "balance_loss_mlp": 1.03690982,
      "epoch": 0.13034721178415754,
      "flos": 21251468453760.0,
      "grad_norm": 1.621918763263411,
      "language_loss": 0.83500075,
      "learning_rate": 3.89529776593877e-06,
      "loss": 0.85665691,
      "num_input_tokens_seen": 46807030,
      "step": 2168,
      "time_per_iteration": 2.6300740242004395
    },
    {
      "auxiliary_loss_clip": 0.01077795,
      "auxiliary_loss_mlp": 0.01059791,
      "balance_loss_clip": 1.04512739,
      "balance_loss_mlp": 1.03598535,
      "epoch": 0.1304073350368255,
      "flos": 18767239931520.0,
      "grad_norm": 1.9985476727546274,
      "language_loss": 0.79849136,
      "learning_rate": 3.8951733693372375e-06,
      "loss": 0.81986725,
      "num_input_tokens_seen": 46826280,
      "step": 2169,
      "time_per_iteration": 2.6715986728668213
    },
    {
      "auxiliary_loss_clip": 0.01164418,
      "auxiliary_loss_mlp": 0.01042472,
      "balance_loss_clip": 1.0521462,
      "balance_loss_mlp": 1.02142012,
      "epoch": 0.13046745828949347,
      "flos": 28364653647360.0,
      "grad_norm": 2.4646733042315048,
      "language_loss": 0.66706544,
      "learning_rate": 3.8950489008705406e-06,
      "loss": 0.68913436,
      "num_input_tokens_seen": 46846505,
      "step": 2170,
      "time_per_iteration": 2.5381481647491455
    },
    {
      "auxiliary_loss_clip": 0.01137071,
      "auxiliary_loss_mlp": 0.01045607,
      "balance_loss_clip": 1.05263996,
      "balance_loss_mlp": 1.02616405,
      "epoch": 0.13052758154216143,
      "flos": 29605044055680.0,
      "grad_norm": 1.721648252953711,
      "language_loss": 0.66952002,
      "learning_rate": 3.8949243605434e-06,
      "loss": 0.69134682,
      "num_input_tokens_seen": 46867380,
      "step": 2171,
      "time_per_iteration": 2.6247501373291016
    },
    {
      "auxiliary_loss_clip": 0.01151564,
      "auxiliary_loss_mlp": 0.01045575,
      "balance_loss_clip": 1.04969692,
      "balance_loss_mlp": 1.02467811,
      "epoch": 0.1305877047948294,
      "flos": 19390864884480.0,
      "grad_norm": 1.9853251306567843,
      "language_loss": 0.72057128,
      "learning_rate": 3.894799748360537e-06,
      "loss": 0.74254262,
      "num_input_tokens_seen": 46886810,
      "step": 2172,
      "time_per_iteration": 2.505753755569458
    },
    {
      "auxiliary_loss_clip": 0.01122335,
      "auxiliary_loss_mlp": 0.01045721,
      "balance_loss_clip": 1.05043888,
      "balance_loss_mlp": 1.02596807,
      "epoch": 0.13064782804749736,
      "flos": 16873527000960.0,
      "grad_norm": 2.1508486852354705,
      "language_loss": 0.75692248,
      "learning_rate": 3.894675064326678e-06,
      "loss": 0.77860296,
      "num_input_tokens_seen": 46905620,
      "step": 2173,
      "time_per_iteration": 2.6044085025787354
    },
    {
      "auxiliary_loss_clip": 0.01131902,
      "auxiliary_loss_mlp": 0.01057356,
      "balance_loss_clip": 1.05675721,
      "balance_loss_mlp": 1.03577876,
      "epoch": 0.13070795130016533,
      "flos": 24499085748480.0,
      "grad_norm": 2.3458353980653803,
      "language_loss": 0.70271695,
      "learning_rate": 3.894550308446551e-06,
      "loss": 0.72460955,
      "num_input_tokens_seen": 46925120,
      "step": 2174,
      "time_per_iteration": 2.615194082260132
    },
    {
      "auxiliary_loss_clip": 0.0106491,
      "auxiliary_loss_mlp": 0.01037933,
      "balance_loss_clip": 1.03661478,
      "balance_loss_mlp": 1.0342133,
      "epoch": 0.13076807455283332,
      "flos": 71054505953280.0,
      "grad_norm": 0.8230570650558096,
      "language_loss": 0.59036255,
      "learning_rate": 3.894425480724886e-06,
      "loss": 0.61139095,
      "num_input_tokens_seen": 46988195,
      "step": 2175,
      "time_per_iteration": 3.2819745540618896
    },
    {
      "auxiliary_loss_clip": 0.0115523,
      "auxiliary_loss_mlp": 0.01053874,
      "balance_loss_clip": 1.05332994,
      "balance_loss_mlp": 1.03453779,
      "epoch": 0.13082819780550128,
      "flos": 20264499475200.0,
      "grad_norm": 2.0294399164933177,
      "language_loss": 0.79678428,
      "learning_rate": 3.894300581166417e-06,
      "loss": 0.81887531,
      "num_input_tokens_seen": 47004720,
      "step": 2176,
      "time_per_iteration": 2.528665542602539
    },
    {
      "auxiliary_loss_clip": 0.01162388,
      "auxiliary_loss_mlp": 0.01048995,
      "balance_loss_clip": 1.05122828,
      "balance_loss_mlp": 1.02729893,
      "epoch": 0.13088832105816925,
      "flos": 34203441231360.0,
      "grad_norm": 1.8895248504771243,
      "language_loss": 0.74679035,
      "learning_rate": 3.894175609775881e-06,
      "loss": 0.76890415,
      "num_input_tokens_seen": 47024255,
      "step": 2177,
      "time_per_iteration": 2.5832173824310303
    },
    {
      "auxiliary_loss_clip": 0.0112434,
      "auxiliary_loss_mlp": 0.01047618,
      "balance_loss_clip": 1.04840779,
      "balance_loss_mlp": 1.02658916,
      "epoch": 0.13094844431083721,
      "flos": 17894970057600.0,
      "grad_norm": 2.07745809167143,
      "language_loss": 0.82303649,
      "learning_rate": 3.894050566558015e-06,
      "loss": 0.84475607,
      "num_input_tokens_seen": 47042465,
      "step": 2178,
      "time_per_iteration": 2.590088129043579
    },
    {
      "auxiliary_loss_clip": 0.01164037,
      "auxiliary_loss_mlp": 0.01042731,
      "balance_loss_clip": 1.05400443,
      "balance_loss_mlp": 1.02232265,
      "epoch": 0.13100856756350518,
      "flos": 17311313963520.0,
      "grad_norm": 2.6079090705551193,
      "language_loss": 0.7456677,
      "learning_rate": 3.893925451517562e-06,
      "loss": 0.76773536,
      "num_input_tokens_seen": 47060370,
      "step": 2179,
      "time_per_iteration": 2.4402098655700684
    },
    {
      "auxiliary_loss_clip": 0.01123597,
      "auxiliary_loss_mlp": 0.01045727,
      "balance_loss_clip": 1.04891157,
      "balance_loss_mlp": 1.02579546,
      "epoch": 0.13106869081617314,
      "flos": 22200551562240.0,
      "grad_norm": 2.3631885842601386,
      "language_loss": 0.84773183,
      "learning_rate": 3.893800264659266e-06,
      "loss": 0.86942506,
      "num_input_tokens_seen": 47081415,
      "step": 2180,
      "time_per_iteration": 2.5952231884002686
    },
    {
      "auxiliary_loss_clip": 0.01153832,
      "auxiliary_loss_mlp": 0.01055493,
      "balance_loss_clip": 1.05470276,
      "balance_loss_mlp": 1.03588367,
      "epoch": 0.13112881406884114,
      "flos": 21763123735680.0,
      "grad_norm": 1.9661799416891512,
      "language_loss": 0.89745599,
      "learning_rate": 3.8936750059878746e-06,
      "loss": 0.91954923,
      "num_input_tokens_seen": 47099860,
      "step": 2181,
      "time_per_iteration": 2.508884906768799
    },
    {
      "auxiliary_loss_clip": 0.01149816,
      "auxiliary_loss_mlp": 0.01051319,
      "balance_loss_clip": 1.0515908,
      "balance_loss_mlp": 1.03170872,
      "epoch": 0.1311889373215091,
      "flos": 23331091201920.0,
      "grad_norm": 1.913852535144799,
      "language_loss": 0.68434632,
      "learning_rate": 3.893549675508137e-06,
      "loss": 0.70635766,
      "num_input_tokens_seen": 47118540,
      "step": 2182,
      "time_per_iteration": 2.5334997177124023
    },
    {
      "auxiliary_loss_clip": 0.01117916,
      "auxiliary_loss_mlp": 0.01053002,
      "balance_loss_clip": 1.04838145,
      "balance_loss_mlp": 1.03208077,
      "epoch": 0.13124906057417707,
      "flos": 21467363149440.0,
      "grad_norm": 2.3167558335095233,
      "language_loss": 0.78465056,
      "learning_rate": 3.893424273224806e-06,
      "loss": 0.80635977,
      "num_input_tokens_seen": 47136710,
      "step": 2183,
      "time_per_iteration": 2.561140298843384
    },
    {
      "auxiliary_loss_clip": 0.01159859,
      "auxiliary_loss_mlp": 0.01044443,
      "balance_loss_clip": 1.04996109,
      "balance_loss_mlp": 1.02463078,
      "epoch": 0.13130918382684503,
      "flos": 23255319461760.0,
      "grad_norm": 1.8070099970127325,
      "language_loss": 0.86052251,
      "learning_rate": 3.893298799142636e-06,
      "loss": 0.8825655,
      "num_input_tokens_seen": 47157155,
      "step": 2184,
      "time_per_iteration": 2.4928627014160156
    },
    {
      "auxiliary_loss_clip": 0.01131249,
      "auxiliary_loss_mlp": 0.01049525,
      "balance_loss_clip": 1.05110741,
      "balance_loss_mlp": 1.02860427,
      "epoch": 0.131369307079513,
      "flos": 20850274471680.0,
      "grad_norm": 2.203560559324091,
      "language_loss": 0.82368201,
      "learning_rate": 3.893173253266387e-06,
      "loss": 0.84548974,
      "num_input_tokens_seen": 47176820,
      "step": 2185,
      "time_per_iteration": 2.610435724258423
    },
    {
      "auxiliary_loss_clip": 0.01138322,
      "auxiliary_loss_mlp": 0.01054274,
      "balance_loss_clip": 1.05181158,
      "balance_loss_mlp": 1.03403187,
      "epoch": 0.13142943033218096,
      "flos": 17858341163520.0,
      "grad_norm": 1.9599731716995554,
      "language_loss": 0.72886199,
      "learning_rate": 3.893047635600818e-06,
      "loss": 0.75078791,
      "num_input_tokens_seen": 47195855,
      "step": 2186,
      "time_per_iteration": 2.5582427978515625
    },
    {
      "auxiliary_loss_clip": 0.01151454,
      "auxiliary_loss_mlp": 0.01048982,
      "balance_loss_clip": 1.05346835,
      "balance_loss_mlp": 1.02735758,
      "epoch": 0.13148955358484893,
      "flos": 20996035862400.0,
      "grad_norm": 2.1140512805635683,
      "language_loss": 0.79912448,
      "learning_rate": 3.892921946150693e-06,
      "loss": 0.82112885,
      "num_input_tokens_seen": 47214535,
      "step": 2187,
      "time_per_iteration": 2.5074751377105713
    },
    {
      "auxiliary_loss_clip": 0.01051035,
      "auxiliary_loss_mlp": 0.01045968,
      "balance_loss_clip": 1.04096484,
      "balance_loss_mlp": 1.0421052,
      "epoch": 0.13154967683751692,
      "flos": 70172467580160.0,
      "grad_norm": 0.8437720511865218,
      "language_loss": 0.59011048,
      "learning_rate": 3.892796184920778e-06,
      "loss": 0.61108047,
      "num_input_tokens_seen": 47270300,
      "step": 2188,
      "time_per_iteration": 3.2248923778533936
    },
    {
      "auxiliary_loss_clip": 0.01090779,
      "auxiliary_loss_mlp": 0.01046624,
      "balance_loss_clip": 1.05075133,
      "balance_loss_mlp": 1.02598882,
      "epoch": 0.1316098000901849,
      "flos": 20376145923840.0,
      "grad_norm": 1.9237537496699424,
      "language_loss": 0.73997676,
      "learning_rate": 3.892670351915842e-06,
      "loss": 0.76135087,
      "num_input_tokens_seen": 47290720,
      "step": 2189,
      "time_per_iteration": 2.6802003383636475
    },
    {
      "auxiliary_loss_clip": 0.01155126,
      "auxiliary_loss_mlp": 0.01042717,
      "balance_loss_clip": 1.05457592,
      "balance_loss_mlp": 1.02209401,
      "epoch": 0.13166992334285285,
      "flos": 23221132692480.0,
      "grad_norm": 2.015733404581712,
      "language_loss": 0.72549021,
      "learning_rate": 3.892544447140657e-06,
      "loss": 0.74746859,
      "num_input_tokens_seen": 47311820,
      "step": 2190,
      "time_per_iteration": 2.5325405597686768
    },
    {
      "auxiliary_loss_clip": 0.01153132,
      "auxiliary_loss_mlp": 0.01047444,
      "balance_loss_clip": 1.05429578,
      "balance_loss_mlp": 1.02745259,
      "epoch": 0.13173004659552082,
      "flos": 23330947547520.0,
      "grad_norm": 2.0957332985675454,
      "language_loss": 0.74930894,
      "learning_rate": 3.892418470599996e-06,
      "loss": 0.77131474,
      "num_input_tokens_seen": 47331605,
      "step": 2191,
      "time_per_iteration": 2.521662473678589
    },
    {
      "auxiliary_loss_clip": 0.01127001,
      "auxiliary_loss_mlp": 0.01051641,
      "balance_loss_clip": 1.05519044,
      "balance_loss_mlp": 1.03044522,
      "epoch": 0.13179016984818878,
      "flos": 21251504367360.0,
      "grad_norm": 3.0975477094958723,
      "language_loss": 0.79357803,
      "learning_rate": 3.892292422298637e-06,
      "loss": 0.81536448,
      "num_input_tokens_seen": 47350455,
      "step": 2192,
      "time_per_iteration": 2.5818686485290527
    },
    {
      "auxiliary_loss_clip": 0.0111124,
      "auxiliary_loss_mlp": 0.01047961,
      "balance_loss_clip": 1.0508976,
      "balance_loss_mlp": 1.02762401,
      "epoch": 0.13185029310085675,
      "flos": 17778690754560.0,
      "grad_norm": 5.026571639057798,
      "language_loss": 0.8513701,
      "learning_rate": 3.892166302241361e-06,
      "loss": 0.87296206,
      "num_input_tokens_seen": 47368225,
      "step": 2193,
      "time_per_iteration": 3.9979944229125977
    },
    {
      "auxiliary_loss_clip": 0.01075375,
      "auxiliary_loss_mlp": 0.01003199,
      "balance_loss_clip": 1.05942488,
      "balance_loss_mlp": 0.99947929,
      "epoch": 0.1319104163535247,
      "flos": 69851785933440.0,
      "grad_norm": 0.7621368862301643,
      "language_loss": 0.54098397,
      "learning_rate": 3.8920401104329475e-06,
      "loss": 0.56176972,
      "num_input_tokens_seen": 47427125,
      "step": 2194,
      "time_per_iteration": 3.115755796432495
    },
    {
      "auxiliary_loss_clip": 0.01156941,
      "auxiliary_loss_mlp": 0.01046521,
      "balance_loss_clip": 1.04784358,
      "balance_loss_mlp": 1.02641034,
      "epoch": 0.1319705396061927,
      "flos": 25193095401600.0,
      "grad_norm": 2.296901037438971,
      "language_loss": 0.72432137,
      "learning_rate": 3.891913846878185e-06,
      "loss": 0.74635601,
      "num_input_tokens_seen": 47450275,
      "step": 2195,
      "time_per_iteration": 3.943492889404297
    },
    {
      "auxiliary_loss_clip": 0.01131241,
      "auxiliary_loss_mlp": 0.01001149,
      "balance_loss_clip": 1.05023837,
      "balance_loss_mlp": 1.33380735,
      "epoch": 0.13203066285886067,
      "flos": 20740459616640.0,
      "grad_norm": 2.255468187619569,
      "language_loss": 0.78381968,
      "learning_rate": 3.891787511581859e-06,
      "loss": 0.80514354,
      "num_input_tokens_seen": 47469155,
      "step": 2196,
      "time_per_iteration": 2.6567015647888184
    },
    {
      "auxiliary_loss_clip": 0.01153103,
      "auxiliary_loss_mlp": 0.01043306,
      "balance_loss_clip": 1.04967487,
      "balance_loss_mlp": 1.02372003,
      "epoch": 0.13209078611152864,
      "flos": 22054395121920.0,
      "grad_norm": 2.086610907915476,
      "language_loss": 0.75126624,
      "learning_rate": 3.89166110454876e-06,
      "loss": 0.77323031,
      "num_input_tokens_seen": 47488405,
      "step": 2197,
      "time_per_iteration": 2.4906678199768066
    },
    {
      "auxiliary_loss_clip": 0.01165387,
      "auxiliary_loss_mlp": 0.01044873,
      "balance_loss_clip": 1.05117321,
      "balance_loss_mlp": 1.0245719,
      "epoch": 0.1321509093641966,
      "flos": 16284950743680.0,
      "grad_norm": 2.1508735889054584,
      "language_loss": 0.79914016,
      "learning_rate": 3.891534625783685e-06,
      "loss": 0.82124275,
      "num_input_tokens_seen": 47505650,
      "step": 2198,
      "time_per_iteration": 3.816161632537842
    },
    {
      "auxiliary_loss_clip": 0.01158741,
      "auxiliary_loss_mlp": 0.01045942,
      "balance_loss_clip": 1.04967189,
      "balance_loss_mlp": 1.02617764,
      "epoch": 0.13221103261686457,
      "flos": 16983018633600.0,
      "grad_norm": 2.545641986529344,
      "language_loss": 0.8295548,
      "learning_rate": 3.891408075291425e-06,
      "loss": 0.8516016,
      "num_input_tokens_seen": 47521540,
      "step": 2199,
      "time_per_iteration": 2.4313833713531494
    },
    {
      "auxiliary_loss_clip": 0.01113431,
      "auxiliary_loss_mlp": 0.01053708,
      "balance_loss_clip": 1.04757524,
      "balance_loss_mlp": 1.03263187,
      "epoch": 0.13227115586953253,
      "flos": 34233605677440.0,
      "grad_norm": 1.7663853326046384,
      "language_loss": 0.69600111,
      "learning_rate": 3.8912814530767826e-06,
      "loss": 0.71767247,
      "num_input_tokens_seen": 47543625,
      "step": 2200,
      "time_per_iteration": 2.6982061862945557
    },
    {
      "auxiliary_loss_clip": 0.01162535,
      "auxiliary_loss_mlp": 0.01049152,
      "balance_loss_clip": 1.05080104,
      "balance_loss_mlp": 1.02829087,
      "epoch": 0.13233127912220052,
      "flos": 20704656735360.0,
      "grad_norm": 2.042289833481919,
      "language_loss": 0.84593868,
      "learning_rate": 3.891154759144557e-06,
      "loss": 0.86805552,
      "num_input_tokens_seen": 47563740,
      "step": 2201,
      "time_per_iteration": 2.461141586303711
    },
    {
      "auxiliary_loss_clip": 0.01164125,
      "auxiliary_loss_mlp": 0.01045651,
      "balance_loss_clip": 1.05157948,
      "balance_loss_mlp": 1.02519453,
      "epoch": 0.1323914023748685,
      "flos": 25805048434560.0,
      "grad_norm": 1.7598336170195388,
      "language_loss": 0.86620033,
      "learning_rate": 3.891027993499554e-06,
      "loss": 0.88829809,
      "num_input_tokens_seen": 47582655,
      "step": 2202,
      "time_per_iteration": 3.9735209941864014
    },
    {
      "auxiliary_loss_clip": 0.01131233,
      "auxiliary_loss_mlp": 0.0104508,
      "balance_loss_clip": 1.05158496,
      "balance_loss_mlp": 1.02536297,
      "epoch": 0.13245152562753645,
      "flos": 21251540280960.0,
      "grad_norm": 2.79915510532735,
      "language_loss": 0.72507739,
      "learning_rate": 3.89090115614658e-06,
      "loss": 0.74684048,
      "num_input_tokens_seen": 47600875,
      "step": 2203,
      "time_per_iteration": 2.5517735481262207
    },
    {
      "auxiliary_loss_clip": 0.01115622,
      "auxiliary_loss_mlp": 0.010484,
      "balance_loss_clip": 1.04834199,
      "balance_loss_mlp": 1.02914798,
      "epoch": 0.13251164888020442,
      "flos": 26610955931520.0,
      "grad_norm": 3.7444895155290223,
      "language_loss": 0.73605746,
      "learning_rate": 3.890774247090444e-06,
      "loss": 0.7576977,
      "num_input_tokens_seen": 47619250,
      "step": 2204,
      "time_per_iteration": 2.643855094909668
    },
    {
      "auxiliary_loss_clip": 0.01161217,
      "auxiliary_loss_mlp": 0.01049598,
      "balance_loss_clip": 1.06220973,
      "balance_loss_mlp": 1.02817631,
      "epoch": 0.13257177213287238,
      "flos": 29826541272960.0,
      "grad_norm": 1.9968036017760995,
      "language_loss": 0.7844488,
      "learning_rate": 3.89064726633596e-06,
      "loss": 0.80655694,
      "num_input_tokens_seen": 47639445,
      "step": 2205,
      "time_per_iteration": 2.5740411281585693
    },
    {
      "auxiliary_loss_clip": 0.01119615,
      "auxiliary_loss_mlp": 0.01050174,
      "balance_loss_clip": 1.04946899,
      "balance_loss_mlp": 1.03025436,
      "epoch": 0.13263189538554035,
      "flos": 21288456483840.0,
      "grad_norm": 1.8496959112258748,
      "language_loss": 0.79164529,
      "learning_rate": 3.890520213887941e-06,
      "loss": 0.81334317,
      "num_input_tokens_seen": 47658740,
      "step": 2206,
      "time_per_iteration": 2.550945281982422
    },
    {
      "auxiliary_loss_clip": 0.01122669,
      "auxiliary_loss_mlp": 0.01047952,
      "balance_loss_clip": 1.05255294,
      "balance_loss_mlp": 1.02933192,
      "epoch": 0.13269201863820831,
      "flos": 16874101618560.0,
      "grad_norm": 2.2509158196116603,
      "language_loss": 0.74600023,
      "learning_rate": 3.890393089751208e-06,
      "loss": 0.76770639,
      "num_input_tokens_seen": 47676880,
      "step": 2207,
      "time_per_iteration": 2.5660173892974854
    },
    {
      "auxiliary_loss_clip": 0.01135705,
      "auxiliary_loss_mlp": 0.01044037,
      "balance_loss_clip": 1.04789925,
      "balance_loss_mlp": 1.0241406,
      "epoch": 0.1327521418908763,
      "flos": 23768914078080.0,
      "grad_norm": 1.9173924079756883,
      "language_loss": 0.84371287,
      "learning_rate": 3.890265893930578e-06,
      "loss": 0.86551028,
      "num_input_tokens_seen": 47696635,
      "step": 2208,
      "time_per_iteration": 2.5439419746398926
    },
    {
      "auxiliary_loss_clip": 0.01144171,
      "auxiliary_loss_mlp": 0.01050499,
      "balance_loss_clip": 1.05369985,
      "balance_loss_mlp": 1.03180647,
      "epoch": 0.13281226514354427,
      "flos": 26505594362880.0,
      "grad_norm": 1.7599145468641564,
      "language_loss": 0.85447544,
      "learning_rate": 3.890138626430876e-06,
      "loss": 0.87642217,
      "num_input_tokens_seen": 47717760,
      "step": 2209,
      "time_per_iteration": 2.5741400718688965
    },
    {
      "auxiliary_loss_clip": 0.01133563,
      "auxiliary_loss_mlp": 0.01015829,
      "balance_loss_clip": 1.05151224,
      "balance_loss_mlp": 1.35585117,
      "epoch": 0.13287238839621224,
      "flos": 24498762526080.0,
      "grad_norm": 3.0084070121474307,
      "language_loss": 0.82481062,
      "learning_rate": 3.890011287256929e-06,
      "loss": 0.84630454,
      "num_input_tokens_seen": 47737685,
      "step": 2210,
      "time_per_iteration": 2.5543212890625
    },
    {
      "auxiliary_loss_clip": 0.01057908,
      "auxiliary_loss_mlp": 0.01921028,
      "balance_loss_clip": 1.05018413,
      "balance_loss_mlp": 3.04678535,
      "epoch": 0.1329325116488802,
      "flos": 67694344369920.0,
      "grad_norm": 0.7721610982941081,
      "language_loss": 0.58026803,
      "learning_rate": 3.889883876413563e-06,
      "loss": 0.61005735,
      "num_input_tokens_seen": 47802415,
      "step": 2211,
      "time_per_iteration": 3.279937744140625
    },
    {
      "auxiliary_loss_clip": 0.01067224,
      "auxiliary_loss_mlp": 0.01022581,
      "balance_loss_clip": 1.03822756,
      "balance_loss_mlp": 1.01936269,
      "epoch": 0.13299263490154817,
      "flos": 72261894741120.0,
      "grad_norm": 0.7968550564030759,
      "language_loss": 0.55341363,
      "learning_rate": 3.889756393905611e-06,
      "loss": 0.57431167,
      "num_input_tokens_seen": 47871485,
      "step": 2212,
      "time_per_iteration": 3.2102363109588623
    },
    {
      "auxiliary_loss_clip": 0.01129196,
      "auxiliary_loss_mlp": 0.01060564,
      "balance_loss_clip": 1.05259466,
      "balance_loss_mlp": 1.0402627,
      "epoch": 0.13305275815421613,
      "flos": 17931275729280.0,
      "grad_norm": 2.6594402613055395,
      "language_loss": 0.73812592,
      "learning_rate": 3.889628839737908e-06,
      "loss": 0.76002353,
      "num_input_tokens_seen": 47888315,
      "step": 2213,
      "time_per_iteration": 2.5477957725524902
    },
    {
      "auxiliary_loss_clip": 0.01112257,
      "auxiliary_loss_mlp": 0.01052116,
      "balance_loss_clip": 1.04930389,
      "balance_loss_mlp": 1.03381753,
      "epoch": 0.13311288140688413,
      "flos": 22340889999360.0,
      "grad_norm": 1.7875951714101368,
      "language_loss": 0.79433095,
      "learning_rate": 3.889501213915291e-06,
      "loss": 0.81597471,
      "num_input_tokens_seen": 47906600,
      "step": 2214,
      "time_per_iteration": 2.612903594970703
    },
    {
      "auxiliary_loss_clip": 0.01137669,
      "auxiliary_loss_mlp": 0.01051392,
      "balance_loss_clip": 1.05491984,
      "balance_loss_mlp": 1.03166258,
      "epoch": 0.1331730046595521,
      "flos": 31868888682240.0,
      "grad_norm": 1.6430102214197357,
      "language_loss": 0.69154382,
      "learning_rate": 3.889373516442597e-06,
      "loss": 0.71343446,
      "num_input_tokens_seen": 47927630,
      "step": 2215,
      "time_per_iteration": 2.613560199737549
    },
    {
      "auxiliary_loss_clip": 0.01154263,
      "auxiliary_loss_mlp": 0.01043917,
      "balance_loss_clip": 1.05273855,
      "balance_loss_mlp": 1.02456939,
      "epoch": 0.13323312791222006,
      "flos": 22566589107840.0,
      "grad_norm": 1.9316770035469533,
      "language_loss": 0.81001699,
      "learning_rate": 3.889245747324671e-06,
      "loss": 0.83199877,
      "num_input_tokens_seen": 47947935,
      "step": 2216,
      "time_per_iteration": 2.543605327606201
    },
    {
      "auxiliary_loss_clip": 0.01148044,
      "auxiliary_loss_mlp": 0.01051757,
      "balance_loss_clip": 1.05174899,
      "balance_loss_mlp": 1.03212333,
      "epoch": 0.13329325116488802,
      "flos": 15085319293440.0,
      "grad_norm": 2.734904696978511,
      "language_loss": 0.87101662,
      "learning_rate": 3.889117906566356e-06,
      "loss": 0.89301455,
      "num_input_tokens_seen": 47965515,
      "step": 2217,
      "time_per_iteration": 2.4840123653411865
    },
    {
      "auxiliary_loss_clip": 0.01135555,
      "auxiliary_loss_mlp": 0.01053901,
      "balance_loss_clip": 1.04998302,
      "balance_loss_mlp": 1.03282475,
      "epoch": 0.133353374417556,
      "flos": 27453671890560.0,
      "grad_norm": 2.9735238678267306,
      "language_loss": 0.73416054,
      "learning_rate": 3.888989994172501e-06,
      "loss": 0.75605512,
      "num_input_tokens_seen": 47985675,
      "step": 2218,
      "time_per_iteration": 2.5783402919769287
    },
    {
      "auxiliary_loss_clip": 0.01118487,
      "auxiliary_loss_mlp": 0.0104656,
      "balance_loss_clip": 1.04812157,
      "balance_loss_mlp": 1.02632976,
      "epoch": 0.13341349767022395,
      "flos": 24094695456000.0,
      "grad_norm": 2.0253535752531118,
      "language_loss": 0.87439203,
      "learning_rate": 3.8888620101479565e-06,
      "loss": 0.89604247,
      "num_input_tokens_seen": 48004985,
      "step": 2219,
      "time_per_iteration": 2.6086602210998535
    },
    {
      "auxiliary_loss_clip": 0.0112659,
      "auxiliary_loss_mlp": 0.01058862,
      "balance_loss_clip": 1.04855645,
      "balance_loss_mlp": 1.04062259,
      "epoch": 0.13347362092289192,
      "flos": 24133335511680.0,
      "grad_norm": 2.023515341012075,
      "language_loss": 0.7716471,
      "learning_rate": 3.888733954497574e-06,
      "loss": 0.79350168,
      "num_input_tokens_seen": 48024965,
      "step": 2220,
      "time_per_iteration": 2.5937328338623047
    },
    {
      "auxiliary_loss_clip": 0.01134386,
      "auxiliary_loss_mlp": 0.01039654,
      "balance_loss_clip": 1.04771125,
      "balance_loss_mlp": 1.0219276,
      "epoch": 0.1335337441755599,
      "flos": 18436538390400.0,
      "grad_norm": 6.662801821267267,
      "language_loss": 0.79203391,
      "learning_rate": 3.888605827226212e-06,
      "loss": 0.81377435,
      "num_input_tokens_seen": 48040890,
      "step": 2221,
      "time_per_iteration": 2.4975669384002686
    },
    {
      "auxiliary_loss_clip": 0.01065102,
      "auxiliary_loss_mlp": 0.01003194,
      "balance_loss_clip": 1.02591014,
      "balance_loss_mlp": 0.99995178,
      "epoch": 0.13359386742822787,
      "flos": 50611997652480.0,
      "grad_norm": 0.9697818298047008,
      "language_loss": 0.68973994,
      "learning_rate": 3.8884776283387275e-06,
      "loss": 0.71042287,
      "num_input_tokens_seen": 48091855,
      "step": 2222,
      "time_per_iteration": 2.9606451988220215
    },
    {
      "auxiliary_loss_clip": 0.0112102,
      "auxiliary_loss_mlp": 0.01045249,
      "balance_loss_clip": 1.05438471,
      "balance_loss_mlp": 1.02770185,
      "epoch": 0.13365399068089584,
      "flos": 22778569221120.0,
      "grad_norm": 1.8345982665629677,
      "language_loss": 0.6713255,
      "learning_rate": 3.888349357839982e-06,
      "loss": 0.69298828,
      "num_input_tokens_seen": 48111350,
      "step": 2223,
      "time_per_iteration": 2.5771777629852295
    },
    {
      "auxiliary_loss_clip": 0.01151218,
      "auxiliary_loss_mlp": 0.01061128,
      "balance_loss_clip": 1.05272293,
      "balance_loss_mlp": 1.04101777,
      "epoch": 0.1337141139335638,
      "flos": 12531603911040.0,
      "grad_norm": 3.3760085059909497,
      "language_loss": 0.83030772,
      "learning_rate": 3.88822101573484e-06,
      "loss": 0.85243118,
      "num_input_tokens_seen": 48129840,
      "step": 2224,
      "time_per_iteration": 2.504643678665161
    },
    {
      "auxiliary_loss_clip": 0.0116237,
      "auxiliary_loss_mlp": 0.01044225,
      "balance_loss_clip": 1.05077386,
      "balance_loss_mlp": 1.02463865,
      "epoch": 0.13377423718623177,
      "flos": 23038957889280.0,
      "grad_norm": 2.5989701310860682,
      "language_loss": 0.66218829,
      "learning_rate": 3.888092602028167e-06,
      "loss": 0.68425429,
      "num_input_tokens_seen": 48149240,
      "step": 2225,
      "time_per_iteration": 2.4821665287017822
    },
    {
      "auxiliary_loss_clip": 0.0114746,
      "auxiliary_loss_mlp": 0.01048514,
      "balance_loss_clip": 1.05018306,
      "balance_loss_mlp": 1.02929759,
      "epoch": 0.13383436043889974,
      "flos": 16216397637120.0,
      "grad_norm": 2.6498775674514436,
      "language_loss": 0.89707607,
      "learning_rate": 3.887964116724835e-06,
      "loss": 0.91903585,
      "num_input_tokens_seen": 48166330,
      "step": 2226,
      "time_per_iteration": 2.4977974891662598
    },
    {
      "auxiliary_loss_clip": 0.01138867,
      "auxiliary_loss_mlp": 0.01054088,
      "balance_loss_clip": 1.05010605,
      "balance_loss_mlp": 1.03507388,
      "epoch": 0.1338944836915677,
      "flos": 24279671520000.0,
      "grad_norm": 2.0737079333817645,
      "language_loss": 0.74059176,
      "learning_rate": 3.887835559829712e-06,
      "loss": 0.76252127,
      "num_input_tokens_seen": 48187600,
      "step": 2227,
      "time_per_iteration": 2.5697526931762695
    },
    {
      "auxiliary_loss_clip": 0.01148276,
      "auxiliary_loss_mlp": 0.01044078,
      "balance_loss_clip": 1.0506593,
      "balance_loss_mlp": 1.02455199,
      "epoch": 0.1339546069442357,
      "flos": 17598742594560.0,
      "grad_norm": 2.4721153153154343,
      "language_loss": 0.85392839,
      "learning_rate": 3.8877069313476764e-06,
      "loss": 0.87585193,
      "num_input_tokens_seen": 48204400,
      "step": 2228,
      "time_per_iteration": 2.470172166824341
    },
    {
      "auxiliary_loss_clip": 0.01131203,
      "auxiliary_loss_mlp": 0.01049779,
      "balance_loss_clip": 1.05337691,
      "balance_loss_mlp": 1.02944207,
      "epoch": 0.13401473019690366,
      "flos": 18990065952000.0,
      "grad_norm": 2.0035308025619023,
      "language_loss": 0.81444836,
      "learning_rate": 3.8875782312836054e-06,
      "loss": 0.83625817,
      "num_input_tokens_seen": 48222180,
      "step": 2229,
      "time_per_iteration": 2.5160717964172363
    },
    {
      "auxiliary_loss_clip": 0.01104894,
      "auxiliary_loss_mlp": 0.01054032,
      "balance_loss_clip": 1.05339384,
      "balance_loss_mlp": 1.03419518,
      "epoch": 0.13407485344957162,
      "flos": 26943812288640.0,
      "grad_norm": 1.9480875743486437,
      "language_loss": 0.74712706,
      "learning_rate": 3.887449459642378e-06,
      "loss": 0.76871634,
      "num_input_tokens_seen": 48243245,
      "step": 2230,
      "time_per_iteration": 2.631298780441284
    },
    {
      "auxiliary_loss_clip": 0.0112011,
      "auxiliary_loss_mlp": 0.01052818,
      "balance_loss_clip": 1.0526402,
      "balance_loss_mlp": 1.03338706,
      "epoch": 0.1341349767022396,
      "flos": 20339373375360.0,
      "grad_norm": 5.388216697466175,
      "language_loss": 0.80247843,
      "learning_rate": 3.8873206164288785e-06,
      "loss": 0.82420766,
      "num_input_tokens_seen": 48262600,
      "step": 2231,
      "time_per_iteration": 2.5515410900115967
    },
    {
      "auxiliary_loss_clip": 0.01108241,
      "auxiliary_loss_mlp": 0.01058047,
      "balance_loss_clip": 1.05312228,
      "balance_loss_mlp": 1.03704262,
      "epoch": 0.13419509995490755,
      "flos": 29862020931840.0,
      "grad_norm": 1.9239776639139006,
      "language_loss": 0.72260904,
      "learning_rate": 3.887191701647992e-06,
      "loss": 0.74427193,
      "num_input_tokens_seen": 48285075,
      "step": 2232,
      "time_per_iteration": 4.059293746948242
    },
    {
      "auxiliary_loss_clip": 0.01121865,
      "auxiliary_loss_mlp": 0.01044097,
      "balance_loss_clip": 1.05296779,
      "balance_loss_mlp": 1.02327108,
      "epoch": 0.13425522320757552,
      "flos": 26942986275840.0,
      "grad_norm": 3.144523999811054,
      "language_loss": 0.65489459,
      "learning_rate": 3.8870627153046066e-06,
      "loss": 0.6765542,
      "num_input_tokens_seen": 48301285,
      "step": 2233,
      "time_per_iteration": 4.025724411010742
    },
    {
      "auxiliary_loss_clip": 0.01161814,
      "auxiliary_loss_mlp": 0.01043739,
      "balance_loss_clip": 1.05109823,
      "balance_loss_mlp": 1.0247606,
      "epoch": 0.1343153464602435,
      "flos": 15777281871360.0,
      "grad_norm": 2.9431583186052763,
      "language_loss": 0.81065691,
      "learning_rate": 3.886933657403615e-06,
      "loss": 0.83271241,
      "num_input_tokens_seen": 48317835,
      "step": 2234,
      "time_per_iteration": 2.4627950191497803
    },
    {
      "auxiliary_loss_clip": 0.0113683,
      "auxiliary_loss_mlp": 0.01048777,
      "balance_loss_clip": 1.05181694,
      "balance_loss_mlp": 1.02946496,
      "epoch": 0.13437546971291148,
      "flos": 24314756129280.0,
      "grad_norm": 4.455221553151614,
      "language_loss": 0.82132041,
      "learning_rate": 3.886804527949909e-06,
      "loss": 0.84317648,
      "num_input_tokens_seen": 48335670,
      "step": 2235,
      "time_per_iteration": 2.5386533737182617
    },
    {
      "auxiliary_loss_clip": 0.01147445,
      "auxiliary_loss_mlp": 0.0104795,
      "balance_loss_clip": 1.05234993,
      "balance_loss_mlp": 1.02694499,
      "epoch": 0.13443559296557944,
      "flos": 26650673395200.0,
      "grad_norm": 1.9119811057177714,
      "language_loss": 0.8639977,
      "learning_rate": 3.8866753269483864e-06,
      "loss": 0.88595164,
      "num_input_tokens_seen": 48357805,
      "step": 2236,
      "time_per_iteration": 2.5484073162078857
    },
    {
      "auxiliary_loss_clip": 0.0116749,
      "auxiliary_loss_mlp": 0.01044766,
      "balance_loss_clip": 1.0559864,
      "balance_loss_mlp": 1.02481079,
      "epoch": 0.1344957162182474,
      "flos": 21796197183360.0,
      "grad_norm": 1.9374236949961903,
      "language_loss": 0.77884907,
      "learning_rate": 3.886546054403946e-06,
      "loss": 0.80097163,
      "num_input_tokens_seen": 48377845,
      "step": 2237,
      "time_per_iteration": 3.8547585010528564
    },
    {
      "auxiliary_loss_clip": 0.01143945,
      "auxiliary_loss_mlp": 0.01046596,
      "balance_loss_clip": 1.05247378,
      "balance_loss_mlp": 1.02535272,
      "epoch": 0.13455583947091537,
      "flos": 19865568049920.0,
      "grad_norm": 2.442783748626539,
      "language_loss": 0.7894069,
      "learning_rate": 3.886416710321491e-06,
      "loss": 0.81131232,
      "num_input_tokens_seen": 48394735,
      "step": 2238,
      "time_per_iteration": 2.513523817062378
    },
    {
      "auxiliary_loss_clip": 0.0114043,
      "auxiliary_loss_mlp": 0.01043693,
      "balance_loss_clip": 1.05284238,
      "balance_loss_mlp": 1.02278376,
      "epoch": 0.13461596272358334,
      "flos": 30846835094400.0,
      "grad_norm": 2.403131163238022,
      "language_loss": 0.67724681,
      "learning_rate": 3.886287294705924e-06,
      "loss": 0.69908798,
      "num_input_tokens_seen": 48414200,
      "step": 2239,
      "time_per_iteration": 2.5896058082580566
    },
    {
      "auxiliary_loss_clip": 0.01148306,
      "auxiliary_loss_mlp": 0.01040335,
      "balance_loss_clip": 1.05783975,
      "balance_loss_mlp": 1.0209161,
      "epoch": 0.1346760859762513,
      "flos": 12494436312960.0,
      "grad_norm": 2.4199869724636653,
      "language_loss": 0.8157593,
      "learning_rate": 3.8861578075621555e-06,
      "loss": 0.83764571,
      "num_input_tokens_seen": 48431065,
      "step": 2240,
      "time_per_iteration": 2.498563766479492
    },
    {
      "auxiliary_loss_clip": 0.01110983,
      "auxiliary_loss_mlp": 0.01049762,
      "balance_loss_clip": 1.05283463,
      "balance_loss_mlp": 1.02866149,
      "epoch": 0.1347362092289193,
      "flos": 21836022387840.0,
      "grad_norm": 1.7659016426780036,
      "language_loss": 0.77705002,
      "learning_rate": 3.886028248895093e-06,
      "loss": 0.79865748,
      "num_input_tokens_seen": 48450335,
      "step": 2241,
      "time_per_iteration": 3.976215124130249
    },
    {
      "auxiliary_loss_clip": 0.01162106,
      "auxiliary_loss_mlp": 0.01037999,
      "balance_loss_clip": 1.05563521,
      "balance_loss_mlp": 1.02053452,
      "epoch": 0.13479633248158726,
      "flos": 23509459163520.0,
      "grad_norm": 1.9375682740849078,
      "language_loss": 0.83558267,
      "learning_rate": 3.88589861870965e-06,
      "loss": 0.8575837,
      "num_input_tokens_seen": 48468555,
      "step": 2242,
      "time_per_iteration": 2.509639263153076
    },
    {
      "auxiliary_loss_clip": 0.0116496,
      "auxiliary_loss_mlp": 0.01051938,
      "balance_loss_clip": 1.05413365,
      "balance_loss_mlp": 1.03084993,
      "epoch": 0.13485645573425523,
      "flos": 29344332165120.0,
      "grad_norm": 3.316340901072225,
      "language_loss": 0.64828765,
      "learning_rate": 3.885768917010744e-06,
      "loss": 0.67045659,
      "num_input_tokens_seen": 48488515,
      "step": 2243,
      "time_per_iteration": 2.5329201221466064
    },
    {
      "auxiliary_loss_clip": 0.01124521,
      "auxiliary_loss_mlp": 0.01044089,
      "balance_loss_clip": 1.04959655,
      "balance_loss_mlp": 1.02416933,
      "epoch": 0.1349165789869232,
      "flos": 28037112503040.0,
      "grad_norm": 1.834418701429945,
      "language_loss": 0.72754025,
      "learning_rate": 3.8856391438032895e-06,
      "loss": 0.74922633,
      "num_input_tokens_seen": 48510515,
      "step": 2244,
      "time_per_iteration": 2.591916084289551
    },
    {
      "auxiliary_loss_clip": 0.01150379,
      "auxiliary_loss_mlp": 0.01046797,
      "balance_loss_clip": 1.05388165,
      "balance_loss_mlp": 1.02922583,
      "epoch": 0.13497670223959116,
      "flos": 22853730430080.0,
      "grad_norm": 1.6825866540382546,
      "language_loss": 0.86239713,
      "learning_rate": 3.88550929909221e-06,
      "loss": 0.8843689,
      "num_input_tokens_seen": 48529940,
      "step": 2245,
      "time_per_iteration": 2.5146827697753906
    },
    {
      "auxiliary_loss_clip": 0.0114862,
      "auxiliary_loss_mlp": 0.01046919,
      "balance_loss_clip": 1.05235291,
      "balance_loss_mlp": 1.02806032,
      "epoch": 0.13503682549225912,
      "flos": 16504580453760.0,
      "grad_norm": 1.8443911738379126,
      "language_loss": 0.78833622,
      "learning_rate": 3.88537938288243e-06,
      "loss": 0.81029165,
      "num_input_tokens_seen": 48548190,
      "step": 2246,
      "time_per_iteration": 2.5010430812835693
    },
    {
      "auxiliary_loss_clip": 0.01034574,
      "auxiliary_loss_mlp": 0.01039933,
      "balance_loss_clip": 1.03767061,
      "balance_loss_mlp": 1.03590381,
      "epoch": 0.1350969487449271,
      "flos": 70756303242240.0,
      "grad_norm": 0.7614789262469613,
      "language_loss": 0.60618865,
      "learning_rate": 3.885249395178874e-06,
      "loss": 0.62693375,
      "num_input_tokens_seen": 48613165,
      "step": 2247,
      "time_per_iteration": 3.318683624267578
    },
    {
      "auxiliary_loss_clip": 0.01161946,
      "auxiliary_loss_mlp": 0.01052983,
      "balance_loss_clip": 1.05788255,
      "balance_loss_mlp": 1.03111982,
      "epoch": 0.13515707199759508,
      "flos": 23075981832960.0,
      "grad_norm": 2.4976555129548865,
      "language_loss": 0.81075227,
      "learning_rate": 3.885119335986473e-06,
      "loss": 0.83290148,
      "num_input_tokens_seen": 48631705,
      "step": 2248,
      "time_per_iteration": 2.5000176429748535
    },
    {
      "auxiliary_loss_clip": 0.01139548,
      "auxiliary_loss_mlp": 0.0104347,
      "balance_loss_clip": 1.0531553,
      "balance_loss_mlp": 1.02530289,
      "epoch": 0.13521719525026304,
      "flos": 23186371305600.0,
      "grad_norm": 1.931658032844801,
      "language_loss": 0.7689631,
      "learning_rate": 3.884989205310157e-06,
      "loss": 0.7907933,
      "num_input_tokens_seen": 48649740,
      "step": 2249,
      "time_per_iteration": 2.567361354827881
    },
    {
      "auxiliary_loss_clip": 0.01123101,
      "auxiliary_loss_mlp": 0.01051764,
      "balance_loss_clip": 1.05355668,
      "balance_loss_mlp": 1.03345311,
      "epoch": 0.135277318502931,
      "flos": 24790931752320.0,
      "grad_norm": 1.4971933233877974,
      "language_loss": 0.84499663,
      "learning_rate": 3.884859003154862e-06,
      "loss": 0.86674529,
      "num_input_tokens_seen": 48671565,
      "step": 2250,
      "time_per_iteration": 2.6351797580718994
    },
    {
      "auxiliary_loss_clip": 0.01153036,
      "auxiliary_loss_mlp": 0.01046047,
      "balance_loss_clip": 1.05263269,
      "balance_loss_mlp": 1.02565026,
      "epoch": 0.13533744175559898,
      "flos": 21908525990400.0,
      "grad_norm": 2.417404150404338,
      "language_loss": 0.8200022,
      "learning_rate": 3.884728729525524e-06,
      "loss": 0.84199297,
      "num_input_tokens_seen": 48690425,
      "step": 2251,
      "time_per_iteration": 2.5219478607177734
    },
    {
      "auxiliary_loss_clip": 0.01162721,
      "auxiliary_loss_mlp": 0.0104685,
      "balance_loss_clip": 1.05275881,
      "balance_loss_mlp": 1.02567828,
      "epoch": 0.13539756500826694,
      "flos": 21211643249280.0,
      "grad_norm": 1.7284399293995605,
      "language_loss": 0.85800153,
      "learning_rate": 3.884598384427084e-06,
      "loss": 0.88009727,
      "num_input_tokens_seen": 48707505,
      "step": 2252,
      "time_per_iteration": 2.466893196105957
    },
    {
      "auxiliary_loss_clip": 0.01076283,
      "auxiliary_loss_mlp": 0.01002599,
      "balance_loss_clip": 1.0430131,
      "balance_loss_mlp": 0.99928534,
      "epoch": 0.1354576882609349,
      "flos": 63242103634560.0,
      "grad_norm": 0.7668495589241298,
      "language_loss": 0.61797643,
      "learning_rate": 3.884467967864485e-06,
      "loss": 0.63876522,
      "num_input_tokens_seen": 48775895,
      "step": 2253,
      "time_per_iteration": 3.204958200454712
    },
    {
      "auxiliary_loss_clip": 0.0115278,
      "auxiliary_loss_mlp": 0.01053616,
      "balance_loss_clip": 1.05585551,
      "balance_loss_mlp": 1.03487659,
      "epoch": 0.1355178115136029,
      "flos": 25483037984640.0,
      "grad_norm": 1.626941027465393,
      "language_loss": 0.89146578,
      "learning_rate": 3.884337479842671e-06,
      "loss": 0.91352975,
      "num_input_tokens_seen": 48798370,
      "step": 2254,
      "time_per_iteration": 2.568239688873291
    },
    {
      "auxiliary_loss_clip": 0.01132267,
      "auxiliary_loss_mlp": 0.0105787,
      "balance_loss_clip": 1.04951119,
      "balance_loss_mlp": 1.03401589,
      "epoch": 0.13557793476627086,
      "flos": 21616967295360.0,
      "grad_norm": 4.340635432900999,
      "language_loss": 0.84370488,
      "learning_rate": 3.884206920366591e-06,
      "loss": 0.86560631,
      "num_input_tokens_seen": 48817955,
      "step": 2255,
      "time_per_iteration": 2.53114652633667
    },
    {
      "auxiliary_loss_clip": 0.01162188,
      "auxiliary_loss_mlp": 0.01048614,
      "balance_loss_clip": 1.05252182,
      "balance_loss_mlp": 1.02911162,
      "epoch": 0.13563805801893883,
      "flos": 24928253447040.0,
      "grad_norm": 3.006749689354804,
      "language_loss": 0.75487387,
      "learning_rate": 3.884076289441196e-06,
      "loss": 0.77698183,
      "num_input_tokens_seen": 48836330,
      "step": 2256,
      "time_per_iteration": 2.4890904426574707
    },
    {
      "auxiliary_loss_clip": 0.01119789,
      "auxiliary_loss_mlp": 0.01047574,
      "balance_loss_clip": 1.04892993,
      "balance_loss_mlp": 1.02675962,
      "epoch": 0.1356981812716068,
      "flos": 14750272206720.0,
      "grad_norm": 2.1474584147154965,
      "language_loss": 0.83619034,
      "learning_rate": 3.88394558707144e-06,
      "loss": 0.85786396,
      "num_input_tokens_seen": 48851890,
      "step": 2257,
      "time_per_iteration": 2.5046608448028564
    },
    {
      "auxiliary_loss_clip": 0.01144622,
      "auxiliary_loss_mlp": 0.01136655,
      "balance_loss_clip": 1.05074501,
      "balance_loss_mlp": 1.59741688,
      "epoch": 0.13575830452427476,
      "flos": 11108571822720.0,
      "grad_norm": 2.964318462281257,
      "language_loss": 0.82139659,
      "learning_rate": 3.883814813262277e-06,
      "loss": 0.84420931,
      "num_input_tokens_seen": 48865510,
      "step": 2258,
      "time_per_iteration": 2.514435291290283
    },
    {
      "auxiliary_loss_clip": 0.01151659,
      "auxiliary_loss_mlp": 0.01049975,
      "balance_loss_clip": 1.04992247,
      "balance_loss_mlp": 1.02764678,
      "epoch": 0.13581842777694272,
      "flos": 17960290940160.0,
      "grad_norm": 2.271053524766643,
      "language_loss": 0.82505202,
      "learning_rate": 3.883683968018669e-06,
      "loss": 0.84706837,
      "num_input_tokens_seen": 48882360,
      "step": 2259,
      "time_per_iteration": 2.4597957134246826
    },
    {
      "auxiliary_loss_clip": 0.01121454,
      "auxiliary_loss_mlp": 0.01054854,
      "balance_loss_clip": 1.04950213,
      "balance_loss_mlp": 1.03634083,
      "epoch": 0.1358785510296107,
      "flos": 22857142222080.0,
      "grad_norm": 2.07385119066564,
      "language_loss": 0.73068082,
      "learning_rate": 3.8835530513455755e-06,
      "loss": 0.75244391,
      "num_input_tokens_seen": 48902700,
      "step": 2260,
      "time_per_iteration": 2.5644311904907227
    },
    {
      "auxiliary_loss_clip": 0.01136024,
      "auxiliary_loss_mlp": 0.0105618,
      "balance_loss_clip": 1.05080497,
      "balance_loss_mlp": 1.03612888,
      "epoch": 0.13593867428227868,
      "flos": 25739404329600.0,
      "grad_norm": 2.714869700607332,
      "language_loss": 0.75204843,
      "learning_rate": 3.883422063247961e-06,
      "loss": 0.77397048,
      "num_input_tokens_seen": 48922525,
      "step": 2261,
      "time_per_iteration": 2.54692006111145
    },
    {
      "auxiliary_loss_clip": 0.01161226,
      "auxiliary_loss_mlp": 0.01047018,
      "balance_loss_clip": 1.04963684,
      "balance_loss_mlp": 1.02776551,
      "epoch": 0.13599879753494665,
      "flos": 31249214225280.0,
      "grad_norm": 1.8550541441133357,
      "language_loss": 0.6327132,
      "learning_rate": 3.883291003730794e-06,
      "loss": 0.65479565,
      "num_input_tokens_seen": 48942510,
      "step": 2262,
      "time_per_iteration": 2.565825939178467
    },
    {
      "auxiliary_loss_clip": 0.01139478,
      "auxiliary_loss_mlp": 0.01044731,
      "balance_loss_clip": 1.04828572,
      "balance_loss_mlp": 1.02507377,
      "epoch": 0.1360589207876146,
      "flos": 23915034604800.0,
      "grad_norm": 2.3077461344029704,
      "language_loss": 0.82582843,
      "learning_rate": 3.883159872799043e-06,
      "loss": 0.84767056,
      "num_input_tokens_seen": 48962625,
      "step": 2263,
      "time_per_iteration": 2.5631628036499023
    },
    {
      "auxiliary_loss_clip": 0.01100338,
      "auxiliary_loss_mlp": 0.01061882,
      "balance_loss_clip": 1.05430818,
      "balance_loss_mlp": 1.03943503,
      "epoch": 0.13611904404028258,
      "flos": 19974197756160.0,
      "grad_norm": 1.9552005827508308,
      "language_loss": 0.8795957,
      "learning_rate": 3.8830286704576815e-06,
      "loss": 0.90121794,
      "num_input_tokens_seen": 48982525,
      "step": 2264,
      "time_per_iteration": 2.6664767265319824
    },
    {
      "auxiliary_loss_clip": 0.01155279,
      "auxiliary_loss_mlp": 0.0104504,
      "balance_loss_clip": 1.05371094,
      "balance_loss_mlp": 1.02339137,
      "epoch": 0.13617916729295054,
      "flos": 15340644144000.0,
      "grad_norm": 3.186987629311333,
      "language_loss": 0.71239662,
      "learning_rate": 3.882897396711683e-06,
      "loss": 0.7343998,
      "num_input_tokens_seen": 48997605,
      "step": 2265,
      "time_per_iteration": 2.464078903198242
    },
    {
      "auxiliary_loss_clip": 0.01106619,
      "auxiliary_loss_mlp": 0.01049545,
      "balance_loss_clip": 1.05154991,
      "balance_loss_mlp": 1.02983952,
      "epoch": 0.1362392905456185,
      "flos": 27451445247360.0,
      "grad_norm": 2.039058178950468,
      "language_loss": 0.66550922,
      "learning_rate": 3.882766051566027e-06,
      "loss": 0.68707091,
      "num_input_tokens_seen": 49018535,
      "step": 2266,
      "time_per_iteration": 2.681840419769287
    },
    {
      "auxiliary_loss_clip": 0.01121816,
      "auxiliary_loss_mlp": 0.01055946,
      "balance_loss_clip": 1.05363643,
      "balance_loss_mlp": 1.03633547,
      "epoch": 0.1362994137982865,
      "flos": 25009017177600.0,
      "grad_norm": 2.168834230618332,
      "language_loss": 0.76659155,
      "learning_rate": 3.882634635025694e-06,
      "loss": 0.78836912,
      "num_input_tokens_seen": 49038865,
      "step": 2267,
      "time_per_iteration": 2.6286492347717285
    },
    {
      "auxiliary_loss_clip": 0.01133054,
      "auxiliary_loss_mlp": 0.01049732,
      "balance_loss_clip": 1.05271792,
      "balance_loss_mlp": 1.02883458,
      "epoch": 0.13635953705095447,
      "flos": 20303031790080.0,
      "grad_norm": 1.9708520209954268,
      "language_loss": 0.82005966,
      "learning_rate": 3.882503147095667e-06,
      "loss": 0.84188753,
      "num_input_tokens_seen": 49058010,
      "step": 2268,
      "time_per_iteration": 2.5393872261047363
    },
    {
      "auxiliary_loss_clip": 0.01149409,
      "auxiliary_loss_mlp": 0.01042817,
      "balance_loss_clip": 1.05357718,
      "balance_loss_mlp": 1.0231117,
      "epoch": 0.13641966030362243,
      "flos": 31358418549120.0,
      "grad_norm": 1.78313009344237,
      "language_loss": 0.76058733,
      "learning_rate": 3.882371587780931e-06,
      "loss": 0.78250957,
      "num_input_tokens_seen": 49080330,
      "step": 2269,
      "time_per_iteration": 2.569965124130249
    },
    {
      "auxiliary_loss_clip": 0.01128153,
      "auxiliary_loss_mlp": 0.01044932,
      "balance_loss_clip": 1.04854822,
      "balance_loss_mlp": 1.02463078,
      "epoch": 0.1364797835562904,
      "flos": 20478095700480.0,
      "grad_norm": 2.108939899666033,
      "language_loss": 0.80979359,
      "learning_rate": 3.882239957086477e-06,
      "loss": 0.83152449,
      "num_input_tokens_seen": 49097035,
      "step": 2270,
      "time_per_iteration": 2.5515968799591064
    },
    {
      "auxiliary_loss_clip": 0.01139119,
      "auxiliary_loss_mlp": 0.01059139,
      "balance_loss_clip": 1.05348706,
      "balance_loss_mlp": 1.03822947,
      "epoch": 0.13653990680895836,
      "flos": 13078343802240.0,
      "grad_norm": 2.424910639243959,
      "language_loss": 0.7549057,
      "learning_rate": 3.882108255017295e-06,
      "loss": 0.77688825,
      "num_input_tokens_seen": 49113945,
      "step": 2271,
      "time_per_iteration": 3.8846676349639893
    },
    {
      "auxiliary_loss_clip": 0.011509,
      "auxiliary_loss_mlp": 0.01053547,
      "balance_loss_clip": 1.04985595,
      "balance_loss_mlp": 1.03287601,
      "epoch": 0.13660003006162633,
      "flos": 16946712961920.0,
      "grad_norm": 1.9492684807401375,
      "language_loss": 0.80239272,
      "learning_rate": 3.881976481578379e-06,
      "loss": 0.82443714,
      "num_input_tokens_seen": 49132855,
      "step": 2272,
      "time_per_iteration": 3.9356212615966797
    },
    {
      "auxiliary_loss_clip": 0.01072031,
      "auxiliary_loss_mlp": 0.01005953,
      "balance_loss_clip": 1.03757954,
      "balance_loss_mlp": 1.00261521,
      "epoch": 0.1366601533142943,
      "flos": 68682749892480.0,
      "grad_norm": 0.6955958389701359,
      "language_loss": 0.60689259,
      "learning_rate": 3.8818446367747255e-06,
      "loss": 0.62767243,
      "num_input_tokens_seen": 49198310,
      "step": 2273,
      "time_per_iteration": 3.2053537368774414
    },
    {
      "auxiliary_loss_clip": 0.0115854,
      "auxiliary_loss_mlp": 0.01006597,
      "balance_loss_clip": 1.05025077,
      "balance_loss_mlp": 1.34720278,
      "epoch": 0.13672027656696228,
      "flos": 19244241567360.0,
      "grad_norm": 1.9366855426528444,
      "language_loss": 0.78017938,
      "learning_rate": 3.881712720611336e-06,
      "loss": 0.80183077,
      "num_input_tokens_seen": 49217250,
      "step": 2274,
      "time_per_iteration": 2.456404209136963
    },
    {
      "auxiliary_loss_clip": 0.01148637,
      "auxiliary_loss_mlp": 0.01047786,
      "balance_loss_clip": 1.0497421,
      "balance_loss_mlp": 1.02690029,
      "epoch": 0.13678039981963025,
      "flos": 24534924543360.0,
      "grad_norm": 1.729096842189411,
      "language_loss": 0.78384799,
      "learning_rate": 3.881580733093211e-06,
      "loss": 0.80581224,
      "num_input_tokens_seen": 49236615,
      "step": 2275,
      "time_per_iteration": 2.52522873878479
    },
    {
      "auxiliary_loss_clip": 0.01148427,
      "auxiliary_loss_mlp": 0.01040375,
      "balance_loss_clip": 1.05041981,
      "balance_loss_mlp": 1.02118206,
      "epoch": 0.13684052307229821,
      "flos": 15669334523520.0,
      "grad_norm": 2.9095049429185926,
      "language_loss": 0.81727517,
      "learning_rate": 3.881448674225356e-06,
      "loss": 0.83916318,
      "num_input_tokens_seen": 49253935,
      "step": 2276,
      "time_per_iteration": 3.882133960723877
    },
    {
      "auxiliary_loss_clip": 0.0115809,
      "auxiliary_loss_mlp": 0.01052592,
      "balance_loss_clip": 1.05173635,
      "balance_loss_mlp": 1.0293467,
      "epoch": 0.13690064632496618,
      "flos": 28364689560960.0,
      "grad_norm": 3.4740559928285606,
      "language_loss": 0.69576287,
      "learning_rate": 3.881316544012779e-06,
      "loss": 0.7178697,
      "num_input_tokens_seen": 49273605,
      "step": 2277,
      "time_per_iteration": 2.5422699451446533
    },
    {
      "auxiliary_loss_clip": 0.01153221,
      "auxiliary_loss_mlp": 0.00991125,
      "balance_loss_clip": 1.05164742,
      "balance_loss_mlp": 1.32036018,
      "epoch": 0.13696076957763414,
      "flos": 23404779953280.0,
      "grad_norm": 2.724646474410736,
      "language_loss": 0.8041321,
      "learning_rate": 3.88118434246049e-06,
      "loss": 0.82557553,
      "num_input_tokens_seen": 49291785,
      "step": 2278,
      "time_per_iteration": 2.5324742794036865
    },
    {
      "auxiliary_loss_clip": 0.01150416,
      "auxiliary_loss_mlp": 0.01055241,
      "balance_loss_clip": 1.05662704,
      "balance_loss_mlp": 1.03504694,
      "epoch": 0.1370208928303021,
      "flos": 37196595601920.0,
      "grad_norm": 3.683030160081133,
      "language_loss": 0.74840397,
      "learning_rate": 3.881052069573502e-06,
      "loss": 0.77046055,
      "num_input_tokens_seen": 49311405,
      "step": 2279,
      "time_per_iteration": 4.1052680015563965
    },
    {
      "auxiliary_loss_clip": 0.01096454,
      "auxiliary_loss_mlp": 0.01050632,
      "balance_loss_clip": 1.05040097,
      "balance_loss_mlp": 1.03006852,
      "epoch": 0.13708101608297008,
      "flos": 26976311118720.0,
      "grad_norm": 1.8697763356274355,
      "language_loss": 0.76655257,
      "learning_rate": 3.880919725356831e-06,
      "loss": 0.78802341,
      "num_input_tokens_seen": 49331835,
      "step": 2280,
      "time_per_iteration": 2.6671535968780518
    },
    {
      "auxiliary_loss_clip": 0.01102642,
      "auxiliary_loss_mlp": 0.01048217,
      "balance_loss_clip": 1.04763889,
      "balance_loss_mlp": 1.02877355,
      "epoch": 0.13714113933563807,
      "flos": 32556864850560.0,
      "grad_norm": 2.8537288063464477,
      "language_loss": 0.80241853,
      "learning_rate": 3.880787309815496e-06,
      "loss": 0.82392704,
      "num_input_tokens_seen": 49352290,
      "step": 2281,
      "time_per_iteration": 2.70766282081604
    },
    {
      "auxiliary_loss_clip": 0.01165081,
      "auxiliary_loss_mlp": 0.01057028,
      "balance_loss_clip": 1.05192637,
      "balance_loss_mlp": 1.03733516,
      "epoch": 0.13720126258830603,
      "flos": 16101267569280.0,
      "grad_norm": 1.9158884362633664,
      "language_loss": 0.83719778,
      "learning_rate": 3.880654822954518e-06,
      "loss": 0.85941887,
      "num_input_tokens_seen": 49370285,
      "step": 2282,
      "time_per_iteration": 2.4461870193481445
    },
    {
      "auxiliary_loss_clip": 0.01136096,
      "auxiliary_loss_mlp": 0.01053781,
      "balance_loss_clip": 1.04872906,
      "balance_loss_mlp": 1.03488588,
      "epoch": 0.137261385840974,
      "flos": 18953544798720.0,
      "grad_norm": 1.7676907249816516,
      "language_loss": 0.73718059,
      "learning_rate": 3.8805222647789195e-06,
      "loss": 0.75907934,
      "num_input_tokens_seen": 49389610,
      "step": 2283,
      "time_per_iteration": 2.522402286529541
    },
    {
      "auxiliary_loss_clip": 0.01148974,
      "auxiliary_loss_mlp": 0.01049373,
      "balance_loss_clip": 1.05155337,
      "balance_loss_mlp": 1.03089523,
      "epoch": 0.13732150909364196,
      "flos": 23295360147840.0,
      "grad_norm": 3.1731879068584425,
      "language_loss": 0.84101921,
      "learning_rate": 3.880389635293729e-06,
      "loss": 0.86300272,
      "num_input_tokens_seen": 49408390,
      "step": 2284,
      "time_per_iteration": 2.51814603805542
    },
    {
      "auxiliary_loss_clip": 0.01139919,
      "auxiliary_loss_mlp": 0.01049953,
      "balance_loss_clip": 1.04867959,
      "balance_loss_mlp": 1.02907896,
      "epoch": 0.13738163234630993,
      "flos": 29351263489920.0,
      "grad_norm": 1.914265745262067,
      "language_loss": 0.75035793,
      "learning_rate": 3.880256934503974e-06,
      "loss": 0.77225661,
      "num_input_tokens_seen": 49427725,
      "step": 2285,
      "time_per_iteration": 2.587636947631836
    },
    {
      "auxiliary_loss_clip": 0.01136329,
      "auxiliary_loss_mlp": 0.01047484,
      "balance_loss_clip": 1.04937112,
      "balance_loss_mlp": 1.02820754,
      "epoch": 0.1374417555989779,
      "flos": 26651319840000.0,
      "grad_norm": 2.02130645464963,
      "language_loss": 0.74751627,
      "learning_rate": 3.880124162414689e-06,
      "loss": 0.76935434,
      "num_input_tokens_seen": 49449000,
      "step": 2286,
      "time_per_iteration": 2.5668067932128906
    },
    {
      "auxiliary_loss_clip": 0.01118516,
      "auxiliary_loss_mlp": 0.01043843,
      "balance_loss_clip": 1.05126083,
      "balance_loss_mlp": 1.02292132,
      "epoch": 0.1375018788516459,
      "flos": 28403401443840.0,
      "grad_norm": 2.715533927091294,
      "language_loss": 0.86311233,
      "learning_rate": 3.879991319030908e-06,
      "loss": 0.88473594,
      "num_input_tokens_seen": 49468360,
      "step": 2287,
      "time_per_iteration": 2.6463422775268555
    },
    {
      "auxiliary_loss_clip": 0.01115225,
      "auxiliary_loss_mlp": 0.01050458,
      "balance_loss_clip": 1.04635429,
      "balance_loss_mlp": 1.02933407,
      "epoch": 0.13756200210431385,
      "flos": 37413783187200.0,
      "grad_norm": 1.8465752684015717,
      "language_loss": 0.68656731,
      "learning_rate": 3.879858404357666e-06,
      "loss": 0.70822418,
      "num_input_tokens_seen": 49493450,
      "step": 2288,
      "time_per_iteration": 2.6856462955474854
    },
    {
      "auxiliary_loss_clip": 0.01103213,
      "auxiliary_loss_mlp": 0.0106579,
      "balance_loss_clip": 1.0493319,
      "balance_loss_mlp": 1.04440439,
      "epoch": 0.13762212535698182,
      "flos": 22711021695360.0,
      "grad_norm": 2.4099122478449,
      "language_loss": 0.86997288,
      "learning_rate": 3.879725418400005e-06,
      "loss": 0.89166296,
      "num_input_tokens_seen": 49511220,
      "step": 2289,
      "time_per_iteration": 2.6050283908843994
    },
    {
      "auxiliary_loss_clip": 0.01123599,
      "auxiliary_loss_mlp": 0.00966133,
      "balance_loss_clip": 1.04715586,
      "balance_loss_mlp": 1.27487338,
      "epoch": 0.13768224860964978,
      "flos": 23952130375680.0,
      "grad_norm": 2.0304873146220057,
      "language_loss": 0.74671936,
      "learning_rate": 3.879592361162969e-06,
      "loss": 0.76761663,
      "num_input_tokens_seen": 49529820,
      "step": 2290,
      "time_per_iteration": 2.554211378097534
    },
    {
      "auxiliary_loss_clip": 0.01068571,
      "auxiliary_loss_mlp": 0.0101112,
      "balance_loss_clip": 1.05483317,
      "balance_loss_mlp": 1.00823462,
      "epoch": 0.13774237186231775,
      "flos": 63590438753280.0,
      "grad_norm": 0.7070347179497773,
      "language_loss": 0.51645386,
      "learning_rate": 3.8794592326516015e-06,
      "loss": 0.53725082,
      "num_input_tokens_seen": 49595325,
      "step": 2291,
      "time_per_iteration": 3.2365164756774902
    },
    {
      "auxiliary_loss_clip": 0.01149434,
      "auxiliary_loss_mlp": 0.01044375,
      "balance_loss_clip": 1.04894972,
      "balance_loss_mlp": 1.02389455,
      "epoch": 0.1378024951149857,
      "flos": 24279456038400.0,
      "grad_norm": 2.0970545367447504,
      "language_loss": 0.71563935,
      "learning_rate": 3.879326032870952e-06,
      "loss": 0.73757744,
      "num_input_tokens_seen": 49615850,
      "step": 2292,
      "time_per_iteration": 2.5200703144073486
    },
    {
      "auxiliary_loss_clip": 0.01146216,
      "auxiliary_loss_mlp": 0.01042112,
      "balance_loss_clip": 1.04830635,
      "balance_loss_mlp": 1.02290738,
      "epoch": 0.13786261836765368,
      "flos": 14021537080320.0,
      "grad_norm": 3.044581848004234,
      "language_loss": 0.79861236,
      "learning_rate": 3.879192761826071e-06,
      "loss": 0.82049572,
      "num_input_tokens_seen": 49631860,
      "step": 2293,
      "time_per_iteration": 2.4658217430114746
    },
    {
      "auxiliary_loss_clip": 0.0114433,
      "auxiliary_loss_mlp": 0.01046829,
      "balance_loss_clip": 1.04671526,
      "balance_loss_mlp": 1.02745759,
      "epoch": 0.13792274162032167,
      "flos": 28878679226880.0,
      "grad_norm": 2.190518617487034,
      "language_loss": 0.78096801,
      "learning_rate": 3.879059419522011e-06,
      "loss": 0.80287957,
      "num_input_tokens_seen": 49652145,
      "step": 2294,
      "time_per_iteration": 2.551774501800537
    },
    {
      "auxiliary_loss_clip": 0.01115299,
      "auxiliary_loss_mlp": 0.01043677,
      "balance_loss_clip": 1.04824758,
      "balance_loss_mlp": 1.02608132,
      "epoch": 0.13798286487298964,
      "flos": 21141150808320.0,
      "grad_norm": 2.3300075200369426,
      "language_loss": 0.80142468,
      "learning_rate": 3.878926005963831e-06,
      "loss": 0.82301444,
      "num_input_tokens_seen": 49669880,
      "step": 2295,
      "time_per_iteration": 2.552255630493164
    },
    {
      "auxiliary_loss_clip": 0.01142868,
      "auxiliary_loss_mlp": 0.0104394,
      "balance_loss_clip": 1.04779553,
      "balance_loss_mlp": 1.02458072,
      "epoch": 0.1380429881256576,
      "flos": 22487477402880.0,
      "grad_norm": 1.905518745913337,
      "language_loss": 0.78820127,
      "learning_rate": 3.878792521156588e-06,
      "loss": 0.81006938,
      "num_input_tokens_seen": 49687255,
      "step": 2296,
      "time_per_iteration": 2.4935050010681152
    },
    {
      "auxiliary_loss_clip": 0.01145252,
      "auxiliary_loss_mlp": 0.0105468,
      "balance_loss_clip": 1.05113292,
      "balance_loss_mlp": 1.03433061,
      "epoch": 0.13810311137832557,
      "flos": 21393674398080.0,
      "grad_norm": 2.141943798818568,
      "language_loss": 0.78492808,
      "learning_rate": 3.8786589651053446e-06,
      "loss": 0.80692744,
      "num_input_tokens_seen": 49706650,
      "step": 2297,
      "time_per_iteration": 2.5190489292144775
    },
    {
      "auxiliary_loss_clip": 0.01102042,
      "auxiliary_loss_mlp": 0.01047479,
      "balance_loss_clip": 1.05065906,
      "balance_loss_mlp": 1.02852464,
      "epoch": 0.13816323463099353,
      "flos": 25989844930560.0,
      "grad_norm": 2.5837008613617636,
      "language_loss": 0.68829018,
      "learning_rate": 3.878525337815164e-06,
      "loss": 0.70978534,
      "num_input_tokens_seen": 49725715,
      "step": 2298,
      "time_per_iteration": 2.6319966316223145
    },
    {
      "auxiliary_loss_clip": 0.01127883,
      "auxiliary_loss_mlp": 0.01046483,
      "balance_loss_clip": 1.04813814,
      "balance_loss_mlp": 1.02681375,
      "epoch": 0.1382233578836615,
      "flos": 19244313394560.0,
      "grad_norm": 2.8443869097795424,
      "language_loss": 0.86913908,
      "learning_rate": 3.878391639291116e-06,
      "loss": 0.89088279,
      "num_input_tokens_seen": 49744710,
      "step": 2299,
      "time_per_iteration": 2.5411484241485596
    },
    {
      "auxiliary_loss_clip": 0.01154852,
      "auxiliary_loss_mlp": 0.01048135,
      "balance_loss_clip": 1.04835176,
      "balance_loss_mlp": 1.02823901,
      "epoch": 0.1382834811363295,
      "flos": 25666290195840.0,
      "grad_norm": 2.248743651878821,
      "language_loss": 0.75596237,
      "learning_rate": 3.878257869538267e-06,
      "loss": 0.77799225,
      "num_input_tokens_seen": 49764300,
      "step": 2300,
      "time_per_iteration": 2.494223117828369
    },
    {
      "auxiliary_loss_clip": 0.01122992,
      "auxiliary_loss_mlp": 0.01040171,
      "balance_loss_clip": 1.05078089,
      "balance_loss_mlp": 1.02152634,
      "epoch": 0.13834360438899745,
      "flos": 19784193788160.0,
      "grad_norm": 2.3585778129568444,
      "language_loss": 0.82362247,
      "learning_rate": 3.878124028561692e-06,
      "loss": 0.845254,
      "num_input_tokens_seen": 49778380,
      "step": 2301,
      "time_per_iteration": 2.5430617332458496
    },
    {
      "auxiliary_loss_clip": 0.01126895,
      "auxiliary_loss_mlp": 0.00955128,
      "balance_loss_clip": 1.0474999,
      "balance_loss_mlp": 1.26831317,
      "epoch": 0.13840372764166542,
      "flos": 26651858544000.0,
      "grad_norm": 5.303328922028776,
      "language_loss": 0.85784739,
      "learning_rate": 3.877990116366466e-06,
      "loss": 0.87866771,
      "num_input_tokens_seen": 49797460,
      "step": 2302,
      "time_per_iteration": 2.5766992568969727
    },
    {
      "auxiliary_loss_clip": 0.01071722,
      "auxiliary_loss_mlp": 0.01033032,
      "balance_loss_clip": 1.04001153,
      "balance_loss_mlp": 1.03040981,
      "epoch": 0.13846385089433338,
      "flos": 70510998286080.0,
      "grad_norm": 0.7569729372277287,
      "language_loss": 0.65610898,
      "learning_rate": 3.877856132957667e-06,
      "loss": 0.67715645,
      "num_input_tokens_seen": 49868005,
      "step": 2303,
      "time_per_iteration": 3.2609856128692627
    },
    {
      "auxiliary_loss_clip": 0.01140572,
      "auxiliary_loss_mlp": 0.01037554,
      "balance_loss_clip": 1.04766881,
      "balance_loss_mlp": 1.01987529,
      "epoch": 0.13852397414700135,
      "flos": 17348732956800.0,
      "grad_norm": 1.9846981662904268,
      "language_loss": 0.78875911,
      "learning_rate": 3.877722078340374e-06,
      "loss": 0.81054038,
      "num_input_tokens_seen": 49885825,
      "step": 2304,
      "time_per_iteration": 2.5090043544769287
    },
    {
      "auxiliary_loss_clip": 0.01146448,
      "auxiliary_loss_mlp": 0.0103769,
      "balance_loss_clip": 1.05027914,
      "balance_loss_mlp": 1.01953459,
      "epoch": 0.13858409739966931,
      "flos": 21543781334400.0,
      "grad_norm": 1.988349723931265,
      "language_loss": 0.7780304,
      "learning_rate": 3.877587952519672e-06,
      "loss": 0.7998718,
      "num_input_tokens_seen": 49905975,
      "step": 2305,
      "time_per_iteration": 2.509162664413452
    },
    {
      "auxiliary_loss_clip": 0.01070102,
      "auxiliary_loss_mlp": 0.01044866,
      "balance_loss_clip": 1.04341972,
      "balance_loss_mlp": 1.02595925,
      "epoch": 0.13864422065233728,
      "flos": 21579907438080.0,
      "grad_norm": 1.9203680995644727,
      "language_loss": 0.87776464,
      "learning_rate": 3.877453755500647e-06,
      "loss": 0.89891434,
      "num_input_tokens_seen": 49925800,
      "step": 2306,
      "time_per_iteration": 2.67519474029541
    },
    {
      "auxiliary_loss_clip": 0.01078296,
      "auxiliary_loss_mlp": 0.01004181,
      "balance_loss_clip": 1.03114533,
      "balance_loss_mlp": 1.00146282,
      "epoch": 0.13870434390500527,
      "flos": 53371156872960.0,
      "grad_norm": 0.8773613800815611,
      "language_loss": 0.59049106,
      "learning_rate": 3.877319487288387e-06,
      "loss": 0.61131585,
      "num_input_tokens_seen": 49977620,
      "step": 2307,
      "time_per_iteration": 3.1401150226593018
    },
    {
      "auxiliary_loss_clip": 0.01160115,
      "auxiliary_loss_mlp": 0.00951028,
      "balance_loss_clip": 1.05057335,
      "balance_loss_mlp": 1.26436496,
      "epoch": 0.13876446715767324,
      "flos": 22565906749440.0,
      "grad_norm": 2.0526371189203036,
      "language_loss": 0.79385793,
      "learning_rate": 3.877185147887984e-06,
      "loss": 0.81496936,
      "num_input_tokens_seen": 49996650,
      "step": 2308,
      "time_per_iteration": 2.5025112628936768
    },
    {
      "auxiliary_loss_clip": 0.01120328,
      "auxiliary_loss_mlp": 0.01040861,
      "balance_loss_clip": 1.04822564,
      "balance_loss_mlp": 1.02187061,
      "epoch": 0.1388245904103412,
      "flos": 20705231352960.0,
      "grad_norm": 2.5174666685254095,
      "language_loss": 0.77688539,
      "learning_rate": 3.877050737304533e-06,
      "loss": 0.7984972,
      "num_input_tokens_seen": 50015640,
      "step": 2309,
      "time_per_iteration": 2.5749220848083496
    },
    {
      "auxiliary_loss_clip": 0.01120434,
      "auxiliary_loss_mlp": 0.01043155,
      "balance_loss_clip": 1.04771185,
      "balance_loss_mlp": 1.02436769,
      "epoch": 0.13888471366300917,
      "flos": 20554729367040.0,
      "grad_norm": 2.410427148926027,
      "language_loss": 0.67824566,
      "learning_rate": 3.876916255543129e-06,
      "loss": 0.69988155,
      "num_input_tokens_seen": 50033500,
      "step": 2310,
      "time_per_iteration": 3.979099988937378
    },
    {
      "auxiliary_loss_clip": 0.01156701,
      "auxiliary_loss_mlp": 0.01045293,
      "balance_loss_clip": 1.0491786,
      "balance_loss_mlp": 1.02521801,
      "epoch": 0.13894483691567713,
      "flos": 13838033473920.0,
      "grad_norm": 2.3216118989655286,
      "language_loss": 0.83904058,
      "learning_rate": 3.8767817026088725e-06,
      "loss": 0.86106056,
      "num_input_tokens_seen": 50050075,
      "step": 2311,
      "time_per_iteration": 3.834423542022705
    },
    {
      "auxiliary_loss_clip": 0.01161533,
      "auxiliary_loss_mlp": 0.01043047,
      "balance_loss_clip": 1.05083394,
      "balance_loss_mlp": 1.02425933,
      "epoch": 0.1390049601683451,
      "flos": 28031186759040.0,
      "grad_norm": 3.841084755412252,
      "language_loss": 0.82023889,
      "learning_rate": 3.876647078506866e-06,
      "loss": 0.84228468,
      "num_input_tokens_seen": 50070080,
      "step": 2312,
      "time_per_iteration": 2.509683847427368
    },
    {
      "auxiliary_loss_clip": 0.01117831,
      "auxiliary_loss_mlp": 0.00940582,
      "balance_loss_clip": 1.04907608,
      "balance_loss_mlp": 1.24452209,
      "epoch": 0.13906508342101306,
      "flos": 26756860976640.0,
      "grad_norm": 2.673175701754439,
      "language_loss": 0.86887813,
      "learning_rate": 3.876512383242215e-06,
      "loss": 0.88946223,
      "num_input_tokens_seen": 50090040,
      "step": 2313,
      "time_per_iteration": 2.6080315113067627
    },
    {
      "auxiliary_loss_clip": 0.01156216,
      "auxiliary_loss_mlp": 0.01047957,
      "balance_loss_clip": 1.05036461,
      "balance_loss_mlp": 1.02903867,
      "epoch": 0.13912520667368106,
      "flos": 24535104111360.0,
      "grad_norm": 2.1793622707873954,
      "language_loss": 0.79971445,
      "learning_rate": 3.876377616820024e-06,
      "loss": 0.82175612,
      "num_input_tokens_seen": 50110595,
      "step": 2314,
      "time_per_iteration": 3.90889310836792
    },
    {
      "auxiliary_loss_clip": 0.01117439,
      "auxiliary_loss_mlp": 0.01040836,
      "balance_loss_clip": 1.04879045,
      "balance_loss_mlp": 1.02220333,
      "epoch": 0.13918532992634902,
      "flos": 19383215287680.0,
      "grad_norm": 2.9176018570663196,
      "language_loss": 0.85337019,
      "learning_rate": 3.876242779245409e-06,
      "loss": 0.87495291,
      "num_input_tokens_seen": 50125430,
      "step": 2315,
      "time_per_iteration": 2.5211880207061768
    },
    {
      "auxiliary_loss_clip": 0.01145014,
      "auxiliary_loss_mlp": 0.01041614,
      "balance_loss_clip": 1.04780459,
      "balance_loss_mlp": 1.02245688,
      "epoch": 0.139245453179017,
      "flos": 21323756574720.0,
      "grad_norm": 2.287500138878055,
      "language_loss": 0.77211845,
      "learning_rate": 3.876107870523477e-06,
      "loss": 0.79398471,
      "num_input_tokens_seen": 50144120,
      "step": 2316,
      "time_per_iteration": 2.5094428062438965
    },
    {
      "auxiliary_loss_clip": 0.01153948,
      "auxiliary_loss_mlp": 0.00939395,
      "balance_loss_clip": 1.04812002,
      "balance_loss_mlp": 1.24493027,
      "epoch": 0.13930557643168495,
      "flos": 19500607912320.0,
      "grad_norm": 1.995780809231942,
      "language_loss": 0.76920104,
      "learning_rate": 3.875972890659349e-06,
      "loss": 0.79013443,
      "num_input_tokens_seen": 50162500,
      "step": 2317,
      "time_per_iteration": 2.474806547164917
    },
    {
      "auxiliary_loss_clip": 0.01133562,
      "auxiliary_loss_mlp": 0.0104122,
      "balance_loss_clip": 1.04748893,
      "balance_loss_mlp": 1.02284956,
      "epoch": 0.13936569968435292,
      "flos": 25410821690880.0,
      "grad_norm": 2.2984703660616828,
      "language_loss": 0.80349118,
      "learning_rate": 3.875837839658139e-06,
      "loss": 0.82523894,
      "num_input_tokens_seen": 50182415,
      "step": 2318,
      "time_per_iteration": 4.003982782363892
    },
    {
      "auxiliary_loss_clip": 0.01051464,
      "auxiliary_loss_mlp": 0.01007544,
      "balance_loss_clip": 1.02659345,
      "balance_loss_mlp": 1.00514817,
      "epoch": 0.13942582293702088,
      "flos": 70771063731840.0,
      "grad_norm": 0.872366834080059,
      "language_loss": 0.59051883,
      "learning_rate": 3.87570271752497e-06,
      "loss": 0.6111089,
      "num_input_tokens_seen": 50245160,
      "step": 2319,
      "time_per_iteration": 3.1743321418762207
    },
    {
      "auxiliary_loss_clip": 0.01111531,
      "auxiliary_loss_mlp": 0.01048904,
      "balance_loss_clip": 1.04519451,
      "balance_loss_mlp": 1.02934158,
      "epoch": 0.13948594618968888,
      "flos": 35590885920000.0,
      "grad_norm": 2.224814398313495,
      "language_loss": 0.6545074,
      "learning_rate": 3.875567524264967e-06,
      "loss": 0.67611176,
      "num_input_tokens_seen": 50268215,
      "step": 2320,
      "time_per_iteration": 2.699268102645874
    },
    {
      "auxiliary_loss_clip": 0.01092786,
      "auxiliary_loss_mlp": 0.01047179,
      "balance_loss_clip": 1.04770553,
      "balance_loss_mlp": 1.02796245,
      "epoch": 0.13954606944235684,
      "flos": 21105204272640.0,
      "grad_norm": 1.5100877764740497,
      "language_loss": 0.7070905,
      "learning_rate": 3.875432259883256e-06,
      "loss": 0.72849011,
      "num_input_tokens_seen": 50288575,
      "step": 2321,
      "time_per_iteration": 2.6275782585144043
    },
    {
      "auxiliary_loss_clip": 0.01110368,
      "auxiliary_loss_mlp": 0.01060198,
      "balance_loss_clip": 1.04382467,
      "balance_loss_mlp": 1.03845382,
      "epoch": 0.1396061926950248,
      "flos": 25044425009280.0,
      "grad_norm": 1.9847061420442231,
      "language_loss": 0.86002183,
      "learning_rate": 3.875296924384965e-06,
      "loss": 0.88172746,
      "num_input_tokens_seen": 50308735,
      "step": 2322,
      "time_per_iteration": 2.6159567832946777
    },
    {
      "auxiliary_loss_clip": 0.0111402,
      "auxiliary_loss_mlp": 0.01047734,
      "balance_loss_clip": 1.04425669,
      "balance_loss_mlp": 1.02957857,
      "epoch": 0.13966631594769277,
      "flos": 37634023428480.0,
      "grad_norm": 2.4228963497799367,
      "language_loss": 0.67399824,
      "learning_rate": 3.875161517775226e-06,
      "loss": 0.69561583,
      "num_input_tokens_seen": 50331025,
      "step": 2323,
      "time_per_iteration": 2.713493824005127
    },
    {
      "auxiliary_loss_clip": 0.01120014,
      "auxiliary_loss_mlp": 0.01046442,
      "balance_loss_clip": 1.04763234,
      "balance_loss_mlp": 1.027071,
      "epoch": 0.13972643920036074,
      "flos": 16690993061760.0,
      "grad_norm": 1.9718052121175653,
      "language_loss": 0.88877928,
      "learning_rate": 3.875026040059175e-06,
      "loss": 0.9104439,
      "num_input_tokens_seen": 50349725,
      "step": 2324,
      "time_per_iteration": 2.5433168411254883
    },
    {
      "auxiliary_loss_clip": 0.01143469,
      "auxiliary_loss_mlp": 0.01050386,
      "balance_loss_clip": 1.04609513,
      "balance_loss_mlp": 1.03112173,
      "epoch": 0.1397865624530287,
      "flos": 23331055288320.0,
      "grad_norm": 3.0263083909692226,
      "language_loss": 0.71444559,
      "learning_rate": 3.8748904912419485e-06,
      "loss": 0.73638415,
      "num_input_tokens_seen": 50367965,
      "step": 2325,
      "time_per_iteration": 2.5170705318450928
    },
    {
      "auxiliary_loss_clip": 0.01126195,
      "auxiliary_loss_mlp": 0.00921269,
      "balance_loss_clip": 1.04889286,
      "balance_loss_mlp": 1.21076131,
      "epoch": 0.13984668570569667,
      "flos": 22778317825920.0,
      "grad_norm": 1.9107831827321762,
      "language_loss": 0.81782711,
      "learning_rate": 3.874754871328688e-06,
      "loss": 0.83830178,
      "num_input_tokens_seen": 50385605,
      "step": 2326,
      "time_per_iteration": 2.5563580989837646
    },
    {
      "auxiliary_loss_clip": 0.01142762,
      "auxiliary_loss_mlp": 0.01042677,
      "balance_loss_clip": 1.04927969,
      "balance_loss_mlp": 1.02476001,
      "epoch": 0.13990680895836466,
      "flos": 19464553635840.0,
      "grad_norm": 2.169482508787621,
      "language_loss": 0.89070171,
      "learning_rate": 3.874619180324534e-06,
      "loss": 0.91255605,
      "num_input_tokens_seen": 50403985,
      "step": 2327,
      "time_per_iteration": 2.5271570682525635
    },
    {
      "auxiliary_loss_clip": 0.01115175,
      "auxiliary_loss_mlp": 0.01059997,
      "balance_loss_clip": 1.04857409,
      "balance_loss_mlp": 1.0397675,
      "epoch": 0.13996693221103262,
      "flos": 20303283185280.0,
      "grad_norm": 2.132554526858539,
      "language_loss": 0.85315973,
      "learning_rate": 3.874483418234632e-06,
      "loss": 0.87491143,
      "num_input_tokens_seen": 50421590,
      "step": 2328,
      "time_per_iteration": 2.554732322692871
    },
    {
      "auxiliary_loss_clip": 0.01142902,
      "auxiliary_loss_mlp": 0.01043216,
      "balance_loss_clip": 1.04781806,
      "balance_loss_mlp": 1.02416646,
      "epoch": 0.1400270554637006,
      "flos": 26617707688320.0,
      "grad_norm": 2.218225721326118,
      "language_loss": 0.73846287,
      "learning_rate": 3.874347585064131e-06,
      "loss": 0.760324,
      "num_input_tokens_seen": 50443945,
      "step": 2329,
      "time_per_iteration": 2.570281505584717
    },
    {
      "auxiliary_loss_clip": 0.01144987,
      "auxiliary_loss_mlp": 0.01047085,
      "balance_loss_clip": 1.04784083,
      "balance_loss_mlp": 1.02715349,
      "epoch": 0.14008717871636855,
      "flos": 19391475415680.0,
      "grad_norm": 2.368847262634538,
      "language_loss": 0.78365093,
      "learning_rate": 3.874211680818183e-06,
      "loss": 0.80557162,
      "num_input_tokens_seen": 50462065,
      "step": 2330,
      "time_per_iteration": 2.4848597049713135
    },
    {
      "auxiliary_loss_clip": 0.01135589,
      "auxiliary_loss_mlp": 0.01042778,
      "balance_loss_clip": 1.04652834,
      "balance_loss_mlp": 1.02407432,
      "epoch": 0.14014730196903652,
      "flos": 15304266645120.0,
      "grad_norm": 2.3785602183077,
      "language_loss": 0.72316593,
      "learning_rate": 3.87407570550194e-06,
      "loss": 0.74494964,
      "num_input_tokens_seen": 50479565,
      "step": 2331,
      "time_per_iteration": 2.51332426071167
    },
    {
      "auxiliary_loss_clip": 0.0115392,
      "auxiliary_loss_mlp": 0.01054106,
      "balance_loss_clip": 1.05134177,
      "balance_loss_mlp": 1.03528285,
      "epoch": 0.14020742522170448,
      "flos": 14939701557120.0,
      "grad_norm": 1.703069969585632,
      "language_loss": 0.7249344,
      "learning_rate": 3.873939659120557e-06,
      "loss": 0.74701464,
      "num_input_tokens_seen": 50497305,
      "step": 2332,
      "time_per_iteration": 2.461261749267578
    },
    {
      "auxiliary_loss_clip": 0.01069512,
      "auxiliary_loss_mlp": 0.01004913,
      "balance_loss_clip": 1.03079426,
      "balance_loss_mlp": 1.00178957,
      "epoch": 0.14026754847437245,
      "flos": 48824580044160.0,
      "grad_norm": 0.8292856690048175,
      "language_loss": 0.56106591,
      "learning_rate": 3.873803541679196e-06,
      "loss": 0.58181012,
      "num_input_tokens_seen": 50549735,
      "step": 2333,
      "time_per_iteration": 2.972109317779541
    },
    {
      "auxiliary_loss_clip": 0.01123975,
      "auxiliary_loss_mlp": 0.01047276,
      "balance_loss_clip": 1.05167317,
      "balance_loss_mlp": 1.02846515,
      "epoch": 0.14032767172704044,
      "flos": 25773267876480.0,
      "grad_norm": 2.3558414455096384,
      "language_loss": 0.82983583,
      "learning_rate": 3.873667353183016e-06,
      "loss": 0.85154831,
      "num_input_tokens_seen": 50570100,
      "step": 2334,
      "time_per_iteration": 2.617372751235962
    },
    {
      "auxiliary_loss_clip": 0.0112794,
      "auxiliary_loss_mlp": 0.01047632,
      "balance_loss_clip": 1.05133891,
      "balance_loss_mlp": 1.02966738,
      "epoch": 0.1403877949797084,
      "flos": 21216312017280.0,
      "grad_norm": 2.6472498371203574,
      "language_loss": 0.80997992,
      "learning_rate": 3.8735310936371825e-06,
      "loss": 0.83173561,
      "num_input_tokens_seen": 50589185,
      "step": 2335,
      "time_per_iteration": 2.616795778274536
    },
    {
      "auxiliary_loss_clip": 0.01109947,
      "auxiliary_loss_mlp": 0.01049077,
      "balance_loss_clip": 1.05299914,
      "balance_loss_mlp": 1.02606964,
      "epoch": 0.14044791823237637,
      "flos": 22747973811840.0,
      "grad_norm": 1.8835121504200492,
      "language_loss": 0.82102263,
      "learning_rate": 3.873394763046862e-06,
      "loss": 0.84261286,
      "num_input_tokens_seen": 50609645,
      "step": 2336,
      "time_per_iteration": 2.614981174468994
    },
    {
      "auxiliary_loss_clip": 0.01147727,
      "auxiliary_loss_mlp": 0.01050173,
      "balance_loss_clip": 1.05568528,
      "balance_loss_mlp": 1.03005028,
      "epoch": 0.14050804148504434,
      "flos": 22964443125120.0,
      "grad_norm": 1.8328119586457146,
      "language_loss": 0.80571866,
      "learning_rate": 3.873258361417225e-06,
      "loss": 0.82769763,
      "num_input_tokens_seen": 50628385,
      "step": 2337,
      "time_per_iteration": 2.5789742469787598
    },
    {
      "auxiliary_loss_clip": 0.01144204,
      "auxiliary_loss_mlp": 0.01047553,
      "balance_loss_clip": 1.05084181,
      "balance_loss_mlp": 1.02886081,
      "epoch": 0.1405681647377123,
      "flos": 22200336080640.0,
      "grad_norm": 1.9742930519734958,
      "language_loss": 0.79585469,
      "learning_rate": 3.873121888753442e-06,
      "loss": 0.81777227,
      "num_input_tokens_seen": 50647260,
      "step": 2338,
      "time_per_iteration": 2.5856409072875977
    },
    {
      "auxiliary_loss_clip": 0.0115211,
      "auxiliary_loss_mlp": 0.0104559,
      "balance_loss_clip": 1.05629122,
      "balance_loss_mlp": 1.02501476,
      "epoch": 0.14062828799038027,
      "flos": 23732787974400.0,
      "grad_norm": 2.442499847692067,
      "language_loss": 0.80119693,
      "learning_rate": 3.87298534506069e-06,
      "loss": 0.82317394,
      "num_input_tokens_seen": 50666130,
      "step": 2339,
      "time_per_iteration": 2.5583200454711914
    },
    {
      "auxiliary_loss_clip": 0.01095591,
      "auxiliary_loss_mlp": 0.01053299,
      "balance_loss_clip": 1.04831469,
      "balance_loss_mlp": 1.0336895,
      "epoch": 0.14068841124304826,
      "flos": 39202493685120.0,
      "grad_norm": 1.8808790666082227,
      "language_loss": 0.65868014,
      "learning_rate": 3.872848730344146e-06,
      "loss": 0.68016905,
      "num_input_tokens_seen": 50687440,
      "step": 2340,
      "time_per_iteration": 2.792349338531494
    },
    {
      "auxiliary_loss_clip": 0.01143736,
      "auxiliary_loss_mlp": 0.01046777,
      "balance_loss_clip": 1.05116522,
      "balance_loss_mlp": 1.02741706,
      "epoch": 0.14074853449571623,
      "flos": 20192283181440.0,
      "grad_norm": 2.8043356207839296,
      "language_loss": 0.78892392,
      "learning_rate": 3.87271204460899e-06,
      "loss": 0.81082904,
      "num_input_tokens_seen": 50704030,
      "step": 2341,
      "time_per_iteration": 2.4812891483306885
    },
    {
      "auxiliary_loss_clip": 0.01156651,
      "auxiliary_loss_mlp": 0.01047817,
      "balance_loss_clip": 1.05123103,
      "balance_loss_mlp": 1.02906525,
      "epoch": 0.1408086577483842,
      "flos": 18405871153920.0,
      "grad_norm": 2.353033530720655,
      "language_loss": 0.80257642,
      "learning_rate": 3.8725752878604066e-06,
      "loss": 0.8246212,
      "num_input_tokens_seen": 50723305,
      "step": 2342,
      "time_per_iteration": 2.516427755355835
    },
    {
      "auxiliary_loss_clip": 0.0114877,
      "auxiliary_loss_mlp": 0.01042877,
      "balance_loss_clip": 1.05727243,
      "balance_loss_mlp": 1.02482891,
      "epoch": 0.14086878100105216,
      "flos": 25264593423360.0,
      "grad_norm": 2.225795725898041,
      "language_loss": 0.77608055,
      "learning_rate": 3.87243846010358e-06,
      "loss": 0.797997,
      "num_input_tokens_seen": 50743270,
      "step": 2343,
      "time_per_iteration": 2.540926456451416
    },
    {
      "auxiliary_loss_clip": 0.01062581,
      "auxiliary_loss_mlp": 0.01003326,
      "balance_loss_clip": 1.03434563,
      "balance_loss_mlp": 1.00046468,
      "epoch": 0.14092890425372012,
      "flos": 65978388869760.0,
      "grad_norm": 0.8436246852505997,
      "language_loss": 0.61455333,
      "learning_rate": 3.872301561343699e-06,
      "loss": 0.63521242,
      "num_input_tokens_seen": 50802710,
      "step": 2344,
      "time_per_iteration": 3.0747182369232178
    },
    {
      "auxiliary_loss_clip": 0.01140755,
      "auxiliary_loss_mlp": 0.01042702,
      "balance_loss_clip": 1.04659581,
      "balance_loss_mlp": 1.02534556,
      "epoch": 0.1409890275063881,
      "flos": 23694973931520.0,
      "grad_norm": 1.5533608229908993,
      "language_loss": 0.64724469,
      "learning_rate": 3.872164591585956e-06,
      "loss": 0.6690793,
      "num_input_tokens_seen": 50822625,
      "step": 2345,
      "time_per_iteration": 2.5403249263763428
    },
    {
      "auxiliary_loss_clip": 0.01151247,
      "auxiliary_loss_mlp": 0.01042777,
      "balance_loss_clip": 1.04772937,
      "balance_loss_mlp": 1.02183163,
      "epoch": 0.14104915075905605,
      "flos": 23623152687360.0,
      "grad_norm": 2.1710525329230896,
      "language_loss": 0.7372272,
      "learning_rate": 3.8720275508355435e-06,
      "loss": 0.75916743,
      "num_input_tokens_seen": 50842330,
      "step": 2346,
      "time_per_iteration": 2.521197557449341
    },
    {
      "auxiliary_loss_clip": 0.01145804,
      "auxiliary_loss_mlp": 0.0104394,
      "balance_loss_clip": 1.05222726,
      "balance_loss_mlp": 1.02422285,
      "epoch": 0.14110927401172405,
      "flos": 20595165102720.0,
      "grad_norm": 2.11144005097669,
      "language_loss": 0.77865791,
      "learning_rate": 3.8718904390976585e-06,
      "loss": 0.80055535,
      "num_input_tokens_seen": 50861035,
      "step": 2347,
      "time_per_iteration": 2.4952824115753174
    },
    {
      "auxiliary_loss_clip": 0.01159488,
      "auxiliary_loss_mlp": 0.01047713,
      "balance_loss_clip": 1.05093908,
      "balance_loss_mlp": 1.02966452,
      "epoch": 0.141169397264392,
      "flos": 28548049512960.0,
      "grad_norm": 2.7973745479466285,
      "language_loss": 0.7680254,
      "learning_rate": 3.8717532563775e-06,
      "loss": 0.79009736,
      "num_input_tokens_seen": 50880105,
      "step": 2348,
      "time_per_iteration": 3.8984763622283936
    },
    {
      "auxiliary_loss_clip": 0.01144058,
      "auxiliary_loss_mlp": 0.01046217,
      "balance_loss_clip": 1.04949188,
      "balance_loss_mlp": 1.02645206,
      "epoch": 0.14122952051705998,
      "flos": 17092258871040.0,
      "grad_norm": 1.7811837062399303,
      "language_loss": 0.86548471,
      "learning_rate": 3.871616002680272e-06,
      "loss": 0.88738751,
      "num_input_tokens_seen": 50897720,
      "step": 2349,
      "time_per_iteration": 2.504899501800537
    },
    {
      "auxiliary_loss_clip": 0.01146344,
      "auxiliary_loss_mlp": 0.01047456,
      "balance_loss_clip": 1.05331147,
      "balance_loss_mlp": 1.0291214,
      "epoch": 0.14128964376972794,
      "flos": 28946801370240.0,
      "grad_norm": 1.744153400578831,
      "language_loss": 0.88802528,
      "learning_rate": 3.871478678011177e-06,
      "loss": 0.90996331,
      "num_input_tokens_seen": 50918385,
      "step": 2350,
      "time_per_iteration": 3.9522485733032227
    },
    {
      "auxiliary_loss_clip": 0.01139015,
      "auxiliary_loss_mlp": 0.01046643,
      "balance_loss_clip": 1.05113626,
      "balance_loss_mlp": 1.02601993,
      "epoch": 0.1413497670223959,
      "flos": 18989778643200.0,
      "grad_norm": 1.9686023028834925,
      "language_loss": 0.80954373,
      "learning_rate": 3.871341282375423e-06,
      "loss": 0.83140033,
      "num_input_tokens_seen": 50938270,
      "step": 2351,
      "time_per_iteration": 2.563307523727417
    },
    {
      "auxiliary_loss_clip": 0.01145942,
      "auxiliary_loss_mlp": 0.01042829,
      "balance_loss_clip": 1.0491271,
      "balance_loss_mlp": 1.02338612,
      "epoch": 0.14140989027506387,
      "flos": 29862236413440.0,
      "grad_norm": 2.9775861775309647,
      "language_loss": 0.84139276,
      "learning_rate": 3.871203815778219e-06,
      "loss": 0.86328042,
      "num_input_tokens_seen": 50958155,
      "step": 2352,
      "time_per_iteration": 2.5839009284973145
    },
    {
      "auxiliary_loss_clip": 0.01074234,
      "auxiliary_loss_mlp": 0.0101349,
      "balance_loss_clip": 1.03581882,
      "balance_loss_mlp": 1.0105691,
      "epoch": 0.14147001352773186,
      "flos": 62079532041600.0,
      "grad_norm": 0.9108708315505235,
      "language_loss": 0.61892247,
      "learning_rate": 3.87106627822478e-06,
      "loss": 0.63979971,
      "num_input_tokens_seen": 51020705,
      "step": 2353,
      "time_per_iteration": 4.441821813583374
    },
    {
      "auxiliary_loss_clip": 0.01132038,
      "auxiliary_loss_mlp": 0.0104884,
      "balance_loss_clip": 1.05330324,
      "balance_loss_mlp": 1.03035092,
      "epoch": 0.14153013678039983,
      "flos": 22017514832640.0,
      "grad_norm": 1.7911081258702715,
      "language_loss": 0.8733592,
      "learning_rate": 3.8709286697203196e-06,
      "loss": 0.89516801,
      "num_input_tokens_seen": 51039995,
      "step": 2354,
      "time_per_iteration": 2.6256446838378906
    },
    {
      "auxiliary_loss_clip": 0.01118874,
      "auxiliary_loss_mlp": 0.01046174,
      "balance_loss_clip": 1.0518589,
      "balance_loss_mlp": 1.0258007,
      "epoch": 0.1415902600330678,
      "flos": 19720093968000.0,
      "grad_norm": 2.353340887008484,
      "language_loss": 0.74907368,
      "learning_rate": 3.870790990270057e-06,
      "loss": 0.77072418,
      "num_input_tokens_seen": 51059075,
      "step": 2355,
      "time_per_iteration": 2.5900442600250244
    },
    {
      "auxiliary_loss_clip": 0.01072228,
      "auxiliary_loss_mlp": 0.01003831,
      "balance_loss_clip": 1.03406572,
      "balance_loss_mlp": 1.0008868,
      "epoch": 0.14165038328573576,
      "flos": 65900929190400.0,
      "grad_norm": 0.6776336419900828,
      "language_loss": 0.5181582,
      "learning_rate": 3.870653239879212e-06,
      "loss": 0.53891879,
      "num_input_tokens_seen": 51120380,
      "step": 2356,
      "time_per_iteration": 4.516743421554565
    },
    {
      "auxiliary_loss_clip": 0.01157027,
      "auxiliary_loss_mlp": 0.01050056,
      "balance_loss_clip": 1.04953122,
      "balance_loss_mlp": 1.03070796,
      "epoch": 0.14171050653840372,
      "flos": 12130158533760.0,
      "grad_norm": 2.432961504733317,
      "language_loss": 0.70673931,
      "learning_rate": 3.8705154185530095e-06,
      "loss": 0.72881019,
      "num_input_tokens_seen": 51136950,
      "step": 2357,
      "time_per_iteration": 2.529110908508301
    },
    {
      "auxiliary_loss_clip": 0.01111861,
      "auxiliary_loss_mlp": 0.01054162,
      "balance_loss_clip": 1.04524815,
      "balance_loss_mlp": 1.03496933,
      "epoch": 0.1417706297910717,
      "flos": 20412487509120.0,
      "grad_norm": 1.8516710233737657,
      "language_loss": 0.82284719,
      "learning_rate": 3.870377526296674e-06,
      "loss": 0.84450746,
      "num_input_tokens_seen": 51155175,
      "step": 2358,
      "time_per_iteration": 2.6158409118652344
    },
    {
      "auxiliary_loss_clip": 0.0113962,
      "auxiliary_loss_mlp": 0.01048205,
      "balance_loss_clip": 1.04979026,
      "balance_loss_mlp": 1.02823782,
      "epoch": 0.14183075304373965,
      "flos": 22380607463040.0,
      "grad_norm": 2.6389974228093855,
      "language_loss": 0.72568476,
      "learning_rate": 3.870239563115436e-06,
      "loss": 0.747563,
      "num_input_tokens_seen": 51174500,
      "step": 2359,
      "time_per_iteration": 2.5967633724212646
    },
    {
      "auxiliary_loss_clip": 0.01114598,
      "auxiliary_loss_mlp": 0.00891882,
      "balance_loss_clip": 1.06378746,
      "balance_loss_mlp": 1.16804254,
      "epoch": 0.14189087629640765,
      "flos": 21580913018880.0,
      "grad_norm": 2.189460943415681,
      "language_loss": 0.75498432,
      "learning_rate": 3.870101529014526e-06,
      "loss": 0.77504909,
      "num_input_tokens_seen": 51194270,
      "step": 2360,
      "time_per_iteration": 2.625854253768921
    },
    {
      "auxiliary_loss_clip": 0.01104416,
      "auxiliary_loss_mlp": 0.01048787,
      "balance_loss_clip": 1.04892302,
      "balance_loss_mlp": 1.02754354,
      "epoch": 0.1419509995490756,
      "flos": 20008564093440.0,
      "grad_norm": 3.1749598234963394,
      "language_loss": 0.81500852,
      "learning_rate": 3.869963423999178e-06,
      "loss": 0.83654052,
      "num_input_tokens_seen": 51211850,
      "step": 2361,
      "time_per_iteration": 2.5813772678375244
    },
    {
      "auxiliary_loss_clip": 0.01143059,
      "auxiliary_loss_mlp": 0.01050351,
      "balance_loss_clip": 1.04883194,
      "balance_loss_mlp": 1.03006184,
      "epoch": 0.14201112280174358,
      "flos": 31941464112000.0,
      "grad_norm": 2.1968476439628852,
      "language_loss": 0.7432307,
      "learning_rate": 3.86982524807463e-06,
      "loss": 0.76516485,
      "num_input_tokens_seen": 51233545,
      "step": 2362,
      "time_per_iteration": 2.5793187618255615
    },
    {
      "auxiliary_loss_clip": 0.01149184,
      "auxiliary_loss_mlp": 0.01045051,
      "balance_loss_clip": 1.05184603,
      "balance_loss_mlp": 1.02479768,
      "epoch": 0.14207124605441154,
      "flos": 41464147582080.0,
      "grad_norm": 2.021978718972557,
      "language_loss": 0.7405777,
      "learning_rate": 3.869687001246122e-06,
      "loss": 0.76252007,
      "num_input_tokens_seen": 51257615,
      "step": 2363,
      "time_per_iteration": 2.6722002029418945
    },
    {
      "auxiliary_loss_clip": 0.01120905,
      "auxiliary_loss_mlp": 0.01045643,
      "balance_loss_clip": 1.04627252,
      "balance_loss_mlp": 1.02530575,
      "epoch": 0.1421313693070795,
      "flos": 31905086613120.0,
      "grad_norm": 1.897988213137336,
      "language_loss": 0.72725266,
      "learning_rate": 3.8695486835188946e-06,
      "loss": 0.74891818,
      "num_input_tokens_seen": 51279645,
      "step": 2364,
      "time_per_iteration": 2.635650396347046
    },
    {
      "auxiliary_loss_clip": 0.01133052,
      "auxiliary_loss_mlp": 0.01043589,
      "balance_loss_clip": 1.04663527,
      "balance_loss_mlp": 1.02571917,
      "epoch": 0.14219149255974747,
      "flos": 26871165031680.0,
      "grad_norm": 1.8514624265372541,
      "language_loss": 0.90988708,
      "learning_rate": 3.869410294898195e-06,
      "loss": 0.9316535,
      "num_input_tokens_seen": 51299775,
      "step": 2365,
      "time_per_iteration": 2.5796475410461426
    },
    {
      "auxiliary_loss_clip": 0.01119466,
      "auxiliary_loss_mlp": 0.01049858,
      "balance_loss_clip": 1.04390836,
      "balance_loss_mlp": 1.02927017,
      "epoch": 0.14225161581241544,
      "flos": 27454426076160.0,
      "grad_norm": 2.10264149058101,
      "language_loss": 0.65041018,
      "learning_rate": 3.869271835389268e-06,
      "loss": 0.67210346,
      "num_input_tokens_seen": 51319430,
      "step": 2366,
      "time_per_iteration": 2.570834159851074
    },
    {
      "auxiliary_loss_clip": 0.01139515,
      "auxiliary_loss_mlp": 0.01049053,
      "balance_loss_clip": 1.05209172,
      "balance_loss_mlp": 1.02900195,
      "epoch": 0.14231173906508343,
      "flos": 10561436881920.0,
      "grad_norm": 2.030600819027385,
      "language_loss": 0.8067565,
      "learning_rate": 3.8691333049973665e-06,
      "loss": 0.82864219,
      "num_input_tokens_seen": 51336045,
      "step": 2367,
      "time_per_iteration": 2.4998860359191895
    },
    {
      "auxiliary_loss_clip": 0.01134484,
      "auxiliary_loss_mlp": 0.01055894,
      "balance_loss_clip": 1.05267739,
      "balance_loss_mlp": 1.03505635,
      "epoch": 0.1423718623177514,
      "flos": 28360882719360.0,
      "grad_norm": 1.838936507219784,
      "language_loss": 0.82819843,
      "learning_rate": 3.868994703727742e-06,
      "loss": 0.85010225,
      "num_input_tokens_seen": 51357030,
      "step": 2368,
      "time_per_iteration": 2.5843539237976074
    },
    {
      "auxiliary_loss_clip": 0.01119932,
      "auxiliary_loss_mlp": 0.01051186,
      "balance_loss_clip": 1.0546186,
      "balance_loss_mlp": 1.03052735,
      "epoch": 0.14243198557041936,
      "flos": 19354235990400.0,
      "grad_norm": 2.2887122716968538,
      "language_loss": 0.870224,
      "learning_rate": 3.868856031585652e-06,
      "loss": 0.89193517,
      "num_input_tokens_seen": 51374890,
      "step": 2369,
      "time_per_iteration": 2.5474843978881836
    },
    {
      "auxiliary_loss_clip": 0.01121296,
      "auxiliary_loss_mlp": 0.01049207,
      "balance_loss_clip": 1.04974151,
      "balance_loss_mlp": 1.02941799,
      "epoch": 0.14249210882308733,
      "flos": 28806857982720.0,
      "grad_norm": 2.387897387838576,
      "language_loss": 0.76191443,
      "learning_rate": 3.868717288576354e-06,
      "loss": 0.7836194,
      "num_input_tokens_seen": 51398100,
      "step": 2370,
      "time_per_iteration": 2.6298346519470215
    },
    {
      "auxiliary_loss_clip": 0.01144442,
      "auxiliary_loss_mlp": 0.00866392,
      "balance_loss_clip": 1.04747415,
      "balance_loss_mlp": 1.12016606,
      "epoch": 0.1425522320757553,
      "flos": 21835016807040.0,
      "grad_norm": 3.1318280629287476,
      "language_loss": 0.8292712,
      "learning_rate": 3.868578474705109e-06,
      "loss": 0.84937954,
      "num_input_tokens_seen": 51418745,
      "step": 2371,
      "time_per_iteration": 2.50667667388916
    },
    {
      "auxiliary_loss_clip": 0.01163409,
      "auxiliary_loss_mlp": 0.01046862,
      "balance_loss_clip": 1.05431151,
      "balance_loss_mlp": 1.02702498,
      "epoch": 0.14261235532842326,
      "flos": 17311457617920.0,
      "grad_norm": 1.8992084806570018,
      "language_loss": 0.82754177,
      "learning_rate": 3.868439589977181e-06,
      "loss": 0.84964448,
      "num_input_tokens_seen": 51437455,
      "step": 2372,
      "time_per_iteration": 2.4611079692840576
    },
    {
      "auxiliary_loss_clip": 0.01162847,
      "auxiliary_loss_mlp": 0.01046862,
      "balance_loss_clip": 1.05429161,
      "balance_loss_mlp": 1.02724028,
      "epoch": 0.14267247858109125,
      "flos": 18806741913600.0,
      "grad_norm": 2.347253838390265,
      "language_loss": 0.84869987,
      "learning_rate": 3.868300634397836e-06,
      "loss": 0.87079692,
      "num_input_tokens_seen": 51455710,
      "step": 2373,
      "time_per_iteration": 2.4502651691436768
    },
    {
      "auxiliary_loss_clip": 0.0113331,
      "auxiliary_loss_mlp": 0.01053804,
      "balance_loss_clip": 1.05015886,
      "balance_loss_mlp": 1.03592253,
      "epoch": 0.14273260183375922,
      "flos": 11358904682880.0,
      "grad_norm": 2.0192949736253407,
      "language_loss": 0.85991794,
      "learning_rate": 3.8681616079723445e-06,
      "loss": 0.88178909,
      "num_input_tokens_seen": 51471270,
      "step": 2374,
      "time_per_iteration": 2.488788604736328
    },
    {
      "auxiliary_loss_clip": 0.01150064,
      "auxiliary_loss_mlp": 0.01052147,
      "balance_loss_clip": 1.04906094,
      "balance_loss_mlp": 1.03208458,
      "epoch": 0.14279272508642718,
      "flos": 27567688636800.0,
      "grad_norm": 1.8171293793217045,
      "language_loss": 0.79282093,
      "learning_rate": 3.868022510705977e-06,
      "loss": 0.81484306,
      "num_input_tokens_seen": 51492705,
      "step": 2375,
      "time_per_iteration": 2.5768489837646484
    },
    {
      "auxiliary_loss_clip": 0.01150901,
      "auxiliary_loss_mlp": 0.01055518,
      "balance_loss_clip": 1.05348802,
      "balance_loss_mlp": 1.03671908,
      "epoch": 0.14285284833909515,
      "flos": 16252559654400.0,
      "grad_norm": 2.9996575294357637,
      "language_loss": 0.76530689,
      "learning_rate": 3.867883342604009e-06,
      "loss": 0.78737104,
      "num_input_tokens_seen": 51510780,
      "step": 2376,
      "time_per_iteration": 2.467540740966797
    },
    {
      "auxiliary_loss_clip": 0.01150554,
      "auxiliary_loss_mlp": 0.01047552,
      "balance_loss_clip": 1.05290425,
      "balance_loss_mlp": 1.02831149,
      "epoch": 0.1429129715917631,
      "flos": 19755609540480.0,
      "grad_norm": 1.746192914125844,
      "language_loss": 0.92959297,
      "learning_rate": 3.867744103671717e-06,
      "loss": 0.95157409,
      "num_input_tokens_seen": 51531400,
      "step": 2377,
      "time_per_iteration": 2.5246336460113525
    },
    {
      "auxiliary_loss_clip": 0.01138522,
      "auxiliary_loss_mlp": 0.01045181,
      "balance_loss_clip": 1.05441332,
      "balance_loss_mlp": 1.0244745,
      "epoch": 0.14297309484443108,
      "flos": 21137092571520.0,
      "grad_norm": 1.9271980538062394,
      "language_loss": 0.9178623,
      "learning_rate": 3.867604793914382e-06,
      "loss": 0.93969929,
      "num_input_tokens_seen": 51548215,
      "step": 2378,
      "time_per_iteration": 2.5184245109558105
    },
    {
      "auxiliary_loss_clip": 0.01157365,
      "auxiliary_loss_mlp": 0.01047978,
      "balance_loss_clip": 1.05692542,
      "balance_loss_mlp": 1.02814174,
      "epoch": 0.14303321809709904,
      "flos": 23586667447680.0,
      "grad_norm": 2.0016679823466017,
      "language_loss": 0.73856318,
      "learning_rate": 3.8674654133372864e-06,
      "loss": 0.7606166,
      "num_input_tokens_seen": 51566820,
      "step": 2379,
      "time_per_iteration": 2.520867109298706
    },
    {
      "auxiliary_loss_clip": 0.0112883,
      "auxiliary_loss_mlp": 0.01053705,
      "balance_loss_clip": 1.05220819,
      "balance_loss_mlp": 1.03351116,
      "epoch": 0.14309334134976703,
      "flos": 15888281875200.0,
      "grad_norm": 1.8646790066835648,
      "language_loss": 0.78695369,
      "learning_rate": 3.867325961945714e-06,
      "loss": 0.808779,
      "num_input_tokens_seen": 51585075,
      "step": 2380,
      "time_per_iteration": 2.5406453609466553
    },
    {
      "auxiliary_loss_clip": 0.01115325,
      "auxiliary_loss_mlp": 0.01051192,
      "balance_loss_clip": 1.05328631,
      "balance_loss_mlp": 1.03139162,
      "epoch": 0.143153464602435,
      "flos": 16325601960960.0,
      "grad_norm": 2.8080046092020066,
      "language_loss": 0.88055819,
      "learning_rate": 3.867186439744955e-06,
      "loss": 0.90222341,
      "num_input_tokens_seen": 51603185,
      "step": 2381,
      "time_per_iteration": 2.545492649078369
    },
    {
      "auxiliary_loss_clip": 0.01134981,
      "auxiliary_loss_mlp": 0.01050853,
      "balance_loss_clip": 1.05584764,
      "balance_loss_mlp": 1.0314219,
      "epoch": 0.14321358785510296,
      "flos": 17092079303040.0,
      "grad_norm": 2.09165048714896,
      "language_loss": 0.76850283,
      "learning_rate": 3.867046846740299e-06,
      "loss": 0.79036117,
      "num_input_tokens_seen": 51620880,
      "step": 2382,
      "time_per_iteration": 2.496264696121216
    },
    {
      "auxiliary_loss_clip": 0.01122472,
      "auxiliary_loss_mlp": 0.01052256,
      "balance_loss_clip": 1.05151629,
      "balance_loss_mlp": 1.03306341,
      "epoch": 0.14327371110777093,
      "flos": 26322916769280.0,
      "grad_norm": 2.0530936131284547,
      "language_loss": 0.76514703,
      "learning_rate": 3.866907182937039e-06,
      "loss": 0.78689432,
      "num_input_tokens_seen": 51640170,
      "step": 2383,
      "time_per_iteration": 2.5881659984588623
    },
    {
      "auxiliary_loss_clip": 0.0112931,
      "auxiliary_loss_mlp": 0.01054031,
      "balance_loss_clip": 1.04982519,
      "balance_loss_mlp": 1.03297865,
      "epoch": 0.1433338343604389,
      "flos": 18076462502400.0,
      "grad_norm": 2.4139628588268756,
      "language_loss": 0.88001651,
      "learning_rate": 3.866767448340471e-06,
      "loss": 0.90184993,
      "num_input_tokens_seen": 51656580,
      "step": 2384,
      "time_per_iteration": 2.511021137237549
    },
    {
      "auxiliary_loss_clip": 0.0115865,
      "auxiliary_loss_mlp": 0.01050277,
      "balance_loss_clip": 1.05780602,
      "balance_loss_mlp": 1.02905774,
      "epoch": 0.14339395761310686,
      "flos": 15522783033600.0,
      "grad_norm": 2.625630882876604,
      "language_loss": 0.80109084,
      "learning_rate": 3.866627642955895e-06,
      "loss": 0.82318014,
      "num_input_tokens_seen": 51674645,
      "step": 2385,
      "time_per_iteration": 2.4832351207733154
    },
    {
      "auxiliary_loss_clip": 0.0114987,
      "auxiliary_loss_mlp": 0.01046667,
      "balance_loss_clip": 1.05168986,
      "balance_loss_mlp": 1.02741504,
      "epoch": 0.14345408086577485,
      "flos": 28548767784960.0,
      "grad_norm": 9.104349935864557,
      "language_loss": 0.75235558,
      "learning_rate": 3.866487766788612e-06,
      "loss": 0.77432096,
      "num_input_tokens_seen": 51695770,
      "step": 2386,
      "time_per_iteration": 2.5584936141967773
    },
    {
      "auxiliary_loss_clip": 0.01163027,
      "auxiliary_loss_mlp": 0.0104409,
      "balance_loss_clip": 1.05462766,
      "balance_loss_mlp": 1.02464676,
      "epoch": 0.14351420411844282,
      "flos": 20230061310720.0,
      "grad_norm": 2.0740357533184173,
      "language_loss": 0.78584349,
      "learning_rate": 3.866347819843925e-06,
      "loss": 0.80791467,
      "num_input_tokens_seen": 51714165,
      "step": 2387,
      "time_per_iteration": 3.848055362701416
    },
    {
      "auxiliary_loss_clip": 0.01135868,
      "auxiliary_loss_mlp": 0.01049618,
      "balance_loss_clip": 1.05529499,
      "balance_loss_mlp": 1.02929306,
      "epoch": 0.14357432737111078,
      "flos": 19865029345920.0,
      "grad_norm": 2.1515238488054638,
      "language_loss": 0.82122087,
      "learning_rate": 3.866207802127143e-06,
      "loss": 0.84307569,
      "num_input_tokens_seen": 51734440,
      "step": 2388,
      "time_per_iteration": 3.93045711517334
    },
    {
      "auxiliary_loss_clip": 0.0115209,
      "auxiliary_loss_mlp": 0.01045783,
      "balance_loss_clip": 1.05611753,
      "balance_loss_mlp": 1.02688837,
      "epoch": 0.14363445062377875,
      "flos": 28256814040320.0,
      "grad_norm": 1.9639154784444621,
      "language_loss": 0.82253081,
      "learning_rate": 3.866067713643573e-06,
      "loss": 0.84450954,
      "num_input_tokens_seen": 51753730,
      "step": 2389,
      "time_per_iteration": 2.552861213684082
    },
    {
      "auxiliary_loss_clip": 0.01142006,
      "auxiliary_loss_mlp": 0.01053346,
      "balance_loss_clip": 1.05546343,
      "balance_loss_mlp": 1.03274703,
      "epoch": 0.1436945738764467,
      "flos": 18186672407040.0,
      "grad_norm": 1.8818231947866149,
      "language_loss": 0.82913232,
      "learning_rate": 3.8659275543985285e-06,
      "loss": 0.8510859,
      "num_input_tokens_seen": 51771195,
      "step": 2390,
      "time_per_iteration": 2.5056447982788086
    },
    {
      "auxiliary_loss_clip": 0.0114839,
      "auxiliary_loss_mlp": 0.01051726,
      "balance_loss_clip": 1.0550282,
      "balance_loss_mlp": 1.03215218,
      "epoch": 0.14375469712911468,
      "flos": 27307910499840.0,
      "grad_norm": 1.7342816294104022,
      "language_loss": 0.74428582,
      "learning_rate": 3.865787324397324e-06,
      "loss": 0.76628697,
      "num_input_tokens_seen": 51792290,
      "step": 2391,
      "time_per_iteration": 3.948801279067993
    },
    {
      "auxiliary_loss_clip": 0.01063771,
      "auxiliary_loss_mlp": 0.01013555,
      "balance_loss_clip": 1.04984844,
      "balance_loss_mlp": 1.00957346,
      "epoch": 0.14381482038178264,
      "flos": 56891445287040.0,
      "grad_norm": 0.8707198708032963,
      "language_loss": 0.61781198,
      "learning_rate": 3.865647023645277e-06,
      "loss": 0.63858521,
      "num_input_tokens_seen": 51843675,
      "step": 2392,
      "time_per_iteration": 3.0174596309661865
    },
    {
      "auxiliary_loss_clip": 0.01154601,
      "auxiliary_loss_mlp": 0.01048503,
      "balance_loss_clip": 1.05302262,
      "balance_loss_mlp": 1.02695012,
      "epoch": 0.14387494363445064,
      "flos": 14282177143680.0,
      "grad_norm": 2.428920892386481,
      "language_loss": 0.7671051,
      "learning_rate": 3.865506652147709e-06,
      "loss": 0.78913611,
      "num_input_tokens_seen": 51860285,
      "step": 2393,
      "time_per_iteration": 2.493244171142578
    },
    {
      "auxiliary_loss_clip": 0.01167249,
      "auxiliary_loss_mlp": 0.0104939,
      "balance_loss_clip": 1.05633152,
      "balance_loss_mlp": 1.02985144,
      "epoch": 0.1439350668871186,
      "flos": 26761493831040.0,
      "grad_norm": 2.077460720931814,
      "language_loss": 0.7646414,
      "learning_rate": 3.865366209909941e-06,
      "loss": 0.78680778,
      "num_input_tokens_seen": 51880105,
      "step": 2394,
      "time_per_iteration": 2.525285005569458
    },
    {
      "auxiliary_loss_clip": 0.01161888,
      "auxiliary_loss_mlp": 0.01049812,
      "balance_loss_clip": 1.05324435,
      "balance_loss_mlp": 1.03041649,
      "epoch": 0.14399519013978657,
      "flos": 40700040537600.0,
      "grad_norm": 1.5784814653527055,
      "language_loss": 0.86147541,
      "learning_rate": 3.8652256969372994e-06,
      "loss": 0.88359243,
      "num_input_tokens_seen": 51905175,
      "step": 2395,
      "time_per_iteration": 4.01458215713501
    },
    {
      "auxiliary_loss_clip": 0.01129677,
      "auxiliary_loss_mlp": 0.01056355,
      "balance_loss_clip": 1.0573082,
      "balance_loss_mlp": 1.03501642,
      "epoch": 0.14405531339245453,
      "flos": 20557530627840.0,
      "grad_norm": 1.772261487307856,
      "language_loss": 0.82859159,
      "learning_rate": 3.865085113235113e-06,
      "loss": 0.85045183,
      "num_input_tokens_seen": 51924490,
      "step": 2396,
      "time_per_iteration": 2.565033435821533
    },
    {
      "auxiliary_loss_clip": 0.01125847,
      "auxiliary_loss_mlp": 0.00846104,
      "balance_loss_clip": 1.0493741,
      "balance_loss_mlp": 1.08699703,
      "epoch": 0.1441154366451225,
      "flos": 19572931946880.0,
      "grad_norm": 2.316187957587271,
      "language_loss": 0.82636178,
      "learning_rate": 3.864944458808712e-06,
      "loss": 0.84608126,
      "num_input_tokens_seen": 51940490,
      "step": 2397,
      "time_per_iteration": 2.5220518112182617
    },
    {
      "auxiliary_loss_clip": 0.01164316,
      "auxiliary_loss_mlp": 0.01045982,
      "balance_loss_clip": 1.05332339,
      "balance_loss_mlp": 1.02589548,
      "epoch": 0.14417555989779046,
      "flos": 18515721922560.0,
      "grad_norm": 1.860365258585905,
      "language_loss": 0.79914647,
      "learning_rate": 3.86480373366343e-06,
      "loss": 0.82124949,
      "num_input_tokens_seen": 51957910,
      "step": 2398,
      "time_per_iteration": 2.4433844089508057
    },
    {
      "auxiliary_loss_clip": 0.01150836,
      "auxiliary_loss_mlp": 0.01051757,
      "balance_loss_clip": 1.05350804,
      "balance_loss_mlp": 1.03277946,
      "epoch": 0.14423568315045843,
      "flos": 26031681296640.0,
      "grad_norm": 2.3800124071001254,
      "language_loss": 0.65255529,
      "learning_rate": 3.864662937804603e-06,
      "loss": 0.67458117,
      "num_input_tokens_seen": 51978010,
      "step": 2399,
      "time_per_iteration": 2.543267011642456
    },
    {
      "auxiliary_loss_clip": 0.01133007,
      "auxiliary_loss_mlp": 0.01047733,
      "balance_loss_clip": 1.05065489,
      "balance_loss_mlp": 1.02777779,
      "epoch": 0.14429580640312642,
      "flos": 21288743792640.0,
      "grad_norm": 1.5113858370378281,
      "language_loss": 0.8198992,
      "learning_rate": 3.864522071237571e-06,
      "loss": 0.84170663,
      "num_input_tokens_seen": 51998515,
      "step": 2400,
      "time_per_iteration": 2.561065673828125
    },
    {
      "auxiliary_loss_clip": 0.01147108,
      "auxiliary_loss_mlp": 0.01054395,
      "balance_loss_clip": 1.05582404,
      "balance_loss_mlp": 1.03236485,
      "epoch": 0.14435592965579438,
      "flos": 25627865621760.0,
      "grad_norm": 1.9762082167977804,
      "language_loss": 0.74498641,
      "learning_rate": 3.864381133967676e-06,
      "loss": 0.76700145,
      "num_input_tokens_seen": 52019270,
      "step": 2401,
      "time_per_iteration": 2.5629281997680664
    },
    {
      "auxiliary_loss_clip": 0.01133008,
      "auxiliary_loss_mlp": 0.01049602,
      "balance_loss_clip": 1.05148911,
      "balance_loss_mlp": 1.03052878,
      "epoch": 0.14441605290846235,
      "flos": 22965053656320.0,
      "grad_norm": 1.5962481101633015,
      "language_loss": 0.8081578,
      "learning_rate": 3.86424012600026e-06,
      "loss": 0.82998383,
      "num_input_tokens_seen": 52039315,
      "step": 2402,
      "time_per_iteration": 2.5392982959747314
    },
    {
      "auxiliary_loss_clip": 0.01120054,
      "auxiliary_loss_mlp": 0.01048627,
      "balance_loss_clip": 1.05200195,
      "balance_loss_mlp": 1.02840948,
      "epoch": 0.14447617616113032,
      "flos": 17347655548800.0,
      "grad_norm": 2.2002969160158266,
      "language_loss": 0.8418709,
      "learning_rate": 3.864099047340673e-06,
      "loss": 0.8635577,
      "num_input_tokens_seen": 52056555,
      "step": 2403,
      "time_per_iteration": 2.533567428588867
    },
    {
      "auxiliary_loss_clip": 0.01126057,
      "auxiliary_loss_mlp": 0.00833696,
      "balance_loss_clip": 1.05106664,
      "balance_loss_mlp": 1.06204689,
      "epoch": 0.14453629941379828,
      "flos": 24060185464320.0,
      "grad_norm": 1.6057215126408448,
      "language_loss": 0.70365715,
      "learning_rate": 3.863957897994262e-06,
      "loss": 0.72325468,
      "num_input_tokens_seen": 52075800,
      "step": 2404,
      "time_per_iteration": 2.571737289428711
    },
    {
      "auxiliary_loss_clip": 0.0113364,
      "auxiliary_loss_mlp": 0.01053691,
      "balance_loss_clip": 1.04930913,
      "balance_loss_mlp": 1.03511882,
      "epoch": 0.14459642266646625,
      "flos": 14429554646400.0,
      "grad_norm": 2.3923370709933223,
      "language_loss": 0.73765504,
      "learning_rate": 3.863816677966381e-06,
      "loss": 0.7595284,
      "num_input_tokens_seen": 52092585,
      "step": 2405,
      "time_per_iteration": 2.506847381591797
    },
    {
      "auxiliary_loss_clip": 0.01104769,
      "auxiliary_loss_mlp": 0.01048467,
      "balance_loss_clip": 1.04967916,
      "balance_loss_mlp": 1.02873778,
      "epoch": 0.14465654591913424,
      "flos": 9867032179200.0,
      "grad_norm": 2.2066935534490004,
      "language_loss": 0.73162353,
      "learning_rate": 3.863675387262386e-06,
      "loss": 0.75315595,
      "num_input_tokens_seen": 52108990,
      "step": 2406,
      "time_per_iteration": 2.613922357559204
    },
    {
      "auxiliary_loss_clip": 0.01150281,
      "auxiliary_loss_mlp": 0.01054468,
      "balance_loss_clip": 1.05164158,
      "balance_loss_mlp": 1.03296304,
      "epoch": 0.1447166691718022,
      "flos": 24972926987520.0,
      "grad_norm": 2.3895689963927667,
      "language_loss": 0.75921392,
      "learning_rate": 3.8635340258876325e-06,
      "loss": 0.78126144,
      "num_input_tokens_seen": 52125385,
      "step": 2407,
      "time_per_iteration": 2.526467800140381
    },
    {
      "auxiliary_loss_clip": 0.01157601,
      "auxiliary_loss_mlp": 0.0104718,
      "balance_loss_clip": 1.0497967,
      "balance_loss_mlp": 1.02808285,
      "epoch": 0.14477679242447017,
      "flos": 21908023200000.0,
      "grad_norm": 1.5588972128959082,
      "language_loss": 0.79662001,
      "learning_rate": 3.8633925938474826e-06,
      "loss": 0.81866777,
      "num_input_tokens_seen": 52144985,
      "step": 2408,
      "time_per_iteration": 2.4909427165985107
    },
    {
      "auxiliary_loss_clip": 0.01150315,
      "auxiliary_loss_mlp": 0.01054395,
      "balance_loss_clip": 1.0530684,
      "balance_loss_mlp": 1.03327072,
      "epoch": 0.14483691567713813,
      "flos": 20740746925440.0,
      "grad_norm": 3.1724136272465686,
      "language_loss": 0.82304269,
      "learning_rate": 3.863251091147299e-06,
      "loss": 0.84508979,
      "num_input_tokens_seen": 52163885,
      "step": 2409,
      "time_per_iteration": 2.509922981262207
    },
    {
      "auxiliary_loss_clip": 0.01108891,
      "auxiliary_loss_mlp": 0.01060451,
      "balance_loss_clip": 1.04645681,
      "balance_loss_mlp": 1.03892159,
      "epoch": 0.1448970389298061,
      "flos": 35407705536000.0,
      "grad_norm": 2.681809018346349,
      "language_loss": 0.75073659,
      "learning_rate": 3.863109517792446e-06,
      "loss": 0.77243,
      "num_input_tokens_seen": 52184325,
      "step": 2410,
      "time_per_iteration": 2.6931052207946777
    },
    {
      "auxiliary_loss_clip": 0.01161249,
      "auxiliary_loss_mlp": 0.01049748,
      "balance_loss_clip": 1.05313826,
      "balance_loss_mlp": 1.03111529,
      "epoch": 0.14495716218247406,
      "flos": 15414368808960.0,
      "grad_norm": 1.8206245857700996,
      "language_loss": 0.81976438,
      "learning_rate": 3.8629678737882945e-06,
      "loss": 0.84187436,
      "num_input_tokens_seen": 52202740,
      "step": 2411,
      "time_per_iteration": 2.4585249423980713
    },
    {
      "auxiliary_loss_clip": 0.01137195,
      "auxiliary_loss_mlp": 0.01055371,
      "balance_loss_clip": 1.0538919,
      "balance_loss_mlp": 1.03501034,
      "epoch": 0.14501728543514203,
      "flos": 33693222493440.0,
      "grad_norm": 1.9941777373460778,
      "language_loss": 0.7031132,
      "learning_rate": 3.862826159140214e-06,
      "loss": 0.72503889,
      "num_input_tokens_seen": 52223100,
      "step": 2412,
      "time_per_iteration": 2.638899087905884
    },
    {
      "auxiliary_loss_clip": 0.0114646,
      "auxiliary_loss_mlp": 0.01051787,
      "balance_loss_clip": 1.05542278,
      "balance_loss_mlp": 1.03167605,
      "epoch": 0.14507740868781002,
      "flos": 15596112648960.0,
      "grad_norm": 2.0913156440252902,
      "language_loss": 0.76611447,
      "learning_rate": 3.862684373853579e-06,
      "loss": 0.78809696,
      "num_input_tokens_seen": 52239690,
      "step": 2413,
      "time_per_iteration": 2.4719831943511963
    },
    {
      "auxiliary_loss_clip": 0.01079824,
      "auxiliary_loss_mlp": 0.01025002,
      "balance_loss_clip": 1.04588723,
      "balance_loss_mlp": 1.02161682,
      "epoch": 0.145137531940478,
      "flos": 66675343438080.0,
      "grad_norm": 0.9007164244795998,
      "language_loss": 0.58844322,
      "learning_rate": 3.8625425179337656e-06,
      "loss": 0.60949147,
      "num_input_tokens_seen": 52296705,
      "step": 2414,
      "time_per_iteration": 3.0326743125915527
    },
    {
      "auxiliary_loss_clip": 0.01066175,
      "auxiliary_loss_mlp": 0.01007473,
      "balance_loss_clip": 1.04564226,
      "balance_loss_mlp": 1.00382519,
      "epoch": 0.14519765519314595,
      "flos": 67521578929920.0,
      "grad_norm": 0.8384556257576324,
      "language_loss": 0.62196672,
      "learning_rate": 3.862400591386154e-06,
      "loss": 0.64270318,
      "num_input_tokens_seen": 52361830,
      "step": 2415,
      "time_per_iteration": 3.1442854404449463
    },
    {
      "auxiliary_loss_clip": 0.01147089,
      "auxiliary_loss_mlp": 0.01046561,
      "balance_loss_clip": 1.05518508,
      "balance_loss_mlp": 1.02602124,
      "epoch": 0.14525777844581392,
      "flos": 17198913329280.0,
      "grad_norm": 2.086668956632305,
      "language_loss": 0.72539961,
      "learning_rate": 3.8622585942161245e-06,
      "loss": 0.74733615,
      "num_input_tokens_seen": 52379420,
      "step": 2416,
      "time_per_iteration": 2.479311466217041
    },
    {
      "auxiliary_loss_clip": 0.01050623,
      "auxiliary_loss_mlp": 0.01004478,
      "balance_loss_clip": 1.03345704,
      "balance_loss_mlp": 1.00111604,
      "epoch": 0.14531790169848188,
      "flos": 65404609015680.0,
      "grad_norm": 0.7160943941984923,
      "language_loss": 0.60390902,
      "learning_rate": 3.8621165264290635e-06,
      "loss": 0.62445998,
      "num_input_tokens_seen": 52446290,
      "step": 2417,
      "time_per_iteration": 3.2052416801452637
    },
    {
      "auxiliary_loss_clip": 0.01165366,
      "auxiliary_loss_mlp": 0.01052089,
      "balance_loss_clip": 1.05416167,
      "balance_loss_mlp": 1.03216887,
      "epoch": 0.14537802495114985,
      "flos": 32562467372160.0,
      "grad_norm": 2.3744759204549495,
      "language_loss": 0.78859735,
      "learning_rate": 3.861974388030356e-06,
      "loss": 0.81077188,
      "num_input_tokens_seen": 52467295,
      "step": 2418,
      "time_per_iteration": 2.566204786300659
    },
    {
      "auxiliary_loss_clip": 0.01115758,
      "auxiliary_loss_mlp": 0.01050018,
      "balance_loss_clip": 1.0517683,
      "balance_loss_mlp": 1.03053975,
      "epoch": 0.1454381482038178,
      "flos": 20226685432320.0,
      "grad_norm": 1.770108998851477,
      "language_loss": 0.72045422,
      "learning_rate": 3.861832179025394e-06,
      "loss": 0.74211198,
      "num_input_tokens_seen": 52487295,
      "step": 2419,
      "time_per_iteration": 2.5769503116607666
    },
    {
      "auxiliary_loss_clip": 0.01139763,
      "auxiliary_loss_mlp": 0.01046222,
      "balance_loss_clip": 1.05569839,
      "balance_loss_mlp": 1.0264926,
      "epoch": 0.1454982714564858,
      "flos": 22893124671360.0,
      "grad_norm": 2.3018782583419393,
      "language_loss": 0.89662445,
      "learning_rate": 3.861689899419569e-06,
      "loss": 0.91848433,
      "num_input_tokens_seen": 52504220,
      "step": 2420,
      "time_per_iteration": 2.534346342086792
    },
    {
      "auxiliary_loss_clip": 0.01150702,
      "auxiliary_loss_mlp": 0.01053266,
      "balance_loss_clip": 1.05225205,
      "balance_loss_mlp": 1.03446674,
      "epoch": 0.14555839470915377,
      "flos": 20229845829120.0,
      "grad_norm": 2.1373906837566974,
      "language_loss": 0.83055341,
      "learning_rate": 3.861547549218276e-06,
      "loss": 0.85259306,
      "num_input_tokens_seen": 52521900,
      "step": 2421,
      "time_per_iteration": 2.516117811203003
    },
    {
      "auxiliary_loss_clip": 0.01104176,
      "auxiliary_loss_mlp": 0.01054217,
      "balance_loss_clip": 1.0535326,
      "balance_loss_mlp": 1.03450036,
      "epoch": 0.14561851796182174,
      "flos": 22236282616320.0,
      "grad_norm": 1.8113527024700427,
      "language_loss": 0.8173281,
      "learning_rate": 3.861405128426914e-06,
      "loss": 0.83891213,
      "num_input_tokens_seen": 52540495,
      "step": 2422,
      "time_per_iteration": 2.6409966945648193
    },
    {
      "auxiliary_loss_clip": 0.01050838,
      "auxiliary_loss_mlp": 0.01339503,
      "balance_loss_clip": 1.04621887,
      "balance_loss_mlp": 2.01205993,
      "epoch": 0.1456786412144897,
      "flos": 52636786289280.0,
      "grad_norm": 0.9469547979371113,
      "language_loss": 0.63309002,
      "learning_rate": 3.861262637050883e-06,
      "loss": 0.65699339,
      "num_input_tokens_seen": 52603305,
      "step": 2423,
      "time_per_iteration": 3.209451675415039
    },
    {
      "auxiliary_loss_clip": 0.01109929,
      "auxiliary_loss_mlp": 0.00839185,
      "balance_loss_clip": 1.05773211,
      "balance_loss_mlp": 1.08094025,
      "epoch": 0.14573876446715767,
      "flos": 23221671396480.0,
      "grad_norm": 1.4795298307862137,
      "language_loss": 0.82603246,
      "learning_rate": 3.861120075095585e-06,
      "loss": 0.8455236,
      "num_input_tokens_seen": 52623435,
      "step": 2424,
      "time_per_iteration": 2.652581214904785
    },
    {
      "auxiliary_loss_clip": 0.01140892,
      "auxiliary_loss_mlp": 0.0105159,
      "balance_loss_clip": 1.0559752,
      "balance_loss_mlp": 1.03276658,
      "epoch": 0.14579888771982563,
      "flos": 18114384286080.0,
      "grad_norm": 2.440243862558632,
      "language_loss": 0.78948212,
      "learning_rate": 3.860977442566429e-06,
      "loss": 0.81140697,
      "num_input_tokens_seen": 52642255,
      "step": 2425,
      "time_per_iteration": 2.578481912612915
    },
    {
      "auxiliary_loss_clip": 0.01154443,
      "auxiliary_loss_mlp": 0.01050828,
      "balance_loss_clip": 1.05824244,
      "balance_loss_mlp": 1.03236294,
      "epoch": 0.14585901097249362,
      "flos": 23001107932800.0,
      "grad_norm": 2.3641579815089186,
      "language_loss": 0.8323037,
      "learning_rate": 3.860834739468821e-06,
      "loss": 0.85435641,
      "num_input_tokens_seen": 52658700,
      "step": 2426,
      "time_per_iteration": 3.9456679821014404
    },
    {
      "auxiliary_loss_clip": 0.01163995,
      "auxiliary_loss_mlp": 0.01048136,
      "balance_loss_clip": 1.0582515,
      "balance_loss_mlp": 1.02945542,
      "epoch": 0.1459191342251616,
      "flos": 21908669644800.0,
      "grad_norm": 1.920247844892834,
      "language_loss": 0.87402934,
      "learning_rate": 3.860691965808173e-06,
      "loss": 0.89615065,
      "num_input_tokens_seen": 52678140,
      "step": 2427,
      "time_per_iteration": 3.8972432613372803
    },
    {
      "auxiliary_loss_clip": 0.01126386,
      "auxiliary_loss_mlp": 0.01051955,
      "balance_loss_clip": 1.05249596,
      "balance_loss_mlp": 1.03005588,
      "epoch": 0.14597925747782955,
      "flos": 14975504438400.0,
      "grad_norm": 2.9763157749945433,
      "language_loss": 0.67227399,
      "learning_rate": 3.8605491215899e-06,
      "loss": 0.69405746,
      "num_input_tokens_seen": 52696825,
      "step": 2428,
      "time_per_iteration": 2.5681796073913574
    },
    {
      "auxiliary_loss_clip": 0.01154114,
      "auxiliary_loss_mlp": 0.01049116,
      "balance_loss_clip": 1.05558991,
      "balance_loss_mlp": 1.0294466,
      "epoch": 0.14603938073049752,
      "flos": 21068898600960.0,
      "grad_norm": 2.237858708414702,
      "language_loss": 0.83956933,
      "learning_rate": 3.860406206819417e-06,
      "loss": 0.86160159,
      "num_input_tokens_seen": 52715125,
      "step": 2429,
      "time_per_iteration": 2.551254987716675
    },
    {
      "auxiliary_loss_clip": 0.0111886,
      "auxiliary_loss_mlp": 0.01049805,
      "balance_loss_clip": 1.0486846,
      "balance_loss_mlp": 1.03144646,
      "epoch": 0.14609950398316549,
      "flos": 19864777950720.0,
      "grad_norm": 1.8981043785102403,
      "language_loss": 0.79172194,
      "learning_rate": 3.860263221502145e-06,
      "loss": 0.81340861,
      "num_input_tokens_seen": 52734015,
      "step": 2430,
      "time_per_iteration": 3.9648454189300537
    },
    {
      "auxiliary_loss_clip": 0.01166718,
      "auxiliary_loss_mlp": 0.01046078,
      "balance_loss_clip": 1.0578624,
      "balance_loss_mlp": 1.02705181,
      "epoch": 0.14615962723583345,
      "flos": 22418852469120.0,
      "grad_norm": 2.0027946257854152,
      "language_loss": 0.82810324,
      "learning_rate": 3.860120165643504e-06,
      "loss": 0.85023123,
      "num_input_tokens_seen": 52753025,
      "step": 2431,
      "time_per_iteration": 2.5266339778900146
    },
    {
      "auxiliary_loss_clip": 0.01158487,
      "auxiliary_loss_mlp": 0.01052292,
      "balance_loss_clip": 1.05544055,
      "balance_loss_mlp": 1.03208649,
      "epoch": 0.14621975048850142,
      "flos": 22346241125760.0,
      "grad_norm": 1.7962107326834569,
      "language_loss": 0.79103637,
      "learning_rate": 3.859977039248921e-06,
      "loss": 0.81314421,
      "num_input_tokens_seen": 52773420,
      "step": 2432,
      "time_per_iteration": 2.523865222930908
    },
    {
      "auxiliary_loss_clip": 0.01161695,
      "auxiliary_loss_mlp": 0.00987369,
      "balance_loss_clip": 1.05491257,
      "balance_loss_mlp": 1.34084308,
      "epoch": 0.1462798737411694,
      "flos": 24389163152640.0,
      "grad_norm": 2.3494882174115386,
      "language_loss": 0.80157524,
      "learning_rate": 3.859833842323822e-06,
      "loss": 0.82306588,
      "num_input_tokens_seen": 52792870,
      "step": 2433,
      "time_per_iteration": 3.980318307876587
    },
    {
      "auxiliary_loss_clip": 0.01124782,
      "auxiliary_loss_mlp": 0.01052281,
      "balance_loss_clip": 1.05803275,
      "balance_loss_mlp": 1.03181303,
      "epoch": 0.14633999699383737,
      "flos": 19244672530560.0,
      "grad_norm": 3.243569300121905,
      "language_loss": 0.78697789,
      "learning_rate": 3.859690574873638e-06,
      "loss": 0.80874854,
      "num_input_tokens_seen": 52811615,
      "step": 2434,
      "time_per_iteration": 2.5750107765197754
    },
    {
      "auxiliary_loss_clip": 0.01053702,
      "auxiliary_loss_mlp": 0.01047445,
      "balance_loss_clip": 1.03591347,
      "balance_loss_mlp": 1.04434586,
      "epoch": 0.14640012024650534,
      "flos": 62660638270080.0,
      "grad_norm": 0.8613434107927777,
      "language_loss": 0.58438826,
      "learning_rate": 3.8595472369038e-06,
      "loss": 0.60539973,
      "num_input_tokens_seen": 52873230,
      "step": 2435,
      "time_per_iteration": 3.1452882289886475
    },
    {
      "auxiliary_loss_clip": 0.01154763,
      "auxiliary_loss_mlp": 0.01048896,
      "balance_loss_clip": 1.04962206,
      "balance_loss_mlp": 1.02990627,
      "epoch": 0.1464602434991733,
      "flos": 12276243146880.0,
      "grad_norm": 2.0919215929649906,
      "language_loss": 0.88277102,
      "learning_rate": 3.859403828419744e-06,
      "loss": 0.90480763,
      "num_input_tokens_seen": 52889325,
      "step": 2436,
      "time_per_iteration": 2.4759373664855957
    },
    {
      "auxiliary_loss_clip": 0.01150011,
      "auxiliary_loss_mlp": 0.00981899,
      "balance_loss_clip": 1.05321515,
      "balance_loss_mlp": 1.33691466,
      "epoch": 0.14652036675184127,
      "flos": 20922311197440.0,
      "grad_norm": 3.740906893719455,
      "language_loss": 0.74748564,
      "learning_rate": 3.85926034942691e-06,
      "loss": 0.76880467,
      "num_input_tokens_seen": 52909705,
      "step": 2437,
      "time_per_iteration": 2.5262320041656494
    },
    {
      "auxiliary_loss_clip": 0.0115998,
      "auxiliary_loss_mlp": 0.01048482,
      "balance_loss_clip": 1.0510428,
      "balance_loss_mlp": 1.02728701,
      "epoch": 0.14658049000450923,
      "flos": 27703681528320.0,
      "grad_norm": 2.196523610987453,
      "language_loss": 0.73657143,
      "learning_rate": 3.859116799930736e-06,
      "loss": 0.75865602,
      "num_input_tokens_seen": 52930300,
      "step": 2438,
      "time_per_iteration": 2.5316321849823
    },
    {
      "auxiliary_loss_clip": 0.01149526,
      "auxiliary_loss_mlp": 0.01041297,
      "balance_loss_clip": 1.0552845,
      "balance_loss_mlp": 1.02271271,
      "epoch": 0.14664061325717723,
      "flos": 24936513575040.0,
      "grad_norm": 1.9644113073953406,
      "language_loss": 0.74373412,
      "learning_rate": 3.858973179936668e-06,
      "loss": 0.76564229,
      "num_input_tokens_seen": 52949955,
      "step": 2439,
      "time_per_iteration": 2.5277037620544434
    },
    {
      "auxiliary_loss_clip": 0.01144026,
      "auxiliary_loss_mlp": 0.01042637,
      "balance_loss_clip": 1.05004454,
      "balance_loss_mlp": 1.02306294,
      "epoch": 0.1467007365098452,
      "flos": 40297661406720.0,
      "grad_norm": 2.770591443615531,
      "language_loss": 0.74084342,
      "learning_rate": 3.85882948945015e-06,
      "loss": 0.76271009,
      "num_input_tokens_seen": 52972905,
      "step": 2440,
      "time_per_iteration": 2.64603590965271
    },
    {
      "auxiliary_loss_clip": 0.01153334,
      "auxiliary_loss_mlp": 0.01049961,
      "balance_loss_clip": 1.04913855,
      "balance_loss_mlp": 1.03112578,
      "epoch": 0.14676085976251316,
      "flos": 26541074021760.0,
      "grad_norm": 1.658176913543238,
      "language_loss": 0.83039391,
      "learning_rate": 3.85868572847663e-06,
      "loss": 0.85242689,
      "num_input_tokens_seen": 52994850,
      "step": 2441,
      "time_per_iteration": 2.5142879486083984
    },
    {
      "auxiliary_loss_clip": 0.01149052,
      "auxiliary_loss_mlp": 0.01047132,
      "balance_loss_clip": 1.05323052,
      "balance_loss_mlp": 1.02632987,
      "epoch": 0.14682098301518112,
      "flos": 23550110380800.0,
      "grad_norm": 2.4767375342427544,
      "language_loss": 0.72192669,
      "learning_rate": 3.858541897021563e-06,
      "loss": 0.74388856,
      "num_input_tokens_seen": 53014740,
      "step": 2442,
      "time_per_iteration": 2.558194875717163
    },
    {
      "auxiliary_loss_clip": 0.0113025,
      "auxiliary_loss_mlp": 0.01041552,
      "balance_loss_clip": 1.05616021,
      "balance_loss_mlp": 1.02137029,
      "epoch": 0.1468811062678491,
      "flos": 11651073909120.0,
      "grad_norm": 3.1335927081817148,
      "language_loss": 0.8168655,
      "learning_rate": 3.8583979950904e-06,
      "loss": 0.83858347,
      "num_input_tokens_seen": 53029780,
      "step": 2443,
      "time_per_iteration": 2.6117029190063477
    },
    {
      "auxiliary_loss_clip": 0.01139413,
      "auxiliary_loss_mlp": 0.01062802,
      "balance_loss_clip": 1.04982018,
      "balance_loss_mlp": 1.04197621,
      "epoch": 0.14694122952051705,
      "flos": 23002616304000.0,
      "grad_norm": 1.6954272839268876,
      "language_loss": 0.82737553,
      "learning_rate": 3.858254022688599e-06,
      "loss": 0.84939766,
      "num_input_tokens_seen": 53048620,
      "step": 2444,
      "time_per_iteration": 2.5117576122283936
    },
    {
      "auxiliary_loss_clip": 0.01127608,
      "auxiliary_loss_mlp": 0.01053611,
      "balance_loss_clip": 1.05170977,
      "balance_loss_mlp": 1.03383398,
      "epoch": 0.14700135277318502,
      "flos": 26502972670080.0,
      "grad_norm": 1.7301628132893199,
      "language_loss": 0.71333551,
      "learning_rate": 3.85810997982162e-06,
      "loss": 0.73514766,
      "num_input_tokens_seen": 53070055,
      "step": 2445,
      "time_per_iteration": 2.568983316421509
    },
    {
      "auxiliary_loss_clip": 0.0107602,
      "auxiliary_loss_mlp": 0.01018441,
      "balance_loss_clip": 1.03388548,
      "balance_loss_mlp": 1.01546121,
      "epoch": 0.147061476025853,
      "flos": 59449434387840.0,
      "grad_norm": 0.8300953499734267,
      "language_loss": 0.63046026,
      "learning_rate": 3.857965866494923e-06,
      "loss": 0.65140486,
      "num_input_tokens_seen": 53126945,
      "step": 2446,
      "time_per_iteration": 2.9828553199768066
    },
    {
      "auxiliary_loss_clip": 0.01116569,
      "auxiliary_loss_mlp": 0.01045948,
      "balance_loss_clip": 1.0556643,
      "balance_loss_mlp": 1.0254674,
      "epoch": 0.14712159927852098,
      "flos": 28330897841280.0,
      "grad_norm": 1.7031534385646947,
      "language_loss": 0.74843848,
      "learning_rate": 3.857821682713975e-06,
      "loss": 0.77006364,
      "num_input_tokens_seen": 53149130,
      "step": 2447,
      "time_per_iteration": 2.6760611534118652
    },
    {
      "auxiliary_loss_clip": 0.01157768,
      "auxiliary_loss_mlp": 0.0104316,
      "balance_loss_clip": 1.0513618,
      "balance_loss_mlp": 1.02423012,
      "epoch": 0.14718172253118894,
      "flos": 27089825074560.0,
      "grad_norm": 2.517532151401068,
      "language_loss": 0.85507625,
      "learning_rate": 3.857677428484242e-06,
      "loss": 0.87708557,
      "num_input_tokens_seen": 53167120,
      "step": 2448,
      "time_per_iteration": 2.509150981903076
    },
    {
      "auxiliary_loss_clip": 0.01071871,
      "auxiliary_loss_mlp": 0.01005493,
      "balance_loss_clip": 1.02900696,
      "balance_loss_mlp": 1.00278652,
      "epoch": 0.1472418457838569,
      "flos": 66706764860160.0,
      "grad_norm": 0.7692393740652745,
      "language_loss": 0.56816018,
      "learning_rate": 3.857533103811195e-06,
      "loss": 0.58893383,
      "num_input_tokens_seen": 53227945,
      "step": 2449,
      "time_per_iteration": 3.0389516353607178
    },
    {
      "auxiliary_loss_clip": 0.01131226,
      "auxiliary_loss_mlp": 0.01044375,
      "balance_loss_clip": 1.05485439,
      "balance_loss_mlp": 1.02397823,
      "epoch": 0.14730196903652487,
      "flos": 19573578391680.0,
      "grad_norm": 1.874171497059279,
      "language_loss": 0.8522138,
      "learning_rate": 3.857388708700307e-06,
      "loss": 0.87396979,
      "num_input_tokens_seen": 53244615,
      "step": 2450,
      "time_per_iteration": 2.524604558944702
    },
    {
      "auxiliary_loss_clip": 0.01150763,
      "auxiliary_loss_mlp": 0.01048961,
      "balance_loss_clip": 1.05203807,
      "balance_loss_mlp": 1.02888668,
      "epoch": 0.14736209228919284,
      "flos": 16071031296000.0,
      "grad_norm": 1.870008240168326,
      "language_loss": 0.74964827,
      "learning_rate": 3.857244243157052e-06,
      "loss": 0.77164555,
      "num_input_tokens_seen": 53262205,
      "step": 2451,
      "time_per_iteration": 2.5117850303649902
    },
    {
      "auxiliary_loss_clip": 0.01130636,
      "auxiliary_loss_mlp": 0.01041167,
      "balance_loss_clip": 1.05098557,
      "balance_loss_mlp": 1.02261829,
      "epoch": 0.1474222155418608,
      "flos": 23039460679680.0,
      "grad_norm": 1.6580491146596554,
      "language_loss": 0.81963849,
      "learning_rate": 3.85709970718691e-06,
      "loss": 0.84135652,
      "num_input_tokens_seen": 53282445,
      "step": 2452,
      "time_per_iteration": 2.5508720874786377
    },
    {
      "auxiliary_loss_clip": 0.01096191,
      "auxiliary_loss_mlp": 0.01036407,
      "balance_loss_clip": 1.0548861,
      "balance_loss_mlp": 1.01807296,
      "epoch": 0.1474823387945288,
      "flos": 17018641946880.0,
      "grad_norm": 1.5109463044096114,
      "language_loss": 0.74402618,
      "learning_rate": 3.856955100795361e-06,
      "loss": 0.76535213,
      "num_input_tokens_seen": 53299060,
      "step": 2453,
      "time_per_iteration": 2.6646926403045654
    },
    {
      "auxiliary_loss_clip": 0.01137097,
      "auxiliary_loss_mlp": 0.01051882,
      "balance_loss_clip": 1.05282497,
      "balance_loss_mlp": 1.03237963,
      "epoch": 0.14754246204719676,
      "flos": 17895041884800.0,
      "grad_norm": 2.1549545530372907,
      "language_loss": 0.76012897,
      "learning_rate": 3.856810423987889e-06,
      "loss": 0.78201866,
      "num_input_tokens_seen": 53315970,
      "step": 2454,
      "time_per_iteration": 2.510106325149536
    },
    {
      "auxiliary_loss_clip": 0.01137557,
      "auxiliary_loss_mlp": 0.01041679,
      "balance_loss_clip": 1.05019379,
      "balance_loss_mlp": 1.0220809,
      "epoch": 0.14760258529986472,
      "flos": 13079097987840.0,
      "grad_norm": 2.265577839602067,
      "language_loss": 0.82894778,
      "learning_rate": 3.856665676769979e-06,
      "loss": 0.85074019,
      "num_input_tokens_seen": 53332940,
      "step": 2455,
      "time_per_iteration": 2.533565044403076
    },
    {
      "auxiliary_loss_clip": 0.0112294,
      "auxiliary_loss_mlp": 0.01053676,
      "balance_loss_clip": 1.04720056,
      "balance_loss_mlp": 1.03449559,
      "epoch": 0.1476627085525327,
      "flos": 30806399358720.0,
      "grad_norm": 1.8823770492251262,
      "language_loss": 0.84305835,
      "learning_rate": 3.85652085914712e-06,
      "loss": 0.86482441,
      "num_input_tokens_seen": 53353295,
      "step": 2456,
      "time_per_iteration": 2.655630350112915
    },
    {
      "auxiliary_loss_clip": 0.01146791,
      "auxiliary_loss_mlp": 0.01045621,
      "balance_loss_clip": 1.05030489,
      "balance_loss_mlp": 1.02690518,
      "epoch": 0.14772283180520066,
      "flos": 21689434984320.0,
      "grad_norm": 1.8060540515808403,
      "language_loss": 0.8435986,
      "learning_rate": 3.856375971124805e-06,
      "loss": 0.86552274,
      "num_input_tokens_seen": 53373410,
      "step": 2457,
      "time_per_iteration": 2.5273449420928955
    },
    {
      "auxiliary_loss_clip": 0.01142774,
      "auxiliary_loss_mlp": 0.0104232,
      "balance_loss_clip": 1.04881418,
      "balance_loss_mlp": 1.02429569,
      "epoch": 0.14778295505786862,
      "flos": 18770400328320.0,
      "grad_norm": 1.9838495513220475,
      "language_loss": 0.75447989,
      "learning_rate": 3.856231012708527e-06,
      "loss": 0.77633083,
      "num_input_tokens_seen": 53391430,
      "step": 2458,
      "time_per_iteration": 2.467681884765625
    },
    {
      "auxiliary_loss_clip": 0.01110542,
      "auxiliary_loss_mlp": 0.01052805,
      "balance_loss_clip": 1.0506289,
      "balance_loss_mlp": 1.03163338,
      "epoch": 0.1478430783105366,
      "flos": 22893555634560.0,
      "grad_norm": 1.834634078240255,
      "language_loss": 0.83709669,
      "learning_rate": 3.856085983903782e-06,
      "loss": 0.85873014,
      "num_input_tokens_seen": 53409960,
      "step": 2459,
      "time_per_iteration": 2.599130392074585
    },
    {
      "auxiliary_loss_clip": 0.01123642,
      "auxiliary_loss_mlp": 0.01041213,
      "balance_loss_clip": 1.04999483,
      "balance_loss_mlp": 1.02315307,
      "epoch": 0.14790320156320458,
      "flos": 15085319293440.0,
      "grad_norm": 2.176376109791904,
      "language_loss": 0.75190997,
      "learning_rate": 3.855940884716071e-06,
      "loss": 0.7735585,
      "num_input_tokens_seen": 53426160,
      "step": 2460,
      "time_per_iteration": 2.4943206310272217
    },
    {
      "auxiliary_loss_clip": 0.01123548,
      "auxiliary_loss_mlp": 0.01043716,
      "balance_loss_clip": 1.04941165,
      "balance_loss_mlp": 1.02422571,
      "epoch": 0.14796332481587254,
      "flos": 26504768350080.0,
      "grad_norm": 1.927374889449342,
      "language_loss": 0.81223053,
      "learning_rate": 3.855795715150896e-06,
      "loss": 0.83390319,
      "num_input_tokens_seen": 53448530,
      "step": 2461,
      "time_per_iteration": 2.6131250858306885
    },
    {
      "auxiliary_loss_clip": 0.01147703,
      "auxiliary_loss_mlp": 0.01049002,
      "balance_loss_clip": 1.04955769,
      "balance_loss_mlp": 1.02865255,
      "epoch": 0.1480234480685405,
      "flos": 17563191108480.0,
      "grad_norm": 2.4233182118796206,
      "language_loss": 0.66306216,
      "learning_rate": 3.855650475213761e-06,
      "loss": 0.68502921,
      "num_input_tokens_seen": 53465915,
      "step": 2462,
      "time_per_iteration": 2.452157735824585
    },
    {
      "auxiliary_loss_clip": 0.01128782,
      "auxiliary_loss_mlp": 0.01049197,
      "balance_loss_clip": 1.04962313,
      "balance_loss_mlp": 1.02872908,
      "epoch": 0.14808357132120847,
      "flos": 53582203232640.0,
      "grad_norm": 2.2887562715226095,
      "language_loss": 0.67260385,
      "learning_rate": 3.8555051649101745e-06,
      "loss": 0.69438362,
      "num_input_tokens_seen": 53496055,
      "step": 2463,
      "time_per_iteration": 2.838212490081787
    },
    {
      "auxiliary_loss_clip": 0.01144863,
      "auxiliary_loss_mlp": 0.01046644,
      "balance_loss_clip": 1.04688454,
      "balance_loss_mlp": 1.0267961,
      "epoch": 0.14814369457387644,
      "flos": 19829190551040.0,
      "grad_norm": 1.8689639233499031,
      "language_loss": 0.7730124,
      "learning_rate": 3.855359784245646e-06,
      "loss": 0.79492748,
      "num_input_tokens_seen": 53513790,
      "step": 2464,
      "time_per_iteration": 2.4797089099884033
    },
    {
      "auxiliary_loss_clip": 0.01130844,
      "auxiliary_loss_mlp": 0.01053821,
      "balance_loss_clip": 1.0510639,
      "balance_loss_mlp": 1.03466368,
      "epoch": 0.1482038178265444,
      "flos": 23914962777600.0,
      "grad_norm": 1.6356893213162957,
      "language_loss": 0.79758656,
      "learning_rate": 3.855214333225688e-06,
      "loss": 0.81943321,
      "num_input_tokens_seen": 53533410,
      "step": 2465,
      "time_per_iteration": 3.919557809829712
    },
    {
      "auxiliary_loss_clip": 0.01163191,
      "auxiliary_loss_mlp": 0.01043837,
      "balance_loss_clip": 1.05319953,
      "balance_loss_mlp": 1.02433479,
      "epoch": 0.1482639410792124,
      "flos": 24170503109760.0,
      "grad_norm": 1.9029268506611359,
      "language_loss": 0.76246303,
      "learning_rate": 3.855068811855817e-06,
      "loss": 0.78453332,
      "num_input_tokens_seen": 53554775,
      "step": 2466,
      "time_per_iteration": 3.8749442100524902
    },
    {
      "auxiliary_loss_clip": 0.01037272,
      "auxiliary_loss_mlp": 0.01014812,
      "balance_loss_clip": 1.05620003,
      "balance_loss_mlp": 1.01156926,
      "epoch": 0.14832406433188036,
      "flos": 66191051341440.0,
      "grad_norm": 0.7998652644299296,
      "language_loss": 0.60061216,
      "learning_rate": 3.854923220141551e-06,
      "loss": 0.62113297,
      "num_input_tokens_seen": 53609675,
      "step": 2467,
      "time_per_iteration": 3.2249529361724854
    },
    {
      "auxiliary_loss_clip": 0.01135875,
      "auxiliary_loss_mlp": 0.0104135,
      "balance_loss_clip": 1.04920852,
      "balance_loss_mlp": 1.02133441,
      "epoch": 0.14838418758454833,
      "flos": 25411252654080.0,
      "grad_norm": 2.085809278772307,
      "language_loss": 0.87784356,
      "learning_rate": 3.85477755808841e-06,
      "loss": 0.89961576,
      "num_input_tokens_seen": 53626950,
      "step": 2468,
      "time_per_iteration": 2.5537025928497314
    },
    {
      "auxiliary_loss_clip": 0.01128679,
      "auxiliary_loss_mlp": 0.010492,
      "balance_loss_clip": 1.05012918,
      "balance_loss_mlp": 1.02901793,
      "epoch": 0.1484443108372163,
      "flos": 23289901280640.0,
      "grad_norm": 2.075383674593907,
      "language_loss": 0.76332307,
      "learning_rate": 3.854631825701919e-06,
      "loss": 0.78510183,
      "num_input_tokens_seen": 53644200,
      "step": 2469,
      "time_per_iteration": 3.9496967792510986
    },
    {
      "auxiliary_loss_clip": 0.01122919,
      "auxiliary_loss_mlp": 0.01042869,
      "balance_loss_clip": 1.04995394,
      "balance_loss_mlp": 1.0239265,
      "epoch": 0.14850443408988426,
      "flos": 14647675985280.0,
      "grad_norm": 2.7376396953558313,
      "language_loss": 0.75922924,
      "learning_rate": 3.854486022987603e-06,
      "loss": 0.78088707,
      "num_input_tokens_seen": 53659650,
      "step": 2470,
      "time_per_iteration": 2.531712055206299
    },
    {
      "auxiliary_loss_clip": 0.01154646,
      "auxiliary_loss_mlp": 0.01042854,
      "balance_loss_clip": 1.05118227,
      "balance_loss_mlp": 1.02382874,
      "epoch": 0.14856455734255222,
      "flos": 23548314700800.0,
      "grad_norm": 2.2479707314336985,
      "language_loss": 0.72155458,
      "learning_rate": 3.8543401499509905e-06,
      "loss": 0.74352956,
      "num_input_tokens_seen": 53680275,
      "step": 2471,
      "time_per_iteration": 2.4988839626312256
    },
    {
      "auxiliary_loss_clip": 0.01132005,
      "auxiliary_loss_mlp": 0.01051792,
      "balance_loss_clip": 1.05407846,
      "balance_loss_mlp": 1.03114533,
      "epoch": 0.1486246805952202,
      "flos": 18077288515200.0,
      "grad_norm": 1.9727013375037021,
      "language_loss": 0.89569962,
      "learning_rate": 3.854194206597615e-06,
      "loss": 0.91753757,
      "num_input_tokens_seen": 53698270,
      "step": 2472,
      "time_per_iteration": 3.9005980491638184
    },
    {
      "auxiliary_loss_clip": 0.01119007,
      "auxiliary_loss_mlp": 0.01042492,
      "balance_loss_clip": 1.04953325,
      "balance_loss_mlp": 1.02333486,
      "epoch": 0.14868480384788818,
      "flos": 19353625459200.0,
      "grad_norm": 2.500654269965786,
      "language_loss": 0.81027508,
      "learning_rate": 3.854048192933008e-06,
      "loss": 0.83189011,
      "num_input_tokens_seen": 53716845,
      "step": 2473,
      "time_per_iteration": 2.579444646835327
    },
    {
      "auxiliary_loss_clip": 0.01152729,
      "auxiliary_loss_mlp": 0.01054864,
      "balance_loss_clip": 1.05374527,
      "balance_loss_mlp": 1.03589797,
      "epoch": 0.14874492710055615,
      "flos": 22200192426240.0,
      "grad_norm": 2.5779842400356,
      "language_loss": 0.77427024,
      "learning_rate": 3.853902108962709e-06,
      "loss": 0.79634619,
      "num_input_tokens_seen": 53734970,
      "step": 2474,
      "time_per_iteration": 2.5281834602355957
    },
    {
      "auxiliary_loss_clip": 0.01124186,
      "auxiliary_loss_mlp": 0.01055664,
      "balance_loss_clip": 1.05612373,
      "balance_loss_mlp": 1.03547025,
      "epoch": 0.1488050503532241,
      "flos": 21103444506240.0,
      "grad_norm": 2.216483796459402,
      "language_loss": 0.82636696,
      "learning_rate": 3.853755954692255e-06,
      "loss": 0.84816551,
      "num_input_tokens_seen": 53753415,
      "step": 2475,
      "time_per_iteration": 2.5758626461029053
    },
    {
      "auxiliary_loss_clip": 0.01106685,
      "auxiliary_loss_mlp": 0.0105121,
      "balance_loss_clip": 1.05677629,
      "balance_loss_mlp": 1.0323987,
      "epoch": 0.14886517360589208,
      "flos": 12786569625600.0,
      "grad_norm": 2.0308729257005633,
      "language_loss": 0.80298829,
      "learning_rate": 3.85360973012719e-06,
      "loss": 0.8245672,
      "num_input_tokens_seen": 53770305,
      "step": 2476,
      "time_per_iteration": 2.578641891479492
    },
    {
      "auxiliary_loss_clip": 0.0114552,
      "auxiliary_loss_mlp": 0.01046262,
      "balance_loss_clip": 1.05288124,
      "balance_loss_mlp": 1.02866721,
      "epoch": 0.14892529685856004,
      "flos": 29022860419200.0,
      "grad_norm": 2.290198847001989,
      "language_loss": 0.77178276,
      "learning_rate": 3.853463435273058e-06,
      "loss": 0.79370058,
      "num_input_tokens_seen": 53788895,
      "step": 2477,
      "time_per_iteration": 2.5667755603790283
    },
    {
      "auxiliary_loss_clip": 0.01066828,
      "auxiliary_loss_mlp": 0.01063099,
      "balance_loss_clip": 1.04164124,
      "balance_loss_mlp": 1.05990434,
      "epoch": 0.148985420111228,
      "flos": 61926121054080.0,
      "grad_norm": 0.8063715614661756,
      "language_loss": 0.60128599,
      "learning_rate": 3.853317070135407e-06,
      "loss": 0.62258524,
      "num_input_tokens_seen": 53850260,
      "step": 2478,
      "time_per_iteration": 3.1874613761901855
    },
    {
      "auxiliary_loss_clip": 0.01105649,
      "auxiliary_loss_mlp": 0.01042552,
      "balance_loss_clip": 1.05316949,
      "balance_loss_mlp": 1.02458727,
      "epoch": 0.149045543363896,
      "flos": 23915106432000.0,
      "grad_norm": 2.535729457658,
      "language_loss": 0.71423948,
      "learning_rate": 3.853170634719787e-06,
      "loss": 0.73572147,
      "num_input_tokens_seen": 53867520,
      "step": 2479,
      "time_per_iteration": 2.60453200340271
    },
    {
      "auxiliary_loss_clip": 0.01137076,
      "auxiliary_loss_mlp": 0.01044636,
      "balance_loss_clip": 1.05295014,
      "balance_loss_mlp": 1.02553868,
      "epoch": 0.14910566661656396,
      "flos": 23654394541440.0,
      "grad_norm": 1.637759596947358,
      "language_loss": 0.80818486,
      "learning_rate": 3.853024129031751e-06,
      "loss": 0.83000201,
      "num_input_tokens_seen": 53886620,
      "step": 2480,
      "time_per_iteration": 2.5721001625061035
    },
    {
      "auxiliary_loss_clip": 0.01132655,
      "auxiliary_loss_mlp": 0.01045492,
      "balance_loss_clip": 1.05246758,
      "balance_loss_mlp": 1.02622771,
      "epoch": 0.14916578986923193,
      "flos": 20515299212160.0,
      "grad_norm": 2.051030781148257,
      "language_loss": 0.84475863,
      "learning_rate": 3.852877553076854e-06,
      "loss": 0.86654007,
      "num_input_tokens_seen": 53902230,
      "step": 2481,
      "time_per_iteration": 2.5518529415130615
    },
    {
      "auxiliary_loss_clip": 0.01144685,
      "auxiliary_loss_mlp": 0.01051144,
      "balance_loss_clip": 1.0528264,
      "balance_loss_mlp": 1.03058028,
      "epoch": 0.1492259131218999,
      "flos": 22491822948480.0,
      "grad_norm": 2.156740679571822,
      "language_loss": 0.77747107,
      "learning_rate": 3.8527309068606546e-06,
      "loss": 0.7994293,
      "num_input_tokens_seen": 53919475,
      "step": 2482,
      "time_per_iteration": 2.517514944076538
    },
    {
      "auxiliary_loss_clip": 0.01134843,
      "auxiliary_loss_mlp": 0.01042443,
      "balance_loss_clip": 1.05538678,
      "balance_loss_mlp": 1.02146244,
      "epoch": 0.14928603637456786,
      "flos": 23185868515200.0,
      "grad_norm": 2.1340305139684514,
      "language_loss": 0.78995121,
      "learning_rate": 3.852584190388713e-06,
      "loss": 0.81172407,
      "num_input_tokens_seen": 53939150,
      "step": 2483,
      "time_per_iteration": 2.59138560295105
    },
    {
      "auxiliary_loss_clip": 0.01144995,
      "auxiliary_loss_mlp": 0.00924096,
      "balance_loss_clip": 1.05083609,
      "balance_loss_mlp": 1.22345781,
      "epoch": 0.14934615962723582,
      "flos": 21653237053440.0,
      "grad_norm": 1.6784635016747396,
      "language_loss": 0.7075941,
      "learning_rate": 3.852437403666595e-06,
      "loss": 0.72828501,
      "num_input_tokens_seen": 53958735,
      "step": 2484,
      "time_per_iteration": 2.521122694015503
    },
    {
      "auxiliary_loss_clip": 0.01135644,
      "auxiliary_loss_mlp": 0.00934919,
      "balance_loss_clip": 1.05124986,
      "balance_loss_mlp": 1.24374747,
      "epoch": 0.1494062828799038,
      "flos": 27010066924800.0,
      "grad_norm": 1.9318308901317545,
      "language_loss": 0.84831786,
      "learning_rate": 3.852290546699863e-06,
      "loss": 0.8690235,
      "num_input_tokens_seen": 53975065,
      "step": 2485,
      "time_per_iteration": 2.5630247592926025
    },
    {
      "auxiliary_loss_clip": 0.01140593,
      "auxiliary_loss_mlp": 0.01047117,
      "balance_loss_clip": 1.05318594,
      "balance_loss_mlp": 1.02663684,
      "epoch": 0.14946640613257178,
      "flos": 21214947300480.0,
      "grad_norm": 2.1147801382617932,
      "language_loss": 0.85118562,
      "learning_rate": 3.8521436194940894e-06,
      "loss": 0.87306267,
      "num_input_tokens_seen": 53993330,
      "step": 2486,
      "time_per_iteration": 2.509166955947876
    },
    {
      "auxiliary_loss_clip": 0.0114271,
      "auxiliary_loss_mlp": 0.01042267,
      "balance_loss_clip": 1.04925942,
      "balance_loss_mlp": 1.02541041,
      "epoch": 0.14952652938523975,
      "flos": 13370872164480.0,
      "grad_norm": 2.0955883566401696,
      "language_loss": 0.74721968,
      "learning_rate": 3.851996622054842e-06,
      "loss": 0.76906949,
      "num_input_tokens_seen": 54010515,
      "step": 2487,
      "time_per_iteration": 2.4926884174346924
    },
    {
      "auxiliary_loss_clip": 0.01145746,
      "auxiliary_loss_mlp": 0.01046898,
      "balance_loss_clip": 1.04930615,
      "balance_loss_mlp": 1.02846837,
      "epoch": 0.1495866526379077,
      "flos": 35517699959040.0,
      "grad_norm": 2.0762590093721163,
      "language_loss": 0.71766913,
      "learning_rate": 3.8518495543877e-06,
      "loss": 0.73959559,
      "num_input_tokens_seen": 54031315,
      "step": 2488,
      "time_per_iteration": 2.627891778945923
    },
    {
      "auxiliary_loss_clip": 0.01134993,
      "auxiliary_loss_mlp": 0.01048274,
      "balance_loss_clip": 1.05414379,
      "balance_loss_mlp": 1.02934313,
      "epoch": 0.14964677589057568,
      "flos": 17632749795840.0,
      "grad_norm": 4.110003360833931,
      "language_loss": 0.70598185,
      "learning_rate": 3.851702416498235e-06,
      "loss": 0.7278145,
      "num_input_tokens_seen": 54045965,
      "step": 2489,
      "time_per_iteration": 2.513213634490967
    },
    {
      "auxiliary_loss_clip": 0.01133772,
      "auxiliary_loss_mlp": 0.01048261,
      "balance_loss_clip": 1.04835331,
      "balance_loss_mlp": 1.02904451,
      "epoch": 0.14970689914324364,
      "flos": 20185280029440.0,
      "grad_norm": 4.649488786577494,
      "language_loss": 0.81687176,
      "learning_rate": 3.8515552083920295e-06,
      "loss": 0.83869207,
      "num_input_tokens_seen": 54059960,
      "step": 2490,
      "time_per_iteration": 2.507173776626587
    },
    {
      "auxiliary_loss_clip": 0.01123518,
      "auxiliary_loss_mlp": 0.01046458,
      "balance_loss_clip": 1.05591297,
      "balance_loss_mlp": 1.0271697,
      "epoch": 0.1497670223959116,
      "flos": 37228699382400.0,
      "grad_norm": 2.167732959182962,
      "language_loss": 0.80146402,
      "learning_rate": 3.851407930074666e-06,
      "loss": 0.82316381,
      "num_input_tokens_seen": 54079330,
      "step": 2491,
      "time_per_iteration": 2.715275526046753
    },
    {
      "auxiliary_loss_clip": 0.01140685,
      "auxiliary_loss_mlp": 0.01053055,
      "balance_loss_clip": 1.0504148,
      "balance_loss_mlp": 1.03179955,
      "epoch": 0.1498271456485796,
      "flos": 24455848752000.0,
      "grad_norm": 2.064398087951897,
      "language_loss": 0.90649939,
      "learning_rate": 3.851260581551727e-06,
      "loss": 0.92843676,
      "num_input_tokens_seen": 54097555,
      "step": 2492,
      "time_per_iteration": 2.518488645553589
    },
    {
      "auxiliary_loss_clip": 0.01148818,
      "auxiliary_loss_mlp": 0.01053491,
      "balance_loss_clip": 1.05246234,
      "balance_loss_mlp": 1.03411913,
      "epoch": 0.14988726890124757,
      "flos": 16253601148800.0,
      "grad_norm": 3.583506721024229,
      "language_loss": 0.79328847,
      "learning_rate": 3.851113162828802e-06,
      "loss": 0.81531155,
      "num_input_tokens_seen": 54115600,
      "step": 2493,
      "time_per_iteration": 2.505173444747925
    },
    {
      "auxiliary_loss_clip": 0.01148176,
      "auxiliary_loss_mlp": 0.01045096,
      "balance_loss_clip": 1.05128086,
      "balance_loss_mlp": 1.02510512,
      "epoch": 0.14994739215391553,
      "flos": 20666555383680.0,
      "grad_norm": 2.5900376005730488,
      "language_loss": 0.80314076,
      "learning_rate": 3.85096567391148e-06,
      "loss": 0.82507348,
      "num_input_tokens_seen": 54135220,
      "step": 2494,
      "time_per_iteration": 2.523502826690674
    },
    {
      "auxiliary_loss_clip": 0.0113976,
      "auxiliary_loss_mlp": 0.010477,
      "balance_loss_clip": 1.05485439,
      "balance_loss_mlp": 1.0276494,
      "epoch": 0.1500075154065835,
      "flos": 70652375239680.0,
      "grad_norm": 1.9666951098265444,
      "language_loss": 0.66467059,
      "learning_rate": 3.850818114805354e-06,
      "loss": 0.68654519,
      "num_input_tokens_seen": 54161065,
      "step": 2495,
      "time_per_iteration": 2.974846601486206
    },
    {
      "auxiliary_loss_clip": 0.01084751,
      "auxiliary_loss_mlp": 0.01019043,
      "balance_loss_clip": 1.04845512,
      "balance_loss_mlp": 1.01605058,
      "epoch": 0.15006763865925146,
      "flos": 68011937447040.0,
      "grad_norm": 0.8982829157603746,
      "language_loss": 0.59466273,
      "learning_rate": 3.850670485516019e-06,
      "loss": 0.61570066,
      "num_input_tokens_seen": 54225095,
      "step": 2496,
      "time_per_iteration": 3.1170949935913086
    },
    {
      "auxiliary_loss_clip": 0.01160417,
      "auxiliary_loss_mlp": 0.0105546,
      "balance_loss_clip": 1.05074561,
      "balance_loss_mlp": 1.03505135,
      "epoch": 0.15012776191191943,
      "flos": 18916269459840.0,
      "grad_norm": 1.7901043711673648,
      "language_loss": 0.65183806,
      "learning_rate": 3.850522786049075e-06,
      "loss": 0.67399681,
      "num_input_tokens_seen": 54243750,
      "step": 2497,
      "time_per_iteration": 2.4556875228881836
    },
    {
      "auxiliary_loss_clip": 0.01124939,
      "auxiliary_loss_mlp": 0.0106176,
      "balance_loss_clip": 1.05286956,
      "balance_loss_mlp": 1.04002798,
      "epoch": 0.1501878851645874,
      "flos": 23701330638720.0,
      "grad_norm": 1.4858828350386342,
      "language_loss": 0.75043976,
      "learning_rate": 3.850375016410121e-06,
      "loss": 0.7723068,
      "num_input_tokens_seen": 54266185,
      "step": 2498,
      "time_per_iteration": 2.6588547229766846
    },
    {
      "auxiliary_loss_clip": 0.01127115,
      "auxiliary_loss_mlp": 0.01050967,
      "balance_loss_clip": 1.05204296,
      "balance_loss_mlp": 1.03031969,
      "epoch": 0.15024800841725539,
      "flos": 20412523422720.0,
      "grad_norm": 4.640523968907937,
      "language_loss": 0.72077155,
      "learning_rate": 3.850227176604761e-06,
      "loss": 0.7425524,
      "num_input_tokens_seen": 54283940,
      "step": 2499,
      "time_per_iteration": 2.542942523956299
    },
    {
      "auxiliary_loss_clip": 0.01132318,
      "auxiliary_loss_mlp": 0.01067322,
      "balance_loss_clip": 1.05226731,
      "balance_loss_mlp": 1.04684246,
      "epoch": 0.15030813166992335,
      "flos": 31831002812160.0,
      "grad_norm": 1.8843672582138238,
      "language_loss": 0.71581495,
      "learning_rate": 3.850079266638601e-06,
      "loss": 0.73781133,
      "num_input_tokens_seen": 54304830,
      "step": 2500,
      "time_per_iteration": 2.6081840991973877
    },
    {
      "auxiliary_loss_clip": 0.01124739,
      "auxiliary_loss_mlp": 0.01070387,
      "balance_loss_clip": 1.04889917,
      "balance_loss_mlp": 1.05043149,
      "epoch": 0.15036825492259132,
      "flos": 35657822914560.0,
      "grad_norm": 1.744526931384772,
      "language_loss": 0.65146405,
      "learning_rate": 3.849931286517249e-06,
      "loss": 0.6734153,
      "num_input_tokens_seen": 54325595,
      "step": 2501,
      "time_per_iteration": 2.73694109916687
    },
    {
      "auxiliary_loss_clip": 0.01134365,
      "auxiliary_loss_mlp": 0.01079556,
      "balance_loss_clip": 1.04861701,
      "balance_loss_mlp": 1.05830109,
      "epoch": 0.15042837817525928,
      "flos": 18838163335680.0,
      "grad_norm": 2.314857911102935,
      "language_loss": 0.83793402,
      "learning_rate": 3.849783236246318e-06,
      "loss": 0.86007321,
      "num_input_tokens_seen": 54342180,
      "step": 2502,
      "time_per_iteration": 2.5477378368377686
    },
    {
      "auxiliary_loss_clip": 0.01118484,
      "auxiliary_loss_mlp": 0.01065443,
      "balance_loss_clip": 1.0469476,
      "balance_loss_mlp": 1.04653668,
      "epoch": 0.15048850142792725,
      "flos": 19535548867200.0,
      "grad_norm": 2.0537044022343305,
      "language_loss": 0.77425718,
      "learning_rate": 3.849635115831421e-06,
      "loss": 0.79609644,
      "num_input_tokens_seen": 54360255,
      "step": 2503,
      "time_per_iteration": 2.5523369312286377
    },
    {
      "auxiliary_loss_clip": 0.01157119,
      "auxiliary_loss_mlp": 0.0105673,
      "balance_loss_clip": 1.05201411,
      "balance_loss_mlp": 1.03819287,
      "epoch": 0.1505486246805952,
      "flos": 22017550746240.0,
      "grad_norm": 2.583759616398654,
      "language_loss": 0.85529566,
      "learning_rate": 3.849486925278176e-06,
      "loss": 0.87743413,
      "num_input_tokens_seen": 54378260,
      "step": 2504,
      "time_per_iteration": 3.9146196842193604
    },
    {
      "auxiliary_loss_clip": 0.01143468,
      "auxiliary_loss_mlp": 0.01049228,
      "balance_loss_clip": 1.05002737,
      "balance_loss_mlp": 1.03102481,
      "epoch": 0.15060874793326318,
      "flos": 20743153136640.0,
      "grad_norm": 1.6398931919116126,
      "language_loss": 0.83018124,
      "learning_rate": 3.8493386645922e-06,
      "loss": 0.85210818,
      "num_input_tokens_seen": 54399745,
      "step": 2505,
      "time_per_iteration": 3.9763729572296143
    },
    {
      "auxiliary_loss_clip": 0.01120317,
      "auxiliary_loss_mlp": 0.01057417,
      "balance_loss_clip": 1.05058157,
      "balance_loss_mlp": 1.0377475,
      "epoch": 0.15066887118593117,
      "flos": 16471902055680.0,
      "grad_norm": 2.3076563837183803,
      "language_loss": 0.76510113,
      "learning_rate": 3.849190333779117e-06,
      "loss": 0.78687847,
      "num_input_tokens_seen": 54417105,
      "step": 2506,
      "time_per_iteration": 2.564502716064453
    },
    {
      "auxiliary_loss_clip": 0.01162492,
      "auxiliary_loss_mlp": 0.01059004,
      "balance_loss_clip": 1.05216813,
      "balance_loss_mlp": 1.03885794,
      "epoch": 0.15072899443859913,
      "flos": 19859319083520.0,
      "grad_norm": 3.6867381562255575,
      "language_loss": 0.768471,
      "learning_rate": 3.849041932844552e-06,
      "loss": 0.79068601,
      "num_input_tokens_seen": 54433920,
      "step": 2507,
      "time_per_iteration": 3.873556613922119
    },
    {
      "auxiliary_loss_clip": 0.01141533,
      "auxiliary_loss_mlp": 0.01053177,
      "balance_loss_clip": 1.04697347,
      "balance_loss_mlp": 1.03407955,
      "epoch": 0.1507891176912671,
      "flos": 20776226584320.0,
      "grad_norm": 2.022889721079339,
      "language_loss": 0.69416809,
      "learning_rate": 3.848893461794131e-06,
      "loss": 0.71611518,
      "num_input_tokens_seen": 54451540,
      "step": 2508,
      "time_per_iteration": 2.4934346675872803
    },
    {
      "auxiliary_loss_clip": 0.01128214,
      "auxiliary_loss_mlp": 0.0106073,
      "balance_loss_clip": 1.05350566,
      "balance_loss_mlp": 1.04187119,
      "epoch": 0.15084924094393506,
      "flos": 23586631534080.0,
      "grad_norm": 1.8281911134299806,
      "language_loss": 0.77949977,
      "learning_rate": 3.8487449206334845e-06,
      "loss": 0.80138922,
      "num_input_tokens_seen": 54470800,
      "step": 2509,
      "time_per_iteration": 2.5723891258239746
    },
    {
      "auxiliary_loss_clip": 0.01137045,
      "auxiliary_loss_mlp": 0.00880045,
      "balance_loss_clip": 1.04927325,
      "balance_loss_mlp": 1.14530373,
      "epoch": 0.15090936419660303,
      "flos": 18911313383040.0,
      "grad_norm": 3.1766073472814895,
      "language_loss": 0.80245662,
      "learning_rate": 3.848596309368246e-06,
      "loss": 0.82262748,
      "num_input_tokens_seen": 54486525,
      "step": 2510,
      "time_per_iteration": 2.488866090774536
    },
    {
      "auxiliary_loss_clip": 0.0114898,
      "auxiliary_loss_mlp": 0.01059693,
      "balance_loss_clip": 1.05192995,
      "balance_loss_mlp": 1.03912961,
      "epoch": 0.150969487449271,
      "flos": 17928223073280.0,
      "grad_norm": 2.5355982351648128,
      "language_loss": 0.7406621,
      "learning_rate": 3.8484476280040495e-06,
      "loss": 0.76274884,
      "num_input_tokens_seen": 54503795,
      "step": 2511,
      "time_per_iteration": 3.987982988357544
    },
    {
      "auxiliary_loss_clip": 0.01092794,
      "auxiliary_loss_mlp": 0.01049698,
      "balance_loss_clip": 1.04707658,
      "balance_loss_mlp": 1.03017139,
      "epoch": 0.151029610701939,
      "flos": 24243078539520.0,
      "grad_norm": 2.2046265161962144,
      "language_loss": 0.69163018,
      "learning_rate": 3.848298876546534e-06,
      "loss": 0.71305513,
      "num_input_tokens_seen": 54523025,
      "step": 2512,
      "time_per_iteration": 2.6761438846588135
    },
    {
      "auxiliary_loss_clip": 0.01146222,
      "auxiliary_loss_mlp": 0.01054872,
      "balance_loss_clip": 1.04937458,
      "balance_loss_mlp": 1.03579879,
      "epoch": 0.15108973395460695,
      "flos": 30262496641920.0,
      "grad_norm": 2.11195926861456,
      "language_loss": 0.73997235,
      "learning_rate": 3.84815005500134e-06,
      "loss": 0.76198328,
      "num_input_tokens_seen": 54545025,
      "step": 2513,
      "time_per_iteration": 2.553710460662842
    },
    {
      "auxiliary_loss_clip": 0.0102298,
      "auxiliary_loss_mlp": 0.01038038,
      "balance_loss_clip": 1.03179371,
      "balance_loss_mlp": 1.03474748,
      "epoch": 0.15114985720727492,
      "flos": 60437624428800.0,
      "grad_norm": 0.8740523425864012,
      "language_loss": 0.64763385,
      "learning_rate": 3.84800116337411e-06,
      "loss": 0.668244,
      "num_input_tokens_seen": 54604545,
      "step": 2514,
      "time_per_iteration": 3.1574530601501465
    },
    {
      "auxiliary_loss_clip": 0.01141346,
      "auxiliary_loss_mlp": 0.0103963,
      "balance_loss_clip": 1.05046594,
      "balance_loss_mlp": 1.02077067,
      "epoch": 0.15120998045994288,
      "flos": 20521691832960.0,
      "grad_norm": 2.055907157330445,
      "language_loss": 0.72729599,
      "learning_rate": 3.8478522016704916e-06,
      "loss": 0.74910581,
      "num_input_tokens_seen": 54620590,
      "step": 2515,
      "time_per_iteration": 2.474595785140991
    },
    {
      "auxiliary_loss_clip": 0.01126926,
      "auxiliary_loss_mlp": 0.01041685,
      "balance_loss_clip": 1.04961264,
      "balance_loss_mlp": 1.0208714,
      "epoch": 0.15127010371261085,
      "flos": 21178893024000.0,
      "grad_norm": 1.8932481374314358,
      "language_loss": 0.77344477,
      "learning_rate": 3.8477031698961325e-06,
      "loss": 0.79513085,
      "num_input_tokens_seen": 54640410,
      "step": 2516,
      "time_per_iteration": 2.562945604324341
    },
    {
      "auxiliary_loss_clip": 0.01068518,
      "auxiliary_loss_mlp": 0.0100878,
      "balance_loss_clip": 1.03562427,
      "balance_loss_mlp": 1.00612187,
      "epoch": 0.1513302269652788,
      "flos": 65320648974720.0,
      "grad_norm": 0.7259082508010725,
      "language_loss": 0.54675132,
      "learning_rate": 3.8475540680566835e-06,
      "loss": 0.56752437,
      "num_input_tokens_seen": 54701430,
      "step": 2517,
      "time_per_iteration": 3.12744402885437
    },
    {
      "auxiliary_loss_clip": 0.01111521,
      "auxiliary_loss_mlp": 0.01049102,
      "balance_loss_clip": 1.04592192,
      "balance_loss_mlp": 1.02680957,
      "epoch": 0.15139035021794678,
      "flos": 19135827342720.0,
      "grad_norm": 2.1264570267527487,
      "language_loss": 0.7876718,
      "learning_rate": 3.8474048961577995e-06,
      "loss": 0.80927801,
      "num_input_tokens_seen": 54720845,
      "step": 2518,
      "time_per_iteration": 2.555051803588867
    },
    {
      "auxiliary_loss_clip": 0.01148479,
      "auxiliary_loss_mlp": 0.01048283,
      "balance_loss_clip": 1.05069709,
      "balance_loss_mlp": 1.02639639,
      "epoch": 0.15145047347061477,
      "flos": 26578564842240.0,
      "grad_norm": 2.452230839740843,
      "language_loss": 0.70433033,
      "learning_rate": 3.847255654205137e-06,
      "loss": 0.72629797,
      "num_input_tokens_seen": 54740495,
      "step": 2519,
      "time_per_iteration": 2.5511722564697266
    },
    {
      "auxiliary_loss_clip": 0.01147159,
      "auxiliary_loss_mlp": 0.01047765,
      "balance_loss_clip": 1.05064321,
      "balance_loss_mlp": 1.0274514,
      "epoch": 0.15151059672328274,
      "flos": 20302959962880.0,
      "grad_norm": 2.0091587017778445,
      "language_loss": 0.7884922,
      "learning_rate": 3.847106342204354e-06,
      "loss": 0.81044143,
      "num_input_tokens_seen": 54758415,
      "step": 2520,
      "time_per_iteration": 2.4996514320373535
    },
    {
      "auxiliary_loss_clip": 0.01139298,
      "auxiliary_loss_mlp": 0.01054512,
      "balance_loss_clip": 1.04932129,
      "balance_loss_mlp": 1.03318512,
      "epoch": 0.1515707199759507,
      "flos": 27228367831680.0,
      "grad_norm": 2.027122658888103,
      "language_loss": 0.74943638,
      "learning_rate": 3.846956960161114e-06,
      "loss": 0.77137446,
      "num_input_tokens_seen": 54779355,
      "step": 2521,
      "time_per_iteration": 2.563396453857422
    },
    {
      "auxiliary_loss_clip": 0.01131464,
      "auxiliary_loss_mlp": 0.01049253,
      "balance_loss_clip": 1.04858923,
      "balance_loss_mlp": 1.02840304,
      "epoch": 0.15163084322861867,
      "flos": 23587349806080.0,
      "grad_norm": 2.5785418498676407,
      "language_loss": 0.82430768,
      "learning_rate": 3.84680750808108e-06,
      "loss": 0.84611481,
      "num_input_tokens_seen": 54799465,
      "step": 2522,
      "time_per_iteration": 2.560260772705078
    },
    {
      "auxiliary_loss_clip": 0.01026068,
      "auxiliary_loss_mlp": 0.01005294,
      "balance_loss_clip": 1.0333817,
      "balance_loss_mlp": 1.0027194,
      "epoch": 0.15169096648128663,
      "flos": 66889622021760.0,
      "grad_norm": 0.8202544696166308,
      "language_loss": 0.57920825,
      "learning_rate": 3.846657985969922e-06,
      "loss": 0.59952188,
      "num_input_tokens_seen": 54857665,
      "step": 2523,
      "time_per_iteration": 3.1694202423095703
    },
    {
      "auxiliary_loss_clip": 0.01138149,
      "auxiliary_loss_mlp": 0.01053853,
      "balance_loss_clip": 1.04839444,
      "balance_loss_mlp": 1.03114343,
      "epoch": 0.1517510897339546,
      "flos": 29095435848960.0,
      "grad_norm": 1.8439307734729804,
      "language_loss": 0.75148451,
      "learning_rate": 3.8465083938333066e-06,
      "loss": 0.7734046,
      "num_input_tokens_seen": 54879895,
      "step": 2524,
      "time_per_iteration": 2.6086173057556152
    },
    {
      "auxiliary_loss_clip": 0.01135205,
      "auxiliary_loss_mlp": 0.01044264,
      "balance_loss_clip": 1.04756415,
      "balance_loss_mlp": 1.0243206,
      "epoch": 0.1518112129866226,
      "flos": 18406553512320.0,
      "grad_norm": 2.1940824193768673,
      "language_loss": 0.74776828,
      "learning_rate": 3.8463587316769085e-06,
      "loss": 0.76956296,
      "num_input_tokens_seen": 54898245,
      "step": 2525,
      "time_per_iteration": 2.52813982963562
    },
    {
      "auxiliary_loss_clip": 0.01144014,
      "auxiliary_loss_mlp": 0.01048067,
      "balance_loss_clip": 1.05349541,
      "balance_loss_mlp": 1.02691936,
      "epoch": 0.15187133623929056,
      "flos": 19425410789760.0,
      "grad_norm": 1.8756965286249285,
      "language_loss": 0.80041063,
      "learning_rate": 3.846208999506402e-06,
      "loss": 0.82233143,
      "num_input_tokens_seen": 54917060,
      "step": 2526,
      "time_per_iteration": 2.4979918003082275
    },
    {
      "auxiliary_loss_clip": 0.01135094,
      "auxiliary_loss_mlp": 0.01050416,
      "balance_loss_clip": 1.05089903,
      "balance_loss_mlp": 1.03113997,
      "epoch": 0.15193145949195852,
      "flos": 17566207850880.0,
      "grad_norm": 1.7793354380422595,
      "language_loss": 0.85154343,
      "learning_rate": 3.846059197327466e-06,
      "loss": 0.87339842,
      "num_input_tokens_seen": 54936365,
      "step": 2527,
      "time_per_iteration": 2.5334696769714355
    },
    {
      "auxiliary_loss_clip": 0.0112417,
      "auxiliary_loss_mlp": 0.010507,
      "balance_loss_clip": 1.0487659,
      "balance_loss_mlp": 1.03019643,
      "epoch": 0.15199158274462649,
      "flos": 36176265866880.0,
      "grad_norm": 1.704276548762612,
      "language_loss": 0.69371343,
      "learning_rate": 3.845909325145779e-06,
      "loss": 0.71546209,
      "num_input_tokens_seen": 54961365,
      "step": 2528,
      "time_per_iteration": 2.6612491607666016
    },
    {
      "auxiliary_loss_clip": 0.01133501,
      "auxiliary_loss_mlp": 0.01056481,
      "balance_loss_clip": 1.05222726,
      "balance_loss_mlp": 1.03640664,
      "epoch": 0.15205170599729445,
      "flos": 23074042498560.0,
      "grad_norm": 1.7518748121359065,
      "language_loss": 0.86890495,
      "learning_rate": 3.845759382967026e-06,
      "loss": 0.89080477,
      "num_input_tokens_seen": 54980750,
      "step": 2529,
      "time_per_iteration": 2.571930408477783
    },
    {
      "auxiliary_loss_clip": 0.01124579,
      "auxiliary_loss_mlp": 0.01042148,
      "balance_loss_clip": 1.04975533,
      "balance_loss_mlp": 1.02213299,
      "epoch": 0.15211182924996242,
      "flos": 21908382336000.0,
      "grad_norm": 2.510939320720816,
      "language_loss": 0.83184993,
      "learning_rate": 3.845609370796893e-06,
      "loss": 0.85351717,
      "num_input_tokens_seen": 54999675,
      "step": 2530,
      "time_per_iteration": 2.585864305496216
    },
    {
      "auxiliary_loss_clip": 0.01125927,
      "auxiliary_loss_mlp": 0.01049013,
      "balance_loss_clip": 1.05047512,
      "balance_loss_mlp": 1.02858031,
      "epoch": 0.15217195250263038,
      "flos": 13881521865600.0,
      "grad_norm": 1.9514312696968228,
      "language_loss": 0.80374604,
      "learning_rate": 3.845459288641066e-06,
      "loss": 0.82549536,
      "num_input_tokens_seen": 55018295,
      "step": 2531,
      "time_per_iteration": 2.5352721214294434
    },
    {
      "auxiliary_loss_clip": 0.01143014,
      "auxiliary_loss_mlp": 0.01052997,
      "balance_loss_clip": 1.04903853,
      "balance_loss_mlp": 1.03401923,
      "epoch": 0.15223207575529837,
      "flos": 24535319592960.0,
      "grad_norm": 1.9844243455813229,
      "language_loss": 0.7896744,
      "learning_rate": 3.8453091365052394e-06,
      "loss": 0.81163454,
      "num_input_tokens_seen": 55037975,
      "step": 2532,
      "time_per_iteration": 2.533263921737671
    },
    {
      "auxiliary_loss_clip": 0.01144622,
      "auxiliary_loss_mlp": 0.01055249,
      "balance_loss_clip": 1.04882407,
      "balance_loss_mlp": 1.03402925,
      "epoch": 0.15229219900796634,
      "flos": 25556798563200.0,
      "grad_norm": 1.822884760811584,
      "language_loss": 0.8783493,
      "learning_rate": 3.845158914395105e-06,
      "loss": 0.90034807,
      "num_input_tokens_seen": 55057135,
      "step": 2533,
      "time_per_iteration": 2.529696226119995
    },
    {
      "auxiliary_loss_clip": 0.01119351,
      "auxiliary_loss_mlp": 0.01056793,
      "balance_loss_clip": 1.05187714,
      "balance_loss_mlp": 1.03645563,
      "epoch": 0.1523523222606343,
      "flos": 18217806520320.0,
      "grad_norm": 2.5763237261782663,
      "language_loss": 0.78841448,
      "learning_rate": 3.84500862231636e-06,
      "loss": 0.8101759,
      "num_input_tokens_seen": 55075525,
      "step": 2534,
      "time_per_iteration": 2.56652569770813
    },
    {
      "auxiliary_loss_clip": 0.01161246,
      "auxiliary_loss_mlp": 0.01054156,
      "balance_loss_clip": 1.05030704,
      "balance_loss_mlp": 1.03329444,
      "epoch": 0.15241244551330227,
      "flos": 13260087642240.0,
      "grad_norm": 3.4762863431661337,
      "language_loss": 0.77463758,
      "learning_rate": 3.844858260274702e-06,
      "loss": 0.79679161,
      "num_input_tokens_seen": 55090845,
      "step": 2535,
      "time_per_iteration": 2.4526448249816895
    },
    {
      "auxiliary_loss_clip": 0.01144788,
      "auxiliary_loss_mlp": 0.01052256,
      "balance_loss_clip": 1.05274749,
      "balance_loss_mlp": 1.03209734,
      "epoch": 0.15247256876597023,
      "flos": 19715568854400.0,
      "grad_norm": 2.554221678649223,
      "language_loss": 0.78224379,
      "learning_rate": 3.844707828275835e-06,
      "loss": 0.80421424,
      "num_input_tokens_seen": 55108750,
      "step": 2536,
      "time_per_iteration": 2.4952940940856934
    },
    {
      "auxiliary_loss_clip": 0.01124818,
      "auxiliary_loss_mlp": 0.01061625,
      "balance_loss_clip": 1.05017316,
      "balance_loss_mlp": 1.04196715,
      "epoch": 0.1525326920186382,
      "flos": 20375858615040.0,
      "grad_norm": 2.852800899099084,
      "language_loss": 0.75776547,
      "learning_rate": 3.844557326325461e-06,
      "loss": 0.77962995,
      "num_input_tokens_seen": 55126750,
      "step": 2537,
      "time_per_iteration": 2.521355628967285
    },
    {
      "auxiliary_loss_clip": 0.01146749,
      "auxiliary_loss_mlp": 0.01058097,
      "balance_loss_clip": 1.05160391,
      "balance_loss_mlp": 1.03852248,
      "epoch": 0.15259281527130616,
      "flos": 13589963170560.0,
      "grad_norm": 2.7656485179567114,
      "language_loss": 0.78022754,
      "learning_rate": 3.8444067544292896e-06,
      "loss": 0.80227602,
      "num_input_tokens_seen": 55144690,
      "step": 2538,
      "time_per_iteration": 2.483314275741577
    },
    {
      "auxiliary_loss_clip": 0.01107465,
      "auxiliary_loss_mlp": 0.01049015,
      "balance_loss_clip": 1.05090356,
      "balance_loss_mlp": 1.0300374,
      "epoch": 0.15265293852397416,
      "flos": 22860374446080.0,
      "grad_norm": 1.5921112778681392,
      "language_loss": 0.89959598,
      "learning_rate": 3.844256112593029e-06,
      "loss": 0.92116082,
      "num_input_tokens_seen": 55166055,
      "step": 2539,
      "time_per_iteration": 2.6397695541381836
    },
    {
      "auxiliary_loss_clip": 0.01142273,
      "auxiliary_loss_mlp": 0.01050259,
      "balance_loss_clip": 1.05230188,
      "balance_loss_mlp": 1.030303,
      "epoch": 0.15271306177664212,
      "flos": 29238108670080.0,
      "grad_norm": 4.493926878854294,
      "language_loss": 0.93545157,
      "learning_rate": 3.844105400822391e-06,
      "loss": 0.95737684,
      "num_input_tokens_seen": 55186285,
      "step": 2540,
      "time_per_iteration": 2.5667154788970947
    },
    {
      "auxiliary_loss_clip": 0.01129979,
      "auxiliary_loss_mlp": 0.01052404,
      "balance_loss_clip": 1.0473243,
      "balance_loss_mlp": 1.033831,
      "epoch": 0.1527731850293101,
      "flos": 31246269310080.0,
      "grad_norm": 1.8158228762175372,
      "language_loss": 0.75441462,
      "learning_rate": 3.843954619123092e-06,
      "loss": 0.77623844,
      "num_input_tokens_seen": 55207915,
      "step": 2541,
      "time_per_iteration": 2.6231472492218018
    },
    {
      "auxiliary_loss_clip": 0.01120513,
      "auxiliary_loss_mlp": 0.01051743,
      "balance_loss_clip": 1.0493474,
      "balance_loss_mlp": 1.03252649,
      "epoch": 0.15283330828197805,
      "flos": 22382079920640.0,
      "grad_norm": 1.7183786107362886,
      "language_loss": 0.81710762,
      "learning_rate": 3.84380376750085e-06,
      "loss": 0.83883023,
      "num_input_tokens_seen": 55227860,
      "step": 2542,
      "time_per_iteration": 2.5596468448638916
    },
    {
      "auxiliary_loss_clip": 0.01160009,
      "auxiliary_loss_mlp": 0.0105614,
      "balance_loss_clip": 1.05316317,
      "balance_loss_mlp": 1.03616071,
      "epoch": 0.15289343153464602,
      "flos": 25520133755520.0,
      "grad_norm": 2.3327947569238834,
      "language_loss": 0.78019363,
      "learning_rate": 3.843652845961383e-06,
      "loss": 0.80235505,
      "num_input_tokens_seen": 55247330,
      "step": 2543,
      "time_per_iteration": 5.326687335968018
    },
    {
      "auxiliary_loss_clip": 0.01145659,
      "auxiliary_loss_mlp": 0.01053998,
      "balance_loss_clip": 1.0511229,
      "balance_loss_mlp": 1.03477001,
      "epoch": 0.15295355478731398,
      "flos": 22710016114560.0,
      "grad_norm": 2.0183930676047583,
      "language_loss": 0.86343306,
      "learning_rate": 3.843501854510416e-06,
      "loss": 0.88542962,
      "num_input_tokens_seen": 55266195,
      "step": 2544,
      "time_per_iteration": 2.527777910232544
    },
    {
      "auxiliary_loss_clip": 0.01148784,
      "auxiliary_loss_mlp": 0.01054293,
      "balance_loss_clip": 1.04916787,
      "balance_loss_mlp": 1.03347957,
      "epoch": 0.15301367803998198,
      "flos": 23251907669760.0,
      "grad_norm": 4.615663077139077,
      "language_loss": 0.82433456,
      "learning_rate": 3.843350793153673e-06,
      "loss": 0.84636533,
      "num_input_tokens_seen": 55283305,
      "step": 2545,
      "time_per_iteration": 2.5196216106414795
    },
    {
      "auxiliary_loss_clip": 0.01158488,
      "auxiliary_loss_mlp": 0.01047582,
      "balance_loss_clip": 1.05189228,
      "balance_loss_mlp": 1.02862787,
      "epoch": 0.15307380129264994,
      "flos": 25886279041920.0,
      "grad_norm": 2.3450245893343964,
      "language_loss": 0.71101475,
      "learning_rate": 3.843199661896884e-06,
      "loss": 0.7330755,
      "num_input_tokens_seen": 55303035,
      "step": 2546,
      "time_per_iteration": 3.9465863704681396
    },
    {
      "auxiliary_loss_clip": 0.01129636,
      "auxiliary_loss_mlp": 0.01051002,
      "balance_loss_clip": 1.04975724,
      "balance_loss_mlp": 1.03081965,
      "epoch": 0.1531339245453179,
      "flos": 46973239205760.0,
      "grad_norm": 1.758471411871533,
      "language_loss": 0.77432281,
      "learning_rate": 3.843048460745779e-06,
      "loss": 0.79612923,
      "num_input_tokens_seen": 55327570,
      "step": 2547,
      "time_per_iteration": 2.7889163494110107
    },
    {
      "auxiliary_loss_clip": 0.01115147,
      "auxiliary_loss_mlp": 0.010451,
      "balance_loss_clip": 1.05391181,
      "balance_loss_mlp": 1.02537072,
      "epoch": 0.15319404779798587,
      "flos": 35882049565440.0,
      "grad_norm": 2.137997111189484,
      "language_loss": 0.74430847,
      "learning_rate": 3.842897189706092e-06,
      "loss": 0.76591098,
      "num_input_tokens_seen": 55351090,
      "step": 2548,
      "time_per_iteration": 2.7302422523498535
    },
    {
      "auxiliary_loss_clip": 0.0113904,
      "auxiliary_loss_mlp": 0.01048724,
      "balance_loss_clip": 1.05182588,
      "balance_loss_mlp": 1.02821994,
      "epoch": 0.15325417105065384,
      "flos": 25664638170240.0,
      "grad_norm": 2.0238430258452698,
      "language_loss": 0.80559909,
      "learning_rate": 3.842745848783558e-06,
      "loss": 0.82747674,
      "num_input_tokens_seen": 55371050,
      "step": 2549,
      "time_per_iteration": 2.5879571437835693
    },
    {
      "auxiliary_loss_clip": 0.01146272,
      "auxiliary_loss_mlp": 0.01047281,
      "balance_loss_clip": 1.050583,
      "balance_loss_mlp": 1.02745676,
      "epoch": 0.1533142943033218,
      "flos": 18770831291520.0,
      "grad_norm": 1.9548365062171056,
      "language_loss": 0.74873894,
      "learning_rate": 3.842594437983917e-06,
      "loss": 0.77067447,
      "num_input_tokens_seen": 55390375,
      "step": 2550,
      "time_per_iteration": 3.9963526725769043
    },
    {
      "auxiliary_loss_clip": 0.01152815,
      "auxiliary_loss_mlp": 0.01044814,
      "balance_loss_clip": 1.05254066,
      "balance_loss_mlp": 1.02457261,
      "epoch": 0.15337441755598977,
      "flos": 23107367341440.0,
      "grad_norm": 2.4050823229418365,
      "language_loss": 0.76782262,
      "learning_rate": 3.8424429573129115e-06,
      "loss": 0.78979892,
      "num_input_tokens_seen": 55408890,
      "step": 2551,
      "time_per_iteration": 2.4923460483551025
    },
    {
      "auxiliary_loss_clip": 0.01076759,
      "auxiliary_loss_mlp": 0.01005417,
      "balance_loss_clip": 1.04264545,
      "balance_loss_mlp": 1.00258017,
      "epoch": 0.15343454080865776,
      "flos": 59861079227520.0,
      "grad_norm": 0.9658240234898465,
      "language_loss": 0.56676167,
      "learning_rate": 3.842291406776283e-06,
      "loss": 0.58758336,
      "num_input_tokens_seen": 55463815,
      "step": 2552,
      "time_per_iteration": 3.0442583560943604
    },
    {
      "auxiliary_loss_clip": 0.01117284,
      "auxiliary_loss_mlp": 0.01043056,
      "balance_loss_clip": 1.05243623,
      "balance_loss_mlp": 1.02281451,
      "epoch": 0.15349466406132573,
      "flos": 11910887959680.0,
      "grad_norm": 2.2322835760058597,
      "language_loss": 0.88332069,
      "learning_rate": 3.84213978637978e-06,
      "loss": 0.90492404,
      "num_input_tokens_seen": 55481050,
      "step": 2553,
      "time_per_iteration": 2.5500214099884033
    },
    {
      "auxiliary_loss_clip": 0.0114822,
      "auxiliary_loss_mlp": 0.01047968,
      "balance_loss_clip": 1.05233574,
      "balance_loss_mlp": 1.02718961,
      "epoch": 0.1535547873139937,
      "flos": 24096922099200.0,
      "grad_norm": 1.9482394906536369,
      "language_loss": 0.78227711,
      "learning_rate": 3.841988096129152e-06,
      "loss": 0.80423903,
      "num_input_tokens_seen": 55500050,
      "step": 2554,
      "time_per_iteration": 2.517559051513672
    },
    {
      "auxiliary_loss_clip": 0.01093373,
      "auxiliary_loss_mlp": 0.01059305,
      "balance_loss_clip": 1.05001712,
      "balance_loss_mlp": 1.03687036,
      "epoch": 0.15361491056666166,
      "flos": 17566459246080.0,
      "grad_norm": 2.485267507428169,
      "language_loss": 0.77934194,
      "learning_rate": 3.841836336030151e-06,
      "loss": 0.80086869,
      "num_input_tokens_seen": 55518125,
      "step": 2555,
      "time_per_iteration": 2.577765703201294
    },
    {
      "auxiliary_loss_clip": 0.0112801,
      "auxiliary_loss_mlp": 0.01047697,
      "balance_loss_clip": 1.05090237,
      "balance_loss_mlp": 1.0291357,
      "epoch": 0.15367503381932962,
      "flos": 25046041121280.0,
      "grad_norm": 1.5491674311202603,
      "language_loss": 0.76928902,
      "learning_rate": 3.8416845060885305e-06,
      "loss": 0.79104602,
      "num_input_tokens_seen": 55540960,
      "step": 2556,
      "time_per_iteration": 2.6282296180725098
    },
    {
      "auxiliary_loss_clip": 0.01139179,
      "auxiliary_loss_mlp": 0.00860632,
      "balance_loss_clip": 1.05250299,
      "balance_loss_mlp": 1.11179781,
      "epoch": 0.15373515707199759,
      "flos": 21507332008320.0,
      "grad_norm": 2.2018105070249288,
      "language_loss": 0.90251791,
      "learning_rate": 3.84153260631005e-06,
      "loss": 0.92251599,
      "num_input_tokens_seen": 55559210,
      "step": 2557,
      "time_per_iteration": 2.4976017475128174
    },
    {
      "auxiliary_loss_clip": 0.01138358,
      "auxiliary_loss_mlp": 0.01049309,
      "balance_loss_clip": 1.04956651,
      "balance_loss_mlp": 1.02872157,
      "epoch": 0.15379528032466555,
      "flos": 25994729180160.0,
      "grad_norm": 1.9548326111825167,
      "language_loss": 0.71053481,
      "learning_rate": 3.841380636700468e-06,
      "loss": 0.73241144,
      "num_input_tokens_seen": 55578925,
      "step": 2558,
      "time_per_iteration": 2.563786029815674
    },
    {
      "auxiliary_loss_clip": 0.01134482,
      "auxiliary_loss_mlp": 0.01047713,
      "balance_loss_clip": 1.05161393,
      "balance_loss_mlp": 1.02787685,
      "epoch": 0.15385540357733354,
      "flos": 19277315015040.0,
      "grad_norm": 1.9826286102375483,
      "language_loss": 0.92315483,
      "learning_rate": 3.841228597265548e-06,
      "loss": 0.94497675,
      "num_input_tokens_seen": 55597255,
      "step": 2559,
      "time_per_iteration": 2.5170583724975586
    },
    {
      "auxiliary_loss_clip": 0.01135743,
      "auxiliary_loss_mlp": 0.01053967,
      "balance_loss_clip": 1.05424476,
      "balance_loss_mlp": 1.03341496,
      "epoch": 0.1539155268300015,
      "flos": 28549126920960.0,
      "grad_norm": 3.08303640166999,
      "language_loss": 0.63885242,
      "learning_rate": 3.841076488011055e-06,
      "loss": 0.66074955,
      "num_input_tokens_seen": 55619515,
      "step": 2560,
      "time_per_iteration": 2.6054844856262207
    },
    {
      "auxiliary_loss_clip": 0.0113726,
      "auxiliary_loss_mlp": 0.01050918,
      "balance_loss_clip": 1.05251837,
      "balance_loss_mlp": 1.03021097,
      "epoch": 0.15397565008266947,
      "flos": 23547883737600.0,
      "grad_norm": 4.994365369489055,
      "language_loss": 0.88214368,
      "learning_rate": 3.8409243089427574e-06,
      "loss": 0.9040255,
      "num_input_tokens_seen": 55640050,
      "step": 2561,
      "time_per_iteration": 2.553596019744873
    },
    {
      "auxiliary_loss_clip": 0.01145112,
      "auxiliary_loss_mlp": 0.01048583,
      "balance_loss_clip": 1.05259371,
      "balance_loss_mlp": 1.02948546,
      "epoch": 0.15403577333533744,
      "flos": 17129821518720.0,
      "grad_norm": 1.7033734633207729,
      "language_loss": 0.8320027,
      "learning_rate": 3.840772060066425e-06,
      "loss": 0.85393965,
      "num_input_tokens_seen": 55658695,
      "step": 2562,
      "time_per_iteration": 2.492812156677246
    },
    {
      "auxiliary_loss_clip": 0.01134478,
      "auxiliary_loss_mlp": 0.00869171,
      "balance_loss_clip": 1.05442357,
      "balance_loss_mlp": 1.12458491,
      "epoch": 0.1540958965880054,
      "flos": 17894503180800.0,
      "grad_norm": 2.4911158370411797,
      "language_loss": 0.74867189,
      "learning_rate": 3.840619741387832e-06,
      "loss": 0.76870835,
      "num_input_tokens_seen": 55676340,
      "step": 2563,
      "time_per_iteration": 2.5173745155334473
    },
    {
      "auxiliary_loss_clip": 0.01119423,
      "auxiliary_loss_mlp": 0.01048466,
      "balance_loss_clip": 1.05064583,
      "balance_loss_mlp": 1.02784252,
      "epoch": 0.15415601984067337,
      "flos": 32161057908480.0,
      "grad_norm": 2.0408635080308675,
      "language_loss": 0.76217675,
      "learning_rate": 3.8404673529127534e-06,
      "loss": 0.78385568,
      "num_input_tokens_seen": 55698890,
      "step": 2564,
      "time_per_iteration": 2.6974117755889893
    },
    {
      "auxiliary_loss_clip": 0.01133136,
      "auxiliary_loss_mlp": 0.01055372,
      "balance_loss_clip": 1.05012202,
      "balance_loss_mlp": 1.03629804,
      "epoch": 0.15421614309334136,
      "flos": 24024418496640.0,
      "grad_norm": 2.0292507503310597,
      "language_loss": 0.70627475,
      "learning_rate": 3.840314894646969e-06,
      "loss": 0.72815984,
      "num_input_tokens_seen": 55718535,
      "step": 2565,
      "time_per_iteration": 2.554295301437378
    },
    {
      "auxiliary_loss_clip": 0.01142283,
      "auxiliary_loss_mlp": 0.01052779,
      "balance_loss_clip": 1.04870152,
      "balance_loss_mlp": 1.03320503,
      "epoch": 0.15427626634600933,
      "flos": 24386290064640.0,
      "grad_norm": 1.9466351950972853,
      "language_loss": 0.71872288,
      "learning_rate": 3.840162366596259e-06,
      "loss": 0.74067342,
      "num_input_tokens_seen": 55738970,
      "step": 2566,
      "time_per_iteration": 2.5477030277252197
    },
    {
      "auxiliary_loss_clip": 0.01153744,
      "auxiliary_loss_mlp": 0.01042989,
      "balance_loss_clip": 1.04815865,
      "balance_loss_mlp": 1.02430868,
      "epoch": 0.1543363895986773,
      "flos": 23331522165120.0,
      "grad_norm": 1.6322097085438834,
      "language_loss": 0.8511951,
      "learning_rate": 3.840009768766408e-06,
      "loss": 0.87316245,
      "num_input_tokens_seen": 55759585,
      "step": 2567,
      "time_per_iteration": 2.4784634113311768
    },
    {
      "auxiliary_loss_clip": 0.01110045,
      "auxiliary_loss_mlp": 0.01054278,
      "balance_loss_clip": 1.04885781,
      "balance_loss_mlp": 1.03478742,
      "epoch": 0.15439651285134526,
      "flos": 24274284480000.0,
      "grad_norm": 2.211941030955953,
      "language_loss": 0.78245115,
      "learning_rate": 3.839857101163202e-06,
      "loss": 0.80409437,
      "num_input_tokens_seen": 55779250,
      "step": 2568,
      "time_per_iteration": 2.6046855449676514
    },
    {
      "auxiliary_loss_clip": 0.01128518,
      "auxiliary_loss_mlp": 0.01038183,
      "balance_loss_clip": 1.05311489,
      "balance_loss_mlp": 1.01730943,
      "epoch": 0.15445663610401322,
      "flos": 22456163721600.0,
      "grad_norm": 2.005929588279904,
      "language_loss": 0.7026242,
      "learning_rate": 3.83970436379243e-06,
      "loss": 0.72429121,
      "num_input_tokens_seen": 55800470,
      "step": 2569,
      "time_per_iteration": 2.547910690307617
    },
    {
      "auxiliary_loss_clip": 0.01130472,
      "auxiliary_loss_mlp": 0.01043024,
      "balance_loss_clip": 1.05044353,
      "balance_loss_mlp": 1.02381897,
      "epoch": 0.1545167593566812,
      "flos": 22049510872320.0,
      "grad_norm": 1.9269957749430744,
      "language_loss": 0.76616973,
      "learning_rate": 3.839551556659884e-06,
      "loss": 0.78790468,
      "num_input_tokens_seen": 55817795,
      "step": 2570,
      "time_per_iteration": 2.5354466438293457
    },
    {
      "auxiliary_loss_clip": 0.01142975,
      "auxiliary_loss_mlp": 0.01045213,
      "balance_loss_clip": 1.05332744,
      "balance_loss_mlp": 1.02488828,
      "epoch": 0.15457688260934915,
      "flos": 19318253541120.0,
      "grad_norm": 2.278589381739884,
      "language_loss": 0.775446,
      "learning_rate": 3.839398679771359e-06,
      "loss": 0.79732788,
      "num_input_tokens_seen": 55836125,
      "step": 2571,
      "time_per_iteration": 2.472949743270874
    },
    {
      "auxiliary_loss_clip": 0.01138649,
      "auxiliary_loss_mlp": 0.01046947,
      "balance_loss_clip": 1.05138683,
      "balance_loss_mlp": 1.02762341,
      "epoch": 0.15463700586201715,
      "flos": 24133981956480.0,
      "grad_norm": 1.9657092301537848,
      "language_loss": 0.82674485,
      "learning_rate": 3.839245733132652e-06,
      "loss": 0.84860086,
      "num_input_tokens_seen": 55855280,
      "step": 2572,
      "time_per_iteration": 2.548285722732544
    },
    {
      "auxiliary_loss_clip": 0.01161078,
      "auxiliary_loss_mlp": 0.01047176,
      "balance_loss_clip": 1.0529201,
      "balance_loss_mlp": 1.02773345,
      "epoch": 0.1546971291146851,
      "flos": 22420935457920.0,
      "grad_norm": 1.5463024952676792,
      "language_loss": 0.90564501,
      "learning_rate": 3.839092716749563e-06,
      "loss": 0.92772758,
      "num_input_tokens_seen": 55875695,
      "step": 2573,
      "time_per_iteration": 2.476029396057129
    },
    {
      "auxiliary_loss_clip": 0.01097899,
      "auxiliary_loss_mlp": 0.01050663,
      "balance_loss_clip": 1.05549121,
      "balance_loss_mlp": 1.02971756,
      "epoch": 0.15475725236735308,
      "flos": 17530225401600.0,
      "grad_norm": 1.8673872232848931,
      "language_loss": 0.70427346,
      "learning_rate": 3.838939630627893e-06,
      "loss": 0.72575909,
      "num_input_tokens_seen": 55894575,
      "step": 2574,
      "time_per_iteration": 2.580514430999756
    },
    {
      "auxiliary_loss_clip": 0.01130388,
      "auxiliary_loss_mlp": 0.01048142,
      "balance_loss_clip": 1.04904199,
      "balance_loss_mlp": 1.0263741,
      "epoch": 0.15481737562002104,
      "flos": 22561740771840.0,
      "grad_norm": 1.8217795631431755,
      "language_loss": 0.82517636,
      "learning_rate": 3.838786474773448e-06,
      "loss": 0.84696174,
      "num_input_tokens_seen": 55912855,
      "step": 2575,
      "time_per_iteration": 2.52793550491333
    },
    {
      "auxiliary_loss_clip": 0.01133294,
      "auxiliary_loss_mlp": 0.01052214,
      "balance_loss_clip": 1.04802489,
      "balance_loss_mlp": 1.03308117,
      "epoch": 0.154877498872689,
      "flos": 24900567039360.0,
      "grad_norm": 1.8544409588099326,
      "language_loss": 0.84676719,
      "learning_rate": 3.838633249192036e-06,
      "loss": 0.86862224,
      "num_input_tokens_seen": 55932375,
      "step": 2576,
      "time_per_iteration": 2.557020425796509
    },
    {
      "auxiliary_loss_clip": 0.01156903,
      "auxiliary_loss_mlp": 0.01045763,
      "balance_loss_clip": 1.04918814,
      "balance_loss_mlp": 1.0260458,
      "epoch": 0.15493762212535697,
      "flos": 28147501975680.0,
      "grad_norm": 1.8340139993553701,
      "language_loss": 0.82088304,
      "learning_rate": 3.838479953889465e-06,
      "loss": 0.84290969,
      "num_input_tokens_seen": 55953970,
      "step": 2577,
      "time_per_iteration": 2.5128233432769775
    },
    {
      "auxiliary_loss_clip": 0.01129615,
      "auxiliary_loss_mlp": 0.01050876,
      "balance_loss_clip": 1.05319309,
      "balance_loss_mlp": 1.0313375,
      "epoch": 0.15499774537802496,
      "flos": 25411073086080.0,
      "grad_norm": 2.953090357528126,
      "language_loss": 0.76897573,
      "learning_rate": 3.8383265888715525e-06,
      "loss": 0.79078066,
      "num_input_tokens_seen": 55973120,
      "step": 2578,
      "time_per_iteration": 2.5880885124206543
    },
    {
      "auxiliary_loss_clip": 0.01121793,
      "auxiliary_loss_mlp": 0.01046604,
      "balance_loss_clip": 1.04757071,
      "balance_loss_mlp": 1.02655268,
      "epoch": 0.15505786863069293,
      "flos": 22091562720000.0,
      "grad_norm": 1.8286879994927463,
      "language_loss": 0.82634807,
      "learning_rate": 3.83817315414411e-06,
      "loss": 0.848032,
      "num_input_tokens_seen": 55993260,
      "step": 2579,
      "time_per_iteration": 2.558387517929077
    },
    {
      "auxiliary_loss_clip": 0.01139256,
      "auxiliary_loss_mlp": 0.01047385,
      "balance_loss_clip": 1.05205345,
      "balance_loss_mlp": 1.02754855,
      "epoch": 0.1551179918833609,
      "flos": 18917131386240.0,
      "grad_norm": 1.9448337536451037,
      "language_loss": 0.80735278,
      "learning_rate": 3.838019649712958e-06,
      "loss": 0.82921916,
      "num_input_tokens_seen": 56012130,
      "step": 2580,
      "time_per_iteration": 2.5290110111236572
    },
    {
      "auxiliary_loss_clip": 0.01067171,
      "auxiliary_loss_mlp": 0.0100769,
      "balance_loss_clip": 1.0338341,
      "balance_loss_mlp": 1.00486457,
      "epoch": 0.15517811513602886,
      "flos": 66239172587520.0,
      "grad_norm": 0.8479080104445912,
      "language_loss": 0.58838534,
      "learning_rate": 3.8378660755839166e-06,
      "loss": 0.60913396,
      "num_input_tokens_seen": 56079045,
      "step": 2581,
      "time_per_iteration": 4.556372165679932
    },
    {
      "auxiliary_loss_clip": 0.01116365,
      "auxiliary_loss_mlp": 0.01050844,
      "balance_loss_clip": 1.05249953,
      "balance_loss_mlp": 1.02869534,
      "epoch": 0.15523823838869683,
      "flos": 24021078531840.0,
      "grad_norm": 1.8857941419449658,
      "language_loss": 0.84954005,
      "learning_rate": 3.8377124317628095e-06,
      "loss": 0.87121218,
      "num_input_tokens_seen": 56098745,
      "step": 2582,
      "time_per_iteration": 3.993600845336914
    },
    {
      "auxiliary_loss_clip": 0.01146047,
      "auxiliary_loss_mlp": 0.01055366,
      "balance_loss_clip": 1.05201161,
      "balance_loss_mlp": 1.03501713,
      "epoch": 0.1552983616413648,
      "flos": 20485062938880.0,
      "grad_norm": 2.2228599941288585,
      "language_loss": 0.78595889,
      "learning_rate": 3.8375587182554625e-06,
      "loss": 0.80797297,
      "num_input_tokens_seen": 56117655,
      "step": 2583,
      "time_per_iteration": 2.518681764602661
    },
    {
      "auxiliary_loss_clip": 0.01143751,
      "auxiliary_loss_mlp": 0.01057379,
      "balance_loss_clip": 1.05177617,
      "balance_loss_mlp": 1.03462195,
      "epoch": 0.15535848489403276,
      "flos": 32123710742400.0,
      "grad_norm": 2.1204701696313,
      "language_loss": 0.76430607,
      "learning_rate": 3.837404935067705e-06,
      "loss": 0.78631735,
      "num_input_tokens_seen": 56141960,
      "step": 2584,
      "time_per_iteration": 3.971644639968872
    },
    {
      "auxiliary_loss_clip": 0.01141764,
      "auxiliary_loss_mlp": 0.01042944,
      "balance_loss_clip": 1.05445266,
      "balance_loss_mlp": 1.02283311,
      "epoch": 0.15541860814670075,
      "flos": 19098444263040.0,
      "grad_norm": 1.7437975099688403,
      "language_loss": 0.75670063,
      "learning_rate": 3.837251082205368e-06,
      "loss": 0.77854764,
      "num_input_tokens_seen": 56161430,
      "step": 2585,
      "time_per_iteration": 2.5077338218688965
    },
    {
      "auxiliary_loss_clip": 0.01115827,
      "auxiliary_loss_mlp": 0.01043049,
      "balance_loss_clip": 1.04827201,
      "balance_loss_mlp": 1.02289057,
      "epoch": 0.1554787313993687,
      "flos": 19172097100800.0,
      "grad_norm": 1.8928225421284093,
      "language_loss": 0.61282724,
      "learning_rate": 3.837097159674286e-06,
      "loss": 0.63441598,
      "num_input_tokens_seen": 56179390,
      "step": 2586,
      "time_per_iteration": 2.550844430923462
    },
    {
      "auxiliary_loss_clip": 0.01133037,
      "auxiliary_loss_mlp": 0.01043845,
      "balance_loss_clip": 1.05139494,
      "balance_loss_mlp": 1.02425861,
      "epoch": 0.15553885465203668,
      "flos": 16143822207360.0,
      "grad_norm": 1.6379651136131848,
      "language_loss": 0.80988228,
      "learning_rate": 3.836943167480296e-06,
      "loss": 0.83165109,
      "num_input_tokens_seen": 56198020,
      "step": 2587,
      "time_per_iteration": 2.511411666870117
    },
    {
      "auxiliary_loss_clip": 0.01163191,
      "auxiliary_loss_mlp": 0.01056787,
      "balance_loss_clip": 1.05253649,
      "balance_loss_mlp": 1.03339815,
      "epoch": 0.15559897790470464,
      "flos": 25337779384320.0,
      "grad_norm": 2.1017308810604987,
      "language_loss": 0.88740063,
      "learning_rate": 3.836789105629236e-06,
      "loss": 0.90960038,
      "num_input_tokens_seen": 56218165,
      "step": 2588,
      "time_per_iteration": 4.034830331802368
    },
    {
      "auxiliary_loss_clip": 0.01103605,
      "auxiliary_loss_mlp": 0.01054511,
      "balance_loss_clip": 1.05438316,
      "balance_loss_mlp": 1.03190923,
      "epoch": 0.1556591011573726,
      "flos": 23148772744320.0,
      "grad_norm": 1.8822686605393952,
      "language_loss": 0.64537483,
      "learning_rate": 3.83663497412695e-06,
      "loss": 0.66695595,
      "num_input_tokens_seen": 56237160,
      "step": 2589,
      "time_per_iteration": 2.635561466217041
    },
    {
      "auxiliary_loss_clip": 0.0110919,
      "auxiliary_loss_mlp": 0.01046284,
      "balance_loss_clip": 1.05230737,
      "balance_loss_mlp": 1.02383757,
      "epoch": 0.15571922441004057,
      "flos": 25370888745600.0,
      "grad_norm": 2.5015777904066767,
      "language_loss": 0.83204043,
      "learning_rate": 3.836480772979281e-06,
      "loss": 0.85359514,
      "num_input_tokens_seen": 56257610,
      "step": 2590,
      "time_per_iteration": 2.60048508644104
    },
    {
      "auxiliary_loss_clip": 0.01121143,
      "auxiliary_loss_mlp": 0.01041085,
      "balance_loss_clip": 1.04782736,
      "balance_loss_mlp": 1.02139163,
      "epoch": 0.15577934766270854,
      "flos": 14501375890560.0,
      "grad_norm": 2.471849204214886,
      "language_loss": 0.79266214,
      "learning_rate": 3.836326502192077e-06,
      "loss": 0.81428438,
      "num_input_tokens_seen": 56275215,
      "step": 2591,
      "time_per_iteration": 2.53827166557312
    },
    {
      "auxiliary_loss_clip": 0.0114508,
      "auxiliary_loss_mlp": 0.01050597,
      "balance_loss_clip": 1.05001843,
      "balance_loss_mlp": 1.03157139,
      "epoch": 0.15583947091537653,
      "flos": 37414537372800.0,
      "grad_norm": 2.8990394122230105,
      "language_loss": 0.64942372,
      "learning_rate": 3.836172161771189e-06,
      "loss": 0.67138052,
      "num_input_tokens_seen": 56297130,
      "step": 2592,
      "time_per_iteration": 2.62737774848938
    },
    {
      "auxiliary_loss_clip": 0.01137842,
      "auxiliary_loss_mlp": 0.01050049,
      "balance_loss_clip": 1.05554557,
      "balance_loss_mlp": 1.0294255,
      "epoch": 0.1558995941680445,
      "flos": 21834729498240.0,
      "grad_norm": 1.9967419005426041,
      "language_loss": 0.82070494,
      "learning_rate": 3.836017751722467e-06,
      "loss": 0.84258384,
      "num_input_tokens_seen": 56314995,
      "step": 2593,
      "time_per_iteration": 2.5337977409362793
    },
    {
      "auxiliary_loss_clip": 0.01147251,
      "auxiliary_loss_mlp": 0.0104324,
      "balance_loss_clip": 1.05376887,
      "balance_loss_mlp": 1.02248597,
      "epoch": 0.15595971742071246,
      "flos": 19792633484160.0,
      "grad_norm": 2.345644485779454,
      "language_loss": 0.72896183,
      "learning_rate": 3.8358632720517695e-06,
      "loss": 0.75086671,
      "num_input_tokens_seen": 56334005,
      "step": 2594,
      "time_per_iteration": 2.490967035293579
    },
    {
      "auxiliary_loss_clip": 0.01122448,
      "auxiliary_loss_mlp": 0.01043658,
      "balance_loss_clip": 1.05042076,
      "balance_loss_mlp": 1.02353597,
      "epoch": 0.15601984067338043,
      "flos": 26722135503360.0,
      "grad_norm": 2.0160700319514806,
      "language_loss": 0.81734759,
      "learning_rate": 3.835708722764952e-06,
      "loss": 0.83900857,
      "num_input_tokens_seen": 56353795,
      "step": 2595,
      "time_per_iteration": 2.584061861038208
    },
    {
      "auxiliary_loss_clip": 0.0115793,
      "auxiliary_loss_mlp": 0.01052562,
      "balance_loss_clip": 1.05071187,
      "balance_loss_mlp": 1.03295159,
      "epoch": 0.1560799639260484,
      "flos": 18369278173440.0,
      "grad_norm": 1.867048680812236,
      "language_loss": 0.86728686,
      "learning_rate": 3.835554103867876e-06,
      "loss": 0.88939178,
      "num_input_tokens_seen": 56373195,
      "step": 2596,
      "time_per_iteration": 2.473767042160034
    },
    {
      "auxiliary_loss_clip": 0.0114731,
      "auxiliary_loss_mlp": 0.01046314,
      "balance_loss_clip": 1.05201685,
      "balance_loss_mlp": 1.02677536,
      "epoch": 0.15614008717871636,
      "flos": 22598980197120.0,
      "grad_norm": 1.7133444338726196,
      "language_loss": 0.68706858,
      "learning_rate": 3.835399415366404e-06,
      "loss": 0.70900482,
      "num_input_tokens_seen": 56391525,
      "step": 2597,
      "time_per_iteration": 2.5299389362335205
    },
    {
      "auxiliary_loss_clip": 0.0112504,
      "auxiliary_loss_mlp": 0.0104955,
      "balance_loss_clip": 1.05099702,
      "balance_loss_mlp": 1.03072715,
      "epoch": 0.15620021043138435,
      "flos": 22746860490240.0,
      "grad_norm": 1.920193386187035,
      "language_loss": 0.7996304,
      "learning_rate": 3.8352446572664035e-06,
      "loss": 0.82137632,
      "num_input_tokens_seen": 56410715,
      "step": 2598,
      "time_per_iteration": 2.5381078720092773
    },
    {
      "auxiliary_loss_clip": 0.01131501,
      "auxiliary_loss_mlp": 0.00830907,
      "balance_loss_clip": 1.04837883,
      "balance_loss_mlp": 1.05966735,
      "epoch": 0.15626033368405232,
      "flos": 13114936782720.0,
      "grad_norm": 1.942526942435539,
      "language_loss": 0.82932025,
      "learning_rate": 3.8350898295737405e-06,
      "loss": 0.84894431,
      "num_input_tokens_seen": 56429170,
      "step": 2599,
      "time_per_iteration": 2.5336642265319824
    },
    {
      "auxiliary_loss_clip": 0.01164703,
      "auxiliary_loss_mlp": 0.01049898,
      "balance_loss_clip": 1.05414498,
      "balance_loss_mlp": 1.02902436,
      "epoch": 0.15632045693672028,
      "flos": 16472297105280.0,
      "grad_norm": 2.2564295678241195,
      "language_loss": 0.81096494,
      "learning_rate": 3.834934932294287e-06,
      "loss": 0.83311093,
      "num_input_tokens_seen": 56445685,
      "step": 2600,
      "time_per_iteration": 2.4484968185424805
    },
    {
      "auxiliary_loss_clip": 0.01162567,
      "auxiliary_loss_mlp": 0.00822825,
      "balance_loss_clip": 1.05523181,
      "balance_loss_mlp": 1.04498982,
      "epoch": 0.15638058018938825,
      "flos": 20850346298880.0,
      "grad_norm": 1.9120580260250803,
      "language_loss": 0.8840012,
      "learning_rate": 3.834779965433917e-06,
      "loss": 0.90385514,
      "num_input_tokens_seen": 56465900,
      "step": 2601,
      "time_per_iteration": 2.5022640228271484
    },
    {
      "auxiliary_loss_clip": 0.0116383,
      "auxiliary_loss_mlp": 0.01070277,
      "balance_loss_clip": 1.05589998,
      "balance_loss_mlp": 1.04824758,
      "epoch": 0.1564407034420562,
      "flos": 21872220318720.0,
      "grad_norm": 2.2427485748434357,
      "language_loss": 0.7833873,
      "learning_rate": 3.834624928998508e-06,
      "loss": 0.80572832,
      "num_input_tokens_seen": 56485020,
      "step": 2602,
      "time_per_iteration": 2.4814453125
    },
    {
      "auxiliary_loss_clip": 0.01127145,
      "auxiliary_loss_mlp": 0.01049286,
      "balance_loss_clip": 1.0498929,
      "balance_loss_mlp": 1.02853215,
      "epoch": 0.15650082669472418,
      "flos": 21834549930240.0,
      "grad_norm": 1.870434730899044,
      "language_loss": 0.73711717,
      "learning_rate": 3.8344698229939376e-06,
      "loss": 0.75888145,
      "num_input_tokens_seen": 56505205,
      "step": 2603,
      "time_per_iteration": 2.584061622619629
    },
    {
      "auxiliary_loss_clip": 0.01149512,
      "auxiliary_loss_mlp": 0.01054733,
      "balance_loss_clip": 1.05262041,
      "balance_loss_mlp": 1.03499222,
      "epoch": 0.15656094994739214,
      "flos": 13800542653440.0,
      "grad_norm": 2.7713403994135657,
      "language_loss": 0.87974572,
      "learning_rate": 3.8343146474260865e-06,
      "loss": 0.90178818,
      "num_input_tokens_seen": 56521495,
      "step": 2604,
      "time_per_iteration": 2.4781904220581055
    },
    {
      "auxiliary_loss_clip": 0.01150665,
      "auxiliary_loss_mlp": 0.01047196,
      "balance_loss_clip": 1.0523665,
      "balance_loss_mlp": 1.02780056,
      "epoch": 0.15662107320006013,
      "flos": 27308197808640.0,
      "grad_norm": 2.059091206466095,
      "language_loss": 0.85477799,
      "learning_rate": 3.834159402300841e-06,
      "loss": 0.87675661,
      "num_input_tokens_seen": 56540665,
      "step": 2605,
      "time_per_iteration": 2.5495266914367676
    },
    {
      "auxiliary_loss_clip": 0.01152897,
      "auxiliary_loss_mlp": 0.01057772,
      "balance_loss_clip": 1.05248463,
      "balance_loss_mlp": 1.0364809,
      "epoch": 0.1566811964527281,
      "flos": 26685075646080.0,
      "grad_norm": 1.89210486431735,
      "language_loss": 0.73640132,
      "learning_rate": 3.834004087624087e-06,
      "loss": 0.75850791,
      "num_input_tokens_seen": 56560805,
      "step": 2606,
      "time_per_iteration": 2.541048288345337
    },
    {
      "auxiliary_loss_clip": 0.01164064,
      "auxiliary_loss_mlp": 0.01053113,
      "balance_loss_clip": 1.05788541,
      "balance_loss_mlp": 1.03430176,
      "epoch": 0.15674131970539606,
      "flos": 16103422385280.0,
      "grad_norm": 2.120554732423527,
      "language_loss": 0.76248789,
      "learning_rate": 3.8338487034017145e-06,
      "loss": 0.78465962,
      "num_input_tokens_seen": 56576335,
      "step": 2607,
      "time_per_iteration": 2.4469385147094727
    },
    {
      "auxiliary_loss_clip": 0.01122012,
      "auxiliary_loss_mlp": 0.01049303,
      "balance_loss_clip": 1.05333447,
      "balance_loss_mlp": 1.03030109,
      "epoch": 0.15680144295806403,
      "flos": 19169690889600.0,
      "grad_norm": 2.3748372323619766,
      "language_loss": 0.82001567,
      "learning_rate": 3.833693249639615e-06,
      "loss": 0.84172881,
      "num_input_tokens_seen": 56595880,
      "step": 2608,
      "time_per_iteration": 2.5525760650634766
    },
    {
      "auxiliary_loss_clip": 0.0113239,
      "auxiliary_loss_mlp": 0.01054914,
      "balance_loss_clip": 1.05113292,
      "balance_loss_mlp": 1.0330863,
      "epoch": 0.156861566210732,
      "flos": 20813430096000.0,
      "grad_norm": 2.300273580300276,
      "language_loss": 0.72607803,
      "learning_rate": 3.833537726343684e-06,
      "loss": 0.74795109,
      "num_input_tokens_seen": 56615130,
      "step": 2609,
      "time_per_iteration": 2.560511350631714
    },
    {
      "auxiliary_loss_clip": 0.01147219,
      "auxiliary_loss_mlp": 0.01046477,
      "balance_loss_clip": 1.05044079,
      "balance_loss_mlp": 1.02612782,
      "epoch": 0.15692168946339996,
      "flos": 20047922421120.0,
      "grad_norm": 1.7517836051760087,
      "language_loss": 0.72035408,
      "learning_rate": 3.833382133519818e-06,
      "loss": 0.74229103,
      "num_input_tokens_seen": 56634005,
      "step": 2610,
      "time_per_iteration": 2.491142511367798
    },
    {
      "auxiliary_loss_clip": 0.01161558,
      "auxiliary_loss_mlp": 0.01056694,
      "balance_loss_clip": 1.05189931,
      "balance_loss_mlp": 1.0352006,
      "epoch": 0.15698181271606793,
      "flos": 21398019943680.0,
      "grad_norm": 1.75391907176345,
      "language_loss": 0.73206174,
      "learning_rate": 3.833226471173919e-06,
      "loss": 0.75424427,
      "num_input_tokens_seen": 56653480,
      "step": 2611,
      "time_per_iteration": 2.4889700412750244
    },
    {
      "auxiliary_loss_clip": 0.01143729,
      "auxiliary_loss_mlp": 0.01044535,
      "balance_loss_clip": 1.0509181,
      "balance_loss_mlp": 1.0248059,
      "epoch": 0.15704193596873592,
      "flos": 20845785271680.0,
      "grad_norm": 2.3213041859120573,
      "language_loss": 0.71256876,
      "learning_rate": 3.833070739311887e-06,
      "loss": 0.73445135,
      "num_input_tokens_seen": 56672270,
      "step": 2612,
      "time_per_iteration": 2.4896185398101807
    },
    {
      "auxiliary_loss_clip": 0.01125027,
      "auxiliary_loss_mlp": 0.010595,
      "balance_loss_clip": 1.05591702,
      "balance_loss_mlp": 1.03861475,
      "epoch": 0.15710205922140388,
      "flos": 21762908254080.0,
      "grad_norm": 1.8910469238577052,
      "language_loss": 0.75777924,
      "learning_rate": 3.83291493793963e-06,
      "loss": 0.77962446,
      "num_input_tokens_seen": 56691510,
      "step": 2613,
      "time_per_iteration": 2.5613925457000732
    },
    {
      "auxiliary_loss_clip": 0.01119408,
      "auxiliary_loss_mlp": 0.01060748,
      "balance_loss_clip": 1.04981589,
      "balance_loss_mlp": 1.040447,
      "epoch": 0.15716218247407185,
      "flos": 25007760201600.0,
      "grad_norm": 1.733329974169963,
      "language_loss": 0.65678358,
      "learning_rate": 3.832759067063055e-06,
      "loss": 0.67858517,
      "num_input_tokens_seen": 56712230,
      "step": 2614,
      "time_per_iteration": 2.5945029258728027
    },
    {
      "auxiliary_loss_clip": 0.01154315,
      "auxiliary_loss_mlp": 0.01047056,
      "balance_loss_clip": 1.05488086,
      "balance_loss_mlp": 1.02637279,
      "epoch": 0.1572223057267398,
      "flos": 20191780391040.0,
      "grad_norm": 2.2321556286864017,
      "language_loss": 0.75306499,
      "learning_rate": 3.832603126688072e-06,
      "loss": 0.77507877,
      "num_input_tokens_seen": 56727490,
      "step": 2615,
      "time_per_iteration": 2.4856534004211426
    },
    {
      "auxiliary_loss_clip": 0.01142303,
      "auxiliary_loss_mlp": 0.0105755,
      "balance_loss_clip": 1.05513799,
      "balance_loss_mlp": 1.03655779,
      "epoch": 0.15728242897940778,
      "flos": 20959514709120.0,
      "grad_norm": 1.8559273202831703,
      "language_loss": 0.72939742,
      "learning_rate": 3.832447116820594e-06,
      "loss": 0.75139594,
      "num_input_tokens_seen": 56747385,
      "step": 2616,
      "time_per_iteration": 2.500152111053467
    },
    {
      "auxiliary_loss_clip": 0.011338,
      "auxiliary_loss_mlp": 0.01048576,
      "balance_loss_clip": 1.0529027,
      "balance_loss_mlp": 1.02795291,
      "epoch": 0.15734255223207574,
      "flos": 23038275530880.0,
      "grad_norm": 2.1843930363104773,
      "language_loss": 0.72392285,
      "learning_rate": 3.832291037466539e-06,
      "loss": 0.74574661,
      "num_input_tokens_seen": 56768055,
      "step": 2617,
      "time_per_iteration": 2.551046848297119
    },
    {
      "auxiliary_loss_clip": 0.01142016,
      "auxiliary_loss_mlp": 0.01056252,
      "balance_loss_clip": 1.05274332,
      "balance_loss_mlp": 1.03493714,
      "epoch": 0.15740267548474374,
      "flos": 20551281661440.0,
      "grad_norm": 2.1206294291603838,
      "language_loss": 0.74147701,
      "learning_rate": 3.8321348886318235e-06,
      "loss": 0.76345974,
      "num_input_tokens_seen": 56785110,
      "step": 2618,
      "time_per_iteration": 2.490840196609497
    },
    {
      "auxiliary_loss_clip": 0.01164222,
      "auxiliary_loss_mlp": 0.0104674,
      "balance_loss_clip": 1.05414999,
      "balance_loss_mlp": 1.02511501,
      "epoch": 0.1574627987374117,
      "flos": 22666922772480.0,
      "grad_norm": 2.6671900837979865,
      "language_loss": 0.78794169,
      "learning_rate": 3.8319786703223695e-06,
      "loss": 0.81005132,
      "num_input_tokens_seen": 56804975,
      "step": 2619,
      "time_per_iteration": 2.488884687423706
    },
    {
      "auxiliary_loss_clip": 0.01132177,
      "auxiliary_loss_mlp": 0.01050029,
      "balance_loss_clip": 1.05761254,
      "balance_loss_mlp": 1.03039479,
      "epoch": 0.15752292199007967,
      "flos": 16800664262400.0,
      "grad_norm": 1.8358367933463895,
      "language_loss": 0.76949549,
      "learning_rate": 3.831822382544101e-06,
      "loss": 0.79131758,
      "num_input_tokens_seen": 56822470,
      "step": 2620,
      "time_per_iteration": 3.9031753540039062
    },
    {
      "auxiliary_loss_clip": 0.01142328,
      "auxiliary_loss_mlp": 0.01052935,
      "balance_loss_clip": 1.05396438,
      "balance_loss_mlp": 1.03175139,
      "epoch": 0.15758304524274763,
      "flos": 29826002568960.0,
      "grad_norm": 1.785181033788137,
      "language_loss": 0.70912296,
      "learning_rate": 3.831666025302944e-06,
      "loss": 0.73107553,
      "num_input_tokens_seen": 56842100,
      "step": 2621,
      "time_per_iteration": 3.987882614135742
    },
    {
      "auxiliary_loss_clip": 0.01102618,
      "auxiliary_loss_mlp": 0.01050125,
      "balance_loss_clip": 1.05142379,
      "balance_loss_mlp": 1.02733171,
      "epoch": 0.1576431684954156,
      "flos": 53577426723840.0,
      "grad_norm": 2.4527770316609967,
      "language_loss": 0.72637117,
      "learning_rate": 3.831509598604828e-06,
      "loss": 0.74789858,
      "num_input_tokens_seen": 56865920,
      "step": 2622,
      "time_per_iteration": 2.8701534271240234
    },
    {
      "auxiliary_loss_clip": 0.0109935,
      "auxiliary_loss_mlp": 0.01047971,
      "balance_loss_clip": 1.05285454,
      "balance_loss_mlp": 1.02808726,
      "epoch": 0.15770329174808356,
      "flos": 20813609664000.0,
      "grad_norm": 1.762116110632521,
      "language_loss": 0.87398648,
      "learning_rate": 3.831353102455684e-06,
      "loss": 0.89545971,
      "num_input_tokens_seen": 56885265,
      "step": 2623,
      "time_per_iteration": 3.9830081462860107
    },
    {
      "auxiliary_loss_clip": 0.01162755,
      "auxiliary_loss_mlp": 0.01044541,
      "balance_loss_clip": 1.05709052,
      "balance_loss_mlp": 1.02549171,
      "epoch": 0.15776341500075153,
      "flos": 24974004395520.0,
      "grad_norm": 3.3152341447330844,
      "language_loss": 0.81674856,
      "learning_rate": 3.831196536861448e-06,
      "loss": 0.83882153,
      "num_input_tokens_seen": 56906710,
      "step": 2624,
      "time_per_iteration": 2.524467945098877
    },
    {
      "auxiliary_loss_clip": 0.01124708,
      "auxiliary_loss_mlp": 0.01052011,
      "balance_loss_clip": 1.05098176,
      "balance_loss_mlp": 1.03155518,
      "epoch": 0.15782353825341952,
      "flos": 21907915459200.0,
      "grad_norm": 2.29388260702212,
      "language_loss": 0.80304754,
      "learning_rate": 3.831039901828054e-06,
      "loss": 0.82481474,
      "num_input_tokens_seen": 56924275,
      "step": 2625,
      "time_per_iteration": 2.5585405826568604
    },
    {
      "auxiliary_loss_clip": 0.01161201,
      "auxiliary_loss_mlp": 0.01044705,
      "balance_loss_clip": 1.05663323,
      "balance_loss_mlp": 1.02571464,
      "epoch": 0.15788366150608749,
      "flos": 26177191292160.0,
      "grad_norm": 2.096946721628493,
      "language_loss": 0.80662417,
      "learning_rate": 3.830883197361445e-06,
      "loss": 0.82868326,
      "num_input_tokens_seen": 56941525,
      "step": 2626,
      "time_per_iteration": 2.514620542526245
    },
    {
      "auxiliary_loss_clip": 0.01099858,
      "auxiliary_loss_mlp": 0.01057298,
      "balance_loss_clip": 1.05168235,
      "balance_loss_mlp": 1.03523231,
      "epoch": 0.15794378475875545,
      "flos": 27709822753920.0,
      "grad_norm": 1.8903550082680596,
      "language_loss": 0.73452771,
      "learning_rate": 3.830726423467561e-06,
      "loss": 0.75609928,
      "num_input_tokens_seen": 56962145,
      "step": 2627,
      "time_per_iteration": 4.030405044555664
    },
    {
      "auxiliary_loss_clip": 0.01119838,
      "auxiliary_loss_mlp": 0.01051729,
      "balance_loss_clip": 1.05533814,
      "balance_loss_mlp": 1.03202415,
      "epoch": 0.15800390801142342,
      "flos": 12130158533760.0,
      "grad_norm": 2.479595297301469,
      "language_loss": 0.84946352,
      "learning_rate": 3.830569580152348e-06,
      "loss": 0.87117916,
      "num_input_tokens_seen": 56977505,
      "step": 2628,
      "time_per_iteration": 2.574367046356201
    },
    {
      "auxiliary_loss_clip": 0.01132314,
      "auxiliary_loss_mlp": 0.01040389,
      "balance_loss_clip": 1.05449271,
      "balance_loss_mlp": 1.02257872,
      "epoch": 0.15806403126409138,
      "flos": 20704728562560.0,
      "grad_norm": 1.9310631673366168,
      "language_loss": 0.7676568,
      "learning_rate": 3.830412667421752e-06,
      "loss": 0.78938383,
      "num_input_tokens_seen": 56996770,
      "step": 2629,
      "time_per_iteration": 2.533308982849121
    },
    {
      "auxiliary_loss_clip": 0.01146697,
      "auxiliary_loss_mlp": 0.01047947,
      "balance_loss_clip": 1.0554018,
      "balance_loss_mlp": 1.02797925,
      "epoch": 0.15812415451675935,
      "flos": 17821712269440.0,
      "grad_norm": 2.617215189417647,
      "language_loss": 0.73822027,
      "learning_rate": 3.8302556852817245e-06,
      "loss": 0.76016676,
      "num_input_tokens_seen": 57014970,
      "step": 2630,
      "time_per_iteration": 2.496027946472168
    },
    {
      "auxiliary_loss_clip": 0.0115365,
      "auxiliary_loss_mlp": 0.01048386,
      "balance_loss_clip": 1.05359316,
      "balance_loss_mlp": 1.02832305,
      "epoch": 0.15818427776942734,
      "flos": 20084048524800.0,
      "grad_norm": 3.4481435477759046,
      "language_loss": 0.83666772,
      "learning_rate": 3.8300986337382184e-06,
      "loss": 0.85868806,
      "num_input_tokens_seen": 57034045,
      "step": 2631,
      "time_per_iteration": 2.4975268840789795
    },
    {
      "auxiliary_loss_clip": 0.01161182,
      "auxiliary_loss_mlp": 0.01047184,
      "balance_loss_clip": 1.05502796,
      "balance_loss_mlp": 1.0283016,
      "epoch": 0.1582444010220953,
      "flos": 21214911386880.0,
      "grad_norm": 1.6876165232554075,
      "language_loss": 0.78216863,
      "learning_rate": 3.8299415127971895e-06,
      "loss": 0.80425227,
      "num_input_tokens_seen": 57053695,
      "step": 2632,
      "time_per_iteration": 2.4962499141693115
    },
    {
      "auxiliary_loss_clip": 0.01154114,
      "auxiliary_loss_mlp": 0.01058494,
      "balance_loss_clip": 1.05756056,
      "balance_loss_mlp": 1.03882432,
      "epoch": 0.15830452427476327,
      "flos": 17858341163520.0,
      "grad_norm": 2.411049366150869,
      "language_loss": 0.83279407,
      "learning_rate": 3.829784322464594e-06,
      "loss": 0.85492015,
      "num_input_tokens_seen": 57071290,
      "step": 2633,
      "time_per_iteration": 2.481889486312866
    },
    {
      "auxiliary_loss_clip": 0.01166307,
      "auxiliary_loss_mlp": 0.0104524,
      "balance_loss_clip": 1.05930114,
      "balance_loss_mlp": 1.02588093,
      "epoch": 0.15836464752743123,
      "flos": 24534960456960.0,
      "grad_norm": 1.633802985447358,
      "language_loss": 0.77220398,
      "learning_rate": 3.829627062746394e-06,
      "loss": 0.79431939,
      "num_input_tokens_seen": 57091465,
      "step": 2634,
      "time_per_iteration": 2.521247625350952
    },
    {
      "auxiliary_loss_clip": 0.01127288,
      "auxiliary_loss_mlp": 0.00822557,
      "balance_loss_clip": 1.05660152,
      "balance_loss_mlp": 1.04901791,
      "epoch": 0.1584247707800992,
      "flos": 20120821073280.0,
      "grad_norm": 2.0312836582656724,
      "language_loss": 0.88842142,
      "learning_rate": 3.829469733648552e-06,
      "loss": 0.90791988,
      "num_input_tokens_seen": 57110075,
      "step": 2635,
      "time_per_iteration": 2.5634281635284424
    },
    {
      "auxiliary_loss_clip": 0.01095231,
      "auxiliary_loss_mlp": 0.01052254,
      "balance_loss_clip": 1.05062938,
      "balance_loss_mlp": 1.03240561,
      "epoch": 0.15848489403276717,
      "flos": 20375966355840.0,
      "grad_norm": 2.161211880770881,
      "language_loss": 0.75658792,
      "learning_rate": 3.829312335177034e-06,
      "loss": 0.77806282,
      "num_input_tokens_seen": 57128945,
      "step": 2636,
      "time_per_iteration": 2.6574482917785645
    },
    {
      "auxiliary_loss_clip": 0.01121429,
      "auxiliary_loss_mlp": 0.01045384,
      "balance_loss_clip": 1.05568707,
      "balance_loss_mlp": 1.02464163,
      "epoch": 0.15854501728543513,
      "flos": 39346890359040.0,
      "grad_norm": 2.4803510457512474,
      "language_loss": 0.72497612,
      "learning_rate": 3.82915486733781e-06,
      "loss": 0.74664426,
      "num_input_tokens_seen": 57152385,
      "step": 2637,
      "time_per_iteration": 2.70961594581604
    },
    {
      "auxiliary_loss_clip": 0.01149333,
      "auxiliary_loss_mlp": 0.01048343,
      "balance_loss_clip": 1.05533028,
      "balance_loss_mlp": 1.02974629,
      "epoch": 0.15860514053810312,
      "flos": 24864225454080.0,
      "grad_norm": 2.1167733594815497,
      "language_loss": 0.77890044,
      "learning_rate": 3.82899733013685e-06,
      "loss": 0.80087721,
      "num_input_tokens_seen": 57172620,
      "step": 2638,
      "time_per_iteration": 2.543856382369995
    },
    {
      "auxiliary_loss_clip": 0.0112673,
      "auxiliary_loss_mlp": 0.0106155,
      "balance_loss_clip": 1.05089581,
      "balance_loss_mlp": 1.0409863,
      "epoch": 0.1586652637907711,
      "flos": 26177694082560.0,
      "grad_norm": 1.9338920979058634,
      "language_loss": 0.75877416,
      "learning_rate": 3.828839723580128e-06,
      "loss": 0.78065705,
      "num_input_tokens_seen": 57194680,
      "step": 2639,
      "time_per_iteration": 2.610563278198242
    },
    {
      "auxiliary_loss_clip": 0.01100198,
      "auxiliary_loss_mlp": 0.01064608,
      "balance_loss_clip": 1.04998398,
      "balance_loss_mlp": 1.04448533,
      "epoch": 0.15872538704343905,
      "flos": 19792058866560.0,
      "grad_norm": 1.824889954406104,
      "language_loss": 0.8116073,
      "learning_rate": 3.82868204767362e-06,
      "loss": 0.83325535,
      "num_input_tokens_seen": 57214675,
      "step": 2640,
      "time_per_iteration": 2.6685256958007812
    },
    {
      "auxiliary_loss_clip": 0.01129693,
      "auxiliary_loss_mlp": 0.01062522,
      "balance_loss_clip": 1.04976082,
      "balance_loss_mlp": 1.04157734,
      "epoch": 0.15878551029610702,
      "flos": 28475366342400.0,
      "grad_norm": 1.3898970448036656,
      "language_loss": 0.66897982,
      "learning_rate": 3.828524302423306e-06,
      "loss": 0.69090199,
      "num_input_tokens_seen": 57235830,
      "step": 2641,
      "time_per_iteration": 2.5933640003204346
    },
    {
      "auxiliary_loss_clip": 0.01141186,
      "auxiliary_loss_mlp": 0.01055763,
      "balance_loss_clip": 1.05005074,
      "balance_loss_mlp": 1.0358665,
      "epoch": 0.15884563354877498,
      "flos": 24206701040640.0,
      "grad_norm": 2.948118422299727,
      "language_loss": 0.75368565,
      "learning_rate": 3.828366487835167e-06,
      "loss": 0.77565515,
      "num_input_tokens_seen": 57255970,
      "step": 2642,
      "time_per_iteration": 2.5654218196868896
    },
    {
      "auxiliary_loss_clip": 0.01152968,
      "auxiliary_loss_mlp": 0.0105584,
      "balance_loss_clip": 1.05715704,
      "balance_loss_mlp": 1.03639734,
      "epoch": 0.15890575680144295,
      "flos": 23949795991680.0,
      "grad_norm": 1.8657306217653318,
      "language_loss": 0.70521414,
      "learning_rate": 3.828208603915186e-06,
      "loss": 0.72730225,
      "num_input_tokens_seen": 57274435,
      "step": 2643,
      "time_per_iteration": 2.5392372608184814
    },
    {
      "auxiliary_loss_clip": 0.01160129,
      "auxiliary_loss_mlp": 0.01047644,
      "balance_loss_clip": 1.05610108,
      "balance_loss_mlp": 1.0296073,
      "epoch": 0.15896588005411091,
      "flos": 21215019127680.0,
      "grad_norm": 1.9825744542407426,
      "language_loss": 0.78503489,
      "learning_rate": 3.828050650669353e-06,
      "loss": 0.80711257,
      "num_input_tokens_seen": 57293115,
      "step": 2644,
      "time_per_iteration": 2.4848146438598633
    },
    {
      "auxiliary_loss_clip": 0.01148743,
      "auxiliary_loss_mlp": 0.01055261,
      "balance_loss_clip": 1.05507874,
      "balance_loss_mlp": 1.03646207,
      "epoch": 0.1590260033067789,
      "flos": 24352390604160.0,
      "grad_norm": 2.1094900718985548,
      "language_loss": 0.82461941,
      "learning_rate": 3.827892628103657e-06,
      "loss": 0.84665942,
      "num_input_tokens_seen": 57312565,
      "step": 2645,
      "time_per_iteration": 2.537905693054199
    },
    {
      "auxiliary_loss_clip": 0.0116056,
      "auxiliary_loss_mlp": 0.010586,
      "balance_loss_clip": 1.05173063,
      "balance_loss_mlp": 1.03841782,
      "epoch": 0.15908612655944687,
      "flos": 32048944583040.0,
      "grad_norm": 2.002229436167892,
      "language_loss": 0.70346981,
      "learning_rate": 3.827734536224087e-06,
      "loss": 0.7256614,
      "num_input_tokens_seen": 57333360,
      "step": 2646,
      "time_per_iteration": 2.582068681716919
    },
    {
      "auxiliary_loss_clip": 0.01131172,
      "auxiliary_loss_mlp": 0.01047677,
      "balance_loss_clip": 1.05218029,
      "balance_loss_mlp": 1.02896059,
      "epoch": 0.15914624981211484,
      "flos": 17785370684160.0,
      "grad_norm": 3.300115635235867,
      "language_loss": 0.62569183,
      "learning_rate": 3.827576375036642e-06,
      "loss": 0.64748031,
      "num_input_tokens_seen": 57350575,
      "step": 2647,
      "time_per_iteration": 2.517565965652466
    },
    {
      "auxiliary_loss_clip": 0.01159617,
      "auxiliary_loss_mlp": 0.01048813,
      "balance_loss_clip": 1.05520296,
      "balance_loss_mlp": 1.02998996,
      "epoch": 0.1592063730647828,
      "flos": 17712507945600.0,
      "grad_norm": 1.9507275686818082,
      "language_loss": 0.89596862,
      "learning_rate": 3.827418144547318e-06,
      "loss": 0.91805291,
      "num_input_tokens_seen": 57367570,
      "step": 2648,
      "time_per_iteration": 2.471160650253296
    },
    {
      "auxiliary_loss_clip": 0.01155235,
      "auxiliary_loss_mlp": 0.0104797,
      "balance_loss_clip": 1.05411148,
      "balance_loss_mlp": 1.03049397,
      "epoch": 0.15926649631745077,
      "flos": 18803545603200.0,
      "grad_norm": 1.8869840243088019,
      "language_loss": 0.91607773,
      "learning_rate": 3.827259844762114e-06,
      "loss": 0.93810976,
      "num_input_tokens_seen": 57383980,
      "step": 2649,
      "time_per_iteration": 2.438542604446411
    },
    {
      "auxiliary_loss_clip": 0.01092358,
      "auxiliary_loss_mlp": 0.01044758,
      "balance_loss_clip": 1.04958189,
      "balance_loss_mlp": 1.02440882,
      "epoch": 0.15932661957011873,
      "flos": 17566243764480.0,
      "grad_norm": 3.2032205815575083,
      "language_loss": 0.71672189,
      "learning_rate": 3.827101475687033e-06,
      "loss": 0.73809302,
      "num_input_tokens_seen": 57400840,
      "step": 2650,
      "time_per_iteration": 2.624683380126953
    },
    {
      "auxiliary_loss_clip": 0.01144273,
      "auxiliary_loss_mlp": 0.01044331,
      "balance_loss_clip": 1.05125725,
      "balance_loss_mlp": 1.02692628,
      "epoch": 0.15938674282278673,
      "flos": 13334351011200.0,
      "grad_norm": 1.946444085560815,
      "language_loss": 0.71285963,
      "learning_rate": 3.826943037328082e-06,
      "loss": 0.73474568,
      "num_input_tokens_seen": 57419230,
      "step": 2651,
      "time_per_iteration": 2.4705519676208496
    },
    {
      "auxiliary_loss_clip": 0.01118748,
      "auxiliary_loss_mlp": 0.00817511,
      "balance_loss_clip": 1.05197144,
      "balance_loss_mlp": 1.03868854,
      "epoch": 0.1594468660754547,
      "flos": 22488842119680.0,
      "grad_norm": 1.7946529271345346,
      "language_loss": 0.799007,
      "learning_rate": 3.8267845296912674e-06,
      "loss": 0.81836963,
      "num_input_tokens_seen": 57439315,
      "step": 2652,
      "time_per_iteration": 2.621919870376587
    },
    {
      "auxiliary_loss_clip": 0.01133714,
      "auxiliary_loss_mlp": 0.00817177,
      "balance_loss_clip": 1.05190325,
      "balance_loss_mlp": 1.04060185,
      "epoch": 0.15950698932812266,
      "flos": 15007320910080.0,
      "grad_norm": 2.5720532530749685,
      "language_loss": 0.70038092,
      "learning_rate": 3.826625952782601e-06,
      "loss": 0.71988988,
      "num_input_tokens_seen": 57454635,
      "step": 2653,
      "time_per_iteration": 2.493992328643799
    },
    {
      "auxiliary_loss_clip": 0.01143734,
      "auxiliary_loss_mlp": 0.01039332,
      "balance_loss_clip": 1.05204964,
      "balance_loss_mlp": 1.02043772,
      "epoch": 0.15956711258079062,
      "flos": 30155052084480.0,
      "grad_norm": 2.0521348935328776,
      "language_loss": 0.76803803,
      "learning_rate": 3.826467306608095e-06,
      "loss": 0.78986865,
      "num_input_tokens_seen": 57476805,
      "step": 2654,
      "time_per_iteration": 2.5978341102600098
    },
    {
      "auxiliary_loss_clip": 0.01115774,
      "auxiliary_loss_mlp": 0.01043785,
      "balance_loss_clip": 1.05055571,
      "balance_loss_mlp": 1.02511716,
      "epoch": 0.1596272358334586,
      "flos": 21032700670080.0,
      "grad_norm": 1.792724133478427,
      "language_loss": 0.82296503,
      "learning_rate": 3.826308591173765e-06,
      "loss": 0.84456062,
      "num_input_tokens_seen": 57496400,
      "step": 2655,
      "time_per_iteration": 2.579822063446045
    },
    {
      "auxiliary_loss_clip": 0.01118326,
      "auxiliary_loss_mlp": 0.01043738,
      "balance_loss_clip": 1.05385828,
      "balance_loss_mlp": 1.02533197,
      "epoch": 0.15968735908612655,
      "flos": 15268032800640.0,
      "grad_norm": 1.829073733520542,
      "language_loss": 0.73182213,
      "learning_rate": 3.826149806485631e-06,
      "loss": 0.75344276,
      "num_input_tokens_seen": 57513700,
      "step": 2656,
      "time_per_iteration": 2.560667037963867
    },
    {
      "auxiliary_loss_clip": 0.01115151,
      "auxiliary_loss_mlp": 0.0104495,
      "balance_loss_clip": 1.05292308,
      "balance_loss_mlp": 1.02675855,
      "epoch": 0.15974748233879452,
      "flos": 52665726695040.0,
      "grad_norm": 2.552769703960878,
      "language_loss": 0.77842367,
      "learning_rate": 3.825990952549713e-06,
      "loss": 0.80002463,
      "num_input_tokens_seen": 57536180,
      "step": 2657,
      "time_per_iteration": 2.8569767475128174
    },
    {
      "auxiliary_loss_clip": 0.01143549,
      "auxiliary_loss_mlp": 0.01048259,
      "balance_loss_clip": 1.05322146,
      "balance_loss_mlp": 1.02888799,
      "epoch": 0.1598076055914625,
      "flos": 18733232730240.0,
      "grad_norm": 1.7255591243680666,
      "language_loss": 0.74017644,
      "learning_rate": 3.825832029372035e-06,
      "loss": 0.76209456,
      "num_input_tokens_seen": 57555025,
      "step": 2658,
      "time_per_iteration": 3.867783784866333
    },
    {
      "auxiliary_loss_clip": 0.01129995,
      "auxiliary_loss_mlp": 0.01051151,
      "balance_loss_clip": 1.05215693,
      "balance_loss_mlp": 1.02891827,
      "epoch": 0.15986772884413047,
      "flos": 34349238535680.0,
      "grad_norm": 1.9309510807958796,
      "language_loss": 0.7556963,
      "learning_rate": 3.825673036958624e-06,
      "loss": 0.77750778,
      "num_input_tokens_seen": 57577660,
      "step": 2659,
      "time_per_iteration": 4.059804916381836
    },
    {
      "auxiliary_loss_clip": 0.01124172,
      "auxiliary_loss_mlp": 0.01058702,
      "balance_loss_clip": 1.05380368,
      "balance_loss_mlp": 1.03868699,
      "epoch": 0.15992785209679844,
      "flos": 22054969739520.0,
      "grad_norm": 2.1394496725422174,
      "language_loss": 0.9068619,
      "learning_rate": 3.825513975315508e-06,
      "loss": 0.92869055,
      "num_input_tokens_seen": 57596335,
      "step": 2660,
      "time_per_iteration": 2.578277587890625
    },
    {
      "auxiliary_loss_clip": 0.01112146,
      "auxiliary_loss_mlp": 0.01062072,
      "balance_loss_clip": 1.05358768,
      "balance_loss_mlp": 1.04035187,
      "epoch": 0.1599879753494664,
      "flos": 33066652625280.0,
      "grad_norm": 1.6972217622954069,
      "language_loss": 0.77824885,
      "learning_rate": 3.82535484444872e-06,
      "loss": 0.79999095,
      "num_input_tokens_seen": 57616830,
      "step": 2661,
      "time_per_iteration": 4.143225908279419
    },
    {
      "auxiliary_loss_clip": 0.01133866,
      "auxiliary_loss_mlp": 0.00810849,
      "balance_loss_clip": 1.04980874,
      "balance_loss_mlp": 1.02856135,
      "epoch": 0.16004809860213437,
      "flos": 28038010343040.0,
      "grad_norm": 1.9451730892349675,
      "language_loss": 0.74215937,
      "learning_rate": 3.825195644364292e-06,
      "loss": 0.76160651,
      "num_input_tokens_seen": 57635515,
      "step": 2662,
      "time_per_iteration": 2.5956945419311523
    },
    {
      "auxiliary_loss_clip": 0.01135404,
      "auxiliary_loss_mlp": 0.00814726,
      "balance_loss_clip": 1.0543046,
      "balance_loss_mlp": 1.02848375,
      "epoch": 0.16010822185480234,
      "flos": 22780113505920.0,
      "grad_norm": 2.2731088562404795,
      "language_loss": 0.82065177,
      "learning_rate": 3.825036375068263e-06,
      "loss": 0.84015298,
      "num_input_tokens_seen": 57654250,
      "step": 2663,
      "time_per_iteration": 2.5595338344573975
    },
    {
      "auxiliary_loss_clip": 0.01119153,
      "auxiliary_loss_mlp": 0.01055359,
      "balance_loss_clip": 1.05884743,
      "balance_loss_mlp": 1.03582036,
      "epoch": 0.16016834510747033,
      "flos": 20084012611200.0,
      "grad_norm": 2.6667364800866964,
      "language_loss": 0.79649299,
      "learning_rate": 3.824877036566672e-06,
      "loss": 0.81823814,
      "num_input_tokens_seen": 57672645,
      "step": 2664,
      "time_per_iteration": 2.596796751022339
    },
    {
      "auxiliary_loss_clip": 0.01151995,
      "auxiliary_loss_mlp": 0.010507,
      "balance_loss_clip": 1.05370629,
      "balance_loss_mlp": 1.03109038,
      "epoch": 0.1602284683601383,
      "flos": 21173829206400.0,
      "grad_norm": 2.0726836612525803,
      "language_loss": 0.94360006,
      "learning_rate": 3.824717628865561e-06,
      "loss": 0.96562696,
      "num_input_tokens_seen": 57691055,
      "step": 2665,
      "time_per_iteration": 3.9174795150756836
    },
    {
      "auxiliary_loss_clip": 0.01124236,
      "auxiliary_loss_mlp": 0.01048907,
      "balance_loss_clip": 1.05014634,
      "balance_loss_mlp": 1.02833188,
      "epoch": 0.16028859161280626,
      "flos": 14647568244480.0,
      "grad_norm": 2.0535486574164223,
      "language_loss": 0.8491565,
      "learning_rate": 3.824558151970974e-06,
      "loss": 0.87088788,
      "num_input_tokens_seen": 57707235,
      "step": 2666,
      "time_per_iteration": 2.5604076385498047
    },
    {
      "auxiliary_loss_clip": 0.01132116,
      "auxiliary_loss_mlp": 0.00805964,
      "balance_loss_clip": 1.05669987,
      "balance_loss_mlp": 1.01958537,
      "epoch": 0.16034871486547422,
      "flos": 20990325600000.0,
      "grad_norm": 3.2321630582335192,
      "language_loss": 0.81585872,
      "learning_rate": 3.8243986058889595e-06,
      "loss": 0.83523953,
      "num_input_tokens_seen": 57724190,
      "step": 2667,
      "time_per_iteration": 2.5674827098846436
    },
    {
      "auxiliary_loss_clip": 0.01160824,
      "auxiliary_loss_mlp": 0.01052651,
      "balance_loss_clip": 1.05561233,
      "balance_loss_mlp": 1.03274298,
      "epoch": 0.1604088381181422,
      "flos": 21397732634880.0,
      "grad_norm": 1.8128773187372504,
      "language_loss": 0.7402916,
      "learning_rate": 3.824238990625567e-06,
      "loss": 0.76242638,
      "num_input_tokens_seen": 57743620,
      "step": 2668,
      "time_per_iteration": 2.492417097091675
    },
    {
      "auxiliary_loss_clip": 0.01147232,
      "auxiliary_loss_mlp": 0.01049533,
      "balance_loss_clip": 1.05566549,
      "balance_loss_mlp": 1.02960134,
      "epoch": 0.16046896137081015,
      "flos": 23877040993920.0,
      "grad_norm": 2.045418817292727,
      "language_loss": 0.77290076,
      "learning_rate": 3.824079306186848e-06,
      "loss": 0.79486835,
      "num_input_tokens_seen": 57764810,
      "step": 2669,
      "time_per_iteration": 2.5273842811584473
    },
    {
      "auxiliary_loss_clip": 0.01078794,
      "auxiliary_loss_mlp": 0.01053905,
      "balance_loss_clip": 1.04636586,
      "balance_loss_mlp": 1.05059052,
      "epoch": 0.16052908462347812,
      "flos": 59806709015040.0,
      "grad_norm": 0.8082869695151813,
      "language_loss": 0.55556238,
      "learning_rate": 3.823919552578861e-06,
      "loss": 0.5768894,
      "num_input_tokens_seen": 57824390,
      "step": 2670,
      "time_per_iteration": 3.0069684982299805
    },
    {
      "auxiliary_loss_clip": 0.01145902,
      "auxiliary_loss_mlp": 0.01045369,
      "balance_loss_clip": 1.05050063,
      "balance_loss_mlp": 1.02577078,
      "epoch": 0.1605892078761461,
      "flos": 18296559089280.0,
      "grad_norm": 2.243360587984137,
      "language_loss": 0.77546376,
      "learning_rate": 3.82375972980766e-06,
      "loss": 0.79737651,
      "num_input_tokens_seen": 57843665,
      "step": 2671,
      "time_per_iteration": 2.4860310554504395
    },
    {
      "auxiliary_loss_clip": 0.01149399,
      "auxiliary_loss_mlp": 0.01045514,
      "balance_loss_clip": 1.05460429,
      "balance_loss_mlp": 1.02607107,
      "epoch": 0.16064933112881408,
      "flos": 32160734686080.0,
      "grad_norm": 1.941381399036769,
      "language_loss": 0.6496228,
      "learning_rate": 3.8235998378793086e-06,
      "loss": 0.67157197,
      "num_input_tokens_seen": 57863305,
      "step": 2672,
      "time_per_iteration": 2.5939865112304688
    },
    {
      "auxiliary_loss_clip": 0.01145771,
      "auxiliary_loss_mlp": 0.01044762,
      "balance_loss_clip": 1.05370796,
      "balance_loss_mlp": 1.02351892,
      "epoch": 0.16070945438148204,
      "flos": 19828795501440.0,
      "grad_norm": 2.064521066290296,
      "language_loss": 0.85700703,
      "learning_rate": 3.8234398767998675e-06,
      "loss": 0.87891233,
      "num_input_tokens_seen": 57883025,
      "step": 2673,
      "time_per_iteration": 2.478292942047119
    },
    {
      "auxiliary_loss_clip": 0.01126271,
      "auxiliary_loss_mlp": 0.01048727,
      "balance_loss_clip": 1.05585647,
      "balance_loss_mlp": 1.02972484,
      "epoch": 0.16076957763415,
      "flos": 18913144976640.0,
      "grad_norm": 2.536638039839282,
      "language_loss": 0.72950184,
      "learning_rate": 3.823279846575403e-06,
      "loss": 0.75125176,
      "num_input_tokens_seen": 57901430,
      "step": 2674,
      "time_per_iteration": 2.5571987628936768
    },
    {
      "auxiliary_loss_clip": 0.01145474,
      "auxiliary_loss_mlp": 0.01043662,
      "balance_loss_clip": 1.04974127,
      "balance_loss_mlp": 1.02212143,
      "epoch": 0.16082970088681797,
      "flos": 16764358590720.0,
      "grad_norm": 1.6752462353462072,
      "language_loss": 0.84308785,
      "learning_rate": 3.823119747211986e-06,
      "loss": 0.86497927,
      "num_input_tokens_seen": 57919550,
      "step": 2675,
      "time_per_iteration": 2.4934885501861572
    },
    {
      "auxiliary_loss_clip": 0.01112207,
      "auxiliary_loss_mlp": 0.01051017,
      "balance_loss_clip": 1.05016637,
      "balance_loss_mlp": 1.02935624,
      "epoch": 0.16088982413948594,
      "flos": 35150261783040.0,
      "grad_norm": 1.964529401307919,
      "language_loss": 0.82749355,
      "learning_rate": 3.822959578715685e-06,
      "loss": 0.84912586,
      "num_input_tokens_seen": 57939890,
      "step": 2676,
      "time_per_iteration": 2.703517198562622
    },
    {
      "auxiliary_loss_clip": 0.01146499,
      "auxiliary_loss_mlp": 0.01046412,
      "balance_loss_clip": 1.05404162,
      "balance_loss_mlp": 1.02838731,
      "epoch": 0.1609499473921539,
      "flos": 18625105814400.0,
      "grad_norm": 1.913289631672295,
      "language_loss": 0.73580086,
      "learning_rate": 3.822799341092573e-06,
      "loss": 0.75773001,
      "num_input_tokens_seen": 57957410,
      "step": 2677,
      "time_per_iteration": 2.4746367931365967
    },
    {
      "auxiliary_loss_clip": 0.01136591,
      "auxiliary_loss_mlp": 0.0104217,
      "balance_loss_clip": 1.05234051,
      "balance_loss_mlp": 1.02236903,
      "epoch": 0.1610100706448219,
      "flos": 33145728416640.0,
      "grad_norm": 1.922411149232369,
      "language_loss": 0.76130587,
      "learning_rate": 3.822639034348728e-06,
      "loss": 0.78309351,
      "num_input_tokens_seen": 57977900,
      "step": 2678,
      "time_per_iteration": 2.642801523208618
    },
    {
      "auxiliary_loss_clip": 0.01146102,
      "auxiliary_loss_mlp": 0.01041986,
      "balance_loss_clip": 1.05253959,
      "balance_loss_mlp": 1.02177954,
      "epoch": 0.16107019389748986,
      "flos": 34676707852800.0,
      "grad_norm": 2.3976193499428806,
      "language_loss": 0.70637459,
      "learning_rate": 3.822478658490228e-06,
      "loss": 0.72825545,
      "num_input_tokens_seen": 57998210,
      "step": 2679,
      "time_per_iteration": 2.6104655265808105
    },
    {
      "auxiliary_loss_clip": 0.0104153,
      "auxiliary_loss_mlp": 0.01292318,
      "balance_loss_clip": 1.03249192,
      "balance_loss_mlp": 1.93311465,
      "epoch": 0.16113031715015783,
      "flos": 65713403260800.0,
      "grad_norm": 0.8203695036103317,
      "language_loss": 0.51809406,
      "learning_rate": 3.822318213523154e-06,
      "loss": 0.54143256,
      "num_input_tokens_seen": 58059420,
      "step": 2680,
      "time_per_iteration": 3.2261202335357666
    },
    {
      "auxiliary_loss_clip": 0.01139868,
      "auxiliary_loss_mlp": 0.01048515,
      "balance_loss_clip": 1.05118573,
      "balance_loss_mlp": 1.02677083,
      "epoch": 0.1611904404028258,
      "flos": 20810413353600.0,
      "grad_norm": 19.173513603471626,
      "language_loss": 0.80556095,
      "learning_rate": 3.8221576994535925e-06,
      "loss": 0.82744479,
      "num_input_tokens_seen": 58078370,
      "step": 2681,
      "time_per_iteration": 2.526144027709961
    },
    {
      "auxiliary_loss_clip": 0.011337,
      "auxiliary_loss_mlp": 0.01055404,
      "balance_loss_clip": 1.05234659,
      "balance_loss_mlp": 1.03596091,
      "epoch": 0.16125056365549376,
      "flos": 27013335062400.0,
      "grad_norm": 1.9641037547900773,
      "language_loss": 0.68823719,
      "learning_rate": 3.821997116287627e-06,
      "loss": 0.71012831,
      "num_input_tokens_seen": 58097395,
      "step": 2682,
      "time_per_iteration": 2.590691089630127
    },
    {
      "auxiliary_loss_clip": 0.01134463,
      "auxiliary_loss_mlp": 0.01052011,
      "balance_loss_clip": 1.05395436,
      "balance_loss_mlp": 1.03147149,
      "epoch": 0.16131068690816172,
      "flos": 19276524915840.0,
      "grad_norm": 1.8904593025900922,
      "language_loss": 0.87582201,
      "learning_rate": 3.821836464031348e-06,
      "loss": 0.89768672,
      "num_input_tokens_seen": 58115630,
      "step": 2683,
      "time_per_iteration": 2.5146102905273438
    },
    {
      "auxiliary_loss_clip": 0.0115975,
      "auxiliary_loss_mlp": 0.01050858,
      "balance_loss_clip": 1.05319107,
      "balance_loss_mlp": 1.03066361,
      "epoch": 0.16137081016082971,
      "flos": 35337931367040.0,
      "grad_norm": 2.7716169471614602,
      "language_loss": 0.74349028,
      "learning_rate": 3.821675742690849e-06,
      "loss": 0.76559633,
      "num_input_tokens_seen": 58138655,
      "step": 2684,
      "time_per_iteration": 2.62308931350708
    },
    {
      "auxiliary_loss_clip": 0.01135941,
      "auxiliary_loss_mlp": 0.00904749,
      "balance_loss_clip": 1.05666447,
      "balance_loss_mlp": 1.20612693,
      "epoch": 0.16143093341349768,
      "flos": 34235257703040.0,
      "grad_norm": 2.0487747295492396,
      "language_loss": 0.70849597,
      "learning_rate": 3.821514952272223e-06,
      "loss": 0.72890282,
      "num_input_tokens_seen": 58157440,
      "step": 2685,
      "time_per_iteration": 2.6421124935150146
    },
    {
      "auxiliary_loss_clip": 0.01108729,
      "auxiliary_loss_mlp": 0.01061225,
      "balance_loss_clip": 1.04943681,
      "balance_loss_mlp": 1.04029191,
      "epoch": 0.16149105666616564,
      "flos": 27999262546560.0,
      "grad_norm": 1.9709972730415162,
      "language_loss": 0.71785879,
      "learning_rate": 3.821354092781567e-06,
      "loss": 0.73955834,
      "num_input_tokens_seen": 58176660,
      "step": 2686,
      "time_per_iteration": 2.628288984298706
    },
    {
      "auxiliary_loss_clip": 0.01151475,
      "auxiliary_loss_mlp": 0.01050293,
      "balance_loss_clip": 1.0534507,
      "balance_loss_mlp": 1.03005087,
      "epoch": 0.1615511799188336,
      "flos": 19422214479360.0,
      "grad_norm": 1.8580722792049957,
      "language_loss": 0.82284951,
      "learning_rate": 3.821193164224981e-06,
      "loss": 0.84486723,
      "num_input_tokens_seen": 58195085,
      "step": 2687,
      "time_per_iteration": 2.4869236946105957
    },
    {
      "auxiliary_loss_clip": 0.01149124,
      "auxiliary_loss_mlp": 0.01053927,
      "balance_loss_clip": 1.04907477,
      "balance_loss_mlp": 1.03201663,
      "epoch": 0.16161130317150157,
      "flos": 22854915578880.0,
      "grad_norm": 1.7549357476938325,
      "language_loss": 0.71679032,
      "learning_rate": 3.821032166608568e-06,
      "loss": 0.73882091,
      "num_input_tokens_seen": 58213540,
      "step": 2688,
      "time_per_iteration": 2.499110460281372
    },
    {
      "auxiliary_loss_clip": 0.01118553,
      "auxiliary_loss_mlp": 0.01048761,
      "balance_loss_clip": 1.05201614,
      "balance_loss_mlp": 1.02966356,
      "epoch": 0.16167142642416954,
      "flos": 26110577520000.0,
      "grad_norm": 1.5681496384520943,
      "language_loss": 0.75843716,
      "learning_rate": 3.8208710999384325e-06,
      "loss": 0.78011024,
      "num_input_tokens_seen": 58236995,
      "step": 2689,
      "time_per_iteration": 2.635648012161255
    },
    {
      "auxiliary_loss_clip": 0.01161245,
      "auxiliary_loss_mlp": 0.0105629,
      "balance_loss_clip": 1.05544615,
      "balance_loss_mlp": 1.03584576,
      "epoch": 0.1617315496768375,
      "flos": 22779646629120.0,
      "grad_norm": 1.8666053779539216,
      "language_loss": 0.87753564,
      "learning_rate": 3.820709964220683e-06,
      "loss": 0.89971095,
      "num_input_tokens_seen": 58257230,
      "step": 2690,
      "time_per_iteration": 2.5588769912719727
    },
    {
      "auxiliary_loss_clip": 0.01143548,
      "auxiliary_loss_mlp": 0.01046307,
      "balance_loss_clip": 1.05520725,
      "balance_loss_mlp": 1.02830637,
      "epoch": 0.1617916729295055,
      "flos": 22017299351040.0,
      "grad_norm": 1.7010302015746661,
      "language_loss": 0.88332486,
      "learning_rate": 3.8205487594614284e-06,
      "loss": 0.90522337,
      "num_input_tokens_seen": 58277080,
      "step": 2691,
      "time_per_iteration": 2.5193605422973633
    },
    {
      "auxiliary_loss_clip": 0.01146197,
      "auxiliary_loss_mlp": 0.01048051,
      "balance_loss_clip": 1.05163872,
      "balance_loss_mlp": 1.02658141,
      "epoch": 0.16185179618217346,
      "flos": 23438248450560.0,
      "grad_norm": 2.532509606446305,
      "language_loss": 0.82574093,
      "learning_rate": 3.820387485666784e-06,
      "loss": 0.84768337,
      "num_input_tokens_seen": 58294815,
      "step": 2692,
      "time_per_iteration": 2.512941837310791
    },
    {
      "auxiliary_loss_clip": 0.01162553,
      "auxiliary_loss_mlp": 0.01054436,
      "balance_loss_clip": 1.051404,
      "balance_loss_mlp": 1.03391981,
      "epoch": 0.16191191943484143,
      "flos": 25666110627840.0,
      "grad_norm": 2.4320561661533247,
      "language_loss": 0.81254935,
      "learning_rate": 3.820226142842862e-06,
      "loss": 0.8347193,
      "num_input_tokens_seen": 58313215,
      "step": 2693,
      "time_per_iteration": 2.5429534912109375
    },
    {
      "auxiliary_loss_clip": 0.01155519,
      "auxiliary_loss_mlp": 0.01058818,
      "balance_loss_clip": 1.05269861,
      "balance_loss_mlp": 1.04014993,
      "epoch": 0.1619720426875094,
      "flos": 23477355383040.0,
      "grad_norm": 1.4476927854723711,
      "language_loss": 0.83949208,
      "learning_rate": 3.820064730995783e-06,
      "loss": 0.86163551,
      "num_input_tokens_seen": 58333215,
      "step": 2694,
      "time_per_iteration": 2.5380687713623047
    },
    {
      "auxiliary_loss_clip": 0.01110841,
      "auxiliary_loss_mlp": 0.01064812,
      "balance_loss_clip": 1.04859245,
      "balance_loss_mlp": 1.04368806,
      "epoch": 0.16203216594017736,
      "flos": 24133658734080.0,
      "grad_norm": 2.0655928621359414,
      "language_loss": 0.68864882,
      "learning_rate": 3.819903250131667e-06,
      "loss": 0.71040535,
      "num_input_tokens_seen": 58351160,
      "step": 2695,
      "time_per_iteration": 2.578021764755249
    },
    {
      "auxiliary_loss_clip": 0.01149614,
      "auxiliary_loss_mlp": 0.01057184,
      "balance_loss_clip": 1.05414295,
      "balance_loss_mlp": 1.03648901,
      "epoch": 0.16209228919284532,
      "flos": 22340889999360.0,
      "grad_norm": 2.1456998644879106,
      "language_loss": 0.82850838,
      "learning_rate": 3.819741700256637e-06,
      "loss": 0.8505764,
      "num_input_tokens_seen": 58368505,
      "step": 2696,
      "time_per_iteration": 2.5328567028045654
    },
    {
      "auxiliary_loss_clip": 0.01167228,
      "auxiliary_loss_mlp": 0.0105814,
      "balance_loss_clip": 1.05514598,
      "balance_loss_mlp": 1.03682566,
      "epoch": 0.1621524124455133,
      "flos": 15815131827840.0,
      "grad_norm": 2.31476607850295,
      "language_loss": 0.88740689,
      "learning_rate": 3.8195800813768194e-06,
      "loss": 0.90966058,
      "num_input_tokens_seen": 58385085,
      "step": 2697,
      "time_per_iteration": 3.868121385574341
    },
    {
      "auxiliary_loss_clip": 0.01152294,
      "auxiliary_loss_mlp": 0.01047971,
      "balance_loss_clip": 1.05025125,
      "balance_loss_mlp": 1.02944541,
      "epoch": 0.16221253569818128,
      "flos": 30186688988160.0,
      "grad_norm": 1.457150840493522,
      "language_loss": 0.80595386,
      "learning_rate": 3.819418393498343e-06,
      "loss": 0.82795656,
      "num_input_tokens_seen": 58406985,
      "step": 2698,
      "time_per_iteration": 3.9901914596557617
    },
    {
      "auxiliary_loss_clip": 0.01143422,
      "auxiliary_loss_mlp": 0.01044197,
      "balance_loss_clip": 1.05300224,
      "balance_loss_mlp": 1.02493238,
      "epoch": 0.16227265895084925,
      "flos": 24605991601920.0,
      "grad_norm": 1.95575769639907,
      "language_loss": 0.77431965,
      "learning_rate": 3.819256636627339e-06,
      "loss": 0.79619586,
      "num_input_tokens_seen": 58426205,
      "step": 2699,
      "time_per_iteration": 3.984302520751953
    },
    {
      "auxiliary_loss_clip": 0.01133137,
      "auxiliary_loss_mlp": 0.01040544,
      "balance_loss_clip": 1.05134153,
      "balance_loss_mlp": 1.02256763,
      "epoch": 0.1623327822035172,
      "flos": 19573326996480.0,
      "grad_norm": 2.172024885654685,
      "language_loss": 0.85671866,
      "learning_rate": 3.81909481076994e-06,
      "loss": 0.8784554,
      "num_input_tokens_seen": 58443830,
      "step": 2700,
      "time_per_iteration": 2.5627851486206055
    },
    {
      "auxiliary_loss_clip": 0.01143,
      "auxiliary_loss_mlp": 0.00950719,
      "balance_loss_clip": 1.0490042,
      "balance_loss_mlp": 1.28568542,
      "epoch": 0.16239290545618518,
      "flos": 26468462678400.0,
      "grad_norm": 1.5941704100801748,
      "language_loss": 0.80655563,
      "learning_rate": 3.818932915932284e-06,
      "loss": 0.82749283,
      "num_input_tokens_seen": 58464405,
      "step": 2701,
      "time_per_iteration": 2.5487163066864014
    },
    {
      "auxiliary_loss_clip": 0.01139772,
      "auxiliary_loss_mlp": 0.010461,
      "balance_loss_clip": 1.0538224,
      "balance_loss_mlp": 1.02662086,
      "epoch": 0.16245302870885314,
      "flos": 15851940289920.0,
      "grad_norm": 2.2729532913807087,
      "language_loss": 0.73288035,
      "learning_rate": 3.818770952120511e-06,
      "loss": 0.75473905,
      "num_input_tokens_seen": 58483295,
      "step": 2702,
      "time_per_iteration": 2.5764291286468506
    },
    {
      "auxiliary_loss_clip": 0.01147615,
      "auxiliary_loss_mlp": 0.01048893,
      "balance_loss_clip": 1.05262816,
      "balance_loss_mlp": 1.0274359,
      "epoch": 0.1625131519615211,
      "flos": 14756521173120.0,
      "grad_norm": 1.9882712654332582,
      "language_loss": 0.73034507,
      "learning_rate": 3.81860891934076e-06,
      "loss": 0.75231016,
      "num_input_tokens_seen": 58501205,
      "step": 2703,
      "time_per_iteration": 2.483491897583008
    },
    {
      "auxiliary_loss_clip": 0.01158105,
      "auxiliary_loss_mlp": 0.01049314,
      "balance_loss_clip": 1.04997206,
      "balance_loss_mlp": 1.02814245,
      "epoch": 0.1625732752141891,
      "flos": 28220508368640.0,
      "grad_norm": 2.0299743179780476,
      "language_loss": 0.70864922,
      "learning_rate": 3.818446817599176e-06,
      "loss": 0.73072338,
      "num_input_tokens_seen": 58522315,
      "step": 2704,
      "time_per_iteration": 3.8935294151306152
    },
    {
      "auxiliary_loss_clip": 0.01028706,
      "auxiliary_loss_mlp": 0.01036622,
      "balance_loss_clip": 1.03337717,
      "balance_loss_mlp": 1.03354633,
      "epoch": 0.16263339846685707,
      "flos": 67327947688320.0,
      "grad_norm": 0.7859325215468879,
      "language_loss": 0.53329688,
      "learning_rate": 3.818284646901907e-06,
      "loss": 0.55395019,
      "num_input_tokens_seen": 58586695,
      "step": 2705,
      "time_per_iteration": 3.1838834285736084
    },
    {
      "auxiliary_loss_clip": 0.01133318,
      "auxiliary_loss_mlp": 0.00917205,
      "balance_loss_clip": 1.05283189,
      "balance_loss_mlp": 1.21622825,
      "epoch": 0.16269352171952503,
      "flos": 14319165173760.0,
      "grad_norm": 6.433154380322071,
      "language_loss": 0.75327158,
      "learning_rate": 3.818122407255102e-06,
      "loss": 0.77377683,
      "num_input_tokens_seen": 58602435,
      "step": 2706,
      "time_per_iteration": 2.5242412090301514
    },
    {
      "auxiliary_loss_clip": 0.01130941,
      "auxiliary_loss_mlp": 0.0105163,
      "balance_loss_clip": 1.05226767,
      "balance_loss_mlp": 1.03231764,
      "epoch": 0.162753644972193,
      "flos": 28361205941760.0,
      "grad_norm": 2.018028329783564,
      "language_loss": 0.7244699,
      "learning_rate": 3.817960098664914e-06,
      "loss": 0.74629557,
      "num_input_tokens_seen": 58621275,
      "step": 2707,
      "time_per_iteration": 2.650559663772583
    },
    {
      "auxiliary_loss_clip": 0.01143647,
      "auxiliary_loss_mlp": 0.01052479,
      "balance_loss_clip": 1.05551732,
      "balance_loss_mlp": 1.03259492,
      "epoch": 0.16281376822486096,
      "flos": 19937856170880.0,
      "grad_norm": 12.305593870749805,
      "language_loss": 0.83561224,
      "learning_rate": 3.817797721137495e-06,
      "loss": 0.85757351,
      "num_input_tokens_seen": 58637550,
      "step": 2708,
      "time_per_iteration": 2.53810453414917
    },
    {
      "auxiliary_loss_clip": 0.01098201,
      "auxiliary_loss_mlp": 0.00896643,
      "balance_loss_clip": 1.04948211,
      "balance_loss_mlp": 1.1797936,
      "epoch": 0.16287389147752893,
      "flos": 21251719848960.0,
      "grad_norm": 2.5043668165943576,
      "language_loss": 0.86525786,
      "learning_rate": 3.817635274679006e-06,
      "loss": 0.88520634,
      "num_input_tokens_seen": 58654135,
      "step": 2709,
      "time_per_iteration": 2.617201566696167
    },
    {
      "auxiliary_loss_clip": 0.01136284,
      "auxiliary_loss_mlp": 0.0088481,
      "balance_loss_clip": 1.05112839,
      "balance_loss_mlp": 1.16346121,
      "epoch": 0.1629340147301969,
      "flos": 19244672530560.0,
      "grad_norm": 1.664927471703942,
      "language_loss": 0.91589355,
      "learning_rate": 3.817472759295605e-06,
      "loss": 0.93610454,
      "num_input_tokens_seen": 58674320,
      "step": 2710,
      "time_per_iteration": 2.5667223930358887
    },
    {
      "auxiliary_loss_clip": 0.01114927,
      "auxiliary_loss_mlp": 0.01056324,
      "balance_loss_clip": 1.05328012,
      "balance_loss_mlp": 1.03635621,
      "epoch": 0.16299413798286488,
      "flos": 21249816428160.0,
      "grad_norm": 2.5136879560345227,
      "language_loss": 0.81871688,
      "learning_rate": 3.817310174993453e-06,
      "loss": 0.84042943,
      "num_input_tokens_seen": 58691000,
      "step": 2711,
      "time_per_iteration": 2.5697169303894043
    },
    {
      "auxiliary_loss_clip": 0.01142942,
      "auxiliary_loss_mlp": 0.01044481,
      "balance_loss_clip": 1.05013776,
      "balance_loss_mlp": 1.02437019,
      "epoch": 0.16305426123553285,
      "flos": 18770579896320.0,
      "grad_norm": 2.8629389741387907,
      "language_loss": 0.80807483,
      "learning_rate": 3.817147521778719e-06,
      "loss": 0.82994908,
      "num_input_tokens_seen": 58710230,
      "step": 2712,
      "time_per_iteration": 2.5495386123657227
    },
    {
      "auxiliary_loss_clip": 0.01166705,
      "auxiliary_loss_mlp": 0.01051614,
      "balance_loss_clip": 1.05663764,
      "balance_loss_mlp": 1.03104997,
      "epoch": 0.16311438448820081,
      "flos": 22087648137600.0,
      "grad_norm": 2.2854744352140224,
      "language_loss": 0.76789558,
      "learning_rate": 3.816984799657568e-06,
      "loss": 0.79007876,
      "num_input_tokens_seen": 58728610,
      "step": 2713,
      "time_per_iteration": 2.4847543239593506
    },
    {
      "auxiliary_loss_clip": 0.01152401,
      "auxiliary_loss_mlp": 0.01060568,
      "balance_loss_clip": 1.05872989,
      "balance_loss_mlp": 1.04069555,
      "epoch": 0.16317450774086878,
      "flos": 16467700164480.0,
      "grad_norm": 2.3405320815460158,
      "language_loss": 0.7922554,
      "learning_rate": 3.8168220086361715e-06,
      "loss": 0.81438506,
      "num_input_tokens_seen": 58744385,
      "step": 2714,
      "time_per_iteration": 2.506089448928833
    },
    {
      "auxiliary_loss_clip": 0.01146366,
      "auxiliary_loss_mlp": 0.01057776,
      "balance_loss_clip": 1.05278206,
      "balance_loss_mlp": 1.03799951,
      "epoch": 0.16323463099353674,
      "flos": 24352929308160.0,
      "grad_norm": 1.6955057359460362,
      "language_loss": 0.77986825,
      "learning_rate": 3.816659148720702e-06,
      "loss": 0.80190957,
      "num_input_tokens_seen": 58763905,
      "step": 2715,
      "time_per_iteration": 2.5383594036102295
    },
    {
      "auxiliary_loss_clip": 0.01128552,
      "auxiliary_loss_mlp": 0.01045344,
      "balance_loss_clip": 1.04975367,
      "balance_loss_mlp": 1.02666366,
      "epoch": 0.1632947542462047,
      "flos": 24900782520960.0,
      "grad_norm": 2.893480651727335,
      "language_loss": 0.81707764,
      "learning_rate": 3.816496219917336e-06,
      "loss": 0.83881664,
      "num_input_tokens_seen": 58785580,
      "step": 2716,
      "time_per_iteration": 2.589437246322632
    },
    {
      "auxiliary_loss_clip": 0.0114013,
      "auxiliary_loss_mlp": 0.0105566,
      "balance_loss_clip": 1.05536127,
      "balance_loss_mlp": 1.03681254,
      "epoch": 0.1633548774988727,
      "flos": 24900279730560.0,
      "grad_norm": 1.8542855053271434,
      "language_loss": 0.85953832,
      "learning_rate": 3.816333222232251e-06,
      "loss": 0.88149625,
      "num_input_tokens_seen": 58806075,
      "step": 2717,
      "time_per_iteration": 2.5721468925476074
    },
    {
      "auxiliary_loss_clip": 0.01134098,
      "auxiliary_loss_mlp": 0.01049251,
      "balance_loss_clip": 1.05409896,
      "balance_loss_mlp": 1.03070211,
      "epoch": 0.16341500075154067,
      "flos": 30441798357120.0,
      "grad_norm": 1.8059476607240434,
      "language_loss": 0.76089865,
      "learning_rate": 3.816170155671629e-06,
      "loss": 0.78273207,
      "num_input_tokens_seen": 58827405,
      "step": 2718,
      "time_per_iteration": 2.6224029064178467
    },
    {
      "auxiliary_loss_clip": 0.01142386,
      "auxiliary_loss_mlp": 0.01045812,
      "balance_loss_clip": 1.05344653,
      "balance_loss_mlp": 1.02741766,
      "epoch": 0.16347512400420863,
      "flos": 22784530878720.0,
      "grad_norm": 2.2112671199320864,
      "language_loss": 0.73815763,
      "learning_rate": 3.816007020241652e-06,
      "loss": 0.76003963,
      "num_input_tokens_seen": 58847205,
      "step": 2719,
      "time_per_iteration": 2.549989938735962
    },
    {
      "auxiliary_loss_clip": 0.01129278,
      "auxiliary_loss_mlp": 0.010507,
      "balance_loss_clip": 1.05130601,
      "balance_loss_mlp": 1.03073192,
      "epoch": 0.1635352472568766,
      "flos": 22633274707200.0,
      "grad_norm": 1.614564790886584,
      "language_loss": 0.72302604,
      "learning_rate": 3.815843815948507e-06,
      "loss": 0.74482584,
      "num_input_tokens_seen": 58866865,
      "step": 2720,
      "time_per_iteration": 2.597231864929199
    },
    {
      "auxiliary_loss_clip": 0.01105314,
      "auxiliary_loss_mlp": 0.01051581,
      "balance_loss_clip": 1.04918301,
      "balance_loss_mlp": 1.02998054,
      "epoch": 0.16359537050954456,
      "flos": 15522998515200.0,
      "grad_norm": 2.2121347083315563,
      "language_loss": 0.75938642,
      "learning_rate": 3.8156805427983824e-06,
      "loss": 0.78095531,
      "num_input_tokens_seen": 58885200,
      "step": 2721,
      "time_per_iteration": 2.5674898624420166
    },
    {
      "auxiliary_loss_clip": 0.01109349,
      "auxiliary_loss_mlp": 0.01059766,
      "balance_loss_clip": 1.04759002,
      "balance_loss_mlp": 1.03749776,
      "epoch": 0.16365549376221253,
      "flos": 22090162089600.0,
      "grad_norm": 1.693655638988614,
      "language_loss": 0.79608893,
      "learning_rate": 3.8155172007974695e-06,
      "loss": 0.81778002,
      "num_input_tokens_seen": 58906385,
      "step": 2722,
      "time_per_iteration": 2.6328768730163574
    },
    {
      "auxiliary_loss_clip": 0.01146882,
      "auxiliary_loss_mlp": 0.0085739,
      "balance_loss_clip": 1.05432165,
      "balance_loss_mlp": 1.10834622,
      "epoch": 0.1637156170148805,
      "flos": 24060400945920.0,
      "grad_norm": 2.1104889977473524,
      "language_loss": 0.84645343,
      "learning_rate": 3.8153537899519624e-06,
      "loss": 0.86649621,
      "num_input_tokens_seen": 58925040,
      "step": 2723,
      "time_per_iteration": 2.522733688354492
    },
    {
      "auxiliary_loss_clip": 0.01105832,
      "auxiliary_loss_mlp": 0.01040816,
      "balance_loss_clip": 1.05046821,
      "balance_loss_mlp": 1.02120614,
      "epoch": 0.1637757402675485,
      "flos": 26685362954880.0,
      "grad_norm": 2.0786211436900954,
      "language_loss": 0.70985049,
      "learning_rate": 3.815190310268058e-06,
      "loss": 0.73131692,
      "num_input_tokens_seen": 58944790,
      "step": 2724,
      "time_per_iteration": 2.6469380855560303
    },
    {
      "auxiliary_loss_clip": 0.01115712,
      "auxiliary_loss_mlp": 0.01048964,
      "balance_loss_clip": 1.05354619,
      "balance_loss_mlp": 1.03049791,
      "epoch": 0.16383586352021645,
      "flos": 16106941918080.0,
      "grad_norm": 2.0033095433008135,
      "language_loss": 0.71085691,
      "learning_rate": 3.815026761751955e-06,
      "loss": 0.73250365,
      "num_input_tokens_seen": 58962500,
      "step": 2725,
      "time_per_iteration": 2.5601398944854736
    },
    {
      "auxiliary_loss_clip": 0.01111557,
      "auxiliary_loss_mlp": 0.01041789,
      "balance_loss_clip": 1.05357718,
      "balance_loss_mlp": 1.02275145,
      "epoch": 0.16389598677288442,
      "flos": 19165991788800.0,
      "grad_norm": 1.9595531103129147,
      "language_loss": 0.88314617,
      "learning_rate": 3.814863144409855e-06,
      "loss": 0.90467966,
      "num_input_tokens_seen": 58980355,
      "step": 2726,
      "time_per_iteration": 2.555626392364502
    },
    {
      "auxiliary_loss_clip": 0.01150561,
      "auxiliary_loss_mlp": 0.01045766,
      "balance_loss_clip": 1.05519652,
      "balance_loss_mlp": 1.02632308,
      "epoch": 0.16395611002555238,
      "flos": 21507008785920.0,
      "grad_norm": 2.1930004731506196,
      "language_loss": 0.73689544,
      "learning_rate": 3.814699458247963e-06,
      "loss": 0.75885874,
      "num_input_tokens_seen": 58999505,
      "step": 2727,
      "time_per_iteration": 2.515079975128174
    },
    {
      "auxiliary_loss_clip": 0.01144875,
      "auxiliary_loss_mlp": 0.0104531,
      "balance_loss_clip": 1.05317283,
      "balance_loss_mlp": 1.02817988,
      "epoch": 0.16401623327822035,
      "flos": 21470918595840.0,
      "grad_norm": 1.5949431818501372,
      "language_loss": 0.82782567,
      "learning_rate": 3.8145357032724855e-06,
      "loss": 0.84972751,
      "num_input_tokens_seen": 59017930,
      "step": 2728,
      "time_per_iteration": 2.524932622909546
    },
    {
      "auxiliary_loss_clip": 0.01151869,
      "auxiliary_loss_mlp": 0.01047286,
      "balance_loss_clip": 1.05423307,
      "balance_loss_mlp": 1.02737808,
      "epoch": 0.1640763565308883,
      "flos": 13626232928640.0,
      "grad_norm": 2.48458424643167,
      "language_loss": 0.8504076,
      "learning_rate": 3.814371879489633e-06,
      "loss": 0.87239915,
      "num_input_tokens_seen": 59035130,
      "step": 2729,
      "time_per_iteration": 2.4818005561828613
    },
    {
      "auxiliary_loss_clip": 0.01159237,
      "auxiliary_loss_mlp": 0.01047934,
      "balance_loss_clip": 1.05298805,
      "balance_loss_mlp": 1.02976656,
      "epoch": 0.16413647978355628,
      "flos": 15451464579840.0,
      "grad_norm": 2.200805089311736,
      "language_loss": 0.73442566,
      "learning_rate": 3.814207986905616e-06,
      "loss": 0.75649738,
      "num_input_tokens_seen": 59053080,
      "step": 2730,
      "time_per_iteration": 2.4542348384857178
    },
    {
      "auxiliary_loss_clip": 0.01136704,
      "auxiliary_loss_mlp": 0.01051464,
      "balance_loss_clip": 1.04773128,
      "balance_loss_mlp": 1.03099561,
      "epoch": 0.16419660303622427,
      "flos": 45878682015360.0,
      "grad_norm": 2.401223118860667,
      "language_loss": 0.74759614,
      "learning_rate": 3.814044025526651e-06,
      "loss": 0.76947778,
      "num_input_tokens_seen": 59075610,
      "step": 2731,
      "time_per_iteration": 2.719670057296753
    },
    {
      "auxiliary_loss_clip": 0.01124269,
      "auxiliary_loss_mlp": 0.01053074,
      "balance_loss_clip": 1.05562377,
      "balance_loss_mlp": 1.03280807,
      "epoch": 0.16425672628889224,
      "flos": 18952826526720.0,
      "grad_norm": 2.0724753605297694,
      "language_loss": 0.79272354,
      "learning_rate": 3.8138799953589548e-06,
      "loss": 0.81449693,
      "num_input_tokens_seen": 59094555,
      "step": 2732,
      "time_per_iteration": 2.564490795135498
    },
    {
      "auxiliary_loss_clip": 0.01137078,
      "auxiliary_loss_mlp": 0.01047394,
      "balance_loss_clip": 1.05202723,
      "balance_loss_mlp": 1.02822483,
      "epoch": 0.1643168495415602,
      "flos": 24312996362880.0,
      "grad_norm": 1.8733094710945413,
      "language_loss": 0.69387954,
      "learning_rate": 3.8137158964087473e-06,
      "loss": 0.71572423,
      "num_input_tokens_seen": 59113515,
      "step": 2733,
      "time_per_iteration": 2.5508759021759033
    },
    {
      "auxiliary_loss_clip": 0.01139445,
      "auxiliary_loss_mlp": 0.01050231,
      "balance_loss_clip": 1.05349338,
      "balance_loss_mlp": 1.02829647,
      "epoch": 0.16437697279422817,
      "flos": 26428421992320.0,
      "grad_norm": 1.678468618487968,
      "language_loss": 0.80700803,
      "learning_rate": 3.8135517286822508e-06,
      "loss": 0.82890475,
      "num_input_tokens_seen": 59133275,
      "step": 2734,
      "time_per_iteration": 2.5930395126342773
    },
    {
      "auxiliary_loss_clip": 0.01130431,
      "auxiliary_loss_mlp": 0.01061516,
      "balance_loss_clip": 1.05277228,
      "balance_loss_mlp": 1.04003453,
      "epoch": 0.16443709604689613,
      "flos": 34532239351680.0,
      "grad_norm": 2.5719655992304484,
      "language_loss": 0.82100368,
      "learning_rate": 3.8133874921856914e-06,
      "loss": 0.84292316,
      "num_input_tokens_seen": 59154095,
      "step": 2735,
      "time_per_iteration": 2.640289068222046
    },
    {
      "auxiliary_loss_clip": 0.01075643,
      "auxiliary_loss_mlp": 0.01044192,
      "balance_loss_clip": 1.05443358,
      "balance_loss_mlp": 1.02566671,
      "epoch": 0.1644972192995641,
      "flos": 23258048895360.0,
      "grad_norm": 2.7720807421999303,
      "language_loss": 0.7813046,
      "learning_rate": 3.813223186925296e-06,
      "loss": 0.80250287,
      "num_input_tokens_seen": 59173795,
      "step": 2736,
      "time_per_iteration": 5.4762349128723145
    },
    {
      "auxiliary_loss_clip": 0.01141235,
      "auxiliary_loss_mlp": 0.01057864,
      "balance_loss_clip": 1.05281842,
      "balance_loss_mlp": 1.03896964,
      "epoch": 0.1645573425522321,
      "flos": 26979543342720.0,
      "grad_norm": 1.9880645836291604,
      "language_loss": 0.81722987,
      "learning_rate": 3.8130588129072964e-06,
      "loss": 0.83922088,
      "num_input_tokens_seen": 59191610,
      "step": 2737,
      "time_per_iteration": 2.5748472213745117
    },
    {
      "auxiliary_loss_clip": 0.0114202,
      "auxiliary_loss_mlp": 0.01061765,
      "balance_loss_clip": 1.04953671,
      "balance_loss_mlp": 1.04118967,
      "epoch": 0.16461746580490005,
      "flos": 28731768600960.0,
      "grad_norm": 1.703648875689689,
      "language_loss": 0.87450475,
      "learning_rate": 3.8128943701379246e-06,
      "loss": 0.89654267,
      "num_input_tokens_seen": 59213000,
      "step": 2738,
      "time_per_iteration": 3.930824041366577
    },
    {
      "auxiliary_loss_clip": 0.01133666,
      "auxiliary_loss_mlp": 0.01055938,
      "balance_loss_clip": 1.05129695,
      "balance_loss_mlp": 1.03600669,
      "epoch": 0.16467758905756802,
      "flos": 24930156867840.0,
      "grad_norm": 1.8475232512129576,
      "language_loss": 0.72380507,
      "learning_rate": 3.8127298586234167e-06,
      "loss": 0.74570119,
      "num_input_tokens_seen": 59232340,
      "step": 2739,
      "time_per_iteration": 2.562509298324585
    },
    {
      "auxiliary_loss_clip": 0.01146177,
      "auxiliary_loss_mlp": 0.01049913,
      "balance_loss_clip": 1.05047858,
      "balance_loss_mlp": 1.03033888,
      "epoch": 0.16473771231023598,
      "flos": 24826519152000.0,
      "grad_norm": 2.163373955545855,
      "language_loss": 0.82031256,
      "learning_rate": 3.8125652783700104e-06,
      "loss": 0.84227347,
      "num_input_tokens_seen": 59253950,
      "step": 2740,
      "time_per_iteration": 2.6030588150024414
    },
    {
      "auxiliary_loss_clip": 0.01108025,
      "auxiliary_loss_mlp": 0.01071131,
      "balance_loss_clip": 1.04780722,
      "balance_loss_mlp": 1.04565573,
      "epoch": 0.16479783556290395,
      "flos": 39896072375040.0,
      "grad_norm": 1.954869785782303,
      "language_loss": 0.68999875,
      "learning_rate": 3.8124006293839475e-06,
      "loss": 0.71179032,
      "num_input_tokens_seen": 59275545,
      "step": 2741,
      "time_per_iteration": 2.7714107036590576
    },
    {
      "auxiliary_loss_clip": 0.01160855,
      "auxiliary_loss_mlp": 0.01046396,
      "balance_loss_clip": 1.05374622,
      "balance_loss_mlp": 1.02709639,
      "epoch": 0.16485795881557191,
      "flos": 19897061299200.0,
      "grad_norm": 1.925985116242982,
      "language_loss": 0.79738945,
      "learning_rate": 3.812235911671472e-06,
      "loss": 0.81946194,
      "num_input_tokens_seen": 59293480,
      "step": 2742,
      "time_per_iteration": 3.853848934173584
    },
    {
      "auxiliary_loss_clip": 0.01135781,
      "auxiliary_loss_mlp": 0.01055787,
      "balance_loss_clip": 1.05246782,
      "balance_loss_mlp": 1.03581953,
      "epoch": 0.16491808206823988,
      "flos": 20556129997440.0,
      "grad_norm": 3.47812481950133,
      "language_loss": 0.84971619,
      "learning_rate": 3.8120711252388274e-06,
      "loss": 0.87163186,
      "num_input_tokens_seen": 59313435,
      "step": 2743,
      "time_per_iteration": 2.5474965572357178
    },
    {
      "auxiliary_loss_clip": 0.01157272,
      "auxiliary_loss_mlp": 0.01054161,
      "balance_loss_clip": 1.05260515,
      "balance_loss_mlp": 1.03405011,
      "epoch": 0.16497820532090787,
      "flos": 23800802376960.0,
      "grad_norm": 1.6094137354799132,
      "language_loss": 0.85668492,
      "learning_rate": 3.811906270092265e-06,
      "loss": 0.8787992,
      "num_input_tokens_seen": 59331535,
      "step": 2744,
      "time_per_iteration": 2.518165111541748
    },
    {
      "auxiliary_loss_clip": 0.011294,
      "auxiliary_loss_mlp": 0.01054432,
      "balance_loss_clip": 1.05277145,
      "balance_loss_mlp": 1.03540587,
      "epoch": 0.16503832857357584,
      "flos": 25482642935040.0,
      "grad_norm": 1.7944534838794501,
      "language_loss": 0.82840991,
      "learning_rate": 3.811741346238036e-06,
      "loss": 0.85024822,
      "num_input_tokens_seen": 59350680,
      "step": 2745,
      "time_per_iteration": 2.5560920238494873
    },
    {
      "auxiliary_loss_clip": 0.01130745,
      "auxiliary_loss_mlp": 0.01057626,
      "balance_loss_clip": 1.05787516,
      "balance_loss_mlp": 1.03807592,
      "epoch": 0.1650984518262438,
      "flos": 17676058619520.0,
      "grad_norm": 1.9164788750900865,
      "language_loss": 0.76553077,
      "learning_rate": 3.8115763536823923e-06,
      "loss": 0.78741455,
      "num_input_tokens_seen": 59367020,
      "step": 2746,
      "time_per_iteration": 2.560053825378418
    },
    {
      "auxiliary_loss_clip": 0.01159431,
      "auxiliary_loss_mlp": 0.01055617,
      "balance_loss_clip": 1.05262733,
      "balance_loss_mlp": 1.03542352,
      "epoch": 0.16515857507891177,
      "flos": 18698327688960.0,
      "grad_norm": 1.587969598327242,
      "language_loss": 0.80787677,
      "learning_rate": 3.811411292431592e-06,
      "loss": 0.83002722,
      "num_input_tokens_seen": 59386075,
      "step": 2747,
      "time_per_iteration": 2.4530060291290283
    },
    {
      "auxiliary_loss_clip": 0.01151838,
      "auxiliary_loss_mlp": 0.01048157,
      "balance_loss_clip": 1.05556643,
      "balance_loss_mlp": 1.02791572,
      "epoch": 0.16521869833157973,
      "flos": 15010481306880.0,
      "grad_norm": 2.107758699088543,
      "language_loss": 0.70029032,
      "learning_rate": 3.8112461624918945e-06,
      "loss": 0.72229028,
      "num_input_tokens_seen": 59402690,
      "step": 2748,
      "time_per_iteration": 2.507517099380493
    },
    {
      "auxiliary_loss_clip": 0.01167088,
      "auxiliary_loss_mlp": 0.00842586,
      "balance_loss_clip": 1.0583694,
      "balance_loss_mlp": 1.08000112,
      "epoch": 0.1652788215842477,
      "flos": 22121152548480.0,
      "grad_norm": 2.303862665908159,
      "language_loss": 0.88183355,
      "learning_rate": 3.811080963869561e-06,
      "loss": 0.90193033,
      "num_input_tokens_seen": 59421130,
      "step": 2749,
      "time_per_iteration": 2.5047249794006348
    },
    {
      "auxiliary_loss_clip": 0.01147185,
      "auxiliary_loss_mlp": 0.01049129,
      "balance_loss_clip": 1.05149901,
      "balance_loss_mlp": 1.02889872,
      "epoch": 0.16533894483691566,
      "flos": 18333080242560.0,
      "grad_norm": 1.9479541715285893,
      "language_loss": 0.78941536,
      "learning_rate": 3.8109156965708557e-06,
      "loss": 0.81137848,
      "num_input_tokens_seen": 59438970,
      "step": 2750,
      "time_per_iteration": 2.488950252532959
    },
    {
      "auxiliary_loss_clip": 0.01150154,
      "auxiliary_loss_mlp": 0.01049764,
      "balance_loss_clip": 1.05504632,
      "balance_loss_mlp": 1.03015387,
      "epoch": 0.16539906808958366,
      "flos": 22382115834240.0,
      "grad_norm": 1.7251288368220443,
      "language_loss": 0.94874507,
      "learning_rate": 3.8107503606020455e-06,
      "loss": 0.97074425,
      "num_input_tokens_seen": 59458510,
      "step": 2751,
      "time_per_iteration": 2.5006072521209717
    },
    {
      "auxiliary_loss_clip": 0.01069651,
      "auxiliary_loss_mlp": 0.01056015,
      "balance_loss_clip": 1.05213952,
      "balance_loss_mlp": 1.03590417,
      "epoch": 0.16545919134225162,
      "flos": 22711093522560.0,
      "grad_norm": 1.9117433547643519,
      "language_loss": 0.70594549,
      "learning_rate": 3.8105849559693997e-06,
      "loss": 0.72720212,
      "num_input_tokens_seen": 59477110,
      "step": 2752,
      "time_per_iteration": 2.661979913711548
    },
    {
      "auxiliary_loss_clip": 0.01066976,
      "auxiliary_loss_mlp": 0.01016386,
      "balance_loss_clip": 1.03223252,
      "balance_loss_mlp": 1.01290512,
      "epoch": 0.1655193145949196,
      "flos": 67802974076160.0,
      "grad_norm": 0.7815292572606064,
      "language_loss": 0.54066873,
      "learning_rate": 3.810419482679192e-06,
      "loss": 0.56150234,
      "num_input_tokens_seen": 59541155,
      "step": 2753,
      "time_per_iteration": 3.227463722229004
    },
    {
      "auxiliary_loss_clip": 0.01160053,
      "auxiliary_loss_mlp": 0.00830464,
      "balance_loss_clip": 1.05320168,
      "balance_loss_mlp": 1.05913115,
      "epoch": 0.16557943784758755,
      "flos": 24280389792000.0,
      "grad_norm": 1.7259836614352193,
      "language_loss": 0.75803649,
      "learning_rate": 3.8102539407376954e-06,
      "loss": 0.7779417,
      "num_input_tokens_seen": 59561155,
      "step": 2754,
      "time_per_iteration": 2.5059049129486084
    },
    {
      "auxiliary_loss_clip": 0.01140449,
      "auxiliary_loss_mlp": 0.01066673,
      "balance_loss_clip": 1.05783117,
      "balance_loss_mlp": 1.04321241,
      "epoch": 0.16563956110025552,
      "flos": 20083617561600.0,
      "grad_norm": 2.3473405878336604,
      "language_loss": 0.86670291,
      "learning_rate": 3.810088330151188e-06,
      "loss": 0.8887741,
      "num_input_tokens_seen": 59580460,
      "step": 2755,
      "time_per_iteration": 2.5629043579101562
    },
    {
      "auxiliary_loss_clip": 0.01119381,
      "auxiliary_loss_mlp": 0.01060666,
      "balance_loss_clip": 1.04720521,
      "balance_loss_mlp": 1.04103231,
      "epoch": 0.16569968435292348,
      "flos": 28034454896640.0,
      "grad_norm": 1.770263404644781,
      "language_loss": 0.73287815,
      "learning_rate": 3.80992265092595e-06,
      "loss": 0.75467861,
      "num_input_tokens_seen": 59600025,
      "step": 2756,
      "time_per_iteration": 2.604966878890991
    },
    {
      "auxiliary_loss_clip": 0.01125309,
      "auxiliary_loss_mlp": 0.01056156,
      "balance_loss_clip": 1.05057299,
      "balance_loss_mlp": 1.03591418,
      "epoch": 0.16575980760559147,
      "flos": 26250233598720.0,
      "grad_norm": 1.5955189974992747,
      "language_loss": 0.74747181,
      "learning_rate": 3.8097569030682636e-06,
      "loss": 0.76928639,
      "num_input_tokens_seen": 59620600,
      "step": 2757,
      "time_per_iteration": 2.637022018432617
    },
    {
      "auxiliary_loss_clip": 0.01140092,
      "auxiliary_loss_mlp": 0.01051765,
      "balance_loss_clip": 1.05561233,
      "balance_loss_mlp": 1.03273916,
      "epoch": 0.16581993085825944,
      "flos": 26943955943040.0,
      "grad_norm": 1.558935458985494,
      "language_loss": 0.84743607,
      "learning_rate": 3.8095910865844137e-06,
      "loss": 0.86935461,
      "num_input_tokens_seen": 59641385,
      "step": 2758,
      "time_per_iteration": 2.5889735221862793
    },
    {
      "auxiliary_loss_clip": 0.011641,
      "auxiliary_loss_mlp": 0.01064298,
      "balance_loss_clip": 1.05810249,
      "balance_loss_mlp": 1.04582095,
      "epoch": 0.1658800541109274,
      "flos": 21653632103040.0,
      "grad_norm": 2.221960236237499,
      "language_loss": 0.79371798,
      "learning_rate": 3.809425201480689e-06,
      "loss": 0.81600201,
      "num_input_tokens_seen": 59659865,
      "step": 2759,
      "time_per_iteration": 2.4944968223571777
    },
    {
      "auxiliary_loss_clip": 0.01102583,
      "auxiliary_loss_mlp": 0.01059153,
      "balance_loss_clip": 1.05547428,
      "balance_loss_mlp": 1.03892279,
      "epoch": 0.16594017736359537,
      "flos": 16435488643200.0,
      "grad_norm": 2.4007074875532606,
      "language_loss": 0.75360668,
      "learning_rate": 3.8092592477633793e-06,
      "loss": 0.77522403,
      "num_input_tokens_seen": 59678780,
      "step": 2760,
      "time_per_iteration": 2.614659547805786
    },
    {
      "auxiliary_loss_clip": 0.01116559,
      "auxiliary_loss_mlp": 0.01045098,
      "balance_loss_clip": 1.05282128,
      "balance_loss_mlp": 1.02582169,
      "epoch": 0.16600030061626334,
      "flos": 22637297030400.0,
      "grad_norm": 2.052881874590094,
      "language_loss": 0.73143721,
      "learning_rate": 3.8090932254387774e-06,
      "loss": 0.75305378,
      "num_input_tokens_seen": 59698795,
      "step": 2761,
      "time_per_iteration": 2.6424593925476074
    },
    {
      "auxiliary_loss_clip": 0.01133773,
      "auxiliary_loss_mlp": 0.01052576,
      "balance_loss_clip": 1.05281854,
      "balance_loss_mlp": 1.03325224,
      "epoch": 0.1660604238689313,
      "flos": 26396569607040.0,
      "grad_norm": 1.9152253881458319,
      "language_loss": 0.88713109,
      "learning_rate": 3.8089271345131788e-06,
      "loss": 0.90899462,
      "num_input_tokens_seen": 59718795,
      "step": 2762,
      "time_per_iteration": 2.56805682182312
    },
    {
      "auxiliary_loss_clip": 0.0111623,
      "auxiliary_loss_mlp": 0.01053285,
      "balance_loss_clip": 1.05409181,
      "balance_loss_mlp": 1.03363895,
      "epoch": 0.16612054712159927,
      "flos": 23039999383680.0,
      "grad_norm": 1.6552529489392105,
      "language_loss": 0.88188791,
      "learning_rate": 3.8087609749928822e-06,
      "loss": 0.90358317,
      "num_input_tokens_seen": 59737555,
      "step": 2763,
      "time_per_iteration": 2.6354095935821533
    },
    {
      "auxiliary_loss_clip": 0.01068965,
      "auxiliary_loss_mlp": 0.01010918,
      "balance_loss_clip": 1.02821684,
      "balance_loss_mlp": 1.00746059,
      "epoch": 0.16618067037426726,
      "flos": 59241225202560.0,
      "grad_norm": 0.7833965595945858,
      "language_loss": 0.59812993,
      "learning_rate": 3.8085947468841885e-06,
      "loss": 0.61892879,
      "num_input_tokens_seen": 59800915,
      "step": 2764,
      "time_per_iteration": 3.097243070602417
    },
    {
      "auxiliary_loss_clip": 0.01153564,
      "auxiliary_loss_mlp": 0.01055013,
      "balance_loss_clip": 1.05557036,
      "balance_loss_mlp": 1.03322196,
      "epoch": 0.16624079362693522,
      "flos": 27198813916800.0,
      "grad_norm": 2.5324096391724056,
      "language_loss": 0.82291096,
      "learning_rate": 3.808428450193401e-06,
      "loss": 0.84499675,
      "num_input_tokens_seen": 59822910,
      "step": 2765,
      "time_per_iteration": 2.5553781986236572
    },
    {
      "auxiliary_loss_clip": 0.01170687,
      "auxiliary_loss_mlp": 0.0104995,
      "balance_loss_clip": 1.05783081,
      "balance_loss_mlp": 1.02845681,
      "epoch": 0.1663009168796032,
      "flos": 10925068216320.0,
      "grad_norm": 6.920890050147014,
      "language_loss": 0.70718098,
      "learning_rate": 3.8082620849268244e-06,
      "loss": 0.72938728,
      "num_input_tokens_seen": 59838805,
      "step": 2766,
      "time_per_iteration": 2.4572510719299316
    },
    {
      "auxiliary_loss_clip": 0.01152187,
      "auxiliary_loss_mlp": 0.01044869,
      "balance_loss_clip": 1.05945277,
      "balance_loss_mlp": 1.02618921,
      "epoch": 0.16636104013227115,
      "flos": 17894431353600.0,
      "grad_norm": 2.302468328561425,
      "language_loss": 0.88861644,
      "learning_rate": 3.808095651090769e-06,
      "loss": 0.91058707,
      "num_input_tokens_seen": 59855345,
      "step": 2767,
      "time_per_iteration": 2.484997034072876
    },
    {
      "auxiliary_loss_clip": 0.01055447,
      "auxiliary_loss_mlp": 0.01000249,
      "balance_loss_clip": 1.02626586,
      "balance_loss_mlp": 0.99674428,
      "epoch": 0.16642116338493912,
      "flos": 66726050463360.0,
      "grad_norm": 0.6430342034551394,
      "language_loss": 0.52914703,
      "learning_rate": 3.8079291486915447e-06,
      "loss": 0.54970396,
      "num_input_tokens_seen": 59917710,
      "step": 2768,
      "time_per_iteration": 3.228600263595581
    },
    {
      "auxiliary_loss_clip": 0.01138346,
      "auxiliary_loss_mlp": 0.01049124,
      "balance_loss_clip": 1.05658722,
      "balance_loss_mlp": 1.0286324,
      "epoch": 0.16648128663760708,
      "flos": 19026048401280.0,
      "grad_norm": 4.971555357494953,
      "language_loss": 0.84875786,
      "learning_rate": 3.8077625777354667e-06,
      "loss": 0.87063265,
      "num_input_tokens_seen": 59935105,
      "step": 2769,
      "time_per_iteration": 2.5082907676696777
    },
    {
      "auxiliary_loss_clip": 0.01048094,
      "auxiliary_loss_mlp": 0.01007213,
      "balance_loss_clip": 1.02633929,
      "balance_loss_mlp": 1.00389886,
      "epoch": 0.16654140989027508,
      "flos": 70134976759680.0,
      "grad_norm": 0.808535131728097,
      "language_loss": 0.57394886,
      "learning_rate": 3.80759593822885e-06,
      "loss": 0.59450191,
      "num_input_tokens_seen": 59984085,
      "step": 2770,
      "time_per_iteration": 2.990462303161621
    },
    {
      "auxiliary_loss_clip": 0.01044578,
      "auxiliary_loss_mlp": 0.01016288,
      "balance_loss_clip": 1.03779888,
      "balance_loss_mlp": 1.0131644,
      "epoch": 0.16660153314294304,
      "flos": 70272406195200.0,
      "grad_norm": 0.8658153599191056,
      "language_loss": 0.56229889,
      "learning_rate": 3.807429230178015e-06,
      "loss": 0.58290756,
      "num_input_tokens_seen": 60043470,
      "step": 2771,
      "time_per_iteration": 3.000206708908081
    },
    {
      "auxiliary_loss_clip": 0.01113926,
      "auxiliary_loss_mlp": 0.01057924,
      "balance_loss_clip": 1.05184734,
      "balance_loss_mlp": 1.03655005,
      "epoch": 0.166661656395611,
      "flos": 23075048079360.0,
      "grad_norm": 2.31771319077267,
      "language_loss": 0.70174706,
      "learning_rate": 3.8072624535892817e-06,
      "loss": 0.72346556,
      "num_input_tokens_seen": 60063045,
      "step": 2772,
      "time_per_iteration": 2.6458218097686768
    },
    {
      "auxiliary_loss_clip": 0.01146642,
      "auxiliary_loss_mlp": 0.01051476,
      "balance_loss_clip": 1.05087829,
      "balance_loss_mlp": 1.03088856,
      "epoch": 0.16672177964827897,
      "flos": 28366341586560.0,
      "grad_norm": 3.553758573174903,
      "language_loss": 0.86218131,
      "learning_rate": 3.807095608468975e-06,
      "loss": 0.88416255,
      "num_input_tokens_seen": 60081945,
      "step": 2773,
      "time_per_iteration": 2.556415557861328
    },
    {
      "auxiliary_loss_clip": 0.0109972,
      "auxiliary_loss_mlp": 0.01048147,
      "balance_loss_clip": 1.0508914,
      "balance_loss_mlp": 1.02863228,
      "epoch": 0.16678190290094694,
      "flos": 19091010147840.0,
      "grad_norm": 2.3103319451787088,
      "language_loss": 0.82341361,
      "learning_rate": 3.8069286948234224e-06,
      "loss": 0.84489226,
      "num_input_tokens_seen": 60096820,
      "step": 2774,
      "time_per_iteration": 5.434317111968994
    },
    {
      "auxiliary_loss_clip": 0.01121414,
      "auxiliary_loss_mlp": 0.0104828,
      "balance_loss_clip": 1.05211473,
      "balance_loss_mlp": 1.02733469,
      "epoch": 0.1668420261536149,
      "flos": 21799106184960.0,
      "grad_norm": 2.259361592647879,
      "language_loss": 0.83034939,
      "learning_rate": 3.806761712658952e-06,
      "loss": 0.85204631,
      "num_input_tokens_seen": 60116140,
      "step": 2775,
      "time_per_iteration": 2.5777127742767334
    },
    {
      "auxiliary_loss_clip": 0.01146371,
      "auxiliary_loss_mlp": 0.01053117,
      "balance_loss_clip": 1.05434072,
      "balance_loss_mlp": 1.0349257,
      "epoch": 0.16690214940628287,
      "flos": 19062533640960.0,
      "grad_norm": 1.9432640804092056,
      "language_loss": 0.80774808,
      "learning_rate": 3.806594661981897e-06,
      "loss": 0.82974303,
      "num_input_tokens_seen": 60134235,
      "step": 2776,
      "time_per_iteration": 2.4948902130126953
    },
    {
      "auxiliary_loss_clip": 0.01137109,
      "auxiliary_loss_mlp": 0.01055926,
      "balance_loss_clip": 1.05468833,
      "balance_loss_mlp": 1.03642368,
      "epoch": 0.16696227265895086,
      "flos": 18588548747520.0,
      "grad_norm": 2.0741148090418626,
      "language_loss": 0.79971433,
      "learning_rate": 3.8064275427985906e-06,
      "loss": 0.8216446,
      "num_input_tokens_seen": 60153275,
      "step": 2777,
      "time_per_iteration": 3.995971441268921
    },
    {
      "auxiliary_loss_clip": 0.01147476,
      "auxiliary_loss_mlp": 0.01051888,
      "balance_loss_clip": 1.05179811,
      "balance_loss_mlp": 1.03181338,
      "epoch": 0.16702239591161883,
      "flos": 23294139085440.0,
      "grad_norm": 1.7653427400063066,
      "language_loss": 0.85309923,
      "learning_rate": 3.806260355115371e-06,
      "loss": 0.87509286,
      "num_input_tokens_seen": 60173215,
      "step": 2778,
      "time_per_iteration": 2.545367956161499
    },
    {
      "auxiliary_loss_clip": 0.01134148,
      "auxiliary_loss_mlp": 0.01044373,
      "balance_loss_clip": 1.05365181,
      "balance_loss_mlp": 1.02503705,
      "epoch": 0.1670825191642868,
      "flos": 24425648392320.0,
      "grad_norm": 1.9797931050808406,
      "language_loss": 0.74465191,
      "learning_rate": 3.8060930989385778e-06,
      "loss": 0.76643711,
      "num_input_tokens_seen": 60190515,
      "step": 2779,
      "time_per_iteration": 2.5579113960266113
    },
    {
      "auxiliary_loss_clip": 0.01117804,
      "auxiliary_loss_mlp": 0.00818864,
      "balance_loss_clip": 1.05177867,
      "balance_loss_mlp": 1.03772378,
      "epoch": 0.16714264241695476,
      "flos": 26797512193920.0,
      "grad_norm": 2.102113771708161,
      "language_loss": 0.65222681,
      "learning_rate": 3.805925774274554e-06,
      "loss": 0.67159349,
      "num_input_tokens_seen": 60211655,
      "step": 2780,
      "time_per_iteration": 2.6560912132263184
    },
    {
      "auxiliary_loss_clip": 0.01124293,
      "auxiliary_loss_mlp": 0.01047569,
      "balance_loss_clip": 1.05295038,
      "balance_loss_mlp": 1.02662373,
      "epoch": 0.16720276566962272,
      "flos": 21835304115840.0,
      "grad_norm": 2.0755797024539584,
      "language_loss": 0.78413785,
      "learning_rate": 3.805758381129643e-06,
      "loss": 0.80585647,
      "num_input_tokens_seen": 60230860,
      "step": 2781,
      "time_per_iteration": 3.9903244972229004
    },
    {
      "auxiliary_loss_clip": 0.01097083,
      "auxiliary_loss_mlp": 0.01052436,
      "balance_loss_clip": 1.04681349,
      "balance_loss_mlp": 1.03207493,
      "epoch": 0.1672628889222907,
      "flos": 21470415805440.0,
      "grad_norm": 1.857862100382753,
      "language_loss": 0.75287533,
      "learning_rate": 3.805590919510193e-06,
      "loss": 0.77437055,
      "num_input_tokens_seen": 60250535,
      "step": 2782,
      "time_per_iteration": 2.623544931411743
    },
    {
      "auxiliary_loss_clip": 0.01134296,
      "auxiliary_loss_mlp": 0.01052835,
      "balance_loss_clip": 1.05419338,
      "balance_loss_mlp": 1.03165174,
      "epoch": 0.16732301217495865,
      "flos": 30774008269440.0,
      "grad_norm": 3.042160868309584,
      "language_loss": 0.6727193,
      "learning_rate": 3.8054233894225547e-06,
      "loss": 0.69459057,
      "num_input_tokens_seen": 60269530,
      "step": 2783,
      "time_per_iteration": 2.6395814418792725
    },
    {
      "auxiliary_loss_clip": 0.01158213,
      "auxiliary_loss_mlp": 0.01054921,
      "balance_loss_clip": 1.05325818,
      "balance_loss_mlp": 1.03605044,
      "epoch": 0.16738313542762664,
      "flos": 23474625949440.0,
      "grad_norm": 1.6621122398479173,
      "language_loss": 0.70134687,
      "learning_rate": 3.805255790873081e-06,
      "loss": 0.72347814,
      "num_input_tokens_seen": 60289900,
      "step": 2784,
      "time_per_iteration": 2.4844400882720947
    },
    {
      "auxiliary_loss_clip": 0.01137966,
      "auxiliary_loss_mlp": 0.0106009,
      "balance_loss_clip": 1.05268896,
      "balance_loss_mlp": 1.03878713,
      "epoch": 0.1674432586802946,
      "flos": 29789086366080.0,
      "grad_norm": 1.9849786131432834,
      "language_loss": 0.60861927,
      "learning_rate": 3.805088123868126e-06,
      "loss": 0.63059986,
      "num_input_tokens_seen": 60310025,
      "step": 2785,
      "time_per_iteration": 2.606717348098755
    },
    {
      "auxiliary_loss_clip": 0.01048988,
      "auxiliary_loss_mlp": 0.01042785,
      "balance_loss_clip": 1.02979243,
      "balance_loss_mlp": 1.03944707,
      "epoch": 0.16750338193296258,
      "flos": 66136073575680.0,
      "grad_norm": 0.7999354336654856,
      "language_loss": 0.58763623,
      "learning_rate": 3.8049203884140492e-06,
      "loss": 0.60855401,
      "num_input_tokens_seen": 60377800,
      "step": 2786,
      "time_per_iteration": 3.1775310039520264
    },
    {
      "auxiliary_loss_clip": 0.01143264,
      "auxiliary_loss_mlp": 0.01050149,
      "balance_loss_clip": 1.05446756,
      "balance_loss_mlp": 1.03014541,
      "epoch": 0.16756350518563054,
      "flos": 25696777864320.0,
      "grad_norm": 2.4271606608275627,
      "language_loss": 0.75829065,
      "learning_rate": 3.80475258451721e-06,
      "loss": 0.7802248,
      "num_input_tokens_seen": 60398215,
      "step": 2787,
      "time_per_iteration": 2.56078839302063
    },
    {
      "auxiliary_loss_clip": 0.01149544,
      "auxiliary_loss_mlp": 0.01043303,
      "balance_loss_clip": 1.05504918,
      "balance_loss_mlp": 1.02442026,
      "epoch": 0.1676236284382985,
      "flos": 23836102467840.0,
      "grad_norm": 1.7878398542518648,
      "language_loss": 0.77468491,
      "learning_rate": 3.804584712183972e-06,
      "loss": 0.79661345,
      "num_input_tokens_seen": 60416910,
      "step": 2788,
      "time_per_iteration": 2.556183338165283
    },
    {
      "auxiliary_loss_clip": 0.01047736,
      "auxiliary_loss_mlp": 0.01012217,
      "balance_loss_clip": 1.02857876,
      "balance_loss_mlp": 1.00909352,
      "epoch": 0.16768375169096647,
      "flos": 59874902985600.0,
      "grad_norm": 0.8608655528684915,
      "language_loss": 0.59292656,
      "learning_rate": 3.8044167714207013e-06,
      "loss": 0.61352611,
      "num_input_tokens_seen": 60468660,
      "step": 2789,
      "time_per_iteration": 2.9948930740356445
    },
    {
      "auxiliary_loss_clip": 0.0114994,
      "auxiliary_loss_mlp": 0.01055117,
      "balance_loss_clip": 1.05471325,
      "balance_loss_mlp": 1.03555465,
      "epoch": 0.16774387494363446,
      "flos": 38435657207040.0,
      "grad_norm": 1.4226874175524844,
      "language_loss": 0.7017808,
      "learning_rate": 3.804248762233765e-06,
      "loss": 0.72383142,
      "num_input_tokens_seen": 60492370,
      "step": 2790,
      "time_per_iteration": 2.683670997619629
    },
    {
      "auxiliary_loss_clip": 0.01129467,
      "auxiliary_loss_mlp": 0.01053769,
      "balance_loss_clip": 1.05511653,
      "balance_loss_mlp": 1.03480268,
      "epoch": 0.16780399819630243,
      "flos": 22637620252800.0,
      "grad_norm": 1.8049917860175317,
      "language_loss": 0.79625928,
      "learning_rate": 3.8040806846295356e-06,
      "loss": 0.81809163,
      "num_input_tokens_seen": 60512655,
      "step": 2791,
      "time_per_iteration": 2.5963921546936035
    },
    {
      "auxiliary_loss_clip": 0.01127295,
      "auxiliary_loss_mlp": 0.01048254,
      "balance_loss_clip": 1.05591583,
      "balance_loss_mlp": 1.02754712,
      "epoch": 0.1678641214489704,
      "flos": 32891516887680.0,
      "grad_norm": 1.878703192623841,
      "language_loss": 0.70956194,
      "learning_rate": 3.8039125386143853e-06,
      "loss": 0.73131746,
      "num_input_tokens_seen": 60533090,
      "step": 2792,
      "time_per_iteration": 2.639566659927368
    },
    {
      "auxiliary_loss_clip": 0.01133869,
      "auxiliary_loss_mlp": 0.01045574,
      "balance_loss_clip": 1.05612993,
      "balance_loss_mlp": 1.02598786,
      "epoch": 0.16792424470163836,
      "flos": 19974916028160.0,
      "grad_norm": 2.0135349324887746,
      "language_loss": 0.7139616,
      "learning_rate": 3.803744324194691e-06,
      "loss": 0.73575604,
      "num_input_tokens_seen": 60553190,
      "step": 2793,
      "time_per_iteration": 2.578856945037842
    },
    {
      "auxiliary_loss_clip": 0.01148071,
      "auxiliary_loss_mlp": 0.01055871,
      "balance_loss_clip": 1.05815482,
      "balance_loss_mlp": 1.03581977,
      "epoch": 0.16798436795430632,
      "flos": 19719878486400.0,
      "grad_norm": 3.7339494185166417,
      "language_loss": 0.77306449,
      "learning_rate": 3.803576041376831e-06,
      "loss": 0.79510397,
      "num_input_tokens_seen": 60571995,
      "step": 2794,
      "time_per_iteration": 2.518659830093384
    },
    {
      "auxiliary_loss_clip": 0.01142254,
      "auxiliary_loss_mlp": 0.01058243,
      "balance_loss_clip": 1.05457461,
      "balance_loss_mlp": 1.03868043,
      "epoch": 0.1680444912069743,
      "flos": 28104839596800.0,
      "grad_norm": 3.0862546171751597,
      "language_loss": 0.71866846,
      "learning_rate": 3.803407690167187e-06,
      "loss": 0.74067342,
      "num_input_tokens_seen": 60591275,
      "step": 2795,
      "time_per_iteration": 2.581310272216797
    },
    {
      "auxiliary_loss_clip": 0.01137226,
      "auxiliary_loss_mlp": 0.01041682,
      "balance_loss_clip": 1.05232096,
      "balance_loss_mlp": 1.02258492,
      "epoch": 0.16810461445964225,
      "flos": 18075205526400.0,
      "grad_norm": 2.8185065503971565,
      "language_loss": 0.84333503,
      "learning_rate": 3.803239270572142e-06,
      "loss": 0.86512411,
      "num_input_tokens_seen": 60609235,
      "step": 2796,
      "time_per_iteration": 2.5261318683624268
    },
    {
      "auxiliary_loss_clip": 0.01103091,
      "auxiliary_loss_mlp": 0.01057202,
      "balance_loss_clip": 1.04967391,
      "balance_loss_mlp": 1.03647184,
      "epoch": 0.16816473771231025,
      "flos": 23878657105920.0,
      "grad_norm": 2.094683133649052,
      "language_loss": 0.81720412,
      "learning_rate": 3.8030707825980838e-06,
      "loss": 0.83880711,
      "num_input_tokens_seen": 60629880,
      "step": 2797,
      "time_per_iteration": 2.6530420780181885
    },
    {
      "auxiliary_loss_clip": 0.01144697,
      "auxiliary_loss_mlp": 0.01052443,
      "balance_loss_clip": 1.05383873,
      "balance_loss_mlp": 1.03475273,
      "epoch": 0.1682248609649782,
      "flos": 22783597125120.0,
      "grad_norm": 1.433874522870038,
      "language_loss": 0.75096226,
      "learning_rate": 3.802902226251401e-06,
      "loss": 0.77293366,
      "num_input_tokens_seen": 60651175,
      "step": 2798,
      "time_per_iteration": 2.575908899307251
    },
    {
      "auxiliary_loss_clip": 0.011616,
      "auxiliary_loss_mlp": 0.01054228,
      "balance_loss_clip": 1.05643892,
      "balance_loss_mlp": 1.03566754,
      "epoch": 0.16828498421764618,
      "flos": 20705123612160.0,
      "grad_norm": 1.4903272763580981,
      "language_loss": 0.79860234,
      "learning_rate": 3.8027336015384845e-06,
      "loss": 0.82076061,
      "num_input_tokens_seen": 60670210,
      "step": 2799,
      "time_per_iteration": 2.484650135040283
    },
    {
      "auxiliary_loss_clip": 0.01078325,
      "auxiliary_loss_mlp": 0.01049711,
      "balance_loss_clip": 1.04935384,
      "balance_loss_mlp": 1.02811027,
      "epoch": 0.16834510747031414,
      "flos": 29420606695680.0,
      "grad_norm": 2.067409789230319,
      "language_loss": 0.7043817,
      "learning_rate": 3.8025649084657296e-06,
      "loss": 0.72566199,
      "num_input_tokens_seen": 60690895,
      "step": 2800,
      "time_per_iteration": 2.7426886558532715
    },
    {
      "auxiliary_loss_clip": 0.01116848,
      "auxiliary_loss_mlp": 0.00815912,
      "balance_loss_clip": 1.05614543,
      "balance_loss_mlp": 1.03632808,
      "epoch": 0.1684052307229821,
      "flos": 18145374744960.0,
      "grad_norm": 1.9589669041468236,
      "language_loss": 0.83796978,
      "learning_rate": 3.8023961470395326e-06,
      "loss": 0.8572973,
      "num_input_tokens_seen": 60708280,
      "step": 2801,
      "time_per_iteration": 2.572870969772339
    },
    {
      "auxiliary_loss_clip": 0.01132135,
      "auxiliary_loss_mlp": 0.01052263,
      "balance_loss_clip": 1.0545789,
      "balance_loss_mlp": 1.03270042,
      "epoch": 0.16846535397565007,
      "flos": 16574929240320.0,
      "grad_norm": 4.35534201899318,
      "language_loss": 0.83268082,
      "learning_rate": 3.8022273172662933e-06,
      "loss": 0.85452479,
      "num_input_tokens_seen": 60724150,
      "step": 2802,
      "time_per_iteration": 2.5374155044555664
    },
    {
      "auxiliary_loss_clip": 0.0114978,
      "auxiliary_loss_mlp": 0.01050041,
      "balance_loss_clip": 1.0542295,
      "balance_loss_mlp": 1.03076482,
      "epoch": 0.16852547722831807,
      "flos": 30408868563840.0,
      "grad_norm": 1.878996672053155,
      "language_loss": 0.80818331,
      "learning_rate": 3.802058419152413e-06,
      "loss": 0.83018148,
      "num_input_tokens_seen": 60746485,
      "step": 2803,
      "time_per_iteration": 2.5968077182769775
    },
    {
      "auxiliary_loss_clip": 0.01149443,
      "auxiliary_loss_mlp": 0.01051143,
      "balance_loss_clip": 1.05551386,
      "balance_loss_mlp": 1.03199804,
      "epoch": 0.16858560048098603,
      "flos": 33507420416640.0,
      "grad_norm": 4.269622778214664,
      "language_loss": 0.76563495,
      "learning_rate": 3.801889452704297e-06,
      "loss": 0.78764075,
      "num_input_tokens_seen": 60762875,
      "step": 2804,
      "time_per_iteration": 2.6066489219665527
    },
    {
      "auxiliary_loss_clip": 0.01032891,
      "auxiliary_loss_mlp": 0.01024183,
      "balance_loss_clip": 1.02185345,
      "balance_loss_mlp": 1.02048743,
      "epoch": 0.168645723733654,
      "flos": 67370502326400.0,
      "grad_norm": 0.841192131715085,
      "language_loss": 0.5541274,
      "learning_rate": 3.8017204179283526e-06,
      "loss": 0.57469809,
      "num_input_tokens_seen": 60825510,
      "step": 2805,
      "time_per_iteration": 3.1317827701568604
    },
    {
      "auxiliary_loss_clip": 0.01140098,
      "auxiliary_loss_mlp": 0.01042308,
      "balance_loss_clip": 1.04931712,
      "balance_loss_mlp": 1.02430749,
      "epoch": 0.16870584698632196,
      "flos": 21324618501120.0,
      "grad_norm": 1.8589829084899674,
      "language_loss": 0.72706515,
      "learning_rate": 3.8015513148309892e-06,
      "loss": 0.74888921,
      "num_input_tokens_seen": 60844440,
      "step": 2806,
      "time_per_iteration": 2.5054423809051514
    },
    {
      "auxiliary_loss_clip": 0.01116756,
      "auxiliary_loss_mlp": 0.01050378,
      "balance_loss_clip": 1.05362415,
      "balance_loss_mlp": 1.03114986,
      "epoch": 0.16876597023898993,
      "flos": 20740746925440.0,
      "grad_norm": 2.709972491241616,
      "language_loss": 0.69998997,
      "learning_rate": 3.80138214341862e-06,
      "loss": 0.72166133,
      "num_input_tokens_seen": 60863210,
      "step": 2807,
      "time_per_iteration": 2.558473587036133
    },
    {
      "auxiliary_loss_clip": 0.01133727,
      "auxiliary_loss_mlp": 0.01054799,
      "balance_loss_clip": 1.05029583,
      "balance_loss_mlp": 1.03434253,
      "epoch": 0.1688260934916579,
      "flos": 20303498666880.0,
      "grad_norm": 2.379129622492227,
      "language_loss": 0.70180202,
      "learning_rate": 3.8012129036976587e-06,
      "loss": 0.72368729,
      "num_input_tokens_seen": 60882510,
      "step": 2808,
      "time_per_iteration": 2.5451200008392334
    },
    {
      "auxiliary_loss_clip": 0.01122449,
      "auxiliary_loss_mlp": 0.01055377,
      "balance_loss_clip": 1.05403125,
      "balance_loss_mlp": 1.03463459,
      "epoch": 0.16888621674432586,
      "flos": 20340702178560.0,
      "grad_norm": 2.8240274502934346,
      "language_loss": 0.8008033,
      "learning_rate": 3.8010435956745236e-06,
      "loss": 0.82258153,
      "num_input_tokens_seen": 60901105,
      "step": 2809,
      "time_per_iteration": 2.5653305053710938
    },
    {
      "auxiliary_loss_clip": 0.01154969,
      "auxiliary_loss_mlp": 0.01051801,
      "balance_loss_clip": 1.05429101,
      "balance_loss_mlp": 1.03225076,
      "epoch": 0.16894633999699385,
      "flos": 16244802316800.0,
      "grad_norm": 2.673546072027647,
      "language_loss": 0.88939863,
      "learning_rate": 3.8008742193556358e-06,
      "loss": 0.91146636,
      "num_input_tokens_seen": 60915340,
      "step": 2810,
      "time_per_iteration": 2.479069948196411
    },
    {
      "auxiliary_loss_clip": 0.01155447,
      "auxiliary_loss_mlp": 0.01059385,
      "balance_loss_clip": 1.0583353,
      "balance_loss_mlp": 1.03914332,
      "epoch": 0.16900646324966181,
      "flos": 19610171372160.0,
      "grad_norm": 2.1569316294488874,
      "language_loss": 0.92818463,
      "learning_rate": 3.800704774747416e-06,
      "loss": 0.950333,
      "num_input_tokens_seen": 60933735,
      "step": 2811,
      "time_per_iteration": 2.495182514190674
    },
    {
      "auxiliary_loss_clip": 0.01143532,
      "auxiliary_loss_mlp": 0.01058253,
      "balance_loss_clip": 1.05532682,
      "balance_loss_mlp": 1.03947771,
      "epoch": 0.16906658650232978,
      "flos": 22018089450240.0,
      "grad_norm": 2.058008745810162,
      "language_loss": 0.78869855,
      "learning_rate": 3.800535261856291e-06,
      "loss": 0.81071639,
      "num_input_tokens_seen": 60953105,
      "step": 2812,
      "time_per_iteration": 2.5183701515197754
    },
    {
      "auxiliary_loss_clip": 0.01148615,
      "auxiliary_loss_mlp": 0.01053824,
      "balance_loss_clip": 1.05564356,
      "balance_loss_mlp": 1.03572762,
      "epoch": 0.16912670975499774,
      "flos": 11763690024960.0,
      "grad_norm": 2.5787391742812424,
      "language_loss": 0.74925363,
      "learning_rate": 3.8003656806886887e-06,
      "loss": 0.77127802,
      "num_input_tokens_seen": 60969150,
      "step": 2813,
      "time_per_iteration": 5.329814910888672
    },
    {
      "auxiliary_loss_clip": 0.01135406,
      "auxiliary_loss_mlp": 0.01054144,
      "balance_loss_clip": 1.05116761,
      "balance_loss_mlp": 1.03453445,
      "epoch": 0.1691868330076657,
      "flos": 17161386595200.0,
      "grad_norm": 2.7575196087928,
      "language_loss": 0.68377042,
      "learning_rate": 3.8001960312510396e-06,
      "loss": 0.70566595,
      "num_input_tokens_seen": 60982825,
      "step": 2814,
      "time_per_iteration": 2.5003573894500732
    },
    {
      "auxiliary_loss_clip": 0.01162158,
      "auxiliary_loss_mlp": 0.01047419,
      "balance_loss_clip": 1.05583191,
      "balance_loss_mlp": 1.02790463,
      "epoch": 0.16924695626033368,
      "flos": 22416553998720.0,
      "grad_norm": 2.093891283805485,
      "language_loss": 0.62375343,
      "learning_rate": 3.800026313549776e-06,
      "loss": 0.64584911,
      "num_input_tokens_seen": 61000875,
      "step": 2815,
      "time_per_iteration": 2.4901442527770996
    },
    {
      "auxiliary_loss_clip": 0.01130296,
      "auxiliary_loss_mlp": 0.01049339,
      "balance_loss_clip": 1.04966307,
      "balance_loss_mlp": 1.02970529,
      "epoch": 0.16930707951300164,
      "flos": 25739655724800.0,
      "grad_norm": 1.7606678870380874,
      "language_loss": 0.82603586,
      "learning_rate": 3.7998565275913342e-06,
      "loss": 0.84783226,
      "num_input_tokens_seen": 61021940,
      "step": 2816,
      "time_per_iteration": 3.9851162433624268
    },
    {
      "auxiliary_loss_clip": 0.01135926,
      "auxiliary_loss_mlp": 0.01052109,
      "balance_loss_clip": 1.05540931,
      "balance_loss_mlp": 1.03303576,
      "epoch": 0.16936720276566963,
      "flos": 22747040058240.0,
      "grad_norm": 2.0481528165656164,
      "language_loss": 0.87482119,
      "learning_rate": 3.799686673382153e-06,
      "loss": 0.89670157,
      "num_input_tokens_seen": 61040285,
      "step": 2817,
      "time_per_iteration": 2.5469205379486084
    },
    {
      "auxiliary_loss_clip": 0.01136878,
      "auxiliary_loss_mlp": 0.01058703,
      "balance_loss_clip": 1.05142415,
      "balance_loss_mlp": 1.0388906,
      "epoch": 0.1694273260183376,
      "flos": 19573973441280.0,
      "grad_norm": 1.6768906303230233,
      "language_loss": 0.81476372,
      "learning_rate": 3.799516750928672e-06,
      "loss": 0.83671951,
      "num_input_tokens_seen": 61059020,
      "step": 2818,
      "time_per_iteration": 2.533008098602295
    },
    {
      "auxiliary_loss_clip": 0.01157952,
      "auxiliary_loss_mlp": 0.01055409,
      "balance_loss_clip": 1.05231798,
      "balance_loss_mlp": 1.03591847,
      "epoch": 0.16948744927100556,
      "flos": 12457843332480.0,
      "grad_norm": 4.918806692755507,
      "language_loss": 0.81568801,
      "learning_rate": 3.799346760237336e-06,
      "loss": 0.83782166,
      "num_input_tokens_seen": 61074245,
      "step": 2819,
      "time_per_iteration": 2.439852476119995
    },
    {
      "auxiliary_loss_clip": 0.01041122,
      "auxiliary_loss_mlp": 0.01015026,
      "balance_loss_clip": 1.01840436,
      "balance_loss_mlp": 1.01229656,
      "epoch": 0.16954757252367353,
      "flos": 71291694435840.0,
      "grad_norm": 0.9584793877451202,
      "language_loss": 0.61029911,
      "learning_rate": 3.7991767013145902e-06,
      "loss": 0.63086057,
      "num_input_tokens_seen": 61127080,
      "step": 2820,
      "time_per_iteration": 4.429564476013184
    },
    {
      "auxiliary_loss_clip": 0.01125681,
      "auxiliary_loss_mlp": 0.010565,
      "balance_loss_clip": 1.0513742,
      "balance_loss_mlp": 1.03662801,
      "epoch": 0.1696076957763415,
      "flos": 29606516513280.0,
      "grad_norm": 1.8761655187640343,
      "language_loss": 0.78699982,
      "learning_rate": 3.7990065741668844e-06,
      "loss": 0.80882168,
      "num_input_tokens_seen": 61146955,
      "step": 2821,
      "time_per_iteration": 2.616596221923828
    },
    {
      "auxiliary_loss_clip": 0.01138749,
      "auxiliary_loss_mlp": 0.01058574,
      "balance_loss_clip": 1.05174422,
      "balance_loss_mlp": 1.03764069,
      "epoch": 0.16966781902900946,
      "flos": 24388588535040.0,
      "grad_norm": 1.9886440315890659,
      "language_loss": 0.78717947,
      "learning_rate": 3.7988363788006685e-06,
      "loss": 0.80915272,
      "num_input_tokens_seen": 61166605,
      "step": 2822,
      "time_per_iteration": 2.5253169536590576
    },
    {
      "auxiliary_loss_clip": 0.01141731,
      "auxiliary_loss_mlp": 0.00815681,
      "balance_loss_clip": 1.05007362,
      "balance_loss_mlp": 1.03610158,
      "epoch": 0.16972794228167745,
      "flos": 23038814234880.0,
      "grad_norm": 1.900436378881157,
      "language_loss": 0.75207156,
      "learning_rate": 3.7986661152223967e-06,
      "loss": 0.77164567,
      "num_input_tokens_seen": 61186535,
      "step": 2823,
      "time_per_iteration": 2.5225589275360107
    },
    {
      "auxiliary_loss_clip": 0.01131181,
      "auxiliary_loss_mlp": 0.01056333,
      "balance_loss_clip": 1.05307794,
      "balance_loss_mlp": 1.03648424,
      "epoch": 0.16978806553434542,
      "flos": 35228691129600.0,
      "grad_norm": 3.176394596317063,
      "language_loss": 0.60262775,
      "learning_rate": 3.7984957834385257e-06,
      "loss": 0.6245029,
      "num_input_tokens_seen": 61208965,
      "step": 2824,
      "time_per_iteration": 2.659937620162964
    },
    {
      "auxiliary_loss_clip": 0.01141142,
      "auxiliary_loss_mlp": 0.01045696,
      "balance_loss_clip": 1.05468094,
      "balance_loss_mlp": 1.02564538,
      "epoch": 0.16984818878701338,
      "flos": 32014290936960.0,
      "grad_norm": 1.922048086476283,
      "language_loss": 0.73146099,
      "learning_rate": 3.7983253834555144e-06,
      "loss": 0.75332934,
      "num_input_tokens_seen": 61230670,
      "step": 2825,
      "time_per_iteration": 2.626640796661377
    },
    {
      "auxiliary_loss_clip": 0.01160219,
      "auxiliary_loss_mlp": 0.01053899,
      "balance_loss_clip": 1.05124629,
      "balance_loss_mlp": 1.03236961,
      "epoch": 0.16990831203968135,
      "flos": 22818609907200.0,
      "grad_norm": 2.256685349965444,
      "language_loss": 0.85578978,
      "learning_rate": 3.7981549152798245e-06,
      "loss": 0.87793094,
      "num_input_tokens_seen": 61249510,
      "step": 2826,
      "time_per_iteration": 2.4618210792541504
    },
    {
      "auxiliary_loss_clip": 0.01139968,
      "auxiliary_loss_mlp": 0.01051969,
      "balance_loss_clip": 1.05083585,
      "balance_loss_mlp": 1.03188169,
      "epoch": 0.1699684352923493,
      "flos": 23039604334080.0,
      "grad_norm": 1.6772071811374327,
      "language_loss": 0.82641202,
      "learning_rate": 3.7979843789179196e-06,
      "loss": 0.84833133,
      "num_input_tokens_seen": 61269440,
      "step": 2827,
      "time_per_iteration": 2.532891273498535
    },
    {
      "auxiliary_loss_clip": 0.01130923,
      "auxiliary_loss_mlp": 0.01048404,
      "balance_loss_clip": 1.05160379,
      "balance_loss_mlp": 1.02763808,
      "epoch": 0.17002855854501728,
      "flos": 21434110133760.0,
      "grad_norm": 1.9489119121400609,
      "language_loss": 0.73699236,
      "learning_rate": 3.797813774376267e-06,
      "loss": 0.75878561,
      "num_input_tokens_seen": 61288195,
      "step": 2828,
      "time_per_iteration": 2.541637897491455
    },
    {
      "auxiliary_loss_clip": 0.01046746,
      "auxiliary_loss_mlp": 0.01012118,
      "balance_loss_clip": 1.03354216,
      "balance_loss_mlp": 1.00924516,
      "epoch": 0.17008868179768524,
      "flos": 71453509205760.0,
      "grad_norm": 0.7692616405283027,
      "language_loss": 0.56463939,
      "learning_rate": 3.797643101661336e-06,
      "loss": 0.58522803,
      "num_input_tokens_seen": 61350850,
      "step": 2829,
      "time_per_iteration": 3.238046169281006
    },
    {
      "auxiliary_loss_clip": 0.01116048,
      "auxiliary_loss_mlp": 0.01058123,
      "balance_loss_clip": 1.0470506,
      "balance_loss_mlp": 1.03696346,
      "epoch": 0.17014880505035324,
      "flos": 24900315644160.0,
      "grad_norm": 2.1169824269108406,
      "language_loss": 0.83226085,
      "learning_rate": 3.7974723607795983e-06,
      "loss": 0.85400259,
      "num_input_tokens_seen": 61370765,
      "step": 2830,
      "time_per_iteration": 2.596724033355713
    },
    {
      "auxiliary_loss_clip": 0.01122765,
      "auxiliary_loss_mlp": 0.01046039,
      "balance_loss_clip": 1.04926872,
      "balance_loss_mlp": 1.025177,
      "epoch": 0.1702089283030212,
      "flos": 29862415981440.0,
      "grad_norm": 2.131226399170026,
      "language_loss": 0.78473425,
      "learning_rate": 3.797301551737529e-06,
      "loss": 0.80642235,
      "num_input_tokens_seen": 61388935,
      "step": 2831,
      "time_per_iteration": 2.6141417026519775
    },
    {
      "auxiliary_loss_clip": 0.01121098,
      "auxiliary_loss_mlp": 0.01050652,
      "balance_loss_clip": 1.04808974,
      "balance_loss_mlp": 1.03020775,
      "epoch": 0.17026905155568917,
      "flos": 17744180762880.0,
      "grad_norm": 1.8909029481621262,
      "language_loss": 0.79606998,
      "learning_rate": 3.7971306745416044e-06,
      "loss": 0.81778753,
      "num_input_tokens_seen": 61407350,
      "step": 2832,
      "time_per_iteration": 2.549471616744995
    },
    {
      "auxiliary_loss_clip": 0.01127174,
      "auxiliary_loss_mlp": 0.01049976,
      "balance_loss_clip": 1.05005372,
      "balance_loss_mlp": 1.03098631,
      "epoch": 0.17032917480835713,
      "flos": 23148665003520.0,
      "grad_norm": 1.61894353972642,
      "language_loss": 0.88773465,
      "learning_rate": 3.7969597291983046e-06,
      "loss": 0.9095062,
      "num_input_tokens_seen": 61429010,
      "step": 2833,
      "time_per_iteration": 2.5802981853485107
    },
    {
      "auxiliary_loss_clip": 0.01156881,
      "auxiliary_loss_mlp": 0.01045145,
      "balance_loss_clip": 1.05127704,
      "balance_loss_mlp": 1.02648854,
      "epoch": 0.1703892980610251,
      "flos": 39202565512320.0,
      "grad_norm": 1.9953578583775617,
      "language_loss": 0.72482467,
      "learning_rate": 3.7967887157141115e-06,
      "loss": 0.74684489,
      "num_input_tokens_seen": 61450040,
      "step": 2834,
      "time_per_iteration": 2.61782169342041
    },
    {
      "auxiliary_loss_clip": 0.01116645,
      "auxiliary_loss_mlp": 0.01053443,
      "balance_loss_clip": 1.05182886,
      "balance_loss_mlp": 1.03550243,
      "epoch": 0.17044942131369306,
      "flos": 23039101543680.0,
      "grad_norm": 1.8730871466290222,
      "language_loss": 0.86474597,
      "learning_rate": 3.7966176340955106e-06,
      "loss": 0.88644683,
      "num_input_tokens_seen": 61468585,
      "step": 2835,
      "time_per_iteration": 2.581347703933716
    },
    {
      "auxiliary_loss_clip": 0.01147046,
      "auxiliary_loss_mlp": 0.01052362,
      "balance_loss_clip": 1.049124,
      "balance_loss_mlp": 1.03123844,
      "epoch": 0.17050954456636103,
      "flos": 17054983532160.0,
      "grad_norm": 2.0804408048148324,
      "language_loss": 0.73723835,
      "learning_rate": 3.796446484348989e-06,
      "loss": 0.7592324,
      "num_input_tokens_seen": 61486330,
      "step": 2836,
      "time_per_iteration": 2.501251220703125
    },
    {
      "auxiliary_loss_clip": 0.01099788,
      "auxiliary_loss_mlp": 0.01050927,
      "balance_loss_clip": 1.04683924,
      "balance_loss_mlp": 1.02983844,
      "epoch": 0.17056966781902902,
      "flos": 16836969934080.0,
      "grad_norm": 2.361696416651369,
      "language_loss": 0.80045748,
      "learning_rate": 3.796275266481036e-06,
      "loss": 0.82196468,
      "num_input_tokens_seen": 61503950,
      "step": 2837,
      "time_per_iteration": 2.574387311935425
    },
    {
      "auxiliary_loss_clip": 0.011433,
      "auxiliary_loss_mlp": 0.0104545,
      "balance_loss_clip": 1.05222154,
      "balance_loss_mlp": 1.02697217,
      "epoch": 0.17062979107169698,
      "flos": 17712543859200.0,
      "grad_norm": 2.2148056290928904,
      "language_loss": 0.83477557,
      "learning_rate": 3.7961039804981456e-06,
      "loss": 0.85666311,
      "num_input_tokens_seen": 61523550,
      "step": 2838,
      "time_per_iteration": 2.493795394897461
    },
    {
      "auxiliary_loss_clip": 0.01101927,
      "auxiliary_loss_mlp": 0.01050641,
      "balance_loss_clip": 1.04810166,
      "balance_loss_mlp": 1.03228331,
      "epoch": 0.17068991432436495,
      "flos": 22525040050560.0,
      "grad_norm": 1.668853732832902,
      "language_loss": 0.93644625,
      "learning_rate": 3.795932626406812e-06,
      "loss": 0.95797193,
      "num_input_tokens_seen": 61542720,
      "step": 2839,
      "time_per_iteration": 2.6169276237487793
    },
    {
      "auxiliary_loss_clip": 0.01126758,
      "auxiliary_loss_mlp": 0.01049011,
      "balance_loss_clip": 1.05036759,
      "balance_loss_mlp": 1.02811384,
      "epoch": 0.17075003757703291,
      "flos": 25882939077120.0,
      "grad_norm": 2.2211664911154076,
      "language_loss": 0.83836174,
      "learning_rate": 3.7957612042135336e-06,
      "loss": 0.86011934,
      "num_input_tokens_seen": 61563040,
      "step": 2840,
      "time_per_iteration": 2.5939884185791016
    },
    {
      "auxiliary_loss_clip": 0.01148966,
      "auxiliary_loss_mlp": 0.01048162,
      "balance_loss_clip": 1.05243576,
      "balance_loss_mlp": 1.0278486,
      "epoch": 0.17081016082970088,
      "flos": 20120713332480.0,
      "grad_norm": 1.8834625867711212,
      "language_loss": 0.7618587,
      "learning_rate": 3.79558971392481e-06,
      "loss": 0.78382993,
      "num_input_tokens_seen": 61581890,
      "step": 2841,
      "time_per_iteration": 2.518474578857422
    },
    {
      "auxiliary_loss_clip": 0.01139046,
      "auxiliary_loss_mlp": 0.01050159,
      "balance_loss_clip": 1.0511148,
      "balance_loss_mlp": 1.03112113,
      "epoch": 0.17087028408236885,
      "flos": 24936477661440.0,
      "grad_norm": 1.9107173062652114,
      "language_loss": 0.76767266,
      "learning_rate": 3.7954181555471443e-06,
      "loss": 0.78956473,
      "num_input_tokens_seen": 61602095,
      "step": 2842,
      "time_per_iteration": 2.5793943405151367
    },
    {
      "auxiliary_loss_clip": 0.01154695,
      "auxiliary_loss_mlp": 0.01049024,
      "balance_loss_clip": 1.05174875,
      "balance_loss_mlp": 1.02985573,
      "epoch": 0.17093040733503684,
      "flos": 19057864872960.0,
      "grad_norm": 2.4634089537561747,
      "language_loss": 0.85774994,
      "learning_rate": 3.795246529087043e-06,
      "loss": 0.87978715,
      "num_input_tokens_seen": 61620400,
      "step": 2843,
      "time_per_iteration": 2.4594357013702393
    },
    {
      "auxiliary_loss_clip": 0.01154128,
      "auxiliary_loss_mlp": 0.01046365,
      "balance_loss_clip": 1.05159092,
      "balance_loss_mlp": 1.02754235,
      "epoch": 0.1709905305877048,
      "flos": 13078954333440.0,
      "grad_norm": 2.2110534605919727,
      "language_loss": 0.68393481,
      "learning_rate": 3.7950748345510126e-06,
      "loss": 0.70593965,
      "num_input_tokens_seen": 61637680,
      "step": 2844,
      "time_per_iteration": 2.4244508743286133
    },
    {
      "auxiliary_loss_clip": 0.01134126,
      "auxiliary_loss_mlp": 0.00819531,
      "balance_loss_clip": 1.05077648,
      "balance_loss_mlp": 1.04483581,
      "epoch": 0.17105065384037277,
      "flos": 19209336526080.0,
      "grad_norm": 1.7250157561601462,
      "language_loss": 0.78357446,
      "learning_rate": 3.7949030719455646e-06,
      "loss": 0.80311108,
      "num_input_tokens_seen": 61655630,
      "step": 2845,
      "time_per_iteration": 2.5046427249908447
    },
    {
      "auxiliary_loss_clip": 0.01143701,
      "auxiliary_loss_mlp": 0.01049833,
      "balance_loss_clip": 1.05107212,
      "balance_loss_mlp": 1.03185618,
      "epoch": 0.17111077709304073,
      "flos": 18515183218560.0,
      "grad_norm": 2.3059295613339157,
      "language_loss": 0.77924454,
      "learning_rate": 3.7947312412772127e-06,
      "loss": 0.80117989,
      "num_input_tokens_seen": 61673475,
      "step": 2846,
      "time_per_iteration": 2.4778501987457275
    },
    {
      "auxiliary_loss_clip": 0.01143761,
      "auxiliary_loss_mlp": 0.01050453,
      "balance_loss_clip": 1.05174589,
      "balance_loss_mlp": 1.0316658,
      "epoch": 0.1711709003457087,
      "flos": 25082670015360.0,
      "grad_norm": 2.4505242993590137,
      "language_loss": 0.79787141,
      "learning_rate": 3.794559342552472e-06,
      "loss": 0.81981349,
      "num_input_tokens_seen": 61693370,
      "step": 2847,
      "time_per_iteration": 2.511660099029541
    },
    {
      "auxiliary_loss_clip": 0.01143354,
      "auxiliary_loss_mlp": 0.01053771,
      "balance_loss_clip": 1.04879951,
      "balance_loss_mlp": 1.03428042,
      "epoch": 0.17123102359837666,
      "flos": 17566387418880.0,
      "grad_norm": 2.2416680243469598,
      "language_loss": 0.86965472,
      "learning_rate": 3.7943873757778614e-06,
      "loss": 0.891626,
      "num_input_tokens_seen": 61710820,
      "step": 2848,
      "time_per_iteration": 2.472407579421997
    },
    {
      "auxiliary_loss_clip": 0.01114898,
      "auxiliary_loss_mlp": 0.01049191,
      "balance_loss_clip": 1.05062985,
      "balance_loss_mlp": 1.02946186,
      "epoch": 0.17129114685104463,
      "flos": 26173635845760.0,
      "grad_norm": 1.813068522178027,
      "language_loss": 0.75010598,
      "learning_rate": 3.794215340959902e-06,
      "loss": 0.77174687,
      "num_input_tokens_seen": 61729855,
      "step": 2849,
      "time_per_iteration": 2.583392381668091
    },
    {
      "auxiliary_loss_clip": 0.01038658,
      "auxiliary_loss_mlp": 0.01007124,
      "balance_loss_clip": 1.03035414,
      "balance_loss_mlp": 1.00442946,
      "epoch": 0.17135127010371262,
      "flos": 69269710037760.0,
      "grad_norm": 0.7959005706383175,
      "language_loss": 0.57507271,
      "learning_rate": 3.7940432381051163e-06,
      "loss": 0.59553051,
      "num_input_tokens_seen": 61790290,
      "step": 2850,
      "time_per_iteration": 3.1467602252960205
    },
    {
      "auxiliary_loss_clip": 0.01116767,
      "auxiliary_loss_mlp": 0.010455,
      "balance_loss_clip": 1.04968834,
      "balance_loss_mlp": 1.02658117,
      "epoch": 0.1714113933563806,
      "flos": 23550110380800.0,
      "grad_norm": 2.11599141610493,
      "language_loss": 0.81807655,
      "learning_rate": 3.793871067220031e-06,
      "loss": 0.83969921,
      "num_input_tokens_seen": 61809265,
      "step": 2851,
      "time_per_iteration": 4.017199516296387
    },
    {
      "auxiliary_loss_clip": 0.01113454,
      "auxiliary_loss_mlp": 0.01045212,
      "balance_loss_clip": 1.04965353,
      "balance_loss_mlp": 1.02694929,
      "epoch": 0.17147151660904855,
      "flos": 21142443697920.0,
      "grad_norm": 1.8869235806991436,
      "language_loss": 0.93335748,
      "learning_rate": 3.7936988283111764e-06,
      "loss": 0.95494413,
      "num_input_tokens_seen": 61828980,
      "step": 2852,
      "time_per_iteration": 2.568448543548584
    },
    {
      "auxiliary_loss_clip": 0.01123863,
      "auxiliary_loss_mlp": 0.01052194,
      "balance_loss_clip": 1.0502429,
      "balance_loss_mlp": 1.03318,
      "epoch": 0.17153163986171652,
      "flos": 18624890332800.0,
      "grad_norm": 9.605017439504994,
      "language_loss": 0.69408739,
      "learning_rate": 3.7935265213850817e-06,
      "loss": 0.71584797,
      "num_input_tokens_seen": 61847915,
      "step": 2853,
      "time_per_iteration": 2.564440965652466
    },
    {
      "auxiliary_loss_clip": 0.01122869,
      "auxiliary_loss_mlp": 0.0104638,
      "balance_loss_clip": 1.05676186,
      "balance_loss_mlp": 1.02805793,
      "epoch": 0.17159176311438448,
      "flos": 18223265387520.0,
      "grad_norm": 2.484251734421179,
      "language_loss": 0.66595155,
      "learning_rate": 3.7933541464482815e-06,
      "loss": 0.687644,
      "num_input_tokens_seen": 61865570,
      "step": 2854,
      "time_per_iteration": 4.05839729309082
    },
    {
      "auxiliary_loss_clip": 0.01124257,
      "auxiliary_loss_mlp": 0.01041639,
      "balance_loss_clip": 1.05196619,
      "balance_loss_mlp": 1.02316189,
      "epoch": 0.17165188636705245,
      "flos": 20738987159040.0,
      "grad_norm": 1.6082338942048273,
      "language_loss": 0.89584315,
      "learning_rate": 3.7931817035073124e-06,
      "loss": 0.91750211,
      "num_input_tokens_seen": 61883340,
      "step": 2855,
      "time_per_iteration": 2.5535194873809814
    },
    {
      "auxiliary_loss_clip": 0.01160299,
      "auxiliary_loss_mlp": 0.01052593,
      "balance_loss_clip": 1.05430984,
      "balance_loss_mlp": 1.03396046,
      "epoch": 0.17171200961972044,
      "flos": 24899884680960.0,
      "grad_norm": 2.17527094775061,
      "language_loss": 0.83149886,
      "learning_rate": 3.7930091925687134e-06,
      "loss": 0.8536278,
      "num_input_tokens_seen": 61900610,
      "step": 2856,
      "time_per_iteration": 2.5007376670837402
    },
    {
      "auxiliary_loss_clip": 0.01148659,
      "auxiliary_loss_mlp": 0.01051232,
      "balance_loss_clip": 1.05486631,
      "balance_loss_mlp": 1.03267086,
      "epoch": 0.1717721328723884,
      "flos": 20157234485760.0,
      "grad_norm": 2.0522905906691995,
      "language_loss": 0.86260235,
      "learning_rate": 3.792836613639026e-06,
      "loss": 0.88460124,
      "num_input_tokens_seen": 61916795,
      "step": 2857,
      "time_per_iteration": 2.510795831680298
    },
    {
      "auxiliary_loss_clip": 0.01148081,
      "auxiliary_loss_mlp": 0.01056964,
      "balance_loss_clip": 1.05462861,
      "balance_loss_mlp": 1.03703237,
      "epoch": 0.17183225612505637,
      "flos": 23361650697600.0,
      "grad_norm": 2.135113707951393,
      "language_loss": 0.78303277,
      "learning_rate": 3.7926639667247947e-06,
      "loss": 0.80508327,
      "num_input_tokens_seen": 61936665,
      "step": 2858,
      "time_per_iteration": 3.926889657974243
    },
    {
      "auxiliary_loss_clip": 0.01160172,
      "auxiliary_loss_mlp": 0.01053156,
      "balance_loss_clip": 1.07450402,
      "balance_loss_mlp": 1.03256857,
      "epoch": 0.17189237937772434,
      "flos": 18114240631680.0,
      "grad_norm": 2.300572422560323,
      "language_loss": 0.77463281,
      "learning_rate": 3.7924912518325663e-06,
      "loss": 0.7967661,
      "num_input_tokens_seen": 61954415,
      "step": 2859,
      "time_per_iteration": 2.5157456398010254
    },
    {
      "auxiliary_loss_clip": 0.01120102,
      "auxiliary_loss_mlp": 0.01041227,
      "balance_loss_clip": 1.06873417,
      "balance_loss_mlp": 1.02162886,
      "epoch": 0.1719525026303923,
      "flos": 23258408031360.0,
      "grad_norm": 2.3104229179389293,
      "language_loss": 0.77066261,
      "learning_rate": 3.7923184689688902e-06,
      "loss": 0.79227591,
      "num_input_tokens_seen": 61973940,
      "step": 2860,
      "time_per_iteration": 2.62282133102417
    },
    {
      "auxiliary_loss_clip": 0.01148398,
      "auxiliary_loss_mlp": 0.0104497,
      "balance_loss_clip": 1.05280316,
      "balance_loss_mlp": 1.02606344,
      "epoch": 0.17201262588306027,
      "flos": 20810413353600.0,
      "grad_norm": 2.6750246423837787,
      "language_loss": 0.81662339,
      "learning_rate": 3.792145618140317e-06,
      "loss": 0.83855712,
      "num_input_tokens_seen": 61991845,
      "step": 2861,
      "time_per_iteration": 2.5068347454071045
    },
    {
      "auxiliary_loss_clip": 0.01132087,
      "auxiliary_loss_mlp": 0.01051852,
      "balance_loss_clip": 1.05180585,
      "balance_loss_mlp": 1.03352928,
      "epoch": 0.17207274913572823,
      "flos": 20375858615040.0,
      "grad_norm": 2.3776555817811356,
      "language_loss": 0.85964882,
      "learning_rate": 3.7919726993534038e-06,
      "loss": 0.8814882,
      "num_input_tokens_seen": 62009395,
      "step": 2862,
      "time_per_iteration": 2.5348544120788574
    },
    {
      "auxiliary_loss_clip": 0.01120728,
      "auxiliary_loss_mlp": 0.01041975,
      "balance_loss_clip": 1.05348825,
      "balance_loss_mlp": 1.0247612,
      "epoch": 0.17213287238839622,
      "flos": 26797727675520.0,
      "grad_norm": 1.8299949932070538,
      "language_loss": 0.78020906,
      "learning_rate": 3.7917997126147054e-06,
      "loss": 0.80183613,
      "num_input_tokens_seen": 62029005,
      "step": 2863,
      "time_per_iteration": 2.620295524597168
    },
    {
      "auxiliary_loss_clip": 0.01125914,
      "auxiliary_loss_mlp": 0.00809431,
      "balance_loss_clip": 1.05410981,
      "balance_loss_mlp": 1.02629542,
      "epoch": 0.1721929956410642,
      "flos": 26030819370240.0,
      "grad_norm": 1.8323399339164812,
      "language_loss": 0.72376615,
      "learning_rate": 3.7916266579307823e-06,
      "loss": 0.7431196,
      "num_input_tokens_seen": 62048730,
      "step": 2864,
      "time_per_iteration": 2.639220952987671
    },
    {
      "auxiliary_loss_clip": 0.01123396,
      "auxiliary_loss_mlp": 0.01052639,
      "balance_loss_clip": 1.05321884,
      "balance_loss_mlp": 1.03326738,
      "epoch": 0.17225311889373215,
      "flos": 22273091078400.0,
      "grad_norm": 1.6501035713580428,
      "language_loss": 0.72621214,
      "learning_rate": 3.7914535353081973e-06,
      "loss": 0.74797249,
      "num_input_tokens_seen": 62069000,
      "step": 2865,
      "time_per_iteration": 2.5765178203582764
    },
    {
      "auxiliary_loss_clip": 0.01146229,
      "auxiliary_loss_mlp": 0.00816258,
      "balance_loss_clip": 1.05602038,
      "balance_loss_mlp": 1.03677559,
      "epoch": 0.17231324214640012,
      "flos": 21287774125440.0,
      "grad_norm": 3.3573541874259583,
      "language_loss": 0.7855742,
      "learning_rate": 3.7912803447535145e-06,
      "loss": 0.80519903,
      "num_input_tokens_seen": 62086750,
      "step": 2866,
      "time_per_iteration": 2.525108814239502
    },
    {
      "auxiliary_loss_clip": 0.01161477,
      "auxiliary_loss_mlp": 0.01048584,
      "balance_loss_clip": 1.05604911,
      "balance_loss_mlp": 1.02875924,
      "epoch": 0.17237336539906808,
      "flos": 19680735640320.0,
      "grad_norm": 1.7022854103544645,
      "language_loss": 0.79753268,
      "learning_rate": 3.7911070862733016e-06,
      "loss": 0.81963325,
      "num_input_tokens_seen": 62106240,
      "step": 2867,
      "time_per_iteration": 2.493462562561035
    },
    {
      "auxiliary_loss_clip": 0.01132599,
      "auxiliary_loss_mlp": 0.01039114,
      "balance_loss_clip": 1.05193472,
      "balance_loss_mlp": 1.02013576,
      "epoch": 0.17243348865173605,
      "flos": 17529650784000.0,
      "grad_norm": 1.8120038090180435,
      "language_loss": 0.79348248,
      "learning_rate": 3.7909337598741276e-06,
      "loss": 0.81519961,
      "num_input_tokens_seen": 62124895,
      "step": 2868,
      "time_per_iteration": 2.519169807434082
    },
    {
      "auxiliary_loss_clip": 0.01130655,
      "auxiliary_loss_mlp": 0.01044807,
      "balance_loss_clip": 1.07659304,
      "balance_loss_mlp": 1.0270927,
      "epoch": 0.17249361190440402,
      "flos": 18259858368000.0,
      "grad_norm": 2.1895452992331887,
      "language_loss": 0.83900356,
      "learning_rate": 3.7907603655625674e-06,
      "loss": 0.86075819,
      "num_input_tokens_seen": 62143510,
      "step": 2869,
      "time_per_iteration": 2.587867498397827
    },
    {
      "auxiliary_loss_clip": 0.01133376,
      "auxiliary_loss_mlp": 0.01049601,
      "balance_loss_clip": 1.05313921,
      "balance_loss_mlp": 1.03039598,
      "epoch": 0.172553735157072,
      "flos": 21174367910400.0,
      "grad_norm": 15.045694308476499,
      "language_loss": 0.7726475,
      "learning_rate": 3.7905869033451932e-06,
      "loss": 0.79447722,
      "num_input_tokens_seen": 62162285,
      "step": 2870,
      "time_per_iteration": 2.5454421043395996
    },
    {
      "auxiliary_loss_clip": 0.01154093,
      "auxiliary_loss_mlp": 0.01043176,
      "balance_loss_clip": 1.0541718,
      "balance_loss_mlp": 1.02612948,
      "epoch": 0.17261385840973997,
      "flos": 22273270646400.0,
      "grad_norm": 1.9954769955860114,
      "language_loss": 0.77207547,
      "learning_rate": 3.7904133732285857e-06,
      "loss": 0.79404813,
      "num_input_tokens_seen": 62180970,
      "step": 2871,
      "time_per_iteration": 2.48834228515625
    },
    {
      "auxiliary_loss_clip": 0.0113783,
      "auxiliary_loss_mlp": 0.01045111,
      "balance_loss_clip": 1.05822349,
      "balance_loss_mlp": 1.02597761,
      "epoch": 0.17267398166240794,
      "flos": 27922233830400.0,
      "grad_norm": 3.3521514224342,
      "language_loss": 0.74119055,
      "learning_rate": 3.7902397752193228e-06,
      "loss": 0.76301998,
      "num_input_tokens_seen": 62198965,
      "step": 2872,
      "time_per_iteration": 2.6126041412353516
    },
    {
      "auxiliary_loss_clip": 0.0115308,
      "auxiliary_loss_mlp": 0.01047126,
      "balance_loss_clip": 1.05242538,
      "balance_loss_mlp": 1.02838659,
      "epoch": 0.1727341049150759,
      "flos": 21945118970880.0,
      "grad_norm": 1.802066445844266,
      "language_loss": 0.82550919,
      "learning_rate": 3.790066109323988e-06,
      "loss": 0.84751129,
      "num_input_tokens_seen": 62219890,
      "step": 2873,
      "time_per_iteration": 2.499656915664673
    },
    {
      "auxiliary_loss_clip": 0.01111993,
      "auxiliary_loss_mlp": 0.01047193,
      "balance_loss_clip": 1.04831696,
      "balance_loss_mlp": 1.02734423,
      "epoch": 0.17279422816774387,
      "flos": 18107883924480.0,
      "grad_norm": 1.9589609037378661,
      "language_loss": 0.75076938,
      "learning_rate": 3.7898923755491678e-06,
      "loss": 0.77236122,
      "num_input_tokens_seen": 62237140,
      "step": 2874,
      "time_per_iteration": 2.5713133811950684
    },
    {
      "auxiliary_loss_clip": 0.01157366,
      "auxiliary_loss_mlp": 0.01044858,
      "balance_loss_clip": 1.05296826,
      "balance_loss_mlp": 1.02504563,
      "epoch": 0.17285435142041183,
      "flos": 21835447770240.0,
      "grad_norm": 2.6546478076809614,
      "language_loss": 0.81097305,
      "learning_rate": 3.7897185739014487e-06,
      "loss": 0.8329953,
      "num_input_tokens_seen": 62255405,
      "step": 2875,
      "time_per_iteration": 2.483370542526245
    },
    {
      "auxiliary_loss_clip": 0.0113856,
      "auxiliary_loss_mlp": 0.01050962,
      "balance_loss_clip": 1.05571795,
      "balance_loss_mlp": 1.03106582,
      "epoch": 0.17291447467307983,
      "flos": 18368452160640.0,
      "grad_norm": 2.519842669443865,
      "language_loss": 0.87820888,
      "learning_rate": 3.7895447043874217e-06,
      "loss": 0.90010417,
      "num_input_tokens_seen": 62271280,
      "step": 2876,
      "time_per_iteration": 2.5212438106536865
    },
    {
      "auxiliary_loss_clip": 0.01137039,
      "auxiliary_loss_mlp": 0.01044408,
      "balance_loss_clip": 1.05570066,
      "balance_loss_mlp": 1.02522755,
      "epoch": 0.1729745979257478,
      "flos": 18624638937600.0,
      "grad_norm": 1.7344409557400629,
      "language_loss": 0.8494249,
      "learning_rate": 3.789370767013681e-06,
      "loss": 0.87123942,
      "num_input_tokens_seen": 62289140,
      "step": 2877,
      "time_per_iteration": 2.522310972213745
    },
    {
      "auxiliary_loss_clip": 0.01134601,
      "auxiliary_loss_mlp": 0.01046224,
      "balance_loss_clip": 1.06645799,
      "balance_loss_mlp": 1.02678108,
      "epoch": 0.17303472117841576,
      "flos": 22998234844800.0,
      "grad_norm": 2.2471510325457085,
      "language_loss": 0.79827631,
      "learning_rate": 3.7891967617868204e-06,
      "loss": 0.82008457,
      "num_input_tokens_seen": 62307490,
      "step": 2878,
      "time_per_iteration": 2.606806993484497
    },
    {
      "auxiliary_loss_clip": 0.01134007,
      "auxiliary_loss_mlp": 0.01048817,
      "balance_loss_clip": 1.05076122,
      "balance_loss_mlp": 1.03045881,
      "epoch": 0.17309484443108372,
      "flos": 25664386775040.0,
      "grad_norm": 3.44638218731754,
      "language_loss": 0.70749921,
      "learning_rate": 3.78902268871344e-06,
      "loss": 0.72932744,
      "num_input_tokens_seen": 62328570,
      "step": 2879,
      "time_per_iteration": 2.59049391746521
    },
    {
      "auxiliary_loss_clip": 0.01129497,
      "auxiliary_loss_mlp": 0.01051114,
      "balance_loss_clip": 1.04865837,
      "balance_loss_mlp": 1.03233874,
      "epoch": 0.1731549676837517,
      "flos": 13552903313280.0,
      "grad_norm": 1.956952823864924,
      "language_loss": 0.83193946,
      "learning_rate": 3.78884854780014e-06,
      "loss": 0.85374558,
      "num_input_tokens_seen": 62345735,
      "step": 2880,
      "time_per_iteration": 2.52827525138855
    },
    {
      "auxiliary_loss_clip": 0.01111483,
      "auxiliary_loss_mlp": 0.01055633,
      "balance_loss_clip": 1.05306923,
      "balance_loss_mlp": 1.03343642,
      "epoch": 0.17321509093641965,
      "flos": 22857070394880.0,
      "grad_norm": 1.882345190860551,
      "language_loss": 0.81325865,
      "learning_rate": 3.7886743390535236e-06,
      "loss": 0.83492982,
      "num_input_tokens_seen": 62365525,
      "step": 2881,
      "time_per_iteration": 2.59496808052063
    },
    {
      "auxiliary_loss_clip": 0.01136789,
      "auxiliary_loss_mlp": 0.01045535,
      "balance_loss_clip": 1.05265951,
      "balance_loss_mlp": 1.02773666,
      "epoch": 0.17327521418908762,
      "flos": 24352785653760.0,
      "grad_norm": 2.075185895255664,
      "language_loss": 0.77298248,
      "learning_rate": 3.788500062480197e-06,
      "loss": 0.79480577,
      "num_input_tokens_seen": 62385160,
      "step": 2882,
      "time_per_iteration": 2.5467443466186523
    },
    {
      "auxiliary_loss_clip": 0.01116555,
      "auxiliary_loss_mlp": 0.01053975,
      "balance_loss_clip": 1.05174637,
      "balance_loss_mlp": 1.03523517,
      "epoch": 0.1733353374417556,
      "flos": 33105651816960.0,
      "grad_norm": 2.2844508223984272,
      "language_loss": 0.76611286,
      "learning_rate": 3.788325718086769e-06,
      "loss": 0.78781819,
      "num_input_tokens_seen": 62405280,
      "step": 2883,
      "time_per_iteration": 2.706714391708374
    },
    {
      "auxiliary_loss_clip": 0.0111167,
      "auxiliary_loss_mlp": 0.01045112,
      "balance_loss_clip": 1.04906106,
      "balance_loss_mlp": 1.02661061,
      "epoch": 0.17339546069442358,
      "flos": 24388947671040.0,
      "grad_norm": 2.0213885815811605,
      "language_loss": 0.85953999,
      "learning_rate": 3.7881513058798503e-06,
      "loss": 0.88110775,
      "num_input_tokens_seen": 62423665,
      "step": 2884,
      "time_per_iteration": 2.589056968688965
    },
    {
      "auxiliary_loss_clip": 0.01138139,
      "auxiliary_loss_mlp": 0.00810753,
      "balance_loss_clip": 1.0543443,
      "balance_loss_mlp": 1.02889562,
      "epoch": 0.17345558394709154,
      "flos": 27454174680960.0,
      "grad_norm": 2.458062701847446,
      "language_loss": 0.74333674,
      "learning_rate": 3.787976825866055e-06,
      "loss": 0.76282567,
      "num_input_tokens_seen": 62445170,
      "step": 2885,
      "time_per_iteration": 2.626267671585083
    },
    {
      "auxiliary_loss_clip": 0.01129591,
      "auxiliary_loss_mlp": 0.01044453,
      "balance_loss_clip": 1.05308473,
      "balance_loss_mlp": 1.02745342,
      "epoch": 0.1735157071997595,
      "flos": 24682158391680.0,
      "grad_norm": 1.5538529015647276,
      "language_loss": 0.70770383,
      "learning_rate": 3.7878022780519998e-06,
      "loss": 0.72944427,
      "num_input_tokens_seen": 62466135,
      "step": 2886,
      "time_per_iteration": 2.5850133895874023
    },
    {
      "auxiliary_loss_clip": 0.01145017,
      "auxiliary_loss_mlp": 0.01043581,
      "balance_loss_clip": 1.05082798,
      "balance_loss_mlp": 1.02468622,
      "epoch": 0.17357583045242747,
      "flos": 21688932193920.0,
      "grad_norm": 2.5645110637781627,
      "language_loss": 0.69637185,
      "learning_rate": 3.7876276624443024e-06,
      "loss": 0.71825778,
      "num_input_tokens_seen": 62483910,
      "step": 2887,
      "time_per_iteration": 2.5200905799865723
    },
    {
      "auxiliary_loss_clip": 0.0111647,
      "auxiliary_loss_mlp": 0.01048838,
      "balance_loss_clip": 1.05194426,
      "balance_loss_mlp": 1.03055155,
      "epoch": 0.17363595370509544,
      "flos": 15375728753280.0,
      "grad_norm": 1.6220846818290815,
      "language_loss": 0.8487035,
      "learning_rate": 3.787452979049585e-06,
      "loss": 0.87035656,
      "num_input_tokens_seen": 62501530,
      "step": 2888,
      "time_per_iteration": 2.555405378341675
    },
    {
      "auxiliary_loss_clip": 0.01093828,
      "auxiliary_loss_mlp": 0.01053272,
      "balance_loss_clip": 1.04740274,
      "balance_loss_mlp": 1.03142107,
      "epoch": 0.1736960769577634,
      "flos": 23440941970560.0,
      "grad_norm": 2.196358872859896,
      "language_loss": 0.78521848,
      "learning_rate": 3.7872782278744718e-06,
      "loss": 0.8066895,
      "num_input_tokens_seen": 62521295,
      "step": 2889,
      "time_per_iteration": 4.0509352684021
    },
    {
      "auxiliary_loss_clip": 0.01112576,
      "auxiliary_loss_mlp": 0.00810837,
      "balance_loss_clip": 1.05193305,
      "balance_loss_mlp": 1.03017211,
      "epoch": 0.1737562002104314,
      "flos": 18587830475520.0,
      "grad_norm": 2.087964727654909,
      "language_loss": 0.8406595,
      "learning_rate": 3.7871034089255883e-06,
      "loss": 0.85989368,
      "num_input_tokens_seen": 62539615,
      "step": 2890,
      "time_per_iteration": 3.9716060161590576
    },
    {
      "auxiliary_loss_clip": 0.01143392,
      "auxiliary_loss_mlp": 0.01046908,
      "balance_loss_clip": 1.05453706,
      "balance_loss_mlp": 1.02801323,
      "epoch": 0.17381632346309936,
      "flos": 15998060816640.0,
      "grad_norm": 2.177067399017797,
      "language_loss": 0.81991309,
      "learning_rate": 3.7869285222095653e-06,
      "loss": 0.84181607,
      "num_input_tokens_seen": 62556820,
      "step": 2891,
      "time_per_iteration": 2.5609354972839355
    },
    {
      "auxiliary_loss_clip": 0.01099412,
      "auxiliary_loss_mlp": 0.01055288,
      "balance_loss_clip": 1.04281652,
      "balance_loss_mlp": 1.03288865,
      "epoch": 0.17387644671576732,
      "flos": 13369830670080.0,
      "grad_norm": 2.209771387820264,
      "language_loss": 0.81475335,
      "learning_rate": 3.7867535677330334e-06,
      "loss": 0.83630037,
      "num_input_tokens_seen": 62572450,
      "step": 2892,
      "time_per_iteration": 2.5467910766601562
    },
    {
      "auxiliary_loss_clip": 0.0115299,
      "auxiliary_loss_mlp": 0.01055426,
      "balance_loss_clip": 1.05900836,
      "balance_loss_mlp": 1.03562522,
      "epoch": 0.1739365699684353,
      "flos": 26615516958720.0,
      "grad_norm": 1.9176163667144206,
      "language_loss": 0.74642324,
      "learning_rate": 3.786578545502627e-06,
      "loss": 0.76850736,
      "num_input_tokens_seen": 62592580,
      "step": 2893,
      "time_per_iteration": 4.018085956573486
    },
    {
      "auxiliary_loss_clip": 0.01134681,
      "auxiliary_loss_mlp": 0.01045023,
      "balance_loss_clip": 1.05330515,
      "balance_loss_mlp": 1.02544916,
      "epoch": 0.17399669322110325,
      "flos": 23367971491200.0,
      "grad_norm": 1.9665593900601435,
      "language_loss": 0.83077586,
      "learning_rate": 3.7864034555249828e-06,
      "loss": 0.85257286,
      "num_input_tokens_seen": 62611220,
      "step": 2894,
      "time_per_iteration": 2.5483009815216064
    },
    {
      "auxiliary_loss_clip": 0.01114045,
      "auxiliary_loss_mlp": 0.01044327,
      "balance_loss_clip": 1.05007482,
      "balance_loss_mlp": 1.02249968,
      "epoch": 0.17405681647377122,
      "flos": 22054107813120.0,
      "grad_norm": 2.5608366414010697,
      "language_loss": 0.74558687,
      "learning_rate": 3.786228297806741e-06,
      "loss": 0.76717061,
      "num_input_tokens_seen": 62629185,
      "step": 2895,
      "time_per_iteration": 2.582894802093506
    },
    {
      "auxiliary_loss_clip": 0.01037035,
      "auxiliary_loss_mlp": 0.01049476,
      "balance_loss_clip": 1.04274738,
      "balance_loss_mlp": 1.04623365,
      "epoch": 0.1741169397264392,
      "flos": 61457559114240.0,
      "grad_norm": 0.9170684436718151,
      "language_loss": 0.62770057,
      "learning_rate": 3.7860530723545435e-06,
      "loss": 0.64856565,
      "num_input_tokens_seen": 62691895,
      "step": 2896,
      "time_per_iteration": 3.274322748184204
    },
    {
      "auxiliary_loss_clip": 0.01131099,
      "auxiliary_loss_mlp": 0.0080351,
      "balance_loss_clip": 1.05140066,
      "balance_loss_mlp": 1.01638937,
      "epoch": 0.17417706297910718,
      "flos": 27017680608000.0,
      "grad_norm": 1.5861034590362757,
      "language_loss": 0.75830138,
      "learning_rate": 3.785877779175034e-06,
      "loss": 0.7776475,
      "num_input_tokens_seen": 62713790,
      "step": 2897,
      "time_per_iteration": 4.00433874130249
    },
    {
      "auxiliary_loss_clip": 0.01143519,
      "auxiliary_loss_mlp": 0.01040146,
      "balance_loss_clip": 1.05438113,
      "balance_loss_mlp": 1.02163267,
      "epoch": 0.17423718623177514,
      "flos": 33508856960640.0,
      "grad_norm": 1.9228870348704274,
      "language_loss": 0.69416428,
      "learning_rate": 3.7857024182748606e-06,
      "loss": 0.71600091,
      "num_input_tokens_seen": 62736285,
      "step": 2898,
      "time_per_iteration": 2.6113712787628174
    },
    {
      "auxiliary_loss_clip": 0.011343,
      "auxiliary_loss_mlp": 0.01042858,
      "balance_loss_clip": 1.05959964,
      "balance_loss_mlp": 1.02355766,
      "epoch": 0.1742973094844431,
      "flos": 27198634348800.0,
      "grad_norm": 2.441435762802727,
      "language_loss": 0.76013374,
      "learning_rate": 3.7855269896606717e-06,
      "loss": 0.78190535,
      "num_input_tokens_seen": 62756240,
      "step": 2899,
      "time_per_iteration": 2.5884125232696533
    },
    {
      "auxiliary_loss_clip": 0.01096951,
      "auxiliary_loss_mlp": 0.01047743,
      "balance_loss_clip": 1.0499835,
      "balance_loss_mlp": 1.02776325,
      "epoch": 0.17435743273711107,
      "flos": 22710734386560.0,
      "grad_norm": 1.8304664900337155,
      "language_loss": 0.72745997,
      "learning_rate": 3.785351493339121e-06,
      "loss": 0.74890685,
      "num_input_tokens_seen": 62775910,
      "step": 2900,
      "time_per_iteration": 2.6228227615356445
    },
    {
      "auxiliary_loss_clip": 0.01110019,
      "auxiliary_loss_mlp": 0.00806612,
      "balance_loss_clip": 1.05024481,
      "balance_loss_mlp": 1.02162075,
      "epoch": 0.17441755598977904,
      "flos": 41646466039680.0,
      "grad_norm": 1.603977032354177,
      "language_loss": 0.69919294,
      "learning_rate": 3.785175929316863e-06,
      "loss": 0.71835923,
      "num_input_tokens_seen": 62799385,
      "step": 2901,
      "time_per_iteration": 2.760504722595215
    },
    {
      "auxiliary_loss_clip": 0.01134691,
      "auxiliary_loss_mlp": 0.01055807,
      "balance_loss_clip": 1.05881882,
      "balance_loss_mlp": 1.03707922,
      "epoch": 0.174477679242447,
      "flos": 26287077974400.0,
      "grad_norm": 2.0881185534782403,
      "language_loss": 0.7653079,
      "learning_rate": 3.7850002976005543e-06,
      "loss": 0.78721285,
      "num_input_tokens_seen": 62819380,
      "step": 2902,
      "time_per_iteration": 2.6073036193847656
    },
    {
      "auxiliary_loss_clip": 0.01147196,
      "auxiliary_loss_mlp": 0.0105225,
      "balance_loss_clip": 1.05186105,
      "balance_loss_mlp": 1.03389192,
      "epoch": 0.174537802495115,
      "flos": 17858412990720.0,
      "grad_norm": 1.9897002853826633,
      "language_loss": 0.81272733,
      "learning_rate": 3.7848245981968558e-06,
      "loss": 0.83472186,
      "num_input_tokens_seen": 62836205,
      "step": 2903,
      "time_per_iteration": 2.4902195930480957
    },
    {
      "auxiliary_loss_clip": 0.01125882,
      "auxiliary_loss_mlp": 0.01042788,
      "balance_loss_clip": 1.05343187,
      "balance_loss_mlp": 1.02392936,
      "epoch": 0.17459792574778296,
      "flos": 16940715390720.0,
      "grad_norm": 1.9274455743239058,
      "language_loss": 0.73514318,
      "learning_rate": 3.784648831112429e-06,
      "loss": 0.75682992,
      "num_input_tokens_seen": 62854045,
      "step": 2904,
      "time_per_iteration": 2.5359294414520264
    },
    {
      "auxiliary_loss_clip": 0.01114866,
      "auxiliary_loss_mlp": 0.01048177,
      "balance_loss_clip": 1.06286395,
      "balance_loss_mlp": 1.02953219,
      "epoch": 0.17465804900045093,
      "flos": 25520026014720.0,
      "grad_norm": 1.819004443431058,
      "language_loss": 0.64677048,
      "learning_rate": 3.7844729963539406e-06,
      "loss": 0.66840088,
      "num_input_tokens_seen": 62873075,
      "step": 2905,
      "time_per_iteration": 2.665915012359619
    },
    {
      "auxiliary_loss_clip": 0.01130966,
      "auxiliary_loss_mlp": 0.01048798,
      "balance_loss_clip": 1.05209064,
      "balance_loss_mlp": 1.0290693,
      "epoch": 0.1747181722531189,
      "flos": 24129708238080.0,
      "grad_norm": 1.7576675873137075,
      "language_loss": 0.79257977,
      "learning_rate": 3.7842970939280566e-06,
      "loss": 0.81437743,
      "num_input_tokens_seen": 62892675,
      "step": 2906,
      "time_per_iteration": 2.589737892150879
    },
    {
      "auxiliary_loss_clip": 0.01147097,
      "auxiliary_loss_mlp": 0.01052327,
      "balance_loss_clip": 1.0552305,
      "balance_loss_mlp": 1.03317058,
      "epoch": 0.17477829550578686,
      "flos": 17748813617280.0,
      "grad_norm": 1.7449451131445253,
      "language_loss": 0.81293046,
      "learning_rate": 3.784121123841449e-06,
      "loss": 0.8349247,
      "num_input_tokens_seen": 62910675,
      "step": 2907,
      "time_per_iteration": 2.483914375305176
    },
    {
      "auxiliary_loss_clip": 0.01144974,
      "auxiliary_loss_mlp": 0.01050364,
      "balance_loss_clip": 1.05353308,
      "balance_loss_mlp": 1.03238726,
      "epoch": 0.17483841875845482,
      "flos": 15377344865280.0,
      "grad_norm": 2.184213289837377,
      "language_loss": 0.8074159,
      "learning_rate": 3.7839450861007886e-06,
      "loss": 0.82936925,
      "num_input_tokens_seen": 62928130,
      "step": 2908,
      "time_per_iteration": 2.5117392539978027
    },
    {
      "auxiliary_loss_clip": 0.01127626,
      "auxiliary_loss_mlp": 0.01053014,
      "balance_loss_clip": 1.05169904,
      "balance_loss_mlp": 1.03341591,
      "epoch": 0.17489854201112282,
      "flos": 17163254102400.0,
      "grad_norm": 2.8160251244991485,
      "language_loss": 0.80545259,
      "learning_rate": 3.7837689807127518e-06,
      "loss": 0.827259,
      "num_input_tokens_seen": 62944290,
      "step": 2909,
      "time_per_iteration": 2.5063986778259277
    },
    {
      "auxiliary_loss_clip": 0.01086176,
      "auxiliary_loss_mlp": 0.01059912,
      "balance_loss_clip": 1.04710054,
      "balance_loss_mlp": 1.03705943,
      "epoch": 0.17495866526379078,
      "flos": 19755286318080.0,
      "grad_norm": 1.6675221592735325,
      "language_loss": 0.76633048,
      "learning_rate": 3.783592807684017e-06,
      "loss": 0.78779137,
      "num_input_tokens_seen": 62963505,
      "step": 2910,
      "time_per_iteration": 2.60888934135437
    },
    {
      "auxiliary_loss_clip": 0.01160387,
      "auxiliary_loss_mlp": 0.01054138,
      "balance_loss_clip": 1.05428052,
      "balance_loss_mlp": 1.03376555,
      "epoch": 0.17501878851645875,
      "flos": 28511133310080.0,
      "grad_norm": 1.8879186430586612,
      "language_loss": 0.87313139,
      "learning_rate": 3.7834165670212645e-06,
      "loss": 0.89527667,
      "num_input_tokens_seen": 62985020,
      "step": 2911,
      "time_per_iteration": 2.546182870864868
    },
    {
      "auxiliary_loss_clip": 0.01154638,
      "auxiliary_loss_mlp": 0.00811498,
      "balance_loss_clip": 1.05016589,
      "balance_loss_mlp": 1.03046119,
      "epoch": 0.1750789117691267,
      "flos": 17931203902080.0,
      "grad_norm": 2.334245540448843,
      "language_loss": 0.90028405,
      "learning_rate": 3.7832402587311764e-06,
      "loss": 0.91994542,
      "num_input_tokens_seen": 63001745,
      "step": 2912,
      "time_per_iteration": 2.4731333255767822
    },
    {
      "auxiliary_loss_clip": 0.01147029,
      "auxiliary_loss_mlp": 0.01046319,
      "balance_loss_clip": 1.05090904,
      "balance_loss_mlp": 1.02691126,
      "epoch": 0.17513903502179468,
      "flos": 18259427404800.0,
      "grad_norm": 1.8446631591636307,
      "language_loss": 0.72207189,
      "learning_rate": 3.783063882820439e-06,
      "loss": 0.74400538,
      "num_input_tokens_seen": 63019750,
      "step": 2913,
      "time_per_iteration": 2.4917943477630615
    },
    {
      "auxiliary_loss_clip": 0.01137521,
      "auxiliary_loss_mlp": 0.01047929,
      "balance_loss_clip": 1.05378938,
      "balance_loss_mlp": 1.02934456,
      "epoch": 0.17519915827446264,
      "flos": 20704728562560.0,
      "grad_norm": 1.9011188384433966,
      "language_loss": 0.69629109,
      "learning_rate": 3.782887439295741e-06,
      "loss": 0.71814555,
      "num_input_tokens_seen": 63039500,
      "step": 2914,
      "time_per_iteration": 2.5492985248565674
    },
    {
      "auxiliary_loss_clip": 0.01142344,
      "auxiliary_loss_mlp": 0.01047109,
      "balance_loss_clip": 1.05402493,
      "balance_loss_mlp": 1.02807164,
      "epoch": 0.1752592815271306,
      "flos": 20523415685760.0,
      "grad_norm": 1.8822013484975568,
      "language_loss": 0.93579727,
      "learning_rate": 3.782710928163772e-06,
      "loss": 0.95769185,
      "num_input_tokens_seen": 63059785,
      "step": 2915,
      "time_per_iteration": 2.5085201263427734
    },
    {
      "auxiliary_loss_clip": 0.01128061,
      "auxiliary_loss_mlp": 0.01044603,
      "balance_loss_clip": 1.06387877,
      "balance_loss_mlp": 1.02508819,
      "epoch": 0.1753194047797986,
      "flos": 21799178012160.0,
      "grad_norm": 1.7790503673452518,
      "language_loss": 0.80966485,
      "learning_rate": 3.782534349431226e-06,
      "loss": 0.83139145,
      "num_input_tokens_seen": 63079385,
      "step": 2916,
      "time_per_iteration": 2.5945425033569336
    },
    {
      "auxiliary_loss_clip": 0.01148648,
      "auxiliary_loss_mlp": 0.0105452,
      "balance_loss_clip": 1.0540539,
      "balance_loss_mlp": 1.03527927,
      "epoch": 0.17537952803246656,
      "flos": 20668351063680.0,
      "grad_norm": 1.6225705648262605,
      "language_loss": 0.73973149,
      "learning_rate": 3.782357703104799e-06,
      "loss": 0.76176322,
      "num_input_tokens_seen": 63098970,
      "step": 2917,
      "time_per_iteration": 2.5192577838897705
    },
    {
      "auxiliary_loss_clip": 0.01133818,
      "auxiliary_loss_mlp": 0.01054567,
      "balance_loss_clip": 1.05175865,
      "balance_loss_mlp": 1.03508854,
      "epoch": 0.17543965128513453,
      "flos": 23295072839040.0,
      "grad_norm": 1.8975282640332778,
      "language_loss": 0.77469969,
      "learning_rate": 3.7821809891911897e-06,
      "loss": 0.79658353,
      "num_input_tokens_seen": 63118750,
      "step": 2918,
      "time_per_iteration": 2.5319983959198
    },
    {
      "auxiliary_loss_clip": 0.01097203,
      "auxiliary_loss_mlp": 0.01051678,
      "balance_loss_clip": 1.0494715,
      "balance_loss_mlp": 1.02964783,
      "epoch": 0.1754997745378025,
      "flos": 29095615416960.0,
      "grad_norm": 3.4502126401328503,
      "language_loss": 0.73827893,
      "learning_rate": 3.782004207697098e-06,
      "loss": 0.75976777,
      "num_input_tokens_seen": 63136865,
      "step": 2919,
      "time_per_iteration": 2.6653575897216797
    },
    {
      "auxiliary_loss_clip": 0.01127333,
      "auxiliary_loss_mlp": 0.01051635,
      "balance_loss_clip": 1.04983449,
      "balance_loss_mlp": 1.03223991,
      "epoch": 0.17555989779047046,
      "flos": 30371844620160.0,
      "grad_norm": 1.7893724690115456,
      "language_loss": 0.74213469,
      "learning_rate": 3.781827358629228e-06,
      "loss": 0.76392436,
      "num_input_tokens_seen": 63158325,
      "step": 2920,
      "time_per_iteration": 2.61352801322937
    },
    {
      "auxiliary_loss_clip": 0.01119797,
      "auxiliary_loss_mlp": 0.01051027,
      "balance_loss_clip": 1.04715121,
      "balance_loss_mlp": 1.03090453,
      "epoch": 0.17562002104313842,
      "flos": 23287746464640.0,
      "grad_norm": 2.616717268931314,
      "language_loss": 0.79901445,
      "learning_rate": 3.7816504419942873e-06,
      "loss": 0.8207227,
      "num_input_tokens_seen": 63173115,
      "step": 2921,
      "time_per_iteration": 2.5060837268829346
    },
    {
      "auxiliary_loss_clip": 0.01124643,
      "auxiliary_loss_mlp": 0.01050056,
      "balance_loss_clip": 1.05328786,
      "balance_loss_mlp": 1.03012466,
      "epoch": 0.1756801442958064,
      "flos": 24790500789120.0,
      "grad_norm": 1.76945238221899,
      "language_loss": 0.87558311,
      "learning_rate": 3.7814734577989823e-06,
      "loss": 0.89733005,
      "num_input_tokens_seen": 63192880,
      "step": 2922,
      "time_per_iteration": 2.594142198562622
    },
    {
      "auxiliary_loss_clip": 0.01148048,
      "auxiliary_loss_mlp": 0.01049887,
      "balance_loss_clip": 1.05295837,
      "balance_loss_mlp": 1.03105187,
      "epoch": 0.17574026754847438,
      "flos": 25771651764480.0,
      "grad_norm": 2.5324387990587587,
      "language_loss": 0.62315881,
      "learning_rate": 3.7812964060500253e-06,
      "loss": 0.6451382,
      "num_input_tokens_seen": 63214395,
      "step": 2923,
      "time_per_iteration": 2.5555741786956787
    },
    {
      "auxiliary_loss_clip": 0.01129741,
      "auxiliary_loss_mlp": 0.01046968,
      "balance_loss_clip": 1.05321574,
      "balance_loss_mlp": 1.02620149,
      "epoch": 0.17580039080114235,
      "flos": 17456608477440.0,
      "grad_norm": 2.220825986283819,
      "language_loss": 0.80323148,
      "learning_rate": 3.78111928675413e-06,
      "loss": 0.82499862,
      "num_input_tokens_seen": 63231020,
      "step": 2924,
      "time_per_iteration": 2.5847175121307373
    },
    {
      "auxiliary_loss_clip": 0.0113489,
      "auxiliary_loss_mlp": 0.01054956,
      "balance_loss_clip": 1.05398726,
      "balance_loss_mlp": 1.03416562,
      "epoch": 0.1758605140538103,
      "flos": 14864648088960.0,
      "grad_norm": 2.222199879307277,
      "language_loss": 0.71246302,
      "learning_rate": 3.7809420999180126e-06,
      "loss": 0.73436141,
      "num_input_tokens_seen": 63246245,
      "step": 2925,
      "time_per_iteration": 2.4941041469573975
    },
    {
      "auxiliary_loss_clip": 0.01120903,
      "auxiliary_loss_mlp": 0.01043914,
      "balance_loss_clip": 1.05313802,
      "balance_loss_mlp": 1.02537644,
      "epoch": 0.17592063730647828,
      "flos": 23004268329600.0,
      "grad_norm": 1.6268663801303818,
      "language_loss": 0.71947503,
      "learning_rate": 3.7807648455483934e-06,
      "loss": 0.74112314,
      "num_input_tokens_seen": 63267790,
      "step": 2926,
      "time_per_iteration": 2.6124072074890137
    },
    {
      "auxiliary_loss_clip": 0.01104773,
      "auxiliary_loss_mlp": 0.01051932,
      "balance_loss_clip": 1.0483458,
      "balance_loss_mlp": 1.02829254,
      "epoch": 0.17598076055914624,
      "flos": 20741501111040.0,
      "grad_norm": 2.0881672192192777,
      "language_loss": 0.84935397,
      "learning_rate": 3.7805875236519918e-06,
      "loss": 0.87092102,
      "num_input_tokens_seen": 63286830,
      "step": 2927,
      "time_per_iteration": 2.553095579147339
    },
    {
      "auxiliary_loss_clip": 0.0111044,
      "auxiliary_loss_mlp": 0.01051211,
      "balance_loss_clip": 1.06372523,
      "balance_loss_mlp": 1.03390181,
      "epoch": 0.1760408838118142,
      "flos": 34092441227520.0,
      "grad_norm": 2.5710698712522313,
      "language_loss": 0.72354817,
      "learning_rate": 3.7804101342355336e-06,
      "loss": 0.74516469,
      "num_input_tokens_seen": 63308870,
      "step": 2928,
      "time_per_iteration": 5.548267364501953
    },
    {
      "auxiliary_loss_clip": 0.01119141,
      "auxiliary_loss_mlp": 0.01049737,
      "balance_loss_clip": 1.05032706,
      "balance_loss_mlp": 1.02990055,
      "epoch": 0.1761010070644822,
      "flos": 24168384207360.0,
      "grad_norm": 1.854679366371153,
      "language_loss": 0.8325026,
      "learning_rate": 3.780232677305744e-06,
      "loss": 0.85419136,
      "num_input_tokens_seen": 63329005,
      "step": 2929,
      "time_per_iteration": 2.597723960876465
    },
    {
      "auxiliary_loss_clip": 0.01131542,
      "auxiliary_loss_mlp": 0.01041608,
      "balance_loss_clip": 1.05742049,
      "balance_loss_mlp": 1.02254689,
      "epoch": 0.17616113031715017,
      "flos": 26576697335040.0,
      "grad_norm": 1.670080609555047,
      "language_loss": 0.79195368,
      "learning_rate": 3.7800551528693535e-06,
      "loss": 0.81368512,
      "num_input_tokens_seen": 63349390,
      "step": 2930,
      "time_per_iteration": 2.60609769821167
    },
    {
      "auxiliary_loss_clip": 0.01159387,
      "auxiliary_loss_mlp": 0.01046545,
      "balance_loss_clip": 1.05531502,
      "balance_loss_mlp": 1.02707839,
      "epoch": 0.17622125356981813,
      "flos": 25666685245440.0,
      "grad_norm": 2.1082321549569887,
      "language_loss": 0.76353794,
      "learning_rate": 3.7798775609330927e-06,
      "loss": 0.78559732,
      "num_input_tokens_seen": 63368835,
      "step": 2931,
      "time_per_iteration": 2.5234673023223877
    },
    {
      "auxiliary_loss_clip": 0.01074287,
      "auxiliary_loss_mlp": 0.01045357,
      "balance_loss_clip": 1.0438714,
      "balance_loss_mlp": 1.02653337,
      "epoch": 0.1762813768224861,
      "flos": 16508530949760.0,
      "grad_norm": 3.6679281437694073,
      "language_loss": 0.7504459,
      "learning_rate": 3.779699901503696e-06,
      "loss": 0.77164233,
      "num_input_tokens_seen": 63385220,
      "step": 2932,
      "time_per_iteration": 4.031884670257568
    },
    {
      "auxiliary_loss_clip": 0.01151382,
      "auxiliary_loss_mlp": 0.01042373,
      "balance_loss_clip": 1.0529952,
      "balance_loss_mlp": 1.02190518,
      "epoch": 0.17634150007515406,
      "flos": 11211850402560.0,
      "grad_norm": 2.8303994654671554,
      "language_loss": 0.89996243,
      "learning_rate": 3.7795221745879016e-06,
      "loss": 0.92189991,
      "num_input_tokens_seen": 63400865,
      "step": 2933,
      "time_per_iteration": 2.484219551086426
    },
    {
      "auxiliary_loss_clip": 0.0115756,
      "auxiliary_loss_mlp": 0.01055157,
      "balance_loss_clip": 1.05555522,
      "balance_loss_mlp": 1.03692961,
      "epoch": 0.17640162332782203,
      "flos": 23659925235840.0,
      "grad_norm": 1.7267112615777147,
      "language_loss": 0.8834486,
      "learning_rate": 3.779344380192448e-06,
      "loss": 0.90557581,
      "num_input_tokens_seen": 63421390,
      "step": 2934,
      "time_per_iteration": 2.5124008655548096
    },
    {
      "auxiliary_loss_clip": 0.01129148,
      "auxiliary_loss_mlp": 0.01048756,
      "balance_loss_clip": 1.05531383,
      "balance_loss_mlp": 1.03086305,
      "epoch": 0.17646174658049,
      "flos": 53796984606720.0,
      "grad_norm": 1.6936556109248069,
      "language_loss": 0.70558494,
      "learning_rate": 3.779166518324077e-06,
      "loss": 0.727364,
      "num_input_tokens_seen": 63444715,
      "step": 2935,
      "time_per_iteration": 4.226144313812256
    },
    {
      "auxiliary_loss_clip": 0.01125911,
      "auxiliary_loss_mlp": 0.01039227,
      "balance_loss_clip": 1.05263793,
      "balance_loss_mlp": 1.02039206,
      "epoch": 0.17652186983315798,
      "flos": 24243868638720.0,
      "grad_norm": 1.9699932267242088,
      "language_loss": 0.69386798,
      "learning_rate": 3.7789885889895325e-06,
      "loss": 0.71551931,
      "num_input_tokens_seen": 63465525,
      "step": 2936,
      "time_per_iteration": 2.573772430419922
    },
    {
      "auxiliary_loss_clip": 0.01108151,
      "auxiliary_loss_mlp": 0.01045841,
      "balance_loss_clip": 1.05289817,
      "balance_loss_mlp": 1.02744734,
      "epoch": 0.17658199308582595,
      "flos": 27454282421760.0,
      "grad_norm": 1.9356638839950495,
      "language_loss": 0.71262199,
      "learning_rate": 3.7788105921955634e-06,
      "loss": 0.73416191,
      "num_input_tokens_seen": 63485815,
      "step": 2937,
      "time_per_iteration": 2.6495559215545654
    },
    {
      "auxiliary_loss_clip": 0.01139355,
      "auxiliary_loss_mlp": 0.01041233,
      "balance_loss_clip": 1.05821943,
      "balance_loss_mlp": 1.02167094,
      "epoch": 0.17664211633849392,
      "flos": 22418672901120.0,
      "grad_norm": 2.858890818710361,
      "language_loss": 0.75788647,
      "learning_rate": 3.7786325279489184e-06,
      "loss": 0.77969241,
      "num_input_tokens_seen": 63503905,
      "step": 2938,
      "time_per_iteration": 2.5561225414276123
    },
    {
      "auxiliary_loss_clip": 0.01150566,
      "auxiliary_loss_mlp": 0.01039174,
      "balance_loss_clip": 1.05687737,
      "balance_loss_mlp": 1.02119684,
      "epoch": 0.17670223959116188,
      "flos": 24715124098560.0,
      "grad_norm": 2.19156361058105,
      "language_loss": 0.7131179,
      "learning_rate": 3.7784543962563495e-06,
      "loss": 0.73501527,
      "num_input_tokens_seen": 63521985,
      "step": 2939,
      "time_per_iteration": 2.530850410461426
    },
    {
      "auxiliary_loss_clip": 0.01161116,
      "auxiliary_loss_mlp": 0.01044606,
      "balance_loss_clip": 1.05687857,
      "balance_loss_mlp": 1.02559209,
      "epoch": 0.17676236284382985,
      "flos": 22527051212160.0,
      "grad_norm": 2.4514631158252187,
      "language_loss": 0.73399305,
      "learning_rate": 3.7782761971246115e-06,
      "loss": 0.75605023,
      "num_input_tokens_seen": 63539830,
      "step": 2940,
      "time_per_iteration": 2.4759058952331543
    },
    {
      "auxiliary_loss_clip": 0.011243,
      "auxiliary_loss_mlp": 0.01048859,
      "balance_loss_clip": 1.05572355,
      "balance_loss_mlp": 1.02896345,
      "epoch": 0.1768224860964978,
      "flos": 12385160161920.0,
      "grad_norm": 2.388246139232071,
      "language_loss": 0.854873,
      "learning_rate": 3.7780979305604616e-06,
      "loss": 0.87660456,
      "num_input_tokens_seen": 63555495,
      "step": 2941,
      "time_per_iteration": 2.543886184692383
    },
    {
      "auxiliary_loss_clip": 0.01159987,
      "auxiliary_loss_mlp": 0.01040707,
      "balance_loss_clip": 1.05449998,
      "balance_loss_mlp": 1.02176499,
      "epoch": 0.1768826093491658,
      "flos": 24353360271360.0,
      "grad_norm": 2.3282555167707026,
      "language_loss": 0.7657854,
      "learning_rate": 3.7779195965706607e-06,
      "loss": 0.78779238,
      "num_input_tokens_seen": 63575290,
      "step": 2942,
      "time_per_iteration": 2.514430522918701
    },
    {
      "auxiliary_loss_clip": 0.01101157,
      "auxiliary_loss_mlp": 0.00823422,
      "balance_loss_clip": 1.04736876,
      "balance_loss_mlp": 1.05175447,
      "epoch": 0.17694273260183377,
      "flos": 23587062497280.0,
      "grad_norm": 1.7995989176202907,
      "language_loss": 0.80293238,
      "learning_rate": 3.77774119516197e-06,
      "loss": 0.82217824,
      "num_input_tokens_seen": 63594670,
      "step": 2943,
      "time_per_iteration": 2.679992437362671
    },
    {
      "auxiliary_loss_clip": 0.01131252,
      "auxiliary_loss_mlp": 0.01050522,
      "balance_loss_clip": 1.05131817,
      "balance_loss_mlp": 1.0295651,
      "epoch": 0.17700285585450173,
      "flos": 26760991040640.0,
      "grad_norm": 1.980561818627125,
      "language_loss": 0.80788219,
      "learning_rate": 3.777562726341155e-06,
      "loss": 0.82969999,
      "num_input_tokens_seen": 63614780,
      "step": 2944,
      "time_per_iteration": 2.5805845260620117
    },
    {
      "auxiliary_loss_clip": 0.01159749,
      "auxiliary_loss_mlp": 0.01056431,
      "balance_loss_clip": 1.05233479,
      "balance_loss_mlp": 1.03763199,
      "epoch": 0.1770629791071697,
      "flos": 42776323320960.0,
      "grad_norm": 2.1220926651857726,
      "language_loss": 0.74408811,
      "learning_rate": 3.7773841901149835e-06,
      "loss": 0.76624995,
      "num_input_tokens_seen": 63637190,
      "step": 2945,
      "time_per_iteration": 2.656189203262329
    },
    {
      "auxiliary_loss_clip": 0.01146003,
      "auxiliary_loss_mlp": 0.01044768,
      "balance_loss_clip": 1.05539978,
      "balance_loss_mlp": 1.02602863,
      "epoch": 0.17712310235983766,
      "flos": 17345572560000.0,
      "grad_norm": 2.495924806849644,
      "language_loss": 0.78333086,
      "learning_rate": 3.7772055864902256e-06,
      "loss": 0.80523854,
      "num_input_tokens_seen": 63652140,
      "step": 2946,
      "time_per_iteration": 2.44804310798645
    },
    {
      "auxiliary_loss_clip": 0.01109647,
      "auxiliary_loss_mlp": 0.01052108,
      "balance_loss_clip": 1.05279922,
      "balance_loss_mlp": 1.03246212,
      "epoch": 0.17718322561250563,
      "flos": 23878477537920.0,
      "grad_norm": 1.695759742234261,
      "language_loss": 0.76017606,
      "learning_rate": 3.7770269154736535e-06,
      "loss": 0.78179359,
      "num_input_tokens_seen": 63671700,
      "step": 2947,
      "time_per_iteration": 2.5979435443878174
    },
    {
      "auxiliary_loss_clip": 0.01146334,
      "auxiliary_loss_mlp": 0.01044093,
      "balance_loss_clip": 1.05201852,
      "balance_loss_mlp": 1.02438784,
      "epoch": 0.1772433488651736,
      "flos": 36466352104320.0,
      "grad_norm": 2.181650091116727,
      "language_loss": 0.72913307,
      "learning_rate": 3.7768481770720424e-06,
      "loss": 0.75103736,
      "num_input_tokens_seen": 63691685,
      "step": 2948,
      "time_per_iteration": 2.6115617752075195
    },
    {
      "auxiliary_loss_clip": 0.01158297,
      "auxiliary_loss_mlp": 0.01044127,
      "balance_loss_clip": 1.06611979,
      "balance_loss_mlp": 1.02545917,
      "epoch": 0.1773034721178416,
      "flos": 26684716510080.0,
      "grad_norm": 1.8233519653521333,
      "language_loss": 0.82257777,
      "learning_rate": 3.776669371292171e-06,
      "loss": 0.84460199,
      "num_input_tokens_seen": 63711720,
      "step": 2949,
      "time_per_iteration": 2.5919477939605713
    },
    {
      "auxiliary_loss_clip": 0.01077799,
      "auxiliary_loss_mlp": 0.01010389,
      "balance_loss_clip": 1.04600704,
      "balance_loss_mlp": 1.00719452,
      "epoch": 0.17736359537050955,
      "flos": 57117467617920.0,
      "grad_norm": 0.7529679864600998,
      "language_loss": 0.65026027,
      "learning_rate": 3.7764904981408186e-06,
      "loss": 0.67114216,
      "num_input_tokens_seen": 63776280,
      "step": 2950,
      "time_per_iteration": 3.190380573272705
    },
    {
      "auxiliary_loss_clip": 0.01120892,
      "auxiliary_loss_mlp": 0.01043549,
      "balance_loss_clip": 1.05293047,
      "balance_loss_mlp": 1.02476192,
      "epoch": 0.17742371862317752,
      "flos": 27198203385600.0,
      "grad_norm": 2.5714958596618422,
      "language_loss": 0.83865118,
      "learning_rate": 3.7763115576247686e-06,
      "loss": 0.86029553,
      "num_input_tokens_seen": 63797535,
      "step": 2951,
      "time_per_iteration": 2.6427111625671387
    },
    {
      "auxiliary_loss_clip": 0.01125666,
      "auxiliary_loss_mlp": 0.01044665,
      "balance_loss_clip": 1.05146229,
      "balance_loss_mlp": 1.02486384,
      "epoch": 0.17748384187584548,
      "flos": 20959694277120.0,
      "grad_norm": 2.5105126934937716,
      "language_loss": 0.80213618,
      "learning_rate": 3.776132549750806e-06,
      "loss": 0.82383943,
      "num_input_tokens_seen": 63817045,
      "step": 2952,
      "time_per_iteration": 2.5700857639312744
    },
    {
      "auxiliary_loss_clip": 0.01160397,
      "auxiliary_loss_mlp": 0.01048424,
      "balance_loss_clip": 1.05545664,
      "balance_loss_mlp": 1.02832568,
      "epoch": 0.17754396512851345,
      "flos": 25009986844800.0,
      "grad_norm": 2.192834126458989,
      "language_loss": 0.79649723,
      "learning_rate": 3.7759534745257194e-06,
      "loss": 0.8185854,
      "num_input_tokens_seen": 63837665,
      "step": 2953,
      "time_per_iteration": 2.5272061824798584
    },
    {
      "auxiliary_loss_clip": 0.01120993,
      "auxiliary_loss_mlp": 0.01043632,
      "balance_loss_clip": 1.05447936,
      "balance_loss_mlp": 1.0244509,
      "epoch": 0.1776040883811814,
      "flos": 32051566275840.0,
      "grad_norm": 1.8340359156995647,
      "language_loss": 0.8778832,
      "learning_rate": 3.7757743319562994e-06,
      "loss": 0.8995294,
      "num_input_tokens_seen": 63858455,
      "step": 2954,
      "time_per_iteration": 2.680642604827881
    },
    {
      "auxiliary_loss_clip": 0.0113957,
      "auxiliary_loss_mlp": 0.01052792,
      "balance_loss_clip": 1.05514598,
      "balance_loss_mlp": 1.0339092,
      "epoch": 0.17766421163384938,
      "flos": 21574125348480.0,
      "grad_norm": 2.103274935313538,
      "language_loss": 0.84631824,
      "learning_rate": 3.7755951220493386e-06,
      "loss": 0.86824191,
      "num_input_tokens_seen": 63876935,
      "step": 2955,
      "time_per_iteration": 2.563159465789795
    },
    {
      "auxiliary_loss_clip": 0.01130116,
      "auxiliary_loss_mlp": 0.01050814,
      "balance_loss_clip": 1.05283022,
      "balance_loss_mlp": 1.03057241,
      "epoch": 0.17772433488651737,
      "flos": 22419319345920.0,
      "grad_norm": 1.7264236232643893,
      "language_loss": 0.70892906,
      "learning_rate": 3.7754158448116327e-06,
      "loss": 0.7307384,
      "num_input_tokens_seen": 63896815,
      "step": 2956,
      "time_per_iteration": 2.5730204582214355
    },
    {
      "auxiliary_loss_clip": 0.01149784,
      "auxiliary_loss_mlp": 0.01053103,
      "balance_loss_clip": 1.05607748,
      "balance_loss_mlp": 1.03377891,
      "epoch": 0.17778445813918534,
      "flos": 25629445820160.0,
      "grad_norm": 1.8600708881288932,
      "language_loss": 0.83256441,
      "learning_rate": 3.7752365002499795e-06,
      "loss": 0.85459328,
      "num_input_tokens_seen": 63916140,
      "step": 2957,
      "time_per_iteration": 2.5378241539001465
    },
    {
      "auxiliary_loss_clip": 0.01101997,
      "auxiliary_loss_mlp": 0.01044855,
      "balance_loss_clip": 1.05079389,
      "balance_loss_mlp": 1.02498317,
      "epoch": 0.1778445813918533,
      "flos": 25628871202560.0,
      "grad_norm": 1.6235078718362561,
      "language_loss": 0.75598133,
      "learning_rate": 3.7750570883711807e-06,
      "loss": 0.77744991,
      "num_input_tokens_seen": 63935220,
      "step": 2958,
      "time_per_iteration": 2.655172348022461
    },
    {
      "auxiliary_loss_clip": 0.01147604,
      "auxiliary_loss_mlp": 0.01048053,
      "balance_loss_clip": 1.05839539,
      "balance_loss_mlp": 1.02917051,
      "epoch": 0.17790470464452127,
      "flos": 22345522853760.0,
      "grad_norm": 5.289409080568085,
      "language_loss": 0.80405802,
      "learning_rate": 3.7748776091820397e-06,
      "loss": 0.82601458,
      "num_input_tokens_seen": 63954550,
      "step": 2959,
      "time_per_iteration": 2.5301895141601562
    },
    {
      "auxiliary_loss_clip": 0.011677,
      "auxiliary_loss_mlp": 0.01051278,
      "balance_loss_clip": 1.05932474,
      "balance_loss_mlp": 1.03131032,
      "epoch": 0.17796482789718923,
      "flos": 18765875214720.0,
      "grad_norm": 2.041852633684074,
      "language_loss": 0.52284062,
      "learning_rate": 3.774698062689362e-06,
      "loss": 0.54503036,
      "num_input_tokens_seen": 63972425,
      "step": 2960,
      "time_per_iteration": 2.4874885082244873
    },
    {
      "auxiliary_loss_clip": 0.0111395,
      "auxiliary_loss_mlp": 0.01062552,
      "balance_loss_clip": 1.05589724,
      "balance_loss_mlp": 1.04159474,
      "epoch": 0.1780249511498572,
      "flos": 23440941970560.0,
      "grad_norm": 1.9261056559183216,
      "language_loss": 0.89137936,
      "learning_rate": 3.7745184488999548e-06,
      "loss": 0.91314441,
      "num_input_tokens_seen": 63992165,
      "step": 2961,
      "time_per_iteration": 2.6163876056671143
    },
    {
      "auxiliary_loss_clip": 0.01115183,
      "auxiliary_loss_mlp": 0.01051772,
      "balance_loss_clip": 1.05429649,
      "balance_loss_mlp": 1.03178072,
      "epoch": 0.1780850744025252,
      "flos": 23367468700800.0,
      "grad_norm": 1.6345170373938518,
      "language_loss": 0.79134417,
      "learning_rate": 3.774338767820631e-06,
      "loss": 0.81301367,
      "num_input_tokens_seen": 64013470,
      "step": 2962,
      "time_per_iteration": 2.6353790760040283
    },
    {
      "auxiliary_loss_clip": 0.01145812,
      "auxiliary_loss_mlp": 0.01055773,
      "balance_loss_clip": 1.05503273,
      "balance_loss_mlp": 1.03351617,
      "epoch": 0.17814519765519315,
      "flos": 13771994319360.0,
      "grad_norm": 1.666630943113837,
      "language_loss": 0.74846959,
      "learning_rate": 3.774159019458203e-06,
      "loss": 0.77048546,
      "num_input_tokens_seen": 64030975,
      "step": 2963,
      "time_per_iteration": 2.494568347930908
    },
    {
      "auxiliary_loss_clip": 0.01142043,
      "auxiliary_loss_mlp": 0.01046798,
      "balance_loss_clip": 1.05823302,
      "balance_loss_mlp": 1.02668691,
      "epoch": 0.17820532090786112,
      "flos": 21976396738560.0,
      "grad_norm": 1.815175761070858,
      "language_loss": 0.78628916,
      "learning_rate": 3.7739792038194877e-06,
      "loss": 0.80817759,
      "num_input_tokens_seen": 64050075,
      "step": 2964,
      "time_per_iteration": 2.5668118000030518
    },
    {
      "auxiliary_loss_clip": 0.01153507,
      "auxiliary_loss_mlp": 0.00810305,
      "balance_loss_clip": 1.05846453,
      "balance_loss_mlp": 1.03116012,
      "epoch": 0.17826544416052909,
      "flos": 24790752184320.0,
      "grad_norm": 2.0596756142120025,
      "language_loss": 0.81510031,
      "learning_rate": 3.7737993209113027e-06,
      "loss": 0.83473849,
      "num_input_tokens_seen": 64071920,
      "step": 2965,
      "time_per_iteration": 2.539760112762451
    },
    {
      "auxiliary_loss_clip": 0.01148169,
      "auxiliary_loss_mlp": 0.01043922,
      "balance_loss_clip": 1.05505729,
      "balance_loss_mlp": 1.02627921,
      "epoch": 0.17832556741319705,
      "flos": 13879582531200.0,
      "grad_norm": 2.2357623177728456,
      "language_loss": 0.94421065,
      "learning_rate": 3.7736193707404698e-06,
      "loss": 0.96613163,
      "num_input_tokens_seen": 64086835,
      "step": 2966,
      "time_per_iteration": 2.463391065597534
    },
    {
      "auxiliary_loss_clip": 0.01116936,
      "auxiliary_loss_mlp": 0.00807526,
      "balance_loss_clip": 1.05628777,
      "balance_loss_mlp": 1.0272001,
      "epoch": 0.17838569066586502,
      "flos": 36641703323520.0,
      "grad_norm": 2.2411834757480515,
      "language_loss": 0.730708,
      "learning_rate": 3.7734393533138127e-06,
      "loss": 0.74995267,
      "num_input_tokens_seen": 64107360,
      "step": 2967,
      "time_per_iteration": 4.2493064403533936
    },
    {
      "auxiliary_loss_clip": 0.01130226,
      "auxiliary_loss_mlp": 0.01050694,
      "balance_loss_clip": 1.05512106,
      "balance_loss_mlp": 1.0313468,
      "epoch": 0.17844581391853298,
      "flos": 18727271072640.0,
      "grad_norm": 1.9294636075824487,
      "language_loss": 0.7694909,
      "learning_rate": 3.773259268638157e-06,
      "loss": 0.79130006,
      "num_input_tokens_seen": 64124690,
      "step": 2968,
      "time_per_iteration": 3.89522647857666
    },
    {
      "auxiliary_loss_clip": 0.01086208,
      "auxiliary_loss_mlp": 0.01046969,
      "balance_loss_clip": 1.049227,
      "balance_loss_mlp": 1.0275141,
      "epoch": 0.17850593717120097,
      "flos": 27378259286400.0,
      "grad_norm": 1.9301122986116048,
      "language_loss": 0.75794005,
      "learning_rate": 3.7730791167203333e-06,
      "loss": 0.77927172,
      "num_input_tokens_seen": 64146315,
      "step": 2969,
      "time_per_iteration": 2.6725833415985107
    },
    {
      "auxiliary_loss_clip": 0.01052277,
      "auxiliary_loss_mlp": 0.01035156,
      "balance_loss_clip": 1.03901553,
      "balance_loss_mlp": 1.03241432,
      "epoch": 0.17856606042386894,
      "flos": 66996025084800.0,
      "grad_norm": 0.8489364856842697,
      "language_loss": 0.69037056,
      "learning_rate": 3.772898897567171e-06,
      "loss": 0.71124494,
      "num_input_tokens_seen": 64210875,
      "step": 2970,
      "time_per_iteration": 3.2114129066467285
    },
    {
      "auxiliary_loss_clip": 0.01127135,
      "auxiliary_loss_mlp": 0.01043091,
      "balance_loss_clip": 1.05182469,
      "balance_loss_mlp": 1.02340984,
      "epoch": 0.1786261836765369,
      "flos": 36977001805440.0,
      "grad_norm": 1.9842444486096198,
      "language_loss": 0.67778349,
      "learning_rate": 3.772718611185505e-06,
      "loss": 0.69948578,
      "num_input_tokens_seen": 64230740,
      "step": 2971,
      "time_per_iteration": 4.137200593948364
    },
    {
      "auxiliary_loss_clip": 0.01117033,
      "auxiliary_loss_mlp": 0.0104903,
      "balance_loss_clip": 1.06215239,
      "balance_loss_mlp": 1.02837086,
      "epoch": 0.17868630692920487,
      "flos": 24825441744000.0,
      "grad_norm": 1.7883837237609734,
      "language_loss": 0.89930332,
      "learning_rate": 3.7725382575821717e-06,
      "loss": 0.92096394,
      "num_input_tokens_seen": 64252300,
      "step": 2972,
      "time_per_iteration": 2.642200469970703
    },
    {
      "auxiliary_loss_clip": 0.01119764,
      "auxiliary_loss_mlp": 0.01063871,
      "balance_loss_clip": 1.05089903,
      "balance_loss_mlp": 1.04244876,
      "epoch": 0.17874643018187283,
      "flos": 16981977139200.0,
      "grad_norm": 2.422009188626755,
      "language_loss": 0.88810194,
      "learning_rate": 3.77235783676401e-06,
      "loss": 0.90993828,
      "num_input_tokens_seen": 64270105,
      "step": 2973,
      "time_per_iteration": 3.967153787612915
    },
    {
      "auxiliary_loss_clip": 0.01159211,
      "auxiliary_loss_mlp": 0.01056544,
      "balance_loss_clip": 1.05417514,
      "balance_loss_mlp": 1.03713655,
      "epoch": 0.1788065534345408,
      "flos": 21032233793280.0,
      "grad_norm": 2.600292413082903,
      "language_loss": 0.76018912,
      "learning_rate": 3.7721773487378615e-06,
      "loss": 0.78234673,
      "num_input_tokens_seen": 64287250,
      "step": 2974,
      "time_per_iteration": 2.476093053817749
    },
    {
      "auxiliary_loss_clip": 0.01137264,
      "auxiliary_loss_mlp": 0.01049182,
      "balance_loss_clip": 1.05524969,
      "balance_loss_mlp": 1.03053725,
      "epoch": 0.17886667668720876,
      "flos": 23987717775360.0,
      "grad_norm": 2.942190638923503,
      "language_loss": 0.7417407,
      "learning_rate": 3.7719967935105705e-06,
      "loss": 0.76360524,
      "num_input_tokens_seen": 64307140,
      "step": 2975,
      "time_per_iteration": 2.5626392364501953
    },
    {
      "auxiliary_loss_clip": 0.01146186,
      "auxiliary_loss_mlp": 0.01056344,
      "balance_loss_clip": 1.05167723,
      "balance_loss_mlp": 1.03784227,
      "epoch": 0.17892679993987676,
      "flos": 25739476156800.0,
      "grad_norm": 1.7722499725876983,
      "language_loss": 0.73453283,
      "learning_rate": 3.7718161710889833e-06,
      "loss": 0.75655812,
      "num_input_tokens_seen": 64328760,
      "step": 2976,
      "time_per_iteration": 2.52687931060791
    },
    {
      "auxiliary_loss_clip": 0.01142246,
      "auxiliary_loss_mlp": 0.01057605,
      "balance_loss_clip": 1.05069447,
      "balance_loss_mlp": 1.04156566,
      "epoch": 0.17898692319254472,
      "flos": 25699686865920.0,
      "grad_norm": 1.6049410023588304,
      "language_loss": 0.77360809,
      "learning_rate": 3.7716354814799495e-06,
      "loss": 0.79560661,
      "num_input_tokens_seen": 64348800,
      "step": 2977,
      "time_per_iteration": 2.538892984390259
    },
    {
      "auxiliary_loss_clip": 0.01131891,
      "auxiliary_loss_mlp": 0.01057257,
      "balance_loss_clip": 1.0691061,
      "balance_loss_mlp": 1.039101,
      "epoch": 0.1790470464452127,
      "flos": 19317786664320.0,
      "grad_norm": 2.002933069468797,
      "language_loss": 0.80088222,
      "learning_rate": 3.7714547246903203e-06,
      "loss": 0.82277369,
      "num_input_tokens_seen": 64367955,
      "step": 2978,
      "time_per_iteration": 2.5949902534484863
    },
    {
      "auxiliary_loss_clip": 0.01139165,
      "auxiliary_loss_mlp": 0.0105252,
      "balance_loss_clip": 1.05300617,
      "balance_loss_mlp": 1.03292179,
      "epoch": 0.17910716969788065,
      "flos": 30044267562240.0,
      "grad_norm": 1.45521750891281,
      "language_loss": 0.7597605,
      "learning_rate": 3.7712739007269508e-06,
      "loss": 0.78167737,
      "num_input_tokens_seen": 64389805,
      "step": 2979,
      "time_per_iteration": 2.618556261062622
    },
    {
      "auxiliary_loss_clip": 0.01116271,
      "auxiliary_loss_mlp": 0.01053324,
      "balance_loss_clip": 1.05147791,
      "balance_loss_mlp": 1.03506088,
      "epoch": 0.17916729295054862,
      "flos": 19427709260160.0,
      "grad_norm": 1.74307087557171,
      "language_loss": 0.69252264,
      "learning_rate": 3.7710930095966976e-06,
      "loss": 0.71421856,
      "num_input_tokens_seen": 64408220,
      "step": 2980,
      "time_per_iteration": 2.5602591037750244
    },
    {
      "auxiliary_loss_clip": 0.0114631,
      "auxiliary_loss_mlp": 0.01052111,
      "balance_loss_clip": 1.05086327,
      "balance_loss_mlp": 1.03139281,
      "epoch": 0.17922741620321658,
      "flos": 14611549881600.0,
      "grad_norm": 1.9401448111318798,
      "language_loss": 0.71133256,
      "learning_rate": 3.7709120513064196e-06,
      "loss": 0.73331678,
      "num_input_tokens_seen": 64426380,
      "step": 2981,
      "time_per_iteration": 2.47927188873291
    },
    {
      "auxiliary_loss_clip": 0.01127728,
      "auxiliary_loss_mlp": 0.01060073,
      "balance_loss_clip": 1.05087996,
      "balance_loss_mlp": 1.0408926,
      "epoch": 0.17928753945588458,
      "flos": 17165301177600.0,
      "grad_norm": 2.1332203709780524,
      "language_loss": 0.82051134,
      "learning_rate": 3.7707310258629796e-06,
      "loss": 0.8423894,
      "num_input_tokens_seen": 64444355,
      "step": 2982,
      "time_per_iteration": 2.491520404815674
    },
    {
      "auxiliary_loss_clip": 0.01156488,
      "auxiliary_loss_mlp": 0.01048009,
      "balance_loss_clip": 1.05254304,
      "balance_loss_mlp": 1.03012717,
      "epoch": 0.17934766270855254,
      "flos": 31395622060800.0,
      "grad_norm": 1.4991512572639922,
      "language_loss": 0.82984984,
      "learning_rate": 3.7705499332732413e-06,
      "loss": 0.8518948,
      "num_input_tokens_seen": 64467800,
      "step": 2983,
      "time_per_iteration": 2.544684410095215
    },
    {
      "auxiliary_loss_clip": 0.01148823,
      "auxiliary_loss_mlp": 0.01055903,
      "balance_loss_clip": 1.04920626,
      "balance_loss_mlp": 1.03657937,
      "epoch": 0.1794077859612205,
      "flos": 20814184281600.0,
      "grad_norm": 2.0758128034831254,
      "language_loss": 0.85186774,
      "learning_rate": 3.7703687735440718e-06,
      "loss": 0.87391508,
      "num_input_tokens_seen": 64487230,
      "step": 2984,
      "time_per_iteration": 2.484694719314575
    },
    {
      "auxiliary_loss_clip": 0.01122614,
      "auxiliary_loss_mlp": 0.01048255,
      "balance_loss_clip": 1.04867697,
      "balance_loss_mlp": 1.02881169,
      "epoch": 0.17946790921388847,
      "flos": 28986447006720.0,
      "grad_norm": 1.6059347019433265,
      "language_loss": 0.89364874,
      "learning_rate": 3.7701875466823416e-06,
      "loss": 0.91535747,
      "num_input_tokens_seen": 64509165,
      "step": 2985,
      "time_per_iteration": 2.634368658065796
    },
    {
      "auxiliary_loss_clip": 0.0115297,
      "auxiliary_loss_mlp": 0.01048757,
      "balance_loss_clip": 1.05143094,
      "balance_loss_mlp": 1.03209198,
      "epoch": 0.17952803246655644,
      "flos": 20737406960640.0,
      "grad_norm": 2.1165325995121353,
      "language_loss": 0.6966691,
      "learning_rate": 3.770006252694922e-06,
      "loss": 0.7186864,
      "num_input_tokens_seen": 64527940,
      "step": 2986,
      "time_per_iteration": 2.4444000720977783
    },
    {
      "auxiliary_loss_clip": 0.01152276,
      "auxiliary_loss_mlp": 0.0079674,
      "balance_loss_clip": 1.04967439,
      "balance_loss_mlp": 1.01363623,
      "epoch": 0.1795881557192244,
      "flos": 28255988027520.0,
      "grad_norm": 2.783359728227293,
      "language_loss": 0.77185631,
      "learning_rate": 3.769824891588688e-06,
      "loss": 0.79134643,
      "num_input_tokens_seen": 64545230,
      "step": 2987,
      "time_per_iteration": 2.52290940284729
    },
    {
      "auxiliary_loss_clip": 0.0115758,
      "auxiliary_loss_mlp": 0.0104509,
      "balance_loss_clip": 1.05029488,
      "balance_loss_mlp": 1.02518225,
      "epoch": 0.17964827897189237,
      "flos": 18552027594240.0,
      "grad_norm": 1.744209969047335,
      "language_loss": 0.78374046,
      "learning_rate": 3.7696434633705164e-06,
      "loss": 0.80576718,
      "num_input_tokens_seen": 64563820,
      "step": 2988,
      "time_per_iteration": 2.4398508071899414
    },
    {
      "auxiliary_loss_clip": 0.01016111,
      "auxiliary_loss_mlp": 0.01181222,
      "balance_loss_clip": 1.03070402,
      "balance_loss_mlp": 1.71153152,
      "epoch": 0.17970840222456036,
      "flos": 58165088711040.0,
      "grad_norm": 0.7704829214551564,
      "language_loss": 0.62785143,
      "learning_rate": 3.7694619680472875e-06,
      "loss": 0.64982474,
      "num_input_tokens_seen": 64621315,
      "step": 2989,
      "time_per_iteration": 3.1377387046813965
    },
    {
      "auxiliary_loss_clip": 0.01134246,
      "auxiliary_loss_mlp": 0.0104258,
      "balance_loss_clip": 1.05122733,
      "balance_loss_mlp": 1.02457988,
      "epoch": 0.17976852547722832,
      "flos": 20300805146880.0,
      "grad_norm": 3.4872757119252746,
      "language_loss": 0.70585704,
      "learning_rate": 3.7692804056258837e-06,
      "loss": 0.72762531,
      "num_input_tokens_seen": 64639885,
      "step": 2990,
      "time_per_iteration": 2.533043146133423
    },
    {
      "auxiliary_loss_clip": 0.01134341,
      "auxiliary_loss_mlp": 0.01043714,
      "balance_loss_clip": 1.05176306,
      "balance_loss_mlp": 1.0259515,
      "epoch": 0.1798286487298963,
      "flos": 39669367685760.0,
      "grad_norm": 1.9639711331875813,
      "language_loss": 0.68470728,
      "learning_rate": 3.7690987761131893e-06,
      "loss": 0.70648777,
      "num_input_tokens_seen": 64661220,
      "step": 2991,
      "time_per_iteration": 2.6862313747406006
    },
    {
      "auxiliary_loss_clip": 0.01096572,
      "auxiliary_loss_mlp": 0.01039946,
      "balance_loss_clip": 1.046386,
      "balance_loss_mlp": 1.02125394,
      "epoch": 0.17988877198256426,
      "flos": 25520313323520.0,
      "grad_norm": 1.462522516451752,
      "language_loss": 0.82618791,
      "learning_rate": 3.7689170795160924e-06,
      "loss": 0.84755307,
      "num_input_tokens_seen": 64682530,
      "step": 2992,
      "time_per_iteration": 2.6289451122283936
    },
    {
      "auxiliary_loss_clip": 0.01138257,
      "auxiliary_loss_mlp": 0.01042597,
      "balance_loss_clip": 1.04864097,
      "balance_loss_mlp": 1.02491856,
      "epoch": 0.17994889523523222,
      "flos": 18807496099200.0,
      "grad_norm": 1.9128477920315141,
      "language_loss": 0.82128233,
      "learning_rate": 3.7687353158414822e-06,
      "loss": 0.84309083,
      "num_input_tokens_seen": 64701025,
      "step": 2993,
      "time_per_iteration": 2.4856393337249756
    },
    {
      "auxiliary_loss_clip": 0.01132431,
      "auxiliary_loss_mlp": 0.01044049,
      "balance_loss_clip": 1.04718649,
      "balance_loss_mlp": 1.02513099,
      "epoch": 0.18000901848790019,
      "flos": 21104450087040.0,
      "grad_norm": 1.6130680162347455,
      "language_loss": 0.78369772,
      "learning_rate": 3.7685534850962517e-06,
      "loss": 0.80546248,
      "num_input_tokens_seen": 64719570,
      "step": 2994,
      "time_per_iteration": 2.5127711296081543
    },
    {
      "auxiliary_loss_clip": 0.01155384,
      "auxiliary_loss_mlp": 0.01042928,
      "balance_loss_clip": 1.05014777,
      "balance_loss_mlp": 1.02520192,
      "epoch": 0.18006914174056818,
      "flos": 19646441130240.0,
      "grad_norm": 2.3191833306215828,
      "language_loss": 0.8078959,
      "learning_rate": 3.768371587287296e-06,
      "loss": 0.82987905,
      "num_input_tokens_seen": 64738110,
      "step": 2995,
      "time_per_iteration": 2.451361894607544
    },
    {
      "auxiliary_loss_clip": 0.01142184,
      "auxiliary_loss_mlp": 0.01046727,
      "balance_loss_clip": 1.05161071,
      "balance_loss_mlp": 1.02946556,
      "epoch": 0.18012926499323614,
      "flos": 19499889640320.0,
      "grad_norm": 1.7660657225275864,
      "language_loss": 0.84471494,
      "learning_rate": 3.768189622421512e-06,
      "loss": 0.86660403,
      "num_input_tokens_seen": 64756345,
      "step": 2996,
      "time_per_iteration": 2.476933717727661
    },
    {
      "auxiliary_loss_clip": 0.01117264,
      "auxiliary_loss_mlp": 0.01041703,
      "balance_loss_clip": 1.05009878,
      "balance_loss_mlp": 1.02391648,
      "epoch": 0.1801893882459041,
      "flos": 19464553635840.0,
      "grad_norm": 1.6173462268177818,
      "language_loss": 0.8791225,
      "learning_rate": 3.7680075905058006e-06,
      "loss": 0.90071219,
      "num_input_tokens_seen": 64776375,
      "step": 2997,
      "time_per_iteration": 2.58276629447937
    },
    {
      "auxiliary_loss_clip": 0.01133796,
      "auxiliary_loss_mlp": 0.01045833,
      "balance_loss_clip": 1.04833353,
      "balance_loss_mlp": 1.02690291,
      "epoch": 0.18024951149857207,
      "flos": 26870590414080.0,
      "grad_norm": 4.081127300282766,
      "language_loss": 0.85438192,
      "learning_rate": 3.7678254915470643e-06,
      "loss": 0.87617826,
      "num_input_tokens_seen": 64796210,
      "step": 2998,
      "time_per_iteration": 2.550574541091919
    },
    {
      "auxiliary_loss_clip": 0.01156082,
      "auxiliary_loss_mlp": 0.01046001,
      "balance_loss_clip": 1.05402803,
      "balance_loss_mlp": 1.02864361,
      "epoch": 0.18030963475124004,
      "flos": 30226621933440.0,
      "grad_norm": 2.170713770723279,
      "language_loss": 0.84199572,
      "learning_rate": 3.7676433255522084e-06,
      "loss": 0.86401659,
      "num_input_tokens_seen": 64818590,
      "step": 2999,
      "time_per_iteration": 2.5431764125823975
    },
    {
      "auxiliary_loss_clip": 0.01141226,
      "auxiliary_loss_mlp": 0.01049772,
      "balance_loss_clip": 1.04998398,
      "balance_loss_mlp": 1.03043675,
      "epoch": 0.180369758003908,
      "flos": 22307493329280.0,
      "grad_norm": 1.7999630525523629,
      "language_loss": 0.74776196,
      "learning_rate": 3.76746109252814e-06,
      "loss": 0.76967198,
      "num_input_tokens_seen": 64838350,
      "step": 3000,
      "time_per_iteration": 2.498661994934082
    },
    {
      "auxiliary_loss_clip": 0.01129777,
      "auxiliary_loss_mlp": 0.00921389,
      "balance_loss_clip": 1.05021095,
      "balance_loss_mlp": 1.24534202,
      "epoch": 0.18042988125657597,
      "flos": 23732033788800.0,
      "grad_norm": 1.8750708883012825,
      "language_loss": 0.71183366,
      "learning_rate": 3.76727879248177e-06,
      "loss": 0.73234534,
      "num_input_tokens_seen": 64858065,
      "step": 3001,
      "time_per_iteration": 2.5668869018554688
    },
    {
      "auxiliary_loss_clip": 0.01145984,
      "auxiliary_loss_mlp": 0.01046731,
      "balance_loss_clip": 1.05052125,
      "balance_loss_mlp": 1.02805054,
      "epoch": 0.18049000450924396,
      "flos": 24093582134400.0,
      "grad_norm": 2.3021249917825073,
      "language_loss": 0.88340354,
      "learning_rate": 3.767096425420011e-06,
      "loss": 0.90533066,
      "num_input_tokens_seen": 64877305,
      "step": 3002,
      "time_per_iteration": 2.496974468231201
    },
    {
      "auxiliary_loss_clip": 0.01154538,
      "auxiliary_loss_mlp": 0.01046526,
      "balance_loss_clip": 1.05124021,
      "balance_loss_mlp": 1.02888346,
      "epoch": 0.18055012776191193,
      "flos": 22163168482560.0,
      "grad_norm": 1.7876322873087913,
      "language_loss": 0.80654389,
      "learning_rate": 3.7669139913497788e-06,
      "loss": 0.82855451,
      "num_input_tokens_seen": 64896955,
      "step": 3003,
      "time_per_iteration": 2.46458101272583
    },
    {
      "auxiliary_loss_clip": 0.0115587,
      "auxiliary_loss_mlp": 0.01045139,
      "balance_loss_clip": 1.05087829,
      "balance_loss_mlp": 1.02735305,
      "epoch": 0.1806102510145799,
      "flos": 28913512440960.0,
      "grad_norm": 2.436893985401114,
      "language_loss": 0.67536902,
      "learning_rate": 3.7667314902779907e-06,
      "loss": 0.69737911,
      "num_input_tokens_seen": 64917080,
      "step": 3004,
      "time_per_iteration": 2.51444149017334
    },
    {
      "auxiliary_loss_clip": 0.01143871,
      "auxiliary_loss_mlp": 0.01047194,
      "balance_loss_clip": 1.05086207,
      "balance_loss_mlp": 1.02872872,
      "epoch": 0.18067037426724786,
      "flos": 19025689265280.0,
      "grad_norm": 2.8838017094796307,
      "language_loss": 0.85379237,
      "learning_rate": 3.7665489222115677e-06,
      "loss": 0.87570298,
      "num_input_tokens_seen": 64935215,
      "step": 3005,
      "time_per_iteration": 2.492448568344116
    },
    {
      "auxiliary_loss_clip": 0.0113753,
      "auxiliary_loss_mlp": 0.01043215,
      "balance_loss_clip": 1.04862356,
      "balance_loss_mlp": 1.02646589,
      "epoch": 0.18073049751991582,
      "flos": 27453635976960.0,
      "grad_norm": 1.63410259564564,
      "language_loss": 0.83076191,
      "learning_rate": 3.766366287157432e-06,
      "loss": 0.85256934,
      "num_input_tokens_seen": 64956275,
      "step": 3006,
      "time_per_iteration": 5.353426456451416
    },
    {
      "auxiliary_loss_clip": 0.01125479,
      "auxiliary_loss_mlp": 0.01051436,
      "balance_loss_clip": 1.04964197,
      "balance_loss_mlp": 1.03229141,
      "epoch": 0.1807906207725838,
      "flos": 28729039167360.0,
      "grad_norm": 1.6617736784832295,
      "language_loss": 0.77349281,
      "learning_rate": 3.7661835851225103e-06,
      "loss": 0.79526198,
      "num_input_tokens_seen": 64979390,
      "step": 3007,
      "time_per_iteration": 2.588491439819336
    },
    {
      "auxiliary_loss_clip": 0.01060962,
      "auxiliary_loss_mlp": 0.01142458,
      "balance_loss_clip": 1.0342288,
      "balance_loss_mlp": 1.13869143,
      "epoch": 0.18085074402525175,
      "flos": 64466515468800.0,
      "grad_norm": 0.832232338729336,
      "language_loss": 0.56997347,
      "learning_rate": 3.7660008161137294e-06,
      "loss": 0.59200776,
      "num_input_tokens_seen": 65043135,
      "step": 3008,
      "time_per_iteration": 3.2681915760040283
    },
    {
      "auxiliary_loss_clip": 0.01129715,
      "auxiliary_loss_mlp": 0.01048708,
      "balance_loss_clip": 1.05132258,
      "balance_loss_mlp": 1.02988529,
      "epoch": 0.18091086727791975,
      "flos": 23476960333440.0,
      "grad_norm": 1.8147284224596485,
      "language_loss": 0.67225385,
      "learning_rate": 3.765817980138021e-06,
      "loss": 0.69403815,
      "num_input_tokens_seen": 65062845,
      "step": 3009,
      "time_per_iteration": 2.5605862140655518
    },
    {
      "auxiliary_loss_clip": 0.01156802,
      "auxiliary_loss_mlp": 0.01040984,
      "balance_loss_clip": 1.05197465,
      "balance_loss_mlp": 1.02343678,
      "epoch": 0.1809709905305877,
      "flos": 24170467196160.0,
      "grad_norm": 1.734316243490035,
      "language_loss": 0.75327349,
      "learning_rate": 3.7656350772023177e-06,
      "loss": 0.77525139,
      "num_input_tokens_seen": 65082110,
      "step": 3010,
      "time_per_iteration": 3.882336139678955
    },
    {
      "auxiliary_loss_clip": 0.01124392,
      "auxiliary_loss_mlp": 0.01038183,
      "balance_loss_clip": 1.0531168,
      "balance_loss_mlp": 1.02146983,
      "epoch": 0.18103111378325568,
      "flos": 21650902669440.0,
      "grad_norm": 1.5432644083753309,
      "language_loss": 0.66847444,
      "learning_rate": 3.7654521073135553e-06,
      "loss": 0.69010019,
      "num_input_tokens_seen": 65101985,
      "step": 3011,
      "time_per_iteration": 2.5455150604248047
    },
    {
      "auxiliary_loss_clip": 0.01104526,
      "auxiliary_loss_mlp": 0.00888436,
      "balance_loss_clip": 1.04520905,
      "balance_loss_mlp": 1.17580414,
      "epoch": 0.18109123703592364,
      "flos": 53686918356480.0,
      "grad_norm": 2.4008576555381205,
      "language_loss": 0.71091032,
      "learning_rate": 3.7652690704786723e-06,
      "loss": 0.73083991,
      "num_input_tokens_seen": 65129295,
      "step": 3012,
      "time_per_iteration": 4.329438209533691
    },
    {
      "auxiliary_loss_clip": 0.01133053,
      "auxiliary_loss_mlp": 0.0104953,
      "balance_loss_clip": 1.05447125,
      "balance_loss_mlp": 1.03175616,
      "epoch": 0.1811513602885916,
      "flos": 35845564325760.0,
      "grad_norm": 2.186009145711596,
      "language_loss": 0.6240747,
      "learning_rate": 3.765085966704609e-06,
      "loss": 0.64590055,
      "num_input_tokens_seen": 65150625,
      "step": 3013,
      "time_per_iteration": 2.6439266204833984
    },
    {
      "auxiliary_loss_clip": 0.011292,
      "auxiliary_loss_mlp": 0.0104913,
      "balance_loss_clip": 1.0508852,
      "balance_loss_mlp": 1.03167796,
      "epoch": 0.18121148354125957,
      "flos": 23732572492800.0,
      "grad_norm": 1.6275097663988767,
      "language_loss": 0.7624439,
      "learning_rate": 3.764902795998309e-06,
      "loss": 0.78422725,
      "num_input_tokens_seen": 65170880,
      "step": 3014,
      "time_per_iteration": 2.5523548126220703
    },
    {
      "auxiliary_loss_clip": 0.01159059,
      "auxiliary_loss_mlp": 0.01050494,
      "balance_loss_clip": 1.05304039,
      "balance_loss_mlp": 1.03151572,
      "epoch": 0.18127160679392756,
      "flos": 28728320895360.0,
      "grad_norm": 1.6048348246258177,
      "language_loss": 0.65874755,
      "learning_rate": 3.7647195583667184e-06,
      "loss": 0.68084306,
      "num_input_tokens_seen": 65192530,
      "step": 3015,
      "time_per_iteration": 2.5248751640319824
    },
    {
      "auxiliary_loss_clip": 0.01127777,
      "auxiliary_loss_mlp": 0.00856035,
      "balance_loss_clip": 1.0529263,
      "balance_loss_mlp": 1.12223363,
      "epoch": 0.18133173004659553,
      "flos": 20485062938880.0,
      "grad_norm": 1.8759364840424069,
      "language_loss": 0.77984363,
      "learning_rate": 3.764536253816785e-06,
      "loss": 0.79968178,
      "num_input_tokens_seen": 65211675,
      "step": 3016,
      "time_per_iteration": 2.5431647300720215
    },
    {
      "auxiliary_loss_clip": 0.01146265,
      "auxiliary_loss_mlp": 0.01052754,
      "balance_loss_clip": 1.05923653,
      "balance_loss_mlp": 1.03317976,
      "epoch": 0.1813918532992635,
      "flos": 22852078404480.0,
      "grad_norm": 1.847018906288768,
      "language_loss": 0.83317745,
      "learning_rate": 3.7643528823554602e-06,
      "loss": 0.85516763,
      "num_input_tokens_seen": 65231185,
      "step": 3017,
      "time_per_iteration": 2.517423629760742
    },
    {
      "auxiliary_loss_clip": 0.01137922,
      "auxiliary_loss_mlp": 0.01044953,
      "balance_loss_clip": 1.05007684,
      "balance_loss_mlp": 1.02729809,
      "epoch": 0.18145197655193146,
      "flos": 36065122208640.0,
      "grad_norm": 2.572438735080716,
      "language_loss": 0.67508125,
      "learning_rate": 3.764169443989697e-06,
      "loss": 0.69691002,
      "num_input_tokens_seen": 65251645,
      "step": 3018,
      "time_per_iteration": 2.6610617637634277
    },
    {
      "auxiliary_loss_clip": 0.01143004,
      "auxiliary_loss_mlp": 0.00836998,
      "balance_loss_clip": 1.05155325,
      "balance_loss_mlp": 1.08633137,
      "epoch": 0.18151209980459942,
      "flos": 24023951619840.0,
      "grad_norm": 3.5959550620740406,
      "language_loss": 0.76214302,
      "learning_rate": 3.7639859387264518e-06,
      "loss": 0.78194308,
      "num_input_tokens_seen": 65271125,
      "step": 3019,
      "time_per_iteration": 2.518836736679077
    },
    {
      "auxiliary_loss_clip": 0.01112079,
      "auxiliary_loss_mlp": 0.01052473,
      "balance_loss_clip": 1.04689407,
      "balance_loss_mlp": 1.03391218,
      "epoch": 0.1815722230572674,
      "flos": 23951627585280.0,
      "grad_norm": 2.0383154848364913,
      "language_loss": 0.8129406,
      "learning_rate": 3.7638023665726834e-06,
      "loss": 0.83458602,
      "num_input_tokens_seen": 65290600,
      "step": 3020,
      "time_per_iteration": 2.5971333980560303
    },
    {
      "auxiliary_loss_clip": 0.01130494,
      "auxiliary_loss_mlp": 0.01046242,
      "balance_loss_clip": 1.05212879,
      "balance_loss_mlp": 1.02644134,
      "epoch": 0.18163234630993536,
      "flos": 24386469632640.0,
      "grad_norm": 2.519290447259399,
      "language_loss": 0.77664095,
      "learning_rate": 3.763618727535352e-06,
      "loss": 0.79840827,
      "num_input_tokens_seen": 65311040,
      "step": 3021,
      "time_per_iteration": 2.571307420730591
    },
    {
      "auxiliary_loss_clip": 0.01139989,
      "auxiliary_loss_mlp": 0.01043632,
      "balance_loss_clip": 1.04967475,
      "balance_loss_mlp": 1.02544022,
      "epoch": 0.18169246956260335,
      "flos": 24681332378880.0,
      "grad_norm": 1.6882240826822532,
      "language_loss": 0.8503387,
      "learning_rate": 3.763435021621422e-06,
      "loss": 0.87217486,
      "num_input_tokens_seen": 65332115,
      "step": 3022,
      "time_per_iteration": 2.5278990268707275
    },
    {
      "auxiliary_loss_clip": 0.01127891,
      "auxiliary_loss_mlp": 0.01047785,
      "balance_loss_clip": 1.04890752,
      "balance_loss_mlp": 1.0276866,
      "epoch": 0.1817525928152713,
      "flos": 24243294021120.0,
      "grad_norm": 1.9335829017026491,
      "language_loss": 0.69473904,
      "learning_rate": 3.763251248837859e-06,
      "loss": 0.71649581,
      "num_input_tokens_seen": 65352210,
      "step": 3023,
      "time_per_iteration": 2.577880620956421
    },
    {
      "auxiliary_loss_clip": 0.01125279,
      "auxiliary_loss_mlp": 0.01042451,
      "balance_loss_clip": 1.0456773,
      "balance_loss_mlp": 1.02419996,
      "epoch": 0.18181271606793928,
      "flos": 16472081623680.0,
      "grad_norm": 1.9938471822198907,
      "language_loss": 0.74280739,
      "learning_rate": 3.7630674091916317e-06,
      "loss": 0.76448464,
      "num_input_tokens_seen": 65370600,
      "step": 3024,
      "time_per_iteration": 2.5060901641845703
    },
    {
      "auxiliary_loss_clip": 0.01139085,
      "auxiliary_loss_mlp": 0.01046804,
      "balance_loss_clip": 1.05050659,
      "balance_loss_mlp": 1.02834988,
      "epoch": 0.18187283932060724,
      "flos": 18581042805120.0,
      "grad_norm": 2.07232941820835,
      "language_loss": 0.88403791,
      "learning_rate": 3.7628835026897123e-06,
      "loss": 0.90589684,
      "num_input_tokens_seen": 65387270,
      "step": 3025,
      "time_per_iteration": 2.487823009490967
    },
    {
      "auxiliary_loss_clip": 0.01128431,
      "auxiliary_loss_mlp": 0.01048904,
      "balance_loss_clip": 1.04842067,
      "balance_loss_mlp": 1.03039062,
      "epoch": 0.1819329625732752,
      "flos": 20266833859200.0,
      "grad_norm": 1.755384225735399,
      "language_loss": 0.78998554,
      "learning_rate": 3.7626995293390735e-06,
      "loss": 0.81175888,
      "num_input_tokens_seen": 65406550,
      "step": 3026,
      "time_per_iteration": 2.5295052528381348
    },
    {
      "auxiliary_loss_clip": 0.01123457,
      "auxiliary_loss_mlp": 0.01050685,
      "balance_loss_clip": 1.05085063,
      "balance_loss_mlp": 1.03212392,
      "epoch": 0.18199308582594317,
      "flos": 25915186512000.0,
      "grad_norm": 1.6237678337750117,
      "language_loss": 0.76157218,
      "learning_rate": 3.762515489146692e-06,
      "loss": 0.78331357,
      "num_input_tokens_seen": 65425955,
      "step": 3027,
      "time_per_iteration": 2.58508563041687
    },
    {
      "auxiliary_loss_clip": 0.01156619,
      "auxiliary_loss_mlp": 0.01050518,
      "balance_loss_clip": 1.05017304,
      "balance_loss_mlp": 1.03144443,
      "epoch": 0.18205320907861114,
      "flos": 15377524433280.0,
      "grad_norm": 1.92110293280215,
      "language_loss": 0.85164356,
      "learning_rate": 3.762331382119546e-06,
      "loss": 0.87371492,
      "num_input_tokens_seen": 65442820,
      "step": 3028,
      "time_per_iteration": 2.461099863052368
    },
    {
      "auxiliary_loss_clip": 0.01150103,
      "auxiliary_loss_mlp": 0.01042013,
      "balance_loss_clip": 1.04958451,
      "balance_loss_mlp": 1.02443004,
      "epoch": 0.18211333233127913,
      "flos": 25624310175360.0,
      "grad_norm": 1.76432801193489,
      "language_loss": 0.82592273,
      "learning_rate": 3.7621472082646183e-06,
      "loss": 0.84784383,
      "num_input_tokens_seen": 65461825,
      "step": 3029,
      "time_per_iteration": 2.530620813369751
    },
    {
      "auxiliary_loss_clip": 0.01112183,
      "auxiliary_loss_mlp": 0.01048256,
      "balance_loss_clip": 1.04885101,
      "balance_loss_mlp": 1.02766812,
      "epoch": 0.1821734555839471,
      "flos": 14976007228800.0,
      "grad_norm": 2.383843248053209,
      "language_loss": 0.77767396,
      "learning_rate": 3.761962967588891e-06,
      "loss": 0.79927832,
      "num_input_tokens_seen": 65479480,
      "step": 3030,
      "time_per_iteration": 2.544044256210327
    },
    {
      "auxiliary_loss_clip": 0.01134863,
      "auxiliary_loss_mlp": 0.01049217,
      "balance_loss_clip": 1.04780674,
      "balance_loss_mlp": 1.03023887,
      "epoch": 0.18223357883661506,
      "flos": 20194007034240.0,
      "grad_norm": 2.147355474320735,
      "language_loss": 0.84680527,
      "learning_rate": 3.761778660099352e-06,
      "loss": 0.86864603,
      "num_input_tokens_seen": 65497775,
      "step": 3031,
      "time_per_iteration": 2.503438711166382
    },
    {
      "auxiliary_loss_clip": 0.01114433,
      "auxiliary_loss_mlp": 0.00799019,
      "balance_loss_clip": 1.04642272,
      "balance_loss_mlp": 1.01588941,
      "epoch": 0.18229370208928303,
      "flos": 15231978524160.0,
      "grad_norm": 1.816780386456844,
      "language_loss": 0.79894567,
      "learning_rate": 3.76159428580299e-06,
      "loss": 0.81808019,
      "num_input_tokens_seen": 65516505,
      "step": 3032,
      "time_per_iteration": 2.555082082748413
    },
    {
      "auxiliary_loss_clip": 0.01157713,
      "auxiliary_loss_mlp": 0.01052839,
      "balance_loss_clip": 1.05180621,
      "balance_loss_mlp": 1.03445685,
      "epoch": 0.182353825341951,
      "flos": 23840483927040.0,
      "grad_norm": 3.5027688852048864,
      "language_loss": 0.81180245,
      "learning_rate": 3.761409844706795e-06,
      "loss": 0.83390796,
      "num_input_tokens_seen": 65536160,
      "step": 3033,
      "time_per_iteration": 2.522714376449585
    },
    {
      "auxiliary_loss_clip": 0.01045753,
      "auxiliary_loss_mlp": 0.01005405,
      "balance_loss_clip": 1.04537964,
      "balance_loss_mlp": 1.00209129,
      "epoch": 0.18241394859461896,
      "flos": 61190957393280.0,
      "grad_norm": 0.8803895618834763,
      "language_loss": 0.63503361,
      "learning_rate": 3.7612253368177625e-06,
      "loss": 0.65554512,
      "num_input_tokens_seen": 65589375,
      "step": 3034,
      "time_per_iteration": 3.121440887451172
    },
    {
      "auxiliary_loss_clip": 0.01120553,
      "auxiliary_loss_mlp": 0.01039728,
      "balance_loss_clip": 1.04749155,
      "balance_loss_mlp": 1.02265751,
      "epoch": 0.18247407184728695,
      "flos": 18471694826880.0,
      "grad_norm": 2.4173441212920204,
      "language_loss": 0.79455721,
      "learning_rate": 3.7610407621428893e-06,
      "loss": 0.81616002,
      "num_input_tokens_seen": 65606720,
      "step": 3035,
      "time_per_iteration": 2.547461748123169
    },
    {
      "auxiliary_loss_clip": 0.01124515,
      "auxiliary_loss_mlp": 0.01043108,
      "balance_loss_clip": 1.04666245,
      "balance_loss_mlp": 1.02616882,
      "epoch": 0.18253419509995492,
      "flos": 21795191602560.0,
      "grad_norm": 1.8993409684698206,
      "language_loss": 0.85167599,
      "learning_rate": 3.7608561206891735e-06,
      "loss": 0.87335217,
      "num_input_tokens_seen": 65625495,
      "step": 3036,
      "time_per_iteration": 2.536027669906616
    },
    {
      "auxiliary_loss_clip": 0.01136625,
      "auxiliary_loss_mlp": 0.01041347,
      "balance_loss_clip": 1.05230117,
      "balance_loss_mlp": 1.02424026,
      "epoch": 0.18259431835262288,
      "flos": 20149764456960.0,
      "grad_norm": 1.941360197198144,
      "language_loss": 0.79997623,
      "learning_rate": 3.760671412463617e-06,
      "loss": 0.82175589,
      "num_input_tokens_seen": 65643515,
      "step": 3037,
      "time_per_iteration": 2.489978075027466
    },
    {
      "auxiliary_loss_clip": 0.01134634,
      "auxiliary_loss_mlp": 0.00802021,
      "balance_loss_clip": 1.05343008,
      "balance_loss_mlp": 1.02217436,
      "epoch": 0.18265444160529085,
      "flos": 16981653916800.0,
      "grad_norm": 3.0855967349767783,
      "language_loss": 0.79188943,
      "learning_rate": 3.7604866374732246e-06,
      "loss": 0.81125599,
      "num_input_tokens_seen": 65658155,
      "step": 3038,
      "time_per_iteration": 2.496943712234497
    },
    {
      "auxiliary_loss_clip": 0.01120084,
      "auxiliary_loss_mlp": 0.01051857,
      "balance_loss_clip": 1.04518008,
      "balance_loss_mlp": 1.03321242,
      "epoch": 0.1827145648579588,
      "flos": 34423250509440.0,
      "grad_norm": 2.3292607438771724,
      "language_loss": 0.67858362,
      "learning_rate": 3.7603017957250023e-06,
      "loss": 0.70030302,
      "num_input_tokens_seen": 65679310,
      "step": 3039,
      "time_per_iteration": 2.644120216369629
    },
    {
      "auxiliary_loss_clip": 0.01125927,
      "auxiliary_loss_mlp": 0.01048927,
      "balance_loss_clip": 1.0490737,
      "balance_loss_mlp": 1.03034258,
      "epoch": 0.18277468811062678,
      "flos": 53287017264000.0,
      "grad_norm": 4.270624063924718,
      "language_loss": 0.73843759,
      "learning_rate": 3.7601168872259593e-06,
      "loss": 0.76018608,
      "num_input_tokens_seen": 65705235,
      "step": 3040,
      "time_per_iteration": 2.813457727432251
    },
    {
      "auxiliary_loss_clip": 0.01137662,
      "auxiliary_loss_mlp": 0.0104323,
      "balance_loss_clip": 1.04861677,
      "balance_loss_mlp": 1.02477646,
      "epoch": 0.18283481136329474,
      "flos": 31650659602560.0,
      "grad_norm": 2.0987730895272594,
      "language_loss": 0.59925443,
      "learning_rate": 3.7599319119831075e-06,
      "loss": 0.62106329,
      "num_input_tokens_seen": 65727575,
      "step": 3041,
      "time_per_iteration": 2.5746612548828125
    },
    {
      "auxiliary_loss_clip": 0.01115929,
      "auxiliary_loss_mlp": 0.01050737,
      "balance_loss_clip": 1.04949045,
      "balance_loss_mlp": 1.03259373,
      "epoch": 0.18289493461596273,
      "flos": 53137664513280.0,
      "grad_norm": 1.7230038706479454,
      "language_loss": 0.59818041,
      "learning_rate": 3.7597468700034616e-06,
      "loss": 0.61984706,
      "num_input_tokens_seen": 65751370,
      "step": 3042,
      "time_per_iteration": 2.856874942779541
    },
    {
      "auxiliary_loss_clip": 0.0112052,
      "auxiliary_loss_mlp": 0.01047073,
      "balance_loss_clip": 1.04972649,
      "balance_loss_mlp": 1.02926278,
      "epoch": 0.1829550578686307,
      "flos": 25589369220480.0,
      "grad_norm": 1.7002060007697757,
      "language_loss": 0.87647045,
      "learning_rate": 3.7595617612940374e-06,
      "loss": 0.89814645,
      "num_input_tokens_seen": 65771040,
      "step": 3043,
      "time_per_iteration": 2.5682122707366943
    },
    {
      "auxiliary_loss_clip": 0.01055352,
      "auxiliary_loss_mlp": 0.01051113,
      "balance_loss_clip": 1.04976773,
      "balance_loss_mlp": 1.03078806,
      "epoch": 0.18301518112129866,
      "flos": 22601422321920.0,
      "grad_norm": 2.222263798331082,
      "language_loss": 0.70991659,
      "learning_rate": 3.7593765858618552e-06,
      "loss": 0.73098123,
      "num_input_tokens_seen": 65789345,
      "step": 3044,
      "time_per_iteration": 4.109682559967041
    },
    {
      "auxiliary_loss_clip": 0.01101691,
      "auxiliary_loss_mlp": 0.01049206,
      "balance_loss_clip": 1.04916775,
      "balance_loss_mlp": 1.03002512,
      "epoch": 0.18307530437396663,
      "flos": 34020799551360.0,
      "grad_norm": 2.2139638467993743,
      "language_loss": 0.64183354,
      "learning_rate": 3.7591913437139365e-06,
      "loss": 0.66334254,
      "num_input_tokens_seen": 65810990,
      "step": 3045,
      "time_per_iteration": 4.099668264389038
    },
    {
      "auxiliary_loss_clip": 0.01151257,
      "auxiliary_loss_mlp": 0.01044936,
      "balance_loss_clip": 1.05058086,
      "balance_loss_mlp": 1.02736449,
      "epoch": 0.1831354276266346,
      "flos": 21279765392640.0,
      "grad_norm": 7.627348590818992,
      "language_loss": 0.79761964,
      "learning_rate": 3.7590060348573066e-06,
      "loss": 0.81958157,
      "num_input_tokens_seen": 65827230,
      "step": 3046,
      "time_per_iteration": 2.4398183822631836
    },
    {
      "auxiliary_loss_clip": 0.01115213,
      "auxiliary_loss_mlp": 0.01044748,
      "balance_loss_clip": 1.04415274,
      "balance_loss_mlp": 1.02628255,
      "epoch": 0.18319555087930256,
      "flos": 21032952065280.0,
      "grad_norm": 1.9832759410969631,
      "language_loss": 0.79380405,
      "learning_rate": 3.7588206592989903e-06,
      "loss": 0.81540358,
      "num_input_tokens_seen": 65845900,
      "step": 3047,
      "time_per_iteration": 2.556051015853882
    },
    {
      "auxiliary_loss_clip": 0.01138514,
      "auxiliary_loss_mlp": 0.01041199,
      "balance_loss_clip": 1.04990268,
      "balance_loss_mlp": 1.02427101,
      "epoch": 0.18325567413197055,
      "flos": 34382958428160.0,
      "grad_norm": 1.4956425557172026,
      "language_loss": 0.81094456,
      "learning_rate": 3.7586352170460194e-06,
      "loss": 0.83274174,
      "num_input_tokens_seen": 65868730,
      "step": 3048,
      "time_per_iteration": 3.9817967414855957
    },
    {
      "auxiliary_loss_clip": 0.01135013,
      "auxiliary_loss_mlp": 0.01045857,
      "balance_loss_clip": 1.04692936,
      "balance_loss_mlp": 1.02587724,
      "epoch": 0.18331579738463852,
      "flos": 20558464381440.0,
      "grad_norm": 1.8157378427220674,
      "language_loss": 0.86672133,
      "learning_rate": 3.758449708105424e-06,
      "loss": 0.88853002,
      "num_input_tokens_seen": 65888420,
      "step": 3049,
      "time_per_iteration": 2.513273239135742
    },
    {
      "auxiliary_loss_clip": 0.01139509,
      "auxiliary_loss_mlp": 0.01049137,
      "balance_loss_clip": 1.04845548,
      "balance_loss_mlp": 1.02899051,
      "epoch": 0.18337592063730648,
      "flos": 19607872901760.0,
      "grad_norm": 2.492652110335141,
      "language_loss": 0.77594405,
      "learning_rate": 3.75826413248424e-06,
      "loss": 0.79783052,
      "num_input_tokens_seen": 65905840,
      "step": 3050,
      "time_per_iteration": 2.465207099914551
    },
    {
      "auxiliary_loss_clip": 0.01123233,
      "auxiliary_loss_mlp": 0.01039915,
      "balance_loss_clip": 1.04478538,
      "balance_loss_mlp": 1.02198553,
      "epoch": 0.18343604388997445,
      "flos": 20850885002880.0,
      "grad_norm": 1.9689854727808425,
      "language_loss": 0.99462783,
      "learning_rate": 3.7580784901895035e-06,
      "loss": 1.01625919,
      "num_input_tokens_seen": 65922845,
      "step": 3051,
      "time_per_iteration": 3.9788222312927246
    },
    {
      "auxiliary_loss_clip": 0.01121681,
      "auxiliary_loss_mlp": 0.01039334,
      "balance_loss_clip": 1.04715991,
      "balance_loss_mlp": 1.02004576,
      "epoch": 0.1834961671426424,
      "flos": 24394370624640.0,
      "grad_norm": 1.672531151101026,
      "language_loss": 0.86072123,
      "learning_rate": 3.7578927812282542e-06,
      "loss": 0.88233137,
      "num_input_tokens_seen": 65945555,
      "step": 3052,
      "time_per_iteration": 2.6032707691192627
    },
    {
      "auxiliary_loss_clip": 0.01148519,
      "auxiliary_loss_mlp": 0.01044179,
      "balance_loss_clip": 1.04846859,
      "balance_loss_mlp": 1.02628589,
      "epoch": 0.18355629039531038,
      "flos": 21251612108160.0,
      "grad_norm": 1.9876039844477378,
      "language_loss": 0.73149574,
      "learning_rate": 3.7577070056075356e-06,
      "loss": 0.75342268,
      "num_input_tokens_seen": 65963965,
      "step": 3053,
      "time_per_iteration": 2.4534664154052734
    },
    {
      "auxiliary_loss_clip": 0.01154114,
      "auxiliary_loss_mlp": 0.01043306,
      "balance_loss_clip": 1.05102777,
      "balance_loss_mlp": 1.02478147,
      "epoch": 0.18361641364797834,
      "flos": 28656499651200.0,
      "grad_norm": 1.9103898394068557,
      "language_loss": 0.61786938,
      "learning_rate": 3.7575211633343902e-06,
      "loss": 0.63984364,
      "num_input_tokens_seen": 65985965,
      "step": 3054,
      "time_per_iteration": 2.54599666595459
    },
    {
      "auxiliary_loss_clip": 0.01103799,
      "auxiliary_loss_mlp": 0.01047439,
      "balance_loss_clip": 1.05103052,
      "balance_loss_mlp": 1.03017771,
      "epoch": 0.18367653690064634,
      "flos": 20918827578240.0,
      "grad_norm": 2.0949643679263215,
      "language_loss": 0.78160024,
      "learning_rate": 3.7573352544158663e-06,
      "loss": 0.80311263,
      "num_input_tokens_seen": 66005645,
      "step": 3055,
      "time_per_iteration": 2.6206202507019043
    },
    {
      "auxiliary_loss_clip": 0.01092097,
      "auxiliary_loss_mlp": 0.01052607,
      "balance_loss_clip": 1.0456965,
      "balance_loss_mlp": 1.0337956,
      "epoch": 0.1837366601533143,
      "flos": 28765596234240.0,
      "grad_norm": 2.579964854704961,
      "language_loss": 0.7009266,
      "learning_rate": 3.757149278859014e-06,
      "loss": 0.72237366,
      "num_input_tokens_seen": 66025675,
      "step": 3056,
      "time_per_iteration": 2.654783248901367
    },
    {
      "auxiliary_loss_clip": 0.01139668,
      "auxiliary_loss_mlp": 0.0104184,
      "balance_loss_clip": 1.04824591,
      "balance_loss_mlp": 1.02466178,
      "epoch": 0.18379678340598227,
      "flos": 21251432540160.0,
      "grad_norm": 1.522596740654723,
      "language_loss": 0.80604124,
      "learning_rate": 3.7569632366708842e-06,
      "loss": 0.8278563,
      "num_input_tokens_seen": 66046125,
      "step": 3057,
      "time_per_iteration": 2.4931893348693848
    },
    {
      "auxiliary_loss_clip": 0.01145774,
      "auxiliary_loss_mlp": 0.01042374,
      "balance_loss_clip": 1.04888403,
      "balance_loss_mlp": 1.02183425,
      "epoch": 0.18385690665865023,
      "flos": 20449619193600.0,
      "grad_norm": 2.1576110602964205,
      "language_loss": 0.82501209,
      "learning_rate": 3.756777127858533e-06,
      "loss": 0.84689355,
      "num_input_tokens_seen": 66064375,
      "step": 3058,
      "time_per_iteration": 2.4946346282958984
    },
    {
      "auxiliary_loss_clip": 0.01113085,
      "auxiliary_loss_mlp": 0.00802743,
      "balance_loss_clip": 1.04399538,
      "balance_loss_mlp": 1.02248967,
      "epoch": 0.1839170299113182,
      "flos": 26140562398080.0,
      "grad_norm": 2.7544625627371118,
      "language_loss": 0.85848188,
      "learning_rate": 3.756590952429017e-06,
      "loss": 0.87764013,
      "num_input_tokens_seen": 66084590,
      "step": 3059,
      "time_per_iteration": 2.58817458152771
    },
    {
      "auxiliary_loss_clip": 0.0114813,
      "auxiliary_loss_mlp": 0.00801536,
      "balance_loss_clip": 1.04827285,
      "balance_loss_mlp": 1.02199769,
      "epoch": 0.18397715316398616,
      "flos": 31758032332800.0,
      "grad_norm": 1.5258600786635184,
      "language_loss": 0.72777092,
      "learning_rate": 3.756404710389396e-06,
      "loss": 0.7472676,
      "num_input_tokens_seen": 66107105,
      "step": 3060,
      "time_per_iteration": 2.5587899684906006
    },
    {
      "auxiliary_loss_clip": 0.01142343,
      "auxiliary_loss_mlp": 0.01037989,
      "balance_loss_clip": 1.04887867,
      "balance_loss_mlp": 1.01915431,
      "epoch": 0.18403727641665413,
      "flos": 24611989173120.0,
      "grad_norm": 1.9628524085829961,
      "language_loss": 0.72702318,
      "learning_rate": 3.7562184017467323e-06,
      "loss": 0.7488265,
      "num_input_tokens_seen": 66129295,
      "step": 3061,
      "time_per_iteration": 2.5410821437835693
    },
    {
      "auxiliary_loss_clip": 0.01133042,
      "auxiliary_loss_mlp": 0.0104823,
      "balance_loss_clip": 1.04785919,
      "balance_loss_mlp": 1.02914417,
      "epoch": 0.18409739966932212,
      "flos": 23439900476160.0,
      "grad_norm": 2.3964887658841176,
      "language_loss": 0.81591642,
      "learning_rate": 3.7560320265080906e-06,
      "loss": 0.8377291,
      "num_input_tokens_seen": 66146910,
      "step": 3062,
      "time_per_iteration": 2.510471820831299
    },
    {
      "auxiliary_loss_clip": 0.01138764,
      "auxiliary_loss_mlp": 0.01042435,
      "balance_loss_clip": 1.05029237,
      "balance_loss_mlp": 1.02426767,
      "epoch": 0.18415752292199009,
      "flos": 21872112577920.0,
      "grad_norm": 2.382880010780113,
      "language_loss": 0.7276963,
      "learning_rate": 3.7558455846805383e-06,
      "loss": 0.74950832,
      "num_input_tokens_seen": 66165370,
      "step": 3063,
      "time_per_iteration": 2.498452663421631
    },
    {
      "auxiliary_loss_clip": 0.0112776,
      "auxiliary_loss_mlp": 0.01037247,
      "balance_loss_clip": 1.04674637,
      "balance_loss_mlp": 1.02147579,
      "epoch": 0.18421764617465805,
      "flos": 25410678036480.0,
      "grad_norm": 1.7063326577187148,
      "language_loss": 0.65845263,
      "learning_rate": 3.7556590762711463e-06,
      "loss": 0.68010271,
      "num_input_tokens_seen": 66186210,
      "step": 3064,
      "time_per_iteration": 2.5683610439300537
    },
    {
      "auxiliary_loss_clip": 0.01136273,
      "auxiliary_loss_mlp": 0.01045549,
      "balance_loss_clip": 1.04829431,
      "balance_loss_mlp": 1.02810907,
      "epoch": 0.18427776942732602,
      "flos": 27198131558400.0,
      "grad_norm": 1.9732491546125797,
      "language_loss": 0.68794978,
      "learning_rate": 3.7554725012869853e-06,
      "loss": 0.709768,
      "num_input_tokens_seen": 66204800,
      "step": 3065,
      "time_per_iteration": 2.5219485759735107
    },
    {
      "auxiliary_loss_clip": 0.01131389,
      "auxiliary_loss_mlp": 0.01041872,
      "balance_loss_clip": 1.04947162,
      "balance_loss_mlp": 1.02321553,
      "epoch": 0.18433789267999398,
      "flos": 27852351920640.0,
      "grad_norm": 3.16796414263879,
      "language_loss": 0.73329222,
      "learning_rate": 3.7552858597351318e-06,
      "loss": 0.75502485,
      "num_input_tokens_seen": 66222195,
      "step": 3066,
      "time_per_iteration": 2.593531847000122
    },
    {
      "auxiliary_loss_clip": 0.01124078,
      "auxiliary_loss_mlp": 0.010383,
      "balance_loss_clip": 1.04920459,
      "balance_loss_mlp": 1.0208714,
      "epoch": 0.18439801593266195,
      "flos": 17856940533120.0,
      "grad_norm": 2.16173903392107,
      "language_loss": 0.81802773,
      "learning_rate": 3.7550991516226622e-06,
      "loss": 0.83965158,
      "num_input_tokens_seen": 66239505,
      "step": 3067,
      "time_per_iteration": 2.5086212158203125
    },
    {
      "auxiliary_loss_clip": 0.01080239,
      "auxiliary_loss_mlp": 0.01033633,
      "balance_loss_clip": 1.03781104,
      "balance_loss_mlp": 1.43463171,
      "epoch": 0.18445813918532994,
      "flos": 56389522590720.0,
      "grad_norm": 0.7940779191365099,
      "language_loss": 0.59711432,
      "learning_rate": 3.754912376956657e-06,
      "loss": 0.61825311,
      "num_input_tokens_seen": 66295695,
      "step": 3068,
      "time_per_iteration": 2.94635009765625
    },
    {
      "auxiliary_loss_clip": 0.01121956,
      "auxiliary_loss_mlp": 0.0103793,
      "balance_loss_clip": 1.05223703,
      "balance_loss_mlp": 1.0216105,
      "epoch": 0.1845182624379979,
      "flos": 20957180325120.0,
      "grad_norm": 1.716663015529881,
      "language_loss": 0.76590329,
      "learning_rate": 3.7547255357441987e-06,
      "loss": 0.78750217,
      "num_input_tokens_seen": 66315315,
      "step": 3069,
      "time_per_iteration": 2.5468287467956543
    },
    {
      "auxiliary_loss_clip": 0.01139705,
      "auxiliary_loss_mlp": 0.01041682,
      "balance_loss_clip": 1.04732788,
      "balance_loss_mlp": 1.02419376,
      "epoch": 0.18457838569066587,
      "flos": 20485170679680.0,
      "grad_norm": 1.9678401891460418,
      "language_loss": 0.84665549,
      "learning_rate": 3.7545386279923718e-06,
      "loss": 0.86846936,
      "num_input_tokens_seen": 66333675,
      "step": 3070,
      "time_per_iteration": 2.5022239685058594
    },
    {
      "auxiliary_loss_clip": 0.0111995,
      "auxiliary_loss_mlp": 0.01045263,
      "balance_loss_clip": 1.04642272,
      "balance_loss_mlp": 1.02695227,
      "epoch": 0.18463850894333383,
      "flos": 25010022758400.0,
      "grad_norm": 2.0233314103901523,
      "language_loss": 0.77879,
      "learning_rate": 3.754351653708265e-06,
      "loss": 0.8004421,
      "num_input_tokens_seen": 66354075,
      "step": 3071,
      "time_per_iteration": 2.5916078090667725
    },
    {
      "auxiliary_loss_clip": 0.01108532,
      "auxiliary_loss_mlp": 0.01044425,
      "balance_loss_clip": 1.05208874,
      "balance_loss_mlp": 1.0270561,
      "epoch": 0.1846986321960018,
      "flos": 16800628348800.0,
      "grad_norm": 1.9459660169171797,
      "language_loss": 0.76967871,
      "learning_rate": 3.7541646128989674e-06,
      "loss": 0.79120827,
      "num_input_tokens_seen": 66372520,
      "step": 3072,
      "time_per_iteration": 2.5603513717651367
    },
    {
      "auxiliary_loss_clip": 0.01135489,
      "auxiliary_loss_mlp": 0.01041117,
      "balance_loss_clip": 1.04911721,
      "balance_loss_mlp": 1.02265108,
      "epoch": 0.18475875544866976,
      "flos": 20814327936000.0,
      "grad_norm": 2.2883955609671034,
      "language_loss": 0.86214983,
      "learning_rate": 3.7539775055715715e-06,
      "loss": 0.88391584,
      "num_input_tokens_seen": 66390745,
      "step": 3073,
      "time_per_iteration": 2.487492322921753
    },
    {
      "auxiliary_loss_clip": 0.0115172,
      "auxiliary_loss_mlp": 0.01042715,
      "balance_loss_clip": 1.05145657,
      "balance_loss_mlp": 1.02616906,
      "epoch": 0.18481887870133773,
      "flos": 22601422321920.0,
      "grad_norm": 2.3549350129025335,
      "language_loss": 0.91308981,
      "learning_rate": 3.7537903317331732e-06,
      "loss": 0.93503416,
      "num_input_tokens_seen": 66410525,
      "step": 3074,
      "time_per_iteration": 2.4682490825653076
    },
    {
      "auxiliary_loss_clip": 0.01101422,
      "auxiliary_loss_mlp": 0.01045007,
      "balance_loss_clip": 1.04643607,
      "balance_loss_mlp": 1.02630329,
      "epoch": 0.18487900195400572,
      "flos": 29458815788160.0,
      "grad_norm": 1.7614725777912241,
      "language_loss": 0.64965224,
      "learning_rate": 3.75360309139087e-06,
      "loss": 0.67111653,
      "num_input_tokens_seen": 66432535,
      "step": 3075,
      "time_per_iteration": 2.6090705394744873
    },
    {
      "auxiliary_loss_clip": 0.01130095,
      "auxiliary_loss_mlp": 0.01040378,
      "balance_loss_clip": 1.05128849,
      "balance_loss_mlp": 1.02317572,
      "epoch": 0.1849391252066737,
      "flos": 20628777254400.0,
      "grad_norm": 1.7856848149354143,
      "language_loss": 0.73173785,
      "learning_rate": 3.753415784551761e-06,
      "loss": 0.75344253,
      "num_input_tokens_seen": 66450620,
      "step": 3076,
      "time_per_iteration": 2.5256187915802
    },
    {
      "auxiliary_loss_clip": 0.01115072,
      "auxiliary_loss_mlp": 0.01037941,
      "balance_loss_clip": 1.05146003,
      "balance_loss_mlp": 1.02145422,
      "epoch": 0.18499924845934165,
      "flos": 14428549065600.0,
      "grad_norm": 2.363784441141818,
      "language_loss": 0.81269217,
      "learning_rate": 3.7532284112229507e-06,
      "loss": 0.83422232,
      "num_input_tokens_seen": 66467865,
      "step": 3077,
      "time_per_iteration": 2.5930373668670654
    },
    {
      "auxiliary_loss_clip": 0.011202,
      "auxiliary_loss_mlp": 0.01043445,
      "balance_loss_clip": 1.04882789,
      "balance_loss_mlp": 1.02664876,
      "epoch": 0.18505937171200962,
      "flos": 23727652329600.0,
      "grad_norm": 2.112828914358928,
      "language_loss": 0.78953946,
      "learning_rate": 3.7530409714115424e-06,
      "loss": 0.81117594,
      "num_input_tokens_seen": 66486245,
      "step": 3078,
      "time_per_iteration": 2.5690486431121826
    },
    {
      "auxiliary_loss_clip": 0.01150232,
      "auxiliary_loss_mlp": 0.01047011,
      "balance_loss_clip": 1.04968452,
      "balance_loss_mlp": 1.03051209,
      "epoch": 0.18511949496467758,
      "flos": 25957489754880.0,
      "grad_norm": 2.086056853094844,
      "language_loss": 0.77418655,
      "learning_rate": 3.7528534651246453e-06,
      "loss": 0.79615897,
      "num_input_tokens_seen": 66506510,
      "step": 3079,
      "time_per_iteration": 2.52986216545105
    },
    {
      "auxiliary_loss_clip": 0.01115944,
      "auxiliary_loss_mlp": 0.01038605,
      "balance_loss_clip": 1.04744911,
      "balance_loss_mlp": 1.02146292,
      "epoch": 0.18517961821734555,
      "flos": 42413553912960.0,
      "grad_norm": 1.8103780844410122,
      "language_loss": 0.82101977,
      "learning_rate": 3.752665892369369e-06,
      "loss": 0.84256518,
      "num_input_tokens_seen": 66530960,
      "step": 3080,
      "time_per_iteration": 2.782543659210205
    },
    {
      "auxiliary_loss_clip": 0.01110571,
      "auxiliary_loss_mlp": 0.01045729,
      "balance_loss_clip": 1.04607666,
      "balance_loss_mlp": 1.02768087,
      "epoch": 0.18523974147001354,
      "flos": 24097568544000.0,
      "grad_norm": 1.8618077101822463,
      "language_loss": 0.74423003,
      "learning_rate": 3.7524782531528266e-06,
      "loss": 0.76579297,
      "num_input_tokens_seen": 66550275,
      "step": 3081,
      "time_per_iteration": 2.615497589111328
    },
    {
      "auxiliary_loss_clip": 0.01123855,
      "auxiliary_loss_mlp": 0.01053373,
      "balance_loss_clip": 1.05157971,
      "balance_loss_mlp": 1.03541994,
      "epoch": 0.1852998647226815,
      "flos": 27375278457600.0,
      "grad_norm": 2.0844464833460634,
      "language_loss": 0.72034919,
      "learning_rate": 3.7522905474821334e-06,
      "loss": 0.74212146,
      "num_input_tokens_seen": 66569040,
      "step": 3082,
      "time_per_iteration": 2.577605724334717
    },
    {
      "auxiliary_loss_clip": 0.01119939,
      "auxiliary_loss_mlp": 0.01049092,
      "balance_loss_clip": 1.05342233,
      "balance_loss_mlp": 1.03066218,
      "epoch": 0.18535998797534947,
      "flos": 18332757020160.0,
      "grad_norm": 2.477494426279642,
      "language_loss": 0.69530118,
      "learning_rate": 3.752102775364407e-06,
      "loss": 0.71699154,
      "num_input_tokens_seen": 66587775,
      "step": 3083,
      "time_per_iteration": 3.943408727645874
    },
    {
      "auxiliary_loss_clip": 0.01119309,
      "auxiliary_loss_mlp": 0.01048838,
      "balance_loss_clip": 1.0493331,
      "balance_loss_mlp": 1.03173113,
      "epoch": 0.18542011122801744,
      "flos": 37845859887360.0,
      "grad_norm": 2.5278872220185153,
      "language_loss": 0.68733144,
      "learning_rate": 3.751914936806767e-06,
      "loss": 0.70901287,
      "num_input_tokens_seen": 66610800,
      "step": 3084,
      "time_per_iteration": 2.6786043643951416
    },
    {
      "auxiliary_loss_clip": 0.01149162,
      "auxiliary_loss_mlp": 0.01041377,
      "balance_loss_clip": 1.04949522,
      "balance_loss_mlp": 1.02436566,
      "epoch": 0.1854802344806854,
      "flos": 25186128163200.0,
      "grad_norm": 1.575877891965453,
      "language_loss": 0.78138876,
      "learning_rate": 3.7517270318163377e-06,
      "loss": 0.80329418,
      "num_input_tokens_seen": 66630960,
      "step": 3085,
      "time_per_iteration": 2.5270917415618896
    },
    {
      "auxiliary_loss_clip": 0.01148856,
      "auxiliary_loss_mlp": 0.0104766,
      "balance_loss_clip": 1.04882574,
      "balance_loss_mlp": 1.03093529,
      "epoch": 0.18554035773335337,
      "flos": 26684788337280.0,
      "grad_norm": 2.204581040648067,
      "language_loss": 0.73339611,
      "learning_rate": 3.751539060400244e-06,
      "loss": 0.7553612,
      "num_input_tokens_seen": 66650585,
      "step": 3086,
      "time_per_iteration": 2.5025787353515625
    },
    {
      "auxiliary_loss_clip": 0.01137533,
      "auxiliary_loss_mlp": 0.01044697,
      "balance_loss_clip": 1.04881334,
      "balance_loss_mlp": 1.02670801,
      "epoch": 0.18560048098602133,
      "flos": 22346887570560.0,
      "grad_norm": 2.7222905177005496,
      "language_loss": 0.69978809,
      "learning_rate": 3.7513510225656132e-06,
      "loss": 0.72161043,
      "num_input_tokens_seen": 66670045,
      "step": 3087,
      "time_per_iteration": 3.920544385910034
    },
    {
      "auxiliary_loss_clip": 0.01111702,
      "auxiliary_loss_mlp": 0.0104868,
      "balance_loss_clip": 1.04850507,
      "balance_loss_mlp": 1.03003573,
      "epoch": 0.18566060423868933,
      "flos": 17748526308480.0,
      "grad_norm": 2.2290092395405114,
      "language_loss": 0.72391838,
      "learning_rate": 3.7511629183195764e-06,
      "loss": 0.7455222,
      "num_input_tokens_seen": 66688790,
      "step": 3088,
      "time_per_iteration": 2.5305416584014893
    },
    {
      "auxiliary_loss_clip": 0.01125026,
      "auxiliary_loss_mlp": 0.01042972,
      "balance_loss_clip": 1.04864609,
      "balance_loss_mlp": 1.02584159,
      "epoch": 0.1857207274913573,
      "flos": 24677274142080.0,
      "grad_norm": 1.9347994149812453,
      "language_loss": 0.92510533,
      "learning_rate": 3.7509747476692663e-06,
      "loss": 0.94678533,
      "num_input_tokens_seen": 66708090,
      "step": 3089,
      "time_per_iteration": 2.5448367595672607
    },
    {
      "auxiliary_loss_clip": 0.01099744,
      "auxiliary_loss_mlp": 0.01041247,
      "balance_loss_clip": 1.04734337,
      "balance_loss_mlp": 1.02311504,
      "epoch": 0.18578085074402526,
      "flos": 28147825198080.0,
      "grad_norm": 2.9771058074317103,
      "language_loss": 0.57897866,
      "learning_rate": 3.7507865106218176e-06,
      "loss": 0.60038859,
      "num_input_tokens_seen": 66727320,
      "step": 3090,
      "time_per_iteration": 4.047714471817017
    },
    {
      "auxiliary_loss_clip": 0.01123541,
      "auxiliary_loss_mlp": 0.01043226,
      "balance_loss_clip": 1.04704332,
      "balance_loss_mlp": 1.02545214,
      "epoch": 0.18584097399669322,
      "flos": 23951878980480.0,
      "grad_norm": 1.7428998132119318,
      "language_loss": 0.81814951,
      "learning_rate": 3.7505982071843695e-06,
      "loss": 0.83981717,
      "num_input_tokens_seen": 66747505,
      "step": 3091,
      "time_per_iteration": 2.548675298690796
    },
    {
      "auxiliary_loss_clip": 0.01100407,
      "auxiliary_loss_mlp": 0.01047538,
      "balance_loss_clip": 1.05060649,
      "balance_loss_mlp": 1.02960908,
      "epoch": 0.18590109724936119,
      "flos": 17201678676480.0,
      "grad_norm": 2.824681383892353,
      "language_loss": 0.84140384,
      "learning_rate": 3.7504098373640617e-06,
      "loss": 0.86288333,
      "num_input_tokens_seen": 66766425,
      "step": 3092,
      "time_per_iteration": 2.594717264175415
    },
    {
      "auxiliary_loss_clip": 0.0113232,
      "auxiliary_loss_mlp": 0.01044708,
      "balance_loss_clip": 1.05010223,
      "balance_loss_mlp": 1.02700543,
      "epoch": 0.18596122050202915,
      "flos": 17234644383360.0,
      "grad_norm": 2.4554560254813222,
      "language_loss": 0.93363714,
      "learning_rate": 3.750221401168038e-06,
      "loss": 0.95540738,
      "num_input_tokens_seen": 66781130,
      "step": 3093,
      "time_per_iteration": 2.483088254928589
    },
    {
      "auxiliary_loss_clip": 0.01125791,
      "auxiliary_loss_mlp": 0.01039276,
      "balance_loss_clip": 1.0513432,
      "balance_loss_mlp": 1.0212276,
      "epoch": 0.18602134375469712,
      "flos": 19020733188480.0,
      "grad_norm": 1.9381110238798234,
      "language_loss": 0.77577317,
      "learning_rate": 3.750032898603443e-06,
      "loss": 0.79742384,
      "num_input_tokens_seen": 66797535,
      "step": 3094,
      "time_per_iteration": 2.520418643951416
    },
    {
      "auxiliary_loss_clip": 0.01090066,
      "auxiliary_loss_mlp": 0.0104444,
      "balance_loss_clip": 1.04865122,
      "balance_loss_mlp": 1.02853799,
      "epoch": 0.1860814670073651,
      "flos": 50950094417280.0,
      "grad_norm": 1.6883284309125253,
      "language_loss": 0.69915903,
      "learning_rate": 3.749844329677425e-06,
      "loss": 0.72050411,
      "num_input_tokens_seen": 66821720,
      "step": 3095,
      "time_per_iteration": 2.8828094005584717
    },
    {
      "auxiliary_loss_clip": 0.01105937,
      "auxiliary_loss_mlp": 0.01047806,
      "balance_loss_clip": 1.04636264,
      "balance_loss_mlp": 1.02764726,
      "epoch": 0.18614159026003307,
      "flos": 19390972625280.0,
      "grad_norm": 2.701053987258419,
      "language_loss": 0.80605948,
      "learning_rate": 3.749655694397135e-06,
      "loss": 0.82759684,
      "num_input_tokens_seen": 66839060,
      "step": 3096,
      "time_per_iteration": 2.5307981967926025
    },
    {
      "auxiliary_loss_clip": 0.01142109,
      "auxiliary_loss_mlp": 0.01044447,
      "balance_loss_clip": 1.05121088,
      "balance_loss_mlp": 1.02619576,
      "epoch": 0.18620171351270104,
      "flos": 21798782962560.0,
      "grad_norm": 2.0599506016461744,
      "language_loss": 0.74928552,
      "learning_rate": 3.7494669927697255e-06,
      "loss": 0.77115113,
      "num_input_tokens_seen": 66857760,
      "step": 3097,
      "time_per_iteration": 2.5017731189727783
    },
    {
      "auxiliary_loss_clip": 0.01126968,
      "auxiliary_loss_mlp": 0.01040437,
      "balance_loss_clip": 1.051736,
      "balance_loss_mlp": 1.02329433,
      "epoch": 0.186261836765369,
      "flos": 16362877299840.0,
      "grad_norm": 3.076231370208987,
      "language_loss": 0.66519624,
      "learning_rate": 3.749278224802352e-06,
      "loss": 0.68687028,
      "num_input_tokens_seen": 66876460,
      "step": 3098,
      "time_per_iteration": 2.5133728981018066
    },
    {
      "auxiliary_loss_clip": 0.01155515,
      "auxiliary_loss_mlp": 0.01047103,
      "balance_loss_clip": 1.05186284,
      "balance_loss_mlp": 1.02750468,
      "epoch": 0.18632196001803697,
      "flos": 23370054480000.0,
      "grad_norm": 1.6806715123253373,
      "language_loss": 0.69731337,
      "learning_rate": 3.7490893905021733e-06,
      "loss": 0.71933955,
      "num_input_tokens_seen": 66897960,
      "step": 3099,
      "time_per_iteration": 2.500502586364746
    },
    {
      "auxiliary_loss_clip": 0.01145787,
      "auxiliary_loss_mlp": 0.01051155,
      "balance_loss_clip": 1.05427432,
      "balance_loss_mlp": 1.03233242,
      "epoch": 0.18638208327070493,
      "flos": 22492002516480.0,
      "grad_norm": 1.5751821065655003,
      "language_loss": 0.72072864,
      "learning_rate": 3.7489004898763494e-06,
      "loss": 0.74269807,
      "num_input_tokens_seen": 66917675,
      "step": 3100,
      "time_per_iteration": 2.5126194953918457
    },
    {
      "auxiliary_loss_clip": 0.01131234,
      "auxiliary_loss_mlp": 0.01049043,
      "balance_loss_clip": 1.0506202,
      "balance_loss_mlp": 1.0305419,
      "epoch": 0.18644220652337293,
      "flos": 29165245931520.0,
      "grad_norm": 2.1709066462401374,
      "language_loss": 0.80442011,
      "learning_rate": 3.7487115229320444e-06,
      "loss": 0.82622284,
      "num_input_tokens_seen": 66936000,
      "step": 3101,
      "time_per_iteration": 2.600951671600342
    },
    {
      "auxiliary_loss_clip": 0.01106484,
      "auxiliary_loss_mlp": 0.01041968,
      "balance_loss_clip": 1.05099702,
      "balance_loss_mlp": 1.02502847,
      "epoch": 0.1865023297760409,
      "flos": 24243796811520.0,
      "grad_norm": 12.509489298863416,
      "language_loss": 0.77376753,
      "learning_rate": 3.7485224896764222e-06,
      "loss": 0.79525208,
      "num_input_tokens_seen": 66955700,
      "step": 3102,
      "time_per_iteration": 2.59336519241333
    },
    {
      "auxiliary_loss_clip": 0.011459,
      "auxiliary_loss_mlp": 0.01039978,
      "balance_loss_clip": 1.05109727,
      "balance_loss_mlp": 1.02238226,
      "epoch": 0.18656245302870886,
      "flos": 19128716449920.0,
      "grad_norm": 2.648559357207612,
      "language_loss": 0.76482046,
      "learning_rate": 3.7483333901166525e-06,
      "loss": 0.78667921,
      "num_input_tokens_seen": 66972815,
      "step": 3103,
      "time_per_iteration": 2.502370595932007
    },
    {
      "auxiliary_loss_clip": 0.01130083,
      "auxiliary_loss_mlp": 0.0104286,
      "balance_loss_clip": 1.05139387,
      "balance_loss_mlp": 1.02502596,
      "epoch": 0.18662257628137682,
      "flos": 17786088956160.0,
      "grad_norm": 2.1434524373939983,
      "language_loss": 0.79194182,
      "learning_rate": 3.7481442242599054e-06,
      "loss": 0.81367123,
      "num_input_tokens_seen": 66992280,
      "step": 3104,
      "time_per_iteration": 2.521860361099243
    },
    {
      "auxiliary_loss_clip": 0.01104575,
      "auxiliary_loss_mlp": 0.01048774,
      "balance_loss_clip": 1.05004692,
      "balance_loss_mlp": 1.02918792,
      "epoch": 0.1866826995340448,
      "flos": 24024382583040.0,
      "grad_norm": 1.8484207999780256,
      "language_loss": 0.85143232,
      "learning_rate": 3.747954992113354e-06,
      "loss": 0.87296581,
      "num_input_tokens_seen": 67012220,
      "step": 3105,
      "time_per_iteration": 2.6337544918060303
    },
    {
      "auxiliary_loss_clip": 0.01122488,
      "auxiliary_loss_mlp": 0.01041457,
      "balance_loss_clip": 1.05227411,
      "balance_loss_mlp": 1.02221656,
      "epoch": 0.18674282278671275,
      "flos": 26141244756480.0,
      "grad_norm": 1.9537184223659891,
      "language_loss": 0.87117696,
      "learning_rate": 3.7477656936841742e-06,
      "loss": 0.89281631,
      "num_input_tokens_seen": 67032030,
      "step": 3106,
      "time_per_iteration": 2.6148219108581543
    },
    {
      "auxiliary_loss_clip": 0.01146064,
      "auxiliary_loss_mlp": 0.01041221,
      "balance_loss_clip": 1.05292046,
      "balance_loss_mlp": 1.02308953,
      "epoch": 0.18680294603938072,
      "flos": 19201938324480.0,
      "grad_norm": 2.0598563380315897,
      "language_loss": 0.78379792,
      "learning_rate": 3.7475763289795445e-06,
      "loss": 0.8056708,
      "num_input_tokens_seen": 67048920,
      "step": 3107,
      "time_per_iteration": 2.506296157836914
    },
    {
      "auxiliary_loss_clip": 0.01147444,
      "auxiliary_loss_mlp": 0.01050681,
      "balance_loss_clip": 1.05191696,
      "balance_loss_mlp": 1.03073752,
      "epoch": 0.1868630692920487,
      "flos": 28544889116160.0,
      "grad_norm": 2.6609821942071554,
      "language_loss": 0.74351448,
      "learning_rate": 3.7473868980066446e-06,
      "loss": 0.76549578,
      "num_input_tokens_seen": 67068645,
      "step": 3108,
      "time_per_iteration": 2.5623955726623535
    },
    {
      "auxiliary_loss_clip": 0.01106227,
      "auxiliary_loss_mlp": 0.01044961,
      "balance_loss_clip": 1.05180824,
      "balance_loss_mlp": 1.02580428,
      "epoch": 0.18692319254471668,
      "flos": 17238020261760.0,
      "grad_norm": 1.621525458217393,
      "language_loss": 0.74324751,
      "learning_rate": 3.747197400772658e-06,
      "loss": 0.76475936,
      "num_input_tokens_seen": 67087075,
      "step": 3109,
      "time_per_iteration": 2.5804126262664795
    },
    {
      "auxiliary_loss_clip": 0.01146294,
      "auxiliary_loss_mlp": 0.0104539,
      "balance_loss_clip": 1.05607414,
      "balance_loss_mlp": 1.02667451,
      "epoch": 0.18698331579738464,
      "flos": 23185186156800.0,
      "grad_norm": 1.50832115242426,
      "language_loss": 0.84388179,
      "learning_rate": 3.747007837284772e-06,
      "loss": 0.86579859,
      "num_input_tokens_seen": 67108040,
      "step": 3110,
      "time_per_iteration": 2.5255074501037598
    },
    {
      "auxiliary_loss_clip": 0.01146019,
      "auxiliary_loss_mlp": 0.01046129,
      "balance_loss_clip": 1.05870283,
      "balance_loss_mlp": 1.02710366,
      "epoch": 0.1870434390500526,
      "flos": 25516721963520.0,
      "grad_norm": 1.585798863719456,
      "language_loss": 0.84511781,
      "learning_rate": 3.7468182075501737e-06,
      "loss": 0.86703926,
      "num_input_tokens_seen": 67127605,
      "step": 3111,
      "time_per_iteration": 2.5558245182037354
    },
    {
      "auxiliary_loss_clip": 0.01129778,
      "auxiliary_loss_mlp": 0.01042266,
      "balance_loss_clip": 1.05585754,
      "balance_loss_mlp": 1.0244081,
      "epoch": 0.18710356230272057,
      "flos": 19500823393920.0,
      "grad_norm": 1.8508455899820442,
      "language_loss": 0.76561999,
      "learning_rate": 3.7466285115760536e-06,
      "loss": 0.7873404,
      "num_input_tokens_seen": 67145785,
      "step": 3112,
      "time_per_iteration": 2.5254390239715576
    },
    {
      "auxiliary_loss_clip": 0.01145501,
      "auxiliary_loss_mlp": 0.01043759,
      "balance_loss_clip": 1.05343652,
      "balance_loss_mlp": 1.02602077,
      "epoch": 0.18716368555538854,
      "flos": 26760847386240.0,
      "grad_norm": 1.8611499340789506,
      "language_loss": 0.64874673,
      "learning_rate": 3.7464387493696046e-06,
      "loss": 0.6706394,
      "num_input_tokens_seen": 67165930,
      "step": 3113,
      "time_per_iteration": 2.5528404712677
    },
    {
      "auxiliary_loss_clip": 0.01154639,
      "auxiliary_loss_mlp": 0.01043146,
      "balance_loss_clip": 1.05802035,
      "balance_loss_mlp": 1.02440596,
      "epoch": 0.1872238088080565,
      "flos": 25189827264000.0,
      "grad_norm": 3.214071877337392,
      "language_loss": 0.81084085,
      "learning_rate": 3.746248920938024e-06,
      "loss": 0.83281869,
      "num_input_tokens_seen": 67185830,
      "step": 3114,
      "time_per_iteration": 2.5382561683654785
    },
    {
      "auxiliary_loss_clip": 0.01110655,
      "auxiliary_loss_mlp": 0.01050046,
      "balance_loss_clip": 1.05757356,
      "balance_loss_mlp": 1.02993512,
      "epoch": 0.1872839320607245,
      "flos": 24134305178880.0,
      "grad_norm": 16.832589873260904,
      "language_loss": 0.57366586,
      "learning_rate": 3.74605902628851e-06,
      "loss": 0.5952729,
      "num_input_tokens_seen": 67206930,
      "step": 3115,
      "time_per_iteration": 2.632845163345337
    },
    {
      "auxiliary_loss_clip": 0.0111921,
      "auxiliary_loss_mlp": 0.01056416,
      "balance_loss_clip": 1.04986918,
      "balance_loss_mlp": 1.03568602,
      "epoch": 0.18734405531339246,
      "flos": 21173793292800.0,
      "grad_norm": 2.2715935179971884,
      "language_loss": 0.71301734,
      "learning_rate": 3.745869065428261e-06,
      "loss": 0.73477364,
      "num_input_tokens_seen": 67226290,
      "step": 3116,
      "time_per_iteration": 2.5672287940979004
    },
    {
      "auxiliary_loss_clip": 0.01152633,
      "auxiliary_loss_mlp": 0.01035792,
      "balance_loss_clip": 1.05427766,
      "balance_loss_mlp": 1.01800644,
      "epoch": 0.18740417856606043,
      "flos": 17237697039360.0,
      "grad_norm": 2.2712902489541302,
      "language_loss": 0.79156202,
      "learning_rate": 3.7456790383644833e-06,
      "loss": 0.81344628,
      "num_input_tokens_seen": 67244410,
      "step": 3117,
      "time_per_iteration": 2.467670202255249
    },
    {
      "auxiliary_loss_clip": 0.01137376,
      "auxiliary_loss_mlp": 0.010459,
      "balance_loss_clip": 1.05664897,
      "balance_loss_mlp": 1.02741027,
      "epoch": 0.1874643018187284,
      "flos": 32558049999360.0,
      "grad_norm": 1.810810998212889,
      "language_loss": 0.84140038,
      "learning_rate": 3.745488945104381e-06,
      "loss": 0.86323315,
      "num_input_tokens_seen": 67264470,
      "step": 3118,
      "time_per_iteration": 2.6290299892425537
    },
    {
      "auxiliary_loss_clip": 0.01149504,
      "auxiliary_loss_mlp": 0.01047661,
      "balance_loss_clip": 1.05797732,
      "balance_loss_mlp": 1.03048337,
      "epoch": 0.18752442507139636,
      "flos": 23258156636160.0,
      "grad_norm": 1.874953408214316,
      "language_loss": 0.76747,
      "learning_rate": 3.7452987856551636e-06,
      "loss": 0.78944159,
      "num_input_tokens_seen": 67284315,
      "step": 3119,
      "time_per_iteration": 2.5235116481781006
    },
    {
      "auxiliary_loss_clip": 0.01157087,
      "auxiliary_loss_mlp": 0.01048192,
      "balance_loss_clip": 1.05461001,
      "balance_loss_mlp": 1.03099012,
      "epoch": 0.18758454832406432,
      "flos": 21760933006080.0,
      "grad_norm": 1.773234312306949,
      "language_loss": 0.82316542,
      "learning_rate": 3.7451085600240406e-06,
      "loss": 0.84521818,
      "num_input_tokens_seen": 67302780,
      "step": 3120,
      "time_per_iteration": 2.4811439514160156
    },
    {
      "auxiliary_loss_clip": 0.01133081,
      "auxiliary_loss_mlp": 0.01036988,
      "balance_loss_clip": 1.05465388,
      "balance_loss_mlp": 1.02001297,
      "epoch": 0.1876446715767323,
      "flos": 29570210841600.0,
      "grad_norm": 2.1301678503823935,
      "language_loss": 0.84947604,
      "learning_rate": 3.7449182682182263e-06,
      "loss": 0.87117672,
      "num_input_tokens_seen": 67323405,
      "step": 3121,
      "time_per_iteration": 2.5861263275146484
    },
    {
      "auxiliary_loss_clip": 0.01102948,
      "auxiliary_loss_mlp": 0.01045671,
      "balance_loss_clip": 1.05325556,
      "balance_loss_mlp": 1.02740812,
      "epoch": 0.18770479482940028,
      "flos": 30339992234880.0,
      "grad_norm": 1.8660693236759498,
      "language_loss": 0.70085263,
      "learning_rate": 3.744727910244937e-06,
      "loss": 0.7223388,
      "num_input_tokens_seen": 67345800,
      "step": 3122,
      "time_per_iteration": 5.541229724884033
    },
    {
      "auxiliary_loss_clip": 0.0115375,
      "auxiliary_loss_mlp": 0.01043349,
      "balance_loss_clip": 1.05394304,
      "balance_loss_mlp": 1.02406061,
      "epoch": 0.18776491808206824,
      "flos": 14465357527680.0,
      "grad_norm": 2.313320283487551,
      "language_loss": 0.70765674,
      "learning_rate": 3.7445374861113905e-06,
      "loss": 0.72962767,
      "num_input_tokens_seen": 67363575,
      "step": 3123,
      "time_per_iteration": 2.4887890815734863
    },
    {
      "auxiliary_loss_clip": 0.01149016,
      "auxiliary_loss_mlp": 0.0104474,
      "balance_loss_clip": 1.05778861,
      "balance_loss_mlp": 1.02766967,
      "epoch": 0.1878250413347362,
      "flos": 24498547044480.0,
      "grad_norm": 2.2689133989347496,
      "language_loss": 0.74061,
      "learning_rate": 3.7443469958248066e-06,
      "loss": 0.76254761,
      "num_input_tokens_seen": 67381765,
      "step": 3124,
      "time_per_iteration": 2.547609329223633
    },
    {
      "auxiliary_loss_clip": 0.01158569,
      "auxiliary_loss_mlp": 0.01050467,
      "balance_loss_clip": 1.05605769,
      "balance_loss_mlp": 1.03122616,
      "epoch": 0.18788516458740417,
      "flos": 39786185692800.0,
      "grad_norm": 1.7449987423048428,
      "language_loss": 0.80309111,
      "learning_rate": 3.7441564393924106e-06,
      "loss": 0.82518148,
      "num_input_tokens_seen": 67405000,
      "step": 3125,
      "time_per_iteration": 2.655449628829956
    },
    {
      "auxiliary_loss_clip": 0.01054827,
      "auxiliary_loss_mlp": 0.0100223,
      "balance_loss_clip": 1.05279732,
      "balance_loss_mlp": 0.99908334,
      "epoch": 0.18794528784007214,
      "flos": 64699250664960.0,
      "grad_norm": 0.9485236786419264,
      "language_loss": 0.63581312,
      "learning_rate": 3.7439658168214273e-06,
      "loss": 0.65638369,
      "num_input_tokens_seen": 67467140,
      "step": 3126,
      "time_per_iteration": 4.642847299575806
    },
    {
      "auxiliary_loss_clip": 0.01129138,
      "auxiliary_loss_mlp": 0.01040972,
      "balance_loss_clip": 1.05679226,
      "balance_loss_mlp": 1.02365065,
      "epoch": 0.1880054110927401,
      "flos": 28622061486720.0,
      "grad_norm": 1.6152107839059424,
      "language_loss": 0.81562865,
      "learning_rate": 3.7437751281190857e-06,
      "loss": 0.83732975,
      "num_input_tokens_seen": 67487980,
      "step": 3127,
      "time_per_iteration": 2.6079609394073486
    },
    {
      "auxiliary_loss_clip": 0.01095939,
      "auxiliary_loss_mlp": 0.01001559,
      "balance_loss_clip": 1.05485034,
      "balance_loss_mlp": 0.99798292,
      "epoch": 0.1880655343454081,
      "flos": 64488958490880.0,
      "grad_norm": 0.7616735039825855,
      "language_loss": 0.61884987,
      "learning_rate": 3.7435843732926164e-06,
      "loss": 0.63982487,
      "num_input_tokens_seen": 67552500,
      "step": 3128,
      "time_per_iteration": 3.158031463623047
    },
    {
      "auxiliary_loss_clip": 0.01105159,
      "auxiliary_loss_mlp": 0.010516,
      "balance_loss_clip": 1.04524243,
      "balance_loss_mlp": 1.03115535,
      "epoch": 0.18812565759807606,
      "flos": 32124464928000.0,
      "grad_norm": 2.1351766080006276,
      "language_loss": 0.71328521,
      "learning_rate": 3.7433935523492536e-06,
      "loss": 0.73485279,
      "num_input_tokens_seen": 67573295,
      "step": 3129,
      "time_per_iteration": 4.139815092086792
    },
    {
      "auxiliary_loss_clip": 0.01153037,
      "auxiliary_loss_mlp": 0.01047578,
      "balance_loss_clip": 1.05226612,
      "balance_loss_mlp": 1.02936244,
      "epoch": 0.18818578085074403,
      "flos": 20624539449600.0,
      "grad_norm": 1.8722798737773303,
      "language_loss": 0.85386759,
      "learning_rate": 3.7432026652962314e-06,
      "loss": 0.87587374,
      "num_input_tokens_seen": 67590010,
      "step": 3130,
      "time_per_iteration": 2.483309268951416
    },
    {
      "auxiliary_loss_clip": 0.01105965,
      "auxiliary_loss_mlp": 0.01052706,
      "balance_loss_clip": 1.04861379,
      "balance_loss_mlp": 1.03434777,
      "epoch": 0.188245904103412,
      "flos": 28840506048000.0,
      "grad_norm": 1.9349992610665965,
      "language_loss": 0.76662588,
      "learning_rate": 3.7430117121407897e-06,
      "loss": 0.7882126,
      "num_input_tokens_seen": 67611110,
      "step": 3131,
      "time_per_iteration": 2.6089823246002197
    },
    {
      "auxiliary_loss_clip": 0.01121369,
      "auxiliary_loss_mlp": 0.01051756,
      "balance_loss_clip": 1.05152321,
      "balance_loss_mlp": 1.03312349,
      "epoch": 0.18830602735607996,
      "flos": 29420319386880.0,
      "grad_norm": 2.739358372754681,
      "language_loss": 0.81386244,
      "learning_rate": 3.74282069289017e-06,
      "loss": 0.8355937,
      "num_input_tokens_seen": 67631990,
      "step": 3132,
      "time_per_iteration": 2.6457366943359375
    },
    {
      "auxiliary_loss_clip": 0.01093236,
      "auxiliary_loss_mlp": 0.00861407,
      "balance_loss_clip": 1.05040884,
      "balance_loss_mlp": 1.12681246,
      "epoch": 0.18836615060874792,
      "flos": 28872933050880.0,
      "grad_norm": 2.429540064108657,
      "language_loss": 0.80056357,
      "learning_rate": 3.742629607551614e-06,
      "loss": 0.82011002,
      "num_input_tokens_seen": 67650490,
      "step": 3133,
      "time_per_iteration": 2.6529829502105713
    },
    {
      "auxiliary_loss_clip": 0.01115145,
      "auxiliary_loss_mlp": 0.01057059,
      "balance_loss_clip": 1.04798937,
      "balance_loss_mlp": 1.03773499,
      "epoch": 0.18842627386141592,
      "flos": 22601673717120.0,
      "grad_norm": 3.218304039361362,
      "language_loss": 0.82768905,
      "learning_rate": 3.7424384561323698e-06,
      "loss": 0.84941113,
      "num_input_tokens_seen": 67668860,
      "step": 3134,
      "time_per_iteration": 2.5607805252075195
    },
    {
      "auxiliary_loss_clip": 0.01125933,
      "auxiliary_loss_mlp": 0.01055015,
      "balance_loss_clip": 1.04672325,
      "balance_loss_mlp": 1.0367645,
      "epoch": 0.18848639711408388,
      "flos": 24573600512640.0,
      "grad_norm": 1.8410570938681095,
      "language_loss": 0.83068079,
      "learning_rate": 3.742247238639684e-06,
      "loss": 0.85249019,
      "num_input_tokens_seen": 67690220,
      "step": 3135,
      "time_per_iteration": 2.588998556137085
    },
    {
      "auxiliary_loss_clip": 0.01137759,
      "auxiliary_loss_mlp": 0.0104639,
      "balance_loss_clip": 1.05010724,
      "balance_loss_mlp": 1.02879429,
      "epoch": 0.18854652036675185,
      "flos": 34166920078080.0,
      "grad_norm": 2.3729982336318973,
      "language_loss": 0.78126276,
      "learning_rate": 3.7420559550808083e-06,
      "loss": 0.80310428,
      "num_input_tokens_seen": 67709820,
      "step": 3136,
      "time_per_iteration": 2.59486722946167
    },
    {
      "auxiliary_loss_clip": 0.01135105,
      "auxiliary_loss_mlp": 0.01045785,
      "balance_loss_clip": 1.05540645,
      "balance_loss_mlp": 1.02753437,
      "epoch": 0.1886066436194198,
      "flos": 24200236592640.0,
      "grad_norm": 2.792536715879011,
      "language_loss": 0.80920219,
      "learning_rate": 3.741864605462996e-06,
      "loss": 0.83101106,
      "num_input_tokens_seen": 67729490,
      "step": 3137,
      "time_per_iteration": 2.55684757232666
    },
    {
      "auxiliary_loss_clip": 0.01154003,
      "auxiliary_loss_mlp": 0.01044509,
      "balance_loss_clip": 1.0529933,
      "balance_loss_mlp": 1.02817726,
      "epoch": 0.18866676687208778,
      "flos": 21251109317760.0,
      "grad_norm": 1.684333843490975,
      "language_loss": 0.80857658,
      "learning_rate": 3.741673189793504e-06,
      "loss": 0.83056176,
      "num_input_tokens_seen": 67749665,
      "step": 3138,
      "time_per_iteration": 2.4614830017089844
    },
    {
      "auxiliary_loss_clip": 0.01142649,
      "auxiliary_loss_mlp": 0.0105575,
      "balance_loss_clip": 1.04961538,
      "balance_loss_mlp": 1.03805912,
      "epoch": 0.18872689012475574,
      "flos": 37308673013760.0,
      "grad_norm": 2.1928299576356514,
      "language_loss": 0.637743,
      "learning_rate": 3.7414817080795896e-06,
      "loss": 0.65972698,
      "num_input_tokens_seen": 67776230,
      "step": 3139,
      "time_per_iteration": 2.6656370162963867
    },
    {
      "auxiliary_loss_clip": 0.01147128,
      "auxiliary_loss_mlp": 0.01041451,
      "balance_loss_clip": 1.04748297,
      "balance_loss_mlp": 1.02390301,
      "epoch": 0.1887870133774237,
      "flos": 21652303299840.0,
      "grad_norm": 2.390915533982515,
      "language_loss": 0.71573037,
      "learning_rate": 3.741290160328514e-06,
      "loss": 0.73761618,
      "num_input_tokens_seen": 67795080,
      "step": 3140,
      "time_per_iteration": 2.448986768722534
    },
    {
      "auxiliary_loss_clip": 0.01149948,
      "auxiliary_loss_mlp": 0.01044863,
      "balance_loss_clip": 1.04775953,
      "balance_loss_mlp": 1.02615869,
      "epoch": 0.1888471366300917,
      "flos": 15924659374080.0,
      "grad_norm": 2.837267779820146,
      "language_loss": 0.87176251,
      "learning_rate": 3.7410985465475412e-06,
      "loss": 0.89371061,
      "num_input_tokens_seen": 67813110,
      "step": 3141,
      "time_per_iteration": 2.4590044021606445
    },
    {
      "auxiliary_loss_clip": 0.01129874,
      "auxiliary_loss_mlp": 0.01042201,
      "balance_loss_clip": 1.05082643,
      "balance_loss_mlp": 1.02291274,
      "epoch": 0.18890725988275966,
      "flos": 18551955767040.0,
      "grad_norm": 1.9230278645302414,
      "language_loss": 0.77511013,
      "learning_rate": 3.7409068667439378e-06,
      "loss": 0.79683083,
      "num_input_tokens_seen": 67831070,
      "step": 3142,
      "time_per_iteration": 2.4964609146118164
    },
    {
      "auxiliary_loss_clip": 0.01124937,
      "auxiliary_loss_mlp": 0.01039824,
      "balance_loss_clip": 1.0519954,
      "balance_loss_mlp": 1.02396965,
      "epoch": 0.18896738313542763,
      "flos": 28840865184000.0,
      "grad_norm": 1.7161157722232474,
      "language_loss": 0.79051793,
      "learning_rate": 3.740715120924971e-06,
      "loss": 0.8121655,
      "num_input_tokens_seen": 67852170,
      "step": 3143,
      "time_per_iteration": 2.604966163635254
    },
    {
      "auxiliary_loss_clip": 0.01117021,
      "auxiliary_loss_mlp": 0.01045017,
      "balance_loss_clip": 1.04959381,
      "balance_loss_mlp": 1.02718329,
      "epoch": 0.1890275063880956,
      "flos": 22412747157120.0,
      "grad_norm": 2.296211626538829,
      "language_loss": 0.71433902,
      "learning_rate": 3.740523309097912e-06,
      "loss": 0.73595941,
      "num_input_tokens_seen": 67869945,
      "step": 3144,
      "time_per_iteration": 2.5658531188964844
    },
    {
      "auxiliary_loss_clip": 0.01121426,
      "auxiliary_loss_mlp": 0.01047033,
      "balance_loss_clip": 1.04734325,
      "balance_loss_mlp": 1.02806687,
      "epoch": 0.18908762964076356,
      "flos": 24243904552320.0,
      "grad_norm": 2.3934980820132044,
      "language_loss": 0.73537242,
      "learning_rate": 3.7403314312700356e-06,
      "loss": 0.75705707,
      "num_input_tokens_seen": 67890240,
      "step": 3145,
      "time_per_iteration": 2.579765796661377
    },
    {
      "auxiliary_loss_clip": 0.01110249,
      "auxiliary_loss_mlp": 0.01042109,
      "balance_loss_clip": 1.04836643,
      "balance_loss_mlp": 1.02463293,
      "epoch": 0.18914775289343153,
      "flos": 16982910892800.0,
      "grad_norm": 2.287367767842877,
      "language_loss": 0.76335245,
      "learning_rate": 3.740139487448616e-06,
      "loss": 0.78487599,
      "num_input_tokens_seen": 67907825,
      "step": 3146,
      "time_per_iteration": 2.5484111309051514
    },
    {
      "auxiliary_loss_clip": 0.01094328,
      "auxiliary_loss_mlp": 0.01048161,
      "balance_loss_clip": 1.04731989,
      "balance_loss_mlp": 1.02873015,
      "epoch": 0.1892078761460995,
      "flos": 21543781334400.0,
      "grad_norm": 1.801036692118683,
      "language_loss": 0.78431177,
      "learning_rate": 3.7399474776409326e-06,
      "loss": 0.80573666,
      "num_input_tokens_seen": 67926670,
      "step": 3147,
      "time_per_iteration": 2.594238758087158
    },
    {
      "auxiliary_loss_clip": 0.01139248,
      "auxiliary_loss_mlp": 0.01043541,
      "balance_loss_clip": 1.05026007,
      "balance_loss_mlp": 1.02651751,
      "epoch": 0.18926799939876748,
      "flos": 23001538896000.0,
      "grad_norm": 2.7881785024500174,
      "language_loss": 0.67721522,
      "learning_rate": 3.739755401854267e-06,
      "loss": 0.69904315,
      "num_input_tokens_seen": 67943645,
      "step": 3148,
      "time_per_iteration": 2.5157783031463623
    },
    {
      "auxiliary_loss_clip": 0.0111221,
      "auxiliary_loss_mlp": 0.01036964,
      "balance_loss_clip": 1.0487802,
      "balance_loss_mlp": 1.02007198,
      "epoch": 0.18932812265143545,
      "flos": 22273019251200.0,
      "grad_norm": 4.335875345147421,
      "language_loss": 0.75466877,
      "learning_rate": 3.739563260095902e-06,
      "loss": 0.77616054,
      "num_input_tokens_seen": 67962345,
      "step": 3149,
      "time_per_iteration": 2.5807857513427734
    },
    {
      "auxiliary_loss_clip": 0.01131255,
      "auxiliary_loss_mlp": 0.01041833,
      "balance_loss_clip": 1.0532701,
      "balance_loss_mlp": 1.02538264,
      "epoch": 0.1893882459041034,
      "flos": 18624423456000.0,
      "grad_norm": 2.6745940393010486,
      "language_loss": 0.80833673,
      "learning_rate": 3.7393710523731245e-06,
      "loss": 0.83006763,
      "num_input_tokens_seen": 67979760,
      "step": 3150,
      "time_per_iteration": 2.5003201961517334
    },
    {
      "auxiliary_loss_clip": 0.01133646,
      "auxiliary_loss_mlp": 0.01048665,
      "balance_loss_clip": 1.05087316,
      "balance_loss_mlp": 1.03124869,
      "epoch": 0.18944836915677138,
      "flos": 22892981016960.0,
      "grad_norm": 2.192040602204963,
      "language_loss": 0.85416484,
      "learning_rate": 3.7391787786932215e-06,
      "loss": 0.87598795,
      "num_input_tokens_seen": 67996895,
      "step": 3151,
      "time_per_iteration": 2.515146493911743
    },
    {
      "auxiliary_loss_clip": 0.01115984,
      "auxiliary_loss_mlp": 0.01046396,
      "balance_loss_clip": 1.0494597,
      "balance_loss_mlp": 1.02882504,
      "epoch": 0.18950849240943934,
      "flos": 26796542526720.0,
      "grad_norm": 2.2682169020949208,
      "language_loss": 0.74344379,
      "learning_rate": 3.7389864390634857e-06,
      "loss": 0.76506758,
      "num_input_tokens_seen": 68018365,
      "step": 3152,
      "time_per_iteration": 2.620814085006714
    },
    {
      "auxiliary_loss_clip": 0.01122908,
      "auxiliary_loss_mlp": 0.01040603,
      "balance_loss_clip": 1.05359054,
      "balance_loss_mlp": 1.02230477,
      "epoch": 0.1895686156621073,
      "flos": 24971239048320.0,
      "grad_norm": 2.1571687451511035,
      "language_loss": 0.75515741,
      "learning_rate": 3.738794033491209e-06,
      "loss": 0.77679253,
      "num_input_tokens_seen": 68037985,
      "step": 3153,
      "time_per_iteration": 2.60398268699646
    },
    {
      "auxiliary_loss_clip": 0.01156851,
      "auxiliary_loss_mlp": 0.01046606,
      "balance_loss_clip": 1.05503583,
      "balance_loss_mlp": 1.02890348,
      "epoch": 0.1896287389147753,
      "flos": 21944544353280.0,
      "grad_norm": 1.991562524014051,
      "language_loss": 0.79263568,
      "learning_rate": 3.7386015619836887e-06,
      "loss": 0.81467026,
      "num_input_tokens_seen": 68057975,
      "step": 3154,
      "time_per_iteration": 2.471313238143921
    },
    {
      "auxiliary_loss_clip": 0.01113229,
      "auxiliary_loss_mlp": 0.01051445,
      "balance_loss_clip": 1.04617906,
      "balance_loss_mlp": 1.03215694,
      "epoch": 0.18968886216744327,
      "flos": 18179058723840.0,
      "grad_norm": 2.459614285361488,
      "language_loss": 0.7260896,
      "learning_rate": 3.738409024548223e-06,
      "loss": 0.74773633,
      "num_input_tokens_seen": 68074175,
      "step": 3155,
      "time_per_iteration": 2.543168783187866
    },
    {
      "auxiliary_loss_clip": 0.01125782,
      "auxiliary_loss_mlp": 0.01040861,
      "balance_loss_clip": 1.05263019,
      "balance_loss_mlp": 1.02346873,
      "epoch": 0.18974898542011123,
      "flos": 20412487509120.0,
      "grad_norm": 1.9262185060066148,
      "language_loss": 0.74085855,
      "learning_rate": 3.7382164211921136e-06,
      "loss": 0.76252496,
      "num_input_tokens_seen": 68095230,
      "step": 3156,
      "time_per_iteration": 2.5644567012786865
    },
    {
      "auxiliary_loss_clip": 0.01154657,
      "auxiliary_loss_mlp": 0.01036422,
      "balance_loss_clip": 1.05285203,
      "balance_loss_mlp": 1.02014995,
      "epoch": 0.1898091086727792,
      "flos": 23985024255360.0,
      "grad_norm": 2.0136307409696945,
      "language_loss": 0.6851272,
      "learning_rate": 3.7380237519226623e-06,
      "loss": 0.70703804,
      "num_input_tokens_seen": 68113805,
      "step": 3157,
      "time_per_iteration": 2.513479709625244
    },
    {
      "auxiliary_loss_clip": 0.01116053,
      "auxiliary_loss_mlp": 0.01042306,
      "balance_loss_clip": 1.05185533,
      "balance_loss_mlp": 1.0250448,
      "epoch": 0.18986923192544716,
      "flos": 27637067756160.0,
      "grad_norm": 1.87248048120988,
      "language_loss": 0.79659534,
      "learning_rate": 3.737831016747176e-06,
      "loss": 0.81817889,
      "num_input_tokens_seen": 68133190,
      "step": 3158,
      "time_per_iteration": 2.5977070331573486
    },
    {
      "auxiliary_loss_clip": 0.01159632,
      "auxiliary_loss_mlp": 0.01038758,
      "balance_loss_clip": 1.05497706,
      "balance_loss_mlp": 1.02059078,
      "epoch": 0.18992935517811513,
      "flos": 25484151306240.0,
      "grad_norm": 1.7985147850493108,
      "language_loss": 0.72232509,
      "learning_rate": 3.737638215672964e-06,
      "loss": 0.74430895,
      "num_input_tokens_seen": 68152330,
      "step": 3159,
      "time_per_iteration": 2.520251750946045
    },
    {
      "auxiliary_loss_clip": 0.0114881,
      "auxiliary_loss_mlp": 0.01043493,
      "balance_loss_clip": 1.05892944,
      "balance_loss_mlp": 1.02518213,
      "epoch": 0.1899894784307831,
      "flos": 17420805596160.0,
      "grad_norm": 2.052464138762349,
      "language_loss": 0.84915459,
      "learning_rate": 3.7374453487073366e-06,
      "loss": 0.87107766,
      "num_input_tokens_seen": 68170185,
      "step": 3160,
      "time_per_iteration": 3.872096300125122
    },
    {
      "auxiliary_loss_clip": 0.01134257,
      "auxiliary_loss_mlp": 0.01043575,
      "balance_loss_clip": 1.05487597,
      "balance_loss_mlp": 1.0267663,
      "epoch": 0.19004960168345109,
      "flos": 27492240119040.0,
      "grad_norm": 1.9732324573788618,
      "language_loss": 0.738029,
      "learning_rate": 3.7372524158576074e-06,
      "loss": 0.75980723,
      "num_input_tokens_seen": 68191665,
      "step": 3161,
      "time_per_iteration": 3.9717118740081787
    },
    {
      "auxiliary_loss_clip": 0.01134354,
      "auxiliary_loss_mlp": 0.01049599,
      "balance_loss_clip": 1.04970181,
      "balance_loss_mlp": 1.03038263,
      "epoch": 0.19010972493611905,
      "flos": 38654676385920.0,
      "grad_norm": 2.584442324043975,
      "language_loss": 0.8078618,
      "learning_rate": 3.7370594171310926e-06,
      "loss": 0.8297013,
      "num_input_tokens_seen": 68214635,
      "step": 3162,
      "time_per_iteration": 2.6530282497406006
    },
    {
      "auxiliary_loss_clip": 0.01156605,
      "auxiliary_loss_mlp": 0.01038415,
      "balance_loss_clip": 1.0555445,
      "balance_loss_mlp": 1.02035534,
      "epoch": 0.19016984818878702,
      "flos": 19244744357760.0,
      "grad_norm": 2.213712549662201,
      "language_loss": 0.75113177,
      "learning_rate": 3.73686635253511e-06,
      "loss": 0.77308202,
      "num_input_tokens_seen": 68232150,
      "step": 3163,
      "time_per_iteration": 2.453986167907715
    },
    {
      "auxiliary_loss_clip": 0.01097874,
      "auxiliary_loss_mlp": 0.01047035,
      "balance_loss_clip": 1.05051947,
      "balance_loss_mlp": 1.02805734,
      "epoch": 0.19022997144145498,
      "flos": 37596891744000.0,
      "grad_norm": 5.5148660894389705,
      "language_loss": 0.74428195,
      "learning_rate": 3.736673222076982e-06,
      "loss": 0.76573104,
      "num_input_tokens_seen": 68253370,
      "step": 3164,
      "time_per_iteration": 2.740471363067627
    },
    {
      "auxiliary_loss_clip": 0.01139773,
      "auxiliary_loss_mlp": 0.01034987,
      "balance_loss_clip": 1.05293989,
      "balance_loss_mlp": 1.01729584,
      "epoch": 0.19029009469412295,
      "flos": 61530921665280.0,
      "grad_norm": 1.4886743482912517,
      "language_loss": 0.66815525,
      "learning_rate": 3.7364800257640313e-06,
      "loss": 0.68990284,
      "num_input_tokens_seen": 68278895,
      "step": 3165,
      "time_per_iteration": 4.264160871505737
    },
    {
      "auxiliary_loss_clip": 0.01144192,
      "auxiliary_loss_mlp": 0.01045761,
      "balance_loss_clip": 1.05406356,
      "balance_loss_mlp": 1.02685475,
      "epoch": 0.1903502179467909,
      "flos": 13954851480960.0,
      "grad_norm": 2.264479347608474,
      "language_loss": 0.74551225,
      "learning_rate": 3.7362867636035835e-06,
      "loss": 0.76741183,
      "num_input_tokens_seen": 68294880,
      "step": 3166,
      "time_per_iteration": 2.49582839012146
    },
    {
      "auxiliary_loss_clip": 0.01051974,
      "auxiliary_loss_mlp": 0.01031637,
      "balance_loss_clip": 1.04178309,
      "balance_loss_mlp": 1.02801287,
      "epoch": 0.1904103411994589,
      "flos": 66899641916160.0,
      "grad_norm": 0.7809703193964058,
      "language_loss": 0.50365973,
      "learning_rate": 3.736093435602968e-06,
      "loss": 0.52449584,
      "num_input_tokens_seen": 68359665,
      "step": 3167,
      "time_per_iteration": 4.57387375831604
    },
    {
      "auxiliary_loss_clip": 0.01134425,
      "auxiliary_loss_mlp": 0.01047349,
      "balance_loss_clip": 1.05223072,
      "balance_loss_mlp": 1.03005219,
      "epoch": 0.19047046445212687,
      "flos": 21908741472000.0,
      "grad_norm": 1.7251222301897158,
      "language_loss": 0.74446106,
      "learning_rate": 3.7359000417695156e-06,
      "loss": 0.7662788,
      "num_input_tokens_seen": 68378950,
      "step": 3168,
      "time_per_iteration": 2.5324606895446777
    },
    {
      "auxiliary_loss_clip": 0.0104654,
      "auxiliary_loss_mlp": 0.01026165,
      "balance_loss_clip": 1.04538429,
      "balance_loss_mlp": 1.02215958,
      "epoch": 0.19053058770479483,
      "flos": 59255156701440.0,
      "grad_norm": 0.8618994983497107,
      "language_loss": 0.60080254,
      "learning_rate": 3.73570658211056e-06,
      "loss": 0.62152958,
      "num_input_tokens_seen": 68434235,
      "step": 3169,
      "time_per_iteration": 3.084881067276001
    },
    {
      "auxiliary_loss_clip": 0.01100922,
      "auxiliary_loss_mlp": 0.01045677,
      "balance_loss_clip": 1.05237222,
      "balance_loss_mlp": 1.02772391,
      "epoch": 0.1905907109574628,
      "flos": 23951304362880.0,
      "grad_norm": 1.528538023908902,
      "language_loss": 0.78332877,
      "learning_rate": 3.735513056633436e-06,
      "loss": 0.80479479,
      "num_input_tokens_seen": 68453830,
      "step": 3170,
      "time_per_iteration": 2.667311429977417
    },
    {
      "auxiliary_loss_clip": 0.01137995,
      "auxiliary_loss_mlp": 0.01040602,
      "balance_loss_clip": 1.05046082,
      "balance_loss_mlp": 1.02337599,
      "epoch": 0.19065083421013077,
      "flos": 20812316774400.0,
      "grad_norm": 1.8022990796524612,
      "language_loss": 0.78557408,
      "learning_rate": 3.7353194653454834e-06,
      "loss": 0.80736005,
      "num_input_tokens_seen": 68473005,
      "step": 3171,
      "time_per_iteration": 2.5501315593719482
    },
    {
      "auxiliary_loss_clip": 0.01154214,
      "auxiliary_loss_mlp": 0.01039188,
      "balance_loss_clip": 1.0514214,
      "balance_loss_mlp": 1.02122295,
      "epoch": 0.19071095746279873,
      "flos": 31284981192960.0,
      "grad_norm": 2.518474228196192,
      "language_loss": 0.7850731,
      "learning_rate": 3.7351258082540426e-06,
      "loss": 0.80700713,
      "num_input_tokens_seen": 68493470,
      "step": 3172,
      "time_per_iteration": 2.6355881690979004
    },
    {
      "auxiliary_loss_clip": 0.01142465,
      "auxiliary_loss_mlp": 0.01049564,
      "balance_loss_clip": 1.0514698,
      "balance_loss_mlp": 1.03132463,
      "epoch": 0.1907710807154667,
      "flos": 14356117290240.0,
      "grad_norm": 1.6148597307678,
      "language_loss": 0.80158269,
      "learning_rate": 3.7349320853664576e-06,
      "loss": 0.82350302,
      "num_input_tokens_seen": 68511290,
      "step": 3173,
      "time_per_iteration": 2.508340835571289
    },
    {
      "auxiliary_loss_clip": 0.01115119,
      "auxiliary_loss_mlp": 0.00820695,
      "balance_loss_clip": 1.05007529,
      "balance_loss_mlp": 1.05415177,
      "epoch": 0.1908312039681347,
      "flos": 26907039740160.0,
      "grad_norm": 3.073745500484498,
      "language_loss": 0.78793609,
      "learning_rate": 3.7347382966900735e-06,
      "loss": 0.80729425,
      "num_input_tokens_seen": 68532575,
      "step": 3174,
      "time_per_iteration": 2.6058740615844727
    },
    {
      "auxiliary_loss_clip": 0.01105906,
      "auxiliary_loss_mlp": 0.01045613,
      "balance_loss_clip": 1.05221391,
      "balance_loss_mlp": 1.02706361,
      "epoch": 0.19089132722080265,
      "flos": 14494695960960.0,
      "grad_norm": 3.2278587820142004,
      "language_loss": 0.81073898,
      "learning_rate": 3.7345444422322395e-06,
      "loss": 0.83225423,
      "num_input_tokens_seen": 68548760,
      "step": 3175,
      "time_per_iteration": 2.589191436767578
    },
    {
      "auxiliary_loss_clip": 0.01079814,
      "auxiliary_loss_mlp": 0.01056029,
      "balance_loss_clip": 1.04832006,
      "balance_loss_mlp": 1.03687191,
      "epoch": 0.19095145047347062,
      "flos": 13952876232960.0,
      "grad_norm": 2.576244461817974,
      "language_loss": 0.85596657,
      "learning_rate": 3.7343505220003067e-06,
      "loss": 0.87732506,
      "num_input_tokens_seen": 68563100,
      "step": 3176,
      "time_per_iteration": 2.5945746898651123
    },
    {
      "auxiliary_loss_clip": 0.01135796,
      "auxiliary_loss_mlp": 0.01048391,
      "balance_loss_clip": 1.05574751,
      "balance_loss_mlp": 1.02893651,
      "epoch": 0.19101157372613858,
      "flos": 25301832848640.0,
      "grad_norm": 4.08661053896423,
      "language_loss": 0.80961251,
      "learning_rate": 3.7341565360016285e-06,
      "loss": 0.8314544,
      "num_input_tokens_seen": 68581650,
      "step": 3177,
      "time_per_iteration": 2.5872857570648193
    },
    {
      "auxiliary_loss_clip": 0.01122665,
      "auxiliary_loss_mlp": 0.01039702,
      "balance_loss_clip": 1.05095506,
      "balance_loss_mlp": 1.02158237,
      "epoch": 0.19107169697880655,
      "flos": 20558212986240.0,
      "grad_norm": 2.1730871014979685,
      "language_loss": 0.74718797,
      "learning_rate": 3.73396248424356e-06,
      "loss": 0.76881164,
      "num_input_tokens_seen": 68600360,
      "step": 3178,
      "time_per_iteration": 2.5505893230438232
    },
    {
      "auxiliary_loss_clip": 0.01141092,
      "auxiliary_loss_mlp": 0.01035392,
      "balance_loss_clip": 1.0502696,
      "balance_loss_mlp": 1.01815486,
      "epoch": 0.19113182023147451,
      "flos": 22163204396160.0,
      "grad_norm": 1.8404925008512711,
      "language_loss": 0.81425065,
      "learning_rate": 3.7337683667334606e-06,
      "loss": 0.83601546,
      "num_input_tokens_seen": 68617885,
      "step": 3179,
      "time_per_iteration": 2.5137524604797363
    },
    {
      "auxiliary_loss_clip": 0.01144295,
      "auxiliary_loss_mlp": 0.0104022,
      "balance_loss_clip": 1.05353594,
      "balance_loss_mlp": 1.02280366,
      "epoch": 0.19119194348414248,
      "flos": 18581796990720.0,
      "grad_norm": 4.669816937423525,
      "language_loss": 0.79197061,
      "learning_rate": 3.733574183478691e-06,
      "loss": 0.81381577,
      "num_input_tokens_seen": 68634550,
      "step": 3180,
      "time_per_iteration": 2.4949803352355957
    },
    {
      "auxiliary_loss_clip": 0.01128206,
      "auxiliary_loss_mlp": 0.01044132,
      "balance_loss_clip": 1.05099142,
      "balance_loss_mlp": 1.02493978,
      "epoch": 0.19125206673681047,
      "flos": 19026623018880.0,
      "grad_norm": 2.9737449335503987,
      "language_loss": 0.79144049,
      "learning_rate": 3.733379934486615e-06,
      "loss": 0.81316388,
      "num_input_tokens_seen": 68651895,
      "step": 3181,
      "time_per_iteration": 2.530320644378662
    },
    {
      "auxiliary_loss_clip": 0.01140117,
      "auxiliary_loss_mlp": 0.01046865,
      "balance_loss_clip": 1.05099607,
      "balance_loss_mlp": 1.02887654,
      "epoch": 0.19131218998947844,
      "flos": 21690153256320.0,
      "grad_norm": 1.9501726010691187,
      "language_loss": 0.73871809,
      "learning_rate": 3.7331856197645973e-06,
      "loss": 0.76058793,
      "num_input_tokens_seen": 68671500,
      "step": 3182,
      "time_per_iteration": 2.5142581462860107
    },
    {
      "auxiliary_loss_clip": 0.01124323,
      "auxiliary_loss_mlp": 0.01040924,
      "balance_loss_clip": 1.0515548,
      "balance_loss_mlp": 1.02261317,
      "epoch": 0.1913723132421464,
      "flos": 18442500048000.0,
      "grad_norm": 1.7390907652565806,
      "language_loss": 0.64848828,
      "learning_rate": 3.7329912393200084e-06,
      "loss": 0.67014074,
      "num_input_tokens_seen": 68690570,
      "step": 3183,
      "time_per_iteration": 2.5721189975738525
    },
    {
      "auxiliary_loss_clip": 0.0112752,
      "auxiliary_loss_mlp": 0.01045033,
      "balance_loss_clip": 1.04731596,
      "balance_loss_mlp": 1.02654326,
      "epoch": 0.19143243649481437,
      "flos": 27160102033920.0,
      "grad_norm": 3.1487972078892126,
      "language_loss": 0.73467326,
      "learning_rate": 3.7327967931602173e-06,
      "loss": 0.7563988,
      "num_input_tokens_seen": 68709735,
      "step": 3184,
      "time_per_iteration": 2.5792317390441895
    },
    {
      "auxiliary_loss_clip": 0.01121718,
      "auxiliary_loss_mlp": 0.01049602,
      "balance_loss_clip": 1.04897296,
      "balance_loss_mlp": 1.02918148,
      "epoch": 0.19149255974748233,
      "flos": 21718952985600.0,
      "grad_norm": 1.8936875711150327,
      "language_loss": 0.88044155,
      "learning_rate": 3.732602281292598e-06,
      "loss": 0.90215474,
      "num_input_tokens_seen": 68727565,
      "step": 3185,
      "time_per_iteration": 2.5387885570526123
    },
    {
      "auxiliary_loss_clip": 0.01150492,
      "auxiliary_loss_mlp": 0.01041556,
      "balance_loss_clip": 1.05008197,
      "balance_loss_mlp": 1.02323329,
      "epoch": 0.1915526830001503,
      "flos": 22963293889920.0,
      "grad_norm": 2.1384034525308677,
      "language_loss": 0.72902596,
      "learning_rate": 3.7324077037245267e-06,
      "loss": 0.7509464,
      "num_input_tokens_seen": 68748110,
      "step": 3186,
      "time_per_iteration": 2.483429193496704
    },
    {
      "auxiliary_loss_clip": 0.01131003,
      "auxiliary_loss_mlp": 0.01041689,
      "balance_loss_clip": 1.05495715,
      "balance_loss_mlp": 1.0213635,
      "epoch": 0.1916128062528183,
      "flos": 26140741966080.0,
      "grad_norm": 1.880594914617483,
      "language_loss": 0.83779466,
      "learning_rate": 3.7322130604633825e-06,
      "loss": 0.85952151,
      "num_input_tokens_seen": 68769765,
      "step": 3187,
      "time_per_iteration": 2.6088030338287354
    },
    {
      "auxiliary_loss_clip": 0.0106546,
      "auxiliary_loss_mlp": 0.01004489,
      "balance_loss_clip": 1.02995658,
      "balance_loss_mlp": 1.00177085,
      "epoch": 0.19167292950548626,
      "flos": 54925767457920.0,
      "grad_norm": 0.8501622468882478,
      "language_loss": 0.5579688,
      "learning_rate": 3.732018351516544e-06,
      "loss": 0.57866824,
      "num_input_tokens_seen": 68826815,
      "step": 3188,
      "time_per_iteration": 3.146822452545166
    },
    {
      "auxiliary_loss_clip": 0.01135721,
      "auxiliary_loss_mlp": 0.01047798,
      "balance_loss_clip": 1.05115223,
      "balance_loss_mlp": 1.03044116,
      "epoch": 0.19173305275815422,
      "flos": 29935601942400.0,
      "grad_norm": 1.75700400607208,
      "language_loss": 0.7000708,
      "learning_rate": 3.731823576891397e-06,
      "loss": 0.72190607,
      "num_input_tokens_seen": 68847585,
      "step": 3189,
      "time_per_iteration": 2.594344139099121
    },
    {
      "auxiliary_loss_clip": 0.01118277,
      "auxiliary_loss_mlp": 0.01036012,
      "balance_loss_clip": 1.04966748,
      "balance_loss_mlp": 1.01957262,
      "epoch": 0.1917931760108222,
      "flos": 24752471264640.0,
      "grad_norm": 2.0413280908600098,
      "language_loss": 0.74097604,
      "learning_rate": 3.7316287365953266e-06,
      "loss": 0.76251894,
      "num_input_tokens_seen": 68866620,
      "step": 3190,
      "time_per_iteration": 2.5525827407836914
    },
    {
      "auxiliary_loss_clip": 0.01108637,
      "auxiliary_loss_mlp": 0.01061104,
      "balance_loss_clip": 1.04744673,
      "balance_loss_mlp": 1.04182839,
      "epoch": 0.19185329926349015,
      "flos": 18843550375680.0,
      "grad_norm": 2.570864697177665,
      "language_loss": 0.84073281,
      "learning_rate": 3.73143383063572e-06,
      "loss": 0.86243021,
      "num_input_tokens_seen": 68885515,
      "step": 3191,
      "time_per_iteration": 2.561770439147949
    },
    {
      "auxiliary_loss_clip": 0.01122262,
      "auxiliary_loss_mlp": 0.01045366,
      "balance_loss_clip": 1.04541671,
      "balance_loss_mlp": 1.02804494,
      "epoch": 0.19191342251615812,
      "flos": 22086858038400.0,
      "grad_norm": 1.875468212231769,
      "language_loss": 0.89403325,
      "learning_rate": 3.73123885901997e-06,
      "loss": 0.91570956,
      "num_input_tokens_seen": 68903225,
      "step": 3192,
      "time_per_iteration": 2.5288219451904297
    },
    {
      "auxiliary_loss_clip": 0.0112405,
      "auxiliary_loss_mlp": 0.01050428,
      "balance_loss_clip": 1.0509758,
      "balance_loss_mlp": 1.03139007,
      "epoch": 0.19197354576882608,
      "flos": 22199115018240.0,
      "grad_norm": 2.24748754906377,
      "language_loss": 0.75186032,
      "learning_rate": 3.7310438217554687e-06,
      "loss": 0.77360511,
      "num_input_tokens_seen": 68922860,
      "step": 3193,
      "time_per_iteration": 2.5830509662628174
    },
    {
      "auxiliary_loss_clip": 0.01124003,
      "auxiliary_loss_mlp": 0.00803102,
      "balance_loss_clip": 1.04957592,
      "balance_loss_mlp": 1.02353334,
      "epoch": 0.19203366902149407,
      "flos": 24896185580160.0,
      "grad_norm": 2.7244440915051524,
      "language_loss": 0.74765766,
      "learning_rate": 3.730848718849612e-06,
      "loss": 0.76692867,
      "num_input_tokens_seen": 68943000,
      "step": 3194,
      "time_per_iteration": 2.5639002323150635
    },
    {
      "auxiliary_loss_clip": 0.01061011,
      "auxiliary_loss_mlp": 0.01016301,
      "balance_loss_clip": 1.02703571,
      "balance_loss_mlp": 1.01308239,
      "epoch": 0.19209379227416204,
      "flos": 68416722789120.0,
      "grad_norm": 0.8596416944852697,
      "language_loss": 0.68554509,
      "learning_rate": 3.7306535503097985e-06,
      "loss": 0.70631814,
      "num_input_tokens_seen": 69000255,
      "step": 3195,
      "time_per_iteration": 3.041114568710327
    },
    {
      "auxiliary_loss_clip": 0.01113685,
      "auxiliary_loss_mlp": 0.01056241,
      "balance_loss_clip": 1.04779851,
      "balance_loss_mlp": 1.03671432,
      "epoch": 0.19215391552683,
      "flos": 22055185221120.0,
      "grad_norm": 2.6900070510889202,
      "language_loss": 0.73099351,
      "learning_rate": 3.730458316143429e-06,
      "loss": 0.7526927,
      "num_input_tokens_seen": 69019665,
      "step": 3196,
      "time_per_iteration": 2.5798418521881104
    },
    {
      "auxiliary_loss_clip": 0.01134024,
      "auxiliary_loss_mlp": 0.01047124,
      "balance_loss_clip": 1.05590618,
      "balance_loss_mlp": 1.02932596,
      "epoch": 0.19221403877949797,
      "flos": 20302959962880.0,
      "grad_norm": 2.024231566370262,
      "language_loss": 0.83458531,
      "learning_rate": 3.7302630163579068e-06,
      "loss": 0.85639679,
      "num_input_tokens_seen": 69039055,
      "step": 3197,
      "time_per_iteration": 2.5140347480773926
    },
    {
      "auxiliary_loss_clip": 0.01089955,
      "auxiliary_loss_mlp": 0.01052249,
      "balance_loss_clip": 1.05185485,
      "balance_loss_mlp": 1.03244877,
      "epoch": 0.19227416203216594,
      "flos": 23185329811200.0,
      "grad_norm": 2.17408933000405,
      "language_loss": 0.80037606,
      "learning_rate": 3.7300676509606373e-06,
      "loss": 0.82179815,
      "num_input_tokens_seen": 69056370,
      "step": 3198,
      "time_per_iteration": 2.627760887145996
    },
    {
      "auxiliary_loss_clip": 0.01130352,
      "auxiliary_loss_mlp": 0.01051281,
      "balance_loss_clip": 1.04923964,
      "balance_loss_mlp": 1.03297019,
      "epoch": 0.1923342852848339,
      "flos": 25776607841280.0,
      "grad_norm": 2.340084961024336,
      "language_loss": 0.78796262,
      "learning_rate": 3.729872219959029e-06,
      "loss": 0.80977893,
      "num_input_tokens_seen": 69075915,
      "step": 3199,
      "time_per_iteration": 5.352616786956787
    },
    {
      "auxiliary_loss_clip": 0.01116488,
      "auxiliary_loss_mlp": 0.01053525,
      "balance_loss_clip": 1.0486263,
      "balance_loss_mlp": 1.03492868,
      "epoch": 0.19239440853750187,
      "flos": 17128349061120.0,
      "grad_norm": 3.2914731610694616,
      "language_loss": 0.84007913,
      "learning_rate": 3.7296767233604934e-06,
      "loss": 0.86177927,
      "num_input_tokens_seen": 69094145,
      "step": 3200,
      "time_per_iteration": 2.539914131164551
    },
    {
      "auxiliary_loss_clip": 0.01153726,
      "auxiliary_loss_mlp": 0.01051028,
      "balance_loss_clip": 1.05207229,
      "balance_loss_mlp": 1.03381407,
      "epoch": 0.19245453179016986,
      "flos": 16435093593600.0,
      "grad_norm": 2.039963542116121,
      "language_loss": 0.79519033,
      "learning_rate": 3.729481161172443e-06,
      "loss": 0.81723785,
      "num_input_tokens_seen": 69111110,
      "step": 3201,
      "time_per_iteration": 2.4444971084594727
    },
    {
      "auxiliary_loss_clip": 0.01100117,
      "auxiliary_loss_mlp": 0.01045062,
      "balance_loss_clip": 1.04675531,
      "balance_loss_mlp": 1.02678728,
      "epoch": 0.19251465504283782,
      "flos": 20230276792320.0,
      "grad_norm": 1.942535423653178,
      "language_loss": 0.69227087,
      "learning_rate": 3.7292855334022927e-06,
      "loss": 0.71372271,
      "num_input_tokens_seen": 69130280,
      "step": 3202,
      "time_per_iteration": 2.5836756229400635
    },
    {
      "auxiliary_loss_clip": 0.01129837,
      "auxiliary_loss_mlp": 0.01038556,
      "balance_loss_clip": 1.05030489,
      "balance_loss_mlp": 1.02079415,
      "epoch": 0.1925747782955058,
      "flos": 19464374067840.0,
      "grad_norm": 1.7872281834036303,
      "language_loss": 0.91255069,
      "learning_rate": 3.7290898400574627e-06,
      "loss": 0.93423462,
      "num_input_tokens_seen": 69149570,
      "step": 3203,
      "time_per_iteration": 2.5184192657470703
    },
    {
      "auxiliary_loss_clip": 0.01141921,
      "auxiliary_loss_mlp": 0.01048901,
      "balance_loss_clip": 1.04880095,
      "balance_loss_mlp": 1.03019667,
      "epoch": 0.19263490154817375,
      "flos": 17785586165760.0,
      "grad_norm": 2.9769886713882427,
      "language_loss": 0.81776732,
      "learning_rate": 3.7288940811453725e-06,
      "loss": 0.83967555,
      "num_input_tokens_seen": 69168190,
      "step": 3204,
      "time_per_iteration": 3.8514444828033447
    },
    {
      "auxiliary_loss_clip": 0.01112742,
      "auxiliary_loss_mlp": 0.01046104,
      "balance_loss_clip": 1.04743505,
      "balance_loss_mlp": 1.0284487,
      "epoch": 0.19269502480084172,
      "flos": 17457075354240.0,
      "grad_norm": 2.297398288266007,
      "language_loss": 0.75651813,
      "learning_rate": 3.7286982566734454e-06,
      "loss": 0.77810657,
      "num_input_tokens_seen": 69186950,
      "step": 3205,
      "time_per_iteration": 2.57621693611145
    },
    {
      "auxiliary_loss_clip": 0.01135708,
      "auxiliary_loss_mlp": 0.01049716,
      "balance_loss_clip": 1.05310225,
      "balance_loss_mlp": 1.03212118,
      "epoch": 0.19275514805350968,
      "flos": 21506901045120.0,
      "grad_norm": 2.917972205610784,
      "language_loss": 0.83352101,
      "learning_rate": 3.728502366649107e-06,
      "loss": 0.85537523,
      "num_input_tokens_seen": 69204850,
      "step": 3206,
      "time_per_iteration": 3.939209461212158
    },
    {
      "auxiliary_loss_clip": 0.01048644,
      "auxiliary_loss_mlp": 0.010037,
      "balance_loss_clip": 1.02158117,
      "balance_loss_mlp": 1.00009978,
      "epoch": 0.19281527130617768,
      "flos": 47695979738880.0,
      "grad_norm": 0.8419473442723836,
      "language_loss": 0.60560095,
      "learning_rate": 3.728306411079786e-06,
      "loss": 0.62612438,
      "num_input_tokens_seen": 69259200,
      "step": 3207,
      "time_per_iteration": 2.947436571121216
    },
    {
      "auxiliary_loss_clip": 0.01117425,
      "auxiliary_loss_mlp": 0.01045364,
      "balance_loss_clip": 1.050758,
      "balance_loss_mlp": 1.02755451,
      "epoch": 0.19287539455884564,
      "flos": 11801252672640.0,
      "grad_norm": 2.4669124251459964,
      "language_loss": 0.75260508,
      "learning_rate": 3.7281103899729125e-06,
      "loss": 0.77423298,
      "num_input_tokens_seen": 69275835,
      "step": 3208,
      "time_per_iteration": 2.5378518104553223
    },
    {
      "auxiliary_loss_clip": 0.01143752,
      "auxiliary_loss_mlp": 0.00796186,
      "balance_loss_clip": 1.04957008,
      "balance_loss_mlp": 1.01260948,
      "epoch": 0.1929355178115136,
      "flos": 20631434860800.0,
      "grad_norm": 2.1296869941936944,
      "language_loss": 0.60824847,
      "learning_rate": 3.7279143033359195e-06,
      "loss": 0.62764788,
      "num_input_tokens_seen": 69294810,
      "step": 3209,
      "time_per_iteration": 2.492734432220459
    },
    {
      "auxiliary_loss_clip": 0.01154999,
      "auxiliary_loss_mlp": 0.01048067,
      "balance_loss_clip": 1.05018878,
      "balance_loss_mlp": 1.02804017,
      "epoch": 0.19299564106418157,
      "flos": 40807916058240.0,
      "grad_norm": 2.3063968620770763,
      "language_loss": 0.79900527,
      "learning_rate": 3.727718151176243e-06,
      "loss": 0.82103598,
      "num_input_tokens_seen": 69316065,
      "step": 3210,
      "time_per_iteration": 2.6449081897735596
    },
    {
      "auxiliary_loss_clip": 0.01114777,
      "auxiliary_loss_mlp": 0.01039429,
      "balance_loss_clip": 1.04499364,
      "balance_loss_mlp": 1.02191734,
      "epoch": 0.19305576431684954,
      "flos": 11361418634880.0,
      "grad_norm": 2.1382958937281096,
      "language_loss": 0.83120441,
      "learning_rate": 3.7275219335013217e-06,
      "loss": 0.85274649,
      "num_input_tokens_seen": 69332900,
      "step": 3211,
      "time_per_iteration": 2.5183794498443604
    },
    {
      "auxiliary_loss_clip": 0.0106314,
      "auxiliary_loss_mlp": 0.0100613,
      "balance_loss_clip": 1.02081192,
      "balance_loss_mlp": 1.00264883,
      "epoch": 0.1931158875695175,
      "flos": 54511895975040.0,
      "grad_norm": 0.9640360044430135,
      "language_loss": 0.6366058,
      "learning_rate": 3.7273256503185953e-06,
      "loss": 0.65729851,
      "num_input_tokens_seen": 69382535,
      "step": 3212,
      "time_per_iteration": 2.9352073669433594
    },
    {
      "auxiliary_loss_clip": 0.0112642,
      "auxiliary_loss_mlp": 0.01047532,
      "balance_loss_clip": 1.05067015,
      "balance_loss_mlp": 1.03002012,
      "epoch": 0.19317601082218547,
      "flos": 19828436365440.0,
      "grad_norm": 1.6604524303089223,
      "language_loss": 0.76276112,
      "learning_rate": 3.7271293016355074e-06,
      "loss": 0.7845006,
      "num_input_tokens_seen": 69400600,
      "step": 3213,
      "time_per_iteration": 2.520739793777466
    },
    {
      "auxiliary_loss_clip": 0.01121666,
      "auxiliary_loss_mlp": 0.01045108,
      "balance_loss_clip": 1.05070162,
      "balance_loss_mlp": 1.02645135,
      "epoch": 0.19323613407485346,
      "flos": 13152068467200.0,
      "grad_norm": 2.3578127155743105,
      "language_loss": 0.70696962,
      "learning_rate": 3.726932887459503e-06,
      "loss": 0.72863734,
      "num_input_tokens_seen": 69417350,
      "step": 3214,
      "time_per_iteration": 2.5383896827697754
    },
    {
      "auxiliary_loss_clip": 0.01151288,
      "auxiliary_loss_mlp": 0.01049499,
      "balance_loss_clip": 1.04934692,
      "balance_loss_mlp": 1.0300802,
      "epoch": 0.19329625732752143,
      "flos": 14027247342720.0,
      "grad_norm": 2.2375838986521313,
      "language_loss": 0.75737697,
      "learning_rate": 3.72673640779803e-06,
      "loss": 0.77938485,
      "num_input_tokens_seen": 69431845,
      "step": 3215,
      "time_per_iteration": 2.4480459690093994
    },
    {
      "auxiliary_loss_clip": 0.01113341,
      "auxiliary_loss_mlp": 0.01052196,
      "balance_loss_clip": 1.0507139,
      "balance_loss_mlp": 1.03456461,
      "epoch": 0.1933563805801894,
      "flos": 23441732069760.0,
      "grad_norm": 2.0965926041438943,
      "language_loss": 0.88312352,
      "learning_rate": 3.72653986265854e-06,
      "loss": 0.90477896,
      "num_input_tokens_seen": 69453275,
      "step": 3216,
      "time_per_iteration": 2.6002695560455322
    },
    {
      "auxiliary_loss_clip": 0.01151288,
      "auxiliary_loss_mlp": 0.01053993,
      "balance_loss_clip": 1.05041397,
      "balance_loss_mlp": 1.03610003,
      "epoch": 0.19341650383285736,
      "flos": 20485314334080.0,
      "grad_norm": 1.9844599361289452,
      "language_loss": 0.80424595,
      "learning_rate": 3.726343252048485e-06,
      "loss": 0.82629871,
      "num_input_tokens_seen": 69471830,
      "step": 3217,
      "time_per_iteration": 2.4625868797302246
    },
    {
      "auxiliary_loss_clip": 0.01137242,
      "auxiliary_loss_mlp": 0.01051701,
      "balance_loss_clip": 1.05226493,
      "balance_loss_mlp": 1.03210318,
      "epoch": 0.19347662708552532,
      "flos": 17858484817920.0,
      "grad_norm": 2.734041486537027,
      "language_loss": 0.61750537,
      "learning_rate": 3.7261465759753206e-06,
      "loss": 0.63939482,
      "num_input_tokens_seen": 69489320,
      "step": 3218,
      "time_per_iteration": 2.5274431705474854
    },
    {
      "auxiliary_loss_clip": 0.01153481,
      "auxiliary_loss_mlp": 0.01045116,
      "balance_loss_clip": 1.05086553,
      "balance_loss_mlp": 1.02736545,
      "epoch": 0.1935367503381933,
      "flos": 18187247024640.0,
      "grad_norm": 1.7150437821310596,
      "language_loss": 0.80289006,
      "learning_rate": 3.7259498344465053e-06,
      "loss": 0.82487607,
      "num_input_tokens_seen": 69506665,
      "step": 3219,
      "time_per_iteration": 2.454946756362915
    },
    {
      "auxiliary_loss_clip": 0.01100649,
      "auxiliary_loss_mlp": 0.01048984,
      "balance_loss_clip": 1.0488708,
      "balance_loss_mlp": 1.0296247,
      "epoch": 0.19359687359086128,
      "flos": 15957122290560.0,
      "grad_norm": 2.2388463255854956,
      "language_loss": 0.85946214,
      "learning_rate": 3.7257530274694993e-06,
      "loss": 0.88095844,
      "num_input_tokens_seen": 69523835,
      "step": 3220,
      "time_per_iteration": 2.6098787784576416
    },
    {
      "auxiliary_loss_clip": 0.01145188,
      "auxiliary_loss_mlp": 0.01039962,
      "balance_loss_clip": 1.04949582,
      "balance_loss_mlp": 1.02370811,
      "epoch": 0.19365699684352924,
      "flos": 21215198695680.0,
      "grad_norm": 1.9365119181076715,
      "language_loss": 0.84057105,
      "learning_rate": 3.725556155051766e-06,
      "loss": 0.86242253,
      "num_input_tokens_seen": 69542620,
      "step": 3221,
      "time_per_iteration": 2.490363359451294
    },
    {
      "auxiliary_loss_clip": 0.01140772,
      "auxiliary_loss_mlp": 0.01045072,
      "balance_loss_clip": 1.05050373,
      "balance_loss_mlp": 1.02841878,
      "epoch": 0.1937171200961972,
      "flos": 17311098481920.0,
      "grad_norm": 2.4014845737807584,
      "language_loss": 0.85819721,
      "learning_rate": 3.7253592172007702e-06,
      "loss": 0.88005567,
      "num_input_tokens_seen": 69561130,
      "step": 3222,
      "time_per_iteration": 2.5092411041259766
    },
    {
      "auxiliary_loss_clip": 0.01072055,
      "auxiliary_loss_mlp": 0.01039787,
      "balance_loss_clip": 1.04618192,
      "balance_loss_mlp": 1.02085686,
      "epoch": 0.19377724334886517,
      "flos": 22635968227200.0,
      "grad_norm": 1.9088949455167903,
      "language_loss": 0.78566861,
      "learning_rate": 3.72516221392398e-06,
      "loss": 0.80678701,
      "num_input_tokens_seen": 69580425,
      "step": 3223,
      "time_per_iteration": 2.6770312786102295
    },
    {
      "auxiliary_loss_clip": 0.01141154,
      "auxiliary_loss_mlp": 0.01045022,
      "balance_loss_clip": 1.05101848,
      "balance_loss_mlp": 1.0268302,
      "epoch": 0.19383736660153314,
      "flos": 15077813351040.0,
      "grad_norm": 1.9814483816718913,
      "language_loss": 0.75482559,
      "learning_rate": 3.7249651452288653e-06,
      "loss": 0.77668738,
      "num_input_tokens_seen": 69597085,
      "step": 3224,
      "time_per_iteration": 2.4869399070739746
    },
    {
      "auxiliary_loss_clip": 0.01100465,
      "auxiliary_loss_mlp": 0.01053385,
      "balance_loss_clip": 1.04548621,
      "balance_loss_mlp": 1.03297591,
      "epoch": 0.1938974898542011,
      "flos": 47119934350080.0,
      "grad_norm": 2.59646414124123,
      "language_loss": 0.70715779,
      "learning_rate": 3.7247680111229e-06,
      "loss": 0.72869629,
      "num_input_tokens_seen": 69618885,
      "step": 3225,
      "time_per_iteration": 2.8448405265808105
    },
    {
      "auxiliary_loss_clip": 0.01123601,
      "auxiliary_loss_mlp": 0.01046427,
      "balance_loss_clip": 1.05631709,
      "balance_loss_mlp": 1.02933264,
      "epoch": 0.19395761310686907,
      "flos": 25812554376960.0,
      "grad_norm": 2.26534696345421,
      "language_loss": 0.69014382,
      "learning_rate": 3.7245708116135585e-06,
      "loss": 0.71184409,
      "num_input_tokens_seen": 69638200,
      "step": 3226,
      "time_per_iteration": 2.6155080795288086
    },
    {
      "auxiliary_loss_clip": 0.01122218,
      "auxiliary_loss_mlp": 0.01043257,
      "balance_loss_clip": 1.05097818,
      "balance_loss_mlp": 1.0242548,
      "epoch": 0.19401773635953706,
      "flos": 23039604334080.0,
      "grad_norm": 2.034673115881795,
      "language_loss": 0.75973368,
      "learning_rate": 3.7243735467083193e-06,
      "loss": 0.7813884,
      "num_input_tokens_seen": 69657550,
      "step": 3227,
      "time_per_iteration": 2.5929644107818604
    },
    {
      "auxiliary_loss_clip": 0.01114558,
      "auxiliary_loss_mlp": 0.01042988,
      "balance_loss_clip": 1.050354,
      "balance_loss_mlp": 1.02586937,
      "epoch": 0.19407785961220503,
      "flos": 15920780705280.0,
      "grad_norm": 3.0062666979892745,
      "language_loss": 0.69543821,
      "learning_rate": 3.724176216414662e-06,
      "loss": 0.71701366,
      "num_input_tokens_seen": 69675005,
      "step": 3228,
      "time_per_iteration": 2.5607962608337402
    },
    {
      "auxiliary_loss_clip": 0.01139728,
      "auxiliary_loss_mlp": 0.01041807,
      "balance_loss_clip": 1.04944313,
      "balance_loss_mlp": 1.02439046,
      "epoch": 0.194137982864873,
      "flos": 25921722787200.0,
      "grad_norm": 1.934839726888585,
      "language_loss": 0.74067461,
      "learning_rate": 3.72397882074007e-06,
      "loss": 0.76249003,
      "num_input_tokens_seen": 69696455,
      "step": 3229,
      "time_per_iteration": 2.5767550468444824
    },
    {
      "auxiliary_loss_clip": 0.01116411,
      "auxiliary_loss_mlp": 0.01043367,
      "balance_loss_clip": 1.05182374,
      "balance_loss_mlp": 1.02539039,
      "epoch": 0.19419810611754096,
      "flos": 13261344618240.0,
      "grad_norm": 1.820750970145502,
      "language_loss": 0.65239704,
      "learning_rate": 3.7237813596920285e-06,
      "loss": 0.67399478,
      "num_input_tokens_seen": 69714245,
      "step": 3230,
      "time_per_iteration": 2.533379554748535
    },
    {
      "auxiliary_loss_clip": 0.01118705,
      "auxiliary_loss_mlp": 0.00800352,
      "balance_loss_clip": 1.04840612,
      "balance_loss_mlp": 1.02206254,
      "epoch": 0.19425822937020892,
      "flos": 15705568368000.0,
      "grad_norm": 1.9641396353098515,
      "language_loss": 0.81872278,
      "learning_rate": 3.7235838332780254e-06,
      "loss": 0.83791339,
      "num_input_tokens_seen": 69731515,
      "step": 3231,
      "time_per_iteration": 2.5613300800323486
    },
    {
      "auxiliary_loss_clip": 0.01127194,
      "auxiliary_loss_mlp": 0.01041467,
      "balance_loss_clip": 1.05091345,
      "balance_loss_mlp": 1.02258384,
      "epoch": 0.1943183526228769,
      "flos": 23105392093440.0,
      "grad_norm": 1.8562655932334644,
      "language_loss": 0.86694682,
      "learning_rate": 3.72338624150555e-06,
      "loss": 0.88863337,
      "num_input_tokens_seen": 69748885,
      "step": 3232,
      "time_per_iteration": 2.5626444816589355
    },
    {
      "auxiliary_loss_clip": 0.01100382,
      "auxiliary_loss_mlp": 0.01052225,
      "balance_loss_clip": 1.05155921,
      "balance_loss_mlp": 1.0325793,
      "epoch": 0.19437847587554485,
      "flos": 24712610146560.0,
      "grad_norm": 1.7543461861919698,
      "language_loss": 0.85026813,
      "learning_rate": 3.723188584382096e-06,
      "loss": 0.87179422,
      "num_input_tokens_seen": 69767540,
      "step": 3233,
      "time_per_iteration": 2.6394073963165283
    },
    {
      "auxiliary_loss_clip": 0.01143603,
      "auxiliary_loss_mlp": 0.01049419,
      "balance_loss_clip": 1.05011392,
      "balance_loss_mlp": 1.03189552,
      "epoch": 0.19443859912821285,
      "flos": 23116130259840.0,
      "grad_norm": 1.7590725628999118,
      "language_loss": 0.89402759,
      "learning_rate": 3.722990861915158e-06,
      "loss": 0.91595781,
      "num_input_tokens_seen": 69789340,
      "step": 3234,
      "time_per_iteration": 2.547166347503662
    },
    {
      "auxiliary_loss_clip": 0.01132913,
      "auxiliary_loss_mlp": 0.01046437,
      "balance_loss_clip": 1.04710889,
      "balance_loss_mlp": 1.02776885,
      "epoch": 0.1944987223808808,
      "flos": 15084385539840.0,
      "grad_norm": 2.3085519844802147,
      "language_loss": 0.78336751,
      "learning_rate": 3.722793074112234e-06,
      "loss": 0.805161,
      "num_input_tokens_seen": 69806470,
      "step": 3235,
      "time_per_iteration": 2.5068070888519287
    },
    {
      "auxiliary_loss_clip": 0.0113062,
      "auxiliary_loss_mlp": 0.01043971,
      "balance_loss_clip": 1.05186641,
      "balance_loss_mlp": 1.02716279,
      "epoch": 0.19455884563354878,
      "flos": 17126876603520.0,
      "grad_norm": 2.1354861256165307,
      "language_loss": 0.79022926,
      "learning_rate": 3.7225952209808233e-06,
      "loss": 0.81197518,
      "num_input_tokens_seen": 69822655,
      "step": 3236,
      "time_per_iteration": 2.5263383388519287
    },
    {
      "auxiliary_loss_clip": 0.01151548,
      "auxiliary_loss_mlp": 0.01040875,
      "balance_loss_clip": 1.05211747,
      "balance_loss_mlp": 1.0224098,
      "epoch": 0.19461896888621674,
      "flos": 20193396503040.0,
      "grad_norm": 1.7912458680857635,
      "language_loss": 0.75773942,
      "learning_rate": 3.72239730252843e-06,
      "loss": 0.77966368,
      "num_input_tokens_seen": 69841895,
      "step": 3237,
      "time_per_iteration": 3.936593770980835
    },
    {
      "auxiliary_loss_clip": 0.01154743,
      "auxiliary_loss_mlp": 0.0105121,
      "balance_loss_clip": 1.05022216,
      "balance_loss_mlp": 1.03385329,
      "epoch": 0.1946790921388847,
      "flos": 25301365971840.0,
      "grad_norm": 4.261608506761535,
      "language_loss": 0.7522608,
      "learning_rate": 3.7221993187625583e-06,
      "loss": 0.77432036,
      "num_input_tokens_seen": 69862220,
      "step": 3238,
      "time_per_iteration": 2.496584415435791
    },
    {
      "auxiliary_loss_clip": 0.01105751,
      "auxiliary_loss_mlp": 0.01040033,
      "balance_loss_clip": 1.04669476,
      "balance_loss_mlp": 1.02146006,
      "epoch": 0.19473921539155267,
      "flos": 20193396503040.0,
      "grad_norm": 3.2733538830422515,
      "language_loss": 0.73250818,
      "learning_rate": 3.7220012696907155e-06,
      "loss": 0.75396597,
      "num_input_tokens_seen": 69881830,
      "step": 3239,
      "time_per_iteration": 3.948404312133789
    },
    {
      "auxiliary_loss_clip": 0.01125231,
      "auxiliary_loss_mlp": 0.01047496,
      "balance_loss_clip": 1.04728794,
      "balance_loss_mlp": 1.02962613,
      "epoch": 0.19479933864422067,
      "flos": 20887549810560.0,
      "grad_norm": 2.3465884681937506,
      "language_loss": 0.73420417,
      "learning_rate": 3.721803155320412e-06,
      "loss": 0.75593144,
      "num_input_tokens_seen": 69900515,
      "step": 3240,
      "time_per_iteration": 2.526747703552246
    },
    {
      "auxiliary_loss_clip": 0.01127418,
      "auxiliary_loss_mlp": 0.0103631,
      "balance_loss_clip": 1.05483294,
      "balance_loss_mlp": 1.01906013,
      "epoch": 0.19485946189688863,
      "flos": 23295072839040.0,
      "grad_norm": 2.0178538609134837,
      "language_loss": 0.66272652,
      "learning_rate": 3.7216049756591606e-06,
      "loss": 0.68436372,
      "num_input_tokens_seen": 69920060,
      "step": 3241,
      "time_per_iteration": 2.5215232372283936
    },
    {
      "auxiliary_loss_clip": 0.01128532,
      "auxiliary_loss_mlp": 0.01040839,
      "balance_loss_clip": 1.05023754,
      "balance_loss_mlp": 1.02332711,
      "epoch": 0.1949195851495566,
      "flos": 23295036925440.0,
      "grad_norm": 1.452531203227124,
      "language_loss": 0.8304584,
      "learning_rate": 3.7214067307144754e-06,
      "loss": 0.85215211,
      "num_input_tokens_seen": 69939820,
      "step": 3242,
      "time_per_iteration": 3.915907859802246
    },
    {
      "auxiliary_loss_clip": 0.01066334,
      "auxiliary_loss_mlp": 0.01012461,
      "balance_loss_clip": 1.02512097,
      "balance_loss_mlp": 1.00924253,
      "epoch": 0.19497970840222456,
      "flos": 64962871557120.0,
      "grad_norm": 0.8419656905917792,
      "language_loss": 0.57497168,
      "learning_rate": 3.721208420493875e-06,
      "loss": 0.59575963,
      "num_input_tokens_seen": 70002145,
      "step": 3243,
      "time_per_iteration": 3.1004536151885986
    },
    {
      "auxiliary_loss_clip": 0.01131209,
      "auxiliary_loss_mlp": 0.01052273,
      "balance_loss_clip": 1.04534578,
      "balance_loss_mlp": 1.03216207,
      "epoch": 0.19503983165489253,
      "flos": 19644717277440.0,
      "grad_norm": 2.0738155526155,
      "language_loss": 0.83651114,
      "learning_rate": 3.7210100450048784e-06,
      "loss": 0.85834599,
      "num_input_tokens_seen": 70020510,
      "step": 3244,
      "time_per_iteration": 3.8990097045898438
    },
    {
      "auxiliary_loss_clip": 0.01137855,
      "auxiliary_loss_mlp": 0.01047761,
      "balance_loss_clip": 1.05025876,
      "balance_loss_mlp": 1.03097618,
      "epoch": 0.1950999549075605,
      "flos": 21141976821120.0,
      "grad_norm": 1.733635934154372,
      "language_loss": 0.7715047,
      "learning_rate": 3.7208116042550088e-06,
      "loss": 0.79336089,
      "num_input_tokens_seen": 70040760,
      "step": 3245,
      "time_per_iteration": 2.496063232421875
    },
    {
      "auxiliary_loss_clip": 0.0113642,
      "auxiliary_loss_mlp": 0.01039543,
      "balance_loss_clip": 1.04862273,
      "balance_loss_mlp": 1.02148271,
      "epoch": 0.19516007816022846,
      "flos": 20884820376960.0,
      "grad_norm": 2.0709844248693745,
      "language_loss": 0.84433544,
      "learning_rate": 3.7206130982517906e-06,
      "loss": 0.86609507,
      "num_input_tokens_seen": 70058720,
      "step": 3246,
      "time_per_iteration": 2.4785845279693604
    },
    {
      "auxiliary_loss_clip": 0.01139756,
      "auxiliary_loss_mlp": 0.00796454,
      "balance_loss_clip": 1.04889941,
      "balance_loss_mlp": 1.01632953,
      "epoch": 0.19522020141289645,
      "flos": 16910515031040.0,
      "grad_norm": 2.1199074476513067,
      "language_loss": 0.75920796,
      "learning_rate": 3.7204145270027514e-06,
      "loss": 0.77857006,
      "num_input_tokens_seen": 70076470,
      "step": 3247,
      "time_per_iteration": 2.4822113513946533
    },
    {
      "auxiliary_loss_clip": 0.01116161,
      "auxiliary_loss_mlp": 0.01042711,
      "balance_loss_clip": 1.05190015,
      "balance_loss_mlp": 1.0252353,
      "epoch": 0.19528032466556441,
      "flos": 26724829023360.0,
      "grad_norm": 2.009152373167641,
      "language_loss": 0.75523454,
      "learning_rate": 3.720215890515421e-06,
      "loss": 0.77682328,
      "num_input_tokens_seen": 70096220,
      "step": 3248,
      "time_per_iteration": 2.604114055633545
    },
    {
      "auxiliary_loss_clip": 0.01148099,
      "auxiliary_loss_mlp": 0.01049609,
      "balance_loss_clip": 1.0467279,
      "balance_loss_mlp": 1.03176355,
      "epoch": 0.19534044791823238,
      "flos": 21032808410880.0,
      "grad_norm": 2.0496814784466566,
      "language_loss": 0.78434885,
      "learning_rate": 3.7200171887973316e-06,
      "loss": 0.80632591,
      "num_input_tokens_seen": 70114800,
      "step": 3249,
      "time_per_iteration": 2.4469313621520996
    },
    {
      "auxiliary_loss_clip": 0.01141007,
      "auxiliary_loss_mlp": 0.01049143,
      "balance_loss_clip": 1.04769468,
      "balance_loss_mlp": 1.0316906,
      "epoch": 0.19540057117090034,
      "flos": 22344050396160.0,
      "grad_norm": 1.5698944768530867,
      "language_loss": 0.73113221,
      "learning_rate": 3.7198184218560176e-06,
      "loss": 0.75303376,
      "num_input_tokens_seen": 70134930,
      "step": 3250,
      "time_per_iteration": 2.488373279571533
    },
    {
      "auxiliary_loss_clip": 0.0110173,
      "auxiliary_loss_mlp": 0.0103971,
      "balance_loss_clip": 1.04771435,
      "balance_loss_mlp": 1.02213907,
      "epoch": 0.1954606944235683,
      "flos": 20301631159680.0,
      "grad_norm": 2.0664699263406376,
      "language_loss": 0.78815746,
      "learning_rate": 3.719619589699017e-06,
      "loss": 0.80957186,
      "num_input_tokens_seen": 70152045,
      "step": 3251,
      "time_per_iteration": 2.5392239093780518
    },
    {
      "auxiliary_loss_clip": 0.01148771,
      "auxiliary_loss_mlp": 0.01044434,
      "balance_loss_clip": 1.04765558,
      "balance_loss_mlp": 1.02679086,
      "epoch": 0.19552081767623627,
      "flos": 17346865449600.0,
      "grad_norm": 2.8671932463487555,
      "language_loss": 0.83283484,
      "learning_rate": 3.7194206923338695e-06,
      "loss": 0.85476685,
      "num_input_tokens_seen": 70169240,
      "step": 3252,
      "time_per_iteration": 2.412708044052124
    },
    {
      "auxiliary_loss_clip": 0.01136603,
      "auxiliary_loss_mlp": 0.01052454,
      "balance_loss_clip": 1.04738414,
      "balance_loss_mlp": 1.03155637,
      "epoch": 0.19558094092890424,
      "flos": 31977626129280.0,
      "grad_norm": 1.6965419249853735,
      "language_loss": 0.7343502,
      "learning_rate": 3.719221729768117e-06,
      "loss": 0.75624073,
      "num_input_tokens_seen": 70192690,
      "step": 3253,
      "time_per_iteration": 2.55769419670105
    },
    {
      "auxiliary_loss_clip": 0.01099952,
      "auxiliary_loss_mlp": 0.01041051,
      "balance_loss_clip": 1.05002451,
      "balance_loss_mlp": 1.0229311,
      "epoch": 0.19564106418157223,
      "flos": 22268889187200.0,
      "grad_norm": 1.8238767103200455,
      "language_loss": 0.76728785,
      "learning_rate": 3.7190227020093037e-06,
      "loss": 0.7886979,
      "num_input_tokens_seen": 70209685,
      "step": 3254,
      "time_per_iteration": 2.584441661834717
    },
    {
      "auxiliary_loss_clip": 0.01023352,
      "auxiliary_loss_mlp": 0.01007753,
      "balance_loss_clip": 1.02680278,
      "balance_loss_mlp": 1.0042721,
      "epoch": 0.1957011874342402,
      "flos": 54364554385920.0,
      "grad_norm": 0.7612769350343391,
      "language_loss": 0.55323118,
      "learning_rate": 3.7188236090649774e-06,
      "loss": 0.5735423,
      "num_input_tokens_seen": 70265050,
      "step": 3255,
      "time_per_iteration": 3.2127888202667236
    },
    {
      "auxiliary_loss_clip": 0.01129108,
      "auxiliary_loss_mlp": 0.01042786,
      "balance_loss_clip": 1.0519352,
      "balance_loss_mlp": 1.0245831,
      "epoch": 0.19576131068690816,
      "flos": 16506699356160.0,
      "grad_norm": 2.6828205080654324,
      "language_loss": 0.70497632,
      "learning_rate": 3.718624450942688e-06,
      "loss": 0.72669524,
      "num_input_tokens_seen": 70281830,
      "step": 3256,
      "time_per_iteration": 2.9588983058929443
    },
    {
      "auxiliary_loss_clip": 0.01147786,
      "auxiliary_loss_mlp": 0.01042627,
      "balance_loss_clip": 1.04783881,
      "balance_loss_mlp": 1.02528226,
      "epoch": 0.19582143393957613,
      "flos": 14719676797440.0,
      "grad_norm": 2.238339897395564,
      "language_loss": 0.80067188,
      "learning_rate": 3.718425227649987e-06,
      "loss": 0.82257605,
      "num_input_tokens_seen": 70297420,
      "step": 3257,
      "time_per_iteration": 2.450559616088867
    },
    {
      "auxiliary_loss_clip": 0.01101482,
      "auxiliary_loss_mlp": 0.01040291,
      "balance_loss_clip": 1.05151486,
      "balance_loss_mlp": 1.02313638,
      "epoch": 0.1958815571922441,
      "flos": 24425504737920.0,
      "grad_norm": 2.01212194941165,
      "language_loss": 0.74935007,
      "learning_rate": 3.7182259391944292e-06,
      "loss": 0.77076781,
      "num_input_tokens_seen": 70319210,
      "step": 3258,
      "time_per_iteration": 2.690911293029785
    },
    {
      "auxiliary_loss_clip": 0.01081562,
      "auxiliary_loss_mlp": 0.01048564,
      "balance_loss_clip": 1.04737234,
      "balance_loss_mlp": 1.02835798,
      "epoch": 0.19594168044491206,
      "flos": 24900279730560.0,
      "grad_norm": 6.003969498498999,
      "language_loss": 0.7383796,
      "learning_rate": 3.7180265855835714e-06,
      "loss": 0.75968081,
      "num_input_tokens_seen": 70339045,
      "step": 3259,
      "time_per_iteration": 2.665027618408203
    },
    {
      "auxiliary_loss_clip": 0.01127066,
      "auxiliary_loss_mlp": 0.0104349,
      "balance_loss_clip": 1.04945016,
      "balance_loss_mlp": 1.02467918,
      "epoch": 0.19600180369758005,
      "flos": 12057008486400.0,
      "grad_norm": 2.8940158183417366,
      "language_loss": 0.77196652,
      "learning_rate": 3.7178271668249735e-06,
      "loss": 0.79367214,
      "num_input_tokens_seen": 70356505,
      "step": 3260,
      "time_per_iteration": 2.5723419189453125
    },
    {
      "auxiliary_loss_clip": 0.01140416,
      "auxiliary_loss_mlp": 0.01049443,
      "balance_loss_clip": 1.04751873,
      "balance_loss_mlp": 1.03125155,
      "epoch": 0.19606192695024802,
      "flos": 20850202644480.0,
      "grad_norm": 2.7742365845602373,
      "language_loss": 0.82137764,
      "learning_rate": 3.7176276829261975e-06,
      "loss": 0.84327614,
      "num_input_tokens_seen": 70375410,
      "step": 3261,
      "time_per_iteration": 2.4897830486297607
    },
    {
      "auxiliary_loss_clip": 0.01117739,
      "auxiliary_loss_mlp": 0.01042568,
      "balance_loss_clip": 1.05407238,
      "balance_loss_mlp": 1.02399564,
      "epoch": 0.19612205020291598,
      "flos": 28475509996800.0,
      "grad_norm": 1.789598330358096,
      "language_loss": 0.76931751,
      "learning_rate": 3.717428133894807e-06,
      "loss": 0.7909205,
      "num_input_tokens_seen": 70396315,
      "step": 3262,
      "time_per_iteration": 2.6371021270751953
    },
    {
      "auxiliary_loss_clip": 0.01141717,
      "auxiliary_loss_mlp": 0.01044113,
      "balance_loss_clip": 1.05248868,
      "balance_loss_mlp": 1.02661324,
      "epoch": 0.19618217345558395,
      "flos": 25556618995200.0,
      "grad_norm": 1.7665637847834788,
      "language_loss": 0.8616538,
      "learning_rate": 3.71722851973837e-06,
      "loss": 0.88351214,
      "num_input_tokens_seen": 70417945,
      "step": 3263,
      "time_per_iteration": 2.5474720001220703
    },
    {
      "auxiliary_loss_clip": 0.01128661,
      "auxiliary_loss_mlp": 0.01038483,
      "balance_loss_clip": 1.05128038,
      "balance_loss_mlp": 1.02098322,
      "epoch": 0.1962422967082519,
      "flos": 25264413855360.0,
      "grad_norm": 1.7996355052996422,
      "language_loss": 0.7414028,
      "learning_rate": 3.717028840464455e-06,
      "loss": 0.76307428,
      "num_input_tokens_seen": 70438690,
      "step": 3264,
      "time_per_iteration": 2.601285457611084
    },
    {
      "auxiliary_loss_clip": 0.01137469,
      "auxiliary_loss_mlp": 0.01044969,
      "balance_loss_clip": 1.05127728,
      "balance_loss_mlp": 1.028005,
      "epoch": 0.19630241996091988,
      "flos": 18807352444800.0,
      "grad_norm": 2.080231413100442,
      "language_loss": 0.78564823,
      "learning_rate": 3.7168290960806344e-06,
      "loss": 0.80747259,
      "num_input_tokens_seen": 70455385,
      "step": 3265,
      "time_per_iteration": 2.4894556999206543
    },
    {
      "auxiliary_loss_clip": 0.01015473,
      "auxiliary_loss_mlp": 0.01007761,
      "balance_loss_clip": 1.01893032,
      "balance_loss_mlp": 1.00510299,
      "epoch": 0.19636254321358784,
      "flos": 62321137896960.0,
      "grad_norm": 0.7864506441228578,
      "language_loss": 0.53439766,
      "learning_rate": 3.716629286594483e-06,
      "loss": 0.55462998,
      "num_input_tokens_seen": 70514280,
      "step": 3266,
      "time_per_iteration": 3.1949355602264404
    },
    {
      "auxiliary_loss_clip": 0.01125634,
      "auxiliary_loss_mlp": 0.00802539,
      "balance_loss_clip": 1.0506525,
      "balance_loss_mlp": 1.02467942,
      "epoch": 0.19642266646625584,
      "flos": 21069329564160.0,
      "grad_norm": 1.9536652125328091,
      "language_loss": 0.80348015,
      "learning_rate": 3.7164294120135767e-06,
      "loss": 0.82276195,
      "num_input_tokens_seen": 70531800,
      "step": 3267,
      "time_per_iteration": 2.5653140544891357
    },
    {
      "auxiliary_loss_clip": 0.0112795,
      "auxiliary_loss_mlp": 0.01043352,
      "balance_loss_clip": 1.04846478,
      "balance_loss_mlp": 1.02557826,
      "epoch": 0.1964827897189238,
      "flos": 14538651229440.0,
      "grad_norm": 2.1332956642785974,
      "language_loss": 0.86523467,
      "learning_rate": 3.7162294723454953e-06,
      "loss": 0.88694769,
      "num_input_tokens_seen": 70550615,
      "step": 3268,
      "time_per_iteration": 2.5538671016693115
    },
    {
      "auxiliary_loss_clip": 0.01102881,
      "auxiliary_loss_mlp": 0.01045207,
      "balance_loss_clip": 1.05378771,
      "balance_loss_mlp": 1.02756429,
      "epoch": 0.19654291297159177,
      "flos": 19244636616960.0,
      "grad_norm": 1.980292437854082,
      "language_loss": 0.69060701,
      "learning_rate": 3.7160294675978197e-06,
      "loss": 0.71208787,
      "num_input_tokens_seen": 70568690,
      "step": 3269,
      "time_per_iteration": 2.590402364730835
    },
    {
      "auxiliary_loss_clip": 0.01115434,
      "auxiliary_loss_mlp": 0.01059585,
      "balance_loss_clip": 1.05122888,
      "balance_loss_mlp": 1.04034424,
      "epoch": 0.19660303622425973,
      "flos": 25775710001280.0,
      "grad_norm": 1.8391123692930271,
      "language_loss": 0.80823714,
      "learning_rate": 3.715829397778135e-06,
      "loss": 0.82998729,
      "num_input_tokens_seen": 70588665,
      "step": 3270,
      "time_per_iteration": 2.590343475341797
    },
    {
      "auxiliary_loss_clip": 0.01136823,
      "auxiliary_loss_mlp": 0.0104708,
      "balance_loss_clip": 1.04876447,
      "balance_loss_mlp": 1.02994931,
      "epoch": 0.1966631594769277,
      "flos": 20595093275520.0,
      "grad_norm": 2.0915679383222003,
      "language_loss": 0.84022164,
      "learning_rate": 3.715629262894028e-06,
      "loss": 0.86206067,
      "num_input_tokens_seen": 70606900,
      "step": 3271,
      "time_per_iteration": 2.498685598373413
    },
    {
      "auxiliary_loss_clip": 0.01136372,
      "auxiliary_loss_mlp": 0.01057328,
      "balance_loss_clip": 1.05152476,
      "balance_loss_mlp": 1.03949463,
      "epoch": 0.19672328272959566,
      "flos": 23623188600960.0,
      "grad_norm": 2.0764233844289346,
      "language_loss": 0.80308515,
      "learning_rate": 3.715429062953087e-06,
      "loss": 0.8250221,
      "num_input_tokens_seen": 70625955,
      "step": 3272,
      "time_per_iteration": 2.5142102241516113
    },
    {
      "auxiliary_loss_clip": 0.01118086,
      "auxiliary_loss_mlp": 0.01065935,
      "balance_loss_clip": 1.04689813,
      "balance_loss_mlp": 1.04437006,
      "epoch": 0.19678340598226365,
      "flos": 23110922787840.0,
      "grad_norm": 1.8189130691948165,
      "language_loss": 0.8047595,
      "learning_rate": 3.7152287979629043e-06,
      "loss": 0.82659972,
      "num_input_tokens_seen": 70646090,
      "step": 3273,
      "time_per_iteration": 2.547788381576538
    },
    {
      "auxiliary_loss_clip": 0.01140584,
      "auxiliary_loss_mlp": 0.01053516,
      "balance_loss_clip": 1.05093765,
      "balance_loss_mlp": 1.03636241,
      "epoch": 0.19684352923493162,
      "flos": 24534852716160.0,
      "grad_norm": 1.9101776131046144,
      "language_loss": 0.77595425,
      "learning_rate": 3.7150284679310735e-06,
      "loss": 0.79789531,
      "num_input_tokens_seen": 70666065,
      "step": 3274,
      "time_per_iteration": 2.5332815647125244
    },
    {
      "auxiliary_loss_clip": 0.01138913,
      "auxiliary_loss_mlp": 0.01052428,
      "balance_loss_clip": 1.05340278,
      "balance_loss_mlp": 1.03376007,
      "epoch": 0.19690365248759958,
      "flos": 21796448578560.0,
      "grad_norm": 7.51699774298835,
      "language_loss": 0.81107926,
      "learning_rate": 3.7148280728651914e-06,
      "loss": 0.83299267,
      "num_input_tokens_seen": 70681580,
      "step": 3275,
      "time_per_iteration": 2.4692206382751465
    },
    {
      "auxiliary_loss_clip": 0.01114769,
      "auxiliary_loss_mlp": 0.01049236,
      "balance_loss_clip": 1.05106986,
      "balance_loss_mlp": 1.03051984,
      "epoch": 0.19696377574026755,
      "flos": 19056643810560.0,
      "grad_norm": 2.3296336930953836,
      "language_loss": 0.81116921,
      "learning_rate": 3.7146276127728563e-06,
      "loss": 0.83280933,
      "num_input_tokens_seen": 70697745,
      "step": 3276,
      "time_per_iteration": 3.912823438644409
    },
    {
      "auxiliary_loss_clip": 0.01136738,
      "auxiliary_loss_mlp": 0.01037518,
      "balance_loss_clip": 1.04681683,
      "balance_loss_mlp": 1.01937485,
      "epoch": 0.19702389899293551,
      "flos": 22820656982400.0,
      "grad_norm": 2.1966898592045774,
      "language_loss": 0.89353323,
      "learning_rate": 3.7144270876616713e-06,
      "loss": 0.91527575,
      "num_input_tokens_seen": 70715110,
      "step": 3277,
      "time_per_iteration": 3.8930046558380127
    },
    {
      "auxiliary_loss_clip": 0.01105764,
      "auxiliary_loss_mlp": 0.01050758,
      "balance_loss_clip": 1.05055058,
      "balance_loss_mlp": 1.03054023,
      "epoch": 0.19708402224560348,
      "flos": 22894237992960.0,
      "grad_norm": 2.1569819907952414,
      "language_loss": 0.62501413,
      "learning_rate": 3.714226497539239e-06,
      "loss": 0.64657938,
      "num_input_tokens_seen": 70734715,
      "step": 3278,
      "time_per_iteration": 2.601461172103882
    },
    {
      "auxiliary_loss_clip": 0.01113778,
      "auxiliary_loss_mlp": 0.01055316,
      "balance_loss_clip": 1.05154347,
      "balance_loss_mlp": 1.03617048,
      "epoch": 0.19714414549827144,
      "flos": 25662519267840.0,
      "grad_norm": 2.262917658758903,
      "language_loss": 0.73796356,
      "learning_rate": 3.714025842413166e-06,
      "loss": 0.75965452,
      "num_input_tokens_seen": 70752650,
      "step": 3279,
      "time_per_iteration": 2.5930287837982178
    },
    {
      "auxiliary_loss_clip": 0.01141249,
      "auxiliary_loss_mlp": 0.01042013,
      "balance_loss_clip": 1.04807723,
      "balance_loss_mlp": 1.02431083,
      "epoch": 0.19720426875093944,
      "flos": 23915824704000.0,
      "grad_norm": 1.8725884200075202,
      "language_loss": 0.8259899,
      "learning_rate": 3.713825122291061e-06,
      "loss": 0.84782255,
      "num_input_tokens_seen": 70772365,
      "step": 3280,
      "time_per_iteration": 2.5292646884918213
    },
    {
      "auxiliary_loss_clip": 0.01106847,
      "auxiliary_loss_mlp": 0.01052886,
      "balance_loss_clip": 1.0516752,
      "balance_loss_mlp": 1.03366971,
      "epoch": 0.1972643920036074,
      "flos": 13881952828800.0,
      "grad_norm": 3.030054514917669,
      "language_loss": 0.77674508,
      "learning_rate": 3.713624337180536e-06,
      "loss": 0.79834247,
      "num_input_tokens_seen": 70790340,
      "step": 3281,
      "time_per_iteration": 4.0080695152282715
    },
    {
      "auxiliary_loss_clip": 0.01122576,
      "auxiliary_loss_mlp": 0.01047162,
      "balance_loss_clip": 1.053738,
      "balance_loss_mlp": 1.03012705,
      "epoch": 0.19732451525627537,
      "flos": 19863592801920.0,
      "grad_norm": 1.7199607620967226,
      "language_loss": 0.79417372,
      "learning_rate": 3.7134234870892045e-06,
      "loss": 0.81587112,
      "num_input_tokens_seen": 70809295,
      "step": 3282,
      "time_per_iteration": 2.536085605621338
    },
    {
      "auxiliary_loss_clip": 0.01110383,
      "auxiliary_loss_mlp": 0.01044473,
      "balance_loss_clip": 1.05225313,
      "balance_loss_mlp": 1.02683043,
      "epoch": 0.19738463850894333,
      "flos": 24973429777920.0,
      "grad_norm": 3.585271668549556,
      "language_loss": 0.71985221,
      "learning_rate": 3.7132225720246826e-06,
      "loss": 0.74140084,
      "num_input_tokens_seen": 70828765,
      "step": 3283,
      "time_per_iteration": 4.006861209869385
    },
    {
      "auxiliary_loss_clip": 0.01135635,
      "auxiliary_loss_mlp": 0.01046812,
      "balance_loss_clip": 1.05172777,
      "balance_loss_mlp": 1.02934802,
      "epoch": 0.1974447617616113,
      "flos": 18368883123840.0,
      "grad_norm": 2.57439658783808,
      "language_loss": 0.78905171,
      "learning_rate": 3.7130215919945886e-06,
      "loss": 0.81087625,
      "num_input_tokens_seen": 70846805,
      "step": 3284,
      "time_per_iteration": 2.473994255065918
    },
    {
      "auxiliary_loss_clip": 0.011254,
      "auxiliary_loss_mlp": 0.00801549,
      "balance_loss_clip": 1.05192971,
      "balance_loss_mlp": 1.02437925,
      "epoch": 0.19750488501427926,
      "flos": 22892945103360.0,
      "grad_norm": 2.0083784989296305,
      "language_loss": 0.86485088,
      "learning_rate": 3.7128205470065445e-06,
      "loss": 0.88412035,
      "num_input_tokens_seen": 70863805,
      "step": 3285,
      "time_per_iteration": 2.5330419540405273
    },
    {
      "auxiliary_loss_clip": 0.01122696,
      "auxiliary_loss_mlp": 0.01044876,
      "balance_loss_clip": 1.05545712,
      "balance_loss_mlp": 1.02672076,
      "epoch": 0.19756500826694723,
      "flos": 21871502046720.0,
      "grad_norm": 2.064065742610555,
      "language_loss": 0.88246822,
      "learning_rate": 3.712619437068174e-06,
      "loss": 0.90414393,
      "num_input_tokens_seen": 70882660,
      "step": 3286,
      "time_per_iteration": 2.5421111583709717
    },
    {
      "auxiliary_loss_clip": 0.01117421,
      "auxiliary_loss_mlp": 0.01053657,
      "balance_loss_clip": 1.05129194,
      "balance_loss_mlp": 1.03209245,
      "epoch": 0.19762513151961522,
      "flos": 15158972131200.0,
      "grad_norm": 2.149687026494698,
      "language_loss": 0.77930146,
      "learning_rate": 3.712418262187102e-06,
      "loss": 0.80101228,
      "num_input_tokens_seen": 70898765,
      "step": 3287,
      "time_per_iteration": 2.529116153717041
    },
    {
      "auxiliary_loss_clip": 0.0112764,
      "auxiliary_loss_mlp": 0.01050901,
      "balance_loss_clip": 1.05326104,
      "balance_loss_mlp": 1.0316608,
      "epoch": 0.1976852547722832,
      "flos": 16979175878400.0,
      "grad_norm": 4.016206525924968,
      "language_loss": 0.8148526,
      "learning_rate": 3.7122170223709584e-06,
      "loss": 0.83663797,
      "num_input_tokens_seen": 70916370,
      "step": 3288,
      "time_per_iteration": 2.4900996685028076
    },
    {
      "auxiliary_loss_clip": 0.01131263,
      "auxiliary_loss_mlp": 0.01051415,
      "balance_loss_clip": 1.05077195,
      "balance_loss_mlp": 1.03343868,
      "epoch": 0.19774537802495115,
      "flos": 20302924049280.0,
      "grad_norm": 1.6543335769540386,
      "language_loss": 0.72898316,
      "learning_rate": 3.712015717627374e-06,
      "loss": 0.75080991,
      "num_input_tokens_seen": 70934870,
      "step": 3289,
      "time_per_iteration": 2.4970452785491943
    },
    {
      "auxiliary_loss_clip": 0.01131723,
      "auxiliary_loss_mlp": 0.01045109,
      "balance_loss_clip": 1.05309796,
      "balance_loss_mlp": 1.02616692,
      "epoch": 0.19780550127761912,
      "flos": 27235478724480.0,
      "grad_norm": 1.7654481887142006,
      "language_loss": 0.79606974,
      "learning_rate": 3.7118143479639813e-06,
      "loss": 0.81783807,
      "num_input_tokens_seen": 70955140,
      "step": 3290,
      "time_per_iteration": 2.571993827819824
    },
    {
      "auxiliary_loss_clip": 0.01045557,
      "auxiliary_loss_mlp": 0.01011122,
      "balance_loss_clip": 1.02707219,
      "balance_loss_mlp": 1.00857115,
      "epoch": 0.19786562453028708,
      "flos": 63550972684800.0,
      "grad_norm": 0.8960846080928302,
      "language_loss": 0.60342979,
      "learning_rate": 3.711612913388418e-06,
      "loss": 0.62399656,
      "num_input_tokens_seen": 71012005,
      "step": 3291,
      "time_per_iteration": 3.167348861694336
    },
    {
      "auxiliary_loss_clip": 0.01155999,
      "auxiliary_loss_mlp": 0.01043635,
      "balance_loss_clip": 1.05201626,
      "balance_loss_mlp": 1.02373922,
      "epoch": 0.19792574778295505,
      "flos": 26286647011200.0,
      "grad_norm": 1.7942676119591836,
      "language_loss": 0.8133986,
      "learning_rate": 3.7114114139083204e-06,
      "loss": 0.83539498,
      "num_input_tokens_seen": 71031140,
      "step": 3292,
      "time_per_iteration": 2.5083141326904297
    },
    {
      "auxiliary_loss_clip": 0.01117678,
      "auxiliary_loss_mlp": 0.00799688,
      "balance_loss_clip": 1.05238008,
      "balance_loss_mlp": 1.01527464,
      "epoch": 0.19798587103562304,
      "flos": 19938107566080.0,
      "grad_norm": 1.9341564156621003,
      "language_loss": 0.81406951,
      "learning_rate": 3.7112098495313313e-06,
      "loss": 0.83324325,
      "num_input_tokens_seen": 71050250,
      "step": 3293,
      "time_per_iteration": 2.543415069580078
    },
    {
      "auxiliary_loss_clip": 0.01136142,
      "auxiliary_loss_mlp": 0.01049126,
      "balance_loss_clip": 1.05260229,
      "balance_loss_mlp": 1.02933717,
      "epoch": 0.198045994288291,
      "flos": 20120282369280.0,
      "grad_norm": 1.982512192608721,
      "language_loss": 0.61448193,
      "learning_rate": 3.711008220265093e-06,
      "loss": 0.6363346,
      "num_input_tokens_seen": 71068665,
      "step": 3294,
      "time_per_iteration": 2.541515588760376
    },
    {
      "auxiliary_loss_clip": 0.01129779,
      "auxiliary_loss_mlp": 0.01039001,
      "balance_loss_clip": 1.05084419,
      "balance_loss_mlp": 1.02164388,
      "epoch": 0.19810611754095897,
      "flos": 17967653228160.0,
      "grad_norm": 1.9813279064381748,
      "language_loss": 0.87191004,
      "learning_rate": 3.710806526117251e-06,
      "loss": 0.89359784,
      "num_input_tokens_seen": 71085320,
      "step": 3295,
      "time_per_iteration": 2.5219264030456543
    },
    {
      "auxiliary_loss_clip": 0.01110873,
      "auxiliary_loss_mlp": 0.01053328,
      "balance_loss_clip": 1.04860854,
      "balance_loss_mlp": 1.03541112,
      "epoch": 0.19816624079362694,
      "flos": 15084996071040.0,
      "grad_norm": 2.4546391364681766,
      "language_loss": 0.80768919,
      "learning_rate": 3.7106047670954544e-06,
      "loss": 0.82933116,
      "num_input_tokens_seen": 71102020,
      "step": 3296,
      "time_per_iteration": 2.5175559520721436
    },
    {
      "auxiliary_loss_clip": 0.01118491,
      "auxiliary_loss_mlp": 0.01048832,
      "balance_loss_clip": 1.04787326,
      "balance_loss_mlp": 1.02841187,
      "epoch": 0.1982263640462949,
      "flos": 24900315644160.0,
      "grad_norm": 2.084209941560928,
      "language_loss": 0.68266779,
      "learning_rate": 3.710402943207354e-06,
      "loss": 0.70434099,
      "num_input_tokens_seen": 71123390,
      "step": 3297,
      "time_per_iteration": 2.59619402885437
    },
    {
      "auxiliary_loss_clip": 0.01148442,
      "auxiliary_loss_mlp": 0.01037363,
      "balance_loss_clip": 1.05175829,
      "balance_loss_mlp": 1.01985145,
      "epoch": 0.19828648729896287,
      "flos": 20376181837440.0,
      "grad_norm": 1.8559053624726098,
      "language_loss": 0.81644136,
      "learning_rate": 3.7102010544606016e-06,
      "loss": 0.83829939,
      "num_input_tokens_seen": 71141800,
      "step": 3298,
      "time_per_iteration": 2.447580337524414
    },
    {
      "auxiliary_loss_clip": 0.01131663,
      "auxiliary_loss_mlp": 0.01041661,
      "balance_loss_clip": 1.05202651,
      "balance_loss_mlp": 1.02133608,
      "epoch": 0.19834661055163083,
      "flos": 18880035615360.0,
      "grad_norm": 2.4752445228323525,
      "language_loss": 0.85185754,
      "learning_rate": 3.7099991008628544e-06,
      "loss": 0.87359071,
      "num_input_tokens_seen": 71159505,
      "step": 3299,
      "time_per_iteration": 2.513279676437378
    },
    {
      "auxiliary_loss_clip": 0.01033989,
      "auxiliary_loss_mlp": 0.01006097,
      "balance_loss_clip": 1.02562237,
      "balance_loss_mlp": 1.00346231,
      "epoch": 0.19840673380429882,
      "flos": 60259184640000.0,
      "grad_norm": 0.7982187505892517,
      "language_loss": 0.53235543,
      "learning_rate": 3.7097970824217706e-06,
      "loss": 0.55275631,
      "num_input_tokens_seen": 71223265,
      "step": 3300,
      "time_per_iteration": 3.141616106033325
    },
    {
      "auxiliary_loss_clip": 0.01105926,
      "auxiliary_loss_mlp": 0.01063595,
      "balance_loss_clip": 1.04684436,
      "balance_loss_mlp": 1.04039681,
      "epoch": 0.1984668570569668,
      "flos": 19902017376000.0,
      "grad_norm": 1.6103771596426892,
      "language_loss": 0.73596382,
      "learning_rate": 3.7095949991450093e-06,
      "loss": 0.75765902,
      "num_input_tokens_seen": 71242385,
      "step": 3301,
      "time_per_iteration": 2.5656323432922363
    },
    {
      "auxiliary_loss_clip": 0.01114868,
      "auxiliary_loss_mlp": 0.01042185,
      "balance_loss_clip": 1.05292201,
      "balance_loss_mlp": 1.02411282,
      "epoch": 0.19852698030963475,
      "flos": 15630766295040.0,
      "grad_norm": 2.409301268279074,
      "language_loss": 0.87857592,
      "learning_rate": 3.709392851040235e-06,
      "loss": 0.90014637,
      "num_input_tokens_seen": 71258990,
      "step": 3302,
      "time_per_iteration": 2.5389819145202637
    },
    {
      "auxiliary_loss_clip": 0.01117342,
      "auxiliary_loss_mlp": 0.01047402,
      "balance_loss_clip": 1.05323255,
      "balance_loss_mlp": 1.02878189,
      "epoch": 0.19858710356230272,
      "flos": 43143007311360.0,
      "grad_norm": 4.02163790287082,
      "language_loss": 0.73582959,
      "learning_rate": 3.709190638115111e-06,
      "loss": 0.75747699,
      "num_input_tokens_seen": 71282770,
      "step": 3303,
      "time_per_iteration": 2.763754367828369
    },
    {
      "auxiliary_loss_clip": 0.01137524,
      "auxiliary_loss_mlp": 0.01046956,
      "balance_loss_clip": 1.051615,
      "balance_loss_mlp": 1.02900338,
      "epoch": 0.19864722681497068,
      "flos": 35144084643840.0,
      "grad_norm": 2.1624385797327514,
      "language_loss": 0.74912369,
      "learning_rate": 3.7089883603773084e-06,
      "loss": 0.7709685,
      "num_input_tokens_seen": 71301410,
      "step": 3304,
      "time_per_iteration": 2.605717420578003
    },
    {
      "auxiliary_loss_clip": 0.01127848,
      "auxiliary_loss_mlp": 0.0103643,
      "balance_loss_clip": 1.04958153,
      "balance_loss_mlp": 1.01901329,
      "epoch": 0.19870735006763865,
      "flos": 19426200888960.0,
      "grad_norm": 1.6697206178558528,
      "language_loss": 0.86044621,
      "learning_rate": 3.7087860178344955e-06,
      "loss": 0.88208902,
      "num_input_tokens_seen": 71319670,
      "step": 3305,
      "time_per_iteration": 2.5096516609191895
    },
    {
      "auxiliary_loss_clip": 0.01126886,
      "auxiliary_loss_mlp": 0.01042047,
      "balance_loss_clip": 1.04734612,
      "balance_loss_mlp": 1.0237844,
      "epoch": 0.19876747332030664,
      "flos": 23547380947200.0,
      "grad_norm": 1.76553115385197,
      "language_loss": 0.68208426,
      "learning_rate": 3.7085836104943445e-06,
      "loss": 0.70377362,
      "num_input_tokens_seen": 71339850,
      "step": 3306,
      "time_per_iteration": 2.573091745376587
    },
    {
      "auxiliary_loss_clip": 0.01116082,
      "auxiliary_loss_mlp": 0.01039677,
      "balance_loss_clip": 1.04718351,
      "balance_loss_mlp": 1.0226059,
      "epoch": 0.1988275965729746,
      "flos": 19829406032640.0,
      "grad_norm": 1.562596760300182,
      "language_loss": 0.76212156,
      "learning_rate": 3.7083811383645332e-06,
      "loss": 0.78367913,
      "num_input_tokens_seen": 71359795,
      "step": 3307,
      "time_per_iteration": 2.540689706802368
    },
    {
      "auxiliary_loss_clip": 0.01150609,
      "auxiliary_loss_mlp": 0.01044448,
      "balance_loss_clip": 1.05207109,
      "balance_loss_mlp": 1.02725768,
      "epoch": 0.19888771982564257,
      "flos": 23513625141120.0,
      "grad_norm": 1.7432862006242638,
      "language_loss": 0.75497031,
      "learning_rate": 3.708178601452737e-06,
      "loss": 0.77692086,
      "num_input_tokens_seen": 71378885,
      "step": 3308,
      "time_per_iteration": 2.514305830001831
    },
    {
      "auxiliary_loss_clip": 0.01105231,
      "auxiliary_loss_mlp": 0.01043465,
      "balance_loss_clip": 1.05255628,
      "balance_loss_mlp": 1.02467752,
      "epoch": 0.19894784307831054,
      "flos": 18150510389760.0,
      "grad_norm": 1.7579892345177726,
      "language_loss": 0.76092207,
      "learning_rate": 3.7079759997666374e-06,
      "loss": 0.78240907,
      "num_input_tokens_seen": 71397285,
      "step": 3309,
      "time_per_iteration": 2.571211099624634
    },
    {
      "auxiliary_loss_clip": 0.01136556,
      "auxiliary_loss_mlp": 0.01052244,
      "balance_loss_clip": 1.04954743,
      "balance_loss_mlp": 1.03232408,
      "epoch": 0.1990079663309785,
      "flos": 24276044246400.0,
      "grad_norm": 1.5608973085550537,
      "language_loss": 0.87897718,
      "learning_rate": 3.707773333313917e-06,
      "loss": 0.9008652,
      "num_input_tokens_seen": 71415775,
      "step": 3310,
      "time_per_iteration": 2.54250168800354
    },
    {
      "auxiliary_loss_clip": 0.01147271,
      "auxiliary_loss_mlp": 0.01038838,
      "balance_loss_clip": 1.0494802,
      "balance_loss_mlp": 1.0209806,
      "epoch": 0.19906808958364647,
      "flos": 34897666366080.0,
      "grad_norm": 2.3759238770042055,
      "language_loss": 0.63846099,
      "learning_rate": 3.70757060210226e-06,
      "loss": 0.66032207,
      "num_input_tokens_seen": 71437315,
      "step": 3311,
      "time_per_iteration": 2.573270320892334
    },
    {
      "auxiliary_loss_clip": 0.01111877,
      "auxiliary_loss_mlp": 0.01040979,
      "balance_loss_clip": 1.04895639,
      "balance_loss_mlp": 1.02271628,
      "epoch": 0.19912821283631443,
      "flos": 24024885373440.0,
      "grad_norm": 3.023810423911428,
      "language_loss": 0.74011111,
      "learning_rate": 3.707367806139355e-06,
      "loss": 0.76163965,
      "num_input_tokens_seen": 71456320,
      "step": 3312,
      "time_per_iteration": 2.5957820415496826
    },
    {
      "auxiliary_loss_clip": 0.01137433,
      "auxiliary_loss_mlp": 0.01048727,
      "balance_loss_clip": 1.05104733,
      "balance_loss_mlp": 1.03085732,
      "epoch": 0.19918833608898243,
      "flos": 19859031774720.0,
      "grad_norm": 2.005599126884347,
      "language_loss": 0.834768,
      "learning_rate": 3.7071649454328915e-06,
      "loss": 0.85662961,
      "num_input_tokens_seen": 71475360,
      "step": 3313,
      "time_per_iteration": 2.5033764839172363
    },
    {
      "auxiliary_loss_clip": 0.01138602,
      "auxiliary_loss_mlp": 0.01045383,
      "balance_loss_clip": 1.0513072,
      "balance_loss_mlp": 1.02751374,
      "epoch": 0.1992484593416504,
      "flos": 29095794984960.0,
      "grad_norm": 2.01811917967288,
      "language_loss": 0.80942428,
      "learning_rate": 3.7069620199905625e-06,
      "loss": 0.83126414,
      "num_input_tokens_seen": 71496155,
      "step": 3314,
      "time_per_iteration": 2.5842645168304443
    },
    {
      "auxiliary_loss_clip": 0.01109761,
      "auxiliary_loss_mlp": 0.01046759,
      "balance_loss_clip": 1.04544187,
      "balance_loss_mlp": 1.02937865,
      "epoch": 0.19930858259431836,
      "flos": 23295001011840.0,
      "grad_norm": 1.665119470094462,
      "language_loss": 0.87711012,
      "learning_rate": 3.7067590298200627e-06,
      "loss": 0.89867526,
      "num_input_tokens_seen": 71517295,
      "step": 3315,
      "time_per_iteration": 4.028660535812378
    },
    {
      "auxiliary_loss_clip": 0.01114626,
      "auxiliary_loss_mlp": 0.00796958,
      "balance_loss_clip": 1.05146539,
      "balance_loss_mlp": 1.01423144,
      "epoch": 0.19936870584698632,
      "flos": 25378825651200.0,
      "grad_norm": 1.8894919509743227,
      "language_loss": 0.71273398,
      "learning_rate": 3.7065559749290892e-06,
      "loss": 0.73184979,
      "num_input_tokens_seen": 71540000,
      "step": 3316,
      "time_per_iteration": 3.9917256832122803
    },
    {
      "auxiliary_loss_clip": 0.01016674,
      "auxiliary_loss_mlp": 0.01008294,
      "balance_loss_clip": 1.02884018,
      "balance_loss_mlp": 1.00561166,
      "epoch": 0.1994288290996543,
      "flos": 62168053109760.0,
      "grad_norm": 0.835463578108911,
      "language_loss": 0.66240323,
      "learning_rate": 3.706352855325342e-06,
      "loss": 0.68265295,
      "num_input_tokens_seen": 71607880,
      "step": 3317,
      "time_per_iteration": 3.3859307765960693
    },
    {
      "auxiliary_loss_clip": 0.01139043,
      "auxiliary_loss_mlp": 0.01049067,
      "balance_loss_clip": 1.04822743,
      "balance_loss_mlp": 1.02998209,
      "epoch": 0.19948895235232225,
      "flos": 19025832919680.0,
      "grad_norm": 2.3023488155875937,
      "language_loss": 0.74178112,
      "learning_rate": 3.7061496710165233e-06,
      "loss": 0.76366222,
      "num_input_tokens_seen": 71625695,
      "step": 3318,
      "time_per_iteration": 2.5921742916107178
    },
    {
      "auxiliary_loss_clip": 0.01113032,
      "auxiliary_loss_mlp": 0.01049382,
      "balance_loss_clip": 1.04779887,
      "balance_loss_mlp": 1.03201294,
      "epoch": 0.19954907560499022,
      "flos": 37815803182080.0,
      "grad_norm": 2.2028137995519024,
      "language_loss": 0.79032218,
      "learning_rate": 3.7059464220103385e-06,
      "loss": 0.81194633,
      "num_input_tokens_seen": 71648520,
      "step": 3319,
      "time_per_iteration": 2.7042741775512695
    },
    {
      "auxiliary_loss_clip": 0.01132079,
      "auxiliary_loss_mlp": 0.01042274,
      "balance_loss_clip": 1.05224919,
      "balance_loss_mlp": 1.0230453,
      "epoch": 0.1996091988576582,
      "flos": 49565199594240.0,
      "grad_norm": 2.5041305526812043,
      "language_loss": 0.76276696,
      "learning_rate": 3.7057431083144945e-06,
      "loss": 0.78451049,
      "num_input_tokens_seen": 71672185,
      "step": 3320,
      "time_per_iteration": 4.230144739151001
    },
    {
      "auxiliary_loss_clip": 0.01116677,
      "auxiliary_loss_mlp": 0.01044356,
      "balance_loss_clip": 1.04906225,
      "balance_loss_mlp": 1.02662969,
      "epoch": 0.19966932211032618,
      "flos": 22635788659200.0,
      "grad_norm": 1.917012296278872,
      "language_loss": 0.80064785,
      "learning_rate": 3.705539729936701e-06,
      "loss": 0.82225817,
      "num_input_tokens_seen": 71692890,
      "step": 3321,
      "time_per_iteration": 3.9482314586639404
    },
    {
      "auxiliary_loss_clip": 0.01029903,
      "auxiliary_loss_mlp": 0.01004601,
      "balance_loss_clip": 1.02104378,
      "balance_loss_mlp": 1.00187135,
      "epoch": 0.19972944536299414,
      "flos": 54082117745280.0,
      "grad_norm": 0.9025459696493671,
      "language_loss": 0.65171719,
      "learning_rate": 3.7053362868846696e-06,
      "loss": 0.67206216,
      "num_input_tokens_seen": 71745815,
      "step": 3322,
      "time_per_iteration": 2.945993423461914
    },
    {
      "auxiliary_loss_clip": 0.01036975,
      "auxiliary_loss_mlp": 0.01013504,
      "balance_loss_clip": 1.02747512,
      "balance_loss_mlp": 1.01085782,
      "epoch": 0.1997895686156621,
      "flos": 69355031817600.0,
      "grad_norm": 0.7864191662094184,
      "language_loss": 0.5695821,
      "learning_rate": 3.7051327791661153e-06,
      "loss": 0.59008694,
      "num_input_tokens_seen": 71806915,
      "step": 3323,
      "time_per_iteration": 3.2409138679504395
    },
    {
      "auxiliary_loss_clip": 0.0112959,
      "auxiliary_loss_mlp": 0.00798671,
      "balance_loss_clip": 1.05184698,
      "balance_loss_mlp": 1.01703179,
      "epoch": 0.19984969186833007,
      "flos": 18552063507840.0,
      "grad_norm": 2.0090650435752044,
      "language_loss": 0.80539191,
      "learning_rate": 3.7049292067887555e-06,
      "loss": 0.82467455,
      "num_input_tokens_seen": 71824645,
      "step": 3324,
      "time_per_iteration": 2.5097837448120117
    },
    {
      "auxiliary_loss_clip": 0.01131123,
      "auxiliary_loss_mlp": 0.01045698,
      "balance_loss_clip": 1.04686594,
      "balance_loss_mlp": 1.02650571,
      "epoch": 0.19990981512099804,
      "flos": 26429678968320.0,
      "grad_norm": 1.7889441359760727,
      "language_loss": 0.53707862,
      "learning_rate": 3.7047255697603092e-06,
      "loss": 0.55884683,
      "num_input_tokens_seen": 71845125,
      "step": 3325,
      "time_per_iteration": 2.555075168609619
    },
    {
      "auxiliary_loss_clip": 0.01127579,
      "auxiliary_loss_mlp": 0.01046714,
      "balance_loss_clip": 1.05075932,
      "balance_loss_mlp": 1.02959538,
      "epoch": 0.19996993837366603,
      "flos": 16325997010560.0,
      "grad_norm": 1.909021284812481,
      "language_loss": 0.8619293,
      "learning_rate": 3.7045218680884984e-06,
      "loss": 0.88367224,
      "num_input_tokens_seen": 71863500,
      "step": 3326,
      "time_per_iteration": 2.5059351921081543
    },
    {
      "auxiliary_loss_clip": 0.01148041,
      "auxiliary_loss_mlp": 0.01041268,
      "balance_loss_clip": 1.05154228,
      "balance_loss_mlp": 1.02360117,
      "epoch": 0.200030061626334,
      "flos": 20844169159680.0,
      "grad_norm": 1.8952223124237986,
      "language_loss": 0.71900713,
      "learning_rate": 3.7043181017810476e-06,
      "loss": 0.74090016,
      "num_input_tokens_seen": 71881845,
      "step": 3327,
      "time_per_iteration": 2.4791247844696045
    },
    {
      "auxiliary_loss_clip": 0.01125862,
      "auxiliary_loss_mlp": 0.01046439,
      "balance_loss_clip": 1.04959095,
      "balance_loss_mlp": 1.0271033,
      "epoch": 0.20009018487900196,
      "flos": 23762629198080.0,
      "grad_norm": 2.305050613111692,
      "language_loss": 0.76387143,
      "learning_rate": 3.7041142708456833e-06,
      "loss": 0.7855944,
      "num_input_tokens_seen": 71900940,
      "step": 3328,
      "time_per_iteration": 2.5441510677337646
    },
    {
      "auxiliary_loss_clip": 0.01120335,
      "auxiliary_loss_mlp": 0.01043594,
      "balance_loss_clip": 1.05129337,
      "balance_loss_mlp": 1.0263567,
      "epoch": 0.20015030813166992,
      "flos": 28111555440000.0,
      "grad_norm": 1.7423102762457783,
      "language_loss": 0.69776535,
      "learning_rate": 3.7039103752901353e-06,
      "loss": 0.71940458,
      "num_input_tokens_seen": 71921925,
      "step": 3329,
      "time_per_iteration": 2.653041362762451
    },
    {
      "auxiliary_loss_clip": 0.01106715,
      "auxiliary_loss_mlp": 0.0106203,
      "balance_loss_clip": 1.04736459,
      "balance_loss_mlp": 1.03905833,
      "epoch": 0.2002104313843379,
      "flos": 26067160955520.0,
      "grad_norm": 1.7803965756444502,
      "language_loss": 0.81644595,
      "learning_rate": 3.7037064151221353e-06,
      "loss": 0.83813334,
      "num_input_tokens_seen": 71941855,
      "step": 3330,
      "time_per_iteration": 2.6207263469696045
    },
    {
      "auxiliary_loss_clip": 0.01134305,
      "auxiliary_loss_mlp": 0.01040921,
      "balance_loss_clip": 1.05068576,
      "balance_loss_mlp": 1.02270579,
      "epoch": 0.20027055463700585,
      "flos": 22966633854720.0,
      "grad_norm": 2.152395380354671,
      "language_loss": 0.76646584,
      "learning_rate": 3.703502390349417e-06,
      "loss": 0.78821808,
      "num_input_tokens_seen": 71960915,
      "step": 3331,
      "time_per_iteration": 2.5592243671417236
    },
    {
      "auxiliary_loss_clip": 0.01095142,
      "auxiliary_loss_mlp": 0.01060408,
      "balance_loss_clip": 1.04886901,
      "balance_loss_mlp": 1.03905725,
      "epoch": 0.20033067788967382,
      "flos": 17165660313600.0,
      "grad_norm": 2.1085381614126235,
      "language_loss": 0.79292601,
      "learning_rate": 3.7032983009797176e-06,
      "loss": 0.8144815,
      "num_input_tokens_seen": 71979220,
      "step": 3332,
      "time_per_iteration": 2.678664445877075
    },
    {
      "auxiliary_loss_clip": 0.01048226,
      "auxiliary_loss_mlp": 0.0101148,
      "balance_loss_clip": 1.02209735,
      "balance_loss_mlp": 1.00869048,
      "epoch": 0.2003908011423418,
      "flos": 60825566292480.0,
      "grad_norm": 1.4618236980177526,
      "language_loss": 0.61940742,
      "learning_rate": 3.703094147020776e-06,
      "loss": 0.64000452,
      "num_input_tokens_seen": 72033950,
      "step": 3333,
      "time_per_iteration": 3.000027656555176
    },
    {
      "auxiliary_loss_clip": 0.01112419,
      "auxiliary_loss_mlp": 0.00799368,
      "balance_loss_clip": 1.05128431,
      "balance_loss_mlp": 1.01804841,
      "epoch": 0.20045092439500978,
      "flos": 24206234163840.0,
      "grad_norm": 2.0536790796252453,
      "language_loss": 0.80988765,
      "learning_rate": 3.7028899284803334e-06,
      "loss": 0.82900548,
      "num_input_tokens_seen": 72051395,
      "step": 3334,
      "time_per_iteration": 2.615004301071167
    },
    {
      "auxiliary_loss_clip": 0.01094649,
      "auxiliary_loss_mlp": 0.01052167,
      "balance_loss_clip": 1.0496279,
      "balance_loss_mlp": 1.03225923,
      "epoch": 0.20051104764767774,
      "flos": 29387605075200.0,
      "grad_norm": 2.6302363892076444,
      "language_loss": 0.74409306,
      "learning_rate": 3.702685645366134e-06,
      "loss": 0.76556122,
      "num_input_tokens_seen": 72071305,
      "step": 3335,
      "time_per_iteration": 2.680752992630005
    },
    {
      "auxiliary_loss_clip": 0.01148995,
      "auxiliary_loss_mlp": 0.01058225,
      "balance_loss_clip": 1.05612075,
      "balance_loss_mlp": 1.03980756,
      "epoch": 0.2005711709003457,
      "flos": 23513804709120.0,
      "grad_norm": 1.6423333662693385,
      "language_loss": 0.80093241,
      "learning_rate": 3.7024812976859243e-06,
      "loss": 0.8230046,
      "num_input_tokens_seen": 72090165,
      "step": 3336,
      "time_per_iteration": 2.5767722129821777
    },
    {
      "auxiliary_loss_clip": 0.01113488,
      "auxiliary_loss_mlp": 0.01054803,
      "balance_loss_clip": 1.04849029,
      "balance_loss_mlp": 1.03441882,
      "epoch": 0.20063129415301367,
      "flos": 22523388024960.0,
      "grad_norm": 2.9492692045898976,
      "language_loss": 0.77535897,
      "learning_rate": 3.7022768854474532e-06,
      "loss": 0.79704189,
      "num_input_tokens_seen": 72107210,
      "step": 3337,
      "time_per_iteration": 2.59751296043396
    },
    {
      "auxiliary_loss_clip": 0.01149483,
      "auxiliary_loss_mlp": 0.01046774,
      "balance_loss_clip": 1.05023205,
      "balance_loss_mlp": 1.02717578,
      "epoch": 0.20069141740568164,
      "flos": 25958243940480.0,
      "grad_norm": 2.2067088808026023,
      "language_loss": 0.6936323,
      "learning_rate": 3.7020724086584724e-06,
      "loss": 0.71559489,
      "num_input_tokens_seen": 72126315,
      "step": 3338,
      "time_per_iteration": 2.5493454933166504
    },
    {
      "auxiliary_loss_clip": 0.01119045,
      "auxiliary_loss_mlp": 0.01052254,
      "balance_loss_clip": 1.05071568,
      "balance_loss_mlp": 1.03356171,
      "epoch": 0.2007515406583496,
      "flos": 24790608529920.0,
      "grad_norm": 2.063211476400424,
      "language_loss": 0.69063288,
      "learning_rate": 3.701867867326735e-06,
      "loss": 0.7123459,
      "num_input_tokens_seen": 72146470,
      "step": 3339,
      "time_per_iteration": 2.644824266433716
    },
    {
      "auxiliary_loss_clip": 0.01119451,
      "auxiliary_loss_mlp": 0.01041795,
      "balance_loss_clip": 1.05320835,
      "balance_loss_mlp": 1.02321076,
      "epoch": 0.2008116639110176,
      "flos": 37925582123520.0,
      "grad_norm": 2.0489630559883545,
      "language_loss": 0.66515994,
      "learning_rate": 3.7016632614599974e-06,
      "loss": 0.68677247,
      "num_input_tokens_seen": 72166600,
      "step": 3340,
      "time_per_iteration": 2.8311378955841064
    },
    {
      "auxiliary_loss_clip": 0.0114176,
      "auxiliary_loss_mlp": 0.01038486,
      "balance_loss_clip": 1.05174339,
      "balance_loss_mlp": 1.01934147,
      "epoch": 0.20087178716368556,
      "flos": 20740531443840.0,
      "grad_norm": 2.338644935214447,
      "language_loss": 0.74338686,
      "learning_rate": 3.701458591066019e-06,
      "loss": 0.76518929,
      "num_input_tokens_seen": 72185160,
      "step": 3341,
      "time_per_iteration": 2.52592396736145
    },
    {
      "auxiliary_loss_clip": 0.01111404,
      "auxiliary_loss_mlp": 0.01051628,
      "balance_loss_clip": 1.05327916,
      "balance_loss_mlp": 1.03279305,
      "epoch": 0.20093191041635353,
      "flos": 23842279607040.0,
      "grad_norm": 1.84177369327494,
      "language_loss": 0.71666443,
      "learning_rate": 3.70125385615256e-06,
      "loss": 0.73829478,
      "num_input_tokens_seen": 72205160,
      "step": 3342,
      "time_per_iteration": 2.670555591583252
    },
    {
      "auxiliary_loss_clip": 0.01112322,
      "auxiliary_loss_mlp": 0.01047178,
      "balance_loss_clip": 1.0468415,
      "balance_loss_mlp": 1.02909446,
      "epoch": 0.2009920336690215,
      "flos": 21792067119360.0,
      "grad_norm": 2.2208321943376967,
      "language_loss": 0.71996355,
      "learning_rate": 3.701049056727384e-06,
      "loss": 0.74155855,
      "num_input_tokens_seen": 72223555,
      "step": 3343,
      "time_per_iteration": 2.6304030418395996
    },
    {
      "auxiliary_loss_clip": 0.01113556,
      "auxiliary_loss_mlp": 0.01053421,
      "balance_loss_clip": 1.04809737,
      "balance_loss_mlp": 1.03414476,
      "epoch": 0.20105215692168946,
      "flos": 26359222440960.0,
      "grad_norm": 1.9789234633667807,
      "language_loss": 0.8015092,
      "learning_rate": 3.7008441927982574e-06,
      "loss": 0.82317901,
      "num_input_tokens_seen": 72242465,
      "step": 3344,
      "time_per_iteration": 2.63647198677063
    },
    {
      "auxiliary_loss_clip": 0.01149291,
      "auxiliary_loss_mlp": 0.01044273,
      "balance_loss_clip": 1.04955482,
      "balance_loss_mlp": 1.02592683,
      "epoch": 0.20111228017435742,
      "flos": 18807280617600.0,
      "grad_norm": 2.1237143386125594,
      "language_loss": 0.83716571,
      "learning_rate": 3.700639264372948e-06,
      "loss": 0.85910136,
      "num_input_tokens_seen": 72260655,
      "step": 3345,
      "time_per_iteration": 2.4709863662719727
    },
    {
      "auxiliary_loss_clip": 0.01093822,
      "auxiliary_loss_mlp": 0.01041785,
      "balance_loss_clip": 1.04705751,
      "balance_loss_mlp": 1.02477407,
      "epoch": 0.20117240342702541,
      "flos": 19975059682560.0,
      "grad_norm": 1.6355700774322997,
      "language_loss": 0.67957461,
      "learning_rate": 3.7004342714592283e-06,
      "loss": 0.70093071,
      "num_input_tokens_seen": 72279055,
      "step": 3346,
      "time_per_iteration": 2.6116139888763428
    },
    {
      "auxiliary_loss_clip": 0.01115378,
      "auxiliary_loss_mlp": 0.01050152,
      "balance_loss_clip": 1.04713345,
      "balance_loss_mlp": 1.03124511,
      "epoch": 0.20123252667969338,
      "flos": 23142703345920.0,
      "grad_norm": 2.058296242408319,
      "language_loss": 0.73106849,
      "learning_rate": 3.70022921406487e-06,
      "loss": 0.75272381,
      "num_input_tokens_seen": 72297895,
      "step": 3347,
      "time_per_iteration": 2.581087112426758
    },
    {
      "auxiliary_loss_clip": 0.01137605,
      "auxiliary_loss_mlp": 0.01053706,
      "balance_loss_clip": 1.05154061,
      "balance_loss_mlp": 1.03670716,
      "epoch": 0.20129264993236134,
      "flos": 23221671396480.0,
      "grad_norm": 1.6423403697671757,
      "language_loss": 0.86628306,
      "learning_rate": 3.70002409219765e-06,
      "loss": 0.88819611,
      "num_input_tokens_seen": 72318385,
      "step": 3348,
      "time_per_iteration": 2.5450663566589355
    },
    {
      "auxiliary_loss_clip": 0.01096207,
      "auxiliary_loss_mlp": 0.01044217,
      "balance_loss_clip": 1.04679334,
      "balance_loss_mlp": 1.02471399,
      "epoch": 0.2013527731850293,
      "flos": 21871466133120.0,
      "grad_norm": 1.7288029526777509,
      "language_loss": 0.71020436,
      "learning_rate": 3.699818905865346e-06,
      "loss": 0.73160863,
      "num_input_tokens_seen": 72338235,
      "step": 3349,
      "time_per_iteration": 2.627180814743042
    },
    {
      "auxiliary_loss_clip": 0.01116914,
      "auxiliary_loss_mlp": 0.01053902,
      "balance_loss_clip": 1.04890513,
      "balance_loss_mlp": 1.03435183,
      "epoch": 0.20141289643769728,
      "flos": 18040803275520.0,
      "grad_norm": 1.83893083213916,
      "language_loss": 0.71196723,
      "learning_rate": 3.6996136550757377e-06,
      "loss": 0.73367536,
      "num_input_tokens_seen": 72357825,
      "step": 3350,
      "time_per_iteration": 2.5875754356384277
    },
    {
      "auxiliary_loss_clip": 0.01127812,
      "auxiliary_loss_mlp": 0.01056028,
      "balance_loss_clip": 1.05205274,
      "balance_loss_mlp": 1.03286529,
      "epoch": 0.20147301969036524,
      "flos": 23951412103680.0,
      "grad_norm": 2.3414474234421454,
      "language_loss": 0.75847638,
      "learning_rate": 3.69940833983661e-06,
      "loss": 0.7803148,
      "num_input_tokens_seen": 72376335,
      "step": 3351,
      "time_per_iteration": 2.5649807453155518
    },
    {
      "auxiliary_loss_clip": 0.01131007,
      "auxiliary_loss_mlp": 0.01046896,
      "balance_loss_clip": 1.05120897,
      "balance_loss_mlp": 1.02697659,
      "epoch": 0.2015331429430332,
      "flos": 25588471380480.0,
      "grad_norm": 1.4758640867766386,
      "language_loss": 0.80542445,
      "learning_rate": 3.699202960155748e-06,
      "loss": 0.82720339,
      "num_input_tokens_seen": 72395440,
      "step": 3352,
      "time_per_iteration": 2.5943145751953125
    },
    {
      "auxiliary_loss_clip": 0.01140844,
      "auxiliary_loss_mlp": 0.01047548,
      "balance_loss_clip": 1.05122375,
      "balance_loss_mlp": 1.02913046,
      "epoch": 0.2015932661957012,
      "flos": 26724972677760.0,
      "grad_norm": 2.690596434950351,
      "language_loss": 0.80268884,
      "learning_rate": 3.6989975160409396e-06,
      "loss": 0.82457274,
      "num_input_tokens_seen": 72414670,
      "step": 3353,
      "time_per_iteration": 4.172781229019165
    },
    {
      "auxiliary_loss_clip": 0.01118083,
      "auxiliary_loss_mlp": 0.01044453,
      "balance_loss_clip": 1.04795408,
      "balance_loss_mlp": 1.02684593,
      "epoch": 0.20165338944836916,
      "flos": 15633136592640.0,
      "grad_norm": 2.320681021136594,
      "language_loss": 0.89928472,
      "learning_rate": 3.6987920074999747e-06,
      "loss": 0.92091012,
      "num_input_tokens_seen": 72432210,
      "step": 3354,
      "time_per_iteration": 3.928866386413574
    },
    {
      "auxiliary_loss_clip": 0.01048553,
      "auxiliary_loss_mlp": 0.00978983,
      "balance_loss_clip": 1.03141248,
      "balance_loss_mlp": 1.33880997,
      "epoch": 0.20171351270103713,
      "flos": 57912529207680.0,
      "grad_norm": 0.8324849460296164,
      "language_loss": 0.55855614,
      "learning_rate": 3.6985864345406465e-06,
      "loss": 0.57883149,
      "num_input_tokens_seen": 72489225,
      "step": 3355,
      "time_per_iteration": 3.11468505859375
    },
    {
      "auxiliary_loss_clip": 0.01127232,
      "auxiliary_loss_mlp": 0.00801082,
      "balance_loss_clip": 1.05213928,
      "balance_loss_mlp": 1.02076769,
      "epoch": 0.2017736359537051,
      "flos": 20814363849600.0,
      "grad_norm": 1.61663359533939,
      "language_loss": 0.84486943,
      "learning_rate": 3.698380797170751e-06,
      "loss": 0.86415255,
      "num_input_tokens_seen": 72508715,
      "step": 3356,
      "time_per_iteration": 2.584040641784668
    },
    {
      "auxiliary_loss_clip": 0.0112414,
      "auxiliary_loss_mlp": 0.01053547,
      "balance_loss_clip": 1.04831445,
      "balance_loss_mlp": 1.02968168,
      "epoch": 0.20183375920637306,
      "flos": 17092043389440.0,
      "grad_norm": 2.8516172506048583,
      "language_loss": 0.68872714,
      "learning_rate": 3.698175095398085e-06,
      "loss": 0.71050406,
      "num_input_tokens_seen": 72525135,
      "step": 3357,
      "time_per_iteration": 3.970665693283081
    },
    {
      "auxiliary_loss_clip": 0.01134433,
      "auxiliary_loss_mlp": 0.01047634,
      "balance_loss_clip": 1.05090189,
      "balance_loss_mlp": 1.02748775,
      "epoch": 0.20189388245904102,
      "flos": 18661339658880.0,
      "grad_norm": 1.7376036906702537,
      "language_loss": 0.72135162,
      "learning_rate": 3.6979693292304493e-06,
      "loss": 0.74317235,
      "num_input_tokens_seen": 72543690,
      "step": 3358,
      "time_per_iteration": 2.5461273193359375
    },
    {
      "auxiliary_loss_clip": 0.01134949,
      "auxiliary_loss_mlp": 0.01054668,
      "balance_loss_clip": 1.04839945,
      "balance_loss_mlp": 1.0371207,
      "epoch": 0.20195400571170902,
      "flos": 16797539779200.0,
      "grad_norm": 1.741878173003893,
      "language_loss": 0.83071709,
      "learning_rate": 3.6977634986756463e-06,
      "loss": 0.85261321,
      "num_input_tokens_seen": 72560725,
      "step": 3359,
      "time_per_iteration": 3.858407497406006
    },
    {
      "auxiliary_loss_clip": 0.01052243,
      "auxiliary_loss_mlp": 0.01024896,
      "balance_loss_clip": 1.02430153,
      "balance_loss_mlp": 1.02177262,
      "epoch": 0.20201412896437698,
      "flos": 67174716268800.0,
      "grad_norm": 0.7847262736373956,
      "language_loss": 0.58985806,
      "learning_rate": 3.697557603741482e-06,
      "loss": 0.61062944,
      "num_input_tokens_seen": 72621940,
      "step": 3360,
      "time_per_iteration": 3.0955324172973633
    },
    {
      "auxiliary_loss_clip": 0.0110051,
      "auxiliary_loss_mlp": 0.01057282,
      "balance_loss_clip": 1.05665207,
      "balance_loss_mlp": 1.03810108,
      "epoch": 0.20207425221704495,
      "flos": 21325013550720.0,
      "grad_norm": 2.257621568237434,
      "language_loss": 0.62243432,
      "learning_rate": 3.697351644435763e-06,
      "loss": 0.64401221,
      "num_input_tokens_seen": 72639135,
      "step": 3361,
      "time_per_iteration": 2.674290418624878
    },
    {
      "auxiliary_loss_clip": 0.01117051,
      "auxiliary_loss_mlp": 0.01063685,
      "balance_loss_clip": 1.05049837,
      "balance_loss_mlp": 1.04370522,
      "epoch": 0.2021343754697129,
      "flos": 22527158952960.0,
      "grad_norm": 1.9952915689815687,
      "language_loss": 0.75854146,
      "learning_rate": 3.6971456207662993e-06,
      "loss": 0.7803489,
      "num_input_tokens_seen": 72658525,
      "step": 3362,
      "time_per_iteration": 2.595655918121338
    },
    {
      "auxiliary_loss_clip": 0.01137211,
      "auxiliary_loss_mlp": 0.00826532,
      "balance_loss_clip": 1.05080938,
      "balance_loss_mlp": 1.06649733,
      "epoch": 0.20219449872238088,
      "flos": 19062785036160.0,
      "grad_norm": 1.6573573086285027,
      "language_loss": 0.76777142,
      "learning_rate": 3.6969395327409035e-06,
      "loss": 0.78740883,
      "num_input_tokens_seen": 72678085,
      "step": 3363,
      "time_per_iteration": 2.5721564292907715
    },
    {
      "auxiliary_loss_clip": 0.01136478,
      "auxiliary_loss_mlp": 0.01048459,
      "balance_loss_clip": 1.04821301,
      "balance_loss_mlp": 1.03129244,
      "epoch": 0.20225462197504884,
      "flos": 24717027519360.0,
      "grad_norm": 1.5058919481013322,
      "language_loss": 0.7544868,
      "learning_rate": 3.696733380367391e-06,
      "loss": 0.77633607,
      "num_input_tokens_seen": 72698695,
      "step": 3364,
      "time_per_iteration": 2.5697171688079834
    },
    {
      "auxiliary_loss_clip": 0.01104986,
      "auxiliary_loss_mlp": 0.01043433,
      "balance_loss_clip": 1.05116248,
      "balance_loss_mlp": 1.02373981,
      "epoch": 0.2023147452277168,
      "flos": 22018304931840.0,
      "grad_norm": 10.323511875801541,
      "language_loss": 0.71414173,
      "learning_rate": 3.6965271636535783e-06,
      "loss": 0.73562586,
      "num_input_tokens_seen": 72717880,
      "step": 3365,
      "time_per_iteration": 2.625694513320923
    },
    {
      "auxiliary_loss_clip": 0.01104407,
      "auxiliary_loss_mlp": 0.01047448,
      "balance_loss_clip": 1.04990864,
      "balance_loss_mlp": 1.02882802,
      "epoch": 0.2023748684803848,
      "flos": 17745365911680.0,
      "grad_norm": 2.704703847054694,
      "language_loss": 0.86421734,
      "learning_rate": 3.696320882607286e-06,
      "loss": 0.88573587,
      "num_input_tokens_seen": 72736410,
      "step": 3366,
      "time_per_iteration": 2.5391907691955566
    },
    {
      "auxiliary_loss_clip": 0.01113048,
      "auxiliary_loss_mlp": 0.01040418,
      "balance_loss_clip": 1.05169654,
      "balance_loss_mlp": 1.0220716,
      "epoch": 0.20243499173305277,
      "flos": 31138932493440.0,
      "grad_norm": 2.098987858712941,
      "language_loss": 0.69654167,
      "learning_rate": 3.696114537236335e-06,
      "loss": 0.71807635,
      "num_input_tokens_seen": 72758295,
      "step": 3367,
      "time_per_iteration": 2.6517255306243896
    },
    {
      "auxiliary_loss_clip": 0.01140334,
      "auxiliary_loss_mlp": 0.01045332,
      "balance_loss_clip": 1.04980648,
      "balance_loss_mlp": 1.02413702,
      "epoch": 0.20249511498572073,
      "flos": 33839235279360.0,
      "grad_norm": 1.9516157635528146,
      "language_loss": 0.68892878,
      "learning_rate": 3.6959081275485512e-06,
      "loss": 0.71078539,
      "num_input_tokens_seen": 72782495,
      "step": 3368,
      "time_per_iteration": 2.6133289337158203
    },
    {
      "auxiliary_loss_clip": 0.01118199,
      "auxiliary_loss_mlp": 0.0105145,
      "balance_loss_clip": 1.05357778,
      "balance_loss_mlp": 1.03205431,
      "epoch": 0.2025552382383887,
      "flos": 21215629658880.0,
      "grad_norm": 2.0170485633658672,
      "language_loss": 0.77734435,
      "learning_rate": 3.6957016535517615e-06,
      "loss": 0.79904079,
      "num_input_tokens_seen": 72801885,
      "step": 3369,
      "time_per_iteration": 2.5903189182281494
    },
    {
      "auxiliary_loss_clip": 0.01132787,
      "auxiliary_loss_mlp": 0.01053794,
      "balance_loss_clip": 1.04888058,
      "balance_loss_mlp": 1.0350666,
      "epoch": 0.20261536149105666,
      "flos": 14647388676480.0,
      "grad_norm": 2.902896481472456,
      "language_loss": 0.65016598,
      "learning_rate": 3.695495115253795e-06,
      "loss": 0.67203176,
      "num_input_tokens_seen": 72816990,
      "step": 3370,
      "time_per_iteration": 2.482893943786621
    },
    {
      "auxiliary_loss_clip": 0.01053408,
      "auxiliary_loss_mlp": 0.01000056,
      "balance_loss_clip": 1.02631986,
      "balance_loss_mlp": 0.99736196,
      "epoch": 0.20267548474372463,
      "flos": 66783649921920.0,
      "grad_norm": 0.6814662131243528,
      "language_loss": 0.58147174,
      "learning_rate": 3.6952885126624834e-06,
      "loss": 0.60200638,
      "num_input_tokens_seen": 72879240,
      "step": 3371,
      "time_per_iteration": 3.182375192642212
    },
    {
      "auxiliary_loss_clip": 0.01117187,
      "auxiliary_loss_mlp": 0.01043003,
      "balance_loss_clip": 1.0477736,
      "balance_loss_mlp": 1.02487159,
      "epoch": 0.2027356079963926,
      "flos": 24680793674880.0,
      "grad_norm": 2.1524260264049198,
      "language_loss": 0.91829985,
      "learning_rate": 3.6950818457856617e-06,
      "loss": 0.93990177,
      "num_input_tokens_seen": 72899030,
      "step": 3372,
      "time_per_iteration": 2.61246657371521
    },
    {
      "auxiliary_loss_clip": 0.01133616,
      "auxiliary_loss_mlp": 0.01051763,
      "balance_loss_clip": 1.04825962,
      "balance_loss_mlp": 1.03030539,
      "epoch": 0.20279573124906058,
      "flos": 26392762765440.0,
      "grad_norm": 1.60092624523118,
      "language_loss": 0.78727973,
      "learning_rate": 3.694875114631167e-06,
      "loss": 0.80913353,
      "num_input_tokens_seen": 72919190,
      "step": 3373,
      "time_per_iteration": 2.5494368076324463
    },
    {
      "auxiliary_loss_clip": 0.01089379,
      "auxiliary_loss_mlp": 0.01045383,
      "balance_loss_clip": 1.0457058,
      "balance_loss_mlp": 1.02406788,
      "epoch": 0.20285585450172855,
      "flos": 33799984692480.0,
      "grad_norm": 1.7725651742733801,
      "language_loss": 0.71680099,
      "learning_rate": 3.6946683192068377e-06,
      "loss": 0.73814863,
      "num_input_tokens_seen": 72939720,
      "step": 3374,
      "time_per_iteration": 2.7234768867492676
    },
    {
      "auxiliary_loss_clip": 0.01047409,
      "auxiliary_loss_mlp": 0.01001296,
      "balance_loss_clip": 1.0273037,
      "balance_loss_mlp": 0.99822026,
      "epoch": 0.20291597775439651,
      "flos": 71164823598720.0,
      "grad_norm": 0.9715764246786197,
      "language_loss": 0.62442183,
      "learning_rate": 3.694461459520516e-06,
      "loss": 0.64490891,
      "num_input_tokens_seen": 73000015,
      "step": 3375,
      "time_per_iteration": 3.1200056076049805
    },
    {
      "auxiliary_loss_clip": 0.01147906,
      "auxiliary_loss_mlp": 0.01045948,
      "balance_loss_clip": 1.05000818,
      "balance_loss_mlp": 1.02793574,
      "epoch": 0.20297610100706448,
      "flos": 19494287118720.0,
      "grad_norm": 1.5463271629893487,
      "language_loss": 0.8215481,
      "learning_rate": 3.6942545355800463e-06,
      "loss": 0.84348667,
      "num_input_tokens_seen": 73017675,
      "step": 3376,
      "time_per_iteration": 2.480059862136841
    },
    {
      "auxiliary_loss_clip": 0.01136414,
      "auxiliary_loss_mlp": 0.01039819,
      "balance_loss_clip": 1.04845893,
      "balance_loss_mlp": 1.02012515,
      "epoch": 0.20303622425973245,
      "flos": 25044245441280.0,
      "grad_norm": 2.221221824254492,
      "language_loss": 0.81850249,
      "learning_rate": 3.6940475473932743e-06,
      "loss": 0.8402648,
      "num_input_tokens_seen": 73036135,
      "step": 3377,
      "time_per_iteration": 2.5193910598754883
    },
    {
      "auxiliary_loss_clip": 0.01120298,
      "auxiliary_loss_mlp": 0.0105242,
      "balance_loss_clip": 1.05216753,
      "balance_loss_mlp": 1.03291714,
      "epoch": 0.2030963475124004,
      "flos": 21979988098560.0,
      "grad_norm": 1.95070036799559,
      "language_loss": 0.7696861,
      "learning_rate": 3.69384049496805e-06,
      "loss": 0.79141325,
      "num_input_tokens_seen": 73054075,
      "step": 3378,
      "time_per_iteration": 2.5563714504241943
    },
    {
      "auxiliary_loss_clip": 0.01088527,
      "auxiliary_loss_mlp": 0.01054459,
      "balance_loss_clip": 1.04987347,
      "balance_loss_mlp": 1.03264368,
      "epoch": 0.2031564707650684,
      "flos": 19500392430720.0,
      "grad_norm": 2.3612405014591786,
      "language_loss": 0.79645985,
      "learning_rate": 3.6936333783122242e-06,
      "loss": 0.81788969,
      "num_input_tokens_seen": 73073530,
      "step": 3379,
      "time_per_iteration": 2.6220247745513916
    },
    {
      "auxiliary_loss_clip": 0.01134756,
      "auxiliary_loss_mlp": 0.01044217,
      "balance_loss_clip": 1.0528307,
      "balance_loss_mlp": 1.02665746,
      "epoch": 0.20321659401773637,
      "flos": 22747075971840.0,
      "grad_norm": 1.9010308785150776,
      "language_loss": 0.86655867,
      "learning_rate": 3.6934261974336505e-06,
      "loss": 0.88834834,
      "num_input_tokens_seen": 73092820,
      "step": 3380,
      "time_per_iteration": 2.5565786361694336
    },
    {
      "auxiliary_loss_clip": 0.01151954,
      "auxiliary_loss_mlp": 0.01048846,
      "balance_loss_clip": 1.05479467,
      "balance_loss_mlp": 1.03076172,
      "epoch": 0.20327671727040433,
      "flos": 22455840499200.0,
      "grad_norm": 2.227354497010493,
      "language_loss": 0.75072646,
      "learning_rate": 3.693218952340186e-06,
      "loss": 0.7727344,
      "num_input_tokens_seen": 73113385,
      "step": 3381,
      "time_per_iteration": 2.5147640705108643
    },
    {
      "auxiliary_loss_clip": 0.01116513,
      "auxiliary_loss_mlp": 0.01058212,
      "balance_loss_clip": 1.04741263,
      "balance_loss_mlp": 1.03953242,
      "epoch": 0.2033368405230723,
      "flos": 19535010163200.0,
      "grad_norm": 1.640897769490943,
      "language_loss": 0.79330939,
      "learning_rate": 3.6930116430396895e-06,
      "loss": 0.81505668,
      "num_input_tokens_seen": 73131195,
      "step": 3382,
      "time_per_iteration": 2.574390172958374
    },
    {
      "auxiliary_loss_clip": 0.01111734,
      "auxiliary_loss_mlp": 0.0080974,
      "balance_loss_clip": 1.04907501,
      "balance_loss_mlp": 1.03494632,
      "epoch": 0.20339696377574026,
      "flos": 13809233744640.0,
      "grad_norm": 2.9480296338349135,
      "language_loss": 0.80346853,
      "learning_rate": 3.6928042695400214e-06,
      "loss": 0.82268327,
      "num_input_tokens_seen": 73148850,
      "step": 3383,
      "time_per_iteration": 2.567694664001465
    },
    {
      "auxiliary_loss_clip": 0.01097145,
      "auxiliary_loss_mlp": 0.01039841,
      "balance_loss_clip": 1.04496074,
      "balance_loss_mlp": 1.02094591,
      "epoch": 0.20345708702840823,
      "flos": 20339409288960.0,
      "grad_norm": 2.023666106868807,
      "language_loss": 0.74414682,
      "learning_rate": 3.6925968318490464e-06,
      "loss": 0.76551664,
      "num_input_tokens_seen": 73166775,
      "step": 3384,
      "time_per_iteration": 2.577505111694336
    },
    {
      "auxiliary_loss_clip": 0.01139534,
      "auxiliary_loss_mlp": 0.01048769,
      "balance_loss_clip": 1.04862809,
      "balance_loss_mlp": 1.0288018,
      "epoch": 0.2035172102810762,
      "flos": 20333950421760.0,
      "grad_norm": 6.119993973992206,
      "language_loss": 0.76722944,
      "learning_rate": 3.6923893299746293e-06,
      "loss": 0.78911245,
      "num_input_tokens_seen": 73183215,
      "step": 3385,
      "time_per_iteration": 2.5399293899536133
    },
    {
      "auxiliary_loss_clip": 0.01107435,
      "auxiliary_loss_mlp": 0.01064399,
      "balance_loss_clip": 1.05225682,
      "balance_loss_mlp": 1.04445577,
      "epoch": 0.2035773335337442,
      "flos": 23330983461120.0,
      "grad_norm": 1.5005830569086331,
      "language_loss": 0.68473536,
      "learning_rate": 3.692181763924639e-06,
      "loss": 0.70645368,
      "num_input_tokens_seen": 73203290,
      "step": 3386,
      "time_per_iteration": 2.642139196395874
    },
    {
      "auxiliary_loss_clip": 0.01101352,
      "auxiliary_loss_mlp": 0.01059582,
      "balance_loss_clip": 1.04929352,
      "balance_loss_mlp": 1.03945947,
      "epoch": 0.20363745678641215,
      "flos": 28330287310080.0,
      "grad_norm": 1.357919103899785,
      "language_loss": 0.81207776,
      "learning_rate": 3.691974133706947e-06,
      "loss": 0.83368713,
      "num_input_tokens_seen": 73226185,
      "step": 3387,
      "time_per_iteration": 2.7188615798950195
    },
    {
      "auxiliary_loss_clip": 0.01115115,
      "auxiliary_loss_mlp": 0.01044149,
      "balance_loss_clip": 1.04717243,
      "balance_loss_mlp": 1.02575457,
      "epoch": 0.20369758003908012,
      "flos": 18915658928640.0,
      "grad_norm": 3.344659379482522,
      "language_loss": 0.79439509,
      "learning_rate": 3.6917664393294262e-06,
      "loss": 0.81598771,
      "num_input_tokens_seen": 73243300,
      "step": 3388,
      "time_per_iteration": 2.5606720447540283
    },
    {
      "auxiliary_loss_clip": 0.01149461,
      "auxiliary_loss_mlp": 0.01044902,
      "balance_loss_clip": 1.05050468,
      "balance_loss_mlp": 1.02574515,
      "epoch": 0.20375770329174808,
      "flos": 19206499351680.0,
      "grad_norm": 1.7902276710875138,
      "language_loss": 0.71979403,
      "learning_rate": 3.6915586807999527e-06,
      "loss": 0.7417376,
      "num_input_tokens_seen": 73261490,
      "step": 3389,
      "time_per_iteration": 2.488412618637085
    },
    {
      "auxiliary_loss_clip": 0.01134344,
      "auxiliary_loss_mlp": 0.01047232,
      "balance_loss_clip": 1.04932392,
      "balance_loss_mlp": 1.02894545,
      "epoch": 0.20381782654441605,
      "flos": 19391008538880.0,
      "grad_norm": 2.1677307904641396,
      "language_loss": 0.86921829,
      "learning_rate": 3.691350858126404e-06,
      "loss": 0.89103413,
      "num_input_tokens_seen": 73280180,
      "step": 3390,
      "time_per_iteration": 2.5605275630950928
    },
    {
      "auxiliary_loss_clip": 0.01117723,
      "auxiliary_loss_mlp": 0.0105723,
      "balance_loss_clip": 1.04772973,
      "balance_loss_mlp": 1.03716719,
      "epoch": 0.203877949797084,
      "flos": 24827704300800.0,
      "grad_norm": 2.3006120129509293,
      "language_loss": 0.70977181,
      "learning_rate": 3.691142971316662e-06,
      "loss": 0.73152137,
      "num_input_tokens_seen": 73300680,
      "step": 3391,
      "time_per_iteration": 4.018676042556763
    },
    {
      "auxiliary_loss_clip": 0.01112741,
      "auxiliary_loss_mlp": 0.01048196,
      "balance_loss_clip": 1.0483191,
      "balance_loss_mlp": 1.02993274,
      "epoch": 0.20393807304975198,
      "flos": 18003707504640.0,
      "grad_norm": 2.4592445544482304,
      "language_loss": 0.86422455,
      "learning_rate": 3.6909350203786086e-06,
      "loss": 0.88583386,
      "num_input_tokens_seen": 73316760,
      "step": 3392,
      "time_per_iteration": 4.007593631744385
    },
    {
      "auxiliary_loss_clip": 0.01137903,
      "auxiliary_loss_mlp": 0.01048985,
      "balance_loss_clip": 1.04854882,
      "balance_loss_mlp": 1.03138924,
      "epoch": 0.20399819630241997,
      "flos": 24206988349440.0,
      "grad_norm": 2.1471200240705466,
      "language_loss": 0.80460072,
      "learning_rate": 3.69072700532013e-06,
      "loss": 0.8264696,
      "num_input_tokens_seen": 73339385,
      "step": 3393,
      "time_per_iteration": 2.566272258758545
    },
    {
      "auxiliary_loss_clip": 0.01117143,
      "auxiliary_loss_mlp": 0.01037235,
      "balance_loss_clip": 1.04702795,
      "balance_loss_mlp": 1.01971138,
      "epoch": 0.20405831955508794,
      "flos": 20777124424320.0,
      "grad_norm": 1.804341172107247,
      "language_loss": 0.86395991,
      "learning_rate": 3.6905189261491137e-06,
      "loss": 0.88550371,
      "num_input_tokens_seen": 73357235,
      "step": 3394,
      "time_per_iteration": 2.545711040496826
    },
    {
      "auxiliary_loss_clip": 0.01131237,
      "auxiliary_loss_mlp": 0.01048289,
      "balance_loss_clip": 1.05103517,
      "balance_loss_mlp": 1.03128982,
      "epoch": 0.2041184428077559,
      "flos": 15486908325120.0,
      "grad_norm": 2.4221781663366357,
      "language_loss": 0.84440231,
      "learning_rate": 3.69031078287345e-06,
      "loss": 0.86619753,
      "num_input_tokens_seen": 73374435,
      "step": 3395,
      "time_per_iteration": 2.475126266479492
    },
    {
      "auxiliary_loss_clip": 0.01136062,
      "auxiliary_loss_mlp": 0.01037693,
      "balance_loss_clip": 1.04932463,
      "balance_loss_mlp": 1.01914358,
      "epoch": 0.20417856606042387,
      "flos": 15588463052160.0,
      "grad_norm": 2.048376689197412,
      "language_loss": 0.84105361,
      "learning_rate": 3.690102575501033e-06,
      "loss": 0.86279112,
      "num_input_tokens_seen": 73391025,
      "step": 3396,
      "time_per_iteration": 3.9363181591033936
    },
    {
      "auxiliary_loss_clip": 0.01108087,
      "auxiliary_loss_mlp": 0.01044157,
      "balance_loss_clip": 1.04722571,
      "balance_loss_mlp": 1.02519059,
      "epoch": 0.20423868931309183,
      "flos": 24279348297600.0,
      "grad_norm": 2.1598500764415394,
      "language_loss": 0.77469385,
      "learning_rate": 3.6898943040397556e-06,
      "loss": 0.79621625,
      "num_input_tokens_seen": 73409270,
      "step": 3397,
      "time_per_iteration": 2.6185953617095947
    },
    {
      "auxiliary_loss_clip": 0.0112147,
      "auxiliary_loss_mlp": 0.01045565,
      "balance_loss_clip": 1.04895675,
      "balance_loss_mlp": 1.02849424,
      "epoch": 0.2042988125657598,
      "flos": 18614870438400.0,
      "grad_norm": 3.192592019943527,
      "language_loss": 0.87949473,
      "learning_rate": 3.689685968497518e-06,
      "loss": 0.90116513,
      "num_input_tokens_seen": 73425225,
      "step": 3398,
      "time_per_iteration": 3.8984038829803467
    },
    {
      "auxiliary_loss_clip": 0.01118541,
      "auxiliary_loss_mlp": 0.01049934,
      "balance_loss_clip": 1.05268598,
      "balance_loss_mlp": 1.03157556,
      "epoch": 0.2043589358184278,
      "flos": 17851230270720.0,
      "grad_norm": 2.181160693169387,
      "language_loss": 0.78011405,
      "learning_rate": 3.6894775688822186e-06,
      "loss": 0.80179876,
      "num_input_tokens_seen": 73440940,
      "step": 3399,
      "time_per_iteration": 2.513711929321289
    },
    {
      "auxiliary_loss_clip": 0.01135573,
      "auxiliary_loss_mlp": 0.01038007,
      "balance_loss_clip": 1.04842651,
      "balance_loss_mlp": 1.0204711,
      "epoch": 0.20441905907109575,
      "flos": 21435223455360.0,
      "grad_norm": 2.125540109030586,
      "language_loss": 0.76421916,
      "learning_rate": 3.6892691052017603e-06,
      "loss": 0.78595495,
      "num_input_tokens_seen": 73458805,
      "step": 3400,
      "time_per_iteration": 2.5543415546417236
    },
    {
      "auxiliary_loss_clip": 0.01113306,
      "auxiliary_loss_mlp": 0.00801861,
      "balance_loss_clip": 1.05056381,
      "balance_loss_mlp": 1.02480555,
      "epoch": 0.20447918232376372,
      "flos": 27707703851520.0,
      "grad_norm": 1.8670433078103281,
      "language_loss": 0.7933653,
      "learning_rate": 3.6890605774640487e-06,
      "loss": 0.81251699,
      "num_input_tokens_seen": 73479380,
      "step": 3401,
      "time_per_iteration": 2.6463019847869873
    },
    {
      "auxiliary_loss_clip": 0.01123329,
      "auxiliary_loss_mlp": 0.01047433,
      "balance_loss_clip": 1.04622209,
      "balance_loss_mlp": 1.02881217,
      "epoch": 0.20453930557643168,
      "flos": 30524214113280.0,
      "grad_norm": 1.7273024926589577,
      "language_loss": 0.69289565,
      "learning_rate": 3.688851985676991e-06,
      "loss": 0.71460325,
      "num_input_tokens_seen": 73505105,
      "step": 3402,
      "time_per_iteration": 2.692523241043091
    },
    {
      "auxiliary_loss_clip": 0.01109087,
      "auxiliary_loss_mlp": 0.01042973,
      "balance_loss_clip": 1.04714298,
      "balance_loss_mlp": 1.0247575,
      "epoch": 0.20459942882909965,
      "flos": 18987767481600.0,
      "grad_norm": 3.172747727781626,
      "language_loss": 0.80823809,
      "learning_rate": 3.688643329848496e-06,
      "loss": 0.8297587,
      "num_input_tokens_seen": 73523700,
      "step": 3403,
      "time_per_iteration": 2.570510149002075
    },
    {
      "auxiliary_loss_clip": 0.01137626,
      "auxiliary_loss_mlp": 0.01045465,
      "balance_loss_clip": 1.05118382,
      "balance_loss_mlp": 1.02814448,
      "epoch": 0.20465955208176762,
      "flos": 20339050152960.0,
      "grad_norm": 2.03621676229595,
      "language_loss": 0.83449239,
      "learning_rate": 3.6884346099864772e-06,
      "loss": 0.8563233,
      "num_input_tokens_seen": 73542625,
      "step": 3404,
      "time_per_iteration": 2.513256549835205
    },
    {
      "auxiliary_loss_clip": 0.01130006,
      "auxiliary_loss_mlp": 0.01047823,
      "balance_loss_clip": 1.0448885,
      "balance_loss_mlp": 1.02935696,
      "epoch": 0.20471967533443558,
      "flos": 21251288885760.0,
      "grad_norm": 1.7714311089115549,
      "language_loss": 0.85973078,
      "learning_rate": 3.6882258260988487e-06,
      "loss": 0.88150907,
      "num_input_tokens_seen": 73561450,
      "step": 3405,
      "time_per_iteration": 2.545488119125366
    },
    {
      "auxiliary_loss_clip": 0.01107884,
      "auxiliary_loss_mlp": 0.01043908,
      "balance_loss_clip": 1.04916358,
      "balance_loss_mlp": 1.02655089,
      "epoch": 0.20477979858710357,
      "flos": 14501555458560.0,
      "grad_norm": 2.144152749989942,
      "language_loss": 0.84130383,
      "learning_rate": 3.6880169781935276e-06,
      "loss": 0.8628217,
      "num_input_tokens_seen": 73577155,
      "step": 3406,
      "time_per_iteration": 2.5298378467559814
    },
    {
      "auxiliary_loss_clip": 0.01147453,
      "auxiliary_loss_mlp": 0.01041267,
      "balance_loss_clip": 1.05124187,
      "balance_loss_mlp": 1.02444661,
      "epoch": 0.20483992183977154,
      "flos": 11400310085760.0,
      "grad_norm": 2.0364324443209885,
      "language_loss": 0.67775559,
      "learning_rate": 3.6878080662784336e-06,
      "loss": 0.69964284,
      "num_input_tokens_seen": 73594900,
      "step": 3407,
      "time_per_iteration": 2.473159074783325
    },
    {
      "auxiliary_loss_clip": 0.01144244,
      "auxiliary_loss_mlp": 0.01047457,
      "balance_loss_clip": 1.04869151,
      "balance_loss_mlp": 1.02981377,
      "epoch": 0.2049000450924395,
      "flos": 19060271084160.0,
      "grad_norm": 2.1749953592653455,
      "language_loss": 0.83954334,
      "learning_rate": 3.6875990903614886e-06,
      "loss": 0.86146033,
      "num_input_tokens_seen": 73613810,
      "step": 3408,
      "time_per_iteration": 2.5187361240386963
    },
    {
      "auxiliary_loss_clip": 0.01148888,
      "auxiliary_loss_mlp": 0.01042384,
      "balance_loss_clip": 1.05118895,
      "balance_loss_mlp": 1.02538431,
      "epoch": 0.20496016834510747,
      "flos": 14574561851520.0,
      "grad_norm": 3.421354100479567,
      "language_loss": 0.63954228,
      "learning_rate": 3.6873900504506166e-06,
      "loss": 0.66145504,
      "num_input_tokens_seen": 73631495,
      "step": 3409,
      "time_per_iteration": 2.466261148452759
    },
    {
      "auxiliary_loss_clip": 0.01129922,
      "auxiliary_loss_mlp": 0.01045837,
      "balance_loss_clip": 1.04741991,
      "balance_loss_mlp": 1.02853978,
      "epoch": 0.20502029159777543,
      "flos": 22126647329280.0,
      "grad_norm": 1.6486648578703644,
      "language_loss": 0.8011961,
      "learning_rate": 3.687180946553745e-06,
      "loss": 0.8229537,
      "num_input_tokens_seen": 73652840,
      "step": 3410,
      "time_per_iteration": 2.520407199859619
    },
    {
      "auxiliary_loss_clip": 0.01092687,
      "auxiliary_loss_mlp": 0.01046492,
      "balance_loss_clip": 1.04999197,
      "balance_loss_mlp": 1.02932596,
      "epoch": 0.2050804148504434,
      "flos": 25367907916800.0,
      "grad_norm": 2.3483070692916925,
      "language_loss": 0.76493222,
      "learning_rate": 3.686971778678803e-06,
      "loss": 0.78632396,
      "num_input_tokens_seen": 73672150,
      "step": 3411,
      "time_per_iteration": 2.6909637451171875
    },
    {
      "auxiliary_loss_clip": 0.01128703,
      "auxiliary_loss_mlp": 0.01048376,
      "balance_loss_clip": 1.04897308,
      "balance_loss_mlp": 1.03116262,
      "epoch": 0.2051405381031114,
      "flos": 23620171858560.0,
      "grad_norm": 1.9353367243453121,
      "language_loss": 0.73537368,
      "learning_rate": 3.686762546833722e-06,
      "loss": 0.75714445,
      "num_input_tokens_seen": 73691940,
      "step": 3412,
      "time_per_iteration": 2.5275936126708984
    },
    {
      "auxiliary_loss_clip": 0.01119609,
      "auxiliary_loss_mlp": 0.01055457,
      "balance_loss_clip": 1.04464936,
      "balance_loss_mlp": 1.03485715,
      "epoch": 0.20520066135577936,
      "flos": 19565533745280.0,
      "grad_norm": 2.484238108751062,
      "language_loss": 0.77972448,
      "learning_rate": 3.6865532510264362e-06,
      "loss": 0.80147517,
      "num_input_tokens_seen": 73709080,
      "step": 3413,
      "time_per_iteration": 2.5265417098999023
    },
    {
      "auxiliary_loss_clip": 0.0110087,
      "auxiliary_loss_mlp": 0.01052146,
      "balance_loss_clip": 1.04630864,
      "balance_loss_mlp": 1.03316784,
      "epoch": 0.20526078460844732,
      "flos": 17676345928320.0,
      "grad_norm": 3.2611945975981427,
      "language_loss": 0.84530413,
      "learning_rate": 3.6863438912648823e-06,
      "loss": 0.86683428,
      "num_input_tokens_seen": 73727670,
      "step": 3414,
      "time_per_iteration": 2.5221221446990967
    },
    {
      "auxiliary_loss_clip": 0.01132518,
      "auxiliary_loss_mlp": 0.01045839,
      "balance_loss_clip": 1.04634273,
      "balance_loss_mlp": 1.02723074,
      "epoch": 0.2053209078611153,
      "flos": 21500328856320.0,
      "grad_norm": 1.8671865703084682,
      "language_loss": 0.81041133,
      "learning_rate": 3.6861344675569986e-06,
      "loss": 0.83219486,
      "num_input_tokens_seen": 73747170,
      "step": 3415,
      "time_per_iteration": 2.514615535736084
    },
    {
      "auxiliary_loss_clip": 0.01077865,
      "auxiliary_loss_mlp": 0.01039683,
      "balance_loss_clip": 1.0477798,
      "balance_loss_mlp": 1.02348268,
      "epoch": 0.20538103111378325,
      "flos": 25663524848640.0,
      "grad_norm": 2.024348109670658,
      "language_loss": 0.72763002,
      "learning_rate": 3.6859249799107275e-06,
      "loss": 0.7488054,
      "num_input_tokens_seen": 73767690,
      "step": 3416,
      "time_per_iteration": 2.6654114723205566
    },
    {
      "auxiliary_loss_clip": 0.01138136,
      "auxiliary_loss_mlp": 0.01041829,
      "balance_loss_clip": 1.0500977,
      "balance_loss_mlp": 1.0232203,
      "epoch": 0.20544115436645122,
      "flos": 23148952312320.0,
      "grad_norm": 2.2291375570070593,
      "language_loss": 0.78635758,
      "learning_rate": 3.6857154283340115e-06,
      "loss": 0.80815721,
      "num_input_tokens_seen": 73786900,
      "step": 3417,
      "time_per_iteration": 2.527583360671997
    },
    {
      "auxiliary_loss_clip": 0.01136522,
      "auxiliary_loss_mlp": 0.01045099,
      "balance_loss_clip": 1.04886508,
      "balance_loss_mlp": 1.02687216,
      "epoch": 0.20550127761911918,
      "flos": 19390433921280.0,
      "grad_norm": 2.9451885098947512,
      "language_loss": 0.8769511,
      "learning_rate": 3.685505812834798e-06,
      "loss": 0.89876729,
      "num_input_tokens_seen": 73804515,
      "step": 3418,
      "time_per_iteration": 2.4742672443389893
    },
    {
      "auxiliary_loss_clip": 0.01128303,
      "auxiliary_loss_mlp": 0.0104354,
      "balance_loss_clip": 1.0485692,
      "balance_loss_mlp": 1.02556312,
      "epoch": 0.20556140087178718,
      "flos": 22893124671360.0,
      "grad_norm": 2.4939992866261087,
      "language_loss": 0.6285187,
      "learning_rate": 3.685296133421035e-06,
      "loss": 0.65023714,
      "num_input_tokens_seen": 73822910,
      "step": 3419,
      "time_per_iteration": 2.542381525039673
    },
    {
      "auxiliary_loss_clip": 0.01134023,
      "auxiliary_loss_mlp": 0.01052252,
      "balance_loss_clip": 1.05439067,
      "balance_loss_mlp": 1.03264165,
      "epoch": 0.20562152412445514,
      "flos": 19789652655360.0,
      "grad_norm": 3.658493039100118,
      "language_loss": 0.86184001,
      "learning_rate": 3.685086390100674e-06,
      "loss": 0.88370275,
      "num_input_tokens_seen": 73841160,
      "step": 3420,
      "time_per_iteration": 2.548982620239258
    },
    {
      "auxiliary_loss_clip": 0.01099273,
      "auxiliary_loss_mlp": 0.00805346,
      "balance_loss_clip": 1.0454874,
      "balance_loss_mlp": 1.01867366,
      "epoch": 0.2056816473771231,
      "flos": 31501989210240.0,
      "grad_norm": 3.7589481241773663,
      "language_loss": 0.71296239,
      "learning_rate": 3.684876582881668e-06,
      "loss": 0.73200864,
      "num_input_tokens_seen": 73862795,
      "step": 3421,
      "time_per_iteration": 2.711793899536133
    },
    {
      "auxiliary_loss_clip": 0.01144457,
      "auxiliary_loss_mlp": 0.01038638,
      "balance_loss_clip": 1.04959691,
      "balance_loss_mlp": 1.02018464,
      "epoch": 0.20574177062979107,
      "flos": 23258372117760.0,
      "grad_norm": 1.909110658739985,
      "language_loss": 0.70873851,
      "learning_rate": 3.6846667117719732e-06,
      "loss": 0.73056948,
      "num_input_tokens_seen": 73881525,
      "step": 3422,
      "time_per_iteration": 2.4635732173919678
    },
    {
      "auxiliary_loss_clip": 0.01054264,
      "auxiliary_loss_mlp": 0.01023029,
      "balance_loss_clip": 1.02866721,
      "balance_loss_mlp": 1.0205611,
      "epoch": 0.20580189388245904,
      "flos": 70312518708480.0,
      "grad_norm": 0.7476236925960988,
      "language_loss": 0.55444312,
      "learning_rate": 3.684456776779548e-06,
      "loss": 0.57521605,
      "num_input_tokens_seen": 73937775,
      "step": 3423,
      "time_per_iteration": 3.1671698093414307
    },
    {
      "auxiliary_loss_clip": 0.01096711,
      "auxiliary_loss_mlp": 0.01039001,
      "balance_loss_clip": 1.04828513,
      "balance_loss_mlp": 1.02071428,
      "epoch": 0.205862017135127,
      "flos": 30737846252160.0,
      "grad_norm": 1.8761180469436132,
      "language_loss": 0.72157598,
      "learning_rate": 3.684246777912353e-06,
      "loss": 0.74293303,
      "num_input_tokens_seen": 73958250,
      "step": 3424,
      "time_per_iteration": 2.6815574169158936
    },
    {
      "auxiliary_loss_clip": 0.01125484,
      "auxiliary_loss_mlp": 0.00804547,
      "balance_loss_clip": 1.06058967,
      "balance_loss_mlp": 1.02473688,
      "epoch": 0.20592214038779497,
      "flos": 21324546673920.0,
      "grad_norm": 1.8806108843194673,
      "language_loss": 0.75443572,
      "learning_rate": 3.684036715178351e-06,
      "loss": 0.77373606,
      "num_input_tokens_seen": 73977775,
      "step": 3425,
      "time_per_iteration": 2.570366621017456
    },
    {
      "auxiliary_loss_clip": 0.01096457,
      "auxiliary_loss_mlp": 0.01055216,
      "balance_loss_clip": 1.05013728,
      "balance_loss_mlp": 1.03726315,
      "epoch": 0.20598226364046296,
      "flos": 22891652213760.0,
      "grad_norm": 1.9586510448334638,
      "language_loss": 0.88175142,
      "learning_rate": 3.683826588585508e-06,
      "loss": 0.90326822,
      "num_input_tokens_seen": 73996590,
      "step": 3426,
      "time_per_iteration": 2.6074695587158203
    },
    {
      "auxiliary_loss_clip": 0.01136003,
      "auxiliary_loss_mlp": 0.01046014,
      "balance_loss_clip": 1.05319548,
      "balance_loss_mlp": 1.02827549,
      "epoch": 0.20604238689313092,
      "flos": 23878549365120.0,
      "grad_norm": 1.4894599063659826,
      "language_loss": 0.76708519,
      "learning_rate": 3.6836163981417926e-06,
      "loss": 0.78890538,
      "num_input_tokens_seen": 74015935,
      "step": 3427,
      "time_per_iteration": 2.541537284851074
    },
    {
      "auxiliary_loss_clip": 0.01150178,
      "auxiliary_loss_mlp": 0.01045321,
      "balance_loss_clip": 1.05271578,
      "balance_loss_mlp": 1.02662897,
      "epoch": 0.2061025101457989,
      "flos": 22491535639680.0,
      "grad_norm": 1.5560277648909873,
      "language_loss": 0.73727775,
      "learning_rate": 3.683406143855174e-06,
      "loss": 0.7592327,
      "num_input_tokens_seen": 74036575,
      "step": 3428,
      "time_per_iteration": 2.5128917694091797
    },
    {
      "auxiliary_loss_clip": 0.01123697,
      "auxiliary_loss_mlp": 0.01046349,
      "balance_loss_clip": 1.04781461,
      "balance_loss_mlp": 1.02729905,
      "epoch": 0.20616263339846685,
      "flos": 22778928357120.0,
      "grad_norm": 2.309891846703111,
      "language_loss": 0.73496026,
      "learning_rate": 3.6831958257336256e-06,
      "loss": 0.7566607,
      "num_input_tokens_seen": 74055365,
      "step": 3429,
      "time_per_iteration": 2.5427491664886475
    },
    {
      "auxiliary_loss_clip": 0.01133198,
      "auxiliary_loss_mlp": 0.01052995,
      "balance_loss_clip": 1.0532074,
      "balance_loss_mlp": 1.03361142,
      "epoch": 0.20622275665113482,
      "flos": 20882198684160.0,
      "grad_norm": 1.9319490438952744,
      "language_loss": 0.85432172,
      "learning_rate": 3.6829854437851237e-06,
      "loss": 0.87618363,
      "num_input_tokens_seen": 74074875,
      "step": 3430,
      "time_per_iteration": 3.859550952911377
    },
    {
      "auxiliary_loss_clip": 0.01077918,
      "auxiliary_loss_mlp": 0.01054753,
      "balance_loss_clip": 1.04829836,
      "balance_loss_mlp": 1.03625226,
      "epoch": 0.20628287990380278,
      "flos": 19354415558400.0,
      "grad_norm": 1.669157853186351,
      "language_loss": 0.68975449,
      "learning_rate": 3.6827749980176444e-06,
      "loss": 0.71108127,
      "num_input_tokens_seen": 74094505,
      "step": 3431,
      "time_per_iteration": 4.034096002578735
    },
    {
      "auxiliary_loss_clip": 0.0102156,
      "auxiliary_loss_mlp": 0.01015087,
      "balance_loss_clip": 1.03331041,
      "balance_loss_mlp": 1.01258373,
      "epoch": 0.20634300315647078,
      "flos": 71517932248320.0,
      "grad_norm": 0.814326978795341,
      "language_loss": 0.60230935,
      "learning_rate": 3.6825644884391693e-06,
      "loss": 0.62267584,
      "num_input_tokens_seen": 74158500,
      "step": 3432,
      "time_per_iteration": 3.3279008865356445
    },
    {
      "auxiliary_loss_clip": 0.01135479,
      "auxiliary_loss_mlp": 0.01041714,
      "balance_loss_clip": 1.05207968,
      "balance_loss_mlp": 1.02470255,
      "epoch": 0.20640312640913874,
      "flos": 21723944976000.0,
      "grad_norm": 2.3646516822946135,
      "language_loss": 0.72169495,
      "learning_rate": 3.682353915057679e-06,
      "loss": 0.74346685,
      "num_input_tokens_seen": 74176685,
      "step": 3433,
      "time_per_iteration": 2.5453410148620605
    },
    {
      "auxiliary_loss_clip": 0.01089425,
      "auxiliary_loss_mlp": 0.01049824,
      "balance_loss_clip": 1.04712129,
      "balance_loss_mlp": 1.03023744,
      "epoch": 0.2064632496618067,
      "flos": 20554621626240.0,
      "grad_norm": 1.8661193936042433,
      "language_loss": 0.86720085,
      "learning_rate": 3.6821432778811604e-06,
      "loss": 0.88859338,
      "num_input_tokens_seen": 74194935,
      "step": 3434,
      "time_per_iteration": 2.6232481002807617
    },
    {
      "auxiliary_loss_clip": 0.01140222,
      "auxiliary_loss_mlp": 0.01038424,
      "balance_loss_clip": 1.05029297,
      "balance_loss_mlp": 1.02134156,
      "epoch": 0.20652337291447467,
      "flos": 29823273135360.0,
      "grad_norm": 2.0777164564453456,
      "language_loss": 0.69520593,
      "learning_rate": 3.6819325769176004e-06,
      "loss": 0.71699238,
      "num_input_tokens_seen": 74215400,
      "step": 3435,
      "time_per_iteration": 4.008605718612671
    },
    {
      "auxiliary_loss_clip": 0.01113363,
      "auxiliary_loss_mlp": 0.01040938,
      "balance_loss_clip": 1.04786062,
      "balance_loss_mlp": 1.02230573,
      "epoch": 0.20658349616714264,
      "flos": 26213640618240.0,
      "grad_norm": 1.7668872442427106,
      "language_loss": 0.89547193,
      "learning_rate": 3.681721812174988e-06,
      "loss": 0.91701502,
      "num_input_tokens_seen": 74234090,
      "step": 3436,
      "time_per_iteration": 3.9448013305664062
    },
    {
      "auxiliary_loss_clip": 0.01110239,
      "auxiliary_loss_mlp": 0.01044392,
      "balance_loss_clip": 1.04932547,
      "balance_loss_mlp": 1.02453148,
      "epoch": 0.2066436194198106,
      "flos": 25994370044160.0,
      "grad_norm": 1.9541562884935573,
      "language_loss": 0.76485837,
      "learning_rate": 3.6815109836613163e-06,
      "loss": 0.78640467,
      "num_input_tokens_seen": 74253345,
      "step": 3437,
      "time_per_iteration": 2.6133923530578613
    },
    {
      "auxiliary_loss_clip": 0.01140275,
      "auxiliary_loss_mlp": 0.0103964,
      "balance_loss_clip": 1.05196643,
      "balance_loss_mlp": 1.02293873,
      "epoch": 0.20670374267247857,
      "flos": 21361067827200.0,
      "grad_norm": 2.5154564032676867,
      "language_loss": 0.7757346,
      "learning_rate": 3.6813000913845795e-06,
      "loss": 0.79753375,
      "num_input_tokens_seen": 74271615,
      "step": 3438,
      "time_per_iteration": 2.512479782104492
    },
    {
      "auxiliary_loss_clip": 0.01053206,
      "auxiliary_loss_mlp": 0.01020467,
      "balance_loss_clip": 1.02777362,
      "balance_loss_mlp": 1.01803529,
      "epoch": 0.20676386592514656,
      "flos": 66383281952640.0,
      "grad_norm": 0.8408839266320254,
      "language_loss": 0.67097104,
      "learning_rate": 3.6810891353527747e-06,
      "loss": 0.69170785,
      "num_input_tokens_seen": 74331390,
      "step": 3439,
      "time_per_iteration": 3.07629656791687
    },
    {
      "auxiliary_loss_clip": 0.01136868,
      "auxiliary_loss_mlp": 0.01037327,
      "balance_loss_clip": 1.05002785,
      "balance_loss_mlp": 1.02005339,
      "epoch": 0.20682398917781453,
      "flos": 17274577328640.0,
      "grad_norm": 2.3200109302949055,
      "language_loss": 0.83991206,
      "learning_rate": 3.6808781155739014e-06,
      "loss": 0.86165398,
      "num_input_tokens_seen": 74347335,
      "step": 3440,
      "time_per_iteration": 2.471365213394165
    },
    {
      "auxiliary_loss_clip": 0.01136674,
      "auxiliary_loss_mlp": 0.0104189,
      "balance_loss_clip": 1.04990864,
      "balance_loss_mlp": 1.02522492,
      "epoch": 0.2068841124304825,
      "flos": 18077288515200.0,
      "grad_norm": 2.019098512322724,
      "language_loss": 0.85298049,
      "learning_rate": 3.6806670320559614e-06,
      "loss": 0.87476611,
      "num_input_tokens_seen": 74366310,
      "step": 3441,
      "time_per_iteration": 2.5025110244750977
    },
    {
      "auxiliary_loss_clip": 0.01103761,
      "auxiliary_loss_mlp": 0.01043418,
      "balance_loss_clip": 1.05373669,
      "balance_loss_mlp": 1.02551305,
      "epoch": 0.20694423568315046,
      "flos": 27347017432320.0,
      "grad_norm": 1.9933889923050232,
      "language_loss": 0.85565841,
      "learning_rate": 3.680455884806959e-06,
      "loss": 0.87713015,
      "num_input_tokens_seen": 74387100,
      "step": 3442,
      "time_per_iteration": 2.6477363109588623
    },
    {
      "auxiliary_loss_clip": 0.01073285,
      "auxiliary_loss_mlp": 0.01040219,
      "balance_loss_clip": 1.05398464,
      "balance_loss_mlp": 1.0221827,
      "epoch": 0.20700435893581842,
      "flos": 20229845829120.0,
      "grad_norm": 2.3055245752461437,
      "language_loss": 0.73040771,
      "learning_rate": 3.6802446738349014e-06,
      "loss": 0.75154275,
      "num_input_tokens_seen": 74404460,
      "step": 3443,
      "time_per_iteration": 2.6486024856567383
    },
    {
      "auxiliary_loss_clip": 0.01118892,
      "auxiliary_loss_mlp": 0.00796983,
      "balance_loss_clip": 1.05034018,
      "balance_loss_mlp": 1.0132705,
      "epoch": 0.2070644821884864,
      "flos": 20631111638400.0,
      "grad_norm": 2.9927212050547807,
      "language_loss": 0.85267627,
      "learning_rate": 3.680033399147797e-06,
      "loss": 0.87183499,
      "num_input_tokens_seen": 74423790,
      "step": 3444,
      "time_per_iteration": 2.5475831031799316
    },
    {
      "auxiliary_loss_clip": 0.01024825,
      "auxiliary_loss_mlp": 0.01000165,
      "balance_loss_clip": 1.03623033,
      "balance_loss_mlp": 0.99786395,
      "epoch": 0.20712460544115438,
      "flos": 65941077617280.0,
      "grad_norm": 0.6949683184703506,
      "language_loss": 0.57120758,
      "learning_rate": 3.6798220607536585e-06,
      "loss": 0.59145755,
      "num_input_tokens_seen": 74488130,
      "step": 3445,
      "time_per_iteration": 3.1430788040161133
    },
    {
      "auxiliary_loss_clip": 0.01145526,
      "auxiliary_loss_mlp": 0.00799872,
      "balance_loss_clip": 1.05042887,
      "balance_loss_mlp": 1.0176065,
      "epoch": 0.20718472869382235,
      "flos": 19425734012160.0,
      "grad_norm": 1.6276757943805396,
      "language_loss": 0.78120536,
      "learning_rate": 3.6796106586604987e-06,
      "loss": 0.8006593,
      "num_input_tokens_seen": 74506720,
      "step": 3446,
      "time_per_iteration": 2.483214855194092
    },
    {
      "auxiliary_loss_clip": 0.0114348,
      "auxiliary_loss_mlp": 0.01044322,
      "balance_loss_clip": 1.05233717,
      "balance_loss_mlp": 1.02381754,
      "epoch": 0.2072448519464903,
      "flos": 24499049834880.0,
      "grad_norm": 3.682693485183034,
      "language_loss": 0.62167156,
      "learning_rate": 3.679399192876334e-06,
      "loss": 0.64354956,
      "num_input_tokens_seen": 74525330,
      "step": 3447,
      "time_per_iteration": 2.5261948108673096
    },
    {
      "auxiliary_loss_clip": 0.01095314,
      "auxiliary_loss_mlp": 0.01052544,
      "balance_loss_clip": 1.04403543,
      "balance_loss_mlp": 1.03258884,
      "epoch": 0.20730497519915828,
      "flos": 23075694524160.0,
      "grad_norm": 2.5452545842051646,
      "language_loss": 0.86180961,
      "learning_rate": 3.679187663409184e-06,
      "loss": 0.88328815,
      "num_input_tokens_seen": 74544535,
      "step": 3448,
      "time_per_iteration": 2.600186586380005
    },
    {
      "auxiliary_loss_clip": 0.01120032,
      "auxiliary_loss_mlp": 0.01041015,
      "balance_loss_clip": 1.04606879,
      "balance_loss_mlp": 1.02104735,
      "epoch": 0.20736509845182624,
      "flos": 21069042255360.0,
      "grad_norm": 3.1937407915631537,
      "language_loss": 0.75697815,
      "learning_rate": 3.6789760702670696e-06,
      "loss": 0.77858865,
      "num_input_tokens_seen": 74562300,
      "step": 3449,
      "time_per_iteration": 2.539281129837036
    },
    {
      "auxiliary_loss_clip": 0.0113192,
      "auxiliary_loss_mlp": 0.01050816,
      "balance_loss_clip": 1.04943419,
      "balance_loss_mlp": 1.03146803,
      "epoch": 0.2074252217044942,
      "flos": 17633288499840.0,
      "grad_norm": 1.8821350292083072,
      "language_loss": 0.76393133,
      "learning_rate": 3.6787644134580134e-06,
      "loss": 0.78575867,
      "num_input_tokens_seen": 74580080,
      "step": 3450,
      "time_per_iteration": 2.516465425491333
    },
    {
      "auxiliary_loss_clip": 0.01116767,
      "auxiliary_loss_mlp": 0.01044907,
      "balance_loss_clip": 1.04848289,
      "balance_loss_mlp": 1.02700186,
      "epoch": 0.20748534495716217,
      "flos": 23546985897600.0,
      "grad_norm": 2.526487579965727,
      "language_loss": 0.8224951,
      "learning_rate": 3.6785526929900436e-06,
      "loss": 0.8441118,
      "num_input_tokens_seen": 74598980,
      "step": 3451,
      "time_per_iteration": 2.590925455093384
    },
    {
      "auxiliary_loss_clip": 0.01059186,
      "auxiliary_loss_mlp": 0.01003685,
      "balance_loss_clip": 1.02382946,
      "balance_loss_mlp": 1.00122893,
      "epoch": 0.20754546820983016,
      "flos": 52252935598080.0,
      "grad_norm": 0.7851238299338253,
      "language_loss": 0.56603658,
      "learning_rate": 3.6783409088711875e-06,
      "loss": 0.58666527,
      "num_input_tokens_seen": 74655275,
      "step": 3452,
      "time_per_iteration": 3.0009827613830566
    },
    {
      "auxiliary_loss_clip": 0.01117395,
      "auxiliary_loss_mlp": 0.00799869,
      "balance_loss_clip": 1.05144322,
      "balance_loss_mlp": 1.01940608,
      "epoch": 0.20760559146249813,
      "flos": 20412379768320.0,
      "grad_norm": 1.919990072486053,
      "language_loss": 0.88029069,
      "learning_rate": 3.6781290611094755e-06,
      "loss": 0.8994633,
      "num_input_tokens_seen": 74674560,
      "step": 3453,
      "time_per_iteration": 2.5796854496002197
    },
    {
      "auxiliary_loss_clip": 0.01144416,
      "auxiliary_loss_mlp": 0.01044419,
      "balance_loss_clip": 1.05596519,
      "balance_loss_mlp": 1.02517867,
      "epoch": 0.2076657147151661,
      "flos": 23186012169600.0,
      "grad_norm": 3.998931972470843,
      "language_loss": 0.79963958,
      "learning_rate": 3.6779171497129407e-06,
      "loss": 0.82152796,
      "num_input_tokens_seen": 74694500,
      "step": 3454,
      "time_per_iteration": 2.556161642074585
    },
    {
      "auxiliary_loss_clip": 0.01102688,
      "auxiliary_loss_mlp": 0.00800396,
      "balance_loss_clip": 1.0480746,
      "balance_loss_mlp": 1.0173912,
      "epoch": 0.20772583796783406,
      "flos": 18293219124480.0,
      "grad_norm": 3.145513064037283,
      "language_loss": 0.76947969,
      "learning_rate": 3.6777051746896202e-06,
      "loss": 0.7885105,
      "num_input_tokens_seen": 74710485,
      "step": 3455,
      "time_per_iteration": 2.528885841369629
    },
    {
      "auxiliary_loss_clip": 0.01116421,
      "auxiliary_loss_mlp": 0.01046977,
      "balance_loss_clip": 1.05324399,
      "balance_loss_mlp": 1.02922714,
      "epoch": 0.20778596122050202,
      "flos": 17602800831360.0,
      "grad_norm": 1.8715827029451806,
      "language_loss": 0.80668479,
      "learning_rate": 3.6774931360475516e-06,
      "loss": 0.82831877,
      "num_input_tokens_seen": 74727450,
      "step": 3456,
      "time_per_iteration": 2.5601418018341064
    },
    {
      "auxiliary_loss_clip": 0.01107812,
      "auxiliary_loss_mlp": 0.00799771,
      "balance_loss_clip": 1.05171108,
      "balance_loss_mlp": 1.01674557,
      "epoch": 0.20784608447317,
      "flos": 23805578885760.0,
      "grad_norm": 1.5918944848958074,
      "language_loss": 0.7773484,
      "learning_rate": 3.6772810337947745e-06,
      "loss": 0.79642427,
      "num_input_tokens_seen": 74746725,
      "step": 3457,
      "time_per_iteration": 2.6408443450927734
    },
    {
      "auxiliary_loss_clip": 0.0107924,
      "auxiliary_loss_mlp": 0.01054229,
      "balance_loss_clip": 1.04806256,
      "balance_loss_mlp": 1.03279519,
      "epoch": 0.20790620772583795,
      "flos": 17639286071040.0,
      "grad_norm": 1.838500144260174,
      "language_loss": 0.83441734,
      "learning_rate": 3.677068867939333e-06,
      "loss": 0.85575211,
      "num_input_tokens_seen": 74765255,
      "step": 3458,
      "time_per_iteration": 2.641970157623291
    },
    {
      "auxiliary_loss_clip": 0.01134122,
      "auxiliary_loss_mlp": 0.00800162,
      "balance_loss_clip": 1.04950285,
      "balance_loss_mlp": 1.01817131,
      "epoch": 0.20796633097850595,
      "flos": 27673481168640.0,
      "grad_norm": 2.7009939442045257,
      "language_loss": 0.75923985,
      "learning_rate": 3.676856638489272e-06,
      "loss": 0.77858269,
      "num_input_tokens_seen": 74785710,
      "step": 3459,
      "time_per_iteration": 2.538094997406006
    },
    {
      "auxiliary_loss_clip": 0.01081538,
      "auxiliary_loss_mlp": 0.01034032,
      "balance_loss_clip": 1.04664731,
      "balance_loss_mlp": 1.01684225,
      "epoch": 0.2080264542311739,
      "flos": 19245606284160.0,
      "grad_norm": 2.6800614517237182,
      "language_loss": 0.77361226,
      "learning_rate": 3.6766443454526382e-06,
      "loss": 0.79476798,
      "num_input_tokens_seen": 74804490,
      "step": 3460,
      "time_per_iteration": 2.662203311920166
    },
    {
      "auxiliary_loss_clip": 0.01085865,
      "auxiliary_loss_mlp": 0.01043899,
      "balance_loss_clip": 1.04721487,
      "balance_loss_mlp": 1.02557683,
      "epoch": 0.20808657748384188,
      "flos": 27525924097920.0,
      "grad_norm": 1.9339180019431084,
      "language_loss": 0.75754774,
      "learning_rate": 3.6764319888374836e-06,
      "loss": 0.77884531,
      "num_input_tokens_seen": 74826340,
      "step": 3461,
      "time_per_iteration": 2.682154893875122
    },
    {
      "auxiliary_loss_clip": 0.01122288,
      "auxiliary_loss_mlp": 0.01041392,
      "balance_loss_clip": 1.04963934,
      "balance_loss_mlp": 1.02255654,
      "epoch": 0.20814670073650984,
      "flos": 26906931999360.0,
      "grad_norm": 1.7561725225975717,
      "language_loss": 0.88405389,
      "learning_rate": 3.6762195686518604e-06,
      "loss": 0.90569067,
      "num_input_tokens_seen": 74844960,
      "step": 3462,
      "time_per_iteration": 2.5886764526367188
    },
    {
      "auxiliary_loss_clip": 0.01020584,
      "auxiliary_loss_mlp": 0.01120873,
      "balance_loss_clip": 1.02479291,
      "balance_loss_mlp": 1.60749757,
      "epoch": 0.2082068239891778,
      "flos": 70175735717760.0,
      "grad_norm": 0.8640369271827509,
      "language_loss": 0.59079254,
      "learning_rate": 3.6760070849038226e-06,
      "loss": 0.61220706,
      "num_input_tokens_seen": 74909075,
      "step": 3463,
      "time_per_iteration": 3.3037097454071045
    },
    {
      "auxiliary_loss_clip": 0.01125607,
      "auxiliary_loss_mlp": 0.01047717,
      "balance_loss_clip": 1.04668713,
      "balance_loss_mlp": 1.02945399,
      "epoch": 0.20826694724184577,
      "flos": 24608074590720.0,
      "grad_norm": 2.5837400195383924,
      "language_loss": 0.66200614,
      "learning_rate": 3.675794537601429e-06,
      "loss": 0.68373942,
      "num_input_tokens_seen": 74928125,
      "step": 3464,
      "time_per_iteration": 2.5874416828155518
    },
    {
      "auxiliary_loss_clip": 0.01112719,
      "auxiliary_loss_mlp": 0.01041402,
      "balance_loss_clip": 1.04595304,
      "balance_loss_mlp": 1.02272153,
      "epoch": 0.20832707049451377,
      "flos": 12892829034240.0,
      "grad_norm": 2.442729049794554,
      "language_loss": 0.83733582,
      "learning_rate": 3.6755819267527373e-06,
      "loss": 0.85887706,
      "num_input_tokens_seen": 74945090,
      "step": 3465,
      "time_per_iteration": 2.5352060794830322
    },
    {
      "auxiliary_loss_clip": 0.01092991,
      "auxiliary_loss_mlp": 0.01039456,
      "balance_loss_clip": 1.04790854,
      "balance_loss_mlp": 1.02151537,
      "epoch": 0.20838719374718173,
      "flos": 22198827709440.0,
      "grad_norm": 2.2527418712445355,
      "language_loss": 0.82081485,
      "learning_rate": 3.6753692523658113e-06,
      "loss": 0.84213924,
      "num_input_tokens_seen": 74963630,
      "step": 3466,
      "time_per_iteration": 2.617039203643799
    },
    {
      "auxiliary_loss_clip": 0.01133547,
      "auxiliary_loss_mlp": 0.01038458,
      "balance_loss_clip": 1.05000508,
      "balance_loss_mlp": 1.02287698,
      "epoch": 0.2084473169998497,
      "flos": 15158648908800.0,
      "grad_norm": 1.9085569339368604,
      "language_loss": 0.81773984,
      "learning_rate": 3.675156514448716e-06,
      "loss": 0.8394599,
      "num_input_tokens_seen": 74981875,
      "step": 3467,
      "time_per_iteration": 2.4802989959716797
    },
    {
      "auxiliary_loss_clip": 0.01142471,
      "auxiliary_loss_mlp": 0.01039368,
      "balance_loss_clip": 1.05050147,
      "balance_loss_mlp": 1.02357268,
      "epoch": 0.20850744025251766,
      "flos": 17456788045440.0,
      "grad_norm": 1.9800416984159306,
      "language_loss": 0.8179723,
      "learning_rate": 3.674943713009518e-06,
      "loss": 0.8397907,
      "num_input_tokens_seen": 74999155,
      "step": 3468,
      "time_per_iteration": 3.8251712322235107
    },
    {
      "auxiliary_loss_clip": 0.01136866,
      "auxiliary_loss_mlp": 0.01047679,
      "balance_loss_clip": 1.04933393,
      "balance_loss_mlp": 1.02747309,
      "epoch": 0.20856756350518563,
      "flos": 25698968593920.0,
      "grad_norm": 2.1381579531805297,
      "language_loss": 0.90033662,
      "learning_rate": 3.6747308480562856e-06,
      "loss": 0.92218208,
      "num_input_tokens_seen": 75017850,
      "step": 3469,
      "time_per_iteration": 2.5651025772094727
    },
    {
      "auxiliary_loss_clip": 0.01112522,
      "auxiliary_loss_mlp": 0.0104036,
      "balance_loss_clip": 1.05138671,
      "balance_loss_mlp": 1.02283657,
      "epoch": 0.2086276867578536,
      "flos": 37889060970240.0,
      "grad_norm": 2.275679910968917,
      "language_loss": 0.76328593,
      "learning_rate": 3.674517919597092e-06,
      "loss": 0.78481472,
      "num_input_tokens_seen": 75039270,
      "step": 3470,
      "time_per_iteration": 4.0824267864227295
    },
    {
      "auxiliary_loss_clip": 0.01124286,
      "auxiliary_loss_mlp": 0.01044249,
      "balance_loss_clip": 1.0514462,
      "balance_loss_mlp": 1.02652287,
      "epoch": 0.20868781001052156,
      "flos": 25557049958400.0,
      "grad_norm": 2.895830467034021,
      "language_loss": 0.75791216,
      "learning_rate": 3.674304927640011e-06,
      "loss": 0.77959746,
      "num_input_tokens_seen": 75059350,
      "step": 3471,
      "time_per_iteration": 2.594482660293579
    },
    {
      "auxiliary_loss_clip": 0.01116278,
      "auxiliary_loss_mlp": 0.01048773,
      "balance_loss_clip": 1.04725099,
      "balance_loss_mlp": 1.02955639,
      "epoch": 0.20874793326318955,
      "flos": 27529192235520.0,
      "grad_norm": 1.7431905605982951,
      "language_loss": 0.75649774,
      "learning_rate": 3.67409187219312e-06,
      "loss": 0.77814823,
      "num_input_tokens_seen": 75080150,
      "step": 3472,
      "time_per_iteration": 2.6478099822998047
    },
    {
      "auxiliary_loss_clip": 0.01140492,
      "auxiliary_loss_mlp": 0.01038848,
      "balance_loss_clip": 1.05323529,
      "balance_loss_mlp": 1.02189624,
      "epoch": 0.20880805651585752,
      "flos": 18548795370240.0,
      "grad_norm": 2.1396975986549878,
      "language_loss": 0.84804034,
      "learning_rate": 3.6738787532644966e-06,
      "loss": 0.86983377,
      "num_input_tokens_seen": 75097920,
      "step": 3473,
      "time_per_iteration": 3.877782106399536
    },
    {
      "auxiliary_loss_clip": 0.01066398,
      "auxiliary_loss_mlp": 0.01000987,
      "balance_loss_clip": 1.06408632,
      "balance_loss_mlp": 0.99860305,
      "epoch": 0.20886817976852548,
      "flos": 65946644225280.0,
      "grad_norm": 0.884051610262058,
      "language_loss": 0.63648605,
      "learning_rate": 3.6736655708622235e-06,
      "loss": 0.65715992,
      "num_input_tokens_seen": 75152410,
      "step": 3474,
      "time_per_iteration": 3.0930633544921875
    },
    {
      "auxiliary_loss_clip": 0.01127642,
      "auxiliary_loss_mlp": 0.01043503,
      "balance_loss_clip": 1.04923463,
      "balance_loss_mlp": 1.02613401,
      "epoch": 0.20892830302119345,
      "flos": 36539178929280.0,
      "grad_norm": 2.3989175930908684,
      "language_loss": 0.70152605,
      "learning_rate": 3.6734523249943844e-06,
      "loss": 0.72323751,
      "num_input_tokens_seen": 75173265,
      "step": 3475,
      "time_per_iteration": 4.052665948867798
    },
    {
      "auxiliary_loss_clip": 0.01149788,
      "auxiliary_loss_mlp": 0.01044503,
      "balance_loss_clip": 1.05270267,
      "balance_loss_mlp": 1.02620435,
      "epoch": 0.2089884262738614,
      "flos": 20956749361920.0,
      "grad_norm": 2.8976780131969218,
      "language_loss": 0.69911337,
      "learning_rate": 3.673239015669065e-06,
      "loss": 0.72105634,
      "num_input_tokens_seen": 75193640,
      "step": 3476,
      "time_per_iteration": 2.4820446968078613
    },
    {
      "auxiliary_loss_clip": 0.0112398,
      "auxiliary_loss_mlp": 0.01043256,
      "balance_loss_clip": 1.04962373,
      "balance_loss_mlp": 1.02624524,
      "epoch": 0.20904854952652938,
      "flos": 22784028088320.0,
      "grad_norm": 2.08871852711378,
      "language_loss": 0.89302534,
      "learning_rate": 3.6730256428943544e-06,
      "loss": 0.91469771,
      "num_input_tokens_seen": 75212545,
      "step": 3477,
      "time_per_iteration": 2.5700876712799072
    },
    {
      "auxiliary_loss_clip": 0.01089968,
      "auxiliary_loss_mlp": 0.01048953,
      "balance_loss_clip": 1.05047798,
      "balance_loss_mlp": 1.03116715,
      "epoch": 0.20910867277919734,
      "flos": 27303277645440.0,
      "grad_norm": 2.424945033212532,
      "language_loss": 0.68271601,
      "learning_rate": 3.672812206678344e-06,
      "loss": 0.70410526,
      "num_input_tokens_seen": 75230865,
      "step": 3478,
      "time_per_iteration": 2.630680561065674
    },
    {
      "auxiliary_loss_clip": 0.01096439,
      "auxiliary_loss_mlp": 0.01050505,
      "balance_loss_clip": 1.04820132,
      "balance_loss_mlp": 1.03079998,
      "epoch": 0.20916879603186533,
      "flos": 14319237000960.0,
      "grad_norm": 3.9872498620182726,
      "language_loss": 0.84516299,
      "learning_rate": 3.672598707029127e-06,
      "loss": 0.86663246,
      "num_input_tokens_seen": 75248285,
      "step": 3479,
      "time_per_iteration": 2.5455973148345947
    },
    {
      "auxiliary_loss_clip": 0.01111411,
      "auxiliary_loss_mlp": 0.01057407,
      "balance_loss_clip": 1.05196404,
      "balance_loss_mlp": 1.03761852,
      "epoch": 0.2092289192845333,
      "flos": 22273019251200.0,
      "grad_norm": 3.0052169094925927,
      "language_loss": 0.74061263,
      "learning_rate": 3.6723851439548003e-06,
      "loss": 0.76230085,
      "num_input_tokens_seen": 75266310,
      "step": 3480,
      "time_per_iteration": 2.5335938930511475
    },
    {
      "auxiliary_loss_clip": 0.01104067,
      "auxiliary_loss_mlp": 0.01047917,
      "balance_loss_clip": 1.04887486,
      "balance_loss_mlp": 1.03188896,
      "epoch": 0.20928904253720126,
      "flos": 14830712714880.0,
      "grad_norm": 2.1338674047506117,
      "language_loss": 0.75857693,
      "learning_rate": 3.67217151746346e-06,
      "loss": 0.78009677,
      "num_input_tokens_seen": 75284175,
      "step": 3481,
      "time_per_iteration": 2.570148229598999
    },
    {
      "auxiliary_loss_clip": 0.01088691,
      "auxiliary_loss_mlp": 0.01051203,
      "balance_loss_clip": 1.0499928,
      "balance_loss_mlp": 1.03358376,
      "epoch": 0.20934916578986923,
      "flos": 23259162216960.0,
      "grad_norm": 1.9284863691121146,
      "language_loss": 0.84975195,
      "learning_rate": 3.671957827563209e-06,
      "loss": 0.87115085,
      "num_input_tokens_seen": 75303465,
      "step": 3482,
      "time_per_iteration": 2.611280918121338
    },
    {
      "auxiliary_loss_clip": 0.01095507,
      "auxiliary_loss_mlp": 0.01047784,
      "balance_loss_clip": 1.05016088,
      "balance_loss_mlp": 1.03027201,
      "epoch": 0.2094092890425372,
      "flos": 32014398677760.0,
      "grad_norm": 2.0939955795071117,
      "language_loss": 0.70713842,
      "learning_rate": 3.6717440742621494e-06,
      "loss": 0.7285713,
      "num_input_tokens_seen": 75325290,
      "step": 3483,
      "time_per_iteration": 2.7167322635650635
    },
    {
      "auxiliary_loss_clip": 0.01121781,
      "auxiliary_loss_mlp": 0.01052635,
      "balance_loss_clip": 1.05267239,
      "balance_loss_mlp": 1.03487253,
      "epoch": 0.20946941229520516,
      "flos": 20010647082240.0,
      "grad_norm": 1.6948069290669017,
      "language_loss": 0.75003344,
      "learning_rate": 3.6715302575683865e-06,
      "loss": 0.77177751,
      "num_input_tokens_seen": 75343895,
      "step": 3484,
      "time_per_iteration": 2.531038999557495
    },
    {
      "auxiliary_loss_clip": 0.01111533,
      "auxiliary_loss_mlp": 0.01045161,
      "balance_loss_clip": 1.05212474,
      "balance_loss_mlp": 1.0281024,
      "epoch": 0.20952953554787315,
      "flos": 30740072895360.0,
      "grad_norm": 2.297391168367102,
      "language_loss": 0.70412374,
      "learning_rate": 3.6713163774900292e-06,
      "loss": 0.72569072,
      "num_input_tokens_seen": 75367100,
      "step": 3485,
      "time_per_iteration": 2.6345601081848145
    },
    {
      "auxiliary_loss_clip": 0.01084735,
      "auxiliary_loss_mlp": 0.00995497,
      "balance_loss_clip": 1.05218363,
      "balance_loss_mlp": 1.38515329,
      "epoch": 0.20958965880054112,
      "flos": 27049209770880.0,
      "grad_norm": 2.007452180720724,
      "language_loss": 0.82957196,
      "learning_rate": 3.6711024340351875e-06,
      "loss": 0.85037428,
      "num_input_tokens_seen": 75389925,
      "step": 3486,
      "time_per_iteration": 2.677536964416504
    },
    {
      "auxiliary_loss_clip": 0.01132715,
      "auxiliary_loss_mlp": 0.01050763,
      "balance_loss_clip": 1.04821658,
      "balance_loss_mlp": 1.03393042,
      "epoch": 0.20964978205320908,
      "flos": 34204123589760.0,
      "grad_norm": 1.9165150088301315,
      "language_loss": 0.86945432,
      "learning_rate": 3.6708884272119737e-06,
      "loss": 0.89128917,
      "num_input_tokens_seen": 75408575,
      "step": 3487,
      "time_per_iteration": 2.607020378112793
    },
    {
      "auxiliary_loss_clip": 0.01110931,
      "auxiliary_loss_mlp": 0.01043318,
      "balance_loss_clip": 1.05018532,
      "balance_loss_mlp": 1.02475691,
      "epoch": 0.20970990530587705,
      "flos": 23477391296640.0,
      "grad_norm": 4.584333532155842,
      "language_loss": 0.72927707,
      "learning_rate": 3.670674357028504e-06,
      "loss": 0.75081956,
      "num_input_tokens_seen": 75427155,
      "step": 3488,
      "time_per_iteration": 2.618422746658325
    },
    {
      "auxiliary_loss_clip": 0.01115491,
      "auxiliary_loss_mlp": 0.01037127,
      "balance_loss_clip": 1.05096412,
      "balance_loss_mlp": 1.02142715,
      "epoch": 0.209770028558545,
      "flos": 18551452976640.0,
      "grad_norm": 2.627647097816227,
      "language_loss": 0.81081307,
      "learning_rate": 3.6704602234928945e-06,
      "loss": 0.83233923,
      "num_input_tokens_seen": 75444450,
      "step": 3489,
      "time_per_iteration": 2.525465726852417
    },
    {
      "auxiliary_loss_clip": 0.01148387,
      "auxiliary_loss_mlp": 0.01043874,
      "balance_loss_clip": 1.05097902,
      "balance_loss_mlp": 1.02657723,
      "epoch": 0.20983015181121298,
      "flos": 21617003208960.0,
      "grad_norm": 2.393596439976296,
      "language_loss": 0.7305547,
      "learning_rate": 3.670246026613266e-06,
      "loss": 0.75247729,
      "num_input_tokens_seen": 75462625,
      "step": 3490,
      "time_per_iteration": 2.5032074451446533
    },
    {
      "auxiliary_loss_clip": 0.01122455,
      "auxiliary_loss_mlp": 0.01051087,
      "balance_loss_clip": 1.0522908,
      "balance_loss_mlp": 1.03444529,
      "epoch": 0.20989027506388094,
      "flos": 16614718531200.0,
      "grad_norm": 1.9587252042989014,
      "language_loss": 0.70764756,
      "learning_rate": 3.6700317663977415e-06,
      "loss": 0.72938299,
      "num_input_tokens_seen": 75480640,
      "step": 3491,
      "time_per_iteration": 2.5132462978363037
    },
    {
      "auxiliary_loss_clip": 0.01135317,
      "auxiliary_loss_mlp": 0.00902845,
      "balance_loss_clip": 1.04903293,
      "balance_loss_mlp": 1.21283817,
      "epoch": 0.20995039831654894,
      "flos": 23216823060480.0,
      "grad_norm": 7.961475652932225,
      "language_loss": 0.79656792,
      "learning_rate": 3.669817442854444e-06,
      "loss": 0.81694949,
      "num_input_tokens_seen": 75494900,
      "step": 3492,
      "time_per_iteration": 2.5188467502593994
    },
    {
      "auxiliary_loss_clip": 0.01138289,
      "auxiliary_loss_mlp": 0.00909516,
      "balance_loss_clip": 1.05258656,
      "balance_loss_mlp": 1.22501016,
      "epoch": 0.2100105215692169,
      "flos": 18147493647360.0,
      "grad_norm": 2.0505421184242154,
      "language_loss": 0.86747658,
      "learning_rate": 3.669603055991502e-06,
      "loss": 0.88795471,
      "num_input_tokens_seen": 75513370,
      "step": 3493,
      "time_per_iteration": 2.505835771560669
    },
    {
      "auxiliary_loss_clip": 0.01109751,
      "auxiliary_loss_mlp": 0.01044329,
      "balance_loss_clip": 1.04797983,
      "balance_loss_mlp": 1.02771163,
      "epoch": 0.21007064482188487,
      "flos": 15961611490560.0,
      "grad_norm": 1.679712941880243,
      "language_loss": 0.68770188,
      "learning_rate": 3.6693886058170455e-06,
      "loss": 0.7092427,
      "num_input_tokens_seen": 75532480,
      "step": 3494,
      "time_per_iteration": 2.5466177463531494
    },
    {
      "auxiliary_loss_clip": 0.01138635,
      "auxiliary_loss_mlp": 0.01041421,
      "balance_loss_clip": 1.04966712,
      "balance_loss_mlp": 1.02437425,
      "epoch": 0.21013076807455283,
      "flos": 32234315696640.0,
      "grad_norm": 2.2017907207780367,
      "language_loss": 0.79248595,
      "learning_rate": 3.6691740923392053e-06,
      "loss": 0.81428647,
      "num_input_tokens_seen": 75552745,
      "step": 3495,
      "time_per_iteration": 2.6143958568573
    },
    {
      "auxiliary_loss_clip": 0.01113852,
      "auxiliary_loss_mlp": 0.01042715,
      "balance_loss_clip": 1.048033,
      "balance_loss_mlp": 1.02498865,
      "epoch": 0.2101908913272208,
      "flos": 23696625957120.0,
      "grad_norm": 1.63401207119576,
      "language_loss": 0.77094662,
      "learning_rate": 3.668959515566116e-06,
      "loss": 0.79251218,
      "num_input_tokens_seen": 75574355,
      "step": 3496,
      "time_per_iteration": 2.563825845718384
    },
    {
      "auxiliary_loss_clip": 0.0112954,
      "auxiliary_loss_mlp": 0.01046132,
      "balance_loss_clip": 1.04858029,
      "balance_loss_mlp": 1.02794075,
      "epoch": 0.21025101457988876,
      "flos": 20375786787840.0,
      "grad_norm": 2.9252521514504894,
      "language_loss": 0.82226348,
      "learning_rate": 3.668744875505915e-06,
      "loss": 0.84402025,
      "num_input_tokens_seen": 75592215,
      "step": 3497,
      "time_per_iteration": 2.5160553455352783
    },
    {
      "auxiliary_loss_clip": 0.01139699,
      "auxiliary_loss_mlp": 0.01045652,
      "balance_loss_clip": 1.05287313,
      "balance_loss_mlp": 1.02833104,
      "epoch": 0.21031113783255675,
      "flos": 25775638174080.0,
      "grad_norm": 1.8937955832744184,
      "language_loss": 0.67519683,
      "learning_rate": 3.668530172166741e-06,
      "loss": 0.69705033,
      "num_input_tokens_seen": 75610740,
      "step": 3498,
      "time_per_iteration": 2.56754207611084
    },
    {
      "auxiliary_loss_clip": 0.01123172,
      "auxiliary_loss_mlp": 0.01039989,
      "balance_loss_clip": 1.05134702,
      "balance_loss_mlp": 1.02221465,
      "epoch": 0.21037126108522472,
      "flos": 22018197191040.0,
      "grad_norm": 1.9410323864550467,
      "language_loss": 0.80622566,
      "learning_rate": 3.6683154055567352e-06,
      "loss": 0.82785726,
      "num_input_tokens_seen": 75631005,
      "step": 3499,
      "time_per_iteration": 2.5532495975494385
    },
    {
      "auxiliary_loss_clip": 0.01137393,
      "auxiliary_loss_mlp": 0.01048415,
      "balance_loss_clip": 1.05228949,
      "balance_loss_mlp": 1.03213739,
      "epoch": 0.21043138433789269,
      "flos": 25334403505920.0,
      "grad_norm": 2.616716500037735,
      "language_loss": 0.78575373,
      "learning_rate": 3.668100575684043e-06,
      "loss": 0.80761182,
      "num_input_tokens_seen": 75650655,
      "step": 3500,
      "time_per_iteration": 2.5483734607696533
    },
    {
      "auxiliary_loss_clip": 0.01125127,
      "auxiliary_loss_mlp": 0.01043496,
      "balance_loss_clip": 1.05135477,
      "balance_loss_mlp": 1.0255903,
      "epoch": 0.21049150759056065,
      "flos": 25556654908800.0,
      "grad_norm": 1.7942351813245856,
      "language_loss": 0.74132001,
      "learning_rate": 3.6678856825568094e-06,
      "loss": 0.76300621,
      "num_input_tokens_seen": 75669895,
      "step": 3501,
      "time_per_iteration": 2.5845720767974854
    },
    {
      "auxiliary_loss_clip": 0.01131982,
      "auxiliary_loss_mlp": 0.01035957,
      "balance_loss_clip": 1.0493896,
      "balance_loss_mlp": 1.01890969,
      "epoch": 0.21055163084322862,
      "flos": 24495602129280.0,
      "grad_norm": 1.6162144415791853,
      "language_loss": 0.75322044,
      "learning_rate": 3.667670726183183e-06,
      "loss": 0.77489978,
      "num_input_tokens_seen": 75689535,
      "step": 3502,
      "time_per_iteration": 2.543369770050049
    },
    {
      "auxiliary_loss_clip": 0.01095476,
      "auxiliary_loss_mlp": 0.01037258,
      "balance_loss_clip": 1.04675198,
      "balance_loss_mlp": 1.01898336,
      "epoch": 0.21061175409589658,
      "flos": 25739045193600.0,
      "grad_norm": 1.8790328207653966,
      "language_loss": 0.77159524,
      "learning_rate": 3.667455706571316e-06,
      "loss": 0.7929225,
      "num_input_tokens_seen": 75709265,
      "step": 3503,
      "time_per_iteration": 2.604196786880493
    },
    {
      "auxiliary_loss_clip": 0.01085188,
      "auxiliary_loss_mlp": 0.01043772,
      "balance_loss_clip": 1.04552209,
      "balance_loss_mlp": 1.02364945,
      "epoch": 0.21067187734856455,
      "flos": 18989168112000.0,
      "grad_norm": 3.260555619688053,
      "language_loss": 0.78570557,
      "learning_rate": 3.6672406237293617e-06,
      "loss": 0.80699515,
      "num_input_tokens_seen": 75727050,
      "step": 3504,
      "time_per_iteration": 2.6142737865448
    },
    {
      "auxiliary_loss_clip": 0.01112166,
      "auxiliary_loss_mlp": 0.01047676,
      "balance_loss_clip": 1.0468111,
      "balance_loss_mlp": 1.02925849,
      "epoch": 0.21073200060123254,
      "flos": 24681368292480.0,
      "grad_norm": 1.5690313262937263,
      "language_loss": 0.76610136,
      "learning_rate": 3.6670254776654754e-06,
      "loss": 0.78769982,
      "num_input_tokens_seen": 75747175,
      "step": 3505,
      "time_per_iteration": 2.5818111896514893
    },
    {
      "auxiliary_loss_clip": 0.01114932,
      "auxiliary_loss_mlp": 0.01045907,
      "balance_loss_clip": 1.04806197,
      "balance_loss_mlp": 1.02863383,
      "epoch": 0.2107921238539005,
      "flos": 28549342402560.0,
      "grad_norm": 1.8588529483175944,
      "language_loss": 0.64074957,
      "learning_rate": 3.6668102683878163e-06,
      "loss": 0.66235799,
      "num_input_tokens_seen": 75767690,
      "step": 3506,
      "time_per_iteration": 2.5834028720855713
    },
    {
      "auxiliary_loss_clip": 0.01134723,
      "auxiliary_loss_mlp": 0.01046698,
      "balance_loss_clip": 1.04999852,
      "balance_loss_mlp": 1.02891171,
      "epoch": 0.21085224710656847,
      "flos": 25885848078720.0,
      "grad_norm": 1.7451141039053462,
      "language_loss": 0.81971765,
      "learning_rate": 3.6665949959045443e-06,
      "loss": 0.84153187,
      "num_input_tokens_seen": 75787255,
      "step": 3507,
      "time_per_iteration": 3.9661033153533936
    },
    {
      "auxiliary_loss_clip": 0.01133643,
      "auxiliary_loss_mlp": 0.01045131,
      "balance_loss_clip": 1.04855299,
      "balance_loss_mlp": 1.02744079,
      "epoch": 0.21091237035923643,
      "flos": 14976294537600.0,
      "grad_norm": 1.692461244655297,
      "language_loss": 0.7549082,
      "learning_rate": 3.666379660223824e-06,
      "loss": 0.77669597,
      "num_input_tokens_seen": 75805890,
      "step": 3508,
      "time_per_iteration": 2.4953200817108154
    },
    {
      "auxiliary_loss_clip": 0.01147844,
      "auxiliary_loss_mlp": 0.01036781,
      "balance_loss_clip": 1.04835439,
      "balance_loss_mlp": 1.01986527,
      "epoch": 0.2109724936119044,
      "flos": 16362518163840.0,
      "grad_norm": 12.334128264821654,
      "language_loss": 0.85571086,
      "learning_rate": 3.6661642613538192e-06,
      "loss": 0.8775571,
      "num_input_tokens_seen": 75821620,
      "step": 3509,
      "time_per_iteration": 3.8399300575256348
    },
    {
      "auxiliary_loss_clip": 0.01113698,
      "auxiliary_loss_mlp": 0.01048644,
      "balance_loss_clip": 1.05233073,
      "balance_loss_mlp": 1.02966559,
      "epoch": 0.21103261686457236,
      "flos": 31502492000640.0,
      "grad_norm": 10.447453394036526,
      "language_loss": 0.68193603,
      "learning_rate": 3.6659487993026987e-06,
      "loss": 0.70355952,
      "num_input_tokens_seen": 75842490,
      "step": 3510,
      "time_per_iteration": 2.6425082683563232
    },
    {
      "auxiliary_loss_clip": 0.01147457,
      "auxiliary_loss_mlp": 0.0104395,
      "balance_loss_clip": 1.04872108,
      "balance_loss_mlp": 1.02656949,
      "epoch": 0.21109274011724033,
      "flos": 27344072517120.0,
      "grad_norm": 2.1970924158165226,
      "language_loss": 0.72436196,
      "learning_rate": 3.6657332740786327e-06,
      "loss": 0.74627602,
      "num_input_tokens_seen": 75865985,
      "step": 3511,
      "time_per_iteration": 2.5369577407836914
    },
    {
      "auxiliary_loss_clip": 0.01064589,
      "auxiliary_loss_mlp": 0.01044876,
      "balance_loss_clip": 1.04728663,
      "balance_loss_mlp": 1.0253135,
      "epoch": 0.21115286336990832,
      "flos": 17820383466240.0,
      "grad_norm": 6.465513340256305,
      "language_loss": 0.69501364,
      "learning_rate": 3.665517685689794e-06,
      "loss": 0.71610832,
      "num_input_tokens_seen": 75882745,
      "step": 3512,
      "time_per_iteration": 4.399074077606201
    },
    {
      "auxiliary_loss_clip": 0.01135645,
      "auxiliary_loss_mlp": 0.01052916,
      "balance_loss_clip": 1.04727292,
      "balance_loss_mlp": 1.03397346,
      "epoch": 0.2112129866225763,
      "flos": 27197987904000.0,
      "grad_norm": 1.80625678094807,
      "language_loss": 0.73233694,
      "learning_rate": 3.6653020341443584e-06,
      "loss": 0.75422257,
      "num_input_tokens_seen": 75904305,
      "step": 3513,
      "time_per_iteration": 4.506604433059692
    },
    {
      "auxiliary_loss_clip": 0.01119423,
      "auxiliary_loss_mlp": 0.01039118,
      "balance_loss_clip": 1.04878569,
      "balance_loss_mlp": 1.02223778,
      "epoch": 0.21127310987524425,
      "flos": 23731279603200.0,
      "grad_norm": 10.452643770281146,
      "language_loss": 0.74398744,
      "learning_rate": 3.665086319450502e-06,
      "loss": 0.76557285,
      "num_input_tokens_seen": 75923710,
      "step": 3514,
      "time_per_iteration": 2.5936524868011475
    },
    {
      "auxiliary_loss_clip": 0.01131753,
      "auxiliary_loss_mlp": 0.01044279,
      "balance_loss_clip": 1.05141842,
      "balance_loss_mlp": 1.02629066,
      "epoch": 0.21133323312791222,
      "flos": 18332505624960.0,
      "grad_norm": 1.7857522943942439,
      "language_loss": 0.76661825,
      "learning_rate": 3.6648705416164062e-06,
      "loss": 0.7883786,
      "num_input_tokens_seen": 75942625,
      "step": 3515,
      "time_per_iteration": 2.5500011444091797
    },
    {
      "auxiliary_loss_clip": 0.01124767,
      "auxiliary_loss_mlp": 0.01046555,
      "balance_loss_clip": 1.05254245,
      "balance_loss_mlp": 1.02868533,
      "epoch": 0.21139335638058018,
      "flos": 17931203902080.0,
      "grad_norm": 2.728392262374262,
      "language_loss": 0.68636107,
      "learning_rate": 3.6646547006502518e-06,
      "loss": 0.70807433,
      "num_input_tokens_seen": 75959930,
      "step": 3516,
      "time_per_iteration": 2.501235008239746
    },
    {
      "auxiliary_loss_clip": 0.01110838,
      "auxiliary_loss_mlp": 0.01051828,
      "balance_loss_clip": 1.0548209,
      "balance_loss_mlp": 1.03368449,
      "epoch": 0.21145347963324815,
      "flos": 24572092141440.0,
      "grad_norm": 2.0239280009973957,
      "language_loss": 0.85076702,
      "learning_rate": 3.664438796560225e-06,
      "loss": 0.87239373,
      "num_input_tokens_seen": 75980335,
      "step": 3517,
      "time_per_iteration": 2.6136910915374756
    },
    {
      "auxiliary_loss_clip": 0.01123624,
      "auxiliary_loss_mlp": 0.01039693,
      "balance_loss_clip": 1.04714513,
      "balance_loss_mlp": 1.02191842,
      "epoch": 0.21151360288591614,
      "flos": 35845959375360.0,
      "grad_norm": 2.3317184645642417,
      "language_loss": 0.6218003,
      "learning_rate": 3.664222829354512e-06,
      "loss": 0.64343351,
      "num_input_tokens_seen": 76002095,
      "step": 3518,
      "time_per_iteration": 2.6331639289855957
    },
    {
      "auxiliary_loss_clip": 0.01083533,
      "auxiliary_loss_mlp": 0.01049899,
      "balance_loss_clip": 1.05089772,
      "balance_loss_mlp": 1.03326988,
      "epoch": 0.2115737261385841,
      "flos": 24641579001600.0,
      "grad_norm": 2.022232231062709,
      "language_loss": 0.8900677,
      "learning_rate": 3.664006799041303e-06,
      "loss": 0.91140199,
      "num_input_tokens_seen": 76020425,
      "step": 3519,
      "time_per_iteration": 2.6855738162994385
    },
    {
      "auxiliary_loss_clip": 0.01131311,
      "auxiliary_loss_mlp": 0.01050373,
      "balance_loss_clip": 1.05117714,
      "balance_loss_mlp": 1.0321219,
      "epoch": 0.21163384939125207,
      "flos": 25226887121280.0,
      "grad_norm": 2.012051516283834,
      "language_loss": 0.81462771,
      "learning_rate": 3.6637907056287886e-06,
      "loss": 0.8364445,
      "num_input_tokens_seen": 76041210,
      "step": 3520,
      "time_per_iteration": 2.5636672973632812
    },
    {
      "auxiliary_loss_clip": 0.01120394,
      "auxiliary_loss_mlp": 0.01047989,
      "balance_loss_clip": 1.04906666,
      "balance_loss_mlp": 1.03066802,
      "epoch": 0.21169397264392004,
      "flos": 26067520091520.0,
      "grad_norm": 1.7195554210974848,
      "language_loss": 0.75815171,
      "learning_rate": 3.6635745491251642e-06,
      "loss": 0.77983558,
      "num_input_tokens_seen": 76062685,
      "step": 3521,
      "time_per_iteration": 2.581467390060425
    },
    {
      "auxiliary_loss_clip": 0.01097727,
      "auxiliary_loss_mlp": 0.0104762,
      "balance_loss_clip": 1.05080175,
      "balance_loss_mlp": 1.03075171,
      "epoch": 0.211754095896588,
      "flos": 23108265181440.0,
      "grad_norm": 1.8091106869172928,
      "language_loss": 0.75670564,
      "learning_rate": 3.663358329538626e-06,
      "loss": 0.77815902,
      "num_input_tokens_seen": 76082300,
      "step": 3522,
      "time_per_iteration": 2.6482675075531006
    },
    {
      "auxiliary_loss_clip": 0.01146532,
      "auxiliary_loss_mlp": 0.01050601,
      "balance_loss_clip": 1.04895616,
      "balance_loss_mlp": 1.03228998,
      "epoch": 0.21181421914925597,
      "flos": 27922341571200.0,
      "grad_norm": 2.4183802005068222,
      "language_loss": 0.70668262,
      "learning_rate": 3.663142046877374e-06,
      "loss": 0.72865397,
      "num_input_tokens_seen": 76101135,
      "step": 3523,
      "time_per_iteration": 2.5337729454040527
    },
    {
      "auxiliary_loss_clip": 0.01135335,
      "auxiliary_loss_mlp": 0.01047722,
      "balance_loss_clip": 1.05021763,
      "balance_loss_mlp": 1.03043699,
      "epoch": 0.21187434240192393,
      "flos": 17128636369920.0,
      "grad_norm": 2.1312922476985507,
      "language_loss": 0.76564622,
      "learning_rate": 3.6629257011496085e-06,
      "loss": 0.78747678,
      "num_input_tokens_seen": 76119320,
      "step": 3524,
      "time_per_iteration": 2.4848852157592773
    },
    {
      "auxiliary_loss_clip": 0.01122453,
      "auxiliary_loss_mlp": 0.010446,
      "balance_loss_clip": 1.04438567,
      "balance_loss_mlp": 1.02727914,
      "epoch": 0.21193446565459192,
      "flos": 22347318533760.0,
      "grad_norm": 2.210570732177641,
      "language_loss": 0.81818151,
      "learning_rate": 3.6627092923635338e-06,
      "loss": 0.83985209,
      "num_input_tokens_seen": 76137445,
      "step": 3525,
      "time_per_iteration": 2.548069477081299
    },
    {
      "auxiliary_loss_clip": 0.01087691,
      "auxiliary_loss_mlp": 0.01040095,
      "balance_loss_clip": 1.04595518,
      "balance_loss_mlp": 1.0220108,
      "epoch": 0.2119945889072599,
      "flos": 27199316707200.0,
      "grad_norm": 5.024190721970387,
      "language_loss": 0.75098658,
      "learning_rate": 3.662492820527356e-06,
      "loss": 0.77226442,
      "num_input_tokens_seen": 76159500,
      "step": 3526,
      "time_per_iteration": 2.6984596252441406
    },
    {
      "auxiliary_loss_clip": 0.01148353,
      "auxiliary_loss_mlp": 0.01039429,
      "balance_loss_clip": 1.04897857,
      "balance_loss_mlp": 1.02151132,
      "epoch": 0.21205471215992786,
      "flos": 20991869884800.0,
      "grad_norm": 2.5588782712728046,
      "language_loss": 0.76503098,
      "learning_rate": 3.662276285649284e-06,
      "loss": 0.78690875,
      "num_input_tokens_seen": 76177990,
      "step": 3527,
      "time_per_iteration": 2.4965102672576904
    },
    {
      "auxiliary_loss_clip": 0.01146794,
      "auxiliary_loss_mlp": 0.01046128,
      "balance_loss_clip": 1.04921532,
      "balance_loss_mlp": 1.02751935,
      "epoch": 0.21211483541259582,
      "flos": 20777663128320.0,
      "grad_norm": 3.708553563873243,
      "language_loss": 0.77640057,
      "learning_rate": 3.662059687737528e-06,
      "loss": 0.79832971,
      "num_input_tokens_seen": 76197125,
      "step": 3528,
      "time_per_iteration": 2.4879233837127686
    },
    {
      "auxiliary_loss_clip": 0.01134044,
      "auxiliary_loss_mlp": 0.01047466,
      "balance_loss_clip": 1.04934907,
      "balance_loss_mlp": 1.03009689,
      "epoch": 0.21217495866526379,
      "flos": 18989994124800.0,
      "grad_norm": 2.1194008089650396,
      "language_loss": 0.81657863,
      "learning_rate": 3.6618430268003024e-06,
      "loss": 0.83839375,
      "num_input_tokens_seen": 76216215,
      "step": 3529,
      "time_per_iteration": 2.5959057807922363
    },
    {
      "auxiliary_loss_clip": 0.01128169,
      "auxiliary_loss_mlp": 0.00897232,
      "balance_loss_clip": 1.04792738,
      "balance_loss_mlp": 1.19988227,
      "epoch": 0.21223508191793175,
      "flos": 20667309569280.0,
      "grad_norm": 3.2401218420683318,
      "language_loss": 0.76727259,
      "learning_rate": 3.6616263028458235e-06,
      "loss": 0.78752661,
      "num_input_tokens_seen": 76237010,
      "step": 3530,
      "time_per_iteration": 2.5688374042510986
    },
    {
      "auxiliary_loss_clip": 0.01145752,
      "auxiliary_loss_mlp": 0.01042723,
      "balance_loss_clip": 1.04892933,
      "balance_loss_mlp": 1.0253545,
      "epoch": 0.21229520517059972,
      "flos": 21616464504960.0,
      "grad_norm": 2.3607547477897066,
      "language_loss": 0.83353627,
      "learning_rate": 3.661409515882308e-06,
      "loss": 0.85542107,
      "num_input_tokens_seen": 76255965,
      "step": 3531,
      "time_per_iteration": 2.4826738834381104
    },
    {
      "auxiliary_loss_clip": 0.0111902,
      "auxiliary_loss_mlp": 0.0104302,
      "balance_loss_clip": 1.04983568,
      "balance_loss_mlp": 1.02335012,
      "epoch": 0.2123553284232677,
      "flos": 13991049411840.0,
      "grad_norm": 2.2670635914066213,
      "language_loss": 0.73539811,
      "learning_rate": 3.661192665917977e-06,
      "loss": 0.75701845,
      "num_input_tokens_seen": 76272150,
      "step": 3532,
      "time_per_iteration": 2.5092530250549316
    },
    {
      "auxiliary_loss_clip": 0.01117113,
      "auxiliary_loss_mlp": 0.01044131,
      "balance_loss_clip": 1.04752171,
      "balance_loss_mlp": 1.02518892,
      "epoch": 0.21241545167593567,
      "flos": 18296774570880.0,
      "grad_norm": 2.0012802607863214,
      "language_loss": 0.74065185,
      "learning_rate": 3.660975752961054e-06,
      "loss": 0.76226437,
      "num_input_tokens_seen": 76291425,
      "step": 3533,
      "time_per_iteration": 2.5427908897399902
    },
    {
      "auxiliary_loss_clip": 0.01140397,
      "auxiliary_loss_mlp": 0.01044,
      "balance_loss_clip": 1.04975021,
      "balance_loss_mlp": 1.02580905,
      "epoch": 0.21247557492860364,
      "flos": 34713121265280.0,
      "grad_norm": 2.166718897009126,
      "language_loss": 0.7107389,
      "learning_rate": 3.6607587770197634e-06,
      "loss": 0.73258287,
      "num_input_tokens_seen": 76313975,
      "step": 3534,
      "time_per_iteration": 2.626420259475708
    },
    {
      "auxiliary_loss_clip": 0.01127606,
      "auxiliary_loss_mlp": 0.01044416,
      "balance_loss_clip": 1.05046082,
      "balance_loss_mlp": 1.02523518,
      "epoch": 0.2125356981812716,
      "flos": 22053820504320.0,
      "grad_norm": 2.058375921061891,
      "language_loss": 0.72007889,
      "learning_rate": 3.6605417381023346e-06,
      "loss": 0.74179912,
      "num_input_tokens_seen": 76330955,
      "step": 3535,
      "time_per_iteration": 2.522433042526245
    },
    {
      "auxiliary_loss_clip": 0.01132182,
      "auxiliary_loss_mlp": 0.01050127,
      "balance_loss_clip": 1.04867887,
      "balance_loss_mlp": 1.03235292,
      "epoch": 0.21259582143393957,
      "flos": 28548336821760.0,
      "grad_norm": 2.285220402853848,
      "language_loss": 0.70483172,
      "learning_rate": 3.660324636216996e-06,
      "loss": 0.72665483,
      "num_input_tokens_seen": 76352680,
      "step": 3536,
      "time_per_iteration": 2.575897693634033
    },
    {
      "auxiliary_loss_clip": 0.01149503,
      "auxiliary_loss_mlp": 0.01044795,
      "balance_loss_clip": 1.04894304,
      "balance_loss_mlp": 1.02585316,
      "epoch": 0.21265594468660753,
      "flos": 20120892900480.0,
      "grad_norm": 1.9038349311812022,
      "language_loss": 0.87563848,
      "learning_rate": 3.660107471371981e-06,
      "loss": 0.89758146,
      "num_input_tokens_seen": 76370750,
      "step": 3537,
      "time_per_iteration": 2.467878580093384
    },
    {
      "auxiliary_loss_clip": 0.01132142,
      "auxiliary_loss_mlp": 0.00865601,
      "balance_loss_clip": 1.04708505,
      "balance_loss_mlp": 1.14377642,
      "epoch": 0.21271606793927553,
      "flos": 23076161400960.0,
      "grad_norm": 1.6774962333781593,
      "language_loss": 0.80186081,
      "learning_rate": 3.659890243575524e-06,
      "loss": 0.82183826,
      "num_input_tokens_seen": 76390610,
      "step": 3538,
      "time_per_iteration": 2.556644916534424
    },
    {
      "auxiliary_loss_clip": 0.01081363,
      "auxiliary_loss_mlp": 0.01052669,
      "balance_loss_clip": 1.04765666,
      "balance_loss_mlp": 1.03268981,
      "epoch": 0.2127761911919435,
      "flos": 26388201738240.0,
      "grad_norm": 2.196758473514269,
      "language_loss": 0.86939758,
      "learning_rate": 3.659672952835863e-06,
      "loss": 0.89073795,
      "num_input_tokens_seen": 76408860,
      "step": 3539,
      "time_per_iteration": 2.684224843978882
    },
    {
      "auxiliary_loss_clip": 0.01118533,
      "auxiliary_loss_mlp": 0.01049966,
      "balance_loss_clip": 1.04890287,
      "balance_loss_mlp": 1.03190613,
      "epoch": 0.21283631444461146,
      "flos": 20228265630720.0,
      "grad_norm": 3.319083503043055,
      "language_loss": 0.58322334,
      "learning_rate": 3.659455599161237e-06,
      "loss": 0.60490835,
      "num_input_tokens_seen": 76424980,
      "step": 3540,
      "time_per_iteration": 2.5448131561279297
    },
    {
      "auxiliary_loss_clip": 0.01148271,
      "auxiliary_loss_mlp": 0.0103944,
      "balance_loss_clip": 1.05048776,
      "balance_loss_mlp": 1.02158225,
      "epoch": 0.21289643769727942,
      "flos": 13516992691200.0,
      "grad_norm": 2.1940855215041424,
      "language_loss": 0.7543779,
      "learning_rate": 3.659238182559888e-06,
      "loss": 0.77625507,
      "num_input_tokens_seen": 76443135,
      "step": 3541,
      "time_per_iteration": 2.465139389038086
    },
    {
      "auxiliary_loss_clip": 0.01103574,
      "auxiliary_loss_mlp": 0.01040576,
      "balance_loss_clip": 1.048949,
      "balance_loss_mlp": 1.02308798,
      "epoch": 0.2129565609499474,
      "flos": 24827021942400.0,
      "grad_norm": 2.1784990222271023,
      "language_loss": 0.69723195,
      "learning_rate": 3.6590207030400615e-06,
      "loss": 0.71867347,
      "num_input_tokens_seen": 76462470,
      "step": 3542,
      "time_per_iteration": 2.687246084213257
    },
    {
      "auxiliary_loss_clip": 0.01144672,
      "auxiliary_loss_mlp": 0.01038039,
      "balance_loss_clip": 1.05009389,
      "balance_loss_mlp": 1.02102828,
      "epoch": 0.21301668420261535,
      "flos": 23659242877440.0,
      "grad_norm": 2.5608506730969682,
      "language_loss": 0.75720978,
      "learning_rate": 3.658803160610004e-06,
      "loss": 0.77903694,
      "num_input_tokens_seen": 76481995,
      "step": 3543,
      "time_per_iteration": 2.5482211112976074
    },
    {
      "auxiliary_loss_clip": 0.01127329,
      "auxiliary_loss_mlp": 0.01041429,
      "balance_loss_clip": 1.05554926,
      "balance_loss_mlp": 1.02359533,
      "epoch": 0.21307680745528332,
      "flos": 16362805472640.0,
      "grad_norm": 1.7051623241506713,
      "language_loss": 0.66536796,
      "learning_rate": 3.6585855552779634e-06,
      "loss": 0.68705559,
      "num_input_tokens_seen": 76500245,
      "step": 3544,
      "time_per_iteration": 2.618896722793579
    },
    {
      "auxiliary_loss_clip": 0.01115115,
      "auxiliary_loss_mlp": 0.01044765,
      "balance_loss_clip": 1.04891706,
      "balance_loss_mlp": 1.02722967,
      "epoch": 0.2131369307079513,
      "flos": 19099054794240.0,
      "grad_norm": 2.9855450697737744,
      "language_loss": 0.7116369,
      "learning_rate": 3.6583678870521934e-06,
      "loss": 0.73323572,
      "num_input_tokens_seen": 76519535,
      "step": 3545,
      "time_per_iteration": 2.5950546264648438
    },
    {
      "auxiliary_loss_clip": 0.01124133,
      "auxiliary_loss_mlp": 0.01045233,
      "balance_loss_clip": 1.05325949,
      "balance_loss_mlp": 1.02772141,
      "epoch": 0.21319705396061928,
      "flos": 30372275583360.0,
      "grad_norm": 2.0413454537432214,
      "language_loss": 0.72064364,
      "learning_rate": 3.658150155940946e-06,
      "loss": 0.74233735,
      "num_input_tokens_seen": 76542065,
      "step": 3546,
      "time_per_iteration": 4.089262008666992
    },
    {
      "auxiliary_loss_clip": 0.01109243,
      "auxiliary_loss_mlp": 0.01043388,
      "balance_loss_clip": 1.05168939,
      "balance_loss_mlp": 1.02544689,
      "epoch": 0.21325717721328724,
      "flos": 21756192410880.0,
      "grad_norm": 2.153903416027601,
      "language_loss": 0.80337214,
      "learning_rate": 3.657932361952479e-06,
      "loss": 0.82489848,
      "num_input_tokens_seen": 76560540,
      "step": 3547,
      "time_per_iteration": 2.657205820083618
    },
    {
      "auxiliary_loss_clip": 0.01152404,
      "auxiliary_loss_mlp": 0.01046033,
      "balance_loss_clip": 1.05268919,
      "balance_loss_mlp": 1.0277698,
      "epoch": 0.2133173004659552,
      "flos": 28730870760960.0,
      "grad_norm": 2.8293262658533975,
      "language_loss": 0.75012743,
      "learning_rate": 3.6577145050950504e-06,
      "loss": 0.77211177,
      "num_input_tokens_seen": 76581760,
      "step": 3548,
      "time_per_iteration": 4.0355565547943115
    },
    {
      "auxiliary_loss_clip": 0.01109683,
      "auxiliary_loss_mlp": 0.01047839,
      "balance_loss_clip": 1.05108166,
      "balance_loss_mlp": 1.02846718,
      "epoch": 0.21337742371862317,
      "flos": 16837077674880.0,
      "grad_norm": 2.042064977488972,
      "language_loss": 0.73910099,
      "learning_rate": 3.657496585376922e-06,
      "loss": 0.76067615,
      "num_input_tokens_seen": 76599940,
      "step": 3549,
      "time_per_iteration": 2.6543850898742676
    },
    {
      "auxiliary_loss_clip": 0.01115846,
      "auxiliary_loss_mlp": 0.01046312,
      "balance_loss_clip": 1.05375147,
      "balance_loss_mlp": 1.02874088,
      "epoch": 0.21343754697129114,
      "flos": 24424930120320.0,
      "grad_norm": 1.6798699983268182,
      "language_loss": 0.80563468,
      "learning_rate": 3.657278602806357e-06,
      "loss": 0.82725626,
      "num_input_tokens_seen": 76619580,
      "step": 3550,
      "time_per_iteration": 2.6947054862976074
    },
    {
      "auxiliary_loss_clip": 0.01147647,
      "auxiliary_loss_mlp": 0.01045743,
      "balance_loss_clip": 1.05299544,
      "balance_loss_mlp": 1.02837431,
      "epoch": 0.21349767022395913,
      "flos": 19277817805440.0,
      "grad_norm": 1.6597811502683146,
      "language_loss": 0.8793155,
      "learning_rate": 3.657060557391621e-06,
      "loss": 0.90124941,
      "num_input_tokens_seen": 76638195,
      "step": 3551,
      "time_per_iteration": 5.267691135406494
    },
    {
      "auxiliary_loss_clip": 0.01146252,
      "auxiliary_loss_mlp": 0.01047582,
      "balance_loss_clip": 1.04950821,
      "balance_loss_mlp": 1.02990377,
      "epoch": 0.2135577934766271,
      "flos": 17347547808000.0,
      "grad_norm": 2.205509168664224,
      "language_loss": 0.8337431,
      "learning_rate": 3.656842449140983e-06,
      "loss": 0.85568142,
      "num_input_tokens_seen": 76656695,
      "step": 3552,
      "time_per_iteration": 2.5200064182281494
    },
    {
      "auxiliary_loss_clip": 0.01129664,
      "auxiliary_loss_mlp": 0.01049758,
      "balance_loss_clip": 1.04717696,
      "balance_loss_mlp": 1.03154278,
      "epoch": 0.21361791672929506,
      "flos": 24057204635520.0,
      "grad_norm": 1.7875544228582751,
      "language_loss": 0.76520455,
      "learning_rate": 3.656624278062713e-06,
      "loss": 0.78699875,
      "num_input_tokens_seen": 76677430,
      "step": 3553,
      "time_per_iteration": 2.5613672733306885
    },
    {
      "auxiliary_loss_clip": 0.01137171,
      "auxiliary_loss_mlp": 0.01042219,
      "balance_loss_clip": 1.05327797,
      "balance_loss_mlp": 1.02586341,
      "epoch": 0.21367803998196302,
      "flos": 22162306556160.0,
      "grad_norm": 1.9114710269566297,
      "language_loss": 0.72625482,
      "learning_rate": 3.6564060441650843e-06,
      "loss": 0.74804872,
      "num_input_tokens_seen": 76697615,
      "step": 3554,
      "time_per_iteration": 2.523019313812256
    },
    {
      "auxiliary_loss_clip": 0.0109325,
      "auxiliary_loss_mlp": 0.00818965,
      "balance_loss_clip": 1.04906988,
      "balance_loss_mlp": 1.05583858,
      "epoch": 0.213738163234631,
      "flos": 20886867452160.0,
      "grad_norm": 2.1388018540589306,
      "language_loss": 0.67226726,
      "learning_rate": 3.6561877474563724e-06,
      "loss": 0.69138944,
      "num_input_tokens_seen": 76715685,
      "step": 3555,
      "time_per_iteration": 2.5881736278533936
    },
    {
      "auxiliary_loss_clip": 0.01108682,
      "auxiliary_loss_mlp": 0.01037701,
      "balance_loss_clip": 1.05199397,
      "balance_loss_mlp": 1.0193553,
      "epoch": 0.21379828648729896,
      "flos": 28403114135040.0,
      "grad_norm": 1.9306632932422454,
      "language_loss": 0.64941859,
      "learning_rate": 3.6559693879448553e-06,
      "loss": 0.67088246,
      "num_input_tokens_seen": 76735405,
      "step": 3556,
      "time_per_iteration": 2.603736639022827
    },
    {
      "auxiliary_loss_clip": 0.01135656,
      "auxiliary_loss_mlp": 0.01044587,
      "balance_loss_clip": 1.05126572,
      "balance_loss_mlp": 1.02649093,
      "epoch": 0.21385840973996692,
      "flos": 25479662106240.0,
      "grad_norm": 1.7214054380470214,
      "language_loss": 0.72507924,
      "learning_rate": 3.6557509656388125e-06,
      "loss": 0.74688166,
      "num_input_tokens_seen": 76754395,
      "step": 3557,
      "time_per_iteration": 2.539219856262207
    },
    {
      "auxiliary_loss_clip": 0.01131707,
      "auxiliary_loss_mlp": 0.00811328,
      "balance_loss_clip": 1.048563,
      "balance_loss_mlp": 1.04187727,
      "epoch": 0.2139185329926349,
      "flos": 28074280101120.0,
      "grad_norm": 2.0584989530610804,
      "language_loss": 0.67162126,
      "learning_rate": 3.655532480546528e-06,
      "loss": 0.6910516,
      "num_input_tokens_seen": 76777210,
      "step": 3558,
      "time_per_iteration": 2.595909833908081
    },
    {
      "auxiliary_loss_clip": 0.01149393,
      "auxiliary_loss_mlp": 0.01041088,
      "balance_loss_clip": 1.0481329,
      "balance_loss_mlp": 1.02294409,
      "epoch": 0.21397865624530288,
      "flos": 19608698914560.0,
      "grad_norm": 1.8805590958814418,
      "language_loss": 0.79990041,
      "learning_rate": 3.655313932676286e-06,
      "loss": 0.82180524,
      "num_input_tokens_seen": 76795830,
      "step": 3559,
      "time_per_iteration": 2.448824882507324
    },
    {
      "auxiliary_loss_clip": 0.01145098,
      "auxiliary_loss_mlp": 0.01046943,
      "balance_loss_clip": 1.04833996,
      "balance_loss_mlp": 1.03015876,
      "epoch": 0.21403877949797084,
      "flos": 24681476033280.0,
      "grad_norm": 1.6901631277413696,
      "language_loss": 0.67429304,
      "learning_rate": 3.655095322036373e-06,
      "loss": 0.69621342,
      "num_input_tokens_seen": 76814700,
      "step": 3560,
      "time_per_iteration": 2.505565643310547
    },
    {
      "auxiliary_loss_clip": 0.01137738,
      "auxiliary_loss_mlp": 0.01046734,
      "balance_loss_clip": 1.05141854,
      "balance_loss_mlp": 1.02884078,
      "epoch": 0.2140989027506388,
      "flos": 19861150677120.0,
      "grad_norm": 2.110417602365459,
      "language_loss": 0.73226529,
      "learning_rate": 3.65487664863508e-06,
      "loss": 0.75410998,
      "num_input_tokens_seen": 76833400,
      "step": 3561,
      "time_per_iteration": 2.474303960800171
    },
    {
      "auxiliary_loss_clip": 0.01125456,
      "auxiliary_loss_mlp": 0.01047409,
      "balance_loss_clip": 1.04931509,
      "balance_loss_mlp": 1.02996838,
      "epoch": 0.21415902600330677,
      "flos": 19135324552320.0,
      "grad_norm": 3.030118217089213,
      "language_loss": 0.77171111,
      "learning_rate": 3.654657912480698e-06,
      "loss": 0.79343975,
      "num_input_tokens_seen": 76850645,
      "step": 3562,
      "time_per_iteration": 2.5068187713623047
    },
    {
      "auxiliary_loss_clip": 0.01147017,
      "auxiliary_loss_mlp": 0.01040073,
      "balance_loss_clip": 1.0510819,
      "balance_loss_mlp": 1.02322865,
      "epoch": 0.21421914925597474,
      "flos": 22272624201600.0,
      "grad_norm": 1.5796076621576305,
      "language_loss": 0.84619159,
      "learning_rate": 3.6544391135815237e-06,
      "loss": 0.8680625,
      "num_input_tokens_seen": 76870135,
      "step": 3563,
      "time_per_iteration": 2.4753429889678955
    },
    {
      "auxiliary_loss_clip": 0.01149568,
      "auxiliary_loss_mlp": 0.01037953,
      "balance_loss_clip": 1.05320954,
      "balance_loss_mlp": 1.02059627,
      "epoch": 0.2142792725086427,
      "flos": 33875109987840.0,
      "grad_norm": 1.7254249634391472,
      "language_loss": 0.76565719,
      "learning_rate": 3.6542202519458507e-06,
      "loss": 0.78753233,
      "num_input_tokens_seen": 76893905,
      "step": 3564,
      "time_per_iteration": 2.6045608520507812
    },
    {
      "auxiliary_loss_clip": 0.01122165,
      "auxiliary_loss_mlp": 0.01043474,
      "balance_loss_clip": 1.05441475,
      "balance_loss_mlp": 1.02567601,
      "epoch": 0.2143393957613107,
      "flos": 19860216923520.0,
      "grad_norm": 1.7289576392983124,
      "language_loss": 0.8846035,
      "learning_rate": 3.654001327581981e-06,
      "loss": 0.90625989,
      "num_input_tokens_seen": 76914205,
      "step": 3565,
      "time_per_iteration": 2.555635690689087
    },
    {
      "auxiliary_loss_clip": 0.01053285,
      "auxiliary_loss_mlp": 0.0102457,
      "balance_loss_clip": 1.04297137,
      "balance_loss_mlp": 1.0224123,
      "epoch": 0.21439951901397866,
      "flos": 68530093090560.0,
      "grad_norm": 0.8361525736553671,
      "language_loss": 0.52211046,
      "learning_rate": 3.653782340498215e-06,
      "loss": 0.54288906,
      "num_input_tokens_seen": 76975650,
      "step": 3566,
      "time_per_iteration": 3.108618974685669
    },
    {
      "auxiliary_loss_clip": 0.01134714,
      "auxiliary_loss_mlp": 0.01039465,
      "balance_loss_clip": 1.05268884,
      "balance_loss_mlp": 1.02297843,
      "epoch": 0.21445964226664663,
      "flos": 19682998197120.0,
      "grad_norm": 2.914095857656631,
      "language_loss": 0.67008561,
      "learning_rate": 3.6535632907028566e-06,
      "loss": 0.69182742,
      "num_input_tokens_seen": 76992615,
      "step": 3567,
      "time_per_iteration": 2.5077567100524902
    },
    {
      "auxiliary_loss_clip": 0.01116864,
      "auxiliary_loss_mlp": 0.01047414,
      "balance_loss_clip": 1.04811716,
      "balance_loss_mlp": 1.03009319,
      "epoch": 0.2145197655193146,
      "flos": 31107259676160.0,
      "grad_norm": 1.6907695347484843,
      "language_loss": 0.74213696,
      "learning_rate": 3.6533441782042126e-06,
      "loss": 0.76377976,
      "num_input_tokens_seen": 77017005,
      "step": 3568,
      "time_per_iteration": 2.6291964054107666
    },
    {
      "auxiliary_loss_clip": 0.01133623,
      "auxiliary_loss_mlp": 0.01044805,
      "balance_loss_clip": 1.05056894,
      "balance_loss_mlp": 1.02701938,
      "epoch": 0.21457988877198256,
      "flos": 20120785159680.0,
      "grad_norm": 2.2628178475000893,
      "language_loss": 0.77461112,
      "learning_rate": 3.6531250030105917e-06,
      "loss": 0.79639536,
      "num_input_tokens_seen": 77034990,
      "step": 3569,
      "time_per_iteration": 2.497225522994995
    },
    {
      "auxiliary_loss_clip": 0.01142836,
      "auxiliary_loss_mlp": 0.0104123,
      "balance_loss_clip": 1.05266833,
      "balance_loss_mlp": 1.02233577,
      "epoch": 0.21464001202465052,
      "flos": 18588045957120.0,
      "grad_norm": 3.0238814510944194,
      "language_loss": 0.69888031,
      "learning_rate": 3.6529057651303053e-06,
      "loss": 0.72072095,
      "num_input_tokens_seen": 77052610,
      "step": 3570,
      "time_per_iteration": 2.48427677154541
    },
    {
      "auxiliary_loss_clip": 0.0115146,
      "auxiliary_loss_mlp": 0.01041125,
      "balance_loss_clip": 1.05305922,
      "balance_loss_mlp": 1.02373242,
      "epoch": 0.21470013527731852,
      "flos": 21835160461440.0,
      "grad_norm": 3.0780937022495483,
      "language_loss": 0.78282779,
      "learning_rate": 3.6526864645716666e-06,
      "loss": 0.80475366,
      "num_input_tokens_seen": 77072475,
      "step": 3571,
      "time_per_iteration": 2.505051374435425
    },
    {
      "auxiliary_loss_clip": 0.0113204,
      "auxiliary_loss_mlp": 0.01047956,
      "balance_loss_clip": 1.05005336,
      "balance_loss_mlp": 1.02736831,
      "epoch": 0.21476025852998648,
      "flos": 17603195880960.0,
      "grad_norm": 2.6421838889047486,
      "language_loss": 0.83038414,
      "learning_rate": 3.652467101342991e-06,
      "loss": 0.85218406,
      "num_input_tokens_seen": 77089930,
      "step": 3572,
      "time_per_iteration": 2.4663052558898926
    },
    {
      "auxiliary_loss_clip": 0.01129119,
      "auxiliary_loss_mlp": 0.01040181,
      "balance_loss_clip": 1.05488324,
      "balance_loss_mlp": 1.02251458,
      "epoch": 0.21482038178265445,
      "flos": 24828135264000.0,
      "grad_norm": 2.4343403785774145,
      "language_loss": 0.64892638,
      "learning_rate": 3.652247675452598e-06,
      "loss": 0.67061937,
      "num_input_tokens_seen": 77108970,
      "step": 3573,
      "time_per_iteration": 2.5658020973205566
    },
    {
      "auxiliary_loss_clip": 0.01145547,
      "auxiliary_loss_mlp": 0.01039631,
      "balance_loss_clip": 1.05213535,
      "balance_loss_mlp": 1.0225004,
      "epoch": 0.2148805050353224,
      "flos": 23258228463360.0,
      "grad_norm": 2.204526760609019,
      "language_loss": 0.75294143,
      "learning_rate": 3.652028186908807e-06,
      "loss": 0.77479321,
      "num_input_tokens_seen": 77126045,
      "step": 3574,
      "time_per_iteration": 2.4684271812438965
    },
    {
      "auxiliary_loss_clip": 0.01134938,
      "auxiliary_loss_mlp": 0.01043702,
      "balance_loss_clip": 1.05164385,
      "balance_loss_mlp": 1.02501035,
      "epoch": 0.21494062828799038,
      "flos": 21321098968320.0,
      "grad_norm": 1.9355297413220018,
      "language_loss": 0.71793222,
      "learning_rate": 3.6518086357199416e-06,
      "loss": 0.73971856,
      "num_input_tokens_seen": 77144600,
      "step": 3575,
      "time_per_iteration": 2.5125670433044434
    },
    {
      "auxiliary_loss_clip": 0.01123089,
      "auxiliary_loss_mlp": 0.01039568,
      "balance_loss_clip": 1.05268288,
      "balance_loss_mlp": 1.0223304,
      "epoch": 0.21500075154065834,
      "flos": 18843334894080.0,
      "grad_norm": 1.7854097914937606,
      "language_loss": 0.68127233,
      "learning_rate": 3.6515890218943277e-06,
      "loss": 0.70289892,
      "num_input_tokens_seen": 77162965,
      "step": 3576,
      "time_per_iteration": 2.5161149501800537
    },
    {
      "auxiliary_loss_clip": 0.01134813,
      "auxiliary_loss_mlp": 0.01049038,
      "balance_loss_clip": 1.04968536,
      "balance_loss_mlp": 1.02850974,
      "epoch": 0.2150608747933263,
      "flos": 18441997257600.0,
      "grad_norm": 2.0608078147134896,
      "language_loss": 0.88615036,
      "learning_rate": 3.651369345440292e-06,
      "loss": 0.90798879,
      "num_input_tokens_seen": 77179960,
      "step": 3577,
      "time_per_iteration": 2.509291172027588
    },
    {
      "auxiliary_loss_clip": 0.01046768,
      "auxiliary_loss_mlp": 0.01007262,
      "balance_loss_clip": 1.02912784,
      "balance_loss_mlp": 1.00444913,
      "epoch": 0.2151209980459943,
      "flos": 66598242894720.0,
      "grad_norm": 0.8100274562646331,
      "language_loss": 0.56144679,
      "learning_rate": 3.6511496063661654e-06,
      "loss": 0.58198714,
      "num_input_tokens_seen": 77239500,
      "step": 3578,
      "time_per_iteration": 3.0738818645477295
    },
    {
      "auxiliary_loss_clip": 0.01136932,
      "auxiliary_loss_mlp": 0.00808638,
      "balance_loss_clip": 1.05210137,
      "balance_loss_mlp": 1.03821027,
      "epoch": 0.21518112129866226,
      "flos": 21575885114880.0,
      "grad_norm": 1.909030279065131,
      "language_loss": 0.88764989,
      "learning_rate": 3.6509298046802807e-06,
      "loss": 0.90710562,
      "num_input_tokens_seen": 77254680,
      "step": 3579,
      "time_per_iteration": 2.5111289024353027
    },
    {
      "auxiliary_loss_clip": 0.01137204,
      "auxiliary_loss_mlp": 0.01043034,
      "balance_loss_clip": 1.04990077,
      "balance_loss_mlp": 1.02481866,
      "epoch": 0.21524124455133023,
      "flos": 20047635112320.0,
      "grad_norm": 1.7060487932929371,
      "language_loss": 0.77799904,
      "learning_rate": 3.650709940390972e-06,
      "loss": 0.79980147,
      "num_input_tokens_seen": 77274060,
      "step": 3580,
      "time_per_iteration": 2.481167793273926
    },
    {
      "auxiliary_loss_clip": 0.01138319,
      "auxiliary_loss_mlp": 0.01041671,
      "balance_loss_clip": 1.05239749,
      "balance_loss_mlp": 1.02313352,
      "epoch": 0.2153013678039982,
      "flos": 23951807153280.0,
      "grad_norm": 1.8217980800183968,
      "language_loss": 0.729801,
      "learning_rate": 3.6504900135065775e-06,
      "loss": 0.75160086,
      "num_input_tokens_seen": 77293255,
      "step": 3581,
      "time_per_iteration": 2.556138753890991
    },
    {
      "auxiliary_loss_clip": 0.01133302,
      "auxiliary_loss_mlp": 0.01045406,
      "balance_loss_clip": 1.05028224,
      "balance_loss_mlp": 1.02562904,
      "epoch": 0.21536149105666616,
      "flos": 20594841880320.0,
      "grad_norm": 2.5782708857365937,
      "language_loss": 0.71116096,
      "learning_rate": 3.6502700240354357e-06,
      "loss": 0.73294806,
      "num_input_tokens_seen": 77312390,
      "step": 3582,
      "time_per_iteration": 2.4891724586486816
    },
    {
      "auxiliary_loss_clip": 0.01145904,
      "auxiliary_loss_mlp": 0.01045244,
      "balance_loss_clip": 1.04980874,
      "balance_loss_mlp": 1.02706432,
      "epoch": 0.21542161430933413,
      "flos": 12860042895360.0,
      "grad_norm": 2.4113552932567783,
      "language_loss": 0.84687179,
      "learning_rate": 3.650049971985889e-06,
      "loss": 0.86878324,
      "num_input_tokens_seen": 77330985,
      "step": 3583,
      "time_per_iteration": 2.4601118564605713
    },
    {
      "auxiliary_loss_clip": 0.01128106,
      "auxiliary_loss_mlp": 0.01045333,
      "balance_loss_clip": 1.05209923,
      "balance_loss_mlp": 1.0281074,
      "epoch": 0.21548173756200212,
      "flos": 26103933504000.0,
      "grad_norm": 8.407530463569582,
      "language_loss": 0.8243736,
      "learning_rate": 3.6498298573662824e-06,
      "loss": 0.84610802,
      "num_input_tokens_seen": 77350770,
      "step": 3584,
      "time_per_iteration": 2.5542404651641846
    },
    {
      "auxiliary_loss_clip": 0.01115106,
      "auxiliary_loss_mlp": 0.0080351,
      "balance_loss_clip": 1.05234396,
      "balance_loss_mlp": 1.02839208,
      "epoch": 0.21554186081467008,
      "flos": 22163779013760.0,
      "grad_norm": 2.1090004363276216,
      "language_loss": 0.90104049,
      "learning_rate": 3.6496096801849625e-06,
      "loss": 0.92022657,
      "num_input_tokens_seen": 77370510,
      "step": 3585,
      "time_per_iteration": 3.924785852432251
    },
    {
      "auxiliary_loss_clip": 0.01137176,
      "auxiliary_loss_mlp": 0.01044971,
      "balance_loss_clip": 1.05354142,
      "balance_loss_mlp": 1.02735233,
      "epoch": 0.21560198406733805,
      "flos": 22966741595520.0,
      "grad_norm": 1.6972166177471275,
      "language_loss": 0.74319875,
      "learning_rate": 3.649389440450277e-06,
      "loss": 0.76502025,
      "num_input_tokens_seen": 77390645,
      "step": 3586,
      "time_per_iteration": 4.003892660140991
    },
    {
      "auxiliary_loss_clip": 0.01110896,
      "auxiliary_loss_mlp": 0.01044621,
      "balance_loss_clip": 1.04933465,
      "balance_loss_mlp": 1.0273242,
      "epoch": 0.215662107320006,
      "flos": 22784064001920.0,
      "grad_norm": 1.7853426327330457,
      "language_loss": 0.83296454,
      "learning_rate": 3.6491691381705804e-06,
      "loss": 0.85451972,
      "num_input_tokens_seen": 77409655,
      "step": 3587,
      "time_per_iteration": 2.6137540340423584
    },
    {
      "auxiliary_loss_clip": 0.01105056,
      "auxiliary_loss_mlp": 0.00800988,
      "balance_loss_clip": 1.05073643,
      "balance_loss_mlp": 1.02053881,
      "epoch": 0.21572223057267398,
      "flos": 30883859038080.0,
      "grad_norm": 1.8657007056620702,
      "language_loss": 0.76136464,
      "learning_rate": 3.648948773354224e-06,
      "loss": 0.78042507,
      "num_input_tokens_seen": 77430560,
      "step": 3588,
      "time_per_iteration": 2.688819408416748
    },
    {
      "auxiliary_loss_clip": 0.01131567,
      "auxiliary_loss_mlp": 0.01041508,
      "balance_loss_clip": 1.04824424,
      "balance_loss_mlp": 1.02288771,
      "epoch": 0.21578235382534194,
      "flos": 26910487445760.0,
      "grad_norm": 2.3306016428580763,
      "language_loss": 0.80879414,
      "learning_rate": 3.6487283460095643e-06,
      "loss": 0.83052492,
      "num_input_tokens_seen": 77455000,
      "step": 3589,
      "time_per_iteration": 2.585813522338867
    },
    {
      "auxiliary_loss_clip": 0.01149942,
      "auxiliary_loss_mlp": 0.0103952,
      "balance_loss_clip": 1.0518074,
      "balance_loss_mlp": 1.02304578,
      "epoch": 0.2158424770780099,
      "flos": 24425720219520.0,
      "grad_norm": 2.1890025284576025,
      "language_loss": 0.72366339,
      "learning_rate": 3.648507856144961e-06,
      "loss": 0.74555802,
      "num_input_tokens_seen": 77475075,
      "step": 3590,
      "time_per_iteration": 3.8883090019226074
    },
    {
      "auxiliary_loss_clip": 0.01130697,
      "auxiliary_loss_mlp": 0.01050389,
      "balance_loss_clip": 1.05158377,
      "balance_loss_mlp": 1.03131568,
      "epoch": 0.2159026003306779,
      "flos": 23949975559680.0,
      "grad_norm": 1.7728891718536726,
      "language_loss": 0.84079504,
      "learning_rate": 3.648287303768775e-06,
      "loss": 0.86260587,
      "num_input_tokens_seen": 77495945,
      "step": 3591,
      "time_per_iteration": 2.5746867656707764
    },
    {
      "auxiliary_loss_clip": 0.01119767,
      "auxiliary_loss_mlp": 0.01048465,
      "balance_loss_clip": 1.05147052,
      "balance_loss_mlp": 1.02858067,
      "epoch": 0.21596272358334587,
      "flos": 30040963511040.0,
      "grad_norm": 2.318526811264304,
      "language_loss": 0.69247723,
      "learning_rate": 3.6480666888893686e-06,
      "loss": 0.71415955,
      "num_input_tokens_seen": 77517140,
      "step": 3592,
      "time_per_iteration": 2.618065595626831
    },
    {
      "auxiliary_loss_clip": 0.01115936,
      "auxiliary_loss_mlp": 0.01051988,
      "balance_loss_clip": 1.05275309,
      "balance_loss_mlp": 1.03308094,
      "epoch": 0.21602284683601383,
      "flos": 20376217751040.0,
      "grad_norm": 2.47167497682882,
      "language_loss": 0.83627129,
      "learning_rate": 3.647846011515108e-06,
      "loss": 0.85795045,
      "num_input_tokens_seen": 77536085,
      "step": 3593,
      "time_per_iteration": 2.5784521102905273
    },
    {
      "auxiliary_loss_clip": 0.01116557,
      "auxiliary_loss_mlp": 0.01056681,
      "balance_loss_clip": 1.04956639,
      "balance_loss_mlp": 1.03806067,
      "epoch": 0.2160829700886818,
      "flos": 20777339905920.0,
      "grad_norm": 2.508969465027594,
      "language_loss": 0.75576353,
      "learning_rate": 3.6476252716543625e-06,
      "loss": 0.77749598,
      "num_input_tokens_seen": 77553675,
      "step": 3594,
      "time_per_iteration": 2.550729990005493
    },
    {
      "auxiliary_loss_clip": 0.01134937,
      "auxiliary_loss_mlp": 0.01047679,
      "balance_loss_clip": 1.05028081,
      "balance_loss_mlp": 1.02882028,
      "epoch": 0.21614309334134976,
      "flos": 22309755886080.0,
      "grad_norm": 1.533239206269284,
      "language_loss": 0.8076545,
      "learning_rate": 3.6474044693155007e-06,
      "loss": 0.82948065,
      "num_input_tokens_seen": 77573360,
      "step": 3595,
      "time_per_iteration": 2.522955894470215
    },
    {
      "auxiliary_loss_clip": 0.01113763,
      "auxiliary_loss_mlp": 0.01044816,
      "balance_loss_clip": 1.05004334,
      "balance_loss_mlp": 1.02674389,
      "epoch": 0.21620321659401773,
      "flos": 19609524927360.0,
      "grad_norm": 1.9991981450478784,
      "language_loss": 0.78598517,
      "learning_rate": 3.647183604506897e-06,
      "loss": 0.80757093,
      "num_input_tokens_seen": 77591865,
      "step": 3596,
      "time_per_iteration": 2.525251626968384
    },
    {
      "auxiliary_loss_clip": 0.01078748,
      "auxiliary_loss_mlp": 0.01048531,
      "balance_loss_clip": 1.05259264,
      "balance_loss_mlp": 1.03166294,
      "epoch": 0.2162633398466857,
      "flos": 18844555956480.0,
      "grad_norm": 1.6465626831070728,
      "language_loss": 0.83148664,
      "learning_rate": 3.6469626772369253e-06,
      "loss": 0.85275942,
      "num_input_tokens_seen": 77611600,
      "step": 3597,
      "time_per_iteration": 2.630080461502075
    },
    {
      "auxiliary_loss_clip": 0.01126917,
      "auxiliary_loss_mlp": 0.00800216,
      "balance_loss_clip": 1.04959917,
      "balance_loss_mlp": 1.02266216,
      "epoch": 0.21632346309935369,
      "flos": 18768820129920.0,
      "grad_norm": 2.7008891878634236,
      "language_loss": 0.80419767,
      "learning_rate": 3.6467416875139642e-06,
      "loss": 0.82346904,
      "num_input_tokens_seen": 77630665,
      "step": 3598,
      "time_per_iteration": 2.529641628265381
    },
    {
      "auxiliary_loss_clip": 0.01118858,
      "auxiliary_loss_mlp": 0.01059012,
      "balance_loss_clip": 1.05036426,
      "balance_loss_mlp": 1.04061818,
      "epoch": 0.21638358635202165,
      "flos": 26324173745280.0,
      "grad_norm": 1.9085715166264319,
      "language_loss": 0.81976426,
      "learning_rate": 3.6465206353463934e-06,
      "loss": 0.8415429,
      "num_input_tokens_seen": 77650835,
      "step": 3599,
      "time_per_iteration": 2.6024491786956787
    },
    {
      "auxiliary_loss_clip": 0.01104995,
      "auxiliary_loss_mlp": 0.008039,
      "balance_loss_clip": 1.04930496,
      "balance_loss_mlp": 1.02896309,
      "epoch": 0.21644370960468962,
      "flos": 20740854666240.0,
      "grad_norm": 2.2273973555042486,
      "language_loss": 0.76484621,
      "learning_rate": 3.6462995207425947e-06,
      "loss": 0.78393513,
      "num_input_tokens_seen": 77669000,
      "step": 3600,
      "time_per_iteration": 2.570035219192505
    },
    {
      "auxiliary_loss_clip": 0.01105758,
      "auxiliary_loss_mlp": 0.01042576,
      "balance_loss_clip": 1.04898727,
      "balance_loss_mlp": 1.02709079,
      "epoch": 0.21650383285735758,
      "flos": 23952238116480.0,
      "grad_norm": 1.9243463688617086,
      "language_loss": 0.79993558,
      "learning_rate": 3.6460783437109533e-06,
      "loss": 0.82141888,
      "num_input_tokens_seen": 77688745,
      "step": 3601,
      "time_per_iteration": 2.5895750522613525
    },
    {
      "auxiliary_loss_clip": 0.01150803,
      "auxiliary_loss_mlp": 0.01054429,
      "balance_loss_clip": 1.05374146,
      "balance_loss_mlp": 1.03690505,
      "epoch": 0.21656395611002555,
      "flos": 23696087253120.0,
      "grad_norm": 9.25625967608789,
      "language_loss": 0.83348989,
      "learning_rate": 3.6458571042598565e-06,
      "loss": 0.85554224,
      "num_input_tokens_seen": 77708445,
      "step": 3602,
      "time_per_iteration": 2.4903314113616943
    },
    {
      "auxiliary_loss_clip": 0.0114789,
      "auxiliary_loss_mlp": 0.0105512,
      "balance_loss_clip": 1.0510596,
      "balance_loss_mlp": 1.03737032,
      "epoch": 0.2166240793626935,
      "flos": 20666052593280.0,
      "grad_norm": 2.2000130581161375,
      "language_loss": 0.74751723,
      "learning_rate": 3.645635802397693e-06,
      "loss": 0.76954734,
      "num_input_tokens_seen": 77728465,
      "step": 3603,
      "time_per_iteration": 2.502960681915283
    },
    {
      "auxiliary_loss_clip": 0.01113705,
      "auxiliary_loss_mlp": 0.0105609,
      "balance_loss_clip": 1.05088377,
      "balance_loss_mlp": 1.03729033,
      "epoch": 0.2166842026153615,
      "flos": 21580410228480.0,
      "grad_norm": 2.2288464709677553,
      "language_loss": 0.73801547,
      "learning_rate": 3.645414438132855e-06,
      "loss": 0.75971341,
      "num_input_tokens_seen": 77746735,
      "step": 3604,
      "time_per_iteration": 2.543144702911377
    },
    {
      "auxiliary_loss_clip": 0.01132347,
      "auxiliary_loss_mlp": 0.01040691,
      "balance_loss_clip": 1.04970789,
      "balance_loss_mlp": 1.02344179,
      "epoch": 0.21674432586802947,
      "flos": 25629948610560.0,
      "grad_norm": 1.7348700177317176,
      "language_loss": 0.79835272,
      "learning_rate": 3.6451930114737366e-06,
      "loss": 0.82008314,
      "num_input_tokens_seen": 77768105,
      "step": 3605,
      "time_per_iteration": 2.5470573902130127
    },
    {
      "auxiliary_loss_clip": 0.01066368,
      "auxiliary_loss_mlp": 0.01012644,
      "balance_loss_clip": 1.02939415,
      "balance_loss_mlp": 1.01006889,
      "epoch": 0.21680444912069743,
      "flos": 56417783616000.0,
      "grad_norm": 0.7137675923000804,
      "language_loss": 0.58322209,
      "learning_rate": 3.6449715224287347e-06,
      "loss": 0.60401219,
      "num_input_tokens_seen": 77833750,
      "step": 3606,
      "time_per_iteration": 3.194392204284668
    },
    {
      "auxiliary_loss_clip": 0.01149821,
      "auxiliary_loss_mlp": 0.01048093,
      "balance_loss_clip": 1.05104542,
      "balance_loss_mlp": 1.02962756,
      "epoch": 0.2168645723733654,
      "flos": 23878944414720.0,
      "grad_norm": 2.6439050529593415,
      "language_loss": 0.73129213,
      "learning_rate": 3.644749971006248e-06,
      "loss": 0.75327122,
      "num_input_tokens_seen": 77853780,
      "step": 3607,
      "time_per_iteration": 2.4901204109191895
    },
    {
      "auxiliary_loss_clip": 0.01131375,
      "auxiliary_loss_mlp": 0.01045014,
      "balance_loss_clip": 1.05294013,
      "balance_loss_mlp": 1.02626204,
      "epoch": 0.21692469562603336,
      "flos": 16946174257920.0,
      "grad_norm": 2.3461093127086,
      "language_loss": 0.76791602,
      "learning_rate": 3.6445283572146765e-06,
      "loss": 0.78967988,
      "num_input_tokens_seen": 77872575,
      "step": 3608,
      "time_per_iteration": 2.5419533252716064
    },
    {
      "auxiliary_loss_clip": 0.01079412,
      "auxiliary_loss_mlp": 0.01053179,
      "balance_loss_clip": 1.05168915,
      "balance_loss_mlp": 1.0354054,
      "epoch": 0.21698481887870133,
      "flos": 25119047514240.0,
      "grad_norm": 1.7218208369904888,
      "language_loss": 0.74195898,
      "learning_rate": 3.6443066810624255e-06,
      "loss": 0.76328492,
      "num_input_tokens_seen": 77892700,
      "step": 3609,
      "time_per_iteration": 2.660496950149536
    },
    {
      "auxiliary_loss_clip": 0.0113581,
      "auxiliary_loss_mlp": 0.01051499,
      "balance_loss_clip": 1.05560172,
      "balance_loss_mlp": 1.033463,
      "epoch": 0.2170449421313693,
      "flos": 17894682748800.0,
      "grad_norm": 2.3557164844240845,
      "language_loss": 0.89041537,
      "learning_rate": 3.6440849425579e-06,
      "loss": 0.91228849,
      "num_input_tokens_seen": 77911060,
      "step": 3610,
      "time_per_iteration": 2.539219856262207
    },
    {
      "auxiliary_loss_clip": 0.01152,
      "auxiliary_loss_mlp": 0.01049602,
      "balance_loss_clip": 1.05507493,
      "balance_loss_mlp": 1.03181565,
      "epoch": 0.2171050653840373,
      "flos": 22638446265600.0,
      "grad_norm": 1.8178916587516818,
      "language_loss": 0.77619052,
      "learning_rate": 3.6438631417095095e-06,
      "loss": 0.79820657,
      "num_input_tokens_seen": 77929930,
      "step": 3611,
      "time_per_iteration": 2.478644847869873
    },
    {
      "auxiliary_loss_clip": 0.01085074,
      "auxiliary_loss_mlp": 0.01049446,
      "balance_loss_clip": 1.04899311,
      "balance_loss_mlp": 1.03166044,
      "epoch": 0.21716518863670525,
      "flos": 19499997381120.0,
      "grad_norm": 1.853776679390499,
      "language_loss": 0.63276482,
      "learning_rate": 3.6436412785256637e-06,
      "loss": 0.65411007,
      "num_input_tokens_seen": 77949060,
      "step": 3612,
      "time_per_iteration": 2.6393883228302
    },
    {
      "auxiliary_loss_clip": 0.01089425,
      "auxiliary_loss_mlp": 0.01048292,
      "balance_loss_clip": 1.04954159,
      "balance_loss_mlp": 1.02843165,
      "epoch": 0.21722531188937322,
      "flos": 19792022952960.0,
      "grad_norm": 1.9334842430400216,
      "language_loss": 0.75866097,
      "learning_rate": 3.643419353014776e-06,
      "loss": 0.78003812,
      "num_input_tokens_seen": 77967920,
      "step": 3613,
      "time_per_iteration": 2.583744764328003
    },
    {
      "auxiliary_loss_clip": 0.01097335,
      "auxiliary_loss_mlp": 0.01051229,
      "balance_loss_clip": 1.04688537,
      "balance_loss_mlp": 1.03117812,
      "epoch": 0.21728543514204118,
      "flos": 13334386924800.0,
      "grad_norm": 1.9151284413011302,
      "language_loss": 0.70878899,
      "learning_rate": 3.643197365185261e-06,
      "loss": 0.73027456,
      "num_input_tokens_seen": 77985330,
      "step": 3614,
      "time_per_iteration": 2.559095859527588
    },
    {
      "auxiliary_loss_clip": 0.01138105,
      "auxiliary_loss_mlp": 0.01048119,
      "balance_loss_clip": 1.05303538,
      "balance_loss_mlp": 1.03005886,
      "epoch": 0.21734555839470915,
      "flos": 15231870783360.0,
      "grad_norm": 2.0413031836150988,
      "language_loss": 0.73730886,
      "learning_rate": 3.6429753150455378e-06,
      "loss": 0.75917107,
      "num_input_tokens_seen": 78003105,
      "step": 3615,
      "time_per_iteration": 2.469573497772217
    },
    {
      "auxiliary_loss_clip": 0.01140979,
      "auxiliary_loss_mlp": 0.01050005,
      "balance_loss_clip": 1.05161059,
      "balance_loss_mlp": 1.02978766,
      "epoch": 0.2174056816473771,
      "flos": 19973982274560.0,
      "grad_norm": 2.8248249920994124,
      "language_loss": 0.8995797,
      "learning_rate": 3.6427532026040263e-06,
      "loss": 0.92148954,
      "num_input_tokens_seen": 78019655,
      "step": 3616,
      "time_per_iteration": 2.5177431106567383
    },
    {
      "auxiliary_loss_clip": 0.01102302,
      "auxiliary_loss_mlp": 0.01046514,
      "balance_loss_clip": 1.05242956,
      "balance_loss_mlp": 1.02748847,
      "epoch": 0.21746580490004508,
      "flos": 16687293960960.0,
      "grad_norm": 2.268065659368606,
      "language_loss": 0.81489325,
      "learning_rate": 3.642531027869148e-06,
      "loss": 0.83638144,
      "num_input_tokens_seen": 78036025,
      "step": 3617,
      "time_per_iteration": 2.571014404296875
    },
    {
      "auxiliary_loss_clip": 0.01128367,
      "auxiliary_loss_mlp": 0.01043254,
      "balance_loss_clip": 1.05191994,
      "balance_loss_mlp": 1.02573013,
      "epoch": 0.21752592815271307,
      "flos": 25772298209280.0,
      "grad_norm": 1.7111843680517322,
      "language_loss": 0.75594056,
      "learning_rate": 3.642308790849329e-06,
      "loss": 0.77765679,
      "num_input_tokens_seen": 78055645,
      "step": 3618,
      "time_per_iteration": 2.5710856914520264
    },
    {
      "auxiliary_loss_clip": 0.01140793,
      "auxiliary_loss_mlp": 0.01049809,
      "balance_loss_clip": 1.05120873,
      "balance_loss_mlp": 1.0310334,
      "epoch": 0.21758605140538104,
      "flos": 11254692349440.0,
      "grad_norm": 2.0285575099806317,
      "language_loss": 0.69402003,
      "learning_rate": 3.642086491552996e-06,
      "loss": 0.71592605,
      "num_input_tokens_seen": 78071660,
      "step": 3619,
      "time_per_iteration": 2.4710757732391357
    },
    {
      "auxiliary_loss_clip": 0.01141697,
      "auxiliary_loss_mlp": 0.01040816,
      "balance_loss_clip": 1.05565464,
      "balance_loss_mlp": 1.02294636,
      "epoch": 0.217646174658049,
      "flos": 19242625455360.0,
      "grad_norm": 1.6533332513566865,
      "language_loss": 0.78564417,
      "learning_rate": 3.641864129988579e-06,
      "loss": 0.80746937,
      "num_input_tokens_seen": 78091265,
      "step": 3620,
      "time_per_iteration": 2.521606683731079
    },
    {
      "auxiliary_loss_clip": 0.01145987,
      "auxiliary_loss_mlp": 0.01041969,
      "balance_loss_clip": 1.05142391,
      "balance_loss_mlp": 1.02479053,
      "epoch": 0.21770629791071697,
      "flos": 21945083057280.0,
      "grad_norm": 1.7652946494592838,
      "language_loss": 0.79635417,
      "learning_rate": 3.641641706164509e-06,
      "loss": 0.81823373,
      "num_input_tokens_seen": 78110095,
      "step": 3621,
      "time_per_iteration": 2.4824273586273193
    },
    {
      "auxiliary_loss_clip": 0.01137506,
      "auxiliary_loss_mlp": 0.01039996,
      "balance_loss_clip": 1.05007041,
      "balance_loss_mlp": 1.02287781,
      "epoch": 0.21776642116338493,
      "flos": 24936764970240.0,
      "grad_norm": 1.6107048326882176,
      "language_loss": 0.87598693,
      "learning_rate": 3.641419220089221e-06,
      "loss": 0.89776194,
      "num_input_tokens_seen": 78129475,
      "step": 3622,
      "time_per_iteration": 2.5649521350860596
    },
    {
      "auxiliary_loss_clip": 0.01142004,
      "auxiliary_loss_mlp": 0.01040682,
      "balance_loss_clip": 1.05157495,
      "balance_loss_mlp": 1.02095306,
      "epoch": 0.2178265444160529,
      "flos": 17821317219840.0,
      "grad_norm": 1.8932488482814362,
      "language_loss": 0.76937318,
      "learning_rate": 3.641196671771152e-06,
      "loss": 0.7912001,
      "num_input_tokens_seen": 78146880,
      "step": 3623,
      "time_per_iteration": 3.834122657775879
    },
    {
      "auxiliary_loss_clip": 0.01122576,
      "auxiliary_loss_mlp": 0.01045708,
      "balance_loss_clip": 1.05417061,
      "balance_loss_mlp": 1.02678919,
      "epoch": 0.2178866676687209,
      "flos": 17712902995200.0,
      "grad_norm": 1.9053231016610352,
      "language_loss": 0.84546435,
      "learning_rate": 3.640974061218741e-06,
      "loss": 0.86714721,
      "num_input_tokens_seen": 78165065,
      "step": 3624,
      "time_per_iteration": 2.5513288974761963
    },
    {
      "auxiliary_loss_clip": 0.01135224,
      "auxiliary_loss_mlp": 0.01057046,
      "balance_loss_clip": 1.05255723,
      "balance_loss_mlp": 1.03894997,
      "epoch": 0.21794679092138886,
      "flos": 16945851035520.0,
      "grad_norm": 2.5074937833666224,
      "language_loss": 0.76958084,
      "learning_rate": 3.640751388440429e-06,
      "loss": 0.79150355,
      "num_input_tokens_seen": 78180005,
      "step": 3625,
      "time_per_iteration": 3.8914358615875244
    },
    {
      "auxiliary_loss_clip": 0.0105276,
      "auxiliary_loss_mlp": 0.01005708,
      "balance_loss_clip": 1.02740872,
      "balance_loss_mlp": 1.00270355,
      "epoch": 0.21800691417405682,
      "flos": 63718566566400.0,
      "grad_norm": 0.8141305506567106,
      "language_loss": 0.6069662,
      "learning_rate": 3.64052865344466e-06,
      "loss": 0.62755084,
      "num_input_tokens_seen": 78245350,
      "step": 3626,
      "time_per_iteration": 3.2012975215911865
    },
    {
      "auxiliary_loss_clip": 0.01121764,
      "auxiliary_loss_mlp": 0.00804473,
      "balance_loss_clip": 1.05081189,
      "balance_loss_mlp": 1.02917778,
      "epoch": 0.21806703742672479,
      "flos": 21616392677760.0,
      "grad_norm": 1.8559892417672197,
      "language_loss": 0.90313089,
      "learning_rate": 3.6403058562398795e-06,
      "loss": 0.92239332,
      "num_input_tokens_seen": 78264165,
      "step": 3627,
      "time_per_iteration": 2.5755021572113037
    },
    {
      "auxiliary_loss_clip": 0.01095613,
      "auxiliary_loss_mlp": 0.01040345,
      "balance_loss_clip": 1.0499599,
      "balance_loss_mlp": 1.02191544,
      "epoch": 0.21812716067939275,
      "flos": 19354882435200.0,
      "grad_norm": 1.5902262841530739,
      "language_loss": 0.73553944,
      "learning_rate": 3.6400829968345365e-06,
      "loss": 0.756899,
      "num_input_tokens_seen": 78283745,
      "step": 3628,
      "time_per_iteration": 3.9810009002685547
    },
    {
      "auxiliary_loss_clip": 0.01147435,
      "auxiliary_loss_mlp": 0.010426,
      "balance_loss_clip": 1.0501852,
      "balance_loss_mlp": 1.0247426,
      "epoch": 0.21818728393206072,
      "flos": 23548063305600.0,
      "grad_norm": 1.7837663822211787,
      "language_loss": 0.77245831,
      "learning_rate": 3.6398600752370826e-06,
      "loss": 0.79435873,
      "num_input_tokens_seen": 78302900,
      "step": 3629,
      "time_per_iteration": 3.853085994720459
    },
    {
      "auxiliary_loss_clip": 0.01139222,
      "auxiliary_loss_mlp": 0.01039834,
      "balance_loss_clip": 1.05215132,
      "balance_loss_mlp": 1.02235818,
      "epoch": 0.21824740718472868,
      "flos": 30225652266240.0,
      "grad_norm": 1.5901199793897753,
      "language_loss": 0.71160948,
      "learning_rate": 3.63963709145597e-06,
      "loss": 0.73340005,
      "num_input_tokens_seen": 78326470,
      "step": 3630,
      "time_per_iteration": 2.614562511444092
    },
    {
      "auxiliary_loss_clip": 0.01086432,
      "auxiliary_loss_mlp": 0.0104,
      "balance_loss_clip": 1.04956794,
      "balance_loss_mlp": 1.02390695,
      "epoch": 0.21830753043739667,
      "flos": 26134672567680.0,
      "grad_norm": 1.6981907148029305,
      "language_loss": 0.76549131,
      "learning_rate": 3.6394140454996544e-06,
      "loss": 0.78675562,
      "num_input_tokens_seen": 78345810,
      "step": 3631,
      "time_per_iteration": 2.673088312149048
    },
    {
      "auxiliary_loss_clip": 0.01148815,
      "auxiliary_loss_mlp": 0.01040357,
      "balance_loss_clip": 1.05059958,
      "balance_loss_mlp": 1.02329779,
      "epoch": 0.21836765369006464,
      "flos": 21720712752000.0,
      "grad_norm": 2.1952068645162144,
      "language_loss": 0.75077474,
      "learning_rate": 3.639190937376594e-06,
      "loss": 0.77266645,
      "num_input_tokens_seen": 78364085,
      "step": 3632,
      "time_per_iteration": 2.492011785507202
    },
    {
      "auxiliary_loss_clip": 0.01146546,
      "auxiliary_loss_mlp": 0.01037936,
      "balance_loss_clip": 1.05101633,
      "balance_loss_mlp": 1.02160406,
      "epoch": 0.2184277769427326,
      "flos": 19937604775680.0,
      "grad_norm": 2.0216932807285284,
      "language_loss": 0.83486921,
      "learning_rate": 3.638967767095249e-06,
      "loss": 0.85671401,
      "num_input_tokens_seen": 78381385,
      "step": 3633,
      "time_per_iteration": 2.4654104709625244
    },
    {
      "auxiliary_loss_clip": 0.01124123,
      "auxiliary_loss_mlp": 0.01057729,
      "balance_loss_clip": 1.05485952,
      "balance_loss_mlp": 1.04073024,
      "epoch": 0.21848790019540057,
      "flos": 20340235301760.0,
      "grad_norm": 1.7246689420399512,
      "language_loss": 0.81501102,
      "learning_rate": 3.6387445346640823e-06,
      "loss": 0.83682954,
      "num_input_tokens_seen": 78400500,
      "step": 3634,
      "time_per_iteration": 2.5703341960906982
    },
    {
      "auxiliary_loss_clip": 0.01143004,
      "auxiliary_loss_mlp": 0.01042236,
      "balance_loss_clip": 1.05510712,
      "balance_loss_mlp": 1.02504647,
      "epoch": 0.21854802344806853,
      "flos": 15450818135040.0,
      "grad_norm": 2.5500429620687934,
      "language_loss": 0.75350702,
      "learning_rate": 3.638521240091558e-06,
      "loss": 0.77535939,
      "num_input_tokens_seen": 78418340,
      "step": 3635,
      "time_per_iteration": 2.4803948402404785
    },
    {
      "auxiliary_loss_clip": 0.01124189,
      "auxiliary_loss_mlp": 0.01054297,
      "balance_loss_clip": 1.05287313,
      "balance_loss_mlp": 1.03717875,
      "epoch": 0.2186081467007365,
      "flos": 16320717711360.0,
      "grad_norm": 2.559445710355263,
      "language_loss": 0.88346934,
      "learning_rate": 3.6382978833861445e-06,
      "loss": 0.90525424,
      "num_input_tokens_seen": 78434375,
      "step": 3636,
      "time_per_iteration": 2.503364324569702
    },
    {
      "auxiliary_loss_clip": 0.01114252,
      "auxiliary_loss_mlp": 0.00793966,
      "balance_loss_clip": 1.05357218,
      "balance_loss_mlp": 1.01198399,
      "epoch": 0.2186682699534045,
      "flos": 21689255416320.0,
      "grad_norm": 2.0331074691652375,
      "language_loss": 0.75864869,
      "learning_rate": 3.638074464556311e-06,
      "loss": 0.77773088,
      "num_input_tokens_seen": 78451735,
      "step": 3637,
      "time_per_iteration": 2.592402219772339
    },
    {
      "auxiliary_loss_clip": 0.01135702,
      "auxiliary_loss_mlp": 0.0104715,
      "balance_loss_clip": 1.05674338,
      "balance_loss_mlp": 1.02862453,
      "epoch": 0.21872839320607246,
      "flos": 17739260599680.0,
      "grad_norm": 3.3966209260532887,
      "language_loss": 0.90023845,
      "learning_rate": 3.63785098361053e-06,
      "loss": 0.92206699,
      "num_input_tokens_seen": 78462730,
      "step": 3638,
      "time_per_iteration": 2.4894158840179443
    },
    {
      "auxiliary_loss_clip": 0.01138474,
      "auxiliary_loss_mlp": 0.01052528,
      "balance_loss_clip": 1.05254579,
      "balance_loss_mlp": 1.03452778,
      "epoch": 0.21878851645874042,
      "flos": 18652289431680.0,
      "grad_norm": 2.6170814372675655,
      "language_loss": 0.89828521,
      "learning_rate": 3.637627440557275e-06,
      "loss": 0.92019522,
      "num_input_tokens_seen": 78476300,
      "step": 3639,
      "time_per_iteration": 2.4726271629333496
    },
    {
      "auxiliary_loss_clip": 0.01130485,
      "auxiliary_loss_mlp": 0.00798461,
      "balance_loss_clip": 1.05296171,
      "balance_loss_mlp": 1.01790214,
      "epoch": 0.2188486397114084,
      "flos": 25557301353600.0,
      "grad_norm": 1.790890356880474,
      "language_loss": 0.79418796,
      "learning_rate": 3.637403835405024e-06,
      "loss": 0.8134774,
      "num_input_tokens_seen": 78496135,
      "step": 3640,
      "time_per_iteration": 2.5811336040496826
    },
    {
      "auxiliary_loss_clip": 0.01145089,
      "auxiliary_loss_mlp": 0.01051979,
      "balance_loss_clip": 1.05585778,
      "balance_loss_mlp": 1.03235674,
      "epoch": 0.21890876296407635,
      "flos": 17892061056000.0,
      "grad_norm": 2.188627488397322,
      "language_loss": 0.72094464,
      "learning_rate": 3.637180168162255e-06,
      "loss": 0.74291539,
      "num_input_tokens_seen": 78513855,
      "step": 3641,
      "time_per_iteration": 2.5281214714050293
    },
    {
      "auxiliary_loss_clip": 0.01128554,
      "auxiliary_loss_mlp": 0.01044531,
      "balance_loss_clip": 1.05507803,
      "balance_loss_mlp": 1.02685237,
      "epoch": 0.21896888621674432,
      "flos": 17749100926080.0,
      "grad_norm": 2.32847833119919,
      "language_loss": 0.80929869,
      "learning_rate": 3.63695643883745e-06,
      "loss": 0.83102953,
      "num_input_tokens_seen": 78531740,
      "step": 3642,
      "time_per_iteration": 2.511493682861328
    },
    {
      "auxiliary_loss_clip": 0.01143124,
      "auxiliary_loss_mlp": 0.01045009,
      "balance_loss_clip": 1.05282879,
      "balance_loss_mlp": 1.02636445,
      "epoch": 0.21902900946941228,
      "flos": 23076161400960.0,
      "grad_norm": 1.8034267510644242,
      "language_loss": 0.71855789,
      "learning_rate": 3.6367326474390928e-06,
      "loss": 0.74043918,
      "num_input_tokens_seen": 78549600,
      "step": 3643,
      "time_per_iteration": 2.5203025341033936
    },
    {
      "auxiliary_loss_clip": 0.0115046,
      "auxiliary_loss_mlp": 0.01052039,
      "balance_loss_clip": 1.05209136,
      "balance_loss_mlp": 1.033288,
      "epoch": 0.21908913272208028,
      "flos": 48178545004800.0,
      "grad_norm": 1.8703035119232514,
      "language_loss": 0.68581176,
      "learning_rate": 3.6365087939756696e-06,
      "loss": 0.70783675,
      "num_input_tokens_seen": 78573350,
      "step": 3644,
      "time_per_iteration": 2.7120461463928223
    },
    {
      "auxiliary_loss_clip": 0.01153036,
      "auxiliary_loss_mlp": 0.0104401,
      "balance_loss_clip": 1.05227304,
      "balance_loss_mlp": 1.02618814,
      "epoch": 0.21914925597474824,
      "flos": 22236749493120.0,
      "grad_norm": 2.0012820553989954,
      "language_loss": 0.78040653,
      "learning_rate": 3.636284878455669e-06,
      "loss": 0.80237699,
      "num_input_tokens_seen": 78591005,
      "step": 3645,
      "time_per_iteration": 2.5040066242218018
    },
    {
      "auxiliary_loss_clip": 0.01139268,
      "auxiliary_loss_mlp": 0.01049955,
      "balance_loss_clip": 1.05549979,
      "balance_loss_mlp": 1.03303885,
      "epoch": 0.2192093792274162,
      "flos": 22125605834880.0,
      "grad_norm": 1.7210858963471396,
      "language_loss": 0.82403046,
      "learning_rate": 3.636060900887582e-06,
      "loss": 0.84592271,
      "num_input_tokens_seen": 78610645,
      "step": 3646,
      "time_per_iteration": 2.513605833053589
    },
    {
      "auxiliary_loss_clip": 0.01135278,
      "auxiliary_loss_mlp": 0.01040399,
      "balance_loss_clip": 1.05169165,
      "balance_loss_mlp": 1.02376974,
      "epoch": 0.21926950248008417,
      "flos": 15669442264320.0,
      "grad_norm": 1.7081906475808488,
      "language_loss": 0.82634318,
      "learning_rate": 3.635836861279901e-06,
      "loss": 0.84809995,
      "num_input_tokens_seen": 78628340,
      "step": 3647,
      "time_per_iteration": 2.4788360595703125
    },
    {
      "auxiliary_loss_clip": 0.01145978,
      "auxiliary_loss_mlp": 0.01052497,
      "balance_loss_clip": 1.05057263,
      "balance_loss_mlp": 1.03539014,
      "epoch": 0.21932962573275214,
      "flos": 30262496641920.0,
      "grad_norm": 1.7749354060663207,
      "language_loss": 0.7323482,
      "learning_rate": 3.635612759641123e-06,
      "loss": 0.75433296,
      "num_input_tokens_seen": 78649355,
      "step": 3648,
      "time_per_iteration": 2.557774066925049
    },
    {
      "auxiliary_loss_clip": 0.01104952,
      "auxiliary_loss_mlp": 0.01058232,
      "balance_loss_clip": 1.04666889,
      "balance_loss_mlp": 1.0370959,
      "epoch": 0.2193897489854201,
      "flos": 10780132838400.0,
      "grad_norm": 2.471219577721094,
      "language_loss": 0.73945308,
      "learning_rate": 3.635388595979745e-06,
      "loss": 0.76108491,
      "num_input_tokens_seen": 78664915,
      "step": 3649,
      "time_per_iteration": 2.5654475688934326
    },
    {
      "auxiliary_loss_clip": 0.01130715,
      "auxiliary_loss_mlp": 0.01047321,
      "balance_loss_clip": 1.04958594,
      "balance_loss_mlp": 1.03069162,
      "epoch": 0.21944987223808807,
      "flos": 19133313390720.0,
      "grad_norm": 2.121476514384449,
      "language_loss": 0.86315417,
      "learning_rate": 3.635164370304267e-06,
      "loss": 0.88493454,
      "num_input_tokens_seen": 78681475,
      "step": 3650,
      "time_per_iteration": 2.482602119445801
    },
    {
      "auxiliary_loss_clip": 0.0112541,
      "auxiliary_loss_mlp": 0.01045072,
      "balance_loss_clip": 1.04912126,
      "balance_loss_mlp": 1.02690494,
      "epoch": 0.21950999549075606,
      "flos": 22711093522560.0,
      "grad_norm": 1.9734162715994552,
      "language_loss": 0.8403163,
      "learning_rate": 3.6349400826231927e-06,
      "loss": 0.86202115,
      "num_input_tokens_seen": 78702300,
      "step": 3651,
      "time_per_iteration": 2.5919017791748047
    },
    {
      "auxiliary_loss_clip": 0.01137732,
      "auxiliary_loss_mlp": 0.01048977,
      "balance_loss_clip": 1.05056357,
      "balance_loss_mlp": 1.03121483,
      "epoch": 0.21957011874342403,
      "flos": 10561329141120.0,
      "grad_norm": 1.8554151170090756,
      "language_loss": 0.74118459,
      "learning_rate": 3.634715732945027e-06,
      "loss": 0.76305175,
      "num_input_tokens_seen": 78720230,
      "step": 3652,
      "time_per_iteration": 2.4985525608062744
    },
    {
      "auxiliary_loss_clip": 0.01026228,
      "auxiliary_loss_mlp": 0.01015326,
      "balance_loss_clip": 1.03012836,
      "balance_loss_mlp": 1.01265609,
      "epoch": 0.219630241996092,
      "flos": 65747913252480.0,
      "grad_norm": 0.736240809805818,
      "language_loss": 0.5153445,
      "learning_rate": 3.6344913212782764e-06,
      "loss": 0.53576005,
      "num_input_tokens_seen": 78780200,
      "step": 3653,
      "time_per_iteration": 3.2133591175079346
    },
    {
      "auxiliary_loss_clip": 0.01122818,
      "auxiliary_loss_mlp": 0.01057069,
      "balance_loss_clip": 1.05662787,
      "balance_loss_mlp": 1.03911567,
      "epoch": 0.21969036524875996,
      "flos": 23696518216320.0,
      "grad_norm": 1.819961328054342,
      "language_loss": 0.75238478,
      "learning_rate": 3.6342668476314514e-06,
      "loss": 0.77418363,
      "num_input_tokens_seen": 78800575,
      "step": 3654,
      "time_per_iteration": 2.568155527114868
    },
    {
      "auxiliary_loss_clip": 0.01148011,
      "auxiliary_loss_mlp": 0.01045644,
      "balance_loss_clip": 1.05581641,
      "balance_loss_mlp": 1.02748823,
      "epoch": 0.21975048850142792,
      "flos": 19640910435840.0,
      "grad_norm": 2.8254009014777206,
      "language_loss": 0.72672415,
      "learning_rate": 3.634042312013064e-06,
      "loss": 0.74866062,
      "num_input_tokens_seen": 78819585,
      "step": 3655,
      "time_per_iteration": 2.517900228500366
    },
    {
      "auxiliary_loss_clip": 0.01125767,
      "auxiliary_loss_mlp": 0.0104805,
      "balance_loss_clip": 1.05538416,
      "balance_loss_mlp": 1.0304184,
      "epoch": 0.21981061175409589,
      "flos": 22448550038400.0,
      "grad_norm": 1.6535189588574208,
      "language_loss": 0.80887252,
      "learning_rate": 3.6338177144316276e-06,
      "loss": 0.83061069,
      "num_input_tokens_seen": 78837330,
      "step": 3656,
      "time_per_iteration": 2.5567877292633057
    },
    {
      "auxiliary_loss_clip": 0.01113478,
      "auxiliary_loss_mlp": 0.00798333,
      "balance_loss_clip": 1.05636406,
      "balance_loss_mlp": 1.01843941,
      "epoch": 0.21987073500676388,
      "flos": 18151049093760.0,
      "grad_norm": 2.417120527051926,
      "language_loss": 0.84727514,
      "learning_rate": 3.63359305489566e-06,
      "loss": 0.86639321,
      "num_input_tokens_seen": 78854955,
      "step": 3657,
      "time_per_iteration": 2.568197011947632
    },
    {
      "auxiliary_loss_clip": 0.01132715,
      "auxiliary_loss_mlp": 0.0103984,
      "balance_loss_clip": 1.05129385,
      "balance_loss_mlp": 1.02212572,
      "epoch": 0.21993085825943184,
      "flos": 25626177682560.0,
      "grad_norm": 2.9191119643299333,
      "language_loss": 0.80479985,
      "learning_rate": 3.6333683334136803e-06,
      "loss": 0.82652545,
      "num_input_tokens_seen": 78874965,
      "step": 3658,
      "time_per_iteration": 2.5533339977264404
    },
    {
      "auxiliary_loss_clip": 0.01040369,
      "auxiliary_loss_mlp": 0.01011592,
      "balance_loss_clip": 1.03445244,
      "balance_loss_mlp": 1.00846851,
      "epoch": 0.2199909815120998,
      "flos": 70923217743360.0,
      "grad_norm": 0.8046671494574841,
      "language_loss": 0.58160567,
      "learning_rate": 3.6331435499942095e-06,
      "loss": 0.60212529,
      "num_input_tokens_seen": 78937740,
      "step": 3659,
      "time_per_iteration": 3.254161834716797
    },
    {
      "auxiliary_loss_clip": 0.01106953,
      "auxiliary_loss_mlp": 0.01042364,
      "balance_loss_clip": 1.04993725,
      "balance_loss_mlp": 1.0243752,
      "epoch": 0.22005110476476777,
      "flos": 21543529939200.0,
      "grad_norm": 2.27163333025809,
      "language_loss": 0.73858714,
      "learning_rate": 3.632918704645772e-06,
      "loss": 0.76008034,
      "num_input_tokens_seen": 78955055,
      "step": 3660,
      "time_per_iteration": 2.5775833129882812
    },
    {
      "auxiliary_loss_clip": 0.01139356,
      "auxiliary_loss_mlp": 0.01042564,
      "balance_loss_clip": 1.05317688,
      "balance_loss_mlp": 1.02477825,
      "epoch": 0.22011122801743574,
      "flos": 22054502862720.0,
      "grad_norm": 1.738165395136322,
      "language_loss": 0.81423593,
      "learning_rate": 3.632693797376893e-06,
      "loss": 0.83605516,
      "num_input_tokens_seen": 78974895,
      "step": 3661,
      "time_per_iteration": 2.5280983448028564
    },
    {
      "auxiliary_loss_clip": 0.0111834,
      "auxiliary_loss_mlp": 0.01049401,
      "balance_loss_clip": 1.04847193,
      "balance_loss_mlp": 1.03149545,
      "epoch": 0.2201713512701037,
      "flos": 26687589598080.0,
      "grad_norm": 1.6748471946560908,
      "language_loss": 0.73097098,
      "learning_rate": 3.632468828196102e-06,
      "loss": 0.75264841,
      "num_input_tokens_seen": 78994990,
      "step": 3662,
      "time_per_iteration": 3.9510345458984375
    },
    {
      "auxiliary_loss_clip": 0.011277,
      "auxiliary_loss_mlp": 0.01051458,
      "balance_loss_clip": 1.05445957,
      "balance_loss_mlp": 1.03554368,
      "epoch": 0.22023147452277167,
      "flos": 22162198815360.0,
      "grad_norm": 1.6042297878706626,
      "language_loss": 0.78651249,
      "learning_rate": 3.632243797111929e-06,
      "loss": 0.80830407,
      "num_input_tokens_seen": 79014405,
      "step": 3663,
      "time_per_iteration": 2.5583837032318115
    },
    {
      "auxiliary_loss_clip": 0.0113569,
      "auxiliary_loss_mlp": 0.01049792,
      "balance_loss_clip": 1.05353284,
      "balance_loss_mlp": 1.02995598,
      "epoch": 0.22029159777543966,
      "flos": 22523280284160.0,
      "grad_norm": 1.9359029100779455,
      "language_loss": 0.80439568,
      "learning_rate": 3.632018704132908e-06,
      "loss": 0.82625055,
      "num_input_tokens_seen": 79032375,
      "step": 3664,
      "time_per_iteration": 3.91141414642334
    },
    {
      "auxiliary_loss_clip": 0.01130591,
      "auxiliary_loss_mlp": 0.01044625,
      "balance_loss_clip": 1.05584669,
      "balance_loss_mlp": 1.02501488,
      "epoch": 0.22035172102810763,
      "flos": 13042469093760.0,
      "grad_norm": 2.6942624598886393,
      "language_loss": 0.76719534,
      "learning_rate": 3.6317935492675742e-06,
      "loss": 0.78894752,
      "num_input_tokens_seen": 79049635,
      "step": 3665,
      "time_per_iteration": 2.5394208431243896
    },
    {
      "auxiliary_loss_clip": 0.01118303,
      "auxiliary_loss_mlp": 0.01051174,
      "balance_loss_clip": 1.05206954,
      "balance_loss_mlp": 1.03386521,
      "epoch": 0.2204118442807756,
      "flos": 12165817760640.0,
      "grad_norm": 2.8606731258114464,
      "language_loss": 0.98223019,
      "learning_rate": 3.631568332524466e-06,
      "loss": 1.00392509,
      "num_input_tokens_seen": 79062890,
      "step": 3666,
      "time_per_iteration": 2.4899590015411377
    },
    {
      "auxiliary_loss_clip": 0.01137924,
      "auxiliary_loss_mlp": 0.00796288,
      "balance_loss_clip": 1.05316877,
      "balance_loss_mlp": 1.01559734,
      "epoch": 0.22047196753344356,
      "flos": 40108806673920.0,
      "grad_norm": 1.7645787496672711,
      "language_loss": 0.80873024,
      "learning_rate": 3.631343053912122e-06,
      "loss": 0.82807231,
      "num_input_tokens_seen": 79085495,
      "step": 3667,
      "time_per_iteration": 2.7084782123565674
    },
    {
      "auxiliary_loss_clip": 0.01138773,
      "auxiliary_loss_mlp": 0.0105124,
      "balance_loss_clip": 1.0560801,
      "balance_loss_mlp": 1.03166616,
      "epoch": 0.22053209078611152,
      "flos": 20701137202560.0,
      "grad_norm": 1.9930794879695934,
      "language_loss": 0.77217424,
      "learning_rate": 3.631117713439087e-06,
      "loss": 0.79407442,
      "num_input_tokens_seen": 79101820,
      "step": 3668,
      "time_per_iteration": 3.941887855529785
    },
    {
      "auxiliary_loss_clip": 0.01134343,
      "auxiliary_loss_mlp": 0.0104531,
      "balance_loss_clip": 1.0558877,
      "balance_loss_mlp": 1.02775013,
      "epoch": 0.2205922140387795,
      "flos": 24716309247360.0,
      "grad_norm": 1.703702445548757,
      "language_loss": 0.71556854,
      "learning_rate": 3.630892311113904e-06,
      "loss": 0.73736507,
      "num_input_tokens_seen": 79123320,
      "step": 3669,
      "time_per_iteration": 2.571268320083618
    },
    {
      "auxiliary_loss_clip": 0.01149285,
      "auxiliary_loss_mlp": 0.0103946,
      "balance_loss_clip": 1.05244219,
      "balance_loss_mlp": 1.02286577,
      "epoch": 0.22065233729144745,
      "flos": 23477247642240.0,
      "grad_norm": 1.7612693055331046,
      "language_loss": 0.8569867,
      "learning_rate": 3.6306668469451215e-06,
      "loss": 0.87887418,
      "num_input_tokens_seen": 79141615,
      "step": 3670,
      "time_per_iteration": 2.477865695953369
    },
    {
      "auxiliary_loss_clip": 0.01130697,
      "auxiliary_loss_mlp": 0.01040876,
      "balance_loss_clip": 1.05380821,
      "balance_loss_mlp": 1.02308989,
      "epoch": 0.22071246054411545,
      "flos": 35225566646400.0,
      "grad_norm": 1.7464844974907983,
      "language_loss": 0.7684592,
      "learning_rate": 3.6304413209412886e-06,
      "loss": 0.79017496,
      "num_input_tokens_seen": 79164910,
      "step": 3671,
      "time_per_iteration": 2.655616521835327
    },
    {
      "auxiliary_loss_clip": 0.01121889,
      "auxiliary_loss_mlp": 0.01038324,
      "balance_loss_clip": 1.05446315,
      "balance_loss_mlp": 1.02138424,
      "epoch": 0.2207725837967834,
      "flos": 18150294908160.0,
      "grad_norm": 2.911520609074442,
      "language_loss": 0.81266344,
      "learning_rate": 3.6302157331109573e-06,
      "loss": 0.83426559,
      "num_input_tokens_seen": 79179685,
      "step": 3672,
      "time_per_iteration": 2.490204334259033
    },
    {
      "auxiliary_loss_clip": 0.01143399,
      "auxiliary_loss_mlp": 0.01047814,
      "balance_loss_clip": 1.05505991,
      "balance_loss_mlp": 1.03031433,
      "epoch": 0.22083270704945138,
      "flos": 20479675898880.0,
      "grad_norm": 1.9924927758171853,
      "language_loss": 0.73793137,
      "learning_rate": 3.629990083462682e-06,
      "loss": 0.75984353,
      "num_input_tokens_seen": 79196285,
      "step": 3673,
      "time_per_iteration": 2.508039712905884
    },
    {
      "auxiliary_loss_clip": 0.01119155,
      "auxiliary_loss_mlp": 0.01042786,
      "balance_loss_clip": 1.05900717,
      "balance_loss_mlp": 1.02364051,
      "epoch": 0.22089283030211934,
      "flos": 34125801984000.0,
      "grad_norm": 2.117045710983467,
      "language_loss": 0.76393187,
      "learning_rate": 3.6297643720050203e-06,
      "loss": 0.78555125,
      "num_input_tokens_seen": 79216060,
      "step": 3674,
      "time_per_iteration": 2.693067789077759
    },
    {
      "auxiliary_loss_clip": 0.01152796,
      "auxiliary_loss_mlp": 0.01043986,
      "balance_loss_clip": 1.05672407,
      "balance_loss_mlp": 1.02574718,
      "epoch": 0.2209529535547873,
      "flos": 18077216688000.0,
      "grad_norm": 2.202236180231166,
      "language_loss": 0.74304384,
      "learning_rate": 3.6295385987465293e-06,
      "loss": 0.76501167,
      "num_input_tokens_seen": 79235145,
      "step": 3675,
      "time_per_iteration": 2.4624598026275635
    },
    {
      "auxiliary_loss_clip": 0.0115188,
      "auxiliary_loss_mlp": 0.01044317,
      "balance_loss_clip": 1.05466366,
      "balance_loss_mlp": 1.02716303,
      "epoch": 0.22101307680745527,
      "flos": 27235335070080.0,
      "grad_norm": 1.772344629817283,
      "language_loss": 0.80377364,
      "learning_rate": 3.629312763695772e-06,
      "loss": 0.82573569,
      "num_input_tokens_seen": 79256960,
      "step": 3676,
      "time_per_iteration": 2.5380258560180664
    },
    {
      "auxiliary_loss_clip": 0.01128244,
      "auxiliary_loss_mlp": 0.01050583,
      "balance_loss_clip": 1.0536648,
      "balance_loss_mlp": 1.03392899,
      "epoch": 0.22107320006012326,
      "flos": 16543256423040.0,
      "grad_norm": 2.1110987433072115,
      "language_loss": 0.7507714,
      "learning_rate": 3.6290868668613107e-06,
      "loss": 0.7725597,
      "num_input_tokens_seen": 79274860,
      "step": 3677,
      "time_per_iteration": 2.521003007888794
    },
    {
      "auxiliary_loss_clip": 0.01116669,
      "auxiliary_loss_mlp": 0.01043728,
      "balance_loss_clip": 1.05517185,
      "balance_loss_mlp": 1.0265981,
      "epoch": 0.22113332331279123,
      "flos": 22054466949120.0,
      "grad_norm": 1.8953834310992432,
      "language_loss": 0.83286393,
      "learning_rate": 3.628860908251712e-06,
      "loss": 0.85446787,
      "num_input_tokens_seen": 79294005,
      "step": 3678,
      "time_per_iteration": 2.6081435680389404
    },
    {
      "auxiliary_loss_clip": 0.010937,
      "auxiliary_loss_mlp": 0.01052785,
      "balance_loss_clip": 1.05674231,
      "balance_loss_mlp": 1.03386664,
      "epoch": 0.2211934465654592,
      "flos": 26612787525120.0,
      "grad_norm": 1.88170368729481,
      "language_loss": 0.89115268,
      "learning_rate": 3.6286348878755452e-06,
      "loss": 0.91261756,
      "num_input_tokens_seen": 79314005,
      "step": 3679,
      "time_per_iteration": 2.6403069496154785
    },
    {
      "auxiliary_loss_clip": 0.0114709,
      "auxiliary_loss_mlp": 0.01056913,
      "balance_loss_clip": 1.05789602,
      "balance_loss_mlp": 1.03835249,
      "epoch": 0.22125356981812716,
      "flos": 16360363347840.0,
      "grad_norm": 2.316225277391517,
      "language_loss": 0.86878163,
      "learning_rate": 3.6284088057413803e-06,
      "loss": 0.8908217,
      "num_input_tokens_seen": 79331030,
      "step": 3680,
      "time_per_iteration": 2.497048854827881
    },
    {
      "auxiliary_loss_clip": 0.01106552,
      "auxiliary_loss_mlp": 0.01044452,
      "balance_loss_clip": 1.05266786,
      "balance_loss_mlp": 1.02716684,
      "epoch": 0.22131369307079513,
      "flos": 21651118151040.0,
      "grad_norm": 1.9952225144895441,
      "language_loss": 0.81557775,
      "learning_rate": 3.6281826618577894e-06,
      "loss": 0.83708787,
      "num_input_tokens_seen": 79348560,
      "step": 3681,
      "time_per_iteration": 2.5601754188537598
    },
    {
      "auxiliary_loss_clip": 0.01148144,
      "auxiliary_loss_mlp": 0.00799366,
      "balance_loss_clip": 1.0557065,
      "balance_loss_mlp": 1.02368104,
      "epoch": 0.2213738163234631,
      "flos": 19609524927360.0,
      "grad_norm": 2.4025946647542833,
      "language_loss": 0.79279947,
      "learning_rate": 3.62795645623335e-06,
      "loss": 0.81227458,
      "num_input_tokens_seen": 79367175,
      "step": 3682,
      "time_per_iteration": 2.5056543350219727
    },
    {
      "auxiliary_loss_clip": 0.01120446,
      "auxiliary_loss_mlp": 0.0104859,
      "balance_loss_clip": 1.05120373,
      "balance_loss_mlp": 1.02979112,
      "epoch": 0.22143393957613106,
      "flos": 23623404082560.0,
      "grad_norm": 1.7884064455188728,
      "language_loss": 0.77415484,
      "learning_rate": 3.627730188876638e-06,
      "loss": 0.79584527,
      "num_input_tokens_seen": 79388435,
      "step": 3683,
      "time_per_iteration": 2.5760154724121094
    },
    {
      "auxiliary_loss_clip": 0.01126291,
      "auxiliary_loss_mlp": 0.01047339,
      "balance_loss_clip": 1.05156612,
      "balance_loss_mlp": 1.0295887,
      "epoch": 0.22149406282879905,
      "flos": 26177801823360.0,
      "grad_norm": 1.8300147832470628,
      "language_loss": 0.7275818,
      "learning_rate": 3.627503859796234e-06,
      "loss": 0.74931812,
      "num_input_tokens_seen": 79407910,
      "step": 3684,
      "time_per_iteration": 2.608365774154663
    },
    {
      "auxiliary_loss_clip": 0.01086047,
      "auxiliary_loss_mlp": 0.01048394,
      "balance_loss_clip": 1.05391777,
      "balance_loss_mlp": 1.02902293,
      "epoch": 0.221554186081467,
      "flos": 14538758970240.0,
      "grad_norm": 3.4272035833049133,
      "language_loss": 0.80310142,
      "learning_rate": 3.6272774690007207e-06,
      "loss": 0.82444578,
      "num_input_tokens_seen": 79424020,
      "step": 3685,
      "time_per_iteration": 2.5939111709594727
    },
    {
      "auxiliary_loss_clip": 0.01144405,
      "auxiliary_loss_mlp": 0.01043214,
      "balance_loss_clip": 1.05217791,
      "balance_loss_mlp": 1.02704895,
      "epoch": 0.22161430933413498,
      "flos": 22238257864320.0,
      "grad_norm": 1.6102466415262626,
      "language_loss": 0.8747015,
      "learning_rate": 3.6270510164986823e-06,
      "loss": 0.89657766,
      "num_input_tokens_seen": 79445605,
      "step": 3686,
      "time_per_iteration": 2.5217339992523193
    },
    {
      "auxiliary_loss_clip": 0.01133111,
      "auxiliary_loss_mlp": 0.01046641,
      "balance_loss_clip": 1.05104399,
      "balance_loss_mlp": 1.02866387,
      "epoch": 0.22167443258680294,
      "flos": 23476529370240.0,
      "grad_norm": 1.859886786993196,
      "language_loss": 0.77638704,
      "learning_rate": 3.626824502298707e-06,
      "loss": 0.79818451,
      "num_input_tokens_seen": 79463850,
      "step": 3687,
      "time_per_iteration": 2.5138392448425293
    },
    {
      "auxiliary_loss_clip": 0.01123905,
      "auxiliary_loss_mlp": 0.01055128,
      "balance_loss_clip": 1.05339372,
      "balance_loss_mlp": 1.03631687,
      "epoch": 0.2217345558394709,
      "flos": 23221132692480.0,
      "grad_norm": 1.9452480644119738,
      "language_loss": 0.84887278,
      "learning_rate": 3.626597926409383e-06,
      "loss": 0.87066317,
      "num_input_tokens_seen": 79482845,
      "step": 3688,
      "time_per_iteration": 2.5564327239990234
    },
    {
      "auxiliary_loss_clip": 0.01106523,
      "auxiliary_loss_mlp": 0.01044934,
      "balance_loss_clip": 1.05073452,
      "balance_loss_mlp": 1.02617025,
      "epoch": 0.22179467909213887,
      "flos": 20011078045440.0,
      "grad_norm": 1.7931838952577865,
      "language_loss": 0.81072587,
      "learning_rate": 3.6263712888393027e-06,
      "loss": 0.83224046,
      "num_input_tokens_seen": 79501550,
      "step": 3689,
      "time_per_iteration": 2.57620906829834
    },
    {
      "auxiliary_loss_clip": 0.01122873,
      "auxiliary_loss_mlp": 0.01045406,
      "balance_loss_clip": 1.05291712,
      "balance_loss_mlp": 1.02758455,
      "epoch": 0.22185480234480687,
      "flos": 19683034110720.0,
      "grad_norm": 1.868179416728469,
      "language_loss": 0.70150167,
      "learning_rate": 3.626144589597061e-06,
      "loss": 0.72318447,
      "num_input_tokens_seen": 79519680,
      "step": 3690,
      "time_per_iteration": 2.542576789855957
    },
    {
      "auxiliary_loss_clip": 0.01142259,
      "auxiliary_loss_mlp": 0.00795306,
      "balance_loss_clip": 1.05227363,
      "balance_loss_mlp": 1.01354825,
      "epoch": 0.22191492559747483,
      "flos": 21981316901760.0,
      "grad_norm": 1.707299056919887,
      "language_loss": 0.72108769,
      "learning_rate": 3.6259178286912528e-06,
      "loss": 0.74046338,
      "num_input_tokens_seen": 79539000,
      "step": 3691,
      "time_per_iteration": 2.5178792476654053
    },
    {
      "auxiliary_loss_clip": 0.01136474,
      "auxiliary_loss_mlp": 0.01045627,
      "balance_loss_clip": 1.05352926,
      "balance_loss_mlp": 1.0269351,
      "epoch": 0.2219750488501428,
      "flos": 23222066446080.0,
      "grad_norm": 1.9451877437974838,
      "language_loss": 0.71676397,
      "learning_rate": 3.625691006130477e-06,
      "loss": 0.738585,
      "num_input_tokens_seen": 79559695,
      "step": 3692,
      "time_per_iteration": 2.5641558170318604
    },
    {
      "auxiliary_loss_clip": 0.01144915,
      "auxiliary_loss_mlp": 0.0104927,
      "balance_loss_clip": 1.05261433,
      "balance_loss_mlp": 1.03159142,
      "epoch": 0.22203517210281076,
      "flos": 22453685683200.0,
      "grad_norm": 1.7647431416611872,
      "language_loss": 0.87377274,
      "learning_rate": 3.6254641219233362e-06,
      "loss": 0.89571464,
      "num_input_tokens_seen": 79579095,
      "step": 3693,
      "time_per_iteration": 2.504927635192871
    },
    {
      "auxiliary_loss_clip": 0.01136663,
      "auxiliary_loss_mlp": 0.01040074,
      "balance_loss_clip": 1.0533886,
      "balance_loss_mlp": 1.02430308,
      "epoch": 0.22209529535547873,
      "flos": 17564555825280.0,
      "grad_norm": 1.967173840771144,
      "language_loss": 0.85631371,
      "learning_rate": 3.6252371760784325e-06,
      "loss": 0.87808108,
      "num_input_tokens_seen": 79596430,
      "step": 3694,
      "time_per_iteration": 2.4905030727386475
    },
    {
      "auxiliary_loss_clip": 0.01107239,
      "auxiliary_loss_mlp": 0.01043881,
      "balance_loss_clip": 1.05040383,
      "balance_loss_mlp": 1.0256418,
      "epoch": 0.2221554186081467,
      "flos": 21469015175040.0,
      "grad_norm": 2.383949844894555,
      "language_loss": 0.68895102,
      "learning_rate": 3.6250101686043725e-06,
      "loss": 0.71046221,
      "num_input_tokens_seen": 79615825,
      "step": 3695,
      "time_per_iteration": 2.5621185302734375
    },
    {
      "auxiliary_loss_clip": 0.0111475,
      "auxiliary_loss_mlp": 0.0104375,
      "balance_loss_clip": 1.05534911,
      "balance_loss_mlp": 1.02731109,
      "epoch": 0.22221554186081466,
      "flos": 27673445255040.0,
      "grad_norm": 1.4445092076790509,
      "language_loss": 0.7184068,
      "learning_rate": 3.6247830995097637e-06,
      "loss": 0.73999178,
      "num_input_tokens_seen": 79637875,
      "step": 3696,
      "time_per_iteration": 2.6141748428344727
    },
    {
      "auxiliary_loss_clip": 0.01134903,
      "auxiliary_loss_mlp": 0.01045553,
      "balance_loss_clip": 1.0508194,
      "balance_loss_mlp": 1.0276835,
      "epoch": 0.22227566511348265,
      "flos": 25958926298880.0,
      "grad_norm": 2.1061061338910796,
      "language_loss": 0.87669015,
      "learning_rate": 3.624555968803217e-06,
      "loss": 0.89849466,
      "num_input_tokens_seen": 79656970,
      "step": 3697,
      "time_per_iteration": 2.554702043533325
    },
    {
      "auxiliary_loss_clip": 0.01117085,
      "auxiliary_loss_mlp": 0.01044974,
      "balance_loss_clip": 1.05005264,
      "balance_loss_mlp": 1.02879679,
      "epoch": 0.22233578836615062,
      "flos": 39203678833920.0,
      "grad_norm": 1.468307383163174,
      "language_loss": 0.65969241,
      "learning_rate": 3.624328776493346e-06,
      "loss": 0.68131304,
      "num_input_tokens_seen": 79680275,
      "step": 3698,
      "time_per_iteration": 2.7362568378448486
    },
    {
      "auxiliary_loss_clip": 0.01134301,
      "auxiliary_loss_mlp": 0.01041843,
      "balance_loss_clip": 1.05388296,
      "balance_loss_mlp": 1.02392554,
      "epoch": 0.22239591161881858,
      "flos": 36283782251520.0,
      "grad_norm": 1.9497695506261954,
      "language_loss": 0.8225503,
      "learning_rate": 3.6241015225887637e-06,
      "loss": 0.84431171,
      "num_input_tokens_seen": 79701255,
      "step": 3699,
      "time_per_iteration": 2.6239187717437744
    },
    {
      "auxiliary_loss_clip": 0.01122347,
      "auxiliary_loss_mlp": 0.0104384,
      "balance_loss_clip": 1.04951191,
      "balance_loss_mlp": 1.02591109,
      "epoch": 0.22245603487148655,
      "flos": 19719591177600.0,
      "grad_norm": 1.4906034976712705,
      "language_loss": 0.79643065,
      "learning_rate": 3.62387420709809e-06,
      "loss": 0.81809253,
      "num_input_tokens_seen": 79721315,
      "step": 3700,
      "time_per_iteration": 2.548328161239624
    },
    {
      "auxiliary_loss_clip": 0.01110668,
      "auxiliary_loss_mlp": 0.01043715,
      "balance_loss_clip": 1.05244303,
      "balance_loss_mlp": 1.02575028,
      "epoch": 0.2225161581241545,
      "flos": 46280450615040.0,
      "grad_norm": 2.1701972428060765,
      "language_loss": 0.71958733,
      "learning_rate": 3.623646830029943e-06,
      "loss": 0.74113107,
      "num_input_tokens_seen": 79742705,
      "step": 3701,
      "time_per_iteration": 4.185058355331421
    },
    {
      "auxiliary_loss_clip": 0.01138335,
      "auxiliary_loss_mlp": 0.01042571,
      "balance_loss_clip": 1.05032611,
      "balance_loss_mlp": 1.02567923,
      "epoch": 0.22257628137682248,
      "flos": 23696194993920.0,
      "grad_norm": 1.8292653964622356,
      "language_loss": 0.80229568,
      "learning_rate": 3.6234193913929454e-06,
      "loss": 0.82410473,
      "num_input_tokens_seen": 79763000,
      "step": 3702,
      "time_per_iteration": 3.9101614952087402
    },
    {
      "auxiliary_loss_clip": 0.01125835,
      "auxiliary_loss_mlp": 0.01042105,
      "balance_loss_clip": 1.04734778,
      "balance_loss_mlp": 1.02462888,
      "epoch": 0.22263640462949044,
      "flos": 19353984595200.0,
      "grad_norm": 1.7873417466984012,
      "language_loss": 0.77789557,
      "learning_rate": 3.623191891195723e-06,
      "loss": 0.79957497,
      "num_input_tokens_seen": 79781335,
      "step": 3703,
      "time_per_iteration": 2.5154995918273926
    },
    {
      "auxiliary_loss_clip": 0.01137204,
      "auxiliary_loss_mlp": 0.01041503,
      "balance_loss_clip": 1.05200076,
      "balance_loss_mlp": 1.02293038,
      "epoch": 0.22269652788215843,
      "flos": 20776047016320.0,
      "grad_norm": 2.040490302330314,
      "language_loss": 0.74058306,
      "learning_rate": 3.6229643294469005e-06,
      "loss": 0.76237017,
      "num_input_tokens_seen": 79800150,
      "step": 3704,
      "time_per_iteration": 2.4972083568573
    },
    {
      "auxiliary_loss_clip": 0.01101439,
      "auxiliary_loss_mlp": 0.01045571,
      "balance_loss_clip": 1.05286741,
      "balance_loss_mlp": 1.02985883,
      "epoch": 0.2227566511348264,
      "flos": 47958843467520.0,
      "grad_norm": 1.9054870683182117,
      "language_loss": 0.64360976,
      "learning_rate": 3.6227367061551074e-06,
      "loss": 0.66507983,
      "num_input_tokens_seen": 79822390,
      "step": 3705,
      "time_per_iteration": 2.795229196548462
    },
    {
      "auxiliary_loss_clip": 0.01039548,
      "auxiliary_loss_mlp": 0.01005406,
      "balance_loss_clip": 1.051301,
      "balance_loss_mlp": 1.0027951,
      "epoch": 0.22281677438749437,
      "flos": 66218953230720.0,
      "grad_norm": 1.2214577721489877,
      "language_loss": 0.6515553,
      "learning_rate": 3.6225090213289766e-06,
      "loss": 0.67200482,
      "num_input_tokens_seen": 79873350,
      "step": 3706,
      "time_per_iteration": 4.421535968780518
    },
    {
      "auxiliary_loss_clip": 0.01112825,
      "auxiliary_loss_mlp": 0.01042955,
      "balance_loss_clip": 1.05370426,
      "balance_loss_mlp": 1.02633715,
      "epoch": 0.22287689764016233,
      "flos": 21871609787520.0,
      "grad_norm": 2.002099612311545,
      "language_loss": 0.80582047,
      "learning_rate": 3.622281274977141e-06,
      "loss": 0.82737827,
      "num_input_tokens_seen": 79891715,
      "step": 3707,
      "time_per_iteration": 3.9179625511169434
    },
    {
      "auxiliary_loss_clip": 0.01147586,
      "auxiliary_loss_mlp": 0.01037651,
      "balance_loss_clip": 1.05311418,
      "balance_loss_mlp": 1.02096164,
      "epoch": 0.2229370208928303,
      "flos": 27672475587840.0,
      "grad_norm": 2.2031729429852986,
      "language_loss": 0.78673404,
      "learning_rate": 3.6220534671082367e-06,
      "loss": 0.80858642,
      "num_input_tokens_seen": 79911175,
      "step": 3708,
      "time_per_iteration": 2.5233566761016846
    },
    {
      "auxiliary_loss_clip": 0.0112594,
      "auxiliary_loss_mlp": 0.01041713,
      "balance_loss_clip": 1.05324852,
      "balance_loss_mlp": 1.02409399,
      "epoch": 0.22299714414549826,
      "flos": 30154657034880.0,
      "grad_norm": 1.9819335441606962,
      "language_loss": 0.80459082,
      "learning_rate": 3.6218255977309024e-06,
      "loss": 0.82626742,
      "num_input_tokens_seen": 79931875,
      "step": 3709,
      "time_per_iteration": 2.629119873046875
    },
    {
      "auxiliary_loss_clip": 0.01134206,
      "auxiliary_loss_mlp": 0.0079543,
      "balance_loss_clip": 1.0529294,
      "balance_loss_mlp": 1.01735461,
      "epoch": 0.22305726739816625,
      "flos": 23143134309120.0,
      "grad_norm": 1.7690933724008897,
      "language_loss": 0.68834007,
      "learning_rate": 3.6215976668537787e-06,
      "loss": 0.70763642,
      "num_input_tokens_seen": 79952445,
      "step": 3710,
      "time_per_iteration": 2.558711051940918
    },
    {
      "auxiliary_loss_clip": 0.01112245,
      "auxiliary_loss_mlp": 0.01049539,
      "balance_loss_clip": 1.05570114,
      "balance_loss_mlp": 1.03184795,
      "epoch": 0.22311739065083422,
      "flos": 19172061187200.0,
      "grad_norm": 2.199210471652172,
      "language_loss": 0.91312939,
      "learning_rate": 3.6213696744855096e-06,
      "loss": 0.93474722,
      "num_input_tokens_seen": 79971030,
      "step": 3711,
      "time_per_iteration": 2.5677599906921387
    },
    {
      "auxiliary_loss_clip": 0.01115644,
      "auxiliary_loss_mlp": 0.01052688,
      "balance_loss_clip": 1.05211163,
      "balance_loss_mlp": 1.03456831,
      "epoch": 0.22317751390350218,
      "flos": 13617757319040.0,
      "grad_norm": 2.949339642494492,
      "language_loss": 0.89604026,
      "learning_rate": 3.6211416206347395e-06,
      "loss": 0.9177236,
      "num_input_tokens_seen": 79982085,
      "step": 3712,
      "time_per_iteration": 2.4656765460968018
    },
    {
      "auxiliary_loss_clip": 0.01148422,
      "auxiliary_loss_mlp": 0.01048396,
      "balance_loss_clip": 1.05564618,
      "balance_loss_mlp": 1.0310744,
      "epoch": 0.22323763715617015,
      "flos": 11029065068160.0,
      "grad_norm": 2.7479700435099805,
      "language_loss": 0.75716138,
      "learning_rate": 3.620913505310117e-06,
      "loss": 0.77912951,
      "num_input_tokens_seen": 79997460,
      "step": 3713,
      "time_per_iteration": 2.4481799602508545
    },
    {
      "auxiliary_loss_clip": 0.0110451,
      "auxiliary_loss_mlp": 0.01045457,
      "balance_loss_clip": 1.05525827,
      "balance_loss_mlp": 1.02851689,
      "epoch": 0.22329776040883811,
      "flos": 41351531466240.0,
      "grad_norm": 3.099933834760393,
      "language_loss": 0.62832606,
      "learning_rate": 3.6206853285202917e-06,
      "loss": 0.64982569,
      "num_input_tokens_seen": 80022450,
      "step": 3714,
      "time_per_iteration": 2.8249099254608154
    },
    {
      "auxiliary_loss_clip": 0.0112088,
      "auxiliary_loss_mlp": 0.01032365,
      "balance_loss_clip": 1.05601287,
      "balance_loss_mlp": 1.01599741,
      "epoch": 0.22335788366150608,
      "flos": 25119478477440.0,
      "grad_norm": 1.7796132430405904,
      "language_loss": 0.7952776,
      "learning_rate": 3.6204570902739164e-06,
      "loss": 0.81681007,
      "num_input_tokens_seen": 80042100,
      "step": 3715,
      "time_per_iteration": 2.6272099018096924
    },
    {
      "auxiliary_loss_clip": 0.01107979,
      "auxiliary_loss_mlp": 0.01052724,
      "balance_loss_clip": 1.05758274,
      "balance_loss_mlp": 1.03586733,
      "epoch": 0.22341800691417404,
      "flos": 16983377769600.0,
      "grad_norm": 1.92139650124282,
      "language_loss": 0.77189517,
      "learning_rate": 3.620228790579645e-06,
      "loss": 0.79350215,
      "num_input_tokens_seen": 80059690,
      "step": 3716,
      "time_per_iteration": 2.566612482070923
    },
    {
      "auxiliary_loss_clip": 0.01121539,
      "auxiliary_loss_mlp": 0.01047567,
      "balance_loss_clip": 1.0515188,
      "balance_loss_mlp": 1.03056788,
      "epoch": 0.22347813016684204,
      "flos": 14136738975360.0,
      "grad_norm": 2.44785926085456,
      "language_loss": 0.7860806,
      "learning_rate": 3.6200004294461367e-06,
      "loss": 0.80777168,
      "num_input_tokens_seen": 80076060,
      "step": 3717,
      "time_per_iteration": 2.5313844680786133
    },
    {
      "auxiliary_loss_clip": 0.01079823,
      "auxiliary_loss_mlp": 0.01045187,
      "balance_loss_clip": 1.05552602,
      "balance_loss_mlp": 1.02686501,
      "epoch": 0.22353825341951,
      "flos": 23583147914880.0,
      "grad_norm": 2.476336774018428,
      "language_loss": 0.6784547,
      "learning_rate": 3.6197720068820497e-06,
      "loss": 0.69970489,
      "num_input_tokens_seen": 80094760,
      "step": 3718,
      "time_per_iteration": 2.6621789932250977
    },
    {
      "auxiliary_loss_clip": 0.01129563,
      "auxiliary_loss_mlp": 0.01041298,
      "balance_loss_clip": 1.05359626,
      "balance_loss_mlp": 1.02198601,
      "epoch": 0.22359837667217797,
      "flos": 29824206888960.0,
      "grad_norm": 2.55547153783583,
      "language_loss": 0.80822825,
      "learning_rate": 3.619543522896045e-06,
      "loss": 0.82993686,
      "num_input_tokens_seen": 80114475,
      "step": 3719,
      "time_per_iteration": 2.627722978591919
    },
    {
      "auxiliary_loss_clip": 0.01123064,
      "auxiliary_loss_mlp": 0.0105337,
      "balance_loss_clip": 1.05091822,
      "balance_loss_mlp": 1.03435588,
      "epoch": 0.22365849992484593,
      "flos": 17603088140160.0,
      "grad_norm": 2.1903534207644846,
      "language_loss": 0.8648268,
      "learning_rate": 3.6193149774967885e-06,
      "loss": 0.88659114,
      "num_input_tokens_seen": 80132920,
      "step": 3720,
      "time_per_iteration": 2.516124963760376
    },
    {
      "auxiliary_loss_clip": 0.0112258,
      "auxiliary_loss_mlp": 0.0104038,
      "balance_loss_clip": 1.0591023,
      "balance_loss_mlp": 1.02198577,
      "epoch": 0.2237186231775139,
      "flos": 22710949868160.0,
      "grad_norm": 1.8042786776808566,
      "language_loss": 0.74390185,
      "learning_rate": 3.619086370692945e-06,
      "loss": 0.76553142,
      "num_input_tokens_seen": 80152845,
      "step": 3721,
      "time_per_iteration": 2.5827584266662598
    },
    {
      "auxiliary_loss_clip": 0.01154226,
      "auxiliary_loss_mlp": 0.01048118,
      "balance_loss_clip": 1.05574286,
      "balance_loss_mlp": 1.03053451,
      "epoch": 0.22377874643018186,
      "flos": 13371518609280.0,
      "grad_norm": 2.050108648041648,
      "language_loss": 0.79282176,
      "learning_rate": 3.6188577024931844e-06,
      "loss": 0.8148452,
      "num_input_tokens_seen": 80170680,
      "step": 3722,
      "time_per_iteration": 2.470303535461426
    },
    {
      "auxiliary_loss_clip": 0.01113211,
      "auxiliary_loss_mlp": 0.01039803,
      "balance_loss_clip": 1.05506945,
      "balance_loss_mlp": 1.02316117,
      "epoch": 0.22383886968284986,
      "flos": 17894970057600.0,
      "grad_norm": 2.1204642292089178,
      "language_loss": 0.820885,
      "learning_rate": 3.618628972906178e-06,
      "loss": 0.84241509,
      "num_input_tokens_seen": 80189030,
      "step": 3723,
      "time_per_iteration": 2.6186423301696777
    },
    {
      "auxiliary_loss_clip": 0.01154455,
      "auxiliary_loss_mlp": 0.01049802,
      "balance_loss_clip": 1.05710661,
      "balance_loss_mlp": 1.03201628,
      "epoch": 0.22389899293551782,
      "flos": 23879123982720.0,
      "grad_norm": 1.8990467955239643,
      "language_loss": 0.84467667,
      "learning_rate": 3.6184001819405984e-06,
      "loss": 0.86671919,
      "num_input_tokens_seen": 80208365,
      "step": 3724,
      "time_per_iteration": 2.5040199756622314
    },
    {
      "auxiliary_loss_clip": 0.01128582,
      "auxiliary_loss_mlp": 0.0103929,
      "balance_loss_clip": 1.05890274,
      "balance_loss_mlp": 1.02188528,
      "epoch": 0.2239591161881858,
      "flos": 27272430840960.0,
      "grad_norm": 2.0217039822730336,
      "language_loss": 0.79168761,
      "learning_rate": 3.618171329605121e-06,
      "loss": 0.81336629,
      "num_input_tokens_seen": 80228685,
      "step": 3725,
      "time_per_iteration": 2.599083662033081
    },
    {
      "auxiliary_loss_clip": 0.01093103,
      "auxiliary_loss_mlp": 0.01040976,
      "balance_loss_clip": 1.05671811,
      "balance_loss_mlp": 1.0236187,
      "epoch": 0.22401923944085375,
      "flos": 22236857233920.0,
      "grad_norm": 1.8041644331285458,
      "language_loss": 0.76951665,
      "learning_rate": 3.6179424159084254e-06,
      "loss": 0.79085743,
      "num_input_tokens_seen": 80247635,
      "step": 3726,
      "time_per_iteration": 2.653315782546997
    },
    {
      "auxiliary_loss_clip": 0.01153221,
      "auxiliary_loss_mlp": 0.0104902,
      "balance_loss_clip": 1.05950832,
      "balance_loss_mlp": 1.02943349,
      "epoch": 0.22407936269352172,
      "flos": 12053668521600.0,
      "grad_norm": 2.367939704489406,
      "language_loss": 0.7255193,
      "learning_rate": 3.6177134408591914e-06,
      "loss": 0.74754167,
      "num_input_tokens_seen": 80260045,
      "step": 3727,
      "time_per_iteration": 2.4843571186065674
    },
    {
      "auxiliary_loss_clip": 0.01156054,
      "auxiliary_loss_mlp": 0.0104632,
      "balance_loss_clip": 1.05639589,
      "balance_loss_mlp": 1.02661514,
      "epoch": 0.22413948594618968,
      "flos": 19353553632000.0,
      "grad_norm": 2.3823967530206778,
      "language_loss": 0.87001526,
      "learning_rate": 3.6174844044661013e-06,
      "loss": 0.89203906,
      "num_input_tokens_seen": 80277680,
      "step": 3728,
      "time_per_iteration": 2.4720895290374756
    },
    {
      "auxiliary_loss_clip": 0.01125501,
      "auxiliary_loss_mlp": 0.01054499,
      "balance_loss_clip": 1.05643523,
      "balance_loss_mlp": 1.03327978,
      "epoch": 0.22419960919885765,
      "flos": 24170000319360.0,
      "grad_norm": 2.154243072499331,
      "language_loss": 0.8043108,
      "learning_rate": 3.6172553067378406e-06,
      "loss": 0.82611078,
      "num_input_tokens_seen": 80294795,
      "step": 3729,
      "time_per_iteration": 2.5827672481536865
    },
    {
      "auxiliary_loss_clip": 0.01127683,
      "auxiliary_loss_mlp": 0.01045647,
      "balance_loss_clip": 1.05689728,
      "balance_loss_mlp": 1.02861166,
      "epoch": 0.22425973245152564,
      "flos": 27378977558400.0,
      "grad_norm": 2.873095235551979,
      "language_loss": 0.86536056,
      "learning_rate": 3.6170261476830964e-06,
      "loss": 0.88709378,
      "num_input_tokens_seen": 80315425,
      "step": 3730,
      "time_per_iteration": 2.607928514480591
    },
    {
      "auxiliary_loss_clip": 0.01128276,
      "auxiliary_loss_mlp": 0.00793415,
      "balance_loss_clip": 1.05574715,
      "balance_loss_mlp": 1.01317406,
      "epoch": 0.2243198557041936,
      "flos": 13735652734080.0,
      "grad_norm": 1.732173958391391,
      "language_loss": 0.73223889,
      "learning_rate": 3.616796927310559e-06,
      "loss": 0.75145578,
      "num_input_tokens_seen": 80333905,
      "step": 3731,
      "time_per_iteration": 2.5272226333618164
    },
    {
      "auxiliary_loss_clip": 0.01124866,
      "auxiliary_loss_mlp": 0.01041073,
      "balance_loss_clip": 1.05654836,
      "balance_loss_mlp": 1.02332306,
      "epoch": 0.22437997895686157,
      "flos": 19530700531200.0,
      "grad_norm": 1.6487240139844148,
      "language_loss": 0.75365263,
      "learning_rate": 3.6165676456289195e-06,
      "loss": 0.77531207,
      "num_input_tokens_seen": 80352165,
      "step": 3732,
      "time_per_iteration": 2.5582118034362793
    },
    {
      "auxiliary_loss_clip": 0.01156782,
      "auxiliary_loss_mlp": 0.01059765,
      "balance_loss_clip": 1.05967486,
      "balance_loss_mlp": 1.04205072,
      "epoch": 0.22444010220952954,
      "flos": 23696230907520.0,
      "grad_norm": 2.0987299000437676,
      "language_loss": 0.87827933,
      "learning_rate": 3.616338302646873e-06,
      "loss": 0.90044481,
      "num_input_tokens_seen": 80371305,
      "step": 3733,
      "time_per_iteration": 2.505507230758667
    },
    {
      "auxiliary_loss_clip": 0.01117646,
      "auxiliary_loss_mlp": 0.01047589,
      "balance_loss_clip": 1.05846262,
      "balance_loss_mlp": 1.02942109,
      "epoch": 0.2245002254621975,
      "flos": 22382905933440.0,
      "grad_norm": 2.0415789591078233,
      "language_loss": 0.84686708,
      "learning_rate": 3.6161088983731166e-06,
      "loss": 0.86851943,
      "num_input_tokens_seen": 80391020,
      "step": 3734,
      "time_per_iteration": 2.6194024085998535
    },
    {
      "auxiliary_loss_clip": 0.01133598,
      "auxiliary_loss_mlp": 0.01049437,
      "balance_loss_clip": 1.05867219,
      "balance_loss_mlp": 1.03131711,
      "epoch": 0.22456034871486547,
      "flos": 26942303917440.0,
      "grad_norm": 1.6212796510693572,
      "language_loss": 0.76850319,
      "learning_rate": 3.6158794328163482e-06,
      "loss": 0.79033351,
      "num_input_tokens_seen": 80411365,
      "step": 3735,
      "time_per_iteration": 2.6044516563415527
    },
    {
      "auxiliary_loss_clip": 0.01132194,
      "auxiliary_loss_mlp": 0.01048712,
      "balance_loss_clip": 1.05861056,
      "balance_loss_mlp": 1.0317719,
      "epoch": 0.22462047196753343,
      "flos": 28983538005120.0,
      "grad_norm": 1.7063430501289794,
      "language_loss": 0.84645194,
      "learning_rate": 3.6156499059852702e-06,
      "loss": 0.86826098,
      "num_input_tokens_seen": 80431075,
      "step": 3736,
      "time_per_iteration": 2.604153633117676
    },
    {
      "auxiliary_loss_clip": 0.01115852,
      "auxiliary_loss_mlp": 0.010427,
      "balance_loss_clip": 1.06077898,
      "balance_loss_mlp": 1.02512896,
      "epoch": 0.22468059522020142,
      "flos": 20011329440640.0,
      "grad_norm": 1.8256528208951255,
      "language_loss": 0.86427402,
      "learning_rate": 3.615420317888586e-06,
      "loss": 0.88585955,
      "num_input_tokens_seen": 80449240,
      "step": 3737,
      "time_per_iteration": 2.562758684158325
    },
    {
      "auxiliary_loss_clip": 0.01155111,
      "auxiliary_loss_mlp": 0.01051877,
      "balance_loss_clip": 1.05735016,
      "balance_loss_mlp": 1.03324461,
      "epoch": 0.2247407184728694,
      "flos": 29314239546240.0,
      "grad_norm": 1.8985445200545226,
      "language_loss": 0.79017413,
      "learning_rate": 3.6151906685350006e-06,
      "loss": 0.812244,
      "num_input_tokens_seen": 80467900,
      "step": 3738,
      "time_per_iteration": 2.5640101432800293
    },
    {
      "auxiliary_loss_clip": 0.01122355,
      "auxiliary_loss_mlp": 0.01045226,
      "balance_loss_clip": 1.05492139,
      "balance_loss_mlp": 1.02767837,
      "epoch": 0.22480084172553735,
      "flos": 22310366417280.0,
      "grad_norm": 1.6454489790048001,
      "language_loss": 0.76213765,
      "learning_rate": 3.614960957933224e-06,
      "loss": 0.78381348,
      "num_input_tokens_seen": 80487100,
      "step": 3739,
      "time_per_iteration": 3.9595401287078857
    },
    {
      "auxiliary_loss_clip": 0.01115621,
      "auxiliary_loss_mlp": 0.01048596,
      "balance_loss_clip": 1.05171108,
      "balance_loss_mlp": 1.02777028,
      "epoch": 0.22486096497820532,
      "flos": 25591272641280.0,
      "grad_norm": 2.731918035872734,
      "language_loss": 0.74390483,
      "learning_rate": 3.6147311860919655e-06,
      "loss": 0.76554704,
      "num_input_tokens_seen": 80508625,
      "step": 3740,
      "time_per_iteration": 2.614309310913086
    },
    {
      "auxiliary_loss_clip": 0.01152084,
      "auxiliary_loss_mlp": 0.01040033,
      "balance_loss_clip": 1.0576247,
      "balance_loss_mlp": 1.02271223,
      "epoch": 0.22492108823087328,
      "flos": 17639824775040.0,
      "grad_norm": 1.705388842671718,
      "language_loss": 0.75909168,
      "learning_rate": 3.614501353019939e-06,
      "loss": 0.78101289,
      "num_input_tokens_seen": 80527345,
      "step": 3741,
      "time_per_iteration": 3.8693768978118896
    },
    {
      "auxiliary_loss_clip": 0.01138823,
      "auxiliary_loss_mlp": 0.01040924,
      "balance_loss_clip": 1.06210303,
      "balance_loss_mlp": 1.02342367,
      "epoch": 0.22498121148354125,
      "flos": 16034653797120.0,
      "grad_norm": 1.762251101844035,
      "language_loss": 0.87479824,
      "learning_rate": 3.6142714587258592e-06,
      "loss": 0.89659572,
      "num_input_tokens_seen": 80545545,
      "step": 3742,
      "time_per_iteration": 2.5557687282562256
    },
    {
      "auxiliary_loss_clip": 0.01104291,
      "auxiliary_loss_mlp": 0.01045521,
      "balance_loss_clip": 1.05937445,
      "balance_loss_mlp": 1.02701962,
      "epoch": 0.22504133473620924,
      "flos": 24023772051840.0,
      "grad_norm": 1.6656173382157309,
      "language_loss": 0.81803781,
      "learning_rate": 3.614041503218444e-06,
      "loss": 0.83953595,
      "num_input_tokens_seen": 80565040,
      "step": 3743,
      "time_per_iteration": 2.6294338703155518
    },
    {
      "auxiliary_loss_clip": 0.011429,
      "auxiliary_loss_mlp": 0.01040597,
      "balance_loss_clip": 1.0567863,
      "balance_loss_mlp": 1.02353764,
      "epoch": 0.2251014579888772,
      "flos": 16763963541120.0,
      "grad_norm": 2.1803182555173977,
      "language_loss": 0.6400159,
      "learning_rate": 3.6138114865064134e-06,
      "loss": 0.66185087,
      "num_input_tokens_seen": 80582815,
      "step": 3744,
      "time_per_iteration": 2.491536855697632
    },
    {
      "auxiliary_loss_clip": 0.01134607,
      "auxiliary_loss_mlp": 0.01039958,
      "balance_loss_clip": 1.0559665,
      "balance_loss_mlp": 1.02249432,
      "epoch": 0.22516158124154517,
      "flos": 13991013498240.0,
      "grad_norm": 2.9263769970283593,
      "language_loss": 0.76143813,
      "learning_rate": 3.613581408598489e-06,
      "loss": 0.78318381,
      "num_input_tokens_seen": 80600865,
      "step": 3745,
      "time_per_iteration": 3.8623361587524414
    },
    {
      "auxiliary_loss_clip": 0.01114466,
      "auxiliary_loss_mlp": 0.01037907,
      "balance_loss_clip": 1.05591094,
      "balance_loss_mlp": 1.02112269,
      "epoch": 0.22522170449421314,
      "flos": 14390016750720.0,
      "grad_norm": 1.9839853955523465,
      "language_loss": 0.81164122,
      "learning_rate": 3.6133512695033965e-06,
      "loss": 0.83316493,
      "num_input_tokens_seen": 80617455,
      "step": 3746,
      "time_per_iteration": 3.975831985473633
    },
    {
      "auxiliary_loss_clip": 0.01139914,
      "auxiliary_loss_mlp": 0.01046998,
      "balance_loss_clip": 1.0541811,
      "balance_loss_mlp": 1.02934313,
      "epoch": 0.2252818277468811,
      "flos": 23805542972160.0,
      "grad_norm": 2.504937815084682,
      "language_loss": 0.86532533,
      "learning_rate": 3.613121069229862e-06,
      "loss": 0.88719451,
      "num_input_tokens_seen": 80635125,
      "step": 3747,
      "time_per_iteration": 2.5332112312316895
    },
    {
      "auxiliary_loss_clip": 0.01139014,
      "auxiliary_loss_mlp": 0.00795194,
      "balance_loss_clip": 1.05376601,
      "balance_loss_mlp": 1.01527572,
      "epoch": 0.22534195099954907,
      "flos": 24718033100160.0,
      "grad_norm": 1.6151594690233029,
      "language_loss": 0.76640296,
      "learning_rate": 3.6128908077866145e-06,
      "loss": 0.78574502,
      "num_input_tokens_seen": 80656370,
      "step": 3748,
      "time_per_iteration": 2.5779221057891846
    },
    {
      "auxiliary_loss_clip": 0.01155238,
      "auxiliary_loss_mlp": 0.01046584,
      "balance_loss_clip": 1.05752385,
      "balance_loss_mlp": 1.02865505,
      "epoch": 0.22540207425221703,
      "flos": 21032341534080.0,
      "grad_norm": 1.5768548681255479,
      "language_loss": 0.79707325,
      "learning_rate": 3.6126604851823864e-06,
      "loss": 0.8190915,
      "num_input_tokens_seen": 80676495,
      "step": 3749,
      "time_per_iteration": 2.4995920658111572
    },
    {
      "auxiliary_loss_clip": 0.01124719,
      "auxiliary_loss_mlp": 0.01042427,
      "balance_loss_clip": 1.05538118,
      "balance_loss_mlp": 1.02621508,
      "epoch": 0.22546219750488503,
      "flos": 19390362094080.0,
      "grad_norm": 1.6579203690508102,
      "language_loss": 0.79747832,
      "learning_rate": 3.6124301014259108e-06,
      "loss": 0.81914985,
      "num_input_tokens_seen": 80694755,
      "step": 3750,
      "time_per_iteration": 2.536231517791748
    },
    {
      "auxiliary_loss_clip": 0.01093638,
      "auxiliary_loss_mlp": 0.01048462,
      "balance_loss_clip": 1.0541209,
      "balance_loss_mlp": 1.03101015,
      "epoch": 0.225522320757553,
      "flos": 25192628524800.0,
      "grad_norm": 1.7315916059129992,
      "language_loss": 0.8184216,
      "learning_rate": 3.6121996565259244e-06,
      "loss": 0.83984256,
      "num_input_tokens_seen": 80713670,
      "step": 3751,
      "time_per_iteration": 2.625833511352539
    },
    {
      "auxiliary_loss_clip": 0.01117888,
      "auxiliary_loss_mlp": 0.01044173,
      "balance_loss_clip": 1.05664742,
      "balance_loss_mlp": 1.02673233,
      "epoch": 0.22558244401022096,
      "flos": 17163110448000.0,
      "grad_norm": 2.1538779973864095,
      "language_loss": 0.83622253,
      "learning_rate": 3.611969150491165e-06,
      "loss": 0.85784316,
      "num_input_tokens_seen": 80731450,
      "step": 3752,
      "time_per_iteration": 2.5240583419799805
    },
    {
      "auxiliary_loss_clip": 0.01147899,
      "auxiliary_loss_mlp": 0.01036912,
      "balance_loss_clip": 1.05460346,
      "balance_loss_mlp": 1.02103317,
      "epoch": 0.22564256726288892,
      "flos": 15231008856960.0,
      "grad_norm": 1.8891383814068339,
      "language_loss": 0.78549361,
      "learning_rate": 3.611738583330375e-06,
      "loss": 0.80734175,
      "num_input_tokens_seen": 80748415,
      "step": 3753,
      "time_per_iteration": 2.4651238918304443
    },
    {
      "auxiliary_loss_clip": 0.01126926,
      "auxiliary_loss_mlp": 0.01040799,
      "balance_loss_clip": 1.05602217,
      "balance_loss_mlp": 1.02325141,
      "epoch": 0.2257026905155569,
      "flos": 34568652764160.0,
      "grad_norm": 2.0229061658016514,
      "language_loss": 0.78974921,
      "learning_rate": 3.611507955052295e-06,
      "loss": 0.81142646,
      "num_input_tokens_seen": 80770835,
      "step": 3754,
      "time_per_iteration": 2.6487679481506348
    },
    {
      "auxiliary_loss_clip": 0.01129785,
      "auxiliary_loss_mlp": 0.01047974,
      "balance_loss_clip": 1.05975795,
      "balance_loss_mlp": 1.03025913,
      "epoch": 0.22576281376822485,
      "flos": 19938430788480.0,
      "grad_norm": 1.748644262666529,
      "language_loss": 0.70582604,
      "learning_rate": 3.6112772656656727e-06,
      "loss": 0.72760355,
      "num_input_tokens_seen": 80787840,
      "step": 3755,
      "time_per_iteration": 2.5304758548736572
    },
    {
      "auxiliary_loss_clip": 0.01123953,
      "auxiliary_loss_mlp": 0.01054906,
      "balance_loss_clip": 1.05806553,
      "balance_loss_mlp": 1.03770423,
      "epoch": 0.22582293702089282,
      "flos": 24602005192320.0,
      "grad_norm": 1.9980308594808178,
      "language_loss": 0.77365804,
      "learning_rate": 3.6110465151792547e-06,
      "loss": 0.79544663,
      "num_input_tokens_seen": 80806335,
      "step": 3756,
      "time_per_iteration": 2.602304220199585
    },
    {
      "auxiliary_loss_clip": 0.01135464,
      "auxiliary_loss_mlp": 0.01047441,
      "balance_loss_clip": 1.06435215,
      "balance_loss_mlp": 1.03006041,
      "epoch": 0.2258830602735608,
      "flos": 23035438356480.0,
      "grad_norm": 1.9232337083973643,
      "language_loss": 0.82484859,
      "learning_rate": 3.6108157036017916e-06,
      "loss": 0.84667766,
      "num_input_tokens_seen": 80825355,
      "step": 3757,
      "time_per_iteration": 2.549470901489258
    },
    {
      "auxiliary_loss_clip": 0.01137561,
      "auxiliary_loss_mlp": 0.01043829,
      "balance_loss_clip": 1.05552542,
      "balance_loss_mlp": 1.02522063,
      "epoch": 0.22594318352622877,
      "flos": 22158427887360.0,
      "grad_norm": 2.143962827059905,
      "language_loss": 0.73425412,
      "learning_rate": 3.6105848309420358e-06,
      "loss": 0.75606805,
      "num_input_tokens_seen": 80842570,
      "step": 3758,
      "time_per_iteration": 2.5137217044830322
    },
    {
      "auxiliary_loss_clip": 0.01133726,
      "auxiliary_loss_mlp": 0.01050141,
      "balance_loss_clip": 1.05688894,
      "balance_loss_mlp": 1.03198552,
      "epoch": 0.22600330677889674,
      "flos": 20594303176320.0,
      "grad_norm": 2.0577999837570955,
      "language_loss": 0.77360106,
      "learning_rate": 3.6103538972087412e-06,
      "loss": 0.79543972,
      "num_input_tokens_seen": 80858745,
      "step": 3759,
      "time_per_iteration": 2.553312063217163
    },
    {
      "auxiliary_loss_clip": 0.01108294,
      "auxiliary_loss_mlp": 0.01042849,
      "balance_loss_clip": 1.0554961,
      "balance_loss_mlp": 1.02536106,
      "epoch": 0.2260634300315647,
      "flos": 35659798162560.0,
      "grad_norm": 1.5948788167608698,
      "language_loss": 0.78859085,
      "learning_rate": 3.6101229024106655e-06,
      "loss": 0.81010228,
      "num_input_tokens_seen": 80880085,
      "step": 3760,
      "time_per_iteration": 2.738917112350464
    },
    {
      "auxiliary_loss_clip": 0.01036548,
      "auxiliary_loss_mlp": 0.01028111,
      "balance_loss_clip": 1.03145194,
      "balance_loss_mlp": 1.02503574,
      "epoch": 0.22612355328423267,
      "flos": 72090455126400.0,
      "grad_norm": 0.9526893316136346,
      "language_loss": 0.60113287,
      "learning_rate": 3.609891846556569e-06,
      "loss": 0.6217795,
      "num_input_tokens_seen": 80937660,
      "step": 3761,
      "time_per_iteration": 3.125507354736328
    },
    {
      "auxiliary_loss_clip": 0.01115378,
      "auxiliary_loss_mlp": 0.01045978,
      "balance_loss_clip": 1.05494428,
      "balance_loss_mlp": 1.02785814,
      "epoch": 0.22618367653690064,
      "flos": 22783776693120.0,
      "grad_norm": 1.9898634493671348,
      "language_loss": 0.7740075,
      "learning_rate": 3.609660729655211e-06,
      "loss": 0.7956211,
      "num_input_tokens_seen": 80956265,
      "step": 3762,
      "time_per_iteration": 2.5678603649139404
    },
    {
      "auxiliary_loss_clip": 0.01128585,
      "auxiliary_loss_mlp": 0.01046592,
      "balance_loss_clip": 1.05545199,
      "balance_loss_mlp": 1.02662432,
      "epoch": 0.22624379978956863,
      "flos": 20448254476800.0,
      "grad_norm": 1.8136800334848775,
      "language_loss": 0.78906608,
      "learning_rate": 3.6094295517153573e-06,
      "loss": 0.81081784,
      "num_input_tokens_seen": 80975185,
      "step": 3763,
      "time_per_iteration": 2.555022716522217
    },
    {
      "auxiliary_loss_clip": 0.01136289,
      "auxiliary_loss_mlp": 0.01048194,
      "balance_loss_clip": 1.05460322,
      "balance_loss_mlp": 1.02995539,
      "epoch": 0.2263039230422366,
      "flos": 17494314779520.0,
      "grad_norm": 1.6216548913829765,
      "language_loss": 0.9167918,
      "learning_rate": 3.6091983127457743e-06,
      "loss": 0.93863666,
      "num_input_tokens_seen": 80992830,
      "step": 3764,
      "time_per_iteration": 2.480250835418701
    },
    {
      "auxiliary_loss_clip": 0.0113231,
      "auxiliary_loss_mlp": 0.01056994,
      "balance_loss_clip": 1.0565393,
      "balance_loss_mlp": 1.03882706,
      "epoch": 0.22636404629490456,
      "flos": 28329748606080.0,
      "grad_norm": 1.9123978371485788,
      "language_loss": 0.75482583,
      "learning_rate": 3.6089670127552293e-06,
      "loss": 0.7767188,
      "num_input_tokens_seen": 81013675,
      "step": 3765,
      "time_per_iteration": 2.5780951976776123
    },
    {
      "auxiliary_loss_clip": 0.01139471,
      "auxiliary_loss_mlp": 0.01047511,
      "balance_loss_clip": 1.05705738,
      "balance_loss_mlp": 1.03052366,
      "epoch": 0.22642416954757252,
      "flos": 17489143221120.0,
      "grad_norm": 2.0299482261223263,
      "language_loss": 0.89814973,
      "learning_rate": 3.608735651752494e-06,
      "loss": 0.92001951,
      "num_input_tokens_seen": 81030345,
      "step": 3766,
      "time_per_iteration": 2.487154006958008
    },
    {
      "auxiliary_loss_clip": 0.01123454,
      "auxiliary_loss_mlp": 0.01041033,
      "balance_loss_clip": 1.05778956,
      "balance_loss_mlp": 1.02401042,
      "epoch": 0.2264842928002405,
      "flos": 24384530298240.0,
      "grad_norm": 1.6620599595742342,
      "language_loss": 0.74797285,
      "learning_rate": 3.6085042297463417e-06,
      "loss": 0.76961774,
      "num_input_tokens_seen": 81051000,
      "step": 3767,
      "time_per_iteration": 2.6078543663024902
    },
    {
      "auxiliary_loss_clip": 0.01139284,
      "auxiliary_loss_mlp": 0.01043375,
      "balance_loss_clip": 1.05449152,
      "balance_loss_mlp": 1.02509987,
      "epoch": 0.22654441605290845,
      "flos": 19830519354240.0,
      "grad_norm": 1.499791435453702,
      "language_loss": 0.71517402,
      "learning_rate": 3.6082727467455477e-06,
      "loss": 0.73700058,
      "num_input_tokens_seen": 81071205,
      "step": 3768,
      "time_per_iteration": 2.5094616413116455
    },
    {
      "auxiliary_loss_clip": 0.01144258,
      "auxiliary_loss_mlp": 0.010608,
      "balance_loss_clip": 1.06025076,
      "balance_loss_mlp": 1.04262006,
      "epoch": 0.22660453930557642,
      "flos": 27454569730560.0,
      "grad_norm": 1.7445075945653372,
      "language_loss": 0.78516221,
      "learning_rate": 3.6080412027588905e-06,
      "loss": 0.80721271,
      "num_input_tokens_seen": 81091880,
      "step": 3769,
      "time_per_iteration": 2.5666685104370117
    },
    {
      "auxiliary_loss_clip": 0.01122675,
      "auxiliary_loss_mlp": 0.01042542,
      "balance_loss_clip": 1.04986739,
      "balance_loss_mlp": 1.02429152,
      "epoch": 0.2266646625582444,
      "flos": 23988148738560.0,
      "grad_norm": 3.047596141797272,
      "language_loss": 0.68240666,
      "learning_rate": 3.6078095977951488e-06,
      "loss": 0.70405889,
      "num_input_tokens_seen": 81113290,
      "step": 3770,
      "time_per_iteration": 2.591033697128296
    },
    {
      "auxiliary_loss_clip": 0.01151815,
      "auxiliary_loss_mlp": 0.01041108,
      "balance_loss_clip": 1.05489385,
      "balance_loss_mlp": 1.02372742,
      "epoch": 0.22672478581091238,
      "flos": 26028054023040.0,
      "grad_norm": 1.5677721500753763,
      "language_loss": 0.80212295,
      "learning_rate": 3.6075779318631067e-06,
      "loss": 0.82405221,
      "num_input_tokens_seen": 81133535,
      "step": 3771,
      "time_per_iteration": 2.5482959747314453
    },
    {
      "auxiliary_loss_clip": 0.01116447,
      "auxiliary_loss_mlp": 0.01053853,
      "balance_loss_clip": 1.05637538,
      "balance_loss_mlp": 1.03626966,
      "epoch": 0.22678490906358034,
      "flos": 23841812730240.0,
      "grad_norm": 1.5389980514277168,
      "language_loss": 0.79163587,
      "learning_rate": 3.6073462049715486e-06,
      "loss": 0.81333888,
      "num_input_tokens_seen": 81154650,
      "step": 3772,
      "time_per_iteration": 2.589308738708496
    },
    {
      "auxiliary_loss_clip": 0.01026231,
      "auxiliary_loss_mlp": 0.01005537,
      "balance_loss_clip": 1.0462656,
      "balance_loss_mlp": 1.00321257,
      "epoch": 0.2268450323162483,
      "flos": 65048088574080.0,
      "grad_norm": 0.6504675534785376,
      "language_loss": 0.5434041,
      "learning_rate": 3.607114417129261e-06,
      "loss": 0.56372178,
      "num_input_tokens_seen": 81221240,
      "step": 3773,
      "time_per_iteration": 3.3104448318481445
    },
    {
      "auxiliary_loss_clip": 0.01123458,
      "auxiliary_loss_mlp": 0.01037692,
      "balance_loss_clip": 1.0590663,
      "balance_loss_mlp": 1.02059782,
      "epoch": 0.22690515556891627,
      "flos": 22526081544960.0,
      "grad_norm": 1.976611121060325,
      "language_loss": 0.70752025,
      "learning_rate": 3.6068825683450334e-06,
      "loss": 0.72913176,
      "num_input_tokens_seen": 81241520,
      "step": 3774,
      "time_per_iteration": 2.5703842639923096
    },
    {
      "auxiliary_loss_clip": 0.011201,
      "auxiliary_loss_mlp": 0.01043341,
      "balance_loss_clip": 1.05472612,
      "balance_loss_mlp": 1.02590048,
      "epoch": 0.22696527882158424,
      "flos": 18223444955520.0,
      "grad_norm": 2.6200067706996495,
      "language_loss": 0.75023556,
      "learning_rate": 3.606650658627658e-06,
      "loss": 0.77187002,
      "num_input_tokens_seen": 81256825,
      "step": 3775,
      "time_per_iteration": 2.5320749282836914
    },
    {
      "auxiliary_loss_clip": 0.01150662,
      "auxiliary_loss_mlp": 0.01044462,
      "balance_loss_clip": 1.05545807,
      "balance_loss_mlp": 1.02770162,
      "epoch": 0.22702540207425223,
      "flos": 17019252478080.0,
      "grad_norm": 2.0404101651530966,
      "language_loss": 0.81924534,
      "learning_rate": 3.606418687985928e-06,
      "loss": 0.8411966,
      "num_input_tokens_seen": 81275695,
      "step": 3776,
      "time_per_iteration": 2.4616827964782715
    },
    {
      "auxiliary_loss_clip": 0.01128242,
      "auxiliary_loss_mlp": 0.01039943,
      "balance_loss_clip": 1.05182219,
      "balance_loss_mlp": 1.02306354,
      "epoch": 0.2270855253269202,
      "flos": 21325731822720.0,
      "grad_norm": 1.743916181644247,
      "language_loss": 0.82684362,
      "learning_rate": 3.606186656428641e-06,
      "loss": 0.84852546,
      "num_input_tokens_seen": 81294920,
      "step": 3777,
      "time_per_iteration": 2.5795741081237793
    },
    {
      "auxiliary_loss_clip": 0.0111951,
      "auxiliary_loss_mlp": 0.0104024,
      "balance_loss_clip": 1.05260158,
      "balance_loss_mlp": 1.02366972,
      "epoch": 0.22714564857958816,
      "flos": 23550469516800.0,
      "grad_norm": 2.307429378772807,
      "language_loss": 0.72486413,
      "learning_rate": 3.6059545639645955e-06,
      "loss": 0.74646163,
      "num_input_tokens_seen": 81314275,
      "step": 3778,
      "time_per_iteration": 3.912975788116455
    },
    {
      "auxiliary_loss_clip": 0.01106393,
      "auxiliary_loss_mlp": 0.01035319,
      "balance_loss_clip": 1.05404961,
      "balance_loss_mlp": 1.0185461,
      "epoch": 0.22720577183225613,
      "flos": 25989880844160.0,
      "grad_norm": 1.9243797342374056,
      "language_loss": 0.63862342,
      "learning_rate": 3.605722410602591e-06,
      "loss": 0.6600405,
      "num_input_tokens_seen": 81333890,
      "step": 3779,
      "time_per_iteration": 2.6188840866088867
    },
    {
      "auxiliary_loss_clip": 0.01130198,
      "auxiliary_loss_mlp": 0.01048303,
      "balance_loss_clip": 1.0551579,
      "balance_loss_mlp": 1.03074396,
      "epoch": 0.2272658950849241,
      "flos": 20814076540800.0,
      "grad_norm": 1.7425441176422654,
      "language_loss": 0.70515758,
      "learning_rate": 3.6054901963514323e-06,
      "loss": 0.72694254,
      "num_input_tokens_seen": 81353640,
      "step": 3780,
      "time_per_iteration": 3.9290170669555664
    },
    {
      "auxiliary_loss_clip": 0.01140331,
      "auxiliary_loss_mlp": 0.01044908,
      "balance_loss_clip": 1.05665541,
      "balance_loss_mlp": 1.02715755,
      "epoch": 0.22732601833759206,
      "flos": 23909324342400.0,
      "grad_norm": 1.6627238788805716,
      "language_loss": 0.89534253,
      "learning_rate": 3.6052579212199246e-06,
      "loss": 0.91719496,
      "num_input_tokens_seen": 81371595,
      "step": 3781,
      "time_per_iteration": 2.5489003658294678
    },
    {
      "auxiliary_loss_clip": 0.01149515,
      "auxiliary_loss_mlp": 0.01044791,
      "balance_loss_clip": 1.053177,
      "balance_loss_mlp": 1.02638555,
      "epoch": 0.22738614159026002,
      "flos": 15924407978880.0,
      "grad_norm": 2.0101787472979775,
      "language_loss": 0.74393141,
      "learning_rate": 3.6050255852168753e-06,
      "loss": 0.76587445,
      "num_input_tokens_seen": 81388435,
      "step": 3782,
      "time_per_iteration": 2.4542014598846436
    },
    {
      "auxiliary_loss_clip": 0.01124379,
      "auxiliary_loss_mlp": 0.01043988,
      "balance_loss_clip": 1.05123305,
      "balance_loss_mlp": 1.02804935,
      "epoch": 0.22744626484292801,
      "flos": 24205515891840.0,
      "grad_norm": 1.7115388504181877,
      "language_loss": 0.82752621,
      "learning_rate": 3.604793188351095e-06,
      "loss": 0.8492099,
      "num_input_tokens_seen": 81410195,
      "step": 3783,
      "time_per_iteration": 2.588388442993164
    },
    {
      "auxiliary_loss_clip": 0.01124848,
      "auxiliary_loss_mlp": 0.01044816,
      "balance_loss_clip": 1.05692506,
      "balance_loss_mlp": 1.02766192,
      "epoch": 0.22750638809559598,
      "flos": 24791614110720.0,
      "grad_norm": 1.7221119712286108,
      "language_loss": 0.75809181,
      "learning_rate": 3.6045607306313964e-06,
      "loss": 0.77978849,
      "num_input_tokens_seen": 81430060,
      "step": 3784,
      "time_per_iteration": 3.943814516067505
    },
    {
      "auxiliary_loss_clip": 0.01141807,
      "auxiliary_loss_mlp": 0.01041504,
      "balance_loss_clip": 1.04939377,
      "balance_loss_mlp": 1.02432573,
      "epoch": 0.22756651134826394,
      "flos": 22236498097920.0,
      "grad_norm": 1.781926831492604,
      "language_loss": 0.70745474,
      "learning_rate": 3.604328212066594e-06,
      "loss": 0.72928786,
      "num_input_tokens_seen": 81447375,
      "step": 3785,
      "time_per_iteration": 3.862828254699707
    },
    {
      "auxiliary_loss_clip": 0.01039577,
      "auxiliary_loss_mlp": 0.01005126,
      "balance_loss_clip": 1.03658938,
      "balance_loss_mlp": 1.00245571,
      "epoch": 0.2276266346009319,
      "flos": 62707466626560.0,
      "grad_norm": 0.8211894122112678,
      "language_loss": 0.61853504,
      "learning_rate": 3.6040956326655047e-06,
      "loss": 0.63898206,
      "num_input_tokens_seen": 81505235,
      "step": 3786,
      "time_per_iteration": 3.1266160011291504
    },
    {
      "auxiliary_loss_clip": 0.01129208,
      "auxiliary_loss_mlp": 0.01042675,
      "balance_loss_clip": 1.05298257,
      "balance_loss_mlp": 1.02484155,
      "epoch": 0.22768675785359987,
      "flos": 18613936684800.0,
      "grad_norm": 3.0265495398563766,
      "language_loss": 0.86368495,
      "learning_rate": 3.6038629924369486e-06,
      "loss": 0.88540375,
      "num_input_tokens_seen": 81518685,
      "step": 3787,
      "time_per_iteration": 2.520141124725342
    },
    {
      "auxiliary_loss_clip": 0.01125065,
      "auxiliary_loss_mlp": 0.01040742,
      "balance_loss_clip": 1.05323696,
      "balance_loss_mlp": 1.0246253,
      "epoch": 0.22774688110626784,
      "flos": 26870195364480.0,
      "grad_norm": 1.261092090482919,
      "language_loss": 0.72619998,
      "learning_rate": 3.6036302913897474e-06,
      "loss": 0.74785805,
      "num_input_tokens_seen": 81538940,
      "step": 3788,
      "time_per_iteration": 2.591799736022949
    },
    {
      "auxiliary_loss_clip": 0.01123089,
      "auxiliary_loss_mlp": 0.01033214,
      "balance_loss_clip": 1.05295992,
      "balance_loss_mlp": 1.0164535,
      "epoch": 0.2278070043589358,
      "flos": 15553593924480.0,
      "grad_norm": 2.7498066098273535,
      "language_loss": 0.67842734,
      "learning_rate": 3.6033975295327243e-06,
      "loss": 0.69999033,
      "num_input_tokens_seen": 81555525,
      "step": 3789,
      "time_per_iteration": 2.5040812492370605
    },
    {
      "auxiliary_loss_clip": 0.01114815,
      "auxiliary_loss_mlp": 0.01045765,
      "balance_loss_clip": 1.05153263,
      "balance_loss_mlp": 1.02850389,
      "epoch": 0.2278671276116038,
      "flos": 22416805393920.0,
      "grad_norm": 2.000027847611294,
      "language_loss": 0.75813323,
      "learning_rate": 3.6031647068747065e-06,
      "loss": 0.77973902,
      "num_input_tokens_seen": 81576305,
      "step": 3790,
      "time_per_iteration": 2.5927207469940186
    },
    {
      "auxiliary_loss_clip": 0.01087043,
      "auxiliary_loss_mlp": 0.01046705,
      "balance_loss_clip": 1.04702282,
      "balance_loss_mlp": 1.02871609,
      "epoch": 0.22792725086427176,
      "flos": 20631363033600.0,
      "grad_norm": 1.9269623444165649,
      "language_loss": 0.91096443,
      "learning_rate": 3.602931823424522e-06,
      "loss": 0.93230188,
      "num_input_tokens_seen": 81594115,
      "step": 3791,
      "time_per_iteration": 2.59020733833313
    },
    {
      "auxiliary_loss_clip": 0.01136919,
      "auxiliary_loss_mlp": 0.01037113,
      "balance_loss_clip": 1.0494771,
      "balance_loss_mlp": 1.01991153,
      "epoch": 0.22798737411693973,
      "flos": 31428946903680.0,
      "grad_norm": 1.681830033947346,
      "language_loss": 0.82667577,
      "learning_rate": 3.6026988791910026e-06,
      "loss": 0.84841609,
      "num_input_tokens_seen": 81615355,
      "step": 3792,
      "time_per_iteration": 2.6188700199127197
    },
    {
      "auxiliary_loss_clip": 0.01062697,
      "auxiliary_loss_mlp": 0.01006751,
      "balance_loss_clip": 1.02853334,
      "balance_loss_mlp": 1.00411665,
      "epoch": 0.2280474973696077,
      "flos": 52396685827200.0,
      "grad_norm": 1.1334899212718617,
      "language_loss": 0.65643871,
      "learning_rate": 3.602465874182981e-06,
      "loss": 0.6771332,
      "num_input_tokens_seen": 81662075,
      "step": 3793,
      "time_per_iteration": 2.8329532146453857
    },
    {
      "auxiliary_loss_clip": 0.01152335,
      "auxiliary_loss_mlp": 0.01050868,
      "balance_loss_clip": 1.05391026,
      "balance_loss_mlp": 1.03253388,
      "epoch": 0.22810762062227566,
      "flos": 26396066816640.0,
      "grad_norm": 3.139828160915369,
      "language_loss": 0.77072072,
      "learning_rate": 3.602232808409293e-06,
      "loss": 0.79275274,
      "num_input_tokens_seen": 81681625,
      "step": 3794,
      "time_per_iteration": 2.5295193195343018
    },
    {
      "auxiliary_loss_clip": 0.01107156,
      "auxiliary_loss_mlp": 0.01048466,
      "balance_loss_clip": 1.05119777,
      "balance_loss_mlp": 1.02934504,
      "epoch": 0.22816774387494362,
      "flos": 25630271832960.0,
      "grad_norm": 1.768792930368063,
      "language_loss": 0.81050599,
      "learning_rate": 3.6019996818787755e-06,
      "loss": 0.83206218,
      "num_input_tokens_seen": 81701170,
      "step": 3795,
      "time_per_iteration": 2.5996851921081543
    },
    {
      "auxiliary_loss_clip": 0.01133733,
      "auxiliary_loss_mlp": 0.01046029,
      "balance_loss_clip": 1.05219698,
      "balance_loss_mlp": 1.02917266,
      "epoch": 0.22822786712761162,
      "flos": 22451602694400.0,
      "grad_norm": 2.089163903812295,
      "language_loss": 0.76966131,
      "learning_rate": 3.6017664946002704e-06,
      "loss": 0.79145896,
      "num_input_tokens_seen": 81721265,
      "step": 3796,
      "time_per_iteration": 2.5228569507598877
    },
    {
      "auxiliary_loss_clip": 0.01103359,
      "auxiliary_loss_mlp": 0.00795969,
      "balance_loss_clip": 1.04818046,
      "balance_loss_mlp": 1.01720834,
      "epoch": 0.22828799038027958,
      "flos": 12202554395520.0,
      "grad_norm": 2.1580096836122515,
      "language_loss": 0.95989692,
      "learning_rate": 3.6015332465826188e-06,
      "loss": 0.97889018,
      "num_input_tokens_seen": 81736565,
      "step": 3797,
      "time_per_iteration": 2.5574705600738525
    },
    {
      "auxiliary_loss_clip": 0.01137323,
      "auxiliary_loss_mlp": 0.00795131,
      "balance_loss_clip": 1.05348182,
      "balance_loss_mlp": 1.01597154,
      "epoch": 0.22834811363294755,
      "flos": 22085708803200.0,
      "grad_norm": 1.5757044299367566,
      "language_loss": 0.81202567,
      "learning_rate": 3.601299937834666e-06,
      "loss": 0.83135021,
      "num_input_tokens_seen": 81756240,
      "step": 3798,
      "time_per_iteration": 2.5479326248168945
    },
    {
      "auxiliary_loss_clip": 0.01113112,
      "auxiliary_loss_mlp": 0.01037209,
      "balance_loss_clip": 1.05063128,
      "balance_loss_mlp": 1.01900554,
      "epoch": 0.2284082368856155,
      "flos": 24860634094080.0,
      "grad_norm": 2.452795832126358,
      "language_loss": 0.79113203,
      "learning_rate": 3.6010665683652596e-06,
      "loss": 0.81263524,
      "num_input_tokens_seen": 81775720,
      "step": 3799,
      "time_per_iteration": 2.5948526859283447
    },
    {
      "auxiliary_loss_clip": 0.01118759,
      "auxiliary_loss_mlp": 0.01054222,
      "balance_loss_clip": 1.05095744,
      "balance_loss_mlp": 1.03586364,
      "epoch": 0.22846836013828348,
      "flos": 23292882109440.0,
      "grad_norm": 1.8155651871840301,
      "language_loss": 0.75414002,
      "learning_rate": 3.6008331381832484e-06,
      "loss": 0.77586985,
      "num_input_tokens_seen": 81795830,
      "step": 3800,
      "time_per_iteration": 2.561786651611328
    },
    {
      "auxiliary_loss_clip": 0.01124746,
      "auxiliary_loss_mlp": 0.0104176,
      "balance_loss_clip": 1.05498254,
      "balance_loss_mlp": 1.02583408,
      "epoch": 0.22852848339095144,
      "flos": 27416288810880.0,
      "grad_norm": 1.6691442648886738,
      "language_loss": 0.63720864,
      "learning_rate": 3.600599647297484e-06,
      "loss": 0.65887368,
      "num_input_tokens_seen": 81815745,
      "step": 3801,
      "time_per_iteration": 2.5844504833221436
    },
    {
      "auxiliary_loss_clip": 0.01127055,
      "auxiliary_loss_mlp": 0.01041513,
      "balance_loss_clip": 1.05653346,
      "balance_loss_mlp": 1.0257175,
      "epoch": 0.2285886066436194,
      "flos": 26321157002880.0,
      "grad_norm": 2.3050139798167013,
      "language_loss": 0.81792516,
      "learning_rate": 3.60036609571682e-06,
      "loss": 0.83961082,
      "num_input_tokens_seen": 81835155,
      "step": 3802,
      "time_per_iteration": 2.611619710922241
    },
    {
      "auxiliary_loss_clip": 0.011281,
      "auxiliary_loss_mlp": 0.0105442,
      "balance_loss_clip": 1.06025064,
      "balance_loss_mlp": 1.0367893,
      "epoch": 0.2286487298962874,
      "flos": 29716475022720.0,
      "grad_norm": 1.7905081121109456,
      "language_loss": 0.78735512,
      "learning_rate": 3.600132483450114e-06,
      "loss": 0.80918032,
      "num_input_tokens_seen": 81855655,
      "step": 3803,
      "time_per_iteration": 2.628741502761841
    },
    {
      "auxiliary_loss_clip": 0.01113139,
      "auxiliary_loss_mlp": 0.01044992,
      "balance_loss_clip": 1.05113792,
      "balance_loss_mlp": 1.02787316,
      "epoch": 0.22870885314895537,
      "flos": 21287199507840.0,
      "grad_norm": 1.5833700352435451,
      "language_loss": 0.84969807,
      "learning_rate": 3.5998988105062235e-06,
      "loss": 0.87127936,
      "num_input_tokens_seen": 81876385,
      "step": 3804,
      "time_per_iteration": 2.586489200592041
    },
    {
      "auxiliary_loss_clip": 0.01138806,
      "auxiliary_loss_mlp": 0.01041581,
      "balance_loss_clip": 1.05276096,
      "balance_loss_mlp": 1.02493906,
      "epoch": 0.22876897640162333,
      "flos": 14939450161920.0,
      "grad_norm": 2.242384506574323,
      "language_loss": 0.76627123,
      "learning_rate": 3.59966507689401e-06,
      "loss": 0.78807509,
      "num_input_tokens_seen": 81893225,
      "step": 3805,
      "time_per_iteration": 2.5013389587402344
    },
    {
      "auxiliary_loss_clip": 0.01129626,
      "auxiliary_loss_mlp": 0.00795506,
      "balance_loss_clip": 1.05583358,
      "balance_loss_mlp": 1.01592612,
      "epoch": 0.2288290996542913,
      "flos": 18113917409280.0,
      "grad_norm": 4.644412032126836,
      "language_loss": 0.7903083,
      "learning_rate": 3.5994312826223363e-06,
      "loss": 0.80955958,
      "num_input_tokens_seen": 81911350,
      "step": 3806,
      "time_per_iteration": 2.525341510772705
    },
    {
      "auxiliary_loss_clip": 0.01121625,
      "auxiliary_loss_mlp": 0.01050926,
      "balance_loss_clip": 1.05316019,
      "balance_loss_mlp": 1.03235292,
      "epoch": 0.22888922290695926,
      "flos": 39855457071360.0,
      "grad_norm": 2.091225503063303,
      "language_loss": 0.69822854,
      "learning_rate": 3.5991974277000684e-06,
      "loss": 0.71995407,
      "num_input_tokens_seen": 81935420,
      "step": 3807,
      "time_per_iteration": 2.7038092613220215
    },
    {
      "auxiliary_loss_clip": 0.0114706,
      "auxiliary_loss_mlp": 0.01050295,
      "balance_loss_clip": 1.06095433,
      "balance_loss_mlp": 1.03198469,
      "epoch": 0.22894934615962723,
      "flos": 23403774372480.0,
      "grad_norm": 2.200221131365818,
      "language_loss": 0.65599227,
      "learning_rate": 3.5989635121360733e-06,
      "loss": 0.67796582,
      "num_input_tokens_seen": 81953845,
      "step": 3808,
      "time_per_iteration": 2.5318214893341064
    },
    {
      "auxiliary_loss_clip": 0.0109752,
      "auxiliary_loss_mlp": 0.01054022,
      "balance_loss_clip": 1.05052137,
      "balance_loss_mlp": 1.03624737,
      "epoch": 0.22900946941229522,
      "flos": 18843011671680.0,
      "grad_norm": 1.8949715307955886,
      "language_loss": 0.74956495,
      "learning_rate": 3.598729535939222e-06,
      "loss": 0.77108037,
      "num_input_tokens_seen": 81972100,
      "step": 3809,
      "time_per_iteration": 2.5904080867767334
    },
    {
      "auxiliary_loss_clip": 0.01125718,
      "auxiliary_loss_mlp": 0.01038884,
      "balance_loss_clip": 1.05510354,
      "balance_loss_mlp": 1.02287471,
      "epoch": 0.22906959266496318,
      "flos": 22929394429440.0,
      "grad_norm": 1.614383918462754,
      "language_loss": 0.8155936,
      "learning_rate": 3.5984954991183862e-06,
      "loss": 0.83723968,
      "num_input_tokens_seen": 81992760,
      "step": 3810,
      "time_per_iteration": 2.5499823093414307
    },
    {
      "auxiliary_loss_clip": 0.01126784,
      "auxiliary_loss_mlp": 0.01032894,
      "balance_loss_clip": 1.05243492,
      "balance_loss_mlp": 1.01790977,
      "epoch": 0.22912971591763115,
      "flos": 19354523299200.0,
      "grad_norm": 2.198637753390495,
      "language_loss": 0.79057354,
      "learning_rate": 3.598261401682441e-06,
      "loss": 0.81217039,
      "num_input_tokens_seen": 82009080,
      "step": 3811,
      "time_per_iteration": 2.5312952995300293
    },
    {
      "auxiliary_loss_clip": 0.01135288,
      "auxiliary_loss_mlp": 0.00795858,
      "balance_loss_clip": 1.06011629,
      "balance_loss_mlp": 1.01558495,
      "epoch": 0.22918983917029911,
      "flos": 19933546538880.0,
      "grad_norm": 5.739862218808734,
      "language_loss": 0.82732236,
      "learning_rate": 3.5980272436402632e-06,
      "loss": 0.84663379,
      "num_input_tokens_seen": 82026705,
      "step": 3812,
      "time_per_iteration": 2.5240819454193115
    },
    {
      "auxiliary_loss_clip": 0.01086842,
      "auxiliary_loss_mlp": 0.01054779,
      "balance_loss_clip": 1.05046439,
      "balance_loss_mlp": 1.03642094,
      "epoch": 0.22924996242296708,
      "flos": 16690885320960.0,
      "grad_norm": 2.4629318187352154,
      "language_loss": 0.82635909,
      "learning_rate": 3.5977930250007324e-06,
      "loss": 0.84777534,
      "num_input_tokens_seen": 82043245,
      "step": 3813,
      "time_per_iteration": 2.6057562828063965
    },
    {
      "auxiliary_loss_clip": 0.01135033,
      "auxiliary_loss_mlp": 0.01044464,
      "balance_loss_clip": 1.05851972,
      "balance_loss_mlp": 1.02757156,
      "epoch": 0.22931008567563504,
      "flos": 33036164956800.0,
      "grad_norm": 1.888609972105612,
      "language_loss": 0.69965899,
      "learning_rate": 3.5975587457727298e-06,
      "loss": 0.72145396,
      "num_input_tokens_seen": 82066870,
      "step": 3814,
      "time_per_iteration": 2.6102564334869385
    },
    {
      "auxiliary_loss_clip": 0.01138264,
      "auxiliary_loss_mlp": 0.01044251,
      "balance_loss_clip": 1.05333614,
      "balance_loss_mlp": 1.02714503,
      "epoch": 0.229370208928303,
      "flos": 23330696152320.0,
      "grad_norm": 2.742251065343541,
      "language_loss": 0.67091596,
      "learning_rate": 3.597324405965139e-06,
      "loss": 0.6927411,
      "num_input_tokens_seen": 82083180,
      "step": 3815,
      "time_per_iteration": 2.53525710105896
    },
    {
      "auxiliary_loss_clip": 0.01148441,
      "auxiliary_loss_mlp": 0.01049799,
      "balance_loss_clip": 1.06018019,
      "balance_loss_mlp": 1.03381324,
      "epoch": 0.229430332180971,
      "flos": 28617213150720.0,
      "grad_norm": 1.708018138405201,
      "language_loss": 0.83380258,
      "learning_rate": 3.597090005586848e-06,
      "loss": 0.85578495,
      "num_input_tokens_seen": 82102950,
      "step": 3816,
      "time_per_iteration": 2.566465139389038
    },
    {
      "auxiliary_loss_clip": 0.01140657,
      "auxiliary_loss_mlp": 0.01034272,
      "balance_loss_clip": 1.058828,
      "balance_loss_mlp": 1.01695037,
      "epoch": 0.22949045543363897,
      "flos": 17238199829760.0,
      "grad_norm": 4.240854013106063,
      "language_loss": 0.87123549,
      "learning_rate": 3.596855544646742e-06,
      "loss": 0.89298475,
      "num_input_tokens_seen": 82119510,
      "step": 3817,
      "time_per_iteration": 3.866034984588623
    },
    {
      "auxiliary_loss_clip": 0.01125226,
      "auxiliary_loss_mlp": 0.01046038,
      "balance_loss_clip": 1.05549395,
      "balance_loss_mlp": 1.02880073,
      "epoch": 0.22955057868630693,
      "flos": 27489438858240.0,
      "grad_norm": 1.6005545995053239,
      "language_loss": 0.74556327,
      "learning_rate": 3.5966210231537154e-06,
      "loss": 0.76727587,
      "num_input_tokens_seen": 82140095,
      "step": 3818,
      "time_per_iteration": 3.9936180114746094
    },
    {
      "auxiliary_loss_clip": 0.01139507,
      "auxiliary_loss_mlp": 0.0103809,
      "balance_loss_clip": 1.05679381,
      "balance_loss_mlp": 1.02143693,
      "epoch": 0.2296107019389749,
      "flos": 23476421629440.0,
      "grad_norm": 1.6731713417265093,
      "language_loss": 0.74736351,
      "learning_rate": 3.596386441116659e-06,
      "loss": 0.76913953,
      "num_input_tokens_seen": 82159510,
      "step": 3819,
      "time_per_iteration": 2.5997955799102783
    },
    {
      "auxiliary_loss_clip": 0.01139898,
      "auxiliary_loss_mlp": 0.01042427,
      "balance_loss_clip": 1.05681157,
      "balance_loss_mlp": 1.02629173,
      "epoch": 0.22967082519164286,
      "flos": 31285160760960.0,
      "grad_norm": 1.7319723288135025,
      "language_loss": 0.80956084,
      "learning_rate": 3.5961517985444684e-06,
      "loss": 0.83138406,
      "num_input_tokens_seen": 82179580,
      "step": 3820,
      "time_per_iteration": 2.5874264240264893
    },
    {
      "auxiliary_loss_clip": 0.01128673,
      "auxiliary_loss_mlp": 0.0104448,
      "balance_loss_clip": 1.05561662,
      "balance_loss_mlp": 1.02642,
      "epoch": 0.22973094844431083,
      "flos": 14642935390080.0,
      "grad_norm": 2.2280449423555724,
      "language_loss": 0.69402218,
      "learning_rate": 3.595917095446042e-06,
      "loss": 0.71575373,
      "num_input_tokens_seen": 82195585,
      "step": 3821,
      "time_per_iteration": 2.5137147903442383
    },
    {
      "auxiliary_loss_clip": 0.01099532,
      "auxiliary_loss_mlp": 0.01035926,
      "balance_loss_clip": 1.05981803,
      "balance_loss_mlp": 1.0186882,
      "epoch": 0.2297910716969788,
      "flos": 22823853292800.0,
      "grad_norm": 1.4825531544775898,
      "language_loss": 0.83136272,
      "learning_rate": 3.5956823318302796e-06,
      "loss": 0.85271722,
      "num_input_tokens_seen": 82217530,
      "step": 3822,
      "time_per_iteration": 2.657780647277832
    },
    {
      "auxiliary_loss_clip": 0.01149851,
      "auxiliary_loss_mlp": 0.01044512,
      "balance_loss_clip": 1.05641174,
      "balance_loss_mlp": 1.02677333,
      "epoch": 0.2298511949496468,
      "flos": 23039029716480.0,
      "grad_norm": 1.5632386111803027,
      "language_loss": 0.66070205,
      "learning_rate": 3.5954475077060833e-06,
      "loss": 0.68264568,
      "num_input_tokens_seen": 82237980,
      "step": 3823,
      "time_per_iteration": 3.971372127532959
    },
    {
      "auxiliary_loss_clip": 0.01051588,
      "auxiliary_loss_mlp": 0.01001549,
      "balance_loss_clip": 1.0325247,
      "balance_loss_mlp": 0.9988429,
      "epoch": 0.22991131820231475,
      "flos": 66890914911360.0,
      "grad_norm": 0.8171841913192446,
      "language_loss": 0.56839406,
      "learning_rate": 3.595212623082357e-06,
      "loss": 0.58892548,
      "num_input_tokens_seen": 82301785,
      "step": 3824,
      "time_per_iteration": 3.1776232719421387
    },
    {
      "auxiliary_loss_clip": 0.01122805,
      "auxiliary_loss_mlp": 0.0103955,
      "balance_loss_clip": 1.0534029,
      "balance_loss_mlp": 1.02327752,
      "epoch": 0.22997144145498272,
      "flos": 17887248633600.0,
      "grad_norm": 2.0750467084293063,
      "language_loss": 0.72595191,
      "learning_rate": 3.594977677968009e-06,
      "loss": 0.74757546,
      "num_input_tokens_seen": 82317355,
      "step": 3825,
      "time_per_iteration": 2.515634775161743
    },
    {
      "auxiliary_loss_clip": 0.01139412,
      "auxiliary_loss_mlp": 0.01049956,
      "balance_loss_clip": 1.05473948,
      "balance_loss_mlp": 1.03153777,
      "epoch": 0.23003156470765068,
      "flos": 24676843178880.0,
      "grad_norm": 1.82366078216413,
      "language_loss": 0.87964362,
      "learning_rate": 3.5947426723719473e-06,
      "loss": 0.9015373,
      "num_input_tokens_seen": 82336645,
      "step": 3826,
      "time_per_iteration": 2.5402939319610596
    },
    {
      "auxiliary_loss_clip": 0.01126882,
      "auxiliary_loss_mlp": 0.01046266,
      "balance_loss_clip": 1.05608273,
      "balance_loss_mlp": 1.02764535,
      "epoch": 0.23009168796031865,
      "flos": 15814126247040.0,
      "grad_norm": 3.0688260964426632,
      "language_loss": 0.81612957,
      "learning_rate": 3.594507606303083e-06,
      "loss": 0.83786106,
      "num_input_tokens_seen": 82354225,
      "step": 3827,
      "time_per_iteration": 2.523902177810669
    },
    {
      "auxiliary_loss_clip": 0.01076117,
      "auxiliary_loss_mlp": 0.01045603,
      "balance_loss_clip": 1.04851866,
      "balance_loss_mlp": 1.02756619,
      "epoch": 0.2301518112129866,
      "flos": 16212842190720.0,
      "grad_norm": 2.3892060762369685,
      "language_loss": 0.86546439,
      "learning_rate": 3.5942724797703314e-06,
      "loss": 0.88668156,
      "num_input_tokens_seen": 82370240,
      "step": 3828,
      "time_per_iteration": 2.589344024658203
    },
    {
      "auxiliary_loss_clip": 0.01125152,
      "auxiliary_loss_mlp": 0.01048758,
      "balance_loss_clip": 1.0567863,
      "balance_loss_mlp": 1.03039944,
      "epoch": 0.2302119344656546,
      "flos": 20595452411520.0,
      "grad_norm": 2.0139500624780555,
      "language_loss": 0.70714861,
      "learning_rate": 3.594037292782607e-06,
      "loss": 0.72888768,
      "num_input_tokens_seen": 82389145,
      "step": 3829,
      "time_per_iteration": 2.5454485416412354
    },
    {
      "auxiliary_loss_clip": 0.01084924,
      "auxiliary_loss_mlp": 0.01041512,
      "balance_loss_clip": 1.05200577,
      "balance_loss_mlp": 1.02558529,
      "epoch": 0.23027205771832257,
      "flos": 26796901662720.0,
      "grad_norm": 1.5972425233474026,
      "language_loss": 0.84025955,
      "learning_rate": 3.5938020453488293e-06,
      "loss": 0.86152393,
      "num_input_tokens_seen": 82409185,
      "step": 3830,
      "time_per_iteration": 2.6867055892944336
    },
    {
      "auxiliary_loss_clip": 0.01127956,
      "auxiliary_loss_mlp": 0.01049759,
      "balance_loss_clip": 1.05144382,
      "balance_loss_mlp": 1.03277194,
      "epoch": 0.23033218097099054,
      "flos": 43873143068160.0,
      "grad_norm": 1.6028440664985804,
      "language_loss": 0.67325199,
      "learning_rate": 3.5935667374779177e-06,
      "loss": 0.69502914,
      "num_input_tokens_seen": 82432070,
      "step": 3831,
      "time_per_iteration": 2.707850456237793
    },
    {
      "auxiliary_loss_clip": 0.01109291,
      "auxiliary_loss_mlp": 0.0105248,
      "balance_loss_clip": 1.0569694,
      "balance_loss_mlp": 1.03405035,
      "epoch": 0.2303923042236585,
      "flos": 26067663745920.0,
      "grad_norm": 3.0344446314632174,
      "language_loss": 0.76166391,
      "learning_rate": 3.5933313691787957e-06,
      "loss": 0.78328156,
      "num_input_tokens_seen": 82450625,
      "step": 3832,
      "time_per_iteration": 2.5985777378082275
    },
    {
      "auxiliary_loss_clip": 0.01099045,
      "auxiliary_loss_mlp": 0.01047682,
      "balance_loss_clip": 1.05293441,
      "balance_loss_mlp": 1.02928817,
      "epoch": 0.23045242747632647,
      "flos": 18296379521280.0,
      "grad_norm": 1.829429721008068,
      "language_loss": 0.87619793,
      "learning_rate": 3.593095940460389e-06,
      "loss": 0.89766526,
      "num_input_tokens_seen": 82468575,
      "step": 3833,
      "time_per_iteration": 2.5935466289520264
    },
    {
      "auxiliary_loss_clip": 0.01112984,
      "auxiliary_loss_mlp": 0.01048962,
      "balance_loss_clip": 1.05210924,
      "balance_loss_mlp": 1.03080678,
      "epoch": 0.23051255072899443,
      "flos": 25520528805120.0,
      "grad_norm": 1.780049537790239,
      "language_loss": 0.74901283,
      "learning_rate": 3.592860451331624e-06,
      "loss": 0.77063227,
      "num_input_tokens_seen": 82488655,
      "step": 3834,
      "time_per_iteration": 2.610978126525879
    },
    {
      "auxiliary_loss_clip": 0.01103886,
      "auxiliary_loss_mlp": 0.01061426,
      "balance_loss_clip": 1.05283308,
      "balance_loss_mlp": 1.04125547,
      "epoch": 0.2305726739816624,
      "flos": 21215198695680.0,
      "grad_norm": 1.7578101009021503,
      "language_loss": 0.85852575,
      "learning_rate": 3.592624901801432e-06,
      "loss": 0.88017881,
      "num_input_tokens_seen": 82507220,
      "step": 3835,
      "time_per_iteration": 2.586960792541504
    },
    {
      "auxiliary_loss_clip": 0.01112557,
      "auxiliary_loss_mlp": 0.01055955,
      "balance_loss_clip": 1.05020642,
      "balance_loss_mlp": 1.03644037,
      "epoch": 0.2306327972343304,
      "flos": 23331127115520.0,
      "grad_norm": 2.491410226603742,
      "language_loss": 0.8185432,
      "learning_rate": 3.5923892918787432e-06,
      "loss": 0.84022826,
      "num_input_tokens_seen": 82527920,
      "step": 3836,
      "time_per_iteration": 2.6100287437438965
    },
    {
      "auxiliary_loss_clip": 0.01136285,
      "auxiliary_loss_mlp": 0.01052514,
      "balance_loss_clip": 1.05724311,
      "balance_loss_mlp": 1.03534746,
      "epoch": 0.23069292048699835,
      "flos": 20666734951680.0,
      "grad_norm": 1.605544813887528,
      "language_loss": 0.79533976,
      "learning_rate": 3.5921536215724934e-06,
      "loss": 0.81722772,
      "num_input_tokens_seen": 82549040,
      "step": 3837,
      "time_per_iteration": 2.5394492149353027
    },
    {
      "auxiliary_loss_clip": 0.01046184,
      "auxiliary_loss_mlp": 0.0102709,
      "balance_loss_clip": 1.04074001,
      "balance_loss_mlp": 1.02447975,
      "epoch": 0.23075304373966632,
      "flos": 70454832393600.0,
      "grad_norm": 0.9346560746316889,
      "language_loss": 0.65387332,
      "learning_rate": 3.5919178908916184e-06,
      "loss": 0.67460603,
      "num_input_tokens_seen": 82604070,
      "step": 3838,
      "time_per_iteration": 3.08566951751709
    },
    {
      "auxiliary_loss_clip": 0.01132201,
      "auxiliary_loss_mlp": 0.01047361,
      "balance_loss_clip": 1.05306172,
      "balance_loss_mlp": 1.03054023,
      "epoch": 0.23081316699233428,
      "flos": 16617986668800.0,
      "grad_norm": 1.9156261730714912,
      "language_loss": 0.75731093,
      "learning_rate": 3.591682099845058e-06,
      "loss": 0.77910656,
      "num_input_tokens_seen": 82619665,
      "step": 3839,
      "time_per_iteration": 2.495717763900757
    },
    {
      "auxiliary_loss_clip": 0.01120692,
      "auxiliary_loss_mlp": 0.01041811,
      "balance_loss_clip": 1.05538511,
      "balance_loss_mlp": 1.02440643,
      "epoch": 0.23087329024500225,
      "flos": 13298081253120.0,
      "grad_norm": 2.035411630268444,
      "language_loss": 0.6884886,
      "learning_rate": 3.591446248441752e-06,
      "loss": 0.71011364,
      "num_input_tokens_seen": 82637530,
      "step": 3840,
      "time_per_iteration": 2.5268445014953613
    },
    {
      "auxiliary_loss_clip": 0.01151082,
      "auxiliary_loss_mlp": 0.0104394,
      "balance_loss_clip": 1.05637467,
      "balance_loss_mlp": 1.02520072,
      "epoch": 0.23093341349767021,
      "flos": 17785729820160.0,
      "grad_norm": 2.025841668078174,
      "language_loss": 0.7971912,
      "learning_rate": 3.591210336690645e-06,
      "loss": 0.81914151,
      "num_input_tokens_seen": 82656130,
      "step": 3841,
      "time_per_iteration": 2.4816672801971436
    },
    {
      "auxiliary_loss_clip": 0.01139139,
      "auxiliary_loss_mlp": 0.01040012,
      "balance_loss_clip": 1.05647695,
      "balance_loss_mlp": 1.02391839,
      "epoch": 0.23099353675033818,
      "flos": 23988076911360.0,
      "grad_norm": 1.8662482736974775,
      "language_loss": 0.83445776,
      "learning_rate": 3.590974364600683e-06,
      "loss": 0.85624921,
      "num_input_tokens_seen": 82675295,
      "step": 3842,
      "time_per_iteration": 2.5472657680511475
    },
    {
      "auxiliary_loss_clip": 0.01137663,
      "auxiliary_loss_mlp": 0.0104281,
      "balance_loss_clip": 1.05154705,
      "balance_loss_mlp": 1.02445161,
      "epoch": 0.23105366000300617,
      "flos": 35995168471680.0,
      "grad_norm": 1.4360807069558068,
      "language_loss": 0.6650964,
      "learning_rate": 3.5907383321808135e-06,
      "loss": 0.68690115,
      "num_input_tokens_seen": 82703260,
      "step": 3843,
      "time_per_iteration": 2.7052645683288574
    },
    {
      "auxiliary_loss_clip": 0.01133906,
      "auxiliary_loss_mlp": 0.01045617,
      "balance_loss_clip": 1.05309486,
      "balance_loss_mlp": 1.02802229,
      "epoch": 0.23111378325567414,
      "flos": 31245335556480.0,
      "grad_norm": 1.615161002875236,
      "language_loss": 0.77084875,
      "learning_rate": 3.590502239439987e-06,
      "loss": 0.79264402,
      "num_input_tokens_seen": 82725060,
      "step": 3844,
      "time_per_iteration": 2.606454610824585
    },
    {
      "auxiliary_loss_clip": 0.01134577,
      "auxiliary_loss_mlp": 0.0104704,
      "balance_loss_clip": 1.05273199,
      "balance_loss_mlp": 1.02871752,
      "epoch": 0.2311739065083421,
      "flos": 19208223204480.0,
      "grad_norm": 1.5631763753814163,
      "language_loss": 0.78123415,
      "learning_rate": 3.590266086387156e-06,
      "loss": 0.80305028,
      "num_input_tokens_seen": 82742960,
      "step": 3845,
      "time_per_iteration": 2.5240631103515625
    },
    {
      "auxiliary_loss_clip": 0.01107608,
      "auxiliary_loss_mlp": 0.01035747,
      "balance_loss_clip": 1.04875386,
      "balance_loss_mlp": 1.01993954,
      "epoch": 0.23123402976101007,
      "flos": 23360178240000.0,
      "grad_norm": 2.0441653389253376,
      "language_loss": 0.75972998,
      "learning_rate": 3.590029873031276e-06,
      "loss": 0.78116357,
      "num_input_tokens_seen": 82760205,
      "step": 3846,
      "time_per_iteration": 2.5820088386535645
    },
    {
      "auxiliary_loss_clip": 0.01126932,
      "auxiliary_loss_mlp": 0.01044352,
      "balance_loss_clip": 1.05349731,
      "balance_loss_mlp": 1.02772248,
      "epoch": 0.23129415301367803,
      "flos": 13735365425280.0,
      "grad_norm": 2.4425711810497073,
      "language_loss": 0.70110404,
      "learning_rate": 3.589793599381304e-06,
      "loss": 0.72281688,
      "num_input_tokens_seen": 82778590,
      "step": 3847,
      "time_per_iteration": 2.5345842838287354
    },
    {
      "auxiliary_loss_clip": 0.01060946,
      "auxiliary_loss_mlp": 0.01006553,
      "balance_loss_clip": 1.0418359,
      "balance_loss_mlp": 1.00420427,
      "epoch": 0.231354276266346,
      "flos": 69737015001600.0,
      "grad_norm": 0.7940712764160209,
      "language_loss": 0.61086202,
      "learning_rate": 3.589557265446198e-06,
      "loss": 0.63153702,
      "num_input_tokens_seen": 82833925,
      "step": 3848,
      "time_per_iteration": 3.030233144760132
    },
    {
      "auxiliary_loss_clip": 0.01137148,
      "auxiliary_loss_mlp": 0.01048688,
      "balance_loss_clip": 1.05557525,
      "balance_loss_mlp": 1.03066349,
      "epoch": 0.231414399519014,
      "flos": 18835900778880.0,
      "grad_norm": 2.094396965048532,
      "language_loss": 0.78266042,
      "learning_rate": 3.589320871234923e-06,
      "loss": 0.80451882,
      "num_input_tokens_seen": 82850625,
      "step": 3849,
      "time_per_iteration": 2.4994523525238037
    },
    {
      "auxiliary_loss_clip": 0.01133121,
      "auxiliary_loss_mlp": 0.01042574,
      "balance_loss_clip": 1.05249631,
      "balance_loss_mlp": 1.0250268,
      "epoch": 0.23147452277168196,
      "flos": 36135470995200.0,
      "grad_norm": 1.8935979820710251,
      "language_loss": 0.71224451,
      "learning_rate": 3.5890844167564405e-06,
      "loss": 0.73400152,
      "num_input_tokens_seen": 82872105,
      "step": 3850,
      "time_per_iteration": 2.652895927429199
    },
    {
      "auxiliary_loss_clip": 0.01121514,
      "auxiliary_loss_mlp": 0.00810802,
      "balance_loss_clip": 1.05772889,
      "balance_loss_mlp": 1.0419687,
      "epoch": 0.23153464602434992,
      "flos": 20812927305600.0,
      "grad_norm": 2.011395541614,
      "language_loss": 0.76376027,
      "learning_rate": 3.588847902019718e-06,
      "loss": 0.78308338,
      "num_input_tokens_seen": 82890595,
      "step": 3851,
      "time_per_iteration": 2.545243740081787
    },
    {
      "auxiliary_loss_clip": 0.01146663,
      "auxiliary_loss_mlp": 0.01042639,
      "balance_loss_clip": 1.05388856,
      "balance_loss_mlp": 1.02479386,
      "epoch": 0.2315947692770179,
      "flos": 19939256801280.0,
      "grad_norm": 1.8115928421958385,
      "language_loss": 0.69626987,
      "learning_rate": 3.588611327033723e-06,
      "loss": 0.71816289,
      "num_input_tokens_seen": 82908910,
      "step": 3852,
      "time_per_iteration": 2.5057969093322754
    },
    {
      "auxiliary_loss_clip": 0.01113036,
      "auxiliary_loss_mlp": 0.01045179,
      "balance_loss_clip": 1.06081414,
      "balance_loss_mlp": 1.02649903,
      "epoch": 0.23165489252968585,
      "flos": 12855553695360.0,
      "grad_norm": 2.650874525338518,
      "language_loss": 0.67401052,
      "learning_rate": 3.588374691807428e-06,
      "loss": 0.69559258,
      "num_input_tokens_seen": 82925405,
      "step": 3853,
      "time_per_iteration": 2.522764205932617
    },
    {
      "auxiliary_loss_clip": 0.01138854,
      "auxiliary_loss_mlp": 0.01036554,
      "balance_loss_clip": 1.05338836,
      "balance_loss_mlp": 1.01857686,
      "epoch": 0.23171501578235382,
      "flos": 30628282792320.0,
      "grad_norm": 2.4119632688366015,
      "language_loss": 0.80101323,
      "learning_rate": 3.5881379963498053e-06,
      "loss": 0.82276726,
      "num_input_tokens_seen": 82945615,
      "step": 3854,
      "time_per_iteration": 2.5823771953582764
    },
    {
      "auxiliary_loss_clip": 0.01115098,
      "auxiliary_loss_mlp": 0.01051409,
      "balance_loss_clip": 1.05072403,
      "balance_loss_mlp": 1.0300827,
      "epoch": 0.23177513903502178,
      "flos": 23842782397440.0,
      "grad_norm": 3.252832112311624,
      "language_loss": 0.65646672,
      "learning_rate": 3.587901240669831e-06,
      "loss": 0.67813182,
      "num_input_tokens_seen": 82967570,
      "step": 3855,
      "time_per_iteration": 4.036926984786987
    },
    {
      "auxiliary_loss_clip": 0.0114918,
      "auxiliary_loss_mlp": 0.01043817,
      "balance_loss_clip": 1.05324948,
      "balance_loss_mlp": 1.02685332,
      "epoch": 0.23183526228768978,
      "flos": 29570282668800.0,
      "grad_norm": 2.0111640426151682,
      "language_loss": 0.71261704,
      "learning_rate": 3.5876644247764815e-06,
      "loss": 0.73454702,
      "num_input_tokens_seen": 82987435,
      "step": 3856,
      "time_per_iteration": 2.5408313274383545
    },
    {
      "auxiliary_loss_clip": 0.011005,
      "auxiliary_loss_mlp": 0.01039564,
      "balance_loss_clip": 1.05382097,
      "balance_loss_mlp": 1.02360165,
      "epoch": 0.23189538554035774,
      "flos": 34458694254720.0,
      "grad_norm": 1.5260220099513135,
      "language_loss": 0.77132583,
      "learning_rate": 3.5874275486787387e-06,
      "loss": 0.79272652,
      "num_input_tokens_seen": 83010505,
      "step": 3857,
      "time_per_iteration": 4.0918004512786865
    },
    {
      "auxiliary_loss_clip": 0.01129478,
      "auxiliary_loss_mlp": 0.00799768,
      "balance_loss_clip": 1.04996872,
      "balance_loss_mlp": 1.02118623,
      "epoch": 0.2319555087930257,
      "flos": 18003815245440.0,
      "grad_norm": 2.4106730137185237,
      "language_loss": 0.9106437,
      "learning_rate": 3.587190612385584e-06,
      "loss": 0.92993617,
      "num_input_tokens_seen": 83026705,
      "step": 3858,
      "time_per_iteration": 2.5487334728240967
    },
    {
      "auxiliary_loss_clip": 0.01090704,
      "auxiliary_loss_mlp": 0.01038561,
      "balance_loss_clip": 1.05095077,
      "balance_loss_mlp": 1.02216983,
      "epoch": 0.23201563204569367,
      "flos": 23143852581120.0,
      "grad_norm": 1.7309755194288259,
      "language_loss": 0.76138818,
      "learning_rate": 3.5869536159060026e-06,
      "loss": 0.78268087,
      "num_input_tokens_seen": 83046500,
      "step": 3859,
      "time_per_iteration": 2.6166861057281494
    },
    {
      "auxiliary_loss_clip": 0.0113217,
      "auxiliary_loss_mlp": 0.01032303,
      "balance_loss_clip": 1.04978406,
      "balance_loss_mlp": 1.01588821,
      "epoch": 0.23207575529836164,
      "flos": 20667991927680.0,
      "grad_norm": 1.656541366781877,
      "language_loss": 0.83928525,
      "learning_rate": 3.58671655924898e-06,
      "loss": 0.86092997,
      "num_input_tokens_seen": 83065280,
      "step": 3860,
      "time_per_iteration": 2.539497137069702
    },
    {
      "auxiliary_loss_clip": 0.01091533,
      "auxiliary_loss_mlp": 0.01040985,
      "balance_loss_clip": 1.05061531,
      "balance_loss_mlp": 1.02250767,
      "epoch": 0.2321358785510296,
      "flos": 16472189364480.0,
      "grad_norm": 1.822780011957372,
      "language_loss": 0.83020008,
      "learning_rate": 3.586479442423508e-06,
      "loss": 0.85152531,
      "num_input_tokens_seen": 83082310,
      "step": 3861,
      "time_per_iteration": 2.5872466564178467
    },
    {
      "auxiliary_loss_clip": 0.01128088,
      "auxiliary_loss_mlp": 0.00796424,
      "balance_loss_clip": 1.05216801,
      "balance_loss_mlp": 1.01638377,
      "epoch": 0.2321960018036976,
      "flos": 21616320850560.0,
      "grad_norm": 1.6843751149850719,
      "language_loss": 0.85838985,
      "learning_rate": 3.586242265438576e-06,
      "loss": 0.877635,
      "num_input_tokens_seen": 83102065,
      "step": 3862,
      "time_per_iteration": 4.0203776359558105
    },
    {
      "auxiliary_loss_clip": 0.01117841,
      "auxiliary_loss_mlp": 0.01039902,
      "balance_loss_clip": 1.05903065,
      "balance_loss_mlp": 1.02445257,
      "epoch": 0.23225612505636556,
      "flos": 22271474966400.0,
      "grad_norm": 1.7204047758070597,
      "language_loss": 0.74910867,
      "learning_rate": 3.5860050283031773e-06,
      "loss": 0.77068615,
      "num_input_tokens_seen": 83121445,
      "step": 3863,
      "time_per_iteration": 2.571600914001465
    },
    {
      "auxiliary_loss_clip": 0.01107992,
      "auxiliary_loss_mlp": 0.01045304,
      "balance_loss_clip": 1.05544806,
      "balance_loss_mlp": 1.02888846,
      "epoch": 0.23231624830903352,
      "flos": 17052325925760.0,
      "grad_norm": 1.8189940122783201,
      "language_loss": 0.74842155,
      "learning_rate": 3.58576773102631e-06,
      "loss": 0.76995444,
      "num_input_tokens_seen": 83138175,
      "step": 3864,
      "time_per_iteration": 2.5869011878967285
    },
    {
      "auxiliary_loss_clip": 0.01144826,
      "auxiliary_loss_mlp": 0.01038351,
      "balance_loss_clip": 1.05242896,
      "balance_loss_mlp": 1.02127969,
      "epoch": 0.2323763715617015,
      "flos": 34640043045120.0,
      "grad_norm": 2.058784779144382,
      "language_loss": 0.70547283,
      "learning_rate": 3.5855303736169714e-06,
      "loss": 0.72730464,
      "num_input_tokens_seen": 83161975,
      "step": 3865,
      "time_per_iteration": 2.5978221893310547
    },
    {
      "auxiliary_loss_clip": 0.01155092,
      "auxiliary_loss_mlp": 0.01048915,
      "balance_loss_clip": 1.05587006,
      "balance_loss_mlp": 1.0303539,
      "epoch": 0.23243649481436945,
      "flos": 25551698832000.0,
      "grad_norm": 1.8017779123713527,
      "language_loss": 0.94576895,
      "learning_rate": 3.5852929560841617e-06,
      "loss": 0.96780908,
      "num_input_tokens_seen": 83180905,
      "step": 3866,
      "time_per_iteration": 2.5655124187469482
    },
    {
      "auxiliary_loss_clip": 0.01128886,
      "auxiliary_loss_mlp": 0.01039209,
      "balance_loss_clip": 1.04990268,
      "balance_loss_mlp": 1.0221982,
      "epoch": 0.23249661806703742,
      "flos": 20483482740480.0,
      "grad_norm": 2.493588391839717,
      "language_loss": 0.73201525,
      "learning_rate": 3.5850554784368846e-06,
      "loss": 0.7536962,
      "num_input_tokens_seen": 83196390,
      "step": 3867,
      "time_per_iteration": 2.482191801071167
    },
    {
      "auxiliary_loss_clip": 0.01127142,
      "auxiliary_loss_mlp": 0.01041987,
      "balance_loss_clip": 1.05097556,
      "balance_loss_mlp": 1.02509546,
      "epoch": 0.23255674131970538,
      "flos": 20376612800640.0,
      "grad_norm": 1.765781478004179,
      "language_loss": 0.82719493,
      "learning_rate": 3.584817940684145e-06,
      "loss": 0.84888625,
      "num_input_tokens_seen": 83216165,
      "step": 3868,
      "time_per_iteration": 2.593801736831665
    },
    {
      "auxiliary_loss_clip": 0.01124397,
      "auxiliary_loss_mlp": 0.0104429,
      "balance_loss_clip": 1.0515275,
      "balance_loss_mlp": 1.02725494,
      "epoch": 0.23261686457237338,
      "flos": 17056096853760.0,
      "grad_norm": 1.9521774646481656,
      "language_loss": 0.73168361,
      "learning_rate": 3.58458034283495e-06,
      "loss": 0.75337046,
      "num_input_tokens_seen": 83233845,
      "step": 3869,
      "time_per_iteration": 2.4829907417297363
    },
    {
      "auxiliary_loss_clip": 0.01132829,
      "auxiliary_loss_mlp": 0.0104785,
      "balance_loss_clip": 1.05462408,
      "balance_loss_mlp": 1.03108931,
      "epoch": 0.23267698782504134,
      "flos": 29169878785920.0,
      "grad_norm": 15.293602618480255,
      "language_loss": 0.79206967,
      "learning_rate": 3.5843426848983097e-06,
      "loss": 0.81387651,
      "num_input_tokens_seen": 83254930,
      "step": 3870,
      "time_per_iteration": 2.619703531265259
    },
    {
      "auxiliary_loss_clip": 0.01147897,
      "auxiliary_loss_mlp": 0.01043658,
      "balance_loss_clip": 1.05296874,
      "balance_loss_mlp": 1.02651596,
      "epoch": 0.2327371110777093,
      "flos": 21174655219200.0,
      "grad_norm": 1.8384305671757024,
      "language_loss": 0.70410174,
      "learning_rate": 3.5841049668832357e-06,
      "loss": 0.72601736,
      "num_input_tokens_seen": 83272095,
      "step": 3871,
      "time_per_iteration": 2.493462085723877
    },
    {
      "auxiliary_loss_clip": 0.01136869,
      "auxiliary_loss_mlp": 0.01052224,
      "balance_loss_clip": 1.05817842,
      "balance_loss_mlp": 1.03324604,
      "epoch": 0.23279723433037727,
      "flos": 24863112132480.0,
      "grad_norm": 6.688714128283287,
      "language_loss": 0.69049025,
      "learning_rate": 3.5838671887987433e-06,
      "loss": 0.71238124,
      "num_input_tokens_seen": 83290980,
      "step": 3872,
      "time_per_iteration": 2.577817678451538
    },
    {
      "auxiliary_loss_clip": 0.01142374,
      "auxiliary_loss_mlp": 0.01043333,
      "balance_loss_clip": 1.05421102,
      "balance_loss_mlp": 1.0249989,
      "epoch": 0.23285735758304524,
      "flos": 38800617344640.0,
      "grad_norm": 1.525020017001597,
      "language_loss": 0.7780028,
      "learning_rate": 3.5836293506538474e-06,
      "loss": 0.79985988,
      "num_input_tokens_seen": 83315175,
      "step": 3873,
      "time_per_iteration": 2.689356565475464
    },
    {
      "auxiliary_loss_clip": 0.01051988,
      "auxiliary_loss_mlp": 0.01010075,
      "balance_loss_clip": 1.03918886,
      "balance_loss_mlp": 1.00741696,
      "epoch": 0.2329174808357132,
      "flos": 53944113692160.0,
      "grad_norm": 0.8490383026407595,
      "language_loss": 0.605214,
      "learning_rate": 3.5833914524575687e-06,
      "loss": 0.62583464,
      "num_input_tokens_seen": 83372060,
      "step": 3874,
      "time_per_iteration": 3.062441110610962
    },
    {
      "auxiliary_loss_clip": 0.01124961,
      "auxiliary_loss_mlp": 0.01042183,
      "balance_loss_clip": 1.05330908,
      "balance_loss_mlp": 1.02436161,
      "epoch": 0.23297760408838117,
      "flos": 21216024708480.0,
      "grad_norm": 3.2240305411525663,
      "language_loss": 0.80500668,
      "learning_rate": 3.583153494218927e-06,
      "loss": 0.82667816,
      "num_input_tokens_seen": 83389795,
      "step": 3875,
      "time_per_iteration": 2.571101665496826
    },
    {
      "auxiliary_loss_clip": 0.01147783,
      "auxiliary_loss_mlp": 0.00798837,
      "balance_loss_clip": 1.05613542,
      "balance_loss_mlp": 1.02486992,
      "epoch": 0.23303772734104916,
      "flos": 28403006394240.0,
      "grad_norm": 1.64310276657619,
      "language_loss": 0.61345834,
      "learning_rate": 3.5829154759469464e-06,
      "loss": 0.6329245,
      "num_input_tokens_seen": 83410005,
      "step": 3876,
      "time_per_iteration": 2.5708000659942627
    },
    {
      "auxiliary_loss_clip": 0.01118109,
      "auxiliary_loss_mlp": 0.01042884,
      "balance_loss_clip": 1.05915928,
      "balance_loss_mlp": 1.02537227,
      "epoch": 0.23309785059371713,
      "flos": 24314720215680.0,
      "grad_norm": 1.931757293229752,
      "language_loss": 0.71062177,
      "learning_rate": 3.5826773976506523e-06,
      "loss": 0.73223174,
      "num_input_tokens_seen": 83430250,
      "step": 3877,
      "time_per_iteration": 2.6000571250915527
    },
    {
      "auxiliary_loss_clip": 0.011399,
      "auxiliary_loss_mlp": 0.01051045,
      "balance_loss_clip": 1.05654049,
      "balance_loss_mlp": 1.03340197,
      "epoch": 0.2331579738463851,
      "flos": 15992925171840.0,
      "grad_norm": 2.2813554562185407,
      "language_loss": 0.81070411,
      "learning_rate": 3.582439259339073e-06,
      "loss": 0.83261347,
      "num_input_tokens_seen": 83447950,
      "step": 3878,
      "time_per_iteration": 2.506186008453369
    },
    {
      "auxiliary_loss_clip": 0.01084864,
      "auxiliary_loss_mlp": 0.01045473,
      "balance_loss_clip": 1.04644668,
      "balance_loss_mlp": 1.02608931,
      "epoch": 0.23321809709905306,
      "flos": 36426957863040.0,
      "grad_norm": 1.7316012393998932,
      "language_loss": 0.74919564,
      "learning_rate": 3.5822010610212374e-06,
      "loss": 0.77049899,
      "num_input_tokens_seen": 83467785,
      "step": 3879,
      "time_per_iteration": 2.7952187061309814
    },
    {
      "auxiliary_loss_clip": 0.01095582,
      "auxiliary_loss_mlp": 0.01039258,
      "balance_loss_clip": 1.05416429,
      "balance_loss_mlp": 1.02093565,
      "epoch": 0.23327822035172102,
      "flos": 21324762155520.0,
      "grad_norm": 2.321457071640085,
      "language_loss": 0.8972072,
      "learning_rate": 3.5819628027061795e-06,
      "loss": 0.91855556,
      "num_input_tokens_seen": 83485390,
      "step": 3880,
      "time_per_iteration": 2.6010191440582275
    },
    {
      "auxiliary_loss_clip": 0.01127161,
      "auxiliary_loss_mlp": 0.01044466,
      "balance_loss_clip": 1.05499732,
      "balance_loss_mlp": 1.02744257,
      "epoch": 0.233338343604389,
      "flos": 19171881619200.0,
      "grad_norm": 1.6927385304705507,
      "language_loss": 0.72089016,
      "learning_rate": 3.5817244844029334e-06,
      "loss": 0.74260652,
      "num_input_tokens_seen": 83504890,
      "step": 3881,
      "time_per_iteration": 2.575547695159912
    },
    {
      "auxiliary_loss_clip": 0.01146485,
      "auxiliary_loss_mlp": 0.01044773,
      "balance_loss_clip": 1.05325842,
      "balance_loss_mlp": 1.02764237,
      "epoch": 0.23339846685705698,
      "flos": 26908368543360.0,
      "grad_norm": 1.6269802434737304,
      "language_loss": 0.67780507,
      "learning_rate": 3.581486106120537e-06,
      "loss": 0.69971764,
      "num_input_tokens_seen": 83526475,
      "step": 3882,
      "time_per_iteration": 2.5681276321411133
    },
    {
      "auxiliary_loss_clip": 0.01106051,
      "auxiliary_loss_mlp": 0.01051955,
      "balance_loss_clip": 1.0474664,
      "balance_loss_mlp": 1.03308427,
      "epoch": 0.23345859010972494,
      "flos": 32343160884480.0,
      "grad_norm": 1.9675810578556254,
      "language_loss": 0.76881325,
      "learning_rate": 3.5812476678680287e-06,
      "loss": 0.79039335,
      "num_input_tokens_seen": 83546620,
      "step": 3883,
      "time_per_iteration": 2.661811590194702
    },
    {
      "auxiliary_loss_clip": 0.0104927,
      "auxiliary_loss_mlp": 0.01005505,
      "balance_loss_clip": 1.03467298,
      "balance_loss_mlp": 1.00277507,
      "epoch": 0.2335187133623929,
      "flos": 58484229050880.0,
      "grad_norm": 0.7777488366157249,
      "language_loss": 0.59151244,
      "learning_rate": 3.58100916965445e-06,
      "loss": 0.61206019,
      "num_input_tokens_seen": 83616160,
      "step": 3884,
      "time_per_iteration": 3.3538589477539062
    },
    {
      "auxiliary_loss_clip": 0.01120725,
      "auxiliary_loss_mlp": 0.01036242,
      "balance_loss_clip": 1.05468392,
      "balance_loss_mlp": 1.0197196,
      "epoch": 0.23357883661506088,
      "flos": 24502317972480.0,
      "grad_norm": 2.6143479579132314,
      "language_loss": 0.80295408,
      "learning_rate": 3.5807706114888455e-06,
      "loss": 0.82452375,
      "num_input_tokens_seen": 83636795,
      "step": 3885,
      "time_per_iteration": 2.618666172027588
    },
    {
      "auxiliary_loss_clip": 0.01131021,
      "auxiliary_loss_mlp": 0.01038131,
      "balance_loss_clip": 1.05429745,
      "balance_loss_mlp": 1.02084613,
      "epoch": 0.23363895986772884,
      "flos": 18948516894720.0,
      "grad_norm": 2.346821077523392,
      "language_loss": 0.88331831,
      "learning_rate": 3.580531993380261e-06,
      "loss": 0.90500987,
      "num_input_tokens_seen": 83654050,
      "step": 3886,
      "time_per_iteration": 2.5157597064971924
    },
    {
      "auxiliary_loss_clip": 0.01149729,
      "auxiliary_loss_mlp": 0.01041353,
      "balance_loss_clip": 1.0557071,
      "balance_loss_mlp": 1.02419889,
      "epoch": 0.2336990831203968,
      "flos": 31686821619840.0,
      "grad_norm": 1.8732007155560442,
      "language_loss": 0.73695683,
      "learning_rate": 3.5802933153377445e-06,
      "loss": 0.75886762,
      "num_input_tokens_seen": 83673720,
      "step": 3887,
      "time_per_iteration": 2.577000379562378
    },
    {
      "auxiliary_loss_clip": 0.01134754,
      "auxiliary_loss_mlp": 0.01039545,
      "balance_loss_clip": 1.05111885,
      "balance_loss_mlp": 1.02322495,
      "epoch": 0.23375920637306477,
      "flos": 27709750926720.0,
      "grad_norm": 1.8635191725501978,
      "language_loss": 0.84497476,
      "learning_rate": 3.5800545773703475e-06,
      "loss": 0.8667177,
      "num_input_tokens_seen": 83693470,
      "step": 3888,
      "time_per_iteration": 2.6348724365234375
    },
    {
      "auxiliary_loss_clip": 0.01119564,
      "auxiliary_loss_mlp": 0.01058134,
      "balance_loss_clip": 1.05093253,
      "balance_loss_mlp": 1.0385952,
      "epoch": 0.23381932962573276,
      "flos": 17675627656320.0,
      "grad_norm": 2.159306801968494,
      "language_loss": 0.87320733,
      "learning_rate": 3.5798157794871225e-06,
      "loss": 0.89498425,
      "num_input_tokens_seen": 83711620,
      "step": 3889,
      "time_per_iteration": 2.526057004928589
    },
    {
      "auxiliary_loss_clip": 0.01137155,
      "auxiliary_loss_mlp": 0.01040449,
      "balance_loss_clip": 1.05318761,
      "balance_loss_mlp": 1.02359271,
      "epoch": 0.23387945287840073,
      "flos": 14390842763520.0,
      "grad_norm": 4.653730390801575,
      "language_loss": 0.7707029,
      "learning_rate": 3.579576921697125e-06,
      "loss": 0.79247892,
      "num_input_tokens_seen": 83727890,
      "step": 3890,
      "time_per_iteration": 2.539618492126465
    },
    {
      "auxiliary_loss_clip": 0.01108339,
      "auxiliary_loss_mlp": 0.00795451,
      "balance_loss_clip": 1.0527426,
      "balance_loss_mlp": 1.01678765,
      "epoch": 0.2339395761310687,
      "flos": 46097988503040.0,
      "grad_norm": 1.6717155662623473,
      "language_loss": 0.73291975,
      "learning_rate": 3.579338004009412e-06,
      "loss": 0.75195765,
      "num_input_tokens_seen": 83749370,
      "step": 3891,
      "time_per_iteration": 2.8195302486419678
    },
    {
      "auxiliary_loss_clip": 0.01143654,
      "auxiliary_loss_mlp": 0.01039023,
      "balance_loss_clip": 1.05354166,
      "balance_loss_mlp": 1.02258372,
      "epoch": 0.23399969938373666,
      "flos": 22382044007040.0,
      "grad_norm": 1.5871962329691258,
      "language_loss": 0.82981378,
      "learning_rate": 3.5790990264330433e-06,
      "loss": 0.85164046,
      "num_input_tokens_seen": 83769560,
      "step": 3892,
      "time_per_iteration": 2.5688040256500244
    },
    {
      "auxiliary_loss_clip": 0.01105339,
      "auxiliary_loss_mlp": 0.0104989,
      "balance_loss_clip": 1.04963768,
      "balance_loss_mlp": 1.02951741,
      "epoch": 0.23405982263640462,
      "flos": 43508542066560.0,
      "grad_norm": 1.453913861858147,
      "language_loss": 0.65050375,
      "learning_rate": 3.578859988977082e-06,
      "loss": 0.67205608,
      "num_input_tokens_seen": 83795635,
      "step": 3893,
      "time_per_iteration": 2.7584879398345947
    },
    {
      "auxiliary_loss_clip": 0.01113923,
      "auxiliary_loss_mlp": 0.0104016,
      "balance_loss_clip": 1.06200314,
      "balance_loss_mlp": 1.02176595,
      "epoch": 0.2341199458890726,
      "flos": 22564685687040.0,
      "grad_norm": 2.868089135926816,
      "language_loss": 0.79227817,
      "learning_rate": 3.5786208916505916e-06,
      "loss": 0.81381899,
      "num_input_tokens_seen": 83814090,
      "step": 3894,
      "time_per_iteration": 4.0035107135772705
    },
    {
      "auxiliary_loss_clip": 0.0113356,
      "auxiliary_loss_mlp": 0.0104073,
      "balance_loss_clip": 1.05239773,
      "balance_loss_mlp": 1.02382612,
      "epoch": 0.23418006914174055,
      "flos": 25633970933760.0,
      "grad_norm": 2.1046401457706976,
      "language_loss": 0.82053024,
      "learning_rate": 3.5783817344626383e-06,
      "loss": 0.84227312,
      "num_input_tokens_seen": 83836870,
      "step": 3895,
      "time_per_iteration": 2.6201205253601074
    },
    {
      "auxiliary_loss_clip": 0.01139076,
      "auxiliary_loss_mlp": 0.01046976,
      "balance_loss_clip": 1.05710959,
      "balance_loss_mlp": 1.02923775,
      "epoch": 0.23424019239440855,
      "flos": 13545936074880.0,
      "grad_norm": 2.0008996418892937,
      "language_loss": 0.80568826,
      "learning_rate": 3.578142517422292e-06,
      "loss": 0.82754874,
      "num_input_tokens_seen": 83853275,
      "step": 3896,
      "time_per_iteration": 3.876016616821289
    },
    {
      "auxiliary_loss_clip": 0.01127004,
      "auxiliary_loss_mlp": 0.01043088,
      "balance_loss_clip": 1.05199289,
      "balance_loss_mlp": 1.02403855,
      "epoch": 0.2343003156470765,
      "flos": 22419498913920.0,
      "grad_norm": 1.5764381503053442,
      "language_loss": 0.82902706,
      "learning_rate": 3.577903240538623e-06,
      "loss": 0.85072803,
      "num_input_tokens_seen": 83872340,
      "step": 3897,
      "time_per_iteration": 2.566791296005249
    },
    {
      "auxiliary_loss_clip": 0.01133539,
      "auxiliary_loss_mlp": 0.01042952,
      "balance_loss_clip": 1.05464339,
      "balance_loss_mlp": 1.02530861,
      "epoch": 0.23436043889974448,
      "flos": 14790815683200.0,
      "grad_norm": 1.5646035849299171,
      "language_loss": 0.79256725,
      "learning_rate": 3.577663903820705e-06,
      "loss": 0.81433213,
      "num_input_tokens_seen": 83888795,
      "step": 3898,
      "time_per_iteration": 2.478588581085205
    },
    {
      "auxiliary_loss_clip": 0.01112753,
      "auxiliary_loss_mlp": 0.01043387,
      "balance_loss_clip": 1.05270946,
      "balance_loss_mlp": 1.02628088,
      "epoch": 0.23442056215241244,
      "flos": 22965700101120.0,
      "grad_norm": 2.064936566284087,
      "language_loss": 0.7366392,
      "learning_rate": 3.577424507277614e-06,
      "loss": 0.75820053,
      "num_input_tokens_seen": 83906820,
      "step": 3899,
      "time_per_iteration": 2.608407497406006
    },
    {
      "auxiliary_loss_clip": 0.01111163,
      "auxiliary_loss_mlp": 0.01047782,
      "balance_loss_clip": 1.05194843,
      "balance_loss_mlp": 1.03019893,
      "epoch": 0.2344806854050804,
      "flos": 23071887682560.0,
      "grad_norm": 1.6676316870048196,
      "language_loss": 0.7541697,
      "learning_rate": 3.5771850509184277e-06,
      "loss": 0.77575922,
      "num_input_tokens_seen": 83926370,
      "step": 3900,
      "time_per_iteration": 2.567404270172119
    },
    {
      "auxiliary_loss_clip": 0.01107523,
      "auxiliary_loss_mlp": 0.01047892,
      "balance_loss_clip": 1.0601145,
      "balance_loss_mlp": 1.03026068,
      "epoch": 0.23454080865774837,
      "flos": 16327074418560.0,
      "grad_norm": 2.012355218575242,
      "language_loss": 0.6691407,
      "learning_rate": 3.5769455347522256e-06,
      "loss": 0.69069481,
      "num_input_tokens_seen": 83944600,
      "step": 3901,
      "time_per_iteration": 5.293537378311157
    },
    {
      "auxiliary_loss_clip": 0.01026844,
      "auxiliary_loss_mlp": 0.00999459,
      "balance_loss_clip": 1.03124702,
      "balance_loss_mlp": 0.99714655,
      "epoch": 0.23460093191041637,
      "flos": 67760958142080.0,
      "grad_norm": 0.7527031580649051,
      "language_loss": 0.58220869,
      "learning_rate": 3.576705958788091e-06,
      "loss": 0.60247171,
      "num_input_tokens_seen": 84005100,
      "step": 3902,
      "time_per_iteration": 3.174189567565918
    },
    {
      "auxiliary_loss_clip": 0.01130002,
      "auxiliary_loss_mlp": 0.01050458,
      "balance_loss_clip": 1.05637133,
      "balance_loss_mlp": 1.03175402,
      "epoch": 0.23466105516308433,
      "flos": 20077619990400.0,
      "grad_norm": 2.3244849516837096,
      "language_loss": 0.80499339,
      "learning_rate": 3.576466323035108e-06,
      "loss": 0.82679808,
      "num_input_tokens_seen": 84023775,
      "step": 3903,
      "time_per_iteration": 2.5738298892974854
    },
    {
      "auxiliary_loss_clip": 0.01099335,
      "auxiliary_loss_mlp": 0.01043526,
      "balance_loss_clip": 1.05619013,
      "balance_loss_mlp": 1.02515554,
      "epoch": 0.2347211784157523,
      "flos": 24535714642560.0,
      "grad_norm": 1.7923512067004113,
      "language_loss": 0.8232035,
      "learning_rate": 3.5762266275023645e-06,
      "loss": 0.84463209,
      "num_input_tokens_seen": 84042605,
      "step": 3904,
      "time_per_iteration": 2.635223388671875
    },
    {
      "auxiliary_loss_clip": 0.01148598,
      "auxiliary_loss_mlp": 0.01047905,
      "balance_loss_clip": 1.05604887,
      "balance_loss_mlp": 1.03087008,
      "epoch": 0.23478130166842026,
      "flos": 23805040181760.0,
      "grad_norm": 1.8997540249887228,
      "language_loss": 0.71538502,
      "learning_rate": 3.57598687219895e-06,
      "loss": 0.73735005,
      "num_input_tokens_seen": 84061520,
      "step": 3905,
      "time_per_iteration": 2.5212044715881348
    },
    {
      "auxiliary_loss_clip": 0.01147522,
      "auxiliary_loss_mlp": 0.01036524,
      "balance_loss_clip": 1.05702829,
      "balance_loss_mlp": 1.01994205,
      "epoch": 0.23484142492108823,
      "flos": 24093618048000.0,
      "grad_norm": 1.761662216246446,
      "language_loss": 0.71436769,
      "learning_rate": 3.5757470571339543e-06,
      "loss": 0.73620814,
      "num_input_tokens_seen": 84081800,
      "step": 3906,
      "time_per_iteration": 2.5279879570007324
    },
    {
      "auxiliary_loss_clip": 0.01141072,
      "auxiliary_loss_mlp": 0.01041947,
      "balance_loss_clip": 1.05402184,
      "balance_loss_mlp": 1.02281404,
      "epoch": 0.2349015481737562,
      "flos": 29095830898560.0,
      "grad_norm": 1.9646820758690333,
      "language_loss": 0.72994077,
      "learning_rate": 3.575507182316473e-06,
      "loss": 0.75177097,
      "num_input_tokens_seen": 84102340,
      "step": 3907,
      "time_per_iteration": 2.6095364093780518
    },
    {
      "auxiliary_loss_clip": 0.01138728,
      "auxiliary_loss_mlp": 0.01054018,
      "balance_loss_clip": 1.05558634,
      "balance_loss_mlp": 1.03568316,
      "epoch": 0.23496167142642416,
      "flos": 18916305373440.0,
      "grad_norm": 1.7842776017656081,
      "language_loss": 0.72878927,
      "learning_rate": 3.575267247755601e-06,
      "loss": 0.75071669,
      "num_input_tokens_seen": 84120370,
      "step": 3908,
      "time_per_iteration": 2.511951446533203
    },
    {
      "auxiliary_loss_clip": 0.01049448,
      "auxiliary_loss_mlp": 0.01012821,
      "balance_loss_clip": 1.03424811,
      "balance_loss_mlp": 1.01041305,
      "epoch": 0.23502179467909215,
      "flos": 55868062896000.0,
      "grad_norm": 1.0197950639386366,
      "language_loss": 0.73358488,
      "learning_rate": 3.5750272534604367e-06,
      "loss": 0.75420761,
      "num_input_tokens_seen": 84165515,
      "step": 3909,
      "time_per_iteration": 2.897395372390747
    },
    {
      "auxiliary_loss_clip": 0.01137857,
      "auxiliary_loss_mlp": 0.01042713,
      "balance_loss_clip": 1.05374527,
      "balance_loss_mlp": 1.02565455,
      "epoch": 0.23508191793176011,
      "flos": 23401763210880.0,
      "grad_norm": 1.7041253606887201,
      "language_loss": 0.88087815,
      "learning_rate": 3.5747871994400822e-06,
      "loss": 0.90268385,
      "num_input_tokens_seen": 84184540,
      "step": 3910,
      "time_per_iteration": 2.5438942909240723
    },
    {
      "auxiliary_loss_clip": 0.01139193,
      "auxiliary_loss_mlp": 0.01040831,
      "balance_loss_clip": 1.0567801,
      "balance_loss_mlp": 1.02406979,
      "epoch": 0.23514204118442808,
      "flos": 20047671025920.0,
      "grad_norm": 2.0276179915671215,
      "language_loss": 0.76595032,
      "learning_rate": 3.5745470857036386e-06,
      "loss": 0.78775054,
      "num_input_tokens_seen": 84202025,
      "step": 3911,
      "time_per_iteration": 2.5497519969940186
    },
    {
      "auxiliary_loss_clip": 0.01132305,
      "auxiliary_loss_mlp": 0.01041499,
      "balance_loss_clip": 1.05387557,
      "balance_loss_mlp": 1.02560854,
      "epoch": 0.23520216443709605,
      "flos": 21580589796480.0,
      "grad_norm": 1.5003743110299204,
      "language_loss": 0.81658304,
      "learning_rate": 3.5743069122602122e-06,
      "loss": 0.83832109,
      "num_input_tokens_seen": 84221895,
      "step": 3912,
      "time_per_iteration": 2.529372453689575
    },
    {
      "auxiliary_loss_clip": 0.01127451,
      "auxiliary_loss_mlp": 0.01048356,
      "balance_loss_clip": 1.05657291,
      "balance_loss_mlp": 1.03071332,
      "epoch": 0.235262287689764,
      "flos": 23185796688000.0,
      "grad_norm": 1.7967396908625806,
      "language_loss": 0.71360731,
      "learning_rate": 3.574066679118909e-06,
      "loss": 0.73536545,
      "num_input_tokens_seen": 84240455,
      "step": 3913,
      "time_per_iteration": 2.5951547622680664
    },
    {
      "auxiliary_loss_clip": 0.01141361,
      "auxiliary_loss_mlp": 0.00793969,
      "balance_loss_clip": 1.05409336,
      "balance_loss_mlp": 1.01293683,
      "epoch": 0.23532241094243198,
      "flos": 23185222070400.0,
      "grad_norm": 1.7296641080301844,
      "language_loss": 0.76219821,
      "learning_rate": 3.57382638628884e-06,
      "loss": 0.78155148,
      "num_input_tokens_seen": 84261605,
      "step": 3914,
      "time_per_iteration": 2.5661492347717285
    },
    {
      "auxiliary_loss_clip": 0.01085353,
      "auxiliary_loss_mlp": 0.01043631,
      "balance_loss_clip": 1.05228651,
      "balance_loss_mlp": 1.02567792,
      "epoch": 0.23538253419509997,
      "flos": 17019324305280.0,
      "grad_norm": 2.8790178696119533,
      "language_loss": 0.89877123,
      "learning_rate": 3.5735860337791174e-06,
      "loss": 0.92006105,
      "num_input_tokens_seen": 84278675,
      "step": 3915,
      "time_per_iteration": 2.62831449508667
    },
    {
      "auxiliary_loss_clip": 0.01042616,
      "auxiliary_loss_mlp": 0.01000677,
      "balance_loss_clip": 1.02608991,
      "balance_loss_mlp": 0.9979946,
      "epoch": 0.23544265744776793,
      "flos": 63448588967040.0,
      "grad_norm": 0.8074520255887843,
      "language_loss": 0.59380126,
      "learning_rate": 3.573345621598854e-06,
      "loss": 0.61423421,
      "num_input_tokens_seen": 84329765,
      "step": 3916,
      "time_per_iteration": 3.0832412242889404
    },
    {
      "auxiliary_loss_clip": 0.01015193,
      "auxiliary_loss_mlp": 0.01002176,
      "balance_loss_clip": 1.03043211,
      "balance_loss_mlp": 0.99936277,
      "epoch": 0.2355027807004359,
      "flos": 70515343831680.0,
      "grad_norm": 0.7702746569586227,
      "language_loss": 0.49471438,
      "learning_rate": 3.5731051497571675e-06,
      "loss": 0.51488805,
      "num_input_tokens_seen": 84393680,
      "step": 3917,
      "time_per_iteration": 3.2329819202423096
    },
    {
      "auxiliary_loss_clip": 0.01116677,
      "auxiliary_loss_mlp": 0.01051243,
      "balance_loss_clip": 1.05844116,
      "balance_loss_mlp": 1.03436255,
      "epoch": 0.23556290395310386,
      "flos": 21434289701760.0,
      "grad_norm": 1.8752454051907863,
      "language_loss": 0.76419842,
      "learning_rate": 3.5728646182631756e-06,
      "loss": 0.78587759,
      "num_input_tokens_seen": 84412640,
      "step": 3918,
      "time_per_iteration": 2.5890719890594482
    },
    {
      "auxiliary_loss_clip": 0.01096085,
      "auxiliary_loss_mlp": 0.0103914,
      "balance_loss_clip": 1.04773235,
      "balance_loss_mlp": 1.02267718,
      "epoch": 0.23562302720577183,
      "flos": 18186421011840.0,
      "grad_norm": 10.554131542572643,
      "language_loss": 0.69566238,
      "learning_rate": 3.5726240271259995e-06,
      "loss": 0.71701461,
      "num_input_tokens_seen": 84431605,
      "step": 3919,
      "time_per_iteration": 2.5676207542419434
    },
    {
      "auxiliary_loss_clip": 0.01110378,
      "auxiliary_loss_mlp": 0.01039202,
      "balance_loss_clip": 1.05501628,
      "balance_loss_mlp": 1.02245307,
      "epoch": 0.2356831504584398,
      "flos": 33730497832320.0,
      "grad_norm": 1.9173113006695695,
      "language_loss": 0.70360518,
      "learning_rate": 3.5723833763547634e-06,
      "loss": 0.72510099,
      "num_input_tokens_seen": 84454210,
      "step": 3920,
      "time_per_iteration": 2.710014820098877
    },
    {
      "auxiliary_loss_clip": 0.01121519,
      "auxiliary_loss_mlp": 0.01048638,
      "balance_loss_clip": 1.05253398,
      "balance_loss_mlp": 1.03183007,
      "epoch": 0.23574327371110776,
      "flos": 24932778560640.0,
      "grad_norm": 1.665346823779534,
      "language_loss": 0.77063107,
      "learning_rate": 3.5721426659585916e-06,
      "loss": 0.79233265,
      "num_input_tokens_seen": 84475540,
      "step": 3921,
      "time_per_iteration": 2.587679624557495
    },
    {
      "auxiliary_loss_clip": 0.01112469,
      "auxiliary_loss_mlp": 0.01045842,
      "balance_loss_clip": 1.05206418,
      "balance_loss_mlp": 1.02816379,
      "epoch": 0.23580339696377575,
      "flos": 17822107319040.0,
      "grad_norm": 2.351883693085046,
      "language_loss": 0.75070214,
      "learning_rate": 3.571901895946612e-06,
      "loss": 0.77228528,
      "num_input_tokens_seen": 84494580,
      "step": 3922,
      "time_per_iteration": 2.576209783554077
    },
    {
      "auxiliary_loss_clip": 0.01114793,
      "auxiliary_loss_mlp": 0.01040974,
      "balance_loss_clip": 1.0519383,
      "balance_loss_mlp": 1.02489233,
      "epoch": 0.23586352021644372,
      "flos": 26286611097600.0,
      "grad_norm": 1.9961203350925847,
      "language_loss": 0.80343777,
      "learning_rate": 3.571661066327956e-06,
      "loss": 0.82499546,
      "num_input_tokens_seen": 84513850,
      "step": 3923,
      "time_per_iteration": 2.5762481689453125
    },
    {
      "auxiliary_loss_clip": 0.01086276,
      "auxiliary_loss_mlp": 0.01049699,
      "balance_loss_clip": 1.04736614,
      "balance_loss_mlp": 1.03242564,
      "epoch": 0.23592364346911168,
      "flos": 14246697484800.0,
      "grad_norm": 1.973201472019396,
      "language_loss": 0.74515307,
      "learning_rate": 3.571420177111754e-06,
      "loss": 0.76651281,
      "num_input_tokens_seen": 84532315,
      "step": 3924,
      "time_per_iteration": 2.592226266860962
    },
    {
      "auxiliary_loss_clip": 0.01146611,
      "auxiliary_loss_mlp": 0.01045704,
      "balance_loss_clip": 1.05387878,
      "balance_loss_mlp": 1.02980185,
      "epoch": 0.23598376672177965,
      "flos": 18587938216320.0,
      "grad_norm": 1.7950237453265165,
      "language_loss": 0.83280671,
      "learning_rate": 3.5711792283071416e-06,
      "loss": 0.85472989,
      "num_input_tokens_seen": 84550970,
      "step": 3925,
      "time_per_iteration": 2.4800949096679688
    },
    {
      "auxiliary_loss_clip": 0.0112533,
      "auxiliary_loss_mlp": 0.01054382,
      "balance_loss_clip": 1.05168962,
      "balance_loss_mlp": 1.03751397,
      "epoch": 0.2360438899744476,
      "flos": 22675542036480.0,
      "grad_norm": 2.1336688701653137,
      "language_loss": 0.59609914,
      "learning_rate": 3.5709382199232564e-06,
      "loss": 0.61789626,
      "num_input_tokens_seen": 84571655,
      "step": 3926,
      "time_per_iteration": 2.5812859535217285
    },
    {
      "auxiliary_loss_clip": 0.01118433,
      "auxiliary_loss_mlp": 0.01041622,
      "balance_loss_clip": 1.04771316,
      "balance_loss_mlp": 1.02655411,
      "epoch": 0.23610401322711558,
      "flos": 29570139014400.0,
      "grad_norm": 1.941478080921335,
      "language_loss": 0.71280909,
      "learning_rate": 3.570697151969235e-06,
      "loss": 0.73440969,
      "num_input_tokens_seen": 84593130,
      "step": 3927,
      "time_per_iteration": 2.5648488998413086
    },
    {
      "auxiliary_loss_clip": 0.01118156,
      "auxiliary_loss_mlp": 0.01046976,
      "balance_loss_clip": 1.04995942,
      "balance_loss_mlp": 1.03168738,
      "epoch": 0.23616413647978354,
      "flos": 17858520731520.0,
      "grad_norm": 1.7763875038189096,
      "language_loss": 0.75305879,
      "learning_rate": 3.570456024454221e-06,
      "loss": 0.77471006,
      "num_input_tokens_seen": 84612410,
      "step": 3928,
      "time_per_iteration": 2.594587564468384
    },
    {
      "auxiliary_loss_clip": 0.01117135,
      "auxiliary_loss_mlp": 0.01051785,
      "balance_loss_clip": 1.05286551,
      "balance_loss_mlp": 1.0336771,
      "epoch": 0.23622425973245154,
      "flos": 11034847157760.0,
      "grad_norm": 2.3426110973547973,
      "language_loss": 0.82109511,
      "learning_rate": 3.5702148373873576e-06,
      "loss": 0.84278429,
      "num_input_tokens_seen": 84627610,
      "step": 3929,
      "time_per_iteration": 2.512575149536133
    },
    {
      "auxiliary_loss_clip": 0.01151447,
      "auxiliary_loss_mlp": 0.01052083,
      "balance_loss_clip": 1.05433226,
      "balance_loss_mlp": 1.03330731,
      "epoch": 0.2362843829851195,
      "flos": 23404061681280.0,
      "grad_norm": 1.8325965753586955,
      "language_loss": 0.71781456,
      "learning_rate": 3.569973590777789e-06,
      "loss": 0.73984987,
      "num_input_tokens_seen": 84648415,
      "step": 3930,
      "time_per_iteration": 2.5138580799102783
    },
    {
      "auxiliary_loss_clip": 0.01143317,
      "auxiliary_loss_mlp": 0.01040959,
      "balance_loss_clip": 1.05108714,
      "balance_loss_mlp": 1.02410316,
      "epoch": 0.23634450623778747,
      "flos": 39529855261440.0,
      "grad_norm": 1.8178673741239038,
      "language_loss": 0.7397995,
      "learning_rate": 3.569732284634665e-06,
      "loss": 0.76164222,
      "num_input_tokens_seen": 84670080,
      "step": 3931,
      "time_per_iteration": 2.6234030723571777
    },
    {
      "auxiliary_loss_clip": 0.01133728,
      "auxiliary_loss_mlp": 0.01043585,
      "balance_loss_clip": 1.05262184,
      "balance_loss_mlp": 1.02609658,
      "epoch": 0.23640462949045543,
      "flos": 24207167917440.0,
      "grad_norm": 2.040284562163236,
      "language_loss": 0.80665421,
      "learning_rate": 3.569490918967136e-06,
      "loss": 0.82842731,
      "num_input_tokens_seen": 84686465,
      "step": 3932,
      "time_per_iteration": 2.549920082092285
    },
    {
      "auxiliary_loss_clip": 0.01108696,
      "auxiliary_loss_mlp": 0.01041547,
      "balance_loss_clip": 1.05572343,
      "balance_loss_mlp": 1.02644277,
      "epoch": 0.2364647527431234,
      "flos": 26177622255360.0,
      "grad_norm": 1.5914436875852116,
      "language_loss": 0.85524285,
      "learning_rate": 3.5692494937843537e-06,
      "loss": 0.87674534,
      "num_input_tokens_seen": 84708825,
      "step": 3933,
      "time_per_iteration": 4.058252573013306
    },
    {
      "auxiliary_loss_clip": 0.01108514,
      "auxiliary_loss_mlp": 0.01045578,
      "balance_loss_clip": 1.05562901,
      "balance_loss_mlp": 1.02611136,
      "epoch": 0.23652487599579136,
      "flos": 22637009721600.0,
      "grad_norm": 2.0669364058690505,
      "language_loss": 0.83576977,
      "learning_rate": 3.5690080090954727e-06,
      "loss": 0.85731071,
      "num_input_tokens_seen": 84726165,
      "step": 3934,
      "time_per_iteration": 2.5937938690185547
    },
    {
      "auxiliary_loss_clip": 0.01146281,
      "auxiliary_loss_mlp": 0.01045086,
      "balance_loss_clip": 1.05272651,
      "balance_loss_mlp": 1.02794361,
      "epoch": 0.23658499924845935,
      "flos": 21762261809280.0,
      "grad_norm": 1.6780617916493208,
      "language_loss": 0.78549421,
      "learning_rate": 3.5687664649096515e-06,
      "loss": 0.80740786,
      "num_input_tokens_seen": 84745815,
      "step": 3935,
      "time_per_iteration": 3.953526735305786
    },
    {
      "auxiliary_loss_clip": 0.01134349,
      "auxiliary_loss_mlp": 0.01039475,
      "balance_loss_clip": 1.05590248,
      "balance_loss_mlp": 1.02364421,
      "epoch": 0.23664512250112732,
      "flos": 21798998444160.0,
      "grad_norm": 1.7477941677946933,
      "language_loss": 0.79014444,
      "learning_rate": 3.5685248612360487e-06,
      "loss": 0.81188273,
      "num_input_tokens_seen": 84765415,
      "step": 3936,
      "time_per_iteration": 2.5164499282836914
    },
    {
      "auxiliary_loss_clip": 0.01124909,
      "auxiliary_loss_mlp": 0.0103769,
      "balance_loss_clip": 1.05274773,
      "balance_loss_mlp": 1.02038085,
      "epoch": 0.23670524575379528,
      "flos": 22637871648000.0,
      "grad_norm": 1.567388653470774,
      "language_loss": 0.79189163,
      "learning_rate": 3.568283198083826e-06,
      "loss": 0.81351763,
      "num_input_tokens_seen": 84787080,
      "step": 3937,
      "time_per_iteration": 2.5767908096313477
    },
    {
      "auxiliary_loss_clip": 0.01135977,
      "auxiliary_loss_mlp": 0.01039394,
      "balance_loss_clip": 1.05879116,
      "balance_loss_mlp": 1.02423072,
      "epoch": 0.23676536900646325,
      "flos": 16725000263040.0,
      "grad_norm": 2.3619585504561544,
      "language_loss": 0.85389519,
      "learning_rate": 3.568041475462147e-06,
      "loss": 0.87564886,
      "num_input_tokens_seen": 84805395,
      "step": 3938,
      "time_per_iteration": 2.5066115856170654
    },
    {
      "auxiliary_loss_clip": 0.01142166,
      "auxiliary_loss_mlp": 0.01045209,
      "balance_loss_clip": 1.05242527,
      "balance_loss_mlp": 1.02867436,
      "epoch": 0.23682549225913122,
      "flos": 11135611785600.0,
      "grad_norm": 2.81832367575203,
      "language_loss": 0.93989241,
      "learning_rate": 3.5677996933801785e-06,
      "loss": 0.96176612,
      "num_input_tokens_seen": 84818090,
      "step": 3939,
      "time_per_iteration": 3.929412603378296
    },
    {
      "auxiliary_loss_clip": 0.0114944,
      "auxiliary_loss_mlp": 0.01046233,
      "balance_loss_clip": 1.05516481,
      "balance_loss_mlp": 1.02879238,
      "epoch": 0.23688561551179918,
      "flos": 22559226819840.0,
      "grad_norm": 1.6538038438310103,
      "language_loss": 0.82099277,
      "learning_rate": 3.567557851847088e-06,
      "loss": 0.84294945,
      "num_input_tokens_seen": 84837695,
      "step": 3940,
      "time_per_iteration": 3.9966132640838623
    },
    {
      "auxiliary_loss_clip": 0.01132047,
      "auxiliary_loss_mlp": 0.00798548,
      "balance_loss_clip": 1.05700278,
      "balance_loss_mlp": 1.02392912,
      "epoch": 0.23694573876446715,
      "flos": 18514895909760.0,
      "grad_norm": 2.2433879758523916,
      "language_loss": 0.89842021,
      "learning_rate": 3.5673159508720464e-06,
      "loss": 0.91772616,
      "num_input_tokens_seen": 84854630,
      "step": 3941,
      "time_per_iteration": 2.546996593475342
    },
    {
      "auxiliary_loss_clip": 0.01146248,
      "auxiliary_loss_mlp": 0.01046029,
      "balance_loss_clip": 1.05281997,
      "balance_loss_mlp": 1.02810001,
      "epoch": 0.23700586201713514,
      "flos": 15335723980800.0,
      "grad_norm": 2.0250797705190244,
      "language_loss": 0.84929091,
      "learning_rate": 3.5670739904642274e-06,
      "loss": 0.87121361,
      "num_input_tokens_seen": 84871805,
      "step": 3942,
      "time_per_iteration": 2.458131790161133
    },
    {
      "auxiliary_loss_clip": 0.01110135,
      "auxiliary_loss_mlp": 0.01046236,
      "balance_loss_clip": 1.05741894,
      "balance_loss_mlp": 1.02792573,
      "epoch": 0.2370659852698031,
      "flos": 23947605262080.0,
      "grad_norm": 1.8123145761167945,
      "language_loss": 0.81469262,
      "learning_rate": 3.5668319706328065e-06,
      "loss": 0.83625627,
      "num_input_tokens_seen": 84889815,
      "step": 3943,
      "time_per_iteration": 2.6367430686950684
    },
    {
      "auxiliary_loss_clip": 0.01114141,
      "auxiliary_loss_mlp": 0.01041359,
      "balance_loss_clip": 1.05388308,
      "balance_loss_mlp": 1.02393079,
      "epoch": 0.23712610852247107,
      "flos": 15332527670400.0,
      "grad_norm": 8.398771358507812,
      "language_loss": 0.67402655,
      "learning_rate": 3.566589891386959e-06,
      "loss": 0.69558156,
      "num_input_tokens_seen": 84904380,
      "step": 3944,
      "time_per_iteration": 2.5157546997070312
    },
    {
      "auxiliary_loss_clip": 0.01123193,
      "auxiliary_loss_mlp": 0.01044875,
      "balance_loss_clip": 1.05832136,
      "balance_loss_mlp": 1.02681518,
      "epoch": 0.23718623177513903,
      "flos": 19682567233920.0,
      "grad_norm": 1.655459837638198,
      "language_loss": 0.75408065,
      "learning_rate": 3.566347752735866e-06,
      "loss": 0.77576137,
      "num_input_tokens_seen": 84922935,
      "step": 3945,
      "time_per_iteration": 2.5526230335235596
    },
    {
      "auxiliary_loss_clip": 0.01129758,
      "auxiliary_loss_mlp": 0.01042504,
      "balance_loss_clip": 1.05482817,
      "balance_loss_mlp": 1.02566004,
      "epoch": 0.237246355027807,
      "flos": 24973322037120.0,
      "grad_norm": 1.4868559536124195,
      "language_loss": 0.63482881,
      "learning_rate": 3.5661055546887094e-06,
      "loss": 0.65655136,
      "num_input_tokens_seen": 84943685,
      "step": 3946,
      "time_per_iteration": 2.5533552169799805
    },
    {
      "auxiliary_loss_clip": 0.01133242,
      "auxiliary_loss_mlp": 0.01040841,
      "balance_loss_clip": 1.05282092,
      "balance_loss_mlp": 1.02303088,
      "epoch": 0.23730647828047496,
      "flos": 15377416692480.0,
      "grad_norm": 4.513801338709062,
      "language_loss": 0.77118856,
      "learning_rate": 3.5658632972546734e-06,
      "loss": 0.79292941,
      "num_input_tokens_seen": 84959505,
      "step": 3947,
      "time_per_iteration": 2.519551992416382
    },
    {
      "auxiliary_loss_clip": 0.0114485,
      "auxiliary_loss_mlp": 0.01039472,
      "balance_loss_clip": 1.05992222,
      "balance_loss_mlp": 1.02204359,
      "epoch": 0.23736660153314296,
      "flos": 28150662372480.0,
      "grad_norm": 1.573190202833209,
      "language_loss": 0.80595517,
      "learning_rate": 3.565620980442944e-06,
      "loss": 0.82779837,
      "num_input_tokens_seen": 84982130,
      "step": 3948,
      "time_per_iteration": 2.606762409210205
    },
    {
      "auxiliary_loss_clip": 0.01129309,
      "auxiliary_loss_mlp": 0.01042978,
      "balance_loss_clip": 1.05615067,
      "balance_loss_mlp": 1.02593112,
      "epoch": 0.23742672478581092,
      "flos": 22086570729600.0,
      "grad_norm": 1.7539736960923562,
      "language_loss": 0.80401385,
      "learning_rate": 3.5653786042627107e-06,
      "loss": 0.82573676,
      "num_input_tokens_seen": 85000640,
      "step": 3949,
      "time_per_iteration": 2.5637118816375732
    },
    {
      "auxiliary_loss_clip": 0.01120531,
      "auxiliary_loss_mlp": 0.01035791,
      "balance_loss_clip": 1.05336428,
      "balance_loss_mlp": 1.01897025,
      "epoch": 0.2374868480384789,
      "flos": 19537093152000.0,
      "grad_norm": 1.876055983493169,
      "language_loss": 0.72634459,
      "learning_rate": 3.565136168723163e-06,
      "loss": 0.74790788,
      "num_input_tokens_seen": 85018970,
      "step": 3950,
      "time_per_iteration": 2.5554633140563965
    },
    {
      "auxiliary_loss_clip": 0.01146244,
      "auxiliary_loss_mlp": 0.01041176,
      "balance_loss_clip": 1.05509329,
      "balance_loss_mlp": 1.02499962,
      "epoch": 0.23754697129114685,
      "flos": 19422501788160.0,
      "grad_norm": 1.8977149808335352,
      "language_loss": 0.73071659,
      "learning_rate": 3.564893673833495e-06,
      "loss": 0.75259078,
      "num_input_tokens_seen": 85035905,
      "step": 3951,
      "time_per_iteration": 2.505060911178589
    },
    {
      "auxiliary_loss_clip": 0.01121671,
      "auxiliary_loss_mlp": 0.01035646,
      "balance_loss_clip": 1.05608916,
      "balance_loss_mlp": 1.01815772,
      "epoch": 0.23760709454381482,
      "flos": 19501002961920.0,
      "grad_norm": 1.708736064361557,
      "language_loss": 0.73918974,
      "learning_rate": 3.564651119602903e-06,
      "loss": 0.76076287,
      "num_input_tokens_seen": 85054560,
      "step": 3952,
      "time_per_iteration": 2.545238971710205
    },
    {
      "auxiliary_loss_clip": 0.01100061,
      "auxiliary_loss_mlp": 0.01046257,
      "balance_loss_clip": 1.04890037,
      "balance_loss_mlp": 1.02861452,
      "epoch": 0.23766721779648278,
      "flos": 27636600879360.0,
      "grad_norm": 1.613676097310807,
      "language_loss": 0.71019685,
      "learning_rate": 3.564408506040583e-06,
      "loss": 0.73166007,
      "num_input_tokens_seen": 85074425,
      "step": 3953,
      "time_per_iteration": 2.6189072132110596
    },
    {
      "auxiliary_loss_clip": 0.01153195,
      "auxiliary_loss_mlp": 0.01046684,
      "balance_loss_clip": 1.05823743,
      "balance_loss_mlp": 1.02792048,
      "epoch": 0.23772734104915075,
      "flos": 23404348990080.0,
      "grad_norm": 2.04090701906158,
      "language_loss": 0.81270242,
      "learning_rate": 3.5641658331557356e-06,
      "loss": 0.83470124,
      "num_input_tokens_seen": 85092865,
      "step": 3954,
      "time_per_iteration": 2.5225765705108643
    },
    {
      "auxiliary_loss_clip": 0.01129264,
      "auxiliary_loss_mlp": 0.01041099,
      "balance_loss_clip": 1.05720687,
      "balance_loss_mlp": 1.02206087,
      "epoch": 0.23778746430181874,
      "flos": 15705496540800.0,
      "grad_norm": 2.2900810058210026,
      "language_loss": 0.65865612,
      "learning_rate": 3.5639231009575634e-06,
      "loss": 0.68035972,
      "num_input_tokens_seen": 85110175,
      "step": 3955,
      "time_per_iteration": 2.522770643234253
    },
    {
      "auxiliary_loss_clip": 0.01149625,
      "auxiliary_loss_mlp": 0.01053921,
      "balance_loss_clip": 1.05615497,
      "balance_loss_mlp": 1.0367074,
      "epoch": 0.2378475875544867,
      "flos": 19426452284160.0,
      "grad_norm": 1.4747629343234192,
      "language_loss": 0.84014642,
      "learning_rate": 3.5636803094552704e-06,
      "loss": 0.86218184,
      "num_input_tokens_seen": 85129925,
      "step": 3956,
      "time_per_iteration": 2.506817102432251
    },
    {
      "auxiliary_loss_clip": 0.0110464,
      "auxiliary_loss_mlp": 0.01040082,
      "balance_loss_clip": 1.05486178,
      "balance_loss_mlp": 1.02334487,
      "epoch": 0.23790771080715467,
      "flos": 22268565964800.0,
      "grad_norm": 2.1075646781951805,
      "language_loss": 0.84762716,
      "learning_rate": 3.5634374586580635e-06,
      "loss": 0.8690744,
      "num_input_tokens_seen": 85147755,
      "step": 3957,
      "time_per_iteration": 2.5956637859344482
    },
    {
      "auxiliary_loss_clip": 0.01087032,
      "auxiliary_loss_mlp": 0.01040214,
      "balance_loss_clip": 1.05194855,
      "balance_loss_mlp": 1.02419281,
      "epoch": 0.23796783405982264,
      "flos": 20047311889920.0,
      "grad_norm": 2.1316937281406236,
      "language_loss": 0.70316124,
      "learning_rate": 3.563194548575151e-06,
      "loss": 0.72443366,
      "num_input_tokens_seen": 85165270,
      "step": 3958,
      "time_per_iteration": 2.6443376541137695
    },
    {
      "auxiliary_loss_clip": 0.01102296,
      "auxiliary_loss_mlp": 0.01045555,
      "balance_loss_clip": 1.05144227,
      "balance_loss_mlp": 1.02600503,
      "epoch": 0.2380279573124906,
      "flos": 14245943299200.0,
      "grad_norm": 2.5360525781370313,
      "language_loss": 0.66396874,
      "learning_rate": 3.562951579215745e-06,
      "loss": 0.68544722,
      "num_input_tokens_seen": 85181555,
      "step": 3959,
      "time_per_iteration": 2.5484368801116943
    },
    {
      "auxiliary_loss_clip": 0.01108935,
      "auxiliary_loss_mlp": 0.01045555,
      "balance_loss_clip": 1.05629826,
      "balance_loss_mlp": 1.02838922,
      "epoch": 0.23808808056515857,
      "flos": 21179180332800.0,
      "grad_norm": 2.496213390672979,
      "language_loss": 0.72332668,
      "learning_rate": 3.5627085505890586e-06,
      "loss": 0.7448715,
      "num_input_tokens_seen": 85199455,
      "step": 3960,
      "time_per_iteration": 2.6329147815704346
    },
    {
      "auxiliary_loss_clip": 0.01059826,
      "auxiliary_loss_mlp": 0.01040472,
      "balance_loss_clip": 1.05036616,
      "balance_loss_mlp": 1.0229125,
      "epoch": 0.23814820381782653,
      "flos": 22528308188160.0,
      "grad_norm": 3.4008596017174764,
      "language_loss": 0.74073315,
      "learning_rate": 3.562465462704307e-06,
      "loss": 0.76173615,
      "num_input_tokens_seen": 85219170,
      "step": 3961,
      "time_per_iteration": 2.9737634658813477
    },
    {
      "auxiliary_loss_clip": 0.01148111,
      "auxiliary_loss_mlp": 0.01051909,
      "balance_loss_clip": 1.05256987,
      "balance_loss_mlp": 1.03200102,
      "epoch": 0.23820832707049452,
      "flos": 22304332932480.0,
      "grad_norm": 1.8528453161028737,
      "language_loss": 0.66160566,
      "learning_rate": 3.5622223155707085e-06,
      "loss": 0.68360579,
      "num_input_tokens_seen": 85238480,
      "step": 3962,
      "time_per_iteration": 2.7489497661590576
    },
    {
      "auxiliary_loss_clip": 0.01118965,
      "auxiliary_loss_mlp": 0.01046523,
      "balance_loss_clip": 1.04937363,
      "balance_loss_mlp": 1.02945185,
      "epoch": 0.2382684503231625,
      "flos": 24864225454080.0,
      "grad_norm": 1.6462070877065913,
      "language_loss": 0.74941844,
      "learning_rate": 3.561979109197483e-06,
      "loss": 0.77107334,
      "num_input_tokens_seen": 85259180,
      "step": 3963,
      "time_per_iteration": 2.62774920463562
    },
    {
      "auxiliary_loss_clip": 0.01117072,
      "auxiliary_loss_mlp": 0.01045797,
      "balance_loss_clip": 1.05496264,
      "balance_loss_mlp": 1.0284996,
      "epoch": 0.23832857357583045,
      "flos": 21871609787520.0,
      "grad_norm": 1.9815962611336266,
      "language_loss": 0.76946729,
      "learning_rate": 3.5617358435938538e-06,
      "loss": 0.79109597,
      "num_input_tokens_seen": 85278550,
      "step": 3964,
      "time_per_iteration": 2.5967934131622314
    },
    {
      "auxiliary_loss_clip": 0.01110521,
      "auxiliary_loss_mlp": 0.01042956,
      "balance_loss_clip": 1.05301023,
      "balance_loss_mlp": 1.02573013,
      "epoch": 0.23838869682849842,
      "flos": 21288061434240.0,
      "grad_norm": 2.680608052499513,
      "language_loss": 0.71604174,
      "learning_rate": 3.561492518769045e-06,
      "loss": 0.73757648,
      "num_input_tokens_seen": 85297345,
      "step": 3965,
      "time_per_iteration": 2.5965473651885986
    },
    {
      "auxiliary_loss_clip": 0.01114531,
      "auxiliary_loss_mlp": 0.0104336,
      "balance_loss_clip": 1.05380714,
      "balance_loss_mlp": 1.02611041,
      "epoch": 0.23844882008116638,
      "flos": 16180594755840.0,
      "grad_norm": 1.7881388585451299,
      "language_loss": 0.78071404,
      "learning_rate": 3.561249134732282e-06,
      "loss": 0.80229294,
      "num_input_tokens_seen": 85315105,
      "step": 3966,
      "time_per_iteration": 2.595555305480957
    },
    {
      "auxiliary_loss_clip": 0.01119392,
      "auxiliary_loss_mlp": 0.01043908,
      "balance_loss_clip": 1.05082631,
      "balance_loss_mlp": 1.02708793,
      "epoch": 0.23850894333383435,
      "flos": 21069724613760.0,
      "grad_norm": 1.9932713058676443,
      "language_loss": 0.68601161,
      "learning_rate": 3.561005691492797e-06,
      "loss": 0.70764464,
      "num_input_tokens_seen": 85334735,
      "step": 3967,
      "time_per_iteration": 2.5344583988189697
    },
    {
      "auxiliary_loss_clip": 0.01116554,
      "auxiliary_loss_mlp": 0.01050236,
      "balance_loss_clip": 1.05269158,
      "balance_loss_mlp": 1.03185439,
      "epoch": 0.23856906658650234,
      "flos": 17201606849280.0,
      "grad_norm": 1.857153899585276,
      "language_loss": 0.67995822,
      "learning_rate": 3.5607621890598185e-06,
      "loss": 0.70162618,
      "num_input_tokens_seen": 85352875,
      "step": 3968,
      "time_per_iteration": 2.575070381164551
    },
    {
      "auxiliary_loss_clip": 0.0108929,
      "auxiliary_loss_mlp": 0.01042932,
      "balance_loss_clip": 1.04837632,
      "balance_loss_mlp": 1.02618313,
      "epoch": 0.2386291898391703,
      "flos": 29494223619840.0,
      "grad_norm": 1.7939964888525397,
      "language_loss": 0.76515961,
      "learning_rate": 3.5605186274425823e-06,
      "loss": 0.78648186,
      "num_input_tokens_seen": 85372205,
      "step": 3969,
      "time_per_iteration": 2.673492193222046
    },
    {
      "auxiliary_loss_clip": 0.01125796,
      "auxiliary_loss_mlp": 0.01040259,
      "balance_loss_clip": 1.05352545,
      "balance_loss_mlp": 1.0237484,
      "epoch": 0.23868931309183827,
      "flos": 21142443697920.0,
      "grad_norm": 2.0149077588297466,
      "language_loss": 0.76286101,
      "learning_rate": 3.5602750066503225e-06,
      "loss": 0.78452158,
      "num_input_tokens_seen": 85389705,
      "step": 3970,
      "time_per_iteration": 2.5753135681152344
    },
    {
      "auxiliary_loss_clip": 0.01108071,
      "auxiliary_loss_mlp": 0.01050822,
      "balance_loss_clip": 1.05032849,
      "balance_loss_mlp": 1.03272593,
      "epoch": 0.23874943634450624,
      "flos": 25659394784640.0,
      "grad_norm": 1.880230780578347,
      "language_loss": 0.84621048,
      "learning_rate": 3.5600313266922793e-06,
      "loss": 0.86779946,
      "num_input_tokens_seen": 85407855,
      "step": 3971,
      "time_per_iteration": 2.583514451980591
    },
    {
      "auxiliary_loss_clip": 0.0105642,
      "auxiliary_loss_mlp": 0.01020033,
      "balance_loss_clip": 1.03157544,
      "balance_loss_mlp": 1.017959,
      "epoch": 0.2388095595971742,
      "flos": 58986618624000.0,
      "grad_norm": 0.7445835779215032,
      "language_loss": 0.62784207,
      "learning_rate": 3.5597875875776915e-06,
      "loss": 0.64860666,
      "num_input_tokens_seen": 85470885,
      "step": 3972,
      "time_per_iteration": 4.796634197235107
    },
    {
      "auxiliary_loss_clip": 0.01121312,
      "auxiliary_loss_mlp": 0.01039659,
      "balance_loss_clip": 1.05333877,
      "balance_loss_mlp": 1.02275538,
      "epoch": 0.23886968284984217,
      "flos": 16800341040000.0,
      "grad_norm": 2.061079039934424,
      "language_loss": 0.8213827,
      "learning_rate": 3.5595437893158013e-06,
      "loss": 0.84299242,
      "num_input_tokens_seen": 85488460,
      "step": 3973,
      "time_per_iteration": 3.9478888511657715
    },
    {
      "auxiliary_loss_clip": 0.01113116,
      "auxiliary_loss_mlp": 0.010504,
      "balance_loss_clip": 1.05104899,
      "balance_loss_mlp": 1.03207803,
      "epoch": 0.23892980610251013,
      "flos": 22382654538240.0,
      "grad_norm": 1.590800791719403,
      "language_loss": 0.79374391,
      "learning_rate": 3.5592999319158546e-06,
      "loss": 0.81537902,
      "num_input_tokens_seen": 85508590,
      "step": 3974,
      "time_per_iteration": 2.595546245574951
    },
    {
      "auxiliary_loss_clip": 0.0113084,
      "auxiliary_loss_mlp": 0.01046006,
      "balance_loss_clip": 1.05526972,
      "balance_loss_mlp": 1.02750468,
      "epoch": 0.23898992935517813,
      "flos": 12823198519680.0,
      "grad_norm": 5.030532161550108,
      "language_loss": 0.84937775,
      "learning_rate": 3.5590560153870984e-06,
      "loss": 0.8711462,
      "num_input_tokens_seen": 85525970,
      "step": 3975,
      "time_per_iteration": 2.5252346992492676
    },
    {
      "auxiliary_loss_clip": 0.01125148,
      "auxiliary_loss_mlp": 0.01047652,
      "balance_loss_clip": 1.05138302,
      "balance_loss_mlp": 1.03055716,
      "epoch": 0.2390500526078461,
      "flos": 22345666508160.0,
      "grad_norm": 2.656566288042889,
      "language_loss": 0.83713007,
      "learning_rate": 3.5588120397387816e-06,
      "loss": 0.85885811,
      "num_input_tokens_seen": 85543700,
      "step": 3976,
      "time_per_iteration": 2.5289478302001953
    },
    {
      "auxiliary_loss_clip": 0.01073264,
      "auxiliary_loss_mlp": 0.01033323,
      "balance_loss_clip": 1.05000925,
      "balance_loss_mlp": 1.01777828,
      "epoch": 0.23911017586051406,
      "flos": 22635142214400.0,
      "grad_norm": 1.8123498407703187,
      "language_loss": 0.74369967,
      "learning_rate": 3.5585680049801566e-06,
      "loss": 0.76476556,
      "num_input_tokens_seen": 85562765,
      "step": 3977,
      "time_per_iteration": 2.683412551879883
    },
    {
      "auxiliary_loss_clip": 0.01146041,
      "auxiliary_loss_mlp": 0.01048486,
      "balance_loss_clip": 1.05231786,
      "balance_loss_mlp": 1.02975798,
      "epoch": 0.23917029911318202,
      "flos": 23653281219840.0,
      "grad_norm": 1.7846829962029425,
      "language_loss": 0.72064781,
      "learning_rate": 3.5583239111204764e-06,
      "loss": 0.74259317,
      "num_input_tokens_seen": 85581755,
      "step": 3978,
      "time_per_iteration": 5.251256465911865
    },
    {
      "auxiliary_loss_clip": 0.01123007,
      "auxiliary_loss_mlp": 0.01049753,
      "balance_loss_clip": 1.05084777,
      "balance_loss_mlp": 1.03174043,
      "epoch": 0.23923042236585,
      "flos": 22783597125120.0,
      "grad_norm": 2.390737430637487,
      "language_loss": 0.78904504,
      "learning_rate": 3.558079758168997e-06,
      "loss": 0.81077266,
      "num_input_tokens_seen": 85599455,
      "step": 3979,
      "time_per_iteration": 2.536644220352173
    },
    {
      "auxiliary_loss_clip": 0.01121063,
      "auxiliary_loss_mlp": 0.01053439,
      "balance_loss_clip": 1.04953194,
      "balance_loss_mlp": 1.03448439,
      "epoch": 0.23929054561851795,
      "flos": 28147717457280.0,
      "grad_norm": 1.6173414134200046,
      "language_loss": 0.81639117,
      "learning_rate": 3.557835546134977e-06,
      "loss": 0.8381362,
      "num_input_tokens_seen": 85619970,
      "step": 3980,
      "time_per_iteration": 2.606170654296875
    },
    {
      "auxiliary_loss_clip": 0.0109376,
      "auxiliary_loss_mlp": 0.01039104,
      "balance_loss_clip": 1.05324113,
      "balance_loss_mlp": 1.02167594,
      "epoch": 0.23935066887118592,
      "flos": 21686525982720.0,
      "grad_norm": 1.928465144549385,
      "language_loss": 0.83992606,
      "learning_rate": 3.5575912750276775e-06,
      "loss": 0.86125469,
      "num_input_tokens_seen": 85638850,
      "step": 3981,
      "time_per_iteration": 2.5904626846313477
    },
    {
      "auxiliary_loss_clip": 0.01122951,
      "auxiliary_loss_mlp": 0.01046126,
      "balance_loss_clip": 1.05135012,
      "balance_loss_mlp": 1.02787542,
      "epoch": 0.2394107921238539,
      "flos": 32122274198400.0,
      "grad_norm": 1.7854685389408194,
      "language_loss": 0.76793194,
      "learning_rate": 3.5573469448563607e-06,
      "loss": 0.78962278,
      "num_input_tokens_seen": 85656285,
      "step": 3982,
      "time_per_iteration": 2.6381239891052246
    },
    {
      "auxiliary_loss_clip": 0.01108236,
      "auxiliary_loss_mlp": 0.01043153,
      "balance_loss_clip": 1.05375719,
      "balance_loss_mlp": 1.02628481,
      "epoch": 0.23947091537652188,
      "flos": 17019180650880.0,
      "grad_norm": 1.6924185059848798,
      "language_loss": 0.78154534,
      "learning_rate": 3.5571025556302915e-06,
      "loss": 0.80305922,
      "num_input_tokens_seen": 85673020,
      "step": 3983,
      "time_per_iteration": 2.583866834640503
    },
    {
      "auxiliary_loss_clip": 0.0112964,
      "auxiliary_loss_mlp": 0.00796828,
      "balance_loss_clip": 1.04988647,
      "balance_loss_mlp": 1.01839876,
      "epoch": 0.23953103862918984,
      "flos": 20593584904320.0,
      "grad_norm": 1.8534213558903507,
      "language_loss": 0.735425,
      "learning_rate": 3.556858107358737e-06,
      "loss": 0.75468969,
      "num_input_tokens_seen": 85692565,
      "step": 3984,
      "time_per_iteration": 2.525561809539795
    },
    {
      "auxiliary_loss_clip": 0.01096018,
      "auxiliary_loss_mlp": 0.0104866,
      "balance_loss_clip": 1.04637766,
      "balance_loss_mlp": 1.02978873,
      "epoch": 0.2395911618818578,
      "flos": 20704405340160.0,
      "grad_norm": 2.100088824649878,
      "language_loss": 0.79106486,
      "learning_rate": 3.5566136000509674e-06,
      "loss": 0.81251168,
      "num_input_tokens_seen": 85709730,
      "step": 3985,
      "time_per_iteration": 2.5827388763427734
    },
    {
      "auxiliary_loss_clip": 0.0109926,
      "auxiliary_loss_mlp": 0.01050179,
      "balance_loss_clip": 1.04866457,
      "balance_loss_mlp": 1.03183222,
      "epoch": 0.23965128513452577,
      "flos": 27053519402880.0,
      "grad_norm": 1.926566711464529,
      "language_loss": 0.73119307,
      "learning_rate": 3.556369033716254e-06,
      "loss": 0.75268745,
      "num_input_tokens_seen": 85730045,
      "step": 3986,
      "time_per_iteration": 2.647474765777588
    },
    {
      "auxiliary_loss_clip": 0.01135635,
      "auxiliary_loss_mlp": 0.01046966,
      "balance_loss_clip": 1.05052495,
      "balance_loss_mlp": 1.02983558,
      "epoch": 0.23971140838719374,
      "flos": 23144319457920.0,
      "grad_norm": 1.838263713942384,
      "language_loss": 0.87676966,
      "learning_rate": 3.556124408363871e-06,
      "loss": 0.89859557,
      "num_input_tokens_seen": 85747590,
      "step": 3987,
      "time_per_iteration": 2.535912275314331
    },
    {
      "auxiliary_loss_clip": 0.01124002,
      "auxiliary_loss_mlp": 0.01044331,
      "balance_loss_clip": 1.04817116,
      "balance_loss_mlp": 1.02847576,
      "epoch": 0.23977153163986173,
      "flos": 18034554309120.0,
      "grad_norm": 2.754663722620775,
      "language_loss": 0.83041346,
      "learning_rate": 3.5558797240030945e-06,
      "loss": 0.8520968,
      "num_input_tokens_seen": 85763460,
      "step": 3988,
      "time_per_iteration": 2.50909161567688
    },
    {
      "auxiliary_loss_clip": 0.01130129,
      "auxiliary_loss_mlp": 0.01042058,
      "balance_loss_clip": 1.05050838,
      "balance_loss_mlp": 1.02414083,
      "epoch": 0.2398316548925297,
      "flos": 18113378705280.0,
      "grad_norm": 1.6467212333622046,
      "language_loss": 0.85133708,
      "learning_rate": 3.5556349806432035e-06,
      "loss": 0.87305892,
      "num_input_tokens_seen": 85782050,
      "step": 3989,
      "time_per_iteration": 2.5063977241516113
    },
    {
      "auxiliary_loss_clip": 0.01139613,
      "auxiliary_loss_mlp": 0.010406,
      "balance_loss_clip": 1.0482645,
      "balance_loss_mlp": 1.02360058,
      "epoch": 0.23989177814519766,
      "flos": 12567730014720.0,
      "grad_norm": 2.1014965764309927,
      "language_loss": 0.84938669,
      "learning_rate": 3.555390178293477e-06,
      "loss": 0.87118882,
      "num_input_tokens_seen": 85797400,
      "step": 3990,
      "time_per_iteration": 2.467935800552368
    },
    {
      "auxiliary_loss_clip": 0.01126497,
      "auxiliary_loss_mlp": 0.01040871,
      "balance_loss_clip": 1.04710507,
      "balance_loss_mlp": 1.02455068,
      "epoch": 0.23995190139786562,
      "flos": 25264593423360.0,
      "grad_norm": 1.4217277974439448,
      "language_loss": 0.7575894,
      "learning_rate": 3.5551453169631994e-06,
      "loss": 0.77926302,
      "num_input_tokens_seen": 85818995,
      "step": 3991,
      "time_per_iteration": 2.6114790439605713
    },
    {
      "auxiliary_loss_clip": 0.01041511,
      "auxiliary_loss_mlp": 0.01009693,
      "balance_loss_clip": 1.03682637,
      "balance_loss_mlp": 1.00698686,
      "epoch": 0.2400120246505336,
      "flos": 61960379650560.0,
      "grad_norm": 0.9105204939209784,
      "language_loss": 0.63676965,
      "learning_rate": 3.554900396661656e-06,
      "loss": 0.6572817,
      "num_input_tokens_seen": 85876695,
      "step": 3992,
      "time_per_iteration": 3.058872938156128
    },
    {
      "auxiliary_loss_clip": 0.01048008,
      "auxiliary_loss_mlp": 0.01012253,
      "balance_loss_clip": 1.02257085,
      "balance_loss_mlp": 1.00989234,
      "epoch": 0.24007214790320155,
      "flos": 66708560540160.0,
      "grad_norm": 0.7584167552123384,
      "language_loss": 0.62989819,
      "learning_rate": 3.5546554173981334e-06,
      "loss": 0.65050083,
      "num_input_tokens_seen": 85940990,
      "step": 3993,
      "time_per_iteration": 3.191061496734619
    },
    {
      "auxiliary_loss_clip": 0.01105097,
      "auxiliary_loss_mlp": 0.01046402,
      "balance_loss_clip": 1.05188179,
      "balance_loss_mlp": 1.02844906,
      "epoch": 0.24013227115586952,
      "flos": 25809070757760.0,
      "grad_norm": 1.7762757403255023,
      "language_loss": 0.76852882,
      "learning_rate": 3.5544103791819218e-06,
      "loss": 0.79004377,
      "num_input_tokens_seen": 85961165,
      "step": 3994,
      "time_per_iteration": 2.614126205444336
    },
    {
      "auxiliary_loss_clip": 0.01117814,
      "auxiliary_loss_mlp": 0.01046809,
      "balance_loss_clip": 1.04799819,
      "balance_loss_mlp": 1.02786684,
      "epoch": 0.2401923944085375,
      "flos": 25557480921600.0,
      "grad_norm": 1.5583007865951757,
      "language_loss": 0.78603429,
      "learning_rate": 3.5541652820223124e-06,
      "loss": 0.80768055,
      "num_input_tokens_seen": 85982710,
      "step": 3995,
      "time_per_iteration": 2.610560178756714
    },
    {
      "auxiliary_loss_clip": 0.01028891,
      "auxiliary_loss_mlp": 0.0100644,
      "balance_loss_clip": 1.02648234,
      "balance_loss_mlp": 1.00379372,
      "epoch": 0.24025251766120548,
      "flos": 54941138478720.0,
      "grad_norm": 1.0258291369588604,
      "language_loss": 0.63426208,
      "learning_rate": 3.5539201259286006e-06,
      "loss": 0.6546154,
      "num_input_tokens_seen": 86046935,
      "step": 3996,
      "time_per_iteration": 3.216639280319214
    },
    {
      "auxiliary_loss_clip": 0.0112158,
      "auxiliary_loss_mlp": 0.01042444,
      "balance_loss_clip": 1.05151033,
      "balance_loss_mlp": 1.02509856,
      "epoch": 0.24031264091387344,
      "flos": 20631075724800.0,
      "grad_norm": 2.8416082661475457,
      "language_loss": 0.69843972,
      "learning_rate": 3.5536749109100808e-06,
      "loss": 0.72007996,
      "num_input_tokens_seen": 86064355,
      "step": 3997,
      "time_per_iteration": 2.5486128330230713
    },
    {
      "auxiliary_loss_clip": 0.01131272,
      "auxiliary_loss_mlp": 0.01045812,
      "balance_loss_clip": 1.04793882,
      "balance_loss_mlp": 1.02838349,
      "epoch": 0.2403727641665414,
      "flos": 20886256920960.0,
      "grad_norm": 1.7538254988423188,
      "language_loss": 0.87081623,
      "learning_rate": 3.5534296369760535e-06,
      "loss": 0.89258707,
      "num_input_tokens_seen": 86081340,
      "step": 3998,
      "time_per_iteration": 2.5194218158721924
    },
    {
      "auxiliary_loss_clip": 0.01123567,
      "auxiliary_loss_mlp": 0.01036504,
      "balance_loss_clip": 1.04552388,
      "balance_loss_mlp": 1.01900411,
      "epoch": 0.24043288741920937,
      "flos": 22820046451200.0,
      "grad_norm": 1.532200940516686,
      "language_loss": 0.75738704,
      "learning_rate": 3.5531843041358183e-06,
      "loss": 0.77898777,
      "num_input_tokens_seen": 86102260,
      "step": 3999,
      "time_per_iteration": 2.5257408618927
    },
    {
      "auxiliary_loss_clip": 0.01118399,
      "auxiliary_loss_mlp": 0.01045206,
      "balance_loss_clip": 1.04893708,
      "balance_loss_mlp": 1.02818251,
      "epoch": 0.24049301067187734,
      "flos": 27959652823680.0,
      "grad_norm": 2.17245677645293,
      "language_loss": 0.7202577,
      "learning_rate": 3.552938912398679e-06,
      "loss": 0.74189377,
      "num_input_tokens_seen": 86123400,
      "step": 4000,
      "time_per_iteration": 2.6110410690307617
    },
    {
      "auxiliary_loss_clip": 0.01133737,
      "auxiliary_loss_mlp": 0.0104128,
      "balance_loss_clip": 1.05354083,
      "balance_loss_mlp": 1.02380371,
      "epoch": 0.24055313392454533,
      "flos": 27451409333760.0,
      "grad_norm": 1.8111866156153111,
      "language_loss": 0.67109108,
      "learning_rate": 3.5526934617739397e-06,
      "loss": 0.69284129,
      "num_input_tokens_seen": 86144060,
      "step": 4001,
      "time_per_iteration": 2.640087127685547
    },
    {
      "auxiliary_loss_clip": 0.011416,
      "auxiliary_loss_mlp": 0.01042102,
      "balance_loss_clip": 1.04832053,
      "balance_loss_mlp": 1.02420866,
      "epoch": 0.2406132571772133,
      "flos": 25556618995200.0,
      "grad_norm": 2.999521716088347,
      "language_loss": 0.8269186,
      "learning_rate": 3.5524479522709095e-06,
      "loss": 0.8487556,
      "num_input_tokens_seen": 86163005,
      "step": 4002,
      "time_per_iteration": 2.623082399368286
    },
    {
      "auxiliary_loss_clip": 0.01105717,
      "auxiliary_loss_mlp": 0.01045616,
      "balance_loss_clip": 1.05172229,
      "balance_loss_mlp": 1.02812767,
      "epoch": 0.24067338042988126,
      "flos": 24791398629120.0,
      "grad_norm": 1.832958449229557,
      "language_loss": 0.83070654,
      "learning_rate": 3.552202383898897e-06,
      "loss": 0.85221988,
      "num_input_tokens_seen": 86182580,
      "step": 4003,
      "time_per_iteration": 2.6036951541900635
    },
    {
      "auxiliary_loss_clip": 0.01114653,
      "auxiliary_loss_mlp": 0.01046296,
      "balance_loss_clip": 1.05497539,
      "balance_loss_mlp": 1.02833104,
      "epoch": 0.24073350368254923,
      "flos": 21177923356800.0,
      "grad_norm": 2.3594217118290124,
      "language_loss": 0.8724547,
      "learning_rate": 3.551956756667215e-06,
      "loss": 0.89406419,
      "num_input_tokens_seen": 86200665,
      "step": 4004,
      "time_per_iteration": 2.588740587234497
    },
    {
      "auxiliary_loss_clip": 0.01112984,
      "auxiliary_loss_mlp": 0.01055246,
      "balance_loss_clip": 1.04746306,
      "balance_loss_mlp": 1.03773427,
      "epoch": 0.2407936269352172,
      "flos": 22494300986880.0,
      "grad_norm": 1.9481871846160752,
      "language_loss": 0.78054601,
      "learning_rate": 3.551711070585177e-06,
      "loss": 0.80222833,
      "num_input_tokens_seen": 86221640,
      "step": 4005,
      "time_per_iteration": 2.6189208030700684
    },
    {
      "auxiliary_loss_clip": 0.01093523,
      "auxiliary_loss_mlp": 0.01041805,
      "balance_loss_clip": 1.0552001,
      "balance_loss_mlp": 1.02361369,
      "epoch": 0.24085375018788516,
      "flos": 18551129754240.0,
      "grad_norm": 1.759248228701162,
      "language_loss": 0.78972691,
      "learning_rate": 3.5514653256620995e-06,
      "loss": 0.81108022,
      "num_input_tokens_seen": 86240795,
      "step": 4006,
      "time_per_iteration": 2.617008686065674
    },
    {
      "auxiliary_loss_clip": 0.01125168,
      "auxiliary_loss_mlp": 0.00802104,
      "balance_loss_clip": 1.04796863,
      "balance_loss_mlp": 1.02371728,
      "epoch": 0.24091387344055312,
      "flos": 24170539023360.0,
      "grad_norm": 1.8154514527029975,
      "language_loss": 0.71510124,
      "learning_rate": 3.551219521907302e-06,
      "loss": 0.73437405,
      "num_input_tokens_seen": 86262000,
      "step": 4007,
      "time_per_iteration": 2.6243021488189697
    },
    {
      "auxiliary_loss_clip": 0.01099673,
      "auxiliary_loss_mlp": 0.01051721,
      "balance_loss_clip": 1.04841959,
      "balance_loss_mlp": 1.03467393,
      "epoch": 0.24097399669322112,
      "flos": 11036319615360.0,
      "grad_norm": 1.7124084580579062,
      "language_loss": 0.76213837,
      "learning_rate": 3.5509736593301042e-06,
      "loss": 0.78365231,
      "num_input_tokens_seen": 86279680,
      "step": 4008,
      "time_per_iteration": 2.5774104595184326
    },
    {
      "auxiliary_loss_clip": 0.01132232,
      "auxiliary_loss_mlp": 0.0103462,
      "balance_loss_clip": 1.05102634,
      "balance_loss_mlp": 1.01781178,
      "epoch": 0.24103411994588908,
      "flos": 17165085696000.0,
      "grad_norm": 3.103374244293146,
      "language_loss": 0.74867499,
      "learning_rate": 3.5507277379398295e-06,
      "loss": 0.77034354,
      "num_input_tokens_seen": 86297180,
      "step": 4009,
      "time_per_iteration": 2.50140380859375
    },
    {
      "auxiliary_loss_clip": 0.01134969,
      "auxiliary_loss_mlp": 0.01042157,
      "balance_loss_clip": 1.05507278,
      "balance_loss_mlp": 1.02599192,
      "epoch": 0.24109424319855705,
      "flos": 20667956014080.0,
      "grad_norm": 2.41954597874374,
      "language_loss": 0.8038913,
      "learning_rate": 3.550481757745804e-06,
      "loss": 0.82566249,
      "num_input_tokens_seen": 86317660,
      "step": 4010,
      "time_per_iteration": 4.076771974563599
    },
    {
      "auxiliary_loss_clip": 0.01112707,
      "auxiliary_loss_mlp": 0.01047437,
      "balance_loss_clip": 1.0464623,
      "balance_loss_mlp": 1.02810097,
      "epoch": 0.241154366451225,
      "flos": 28181796485760.0,
      "grad_norm": 1.8691953210459238,
      "language_loss": 0.70688415,
      "learning_rate": 3.5502357187573555e-06,
      "loss": 0.72848558,
      "num_input_tokens_seen": 86338325,
      "step": 4011,
      "time_per_iteration": 2.614694118499756
    },
    {
      "auxiliary_loss_clip": 0.0106501,
      "auxiliary_loss_mlp": 0.01045721,
      "balance_loss_clip": 1.04901385,
      "balance_loss_mlp": 1.0272795,
      "epoch": 0.24121448970389298,
      "flos": 21689722293120.0,
      "grad_norm": 1.547327249856818,
      "language_loss": 0.69206476,
      "learning_rate": 3.5499896209838118e-06,
      "loss": 0.71317208,
      "num_input_tokens_seen": 86357615,
      "step": 4012,
      "time_per_iteration": 4.117403268814087
    },
    {
      "auxiliary_loss_clip": 0.01136374,
      "auxiliary_loss_mlp": 0.01039232,
      "balance_loss_clip": 1.05388105,
      "balance_loss_mlp": 1.02054036,
      "epoch": 0.24127461295656094,
      "flos": 39676191269760.0,
      "grad_norm": 2.546662546725643,
      "language_loss": 0.73377073,
      "learning_rate": 3.5497434644345073e-06,
      "loss": 0.75552678,
      "num_input_tokens_seen": 86380355,
      "step": 4013,
      "time_per_iteration": 2.7572360038757324
    },
    {
      "auxiliary_loss_clip": 0.01145021,
      "auxiliary_loss_mlp": 0.01036892,
      "balance_loss_clip": 1.05215931,
      "balance_loss_mlp": 1.0200243,
      "epoch": 0.2413347362092289,
      "flos": 19135863256320.0,
      "grad_norm": 1.8766334035211751,
      "language_loss": 0.88072759,
      "learning_rate": 3.5494972491187753e-06,
      "loss": 0.90254664,
      "num_input_tokens_seen": 86399125,
      "step": 4014,
      "time_per_iteration": 2.584608316421509
    },
    {
      "auxiliary_loss_clip": 0.01112331,
      "auxiliary_loss_mlp": 0.01045714,
      "balance_loss_clip": 1.04684353,
      "balance_loss_mlp": 1.02728415,
      "epoch": 0.2413948594618969,
      "flos": 26939430829440.0,
      "grad_norm": 3.963401747993656,
      "language_loss": 0.94899535,
      "learning_rate": 3.549250975045952e-06,
      "loss": 0.97057581,
      "num_input_tokens_seen": 86418625,
      "step": 4015,
      "time_per_iteration": 2.6074790954589844
    },
    {
      "auxiliary_loss_clip": 0.01118639,
      "auxiliary_loss_mlp": 0.01041119,
      "balance_loss_clip": 1.04913712,
      "balance_loss_mlp": 1.0239768,
      "epoch": 0.24145498271456486,
      "flos": 25228108183680.0,
      "grad_norm": 1.5974701962007962,
      "language_loss": 0.83014596,
      "learning_rate": 3.5490046422253768e-06,
      "loss": 0.85174346,
      "num_input_tokens_seen": 86438375,
      "step": 4016,
      "time_per_iteration": 3.981218099594116
    },
    {
      "auxiliary_loss_clip": 0.01098541,
      "auxiliary_loss_mlp": 0.01045398,
      "balance_loss_clip": 1.04932272,
      "balance_loss_mlp": 1.02787423,
      "epoch": 0.24151510596723283,
      "flos": 40661759617920.0,
      "grad_norm": 1.7830423585222162,
      "language_loss": 0.69282365,
      "learning_rate": 3.54875825066639e-06,
      "loss": 0.71426308,
      "num_input_tokens_seen": 86463230,
      "step": 4017,
      "time_per_iteration": 4.1247382164001465
    },
    {
      "auxiliary_loss_clip": 0.01136839,
      "auxiliary_loss_mlp": 0.01050121,
      "balance_loss_clip": 1.05055165,
      "balance_loss_mlp": 1.03202534,
      "epoch": 0.2415752292199008,
      "flos": 18146667634560.0,
      "grad_norm": 1.6593854001122648,
      "language_loss": 0.85084385,
      "learning_rate": 3.5485118003783353e-06,
      "loss": 0.87271345,
      "num_input_tokens_seen": 86481230,
      "step": 4018,
      "time_per_iteration": 2.4732067584991455
    },
    {
      "auxiliary_loss_clip": 0.0104496,
      "auxiliary_loss_mlp": 0.01027712,
      "balance_loss_clip": 1.02184987,
      "balance_loss_mlp": 1.02508914,
      "epoch": 0.24163535247256876,
      "flos": 67288409792640.0,
      "grad_norm": 0.8923993929372823,
      "language_loss": 0.60657042,
      "learning_rate": 3.548265291370558e-06,
      "loss": 0.6272971,
      "num_input_tokens_seen": 86541260,
      "step": 4019,
      "time_per_iteration": 3.180785894393921
    },
    {
      "auxiliary_loss_clip": 0.01114539,
      "auxiliary_loss_mlp": 0.0104233,
      "balance_loss_clip": 1.04849195,
      "balance_loss_mlp": 1.02413881,
      "epoch": 0.24169547572523672,
      "flos": 24929941386240.0,
      "grad_norm": 1.815298460670813,
      "language_loss": 0.72977412,
      "learning_rate": 3.5480187236524055e-06,
      "loss": 0.75134283,
      "num_input_tokens_seen": 86559580,
      "step": 4020,
      "time_per_iteration": 2.5616390705108643
    },
    {
      "auxiliary_loss_clip": 0.01109041,
      "auxiliary_loss_mlp": 0.01043272,
      "balance_loss_clip": 1.05300891,
      "balance_loss_mlp": 1.02651095,
      "epoch": 0.24175559897790472,
      "flos": 18728312567040.0,
      "grad_norm": 2.0251892274081134,
      "language_loss": 0.8179847,
      "learning_rate": 3.5477720972332285e-06,
      "loss": 0.83950788,
      "num_input_tokens_seen": 86577560,
      "step": 4021,
      "time_per_iteration": 2.6048834323883057
    },
    {
      "auxiliary_loss_clip": 0.01146623,
      "auxiliary_loss_mlp": 0.01044586,
      "balance_loss_clip": 1.05097163,
      "balance_loss_mlp": 1.02570295,
      "epoch": 0.24181572223057268,
      "flos": 23039281111680.0,
      "grad_norm": 3.275335023822472,
      "language_loss": 0.76536536,
      "learning_rate": 3.547525412122378e-06,
      "loss": 0.78727746,
      "num_input_tokens_seen": 86595350,
      "step": 4022,
      "time_per_iteration": 2.5886471271514893
    },
    {
      "auxiliary_loss_clip": 0.01102041,
      "auxiliary_loss_mlp": 0.01055398,
      "balance_loss_clip": 1.04630566,
      "balance_loss_mlp": 1.03450108,
      "epoch": 0.24187584548324065,
      "flos": 20376145923840.0,
      "grad_norm": 3.08554944544278,
      "language_loss": 0.7518698,
      "learning_rate": 3.5472786683292083e-06,
      "loss": 0.77344418,
      "num_input_tokens_seen": 86614805,
      "step": 4023,
      "time_per_iteration": 2.647296190261841
    },
    {
      "auxiliary_loss_clip": 0.01116421,
      "auxiliary_loss_mlp": 0.0104822,
      "balance_loss_clip": 1.04828644,
      "balance_loss_mlp": 1.03141165,
      "epoch": 0.2419359687359086,
      "flos": 21397517153280.0,
      "grad_norm": 1.8786778732475815,
      "language_loss": 0.82211012,
      "learning_rate": 3.5470318658630766e-06,
      "loss": 0.84375656,
      "num_input_tokens_seen": 86633700,
      "step": 4024,
      "time_per_iteration": 2.5474495887756348
    },
    {
      "auxiliary_loss_clip": 0.01130209,
      "auxiliary_loss_mlp": 0.01048,
      "balance_loss_clip": 1.05083799,
      "balance_loss_mlp": 1.03023815,
      "epoch": 0.24199609198857658,
      "flos": 18369385914240.0,
      "grad_norm": 1.8363904911614037,
      "language_loss": 0.86203313,
      "learning_rate": 3.5467850047333424e-06,
      "loss": 0.88381511,
      "num_input_tokens_seen": 86650905,
      "step": 4025,
      "time_per_iteration": 2.5017287731170654
    },
    {
      "auxiliary_loss_clip": 0.01088591,
      "auxiliary_loss_mlp": 0.01060871,
      "balance_loss_clip": 1.04398167,
      "balance_loss_mlp": 1.03986657,
      "epoch": 0.24205621524124454,
      "flos": 19463871277440.0,
      "grad_norm": 2.291313034439703,
      "language_loss": 0.71767557,
      "learning_rate": 3.546538084949365e-06,
      "loss": 0.73917019,
      "num_input_tokens_seen": 86669185,
      "step": 4026,
      "time_per_iteration": 2.5372469425201416
    },
    {
      "auxiliary_loss_clip": 0.01130568,
      "auxiliary_loss_mlp": 0.01044265,
      "balance_loss_clip": 1.04912758,
      "balance_loss_mlp": 1.02765906,
      "epoch": 0.2421163384939125,
      "flos": 14976330451200.0,
      "grad_norm": 1.932385070122097,
      "language_loss": 0.64375663,
      "learning_rate": 3.546291106520509e-06,
      "loss": 0.66550493,
      "num_input_tokens_seen": 86686805,
      "step": 4027,
      "time_per_iteration": 2.497695207595825
    },
    {
      "auxiliary_loss_clip": 0.01130882,
      "auxiliary_loss_mlp": 0.00796493,
      "balance_loss_clip": 1.05233431,
      "balance_loss_mlp": 1.01803195,
      "epoch": 0.2421764617465805,
      "flos": 18662057930880.0,
      "grad_norm": 2.0747926514283717,
      "language_loss": 0.70681512,
      "learning_rate": 3.5460440694561388e-06,
      "loss": 0.72608888,
      "num_input_tokens_seen": 86705520,
      "step": 4028,
      "time_per_iteration": 2.4857304096221924
    },
    {
      "auxiliary_loss_clip": 0.01047859,
      "auxiliary_loss_mlp": 0.01002631,
      "balance_loss_clip": 1.02258825,
      "balance_loss_mlp": 0.99987727,
      "epoch": 0.24223658499924847,
      "flos": 64347327164160.0,
      "grad_norm": 0.8596866466051839,
      "language_loss": 0.55335605,
      "learning_rate": 3.545796973765623e-06,
      "loss": 0.57386094,
      "num_input_tokens_seen": 86767320,
      "step": 4029,
      "time_per_iteration": 3.1139018535614014
    },
    {
      "auxiliary_loss_clip": 0.01128465,
      "auxiliary_loss_mlp": 0.01043135,
      "balance_loss_clip": 1.04789257,
      "balance_loss_mlp": 1.02534914,
      "epoch": 0.24229670825191643,
      "flos": 25775243124480.0,
      "grad_norm": 2.0070647139598807,
      "language_loss": 0.74112433,
      "learning_rate": 3.54554981945833e-06,
      "loss": 0.76284033,
      "num_input_tokens_seen": 86788110,
      "step": 4030,
      "time_per_iteration": 2.5590341091156006
    },
    {
      "auxiliary_loss_clip": 0.01143192,
      "auxiliary_loss_mlp": 0.01047872,
      "balance_loss_clip": 1.05018294,
      "balance_loss_mlp": 1.03051519,
      "epoch": 0.2423568315045844,
      "flos": 20667094087680.0,
      "grad_norm": 2.308389388341383,
      "language_loss": 0.76584744,
      "learning_rate": 3.5453026065436343e-06,
      "loss": 0.78775811,
      "num_input_tokens_seen": 86807640,
      "step": 4031,
      "time_per_iteration": 2.477060556411743
    },
    {
      "auxiliary_loss_clip": 0.01126788,
      "auxiliary_loss_mlp": 0.00794528,
      "balance_loss_clip": 1.05056179,
      "balance_loss_mlp": 1.01542914,
      "epoch": 0.24241695475725236,
      "flos": 22416805393920.0,
      "grad_norm": 2.0302274532160993,
      "language_loss": 0.65499628,
      "learning_rate": 3.5450553350309083e-06,
      "loss": 0.67420948,
      "num_input_tokens_seen": 86826795,
      "step": 4032,
      "time_per_iteration": 2.539369821548462
    },
    {
      "auxiliary_loss_clip": 0.01128533,
      "auxiliary_loss_mlp": 0.0104903,
      "balance_loss_clip": 1.04847717,
      "balance_loss_mlp": 1.03108931,
      "epoch": 0.24247707800992033,
      "flos": 17128995505920.0,
      "grad_norm": 2.0138608647254097,
      "language_loss": 0.81748688,
      "learning_rate": 3.5448080049295286e-06,
      "loss": 0.83926249,
      "num_input_tokens_seen": 86843175,
      "step": 4033,
      "time_per_iteration": 2.47478985786438
    },
    {
      "auxiliary_loss_clip": 0.01101277,
      "auxiliary_loss_mlp": 0.01040001,
      "balance_loss_clip": 1.04493499,
      "balance_loss_mlp": 1.02263224,
      "epoch": 0.2425372012625883,
      "flos": 31613743399680.0,
      "grad_norm": 1.9399888389233693,
      "language_loss": 0.69009948,
      "learning_rate": 3.5445606162488754e-06,
      "loss": 0.71151227,
      "num_input_tokens_seen": 86863185,
      "step": 4034,
      "time_per_iteration": 2.6520516872406006
    },
    {
      "auxiliary_loss_clip": 0.01121607,
      "auxiliary_loss_mlp": 0.01035449,
      "balance_loss_clip": 1.05028987,
      "balance_loss_mlp": 1.01788962,
      "epoch": 0.24259732451525629,
      "flos": 16326032924160.0,
      "grad_norm": 2.257829122812066,
      "language_loss": 0.96469712,
      "learning_rate": 3.5443131689983283e-06,
      "loss": 0.98626763,
      "num_input_tokens_seen": 86880040,
      "step": 4035,
      "time_per_iteration": 2.524733066558838
    },
    {
      "auxiliary_loss_clip": 0.01109954,
      "auxiliary_loss_mlp": 0.01043292,
      "balance_loss_clip": 1.0469228,
      "balance_loss_mlp": 1.02697241,
      "epoch": 0.24265744776792425,
      "flos": 22856639431680.0,
      "grad_norm": 1.7216986085237018,
      "language_loss": 0.77888983,
      "learning_rate": 3.5440656631872715e-06,
      "loss": 0.80042225,
      "num_input_tokens_seen": 86900610,
      "step": 4036,
      "time_per_iteration": 2.594829797744751
    },
    {
      "auxiliary_loss_clip": 0.01131105,
      "auxiliary_loss_mlp": 0.01045309,
      "balance_loss_clip": 1.0494417,
      "balance_loss_mlp": 1.02780914,
      "epoch": 0.24271757102059222,
      "flos": 21871573873920.0,
      "grad_norm": 1.660775597199867,
      "language_loss": 0.74402273,
      "learning_rate": 3.5438180988250898e-06,
      "loss": 0.76578683,
      "num_input_tokens_seen": 86919385,
      "step": 4037,
      "time_per_iteration": 2.5148277282714844
    },
    {
      "auxiliary_loss_clip": 0.01100191,
      "auxiliary_loss_mlp": 0.01044572,
      "balance_loss_clip": 1.04295444,
      "balance_loss_mlp": 1.02609491,
      "epoch": 0.24277769427326018,
      "flos": 19208582340480.0,
      "grad_norm": 2.5828722719838333,
      "language_loss": 0.76590365,
      "learning_rate": 3.543570475921171e-06,
      "loss": 0.78735125,
      "num_input_tokens_seen": 86938885,
      "step": 4038,
      "time_per_iteration": 2.56931471824646
    },
    {
      "auxiliary_loss_clip": 0.01120516,
      "auxiliary_loss_mlp": 0.01043683,
      "balance_loss_clip": 1.04519093,
      "balance_loss_mlp": 1.02588558,
      "epoch": 0.24283781752592815,
      "flos": 19499889640320.0,
      "grad_norm": 1.8754543105885366,
      "language_loss": 0.719172,
      "learning_rate": 3.543322794484905e-06,
      "loss": 0.74081397,
      "num_input_tokens_seen": 86957705,
      "step": 4039,
      "time_per_iteration": 2.5069429874420166
    },
    {
      "auxiliary_loss_clip": 0.01122641,
      "auxiliary_loss_mlp": 0.01047942,
      "balance_loss_clip": 1.04563642,
      "balance_loss_mlp": 1.02970314,
      "epoch": 0.2428979407785961,
      "flos": 19902196944000.0,
      "grad_norm": 1.6458444554386125,
      "language_loss": 0.78198588,
      "learning_rate": 3.5430750545256843e-06,
      "loss": 0.80369174,
      "num_input_tokens_seen": 86975845,
      "step": 4040,
      "time_per_iteration": 2.532717704772949
    },
    {
      "auxiliary_loss_clip": 0.0108711,
      "auxiliary_loss_mlp": 0.01038578,
      "balance_loss_clip": 1.04925156,
      "balance_loss_mlp": 1.0232476,
      "epoch": 0.2429580640312641,
      "flos": 24715878284160.0,
      "grad_norm": 1.745798873066938,
      "language_loss": 0.80524886,
      "learning_rate": 3.5428272560529027e-06,
      "loss": 0.82650578,
      "num_input_tokens_seen": 86994800,
      "step": 4041,
      "time_per_iteration": 2.6474568843841553
    },
    {
      "auxiliary_loss_clip": 0.01102459,
      "auxiliary_loss_mlp": 0.01047221,
      "balance_loss_clip": 1.04431391,
      "balance_loss_mlp": 1.03077054,
      "epoch": 0.24301818728393207,
      "flos": 25630343660160.0,
      "grad_norm": 2.3407593573194627,
      "language_loss": 0.77206039,
      "learning_rate": 3.542579399075957e-06,
      "loss": 0.79355717,
      "num_input_tokens_seen": 87016845,
      "step": 4042,
      "time_per_iteration": 2.6184520721435547
    },
    {
      "auxiliary_loss_clip": 0.01057189,
      "auxiliary_loss_mlp": 0.01034831,
      "balance_loss_clip": 1.04875016,
      "balance_loss_mlp": 1.01901221,
      "epoch": 0.24307831053660003,
      "flos": 26141388410880.0,
      "grad_norm": 1.7904267078337692,
      "language_loss": 0.81468761,
      "learning_rate": 3.542331483604246e-06,
      "loss": 0.83560789,
      "num_input_tokens_seen": 87036270,
      "step": 4043,
      "time_per_iteration": 2.9986908435821533
    },
    {
      "auxiliary_loss_clip": 0.01120177,
      "auxiliary_loss_mlp": 0.01035618,
      "balance_loss_clip": 1.04476559,
      "balance_loss_mlp": 1.01786816,
      "epoch": 0.243138433789268,
      "flos": 14972415868800.0,
      "grad_norm": 2.538662212436332,
      "language_loss": 0.73247063,
      "learning_rate": 3.5420835096471706e-06,
      "loss": 0.75402856,
      "num_input_tokens_seen": 87049920,
      "step": 4044,
      "time_per_iteration": 2.6385445594787598
    },
    {
      "auxiliary_loss_clip": 0.0113084,
      "auxiliary_loss_mlp": 0.01043514,
      "balance_loss_clip": 1.0503329,
      "balance_loss_mlp": 1.02607369,
      "epoch": 0.24319855704193596,
      "flos": 25191694771200.0,
      "grad_norm": 1.7314338749369147,
      "language_loss": 0.8323797,
      "learning_rate": 3.5418354772141337e-06,
      "loss": 0.85412323,
      "num_input_tokens_seen": 87068230,
      "step": 4045,
      "time_per_iteration": 2.548265218734741
    },
    {
      "auxiliary_loss_clip": 0.01069807,
      "auxiliary_loss_mlp": 0.01046837,
      "balance_loss_clip": 1.04905891,
      "balance_loss_mlp": 1.02946854,
      "epoch": 0.24325868029460393,
      "flos": 22127221946880.0,
      "grad_norm": 1.6540377309510772,
      "language_loss": 0.86619663,
      "learning_rate": 3.541587386314541e-06,
      "loss": 0.88736308,
      "num_input_tokens_seen": 87086435,
      "step": 4046,
      "time_per_iteration": 2.665343999862671
    },
    {
      "auxiliary_loss_clip": 0.01113428,
      "auxiliary_loss_mlp": 0.01040588,
      "balance_loss_clip": 1.04386544,
      "balance_loss_mlp": 1.02295732,
      "epoch": 0.2433188035472719,
      "flos": 23582106420480.0,
      "grad_norm": 2.8255504732731103,
      "language_loss": 0.72904557,
      "learning_rate": 3.5413392369578e-06,
      "loss": 0.75058573,
      "num_input_tokens_seen": 87105340,
      "step": 4047,
      "time_per_iteration": 2.5578064918518066
    },
    {
      "auxiliary_loss_clip": 0.0111971,
      "auxiliary_loss_mlp": 0.01044445,
      "balance_loss_clip": 1.04478192,
      "balance_loss_mlp": 1.02528834,
      "epoch": 0.2433789267999399,
      "flos": 24462815990400.0,
      "grad_norm": 1.9350315978084784,
      "language_loss": 0.73341578,
      "learning_rate": 3.5410910291533213e-06,
      "loss": 0.75505728,
      "num_input_tokens_seen": 87125780,
      "step": 4048,
      "time_per_iteration": 3.9934537410736084
    },
    {
      "auxiliary_loss_clip": 0.01105718,
      "auxiliary_loss_mlp": 0.0104604,
      "balance_loss_clip": 1.04889297,
      "balance_loss_mlp": 1.02975655,
      "epoch": 0.24343905005260785,
      "flos": 16727909264640.0,
      "grad_norm": 2.0939950210655387,
      "language_loss": 0.73073602,
      "learning_rate": 3.5408427629105155e-06,
      "loss": 0.75225365,
      "num_input_tokens_seen": 87144470,
      "step": 4049,
      "time_per_iteration": 2.582179069519043
    },
    {
      "auxiliary_loss_clip": 0.01094697,
      "auxiliary_loss_mlp": 0.01045239,
      "balance_loss_clip": 1.0494746,
      "balance_loss_mlp": 1.0281086,
      "epoch": 0.24349917330527582,
      "flos": 20043756443520.0,
      "grad_norm": 1.6451586132693121,
      "language_loss": 0.73533195,
      "learning_rate": 3.5405944382387985e-06,
      "loss": 0.75673133,
      "num_input_tokens_seen": 87162830,
      "step": 4050,
      "time_per_iteration": 2.5592751502990723
    },
    {
      "auxiliary_loss_clip": 0.01112302,
      "auxiliary_loss_mlp": 0.01039314,
      "balance_loss_clip": 1.04630661,
      "balance_loss_mlp": 1.02400756,
      "epoch": 0.24355929655794378,
      "flos": 17420554200960.0,
      "grad_norm": 3.552410221398559,
      "language_loss": 0.75445408,
      "learning_rate": 3.5403460551475854e-06,
      "loss": 0.77597028,
      "num_input_tokens_seen": 87180905,
      "step": 4051,
      "time_per_iteration": 4.261563539505005
    },
    {
      "auxiliary_loss_clip": 0.01088596,
      "auxiliary_loss_mlp": 0.01040579,
      "balance_loss_clip": 1.04402041,
      "balance_loss_mlp": 1.02309132,
      "epoch": 0.24361941981061175,
      "flos": 25410929431680.0,
      "grad_norm": 1.964121489417537,
      "language_loss": 0.70954454,
      "learning_rate": 3.540097613646296e-06,
      "loss": 0.73083627,
      "num_input_tokens_seen": 87202290,
      "step": 4052,
      "time_per_iteration": 2.6766762733459473
    },
    {
      "auxiliary_loss_clip": 0.01111934,
      "auxiliary_loss_mlp": 0.01047906,
      "balance_loss_clip": 1.04940224,
      "balance_loss_mlp": 1.03091884,
      "epoch": 0.2436795430632797,
      "flos": 22820800636800.0,
      "grad_norm": 2.0452220700371297,
      "language_loss": 0.80944443,
      "learning_rate": 3.539849113744351e-06,
      "loss": 0.83104277,
      "num_input_tokens_seen": 87221650,
      "step": 4053,
      "time_per_iteration": 2.546811819076538
    },
    {
      "auxiliary_loss_clip": 0.011419,
      "auxiliary_loss_mlp": 0.01033731,
      "balance_loss_clip": 1.04839897,
      "balance_loss_mlp": 1.0166961,
      "epoch": 0.2437396663159477,
      "flos": 15157786982400.0,
      "grad_norm": 1.486131170278907,
      "language_loss": 0.78313071,
      "learning_rate": 3.539600555451172e-06,
      "loss": 0.80488706,
      "num_input_tokens_seen": 87238515,
      "step": 4054,
      "time_per_iteration": 3.9412338733673096
    },
    {
      "auxiliary_loss_clip": 0.01092707,
      "auxiliary_loss_mlp": 0.01050961,
      "balance_loss_clip": 1.04194045,
      "balance_loss_mlp": 1.03441453,
      "epoch": 0.24379978956861567,
      "flos": 22091131756800.0,
      "grad_norm": 1.7563621562144864,
      "language_loss": 0.83513236,
      "learning_rate": 3.5393519387761866e-06,
      "loss": 0.85656905,
      "num_input_tokens_seen": 87256290,
      "step": 4055,
      "time_per_iteration": 2.584815263748169
    },
    {
      "auxiliary_loss_clip": 0.01107682,
      "auxiliary_loss_mlp": 0.01042001,
      "balance_loss_clip": 1.0457263,
      "balance_loss_mlp": 1.02473927,
      "epoch": 0.24385991282128364,
      "flos": 31467766527360.0,
      "grad_norm": 2.578589499630782,
      "language_loss": 0.54837275,
      "learning_rate": 3.5391032637288217e-06,
      "loss": 0.56986952,
      "num_input_tokens_seen": 87277085,
      "step": 4056,
      "time_per_iteration": 4.116791486740112
    },
    {
      "auxiliary_loss_clip": 0.01131829,
      "auxiliary_loss_mlp": 0.01045558,
      "balance_loss_clip": 1.04693413,
      "balance_loss_mlp": 1.02851152,
      "epoch": 0.2439200360739516,
      "flos": 23838795987840.0,
      "grad_norm": 2.3497757278599054,
      "language_loss": 0.80489063,
      "learning_rate": 3.538854530318506e-06,
      "loss": 0.82666445,
      "num_input_tokens_seen": 87293020,
      "step": 4057,
      "time_per_iteration": 2.555865526199341
    },
    {
      "auxiliary_loss_clip": 0.01129609,
      "auxiliary_loss_mlp": 0.01042933,
      "balance_loss_clip": 1.04953766,
      "balance_loss_mlp": 1.02571964,
      "epoch": 0.24398015932661957,
      "flos": 19169978198400.0,
      "grad_norm": 1.857725121557648,
      "language_loss": 0.79824817,
      "learning_rate": 3.538605738554673e-06,
      "loss": 0.81997359,
      "num_input_tokens_seen": 87311445,
      "step": 4058,
      "time_per_iteration": 2.508112907409668
    },
    {
      "auxiliary_loss_clip": 0.01143594,
      "auxiliary_loss_mlp": 0.01044896,
      "balance_loss_clip": 1.04854465,
      "balance_loss_mlp": 1.02873147,
      "epoch": 0.24404028257928753,
      "flos": 25262474520960.0,
      "grad_norm": 1.6853755783741657,
      "language_loss": 0.85349518,
      "learning_rate": 3.538356888446756e-06,
      "loss": 0.8753801,
      "num_input_tokens_seen": 87332055,
      "step": 4059,
      "time_per_iteration": 2.5445706844329834
    },
    {
      "auxiliary_loss_clip": 0.01119939,
      "auxiliary_loss_mlp": 0.01040853,
      "balance_loss_clip": 1.04820955,
      "balance_loss_mlp": 1.02480769,
      "epoch": 0.2441004058319555,
      "flos": 26467600752000.0,
      "grad_norm": 1.5792654594178297,
      "language_loss": 0.74005854,
      "learning_rate": 3.5381079800041913e-06,
      "loss": 0.76166648,
      "num_input_tokens_seen": 87351295,
      "step": 4060,
      "time_per_iteration": 2.533621311187744
    },
    {
      "auxiliary_loss_clip": 0.01111929,
      "auxiliary_loss_mlp": 0.01051804,
      "balance_loss_clip": 1.05061626,
      "balance_loss_mlp": 1.032444,
      "epoch": 0.2441605290846235,
      "flos": 26760524163840.0,
      "grad_norm": 1.6863196326097842,
      "language_loss": 0.73267519,
      "learning_rate": 3.5378590132364182e-06,
      "loss": 0.75431252,
      "num_input_tokens_seen": 87370650,
      "step": 4061,
      "time_per_iteration": 2.6412577629089355
    },
    {
      "auxiliary_loss_clip": 0.01141652,
      "auxiliary_loss_mlp": 0.01043399,
      "balance_loss_clip": 1.0519197,
      "balance_loss_mlp": 1.02722239,
      "epoch": 0.24422065233729146,
      "flos": 21105850717440.0,
      "grad_norm": 1.6919656136647816,
      "language_loss": 0.75998342,
      "learning_rate": 3.5376099881528768e-06,
      "loss": 0.78183395,
      "num_input_tokens_seen": 87389020,
      "step": 4062,
      "time_per_iteration": 2.489513635635376
    },
    {
      "auxiliary_loss_clip": 0.010969,
      "auxiliary_loss_mlp": 0.01040973,
      "balance_loss_clip": 1.04417419,
      "balance_loss_mlp": 1.0239265,
      "epoch": 0.24428077558995942,
      "flos": 25263156879360.0,
      "grad_norm": 2.8224101995053212,
      "language_loss": 0.85319197,
      "learning_rate": 3.537360904763011e-06,
      "loss": 0.87457067,
      "num_input_tokens_seen": 87409695,
      "step": 4063,
      "time_per_iteration": 2.6377546787261963
    },
    {
      "auxiliary_loss_clip": 0.01115781,
      "auxiliary_loss_mlp": 0.0104536,
      "balance_loss_clip": 1.04801059,
      "balance_loss_mlp": 1.02638149,
      "epoch": 0.24434089884262739,
      "flos": 20485278420480.0,
      "grad_norm": 2.330424838905176,
      "language_loss": 0.68731987,
      "learning_rate": 3.5371117630762656e-06,
      "loss": 0.70893133,
      "num_input_tokens_seen": 87428250,
      "step": 4064,
      "time_per_iteration": 2.5396385192871094
    },
    {
      "auxiliary_loss_clip": 0.01131381,
      "auxiliary_loss_mlp": 0.01037037,
      "balance_loss_clip": 1.04562259,
      "balance_loss_mlp": 1.01938248,
      "epoch": 0.24440102209529535,
      "flos": 23621895711360.0,
      "grad_norm": 1.5977548026969297,
      "language_loss": 0.6986832,
      "learning_rate": 3.536862563102088e-06,
      "loss": 0.72036737,
      "num_input_tokens_seen": 87449380,
      "step": 4065,
      "time_per_iteration": 2.5446460247039795
    },
    {
      "auxiliary_loss_clip": 0.01144215,
      "auxiliary_loss_mlp": 0.01050484,
      "balance_loss_clip": 1.04831016,
      "balance_loss_mlp": 1.03099322,
      "epoch": 0.24446114534796332,
      "flos": 20554729367040.0,
      "grad_norm": 1.8400148046646443,
      "language_loss": 0.84147298,
      "learning_rate": 3.5366133048499282e-06,
      "loss": 0.86342001,
      "num_input_tokens_seen": 87465365,
      "step": 4066,
      "time_per_iteration": 2.465567111968994
    },
    {
      "auxiliary_loss_clip": 0.01054569,
      "auxiliary_loss_mlp": 0.01006142,
      "balance_loss_clip": 1.0221839,
      "balance_loss_mlp": 1.00363851,
      "epoch": 0.24452126860063128,
      "flos": 60389575009920.0,
      "grad_norm": 0.735897067806973,
      "language_loss": 0.52292848,
      "learning_rate": 3.5363639883292374e-06,
      "loss": 0.54353559,
      "num_input_tokens_seen": 87522525,
      "step": 4067,
      "time_per_iteration": 2.9932448863983154
    },
    {
      "auxiliary_loss_clip": 0.01115961,
      "auxiliary_loss_mlp": 0.01043709,
      "balance_loss_clip": 1.04977655,
      "balance_loss_mlp": 1.02629256,
      "epoch": 0.24458139185329927,
      "flos": 15121660878720.0,
      "grad_norm": 5.6492349172965355,
      "language_loss": 0.72757614,
      "learning_rate": 3.5361146135494706e-06,
      "loss": 0.74917281,
      "num_input_tokens_seen": 87539170,
      "step": 4068,
      "time_per_iteration": 2.5147202014923096
    },
    {
      "auxiliary_loss_clip": 0.01081586,
      "auxiliary_loss_mlp": 0.01049372,
      "balance_loss_clip": 1.04547429,
      "balance_loss_mlp": 1.03122818,
      "epoch": 0.24464151510596724,
      "flos": 27998723842560.0,
      "grad_norm": 1.4615960965478452,
      "language_loss": 0.77617294,
      "learning_rate": 3.5358651805200835e-06,
      "loss": 0.79748249,
      "num_input_tokens_seen": 87558875,
      "step": 4069,
      "time_per_iteration": 2.658329725265503
    },
    {
      "auxiliary_loss_clip": 0.01109904,
      "auxiliary_loss_mlp": 0.01046352,
      "balance_loss_clip": 1.04953909,
      "balance_loss_mlp": 1.02866125,
      "epoch": 0.2447016383586352,
      "flos": 19792884879360.0,
      "grad_norm": 1.9226532412656518,
      "language_loss": 0.80536497,
      "learning_rate": 3.5356156892505347e-06,
      "loss": 0.82692748,
      "num_input_tokens_seen": 87576485,
      "step": 4070,
      "time_per_iteration": 2.513995885848999
    },
    {
      "auxiliary_loss_clip": 0.01115447,
      "auxiliary_loss_mlp": 0.01044847,
      "balance_loss_clip": 1.04194021,
      "balance_loss_mlp": 1.02782345,
      "epoch": 0.24476176161130317,
      "flos": 26067340523520.0,
      "grad_norm": 1.4795252715687568,
      "language_loss": 0.84157377,
      "learning_rate": 3.5353661397502854e-06,
      "loss": 0.8631767,
      "num_input_tokens_seen": 87598620,
      "step": 4071,
      "time_per_iteration": 2.5597803592681885
    },
    {
      "auxiliary_loss_clip": 0.01111101,
      "auxiliary_loss_mlp": 0.01058603,
      "balance_loss_clip": 1.04430556,
      "balance_loss_mlp": 1.03756237,
      "epoch": 0.24482188486397113,
      "flos": 18843550375680.0,
      "grad_norm": 3.848192691463926,
      "language_loss": 0.79799104,
      "learning_rate": 3.535116532028798e-06,
      "loss": 0.81968808,
      "num_input_tokens_seen": 87616595,
      "step": 4072,
      "time_per_iteration": 2.5026051998138428
    },
    {
      "auxiliary_loss_clip": 0.01125416,
      "auxiliary_loss_mlp": 0.01042288,
      "balance_loss_clip": 1.0474143,
      "balance_loss_mlp": 1.02625394,
      "epoch": 0.2448820081166391,
      "flos": 21251791676160.0,
      "grad_norm": 1.8016006957344795,
      "language_loss": 0.70453238,
      "learning_rate": 3.5348668660955382e-06,
      "loss": 0.7262094,
      "num_input_tokens_seen": 87635755,
      "step": 4073,
      "time_per_iteration": 2.5232815742492676
    },
    {
      "auxiliary_loss_clip": 0.01104844,
      "auxiliary_loss_mlp": 0.01044772,
      "balance_loss_clip": 1.04434156,
      "balance_loss_mlp": 1.02796352,
      "epoch": 0.2449421313693071,
      "flos": 23950586090880.0,
      "grad_norm": 2.2536831441202634,
      "language_loss": 0.67355007,
      "learning_rate": 3.5346171419599728e-06,
      "loss": 0.69504625,
      "num_input_tokens_seen": 87652885,
      "step": 4074,
      "time_per_iteration": 2.559296131134033
    },
    {
      "auxiliary_loss_clip": 0.01052857,
      "auxiliary_loss_mlp": 0.01002022,
      "balance_loss_clip": 1.02016497,
      "balance_loss_mlp": 0.99955481,
      "epoch": 0.24500225462197506,
      "flos": 60687669980160.0,
      "grad_norm": 0.9704253833200763,
      "language_loss": 0.68699533,
      "learning_rate": 3.5343673596315718e-06,
      "loss": 0.70754409,
      "num_input_tokens_seen": 87713220,
      "step": 4075,
      "time_per_iteration": 3.171764850616455
    },
    {
      "auxiliary_loss_clip": 0.01137022,
      "auxiliary_loss_mlp": 0.01039808,
      "balance_loss_clip": 1.04742277,
      "balance_loss_mlp": 1.02258253,
      "epoch": 0.24506237787464302,
      "flos": 26284204886400.0,
      "grad_norm": 1.9466558058527834,
      "language_loss": 0.79569173,
      "learning_rate": 3.5341175191198063e-06,
      "loss": 0.81746,
      "num_input_tokens_seen": 87732680,
      "step": 4076,
      "time_per_iteration": 2.5383574962615967
    },
    {
      "auxiliary_loss_clip": 0.01117743,
      "auxiliary_loss_mlp": 0.00799091,
      "balance_loss_clip": 1.04427421,
      "balance_loss_mlp": 1.01950848,
      "epoch": 0.245122501127311,
      "flos": 20552287242240.0,
      "grad_norm": 1.9212466110137036,
      "language_loss": 0.82518983,
      "learning_rate": 3.533867620434151e-06,
      "loss": 0.84435821,
      "num_input_tokens_seen": 87751880,
      "step": 4077,
      "time_per_iteration": 2.530695676803589
    },
    {
      "auxiliary_loss_clip": 0.01139553,
      "auxiliary_loss_mlp": 0.01050621,
      "balance_loss_clip": 1.04718626,
      "balance_loss_mlp": 1.03211927,
      "epoch": 0.24518262437997895,
      "flos": 29132603447040.0,
      "grad_norm": 2.119799103636231,
      "language_loss": 0.61810875,
      "learning_rate": 3.533617663584082e-06,
      "loss": 0.64001048,
      "num_input_tokens_seen": 87771795,
      "step": 4078,
      "time_per_iteration": 2.584671974182129
    },
    {
      "auxiliary_loss_clip": 0.01106838,
      "auxiliary_loss_mlp": 0.01038334,
      "balance_loss_clip": 1.0468055,
      "balance_loss_mlp": 1.02164483,
      "epoch": 0.24524274763264692,
      "flos": 23476924419840.0,
      "grad_norm": 1.541652739336377,
      "language_loss": 0.7580663,
      "learning_rate": 3.5333676485790765e-06,
      "loss": 0.77951795,
      "num_input_tokens_seen": 87793640,
      "step": 4079,
      "time_per_iteration": 2.600055456161499
    },
    {
      "auxiliary_loss_clip": 0.01135982,
      "auxiliary_loss_mlp": 0.01048022,
      "balance_loss_clip": 1.04581106,
      "balance_loss_mlp": 1.02946115,
      "epoch": 0.24530287088531488,
      "flos": 17201175886080.0,
      "grad_norm": 1.9531432142201848,
      "language_loss": 0.75416845,
      "learning_rate": 3.5331175754286173e-06,
      "loss": 0.77600849,
      "num_input_tokens_seen": 87812390,
      "step": 4080,
      "time_per_iteration": 2.4901764392852783
    },
    {
      "auxiliary_loss_clip": 0.01110847,
      "auxiliary_loss_mlp": 0.01047747,
      "balance_loss_clip": 1.04658723,
      "balance_loss_mlp": 1.03007984,
      "epoch": 0.24536299413798288,
      "flos": 14867449349760.0,
      "grad_norm": 1.8816592413859812,
      "language_loss": 0.8332175,
      "learning_rate": 3.532867444142186e-06,
      "loss": 0.85480344,
      "num_input_tokens_seen": 87830640,
      "step": 4081,
      "time_per_iteration": 2.522614002227783
    },
    {
      "auxiliary_loss_clip": 0.01111238,
      "auxiliary_loss_mlp": 0.01040632,
      "balance_loss_clip": 1.04979253,
      "balance_loss_mlp": 1.02455103,
      "epoch": 0.24542311739065084,
      "flos": 35262051886080.0,
      "grad_norm": 2.2200492557024103,
      "language_loss": 0.73291588,
      "learning_rate": 3.532617254729267e-06,
      "loss": 0.75443459,
      "num_input_tokens_seen": 87850450,
      "step": 4082,
      "time_per_iteration": 2.6994357109069824
    },
    {
      "auxiliary_loss_clip": 0.01098353,
      "auxiliary_loss_mlp": 0.01042265,
      "balance_loss_clip": 1.04277372,
      "balance_loss_mlp": 1.02613568,
      "epoch": 0.2454832406433188,
      "flos": 21503130117120.0,
      "grad_norm": 2.008526572057144,
      "language_loss": 0.71873218,
      "learning_rate": 3.5323670071993485e-06,
      "loss": 0.74013841,
      "num_input_tokens_seen": 87868810,
      "step": 4083,
      "time_per_iteration": 2.5573463439941406
    },
    {
      "auxiliary_loss_clip": 0.01114321,
      "auxiliary_loss_mlp": 0.0104478,
      "balance_loss_clip": 1.04943562,
      "balance_loss_mlp": 1.02573085,
      "epoch": 0.24554336389598677,
      "flos": 14756664827520.0,
      "grad_norm": 2.3963013424583175,
      "language_loss": 0.74696505,
      "learning_rate": 3.532116701561919e-06,
      "loss": 0.768556,
      "num_input_tokens_seen": 87885685,
      "step": 4084,
      "time_per_iteration": 2.5822641849517822
    },
    {
      "auxiliary_loss_clip": 0.01121863,
      "auxiliary_loss_mlp": 0.01038712,
      "balance_loss_clip": 1.0455904,
      "balance_loss_mlp": 1.02102089,
      "epoch": 0.24560348714865474,
      "flos": 14976402278400.0,
      "grad_norm": 1.837033736313879,
      "language_loss": 0.853522,
      "learning_rate": 3.531866337826471e-06,
      "loss": 0.87512779,
      "num_input_tokens_seen": 87903715,
      "step": 4085,
      "time_per_iteration": 2.4841411113739014
    },
    {
      "auxiliary_loss_clip": 0.01106379,
      "auxiliary_loss_mlp": 0.01050924,
      "balance_loss_clip": 1.04965615,
      "balance_loss_mlp": 1.03422308,
      "epoch": 0.2456636104013227,
      "flos": 22675326554880.0,
      "grad_norm": 2.17373593527973,
      "language_loss": 0.7919153,
      "learning_rate": 3.5316159160024982e-06,
      "loss": 0.81348836,
      "num_input_tokens_seen": 87923375,
      "step": 4086,
      "time_per_iteration": 2.5669467449188232
    },
    {
      "auxiliary_loss_clip": 0.01084263,
      "auxiliary_loss_mlp": 0.01041736,
      "balance_loss_clip": 1.05082965,
      "balance_loss_mlp": 1.02468944,
      "epoch": 0.2457237336539907,
      "flos": 27417869009280.0,
      "grad_norm": 1.9327504568533138,
      "language_loss": 0.74810493,
      "learning_rate": 3.531365436099496e-06,
      "loss": 0.76936489,
      "num_input_tokens_seen": 87943115,
      "step": 4087,
      "time_per_iteration": 2.6562256813049316
    },
    {
      "auxiliary_loss_clip": 0.01080461,
      "auxiliary_loss_mlp": 0.01041709,
      "balance_loss_clip": 1.04579067,
      "balance_loss_mlp": 1.02373266,
      "epoch": 0.24578385690665866,
      "flos": 20412379768320.0,
      "grad_norm": 2.301462338837411,
      "language_loss": 0.79594231,
      "learning_rate": 3.5311148981269635e-06,
      "loss": 0.81716394,
      "num_input_tokens_seen": 87959505,
      "step": 4088,
      "time_per_iteration": 4.188084602355957
    },
    {
      "auxiliary_loss_clip": 0.01095449,
      "auxiliary_loss_mlp": 0.01032387,
      "balance_loss_clip": 1.04577804,
      "balance_loss_mlp": 1.01668727,
      "epoch": 0.24584398015932662,
      "flos": 23915393740800.0,
      "grad_norm": 2.7966761098355954,
      "language_loss": 0.77486593,
      "learning_rate": 3.5308643020944e-06,
      "loss": 0.79614425,
      "num_input_tokens_seen": 87979725,
      "step": 4089,
      "time_per_iteration": 4.057018518447876
    },
    {
      "auxiliary_loss_clip": 0.01117572,
      "auxiliary_loss_mlp": 0.01045765,
      "balance_loss_clip": 1.04462266,
      "balance_loss_mlp": 1.02884936,
      "epoch": 0.2459041034119946,
      "flos": 41496359103360.0,
      "grad_norm": 2.4218106904578476,
      "language_loss": 0.81431168,
      "learning_rate": 3.530613648011309e-06,
      "loss": 0.83594513,
      "num_input_tokens_seen": 87998270,
      "step": 4090,
      "time_per_iteration": 2.7036917209625244
    },
    {
      "auxiliary_loss_clip": 0.01116544,
      "auxiliary_loss_mlp": 0.01044538,
      "balance_loss_clip": 1.04748869,
      "balance_loss_mlp": 1.02670467,
      "epoch": 0.24596422666466256,
      "flos": 19936814676480.0,
      "grad_norm": 1.6863877575491022,
      "language_loss": 0.73427761,
      "learning_rate": 3.5303629358871946e-06,
      "loss": 0.75588834,
      "num_input_tokens_seen": 88016760,
      "step": 4091,
      "time_per_iteration": 2.564976692199707
    },
    {
      "auxiliary_loss_clip": 0.01105134,
      "auxiliary_loss_mlp": 0.01047622,
      "balance_loss_clip": 1.05047441,
      "balance_loss_mlp": 1.03002667,
      "epoch": 0.24602434991733052,
      "flos": 21544391865600.0,
      "grad_norm": 1.9259740970569186,
      "language_loss": 0.76788026,
      "learning_rate": 3.5301121657315653e-06,
      "loss": 0.78940785,
      "num_input_tokens_seen": 88036465,
      "step": 4092,
      "time_per_iteration": 2.602738618850708
    },
    {
      "auxiliary_loss_clip": 0.01111087,
      "auxiliary_loss_mlp": 0.01038944,
      "balance_loss_clip": 1.04835844,
      "balance_loss_mlp": 1.02093172,
      "epoch": 0.24608447316999849,
      "flos": 23185078416000.0,
      "grad_norm": 2.6535042907297384,
      "language_loss": 0.81777549,
      "learning_rate": 3.5298613375539287e-06,
      "loss": 0.83927578,
      "num_input_tokens_seen": 88053270,
      "step": 4093,
      "time_per_iteration": 2.565294027328491
    },
    {
      "auxiliary_loss_clip": 0.0112937,
      "auxiliary_loss_mlp": 0.01046147,
      "balance_loss_clip": 1.04659247,
      "balance_loss_mlp": 1.02832544,
      "epoch": 0.24614459642266648,
      "flos": 19641951930240.0,
      "grad_norm": 2.763278392957555,
      "language_loss": 0.87075579,
      "learning_rate": 3.529610451363797e-06,
      "loss": 0.89251095,
      "num_input_tokens_seen": 88072305,
      "step": 4094,
      "time_per_iteration": 5.4436821937561035
    },
    {
      "auxiliary_loss_clip": 0.01005841,
      "auxiliary_loss_mlp": 0.01022095,
      "balance_loss_clip": 1.03511453,
      "balance_loss_mlp": 1.01962698,
      "epoch": 0.24620471967533444,
      "flos": 61739816186880.0,
      "grad_norm": 0.761623042357955,
      "language_loss": 0.57535207,
      "learning_rate": 3.5293595071706833e-06,
      "loss": 0.59563142,
      "num_input_tokens_seen": 88137995,
      "step": 4095,
      "time_per_iteration": 3.380734920501709
    },
    {
      "auxiliary_loss_clip": 0.01045027,
      "auxiliary_loss_mlp": 0.01020194,
      "balance_loss_clip": 1.02984977,
      "balance_loss_mlp": 1.01764345,
      "epoch": 0.2462648429280024,
      "flos": 69154436315520.0,
      "grad_norm": 0.6468204870944155,
      "language_loss": 0.56254905,
      "learning_rate": 3.5291085049841042e-06,
      "loss": 0.58320123,
      "num_input_tokens_seen": 88208490,
      "step": 4096,
      "time_per_iteration": 3.386735439300537
    },
    {
      "auxiliary_loss_clip": 0.01124918,
      "auxiliary_loss_mlp": 0.01037995,
      "balance_loss_clip": 1.05461454,
      "balance_loss_mlp": 1.02146661,
      "epoch": 0.24632496618067037,
      "flos": 29459605887360.0,
      "grad_norm": 1.78537155839086,
      "language_loss": 0.77265137,
      "learning_rate": 3.5288574448135773e-06,
      "loss": 0.79428053,
      "num_input_tokens_seen": 88228050,
      "step": 4097,
      "time_per_iteration": 2.62400484085083
    },
    {
      "auxiliary_loss_clip": 0.01107265,
      "auxiliary_loss_mlp": 0.01047272,
      "balance_loss_clip": 1.0481807,
      "balance_loss_mlp": 1.02796006,
      "epoch": 0.24638508943333834,
      "flos": 24316444068480.0,
      "grad_norm": 5.137470019199943,
      "language_loss": 0.76399124,
      "learning_rate": 3.5286063266686235e-06,
      "loss": 0.78553665,
      "num_input_tokens_seen": 88248090,
      "step": 4098,
      "time_per_iteration": 2.634230375289917
    },
    {
      "auxiliary_loss_clip": 0.01119,
      "auxiliary_loss_mlp": 0.01042201,
      "balance_loss_clip": 1.04897356,
      "balance_loss_mlp": 1.02609622,
      "epoch": 0.2464452126860063,
      "flos": 26613254401920.0,
      "grad_norm": 2.05460511762004,
      "language_loss": 0.6777097,
      "learning_rate": 3.528355150558764e-06,
      "loss": 0.69932169,
      "num_input_tokens_seen": 88267545,
      "step": 4099,
      "time_per_iteration": 2.6085078716278076
    },
    {
      "auxiliary_loss_clip": 0.01121189,
      "auxiliary_loss_mlp": 0.01040258,
      "balance_loss_clip": 1.04561186,
      "balance_loss_mlp": 1.02422416,
      "epoch": 0.24650533593867427,
      "flos": 31212405763200.0,
      "grad_norm": 2.5931873230484506,
      "language_loss": 0.6583665,
      "learning_rate": 3.5281039164935237e-06,
      "loss": 0.67998099,
      "num_input_tokens_seen": 88289785,
      "step": 4100,
      "time_per_iteration": 2.6501948833465576
    },
    {
      "auxiliary_loss_clip": 0.01037679,
      "auxiliary_loss_mlp": 0.01011164,
      "balance_loss_clip": 1.0251236,
      "balance_loss_mlp": 1.00838697,
      "epoch": 0.24656545919134226,
      "flos": 68494002900480.0,
      "grad_norm": 0.7163803846660457,
      "language_loss": 0.6150279,
      "learning_rate": 3.5278526244824304e-06,
      "loss": 0.63551641,
      "num_input_tokens_seen": 88357320,
      "step": 4101,
      "time_per_iteration": 3.335144281387329
    },
    {
      "auxiliary_loss_clip": 0.01133851,
      "auxiliary_loss_mlp": 0.01040431,
      "balance_loss_clip": 1.04597139,
      "balance_loss_mlp": 1.02371776,
      "epoch": 0.24662558244401023,
      "flos": 20084192179200.0,
      "grad_norm": 1.6621046723805768,
      "language_loss": 0.73425752,
      "learning_rate": 3.527601274535012e-06,
      "loss": 0.75600034,
      "num_input_tokens_seen": 88377040,
      "step": 4102,
      "time_per_iteration": 2.502042531967163
    },
    {
      "auxiliary_loss_clip": 0.01122108,
      "auxiliary_loss_mlp": 0.01034935,
      "balance_loss_clip": 1.05115032,
      "balance_loss_mlp": 1.01866329,
      "epoch": 0.2466857056966782,
      "flos": 30701361012480.0,
      "grad_norm": 2.0846927384935237,
      "language_loss": 0.76391196,
      "learning_rate": 3.5273498666608004e-06,
      "loss": 0.78548235,
      "num_input_tokens_seen": 88395085,
      "step": 4103,
      "time_per_iteration": 2.628417491912842
    },
    {
      "auxiliary_loss_clip": 0.01124086,
      "auxiliary_loss_mlp": 0.01040225,
      "balance_loss_clip": 1.04939628,
      "balance_loss_mlp": 1.02327991,
      "epoch": 0.24674582894934616,
      "flos": 22528523669760.0,
      "grad_norm": 2.2781281745005124,
      "language_loss": 0.78875113,
      "learning_rate": 3.5270984008693288e-06,
      "loss": 0.81039417,
      "num_input_tokens_seen": 88413205,
      "step": 4104,
      "time_per_iteration": 2.5066943168640137
    },
    {
      "auxiliary_loss_clip": 0.01125264,
      "auxiliary_loss_mlp": 0.01041177,
      "balance_loss_clip": 1.04759538,
      "balance_loss_mlp": 1.02279544,
      "epoch": 0.24680595220201412,
      "flos": 20704297599360.0,
      "grad_norm": 1.75821864123404,
      "language_loss": 0.83185595,
      "learning_rate": 3.526846877170133e-06,
      "loss": 0.85352039,
      "num_input_tokens_seen": 88431525,
      "step": 4105,
      "time_per_iteration": 2.517305850982666
    },
    {
      "auxiliary_loss_clip": 0.01138692,
      "auxiliary_loss_mlp": 0.01040378,
      "balance_loss_clip": 1.04915869,
      "balance_loss_mlp": 1.02396274,
      "epoch": 0.2468660754546821,
      "flos": 21831174051840.0,
      "grad_norm": 1.7821254374915958,
      "language_loss": 0.76362121,
      "learning_rate": 3.52659529557275e-06,
      "loss": 0.78541195,
      "num_input_tokens_seen": 88451210,
      "step": 4106,
      "time_per_iteration": 2.50044322013855
    },
    {
      "auxiliary_loss_clip": 0.01101183,
      "auxiliary_loss_mlp": 0.01057149,
      "balance_loss_clip": 1.04281211,
      "balance_loss_mlp": 1.03598893,
      "epoch": 0.24692619870735008,
      "flos": 15267709578240.0,
      "grad_norm": 3.061023790765575,
      "language_loss": 0.72957838,
      "learning_rate": 3.5263436560867205e-06,
      "loss": 0.75116169,
      "num_input_tokens_seen": 88467790,
      "step": 4107,
      "time_per_iteration": 2.509249687194824
    },
    {
      "auxiliary_loss_clip": 0.01139325,
      "auxiliary_loss_mlp": 0.01048073,
      "balance_loss_clip": 1.04902554,
      "balance_loss_mlp": 1.03102636,
      "epoch": 0.24698632196001805,
      "flos": 29680097523840.0,
      "grad_norm": 2.0717804190004134,
      "language_loss": 0.65264738,
      "learning_rate": 3.526091958721587e-06,
      "loss": 0.67452133,
      "num_input_tokens_seen": 88490330,
      "step": 4108,
      "time_per_iteration": 2.565877914428711
    },
    {
      "auxiliary_loss_clip": 0.01088916,
      "auxiliary_loss_mlp": 0.01046132,
      "balance_loss_clip": 1.04593635,
      "balance_loss_mlp": 1.02897763,
      "epoch": 0.247046445212686,
      "flos": 39165469741440.0,
      "grad_norm": 1.769755366651145,
      "language_loss": 0.72741055,
      "learning_rate": 3.5258402034868936e-06,
      "loss": 0.74876094,
      "num_input_tokens_seen": 88512435,
      "step": 4109,
      "time_per_iteration": 2.755033493041992
    },
    {
      "auxiliary_loss_clip": 0.011054,
      "auxiliary_loss_mlp": 0.01043577,
      "balance_loss_clip": 1.04649365,
      "balance_loss_mlp": 1.0265305,
      "epoch": 0.24710656846535398,
      "flos": 22998845376000.0,
      "grad_norm": 2.411221865263363,
      "language_loss": 0.79137492,
      "learning_rate": 3.5255883903921866e-06,
      "loss": 0.81286472,
      "num_input_tokens_seen": 88529780,
      "step": 4110,
      "time_per_iteration": 2.585092782974243
    },
    {
      "auxiliary_loss_clip": 0.01113201,
      "auxiliary_loss_mlp": 0.01040365,
      "balance_loss_clip": 1.05533838,
      "balance_loss_mlp": 1.02260327,
      "epoch": 0.24716669171802194,
      "flos": 26432803451520.0,
      "grad_norm": 2.4366560444076923,
      "language_loss": 0.80910999,
      "learning_rate": 3.5253365194470144e-06,
      "loss": 0.83064568,
      "num_input_tokens_seen": 88547200,
      "step": 4111,
      "time_per_iteration": 2.575685739517212
    },
    {
      "auxiliary_loss_clip": 0.0113651,
      "auxiliary_loss_mlp": 0.01042857,
      "balance_loss_clip": 1.04611444,
      "balance_loss_mlp": 1.02690661,
      "epoch": 0.2472268149706899,
      "flos": 23329870139520.0,
      "grad_norm": 1.8703946250913648,
      "language_loss": 0.75027156,
      "learning_rate": 3.5250845906609294e-06,
      "loss": 0.77206522,
      "num_input_tokens_seen": 88566415,
      "step": 4112,
      "time_per_iteration": 2.494840383529663
    },
    {
      "auxiliary_loss_clip": 0.01106586,
      "auxiliary_loss_mlp": 0.00796106,
      "balance_loss_clip": 1.0466553,
      "balance_loss_mlp": 1.01570094,
      "epoch": 0.24728693822335787,
      "flos": 23768734510080.0,
      "grad_norm": 2.1896344182442604,
      "language_loss": 0.82396328,
      "learning_rate": 3.5248326040434835e-06,
      "loss": 0.84299016,
      "num_input_tokens_seen": 88585225,
      "step": 4113,
      "time_per_iteration": 2.614182710647583
    },
    {
      "auxiliary_loss_clip": 0.01136864,
      "auxiliary_loss_mlp": 0.0104037,
      "balance_loss_clip": 1.0464623,
      "balance_loss_mlp": 1.02291799,
      "epoch": 0.24734706147602586,
      "flos": 19317499355520.0,
      "grad_norm": 2.375625929213814,
      "language_loss": 0.87100542,
      "learning_rate": 3.5245805596042322e-06,
      "loss": 0.8927778,
      "num_input_tokens_seen": 88603280,
      "step": 4114,
      "time_per_iteration": 2.4783382415771484
    },
    {
      "auxiliary_loss_clip": 0.01090178,
      "auxiliary_loss_mlp": 0.01038062,
      "balance_loss_clip": 1.04879808,
      "balance_loss_mlp": 1.02192092,
      "epoch": 0.24740718472869383,
      "flos": 28036932935040.0,
      "grad_norm": 1.8014719145348934,
      "language_loss": 0.75647044,
      "learning_rate": 3.524328457352734e-06,
      "loss": 0.77775288,
      "num_input_tokens_seen": 88624925,
      "step": 4115,
      "time_per_iteration": 2.6645374298095703
    },
    {
      "auxiliary_loss_clip": 0.0100418,
      "auxiliary_loss_mlp": 0.01004414,
      "balance_loss_clip": 1.02191377,
      "balance_loss_mlp": 1.00146914,
      "epoch": 0.2474673079813618,
      "flos": 68107569408000.0,
      "grad_norm": 0.6884371776921112,
      "language_loss": 0.58225501,
      "learning_rate": 3.5240762972985475e-06,
      "loss": 0.60234094,
      "num_input_tokens_seen": 88691475,
      "step": 4116,
      "time_per_iteration": 3.2892348766326904
    },
    {
      "auxiliary_loss_clip": 0.01117571,
      "auxiliary_loss_mlp": 0.01036068,
      "balance_loss_clip": 1.0489924,
      "balance_loss_mlp": 1.01941454,
      "epoch": 0.24752743123402976,
      "flos": 29462119839360.0,
      "grad_norm": 1.5755206340169294,
      "language_loss": 0.83955437,
      "learning_rate": 3.523824079451235e-06,
      "loss": 0.86109078,
      "num_input_tokens_seen": 88713425,
      "step": 4117,
      "time_per_iteration": 2.6707966327667236
    },
    {
      "auxiliary_loss_clip": 0.010429,
      "auxiliary_loss_mlp": 0.01120672,
      "balance_loss_clip": 1.0318023,
      "balance_loss_mlp": 1.60832429,
      "epoch": 0.24758755448669773,
      "flos": 58350459824640.0,
      "grad_norm": 0.9183164676287245,
      "language_loss": 0.63480014,
      "learning_rate": 3.5235718038203602e-06,
      "loss": 0.65643585,
      "num_input_tokens_seen": 88769995,
      "step": 4118,
      "time_per_iteration": 2.9928536415100098
    },
    {
      "auxiliary_loss_clip": 0.01124542,
      "auxiliary_loss_mlp": 0.01043981,
      "balance_loss_clip": 1.04752135,
      "balance_loss_mlp": 1.02767289,
      "epoch": 0.2476476777393657,
      "flos": 20484416494080.0,
      "grad_norm": 2.751520021795543,
      "language_loss": 0.7927981,
      "learning_rate": 3.523319470415491e-06,
      "loss": 0.81448328,
      "num_input_tokens_seen": 88789970,
      "step": 4119,
      "time_per_iteration": 2.5344390869140625
    },
    {
      "auxiliary_loss_clip": 0.01125539,
      "auxiliary_loss_mlp": 0.01044673,
      "balance_loss_clip": 1.0487349,
      "balance_loss_mlp": 1.027578,
      "epoch": 0.24770780099203366,
      "flos": 20485853038080.0,
      "grad_norm": 2.5427382362908157,
      "language_loss": 0.74193549,
      "learning_rate": 3.5230670792461943e-06,
      "loss": 0.76363766,
      "num_input_tokens_seen": 88810000,
      "step": 4120,
      "time_per_iteration": 2.522059679031372
    },
    {
      "auxiliary_loss_clip": 0.01135088,
      "auxiliary_loss_mlp": 0.01046808,
      "balance_loss_clip": 1.05156136,
      "balance_loss_mlp": 1.02946293,
      "epoch": 0.24776792424470165,
      "flos": 15153405523200.0,
      "grad_norm": 2.4358877409098048,
      "language_loss": 0.88883543,
      "learning_rate": 3.522814630322041e-06,
      "loss": 0.91065437,
      "num_input_tokens_seen": 88827515,
      "step": 4121,
      "time_per_iteration": 2.5243165493011475
    },
    {
      "auxiliary_loss_clip": 0.01140053,
      "auxiliary_loss_mlp": 0.01042954,
      "balance_loss_clip": 1.04703283,
      "balance_loss_mlp": 1.02527499,
      "epoch": 0.2478280474973696,
      "flos": 21725453347200.0,
      "grad_norm": 1.9237126277521899,
      "language_loss": 0.69383675,
      "learning_rate": 3.5225621236526045e-06,
      "loss": 0.71566677,
      "num_input_tokens_seen": 88845025,
      "step": 4122,
      "time_per_iteration": 2.475464344024658
    },
    {
      "auxiliary_loss_clip": 0.01140506,
      "auxiliary_loss_mlp": 0.01044107,
      "balance_loss_clip": 1.04741681,
      "balance_loss_mlp": 1.02492595,
      "epoch": 0.24788817075003758,
      "flos": 20412200200320.0,
      "grad_norm": 2.2434594038346307,
      "language_loss": 0.80372918,
      "learning_rate": 3.5223095592474596e-06,
      "loss": 0.82557535,
      "num_input_tokens_seen": 88861740,
      "step": 4123,
      "time_per_iteration": 2.4736177921295166
    },
    {
      "auxiliary_loss_clip": 0.01081597,
      "auxiliary_loss_mlp": 0.01048962,
      "balance_loss_clip": 1.05137062,
      "balance_loss_mlp": 1.03181958,
      "epoch": 0.24794829400270554,
      "flos": 22594455083520.0,
      "grad_norm": 1.847149334970972,
      "language_loss": 0.74529076,
      "learning_rate": 3.5220569371161846e-06,
      "loss": 0.76659632,
      "num_input_tokens_seen": 88879740,
      "step": 4124,
      "time_per_iteration": 2.618082046508789
    },
    {
      "auxiliary_loss_clip": 0.01126016,
      "auxiliary_loss_mlp": 0.01043572,
      "balance_loss_clip": 1.04680741,
      "balance_loss_mlp": 1.02749085,
      "epoch": 0.2480084172553735,
      "flos": 39676047615360.0,
      "grad_norm": 1.4149046635443423,
      "language_loss": 0.73537505,
      "learning_rate": 3.521804257268357e-06,
      "loss": 0.7570709,
      "num_input_tokens_seen": 88904095,
      "step": 4125,
      "time_per_iteration": 2.7050702571868896
    },
    {
      "auxiliary_loss_clip": 0.01110037,
      "auxiliary_loss_mlp": 0.00891478,
      "balance_loss_clip": 1.05011868,
      "balance_loss_mlp": 1.19141185,
      "epoch": 0.24806854050804147,
      "flos": 22053712763520.0,
      "grad_norm": 1.683684435761037,
      "language_loss": 0.69425464,
      "learning_rate": 3.5215515197135595e-06,
      "loss": 0.71426976,
      "num_input_tokens_seen": 88920740,
      "step": 4126,
      "time_per_iteration": 2.562877655029297
    },
    {
      "auxiliary_loss_clip": 0.01127104,
      "auxiliary_loss_mlp": 0.01046934,
      "balance_loss_clip": 1.04586124,
      "balance_loss_mlp": 1.02923131,
      "epoch": 0.24812866376070947,
      "flos": 15486764670720.0,
      "grad_norm": 2.153427807835069,
      "language_loss": 0.80948591,
      "learning_rate": 3.5212987244613764e-06,
      "loss": 0.83122635,
      "num_input_tokens_seen": 88938510,
      "step": 4127,
      "time_per_iteration": 3.8914225101470947
    },
    {
      "auxiliary_loss_clip": 0.01136646,
      "auxiliary_loss_mlp": 0.00880402,
      "balance_loss_clip": 1.05215406,
      "balance_loss_mlp": 1.17223668,
      "epoch": 0.24818878701337743,
      "flos": 14757419013120.0,
      "grad_norm": 2.108074144190643,
      "language_loss": 0.84281337,
      "learning_rate": 3.5210458715213927e-06,
      "loss": 0.86298382,
      "num_input_tokens_seen": 88955235,
      "step": 4128,
      "time_per_iteration": 3.9490010738372803
    },
    {
      "auxiliary_loss_clip": 0.01111183,
      "auxiliary_loss_mlp": 0.01060146,
      "balance_loss_clip": 1.04542506,
      "balance_loss_mlp": 1.04185939,
      "epoch": 0.2482489102660454,
      "flos": 27089501852160.0,
      "grad_norm": 2.207465673456794,
      "language_loss": 0.65040553,
      "learning_rate": 3.5207929609031973e-06,
      "loss": 0.67211884,
      "num_input_tokens_seen": 88975210,
      "step": 4129,
      "time_per_iteration": 2.572672128677368
    },
    {
      "auxiliary_loss_clip": 0.01100023,
      "auxiliary_loss_mlp": 0.01041027,
      "balance_loss_clip": 1.05255604,
      "balance_loss_mlp": 1.02289522,
      "epoch": 0.24830903351871336,
      "flos": 26467528924800.0,
      "grad_norm": 1.6752917617675114,
      "language_loss": 0.76018584,
      "learning_rate": 3.5205399926163806e-06,
      "loss": 0.78159642,
      "num_input_tokens_seen": 88996120,
      "step": 4130,
      "time_per_iteration": 2.6761093139648438
    },
    {
      "auxiliary_loss_clip": 0.0108235,
      "auxiliary_loss_mlp": 0.01048149,
      "balance_loss_clip": 1.05399072,
      "balance_loss_mlp": 1.02996993,
      "epoch": 0.24836915677138133,
      "flos": 10228436870400.0,
      "grad_norm": 2.1154183824550983,
      "language_loss": 0.76766288,
      "learning_rate": 3.520286966670535e-06,
      "loss": 0.78896797,
      "num_input_tokens_seen": 89008685,
      "step": 4131,
      "time_per_iteration": 2.6322720050811768
    },
    {
      "auxiliary_loss_clip": 0.01128727,
      "auxiliary_loss_mlp": 0.01038182,
      "balance_loss_clip": 1.04962575,
      "balance_loss_mlp": 1.02225518,
      "epoch": 0.2484292800240493,
      "flos": 30080429579520.0,
      "grad_norm": 1.6462772087836217,
      "language_loss": 0.83701503,
      "learning_rate": 3.520033883075255e-06,
      "loss": 0.85868406,
      "num_input_tokens_seen": 89031160,
      "step": 4132,
      "time_per_iteration": 4.002025604248047
    },
    {
      "auxiliary_loss_clip": 0.0111566,
      "auxiliary_loss_mlp": 0.01040046,
      "balance_loss_clip": 1.04667783,
      "balance_loss_mlp": 1.02224755,
      "epoch": 0.24848940327671726,
      "flos": 13442944803840.0,
      "grad_norm": 1.6530696289927114,
      "language_loss": 0.71028817,
      "learning_rate": 3.5197807418401386e-06,
      "loss": 0.73184526,
      "num_input_tokens_seen": 89047235,
      "step": 4133,
      "time_per_iteration": 3.8995234966278076
    },
    {
      "auxiliary_loss_clip": 0.01144755,
      "auxiliary_loss_mlp": 0.01041711,
      "balance_loss_clip": 1.04837191,
      "balance_loss_mlp": 1.02170789,
      "epoch": 0.24854952652938525,
      "flos": 19970247260160.0,
      "grad_norm": 2.852315598683667,
      "language_loss": 0.61901885,
      "learning_rate": 3.5195275429747834e-06,
      "loss": 0.64088356,
      "num_input_tokens_seen": 89064790,
      "step": 4134,
      "time_per_iteration": 2.4875385761260986
    },
    {
      "auxiliary_loss_clip": 0.01133761,
      "auxiliary_loss_mlp": 0.01038938,
      "balance_loss_clip": 1.04852426,
      "balance_loss_mlp": 1.02216494,
      "epoch": 0.24860964978205322,
      "flos": 18150187167360.0,
      "grad_norm": 2.2937288186725135,
      "language_loss": 0.78303152,
      "learning_rate": 3.5192742864887914e-06,
      "loss": 0.80475849,
      "num_input_tokens_seen": 89083250,
      "step": 4135,
      "time_per_iteration": 2.4985318183898926
    },
    {
      "auxiliary_loss_clip": 0.01121386,
      "auxiliary_loss_mlp": 0.01033319,
      "balance_loss_clip": 1.05619216,
      "balance_loss_mlp": 1.01707041,
      "epoch": 0.24866977303472118,
      "flos": 11728641329280.0,
      "grad_norm": 2.3500161314053094,
      "language_loss": 0.82600498,
      "learning_rate": 3.5190209723917662e-06,
      "loss": 0.847552,
      "num_input_tokens_seen": 89100905,
      "step": 4136,
      "time_per_iteration": 2.5383379459381104
    },
    {
      "auxiliary_loss_clip": 0.01112691,
      "auxiliary_loss_mlp": 0.01039828,
      "balance_loss_clip": 1.04932618,
      "balance_loss_mlp": 1.02349591,
      "epoch": 0.24872989628738915,
      "flos": 34823582565120.0,
      "grad_norm": 1.7399471198848842,
      "language_loss": 0.70878381,
      "learning_rate": 3.518767600693314e-06,
      "loss": 0.73030901,
      "num_input_tokens_seen": 89122630,
      "step": 4137,
      "time_per_iteration": 2.681018590927124
    },
    {
      "auxiliary_loss_clip": 0.01129554,
      "auxiliary_loss_mlp": 0.00836255,
      "balance_loss_clip": 1.04443419,
      "balance_loss_mlp": 1.08970308,
      "epoch": 0.2487900195400571,
      "flos": 13699347062400.0,
      "grad_norm": 2.173309315743969,
      "language_loss": 0.67251742,
      "learning_rate": 3.518514171403042e-06,
      "loss": 0.69217551,
      "num_input_tokens_seen": 89141050,
      "step": 4138,
      "time_per_iteration": 2.5236244201660156
    },
    {
      "auxiliary_loss_clip": 0.01102859,
      "auxiliary_loss_mlp": 0.01036789,
      "balance_loss_clip": 1.05030572,
      "balance_loss_mlp": 1.02089214,
      "epoch": 0.24885014279272508,
      "flos": 25337815297920.0,
      "grad_norm": 2.0091220993289305,
      "language_loss": 0.83943886,
      "learning_rate": 3.51826068453056e-06,
      "loss": 0.86083537,
      "num_input_tokens_seen": 89160810,
      "step": 4139,
      "time_per_iteration": 2.616950750350952
    },
    {
      "auxiliary_loss_clip": 0.01114385,
      "auxiliary_loss_mlp": 0.01044829,
      "balance_loss_clip": 1.04837298,
      "balance_loss_mlp": 1.02660227,
      "epoch": 0.24891026604539307,
      "flos": 20631434860800.0,
      "grad_norm": 1.6088566022062087,
      "language_loss": 0.78956091,
      "learning_rate": 3.518007140085481e-06,
      "loss": 0.81115305,
      "num_input_tokens_seen": 89180610,
      "step": 4140,
      "time_per_iteration": 2.631852626800537
    },
    {
      "auxiliary_loss_clip": 0.01042773,
      "auxiliary_loss_mlp": 0.01027477,
      "balance_loss_clip": 1.02544045,
      "balance_loss_mlp": 1.02479517,
      "epoch": 0.24897038929806103,
      "flos": 66960294030720.0,
      "grad_norm": 0.835416219623163,
      "language_loss": 0.61016411,
      "learning_rate": 3.51775353807742e-06,
      "loss": 0.63086665,
      "num_input_tokens_seen": 89241880,
      "step": 4141,
      "time_per_iteration": 3.165205955505371
    },
    {
      "auxiliary_loss_clip": 0.01145305,
      "auxiliary_loss_mlp": 0.01047435,
      "balance_loss_clip": 1.05094647,
      "balance_loss_mlp": 1.02936292,
      "epoch": 0.249030512550729,
      "flos": 36392555612160.0,
      "grad_norm": 1.7772766714269916,
      "language_loss": 0.72956634,
      "learning_rate": 3.5174998785159913e-06,
      "loss": 0.75149363,
      "num_input_tokens_seen": 89263340,
      "step": 4142,
      "time_per_iteration": 2.6350324153900146
    },
    {
      "auxiliary_loss_clip": 0.01129919,
      "auxiliary_loss_mlp": 0.01043132,
      "balance_loss_clip": 1.04777157,
      "balance_loss_mlp": 1.02578723,
      "epoch": 0.24909063580339696,
      "flos": 20154576879360.0,
      "grad_norm": 1.921181892071923,
      "language_loss": 0.81097233,
      "learning_rate": 3.5172461614108157e-06,
      "loss": 0.83270288,
      "num_input_tokens_seen": 89282870,
      "step": 4143,
      "time_per_iteration": 2.4944663047790527
    },
    {
      "auxiliary_loss_clip": 0.01115008,
      "auxiliary_loss_mlp": 0.01032831,
      "balance_loss_clip": 1.04672313,
      "balance_loss_mlp": 1.01620126,
      "epoch": 0.24915075905606493,
      "flos": 26396569607040.0,
      "grad_norm": 1.7852390662741324,
      "language_loss": 0.58778453,
      "learning_rate": 3.5169923867715137e-06,
      "loss": 0.60926288,
      "num_input_tokens_seen": 89303830,
      "step": 4144,
      "time_per_iteration": 2.601059675216675
    },
    {
      "auxiliary_loss_clip": 0.01125648,
      "auxiliary_loss_mlp": 0.01044358,
      "balance_loss_clip": 1.04463744,
      "balance_loss_mlp": 1.02737117,
      "epoch": 0.2492108823087329,
      "flos": 27527216987520.0,
      "grad_norm": 1.9815387158588953,
      "language_loss": 0.78678077,
      "learning_rate": 3.516738554607708e-06,
      "loss": 0.80848086,
      "num_input_tokens_seen": 89324350,
      "step": 4145,
      "time_per_iteration": 2.5409960746765137
    },
    {
      "auxiliary_loss_clip": 0.011365,
      "auxiliary_loss_mlp": 0.00824089,
      "balance_loss_clip": 1.04778695,
      "balance_loss_mlp": 1.06460738,
      "epoch": 0.24927100556140086,
      "flos": 16691388111360.0,
      "grad_norm": 2.4471496142617344,
      "language_loss": 0.64958751,
      "learning_rate": 3.5164846649290253e-06,
      "loss": 0.66919339,
      "num_input_tokens_seen": 89342875,
      "step": 4146,
      "time_per_iteration": 2.541562080383301
    },
    {
      "auxiliary_loss_clip": 0.01044138,
      "auxiliary_loss_mlp": 0.01007488,
      "balance_loss_clip": 1.03918529,
      "balance_loss_mlp": 1.00480556,
      "epoch": 0.24933112881406885,
      "flos": 62772464286720.0,
      "grad_norm": 0.9581606018026156,
      "language_loss": 0.67360258,
      "learning_rate": 3.5162307177450915e-06,
      "loss": 0.69411874,
      "num_input_tokens_seen": 89404925,
      "step": 4147,
      "time_per_iteration": 3.278383255004883
    },
    {
      "auxiliary_loss_clip": 0.01121499,
      "auxiliary_loss_mlp": 0.01050557,
      "balance_loss_clip": 1.05102086,
      "balance_loss_mlp": 1.03204358,
      "epoch": 0.24939125206673682,
      "flos": 26651894457600.0,
      "grad_norm": 5.8099437677705765,
      "language_loss": 0.8870182,
      "learning_rate": 3.5159767130655366e-06,
      "loss": 0.90873873,
      "num_input_tokens_seen": 89425090,
      "step": 4148,
      "time_per_iteration": 2.6115939617156982
    },
    {
      "auxiliary_loss_clip": 0.01098408,
      "auxiliary_loss_mlp": 0.01045386,
      "balance_loss_clip": 1.05248904,
      "balance_loss_mlp": 1.02484572,
      "epoch": 0.24945137531940478,
      "flos": 20704333512960.0,
      "grad_norm": 2.1358873357954065,
      "language_loss": 0.6780448,
      "learning_rate": 3.5157226508999935e-06,
      "loss": 0.6994828,
      "num_input_tokens_seen": 89442615,
      "step": 4149,
      "time_per_iteration": 2.6600887775421143
    },
    {
      "auxiliary_loss_clip": 0.01132804,
      "auxiliary_loss_mlp": 0.01043699,
      "balance_loss_clip": 1.05065155,
      "balance_loss_mlp": 1.02632987,
      "epoch": 0.24951149857207275,
      "flos": 23768662682880.0,
      "grad_norm": 1.8246636469997897,
      "language_loss": 0.71436256,
      "learning_rate": 3.515468531258095e-06,
      "loss": 0.73612756,
      "num_input_tokens_seen": 89463025,
      "step": 4150,
      "time_per_iteration": 2.586015462875366
    },
    {
      "auxiliary_loss_clip": 0.01099992,
      "auxiliary_loss_mlp": 0.01052415,
      "balance_loss_clip": 1.05100632,
      "balance_loss_mlp": 1.03419948,
      "epoch": 0.2495716218247407,
      "flos": 15664881237120.0,
      "grad_norm": 1.8003030563349758,
      "language_loss": 0.7278831,
      "learning_rate": 3.515214354149478e-06,
      "loss": 0.74940717,
      "num_input_tokens_seen": 89480225,
      "step": 4151,
      "time_per_iteration": 2.6327946186065674
    },
    {
      "auxiliary_loss_clip": 0.01139202,
      "auxiliary_loss_mlp": 0.01051128,
      "balance_loss_clip": 1.04965281,
      "balance_loss_mlp": 1.0335331,
      "epoch": 0.24963174507740868,
      "flos": 24052499953920.0,
      "grad_norm": 5.654057858241154,
      "language_loss": 0.63348258,
      "learning_rate": 3.514960119583781e-06,
      "loss": 0.65538591,
      "num_input_tokens_seen": 89496985,
      "step": 4152,
      "time_per_iteration": 2.5183281898498535
    },
    {
      "auxiliary_loss_clip": 0.01125002,
      "auxiliary_loss_mlp": 0.01039395,
      "balance_loss_clip": 1.05073178,
      "balance_loss_mlp": 1.02300406,
      "epoch": 0.24969186833007664,
      "flos": 21799501234560.0,
      "grad_norm": 2.0962964827423685,
      "language_loss": 0.77338862,
      "learning_rate": 3.514705827570645e-06,
      "loss": 0.79503262,
      "num_input_tokens_seen": 89514420,
      "step": 4153,
      "time_per_iteration": 2.5620808601379395
    },
    {
      "auxiliary_loss_clip": 0.01132082,
      "auxiliary_loss_mlp": 0.0104217,
      "balance_loss_clip": 1.05042517,
      "balance_loss_mlp": 1.02422905,
      "epoch": 0.24975199158274464,
      "flos": 19938143479680.0,
      "grad_norm": 2.0786116446902683,
      "language_loss": 0.76931506,
      "learning_rate": 3.514451478119711e-06,
      "loss": 0.79105759,
      "num_input_tokens_seen": 89532925,
      "step": 4154,
      "time_per_iteration": 2.5017600059509277
    },
    {
      "auxiliary_loss_clip": 0.01132313,
      "auxiliary_loss_mlp": 0.01046606,
      "balance_loss_clip": 1.05098653,
      "balance_loss_mlp": 1.02740121,
      "epoch": 0.2498121148354126,
      "flos": 25338389915520.0,
      "grad_norm": 1.6686039247396394,
      "language_loss": 0.70890975,
      "learning_rate": 3.5141970712406258e-06,
      "loss": 0.73069894,
      "num_input_tokens_seen": 89552855,
      "step": 4155,
      "time_per_iteration": 2.573125123977661
    },
    {
      "auxiliary_loss_clip": 0.0112671,
      "auxiliary_loss_mlp": 0.01053332,
      "balance_loss_clip": 1.05419278,
      "balance_loss_mlp": 1.03529525,
      "epoch": 0.24987223808808057,
      "flos": 20558787603840.0,
      "grad_norm": 1.6602888993873777,
      "language_loss": 0.74720329,
      "learning_rate": 3.513942606943036e-06,
      "loss": 0.76900375,
      "num_input_tokens_seen": 89572830,
      "step": 4156,
      "time_per_iteration": 2.5292437076568604
    },
    {
      "auxiliary_loss_clip": 0.01124922,
      "auxiliary_loss_mlp": 0.01041967,
      "balance_loss_clip": 1.0495472,
      "balance_loss_mlp": 1.02584982,
      "epoch": 0.24993236134074853,
      "flos": 19749037351680.0,
      "grad_norm": 2.0597851096726227,
      "language_loss": 0.76852906,
      "learning_rate": 3.513688085236591e-06,
      "loss": 0.79019797,
      "num_input_tokens_seen": 89590345,
      "step": 4157,
      "time_per_iteration": 2.5438830852508545
    },
    {
      "auxiliary_loss_clip": 0.010848,
      "auxiliary_loss_mlp": 0.01048115,
      "balance_loss_clip": 1.0501945,
      "balance_loss_mlp": 1.03086579,
      "epoch": 0.2499924845934165,
      "flos": 18770292587520.0,
      "grad_norm": 1.7501006942913941,
      "language_loss": 0.81622154,
      "learning_rate": 3.513433506130942e-06,
      "loss": 0.83755064,
      "num_input_tokens_seen": 89610295,
      "step": 4158,
      "time_per_iteration": 2.631075382232666
    },
    {
      "auxiliary_loss_clip": 0.01113003,
      "auxiliary_loss_mlp": 0.01041548,
      "balance_loss_clip": 1.04756236,
      "balance_loss_mlp": 1.0242269,
      "epoch": 0.25005260784608446,
      "flos": 16872198197760.0,
      "grad_norm": 1.7915506440459277,
      "language_loss": 0.75868756,
      "learning_rate": 3.5131788696357427e-06,
      "loss": 0.78023303,
      "num_input_tokens_seen": 89627795,
      "step": 4159,
      "time_per_iteration": 2.5396616458892822
    },
    {
      "auxiliary_loss_clip": 0.01137981,
      "auxiliary_loss_mlp": 0.01040365,
      "balance_loss_clip": 1.04924023,
      "balance_loss_mlp": 1.02199435,
      "epoch": 0.2501127310987524,
      "flos": 22124923476480.0,
      "grad_norm": 1.7617842334061926,
      "language_loss": 0.71579945,
      "learning_rate": 3.512924175760649e-06,
      "loss": 0.73758292,
      "num_input_tokens_seen": 89648090,
      "step": 4160,
      "time_per_iteration": 2.5165696144104004
    },
    {
      "auxiliary_loss_clip": 0.01053899,
      "auxiliary_loss_mlp": 0.01000773,
      "balance_loss_clip": 1.02166438,
      "balance_loss_mlp": 0.99829334,
      "epoch": 0.2501728543514204,
      "flos": 69458061980160.0,
      "grad_norm": 0.7522567263675518,
      "language_loss": 0.56782573,
      "learning_rate": 3.5126694245153186e-06,
      "loss": 0.58837247,
      "num_input_tokens_seen": 89710345,
      "step": 4161,
      "time_per_iteration": 3.170154571533203
    },
    {
      "auxiliary_loss_clip": 0.01136388,
      "auxiliary_loss_mlp": 0.0104707,
      "balance_loss_clip": 1.05113387,
      "balance_loss_mlp": 1.02952278,
      "epoch": 0.25023297760408836,
      "flos": 16289978647680.0,
      "grad_norm": 2.670184363257036,
      "language_loss": 0.80400395,
      "learning_rate": 3.5124146159094125e-06,
      "loss": 0.82583857,
      "num_input_tokens_seen": 89729390,
      "step": 4162,
      "time_per_iteration": 2.4993739128112793
    },
    {
      "auxiliary_loss_clip": 0.01123698,
      "auxiliary_loss_mlp": 0.00809538,
      "balance_loss_clip": 1.04711008,
      "balance_loss_mlp": 1.03849638,
      "epoch": 0.2502931008567563,
      "flos": 12237998140800.0,
      "grad_norm": 2.8625025997031637,
      "language_loss": 0.8788147,
      "learning_rate": 3.5121597499525927e-06,
      "loss": 0.89814705,
      "num_input_tokens_seen": 89742805,
      "step": 4163,
      "time_per_iteration": 2.560518264770508
    },
    {
      "auxiliary_loss_clip": 0.01123111,
      "auxiliary_loss_mlp": 0.01040794,
      "balance_loss_clip": 1.04742765,
      "balance_loss_mlp": 1.02399731,
      "epoch": 0.25035322410942434,
      "flos": 23181882105600.0,
      "grad_norm": 1.6169879629797883,
      "language_loss": 0.83554333,
      "learning_rate": 3.5119048266545232e-06,
      "loss": 0.85718238,
      "num_input_tokens_seen": 89761145,
      "step": 4164,
      "time_per_iteration": 2.5147740840911865
    },
    {
      "auxiliary_loss_clip": 0.01119927,
      "auxiliary_loss_mlp": 0.0104443,
      "balance_loss_clip": 1.04894054,
      "balance_loss_mlp": 1.02888536,
      "epoch": 0.2504133473620923,
      "flos": 20917534688640.0,
      "grad_norm": 1.739495066376407,
      "language_loss": 0.74490547,
      "learning_rate": 3.5116498460248716e-06,
      "loss": 0.76654899,
      "num_input_tokens_seen": 89780905,
      "step": 4165,
      "time_per_iteration": 3.994152069091797
    },
    {
      "auxiliary_loss_clip": 0.01109691,
      "auxiliary_loss_mlp": 0.01040661,
      "balance_loss_clip": 1.05149627,
      "balance_loss_mlp": 1.02345955,
      "epoch": 0.2504734706147603,
      "flos": 20776549806720.0,
      "grad_norm": 2.58045330400646,
      "language_loss": 0.74187362,
      "learning_rate": 3.5113948080733062e-06,
      "loss": 0.76337713,
      "num_input_tokens_seen": 89799230,
      "step": 4166,
      "time_per_iteration": 2.593127727508545
    },
    {
      "auxiliary_loss_clip": 0.01108854,
      "auxiliary_loss_mlp": 0.01046142,
      "balance_loss_clip": 1.05197155,
      "balance_loss_mlp": 1.02971506,
      "epoch": 0.25053359386742824,
      "flos": 24349373861760.0,
      "grad_norm": 1.8106054794924564,
      "language_loss": 0.81629688,
      "learning_rate": 3.5111397128094973e-06,
      "loss": 0.83784682,
      "num_input_tokens_seen": 89818240,
      "step": 4167,
      "time_per_iteration": 4.06223726272583
    },
    {
      "auxiliary_loss_clip": 0.0112493,
      "auxiliary_loss_mlp": 0.01040943,
      "balance_loss_clip": 1.04999757,
      "balance_loss_mlp": 1.02440846,
      "epoch": 0.2505937171200962,
      "flos": 21214336769280.0,
      "grad_norm": 2.4213369307678496,
      "language_loss": 0.80078799,
      "learning_rate": 3.51088456024312e-06,
      "loss": 0.8224467,
      "num_input_tokens_seen": 89834485,
      "step": 4168,
      "time_per_iteration": 2.5175251960754395
    },
    {
      "auxiliary_loss_clip": 0.01133528,
      "auxiliary_loss_mlp": 0.010419,
      "balance_loss_clip": 1.05065882,
      "balance_loss_mlp": 1.0231719,
      "epoch": 0.25065384037276417,
      "flos": 41427231379200.0,
      "grad_norm": 2.5635899941483857,
      "language_loss": 0.69463646,
      "learning_rate": 3.510629350383849e-06,
      "loss": 0.71639079,
      "num_input_tokens_seen": 89855645,
      "step": 4169,
      "time_per_iteration": 2.712191581726074
    },
    {
      "auxiliary_loss_clip": 0.01114254,
      "auxiliary_loss_mlp": 0.01045792,
      "balance_loss_clip": 1.05119789,
      "balance_loss_mlp": 1.02924621,
      "epoch": 0.25071396362543213,
      "flos": 26102389219200.0,
      "grad_norm": 1.8757010133403997,
      "language_loss": 0.77773982,
      "learning_rate": 3.510374083241361e-06,
      "loss": 0.79934031,
      "num_input_tokens_seen": 89874895,
      "step": 4170,
      "time_per_iteration": 2.5920393466949463
    },
    {
      "auxiliary_loss_clip": 0.01124444,
      "auxiliary_loss_mlp": 0.01042644,
      "balance_loss_clip": 1.05373669,
      "balance_loss_mlp": 1.0256207,
      "epoch": 0.2507740868781001,
      "flos": 19098982967040.0,
      "grad_norm": 2.656063587996473,
      "language_loss": 0.76917732,
      "learning_rate": 3.5101187588253368e-06,
      "loss": 0.79084826,
      "num_input_tokens_seen": 89891700,
      "step": 4171,
      "time_per_iteration": 3.9274473190307617
    },
    {
      "auxiliary_loss_clip": 0.01054623,
      "auxiliary_loss_mlp": 0.01002679,
      "balance_loss_clip": 1.02209496,
      "balance_loss_mlp": 1.0004853,
      "epoch": 0.25083421013076806,
      "flos": 64341868296960.0,
      "grad_norm": 0.8337183342791503,
      "language_loss": 0.60019076,
      "learning_rate": 3.509863377145458e-06,
      "loss": 0.62076378,
      "num_input_tokens_seen": 89955775,
      "step": 4172,
      "time_per_iteration": 4.542768955230713
    },
    {
      "auxiliary_loss_clip": 0.01119046,
      "auxiliary_loss_mlp": 0.01047396,
      "balance_loss_clip": 1.04860425,
      "balance_loss_mlp": 1.03008687,
      "epoch": 0.25089433338343603,
      "flos": 24279599692800.0,
      "grad_norm": 2.0876871392154817,
      "language_loss": 0.78822267,
      "learning_rate": 3.509607938211409e-06,
      "loss": 0.80988717,
      "num_input_tokens_seen": 89977150,
      "step": 4173,
      "time_per_iteration": 2.556041955947876
    },
    {
      "auxiliary_loss_clip": 0.01143433,
      "auxiliary_loss_mlp": 0.01046027,
      "balance_loss_clip": 1.05179024,
      "balance_loss_mlp": 1.02901638,
      "epoch": 0.250954456636104,
      "flos": 14721472477440.0,
      "grad_norm": 2.4155455099056615,
      "language_loss": 0.83314586,
      "learning_rate": 3.509352442032875e-06,
      "loss": 0.85504043,
      "num_input_tokens_seen": 89994925,
      "step": 4174,
      "time_per_iteration": 2.4972116947174072
    },
    {
      "auxiliary_loss_clip": 0.01093314,
      "auxiliary_loss_mlp": 0.01044753,
      "balance_loss_clip": 1.0524497,
      "balance_loss_mlp": 1.02695549,
      "epoch": 0.25101457988877196,
      "flos": 22273593868800.0,
      "grad_norm": 1.9115259935718238,
      "language_loss": 0.7098121,
      "learning_rate": 3.509096888619545e-06,
      "loss": 0.73119277,
      "num_input_tokens_seen": 90013235,
      "step": 4175,
      "time_per_iteration": 2.59843111038208
    },
    {
      "auxiliary_loss_clip": 0.01108247,
      "auxiliary_loss_mlp": 0.01039456,
      "balance_loss_clip": 1.0450871,
      "balance_loss_mlp": 1.0218488,
      "epoch": 0.2510747031414399,
      "flos": 25188929424000.0,
      "grad_norm": 2.3950360202614425,
      "language_loss": 0.81089795,
      "learning_rate": 3.50884127798111e-06,
      "loss": 0.83237493,
      "num_input_tokens_seen": 90032150,
      "step": 4176,
      "time_per_iteration": 2.635676860809326
    },
    {
      "auxiliary_loss_clip": 0.01123898,
      "auxiliary_loss_mlp": 0.01044524,
      "balance_loss_clip": 1.05265474,
      "balance_loss_mlp": 1.02623737,
      "epoch": 0.25113482639410795,
      "flos": 20704189858560.0,
      "grad_norm": 1.9689323626036777,
      "language_loss": 0.82736754,
      "learning_rate": 3.5085856101272623e-06,
      "loss": 0.84905171,
      "num_input_tokens_seen": 90049085,
      "step": 4177,
      "time_per_iteration": 2.5302979946136475
    },
    {
      "auxiliary_loss_clip": 0.0110562,
      "auxiliary_loss_mlp": 0.01046526,
      "balance_loss_clip": 1.05345774,
      "balance_loss_mlp": 1.02896667,
      "epoch": 0.2511949496467759,
      "flos": 21506936958720.0,
      "grad_norm": 2.873094117994669,
      "language_loss": 0.83124399,
      "learning_rate": 3.508329885067698e-06,
      "loss": 0.85276556,
      "num_input_tokens_seen": 90067695,
      "step": 4178,
      "time_per_iteration": 2.5873606204986572
    },
    {
      "auxiliary_loss_clip": 0.01136461,
      "auxiliary_loss_mlp": 0.00813418,
      "balance_loss_clip": 1.04790998,
      "balance_loss_mlp": 1.04619706,
      "epoch": 0.2512550728994439,
      "flos": 20701999128960.0,
      "grad_norm": 2.177709512125536,
      "language_loss": 0.75745499,
      "learning_rate": 3.508074102812112e-06,
      "loss": 0.77695382,
      "num_input_tokens_seen": 90083890,
      "step": 4179,
      "time_per_iteration": 2.4612603187561035
    },
    {
      "auxiliary_loss_clip": 0.01110062,
      "auxiliary_loss_mlp": 0.01049512,
      "balance_loss_clip": 1.05662632,
      "balance_loss_mlp": 1.03147602,
      "epoch": 0.25131519615211184,
      "flos": 18478626151680.0,
      "grad_norm": 2.0407728053029084,
      "language_loss": 0.70052803,
      "learning_rate": 3.507818263370206e-06,
      "loss": 0.72212374,
      "num_input_tokens_seen": 90100995,
      "step": 4180,
      "time_per_iteration": 2.5649971961975098
    },
    {
      "auxiliary_loss_clip": 0.01142537,
      "auxiliary_loss_mlp": 0.01044082,
      "balance_loss_clip": 1.05112052,
      "balance_loss_mlp": 1.02691579,
      "epoch": 0.2513753194047798,
      "flos": 20484955198080.0,
      "grad_norm": 1.8549944600581727,
      "language_loss": 0.86271638,
      "learning_rate": 3.5075623667516796e-06,
      "loss": 0.88458258,
      "num_input_tokens_seen": 90120365,
      "step": 4181,
      "time_per_iteration": 2.4703288078308105
    },
    {
      "auxiliary_loss_clip": 0.01142255,
      "auxiliary_loss_mlp": 0.01041752,
      "balance_loss_clip": 1.05076313,
      "balance_loss_mlp": 1.02537239,
      "epoch": 0.25143544265744777,
      "flos": 37670077704960.0,
      "grad_norm": 1.9556720494717799,
      "language_loss": 0.68327421,
      "learning_rate": 3.507306412966238e-06,
      "loss": 0.70511431,
      "num_input_tokens_seen": 90142610,
      "step": 4182,
      "time_per_iteration": 2.6250925064086914
    },
    {
      "auxiliary_loss_clip": 0.01036658,
      "auxiliary_loss_mlp": 0.01019785,
      "balance_loss_clip": 1.02276802,
      "balance_loss_mlp": 1.01755607,
      "epoch": 0.25149556591011574,
      "flos": 69367457923200.0,
      "grad_norm": 0.856689785132992,
      "language_loss": 0.70153159,
      "learning_rate": 3.5070504020235853e-06,
      "loss": 0.72209603,
      "num_input_tokens_seen": 90200555,
      "step": 4183,
      "time_per_iteration": 3.1446096897125244
    },
    {
      "auxiliary_loss_clip": 0.01119631,
      "auxiliary_loss_mlp": 0.01037726,
      "balance_loss_clip": 1.04457963,
      "balance_loss_mlp": 1.0196898,
      "epoch": 0.2515556891627837,
      "flos": 13990402967040.0,
      "grad_norm": 1.871453909510983,
      "language_loss": 0.74295533,
      "learning_rate": 3.506794333933431e-06,
      "loss": 0.76452893,
      "num_input_tokens_seen": 90218120,
      "step": 4184,
      "time_per_iteration": 2.5237274169921875
    },
    {
      "auxiliary_loss_clip": 0.01133871,
      "auxiliary_loss_mlp": 0.01047591,
      "balance_loss_clip": 1.0536691,
      "balance_loss_mlp": 1.03055596,
      "epoch": 0.25161581241545167,
      "flos": 22163527618560.0,
      "grad_norm": 1.73834114088923,
      "language_loss": 0.83287394,
      "learning_rate": 3.506538208705484e-06,
      "loss": 0.85468853,
      "num_input_tokens_seen": 90236790,
      "step": 4185,
      "time_per_iteration": 2.5087890625
    },
    {
      "auxiliary_loss_clip": 0.01018739,
      "auxiliary_loss_mlp": 0.01007103,
      "balance_loss_clip": 1.0406152,
      "balance_loss_mlp": 1.00473106,
      "epoch": 0.25167593566811963,
      "flos": 69358407696000.0,
      "grad_norm": 0.7876503935608078,
      "language_loss": 0.61452711,
      "learning_rate": 3.5062820263494574e-06,
      "loss": 0.63478553,
      "num_input_tokens_seen": 90297070,
      "step": 4186,
      "time_per_iteration": 3.173625946044922
    },
    {
      "auxiliary_loss_clip": 0.01105394,
      "auxiliary_loss_mlp": 0.01039698,
      "balance_loss_clip": 1.0530982,
      "balance_loss_mlp": 1.02234066,
      "epoch": 0.2517360589207876,
      "flos": 13261452359040.0,
      "grad_norm": 2.152180640305493,
      "language_loss": 0.7882688,
      "learning_rate": 3.5060257868750656e-06,
      "loss": 0.80971968,
      "num_input_tokens_seen": 90315255,
      "step": 4187,
      "time_per_iteration": 2.7467048168182373
    },
    {
      "auxiliary_loss_clip": 0.0108929,
      "auxiliary_loss_mlp": 0.01047022,
      "balance_loss_clip": 1.05331969,
      "balance_loss_mlp": 1.02997482,
      "epoch": 0.25179618217345556,
      "flos": 20376828282240.0,
      "grad_norm": 1.5305206492882368,
      "language_loss": 0.80367285,
      "learning_rate": 3.5057694902920244e-06,
      "loss": 0.82503593,
      "num_input_tokens_seen": 90334990,
      "step": 4188,
      "time_per_iteration": 2.6302270889282227
    },
    {
      "auxiliary_loss_clip": 0.01131087,
      "auxiliary_loss_mlp": 0.01041819,
      "balance_loss_clip": 1.05176997,
      "balance_loss_mlp": 1.0250107,
      "epoch": 0.25185630542612353,
      "flos": 27664718250240.0,
      "grad_norm": 1.8245589903071278,
      "language_loss": 0.74277329,
      "learning_rate": 3.5055131366100534e-06,
      "loss": 0.76450241,
      "num_input_tokens_seen": 90351825,
      "step": 4189,
      "time_per_iteration": 2.5701334476470947
    },
    {
      "auxiliary_loss_clip": 0.01116179,
      "auxiliary_loss_mlp": 0.01038152,
      "balance_loss_clip": 1.050457,
      "balance_loss_mlp": 1.02193975,
      "epoch": 0.25191642867879155,
      "flos": 20996430912000.0,
      "grad_norm": 1.9439584487519277,
      "language_loss": 0.84726453,
      "learning_rate": 3.5052567258388745e-06,
      "loss": 0.86880785,
      "num_input_tokens_seen": 90369860,
      "step": 4190,
      "time_per_iteration": 2.5372204780578613
    },
    {
      "auxiliary_loss_clip": 0.01114883,
      "auxiliary_loss_mlp": 0.01041187,
      "balance_loss_clip": 1.04696929,
      "balance_loss_mlp": 1.02240014,
      "epoch": 0.2519765519314595,
      "flos": 21105671149440.0,
      "grad_norm": 2.047462002790528,
      "language_loss": 0.75713044,
      "learning_rate": 3.5050002579882082e-06,
      "loss": 0.77869117,
      "num_input_tokens_seen": 90389245,
      "step": 4191,
      "time_per_iteration": 2.556689977645874
    },
    {
      "auxiliary_loss_clip": 0.01040592,
      "auxiliary_loss_mlp": 0.01014081,
      "balance_loss_clip": 1.01856768,
      "balance_loss_mlp": 1.01176834,
      "epoch": 0.2520366751841275,
      "flos": 62744993360640.0,
      "grad_norm": 0.7225712115793018,
      "language_loss": 0.57186699,
      "learning_rate": 3.5047437330677823e-06,
      "loss": 0.59241378,
      "num_input_tokens_seen": 90456735,
      "step": 4192,
      "time_per_iteration": 3.1918184757232666
    },
    {
      "auxiliary_loss_clip": 0.01113354,
      "auxiliary_loss_mlp": 0.01037803,
      "balance_loss_clip": 1.05171919,
      "balance_loss_mlp": 1.02052915,
      "epoch": 0.25209679843679544,
      "flos": 22230716008320.0,
      "grad_norm": 2.1431237444523386,
      "language_loss": 0.76147521,
      "learning_rate": 3.504487151087323e-06,
      "loss": 0.78298676,
      "num_input_tokens_seen": 90474165,
      "step": 4193,
      "time_per_iteration": 2.5616514682769775
    },
    {
      "auxiliary_loss_clip": 0.01133836,
      "auxiliary_loss_mlp": 0.01042961,
      "balance_loss_clip": 1.0520308,
      "balance_loss_mlp": 1.02625978,
      "epoch": 0.2521569216894634,
      "flos": 12166643773440.0,
      "grad_norm": 2.411381985503307,
      "language_loss": 0.84560609,
      "learning_rate": 3.5042305120565598e-06,
      "loss": 0.86737406,
      "num_input_tokens_seen": 90491660,
      "step": 4194,
      "time_per_iteration": 2.499225616455078
    },
    {
      "auxiliary_loss_clip": 0.0114583,
      "auxiliary_loss_mlp": 0.01049245,
      "balance_loss_clip": 1.05131829,
      "balance_loss_mlp": 1.03294885,
      "epoch": 0.2522170449421314,
      "flos": 23699786353920.0,
      "grad_norm": 1.6507998675592224,
      "language_loss": 0.88385773,
      "learning_rate": 3.5039738159852253e-06,
      "loss": 0.90580851,
      "num_input_tokens_seen": 90514025,
      "step": 4195,
      "time_per_iteration": 2.550368070602417
    },
    {
      "auxiliary_loss_clip": 0.01142552,
      "auxiliary_loss_mlp": 0.01043084,
      "balance_loss_clip": 1.04944646,
      "balance_loss_mlp": 1.02345037,
      "epoch": 0.25227716819479934,
      "flos": 20955456472320.0,
      "grad_norm": 1.9692881368749846,
      "language_loss": 0.85777748,
      "learning_rate": 3.503717062883053e-06,
      "loss": 0.8796339,
      "num_input_tokens_seen": 90533530,
      "step": 4196,
      "time_per_iteration": 2.465139865875244
    },
    {
      "auxiliary_loss_clip": 0.01134465,
      "auxiliary_loss_mlp": 0.01045389,
      "balance_loss_clip": 1.05269337,
      "balance_loss_mlp": 1.02849722,
      "epoch": 0.2523372914474673,
      "flos": 23331342597120.0,
      "grad_norm": 1.8424119494160511,
      "language_loss": 0.82944691,
      "learning_rate": 3.5034602527597786e-06,
      "loss": 0.8512454,
      "num_input_tokens_seen": 90554025,
      "step": 4197,
      "time_per_iteration": 2.5739097595214844
    },
    {
      "auxiliary_loss_clip": 0.01134259,
      "auxiliary_loss_mlp": 0.01053187,
      "balance_loss_clip": 1.05059266,
      "balance_loss_mlp": 1.03356552,
      "epoch": 0.25239741470013527,
      "flos": 36970321875840.0,
      "grad_norm": 1.9307504459478506,
      "language_loss": 0.7284075,
      "learning_rate": 3.5032033856251405e-06,
      "loss": 0.75028193,
      "num_input_tokens_seen": 90576930,
      "step": 4198,
      "time_per_iteration": 2.6197543144226074
    },
    {
      "auxiliary_loss_clip": 0.01144958,
      "auxiliary_loss_mlp": 0.01047975,
      "balance_loss_clip": 1.04897666,
      "balance_loss_mlp": 1.03010535,
      "epoch": 0.25245753795280323,
      "flos": 18515757836160.0,
      "grad_norm": 3.850166073254288,
      "language_loss": 0.77104497,
      "learning_rate": 3.50294646148888e-06,
      "loss": 0.79297435,
      "num_input_tokens_seen": 90595710,
      "step": 4199,
      "time_per_iteration": 2.481710910797119
    },
    {
      "auxiliary_loss_clip": 0.01119214,
      "auxiliary_loss_mlp": 0.00803683,
      "balance_loss_clip": 1.05081201,
      "balance_loss_mlp": 1.02567899,
      "epoch": 0.2525176612054712,
      "flos": 32344884737280.0,
      "grad_norm": 1.9109894482105225,
      "language_loss": 0.73717773,
      "learning_rate": 3.502689480360739e-06,
      "loss": 0.75640666,
      "num_input_tokens_seen": 90617945,
      "step": 4200,
      "time_per_iteration": 2.6304562091827393
    },
    {
      "auxiliary_loss_clip": 0.01130258,
      "auxiliary_loss_mlp": 0.01049186,
      "balance_loss_clip": 1.04807377,
      "balance_loss_mlp": 1.0331161,
      "epoch": 0.25257778445813917,
      "flos": 45258217459200.0,
      "grad_norm": 1.6124490132601998,
      "language_loss": 0.82179862,
      "learning_rate": 3.5024324422504616e-06,
      "loss": 0.84359306,
      "num_input_tokens_seen": 90640855,
      "step": 4201,
      "time_per_iteration": 2.713648796081543
    },
    {
      "auxiliary_loss_clip": 0.01088373,
      "auxiliary_loss_mlp": 0.01048496,
      "balance_loss_clip": 1.04990649,
      "balance_loss_mlp": 1.0314486,
      "epoch": 0.25263790771080713,
      "flos": 23367791923200.0,
      "grad_norm": 1.8377893790834416,
      "language_loss": 0.75092006,
      "learning_rate": 3.5021753471677965e-06,
      "loss": 0.77228868,
      "num_input_tokens_seen": 90661350,
      "step": 4202,
      "time_per_iteration": 2.6195433139801025
    },
    {
      "auxiliary_loss_clip": 0.01127139,
      "auxiliary_loss_mlp": 0.01039127,
      "balance_loss_clip": 1.0486691,
      "balance_loss_mlp": 1.02292621,
      "epoch": 0.25269803096347515,
      "flos": 18515039564160.0,
      "grad_norm": 8.051313289538987,
      "language_loss": 0.73033226,
      "learning_rate": 3.501918195122491e-06,
      "loss": 0.75199485,
      "num_input_tokens_seen": 90680540,
      "step": 4203,
      "time_per_iteration": 2.4859166145324707
    },
    {
      "auxiliary_loss_clip": 0.01120245,
      "auxiliary_loss_mlp": 0.01042484,
      "balance_loss_clip": 1.04730225,
      "balance_loss_mlp": 1.02513337,
      "epoch": 0.2527581542161431,
      "flos": 24610552629120.0,
      "grad_norm": 1.5576353205722957,
      "language_loss": 0.77772391,
      "learning_rate": 3.501660986124297e-06,
      "loss": 0.79935122,
      "num_input_tokens_seen": 90703460,
      "step": 4204,
      "time_per_iteration": 4.1854846477508545
    },
    {
      "auxiliary_loss_clip": 0.01108976,
      "auxiliary_loss_mlp": 0.01051655,
      "balance_loss_clip": 1.0509088,
      "balance_loss_mlp": 1.03436995,
      "epoch": 0.2528182774688111,
      "flos": 12641275111680.0,
      "grad_norm": 2.002012198550442,
      "language_loss": 0.72438425,
      "learning_rate": 3.5014037201829684e-06,
      "loss": 0.74599057,
      "num_input_tokens_seen": 90718815,
      "step": 4205,
      "time_per_iteration": 4.00794529914856
    },
    {
      "auxiliary_loss_clip": 0.01120223,
      "auxiliary_loss_mlp": 0.01047128,
      "balance_loss_clip": 1.04895735,
      "balance_loss_mlp": 1.03080821,
      "epoch": 0.25287840072147905,
      "flos": 46936789879680.0,
      "grad_norm": 1.4987666619521824,
      "language_loss": 0.75560892,
      "learning_rate": 3.50114639730826e-06,
      "loss": 0.77728236,
      "num_input_tokens_seen": 90742125,
      "step": 4206,
      "time_per_iteration": 2.7286767959594727
    },
    {
      "auxiliary_loss_clip": 0.01103882,
      "auxiliary_loss_mlp": 0.01042445,
      "balance_loss_clip": 1.05054867,
      "balance_loss_mlp": 1.02552915,
      "epoch": 0.252938523974147,
      "flos": 18879712392960.0,
      "grad_norm": 1.704649105745411,
      "language_loss": 0.79357755,
      "learning_rate": 3.5008890175099296e-06,
      "loss": 0.81504083,
      "num_input_tokens_seen": 90760785,
      "step": 4207,
      "time_per_iteration": 2.5845463275909424
    },
    {
      "auxiliary_loss_clip": 0.01123209,
      "auxiliary_loss_mlp": 0.01045852,
      "balance_loss_clip": 1.04920232,
      "balance_loss_mlp": 1.02942514,
      "epoch": 0.252998647226815,
      "flos": 21434720664960.0,
      "grad_norm": 1.4914187430649963,
      "language_loss": 0.76483417,
      "learning_rate": 3.5006315807977375e-06,
      "loss": 0.78652477,
      "num_input_tokens_seen": 90780045,
      "step": 4208,
      "time_per_iteration": 2.5067555904388428
    },
    {
      "auxiliary_loss_clip": 0.01123771,
      "auxiliary_loss_mlp": 0.01036535,
      "balance_loss_clip": 1.0482868,
      "balance_loss_mlp": 1.02033424,
      "epoch": 0.25305877047948294,
      "flos": 25442171285760.0,
      "grad_norm": 2.0227314915176713,
      "language_loss": 0.69580591,
      "learning_rate": 3.5003740871814456e-06,
      "loss": 0.71740901,
      "num_input_tokens_seen": 90797980,
      "step": 4209,
      "time_per_iteration": 2.5619609355926514
    },
    {
      "auxiliary_loss_clip": 0.01045357,
      "auxiliary_loss_mlp": 0.01008836,
      "balance_loss_clip": 1.02289116,
      "balance_loss_mlp": 1.00672579,
      "epoch": 0.2531188937321509,
      "flos": 60185603629440.0,
      "grad_norm": 0.7535851087009179,
      "language_loss": 0.55069059,
      "learning_rate": 3.5001165366708175e-06,
      "loss": 0.57123256,
      "num_input_tokens_seen": 90864865,
      "step": 4210,
      "time_per_iteration": 4.6422340869903564
    },
    {
      "auxiliary_loss_clip": 0.01108315,
      "auxiliary_loss_mlp": 0.01033944,
      "balance_loss_clip": 1.04945707,
      "balance_loss_mlp": 1.01791084,
      "epoch": 0.25317901698481887,
      "flos": 19682387665920.0,
      "grad_norm": 1.9143855402434136,
      "language_loss": 0.80569184,
      "learning_rate": 3.4998589292756204e-06,
      "loss": 0.8271144,
      "num_input_tokens_seen": 90882885,
      "step": 4211,
      "time_per_iteration": 2.5754833221435547
    },
    {
      "auxiliary_loss_clip": 0.01092567,
      "auxiliary_loss_mlp": 0.01040711,
      "balance_loss_clip": 1.04849422,
      "balance_loss_mlp": 1.02459407,
      "epoch": 0.25323914023748684,
      "flos": 24424355502720.0,
      "grad_norm": 1.6880604248252724,
      "language_loss": 0.78215683,
      "learning_rate": 3.499601265005622e-06,
      "loss": 0.80348957,
      "num_input_tokens_seen": 90902985,
      "step": 4212,
      "time_per_iteration": 2.6975791454315186
    },
    {
      "auxiliary_loss_clip": 0.01129892,
      "auxiliary_loss_mlp": 0.01038437,
      "balance_loss_clip": 1.04700506,
      "balance_loss_mlp": 1.02102077,
      "epoch": 0.2532992634901548,
      "flos": 25447450584960.0,
      "grad_norm": 1.73914296717674,
      "language_loss": 0.53461695,
      "learning_rate": 3.4993435438705938e-06,
      "loss": 0.55630022,
      "num_input_tokens_seen": 90923550,
      "step": 4213,
      "time_per_iteration": 2.534336566925049
    },
    {
      "auxiliary_loss_clip": 0.01115557,
      "auxiliary_loss_mlp": 0.010442,
      "balance_loss_clip": 1.04790354,
      "balance_loss_mlp": 1.02574682,
      "epoch": 0.25335938674282277,
      "flos": 18880538405760.0,
      "grad_norm": 2.161212914563186,
      "language_loss": 0.65408087,
      "learning_rate": 3.499085765880308e-06,
      "loss": 0.67567849,
      "num_input_tokens_seen": 90943260,
      "step": 4214,
      "time_per_iteration": 2.535834312438965
    },
    {
      "auxiliary_loss_clip": 0.01044738,
      "auxiliary_loss_mlp": 0.01003893,
      "balance_loss_clip": 1.02401614,
      "balance_loss_mlp": 1.0018425,
      "epoch": 0.25341950999549073,
      "flos": 53062649936640.0,
      "grad_norm": 0.8548362707507434,
      "language_loss": 0.58043808,
      "learning_rate": 3.4988279310445396e-06,
      "loss": 0.60092437,
      "num_input_tokens_seen": 90996295,
      "step": 4215,
      "time_per_iteration": 2.8520216941833496
    },
    {
      "auxiliary_loss_clip": 0.01118991,
      "auxiliary_loss_mlp": 0.0104383,
      "balance_loss_clip": 1.05027294,
      "balance_loss_mlp": 1.02648449,
      "epoch": 0.2534796332481587,
      "flos": 39020247054720.0,
      "grad_norm": 1.6420361991509769,
      "language_loss": 0.83592808,
      "learning_rate": 3.498570039373066e-06,
      "loss": 0.85755634,
      "num_input_tokens_seen": 91017545,
      "step": 4216,
      "time_per_iteration": 2.7138257026672363
    },
    {
      "auxiliary_loss_clip": 0.01125245,
      "auxiliary_loss_mlp": 0.01039168,
      "balance_loss_clip": 1.04997885,
      "balance_loss_mlp": 1.02191842,
      "epoch": 0.2535397565008267,
      "flos": 23586990670080.0,
      "grad_norm": 1.7531630824152382,
      "language_loss": 0.80319643,
      "learning_rate": 3.498312090875666e-06,
      "loss": 0.82484061,
      "num_input_tokens_seen": 91037715,
      "step": 4217,
      "time_per_iteration": 2.584118366241455
    },
    {
      "auxiliary_loss_clip": 0.01113155,
      "auxiliary_loss_mlp": 0.01045231,
      "balance_loss_clip": 1.04432559,
      "balance_loss_mlp": 1.02760041,
      "epoch": 0.2535998797534947,
      "flos": 19281373251840.0,
      "grad_norm": 1.9958812293465191,
      "language_loss": 0.74772942,
      "learning_rate": 3.4980540855621218e-06,
      "loss": 0.76931334,
      "num_input_tokens_seen": 91055295,
      "step": 4218,
      "time_per_iteration": 2.521731376647949
    },
    {
      "auxiliary_loss_clip": 0.01132137,
      "auxiliary_loss_mlp": 0.01041681,
      "balance_loss_clip": 1.04777539,
      "balance_loss_mlp": 1.02412152,
      "epoch": 0.25366000300616265,
      "flos": 24024382583040.0,
      "grad_norm": 1.7903530623944606,
      "language_loss": 0.74492151,
      "learning_rate": 3.4977960234422167e-06,
      "loss": 0.76665974,
      "num_input_tokens_seen": 91075485,
      "step": 4219,
      "time_per_iteration": 2.515061140060425
    },
    {
      "auxiliary_loss_clip": 0.01139753,
      "auxiliary_loss_mlp": 0.01047329,
      "balance_loss_clip": 1.05136347,
      "balance_loss_mlp": 1.02923298,
      "epoch": 0.2537201262588306,
      "flos": 16289368116480.0,
      "grad_norm": 1.7236782934895805,
      "language_loss": 0.81369126,
      "learning_rate": 3.497537904525736e-06,
      "loss": 0.83556211,
      "num_input_tokens_seen": 91093620,
      "step": 4220,
      "time_per_iteration": 2.503474235534668
    },
    {
      "auxiliary_loss_clip": 0.0109648,
      "auxiliary_loss_mlp": 0.0104833,
      "balance_loss_clip": 1.04748201,
      "balance_loss_mlp": 1.02880335,
      "epoch": 0.2537802495114986,
      "flos": 23294677789440.0,
      "grad_norm": 2.3592018529084378,
      "language_loss": 0.70802414,
      "learning_rate": 3.497279728822468e-06,
      "loss": 0.72947222,
      "num_input_tokens_seen": 91114110,
      "step": 4221,
      "time_per_iteration": 2.603095054626465
    },
    {
      "auxiliary_loss_clip": 0.01141229,
      "auxiliary_loss_mlp": 0.0103788,
      "balance_loss_clip": 1.04879808,
      "balance_loss_mlp": 1.02061832,
      "epoch": 0.25384037276416654,
      "flos": 17639142416640.0,
      "grad_norm": 1.5466148007371248,
      "language_loss": 0.61762375,
      "learning_rate": 3.497021496342202e-06,
      "loss": 0.63941485,
      "num_input_tokens_seen": 91133135,
      "step": 4222,
      "time_per_iteration": 2.4434754848480225
    },
    {
      "auxiliary_loss_clip": 0.01133249,
      "auxiliary_loss_mlp": 0.01045784,
      "balance_loss_clip": 1.05050397,
      "balance_loss_mlp": 1.02783108,
      "epoch": 0.2539004960168345,
      "flos": 21507044699520.0,
      "grad_norm": 1.669617481222977,
      "language_loss": 0.74588066,
      "learning_rate": 3.496763207094731e-06,
      "loss": 0.76767099,
      "num_input_tokens_seen": 91151805,
      "step": 4223,
      "time_per_iteration": 2.48734712600708
    },
    {
      "auxiliary_loss_clip": 0.01093433,
      "auxiliary_loss_mlp": 0.01036521,
      "balance_loss_clip": 1.05494618,
      "balance_loss_mlp": 1.01915169,
      "epoch": 0.2539606192695025,
      "flos": 23950909313280.0,
      "grad_norm": 1.8854226523158253,
      "language_loss": 0.79919016,
      "learning_rate": 3.49650486108985e-06,
      "loss": 0.82048976,
      "num_input_tokens_seen": 91172270,
      "step": 4224,
      "time_per_iteration": 2.6407740116119385
    },
    {
      "auxiliary_loss_clip": 0.0112612,
      "auxiliary_loss_mlp": 0.00799344,
      "balance_loss_clip": 1.04775882,
      "balance_loss_mlp": 1.02086532,
      "epoch": 0.25402074252217044,
      "flos": 24169784837760.0,
      "grad_norm": 1.3934992558251114,
      "language_loss": 0.77413338,
      "learning_rate": 3.496246458337354e-06,
      "loss": 0.79338801,
      "num_input_tokens_seen": 91192080,
      "step": 4225,
      "time_per_iteration": 2.5812244415283203
    },
    {
      "auxiliary_loss_clip": 0.0112889,
      "auxiliary_loss_mlp": 0.01055226,
      "balance_loss_clip": 1.05014372,
      "balance_loss_mlp": 1.03689146,
      "epoch": 0.2540808657748384,
      "flos": 22303758314880.0,
      "grad_norm": 1.6561668322737146,
      "language_loss": 0.84560812,
      "learning_rate": 3.4959879988470426e-06,
      "loss": 0.86744922,
      "num_input_tokens_seen": 91211450,
      "step": 4226,
      "time_per_iteration": 2.524010181427002
    },
    {
      "auxiliary_loss_clip": 0.01139176,
      "auxiliary_loss_mlp": 0.01043821,
      "balance_loss_clip": 1.04824162,
      "balance_loss_mlp": 1.02574909,
      "epoch": 0.25414098902750637,
      "flos": 27599541022080.0,
      "grad_norm": 1.6127510479819556,
      "language_loss": 0.71263063,
      "learning_rate": 3.4957294826287164e-06,
      "loss": 0.73446059,
      "num_input_tokens_seen": 91231835,
      "step": 4227,
      "time_per_iteration": 2.5184755325317383
    },
    {
      "auxiliary_loss_clip": 0.01056297,
      "auxiliary_loss_mlp": 0.01002949,
      "balance_loss_clip": 1.02424264,
      "balance_loss_mlp": 1.00091004,
      "epoch": 0.25420111228017434,
      "flos": 58170834887040.0,
      "grad_norm": 1.0028746730980762,
      "language_loss": 0.61823136,
      "learning_rate": 3.4954709096921785e-06,
      "loss": 0.63882381,
      "num_input_tokens_seen": 91288755,
      "step": 4228,
      "time_per_iteration": 2.93076491355896
    },
    {
      "auxiliary_loss_clip": 0.01123852,
      "auxiliary_loss_mlp": 0.01039323,
      "balance_loss_clip": 1.04733372,
      "balance_loss_mlp": 1.0201062,
      "epoch": 0.2542612355328423,
      "flos": 11464409905920.0,
      "grad_norm": 3.488721138519077,
      "language_loss": 0.86304027,
      "learning_rate": 3.4952122800472336e-06,
      "loss": 0.88467205,
      "num_input_tokens_seen": 91302485,
      "step": 4229,
      "time_per_iteration": 2.4584848880767822
    },
    {
      "auxiliary_loss_clip": 0.01101028,
      "auxiliary_loss_mlp": 0.0104724,
      "balance_loss_clip": 1.05259359,
      "balance_loss_mlp": 1.0292635,
      "epoch": 0.2543213587855103,
      "flos": 22965879669120.0,
      "grad_norm": 2.1801011401635413,
      "language_loss": 0.77314049,
      "learning_rate": 3.4949535937036892e-06,
      "loss": 0.79462314,
      "num_input_tokens_seen": 91321120,
      "step": 4230,
      "time_per_iteration": 2.571901559829712
    },
    {
      "auxiliary_loss_clip": 0.01129366,
      "auxiliary_loss_mlp": 0.01045416,
      "balance_loss_clip": 1.04849374,
      "balance_loss_mlp": 1.0274272,
      "epoch": 0.2543814820381783,
      "flos": 18253178438400.0,
      "grad_norm": 3.5951308631028076,
      "language_loss": 0.75364429,
      "learning_rate": 3.4946948506713544e-06,
      "loss": 0.77539212,
      "num_input_tokens_seen": 91338575,
      "step": 4231,
      "time_per_iteration": 2.4861953258514404
    },
    {
      "auxiliary_loss_clip": 0.01133144,
      "auxiliary_loss_mlp": 0.01038405,
      "balance_loss_clip": 1.04841709,
      "balance_loss_mlp": 1.02114367,
      "epoch": 0.25444160529084625,
      "flos": 15632705629440.0,
      "grad_norm": 1.7876726972299433,
      "language_loss": 0.74301231,
      "learning_rate": 3.4944360509600416e-06,
      "loss": 0.76472783,
      "num_input_tokens_seen": 91357355,
      "step": 4232,
      "time_per_iteration": 2.4565417766571045
    },
    {
      "auxiliary_loss_clip": 0.01143802,
      "auxiliary_loss_mlp": 0.01041737,
      "balance_loss_clip": 1.05154347,
      "balance_loss_mlp": 1.02328348,
      "epoch": 0.2545017285435142,
      "flos": 24601610142720.0,
      "grad_norm": 1.9866010124374502,
      "language_loss": 0.86342567,
      "learning_rate": 3.4941771945795637e-06,
      "loss": 0.88528109,
      "num_input_tokens_seen": 91376515,
      "step": 4233,
      "time_per_iteration": 2.5158069133758545
    },
    {
      "auxiliary_loss_clip": 0.01070143,
      "auxiliary_loss_mlp": 0.01039907,
      "balance_loss_clip": 1.04045379,
      "balance_loss_mlp": 1.0234561,
      "epoch": 0.2545618517961822,
      "flos": 24679069822080.0,
      "grad_norm": 1.4879481264817707,
      "language_loss": 0.75046974,
      "learning_rate": 3.493918281539737e-06,
      "loss": 0.77157027,
      "num_input_tokens_seen": 91397595,
      "step": 4234,
      "time_per_iteration": 2.638753890991211
    },
    {
      "auxiliary_loss_clip": 0.01112207,
      "auxiliary_loss_mlp": 0.01045045,
      "balance_loss_clip": 1.05060732,
      "balance_loss_mlp": 1.02861845,
      "epoch": 0.25462197504885015,
      "flos": 23915106432000.0,
      "grad_norm": 1.5828786804226291,
      "language_loss": 0.75011247,
      "learning_rate": 3.493659311850379e-06,
      "loss": 0.771685,
      "num_input_tokens_seen": 91417775,
      "step": 4235,
      "time_per_iteration": 2.563563823699951
    },
    {
      "auxiliary_loss_clip": 0.01120774,
      "auxiliary_loss_mlp": 0.00802774,
      "balance_loss_clip": 1.05089688,
      "balance_loss_mlp": 1.02229548,
      "epoch": 0.2546820983015181,
      "flos": 24789387467520.0,
      "grad_norm": 1.867689288782314,
      "language_loss": 0.65187043,
      "learning_rate": 3.4934002855213106e-06,
      "loss": 0.67110592,
      "num_input_tokens_seen": 91437665,
      "step": 4236,
      "time_per_iteration": 2.590193271636963
    },
    {
      "auxiliary_loss_clip": 0.01140032,
      "auxiliary_loss_mlp": 0.01035623,
      "balance_loss_clip": 1.04862189,
      "balance_loss_mlp": 1.01972079,
      "epoch": 0.2547422215541861,
      "flos": 18734130570240.0,
      "grad_norm": 1.8405734796122653,
      "language_loss": 0.66651976,
      "learning_rate": 3.493141202562354e-06,
      "loss": 0.68827635,
      "num_input_tokens_seen": 91456705,
      "step": 4237,
      "time_per_iteration": 2.466362476348877
    },
    {
      "auxiliary_loss_clip": 0.01141966,
      "auxiliary_loss_mlp": 0.01045092,
      "balance_loss_clip": 1.0498637,
      "balance_loss_mlp": 1.02744937,
      "epoch": 0.25480234480685404,
      "flos": 21032449274880.0,
      "grad_norm": 2.1893778627937293,
      "language_loss": 0.74857002,
      "learning_rate": 3.492882062983333e-06,
      "loss": 0.77044058,
      "num_input_tokens_seen": 91475535,
      "step": 4238,
      "time_per_iteration": 2.478989839553833
    },
    {
      "auxiliary_loss_clip": 0.01134811,
      "auxiliary_loss_mlp": 0.01044884,
      "balance_loss_clip": 1.05262041,
      "balance_loss_mlp": 1.02628744,
      "epoch": 0.254862468059522,
      "flos": 25082167224960.0,
      "grad_norm": 1.7547210609300186,
      "language_loss": 0.80364847,
      "learning_rate": 3.492622866794074e-06,
      "loss": 0.82544541,
      "num_input_tokens_seen": 91499140,
      "step": 4239,
      "time_per_iteration": 2.567063093185425
    },
    {
      "auxiliary_loss_clip": 0.01122458,
      "auxiliary_loss_mlp": 0.01043735,
      "balance_loss_clip": 1.04824591,
      "balance_loss_mlp": 1.02604461,
      "epoch": 0.25492259131219,
      "flos": 20558392554240.0,
      "grad_norm": 1.8338299195480898,
      "language_loss": 0.77298915,
      "learning_rate": 3.492363614004407e-06,
      "loss": 0.79465115,
      "num_input_tokens_seen": 91518335,
      "step": 4240,
      "time_per_iteration": 2.4843051433563232
    },
    {
      "auxiliary_loss_clip": 0.01147008,
      "auxiliary_loss_mlp": 0.010414,
      "balance_loss_clip": 1.05006981,
      "balance_loss_mlp": 1.0221591,
      "epoch": 0.25498271456485794,
      "flos": 25042485674880.0,
      "grad_norm": 1.7745837004290628,
      "language_loss": 0.83331883,
      "learning_rate": 3.492104304624162e-06,
      "loss": 0.85520285,
      "num_input_tokens_seen": 91537655,
      "step": 4241,
      "time_per_iteration": 2.5271737575531006
    },
    {
      "auxiliary_loss_clip": 0.01134377,
      "auxiliary_loss_mlp": 0.01047659,
      "balance_loss_clip": 1.05211878,
      "balance_loss_mlp": 1.02976608,
      "epoch": 0.2550428378175259,
      "flos": 26178412354560.0,
      "grad_norm": 1.643603492614393,
      "language_loss": 0.7349869,
      "learning_rate": 3.4918449386631725e-06,
      "loss": 0.75680727,
      "num_input_tokens_seen": 91557545,
      "step": 4242,
      "time_per_iteration": 3.9310996532440186
    },
    {
      "auxiliary_loss_clip": 0.01143143,
      "auxiliary_loss_mlp": 0.008008,
      "balance_loss_clip": 1.04972279,
      "balance_loss_mlp": 1.02218652,
      "epoch": 0.2551029610701939,
      "flos": 15267170874240.0,
      "grad_norm": 2.50326989937717,
      "language_loss": 0.72598886,
      "learning_rate": 3.491585516131273e-06,
      "loss": 0.74542832,
      "num_input_tokens_seen": 91574405,
      "step": 4243,
      "time_per_iteration": 2.4698591232299805
    },
    {
      "auxiliary_loss_clip": 0.01131941,
      "auxiliary_loss_mlp": 0.01043677,
      "balance_loss_clip": 1.05228317,
      "balance_loss_mlp": 1.02566457,
      "epoch": 0.2551630843228619,
      "flos": 18112193556480.0,
      "grad_norm": 1.7138235286588777,
      "language_loss": 0.81636965,
      "learning_rate": 3.491326037038301e-06,
      "loss": 0.83812582,
      "num_input_tokens_seen": 91593755,
      "step": 4244,
      "time_per_iteration": 2.4953184127807617
    },
    {
      "auxiliary_loss_clip": 0.01046613,
      "auxiliary_loss_mlp": 0.00999958,
      "balance_loss_clip": 1.03326035,
      "balance_loss_mlp": 0.99769288,
      "epoch": 0.25522320757552985,
      "flos": 70520192167680.0,
      "grad_norm": 0.6836146490701756,
      "language_loss": 0.57725215,
      "learning_rate": 3.4910665013940967e-06,
      "loss": 0.59771788,
      "num_input_tokens_seen": 91660335,
      "step": 4245,
      "time_per_iteration": 4.584337949752808
    },
    {
      "auxiliary_loss_clip": 0.01141583,
      "auxiliary_loss_mlp": 0.01048247,
      "balance_loss_clip": 1.04727948,
      "balance_loss_mlp": 1.03031814,
      "epoch": 0.2552833308281978,
      "flos": 22893088757760.0,
      "grad_norm": 8.04411389754988,
      "language_loss": 0.6569916,
      "learning_rate": 3.4908069092085015e-06,
      "loss": 0.67888993,
      "num_input_tokens_seen": 91678500,
      "step": 4246,
      "time_per_iteration": 2.5026135444641113
    },
    {
      "auxiliary_loss_clip": 0.0112532,
      "auxiliary_loss_mlp": 0.01042422,
      "balance_loss_clip": 1.0478363,
      "balance_loss_mlp": 1.02589989,
      "epoch": 0.2553434540808658,
      "flos": 22053605022720.0,
      "grad_norm": 1.7248728832006013,
      "language_loss": 0.81413066,
      "learning_rate": 3.4905472604913585e-06,
      "loss": 0.8358081,
      "num_input_tokens_seen": 91696430,
      "step": 4247,
      "time_per_iteration": 2.5147736072540283
    },
    {
      "auxiliary_loss_clip": 0.01138112,
      "auxiliary_loss_mlp": 0.01047619,
      "balance_loss_clip": 1.05128479,
      "balance_loss_mlp": 1.02738893,
      "epoch": 0.25540357733353375,
      "flos": 16544190176640.0,
      "grad_norm": 12.174063335875717,
      "language_loss": 0.83309078,
      "learning_rate": 3.490287555252514e-06,
      "loss": 0.85494804,
      "num_input_tokens_seen": 91713270,
      "step": 4248,
      "time_per_iteration": 2.496375322341919
    },
    {
      "auxiliary_loss_clip": 0.0111429,
      "auxiliary_loss_mlp": 0.01048862,
      "balance_loss_clip": 1.05105305,
      "balance_loss_mlp": 1.03027773,
      "epoch": 0.2554637005862017,
      "flos": 17565022702080.0,
      "grad_norm": 1.9263832984929832,
      "language_loss": 0.84380645,
      "learning_rate": 3.4900277935018166e-06,
      "loss": 0.86543792,
      "num_input_tokens_seen": 91728865,
      "step": 4249,
      "time_per_iteration": 3.9953527450561523
    },
    {
      "auxiliary_loss_clip": 0.01016211,
      "auxiliary_loss_mlp": 0.01006234,
      "balance_loss_clip": 1.04986334,
      "balance_loss_mlp": 1.00369477,
      "epoch": 0.2555238238388697,
      "flos": 72244763953920.0,
      "grad_norm": 0.8760033519730654,
      "language_loss": 0.56300676,
      "learning_rate": 3.489767975249115e-06,
      "loss": 0.58323121,
      "num_input_tokens_seen": 91787470,
      "step": 4250,
      "time_per_iteration": 3.2334401607513428
    },
    {
      "auxiliary_loss_clip": 0.01117793,
      "auxiliary_loss_mlp": 0.01037482,
      "balance_loss_clip": 1.0520196,
      "balance_loss_mlp": 1.01876628,
      "epoch": 0.25558394709153764,
      "flos": 24389414547840.0,
      "grad_norm": 2.3934817870936693,
      "language_loss": 0.80622435,
      "learning_rate": 3.4895081005042632e-06,
      "loss": 0.82777715,
      "num_input_tokens_seen": 91805640,
      "step": 4251,
      "time_per_iteration": 2.7421164512634277
    },
    {
      "auxiliary_loss_clip": 0.01030078,
      "auxiliary_loss_mlp": 0.01004369,
      "balance_loss_clip": 1.02314687,
      "balance_loss_mlp": 1.00230694,
      "epoch": 0.2556440703442056,
      "flos": 69231213636480.0,
      "grad_norm": 0.7902388601062489,
      "language_loss": 0.66075605,
      "learning_rate": 3.4892481692771146e-06,
      "loss": 0.68110049,
      "num_input_tokens_seen": 91869695,
      "step": 4252,
      "time_per_iteration": 3.1808204650878906
    },
    {
      "auxiliary_loss_clip": 0.01128658,
      "auxiliary_loss_mlp": 0.01037519,
      "balance_loss_clip": 1.0511682,
      "balance_loss_mlp": 1.0214262,
      "epoch": 0.2557041935968736,
      "flos": 24863902231680.0,
      "grad_norm": 1.9077071824480705,
      "language_loss": 0.73546243,
      "learning_rate": 3.4889881815775267e-06,
      "loss": 0.75712419,
      "num_input_tokens_seen": 91889920,
      "step": 4253,
      "time_per_iteration": 2.548811197280884
    },
    {
      "auxiliary_loss_clip": 0.01099403,
      "auxiliary_loss_mlp": 0.01049412,
      "balance_loss_clip": 1.04519558,
      "balance_loss_mlp": 1.03124452,
      "epoch": 0.25576431684954154,
      "flos": 22492110257280.0,
      "grad_norm": 1.9289298777358568,
      "language_loss": 0.72460449,
      "learning_rate": 3.488728137415357e-06,
      "loss": 0.74609256,
      "num_input_tokens_seen": 91908665,
      "step": 4254,
      "time_per_iteration": 2.5426926612854004
    },
    {
      "auxiliary_loss_clip": 0.01092867,
      "auxiliary_loss_mlp": 0.00799693,
      "balance_loss_clip": 1.04223788,
      "balance_loss_mlp": 1.02061284,
      "epoch": 0.2558244401022095,
      "flos": 19826748426240.0,
      "grad_norm": 1.95712882892548,
      "language_loss": 0.81002963,
      "learning_rate": 3.4884680368004675e-06,
      "loss": 0.82895517,
      "num_input_tokens_seen": 91927855,
      "step": 4255,
      "time_per_iteration": 2.6083834171295166
    },
    {
      "auxiliary_loss_clip": 0.0111504,
      "auxiliary_loss_mlp": 0.01044827,
      "balance_loss_clip": 1.04491186,
      "balance_loss_mlp": 1.02766049,
      "epoch": 0.2558845633548775,
      "flos": 23220486247680.0,
      "grad_norm": 1.45556324342878,
      "language_loss": 0.85644722,
      "learning_rate": 3.488207879742721e-06,
      "loss": 0.87804586,
      "num_input_tokens_seen": 91948500,
      "step": 4256,
      "time_per_iteration": 2.5286006927490234
    },
    {
      "auxiliary_loss_clip": 0.01099012,
      "auxiliary_loss_mlp": 0.01053253,
      "balance_loss_clip": 1.04855323,
      "balance_loss_mlp": 1.03416717,
      "epoch": 0.2559446866075455,
      "flos": 16837867774080.0,
      "grad_norm": 1.7247652117730614,
      "language_loss": 0.7494669,
      "learning_rate": 3.4879476662519826e-06,
      "loss": 0.77098948,
      "num_input_tokens_seen": 91968375,
      "step": 4257,
      "time_per_iteration": 2.8045501708984375
    },
    {
      "auxiliary_loss_clip": 0.01019017,
      "auxiliary_loss_mlp": 0.01006111,
      "balance_loss_clip": 1.03457105,
      "balance_loss_mlp": 1.00381017,
      "epoch": 0.25600480986021346,
      "flos": 57593786895360.0,
      "grad_norm": 0.8043296708150767,
      "language_loss": 0.65303993,
      "learning_rate": 3.4876873963381196e-06,
      "loss": 0.67329121,
      "num_input_tokens_seen": 92028490,
      "step": 4258,
      "time_per_iteration": 3.1575746536254883
    },
    {
      "auxiliary_loss_clip": 0.01090975,
      "auxiliary_loss_mlp": 0.00807163,
      "balance_loss_clip": 1.0481627,
      "balance_loss_mlp": 1.03512788,
      "epoch": 0.2560649331128814,
      "flos": 27819529868160.0,
      "grad_norm": 1.6044212836112213,
      "language_loss": 0.7658385,
      "learning_rate": 3.4874270700110013e-06,
      "loss": 0.78481984,
      "num_input_tokens_seen": 92048060,
      "step": 4259,
      "time_per_iteration": 2.6379659175872803
    },
    {
      "auxiliary_loss_clip": 0.01023861,
      "auxiliary_loss_mlp": 0.0101671,
      "balance_loss_clip": 1.02979696,
      "balance_loss_mlp": 1.01424217,
      "epoch": 0.2561250563655494,
      "flos": 70950509101440.0,
      "grad_norm": 0.7943348395857789,
      "language_loss": 0.58453786,
      "learning_rate": 3.4871666872804994e-06,
      "loss": 0.60494357,
      "num_input_tokens_seen": 92118180,
      "step": 4260,
      "time_per_iteration": 3.2660748958587646
    },
    {
      "auxiliary_loss_clip": 0.01127002,
      "auxiliary_loss_mlp": 0.01045124,
      "balance_loss_clip": 1.04498827,
      "balance_loss_mlp": 1.02708793,
      "epoch": 0.25618517961821735,
      "flos": 27012329481600.0,
      "grad_norm": 1.8419885962491196,
      "language_loss": 0.76981723,
      "learning_rate": 3.4869062481564875e-06,
      "loss": 0.79153848,
      "num_input_tokens_seen": 92137570,
      "step": 4261,
      "time_per_iteration": 2.548661947250366
    },
    {
      "auxiliary_loss_clip": 0.01138182,
      "auxiliary_loss_mlp": 0.01036346,
      "balance_loss_clip": 1.04749751,
      "balance_loss_mlp": 1.02032471,
      "epoch": 0.2562453028708853,
      "flos": 23068296322560.0,
      "grad_norm": 1.755725320850985,
      "language_loss": 0.83410358,
      "learning_rate": 3.486645752648842e-06,
      "loss": 0.85584885,
      "num_input_tokens_seen": 92157625,
      "step": 4262,
      "time_per_iteration": 2.5282111167907715
    },
    {
      "auxiliary_loss_clip": 0.01124862,
      "auxiliary_loss_mlp": 0.01046502,
      "balance_loss_clip": 1.04495347,
      "balance_loss_mlp": 1.02807212,
      "epoch": 0.2563054261235533,
      "flos": 15120942606720.0,
      "grad_norm": 2.956504339816012,
      "language_loss": 0.73483956,
      "learning_rate": 3.4863852007674405e-06,
      "loss": 0.75655317,
      "num_input_tokens_seen": 92175350,
      "step": 4263,
      "time_per_iteration": 2.4598124027252197
    },
    {
      "auxiliary_loss_clip": 0.01107571,
      "auxiliary_loss_mlp": 0.00799964,
      "balance_loss_clip": 1.05061722,
      "balance_loss_mlp": 1.02069402,
      "epoch": 0.25636554937622125,
      "flos": 27854865872640.0,
      "grad_norm": 2.196645328202278,
      "language_loss": 0.82722867,
      "learning_rate": 3.486124592522163e-06,
      "loss": 0.846304,
      "num_input_tokens_seen": 92196070,
      "step": 4264,
      "time_per_iteration": 2.594109058380127
    },
    {
      "auxiliary_loss_clip": 0.01122692,
      "auxiliary_loss_mlp": 0.01046667,
      "balance_loss_clip": 1.04783094,
      "balance_loss_mlp": 1.02864218,
      "epoch": 0.2564256726288892,
      "flos": 28906509288960.0,
      "grad_norm": 1.716823220388212,
      "language_loss": 0.74581778,
      "learning_rate": 3.4858639279228924e-06,
      "loss": 0.76751137,
      "num_input_tokens_seen": 92216310,
      "step": 4265,
      "time_per_iteration": 2.5553011894226074
    },
    {
      "auxiliary_loss_clip": 0.01103464,
      "auxiliary_loss_mlp": 0.01038492,
      "balance_loss_clip": 1.04165602,
      "balance_loss_mlp": 1.02113521,
      "epoch": 0.2564857958815572,
      "flos": 18514931823360.0,
      "grad_norm": 1.659260842420772,
      "language_loss": 0.81608248,
      "learning_rate": 3.485603206979513e-06,
      "loss": 0.83750206,
      "num_input_tokens_seen": 92234510,
      "step": 4266,
      "time_per_iteration": 2.5572922229766846
    },
    {
      "auxiliary_loss_clip": 0.01076585,
      "auxiliary_loss_mlp": 0.0104554,
      "balance_loss_clip": 1.04594505,
      "balance_loss_mlp": 1.02685964,
      "epoch": 0.25654591913422514,
      "flos": 25808280658560.0,
      "grad_norm": 1.5922476894083235,
      "language_loss": 0.79321229,
      "learning_rate": 3.4853424297019103e-06,
      "loss": 0.81443357,
      "num_input_tokens_seen": 92254070,
      "step": 4267,
      "time_per_iteration": 2.605055809020996
    },
    {
      "auxiliary_loss_clip": 0.01089267,
      "auxiliary_loss_mlp": 0.01052449,
      "balance_loss_clip": 1.04506159,
      "balance_loss_mlp": 1.03398395,
      "epoch": 0.2566060423868931,
      "flos": 19099665325440.0,
      "grad_norm": 1.6587867457352752,
      "language_loss": 0.79010189,
      "learning_rate": 3.4850815960999736e-06,
      "loss": 0.81151903,
      "num_input_tokens_seen": 92275060,
      "step": 4268,
      "time_per_iteration": 2.5922610759735107
    },
    {
      "auxiliary_loss_clip": 0.01100504,
      "auxiliary_loss_mlp": 0.00801056,
      "balance_loss_clip": 1.0475899,
      "balance_loss_mlp": 1.01867104,
      "epoch": 0.25666616563956113,
      "flos": 23842674656640.0,
      "grad_norm": 1.6293652573550652,
      "language_loss": 0.68179679,
      "learning_rate": 3.484820706183595e-06,
      "loss": 0.70081234,
      "num_input_tokens_seen": 92293610,
      "step": 4269,
      "time_per_iteration": 2.5811843872070312
    },
    {
      "auxiliary_loss_clip": 0.01118993,
      "auxiliary_loss_mlp": 0.01040183,
      "balance_loss_clip": 1.049788,
      "balance_loss_mlp": 1.02230191,
      "epoch": 0.2567262888922291,
      "flos": 14604259420800.0,
      "grad_norm": 2.68568062447982,
      "language_loss": 0.79041356,
      "learning_rate": 3.484559759962666e-06,
      "loss": 0.81200534,
      "num_input_tokens_seen": 92308305,
      "step": 4270,
      "time_per_iteration": 2.4964094161987305
    },
    {
      "auxiliary_loss_clip": 0.01091696,
      "auxiliary_loss_mlp": 0.0104358,
      "balance_loss_clip": 1.04449821,
      "balance_loss_mlp": 1.02346981,
      "epoch": 0.25678641214489706,
      "flos": 32923117877760.0,
      "grad_norm": 2.250891638801971,
      "language_loss": 0.68206286,
      "learning_rate": 3.4842987574470816e-06,
      "loss": 0.70341569,
      "num_input_tokens_seen": 92329875,
      "step": 4271,
      "time_per_iteration": 2.7033114433288574
    },
    {
      "auxiliary_loss_clip": 0.0112868,
      "auxiliary_loss_mlp": 0.00799538,
      "balance_loss_clip": 1.04659271,
      "balance_loss_mlp": 1.021626,
      "epoch": 0.256846535397565,
      "flos": 24098933260800.0,
      "grad_norm": 1.3387401410314024,
      "language_loss": 0.87680429,
      "learning_rate": 3.4840376986467403e-06,
      "loss": 0.89608645,
      "num_input_tokens_seen": 92348780,
      "step": 4272,
      "time_per_iteration": 2.5421268939971924
    },
    {
      "auxiliary_loss_clip": 0.011139,
      "auxiliary_loss_mlp": 0.0104264,
      "balance_loss_clip": 1.04792595,
      "balance_loss_mlp": 1.02469945,
      "epoch": 0.256906658650233,
      "flos": 19718441942400.0,
      "grad_norm": 2.181205162885094,
      "language_loss": 0.81756425,
      "learning_rate": 3.483776583571541e-06,
      "loss": 0.83912969,
      "num_input_tokens_seen": 92368175,
      "step": 4273,
      "time_per_iteration": 2.5358664989471436
    },
    {
      "auxiliary_loss_clip": 0.01093086,
      "auxiliary_loss_mlp": 0.01043857,
      "balance_loss_clip": 1.04305995,
      "balance_loss_mlp": 1.02596366,
      "epoch": 0.25696678190290095,
      "flos": 22926018551040.0,
      "grad_norm": 1.5090656170182468,
      "language_loss": 0.77242714,
      "learning_rate": 3.4835154122313846e-06,
      "loss": 0.79379654,
      "num_input_tokens_seen": 92387755,
      "step": 4274,
      "time_per_iteration": 2.5790083408355713
    },
    {
      "auxiliary_loss_clip": 0.01108341,
      "auxiliary_loss_mlp": 0.01036086,
      "balance_loss_clip": 1.04319179,
      "balance_loss_mlp": 1.01865816,
      "epoch": 0.2570269051555689,
      "flos": 27307838672640.0,
      "grad_norm": 1.6144211663080767,
      "language_loss": 0.83899903,
      "learning_rate": 3.4832541846361743e-06,
      "loss": 0.86044335,
      "num_input_tokens_seen": 92409850,
      "step": 4275,
      "time_per_iteration": 2.6044628620147705
    },
    {
      "auxiliary_loss_clip": 0.01113526,
      "auxiliary_loss_mlp": 0.01039132,
      "balance_loss_clip": 1.04747784,
      "balance_loss_mlp": 1.02103555,
      "epoch": 0.2570870284082369,
      "flos": 27563414918400.0,
      "grad_norm": 1.9785093636167983,
      "language_loss": 0.78262246,
      "learning_rate": 3.4829929007958175e-06,
      "loss": 0.80414903,
      "num_input_tokens_seen": 92431250,
      "step": 4276,
      "time_per_iteration": 2.5839807987213135
    },
    {
      "auxiliary_loss_clip": 0.01130526,
      "auxiliary_loss_mlp": 0.01045013,
      "balance_loss_clip": 1.05004811,
      "balance_loss_mlp": 1.02787101,
      "epoch": 0.25714715166090485,
      "flos": 28730834847360.0,
      "grad_norm": 1.5836912341271765,
      "language_loss": 0.79331613,
      "learning_rate": 3.4827315607202214e-06,
      "loss": 0.81507158,
      "num_input_tokens_seen": 92452065,
      "step": 4277,
      "time_per_iteration": 2.5692403316497803
    },
    {
      "auxiliary_loss_clip": 0.01135525,
      "auxiliary_loss_mlp": 0.01038625,
      "balance_loss_clip": 1.0456754,
      "balance_loss_mlp": 1.02232885,
      "epoch": 0.2572072749135728,
      "flos": 20116152305280.0,
      "grad_norm": 2.2702311326926,
      "language_loss": 0.78513592,
      "learning_rate": 3.482470164419295e-06,
      "loss": 0.80687749,
      "num_input_tokens_seen": 92470025,
      "step": 4278,
      "time_per_iteration": 2.467740774154663
    },
    {
      "auxiliary_loss_clip": 0.01119023,
      "auxiliary_loss_mlp": 0.0104335,
      "balance_loss_clip": 1.04783249,
      "balance_loss_mlp": 1.02610004,
      "epoch": 0.2572673981662408,
      "flos": 26030855283840.0,
      "grad_norm": 3.1081899400013464,
      "language_loss": 0.74423659,
      "learning_rate": 3.482208711902952e-06,
      "loss": 0.76586032,
      "num_input_tokens_seen": 92489825,
      "step": 4279,
      "time_per_iteration": 2.571587324142456
    },
    {
      "auxiliary_loss_clip": 0.0112701,
      "auxiliary_loss_mlp": 0.01051158,
      "balance_loss_clip": 1.04582095,
      "balance_loss_mlp": 1.03405118,
      "epoch": 0.25732752141890874,
      "flos": 16106618695680.0,
      "grad_norm": 2.7488843339333235,
      "language_loss": 0.85104078,
      "learning_rate": 3.4819472031811065e-06,
      "loss": 0.8728224,
      "num_input_tokens_seen": 92507270,
      "step": 4280,
      "time_per_iteration": 2.4687283039093018
    },
    {
      "auxiliary_loss_clip": 0.01127492,
      "auxiliary_loss_mlp": 0.01040304,
      "balance_loss_clip": 1.04679573,
      "balance_loss_mlp": 1.02262545,
      "epoch": 0.2573876446715767,
      "flos": 22524429519360.0,
      "grad_norm": 2.4675384330036407,
      "language_loss": 0.78568369,
      "learning_rate": 3.4816856382636744e-06,
      "loss": 0.8073616,
      "num_input_tokens_seen": 92526300,
      "step": 4281,
      "time_per_iteration": 3.886716842651367
    },
    {
      "auxiliary_loss_clip": 0.01104253,
      "auxiliary_loss_mlp": 0.01038146,
      "balance_loss_clip": 1.04692113,
      "balance_loss_mlp": 1.02089643,
      "epoch": 0.2574477679242447,
      "flos": 23950837486080.0,
      "grad_norm": 1.6965105252517332,
      "language_loss": 0.87348092,
      "learning_rate": 3.4814240171605737e-06,
      "loss": 0.89490497,
      "num_input_tokens_seen": 92546465,
      "step": 4282,
      "time_per_iteration": 2.5583720207214355
    },
    {
      "auxiliary_loss_clip": 0.01140688,
      "auxiliary_loss_mlp": 0.01047402,
      "balance_loss_clip": 1.0486536,
      "balance_loss_mlp": 1.03050983,
      "epoch": 0.2575078911769127,
      "flos": 21981711951360.0,
      "grad_norm": 1.4343370138194094,
      "language_loss": 0.70238447,
      "learning_rate": 3.4811623398817267e-06,
      "loss": 0.72426534,
      "num_input_tokens_seen": 92567260,
      "step": 4283,
      "time_per_iteration": 3.8846657276153564
    },
    {
      "auxiliary_loss_clip": 0.0113632,
      "auxiliary_loss_mlp": 0.00797711,
      "balance_loss_clip": 1.05005956,
      "balance_loss_mlp": 1.02026296,
      "epoch": 0.25756801442958066,
      "flos": 21945406279680.0,
      "grad_norm": 1.8586936704082038,
      "language_loss": 0.80360806,
      "learning_rate": 3.4809006064370553e-06,
      "loss": 0.82294834,
      "num_input_tokens_seen": 92585425,
      "step": 4284,
      "time_per_iteration": 2.485323190689087
    },
    {
      "auxiliary_loss_clip": 0.01104155,
      "auxiliary_loss_mlp": 0.01045104,
      "balance_loss_clip": 1.06230116,
      "balance_loss_mlp": 1.02959478,
      "epoch": 0.2576281376822486,
      "flos": 35261980058880.0,
      "grad_norm": 2.2164560290648585,
      "language_loss": 0.69986564,
      "learning_rate": 3.4806388168364835e-06,
      "loss": 0.72135824,
      "num_input_tokens_seen": 92604770,
      "step": 4285,
      "time_per_iteration": 2.7337234020233154
    },
    {
      "auxiliary_loss_clip": 0.01120559,
      "auxiliary_loss_mlp": 0.01041919,
      "balance_loss_clip": 1.04921699,
      "balance_loss_mlp": 1.02569485,
      "epoch": 0.2576882609349166,
      "flos": 14132285688960.0,
      "grad_norm": 2.131717672758683,
      "language_loss": 0.58454341,
      "learning_rate": 3.4803769710899402e-06,
      "loss": 0.60616821,
      "num_input_tokens_seen": 92622635,
      "step": 4286,
      "time_per_iteration": 2.5230324268341064
    },
    {
      "auxiliary_loss_clip": 0.01137081,
      "auxiliary_loss_mlp": 0.01050862,
      "balance_loss_clip": 1.05513835,
      "balance_loss_mlp": 1.03364861,
      "epoch": 0.25774838418758456,
      "flos": 23258336204160.0,
      "grad_norm": 1.9921226661956455,
      "language_loss": 0.64447463,
      "learning_rate": 3.480115069207354e-06,
      "loss": 0.66635406,
      "num_input_tokens_seen": 92642960,
      "step": 4287,
      "time_per_iteration": 3.9076879024505615
    },
    {
      "auxiliary_loss_clip": 0.01121383,
      "auxiliary_loss_mlp": 0.01041021,
      "balance_loss_clip": 1.0517416,
      "balance_loss_mlp": 1.02318752,
      "epoch": 0.2578085074402525,
      "flos": 22601745544320.0,
      "grad_norm": 1.907799843331985,
      "language_loss": 0.71493471,
      "learning_rate": 3.4798531111986557e-06,
      "loss": 0.73655874,
      "num_input_tokens_seen": 92662455,
      "step": 4288,
      "time_per_iteration": 2.5470919609069824
    },
    {
      "auxiliary_loss_clip": 0.0110921,
      "auxiliary_loss_mlp": 0.01040513,
      "balance_loss_clip": 1.05219197,
      "balance_loss_mlp": 1.02460492,
      "epoch": 0.2578686306929205,
      "flos": 24571840746240.0,
      "grad_norm": 1.582076261160591,
      "language_loss": 0.77113163,
      "learning_rate": 3.4795910970737786e-06,
      "loss": 0.79262882,
      "num_input_tokens_seen": 92683520,
      "step": 4289,
      "time_per_iteration": 2.6136324405670166
    },
    {
      "auxiliary_loss_clip": 0.01138363,
      "auxiliary_loss_mlp": 0.00797868,
      "balance_loss_clip": 1.04776001,
      "balance_loss_mlp": 1.01614797,
      "epoch": 0.25792875394558845,
      "flos": 18113953322880.0,
      "grad_norm": 2.169361696540674,
      "language_loss": 0.85465837,
      "learning_rate": 3.4793290268426592e-06,
      "loss": 0.87402064,
      "num_input_tokens_seen": 92701450,
      "step": 4290,
      "time_per_iteration": 2.4610414505004883
    },
    {
      "auxiliary_loss_clip": 0.01106441,
      "auxiliary_loss_mlp": 0.01056347,
      "balance_loss_clip": 1.04692769,
      "balance_loss_mlp": 1.03630793,
      "epoch": 0.2579888771982564,
      "flos": 17712902995200.0,
      "grad_norm": 2.240185939215405,
      "language_loss": 0.72462511,
      "learning_rate": 3.4790669005152354e-06,
      "loss": 0.74625295,
      "num_input_tokens_seen": 92720355,
      "step": 4291,
      "time_per_iteration": 2.52471661567688
    },
    {
      "auxiliary_loss_clip": 0.01145325,
      "auxiliary_loss_mlp": 0.01044476,
      "balance_loss_clip": 1.0512569,
      "balance_loss_mlp": 1.02605844,
      "epoch": 0.2580490004509244,
      "flos": 16434878112000.0,
      "grad_norm": 2.374984192610286,
      "language_loss": 0.8021847,
      "learning_rate": 3.4788047181014458e-06,
      "loss": 0.82408273,
      "num_input_tokens_seen": 92736755,
      "step": 4292,
      "time_per_iteration": 2.434584617614746
    },
    {
      "auxiliary_loss_clip": 0.01145265,
      "auxiliary_loss_mlp": 0.01041032,
      "balance_loss_clip": 1.05210721,
      "balance_loss_mlp": 1.02332926,
      "epoch": 0.25810912370359235,
      "flos": 33835141128960.0,
      "grad_norm": 2.6186875607541067,
      "language_loss": 0.67857945,
      "learning_rate": 3.4785424796112337e-06,
      "loss": 0.70044249,
      "num_input_tokens_seen": 92757655,
      "step": 4293,
      "time_per_iteration": 2.571488380432129
    },
    {
      "auxiliary_loss_clip": 0.01100147,
      "auxiliary_loss_mlp": 0.01041131,
      "balance_loss_clip": 1.04961753,
      "balance_loss_mlp": 1.02435827,
      "epoch": 0.2581692469562603,
      "flos": 25192197561600.0,
      "grad_norm": 1.8780374950730028,
      "language_loss": 0.75602859,
      "learning_rate": 3.478280185054542e-06,
      "loss": 0.77744138,
      "num_input_tokens_seen": 92776100,
      "step": 4294,
      "time_per_iteration": 2.598740577697754
    },
    {
      "auxiliary_loss_clip": 0.01094854,
      "auxiliary_loss_mlp": 0.01051318,
      "balance_loss_clip": 1.04619396,
      "balance_loss_mlp": 1.03144598,
      "epoch": 0.2582293702089283,
      "flos": 34932212271360.0,
      "grad_norm": 1.9953637378804363,
      "language_loss": 0.81120205,
      "learning_rate": 3.478017834441318e-06,
      "loss": 0.83266377,
      "num_input_tokens_seen": 92798880,
      "step": 4295,
      "time_per_iteration": 2.673055648803711
    },
    {
      "auxiliary_loss_clip": 0.01053476,
      "auxiliary_loss_mlp": 0.01044454,
      "balance_loss_clip": 1.05398846,
      "balance_loss_mlp": 1.02577424,
      "epoch": 0.2582894934615963,
      "flos": 26833746038400.0,
      "grad_norm": 1.9118231748290029,
      "language_loss": 0.72756565,
      "learning_rate": 3.4777554277815096e-06,
      "loss": 0.74854493,
      "num_input_tokens_seen": 92817750,
      "step": 4296,
      "time_per_iteration": 2.9798176288604736
    },
    {
      "auxiliary_loss_clip": 0.01087787,
      "auxiliary_loss_mlp": 0.01045803,
      "balance_loss_clip": 1.05587077,
      "balance_loss_mlp": 1.02731407,
      "epoch": 0.25834961671426426,
      "flos": 23515241253120.0,
      "grad_norm": 1.6239038428685129,
      "language_loss": 0.86971539,
      "learning_rate": 3.477492965085067e-06,
      "loss": 0.89105129,
      "num_input_tokens_seen": 92837995,
      "step": 4297,
      "time_per_iteration": 2.830334186553955
    },
    {
      "auxiliary_loss_clip": 0.0114407,
      "auxiliary_loss_mlp": 0.01051685,
      "balance_loss_clip": 1.05137229,
      "balance_loss_mlp": 1.0345192,
      "epoch": 0.25840973996693223,
      "flos": 22451028076800.0,
      "grad_norm": 1.7817813381239145,
      "language_loss": 0.84521413,
      "learning_rate": 3.477230446361943e-06,
      "loss": 0.86717165,
      "num_input_tokens_seen": 92857245,
      "step": 4298,
      "time_per_iteration": 2.4782094955444336
    },
    {
      "auxiliary_loss_clip": 0.01131021,
      "auxiliary_loss_mlp": 0.00797932,
      "balance_loss_clip": 1.0501492,
      "balance_loss_mlp": 1.01819611,
      "epoch": 0.2584698632196002,
      "flos": 11290854366720.0,
      "grad_norm": 2.264361981335166,
      "language_loss": 0.83932018,
      "learning_rate": 3.4769678716220927e-06,
      "loss": 0.85860968,
      "num_input_tokens_seen": 92873265,
      "step": 4299,
      "time_per_iteration": 2.491185188293457
    },
    {
      "auxiliary_loss_clip": 0.01114915,
      "auxiliary_loss_mlp": 0.01037871,
      "balance_loss_clip": 1.05451441,
      "balance_loss_mlp": 1.02119339,
      "epoch": 0.25852998647226816,
      "flos": 17929982839680.0,
      "grad_norm": 3.709325899304579,
      "language_loss": 0.82388949,
      "learning_rate": 3.4767052408754726e-06,
      "loss": 0.84541738,
      "num_input_tokens_seen": 92890880,
      "step": 4300,
      "time_per_iteration": 2.5094621181488037
    },
    {
      "auxiliary_loss_clip": 0.01131524,
      "auxiliary_loss_mlp": 0.01041505,
      "balance_loss_clip": 1.04924357,
      "balance_loss_mlp": 1.02387381,
      "epoch": 0.2585901097249361,
      "flos": 33256117889280.0,
      "grad_norm": 1.874298841074203,
      "language_loss": 0.67626262,
      "learning_rate": 3.4764425541320417e-06,
      "loss": 0.69799292,
      "num_input_tokens_seen": 92910770,
      "step": 4301,
      "time_per_iteration": 2.6223793029785156
    },
    {
      "auxiliary_loss_clip": 0.01135615,
      "auxiliary_loss_mlp": 0.01042044,
      "balance_loss_clip": 1.05178702,
      "balance_loss_mlp": 1.02372169,
      "epoch": 0.2586502329776041,
      "flos": 18441278985600.0,
      "grad_norm": 2.4610688310578746,
      "language_loss": 0.81512749,
      "learning_rate": 3.4761798114017617e-06,
      "loss": 0.83690405,
      "num_input_tokens_seen": 92929520,
      "step": 4302,
      "time_per_iteration": 2.4644076824188232
    },
    {
      "auxiliary_loss_clip": 0.01101069,
      "auxiliary_loss_mlp": 0.01051676,
      "balance_loss_clip": 1.05248427,
      "balance_loss_mlp": 1.03374708,
      "epoch": 0.25871035623027205,
      "flos": 17968120104960.0,
      "grad_norm": 1.7571093141848448,
      "language_loss": 0.92334336,
      "learning_rate": 3.475917012694595e-06,
      "loss": 0.94487083,
      "num_input_tokens_seen": 92947890,
      "step": 4303,
      "time_per_iteration": 2.5807883739471436
    },
    {
      "auxiliary_loss_clip": 0.01137401,
      "auxiliary_loss_mlp": 0.01040152,
      "balance_loss_clip": 1.05324376,
      "balance_loss_mlp": 1.02227068,
      "epoch": 0.25877047948294,
      "flos": 27777729415680.0,
      "grad_norm": 2.099282250872492,
      "language_loss": 0.67070353,
      "learning_rate": 3.475654158020507e-06,
      "loss": 0.69247907,
      "num_input_tokens_seen": 92967690,
      "step": 4304,
      "time_per_iteration": 2.567720413208008
    },
    {
      "auxiliary_loss_clip": 0.0110925,
      "auxiliary_loss_mlp": 0.01045098,
      "balance_loss_clip": 1.04842925,
      "balance_loss_mlp": 1.02817082,
      "epoch": 0.258830602735608,
      "flos": 27125843437440.0,
      "grad_norm": 2.0658529677280133,
      "language_loss": 0.72147042,
      "learning_rate": 3.4753912473894657e-06,
      "loss": 0.74301392,
      "num_input_tokens_seen": 92986830,
      "step": 4305,
      "time_per_iteration": 2.625807285308838
    },
    {
      "auxiliary_loss_clip": 0.01112111,
      "auxiliary_loss_mlp": 0.00797522,
      "balance_loss_clip": 1.05853105,
      "balance_loss_mlp": 1.01642895,
      "epoch": 0.25889072598827595,
      "flos": 17891486438400.0,
      "grad_norm": 1.9310047252969054,
      "language_loss": 0.76028717,
      "learning_rate": 3.4751282808114403e-06,
      "loss": 0.77938348,
      "num_input_tokens_seen": 93002740,
      "step": 4306,
      "time_per_iteration": 2.569707155227661
    },
    {
      "auxiliary_loss_clip": 0.01038845,
      "auxiliary_loss_mlp": 0.0102569,
      "balance_loss_clip": 1.02955842,
      "balance_loss_mlp": 1.02372324,
      "epoch": 0.2589508492409439,
      "flos": 53934955724160.0,
      "grad_norm": 0.8684610995872517,
      "language_loss": 0.57161283,
      "learning_rate": 3.474865258296403e-06,
      "loss": 0.59225821,
      "num_input_tokens_seen": 93058645,
      "step": 4307,
      "time_per_iteration": 3.0661473274230957
    },
    {
      "auxiliary_loss_clip": 0.01120957,
      "auxiliary_loss_mlp": 0.0103979,
      "balance_loss_clip": 1.05059099,
      "balance_loss_mlp": 1.02249253,
      "epoch": 0.2590109724936119,
      "flos": 22125785402880.0,
      "grad_norm": 1.5360734267448695,
      "language_loss": 0.71434838,
      "learning_rate": 3.474602179854327e-06,
      "loss": 0.73595583,
      "num_input_tokens_seen": 93077140,
      "step": 4308,
      "time_per_iteration": 2.555570125579834
    },
    {
      "auxiliary_loss_clip": 0.01144346,
      "auxiliary_loss_mlp": 0.01048134,
      "balance_loss_clip": 1.04951274,
      "balance_loss_mlp": 1.03026462,
      "epoch": 0.2590710957462799,
      "flos": 13474294398720.0,
      "grad_norm": 1.9383156167940314,
      "language_loss": 0.8453536,
      "learning_rate": 3.4743390454951886e-06,
      "loss": 0.8672784,
      "num_input_tokens_seen": 93093580,
      "step": 4309,
      "time_per_iteration": 2.469115734100342
    },
    {
      "auxiliary_loss_clip": 0.01131582,
      "auxiliary_loss_mlp": 0.01043641,
      "balance_loss_clip": 1.05199981,
      "balance_loss_mlp": 1.02739322,
      "epoch": 0.25913121899894787,
      "flos": 22307098279680.0,
      "grad_norm": 1.4364121391358387,
      "language_loss": 0.84343874,
      "learning_rate": 3.474075855228966e-06,
      "loss": 0.86519098,
      "num_input_tokens_seen": 93112345,
      "step": 4310,
      "time_per_iteration": 2.522749185562134
    },
    {
      "auxiliary_loss_clip": 0.01135287,
      "auxiliary_loss_mlp": 0.01050611,
      "balance_loss_clip": 1.05208492,
      "balance_loss_mlp": 1.03299189,
      "epoch": 0.25919134225161583,
      "flos": 25811728364160.0,
      "grad_norm": 1.8042846745674028,
      "language_loss": 0.77197385,
      "learning_rate": 3.473812609065639e-06,
      "loss": 0.79383278,
      "num_input_tokens_seen": 93131545,
      "step": 4311,
      "time_per_iteration": 2.5374698638916016
    },
    {
      "auxiliary_loss_clip": 0.01107237,
      "auxiliary_loss_mlp": 0.01051686,
      "balance_loss_clip": 1.04918838,
      "balance_loss_mlp": 1.03307748,
      "epoch": 0.2592514655042838,
      "flos": 31212262108800.0,
      "grad_norm": 1.9830086721589142,
      "language_loss": 0.72743535,
      "learning_rate": 3.4735493070151904e-06,
      "loss": 0.74902457,
      "num_input_tokens_seen": 93150730,
      "step": 4312,
      "time_per_iteration": 2.6053314208984375
    },
    {
      "auxiliary_loss_clip": 0.01140907,
      "auxiliary_loss_mlp": 0.01047105,
      "balance_loss_clip": 1.04838264,
      "balance_loss_mlp": 1.02943873,
      "epoch": 0.25931158875695176,
      "flos": 18474998878080.0,
      "grad_norm": 1.7922671042561338,
      "language_loss": 0.7009573,
      "learning_rate": 3.4732859490876044e-06,
      "loss": 0.72283745,
      "num_input_tokens_seen": 93167895,
      "step": 4313,
      "time_per_iteration": 2.456695318222046
    },
    {
      "auxiliary_loss_clip": 0.01142924,
      "auxiliary_loss_mlp": 0.01054284,
      "balance_loss_clip": 1.05155659,
      "balance_loss_mlp": 1.03814328,
      "epoch": 0.2593717120096197,
      "flos": 19207935895680.0,
      "grad_norm": 1.625527915626716,
      "language_loss": 0.80365914,
      "learning_rate": 3.473022535292867e-06,
      "loss": 0.82563126,
      "num_input_tokens_seen": 93187650,
      "step": 4314,
      "time_per_iteration": 2.4841270446777344
    },
    {
      "auxiliary_loss_clip": 0.01111106,
      "auxiliary_loss_mlp": 0.0105176,
      "balance_loss_clip": 1.05653501,
      "balance_loss_mlp": 1.03298509,
      "epoch": 0.2594318352622877,
      "flos": 31248100903680.0,
      "grad_norm": 2.1540535133644867,
      "language_loss": 0.67564428,
      "learning_rate": 3.472759065640968e-06,
      "loss": 0.69727296,
      "num_input_tokens_seen": 93207370,
      "step": 4315,
      "time_per_iteration": 2.655148506164551
    },
    {
      "auxiliary_loss_clip": 0.0109984,
      "auxiliary_loss_mlp": 0.01050248,
      "balance_loss_clip": 1.05578899,
      "balance_loss_mlp": 1.03348696,
      "epoch": 0.25949195851495566,
      "flos": 22237144542720.0,
      "grad_norm": 1.536865622680474,
      "language_loss": 0.79594028,
      "learning_rate": 3.4724955401418976e-06,
      "loss": 0.81744123,
      "num_input_tokens_seen": 93227925,
      "step": 4316,
      "time_per_iteration": 2.5955095291137695
    },
    {
      "auxiliary_loss_clip": 0.01099143,
      "auxiliary_loss_mlp": 0.01047348,
      "balance_loss_clip": 1.05579841,
      "balance_loss_mlp": 1.03000307,
      "epoch": 0.2595520817676236,
      "flos": 28075716645120.0,
      "grad_norm": 1.7113342047685036,
      "language_loss": 0.78233808,
      "learning_rate": 3.4722319588056487e-06,
      "loss": 0.80380297,
      "num_input_tokens_seen": 93250020,
      "step": 4317,
      "time_per_iteration": 2.6446585655212402
    },
    {
      "auxiliary_loss_clip": 0.01146443,
      "auxiliary_loss_mlp": 0.01055735,
      "balance_loss_clip": 1.05391908,
      "balance_loss_mlp": 1.03741264,
      "epoch": 0.2596122050202916,
      "flos": 20190954378240.0,
      "grad_norm": 1.905227505821882,
      "language_loss": 0.77810329,
      "learning_rate": 3.4719683216422163e-06,
      "loss": 0.80012506,
      "num_input_tokens_seen": 93269070,
      "step": 4318,
      "time_per_iteration": 2.4877078533172607
    },
    {
      "auxiliary_loss_clip": 0.0114319,
      "auxiliary_loss_mlp": 0.01047727,
      "balance_loss_clip": 1.05140805,
      "balance_loss_mlp": 1.02848649,
      "epoch": 0.25967232827295955,
      "flos": 22527949052160.0,
      "grad_norm": 1.7594511300737965,
      "language_loss": 0.76270759,
      "learning_rate": 3.471704628661598e-06,
      "loss": 0.78461671,
      "num_input_tokens_seen": 93290250,
      "step": 4319,
      "time_per_iteration": 2.504451274871826
    },
    {
      "auxiliary_loss_clip": 0.01119863,
      "auxiliary_loss_mlp": 0.01043001,
      "balance_loss_clip": 1.05329406,
      "balance_loss_mlp": 1.02589452,
      "epoch": 0.2597324515256275,
      "flos": 21068252156160.0,
      "grad_norm": 1.7414394983796198,
      "language_loss": 0.76944822,
      "learning_rate": 3.4714408798737925e-06,
      "loss": 0.7910769,
      "num_input_tokens_seen": 93310090,
      "step": 4320,
      "time_per_iteration": 4.719043493270874
    },
    {
      "auxiliary_loss_clip": 0.01113514,
      "auxiliary_loss_mlp": 0.01043802,
      "balance_loss_clip": 1.05157399,
      "balance_loss_mlp": 1.02576518,
      "epoch": 0.2597925747782955,
      "flos": 22050013662720.0,
      "grad_norm": 1.4556809847023873,
      "language_loss": 0.71682805,
      "learning_rate": 3.471177075288801e-06,
      "loss": 0.73840123,
      "num_input_tokens_seen": 93329570,
      "step": 4321,
      "time_per_iteration": 2.597766876220703
    },
    {
      "auxiliary_loss_clip": 0.01121439,
      "auxiliary_loss_mlp": 0.01054612,
      "balance_loss_clip": 1.05148494,
      "balance_loss_mlp": 1.03470397,
      "epoch": 0.2598526980309635,
      "flos": 19536949497600.0,
      "grad_norm": 2.5833491319651958,
      "language_loss": 0.74848419,
      "learning_rate": 3.4709132149166277e-06,
      "loss": 0.77024472,
      "num_input_tokens_seen": 93347920,
      "step": 4322,
      "time_per_iteration": 3.9139814376831055
    },
    {
      "auxiliary_loss_clip": 0.01113039,
      "auxiliary_loss_mlp": 0.01045727,
      "balance_loss_clip": 1.05149674,
      "balance_loss_mlp": 1.02769089,
      "epoch": 0.25991282128363147,
      "flos": 24495207079680.0,
      "grad_norm": 2.0147062610199304,
      "language_loss": 0.73401642,
      "learning_rate": 3.470649298767278e-06,
      "loss": 0.75560403,
      "num_input_tokens_seen": 93367145,
      "step": 4323,
      "time_per_iteration": 2.606518507003784
    },
    {
      "auxiliary_loss_clip": 0.01139452,
      "auxiliary_loss_mlp": 0.00794283,
      "balance_loss_clip": 1.05182719,
      "balance_loss_mlp": 1.01093149,
      "epoch": 0.25997294453629943,
      "flos": 24201457655040.0,
      "grad_norm": 2.3369812444527946,
      "language_loss": 0.67120755,
      "learning_rate": 3.4703853268507597e-06,
      "loss": 0.69054484,
      "num_input_tokens_seen": 93386555,
      "step": 4324,
      "time_per_iteration": 2.526766538619995
    },
    {
      "auxiliary_loss_clip": 0.01094633,
      "auxiliary_loss_mlp": 0.01043809,
      "balance_loss_clip": 1.0535562,
      "balance_loss_mlp": 1.0274539,
      "epoch": 0.2600330677889674,
      "flos": 31431460855680.0,
      "grad_norm": 1.948711840672533,
      "language_loss": 0.70369822,
      "learning_rate": 3.470121299177082e-06,
      "loss": 0.72508264,
      "num_input_tokens_seen": 93405590,
      "step": 4325,
      "time_per_iteration": 2.689810037612915
    },
    {
      "auxiliary_loss_clip": 0.01133219,
      "auxiliary_loss_mlp": 0.01037349,
      "balance_loss_clip": 1.05007637,
      "balance_loss_mlp": 1.01938391,
      "epoch": 0.26009319104163536,
      "flos": 32266527217920.0,
      "grad_norm": 1.7884317276415618,
      "language_loss": 0.73248255,
      "learning_rate": 3.469857215756257e-06,
      "loss": 0.7541883,
      "num_input_tokens_seen": 93424750,
      "step": 4326,
      "time_per_iteration": 5.432117462158203
    },
    {
      "auxiliary_loss_clip": 0.0111517,
      "auxiliary_loss_mlp": 0.00794392,
      "balance_loss_clip": 1.04967153,
      "balance_loss_mlp": 1.01086843,
      "epoch": 0.26015331429430333,
      "flos": 26286754752000.0,
      "grad_norm": 1.9501609860226394,
      "language_loss": 0.86770654,
      "learning_rate": 3.4695930765982997e-06,
      "loss": 0.88680214,
      "num_input_tokens_seen": 93443465,
      "step": 4327,
      "time_per_iteration": 2.5746920108795166
    },
    {
      "auxiliary_loss_clip": 0.01150133,
      "auxiliary_loss_mlp": 0.00796744,
      "balance_loss_clip": 1.05507922,
      "balance_loss_mlp": 1.01390171,
      "epoch": 0.2602134375469713,
      "flos": 21142335957120.0,
      "grad_norm": 1.5786999783214695,
      "language_loss": 0.80524671,
      "learning_rate": 3.4693288817132255e-06,
      "loss": 0.8247155,
      "num_input_tokens_seen": 93462580,
      "step": 4328,
      "time_per_iteration": 2.5092906951904297
    },
    {
      "auxiliary_loss_clip": 0.01118202,
      "auxiliary_loss_mlp": 0.00792321,
      "balance_loss_clip": 1.04910231,
      "balance_loss_mlp": 1.01003909,
      "epoch": 0.26027356079963926,
      "flos": 25921327737600.0,
      "grad_norm": 1.438560125490664,
      "language_loss": 0.88107479,
      "learning_rate": 3.4690646311110525e-06,
      "loss": 0.90017998,
      "num_input_tokens_seen": 93482790,
      "step": 4329,
      "time_per_iteration": 2.5702037811279297
    },
    {
      "auxiliary_loss_clip": 0.01143982,
      "auxiliary_loss_mlp": 0.0104051,
      "balance_loss_clip": 1.05399287,
      "balance_loss_mlp": 1.02384424,
      "epoch": 0.2603336840523072,
      "flos": 26359222440960.0,
      "grad_norm": 2.0050833458655397,
      "language_loss": 0.7769537,
      "learning_rate": 3.468800324801802e-06,
      "loss": 0.79879868,
      "num_input_tokens_seen": 93498795,
      "step": 4330,
      "time_per_iteration": 2.524470329284668
    },
    {
      "auxiliary_loss_clip": 0.01146425,
      "auxiliary_loss_mlp": 0.01050687,
      "balance_loss_clip": 1.0536449,
      "balance_loss_mlp": 1.03328288,
      "epoch": 0.2603938073049752,
      "flos": 23513661054720.0,
      "grad_norm": 1.4580356957206388,
      "language_loss": 0.75740635,
      "learning_rate": 3.4685359627954958e-06,
      "loss": 0.7793774,
      "num_input_tokens_seen": 93518335,
      "step": 4331,
      "time_per_iteration": 2.474468231201172
    },
    {
      "auxiliary_loss_clip": 0.01130942,
      "auxiliary_loss_mlp": 0.01043477,
      "balance_loss_clip": 1.06179154,
      "balance_loss_mlp": 1.02670467,
      "epoch": 0.26045393055764315,
      "flos": 25374300537600.0,
      "grad_norm": 1.3750957576397314,
      "language_loss": 0.69333494,
      "learning_rate": 3.4682715451021584e-06,
      "loss": 0.71507913,
      "num_input_tokens_seen": 93539170,
      "step": 4332,
      "time_per_iteration": 2.5712621212005615
    },
    {
      "auxiliary_loss_clip": 0.0111626,
      "auxiliary_loss_mlp": 0.01049058,
      "balance_loss_clip": 1.0506351,
      "balance_loss_mlp": 1.0312717,
      "epoch": 0.2605140538103111,
      "flos": 27635272076160.0,
      "grad_norm": 2.131085289059811,
      "language_loss": 0.80100262,
      "learning_rate": 3.4680070717318174e-06,
      "loss": 0.8226558,
      "num_input_tokens_seen": 93558480,
      "step": 4333,
      "time_per_iteration": 2.6174464225769043
    },
    {
      "auxiliary_loss_clip": 0.0114315,
      "auxiliary_loss_mlp": 0.01040593,
      "balance_loss_clip": 1.05362654,
      "balance_loss_mlp": 1.02393973,
      "epoch": 0.2605741770629791,
      "flos": 13769839503360.0,
      "grad_norm": 2.2194680530795625,
      "language_loss": 0.8105405,
      "learning_rate": 3.467742542694501e-06,
      "loss": 0.83237791,
      "num_input_tokens_seen": 93575220,
      "step": 4334,
      "time_per_iteration": 2.425631284713745
    },
    {
      "auxiliary_loss_clip": 0.01124042,
      "auxiliary_loss_mlp": 0.0104175,
      "balance_loss_clip": 1.05481255,
      "balance_loss_mlp": 1.02352262,
      "epoch": 0.26063430031564705,
      "flos": 26031681296640.0,
      "grad_norm": 1.733119612950829,
      "language_loss": 0.7937693,
      "learning_rate": 3.46747795800024e-06,
      "loss": 0.81542724,
      "num_input_tokens_seen": 93597015,
      "step": 4335,
      "time_per_iteration": 2.5666167736053467
    },
    {
      "auxiliary_loss_clip": 0.0104311,
      "auxiliary_loss_mlp": 0.01021426,
      "balance_loss_clip": 1.02145386,
      "balance_loss_mlp": 1.0196023,
      "epoch": 0.26069442356831507,
      "flos": 62443809820800.0,
      "grad_norm": 0.8523053376293434,
      "language_loss": 0.60719335,
      "learning_rate": 3.467213317659068e-06,
      "loss": 0.62783873,
      "num_input_tokens_seen": 93657775,
      "step": 4336,
      "time_per_iteration": 3.090118169784546
    },
    {
      "auxiliary_loss_clip": 0.01112759,
      "auxiliary_loss_mlp": 0.01049381,
      "balance_loss_clip": 1.05626202,
      "balance_loss_mlp": 1.03209531,
      "epoch": 0.26075454682098304,
      "flos": 13626376583040.0,
      "grad_norm": 1.8067748553068412,
      "language_loss": 0.77059507,
      "learning_rate": 3.46694862168102e-06,
      "loss": 0.79221648,
      "num_input_tokens_seen": 93676145,
      "step": 4337,
      "time_per_iteration": 2.570322036743164
    },
    {
      "auxiliary_loss_clip": 0.01124625,
      "auxiliary_loss_mlp": 0.01046805,
      "balance_loss_clip": 1.05298734,
      "balance_loss_mlp": 1.02878118,
      "epoch": 0.260814670073651,
      "flos": 12126531260160.0,
      "grad_norm": 2.0938754207217953,
      "language_loss": 0.74802327,
      "learning_rate": 3.4666838700761334e-06,
      "loss": 0.7697376,
      "num_input_tokens_seen": 93692480,
      "step": 4338,
      "time_per_iteration": 2.553378105163574
    },
    {
      "auxiliary_loss_clip": 0.01139183,
      "auxiliary_loss_mlp": 0.0104467,
      "balance_loss_clip": 1.05332136,
      "balance_loss_mlp": 1.02632356,
      "epoch": 0.26087479332631897,
      "flos": 15122522805120.0,
      "grad_norm": 2.0903182747107873,
      "language_loss": 0.80266958,
      "learning_rate": 3.466419062854447e-06,
      "loss": 0.82450807,
      "num_input_tokens_seen": 93710165,
      "step": 4339,
      "time_per_iteration": 2.565016031265259
    },
    {
      "auxiliary_loss_clip": 0.01092035,
      "auxiliary_loss_mlp": 0.01042612,
      "balance_loss_clip": 1.05473995,
      "balance_loss_mlp": 1.02606559,
      "epoch": 0.26093491657898693,
      "flos": 24680937329280.0,
      "grad_norm": 1.5283632930736504,
      "language_loss": 0.76621288,
      "learning_rate": 3.4661542000260033e-06,
      "loss": 0.78755933,
      "num_input_tokens_seen": 93730185,
      "step": 4340,
      "time_per_iteration": 2.6806702613830566
    },
    {
      "auxiliary_loss_clip": 0.01079926,
      "auxiliary_loss_mlp": 0.01045089,
      "balance_loss_clip": 1.0510751,
      "balance_loss_mlp": 1.02695775,
      "epoch": 0.2609950398316549,
      "flos": 25116138512640.0,
      "grad_norm": 1.4507369348382404,
      "language_loss": 0.82588172,
      "learning_rate": 3.465889281600845e-06,
      "loss": 0.84713185,
      "num_input_tokens_seen": 93747690,
      "step": 4341,
      "time_per_iteration": 2.7596187591552734
    },
    {
      "auxiliary_loss_clip": 0.01145798,
      "auxiliary_loss_mlp": 0.01044881,
      "balance_loss_clip": 1.05402863,
      "balance_loss_mlp": 1.02651131,
      "epoch": 0.26105516308432286,
      "flos": 28548588216960.0,
      "grad_norm": 1.8590792570963286,
      "language_loss": 0.76490748,
      "learning_rate": 3.4656243075890183e-06,
      "loss": 0.78681433,
      "num_input_tokens_seen": 93767405,
      "step": 4342,
      "time_per_iteration": 2.628211259841919
    },
    {
      "auxiliary_loss_clip": 0.01139623,
      "auxiliary_loss_mlp": 0.01034317,
      "balance_loss_clip": 1.05396891,
      "balance_loss_mlp": 1.01586318,
      "epoch": 0.2611152863369908,
      "flos": 39530609447040.0,
      "grad_norm": 2.5263104630045206,
      "language_loss": 0.65793741,
      "learning_rate": 3.4653592780005707e-06,
      "loss": 0.67967677,
      "num_input_tokens_seen": 93789950,
      "step": 4343,
      "time_per_iteration": 2.6799204349517822
    },
    {
      "auxiliary_loss_clip": 0.01078756,
      "auxiliary_loss_mlp": 0.01053407,
      "balance_loss_clip": 1.05614293,
      "balance_loss_mlp": 1.03372598,
      "epoch": 0.2611754095896588,
      "flos": 13735329511680.0,
      "grad_norm": 1.8845952236487313,
      "language_loss": 0.73191047,
      "learning_rate": 3.465094192845553e-06,
      "loss": 0.75323212,
      "num_input_tokens_seen": 93807835,
      "step": 4344,
      "time_per_iteration": 2.5885727405548096
    },
    {
      "auxiliary_loss_clip": 0.01146157,
      "auxiliary_loss_mlp": 0.01043,
      "balance_loss_clip": 1.0541023,
      "balance_loss_mlp": 1.02473712,
      "epoch": 0.26123553284232676,
      "flos": 21506649649920.0,
      "grad_norm": 1.9991646639234057,
      "language_loss": 0.86644703,
      "learning_rate": 3.4648290521340165e-06,
      "loss": 0.88833863,
      "num_input_tokens_seen": 93825670,
      "step": 4345,
      "time_per_iteration": 2.4916841983795166
    },
    {
      "auxiliary_loss_clip": 0.01120438,
      "auxiliary_loss_mlp": 0.01041875,
      "balance_loss_clip": 1.0535959,
      "balance_loss_mlp": 1.02472043,
      "epoch": 0.2612956560949947,
      "flos": 21139786091520.0,
      "grad_norm": 1.84400567060217,
      "language_loss": 0.76431578,
      "learning_rate": 3.464563855876015e-06,
      "loss": 0.78593886,
      "num_input_tokens_seen": 93844045,
      "step": 4346,
      "time_per_iteration": 2.5186007022857666
    },
    {
      "auxiliary_loss_clip": 0.011354,
      "auxiliary_loss_mlp": 0.01043701,
      "balance_loss_clip": 1.05299759,
      "balance_loss_mlp": 1.02614164,
      "epoch": 0.2613557793476627,
      "flos": 25119011600640.0,
      "grad_norm": 1.446564548211201,
      "language_loss": 0.75851655,
      "learning_rate": 3.464298604081606e-06,
      "loss": 0.78030759,
      "num_input_tokens_seen": 93864380,
      "step": 4347,
      "time_per_iteration": 2.592986583709717
    },
    {
      "auxiliary_loss_clip": 0.01101102,
      "auxiliary_loss_mlp": 0.01039296,
      "balance_loss_clip": 1.04844522,
      "balance_loss_mlp": 1.02100956,
      "epoch": 0.26141590260033065,
      "flos": 26067699659520.0,
      "grad_norm": 1.6886710965101979,
      "language_loss": 0.73634154,
      "learning_rate": 3.4640332967608476e-06,
      "loss": 0.7577455,
      "num_input_tokens_seen": 93885475,
      "step": 4348,
      "time_per_iteration": 2.600609302520752
    },
    {
      "auxiliary_loss_clip": 0.01107802,
      "auxiliary_loss_mlp": 0.01047118,
      "balance_loss_clip": 1.05433822,
      "balance_loss_mlp": 1.02903438,
      "epoch": 0.2614760258529987,
      "flos": 25701518459520.0,
      "grad_norm": 2.0576385305303595,
      "language_loss": 0.90948963,
      "learning_rate": 3.463767933923799e-06,
      "loss": 0.93103886,
      "num_input_tokens_seen": 93905545,
      "step": 4349,
      "time_per_iteration": 2.585158109664917
    },
    {
      "auxiliary_loss_clip": 0.01138384,
      "auxiliary_loss_mlp": 0.0104717,
      "balance_loss_clip": 1.05462372,
      "balance_loss_mlp": 1.02981365,
      "epoch": 0.26153614910566664,
      "flos": 17457147181440.0,
      "grad_norm": 1.9299011807650872,
      "language_loss": 0.80198544,
      "learning_rate": 3.463502515580524e-06,
      "loss": 0.82384098,
      "num_input_tokens_seen": 93924185,
      "step": 4350,
      "time_per_iteration": 2.4858100414276123
    },
    {
      "auxiliary_loss_clip": 0.01128008,
      "auxiliary_loss_mlp": 0.01046501,
      "balance_loss_clip": 1.05072606,
      "balance_loss_mlp": 1.02878666,
      "epoch": 0.2615962723583346,
      "flos": 17712831168000.0,
      "grad_norm": 1.9843122876684063,
      "language_loss": 0.62930971,
      "learning_rate": 3.4632370417410866e-06,
      "loss": 0.65105486,
      "num_input_tokens_seen": 93942825,
      "step": 4351,
      "time_per_iteration": 2.487734317779541
    },
    {
      "auxiliary_loss_clip": 0.01134059,
      "auxiliary_loss_mlp": 0.01040894,
      "balance_loss_clip": 1.05019701,
      "balance_loss_mlp": 1.0232389,
      "epoch": 0.26165639561100257,
      "flos": 23257725672960.0,
      "grad_norm": 2.092744981865366,
      "language_loss": 0.84091425,
      "learning_rate": 3.462971512415555e-06,
      "loss": 0.86266381,
      "num_input_tokens_seen": 93962045,
      "step": 4352,
      "time_per_iteration": 2.4973349571228027
    },
    {
      "auxiliary_loss_clip": 0.01047516,
      "auxiliary_loss_mlp": 0.01003382,
      "balance_loss_clip": 1.02565098,
      "balance_loss_mlp": 1.0014981,
      "epoch": 0.26171651886367053,
      "flos": 66737970800640.0,
      "grad_norm": 0.8083682155822187,
      "language_loss": 0.705567,
      "learning_rate": 3.462705927613996e-06,
      "loss": 0.72607595,
      "num_input_tokens_seen": 94021175,
      "step": 4353,
      "time_per_iteration": 3.005462884902954
    },
    {
      "auxiliary_loss_clip": 0.01109624,
      "auxiliary_loss_mlp": 0.01054865,
      "balance_loss_clip": 1.04579782,
      "balance_loss_mlp": 1.03511214,
      "epoch": 0.2617766421163385,
      "flos": 22349581090560.0,
      "grad_norm": 1.6301351895871359,
      "language_loss": 0.77481627,
      "learning_rate": 3.4624402873464816e-06,
      "loss": 0.79646116,
      "num_input_tokens_seen": 94043370,
      "step": 4354,
      "time_per_iteration": 2.561483860015869
    },
    {
      "auxiliary_loss_clip": 0.01085974,
      "auxiliary_loss_mlp": 0.01048217,
      "balance_loss_clip": 1.0521884,
      "balance_loss_mlp": 1.02998936,
      "epoch": 0.26183676536900646,
      "flos": 26067125041920.0,
      "grad_norm": 1.8338169443674612,
      "language_loss": 0.68119442,
      "learning_rate": 3.462174591623085e-06,
      "loss": 0.70253634,
      "num_input_tokens_seen": 94063510,
      "step": 4355,
      "time_per_iteration": 2.611246347427368
    },
    {
      "auxiliary_loss_clip": 0.0109416,
      "auxiliary_loss_mlp": 0.01039403,
      "balance_loss_clip": 1.05252719,
      "balance_loss_mlp": 1.02012658,
      "epoch": 0.26189688862167443,
      "flos": 20996466825600.0,
      "grad_norm": 1.7108487844870557,
      "language_loss": 0.67281187,
      "learning_rate": 3.4619088404538815e-06,
      "loss": 0.69414747,
      "num_input_tokens_seen": 94083865,
      "step": 4356,
      "time_per_iteration": 2.6090705394744873
    },
    {
      "auxiliary_loss_clip": 0.01042452,
      "auxiliary_loss_mlp": 0.01003951,
      "balance_loss_clip": 1.02172256,
      "balance_loss_mlp": 1.00213933,
      "epoch": 0.2619570118743424,
      "flos": 65798261141760.0,
      "grad_norm": 0.6824195649184003,
      "language_loss": 0.53114581,
      "learning_rate": 3.4616430338489487e-06,
      "loss": 0.55160987,
      "num_input_tokens_seen": 94144095,
      "step": 4357,
      "time_per_iteration": 2.9986743927001953
    },
    {
      "auxiliary_loss_clip": 0.01135305,
      "auxiliary_loss_mlp": 0.01049319,
      "balance_loss_clip": 1.051319,
      "balance_loss_mlp": 1.03131807,
      "epoch": 0.26201713512701036,
      "flos": 28766817296640.0,
      "grad_norm": 1.9618307246744673,
      "language_loss": 0.83876789,
      "learning_rate": 3.4613771718183654e-06,
      "loss": 0.86061418,
      "num_input_tokens_seen": 94163035,
      "step": 4358,
      "time_per_iteration": 3.982941150665283
    },
    {
      "auxiliary_loss_clip": 0.01121418,
      "auxiliary_loss_mlp": 0.01046412,
      "balance_loss_clip": 1.04700947,
      "balance_loss_mlp": 1.02569306,
      "epoch": 0.2620772583796783,
      "flos": 26432516142720.0,
      "grad_norm": 2.79078183242566,
      "language_loss": 0.6763202,
      "learning_rate": 3.4611112543722127e-06,
      "loss": 0.69799852,
      "num_input_tokens_seen": 94182520,
      "step": 4359,
      "time_per_iteration": 2.5809645652770996
    },
    {
      "auxiliary_loss_clip": 0.01115076,
      "auxiliary_loss_mlp": 0.01044168,
      "balance_loss_clip": 1.04576993,
      "balance_loss_mlp": 1.02685857,
      "epoch": 0.2621373816323463,
      "flos": 20156552127360.0,
      "grad_norm": 1.9302512057123988,
      "language_loss": 0.78561997,
      "learning_rate": 3.4608452815205757e-06,
      "loss": 0.80721241,
      "num_input_tokens_seen": 94201795,
      "step": 4360,
      "time_per_iteration": 2.5514204502105713
    },
    {
      "auxiliary_loss_clip": 0.01116739,
      "auxiliary_loss_mlp": 0.01045565,
      "balance_loss_clip": 1.04817796,
      "balance_loss_mlp": 1.02848291,
      "epoch": 0.26219750488501425,
      "flos": 28621235473920.0,
      "grad_norm": 1.8295380028832282,
      "language_loss": 0.68313551,
      "learning_rate": 3.4605792532735387e-06,
      "loss": 0.70475858,
      "num_input_tokens_seen": 94222390,
      "step": 4361,
      "time_per_iteration": 3.999147891998291
    },
    {
      "auxiliary_loss_clip": 0.01135038,
      "auxiliary_loss_mlp": 0.01062518,
      "balance_loss_clip": 1.05126452,
      "balance_loss_mlp": 1.04437387,
      "epoch": 0.2622576281376823,
      "flos": 15042549173760.0,
      "grad_norm": 2.424189474497625,
      "language_loss": 0.84023184,
      "learning_rate": 3.46031316964119e-06,
      "loss": 0.86220741,
      "num_input_tokens_seen": 94239980,
      "step": 4362,
      "time_per_iteration": 2.479478120803833
    },
    {
      "auxiliary_loss_clip": 0.0110158,
      "auxiliary_loss_mlp": 0.01046658,
      "balance_loss_clip": 1.04843307,
      "balance_loss_mlp": 1.02820456,
      "epoch": 0.26231775139035024,
      "flos": 26396174557440.0,
      "grad_norm": 1.7016252825507183,
      "language_loss": 0.65391695,
      "learning_rate": 3.4600470306336197e-06,
      "loss": 0.67539942,
      "num_input_tokens_seen": 94260715,
      "step": 4363,
      "time_per_iteration": 2.6306166648864746
    },
    {
      "auxiliary_loss_clip": 0.01036756,
      "auxiliary_loss_mlp": 0.01005948,
      "balance_loss_clip": 1.03407109,
      "balance_loss_mlp": 1.00387383,
      "epoch": 0.2623778746430182,
      "flos": 65408918647680.0,
      "grad_norm": 0.8955280228033217,
      "language_loss": 0.61087322,
      "learning_rate": 3.4597808362609194e-06,
      "loss": 0.63130033,
      "num_input_tokens_seen": 94321285,
      "step": 4364,
      "time_per_iteration": 4.630819082260132
    },
    {
      "auxiliary_loss_clip": 0.01148628,
      "auxiliary_loss_mlp": 0.01053271,
      "balance_loss_clip": 1.05364084,
      "balance_loss_mlp": 1.03369665,
      "epoch": 0.26243799789568617,
      "flos": 12604215254400.0,
      "grad_norm": 2.4539086292477594,
      "language_loss": 0.71765798,
      "learning_rate": 3.459514586533184e-06,
      "loss": 0.73967701,
      "num_input_tokens_seen": 94335420,
      "step": 4365,
      "time_per_iteration": 3.8484385013580322
    },
    {
      "auxiliary_loss_clip": 0.0112043,
      "auxiliary_loss_mlp": 0.00812219,
      "balance_loss_clip": 1.052544,
      "balance_loss_mlp": 1.04350734,
      "epoch": 0.26249812114835414,
      "flos": 28623821253120.0,
      "grad_norm": 1.8384531948395009,
      "language_loss": 0.77336061,
      "learning_rate": 3.459248281460509e-06,
      "loss": 0.79268706,
      "num_input_tokens_seen": 94357440,
      "step": 4366,
      "time_per_iteration": 2.6529324054718018
    },
    {
      "auxiliary_loss_clip": 0.01147554,
      "auxiliary_loss_mlp": 0.0105079,
      "balance_loss_clip": 1.05417681,
      "balance_loss_mlp": 1.03383887,
      "epoch": 0.2625582444010221,
      "flos": 14465393441280.0,
      "grad_norm": 1.7739601521421118,
      "language_loss": 0.76445055,
      "learning_rate": 3.4589819210529927e-06,
      "loss": 0.78643394,
      "num_input_tokens_seen": 94375690,
      "step": 4367,
      "time_per_iteration": 2.5249221324920654
    },
    {
      "auxiliary_loss_clip": 0.0113331,
      "auxiliary_loss_mlp": 0.01051582,
      "balance_loss_clip": 1.05135322,
      "balance_loss_mlp": 1.03494,
      "epoch": 0.26261836765369007,
      "flos": 16613174246400.0,
      "grad_norm": 1.559255083678225,
      "language_loss": 0.6986447,
      "learning_rate": 3.458715505320736e-06,
      "loss": 0.72049367,
      "num_input_tokens_seen": 94393190,
      "step": 4368,
      "time_per_iteration": 2.5003228187561035
    },
    {
      "auxiliary_loss_clip": 0.01121761,
      "auxiliary_loss_mlp": 0.01048257,
      "balance_loss_clip": 1.05187249,
      "balance_loss_mlp": 1.02986312,
      "epoch": 0.26267849090635803,
      "flos": 20519932066560.0,
      "grad_norm": 2.0433897832995447,
      "language_loss": 0.78875363,
      "learning_rate": 3.458449034273841e-06,
      "loss": 0.81045383,
      "num_input_tokens_seen": 94410975,
      "step": 4369,
      "time_per_iteration": 2.539787530899048
    },
    {
      "auxiliary_loss_clip": 0.01118733,
      "auxiliary_loss_mlp": 0.01043924,
      "balance_loss_clip": 1.05358374,
      "balance_loss_mlp": 1.0267694,
      "epoch": 0.262738614159026,
      "flos": 21323936142720.0,
      "grad_norm": 1.9060299312410807,
      "language_loss": 0.83110225,
      "learning_rate": 3.4581825079224133e-06,
      "loss": 0.85272884,
      "num_input_tokens_seen": 94429985,
      "step": 4370,
      "time_per_iteration": 2.545135021209717
    },
    {
      "auxiliary_loss_clip": 0.01138646,
      "auxiliary_loss_mlp": 0.01055747,
      "balance_loss_clip": 1.05306959,
      "balance_loss_mlp": 1.03512406,
      "epoch": 0.26279873741169396,
      "flos": 17603590930560.0,
      "grad_norm": 1.6123663783188398,
      "language_loss": 0.7141428,
      "learning_rate": 3.4579159262765575e-06,
      "loss": 0.73608673,
      "num_input_tokens_seen": 94448660,
      "step": 4371,
      "time_per_iteration": 2.572695255279541
    },
    {
      "auxiliary_loss_clip": 0.01055633,
      "auxiliary_loss_mlp": 0.01014799,
      "balance_loss_clip": 1.02339172,
      "balance_loss_mlp": 1.01285636,
      "epoch": 0.2628588606643619,
      "flos": 60949746587520.0,
      "grad_norm": 0.7007925315149038,
      "language_loss": 0.56430948,
      "learning_rate": 3.457649289346384e-06,
      "loss": 0.58501387,
      "num_input_tokens_seen": 94515630,
      "step": 4372,
      "time_per_iteration": 3.2151167392730713
    },
    {
      "auxiliary_loss_clip": 0.01120117,
      "auxiliary_loss_mlp": 0.01036166,
      "balance_loss_clip": 1.05193114,
      "balance_loss_mlp": 1.01989365,
      "epoch": 0.2629189839170299,
      "flos": 27016315891200.0,
      "grad_norm": 2.109915434050956,
      "language_loss": 0.77628684,
      "learning_rate": 3.4573825971420042e-06,
      "loss": 0.79784966,
      "num_input_tokens_seen": 94535385,
      "step": 4373,
      "time_per_iteration": 2.6567587852478027
    },
    {
      "auxiliary_loss_clip": 0.01101984,
      "auxiliary_loss_mlp": 0.01043854,
      "balance_loss_clip": 1.05317998,
      "balance_loss_mlp": 1.02723026,
      "epoch": 0.26297910716969786,
      "flos": 17019863009280.0,
      "grad_norm": 2.463689670877103,
      "language_loss": 0.70781839,
      "learning_rate": 3.4571158496735294e-06,
      "loss": 0.72927678,
      "num_input_tokens_seen": 94552650,
      "step": 4374,
      "time_per_iteration": 2.5929269790649414
    },
    {
      "auxiliary_loss_clip": 0.01117506,
      "auxiliary_loss_mlp": 0.0104455,
      "balance_loss_clip": 1.0537827,
      "balance_loss_mlp": 1.02587008,
      "epoch": 0.2630392304223659,
      "flos": 24897370728960.0,
      "grad_norm": 1.687008300326169,
      "language_loss": 0.8089484,
      "learning_rate": 3.4568490469510756e-06,
      "loss": 0.83056897,
      "num_input_tokens_seen": 94574075,
      "step": 4375,
      "time_per_iteration": 2.590388536453247
    },
    {
      "auxiliary_loss_clip": 0.01118027,
      "auxiliary_loss_mlp": 0.01046868,
      "balance_loss_clip": 1.04994369,
      "balance_loss_mlp": 1.03025043,
      "epoch": 0.26309935367503384,
      "flos": 32854026067200.0,
      "grad_norm": 2.1003347436398676,
      "language_loss": 0.660061,
      "learning_rate": 3.4565821889847603e-06,
      "loss": 0.68170989,
      "num_input_tokens_seen": 94594255,
      "step": 4376,
      "time_per_iteration": 2.618356943130493
    },
    {
      "auxiliary_loss_clip": 0.01102788,
      "auxiliary_loss_mlp": 0.01055949,
      "balance_loss_clip": 1.05451763,
      "balance_loss_mlp": 1.03682756,
      "epoch": 0.2631594769277018,
      "flos": 15887958652800.0,
      "grad_norm": 2.1136069737120486,
      "language_loss": 0.69312197,
      "learning_rate": 3.4563152757847026e-06,
      "loss": 0.71470928,
      "num_input_tokens_seen": 94611410,
      "step": 4377,
      "time_per_iteration": 2.60772705078125
    },
    {
      "auxiliary_loss_clip": 0.0113318,
      "auxiliary_loss_mlp": 0.01043379,
      "balance_loss_clip": 1.05162215,
      "balance_loss_mlp": 1.02617693,
      "epoch": 0.2632196001803698,
      "flos": 50804943557760.0,
      "grad_norm": 1.7490173290015092,
      "language_loss": 0.79582822,
      "learning_rate": 3.4560483073610233e-06,
      "loss": 0.81759381,
      "num_input_tokens_seen": 94636575,
      "step": 4378,
      "time_per_iteration": 2.760911464691162
    },
    {
      "auxiliary_loss_clip": 0.01119675,
      "auxiliary_loss_mlp": 0.0104864,
      "balance_loss_clip": 1.05175352,
      "balance_loss_mlp": 1.03301156,
      "epoch": 0.26327972343303774,
      "flos": 13733031041280.0,
      "grad_norm": 1.982443517710993,
      "language_loss": 0.7683928,
      "learning_rate": 3.455781283723846e-06,
      "loss": 0.7900759,
      "num_input_tokens_seen": 94654345,
      "step": 4379,
      "time_per_iteration": 2.543760061264038
    },
    {
      "auxiliary_loss_clip": 0.01111095,
      "auxiliary_loss_mlp": 0.01043194,
      "balance_loss_clip": 1.05439043,
      "balance_loss_mlp": 1.0242877,
      "epoch": 0.2633398466857057,
      "flos": 23769057732480.0,
      "grad_norm": 3.1073610659824027,
      "language_loss": 0.77994424,
      "learning_rate": 3.4555142048832975e-06,
      "loss": 0.80148715,
      "num_input_tokens_seen": 94673985,
      "step": 4380,
      "time_per_iteration": 2.6417787075042725
    },
    {
      "auxiliary_loss_clip": 0.01117362,
      "auxiliary_loss_mlp": 0.01041985,
      "balance_loss_clip": 1.04558921,
      "balance_loss_mlp": 1.02461624,
      "epoch": 0.26339996993837367,
      "flos": 27600223380480.0,
      "grad_norm": 1.924105774372334,
      "language_loss": 0.64101404,
      "learning_rate": 3.4552470708495036e-06,
      "loss": 0.66260749,
      "num_input_tokens_seen": 94693145,
      "step": 4381,
      "time_per_iteration": 2.577605724334717
    },
    {
      "auxiliary_loss_clip": 0.01130085,
      "auxiliary_loss_mlp": 0.01040754,
      "balance_loss_clip": 1.04822731,
      "balance_loss_mlp": 1.02404118,
      "epoch": 0.26346009319104163,
      "flos": 16946317912320.0,
      "grad_norm": 1.8722766535643167,
      "language_loss": 0.82916325,
      "learning_rate": 3.454979881632595e-06,
      "loss": 0.85087168,
      "num_input_tokens_seen": 94710185,
      "step": 4382,
      "time_per_iteration": 2.4710776805877686
    },
    {
      "auxiliary_loss_clip": 0.01105921,
      "auxiliary_loss_mlp": 0.01055454,
      "balance_loss_clip": 1.05695319,
      "balance_loss_mlp": 1.03619003,
      "epoch": 0.2635202164437096,
      "flos": 37232218915200.0,
      "grad_norm": 1.8627329063786284,
      "language_loss": 0.69699407,
      "learning_rate": 3.4547126372427035e-06,
      "loss": 0.71860778,
      "num_input_tokens_seen": 94730280,
      "step": 4383,
      "time_per_iteration": 2.6879851818084717
    },
    {
      "auxiliary_loss_clip": 0.011316,
      "auxiliary_loss_mlp": 0.01047104,
      "balance_loss_clip": 1.049927,
      "balance_loss_mlp": 1.03077209,
      "epoch": 0.26358033969637756,
      "flos": 20996359084800.0,
      "grad_norm": 2.1131313905581206,
      "language_loss": 0.69373494,
      "learning_rate": 3.4544453376899638e-06,
      "loss": 0.71552199,
      "num_input_tokens_seen": 94748560,
      "step": 4384,
      "time_per_iteration": 2.501309394836426
    },
    {
      "auxiliary_loss_clip": 0.0112889,
      "auxiliary_loss_mlp": 0.01043778,
      "balance_loss_clip": 1.04798007,
      "balance_loss_mlp": 1.02658832,
      "epoch": 0.26364046294904553,
      "flos": 27746092512000.0,
      "grad_norm": 7.2107371007278775,
      "language_loss": 0.70297003,
      "learning_rate": 3.45417798298451e-06,
      "loss": 0.7246967,
      "num_input_tokens_seen": 94767570,
      "step": 4385,
      "time_per_iteration": 2.626553535461426
    },
    {
      "auxiliary_loss_clip": 0.01110324,
      "auxiliary_loss_mlp": 0.01051196,
      "balance_loss_clip": 1.04985929,
      "balance_loss_mlp": 1.03326643,
      "epoch": 0.2637005862017135,
      "flos": 22893088757760.0,
      "grad_norm": 1.8376093900720654,
      "language_loss": 0.85618889,
      "learning_rate": 3.453910573136482e-06,
      "loss": 0.87780404,
      "num_input_tokens_seen": 94784985,
      "step": 4386,
      "time_per_iteration": 2.5771591663360596
    },
    {
      "auxiliary_loss_clip": 0.01124331,
      "auxiliary_loss_mlp": 0.0104394,
      "balance_loss_clip": 1.05228853,
      "balance_loss_mlp": 1.02696466,
      "epoch": 0.26376070945438146,
      "flos": 15048834053760.0,
      "grad_norm": 3.0277282317871252,
      "language_loss": 0.7747761,
      "learning_rate": 3.4536431081560196e-06,
      "loss": 0.79645884,
      "num_input_tokens_seen": 94802545,
      "step": 4387,
      "time_per_iteration": 2.5180513858795166
    },
    {
      "auxiliary_loss_clip": 0.01129837,
      "auxiliary_loss_mlp": 0.01050696,
      "balance_loss_clip": 1.05579758,
      "balance_loss_mlp": 1.03404272,
      "epoch": 0.2638208327070494,
      "flos": 21141833166720.0,
      "grad_norm": 1.937737761135073,
      "language_loss": 0.76291931,
      "learning_rate": 3.453375588053264e-06,
      "loss": 0.78472459,
      "num_input_tokens_seen": 94820730,
      "step": 4388,
      "time_per_iteration": 2.5022358894348145
    },
    {
      "auxiliary_loss_clip": 0.01141735,
      "auxiliary_loss_mlp": 0.01040995,
      "balance_loss_clip": 1.05024266,
      "balance_loss_mlp": 1.02437782,
      "epoch": 0.26388095595971744,
      "flos": 21725597001600.0,
      "grad_norm": 2.881053240785093,
      "language_loss": 0.86369467,
      "learning_rate": 3.4531080128383617e-06,
      "loss": 0.88552201,
      "num_input_tokens_seen": 94839175,
      "step": 4389,
      "time_per_iteration": 2.46939754486084
    },
    {
      "auxiliary_loss_clip": 0.01044334,
      "auxiliary_loss_mlp": 0.01010081,
      "balance_loss_clip": 1.02608502,
      "balance_loss_mlp": 1.00769639,
      "epoch": 0.2639410792123854,
      "flos": 65515537192320.0,
      "grad_norm": 0.8098718678112861,
      "language_loss": 0.60292989,
      "learning_rate": 3.452840382521457e-06,
      "loss": 0.623474,
      "num_input_tokens_seen": 94898865,
      "step": 4390,
      "time_per_iteration": 3.1140854358673096
    },
    {
      "auxiliary_loss_clip": 0.0112849,
      "auxiliary_loss_mlp": 0.01042392,
      "balance_loss_clip": 1.0505836,
      "balance_loss_mlp": 1.02472544,
      "epoch": 0.2640012024650534,
      "flos": 23948574929280.0,
      "grad_norm": 2.0234689527115437,
      "language_loss": 0.77745348,
      "learning_rate": 3.4525726971127e-06,
      "loss": 0.79916233,
      "num_input_tokens_seen": 94917490,
      "step": 4391,
      "time_per_iteration": 2.5452842712402344
    },
    {
      "auxiliary_loss_clip": 0.01024267,
      "auxiliary_loss_mlp": 0.01034943,
      "balance_loss_clip": 1.02411127,
      "balance_loss_mlp": 1.44962382,
      "epoch": 0.26406132571772134,
      "flos": 56441163369600.0,
      "grad_norm": 0.8980062902404639,
      "language_loss": 0.58728611,
      "learning_rate": 3.45230495662224e-06,
      "loss": 0.60787821,
      "num_input_tokens_seen": 94969065,
      "step": 4392,
      "time_per_iteration": 3.136406183242798
    },
    {
      "auxiliary_loss_clip": 0.01134592,
      "auxiliary_loss_mlp": 0.0105264,
      "balance_loss_clip": 1.05362105,
      "balance_loss_mlp": 1.03535509,
      "epoch": 0.2641214489703893,
      "flos": 22090557139200.0,
      "grad_norm": 1.9192418706603087,
      "language_loss": 0.68714941,
      "learning_rate": 3.4520371610602306e-06,
      "loss": 0.70902181,
      "num_input_tokens_seen": 94988540,
      "step": 4393,
      "time_per_iteration": 2.5015244483947754
    },
    {
      "auxiliary_loss_clip": 0.01136645,
      "auxiliary_loss_mlp": 0.01043828,
      "balance_loss_clip": 1.05249703,
      "balance_loss_mlp": 1.02570808,
      "epoch": 0.26418157222305727,
      "flos": 16544764794240.0,
      "grad_norm": 1.7876586007322226,
      "language_loss": 0.8408162,
      "learning_rate": 3.4517693104368267e-06,
      "loss": 0.86262095,
      "num_input_tokens_seen": 95004810,
      "step": 4394,
      "time_per_iteration": 2.482170343399048
    },
    {
      "auxiliary_loss_clip": 0.01121145,
      "auxiliary_loss_mlp": 0.01047799,
      "balance_loss_clip": 1.05023146,
      "balance_loss_mlp": 1.02851093,
      "epoch": 0.26424169547572524,
      "flos": 18002486442240.0,
      "grad_norm": 2.27239914833793,
      "language_loss": 0.69740576,
      "learning_rate": 3.4515014047621856e-06,
      "loss": 0.71909517,
      "num_input_tokens_seen": 95024085,
      "step": 4395,
      "time_per_iteration": 2.5033183097839355
    },
    {
      "auxiliary_loss_clip": 0.01110827,
      "auxiliary_loss_mlp": 0.01038454,
      "balance_loss_clip": 1.05107808,
      "balance_loss_mlp": 1.02119207,
      "epoch": 0.2643018187283932,
      "flos": 16983162288000.0,
      "grad_norm": 1.8901670889805686,
      "language_loss": 0.86793739,
      "learning_rate": 3.4512334440464655e-06,
      "loss": 0.88943022,
      "num_input_tokens_seen": 95042515,
      "step": 4396,
      "time_per_iteration": 2.5593998432159424
    },
    {
      "auxiliary_loss_clip": 0.01006689,
      "auxiliary_loss_mlp": 0.0102546,
      "balance_loss_clip": 1.02342391,
      "balance_loss_mlp": 1.02267098,
      "epoch": 0.26436194198106117,
      "flos": 59664359416320.0,
      "grad_norm": 0.7910875031223654,
      "language_loss": 0.55005282,
      "learning_rate": 3.4509654282998277e-06,
      "loss": 0.57037431,
      "num_input_tokens_seen": 95094835,
      "step": 4397,
      "time_per_iteration": 4.526848316192627
    },
    {
      "auxiliary_loss_clip": 0.0113141,
      "auxiliary_loss_mlp": 0.01051345,
      "balance_loss_clip": 1.050735,
      "balance_loss_mlp": 1.03476334,
      "epoch": 0.26442206523372913,
      "flos": 32921322197760.0,
      "grad_norm": 2.0398499259582565,
      "language_loss": 0.77977586,
      "learning_rate": 3.450697357532435e-06,
      "loss": 0.80160338,
      "num_input_tokens_seen": 95113480,
      "step": 4398,
      "time_per_iteration": 2.7783992290496826
    },
    {
      "auxiliary_loss_clip": 0.01139459,
      "auxiliary_loss_mlp": 0.01046485,
      "balance_loss_clip": 1.05867553,
      "balance_loss_mlp": 1.02874649,
      "epoch": 0.2644821884863971,
      "flos": 21031300039680.0,
      "grad_norm": 1.6584463220502164,
      "language_loss": 0.67016208,
      "learning_rate": 3.4504292317544534e-06,
      "loss": 0.69202143,
      "num_input_tokens_seen": 95132580,
      "step": 4399,
      "time_per_iteration": 3.8913605213165283
    },
    {
      "auxiliary_loss_clip": 0.01096951,
      "auxiliary_loss_mlp": 0.01049858,
      "balance_loss_clip": 1.05236864,
      "balance_loss_mlp": 1.03185773,
      "epoch": 0.26454231173906506,
      "flos": 20776801201920.0,
      "grad_norm": 1.6893011635552766,
      "language_loss": 0.86926705,
      "learning_rate": 3.4501610509760504e-06,
      "loss": 0.89073515,
      "num_input_tokens_seen": 95152375,
      "step": 4400,
      "time_per_iteration": 2.605315923690796
    },
    {
      "auxiliary_loss_clip": 0.01117111,
      "auxiliary_loss_mlp": 0.01041909,
      "balance_loss_clip": 1.04945254,
      "balance_loss_mlp": 1.02403998,
      "epoch": 0.264602434991733,
      "flos": 16618669027200.0,
      "grad_norm": 2.0091643639118835,
      "language_loss": 0.76037884,
      "learning_rate": 3.4498928152073944e-06,
      "loss": 0.78196907,
      "num_input_tokens_seen": 95170265,
      "step": 4401,
      "time_per_iteration": 2.5109434127807617
    },
    {
      "auxiliary_loss_clip": 0.01103023,
      "auxiliary_loss_mlp": 0.01052503,
      "balance_loss_clip": 1.05305493,
      "balance_loss_mlp": 1.03376341,
      "epoch": 0.26466255824440105,
      "flos": 19062677295360.0,
      "grad_norm": 1.6961170679443283,
      "language_loss": 0.87897581,
      "learning_rate": 3.4496245244586577e-06,
      "loss": 0.90053105,
      "num_input_tokens_seen": 95188655,
      "step": 4402,
      "time_per_iteration": 3.967491626739502
    },
    {
      "auxiliary_loss_clip": 0.01111241,
      "auxiliary_loss_mlp": 0.01047223,
      "balance_loss_clip": 1.05405986,
      "balance_loss_mlp": 1.02938914,
      "epoch": 0.264722681497069,
      "flos": 22638554006400.0,
      "grad_norm": 1.5282527709412101,
      "language_loss": 0.78206861,
      "learning_rate": 3.4493561787400137e-06,
      "loss": 0.80365324,
      "num_input_tokens_seen": 95209615,
      "step": 4403,
      "time_per_iteration": 2.6071062088012695
    },
    {
      "auxiliary_loss_clip": 0.01130712,
      "auxiliary_loss_mlp": 0.01040815,
      "balance_loss_clip": 1.05544055,
      "balance_loss_mlp": 1.02262378,
      "epoch": 0.264782804749737,
      "flos": 22492253911680.0,
      "grad_norm": 2.0038718210069284,
      "language_loss": 0.8797586,
      "learning_rate": 3.4490877780616387e-06,
      "loss": 0.90147388,
      "num_input_tokens_seen": 95227810,
      "step": 4404,
      "time_per_iteration": 3.8886563777923584
    },
    {
      "auxiliary_loss_clip": 0.01122966,
      "auxiliary_loss_mlp": 0.01041368,
      "balance_loss_clip": 1.04914522,
      "balance_loss_mlp": 1.02419043,
      "epoch": 0.26484292800240494,
      "flos": 16800269212800.0,
      "grad_norm": 1.7938719653886217,
      "language_loss": 0.76728553,
      "learning_rate": 3.448819322433709e-06,
      "loss": 0.78892881,
      "num_input_tokens_seen": 95245890,
      "step": 4405,
      "time_per_iteration": 2.520015239715576
    },
    {
      "auxiliary_loss_clip": 0.01146226,
      "auxiliary_loss_mlp": 0.01040581,
      "balance_loss_clip": 1.0532428,
      "balance_loss_mlp": 1.02218664,
      "epoch": 0.2649030512550729,
      "flos": 20449583280000.0,
      "grad_norm": 1.9651212866756151,
      "language_loss": 0.70241761,
      "learning_rate": 3.4485508118664066e-06,
      "loss": 0.72428572,
      "num_input_tokens_seen": 95264955,
      "step": 4406,
      "time_per_iteration": 2.469975233078003
    },
    {
      "auxiliary_loss_clip": 0.01120805,
      "auxiliary_loss_mlp": 0.01049514,
      "balance_loss_clip": 1.05607772,
      "balance_loss_mlp": 1.03309894,
      "epoch": 0.2649631745077409,
      "flos": 22416123035520.0,
      "grad_norm": 1.7170309932055356,
      "language_loss": 0.83698618,
      "learning_rate": 3.448282246369912e-06,
      "loss": 0.85868937,
      "num_input_tokens_seen": 95284245,
      "step": 4407,
      "time_per_iteration": 2.578608989715576
    },
    {
      "auxiliary_loss_clip": 0.01107093,
      "auxiliary_loss_mlp": 0.0103423,
      "balance_loss_clip": 1.04968333,
      "balance_loss_mlp": 1.01645565,
      "epoch": 0.26502329776040884,
      "flos": 35116110927360.0,
      "grad_norm": 1.8913692257173007,
      "language_loss": 0.76059258,
      "learning_rate": 3.4480136259544084e-06,
      "loss": 0.78200579,
      "num_input_tokens_seen": 95307125,
      "step": 4408,
      "time_per_iteration": 2.683703660964966
    },
    {
      "auxiliary_loss_clip": 0.01095089,
      "auxiliary_loss_mlp": 0.0104255,
      "balance_loss_clip": 1.05363941,
      "balance_loss_mlp": 1.02447772,
      "epoch": 0.2650834210130768,
      "flos": 38687498438400.0,
      "grad_norm": 1.7176394478630326,
      "language_loss": 0.70982969,
      "learning_rate": 3.447744950630084e-06,
      "loss": 0.73120612,
      "num_input_tokens_seen": 95329150,
      "step": 4409,
      "time_per_iteration": 2.725533962249756
    },
    {
      "auxiliary_loss_clip": 0.01131844,
      "auxiliary_loss_mlp": 0.01041131,
      "balance_loss_clip": 1.05357516,
      "balance_loss_mlp": 1.02268934,
      "epoch": 0.26514354426574477,
      "flos": 24716847951360.0,
      "grad_norm": 1.6030140374413975,
      "language_loss": 0.72768289,
      "learning_rate": 3.4474762204071253e-06,
      "loss": 0.7494126,
      "num_input_tokens_seen": 95349880,
      "step": 4410,
      "time_per_iteration": 2.538592576980591
    },
    {
      "auxiliary_loss_clip": 0.01138845,
      "auxiliary_loss_mlp": 0.01052984,
      "balance_loss_clip": 1.05246949,
      "balance_loss_mlp": 1.0360086,
      "epoch": 0.26520366751841273,
      "flos": 20340055733760.0,
      "grad_norm": 1.7430946471764588,
      "language_loss": 0.7333802,
      "learning_rate": 3.4472074352957244e-06,
      "loss": 0.7552985,
      "num_input_tokens_seen": 95368570,
      "step": 4411,
      "time_per_iteration": 2.529890537261963
    },
    {
      "auxiliary_loss_clip": 0.01107955,
      "auxiliary_loss_mlp": 0.01043679,
      "balance_loss_clip": 1.05844247,
      "balance_loss_mlp": 1.02677476,
      "epoch": 0.2652637907710807,
      "flos": 22343870828160.0,
      "grad_norm": 2.853722657739487,
      "language_loss": 0.82264185,
      "learning_rate": 3.446938595306071e-06,
      "loss": 0.84415817,
      "num_input_tokens_seen": 95387065,
      "step": 4412,
      "time_per_iteration": 2.607990026473999
    },
    {
      "auxiliary_loss_clip": 0.01133394,
      "auxiliary_loss_mlp": 0.01050864,
      "balance_loss_clip": 1.05392146,
      "balance_loss_mlp": 1.03444862,
      "epoch": 0.26532391402374866,
      "flos": 19354235990400.0,
      "grad_norm": 2.090971702600164,
      "language_loss": 0.74601042,
      "learning_rate": 3.4466697004483622e-06,
      "loss": 0.76785302,
      "num_input_tokens_seen": 95406345,
      "step": 4413,
      "time_per_iteration": 2.5071966648101807
    },
    {
      "auxiliary_loss_clip": 0.01044692,
      "auxiliary_loss_mlp": 0.01003874,
      "balance_loss_clip": 1.02066255,
      "balance_loss_mlp": 1.00160933,
      "epoch": 0.26538403727641663,
      "flos": 44787611422080.0,
      "grad_norm": 0.8721646797899274,
      "language_loss": 0.56900698,
      "learning_rate": 3.446400750732793e-06,
      "loss": 0.58949262,
      "num_input_tokens_seen": 95463595,
      "step": 4414,
      "time_per_iteration": 3.023125171661377
    },
    {
      "auxiliary_loss_clip": 0.01111193,
      "auxiliary_loss_mlp": 0.0104508,
      "balance_loss_clip": 1.05856359,
      "balance_loss_mlp": 1.02932024,
      "epoch": 0.26544416052908465,
      "flos": 28182119708160.0,
      "grad_norm": 1.730570216326672,
      "language_loss": 0.74342585,
      "learning_rate": 3.4461317461695625e-06,
      "loss": 0.7649886,
      "num_input_tokens_seen": 95484115,
      "step": 4415,
      "time_per_iteration": 2.6364481449127197
    },
    {
      "auxiliary_loss_clip": 0.01093386,
      "auxiliary_loss_mlp": 0.01044109,
      "balance_loss_clip": 1.04633212,
      "balance_loss_mlp": 1.02414107,
      "epoch": 0.2655042837817526,
      "flos": 17565274097280.0,
      "grad_norm": 2.1555280244232438,
      "language_loss": 0.86553347,
      "learning_rate": 3.4458626867688707e-06,
      "loss": 0.88690841,
      "num_input_tokens_seen": 95501435,
      "step": 4416,
      "time_per_iteration": 2.568256378173828
    },
    {
      "auxiliary_loss_clip": 0.01140352,
      "auxiliary_loss_mlp": 0.01044602,
      "balance_loss_clip": 1.05521917,
      "balance_loss_mlp": 1.02589798,
      "epoch": 0.2655644070344206,
      "flos": 23404636298880.0,
      "grad_norm": 1.647706326575447,
      "language_loss": 0.76408511,
      "learning_rate": 3.4455935725409217e-06,
      "loss": 0.78593463,
      "num_input_tokens_seen": 95520135,
      "step": 4417,
      "time_per_iteration": 2.536073684692383
    },
    {
      "auxiliary_loss_clip": 0.01116712,
      "auxiliary_loss_mlp": 0.01041302,
      "balance_loss_clip": 1.0505116,
      "balance_loss_mlp": 1.02239561,
      "epoch": 0.26562453028708854,
      "flos": 26468462678400.0,
      "grad_norm": 1.4614490570466478,
      "language_loss": 0.79934263,
      "learning_rate": 3.4453244034959196e-06,
      "loss": 0.82092279,
      "num_input_tokens_seen": 95541705,
      "step": 4418,
      "time_per_iteration": 2.585609197616577
    },
    {
      "auxiliary_loss_clip": 0.01134012,
      "auxiliary_loss_mlp": 0.01051816,
      "balance_loss_clip": 1.05209827,
      "balance_loss_mlp": 1.03411365,
      "epoch": 0.2656846535397565,
      "flos": 19207576759680.0,
      "grad_norm": 2.047124782234769,
      "language_loss": 0.66981399,
      "learning_rate": 3.445055179644071e-06,
      "loss": 0.69167227,
      "num_input_tokens_seen": 95560300,
      "step": 4419,
      "time_per_iteration": 2.4991655349731445
    },
    {
      "auxiliary_loss_clip": 0.01144906,
      "auxiliary_loss_mlp": 0.01051613,
      "balance_loss_clip": 1.05178988,
      "balance_loss_mlp": 1.03281355,
      "epoch": 0.2657447767924245,
      "flos": 30551325903360.0,
      "grad_norm": 2.1283391962267633,
      "language_loss": 0.79438108,
      "learning_rate": 3.444785900995585e-06,
      "loss": 0.81634629,
      "num_input_tokens_seen": 95580150,
      "step": 4420,
      "time_per_iteration": 2.539354085922241
    },
    {
      "auxiliary_loss_clip": 0.01122686,
      "auxiliary_loss_mlp": 0.01054445,
      "balance_loss_clip": 1.05308604,
      "balance_loss_mlp": 1.03416729,
      "epoch": 0.26580490004509244,
      "flos": 20922742160640.0,
      "grad_norm": 2.032274391641846,
      "language_loss": 0.81626356,
      "learning_rate": 3.444516567560673e-06,
      "loss": 0.83803487,
      "num_input_tokens_seen": 95597570,
      "step": 4421,
      "time_per_iteration": 2.5403130054473877
    },
    {
      "auxiliary_loss_clip": 0.01129056,
      "auxiliary_loss_mlp": 0.01041749,
      "balance_loss_clip": 1.05126178,
      "balance_loss_mlp": 1.02451086,
      "epoch": 0.2658650232977604,
      "flos": 43945682584320.0,
      "grad_norm": 1.5914586562586155,
      "language_loss": 0.65694618,
      "learning_rate": 3.444247179349548e-06,
      "loss": 0.67865425,
      "num_input_tokens_seen": 95619415,
      "step": 4422,
      "time_per_iteration": 2.6843690872192383
    },
    {
      "auxiliary_loss_clip": 0.01128207,
      "auxiliary_loss_mlp": 0.01052475,
      "balance_loss_clip": 1.05145574,
      "balance_loss_mlp": 1.03567886,
      "epoch": 0.26592514655042837,
      "flos": 29716439109120.0,
      "grad_norm": 2.0949678428956133,
      "language_loss": 0.74429774,
      "learning_rate": 3.4439777363724252e-06,
      "loss": 0.76610452,
      "num_input_tokens_seen": 95639155,
      "step": 4423,
      "time_per_iteration": 2.580152988433838
    },
    {
      "auxiliary_loss_clip": 0.01126843,
      "auxiliary_loss_mlp": 0.01064823,
      "balance_loss_clip": 1.04784322,
      "balance_loss_mlp": 1.04547548,
      "epoch": 0.26598526980309634,
      "flos": 46677730014720.0,
      "grad_norm": 1.7126677816458091,
      "language_loss": 0.77900314,
      "learning_rate": 3.443708238639522e-06,
      "loss": 0.80091971,
      "num_input_tokens_seen": 95663320,
      "step": 4424,
      "time_per_iteration": 2.7113468647003174
    },
    {
      "auxiliary_loss_clip": 0.01130611,
      "auxiliary_loss_mlp": 0.01053883,
      "balance_loss_clip": 1.05677748,
      "balance_loss_mlp": 1.0362637,
      "epoch": 0.2660453930557643,
      "flos": 11509442582400.0,
      "grad_norm": 2.0503481042831684,
      "language_loss": 0.79425108,
      "learning_rate": 3.4434386861610573e-06,
      "loss": 0.81609607,
      "num_input_tokens_seen": 95680260,
      "step": 4425,
      "time_per_iteration": 2.464238166809082
    },
    {
      "auxiliary_loss_clip": 0.01118916,
      "auxiliary_loss_mlp": 0.01050925,
      "balance_loss_clip": 1.05207205,
      "balance_loss_mlp": 1.03504646,
      "epoch": 0.26610551630843227,
      "flos": 24791578197120.0,
      "grad_norm": 1.503152849994541,
      "language_loss": 0.80416524,
      "learning_rate": 3.4431690789472532e-06,
      "loss": 0.8258636,
      "num_input_tokens_seen": 95701140,
      "step": 4426,
      "time_per_iteration": 2.579782724380493
    },
    {
      "auxiliary_loss_clip": 0.01149112,
      "auxiliary_loss_mlp": 0.01054087,
      "balance_loss_clip": 1.05695677,
      "balance_loss_mlp": 1.03723061,
      "epoch": 0.26616563956110023,
      "flos": 27636385397760.0,
      "grad_norm": 1.537056116638259,
      "language_loss": 0.77172869,
      "learning_rate": 3.442899417008333e-06,
      "loss": 0.79376066,
      "num_input_tokens_seen": 95722060,
      "step": 4427,
      "time_per_iteration": 2.520817279815674
    },
    {
      "auxiliary_loss_clip": 0.01107953,
      "auxiliary_loss_mlp": 0.01037954,
      "balance_loss_clip": 1.05090678,
      "balance_loss_mlp": 1.02155089,
      "epoch": 0.26622576281376825,
      "flos": 28362893880960.0,
      "grad_norm": 1.5792828291709775,
      "language_loss": 0.76619202,
      "learning_rate": 3.4426297003545227e-06,
      "loss": 0.78765112,
      "num_input_tokens_seen": 95742495,
      "step": 4428,
      "time_per_iteration": 2.627612352371216
    },
    {
      "auxiliary_loss_clip": 0.01107725,
      "auxiliary_loss_mlp": 0.01002234,
      "balance_loss_clip": 1.050843,
      "balance_loss_mlp": 1.40964615,
      "epoch": 0.2662858860664362,
      "flos": 18041341979520.0,
      "grad_norm": 2.236605847511358,
      "language_loss": 0.82639253,
      "learning_rate": 3.4423599289960495e-06,
      "loss": 0.8474921,
      "num_input_tokens_seen": 95761510,
      "step": 4429,
      "time_per_iteration": 2.525341510772705
    },
    {
      "auxiliary_loss_clip": 0.01099685,
      "auxiliary_loss_mlp": 0.01046622,
      "balance_loss_clip": 1.04895079,
      "balance_loss_mlp": 1.02978969,
      "epoch": 0.2663460093191042,
      "flos": 22745818995840.0,
      "grad_norm": 1.544496983474864,
      "language_loss": 0.72031105,
      "learning_rate": 3.442090102943143e-06,
      "loss": 0.7417742,
      "num_input_tokens_seen": 95782385,
      "step": 4430,
      "time_per_iteration": 2.5926504135131836
    },
    {
      "auxiliary_loss_clip": 0.01144431,
      "auxiliary_loss_mlp": 0.0104679,
      "balance_loss_clip": 1.05136895,
      "balance_loss_mlp": 1.0277524,
      "epoch": 0.26640613257177215,
      "flos": 16508782344960.0,
      "grad_norm": 1.9216805620612056,
      "language_loss": 0.82211339,
      "learning_rate": 3.441820222206035e-06,
      "loss": 0.84402561,
      "num_input_tokens_seen": 95800595,
      "step": 4431,
      "time_per_iteration": 2.4560351371765137
    },
    {
      "auxiliary_loss_clip": 0.0113755,
      "auxiliary_loss_mlp": 0.01054675,
      "balance_loss_clip": 1.05209875,
      "balance_loss_mlp": 1.03663838,
      "epoch": 0.2664662558244401,
      "flos": 23075945919360.0,
      "grad_norm": 2.145849961411832,
      "language_loss": 0.76728427,
      "learning_rate": 3.44155028679496e-06,
      "loss": 0.78920656,
      "num_input_tokens_seen": 95818480,
      "step": 4432,
      "time_per_iteration": 2.5234768390655518
    },
    {
      "auxiliary_loss_clip": 0.01086313,
      "auxiliary_loss_mlp": 0.01050933,
      "balance_loss_clip": 1.05426133,
      "balance_loss_mlp": 1.03097689,
      "epoch": 0.2665263790771081,
      "flos": 23769273214080.0,
      "grad_norm": 1.7952334988097736,
      "language_loss": 0.82375765,
      "learning_rate": 3.441280296720154e-06,
      "loss": 0.84513009,
      "num_input_tokens_seen": 95837205,
      "step": 4433,
      "time_per_iteration": 2.6081361770629883
    },
    {
      "auxiliary_loss_clip": 0.01133106,
      "auxiliary_loss_mlp": 0.01045396,
      "balance_loss_clip": 1.05154896,
      "balance_loss_mlp": 1.02727675,
      "epoch": 0.26658650232977604,
      "flos": 28001273708160.0,
      "grad_norm": 1.91425509807709,
      "language_loss": 0.76696533,
      "learning_rate": 3.441010251991854e-06,
      "loss": 0.78875041,
      "num_input_tokens_seen": 95858395,
      "step": 4434,
      "time_per_iteration": 2.553297519683838
    },
    {
      "auxiliary_loss_clip": 0.01138661,
      "auxiliary_loss_mlp": 0.01041866,
      "balance_loss_clip": 1.04942489,
      "balance_loss_mlp": 1.02570164,
      "epoch": 0.266646625582444,
      "flos": 22163635359360.0,
      "grad_norm": 1.6939232937855473,
      "language_loss": 0.82330251,
      "learning_rate": 3.440740152620301e-06,
      "loss": 0.84510779,
      "num_input_tokens_seen": 95877875,
      "step": 4435,
      "time_per_iteration": 3.9191555976867676
    },
    {
      "auxiliary_loss_clip": 0.01101275,
      "auxiliary_loss_mlp": 0.01054708,
      "balance_loss_clip": 1.04687381,
      "balance_loss_mlp": 1.03688645,
      "epoch": 0.266706748835112,
      "flos": 27853537069440.0,
      "grad_norm": 2.2715279163038695,
      "language_loss": 0.87753206,
      "learning_rate": 3.4404699986157376e-06,
      "loss": 0.8990919,
      "num_input_tokens_seen": 95895820,
      "step": 4436,
      "time_per_iteration": 2.6219725608825684
    },
    {
      "auxiliary_loss_clip": 0.0111722,
      "auxiliary_loss_mlp": 0.01047436,
      "balance_loss_clip": 1.04947853,
      "balance_loss_mlp": 1.03062761,
      "epoch": 0.26676687208777994,
      "flos": 25812123413760.0,
      "grad_norm": 1.4205169612174395,
      "language_loss": 0.78481591,
      "learning_rate": 3.440199789988407e-06,
      "loss": 0.80646253,
      "num_input_tokens_seen": 95918025,
      "step": 4437,
      "time_per_iteration": 2.6007931232452393
    },
    {
      "auxiliary_loss_clip": 0.01082908,
      "auxiliary_loss_mlp": 0.01044531,
      "balance_loss_clip": 1.04704571,
      "balance_loss_mlp": 1.02796054,
      "epoch": 0.2668269953404479,
      "flos": 36064583504640.0,
      "grad_norm": 2.121767228001481,
      "language_loss": 0.6386252,
      "learning_rate": 3.439929526748556e-06,
      "loss": 0.65989959,
      "num_input_tokens_seen": 95937725,
      "step": 4438,
      "time_per_iteration": 4.1463892459869385
    },
    {
      "auxiliary_loss_clip": 0.01076665,
      "auxiliary_loss_mlp": 0.01041793,
      "balance_loss_clip": 1.04625285,
      "balance_loss_mlp": 1.02503181,
      "epoch": 0.26688711859311587,
      "flos": 26570987072640.0,
      "grad_norm": 1.7623683861652961,
      "language_loss": 0.7558741,
      "learning_rate": 3.4396592089064334e-06,
      "loss": 0.77705866,
      "num_input_tokens_seen": 95956335,
      "step": 4439,
      "time_per_iteration": 2.661316156387329
    },
    {
      "auxiliary_loss_clip": 0.01082716,
      "auxiliary_loss_mlp": 0.01039724,
      "balance_loss_clip": 1.05045295,
      "balance_loss_mlp": 1.02135396,
      "epoch": 0.26694724184578383,
      "flos": 26761565658240.0,
      "grad_norm": 1.8028190082303404,
      "language_loss": 0.71727204,
      "learning_rate": 3.4393888364722897e-06,
      "loss": 0.73849642,
      "num_input_tokens_seen": 95977135,
      "step": 4440,
      "time_per_iteration": 2.8228225708007812
    },
    {
      "auxiliary_loss_clip": 0.01117812,
      "auxiliary_loss_mlp": 0.01043118,
      "balance_loss_clip": 1.04706824,
      "balance_loss_mlp": 1.0243907,
      "epoch": 0.2670073650984518,
      "flos": 20959586536320.0,
      "grad_norm": 2.0135333594781226,
      "language_loss": 0.66859025,
      "learning_rate": 3.439118409456376e-06,
      "loss": 0.69019961,
      "num_input_tokens_seen": 95995435,
      "step": 4441,
      "time_per_iteration": 4.0875084400177
    },
    {
      "auxiliary_loss_clip": 0.01132056,
      "auxiliary_loss_mlp": 0.01044278,
      "balance_loss_clip": 1.05036354,
      "balance_loss_mlp": 1.0262413,
      "epoch": 0.2670674883511198,
      "flos": 28366054277760.0,
      "grad_norm": 1.4930384976745803,
      "language_loss": 0.76444209,
      "learning_rate": 3.4388479278689486e-06,
      "loss": 0.78620547,
      "num_input_tokens_seen": 96016340,
      "step": 4442,
      "time_per_iteration": 2.5793521404266357
    },
    {
      "auxiliary_loss_clip": 0.01015655,
      "auxiliary_loss_mlp": 0.01006796,
      "balance_loss_clip": 1.04582334,
      "balance_loss_mlp": 1.00414968,
      "epoch": 0.2671276116037878,
      "flos": 58971319430400.0,
      "grad_norm": 0.9390549358690226,
      "language_loss": 0.61215156,
      "learning_rate": 3.4385773917202637e-06,
      "loss": 0.63237613,
      "num_input_tokens_seen": 96071205,
      "step": 4443,
      "time_per_iteration": 4.521975040435791
    },
    {
      "auxiliary_loss_clip": 0.01110311,
      "auxiliary_loss_mlp": 0.01038758,
      "balance_loss_clip": 1.04921758,
      "balance_loss_mlp": 1.02234268,
      "epoch": 0.26718773485645575,
      "flos": 43945072053120.0,
      "grad_norm": 1.539144949862099,
      "language_loss": 0.76472688,
      "learning_rate": 3.4383068010205793e-06,
      "loss": 0.78621757,
      "num_input_tokens_seen": 96094240,
      "step": 4444,
      "time_per_iteration": 2.9049065113067627
    },
    {
      "auxiliary_loss_clip": 0.01131106,
      "auxiliary_loss_mlp": 0.01041508,
      "balance_loss_clip": 1.04940248,
      "balance_loss_mlp": 1.02310228,
      "epoch": 0.2672478581091237,
      "flos": 25228323665280.0,
      "grad_norm": 1.7877005171646547,
      "language_loss": 0.80601805,
      "learning_rate": 3.438036155780158e-06,
      "loss": 0.82774419,
      "num_input_tokens_seen": 96114105,
      "step": 4445,
      "time_per_iteration": 2.544567346572876
    },
    {
      "auxiliary_loss_clip": 0.01117477,
      "auxiliary_loss_mlp": 0.01041654,
      "balance_loss_clip": 1.05175412,
      "balance_loss_mlp": 1.02356958,
      "epoch": 0.2673079813617917,
      "flos": 15268176455040.0,
      "grad_norm": 1.6900831222016517,
      "language_loss": 0.88785595,
      "learning_rate": 3.43776545600926e-06,
      "loss": 0.90944731,
      "num_input_tokens_seen": 96132140,
      "step": 4446,
      "time_per_iteration": 2.506457567214966
    },
    {
      "auxiliary_loss_clip": 0.01137587,
      "auxiliary_loss_mlp": 0.01045312,
      "balance_loss_clip": 1.05390275,
      "balance_loss_mlp": 1.02822983,
      "epoch": 0.26736810461445965,
      "flos": 25812733944960.0,
      "grad_norm": 1.712450017647839,
      "language_loss": 0.68322051,
      "learning_rate": 3.437494701718153e-06,
      "loss": 0.70504946,
      "num_input_tokens_seen": 96152090,
      "step": 4447,
      "time_per_iteration": 2.5620434284210205
    },
    {
      "auxiliary_loss_clip": 0.01131958,
      "auxiliary_loss_mlp": 0.0103367,
      "balance_loss_clip": 1.04934859,
      "balance_loss_mlp": 1.01625395,
      "epoch": 0.2674282278671276,
      "flos": 24312709054080.0,
      "grad_norm": 1.84329767421907,
      "language_loss": 0.83499849,
      "learning_rate": 3.4372238929171026e-06,
      "loss": 0.85665476,
      "num_input_tokens_seen": 96170015,
      "step": 4448,
      "time_per_iteration": 2.5200748443603516
    },
    {
      "auxiliary_loss_clip": 0.01106577,
      "auxiliary_loss_mlp": 0.01047458,
      "balance_loss_clip": 1.05090284,
      "balance_loss_mlp": 1.0299108,
      "epoch": 0.2674883511197956,
      "flos": 22815521337600.0,
      "grad_norm": 1.5881662232920335,
      "language_loss": 0.84498179,
      "learning_rate": 3.436953029616378e-06,
      "loss": 0.86652213,
      "num_input_tokens_seen": 96188065,
      "step": 4449,
      "time_per_iteration": 2.5832161903381348
    },
    {
      "auxiliary_loss_clip": 0.01125331,
      "auxiliary_loss_mlp": 0.01050869,
      "balance_loss_clip": 1.0520699,
      "balance_loss_mlp": 1.03134286,
      "epoch": 0.26754847437246354,
      "flos": 25370170473600.0,
      "grad_norm": 1.702608161601835,
      "language_loss": 0.84063292,
      "learning_rate": 3.4366821118262506e-06,
      "loss": 0.86239493,
      "num_input_tokens_seen": 96205780,
      "step": 4450,
      "time_per_iteration": 2.556450605392456
    },
    {
      "auxiliary_loss_clip": 0.01099728,
      "auxiliary_loss_mlp": 0.01046402,
      "balance_loss_clip": 1.04524529,
      "balance_loss_mlp": 1.03057146,
      "epoch": 0.2676085976251315,
      "flos": 20230420446720.0,
      "grad_norm": 1.8332219727447805,
      "language_loss": 0.80917144,
      "learning_rate": 3.4364111395569937e-06,
      "loss": 0.83063275,
      "num_input_tokens_seen": 96224990,
      "step": 4451,
      "time_per_iteration": 2.5719053745269775
    },
    {
      "auxiliary_loss_clip": 0.01131679,
      "auxiliary_loss_mlp": 0.0104383,
      "balance_loss_clip": 1.05191159,
      "balance_loss_mlp": 1.02753377,
      "epoch": 0.26766872087779947,
      "flos": 28038225824640.0,
      "grad_norm": 1.747725974215543,
      "language_loss": 0.86410314,
      "learning_rate": 3.436140112818882e-06,
      "loss": 0.88585824,
      "num_input_tokens_seen": 96245345,
      "step": 4452,
      "time_per_iteration": 2.5630733966827393
    },
    {
      "auxiliary_loss_clip": 0.01128667,
      "auxiliary_loss_mlp": 0.01041628,
      "balance_loss_clip": 1.05152178,
      "balance_loss_mlp": 1.02424717,
      "epoch": 0.26772884413046744,
      "flos": 18325179250560.0,
      "grad_norm": 1.9421305910331945,
      "language_loss": 0.83588606,
      "learning_rate": 3.435869031622194e-06,
      "loss": 0.85758901,
      "num_input_tokens_seen": 96259000,
      "step": 4453,
      "time_per_iteration": 2.5198562145233154
    },
    {
      "auxiliary_loss_clip": 0.01129044,
      "auxiliary_loss_mlp": 0.01053853,
      "balance_loss_clip": 1.0491432,
      "balance_loss_mlp": 1.0359242,
      "epoch": 0.2677889673831354,
      "flos": 22127509255680.0,
      "grad_norm": 1.5747280391345864,
      "language_loss": 0.7953862,
      "learning_rate": 3.435597895977208e-06,
      "loss": 0.8172152,
      "num_input_tokens_seen": 96277000,
      "step": 4454,
      "time_per_iteration": 2.506873369216919
    },
    {
      "auxiliary_loss_clip": 0.01121252,
      "auxiliary_loss_mlp": 0.01045283,
      "balance_loss_clip": 1.04929137,
      "balance_loss_mlp": 1.02861774,
      "epoch": 0.2678490906358034,
      "flos": 23729699404800.0,
      "grad_norm": 1.5975597461759021,
      "language_loss": 0.72825682,
      "learning_rate": 3.435326705894206e-06,
      "loss": 0.74992222,
      "num_input_tokens_seen": 96297010,
      "step": 4455,
      "time_per_iteration": 2.5931975841522217
    },
    {
      "auxiliary_loss_clip": 0.01103411,
      "auxiliary_loss_mlp": 0.0103854,
      "balance_loss_clip": 1.04691625,
      "balance_loss_mlp": 1.02189851,
      "epoch": 0.2679092138884714,
      "flos": 21762872340480.0,
      "grad_norm": 1.7569992112633872,
      "language_loss": 0.73731369,
      "learning_rate": 3.435055461383471e-06,
      "loss": 0.75873321,
      "num_input_tokens_seen": 96315780,
      "step": 4456,
      "time_per_iteration": 2.5601377487182617
    },
    {
      "auxiliary_loss_clip": 0.01132101,
      "auxiliary_loss_mlp": 0.01043958,
      "balance_loss_clip": 1.05058622,
      "balance_loss_mlp": 1.02677965,
      "epoch": 0.26796933714113935,
      "flos": 19861186590720.0,
      "grad_norm": 2.3166750412910373,
      "language_loss": 0.71135765,
      "learning_rate": 3.4347841624552896e-06,
      "loss": 0.73311824,
      "num_input_tokens_seen": 96333465,
      "step": 4457,
      "time_per_iteration": 2.496516704559326
    },
    {
      "auxiliary_loss_clip": 0.01104468,
      "auxiliary_loss_mlp": 0.01050712,
      "balance_loss_clip": 1.05158019,
      "balance_loss_mlp": 1.03297377,
      "epoch": 0.2680294603938073,
      "flos": 20047886507520.0,
      "grad_norm": 1.7088232607677534,
      "language_loss": 0.7907716,
      "learning_rate": 3.4345128091199493e-06,
      "loss": 0.81232339,
      "num_input_tokens_seen": 96352005,
      "step": 4458,
      "time_per_iteration": 2.549220323562622
    },
    {
      "auxiliary_loss_clip": 0.01025929,
      "auxiliary_loss_mlp": 0.01010182,
      "balance_loss_clip": 1.03887081,
      "balance_loss_mlp": 1.00791728,
      "epoch": 0.2680895836464753,
      "flos": 72113763052800.0,
      "grad_norm": 0.8638735331751527,
      "language_loss": 0.58753645,
      "learning_rate": 3.434241401387739e-06,
      "loss": 0.60789752,
      "num_input_tokens_seen": 96406265,
      "step": 4459,
      "time_per_iteration": 3.134660005569458
    },
    {
      "auxiliary_loss_clip": 0.01087013,
      "auxiliary_loss_mlp": 0.01048586,
      "balance_loss_clip": 1.04771137,
      "balance_loss_mlp": 1.03096747,
      "epoch": 0.26814970689914325,
      "flos": 20449044576000.0,
      "grad_norm": 2.1167551641864364,
      "language_loss": 0.8488428,
      "learning_rate": 3.4339699392689507e-06,
      "loss": 0.87019873,
      "num_input_tokens_seen": 96425225,
      "step": 4460,
      "time_per_iteration": 2.594700813293457
    },
    {
      "auxiliary_loss_clip": 0.01123927,
      "auxiliary_loss_mlp": 0.0105005,
      "balance_loss_clip": 1.04747891,
      "balance_loss_mlp": 1.03127444,
      "epoch": 0.2682098301518112,
      "flos": 17566674727680.0,
      "grad_norm": 1.6824790048684086,
      "language_loss": 0.69000703,
      "learning_rate": 3.4336984227738796e-06,
      "loss": 0.71174681,
      "num_input_tokens_seen": 96443780,
      "step": 4461,
      "time_per_iteration": 2.496055841445923
    },
    {
      "auxiliary_loss_clip": 0.01110421,
      "auxiliary_loss_mlp": 0.01051103,
      "balance_loss_clip": 1.04753959,
      "balance_loss_mlp": 1.03328133,
      "epoch": 0.2682699534044792,
      "flos": 18333259810560.0,
      "grad_norm": 1.5007190295104833,
      "language_loss": 0.67323077,
      "learning_rate": 3.43342685191282e-06,
      "loss": 0.69484597,
      "num_input_tokens_seen": 96464530,
      "step": 4462,
      "time_per_iteration": 2.574934244155884
    },
    {
      "auxiliary_loss_clip": 0.01108192,
      "auxiliary_loss_mlp": 0.01043412,
      "balance_loss_clip": 1.04988587,
      "balance_loss_mlp": 1.02579343,
      "epoch": 0.26833007665714714,
      "flos": 25301294144640.0,
      "grad_norm": 1.68020691687763,
      "language_loss": 0.69699466,
      "learning_rate": 3.4331552266960705e-06,
      "loss": 0.71851069,
      "num_input_tokens_seen": 96483345,
      "step": 4463,
      "time_per_iteration": 2.5978140830993652
    },
    {
      "auxiliary_loss_clip": 0.01111255,
      "auxiliary_loss_mlp": 0.01042828,
      "balance_loss_clip": 1.05031061,
      "balance_loss_mlp": 1.02381396,
      "epoch": 0.2683901999098151,
      "flos": 16099759198080.0,
      "grad_norm": 4.196534262942418,
      "language_loss": 0.77649295,
      "learning_rate": 3.432883547133931e-06,
      "loss": 0.79803383,
      "num_input_tokens_seen": 96498305,
      "step": 4464,
      "time_per_iteration": 2.511720895767212
    },
    {
      "auxiliary_loss_clip": 0.01123149,
      "auxiliary_loss_mlp": 0.01046319,
      "balance_loss_clip": 1.04771745,
      "balance_loss_mlp": 1.02816391,
      "epoch": 0.2684503231624831,
      "flos": 27308054154240.0,
      "grad_norm": 1.84370009639088,
      "language_loss": 0.71120918,
      "learning_rate": 3.432611813236704e-06,
      "loss": 0.73290384,
      "num_input_tokens_seen": 96519740,
      "step": 4465,
      "time_per_iteration": 2.537893056869507
    },
    {
      "auxiliary_loss_clip": 0.0103385,
      "auxiliary_loss_mlp": 0.01008438,
      "balance_loss_clip": 1.02243781,
      "balance_loss_mlp": 1.00634027,
      "epoch": 0.26851044641515104,
      "flos": 71858007239040.0,
      "grad_norm": 0.7024333089637484,
      "language_loss": 0.53084028,
      "learning_rate": 3.4323400250146943e-06,
      "loss": 0.55126315,
      "num_input_tokens_seen": 96588870,
      "step": 4466,
      "time_per_iteration": 3.266301155090332
    },
    {
      "auxiliary_loss_clip": 0.01112109,
      "auxiliary_loss_mlp": 0.01055079,
      "balance_loss_clip": 1.04691613,
      "balance_loss_mlp": 1.03520691,
      "epoch": 0.268570569667819,
      "flos": 18733771434240.0,
      "grad_norm": 2.0183857375358683,
      "language_loss": 0.7368573,
      "learning_rate": 3.4320681824782057e-06,
      "loss": 0.75852919,
      "num_input_tokens_seen": 96605100,
      "step": 4467,
      "time_per_iteration": 2.496220827102661
    },
    {
      "auxiliary_loss_clip": 0.01116474,
      "auxiliary_loss_mlp": 0.00817638,
      "balance_loss_clip": 1.04721522,
      "balance_loss_mlp": 1.05253947,
      "epoch": 0.268630692920487,
      "flos": 18178376365440.0,
      "grad_norm": 2.09820191206727,
      "language_loss": 0.8037039,
      "learning_rate": 3.4317962856375493e-06,
      "loss": 0.82304496,
      "num_input_tokens_seen": 96621410,
      "step": 4468,
      "time_per_iteration": 2.5150015354156494
    },
    {
      "auxiliary_loss_clip": 0.01047537,
      "auxiliary_loss_mlp": 0.01002468,
      "balance_loss_clip": 1.01688409,
      "balance_loss_mlp": 1.00052536,
      "epoch": 0.268690816173155,
      "flos": 68731768978560.0,
      "grad_norm": 0.8433721372102897,
      "language_loss": 0.5952903,
      "learning_rate": 3.4315243345030334e-06,
      "loss": 0.61579037,
      "num_input_tokens_seen": 96684810,
      "step": 4469,
      "time_per_iteration": 3.1412925720214844
    },
    {
      "auxiliary_loss_clip": 0.01144386,
      "auxiliary_loss_mlp": 0.01051764,
      "balance_loss_clip": 1.05051684,
      "balance_loss_mlp": 1.0324403,
      "epoch": 0.26875093942582295,
      "flos": 23293636295040.0,
      "grad_norm": 1.8171657653953754,
      "language_loss": 0.81470454,
      "learning_rate": 3.431252329084972e-06,
      "loss": 0.83666605,
      "num_input_tokens_seen": 96701920,
      "step": 4470,
      "time_per_iteration": 2.4732558727264404
    },
    {
      "auxiliary_loss_clip": 0.011076,
      "auxiliary_loss_mlp": 0.01038979,
      "balance_loss_clip": 1.04698515,
      "balance_loss_mlp": 1.0210855,
      "epoch": 0.2688110626784909,
      "flos": 21543458112000.0,
      "grad_norm": 1.6020260168754987,
      "language_loss": 0.82371807,
      "learning_rate": 3.4309802693936786e-06,
      "loss": 0.84518391,
      "num_input_tokens_seen": 96721260,
      "step": 4471,
      "time_per_iteration": 2.5486056804656982
    },
    {
      "auxiliary_loss_clip": 0.01123472,
      "auxiliary_loss_mlp": 0.01034752,
      "balance_loss_clip": 1.04969656,
      "balance_loss_mlp": 1.01808643,
      "epoch": 0.2688711859311589,
      "flos": 28400600183040.0,
      "grad_norm": 2.000035490320257,
      "language_loss": 0.69287491,
      "learning_rate": 3.43070815543947e-06,
      "loss": 0.71445715,
      "num_input_tokens_seen": 96740385,
      "step": 4472,
      "time_per_iteration": 2.5673751831054688
    },
    {
      "auxiliary_loss_clip": 0.01141421,
      "auxiliary_loss_mlp": 0.01046354,
      "balance_loss_clip": 1.05152702,
      "balance_loss_mlp": 1.02948558,
      "epoch": 0.26893130918382685,
      "flos": 25994944661760.0,
      "grad_norm": 1.6047456296616176,
      "language_loss": 0.6793195,
      "learning_rate": 3.4304359872326656e-06,
      "loss": 0.70119727,
      "num_input_tokens_seen": 96761860,
      "step": 4473,
      "time_per_iteration": 2.5278029441833496
    },
    {
      "auxiliary_loss_clip": 0.01110286,
      "auxiliary_loss_mlp": 0.01044422,
      "balance_loss_clip": 1.05326653,
      "balance_loss_mlp": 1.02789927,
      "epoch": 0.2689914324364948,
      "flos": 20339624770560.0,
      "grad_norm": 2.5591362537211557,
      "language_loss": 0.83369005,
      "learning_rate": 3.4301637647835843e-06,
      "loss": 0.85523713,
      "num_input_tokens_seen": 96781890,
      "step": 4474,
      "time_per_iteration": 3.988219738006592
    },
    {
      "auxiliary_loss_clip": 0.01133581,
      "auxiliary_loss_mlp": 0.01048487,
      "balance_loss_clip": 1.05033731,
      "balance_loss_mlp": 1.03160667,
      "epoch": 0.2690515556891628,
      "flos": 19464553635840.0,
      "grad_norm": 1.7354524440223338,
      "language_loss": 0.70562327,
      "learning_rate": 3.4298914881025494e-06,
      "loss": 0.72744393,
      "num_input_tokens_seen": 96800390,
      "step": 4475,
      "time_per_iteration": 2.515655994415283
    },
    {
      "auxiliary_loss_clip": 0.01106248,
      "auxiliary_loss_mlp": 0.00799418,
      "balance_loss_clip": 1.05015504,
      "balance_loss_mlp": 1.01873732,
      "epoch": 0.26911167894183075,
      "flos": 18146631720960.0,
      "grad_norm": 1.688227134106577,
      "language_loss": 0.73619485,
      "learning_rate": 3.4296191571998863e-06,
      "loss": 0.75525153,
      "num_input_tokens_seen": 96816685,
      "step": 4476,
      "time_per_iteration": 2.5385799407958984
    },
    {
      "auxiliary_loss_clip": 0.01110408,
      "auxiliary_loss_mlp": 0.01041315,
      "balance_loss_clip": 1.0488379,
      "balance_loss_mlp": 1.02450597,
      "epoch": 0.2691718021944987,
      "flos": 19975131509760.0,
      "grad_norm": 1.598061615280744,
      "language_loss": 0.80687028,
      "learning_rate": 3.429346772085922e-06,
      "loss": 0.8283875,
      "num_input_tokens_seen": 96836285,
      "step": 4477,
      "time_per_iteration": 3.918288230895996
    },
    {
      "auxiliary_loss_clip": 0.01090335,
      "auxiliary_loss_mlp": 0.0104428,
      "balance_loss_clip": 1.05206299,
      "balance_loss_mlp": 1.02730441,
      "epoch": 0.2692319254471667,
      "flos": 37447215770880.0,
      "grad_norm": 1.609309327645889,
      "language_loss": 0.65358973,
      "learning_rate": 3.429074332770984e-06,
      "loss": 0.67493588,
      "num_input_tokens_seen": 96857745,
      "step": 4478,
      "time_per_iteration": 2.7518298625946045
    },
    {
      "auxiliary_loss_clip": 0.01125696,
      "auxiliary_loss_mlp": 0.0104342,
      "balance_loss_clip": 1.0526123,
      "balance_loss_mlp": 1.02607477,
      "epoch": 0.26929204869983464,
      "flos": 22127796564480.0,
      "grad_norm": 2.5295545498049075,
      "language_loss": 0.80842757,
      "learning_rate": 3.4288018392654047e-06,
      "loss": 0.83011866,
      "num_input_tokens_seen": 96877295,
      "step": 4479,
      "time_per_iteration": 2.532438278198242
    },
    {
      "auxiliary_loss_clip": 0.0111564,
      "auxiliary_loss_mlp": 0.00803326,
      "balance_loss_clip": 1.04490662,
      "balance_loss_mlp": 1.02934241,
      "epoch": 0.2693521719525026,
      "flos": 19792813052160.0,
      "grad_norm": 1.932924500821838,
      "language_loss": 0.80649167,
      "learning_rate": 3.4285292915795166e-06,
      "loss": 0.82568139,
      "num_input_tokens_seen": 96896160,
      "step": 4480,
      "time_per_iteration": 3.9151158332824707
    },
    {
      "auxiliary_loss_clip": 0.01090093,
      "auxiliary_loss_mlp": 0.01043445,
      "balance_loss_clip": 1.04528522,
      "balance_loss_mlp": 1.02624357,
      "epoch": 0.2694122952051706,
      "flos": 20994383836800.0,
      "grad_norm": 1.5497070024897408,
      "language_loss": 0.78247821,
      "learning_rate": 3.4282566897236543e-06,
      "loss": 0.80381358,
      "num_input_tokens_seen": 96915410,
      "step": 4481,
      "time_per_iteration": 3.9661073684692383
    },
    {
      "auxiliary_loss_clip": 0.01128721,
      "auxiliary_loss_mlp": 0.01045143,
      "balance_loss_clip": 1.04769182,
      "balance_loss_mlp": 1.02771509,
      "epoch": 0.2694724184578386,
      "flos": 25849291011840.0,
      "grad_norm": 1.6268377659048607,
      "language_loss": 0.74103439,
      "learning_rate": 3.4279840337081547e-06,
      "loss": 0.76277304,
      "num_input_tokens_seen": 96937865,
      "step": 4482,
      "time_per_iteration": 2.554187297821045
    },
    {
      "auxiliary_loss_clip": 0.01114032,
      "auxiliary_loss_mlp": 0.0104006,
      "balance_loss_clip": 1.05001283,
      "balance_loss_mlp": 1.02277493,
      "epoch": 0.26953254171050656,
      "flos": 21726961718400.0,
      "grad_norm": 1.845516773710618,
      "language_loss": 0.72103763,
      "learning_rate": 3.4277113235433584e-06,
      "loss": 0.74257851,
      "num_input_tokens_seen": 96957710,
      "step": 4483,
      "time_per_iteration": 2.551866292953491
    },
    {
      "auxiliary_loss_clip": 0.01127827,
      "auxiliary_loss_mlp": 0.01043592,
      "balance_loss_clip": 1.04464364,
      "balance_loss_mlp": 1.02548409,
      "epoch": 0.2695926649631745,
      "flos": 19682926369920.0,
      "grad_norm": 2.243586414976352,
      "language_loss": 0.86571479,
      "learning_rate": 3.427438559239605e-06,
      "loss": 0.887429,
      "num_input_tokens_seen": 96975890,
      "step": 4484,
      "time_per_iteration": 2.4865710735321045
    },
    {
      "auxiliary_loss_clip": 0.01131473,
      "auxiliary_loss_mlp": 0.01041479,
      "balance_loss_clip": 1.04900467,
      "balance_loss_mlp": 1.02479017,
      "epoch": 0.2696527882158425,
      "flos": 32886596724480.0,
      "grad_norm": 1.4793412766113272,
      "language_loss": 0.66303736,
      "learning_rate": 3.427165740807239e-06,
      "loss": 0.68476689,
      "num_input_tokens_seen": 96998595,
      "step": 4485,
      "time_per_iteration": 2.6156177520751953
    },
    {
      "auxiliary_loss_clip": 0.01106626,
      "auxiliary_loss_mlp": 0.01043923,
      "balance_loss_clip": 1.04639256,
      "balance_loss_mlp": 1.02619624,
      "epoch": 0.26971291146851045,
      "flos": 12124843320960.0,
      "grad_norm": 1.9959194863007592,
      "language_loss": 0.72878277,
      "learning_rate": 3.426892868256604e-06,
      "loss": 0.75028825,
      "num_input_tokens_seen": 97013715,
      "step": 4486,
      "time_per_iteration": 2.494731903076172
    },
    {
      "auxiliary_loss_clip": 0.01143223,
      "auxiliary_loss_mlp": 0.01042378,
      "balance_loss_clip": 1.04994547,
      "balance_loss_mlp": 1.02579641,
      "epoch": 0.2697730347211784,
      "flos": 22634459856000.0,
      "grad_norm": 2.2394008333884927,
      "language_loss": 0.84021556,
      "learning_rate": 3.4266199415980495e-06,
      "loss": 0.86207151,
      "num_input_tokens_seen": 97031570,
      "step": 4487,
      "time_per_iteration": 2.476351499557495
    },
    {
      "auxiliary_loss_clip": 0.01112031,
      "auxiliary_loss_mlp": 0.01048766,
      "balance_loss_clip": 1.04656935,
      "balance_loss_mlp": 1.03122997,
      "epoch": 0.2698331579738464,
      "flos": 23513050523520.0,
      "grad_norm": 1.9792413235406863,
      "language_loss": 0.71202135,
      "learning_rate": 3.4263469608419234e-06,
      "loss": 0.73362935,
      "num_input_tokens_seen": 97049815,
      "step": 4488,
      "time_per_iteration": 2.5568771362304688
    },
    {
      "auxiliary_loss_clip": 0.01064416,
      "auxiliary_loss_mlp": 0.01054286,
      "balance_loss_clip": 1.04693699,
      "balance_loss_mlp": 1.03470039,
      "epoch": 0.26989328122651435,
      "flos": 24641040297600.0,
      "grad_norm": 1.6920952147906192,
      "language_loss": 0.83591771,
      "learning_rate": 3.426073925998578e-06,
      "loss": 0.85710466,
      "num_input_tokens_seen": 97067570,
      "step": 4489,
      "time_per_iteration": 2.7415268421173096
    },
    {
      "auxiliary_loss_clip": 0.01117084,
      "auxiliary_loss_mlp": 0.01053478,
      "balance_loss_clip": 1.05060363,
      "balance_loss_mlp": 1.03547716,
      "epoch": 0.2699534044791823,
      "flos": 10772555068800.0,
      "grad_norm": 2.645649836290693,
      "language_loss": 0.89948237,
      "learning_rate": 3.4258008370783656e-06,
      "loss": 0.92118794,
      "num_input_tokens_seen": 97082180,
      "step": 4490,
      "time_per_iteration": 2.7907402515411377
    },
    {
      "auxiliary_loss_clip": 0.01067402,
      "auxiliary_loss_mlp": 0.01047596,
      "balance_loss_clip": 1.04283535,
      "balance_loss_mlp": 1.02870131,
      "epoch": 0.2700135277318503,
      "flos": 36171597098880.0,
      "grad_norm": 2.1400600397898684,
      "language_loss": 0.73436946,
      "learning_rate": 3.4255276940916434e-06,
      "loss": 0.75551939,
      "num_input_tokens_seen": 97103470,
      "step": 4491,
      "time_per_iteration": 2.7784411907196045
    },
    {
      "auxiliary_loss_clip": 0.01143226,
      "auxiliary_loss_mlp": 0.01043648,
      "balance_loss_clip": 1.05035329,
      "balance_loss_mlp": 1.02587366,
      "epoch": 0.27007365098451824,
      "flos": 17418614866560.0,
      "grad_norm": 2.6454558244736437,
      "language_loss": 0.74234641,
      "learning_rate": 3.4252544970487676e-06,
      "loss": 0.76421517,
      "num_input_tokens_seen": 97118100,
      "step": 4492,
      "time_per_iteration": 2.4473721981048584
    },
    {
      "auxiliary_loss_clip": 0.01119635,
      "auxiliary_loss_mlp": 0.01039173,
      "balance_loss_clip": 1.04784238,
      "balance_loss_mlp": 1.02107668,
      "epoch": 0.2701337742371862,
      "flos": 23185688947200.0,
      "grad_norm": 1.829117691457779,
      "language_loss": 0.89527071,
      "learning_rate": 3.4249812459600986e-06,
      "loss": 0.91685879,
      "num_input_tokens_seen": 97136765,
      "step": 4493,
      "time_per_iteration": 2.5625925064086914
    },
    {
      "auxiliary_loss_clip": 0.01131222,
      "auxiliary_loss_mlp": 0.01040813,
      "balance_loss_clip": 1.04900217,
      "balance_loss_mlp": 1.02482677,
      "epoch": 0.2701938974898542,
      "flos": 24389450461440.0,
      "grad_norm": 1.3998316055978226,
      "language_loss": 0.71245718,
      "learning_rate": 3.424707940835998e-06,
      "loss": 0.73417759,
      "num_input_tokens_seen": 97157470,
      "step": 4494,
      "time_per_iteration": 2.5274531841278076
    },
    {
      "auxiliary_loss_clip": 0.01109175,
      "auxiliary_loss_mlp": 0.01033513,
      "balance_loss_clip": 1.04669833,
      "balance_loss_mlp": 1.01779544,
      "epoch": 0.2702540207425222,
      "flos": 26214322976640.0,
      "grad_norm": 1.9982972689288083,
      "language_loss": 0.86037445,
      "learning_rate": 3.42443458168683e-06,
      "loss": 0.88180137,
      "num_input_tokens_seen": 97176905,
      "step": 4495,
      "time_per_iteration": 2.573192834854126
    },
    {
      "auxiliary_loss_clip": 0.01141665,
      "auxiliary_loss_mlp": 0.0104837,
      "balance_loss_clip": 1.05089402,
      "balance_loss_mlp": 1.03108478,
      "epoch": 0.27031414399519016,
      "flos": 22926377687040.0,
      "grad_norm": 1.6995144344735065,
      "language_loss": 0.76506186,
      "learning_rate": 3.424161168522959e-06,
      "loss": 0.78696221,
      "num_input_tokens_seen": 97196380,
      "step": 4496,
      "time_per_iteration": 2.4609932899475098
    },
    {
      "auxiliary_loss_clip": 0.01048548,
      "auxiliary_loss_mlp": 0.01021007,
      "balance_loss_clip": 1.01847565,
      "balance_loss_mlp": 1.01903963,
      "epoch": 0.2703742672478581,
      "flos": 63019780404480.0,
      "grad_norm": 0.740758459414504,
      "language_loss": 0.50290072,
      "learning_rate": 3.423887701354754e-06,
      "loss": 0.52359623,
      "num_input_tokens_seen": 97260100,
      "step": 4497,
      "time_per_iteration": 3.1563968658447266
    },
    {
      "auxiliary_loss_clip": 0.01105101,
      "auxiliary_loss_mlp": 0.01048364,
      "balance_loss_clip": 1.05334675,
      "balance_loss_mlp": 1.03175807,
      "epoch": 0.2704343905005261,
      "flos": 18840820942080.0,
      "grad_norm": 1.8983785311463663,
      "language_loss": 0.72455627,
      "learning_rate": 3.4236141801925847e-06,
      "loss": 0.74609089,
      "num_input_tokens_seen": 97277935,
      "step": 4498,
      "time_per_iteration": 2.6308045387268066
    },
    {
      "auxiliary_loss_clip": 0.01027131,
      "auxiliary_loss_mlp": 0.01007791,
      "balance_loss_clip": 1.02563977,
      "balance_loss_mlp": 1.00577629,
      "epoch": 0.27049451375319405,
      "flos": 71233412618880.0,
      "grad_norm": 0.7508698867175054,
      "language_loss": 0.59205633,
      "learning_rate": 3.4233406050468237e-06,
      "loss": 0.61240554,
      "num_input_tokens_seen": 97338845,
      "step": 4499,
      "time_per_iteration": 3.1805806159973145
    },
    {
      "auxiliary_loss_clip": 0.0111615,
      "auxiliary_loss_mlp": 0.01038228,
      "balance_loss_clip": 1.04770088,
      "balance_loss_mlp": 1.0210619,
      "epoch": 0.270554637005862,
      "flos": 24278594112000.0,
      "grad_norm": 1.8887649400601636,
      "language_loss": 0.7335906,
      "learning_rate": 3.4230669759278438e-06,
      "loss": 0.75513434,
      "num_input_tokens_seen": 97356640,
      "step": 4500,
      "time_per_iteration": 2.548063278198242
    },
    {
      "auxiliary_loss_clip": 0.0110765,
      "auxiliary_loss_mlp": 0.01039438,
      "balance_loss_clip": 1.04454708,
      "balance_loss_mlp": 1.02140117,
      "epoch": 0.27061476025853,
      "flos": 17632318832640.0,
      "grad_norm": 2.7793610667265596,
      "language_loss": 0.82141984,
      "learning_rate": 3.4227932928460215e-06,
      "loss": 0.84289068,
      "num_input_tokens_seen": 97372585,
      "step": 4501,
      "time_per_iteration": 2.483837842941284
    },
    {
      "auxiliary_loss_clip": 0.01096542,
      "auxiliary_loss_mlp": 0.01049972,
      "balance_loss_clip": 1.04841316,
      "balance_loss_mlp": 1.03074312,
      "epoch": 0.27067488351119795,
      "flos": 22710123855360.0,
      "grad_norm": 1.6585083974985475,
      "language_loss": 0.72276747,
      "learning_rate": 3.422519555811735e-06,
      "loss": 0.74423254,
      "num_input_tokens_seen": 97393315,
      "step": 4502,
      "time_per_iteration": 2.5873076915740967
    },
    {
      "auxiliary_loss_clip": 0.01119053,
      "auxiliary_loss_mlp": 0.01040381,
      "balance_loss_clip": 1.04458654,
      "balance_loss_mlp": 1.02185643,
      "epoch": 0.2707350067638659,
      "flos": 41719616087040.0,
      "grad_norm": 2.672846836716759,
      "language_loss": 0.6845091,
      "learning_rate": 3.4222457648353642e-06,
      "loss": 0.70610344,
      "num_input_tokens_seen": 97417860,
      "step": 4503,
      "time_per_iteration": 2.7004613876342773
    },
    {
      "auxiliary_loss_clip": 0.01093346,
      "auxiliary_loss_mlp": 0.0104818,
      "balance_loss_clip": 1.0456419,
      "balance_loss_mlp": 1.02846241,
      "epoch": 0.2707951300165339,
      "flos": 20193037367040.0,
      "grad_norm": 1.9340405421063862,
      "language_loss": 0.68538749,
      "learning_rate": 3.4219719199272918e-06,
      "loss": 0.70680273,
      "num_input_tokens_seen": 97436780,
      "step": 4504,
      "time_per_iteration": 2.5856242179870605
    },
    {
      "auxiliary_loss_clip": 0.01134226,
      "auxiliary_loss_mlp": 0.01045619,
      "balance_loss_clip": 1.053473,
      "balance_loss_mlp": 1.02932334,
      "epoch": 0.27085525326920185,
      "flos": 21433966479360.0,
      "grad_norm": 2.2360337874702956,
      "language_loss": 0.75650901,
      "learning_rate": 3.421698021097902e-06,
      "loss": 0.77830744,
      "num_input_tokens_seen": 97456190,
      "step": 4505,
      "time_per_iteration": 2.5162622928619385
    },
    {
      "auxiliary_loss_clip": 0.01142935,
      "auxiliary_loss_mlp": 0.01049197,
      "balance_loss_clip": 1.04793525,
      "balance_loss_mlp": 1.03076684,
      "epoch": 0.2709153765218698,
      "flos": 17675232606720.0,
      "grad_norm": 2.5215357577653514,
      "language_loss": 0.74191433,
      "learning_rate": 3.42142406835758e-06,
      "loss": 0.76383567,
      "num_input_tokens_seen": 97474545,
      "step": 4506,
      "time_per_iteration": 2.454254388809204
    },
    {
      "auxiliary_loss_clip": 0.0112089,
      "auxiliary_loss_mlp": 0.01045417,
      "balance_loss_clip": 1.04666853,
      "balance_loss_mlp": 1.02709484,
      "epoch": 0.2709754997745378,
      "flos": 24456243801600.0,
      "grad_norm": 1.6943622793621906,
      "language_loss": 0.80773634,
      "learning_rate": 3.421150061716715e-06,
      "loss": 0.82939935,
      "num_input_tokens_seen": 97494520,
      "step": 4507,
      "time_per_iteration": 2.5464677810668945
    },
    {
      "auxiliary_loss_clip": 0.01029837,
      "auxiliary_loss_mlp": 0.01005558,
      "balance_loss_clip": 1.01846051,
      "balance_loss_mlp": 1.00353193,
      "epoch": 0.2710356230272058,
      "flos": 65210798206080.0,
      "grad_norm": 0.7353174210334501,
      "language_loss": 0.50890005,
      "learning_rate": 3.420876001185698e-06,
      "loss": 0.52925396,
      "num_input_tokens_seen": 97552455,
      "step": 4508,
      "time_per_iteration": 3.0610549449920654
    },
    {
      "auxiliary_loss_clip": 0.01073424,
      "auxiliary_loss_mlp": 0.01043065,
      "balance_loss_clip": 1.04831767,
      "balance_loss_mlp": 1.02598262,
      "epoch": 0.27109574627987376,
      "flos": 25484438615040.0,
      "grad_norm": 1.8757852316739143,
      "language_loss": 0.74867398,
      "learning_rate": 3.4206018867749197e-06,
      "loss": 0.76983887,
      "num_input_tokens_seen": 97572650,
      "step": 4509,
      "time_per_iteration": 2.6566059589385986
    },
    {
      "auxiliary_loss_clip": 0.01122941,
      "auxiliary_loss_mlp": 0.01042544,
      "balance_loss_clip": 1.04699922,
      "balance_loss_mlp": 1.026546,
      "epoch": 0.2711558695325417,
      "flos": 19682782715520.0,
      "grad_norm": 1.8402324854289138,
      "language_loss": 0.72025633,
      "learning_rate": 3.4203277184947757e-06,
      "loss": 0.74191129,
      "num_input_tokens_seen": 97591150,
      "step": 4510,
      "time_per_iteration": 2.505190849304199
    },
    {
      "auxiliary_loss_clip": 0.01127741,
      "auxiliary_loss_mlp": 0.01036116,
      "balance_loss_clip": 1.05075705,
      "balance_loss_mlp": 1.01954556,
      "epoch": 0.2712159927852097,
      "flos": 18587758648320.0,
      "grad_norm": 2.6187146391217326,
      "language_loss": 0.70010448,
      "learning_rate": 3.4200534963556627e-06,
      "loss": 0.72174305,
      "num_input_tokens_seen": 97607410,
      "step": 4511,
      "time_per_iteration": 2.4943504333496094
    },
    {
      "auxiliary_loss_clip": 0.01120763,
      "auxiliary_loss_mlp": 0.01047372,
      "balance_loss_clip": 1.05080712,
      "balance_loss_mlp": 1.02955031,
      "epoch": 0.27127611603787766,
      "flos": 25630235919360.0,
      "grad_norm": 2.313272511447112,
      "language_loss": 0.81138587,
      "learning_rate": 3.419779220367979e-06,
      "loss": 0.83306718,
      "num_input_tokens_seen": 97626870,
      "step": 4512,
      "time_per_iteration": 2.5854361057281494
    },
    {
      "auxiliary_loss_clip": 0.01138507,
      "auxiliary_loss_mlp": 0.01036222,
      "balance_loss_clip": 1.0501523,
      "balance_loss_mlp": 1.02109456,
      "epoch": 0.2713362392905456,
      "flos": 23148952312320.0,
      "grad_norm": 1.469728648635988,
      "language_loss": 0.80464095,
      "learning_rate": 3.419504890542124e-06,
      "loss": 0.82638824,
      "num_input_tokens_seen": 97646595,
      "step": 4513,
      "time_per_iteration": 3.8704352378845215
    },
    {
      "auxiliary_loss_clip": 0.01114196,
      "auxiliary_loss_mlp": 0.01042347,
      "balance_loss_clip": 1.04599738,
      "balance_loss_mlp": 1.02607524,
      "epoch": 0.2713963625432136,
      "flos": 18366045949440.0,
      "grad_norm": 1.7141189597958044,
      "language_loss": 0.88035959,
      "learning_rate": 3.4192305068885026e-06,
      "loss": 0.90192503,
      "num_input_tokens_seen": 97665485,
      "step": 4514,
      "time_per_iteration": 2.501976490020752
    },
    {
      "auxiliary_loss_clip": 0.01121125,
      "auxiliary_loss_mlp": 0.01045801,
      "balance_loss_clip": 1.05205917,
      "balance_loss_mlp": 1.02797914,
      "epoch": 0.27145648579588155,
      "flos": 22491751121280.0,
      "grad_norm": 1.6189721178620045,
      "language_loss": 0.92009163,
      "learning_rate": 3.418956069417517e-06,
      "loss": 0.9417609,
      "num_input_tokens_seen": 97683800,
      "step": 4515,
      "time_per_iteration": 2.5290305614471436
    },
    {
      "auxiliary_loss_clip": 0.01094956,
      "auxiliary_loss_mlp": 0.01054977,
      "balance_loss_clip": 1.04787481,
      "balance_loss_mlp": 1.03503323,
      "epoch": 0.2715166090485495,
      "flos": 19239177749760.0,
      "grad_norm": 2.0764666251076145,
      "language_loss": 0.73953557,
      "learning_rate": 3.4186815781395756e-06,
      "loss": 0.76103497,
      "num_input_tokens_seen": 97700505,
      "step": 4516,
      "time_per_iteration": 3.9668712615966797
    },
    {
      "auxiliary_loss_clip": 0.01130694,
      "auxiliary_loss_mlp": 0.01045234,
      "balance_loss_clip": 1.0497849,
      "balance_loss_mlp": 1.02756715,
      "epoch": 0.2715767323012175,
      "flos": 17709598944000.0,
      "grad_norm": 1.794788833674979,
      "language_loss": 0.7601161,
      "learning_rate": 3.4184070330650866e-06,
      "loss": 0.78187537,
      "num_input_tokens_seen": 97717410,
      "step": 4517,
      "time_per_iteration": 2.5013344287872314
    },
    {
      "auxiliary_loss_clip": 0.01095685,
      "auxiliary_loss_mlp": 0.01050651,
      "balance_loss_clip": 1.05052805,
      "balance_loss_mlp": 1.03206658,
      "epoch": 0.27163685555388545,
      "flos": 22382834106240.0,
      "grad_norm": 2.4518279809822907,
      "language_loss": 0.76961941,
      "learning_rate": 3.4181324342044607e-06,
      "loss": 0.79108274,
      "num_input_tokens_seen": 97734545,
      "step": 4518,
      "time_per_iteration": 3.9212284088134766
    },
    {
      "auxiliary_loss_clip": 0.01108027,
      "auxiliary_loss_mlp": 0.01042431,
      "balance_loss_clip": 1.04692054,
      "balance_loss_mlp": 1.02638555,
      "epoch": 0.2716969788065534,
      "flos": 22346708002560.0,
      "grad_norm": 1.570556671116304,
      "language_loss": 0.68091941,
      "learning_rate": 3.41785778156811e-06,
      "loss": 0.70242393,
      "num_input_tokens_seen": 97754000,
      "step": 4519,
      "time_per_iteration": 2.5338449478149414
    },
    {
      "auxiliary_loss_clip": 0.01128838,
      "auxiliary_loss_mlp": 0.01037237,
      "balance_loss_clip": 1.04917455,
      "balance_loss_mlp": 1.02079773,
      "epoch": 0.2717571020592214,
      "flos": 25228467319680.0,
      "grad_norm": 1.9654778561120358,
      "language_loss": 0.7564159,
      "learning_rate": 3.417583075166451e-06,
      "loss": 0.77807665,
      "num_input_tokens_seen": 97772080,
      "step": 4520,
      "time_per_iteration": 3.893636703491211
    },
    {
      "auxiliary_loss_clip": 0.01130925,
      "auxiliary_loss_mlp": 0.01050621,
      "balance_loss_clip": 1.0494076,
      "balance_loss_mlp": 1.03163075,
      "epoch": 0.2718172253118894,
      "flos": 20189769229440.0,
      "grad_norm": 2.2897556694385237,
      "language_loss": 0.76804525,
      "learning_rate": 3.4173083150099e-06,
      "loss": 0.78986073,
      "num_input_tokens_seen": 97789370,
      "step": 4521,
      "time_per_iteration": 2.485511541366577
    },
    {
      "auxiliary_loss_clip": 0.01114827,
      "auxiliary_loss_mlp": 0.01055841,
      "balance_loss_clip": 1.04888737,
      "balance_loss_mlp": 1.03774524,
      "epoch": 0.27187734856455736,
      "flos": 14319129260160.0,
      "grad_norm": 2.1644352252469066,
      "language_loss": 0.7536062,
      "learning_rate": 3.417033501108875e-06,
      "loss": 0.7753129,
      "num_input_tokens_seen": 97807385,
      "step": 4522,
      "time_per_iteration": 2.5214123725891113
    },
    {
      "auxiliary_loss_clip": 0.01148137,
      "auxiliary_loss_mlp": 0.01042453,
      "balance_loss_clip": 1.0544132,
      "balance_loss_mlp": 1.02498937,
      "epoch": 0.27193747181722533,
      "flos": 21107682311040.0,
      "grad_norm": 1.7594104124888108,
      "language_loss": 0.73183894,
      "learning_rate": 3.416758633473798e-06,
      "loss": 0.75374484,
      "num_input_tokens_seen": 97827930,
      "step": 4523,
      "time_per_iteration": 2.4845175743103027
    },
    {
      "auxiliary_loss_clip": 0.01115382,
      "auxiliary_loss_mlp": 0.01042952,
      "balance_loss_clip": 1.05120468,
      "balance_loss_mlp": 1.02508235,
      "epoch": 0.2719975950698933,
      "flos": 19682782715520.0,
      "grad_norm": 1.5732697287025224,
      "language_loss": 0.74388039,
      "learning_rate": 3.4164837121150915e-06,
      "loss": 0.76546371,
      "num_input_tokens_seen": 97847440,
      "step": 4524,
      "time_per_iteration": 2.533616065979004
    },
    {
      "auxiliary_loss_clip": 0.01144623,
      "auxiliary_loss_mlp": 0.01043077,
      "balance_loss_clip": 1.05292606,
      "balance_loss_mlp": 1.02628016,
      "epoch": 0.27205771832256126,
      "flos": 24754482426240.0,
      "grad_norm": 1.5974851896144944,
      "language_loss": 0.76450539,
      "learning_rate": 3.4162087370431803e-06,
      "loss": 0.78638244,
      "num_input_tokens_seen": 97867620,
      "step": 4525,
      "time_per_iteration": 2.4912586212158203
    },
    {
      "auxiliary_loss_clip": 0.01128959,
      "auxiliary_loss_mlp": 0.01056791,
      "balance_loss_clip": 1.04988551,
      "balance_loss_mlp": 1.03914773,
      "epoch": 0.2721178415752292,
      "flos": 21755581879680.0,
      "grad_norm": 1.8085598090627317,
      "language_loss": 0.81979275,
      "learning_rate": 3.4159337082684926e-06,
      "loss": 0.84165019,
      "num_input_tokens_seen": 97884345,
      "step": 4526,
      "time_per_iteration": 2.5021679401397705
    },
    {
      "auxiliary_loss_clip": 0.01145838,
      "auxiliary_loss_mlp": 0.01044938,
      "balance_loss_clip": 1.04952478,
      "balance_loss_mlp": 1.02648473,
      "epoch": 0.2721779648278972,
      "flos": 12676826597760.0,
      "grad_norm": 1.9456468450755564,
      "language_loss": 0.77229297,
      "learning_rate": 3.4156586258014566e-06,
      "loss": 0.79420072,
      "num_input_tokens_seen": 97901500,
      "step": 4527,
      "time_per_iteration": 2.4227097034454346
    },
    {
      "auxiliary_loss_clip": 0.01103454,
      "auxiliary_loss_mlp": 0.00808735,
      "balance_loss_clip": 1.04762781,
      "balance_loss_mlp": 1.03469324,
      "epoch": 0.27223808808056515,
      "flos": 16253206099200.0,
      "grad_norm": 2.091462288367161,
      "language_loss": 0.81780398,
      "learning_rate": 3.415383489652503e-06,
      "loss": 0.83692586,
      "num_input_tokens_seen": 97917800,
      "step": 4528,
      "time_per_iteration": 2.5451056957244873
    },
    {
      "auxiliary_loss_clip": 0.01110182,
      "auxiliary_loss_mlp": 0.0104906,
      "balance_loss_clip": 1.04880047,
      "balance_loss_mlp": 1.0315119,
      "epoch": 0.2722982113332331,
      "flos": 27745805203200.0,
      "grad_norm": 1.6882648698834963,
      "language_loss": 0.77496833,
      "learning_rate": 3.4151082998320666e-06,
      "loss": 0.79656076,
      "num_input_tokens_seen": 97937225,
      "step": 4529,
      "time_per_iteration": 2.5883820056915283
    },
    {
      "auxiliary_loss_clip": 0.01116165,
      "auxiliary_loss_mlp": 0.01049023,
      "balance_loss_clip": 1.05065322,
      "balance_loss_mlp": 1.03229761,
      "epoch": 0.2723583345859011,
      "flos": 21726243446400.0,
      "grad_norm": 2.5601133053175107,
      "language_loss": 0.82266498,
      "learning_rate": 3.4148330563505805e-06,
      "loss": 0.84431684,
      "num_input_tokens_seen": 97956845,
      "step": 4530,
      "time_per_iteration": 2.5457839965820312
    },
    {
      "auxiliary_loss_clip": 0.01130935,
      "auxiliary_loss_mlp": 0.0104324,
      "balance_loss_clip": 1.05029082,
      "balance_loss_mlp": 1.02558541,
      "epoch": 0.27241845783856905,
      "flos": 17347260499200.0,
      "grad_norm": 2.1270422739159294,
      "language_loss": 0.9153102,
      "learning_rate": 3.4145577592184838e-06,
      "loss": 0.93705189,
      "num_input_tokens_seen": 97972465,
      "step": 4531,
      "time_per_iteration": 2.4806768894195557
    },
    {
      "auxiliary_loss_clip": 0.01135778,
      "auxiliary_loss_mlp": 0.01053113,
      "balance_loss_clip": 1.05317128,
      "balance_loss_mlp": 1.03550553,
      "epoch": 0.272478581091237,
      "flos": 24754302858240.0,
      "grad_norm": 2.013553690636943,
      "language_loss": 0.76224649,
      "learning_rate": 3.4142824084462155e-06,
      "loss": 0.78413534,
      "num_input_tokens_seen": 97990770,
      "step": 4532,
      "time_per_iteration": 2.524092197418213
    },
    {
      "auxiliary_loss_clip": 0.01105198,
      "auxiliary_loss_mlp": 0.01038354,
      "balance_loss_clip": 1.04765201,
      "balance_loss_mlp": 1.02158141,
      "epoch": 0.272538704343905,
      "flos": 17890624512000.0,
      "grad_norm": 2.811896406683421,
      "language_loss": 0.88966048,
      "learning_rate": 3.4140070040442162e-06,
      "loss": 0.91109598,
      "num_input_tokens_seen": 98005775,
      "step": 4533,
      "time_per_iteration": 2.579923391342163
    },
    {
      "auxiliary_loss_clip": 0.01115748,
      "auxiliary_loss_mlp": 0.01037686,
      "balance_loss_clip": 1.04822183,
      "balance_loss_mlp": 1.02084208,
      "epoch": 0.272598827596573,
      "flos": 22932016122240.0,
      "grad_norm": 2.2947188796784252,
      "language_loss": 0.71111238,
      "learning_rate": 3.413731546022929e-06,
      "loss": 0.7326467,
      "num_input_tokens_seen": 98025750,
      "step": 4534,
      "time_per_iteration": 2.5526938438415527
    },
    {
      "auxiliary_loss_clip": 0.01114558,
      "auxiliary_loss_mlp": 0.01042451,
      "balance_loss_clip": 1.04923487,
      "balance_loss_mlp": 1.02386642,
      "epoch": 0.27265895084924097,
      "flos": 24238409771520.0,
      "grad_norm": 2.304384910111777,
      "language_loss": 0.91507041,
      "learning_rate": 3.4134560343928005e-06,
      "loss": 0.9366405,
      "num_input_tokens_seen": 98044955,
      "step": 4535,
      "time_per_iteration": 2.564807176589966
    },
    {
      "auxiliary_loss_clip": 0.01124425,
      "auxiliary_loss_mlp": 0.01044843,
      "balance_loss_clip": 1.0526731,
      "balance_loss_mlp": 1.0268538,
      "epoch": 0.27271907410190893,
      "flos": 27013155494400.0,
      "grad_norm": 1.8528563976854204,
      "language_loss": 0.72931945,
      "learning_rate": 3.4131804691642778e-06,
      "loss": 0.75101215,
      "num_input_tokens_seen": 98065860,
      "step": 4536,
      "time_per_iteration": 2.5818119049072266
    },
    {
      "auxiliary_loss_clip": 0.01130097,
      "auxiliary_loss_mlp": 0.01042362,
      "balance_loss_clip": 1.04721761,
      "balance_loss_mlp": 1.02488577,
      "epoch": 0.2727791973545769,
      "flos": 34452588942720.0,
      "grad_norm": 1.9944316080066942,
      "language_loss": 0.71762562,
      "learning_rate": 3.41290485034781e-06,
      "loss": 0.7393502,
      "num_input_tokens_seen": 98085450,
      "step": 4537,
      "time_per_iteration": 2.632749080657959
    },
    {
      "auxiliary_loss_clip": 0.01116822,
      "auxiliary_loss_mlp": 0.0104377,
      "balance_loss_clip": 1.05132604,
      "balance_loss_mlp": 1.02549505,
      "epoch": 0.27283932060724486,
      "flos": 15041723160960.0,
      "grad_norm": 2.1545889174517243,
      "language_loss": 0.78586113,
      "learning_rate": 3.4126291779538485e-06,
      "loss": 0.80746704,
      "num_input_tokens_seen": 98099115,
      "step": 4538,
      "time_per_iteration": 2.499117851257324
    },
    {
      "auxiliary_loss_clip": 0.01128334,
      "auxiliary_loss_mlp": 0.01052535,
      "balance_loss_clip": 1.05050778,
      "balance_loss_mlp": 1.03545225,
      "epoch": 0.2728994438599128,
      "flos": 21652411040640.0,
      "grad_norm": 1.5179070909798678,
      "language_loss": 0.90235782,
      "learning_rate": 3.412353451992847e-06,
      "loss": 0.9241665,
      "num_input_tokens_seen": 98118415,
      "step": 4539,
      "time_per_iteration": 2.5019781589508057
    },
    {
      "auxiliary_loss_clip": 0.01120072,
      "auxiliary_loss_mlp": 0.01045068,
      "balance_loss_clip": 1.04940486,
      "balance_loss_mlp": 1.02647185,
      "epoch": 0.2729595671125808,
      "flos": 17488424949120.0,
      "grad_norm": 2.156731876388365,
      "language_loss": 0.8804332,
      "learning_rate": 3.4120776724752607e-06,
      "loss": 0.90208459,
      "num_input_tokens_seen": 98136300,
      "step": 4540,
      "time_per_iteration": 2.5047340393066406
    },
    {
      "auxiliary_loss_clip": 0.01128716,
      "auxiliary_loss_mlp": 0.00801765,
      "balance_loss_clip": 1.04611623,
      "balance_loss_mlp": 1.02379584,
      "epoch": 0.27301969036524876,
      "flos": 19318145800320.0,
      "grad_norm": 1.9747411159870862,
      "language_loss": 0.82196605,
      "learning_rate": 3.4118018394115476e-06,
      "loss": 0.84127086,
      "num_input_tokens_seen": 98154580,
      "step": 4541,
      "time_per_iteration": 2.499253034591675
    },
    {
      "auxiliary_loss_clip": 0.01119741,
      "auxiliary_loss_mlp": 0.01046854,
      "balance_loss_clip": 1.05043793,
      "balance_loss_mlp": 1.02998614,
      "epoch": 0.2730798136179167,
      "flos": 21065666376960.0,
      "grad_norm": 1.7268216415905464,
      "language_loss": 0.80162841,
      "learning_rate": 3.4115259528121678e-06,
      "loss": 0.82329428,
      "num_input_tokens_seen": 98173115,
      "step": 4542,
      "time_per_iteration": 2.520578145980835
    },
    {
      "auxiliary_loss_clip": 0.01124253,
      "auxiliary_loss_mlp": 0.01044221,
      "balance_loss_clip": 1.05272257,
      "balance_loss_mlp": 1.02685237,
      "epoch": 0.2731399368705847,
      "flos": 19171737964800.0,
      "grad_norm": 2.752868616583742,
      "language_loss": 0.89792156,
      "learning_rate": 3.411250012687582e-06,
      "loss": 0.91960633,
      "num_input_tokens_seen": 98190260,
      "step": 4543,
      "time_per_iteration": 2.5313565731048584
    },
    {
      "auxiliary_loss_clip": 0.01112592,
      "auxiliary_loss_mlp": 0.0080349,
      "balance_loss_clip": 1.04763103,
      "balance_loss_mlp": 1.02365911,
      "epoch": 0.27320006012325265,
      "flos": 18290130554880.0,
      "grad_norm": 1.9448178398491165,
      "language_loss": 0.63466549,
      "learning_rate": 3.410974019048255e-06,
      "loss": 0.65382636,
      "num_input_tokens_seen": 98207115,
      "step": 4544,
      "time_per_iteration": 2.500349998474121
    },
    {
      "auxiliary_loss_clip": 0.01114489,
      "auxiliary_loss_mlp": 0.01050742,
      "balance_loss_clip": 1.05041945,
      "balance_loss_mlp": 1.03249109,
      "epoch": 0.2732601833759206,
      "flos": 34860929731200.0,
      "grad_norm": 2.664600784336412,
      "language_loss": 0.69951761,
      "learning_rate": 3.410697971904651e-06,
      "loss": 0.72116995,
      "num_input_tokens_seen": 98230610,
      "step": 4545,
      "time_per_iteration": 2.6435632705688477
    },
    {
      "auxiliary_loss_clip": 0.01037535,
      "auxiliary_loss_mlp": 0.0102222,
      "balance_loss_clip": 1.02492678,
      "balance_loss_mlp": 1.01981187,
      "epoch": 0.2733203066285886,
      "flos": 53910824762880.0,
      "grad_norm": 0.7178554000585645,
      "language_loss": 0.61609846,
      "learning_rate": 3.4104218712672383e-06,
      "loss": 0.63669598,
      "num_input_tokens_seen": 98293585,
      "step": 4546,
      "time_per_iteration": 3.147480010986328
    },
    {
      "auxiliary_loss_clip": 0.01059046,
      "auxiliary_loss_mlp": 0.01049629,
      "balance_loss_clip": 1.0512315,
      "balance_loss_mlp": 1.03149772,
      "epoch": 0.2733804298812566,
      "flos": 20660378244480.0,
      "grad_norm": 1.878604642393384,
      "language_loss": 0.65212297,
      "learning_rate": 3.410145717146488e-06,
      "loss": 0.67320967,
      "num_input_tokens_seen": 98311680,
      "step": 4547,
      "time_per_iteration": 2.6773836612701416
    },
    {
      "auxiliary_loss_clip": 0.01114092,
      "auxiliary_loss_mlp": 0.00798582,
      "balance_loss_clip": 1.04781163,
      "balance_loss_mlp": 1.02120054,
      "epoch": 0.27344055313392457,
      "flos": 25884339707520.0,
      "grad_norm": 1.9953491448887073,
      "language_loss": 0.77718008,
      "learning_rate": 3.4098695095528694e-06,
      "loss": 0.79630679,
      "num_input_tokens_seen": 98330770,
      "step": 4548,
      "time_per_iteration": 2.5650501251220703
    },
    {
      "auxiliary_loss_clip": 0.01118547,
      "auxiliary_loss_mlp": 0.01043897,
      "balance_loss_clip": 1.05588746,
      "balance_loss_mlp": 1.02886498,
      "epoch": 0.27350067638659253,
      "flos": 22929753565440.0,
      "grad_norm": 1.9571312809506023,
      "language_loss": 0.82959986,
      "learning_rate": 3.4095932484968585e-06,
      "loss": 0.8512243,
      "num_input_tokens_seen": 98349860,
      "step": 4549,
      "time_per_iteration": 2.5703437328338623
    },
    {
      "auxiliary_loss_clip": 0.0113131,
      "auxiliary_loss_mlp": 0.01046858,
      "balance_loss_clip": 1.04935944,
      "balance_loss_mlp": 1.02826118,
      "epoch": 0.2735607996392605,
      "flos": 16574821499520.0,
      "grad_norm": 1.9110335880786171,
      "language_loss": 0.71209204,
      "learning_rate": 3.4093169339889305e-06,
      "loss": 0.73387367,
      "num_input_tokens_seen": 98367040,
      "step": 4550,
      "time_per_iteration": 2.47241473197937
    },
    {
      "auxiliary_loss_clip": 0.01101262,
      "auxiliary_loss_mlp": 0.01040311,
      "balance_loss_clip": 1.05056798,
      "balance_loss_mlp": 1.02452779,
      "epoch": 0.27362092289192846,
      "flos": 19645291895040.0,
      "grad_norm": 3.5496042573401483,
      "language_loss": 0.7877121,
      "learning_rate": 3.409040566039563e-06,
      "loss": 0.80912781,
      "num_input_tokens_seen": 98384010,
      "step": 4551,
      "time_per_iteration": 3.9322588443756104
    },
    {
      "auxiliary_loss_clip": 0.0110168,
      "auxiliary_loss_mlp": 0.01061336,
      "balance_loss_clip": 1.04672778,
      "balance_loss_mlp": 1.04309726,
      "epoch": 0.27368104614459643,
      "flos": 17639142416640.0,
      "grad_norm": 2.498622408224486,
      "language_loss": 0.70441055,
      "learning_rate": 3.4087641446592362e-06,
      "loss": 0.72604072,
      "num_input_tokens_seen": 98399625,
      "step": 4552,
      "time_per_iteration": 2.5511727333068848
    },
    {
      "auxiliary_loss_clip": 0.01118855,
      "auxiliary_loss_mlp": 0.01039164,
      "balance_loss_clip": 1.05241859,
      "balance_loss_mlp": 1.0224154,
      "epoch": 0.2737411693972644,
      "flos": 21580015178880.0,
      "grad_norm": 1.9068315580108788,
      "language_loss": 0.71525764,
      "learning_rate": 3.408487669858431e-06,
      "loss": 0.73683786,
      "num_input_tokens_seen": 98417310,
      "step": 4553,
      "time_per_iteration": 2.5350074768066406
    },
    {
      "auxiliary_loss_clip": 0.01130031,
      "auxiliary_loss_mlp": 0.01040073,
      "balance_loss_clip": 1.05045605,
      "balance_loss_mlp": 1.02257335,
      "epoch": 0.27380129264993236,
      "flos": 25484043565440.0,
      "grad_norm": 1.5614163550254527,
      "language_loss": 0.59273994,
      "learning_rate": 3.4082111416476337e-06,
      "loss": 0.61444098,
      "num_input_tokens_seen": 98438670,
      "step": 4554,
      "time_per_iteration": 3.9525089263916016
    },
    {
      "auxiliary_loss_clip": 0.01124878,
      "auxiliary_loss_mlp": 0.01041437,
      "balance_loss_clip": 1.05314243,
      "balance_loss_mlp": 1.02425921,
      "epoch": 0.2738614159026003,
      "flos": 18661196004480.0,
      "grad_norm": 1.6774875614747509,
      "language_loss": 0.74197125,
      "learning_rate": 3.4079345600373275e-06,
      "loss": 0.76363432,
      "num_input_tokens_seen": 98456060,
      "step": 4555,
      "time_per_iteration": 2.516784906387329
    },
    {
      "auxiliary_loss_clip": 0.01135253,
      "auxiliary_loss_mlp": 0.0104142,
      "balance_loss_clip": 1.05415702,
      "balance_loss_mlp": 1.02432489,
      "epoch": 0.2739215391552683,
      "flos": 23477139901440.0,
      "grad_norm": 1.810006062912045,
      "language_loss": 0.78252065,
      "learning_rate": 3.407657925038002e-06,
      "loss": 0.80428737,
      "num_input_tokens_seen": 98473765,
      "step": 4556,
      "time_per_iteration": 2.525484085083008
    },
    {
      "auxiliary_loss_clip": 0.01141809,
      "auxiliary_loss_mlp": 0.01050109,
      "balance_loss_clip": 1.0516057,
      "balance_loss_mlp": 1.03157234,
      "epoch": 0.27398166240793626,
      "flos": 17128636369920.0,
      "grad_norm": 1.8441261448157684,
      "language_loss": 0.81953609,
      "learning_rate": 3.4073812366601473e-06,
      "loss": 0.84145528,
      "num_input_tokens_seen": 98490590,
      "step": 4557,
      "time_per_iteration": 3.840811252593994
    },
    {
      "auxiliary_loss_clip": 0.01089759,
      "auxiliary_loss_mlp": 0.01047343,
      "balance_loss_clip": 1.04805732,
      "balance_loss_mlp": 1.03030801,
      "epoch": 0.2740417856606042,
      "flos": 23404744039680.0,
      "grad_norm": 1.804654663251744,
      "language_loss": 0.72688043,
      "learning_rate": 3.4071044949142547e-06,
      "loss": 0.74825156,
      "num_input_tokens_seen": 98510590,
      "step": 4558,
      "time_per_iteration": 2.583707332611084
    },
    {
      "auxiliary_loss_clip": 0.01119497,
      "auxiliary_loss_mlp": 0.01050064,
      "balance_loss_clip": 1.05051875,
      "balance_loss_mlp": 1.03308892,
      "epoch": 0.2741019089132722,
      "flos": 12780428400000.0,
      "grad_norm": 2.2034832489796012,
      "language_loss": 0.67164522,
      "learning_rate": 3.406827699810819e-06,
      "loss": 0.69334078,
      "num_input_tokens_seen": 98527875,
      "step": 4559,
      "time_per_iteration": 3.9095847606658936
    },
    {
      "auxiliary_loss_clip": 0.01119445,
      "auxiliary_loss_mlp": 0.0105611,
      "balance_loss_clip": 1.05226851,
      "balance_loss_mlp": 1.03836012,
      "epoch": 0.27416203216594015,
      "flos": 20631542601600.0,
      "grad_norm": 1.6938784733853083,
      "language_loss": 0.72141016,
      "learning_rate": 3.4065508513603353e-06,
      "loss": 0.74316573,
      "num_input_tokens_seen": 98547575,
      "step": 4560,
      "time_per_iteration": 2.5518782138824463
    },
    {
      "auxiliary_loss_clip": 0.01120571,
      "auxiliary_loss_mlp": 0.01047266,
      "balance_loss_clip": 1.05052269,
      "balance_loss_mlp": 1.03018284,
      "epoch": 0.27422215541860817,
      "flos": 26541576812160.0,
      "grad_norm": 1.963117285069061,
      "language_loss": 0.8123908,
      "learning_rate": 3.406273949573303e-06,
      "loss": 0.83406925,
      "num_input_tokens_seen": 98566290,
      "step": 4561,
      "time_per_iteration": 2.5771610736846924
    },
    {
      "auxiliary_loss_clip": 0.01146238,
      "auxiliary_loss_mlp": 0.01040169,
      "balance_loss_clip": 1.05289841,
      "balance_loss_mlp": 1.02364707,
      "epoch": 0.27428227867127614,
      "flos": 23331163029120.0,
      "grad_norm": 1.5346026029762971,
      "language_loss": 0.75317484,
      "learning_rate": 3.4059969944602214e-06,
      "loss": 0.77503896,
      "num_input_tokens_seen": 98586255,
      "step": 4562,
      "time_per_iteration": 2.490492582321167
    },
    {
      "auxiliary_loss_clip": 0.01144916,
      "auxiliary_loss_mlp": 0.01035383,
      "balance_loss_clip": 1.05402851,
      "balance_loss_mlp": 1.01917028,
      "epoch": 0.2743424019239441,
      "flos": 23035115134080.0,
      "grad_norm": 1.6627032782671975,
      "language_loss": 0.74494636,
      "learning_rate": 3.4057199860315928e-06,
      "loss": 0.76674938,
      "num_input_tokens_seen": 98606030,
      "step": 4563,
      "time_per_iteration": 2.4725401401519775
    },
    {
      "auxiliary_loss_clip": 0.01114398,
      "auxiliary_loss_mlp": 0.01045745,
      "balance_loss_clip": 1.0489639,
      "balance_loss_mlp": 1.02767324,
      "epoch": 0.27440252517661207,
      "flos": 21981101420160.0,
      "grad_norm": 1.7547496909436608,
      "language_loss": 0.62307405,
      "learning_rate": 3.4054429242979213e-06,
      "loss": 0.64467549,
      "num_input_tokens_seen": 98625225,
      "step": 4564,
      "time_per_iteration": 2.568338394165039
    },
    {
      "auxiliary_loss_clip": 0.01119935,
      "auxiliary_loss_mlp": 0.0103954,
      "balance_loss_clip": 1.0504005,
      "balance_loss_mlp": 1.02190936,
      "epoch": 0.27446264842928003,
      "flos": 40187451502080.0,
      "grad_norm": 1.8354037759692108,
      "language_loss": 0.78316224,
      "learning_rate": 3.4051658092697135e-06,
      "loss": 0.804757,
      "num_input_tokens_seen": 98649470,
      "step": 4565,
      "time_per_iteration": 2.7361390590667725
    },
    {
      "auxiliary_loss_clip": 0.01088838,
      "auxiliary_loss_mlp": 0.01043462,
      "balance_loss_clip": 1.04926825,
      "balance_loss_mlp": 1.02648675,
      "epoch": 0.274522771681948,
      "flos": 13479681438720.0,
      "grad_norm": 1.8532940803934566,
      "language_loss": 0.69197786,
      "learning_rate": 3.404888640957477e-06,
      "loss": 0.71330082,
      "num_input_tokens_seen": 98666915,
      "step": 4566,
      "time_per_iteration": 2.5790462493896484
    },
    {
      "auxiliary_loss_clip": 0.01134066,
      "auxiliary_loss_mlp": 0.01048642,
      "balance_loss_clip": 1.05532503,
      "balance_loss_mlp": 1.03276348,
      "epoch": 0.27458289493461596,
      "flos": 28622133313920.0,
      "grad_norm": 1.6155001708637973,
      "language_loss": 0.61191463,
      "learning_rate": 3.404611419371723e-06,
      "loss": 0.63374174,
      "num_input_tokens_seen": 98688240,
      "step": 4567,
      "time_per_iteration": 2.569058895111084
    },
    {
      "auxiliary_loss_clip": 0.01128143,
      "auxiliary_loss_mlp": 0.01051064,
      "balance_loss_clip": 1.05105937,
      "balance_loss_mlp": 1.03238344,
      "epoch": 0.2746430181872839,
      "flos": 20119815492480.0,
      "grad_norm": 1.840314437980144,
      "language_loss": 0.82663429,
      "learning_rate": 3.4043341445229627e-06,
      "loss": 0.8484264,
      "num_input_tokens_seen": 98708245,
      "step": 4568,
      "time_per_iteration": 2.522486686706543
    },
    {
      "auxiliary_loss_clip": 0.01133863,
      "auxiliary_loss_mlp": 0.01038972,
      "balance_loss_clip": 1.05766308,
      "balance_loss_mlp": 1.02188957,
      "epoch": 0.2747031414399519,
      "flos": 20193468330240.0,
      "grad_norm": 1.909086822001717,
      "language_loss": 0.68361425,
      "learning_rate": 3.4040568164217117e-06,
      "loss": 0.70534259,
      "num_input_tokens_seen": 98724575,
      "step": 4569,
      "time_per_iteration": 2.4911868572235107
    },
    {
      "auxiliary_loss_clip": 0.01110755,
      "auxiliary_loss_mlp": 0.0104203,
      "balance_loss_clip": 1.04618526,
      "balance_loss_mlp": 1.02466154,
      "epoch": 0.27476326469261986,
      "flos": 13516346246400.0,
      "grad_norm": 2.0712356786557726,
      "language_loss": 0.70915002,
      "learning_rate": 3.4037794350784848e-06,
      "loss": 0.7306779,
      "num_input_tokens_seen": 98740700,
      "step": 4570,
      "time_per_iteration": 2.5152506828308105
    },
    {
      "auxiliary_loss_clip": 0.0102923,
      "auxiliary_loss_mlp": 0.01012619,
      "balance_loss_clip": 1.02956676,
      "balance_loss_mlp": 1.01053262,
      "epoch": 0.2748233879452878,
      "flos": 65937127121280.0,
      "grad_norm": 0.729525008789145,
      "language_loss": 0.55786008,
      "learning_rate": 3.4035020005038014e-06,
      "loss": 0.57827854,
      "num_input_tokens_seen": 98803030,
      "step": 4571,
      "time_per_iteration": 3.283519983291626
    },
    {
      "auxiliary_loss_clip": 0.01089561,
      "auxiliary_loss_mlp": 0.01050093,
      "balance_loss_clip": 1.04630232,
      "balance_loss_mlp": 1.03330863,
      "epoch": 0.2748835111979558,
      "flos": 17384212615680.0,
      "grad_norm": 2.2123016574130903,
      "language_loss": 0.78179109,
      "learning_rate": 3.4032245127081812e-06,
      "loss": 0.80318761,
      "num_input_tokens_seen": 98820505,
      "step": 4572,
      "time_per_iteration": 2.5635368824005127
    },
    {
      "auxiliary_loss_clip": 0.01138101,
      "auxiliary_loss_mlp": 0.01035728,
      "balance_loss_clip": 1.05242419,
      "balance_loss_mlp": 1.02102923,
      "epoch": 0.27494363445062375,
      "flos": 23587565287680.0,
      "grad_norm": 1.5534112849435615,
      "language_loss": 0.8169744,
      "learning_rate": 3.402946971702147e-06,
      "loss": 0.83871275,
      "num_input_tokens_seen": 98842150,
      "step": 4573,
      "time_per_iteration": 2.495182514190674
    },
    {
      "auxiliary_loss_clip": 0.01128378,
      "auxiliary_loss_mlp": 0.01036518,
      "balance_loss_clip": 1.05076098,
      "balance_loss_mlp": 1.02046061,
      "epoch": 0.2750037577032918,
      "flos": 17164582905600.0,
      "grad_norm": 1.7352847900596808,
      "language_loss": 0.79025495,
      "learning_rate": 3.402669377496223e-06,
      "loss": 0.81190389,
      "num_input_tokens_seen": 98861050,
      "step": 4574,
      "time_per_iteration": 2.526484727859497
    },
    {
      "auxiliary_loss_clip": 0.01096269,
      "auxiliary_loss_mlp": 0.01046291,
      "balance_loss_clip": 1.05016243,
      "balance_loss_mlp": 1.03067446,
      "epoch": 0.27506388095595974,
      "flos": 24491903028480.0,
      "grad_norm": 1.9179857475411626,
      "language_loss": 0.7433297,
      "learning_rate": 3.402391730100936e-06,
      "loss": 0.76475531,
      "num_input_tokens_seen": 98879695,
      "step": 4575,
      "time_per_iteration": 2.6148297786712646
    },
    {
      "auxiliary_loss_clip": 0.01116613,
      "auxiliary_loss_mlp": 0.01039832,
      "balance_loss_clip": 1.04837322,
      "balance_loss_mlp": 1.02459693,
      "epoch": 0.2751240042086277,
      "flos": 38764706722560.0,
      "grad_norm": 1.9037872313967157,
      "language_loss": 0.71472859,
      "learning_rate": 3.402114029526814e-06,
      "loss": 0.73629296,
      "num_input_tokens_seen": 98902035,
      "step": 4576,
      "time_per_iteration": 2.668282985687256
    },
    {
      "auxiliary_loss_clip": 0.01093512,
      "auxiliary_loss_mlp": 0.0079766,
      "balance_loss_clip": 1.04820681,
      "balance_loss_mlp": 1.01608515,
      "epoch": 0.27518412746129567,
      "flos": 26907039740160.0,
      "grad_norm": 2.3717910777771403,
      "language_loss": 0.7332589,
      "learning_rate": 3.4018362757843866e-06,
      "loss": 0.75217062,
      "num_input_tokens_seen": 98921835,
      "step": 4577,
      "time_per_iteration": 2.609741687774658
    },
    {
      "auxiliary_loss_clip": 0.01119014,
      "auxiliary_loss_mlp": 0.01035019,
      "balance_loss_clip": 1.04985094,
      "balance_loss_mlp": 1.01841295,
      "epoch": 0.27524425071396363,
      "flos": 24900531125760.0,
      "grad_norm": 1.7459002502087733,
      "language_loss": 0.75670123,
      "learning_rate": 3.401558468884188e-06,
      "loss": 0.77824152,
      "num_input_tokens_seen": 98939610,
      "step": 4578,
      "time_per_iteration": 2.6070759296417236
    },
    {
      "auxiliary_loss_clip": 0.01112081,
      "auxiliary_loss_mlp": 0.01049131,
      "balance_loss_clip": 1.05056262,
      "balance_loss_mlp": 1.03012896,
      "epoch": 0.2753043739666316,
      "flos": 26288047641600.0,
      "grad_norm": 1.4042292350543357,
      "language_loss": 0.66100872,
      "learning_rate": 3.4012806088367516e-06,
      "loss": 0.68262082,
      "num_input_tokens_seen": 98962250,
      "step": 4579,
      "time_per_iteration": 2.605952262878418
    },
    {
      "auxiliary_loss_clip": 0.01108784,
      "auxiliary_loss_mlp": 0.01053297,
      "balance_loss_clip": 1.04921377,
      "balance_loss_mlp": 1.03489113,
      "epoch": 0.27536449721929956,
      "flos": 24206772867840.0,
      "grad_norm": 1.7835397570030327,
      "language_loss": 0.79870218,
      "learning_rate": 3.4010026956526137e-06,
      "loss": 0.82032299,
      "num_input_tokens_seen": 98981845,
      "step": 4580,
      "time_per_iteration": 2.5854430198669434
    },
    {
      "auxiliary_loss_clip": 0.01129091,
      "auxiliary_loss_mlp": 0.01045212,
      "balance_loss_clip": 1.05021822,
      "balance_loss_mlp": 1.02716339,
      "epoch": 0.27542462047196753,
      "flos": 19537272720000.0,
      "grad_norm": 1.4597107140874,
      "language_loss": 0.67508006,
      "learning_rate": 3.4007247293423137e-06,
      "loss": 0.69682312,
      "num_input_tokens_seen": 99001855,
      "step": 4581,
      "time_per_iteration": 2.51517391204834
    },
    {
      "auxiliary_loss_clip": 0.01137224,
      "auxiliary_loss_mlp": 0.01043672,
      "balance_loss_clip": 1.07608891,
      "balance_loss_mlp": 1.02787638,
      "epoch": 0.2754847437246355,
      "flos": 14319165173760.0,
      "grad_norm": 1.6345952899159173,
      "language_loss": 0.78349888,
      "learning_rate": 3.400446709916392e-06,
      "loss": 0.80530781,
      "num_input_tokens_seen": 99019880,
      "step": 4582,
      "time_per_iteration": 2.5103378295898438
    },
    {
      "auxiliary_loss_clip": 0.0109258,
      "auxiliary_loss_mlp": 0.01041821,
      "balance_loss_clip": 1.04995096,
      "balance_loss_mlp": 1.02639556,
      "epoch": 0.27554486697730346,
      "flos": 18838773866880.0,
      "grad_norm": 1.689995703652388,
      "language_loss": 0.84217536,
      "learning_rate": 3.4001686373853895e-06,
      "loss": 0.86351943,
      "num_input_tokens_seen": 99037570,
      "step": 4583,
      "time_per_iteration": 2.5703258514404297
    },
    {
      "auxiliary_loss_clip": 0.01129991,
      "auxiliary_loss_mlp": 0.01039308,
      "balance_loss_clip": 1.04892838,
      "balance_loss_mlp": 1.02340567,
      "epoch": 0.2756049902299714,
      "flos": 22382295402240.0,
      "grad_norm": 1.6670971472417908,
      "language_loss": 0.67117977,
      "learning_rate": 3.3998905117598528e-06,
      "loss": 0.69287276,
      "num_input_tokens_seen": 99056875,
      "step": 4584,
      "time_per_iteration": 2.5056724548339844
    },
    {
      "auxiliary_loss_clip": 0.01075734,
      "auxiliary_loss_mlp": 0.01058966,
      "balance_loss_clip": 1.04437709,
      "balance_loss_mlp": 1.04119158,
      "epoch": 0.2756651134826394,
      "flos": 19573901614080.0,
      "grad_norm": 1.7901573065808485,
      "language_loss": 0.76945627,
      "learning_rate": 3.399612333050327e-06,
      "loss": 0.79080331,
      "num_input_tokens_seen": 99074685,
      "step": 4585,
      "time_per_iteration": 2.571514844894409
    },
    {
      "auxiliary_loss_clip": 0.01132056,
      "auxiliary_loss_mlp": 0.00795758,
      "balance_loss_clip": 1.05190325,
      "balance_loss_mlp": 1.01412296,
      "epoch": 0.27572523673530736,
      "flos": 23586559706880.0,
      "grad_norm": 2.502688797944457,
      "language_loss": 0.71783203,
      "learning_rate": 3.399334101267362e-06,
      "loss": 0.73711014,
      "num_input_tokens_seen": 99095300,
      "step": 4586,
      "time_per_iteration": 2.55117130279541
    },
    {
      "auxiliary_loss_clip": 0.01116307,
      "auxiliary_loss_mlp": 0.0103393,
      "balance_loss_clip": 1.04923749,
      "balance_loss_mlp": 1.01833105,
      "epoch": 0.2757853599879754,
      "flos": 22820118278400.0,
      "grad_norm": 1.5021314304396445,
      "language_loss": 0.80737627,
      "learning_rate": 3.3990558164215073e-06,
      "loss": 0.82887852,
      "num_input_tokens_seen": 99115965,
      "step": 4587,
      "time_per_iteration": 2.553982973098755
    },
    {
      "auxiliary_loss_clip": 0.01124902,
      "auxiliary_loss_mlp": 0.01037289,
      "balance_loss_clip": 1.04641604,
      "balance_loss_mlp": 1.0215714,
      "epoch": 0.27584548324064334,
      "flos": 18551704371840.0,
      "grad_norm": 1.8384515501312855,
      "language_loss": 0.82731664,
      "learning_rate": 3.398777478523316e-06,
      "loss": 0.84893858,
      "num_input_tokens_seen": 99134265,
      "step": 4588,
      "time_per_iteration": 2.4975743293762207
    },
    {
      "auxiliary_loss_clip": 0.0110073,
      "auxiliary_loss_mlp": 0.01038644,
      "balance_loss_clip": 1.04469192,
      "balance_loss_mlp": 1.0222528,
      "epoch": 0.2759056064933113,
      "flos": 23769883745280.0,
      "grad_norm": 1.4391082804463773,
      "language_loss": 0.75537062,
      "learning_rate": 3.398499087583342e-06,
      "loss": 0.77676433,
      "num_input_tokens_seen": 99156185,
      "step": 4589,
      "time_per_iteration": 2.6969096660614014
    },
    {
      "auxiliary_loss_clip": 0.0113205,
      "auxiliary_loss_mlp": 0.01042943,
      "balance_loss_clip": 1.05382586,
      "balance_loss_mlp": 1.02686214,
      "epoch": 0.27596572974597927,
      "flos": 24281898163200.0,
      "grad_norm": 1.7344916579577159,
      "language_loss": 0.88667816,
      "learning_rate": 3.398220643612143e-06,
      "loss": 0.90842807,
      "num_input_tokens_seen": 99176735,
      "step": 4590,
      "time_per_iteration": 4.020005702972412
    },
    {
      "auxiliary_loss_clip": 0.01120574,
      "auxiliary_loss_mlp": 0.01046791,
      "balance_loss_clip": 1.04740644,
      "balance_loss_mlp": 1.02985179,
      "epoch": 0.27602585299864724,
      "flos": 35040985632000.0,
      "grad_norm": 1.5917515285143642,
      "language_loss": 0.71352559,
      "learning_rate": 3.397942146620277e-06,
      "loss": 0.73519927,
      "num_input_tokens_seen": 99199765,
      "step": 4591,
      "time_per_iteration": 2.6162924766540527
    },
    {
      "auxiliary_loss_clip": 0.01103177,
      "auxiliary_loss_mlp": 0.01042908,
      "balance_loss_clip": 1.04735827,
      "balance_loss_mlp": 1.02623081,
      "epoch": 0.2760859762513152,
      "flos": 24309405002880.0,
      "grad_norm": 1.7693090981193267,
      "language_loss": 0.80589068,
      "learning_rate": 3.3976635966183046e-06,
      "loss": 0.82735151,
      "num_input_tokens_seen": 99218435,
      "step": 4592,
      "time_per_iteration": 2.5975236892700195
    },
    {
      "auxiliary_loss_clip": 0.01042429,
      "auxiliary_loss_mlp": 0.01064954,
      "balance_loss_clip": 1.02105987,
      "balance_loss_mlp": 1.51519334,
      "epoch": 0.27614609950398317,
      "flos": 71260739890560.0,
      "grad_norm": 0.7262632233473194,
      "language_loss": 0.61610973,
      "learning_rate": 3.3973849936167886e-06,
      "loss": 0.63718367,
      "num_input_tokens_seen": 99276200,
      "step": 4593,
      "time_per_iteration": 3.071500778198242
    },
    {
      "auxiliary_loss_clip": 0.01122033,
      "auxiliary_loss_mlp": 0.01044588,
      "balance_loss_clip": 1.04893804,
      "balance_loss_mlp": 1.0290314,
      "epoch": 0.27620622275665113,
      "flos": 29674854138240.0,
      "grad_norm": 1.9477870890362081,
      "language_loss": 0.77191192,
      "learning_rate": 3.3971063376262937e-06,
      "loss": 0.79357809,
      "num_input_tokens_seen": 99297625,
      "step": 4594,
      "time_per_iteration": 3.9729862213134766
    },
    {
      "auxiliary_loss_clip": 0.01126328,
      "auxiliary_loss_mlp": 0.01035069,
      "balance_loss_clip": 1.04918838,
      "balance_loss_mlp": 1.01904774,
      "epoch": 0.2762663460093191,
      "flos": 15378063137280.0,
      "grad_norm": 1.5727802020094417,
      "language_loss": 0.91514361,
      "learning_rate": 3.3968276286573866e-06,
      "loss": 0.93675756,
      "num_input_tokens_seen": 99315790,
      "step": 4595,
      "time_per_iteration": 3.844413995742798
    },
    {
      "auxiliary_loss_clip": 0.01125853,
      "auxiliary_loss_mlp": 0.01047307,
      "balance_loss_clip": 1.05224323,
      "balance_loss_mlp": 1.03037965,
      "epoch": 0.27632646926198706,
      "flos": 20704082117760.0,
      "grad_norm": 1.794384131044761,
      "language_loss": 0.69423991,
      "learning_rate": 3.3965488667206353e-06,
      "loss": 0.71597153,
      "num_input_tokens_seen": 99334615,
      "step": 4596,
      "time_per_iteration": 2.482381820678711
    },
    {
      "auxiliary_loss_clip": 0.01114096,
      "auxiliary_loss_mlp": 0.01043073,
      "balance_loss_clip": 1.04718077,
      "balance_loss_mlp": 1.02651477,
      "epoch": 0.276386592514655,
      "flos": 32813374849920.0,
      "grad_norm": 2.098328341184232,
      "language_loss": 0.6364525,
      "learning_rate": 3.3962700518266113e-06,
      "loss": 0.65802419,
      "num_input_tokens_seen": 99356685,
      "step": 4597,
      "time_per_iteration": 4.0288896560668945
    },
    {
      "auxiliary_loss_clip": 0.01136141,
      "auxiliary_loss_mlp": 0.01040724,
      "balance_loss_clip": 1.04940391,
      "balance_loss_mlp": 1.02554893,
      "epoch": 0.276446715767323,
      "flos": 18551704371840.0,
      "grad_norm": 2.0482349351030797,
      "language_loss": 0.86713678,
      "learning_rate": 3.395991183985887e-06,
      "loss": 0.88890541,
      "num_input_tokens_seen": 99374810,
      "step": 4598,
      "time_per_iteration": 2.435037136077881
    },
    {
      "auxiliary_loss_clip": 0.01137243,
      "auxiliary_loss_mlp": 0.01045467,
      "balance_loss_clip": 1.04756272,
      "balance_loss_mlp": 1.0284555,
      "epoch": 0.27650683901999096,
      "flos": 22819615488000.0,
      "grad_norm": 2.282877417620533,
      "language_loss": 0.79980391,
      "learning_rate": 3.395712263209037e-06,
      "loss": 0.82163101,
      "num_input_tokens_seen": 99391290,
      "step": 4599,
      "time_per_iteration": 2.4718916416168213
    },
    {
      "auxiliary_loss_clip": 0.01117317,
      "auxiliary_loss_mlp": 0.01045297,
      "balance_loss_clip": 1.04641652,
      "balance_loss_mlp": 1.02933502,
      "epoch": 0.276566962272659,
      "flos": 21361534704000.0,
      "grad_norm": 1.9949490391617812,
      "language_loss": 0.79224294,
      "learning_rate": 3.395433289506639e-06,
      "loss": 0.81386912,
      "num_input_tokens_seen": 99409120,
      "step": 4600,
      "time_per_iteration": 2.5263750553131104
    },
    {
      "auxiliary_loss_clip": 0.01107023,
      "auxiliary_loss_mlp": 0.01045421,
      "balance_loss_clip": 1.04740596,
      "balance_loss_mlp": 1.02910113,
      "epoch": 0.27662708552532694,
      "flos": 17710604524800.0,
      "grad_norm": 1.745096568033291,
      "language_loss": 0.72896874,
      "learning_rate": 3.3951542628892694e-06,
      "loss": 0.75049317,
      "num_input_tokens_seen": 99426180,
      "step": 4601,
      "time_per_iteration": 2.5388035774230957
    },
    {
      "auxiliary_loss_clip": 0.01124964,
      "auxiliary_loss_mlp": 0.01044537,
      "balance_loss_clip": 1.04663622,
      "balance_loss_mlp": 1.02746677,
      "epoch": 0.2766872087779949,
      "flos": 21252725429760.0,
      "grad_norm": 1.6226448699763845,
      "language_loss": 0.80026096,
      "learning_rate": 3.3948751833675113e-06,
      "loss": 0.82195592,
      "num_input_tokens_seen": 99447720,
      "step": 4602,
      "time_per_iteration": 2.541095495223999
    },
    {
      "auxiliary_loss_clip": 0.01120022,
      "auxiliary_loss_mlp": 0.0105208,
      "balance_loss_clip": 1.04730737,
      "balance_loss_mlp": 1.03398371,
      "epoch": 0.2767473320306629,
      "flos": 12931900053120.0,
      "grad_norm": 2.0494422499463547,
      "language_loss": 0.77083993,
      "learning_rate": 3.3945960509519455e-06,
      "loss": 0.79256094,
      "num_input_tokens_seen": 99464720,
      "step": 4603,
      "time_per_iteration": 2.503892660140991
    },
    {
      "auxiliary_loss_clip": 0.0110519,
      "auxiliary_loss_mlp": 0.01042108,
      "balance_loss_clip": 1.04653883,
      "balance_loss_mlp": 1.0270164,
      "epoch": 0.27680745528333084,
      "flos": 15012851604480.0,
      "grad_norm": 1.545044163277421,
      "language_loss": 0.81708497,
      "learning_rate": 3.3943168656531585e-06,
      "loss": 0.83855796,
      "num_input_tokens_seen": 99482310,
      "step": 4604,
      "time_per_iteration": 2.505063772201538
    },
    {
      "auxiliary_loss_clip": 0.01090317,
      "auxiliary_loss_mlp": 0.01034344,
      "balance_loss_clip": 1.04821622,
      "balance_loss_mlp": 1.01755977,
      "epoch": 0.2768675785359988,
      "flos": 22637835734400.0,
      "grad_norm": 1.879089817930719,
      "language_loss": 0.70047253,
      "learning_rate": 3.3940376274817363e-06,
      "loss": 0.72171915,
      "num_input_tokens_seen": 99501255,
      "step": 4605,
      "time_per_iteration": 2.6266467571258545
    },
    {
      "auxiliary_loss_clip": 0.01037719,
      "auxiliary_loss_mlp": 0.01014949,
      "balance_loss_clip": 1.01658928,
      "balance_loss_mlp": 1.01294613,
      "epoch": 0.27692770178866677,
      "flos": 66130542881280.0,
      "grad_norm": 0.6991732294776043,
      "language_loss": 0.57192135,
      "learning_rate": 3.3937583364482673e-06,
      "loss": 0.592448,
      "num_input_tokens_seen": 99568925,
      "step": 4606,
      "time_per_iteration": 3.1990973949432373
    },
    {
      "auxiliary_loss_clip": 0.01118391,
      "auxiliary_loss_mlp": 0.01049217,
      "balance_loss_clip": 1.05202675,
      "balance_loss_mlp": 1.03230166,
      "epoch": 0.27698782504133473,
      "flos": 26464979059200.0,
      "grad_norm": 2.007095824110389,
      "language_loss": 0.69247019,
      "learning_rate": 3.3934789925633424e-06,
      "loss": 0.71414638,
      "num_input_tokens_seen": 99588455,
      "step": 4607,
      "time_per_iteration": 2.656494379043579
    },
    {
      "auxiliary_loss_clip": 0.01120114,
      "auxiliary_loss_mlp": 0.01038531,
      "balance_loss_clip": 1.04959011,
      "balance_loss_mlp": 1.02318907,
      "epoch": 0.2770479482940027,
      "flos": 25884806584320.0,
      "grad_norm": 1.8633162070389817,
      "language_loss": 0.69553816,
      "learning_rate": 3.393199595837555e-06,
      "loss": 0.71712458,
      "num_input_tokens_seen": 99609355,
      "step": 4608,
      "time_per_iteration": 2.5660226345062256
    },
    {
      "auxiliary_loss_clip": 0.0108775,
      "auxiliary_loss_mlp": 0.01035153,
      "balance_loss_clip": 1.04684854,
      "balance_loss_mlp": 1.01942933,
      "epoch": 0.27710807154667066,
      "flos": 22857249962880.0,
      "grad_norm": 1.7415987207604806,
      "language_loss": 0.72589314,
      "learning_rate": 3.392920146281499e-06,
      "loss": 0.74712217,
      "num_input_tokens_seen": 99628780,
      "step": 4609,
      "time_per_iteration": 2.6270017623901367
    },
    {
      "auxiliary_loss_clip": 0.01098854,
      "auxiliary_loss_mlp": 0.01049616,
      "balance_loss_clip": 1.04445148,
      "balance_loss_mlp": 1.0321641,
      "epoch": 0.27716819479933863,
      "flos": 17711071401600.0,
      "grad_norm": 2.4332284724049185,
      "language_loss": 0.84144986,
      "learning_rate": 3.3926406439057714e-06,
      "loss": 0.86293459,
      "num_input_tokens_seen": 99644545,
      "step": 4610,
      "time_per_iteration": 2.5473031997680664
    },
    {
      "auxiliary_loss_clip": 0.01070351,
      "auxiliary_loss_mlp": 0.00970023,
      "balance_loss_clip": 1.04251122,
      "balance_loss_mlp": 1.32770705,
      "epoch": 0.2772283180520066,
      "flos": 19646046080640.0,
      "grad_norm": 1.9969084022953179,
      "language_loss": 0.68949115,
      "learning_rate": 3.3923610887209705e-06,
      "loss": 0.7098949,
      "num_input_tokens_seen": 99663125,
      "step": 4611,
      "time_per_iteration": 2.6056246757507324
    },
    {
      "auxiliary_loss_clip": 0.01132708,
      "auxiliary_loss_mlp": 0.01040555,
      "balance_loss_clip": 1.04773974,
      "balance_loss_mlp": 1.02491426,
      "epoch": 0.27728844130467456,
      "flos": 21032628842880.0,
      "grad_norm": 2.2208598116921334,
      "language_loss": 0.73927277,
      "learning_rate": 3.392081480737698e-06,
      "loss": 0.7610054,
      "num_input_tokens_seen": 99682645,
      "step": 4612,
      "time_per_iteration": 2.4833269119262695
    },
    {
      "auxiliary_loss_clip": 0.01126487,
      "auxiliary_loss_mlp": 0.00926401,
      "balance_loss_clip": 1.04549122,
      "balance_loss_mlp": 1.2558229,
      "epoch": 0.2773485645573425,
      "flos": 18989204025600.0,
      "grad_norm": 2.1654356304366074,
      "language_loss": 0.66371137,
      "learning_rate": 3.3918018199665563e-06,
      "loss": 0.68424022,
      "num_input_tokens_seen": 99700520,
      "step": 4613,
      "time_per_iteration": 2.489562749862671
    },
    {
      "auxiliary_loss_clip": 0.01089532,
      "auxiliary_loss_mlp": 0.01048419,
      "balance_loss_clip": 1.04482877,
      "balance_loss_mlp": 1.03044188,
      "epoch": 0.27740868781001055,
      "flos": 21468440557440.0,
      "grad_norm": 1.5999895911122677,
      "language_loss": 0.79339874,
      "learning_rate": 3.39152210641815e-06,
      "loss": 0.81477821,
      "num_input_tokens_seen": 99720355,
      "step": 4614,
      "time_per_iteration": 2.5995330810546875
    },
    {
      "auxiliary_loss_clip": 0.01122576,
      "auxiliary_loss_mlp": 0.01048232,
      "balance_loss_clip": 1.0471983,
      "balance_loss_mlp": 1.0308274,
      "epoch": 0.2774688110626785,
      "flos": 19827825834240.0,
      "grad_norm": 2.4625342958150083,
      "language_loss": 0.80370331,
      "learning_rate": 3.3912423401030865e-06,
      "loss": 0.82541144,
      "num_input_tokens_seen": 99736090,
      "step": 4615,
      "time_per_iteration": 2.4763236045837402
    },
    {
      "auxiliary_loss_clip": 0.01105885,
      "auxiliary_loss_mlp": 0.0104815,
      "balance_loss_clip": 1.04328609,
      "balance_loss_mlp": 1.03128231,
      "epoch": 0.2775289343153465,
      "flos": 18216226321920.0,
      "grad_norm": 2.5065406414970672,
      "language_loss": 0.64132905,
      "learning_rate": 3.3909625210319735e-06,
      "loss": 0.66286939,
      "num_input_tokens_seen": 99751805,
      "step": 4616,
      "time_per_iteration": 2.526428461074829
    },
    {
      "auxiliary_loss_clip": 0.01119088,
      "auxiliary_loss_mlp": 0.01041888,
      "balance_loss_clip": 1.04597723,
      "balance_loss_mlp": 1.02572346,
      "epoch": 0.27758905756801444,
      "flos": 16472476673280.0,
      "grad_norm": 1.9628791553878375,
      "language_loss": 0.82244915,
      "learning_rate": 3.3906826492154226e-06,
      "loss": 0.84405893,
      "num_input_tokens_seen": 99770610,
      "step": 4617,
      "time_per_iteration": 2.463698387145996
    },
    {
      "auxiliary_loss_clip": 0.0113526,
      "auxiliary_loss_mlp": 0.01046815,
      "balance_loss_clip": 1.04471505,
      "balance_loss_mlp": 1.0306505,
      "epoch": 0.2776491808206824,
      "flos": 18728240739840.0,
      "grad_norm": 2.1035555178318246,
      "language_loss": 0.76906061,
      "learning_rate": 3.3904027246640458e-06,
      "loss": 0.7908814,
      "num_input_tokens_seen": 99787305,
      "step": 4618,
      "time_per_iteration": 2.4476096630096436
    },
    {
      "auxiliary_loss_clip": 0.01140291,
      "auxiliary_loss_mlp": 0.01043357,
      "balance_loss_clip": 1.04953671,
      "balance_loss_mlp": 1.02762127,
      "epoch": 0.27770930407335037,
      "flos": 28038189911040.0,
      "grad_norm": 1.7186766072974553,
      "language_loss": 0.84697956,
      "learning_rate": 3.390122747388459e-06,
      "loss": 0.86881602,
      "num_input_tokens_seen": 99808940,
      "step": 4619,
      "time_per_iteration": 2.4974989891052246
    },
    {
      "auxiliary_loss_clip": 0.01115397,
      "auxiliary_loss_mlp": 0.01044371,
      "balance_loss_clip": 1.04970312,
      "balance_loss_mlp": 1.02936208,
      "epoch": 0.27776942732601834,
      "flos": 23549823072000.0,
      "grad_norm": 1.4532002320895863,
      "language_loss": 0.77073061,
      "learning_rate": 3.3898427173992778e-06,
      "loss": 0.79232824,
      "num_input_tokens_seen": 99829575,
      "step": 4620,
      "time_per_iteration": 2.553058385848999
    },
    {
      "auxiliary_loss_clip": 0.01083806,
      "auxiliary_loss_mlp": 0.01045214,
      "balance_loss_clip": 1.04345798,
      "balance_loss_mlp": 1.02858424,
      "epoch": 0.2778295505786863,
      "flos": 23908713811200.0,
      "grad_norm": 1.9414833436103471,
      "language_loss": 0.78531361,
      "learning_rate": 3.389562634707122e-06,
      "loss": 0.80660379,
      "num_input_tokens_seen": 99847575,
      "step": 4621,
      "time_per_iteration": 2.5575671195983887
    },
    {
      "auxiliary_loss_clip": 0.01105024,
      "auxiliary_loss_mlp": 0.01048569,
      "balance_loss_clip": 1.04612386,
      "balance_loss_mlp": 1.03178477,
      "epoch": 0.27788967383135427,
      "flos": 25554571920000.0,
      "grad_norm": 2.054327389108337,
      "language_loss": 0.88047147,
      "learning_rate": 3.389282499322611e-06,
      "loss": 0.90200746,
      "num_input_tokens_seen": 99864995,
      "step": 4622,
      "time_per_iteration": 2.5794317722320557
    },
    {
      "auxiliary_loss_clip": 0.01092365,
      "auxiliary_loss_mlp": 0.01046548,
      "balance_loss_clip": 1.05089056,
      "balance_loss_mlp": 1.02983463,
      "epoch": 0.27794979708402223,
      "flos": 16252631481600.0,
      "grad_norm": 1.832885730134156,
      "language_loss": 0.81256199,
      "learning_rate": 3.389002311256369e-06,
      "loss": 0.83395118,
      "num_input_tokens_seen": 99881540,
      "step": 4623,
      "time_per_iteration": 2.566901683807373
    },
    {
      "auxiliary_loss_clip": 0.01112504,
      "auxiliary_loss_mlp": 0.01041143,
      "balance_loss_clip": 1.05366421,
      "balance_loss_mlp": 1.02533627,
      "epoch": 0.2780099203366902,
      "flos": 20667632791680.0,
      "grad_norm": 1.9271216548617651,
      "language_loss": 0.81662834,
      "learning_rate": 3.3887220705190204e-06,
      "loss": 0.83816481,
      "num_input_tokens_seen": 99899595,
      "step": 4624,
      "time_per_iteration": 2.5724620819091797
    },
    {
      "auxiliary_loss_clip": 0.01103994,
      "auxiliary_loss_mlp": 0.0086115,
      "balance_loss_clip": 1.04751551,
      "balance_loss_mlp": 1.13005877,
      "epoch": 0.27807004358935816,
      "flos": 17739583822080.0,
      "grad_norm": 2.333408868810702,
      "language_loss": 0.76850295,
      "learning_rate": 3.388441777121191e-06,
      "loss": 0.78815436,
      "num_input_tokens_seen": 99913020,
      "step": 4625,
      "time_per_iteration": 2.4856669902801514
    },
    {
      "auxiliary_loss_clip": 0.01101894,
      "auxiliary_loss_mlp": 0.01045432,
      "balance_loss_clip": 1.04383314,
      "balance_loss_mlp": 1.028409,
      "epoch": 0.2781301668420261,
      "flos": 16727119165440.0,
      "grad_norm": 1.9236987067820803,
      "language_loss": 0.70091903,
      "learning_rate": 3.388161431073511e-06,
      "loss": 0.72239232,
      "num_input_tokens_seen": 99931405,
      "step": 4626,
      "time_per_iteration": 2.5033881664276123
    },
    {
      "auxiliary_loss_clip": 0.01100071,
      "auxiliary_loss_mlp": 0.01037942,
      "balance_loss_clip": 1.05040431,
      "balance_loss_mlp": 1.02050173,
      "epoch": 0.27819029009469415,
      "flos": 13844749317120.0,
      "grad_norm": 2.220480798840469,
      "language_loss": 0.92374498,
      "learning_rate": 3.38788103238661e-06,
      "loss": 0.9451251,
      "num_input_tokens_seen": 99948100,
      "step": 4627,
      "time_per_iteration": 2.590651512145996
    },
    {
      "auxiliary_loss_clip": 0.01138132,
      "auxiliary_loss_mlp": 0.01036755,
      "balance_loss_clip": 1.04818082,
      "balance_loss_mlp": 1.02204502,
      "epoch": 0.2782504133473621,
      "flos": 27089286370560.0,
      "grad_norm": 1.8443621619033856,
      "language_loss": 0.8555727,
      "learning_rate": 3.387600581071121e-06,
      "loss": 0.8773216,
      "num_input_tokens_seen": 99966470,
      "step": 4628,
      "time_per_iteration": 3.9970452785491943
    },
    {
      "auxiliary_loss_clip": 0.01104961,
      "auxiliary_loss_mlp": 0.01037681,
      "balance_loss_clip": 1.04529834,
      "balance_loss_mlp": 1.02195776,
      "epoch": 0.2783105366000301,
      "flos": 21068826773760.0,
      "grad_norm": 1.6773352140184754,
      "language_loss": 0.79432112,
      "learning_rate": 3.387320077137679e-06,
      "loss": 0.81574756,
      "num_input_tokens_seen": 99985930,
      "step": 4629,
      "time_per_iteration": 2.60141921043396
    },
    {
      "auxiliary_loss_clip": 0.01096186,
      "auxiliary_loss_mlp": 0.01035997,
      "balance_loss_clip": 1.04897237,
      "balance_loss_mlp": 1.02100015,
      "epoch": 0.27837065985269804,
      "flos": 26501823434880.0,
      "grad_norm": 1.5225963087161418,
      "language_loss": 0.84674692,
      "learning_rate": 3.3870395205969208e-06,
      "loss": 0.8680687,
      "num_input_tokens_seen": 100006235,
      "step": 4630,
      "time_per_iteration": 2.5954160690307617
    },
    {
      "auxiliary_loss_clip": 0.01116265,
      "auxiliary_loss_mlp": 0.01036389,
      "balance_loss_clip": 1.04775929,
      "balance_loss_mlp": 1.02040255,
      "epoch": 0.278430783105366,
      "flos": 20223201813120.0,
      "grad_norm": 2.056054936080906,
      "language_loss": 0.80716789,
      "learning_rate": 3.386758911459485e-06,
      "loss": 0.8286944,
      "num_input_tokens_seen": 100023655,
      "step": 4631,
      "time_per_iteration": 2.5210793018341064
    },
    {
      "auxiliary_loss_clip": 0.01140955,
      "auxiliary_loss_mlp": 0.01044766,
      "balance_loss_clip": 1.05197287,
      "balance_loss_mlp": 1.0289588,
      "epoch": 0.278490906358034,
      "flos": 25592888753280.0,
      "grad_norm": 2.0131091829862107,
      "language_loss": 0.71281576,
      "learning_rate": 3.3864782497360126e-06,
      "loss": 0.73467302,
      "num_input_tokens_seen": 100043280,
      "step": 4632,
      "time_per_iteration": 3.8958778381347656
    },
    {
      "auxiliary_loss_clip": 0.01124893,
      "auxiliary_loss_mlp": 0.0103816,
      "balance_loss_clip": 1.05348122,
      "balance_loss_mlp": 1.02353859,
      "epoch": 0.27855102961070194,
      "flos": 16171544528640.0,
      "grad_norm": 2.8729806000309006,
      "language_loss": 0.82598174,
      "learning_rate": 3.386197535437145e-06,
      "loss": 0.84761232,
      "num_input_tokens_seen": 100057690,
      "step": 4633,
      "time_per_iteration": 2.4715912342071533
    },
    {
      "auxiliary_loss_clip": 0.01114567,
      "auxiliary_loss_mlp": 0.01036745,
      "balance_loss_clip": 1.04612207,
      "balance_loss_mlp": 1.01996088,
      "epoch": 0.2786111528633699,
      "flos": 22927598749440.0,
      "grad_norm": 1.7099276805900336,
      "language_loss": 0.88070834,
      "learning_rate": 3.385916768573529e-06,
      "loss": 0.9022215,
      "num_input_tokens_seen": 100075875,
      "step": 4634,
      "time_per_iteration": 3.918361186981201
    },
    {
      "auxiliary_loss_clip": 0.01111288,
      "auxiliary_loss_mlp": 0.01038052,
      "balance_loss_clip": 1.05041873,
      "balance_loss_mlp": 1.02167225,
      "epoch": 0.27867127611603787,
      "flos": 23404205335680.0,
      "grad_norm": 1.5728230436888795,
      "language_loss": 0.77092016,
      "learning_rate": 3.38563594915581e-06,
      "loss": 0.79241359,
      "num_input_tokens_seen": 100092930,
      "step": 4635,
      "time_per_iteration": 2.525675058364868
    },
    {
      "auxiliary_loss_clip": 0.01137117,
      "auxiliary_loss_mlp": 0.01041733,
      "balance_loss_clip": 1.04662895,
      "balance_loss_mlp": 1.02541304,
      "epoch": 0.27873139936870583,
      "flos": 19829010983040.0,
      "grad_norm": 1.8168896741910923,
      "language_loss": 0.65567625,
      "learning_rate": 3.385355077194637e-06,
      "loss": 0.67746478,
      "num_input_tokens_seen": 100110790,
      "step": 4636,
      "time_per_iteration": 3.8618345260620117
    },
    {
      "auxiliary_loss_clip": 0.01123143,
      "auxiliary_loss_mlp": 0.01041048,
      "balance_loss_clip": 1.04482627,
      "balance_loss_mlp": 1.02439427,
      "epoch": 0.2787915226213738,
      "flos": 17707659609600.0,
      "grad_norm": 2.338718946525157,
      "language_loss": 0.8375017,
      "learning_rate": 3.3850741527006604e-06,
      "loss": 0.85914361,
      "num_input_tokens_seen": 100126970,
      "step": 4637,
      "time_per_iteration": 2.501603126525879
    },
    {
      "auxiliary_loss_clip": 0.0110914,
      "auxiliary_loss_mlp": 0.01038811,
      "balance_loss_clip": 1.0451746,
      "balance_loss_mlp": 1.0236001,
      "epoch": 0.27885164587404176,
      "flos": 22090557139200.0,
      "grad_norm": 1.560462396490986,
      "language_loss": 0.75961751,
      "learning_rate": 3.384793175684533e-06,
      "loss": 0.78109705,
      "num_input_tokens_seen": 100146720,
      "step": 4638,
      "time_per_iteration": 2.5474679470062256
    },
    {
      "auxiliary_loss_clip": 0.0112359,
      "auxiliary_loss_mlp": 0.01046815,
      "balance_loss_clip": 1.04570699,
      "balance_loss_mlp": 1.03063846,
      "epoch": 0.27891176912670973,
      "flos": 19207684500480.0,
      "grad_norm": 1.4802483772938428,
      "language_loss": 0.71746266,
      "learning_rate": 3.38451214615691e-06,
      "loss": 0.73916662,
      "num_input_tokens_seen": 100165920,
      "step": 4639,
      "time_per_iteration": 2.56996750831604
    },
    {
      "auxiliary_loss_clip": 0.01125007,
      "auxiliary_loss_mlp": 0.01031937,
      "balance_loss_clip": 1.0462656,
      "balance_loss_mlp": 1.01592779,
      "epoch": 0.27897189237937775,
      "flos": 27600007898880.0,
      "grad_norm": 1.9886332186873583,
      "language_loss": 0.65409517,
      "learning_rate": 3.384231064128447e-06,
      "loss": 0.67566466,
      "num_input_tokens_seen": 100185525,
      "step": 4640,
      "time_per_iteration": 2.621410846710205
    },
    {
      "auxiliary_loss_clip": 0.01127101,
      "auxiliary_loss_mlp": 0.01035471,
      "balance_loss_clip": 1.04912806,
      "balance_loss_mlp": 1.02020025,
      "epoch": 0.2790320156320457,
      "flos": 21178210665600.0,
      "grad_norm": 2.132173514709288,
      "language_loss": 0.72568685,
      "learning_rate": 3.383949929609804e-06,
      "loss": 0.74731255,
      "num_input_tokens_seen": 100204850,
      "step": 4641,
      "time_per_iteration": 2.515049695968628
    },
    {
      "auxiliary_loss_clip": 0.01104436,
      "auxiliary_loss_mlp": 0.01038865,
      "balance_loss_clip": 1.05019653,
      "balance_loss_mlp": 1.02145994,
      "epoch": 0.2790921388847137,
      "flos": 22783920347520.0,
      "grad_norm": 1.8084714821666483,
      "language_loss": 0.75005323,
      "learning_rate": 3.383668742611641e-06,
      "loss": 0.77148622,
      "num_input_tokens_seen": 100224520,
      "step": 4642,
      "time_per_iteration": 2.5699260234832764
    },
    {
      "auxiliary_loss_clip": 0.01099257,
      "auxiliary_loss_mlp": 0.01041271,
      "balance_loss_clip": 1.04186738,
      "balance_loss_mlp": 1.02415264,
      "epoch": 0.27915226213738165,
      "flos": 23400649889280.0,
      "grad_norm": 1.7883942494773621,
      "language_loss": 0.86328977,
      "learning_rate": 3.3833875031446205e-06,
      "loss": 0.88469505,
      "num_input_tokens_seen": 100243935,
      "step": 4643,
      "time_per_iteration": 2.5711557865142822
    },
    {
      "auxiliary_loss_clip": 0.01097974,
      "auxiliary_loss_mlp": 0.01044078,
      "balance_loss_clip": 1.04903412,
      "balance_loss_mlp": 1.02775788,
      "epoch": 0.2792123853900496,
      "flos": 22747794243840.0,
      "grad_norm": 2.0920994559622472,
      "language_loss": 0.83128965,
      "learning_rate": 3.383106211219407e-06,
      "loss": 0.85271019,
      "num_input_tokens_seen": 100262290,
      "step": 4644,
      "time_per_iteration": 2.5910048484802246
    },
    {
      "auxiliary_loss_clip": 0.01126353,
      "auxiliary_loss_mlp": 0.01036884,
      "balance_loss_clip": 1.04743123,
      "balance_loss_mlp": 1.02117229,
      "epoch": 0.2792725086427176,
      "flos": 15049372757760.0,
      "grad_norm": 2.141806993340925,
      "language_loss": 0.78821397,
      "learning_rate": 3.3828248668466673e-06,
      "loss": 0.80984634,
      "num_input_tokens_seen": 100280015,
      "step": 4645,
      "time_per_iteration": 2.4989938735961914
    },
    {
      "auxiliary_loss_clip": 0.01038937,
      "auxiliary_loss_mlp": 0.01002614,
      "balance_loss_clip": 1.02718139,
      "balance_loss_mlp": 1.00020587,
      "epoch": 0.27933263189538554,
      "flos": 62544861757440.0,
      "grad_norm": 0.77199471251545,
      "language_loss": 0.62278366,
      "learning_rate": 3.3825434700370705e-06,
      "loss": 0.64319915,
      "num_input_tokens_seen": 100338935,
      "step": 4646,
      "time_per_iteration": 3.1178700923919678
    },
    {
      "auxiliary_loss_clip": 0.01106299,
      "auxiliary_loss_mlp": 0.01035882,
      "balance_loss_clip": 1.04774046,
      "balance_loss_mlp": 1.02127862,
      "epoch": 0.2793927551480535,
      "flos": 25118365155840.0,
      "grad_norm": 2.140834403548063,
      "language_loss": 0.89443445,
      "learning_rate": 3.3822620208012865e-06,
      "loss": 0.91585624,
      "num_input_tokens_seen": 100359905,
      "step": 4647,
      "time_per_iteration": 2.6019418239593506
    },
    {
      "auxiliary_loss_clip": 0.01128825,
      "auxiliary_loss_mlp": 0.01042754,
      "balance_loss_clip": 1.04668164,
      "balance_loss_mlp": 1.02680373,
      "epoch": 0.27945287840072147,
      "flos": 21324582587520.0,
      "grad_norm": 1.9868112668555429,
      "language_loss": 0.87038803,
      "learning_rate": 3.381980519149988e-06,
      "loss": 0.89210391,
      "num_input_tokens_seen": 100376955,
      "step": 4648,
      "time_per_iteration": 2.5379035472869873
    },
    {
      "auxiliary_loss_clip": 0.01127397,
      "auxiliary_loss_mlp": 0.01035924,
      "balance_loss_clip": 1.04859483,
      "balance_loss_mlp": 1.02043843,
      "epoch": 0.27951300165338944,
      "flos": 27450547407360.0,
      "grad_norm": 2.647708062291757,
      "language_loss": 0.72879887,
      "learning_rate": 3.38169896509385e-06,
      "loss": 0.75043207,
      "num_input_tokens_seen": 100397545,
      "step": 4649,
      "time_per_iteration": 2.6446566581726074
    },
    {
      "auxiliary_loss_clip": 0.01104258,
      "auxiliary_loss_mlp": 0.01037894,
      "balance_loss_clip": 1.04621673,
      "balance_loss_mlp": 1.02039361,
      "epoch": 0.2795731249060574,
      "flos": 15159008044800.0,
      "grad_norm": 2.016975520561866,
      "language_loss": 0.80387753,
      "learning_rate": 3.381417358643549e-06,
      "loss": 0.82529902,
      "num_input_tokens_seen": 100415080,
      "step": 4650,
      "time_per_iteration": 2.559195041656494
    },
    {
      "auxiliary_loss_clip": 0.01041749,
      "auxiliary_loss_mlp": 0.01101311,
      "balance_loss_clip": 1.03725791,
      "balance_loss_mlp": 1.57935834,
      "epoch": 0.27963324815872537,
      "flos": 60120103178880.0,
      "grad_norm": 0.833010792876427,
      "language_loss": 0.58868992,
      "learning_rate": 3.3811356998097624e-06,
      "loss": 0.61012053,
      "num_input_tokens_seen": 100471105,
      "step": 4651,
      "time_per_iteration": 3.195953845977783
    },
    {
      "auxiliary_loss_clip": 0.01124402,
      "auxiliary_loss_mlp": 0.01040942,
      "balance_loss_clip": 1.04370177,
      "balance_loss_mlp": 1.02368093,
      "epoch": 0.27969337141139333,
      "flos": 21765960910080.0,
      "grad_norm": 2.2462954633140018,
      "language_loss": 0.73860872,
      "learning_rate": 3.3808539886031726e-06,
      "loss": 0.76026213,
      "num_input_tokens_seen": 100492520,
      "step": 4652,
      "time_per_iteration": 2.527059555053711
    },
    {
      "auxiliary_loss_clip": 0.01139726,
      "auxiliary_loss_mlp": 0.01042486,
      "balance_loss_clip": 1.04931331,
      "balance_loss_mlp": 1.02638078,
      "epoch": 0.27975349466406135,
      "flos": 39851398834560.0,
      "grad_norm": 2.2423148305832967,
      "language_loss": 0.79883558,
      "learning_rate": 3.380572225034461e-06,
      "loss": 0.82065773,
      "num_input_tokens_seen": 100512870,
      "step": 4653,
      "time_per_iteration": 2.610903263092041
    },
    {
      "auxiliary_loss_clip": 0.01113825,
      "auxiliary_loss_mlp": 0.01042194,
      "balance_loss_clip": 1.04847276,
      "balance_loss_mlp": 1.02607083,
      "epoch": 0.2798136179167293,
      "flos": 21579799697280.0,
      "grad_norm": 2.361965493090918,
      "language_loss": 0.7899189,
      "learning_rate": 3.380290409114312e-06,
      "loss": 0.81147909,
      "num_input_tokens_seen": 100531655,
      "step": 4654,
      "time_per_iteration": 2.5414981842041016
    },
    {
      "auxiliary_loss_clip": 0.01088722,
      "auxiliary_loss_mlp": 0.01042359,
      "balance_loss_clip": 1.04437375,
      "balance_loss_mlp": 1.02504945,
      "epoch": 0.2798737411693973,
      "flos": 21537676022400.0,
      "grad_norm": 1.7510041641377763,
      "language_loss": 0.80720276,
      "learning_rate": 3.3800085408534127e-06,
      "loss": 0.8285135,
      "num_input_tokens_seen": 100548005,
      "step": 4655,
      "time_per_iteration": 2.612304925918579
    },
    {
      "auxiliary_loss_clip": 0.0110336,
      "auxiliary_loss_mlp": 0.00819539,
      "balance_loss_clip": 1.04477298,
      "balance_loss_mlp": 1.05953538,
      "epoch": 0.27993386442206525,
      "flos": 26981051713920.0,
      "grad_norm": 1.8097650784293082,
      "language_loss": 0.818385,
      "learning_rate": 3.3797266202624506e-06,
      "loss": 0.83761394,
      "num_input_tokens_seen": 100567980,
      "step": 4656,
      "time_per_iteration": 2.6301631927490234
    },
    {
      "auxiliary_loss_clip": 0.01107626,
      "auxiliary_loss_mlp": 0.01035967,
      "balance_loss_clip": 1.04694164,
      "balance_loss_mlp": 1.01987422,
      "epoch": 0.2799939876747332,
      "flos": 24349876652160.0,
      "grad_norm": 1.6411319824669215,
      "language_loss": 0.83330786,
      "learning_rate": 3.3794446473521176e-06,
      "loss": 0.85474384,
      "num_input_tokens_seen": 100588630,
      "step": 4657,
      "time_per_iteration": 2.578326940536499
    },
    {
      "auxiliary_loss_clip": 0.01101136,
      "auxiliary_loss_mlp": 0.01044654,
      "balance_loss_clip": 1.04536581,
      "balance_loss_mlp": 1.02764308,
      "epoch": 0.2800541109274012,
      "flos": 33656988648960.0,
      "grad_norm": 1.982101006700611,
      "language_loss": 0.63934344,
      "learning_rate": 3.379162622133105e-06,
      "loss": 0.66080135,
      "num_input_tokens_seen": 100608775,
      "step": 4658,
      "time_per_iteration": 2.6612305641174316
    },
    {
      "auxiliary_loss_clip": 0.01124837,
      "auxiliary_loss_mlp": 0.01047489,
      "balance_loss_clip": 1.04482651,
      "balance_loss_mlp": 1.03096616,
      "epoch": 0.28011423418006914,
      "flos": 21614417429760.0,
      "grad_norm": 1.7052387776381874,
      "language_loss": 0.78747398,
      "learning_rate": 3.3788805446161073e-06,
      "loss": 0.80919725,
      "num_input_tokens_seen": 100627975,
      "step": 4659,
      "time_per_iteration": 2.5252163410186768
    },
    {
      "auxiliary_loss_clip": 0.01099112,
      "auxiliary_loss_mlp": 0.01052073,
      "balance_loss_clip": 1.04606164,
      "balance_loss_mlp": 1.03557467,
      "epoch": 0.2801743574327371,
      "flos": 23112431159040.0,
      "grad_norm": 1.669471323442172,
      "language_loss": 0.7909928,
      "learning_rate": 3.3785984148118215e-06,
      "loss": 0.81250465,
      "num_input_tokens_seen": 100645430,
      "step": 4660,
      "time_per_iteration": 2.582948684692383
    },
    {
      "auxiliary_loss_clip": 0.01095089,
      "auxiliary_loss_mlp": 0.01040743,
      "balance_loss_clip": 1.04538989,
      "balance_loss_mlp": 1.02549028,
      "epoch": 0.2802344806854051,
      "flos": 12641418766080.0,
      "grad_norm": 1.7014261992680957,
      "language_loss": 0.80440485,
      "learning_rate": 3.3783162327309453e-06,
      "loss": 0.82576323,
      "num_input_tokens_seen": 100663775,
      "step": 4661,
      "time_per_iteration": 2.550361394882202
    },
    {
      "auxiliary_loss_clip": 0.01117089,
      "auxiliary_loss_mlp": 0.01051152,
      "balance_loss_clip": 1.04930079,
      "balance_loss_mlp": 1.03491592,
      "epoch": 0.28029460393807304,
      "flos": 37267878142080.0,
      "grad_norm": 2.084969429345593,
      "language_loss": 0.78964412,
      "learning_rate": 3.3780339983841794e-06,
      "loss": 0.8113265,
      "num_input_tokens_seen": 100686085,
      "step": 4662,
      "time_per_iteration": 2.670490026473999
    },
    {
      "auxiliary_loss_clip": 0.01118154,
      "auxiliary_loss_mlp": 0.010417,
      "balance_loss_clip": 1.0430541,
      "balance_loss_mlp": 1.02335382,
      "epoch": 0.280354727190741,
      "flos": 20741106061440.0,
      "grad_norm": 1.6029331798194923,
      "language_loss": 0.69727182,
      "learning_rate": 3.377751711782227e-06,
      "loss": 0.71887034,
      "num_input_tokens_seen": 100705135,
      "step": 4663,
      "time_per_iteration": 2.5277862548828125
    },
    {
      "auxiliary_loss_clip": 0.01129237,
      "auxiliary_loss_mlp": 0.01046153,
      "balance_loss_clip": 1.06351531,
      "balance_loss_mlp": 1.02902293,
      "epoch": 0.28041485044340897,
      "flos": 21471026336640.0,
      "grad_norm": 1.6553843075939263,
      "language_loss": 0.77396965,
      "learning_rate": 3.377469372935791e-06,
      "loss": 0.79572356,
      "num_input_tokens_seen": 100724960,
      "step": 4664,
      "time_per_iteration": 2.5541462898254395
    },
    {
      "auxiliary_loss_clip": 0.01107859,
      "auxiliary_loss_mlp": 0.01044576,
      "balance_loss_clip": 1.05315757,
      "balance_loss_mlp": 1.0288403,
      "epoch": 0.28047497369607693,
      "flos": 14794263388800.0,
      "grad_norm": 1.69870994289954,
      "language_loss": 0.79033214,
      "learning_rate": 3.377186981855578e-06,
      "loss": 0.81185651,
      "num_input_tokens_seen": 100741995,
      "step": 4665,
      "time_per_iteration": 2.520967483520508
    },
    {
      "auxiliary_loss_clip": 0.01120048,
      "auxiliary_loss_mlp": 0.01040802,
      "balance_loss_clip": 1.04307604,
      "balance_loss_mlp": 1.02476859,
      "epoch": 0.2805350969487449,
      "flos": 23070738447360.0,
      "grad_norm": 1.749609877681062,
      "language_loss": 0.81003022,
      "learning_rate": 3.3769045385522968e-06,
      "loss": 0.83163869,
      "num_input_tokens_seen": 100758985,
      "step": 4666,
      "time_per_iteration": 3.872018337249756
    },
    {
      "auxiliary_loss_clip": 0.01097885,
      "auxiliary_loss_mlp": 0.01054708,
      "balance_loss_clip": 1.04372501,
      "balance_loss_mlp": 1.03642142,
      "epoch": 0.2805952202014129,
      "flos": 20479855466880.0,
      "grad_norm": 1.8043685598694792,
      "language_loss": 0.85038841,
      "learning_rate": 3.376622043036658e-06,
      "loss": 0.87191433,
      "num_input_tokens_seen": 100777820,
      "step": 4667,
      "time_per_iteration": 2.5516834259033203
    },
    {
      "auxiliary_loss_clip": 0.01109251,
      "auxiliary_loss_mlp": 0.00856152,
      "balance_loss_clip": 1.04683554,
      "balance_loss_mlp": 1.12841833,
      "epoch": 0.2806553434540809,
      "flos": 27417330305280.0,
      "grad_norm": 1.5976279354310383,
      "language_loss": 0.79502612,
      "learning_rate": 3.376339495319373e-06,
      "loss": 0.8146801,
      "num_input_tokens_seen": 100798205,
      "step": 4668,
      "time_per_iteration": 2.6213877201080322
    },
    {
      "auxiliary_loss_clip": 0.01090446,
      "auxiliary_loss_mlp": 0.01042703,
      "balance_loss_clip": 1.05952621,
      "balance_loss_mlp": 1.02604938,
      "epoch": 0.28071546670674885,
      "flos": 26505019745280.0,
      "grad_norm": 1.3921098550598912,
      "language_loss": 0.76412952,
      "learning_rate": 3.3760568954111563e-06,
      "loss": 0.78546101,
      "num_input_tokens_seen": 100819800,
      "step": 4669,
      "time_per_iteration": 2.6618430614471436
    },
    {
      "auxiliary_loss_clip": 0.01124576,
      "auxiliary_loss_mlp": 0.01042096,
      "balance_loss_clip": 1.04607916,
      "balance_loss_mlp": 1.02585995,
      "epoch": 0.2807755899594168,
      "flos": 20558679863040.0,
      "grad_norm": 1.9127563800949565,
      "language_loss": 0.78800786,
      "learning_rate": 3.375774243322725e-06,
      "loss": 0.8096745,
      "num_input_tokens_seen": 100837880,
      "step": 4670,
      "time_per_iteration": 3.892761707305908
    },
    {
      "auxiliary_loss_clip": 0.01096477,
      "auxiliary_loss_mlp": 0.010456,
      "balance_loss_clip": 1.04462171,
      "balance_loss_mlp": 1.02771902,
      "epoch": 0.2808357132120848,
      "flos": 24313319585280.0,
      "grad_norm": 1.9736450835321329,
      "language_loss": 0.78751451,
      "learning_rate": 3.3754915390647955e-06,
      "loss": 0.80893528,
      "num_input_tokens_seen": 100856350,
      "step": 4671,
      "time_per_iteration": 2.594900131225586
    },
    {
      "auxiliary_loss_clip": 0.0111549,
      "auxiliary_loss_mlp": 0.01041309,
      "balance_loss_clip": 1.04449296,
      "balance_loss_mlp": 1.02495337,
      "epoch": 0.28089583646475275,
      "flos": 26432408401920.0,
      "grad_norm": 1.6749655398289727,
      "language_loss": 0.75135559,
      "learning_rate": 3.37520878264809e-06,
      "loss": 0.77292353,
      "num_input_tokens_seen": 100876135,
      "step": 4672,
      "time_per_iteration": 2.542782783508301
    },
    {
      "auxiliary_loss_clip": 0.01115214,
      "auxiliary_loss_mlp": 0.01041074,
      "balance_loss_clip": 1.04347706,
      "balance_loss_mlp": 1.02238178,
      "epoch": 0.2809559597174207,
      "flos": 23111820627840.0,
      "grad_norm": 7.019177576679457,
      "language_loss": 0.75139004,
      "learning_rate": 3.3749259740833286e-06,
      "loss": 0.77295291,
      "num_input_tokens_seen": 100894790,
      "step": 4673,
      "time_per_iteration": 3.936271905899048
    },
    {
      "auxiliary_loss_clip": 0.01123439,
      "auxiliary_loss_mlp": 0.01037698,
      "balance_loss_clip": 1.04712594,
      "balance_loss_mlp": 1.02110422,
      "epoch": 0.2810160829700887,
      "flos": 20923496346240.0,
      "grad_norm": 1.7859653993196654,
      "language_loss": 0.727992,
      "learning_rate": 3.374643113381237e-06,
      "loss": 0.74960339,
      "num_input_tokens_seen": 100915100,
      "step": 4674,
      "time_per_iteration": 3.9172489643096924
    },
    {
      "auxiliary_loss_clip": 0.01129035,
      "auxiliary_loss_mlp": 0.0103869,
      "balance_loss_clip": 1.04793084,
      "balance_loss_mlp": 1.02123761,
      "epoch": 0.28107620622275664,
      "flos": 14355901808640.0,
      "grad_norm": 1.946414113035512,
      "language_loss": 0.77284235,
      "learning_rate": 3.374360200552541e-06,
      "loss": 0.79451954,
      "num_input_tokens_seen": 100932795,
      "step": 4675,
      "time_per_iteration": 2.5007379055023193
    },
    {
      "auxiliary_loss_clip": 0.01136569,
      "auxiliary_loss_mlp": 0.01040781,
      "balance_loss_clip": 1.04534388,
      "balance_loss_mlp": 1.02379346,
      "epoch": 0.2811363294754246,
      "flos": 20919078973440.0,
      "grad_norm": 1.998049523469888,
      "language_loss": 0.70266777,
      "learning_rate": 3.374077235607968e-06,
      "loss": 0.72444129,
      "num_input_tokens_seen": 100950505,
      "step": 4676,
      "time_per_iteration": 2.4650843143463135
    },
    {
      "auxiliary_loss_clip": 0.01134299,
      "auxiliary_loss_mlp": 0.01036167,
      "balance_loss_clip": 1.04904854,
      "balance_loss_mlp": 1.02105153,
      "epoch": 0.28119645272809257,
      "flos": 20594841880320.0,
      "grad_norm": 1.5541214466072006,
      "language_loss": 0.69771254,
      "learning_rate": 3.3737942185582487e-06,
      "loss": 0.71941721,
      "num_input_tokens_seen": 100968790,
      "step": 4677,
      "time_per_iteration": 2.454850912094116
    },
    {
      "auxiliary_loss_clip": 0.01118687,
      "auxiliary_loss_mlp": 0.01044677,
      "balance_loss_clip": 1.04783058,
      "balance_loss_mlp": 1.02660513,
      "epoch": 0.28125657598076054,
      "flos": 25337420248320.0,
      "grad_norm": 1.5466186628135945,
      "language_loss": 0.63499784,
      "learning_rate": 3.3735111494141153e-06,
      "loss": 0.65663147,
      "num_input_tokens_seen": 100990205,
      "step": 4678,
      "time_per_iteration": 2.5450096130371094
    },
    {
      "auxiliary_loss_clip": 0.01124337,
      "auxiliary_loss_mlp": 0.01041046,
      "balance_loss_clip": 1.04573441,
      "balance_loss_mlp": 1.0253818,
      "epoch": 0.2813166992334285,
      "flos": 24827093769600.0,
      "grad_norm": 1.5374594470617675,
      "language_loss": 0.70330453,
      "learning_rate": 3.3732280281863013e-06,
      "loss": 0.7249583,
      "num_input_tokens_seen": 101009815,
      "step": 4679,
      "time_per_iteration": 2.5043954849243164
    },
    {
      "auxiliary_loss_clip": 0.01124017,
      "auxiliary_loss_mlp": 0.01040753,
      "balance_loss_clip": 1.04533398,
      "balance_loss_mlp": 1.0235157,
      "epoch": 0.2813768224860965,
      "flos": 21760753438080.0,
      "grad_norm": 1.875750232638296,
      "language_loss": 0.74803919,
      "learning_rate": 3.3729448548855422e-06,
      "loss": 0.76968694,
      "num_input_tokens_seen": 101026780,
      "step": 4680,
      "time_per_iteration": 2.4857938289642334
    },
    {
      "auxiliary_loss_clip": 0.01136772,
      "auxiliary_loss_mlp": 0.01040502,
      "balance_loss_clip": 1.04715753,
      "balance_loss_mlp": 1.02531505,
      "epoch": 0.2814369457387645,
      "flos": 24316803204480.0,
      "grad_norm": 1.6016860899376817,
      "language_loss": 0.77811068,
      "learning_rate": 3.3726616295225774e-06,
      "loss": 0.79988343,
      "num_input_tokens_seen": 101046215,
      "step": 4681,
      "time_per_iteration": 2.485274076461792
    },
    {
      "auxiliary_loss_clip": 0.01125733,
      "auxiliary_loss_mlp": 0.01034027,
      "balance_loss_clip": 1.04688406,
      "balance_loss_mlp": 1.01751649,
      "epoch": 0.28149706899143245,
      "flos": 18515326872960.0,
      "grad_norm": 2.31702337665741,
      "language_loss": 0.73562539,
      "learning_rate": 3.372378352108146e-06,
      "loss": 0.75722301,
      "num_input_tokens_seen": 101063365,
      "step": 4682,
      "time_per_iteration": 2.4952099323272705
    },
    {
      "auxiliary_loss_clip": 0.01132415,
      "auxiliary_loss_mlp": 0.01036588,
      "balance_loss_clip": 1.04593492,
      "balance_loss_mlp": 1.02149618,
      "epoch": 0.2815571922441004,
      "flos": 24863255786880.0,
      "grad_norm": 1.4676256436812118,
      "language_loss": 0.80541968,
      "learning_rate": 3.3720950226529894e-06,
      "loss": 0.82710969,
      "num_input_tokens_seen": 101083835,
      "step": 4683,
      "time_per_iteration": 2.4969868659973145
    },
    {
      "auxiliary_loss_clip": 0.01083239,
      "auxiliary_loss_mlp": 0.0104281,
      "balance_loss_clip": 1.04940498,
      "balance_loss_mlp": 1.02556038,
      "epoch": 0.2816173154967684,
      "flos": 19901622326400.0,
      "grad_norm": 3.488422867779134,
      "language_loss": 0.76347148,
      "learning_rate": 3.371811641167852e-06,
      "loss": 0.78473198,
      "num_input_tokens_seen": 101101740,
      "step": 4684,
      "time_per_iteration": 2.5843777656555176
    },
    {
      "auxiliary_loss_clip": 0.01082383,
      "auxiliary_loss_mlp": 0.01039119,
      "balance_loss_clip": 1.04645956,
      "balance_loss_mlp": 1.02334714,
      "epoch": 0.28167743874943635,
      "flos": 17491333950720.0,
      "grad_norm": 1.7720763514361482,
      "language_loss": 0.76182789,
      "learning_rate": 3.3715282076634807e-06,
      "loss": 0.78304297,
      "num_input_tokens_seen": 101120480,
      "step": 4685,
      "time_per_iteration": 2.558091640472412
    },
    {
      "auxiliary_loss_clip": 0.01108652,
      "auxiliary_loss_mlp": 0.01040358,
      "balance_loss_clip": 1.04886866,
      "balance_loss_mlp": 1.02524209,
      "epoch": 0.2817375620021043,
      "flos": 25302120157440.0,
      "grad_norm": 1.41240106607788,
      "language_loss": 0.75407463,
      "learning_rate": 3.3712447221506218e-06,
      "loss": 0.77556467,
      "num_input_tokens_seen": 101142910,
      "step": 4686,
      "time_per_iteration": 2.582836389541626
    },
    {
      "auxiliary_loss_clip": 0.01107955,
      "auxiliary_loss_mlp": 0.01047543,
      "balance_loss_clip": 1.045753,
      "balance_loss_mlp": 1.03038836,
      "epoch": 0.2817976852547723,
      "flos": 18693227957760.0,
      "grad_norm": 2.383655571595673,
      "language_loss": 0.62683159,
      "learning_rate": 3.370961184640025e-06,
      "loss": 0.6483866,
      "num_input_tokens_seen": 101160030,
      "step": 4687,
      "time_per_iteration": 2.4957659244537354
    },
    {
      "auxiliary_loss_clip": 0.01114553,
      "auxiliary_loss_mlp": 0.01046096,
      "balance_loss_clip": 1.04716611,
      "balance_loss_mlp": 1.03037262,
      "epoch": 0.28185780850744024,
      "flos": 22742263549440.0,
      "grad_norm": 2.118160622745849,
      "language_loss": 0.76500183,
      "learning_rate": 3.3706775951424433e-06,
      "loss": 0.78660834,
      "num_input_tokens_seen": 101177675,
      "step": 4688,
      "time_per_iteration": 2.542945146560669
    },
    {
      "auxiliary_loss_clip": 0.01107846,
      "auxiliary_loss_mlp": 0.01038181,
      "balance_loss_clip": 1.05424023,
      "balance_loss_mlp": 1.02246368,
      "epoch": 0.2819179317601082,
      "flos": 14933919467520.0,
      "grad_norm": 1.9221042875729457,
      "language_loss": 0.78427291,
      "learning_rate": 3.37039395366863e-06,
      "loss": 0.8057332,
      "num_input_tokens_seen": 101192225,
      "step": 4689,
      "time_per_iteration": 2.527785301208496
    },
    {
      "auxiliary_loss_clip": 0.01097337,
      "auxiliary_loss_mlp": 0.0104165,
      "balance_loss_clip": 1.04753399,
      "balance_loss_mlp": 1.02481723,
      "epoch": 0.2819780550127762,
      "flos": 23145325038720.0,
      "grad_norm": 1.8845949920405425,
      "language_loss": 0.78016448,
      "learning_rate": 3.37011026022934e-06,
      "loss": 0.80155432,
      "num_input_tokens_seen": 101210870,
      "step": 4690,
      "time_per_iteration": 2.578328847885132
    },
    {
      "auxiliary_loss_clip": 0.01135913,
      "auxiliary_loss_mlp": 0.00820802,
      "balance_loss_clip": 1.04669869,
      "balance_loss_mlp": 1.06502342,
      "epoch": 0.28203817826544414,
      "flos": 21616356764160.0,
      "grad_norm": 1.773554879048875,
      "language_loss": 0.87545443,
      "learning_rate": 3.369826514835332e-06,
      "loss": 0.89502156,
      "num_input_tokens_seen": 101229965,
      "step": 4691,
      "time_per_iteration": 2.4609627723693848
    },
    {
      "auxiliary_loss_clip": 0.01114087,
      "auxiliary_loss_mlp": 0.01046591,
      "balance_loss_clip": 1.04700041,
      "balance_loss_mlp": 1.02894855,
      "epoch": 0.2820983015181121,
      "flos": 24026788794240.0,
      "grad_norm": 3.5106192650424615,
      "language_loss": 0.81812793,
      "learning_rate": 3.3695427174973654e-06,
      "loss": 0.83973467,
      "num_input_tokens_seen": 101250980,
      "step": 4692,
      "time_per_iteration": 2.558096170425415
    },
    {
      "auxiliary_loss_clip": 0.01106792,
      "auxiliary_loss_mlp": 0.01032129,
      "balance_loss_clip": 1.05386162,
      "balance_loss_mlp": 1.01550519,
      "epoch": 0.2821584247707801,
      "flos": 30007925976960.0,
      "grad_norm": 1.533300618251183,
      "language_loss": 0.74624848,
      "learning_rate": 3.3692588682262022e-06,
      "loss": 0.76763773,
      "num_input_tokens_seen": 101273335,
      "step": 4693,
      "time_per_iteration": 2.6105990409851074
    },
    {
      "auxiliary_loss_clip": 0.01102975,
      "auxiliary_loss_mlp": 0.01034081,
      "balance_loss_clip": 1.04758358,
      "balance_loss_mlp": 1.0176065,
      "epoch": 0.2822185480234481,
      "flos": 21396762967680.0,
      "grad_norm": 1.6477652061037509,
      "language_loss": 0.77372909,
      "learning_rate": 3.3689749670326046e-06,
      "loss": 0.79509968,
      "num_input_tokens_seen": 101292110,
      "step": 4694,
      "time_per_iteration": 2.5589561462402344
    },
    {
      "auxiliary_loss_clip": 0.01124383,
      "auxiliary_loss_mlp": 0.0104021,
      "balance_loss_clip": 1.04910028,
      "balance_loss_mlp": 1.02416396,
      "epoch": 0.28227867127611606,
      "flos": 27452809964160.0,
      "grad_norm": 1.9784256864313345,
      "language_loss": 0.67364204,
      "learning_rate": 3.3686910139273392e-06,
      "loss": 0.69528806,
      "num_input_tokens_seen": 101312815,
      "step": 4695,
      "time_per_iteration": 2.5303614139556885
    },
    {
      "auxiliary_loss_clip": 0.01120439,
      "auxiliary_loss_mlp": 0.0105031,
      "balance_loss_clip": 1.04867887,
      "balance_loss_mlp": 1.03201151,
      "epoch": 0.282338794528784,
      "flos": 22593736811520.0,
      "grad_norm": 2.5065640111625216,
      "language_loss": 0.75814497,
      "learning_rate": 3.3684070089211736e-06,
      "loss": 0.77985251,
      "num_input_tokens_seen": 101329045,
      "step": 4696,
      "time_per_iteration": 2.51064133644104
    },
    {
      "auxiliary_loss_clip": 0.01107956,
      "auxiliary_loss_mlp": 0.01042798,
      "balance_loss_clip": 1.05199051,
      "balance_loss_mlp": 1.02649617,
      "epoch": 0.282398917781452,
      "flos": 42010923386880.0,
      "grad_norm": 1.485280318069343,
      "language_loss": 0.6243881,
      "learning_rate": 3.368122952024877e-06,
      "loss": 0.6458956,
      "num_input_tokens_seen": 101352715,
      "step": 4697,
      "time_per_iteration": 2.7115960121154785
    },
    {
      "auxiliary_loss_clip": 0.01094036,
      "auxiliary_loss_mlp": 0.01033836,
      "balance_loss_clip": 1.04951119,
      "balance_loss_mlp": 1.01824379,
      "epoch": 0.28245904103411995,
      "flos": 23224724052480.0,
      "grad_norm": 2.0105385650668994,
      "language_loss": 0.73338199,
      "learning_rate": 3.3678388432492214e-06,
      "loss": 0.75466067,
      "num_input_tokens_seen": 101374640,
      "step": 4698,
      "time_per_iteration": 2.623849391937256
    },
    {
      "auxiliary_loss_clip": 0.01133746,
      "auxiliary_loss_mlp": 0.01040876,
      "balance_loss_clip": 1.04752636,
      "balance_loss_mlp": 1.02603436,
      "epoch": 0.2825191642867879,
      "flos": 25374623760000.0,
      "grad_norm": 1.7661338011154286,
      "language_loss": 0.75224173,
      "learning_rate": 3.3675546826049788e-06,
      "loss": 0.77398801,
      "num_input_tokens_seen": 101393595,
      "step": 4699,
      "time_per_iteration": 2.487856864929199
    },
    {
      "auxiliary_loss_clip": 0.01126496,
      "auxiliary_loss_mlp": 0.01036838,
      "balance_loss_clip": 1.04680598,
      "balance_loss_mlp": 1.01949322,
      "epoch": 0.2825792875394559,
      "flos": 17236799199360.0,
      "grad_norm": 2.5361200078159456,
      "language_loss": 0.79775268,
      "learning_rate": 3.3672704701029265e-06,
      "loss": 0.81938601,
      "num_input_tokens_seen": 101409265,
      "step": 4700,
      "time_per_iteration": 2.475954055786133
    },
    {
      "auxiliary_loss_clip": 0.01113311,
      "auxiliary_loss_mlp": 0.01041458,
      "balance_loss_clip": 1.05102527,
      "balance_loss_mlp": 1.02766001,
      "epoch": 0.28263941079212385,
      "flos": 26723967096960.0,
      "grad_norm": 1.7845390147533655,
      "language_loss": 0.81643629,
      "learning_rate": 3.3669862057538402e-06,
      "loss": 0.83798397,
      "num_input_tokens_seen": 101428365,
      "step": 4701,
      "time_per_iteration": 2.5704147815704346
    },
    {
      "auxiliary_loss_clip": 0.01068657,
      "auxiliary_loss_mlp": 0.01039108,
      "balance_loss_clip": 1.04364395,
      "balance_loss_mlp": 1.02336049,
      "epoch": 0.2826995340447918,
      "flos": 25921327737600.0,
      "grad_norm": 1.9722216417776959,
      "language_loss": 0.73069978,
      "learning_rate": 3.3667018895685004e-06,
      "loss": 0.75177741,
      "num_input_tokens_seen": 101447280,
      "step": 4702,
      "time_per_iteration": 2.706874370574951
    },
    {
      "auxiliary_loss_clip": 0.0113728,
      "auxiliary_loss_mlp": 0.01040513,
      "balance_loss_clip": 1.0504545,
      "balance_loss_mlp": 1.02455115,
      "epoch": 0.2827596572974598,
      "flos": 22379709623040.0,
      "grad_norm": 1.7354510846689797,
      "language_loss": 0.7829591,
      "learning_rate": 3.3664175215576886e-06,
      "loss": 0.80473703,
      "num_input_tokens_seen": 101465435,
      "step": 4703,
      "time_per_iteration": 2.4871652126312256
    },
    {
      "auxiliary_loss_clip": 0.01106803,
      "auxiliary_loss_mlp": 0.01043875,
      "balance_loss_clip": 1.04510415,
      "balance_loss_mlp": 1.02731669,
      "epoch": 0.28281978055012774,
      "flos": 33547137880320.0,
      "grad_norm": 1.522973924737269,
      "language_loss": 0.69374758,
      "learning_rate": 3.3661331017321867e-06,
      "loss": 0.71525437,
      "num_input_tokens_seen": 101486355,
      "step": 4704,
      "time_per_iteration": 2.6305463314056396
    },
    {
      "auxiliary_loss_clip": 0.0110109,
      "auxiliary_loss_mlp": 0.01038708,
      "balance_loss_clip": 1.04865193,
      "balance_loss_mlp": 1.02173281,
      "epoch": 0.2828799038027957,
      "flos": 23440870143360.0,
      "grad_norm": 1.915696614425445,
      "language_loss": 0.70383757,
      "learning_rate": 3.3658486301027807e-06,
      "loss": 0.72523558,
      "num_input_tokens_seen": 101505875,
      "step": 4705,
      "time_per_iteration": 4.070115089416504
    },
    {
      "auxiliary_loss_clip": 0.01046845,
      "auxiliary_loss_mlp": 0.01001626,
      "balance_loss_clip": 1.02562094,
      "balance_loss_mlp": 0.99970675,
      "epoch": 0.2829400270554637,
      "flos": 69873690251520.0,
      "grad_norm": 0.7242747442982526,
      "language_loss": 0.59198987,
      "learning_rate": 3.3655641066802577e-06,
      "loss": 0.61247456,
      "num_input_tokens_seen": 101565045,
      "step": 4706,
      "time_per_iteration": 3.142594814300537
    },
    {
      "auxiliary_loss_clip": 0.01108345,
      "auxiliary_loss_mlp": 0.01038592,
      "balance_loss_clip": 1.04421616,
      "balance_loss_mlp": 1.02442372,
      "epoch": 0.2830001503081317,
      "flos": 24789028331520.0,
      "grad_norm": 1.4224633930716837,
      "language_loss": 0.82220376,
      "learning_rate": 3.365279531475407e-06,
      "loss": 0.84367311,
      "num_input_tokens_seen": 101585825,
      "step": 4707,
      "time_per_iteration": 2.5846662521362305
    },
    {
      "auxiliary_loss_clip": 0.01115389,
      "auxiliary_loss_mlp": 0.01035427,
      "balance_loss_clip": 1.04521322,
      "balance_loss_mlp": 1.01896417,
      "epoch": 0.28306027356079966,
      "flos": 27669387018240.0,
      "grad_norm": 1.7313707110935654,
      "language_loss": 0.8031528,
      "learning_rate": 3.36499490449902e-06,
      "loss": 0.82466096,
      "num_input_tokens_seen": 101606105,
      "step": 4708,
      "time_per_iteration": 2.608454704284668
    },
    {
      "auxiliary_loss_clip": 0.01037216,
      "auxiliary_loss_mlp": 0.01002154,
      "balance_loss_clip": 1.02806938,
      "balance_loss_mlp": 0.99988872,
      "epoch": 0.2831203968134676,
      "flos": 60527938199040.0,
      "grad_norm": 0.8857249519734655,
      "language_loss": 0.62818146,
      "learning_rate": 3.3647102257618895e-06,
      "loss": 0.64857513,
      "num_input_tokens_seen": 101656875,
      "step": 4709,
      "time_per_iteration": 4.357207536697388
    },
    {
      "auxiliary_loss_clip": 0.01108642,
      "auxiliary_loss_mlp": 0.01047613,
      "balance_loss_clip": 1.04767466,
      "balance_loss_mlp": 1.03005326,
      "epoch": 0.2831805200661356,
      "flos": 22054790171520.0,
      "grad_norm": 1.356831537212952,
      "language_loss": 0.74126446,
      "learning_rate": 3.3644254952748103e-06,
      "loss": 0.76282704,
      "num_input_tokens_seen": 101676225,
      "step": 4710,
      "time_per_iteration": 2.5286450386047363
    },
    {
      "auxiliary_loss_clip": 0.01108941,
      "auxiliary_loss_mlp": 0.01060804,
      "balance_loss_clip": 1.04743147,
      "balance_loss_mlp": 1.04306543,
      "epoch": 0.28324064331880355,
      "flos": 22600668136320.0,
      "grad_norm": 1.7897364385047985,
      "language_loss": 0.79235715,
      "learning_rate": 3.364140713048579e-06,
      "loss": 0.81405461,
      "num_input_tokens_seen": 101693710,
      "step": 4711,
      "time_per_iteration": 2.5403249263763428
    },
    {
      "auxiliary_loss_clip": 0.01129004,
      "auxiliary_loss_mlp": 0.00802997,
      "balance_loss_clip": 1.0504998,
      "balance_loss_mlp": 1.0295794,
      "epoch": 0.2833007665714715,
      "flos": 30404127968640.0,
      "grad_norm": 2.015063467950678,
      "language_loss": 0.71361828,
      "learning_rate": 3.363855879093996e-06,
      "loss": 0.73293829,
      "num_input_tokens_seen": 101714010,
      "step": 4712,
      "time_per_iteration": 3.9620487689971924
    },
    {
      "auxiliary_loss_clip": 0.01137962,
      "auxiliary_loss_mlp": 0.01052891,
      "balance_loss_clip": 1.0485605,
      "balance_loss_mlp": 1.03589201,
      "epoch": 0.2833608898241395,
      "flos": 23549499849600.0,
      "grad_norm": 1.9333888296346071,
      "language_loss": 0.81510329,
      "learning_rate": 3.3635709934218605e-06,
      "loss": 0.83701181,
      "num_input_tokens_seen": 101732995,
      "step": 4713,
      "time_per_iteration": 3.8527326583862305
    },
    {
      "auxiliary_loss_clip": 0.01115652,
      "auxiliary_loss_mlp": 0.01041132,
      "balance_loss_clip": 1.04942465,
      "balance_loss_mlp": 1.02419233,
      "epoch": 0.28342101307680745,
      "flos": 20266726118400.0,
      "grad_norm": 1.7018411304733572,
      "language_loss": 0.75099349,
      "learning_rate": 3.3632860560429766e-06,
      "loss": 0.77256131,
      "num_input_tokens_seen": 101751385,
      "step": 4714,
      "time_per_iteration": 2.524751901626587
    },
    {
      "auxiliary_loss_clip": 0.01122886,
      "auxiliary_loss_mlp": 0.01047364,
      "balance_loss_clip": 1.0474565,
      "balance_loss_mlp": 1.03168809,
      "epoch": 0.2834811363294754,
      "flos": 30847050576000.0,
      "grad_norm": 1.3818759164232401,
      "language_loss": 0.78353739,
      "learning_rate": 3.3630010669681494e-06,
      "loss": 0.80523992,
      "num_input_tokens_seen": 101773825,
      "step": 4715,
      "time_per_iteration": 2.5830488204956055
    },
    {
      "auxiliary_loss_clip": 0.01113898,
      "auxiliary_loss_mlp": 0.01039894,
      "balance_loss_clip": 1.04743803,
      "balance_loss_mlp": 1.02382493,
      "epoch": 0.2835412595821434,
      "flos": 22711021695360.0,
      "grad_norm": 2.1178468371219052,
      "language_loss": 0.73809367,
      "learning_rate": 3.3627160262081845e-06,
      "loss": 0.75963157,
      "num_input_tokens_seen": 101791920,
      "step": 4716,
      "time_per_iteration": 2.5343379974365234
    },
    {
      "auxiliary_loss_clip": 0.01110655,
      "auxiliary_loss_mlp": 0.01051172,
      "balance_loss_clip": 1.04374957,
      "balance_loss_mlp": 1.03195524,
      "epoch": 0.28360138283481134,
      "flos": 18077719478400.0,
      "grad_norm": 2.2364903333020583,
      "language_loss": 0.7421577,
      "learning_rate": 3.3624309337738917e-06,
      "loss": 0.76377606,
      "num_input_tokens_seen": 101809515,
      "step": 4717,
      "time_per_iteration": 2.515613317489624
    },
    {
      "auxiliary_loss_clip": 0.01106143,
      "auxiliary_loss_mlp": 0.01043526,
      "balance_loss_clip": 1.04614985,
      "balance_loss_mlp": 1.02755249,
      "epoch": 0.2836615060874793,
      "flos": 17854785717120.0,
      "grad_norm": 1.4271762280587414,
      "language_loss": 0.67124969,
      "learning_rate": 3.3621457896760813e-06,
      "loss": 0.69274634,
      "num_input_tokens_seen": 101827735,
      "step": 4718,
      "time_per_iteration": 2.5307796001434326
    },
    {
      "auxiliary_loss_clip": 0.01115475,
      "auxiliary_loss_mlp": 0.01043642,
      "balance_loss_clip": 1.0456841,
      "balance_loss_mlp": 1.02697623,
      "epoch": 0.2837216293401473,
      "flos": 25740302169600.0,
      "grad_norm": 1.787453020258257,
      "language_loss": 0.7235682,
      "learning_rate": 3.361860593925566e-06,
      "loss": 0.74515939,
      "num_input_tokens_seen": 101845970,
      "step": 4719,
      "time_per_iteration": 2.562986373901367
    },
    {
      "auxiliary_loss_clip": 0.01124225,
      "auxiliary_loss_mlp": 0.01045924,
      "balance_loss_clip": 1.04766309,
      "balance_loss_mlp": 1.03020036,
      "epoch": 0.2837817525928153,
      "flos": 20923532259840.0,
      "grad_norm": 1.4876309566993318,
      "language_loss": 0.80460107,
      "learning_rate": 3.3615753465331605e-06,
      "loss": 0.82630259,
      "num_input_tokens_seen": 101865040,
      "step": 4720,
      "time_per_iteration": 2.515974521636963
    },
    {
      "auxiliary_loss_clip": 0.01124894,
      "auxiliary_loss_mlp": 0.01042739,
      "balance_loss_clip": 1.04734015,
      "balance_loss_mlp": 1.02576399,
      "epoch": 0.28384187584548326,
      "flos": 18916700423040.0,
      "grad_norm": 1.891617944669613,
      "language_loss": 0.794743,
      "learning_rate": 3.3612900475096817e-06,
      "loss": 0.8164193,
      "num_input_tokens_seen": 101883735,
      "step": 4721,
      "time_per_iteration": 2.4790003299713135
    },
    {
      "auxiliary_loss_clip": 0.01085203,
      "auxiliary_loss_mlp": 0.00797088,
      "balance_loss_clip": 1.04421127,
      "balance_loss_mlp": 1.01730049,
      "epoch": 0.2839019990981512,
      "flos": 27343964776320.0,
      "grad_norm": 1.8005827840091124,
      "language_loss": 0.82781452,
      "learning_rate": 3.3610046968659474e-06,
      "loss": 0.84663749,
      "num_input_tokens_seen": 101903025,
      "step": 4722,
      "time_per_iteration": 2.6383211612701416
    },
    {
      "auxiliary_loss_clip": 0.0113804,
      "auxiliary_loss_mlp": 0.01040662,
      "balance_loss_clip": 1.04901862,
      "balance_loss_mlp": 1.02479529,
      "epoch": 0.2839621223508192,
      "flos": 18114312458880.0,
      "grad_norm": 2.261307870988022,
      "language_loss": 0.70257539,
      "learning_rate": 3.3607192946127785e-06,
      "loss": 0.72436243,
      "num_input_tokens_seen": 101922255,
      "step": 4723,
      "time_per_iteration": 2.4621448516845703
    },
    {
      "auxiliary_loss_clip": 0.01107496,
      "auxiliary_loss_mlp": 0.01040536,
      "balance_loss_clip": 1.04707062,
      "balance_loss_mlp": 1.02292848,
      "epoch": 0.28402224560348716,
      "flos": 26358360514560.0,
      "grad_norm": 1.3865925725096342,
      "language_loss": 0.78444624,
      "learning_rate": 3.360433840760998e-06,
      "loss": 0.8059265,
      "num_input_tokens_seen": 101943100,
      "step": 4724,
      "time_per_iteration": 2.589341640472412
    },
    {
      "auxiliary_loss_clip": 0.01108982,
      "auxiliary_loss_mlp": 0.01051436,
      "balance_loss_clip": 1.0454309,
      "balance_loss_mlp": 1.03399611,
      "epoch": 0.2840823688561551,
      "flos": 24060795995520.0,
      "grad_norm": 1.5028282835718851,
      "language_loss": 0.92422891,
      "learning_rate": 3.36014833532143e-06,
      "loss": 0.94583309,
      "num_input_tokens_seen": 101963160,
      "step": 4725,
      "time_per_iteration": 2.5751779079437256
    },
    {
      "auxiliary_loss_clip": 0.01130655,
      "auxiliary_loss_mlp": 0.01042074,
      "balance_loss_clip": 1.04940116,
      "balance_loss_mlp": 1.02546811,
      "epoch": 0.2841424921088231,
      "flos": 29459821368960.0,
      "grad_norm": 1.4967415413497736,
      "language_loss": 0.88623804,
      "learning_rate": 3.3598627783049e-06,
      "loss": 0.90796536,
      "num_input_tokens_seen": 101984300,
      "step": 4726,
      "time_per_iteration": 2.578953266143799
    },
    {
      "auxiliary_loss_clip": 0.01127487,
      "auxiliary_loss_mlp": 0.01041638,
      "balance_loss_clip": 1.05270338,
      "balance_loss_mlp": 1.02503181,
      "epoch": 0.28420261536149105,
      "flos": 48100367053440.0,
      "grad_norm": 1.9979955827616438,
      "language_loss": 0.78705853,
      "learning_rate": 3.359577169722238e-06,
      "loss": 0.80874979,
      "num_input_tokens_seen": 102005765,
      "step": 4727,
      "time_per_iteration": 2.732815742492676
    },
    {
      "auxiliary_loss_clip": 0.01125111,
      "auxiliary_loss_mlp": 0.01038933,
      "balance_loss_clip": 1.04780173,
      "balance_loss_mlp": 1.02376997,
      "epoch": 0.284262738614159,
      "flos": 25666146541440.0,
      "grad_norm": 2.2446077694076565,
      "language_loss": 0.66971147,
      "learning_rate": 3.3592915095842733e-06,
      "loss": 0.69135189,
      "num_input_tokens_seen": 102022755,
      "step": 4728,
      "time_per_iteration": 2.526355266571045
    },
    {
      "auxiliary_loss_clip": 0.01100015,
      "auxiliary_loss_mlp": 0.01046007,
      "balance_loss_clip": 1.04468679,
      "balance_loss_mlp": 1.02922189,
      "epoch": 0.284322861866827,
      "flos": 19718980646400.0,
      "grad_norm": 1.6615122135184746,
      "language_loss": 0.76435685,
      "learning_rate": 3.3590057979018386e-06,
      "loss": 0.78581703,
      "num_input_tokens_seen": 102041850,
      "step": 4729,
      "time_per_iteration": 2.5324559211730957
    },
    {
      "auxiliary_loss_clip": 0.01117082,
      "auxiliary_loss_mlp": 0.01047722,
      "balance_loss_clip": 1.05305576,
      "balance_loss_mlp": 1.03093743,
      "epoch": 0.28438298511949495,
      "flos": 23915250086400.0,
      "grad_norm": 1.6377745214442347,
      "language_loss": 0.66508847,
      "learning_rate": 3.3587200346857674e-06,
      "loss": 0.68673652,
      "num_input_tokens_seen": 102059500,
      "step": 4730,
      "time_per_iteration": 2.5561037063598633
    },
    {
      "auxiliary_loss_clip": 0.0111587,
      "auxiliary_loss_mlp": 0.01035587,
      "balance_loss_clip": 1.04831815,
      "balance_loss_mlp": 1.01863527,
      "epoch": 0.2844431083721629,
      "flos": 26067340523520.0,
      "grad_norm": 1.8494322909263117,
      "language_loss": 0.74553567,
      "learning_rate": 3.3584342199468965e-06,
      "loss": 0.76705027,
      "num_input_tokens_seen": 102080460,
      "step": 4731,
      "time_per_iteration": 2.569830894470215
    },
    {
      "auxiliary_loss_clip": 0.01098977,
      "auxiliary_loss_mlp": 0.0103818,
      "balance_loss_clip": 1.05490744,
      "balance_loss_mlp": 1.02212214,
      "epoch": 0.2845032316248309,
      "flos": 25810435474560.0,
      "grad_norm": 1.502885318910816,
      "language_loss": 0.83617097,
      "learning_rate": 3.3581483536960638e-06,
      "loss": 0.85754251,
      "num_input_tokens_seen": 102100950,
      "step": 4732,
      "time_per_iteration": 2.6337873935699463
    },
    {
      "auxiliary_loss_clip": 0.01130291,
      "auxiliary_loss_mlp": 0.01050253,
      "balance_loss_clip": 1.05007625,
      "balance_loss_mlp": 1.03226411,
      "epoch": 0.2845633548774989,
      "flos": 19823192979840.0,
      "grad_norm": 1.570658103313605,
      "language_loss": 0.79042763,
      "learning_rate": 3.357862435944109e-06,
      "loss": 0.81223309,
      "num_input_tokens_seen": 102119345,
      "step": 4733,
      "time_per_iteration": 2.491194248199463
    },
    {
      "auxiliary_loss_clip": 0.01146298,
      "auxiliary_loss_mlp": 0.01053606,
      "balance_loss_clip": 1.05172348,
      "balance_loss_mlp": 1.03618932,
      "epoch": 0.28462347813016686,
      "flos": 23182815859200.0,
      "grad_norm": 2.765358066672143,
      "language_loss": 0.71167308,
      "learning_rate": 3.357576466701875e-06,
      "loss": 0.73367214,
      "num_input_tokens_seen": 102139050,
      "step": 4734,
      "time_per_iteration": 2.5084898471832275
    },
    {
      "auxiliary_loss_clip": 0.01117554,
      "auxiliary_loss_mlp": 0.01037935,
      "balance_loss_clip": 1.0494473,
      "balance_loss_mlp": 1.02160299,
      "epoch": 0.2846836013828348,
      "flos": 18660477732480.0,
      "grad_norm": 1.7565704609117334,
      "language_loss": 0.74026042,
      "learning_rate": 3.3572904459802056e-06,
      "loss": 0.76181537,
      "num_input_tokens_seen": 102157935,
      "step": 4735,
      "time_per_iteration": 2.516871690750122
    },
    {
      "auxiliary_loss_clip": 0.01119281,
      "auxiliary_loss_mlp": 0.01043626,
      "balance_loss_clip": 1.05183434,
      "balance_loss_mlp": 1.02800918,
      "epoch": 0.2847437246355028,
      "flos": 14173511523840.0,
      "grad_norm": 1.8301074254338343,
      "language_loss": 0.79881096,
      "learning_rate": 3.357004373789946e-06,
      "loss": 0.82044005,
      "num_input_tokens_seen": 102175325,
      "step": 4736,
      "time_per_iteration": 2.522127866744995
    },
    {
      "auxiliary_loss_clip": 0.01142373,
      "auxiliary_loss_mlp": 0.0104719,
      "balance_loss_clip": 1.05098569,
      "balance_loss_mlp": 1.02941585,
      "epoch": 0.28480384788817076,
      "flos": 29278364837760.0,
      "grad_norm": 2.586107812262861,
      "language_loss": 0.59716618,
      "learning_rate": 3.3567182501419453e-06,
      "loss": 0.61906189,
      "num_input_tokens_seen": 102196625,
      "step": 4737,
      "time_per_iteration": 2.5250797271728516
    },
    {
      "auxiliary_loss_clip": 0.01124903,
      "auxiliary_loss_mlp": 0.01041208,
      "balance_loss_clip": 1.04805982,
      "balance_loss_mlp": 1.02462602,
      "epoch": 0.2848639711408387,
      "flos": 22601314581120.0,
      "grad_norm": 1.8761504956160868,
      "language_loss": 0.86821061,
      "learning_rate": 3.356432075047052e-06,
      "loss": 0.88987172,
      "num_input_tokens_seen": 102214975,
      "step": 4738,
      "time_per_iteration": 2.5215678215026855
    },
    {
      "auxiliary_loss_clip": 0.01116746,
      "auxiliary_loss_mlp": 0.01044634,
      "balance_loss_clip": 1.05256605,
      "balance_loss_mlp": 1.02564359,
      "epoch": 0.2849240943935067,
      "flos": 17599460866560.0,
      "grad_norm": 1.8981088897526033,
      "language_loss": 0.90005398,
      "learning_rate": 3.356145848516118e-06,
      "loss": 0.92166775,
      "num_input_tokens_seen": 102231885,
      "step": 4739,
      "time_per_iteration": 2.5362894535064697
    },
    {
      "auxiliary_loss_clip": 0.01133041,
      "auxiliary_loss_mlp": 0.01044793,
      "balance_loss_clip": 1.05736172,
      "balance_loss_mlp": 1.02806759,
      "epoch": 0.28498421764617465,
      "flos": 24862573428480.0,
      "grad_norm": 1.378518139615513,
      "language_loss": 0.72308254,
      "learning_rate": 3.355859570559998e-06,
      "loss": 0.74486089,
      "num_input_tokens_seen": 102252725,
      "step": 4740,
      "time_per_iteration": 2.5610649585723877
    },
    {
      "auxiliary_loss_clip": 0.0111576,
      "auxiliary_loss_mlp": 0.01038,
      "balance_loss_clip": 1.04853237,
      "balance_loss_mlp": 1.02159643,
      "epoch": 0.2850443408988426,
      "flos": 22782555630720.0,
      "grad_norm": 1.4968429676091495,
      "language_loss": 0.77567494,
      "learning_rate": 3.3555732411895477e-06,
      "loss": 0.79721248,
      "num_input_tokens_seen": 102271730,
      "step": 4741,
      "time_per_iteration": 2.5290775299072266
    },
    {
      "auxiliary_loss_clip": 0.01105816,
      "auxiliary_loss_mlp": 0.01044735,
      "balance_loss_clip": 1.04954851,
      "balance_loss_mlp": 1.02693701,
      "epoch": 0.2851044641515106,
      "flos": 18844053166080.0,
      "grad_norm": 1.794811285141452,
      "language_loss": 0.76328015,
      "learning_rate": 3.3552868604156235e-06,
      "loss": 0.78478563,
      "num_input_tokens_seen": 102291325,
      "step": 4742,
      "time_per_iteration": 2.573063373565674
    },
    {
      "auxiliary_loss_clip": 0.01147761,
      "auxiliary_loss_mlp": 0.0105458,
      "balance_loss_clip": 1.05183744,
      "balance_loss_mlp": 1.03474402,
      "epoch": 0.28516458740417855,
      "flos": 18880502492160.0,
      "grad_norm": 1.8140698017504322,
      "language_loss": 0.57834566,
      "learning_rate": 3.355000428249086e-06,
      "loss": 0.60036904,
      "num_input_tokens_seen": 102309000,
      "step": 4743,
      "time_per_iteration": 3.9263756275177
    },
    {
      "auxiliary_loss_clip": 0.01111528,
      "auxiliary_loss_mlp": 0.01053832,
      "balance_loss_clip": 1.05152857,
      "balance_loss_mlp": 1.03597403,
      "epoch": 0.2852247106568465,
      "flos": 25299821687040.0,
      "grad_norm": 1.6190569928254324,
      "language_loss": 0.74562025,
      "learning_rate": 3.354713944700797e-06,
      "loss": 0.76727378,
      "num_input_tokens_seen": 102329240,
      "step": 4744,
      "time_per_iteration": 2.591512441635132
    },
    {
      "auxiliary_loss_clip": 0.01127608,
      "auxiliary_loss_mlp": 0.01042876,
      "balance_loss_clip": 1.05403781,
      "balance_loss_mlp": 1.02658033,
      "epoch": 0.2852848339095145,
      "flos": 11655383541120.0,
      "grad_norm": 2.1556536735844958,
      "language_loss": 0.77555263,
      "learning_rate": 3.3544274097816185e-06,
      "loss": 0.79725748,
      "num_input_tokens_seen": 102344440,
      "step": 4745,
      "time_per_iteration": 2.4967527389526367
    },
    {
      "auxiliary_loss_clip": 0.01119628,
      "auxiliary_loss_mlp": 0.01041223,
      "balance_loss_clip": 1.05126548,
      "balance_loss_mlp": 1.0254035,
      "epoch": 0.2853449571621825,
      "flos": 12933228856320.0,
      "grad_norm": 1.7574636648395323,
      "language_loss": 0.82555014,
      "learning_rate": 3.3541408235024173e-06,
      "loss": 0.84715867,
      "num_input_tokens_seen": 102360985,
      "step": 4746,
      "time_per_iteration": 2.455691337585449
    },
    {
      "auxiliary_loss_clip": 0.01096621,
      "auxiliary_loss_mlp": 0.01040639,
      "balance_loss_clip": 1.04853284,
      "balance_loss_mlp": 1.02260303,
      "epoch": 0.28540508041485046,
      "flos": 20010575255040.0,
      "grad_norm": 1.7242825725443152,
      "language_loss": 0.79685938,
      "learning_rate": 3.3538541858740604e-06,
      "loss": 0.81823194,
      "num_input_tokens_seen": 102380320,
      "step": 4747,
      "time_per_iteration": 3.998424530029297
    },
    {
      "auxiliary_loss_clip": 0.0105427,
      "auxiliary_loss_mlp": 0.00998739,
      "balance_loss_clip": 1.04021525,
      "balance_loss_mlp": 0.99655706,
      "epoch": 0.28546520366751843,
      "flos": 68139349966080.0,
      "grad_norm": 0.7803066287920077,
      "language_loss": 0.60478145,
      "learning_rate": 3.3535674969074173e-06,
      "loss": 0.62531155,
      "num_input_tokens_seen": 102439140,
      "step": 4748,
      "time_per_iteration": 3.059692859649658
    },
    {
      "auxiliary_loss_clip": 0.01141742,
      "auxiliary_loss_mlp": 0.01045095,
      "balance_loss_clip": 1.0509851,
      "balance_loss_mlp": 1.02791691,
      "epoch": 0.2855253269201864,
      "flos": 13251540205440.0,
      "grad_norm": 2.1477024829352995,
      "language_loss": 0.80526704,
      "learning_rate": 3.3532807566133592e-06,
      "loss": 0.82713538,
      "num_input_tokens_seen": 102450990,
      "step": 4749,
      "time_per_iteration": 2.4262030124664307
    },
    {
      "auxiliary_loss_clip": 0.01128148,
      "auxiliary_loss_mlp": 0.01038797,
      "balance_loss_clip": 1.04833663,
      "balance_loss_mlp": 1.02245343,
      "epoch": 0.28558545017285436,
      "flos": 28620876337920.0,
      "grad_norm": 2.0494126843266978,
      "language_loss": 0.70746493,
      "learning_rate": 3.3529939650027587e-06,
      "loss": 0.72913432,
      "num_input_tokens_seen": 102471820,
      "step": 4750,
      "time_per_iteration": 2.546436309814453
    },
    {
      "auxiliary_loss_clip": 0.01125916,
      "auxiliary_loss_mlp": 0.01037574,
      "balance_loss_clip": 1.05005288,
      "balance_loss_mlp": 1.02033651,
      "epoch": 0.2856455734255223,
      "flos": 34130470752000.0,
      "grad_norm": 1.5201751613753034,
      "language_loss": 0.82011378,
      "learning_rate": 3.3527071220864917e-06,
      "loss": 0.84174865,
      "num_input_tokens_seen": 102492625,
      "step": 4751,
      "time_per_iteration": 3.978419542312622
    },
    {
      "auxiliary_loss_clip": 0.01137419,
      "auxiliary_loss_mlp": 0.01043643,
      "balance_loss_clip": 1.04902041,
      "balance_loss_mlp": 1.02720404,
      "epoch": 0.2857056966781903,
      "flos": 39786149779200.0,
      "grad_norm": 1.8596793578360746,
      "language_loss": 0.79753631,
      "learning_rate": 3.3524202278754353e-06,
      "loss": 0.81934696,
      "num_input_tokens_seen": 102514145,
      "step": 4752,
      "time_per_iteration": 3.9630253314971924
    },
    {
      "auxiliary_loss_clip": 0.01123437,
      "auxiliary_loss_mlp": 0.01041,
      "balance_loss_clip": 1.04618716,
      "balance_loss_mlp": 1.02333331,
      "epoch": 0.28576581993085826,
      "flos": 21872292145920.0,
      "grad_norm": 3.4673473915228237,
      "language_loss": 0.78727859,
      "learning_rate": 3.3521332823804676e-06,
      "loss": 0.80892301,
      "num_input_tokens_seen": 102532365,
      "step": 4753,
      "time_per_iteration": 2.495924711227417
    },
    {
      "auxiliary_loss_clip": 0.0114203,
      "auxiliary_loss_mlp": 0.01046793,
      "balance_loss_clip": 1.04894924,
      "balance_loss_mlp": 1.02838707,
      "epoch": 0.2858259431835262,
      "flos": 19091656592640.0,
      "grad_norm": 2.2123475876088414,
      "language_loss": 0.89530432,
      "learning_rate": 3.3518462856124704e-06,
      "loss": 0.91719252,
      "num_input_tokens_seen": 102548425,
      "step": 4754,
      "time_per_iteration": 2.442079782485962
    },
    {
      "auxiliary_loss_clip": 0.01122922,
      "auxiliary_loss_mlp": 0.01044951,
      "balance_loss_clip": 1.04663658,
      "balance_loss_mlp": 1.02917945,
      "epoch": 0.2858860664361942,
      "flos": 20334309557760.0,
      "grad_norm": 1.6517408929846047,
      "language_loss": 0.82434464,
      "learning_rate": 3.3515592375823267e-06,
      "loss": 0.84602338,
      "num_input_tokens_seen": 102566370,
      "step": 4755,
      "time_per_iteration": 2.4942989349365234
    },
    {
      "auxiliary_loss_clip": 0.01091851,
      "auxiliary_loss_mlp": 0.01045617,
      "balance_loss_clip": 1.04487181,
      "balance_loss_mlp": 1.02874875,
      "epoch": 0.28594618968886215,
      "flos": 24461738582400.0,
      "grad_norm": 1.5061533913788643,
      "language_loss": 0.83860207,
      "learning_rate": 3.351272138300922e-06,
      "loss": 0.85997677,
      "num_input_tokens_seen": 102588715,
      "step": 4756,
      "time_per_iteration": 2.6073732376098633
    },
    {
      "auxiliary_loss_clip": 0.01025286,
      "auxiliary_loss_mlp": 0.01045071,
      "balance_loss_clip": 1.02326345,
      "balance_loss_mlp": 1.04293716,
      "epoch": 0.2860063129415301,
      "flos": 71652850709760.0,
      "grad_norm": 0.8717787889320826,
      "language_loss": 0.61005175,
      "learning_rate": 3.350984987779142e-06,
      "loss": 0.63075531,
      "num_input_tokens_seen": 102656715,
      "step": 4757,
      "time_per_iteration": 3.294732093811035
    },
    {
      "auxiliary_loss_clip": 0.01138982,
      "auxiliary_loss_mlp": 0.01039281,
      "balance_loss_clip": 1.05072808,
      "balance_loss_mlp": 1.02317548,
      "epoch": 0.2860664361941981,
      "flos": 20558679863040.0,
      "grad_norm": 2.3340868643179875,
      "language_loss": 0.65656912,
      "learning_rate": 3.3506977860278756e-06,
      "loss": 0.67835176,
      "num_input_tokens_seen": 102676545,
      "step": 4758,
      "time_per_iteration": 2.456272602081299
    },
    {
      "auxiliary_loss_clip": 0.01127335,
      "auxiliary_loss_mlp": 0.0104159,
      "balance_loss_clip": 1.04701948,
      "balance_loss_mlp": 1.02526999,
      "epoch": 0.2861265594468661,
      "flos": 35996389534080.0,
      "grad_norm": 1.7138502004116387,
      "language_loss": 0.63186759,
      "learning_rate": 3.3504105330580143e-06,
      "loss": 0.65355682,
      "num_input_tokens_seen": 102702875,
      "step": 4759,
      "time_per_iteration": 2.651405096054077
    },
    {
      "auxiliary_loss_clip": 0.01121616,
      "auxiliary_loss_mlp": 0.00808024,
      "balance_loss_clip": 1.04575109,
      "balance_loss_mlp": 1.03912568,
      "epoch": 0.28618668269953407,
      "flos": 20047419630720.0,
      "grad_norm": 1.7105191328054878,
      "language_loss": 0.74347973,
      "learning_rate": 3.3501232288804496e-06,
      "loss": 0.76277614,
      "num_input_tokens_seen": 102723160,
      "step": 4760,
      "time_per_iteration": 2.503000020980835
    },
    {
      "auxiliary_loss_clip": 0.01111423,
      "auxiliary_loss_mlp": 0.01037704,
      "balance_loss_clip": 1.04649949,
      "balance_loss_mlp": 1.02224302,
      "epoch": 0.28624680595220203,
      "flos": 24971849579520.0,
      "grad_norm": 1.8335614924683528,
      "language_loss": 0.72312033,
      "learning_rate": 3.3498358735060773e-06,
      "loss": 0.74461162,
      "num_input_tokens_seen": 102743855,
      "step": 4761,
      "time_per_iteration": 2.5940070152282715
    },
    {
      "auxiliary_loss_clip": 0.01071955,
      "auxiliary_loss_mlp": 0.01051225,
      "balance_loss_clip": 1.04343772,
      "balance_loss_mlp": 1.03408265,
      "epoch": 0.28630692920487,
      "flos": 22492253911680.0,
      "grad_norm": 1.9268016941546922,
      "language_loss": 0.74752426,
      "learning_rate": 3.349548466945793e-06,
      "loss": 0.76875603,
      "num_input_tokens_seen": 102761370,
      "step": 4762,
      "time_per_iteration": 2.64593505859375
    },
    {
      "auxiliary_loss_clip": 0.01104164,
      "auxiliary_loss_mlp": 0.01047436,
      "balance_loss_clip": 1.04861796,
      "balance_loss_mlp": 1.03081822,
      "epoch": 0.28636705245753796,
      "flos": 21249888255360.0,
      "grad_norm": 1.4706477794068227,
      "language_loss": 0.76228422,
      "learning_rate": 3.349261009210496e-06,
      "loss": 0.78380024,
      "num_input_tokens_seen": 102780885,
      "step": 4763,
      "time_per_iteration": 2.5804033279418945
    },
    {
      "auxiliary_loss_clip": 0.01098643,
      "auxiliary_loss_mlp": 0.01040866,
      "balance_loss_clip": 1.04492235,
      "balance_loss_mlp": 1.0231154,
      "epoch": 0.28642717571020593,
      "flos": 24095772864000.0,
      "grad_norm": 1.6809529045282812,
      "language_loss": 0.77258694,
      "learning_rate": 3.348973500311086e-06,
      "loss": 0.79398209,
      "num_input_tokens_seen": 102801000,
      "step": 4764,
      "time_per_iteration": 2.59322190284729
    },
    {
      "auxiliary_loss_clip": 0.0111101,
      "auxiliary_loss_mlp": 0.0104964,
      "balance_loss_clip": 1.04797757,
      "balance_loss_mlp": 1.03056645,
      "epoch": 0.2864872989628739,
      "flos": 22601386408320.0,
      "grad_norm": 2.145650389309795,
      "language_loss": 0.71602571,
      "learning_rate": 3.348685940258466e-06,
      "loss": 0.73763216,
      "num_input_tokens_seen": 102820230,
      "step": 4765,
      "time_per_iteration": 2.5628535747528076
    },
    {
      "auxiliary_loss_clip": 0.0112123,
      "auxiliary_loss_mlp": 0.01037512,
      "balance_loss_clip": 1.04530001,
      "balance_loss_mlp": 1.02103722,
      "epoch": 0.28654742221554186,
      "flos": 32745073138560.0,
      "grad_norm": 1.4741809827318884,
      "language_loss": 0.76299542,
      "learning_rate": 3.3483983290635395e-06,
      "loss": 0.78458285,
      "num_input_tokens_seen": 102842670,
      "step": 4766,
      "time_per_iteration": 2.587432384490967
    },
    {
      "auxiliary_loss_clip": 0.01126043,
      "auxiliary_loss_mlp": 0.01037963,
      "balance_loss_clip": 1.04860449,
      "balance_loss_mlp": 1.02090383,
      "epoch": 0.2866075454682098,
      "flos": 26981626331520.0,
      "grad_norm": 1.6391728387880309,
      "language_loss": 0.77360928,
      "learning_rate": 3.348110666737214e-06,
      "loss": 0.79524934,
      "num_input_tokens_seen": 102864480,
      "step": 4767,
      "time_per_iteration": 2.546358346939087
    },
    {
      "auxiliary_loss_clip": 0.01138345,
      "auxiliary_loss_mlp": 0.0104506,
      "balance_loss_clip": 1.04898131,
      "balance_loss_mlp": 1.02763152,
      "epoch": 0.2866676687208778,
      "flos": 23253847004160.0,
      "grad_norm": 7.046884905568304,
      "language_loss": 0.65011466,
      "learning_rate": 3.3478229532903956e-06,
      "loss": 0.67194867,
      "num_input_tokens_seen": 102883740,
      "step": 4768,
      "time_per_iteration": 2.4628591537475586
    },
    {
      "auxiliary_loss_clip": 0.01122376,
      "auxiliary_loss_mlp": 0.01043338,
      "balance_loss_clip": 1.04617262,
      "balance_loss_mlp": 1.0260886,
      "epoch": 0.28672779197354575,
      "flos": 21579727870080.0,
      "grad_norm": 1.4817828808367253,
      "language_loss": 0.70423031,
      "learning_rate": 3.3475351887339967e-06,
      "loss": 0.72588742,
      "num_input_tokens_seen": 102902945,
      "step": 4769,
      "time_per_iteration": 2.5374674797058105
    },
    {
      "auxiliary_loss_clip": 0.01079071,
      "auxiliary_loss_mlp": 0.01036974,
      "balance_loss_clip": 1.04922342,
      "balance_loss_mlp": 1.02066636,
      "epoch": 0.2867879152262137,
      "flos": 19865568049920.0,
      "grad_norm": 1.8495768809499047,
      "language_loss": 0.74623311,
      "learning_rate": 3.3472473730789288e-06,
      "loss": 0.76739353,
      "num_input_tokens_seen": 102922405,
      "step": 4770,
      "time_per_iteration": 2.572683334350586
    },
    {
      "auxiliary_loss_clip": 0.01095659,
      "auxiliary_loss_mlp": 0.01039514,
      "balance_loss_clip": 1.04554141,
      "balance_loss_mlp": 1.02244353,
      "epoch": 0.2868480384788817,
      "flos": 28213325648640.0,
      "grad_norm": 2.1929149219131046,
      "language_loss": 0.67686105,
      "learning_rate": 3.3469595063361045e-06,
      "loss": 0.69821274,
      "num_input_tokens_seen": 102938980,
      "step": 4771,
      "time_per_iteration": 2.606570243835449
    },
    {
      "auxiliary_loss_clip": 0.0104232,
      "auxiliary_loss_mlp": 0.01013457,
      "balance_loss_clip": 1.02050531,
      "balance_loss_mlp": 1.01083457,
      "epoch": 0.2869081617315497,
      "flos": 65424286690560.0,
      "grad_norm": 0.7771358482650962,
      "language_loss": 0.56848186,
      "learning_rate": 3.3466715885164414e-06,
      "loss": 0.58903962,
      "num_input_tokens_seen": 103000405,
      "step": 4772,
      "time_per_iteration": 3.0561890602111816
    },
    {
      "auxiliary_loss_clip": 0.01068197,
      "auxiliary_loss_mlp": 0.00802564,
      "balance_loss_clip": 1.04791284,
      "balance_loss_mlp": 1.02087307,
      "epoch": 0.28696828498421767,
      "flos": 18660729127680.0,
      "grad_norm": 3.514771486332438,
      "language_loss": 0.83102655,
      "learning_rate": 3.346383619630856e-06,
      "loss": 0.84973413,
      "num_input_tokens_seen": 103017970,
      "step": 4773,
      "time_per_iteration": 2.589796543121338
    },
    {
      "auxiliary_loss_clip": 0.01135772,
      "auxiliary_loss_mlp": 0.01044142,
      "balance_loss_clip": 1.04373336,
      "balance_loss_mlp": 1.02660584,
      "epoch": 0.28702840823688563,
      "flos": 23659745667840.0,
      "grad_norm": 2.236332573757313,
      "language_loss": 0.78125864,
      "learning_rate": 3.34609559969027e-06,
      "loss": 0.80305773,
      "num_input_tokens_seen": 103036385,
      "step": 4774,
      "time_per_iteration": 2.4978747367858887
    },
    {
      "auxiliary_loss_clip": 0.01120781,
      "auxiliary_loss_mlp": 0.01040058,
      "balance_loss_clip": 1.04889321,
      "balance_loss_mlp": 1.0228802,
      "epoch": 0.2870885314895536,
      "flos": 13804744544640.0,
      "grad_norm": 3.2303872857382094,
      "language_loss": 0.73551333,
      "learning_rate": 3.3458075287056034e-06,
      "loss": 0.75712168,
      "num_input_tokens_seen": 103052170,
      "step": 4775,
      "time_per_iteration": 2.487281322479248
    },
    {
      "auxiliary_loss_clip": 0.01128898,
      "auxiliary_loss_mlp": 0.01043169,
      "balance_loss_clip": 1.04778576,
      "balance_loss_mlp": 1.02615762,
      "epoch": 0.28714865474222157,
      "flos": 17786771314560.0,
      "grad_norm": 1.6660118843038847,
      "language_loss": 0.88346696,
      "learning_rate": 3.34551940668778e-06,
      "loss": 0.90518761,
      "num_input_tokens_seen": 103070510,
      "step": 4776,
      "time_per_iteration": 2.49529767036438
    },
    {
      "auxiliary_loss_clip": 0.01130966,
      "auxiliary_loss_mlp": 0.01042449,
      "balance_loss_clip": 1.04798794,
      "balance_loss_mlp": 1.02599788,
      "epoch": 0.28720877799488953,
      "flos": 15997486199040.0,
      "grad_norm": 1.7433575012401514,
      "language_loss": 0.74455446,
      "learning_rate": 3.345231233647726e-06,
      "loss": 0.76628864,
      "num_input_tokens_seen": 103089590,
      "step": 4777,
      "time_per_iteration": 2.4783732891082764
    },
    {
      "auxiliary_loss_clip": 0.01121369,
      "auxiliary_loss_mlp": 0.01045536,
      "balance_loss_clip": 1.04871154,
      "balance_loss_mlp": 1.02751124,
      "epoch": 0.2872689012475575,
      "flos": 20923137210240.0,
      "grad_norm": 1.9333975098312768,
      "language_loss": 0.80073142,
      "learning_rate": 3.3449430095963696e-06,
      "loss": 0.82240045,
      "num_input_tokens_seen": 103109080,
      "step": 4778,
      "time_per_iteration": 2.5239548683166504
    },
    {
      "auxiliary_loss_clip": 0.01115648,
      "auxiliary_loss_mlp": 0.01045992,
      "balance_loss_clip": 1.05035591,
      "balance_loss_mlp": 1.02875447,
      "epoch": 0.28732902450022546,
      "flos": 21325121291520.0,
      "grad_norm": 2.3268660464597133,
      "language_loss": 0.73681831,
      "learning_rate": 3.3446547345446386e-06,
      "loss": 0.75843465,
      "num_input_tokens_seen": 103127755,
      "step": 4779,
      "time_per_iteration": 2.5321717262268066
    },
    {
      "auxiliary_loss_clip": 0.01122971,
      "auxiliary_loss_mlp": 0.01047601,
      "balance_loss_clip": 1.04734063,
      "balance_loss_mlp": 1.02939737,
      "epoch": 0.2873891477528934,
      "flos": 20850382212480.0,
      "grad_norm": 1.519506602262648,
      "language_loss": 0.7599743,
      "learning_rate": 3.3443664085034656e-06,
      "loss": 0.78167999,
      "num_input_tokens_seen": 103147035,
      "step": 4780,
      "time_per_iteration": 2.519141912460327
    },
    {
      "auxiliary_loss_clip": 0.01098947,
      "auxiliary_loss_mlp": 0.01044288,
      "balance_loss_clip": 1.04321027,
      "balance_loss_mlp": 1.02838564,
      "epoch": 0.2874492710055614,
      "flos": 17420051410560.0,
      "grad_norm": 1.788816116988456,
      "language_loss": 0.81210744,
      "learning_rate": 3.344078031483784e-06,
      "loss": 0.83353978,
      "num_input_tokens_seen": 103165410,
      "step": 4781,
      "time_per_iteration": 2.5388567447662354
    },
    {
      "auxiliary_loss_clip": 0.0110773,
      "auxiliary_loss_mlp": 0.01045687,
      "balance_loss_clip": 1.04744744,
      "balance_loss_mlp": 1.0269351,
      "epoch": 0.28750939425822936,
      "flos": 13406818700160.0,
      "grad_norm": 2.0634908323296646,
      "language_loss": 0.86759174,
      "learning_rate": 3.3437896034965283e-06,
      "loss": 0.88912594,
      "num_input_tokens_seen": 103183710,
      "step": 4782,
      "time_per_iteration": 3.931670665740967
    },
    {
      "auxiliary_loss_clip": 0.01105331,
      "auxiliary_loss_mlp": 0.01042782,
      "balance_loss_clip": 1.04790056,
      "balance_loss_mlp": 1.02553844,
      "epoch": 0.2875695175108973,
      "flos": 21870029589120.0,
      "grad_norm": 1.431989918082156,
      "language_loss": 0.71197486,
      "learning_rate": 3.3435011245526357e-06,
      "loss": 0.73345602,
      "num_input_tokens_seen": 103203790,
      "step": 4783,
      "time_per_iteration": 2.562919855117798
    },
    {
      "auxiliary_loss_clip": 0.01116754,
      "auxiliary_loss_mlp": 0.01043158,
      "balance_loss_clip": 1.0515151,
      "balance_loss_mlp": 1.02634978,
      "epoch": 0.2876296407635653,
      "flos": 26245457089920.0,
      "grad_norm": 1.6348801775613293,
      "language_loss": 0.76708436,
      "learning_rate": 3.343212594663047e-06,
      "loss": 0.78868347,
      "num_input_tokens_seen": 103223925,
      "step": 4784,
      "time_per_iteration": 2.553053140640259
    },
    {
      "auxiliary_loss_clip": 0.01093478,
      "auxiliary_loss_mlp": 0.01050314,
      "balance_loss_clip": 1.04737186,
      "balance_loss_mlp": 1.03230166,
      "epoch": 0.28768976401623325,
      "flos": 25373654092800.0,
      "grad_norm": 1.4839660650445712,
      "language_loss": 0.76311821,
      "learning_rate": 3.3429240138387015e-06,
      "loss": 0.78455609,
      "num_input_tokens_seen": 103244760,
      "step": 4785,
      "time_per_iteration": 2.5848641395568848
    },
    {
      "auxiliary_loss_clip": 0.01139317,
      "auxiliary_loss_mlp": 0.01048424,
      "balance_loss_clip": 1.04904962,
      "balance_loss_mlp": 1.0314486,
      "epoch": 0.28774988726890127,
      "flos": 30664372982400.0,
      "grad_norm": 2.347020711878431,
      "language_loss": 0.83030087,
      "learning_rate": 3.3426353820905425e-06,
      "loss": 0.85217828,
      "num_input_tokens_seen": 103261995,
      "step": 4786,
      "time_per_iteration": 3.9360263347625732
    },
    {
      "auxiliary_loss_clip": 0.01104715,
      "auxiliary_loss_mlp": 0.00800125,
      "balance_loss_clip": 1.04583335,
      "balance_loss_mlp": 1.02445149,
      "epoch": 0.28781001052156924,
      "flos": 20595452411520.0,
      "grad_norm": 1.8640251828811247,
      "language_loss": 0.79684925,
      "learning_rate": 3.342346699429516e-06,
      "loss": 0.8158977,
      "num_input_tokens_seen": 103279780,
      "step": 4787,
      "time_per_iteration": 2.5587966442108154
    },
    {
      "auxiliary_loss_clip": 0.01116317,
      "auxiliary_loss_mlp": 0.01042927,
      "balance_loss_clip": 1.04686737,
      "balance_loss_mlp": 1.02608252,
      "epoch": 0.2878701337742372,
      "flos": 26542330997760.0,
      "grad_norm": 5.033182495816941,
      "language_loss": 0.83516729,
      "learning_rate": 3.3420579658665677e-06,
      "loss": 0.85675967,
      "num_input_tokens_seen": 103300580,
      "step": 4788,
      "time_per_iteration": 2.561777114868164
    },
    {
      "auxiliary_loss_clip": 0.01102843,
      "auxiliary_loss_mlp": 0.01051308,
      "balance_loss_clip": 1.04956782,
      "balance_loss_mlp": 1.03402257,
      "epoch": 0.28793025702690517,
      "flos": 28146855530880.0,
      "grad_norm": 1.6856530365115436,
      "language_loss": 0.73589593,
      "learning_rate": 3.3417691814126468e-06,
      "loss": 0.75743747,
      "num_input_tokens_seen": 103320430,
      "step": 4789,
      "time_per_iteration": 4.003628730773926
    },
    {
      "auxiliary_loss_clip": 0.01119623,
      "auxiliary_loss_mlp": 0.01048891,
      "balance_loss_clip": 1.04887676,
      "balance_loss_mlp": 1.03243947,
      "epoch": 0.28799038027957313,
      "flos": 23805471144960.0,
      "grad_norm": 1.5889162720658165,
      "language_loss": 0.83605951,
      "learning_rate": 3.341480346078704e-06,
      "loss": 0.85774469,
      "num_input_tokens_seen": 103337695,
      "step": 4790,
      "time_per_iteration": 2.5266106128692627
    },
    {
      "auxiliary_loss_clip": 0.01130859,
      "auxiliary_loss_mlp": 0.01042825,
      "balance_loss_clip": 1.04926395,
      "balance_loss_mlp": 1.02594507,
      "epoch": 0.2880505035322411,
      "flos": 22344122223360.0,
      "grad_norm": 2.1992852955532403,
      "language_loss": 0.7786116,
      "learning_rate": 3.3411914598756922e-06,
      "loss": 0.8003484,
      "num_input_tokens_seen": 103357010,
      "step": 4791,
      "time_per_iteration": 3.864971876144409
    },
    {
      "auxiliary_loss_clip": 0.01118893,
      "auxiliary_loss_mlp": 0.01036569,
      "balance_loss_clip": 1.05363584,
      "balance_loss_mlp": 1.01983249,
      "epoch": 0.28811062678490906,
      "flos": 18004246208640.0,
      "grad_norm": 1.69083438806317,
      "language_loss": 0.70408237,
      "learning_rate": 3.3409025228145654e-06,
      "loss": 0.72563696,
      "num_input_tokens_seen": 103375600,
      "step": 4792,
      "time_per_iteration": 2.5106186866760254
    },
    {
      "auxiliary_loss_clip": 0.01096941,
      "auxiliary_loss_mlp": 0.01036645,
      "balance_loss_clip": 1.04831147,
      "balance_loss_mlp": 1.02021778,
      "epoch": 0.28817075003757703,
      "flos": 22090880361600.0,
      "grad_norm": 1.852421046620153,
      "language_loss": 0.79528248,
      "learning_rate": 3.3406135349062812e-06,
      "loss": 0.81661832,
      "num_input_tokens_seen": 103395225,
      "step": 4793,
      "time_per_iteration": 2.597184658050537
    },
    {
      "auxiliary_loss_clip": 0.01114566,
      "auxiliary_loss_mlp": 0.01042947,
      "balance_loss_clip": 1.04845262,
      "balance_loss_mlp": 1.02672255,
      "epoch": 0.288230873290245,
      "flos": 41683130847360.0,
      "grad_norm": 1.5850170885615402,
      "language_loss": 0.77554369,
      "learning_rate": 3.340324496161797e-06,
      "loss": 0.79711878,
      "num_input_tokens_seen": 103417245,
      "step": 4794,
      "time_per_iteration": 2.7076923847198486
    },
    {
      "auxiliary_loss_clip": 0.01130233,
      "auxiliary_loss_mlp": 0.0104739,
      "balance_loss_clip": 1.05089021,
      "balance_loss_mlp": 1.03017652,
      "epoch": 0.28829099654291296,
      "flos": 18624423456000.0,
      "grad_norm": 2.2091357316789466,
      "language_loss": 0.8344444,
      "learning_rate": 3.340035406592074e-06,
      "loss": 0.8562206,
      "num_input_tokens_seen": 103435500,
      "step": 4795,
      "time_per_iteration": 2.4757022857666016
    },
    {
      "auxiliary_loss_clip": 0.0112449,
      "auxiliary_loss_mlp": 0.01041411,
      "balance_loss_clip": 1.04961681,
      "balance_loss_mlp": 1.02561593,
      "epoch": 0.2883511197955809,
      "flos": 24674832017280.0,
      "grad_norm": 2.7237936312301705,
      "language_loss": 0.74602813,
      "learning_rate": 3.339746266208074e-06,
      "loss": 0.76768708,
      "num_input_tokens_seen": 103451040,
      "step": 4796,
      "time_per_iteration": 2.5243303775787354
    },
    {
      "auxiliary_loss_clip": 0.01134043,
      "auxiliary_loss_mlp": 0.01042604,
      "balance_loss_clip": 1.0512476,
      "balance_loss_mlp": 1.02428222,
      "epoch": 0.2884112430482489,
      "flos": 23112143850240.0,
      "grad_norm": 2.1808058552661307,
      "language_loss": 0.72897542,
      "learning_rate": 3.3394570750207614e-06,
      "loss": 0.7507419,
      "num_input_tokens_seen": 103471330,
      "step": 4797,
      "time_per_iteration": 2.5189762115478516
    },
    {
      "auxiliary_loss_clip": 0.01102617,
      "auxiliary_loss_mlp": 0.0079913,
      "balance_loss_clip": 1.04585314,
      "balance_loss_mlp": 1.01926136,
      "epoch": 0.28847136630091685,
      "flos": 16873347432960.0,
      "grad_norm": 1.888846599443049,
      "language_loss": 0.74155521,
      "learning_rate": 3.3391678330411017e-06,
      "loss": 0.76057273,
      "num_input_tokens_seen": 103488060,
      "step": 4798,
      "time_per_iteration": 2.5525143146514893
    },
    {
      "auxiliary_loss_clip": 0.01129831,
      "auxiliary_loss_mlp": 0.01041643,
      "balance_loss_clip": 1.0475533,
      "balance_loss_mlp": 1.0235467,
      "epoch": 0.2885314895535849,
      "flos": 25657527277440.0,
      "grad_norm": 2.7295345992437485,
      "language_loss": 0.64997059,
      "learning_rate": 3.3388785402800642e-06,
      "loss": 0.67168534,
      "num_input_tokens_seen": 103503600,
      "step": 4799,
      "time_per_iteration": 2.5174527168273926
    },
    {
      "auxiliary_loss_clip": 0.01142333,
      "auxiliary_loss_mlp": 0.01045959,
      "balance_loss_clip": 1.05084682,
      "balance_loss_mlp": 1.02903104,
      "epoch": 0.28859161280625284,
      "flos": 21107251347840.0,
      "grad_norm": 1.7006293787646176,
      "language_loss": 0.82455748,
      "learning_rate": 3.3385891967486178e-06,
      "loss": 0.84644043,
      "num_input_tokens_seen": 103524195,
      "step": 4800,
      "time_per_iteration": 2.545185089111328
    },
    {
      "auxiliary_loss_clip": 0.01100575,
      "auxiliary_loss_mlp": 0.01045044,
      "balance_loss_clip": 1.04718399,
      "balance_loss_mlp": 1.02800894,
      "epoch": 0.2886517360589208,
      "flos": 26469540086400.0,
      "grad_norm": 2.5224837067113692,
      "language_loss": 0.91113305,
      "learning_rate": 3.3382998024577347e-06,
      "loss": 0.93258929,
      "num_input_tokens_seen": 103545235,
      "step": 4801,
      "time_per_iteration": 2.606602907180786
    },
    {
      "auxiliary_loss_clip": 0.01121485,
      "auxiliary_loss_mlp": 0.0079829,
      "balance_loss_clip": 1.052791,
      "balance_loss_mlp": 1.01896858,
      "epoch": 0.28871185931158877,
      "flos": 25265275781760.0,
      "grad_norm": 1.948281667334705,
      "language_loss": 0.73326087,
      "learning_rate": 3.33801035741839e-06,
      "loss": 0.75245863,
      "num_input_tokens_seen": 103563305,
      "step": 4802,
      "time_per_iteration": 2.572216749191284
    },
    {
      "auxiliary_loss_clip": 0.01023632,
      "auxiliary_loss_mlp": 0.01007671,
      "balance_loss_clip": 1.02162278,
      "balance_loss_mlp": 1.00567997,
      "epoch": 0.28877198256425674,
      "flos": 66665431284480.0,
      "grad_norm": 0.7828172742007847,
      "language_loss": 0.62973672,
      "learning_rate": 3.337720861641558e-06,
      "loss": 0.65004975,
      "num_input_tokens_seen": 103625025,
      "step": 4803,
      "time_per_iteration": 3.1200687885284424
    },
    {
      "auxiliary_loss_clip": 0.01083662,
      "auxiliary_loss_mlp": 0.01045715,
      "balance_loss_clip": 1.04621577,
      "balance_loss_mlp": 1.02947867,
      "epoch": 0.2888321058169247,
      "flos": 20303031790080.0,
      "grad_norm": 1.7674035801860448,
      "language_loss": 0.70588827,
      "learning_rate": 3.3374313151382165e-06,
      "loss": 0.72718203,
      "num_input_tokens_seen": 103644235,
      "step": 4804,
      "time_per_iteration": 2.5733144283294678
    },
    {
      "auxiliary_loss_clip": 0.01130126,
      "auxiliary_loss_mlp": 0.0104538,
      "balance_loss_clip": 1.04806495,
      "balance_loss_mlp": 1.02731931,
      "epoch": 0.28889222906959267,
      "flos": 25516721963520.0,
      "grad_norm": 2.0708928721694986,
      "language_loss": 0.68365002,
      "learning_rate": 3.337141717919346e-06,
      "loss": 0.70540506,
      "num_input_tokens_seen": 103664700,
      "step": 4805,
      "time_per_iteration": 2.5465078353881836
    },
    {
      "auxiliary_loss_clip": 0.01129149,
      "auxiliary_loss_mlp": 0.01042849,
      "balance_loss_clip": 1.04949367,
      "balance_loss_mlp": 1.026196,
      "epoch": 0.28895235232226063,
      "flos": 32671312560000.0,
      "grad_norm": 1.5028380745726795,
      "language_loss": 0.69150245,
      "learning_rate": 3.3368520699959272e-06,
      "loss": 0.71322244,
      "num_input_tokens_seen": 103686595,
      "step": 4806,
      "time_per_iteration": 2.585876226425171
    },
    {
      "auxiliary_loss_clip": 0.01116833,
      "auxiliary_loss_mlp": 0.01044133,
      "balance_loss_clip": 1.04845011,
      "balance_loss_mlp": 1.02753949,
      "epoch": 0.2890124755749286,
      "flos": 29714679342720.0,
      "grad_norm": 1.4605569665929177,
      "language_loss": 0.71404994,
      "learning_rate": 3.3365623713789443e-06,
      "loss": 0.7356596,
      "num_input_tokens_seen": 103707525,
      "step": 4807,
      "time_per_iteration": 2.6029093265533447
    },
    {
      "auxiliary_loss_clip": 0.01103077,
      "auxiliary_loss_mlp": 0.01042189,
      "balance_loss_clip": 1.05245352,
      "balance_loss_mlp": 1.02510667,
      "epoch": 0.28907259882759656,
      "flos": 22674464628480.0,
      "grad_norm": 1.7863235905482282,
      "language_loss": 0.81668806,
      "learning_rate": 3.336272622079382e-06,
      "loss": 0.83814073,
      "num_input_tokens_seen": 103727905,
      "step": 4808,
      "time_per_iteration": 2.592531204223633
    },
    {
      "auxiliary_loss_clip": 0.01096128,
      "auxiliary_loss_mlp": 0.01046816,
      "balance_loss_clip": 1.05126381,
      "balance_loss_mlp": 1.02999592,
      "epoch": 0.2891327220802645,
      "flos": 22566050403840.0,
      "grad_norm": 1.6796839893116917,
      "language_loss": 0.78443199,
      "learning_rate": 3.3359828221082276e-06,
      "loss": 0.80586141,
      "num_input_tokens_seen": 103748335,
      "step": 4809,
      "time_per_iteration": 2.589555501937866
    },
    {
      "auxiliary_loss_clip": 0.01086526,
      "auxiliary_loss_mlp": 0.01047285,
      "balance_loss_clip": 1.04476142,
      "balance_loss_mlp": 1.02911699,
      "epoch": 0.2891928453329325,
      "flos": 21652806090240.0,
      "grad_norm": 1.7242362731346326,
      "language_loss": 0.787103,
      "learning_rate": 3.3356929714764714e-06,
      "loss": 0.80844104,
      "num_input_tokens_seen": 103767020,
      "step": 4810,
      "time_per_iteration": 2.592916250228882
    },
    {
      "auxiliary_loss_clip": 0.01087993,
      "auxiliary_loss_mlp": 0.01038983,
      "balance_loss_clip": 1.04615712,
      "balance_loss_mlp": 1.02251434,
      "epoch": 0.28925296858560046,
      "flos": 23222102359680.0,
      "grad_norm": 1.7067608554091012,
      "language_loss": 0.7702812,
      "learning_rate": 3.3354030701951032e-06,
      "loss": 0.79155087,
      "num_input_tokens_seen": 103786355,
      "step": 4811,
      "time_per_iteration": 2.6138651371002197
    },
    {
      "auxiliary_loss_clip": 0.01123123,
      "auxiliary_loss_mlp": 0.01040523,
      "balance_loss_clip": 1.04570723,
      "balance_loss_mlp": 1.02288032,
      "epoch": 0.2893130918382685,
      "flos": 28621666437120.0,
      "grad_norm": 1.5918208177061774,
      "language_loss": 0.77299571,
      "learning_rate": 3.335113118275117e-06,
      "loss": 0.7946322,
      "num_input_tokens_seen": 103809345,
      "step": 4812,
      "time_per_iteration": 2.5525853633880615
    },
    {
      "auxiliary_loss_clip": 0.0103715,
      "auxiliary_loss_mlp": 0.01005657,
      "balance_loss_clip": 1.04362643,
      "balance_loss_mlp": 1.00334454,
      "epoch": 0.28937321509093644,
      "flos": 72301288982400.0,
      "grad_norm": 0.8203812397574088,
      "language_loss": 0.60211718,
      "learning_rate": 3.3348231157275085e-06,
      "loss": 0.62254524,
      "num_input_tokens_seen": 103871180,
      "step": 4813,
      "time_per_iteration": 3.350177049636841
    },
    {
      "auxiliary_loss_clip": 0.01093125,
      "auxiliary_loss_mlp": 0.01042086,
      "balance_loss_clip": 1.04585767,
      "balance_loss_mlp": 1.02412081,
      "epoch": 0.2894333383436044,
      "flos": 16216397637120.0,
      "grad_norm": 3.590983818089054,
      "language_loss": 0.82785547,
      "learning_rate": 3.3345330625632725e-06,
      "loss": 0.84920758,
      "num_input_tokens_seen": 103889040,
      "step": 4814,
      "time_per_iteration": 2.5532994270324707
    },
    {
      "auxiliary_loss_clip": 0.01093454,
      "auxiliary_loss_mlp": 0.01043973,
      "balance_loss_clip": 1.04635572,
      "balance_loss_mlp": 1.02631855,
      "epoch": 0.2894934615962724,
      "flos": 24828278918400.0,
      "grad_norm": 1.5385135063197175,
      "language_loss": 0.72702378,
      "learning_rate": 3.3342429587934094e-06,
      "loss": 0.74839807,
      "num_input_tokens_seen": 103910380,
      "step": 4815,
      "time_per_iteration": 2.646898031234741
    },
    {
      "auxiliary_loss_clip": 0.01124896,
      "auxiliary_loss_mlp": 0.01049842,
      "balance_loss_clip": 1.04897428,
      "balance_loss_mlp": 1.03420162,
      "epoch": 0.28955358484894034,
      "flos": 20449978329600.0,
      "grad_norm": 1.633085913960417,
      "language_loss": 0.71300352,
      "learning_rate": 3.3339528044289198e-06,
      "loss": 0.73475087,
      "num_input_tokens_seen": 103929955,
      "step": 4816,
      "time_per_iteration": 2.5361380577087402
    },
    {
      "auxiliary_loss_clip": 0.01113518,
      "auxiliary_loss_mlp": 0.01048274,
      "balance_loss_clip": 1.04718804,
      "balance_loss_mlp": 1.03040504,
      "epoch": 0.2896137081016083,
      "flos": 22565188477440.0,
      "grad_norm": 2.0464038254033277,
      "language_loss": 0.74567235,
      "learning_rate": 3.3336625994808055e-06,
      "loss": 0.76729023,
      "num_input_tokens_seen": 103948020,
      "step": 4817,
      "time_per_iteration": 2.54116153717041
    },
    {
      "auxiliary_loss_clip": 0.01105473,
      "auxiliary_loss_mlp": 0.01050379,
      "balance_loss_clip": 1.04829526,
      "balance_loss_mlp": 1.03217554,
      "epoch": 0.28967383135427627,
      "flos": 26687948734080.0,
      "grad_norm": 1.7654509799217897,
      "language_loss": 0.76455295,
      "learning_rate": 3.3333723439600723e-06,
      "loss": 0.78611147,
      "num_input_tokens_seen": 103968740,
      "step": 4818,
      "time_per_iteration": 2.641082763671875
    },
    {
      "auxiliary_loss_clip": 0.01078386,
      "auxiliary_loss_mlp": 0.01041792,
      "balance_loss_clip": 1.05289197,
      "balance_loss_mlp": 1.02443552,
      "epoch": 0.28973395460694423,
      "flos": 15558262692480.0,
      "grad_norm": 1.8055708611924384,
      "language_loss": 0.79841411,
      "learning_rate": 3.3330820378777263e-06,
      "loss": 0.81961584,
      "num_input_tokens_seen": 103986005,
      "step": 4819,
      "time_per_iteration": 2.600611925125122
    },
    {
      "auxiliary_loss_clip": 0.01107773,
      "auxiliary_loss_mlp": 0.01045481,
      "balance_loss_clip": 1.05002475,
      "balance_loss_mlp": 1.0266335,
      "epoch": 0.2897940778596122,
      "flos": 18697465762560.0,
      "grad_norm": 2.19313585076955,
      "language_loss": 0.78800124,
      "learning_rate": 3.332791681244776e-06,
      "loss": 0.80953377,
      "num_input_tokens_seen": 104005070,
      "step": 4820,
      "time_per_iteration": 3.9668190479278564
    },
    {
      "auxiliary_loss_clip": 0.01098833,
      "auxiliary_loss_mlp": 0.01039183,
      "balance_loss_clip": 1.04956484,
      "balance_loss_mlp": 1.0221957,
      "epoch": 0.28985420111228016,
      "flos": 18770292587520.0,
      "grad_norm": 2.0852321246101893,
      "language_loss": 0.72399557,
      "learning_rate": 3.332501274072231e-06,
      "loss": 0.74537575,
      "num_input_tokens_seen": 104022945,
      "step": 4821,
      "time_per_iteration": 2.6255605220794678
    },
    {
      "auxiliary_loss_clip": 0.01123572,
      "auxiliary_loss_mlp": 0.01043386,
      "balance_loss_clip": 1.04462767,
      "balance_loss_mlp": 1.02573073,
      "epoch": 0.28991432436494813,
      "flos": 23069840607360.0,
      "grad_norm": 1.6345772508501442,
      "language_loss": 0.72151625,
      "learning_rate": 3.332210816371104e-06,
      "loss": 0.74318576,
      "num_input_tokens_seen": 104042080,
      "step": 4822,
      "time_per_iteration": 2.518812417984009
    },
    {
      "auxiliary_loss_clip": 0.01127346,
      "auxiliary_loss_mlp": 0.01052542,
      "balance_loss_clip": 1.05389559,
      "balance_loss_mlp": 1.03572178,
      "epoch": 0.2899744476176161,
      "flos": 17603195880960.0,
      "grad_norm": 1.9413932778065264,
      "language_loss": 0.66026402,
      "learning_rate": 3.3319203081524102e-06,
      "loss": 0.68206298,
      "num_input_tokens_seen": 104060975,
      "step": 4823,
      "time_per_iteration": 2.480194330215454
    },
    {
      "auxiliary_loss_clip": 0.01104906,
      "auxiliary_loss_mlp": 0.01050483,
      "balance_loss_clip": 1.04491401,
      "balance_loss_mlp": 1.03332925,
      "epoch": 0.29003457087028406,
      "flos": 22309360836480.0,
      "grad_norm": 1.9682731472528185,
      "language_loss": 0.80678505,
      "learning_rate": 3.331629749427164e-06,
      "loss": 0.82833898,
      "num_input_tokens_seen": 104081395,
      "step": 4824,
      "time_per_iteration": 2.5336921215057373
    },
    {
      "auxiliary_loss_clip": 0.01136256,
      "auxiliary_loss_mlp": 0.01045007,
      "balance_loss_clip": 1.04539847,
      "balance_loss_mlp": 1.02679217,
      "epoch": 0.2900946941229521,
      "flos": 21944975316480.0,
      "grad_norm": 2.432648501077168,
      "language_loss": 0.72968924,
      "learning_rate": 3.331339140206385e-06,
      "loss": 0.75150186,
      "num_input_tokens_seen": 104099995,
      "step": 4825,
      "time_per_iteration": 3.908147096633911
    },
    {
      "auxiliary_loss_clip": 0.01140351,
      "auxiliary_loss_mlp": 0.01042519,
      "balance_loss_clip": 1.04924107,
      "balance_loss_mlp": 1.02509022,
      "epoch": 0.29015481737562004,
      "flos": 17932173569280.0,
      "grad_norm": 2.0342936156330054,
      "language_loss": 0.73755825,
      "learning_rate": 3.331048480501092e-06,
      "loss": 0.7593869,
      "num_input_tokens_seen": 104118930,
      "step": 4826,
      "time_per_iteration": 2.454428195953369
    },
    {
      "auxiliary_loss_clip": 0.01122694,
      "auxiliary_loss_mlp": 0.01044601,
      "balance_loss_clip": 1.04481852,
      "balance_loss_mlp": 1.0286268,
      "epoch": 0.290214940628288,
      "flos": 22783525297920.0,
      "grad_norm": 1.9919390217428057,
      "language_loss": 0.68618894,
      "learning_rate": 3.3307577703223073e-06,
      "loss": 0.7078619,
      "num_input_tokens_seen": 104136940,
      "step": 4827,
      "time_per_iteration": 2.491593360900879
    },
    {
      "auxiliary_loss_clip": 0.01120912,
      "auxiliary_loss_mlp": 0.01047738,
      "balance_loss_clip": 1.04821396,
      "balance_loss_mlp": 1.02971363,
      "epoch": 0.290275063880956,
      "flos": 20006481104640.0,
      "grad_norm": 1.7051283393804744,
      "language_loss": 0.80255914,
      "learning_rate": 3.3304670096810545e-06,
      "loss": 0.82424569,
      "num_input_tokens_seen": 104154280,
      "step": 4828,
      "time_per_iteration": 3.8448123931884766
    },
    {
      "auxiliary_loss_clip": 0.01135672,
      "auxiliary_loss_mlp": 0.01052731,
      "balance_loss_clip": 1.04773808,
      "balance_loss_mlp": 1.03608894,
      "epoch": 0.29033518713362394,
      "flos": 22053605022720.0,
      "grad_norm": 2.398962496707361,
      "language_loss": 0.8048256,
      "learning_rate": 3.33017619858836e-06,
      "loss": 0.82670963,
      "num_input_tokens_seen": 104172605,
      "step": 4829,
      "time_per_iteration": 2.45902419090271
    },
    {
      "auxiliary_loss_clip": 0.01108453,
      "auxiliary_loss_mlp": 0.01038963,
      "balance_loss_clip": 1.04495692,
      "balance_loss_mlp": 1.02246451,
      "epoch": 0.2903953103862919,
      "flos": 25630056351360.0,
      "grad_norm": 1.5005968325389218,
      "language_loss": 0.82667208,
      "learning_rate": 3.329885337055249e-06,
      "loss": 0.84814626,
      "num_input_tokens_seen": 104194120,
      "step": 4830,
      "time_per_iteration": 3.9271233081817627
    },
    {
      "auxiliary_loss_clip": 0.01129852,
      "auxiliary_loss_mlp": 0.01050254,
      "balance_loss_clip": 1.04896581,
      "balance_loss_mlp": 1.03273034,
      "epoch": 0.29045543363895987,
      "flos": 16945851035520.0,
      "grad_norm": 2.238050671510261,
      "language_loss": 0.78937465,
      "learning_rate": 3.3295944250927546e-06,
      "loss": 0.8111757,
      "num_input_tokens_seen": 104210875,
      "step": 4831,
      "time_per_iteration": 2.4494216442108154
    },
    {
      "auxiliary_loss_clip": 0.01132521,
      "auxiliary_loss_mlp": 0.01044349,
      "balance_loss_clip": 1.04658771,
      "balance_loss_mlp": 1.02891135,
      "epoch": 0.29051555689162784,
      "flos": 26395492199040.0,
      "grad_norm": 1.785915045387044,
      "language_loss": 0.74273837,
      "learning_rate": 3.3293034627119055e-06,
      "loss": 0.76450706,
      "num_input_tokens_seen": 104229875,
      "step": 4832,
      "time_per_iteration": 2.505241870880127
    },
    {
      "auxiliary_loss_clip": 0.01109624,
      "auxiliary_loss_mlp": 0.01035142,
      "balance_loss_clip": 1.04497075,
      "balance_loss_mlp": 1.02061009,
      "epoch": 0.2905756801442958,
      "flos": 21103875469440.0,
      "grad_norm": 1.6257621918238705,
      "language_loss": 0.76184607,
      "learning_rate": 3.329012449923736e-06,
      "loss": 0.78329372,
      "num_input_tokens_seen": 104250405,
      "step": 4833,
      "time_per_iteration": 2.5236973762512207
    },
    {
      "auxiliary_loss_clip": 0.01102174,
      "auxiliary_loss_mlp": 0.01043579,
      "balance_loss_clip": 1.04461467,
      "balance_loss_mlp": 1.02711582,
      "epoch": 0.29063580339696377,
      "flos": 15706071158400.0,
      "grad_norm": 1.638458583624717,
      "language_loss": 0.64486355,
      "learning_rate": 3.3287213867392813e-06,
      "loss": 0.6663211,
      "num_input_tokens_seen": 104269185,
      "step": 4834,
      "time_per_iteration": 2.4966297149658203
    },
    {
      "auxiliary_loss_clip": 0.01114374,
      "auxiliary_loss_mlp": 0.01031664,
      "balance_loss_clip": 1.04618764,
      "balance_loss_mlp": 1.0166018,
      "epoch": 0.29069592664963173,
      "flos": 24644990793600.0,
      "grad_norm": 1.6645229963984465,
      "language_loss": 0.71389157,
      "learning_rate": 3.3284302731695783e-06,
      "loss": 0.73535192,
      "num_input_tokens_seen": 104289400,
      "step": 4835,
      "time_per_iteration": 2.536180257797241
    },
    {
      "auxiliary_loss_clip": 0.01108478,
      "auxiliary_loss_mlp": 0.01038209,
      "balance_loss_clip": 1.04437828,
      "balance_loss_mlp": 1.02318835,
      "epoch": 0.2907560499022997,
      "flos": 24973753000320.0,
      "grad_norm": 1.7098391207988692,
      "language_loss": 0.79219645,
      "learning_rate": 3.3281391092256668e-06,
      "loss": 0.81366336,
      "num_input_tokens_seen": 104310485,
      "step": 4836,
      "time_per_iteration": 2.5574657917022705
    },
    {
      "auxiliary_loss_clip": 0.01095918,
      "auxiliary_loss_mlp": 0.01048228,
      "balance_loss_clip": 1.04634571,
      "balance_loss_mlp": 1.03079975,
      "epoch": 0.29081617315496766,
      "flos": 18657496903680.0,
      "grad_norm": 1.7853821310553506,
      "language_loss": 0.8084169,
      "learning_rate": 3.3278478949185865e-06,
      "loss": 0.8298583,
      "num_input_tokens_seen": 104327330,
      "step": 4837,
      "time_per_iteration": 2.526125192642212
    },
    {
      "auxiliary_loss_clip": 0.01106894,
      "auxiliary_loss_mlp": 0.0103912,
      "balance_loss_clip": 1.04448175,
      "balance_loss_mlp": 1.02309239,
      "epoch": 0.2908762964076356,
      "flos": 35331035955840.0,
      "grad_norm": 1.8969058082127794,
      "language_loss": 0.67205465,
      "learning_rate": 3.327556630259381e-06,
      "loss": 0.69351482,
      "num_input_tokens_seen": 104350350,
      "step": 4838,
      "time_per_iteration": 2.6529359817504883
    },
    {
      "auxiliary_loss_clip": 0.01140035,
      "auxiliary_loss_mlp": 0.00796586,
      "balance_loss_clip": 1.04971874,
      "balance_loss_mlp": 1.01541865,
      "epoch": 0.29093641966030365,
      "flos": 23076305055360.0,
      "grad_norm": 1.6240014906662916,
      "language_loss": 0.71291816,
      "learning_rate": 3.327265315259095e-06,
      "loss": 0.73228437,
      "num_input_tokens_seen": 104369995,
      "step": 4839,
      "time_per_iteration": 2.475241184234619
    },
    {
      "auxiliary_loss_clip": 0.01135377,
      "auxiliary_loss_mlp": 0.01034313,
      "balance_loss_clip": 1.04671979,
      "balance_loss_mlp": 1.01887584,
      "epoch": 0.2909965429129716,
      "flos": 35955415094400.0,
      "grad_norm": 2.104541484353523,
      "language_loss": 0.76117986,
      "learning_rate": 3.326973949928776e-06,
      "loss": 0.78287673,
      "num_input_tokens_seen": 104392285,
      "step": 4840,
      "time_per_iteration": 2.6014418601989746
    },
    {
      "auxiliary_loss_clip": 0.01092254,
      "auxiliary_loss_mlp": 0.01044263,
      "balance_loss_clip": 1.04911077,
      "balance_loss_mlp": 1.02846766,
      "epoch": 0.2910566661656396,
      "flos": 30880231764480.0,
      "grad_norm": 1.891739371992281,
      "language_loss": 0.60350764,
      "learning_rate": 3.326682534279471e-06,
      "loss": 0.62487274,
      "num_input_tokens_seen": 104412640,
      "step": 4841,
      "time_per_iteration": 2.6110615730285645
    },
    {
      "auxiliary_loss_clip": 0.01113767,
      "auxiliary_loss_mlp": 0.01034628,
      "balance_loss_clip": 1.04790914,
      "balance_loss_mlp": 1.01787901,
      "epoch": 0.29111678941830754,
      "flos": 30010188533760.0,
      "grad_norm": 1.2835636525615306,
      "language_loss": 0.71131843,
      "learning_rate": 3.326391068322232e-06,
      "loss": 0.73280239,
      "num_input_tokens_seen": 104435245,
      "step": 4842,
      "time_per_iteration": 2.589752435684204
    },
    {
      "auxiliary_loss_clip": 0.01120649,
      "auxiliary_loss_mlp": 0.01035664,
      "balance_loss_clip": 1.04453254,
      "balance_loss_mlp": 1.02083421,
      "epoch": 0.2911769126709755,
      "flos": 22857393617280.0,
      "grad_norm": 1.6210651385791541,
      "language_loss": 0.73533225,
      "learning_rate": 3.3260995520681098e-06,
      "loss": 0.7568953,
      "num_input_tokens_seen": 104455395,
      "step": 4843,
      "time_per_iteration": 2.4801993370056152
    },
    {
      "auxiliary_loss_clip": 0.01089906,
      "auxiliary_loss_mlp": 0.01035123,
      "balance_loss_clip": 1.05082464,
      "balance_loss_mlp": 1.01997781,
      "epoch": 0.2912370359236435,
      "flos": 21650507619840.0,
      "grad_norm": 2.4135522728095133,
      "language_loss": 0.58341634,
      "learning_rate": 3.3258079855281602e-06,
      "loss": 0.60466671,
      "num_input_tokens_seen": 104473350,
      "step": 4844,
      "time_per_iteration": 2.5971016883850098
    },
    {
      "auxiliary_loss_clip": 0.01130059,
      "auxiliary_loss_mlp": 0.01037611,
      "balance_loss_clip": 1.05085087,
      "balance_loss_mlp": 1.02022982,
      "epoch": 0.29129715917631144,
      "flos": 22893340152960.0,
      "grad_norm": 2.2220433398646957,
      "language_loss": 0.86151212,
      "learning_rate": 3.3255163687134396e-06,
      "loss": 0.88318884,
      "num_input_tokens_seen": 104492265,
      "step": 4845,
      "time_per_iteration": 2.4858081340789795
    },
    {
      "auxiliary_loss_clip": 0.01113229,
      "auxiliary_loss_mlp": 0.01048283,
      "balance_loss_clip": 1.05085874,
      "balance_loss_mlp": 1.03112876,
      "epoch": 0.2913572824289794,
      "flos": 22674464628480.0,
      "grad_norm": 1.8785829465028014,
      "language_loss": 0.66771609,
      "learning_rate": 3.3252247016350046e-06,
      "loss": 0.68933117,
      "num_input_tokens_seen": 104510755,
      "step": 4846,
      "time_per_iteration": 2.5500428676605225
    },
    {
      "auxiliary_loss_clip": 0.01120072,
      "auxiliary_loss_mlp": 0.01037171,
      "balance_loss_clip": 1.04981589,
      "balance_loss_mlp": 1.0222466,
      "epoch": 0.29141740568164737,
      "flos": 23107403255040.0,
      "grad_norm": 1.935802944504872,
      "language_loss": 0.70473295,
      "learning_rate": 3.3249329843039166e-06,
      "loss": 0.72630537,
      "num_input_tokens_seen": 104530830,
      "step": 4847,
      "time_per_iteration": 2.521238327026367
    },
    {
      "auxiliary_loss_clip": 0.0112111,
      "auxiliary_loss_mlp": 0.01030208,
      "balance_loss_clip": 1.04434049,
      "balance_loss_mlp": 1.01468706,
      "epoch": 0.29147752893431533,
      "flos": 23587026583680.0,
      "grad_norm": 1.699794864283418,
      "language_loss": 0.7387647,
      "learning_rate": 3.324641216731237e-06,
      "loss": 0.76027787,
      "num_input_tokens_seen": 104550115,
      "step": 4848,
      "time_per_iteration": 2.535477876663208
    },
    {
      "auxiliary_loss_clip": 0.01117412,
      "auxiliary_loss_mlp": 0.01042526,
      "balance_loss_clip": 1.04371822,
      "balance_loss_mlp": 1.02561033,
      "epoch": 0.2915376521869833,
      "flos": 20591968792320.0,
      "grad_norm": 2.185202372781586,
      "language_loss": 0.76977265,
      "learning_rate": 3.3243493989280295e-06,
      "loss": 0.79137206,
      "num_input_tokens_seen": 104566255,
      "step": 4849,
      "time_per_iteration": 2.47027587890625
    },
    {
      "auxiliary_loss_clip": 0.0111774,
      "auxiliary_loss_mlp": 0.01039357,
      "balance_loss_clip": 1.04584193,
      "balance_loss_mlp": 1.02374077,
      "epoch": 0.29159777543965126,
      "flos": 20811490761600.0,
      "grad_norm": 1.618833453631688,
      "language_loss": 0.78362465,
      "learning_rate": 3.3240575309053596e-06,
      "loss": 0.80519569,
      "num_input_tokens_seen": 104585235,
      "step": 4850,
      "time_per_iteration": 2.505894184112549
    },
    {
      "auxiliary_loss_clip": 0.01107214,
      "auxiliary_loss_mlp": 0.01038631,
      "balance_loss_clip": 1.05103838,
      "balance_loss_mlp": 1.02253222,
      "epoch": 0.29165789869231923,
      "flos": 24244155947520.0,
      "grad_norm": 1.6513055039882796,
      "language_loss": 0.75652337,
      "learning_rate": 3.323765612674296e-06,
      "loss": 0.77798176,
      "num_input_tokens_seen": 104605315,
      "step": 4851,
      "time_per_iteration": 2.5365350246429443
    },
    {
      "auxiliary_loss_clip": 0.01120916,
      "auxiliary_loss_mlp": 0.0104321,
      "balance_loss_clip": 1.04728937,
      "balance_loss_mlp": 1.02900028,
      "epoch": 0.29171802194498725,
      "flos": 28949925853440.0,
      "grad_norm": 1.4122880165415193,
      "language_loss": 0.77309263,
      "learning_rate": 3.3234736442459078e-06,
      "loss": 0.79473388,
      "num_input_tokens_seen": 104626055,
      "step": 4852,
      "time_per_iteration": 2.5700159072875977
    },
    {
      "auxiliary_loss_clip": 0.01111467,
      "auxiliary_loss_mlp": 0.0104185,
      "balance_loss_clip": 1.04714632,
      "balance_loss_mlp": 1.02643657,
      "epoch": 0.2917781451976552,
      "flos": 22598226011520.0,
      "grad_norm": 1.576679825996922,
      "language_loss": 0.78142083,
      "learning_rate": 3.3231816256312665e-06,
      "loss": 0.80295396,
      "num_input_tokens_seen": 104646005,
      "step": 4853,
      "time_per_iteration": 2.558647871017456
    },
    {
      "auxiliary_loss_clip": 0.01100322,
      "auxiliary_loss_mlp": 0.01036277,
      "balance_loss_clip": 1.0474267,
      "balance_loss_mlp": 1.02079189,
      "epoch": 0.2918382684503232,
      "flos": 21574448570880.0,
      "grad_norm": 2.7192614644424142,
      "language_loss": 0.88506591,
      "learning_rate": 3.322889556841445e-06,
      "loss": 0.90643191,
      "num_input_tokens_seen": 104661620,
      "step": 4854,
      "time_per_iteration": 2.6070668697357178
    },
    {
      "auxiliary_loss_clip": 0.0111831,
      "auxiliary_loss_mlp": 0.01056007,
      "balance_loss_clip": 1.0480063,
      "balance_loss_mlp": 1.03791153,
      "epoch": 0.29189839170299114,
      "flos": 24353503925760.0,
      "grad_norm": 2.2972828860732806,
      "language_loss": 0.86293477,
      "learning_rate": 3.322597437887519e-06,
      "loss": 0.88467795,
      "num_input_tokens_seen": 104681445,
      "step": 4855,
      "time_per_iteration": 2.527416467666626
    },
    {
      "auxiliary_loss_clip": 0.01050791,
      "auxiliary_loss_mlp": 0.01041844,
      "balance_loss_clip": 1.02596068,
      "balance_loss_mlp": 1.03998458,
      "epoch": 0.2919585149556591,
      "flos": 71316726215040.0,
      "grad_norm": 0.807103294701614,
      "language_loss": 0.60243952,
      "learning_rate": 3.322305268780566e-06,
      "loss": 0.62336588,
      "num_input_tokens_seen": 104747945,
      "step": 4856,
      "time_per_iteration": 3.242067813873291
    },
    {
      "auxiliary_loss_clip": 0.01109,
      "auxiliary_loss_mlp": 0.00793662,
      "balance_loss_clip": 1.04584336,
      "balance_loss_mlp": 1.0141933,
      "epoch": 0.2920186382083271,
      "flos": 15633208419840.0,
      "grad_norm": 1.9073947903075221,
      "language_loss": 0.6844548,
      "learning_rate": 3.322013049531664e-06,
      "loss": 0.70348144,
      "num_input_tokens_seen": 104766225,
      "step": 4857,
      "time_per_iteration": 2.5312864780426025
    },
    {
      "auxiliary_loss_clip": 0.01120108,
      "auxiliary_loss_mlp": 0.0079238,
      "balance_loss_clip": 1.04651284,
      "balance_loss_mlp": 1.01410019,
      "epoch": 0.29207876146099504,
      "flos": 28366018364160.0,
      "grad_norm": 2.4218959479493045,
      "language_loss": 0.84190947,
      "learning_rate": 3.321720780151895e-06,
      "loss": 0.86103427,
      "num_input_tokens_seen": 104785345,
      "step": 4858,
      "time_per_iteration": 2.6087169647216797
    },
    {
      "auxiliary_loss_clip": 0.01134357,
      "auxiliary_loss_mlp": 0.01036867,
      "balance_loss_clip": 1.04824066,
      "balance_loss_mlp": 1.02223992,
      "epoch": 0.292138884713663,
      "flos": 21870963342720.0,
      "grad_norm": 2.0761692899655078,
      "language_loss": 0.77788353,
      "learning_rate": 3.321428460652342e-06,
      "loss": 0.79959583,
      "num_input_tokens_seen": 104804560,
      "step": 4859,
      "time_per_iteration": 3.855374336242676
    },
    {
      "auxiliary_loss_clip": 0.010975,
      "auxiliary_loss_mlp": 0.01039493,
      "balance_loss_clip": 1.04785335,
      "balance_loss_mlp": 1.0231142,
      "epoch": 0.29219900796633097,
      "flos": 20992552243200.0,
      "grad_norm": 2.9049728651643685,
      "language_loss": 0.68475497,
      "learning_rate": 3.3211360910440885e-06,
      "loss": 0.7061249,
      "num_input_tokens_seen": 104821105,
      "step": 4860,
      "time_per_iteration": 2.620515823364258
    },
    {
      "auxiliary_loss_clip": 0.01111516,
      "auxiliary_loss_mlp": 0.01042436,
      "balance_loss_clip": 1.04934347,
      "balance_loss_mlp": 1.028054,
      "epoch": 0.29225913121899894,
      "flos": 35004608133120.0,
      "grad_norm": 2.0956799027611943,
      "language_loss": 0.75221592,
      "learning_rate": 3.320843671338222e-06,
      "loss": 0.77375549,
      "num_input_tokens_seen": 104841440,
      "step": 4861,
      "time_per_iteration": 2.6450259685516357
    },
    {
      "auxiliary_loss_clip": 0.01120963,
      "auxiliary_loss_mlp": 0.01045138,
      "balance_loss_clip": 1.04619575,
      "balance_loss_mlp": 1.03071976,
      "epoch": 0.2923192544716669,
      "flos": 13515663888000.0,
      "grad_norm": 1.7534340313707848,
      "language_loss": 0.91651678,
      "learning_rate": 3.320551201545832e-06,
      "loss": 0.93817782,
      "num_input_tokens_seen": 104858210,
      "step": 4862,
      "time_per_iteration": 2.5155978202819824
    },
    {
      "auxiliary_loss_clip": 0.01122067,
      "auxiliary_loss_mlp": 0.01038437,
      "balance_loss_clip": 1.04535699,
      "balance_loss_mlp": 1.0236311,
      "epoch": 0.29237937772433487,
      "flos": 19463512141440.0,
      "grad_norm": 2.736432102024755,
      "language_loss": 0.73195994,
      "learning_rate": 3.320258681678008e-06,
      "loss": 0.75356501,
      "num_input_tokens_seen": 104875620,
      "step": 4863,
      "time_per_iteration": 3.9440033435821533
    },
    {
      "auxiliary_loss_clip": 0.010617,
      "auxiliary_loss_mlp": 0.01035244,
      "balance_loss_clip": 1.04624104,
      "balance_loss_mlp": 1.02029502,
      "epoch": 0.29243950097700283,
      "flos": 20850597694080.0,
      "grad_norm": 1.8359665266469698,
      "language_loss": 0.7779513,
      "learning_rate": 3.319966111745842e-06,
      "loss": 0.79892075,
      "num_input_tokens_seen": 104894600,
      "step": 4864,
      "time_per_iteration": 2.6766703128814697
    },
    {
      "auxiliary_loss_clip": 0.01094848,
      "auxiliary_loss_mlp": 0.01042284,
      "balance_loss_clip": 1.04344225,
      "balance_loss_mlp": 1.02491522,
      "epoch": 0.29249962422967085,
      "flos": 23584225322880.0,
      "grad_norm": 2.2018536745064647,
      "language_loss": 0.81772017,
      "learning_rate": 3.319673491760429e-06,
      "loss": 0.83909142,
      "num_input_tokens_seen": 104914530,
      "step": 4865,
      "time_per_iteration": 2.5476834774017334
    },
    {
      "auxiliary_loss_clip": 0.01077071,
      "auxiliary_loss_mlp": 0.01040087,
      "balance_loss_clip": 1.04262638,
      "balance_loss_mlp": 1.02388656,
      "epoch": 0.2925597474823388,
      "flos": 22273342473600.0,
      "grad_norm": 2.162105808028683,
      "language_loss": 0.85104614,
      "learning_rate": 3.3193808217328645e-06,
      "loss": 0.87221777,
      "num_input_tokens_seen": 104933460,
      "step": 4866,
      "time_per_iteration": 2.6124978065490723
    },
    {
      "auxiliary_loss_clip": 0.0110476,
      "auxiliary_loss_mlp": 0.01034294,
      "balance_loss_clip": 1.04799366,
      "balance_loss_mlp": 1.01959538,
      "epoch": 0.2926198707350068,
      "flos": 34456108475520.0,
      "grad_norm": 1.6778864828391444,
      "language_loss": 0.75531071,
      "learning_rate": 3.3190881016742476e-06,
      "loss": 0.77670127,
      "num_input_tokens_seen": 104954495,
      "step": 4867,
      "time_per_iteration": 3.9887735843658447
    },
    {
      "auxiliary_loss_clip": 0.01073313,
      "auxiliary_loss_mlp": 0.01045394,
      "balance_loss_clip": 1.04441845,
      "balance_loss_mlp": 1.02963471,
      "epoch": 0.29267999398767475,
      "flos": 20704153944960.0,
      "grad_norm": 1.77119559111657,
      "language_loss": 0.73429871,
      "learning_rate": 3.3187953315956776e-06,
      "loss": 0.75548583,
      "num_input_tokens_seen": 104971915,
      "step": 4868,
      "time_per_iteration": 3.9865503311157227
    },
    {
      "auxiliary_loss_clip": 0.01084403,
      "auxiliary_loss_mlp": 0.01040676,
      "balance_loss_clip": 1.04842043,
      "balance_loss_mlp": 1.02386785,
      "epoch": 0.2927401172403427,
      "flos": 18368667642240.0,
      "grad_norm": 1.3192807070804216,
      "language_loss": 0.74628031,
      "learning_rate": 3.3185025115082566e-06,
      "loss": 0.7675311,
      "num_input_tokens_seen": 104991335,
      "step": 4869,
      "time_per_iteration": 2.599173069000244
    },
    {
      "auxiliary_loss_clip": 0.01107454,
      "auxiliary_loss_mlp": 0.01035869,
      "balance_loss_clip": 1.04796124,
      "balance_loss_mlp": 1.02022839,
      "epoch": 0.2928002404930107,
      "flos": 26104041244800.0,
      "grad_norm": 1.4750428906256239,
      "language_loss": 0.76594031,
      "learning_rate": 3.318209641423088e-06,
      "loss": 0.78737354,
      "num_input_tokens_seen": 105012015,
      "step": 4870,
      "time_per_iteration": 2.5542151927948
    },
    {
      "auxiliary_loss_clip": 0.01122903,
      "auxiliary_loss_mlp": 0.01044719,
      "balance_loss_clip": 1.04645526,
      "balance_loss_mlp": 1.02818465,
      "epoch": 0.29286036374567864,
      "flos": 21324726241920.0,
      "grad_norm": 2.032644808503494,
      "language_loss": 0.67701113,
      "learning_rate": 3.3179167213512777e-06,
      "loss": 0.69868743,
      "num_input_tokens_seen": 105031460,
      "step": 4871,
      "time_per_iteration": 2.5021960735321045
    },
    {
      "auxiliary_loss_clip": 0.01102426,
      "auxiliary_loss_mlp": 0.01047874,
      "balance_loss_clip": 1.04175055,
      "balance_loss_mlp": 1.03131557,
      "epoch": 0.2929204869983466,
      "flos": 29569492569600.0,
      "grad_norm": 1.7439313058417694,
      "language_loss": 0.77657032,
      "learning_rate": 3.317623751303933e-06,
      "loss": 0.79807335,
      "num_input_tokens_seen": 105052965,
      "step": 4872,
      "time_per_iteration": 2.5804905891418457
    },
    {
      "auxiliary_loss_clip": 0.01074211,
      "auxiliary_loss_mlp": 0.01041091,
      "balance_loss_clip": 1.05210352,
      "balance_loss_mlp": 1.02417541,
      "epoch": 0.2929806102510146,
      "flos": 19058259922560.0,
      "grad_norm": 1.9746360033515076,
      "language_loss": 0.72462058,
      "learning_rate": 3.317330731292164e-06,
      "loss": 0.74577367,
      "num_input_tokens_seen": 105071840,
      "step": 4873,
      "time_per_iteration": 2.6133391857147217
    },
    {
      "auxiliary_loss_clip": 0.01123673,
      "auxiliary_loss_mlp": 0.01039377,
      "balance_loss_clip": 1.04591727,
      "balance_loss_mlp": 1.0232836,
      "epoch": 0.29304073350368254,
      "flos": 21944221130880.0,
      "grad_norm": 1.8474491602066785,
      "language_loss": 0.78506398,
      "learning_rate": 3.3170376613270812e-06,
      "loss": 0.80669451,
      "num_input_tokens_seen": 105089445,
      "step": 4874,
      "time_per_iteration": 2.470649480819702
    },
    {
      "auxiliary_loss_clip": 0.01084308,
      "auxiliary_loss_mlp": 0.01045147,
      "balance_loss_clip": 1.04921818,
      "balance_loss_mlp": 1.02864826,
      "epoch": 0.2931008567563505,
      "flos": 15450818135040.0,
      "grad_norm": 2.431306686061909,
      "language_loss": 0.77197063,
      "learning_rate": 3.3167445414197985e-06,
      "loss": 0.79326522,
      "num_input_tokens_seen": 105106210,
      "step": 4875,
      "time_per_iteration": 2.5743253231048584
    },
    {
      "auxiliary_loss_clip": 0.01129272,
      "auxiliary_loss_mlp": 0.01033662,
      "balance_loss_clip": 1.05002117,
      "balance_loss_mlp": 1.01819503,
      "epoch": 0.29316098000901847,
      "flos": 16983162288000.0,
      "grad_norm": 1.6622948697815456,
      "language_loss": 0.69379979,
      "learning_rate": 3.316451371581431e-06,
      "loss": 0.71542919,
      "num_input_tokens_seen": 105124200,
      "step": 4876,
      "time_per_iteration": 2.453495502471924
    },
    {
      "auxiliary_loss_clip": 0.0111459,
      "auxiliary_loss_mlp": 0.01046379,
      "balance_loss_clip": 1.04559231,
      "balance_loss_mlp": 1.03116751,
      "epoch": 0.29322110326168643,
      "flos": 16357705741440.0,
      "grad_norm": 2.015632254337879,
      "language_loss": 0.81755853,
      "learning_rate": 3.316158151823096e-06,
      "loss": 0.83916819,
      "num_input_tokens_seen": 105140400,
      "step": 4877,
      "time_per_iteration": 2.470668077468872
    },
    {
      "auxiliary_loss_clip": 0.01128089,
      "auxiliary_loss_mlp": 0.01040921,
      "balance_loss_clip": 1.04780734,
      "balance_loss_mlp": 1.02525711,
      "epoch": 0.29328122651435445,
      "flos": 13990869843840.0,
      "grad_norm": 2.688030710122657,
      "language_loss": 0.67930096,
      "learning_rate": 3.315864882155911e-06,
      "loss": 0.70099103,
      "num_input_tokens_seen": 105157535,
      "step": 4878,
      "time_per_iteration": 2.4566025733947754
    },
    {
      "auxiliary_loss_clip": 0.01102792,
      "auxiliary_loss_mlp": 0.01046594,
      "balance_loss_clip": 1.04958808,
      "balance_loss_mlp": 1.03110838,
      "epoch": 0.2933413497670224,
      "flos": 25264593423360.0,
      "grad_norm": 1.7738488176787937,
      "language_loss": 0.73789984,
      "learning_rate": 3.3155715625909982e-06,
      "loss": 0.75939369,
      "num_input_tokens_seen": 105175185,
      "step": 4879,
      "time_per_iteration": 2.5683116912841797
    },
    {
      "auxiliary_loss_clip": 0.01094622,
      "auxiliary_loss_mlp": 0.00796211,
      "balance_loss_clip": 1.04959798,
      "balance_loss_mlp": 1.01395142,
      "epoch": 0.2934014730196904,
      "flos": 32123746656000.0,
      "grad_norm": 4.170465141648543,
      "language_loss": 0.66318065,
      "learning_rate": 3.3152781931394803e-06,
      "loss": 0.68208897,
      "num_input_tokens_seen": 105194540,
      "step": 4880,
      "time_per_iteration": 2.629755973815918
    },
    {
      "auxiliary_loss_clip": 0.01122982,
      "auxiliary_loss_mlp": 0.01050818,
      "balance_loss_clip": 1.04577231,
      "balance_loss_mlp": 1.03466535,
      "epoch": 0.29346159627235835,
      "flos": 24352498344960.0,
      "grad_norm": 2.2275853353298736,
      "language_loss": 0.70025432,
      "learning_rate": 3.314984773812481e-06,
      "loss": 0.72199231,
      "num_input_tokens_seen": 105213215,
      "step": 4881,
      "time_per_iteration": 2.5017483234405518
    },
    {
      "auxiliary_loss_clip": 0.01105357,
      "auxiliary_loss_mlp": 0.00793235,
      "balance_loss_clip": 1.04586196,
      "balance_loss_mlp": 1.01241755,
      "epoch": 0.2935217195250263,
      "flos": 22746752749440.0,
      "grad_norm": 1.6120395084743337,
      "language_loss": 0.83769453,
      "learning_rate": 3.314691304621127e-06,
      "loss": 0.85668045,
      "num_input_tokens_seen": 105231585,
      "step": 4882,
      "time_per_iteration": 2.520005226135254
    },
    {
      "auxiliary_loss_clip": 0.01139105,
      "auxiliary_loss_mlp": 0.01041903,
      "balance_loss_clip": 1.04858851,
      "balance_loss_mlp": 1.02592885,
      "epoch": 0.2935818427776943,
      "flos": 21725561088000.0,
      "grad_norm": 2.456426184981771,
      "language_loss": 0.71473742,
      "learning_rate": 3.314397785576548e-06,
      "loss": 0.73654747,
      "num_input_tokens_seen": 105250120,
      "step": 4883,
      "time_per_iteration": 2.4620766639709473
    },
    {
      "auxiliary_loss_clip": 0.01111876,
      "auxiliary_loss_mlp": 0.01039114,
      "balance_loss_clip": 1.04647648,
      "balance_loss_mlp": 1.02358127,
      "epoch": 0.29364196603036224,
      "flos": 23804968354560.0,
      "grad_norm": 2.0907261016749437,
      "language_loss": 0.92236882,
      "learning_rate": 3.3141042166898726e-06,
      "loss": 0.94387877,
      "num_input_tokens_seen": 105266065,
      "step": 4884,
      "time_per_iteration": 2.5337040424346924
    },
    {
      "auxiliary_loss_clip": 0.01128784,
      "auxiliary_loss_mlp": 0.01038558,
      "balance_loss_clip": 1.05069876,
      "balance_loss_mlp": 1.02288151,
      "epoch": 0.2937020892830302,
      "flos": 23470064922240.0,
      "grad_norm": 6.138772849755271,
      "language_loss": 0.73307711,
      "learning_rate": 3.313810597972234e-06,
      "loss": 0.75475055,
      "num_input_tokens_seen": 105282155,
      "step": 4885,
      "time_per_iteration": 2.4846506118774414
    },
    {
      "auxiliary_loss_clip": 0.01117812,
      "auxiliary_loss_mlp": 0.01050858,
      "balance_loss_clip": 1.04821086,
      "balance_loss_mlp": 1.03486025,
      "epoch": 0.2937622125356982,
      "flos": 24272740195200.0,
      "grad_norm": 1.9289476504170726,
      "language_loss": 0.8539021,
      "learning_rate": 3.3135169294347655e-06,
      "loss": 0.87558872,
      "num_input_tokens_seen": 105299225,
      "step": 4886,
      "time_per_iteration": 2.495817184448242
    },
    {
      "auxiliary_loss_clip": 0.01103479,
      "auxiliary_loss_mlp": 0.01040423,
      "balance_loss_clip": 1.04361868,
      "balance_loss_mlp": 1.02521205,
      "epoch": 0.29382233578836614,
      "flos": 20662461233280.0,
      "grad_norm": 2.501634940459168,
      "language_loss": 0.77050245,
      "learning_rate": 3.313223211088603e-06,
      "loss": 0.79194146,
      "num_input_tokens_seen": 105315710,
      "step": 4887,
      "time_per_iteration": 2.522519826889038
    },
    {
      "auxiliary_loss_clip": 0.01114905,
      "auxiliary_loss_mlp": 0.01045588,
      "balance_loss_clip": 1.05093074,
      "balance_loss_mlp": 1.03001952,
      "epoch": 0.2938824590410341,
      "flos": 16545052103040.0,
      "grad_norm": 2.7373356755599896,
      "language_loss": 0.79280293,
      "learning_rate": 3.3129294429448855e-06,
      "loss": 0.81440789,
      "num_input_tokens_seen": 105333505,
      "step": 4888,
      "time_per_iteration": 2.4880967140197754
    },
    {
      "auxiliary_loss_clip": 0.0110834,
      "auxiliary_loss_mlp": 0.01035148,
      "balance_loss_clip": 1.04861164,
      "balance_loss_mlp": 1.01996112,
      "epoch": 0.29394258229370207,
      "flos": 37925474382720.0,
      "grad_norm": 1.4438986865716363,
      "language_loss": 0.55357742,
      "learning_rate": 3.3126356250147517e-06,
      "loss": 0.57501233,
      "num_input_tokens_seen": 105355605,
      "step": 4889,
      "time_per_iteration": 2.6691501140594482
    },
    {
      "auxiliary_loss_clip": 0.01125455,
      "auxiliary_loss_mlp": 0.01039283,
      "balance_loss_clip": 1.04619408,
      "balance_loss_mlp": 1.02261806,
      "epoch": 0.29400270554637004,
      "flos": 20044690197120.0,
      "grad_norm": 1.8285535452983845,
      "language_loss": 0.84365165,
      "learning_rate": 3.3123417573093434e-06,
      "loss": 0.86529905,
      "num_input_tokens_seen": 105374225,
      "step": 4890,
      "time_per_iteration": 2.4761905670166016
    },
    {
      "auxiliary_loss_clip": 0.01129133,
      "auxiliary_loss_mlp": 0.01047197,
      "balance_loss_clip": 1.04979587,
      "balance_loss_mlp": 1.03152084,
      "epoch": 0.294062828799038,
      "flos": 15266380775040.0,
      "grad_norm": 1.8088010657157847,
      "language_loss": 0.72714233,
      "learning_rate": 3.3120478398398046e-06,
      "loss": 0.7489056,
      "num_input_tokens_seen": 105391565,
      "step": 4891,
      "time_per_iteration": 2.46941876411438
    },
    {
      "auxiliary_loss_clip": 0.0113805,
      "auxiliary_loss_mlp": 0.01044374,
      "balance_loss_clip": 1.04875684,
      "balance_loss_mlp": 1.02786374,
      "epoch": 0.294122952051706,
      "flos": 22747147799040.0,
      "grad_norm": 1.727525484627151,
      "language_loss": 0.7727828,
      "learning_rate": 3.3117538726172797e-06,
      "loss": 0.79460704,
      "num_input_tokens_seen": 105409840,
      "step": 4892,
      "time_per_iteration": 2.4586985111236572
    },
    {
      "auxiliary_loss_clip": 0.01131748,
      "auxiliary_loss_mlp": 0.01034006,
      "balance_loss_clip": 1.0444119,
      "balance_loss_mlp": 1.01788926,
      "epoch": 0.294183075304374,
      "flos": 24972891073920.0,
      "grad_norm": 3.1238776553728007,
      "language_loss": 0.78405607,
      "learning_rate": 3.3114598556529164e-06,
      "loss": 0.80571359,
      "num_input_tokens_seen": 105428645,
      "step": 4893,
      "time_per_iteration": 2.4751572608947754
    },
    {
      "auxiliary_loss_clip": 0.01099542,
      "auxiliary_loss_mlp": 0.01042052,
      "balance_loss_clip": 1.04928422,
      "balance_loss_mlp": 1.02630448,
      "epoch": 0.29424319855704195,
      "flos": 30952986762240.0,
      "grad_norm": 1.931257378100444,
      "language_loss": 0.85214335,
      "learning_rate": 3.311165788957864e-06,
      "loss": 0.87355924,
      "num_input_tokens_seen": 105447480,
      "step": 4894,
      "time_per_iteration": 2.6131014823913574
    },
    {
      "auxiliary_loss_clip": 0.01121813,
      "auxiliary_loss_mlp": 0.01039262,
      "balance_loss_clip": 1.0467608,
      "balance_loss_mlp": 1.02369308,
      "epoch": 0.2943033218097099,
      "flos": 15231583474560.0,
      "grad_norm": 2.5654195386417964,
      "language_loss": 0.90317953,
      "learning_rate": 3.310871672543274e-06,
      "loss": 0.92479026,
      "num_input_tokens_seen": 105464600,
      "step": 4895,
      "time_per_iteration": 2.4986560344696045
    },
    {
      "auxiliary_loss_clip": 0.0112748,
      "auxiliary_loss_mlp": 0.01042008,
      "balance_loss_clip": 1.04756224,
      "balance_loss_mlp": 1.02504468,
      "epoch": 0.2943634450623779,
      "flos": 21725884310400.0,
      "grad_norm": 1.8595131465716532,
      "language_loss": 0.86807722,
      "learning_rate": 3.3105775064202982e-06,
      "loss": 0.88977206,
      "num_input_tokens_seen": 105481510,
      "step": 4896,
      "time_per_iteration": 2.4746692180633545
    },
    {
      "auxiliary_loss_clip": 0.01126894,
      "auxiliary_loss_mlp": 0.01044427,
      "balance_loss_clip": 1.04753733,
      "balance_loss_mlp": 1.02714169,
      "epoch": 0.29442356831504585,
      "flos": 22602104680320.0,
      "grad_norm": 1.781062642290452,
      "language_loss": 0.73368102,
      "learning_rate": 3.3102832906000924e-06,
      "loss": 0.75539422,
      "num_input_tokens_seen": 105501390,
      "step": 4897,
      "time_per_iteration": 3.878309488296509
    },
    {
      "auxiliary_loss_clip": 0.01122727,
      "auxiliary_loss_mlp": 0.01044693,
      "balance_loss_clip": 1.0461216,
      "balance_loss_mlp": 1.02657342,
      "epoch": 0.2944836915677138,
      "flos": 20011401267840.0,
      "grad_norm": 1.969961281753076,
      "language_loss": 0.73796952,
      "learning_rate": 3.309989025093813e-06,
      "loss": 0.75964367,
      "num_input_tokens_seen": 105519600,
      "step": 4898,
      "time_per_iteration": 2.4916224479675293
    },
    {
      "auxiliary_loss_clip": 0.01129737,
      "auxiliary_loss_mlp": 0.01040618,
      "balance_loss_clip": 1.05199862,
      "balance_loss_mlp": 1.02171135,
      "epoch": 0.2945438148203818,
      "flos": 20045875345920.0,
      "grad_norm": 2.4246476949459534,
      "language_loss": 0.69957542,
      "learning_rate": 3.309694709912618e-06,
      "loss": 0.72127903,
      "num_input_tokens_seen": 105535970,
      "step": 4899,
      "time_per_iteration": 2.4598538875579834
    },
    {
      "auxiliary_loss_clip": 0.01112202,
      "auxiliary_loss_mlp": 0.00799019,
      "balance_loss_clip": 1.04465938,
      "balance_loss_mlp": 1.02145696,
      "epoch": 0.29460393807304974,
      "flos": 23733542160000.0,
      "grad_norm": 1.9578903959633245,
      "language_loss": 0.79285479,
      "learning_rate": 3.3094003450676685e-06,
      "loss": 0.81196702,
      "num_input_tokens_seen": 105556735,
      "step": 4900,
      "time_per_iteration": 2.5395562648773193
    },
    {
      "auxiliary_loss_clip": 0.01096787,
      "auxiliary_loss_mlp": 0.01055943,
      "balance_loss_clip": 1.04089987,
      "balance_loss_mlp": 1.03783512,
      "epoch": 0.2946640613257177,
      "flos": 14976079056000.0,
      "grad_norm": 1.6438298621403855,
      "language_loss": 0.80785561,
      "learning_rate": 3.3091059305701268e-06,
      "loss": 0.82938284,
      "num_input_tokens_seen": 105574875,
      "step": 4901,
      "time_per_iteration": 2.474019765853882
    },
    {
      "auxiliary_loss_clip": 0.01104026,
      "auxiliary_loss_mlp": 0.01032829,
      "balance_loss_clip": 1.04525256,
      "balance_loss_mlp": 1.01814222,
      "epoch": 0.2947241845783857,
      "flos": 24243904552320.0,
      "grad_norm": 2.4398297654725662,
      "language_loss": 0.58460808,
      "learning_rate": 3.308811466431157e-06,
      "loss": 0.6059767,
      "num_input_tokens_seen": 105594225,
      "step": 4902,
      "time_per_iteration": 3.9335074424743652
    },
    {
      "auxiliary_loss_clip": 0.01113038,
      "auxiliary_loss_mlp": 0.01039568,
      "balance_loss_clip": 1.04633045,
      "balance_loss_mlp": 1.02439249,
      "epoch": 0.29478430783105364,
      "flos": 19938394874880.0,
      "grad_norm": 1.7611037090354738,
      "language_loss": 0.7548719,
      "learning_rate": 3.308516952661925e-06,
      "loss": 0.77639794,
      "num_input_tokens_seen": 105614000,
      "step": 4903,
      "time_per_iteration": 2.494096517562866
    },
    {
      "auxiliary_loss_clip": 0.01108475,
      "auxiliary_loss_mlp": 0.01055923,
      "balance_loss_clip": 1.04559398,
      "balance_loss_mlp": 1.0368849,
      "epoch": 0.2948444310837216,
      "flos": 27381347856000.0,
      "grad_norm": 2.1661939074750842,
      "language_loss": 0.62255424,
      "learning_rate": 3.3082223892736e-06,
      "loss": 0.64419824,
      "num_input_tokens_seen": 105634575,
      "step": 4904,
      "time_per_iteration": 2.56199312210083
    },
    {
      "auxiliary_loss_clip": 0.0112356,
      "auxiliary_loss_mlp": 0.01044683,
      "balance_loss_clip": 1.04610598,
      "balance_loss_mlp": 1.02823246,
      "epoch": 0.2949045543363896,
      "flos": 23405462311680.0,
      "grad_norm": 1.44627143917231,
      "language_loss": 0.73085648,
      "learning_rate": 3.3079277762773496e-06,
      "loss": 0.75253892,
      "num_input_tokens_seen": 105654385,
      "step": 4905,
      "time_per_iteration": 3.8679378032684326
    },
    {
      "auxiliary_loss_clip": 0.01100874,
      "auxiliary_loss_mlp": 0.01042058,
      "balance_loss_clip": 1.04384816,
      "balance_loss_mlp": 1.02495158,
      "epoch": 0.2949646775890576,
      "flos": 23951483930880.0,
      "grad_norm": 1.6266736324131077,
      "language_loss": 0.81753802,
      "learning_rate": 3.3076331136843476e-06,
      "loss": 0.83896732,
      "num_input_tokens_seen": 105673570,
      "step": 4906,
      "time_per_iteration": 3.9810681343078613
    },
    {
      "auxiliary_loss_clip": 0.01089943,
      "auxiliary_loss_mlp": 0.01042727,
      "balance_loss_clip": 1.04213428,
      "balance_loss_mlp": 1.02686024,
      "epoch": 0.29502480084172555,
      "flos": 22784315397120.0,
      "grad_norm": 1.8527376765501666,
      "language_loss": 0.8753252,
      "learning_rate": 3.3073384015057667e-06,
      "loss": 0.89665186,
      "num_input_tokens_seen": 105691940,
      "step": 4907,
      "time_per_iteration": 2.584670305252075
    },
    {
      "auxiliary_loss_clip": 0.01138108,
      "auxiliary_loss_mlp": 0.01044809,
      "balance_loss_clip": 1.04699647,
      "balance_loss_mlp": 1.0272733,
      "epoch": 0.2950849240943935,
      "flos": 19646656611840.0,
      "grad_norm": 2.246177546917773,
      "language_loss": 0.81753218,
      "learning_rate": 3.307043639752782e-06,
      "loss": 0.83936131,
      "num_input_tokens_seen": 105709825,
      "step": 4908,
      "time_per_iteration": 2.4579389095306396
    },
    {
      "auxiliary_loss_clip": 0.01050203,
      "auxiliary_loss_mlp": 0.01013942,
      "balance_loss_clip": 1.01877642,
      "balance_loss_mlp": 1.01147389,
      "epoch": 0.2951450473470615,
      "flos": 71002829260800.0,
      "grad_norm": 0.8739148813883363,
      "language_loss": 0.57220006,
      "learning_rate": 3.3067488284365728e-06,
      "loss": 0.59284151,
      "num_input_tokens_seen": 105766880,
      "step": 4909,
      "time_per_iteration": 2.937335968017578
    },
    {
      "auxiliary_loss_clip": 0.01123155,
      "auxiliary_loss_mlp": 0.00794712,
      "balance_loss_clip": 1.04816902,
      "balance_loss_mlp": 1.01533914,
      "epoch": 0.29520517059972945,
      "flos": 22966310632320.0,
      "grad_norm": 1.4832857933438257,
      "language_loss": 0.86776012,
      "learning_rate": 3.3064539675683163e-06,
      "loss": 0.88693881,
      "num_input_tokens_seen": 105786875,
      "step": 4910,
      "time_per_iteration": 2.5236947536468506
    },
    {
      "auxiliary_loss_clip": 0.01119667,
      "auxiliary_loss_mlp": 0.01042815,
      "balance_loss_clip": 1.04634047,
      "balance_loss_mlp": 1.02769291,
      "epoch": 0.2952652938523974,
      "flos": 20485673470080.0,
      "grad_norm": 1.9037110659509064,
      "language_loss": 0.72981197,
      "learning_rate": 3.3061590571591946e-06,
      "loss": 0.75143677,
      "num_input_tokens_seen": 105805315,
      "step": 4911,
      "time_per_iteration": 2.493351697921753
    },
    {
      "auxiliary_loss_clip": 0.01122406,
      "auxiliary_loss_mlp": 0.01038456,
      "balance_loss_clip": 1.04755485,
      "balance_loss_mlp": 1.02301884,
      "epoch": 0.2953254171050654,
      "flos": 19646584784640.0,
      "grad_norm": 1.6861268417955038,
      "language_loss": 0.89695966,
      "learning_rate": 3.3058640972203904e-06,
      "loss": 0.91856831,
      "num_input_tokens_seen": 105825125,
      "step": 4912,
      "time_per_iteration": 2.5040273666381836
    },
    {
      "auxiliary_loss_clip": 0.01104328,
      "auxiliary_loss_mlp": 0.01052079,
      "balance_loss_clip": 1.04452348,
      "balance_loss_mlp": 1.03589022,
      "epoch": 0.29538554035773334,
      "flos": 22747973811840.0,
      "grad_norm": 1.3662345777095246,
      "language_loss": 0.83456069,
      "learning_rate": 3.3055690877630894e-06,
      "loss": 0.85612476,
      "num_input_tokens_seen": 105846085,
      "step": 4913,
      "time_per_iteration": 2.5472917556762695
    },
    {
      "auxiliary_loss_clip": 0.01133839,
      "auxiliary_loss_mlp": 0.01044669,
      "balance_loss_clip": 1.04531503,
      "balance_loss_mlp": 1.02942204,
      "epoch": 0.2954456636104013,
      "flos": 21871861182720.0,
      "grad_norm": 1.6615543958123002,
      "language_loss": 0.77000606,
      "learning_rate": 3.3052740287984765e-06,
      "loss": 0.79179108,
      "num_input_tokens_seen": 105865400,
      "step": 4914,
      "time_per_iteration": 2.482278347015381
    },
    {
      "auxiliary_loss_clip": 0.0111123,
      "auxiliary_loss_mlp": 0.01042056,
      "balance_loss_clip": 1.04507804,
      "balance_loss_mlp": 1.02577162,
      "epoch": 0.2955057868630693,
      "flos": 40442560871040.0,
      "grad_norm": 1.735280153274108,
      "language_loss": 0.81486237,
      "learning_rate": 3.3049789203377424e-06,
      "loss": 0.83639526,
      "num_input_tokens_seen": 105887920,
      "step": 4915,
      "time_per_iteration": 2.6914052963256836
    },
    {
      "auxiliary_loss_clip": 0.01063129,
      "auxiliary_loss_mlp": 0.01040463,
      "balance_loss_clip": 1.04851794,
      "balance_loss_mlp": 1.0246557,
      "epoch": 0.29556591011573724,
      "flos": 22564506119040.0,
      "grad_norm": 1.8573261972168285,
      "language_loss": 0.84808469,
      "learning_rate": 3.3046837623920772e-06,
      "loss": 0.8691206,
      "num_input_tokens_seen": 105904035,
      "step": 4916,
      "time_per_iteration": 2.663374423980713
    },
    {
      "auxiliary_loss_clip": 0.01117781,
      "auxiliary_loss_mlp": 0.01038662,
      "balance_loss_clip": 1.04346943,
      "balance_loss_mlp": 1.02290249,
      "epoch": 0.2956260333684052,
      "flos": 22089300163200.0,
      "grad_norm": 1.918522518270424,
      "language_loss": 0.69340903,
      "learning_rate": 3.3043885549726723e-06,
      "loss": 0.71497345,
      "num_input_tokens_seen": 105922685,
      "step": 4917,
      "time_per_iteration": 2.5277392864227295
    },
    {
      "auxiliary_loss_clip": 0.01116632,
      "auxiliary_loss_mlp": 0.01043496,
      "balance_loss_clip": 1.0482322,
      "balance_loss_mlp": 1.0271759,
      "epoch": 0.2956861566210732,
      "flos": 16435488643200.0,
      "grad_norm": 2.1431860523004915,
      "language_loss": 0.91212702,
      "learning_rate": 3.3040932980907226e-06,
      "loss": 0.93372834,
      "num_input_tokens_seen": 105940425,
      "step": 4918,
      "time_per_iteration": 2.5113093852996826
    },
    {
      "auxiliary_loss_clip": 0.01136974,
      "auxiliary_loss_mlp": 0.01040716,
      "balance_loss_clip": 1.04891634,
      "balance_loss_mlp": 1.02447987,
      "epoch": 0.2957462798737412,
      "flos": 25812087500160.0,
      "grad_norm": 1.9192377883004035,
      "language_loss": 0.72685957,
      "learning_rate": 3.303797991757425e-06,
      "loss": 0.74863648,
      "num_input_tokens_seen": 105960550,
      "step": 4919,
      "time_per_iteration": 2.501753091812134
    },
    {
      "auxiliary_loss_clip": 0.01111154,
      "auxiliary_loss_mlp": 0.01045876,
      "balance_loss_clip": 1.04485106,
      "balance_loss_mlp": 1.02973485,
      "epoch": 0.29580640312640916,
      "flos": 16690849407360.0,
      "grad_norm": 1.663954480290254,
      "language_loss": 0.75656331,
      "learning_rate": 3.3035026359839763e-06,
      "loss": 0.77813363,
      "num_input_tokens_seen": 105978820,
      "step": 4920,
      "time_per_iteration": 2.5170164108276367
    },
    {
      "auxiliary_loss_clip": 0.01114554,
      "auxiliary_loss_mlp": 0.01048129,
      "balance_loss_clip": 1.05274725,
      "balance_loss_mlp": 1.03098667,
      "epoch": 0.2958665263790771,
      "flos": 23945594100480.0,
      "grad_norm": 1.9891320726188144,
      "language_loss": 0.68606687,
      "learning_rate": 3.3032072307815774e-06,
      "loss": 0.7076937,
      "num_input_tokens_seen": 105997545,
      "step": 4921,
      "time_per_iteration": 2.547312021255493
    },
    {
      "auxiliary_loss_clip": 0.01113835,
      "auxiliary_loss_mlp": 0.0104726,
      "balance_loss_clip": 1.04876328,
      "balance_loss_mlp": 1.02924728,
      "epoch": 0.2959266496317451,
      "flos": 18478410670080.0,
      "grad_norm": 1.8585661098005761,
      "language_loss": 0.74796164,
      "learning_rate": 3.3029117761614298e-06,
      "loss": 0.76957256,
      "num_input_tokens_seen": 106015320,
      "step": 4922,
      "time_per_iteration": 2.5280961990356445
    },
    {
      "auxiliary_loss_clip": 0.01140753,
      "auxiliary_loss_mlp": 0.00796917,
      "balance_loss_clip": 1.04812646,
      "balance_loss_mlp": 1.01910353,
      "epoch": 0.29598677288441305,
      "flos": 25957489754880.0,
      "grad_norm": 1.8369715669648106,
      "language_loss": 0.7659955,
      "learning_rate": 3.302616272134737e-06,
      "loss": 0.7853722,
      "num_input_tokens_seen": 106034555,
      "step": 4923,
      "time_per_iteration": 2.5392491817474365
    },
    {
      "auxiliary_loss_clip": 0.01109356,
      "auxiliary_loss_mlp": 0.01042321,
      "balance_loss_clip": 1.04817212,
      "balance_loss_mlp": 1.02591789,
      "epoch": 0.296046896137081,
      "flos": 25155999630720.0,
      "grad_norm": 1.5358678528472514,
      "language_loss": 0.86043108,
      "learning_rate": 3.3023207187127042e-06,
      "loss": 0.88194788,
      "num_input_tokens_seen": 106054200,
      "step": 4924,
      "time_per_iteration": 2.5724589824676514
    },
    {
      "auxiliary_loss_clip": 0.01124706,
      "auxiliary_loss_mlp": 0.01036357,
      "balance_loss_clip": 1.04624474,
      "balance_loss_mlp": 1.01987004,
      "epoch": 0.296107019389749,
      "flos": 21761148487680.0,
      "grad_norm": 1.5506582979546162,
      "language_loss": 0.82045966,
      "learning_rate": 3.3020251159065396e-06,
      "loss": 0.84207028,
      "num_input_tokens_seen": 106074700,
      "step": 4925,
      "time_per_iteration": 2.493276596069336
    },
    {
      "auxiliary_loss_clip": 0.01073246,
      "auxiliary_loss_mlp": 0.0105467,
      "balance_loss_clip": 1.04146099,
      "balance_loss_mlp": 1.0360496,
      "epoch": 0.29616714264241695,
      "flos": 17960039544960.0,
      "grad_norm": 2.765801889007856,
      "language_loss": 0.85679293,
      "learning_rate": 3.301729463727452e-06,
      "loss": 0.87807202,
      "num_input_tokens_seen": 106091415,
      "step": 4926,
      "time_per_iteration": 2.5829362869262695
    },
    {
      "auxiliary_loss_clip": 0.01102715,
      "auxiliary_loss_mlp": 0.01034584,
      "balance_loss_clip": 1.04951811,
      "balance_loss_mlp": 1.0181334,
      "epoch": 0.2962272658950849,
      "flos": 15012779777280.0,
      "grad_norm": 1.8693649389657263,
      "language_loss": 0.85876894,
      "learning_rate": 3.3014337621866527e-06,
      "loss": 0.88014197,
      "num_input_tokens_seen": 106109135,
      "step": 4927,
      "time_per_iteration": 2.5275156497955322
    },
    {
      "auxiliary_loss_clip": 0.01124483,
      "auxiliary_loss_mlp": 0.01035881,
      "balance_loss_clip": 1.04768372,
      "balance_loss_mlp": 1.02021694,
      "epoch": 0.2962873891477529,
      "flos": 14720861946240.0,
      "grad_norm": 1.9016280820736657,
      "language_loss": 0.80796528,
      "learning_rate": 3.3011380112953553e-06,
      "loss": 0.82956898,
      "num_input_tokens_seen": 106125750,
      "step": 4928,
      "time_per_iteration": 2.4751455783843994
    },
    {
      "auxiliary_loss_clip": 0.01118862,
      "auxiliary_loss_mlp": 0.01045489,
      "balance_loss_clip": 1.04860318,
      "balance_loss_mlp": 1.025617,
      "epoch": 0.29634751240042084,
      "flos": 26723787528960.0,
      "grad_norm": 2.727473683610809,
      "language_loss": 0.73132175,
      "learning_rate": 3.300842211064773e-06,
      "loss": 0.75296527,
      "num_input_tokens_seen": 106142835,
      "step": 4929,
      "time_per_iteration": 2.5643842220306396
    },
    {
      "auxiliary_loss_clip": 0.01112046,
      "auxiliary_loss_mlp": 0.01051851,
      "balance_loss_clip": 1.04593229,
      "balance_loss_mlp": 1.03267026,
      "epoch": 0.2964076356530888,
      "flos": 14571293713920.0,
      "grad_norm": 2.5975991010604322,
      "language_loss": 0.72156721,
      "learning_rate": 3.3005463615061246e-06,
      "loss": 0.74320614,
      "num_input_tokens_seen": 106160680,
      "step": 4930,
      "time_per_iteration": 2.5074079036712646
    },
    {
      "auxiliary_loss_clip": 0.01024303,
      "auxiliary_loss_mlp": 0.01012113,
      "balance_loss_clip": 1.03160453,
      "balance_loss_mlp": 1.00978816,
      "epoch": 0.29646775890575683,
      "flos": 63104315063040.0,
      "grad_norm": 0.839756314478916,
      "language_loss": 0.60666978,
      "learning_rate": 3.3002504626306275e-06,
      "loss": 0.62703395,
      "num_input_tokens_seen": 106224415,
      "step": 4931,
      "time_per_iteration": 3.101501941680908
    },
    {
      "auxiliary_loss_clip": 0.01016519,
      "auxiliary_loss_mlp": 0.0100333,
      "balance_loss_clip": 1.03191793,
      "balance_loss_mlp": 1.00133944,
      "epoch": 0.2965278821584248,
      "flos": 63067686168960.0,
      "grad_norm": 0.7358175170088151,
      "language_loss": 0.52359688,
      "learning_rate": 3.2999545144495023e-06,
      "loss": 0.54379535,
      "num_input_tokens_seen": 106279140,
      "step": 4932,
      "time_per_iteration": 3.108490467071533
    },
    {
      "auxiliary_loss_clip": 0.01120713,
      "auxiliary_loss_mlp": 0.01040023,
      "balance_loss_clip": 1.04415369,
      "balance_loss_mlp": 1.02402496,
      "epoch": 0.29658800541109276,
      "flos": 23768734510080.0,
      "grad_norm": 1.8313847156801686,
      "language_loss": 0.81668144,
      "learning_rate": 3.299658516973972e-06,
      "loss": 0.83828884,
      "num_input_tokens_seen": 106298190,
      "step": 4933,
      "time_per_iteration": 2.508192777633667
    },
    {
      "auxiliary_loss_clip": 0.0108748,
      "auxiliary_loss_mlp": 0.01033938,
      "balance_loss_clip": 1.04446614,
      "balance_loss_mlp": 1.0176537,
      "epoch": 0.2966481286637607,
      "flos": 23988543788160.0,
      "grad_norm": 1.6980827793485707,
      "language_loss": 0.75080633,
      "learning_rate": 3.299362470215261e-06,
      "loss": 0.77202052,
      "num_input_tokens_seen": 106319065,
      "step": 4934,
      "time_per_iteration": 2.5856339931488037
    },
    {
      "auxiliary_loss_clip": 0.01117362,
      "auxiliary_loss_mlp": 0.010438,
      "balance_loss_clip": 1.04535282,
      "balance_loss_mlp": 1.02688408,
      "epoch": 0.2967082519164287,
      "flos": 17165157523200.0,
      "grad_norm": 2.561484325445196,
      "language_loss": 0.62242049,
      "learning_rate": 3.299066374184594e-06,
      "loss": 0.64403206,
      "num_input_tokens_seen": 106338040,
      "step": 4935,
      "time_per_iteration": 2.5256917476654053
    },
    {
      "auxiliary_loss_clip": 0.01120715,
      "auxiliary_loss_mlp": 0.01044921,
      "balance_loss_clip": 1.04529619,
      "balance_loss_mlp": 1.02814782,
      "epoch": 0.29676837516909665,
      "flos": 29387712816000.0,
      "grad_norm": 1.4627948897126581,
      "language_loss": 0.79835713,
      "learning_rate": 3.2987702288932e-06,
      "loss": 0.82001352,
      "num_input_tokens_seen": 106358900,
      "step": 4936,
      "time_per_iteration": 3.982489585876465
    },
    {
      "auxiliary_loss_clip": 0.01092786,
      "auxiliary_loss_mlp": 0.01043847,
      "balance_loss_clip": 1.04642642,
      "balance_loss_mlp": 1.02703881,
      "epoch": 0.2968284984217646,
      "flos": 34751222616960.0,
      "grad_norm": 1.5205678510468719,
      "language_loss": 0.74239337,
      "learning_rate": 3.298474034352309e-06,
      "loss": 0.76375973,
      "num_input_tokens_seen": 106381805,
      "step": 4937,
      "time_per_iteration": 2.709940195083618
    },
    {
      "auxiliary_loss_clip": 0.01085801,
      "auxiliary_loss_mlp": 0.01038702,
      "balance_loss_clip": 1.04770231,
      "balance_loss_mlp": 1.02221584,
      "epoch": 0.2968886216744326,
      "flos": 21544104556800.0,
      "grad_norm": 1.749637238471787,
      "language_loss": 0.78230458,
      "learning_rate": 3.2981777905731526e-06,
      "loss": 0.80354965,
      "num_input_tokens_seen": 106402365,
      "step": 4938,
      "time_per_iteration": 2.591019630432129
    },
    {
      "auxiliary_loss_clip": 0.01112082,
      "auxiliary_loss_mlp": 0.01053028,
      "balance_loss_clip": 1.04882312,
      "balance_loss_mlp": 1.03456235,
      "epoch": 0.29694874492710055,
      "flos": 12787323811200.0,
      "grad_norm": 2.5581979510543817,
      "language_loss": 0.76968861,
      "learning_rate": 3.297881497566964e-06,
      "loss": 0.79133964,
      "num_input_tokens_seen": 106419800,
      "step": 4939,
      "time_per_iteration": 2.501011848449707
    },
    {
      "auxiliary_loss_clip": 0.01103574,
      "auxiliary_loss_mlp": 0.01042692,
      "balance_loss_clip": 1.04527116,
      "balance_loss_mlp": 1.02581215,
      "epoch": 0.2970088681797685,
      "flos": 24569973239040.0,
      "grad_norm": 1.5268105538887367,
      "language_loss": 0.77793717,
      "learning_rate": 3.297585155344979e-06,
      "loss": 0.79939985,
      "num_input_tokens_seen": 106440300,
      "step": 4940,
      "time_per_iteration": 3.985199451446533
    },
    {
      "auxiliary_loss_clip": 0.01110712,
      "auxiliary_loss_mlp": 0.01042539,
      "balance_loss_clip": 1.04663587,
      "balance_loss_mlp": 1.02454996,
      "epoch": 0.2970689914324365,
      "flos": 23659171050240.0,
      "grad_norm": 1.5910322643189212,
      "language_loss": 0.75560927,
      "learning_rate": 3.297288763918435e-06,
      "loss": 0.77714175,
      "num_input_tokens_seen": 106460035,
      "step": 4941,
      "time_per_iteration": 2.5265209674835205
    },
    {
      "auxiliary_loss_clip": 0.01129836,
      "auxiliary_loss_mlp": 0.01052822,
      "balance_loss_clip": 1.04919887,
      "balance_loss_mlp": 1.03523827,
      "epoch": 0.29712911468510445,
      "flos": 39670301439360.0,
      "grad_norm": 2.2224279724879494,
      "language_loss": 0.736848,
      "learning_rate": 3.2969923232985712e-06,
      "loss": 0.75867462,
      "num_input_tokens_seen": 106481095,
      "step": 4942,
      "time_per_iteration": 2.6539859771728516
    },
    {
      "auxiliary_loss_clip": 0.01103854,
      "auxiliary_loss_mlp": 0.01049522,
      "balance_loss_clip": 1.04729104,
      "balance_loss_mlp": 1.03149712,
      "epoch": 0.2971892379377724,
      "flos": 26395312631040.0,
      "grad_norm": 1.8462232639750258,
      "language_loss": 0.70156109,
      "learning_rate": 3.2966958334966287e-06,
      "loss": 0.72309488,
      "num_input_tokens_seen": 106501590,
      "step": 4943,
      "time_per_iteration": 2.584354877471924
    },
    {
      "auxiliary_loss_clip": 0.0111667,
      "auxiliary_loss_mlp": 0.01038744,
      "balance_loss_clip": 1.04889107,
      "balance_loss_mlp": 1.02176857,
      "epoch": 0.2972493611904404,
      "flos": 17603195880960.0,
      "grad_norm": 5.43307363628931,
      "language_loss": 0.80035192,
      "learning_rate": 3.2963992945238497e-06,
      "loss": 0.82190609,
      "num_input_tokens_seen": 106519430,
      "step": 4944,
      "time_per_iteration": 3.8918614387512207
    },
    {
      "auxiliary_loss_clip": 0.01109374,
      "auxiliary_loss_mlp": 0.01046265,
      "balance_loss_clip": 1.04611421,
      "balance_loss_mlp": 1.03054166,
      "epoch": 0.2973094844431084,
      "flos": 20412774817920.0,
      "grad_norm": 2.074660948512871,
      "language_loss": 0.83114612,
      "learning_rate": 3.2961027063914795e-06,
      "loss": 0.8527025,
      "num_input_tokens_seen": 106535870,
      "step": 4945,
      "time_per_iteration": 3.9237940311431885
    },
    {
      "auxiliary_loss_clip": 0.01084488,
      "auxiliary_loss_mlp": 0.01042055,
      "balance_loss_clip": 1.0454663,
      "balance_loss_mlp": 1.02603281,
      "epoch": 0.29736960769577636,
      "flos": 17493488766720.0,
      "grad_norm": 1.7190780776521672,
      "language_loss": 0.6724177,
      "learning_rate": 3.2958060691107654e-06,
      "loss": 0.69368315,
      "num_input_tokens_seen": 106553560,
      "step": 4946,
      "time_per_iteration": 2.5634496212005615
    },
    {
      "auxiliary_loss_clip": 0.01120952,
      "auxiliary_loss_mlp": 0.00796396,
      "balance_loss_clip": 1.04690695,
      "balance_loss_mlp": 1.01498115,
      "epoch": 0.2974297309484443,
      "flos": 26103969417600.0,
      "grad_norm": 2.2294357679976953,
      "language_loss": 0.73810446,
      "learning_rate": 3.2955093826929547e-06,
      "loss": 0.75727797,
      "num_input_tokens_seen": 106574115,
      "step": 4947,
      "time_per_iteration": 2.542558431625366
    },
    {
      "auxiliary_loss_clip": 0.01107005,
      "auxiliary_loss_mlp": 0.0104565,
      "balance_loss_clip": 1.05087936,
      "balance_loss_mlp": 1.02879381,
      "epoch": 0.2974898542011123,
      "flos": 25666433850240.0,
      "grad_norm": 1.9620140727914286,
      "language_loss": 0.73488063,
      "learning_rate": 3.2952126471492985e-06,
      "loss": 0.75640714,
      "num_input_tokens_seen": 106593070,
      "step": 4948,
      "time_per_iteration": 2.574901819229126
    },
    {
      "auxiliary_loss_clip": 0.01134555,
      "auxiliary_loss_mlp": 0.01034652,
      "balance_loss_clip": 1.04622447,
      "balance_loss_mlp": 1.01902342,
      "epoch": 0.29754997745378026,
      "flos": 18661339658880.0,
      "grad_norm": 2.059281226990047,
      "language_loss": 0.84081614,
      "learning_rate": 3.2949158624910497e-06,
      "loss": 0.8625083,
      "num_input_tokens_seen": 106610695,
      "step": 4949,
      "time_per_iteration": 2.434049129486084
    },
    {
      "auxiliary_loss_clip": 0.01120207,
      "auxiliary_loss_mlp": 0.01040125,
      "balance_loss_clip": 1.04426491,
      "balance_loss_mlp": 1.02424049,
      "epoch": 0.2976101007064482,
      "flos": 22274599449600.0,
      "grad_norm": 1.7927366131895477,
      "language_loss": 0.71439672,
      "learning_rate": 3.2946190287294603e-06,
      "loss": 0.73600006,
      "num_input_tokens_seen": 106631300,
      "step": 4950,
      "time_per_iteration": 2.5180699825286865
    },
    {
      "auxiliary_loss_clip": 0.0108866,
      "auxiliary_loss_mlp": 0.01047115,
      "balance_loss_clip": 1.047786,
      "balance_loss_mlp": 1.03222561,
      "epoch": 0.2976702239591162,
      "flos": 21945657674880.0,
      "grad_norm": 3.600264950959555,
      "language_loss": 0.82365501,
      "learning_rate": 3.294322145875789e-06,
      "loss": 0.84501278,
      "num_input_tokens_seen": 106650065,
      "step": 4951,
      "time_per_iteration": 2.613265037536621
    },
    {
      "auxiliary_loss_clip": 0.01107971,
      "auxiliary_loss_mlp": 0.01037204,
      "balance_loss_clip": 1.04129207,
      "balance_loss_mlp": 1.02049077,
      "epoch": 0.29773034721178415,
      "flos": 24637197542400.0,
      "grad_norm": 2.4577561368441456,
      "language_loss": 0.74009323,
      "learning_rate": 3.2940252139412912e-06,
      "loss": 0.76154494,
      "num_input_tokens_seen": 106668230,
      "step": 4952,
      "time_per_iteration": 2.5388505458831787
    },
    {
      "auxiliary_loss_clip": 0.01057955,
      "auxiliary_loss_mlp": 0.01040764,
      "balance_loss_clip": 1.04443014,
      "balance_loss_mlp": 1.02270341,
      "epoch": 0.2977904704644521,
      "flos": 20557566541440.0,
      "grad_norm": 1.6479531658806863,
      "language_loss": 0.8396914,
      "learning_rate": 3.293728232937228e-06,
      "loss": 0.86067855,
      "num_input_tokens_seen": 106687785,
      "step": 4953,
      "time_per_iteration": 2.661498546600342
    },
    {
      "auxiliary_loss_clip": 0.01112075,
      "auxiliary_loss_mlp": 0.01045485,
      "balance_loss_clip": 1.0459528,
      "balance_loss_mlp": 1.0298326,
      "epoch": 0.2978505937171201,
      "flos": 18916449027840.0,
      "grad_norm": 2.0291355196395164,
      "language_loss": 0.74258542,
      "learning_rate": 3.2934312028748597e-06,
      "loss": 0.76416099,
      "num_input_tokens_seen": 106706875,
      "step": 4954,
      "time_per_iteration": 2.5045459270477295
    },
    {
      "auxiliary_loss_clip": 0.01134979,
      "auxiliary_loss_mlp": 0.01037066,
      "balance_loss_clip": 1.04831278,
      "balance_loss_mlp": 1.02199793,
      "epoch": 0.29791071696978805,
      "flos": 19317750750720.0,
      "grad_norm": 1.872131220493492,
      "language_loss": 0.7530753,
      "learning_rate": 3.293134123765452e-06,
      "loss": 0.77479577,
      "num_input_tokens_seen": 106725105,
      "step": 4955,
      "time_per_iteration": 2.4464757442474365
    },
    {
      "auxiliary_loss_clip": 0.01090437,
      "auxiliary_loss_mlp": 0.01039127,
      "balance_loss_clip": 1.04950976,
      "balance_loss_mlp": 1.02297449,
      "epoch": 0.297970840222456,
      "flos": 18806813740800.0,
      "grad_norm": 1.7485968961253566,
      "language_loss": 0.72980022,
      "learning_rate": 3.2928369956202684e-06,
      "loss": 0.75109583,
      "num_input_tokens_seen": 106744780,
      "step": 4956,
      "time_per_iteration": 2.5750458240509033
    },
    {
      "auxiliary_loss_clip": 0.01127841,
      "auxiliary_loss_mlp": 0.01043474,
      "balance_loss_clip": 1.04751933,
      "balance_loss_mlp": 1.02626014,
      "epoch": 0.298030963475124,
      "flos": 22852760762880.0,
      "grad_norm": 1.9824948840039616,
      "language_loss": 0.78920531,
      "learning_rate": 3.2925398184505754e-06,
      "loss": 0.81091845,
      "num_input_tokens_seen": 106764670,
      "step": 4957,
      "time_per_iteration": 2.50779390335083
    },
    {
      "auxiliary_loss_clip": 0.01129331,
      "auxiliary_loss_mlp": 0.01039878,
      "balance_loss_clip": 1.04948533,
      "balance_loss_mlp": 1.02233028,
      "epoch": 0.298091086727792,
      "flos": 21868485304320.0,
      "grad_norm": 1.4954502273126786,
      "language_loss": 0.70709717,
      "learning_rate": 3.2922425922676437e-06,
      "loss": 0.72878921,
      "num_input_tokens_seen": 106783695,
      "step": 4958,
      "time_per_iteration": 2.4895408153533936
    },
    {
      "auxiliary_loss_clip": 0.0110244,
      "auxiliary_loss_mlp": 0.01051433,
      "balance_loss_clip": 1.05049324,
      "balance_loss_mlp": 1.03514886,
      "epoch": 0.29815120998045996,
      "flos": 21175014355200.0,
      "grad_norm": 1.535492812525439,
      "language_loss": 0.79017234,
      "learning_rate": 3.291945317082743e-06,
      "loss": 0.81171107,
      "num_input_tokens_seen": 106803150,
      "step": 4959,
      "time_per_iteration": 2.5596766471862793
    },
    {
      "auxiliary_loss_clip": 0.01123787,
      "auxiliary_loss_mlp": 0.01053742,
      "balance_loss_clip": 1.04743814,
      "balance_loss_mlp": 1.03761268,
      "epoch": 0.29821133323312793,
      "flos": 19896271200000.0,
      "grad_norm": 1.9875378260786563,
      "language_loss": 0.79875195,
      "learning_rate": 3.291647992907147e-06,
      "loss": 0.8205272,
      "num_input_tokens_seen": 106820705,
      "step": 4960,
      "time_per_iteration": 2.4760794639587402
    },
    {
      "auxiliary_loss_clip": 0.01113633,
      "auxiliary_loss_mlp": 0.01046445,
      "balance_loss_clip": 1.05124331,
      "balance_loss_mlp": 1.0291357,
      "epoch": 0.2982714564857959,
      "flos": 12750766744320.0,
      "grad_norm": 2.2486105599802104,
      "language_loss": 0.73531622,
      "learning_rate": 3.291350619752129e-06,
      "loss": 0.756917,
      "num_input_tokens_seen": 106837335,
      "step": 4961,
      "time_per_iteration": 2.531867742538452
    },
    {
      "auxiliary_loss_clip": 0.01126822,
      "auxiliary_loss_mlp": 0.01039491,
      "balance_loss_clip": 1.0498147,
      "balance_loss_mlp": 1.02360058,
      "epoch": 0.29833157973846386,
      "flos": 22271905929600.0,
      "grad_norm": 1.8868443884366033,
      "language_loss": 0.62142742,
      "learning_rate": 3.291053197628967e-06,
      "loss": 0.64309055,
      "num_input_tokens_seen": 106856250,
      "step": 4962,
      "time_per_iteration": 2.495713472366333
    },
    {
      "auxiliary_loss_clip": 0.01126938,
      "auxiliary_loss_mlp": 0.01045824,
      "balance_loss_clip": 1.05467582,
      "balance_loss_mlp": 1.02828836,
      "epoch": 0.2983917029911318,
      "flos": 15372999319680.0,
      "grad_norm": 1.7620275376838912,
      "language_loss": 0.83093548,
      "learning_rate": 3.2907557265489375e-06,
      "loss": 0.85266304,
      "num_input_tokens_seen": 106873370,
      "step": 4963,
      "time_per_iteration": 2.493417501449585
    },
    {
      "auxiliary_loss_clip": 0.01106039,
      "auxiliary_loss_mlp": 0.01039411,
      "balance_loss_clip": 1.05155969,
      "balance_loss_mlp": 1.02332926,
      "epoch": 0.2984518262437998,
      "flos": 15377632174080.0,
      "grad_norm": 2.2934563308067455,
      "language_loss": 0.65996009,
      "learning_rate": 3.290458206523322e-06,
      "loss": 0.6814146,
      "num_input_tokens_seen": 106890330,
      "step": 4964,
      "time_per_iteration": 2.5364348888397217
    },
    {
      "auxiliary_loss_clip": 0.01124171,
      "auxiliary_loss_mlp": 0.01037257,
      "balance_loss_clip": 1.04908979,
      "balance_loss_mlp": 1.022928,
      "epoch": 0.29851194949646775,
      "flos": 18108458542080.0,
      "grad_norm": 1.7134842790634002,
      "language_loss": 0.71102113,
      "learning_rate": 3.2901606375634015e-06,
      "loss": 0.73263538,
      "num_input_tokens_seen": 106909190,
      "step": 4965,
      "time_per_iteration": 2.507201910018921
    },
    {
      "auxiliary_loss_clip": 0.01145099,
      "auxiliary_loss_mlp": 0.01049926,
      "balance_loss_clip": 1.05561697,
      "balance_loss_mlp": 1.03374922,
      "epoch": 0.2985720727491357,
      "flos": 22018233104640.0,
      "grad_norm": 1.7465980719507772,
      "language_loss": 0.6666913,
      "learning_rate": 3.289863019680461e-06,
      "loss": 0.68864149,
      "num_input_tokens_seen": 106927825,
      "step": 4966,
      "time_per_iteration": 2.4612295627593994
    },
    {
      "auxiliary_loss_clip": 0.01143995,
      "auxiliary_loss_mlp": 0.01042389,
      "balance_loss_clip": 1.05476546,
      "balance_loss_mlp": 1.02617693,
      "epoch": 0.2986321960018037,
      "flos": 13041355772160.0,
      "grad_norm": 2.496527951257142,
      "language_loss": 0.73763889,
      "learning_rate": 3.289565352885785e-06,
      "loss": 0.75950271,
      "num_input_tokens_seen": 106943155,
      "step": 4967,
      "time_per_iteration": 2.4381401538848877
    },
    {
      "auxiliary_loss_clip": 0.0110763,
      "auxiliary_loss_mlp": 0.01033817,
      "balance_loss_clip": 1.04713082,
      "balance_loss_mlp": 1.01889229,
      "epoch": 0.29869231925447165,
      "flos": 14465034305280.0,
      "grad_norm": 1.9622410450331167,
      "language_loss": 0.71063399,
      "learning_rate": 3.2892676371906614e-06,
      "loss": 0.73204845,
      "num_input_tokens_seen": 106960295,
      "step": 4968,
      "time_per_iteration": 2.5260169506073
    },
    {
      "auxiliary_loss_clip": 0.01121578,
      "auxiliary_loss_mlp": 0.01039779,
      "balance_loss_clip": 1.0515306,
      "balance_loss_mlp": 1.0234828,
      "epoch": 0.2987524425071396,
      "flos": 31650228639360.0,
      "grad_norm": 1.5880547015350621,
      "language_loss": 0.76861179,
      "learning_rate": 3.2889698726063805e-06,
      "loss": 0.79022533,
      "num_input_tokens_seen": 106982870,
      "step": 4969,
      "time_per_iteration": 2.583770275115967
    },
    {
      "auxiliary_loss_clip": 0.01138014,
      "auxiliary_loss_mlp": 0.01034243,
      "balance_loss_clip": 1.05114555,
      "balance_loss_mlp": 1.01953244,
      "epoch": 0.2988125657598076,
      "flos": 21433427775360.0,
      "grad_norm": 1.803237038759876,
      "language_loss": 0.69812667,
      "learning_rate": 3.2886720591442327e-06,
      "loss": 0.71984923,
      "num_input_tokens_seen": 107002405,
      "step": 4970,
      "time_per_iteration": 2.448671340942383
    },
    {
      "auxiliary_loss_clip": 0.01130592,
      "auxiliary_loss_mlp": 0.01042155,
      "balance_loss_clip": 1.05102003,
      "balance_loss_mlp": 1.024786,
      "epoch": 0.2988726890124756,
      "flos": 18076965292800.0,
      "grad_norm": 1.9330718555226658,
      "language_loss": 0.85060537,
      "learning_rate": 3.2883741968155103e-06,
      "loss": 0.87233281,
      "num_input_tokens_seen": 107017310,
      "step": 4971,
      "time_per_iteration": 2.475008726119995
    },
    {
      "auxiliary_loss_clip": 0.01108282,
      "auxiliary_loss_mlp": 0.01050545,
      "balance_loss_clip": 1.04856646,
      "balance_loss_mlp": 1.032938,
      "epoch": 0.29893281226514357,
      "flos": 21755653706880.0,
      "grad_norm": 2.326055663764313,
      "language_loss": 0.7943964,
      "learning_rate": 3.2880762856315107e-06,
      "loss": 0.81598467,
      "num_input_tokens_seen": 107034645,
      "step": 4972,
      "time_per_iteration": 2.536632776260376
    },
    {
      "auxiliary_loss_clip": 0.01139338,
      "auxiliary_loss_mlp": 0.01045256,
      "balance_loss_clip": 1.05158985,
      "balance_loss_mlp": 1.02937722,
      "epoch": 0.29899293551781153,
      "flos": 16836718538880.0,
      "grad_norm": 1.788750696460857,
      "language_loss": 0.85357928,
      "learning_rate": 3.2877783256035285e-06,
      "loss": 0.87542522,
      "num_input_tokens_seen": 107051125,
      "step": 4973,
      "time_per_iteration": 2.452819347381592
    },
    {
      "auxiliary_loss_clip": 0.01106943,
      "auxiliary_loss_mlp": 0.01035399,
      "balance_loss_clip": 1.0482924,
      "balance_loss_mlp": 1.01921034,
      "epoch": 0.2990530587704795,
      "flos": 11729215946880.0,
      "grad_norm": 1.9519118679618412,
      "language_loss": 0.77768046,
      "learning_rate": 3.287480316742863e-06,
      "loss": 0.79910386,
      "num_input_tokens_seen": 107068815,
      "step": 4974,
      "time_per_iteration": 2.5142340660095215
    },
    {
      "auxiliary_loss_clip": 0.01111197,
      "auxiliary_loss_mlp": 0.00804885,
      "balance_loss_clip": 1.0485791,
      "balance_loss_mlp": 1.02873158,
      "epoch": 0.29911318202314746,
      "flos": 28039877850240.0,
      "grad_norm": 1.6282605991294536,
      "language_loss": 0.72612631,
      "learning_rate": 3.287182259060815e-06,
      "loss": 0.74528718,
      "num_input_tokens_seen": 107090420,
      "step": 4975,
      "time_per_iteration": 4.041979074478149
    },
    {
      "auxiliary_loss_clip": 0.01130138,
      "auxiliary_loss_mlp": 0.01039758,
      "balance_loss_clip": 1.05351257,
      "balance_loss_mlp": 1.02342653,
      "epoch": 0.2991733052758154,
      "flos": 18733555952640.0,
      "grad_norm": 2.0560499047342864,
      "language_loss": 0.76485431,
      "learning_rate": 3.286884152568687e-06,
      "loss": 0.78655332,
      "num_input_tokens_seen": 107107255,
      "step": 4976,
      "time_per_iteration": 2.4753096103668213
    },
    {
      "auxiliary_loss_clip": 0.0112518,
      "auxiliary_loss_mlp": 0.01039798,
      "balance_loss_clip": 1.050264,
      "balance_loss_mlp": 1.02437198,
      "epoch": 0.2992334285284834,
      "flos": 15559160532480.0,
      "grad_norm": 1.9670025423908344,
      "language_loss": 0.86251616,
      "learning_rate": 3.2865859972777827e-06,
      "loss": 0.884166,
      "num_input_tokens_seen": 107123840,
      "step": 4977,
      "time_per_iteration": 2.464825391769409
    },
    {
      "auxiliary_loss_clip": 0.01116371,
      "auxiliary_loss_mlp": 0.010414,
      "balance_loss_clip": 1.05296421,
      "balance_loss_mlp": 1.02494907,
      "epoch": 0.29929355178115136,
      "flos": 21797561900160.0,
      "grad_norm": 1.5526784782249545,
      "language_loss": 0.68555915,
      "learning_rate": 3.2862877931994088e-06,
      "loss": 0.70713687,
      "num_input_tokens_seen": 107143475,
      "step": 4978,
      "time_per_iteration": 2.5562891960144043
    },
    {
      "auxiliary_loss_clip": 0.01120986,
      "auxiliary_loss_mlp": 0.01036255,
      "balance_loss_clip": 1.05344117,
      "balance_loss_mlp": 1.02008998,
      "epoch": 0.2993536750338193,
      "flos": 21178533888000.0,
      "grad_norm": 2.443722641167502,
      "language_loss": 0.76693004,
      "learning_rate": 3.2859895403448726e-06,
      "loss": 0.78850245,
      "num_input_tokens_seen": 107161725,
      "step": 4979,
      "time_per_iteration": 3.898388385772705
    },
    {
      "auxiliary_loss_clip": 0.01090218,
      "auxiliary_loss_mlp": 0.010413,
      "balance_loss_clip": 1.04398394,
      "balance_loss_mlp": 1.02354956,
      "epoch": 0.2994137982864873,
      "flos": 32122130544000.0,
      "grad_norm": 1.5650390116049888,
      "language_loss": 0.68047053,
      "learning_rate": 3.285691238725484e-06,
      "loss": 0.70178568,
      "num_input_tokens_seen": 107183935,
      "step": 4980,
      "time_per_iteration": 2.6674156188964844
    },
    {
      "auxiliary_loss_clip": 0.01125998,
      "auxiliary_loss_mlp": 0.00793846,
      "balance_loss_clip": 1.05301797,
      "balance_loss_mlp": 1.01659429,
      "epoch": 0.29947392153915525,
      "flos": 21105419754240.0,
      "grad_norm": 1.7367779102422212,
      "language_loss": 0.73491484,
      "learning_rate": 3.285392888352555e-06,
      "loss": 0.75411332,
      "num_input_tokens_seen": 107204285,
      "step": 4981,
      "time_per_iteration": 2.565366268157959
    },
    {
      "auxiliary_loss_clip": 0.01128904,
      "auxiliary_loss_mlp": 0.01040841,
      "balance_loss_clip": 1.04977846,
      "balance_loss_mlp": 1.02514136,
      "epoch": 0.2995340447918232,
      "flos": 21542632099200.0,
      "grad_norm": 1.5152469998882911,
      "language_loss": 0.86179805,
      "learning_rate": 3.2850944892373987e-06,
      "loss": 0.88349557,
      "num_input_tokens_seen": 107225265,
      "step": 4982,
      "time_per_iteration": 3.9573800563812256
    },
    {
      "auxiliary_loss_clip": 0.01124265,
      "auxiliary_loss_mlp": 0.01039309,
      "balance_loss_clip": 1.05501938,
      "balance_loss_mlp": 1.02183306,
      "epoch": 0.2995941680444912,
      "flos": 16725143917440.0,
      "grad_norm": 2.178267434296674,
      "language_loss": 0.86945653,
      "learning_rate": 3.2847960413913307e-06,
      "loss": 0.8910923,
      "num_input_tokens_seen": 107241335,
      "step": 4983,
      "time_per_iteration": 3.9267542362213135
    },
    {
      "auxiliary_loss_clip": 0.01124063,
      "auxiliary_loss_mlp": 0.01044364,
      "balance_loss_clip": 1.05308414,
      "balance_loss_mlp": 1.02948678,
      "epoch": 0.2996542912971592,
      "flos": 20923496346240.0,
      "grad_norm": 2.00736407207815,
      "language_loss": 0.78666437,
      "learning_rate": 3.284497544825668e-06,
      "loss": 0.8083486,
      "num_input_tokens_seen": 107259375,
      "step": 4984,
      "time_per_iteration": 2.511950969696045
    },
    {
      "auxiliary_loss_clip": 0.01108388,
      "auxiliary_loss_mlp": 0.01045447,
      "balance_loss_clip": 1.05400062,
      "balance_loss_mlp": 1.0286746,
      "epoch": 0.29971441454982717,
      "flos": 25079868754560.0,
      "grad_norm": 1.5653538001627576,
      "language_loss": 0.78252494,
      "learning_rate": 3.2841989995517303e-06,
      "loss": 0.80406332,
      "num_input_tokens_seen": 107279890,
      "step": 4985,
      "time_per_iteration": 2.588338851928711
    },
    {
      "auxiliary_loss_clip": 0.01083459,
      "auxiliary_loss_mlp": 0.01055652,
      "balance_loss_clip": 1.04852283,
      "balance_loss_mlp": 1.03530347,
      "epoch": 0.29977453780249513,
      "flos": 52555911840000.0,
      "grad_norm": 2.261501183930516,
      "language_loss": 0.71658659,
      "learning_rate": 3.283900405580837e-06,
      "loss": 0.73797762,
      "num_input_tokens_seen": 107303430,
      "step": 4986,
      "time_per_iteration": 2.897321939468384
    },
    {
      "auxiliary_loss_clip": 0.01117725,
      "auxiliary_loss_mlp": 0.01043639,
      "balance_loss_clip": 1.0510875,
      "balance_loss_mlp": 1.02681816,
      "epoch": 0.2998346610551631,
      "flos": 22237144542720.0,
      "grad_norm": 1.6516946314641412,
      "language_loss": 0.73302639,
      "learning_rate": 3.283601762924312e-06,
      "loss": 0.7546401,
      "num_input_tokens_seen": 107323700,
      "step": 4987,
      "time_per_iteration": 2.531484365463257
    },
    {
      "auxiliary_loss_clip": 0.01108881,
      "auxiliary_loss_mlp": 0.01039982,
      "balance_loss_clip": 1.05021739,
      "balance_loss_mlp": 1.02366257,
      "epoch": 0.29989478430783106,
      "flos": 16873203778560.0,
      "grad_norm": 1.6716986159150196,
      "language_loss": 0.79939735,
      "learning_rate": 3.2833030715934793e-06,
      "loss": 0.8208859,
      "num_input_tokens_seen": 107341965,
      "step": 4988,
      "time_per_iteration": 2.53928804397583
    },
    {
      "auxiliary_loss_clip": 0.01115922,
      "auxiliary_loss_mlp": 0.00795748,
      "balance_loss_clip": 1.0486393,
      "balance_loss_mlp": 1.01360512,
      "epoch": 0.29995490756049903,
      "flos": 23768878164480.0,
      "grad_norm": 1.6730632093885551,
      "language_loss": 0.70436561,
      "learning_rate": 3.2830043315996658e-06,
      "loss": 0.72348231,
      "num_input_tokens_seen": 107362615,
      "step": 4989,
      "time_per_iteration": 2.56370210647583
    },
    {
      "auxiliary_loss_clip": 0.01104126,
      "auxiliary_loss_mlp": 0.01045497,
      "balance_loss_clip": 1.05030847,
      "balance_loss_mlp": 1.02840257,
      "epoch": 0.300015030813167,
      "flos": 14465321614080.0,
      "grad_norm": 2.0264694424956295,
      "language_loss": 0.85662687,
      "learning_rate": 3.282705542954199e-06,
      "loss": 0.8781231,
      "num_input_tokens_seen": 107378980,
      "step": 4990,
      "time_per_iteration": 2.5663490295410156
    },
    {
      "auxiliary_loss_clip": 0.01128654,
      "auxiliary_loss_mlp": 0.01040321,
      "balance_loss_clip": 1.05016279,
      "balance_loss_mlp": 1.02319098,
      "epoch": 0.30007515406583496,
      "flos": 25191982080000.0,
      "grad_norm": 1.6170242131856498,
      "language_loss": 0.66896319,
      "learning_rate": 3.28240670566841e-06,
      "loss": 0.69065297,
      "num_input_tokens_seen": 107397640,
      "step": 4991,
      "time_per_iteration": 2.5176143646240234
    },
    {
      "auxiliary_loss_clip": 0.01114466,
      "auxiliary_loss_mlp": 0.01037011,
      "balance_loss_clip": 1.04525542,
      "balance_loss_mlp": 1.01949954,
      "epoch": 0.3001352773185029,
      "flos": 19391188106880.0,
      "grad_norm": 1.8438565291232796,
      "language_loss": 0.78738689,
      "learning_rate": 3.28210781975363e-06,
      "loss": 0.80890167,
      "num_input_tokens_seen": 107416020,
      "step": 4992,
      "time_per_iteration": 2.520667314529419
    },
    {
      "auxiliary_loss_clip": 0.01138573,
      "auxiliary_loss_mlp": 0.01037363,
      "balance_loss_clip": 1.05084538,
      "balance_loss_mlp": 1.02092373,
      "epoch": 0.3001954005711709,
      "flos": 21543853161600.0,
      "grad_norm": 2.417657915636467,
      "language_loss": 0.8233614,
      "learning_rate": 3.281808885221193e-06,
      "loss": 0.84512079,
      "num_input_tokens_seen": 107436340,
      "step": 4993,
      "time_per_iteration": 2.4865798950195312
    },
    {
      "auxiliary_loss_clip": 0.01093742,
      "auxiliary_loss_mlp": 0.01051245,
      "balance_loss_clip": 1.04945588,
      "balance_loss_mlp": 1.03339922,
      "epoch": 0.30025552382383885,
      "flos": 17384320356480.0,
      "grad_norm": 2.2166145976427645,
      "language_loss": 0.86553121,
      "learning_rate": 3.2815099020824345e-06,
      "loss": 0.88698113,
      "num_input_tokens_seen": 107454585,
      "step": 4994,
      "time_per_iteration": 2.581536054611206
    },
    {
      "auxiliary_loss_clip": 0.01114301,
      "auxiliary_loss_mlp": 0.01036063,
      "balance_loss_clip": 1.05031097,
      "balance_loss_mlp": 1.02043438,
      "epoch": 0.3003156470765068,
      "flos": 29533330552320.0,
      "grad_norm": 1.6246550986235235,
      "language_loss": 0.80989408,
      "learning_rate": 3.2812108703486924e-06,
      "loss": 0.83139765,
      "num_input_tokens_seen": 107477180,
      "step": 4995,
      "time_per_iteration": 2.6027274131774902
    },
    {
      "auxiliary_loss_clip": 0.01119483,
      "auxiliary_loss_mlp": 0.01038831,
      "balance_loss_clip": 1.04974723,
      "balance_loss_mlp": 1.02235603,
      "epoch": 0.3003757703291748,
      "flos": 43646402465280.0,
      "grad_norm": 1.6735142631515005,
      "language_loss": 0.67354631,
      "learning_rate": 3.2809117900313055e-06,
      "loss": 0.69512945,
      "num_input_tokens_seen": 107500250,
      "step": 4996,
      "time_per_iteration": 2.6975860595703125
    },
    {
      "auxiliary_loss_clip": 0.0111322,
      "auxiliary_loss_mlp": 0.01043185,
      "balance_loss_clip": 1.05123568,
      "balance_loss_mlp": 1.02649558,
      "epoch": 0.30043589358184275,
      "flos": 22528380015360.0,
      "grad_norm": 1.6793152613844229,
      "language_loss": 0.75373858,
      "learning_rate": 3.280612661141615e-06,
      "loss": 0.77530265,
      "num_input_tokens_seen": 107520070,
      "step": 4997,
      "time_per_iteration": 2.5222723484039307
    },
    {
      "auxiliary_loss_clip": 0.01127707,
      "auxiliary_loss_mlp": 0.01046694,
      "balance_loss_clip": 1.0495832,
      "balance_loss_mlp": 1.03103006,
      "epoch": 0.30049601683451077,
      "flos": 20995892208000.0,
      "grad_norm": 2.5231291562388822,
      "language_loss": 0.7796188,
      "learning_rate": 3.2803134836909646e-06,
      "loss": 0.80136281,
      "num_input_tokens_seen": 107539285,
      "step": 4998,
      "time_per_iteration": 2.4837443828582764
    },
    {
      "auxiliary_loss_clip": 0.01135992,
      "auxiliary_loss_mlp": 0.01042176,
      "balance_loss_clip": 1.05157721,
      "balance_loss_mlp": 1.02689362,
      "epoch": 0.30055614008717874,
      "flos": 23916004272000.0,
      "grad_norm": 1.7507611900988742,
      "language_loss": 0.73029566,
      "learning_rate": 3.2800142576906985e-06,
      "loss": 0.75207734,
      "num_input_tokens_seen": 107560260,
      "step": 4999,
      "time_per_iteration": 2.482285737991333
    },
    {
      "auxiliary_loss_clip": 0.01127361,
      "auxiliary_loss_mlp": 0.0104265,
      "balance_loss_clip": 1.04840231,
      "balance_loss_mlp": 1.02654457,
      "epoch": 0.3006162633398467,
      "flos": 19169798630400.0,
      "grad_norm": 1.7252286233426317,
      "language_loss": 0.75548553,
      "learning_rate": 3.2797149831521626e-06,
      "loss": 0.77718562,
      "num_input_tokens_seen": 107579260,
      "step": 5000,
      "time_per_iteration": 2.472318649291992
    },
    {
      "auxiliary_loss_clip": 0.01134743,
      "auxiliary_loss_mlp": 0.01041564,
      "balance_loss_clip": 1.04956388,
      "balance_loss_mlp": 1.0274024,
      "epoch": 0.30067638659251467,
      "flos": 14679241061760.0,
      "grad_norm": 1.8025304270448066,
      "language_loss": 0.82174647,
      "learning_rate": 3.2794156600867073e-06,
      "loss": 0.84350955,
      "num_input_tokens_seen": 107595245,
      "step": 5001,
      "time_per_iteration": 2.4110982418060303
    },
    {
      "auxiliary_loss_clip": 0.01126196,
      "auxiliary_loss_mlp": 0.01046005,
      "balance_loss_clip": 1.05058146,
      "balance_loss_mlp": 1.02982819,
      "epoch": 0.30073650984518263,
      "flos": 23368007404800.0,
      "grad_norm": 1.6976251400224927,
      "language_loss": 0.80829048,
      "learning_rate": 3.2791162885056815e-06,
      "loss": 0.83001256,
      "num_input_tokens_seen": 107613985,
      "step": 5002,
      "time_per_iteration": 2.4938101768493652
    },
    {
      "auxiliary_loss_clip": 0.01089451,
      "auxiliary_loss_mlp": 0.01041605,
      "balance_loss_clip": 1.05266523,
      "balance_loss_mlp": 1.02323461,
      "epoch": 0.3007966330978506,
      "flos": 22966633854720.0,
      "grad_norm": 1.8980960854875124,
      "language_loss": 0.70817143,
      "learning_rate": 3.2788168684204376e-06,
      "loss": 0.729482,
      "num_input_tokens_seen": 107631435,
      "step": 5003,
      "time_per_iteration": 2.572446584701538
    },
    {
      "auxiliary_loss_clip": 0.01104864,
      "auxiliary_loss_mlp": 0.01044812,
      "balance_loss_clip": 1.04621911,
      "balance_loss_mlp": 1.02867079,
      "epoch": 0.30085675635051856,
      "flos": 27818452460160.0,
      "grad_norm": 2.3351197717296683,
      "language_loss": 0.70021564,
      "learning_rate": 3.27851739984233e-06,
      "loss": 0.72171235,
      "num_input_tokens_seen": 107650530,
      "step": 5004,
      "time_per_iteration": 2.597621440887451
    },
    {
      "auxiliary_loss_clip": 0.01116068,
      "auxiliary_loss_mlp": 0.01042147,
      "balance_loss_clip": 1.04942274,
      "balance_loss_mlp": 1.0262326,
      "epoch": 0.3009168796031865,
      "flos": 10882729059840.0,
      "grad_norm": 2.671529055966598,
      "language_loss": 0.82086205,
      "learning_rate": 3.278217882782715e-06,
      "loss": 0.84244418,
      "num_input_tokens_seen": 107662240,
      "step": 5005,
      "time_per_iteration": 2.4591000080108643
    },
    {
      "auxiliary_loss_clip": 0.0112328,
      "auxiliary_loss_mlp": 0.0103599,
      "balance_loss_clip": 1.04755902,
      "balance_loss_mlp": 1.02119589,
      "epoch": 0.3009770028558545,
      "flos": 23805399317760.0,
      "grad_norm": 2.5033797960099102,
      "language_loss": 0.7476567,
      "learning_rate": 3.2779183172529497e-06,
      "loss": 0.76924944,
      "num_input_tokens_seen": 107680330,
      "step": 5006,
      "time_per_iteration": 2.5223889350891113
    },
    {
      "auxiliary_loss_clip": 0.01095068,
      "auxiliary_loss_mlp": 0.00791997,
      "balance_loss_clip": 1.04681659,
      "balance_loss_mlp": 1.00850153,
      "epoch": 0.30103712610852246,
      "flos": 26468211283200.0,
      "grad_norm": 1.8442660899411236,
      "language_loss": 0.70984352,
      "learning_rate": 3.2776187032643932e-06,
      "loss": 0.72871411,
      "num_input_tokens_seen": 107700020,
      "step": 5007,
      "time_per_iteration": 2.5825209617614746
    },
    {
      "auxiliary_loss_clip": 0.01123255,
      "auxiliary_loss_mlp": 0.01039942,
      "balance_loss_clip": 1.0475539,
      "balance_loss_mlp": 1.02313316,
      "epoch": 0.3010972493611904,
      "flos": 22856459863680.0,
      "grad_norm": 2.435830217006068,
      "language_loss": 0.7587499,
      "learning_rate": 3.2773190408284075e-06,
      "loss": 0.7803818,
      "num_input_tokens_seen": 107718575,
      "step": 5008,
      "time_per_iteration": 2.5135397911071777
    },
    {
      "auxiliary_loss_clip": 0.01122428,
      "auxiliary_loss_mlp": 0.01037852,
      "balance_loss_clip": 1.05083084,
      "balance_loss_mlp": 1.02185404,
      "epoch": 0.3011573726138584,
      "flos": 24053685102720.0,
      "grad_norm": 1.8134224701260682,
      "language_loss": 0.84954768,
      "learning_rate": 3.2770193299563564e-06,
      "loss": 0.87115049,
      "num_input_tokens_seen": 107738635,
      "step": 5009,
      "time_per_iteration": 2.5063834190368652
    },
    {
      "auxiliary_loss_clip": 0.01127798,
      "auxiliary_loss_mlp": 0.01039477,
      "balance_loss_clip": 1.04699087,
      "balance_loss_mlp": 1.0223701,
      "epoch": 0.30121749586652635,
      "flos": 20259687052800.0,
      "grad_norm": 1.9780798275027232,
      "language_loss": 0.83691001,
      "learning_rate": 3.276719570659604e-06,
      "loss": 0.85858279,
      "num_input_tokens_seen": 107753415,
      "step": 5010,
      "time_per_iteration": 2.49542236328125
    },
    {
      "auxiliary_loss_clip": 0.01101936,
      "auxiliary_loss_mlp": 0.01034523,
      "balance_loss_clip": 1.0454936,
      "balance_loss_mlp": 1.01976514,
      "epoch": 0.3012776191191944,
      "flos": 26943058103040.0,
      "grad_norm": 3.9079167958940237,
      "language_loss": 0.84785831,
      "learning_rate": 3.2764197629495176e-06,
      "loss": 0.86922294,
      "num_input_tokens_seen": 107773840,
      "step": 5011,
      "time_per_iteration": 2.5702877044677734
    },
    {
      "auxiliary_loss_clip": 0.01114302,
      "auxiliary_loss_mlp": 0.01038658,
      "balance_loss_clip": 1.04278541,
      "balance_loss_mlp": 1.02237439,
      "epoch": 0.30133774237186234,
      "flos": 20412307941120.0,
      "grad_norm": 1.895589823278429,
      "language_loss": 0.71950364,
      "learning_rate": 3.2761199068374656e-06,
      "loss": 0.74103326,
      "num_input_tokens_seen": 107792020,
      "step": 5012,
      "time_per_iteration": 2.524195432662964
    },
    {
      "auxiliary_loss_clip": 0.01124124,
      "auxiliary_loss_mlp": 0.01039351,
      "balance_loss_clip": 1.04910159,
      "balance_loss_mlp": 1.02430725,
      "epoch": 0.3013978656245303,
      "flos": 19792453916160.0,
      "grad_norm": 2.1164283540766493,
      "language_loss": 0.87311447,
      "learning_rate": 3.275820002334819e-06,
      "loss": 0.89474922,
      "num_input_tokens_seen": 107809595,
      "step": 5013,
      "time_per_iteration": 2.4812512397766113
    },
    {
      "auxiliary_loss_clip": 0.01107591,
      "auxiliary_loss_mlp": 0.01047325,
      "balance_loss_clip": 1.04383719,
      "balance_loss_mlp": 1.02770281,
      "epoch": 0.30145798887719827,
      "flos": 16249650652800.0,
      "grad_norm": 2.8376924339131664,
      "language_loss": 0.82869661,
      "learning_rate": 3.2755200494529496e-06,
      "loss": 0.85024571,
      "num_input_tokens_seen": 107827230,
      "step": 5014,
      "time_per_iteration": 3.91120982170105
    },
    {
      "auxiliary_loss_clip": 0.01090462,
      "auxiliary_loss_mlp": 0.01042259,
      "balance_loss_clip": 1.045784,
      "balance_loss_mlp": 1.02535486,
      "epoch": 0.30151811212986623,
      "flos": 24571733005440.0,
      "grad_norm": 1.5379268162300745,
      "language_loss": 0.6837967,
      "learning_rate": 3.2752200482032323e-06,
      "loss": 0.7051239,
      "num_input_tokens_seen": 107847195,
      "step": 5015,
      "time_per_iteration": 2.5861940383911133
    },
    {
      "auxiliary_loss_clip": 0.01108581,
      "auxiliary_loss_mlp": 0.0103589,
      "balance_loss_clip": 1.04680753,
      "balance_loss_mlp": 1.01967168,
      "epoch": 0.3015782353825342,
      "flos": 21872076664320.0,
      "grad_norm": 3.4891791683063134,
      "language_loss": 0.75034726,
      "learning_rate": 3.2749199985970436e-06,
      "loss": 0.77179193,
      "num_input_tokens_seen": 107866420,
      "step": 5016,
      "time_per_iteration": 2.540588855743408
    },
    {
      "auxiliary_loss_clip": 0.01126164,
      "auxiliary_loss_mlp": 0.0103761,
      "balance_loss_clip": 1.04735529,
      "balance_loss_mlp": 1.02213645,
      "epoch": 0.30163835863520216,
      "flos": 28769331248640.0,
      "grad_norm": 2.106984321978422,
      "language_loss": 0.65675789,
      "learning_rate": 3.2746199006457603e-06,
      "loss": 0.67839563,
      "num_input_tokens_seen": 107889090,
      "step": 5017,
      "time_per_iteration": 2.5641114711761475
    },
    {
      "auxiliary_loss_clip": 0.01097453,
      "auxiliary_loss_mlp": 0.01051742,
      "balance_loss_clip": 1.04973161,
      "balance_loss_mlp": 1.03409863,
      "epoch": 0.30169848188787013,
      "flos": 22966202891520.0,
      "grad_norm": 2.0157483127875184,
      "language_loss": 0.68887591,
      "learning_rate": 3.2743197543607628e-06,
      "loss": 0.7103678,
      "num_input_tokens_seen": 107907520,
      "step": 5018,
      "time_per_iteration": 3.923494577407837
    },
    {
      "auxiliary_loss_clip": 0.01129174,
      "auxiliary_loss_mlp": 0.01044044,
      "balance_loss_clip": 1.04529071,
      "balance_loss_mlp": 1.03010893,
      "epoch": 0.3017586051405381,
      "flos": 21835268202240.0,
      "grad_norm": 2.1578167382580227,
      "language_loss": 0.79096758,
      "learning_rate": 3.2740195597534327e-06,
      "loss": 0.81269979,
      "num_input_tokens_seen": 107925650,
      "step": 5019,
      "time_per_iteration": 2.474189281463623
    },
    {
      "auxiliary_loss_clip": 0.01107469,
      "auxiliary_loss_mlp": 0.01041087,
      "balance_loss_clip": 1.04696059,
      "balance_loss_mlp": 1.02582777,
      "epoch": 0.30181872839320606,
      "flos": 22160403135360.0,
      "grad_norm": 1.995824270528882,
      "language_loss": 0.69660342,
      "learning_rate": 3.2737193168351527e-06,
      "loss": 0.71808898,
      "num_input_tokens_seen": 107943975,
      "step": 5020,
      "time_per_iteration": 2.529404640197754
    },
    {
      "auxiliary_loss_clip": 0.01137148,
      "auxiliary_loss_mlp": 0.01044588,
      "balance_loss_clip": 1.04687035,
      "balance_loss_mlp": 1.02954364,
      "epoch": 0.301878851645874,
      "flos": 18114168804480.0,
      "grad_norm": 2.553762834764847,
      "language_loss": 0.78794825,
      "learning_rate": 3.2734190256173085e-06,
      "loss": 0.80976564,
      "num_input_tokens_seen": 107962950,
      "step": 5021,
      "time_per_iteration": 3.829279899597168
    },
    {
      "auxiliary_loss_clip": 0.01122893,
      "auxiliary_loss_mlp": 0.01033362,
      "balance_loss_clip": 1.04386508,
      "balance_loss_mlp": 1.01852083,
      "epoch": 0.301938974898542,
      "flos": 17602226213760.0,
      "grad_norm": 3.2362313691780247,
      "language_loss": 0.76031899,
      "learning_rate": 3.2731186861112877e-06,
      "loss": 0.78188157,
      "num_input_tokens_seen": 107979700,
      "step": 5022,
      "time_per_iteration": 3.8037078380584717
    },
    {
      "auxiliary_loss_clip": 0.01135612,
      "auxiliary_loss_mlp": 0.01041764,
      "balance_loss_clip": 1.0463872,
      "balance_loss_mlp": 1.02574229,
      "epoch": 0.30199909815120995,
      "flos": 11181219079680.0,
      "grad_norm": 2.4269824877013972,
      "language_loss": 0.69675899,
      "learning_rate": 3.2728182983284793e-06,
      "loss": 0.71853268,
      "num_input_tokens_seen": 107996645,
      "step": 5023,
      "time_per_iteration": 2.4245004653930664
    },
    {
      "auxiliary_loss_clip": 0.0111028,
      "auxiliary_loss_mlp": 0.0103742,
      "balance_loss_clip": 1.04728556,
      "balance_loss_mlp": 1.0223875,
      "epoch": 0.302059221403878,
      "flos": 21907843632000.0,
      "grad_norm": 1.9877278401528358,
      "language_loss": 0.71320581,
      "learning_rate": 3.2725178622802724e-06,
      "loss": 0.7346828,
      "num_input_tokens_seen": 108015020,
      "step": 5024,
      "time_per_iteration": 2.506767511367798
    },
    {
      "auxiliary_loss_clip": 0.01120946,
      "auxiliary_loss_mlp": 0.01041287,
      "balance_loss_clip": 1.04641795,
      "balance_loss_mlp": 1.02581358,
      "epoch": 0.30211934465654594,
      "flos": 26396390039040.0,
      "grad_norm": 1.7897725998702476,
      "language_loss": 0.74343699,
      "learning_rate": 3.272217377978061e-06,
      "loss": 0.76505935,
      "num_input_tokens_seen": 108036430,
      "step": 5025,
      "time_per_iteration": 2.5245823860168457
    },
    {
      "auxiliary_loss_clip": 0.01122714,
      "auxiliary_loss_mlp": 0.01040391,
      "balance_loss_clip": 1.04840565,
      "balance_loss_mlp": 1.02568614,
      "epoch": 0.3021794679092139,
      "flos": 23400470321280.0,
      "grad_norm": 1.9753713599213931,
      "language_loss": 0.67020118,
      "learning_rate": 3.2719168454332387e-06,
      "loss": 0.69183218,
      "num_input_tokens_seen": 108054250,
      "step": 5026,
      "time_per_iteration": 2.499821901321411
    },
    {
      "auxiliary_loss_clip": 0.01124726,
      "auxiliary_loss_mlp": 0.0104079,
      "balance_loss_clip": 1.04773068,
      "balance_loss_mlp": 1.02560258,
      "epoch": 0.30223959116188187,
      "flos": 20260979942400.0,
      "grad_norm": 1.7883321615784689,
      "language_loss": 0.85508227,
      "learning_rate": 3.2716162646572034e-06,
      "loss": 0.87673748,
      "num_input_tokens_seen": 108071495,
      "step": 5027,
      "time_per_iteration": 2.488584280014038
    },
    {
      "auxiliary_loss_clip": 0.01102362,
      "auxiliary_loss_mlp": 0.01040236,
      "balance_loss_clip": 1.05059242,
      "balance_loss_mlp": 1.02606213,
      "epoch": 0.30229971441454984,
      "flos": 26687840993280.0,
      "grad_norm": 1.655398840970718,
      "language_loss": 0.79208744,
      "learning_rate": 3.271315635661351e-06,
      "loss": 0.8135134,
      "num_input_tokens_seen": 108092135,
      "step": 5028,
      "time_per_iteration": 2.5646305084228516
    },
    {
      "auxiliary_loss_clip": 0.01108867,
      "auxiliary_loss_mlp": 0.01044683,
      "balance_loss_clip": 1.04677677,
      "balance_loss_mlp": 1.02911472,
      "epoch": 0.3023598376672178,
      "flos": 34345323953280.0,
      "grad_norm": 1.74509112611865,
      "language_loss": 0.77176285,
      "learning_rate": 3.2710149584570826e-06,
      "loss": 0.79329836,
      "num_input_tokens_seen": 108112945,
      "step": 5029,
      "time_per_iteration": 2.638576030731201
    },
    {
      "auxiliary_loss_clip": 0.01103725,
      "auxiliary_loss_mlp": 0.01043424,
      "balance_loss_clip": 1.04503465,
      "balance_loss_mlp": 1.02637756,
      "epoch": 0.30241996091988577,
      "flos": 23112143850240.0,
      "grad_norm": 1.8270087008066194,
      "language_loss": 0.8208282,
      "learning_rate": 3.2707142330557993e-06,
      "loss": 0.8422997,
      "num_input_tokens_seen": 108130325,
      "step": 5030,
      "time_per_iteration": 2.5418989658355713
    },
    {
      "auxiliary_loss_clip": 0.01087978,
      "auxiliary_loss_mlp": 0.00792286,
      "balance_loss_clip": 1.04779077,
      "balance_loss_mlp": 1.01254249,
      "epoch": 0.30248008417255373,
      "flos": 19390002958080.0,
      "grad_norm": 1.5971244985502155,
      "language_loss": 0.69760072,
      "learning_rate": 3.270413459468905e-06,
      "loss": 0.71640337,
      "num_input_tokens_seen": 108150300,
      "step": 5031,
      "time_per_iteration": 2.6196177005767822
    },
    {
      "auxiliary_loss_clip": 0.01120808,
      "auxiliary_loss_mlp": 0.01037504,
      "balance_loss_clip": 1.05005407,
      "balance_loss_mlp": 1.02197075,
      "epoch": 0.3025402074252217,
      "flos": 23769704177280.0,
      "grad_norm": 1.9695219055555078,
      "language_loss": 0.82472879,
      "learning_rate": 3.2701126377078047e-06,
      "loss": 0.84631193,
      "num_input_tokens_seen": 108170330,
      "step": 5032,
      "time_per_iteration": 2.505718946456909
    },
    {
      "auxiliary_loss_clip": 0.01101418,
      "auxiliary_loss_mlp": 0.01051626,
      "balance_loss_clip": 1.05340934,
      "balance_loss_mlp": 1.03347087,
      "epoch": 0.30260033067788966,
      "flos": 25994118648960.0,
      "grad_norm": 2.8175084237099606,
      "language_loss": 0.73096377,
      "learning_rate": 3.269811767783906e-06,
      "loss": 0.75249422,
      "num_input_tokens_seen": 108191265,
      "step": 5033,
      "time_per_iteration": 2.608933210372925
    },
    {
      "auxiliary_loss_clip": 0.01119856,
      "auxiliary_loss_mlp": 0.01048618,
      "balance_loss_clip": 1.04748499,
      "balance_loss_mlp": 1.0324651,
      "epoch": 0.3026604539305576,
      "flos": 25374551932800.0,
      "grad_norm": 1.469774908239953,
      "language_loss": 0.73938739,
      "learning_rate": 3.2695108497086185e-06,
      "loss": 0.76107216,
      "num_input_tokens_seen": 108211615,
      "step": 5034,
      "time_per_iteration": 2.516589879989624
    },
    {
      "auxiliary_loss_clip": 0.01134623,
      "auxiliary_loss_mlp": 0.01035083,
      "balance_loss_clip": 1.04735935,
      "balance_loss_mlp": 1.01980042,
      "epoch": 0.3027205771832256,
      "flos": 25812733944960.0,
      "grad_norm": 1.6855038005244924,
      "language_loss": 0.71866202,
      "learning_rate": 3.269209883493352e-06,
      "loss": 0.74035919,
      "num_input_tokens_seen": 108231080,
      "step": 5035,
      "time_per_iteration": 2.5019819736480713
    },
    {
      "auxiliary_loss_clip": 0.01119043,
      "auxiliary_loss_mlp": 0.01037384,
      "balance_loss_clip": 1.04488468,
      "balance_loss_mlp": 1.02359128,
      "epoch": 0.30278070043589356,
      "flos": 27344539393920.0,
      "grad_norm": 10.817542083870004,
      "language_loss": 0.87663853,
      "learning_rate": 3.2689088691495196e-06,
      "loss": 0.89820278,
      "num_input_tokens_seen": 108251125,
      "step": 5036,
      "time_per_iteration": 2.525693655014038
    },
    {
      "auxiliary_loss_clip": 0.01093018,
      "auxiliary_loss_mlp": 0.01047245,
      "balance_loss_clip": 1.04703236,
      "balance_loss_mlp": 1.03094947,
      "epoch": 0.3028408236885616,
      "flos": 24786227070720.0,
      "grad_norm": 1.588445482514792,
      "language_loss": 0.77619648,
      "learning_rate": 3.268607806688536e-06,
      "loss": 0.7975992,
      "num_input_tokens_seen": 108272545,
      "step": 5037,
      "time_per_iteration": 2.571694850921631
    },
    {
      "auxiliary_loss_clip": 0.01104658,
      "auxiliary_loss_mlp": 0.01046705,
      "balance_loss_clip": 1.05111504,
      "balance_loss_mlp": 1.03108907,
      "epoch": 0.30290094694122954,
      "flos": 12932474670720.0,
      "grad_norm": 4.076112072211769,
      "language_loss": 0.77324855,
      "learning_rate": 3.268306696121816e-06,
      "loss": 0.79476219,
      "num_input_tokens_seen": 108289725,
      "step": 5038,
      "time_per_iteration": 2.5026886463165283
    },
    {
      "auxiliary_loss_clip": 0.01110972,
      "auxiliary_loss_mlp": 0.0103866,
      "balance_loss_clip": 1.04870868,
      "balance_loss_mlp": 1.02365112,
      "epoch": 0.3029610701938975,
      "flos": 25916443488000.0,
      "grad_norm": 2.3488857310617965,
      "language_loss": 0.74242091,
      "learning_rate": 3.2680055374607804e-06,
      "loss": 0.76391727,
      "num_input_tokens_seen": 108310690,
      "step": 5039,
      "time_per_iteration": 2.5622355937957764
    },
    {
      "auxiliary_loss_clip": 0.0113231,
      "auxiliary_loss_mlp": 0.00789317,
      "balance_loss_clip": 1.04859507,
      "balance_loss_mlp": 1.00950956,
      "epoch": 0.3030211934465655,
      "flos": 21980993679360.0,
      "grad_norm": 1.7824040690755476,
      "language_loss": 0.79765821,
      "learning_rate": 3.267704330716847e-06,
      "loss": 0.8168745,
      "num_input_tokens_seen": 108328905,
      "step": 5040,
      "time_per_iteration": 2.4578700065612793
    },
    {
      "auxiliary_loss_clip": 0.01111208,
      "auxiliary_loss_mlp": 0.0104063,
      "balance_loss_clip": 1.05015945,
      "balance_loss_mlp": 1.02638423,
      "epoch": 0.30308131669923344,
      "flos": 20991977625600.0,
      "grad_norm": 2.9039556321710083,
      "language_loss": 0.82131392,
      "learning_rate": 3.267403075901438e-06,
      "loss": 0.84283227,
      "num_input_tokens_seen": 108346680,
      "step": 5041,
      "time_per_iteration": 2.533137798309326
    },
    {
      "auxiliary_loss_clip": 0.01027592,
      "auxiliary_loss_mlp": 0.01005447,
      "balance_loss_clip": 1.03622508,
      "balance_loss_mlp": 1.00365877,
      "epoch": 0.3031414399519014,
      "flos": 60548875827840.0,
      "grad_norm": 0.7545294962064434,
      "language_loss": 0.59510523,
      "learning_rate": 3.267101773025978e-06,
      "loss": 0.6154356,
      "num_input_tokens_seen": 108413885,
      "step": 5042,
      "time_per_iteration": 3.2799031734466553
    },
    {
      "auxiliary_loss_clip": 0.01138036,
      "auxiliary_loss_mlp": 0.01036116,
      "balance_loss_clip": 1.04932892,
      "balance_loss_mlp": 1.02099395,
      "epoch": 0.30320156320456937,
      "flos": 21907664064000.0,
      "grad_norm": 1.7929005585552833,
      "language_loss": 0.7137714,
      "learning_rate": 3.266800422101892e-06,
      "loss": 0.73551291,
      "num_input_tokens_seen": 108433640,
      "step": 5043,
      "time_per_iteration": 2.486018657684326
    },
    {
      "auxiliary_loss_clip": 0.01090154,
      "auxiliary_loss_mlp": 0.01032985,
      "balance_loss_clip": 1.04562855,
      "balance_loss_mlp": 1.01825702,
      "epoch": 0.30326168645723733,
      "flos": 21652770176640.0,
      "grad_norm": 1.759179725413549,
      "language_loss": 0.69137061,
      "learning_rate": 3.266499023140606e-06,
      "loss": 0.71260202,
      "num_input_tokens_seen": 108452640,
      "step": 5044,
      "time_per_iteration": 2.5620553493499756
    },
    {
      "auxiliary_loss_clip": 0.01121967,
      "auxiliary_loss_mlp": 0.01038655,
      "balance_loss_clip": 1.04672647,
      "balance_loss_mlp": 1.02316999,
      "epoch": 0.3033218097099053,
      "flos": 21871286565120.0,
      "grad_norm": 1.3922843749442546,
      "language_loss": 0.77247328,
      "learning_rate": 3.2661975761535513e-06,
      "loss": 0.79407954,
      "num_input_tokens_seen": 108472470,
      "step": 5045,
      "time_per_iteration": 2.49916934967041
    },
    {
      "auxiliary_loss_clip": 0.01133578,
      "auxiliary_loss_mlp": 0.0079288,
      "balance_loss_clip": 1.04670262,
      "balance_loss_mlp": 1.01428211,
      "epoch": 0.30338193296257326,
      "flos": 27089717333760.0,
      "grad_norm": 1.5981280459189566,
      "language_loss": 0.7277652,
      "learning_rate": 3.2658960811521564e-06,
      "loss": 0.74702978,
      "num_input_tokens_seen": 108493025,
      "step": 5046,
      "time_per_iteration": 2.4955668449401855
    },
    {
      "auxiliary_loss_clip": 0.01129494,
      "auxiliary_loss_mlp": 0.01039415,
      "balance_loss_clip": 1.04668891,
      "balance_loss_mlp": 1.02180791,
      "epoch": 0.30344205621524123,
      "flos": 19534363718400.0,
      "grad_norm": 2.022494985953209,
      "language_loss": 0.80976367,
      "learning_rate": 3.2655945381478564e-06,
      "loss": 0.83145273,
      "num_input_tokens_seen": 108513480,
      "step": 5047,
      "time_per_iteration": 2.5077357292175293
    },
    {
      "auxiliary_loss_clip": 0.01076565,
      "auxiliary_loss_mlp": 0.01043602,
      "balance_loss_clip": 1.04660368,
      "balance_loss_mlp": 1.02864122,
      "epoch": 0.3035021794679092,
      "flos": 23910976368000.0,
      "grad_norm": 1.74240832908338,
      "language_loss": 0.72610414,
      "learning_rate": 3.265292947152084e-06,
      "loss": 0.74730587,
      "num_input_tokens_seen": 108533155,
      "step": 5048,
      "time_per_iteration": 2.605783700942993
    },
    {
      "auxiliary_loss_clip": 0.01109763,
      "auxiliary_loss_mlp": 0.01031945,
      "balance_loss_clip": 1.04464102,
      "balance_loss_mlp": 1.01777101,
      "epoch": 0.30356230272057716,
      "flos": 16143606725760.0,
      "grad_norm": 2.109818493952084,
      "language_loss": 0.75320429,
      "learning_rate": 3.2649913081762763e-06,
      "loss": 0.77462137,
      "num_input_tokens_seen": 108551900,
      "step": 5049,
      "time_per_iteration": 2.5007612705230713
    },
    {
      "auxiliary_loss_clip": 0.01123869,
      "auxiliary_loss_mlp": 0.01035761,
      "balance_loss_clip": 1.0444746,
      "balance_loss_mlp": 1.02084827,
      "epoch": 0.3036224259732452,
      "flos": 28914697589760.0,
      "grad_norm": 1.6689503744029792,
      "language_loss": 0.81846428,
      "learning_rate": 3.2646896212318717e-06,
      "loss": 0.84006059,
      "num_input_tokens_seen": 108574005,
      "step": 5050,
      "time_per_iteration": 2.531283378601074
    },
    {
      "auxiliary_loss_clip": 0.01097535,
      "auxiliary_loss_mlp": 0.01038065,
      "balance_loss_clip": 1.04568982,
      "balance_loss_mlp": 1.02179301,
      "epoch": 0.30368254922591315,
      "flos": 21105599322240.0,
      "grad_norm": 2.1983781286528616,
      "language_loss": 0.7373625,
      "learning_rate": 3.2643878863303106e-06,
      "loss": 0.75871855,
      "num_input_tokens_seen": 108592715,
      "step": 5051,
      "time_per_iteration": 2.5411248207092285
    },
    {
      "auxiliary_loss_clip": 0.01075062,
      "auxiliary_loss_mlp": 0.00792748,
      "balance_loss_clip": 1.04527879,
      "balance_loss_mlp": 1.01166964,
      "epoch": 0.3037426724785811,
      "flos": 23002293081600.0,
      "grad_norm": 1.6831582637027438,
      "language_loss": 0.76367122,
      "learning_rate": 3.264086103483033e-06,
      "loss": 0.78234935,
      "num_input_tokens_seen": 108611770,
      "step": 5052,
      "time_per_iteration": 2.618191957473755
    },
    {
      "auxiliary_loss_clip": 0.01135883,
      "auxiliary_loss_mlp": 0.01043895,
      "balance_loss_clip": 1.04660511,
      "balance_loss_mlp": 1.02824903,
      "epoch": 0.3038027957312491,
      "flos": 15632705629440.0,
      "grad_norm": 2.8262253379026636,
      "language_loss": 0.82604003,
      "learning_rate": 3.2637842727014836e-06,
      "loss": 0.84783787,
      "num_input_tokens_seen": 108629070,
      "step": 5053,
      "time_per_iteration": 3.827930450439453
    },
    {
      "auxiliary_loss_clip": 0.0111127,
      "auxiliary_loss_mlp": 0.01037155,
      "balance_loss_clip": 1.04815972,
      "balance_loss_mlp": 1.02140749,
      "epoch": 0.30386291898391704,
      "flos": 12713994195840.0,
      "grad_norm": 1.5932889222189386,
      "language_loss": 0.70945603,
      "learning_rate": 3.2634823939971083e-06,
      "loss": 0.73094028,
      "num_input_tokens_seen": 108646315,
      "step": 5054,
      "time_per_iteration": 2.5225436687469482
    },
    {
      "auxiliary_loss_clip": 0.01134548,
      "auxiliary_loss_mlp": 0.01039848,
      "balance_loss_clip": 1.04748714,
      "balance_loss_mlp": 1.02410674,
      "epoch": 0.303923042236585,
      "flos": 26359437922560.0,
      "grad_norm": 2.0166861590588443,
      "language_loss": 0.69648457,
      "learning_rate": 3.2631804673813545e-06,
      "loss": 0.71822852,
      "num_input_tokens_seen": 108665920,
      "step": 5055,
      "time_per_iteration": 2.4945602416992188
    },
    {
      "auxiliary_loss_clip": 0.01112124,
      "auxiliary_loss_mlp": 0.01037239,
      "balance_loss_clip": 1.05117643,
      "balance_loss_mlp": 1.02093148,
      "epoch": 0.30398316548925297,
      "flos": 19719232041600.0,
      "grad_norm": 1.7982401703821733,
      "language_loss": 0.68174469,
      "learning_rate": 3.2628784928656707e-06,
      "loss": 0.70323837,
      "num_input_tokens_seen": 108683485,
      "step": 5056,
      "time_per_iteration": 3.88393235206604
    },
    {
      "auxiliary_loss_clip": 0.01109559,
      "auxiliary_loss_mlp": 0.01042475,
      "balance_loss_clip": 1.04798222,
      "balance_loss_mlp": 1.02721667,
      "epoch": 0.30404328874192094,
      "flos": 24239846315520.0,
      "grad_norm": 1.895815421600517,
      "language_loss": 0.82551587,
      "learning_rate": 3.262576470461507e-06,
      "loss": 0.84703624,
      "num_input_tokens_seen": 108702700,
      "step": 5057,
      "time_per_iteration": 2.5427966117858887
    },
    {
      "auxiliary_loss_clip": 0.0110859,
      "auxiliary_loss_mlp": 0.01036634,
      "balance_loss_clip": 1.04384065,
      "balance_loss_mlp": 1.0211488,
      "epoch": 0.3041034119945889,
      "flos": 24498942094080.0,
      "grad_norm": 1.695535926361367,
      "language_loss": 0.8886981,
      "learning_rate": 3.2622744001803176e-06,
      "loss": 0.91015029,
      "num_input_tokens_seen": 108721860,
      "step": 5058,
      "time_per_iteration": 2.544731616973877
    },
    {
      "auxiliary_loss_clip": 0.01099867,
      "auxiliary_loss_mlp": 0.0103833,
      "balance_loss_clip": 1.04681528,
      "balance_loss_mlp": 1.0227437,
      "epoch": 0.30416353524725687,
      "flos": 28288881907200.0,
      "grad_norm": 1.912162141114911,
      "language_loss": 0.71224678,
      "learning_rate": 3.2619722820335564e-06,
      "loss": 0.73362875,
      "num_input_tokens_seen": 108743215,
      "step": 5059,
      "time_per_iteration": 3.988891839981079
    },
    {
      "auxiliary_loss_clip": 0.01075123,
      "auxiliary_loss_mlp": 0.01036582,
      "balance_loss_clip": 1.04522324,
      "balance_loss_mlp": 1.02160943,
      "epoch": 0.30422365849992483,
      "flos": 23660392112640.0,
      "grad_norm": 1.5256326920730554,
      "language_loss": 0.72842723,
      "learning_rate": 3.26167011603268e-06,
      "loss": 0.74954426,
      "num_input_tokens_seen": 108765505,
      "step": 5060,
      "time_per_iteration": 4.0895915031433105
    },
    {
      "auxiliary_loss_clip": 0.0113517,
      "auxiliary_loss_mlp": 0.01036365,
      "balance_loss_clip": 1.04798388,
      "balance_loss_mlp": 1.02126122,
      "epoch": 0.3042837817525928,
      "flos": 22998773548800.0,
      "grad_norm": 2.0403968921607216,
      "language_loss": 0.76854384,
      "learning_rate": 3.2613679021891463e-06,
      "loss": 0.79025924,
      "num_input_tokens_seen": 108783370,
      "step": 5061,
      "time_per_iteration": 2.4672601222991943
    },
    {
      "auxiliary_loss_clip": 0.01101666,
      "auxiliary_loss_mlp": 0.01039382,
      "balance_loss_clip": 1.0544219,
      "balance_loss_mlp": 1.02262068,
      "epoch": 0.30434390500526076,
      "flos": 22082332924800.0,
      "grad_norm": 4.613734378002912,
      "language_loss": 0.81691206,
      "learning_rate": 3.261065640514415e-06,
      "loss": 0.83832258,
      "num_input_tokens_seen": 108797430,
      "step": 5062,
      "time_per_iteration": 2.538012981414795
    },
    {
      "auxiliary_loss_clip": 0.01128933,
      "auxiliary_loss_mlp": 0.01028974,
      "balance_loss_clip": 1.0445447,
      "balance_loss_mlp": 1.01490736,
      "epoch": 0.3044040282579287,
      "flos": 25483504861440.0,
      "grad_norm": 2.415443361972339,
      "language_loss": 0.74785197,
      "learning_rate": 3.2607633310199483e-06,
      "loss": 0.76943111,
      "num_input_tokens_seen": 108816945,
      "step": 5063,
      "time_per_iteration": 2.493812084197998
    },
    {
      "auxiliary_loss_clip": 0.01122729,
      "auxiliary_loss_mlp": 0.00793142,
      "balance_loss_clip": 1.0479188,
      "balance_loss_mlp": 1.0148586,
      "epoch": 0.30446415151059675,
      "flos": 21945478106880.0,
      "grad_norm": 1.5590367185508835,
      "language_loss": 0.84223461,
      "learning_rate": 3.26046097371721e-06,
      "loss": 0.86139333,
      "num_input_tokens_seen": 108836615,
      "step": 5064,
      "time_per_iteration": 2.545823574066162
    },
    {
      "auxiliary_loss_clip": 0.01123941,
      "auxiliary_loss_mlp": 0.01035402,
      "balance_loss_clip": 1.04884815,
      "balance_loss_mlp": 1.01945138,
      "epoch": 0.3045242747632647,
      "flos": 16435416816000.0,
      "grad_norm": 2.152686934158601,
      "language_loss": 0.7569294,
      "learning_rate": 3.2601585686176655e-06,
      "loss": 0.77852273,
      "num_input_tokens_seen": 108855165,
      "step": 5065,
      "time_per_iteration": 2.475019693374634
    },
    {
      "auxiliary_loss_clip": 0.01111469,
      "auxiliary_loss_mlp": 0.01036756,
      "balance_loss_clip": 1.04949522,
      "balance_loss_mlp": 1.02088904,
      "epoch": 0.3045843980159327,
      "flos": 31540341957120.0,
      "grad_norm": 2.130802735175459,
      "language_loss": 0.62170458,
      "learning_rate": 3.2598561157327814e-06,
      "loss": 0.64318681,
      "num_input_tokens_seen": 108874690,
      "step": 5066,
      "time_per_iteration": 2.6204943656921387
    },
    {
      "auxiliary_loss_clip": 0.01116099,
      "auxiliary_loss_mlp": 0.01047935,
      "balance_loss_clip": 1.05782771,
      "balance_loss_mlp": 1.03217566,
      "epoch": 0.30464452126860064,
      "flos": 17853636481920.0,
      "grad_norm": 1.9852801388067143,
      "language_loss": 0.82529271,
      "learning_rate": 3.2595536150740265e-06,
      "loss": 0.84693301,
      "num_input_tokens_seen": 108893140,
      "step": 5067,
      "time_per_iteration": 2.5068275928497314
    },
    {
      "auxiliary_loss_clip": 0.01134223,
      "auxiliary_loss_mlp": 0.01038643,
      "balance_loss_clip": 1.04961383,
      "balance_loss_mlp": 1.02350366,
      "epoch": 0.3047046445212686,
      "flos": 20631398947200.0,
      "grad_norm": 1.845595786188185,
      "language_loss": 0.63034922,
      "learning_rate": 3.259251066652873e-06,
      "loss": 0.65207791,
      "num_input_tokens_seen": 108911880,
      "step": 5068,
      "time_per_iteration": 2.4847700595855713
    },
    {
      "auxiliary_loss_clip": 0.01126455,
      "auxiliary_loss_mlp": 0.01033221,
      "balance_loss_clip": 1.04755592,
      "balance_loss_mlp": 1.01761651,
      "epoch": 0.3047647677739366,
      "flos": 21287594557440.0,
      "grad_norm": 1.6034764769250842,
      "language_loss": 0.74536997,
      "learning_rate": 3.258948470480793e-06,
      "loss": 0.7669667,
      "num_input_tokens_seen": 108930440,
      "step": 5069,
      "time_per_iteration": 2.4868061542510986
    },
    {
      "auxiliary_loss_clip": 0.01106698,
      "auxiliary_loss_mlp": 0.01045708,
      "balance_loss_clip": 1.04889965,
      "balance_loss_mlp": 1.03099179,
      "epoch": 0.30482489102660454,
      "flos": 20995928121600.0,
      "grad_norm": 2.909800978394936,
      "language_loss": 0.76006079,
      "learning_rate": 3.258645826569261e-06,
      "loss": 0.78158474,
      "num_input_tokens_seen": 108949125,
      "step": 5070,
      "time_per_iteration": 2.5581939220428467
    },
    {
      "auxiliary_loss_clip": 0.01138893,
      "auxiliary_loss_mlp": 0.00791519,
      "balance_loss_clip": 1.04957914,
      "balance_loss_mlp": 1.01184595,
      "epoch": 0.3048850142792725,
      "flos": 26290812988800.0,
      "grad_norm": 1.6507559660289683,
      "language_loss": 0.81460655,
      "learning_rate": 3.2583431349297527e-06,
      "loss": 0.83391058,
      "num_input_tokens_seen": 108972190,
      "step": 5071,
      "time_per_iteration": 2.52705979347229
    },
    {
      "auxiliary_loss_clip": 0.0111047,
      "auxiliary_loss_mlp": 0.01044877,
      "balance_loss_clip": 1.05056286,
      "balance_loss_mlp": 1.02825928,
      "epoch": 0.30494513753194047,
      "flos": 22346241125760.0,
      "grad_norm": 1.6053323569643994,
      "language_loss": 0.7587418,
      "learning_rate": 3.2580403955737467e-06,
      "loss": 0.78029525,
      "num_input_tokens_seen": 108990325,
      "step": 5072,
      "time_per_iteration": 2.5522186756134033
    },
    {
      "auxiliary_loss_clip": 0.01100834,
      "auxiliary_loss_mlp": 0.01047178,
      "balance_loss_clip": 1.05258882,
      "balance_loss_mlp": 1.03095388,
      "epoch": 0.30500526078460843,
      "flos": 19537667769600.0,
      "grad_norm": 1.9747044809637038,
      "language_loss": 0.71797055,
      "learning_rate": 3.257737608512723e-06,
      "loss": 0.73945069,
      "num_input_tokens_seen": 109009505,
      "step": 5073,
      "time_per_iteration": 2.545243978500366
    },
    {
      "auxiliary_loss_clip": 0.01131541,
      "auxiliary_loss_mlp": 0.01048105,
      "balance_loss_clip": 1.0511663,
      "balance_loss_mlp": 1.03203571,
      "epoch": 0.3050653840372764,
      "flos": 14465321614080.0,
      "grad_norm": 2.0679825755195047,
      "language_loss": 0.7684505,
      "learning_rate": 3.257434773758163e-06,
      "loss": 0.79024696,
      "num_input_tokens_seen": 109026350,
      "step": 5074,
      "time_per_iteration": 2.462851047515869
    },
    {
      "auxiliary_loss_clip": 0.01116082,
      "auxiliary_loss_mlp": 0.01032161,
      "balance_loss_clip": 1.05245876,
      "balance_loss_mlp": 1.01736689,
      "epoch": 0.30512550728994436,
      "flos": 24243796811520.0,
      "grad_norm": 2.170565671155647,
      "language_loss": 0.74483377,
      "learning_rate": 3.25713189132155e-06,
      "loss": 0.76631618,
      "num_input_tokens_seen": 109044165,
      "step": 5075,
      "time_per_iteration": 2.5491833686828613
    },
    {
      "auxiliary_loss_clip": 0.0113848,
      "auxiliary_loss_mlp": 0.01046444,
      "balance_loss_clip": 1.04955614,
      "balance_loss_mlp": 1.02887249,
      "epoch": 0.30518563054261233,
      "flos": 16360542915840.0,
      "grad_norm": 2.141357883454221,
      "language_loss": 0.75773489,
      "learning_rate": 3.2568289612143703e-06,
      "loss": 0.77958411,
      "num_input_tokens_seen": 109060665,
      "step": 5076,
      "time_per_iteration": 2.4443511962890625
    },
    {
      "auxiliary_loss_clip": 0.0112506,
      "auxiliary_loss_mlp": 0.01037491,
      "balance_loss_clip": 1.05398703,
      "balance_loss_mlp": 1.02194643,
      "epoch": 0.30524575379528035,
      "flos": 21579584215680.0,
      "grad_norm": 1.3949293248851937,
      "language_loss": 0.79628909,
      "learning_rate": 3.25652598344811e-06,
      "loss": 0.81791461,
      "num_input_tokens_seen": 109080035,
      "step": 5077,
      "time_per_iteration": 2.523547649383545
    },
    {
      "auxiliary_loss_clip": 0.01090009,
      "auxiliary_loss_mlp": 0.01034945,
      "balance_loss_clip": 1.05000806,
      "balance_loss_mlp": 1.02030575,
      "epoch": 0.3053058770479483,
      "flos": 16545231671040.0,
      "grad_norm": 1.5383924912404239,
      "language_loss": 0.74878854,
      "learning_rate": 3.256222958034259e-06,
      "loss": 0.77003813,
      "num_input_tokens_seen": 109097385,
      "step": 5078,
      "time_per_iteration": 2.5725393295288086
    },
    {
      "auxiliary_loss_clip": 0.01087735,
      "auxiliary_loss_mlp": 0.0105293,
      "balance_loss_clip": 1.05243587,
      "balance_loss_mlp": 1.03696179,
      "epoch": 0.3053660003006163,
      "flos": 12312907954560.0,
      "grad_norm": 1.9493642566642697,
      "language_loss": 0.66645539,
      "learning_rate": 3.255919884984307e-06,
      "loss": 0.68786204,
      "num_input_tokens_seen": 109115495,
      "step": 5079,
      "time_per_iteration": 2.5597622394561768
    },
    {
      "auxiliary_loss_clip": 0.01125055,
      "auxiliary_loss_mlp": 0.01037471,
      "balance_loss_clip": 1.04903173,
      "balance_loss_mlp": 1.02276051,
      "epoch": 0.30542612355328425,
      "flos": 23112287504640.0,
      "grad_norm": 1.9441253088566703,
      "language_loss": 0.79768336,
      "learning_rate": 3.2556167643097477e-06,
      "loss": 0.81930864,
      "num_input_tokens_seen": 109134235,
      "step": 5080,
      "time_per_iteration": 2.509190082550049
    },
    {
      "auxiliary_loss_clip": 0.01124915,
      "auxiliary_loss_mlp": 0.00791136,
      "balance_loss_clip": 1.05104709,
      "balance_loss_mlp": 1.01269221,
      "epoch": 0.3054862468059522,
      "flos": 24389450461440.0,
      "grad_norm": 2.8986846170070724,
      "language_loss": 0.81030858,
      "learning_rate": 3.255313596022074e-06,
      "loss": 0.82946908,
      "num_input_tokens_seen": 109152760,
      "step": 5081,
      "time_per_iteration": 2.529574155807495
    },
    {
      "auxiliary_loss_clip": 0.01121627,
      "auxiliary_loss_mlp": 0.01043048,
      "balance_loss_clip": 1.05218542,
      "balance_loss_mlp": 1.02803993,
      "epoch": 0.3055463700586202,
      "flos": 29386096704000.0,
      "grad_norm": 1.6520753483990296,
      "language_loss": 0.71927911,
      "learning_rate": 3.255010380132783e-06,
      "loss": 0.74092585,
      "num_input_tokens_seen": 109173925,
      "step": 5082,
      "time_per_iteration": 2.562056541442871
    },
    {
      "auxiliary_loss_clip": 0.01125385,
      "auxiliary_loss_mlp": 0.01036816,
      "balance_loss_clip": 1.04792964,
      "balance_loss_mlp": 1.02022195,
      "epoch": 0.30560649331128814,
      "flos": 25591775431680.0,
      "grad_norm": 1.8211269826107046,
      "language_loss": 0.72992283,
      "learning_rate": 3.2547071166533736e-06,
      "loss": 0.75154483,
      "num_input_tokens_seen": 109192510,
      "step": 5083,
      "time_per_iteration": 2.5121631622314453
    },
    {
      "auxiliary_loss_clip": 0.01107363,
      "auxiliary_loss_mlp": 0.00791623,
      "balance_loss_clip": 1.04900575,
      "balance_loss_mlp": 1.00923359,
      "epoch": 0.3056666165639561,
      "flos": 19128321400320.0,
      "grad_norm": 1.7299783538864115,
      "language_loss": 0.70932376,
      "learning_rate": 3.254403805595344e-06,
      "loss": 0.72831368,
      "num_input_tokens_seen": 109210885,
      "step": 5084,
      "time_per_iteration": 2.52413010597229
    },
    {
      "auxiliary_loss_clip": 0.01100134,
      "auxiliary_loss_mlp": 0.0104252,
      "balance_loss_clip": 1.04889369,
      "balance_loss_mlp": 1.02448392,
      "epoch": 0.30572673981662407,
      "flos": 15523860441600.0,
      "grad_norm": 1.9101004536408241,
      "language_loss": 0.78309613,
      "learning_rate": 3.2541004469701962e-06,
      "loss": 0.80452263,
      "num_input_tokens_seen": 109229180,
      "step": 5085,
      "time_per_iteration": 2.5433623790740967
    },
    {
      "auxiliary_loss_clip": 0.0112963,
      "auxiliary_loss_mlp": 0.01035988,
      "balance_loss_clip": 1.0462749,
      "balance_loss_mlp": 1.02143264,
      "epoch": 0.30578686306929204,
      "flos": 21506541909120.0,
      "grad_norm": 1.566912184449944,
      "language_loss": 0.78487313,
      "learning_rate": 3.2537970407894342e-06,
      "loss": 0.80652928,
      "num_input_tokens_seen": 109249510,
      "step": 5086,
      "time_per_iteration": 2.4827637672424316
    },
    {
      "auxiliary_loss_clip": 0.01105398,
      "auxiliary_loss_mlp": 0.01045685,
      "balance_loss_clip": 1.05138218,
      "balance_loss_mlp": 1.02816749,
      "epoch": 0.30584698632196,
      "flos": 20954271323520.0,
      "grad_norm": 2.239845278641564,
      "language_loss": 0.76728064,
      "learning_rate": 3.253493587064563e-06,
      "loss": 0.78879148,
      "num_input_tokens_seen": 109268200,
      "step": 5087,
      "time_per_iteration": 2.527249336242676
    },
    {
      "auxiliary_loss_clip": 0.01122796,
      "auxiliary_loss_mlp": 0.01040038,
      "balance_loss_clip": 1.04530907,
      "balance_loss_mlp": 1.02387357,
      "epoch": 0.30590710957462797,
      "flos": 24681116897280.0,
      "grad_norm": 1.9662993075073387,
      "language_loss": 0.72105432,
      "learning_rate": 3.2531900858070885e-06,
      "loss": 0.7426827,
      "num_input_tokens_seen": 109288370,
      "step": 5088,
      "time_per_iteration": 2.5317161083221436
    },
    {
      "auxiliary_loss_clip": 0.01124637,
      "auxiliary_loss_mlp": 0.01038311,
      "balance_loss_clip": 1.04650855,
      "balance_loss_mlp": 1.02218151,
      "epoch": 0.30596723282729593,
      "flos": 17086907744640.0,
      "grad_norm": 2.2247932546184495,
      "language_loss": 0.79311979,
      "learning_rate": 3.252886537028521e-06,
      "loss": 0.81474924,
      "num_input_tokens_seen": 109306730,
      "step": 5089,
      "time_per_iteration": 2.4795587062835693
    },
    {
      "auxiliary_loss_clip": 0.01110333,
      "auxiliary_loss_mlp": 0.01039859,
      "balance_loss_clip": 1.05140638,
      "balance_loss_mlp": 1.02414107,
      "epoch": 0.30602735607996395,
      "flos": 22857106308480.0,
      "grad_norm": 1.7530407327365238,
      "language_loss": 0.77307272,
      "learning_rate": 3.2525829407403703e-06,
      "loss": 0.79457462,
      "num_input_tokens_seen": 109327360,
      "step": 5090,
      "time_per_iteration": 2.5709004402160645
    },
    {
      "auxiliary_loss_clip": 0.01114679,
      "auxiliary_loss_mlp": 0.0105012,
      "balance_loss_clip": 1.04628253,
      "balance_loss_mlp": 1.03403854,
      "epoch": 0.3060874793326319,
      "flos": 29861482227840.0,
      "grad_norm": 1.8669347185967908,
      "language_loss": 0.7638979,
      "learning_rate": 3.2522792969541488e-06,
      "loss": 0.78554583,
      "num_input_tokens_seen": 109348135,
      "step": 5091,
      "time_per_iteration": 4.047200918197632
    },
    {
      "auxiliary_loss_clip": 0.01068696,
      "auxiliary_loss_mlp": 0.01043239,
      "balance_loss_clip": 1.04349279,
      "balance_loss_mlp": 1.02540553,
      "epoch": 0.3061476025852999,
      "flos": 20448577699200.0,
      "grad_norm": 1.8826241953480907,
      "language_loss": 0.71840894,
      "learning_rate": 3.2519756056813705e-06,
      "loss": 0.73952836,
      "num_input_tokens_seen": 109366220,
      "step": 5092,
      "time_per_iteration": 2.656700849533081
    },
    {
      "auxiliary_loss_clip": 0.01111082,
      "auxiliary_loss_mlp": 0.01036918,
      "balance_loss_clip": 1.04739928,
      "balance_loss_mlp": 1.02256536,
      "epoch": 0.30620772583796785,
      "flos": 19391475415680.0,
      "grad_norm": 1.7260120045400271,
      "language_loss": 0.82041371,
      "learning_rate": 3.2516718669335522e-06,
      "loss": 0.84189373,
      "num_input_tokens_seen": 109385260,
      "step": 5093,
      "time_per_iteration": 2.5346693992614746
    },
    {
      "auxiliary_loss_clip": 0.01133573,
      "auxiliary_loss_mlp": 0.00790223,
      "balance_loss_clip": 1.0484128,
      "balance_loss_mlp": 1.01192605,
      "epoch": 0.3062678490906358,
      "flos": 24024562151040.0,
      "grad_norm": 1.8713740157555094,
      "language_loss": 0.75356257,
      "learning_rate": 3.2513680807222114e-06,
      "loss": 0.77280056,
      "num_input_tokens_seen": 109405025,
      "step": 5094,
      "time_per_iteration": 2.4772167205810547
    },
    {
      "auxiliary_loss_clip": 0.01112545,
      "auxiliary_loss_mlp": 0.01039696,
      "balance_loss_clip": 1.05037808,
      "balance_loss_mlp": 1.02527213,
      "epoch": 0.3063279723433038,
      "flos": 19754639873280.0,
      "grad_norm": 2.1314616778106443,
      "language_loss": 0.76144707,
      "learning_rate": 3.251064247058868e-06,
      "loss": 0.78296947,
      "num_input_tokens_seen": 109422465,
      "step": 5095,
      "time_per_iteration": 3.911518096923828
    },
    {
      "auxiliary_loss_clip": 0.01120638,
      "auxiliary_loss_mlp": 0.01043504,
      "balance_loss_clip": 1.04719055,
      "balance_loss_mlp": 1.02860284,
      "epoch": 0.30638809559597174,
      "flos": 22450022496000.0,
      "grad_norm": 1.6541401228138783,
      "language_loss": 0.80679047,
      "learning_rate": 3.250760365955042e-06,
      "loss": 0.8284319,
      "num_input_tokens_seen": 109440575,
      "step": 5096,
      "time_per_iteration": 2.4915823936462402
    },
    {
      "auxiliary_loss_clip": 0.01125517,
      "auxiliary_loss_mlp": 0.01035554,
      "balance_loss_clip": 1.04893947,
      "balance_loss_mlp": 1.02078414,
      "epoch": 0.3064482188486397,
      "flos": 17165157523200.0,
      "grad_norm": 1.9411747073422465,
      "language_loss": 0.81598175,
      "learning_rate": 3.250456437422258e-06,
      "loss": 0.83759242,
      "num_input_tokens_seen": 109459050,
      "step": 5097,
      "time_per_iteration": 2.465376615524292
    },
    {
      "auxiliary_loss_clip": 0.01135781,
      "auxiliary_loss_mlp": 0.0104149,
      "balance_loss_clip": 1.04894066,
      "balance_loss_mlp": 1.02482486,
      "epoch": 0.3065083421013077,
      "flos": 23768483114880.0,
      "grad_norm": 1.93054872045021,
      "language_loss": 0.77965021,
      "learning_rate": 3.250152461472041e-06,
      "loss": 0.80142295,
      "num_input_tokens_seen": 109475860,
      "step": 5098,
      "time_per_iteration": 5.191356897354126
    },
    {
      "auxiliary_loss_clip": 0.01096508,
      "auxiliary_loss_mlp": 0.01038897,
      "balance_loss_clip": 1.05094504,
      "balance_loss_mlp": 1.02330399,
      "epoch": 0.30656846535397564,
      "flos": 26431833784320.0,
      "grad_norm": 1.806936067939058,
      "language_loss": 0.84155428,
      "learning_rate": 3.249848438115917e-06,
      "loss": 0.86290836,
      "num_input_tokens_seen": 109494760,
      "step": 5099,
      "time_per_iteration": 2.616321086883545
    },
    {
      "auxiliary_loss_clip": 0.01137024,
      "auxiliary_loss_mlp": 0.01042706,
      "balance_loss_clip": 1.04791307,
      "balance_loss_mlp": 1.0272032,
      "epoch": 0.3066285886066436,
      "flos": 26651786716800.0,
      "grad_norm": 1.5669882924396057,
      "language_loss": 0.85359192,
      "learning_rate": 3.2495443673654148e-06,
      "loss": 0.87538922,
      "num_input_tokens_seen": 109516480,
      "step": 5100,
      "time_per_iteration": 2.5169858932495117
    },
    {
      "auxiliary_loss_clip": 0.01099296,
      "auxiliary_loss_mlp": 0.01035528,
      "balance_loss_clip": 1.04454994,
      "balance_loss_mlp": 1.01904178,
      "epoch": 0.30668871185931157,
      "flos": 15049947375360.0,
      "grad_norm": 1.6930061357631445,
      "language_loss": 0.78801191,
      "learning_rate": 3.249240249232065e-06,
      "loss": 0.80936015,
      "num_input_tokens_seen": 109534615,
      "step": 5101,
      "time_per_iteration": 2.5151965618133545
    },
    {
      "auxiliary_loss_clip": 0.01102271,
      "auxiliary_loss_mlp": 0.01043094,
      "balance_loss_clip": 1.05182481,
      "balance_loss_mlp": 1.02558827,
      "epoch": 0.30674883511197953,
      "flos": 20082109190400.0,
      "grad_norm": 1.8354918150005812,
      "language_loss": 0.79608208,
      "learning_rate": 3.2489360837273998e-06,
      "loss": 0.81753576,
      "num_input_tokens_seen": 109554040,
      "step": 5102,
      "time_per_iteration": 2.5565876960754395
    },
    {
      "auxiliary_loss_clip": 0.01141002,
      "auxiliary_loss_mlp": 0.01043204,
      "balance_loss_clip": 1.05367947,
      "balance_loss_mlp": 1.02634764,
      "epoch": 0.30680895836464755,
      "flos": 22893807029760.0,
      "grad_norm": 1.8010833685374177,
      "language_loss": 0.88698995,
      "learning_rate": 3.2486318708629532e-06,
      "loss": 0.90883195,
      "num_input_tokens_seen": 109574345,
      "step": 5103,
      "time_per_iteration": 2.4712846279144287
    },
    {
      "auxiliary_loss_clip": 0.01120604,
      "auxiliary_loss_mlp": 0.01041874,
      "balance_loss_clip": 1.05347633,
      "balance_loss_mlp": 1.02596498,
      "epoch": 0.3068690816173155,
      "flos": 23696159080320.0,
      "grad_norm": 1.6834241723641357,
      "language_loss": 0.73778701,
      "learning_rate": 3.2483276106502607e-06,
      "loss": 0.75941181,
      "num_input_tokens_seen": 109593670,
      "step": 5104,
      "time_per_iteration": 2.5241332054138184
    },
    {
      "auxiliary_loss_clip": 0.01124122,
      "auxiliary_loss_mlp": 0.00794398,
      "balance_loss_clip": 1.04693866,
      "balance_loss_mlp": 1.0146991,
      "epoch": 0.3069292048699835,
      "flos": 23551044134400.0,
      "grad_norm": 1.8491751783777972,
      "language_loss": 0.72913712,
      "learning_rate": 3.2480233031008605e-06,
      "loss": 0.74832237,
      "num_input_tokens_seen": 109613385,
      "step": 5105,
      "time_per_iteration": 2.522522211074829
    },
    {
      "auxiliary_loss_clip": 0.01115083,
      "auxiliary_loss_mlp": 0.01041374,
      "balance_loss_clip": 1.05421078,
      "balance_loss_mlp": 1.02469611,
      "epoch": 0.30698932812265145,
      "flos": 24531656405760.0,
      "grad_norm": 1.8758791585399968,
      "language_loss": 0.87503719,
      "learning_rate": 3.2477189482262916e-06,
      "loss": 0.8966018,
      "num_input_tokens_seen": 109632395,
      "step": 5106,
      "time_per_iteration": 2.5765206813812256
    },
    {
      "auxiliary_loss_clip": 0.01107248,
      "auxiliary_loss_mlp": 0.01048216,
      "balance_loss_clip": 1.04846263,
      "balance_loss_mlp": 1.03139567,
      "epoch": 0.3070494513753194,
      "flos": 20996430912000.0,
      "grad_norm": 2.151922630430958,
      "language_loss": 0.71787798,
      "learning_rate": 3.2474145460380945e-06,
      "loss": 0.73943257,
      "num_input_tokens_seen": 109651380,
      "step": 5107,
      "time_per_iteration": 2.549423933029175
    },
    {
      "auxiliary_loss_clip": 0.01100851,
      "auxiliary_loss_mlp": 0.01052218,
      "balance_loss_clip": 1.04795861,
      "balance_loss_mlp": 1.03629148,
      "epoch": 0.3071095746279874,
      "flos": 19025940660480.0,
      "grad_norm": 2.1866667587905955,
      "language_loss": 0.72334552,
      "learning_rate": 3.247110096547814e-06,
      "loss": 0.74487627,
      "num_input_tokens_seen": 109670240,
      "step": 5108,
      "time_per_iteration": 2.5428662300109863
    },
    {
      "auxiliary_loss_clip": 0.0111299,
      "auxiliary_loss_mlp": 0.01038792,
      "balance_loss_clip": 1.0524236,
      "balance_loss_mlp": 1.02285981,
      "epoch": 0.30716969788065535,
      "flos": 21215521918080.0,
      "grad_norm": 1.5255138991223882,
      "language_loss": 0.85776103,
      "learning_rate": 3.2468055997669926e-06,
      "loss": 0.87927878,
      "num_input_tokens_seen": 109690810,
      "step": 5109,
      "time_per_iteration": 2.5429224967956543
    },
    {
      "auxiliary_loss_clip": 0.01109672,
      "auxiliary_loss_mlp": 0.01035608,
      "balance_loss_clip": 1.04936838,
      "balance_loss_mlp": 1.0202539,
      "epoch": 0.3072298211333233,
      "flos": 25772765086080.0,
      "grad_norm": 2.1514407454316293,
      "language_loss": 0.67383152,
      "learning_rate": 3.2465010557071788e-06,
      "loss": 0.69528437,
      "num_input_tokens_seen": 109711145,
      "step": 5110,
      "time_per_iteration": 2.5701611042022705
    },
    {
      "auxiliary_loss_clip": 0.01122323,
      "auxiliary_loss_mlp": 0.01035513,
      "balance_loss_clip": 1.04767251,
      "balance_loss_mlp": 1.02108884,
      "epoch": 0.3072899443859913,
      "flos": 25848931875840.0,
      "grad_norm": 1.600201760907957,
      "language_loss": 0.77334571,
      "learning_rate": 3.246196464379919e-06,
      "loss": 0.79492408,
      "num_input_tokens_seen": 109731425,
      "step": 5111,
      "time_per_iteration": 2.552011489868164
    },
    {
      "auxiliary_loss_clip": 0.01136135,
      "auxiliary_loss_mlp": 0.01041675,
      "balance_loss_clip": 1.04920578,
      "balance_loss_mlp": 1.02643371,
      "epoch": 0.30735006763865924,
      "flos": 25922800195200.0,
      "grad_norm": 1.7096686040645723,
      "language_loss": 0.67159903,
      "learning_rate": 3.245891825796765e-06,
      "loss": 0.69337714,
      "num_input_tokens_seen": 109752720,
      "step": 5112,
      "time_per_iteration": 2.5191309452056885
    },
    {
      "auxiliary_loss_clip": 0.01131201,
      "auxiliary_loss_mlp": 0.01042866,
      "balance_loss_clip": 1.05047989,
      "balance_loss_mlp": 1.02529418,
      "epoch": 0.3074101908913272,
      "flos": 30917004312960.0,
      "grad_norm": 1.7431548811247024,
      "language_loss": 0.79471433,
      "learning_rate": 3.2455871399692678e-06,
      "loss": 0.81645501,
      "num_input_tokens_seen": 109772840,
      "step": 5113,
      "time_per_iteration": 2.5691511631011963
    },
    {
      "auxiliary_loss_clip": 0.01106529,
      "auxiliary_loss_mlp": 0.00791205,
      "balance_loss_clip": 1.051126,
      "balance_loss_mlp": 1.01013219,
      "epoch": 0.30747031414399517,
      "flos": 18401058731520.0,
      "grad_norm": 2.056844458065421,
      "language_loss": 0.77346069,
      "learning_rate": 3.2452824069089815e-06,
      "loss": 0.79243803,
      "num_input_tokens_seen": 109790150,
      "step": 5114,
      "time_per_iteration": 2.5351390838623047
    },
    {
      "auxiliary_loss_clip": 0.01100895,
      "auxiliary_loss_mlp": 0.01039742,
      "balance_loss_clip": 1.04842448,
      "balance_loss_mlp": 1.02239728,
      "epoch": 0.30753043739666314,
      "flos": 22633166966400.0,
      "grad_norm": 1.7859074538728787,
      "language_loss": 0.62136102,
      "learning_rate": 3.2449776266274623e-06,
      "loss": 0.64276743,
      "num_input_tokens_seen": 109807985,
      "step": 5115,
      "time_per_iteration": 2.5567257404327393
    },
    {
      "auxiliary_loss_clip": 0.01128583,
      "auxiliary_loss_mlp": 0.01044244,
      "balance_loss_clip": 1.04797733,
      "balance_loss_mlp": 1.02856815,
      "epoch": 0.3075905606493311,
      "flos": 27344072517120.0,
      "grad_norm": 1.8624379822359596,
      "language_loss": 0.82865864,
      "learning_rate": 3.2446727991362657e-06,
      "loss": 0.85038686,
      "num_input_tokens_seen": 109825920,
      "step": 5116,
      "time_per_iteration": 2.5321226119995117
    },
    {
      "auxiliary_loss_clip": 0.01113358,
      "auxiliary_loss_mlp": 0.01041353,
      "balance_loss_clip": 1.04987669,
      "balance_loss_mlp": 1.02555776,
      "epoch": 0.3076506839019991,
      "flos": 22090808534400.0,
      "grad_norm": 1.7580104221247757,
      "language_loss": 0.76169503,
      "learning_rate": 3.244367924446952e-06,
      "loss": 0.78324217,
      "num_input_tokens_seen": 109846220,
      "step": 5117,
      "time_per_iteration": 2.5518136024475098
    },
    {
      "auxiliary_loss_clip": 0.01096852,
      "auxiliary_loss_mlp": 0.01041227,
      "balance_loss_clip": 1.0507102,
      "balance_loss_mlp": 1.0226903,
      "epoch": 0.3077108071546671,
      "flos": 21289533891840.0,
      "grad_norm": 2.507907047117693,
      "language_loss": 0.71942323,
      "learning_rate": 3.2440630025710826e-06,
      "loss": 0.74080408,
      "num_input_tokens_seen": 109863870,
      "step": 5118,
      "time_per_iteration": 2.5447747707366943
    },
    {
      "auxiliary_loss_clip": 0.01091868,
      "auxiliary_loss_mlp": 0.01037386,
      "balance_loss_clip": 1.04947543,
      "balance_loss_mlp": 1.02165008,
      "epoch": 0.30777093040733505,
      "flos": 21430985650560.0,
      "grad_norm": 1.6267558602890053,
      "language_loss": 0.74420536,
      "learning_rate": 3.243758033520219e-06,
      "loss": 0.76549792,
      "num_input_tokens_seen": 109883500,
      "step": 5119,
      "time_per_iteration": 2.60062313079834
    },
    {
      "auxiliary_loss_clip": 0.01128956,
      "auxiliary_loss_mlp": 0.01052199,
      "balance_loss_clip": 1.04961932,
      "balance_loss_mlp": 1.03473532,
      "epoch": 0.307831053660003,
      "flos": 23149275534720.0,
      "grad_norm": 1.8159886058043166,
      "language_loss": 0.80395091,
      "learning_rate": 3.243453017305926e-06,
      "loss": 0.82576251,
      "num_input_tokens_seen": 109904620,
      "step": 5120,
      "time_per_iteration": 2.5535271167755127
    },
    {
      "auxiliary_loss_clip": 0.0112246,
      "auxiliary_loss_mlp": 0.01039652,
      "balance_loss_clip": 1.04454899,
      "balance_loss_mlp": 1.02459598,
      "epoch": 0.307891176912671,
      "flos": 17019755268480.0,
      "grad_norm": 1.5933604436822906,
      "language_loss": 0.79914868,
      "learning_rate": 3.24314795393977e-06,
      "loss": 0.82076979,
      "num_input_tokens_seen": 109922275,
      "step": 5121,
      "time_per_iteration": 2.4733738899230957
    },
    {
      "auxiliary_loss_clip": 0.01105071,
      "auxiliary_loss_mlp": 0.01035385,
      "balance_loss_clip": 1.05426407,
      "balance_loss_mlp": 1.01998281,
      "epoch": 0.30795130016533895,
      "flos": 27705046245120.0,
      "grad_norm": 1.437674598612212,
      "language_loss": 0.82692528,
      "learning_rate": 3.242842843433319e-06,
      "loss": 0.84832978,
      "num_input_tokens_seen": 109944265,
      "step": 5122,
      "time_per_iteration": 2.5651676654815674
    },
    {
      "auxiliary_loss_clip": 0.01046516,
      "auxiliary_loss_mlp": 0.01004003,
      "balance_loss_clip": 1.02295542,
      "balance_loss_mlp": 1.00183296,
      "epoch": 0.3080114234180069,
      "flos": 69058699591680.0,
      "grad_norm": 0.7403419678183805,
      "language_loss": 0.58598638,
      "learning_rate": 3.242537685798143e-06,
      "loss": 0.60649157,
      "num_input_tokens_seen": 110014160,
      "step": 5123,
      "time_per_iteration": 3.26607346534729
    },
    {
      "auxiliary_loss_clip": 0.01131117,
      "auxiliary_loss_mlp": 0.00791954,
      "balance_loss_clip": 1.04846776,
      "balance_loss_mlp": 1.01004481,
      "epoch": 0.3080715466706749,
      "flos": 24060221377920.0,
      "grad_norm": 2.131265699194189,
      "language_loss": 0.83690172,
      "learning_rate": 3.242232481045813e-06,
      "loss": 0.85613251,
      "num_input_tokens_seen": 110034865,
      "step": 5124,
      "time_per_iteration": 2.5511813163757324
    },
    {
      "auxiliary_loss_clip": 0.01139511,
      "auxiliary_loss_mlp": 0.010389,
      "balance_loss_clip": 1.04996657,
      "balance_loss_mlp": 1.0232718,
      "epoch": 0.30813166992334284,
      "flos": 25848680480640.0,
      "grad_norm": 2.002212414980544,
      "language_loss": 0.79170126,
      "learning_rate": 3.2419272291879035e-06,
      "loss": 0.81348538,
      "num_input_tokens_seen": 110052930,
      "step": 5125,
      "time_per_iteration": 2.5060949325561523
    },
    {
      "auxiliary_loss_clip": 0.01124322,
      "auxiliary_loss_mlp": 0.01042048,
      "balance_loss_clip": 1.04877329,
      "balance_loss_mlp": 1.02411926,
      "epoch": 0.3081917931760108,
      "flos": 20449619193600.0,
      "grad_norm": 1.8925401924383416,
      "language_loss": 0.64502287,
      "learning_rate": 3.241621930235989e-06,
      "loss": 0.66668665,
      "num_input_tokens_seen": 110071765,
      "step": 5126,
      "time_per_iteration": 2.465609073638916
    },
    {
      "auxiliary_loss_clip": 0.0108997,
      "auxiliary_loss_mlp": 0.01039946,
      "balance_loss_clip": 1.04512811,
      "balance_loss_mlp": 1.02438879,
      "epoch": 0.3082519164286788,
      "flos": 22166257052160.0,
      "grad_norm": 1.5359451701440519,
      "language_loss": 0.86318147,
      "learning_rate": 3.241316584201646e-06,
      "loss": 0.8844806,
      "num_input_tokens_seen": 110092660,
      "step": 5127,
      "time_per_iteration": 2.6120364665985107
    },
    {
      "auxiliary_loss_clip": 0.01088781,
      "auxiliary_loss_mlp": 0.0104144,
      "balance_loss_clip": 1.04430413,
      "balance_loss_mlp": 1.02529883,
      "epoch": 0.30831203968134674,
      "flos": 28913404700160.0,
      "grad_norm": 1.751514660264318,
      "language_loss": 0.69138956,
      "learning_rate": 3.2410111910964538e-06,
      "loss": 0.71269178,
      "num_input_tokens_seen": 110114960,
      "step": 5128,
      "time_per_iteration": 2.6283555030822754
    },
    {
      "auxiliary_loss_clip": 0.01124886,
      "auxiliary_loss_mlp": 0.00790957,
      "balance_loss_clip": 1.04735279,
      "balance_loss_mlp": 1.00913072,
      "epoch": 0.3083721629340147,
      "flos": 25667726739840.0,
      "grad_norm": 1.9199537416804526,
      "language_loss": 0.71370411,
      "learning_rate": 3.240705750931993e-06,
      "loss": 0.73286253,
      "num_input_tokens_seen": 110135750,
      "step": 5129,
      "time_per_iteration": 2.535334825515747
    },
    {
      "auxiliary_loss_clip": 0.01021848,
      "auxiliary_loss_mlp": 0.01004726,
      "balance_loss_clip": 1.01862228,
      "balance_loss_mlp": 1.00237751,
      "epoch": 0.3084322861866827,
      "flos": 68212679581440.0,
      "grad_norm": 0.8308423309343258,
      "language_loss": 0.59223545,
      "learning_rate": 3.240400263719846e-06,
      "loss": 0.6125012,
      "num_input_tokens_seen": 110189480,
      "step": 5130,
      "time_per_iteration": 4.4779157638549805
    },
    {
      "auxiliary_loss_clip": 0.01114869,
      "auxiliary_loss_mlp": 0.01038897,
      "balance_loss_clip": 1.0476954,
      "balance_loss_mlp": 1.0217073,
      "epoch": 0.3084924094393507,
      "flos": 20296495514880.0,
      "grad_norm": 2.2726027329164356,
      "language_loss": 0.72954559,
      "learning_rate": 3.2400947294715957e-06,
      "loss": 0.75108331,
      "num_input_tokens_seen": 110206445,
      "step": 5131,
      "time_per_iteration": 2.5157718658447266
    },
    {
      "auxiliary_loss_clip": 0.01097355,
      "auxiliary_loss_mlp": 0.01036661,
      "balance_loss_clip": 1.04471791,
      "balance_loss_mlp": 1.02164114,
      "epoch": 0.30855253269201866,
      "flos": 23949831905280.0,
      "grad_norm": 1.4976716853155425,
      "language_loss": 0.70680654,
      "learning_rate": 3.2397891481988303e-06,
      "loss": 0.72814673,
      "num_input_tokens_seen": 110226845,
      "step": 5132,
      "time_per_iteration": 2.578185558319092
    },
    {
      "auxiliary_loss_clip": 0.01131491,
      "auxiliary_loss_mlp": 0.00791007,
      "balance_loss_clip": 1.04714143,
      "balance_loss_mlp": 1.00972545,
      "epoch": 0.3086126559446866,
      "flos": 19281876042240.0,
      "grad_norm": 1.6878551322865813,
      "language_loss": 0.89759606,
      "learning_rate": 3.239483519913136e-06,
      "loss": 0.916821,
      "num_input_tokens_seen": 110244095,
      "step": 5133,
      "time_per_iteration": 2.4394192695617676
    },
    {
      "auxiliary_loss_clip": 0.01116362,
      "auxiliary_loss_mlp": 0.01045,
      "balance_loss_clip": 1.0451231,
      "balance_loss_mlp": 1.02867997,
      "epoch": 0.3086727791973546,
      "flos": 33760770019200.0,
      "grad_norm": 2.0616612895315996,
      "language_loss": 0.67234379,
      "learning_rate": 3.239177844626102e-06,
      "loss": 0.69395745,
      "num_input_tokens_seen": 110264240,
      "step": 5134,
      "time_per_iteration": 3.992511034011841
    },
    {
      "auxiliary_loss_clip": 0.01122975,
      "auxiliary_loss_mlp": 0.01043144,
      "balance_loss_clip": 1.04870009,
      "balance_loss_mlp": 1.02612114,
      "epoch": 0.30873290245002255,
      "flos": 16034151006720.0,
      "grad_norm": 1.8803038549908033,
      "language_loss": 0.82805687,
      "learning_rate": 3.2388721223493197e-06,
      "loss": 0.84971809,
      "num_input_tokens_seen": 110282450,
      "step": 5135,
      "time_per_iteration": 2.466834306716919
    },
    {
      "auxiliary_loss_clip": 0.01018859,
      "auxiliary_loss_mlp": 0.01008721,
      "balance_loss_clip": 1.01553559,
      "balance_loss_mlp": 1.00664663,
      "epoch": 0.3087930257026905,
      "flos": 65048304055680.0,
      "grad_norm": 0.7027327457366329,
      "language_loss": 0.553177,
      "learning_rate": 3.2385663530943824e-06,
      "loss": 0.57345283,
      "num_input_tokens_seen": 110343715,
      "step": 5136,
      "time_per_iteration": 3.186471939086914
    },
    {
      "auxiliary_loss_clip": 0.01111712,
      "auxiliary_loss_mlp": 0.00791801,
      "balance_loss_clip": 1.04443812,
      "balance_loss_mlp": 1.01110983,
      "epoch": 0.3088531489553585,
      "flos": 74738829824640.0,
      "grad_norm": 3.6141860830208703,
      "language_loss": 0.76095974,
      "learning_rate": 3.2382605368728852e-06,
      "loss": 0.7799949,
      "num_input_tokens_seen": 110368430,
      "step": 5137,
      "time_per_iteration": 5.66956901550293
    },
    {
      "auxiliary_loss_clip": 0.01096748,
      "auxiliary_loss_mlp": 0.01035591,
      "balance_loss_clip": 1.04324245,
      "balance_loss_mlp": 1.0208863,
      "epoch": 0.30891327220802645,
      "flos": 21142300043520.0,
      "grad_norm": 1.611292880470574,
      "language_loss": 0.79480058,
      "learning_rate": 3.237954673696424e-06,
      "loss": 0.81612396,
      "num_input_tokens_seen": 110386735,
      "step": 5138,
      "time_per_iteration": 2.544736623764038
    },
    {
      "auxiliary_loss_clip": 0.01080991,
      "auxiliary_loss_mlp": 0.01046534,
      "balance_loss_clip": 1.04888475,
      "balance_loss_mlp": 1.02918911,
      "epoch": 0.3089733954606944,
      "flos": 25664494515840.0,
      "grad_norm": 1.4685043975493102,
      "language_loss": 0.81352645,
      "learning_rate": 3.2376487635765983e-06,
      "loss": 0.83480167,
      "num_input_tokens_seen": 110406820,
      "step": 5139,
      "time_per_iteration": 2.615164279937744
    },
    {
      "auxiliary_loss_clip": 0.01124264,
      "auxiliary_loss_mlp": 0.01043013,
      "balance_loss_clip": 1.04623747,
      "balance_loss_mlp": 1.02514327,
      "epoch": 0.3090335187133624,
      "flos": 19427350124160.0,
      "grad_norm": 2.026754591581606,
      "language_loss": 0.77583021,
      "learning_rate": 3.2373428065250067e-06,
      "loss": 0.79750299,
      "num_input_tokens_seen": 110424225,
      "step": 5140,
      "time_per_iteration": 2.475653886795044
    },
    {
      "auxiliary_loss_clip": 0.01095993,
      "auxiliary_loss_mlp": 0.01048035,
      "balance_loss_clip": 1.04417276,
      "balance_loss_mlp": 1.03270459,
      "epoch": 0.30909364196603034,
      "flos": 20011329440640.0,
      "grad_norm": 2.5521180203314375,
      "language_loss": 0.78292257,
      "learning_rate": 3.237036802553252e-06,
      "loss": 0.80436289,
      "num_input_tokens_seen": 110443310,
      "step": 5141,
      "time_per_iteration": 2.537550687789917
    },
    {
      "auxiliary_loss_clip": 0.01121276,
      "auxiliary_loss_mlp": 0.01043065,
      "balance_loss_clip": 1.04823518,
      "balance_loss_mlp": 1.02662611,
      "epoch": 0.3091537652186983,
      "flos": 19677575243520.0,
      "grad_norm": 2.189522433261539,
      "language_loss": 0.87047809,
      "learning_rate": 3.2367307516729377e-06,
      "loss": 0.89212149,
      "num_input_tokens_seen": 110460215,
      "step": 5142,
      "time_per_iteration": 2.516071319580078
    },
    {
      "auxiliary_loss_clip": 0.01123133,
      "auxiliary_loss_mlp": 0.01043572,
      "balance_loss_clip": 1.04517365,
      "balance_loss_mlp": 1.02788424,
      "epoch": 0.3092138884713663,
      "flos": 17020042577280.0,
      "grad_norm": 1.828304621497409,
      "language_loss": 0.791161,
      "learning_rate": 3.23642465389567e-06,
      "loss": 0.81282806,
      "num_input_tokens_seen": 110479385,
      "step": 5143,
      "time_per_iteration": 2.470850706100464
    },
    {
      "auxiliary_loss_clip": 0.01103064,
      "auxiliary_loss_mlp": 0.01038895,
      "balance_loss_clip": 1.04684007,
      "balance_loss_mlp": 1.02259874,
      "epoch": 0.3092740117240343,
      "flos": 25009986844800.0,
      "grad_norm": 1.6279856290545514,
      "language_loss": 0.71849293,
      "learning_rate": 3.236118509233055e-06,
      "loss": 0.73991251,
      "num_input_tokens_seen": 110499885,
      "step": 5144,
      "time_per_iteration": 2.5860989093780518
    },
    {
      "auxiliary_loss_clip": 0.01125529,
      "auxiliary_loss_mlp": 0.01042406,
      "balance_loss_clip": 1.04461944,
      "balance_loss_mlp": 1.02647996,
      "epoch": 0.30933413497670226,
      "flos": 25590410714880.0,
      "grad_norm": 1.6875416683525042,
      "language_loss": 0.7436403,
      "learning_rate": 3.235812317696702e-06,
      "loss": 0.76531965,
      "num_input_tokens_seen": 110519690,
      "step": 5145,
      "time_per_iteration": 2.518439531326294
    },
    {
      "auxiliary_loss_clip": 0.0111389,
      "auxiliary_loss_mlp": 0.01049004,
      "balance_loss_clip": 1.04682899,
      "balance_loss_mlp": 1.03260112,
      "epoch": 0.3093942582293702,
      "flos": 24389665943040.0,
      "grad_norm": 4.610241339881009,
      "language_loss": 0.76425201,
      "learning_rate": 3.2355060792982224e-06,
      "loss": 0.78588092,
      "num_input_tokens_seen": 110540520,
      "step": 5146,
      "time_per_iteration": 2.547069549560547
    },
    {
      "auxiliary_loss_clip": 0.0110674,
      "auxiliary_loss_mlp": 0.01038721,
      "balance_loss_clip": 1.04276359,
      "balance_loss_mlp": 1.02328908,
      "epoch": 0.3094543814820382,
      "flos": 19646441130240.0,
      "grad_norm": 1.7872494046337706,
      "language_loss": 0.66106921,
      "learning_rate": 3.2351997940492286e-06,
      "loss": 0.68252385,
      "num_input_tokens_seen": 110557950,
      "step": 5147,
      "time_per_iteration": 2.4985668659210205
    },
    {
      "auxiliary_loss_clip": 0.01128481,
      "auxiliary_loss_mlp": 0.01041074,
      "balance_loss_clip": 1.04848289,
      "balance_loss_mlp": 1.02642322,
      "epoch": 0.30951450473470615,
      "flos": 25663812157440.0,
      "grad_norm": 2.0710085825006144,
      "language_loss": 0.74743724,
      "learning_rate": 3.2348934619613346e-06,
      "loss": 0.76913279,
      "num_input_tokens_seen": 110578215,
      "step": 5148,
      "time_per_iteration": 2.5143988132476807
    },
    {
      "auxiliary_loss_clip": 0.01129538,
      "auxiliary_loss_mlp": 0.01046454,
      "balance_loss_clip": 1.04654431,
      "balance_loss_mlp": 1.03009832,
      "epoch": 0.3095746279873741,
      "flos": 12020415505920.0,
      "grad_norm": 2.6134496639525837,
      "language_loss": 0.72102737,
      "learning_rate": 3.2345870830461567e-06,
      "loss": 0.74278736,
      "num_input_tokens_seen": 110592990,
      "step": 5149,
      "time_per_iteration": 2.420346736907959
    },
    {
      "auxiliary_loss_clip": 0.01091789,
      "auxiliary_loss_mlp": 0.01039949,
      "balance_loss_clip": 1.04417312,
      "balance_loss_mlp": 1.02280688,
      "epoch": 0.3096347512400421,
      "flos": 23623044946560.0,
      "grad_norm": 1.8349776678294214,
      "language_loss": 0.84674132,
      "learning_rate": 3.2342806573153132e-06,
      "loss": 0.86805868,
      "num_input_tokens_seen": 110612130,
      "step": 5150,
      "time_per_iteration": 2.5832021236419678
    },
    {
      "auxiliary_loss_clip": 0.0108541,
      "auxiliary_loss_mlp": 0.01045762,
      "balance_loss_clip": 1.04637694,
      "balance_loss_mlp": 1.02833378,
      "epoch": 0.30969487449271005,
      "flos": 22529313768960.0,
      "grad_norm": 1.933581818493374,
      "language_loss": 0.78832352,
      "learning_rate": 3.233974184780424e-06,
      "loss": 0.80963528,
      "num_input_tokens_seen": 110632045,
      "step": 5151,
      "time_per_iteration": 2.5732645988464355
    },
    {
      "auxiliary_loss_clip": 0.01126874,
      "auxiliary_loss_mlp": 0.01042425,
      "balance_loss_clip": 1.04638541,
      "balance_loss_mlp": 1.02490151,
      "epoch": 0.309754997745378,
      "flos": 15267925059840.0,
      "grad_norm": 3.2489503518730807,
      "language_loss": 0.66898882,
      "learning_rate": 3.2336676654531084e-06,
      "loss": 0.69068182,
      "num_input_tokens_seen": 110649340,
      "step": 5152,
      "time_per_iteration": 2.4647090435028076
    },
    {
      "auxiliary_loss_clip": 0.01078987,
      "auxiliary_loss_mlp": 0.0104301,
      "balance_loss_clip": 1.04586983,
      "balance_loss_mlp": 1.02673817,
      "epoch": 0.309815120998046,
      "flos": 26979291947520.0,
      "grad_norm": 2.0314345938037883,
      "language_loss": 0.82136017,
      "learning_rate": 3.2333610993449926e-06,
      "loss": 0.8425802,
      "num_input_tokens_seen": 110668450,
      "step": 5153,
      "time_per_iteration": 2.63582706451416
    },
    {
      "auxiliary_loss_clip": 0.01110126,
      "auxiliary_loss_mlp": 0.00795524,
      "balance_loss_clip": 1.0471034,
      "balance_loss_mlp": 1.01592326,
      "epoch": 0.30987524425071394,
      "flos": 21143161969920.0,
      "grad_norm": 1.9398322491677165,
      "language_loss": 0.74210793,
      "learning_rate": 3.2330544864676997e-06,
      "loss": 0.76116443,
      "num_input_tokens_seen": 110689410,
      "step": 5154,
      "time_per_iteration": 2.5508196353912354
    },
    {
      "auxiliary_loss_clip": 0.0112657,
      "auxiliary_loss_mlp": 0.01036252,
      "balance_loss_clip": 1.04715347,
      "balance_loss_mlp": 1.01976526,
      "epoch": 0.3099353675033819,
      "flos": 15268284195840.0,
      "grad_norm": 2.909941683543604,
      "language_loss": 0.76229084,
      "learning_rate": 3.232747826832858e-06,
      "loss": 0.78391904,
      "num_input_tokens_seen": 110707350,
      "step": 5155,
      "time_per_iteration": 2.460416793823242
    },
    {
      "auxiliary_loss_clip": 0.01125444,
      "auxiliary_loss_mlp": 0.01039499,
      "balance_loss_clip": 1.0519228,
      "balance_loss_mlp": 1.02288115,
      "epoch": 0.30999549075604993,
      "flos": 15413794191360.0,
      "grad_norm": 2.3342258746411493,
      "language_loss": 0.79106486,
      "learning_rate": 3.232441120452094e-06,
      "loss": 0.81271434,
      "num_input_tokens_seen": 110724910,
      "step": 5156,
      "time_per_iteration": 2.507646083831787
    },
    {
      "auxiliary_loss_clip": 0.01124735,
      "auxiliary_loss_mlp": 0.01044663,
      "balance_loss_clip": 1.04945016,
      "balance_loss_mlp": 1.02774715,
      "epoch": 0.3100556140087179,
      "flos": 23184539712000.0,
      "grad_norm": 2.166944937986495,
      "language_loss": 0.75024378,
      "learning_rate": 3.23213436733704e-06,
      "loss": 0.77193773,
      "num_input_tokens_seen": 110744010,
      "step": 5157,
      "time_per_iteration": 2.5005106925964355
    },
    {
      "auxiliary_loss_clip": 0.01099555,
      "auxiliary_loss_mlp": 0.01041818,
      "balance_loss_clip": 1.04467249,
      "balance_loss_mlp": 1.02616549,
      "epoch": 0.31011573726138586,
      "flos": 25742169676800.0,
      "grad_norm": 1.5696421822926878,
      "language_loss": 0.69579256,
      "learning_rate": 3.231827567499327e-06,
      "loss": 0.71720624,
      "num_input_tokens_seen": 110765835,
      "step": 5158,
      "time_per_iteration": 2.599217414855957
    },
    {
      "auxiliary_loss_clip": 0.01088321,
      "auxiliary_loss_mlp": 0.01037809,
      "balance_loss_clip": 1.04163313,
      "balance_loss_mlp": 1.02296734,
      "epoch": 0.3101758605140538,
      "flos": 20011329440640.0,
      "grad_norm": 2.4712486636239137,
      "language_loss": 0.84994459,
      "learning_rate": 3.2315207209505896e-06,
      "loss": 0.87120587,
      "num_input_tokens_seen": 110784655,
      "step": 5159,
      "time_per_iteration": 2.5482935905456543
    },
    {
      "auxiliary_loss_clip": 0.01112414,
      "auxiliary_loss_mlp": 0.01043839,
      "balance_loss_clip": 1.04609013,
      "balance_loss_mlp": 1.02757919,
      "epoch": 0.3102359837667218,
      "flos": 19135683688320.0,
      "grad_norm": 1.7317939008168393,
      "language_loss": 0.85188442,
      "learning_rate": 3.231213827702462e-06,
      "loss": 0.87344694,
      "num_input_tokens_seen": 110802545,
      "step": 5160,
      "time_per_iteration": 2.5181961059570312
    },
    {
      "auxiliary_loss_clip": 0.01123575,
      "auxiliary_loss_mlp": 0.01038954,
      "balance_loss_clip": 1.04579222,
      "balance_loss_mlp": 1.02354002,
      "epoch": 0.31029610701938976,
      "flos": 22265405568000.0,
      "grad_norm": 1.8977180283170363,
      "language_loss": 0.7595709,
      "learning_rate": 3.230906887766584e-06,
      "loss": 0.78119624,
      "num_input_tokens_seen": 110820265,
      "step": 5161,
      "time_per_iteration": 2.4689128398895264
    },
    {
      "auxiliary_loss_clip": 0.01126974,
      "auxiliary_loss_mlp": 0.01040086,
      "balance_loss_clip": 1.0475775,
      "balance_loss_mlp": 1.02377796,
      "epoch": 0.3103562302720577,
      "flos": 20805349536000.0,
      "grad_norm": 2.0087336665127205,
      "language_loss": 0.81976295,
      "learning_rate": 3.2305999011545924e-06,
      "loss": 0.84143353,
      "num_input_tokens_seen": 110836195,
      "step": 5162,
      "time_per_iteration": 2.501729965209961
    },
    {
      "auxiliary_loss_clip": 0.01122791,
      "auxiliary_loss_mlp": 0.01036481,
      "balance_loss_clip": 1.04750407,
      "balance_loss_mlp": 1.02222919,
      "epoch": 0.3104163535247257,
      "flos": 22344158136960.0,
      "grad_norm": 1.7623342700956923,
      "language_loss": 0.83124697,
      "learning_rate": 3.2302928678781295e-06,
      "loss": 0.85283971,
      "num_input_tokens_seen": 110856420,
      "step": 5163,
      "time_per_iteration": 2.493786334991455
    },
    {
      "auxiliary_loss_clip": 0.01139342,
      "auxiliary_loss_mlp": 0.01042755,
      "balance_loss_clip": 1.04878068,
      "balance_loss_mlp": 1.02688789,
      "epoch": 0.31047647677739365,
      "flos": 21689363157120.0,
      "grad_norm": 1.8421898312037124,
      "language_loss": 0.75828815,
      "learning_rate": 3.2299857879488376e-06,
      "loss": 0.78010905,
      "num_input_tokens_seen": 110876650,
      "step": 5164,
      "time_per_iteration": 2.483548879623413
    },
    {
      "auxiliary_loss_clip": 0.01101012,
      "auxiliary_loss_mlp": 0.01040371,
      "balance_loss_clip": 1.0478344,
      "balance_loss_mlp": 1.02400351,
      "epoch": 0.3105366000300616,
      "flos": 18917275040640.0,
      "grad_norm": 1.8030089877008284,
      "language_loss": 0.74250633,
      "learning_rate": 3.2296786613783626e-06,
      "loss": 0.76392019,
      "num_input_tokens_seen": 110894445,
      "step": 5165,
      "time_per_iteration": 2.544454574584961
    },
    {
      "auxiliary_loss_clip": 0.01100525,
      "auxiliary_loss_mlp": 0.01046593,
      "balance_loss_clip": 1.05296779,
      "balance_loss_mlp": 1.02886653,
      "epoch": 0.3105967232827296,
      "flos": 18260397072000.0,
      "grad_norm": 1.4650576728571867,
      "language_loss": 0.76115322,
      "learning_rate": 3.229371488178348e-06,
      "loss": 0.78262436,
      "num_input_tokens_seen": 110912855,
      "step": 5166,
      "time_per_iteration": 2.5684804916381836
    },
    {
      "auxiliary_loss_clip": 0.01117025,
      "auxiliary_loss_mlp": 0.01041485,
      "balance_loss_clip": 1.04738986,
      "balance_loss_mlp": 1.02471232,
      "epoch": 0.31065684653539755,
      "flos": 17672144037120.0,
      "grad_norm": 3.122357064853797,
      "language_loss": 0.73346955,
      "learning_rate": 3.229064268360444e-06,
      "loss": 0.75505465,
      "num_input_tokens_seen": 110928025,
      "step": 5167,
      "time_per_iteration": 2.47337007522583
    },
    {
      "auxiliary_loss_clip": 0.01028684,
      "auxiliary_loss_mlp": 0.01057931,
      "balance_loss_clip": 1.05697787,
      "balance_loss_mlp": 1.05545127,
      "epoch": 0.3107169697880655,
      "flos": 68531996511360.0,
      "grad_norm": 0.7205830673479034,
      "language_loss": 0.52990001,
      "learning_rate": 3.2287570019362997e-06,
      "loss": 0.55076617,
      "num_input_tokens_seen": 110992215,
      "step": 5168,
      "time_per_iteration": 4.648841619491577
    },
    {
      "auxiliary_loss_clip": 0.01128057,
      "auxiliary_loss_mlp": 0.01041426,
      "balance_loss_clip": 1.04900575,
      "balance_loss_mlp": 1.02473664,
      "epoch": 0.3107770930407335,
      "flos": 13188733274880.0,
      "grad_norm": 2.611895628033281,
      "language_loss": 0.78243387,
      "learning_rate": 3.2284496889175668e-06,
      "loss": 0.80412877,
      "num_input_tokens_seen": 111010400,
      "step": 5169,
      "time_per_iteration": 2.602570056915283
    },
    {
      "auxiliary_loss_clip": 0.01115759,
      "auxiliary_loss_mlp": 0.01040141,
      "balance_loss_clip": 1.04538488,
      "balance_loss_mlp": 1.02415502,
      "epoch": 0.3108372162934015,
      "flos": 31580849520000.0,
      "grad_norm": 1.484947480576105,
      "language_loss": 0.63738811,
      "learning_rate": 3.2281423293158986e-06,
      "loss": 0.65894711,
      "num_input_tokens_seen": 111033960,
      "step": 5170,
      "time_per_iteration": 2.627995491027832
    },
    {
      "auxiliary_loss_clip": 0.01102077,
      "auxiliary_loss_mlp": 0.00794083,
      "balance_loss_clip": 1.05027497,
      "balance_loss_mlp": 1.01183414,
      "epoch": 0.31089733954606946,
      "flos": 28729829266560.0,
      "grad_norm": 2.394576409658571,
      "language_loss": 0.78137171,
      "learning_rate": 3.22783492314295e-06,
      "loss": 0.80033332,
      "num_input_tokens_seen": 111053265,
      "step": 5171,
      "time_per_iteration": 2.617152690887451
    },
    {
      "auxiliary_loss_clip": 0.01100952,
      "auxiliary_loss_mlp": 0.01054239,
      "balance_loss_clip": 1.05047846,
      "balance_loss_mlp": 1.03733563,
      "epoch": 0.3109574627987374,
      "flos": 19683249592320.0,
      "grad_norm": 1.8125294233796099,
      "language_loss": 0.83897364,
      "learning_rate": 3.2275274704103785e-06,
      "loss": 0.86052561,
      "num_input_tokens_seen": 111071130,
      "step": 5172,
      "time_per_iteration": 3.9540529251098633
    },
    {
      "auxiliary_loss_clip": 0.01092156,
      "auxiliary_loss_mlp": 0.01042326,
      "balance_loss_clip": 1.05263233,
      "balance_loss_mlp": 1.02541041,
      "epoch": 0.3110175860514054,
      "flos": 14683981656960.0,
      "grad_norm": 2.063805517736345,
      "language_loss": 0.8437928,
      "learning_rate": 3.227219971129842e-06,
      "loss": 0.86513764,
      "num_input_tokens_seen": 111089560,
      "step": 5173,
      "time_per_iteration": 2.568983316421509
    },
    {
      "auxiliary_loss_clip": 0.0113316,
      "auxiliary_loss_mlp": 0.01035803,
      "balance_loss_clip": 1.04829097,
      "balance_loss_mlp": 1.02085984,
      "epoch": 0.31107770930407336,
      "flos": 25739655724800.0,
      "grad_norm": 1.8270525156300017,
      "language_loss": 0.83645487,
      "learning_rate": 3.226912425313001e-06,
      "loss": 0.85814452,
      "num_input_tokens_seen": 111109960,
      "step": 5174,
      "time_per_iteration": 2.517989158630371
    },
    {
      "auxiliary_loss_clip": 0.01115388,
      "auxiliary_loss_mlp": 0.01043589,
      "balance_loss_clip": 1.04775548,
      "balance_loss_mlp": 1.02777028,
      "epoch": 0.3111378325567413,
      "flos": 19208259118080.0,
      "grad_norm": 2.0653210626358343,
      "language_loss": 0.85410631,
      "learning_rate": 3.2266048329715183e-06,
      "loss": 0.87569606,
      "num_input_tokens_seen": 111127960,
      "step": 5175,
      "time_per_iteration": 3.965095043182373
    },
    {
      "auxiliary_loss_clip": 0.01080534,
      "auxiliary_loss_mlp": 0.01051819,
      "balance_loss_clip": 1.04671264,
      "balance_loss_mlp": 1.03428352,
      "epoch": 0.3111979558094093,
      "flos": 23696374561920.0,
      "grad_norm": 1.9875723257700844,
      "language_loss": 0.8326515,
      "learning_rate": 3.2262971941170575e-06,
      "loss": 0.85397494,
      "num_input_tokens_seen": 111146730,
      "step": 5176,
      "time_per_iteration": 4.029592514038086
    },
    {
      "auxiliary_loss_clip": 0.01117786,
      "auxiliary_loss_mlp": 0.0104096,
      "balance_loss_clip": 1.04273641,
      "balance_loss_mlp": 1.0238061,
      "epoch": 0.31125807906207725,
      "flos": 21033023892480.0,
      "grad_norm": 1.7975039537812783,
      "language_loss": 0.8059147,
      "learning_rate": 3.2259895087612837e-06,
      "loss": 0.82750213,
      "num_input_tokens_seen": 111166295,
      "step": 5177,
      "time_per_iteration": 2.499922037124634
    },
    {
      "auxiliary_loss_clip": 0.01124084,
      "auxiliary_loss_mlp": 0.00794124,
      "balance_loss_clip": 1.04842472,
      "balance_loss_mlp": 1.0129987,
      "epoch": 0.3113182023147452,
      "flos": 23076628277760.0,
      "grad_norm": 1.603204742098221,
      "language_loss": 0.80778694,
      "learning_rate": 3.2256817769158657e-06,
      "loss": 0.82696897,
      "num_input_tokens_seen": 111185665,
      "step": 5178,
      "time_per_iteration": 2.486328125
    },
    {
      "auxiliary_loss_clip": 0.01111816,
      "auxiliary_loss_mlp": 0.01045516,
      "balance_loss_clip": 1.04913521,
      "balance_loss_mlp": 1.0301621,
      "epoch": 0.3113783255674132,
      "flos": 11838994888320.0,
      "grad_norm": 1.7696422578055893,
      "language_loss": 0.8110491,
      "learning_rate": 3.225373998592471e-06,
      "loss": 0.83262247,
      "num_input_tokens_seen": 111201615,
      "step": 5179,
      "time_per_iteration": 2.519951820373535
    },
    {
      "auxiliary_loss_clip": 0.01102373,
      "auxiliary_loss_mlp": 0.01048377,
      "balance_loss_clip": 1.04727674,
      "balance_loss_mlp": 1.03215837,
      "epoch": 0.31143844882008115,
      "flos": 16289547684480.0,
      "grad_norm": 1.7544835344503964,
      "language_loss": 0.7843743,
      "learning_rate": 3.2250661738027715e-06,
      "loss": 0.80588174,
      "num_input_tokens_seen": 111220515,
      "step": 5180,
      "time_per_iteration": 2.5336902141571045
    },
    {
      "auxiliary_loss_clip": 0.01100611,
      "auxiliary_loss_mlp": 0.01038562,
      "balance_loss_clip": 1.0486815,
      "balance_loss_mlp": 1.02274334,
      "epoch": 0.3114985720727491,
      "flos": 23217792727680.0,
      "grad_norm": 1.701483802239429,
      "language_loss": 0.82973057,
      "learning_rate": 3.22475830255844e-06,
      "loss": 0.85112232,
      "num_input_tokens_seen": 111240395,
      "step": 5181,
      "time_per_iteration": 2.6240293979644775
    },
    {
      "auxiliary_loss_clip": 0.01104904,
      "auxiliary_loss_mlp": 0.01046827,
      "balance_loss_clip": 1.04833639,
      "balance_loss_mlp": 1.03190184,
      "epoch": 0.3115586953254171,
      "flos": 30044626698240.0,
      "grad_norm": 1.5838323050963508,
      "language_loss": 0.73865235,
      "learning_rate": 3.2244503848711516e-06,
      "loss": 0.76016974,
      "num_input_tokens_seen": 111261100,
      "step": 5182,
      "time_per_iteration": 2.6070053577423096
    },
    {
      "auxiliary_loss_clip": 0.01091651,
      "auxiliary_loss_mlp": 0.00794192,
      "balance_loss_clip": 1.05143619,
      "balance_loss_mlp": 1.01375818,
      "epoch": 0.3116188185780851,
      "flos": 25666326109440.0,
      "grad_norm": 2.0452499501317933,
      "language_loss": 0.70419228,
      "learning_rate": 3.2241424207525815e-06,
      "loss": 0.72305071,
      "num_input_tokens_seen": 111281320,
      "step": 5183,
      "time_per_iteration": 2.6470866203308105
    },
    {
      "auxiliary_loss_clip": 0.01032647,
      "auxiliary_loss_mlp": 0.01027089,
      "balance_loss_clip": 1.03655267,
      "balance_loss_mlp": 1.0250144,
      "epoch": 0.31167894183075306,
      "flos": 69510058917120.0,
      "grad_norm": 0.948472405818638,
      "language_loss": 0.59616172,
      "learning_rate": 3.223834410214408e-06,
      "loss": 0.61675906,
      "num_input_tokens_seen": 111341405,
      "step": 5184,
      "time_per_iteration": 3.1720921993255615
    },
    {
      "auxiliary_loss_clip": 0.01110505,
      "auxiliary_loss_mlp": 0.01047951,
      "balance_loss_clip": 1.04540801,
      "balance_loss_mlp": 1.03231096,
      "epoch": 0.31173906508342103,
      "flos": 14939845211520.0,
      "grad_norm": 2.221267139509657,
      "language_loss": 0.7008751,
      "learning_rate": 3.223526353268311e-06,
      "loss": 0.72245967,
      "num_input_tokens_seen": 111358975,
      "step": 5185,
      "time_per_iteration": 2.4985976219177246
    },
    {
      "auxiliary_loss_clip": 0.01117424,
      "auxiliary_loss_mlp": 0.01047401,
      "balance_loss_clip": 1.05180252,
      "balance_loss_mlp": 1.03086674,
      "epoch": 0.311799188336089,
      "flos": 16176033728640.0,
      "grad_norm": 2.2806412933373763,
      "language_loss": 0.64010477,
      "learning_rate": 3.2232182499259725e-06,
      "loss": 0.661753,
      "num_input_tokens_seen": 111375845,
      "step": 5186,
      "time_per_iteration": 2.55678129196167
    },
    {
      "auxiliary_loss_clip": 0.0111856,
      "auxiliary_loss_mlp": 0.01046602,
      "balance_loss_clip": 1.04853773,
      "balance_loss_mlp": 1.02965021,
      "epoch": 0.31185931158875696,
      "flos": 25009627708800.0,
      "grad_norm": 2.7785771110718502,
      "language_loss": 0.87207204,
      "learning_rate": 3.2229101001990747e-06,
      "loss": 0.89372361,
      "num_input_tokens_seen": 111394150,
      "step": 5187,
      "time_per_iteration": 2.5398917198181152
    },
    {
      "auxiliary_loss_clip": 0.0113274,
      "auxiliary_loss_mlp": 0.00793539,
      "balance_loss_clip": 1.04485118,
      "balance_loss_mlp": 1.01490855,
      "epoch": 0.3119194348414249,
      "flos": 37232901273600.0,
      "grad_norm": 1.6758342796593326,
      "language_loss": 0.63002717,
      "learning_rate": 3.2226019040993036e-06,
      "loss": 0.64928997,
      "num_input_tokens_seen": 111418355,
      "step": 5188,
      "time_per_iteration": 2.626851797103882
    },
    {
      "auxiliary_loss_clip": 0.0110431,
      "auxiliary_loss_mlp": 0.01038539,
      "balance_loss_clip": 1.05511713,
      "balance_loss_mlp": 1.02296984,
      "epoch": 0.3119795580940929,
      "flos": 15012779777280.0,
      "grad_norm": 2.5274316357320616,
      "language_loss": 0.8294239,
      "learning_rate": 3.222293661638346e-06,
      "loss": 0.85085237,
      "num_input_tokens_seen": 111435445,
      "step": 5189,
      "time_per_iteration": 2.5121705532073975
    },
    {
      "auxiliary_loss_clip": 0.01034089,
      "auxiliary_loss_mlp": 0.01040511,
      "balance_loss_clip": 1.05023551,
      "balance_loss_mlp": 1.02367854,
      "epoch": 0.31203968134676086,
      "flos": 15998168557440.0,
      "grad_norm": 1.7172832206746698,
      "language_loss": 0.79305208,
      "learning_rate": 3.22198537282789e-06,
      "loss": 0.81379807,
      "num_input_tokens_seen": 111453430,
      "step": 5190,
      "time_per_iteration": 2.8487329483032227
    },
    {
      "auxiliary_loss_clip": 0.01083361,
      "auxiliary_loss_mlp": 0.0105712,
      "balance_loss_clip": 1.04773653,
      "balance_loss_mlp": 1.03911948,
      "epoch": 0.3120998045994288,
      "flos": 23837359443840.0,
      "grad_norm": 1.54532643164188,
      "language_loss": 0.75090659,
      "learning_rate": 3.2216770376796262e-06,
      "loss": 0.77231145,
      "num_input_tokens_seen": 111475325,
      "step": 5191,
      "time_per_iteration": 2.8717501163482666
    },
    {
      "auxiliary_loss_clip": 0.01040946,
      "auxiliary_loss_mlp": 0.00950794,
      "balance_loss_clip": 1.01857841,
      "balance_loss_mlp": 1.3100661,
      "epoch": 0.3121599278520968,
      "flos": 69184205712000.0,
      "grad_norm": 0.8479726393477088,
      "language_loss": 0.6389336,
      "learning_rate": 3.221368656205247e-06,
      "loss": 0.65885103,
      "num_input_tokens_seen": 111533960,
      "step": 5192,
      "time_per_iteration": 3.204542875289917
    },
    {
      "auxiliary_loss_clip": 0.01126482,
      "auxiliary_loss_mlp": 0.01039471,
      "balance_loss_clip": 1.04712534,
      "balance_loss_mlp": 1.022686,
      "epoch": 0.31222005110476475,
      "flos": 23806368984960.0,
      "grad_norm": 1.6343880469172782,
      "language_loss": 0.80250365,
      "learning_rate": 3.221060228416446e-06,
      "loss": 0.8241632,
      "num_input_tokens_seen": 111554055,
      "step": 5193,
      "time_per_iteration": 2.5219171047210693
    },
    {
      "auxiliary_loss_clip": 0.01107096,
      "auxiliary_loss_mlp": 0.01049755,
      "balance_loss_clip": 1.04404974,
      "balance_loss_mlp": 1.03155184,
      "epoch": 0.3122801743574327,
      "flos": 25226132935680.0,
      "grad_norm": 2.712552235021476,
      "language_loss": 0.72168684,
      "learning_rate": 3.2207517543249183e-06,
      "loss": 0.74325538,
      "num_input_tokens_seen": 111574305,
      "step": 5194,
      "time_per_iteration": 2.5647292137145996
    },
    {
      "auxiliary_loss_clip": 0.01137359,
      "auxiliary_loss_mlp": 0.0103993,
      "balance_loss_clip": 1.04955387,
      "balance_loss_mlp": 1.02481389,
      "epoch": 0.3123402976101007,
      "flos": 22966490200320.0,
      "grad_norm": 1.4868172919614784,
      "language_loss": 0.76554722,
      "learning_rate": 3.2204432339423616e-06,
      "loss": 0.78732008,
      "num_input_tokens_seen": 111595680,
      "step": 5195,
      "time_per_iteration": 2.486654281616211
    },
    {
      "auxiliary_loss_clip": 0.01134167,
      "auxiliary_loss_mlp": 0.01038979,
      "balance_loss_clip": 1.04447699,
      "balance_loss_mlp": 1.02372026,
      "epoch": 0.3124004208627687,
      "flos": 25192089820800.0,
      "grad_norm": 1.3433196674166674,
      "language_loss": 0.77865672,
      "learning_rate": 3.220134667280476e-06,
      "loss": 0.80038822,
      "num_input_tokens_seen": 111618135,
      "step": 5196,
      "time_per_iteration": 2.5166382789611816
    },
    {
      "auxiliary_loss_clip": 0.01036089,
      "auxiliary_loss_mlp": 0.00882511,
      "balance_loss_clip": 1.03528142,
      "balance_loss_mlp": 1.19397402,
      "epoch": 0.31246054411543667,
      "flos": 67485165517440.0,
      "grad_norm": 0.7686288548926228,
      "language_loss": 0.54783052,
      "learning_rate": 3.2198260543509613e-06,
      "loss": 0.56701654,
      "num_input_tokens_seen": 111682220,
      "step": 5197,
      "time_per_iteration": 3.1676833629608154
    },
    {
      "auxiliary_loss_clip": 0.01133354,
      "auxiliary_loss_mlp": 0.01039928,
      "balance_loss_clip": 1.04703832,
      "balance_loss_mlp": 1.02497935,
      "epoch": 0.31252066736810463,
      "flos": 17858520731520.0,
      "grad_norm": 1.6689481791373746,
      "language_loss": 0.66449761,
      "learning_rate": 3.21951739516552e-06,
      "loss": 0.68623042,
      "num_input_tokens_seen": 111700815,
      "step": 5198,
      "time_per_iteration": 2.4347949028015137
    },
    {
      "auxiliary_loss_clip": 0.01097468,
      "auxiliary_loss_mlp": 0.01044707,
      "balance_loss_clip": 1.0480839,
      "balance_loss_mlp": 1.02832723,
      "epoch": 0.3125807906207726,
      "flos": 18475034791680.0,
      "grad_norm": 2.0696562408039165,
      "language_loss": 0.69305652,
      "learning_rate": 3.219208689735857e-06,
      "loss": 0.71447825,
      "num_input_tokens_seen": 111718195,
      "step": 5199,
      "time_per_iteration": 2.5272793769836426
    },
    {
      "auxiliary_loss_clip": 0.01125201,
      "auxiliary_loss_mlp": 0.01045065,
      "balance_loss_clip": 1.04653728,
      "balance_loss_mlp": 1.02868581,
      "epoch": 0.31264091387344056,
      "flos": 18946541646720.0,
      "grad_norm": 2.025599793661542,
      "language_loss": 0.78516227,
      "learning_rate": 3.2188999380736785e-06,
      "loss": 0.80686492,
      "num_input_tokens_seen": 111734440,
      "step": 5200,
      "time_per_iteration": 2.4818742275238037
    },
    {
      "auxiliary_loss_clip": 0.011226,
      "auxiliary_loss_mlp": 0.01036195,
      "balance_loss_clip": 1.04734683,
      "balance_loss_mlp": 1.02069747,
      "epoch": 0.3127010371261085,
      "flos": 21468512384640.0,
      "grad_norm": 2.111240648917942,
      "language_loss": 0.83747625,
      "learning_rate": 3.2185911401906917e-06,
      "loss": 0.85906422,
      "num_input_tokens_seen": 111751960,
      "step": 5201,
      "time_per_iteration": 2.5128326416015625
    },
    {
      "auxiliary_loss_clip": 0.01137302,
      "auxiliary_loss_mlp": 0.01042565,
      "balance_loss_clip": 1.05001748,
      "balance_loss_mlp": 1.02639413,
      "epoch": 0.3127611603787765,
      "flos": 15336047203200.0,
      "grad_norm": 2.169256429211964,
      "language_loss": 0.69747806,
      "learning_rate": 3.2182822960986072e-06,
      "loss": 0.71927673,
      "num_input_tokens_seen": 111769585,
      "step": 5202,
      "time_per_iteration": 2.450622797012329
    },
    {
      "auxiliary_loss_clip": 0.01136343,
      "auxiliary_loss_mlp": 0.01035204,
      "balance_loss_clip": 1.0467298,
      "balance_loss_mlp": 1.02138734,
      "epoch": 0.31282128363144446,
      "flos": 17602980399360.0,
      "grad_norm": 1.690486940878734,
      "language_loss": 0.84001487,
      "learning_rate": 3.2179734058091358e-06,
      "loss": 0.86173034,
      "num_input_tokens_seen": 111787880,
      "step": 5203,
      "time_per_iteration": 2.4644951820373535
    },
    {
      "auxiliary_loss_clip": 0.01086346,
      "auxiliary_loss_mlp": 0.01040608,
      "balance_loss_clip": 1.05260599,
      "balance_loss_mlp": 1.02448487,
      "epoch": 0.3128814068841124,
      "flos": 26756753235840.0,
      "grad_norm": 2.0364480852359685,
      "language_loss": 0.60868442,
      "learning_rate": 3.2176644693339913e-06,
      "loss": 0.62995398,
      "num_input_tokens_seen": 111805950,
      "step": 5204,
      "time_per_iteration": 2.636850118637085
    },
    {
      "auxiliary_loss_clip": 0.01102569,
      "auxiliary_loss_mlp": 0.01041293,
      "balance_loss_clip": 1.04868245,
      "balance_loss_mlp": 1.0271486,
      "epoch": 0.3129415301367804,
      "flos": 22272372806400.0,
      "grad_norm": 2.4592486453288656,
      "language_loss": 0.65781641,
      "learning_rate": 3.217355486684887e-06,
      "loss": 0.67925501,
      "num_input_tokens_seen": 111826135,
      "step": 5205,
      "time_per_iteration": 2.537339925765991
    },
    {
      "auxiliary_loss_clip": 0.01126912,
      "auxiliary_loss_mlp": 0.01045894,
      "balance_loss_clip": 1.04904854,
      "balance_loss_mlp": 1.0296216,
      "epoch": 0.31300165338944835,
      "flos": 26464907232000.0,
      "grad_norm": 1.5442540569090208,
      "language_loss": 0.76642656,
      "learning_rate": 3.2170464578735414e-06,
      "loss": 0.7881546,
      "num_input_tokens_seen": 111844700,
      "step": 5206,
      "time_per_iteration": 2.5433506965637207
    },
    {
      "auxiliary_loss_clip": 0.01134291,
      "auxiliary_loss_mlp": 0.01037572,
      "balance_loss_clip": 1.04710698,
      "balance_loss_mlp": 1.02232492,
      "epoch": 0.3130617766421163,
      "flos": 21944652094080.0,
      "grad_norm": 2.0491924909917825,
      "language_loss": 0.83346403,
      "learning_rate": 3.216737382911672e-06,
      "loss": 0.85518265,
      "num_input_tokens_seen": 111861585,
      "step": 5207,
      "time_per_iteration": 4.298353672027588
    },
    {
      "auxiliary_loss_clip": 0.01120945,
      "auxiliary_loss_mlp": 0.01041495,
      "balance_loss_clip": 1.04804385,
      "balance_loss_mlp": 1.02761316,
      "epoch": 0.3131218998947843,
      "flos": 23292774368640.0,
      "grad_norm": 1.785733933756896,
      "language_loss": 0.71241498,
      "learning_rate": 3.216428261810999e-06,
      "loss": 0.73403931,
      "num_input_tokens_seen": 111882950,
      "step": 5208,
      "time_per_iteration": 2.5257740020751953
    },
    {
      "auxiliary_loss_clip": 0.01113951,
      "auxiliary_loss_mlp": 0.01043194,
      "balance_loss_clip": 1.04694748,
      "balance_loss_mlp": 1.02757788,
      "epoch": 0.3131820231474523,
      "flos": 21139642437120.0,
      "grad_norm": 1.9897925635685254,
      "language_loss": 0.74732858,
      "learning_rate": 3.2161190945832445e-06,
      "loss": 0.76889998,
      "num_input_tokens_seen": 111901640,
      "step": 5209,
      "time_per_iteration": 2.5442450046539307
    },
    {
      "auxiliary_loss_clip": 0.01134312,
      "auxiliary_loss_mlp": 0.01044211,
      "balance_loss_clip": 1.04607558,
      "balance_loss_mlp": 1.02956009,
      "epoch": 0.31324214640012027,
      "flos": 23909863046400.0,
      "grad_norm": 1.7292165498821463,
      "language_loss": 0.77616775,
      "learning_rate": 3.2158098812401325e-06,
      "loss": 0.79795301,
      "num_input_tokens_seen": 111919615,
      "step": 5210,
      "time_per_iteration": 2.485924005508423
    },
    {
      "auxiliary_loss_clip": 0.01118425,
      "auxiliary_loss_mlp": 0.01039822,
      "balance_loss_clip": 1.04493189,
      "balance_loss_mlp": 1.02495706,
      "epoch": 0.31330226965278823,
      "flos": 22236929061120.0,
      "grad_norm": 1.645380940549625,
      "language_loss": 0.79459143,
      "learning_rate": 3.2155006217933874e-06,
      "loss": 0.81617391,
      "num_input_tokens_seen": 111938485,
      "step": 5211,
      "time_per_iteration": 3.9138388633728027
    },
    {
      "auxiliary_loss_clip": 0.01120589,
      "auxiliary_loss_mlp": 0.01037009,
      "balance_loss_clip": 1.0448072,
      "balance_loss_mlp": 1.02248931,
      "epoch": 0.3133623929054562,
      "flos": 19753993428480.0,
      "grad_norm": 1.7210271460844149,
      "language_loss": 0.79722607,
      "learning_rate": 3.2151913162547367e-06,
      "loss": 0.81880212,
      "num_input_tokens_seen": 111956425,
      "step": 5212,
      "time_per_iteration": 2.508111000061035
    },
    {
      "auxiliary_loss_clip": 0.01122775,
      "auxiliary_loss_mlp": 0.01047595,
      "balance_loss_clip": 1.04892015,
      "balance_loss_mlp": 1.03197837,
      "epoch": 0.31342251615812416,
      "flos": 27162256849920.0,
      "grad_norm": 2.1240796469420564,
      "language_loss": 0.7106024,
      "learning_rate": 3.2148819646359097e-06,
      "loss": 0.73230612,
      "num_input_tokens_seen": 111975915,
      "step": 5213,
      "time_per_iteration": 4.019595146179199
    },
    {
      "auxiliary_loss_clip": 0.01127092,
      "auxiliary_loss_mlp": 0.01045638,
      "balance_loss_clip": 1.05091286,
      "balance_loss_mlp": 1.03040349,
      "epoch": 0.31348263941079213,
      "flos": 20229809915520.0,
      "grad_norm": 2.1363712396767416,
      "language_loss": 0.77613342,
      "learning_rate": 3.2145725669486374e-06,
      "loss": 0.79786074,
      "num_input_tokens_seen": 111995055,
      "step": 5214,
      "time_per_iteration": 3.8577523231506348
    },
    {
      "auxiliary_loss_clip": 0.0108869,
      "auxiliary_loss_mlp": 0.01031607,
      "balance_loss_clip": 1.0465889,
      "balance_loss_mlp": 1.01761806,
      "epoch": 0.3135427626634601,
      "flos": 24607643627520.0,
      "grad_norm": 1.6006245861298873,
      "language_loss": 0.82704324,
      "learning_rate": 3.2142631232046517e-06,
      "loss": 0.84824622,
      "num_input_tokens_seen": 112015830,
      "step": 5215,
      "time_per_iteration": 2.6515002250671387
    },
    {
      "auxiliary_loss_clip": 0.01124746,
      "auxiliary_loss_mlp": 0.01038224,
      "balance_loss_clip": 1.04784989,
      "balance_loss_mlp": 1.02275038,
      "epoch": 0.31360288591612806,
      "flos": 20959873845120.0,
      "grad_norm": 2.0668590643427236,
      "language_loss": 0.79566288,
      "learning_rate": 3.213953633415686e-06,
      "loss": 0.81729257,
      "num_input_tokens_seen": 112035065,
      "step": 5216,
      "time_per_iteration": 2.4976272583007812
    },
    {
      "auxiliary_loss_clip": 0.01113977,
      "auxiliary_loss_mlp": 0.01046292,
      "balance_loss_clip": 1.04644322,
      "balance_loss_mlp": 1.0295074,
      "epoch": 0.313663009168796,
      "flos": 26980513009920.0,
      "grad_norm": 2.4569149726963775,
      "language_loss": 0.68258768,
      "learning_rate": 3.213644097593477e-06,
      "loss": 0.70419037,
      "num_input_tokens_seen": 112058405,
      "step": 5217,
      "time_per_iteration": 2.603549003601074
    },
    {
      "auxiliary_loss_clip": 0.01114681,
      "auxiliary_loss_mlp": 0.01035806,
      "balance_loss_clip": 1.04653287,
      "balance_loss_mlp": 1.0209105,
      "epoch": 0.313723132421464,
      "flos": 18040911016320.0,
      "grad_norm": 1.5895465305197436,
      "language_loss": 0.80903047,
      "learning_rate": 3.2133345157497624e-06,
      "loss": 0.83053529,
      "num_input_tokens_seen": 112076420,
      "step": 5218,
      "time_per_iteration": 2.492814302444458
    },
    {
      "auxiliary_loss_clip": 0.01133986,
      "auxiliary_loss_mlp": 0.01041928,
      "balance_loss_clip": 1.04660439,
      "balance_loss_mlp": 1.02551925,
      "epoch": 0.31378325567413196,
      "flos": 22488913946880.0,
      "grad_norm": 2.2846070132117826,
      "language_loss": 0.6874423,
      "learning_rate": 3.2130248878962813e-06,
      "loss": 0.70920146,
      "num_input_tokens_seen": 112090775,
      "step": 5219,
      "time_per_iteration": 2.4423274993896484
    },
    {
      "auxiliary_loss_clip": 0.01113631,
      "auxiliary_loss_mlp": 0.01035623,
      "balance_loss_clip": 1.04880488,
      "balance_loss_mlp": 1.02168107,
      "epoch": 0.3138433789267999,
      "flos": 22419247518720.0,
      "grad_norm": 1.9857780337760358,
      "language_loss": 0.79602402,
      "learning_rate": 3.2127152140447747e-06,
      "loss": 0.81751657,
      "num_input_tokens_seen": 112110980,
      "step": 5220,
      "time_per_iteration": 2.5338916778564453
    },
    {
      "auxiliary_loss_clip": 0.01125815,
      "auxiliary_loss_mlp": 0.01037908,
      "balance_loss_clip": 1.04784584,
      "balance_loss_mlp": 1.02340019,
      "epoch": 0.3139035021794679,
      "flos": 13005912026880.0,
      "grad_norm": 1.6462924290557253,
      "language_loss": 0.72998732,
      "learning_rate": 3.212405494206986e-06,
      "loss": 0.75162458,
      "num_input_tokens_seen": 112129020,
      "step": 5221,
      "time_per_iteration": 2.4895148277282715
    },
    {
      "auxiliary_loss_clip": 0.01106429,
      "auxiliary_loss_mlp": 0.0104227,
      "balance_loss_clip": 1.05023623,
      "balance_loss_mlp": 1.02719557,
      "epoch": 0.31396362543213585,
      "flos": 16945994689920.0,
      "grad_norm": 1.7472916931131102,
      "language_loss": 0.81640959,
      "learning_rate": 3.2120957283946588e-06,
      "loss": 0.83789653,
      "num_input_tokens_seen": 112147865,
      "step": 5222,
      "time_per_iteration": 2.505892753601074
    },
    {
      "auxiliary_loss_clip": 0.0112837,
      "auxiliary_loss_mlp": 0.01043166,
      "balance_loss_clip": 1.04827607,
      "balance_loss_mlp": 1.02633417,
      "epoch": 0.31402374868480387,
      "flos": 20156731695360.0,
      "grad_norm": 1.9738925164338472,
      "language_loss": 0.69880319,
      "learning_rate": 3.2117859166195407e-06,
      "loss": 0.72051847,
      "num_input_tokens_seen": 112166745,
      "step": 5223,
      "time_per_iteration": 2.478740930557251
    },
    {
      "auxiliary_loss_clip": 0.01119965,
      "auxiliary_loss_mlp": 0.00947269,
      "balance_loss_clip": 1.0488162,
      "balance_loss_mlp": 1.31515539,
      "epoch": 0.31408387193747184,
      "flos": 21251073404160.0,
      "grad_norm": 1.489419243883012,
      "language_loss": 0.80575806,
      "learning_rate": 3.211476058893379e-06,
      "loss": 0.82643044,
      "num_input_tokens_seen": 112185895,
      "step": 5224,
      "time_per_iteration": 2.537212371826172
    },
    {
      "auxiliary_loss_clip": 0.01135137,
      "auxiliary_loss_mlp": 0.01039503,
      "balance_loss_clip": 1.05306709,
      "balance_loss_mlp": 1.02350497,
      "epoch": 0.3141439951901398,
      "flos": 27484267299840.0,
      "grad_norm": 2.019719196773591,
      "language_loss": 0.5812881,
      "learning_rate": 3.2111661552279243e-06,
      "loss": 0.6030345,
      "num_input_tokens_seen": 112204465,
      "step": 5225,
      "time_per_iteration": 2.542659282684326
    },
    {
      "auxiliary_loss_clip": 0.01093186,
      "auxiliary_loss_mlp": 0.01034847,
      "balance_loss_clip": 1.04801822,
      "balance_loss_mlp": 1.02054787,
      "epoch": 0.31420411844280777,
      "flos": 17852235851520.0,
      "grad_norm": 1.7888639176679975,
      "language_loss": 0.81622255,
      "learning_rate": 3.2108562056349273e-06,
      "loss": 0.83750296,
      "num_input_tokens_seen": 112221635,
      "step": 5226,
      "time_per_iteration": 2.5989136695861816
    },
    {
      "auxiliary_loss_clip": 0.01121127,
      "auxiliary_loss_mlp": 0.01052735,
      "balance_loss_clip": 1.04862201,
      "balance_loss_mlp": 1.03587866,
      "epoch": 0.31426424169547573,
      "flos": 21616967295360.0,
      "grad_norm": 2.1020661112241776,
      "language_loss": 0.74108303,
      "learning_rate": 3.210546210126141e-06,
      "loss": 0.76282161,
      "num_input_tokens_seen": 112241240,
      "step": 5227,
      "time_per_iteration": 2.55983567237854
    },
    {
      "auxiliary_loss_clip": 0.01126166,
      "auxiliary_loss_mlp": 0.01036659,
      "balance_loss_clip": 1.05428696,
      "balance_loss_mlp": 1.02095902,
      "epoch": 0.3143243649481437,
      "flos": 30920631586560.0,
      "grad_norm": 1.6075864047314838,
      "language_loss": 0.6770004,
      "learning_rate": 3.2102361687133213e-06,
      "loss": 0.69862866,
      "num_input_tokens_seen": 112262350,
      "step": 5228,
      "time_per_iteration": 2.6122848987579346
    },
    {
      "auxiliary_loss_clip": 0.01113854,
      "auxiliary_loss_mlp": 0.01040202,
      "balance_loss_clip": 1.04726195,
      "balance_loss_mlp": 1.02512217,
      "epoch": 0.31438448820081166,
      "flos": 22821411168000.0,
      "grad_norm": 1.7897783144060964,
      "language_loss": 0.79885,
      "learning_rate": 3.2099260814082254e-06,
      "loss": 0.82039058,
      "num_input_tokens_seen": 112283710,
      "step": 5229,
      "time_per_iteration": 2.6531636714935303
    },
    {
      "auxiliary_loss_clip": 0.0111344,
      "auxiliary_loss_mlp": 0.01035148,
      "balance_loss_clip": 1.04995275,
      "balance_loss_mlp": 1.0194838,
      "epoch": 0.3144446114534796,
      "flos": 23292127923840.0,
      "grad_norm": 1.746706910547038,
      "language_loss": 0.70040798,
      "learning_rate": 3.209615948222611e-06,
      "loss": 0.72189391,
      "num_input_tokens_seen": 112304285,
      "step": 5230,
      "time_per_iteration": 2.6111514568328857
    },
    {
      "auxiliary_loss_clip": 0.01092543,
      "auxiliary_loss_mlp": 0.01048427,
      "balance_loss_clip": 1.04470325,
      "balance_loss_mlp": 1.03048611,
      "epoch": 0.3145047347061476,
      "flos": 31355976424320.0,
      "grad_norm": 1.5305369504833517,
      "language_loss": 0.79399204,
      "learning_rate": 3.209305769168239e-06,
      "loss": 0.81540167,
      "num_input_tokens_seen": 112325110,
      "step": 5231,
      "time_per_iteration": 2.692861557006836
    },
    {
      "auxiliary_loss_clip": 0.01109043,
      "auxiliary_loss_mlp": 0.01041283,
      "balance_loss_clip": 1.05148053,
      "balance_loss_mlp": 1.02499938,
      "epoch": 0.31456485795881556,
      "flos": 10889552643840.0,
      "grad_norm": 2.814438173731319,
      "language_loss": 0.84686923,
      "learning_rate": 3.2089955442568704e-06,
      "loss": 0.8683725,
      "num_input_tokens_seen": 112339855,
      "step": 5232,
      "time_per_iteration": 2.5521678924560547
    },
    {
      "auxiliary_loss_clip": 0.01078151,
      "auxiliary_loss_mlp": 0.0106228,
      "balance_loss_clip": 1.04185772,
      "balance_loss_mlp": 1.04542434,
      "epoch": 0.3146249812114835,
      "flos": 17092438439040.0,
      "grad_norm": 1.5429441291037251,
      "language_loss": 0.80080622,
      "learning_rate": 3.2086852735002692e-06,
      "loss": 0.82221055,
      "num_input_tokens_seen": 112358480,
      "step": 5233,
      "time_per_iteration": 2.6107914447784424
    },
    {
      "auxiliary_loss_clip": 0.01094156,
      "auxiliary_loss_mlp": 0.01038063,
      "balance_loss_clip": 1.04938102,
      "balance_loss_mlp": 1.02281618,
      "epoch": 0.3146851044641515,
      "flos": 55291442889600.0,
      "grad_norm": 1.694335972621516,
      "language_loss": 0.70801699,
      "learning_rate": 3.2083749569102024e-06,
      "loss": 0.72933924,
      "num_input_tokens_seen": 112382350,
      "step": 5234,
      "time_per_iteration": 2.9402360916137695
    },
    {
      "auxiliary_loss_clip": 0.01103781,
      "auxiliary_loss_mlp": 0.01036934,
      "balance_loss_clip": 1.05213642,
      "balance_loss_mlp": 1.02124047,
      "epoch": 0.31474522771681945,
      "flos": 27015884928000.0,
      "grad_norm": 2.129473387359332,
      "language_loss": 0.72344303,
      "learning_rate": 3.2080645944984356e-06,
      "loss": 0.74485016,
      "num_input_tokens_seen": 112400260,
      "step": 5235,
      "time_per_iteration": 2.6120810508728027
    },
    {
      "auxiliary_loss_clip": 0.01126101,
      "auxiliary_loss_mlp": 0.0103693,
      "balance_loss_clip": 1.047979,
      "balance_loss_mlp": 1.02139735,
      "epoch": 0.3148053509694875,
      "flos": 21251935330560.0,
      "grad_norm": 1.9028456087878876,
      "language_loss": 0.79160875,
      "learning_rate": 3.2077541862767384e-06,
      "loss": 0.81323904,
      "num_input_tokens_seen": 112419400,
      "step": 5236,
      "time_per_iteration": 2.5714783668518066
    },
    {
      "auxiliary_loss_clip": 0.01138928,
      "auxiliary_loss_mlp": 0.01041531,
      "balance_loss_clip": 1.04921556,
      "balance_loss_mlp": 1.02584291,
      "epoch": 0.31486547422215544,
      "flos": 31248675521280.0,
      "grad_norm": 1.5615004086982307,
      "language_loss": 0.75871325,
      "learning_rate": 3.207443732256881e-06,
      "loss": 0.78051782,
      "num_input_tokens_seen": 112440825,
      "step": 5237,
      "time_per_iteration": 2.5645601749420166
    },
    {
      "auxiliary_loss_clip": 0.01131906,
      "auxiliary_loss_mlp": 0.01035893,
      "balance_loss_clip": 1.04820347,
      "balance_loss_mlp": 1.02196932,
      "epoch": 0.3149255974748234,
      "flos": 19828615933440.0,
      "grad_norm": 1.710891802267315,
      "language_loss": 0.80179644,
      "learning_rate": 3.2071332324506372e-06,
      "loss": 0.82347441,
      "num_input_tokens_seen": 112459180,
      "step": 5238,
      "time_per_iteration": 2.5048158168792725
    },
    {
      "auxiliary_loss_clip": 0.01044726,
      "auxiliary_loss_mlp": 0.01002198,
      "balance_loss_clip": 1.02142191,
      "balance_loss_mlp": 0.99979037,
      "epoch": 0.31498572072749137,
      "flos": 67683965339520.0,
      "grad_norm": 0.8349325091043307,
      "language_loss": 0.67923057,
      "learning_rate": 3.2068226868697795e-06,
      "loss": 0.69969988,
      "num_input_tokens_seen": 112516680,
      "step": 5239,
      "time_per_iteration": 3.1070854663848877
    },
    {
      "auxiliary_loss_clip": 0.01116887,
      "auxiliary_loss_mlp": 0.01041061,
      "balance_loss_clip": 1.04814386,
      "balance_loss_mlp": 1.02428877,
      "epoch": 0.31504584398015933,
      "flos": 19793136274560.0,
      "grad_norm": 2.241149557686202,
      "language_loss": 0.83087802,
      "learning_rate": 3.2065120955260846e-06,
      "loss": 0.85245752,
      "num_input_tokens_seen": 112535895,
      "step": 5240,
      "time_per_iteration": 2.5836524963378906
    },
    {
      "auxiliary_loss_clip": 0.01113925,
      "auxiliary_loss_mlp": 0.00842079,
      "balance_loss_clip": 1.05128574,
      "balance_loss_mlp": 1.10492444,
      "epoch": 0.3151059672328273,
      "flos": 26615409217920.0,
      "grad_norm": 1.6545194853361676,
      "language_loss": 0.81287932,
      "learning_rate": 3.2062014584313302e-06,
      "loss": 0.83243942,
      "num_input_tokens_seen": 112557490,
      "step": 5241,
      "time_per_iteration": 2.578056812286377
    },
    {
      "auxiliary_loss_clip": 0.01136684,
      "auxiliary_loss_mlp": 0.01040504,
      "balance_loss_clip": 1.05189133,
      "balance_loss_mlp": 1.0254482,
      "epoch": 0.31516609048549526,
      "flos": 24204438483840.0,
      "grad_norm": 1.544403946190781,
      "language_loss": 0.74196118,
      "learning_rate": 3.2058907755972956e-06,
      "loss": 0.76373315,
      "num_input_tokens_seen": 112577075,
      "step": 5242,
      "time_per_iteration": 2.511141061782837
    },
    {
      "auxiliary_loss_clip": 0.01102535,
      "auxiliary_loss_mlp": 0.01040126,
      "balance_loss_clip": 1.04516459,
      "balance_loss_mlp": 1.02375901,
      "epoch": 0.31522621373816323,
      "flos": 25958710817280.0,
      "grad_norm": 1.9585354240595878,
      "language_loss": 0.74000096,
      "learning_rate": 3.2055800470357626e-06,
      "loss": 0.76142758,
      "num_input_tokens_seen": 112597620,
      "step": 5243,
      "time_per_iteration": 2.5963993072509766
    },
    {
      "auxiliary_loss_clip": 0.01124346,
      "auxiliary_loss_mlp": 0.01039557,
      "balance_loss_clip": 1.04975796,
      "balance_loss_mlp": 1.02440584,
      "epoch": 0.3152863369908312,
      "flos": 21908813299200.0,
      "grad_norm": 1.6939482005602378,
      "language_loss": 0.64090401,
      "learning_rate": 3.205269272758513e-06,
      "loss": 0.66254306,
      "num_input_tokens_seen": 112617150,
      "step": 5244,
      "time_per_iteration": 2.532068967819214
    },
    {
      "auxiliary_loss_clip": 0.01088929,
      "auxiliary_loss_mlp": 0.0103733,
      "balance_loss_clip": 1.0511024,
      "balance_loss_mlp": 1.022995,
      "epoch": 0.31534646024349916,
      "flos": 16281072074880.0,
      "grad_norm": 2.0787849358504777,
      "language_loss": 0.9110043,
      "learning_rate": 3.2049584527773313e-06,
      "loss": 0.93226689,
      "num_input_tokens_seen": 112631090,
      "step": 5245,
      "time_per_iteration": 4.014739274978638
    },
    {
      "auxiliary_loss_clip": 0.011276,
      "auxiliary_loss_mlp": 0.01042944,
      "balance_loss_clip": 1.04998124,
      "balance_loss_mlp": 1.02754807,
      "epoch": 0.3154065834961671,
      "flos": 24717243000960.0,
      "grad_norm": 1.612091820394434,
      "language_loss": 0.75035632,
      "learning_rate": 3.2046475871040048e-06,
      "loss": 0.77206177,
      "num_input_tokens_seen": 112651220,
      "step": 5246,
      "time_per_iteration": 2.5744895935058594
    },
    {
      "auxiliary_loss_clip": 0.01137869,
      "auxiliary_loss_mlp": 0.01042182,
      "balance_loss_clip": 1.05014086,
      "balance_loss_mlp": 1.02705455,
      "epoch": 0.3154667067488351,
      "flos": 35371148469120.0,
      "grad_norm": 1.4928941340316568,
      "language_loss": 0.61130077,
      "learning_rate": 3.204336675750321e-06,
      "loss": 0.63310128,
      "num_input_tokens_seen": 112671560,
      "step": 5247,
      "time_per_iteration": 2.6202356815338135
    },
    {
      "auxiliary_loss_clip": 0.01127343,
      "auxiliary_loss_mlp": 0.0104193,
      "balance_loss_clip": 1.04911995,
      "balance_loss_mlp": 1.02683771,
      "epoch": 0.31552683000150306,
      "flos": 17456464823040.0,
      "grad_norm": 2.423156577136237,
      "language_loss": 0.82155997,
      "learning_rate": 3.2040257187280693e-06,
      "loss": 0.84325266,
      "num_input_tokens_seen": 112689790,
      "step": 5248,
      "time_per_iteration": 2.4615566730499268
    },
    {
      "auxiliary_loss_clip": 0.01117163,
      "auxiliary_loss_mlp": 0.01050852,
      "balance_loss_clip": 1.04990673,
      "balance_loss_mlp": 1.03442538,
      "epoch": 0.3155869532541711,
      "flos": 18405763413120.0,
      "grad_norm": 2.278820933064458,
      "language_loss": 0.84688592,
      "learning_rate": 3.2037147160490423e-06,
      "loss": 0.86856604,
      "num_input_tokens_seen": 112708265,
      "step": 5249,
      "time_per_iteration": 2.5241925716400146
    },
    {
      "auxiliary_loss_clip": 0.01112543,
      "auxiliary_loss_mlp": 0.01043778,
      "balance_loss_clip": 1.05397642,
      "balance_loss_mlp": 1.02741051,
      "epoch": 0.31564707650683904,
      "flos": 21579763783680.0,
      "grad_norm": 1.6921497749167889,
      "language_loss": 0.85471606,
      "learning_rate": 3.2034036677250322e-06,
      "loss": 0.87627929,
      "num_input_tokens_seen": 112727820,
      "step": 5250,
      "time_per_iteration": 3.967754602432251
    },
    {
      "auxiliary_loss_clip": 0.01111459,
      "auxiliary_loss_mlp": 0.01043965,
      "balance_loss_clip": 1.05130935,
      "balance_loss_mlp": 1.02826476,
      "epoch": 0.315707199759507,
      "flos": 21030976817280.0,
      "grad_norm": 2.4246814535551846,
      "language_loss": 0.68575168,
      "learning_rate": 3.203092573767835e-06,
      "loss": 0.70730585,
      "num_input_tokens_seen": 112743140,
      "step": 5251,
      "time_per_iteration": 2.525578498840332
    },
    {
      "auxiliary_loss_clip": 0.01138099,
      "auxiliary_loss_mlp": 0.01044718,
      "balance_loss_clip": 1.05258727,
      "balance_loss_mlp": 1.02998412,
      "epoch": 0.31576732301217497,
      "flos": 26828861788800.0,
      "grad_norm": 1.87358731195123,
      "language_loss": 0.78980613,
      "learning_rate": 3.202781434189246e-06,
      "loss": 0.8116343,
      "num_input_tokens_seen": 112764705,
      "step": 5252,
      "time_per_iteration": 3.916820764541626
    },
    {
      "auxiliary_loss_clip": 0.01123084,
      "auxiliary_loss_mlp": 0.01056635,
      "balance_loss_clip": 1.05221581,
      "balance_loss_mlp": 1.04013658,
      "epoch": 0.31582744626484294,
      "flos": 22711165349760.0,
      "grad_norm": 1.5234422657884983,
      "language_loss": 0.74199605,
      "learning_rate": 3.202470249001066e-06,
      "loss": 0.76379329,
      "num_input_tokens_seen": 112785310,
      "step": 5253,
      "time_per_iteration": 3.900736093521118
    },
    {
      "auxiliary_loss_clip": 0.01122104,
      "auxiliary_loss_mlp": 0.0104031,
      "balance_loss_clip": 1.05114126,
      "balance_loss_mlp": 1.0249083,
      "epoch": 0.3158875695175109,
      "flos": 23951914894080.0,
      "grad_norm": 1.886348174412362,
      "language_loss": 0.73591691,
      "learning_rate": 3.2021590182150924e-06,
      "loss": 0.75754106,
      "num_input_tokens_seen": 112802905,
      "step": 5254,
      "time_per_iteration": 2.555074691772461
    },
    {
      "auxiliary_loss_clip": 0.01128137,
      "auxiliary_loss_mlp": 0.01042581,
      "balance_loss_clip": 1.05179739,
      "balance_loss_mlp": 1.02750659,
      "epoch": 0.31594769277017887,
      "flos": 13261883322240.0,
      "grad_norm": 1.7992620430986601,
      "language_loss": 0.77799451,
      "learning_rate": 3.201847741843128e-06,
      "loss": 0.79970169,
      "num_input_tokens_seen": 112820305,
      "step": 5255,
      "time_per_iteration": 2.4864346981048584
    },
    {
      "auxiliary_loss_clip": 0.01113901,
      "auxiliary_loss_mlp": 0.01043039,
      "balance_loss_clip": 1.05044961,
      "balance_loss_mlp": 1.02624285,
      "epoch": 0.31600781602284683,
      "flos": 23368258800000.0,
      "grad_norm": 1.9767084181228005,
      "language_loss": 0.78144288,
      "learning_rate": 3.2015364198969772e-06,
      "loss": 0.80301237,
      "num_input_tokens_seen": 112841185,
      "step": 5256,
      "time_per_iteration": 2.531808853149414
    },
    {
      "auxiliary_loss_clip": 0.01095894,
      "auxiliary_loss_mlp": 0.01038001,
      "balance_loss_clip": 1.05213141,
      "balance_loss_mlp": 1.02404106,
      "epoch": 0.3160679392755148,
      "flos": 19828580019840.0,
      "grad_norm": 1.5674103021189727,
      "language_loss": 0.71607172,
      "learning_rate": 3.2012250523884453e-06,
      "loss": 0.73741066,
      "num_input_tokens_seen": 112860570,
      "step": 5257,
      "time_per_iteration": 2.58862566947937
    },
    {
      "auxiliary_loss_clip": 0.01130246,
      "auxiliary_loss_mlp": 0.01043109,
      "balance_loss_clip": 1.0523386,
      "balance_loss_mlp": 1.02700412,
      "epoch": 0.31612806252818276,
      "flos": 20193216935040.0,
      "grad_norm": 1.863668782142547,
      "language_loss": 0.76839578,
      "learning_rate": 3.2009136393293393e-06,
      "loss": 0.79012936,
      "num_input_tokens_seen": 112877975,
      "step": 5258,
      "time_per_iteration": 2.479698896408081
    },
    {
      "auxiliary_loss_clip": 0.01107451,
      "auxiliary_loss_mlp": 0.01045763,
      "balance_loss_clip": 1.04908991,
      "balance_loss_mlp": 1.02974093,
      "epoch": 0.31618818578085073,
      "flos": 24235967646720.0,
      "grad_norm": 2.180268787874204,
      "language_loss": 0.72608805,
      "learning_rate": 3.200602180731467e-06,
      "loss": 0.74762022,
      "num_input_tokens_seen": 112896170,
      "step": 5259,
      "time_per_iteration": 2.5537657737731934
    },
    {
      "auxiliary_loss_clip": 0.01114652,
      "auxiliary_loss_mlp": 0.00811725,
      "balance_loss_clip": 1.05337918,
      "balance_loss_mlp": 1.04783964,
      "epoch": 0.3162483090335187,
      "flos": 25081844002560.0,
      "grad_norm": 1.912829640413767,
      "language_loss": 0.66653097,
      "learning_rate": 3.20029067660664e-06,
      "loss": 0.68579477,
      "num_input_tokens_seen": 112916180,
      "step": 5260,
      "time_per_iteration": 2.571277618408203
    },
    {
      "auxiliary_loss_clip": 0.01126576,
      "auxiliary_loss_mlp": 0.01032097,
      "balance_loss_clip": 1.04818964,
      "balance_loss_mlp": 1.01693392,
      "epoch": 0.31630843228618666,
      "flos": 26323383646080.0,
      "grad_norm": 1.8024607866830336,
      "language_loss": 0.72465134,
      "learning_rate": 3.1999791269666706e-06,
      "loss": 0.74623805,
      "num_input_tokens_seen": 112936745,
      "step": 5261,
      "time_per_iteration": 2.5579285621643066
    },
    {
      "auxiliary_loss_clip": 0.01043865,
      "auxiliary_loss_mlp": 0.01003024,
      "balance_loss_clip": 1.02235746,
      "balance_loss_mlp": 1.00068712,
      "epoch": 0.3163685555388547,
      "flos": 66758441552640.0,
      "grad_norm": 0.7534593805933323,
      "language_loss": 0.50716299,
      "learning_rate": 3.1996675318233716e-06,
      "loss": 0.52763188,
      "num_input_tokens_seen": 112994845,
      "step": 5262,
      "time_per_iteration": 3.1014809608459473
    },
    {
      "auxiliary_loss_clip": 0.01128076,
      "auxiliary_loss_mlp": 0.01038795,
      "balance_loss_clip": 1.05631959,
      "balance_loss_mlp": 1.02357197,
      "epoch": 0.31642867879152264,
      "flos": 25995662933760.0,
      "grad_norm": 1.627835951349252,
      "language_loss": 0.85454941,
      "learning_rate": 3.19935589118856e-06,
      "loss": 0.87621808,
      "num_input_tokens_seen": 113015125,
      "step": 5263,
      "time_per_iteration": 2.5535800457000732
    },
    {
      "auxiliary_loss_clip": 0.01109893,
      "auxiliary_loss_mlp": 0.01040706,
      "balance_loss_clip": 1.05085254,
      "balance_loss_mlp": 1.02637732,
      "epoch": 0.3164888020441906,
      "flos": 25774955815680.0,
      "grad_norm": 1.7541471067806753,
      "language_loss": 0.81845057,
      "learning_rate": 3.1990442050740535e-06,
      "loss": 0.83995658,
      "num_input_tokens_seen": 113035535,
      "step": 5264,
      "time_per_iteration": 2.5572218894958496
    },
    {
      "auxiliary_loss_clip": 0.01119606,
      "auxiliary_loss_mlp": 0.0103565,
      "balance_loss_clip": 1.05175257,
      "balance_loss_mlp": 1.01947355,
      "epoch": 0.3165489252968586,
      "flos": 19756220071680.0,
      "grad_norm": 2.0435652096615864,
      "language_loss": 0.79744875,
      "learning_rate": 3.19873247349167e-06,
      "loss": 0.81900132,
      "num_input_tokens_seen": 113052720,
      "step": 5265,
      "time_per_iteration": 2.556729793548584
    },
    {
      "auxiliary_loss_clip": 0.01132794,
      "auxiliary_loss_mlp": 0.01039446,
      "balance_loss_clip": 1.05561113,
      "balance_loss_mlp": 1.02374589,
      "epoch": 0.31660904854952654,
      "flos": 23183929180800.0,
      "grad_norm": 1.8881313936000494,
      "language_loss": 0.75263774,
      "learning_rate": 3.1984206964532307e-06,
      "loss": 0.77436012,
      "num_input_tokens_seen": 113071435,
      "step": 5266,
      "time_per_iteration": 2.508272409439087
    },
    {
      "auxiliary_loss_clip": 0.01108979,
      "auxiliary_loss_mlp": 0.01039507,
      "balance_loss_clip": 1.05234218,
      "balance_loss_mlp": 1.02417636,
      "epoch": 0.3166691718021945,
      "flos": 20408501099520.0,
      "grad_norm": 2.0441762575836893,
      "language_loss": 0.79174393,
      "learning_rate": 3.1981088739705585e-06,
      "loss": 0.81322879,
      "num_input_tokens_seen": 113088645,
      "step": 5267,
      "time_per_iteration": 2.5711276531219482
    },
    {
      "auxiliary_loss_clip": 0.0103884,
      "auxiliary_loss_mlp": 0.01001225,
      "balance_loss_clip": 1.0280776,
      "balance_loss_mlp": 0.99892408,
      "epoch": 0.31672929505486247,
      "flos": 70144781172480.0,
      "grad_norm": 0.7385054836274102,
      "language_loss": 0.57819247,
      "learning_rate": 3.197797006055478e-06,
      "loss": 0.59859312,
      "num_input_tokens_seen": 113152775,
      "step": 5268,
      "time_per_iteration": 3.141050338745117
    },
    {
      "auxiliary_loss_clip": 0.01142816,
      "auxiliary_loss_mlp": 0.01038485,
      "balance_loss_clip": 1.05329776,
      "balance_loss_mlp": 1.02314305,
      "epoch": 0.31678941830753043,
      "flos": 14355758154240.0,
      "grad_norm": 2.187262618804117,
      "language_loss": 0.73038757,
      "learning_rate": 3.197485092719815e-06,
      "loss": 0.7522006,
      "num_input_tokens_seen": 113171410,
      "step": 5269,
      "time_per_iteration": 2.4694061279296875
    },
    {
      "auxiliary_loss_clip": 0.01109208,
      "auxiliary_loss_mlp": 0.0104558,
      "balance_loss_clip": 1.05336595,
      "balance_loss_mlp": 1.03007102,
      "epoch": 0.3168495415601984,
      "flos": 22747722416640.0,
      "grad_norm": 1.92615857898234,
      "language_loss": 0.80228651,
      "learning_rate": 3.1971731339753973e-06,
      "loss": 0.82383442,
      "num_input_tokens_seen": 113189965,
      "step": 5270,
      "time_per_iteration": 2.568838119506836
    },
    {
      "auxiliary_loss_clip": 0.01145612,
      "auxiliary_loss_mlp": 0.01044615,
      "balance_loss_clip": 1.05459249,
      "balance_loss_mlp": 1.02792609,
      "epoch": 0.31690966481286637,
      "flos": 20115254465280.0,
      "grad_norm": 2.641262920522099,
      "language_loss": 0.79612768,
      "learning_rate": 3.1968611298340545e-06,
      "loss": 0.81803,
      "num_input_tokens_seen": 113206355,
      "step": 5271,
      "time_per_iteration": 2.449613571166992
    },
    {
      "auxiliary_loss_clip": 0.01142161,
      "auxiliary_loss_mlp": 0.01040242,
      "balance_loss_clip": 1.05404818,
      "balance_loss_mlp": 1.02460766,
      "epoch": 0.31696978806553433,
      "flos": 21178928937600.0,
      "grad_norm": 2.2520840173029733,
      "language_loss": 0.73129141,
      "learning_rate": 3.1965490803076173e-06,
      "loss": 0.75311548,
      "num_input_tokens_seen": 113225440,
      "step": 5272,
      "time_per_iteration": 2.4721462726593018
    },
    {
      "auxiliary_loss_clip": 0.01119199,
      "auxiliary_loss_mlp": 0.01045983,
      "balance_loss_clip": 1.0494206,
      "balance_loss_mlp": 1.02864981,
      "epoch": 0.3170299113182023,
      "flos": 42997030439040.0,
      "grad_norm": 2.0834405479839497,
      "language_loss": 0.68894649,
      "learning_rate": 3.1962369854079194e-06,
      "loss": 0.71059823,
      "num_input_tokens_seen": 113248840,
      "step": 5273,
      "time_per_iteration": 2.72156023979187
    },
    {
      "auxiliary_loss_clip": 0.01126228,
      "auxiliary_loss_mlp": 0.00804013,
      "balance_loss_clip": 1.0506283,
      "balance_loss_mlp": 1.03286171,
      "epoch": 0.31709003457087026,
      "flos": 24460158384000.0,
      "grad_norm": 2.2261655645941594,
      "language_loss": 0.67693901,
      "learning_rate": 3.195924845146795e-06,
      "loss": 0.69624138,
      "num_input_tokens_seen": 113269630,
      "step": 5274,
      "time_per_iteration": 2.535186767578125
    },
    {
      "auxiliary_loss_clip": 0.01094616,
      "auxiliary_loss_mlp": 0.01061929,
      "balance_loss_clip": 1.05058467,
      "balance_loss_mlp": 1.04481101,
      "epoch": 0.3171501578235382,
      "flos": 24135310759680.0,
      "grad_norm": 1.4858629231712213,
      "language_loss": 0.81048024,
      "learning_rate": 3.195612659536081e-06,
      "loss": 0.83204567,
      "num_input_tokens_seen": 113291200,
      "step": 5275,
      "time_per_iteration": 2.5741326808929443
    },
    {
      "auxiliary_loss_clip": 0.01128344,
      "auxiliary_loss_mlp": 0.01045838,
      "balance_loss_clip": 1.04925823,
      "balance_loss_mlp": 1.03026938,
      "epoch": 0.31721028107620625,
      "flos": 18879712392960.0,
      "grad_norm": 1.8129936355138583,
      "language_loss": 0.72861731,
      "learning_rate": 3.1953004285876147e-06,
      "loss": 0.75035912,
      "num_input_tokens_seen": 113310170,
      "step": 5276,
      "time_per_iteration": 2.4873099327087402
    },
    {
      "auxiliary_loss_clip": 0.01114059,
      "auxiliary_loss_mlp": 0.01040161,
      "balance_loss_clip": 1.04892099,
      "balance_loss_mlp": 1.02558148,
      "epoch": 0.3172704043288742,
      "flos": 23147874904320.0,
      "grad_norm": 1.4706361100939296,
      "language_loss": 0.78108144,
      "learning_rate": 3.194988152313236e-06,
      "loss": 0.80262363,
      "num_input_tokens_seen": 113331140,
      "step": 5277,
      "time_per_iteration": 2.557978630065918
    },
    {
      "auxiliary_loss_clip": 0.01109114,
      "auxiliary_loss_mlp": 0.01049966,
      "balance_loss_clip": 1.04867578,
      "balance_loss_mlp": 1.03104806,
      "epoch": 0.3173305275815422,
      "flos": 17858520731520.0,
      "grad_norm": 1.6391296860853455,
      "language_loss": 0.7918067,
      "learning_rate": 3.1946758307247878e-06,
      "loss": 0.81339753,
      "num_input_tokens_seen": 113350030,
      "step": 5278,
      "time_per_iteration": 2.5151944160461426
    },
    {
      "auxiliary_loss_clip": 0.01044267,
      "auxiliary_loss_mlp": 0.01013422,
      "balance_loss_clip": 1.02925158,
      "balance_loss_mlp": 1.01088333,
      "epoch": 0.31739065083421014,
      "flos": 59973476883840.0,
      "grad_norm": 0.8785415952885072,
      "language_loss": 0.62846756,
      "learning_rate": 3.1943634638341114e-06,
      "loss": 0.64904439,
      "num_input_tokens_seen": 113395820,
      "step": 5279,
      "time_per_iteration": 2.8708362579345703
    },
    {
      "auxiliary_loss_clip": 0.01144465,
      "auxiliary_loss_mlp": 0.01047525,
      "balance_loss_clip": 1.0523169,
      "balance_loss_mlp": 1.0304898,
      "epoch": 0.3174507740868781,
      "flos": 23800981944960.0,
      "grad_norm": 1.8091601046855692,
      "language_loss": 0.80960584,
      "learning_rate": 3.194051051653053e-06,
      "loss": 0.8315258,
      "num_input_tokens_seen": 113416835,
      "step": 5280,
      "time_per_iteration": 2.526648998260498
    },
    {
      "auxiliary_loss_clip": 0.01107991,
      "auxiliary_loss_mlp": 0.01045975,
      "balance_loss_clip": 1.05423522,
      "balance_loss_mlp": 1.03134799,
      "epoch": 0.31751089733954607,
      "flos": 27638899349760.0,
      "grad_norm": 1.9958910987416494,
      "language_loss": 0.77758092,
      "learning_rate": 3.19373859419346e-06,
      "loss": 0.79912055,
      "num_input_tokens_seen": 113440850,
      "step": 5281,
      "time_per_iteration": 2.658937454223633
    },
    {
      "auxiliary_loss_clip": 0.01118647,
      "auxiliary_loss_mlp": 0.01043743,
      "balance_loss_clip": 1.05239916,
      "balance_loss_mlp": 1.02719104,
      "epoch": 0.31757102059221404,
      "flos": 23769273214080.0,
      "grad_norm": 1.5867537340751843,
      "language_loss": 0.78144634,
      "learning_rate": 3.193426091467179e-06,
      "loss": 0.80307019,
      "num_input_tokens_seen": 113461000,
      "step": 5282,
      "time_per_iteration": 2.559037923812866
    },
    {
      "auxiliary_loss_clip": 0.01117286,
      "auxiliary_loss_mlp": 0.01052174,
      "balance_loss_clip": 1.05065203,
      "balance_loss_mlp": 1.03454316,
      "epoch": 0.317631143844882,
      "flos": 25264521596160.0,
      "grad_norm": 3.343983957483309,
      "language_loss": 0.67286217,
      "learning_rate": 3.193113543486061e-06,
      "loss": 0.69455671,
      "num_input_tokens_seen": 113480820,
      "step": 5283,
      "time_per_iteration": 2.559706687927246
    },
    {
      "auxiliary_loss_clip": 0.01040139,
      "auxiliary_loss_mlp": 0.01006966,
      "balance_loss_clip": 1.02919245,
      "balance_loss_mlp": 1.0049988,
      "epoch": 0.31769126709754997,
      "flos": 55825939221120.0,
      "grad_norm": 0.7513958323200097,
      "language_loss": 0.52788138,
      "learning_rate": 3.192800950261958e-06,
      "loss": 0.54835248,
      "num_input_tokens_seen": 113536910,
      "step": 5284,
      "time_per_iteration": 4.431357383728027
    },
    {
      "auxiliary_loss_clip": 0.01122683,
      "auxiliary_loss_mlp": 0.01037918,
      "balance_loss_clip": 1.05562294,
      "balance_loss_mlp": 1.02287376,
      "epoch": 0.31775139035021793,
      "flos": 16690562098560.0,
      "grad_norm": 1.6448547737019368,
      "language_loss": 0.70651275,
      "learning_rate": 3.1924883118067235e-06,
      "loss": 0.72811872,
      "num_input_tokens_seen": 113555480,
      "step": 5285,
      "time_per_iteration": 2.5310556888580322
    },
    {
      "auxiliary_loss_clip": 0.01048465,
      "auxiliary_loss_mlp": 0.01002051,
      "balance_loss_clip": 1.01780212,
      "balance_loss_mlp": 0.99988097,
      "epoch": 0.3178115136028859,
      "flos": 64227241019520.0,
      "grad_norm": 0.8206012940865721,
      "language_loss": 0.60512167,
      "learning_rate": 3.1921756281322123e-06,
      "loss": 0.6256268,
      "num_input_tokens_seen": 113616790,
      "step": 5286,
      "time_per_iteration": 3.0809075832366943
    },
    {
      "auxiliary_loss_clip": 0.01139758,
      "auxiliary_loss_mlp": 0.01043456,
      "balance_loss_clip": 1.05014896,
      "balance_loss_mlp": 1.02757764,
      "epoch": 0.31787163685555386,
      "flos": 18697465762560.0,
      "grad_norm": 1.816893621436008,
      "language_loss": 0.72385061,
      "learning_rate": 3.1918628992502826e-06,
      "loss": 0.74568278,
      "num_input_tokens_seen": 113635320,
      "step": 5287,
      "time_per_iteration": 2.4503214359283447
    },
    {
      "auxiliary_loss_clip": 0.01129737,
      "auxiliary_loss_mlp": 0.01047353,
      "balance_loss_clip": 1.05023813,
      "balance_loss_mlp": 1.02976942,
      "epoch": 0.31793176010822183,
      "flos": 21324762155520.0,
      "grad_norm": 2.276961695995455,
      "language_loss": 0.75566912,
      "learning_rate": 3.191550125172792e-06,
      "loss": 0.77744007,
      "num_input_tokens_seen": 113654000,
      "step": 5288,
      "time_per_iteration": 2.478504180908203
    },
    {
      "auxiliary_loss_clip": 0.01124303,
      "auxiliary_loss_mlp": 0.01034931,
      "balance_loss_clip": 1.04785633,
      "balance_loss_mlp": 1.02043498,
      "epoch": 0.31799188336088985,
      "flos": 20958688696320.0,
      "grad_norm": 1.6277909052519561,
      "language_loss": 0.87596345,
      "learning_rate": 3.1912373059116007e-06,
      "loss": 0.89755583,
      "num_input_tokens_seen": 113672375,
      "step": 5289,
      "time_per_iteration": 2.508985757827759
    },
    {
      "auxiliary_loss_clip": 0.01124259,
      "auxiliary_loss_mlp": 0.01037444,
      "balance_loss_clip": 1.05566287,
      "balance_loss_mlp": 1.02322221,
      "epoch": 0.3180520066135578,
      "flos": 22491930689280.0,
      "grad_norm": 1.7222326762923867,
      "language_loss": 0.68135965,
      "learning_rate": 3.190924441478572e-06,
      "loss": 0.70297658,
      "num_input_tokens_seen": 113692385,
      "step": 5290,
      "time_per_iteration": 3.8745224475860596
    },
    {
      "auxiliary_loss_clip": 0.01119708,
      "auxiliary_loss_mlp": 0.01045289,
      "balance_loss_clip": 1.05047846,
      "balance_loss_mlp": 1.02962506,
      "epoch": 0.3181121298662258,
      "flos": 27235335070080.0,
      "grad_norm": 1.913246041662527,
      "language_loss": 0.79831827,
      "learning_rate": 3.1906115318855687e-06,
      "loss": 0.81996828,
      "num_input_tokens_seen": 113712145,
      "step": 5291,
      "time_per_iteration": 3.9618608951568604
    },
    {
      "auxiliary_loss_clip": 0.01101437,
      "auxiliary_loss_mlp": 0.01039213,
      "balance_loss_clip": 1.05205846,
      "balance_loss_mlp": 1.02226186,
      "epoch": 0.31817225311889374,
      "flos": 23180158252800.0,
      "grad_norm": 2.111259424987871,
      "language_loss": 0.79828715,
      "learning_rate": 3.1902985771444577e-06,
      "loss": 0.81969362,
      "num_input_tokens_seen": 113731435,
      "step": 5292,
      "time_per_iteration": 3.992375135421753
    },
    {
      "auxiliary_loss_clip": 0.01128024,
      "auxiliary_loss_mlp": 0.01037086,
      "balance_loss_clip": 1.05051756,
      "balance_loss_mlp": 1.02261412,
      "epoch": 0.3182323763715617,
      "flos": 23258803080960.0,
      "grad_norm": 1.7149714046775084,
      "language_loss": 0.74992615,
      "learning_rate": 3.1899855772671043e-06,
      "loss": 0.77157724,
      "num_input_tokens_seen": 113750825,
      "step": 5293,
      "time_per_iteration": 2.511645555496216
    },
    {
      "auxiliary_loss_clip": 0.0112787,
      "auxiliary_loss_mlp": 0.01041326,
      "balance_loss_clip": 1.0545733,
      "balance_loss_mlp": 1.02655578,
      "epoch": 0.3182924996242297,
      "flos": 29016683280000.0,
      "grad_norm": 2.0920477972873273,
      "language_loss": 0.74227959,
      "learning_rate": 3.189672532265379e-06,
      "loss": 0.76397157,
      "num_input_tokens_seen": 113770010,
      "step": 5294,
      "time_per_iteration": 2.5681374073028564
    },
    {
      "auxiliary_loss_clip": 0.01143118,
      "auxiliary_loss_mlp": 0.01038744,
      "balance_loss_clip": 1.0529449,
      "balance_loss_mlp": 1.02179301,
      "epoch": 0.31835262287689764,
      "flos": 20449188230400.0,
      "grad_norm": 1.8813171065727299,
      "language_loss": 0.7593506,
      "learning_rate": 3.189359442151152e-06,
      "loss": 0.78116918,
      "num_input_tokens_seen": 113788640,
      "step": 5295,
      "time_per_iteration": 2.4738216400146484
    },
    {
      "auxiliary_loss_clip": 0.01115236,
      "auxiliary_loss_mlp": 0.01042768,
      "balance_loss_clip": 1.05407917,
      "balance_loss_mlp": 1.02688968,
      "epoch": 0.3184127461295656,
      "flos": 25119478477440.0,
      "grad_norm": 1.472895175501562,
      "language_loss": 0.69145858,
      "learning_rate": 3.189046306936296e-06,
      "loss": 0.71303868,
      "num_input_tokens_seen": 113809515,
      "step": 5296,
      "time_per_iteration": 2.58266544342041
    },
    {
      "auxiliary_loss_clip": 0.01114129,
      "auxiliary_loss_mlp": 0.0103982,
      "balance_loss_clip": 1.05432558,
      "balance_loss_mlp": 1.02481151,
      "epoch": 0.31847286938223357,
      "flos": 25551231955200.0,
      "grad_norm": 1.566883757979649,
      "language_loss": 0.77718604,
      "learning_rate": 3.1887331266326846e-06,
      "loss": 0.79872561,
      "num_input_tokens_seen": 113829770,
      "step": 5297,
      "time_per_iteration": 2.5843825340270996
    },
    {
      "auxiliary_loss_clip": 0.01105769,
      "auxiliary_loss_mlp": 0.01032014,
      "balance_loss_clip": 1.05038452,
      "balance_loss_mlp": 1.01627886,
      "epoch": 0.31853299263490154,
      "flos": 27782470010880.0,
      "grad_norm": 1.830209427058319,
      "language_loss": 0.79226089,
      "learning_rate": 3.1884199012521942e-06,
      "loss": 0.81363875,
      "num_input_tokens_seen": 113849320,
      "step": 5298,
      "time_per_iteration": 2.609867811203003
    },
    {
      "auxiliary_loss_clip": 0.01124506,
      "auxiliary_loss_mlp": 0.01042221,
      "balance_loss_clip": 1.05293727,
      "balance_loss_mlp": 1.0269382,
      "epoch": 0.3185931158875695,
      "flos": 22706747976960.0,
      "grad_norm": 1.6854594378553205,
      "language_loss": 0.74064636,
      "learning_rate": 3.1881066308067016e-06,
      "loss": 0.7623136,
      "num_input_tokens_seen": 113867860,
      "step": 5299,
      "time_per_iteration": 2.554365634918213
    },
    {
      "auxiliary_loss_clip": 0.01123902,
      "auxiliary_loss_mlp": 0.01047246,
      "balance_loss_clip": 1.05051339,
      "balance_loss_mlp": 1.03156996,
      "epoch": 0.31865323914023747,
      "flos": 24571517523840.0,
      "grad_norm": 1.9470045963142832,
      "language_loss": 0.78165495,
      "learning_rate": 3.1877933153080873e-06,
      "loss": 0.80336642,
      "num_input_tokens_seen": 113886375,
      "step": 5300,
      "time_per_iteration": 2.597076177597046
    },
    {
      "auxiliary_loss_clip": 0.01117275,
      "auxiliary_loss_mlp": 0.01042573,
      "balance_loss_clip": 1.05001724,
      "balance_loss_mlp": 1.02528751,
      "epoch": 0.31871336239290543,
      "flos": 18186564666240.0,
      "grad_norm": 2.077998744691648,
      "language_loss": 0.8387593,
      "learning_rate": 3.1874799547682304e-06,
      "loss": 0.86035776,
      "num_input_tokens_seen": 113904065,
      "step": 5301,
      "time_per_iteration": 2.5344371795654297
    },
    {
      "auxiliary_loss_clip": 0.01129491,
      "auxiliary_loss_mlp": 0.01043421,
      "balance_loss_clip": 1.05492163,
      "balance_loss_mlp": 1.02754235,
      "epoch": 0.31877348564557345,
      "flos": 21826756679040.0,
      "grad_norm": 2.365335892215333,
      "language_loss": 0.77421021,
      "learning_rate": 3.187166549199015e-06,
      "loss": 0.79593933,
      "num_input_tokens_seen": 113918415,
      "step": 5302,
      "time_per_iteration": 2.4855964183807373
    },
    {
      "auxiliary_loss_clip": 0.01139179,
      "auxiliary_loss_mlp": 0.01040018,
      "balance_loss_clip": 1.05411005,
      "balance_loss_mlp": 1.02410388,
      "epoch": 0.3188336088982414,
      "flos": 22015252275840.0,
      "grad_norm": 1.7209117075342533,
      "language_loss": 0.80104727,
      "learning_rate": 3.1868530986123255e-06,
      "loss": 0.82283926,
      "num_input_tokens_seen": 113938135,
      "step": 5303,
      "time_per_iteration": 2.6323611736297607
    },
    {
      "auxiliary_loss_clip": 0.01134059,
      "auxiliary_loss_mlp": 0.01042795,
      "balance_loss_clip": 1.05625963,
      "balance_loss_mlp": 1.02647519,
      "epoch": 0.3188937321509094,
      "flos": 20047886507520.0,
      "grad_norm": 3.1932599498675076,
      "language_loss": 0.73010331,
      "learning_rate": 3.186539603020047e-06,
      "loss": 0.75187182,
      "num_input_tokens_seen": 113957125,
      "step": 5304,
      "time_per_iteration": 2.4961016178131104
    },
    {
      "auxiliary_loss_clip": 0.01106329,
      "auxiliary_loss_mlp": 0.01042053,
      "balance_loss_clip": 1.05100846,
      "balance_loss_mlp": 1.0271039,
      "epoch": 0.31895385540357735,
      "flos": 25848105863040.0,
      "grad_norm": 2.05337772032714,
      "language_loss": 0.71786052,
      "learning_rate": 3.186226062434068e-06,
      "loss": 0.7393443,
      "num_input_tokens_seen": 113974875,
      "step": 5305,
      "time_per_iteration": 2.6048390865325928
    },
    {
      "auxiliary_loss_clip": 0.01118405,
      "auxiliary_loss_mlp": 0.0103943,
      "balance_loss_clip": 1.05146778,
      "balance_loss_mlp": 1.02539277,
      "epoch": 0.3190139786562453,
      "flos": 23477714519040.0,
      "grad_norm": 1.672105837745802,
      "language_loss": 0.64849806,
      "learning_rate": 3.1859124768662778e-06,
      "loss": 0.67007643,
      "num_input_tokens_seen": 113994450,
      "step": 5306,
      "time_per_iteration": 2.5436275005340576
    },
    {
      "auxiliary_loss_clip": 0.0111645,
      "auxiliary_loss_mlp": 0.01046225,
      "balance_loss_clip": 1.05560458,
      "balance_loss_mlp": 1.03034639,
      "epoch": 0.3190741019089133,
      "flos": 29095543589760.0,
      "grad_norm": 2.2694018674988277,
      "language_loss": 0.79467201,
      "learning_rate": 3.1855988463285678e-06,
      "loss": 0.81629872,
      "num_input_tokens_seen": 114013945,
      "step": 5307,
      "time_per_iteration": 2.588643789291382
    },
    {
      "auxiliary_loss_clip": 0.01111101,
      "auxiliary_loss_mlp": 0.01043503,
      "balance_loss_clip": 1.05141211,
      "balance_loss_mlp": 1.02712345,
      "epoch": 0.31913422516158124,
      "flos": 17129534209920.0,
      "grad_norm": 1.6861403002693403,
      "language_loss": 0.77331603,
      "learning_rate": 3.1852851708328308e-06,
      "loss": 0.79486209,
      "num_input_tokens_seen": 114031375,
      "step": 5308,
      "time_per_iteration": 2.50050687789917
    },
    {
      "auxiliary_loss_clip": 0.01143981,
      "auxiliary_loss_mlp": 0.01055543,
      "balance_loss_clip": 1.05885863,
      "balance_loss_mlp": 1.03774524,
      "epoch": 0.3191943484142492,
      "flos": 16069846147200.0,
      "grad_norm": 2.2221068614279353,
      "language_loss": 0.74622744,
      "learning_rate": 3.184971450390961e-06,
      "loss": 0.76822269,
      "num_input_tokens_seen": 114048465,
      "step": 5309,
      "time_per_iteration": 2.4778504371643066
    },
    {
      "auxiliary_loss_clip": 0.01129853,
      "auxiliary_loss_mlp": 0.01040547,
      "balance_loss_clip": 1.05107045,
      "balance_loss_mlp": 1.02562773,
      "epoch": 0.3192544716669172,
      "flos": 22966166977920.0,
      "grad_norm": 1.7513964125083117,
      "language_loss": 0.83157957,
      "learning_rate": 3.184657685014856e-06,
      "loss": 0.85328352,
      "num_input_tokens_seen": 114068415,
      "step": 5310,
      "time_per_iteration": 2.5016674995422363
    },
    {
      "auxiliary_loss_clip": 0.01116101,
      "auxiliary_loss_mlp": 0.01038631,
      "balance_loss_clip": 1.05126595,
      "balance_loss_mlp": 1.02484465,
      "epoch": 0.31931459491958514,
      "flos": 26870339018880.0,
      "grad_norm": 1.5550780244996725,
      "language_loss": 0.78235042,
      "learning_rate": 3.184343874716412e-06,
      "loss": 0.80389774,
      "num_input_tokens_seen": 114088565,
      "step": 5311,
      "time_per_iteration": 2.587149143218994
    },
    {
      "auxiliary_loss_clip": 0.01104121,
      "auxiliary_loss_mlp": 0.01043512,
      "balance_loss_clip": 1.04928446,
      "balance_loss_mlp": 1.02771628,
      "epoch": 0.3193747181722531,
      "flos": 21836525178240.0,
      "grad_norm": 1.7254339291869705,
      "language_loss": 0.84810984,
      "learning_rate": 3.1840300195075295e-06,
      "loss": 0.86958611,
      "num_input_tokens_seen": 114107160,
      "step": 5312,
      "time_per_iteration": 2.5641889572143555
    },
    {
      "auxiliary_loss_clip": 0.01098054,
      "auxiliary_loss_mlp": 0.01052846,
      "balance_loss_clip": 1.05305123,
      "balance_loss_mlp": 1.0354768,
      "epoch": 0.31943484142492107,
      "flos": 18324999682560.0,
      "grad_norm": 2.491688850830187,
      "language_loss": 0.78091627,
      "learning_rate": 3.1837161194001102e-06,
      "loss": 0.80242527,
      "num_input_tokens_seen": 114123420,
      "step": 5313,
      "time_per_iteration": 2.5364279747009277
    },
    {
      "auxiliary_loss_clip": 0.01128056,
      "auxiliary_loss_mlp": 0.01036689,
      "balance_loss_clip": 1.05131292,
      "balance_loss_mlp": 1.02170467,
      "epoch": 0.31949496467758903,
      "flos": 21615818060160.0,
      "grad_norm": 2.3651395868594536,
      "language_loss": 0.86310947,
      "learning_rate": 3.183402174406057e-06,
      "loss": 0.88475692,
      "num_input_tokens_seen": 114139230,
      "step": 5314,
      "time_per_iteration": 2.4928359985351562
    },
    {
      "auxiliary_loss_clip": 0.01112235,
      "auxiliary_loss_mlp": 0.01055268,
      "balance_loss_clip": 1.04897904,
      "balance_loss_mlp": 1.0377686,
      "epoch": 0.31955508793025705,
      "flos": 21760214734080.0,
      "grad_norm": 1.7504162293059833,
      "language_loss": 0.79702604,
      "learning_rate": 3.1830881845372747e-06,
      "loss": 0.81870109,
      "num_input_tokens_seen": 114159290,
      "step": 5315,
      "time_per_iteration": 2.549226999282837
    },
    {
      "auxiliary_loss_clip": 0.01104542,
      "auxiliary_loss_mlp": 0.01057435,
      "balance_loss_clip": 1.04989696,
      "balance_loss_mlp": 1.03982735,
      "epoch": 0.319615211182925,
      "flos": 17164331510400.0,
      "grad_norm": 1.9168739887784296,
      "language_loss": 0.67594182,
      "learning_rate": 3.18277414980567e-06,
      "loss": 0.69756162,
      "num_input_tokens_seen": 114177655,
      "step": 5316,
      "time_per_iteration": 2.5399420261383057
    },
    {
      "auxiliary_loss_clip": 0.01129004,
      "auxiliary_loss_mlp": 0.01039022,
      "balance_loss_clip": 1.05007148,
      "balance_loss_mlp": 1.02466953,
      "epoch": 0.319675334435593,
      "flos": 28112812416000.0,
      "grad_norm": 1.5121073909309855,
      "language_loss": 0.69194674,
      "learning_rate": 3.1824600702231515e-06,
      "loss": 0.71362698,
      "num_input_tokens_seen": 114200880,
      "step": 5317,
      "time_per_iteration": 2.5801501274108887
    },
    {
      "auxiliary_loss_clip": 0.01031301,
      "auxiliary_loss_mlp": 0.01008252,
      "balance_loss_clip": 1.02216911,
      "balance_loss_mlp": 1.00598681,
      "epoch": 0.31973545768826095,
      "flos": 69501119408640.0,
      "grad_norm": 0.72898209453558,
      "language_loss": 0.5301075,
      "learning_rate": 3.182145945801628e-06,
      "loss": 0.55050308,
      "num_input_tokens_seen": 114267145,
      "step": 5318,
      "time_per_iteration": 3.263522148132324
    },
    {
      "auxiliary_loss_clip": 0.01137497,
      "auxiliary_loss_mlp": 0.01034621,
      "balance_loss_clip": 1.05191064,
      "balance_loss_mlp": 1.0202148,
      "epoch": 0.3197955809409289,
      "flos": 13699203408000.0,
      "grad_norm": 2.615331686557557,
      "language_loss": 0.84313393,
      "learning_rate": 3.181831776553012e-06,
      "loss": 0.86485505,
      "num_input_tokens_seen": 114284630,
      "step": 5319,
      "time_per_iteration": 2.469848394393921
    },
    {
      "auxiliary_loss_clip": 0.01124364,
      "auxiliary_loss_mlp": 0.01039297,
      "balance_loss_clip": 1.04898071,
      "balance_loss_mlp": 1.02430665,
      "epoch": 0.3198557041935969,
      "flos": 33218124278400.0,
      "grad_norm": 1.6478802560791717,
      "language_loss": 0.63309526,
      "learning_rate": 3.1815175624892165e-06,
      "loss": 0.65473187,
      "num_input_tokens_seen": 114305830,
      "step": 5320,
      "time_per_iteration": 2.6162517070770264
    },
    {
      "auxiliary_loss_clip": 0.01122192,
      "auxiliary_loss_mlp": 0.01039653,
      "balance_loss_clip": 1.05026591,
      "balance_loss_mlp": 1.02457333,
      "epoch": 0.31991582744626484,
      "flos": 23732033788800.0,
      "grad_norm": 1.910989794333111,
      "language_loss": 0.7054162,
      "learning_rate": 3.1812033036221567e-06,
      "loss": 0.72703469,
      "num_input_tokens_seen": 114325165,
      "step": 5321,
      "time_per_iteration": 2.5638370513916016
    },
    {
      "auxiliary_loss_clip": 0.01146889,
      "auxiliary_loss_mlp": 0.00799634,
      "balance_loss_clip": 1.05285335,
      "balance_loss_mlp": 1.02231359,
      "epoch": 0.3199759506989328,
      "flos": 18550842445440.0,
      "grad_norm": 3.1853128897146705,
      "language_loss": 0.86737156,
      "learning_rate": 3.180888999963749e-06,
      "loss": 0.88683677,
      "num_input_tokens_seen": 114341310,
      "step": 5322,
      "time_per_iteration": 2.4539544582366943
    },
    {
      "auxiliary_loss_clip": 0.01115555,
      "auxiliary_loss_mlp": 0.01036456,
      "balance_loss_clip": 1.05041933,
      "balance_loss_mlp": 1.02150702,
      "epoch": 0.3200360739516008,
      "flos": 22418888382720.0,
      "grad_norm": 1.6770846466241798,
      "language_loss": 0.83398426,
      "learning_rate": 3.1805746515259123e-06,
      "loss": 0.85550433,
      "num_input_tokens_seen": 114360355,
      "step": 5323,
      "time_per_iteration": 4.040741920471191
    },
    {
      "auxiliary_loss_clip": 0.01130072,
      "auxiliary_loss_mlp": 0.01035825,
      "balance_loss_clip": 1.04988706,
      "balance_loss_mlp": 1.01966047,
      "epoch": 0.32009619720426874,
      "flos": 20595236929920.0,
      "grad_norm": 1.901155423949736,
      "language_loss": 0.77968752,
      "learning_rate": 3.1802602583205663e-06,
      "loss": 0.80134642,
      "num_input_tokens_seen": 114379220,
      "step": 5324,
      "time_per_iteration": 2.5038719177246094
    },
    {
      "auxiliary_loss_clip": 0.01114869,
      "auxiliary_loss_mlp": 0.01035491,
      "balance_loss_clip": 1.04962683,
      "balance_loss_mlp": 1.01945698,
      "epoch": 0.3201563204569367,
      "flos": 18147637301760.0,
      "grad_norm": 2.1939150842033075,
      "language_loss": 0.8018952,
      "learning_rate": 3.1799458203596333e-06,
      "loss": 0.82339877,
      "num_input_tokens_seen": 114396365,
      "step": 5325,
      "time_per_iteration": 2.5125627517700195
    },
    {
      "auxiliary_loss_clip": 0.01128205,
      "auxiliary_loss_mlp": 0.01040941,
      "balance_loss_clip": 1.05028224,
      "balance_loss_mlp": 1.0260042,
      "epoch": 0.32021644370960467,
      "flos": 31684235840640.0,
      "grad_norm": 1.9755504813661542,
      "language_loss": 0.74862838,
      "learning_rate": 3.179631337655037e-06,
      "loss": 0.77031988,
      "num_input_tokens_seen": 114416780,
      "step": 5326,
      "time_per_iteration": 2.5965280532836914
    },
    {
      "auxiliary_loss_clip": 0.01101284,
      "auxiliary_loss_mlp": 0.0104144,
      "balance_loss_clip": 1.05248737,
      "balance_loss_mlp": 1.02600193,
      "epoch": 0.32027656696227264,
      "flos": 26865921646080.0,
      "grad_norm": 1.5091304415708018,
      "language_loss": 0.80718458,
      "learning_rate": 3.179316810218701e-06,
      "loss": 0.82861185,
      "num_input_tokens_seen": 114437405,
      "step": 5327,
      "time_per_iteration": 2.6063945293426514
    },
    {
      "auxiliary_loss_clip": 0.01110497,
      "auxiliary_loss_mlp": 0.01041107,
      "balance_loss_clip": 1.05058551,
      "balance_loss_mlp": 1.02526414,
      "epoch": 0.32033669021494066,
      "flos": 24169928492160.0,
      "grad_norm": 1.4696409075538084,
      "language_loss": 0.77938831,
      "learning_rate": 3.179002238062554e-06,
      "loss": 0.80090433,
      "num_input_tokens_seen": 114458505,
      "step": 5328,
      "time_per_iteration": 2.6059303283691406
    },
    {
      "auxiliary_loss_clip": 0.01087584,
      "auxiliary_loss_mlp": 0.01040674,
      "balance_loss_clip": 1.05348194,
      "balance_loss_mlp": 1.02402008,
      "epoch": 0.3203968134676086,
      "flos": 24460768915200.0,
      "grad_norm": 1.8322120430742486,
      "language_loss": 0.74189055,
      "learning_rate": 3.178687621198524e-06,
      "loss": 0.76317316,
      "num_input_tokens_seen": 114479050,
      "step": 5329,
      "time_per_iteration": 4.044207334518433
    },
    {
      "auxiliary_loss_clip": 0.01107438,
      "auxiliary_loss_mlp": 0.01032578,
      "balance_loss_clip": 1.04697418,
      "balance_loss_mlp": 1.01823723,
      "epoch": 0.3204569367202766,
      "flos": 18004713085440.0,
      "grad_norm": 1.7709617261076471,
      "language_loss": 0.70846403,
      "learning_rate": 3.1783729596385415e-06,
      "loss": 0.72986424,
      "num_input_tokens_seen": 114497415,
      "step": 5330,
      "time_per_iteration": 3.90510630607605
    },
    {
      "auxiliary_loss_clip": 0.01089577,
      "auxiliary_loss_mlp": 0.01054094,
      "balance_loss_clip": 1.05086565,
      "balance_loss_mlp": 1.03518701,
      "epoch": 0.32051705997294455,
      "flos": 30589678650240.0,
      "grad_norm": 1.8801279049244408,
      "language_loss": 0.79911876,
      "learning_rate": 3.1780582533945376e-06,
      "loss": 0.82055545,
      "num_input_tokens_seen": 114518785,
      "step": 5331,
      "time_per_iteration": 4.064019441604614
    },
    {
      "auxiliary_loss_clip": 0.01035931,
      "auxiliary_loss_mlp": 0.01006097,
      "balance_loss_clip": 1.02032804,
      "balance_loss_mlp": 1.00418997,
      "epoch": 0.3205771832256125,
      "flos": 68417979765120.0,
      "grad_norm": 0.8238478946142479,
      "language_loss": 0.57755977,
      "learning_rate": 3.177743502478447e-06,
      "loss": 0.59798002,
      "num_input_tokens_seen": 114577710,
      "step": 5332,
      "time_per_iteration": 3.0372300148010254
    },
    {
      "auxiliary_loss_clip": 0.0110207,
      "auxiliary_loss_mlp": 0.01040571,
      "balance_loss_clip": 1.05117512,
      "balance_loss_mlp": 1.02517521,
      "epoch": 0.3206373064782805,
      "flos": 30443953173120.0,
      "grad_norm": 1.621781334019397,
      "language_loss": 0.72995144,
      "learning_rate": 3.177428706902205e-06,
      "loss": 0.75137782,
      "num_input_tokens_seen": 114598640,
      "step": 5333,
      "time_per_iteration": 2.6352834701538086
    },
    {
      "auxiliary_loss_clip": 0.01114954,
      "auxiliary_loss_mlp": 0.01039949,
      "balance_loss_clip": 1.04790461,
      "balance_loss_mlp": 1.0242548,
      "epoch": 0.32069742973094845,
      "flos": 22054502862720.0,
      "grad_norm": 1.562523781130142,
      "language_loss": 0.70477021,
      "learning_rate": 3.1771138666777485e-06,
      "loss": 0.72631931,
      "num_input_tokens_seen": 114618780,
      "step": 5334,
      "time_per_iteration": 2.5306146144866943
    },
    {
      "auxiliary_loss_clip": 0.0109793,
      "auxiliary_loss_mlp": 0.01040398,
      "balance_loss_clip": 1.0482707,
      "balance_loss_mlp": 1.02493656,
      "epoch": 0.3207575529836164,
      "flos": 22054000072320.0,
      "grad_norm": 1.8278808276735463,
      "language_loss": 0.77262294,
      "learning_rate": 3.1767989818170156e-06,
      "loss": 0.79400623,
      "num_input_tokens_seen": 114637525,
      "step": 5335,
      "time_per_iteration": 2.5924525260925293
    },
    {
      "auxiliary_loss_clip": 0.01127958,
      "auxiliary_loss_mlp": 0.01041475,
      "balance_loss_clip": 1.05114889,
      "balance_loss_mlp": 1.02640748,
      "epoch": 0.3208176762362844,
      "flos": 34057536186240.0,
      "grad_norm": 1.4928934711136235,
      "language_loss": 0.68518162,
      "learning_rate": 3.1764840523319477e-06,
      "loss": 0.70687604,
      "num_input_tokens_seen": 114659705,
      "step": 5336,
      "time_per_iteration": 2.6331582069396973
    },
    {
      "auxiliary_loss_clip": 0.01101639,
      "auxiliary_loss_mlp": 0.01044756,
      "balance_loss_clip": 1.05025053,
      "balance_loss_mlp": 1.02867436,
      "epoch": 0.32087779948895234,
      "flos": 21798711135360.0,
      "grad_norm": 1.641478190657808,
      "language_loss": 0.78838027,
      "learning_rate": 3.176169078234487e-06,
      "loss": 0.80984426,
      "num_input_tokens_seen": 114678340,
      "step": 5337,
      "time_per_iteration": 2.5509026050567627
    },
    {
      "auxiliary_loss_clip": 0.01120776,
      "auxiliary_loss_mlp": 0.01035852,
      "balance_loss_clip": 1.04822421,
      "balance_loss_mlp": 1.02163029,
      "epoch": 0.3209379227416203,
      "flos": 21434110133760.0,
      "grad_norm": 1.5052151443654458,
      "language_loss": 0.74244797,
      "learning_rate": 3.1758540595365766e-06,
      "loss": 0.76401424,
      "num_input_tokens_seen": 114696980,
      "step": 5338,
      "time_per_iteration": 2.575389862060547
    },
    {
      "auxiliary_loss_clip": 0.01122734,
      "auxiliary_loss_mlp": 0.01042738,
      "balance_loss_clip": 1.04887116,
      "balance_loss_mlp": 1.02665114,
      "epoch": 0.3209980459942883,
      "flos": 25849075530240.0,
      "grad_norm": 2.1612925551730626,
      "language_loss": 0.63173604,
      "learning_rate": 3.1755389962501626e-06,
      "loss": 0.65339071,
      "num_input_tokens_seen": 114717330,
      "step": 5339,
      "time_per_iteration": 2.555581569671631
    },
    {
      "auxiliary_loss_clip": 0.01139641,
      "auxiliary_loss_mlp": 0.01037485,
      "balance_loss_clip": 1.05055857,
      "balance_loss_mlp": 1.02215457,
      "epoch": 0.32105816924695624,
      "flos": 19099162535040.0,
      "grad_norm": 2.164286259137005,
      "language_loss": 0.81401825,
      "learning_rate": 3.175223888387192e-06,
      "loss": 0.8357895,
      "num_input_tokens_seen": 114736320,
      "step": 5340,
      "time_per_iteration": 2.4604363441467285
    },
    {
      "auxiliary_loss_clip": 0.01111345,
      "auxiliary_loss_mlp": 0.0104628,
      "balance_loss_clip": 1.04795384,
      "balance_loss_mlp": 1.03152823,
      "epoch": 0.3211182924996242,
      "flos": 16581860565120.0,
      "grad_norm": 1.955691480428128,
      "language_loss": 0.76468194,
      "learning_rate": 3.1749087359596137e-06,
      "loss": 0.78625816,
      "num_input_tokens_seen": 114754575,
      "step": 5341,
      "time_per_iteration": 2.5424680709838867
    },
    {
      "auxiliary_loss_clip": 0.01101984,
      "auxiliary_loss_mlp": 0.01036111,
      "balance_loss_clip": 1.05153632,
      "balance_loss_mlp": 1.02109015,
      "epoch": 0.3211784157522922,
      "flos": 22672202071680.0,
      "grad_norm": 1.7240146233484375,
      "language_loss": 0.7895838,
      "learning_rate": 3.1745935389793786e-06,
      "loss": 0.81096476,
      "num_input_tokens_seen": 114773590,
      "step": 5342,
      "time_per_iteration": 2.5626773834228516
    },
    {
      "auxiliary_loss_clip": 0.01115964,
      "auxiliary_loss_mlp": 0.01037755,
      "balance_loss_clip": 1.05250466,
      "balance_loss_mlp": 1.021626,
      "epoch": 0.3212385390049602,
      "flos": 20558787603840.0,
      "grad_norm": 2.496211081863201,
      "language_loss": 0.7521311,
      "learning_rate": 3.174278297458438e-06,
      "loss": 0.77366829,
      "num_input_tokens_seen": 114790775,
      "step": 5343,
      "time_per_iteration": 2.5242292881011963
    },
    {
      "auxiliary_loss_clip": 0.01075985,
      "auxiliary_loss_mlp": 0.01035953,
      "balance_loss_clip": 1.04667902,
      "balance_loss_mlp": 1.01988316,
      "epoch": 0.32129866225762815,
      "flos": 24791147233920.0,
      "grad_norm": 1.5357212026823333,
      "language_loss": 0.82633555,
      "learning_rate": 3.173963011408748e-06,
      "loss": 0.84745491,
      "num_input_tokens_seen": 114809835,
      "step": 5344,
      "time_per_iteration": 2.6325573921203613
    },
    {
      "auxiliary_loss_clip": 0.01086207,
      "auxiliary_loss_mlp": 0.01034348,
      "balance_loss_clip": 1.05107391,
      "balance_loss_mlp": 1.01874375,
      "epoch": 0.3213587855102961,
      "flos": 18366871962240.0,
      "grad_norm": 2.1405973518978594,
      "language_loss": 0.79600996,
      "learning_rate": 3.173647680842262e-06,
      "loss": 0.8172155,
      "num_input_tokens_seen": 114826505,
      "step": 5345,
      "time_per_iteration": 2.5617942810058594
    },
    {
      "auxiliary_loss_clip": 0.01110149,
      "auxiliary_loss_mlp": 0.01039298,
      "balance_loss_clip": 1.04668641,
      "balance_loss_mlp": 1.02439046,
      "epoch": 0.3214189087629641,
      "flos": 27015992668800.0,
      "grad_norm": 1.6892866632581776,
      "language_loss": 0.83098543,
      "learning_rate": 3.1733323057709384e-06,
      "loss": 0.85247982,
      "num_input_tokens_seen": 114846140,
      "step": 5346,
      "time_per_iteration": 2.5525989532470703
    },
    {
      "auxiliary_loss_clip": 0.01113606,
      "auxiliary_loss_mlp": 0.01038043,
      "balance_loss_clip": 1.04918456,
      "balance_loss_mlp": 1.02189612,
      "epoch": 0.32147903201563205,
      "flos": 23148269953920.0,
      "grad_norm": 1.4445545823993287,
      "language_loss": 0.8136065,
      "learning_rate": 3.1730168862067366e-06,
      "loss": 0.83512306,
      "num_input_tokens_seen": 114866660,
      "step": 5347,
      "time_per_iteration": 2.5763821601867676
    },
    {
      "auxiliary_loss_clip": 0.01123271,
      "auxiliary_loss_mlp": 0.01041363,
      "balance_loss_clip": 1.04748619,
      "balance_loss_mlp": 1.02470958,
      "epoch": 0.3215391552683,
      "flos": 16580747243520.0,
      "grad_norm": 1.9302074279259138,
      "language_loss": 0.79531425,
      "learning_rate": 3.1727014221616164e-06,
      "loss": 0.81696057,
      "num_input_tokens_seen": 114882820,
      "step": 5348,
      "time_per_iteration": 2.4493649005889893
    },
    {
      "auxiliary_loss_clip": 0.01114097,
      "auxiliary_loss_mlp": 0.01051344,
      "balance_loss_clip": 1.05181479,
      "balance_loss_mlp": 1.03525114,
      "epoch": 0.321599278520968,
      "flos": 17821820010240.0,
      "grad_norm": 1.9598161818453348,
      "language_loss": 0.848028,
      "learning_rate": 3.172385913647542e-06,
      "loss": 0.86968243,
      "num_input_tokens_seen": 114900745,
      "step": 5349,
      "time_per_iteration": 2.524142026901245
    },
    {
      "auxiliary_loss_clip": 0.01109744,
      "auxiliary_loss_mlp": 0.01046802,
      "balance_loss_clip": 1.04919636,
      "balance_loss_mlp": 1.03145981,
      "epoch": 0.32165940177363594,
      "flos": 16251769555200.0,
      "grad_norm": 2.058989106487354,
      "language_loss": 0.80391401,
      "learning_rate": 3.172070360676475e-06,
      "loss": 0.82547951,
      "num_input_tokens_seen": 114917940,
      "step": 5350,
      "time_per_iteration": 2.4989986419677734
    },
    {
      "auxiliary_loss_clip": 0.01125949,
      "auxiliary_loss_mlp": 0.01043299,
      "balance_loss_clip": 1.04984593,
      "balance_loss_mlp": 1.02862453,
      "epoch": 0.3217195250263039,
      "flos": 27599900158080.0,
      "grad_norm": 1.5445946804618296,
      "language_loss": 0.80306184,
      "learning_rate": 3.1717547632603828e-06,
      "loss": 0.82475436,
      "num_input_tokens_seen": 114937735,
      "step": 5351,
      "time_per_iteration": 2.5497188568115234
    },
    {
      "auxiliary_loss_clip": 0.01103731,
      "auxiliary_loss_mlp": 0.01045088,
      "balance_loss_clip": 1.05010498,
      "balance_loss_mlp": 1.02891099,
      "epoch": 0.3217796482789719,
      "flos": 21470595373440.0,
      "grad_norm": 1.669699508514137,
      "language_loss": 0.75719404,
      "learning_rate": 3.1714391214112326e-06,
      "loss": 0.77868223,
      "num_input_tokens_seen": 114956630,
      "step": 5352,
      "time_per_iteration": 2.5581398010253906
    },
    {
      "auxiliary_loss_clip": 0.0109709,
      "auxiliary_loss_mlp": 0.01041892,
      "balance_loss_clip": 1.05276668,
      "balance_loss_mlp": 1.02532208,
      "epoch": 0.32183977153163984,
      "flos": 21215593745280.0,
      "grad_norm": 2.085890552989524,
      "language_loss": 0.8239221,
      "learning_rate": 3.1711234351409933e-06,
      "loss": 0.84531188,
      "num_input_tokens_seen": 114976470,
      "step": 5353,
      "time_per_iteration": 2.574118137359619
    },
    {
      "auxiliary_loss_clip": 0.01074557,
      "auxiliary_loss_mlp": 0.01039199,
      "balance_loss_clip": 1.05164945,
      "balance_loss_mlp": 1.02341616,
      "epoch": 0.3218998947843078,
      "flos": 24608182331520.0,
      "grad_norm": 1.6226918520960243,
      "language_loss": 0.73130733,
      "learning_rate": 3.1708077044616365e-06,
      "loss": 0.75244486,
      "num_input_tokens_seen": 114996710,
      "step": 5354,
      "time_per_iteration": 2.69602632522583
    },
    {
      "auxiliary_loss_clip": 0.0110178,
      "auxiliary_loss_mlp": 0.01033349,
      "balance_loss_clip": 1.04649448,
      "balance_loss_mlp": 1.0189482,
      "epoch": 0.3219600180369758,
      "flos": 22270577126400.0,
      "grad_norm": 1.6632501529294739,
      "language_loss": 0.83869004,
      "learning_rate": 3.1704919293851334e-06,
      "loss": 0.86004132,
      "num_input_tokens_seen": 115015775,
      "step": 5355,
      "time_per_iteration": 2.5932302474975586
    },
    {
      "auxiliary_loss_clip": 0.01142714,
      "auxiliary_loss_mlp": 0.01045623,
      "balance_loss_clip": 1.05234218,
      "balance_loss_mlp": 1.03026938,
      "epoch": 0.3220201412896438,
      "flos": 14939126939520.0,
      "grad_norm": 1.990411790303877,
      "language_loss": 0.71202266,
      "learning_rate": 3.1701761099234597e-06,
      "loss": 0.73390603,
      "num_input_tokens_seen": 115034265,
      "step": 5356,
      "time_per_iteration": 2.4818005561828613
    },
    {
      "auxiliary_loss_clip": 0.01099335,
      "auxiliary_loss_mlp": 0.01042472,
      "balance_loss_clip": 1.05169332,
      "balance_loss_mlp": 1.02585387,
      "epoch": 0.32208026454231176,
      "flos": 22667389649280.0,
      "grad_norm": 2.241121934273558,
      "language_loss": 0.67085183,
      "learning_rate": 3.1698602460885903e-06,
      "loss": 0.69226992,
      "num_input_tokens_seen": 115051945,
      "step": 5357,
      "time_per_iteration": 2.607755422592163
    },
    {
      "auxiliary_loss_clip": 0.0103618,
      "auxiliary_loss_mlp": 0.01003302,
      "balance_loss_clip": 1.02293539,
      "balance_loss_mlp": 1.00119209,
      "epoch": 0.3221403877949797,
      "flos": 64605130053120.0,
      "grad_norm": 0.7103009869340805,
      "language_loss": 0.58286023,
      "learning_rate": 3.1695443378925035e-06,
      "loss": 0.60325503,
      "num_input_tokens_seen": 115119090,
      "step": 5358,
      "time_per_iteration": 3.2216641902923584
    },
    {
      "auxiliary_loss_clip": 0.01078088,
      "auxiliary_loss_mlp": 0.01040192,
      "balance_loss_clip": 1.05014145,
      "balance_loss_mlp": 1.02426553,
      "epoch": 0.3222005110476477,
      "flos": 20157019004160.0,
      "grad_norm": 1.7872036174454586,
      "language_loss": 0.83999765,
      "learning_rate": 3.1692283853471777e-06,
      "loss": 0.86118054,
      "num_input_tokens_seen": 115137755,
      "step": 5359,
      "time_per_iteration": 2.636096715927124
    },
    {
      "auxiliary_loss_clip": 0.01126916,
      "auxiliary_loss_mlp": 0.01037289,
      "balance_loss_clip": 1.04760993,
      "balance_loss_mlp": 1.02204227,
      "epoch": 0.32226063430031565,
      "flos": 22674177319680.0,
      "grad_norm": 1.6149637296728692,
      "language_loss": 0.79648942,
      "learning_rate": 3.168912388464595e-06,
      "loss": 0.81813145,
      "num_input_tokens_seen": 115158150,
      "step": 5360,
      "time_per_iteration": 2.5184459686279297
    },
    {
      "auxiliary_loss_clip": 0.01042647,
      "auxiliary_loss_mlp": 0.01005537,
      "balance_loss_clip": 1.0236603,
      "balance_loss_mlp": 1.00363004,
      "epoch": 0.3223207575529836,
      "flos": 63828525075840.0,
      "grad_norm": 0.6567233708891804,
      "language_loss": 0.56999826,
      "learning_rate": 3.168596347256737e-06,
      "loss": 0.59048009,
      "num_input_tokens_seen": 115212755,
      "step": 5361,
      "time_per_iteration": 2.9626264572143555
    },
    {
      "auxiliary_loss_clip": 0.0107903,
      "auxiliary_loss_mlp": 0.01041335,
      "balance_loss_clip": 1.04830849,
      "balance_loss_mlp": 1.02548027,
      "epoch": 0.3223808808056516,
      "flos": 26870123537280.0,
      "grad_norm": 1.7762274513735208,
      "language_loss": 0.71125782,
      "learning_rate": 3.168280261735588e-06,
      "loss": 0.73246151,
      "num_input_tokens_seen": 115233090,
      "step": 5362,
      "time_per_iteration": 4.108921527862549
    },
    {
      "auxiliary_loss_clip": 0.01122888,
      "auxiliary_loss_mlp": 0.01045759,
      "balance_loss_clip": 1.04927039,
      "balance_loss_mlp": 1.0295465,
      "epoch": 0.32244100405831955,
      "flos": 26761350176640.0,
      "grad_norm": 1.9001845790567662,
      "language_loss": 0.74009669,
      "learning_rate": 3.167964131913135e-06,
      "loss": 0.76178318,
      "num_input_tokens_seen": 115252645,
      "step": 5363,
      "time_per_iteration": 2.535872459411621
    },
    {
      "auxiliary_loss_clip": 0.01130156,
      "auxiliary_loss_mlp": 0.01044466,
      "balance_loss_clip": 1.04879737,
      "balance_loss_mlp": 1.02911234,
      "epoch": 0.3225011273109875,
      "flos": 23803029020160.0,
      "grad_norm": 2.366702561285205,
      "language_loss": 0.76950383,
      "learning_rate": 3.167647957801365e-06,
      "loss": 0.79125005,
      "num_input_tokens_seen": 115269085,
      "step": 5364,
      "time_per_iteration": 2.5045273303985596
    },
    {
      "auxiliary_loss_clip": 0.01116345,
      "auxiliary_loss_mlp": 0.01041332,
      "balance_loss_clip": 1.04888296,
      "balance_loss_mlp": 1.02527416,
      "epoch": 0.3225612505636555,
      "flos": 17274505501440.0,
      "grad_norm": 2.4014437908881545,
      "language_loss": 0.77007008,
      "learning_rate": 3.1673317394122672e-06,
      "loss": 0.79164684,
      "num_input_tokens_seen": 115286470,
      "step": 5365,
      "time_per_iteration": 2.5036580562591553
    },
    {
      "auxiliary_loss_clip": 0.01123807,
      "auxiliary_loss_mlp": 0.0104668,
      "balance_loss_clip": 1.05610359,
      "balance_loss_mlp": 1.03100348,
      "epoch": 0.32262137381632344,
      "flos": 23366247638400.0,
      "grad_norm": 1.6573394377358865,
      "language_loss": 0.7671634,
      "learning_rate": 3.1670154767578333e-06,
      "loss": 0.78886825,
      "num_input_tokens_seen": 115307000,
      "step": 5366,
      "time_per_iteration": 2.5649216175079346
    },
    {
      "auxiliary_loss_clip": 0.01110219,
      "auxiliary_loss_mlp": 0.01037726,
      "balance_loss_clip": 1.04839659,
      "balance_loss_mlp": 1.0220201,
      "epoch": 0.3226814970689914,
      "flos": 23258803080960.0,
      "grad_norm": 2.291122162140556,
      "language_loss": 0.72452486,
      "learning_rate": 3.166699169850055e-06,
      "loss": 0.74600434,
      "num_input_tokens_seen": 115325925,
      "step": 5367,
      "time_per_iteration": 3.936999559402466
    },
    {
      "auxiliary_loss_clip": 0.01136197,
      "auxiliary_loss_mlp": 0.01039917,
      "balance_loss_clip": 1.0500015,
      "balance_loss_mlp": 1.02532589,
      "epoch": 0.32274162032165943,
      "flos": 16395196561920.0,
      "grad_norm": 1.7984507437287314,
      "language_loss": 0.74638617,
      "learning_rate": 3.1663828187009274e-06,
      "loss": 0.76814729,
      "num_input_tokens_seen": 115343705,
      "step": 5368,
      "time_per_iteration": 3.825505495071411
    },
    {
      "auxiliary_loss_clip": 0.0110321,
      "auxiliary_loss_mlp": 0.01046874,
      "balance_loss_clip": 1.04709005,
      "balance_loss_mlp": 1.03147805,
      "epoch": 0.3228017435743274,
      "flos": 27855081354240.0,
      "grad_norm": 1.5525925577263395,
      "language_loss": 0.78619456,
      "learning_rate": 3.1660664233224467e-06,
      "loss": 0.80769545,
      "num_input_tokens_seen": 115364170,
      "step": 5369,
      "time_per_iteration": 2.581200361251831
    },
    {
      "auxiliary_loss_clip": 0.0109956,
      "auxiliary_loss_mlp": 0.01036801,
      "balance_loss_clip": 1.05190003,
      "balance_loss_mlp": 1.02235258,
      "epoch": 0.32286186682699536,
      "flos": 19608770741760.0,
      "grad_norm": 1.8485307558252508,
      "language_loss": 0.83428025,
      "learning_rate": 3.16574998372661e-06,
      "loss": 0.85564387,
      "num_input_tokens_seen": 115382495,
      "step": 5370,
      "time_per_iteration": 4.0455451011657715
    },
    {
      "auxiliary_loss_clip": 0.01138555,
      "auxiliary_loss_mlp": 0.01038871,
      "balance_loss_clip": 1.05154407,
      "balance_loss_mlp": 1.02369523,
      "epoch": 0.3229219900796633,
      "flos": 24134017870080.0,
      "grad_norm": 3.7810677752852175,
      "language_loss": 0.8308112,
      "learning_rate": 3.1654334999254177e-06,
      "loss": 0.85258543,
      "num_input_tokens_seen": 115399450,
      "step": 5371,
      "time_per_iteration": 2.4780356884002686
    },
    {
      "auxiliary_loss_clip": 0.01130216,
      "auxiliary_loss_mlp": 0.00802373,
      "balance_loss_clip": 1.05070186,
      "balance_loss_mlp": 1.0266993,
      "epoch": 0.3229821133323313,
      "flos": 17748705876480.0,
      "grad_norm": 2.0557034085707735,
      "language_loss": 0.88297474,
      "learning_rate": 3.1651169719308695e-06,
      "loss": 0.9023006,
      "num_input_tokens_seen": 115417700,
      "step": 5372,
      "time_per_iteration": 2.4758169651031494
    },
    {
      "auxiliary_loss_clip": 0.01139235,
      "auxiliary_loss_mlp": 0.01045029,
      "balance_loss_clip": 1.05218077,
      "balance_loss_mlp": 1.02954328,
      "epoch": 0.32304223658499925,
      "flos": 22346025644160.0,
      "grad_norm": 1.916731588964195,
      "language_loss": 0.73154128,
      "learning_rate": 3.1648003997549694e-06,
      "loss": 0.75338393,
      "num_input_tokens_seen": 115435840,
      "step": 5373,
      "time_per_iteration": 2.4670889377593994
    },
    {
      "auxiliary_loss_clip": 0.01109568,
      "auxiliary_loss_mlp": 0.01035237,
      "balance_loss_clip": 1.05046034,
      "balance_loss_mlp": 1.0201329,
      "epoch": 0.3231023598376672,
      "flos": 18478302929280.0,
      "grad_norm": 1.9366577151247066,
      "language_loss": 0.81443393,
      "learning_rate": 3.1644837834097214e-06,
      "loss": 0.83588195,
      "num_input_tokens_seen": 115454210,
      "step": 5374,
      "time_per_iteration": 2.512446165084839
    },
    {
      "auxiliary_loss_clip": 0.01091712,
      "auxiliary_loss_mlp": 0.01037771,
      "balance_loss_clip": 1.04674721,
      "balance_loss_mlp": 1.02228534,
      "epoch": 0.3231624830903352,
      "flos": 27636313570560.0,
      "grad_norm": 2.1493220726945865,
      "language_loss": 0.8738156,
      "learning_rate": 3.1641671229071317e-06,
      "loss": 0.89511043,
      "num_input_tokens_seen": 115471785,
      "step": 5375,
      "time_per_iteration": 2.596646547317505
    },
    {
      "auxiliary_loss_clip": 0.01139871,
      "auxiliary_loss_mlp": 0.01037742,
      "balance_loss_clip": 1.04915297,
      "balance_loss_mlp": 1.02195919,
      "epoch": 0.32322260634300315,
      "flos": 21726423014400.0,
      "grad_norm": 2.059544517882633,
      "language_loss": 0.75919485,
      "learning_rate": 3.1638504182592076e-06,
      "loss": 0.78097105,
      "num_input_tokens_seen": 115491405,
      "step": 5376,
      "time_per_iteration": 2.4884703159332275
    },
    {
      "auxiliary_loss_clip": 0.01099338,
      "auxiliary_loss_mlp": 0.01032746,
      "balance_loss_clip": 1.04974127,
      "balance_loss_mlp": 1.01879287,
      "epoch": 0.3232827295956711,
      "flos": 22637656166400.0,
      "grad_norm": 1.4961108394962088,
      "language_loss": 0.66752285,
      "learning_rate": 3.1635336694779594e-06,
      "loss": 0.68884373,
      "num_input_tokens_seen": 115511555,
      "step": 5377,
      "time_per_iteration": 2.5662660598754883
    },
    {
      "auxiliary_loss_clip": 0.01100977,
      "auxiliary_loss_mlp": 0.01050163,
      "balance_loss_clip": 1.0465579,
      "balance_loss_mlp": 1.03236496,
      "epoch": 0.3233428528483391,
      "flos": 26322593546880.0,
      "grad_norm": 1.40495165123392,
      "language_loss": 0.72054923,
      "learning_rate": 3.1632168765753982e-06,
      "loss": 0.74206066,
      "num_input_tokens_seen": 115532860,
      "step": 5378,
      "time_per_iteration": 2.600036382675171
    },
    {
      "auxiliary_loss_clip": 0.0112097,
      "auxiliary_loss_mlp": 0.01037339,
      "balance_loss_clip": 1.04930127,
      "balance_loss_mlp": 1.02254486,
      "epoch": 0.32340297610100704,
      "flos": 28585217111040.0,
      "grad_norm": 2.196769333378244,
      "language_loss": 0.82174897,
      "learning_rate": 3.1629000395635357e-06,
      "loss": 0.84333211,
      "num_input_tokens_seen": 115553850,
      "step": 5379,
      "time_per_iteration": 2.6044538021087646
    },
    {
      "auxiliary_loss_clip": 0.01136481,
      "auxiliary_loss_mlp": 0.01039561,
      "balance_loss_clip": 1.05703568,
      "balance_loss_mlp": 1.02506518,
      "epoch": 0.323463099353675,
      "flos": 30773792787840.0,
      "grad_norm": 1.871747138226317,
      "language_loss": 0.78792357,
      "learning_rate": 3.162583158454388e-06,
      "loss": 0.80968398,
      "num_input_tokens_seen": 115575530,
      "step": 5380,
      "time_per_iteration": 2.5817179679870605
    },
    {
      "auxiliary_loss_clip": 0.01122744,
      "auxiliary_loss_mlp": 0.01038527,
      "balance_loss_clip": 1.05090952,
      "balance_loss_mlp": 1.02402472,
      "epoch": 0.32352322260634303,
      "flos": 25228610974080.0,
      "grad_norm": 1.6978516562193577,
      "language_loss": 0.77136946,
      "learning_rate": 3.1622662332599697e-06,
      "loss": 0.7929821,
      "num_input_tokens_seen": 115594885,
      "step": 5381,
      "time_per_iteration": 2.5174336433410645
    },
    {
      "auxiliary_loss_clip": 0.01120248,
      "auxiliary_loss_mlp": 0.01037438,
      "balance_loss_clip": 1.04743195,
      "balance_loss_mlp": 1.02362716,
      "epoch": 0.323583345859011,
      "flos": 23330480670720.0,
      "grad_norm": 1.7718215324592568,
      "language_loss": 0.7114076,
      "learning_rate": 3.1619492639922998e-06,
      "loss": 0.73298442,
      "num_input_tokens_seen": 115614080,
      "step": 5382,
      "time_per_iteration": 2.5249316692352295
    },
    {
      "auxiliary_loss_clip": 0.01117105,
      "auxiliary_loss_mlp": 0.01045785,
      "balance_loss_clip": 1.04847646,
      "balance_loss_mlp": 1.02975094,
      "epoch": 0.32364346911167896,
      "flos": 26207499392640.0,
      "grad_norm": 2.5260313995271755,
      "language_loss": 0.71033585,
      "learning_rate": 3.1616322506633964e-06,
      "loss": 0.73196477,
      "num_input_tokens_seen": 115632820,
      "step": 5383,
      "time_per_iteration": 2.548656702041626
    },
    {
      "auxiliary_loss_clip": 0.01120953,
      "auxiliary_loss_mlp": 0.01039501,
      "balance_loss_clip": 1.04777932,
      "balance_loss_mlp": 1.02587533,
      "epoch": 0.3237035923643469,
      "flos": 23695764030720.0,
      "grad_norm": 1.7984304122905868,
      "language_loss": 0.78510875,
      "learning_rate": 3.161315193285283e-06,
      "loss": 0.80671328,
      "num_input_tokens_seen": 115652860,
      "step": 5384,
      "time_per_iteration": 2.5162243843078613
    },
    {
      "auxiliary_loss_clip": 0.01080513,
      "auxiliary_loss_mlp": 0.01042879,
      "balance_loss_clip": 1.05090392,
      "balance_loss_mlp": 1.0260942,
      "epoch": 0.3237637156170149,
      "flos": 14428728633600.0,
      "grad_norm": 2.6447946742184496,
      "language_loss": 0.74938333,
      "learning_rate": 3.16099809186998e-06,
      "loss": 0.77061725,
      "num_input_tokens_seen": 115670940,
      "step": 5385,
      "time_per_iteration": 2.588855504989624
    },
    {
      "auxiliary_loss_clip": 0.01110195,
      "auxiliary_loss_mlp": 0.01038839,
      "balance_loss_clip": 1.05471933,
      "balance_loss_mlp": 1.02387226,
      "epoch": 0.32382383886968286,
      "flos": 31062981185280.0,
      "grad_norm": 1.7155073105094663,
      "language_loss": 0.71663594,
      "learning_rate": 3.1606809464295145e-06,
      "loss": 0.73812628,
      "num_input_tokens_seen": 115691155,
      "step": 5386,
      "time_per_iteration": 2.6090614795684814
    },
    {
      "auxiliary_loss_clip": 0.01137753,
      "auxiliary_loss_mlp": 0.01037638,
      "balance_loss_clip": 1.04708409,
      "balance_loss_mlp": 1.02227223,
      "epoch": 0.3238839621223508,
      "flos": 23256935573760.0,
      "grad_norm": 2.0879008397508803,
      "language_loss": 0.948394,
      "learning_rate": 3.1603637569759095e-06,
      "loss": 0.97014797,
      "num_input_tokens_seen": 115710340,
      "step": 5387,
      "time_per_iteration": 2.4672648906707764
    },
    {
      "auxiliary_loss_clip": 0.01128423,
      "auxiliary_loss_mlp": 0.01046175,
      "balance_loss_clip": 1.04990029,
      "balance_loss_mlp": 1.03041542,
      "epoch": 0.3239440853750188,
      "flos": 22964658606720.0,
      "grad_norm": 2.014040467567363,
      "language_loss": 0.77483904,
      "learning_rate": 3.1600465235211956e-06,
      "loss": 0.79658496,
      "num_input_tokens_seen": 115726745,
      "step": 5388,
      "time_per_iteration": 2.4844744205474854
    },
    {
      "auxiliary_loss_clip": 0.01110109,
      "auxiliary_loss_mlp": 0.01035677,
      "balance_loss_clip": 1.04855311,
      "balance_loss_mlp": 1.02029848,
      "epoch": 0.32400420862768675,
      "flos": 36246614653440.0,
      "grad_norm": 1.833516454365437,
      "language_loss": 0.71096838,
      "learning_rate": 3.1597292460774006e-06,
      "loss": 0.73242629,
      "num_input_tokens_seen": 115749385,
      "step": 5389,
      "time_per_iteration": 2.6484270095825195
    },
    {
      "auxiliary_loss_clip": 0.01098278,
      "auxiliary_loss_mlp": 0.01041918,
      "balance_loss_clip": 1.04968739,
      "balance_loss_mlp": 1.02626598,
      "epoch": 0.3240643318803547,
      "flos": 21616500418560.0,
      "grad_norm": 1.6988650048734761,
      "language_loss": 0.80835056,
      "learning_rate": 3.159411924656557e-06,
      "loss": 0.82975256,
      "num_input_tokens_seen": 115768105,
      "step": 5390,
      "time_per_iteration": 2.5615243911743164
    },
    {
      "auxiliary_loss_clip": 0.01110034,
      "auxiliary_loss_mlp": 0.01047814,
      "balance_loss_clip": 1.05075431,
      "balance_loss_mlp": 1.03247154,
      "epoch": 0.3241244551330227,
      "flos": 23295611543040.0,
      "grad_norm": 1.8010368707071807,
      "language_loss": 0.73385483,
      "learning_rate": 3.1590945592706967e-06,
      "loss": 0.75543332,
      "num_input_tokens_seen": 115787340,
      "step": 5391,
      "time_per_iteration": 2.550978422164917
    },
    {
      "auxiliary_loss_clip": 0.01111601,
      "auxiliary_loss_mlp": 0.01040955,
      "balance_loss_clip": 1.04781353,
      "balance_loss_mlp": 1.02645862,
      "epoch": 0.32418457838569065,
      "flos": 14097236993280.0,
      "grad_norm": 1.5497678118827314,
      "language_loss": 0.77146083,
      "learning_rate": 3.158777149931855e-06,
      "loss": 0.79298639,
      "num_input_tokens_seen": 115805565,
      "step": 5392,
      "time_per_iteration": 2.515350103378296
    },
    {
      "auxiliary_loss_clip": 0.01109367,
      "auxiliary_loss_mlp": 0.01046567,
      "balance_loss_clip": 1.04539609,
      "balance_loss_mlp": 1.02900708,
      "epoch": 0.3242447016383586,
      "flos": 29752672953600.0,
      "grad_norm": 2.2128977512186236,
      "language_loss": 0.62530446,
      "learning_rate": 3.158459696652067e-06,
      "loss": 0.64686382,
      "num_input_tokens_seen": 115826725,
      "step": 5393,
      "time_per_iteration": 2.594237804412842
    },
    {
      "auxiliary_loss_clip": 0.01119538,
      "auxiliary_loss_mlp": 0.01039975,
      "balance_loss_clip": 1.05036557,
      "balance_loss_mlp": 1.02506816,
      "epoch": 0.3243048248910266,
      "flos": 24351205455360.0,
      "grad_norm": 1.5329086985066331,
      "language_loss": 0.82812119,
      "learning_rate": 3.158142199443371e-06,
      "loss": 0.84971631,
      "num_input_tokens_seen": 115846955,
      "step": 5394,
      "time_per_iteration": 2.535346031188965
    },
    {
      "auxiliary_loss_clip": 0.01111743,
      "auxiliary_loss_mlp": 0.01046653,
      "balance_loss_clip": 1.04980302,
      "balance_loss_mlp": 1.03200173,
      "epoch": 0.3243649481436946,
      "flos": 24353037048960.0,
      "grad_norm": 1.6991828606616228,
      "language_loss": 0.82083082,
      "learning_rate": 3.1578246583178076e-06,
      "loss": 0.84241474,
      "num_input_tokens_seen": 115865975,
      "step": 5395,
      "time_per_iteration": 2.540802478790283
    },
    {
      "auxiliary_loss_clip": 0.01122544,
      "auxiliary_loss_mlp": 0.01042073,
      "balance_loss_clip": 1.04965329,
      "balance_loss_mlp": 1.02741051,
      "epoch": 0.32442507139636256,
      "flos": 22925228451840.0,
      "grad_norm": 1.7295680706852552,
      "language_loss": 0.83241618,
      "learning_rate": 3.157507073287417e-06,
      "loss": 0.85406232,
      "num_input_tokens_seen": 115884950,
      "step": 5396,
      "time_per_iteration": 2.523560047149658
    },
    {
      "auxiliary_loss_clip": 0.0110063,
      "auxiliary_loss_mlp": 0.01048673,
      "balance_loss_clip": 1.04907346,
      "balance_loss_mlp": 1.03101778,
      "epoch": 0.32448519464903053,
      "flos": 22200192426240.0,
      "grad_norm": 1.7984303761815954,
      "language_loss": 0.7567296,
      "learning_rate": 3.1571894443642414e-06,
      "loss": 0.77822262,
      "num_input_tokens_seen": 115904170,
      "step": 5397,
      "time_per_iteration": 2.5513064861297607
    },
    {
      "auxiliary_loss_clip": 0.01101676,
      "auxiliary_loss_mlp": 0.01032844,
      "balance_loss_clip": 1.04938245,
      "balance_loss_mlp": 1.01792479,
      "epoch": 0.3245453179016985,
      "flos": 18838450644480.0,
      "grad_norm": 2.2606459417251124,
      "language_loss": 0.67263281,
      "learning_rate": 3.1568717715603263e-06,
      "loss": 0.69397807,
      "num_input_tokens_seen": 115919255,
      "step": 5398,
      "time_per_iteration": 2.5438036918640137
    },
    {
      "auxiliary_loss_clip": 0.01108012,
      "auxiliary_loss_mlp": 0.01037053,
      "balance_loss_clip": 1.04749513,
      "balance_loss_mlp": 1.02154374,
      "epoch": 0.32460544115436646,
      "flos": 21178390233600.0,
      "grad_norm": 1.3847606855341306,
      "language_loss": 0.73305732,
      "learning_rate": 3.156554054887718e-06,
      "loss": 0.75450802,
      "num_input_tokens_seen": 115938535,
      "step": 5399,
      "time_per_iteration": 2.568788766860962
    },
    {
      "auxiliary_loss_clip": 0.01100833,
      "auxiliary_loss_mlp": 0.01042608,
      "balance_loss_clip": 1.04767179,
      "balance_loss_mlp": 1.02678919,
      "epoch": 0.3246655644070344,
      "flos": 21981137333760.0,
      "grad_norm": 2.02864900707376,
      "language_loss": 0.71010518,
      "learning_rate": 3.1562362943584645e-06,
      "loss": 0.73153961,
      "num_input_tokens_seen": 115955005,
      "step": 5400,
      "time_per_iteration": 2.567824602127075
    },
    {
      "auxiliary_loss_clip": 0.0112763,
      "auxiliary_loss_mlp": 0.01038398,
      "balance_loss_clip": 1.04844737,
      "balance_loss_mlp": 1.02331781,
      "epoch": 0.3247256876597024,
      "flos": 32159729105280.0,
      "grad_norm": 1.9672805156507402,
      "language_loss": 0.79801124,
      "learning_rate": 3.155918489984614e-06,
      "loss": 0.81967151,
      "num_input_tokens_seen": 115975305,
      "step": 5401,
      "time_per_iteration": 3.9444713592529297
    },
    {
      "auxiliary_loss_clip": 0.01109869,
      "auxiliary_loss_mlp": 0.0104594,
      "balance_loss_clip": 1.04949403,
      "balance_loss_mlp": 1.02938235,
      "epoch": 0.32478581091237035,
      "flos": 20997544233600.0,
      "grad_norm": 1.4911067207729836,
      "language_loss": 0.87500536,
      "learning_rate": 3.1556006417782196e-06,
      "loss": 0.89656341,
      "num_input_tokens_seen": 115994810,
      "step": 5402,
      "time_per_iteration": 2.5793704986572266
    },
    {
      "auxiliary_loss_clip": 0.01078728,
      "auxiliary_loss_mlp": 0.01047086,
      "balance_loss_clip": 1.04276025,
      "balance_loss_mlp": 1.03030133,
      "epoch": 0.3248459341650383,
      "flos": 17924990849280.0,
      "grad_norm": 1.8597282856031376,
      "language_loss": 0.84581208,
      "learning_rate": 3.155282749751332e-06,
      "loss": 0.8670702,
      "num_input_tokens_seen": 116011095,
      "step": 5403,
      "time_per_iteration": 2.5595030784606934
    },
    {
      "auxiliary_loss_clip": 0.01104991,
      "auxiliary_loss_mlp": 0.01041625,
      "balance_loss_clip": 1.04949415,
      "balance_loss_mlp": 1.0271349,
      "epoch": 0.3249060574177063,
      "flos": 24535606901760.0,
      "grad_norm": 2.1034827846191497,
      "language_loss": 0.87582874,
      "learning_rate": 3.154964813916007e-06,
      "loss": 0.89729482,
      "num_input_tokens_seen": 116028805,
      "step": 5404,
      "time_per_iteration": 2.5495753288269043
    },
    {
      "auxiliary_loss_clip": 0.011242,
      "auxiliary_loss_mlp": 0.01039234,
      "balance_loss_clip": 1.0506103,
      "balance_loss_mlp": 1.02454805,
      "epoch": 0.32496618067037425,
      "flos": 25994765093760.0,
      "grad_norm": 1.6080156173463307,
      "language_loss": 0.72640991,
      "learning_rate": 3.1546468342843008e-06,
      "loss": 0.74804419,
      "num_input_tokens_seen": 116047765,
      "step": 5405,
      "time_per_iteration": 3.980363607406616
    },
    {
      "auxiliary_loss_clip": 0.01095306,
      "auxiliary_loss_mlp": 0.0103583,
      "balance_loss_clip": 1.05306613,
      "balance_loss_mlp": 1.02086926,
      "epoch": 0.3250263039230422,
      "flos": 19573757959680.0,
      "grad_norm": 1.7464790925081126,
      "language_loss": 0.83020937,
      "learning_rate": 3.1543288108682707e-06,
      "loss": 0.85152066,
      "num_input_tokens_seen": 116068385,
      "step": 5406,
      "time_per_iteration": 2.584517240524292
    },
    {
      "auxiliary_loss_clip": 0.01134825,
      "auxiliary_loss_mlp": 0.01032415,
      "balance_loss_clip": 1.05068731,
      "balance_loss_mlp": 1.01855707,
      "epoch": 0.3250864271757102,
      "flos": 16763640318720.0,
      "grad_norm": 1.6595417558651204,
      "language_loss": 0.87602508,
      "learning_rate": 3.1540107436799764e-06,
      "loss": 0.89769745,
      "num_input_tokens_seen": 116085350,
      "step": 5407,
      "time_per_iteration": 3.7864737510681152
    },
    {
      "auxiliary_loss_clip": 0.01112192,
      "auxiliary_loss_mlp": 0.01039493,
      "balance_loss_clip": 1.05003524,
      "balance_loss_mlp": 1.02455592,
      "epoch": 0.3251465504283782,
      "flos": 27819458040960.0,
      "grad_norm": 1.311560653971944,
      "language_loss": 0.69671822,
      "learning_rate": 3.153692632731479e-06,
      "loss": 0.71823502,
      "num_input_tokens_seen": 116107560,
      "step": 5408,
      "time_per_iteration": 4.010082244873047
    },
    {
      "auxiliary_loss_clip": 0.01126731,
      "auxiliary_loss_mlp": 0.01032743,
      "balance_loss_clip": 1.04720068,
      "balance_loss_mlp": 1.01800895,
      "epoch": 0.32520667368104617,
      "flos": 19063144172160.0,
      "grad_norm": 1.7032492064776585,
      "language_loss": 0.77659476,
      "learning_rate": 3.153374478034841e-06,
      "loss": 0.79818946,
      "num_input_tokens_seen": 116125980,
      "step": 5409,
      "time_per_iteration": 2.4756855964660645
    },
    {
      "auxiliary_loss_clip": 0.01079239,
      "auxiliary_loss_mlp": 0.0104366,
      "balance_loss_clip": 1.04248738,
      "balance_loss_mlp": 1.02811515,
      "epoch": 0.32526679693371413,
      "flos": 29382146208000.0,
      "grad_norm": 1.8782008808424644,
      "language_loss": 0.83318877,
      "learning_rate": 3.1530562796021285e-06,
      "loss": 0.85441774,
      "num_input_tokens_seen": 116146530,
      "step": 5410,
      "time_per_iteration": 2.66770601272583
    },
    {
      "auxiliary_loss_clip": 0.01084414,
      "auxiliary_loss_mlp": 0.01035035,
      "balance_loss_clip": 1.0484302,
      "balance_loss_mlp": 1.02063417,
      "epoch": 0.3253269201863821,
      "flos": 20704513080960.0,
      "grad_norm": 1.6059131418251358,
      "language_loss": 0.71326685,
      "learning_rate": 3.152738037445405e-06,
      "loss": 0.73446137,
      "num_input_tokens_seen": 116165695,
      "step": 5411,
      "time_per_iteration": 2.596151828765869
    },
    {
      "auxiliary_loss_clip": 0.01091775,
      "auxiliary_loss_mlp": 0.01040423,
      "balance_loss_clip": 1.05682755,
      "balance_loss_mlp": 1.02633262,
      "epoch": 0.32538704343905006,
      "flos": 29094142959360.0,
      "grad_norm": 1.490604705913697,
      "language_loss": 0.83218026,
      "learning_rate": 3.1524197515767403e-06,
      "loss": 0.85350227,
      "num_input_tokens_seen": 116185375,
      "step": 5412,
      "time_per_iteration": 2.6373658180236816
    },
    {
      "auxiliary_loss_clip": 0.01102876,
      "auxiliary_loss_mlp": 0.01037716,
      "balance_loss_clip": 1.04616272,
      "balance_loss_mlp": 1.02163482,
      "epoch": 0.325447166691718,
      "flos": 24676124906880.0,
      "grad_norm": 1.7095817306612302,
      "language_loss": 0.81044722,
      "learning_rate": 3.152101422008203e-06,
      "loss": 0.83185315,
      "num_input_tokens_seen": 116204335,
      "step": 5413,
      "time_per_iteration": 2.5852599143981934
    },
    {
      "auxiliary_loss_clip": 0.01110685,
      "auxiliary_loss_mlp": 0.01036913,
      "balance_loss_clip": 1.04697657,
      "balance_loss_mlp": 1.02099824,
      "epoch": 0.325507289944386,
      "flos": 21543134889600.0,
      "grad_norm": 1.5573136731519273,
      "language_loss": 0.76794767,
      "learning_rate": 3.151783048751864e-06,
      "loss": 0.78942358,
      "num_input_tokens_seen": 116222840,
      "step": 5414,
      "time_per_iteration": 2.5623910427093506
    },
    {
      "auxiliary_loss_clip": 0.01020807,
      "auxiliary_loss_mlp": 0.01010425,
      "balance_loss_clip": 1.02352691,
      "balance_loss_mlp": 1.00862527,
      "epoch": 0.32556741319705396,
      "flos": 71518722347520.0,
      "grad_norm": 0.9085066961436206,
      "language_loss": 0.64000785,
      "learning_rate": 3.1514646318197965e-06,
      "loss": 0.66032016,
      "num_input_tokens_seen": 116274940,
      "step": 5415,
      "time_per_iteration": 3.066683292388916
    },
    {
      "auxiliary_loss_clip": 0.01089423,
      "auxiliary_loss_mlp": 0.01040066,
      "balance_loss_clip": 1.04952765,
      "balance_loss_mlp": 1.02417564,
      "epoch": 0.3256275364497219,
      "flos": 23732428838400.0,
      "grad_norm": 2.155493165527598,
      "language_loss": 0.7401213,
      "learning_rate": 3.151146171224075e-06,
      "loss": 0.7614162,
      "num_input_tokens_seen": 116297300,
      "step": 5416,
      "time_per_iteration": 2.6062815189361572
    },
    {
      "auxiliary_loss_clip": 0.01045711,
      "auxiliary_loss_mlp": 0.01004378,
      "balance_loss_clip": 1.01606297,
      "balance_loss_mlp": 1.00255382,
      "epoch": 0.3256876597023899,
      "flos": 67289199891840.0,
      "grad_norm": 0.7789619128187606,
      "language_loss": 0.57983875,
      "learning_rate": 3.1508276669767757e-06,
      "loss": 0.60033959,
      "num_input_tokens_seen": 116362370,
      "step": 5417,
      "time_per_iteration": 3.1594011783599854
    },
    {
      "auxiliary_loss_clip": 0.01027927,
      "auxiliary_loss_mlp": 0.01003781,
      "balance_loss_clip": 1.01858234,
      "balance_loss_mlp": 1.0017184,
      "epoch": 0.32574778295505785,
      "flos": 71282323964160.0,
      "grad_norm": 0.8042363563572404,
      "language_loss": 0.63395613,
      "learning_rate": 3.150509119089975e-06,
      "loss": 0.65427321,
      "num_input_tokens_seen": 116430365,
      "step": 5418,
      "time_per_iteration": 3.233729362487793
    },
    {
      "auxiliary_loss_clip": 0.0111788,
      "auxiliary_loss_mlp": 0.01042594,
      "balance_loss_clip": 1.05879796,
      "balance_loss_mlp": 1.02790761,
      "epoch": 0.3258079062077258,
      "flos": 20776370238720.0,
      "grad_norm": 2.10447166520173,
      "language_loss": 0.69397229,
      "learning_rate": 3.1501905275757537e-06,
      "loss": 0.71557701,
      "num_input_tokens_seen": 116447525,
      "step": 5419,
      "time_per_iteration": 2.5184731483459473
    },
    {
      "auxiliary_loss_clip": 0.0113196,
      "auxiliary_loss_mlp": 0.01038813,
      "balance_loss_clip": 1.05446255,
      "balance_loss_mlp": 1.02288711,
      "epoch": 0.3258680294603938,
      "flos": 22235456603520.0,
      "grad_norm": 1.6998999618423507,
      "language_loss": 0.77501321,
      "learning_rate": 3.1498718924461926e-06,
      "loss": 0.79672098,
      "num_input_tokens_seen": 116466310,
      "step": 5420,
      "time_per_iteration": 2.5011355876922607
    },
    {
      "auxiliary_loss_clip": 0.01122544,
      "auxiliary_loss_mlp": 0.00801869,
      "balance_loss_clip": 1.04578781,
      "balance_loss_mlp": 1.02611065,
      "epoch": 0.3259281527130618,
      "flos": 26979974305920.0,
      "grad_norm": 1.4607310284309045,
      "language_loss": 0.80082506,
      "learning_rate": 3.1495532137133736e-06,
      "loss": 0.82006919,
      "num_input_tokens_seen": 116487825,
      "step": 5421,
      "time_per_iteration": 2.535216808319092
    },
    {
      "auxiliary_loss_clip": 0.01132063,
      "auxiliary_loss_mlp": 0.01038576,
      "balance_loss_clip": 1.04777586,
      "balance_loss_mlp": 1.02429509,
      "epoch": 0.32598827596572977,
      "flos": 26214251149440.0,
      "grad_norm": 2.2485898833957925,
      "language_loss": 0.75884289,
      "learning_rate": 3.149234491389381e-06,
      "loss": 0.78054929,
      "num_input_tokens_seen": 116509950,
      "step": 5422,
      "time_per_iteration": 2.5255470275878906
    },
    {
      "auxiliary_loss_clip": 0.01101564,
      "auxiliary_loss_mlp": 0.00797715,
      "balance_loss_clip": 1.04816091,
      "balance_loss_mlp": 1.02164102,
      "epoch": 0.32604839921839773,
      "flos": 17639752947840.0,
      "grad_norm": 2.1068187865673935,
      "language_loss": 0.62819993,
      "learning_rate": 3.1489157254863026e-06,
      "loss": 0.64719272,
      "num_input_tokens_seen": 116527695,
      "step": 5423,
      "time_per_iteration": 2.5277020931243896
    },
    {
      "auxiliary_loss_clip": 0.01097629,
      "auxiliary_loss_mlp": 0.01035189,
      "balance_loss_clip": 1.04460049,
      "balance_loss_mlp": 1.02171874,
      "epoch": 0.3261085224710657,
      "flos": 23622721724160.0,
      "grad_norm": 1.660461862129649,
      "language_loss": 0.74589765,
      "learning_rate": 3.148596916016224e-06,
      "loss": 0.7672258,
      "num_input_tokens_seen": 116547800,
      "step": 5424,
      "time_per_iteration": 2.5676016807556152
    },
    {
      "auxiliary_loss_clip": 0.01103056,
      "auxiliary_loss_mlp": 0.01035086,
      "balance_loss_clip": 1.04837132,
      "balance_loss_mlp": 1.02098346,
      "epoch": 0.32616864572373366,
      "flos": 23260455106560.0,
      "grad_norm": 1.6098004539572819,
      "language_loss": 0.77129918,
      "learning_rate": 3.1482780629912355e-06,
      "loss": 0.79268062,
      "num_input_tokens_seen": 116568460,
      "step": 5425,
      "time_per_iteration": 2.549919843673706
    },
    {
      "auxiliary_loss_clip": 0.01105564,
      "auxiliary_loss_mlp": 0.01044419,
      "balance_loss_clip": 1.04815006,
      "balance_loss_mlp": 1.02777767,
      "epoch": 0.32622876897640163,
      "flos": 25593427457280.0,
      "grad_norm": 2.215612796829964,
      "language_loss": 0.77805722,
      "learning_rate": 3.147959166423428e-06,
      "loss": 0.79955697,
      "num_input_tokens_seen": 116588705,
      "step": 5426,
      "time_per_iteration": 2.604257345199585
    },
    {
      "auxiliary_loss_clip": 0.01088375,
      "auxiliary_loss_mlp": 0.01039826,
      "balance_loss_clip": 1.04580176,
      "balance_loss_mlp": 1.02336287,
      "epoch": 0.3262888922290696,
      "flos": 22418996123520.0,
      "grad_norm": 1.712001980523705,
      "language_loss": 0.7452848,
      "learning_rate": 3.147640226324893e-06,
      "loss": 0.76656681,
      "num_input_tokens_seen": 116608845,
      "step": 5427,
      "time_per_iteration": 2.6226346492767334
    },
    {
      "auxiliary_loss_clip": 0.01101648,
      "auxiliary_loss_mlp": 0.01047673,
      "balance_loss_clip": 1.04464817,
      "balance_loss_mlp": 1.03185356,
      "epoch": 0.32634901548173756,
      "flos": 19718908819200.0,
      "grad_norm": 1.5602753641467806,
      "language_loss": 0.79263842,
      "learning_rate": 3.1473212427077266e-06,
      "loss": 0.81413168,
      "num_input_tokens_seen": 116628145,
      "step": 5428,
      "time_per_iteration": 2.547834873199463
    },
    {
      "auxiliary_loss_clip": 0.01119582,
      "auxiliary_loss_mlp": 0.01041212,
      "balance_loss_clip": 1.04523325,
      "balance_loss_mlp": 1.02656114,
      "epoch": 0.3264091387344055,
      "flos": 16142924367360.0,
      "grad_norm": 1.5508952122638706,
      "language_loss": 0.71127367,
      "learning_rate": 3.147002215584023e-06,
      "loss": 0.73288161,
      "num_input_tokens_seen": 116646920,
      "step": 5429,
      "time_per_iteration": 2.4773359298706055
    },
    {
      "auxiliary_loss_clip": 0.01098365,
      "auxiliary_loss_mlp": 0.01039668,
      "balance_loss_clip": 1.04940236,
      "balance_loss_mlp": 1.02560127,
      "epoch": 0.3264692619870735,
      "flos": 16399075230720.0,
      "grad_norm": 1.6249751062640256,
      "language_loss": 0.78498638,
      "learning_rate": 3.146683144965881e-06,
      "loss": 0.80636668,
      "num_input_tokens_seen": 116665100,
      "step": 5430,
      "time_per_iteration": 2.552614450454712
    },
    {
      "auxiliary_loss_clip": 0.01084611,
      "auxiliary_loss_mlp": 0.01036515,
      "balance_loss_clip": 1.05107343,
      "balance_loss_mlp": 1.02036238,
      "epoch": 0.32652938523974145,
      "flos": 22382331315840.0,
      "grad_norm": 2.0332998859557017,
      "language_loss": 0.8394624,
      "learning_rate": 3.146364030865399e-06,
      "loss": 0.86067367,
      "num_input_tokens_seen": 116682205,
      "step": 5431,
      "time_per_iteration": 2.6155076026916504
    },
    {
      "auxiliary_loss_clip": 0.01116859,
      "auxiliary_loss_mlp": 0.01038375,
      "balance_loss_clip": 1.04940915,
      "balance_loss_mlp": 1.02385497,
      "epoch": 0.3265895084924094,
      "flos": 21908059113600.0,
      "grad_norm": 1.9017484206687336,
      "language_loss": 0.70642662,
      "learning_rate": 3.146044873294678e-06,
      "loss": 0.72797894,
      "num_input_tokens_seen": 116702575,
      "step": 5432,
      "time_per_iteration": 2.527967929840088
    },
    {
      "auxiliary_loss_clip": 0.0107445,
      "auxiliary_loss_mlp": 0.01041284,
      "balance_loss_clip": 1.04551268,
      "balance_loss_mlp": 1.0251193,
      "epoch": 0.3266496317450774,
      "flos": 16067152627200.0,
      "grad_norm": 1.369103485547817,
      "language_loss": 0.84196103,
      "learning_rate": 3.1457256722658203e-06,
      "loss": 0.86311841,
      "num_input_tokens_seen": 116720885,
      "step": 5433,
      "time_per_iteration": 2.610870122909546
    },
    {
      "auxiliary_loss_clip": 0.01111584,
      "auxiliary_loss_mlp": 0.01034753,
      "balance_loss_clip": 1.05118942,
      "balance_loss_mlp": 1.01999533,
      "epoch": 0.3267097549977454,
      "flos": 22528236360960.0,
      "grad_norm": 1.3559662597362263,
      "language_loss": 0.85867852,
      "learning_rate": 3.145406427790931e-06,
      "loss": 0.88014185,
      "num_input_tokens_seen": 116740395,
      "step": 5434,
      "time_per_iteration": 2.5827043056488037
    },
    {
      "auxiliary_loss_clip": 0.01112306,
      "auxiliary_loss_mlp": 0.01037931,
      "balance_loss_clip": 1.04578376,
      "balance_loss_mlp": 1.02221906,
      "epoch": 0.32676987825041337,
      "flos": 27270419679360.0,
      "grad_norm": 1.7947325827272111,
      "language_loss": 0.87822509,
      "learning_rate": 3.1450871398821147e-06,
      "loss": 0.8997274,
      "num_input_tokens_seen": 116758870,
      "step": 5435,
      "time_per_iteration": 2.5690951347351074
    },
    {
      "auxiliary_loss_clip": 0.01131822,
      "auxiliary_loss_mlp": 0.01035188,
      "balance_loss_clip": 1.04706204,
      "balance_loss_mlp": 1.02027452,
      "epoch": 0.32683000150308134,
      "flos": 11508257433600.0,
      "grad_norm": 2.5481723738390882,
      "language_loss": 0.76462311,
      "learning_rate": 3.144767808551479e-06,
      "loss": 0.78629321,
      "num_input_tokens_seen": 116773440,
      "step": 5436,
      "time_per_iteration": 2.4680943489074707
    },
    {
      "auxiliary_loss_clip": 0.01133053,
      "auxiliary_loss_mlp": 0.01038672,
      "balance_loss_clip": 1.04929042,
      "balance_loss_mlp": 1.02405143,
      "epoch": 0.3268901247557493,
      "flos": 25630200005760.0,
      "grad_norm": 1.5794215836469658,
      "language_loss": 0.72003484,
      "learning_rate": 3.144448433811134e-06,
      "loss": 0.74175215,
      "num_input_tokens_seen": 116794375,
      "step": 5437,
      "time_per_iteration": 2.5115909576416016
    },
    {
      "auxiliary_loss_clip": 0.01100683,
      "auxiliary_loss_mlp": 0.01039723,
      "balance_loss_clip": 1.04990578,
      "balance_loss_mlp": 1.0233438,
      "epoch": 0.32695024800841727,
      "flos": 24860849575680.0,
      "grad_norm": 1.7105248071197225,
      "language_loss": 0.63879526,
      "learning_rate": 3.144129015673189e-06,
      "loss": 0.66019928,
      "num_input_tokens_seen": 116815095,
      "step": 5438,
      "time_per_iteration": 2.5647690296173096
    },
    {
      "auxiliary_loss_clip": 0.0112281,
      "auxiliary_loss_mlp": 0.01038046,
      "balance_loss_clip": 1.04870951,
      "balance_loss_mlp": 1.02248907,
      "epoch": 0.32701037126108523,
      "flos": 28839249072000.0,
      "grad_norm": 2.366430480314041,
      "language_loss": 0.74460363,
      "learning_rate": 3.1438095541497576e-06,
      "loss": 0.76621211,
      "num_input_tokens_seen": 116836630,
      "step": 5439,
      "time_per_iteration": 2.5459823608398438
    },
    {
      "auxiliary_loss_clip": 0.0112639,
      "auxiliary_loss_mlp": 0.01048965,
      "balance_loss_clip": 1.0502249,
      "balance_loss_mlp": 1.03284156,
      "epoch": 0.3270704945137532,
      "flos": 27965075777280.0,
      "grad_norm": 1.9649315998866363,
      "language_loss": 0.74867594,
      "learning_rate": 3.1434900492529527e-06,
      "loss": 0.77042949,
      "num_input_tokens_seen": 116856880,
      "step": 5440,
      "time_per_iteration": 3.930870532989502
    },
    {
      "auxiliary_loss_clip": 0.01119808,
      "auxiliary_loss_mlp": 0.00798639,
      "balance_loss_clip": 1.04802608,
      "balance_loss_mlp": 1.01993322,
      "epoch": 0.32713061776642116,
      "flos": 23690700213120.0,
      "grad_norm": 1.9309873447422525,
      "language_loss": 0.85071123,
      "learning_rate": 3.1431705009948914e-06,
      "loss": 0.8698957,
      "num_input_tokens_seen": 116873770,
      "step": 5441,
      "time_per_iteration": 2.497464179992676
    },
    {
      "auxiliary_loss_clip": 0.01119024,
      "auxiliary_loss_mlp": 0.01042892,
      "balance_loss_clip": 1.04390383,
      "balance_loss_mlp": 1.02683449,
      "epoch": 0.3271907410190891,
      "flos": 22455625017600.0,
      "grad_norm": 1.9894046400053618,
      "language_loss": 0.86729056,
      "learning_rate": 3.1428509093876897e-06,
      "loss": 0.88890976,
      "num_input_tokens_seen": 116891225,
      "step": 5442,
      "time_per_iteration": 2.4799108505249023
    },
    {
      "auxiliary_loss_clip": 0.01104169,
      "auxiliary_loss_mlp": 0.01042778,
      "balance_loss_clip": 1.05224967,
      "balance_loss_mlp": 1.02576685,
      "epoch": 0.3272508642717571,
      "flos": 22820118278400.0,
      "grad_norm": 2.513213694436785,
      "language_loss": 0.77500743,
      "learning_rate": 3.1425312744434668e-06,
      "loss": 0.79647684,
      "num_input_tokens_seen": 116912300,
      "step": 5443,
      "time_per_iteration": 3.9552462100982666
    },
    {
      "auxiliary_loss_clip": 0.01104283,
      "auxiliary_loss_mlp": 0.00797833,
      "balance_loss_clip": 1.04864311,
      "balance_loss_mlp": 1.01857197,
      "epoch": 0.32731098752442506,
      "flos": 11801360413440.0,
      "grad_norm": 2.0536627263236076,
      "language_loss": 0.81485456,
      "learning_rate": 3.142211596174343e-06,
      "loss": 0.83387572,
      "num_input_tokens_seen": 116929425,
      "step": 5444,
      "time_per_iteration": 2.5245606899261475
    },
    {
      "auxiliary_loss_clip": 0.01090457,
      "auxiliary_loss_mlp": 0.01039387,
      "balance_loss_clip": 1.04617047,
      "balance_loss_mlp": 1.02422345,
      "epoch": 0.327371110777093,
      "flos": 21027780506880.0,
      "grad_norm": 1.9770537238621573,
      "language_loss": 0.5920319,
      "learning_rate": 3.1418918745924423e-06,
      "loss": 0.61333036,
      "num_input_tokens_seen": 116948255,
      "step": 5445,
      "time_per_iteration": 2.598482131958008
    },
    {
      "auxiliary_loss_clip": 0.01126528,
      "auxiliary_loss_mlp": 0.01044067,
      "balance_loss_clip": 1.050138,
      "balance_loss_mlp": 1.02768743,
      "epoch": 0.327431234029761,
      "flos": 19062102677760.0,
      "grad_norm": 2.238430077055999,
      "language_loss": 0.88429976,
      "learning_rate": 3.1415721097098865e-06,
      "loss": 0.90600574,
      "num_input_tokens_seen": 116964905,
      "step": 5446,
      "time_per_iteration": 3.8755714893341064
    },
    {
      "auxiliary_loss_clip": 0.01123276,
      "auxiliary_loss_mlp": 0.01045038,
      "balance_loss_clip": 1.04955482,
      "balance_loss_mlp": 1.02756238,
      "epoch": 0.32749135728242895,
      "flos": 25849219184640.0,
      "grad_norm": 1.5563423811742216,
      "language_loss": 0.78900486,
      "learning_rate": 3.141252301538802e-06,
      "loss": 0.81068802,
      "num_input_tokens_seen": 116983650,
      "step": 5447,
      "time_per_iteration": 3.9296629428863525
    },
    {
      "auxiliary_loss_clip": 0.0110489,
      "auxiliary_loss_mlp": 0.00794229,
      "balance_loss_clip": 1.04587197,
      "balance_loss_mlp": 1.01126957,
      "epoch": 0.327551480535097,
      "flos": 20120533764480.0,
      "grad_norm": 1.7718813235975865,
      "language_loss": 0.73357046,
      "learning_rate": 3.1409324500913157e-06,
      "loss": 0.75256169,
      "num_input_tokens_seen": 117003265,
      "step": 5448,
      "time_per_iteration": 2.5481104850769043
    },
    {
      "auxiliary_loss_clip": 0.01134639,
      "auxiliary_loss_mlp": 0.0104581,
      "balance_loss_clip": 1.04820228,
      "balance_loss_mlp": 1.02959728,
      "epoch": 0.32761160378776494,
      "flos": 28803553931520.0,
      "grad_norm": 1.3685217537393823,
      "language_loss": 0.66924495,
      "learning_rate": 3.1406125553795567e-06,
      "loss": 0.6910494,
      "num_input_tokens_seen": 117025370,
      "step": 5449,
      "time_per_iteration": 2.5778377056121826
    },
    {
      "auxiliary_loss_clip": 0.01099978,
      "auxiliary_loss_mlp": 0.01038448,
      "balance_loss_clip": 1.04685152,
      "balance_loss_mlp": 1.02384531,
      "epoch": 0.3276717270404329,
      "flos": 26937778803840.0,
      "grad_norm": 1.443796868871506,
      "language_loss": 0.65302283,
      "learning_rate": 3.1402926174156556e-06,
      "loss": 0.67440712,
      "num_input_tokens_seen": 117044350,
      "step": 5450,
      "time_per_iteration": 2.6024248600006104
    },
    {
      "auxiliary_loss_clip": 0.01131805,
      "auxiliary_loss_mlp": 0.01046314,
      "balance_loss_clip": 1.05242252,
      "balance_loss_mlp": 1.03031635,
      "epoch": 0.32773185029310087,
      "flos": 25338425829120.0,
      "grad_norm": 1.4275192776950087,
      "language_loss": 0.77370751,
      "learning_rate": 3.1399726362117437e-06,
      "loss": 0.79548866,
      "num_input_tokens_seen": 117064450,
      "step": 5451,
      "time_per_iteration": 2.531707286834717
    },
    {
      "auxiliary_loss_clip": 0.01124713,
      "auxiliary_loss_mlp": 0.01045606,
      "balance_loss_clip": 1.04846179,
      "balance_loss_mlp": 1.0289284,
      "epoch": 0.32779197354576883,
      "flos": 26391721271040.0,
      "grad_norm": 2.0463258689533883,
      "language_loss": 0.70519453,
      "learning_rate": 3.1396526117799555e-06,
      "loss": 0.72689772,
      "num_input_tokens_seen": 117083060,
      "step": 5452,
      "time_per_iteration": 2.5307023525238037
    },
    {
      "auxiliary_loss_clip": 0.01108002,
      "auxiliary_loss_mlp": 0.01038461,
      "balance_loss_clip": 1.04552972,
      "balance_loss_mlp": 1.02330375,
      "epoch": 0.3278520967984368,
      "flos": 24899381890560.0,
      "grad_norm": 1.5988798846792438,
      "language_loss": 0.78585583,
      "learning_rate": 3.1393325441324256e-06,
      "loss": 0.80732048,
      "num_input_tokens_seen": 117101860,
      "step": 5453,
      "time_per_iteration": 2.560089349746704
    },
    {
      "auxiliary_loss_clip": 0.01128259,
      "auxiliary_loss_mlp": 0.0103206,
      "balance_loss_clip": 1.05044651,
      "balance_loss_mlp": 1.01687837,
      "epoch": 0.32791222005110476,
      "flos": 29752996176000.0,
      "grad_norm": 1.9969561284658575,
      "language_loss": 0.75538445,
      "learning_rate": 3.1390124332812916e-06,
      "loss": 0.77698767,
      "num_input_tokens_seen": 117123100,
      "step": 5454,
      "time_per_iteration": 2.5719680786132812
    },
    {
      "auxiliary_loss_clip": 0.01073759,
      "auxiliary_loss_mlp": 0.01042311,
      "balance_loss_clip": 1.04614854,
      "balance_loss_mlp": 1.02767217,
      "epoch": 0.32797234330377273,
      "flos": 16508064072960.0,
      "grad_norm": 1.8239852247166373,
      "language_loss": 0.77065706,
      "learning_rate": 3.1386922792386924e-06,
      "loss": 0.79181778,
      "num_input_tokens_seen": 117140515,
      "step": 5455,
      "time_per_iteration": 2.5900113582611084
    },
    {
      "auxiliary_loss_clip": 0.01134019,
      "auxiliary_loss_mlp": 0.01043351,
      "balance_loss_clip": 1.0518527,
      "balance_loss_mlp": 1.02686453,
      "epoch": 0.3280324665564407,
      "flos": 26577918397440.0,
      "grad_norm": 1.6660752178931932,
      "language_loss": 0.73867655,
      "learning_rate": 3.138372082016768e-06,
      "loss": 0.76045024,
      "num_input_tokens_seen": 117161485,
      "step": 5456,
      "time_per_iteration": 2.560990333557129
    },
    {
      "auxiliary_loss_clip": 0.01137242,
      "auxiliary_loss_mlp": 0.0104692,
      "balance_loss_clip": 1.05023348,
      "balance_loss_mlp": 1.03167319,
      "epoch": 0.32809258980910866,
      "flos": 22929969047040.0,
      "grad_norm": 1.4650735136819832,
      "language_loss": 0.78203988,
      "learning_rate": 3.1380518416276596e-06,
      "loss": 0.80388147,
      "num_input_tokens_seen": 117181870,
      "step": 5457,
      "time_per_iteration": 2.490001678466797
    },
    {
      "auxiliary_loss_clip": 0.01100581,
      "auxiliary_loss_mlp": 0.01040461,
      "balance_loss_clip": 1.04724693,
      "balance_loss_mlp": 1.02556038,
      "epoch": 0.3281527130617766,
      "flos": 22783848520320.0,
      "grad_norm": 2.260344235111836,
      "language_loss": 0.78703022,
      "learning_rate": 3.1377315580835115e-06,
      "loss": 0.80844069,
      "num_input_tokens_seen": 117201380,
      "step": 5458,
      "time_per_iteration": 2.5651304721832275
    },
    {
      "auxiliary_loss_clip": 0.01121333,
      "auxiliary_loss_mlp": 0.01039801,
      "balance_loss_clip": 1.05292201,
      "balance_loss_mlp": 1.02433956,
      "epoch": 0.3282128363144446,
      "flos": 21250678354560.0,
      "grad_norm": 1.6807926449982848,
      "language_loss": 0.73032379,
      "learning_rate": 3.1374112313964686e-06,
      "loss": 0.75193512,
      "num_input_tokens_seen": 117221040,
      "step": 5459,
      "time_per_iteration": 2.5186305046081543
    },
    {
      "auxiliary_loss_clip": 0.01117413,
      "auxiliary_loss_mlp": 0.01040085,
      "balance_loss_clip": 1.05157149,
      "balance_loss_mlp": 1.02517223,
      "epoch": 0.32827295956711255,
      "flos": 30843064166400.0,
      "grad_norm": 1.9309464297881194,
      "language_loss": 0.84114075,
      "learning_rate": 3.1370908615786783e-06,
      "loss": 0.86271572,
      "num_input_tokens_seen": 117241395,
      "step": 5460,
      "time_per_iteration": 2.6068003177642822
    },
    {
      "auxiliary_loss_clip": 0.01135913,
      "auxiliary_loss_mlp": 0.01036105,
      "balance_loss_clip": 1.04918694,
      "balance_loss_mlp": 1.02144241,
      "epoch": 0.3283330828197806,
      "flos": 25915006944000.0,
      "grad_norm": 1.8227541916770955,
      "language_loss": 0.77131879,
      "learning_rate": 3.136770448642288e-06,
      "loss": 0.79303896,
      "num_input_tokens_seen": 117259340,
      "step": 5461,
      "time_per_iteration": 2.518712043762207
    },
    {
      "auxiliary_loss_clip": 0.01120089,
      "auxiliary_loss_mlp": 0.01040536,
      "balance_loss_clip": 1.05017662,
      "balance_loss_mlp": 1.02315497,
      "epoch": 0.32839320607244854,
      "flos": 38582065042560.0,
      "grad_norm": 1.9794243956468336,
      "language_loss": 0.62745833,
      "learning_rate": 3.1364499925994484e-06,
      "loss": 0.64906454,
      "num_input_tokens_seen": 117282375,
      "step": 5462,
      "time_per_iteration": 2.6235275268554688
    },
    {
      "auxiliary_loss_clip": 0.01135017,
      "auxiliary_loss_mlp": 0.00799753,
      "balance_loss_clip": 1.05078566,
      "balance_loss_mlp": 1.02517772,
      "epoch": 0.3284533293251165,
      "flos": 26650888876800.0,
      "grad_norm": 1.362325422267679,
      "language_loss": 0.78294134,
      "learning_rate": 3.1361294934623115e-06,
      "loss": 0.80228907,
      "num_input_tokens_seen": 117303830,
      "step": 5463,
      "time_per_iteration": 2.5316519737243652
    },
    {
      "auxiliary_loss_clip": 0.01103147,
      "auxiliary_loss_mlp": 0.01039787,
      "balance_loss_clip": 1.04568648,
      "balance_loss_mlp": 1.02405155,
      "epoch": 0.32851345257778447,
      "flos": 15304158904320.0,
      "grad_norm": 1.8656324518084828,
      "language_loss": 0.69784272,
      "learning_rate": 3.1358089512430303e-06,
      "loss": 0.71927208,
      "num_input_tokens_seen": 117320665,
      "step": 5464,
      "time_per_iteration": 2.498030185699463
    },
    {
      "auxiliary_loss_clip": 0.01126028,
      "auxiliary_loss_mlp": 0.01039802,
      "balance_loss_clip": 1.05301952,
      "balance_loss_mlp": 1.02475786,
      "epoch": 0.32857357583045244,
      "flos": 23513732881920.0,
      "grad_norm": 1.6248117913851012,
      "language_loss": 0.72404552,
      "learning_rate": 3.1354883659537594e-06,
      "loss": 0.74570388,
      "num_input_tokens_seen": 117339795,
      "step": 5465,
      "time_per_iteration": 2.5114364624023438
    },
    {
      "auxiliary_loss_clip": 0.01111335,
      "auxiliary_loss_mlp": 0.010484,
      "balance_loss_clip": 1.05029297,
      "balance_loss_mlp": 1.03153169,
      "epoch": 0.3286336990831204,
      "flos": 20995209849600.0,
      "grad_norm": 1.4150920753040914,
      "language_loss": 0.82816505,
      "learning_rate": 3.1351677376066567e-06,
      "loss": 0.84976244,
      "num_input_tokens_seen": 117359525,
      "step": 5466,
      "time_per_iteration": 2.518624782562256
    },
    {
      "auxiliary_loss_clip": 0.01118214,
      "auxiliary_loss_mlp": 0.0103629,
      "balance_loss_clip": 1.05259144,
      "balance_loss_mlp": 1.02105546,
      "epoch": 0.32869382233578837,
      "flos": 23658811914240.0,
      "grad_norm": 1.919160953585714,
      "language_loss": 0.79567707,
      "learning_rate": 3.134847066213879e-06,
      "loss": 0.81722212,
      "num_input_tokens_seen": 117380320,
      "step": 5467,
      "time_per_iteration": 2.5562973022460938
    },
    {
      "auxiliary_loss_clip": 0.01112083,
      "auxiliary_loss_mlp": 0.01035639,
      "balance_loss_clip": 1.047364,
      "balance_loss_mlp": 1.02036786,
      "epoch": 0.32875394558845633,
      "flos": 25336522408320.0,
      "grad_norm": 1.5575206248847773,
      "language_loss": 0.74513215,
      "learning_rate": 3.134526351787587e-06,
      "loss": 0.76660943,
      "num_input_tokens_seen": 117400695,
      "step": 5468,
      "time_per_iteration": 2.560326337814331
    },
    {
      "auxiliary_loss_clip": 0.01112494,
      "auxiliary_loss_mlp": 0.0104384,
      "balance_loss_clip": 1.04939926,
      "balance_loss_mlp": 1.02666199,
      "epoch": 0.3288140688411243,
      "flos": 14903108576640.0,
      "grad_norm": 2.103159118987981,
      "language_loss": 0.78183293,
      "learning_rate": 3.134205594339942e-06,
      "loss": 0.80339634,
      "num_input_tokens_seen": 117418800,
      "step": 5469,
      "time_per_iteration": 2.506467819213867
    },
    {
      "auxiliary_loss_clip": 0.0110807,
      "auxiliary_loss_mlp": 0.01039681,
      "balance_loss_clip": 1.0493933,
      "balance_loss_mlp": 1.02528012,
      "epoch": 0.32887419209379226,
      "flos": 18551345235840.0,
      "grad_norm": 1.7485832352108162,
      "language_loss": 0.81797308,
      "learning_rate": 3.133884793883107e-06,
      "loss": 0.8394506,
      "num_input_tokens_seen": 117438220,
      "step": 5470,
      "time_per_iteration": 2.5368640422821045
    },
    {
      "auxiliary_loss_clip": 0.01135829,
      "auxiliary_loss_mlp": 0.01043031,
      "balance_loss_clip": 1.04765964,
      "balance_loss_mlp": 1.02747416,
      "epoch": 0.3289343153464602,
      "flos": 48105610439040.0,
      "grad_norm": 1.8208787496516754,
      "language_loss": 0.67756569,
      "learning_rate": 3.1335639504292478e-06,
      "loss": 0.69935435,
      "num_input_tokens_seen": 117462560,
      "step": 5471,
      "time_per_iteration": 2.688544511795044
    },
    {
      "auxiliary_loss_clip": 0.0114126,
      "auxiliary_loss_mlp": 0.01045364,
      "balance_loss_clip": 1.05116487,
      "balance_loss_mlp": 1.02809083,
      "epoch": 0.3289944385991282,
      "flos": 27600295207680.0,
      "grad_norm": 1.7151019450587666,
      "language_loss": 0.64985889,
      "learning_rate": 3.1332430639905288e-06,
      "loss": 0.67172515,
      "num_input_tokens_seen": 117483665,
      "step": 5472,
      "time_per_iteration": 2.493891477584839
    },
    {
      "auxiliary_loss_clip": 0.01122796,
      "auxiliary_loss_mlp": 0.01052658,
      "balance_loss_clip": 1.05241513,
      "balance_loss_mlp": 1.03561151,
      "epoch": 0.32905456185179616,
      "flos": 20120318282880.0,
      "grad_norm": 1.6980640494651191,
      "language_loss": 0.88304245,
      "learning_rate": 3.13292213457912e-06,
      "loss": 0.90479702,
      "num_input_tokens_seen": 117503565,
      "step": 5473,
      "time_per_iteration": 2.4941635131835938
    },
    {
      "auxiliary_loss_clip": 0.01099416,
      "auxiliary_loss_mlp": 0.01041865,
      "balance_loss_clip": 1.0486846,
      "balance_loss_mlp": 1.02469885,
      "epoch": 0.3291146851044642,
      "flos": 23180230080000.0,
      "grad_norm": 1.610818880910699,
      "language_loss": 0.77934992,
      "learning_rate": 3.1326011622071903e-06,
      "loss": 0.80076277,
      "num_input_tokens_seen": 117521460,
      "step": 5474,
      "time_per_iteration": 2.5681374073028564
    },
    {
      "auxiliary_loss_clip": 0.01033964,
      "auxiliary_loss_mlp": 0.01004579,
      "balance_loss_clip": 1.02615333,
      "balance_loss_mlp": 1.00250518,
      "epoch": 0.32917480835713214,
      "flos": 67621912594560.0,
      "grad_norm": 0.8086427566173015,
      "language_loss": 0.60197896,
      "learning_rate": 3.132280146886911e-06,
      "loss": 0.62236434,
      "num_input_tokens_seen": 117580550,
      "step": 5475,
      "time_per_iteration": 3.1031100749969482
    },
    {
      "auxiliary_loss_clip": 0.01090035,
      "auxiliary_loss_mlp": 0.01064085,
      "balance_loss_clip": 1.04547465,
      "balance_loss_mlp": 1.04391456,
      "epoch": 0.3292349316098001,
      "flos": 27964537073280.0,
      "grad_norm": 2.8020759869830494,
      "language_loss": 0.76461595,
      "learning_rate": 3.131959088630455e-06,
      "loss": 0.78615719,
      "num_input_tokens_seen": 117600645,
      "step": 5476,
      "time_per_iteration": 2.651751756668091
    },
    {
      "auxiliary_loss_clip": 0.01101078,
      "auxiliary_loss_mlp": 0.01043622,
      "balance_loss_clip": 1.05127311,
      "balance_loss_mlp": 1.02788043,
      "epoch": 0.3292950548624681,
      "flos": 20263673462400.0,
      "grad_norm": 1.7600088659618365,
      "language_loss": 0.747738,
      "learning_rate": 3.131637987449997e-06,
      "loss": 0.76918501,
      "num_input_tokens_seen": 117618880,
      "step": 5477,
      "time_per_iteration": 2.545360803604126
    },
    {
      "auxiliary_loss_clip": 0.0113016,
      "auxiliary_loss_mlp": 0.01037557,
      "balance_loss_clip": 1.04784095,
      "balance_loss_mlp": 1.02369285,
      "epoch": 0.32935517811513604,
      "flos": 20812999132800.0,
      "grad_norm": 2.1056019745433394,
      "language_loss": 0.76023293,
      "learning_rate": 3.131316843357713e-06,
      "loss": 0.78191,
      "num_input_tokens_seen": 117636445,
      "step": 5478,
      "time_per_iteration": 2.4514975547790527
    },
    {
      "auxiliary_loss_clip": 0.01123547,
      "auxiliary_loss_mlp": 0.01038072,
      "balance_loss_clip": 1.04846287,
      "balance_loss_mlp": 1.02303922,
      "epoch": 0.329415301367804,
      "flos": 18441853603200.0,
      "grad_norm": 1.6813502205150128,
      "language_loss": 0.80191004,
      "learning_rate": 3.1309956563657807e-06,
      "loss": 0.82352614,
      "num_input_tokens_seen": 117653105,
      "step": 5479,
      "time_per_iteration": 3.837000608444214
    },
    {
      "auxiliary_loss_clip": 0.01031,
      "auxiliary_loss_mlp": 0.01002052,
      "balance_loss_clip": 1.02065659,
      "balance_loss_mlp": 0.99990654,
      "epoch": 0.32947542462047197,
      "flos": 66323024887680.0,
      "grad_norm": 0.742993163833018,
      "language_loss": 0.56541169,
      "learning_rate": 3.1306744264863804e-06,
      "loss": 0.58574218,
      "num_input_tokens_seen": 117719225,
      "step": 5480,
      "time_per_iteration": 3.1774446964263916
    },
    {
      "auxiliary_loss_clip": 0.0111852,
      "auxiliary_loss_mlp": 0.00799337,
      "balance_loss_clip": 1.044438,
      "balance_loss_mlp": 1.02188611,
      "epoch": 0.32953554787313993,
      "flos": 23221599569280.0,
      "grad_norm": 1.5796536615293784,
      "language_loss": 0.76773661,
      "learning_rate": 3.1303531537316915e-06,
      "loss": 0.78691512,
      "num_input_tokens_seen": 117738725,
      "step": 5481,
      "time_per_iteration": 2.520198345184326
    },
    {
      "auxiliary_loss_clip": 0.01115462,
      "auxiliary_loss_mlp": 0.01039739,
      "balance_loss_clip": 1.04979312,
      "balance_loss_mlp": 1.02512407,
      "epoch": 0.3295956711258079,
      "flos": 27009492307200.0,
      "grad_norm": 1.6892566733503858,
      "language_loss": 0.78874493,
      "learning_rate": 3.130031838113899e-06,
      "loss": 0.81029695,
      "num_input_tokens_seen": 117757765,
      "step": 5482,
      "time_per_iteration": 3.9484097957611084
    },
    {
      "auxiliary_loss_clip": 0.01125318,
      "auxiliary_loss_mlp": 0.0104137,
      "balance_loss_clip": 1.04666817,
      "balance_loss_mlp": 1.02522278,
      "epoch": 0.32965579437847586,
      "flos": 19171702051200.0,
      "grad_norm": 1.700504979403609,
      "language_loss": 0.73801476,
      "learning_rate": 3.129710479645185e-06,
      "loss": 0.75968158,
      "num_input_tokens_seen": 117776810,
      "step": 5483,
      "time_per_iteration": 2.492429256439209
    },
    {
      "auxiliary_loss_clip": 0.01118639,
      "auxiliary_loss_mlp": 0.01040903,
      "balance_loss_clip": 1.04778647,
      "balance_loss_mlp": 1.02504802,
      "epoch": 0.32971591763114383,
      "flos": 30482521401600.0,
      "grad_norm": 2.0638169605851466,
      "language_loss": 0.75515503,
      "learning_rate": 3.1293890783377366e-06,
      "loss": 0.77675045,
      "num_input_tokens_seen": 117797730,
      "step": 5484,
      "time_per_iteration": 3.972377061843872
    },
    {
      "auxiliary_loss_clip": 0.01136109,
      "auxiliary_loss_mlp": 0.01042292,
      "balance_loss_clip": 1.05119514,
      "balance_loss_mlp": 1.02743912,
      "epoch": 0.3297760408838118,
      "flos": 16289583598080.0,
      "grad_norm": 1.9152973663677686,
      "language_loss": 0.71589315,
      "learning_rate": 3.129067634203742e-06,
      "loss": 0.73767716,
      "num_input_tokens_seen": 117815365,
      "step": 5485,
      "time_per_iteration": 2.453735828399658
    },
    {
      "auxiliary_loss_clip": 0.01079009,
      "auxiliary_loss_mlp": 0.01043771,
      "balance_loss_clip": 1.04968822,
      "balance_loss_mlp": 1.0291853,
      "epoch": 0.32983616413647976,
      "flos": 29530924341120.0,
      "grad_norm": 1.5433290961866533,
      "language_loss": 0.80142546,
      "learning_rate": 3.128746147255388e-06,
      "loss": 0.82265329,
      "num_input_tokens_seen": 117836095,
      "step": 5486,
      "time_per_iteration": 4.129487037658691
    },
    {
      "auxiliary_loss_clip": 0.01103452,
      "auxiliary_loss_mlp": 0.01043017,
      "balance_loss_clip": 1.04526496,
      "balance_loss_mlp": 1.02693594,
      "epoch": 0.3298962873891478,
      "flos": 20631398947200.0,
      "grad_norm": 2.3778133122296223,
      "language_loss": 0.84491414,
      "learning_rate": 3.1284246175048683e-06,
      "loss": 0.86637884,
      "num_input_tokens_seen": 117854655,
      "step": 5487,
      "time_per_iteration": 2.5164694786071777
    },
    {
      "auxiliary_loss_clip": 0.01086335,
      "auxiliary_loss_mlp": 0.01047651,
      "balance_loss_clip": 1.04621315,
      "balance_loss_mlp": 1.03029442,
      "epoch": 0.32995641064181574,
      "flos": 14976007228800.0,
      "grad_norm": 4.638093314421904,
      "language_loss": 0.7443555,
      "learning_rate": 3.1281030449643735e-06,
      "loss": 0.76569539,
      "num_input_tokens_seen": 117873300,
      "step": 5488,
      "time_per_iteration": 2.578690528869629
    },
    {
      "auxiliary_loss_clip": 0.01135722,
      "auxiliary_loss_mlp": 0.01046155,
      "balance_loss_clip": 1.04939377,
      "balance_loss_mlp": 1.030604,
      "epoch": 0.3300165338944837,
      "flos": 18661447399680.0,
      "grad_norm": 1.9638895865791772,
      "language_loss": 0.72278631,
      "learning_rate": 3.127781429646098e-06,
      "loss": 0.74460506,
      "num_input_tokens_seen": 117891540,
      "step": 5489,
      "time_per_iteration": 2.4486260414123535
    },
    {
      "auxiliary_loss_clip": 0.01129293,
      "auxiliary_loss_mlp": 0.01037101,
      "balance_loss_clip": 1.04441094,
      "balance_loss_mlp": 1.02222323,
      "epoch": 0.3300766571471517,
      "flos": 25583730785280.0,
      "grad_norm": 3.104893868167829,
      "language_loss": 0.88910258,
      "learning_rate": 3.127459771562238e-06,
      "loss": 0.91076654,
      "num_input_tokens_seen": 117907690,
      "step": 5490,
      "time_per_iteration": 2.4767656326293945
    },
    {
      "auxiliary_loss_clip": 0.01121905,
      "auxiliary_loss_mlp": 0.01036712,
      "balance_loss_clip": 1.04669809,
      "balance_loss_mlp": 1.02192378,
      "epoch": 0.33013678039981964,
      "flos": 11363501623680.0,
      "grad_norm": 2.09878642216636,
      "language_loss": 0.83256477,
      "learning_rate": 3.1271380707249907e-06,
      "loss": 0.85415089,
      "num_input_tokens_seen": 117925640,
      "step": 5491,
      "time_per_iteration": 2.4777300357818604
    },
    {
      "auxiliary_loss_clip": 0.01104845,
      "auxiliary_loss_mlp": 0.01042442,
      "balance_loss_clip": 1.04622126,
      "balance_loss_mlp": 1.02684903,
      "epoch": 0.3301969036524876,
      "flos": 24821203939200.0,
      "grad_norm": 2.1355899255731976,
      "language_loss": 0.77415788,
      "learning_rate": 3.126816327146554e-06,
      "loss": 0.79563075,
      "num_input_tokens_seen": 117944525,
      "step": 5492,
      "time_per_iteration": 2.550823211669922
    },
    {
      "auxiliary_loss_clip": 0.01139042,
      "auxiliary_loss_mlp": 0.0104954,
      "balance_loss_clip": 1.0506618,
      "balance_loss_mlp": 1.03301787,
      "epoch": 0.33025702690515557,
      "flos": 15961144613760.0,
      "grad_norm": 2.0169818605182495,
      "language_loss": 0.74499428,
      "learning_rate": 3.12649454083913e-06,
      "loss": 0.76688015,
      "num_input_tokens_seen": 117962515,
      "step": 5493,
      "time_per_iteration": 2.442042112350464
    },
    {
      "auxiliary_loss_clip": 0.01001518,
      "auxiliary_loss_mlp": 0.0101201,
      "balance_loss_clip": 1.02339768,
      "balance_loss_mlp": 1.0100435,
      "epoch": 0.33031715015782354,
      "flos": 59416755989760.0,
      "grad_norm": 0.8203395257930374,
      "language_loss": 0.53945386,
      "learning_rate": 3.12617271181492e-06,
      "loss": 0.55958915,
      "num_input_tokens_seen": 118018780,
      "step": 5494,
      "time_per_iteration": 3.1261751651763916
    },
    {
      "auxiliary_loss_clip": 0.0111175,
      "auxiliary_loss_mlp": 0.01038592,
      "balance_loss_clip": 1.0459758,
      "balance_loss_mlp": 1.02304733,
      "epoch": 0.3303772734104915,
      "flos": 23184360144000.0,
      "grad_norm": 1.4864332269576475,
      "language_loss": 0.86923397,
      "learning_rate": 3.1258508400861276e-06,
      "loss": 0.89073741,
      "num_input_tokens_seen": 118038610,
      "step": 5495,
      "time_per_iteration": 2.5393624305725098
    },
    {
      "auxiliary_loss_clip": 0.0110108,
      "auxiliary_loss_mlp": 0.01049143,
      "balance_loss_clip": 1.05028534,
      "balance_loss_mlp": 1.03208458,
      "epoch": 0.33043739666315947,
      "flos": 33071896010880.0,
      "grad_norm": 1.935474501440665,
      "language_loss": 0.74103951,
      "learning_rate": 3.1255289256649587e-06,
      "loss": 0.76254177,
      "num_input_tokens_seen": 118055905,
      "step": 5496,
      "time_per_iteration": 2.632340431213379
    },
    {
      "auxiliary_loss_clip": 0.01101942,
      "auxiliary_loss_mlp": 0.01035967,
      "balance_loss_clip": 1.04539526,
      "balance_loss_mlp": 1.02139401,
      "epoch": 0.33049751991582743,
      "flos": 24895431394560.0,
      "grad_norm": 2.010564387297931,
      "language_loss": 0.72076875,
      "learning_rate": 3.1252069685636196e-06,
      "loss": 0.74214786,
      "num_input_tokens_seen": 118073695,
      "step": 5497,
      "time_per_iteration": 2.5323538780212402
    },
    {
      "auxiliary_loss_clip": 0.0110802,
      "auxiliary_loss_mlp": 0.01037398,
      "balance_loss_clip": 1.04930186,
      "balance_loss_mlp": 1.02232957,
      "epoch": 0.3305576431684954,
      "flos": 29460575554560.0,
      "grad_norm": 1.7679144991047733,
      "language_loss": 0.80004895,
      "learning_rate": 3.124884968794321e-06,
      "loss": 0.82150304,
      "num_input_tokens_seen": 118094030,
      "step": 5498,
      "time_per_iteration": 2.582070827484131
    },
    {
      "auxiliary_loss_clip": 0.01118463,
      "auxiliary_loss_mlp": 0.01040138,
      "balance_loss_clip": 1.0480516,
      "balance_loss_mlp": 1.02414012,
      "epoch": 0.33061776642116336,
      "flos": 22632305040000.0,
      "grad_norm": 1.882264977955824,
      "language_loss": 0.75789982,
      "learning_rate": 3.12456292636927e-06,
      "loss": 0.77948582,
      "num_input_tokens_seen": 118111665,
      "step": 5499,
      "time_per_iteration": 2.4868319034576416
    },
    {
      "auxiliary_loss_clip": 0.01112597,
      "auxiliary_loss_mlp": 0.01037118,
      "balance_loss_clip": 1.04704332,
      "balance_loss_mlp": 1.02191901,
      "epoch": 0.3306778896738313,
      "flos": 25776320532480.0,
      "grad_norm": 1.5065644749892309,
      "language_loss": 0.78804672,
      "learning_rate": 3.124240841300681e-06,
      "loss": 0.80954385,
      "num_input_tokens_seen": 118132435,
      "step": 5500,
      "time_per_iteration": 2.6027326583862305
    },
    {
      "auxiliary_loss_clip": 0.01124093,
      "auxiliary_loss_mlp": 0.01034011,
      "balance_loss_clip": 1.04829073,
      "balance_loss_mlp": 1.01775098,
      "epoch": 0.33073801292649935,
      "flos": 36940552479360.0,
      "grad_norm": 2.369776800035216,
      "language_loss": 0.65798068,
      "learning_rate": 3.1239187136007665e-06,
      "loss": 0.67956173,
      "num_input_tokens_seen": 118155255,
      "step": 5501,
      "time_per_iteration": 2.625359058380127
    },
    {
      "auxiliary_loss_clip": 0.01124953,
      "auxiliary_loss_mlp": 0.01042176,
      "balance_loss_clip": 1.04852605,
      "balance_loss_mlp": 1.0264703,
      "epoch": 0.3307981361791673,
      "flos": 12967738848000.0,
      "grad_norm": 2.025538435449352,
      "language_loss": 0.77407002,
      "learning_rate": 3.1235965432817417e-06,
      "loss": 0.79574132,
      "num_input_tokens_seen": 118169865,
      "step": 5502,
      "time_per_iteration": 2.4394137859344482
    },
    {
      "auxiliary_loss_clip": 0.01112345,
      "auxiliary_loss_mlp": 0.01037162,
      "balance_loss_clip": 1.05330682,
      "balance_loss_mlp": 1.02119946,
      "epoch": 0.3308582594318353,
      "flos": 25374372364800.0,
      "grad_norm": 1.675536065866582,
      "language_loss": 0.7223888,
      "learning_rate": 3.123274330355824e-06,
      "loss": 0.74388385,
      "num_input_tokens_seen": 118190760,
      "step": 5503,
      "time_per_iteration": 2.576261520385742
    },
    {
      "auxiliary_loss_clip": 0.01106336,
      "auxiliary_loss_mlp": 0.01039226,
      "balance_loss_clip": 1.04572856,
      "balance_loss_mlp": 1.02251315,
      "epoch": 0.33091838268450324,
      "flos": 26468570419200.0,
      "grad_norm": 1.491195717388614,
      "language_loss": 0.75358522,
      "learning_rate": 3.12295207483523e-06,
      "loss": 0.77504086,
      "num_input_tokens_seen": 118213620,
      "step": 5504,
      "time_per_iteration": 2.628840446472168
    },
    {
      "auxiliary_loss_clip": 0.0111595,
      "auxiliary_loss_mlp": 0.01040822,
      "balance_loss_clip": 1.04766142,
      "balance_loss_mlp": 1.02623641,
      "epoch": 0.3309785059371712,
      "flos": 24971167221120.0,
      "grad_norm": 1.5506213646942966,
      "language_loss": 0.69952393,
      "learning_rate": 3.1226297767321816e-06,
      "loss": 0.72109157,
      "num_input_tokens_seen": 118235010,
      "step": 5505,
      "time_per_iteration": 2.584027051925659
    },
    {
      "auxiliary_loss_clip": 0.01121121,
      "auxiliary_loss_mlp": 0.01043607,
      "balance_loss_clip": 1.05177379,
      "balance_loss_mlp": 1.02830088,
      "epoch": 0.3310386291898392,
      "flos": 20446710192000.0,
      "grad_norm": 2.3302474517869074,
      "language_loss": 0.81918979,
      "learning_rate": 3.122307436058899e-06,
      "loss": 0.840837,
      "num_input_tokens_seen": 118255820,
      "step": 5506,
      "time_per_iteration": 2.5205061435699463
    },
    {
      "auxiliary_loss_clip": 0.01122387,
      "auxiliary_loss_mlp": 0.01036392,
      "balance_loss_clip": 1.04939198,
      "balance_loss_mlp": 1.02075195,
      "epoch": 0.33109875244250714,
      "flos": 23182672204800.0,
      "grad_norm": 1.7511576720064914,
      "language_loss": 0.79148108,
      "learning_rate": 3.121985052827606e-06,
      "loss": 0.81306887,
      "num_input_tokens_seen": 118274160,
      "step": 5507,
      "time_per_iteration": 2.5357484817504883
    },
    {
      "auxiliary_loss_clip": 0.01108696,
      "auxiliary_loss_mlp": 0.01046038,
      "balance_loss_clip": 1.04426932,
      "balance_loss_mlp": 1.0304929,
      "epoch": 0.3311588756951751,
      "flos": 24168384207360.0,
      "grad_norm": 1.4683645976318707,
      "language_loss": 0.71438825,
      "learning_rate": 3.1216626270505274e-06,
      "loss": 0.73593557,
      "num_input_tokens_seen": 118294385,
      "step": 5508,
      "time_per_iteration": 2.5443015098571777
    },
    {
      "auxiliary_loss_clip": 0.01104344,
      "auxiliary_loss_mlp": 0.01034319,
      "balance_loss_clip": 1.04953694,
      "balance_loss_mlp": 1.01971579,
      "epoch": 0.33121899894784307,
      "flos": 28145742209280.0,
      "grad_norm": 1.88543403506718,
      "language_loss": 0.71630645,
      "learning_rate": 3.12134015873989e-06,
      "loss": 0.73769307,
      "num_input_tokens_seen": 118313105,
      "step": 5509,
      "time_per_iteration": 2.5685813426971436
    },
    {
      "auxiliary_loss_clip": 0.0111882,
      "auxiliary_loss_mlp": 0.0103215,
      "balance_loss_clip": 1.05020237,
      "balance_loss_mlp": 1.01719546,
      "epoch": 0.33127912220051103,
      "flos": 29567660976000.0,
      "grad_norm": 1.6244961402668463,
      "language_loss": 0.72920591,
      "learning_rate": 3.121017647907921e-06,
      "loss": 0.75071555,
      "num_input_tokens_seen": 118335250,
      "step": 5510,
      "time_per_iteration": 2.567974328994751
    },
    {
      "auxiliary_loss_clip": 0.010981,
      "auxiliary_loss_mlp": 0.01038685,
      "balance_loss_clip": 1.04612899,
      "balance_loss_mlp": 1.02391493,
      "epoch": 0.331339245453179,
      "flos": 14428836374400.0,
      "grad_norm": 9.010627189791242,
      "language_loss": 0.88039297,
      "learning_rate": 3.1206950945668508e-06,
      "loss": 0.90176082,
      "num_input_tokens_seen": 118351470,
      "step": 5511,
      "time_per_iteration": 2.5276591777801514
    },
    {
      "auxiliary_loss_clip": 0.01073807,
      "auxiliary_loss_mlp": 0.01040851,
      "balance_loss_clip": 1.04407167,
      "balance_loss_mlp": 1.02567554,
      "epoch": 0.33139936870584696,
      "flos": 20887118847360.0,
      "grad_norm": 1.6778755553351334,
      "language_loss": 0.72861725,
      "learning_rate": 3.12037249872891e-06,
      "loss": 0.74976385,
      "num_input_tokens_seen": 118370970,
      "step": 5512,
      "time_per_iteration": 2.5764427185058594
    },
    {
      "auxiliary_loss_clip": 0.01089968,
      "auxiliary_loss_mlp": 0.0103928,
      "balance_loss_clip": 1.04646754,
      "balance_loss_mlp": 1.0249157,
      "epoch": 0.33145949195851493,
      "flos": 36284356869120.0,
      "grad_norm": 1.898640615628319,
      "language_loss": 0.72444129,
      "learning_rate": 3.1200498604063317e-06,
      "loss": 0.7457338,
      "num_input_tokens_seen": 118393125,
      "step": 5513,
      "time_per_iteration": 2.707489490509033
    },
    {
      "auxiliary_loss_clip": 0.0110083,
      "auxiliary_loss_mlp": 0.01035725,
      "balance_loss_clip": 1.05102122,
      "balance_loss_mlp": 1.02026367,
      "epoch": 0.33151961521118295,
      "flos": 14279735018880.0,
      "grad_norm": 2.002162609009222,
      "language_loss": 0.68207383,
      "learning_rate": 3.1197271796113507e-06,
      "loss": 0.70343941,
      "num_input_tokens_seen": 118410860,
      "step": 5514,
      "time_per_iteration": 2.567573070526123
    },
    {
      "auxiliary_loss_clip": 0.01107486,
      "auxiliary_loss_mlp": 0.01042726,
      "balance_loss_clip": 1.05014658,
      "balance_loss_mlp": 1.02573872,
      "epoch": 0.3315797384638509,
      "flos": 20774323163520.0,
      "grad_norm": 2.3633346886566207,
      "language_loss": 0.66522485,
      "learning_rate": 3.1194044563562026e-06,
      "loss": 0.68672705,
      "num_input_tokens_seen": 118429570,
      "step": 5515,
      "time_per_iteration": 2.5206732749938965
    },
    {
      "auxiliary_loss_clip": 0.01117267,
      "auxiliary_loss_mlp": 0.01036797,
      "balance_loss_clip": 1.04919469,
      "balance_loss_mlp": 1.02145493,
      "epoch": 0.3316398617165189,
      "flos": 24679464871680.0,
      "grad_norm": 1.7031284162875397,
      "language_loss": 0.68992853,
      "learning_rate": 3.1190816906531257e-06,
      "loss": 0.71146917,
      "num_input_tokens_seen": 118450285,
      "step": 5516,
      "time_per_iteration": 2.4967901706695557
    },
    {
      "auxiliary_loss_clip": 0.01122491,
      "auxiliary_loss_mlp": 0.01039129,
      "balance_loss_clip": 1.04584599,
      "balance_loss_mlp": 1.0242939,
      "epoch": 0.33169998496918685,
      "flos": 18587974129920.0,
      "grad_norm": 2.875536071878683,
      "language_loss": 0.80750042,
      "learning_rate": 3.118758882514359e-06,
      "loss": 0.82911658,
      "num_input_tokens_seen": 118468270,
      "step": 5517,
      "time_per_iteration": 2.471675157546997
    },
    {
      "auxiliary_loss_clip": 0.01112029,
      "auxiliary_loss_mlp": 0.01037728,
      "balance_loss_clip": 1.04419458,
      "balance_loss_mlp": 1.02232611,
      "epoch": 0.3317601082218548,
      "flos": 20193647898240.0,
      "grad_norm": 1.682872415106335,
      "language_loss": 0.74316704,
      "learning_rate": 3.118436031952143e-06,
      "loss": 0.76466459,
      "num_input_tokens_seen": 118486615,
      "step": 5518,
      "time_per_iteration": 3.852931261062622
    },
    {
      "auxiliary_loss_clip": 0.01028092,
      "auxiliary_loss_mlp": 0.01009363,
      "balance_loss_clip": 1.01982927,
      "balance_loss_mlp": 1.00719392,
      "epoch": 0.3318202314745228,
      "flos": 68974703637120.0,
      "grad_norm": 0.6218892697076962,
      "language_loss": 0.54361409,
      "learning_rate": 3.1181131389787206e-06,
      "loss": 0.56398863,
      "num_input_tokens_seen": 118553580,
      "step": 5519,
      "time_per_iteration": 3.2361485958099365
    },
    {
      "auxiliary_loss_clip": 0.01119478,
      "auxiliary_loss_mlp": 0.01038891,
      "balance_loss_clip": 1.04561961,
      "balance_loss_mlp": 1.02357316,
      "epoch": 0.33188035472719074,
      "flos": 21500113374720.0,
      "grad_norm": 2.447614561413274,
      "language_loss": 0.78667092,
      "learning_rate": 3.117790203606336e-06,
      "loss": 0.8082546,
      "num_input_tokens_seen": 118570280,
      "step": 5520,
      "time_per_iteration": 2.484006404876709
    },
    {
      "auxiliary_loss_clip": 0.01106089,
      "auxiliary_loss_mlp": 0.01034254,
      "balance_loss_clip": 1.04800844,
      "balance_loss_mlp": 1.01973403,
      "epoch": 0.3319404779798587,
      "flos": 28870490926080.0,
      "grad_norm": 1.77765321226945,
      "language_loss": 0.76303041,
      "learning_rate": 3.1174672258472344e-06,
      "loss": 0.7844339,
      "num_input_tokens_seen": 118590455,
      "step": 5521,
      "time_per_iteration": 3.977576971054077
    },
    {
      "auxiliary_loss_clip": 0.01120946,
      "auxiliary_loss_mlp": 0.0104277,
      "balance_loss_clip": 1.04569387,
      "balance_loss_mlp": 1.02692688,
      "epoch": 0.33200060123252667,
      "flos": 23076915586560.0,
      "grad_norm": 1.851364202467023,
      "language_loss": 0.7035324,
      "learning_rate": 3.117144205713664e-06,
      "loss": 0.72516954,
      "num_input_tokens_seen": 118609495,
      "step": 5522,
      "time_per_iteration": 2.494180202484131
    },
    {
      "auxiliary_loss_clip": 0.01107221,
      "auxiliary_loss_mlp": 0.01032792,
      "balance_loss_clip": 1.0460093,
      "balance_loss_mlp": 1.01864171,
      "epoch": 0.33206072448519464,
      "flos": 21142479611520.0,
      "grad_norm": 2.0169205581593617,
      "language_loss": 0.73715079,
      "learning_rate": 3.1168211432178735e-06,
      "loss": 0.75855082,
      "num_input_tokens_seen": 118628720,
      "step": 5523,
      "time_per_iteration": 4.068978309631348
    },
    {
      "auxiliary_loss_clip": 0.0110103,
      "auxiliary_loss_mlp": 0.01034125,
      "balance_loss_clip": 1.04377627,
      "balance_loss_mlp": 1.01887846,
      "epoch": 0.3321208477378626,
      "flos": 13079097987840.0,
      "grad_norm": 1.8147235516209936,
      "language_loss": 0.81997728,
      "learning_rate": 3.116498038372114e-06,
      "loss": 0.84132886,
      "num_input_tokens_seen": 118645955,
      "step": 5524,
      "time_per_iteration": 2.500979423522949
    },
    {
      "auxiliary_loss_clip": 0.01088461,
      "auxiliary_loss_mlp": 0.00795264,
      "balance_loss_clip": 1.04533839,
      "balance_loss_mlp": 1.0185895,
      "epoch": 0.33218097099053057,
      "flos": 21215414177280.0,
      "grad_norm": 2.136733840343463,
      "language_loss": 0.82797927,
      "learning_rate": 3.116174891188636e-06,
      "loss": 0.84681648,
      "num_input_tokens_seen": 118665605,
      "step": 5525,
      "time_per_iteration": 3.9312376976013184
    },
    {
      "auxiliary_loss_clip": 0.01044639,
      "auxiliary_loss_mlp": 0.01004166,
      "balance_loss_clip": 1.01571834,
      "balance_loss_mlp": 1.00182998,
      "epoch": 0.33224109424319853,
      "flos": 64348979189760.0,
      "grad_norm": 0.7696809073821715,
      "language_loss": 0.52713746,
      "learning_rate": 3.1158517016796945e-06,
      "loss": 0.54762548,
      "num_input_tokens_seen": 118728155,
      "step": 5526,
      "time_per_iteration": 3.037161350250244
    },
    {
      "auxiliary_loss_clip": 0.01091373,
      "auxiliary_loss_mlp": 0.00797034,
      "balance_loss_clip": 1.04797339,
      "balance_loss_mlp": 1.01835501,
      "epoch": 0.33230121749586655,
      "flos": 17346003523200.0,
      "grad_norm": 2.280891534753282,
      "language_loss": 0.77977747,
      "learning_rate": 3.1155284698575445e-06,
      "loss": 0.79866153,
      "num_input_tokens_seen": 118743955,
      "step": 5527,
      "time_per_iteration": 2.5080807209014893
    },
    {
      "auxiliary_loss_clip": 0.0108338,
      "auxiliary_loss_mlp": 0.01041682,
      "balance_loss_clip": 1.04566407,
      "balance_loss_mlp": 1.02744269,
      "epoch": 0.3323613407485345,
      "flos": 20997041443200.0,
      "grad_norm": 1.7771673260392324,
      "language_loss": 0.71714807,
      "learning_rate": 3.1152051957344434e-06,
      "loss": 0.73839861,
      "num_input_tokens_seen": 118763275,
      "step": 5528,
      "time_per_iteration": 2.589385747909546
    },
    {
      "auxiliary_loss_clip": 0.01106957,
      "auxiliary_loss_mlp": 0.01032315,
      "balance_loss_clip": 1.04449821,
      "balance_loss_mlp": 1.01852846,
      "epoch": 0.3324214640012025,
      "flos": 13152535344000.0,
      "grad_norm": 1.6043166301368745,
      "language_loss": 0.82613438,
      "learning_rate": 3.1148818793226497e-06,
      "loss": 0.84752715,
      "num_input_tokens_seen": 118781110,
      "step": 5529,
      "time_per_iteration": 2.496351718902588
    },
    {
      "auxiliary_loss_clip": 0.01108919,
      "auxiliary_loss_mlp": 0.00798967,
      "balance_loss_clip": 1.05034125,
      "balance_loss_mlp": 1.02298546,
      "epoch": 0.33248158725387045,
      "flos": 22273522041600.0,
      "grad_norm": 1.752755529474781,
      "language_loss": 0.69842148,
      "learning_rate": 3.114558520634423e-06,
      "loss": 0.71750033,
      "num_input_tokens_seen": 118800620,
      "step": 5530,
      "time_per_iteration": 2.5433077812194824
    },
    {
      "auxiliary_loss_clip": 0.01118635,
      "auxiliary_loss_mlp": 0.010425,
      "balance_loss_clip": 1.04502964,
      "balance_loss_mlp": 1.0268718,
      "epoch": 0.3325417105065384,
      "flos": 20740998320640.0,
      "grad_norm": 2.6615035578931145,
      "language_loss": 0.76388711,
      "learning_rate": 3.1142351196820256e-06,
      "loss": 0.7854985,
      "num_input_tokens_seen": 118818725,
      "step": 5531,
      "time_per_iteration": 2.481428861618042
    },
    {
      "auxiliary_loss_clip": 0.01111239,
      "auxiliary_loss_mlp": 0.0103751,
      "balance_loss_clip": 1.04772425,
      "balance_loss_mlp": 1.02194083,
      "epoch": 0.3326018337592064,
      "flos": 24790536702720.0,
      "grad_norm": 1.7330229239142028,
      "language_loss": 0.73223042,
      "learning_rate": 3.1139116764777206e-06,
      "loss": 0.7537179,
      "num_input_tokens_seen": 118839390,
      "step": 5532,
      "time_per_iteration": 2.588430404663086
    },
    {
      "auxiliary_loss_clip": 0.01109561,
      "auxiliary_loss_mlp": 0.01031368,
      "balance_loss_clip": 1.050596,
      "balance_loss_mlp": 1.01672876,
      "epoch": 0.33266195701187434,
      "flos": 14501699112960.0,
      "grad_norm": 2.956639341064044,
      "language_loss": 0.65750384,
      "learning_rate": 3.1135881910337735e-06,
      "loss": 0.67891312,
      "num_input_tokens_seen": 118856275,
      "step": 5533,
      "time_per_iteration": 2.4874658584594727
    },
    {
      "auxiliary_loss_clip": 0.0107763,
      "auxiliary_loss_mlp": 0.01036218,
      "balance_loss_clip": 1.04673672,
      "balance_loss_mlp": 1.02085185,
      "epoch": 0.3327220802645423,
      "flos": 15304410299520.0,
      "grad_norm": 1.6137793026537122,
      "language_loss": 0.70894378,
      "learning_rate": 3.113264663362451e-06,
      "loss": 0.73008227,
      "num_input_tokens_seen": 118873830,
      "step": 5534,
      "time_per_iteration": 2.6081204414367676
    },
    {
      "auxiliary_loss_clip": 0.01091323,
      "auxiliary_loss_mlp": 0.01038561,
      "balance_loss_clip": 1.0479908,
      "balance_loss_mlp": 1.02381516,
      "epoch": 0.3327822035172103,
      "flos": 23477534951040.0,
      "grad_norm": 1.741078651829642,
      "language_loss": 0.67151392,
      "learning_rate": 3.1129410934760204e-06,
      "loss": 0.6928128,
      "num_input_tokens_seen": 118891560,
      "step": 5535,
      "time_per_iteration": 2.5801644325256348
    },
    {
      "auxiliary_loss_clip": 0.01118069,
      "auxiliary_loss_mlp": 0.00792287,
      "balance_loss_clip": 1.04384804,
      "balance_loss_mlp": 1.01301372,
      "epoch": 0.33284232676987824,
      "flos": 25374516019200.0,
      "grad_norm": 2.0671010275432335,
      "language_loss": 0.73064828,
      "learning_rate": 3.1126174813867517e-06,
      "loss": 0.74975181,
      "num_input_tokens_seen": 118910260,
      "step": 5536,
      "time_per_iteration": 2.5389885902404785
    },
    {
      "auxiliary_loss_clip": 0.01119597,
      "auxiliary_loss_mlp": 0.01037095,
      "balance_loss_clip": 1.04611707,
      "balance_loss_mlp": 1.02311754,
      "epoch": 0.3329024500225462,
      "flos": 23694363400320.0,
      "grad_norm": 1.5191004388462481,
      "language_loss": 0.81648958,
      "learning_rate": 3.112293827106917e-06,
      "loss": 0.83805645,
      "num_input_tokens_seen": 118929985,
      "step": 5537,
      "time_per_iteration": 2.509449005126953
    },
    {
      "auxiliary_loss_clip": 0.01124985,
      "auxiliary_loss_mlp": 0.01040009,
      "balance_loss_clip": 1.04992783,
      "balance_loss_mlp": 1.02475059,
      "epoch": 0.33296257327521417,
      "flos": 31723163205120.0,
      "grad_norm": 2.049255837108728,
      "language_loss": 0.72100633,
      "learning_rate": 3.111970130648789e-06,
      "loss": 0.74265629,
      "num_input_tokens_seen": 118951355,
      "step": 5538,
      "time_per_iteration": 2.572175979614258
    },
    {
      "auxiliary_loss_clip": 0.01115995,
      "auxiliary_loss_mlp": 0.01033165,
      "balance_loss_clip": 1.0454154,
      "balance_loss_mlp": 1.01885426,
      "epoch": 0.33302269652788213,
      "flos": 22744705674240.0,
      "grad_norm": 1.9370743266214556,
      "language_loss": 0.74407899,
      "learning_rate": 3.1116463920246424e-06,
      "loss": 0.76557064,
      "num_input_tokens_seen": 118970910,
      "step": 5539,
      "time_per_iteration": 2.5058515071868896
    },
    {
      "auxiliary_loss_clip": 0.01137562,
      "auxiliary_loss_mlp": 0.01045255,
      "balance_loss_clip": 1.04889858,
      "balance_loss_mlp": 1.02996063,
      "epoch": 0.33308281978055015,
      "flos": 11473747441920.0,
      "grad_norm": 1.7925757816038796,
      "language_loss": 0.71246433,
      "learning_rate": 3.1113226112467527e-06,
      "loss": 0.73429251,
      "num_input_tokens_seen": 118989200,
      "step": 5540,
      "time_per_iteration": 2.4454381465911865
    },
    {
      "auxiliary_loss_clip": 0.01116634,
      "auxiliary_loss_mlp": 0.0103525,
      "balance_loss_clip": 1.04290128,
      "balance_loss_mlp": 1.02064705,
      "epoch": 0.3331429430332181,
      "flos": 38213693112960.0,
      "grad_norm": 1.5790520466337679,
      "language_loss": 0.6082598,
      "learning_rate": 3.1109987883273983e-06,
      "loss": 0.62977868,
      "num_input_tokens_seen": 119011030,
      "step": 5541,
      "time_per_iteration": 2.636073589324951
    },
    {
      "auxiliary_loss_clip": 0.01111752,
      "auxiliary_loss_mlp": 0.01043605,
      "balance_loss_clip": 1.04666042,
      "balance_loss_mlp": 1.0276072,
      "epoch": 0.3332030662858861,
      "flos": 22528667324160.0,
      "grad_norm": 1.6298898735155893,
      "language_loss": 0.68772066,
      "learning_rate": 3.1106749232788584e-06,
      "loss": 0.70927423,
      "num_input_tokens_seen": 119030620,
      "step": 5542,
      "time_per_iteration": 2.5379576683044434
    },
    {
      "auxiliary_loss_clip": 0.01118784,
      "auxiliary_loss_mlp": 0.010397,
      "balance_loss_clip": 1.045367,
      "balance_loss_mlp": 1.0254786,
      "epoch": 0.33326318953855405,
      "flos": 15997773507840.0,
      "grad_norm": 1.8484241478748256,
      "language_loss": 0.7508536,
      "learning_rate": 3.110351016113414e-06,
      "loss": 0.77243853,
      "num_input_tokens_seen": 119048015,
      "step": 5543,
      "time_per_iteration": 2.449554204940796
    },
    {
      "auxiliary_loss_clip": 0.01056054,
      "auxiliary_loss_mlp": 0.01044626,
      "balance_loss_clip": 1.04643583,
      "balance_loss_mlp": 1.02790082,
      "epoch": 0.333323312791222,
      "flos": 25593535198080.0,
      "grad_norm": 2.06389028723006,
      "language_loss": 0.74882305,
      "learning_rate": 3.110027066843348e-06,
      "loss": 0.76982975,
      "num_input_tokens_seen": 119066280,
      "step": 5544,
      "time_per_iteration": 2.7156589031219482
    },
    {
      "auxiliary_loss_clip": 0.01128864,
      "auxiliary_loss_mlp": 0.01030931,
      "balance_loss_clip": 1.04639554,
      "balance_loss_mlp": 1.01706696,
      "epoch": 0.33338343604389,
      "flos": 25119550304640.0,
      "grad_norm": 1.5600650451093327,
      "language_loss": 0.70888126,
      "learning_rate": 3.1097030754809456e-06,
      "loss": 0.73047924,
      "num_input_tokens_seen": 119087680,
      "step": 5545,
      "time_per_iteration": 2.5162851810455322
    },
    {
      "auxiliary_loss_clip": 0.01093825,
      "auxiliary_loss_mlp": 0.01036526,
      "balance_loss_clip": 1.04973483,
      "balance_loss_mlp": 1.02223241,
      "epoch": 0.33344355929655795,
      "flos": 16947287579520.0,
      "grad_norm": 1.753661732973528,
      "language_loss": 0.69056153,
      "learning_rate": 3.1093790420384894e-06,
      "loss": 0.71186501,
      "num_input_tokens_seen": 119105820,
      "step": 5546,
      "time_per_iteration": 2.543632984161377
    },
    {
      "auxiliary_loss_clip": 0.01095664,
      "auxiliary_loss_mlp": 0.01036968,
      "balance_loss_clip": 1.04292643,
      "balance_loss_mlp": 1.02212656,
      "epoch": 0.3335036825492259,
      "flos": 27889591345920.0,
      "grad_norm": 1.5377883972955524,
      "language_loss": 0.64645171,
      "learning_rate": 3.1090549665282702e-06,
      "loss": 0.66777802,
      "num_input_tokens_seen": 119126630,
      "step": 5547,
      "time_per_iteration": 2.613478183746338
    },
    {
      "auxiliary_loss_clip": 0.01109725,
      "auxiliary_loss_mlp": 0.01032824,
      "balance_loss_clip": 1.04822612,
      "balance_loss_mlp": 1.01918697,
      "epoch": 0.3335638058018939,
      "flos": 16179553261440.0,
      "grad_norm": 2.132337652019009,
      "language_loss": 0.8573246,
      "learning_rate": 3.1087308489625742e-06,
      "loss": 0.87875009,
      "num_input_tokens_seen": 119143375,
      "step": 5548,
      "time_per_iteration": 2.508316993713379
    },
    {
      "auxiliary_loss_clip": 0.01121636,
      "auxiliary_loss_mlp": 0.01037151,
      "balance_loss_clip": 1.04691625,
      "balance_loss_mlp": 1.02140307,
      "epoch": 0.33362392905456184,
      "flos": 39896108288640.0,
      "grad_norm": 1.9634304924993615,
      "language_loss": 0.74345231,
      "learning_rate": 3.1084066893536945e-06,
      "loss": 0.76504016,
      "num_input_tokens_seen": 119166450,
      "step": 5549,
      "time_per_iteration": 2.657221555709839
    },
    {
      "auxiliary_loss_clip": 0.01123783,
      "auxiliary_loss_mlp": 0.0104177,
      "balance_loss_clip": 1.04752827,
      "balance_loss_mlp": 1.02597439,
      "epoch": 0.3336840523072298,
      "flos": 44271212567040.0,
      "grad_norm": 1.9283372487541417,
      "language_loss": 0.68697447,
      "learning_rate": 3.108082487713921e-06,
      "loss": 0.70862997,
      "num_input_tokens_seen": 119189645,
      "step": 5550,
      "time_per_iteration": 2.6794331073760986
    },
    {
      "auxiliary_loss_clip": 0.01089728,
      "auxiliary_loss_mlp": 0.01044037,
      "balance_loss_clip": 1.04591775,
      "balance_loss_mlp": 1.02873647,
      "epoch": 0.33374417555989777,
      "flos": 15085678429440.0,
      "grad_norm": 1.7774529748197163,
      "language_loss": 0.60163587,
      "learning_rate": 3.1077582440555495e-06,
      "loss": 0.6229735,
      "num_input_tokens_seen": 119208045,
      "step": 5551,
      "time_per_iteration": 2.527299404144287
    },
    {
      "auxiliary_loss_clip": 0.01090977,
      "auxiliary_loss_mlp": 0.01036147,
      "balance_loss_clip": 1.04778361,
      "balance_loss_mlp": 1.02181196,
      "epoch": 0.33380429881256574,
      "flos": 15849174942720.0,
      "grad_norm": 1.7950518390477725,
      "language_loss": 0.70595276,
      "learning_rate": 3.1074339583908746e-06,
      "loss": 0.72722399,
      "num_input_tokens_seen": 119224910,
      "step": 5552,
      "time_per_iteration": 2.574974775314331
    },
    {
      "auxiliary_loss_clip": 0.01096507,
      "auxiliary_loss_mlp": 0.01035807,
      "balance_loss_clip": 1.04485524,
      "balance_loss_mlp": 1.02123356,
      "epoch": 0.33386442206523376,
      "flos": 13480327883520.0,
      "grad_norm": 1.9877689899076725,
      "language_loss": 0.82570982,
      "learning_rate": 3.107109630732192e-06,
      "loss": 0.8470329,
      "num_input_tokens_seen": 119243290,
      "step": 5553,
      "time_per_iteration": 2.5344736576080322
    },
    {
      "auxiliary_loss_clip": 0.01112315,
      "auxiliary_loss_mlp": 0.00794505,
      "balance_loss_clip": 1.0499388,
      "balance_loss_mlp": 1.01665354,
      "epoch": 0.3339245453179017,
      "flos": 16690669839360.0,
      "grad_norm": 1.8657876204818176,
      "language_loss": 0.80720991,
      "learning_rate": 3.1067852610918017e-06,
      "loss": 0.82627815,
      "num_input_tokens_seen": 119261195,
      "step": 5554,
      "time_per_iteration": 2.5218353271484375
    },
    {
      "auxiliary_loss_clip": 0.01121488,
      "auxiliary_loss_mlp": 0.01041057,
      "balance_loss_clip": 1.04838514,
      "balance_loss_mlp": 1.02685308,
      "epoch": 0.3339846685705697,
      "flos": 24610624456320.0,
      "grad_norm": 1.4919463605556968,
      "language_loss": 0.81442487,
      "learning_rate": 3.1064608494820032e-06,
      "loss": 0.83605033,
      "num_input_tokens_seen": 119282845,
      "step": 5555,
      "time_per_iteration": 2.542682409286499
    },
    {
      "auxiliary_loss_clip": 0.0111842,
      "auxiliary_loss_mlp": 0.01040672,
      "balance_loss_clip": 1.04644299,
      "balance_loss_mlp": 1.02645588,
      "epoch": 0.33404479182323765,
      "flos": 30953812775040.0,
      "grad_norm": 1.7407386105171407,
      "language_loss": 0.74456275,
      "learning_rate": 3.106136395915099e-06,
      "loss": 0.76615369,
      "num_input_tokens_seen": 119304430,
      "step": 5556,
      "time_per_iteration": 3.956327438354492
    },
    {
      "auxiliary_loss_clip": 0.01120097,
      "auxiliary_loss_mlp": 0.01038467,
      "balance_loss_clip": 1.04759526,
      "balance_loss_mlp": 1.02457333,
      "epoch": 0.3341049150759056,
      "flos": 23513301918720.0,
      "grad_norm": 1.420117107680523,
      "language_loss": 0.82470226,
      "learning_rate": 3.105811900403391e-06,
      "loss": 0.84628797,
      "num_input_tokens_seen": 119323830,
      "step": 5557,
      "time_per_iteration": 2.537945508956909
    },
    {
      "auxiliary_loss_clip": 0.01111923,
      "auxiliary_loss_mlp": 0.01037463,
      "balance_loss_clip": 1.04855561,
      "balance_loss_mlp": 1.02265704,
      "epoch": 0.3341650383285736,
      "flos": 24026824707840.0,
      "grad_norm": 1.5381210144178443,
      "language_loss": 0.80001342,
      "learning_rate": 3.1054873629591855e-06,
      "loss": 0.82150733,
      "num_input_tokens_seen": 119346340,
      "step": 5558,
      "time_per_iteration": 2.6029341220855713
    },
    {
      "auxiliary_loss_clip": 0.0110716,
      "auxiliary_loss_mlp": 0.01042119,
      "balance_loss_clip": 1.04868066,
      "balance_loss_mlp": 1.02776659,
      "epoch": 0.33422516158124155,
      "flos": 24901967669760.0,
      "grad_norm": 1.586762045545281,
      "language_loss": 0.81584889,
      "learning_rate": 3.105162783594788e-06,
      "loss": 0.83734167,
      "num_input_tokens_seen": 119367285,
      "step": 5559,
      "time_per_iteration": 2.571357011795044
    },
    {
      "auxiliary_loss_clip": 0.01095886,
      "auxiliary_loss_mlp": 0.0104257,
      "balance_loss_clip": 1.04834604,
      "balance_loss_mlp": 1.0279789,
      "epoch": 0.3342852848339095,
      "flos": 18333403464960.0,
      "grad_norm": 1.6821669582844299,
      "language_loss": 0.71647573,
      "learning_rate": 3.1048381623225074e-06,
      "loss": 0.73786026,
      "num_input_tokens_seen": 119385370,
      "step": 5560,
      "time_per_iteration": 3.9117767810821533
    },
    {
      "auxiliary_loss_clip": 0.01114421,
      "auxiliary_loss_mlp": 0.01044046,
      "balance_loss_clip": 1.04834878,
      "balance_loss_mlp": 1.02909684,
      "epoch": 0.3343454080865775,
      "flos": 30046530119040.0,
      "grad_norm": 1.5268256802125977,
      "language_loss": 0.75024223,
      "learning_rate": 3.1045134991546526e-06,
      "loss": 0.77182686,
      "num_input_tokens_seen": 119409150,
      "step": 5561,
      "time_per_iteration": 2.5881669521331787
    },
    {
      "auxiliary_loss_clip": 0.01110901,
      "auxiliary_loss_mlp": 0.01033835,
      "balance_loss_clip": 1.05188227,
      "balance_loss_mlp": 1.01893342,
      "epoch": 0.33440553133924544,
      "flos": 16398823835520.0,
      "grad_norm": 1.7589936363340548,
      "language_loss": 0.69297361,
      "learning_rate": 3.1041887941035355e-06,
      "loss": 0.71442103,
      "num_input_tokens_seen": 119426475,
      "step": 5562,
      "time_per_iteration": 3.9318058490753174
    },
    {
      "auxiliary_loss_clip": 0.01123188,
      "auxiliary_loss_mlp": 0.01039984,
      "balance_loss_clip": 1.05061877,
      "balance_loss_mlp": 1.02641749,
      "epoch": 0.3344656545919134,
      "flos": 24242072958720.0,
      "grad_norm": 1.649529676026584,
      "language_loss": 0.64662284,
      "learning_rate": 3.1038640471814685e-06,
      "loss": 0.66825455,
      "num_input_tokens_seen": 119446900,
      "step": 5563,
      "time_per_iteration": 2.523895025253296
    },
    {
      "auxiliary_loss_clip": 0.01077403,
      "auxiliary_loss_mlp": 0.01044019,
      "balance_loss_clip": 1.05174661,
      "balance_loss_mlp": 1.02840304,
      "epoch": 0.3345257778445814,
      "flos": 52118843149440.0,
      "grad_norm": 1.360535054000048,
      "language_loss": 0.74472356,
      "learning_rate": 3.103539258400766e-06,
      "loss": 0.76593781,
      "num_input_tokens_seen": 119470945,
      "step": 5564,
      "time_per_iteration": 4.205468654632568
    },
    {
      "auxiliary_loss_clip": 0.01023826,
      "auxiliary_loss_mlp": 0.01008086,
      "balance_loss_clip": 1.0218792,
      "balance_loss_mlp": 1.005988,
      "epoch": 0.33458590109724934,
      "flos": 68048602254720.0,
      "grad_norm": 0.778106716183215,
      "language_loss": 0.55579448,
      "learning_rate": 3.103214427773745e-06,
      "loss": 0.57611358,
      "num_input_tokens_seen": 119529925,
      "step": 5565,
      "time_per_iteration": 3.114863395690918
    },
    {
      "auxiliary_loss_clip": 0.01134208,
      "auxiliary_loss_mlp": 0.01037207,
      "balance_loss_clip": 1.05202937,
      "balance_loss_mlp": 1.02309895,
      "epoch": 0.3346460243499173,
      "flos": 37414788768000.0,
      "grad_norm": 1.8526630471417884,
      "language_loss": 0.65061134,
      "learning_rate": 3.102889555312721e-06,
      "loss": 0.67232549,
      "num_input_tokens_seen": 119550700,
      "step": 5566,
      "time_per_iteration": 2.61820387840271
    },
    {
      "auxiliary_loss_clip": 0.01114922,
      "auxiliary_loss_mlp": 0.01040118,
      "balance_loss_clip": 1.05024755,
      "balance_loss_mlp": 1.0253005,
      "epoch": 0.3347061476025853,
      "flos": 18697358021760.0,
      "grad_norm": 1.7838722494845207,
      "language_loss": 0.77318799,
      "learning_rate": 3.102564641030016e-06,
      "loss": 0.79473841,
      "num_input_tokens_seen": 119569295,
      "step": 5567,
      "time_per_iteration": 2.5359787940979004
    },
    {
      "auxiliary_loss_clip": 0.01111773,
      "auxiliary_loss_mlp": 0.01036391,
      "balance_loss_clip": 1.04934549,
      "balance_loss_mlp": 1.02070296,
      "epoch": 0.3347662708552533,
      "flos": 13917827537280.0,
      "grad_norm": 1.6429466092176976,
      "language_loss": 0.76172352,
      "learning_rate": 3.102239684937949e-06,
      "loss": 0.78320515,
      "num_input_tokens_seen": 119587375,
      "step": 5568,
      "time_per_iteration": 2.499683380126953
    },
    {
      "auxiliary_loss_clip": 0.01098225,
      "auxiliary_loss_mlp": 0.01039903,
      "balance_loss_clip": 1.05071259,
      "balance_loss_mlp": 1.02482319,
      "epoch": 0.33482639410792125,
      "flos": 19750402068480.0,
      "grad_norm": 1.9446663549803862,
      "language_loss": 0.71489513,
      "learning_rate": 3.101914687048842e-06,
      "loss": 0.73627645,
      "num_input_tokens_seen": 119604530,
      "step": 5569,
      "time_per_iteration": 2.5983738899230957
    },
    {
      "auxiliary_loss_clip": 0.01095767,
      "auxiliary_loss_mlp": 0.01036052,
      "balance_loss_clip": 1.0444659,
      "balance_loss_mlp": 1.01945794,
      "epoch": 0.3348865173605892,
      "flos": 16102991422080.0,
      "grad_norm": 1.9522498508532315,
      "language_loss": 0.89482403,
      "learning_rate": 3.10158964737502e-06,
      "loss": 0.91614217,
      "num_input_tokens_seen": 119621025,
      "step": 5570,
      "time_per_iteration": 2.5375895500183105
    },
    {
      "auxiliary_loss_clip": 0.01099642,
      "auxiliary_loss_mlp": 0.01034229,
      "balance_loss_clip": 1.04970312,
      "balance_loss_mlp": 1.01899457,
      "epoch": 0.3349466406132572,
      "flos": 25008945350400.0,
      "grad_norm": 1.448152600262487,
      "language_loss": 0.7962206,
      "learning_rate": 3.101264565928808e-06,
      "loss": 0.81755936,
      "num_input_tokens_seen": 119641725,
      "step": 5571,
      "time_per_iteration": 2.600584030151367
    },
    {
      "auxiliary_loss_clip": 0.01048437,
      "auxiliary_loss_mlp": 0.00798853,
      "balance_loss_clip": 1.01875937,
      "balance_loss_mlp": 1.04525566,
      "epoch": 0.33500676386592515,
      "flos": 54319991564160.0,
      "grad_norm": 0.90130250939553,
      "language_loss": 0.55901939,
      "learning_rate": 3.1009394427225335e-06,
      "loss": 0.5774923,
      "num_input_tokens_seen": 119693560,
      "step": 5572,
      "time_per_iteration": 3.036715507507324
    },
    {
      "auxiliary_loss_clip": 0.01135315,
      "auxiliary_loss_mlp": 0.01046303,
      "balance_loss_clip": 1.05176187,
      "balance_loss_mlp": 1.03098512,
      "epoch": 0.3350668871185931,
      "flos": 26797332625920.0,
      "grad_norm": 1.7792847980775874,
      "language_loss": 0.78341997,
      "learning_rate": 3.1006142777685257e-06,
      "loss": 0.8052361,
      "num_input_tokens_seen": 119712935,
      "step": 5573,
      "time_per_iteration": 2.5151069164276123
    },
    {
      "auxiliary_loss_clip": 0.01102217,
      "auxiliary_loss_mlp": 0.01044989,
      "balance_loss_clip": 1.04797196,
      "balance_loss_mlp": 1.02928901,
      "epoch": 0.3351270103712611,
      "flos": 33510508986240.0,
      "grad_norm": 2.4643976192808883,
      "language_loss": 0.724401,
      "learning_rate": 3.1002890710791133e-06,
      "loss": 0.74587309,
      "num_input_tokens_seen": 119731680,
      "step": 5574,
      "time_per_iteration": 2.658658742904663
    },
    {
      "auxiliary_loss_clip": 0.01120474,
      "auxiliary_loss_mlp": 0.01032721,
      "balance_loss_clip": 1.04919863,
      "balance_loss_mlp": 1.01795125,
      "epoch": 0.33518713362392905,
      "flos": 26506240807680.0,
      "grad_norm": 1.5527502688454318,
      "language_loss": 0.88050395,
      "learning_rate": 3.0999638226666287e-06,
      "loss": 0.90203589,
      "num_input_tokens_seen": 119752155,
      "step": 5575,
      "time_per_iteration": 2.5395758152008057
    },
    {
      "auxiliary_loss_clip": 0.01114346,
      "auxiliary_loss_mlp": 0.01039097,
      "balance_loss_clip": 1.0479871,
      "balance_loss_mlp": 1.02225244,
      "epoch": 0.335247256876597,
      "flos": 17232345912960.0,
      "grad_norm": 2.0553776260834407,
      "language_loss": 0.82291591,
      "learning_rate": 3.0996385325434063e-06,
      "loss": 0.84445029,
      "num_input_tokens_seen": 119769195,
      "step": 5576,
      "time_per_iteration": 2.5089094638824463
    },
    {
      "auxiliary_loss_clip": 0.01120829,
      "auxiliary_loss_mlp": 0.01035573,
      "balance_loss_clip": 1.0480938,
      "balance_loss_mlp": 1.02008772,
      "epoch": 0.335307380129265,
      "flos": 25629373992960.0,
      "grad_norm": 2.131808297183543,
      "language_loss": 0.72159052,
      "learning_rate": 3.0993132007217806e-06,
      "loss": 0.74315459,
      "num_input_tokens_seen": 119786810,
      "step": 5577,
      "time_per_iteration": 2.522834539413452
    },
    {
      "auxiliary_loss_clip": 0.0110506,
      "auxiliary_loss_mlp": 0.01037695,
      "balance_loss_clip": 1.05348182,
      "balance_loss_mlp": 1.0222578,
      "epoch": 0.33536750338193294,
      "flos": 19680089195520.0,
      "grad_norm": 1.5970009651873587,
      "language_loss": 0.81792861,
      "learning_rate": 3.0989878272140883e-06,
      "loss": 0.83935612,
      "num_input_tokens_seen": 119805395,
      "step": 5578,
      "time_per_iteration": 2.5687131881713867
    },
    {
      "auxiliary_loss_clip": 0.01072572,
      "auxiliary_loss_mlp": 0.00820626,
      "balance_loss_clip": 1.04605198,
      "balance_loss_mlp": 1.0667789,
      "epoch": 0.3354276266346009,
      "flos": 18332613365760.0,
      "grad_norm": 1.740764083642832,
      "language_loss": 0.71580797,
      "learning_rate": 3.0986624120326676e-06,
      "loss": 0.73474002,
      "num_input_tokens_seen": 119823135,
      "step": 5579,
      "time_per_iteration": 2.5987608432769775
    },
    {
      "auxiliary_loss_clip": 0.01082474,
      "auxiliary_loss_mlp": 0.01037523,
      "balance_loss_clip": 1.05036891,
      "balance_loss_mlp": 1.02135849,
      "epoch": 0.3354877498872689,
      "flos": 17858556645120.0,
      "grad_norm": 1.7050273367615514,
      "language_loss": 0.81352794,
      "learning_rate": 3.0983369551898573e-06,
      "loss": 0.83472788,
      "num_input_tokens_seen": 119842265,
      "step": 5580,
      "time_per_iteration": 2.6142852306365967
    },
    {
      "auxiliary_loss_clip": 0.01118314,
      "auxiliary_loss_mlp": 0.0103126,
      "balance_loss_clip": 1.04947257,
      "balance_loss_mlp": 1.01603651,
      "epoch": 0.3355478731399369,
      "flos": 24717745791360.0,
      "grad_norm": 1.7177670393343658,
      "language_loss": 0.77889067,
      "learning_rate": 3.0980114566980003e-06,
      "loss": 0.80038643,
      "num_input_tokens_seen": 119862500,
      "step": 5581,
      "time_per_iteration": 2.5574939250946045
    },
    {
      "auxiliary_loss_clip": 0.01100682,
      "auxiliary_loss_mlp": 0.01045569,
      "balance_loss_clip": 1.04710031,
      "balance_loss_mlp": 1.02841449,
      "epoch": 0.33560799639260486,
      "flos": 16873886136960.0,
      "grad_norm": 2.466293118509431,
      "language_loss": 0.7464571,
      "learning_rate": 3.0976859165694384e-06,
      "loss": 0.76791966,
      "num_input_tokens_seen": 119880160,
      "step": 5582,
      "time_per_iteration": 2.5395913124084473
    },
    {
      "auxiliary_loss_clip": 0.01108893,
      "auxiliary_loss_mlp": 0.01044313,
      "balance_loss_clip": 1.04518175,
      "balance_loss_mlp": 1.02850592,
      "epoch": 0.3356681196452728,
      "flos": 18333511205760.0,
      "grad_norm": 1.5615663718935355,
      "language_loss": 0.81778061,
      "learning_rate": 3.0973603348165166e-06,
      "loss": 0.83931267,
      "num_input_tokens_seen": 119899040,
      "step": 5583,
      "time_per_iteration": 2.521315336227417
    },
    {
      "auxiliary_loss_clip": 0.0111158,
      "auxiliary_loss_mlp": 0.0104567,
      "balance_loss_clip": 1.0484463,
      "balance_loss_mlp": 1.0311439,
      "epoch": 0.3357282428979408,
      "flos": 34750612085760.0,
      "grad_norm": 1.8169661673799984,
      "language_loss": 0.77997488,
      "learning_rate": 3.097034711451581e-06,
      "loss": 0.80154741,
      "num_input_tokens_seen": 119921120,
      "step": 5584,
      "time_per_iteration": 2.6589980125427246
    },
    {
      "auxiliary_loss_clip": 0.01112227,
      "auxiliary_loss_mlp": 0.01037787,
      "balance_loss_clip": 1.0474087,
      "balance_loss_mlp": 1.02262402,
      "epoch": 0.33578836615060875,
      "flos": 21580087006080.0,
      "grad_norm": 1.5371972701112662,
      "language_loss": 0.76172698,
      "learning_rate": 3.0967090464869795e-06,
      "loss": 0.78322709,
      "num_input_tokens_seen": 119940165,
      "step": 5585,
      "time_per_iteration": 2.54693865776062
    },
    {
      "auxiliary_loss_clip": 0.01117767,
      "auxiliary_loss_mlp": 0.01039569,
      "balance_loss_clip": 1.04538321,
      "balance_loss_mlp": 1.02404773,
      "epoch": 0.3358484894032767,
      "flos": 24530291688960.0,
      "grad_norm": 1.507248960338949,
      "language_loss": 0.7747575,
      "learning_rate": 3.0963833399350608e-06,
      "loss": 0.79633081,
      "num_input_tokens_seen": 119959730,
      "step": 5586,
      "time_per_iteration": 2.5451254844665527
    },
    {
      "auxiliary_loss_clip": 0.01098655,
      "auxiliary_loss_mlp": 0.01055631,
      "balance_loss_clip": 1.05192041,
      "balance_loss_mlp": 1.03704643,
      "epoch": 0.3359086126559447,
      "flos": 22455589104000.0,
      "grad_norm": 1.684145280134752,
      "language_loss": 0.80838186,
      "learning_rate": 3.0960575918081756e-06,
      "loss": 0.8299247,
      "num_input_tokens_seen": 119979315,
      "step": 5587,
      "time_per_iteration": 2.556368589401245
    },
    {
      "auxiliary_loss_clip": 0.01131141,
      "auxiliary_loss_mlp": 0.01042801,
      "balance_loss_clip": 1.05027795,
      "balance_loss_mlp": 1.02887154,
      "epoch": 0.33596873590861265,
      "flos": 16543687386240.0,
      "grad_norm": 1.7010533526278973,
      "language_loss": 0.67026949,
      "learning_rate": 3.095731802118677e-06,
      "loss": 0.69200897,
      "num_input_tokens_seen": 119996140,
      "step": 5588,
      "time_per_iteration": 2.459420680999756
    },
    {
      "auxiliary_loss_clip": 0.01109811,
      "auxiliary_loss_mlp": 0.00807538,
      "balance_loss_clip": 1.04602671,
      "balance_loss_mlp": 1.04038262,
      "epoch": 0.3360288591612806,
      "flos": 31175812782720.0,
      "grad_norm": 2.0907640342145704,
      "language_loss": 0.70187289,
      "learning_rate": 3.095405970878919e-06,
      "loss": 0.72104639,
      "num_input_tokens_seen": 120017720,
      "step": 5589,
      "time_per_iteration": 2.610053539276123
    },
    {
      "auxiliary_loss_clip": 0.01111767,
      "auxiliary_loss_mlp": 0.01045602,
      "balance_loss_clip": 1.04832494,
      "balance_loss_mlp": 1.02929378,
      "epoch": 0.3360889824139486,
      "flos": 23696913265920.0,
      "grad_norm": 1.626171432883623,
      "language_loss": 0.67269981,
      "learning_rate": 3.0950800981012567e-06,
      "loss": 0.69427341,
      "num_input_tokens_seen": 120036335,
      "step": 5590,
      "time_per_iteration": 2.557910919189453
    },
    {
      "auxiliary_loss_clip": 0.01103046,
      "auxiliary_loss_mlp": 0.01049558,
      "balance_loss_clip": 1.05045378,
      "balance_loss_mlp": 1.03430474,
      "epoch": 0.33614910566661654,
      "flos": 19318109886720.0,
      "grad_norm": 2.946946472153406,
      "language_loss": 0.73538172,
      "learning_rate": 3.094754183798047e-06,
      "loss": 0.7569077,
      "num_input_tokens_seen": 120056120,
      "step": 5591,
      "time_per_iteration": 2.565479278564453
    },
    {
      "auxiliary_loss_clip": 0.01133683,
      "auxiliary_loss_mlp": 0.01043082,
      "balance_loss_clip": 1.05036747,
      "balance_loss_mlp": 1.02838409,
      "epoch": 0.3362092289192845,
      "flos": 16472261191680.0,
      "grad_norm": 2.0663762409336983,
      "language_loss": 0.69870663,
      "learning_rate": 3.0944282279816493e-06,
      "loss": 0.72047424,
      "num_input_tokens_seen": 120073650,
      "step": 5592,
      "time_per_iteration": 2.4401934146881104
    },
    {
      "auxiliary_loss_clip": 0.01111891,
      "auxiliary_loss_mlp": 0.01038639,
      "balance_loss_clip": 1.05168164,
      "balance_loss_mlp": 1.02428555,
      "epoch": 0.33626935217195253,
      "flos": 24243581329920.0,
      "grad_norm": 1.9110978116613533,
      "language_loss": 0.76674455,
      "learning_rate": 3.094102230664423e-06,
      "loss": 0.78824985,
      "num_input_tokens_seen": 120093260,
      "step": 5593,
      "time_per_iteration": 2.5452418327331543
    },
    {
      "auxiliary_loss_clip": 0.01102706,
      "auxiliary_loss_mlp": 0.0079502,
      "balance_loss_clip": 1.04755425,
      "balance_loss_mlp": 1.01427686,
      "epoch": 0.3363294754246205,
      "flos": 19718765164800.0,
      "grad_norm": 1.9597891829019363,
      "language_loss": 0.72036928,
      "learning_rate": 3.093776191858731e-06,
      "loss": 0.7393465,
      "num_input_tokens_seen": 120111830,
      "step": 5594,
      "time_per_iteration": 2.552299976348877
    },
    {
      "auxiliary_loss_clip": 0.01081368,
      "auxiliary_loss_mlp": 0.00799073,
      "balance_loss_clip": 1.04908395,
      "balance_loss_mlp": 1.02090669,
      "epoch": 0.33638959867728846,
      "flos": 22596286677120.0,
      "grad_norm": 1.524403703537189,
      "language_loss": 0.800354,
      "learning_rate": 3.0934501115769363e-06,
      "loss": 0.81915838,
      "num_input_tokens_seen": 120130470,
      "step": 5595,
      "time_per_iteration": 3.9465320110321045
    },
    {
      "auxiliary_loss_clip": 0.01112716,
      "auxiliary_loss_mlp": 0.01036032,
      "balance_loss_clip": 1.05174124,
      "balance_loss_mlp": 1.02254319,
      "epoch": 0.3364497219299564,
      "flos": 20994742972800.0,
      "grad_norm": 1.5329593644891475,
      "language_loss": 0.81005889,
      "learning_rate": 3.0931239898314037e-06,
      "loss": 0.83154637,
      "num_input_tokens_seen": 120150735,
      "step": 5596,
      "time_per_iteration": 2.5364816188812256
    },
    {
      "auxiliary_loss_clip": 0.01113971,
      "auxiliary_loss_mlp": 0.01041187,
      "balance_loss_clip": 1.05156195,
      "balance_loss_mlp": 1.02699518,
      "epoch": 0.3365098451826244,
      "flos": 25228610974080.0,
      "grad_norm": 1.860812093866987,
      "language_loss": 0.75630754,
      "learning_rate": 3.0927978266344995e-06,
      "loss": 0.77785915,
      "num_input_tokens_seen": 120173230,
      "step": 5597,
      "time_per_iteration": 2.597223997116089
    },
    {
      "auxiliary_loss_clip": 0.01124911,
      "auxiliary_loss_mlp": 0.01035959,
      "balance_loss_clip": 1.05158329,
      "balance_loss_mlp": 1.02152872,
      "epoch": 0.33656996843529235,
      "flos": 24571697091840.0,
      "grad_norm": 1.657119266375199,
      "language_loss": 0.78443009,
      "learning_rate": 3.0924716219985916e-06,
      "loss": 0.8060388,
      "num_input_tokens_seen": 120191860,
      "step": 5598,
      "time_per_iteration": 3.9629456996917725
    },
    {
      "auxiliary_loss_clip": 0.01141324,
      "auxiliary_loss_mlp": 0.01039923,
      "balance_loss_clip": 1.05237591,
      "balance_loss_mlp": 1.02372265,
      "epoch": 0.3366300916879603,
      "flos": 44091120752640.0,
      "grad_norm": 1.4033009644008196,
      "language_loss": 0.64717185,
      "learning_rate": 3.0921453759360514e-06,
      "loss": 0.66898429,
      "num_input_tokens_seen": 120219195,
      "step": 5599,
      "time_per_iteration": 2.6622812747955322
    },
    {
      "auxiliary_loss_clip": 0.01102727,
      "auxiliary_loss_mlp": 0.0105331,
      "balance_loss_clip": 1.04849589,
      "balance_loss_mlp": 1.03532124,
      "epoch": 0.3366902149406283,
      "flos": 13879869840000.0,
      "grad_norm": 2.5421862549736733,
      "language_loss": 0.82373261,
      "learning_rate": 3.091819088459249e-06,
      "loss": 0.84529305,
      "num_input_tokens_seen": 120232950,
      "step": 5600,
      "time_per_iteration": 3.998006820678711
    },
    {
      "auxiliary_loss_clip": 0.01130753,
      "auxiliary_loss_mlp": 0.01047723,
      "balance_loss_clip": 1.05226827,
      "balance_loss_mlp": 1.03139734,
      "epoch": 0.33675033819329625,
      "flos": 16253098358400.0,
      "grad_norm": 2.176504053391658,
      "language_loss": 0.83086812,
      "learning_rate": 3.0914927595805573e-06,
      "loss": 0.85265285,
      "num_input_tokens_seen": 120248865,
      "step": 5601,
      "time_per_iteration": 2.49052095413208
    },
    {
      "auxiliary_loss_clip": 0.01127646,
      "auxiliary_loss_mlp": 0.01039182,
      "balance_loss_clip": 1.05613232,
      "balance_loss_mlp": 1.02495408,
      "epoch": 0.3368104614459642,
      "flos": 17055809544960.0,
      "grad_norm": 1.672505156808659,
      "language_loss": 0.8374517,
      "learning_rate": 3.0911663893123507e-06,
      "loss": 0.85912007,
      "num_input_tokens_seen": 120267820,
      "step": 5602,
      "time_per_iteration": 3.850393533706665
    },
    {
      "auxiliary_loss_clip": 0.01136329,
      "auxiliary_loss_mlp": 0.01054165,
      "balance_loss_clip": 1.05029893,
      "balance_loss_mlp": 1.03925204,
      "epoch": 0.3368705846986322,
      "flos": 17858628472320.0,
      "grad_norm": 1.842382696239513,
      "language_loss": 0.69786656,
      "learning_rate": 3.0908399776670048e-06,
      "loss": 0.71977144,
      "num_input_tokens_seen": 120286540,
      "step": 5603,
      "time_per_iteration": 2.4520161151885986
    },
    {
      "auxiliary_loss_clip": 0.01116307,
      "auxiliary_loss_mlp": 0.01038306,
      "balance_loss_clip": 1.05347061,
      "balance_loss_mlp": 1.02311826,
      "epoch": 0.33693070795130015,
      "flos": 22929502170240.0,
      "grad_norm": 1.4271207909988064,
      "language_loss": 0.83068252,
      "learning_rate": 3.090513524656898e-06,
      "loss": 0.85222864,
      "num_input_tokens_seen": 120307305,
      "step": 5604,
      "time_per_iteration": 2.5279109477996826
    },
    {
      "auxiliary_loss_clip": 0.01098917,
      "auxiliary_loss_mlp": 0.01044586,
      "balance_loss_clip": 1.05080986,
      "balance_loss_mlp": 1.02899337,
      "epoch": 0.3369908312039681,
      "flos": 22017443005440.0,
      "grad_norm": 1.7583111438084957,
      "language_loss": 0.73505312,
      "learning_rate": 3.090187030294409e-06,
      "loss": 0.75648814,
      "num_input_tokens_seen": 120327845,
      "step": 5605,
      "time_per_iteration": 2.55694842338562
    },
    {
      "auxiliary_loss_clip": 0.01121052,
      "auxiliary_loss_mlp": 0.01038281,
      "balance_loss_clip": 1.05290556,
      "balance_loss_mlp": 1.02312946,
      "epoch": 0.33705095445663613,
      "flos": 11801970944640.0,
      "grad_norm": 2.511059501331598,
      "language_loss": 0.83051419,
      "learning_rate": 3.089860494591919e-06,
      "loss": 0.85210752,
      "num_input_tokens_seen": 120343255,
      "step": 5606,
      "time_per_iteration": 2.51479434967041
    },
    {
      "auxiliary_loss_clip": 0.01111624,
      "auxiliary_loss_mlp": 0.01038127,
      "balance_loss_clip": 1.04885507,
      "balance_loss_mlp": 1.02338696,
      "epoch": 0.3371110777093041,
      "flos": 25046400257280.0,
      "grad_norm": 1.5069068165700772,
      "language_loss": 0.67864496,
      "learning_rate": 3.089533917561809e-06,
      "loss": 0.70014244,
      "num_input_tokens_seen": 120361745,
      "step": 5607,
      "time_per_iteration": 2.563803195953369
    },
    {
      "auxiliary_loss_clip": 0.01120993,
      "auxiliary_loss_mlp": 0.01053168,
      "balance_loss_clip": 1.04845631,
      "balance_loss_mlp": 1.0363003,
      "epoch": 0.33717120096197206,
      "flos": 26579031719040.0,
      "grad_norm": 1.8483661854527487,
      "language_loss": 0.71059465,
      "learning_rate": 3.089207299216464e-06,
      "loss": 0.73233628,
      "num_input_tokens_seen": 120380565,
      "step": 5608,
      "time_per_iteration": 2.600996494293213
    },
    {
      "auxiliary_loss_clip": 0.01059139,
      "auxiliary_loss_mlp": 0.01041256,
      "balance_loss_clip": 1.05147934,
      "balance_loss_mlp": 1.02631342,
      "epoch": 0.33723132421464,
      "flos": 15158541168000.0,
      "grad_norm": 1.7490388323601793,
      "language_loss": 0.7947253,
      "learning_rate": 3.088880639568269e-06,
      "loss": 0.81572926,
      "num_input_tokens_seen": 120399235,
      "step": 5609,
      "time_per_iteration": 2.6472108364105225
    },
    {
      "auxiliary_loss_clip": 0.01127094,
      "auxiliary_loss_mlp": 0.01042523,
      "balance_loss_clip": 1.05203962,
      "balance_loss_mlp": 1.02662086,
      "epoch": 0.337291447467308,
      "flos": 23436093634560.0,
      "grad_norm": 1.6756335469309984,
      "language_loss": 0.82870519,
      "learning_rate": 3.0885539386296114e-06,
      "loss": 0.8504014,
      "num_input_tokens_seen": 120420095,
      "step": 5610,
      "time_per_iteration": 2.544128179550171
    },
    {
      "auxiliary_loss_clip": 0.01124734,
      "auxiliary_loss_mlp": 0.01039408,
      "balance_loss_clip": 1.05186987,
      "balance_loss_mlp": 1.02318382,
      "epoch": 0.33735157071997596,
      "flos": 17238163916160.0,
      "grad_norm": 1.7585377162940603,
      "language_loss": 0.81921387,
      "learning_rate": 3.088227196412879e-06,
      "loss": 0.8408553,
      "num_input_tokens_seen": 120437690,
      "step": 5611,
      "time_per_iteration": 2.4829158782958984
    },
    {
      "auxiliary_loss_clip": 0.01119892,
      "auxiliary_loss_mlp": 0.01040604,
      "balance_loss_clip": 1.0546484,
      "balance_loss_mlp": 1.0237,
      "epoch": 0.3374116939726439,
      "flos": 28257388657920.0,
      "grad_norm": 1.6092575759631285,
      "language_loss": 0.79475796,
      "learning_rate": 3.0879004129304626e-06,
      "loss": 0.81636286,
      "num_input_tokens_seen": 120459240,
      "step": 5612,
      "time_per_iteration": 2.5990817546844482
    },
    {
      "auxiliary_loss_clip": 0.01080523,
      "auxiliary_loss_mlp": 0.01034908,
      "balance_loss_clip": 1.04838264,
      "balance_loss_mlp": 1.01961362,
      "epoch": 0.3374718172253119,
      "flos": 35919396731520.0,
      "grad_norm": 5.218837287786804,
      "language_loss": 0.70304585,
      "learning_rate": 3.087573588194753e-06,
      "loss": 0.72420013,
      "num_input_tokens_seen": 120481090,
      "step": 5613,
      "time_per_iteration": 2.747260808944702
    },
    {
      "auxiliary_loss_clip": 0.01118501,
      "auxiliary_loss_mlp": 0.01032058,
      "balance_loss_clip": 1.05151856,
      "balance_loss_mlp": 1.01621521,
      "epoch": 0.33753194047797985,
      "flos": 18186672407040.0,
      "grad_norm": 1.6778090013766145,
      "language_loss": 0.79668695,
      "learning_rate": 3.087246722218144e-06,
      "loss": 0.81819254,
      "num_input_tokens_seen": 120500045,
      "step": 5614,
      "time_per_iteration": 2.554368019104004
    },
    {
      "auxiliary_loss_clip": 0.01105165,
      "auxiliary_loss_mlp": 0.01040649,
      "balance_loss_clip": 1.04732358,
      "balance_loss_mlp": 1.02388835,
      "epoch": 0.3375920637306478,
      "flos": 23148916398720.0,
      "grad_norm": 1.6209061594699703,
      "language_loss": 0.91321743,
      "learning_rate": 3.086919815013031e-06,
      "loss": 0.93467557,
      "num_input_tokens_seen": 120521125,
      "step": 5615,
      "time_per_iteration": 2.5925652980804443
    },
    {
      "auxiliary_loss_clip": 0.01119322,
      "auxiliary_loss_mlp": 0.01032308,
      "balance_loss_clip": 1.04715121,
      "balance_loss_mlp": 1.01812196,
      "epoch": 0.3376521869833158,
      "flos": 23112215677440.0,
      "grad_norm": 1.6616081461857979,
      "language_loss": 0.80633163,
      "learning_rate": 3.086592866591809e-06,
      "loss": 0.8278479,
      "num_input_tokens_seen": 120539180,
      "step": 5616,
      "time_per_iteration": 2.504072666168213
    },
    {
      "auxiliary_loss_clip": 0.01126561,
      "auxiliary_loss_mlp": 0.00811505,
      "balance_loss_clip": 1.04820061,
      "balance_loss_mlp": 1.04626822,
      "epoch": 0.33771231023598375,
      "flos": 19274585581440.0,
      "grad_norm": 1.6969100693799928,
      "language_loss": 0.83831745,
      "learning_rate": 3.0862658769668774e-06,
      "loss": 0.85769808,
      "num_input_tokens_seen": 120556280,
      "step": 5617,
      "time_per_iteration": 2.5130770206451416
    },
    {
      "auxiliary_loss_clip": 0.0106806,
      "auxiliary_loss_mlp": 0.01040065,
      "balance_loss_clip": 1.04471517,
      "balance_loss_mlp": 1.02387691,
      "epoch": 0.3377724334886517,
      "flos": 18150187167360.0,
      "grad_norm": 1.5580699097318684,
      "language_loss": 0.80092418,
      "learning_rate": 3.0859388461506343e-06,
      "loss": 0.82200539,
      "num_input_tokens_seen": 120575395,
      "step": 5618,
      "time_per_iteration": 2.647524118423462
    },
    {
      "auxiliary_loss_clip": 0.01090069,
      "auxiliary_loss_mlp": 0.01036784,
      "balance_loss_clip": 1.04767537,
      "balance_loss_mlp": 1.02165604,
      "epoch": 0.3378325567413197,
      "flos": 25775997310080.0,
      "grad_norm": 1.6197731468412164,
      "language_loss": 0.70302761,
      "learning_rate": 3.085611774155481e-06,
      "loss": 0.72429609,
      "num_input_tokens_seen": 120596075,
      "step": 5619,
      "time_per_iteration": 2.641116142272949
    },
    {
      "auxiliary_loss_clip": 0.0111271,
      "auxiliary_loss_mlp": 0.0104642,
      "balance_loss_clip": 1.04772329,
      "balance_loss_mlp": 1.03142321,
      "epoch": 0.3378926799939877,
      "flos": 21317112558720.0,
      "grad_norm": 2.3026418000394764,
      "language_loss": 0.69961345,
      "learning_rate": 3.085284660993821e-06,
      "loss": 0.7212047,
      "num_input_tokens_seen": 120614195,
      "step": 5620,
      "time_per_iteration": 2.604933500289917
    },
    {
      "auxiliary_loss_clip": 0.01132738,
      "auxiliary_loss_mlp": 0.01041644,
      "balance_loss_clip": 1.04900551,
      "balance_loss_mlp": 1.02747619,
      "epoch": 0.33795280324665566,
      "flos": 24900028335360.0,
      "grad_norm": 1.64566407212675,
      "language_loss": 0.68022966,
      "learning_rate": 3.084957506678058e-06,
      "loss": 0.70197344,
      "num_input_tokens_seen": 120634475,
      "step": 5621,
      "time_per_iteration": 2.5334160327911377
    },
    {
      "auxiliary_loss_clip": 0.01106209,
      "auxiliary_loss_mlp": 0.01037532,
      "balance_loss_clip": 1.05052209,
      "balance_loss_mlp": 1.02364469,
      "epoch": 0.33801292649932363,
      "flos": 24753943722240.0,
      "grad_norm": 1.6166646234589637,
      "language_loss": 0.82590806,
      "learning_rate": 3.0846303112205975e-06,
      "loss": 0.84734547,
      "num_input_tokens_seen": 120654980,
      "step": 5622,
      "time_per_iteration": 2.5767462253570557
    },
    {
      "auxiliary_loss_clip": 0.01097359,
      "auxiliary_loss_mlp": 0.01034669,
      "balance_loss_clip": 1.04623485,
      "balance_loss_mlp": 1.02047706,
      "epoch": 0.3380730497519916,
      "flos": 26723967096960.0,
      "grad_norm": 1.459819892968621,
      "language_loss": 0.73540354,
      "learning_rate": 3.0843030746338464e-06,
      "loss": 0.75672376,
      "num_input_tokens_seen": 120676245,
      "step": 5623,
      "time_per_iteration": 2.646000862121582
    },
    {
      "auxiliary_loss_clip": 0.01031737,
      "auxiliary_loss_mlp": 0.01002853,
      "balance_loss_clip": 1.02056813,
      "balance_loss_mlp": 1.00073111,
      "epoch": 0.33813317300465956,
      "flos": 70035756416640.0,
      "grad_norm": 0.7446325139577281,
      "language_loss": 0.5490734,
      "learning_rate": 3.083975796930215e-06,
      "loss": 0.56941932,
      "num_input_tokens_seen": 120741965,
      "step": 5624,
      "time_per_iteration": 3.2677204608917236
    },
    {
      "auxiliary_loss_clip": 0.01090429,
      "auxiliary_loss_mlp": 0.01050483,
      "balance_loss_clip": 1.04867184,
      "balance_loss_mlp": 1.03347182,
      "epoch": 0.3381932962573275,
      "flos": 24097317148800.0,
      "grad_norm": 2.451999518187238,
      "language_loss": 0.73290044,
      "learning_rate": 3.083648478122111e-06,
      "loss": 0.75430954,
      "num_input_tokens_seen": 120760410,
      "step": 5625,
      "time_per_iteration": 2.604811668395996
    },
    {
      "auxiliary_loss_clip": 0.01124962,
      "auxiliary_loss_mlp": 0.01040454,
      "balance_loss_clip": 1.04822373,
      "balance_loss_mlp": 1.02457523,
      "epoch": 0.3382534195099955,
      "flos": 19278248768640.0,
      "grad_norm": 1.8628169889081865,
      "language_loss": 0.70656943,
      "learning_rate": 3.0833211182219497e-06,
      "loss": 0.72822362,
      "num_input_tokens_seen": 120777705,
      "step": 5626,
      "time_per_iteration": 2.4951937198638916
    },
    {
      "auxiliary_loss_clip": 0.01109109,
      "auxiliary_loss_mlp": 0.0103453,
      "balance_loss_clip": 1.04828393,
      "balance_loss_mlp": 1.0188657,
      "epoch": 0.33831354276266346,
      "flos": 25226240676480.0,
      "grad_norm": 1.5739589405002028,
      "language_loss": 0.8094635,
      "learning_rate": 3.0829937172421425e-06,
      "loss": 0.83089983,
      "num_input_tokens_seen": 120798660,
      "step": 5627,
      "time_per_iteration": 2.6195735931396484
    },
    {
      "auxiliary_loss_clip": 0.01126243,
      "auxiliary_loss_mlp": 0.00795753,
      "balance_loss_clip": 1.05038667,
      "balance_loss_mlp": 1.01896024,
      "epoch": 0.3383736660153314,
      "flos": 23112000195840.0,
      "grad_norm": 1.6796445860542126,
      "language_loss": 0.80545831,
      "learning_rate": 3.0826662751951055e-06,
      "loss": 0.8246783,
      "num_input_tokens_seen": 120816705,
      "step": 5628,
      "time_per_iteration": 2.529750347137451
    },
    {
      "auxiliary_loss_clip": 0.01079856,
      "auxiliary_loss_mlp": 0.01042395,
      "balance_loss_clip": 1.0477562,
      "balance_loss_mlp": 1.02546692,
      "epoch": 0.3384337892679994,
      "flos": 23477139901440.0,
      "grad_norm": 1.9767443800089315,
      "language_loss": 0.77003777,
      "learning_rate": 3.082338792093254e-06,
      "loss": 0.79126024,
      "num_input_tokens_seen": 120835375,
      "step": 5629,
      "time_per_iteration": 2.6379587650299072
    },
    {
      "auxiliary_loss_clip": 0.01112735,
      "auxiliary_loss_mlp": 0.0104129,
      "balance_loss_clip": 1.04561138,
      "balance_loss_mlp": 1.02519703,
      "epoch": 0.33849391252066735,
      "flos": 19425805839360.0,
      "grad_norm": 1.7596117868287955,
      "language_loss": 0.84723222,
      "learning_rate": 3.0820112679490074e-06,
      "loss": 0.86877245,
      "num_input_tokens_seen": 120854260,
      "step": 5630,
      "time_per_iteration": 2.585627555847168
    },
    {
      "auxiliary_loss_clip": 0.01083255,
      "auxiliary_loss_mlp": 0.01053979,
      "balance_loss_clip": 1.04597759,
      "balance_loss_mlp": 1.03845739,
      "epoch": 0.3385540357733353,
      "flos": 21064840364160.0,
      "grad_norm": 1.8719353275883275,
      "language_loss": 0.72145414,
      "learning_rate": 3.0816837027747857e-06,
      "loss": 0.74282646,
      "num_input_tokens_seen": 120871590,
      "step": 5631,
      "time_per_iteration": 2.592647075653076
    },
    {
      "auxiliary_loss_clip": 0.01029,
      "auxiliary_loss_mlp": 0.01002137,
      "balance_loss_clip": 1.02049911,
      "balance_loss_mlp": 1.00003898,
      "epoch": 0.3386141590260033,
      "flos": 69208013450880.0,
      "grad_norm": 0.8596488056296706,
      "language_loss": 0.56211817,
      "learning_rate": 3.0813560965830084e-06,
      "loss": 0.58242953,
      "num_input_tokens_seen": 120925550,
      "step": 5632,
      "time_per_iteration": 3.202526092529297
    },
    {
      "auxiliary_loss_clip": 0.01115278,
      "auxiliary_loss_mlp": 0.01036377,
      "balance_loss_clip": 1.04706883,
      "balance_loss_mlp": 1.02099848,
      "epoch": 0.3386742822786713,
      "flos": 25519487310720.0,
      "grad_norm": 2.033594353082862,
      "language_loss": 0.80482674,
      "learning_rate": 3.0810284493861005e-06,
      "loss": 0.8263433,
      "num_input_tokens_seen": 120947620,
      "step": 5633,
      "time_per_iteration": 2.5611870288848877
    },
    {
      "auxiliary_loss_clip": 0.0109425,
      "auxiliary_loss_mlp": 0.01040498,
      "balance_loss_clip": 1.04265249,
      "balance_loss_mlp": 1.02507257,
      "epoch": 0.33873440553133927,
      "flos": 23623116773760.0,
      "grad_norm": 1.8851048010864473,
      "language_loss": 0.58898383,
      "learning_rate": 3.0807007611964855e-06,
      "loss": 0.6103313,
      "num_input_tokens_seen": 120965205,
      "step": 5634,
      "time_per_iteration": 4.003119468688965
    },
    {
      "auxiliary_loss_clip": 0.01099103,
      "auxiliary_loss_mlp": 0.010393,
      "balance_loss_clip": 1.04838395,
      "balance_loss_mlp": 1.02430916,
      "epoch": 0.33879452878400723,
      "flos": 17088882992640.0,
      "grad_norm": 1.8212017000633076,
      "language_loss": 0.92448294,
      "learning_rate": 3.080373032026589e-06,
      "loss": 0.945867,
      "num_input_tokens_seen": 120983560,
      "step": 5635,
      "time_per_iteration": 2.5260612964630127
    },
    {
      "auxiliary_loss_clip": 0.01085007,
      "auxiliary_loss_mlp": 0.01033717,
      "balance_loss_clip": 1.04764175,
      "balance_loss_mlp": 1.01878035,
      "epoch": 0.3388546520366752,
      "flos": 15742053607680.0,
      "grad_norm": 1.8010296583263754,
      "language_loss": 0.75782037,
      "learning_rate": 3.0800452618888386e-06,
      "loss": 0.77900755,
      "num_input_tokens_seen": 121001400,
      "step": 5636,
      "time_per_iteration": 2.544010877609253
    },
    {
      "auxiliary_loss_clip": 0.01118783,
      "auxiliary_loss_mlp": 0.01041531,
      "balance_loss_clip": 1.04614305,
      "balance_loss_mlp": 1.02648056,
      "epoch": 0.33891477528934316,
      "flos": 22418744728320.0,
      "grad_norm": 1.9056574527925345,
      "language_loss": 0.83514988,
      "learning_rate": 3.0797174507956637e-06,
      "loss": 0.85675299,
      "num_input_tokens_seen": 121021760,
      "step": 5637,
      "time_per_iteration": 3.8791682720184326
    },
    {
      "auxiliary_loss_clip": 0.01084417,
      "auxiliary_loss_mlp": 0.01044177,
      "balance_loss_clip": 1.04610491,
      "balance_loss_mlp": 1.02585459,
      "epoch": 0.3389748985420111,
      "flos": 17274828723840.0,
      "grad_norm": 1.7180383549100888,
      "language_loss": 0.70061815,
      "learning_rate": 3.079389598759495e-06,
      "loss": 0.7219041,
      "num_input_tokens_seen": 121041070,
      "step": 5638,
      "time_per_iteration": 4.015904188156128
    },
    {
      "auxiliary_loss_clip": 0.01103636,
      "auxiliary_loss_mlp": 0.01055274,
      "balance_loss_clip": 1.04724908,
      "balance_loss_mlp": 1.0397054,
      "epoch": 0.3390350217946791,
      "flos": 27744979190400.0,
      "grad_norm": 1.7274015247509964,
      "language_loss": 0.81084299,
      "learning_rate": 3.079061705792765e-06,
      "loss": 0.83243203,
      "num_input_tokens_seen": 121060890,
      "step": 5639,
      "time_per_iteration": 2.594317674636841
    },
    {
      "auxiliary_loss_clip": 0.011338,
      "auxiliary_loss_mlp": 0.01048015,
      "balance_loss_clip": 1.0469203,
      "balance_loss_mlp": 1.03247023,
      "epoch": 0.33909514504734706,
      "flos": 20339804338560.0,
      "grad_norm": 2.2522383901580962,
      "language_loss": 0.67834413,
      "learning_rate": 3.078733771907907e-06,
      "loss": 0.70016229,
      "num_input_tokens_seen": 121079135,
      "step": 5640,
      "time_per_iteration": 2.4960238933563232
    },
    {
      "auxiliary_loss_clip": 0.0110738,
      "auxiliary_loss_mlp": 0.01041151,
      "balance_loss_clip": 1.04692912,
      "balance_loss_mlp": 1.02629137,
      "epoch": 0.339155268300015,
      "flos": 14830030356480.0,
      "grad_norm": 1.6752566842807726,
      "language_loss": 0.69855046,
      "learning_rate": 3.0784057971173554e-06,
      "loss": 0.72003573,
      "num_input_tokens_seen": 121097685,
      "step": 5641,
      "time_per_iteration": 3.927908420562744
    },
    {
      "auxiliary_loss_clip": 0.01134664,
      "auxiliary_loss_mlp": 0.01046569,
      "balance_loss_clip": 1.04909289,
      "balance_loss_mlp": 1.03167367,
      "epoch": 0.339215391552683,
      "flos": 26067951054720.0,
      "grad_norm": 1.7959978619916872,
      "language_loss": 0.8766107,
      "learning_rate": 3.0780777814335483e-06,
      "loss": 0.89842302,
      "num_input_tokens_seen": 121115640,
      "step": 5642,
      "time_per_iteration": 2.5210633277893066
    },
    {
      "auxiliary_loss_clip": 0.01113924,
      "auxiliary_loss_mlp": 0.01031567,
      "balance_loss_clip": 1.04663229,
      "balance_loss_mlp": 1.01874006,
      "epoch": 0.33927551480535095,
      "flos": 14574705505920.0,
      "grad_norm": 1.7100258677102338,
      "language_loss": 0.83968282,
      "learning_rate": 3.077749724868924e-06,
      "loss": 0.86113775,
      "num_input_tokens_seen": 121132485,
      "step": 5643,
      "time_per_iteration": 2.525709867477417
    },
    {
      "auxiliary_loss_clip": 0.01104341,
      "auxiliary_loss_mlp": 0.01047876,
      "balance_loss_clip": 1.04680371,
      "balance_loss_mlp": 1.03332055,
      "epoch": 0.3393356380580189,
      "flos": 23805578885760.0,
      "grad_norm": 1.5086519357633559,
      "language_loss": 0.77168912,
      "learning_rate": 3.077421627435922e-06,
      "loss": 0.79321122,
      "num_input_tokens_seen": 121152935,
      "step": 5644,
      "time_per_iteration": 2.642129421234131
    },
    {
      "auxiliary_loss_clip": 0.01112996,
      "auxiliary_loss_mlp": 0.01044638,
      "balance_loss_clip": 1.04521394,
      "balance_loss_mlp": 1.03013015,
      "epoch": 0.3393957613106869,
      "flos": 17347871030400.0,
      "grad_norm": 4.768041122555143,
      "language_loss": 0.63128334,
      "learning_rate": 3.0770934891469832e-06,
      "loss": 0.65285969,
      "num_input_tokens_seen": 121169835,
      "step": 5645,
      "time_per_iteration": 2.4716014862060547
    },
    {
      "auxiliary_loss_clip": 0.01117364,
      "auxiliary_loss_mlp": 0.01041621,
      "balance_loss_clip": 1.04601395,
      "balance_loss_mlp": 1.02759576,
      "epoch": 0.3394558845633549,
      "flos": 28433960939520.0,
      "grad_norm": 2.530394190132794,
      "language_loss": 0.76728874,
      "learning_rate": 3.076765310014552e-06,
      "loss": 0.78887856,
      "num_input_tokens_seen": 121190290,
      "step": 5646,
      "time_per_iteration": 2.572298765182495
    },
    {
      "auxiliary_loss_clip": 0.01118916,
      "auxiliary_loss_mlp": 0.01042535,
      "balance_loss_clip": 1.04768991,
      "balance_loss_mlp": 1.02712107,
      "epoch": 0.33951600781602287,
      "flos": 22086929865600.0,
      "grad_norm": 3.1281602088508187,
      "language_loss": 0.78474402,
      "learning_rate": 3.0764370900510727e-06,
      "loss": 0.80635858,
      "num_input_tokens_seen": 121209060,
      "step": 5647,
      "time_per_iteration": 2.550107717514038
    },
    {
      "auxiliary_loss_clip": 0.01107105,
      "auxiliary_loss_mlp": 0.00796549,
      "balance_loss_clip": 1.05308437,
      "balance_loss_mlp": 1.02542078,
      "epoch": 0.33957613106869083,
      "flos": 23878262056320.0,
      "grad_norm": 1.8367939846392098,
      "language_loss": 0.77445722,
      "learning_rate": 3.0761088292689904e-06,
      "loss": 0.79349375,
      "num_input_tokens_seen": 121227480,
      "step": 5648,
      "time_per_iteration": 2.58085298538208
    },
    {
      "auxiliary_loss_clip": 0.00994624,
      "auxiliary_loss_mlp": 0.01005494,
      "balance_loss_clip": 1.02797484,
      "balance_loss_mlp": 1.00328827,
      "epoch": 0.3396362543213588,
      "flos": 71242642414080.0,
      "grad_norm": 0.7885268529206491,
      "language_loss": 0.56328243,
      "learning_rate": 3.075780527680754e-06,
      "loss": 0.5832836,
      "num_input_tokens_seen": 121291305,
      "step": 5649,
      "time_per_iteration": 3.436791181564331
    },
    {
      "auxiliary_loss_clip": 0.0110223,
      "auxiliary_loss_mlp": 0.00797917,
      "balance_loss_clip": 1.04392338,
      "balance_loss_mlp": 1.01979637,
      "epoch": 0.33969637757402676,
      "flos": 25921615046400.0,
      "grad_norm": 1.3750576955834768,
      "language_loss": 0.85620612,
      "learning_rate": 3.0754521852988117e-06,
      "loss": 0.87520754,
      "num_input_tokens_seen": 121312740,
      "step": 5650,
      "time_per_iteration": 2.7819693088531494
    },
    {
      "auxiliary_loss_clip": 0.01119457,
      "auxiliary_loss_mlp": 0.01033573,
      "balance_loss_clip": 1.04568434,
      "balance_loss_mlp": 1.01961327,
      "epoch": 0.33975650082669473,
      "flos": 35261728663680.0,
      "grad_norm": 1.560562624615696,
      "language_loss": 0.70549858,
      "learning_rate": 3.0751238021356152e-06,
      "loss": 0.72702891,
      "num_input_tokens_seen": 121334220,
      "step": 5651,
      "time_per_iteration": 2.6545698642730713
    },
    {
      "auxiliary_loss_clip": 0.01082435,
      "auxiliary_loss_mlp": 0.01040478,
      "balance_loss_clip": 1.04599142,
      "balance_loss_mlp": 1.02591109,
      "epoch": 0.3398166240793627,
      "flos": 16647001879680.0,
      "grad_norm": 1.993898850316831,
      "language_loss": 0.81301177,
      "learning_rate": 3.074795378203616e-06,
      "loss": 0.83424085,
      "num_input_tokens_seen": 121351870,
      "step": 5652,
      "time_per_iteration": 2.5853075981140137
    },
    {
      "auxiliary_loss_clip": 0.01135844,
      "auxiliary_loss_mlp": 0.01036146,
      "balance_loss_clip": 1.04973626,
      "balance_loss_mlp": 1.02120948,
      "epoch": 0.33987674733203066,
      "flos": 24062196625920.0,
      "grad_norm": 1.6504144076819252,
      "language_loss": 0.76946211,
      "learning_rate": 3.0744669135152685e-06,
      "loss": 0.79118204,
      "num_input_tokens_seen": 121373400,
      "step": 5653,
      "time_per_iteration": 2.5696804523468018
    },
    {
      "auxiliary_loss_clip": 0.01114578,
      "auxiliary_loss_mlp": 0.01046614,
      "balance_loss_clip": 1.04516721,
      "balance_loss_mlp": 1.03152204,
      "epoch": 0.3399368705846986,
      "flos": 13250678279040.0,
      "grad_norm": 2.7524037642974943,
      "language_loss": 0.85278833,
      "learning_rate": 3.0741384080830278e-06,
      "loss": 0.8744002,
      "num_input_tokens_seen": 121385225,
      "step": 5654,
      "time_per_iteration": 2.4556944370269775
    },
    {
      "auxiliary_loss_clip": 0.011162,
      "auxiliary_loss_mlp": 0.01036527,
      "balance_loss_clip": 1.04835129,
      "balance_loss_mlp": 1.02166128,
      "epoch": 0.3399969938373666,
      "flos": 27012832272000.0,
      "grad_norm": 2.336365321478241,
      "language_loss": 0.65092331,
      "learning_rate": 3.073809861919351e-06,
      "loss": 0.67245054,
      "num_input_tokens_seen": 121404735,
      "step": 5655,
      "time_per_iteration": 2.5661087036132812
    },
    {
      "auxiliary_loss_clip": 0.01122172,
      "auxiliary_loss_mlp": 0.01039703,
      "balance_loss_clip": 1.05074382,
      "balance_loss_mlp": 1.02605379,
      "epoch": 0.34005711709003456,
      "flos": 28550096588160.0,
      "grad_norm": 1.5668395255716543,
      "language_loss": 0.76854104,
      "learning_rate": 3.073481275036697e-06,
      "loss": 0.79015982,
      "num_input_tokens_seen": 121426780,
      "step": 5656,
      "time_per_iteration": 2.6227197647094727
    },
    {
      "auxiliary_loss_clip": 0.0109813,
      "auxiliary_loss_mlp": 0.01039892,
      "balance_loss_clip": 1.04587126,
      "balance_loss_mlp": 1.02422786,
      "epoch": 0.3401172403427025,
      "flos": 21617003208960.0,
      "grad_norm": 1.7656283860282789,
      "language_loss": 0.8339889,
      "learning_rate": 3.073152647447525e-06,
      "loss": 0.85536909,
      "num_input_tokens_seen": 121447245,
      "step": 5657,
      "time_per_iteration": 2.5967695713043213
    },
    {
      "auxiliary_loss_clip": 0.01104815,
      "auxiliary_loss_mlp": 0.01039551,
      "balance_loss_clip": 1.04742038,
      "balance_loss_mlp": 1.0261997,
      "epoch": 0.3401773635953705,
      "flos": 25885776251520.0,
      "grad_norm": 1.9021899847783224,
      "language_loss": 0.85538173,
      "learning_rate": 3.0728239791642976e-06,
      "loss": 0.87682533,
      "num_input_tokens_seen": 121468165,
      "step": 5658,
      "time_per_iteration": 2.5734925270080566
    },
    {
      "auxiliary_loss_clip": 0.01037994,
      "auxiliary_loss_mlp": 0.01022975,
      "balance_loss_clip": 1.01932549,
      "balance_loss_mlp": 1.0207938,
      "epoch": 0.3402374868480385,
      "flos": 65507995336320.0,
      "grad_norm": 0.817092038647145,
      "language_loss": 0.60063612,
      "learning_rate": 3.072495270199477e-06,
      "loss": 0.62124586,
      "num_input_tokens_seen": 121523795,
      "step": 5659,
      "time_per_iteration": 3.069321870803833
    },
    {
      "auxiliary_loss_clip": 0.01129145,
      "auxiliary_loss_mlp": 0.01035345,
      "balance_loss_clip": 1.04882216,
      "balance_loss_mlp": 1.02192843,
      "epoch": 0.34029761010070647,
      "flos": 24060580513920.0,
      "grad_norm": 1.9257947345530304,
      "language_loss": 0.6749649,
      "learning_rate": 3.0721665205655284e-06,
      "loss": 0.69660985,
      "num_input_tokens_seen": 121542950,
      "step": 5660,
      "time_per_iteration": 2.5340828895568848
    },
    {
      "auxiliary_loss_clip": 0.01134185,
      "auxiliary_loss_mlp": 0.01045857,
      "balance_loss_clip": 1.05167961,
      "balance_loss_mlp": 1.03097963,
      "epoch": 0.34035773335337444,
      "flos": 27599720590080.0,
      "grad_norm": 2.0367146888013137,
      "language_loss": 0.67046511,
      "learning_rate": 3.071837730274918e-06,
      "loss": 0.69226551,
      "num_input_tokens_seen": 121562765,
      "step": 5661,
      "time_per_iteration": 2.559863328933716
    },
    {
      "auxiliary_loss_clip": 0.01108903,
      "auxiliary_loss_mlp": 0.01039425,
      "balance_loss_clip": 1.04771423,
      "balance_loss_mlp": 1.02567434,
      "epoch": 0.3404178566060424,
      "flos": 20812783651200.0,
      "grad_norm": 1.7239698492965985,
      "language_loss": 0.78973424,
      "learning_rate": 3.071508899340113e-06,
      "loss": 0.81121755,
      "num_input_tokens_seen": 121581610,
      "step": 5662,
      "time_per_iteration": 2.558148145675659
    },
    {
      "auxiliary_loss_clip": 0.01098955,
      "auxiliary_loss_mlp": 0.01038596,
      "balance_loss_clip": 1.05207086,
      "balance_loss_mlp": 1.02348018,
      "epoch": 0.34047797985871037,
      "flos": 26833566470400.0,
      "grad_norm": 2.052164418592098,
      "language_loss": 0.73377538,
      "learning_rate": 3.0711800277735833e-06,
      "loss": 0.75515079,
      "num_input_tokens_seen": 121601885,
      "step": 5663,
      "time_per_iteration": 2.5971908569335938
    },
    {
      "auxiliary_loss_clip": 0.01087352,
      "auxiliary_loss_mlp": 0.01038047,
      "balance_loss_clip": 1.04658115,
      "balance_loss_mlp": 1.02482033,
      "epoch": 0.34053810311137833,
      "flos": 19682639061120.0,
      "grad_norm": 1.6984642450855774,
      "language_loss": 0.86486512,
      "learning_rate": 3.0708511155877997e-06,
      "loss": 0.88611913,
      "num_input_tokens_seen": 121621335,
      "step": 5664,
      "time_per_iteration": 2.5762150287628174
    },
    {
      "auxiliary_loss_clip": 0.01133178,
      "auxiliary_loss_mlp": 0.0103634,
      "balance_loss_clip": 1.04900193,
      "balance_loss_mlp": 1.0224818,
      "epoch": 0.3405982263640463,
      "flos": 21725740656000.0,
      "grad_norm": 1.8640738577462617,
      "language_loss": 0.68651927,
      "learning_rate": 3.070522162795235e-06,
      "loss": 0.7082144,
      "num_input_tokens_seen": 121641310,
      "step": 5665,
      "time_per_iteration": 2.4929463863372803
    },
    {
      "auxiliary_loss_clip": 0.01133303,
      "auxiliary_loss_mlp": 0.01040057,
      "balance_loss_clip": 1.04895532,
      "balance_loss_mlp": 1.02404714,
      "epoch": 0.34065834961671426,
      "flos": 18041629288320.0,
      "grad_norm": 2.5037769846412226,
      "language_loss": 0.73076272,
      "learning_rate": 3.0701931694083626e-06,
      "loss": 0.75249636,
      "num_input_tokens_seen": 121659625,
      "step": 5666,
      "time_per_iteration": 2.4856858253479004
    },
    {
      "auxiliary_loss_clip": 0.01125123,
      "auxiliary_loss_mlp": 0.01036447,
      "balance_loss_clip": 1.04790783,
      "balance_loss_mlp": 1.02224898,
      "epoch": 0.3407184728693822,
      "flos": 21397337585280.0,
      "grad_norm": 1.5060264965876597,
      "language_loss": 0.73254758,
      "learning_rate": 3.0698641354396576e-06,
      "loss": 0.75416327,
      "num_input_tokens_seen": 121679205,
      "step": 5667,
      "time_per_iteration": 2.537642478942871
    },
    {
      "auxiliary_loss_clip": 0.01038983,
      "auxiliary_loss_mlp": 0.01013917,
      "balance_loss_clip": 1.02748215,
      "balance_loss_mlp": 1.01152074,
      "epoch": 0.3407785961220502,
      "flos": 68688101018880.0,
      "grad_norm": 0.8434254783907549,
      "language_loss": 0.63303459,
      "learning_rate": 3.069535060901597e-06,
      "loss": 0.65356362,
      "num_input_tokens_seen": 121751085,
      "step": 5668,
      "time_per_iteration": 3.292513370513916
    },
    {
      "auxiliary_loss_clip": 0.01043531,
      "auxiliary_loss_mlp": 0.01040841,
      "balance_loss_clip": 1.05030179,
      "balance_loss_mlp": 1.02492654,
      "epoch": 0.34083871937471816,
      "flos": 14064379027200.0,
      "grad_norm": 2.209231736193042,
      "language_loss": 0.72364616,
      "learning_rate": 3.0692059458066596e-06,
      "loss": 0.74448985,
      "num_input_tokens_seen": 121768565,
      "step": 5669,
      "time_per_iteration": 2.8021633625030518
    },
    {
      "auxiliary_loss_clip": 0.01097468,
      "auxiliary_loss_mlp": 0.00792473,
      "balance_loss_clip": 1.04791343,
      "balance_loss_mlp": 1.01242328,
      "epoch": 0.3408988426273861,
      "flos": 17085435287040.0,
      "grad_norm": 1.9300508828493999,
      "language_loss": 0.80458981,
      "learning_rate": 3.0688767901673265e-06,
      "loss": 0.82348919,
      "num_input_tokens_seen": 121784925,
      "step": 5670,
      "time_per_iteration": 3.548823595046997
    },
    {
      "auxiliary_loss_clip": 0.01084916,
      "auxiliary_loss_mlp": 0.0103529,
      "balance_loss_clip": 1.0431633,
      "balance_loss_mlp": 1.0206691,
      "epoch": 0.3409589658800541,
      "flos": 24024562151040.0,
      "grad_norm": 1.8327681834622422,
      "language_loss": 0.77530444,
      "learning_rate": 3.068547593996078e-06,
      "loss": 0.79650652,
      "num_input_tokens_seen": 121804425,
      "step": 5671,
      "time_per_iteration": 2.6493420600891113
    },
    {
      "auxiliary_loss_clip": 0.0113367,
      "auxiliary_loss_mlp": 0.00793258,
      "balance_loss_clip": 1.04912543,
      "balance_loss_mlp": 1.01493692,
      "epoch": 0.34101908913272205,
      "flos": 21142012734720.0,
      "grad_norm": 2.1735867572850687,
      "language_loss": 0.74009454,
      "learning_rate": 3.0682183573053974e-06,
      "loss": 0.75936383,
      "num_input_tokens_seen": 121825145,
      "step": 5672,
      "time_per_iteration": 2.505359411239624
    },
    {
      "auxiliary_loss_clip": 0.01116932,
      "auxiliary_loss_mlp": 0.0104162,
      "balance_loss_clip": 1.04614782,
      "balance_loss_mlp": 1.02607548,
      "epoch": 0.3410792123853901,
      "flos": 15702012921600.0,
      "grad_norm": 1.872715424960896,
      "language_loss": 0.73440754,
      "learning_rate": 3.06788908010777e-06,
      "loss": 0.75599307,
      "num_input_tokens_seen": 121842185,
      "step": 5673,
      "time_per_iteration": 4.238344669342041
    },
    {
      "auxiliary_loss_clip": 0.01122337,
      "auxiliary_loss_mlp": 0.01036204,
      "balance_loss_clip": 1.04915595,
      "balance_loss_mlp": 1.02151752,
      "epoch": 0.34113933563805804,
      "flos": 23036012974080.0,
      "grad_norm": 1.7698245463823359,
      "language_loss": 0.79754424,
      "learning_rate": 3.067559762415682e-06,
      "loss": 0.81912965,
      "num_input_tokens_seen": 121862260,
      "step": 5674,
      "time_per_iteration": 2.549967050552368
    },
    {
      "auxiliary_loss_clip": 0.01048861,
      "auxiliary_loss_mlp": 0.01002786,
      "balance_loss_clip": 1.02037501,
      "balance_loss_mlp": 1.00025868,
      "epoch": 0.341199458890726,
      "flos": 69614235336960.0,
      "grad_norm": 1.245975616558794,
      "language_loss": 0.56058705,
      "learning_rate": 3.0672304042416198e-06,
      "loss": 0.58110356,
      "num_input_tokens_seen": 121923560,
      "step": 5675,
      "time_per_iteration": 3.234285593032837
    },
    {
      "auxiliary_loss_clip": 0.01111321,
      "auxiliary_loss_mlp": 0.00792577,
      "balance_loss_clip": 1.04987335,
      "balance_loss_mlp": 1.01455426,
      "epoch": 0.34125958214339397,
      "flos": 22346348866560.0,
      "grad_norm": 1.8667831734763414,
      "language_loss": 0.79397911,
      "learning_rate": 3.0669010055980734e-06,
      "loss": 0.81301808,
      "num_input_tokens_seen": 121943515,
      "step": 5676,
      "time_per_iteration": 3.9644792079925537
    },
    {
      "auxiliary_loss_clip": 0.01116355,
      "auxiliary_loss_mlp": 0.01031173,
      "balance_loss_clip": 1.0438205,
      "balance_loss_mlp": 1.0158782,
      "epoch": 0.34131970539606193,
      "flos": 21871933009920.0,
      "grad_norm": 2.063781887135893,
      "language_loss": 0.85403764,
      "learning_rate": 3.0665715664975357e-06,
      "loss": 0.87551296,
      "num_input_tokens_seen": 121962540,
      "step": 5677,
      "time_per_iteration": 3.9514260292053223
    },
    {
      "auxiliary_loss_clip": 0.01106829,
      "auxiliary_loss_mlp": 0.01037491,
      "balance_loss_clip": 1.04931426,
      "balance_loss_mlp": 1.02235746,
      "epoch": 0.3413798286487299,
      "flos": 24935723475840.0,
      "grad_norm": 1.851689470051099,
      "language_loss": 0.79373002,
      "learning_rate": 3.0662420869524966e-06,
      "loss": 0.81517327,
      "num_input_tokens_seen": 121979830,
      "step": 5678,
      "time_per_iteration": 2.5567758083343506
    },
    {
      "auxiliary_loss_clip": 0.01117989,
      "auxiliary_loss_mlp": 0.01032435,
      "balance_loss_clip": 1.04474688,
      "balance_loss_mlp": 1.01770043,
      "epoch": 0.34143995190139786,
      "flos": 25374372364800.0,
      "grad_norm": 1.7922020358948438,
      "language_loss": 0.74979281,
      "learning_rate": 3.0659125669754506e-06,
      "loss": 0.77129698,
      "num_input_tokens_seen": 121999055,
      "step": 5679,
      "time_per_iteration": 3.929081678390503
    },
    {
      "auxiliary_loss_clip": 0.01041389,
      "auxiliary_loss_mlp": 0.01002144,
      "balance_loss_clip": 1.02421856,
      "balance_loss_mlp": 0.99999827,
      "epoch": 0.34150007515406583,
      "flos": 67782578129280.0,
      "grad_norm": 0.7146282807856524,
      "language_loss": 0.59485769,
      "learning_rate": 3.0655830065788923e-06,
      "loss": 0.61529303,
      "num_input_tokens_seen": 122067015,
      "step": 5680,
      "time_per_iteration": 3.1907830238342285
    },
    {
      "auxiliary_loss_clip": 0.01106596,
      "auxiliary_loss_mlp": 0.01030639,
      "balance_loss_clip": 1.04587626,
      "balance_loss_mlp": 1.01654291,
      "epoch": 0.3415601984067338,
      "flos": 20302421258880.0,
      "grad_norm": 1.7535210232065053,
      "language_loss": 0.72432548,
      "learning_rate": 3.0652534057753206e-06,
      "loss": 0.74569786,
      "num_input_tokens_seen": 122085295,
      "step": 5681,
      "time_per_iteration": 2.5356457233428955
    },
    {
      "auxiliary_loss_clip": 0.01108874,
      "auxiliary_loss_mlp": 0.01044194,
      "balance_loss_clip": 1.04768205,
      "balance_loss_mlp": 1.03000855,
      "epoch": 0.34162032165940176,
      "flos": 26031178506240.0,
      "grad_norm": 3.4828876579297767,
      "language_loss": 0.71029687,
      "learning_rate": 3.064923764577233e-06,
      "loss": 0.7318275,
      "num_input_tokens_seen": 122104020,
      "step": 5682,
      "time_per_iteration": 2.5861222743988037
    },
    {
      "auxiliary_loss_clip": 0.01129512,
      "auxiliary_loss_mlp": 0.01037754,
      "balance_loss_clip": 1.0455128,
      "balance_loss_mlp": 1.02305579,
      "epoch": 0.3416804449120697,
      "flos": 28803338449920.0,
      "grad_norm": 1.6045328575230626,
      "language_loss": 0.83801365,
      "learning_rate": 3.0645940829971295e-06,
      "loss": 0.85968637,
      "num_input_tokens_seen": 122125080,
      "step": 5683,
      "time_per_iteration": 2.530492067337036
    },
    {
      "auxiliary_loss_clip": 0.01111478,
      "auxiliary_loss_mlp": 0.01047235,
      "balance_loss_clip": 1.04813743,
      "balance_loss_mlp": 1.03218484,
      "epoch": 0.3417405681647377,
      "flos": 22601601889920.0,
      "grad_norm": 1.6631889328829361,
      "language_loss": 0.70458078,
      "learning_rate": 3.0642643610475116e-06,
      "loss": 0.72616798,
      "num_input_tokens_seen": 122146350,
      "step": 5684,
      "time_per_iteration": 2.629939317703247
    },
    {
      "auxiliary_loss_clip": 0.01128772,
      "auxiliary_loss_mlp": 0.01031531,
      "balance_loss_clip": 1.04715002,
      "balance_loss_mlp": 1.01769042,
      "epoch": 0.34180069141740566,
      "flos": 24716237420160.0,
      "grad_norm": 1.3529969663233177,
      "language_loss": 0.75151646,
      "learning_rate": 3.0639345987408823e-06,
      "loss": 0.77311951,
      "num_input_tokens_seen": 122168085,
      "step": 5685,
      "time_per_iteration": 2.5329973697662354
    },
    {
      "auxiliary_loss_clip": 0.01112496,
      "auxiliary_loss_mlp": 0.01041535,
      "balance_loss_clip": 1.0444175,
      "balance_loss_mlp": 1.02691436,
      "epoch": 0.3418608146700737,
      "flos": 30518755246080.0,
      "grad_norm": 2.1529275824714267,
      "language_loss": 0.70438361,
      "learning_rate": 3.0636047960897468e-06,
      "loss": 0.7259239,
      "num_input_tokens_seen": 122191040,
      "step": 5686,
      "time_per_iteration": 2.5817360877990723
    },
    {
      "auxiliary_loss_clip": 0.01120411,
      "auxiliary_loss_mlp": 0.01044068,
      "balance_loss_clip": 1.0471158,
      "balance_loss_mlp": 1.02877307,
      "epoch": 0.34192093792274164,
      "flos": 15122343237120.0,
      "grad_norm": 1.9392015101583955,
      "language_loss": 0.7749486,
      "learning_rate": 3.06327495310661e-06,
      "loss": 0.79659337,
      "num_input_tokens_seen": 122209225,
      "step": 5687,
      "time_per_iteration": 2.4860329627990723
    },
    {
      "auxiliary_loss_clip": 0.01103214,
      "auxiliary_loss_mlp": 0.01036005,
      "balance_loss_clip": 1.04832828,
      "balance_loss_mlp": 1.02118707,
      "epoch": 0.3419810611754096,
      "flos": 13187799521280.0,
      "grad_norm": 1.750346626851128,
      "language_loss": 0.86822718,
      "learning_rate": 3.062945069803981e-06,
      "loss": 0.88961935,
      "num_input_tokens_seen": 122226160,
      "step": 5688,
      "time_per_iteration": 2.5126590728759766
    },
    {
      "auxiliary_loss_clip": 0.01119674,
      "auxiliary_loss_mlp": 0.01038386,
      "balance_loss_clip": 1.04751062,
      "balance_loss_mlp": 1.02284098,
      "epoch": 0.34204118442807757,
      "flos": 19536267139200.0,
      "grad_norm": 1.6894257675095075,
      "language_loss": 0.79903233,
      "learning_rate": 3.0626151461943684e-06,
      "loss": 0.82061291,
      "num_input_tokens_seen": 122243115,
      "step": 5689,
      "time_per_iteration": 2.515993356704712
    },
    {
      "auxiliary_loss_clip": 0.01123746,
      "auxiliary_loss_mlp": 0.01039358,
      "balance_loss_clip": 1.04777002,
      "balance_loss_mlp": 1.02437329,
      "epoch": 0.34210130768074554,
      "flos": 15194846839680.0,
      "grad_norm": 1.92160976041539,
      "language_loss": 0.7376225,
      "learning_rate": 3.0622851822902834e-06,
      "loss": 0.75925356,
      "num_input_tokens_seen": 122261105,
      "step": 5690,
      "time_per_iteration": 2.4873716831207275
    },
    {
      "auxiliary_loss_clip": 0.0111335,
      "auxiliary_loss_mlp": 0.0103636,
      "balance_loss_clip": 1.04337406,
      "balance_loss_mlp": 1.02130353,
      "epoch": 0.3421614309334135,
      "flos": 24936226266240.0,
      "grad_norm": 1.7211127251650544,
      "language_loss": 0.75679612,
      "learning_rate": 3.061955178104237e-06,
      "loss": 0.77829319,
      "num_input_tokens_seen": 122279995,
      "step": 5691,
      "time_per_iteration": 2.5226752758026123
    },
    {
      "auxiliary_loss_clip": 0.01115209,
      "auxiliary_loss_mlp": 0.0103472,
      "balance_loss_clip": 1.045187,
      "balance_loss_mlp": 1.02133286,
      "epoch": 0.34222155418608147,
      "flos": 21908633731200.0,
      "grad_norm": 1.8235103448354448,
      "language_loss": 0.6795457,
      "learning_rate": 3.0616251336487447e-06,
      "loss": 0.70104498,
      "num_input_tokens_seen": 122299070,
      "step": 5692,
      "time_per_iteration": 2.522111177444458
    },
    {
      "auxiliary_loss_clip": 0.01119891,
      "auxiliary_loss_mlp": 0.01036678,
      "balance_loss_clip": 1.0467689,
      "balance_loss_mlp": 1.02169359,
      "epoch": 0.34228167743874943,
      "flos": 18114061063680.0,
      "grad_norm": 2.108045862238099,
      "language_loss": 0.72894788,
      "learning_rate": 3.06129504893632e-06,
      "loss": 0.75051355,
      "num_input_tokens_seen": 122316800,
      "step": 5693,
      "time_per_iteration": 2.4905612468719482
    },
    {
      "auxiliary_loss_clip": 0.01087615,
      "auxiliary_loss_mlp": 0.01041149,
      "balance_loss_clip": 1.04446435,
      "balance_loss_mlp": 1.02713013,
      "epoch": 0.3423418006914174,
      "flos": 21288600138240.0,
      "grad_norm": 7.5947264208531085,
      "language_loss": 0.75097442,
      "learning_rate": 3.0609649239794813e-06,
      "loss": 0.77226204,
      "num_input_tokens_seen": 122335275,
      "step": 5694,
      "time_per_iteration": 2.561955451965332
    },
    {
      "auxiliary_loss_clip": 0.01089518,
      "auxiliary_loss_mlp": 0.010373,
      "balance_loss_clip": 1.04769492,
      "balance_loss_mlp": 1.02375841,
      "epoch": 0.34240192394408536,
      "flos": 19823480288640.0,
      "grad_norm": 1.849644282803971,
      "language_loss": 0.79685074,
      "learning_rate": 3.060634758790747e-06,
      "loss": 0.81811887,
      "num_input_tokens_seen": 122353215,
      "step": 5695,
      "time_per_iteration": 2.5597431659698486
    },
    {
      "auxiliary_loss_clip": 0.01077687,
      "auxiliary_loss_mlp": 0.01040897,
      "balance_loss_clip": 1.04354024,
      "balance_loss_mlp": 1.02640128,
      "epoch": 0.3424620471967533,
      "flos": 24535535074560.0,
      "grad_norm": 1.6324055721219808,
      "language_loss": 0.7356804,
      "learning_rate": 3.060304553382635e-06,
      "loss": 0.75686622,
      "num_input_tokens_seen": 122372495,
      "step": 5696,
      "time_per_iteration": 2.624439001083374
    },
    {
      "auxiliary_loss_clip": 0.01094564,
      "auxiliary_loss_mlp": 0.01055637,
      "balance_loss_clip": 1.04829872,
      "balance_loss_mlp": 1.03990149,
      "epoch": 0.3425221704494213,
      "flos": 25848895962240.0,
      "grad_norm": 1.6897783954598737,
      "language_loss": 0.70765203,
      "learning_rate": 3.0599743077676685e-06,
      "loss": 0.72915399,
      "num_input_tokens_seen": 122394600,
      "step": 5697,
      "time_per_iteration": 2.6139001846313477
    },
    {
      "auxiliary_loss_clip": 0.01111009,
      "auxiliary_loss_mlp": 0.01032898,
      "balance_loss_clip": 1.04961157,
      "balance_loss_mlp": 1.01870036,
      "epoch": 0.34258229370208926,
      "flos": 21540513196800.0,
      "grad_norm": 1.94705850959536,
      "language_loss": 0.82419479,
      "learning_rate": 3.05964402195837e-06,
      "loss": 0.84563386,
      "num_input_tokens_seen": 122414700,
      "step": 5698,
      "time_per_iteration": 2.5791268348693848
    },
    {
      "auxiliary_loss_clip": 0.01079567,
      "auxiliary_loss_mlp": 0.01049956,
      "balance_loss_clip": 1.04466009,
      "balance_loss_mlp": 1.03151441,
      "epoch": 0.3426424169547573,
      "flos": 23652778429440.0,
      "grad_norm": 1.7503673228858105,
      "language_loss": 0.69028735,
      "learning_rate": 3.0593136959672645e-06,
      "loss": 0.71158254,
      "num_input_tokens_seen": 122432760,
      "step": 5699,
      "time_per_iteration": 2.6564457416534424
    },
    {
      "auxiliary_loss_clip": 0.01108427,
      "auxiliary_loss_mlp": 0.01035567,
      "balance_loss_clip": 1.04644513,
      "balance_loss_mlp": 1.02148223,
      "epoch": 0.34270254020742524,
      "flos": 24644883052800.0,
      "grad_norm": 2.0294204050599003,
      "language_loss": 0.72586113,
      "learning_rate": 3.058983329806877e-06,
      "loss": 0.74730098,
      "num_input_tokens_seen": 122449105,
      "step": 5700,
      "time_per_iteration": 2.579179048538208
    },
    {
      "auxiliary_loss_clip": 0.01102843,
      "auxiliary_loss_mlp": 0.01033153,
      "balance_loss_clip": 1.0489527,
      "balance_loss_mlp": 1.01965237,
      "epoch": 0.3427626634600932,
      "flos": 20996754134400.0,
      "grad_norm": 1.808419279637055,
      "language_loss": 0.82007027,
      "learning_rate": 3.0586529234897354e-06,
      "loss": 0.84143019,
      "num_input_tokens_seen": 122468700,
      "step": 5701,
      "time_per_iteration": 2.5421142578125
    },
    {
      "auxiliary_loss_clip": 0.01113807,
      "auxiliary_loss_mlp": 0.01032218,
      "balance_loss_clip": 1.04644346,
      "balance_loss_mlp": 1.0183177,
      "epoch": 0.3428227867127612,
      "flos": 21433786911360.0,
      "grad_norm": 1.6211516508197004,
      "language_loss": 0.71206844,
      "learning_rate": 3.0583224770283694e-06,
      "loss": 0.73352861,
      "num_input_tokens_seen": 122488160,
      "step": 5702,
      "time_per_iteration": 2.513777494430542
    },
    {
      "auxiliary_loss_clip": 0.01035916,
      "auxiliary_loss_mlp": 0.01006506,
      "balance_loss_clip": 1.03588808,
      "balance_loss_mlp": 1.0043366,
      "epoch": 0.34288290996542914,
      "flos": 55731782695680.0,
      "grad_norm": 0.7929841012532431,
      "language_loss": 0.57401109,
      "learning_rate": 3.057991990435309e-06,
      "loss": 0.59443533,
      "num_input_tokens_seen": 122542890,
      "step": 5703,
      "time_per_iteration": 3.0223841667175293
    },
    {
      "auxiliary_loss_clip": 0.01119227,
      "auxiliary_loss_mlp": 0.01040621,
      "balance_loss_clip": 1.04743934,
      "balance_loss_mlp": 1.0249989,
      "epoch": 0.3429430332180971,
      "flos": 20156803522560.0,
      "grad_norm": 1.7997286723572108,
      "language_loss": 0.74340427,
      "learning_rate": 3.057661463723086e-06,
      "loss": 0.76500273,
      "num_input_tokens_seen": 122561770,
      "step": 5704,
      "time_per_iteration": 2.521563768386841
    },
    {
      "auxiliary_loss_clip": 0.01100733,
      "auxiliary_loss_mlp": 0.01041631,
      "balance_loss_clip": 1.04660726,
      "balance_loss_mlp": 1.02860093,
      "epoch": 0.34300315647076507,
      "flos": 17965857548160.0,
      "grad_norm": 1.9523696512974578,
      "language_loss": 0.72830284,
      "learning_rate": 3.0573308969042346e-06,
      "loss": 0.74972641,
      "num_input_tokens_seen": 122580580,
      "step": 5705,
      "time_per_iteration": 2.597683906555176
    },
    {
      "auxiliary_loss_clip": 0.01085546,
      "auxiliary_loss_mlp": 0.01035446,
      "balance_loss_clip": 1.04298496,
      "balance_loss_mlp": 1.02101541,
      "epoch": 0.34306327972343303,
      "flos": 22086822124800.0,
      "grad_norm": 1.98706037545995,
      "language_loss": 0.80058694,
      "learning_rate": 3.057000289991289e-06,
      "loss": 0.82179689,
      "num_input_tokens_seen": 122599810,
      "step": 5706,
      "time_per_iteration": 2.589869737625122
    },
    {
      "auxiliary_loss_clip": 0.01113707,
      "auxiliary_loss_mlp": 0.01033211,
      "balance_loss_clip": 1.04834807,
      "balance_loss_mlp": 1.01878083,
      "epoch": 0.343123402976101,
      "flos": 18442679616000.0,
      "grad_norm": 3.3997122089974274,
      "language_loss": 0.82996064,
      "learning_rate": 3.056669642996787e-06,
      "loss": 0.85142982,
      "num_input_tokens_seen": 122616035,
      "step": 5707,
      "time_per_iteration": 2.4882290363311768
    },
    {
      "auxiliary_loss_clip": 0.01120917,
      "auxiliary_loss_mlp": 0.01031263,
      "balance_loss_clip": 1.04911935,
      "balance_loss_mlp": 1.01712477,
      "epoch": 0.34318352622876896,
      "flos": 17163685065600.0,
      "grad_norm": 1.8754806185084165,
      "language_loss": 0.74905574,
      "learning_rate": 3.056338955933266e-06,
      "loss": 0.77057755,
      "num_input_tokens_seen": 122633785,
      "step": 5708,
      "time_per_iteration": 2.4715514183044434
    },
    {
      "auxiliary_loss_clip": 0.01097198,
      "auxiliary_loss_mlp": 0.01036841,
      "balance_loss_clip": 1.04433513,
      "balance_loss_mlp": 1.02314353,
      "epoch": 0.34324364948143693,
      "flos": 26688164215680.0,
      "grad_norm": 1.6356228858761876,
      "language_loss": 0.81198543,
      "learning_rate": 3.0560082288132662e-06,
      "loss": 0.8333258,
      "num_input_tokens_seen": 122652100,
      "step": 5709,
      "time_per_iteration": 2.561706304550171
    },
    {
      "auxiliary_loss_clip": 0.01109225,
      "auxiliary_loss_mlp": 0.01041432,
      "balance_loss_clip": 1.04690897,
      "balance_loss_mlp": 1.02567863,
      "epoch": 0.3433037727341049,
      "flos": 21251576194560.0,
      "grad_norm": 2.095245398137087,
      "language_loss": 0.78742772,
      "learning_rate": 3.055677461649329e-06,
      "loss": 0.80893427,
      "num_input_tokens_seen": 122669720,
      "step": 5710,
      "time_per_iteration": 2.5305063724517822
    },
    {
      "auxiliary_loss_clip": 0.01120379,
      "auxiliary_loss_mlp": 0.01036491,
      "balance_loss_clip": 1.04589629,
      "balance_loss_mlp": 1.02122593,
      "epoch": 0.34336389598677286,
      "flos": 20629423699200.0,
      "grad_norm": 1.8374974785896627,
      "language_loss": 0.70443261,
      "learning_rate": 3.055346654453996e-06,
      "loss": 0.72600126,
      "num_input_tokens_seen": 122688715,
      "step": 5711,
      "time_per_iteration": 3.9133830070495605
    },
    {
      "auxiliary_loss_clip": 0.01094939,
      "auxiliary_loss_mlp": 0.00793821,
      "balance_loss_clip": 1.04353321,
      "balance_loss_mlp": 1.01639199,
      "epoch": 0.3434240192394409,
      "flos": 14538579402240.0,
      "grad_norm": 1.7961413868105884,
      "language_loss": 0.66999364,
      "learning_rate": 3.055015807239812e-06,
      "loss": 0.68888128,
      "num_input_tokens_seen": 122706970,
      "step": 5712,
      "time_per_iteration": 2.5162129402160645
    },
    {
      "auxiliary_loss_clip": 0.01018826,
      "auxiliary_loss_mlp": 0.01011539,
      "balance_loss_clip": 1.02230716,
      "balance_loss_mlp": 1.00950098,
      "epoch": 0.34348414249210885,
      "flos": 58051538841600.0,
      "grad_norm": 0.8469514412611118,
      "language_loss": 0.58112222,
      "learning_rate": 3.0546849200193226e-06,
      "loss": 0.60142589,
      "num_input_tokens_seen": 122758095,
      "step": 5713,
      "time_per_iteration": 3.1269378662109375
    },
    {
      "auxiliary_loss_clip": 0.01130573,
      "auxiliary_loss_mlp": 0.01039821,
      "balance_loss_clip": 1.04695427,
      "balance_loss_mlp": 1.02540803,
      "epoch": 0.3435442657447768,
      "flos": 20704441253760.0,
      "grad_norm": 1.7848104135568157,
      "language_loss": 0.80742419,
      "learning_rate": 3.054353992805076e-06,
      "loss": 0.82912809,
      "num_input_tokens_seen": 122777815,
      "step": 5714,
      "time_per_iteration": 3.9391515254974365
    },
    {
      "auxiliary_loss_clip": 0.01129218,
      "auxiliary_loss_mlp": 0.01039389,
      "balance_loss_clip": 1.04644632,
      "balance_loss_mlp": 1.02446985,
      "epoch": 0.3436043889974448,
      "flos": 22930256355840.0,
      "grad_norm": 1.8432215189574783,
      "language_loss": 0.72314036,
      "learning_rate": 3.05402302560962e-06,
      "loss": 0.74482644,
      "num_input_tokens_seen": 122797555,
      "step": 5715,
      "time_per_iteration": 3.881129503250122
    },
    {
      "auxiliary_loss_clip": 0.01032908,
      "auxiliary_loss_mlp": 0.0100322,
      "balance_loss_clip": 1.020015,
      "balance_loss_mlp": 1.00106263,
      "epoch": 0.34366451225011274,
      "flos": 58403285752320.0,
      "grad_norm": 0.894190752780206,
      "language_loss": 0.65825307,
      "learning_rate": 3.053692018445505e-06,
      "loss": 0.67861432,
      "num_input_tokens_seen": 122863955,
      "step": 5716,
      "time_per_iteration": 3.132615804672241
    },
    {
      "auxiliary_loss_clip": 0.01118548,
      "auxiliary_loss_mlp": 0.01042479,
      "balance_loss_clip": 1.05299509,
      "balance_loss_mlp": 1.02757812,
      "epoch": 0.3437246355027807,
      "flos": 15596292216960.0,
      "grad_norm": 1.7165858454587368,
      "language_loss": 0.73807061,
      "learning_rate": 3.0533609713252838e-06,
      "loss": 0.75968087,
      "num_input_tokens_seen": 122883000,
      "step": 5717,
      "time_per_iteration": 2.5186049938201904
    },
    {
      "auxiliary_loss_clip": 0.01075882,
      "auxiliary_loss_mlp": 0.01036853,
      "balance_loss_clip": 1.04658663,
      "balance_loss_mlp": 1.0232091,
      "epoch": 0.34378475875544867,
      "flos": 27672260106240.0,
      "grad_norm": 1.6239063083827567,
      "language_loss": 0.7564429,
      "learning_rate": 3.0530298842615077e-06,
      "loss": 0.77757031,
      "num_input_tokens_seen": 122903265,
      "step": 5718,
      "time_per_iteration": 4.009954214096069
    },
    {
      "auxiliary_loss_clip": 0.0109336,
      "auxiliary_loss_mlp": 0.01046333,
      "balance_loss_clip": 1.04817104,
      "balance_loss_mlp": 1.03115189,
      "epoch": 0.34384488200811664,
      "flos": 31431496769280.0,
      "grad_norm": 1.8350036771739286,
      "language_loss": 0.6398294,
      "learning_rate": 3.052698757266734e-06,
      "loss": 0.66122639,
      "num_input_tokens_seen": 122923860,
      "step": 5719,
      "time_per_iteration": 2.6446497440338135
    },
    {
      "auxiliary_loss_clip": 0.01087221,
      "auxiliary_loss_mlp": 0.01038605,
      "balance_loss_clip": 1.04416358,
      "balance_loss_mlp": 1.02190387,
      "epoch": 0.3439050052607846,
      "flos": 24899920594560.0,
      "grad_norm": 1.6799516539082462,
      "language_loss": 0.73821795,
      "learning_rate": 3.0523675903535183e-06,
      "loss": 0.75947618,
      "num_input_tokens_seen": 122945305,
      "step": 5720,
      "time_per_iteration": 2.6081628799438477
    },
    {
      "auxiliary_loss_clip": 0.01113698,
      "auxiliary_loss_mlp": 0.01049324,
      "balance_loss_clip": 1.0437721,
      "balance_loss_mlp": 1.03251576,
      "epoch": 0.34396512851345257,
      "flos": 18150079426560.0,
      "grad_norm": 1.6222564078771202,
      "language_loss": 0.74273765,
      "learning_rate": 3.0520363835344173e-06,
      "loss": 0.76436788,
      "num_input_tokens_seen": 122962535,
      "step": 5721,
      "time_per_iteration": 2.474538564682007
    },
    {
      "auxiliary_loss_clip": 0.01111049,
      "auxiliary_loss_mlp": 0.00796407,
      "balance_loss_clip": 1.05333543,
      "balance_loss_mlp": 1.01797044,
      "epoch": 0.34402525176612053,
      "flos": 16034438315520.0,
      "grad_norm": 1.9726514238743618,
      "language_loss": 0.79943669,
      "learning_rate": 3.051705136821992e-06,
      "loss": 0.81851125,
      "num_input_tokens_seen": 122979750,
      "step": 5722,
      "time_per_iteration": 2.5066146850585938
    },
    {
      "auxiliary_loss_clip": 0.01079007,
      "auxiliary_loss_mlp": 0.01032068,
      "balance_loss_clip": 1.04562068,
      "balance_loss_mlp": 1.01846611,
      "epoch": 0.3440853750187885,
      "flos": 21178641628800.0,
      "grad_norm": 2.212307655994471,
      "language_loss": 0.81790084,
      "learning_rate": 3.051373850228801e-06,
      "loss": 0.83901155,
      "num_input_tokens_seen": 122998955,
      "step": 5723,
      "time_per_iteration": 2.6435256004333496
    },
    {
      "auxiliary_loss_clip": 0.0109358,
      "auxiliary_loss_mlp": 0.01050386,
      "balance_loss_clip": 1.04387426,
      "balance_loss_mlp": 1.03409004,
      "epoch": 0.34414549827145646,
      "flos": 12677868092160.0,
      "grad_norm": 1.8472328405445384,
      "language_loss": 0.80935365,
      "learning_rate": 3.0510425237674096e-06,
      "loss": 0.83079338,
      "num_input_tokens_seen": 123016165,
      "step": 5724,
      "time_per_iteration": 2.515774965286255
    },
    {
      "auxiliary_loss_clip": 0.01102317,
      "auxiliary_loss_mlp": 0.01043338,
      "balance_loss_clip": 1.04343581,
      "balance_loss_mlp": 1.02682769,
      "epoch": 0.3442056215241244,
      "flos": 31284514316160.0,
      "grad_norm": 1.71552655143069,
      "language_loss": 0.68730938,
      "learning_rate": 3.05071115745038e-06,
      "loss": 0.70876598,
      "num_input_tokens_seen": 123036900,
      "step": 5725,
      "time_per_iteration": 2.60569429397583
    },
    {
      "auxiliary_loss_clip": 0.01120135,
      "auxiliary_loss_mlp": 0.01044824,
      "balance_loss_clip": 1.04667759,
      "balance_loss_mlp": 1.02792013,
      "epoch": 0.34426574477679245,
      "flos": 23367289132800.0,
      "grad_norm": 1.3409587827898828,
      "language_loss": 0.69242626,
      "learning_rate": 3.0503797512902773e-06,
      "loss": 0.71407586,
      "num_input_tokens_seen": 123057480,
      "step": 5726,
      "time_per_iteration": 2.5077943801879883
    },
    {
      "auxiliary_loss_clip": 0.01099304,
      "auxiliary_loss_mlp": 0.01041826,
      "balance_loss_clip": 1.04617119,
      "balance_loss_mlp": 1.0280093,
      "epoch": 0.3443258680294604,
      "flos": 24535427333760.0,
      "grad_norm": 1.6817234128163239,
      "language_loss": 0.73005056,
      "learning_rate": 3.0500483052996703e-06,
      "loss": 0.75146192,
      "num_input_tokens_seen": 123076890,
      "step": 5727,
      "time_per_iteration": 2.597428321838379
    },
    {
      "auxiliary_loss_clip": 0.01090028,
      "auxiliary_loss_mlp": 0.01039791,
      "balance_loss_clip": 1.04428828,
      "balance_loss_mlp": 1.02435303,
      "epoch": 0.3443859912821284,
      "flos": 20230133137920.0,
      "grad_norm": 1.7785759388353075,
      "language_loss": 0.88125885,
      "learning_rate": 3.0497168194911257e-06,
      "loss": 0.90255702,
      "num_input_tokens_seen": 123092530,
      "step": 5728,
      "time_per_iteration": 2.548021078109741
    },
    {
      "auxiliary_loss_clip": 0.01086334,
      "auxiliary_loss_mlp": 0.01043323,
      "balance_loss_clip": 1.05194914,
      "balance_loss_mlp": 1.02857649,
      "epoch": 0.34444611453479634,
      "flos": 24316515895680.0,
      "grad_norm": 1.937217359294808,
      "language_loss": 0.70012105,
      "learning_rate": 3.0493852938772143e-06,
      "loss": 0.72141767,
      "num_input_tokens_seen": 123110560,
      "step": 5729,
      "time_per_iteration": 2.61147403717041
    },
    {
      "auxiliary_loss_clip": 0.01115147,
      "auxiliary_loss_mlp": 0.01032406,
      "balance_loss_clip": 1.04565525,
      "balance_loss_mlp": 1.01780331,
      "epoch": 0.3445062377874643,
      "flos": 16983413683200.0,
      "grad_norm": 2.2589660532263927,
      "language_loss": 0.73904788,
      "learning_rate": 3.0490537284705078e-06,
      "loss": 0.76052344,
      "num_input_tokens_seen": 123128655,
      "step": 5730,
      "time_per_iteration": 2.465240716934204
    },
    {
      "auxiliary_loss_clip": 0.01091006,
      "auxiliary_loss_mlp": 0.01048322,
      "balance_loss_clip": 1.04208827,
      "balance_loss_mlp": 1.03193116,
      "epoch": 0.3445663610401323,
      "flos": 20302708567680.0,
      "grad_norm": 2.043328677408356,
      "language_loss": 0.79610717,
      "learning_rate": 3.048722123283578e-06,
      "loss": 0.81750053,
      "num_input_tokens_seen": 123145130,
      "step": 5731,
      "time_per_iteration": 2.5481083393096924
    },
    {
      "auxiliary_loss_clip": 0.01118139,
      "auxiliary_loss_mlp": 0.01036844,
      "balance_loss_clip": 1.04501152,
      "balance_loss_mlp": 1.02233028,
      "epoch": 0.34462648429280024,
      "flos": 15888102307200.0,
      "grad_norm": 2.0607995050496863,
      "language_loss": 0.78710258,
      "learning_rate": 3.0483904783290006e-06,
      "loss": 0.8086524,
      "num_input_tokens_seen": 123162265,
      "step": 5732,
      "time_per_iteration": 2.4600865840911865
    },
    {
      "auxiliary_loss_clip": 0.01016771,
      "auxiliary_loss_mlp": 0.01003446,
      "balance_loss_clip": 1.0188992,
      "balance_loss_mlp": 1.00147939,
      "epoch": 0.3446866075454682,
      "flos": 59311035285120.0,
      "grad_norm": 0.7427556907238304,
      "language_loss": 0.53545117,
      "learning_rate": 3.0480587936193505e-06,
      "loss": 0.55565333,
      "num_input_tokens_seen": 123218620,
      "step": 5733,
      "time_per_iteration": 3.1845645904541016
    },
    {
      "auxiliary_loss_clip": 0.01111863,
      "auxiliary_loss_mlp": 0.01038275,
      "balance_loss_clip": 1.04751742,
      "balance_loss_mlp": 1.02314782,
      "epoch": 0.34474673079813617,
      "flos": 22343799000960.0,
      "grad_norm": 2.055627067095104,
      "language_loss": 0.83521259,
      "learning_rate": 3.047727069167207e-06,
      "loss": 0.85671401,
      "num_input_tokens_seen": 123237325,
      "step": 5734,
      "time_per_iteration": 2.525757312774658
    },
    {
      "auxiliary_loss_clip": 0.01108431,
      "auxiliary_loss_mlp": 0.01034507,
      "balance_loss_clip": 1.04631579,
      "balance_loss_mlp": 1.01990414,
      "epoch": 0.34480685405080413,
      "flos": 27670141203840.0,
      "grad_norm": 1.8371320533577185,
      "language_loss": 0.92572904,
      "learning_rate": 3.0473953049851478e-06,
      "loss": 0.9471584,
      "num_input_tokens_seen": 123258650,
      "step": 5735,
      "time_per_iteration": 2.580904960632324
    },
    {
      "auxiliary_loss_clip": 0.01090967,
      "auxiliary_loss_mlp": 0.01038424,
      "balance_loss_clip": 1.04761338,
      "balance_loss_mlp": 1.0231117,
      "epoch": 0.3448669773034721,
      "flos": 22456020067200.0,
      "grad_norm": 2.1082194557465805,
      "language_loss": 0.76913357,
      "learning_rate": 3.0470635010857533e-06,
      "loss": 0.79042745,
      "num_input_tokens_seen": 123277155,
      "step": 5736,
      "time_per_iteration": 2.6083786487579346
    },
    {
      "auxiliary_loss_clip": 0.01109842,
      "auxiliary_loss_mlp": 0.01039175,
      "balance_loss_clip": 1.04582214,
      "balance_loss_mlp": 1.02478611,
      "epoch": 0.34492710055614006,
      "flos": 24936190352640.0,
      "grad_norm": 1.583783552807668,
      "language_loss": 0.78856671,
      "learning_rate": 3.0467316574816064e-06,
      "loss": 0.81005687,
      "num_input_tokens_seen": 123297640,
      "step": 5737,
      "time_per_iteration": 2.5422370433807373
    },
    {
      "auxiliary_loss_clip": 0.01084691,
      "auxiliary_loss_mlp": 0.01048588,
      "balance_loss_clip": 1.04155731,
      "balance_loss_mlp": 1.03093326,
      "epoch": 0.34498722380880803,
      "flos": 20120821073280.0,
      "grad_norm": 1.7780168905197622,
      "language_loss": 0.71362531,
      "learning_rate": 3.0463997741852893e-06,
      "loss": 0.73495805,
      "num_input_tokens_seen": 123314370,
      "step": 5738,
      "time_per_iteration": 2.569952964782715
    },
    {
      "auxiliary_loss_clip": 0.01092197,
      "auxiliary_loss_mlp": 0.0104044,
      "balance_loss_clip": 1.04213834,
      "balance_loss_mlp": 1.02439475,
      "epoch": 0.34504734706147605,
      "flos": 28438126917120.0,
      "grad_norm": 2.054872494098109,
      "language_loss": 0.81606579,
      "learning_rate": 3.046067851209389e-06,
      "loss": 0.83739215,
      "num_input_tokens_seen": 123336085,
      "step": 5739,
      "time_per_iteration": 2.6183180809020996
    },
    {
      "auxiliary_loss_clip": 0.0109791,
      "auxiliary_loss_mlp": 0.0103903,
      "balance_loss_clip": 1.04581046,
      "balance_loss_mlp": 1.02318728,
      "epoch": 0.345107470314144,
      "flos": 22674464628480.0,
      "grad_norm": 1.993210083260518,
      "language_loss": 0.83063483,
      "learning_rate": 3.0457358885664898e-06,
      "loss": 0.85200429,
      "num_input_tokens_seen": 123354460,
      "step": 5740,
      "time_per_iteration": 2.563612461090088
    },
    {
      "auxiliary_loss_clip": 0.01121295,
      "auxiliary_loss_mlp": 0.01036688,
      "balance_loss_clip": 1.04820204,
      "balance_loss_mlp": 1.0203445,
      "epoch": 0.345167593566812,
      "flos": 20630716588800.0,
      "grad_norm": 4.075478657409653,
      "language_loss": 0.76558262,
      "learning_rate": 3.045403886269181e-06,
      "loss": 0.78716242,
      "num_input_tokens_seen": 123373420,
      "step": 5741,
      "time_per_iteration": 2.495298385620117
    },
    {
      "auxiliary_loss_clip": 0.01106825,
      "auxiliary_loss_mlp": 0.01031663,
      "balance_loss_clip": 1.04618788,
      "balance_loss_mlp": 1.01694012,
      "epoch": 0.34522771681947995,
      "flos": 26214358890240.0,
      "grad_norm": 2.188410888477557,
      "language_loss": 0.77120763,
      "learning_rate": 3.045071844330053e-06,
      "loss": 0.79259253,
      "num_input_tokens_seen": 123394730,
      "step": 5742,
      "time_per_iteration": 2.581979274749756
    },
    {
      "auxiliary_loss_clip": 0.01118132,
      "auxiliary_loss_mlp": 0.01035684,
      "balance_loss_clip": 1.04389036,
      "balance_loss_mlp": 1.02091432,
      "epoch": 0.3452878400721479,
      "flos": 19062354072960.0,
      "grad_norm": 2.046197052403095,
      "language_loss": 0.75556755,
      "learning_rate": 3.0447397627616955e-06,
      "loss": 0.77710575,
      "num_input_tokens_seen": 123412895,
      "step": 5743,
      "time_per_iteration": 2.5174050331115723
    },
    {
      "auxiliary_loss_clip": 0.01117526,
      "auxiliary_loss_mlp": 0.01038812,
      "balance_loss_clip": 1.04565287,
      "balance_loss_mlp": 1.02477479,
      "epoch": 0.3453479633248159,
      "flos": 27929739772800.0,
      "grad_norm": 2.0706179695993274,
      "language_loss": 0.7048558,
      "learning_rate": 3.0444076415767016e-06,
      "loss": 0.72641921,
      "num_input_tokens_seen": 123432320,
      "step": 5744,
      "time_per_iteration": 2.564758539199829
    },
    {
      "auxiliary_loss_clip": 0.01127736,
      "auxiliary_loss_mlp": 0.01033516,
      "balance_loss_clip": 1.04577875,
      "balance_loss_mlp": 1.01846027,
      "epoch": 0.34540808657748384,
      "flos": 19606113135360.0,
      "grad_norm": 1.6240809421625528,
      "language_loss": 0.79789662,
      "learning_rate": 3.044075480787665e-06,
      "loss": 0.81950915,
      "num_input_tokens_seen": 123450980,
      "step": 5745,
      "time_per_iteration": 2.4474313259124756
    },
    {
      "auxiliary_loss_clip": 0.0108774,
      "auxiliary_loss_mlp": 0.01038068,
      "balance_loss_clip": 1.04627085,
      "balance_loss_mlp": 1.02247524,
      "epoch": 0.3454682098301518,
      "flos": 20411661496320.0,
      "grad_norm": 1.997813699322109,
      "language_loss": 0.89171344,
      "learning_rate": 3.043743280407182e-06,
      "loss": 0.9129715,
      "num_input_tokens_seen": 123469365,
      "step": 5746,
      "time_per_iteration": 2.6046202182769775
    },
    {
      "auxiliary_loss_clip": 0.01124175,
      "auxiliary_loss_mlp": 0.01038383,
      "balance_loss_clip": 1.04847503,
      "balance_loss_mlp": 1.0228498,
      "epoch": 0.34552833308281977,
      "flos": 21325121291520.0,
      "grad_norm": 1.8169452728700954,
      "language_loss": 0.64456427,
      "learning_rate": 3.043411040447849e-06,
      "loss": 0.66618979,
      "num_input_tokens_seen": 123489425,
      "step": 5747,
      "time_per_iteration": 2.492297410964966
    },
    {
      "auxiliary_loss_clip": 0.01109849,
      "auxiliary_loss_mlp": 0.01034783,
      "balance_loss_clip": 1.04472148,
      "balance_loss_mlp": 1.02013254,
      "epoch": 0.34558845633548774,
      "flos": 36243633824640.0,
      "grad_norm": 1.5217327850553932,
      "language_loss": 0.73002928,
      "learning_rate": 3.043078760922264e-06,
      "loss": 0.75147557,
      "num_input_tokens_seen": 123509970,
      "step": 5748,
      "time_per_iteration": 2.6488916873931885
    },
    {
      "auxiliary_loss_clip": 0.0107795,
      "auxiliary_loss_mlp": 0.01035976,
      "balance_loss_clip": 1.04922724,
      "balance_loss_mlp": 1.02225494,
      "epoch": 0.3456485795881557,
      "flos": 22450561200000.0,
      "grad_norm": 3.022688996848376,
      "language_loss": 0.7558614,
      "learning_rate": 3.042746441843029e-06,
      "loss": 0.77700067,
      "num_input_tokens_seen": 123531055,
      "step": 5749,
      "time_per_iteration": 2.632967233657837
    },
    {
      "auxiliary_loss_clip": 0.01027975,
      "auxiliary_loss_mlp": 0.01004699,
      "balance_loss_clip": 1.01792622,
      "balance_loss_mlp": 1.00244546,
      "epoch": 0.34570870284082367,
      "flos": 62004299005440.0,
      "grad_norm": 0.8827388039971793,
      "language_loss": 0.62660539,
      "learning_rate": 3.0424140832227437e-06,
      "loss": 0.64693213,
      "num_input_tokens_seen": 123584720,
      "step": 5750,
      "time_per_iteration": 4.349289417266846
    },
    {
      "auxiliary_loss_clip": 0.01104448,
      "auxiliary_loss_mlp": 0.01033892,
      "balance_loss_clip": 1.04689884,
      "balance_loss_mlp": 1.0198369,
      "epoch": 0.34576882609349163,
      "flos": 22782196494720.0,
      "grad_norm": 1.7625318178431404,
      "language_loss": 0.80857599,
      "learning_rate": 3.042081685074012e-06,
      "loss": 0.82995939,
      "num_input_tokens_seen": 123604465,
      "step": 5751,
      "time_per_iteration": 2.544708728790283
    },
    {
      "auxiliary_loss_clip": 0.01125335,
      "auxiliary_loss_mlp": 0.01042696,
      "balance_loss_clip": 1.04442215,
      "balance_loss_mlp": 1.02790177,
      "epoch": 0.34582894934615965,
      "flos": 12348818576640.0,
      "grad_norm": 2.2304129204516587,
      "language_loss": 0.83836287,
      "learning_rate": 3.041749247409439e-06,
      "loss": 0.86004317,
      "num_input_tokens_seen": 123622320,
      "step": 5752,
      "time_per_iteration": 2.4377377033233643
    },
    {
      "auxiliary_loss_clip": 0.01022518,
      "auxiliary_loss_mlp": 0.00861033,
      "balance_loss_clip": 1.01479959,
      "balance_loss_mlp": 1.17114151,
      "epoch": 0.3458890725988276,
      "flos": 70167691071360.0,
      "grad_norm": 0.7599886160242099,
      "language_loss": 0.63112593,
      "learning_rate": 3.0414167702416296e-06,
      "loss": 0.64996141,
      "num_input_tokens_seen": 123678010,
      "step": 5753,
      "time_per_iteration": 4.451346158981323
    },
    {
      "auxiliary_loss_clip": 0.01102927,
      "auxiliary_loss_mlp": 0.0103607,
      "balance_loss_clip": 1.04592156,
      "balance_loss_mlp": 1.02096605,
      "epoch": 0.3459491958514956,
      "flos": 17092582093440.0,
      "grad_norm": 1.9421549691834499,
      "language_loss": 0.70807165,
      "learning_rate": 3.0410842535831914e-06,
      "loss": 0.72946167,
      "num_input_tokens_seen": 123696830,
      "step": 5754,
      "time_per_iteration": 3.9976775646209717
    },
    {
      "auxiliary_loss_clip": 0.0111903,
      "auxiliary_loss_mlp": 0.01035601,
      "balance_loss_clip": 1.04542267,
      "balance_loss_mlp": 1.02084327,
      "epoch": 0.34600931910416355,
      "flos": 16650952375680.0,
      "grad_norm": 1.7656342507534244,
      "language_loss": 0.72987127,
      "learning_rate": 3.0407516974467343e-06,
      "loss": 0.75141758,
      "num_input_tokens_seen": 123714360,
      "step": 5755,
      "time_per_iteration": 2.4671053886413574
    },
    {
      "auxiliary_loss_clip": 0.01117104,
      "auxiliary_loss_mlp": 0.01032361,
      "balance_loss_clip": 1.04573309,
      "balance_loss_mlp": 1.0181278,
      "epoch": 0.3460694423568315,
      "flos": 38546190334080.0,
      "grad_norm": 1.7309596881342568,
      "language_loss": 0.72141677,
      "learning_rate": 3.040419101844869e-06,
      "loss": 0.74291146,
      "num_input_tokens_seen": 123739250,
      "step": 5756,
      "time_per_iteration": 4.001556634902954
    },
    {
      "auxiliary_loss_clip": 0.01033947,
      "auxiliary_loss_mlp": 0.010035,
      "balance_loss_clip": 1.01482129,
      "balance_loss_mlp": 1.00140178,
      "epoch": 0.3461295656094995,
      "flos": 72081479704320.0,
      "grad_norm": 0.718414390385618,
      "language_loss": 0.62559074,
      "learning_rate": 3.040086466790207e-06,
      "loss": 0.64596522,
      "num_input_tokens_seen": 123802845,
      "step": 5757,
      "time_per_iteration": 3.1140048503875732
    },
    {
      "auxiliary_loss_clip": 0.01016922,
      "auxiliary_loss_mlp": 0.00850105,
      "balance_loss_clip": 1.01830029,
      "balance_loss_mlp": 1.14248848,
      "epoch": 0.34618968886216744,
      "flos": 65460089571840.0,
      "grad_norm": 0.8611784330914724,
      "language_loss": 0.59212428,
      "learning_rate": 3.039753792295362e-06,
      "loss": 0.6107946,
      "num_input_tokens_seen": 123861805,
      "step": 5758,
      "time_per_iteration": 3.134035587310791
    },
    {
      "auxiliary_loss_clip": 0.01112544,
      "auxiliary_loss_mlp": 0.01039175,
      "balance_loss_clip": 1.04846597,
      "balance_loss_mlp": 1.02516222,
      "epoch": 0.3462498121148354,
      "flos": 23472542960640.0,
      "grad_norm": 1.8005421841541807,
      "language_loss": 0.71731806,
      "learning_rate": 3.0394210783729487e-06,
      "loss": 0.73883533,
      "num_input_tokens_seen": 123881820,
      "step": 5759,
      "time_per_iteration": 2.5734035968780518
    },
    {
      "auxiliary_loss_clip": 0.01077395,
      "auxiliary_loss_mlp": 0.01050056,
      "balance_loss_clip": 1.0420351,
      "balance_loss_mlp": 1.03360546,
      "epoch": 0.3463099353675034,
      "flos": 24170790418560.0,
      "grad_norm": 1.6516304241136652,
      "language_loss": 0.82822049,
      "learning_rate": 3.0390883250355836e-06,
      "loss": 0.84949499,
      "num_input_tokens_seen": 123903700,
      "step": 5760,
      "time_per_iteration": 2.6013007164001465
    },
    {
      "auxiliary_loss_clip": 0.01013407,
      "auxiliary_loss_mlp": 0.01003671,
      "balance_loss_clip": 1.01572824,
      "balance_loss_mlp": 1.00148952,
      "epoch": 0.34637005862017134,
      "flos": 63700609766400.0,
      "grad_norm": 0.8308538304551683,
      "language_loss": 0.56556827,
      "learning_rate": 3.0387555322958865e-06,
      "loss": 0.58573902,
      "num_input_tokens_seen": 123960075,
      "step": 5761,
      "time_per_iteration": 3.245985507965088
    },
    {
      "auxiliary_loss_clip": 0.01113061,
      "auxiliary_loss_mlp": 0.00870884,
      "balance_loss_clip": 1.04109764,
      "balance_loss_mlp": 1.1615802,
      "epoch": 0.3464301818728393,
      "flos": 13145532192000.0,
      "grad_norm": 2.0128922236619107,
      "language_loss": 0.94299078,
      "learning_rate": 3.038422700166474e-06,
      "loss": 0.96283019,
      "num_input_tokens_seen": 123975805,
      "step": 5762,
      "time_per_iteration": 2.4791805744171143
    },
    {
      "auxiliary_loss_clip": 0.01098874,
      "auxiliary_loss_mlp": 0.01041792,
      "balance_loss_clip": 1.04270649,
      "balance_loss_mlp": 1.02691448,
      "epoch": 0.34649030512550727,
      "flos": 29315173299840.0,
      "grad_norm": 1.596381651265668,
      "language_loss": 0.6982922,
      "learning_rate": 3.0380898286599692e-06,
      "loss": 0.71969891,
      "num_input_tokens_seen": 123997530,
      "step": 5763,
      "time_per_iteration": 2.622706651687622
    },
    {
      "auxiliary_loss_clip": 0.01120603,
      "auxiliary_loss_mlp": 0.01044863,
      "balance_loss_clip": 1.0452944,
      "balance_loss_mlp": 1.02791142,
      "epoch": 0.34655042837817523,
      "flos": 23730884553600.0,
      "grad_norm": 1.743155019310353,
      "language_loss": 0.84081578,
      "learning_rate": 3.0377569177889945e-06,
      "loss": 0.86247039,
      "num_input_tokens_seen": 124016375,
      "step": 5764,
      "time_per_iteration": 2.5066659450531006
    },
    {
      "auxiliary_loss_clip": 0.01094207,
      "auxiliary_loss_mlp": 0.01039569,
      "balance_loss_clip": 1.04269552,
      "balance_loss_mlp": 1.02469122,
      "epoch": 0.34661055163084326,
      "flos": 22054215553920.0,
      "grad_norm": 2.8018435424433927,
      "language_loss": 0.675928,
      "learning_rate": 3.0374239675661722e-06,
      "loss": 0.6972658,
      "num_input_tokens_seen": 124033975,
      "step": 5765,
      "time_per_iteration": 2.501084089279175
    },
    {
      "auxiliary_loss_clip": 0.01106865,
      "auxiliary_loss_mlp": 0.0104891,
      "balance_loss_clip": 1.04815078,
      "balance_loss_mlp": 1.03205407,
      "epoch": 0.3466706748835112,
      "flos": 21799213925760.0,
      "grad_norm": 1.8691693206537439,
      "language_loss": 0.77180433,
      "learning_rate": 3.03709097800413e-06,
      "loss": 0.79336214,
      "num_input_tokens_seen": 124051930,
      "step": 5766,
      "time_per_iteration": 2.52838397026062
    },
    {
      "auxiliary_loss_clip": 0.01076385,
      "auxiliary_loss_mlp": 0.01041478,
      "balance_loss_clip": 1.04993629,
      "balance_loss_mlp": 1.02704823,
      "epoch": 0.3467307981361792,
      "flos": 19461680547840.0,
      "grad_norm": 1.5288652104895275,
      "language_loss": 0.73265833,
      "learning_rate": 3.0367579491154943e-06,
      "loss": 0.75383699,
      "num_input_tokens_seen": 124071220,
      "step": 5767,
      "time_per_iteration": 2.5667567253112793
    },
    {
      "auxiliary_loss_clip": 0.01103977,
      "auxiliary_loss_mlp": 0.01043759,
      "balance_loss_clip": 1.04609942,
      "balance_loss_mlp": 1.02764165,
      "epoch": 0.34679092138884715,
      "flos": 24827452905600.0,
      "grad_norm": 2.1591011923880408,
      "language_loss": 0.7798444,
      "learning_rate": 3.036424880912893e-06,
      "loss": 0.80132174,
      "num_input_tokens_seen": 124090140,
      "step": 5768,
      "time_per_iteration": 2.5613629817962646
    },
    {
      "auxiliary_loss_clip": 0.01031981,
      "auxiliary_loss_mlp": 0.01006437,
      "balance_loss_clip": 1.01456833,
      "balance_loss_mlp": 1.00443411,
      "epoch": 0.3468510446415151,
      "flos": 63236070149760.0,
      "grad_norm": 0.7706090516906573,
      "language_loss": 0.57498252,
      "learning_rate": 3.036091773408956e-06,
      "loss": 0.59536672,
      "num_input_tokens_seen": 124152025,
      "step": 5769,
      "time_per_iteration": 3.115565299987793
    },
    {
      "auxiliary_loss_clip": 0.01104151,
      "auxiliary_loss_mlp": 0.01038226,
      "balance_loss_clip": 1.04746962,
      "balance_loss_mlp": 1.02077413,
      "epoch": 0.3469111678941831,
      "flos": 12120713256960.0,
      "grad_norm": 2.668064288145012,
      "language_loss": 0.86062962,
      "learning_rate": 3.0357586266163154e-06,
      "loss": 0.88205338,
      "num_input_tokens_seen": 124165795,
      "step": 5770,
      "time_per_iteration": 2.5307531356811523
    },
    {
      "auxiliary_loss_clip": 0.0102093,
      "auxiliary_loss_mlp": 0.01019774,
      "balance_loss_clip": 1.01796949,
      "balance_loss_mlp": 1.0174731,
      "epoch": 0.34697129114685105,
      "flos": 65934110378880.0,
      "grad_norm": 0.7724028224132267,
      "language_loss": 0.59851867,
      "learning_rate": 3.0354254405476036e-06,
      "loss": 0.61892575,
      "num_input_tokens_seen": 124222925,
      "step": 5771,
      "time_per_iteration": 2.8933849334716797
    },
    {
      "auxiliary_loss_clip": 0.01119235,
      "auxiliary_loss_mlp": 0.01049123,
      "balance_loss_clip": 1.04439998,
      "balance_loss_mlp": 1.0339359,
      "epoch": 0.347031414399519,
      "flos": 34454205054720.0,
      "grad_norm": 1.8465390109194797,
      "language_loss": 0.71433222,
      "learning_rate": 3.0350922152154557e-06,
      "loss": 0.7360158,
      "num_input_tokens_seen": 124240915,
      "step": 5772,
      "time_per_iteration": 2.5881481170654297
    },
    {
      "auxiliary_loss_clip": 0.01101113,
      "auxiliary_loss_mlp": 0.00864464,
      "balance_loss_clip": 1.0436089,
      "balance_loss_mlp": 1.14564204,
      "epoch": 0.347091537652187,
      "flos": 26944135511040.0,
      "grad_norm": 1.7220906261969382,
      "language_loss": 0.76238954,
      "learning_rate": 3.034758950632507e-06,
      "loss": 0.78204536,
      "num_input_tokens_seen": 124262770,
      "step": 5773,
      "time_per_iteration": 2.6080851554870605
    },
    {
      "auxiliary_loss_clip": 0.01120899,
      "auxiliary_loss_mlp": 0.01041316,
      "balance_loss_clip": 1.0450263,
      "balance_loss_mlp": 1.0255568,
      "epoch": 0.34715166090485494,
      "flos": 21142228216320.0,
      "grad_norm": 2.054189394926556,
      "language_loss": 0.69683087,
      "learning_rate": 3.034425646811396e-06,
      "loss": 0.71845305,
      "num_input_tokens_seen": 124280950,
      "step": 5774,
      "time_per_iteration": 2.524069309234619
    },
    {
      "auxiliary_loss_clip": 0.01108204,
      "auxiliary_loss_mlp": 0.00857915,
      "balance_loss_clip": 1.04639173,
      "balance_loss_mlp": 1.13276374,
      "epoch": 0.3472117841575229,
      "flos": 23478001827840.0,
      "grad_norm": 1.577680715417953,
      "language_loss": 0.7659058,
      "learning_rate": 3.0340923037647602e-06,
      "loss": 0.78556705,
      "num_input_tokens_seen": 124299540,
      "step": 5775,
      "time_per_iteration": 2.5543456077575684
    },
    {
      "auxiliary_loss_clip": 0.01109875,
      "auxiliary_loss_mlp": 0.01042385,
      "balance_loss_clip": 1.04392266,
      "balance_loss_mlp": 1.02595842,
      "epoch": 0.34727190741019087,
      "flos": 17492806408320.0,
      "grad_norm": 1.9336255500658956,
      "language_loss": 0.77338493,
      "learning_rate": 3.0337589215052404e-06,
      "loss": 0.79490757,
      "num_input_tokens_seen": 124316285,
      "step": 5776,
      "time_per_iteration": 2.5105631351470947
    },
    {
      "auxiliary_loss_clip": 0.01020786,
      "auxiliary_loss_mlp": 0.01007584,
      "balance_loss_clip": 1.01212549,
      "balance_loss_mlp": 1.00545049,
      "epoch": 0.34733203066285884,
      "flos": 65265491640960.0,
      "grad_norm": 0.84534566984862,
      "language_loss": 0.63361847,
      "learning_rate": 3.033425500045478e-06,
      "loss": 0.65390223,
      "num_input_tokens_seen": 124376650,
      "step": 5777,
      "time_per_iteration": 3.1819326877593994
    },
    {
      "auxiliary_loss_clip": 0.01093596,
      "auxiliary_loss_mlp": 0.0104532,
      "balance_loss_clip": 1.04659152,
      "balance_loss_mlp": 1.0292511,
      "epoch": 0.3473921539155268,
      "flos": 28658726294400.0,
      "grad_norm": 2.01819356058731,
      "language_loss": 0.65127444,
      "learning_rate": 3.033092039398119e-06,
      "loss": 0.67266357,
      "num_input_tokens_seen": 124396475,
      "step": 5778,
      "time_per_iteration": 2.6209211349487305
    },
    {
      "auxiliary_loss_clip": 0.01109957,
      "auxiliary_loss_mlp": 0.0105062,
      "balance_loss_clip": 1.04618764,
      "balance_loss_mlp": 1.0352416,
      "epoch": 0.3474522771681948,
      "flos": 40836895355520.0,
      "grad_norm": 1.707816447829896,
      "language_loss": 0.71077502,
      "learning_rate": 3.0327585395758046e-06,
      "loss": 0.73238081,
      "num_input_tokens_seen": 124416480,
      "step": 5779,
      "time_per_iteration": 2.7074155807495117
    },
    {
      "auxiliary_loss_clip": 0.01136369,
      "auxiliary_loss_mlp": 0.01048829,
      "balance_loss_clip": 1.0480268,
      "balance_loss_mlp": 1.03330827,
      "epoch": 0.3475124004208628,
      "flos": 24608577381120.0,
      "grad_norm": 1.9760099889815457,
      "language_loss": 0.6216538,
      "learning_rate": 3.0324250005911837e-06,
      "loss": 0.64350569,
      "num_input_tokens_seen": 124435950,
      "step": 5780,
      "time_per_iteration": 2.5170717239379883
    },
    {
      "auxiliary_loss_clip": 0.01092941,
      "auxiliary_loss_mlp": 0.01047586,
      "balance_loss_clip": 1.04467273,
      "balance_loss_mlp": 1.032673,
      "epoch": 0.34757252367353075,
      "flos": 22711309004160.0,
      "grad_norm": 1.8609861672635897,
      "language_loss": 0.72238052,
      "learning_rate": 3.0320914224569033e-06,
      "loss": 0.74378574,
      "num_input_tokens_seen": 124455410,
      "step": 5781,
      "time_per_iteration": 2.594478130340576
    },
    {
      "auxiliary_loss_clip": 0.0107505,
      "auxiliary_loss_mlp": 0.01052134,
      "balance_loss_clip": 1.04395282,
      "balance_loss_mlp": 1.03484249,
      "epoch": 0.3476326469261987,
      "flos": 19828184970240.0,
      "grad_norm": 2.033933592376054,
      "language_loss": 0.76910055,
      "learning_rate": 3.031757805185612e-06,
      "loss": 0.79037243,
      "num_input_tokens_seen": 124474870,
      "step": 5782,
      "time_per_iteration": 2.596741199493408
    },
    {
      "auxiliary_loss_clip": 0.01102929,
      "auxiliary_loss_mlp": 0.01035666,
      "balance_loss_clip": 1.04703379,
      "balance_loss_mlp": 1.02042556,
      "epoch": 0.3476927701788667,
      "flos": 19938107566080.0,
      "grad_norm": 1.9138508804933967,
      "language_loss": 0.62517834,
      "learning_rate": 3.0314241487899622e-06,
      "loss": 0.64656436,
      "num_input_tokens_seen": 124494105,
      "step": 5783,
      "time_per_iteration": 2.526453971862793
    },
    {
      "auxiliary_loss_clip": 0.01089424,
      "auxiliary_loss_mlp": 0.01032492,
      "balance_loss_clip": 1.04706788,
      "balance_loss_mlp": 1.01807988,
      "epoch": 0.34775289343153465,
      "flos": 20735108490240.0,
      "grad_norm": 1.7739881976679155,
      "language_loss": 0.88447928,
      "learning_rate": 3.031090453282605e-06,
      "loss": 0.90569842,
      "num_input_tokens_seen": 124512030,
      "step": 5784,
      "time_per_iteration": 2.6694540977478027
    },
    {
      "auxiliary_loss_clip": 0.01084108,
      "auxiliary_loss_mlp": 0.0103542,
      "balance_loss_clip": 1.04946375,
      "balance_loss_mlp": 1.02036428,
      "epoch": 0.3478130166842026,
      "flos": 19354846521600.0,
      "grad_norm": 1.9744060591984811,
      "language_loss": 0.81722569,
      "learning_rate": 3.0307567186761946e-06,
      "loss": 0.83842099,
      "num_input_tokens_seen": 124530980,
      "step": 5785,
      "time_per_iteration": 2.610063076019287
    },
    {
      "auxiliary_loss_clip": 0.01111672,
      "auxiliary_loss_mlp": 0.01043322,
      "balance_loss_clip": 1.05535197,
      "balance_loss_mlp": 1.02803946,
      "epoch": 0.3478731399368706,
      "flos": 22051198811520.0,
      "grad_norm": 2.021910221983646,
      "language_loss": 0.80195284,
      "learning_rate": 3.0304229449833862e-06,
      "loss": 0.82350278,
      "num_input_tokens_seen": 124549330,
      "step": 5786,
      "time_per_iteration": 2.530327796936035
    },
    {
      "auxiliary_loss_clip": 0.01133143,
      "auxiliary_loss_mlp": 0.00799112,
      "balance_loss_clip": 1.04982138,
      "balance_loss_mlp": 1.02110016,
      "epoch": 0.34793326318953854,
      "flos": 18041449720320.0,
      "grad_norm": 1.831576835483207,
      "language_loss": 0.74861562,
      "learning_rate": 3.030089132216836e-06,
      "loss": 0.76793814,
      "num_input_tokens_seen": 124567200,
      "step": 5787,
      "time_per_iteration": 2.4620914459228516
    },
    {
      "auxiliary_loss_clip": 0.01109155,
      "auxiliary_loss_mlp": 0.0079791,
      "balance_loss_clip": 1.05137634,
      "balance_loss_mlp": 1.01979852,
      "epoch": 0.3479933864422065,
      "flos": 29314670509440.0,
      "grad_norm": 1.8561199133068014,
      "language_loss": 0.80901581,
      "learning_rate": 3.029755280389203e-06,
      "loss": 0.8280865,
      "num_input_tokens_seen": 124587025,
      "step": 5788,
      "time_per_iteration": 2.6011006832122803
    },
    {
      "auxiliary_loss_clip": 0.01140229,
      "auxiliary_loss_mlp": 0.01036223,
      "balance_loss_clip": 1.05208719,
      "balance_loss_mlp": 1.01998055,
      "epoch": 0.3480535096948745,
      "flos": 20120713332480.0,
      "grad_norm": 1.767061898422289,
      "language_loss": 0.85762572,
      "learning_rate": 3.029421389513147e-06,
      "loss": 0.87939024,
      "num_input_tokens_seen": 124605860,
      "step": 5789,
      "time_per_iteration": 3.871298313140869
    },
    {
      "auxiliary_loss_clip": 0.01129005,
      "auxiliary_loss_mlp": 0.01056979,
      "balance_loss_clip": 1.05255389,
      "balance_loss_mlp": 1.04114771,
      "epoch": 0.34811363294754244,
      "flos": 18548974938240.0,
      "grad_norm": 1.7842060875407577,
      "language_loss": 0.85311353,
      "learning_rate": 3.029087459601328e-06,
      "loss": 0.87497342,
      "num_input_tokens_seen": 124624270,
      "step": 5790,
      "time_per_iteration": 2.500088691711426
    },
    {
      "auxiliary_loss_clip": 0.01131189,
      "auxiliary_loss_mlp": 0.01044778,
      "balance_loss_clip": 1.05405951,
      "balance_loss_mlp": 1.02855408,
      "epoch": 0.3481737562002104,
      "flos": 26870303105280.0,
      "grad_norm": 1.8647454294760415,
      "language_loss": 0.81413198,
      "learning_rate": 3.0287534906664097e-06,
      "loss": 0.83589172,
      "num_input_tokens_seen": 124644005,
      "step": 5791,
      "time_per_iteration": 2.548558473587036
    },
    {
      "auxiliary_loss_clip": 0.01125297,
      "auxiliary_loss_mlp": 0.01039113,
      "balance_loss_clip": 1.04978228,
      "balance_loss_mlp": 1.02325833,
      "epoch": 0.3482338794528784,
      "flos": 28908664104960.0,
      "grad_norm": 1.9287841113986646,
      "language_loss": 0.77678001,
      "learning_rate": 3.028419482721056e-06,
      "loss": 0.79842407,
      "num_input_tokens_seen": 124663020,
      "step": 5792,
      "time_per_iteration": 5.296192407608032
    },
    {
      "auxiliary_loss_clip": 0.01109868,
      "auxiliary_loss_mlp": 0.0103031,
      "balance_loss_clip": 1.04627991,
      "balance_loss_mlp": 1.01533723,
      "epoch": 0.3482940027055464,
      "flos": 22200767043840.0,
      "grad_norm": 1.8579055052979125,
      "language_loss": 0.81918228,
      "learning_rate": 3.0280854357779325e-06,
      "loss": 0.84058404,
      "num_input_tokens_seen": 124682975,
      "step": 5793,
      "time_per_iteration": 2.55381441116333
    },
    {
      "auxiliary_loss_clip": 0.01125094,
      "auxiliary_loss_mlp": 0.01046942,
      "balance_loss_clip": 1.05285621,
      "balance_loss_mlp": 1.03096747,
      "epoch": 0.34835412595821436,
      "flos": 20302708567680.0,
      "grad_norm": 2.242787635682421,
      "language_loss": 0.76371729,
      "learning_rate": 3.027751349849706e-06,
      "loss": 0.7854377,
      "num_input_tokens_seen": 124701340,
      "step": 5794,
      "time_per_iteration": 2.4923934936523438
    },
    {
      "auxiliary_loss_clip": 0.01123688,
      "auxiliary_loss_mlp": 0.01043784,
      "balance_loss_clip": 1.04934585,
      "balance_loss_mlp": 1.02801216,
      "epoch": 0.3484142492108823,
      "flos": 20449691020800.0,
      "grad_norm": 1.835860575302882,
      "language_loss": 0.57348204,
      "learning_rate": 3.0274172249490456e-06,
      "loss": 0.59515673,
      "num_input_tokens_seen": 124719165,
      "step": 5795,
      "time_per_iteration": 3.8540825843811035
    },
    {
      "auxiliary_loss_clip": 0.01108336,
      "auxiliary_loss_mlp": 0.0103216,
      "balance_loss_clip": 1.05127144,
      "balance_loss_mlp": 1.01734269,
      "epoch": 0.3484743724635503,
      "flos": 24352929308160.0,
      "grad_norm": 1.7925700486875196,
      "language_loss": 0.8248685,
      "learning_rate": 3.0270830610886213e-06,
      "loss": 0.84627342,
      "num_input_tokens_seen": 124738670,
      "step": 5796,
      "time_per_iteration": 2.565178394317627
    },
    {
      "auxiliary_loss_clip": 0.01121693,
      "auxiliary_loss_mlp": 0.01031726,
      "balance_loss_clip": 1.05139971,
      "balance_loss_mlp": 1.01755238,
      "epoch": 0.34853449571621825,
      "flos": 24353001135360.0,
      "grad_norm": 1.6650060667081392,
      "language_loss": 0.83196414,
      "learning_rate": 3.0267488582811033e-06,
      "loss": 0.85349834,
      "num_input_tokens_seen": 124758760,
      "step": 5797,
      "time_per_iteration": 2.561216115951538
    },
    {
      "auxiliary_loss_clip": 0.01132828,
      "auxiliary_loss_mlp": 0.01038517,
      "balance_loss_clip": 1.0504092,
      "balance_loss_mlp": 1.02280545,
      "epoch": 0.3485946189688862,
      "flos": 27267690245760.0,
      "grad_norm": 1.6607571104170704,
      "language_loss": 0.73319948,
      "learning_rate": 3.026414616539167e-06,
      "loss": 0.75491291,
      "num_input_tokens_seen": 124777765,
      "step": 5798,
      "time_per_iteration": 2.5022478103637695
    },
    {
      "auxiliary_loss_clip": 0.01132102,
      "auxiliary_loss_mlp": 0.01042611,
      "balance_loss_clip": 1.04675293,
      "balance_loss_mlp": 1.0271616,
      "epoch": 0.3486547422215542,
      "flos": 20156695781760.0,
      "grad_norm": 2.0449474544609987,
      "language_loss": 0.76093674,
      "learning_rate": 3.026080335875485e-06,
      "loss": 0.78268385,
      "num_input_tokens_seen": 124796775,
      "step": 5799,
      "time_per_iteration": 2.525991201400757
    },
    {
      "auxiliary_loss_clip": 0.01074831,
      "auxiliary_loss_mlp": 0.01033106,
      "balance_loss_clip": 1.05060577,
      "balance_loss_mlp": 1.01847911,
      "epoch": 0.34871486547422215,
      "flos": 20230348619520.0,
      "grad_norm": 1.7795970406062929,
      "language_loss": 0.75494033,
      "learning_rate": 3.025746016302734e-06,
      "loss": 0.77601969,
      "num_input_tokens_seen": 124815825,
      "step": 5800,
      "time_per_iteration": 2.86460280418396
    },
    {
      "auxiliary_loss_clip": 0.0111391,
      "auxiliary_loss_mlp": 0.00798067,
      "balance_loss_clip": 1.0486275,
      "balance_loss_mlp": 1.02049971,
      "epoch": 0.3487749887268901,
      "flos": 44053234882560.0,
      "grad_norm": 1.8740935404394683,
      "language_loss": 0.67088264,
      "learning_rate": 3.025411657833591e-06,
      "loss": 0.69000244,
      "num_input_tokens_seen": 124838420,
      "step": 5801,
      "time_per_iteration": 2.8939156532287598
    },
    {
      "auxiliary_loss_clip": 0.01107945,
      "auxiliary_loss_mlp": 0.01041383,
      "balance_loss_clip": 1.05328321,
      "balance_loss_mlp": 1.02563608,
      "epoch": 0.3488351119795581,
      "flos": 23295144666240.0,
      "grad_norm": 1.7423924733077127,
      "language_loss": 0.7668823,
      "learning_rate": 3.025077260480735e-06,
      "loss": 0.78837562,
      "num_input_tokens_seen": 124857320,
      "step": 5802,
      "time_per_iteration": 2.5549662113189697
    },
    {
      "auxiliary_loss_clip": 0.01060221,
      "auxiliary_loss_mlp": 0.01037,
      "balance_loss_clip": 1.04546428,
      "balance_loss_mlp": 1.02183628,
      "epoch": 0.34889523523222604,
      "flos": 19934839428480.0,
      "grad_norm": 1.5844279791073597,
      "language_loss": 0.78726411,
      "learning_rate": 3.0247428242568474e-06,
      "loss": 0.80823636,
      "num_input_tokens_seen": 124875685,
      "step": 5803,
      "time_per_iteration": 2.769475221633911
    },
    {
      "auxiliary_loss_clip": 0.01108528,
      "auxiliary_loss_mlp": 0.0079545,
      "balance_loss_clip": 1.04465592,
      "balance_loss_mlp": 1.01478028,
      "epoch": 0.348955358484894,
      "flos": 30446179816320.0,
      "grad_norm": 1.6736892078054633,
      "language_loss": 0.67507672,
      "learning_rate": 3.0244083491746085e-06,
      "loss": 0.69411653,
      "num_input_tokens_seen": 124895960,
      "step": 5804,
      "time_per_iteration": 2.696096897125244
    },
    {
      "auxiliary_loss_clip": 0.01106645,
      "auxiliary_loss_mlp": 0.01040207,
      "balance_loss_clip": 1.04962993,
      "balance_loss_mlp": 1.02486515,
      "epoch": 0.349015481737562,
      "flos": 17999972490240.0,
      "grad_norm": 1.7141471422876335,
      "language_loss": 0.7586593,
      "learning_rate": 3.024073835246702e-06,
      "loss": 0.78012788,
      "num_input_tokens_seen": 124914140,
      "step": 5805,
      "time_per_iteration": 2.5204193592071533
    },
    {
      "auxiliary_loss_clip": 0.01091618,
      "auxiliary_loss_mlp": 0.01037574,
      "balance_loss_clip": 1.05297244,
      "balance_loss_mlp": 1.02182603,
      "epoch": 0.34907560499023,
      "flos": 27198490694400.0,
      "grad_norm": 2.54021133736412,
      "language_loss": 0.66622251,
      "learning_rate": 3.023739282485814e-06,
      "loss": 0.68751442,
      "num_input_tokens_seen": 124934180,
      "step": 5806,
      "time_per_iteration": 2.630375862121582
    },
    {
      "auxiliary_loss_clip": 0.01118202,
      "auxiliary_loss_mlp": 0.0104249,
      "balance_loss_clip": 1.04937863,
      "balance_loss_mlp": 1.02717161,
      "epoch": 0.34913572824289796,
      "flos": 30226873328640.0,
      "grad_norm": 1.7744080983246713,
      "language_loss": 0.72320211,
      "learning_rate": 3.023404690904629e-06,
      "loss": 0.74480897,
      "num_input_tokens_seen": 124956060,
      "step": 5807,
      "time_per_iteration": 2.5776526927948
    },
    {
      "auxiliary_loss_clip": 0.01130935,
      "auxiliary_loss_mlp": 0.0103572,
      "balance_loss_clip": 1.04565489,
      "balance_loss_mlp": 1.02048469,
      "epoch": 0.3491958514955659,
      "flos": 29971907614080.0,
      "grad_norm": 1.929717596915313,
      "language_loss": 0.74174094,
      "learning_rate": 3.0230700605158364e-06,
      "loss": 0.76340759,
      "num_input_tokens_seen": 124976070,
      "step": 5808,
      "time_per_iteration": 2.584321975708008
    },
    {
      "auxiliary_loss_clip": 0.01130106,
      "auxiliary_loss_mlp": 0.01047369,
      "balance_loss_clip": 1.04965067,
      "balance_loss_mlp": 1.0324564,
      "epoch": 0.3492559747482339,
      "flos": 22783273902720.0,
      "grad_norm": 1.8580407593026143,
      "language_loss": 0.84422874,
      "learning_rate": 3.0227353913321238e-06,
      "loss": 0.86600345,
      "num_input_tokens_seen": 124996995,
      "step": 5809,
      "time_per_iteration": 2.479323387145996
    },
    {
      "auxiliary_loss_clip": 0.01102738,
      "auxiliary_loss_mlp": 0.01032966,
      "balance_loss_clip": 1.04489481,
      "balance_loss_mlp": 1.01965022,
      "epoch": 0.34931609800090185,
      "flos": 26068022881920.0,
      "grad_norm": 3.678988773609896,
      "language_loss": 0.8056525,
      "learning_rate": 3.0224006833661835e-06,
      "loss": 0.82700956,
      "num_input_tokens_seen": 125015600,
      "step": 5810,
      "time_per_iteration": 2.5717360973358154
    },
    {
      "auxiliary_loss_clip": 0.01129266,
      "auxiliary_loss_mlp": 0.01041557,
      "balance_loss_clip": 1.04611552,
      "balance_loss_mlp": 1.02780628,
      "epoch": 0.3493762212535698,
      "flos": 29242023252480.0,
      "grad_norm": 1.9983467185828794,
      "language_loss": 0.75277495,
      "learning_rate": 3.0220659366307057e-06,
      "loss": 0.7744832,
      "num_input_tokens_seen": 125035290,
      "step": 5811,
      "time_per_iteration": 2.5317916870117188
    },
    {
      "auxiliary_loss_clip": 0.01111557,
      "auxiliary_loss_mlp": 0.01043013,
      "balance_loss_clip": 1.04503679,
      "balance_loss_mlp": 1.02828443,
      "epoch": 0.3494363445062378,
      "flos": 27126058919040.0,
      "grad_norm": 1.4352897368207767,
      "language_loss": 0.80294013,
      "learning_rate": 3.021731151138386e-06,
      "loss": 0.82448572,
      "num_input_tokens_seen": 125057130,
      "step": 5812,
      "time_per_iteration": 2.593714475631714
    },
    {
      "auxiliary_loss_clip": 0.01071222,
      "auxiliary_loss_mlp": 0.01034097,
      "balance_loss_clip": 1.04467392,
      "balance_loss_mlp": 1.01834393,
      "epoch": 0.34949646775890575,
      "flos": 12276207233280.0,
      "grad_norm": 2.7080529463812573,
      "language_loss": 0.69415176,
      "learning_rate": 3.021396326901918e-06,
      "loss": 0.71520495,
      "num_input_tokens_seen": 125073720,
      "step": 5813,
      "time_per_iteration": 2.5898845195770264
    },
    {
      "auxiliary_loss_clip": 0.01099666,
      "auxiliary_loss_mlp": 0.00792923,
      "balance_loss_clip": 1.04163742,
      "balance_loss_mlp": 1.01363087,
      "epoch": 0.3495565910115737,
      "flos": 17165516659200.0,
      "grad_norm": 1.9451867545086052,
      "language_loss": 0.77130079,
      "learning_rate": 3.0210614639339998e-06,
      "loss": 0.7902267,
      "num_input_tokens_seen": 125090635,
      "step": 5814,
      "time_per_iteration": 2.537668228149414
    },
    {
      "auxiliary_loss_clip": 0.01118534,
      "auxiliary_loss_mlp": 0.00792767,
      "balance_loss_clip": 1.04918635,
      "balance_loss_mlp": 1.01377869,
      "epoch": 0.3496167142642417,
      "flos": 26465661417600.0,
      "grad_norm": 2.2087955978273204,
      "language_loss": 0.84412062,
      "learning_rate": 3.020726562247328e-06,
      "loss": 0.86323369,
      "num_input_tokens_seen": 125110070,
      "step": 5815,
      "time_per_iteration": 2.576323986053467
    },
    {
      "auxiliary_loss_clip": 0.01113641,
      "auxiliary_loss_mlp": 0.01032989,
      "balance_loss_clip": 1.04621542,
      "balance_loss_mlp": 1.01929188,
      "epoch": 0.34967683751690964,
      "flos": 17414843938560.0,
      "grad_norm": 1.9922287939743144,
      "language_loss": 0.77405334,
      "learning_rate": 3.0203916218546024e-06,
      "loss": 0.79551959,
      "num_input_tokens_seen": 125125730,
      "step": 5816,
      "time_per_iteration": 2.480428695678711
    },
    {
      "auxiliary_loss_clip": 0.01116871,
      "auxiliary_loss_mlp": 0.01040529,
      "balance_loss_clip": 1.04756451,
      "balance_loss_mlp": 1.02593756,
      "epoch": 0.3497369607695776,
      "flos": 22600021691520.0,
      "grad_norm": 1.9437217181867819,
      "language_loss": 0.58660495,
      "learning_rate": 3.0200566427685246e-06,
      "loss": 0.60817897,
      "num_input_tokens_seen": 125146195,
      "step": 5817,
      "time_per_iteration": 2.5158848762512207
    },
    {
      "auxiliary_loss_clip": 0.01045738,
      "auxiliary_loss_mlp": 0.01005046,
      "balance_loss_clip": 1.01679444,
      "balance_loss_mlp": 1.00342524,
      "epoch": 0.34979708402224563,
      "flos": 68529374818560.0,
      "grad_norm": 0.8716650635917494,
      "language_loss": 0.59910929,
      "learning_rate": 3.0197216250017975e-06,
      "loss": 0.61961722,
      "num_input_tokens_seen": 125207790,
      "step": 5818,
      "time_per_iteration": 3.145456314086914
    },
    {
      "auxiliary_loss_clip": 0.01098015,
      "auxiliary_loss_mlp": 0.01040814,
      "balance_loss_clip": 1.05323076,
      "balance_loss_mlp": 1.02569246,
      "epoch": 0.3498572072749136,
      "flos": 18989634988800.0,
      "grad_norm": 1.6392515976083797,
      "language_loss": 0.83438218,
      "learning_rate": 3.019386568567123e-06,
      "loss": 0.85577047,
      "num_input_tokens_seen": 125226220,
      "step": 5819,
      "time_per_iteration": 2.5575475692749023
    },
    {
      "auxiliary_loss_clip": 0.0110541,
      "auxiliary_loss_mlp": 0.01031489,
      "balance_loss_clip": 1.04469657,
      "balance_loss_mlp": 1.01749945,
      "epoch": 0.34991733052758156,
      "flos": 27818883423360.0,
      "grad_norm": 1.6533854029478456,
      "language_loss": 0.70815057,
      "learning_rate": 3.0190514734772083e-06,
      "loss": 0.72951955,
      "num_input_tokens_seen": 125247485,
      "step": 5820,
      "time_per_iteration": 2.6225826740264893
    },
    {
      "auxiliary_loss_clip": 0.01118616,
      "auxiliary_loss_mlp": 0.01040835,
      "balance_loss_clip": 1.0442735,
      "balance_loss_mlp": 1.02675021,
      "epoch": 0.3499774537802495,
      "flos": 33584197737600.0,
      "grad_norm": 1.6580439431033682,
      "language_loss": 0.7043128,
      "learning_rate": 3.018716339744759e-06,
      "loss": 0.72590733,
      "num_input_tokens_seen": 125268625,
      "step": 5821,
      "time_per_iteration": 2.6092538833618164
    },
    {
      "auxiliary_loss_clip": 0.01124586,
      "auxiliary_loss_mlp": 0.01042473,
      "balance_loss_clip": 1.04758298,
      "balance_loss_mlp": 1.02673149,
      "epoch": 0.3500375770329175,
      "flos": 23476744851840.0,
      "grad_norm": 2.069446735259659,
      "language_loss": 0.73758411,
      "learning_rate": 3.0183811673824842e-06,
      "loss": 0.75925469,
      "num_input_tokens_seen": 125287530,
      "step": 5822,
      "time_per_iteration": 2.504305839538574
    },
    {
      "auxiliary_loss_clip": 0.01109761,
      "auxiliary_loss_mlp": 0.01034363,
      "balance_loss_clip": 1.04915643,
      "balance_loss_mlp": 1.01863956,
      "epoch": 0.35009770028558546,
      "flos": 19026048401280.0,
      "grad_norm": 1.5812623175025033,
      "language_loss": 0.78563869,
      "learning_rate": 3.018045956403094e-06,
      "loss": 0.80707991,
      "num_input_tokens_seen": 125307020,
      "step": 5823,
      "time_per_iteration": 2.5024831295013428
    },
    {
      "auxiliary_loss_clip": 0.01033916,
      "auxiliary_loss_mlp": 0.01001325,
      "balance_loss_clip": 1.01549864,
      "balance_loss_mlp": 0.99957287,
      "epoch": 0.3501578235382534,
      "flos": 68351868783360.0,
      "grad_norm": 0.7148670750866176,
      "language_loss": 0.59281546,
      "learning_rate": 3.017710706819298e-06,
      "loss": 0.61316788,
      "num_input_tokens_seen": 125370445,
      "step": 5824,
      "time_per_iteration": 3.11738657951355
    },
    {
      "auxiliary_loss_clip": 0.0111012,
      "auxiliary_loss_mlp": 0.01036676,
      "balance_loss_clip": 1.04720461,
      "balance_loss_mlp": 1.02067828,
      "epoch": 0.3502179467909214,
      "flos": 21250893836160.0,
      "grad_norm": 1.7707764067077065,
      "language_loss": 0.84662026,
      "learning_rate": 3.017375418643811e-06,
      "loss": 0.86808825,
      "num_input_tokens_seen": 125388900,
      "step": 5825,
      "time_per_iteration": 2.5378222465515137
    },
    {
      "auxiliary_loss_clip": 0.0112165,
      "auxiliary_loss_mlp": 0.00795293,
      "balance_loss_clip": 1.04814231,
      "balance_loss_mlp": 1.01630867,
      "epoch": 0.35027807004358935,
      "flos": 11942955826560.0,
      "grad_norm": 2.3081970940901715,
      "language_loss": 0.83331585,
      "learning_rate": 3.0170400918893464e-06,
      "loss": 0.85248536,
      "num_input_tokens_seen": 125402675,
      "step": 5826,
      "time_per_iteration": 2.4486277103424072
    },
    {
      "auxiliary_loss_clip": 0.01111345,
      "auxiliary_loss_mlp": 0.01045587,
      "balance_loss_clip": 1.04954314,
      "balance_loss_mlp": 1.03068018,
      "epoch": 0.3503381932962573,
      "flos": 21470918595840.0,
      "grad_norm": 1.7038488778909955,
      "language_loss": 0.81126523,
      "learning_rate": 3.0167047265686186e-06,
      "loss": 0.8328346,
      "num_input_tokens_seen": 125421360,
      "step": 5827,
      "time_per_iteration": 2.537687301635742
    },
    {
      "auxiliary_loss_clip": 0.01080721,
      "auxiliary_loss_mlp": 0.01043377,
      "balance_loss_clip": 1.04439294,
      "balance_loss_mlp": 1.02807057,
      "epoch": 0.3503983165489253,
      "flos": 21251109317760.0,
      "grad_norm": 2.7796788375227526,
      "language_loss": 0.71324801,
      "learning_rate": 3.0163693226943467e-06,
      "loss": 0.73448896,
      "num_input_tokens_seen": 125440000,
      "step": 5828,
      "time_per_iteration": 4.172318458557129
    },
    {
      "auxiliary_loss_clip": 0.01125238,
      "auxiliary_loss_mlp": 0.01046498,
      "balance_loss_clip": 1.04939294,
      "balance_loss_mlp": 1.02923644,
      "epoch": 0.35045843980159325,
      "flos": 27815723026560.0,
      "grad_norm": 1.7021763742408673,
      "language_loss": 0.79749548,
      "learning_rate": 3.016033880279248e-06,
      "loss": 0.81921285,
      "num_input_tokens_seen": 125460390,
      "step": 5829,
      "time_per_iteration": 2.573141098022461
    },
    {
      "auxiliary_loss_clip": 0.01095526,
      "auxiliary_loss_mlp": 0.01051669,
      "balance_loss_clip": 1.0508182,
      "balance_loss_mlp": 1.03481269,
      "epoch": 0.3505185630542612,
      "flos": 25921148169600.0,
      "grad_norm": 1.8031161144613588,
      "language_loss": 0.72547805,
      "learning_rate": 3.0156983993360417e-06,
      "loss": 0.74695003,
      "num_input_tokens_seen": 125478410,
      "step": 5830,
      "time_per_iteration": 3.996628999710083
    },
    {
      "auxiliary_loss_clip": 0.01094764,
      "auxiliary_loss_mlp": 0.01036608,
      "balance_loss_clip": 1.04997563,
      "balance_loss_mlp": 1.020473,
      "epoch": 0.35057868630692923,
      "flos": 20521763660160.0,
      "grad_norm": 2.0375878174280686,
      "language_loss": 0.88555193,
      "learning_rate": 3.0153628798774513e-06,
      "loss": 0.90686566,
      "num_input_tokens_seen": 125495975,
      "step": 5831,
      "time_per_iteration": 3.968332529067993
    },
    {
      "auxiliary_loss_clip": 0.0107764,
      "auxiliary_loss_mlp": 0.01048227,
      "balance_loss_clip": 1.0463388,
      "balance_loss_mlp": 1.0322051,
      "epoch": 0.3506388095595972,
      "flos": 20448649526400.0,
      "grad_norm": 2.0451708222432665,
      "language_loss": 0.78637874,
      "learning_rate": 3.0150273219161985e-06,
      "loss": 0.80763745,
      "num_input_tokens_seen": 125515035,
      "step": 5832,
      "time_per_iteration": 2.606616973876953
    },
    {
      "auxiliary_loss_clip": 0.01097098,
      "auxiliary_loss_mlp": 0.01041954,
      "balance_loss_clip": 1.04672647,
      "balance_loss_mlp": 1.02420378,
      "epoch": 0.35069893281226516,
      "flos": 23109665811840.0,
      "grad_norm": 1.699179386386844,
      "language_loss": 0.70954883,
      "learning_rate": 3.014691725465008e-06,
      "loss": 0.73093939,
      "num_input_tokens_seen": 125535555,
      "step": 5833,
      "time_per_iteration": 2.5976011753082275
    },
    {
      "auxiliary_loss_clip": 0.0112194,
      "auxiliary_loss_mlp": 0.01035166,
      "balance_loss_clip": 1.04855418,
      "balance_loss_mlp": 1.02032459,
      "epoch": 0.35075905606493313,
      "flos": 27271999877760.0,
      "grad_norm": 1.323501883759618,
      "language_loss": 0.80836153,
      "learning_rate": 3.014356090536606e-06,
      "loss": 0.82993257,
      "num_input_tokens_seen": 125558195,
      "step": 5834,
      "time_per_iteration": 3.9552626609802246
    },
    {
      "auxiliary_loss_clip": 0.01084504,
      "auxiliary_loss_mlp": 0.01048013,
      "balance_loss_clip": 1.05150414,
      "balance_loss_mlp": 1.03197932,
      "epoch": 0.3508191793176011,
      "flos": 19128608709120.0,
      "grad_norm": 1.9878917957925786,
      "language_loss": 0.84337652,
      "learning_rate": 3.0140204171437183e-06,
      "loss": 0.86470163,
      "num_input_tokens_seen": 125575375,
      "step": 5835,
      "time_per_iteration": 2.6010093688964844
    },
    {
      "auxiliary_loss_clip": 0.01080928,
      "auxiliary_loss_mlp": 0.01044722,
      "balance_loss_clip": 1.0477922,
      "balance_loss_mlp": 1.02954698,
      "epoch": 0.35087930257026906,
      "flos": 25557588662400.0,
      "grad_norm": 1.525277146120332,
      "language_loss": 0.76577449,
      "learning_rate": 3.0136847052990754e-06,
      "loss": 0.78703099,
      "num_input_tokens_seen": 125596745,
      "step": 5836,
      "time_per_iteration": 2.6478371620178223
    },
    {
      "auxiliary_loss_clip": 0.01096868,
      "auxiliary_loss_mlp": 0.0104305,
      "balance_loss_clip": 1.05297875,
      "balance_loss_mlp": 1.02635467,
      "epoch": 0.350939425822937,
      "flos": 18004246208640.0,
      "grad_norm": 1.8907688249638006,
      "language_loss": 0.7710861,
      "learning_rate": 3.0133489550154074e-06,
      "loss": 0.79248524,
      "num_input_tokens_seen": 125613980,
      "step": 5837,
      "time_per_iteration": 2.5695292949676514
    },
    {
      "auxiliary_loss_clip": 0.01124088,
      "auxiliary_loss_mlp": 0.01043963,
      "balance_loss_clip": 1.05110526,
      "balance_loss_mlp": 1.02883518,
      "epoch": 0.350999549075605,
      "flos": 22273198819200.0,
      "grad_norm": 1.6935788391722322,
      "language_loss": 0.68174243,
      "learning_rate": 3.0130131663054442e-06,
      "loss": 0.70342302,
      "num_input_tokens_seen": 125632100,
      "step": 5838,
      "time_per_iteration": 2.5336806774139404
    },
    {
      "auxiliary_loss_clip": 0.01134765,
      "auxiliary_loss_mlp": 0.01039177,
      "balance_loss_clip": 1.04972744,
      "balance_loss_mlp": 1.02302456,
      "epoch": 0.35105967232827295,
      "flos": 14392279307520.0,
      "grad_norm": 1.771987844612919,
      "language_loss": 0.82979786,
      "learning_rate": 3.0126773391819215e-06,
      "loss": 0.85153723,
      "num_input_tokens_seen": 125649190,
      "step": 5839,
      "time_per_iteration": 2.483438014984131
    },
    {
      "auxiliary_loss_clip": 0.01129713,
      "auxiliary_loss_mlp": 0.01039239,
      "balance_loss_clip": 1.04855597,
      "balance_loss_mlp": 1.02336025,
      "epoch": 0.3511197955809409,
      "flos": 25082346792960.0,
      "grad_norm": 1.6762890116167581,
      "language_loss": 0.58736056,
      "learning_rate": 3.012341473657572e-06,
      "loss": 0.60905004,
      "num_input_tokens_seen": 125668680,
      "step": 5840,
      "time_per_iteration": 2.543370008468628
    },
    {
      "auxiliary_loss_clip": 0.01093623,
      "auxiliary_loss_mlp": 0.01040527,
      "balance_loss_clip": 1.04599118,
      "balance_loss_mlp": 1.02455294,
      "epoch": 0.3511799188336089,
      "flos": 25884160139520.0,
      "grad_norm": 2.217167887623517,
      "language_loss": 0.87129503,
      "learning_rate": 3.0120055697451322e-06,
      "loss": 0.89263654,
      "num_input_tokens_seen": 125686935,
      "step": 5841,
      "time_per_iteration": 2.575432300567627
    },
    {
      "auxiliary_loss_clip": 0.01123106,
      "auxiliary_loss_mlp": 0.01043995,
      "balance_loss_clip": 1.05268312,
      "balance_loss_mlp": 1.02630413,
      "epoch": 0.35124004208627685,
      "flos": 20083725302400.0,
      "grad_norm": 1.5778138073697603,
      "language_loss": 0.74827182,
      "learning_rate": 3.0116696274573406e-06,
      "loss": 0.76994282,
      "num_input_tokens_seen": 125707180,
      "step": 5842,
      "time_per_iteration": 2.594163656234741
    },
    {
      "auxiliary_loss_clip": 0.01129127,
      "auxiliary_loss_mlp": 0.01042584,
      "balance_loss_clip": 1.0499351,
      "balance_loss_mlp": 1.02713406,
      "epoch": 0.3513001653389448,
      "flos": 17783431349760.0,
      "grad_norm": 2.137753022959898,
      "language_loss": 0.6872648,
      "learning_rate": 3.0113336468069346e-06,
      "loss": 0.70898193,
      "num_input_tokens_seen": 125722780,
      "step": 5843,
      "time_per_iteration": 2.4681665897369385
    },
    {
      "auxiliary_loss_clip": 0.01136087,
      "auxiliary_loss_mlp": 0.01041086,
      "balance_loss_clip": 1.05203342,
      "balance_loss_mlp": 1.02523136,
      "epoch": 0.3513602885916128,
      "flos": 29387138198400.0,
      "grad_norm": 1.8466012567572074,
      "language_loss": 0.65399778,
      "learning_rate": 3.010997627806655e-06,
      "loss": 0.67576945,
      "num_input_tokens_seen": 125742110,
      "step": 5844,
      "time_per_iteration": 2.5207743644714355
    },
    {
      "auxiliary_loss_clip": 0.01122659,
      "auxiliary_loss_mlp": 0.01044471,
      "balance_loss_clip": 1.05318165,
      "balance_loss_mlp": 1.02801991,
      "epoch": 0.3514204118442808,
      "flos": 16179876483840.0,
      "grad_norm": 2.449804850926252,
      "language_loss": 0.74935693,
      "learning_rate": 3.010661570469245e-06,
      "loss": 0.77102822,
      "num_input_tokens_seen": 125759980,
      "step": 5845,
      "time_per_iteration": 2.477585554122925
    },
    {
      "auxiliary_loss_clip": 0.01123938,
      "auxiliary_loss_mlp": 0.01040458,
      "balance_loss_clip": 1.05140877,
      "balance_loss_mlp": 1.02488387,
      "epoch": 0.35148053509694877,
      "flos": 23834665923840.0,
      "grad_norm": 2.5118975235667684,
      "language_loss": 0.73316616,
      "learning_rate": 3.0103254748074465e-06,
      "loss": 0.75481004,
      "num_input_tokens_seen": 125772660,
      "step": 5846,
      "time_per_iteration": 2.499051570892334
    },
    {
      "auxiliary_loss_clip": 0.0110007,
      "auxiliary_loss_mlp": 0.01039704,
      "balance_loss_clip": 1.04864621,
      "balance_loss_mlp": 1.02402782,
      "epoch": 0.35154065834961673,
      "flos": 20991295267200.0,
      "grad_norm": 1.5676039781792857,
      "language_loss": 0.75464141,
      "learning_rate": 3.0099893408340046e-06,
      "loss": 0.77603912,
      "num_input_tokens_seen": 125791935,
      "step": 5847,
      "time_per_iteration": 2.555535316467285
    },
    {
      "auxiliary_loss_clip": 0.01109209,
      "auxiliary_loss_mlp": 0.01037619,
      "balance_loss_clip": 1.04650044,
      "balance_loss_mlp": 1.02246165,
      "epoch": 0.3516007816022847,
      "flos": 33255471444480.0,
      "grad_norm": 1.9264019898579687,
      "language_loss": 0.72563243,
      "learning_rate": 3.009653168561666e-06,
      "loss": 0.74710065,
      "num_input_tokens_seen": 125813455,
      "step": 5848,
      "time_per_iteration": 2.6338155269622803
    },
    {
      "auxiliary_loss_clip": 0.01117002,
      "auxiliary_loss_mlp": 0.01045767,
      "balance_loss_clip": 1.04781115,
      "balance_loss_mlp": 1.02925694,
      "epoch": 0.35166090485495266,
      "flos": 11726953390080.0,
      "grad_norm": 6.875120398005647,
      "language_loss": 0.89234543,
      "learning_rate": 3.009316958003178e-06,
      "loss": 0.91397309,
      "num_input_tokens_seen": 125827660,
      "step": 5849,
      "time_per_iteration": 2.475409984588623
    },
    {
      "auxiliary_loss_clip": 0.01113607,
      "auxiliary_loss_mlp": 0.01036429,
      "balance_loss_clip": 1.05029523,
      "balance_loss_mlp": 1.02103949,
      "epoch": 0.3517210281076206,
      "flos": 22638446265600.0,
      "grad_norm": 1.6967685486415967,
      "language_loss": 0.74683148,
      "learning_rate": 3.0089807091712897e-06,
      "loss": 0.76833189,
      "num_input_tokens_seen": 125846655,
      "step": 5850,
      "time_per_iteration": 2.5845816135406494
    },
    {
      "auxiliary_loss_clip": 0.01123184,
      "auxiliary_loss_mlp": 0.01036262,
      "balance_loss_clip": 1.05038047,
      "balance_loss_mlp": 1.0206697,
      "epoch": 0.3517811513602886,
      "flos": 21322750993920.0,
      "grad_norm": 1.3731079826195953,
      "language_loss": 0.75315857,
      "learning_rate": 3.0086444220787515e-06,
      "loss": 0.77475303,
      "num_input_tokens_seen": 125866290,
      "step": 5851,
      "time_per_iteration": 2.5103023052215576
    },
    {
      "auxiliary_loss_clip": 0.01109473,
      "auxiliary_loss_mlp": 0.01041263,
      "balance_loss_clip": 1.05027366,
      "balance_loss_mlp": 1.02397728,
      "epoch": 0.35184127461295656,
      "flos": 21032880238080.0,
      "grad_norm": 1.866952071108202,
      "language_loss": 0.87552691,
      "learning_rate": 3.0083080967383165e-06,
      "loss": 0.89703429,
      "num_input_tokens_seen": 125884620,
      "step": 5852,
      "time_per_iteration": 2.6041038036346436
    },
    {
      "auxiliary_loss_clip": 0.01131535,
      "auxiliary_loss_mlp": 0.01032717,
      "balance_loss_clip": 1.04928529,
      "balance_loss_mlp": 1.01777458,
      "epoch": 0.3519013978656245,
      "flos": 22455265881600.0,
      "grad_norm": 2.0272198145544684,
      "language_loss": 0.6730237,
      "learning_rate": 3.007971733162737e-06,
      "loss": 0.69466615,
      "num_input_tokens_seen": 125902430,
      "step": 5853,
      "time_per_iteration": 2.4967010021209717
    },
    {
      "auxiliary_loss_clip": 0.01109041,
      "auxiliary_loss_mlp": 0.01036568,
      "balance_loss_clip": 1.04612279,
      "balance_loss_mlp": 1.02073658,
      "epoch": 0.3519615211182925,
      "flos": 13115295918720.0,
      "grad_norm": 1.7281458650852382,
      "language_loss": 0.8122561,
      "learning_rate": 3.0076353313647686e-06,
      "loss": 0.83371216,
      "num_input_tokens_seen": 125920570,
      "step": 5854,
      "time_per_iteration": 2.5506367683410645
    },
    {
      "auxiliary_loss_clip": 0.0111036,
      "auxiliary_loss_mlp": 0.01033287,
      "balance_loss_clip": 1.04889774,
      "balance_loss_mlp": 1.01917291,
      "epoch": 0.35202164437096045,
      "flos": 19135144984320.0,
      "grad_norm": 1.5272123834975948,
      "language_loss": 0.73169172,
      "learning_rate": 3.0072988913571666e-06,
      "loss": 0.75312817,
      "num_input_tokens_seen": 125939800,
      "step": 5855,
      "time_per_iteration": 2.5319435596466064
    },
    {
      "auxiliary_loss_clip": 0.01128804,
      "auxiliary_loss_mlp": 0.01035421,
      "balance_loss_clip": 1.04746926,
      "balance_loss_mlp": 1.02150965,
      "epoch": 0.3520817676236284,
      "flos": 26542187343360.0,
      "grad_norm": 1.8741235134212486,
      "language_loss": 0.70963943,
      "learning_rate": 3.006962413152691e-06,
      "loss": 0.7312817,
      "num_input_tokens_seen": 125958720,
      "step": 5856,
      "time_per_iteration": 2.517080068588257
    },
    {
      "auxiliary_loss_clip": 0.01120545,
      "auxiliary_loss_mlp": 0.01044875,
      "balance_loss_clip": 1.05249798,
      "balance_loss_mlp": 1.02811408,
      "epoch": 0.3521418908762964,
      "flos": 44893472803200.0,
      "grad_norm": 1.6767013858738935,
      "language_loss": 0.6127106,
      "learning_rate": 3.0066258967640987e-06,
      "loss": 0.63436478,
      "num_input_tokens_seen": 125984310,
      "step": 5857,
      "time_per_iteration": 2.7190990447998047
    },
    {
      "auxiliary_loss_clip": 0.0112255,
      "auxiliary_loss_mlp": 0.01039791,
      "balance_loss_clip": 1.0490396,
      "balance_loss_mlp": 1.02407897,
      "epoch": 0.3522020141289644,
      "flos": 20187398931840.0,
      "grad_norm": 1.724283252075808,
      "language_loss": 0.73867416,
      "learning_rate": 3.006289342204152e-06,
      "loss": 0.76029754,
      "num_input_tokens_seen": 126002410,
      "step": 5858,
      "time_per_iteration": 2.5223817825317383
    },
    {
      "auxiliary_loss_clip": 0.01133931,
      "auxiliary_loss_mlp": 0.0103748,
      "balance_loss_clip": 1.04796326,
      "balance_loss_mlp": 1.02264476,
      "epoch": 0.35226213738163237,
      "flos": 27563917708800.0,
      "grad_norm": 1.794022223030072,
      "language_loss": 0.76146573,
      "learning_rate": 3.0059527494856126e-06,
      "loss": 0.78317982,
      "num_input_tokens_seen": 126022490,
      "step": 5859,
      "time_per_iteration": 2.5265674591064453
    },
    {
      "auxiliary_loss_clip": 0.01116168,
      "auxiliary_loss_mlp": 0.01043438,
      "balance_loss_clip": 1.05037916,
      "balance_loss_mlp": 1.02662909,
      "epoch": 0.35232226063430033,
      "flos": 22966310632320.0,
      "grad_norm": 2.178415419128804,
      "language_loss": 0.7196883,
      "learning_rate": 3.0056161186212435e-06,
      "loss": 0.74128425,
      "num_input_tokens_seen": 126042895,
      "step": 5860,
      "time_per_iteration": 2.5323452949523926
    },
    {
      "auxiliary_loss_clip": 0.0110542,
      "auxiliary_loss_mlp": 0.01039849,
      "balance_loss_clip": 1.05077481,
      "balance_loss_mlp": 1.02312422,
      "epoch": 0.3523823838869683,
      "flos": 19168290259200.0,
      "grad_norm": 6.363159385561328,
      "language_loss": 0.663679,
      "learning_rate": 3.005279449623811e-06,
      "loss": 0.68513173,
      "num_input_tokens_seen": 126060130,
      "step": 5861,
      "time_per_iteration": 2.520472288131714
    },
    {
      "auxiliary_loss_clip": 0.01110679,
      "auxiliary_loss_mlp": 0.01033544,
      "balance_loss_clip": 1.04919958,
      "balance_loss_mlp": 1.01927423,
      "epoch": 0.35244250713963626,
      "flos": 17930988420480.0,
      "grad_norm": 2.2330287333610985,
      "language_loss": 0.66732037,
      "learning_rate": 3.0049427425060815e-06,
      "loss": 0.68876261,
      "num_input_tokens_seen": 126077850,
      "step": 5862,
      "time_per_iteration": 2.5048139095306396
    },
    {
      "auxiliary_loss_clip": 0.01109281,
      "auxiliary_loss_mlp": 0.01038882,
      "balance_loss_clip": 1.05169249,
      "balance_loss_mlp": 1.02257431,
      "epoch": 0.35250263039230423,
      "flos": 21432529935360.0,
      "grad_norm": 1.8897853429901486,
      "language_loss": 0.76899844,
      "learning_rate": 3.0046059972808215e-06,
      "loss": 0.79048002,
      "num_input_tokens_seen": 126095985,
      "step": 5863,
      "time_per_iteration": 2.533298969268799
    },
    {
      "auxiliary_loss_clip": 0.01118531,
      "auxiliary_loss_mlp": 0.01038936,
      "balance_loss_clip": 1.05049229,
      "balance_loss_mlp": 1.02432644,
      "epoch": 0.3525627536449722,
      "flos": 27416863428480.0,
      "grad_norm": 1.8160782879304753,
      "language_loss": 0.75213611,
      "learning_rate": 3.0042692139608024e-06,
      "loss": 0.77371085,
      "num_input_tokens_seen": 126116070,
      "step": 5864,
      "time_per_iteration": 2.5390617847442627
    },
    {
      "auxiliary_loss_clip": 0.01120196,
      "auxiliary_loss_mlp": 0.01046467,
      "balance_loss_clip": 1.04623139,
      "balance_loss_mlp": 1.03156555,
      "epoch": 0.35262287689764016,
      "flos": 24789818430720.0,
      "grad_norm": 2.9143774887190306,
      "language_loss": 0.79060072,
      "learning_rate": 3.003932392558793e-06,
      "loss": 0.8122673,
      "num_input_tokens_seen": 126135205,
      "step": 5865,
      "time_per_iteration": 2.531221866607666
    },
    {
      "auxiliary_loss_clip": 0.01126557,
      "auxiliary_loss_mlp": 0.01040799,
      "balance_loss_clip": 1.05069423,
      "balance_loss_mlp": 1.02510512,
      "epoch": 0.3526830001503081,
      "flos": 17821604528640.0,
      "grad_norm": 2.0126835174724964,
      "language_loss": 0.8122918,
      "learning_rate": 3.0035955330875677e-06,
      "loss": 0.83396536,
      "num_input_tokens_seen": 126151895,
      "step": 5866,
      "time_per_iteration": 3.898078441619873
    },
    {
      "auxiliary_loss_clip": 0.01090459,
      "auxiliary_loss_mlp": 0.01036697,
      "balance_loss_clip": 1.05034804,
      "balance_loss_mlp": 1.01985228,
      "epoch": 0.3527431234029761,
      "flos": 18078114528000.0,
      "grad_norm": 2.0978919941418677,
      "language_loss": 0.845595,
      "learning_rate": 3.0032586355598986e-06,
      "loss": 0.86686653,
      "num_input_tokens_seen": 126168515,
      "step": 5867,
      "time_per_iteration": 2.575634717941284
    },
    {
      "auxiliary_loss_clip": 0.01136033,
      "auxiliary_loss_mlp": 0.01048109,
      "balance_loss_clip": 1.04980278,
      "balance_loss_mlp": 1.03254092,
      "epoch": 0.35280324665564405,
      "flos": 19427350124160.0,
      "grad_norm": 1.7278714421264112,
      "language_loss": 0.74146622,
      "learning_rate": 3.0029216999885613e-06,
      "loss": 0.76330769,
      "num_input_tokens_seen": 126186460,
      "step": 5868,
      "time_per_iteration": 2.4642903804779053
    },
    {
      "auxiliary_loss_clip": 0.01125515,
      "auxiliary_loss_mlp": 0.01040246,
      "balance_loss_clip": 1.05155182,
      "balance_loss_mlp": 1.02457047,
      "epoch": 0.352863369908312,
      "flos": 21504027957120.0,
      "grad_norm": 1.8903506467184195,
      "language_loss": 0.61618161,
      "learning_rate": 3.0025847263863327e-06,
      "loss": 0.63783926,
      "num_input_tokens_seen": 126206170,
      "step": 5869,
      "time_per_iteration": 3.8837032318115234
    },
    {
      "auxiliary_loss_clip": 0.01123496,
      "auxiliary_loss_mlp": 0.01040731,
      "balance_loss_clip": 1.04868746,
      "balance_loss_mlp": 1.02541852,
      "epoch": 0.35292349316098,
      "flos": 22309504490880.0,
      "grad_norm": 1.948616548591169,
      "language_loss": 0.74225891,
      "learning_rate": 3.0022477147659917e-06,
      "loss": 0.76390111,
      "num_input_tokens_seen": 126225605,
      "step": 5870,
      "time_per_iteration": 3.9331157207489014
    },
    {
      "auxiliary_loss_clip": 0.01120421,
      "auxiliary_loss_mlp": 0.01033814,
      "balance_loss_clip": 1.04699183,
      "balance_loss_mlp": 1.01885319,
      "epoch": 0.352983616413648,
      "flos": 33109745967360.0,
      "grad_norm": 1.4252127344938181,
      "language_loss": 0.71892953,
      "learning_rate": 3.001910665140316e-06,
      "loss": 0.74047184,
      "num_input_tokens_seen": 126250230,
      "step": 5871,
      "time_per_iteration": 2.7069175243377686
    },
    {
      "auxiliary_loss_clip": 0.01114977,
      "auxiliary_loss_mlp": 0.01034465,
      "balance_loss_clip": 1.0459249,
      "balance_loss_mlp": 1.0204761,
      "epoch": 0.35304373966631597,
      "flos": 18696603836160.0,
      "grad_norm": 2.2074380017367297,
      "language_loss": 0.73791277,
      "learning_rate": 3.0015735775220873e-06,
      "loss": 0.75940716,
      "num_input_tokens_seen": 126268315,
      "step": 5872,
      "time_per_iteration": 4.041405916213989
    },
    {
      "auxiliary_loss_clip": 0.01110694,
      "auxiliary_loss_mlp": 0.00794405,
      "balance_loss_clip": 1.04791975,
      "balance_loss_mlp": 1.01620209,
      "epoch": 0.35310386291898394,
      "flos": 23364954748800.0,
      "grad_norm": 1.5770157746660816,
      "language_loss": 0.82102799,
      "learning_rate": 3.001236451924089e-06,
      "loss": 0.84007901,
      "num_input_tokens_seen": 126288390,
      "step": 5873,
      "time_per_iteration": 2.618715763092041
    },
    {
      "auxiliary_loss_clip": 0.01114701,
      "auxiliary_loss_mlp": 0.0104482,
      "balance_loss_clip": 1.04773211,
      "balance_loss_mlp": 1.02810645,
      "epoch": 0.3531639861716519,
      "flos": 24461954064000.0,
      "grad_norm": 1.8312039007593897,
      "language_loss": 0.65897655,
      "learning_rate": 3.000899288359104e-06,
      "loss": 0.68057179,
      "num_input_tokens_seen": 126305750,
      "step": 5874,
      "time_per_iteration": 2.621669292449951
    },
    {
      "auxiliary_loss_clip": 0.01043001,
      "auxiliary_loss_mlp": 0.01002705,
      "balance_loss_clip": 1.02338457,
      "balance_loss_mlp": 1.00024915,
      "epoch": 0.35322410942431987,
      "flos": 70312446881280.0,
      "grad_norm": 0.7589241803489293,
      "language_loss": 0.61531889,
      "learning_rate": 3.000562086839917e-06,
      "loss": 0.63577598,
      "num_input_tokens_seen": 126362495,
      "step": 5875,
      "time_per_iteration": 3.029635429382324
    },
    {
      "auxiliary_loss_clip": 0.01074251,
      "auxiliary_loss_mlp": 0.01046761,
      "balance_loss_clip": 1.04577875,
      "balance_loss_mlp": 1.03163362,
      "epoch": 0.35328423267698783,
      "flos": 19820894509440.0,
      "grad_norm": 1.893886986818516,
      "language_loss": 0.79803073,
      "learning_rate": 3.0002248473793163e-06,
      "loss": 0.81924081,
      "num_input_tokens_seen": 126378320,
      "step": 5876,
      "time_per_iteration": 2.5927631855010986
    },
    {
      "auxiliary_loss_clip": 0.01030166,
      "auxiliary_loss_mlp": 0.00833022,
      "balance_loss_clip": 1.03183699,
      "balance_loss_mlp": 1.11425138,
      "epoch": 0.3533443559296558,
      "flos": 60826356391680.0,
      "grad_norm": 0.6874110913094893,
      "language_loss": 0.56770647,
      "learning_rate": 2.999887569990088e-06,
      "loss": 0.58633828,
      "num_input_tokens_seen": 126442735,
      "step": 5877,
      "time_per_iteration": 3.2602016925811768
    },
    {
      "auxiliary_loss_clip": 0.01100188,
      "auxiliary_loss_mlp": 0.01033407,
      "balance_loss_clip": 1.04741549,
      "balance_loss_mlp": 1.018381,
      "epoch": 0.35340447918232376,
      "flos": 24755775315840.0,
      "grad_norm": 1.628929703377693,
      "language_loss": 0.71471298,
      "learning_rate": 2.999550254685024e-06,
      "loss": 0.73604894,
      "num_input_tokens_seen": 126463090,
      "step": 5878,
      "time_per_iteration": 2.5671818256378174
    },
    {
      "auxiliary_loss_clip": 0.01104026,
      "auxiliary_loss_mlp": 0.0103634,
      "balance_loss_clip": 1.04665756,
      "balance_loss_mlp": 1.02128363,
      "epoch": 0.3534646024349917,
      "flos": 21796304924160.0,
      "grad_norm": 1.7268710787497756,
      "language_loss": 0.78776789,
      "learning_rate": 2.9992129014769136e-06,
      "loss": 0.80917156,
      "num_input_tokens_seen": 126482105,
      "step": 5879,
      "time_per_iteration": 2.5543177127838135
    },
    {
      "auxiliary_loss_clip": 0.01103535,
      "auxiliary_loss_mlp": 0.01046811,
      "balance_loss_clip": 1.0485642,
      "balance_loss_mlp": 1.0295136,
      "epoch": 0.3535247256876597,
      "flos": 20012119539840.0,
      "grad_norm": 2.068417698749051,
      "language_loss": 0.63326859,
      "learning_rate": 2.9988755103785493e-06,
      "loss": 0.65477204,
      "num_input_tokens_seen": 126502125,
      "step": 5880,
      "time_per_iteration": 2.566581964492798
    },
    {
      "auxiliary_loss_clip": 0.01110784,
      "auxiliary_loss_mlp": 0.01033537,
      "balance_loss_clip": 1.04823625,
      "balance_loss_mlp": 1.0179683,
      "epoch": 0.35358484894032766,
      "flos": 18187929383040.0,
      "grad_norm": 3.4726871088134543,
      "language_loss": 0.65531564,
      "learning_rate": 2.998538081402727e-06,
      "loss": 0.67675883,
      "num_input_tokens_seen": 126521950,
      "step": 5881,
      "time_per_iteration": 2.545579433441162
    },
    {
      "auxiliary_loss_clip": 0.01111761,
      "auxiliary_loss_mlp": 0.01032703,
      "balance_loss_clip": 1.04682302,
      "balance_loss_mlp": 1.01876116,
      "epoch": 0.3536449721929956,
      "flos": 22820369673600.0,
      "grad_norm": 1.3711597106571851,
      "language_loss": 0.75844944,
      "learning_rate": 2.998200614562239e-06,
      "loss": 0.77989405,
      "num_input_tokens_seen": 126542445,
      "step": 5882,
      "time_per_iteration": 2.525022268295288
    },
    {
      "auxiliary_loss_clip": 0.01107642,
      "auxiliary_loss_mlp": 0.01047776,
      "balance_loss_clip": 1.04686475,
      "balance_loss_mlp": 1.03070498,
      "epoch": 0.3537050954456636,
      "flos": 26432336574720.0,
      "grad_norm": 2.442342580905341,
      "language_loss": 0.7051248,
      "learning_rate": 2.9978631098698847e-06,
      "loss": 0.72667897,
      "num_input_tokens_seen": 126560690,
      "step": 5883,
      "time_per_iteration": 2.586571216583252
    },
    {
      "auxiliary_loss_clip": 0.01100512,
      "auxiliary_loss_mlp": 0.01037263,
      "balance_loss_clip": 1.0495882,
      "balance_loss_mlp": 1.02146804,
      "epoch": 0.3537652186983316,
      "flos": 17197153562880.0,
      "grad_norm": 2.251627720735446,
      "language_loss": 0.78039622,
      "learning_rate": 2.9975255673384614e-06,
      "loss": 0.80177391,
      "num_input_tokens_seen": 126577620,
      "step": 5884,
      "time_per_iteration": 2.535024642944336
    },
    {
      "auxiliary_loss_clip": 0.01111462,
      "auxiliary_loss_mlp": 0.01036128,
      "balance_loss_clip": 1.04832029,
      "balance_loss_mlp": 1.02187634,
      "epoch": 0.3538253419509996,
      "flos": 19536769929600.0,
      "grad_norm": 1.971236449653369,
      "language_loss": 0.75404173,
      "learning_rate": 2.9971879869807673e-06,
      "loss": 0.77551764,
      "num_input_tokens_seen": 126596235,
      "step": 5885,
      "time_per_iteration": 2.5211470127105713
    },
    {
      "auxiliary_loss_clip": 0.01088748,
      "auxiliary_loss_mlp": 0.0104033,
      "balance_loss_clip": 1.05036426,
      "balance_loss_mlp": 1.02408135,
      "epoch": 0.35388546520366754,
      "flos": 12128578335360.0,
      "grad_norm": 2.179476376925401,
      "language_loss": 0.83326113,
      "learning_rate": 2.996850368809606e-06,
      "loss": 0.85455191,
      "num_input_tokens_seen": 126612830,
      "step": 5886,
      "time_per_iteration": 2.6037914752960205
    },
    {
      "auxiliary_loss_clip": 0.01131002,
      "auxiliary_loss_mlp": 0.01032639,
      "balance_loss_clip": 1.04753017,
      "balance_loss_mlp": 1.01695132,
      "epoch": 0.3539455884563355,
      "flos": 19678149861120.0,
      "grad_norm": 3.043492202349756,
      "language_loss": 0.78383017,
      "learning_rate": 2.9965127128377787e-06,
      "loss": 0.80546653,
      "num_input_tokens_seen": 126630910,
      "step": 5887,
      "time_per_iteration": 2.4791693687438965
    },
    {
      "auxiliary_loss_clip": 0.01075351,
      "auxiliary_loss_mlp": 0.01045108,
      "balance_loss_clip": 1.0453614,
      "balance_loss_mlp": 1.02974212,
      "epoch": 0.35400571170900347,
      "flos": 18072045129600.0,
      "grad_norm": 1.70557531171087,
      "language_loss": 0.65783507,
      "learning_rate": 2.996175019078089e-06,
      "loss": 0.67903966,
      "num_input_tokens_seen": 126648365,
      "step": 5888,
      "time_per_iteration": 2.6191439628601074
    },
    {
      "auxiliary_loss_clip": 0.01103213,
      "auxiliary_loss_mlp": 0.01040734,
      "balance_loss_clip": 1.04580677,
      "balance_loss_mlp": 1.02584457,
      "epoch": 0.35406583496167143,
      "flos": 26068058795520.0,
      "grad_norm": 1.6041871621726798,
      "language_loss": 0.76967895,
      "learning_rate": 2.9958372875433437e-06,
      "loss": 0.79111838,
      "num_input_tokens_seen": 126667500,
      "step": 5889,
      "time_per_iteration": 2.5735135078430176
    },
    {
      "auxiliary_loss_clip": 0.01096952,
      "auxiliary_loss_mlp": 0.0103385,
      "balance_loss_clip": 1.04932606,
      "balance_loss_mlp": 1.0192703,
      "epoch": 0.3541259582143394,
      "flos": 19792453916160.0,
      "grad_norm": 1.698733867397149,
      "language_loss": 0.8066355,
      "learning_rate": 2.9954995182463478e-06,
      "loss": 0.8279435,
      "num_input_tokens_seen": 126686820,
      "step": 5890,
      "time_per_iteration": 2.5600216388702393
    },
    {
      "auxiliary_loss_clip": 0.01111626,
      "auxiliary_loss_mlp": 0.01035016,
      "balance_loss_clip": 1.04650974,
      "balance_loss_mlp": 1.0221231,
      "epoch": 0.35418608146700736,
      "flos": 24022084112640.0,
      "grad_norm": 1.7579952005499295,
      "language_loss": 0.79573357,
      "learning_rate": 2.99516171119991e-06,
      "loss": 0.81719995,
      "num_input_tokens_seen": 126706965,
      "step": 5891,
      "time_per_iteration": 2.5740268230438232
    },
    {
      "auxiliary_loss_clip": 0.0109341,
      "auxiliary_loss_mlp": 0.01044268,
      "balance_loss_clip": 1.0453887,
      "balance_loss_mlp": 1.02803206,
      "epoch": 0.35424620471967533,
      "flos": 12385770693120.0,
      "grad_norm": 1.8392645668761929,
      "language_loss": 0.72995037,
      "learning_rate": 2.9948238664168415e-06,
      "loss": 0.75132716,
      "num_input_tokens_seen": 126724015,
      "step": 5892,
      "time_per_iteration": 2.56139874458313
    },
    {
      "auxiliary_loss_clip": 0.01107719,
      "auxiliary_loss_mlp": 0.01039443,
      "balance_loss_clip": 1.04696393,
      "balance_loss_mlp": 1.02420855,
      "epoch": 0.3543063279723433,
      "flos": 19673624747520.0,
      "grad_norm": 1.9777894333065336,
      "language_loss": 0.67104053,
      "learning_rate": 2.9944859839099518e-06,
      "loss": 0.69251215,
      "num_input_tokens_seen": 126737565,
      "step": 5893,
      "time_per_iteration": 2.528381586074829
    },
    {
      "auxiliary_loss_clip": 0.01081039,
      "auxiliary_loss_mlp": 0.01036667,
      "balance_loss_clip": 1.04858756,
      "balance_loss_mlp": 1.02089596,
      "epoch": 0.35436645122501126,
      "flos": 21909208348800.0,
      "grad_norm": 1.830654292283894,
      "language_loss": 0.69861346,
      "learning_rate": 2.9941480636920533e-06,
      "loss": 0.71979052,
      "num_input_tokens_seen": 126756095,
      "step": 5894,
      "time_per_iteration": 2.5980918407440186
    },
    {
      "auxiliary_loss_clip": 0.011106,
      "auxiliary_loss_mlp": 0.00850742,
      "balance_loss_clip": 1.04706383,
      "balance_loss_mlp": 1.12833142,
      "epoch": 0.3544265744776792,
      "flos": 21719527603200.0,
      "grad_norm": 2.510381342547178,
      "language_loss": 0.74569404,
      "learning_rate": 2.9938101057759615e-06,
      "loss": 0.76530749,
      "num_input_tokens_seen": 126775455,
      "step": 5895,
      "time_per_iteration": 2.572514533996582
    },
    {
      "auxiliary_loss_clip": 0.0110526,
      "auxiliary_loss_mlp": 0.01039974,
      "balance_loss_clip": 1.04980445,
      "balance_loss_mlp": 1.02502525,
      "epoch": 0.3544866977303472,
      "flos": 21213223447680.0,
      "grad_norm": 2.005942085986966,
      "language_loss": 0.8361218,
      "learning_rate": 2.993472110174491e-06,
      "loss": 0.85757411,
      "num_input_tokens_seen": 126792320,
      "step": 5896,
      "time_per_iteration": 2.532672643661499
    },
    {
      "auxiliary_loss_clip": 0.0110629,
      "auxiliary_loss_mlp": 0.00845838,
      "balance_loss_clip": 1.04585528,
      "balance_loss_mlp": 1.11739922,
      "epoch": 0.35454682098301515,
      "flos": 29311402371840.0,
      "grad_norm": 2.1804086194929324,
      "language_loss": 0.70109695,
      "learning_rate": 2.9931340769004576e-06,
      "loss": 0.72061819,
      "num_input_tokens_seen": 126813680,
      "step": 5897,
      "time_per_iteration": 2.636375904083252
    },
    {
      "auxiliary_loss_clip": 0.01104333,
      "auxiliary_loss_mlp": 0.01043822,
      "balance_loss_clip": 1.04513574,
      "balance_loss_mlp": 1.02729964,
      "epoch": 0.3546069442356832,
      "flos": 24316587722880.0,
      "grad_norm": 1.5892179546303355,
      "language_loss": 0.81791103,
      "learning_rate": 2.9927960059666816e-06,
      "loss": 0.8393926,
      "num_input_tokens_seen": 126834395,
      "step": 5898,
      "time_per_iteration": 2.567105770111084
    },
    {
      "auxiliary_loss_clip": 0.01125104,
      "auxiliary_loss_mlp": 0.01041448,
      "balance_loss_clip": 1.0442009,
      "balance_loss_mlp": 1.02750611,
      "epoch": 0.35466706748835114,
      "flos": 22857285876480.0,
      "grad_norm": 1.4778620150762192,
      "language_loss": 0.74151313,
      "learning_rate": 2.9924578973859804e-06,
      "loss": 0.76317859,
      "num_input_tokens_seen": 126855145,
      "step": 5899,
      "time_per_iteration": 2.5139100551605225
    },
    {
      "auxiliary_loss_clip": 0.01128217,
      "auxiliary_loss_mlp": 0.00815382,
      "balance_loss_clip": 1.04502213,
      "balance_loss_mlp": 1.06092286,
      "epoch": 0.3547271907410191,
      "flos": 28330107742080.0,
      "grad_norm": 1.8180254913807365,
      "language_loss": 0.79545325,
      "learning_rate": 2.9921197511711763e-06,
      "loss": 0.81488931,
      "num_input_tokens_seen": 126873790,
      "step": 5900,
      "time_per_iteration": 2.5379104614257812
    },
    {
      "auxiliary_loss_clip": 0.01105229,
      "auxiliary_loss_mlp": 0.01040203,
      "balance_loss_clip": 1.04336488,
      "balance_loss_mlp": 1.02445531,
      "epoch": 0.35478731399368707,
      "flos": 23514092017920.0,
      "grad_norm": 1.8692506618451001,
      "language_loss": 0.81474996,
      "learning_rate": 2.991781567335093e-06,
      "loss": 0.83620429,
      "num_input_tokens_seen": 126892865,
      "step": 5901,
      "time_per_iteration": 2.5517306327819824
    },
    {
      "auxiliary_loss_clip": 0.01117284,
      "auxiliary_loss_mlp": 0.00804576,
      "balance_loss_clip": 1.04686153,
      "balance_loss_mlp": 1.04012465,
      "epoch": 0.35484743724635504,
      "flos": 18624315715200.0,
      "grad_norm": 1.815426327020132,
      "language_loss": 0.76073986,
      "learning_rate": 2.9914433458905525e-06,
      "loss": 0.77995849,
      "num_input_tokens_seen": 126911935,
      "step": 5902,
      "time_per_iteration": 2.524336576461792
    },
    {
      "auxiliary_loss_clip": 0.01117417,
      "auxiliary_loss_mlp": 0.01038539,
      "balance_loss_clip": 1.04483438,
      "balance_loss_mlp": 1.02487159,
      "epoch": 0.354907560499023,
      "flos": 17384499924480.0,
      "grad_norm": 3.2326892984657474,
      "language_loss": 0.70299548,
      "learning_rate": 2.991105086850381e-06,
      "loss": 0.72455502,
      "num_input_tokens_seen": 126930040,
      "step": 5903,
      "time_per_iteration": 2.478618860244751
    },
    {
      "auxiliary_loss_clip": 0.01117487,
      "auxiliary_loss_mlp": 0.01034057,
      "balance_loss_clip": 1.04408693,
      "balance_loss_mlp": 1.01952553,
      "epoch": 0.35496768375169097,
      "flos": 19208546426880.0,
      "grad_norm": 2.7251896547937067,
      "language_loss": 0.75037169,
      "learning_rate": 2.9907667902274053e-06,
      "loss": 0.77188706,
      "num_input_tokens_seen": 126948390,
      "step": 5904,
      "time_per_iteration": 3.8400659561157227
    },
    {
      "auxiliary_loss_clip": 0.01109055,
      "auxiliary_loss_mlp": 0.00797305,
      "balance_loss_clip": 1.04735982,
      "balance_loss_mlp": 1.02403474,
      "epoch": 0.35502780700435893,
      "flos": 18332792933760.0,
      "grad_norm": 1.845250381263542,
      "language_loss": 0.78647125,
      "learning_rate": 2.9904284560344536e-06,
      "loss": 0.80553484,
      "num_input_tokens_seen": 126964905,
      "step": 5905,
      "time_per_iteration": 2.5221595764160156
    },
    {
      "auxiliary_loss_clip": 0.01091287,
      "auxiliary_loss_mlp": 0.01035918,
      "balance_loss_clip": 1.04293203,
      "balance_loss_mlp": 1.02301955,
      "epoch": 0.3550879302570269,
      "flos": 15448555578240.0,
      "grad_norm": 1.9052570640456417,
      "language_loss": 0.72377002,
      "learning_rate": 2.990090084284356e-06,
      "loss": 0.74504203,
      "num_input_tokens_seen": 126982000,
      "step": 5906,
      "time_per_iteration": 2.5466530323028564
    },
    {
      "auxiliary_loss_clip": 0.01101101,
      "auxiliary_loss_mlp": 0.01036428,
      "balance_loss_clip": 1.04728007,
      "balance_loss_mlp": 1.02021599,
      "epoch": 0.35514805350969486,
      "flos": 21979197999360.0,
      "grad_norm": 2.1334005490454646,
      "language_loss": 0.74469888,
      "learning_rate": 2.9897516749899426e-06,
      "loss": 0.76607412,
      "num_input_tokens_seen": 126998390,
      "step": 5907,
      "time_per_iteration": 2.5080583095550537
    },
    {
      "auxiliary_loss_clip": 0.01063269,
      "auxiliary_loss_mlp": 0.01038131,
      "balance_loss_clip": 1.04331481,
      "balance_loss_mlp": 1.02039289,
      "epoch": 0.3552081767623628,
      "flos": 29861949104640.0,
      "grad_norm": 1.785159971212328,
      "language_loss": 0.75681353,
      "learning_rate": 2.989413228164047e-06,
      "loss": 0.7778275,
      "num_input_tokens_seen": 127020220,
      "step": 5908,
      "time_per_iteration": 4.0817930698394775
    },
    {
      "auxiliary_loss_clip": 0.01106872,
      "auxiliary_loss_mlp": 0.01038946,
      "balance_loss_clip": 1.04797769,
      "balance_loss_mlp": 1.02436662,
      "epoch": 0.3552683000150308,
      "flos": 26432264747520.0,
      "grad_norm": 2.343384075611453,
      "language_loss": 0.68337584,
      "learning_rate": 2.989074743819502e-06,
      "loss": 0.7048341,
      "num_input_tokens_seen": 127038585,
      "step": 5909,
      "time_per_iteration": 4.025432586669922
    },
    {
      "auxiliary_loss_clip": 0.01114331,
      "auxiliary_loss_mlp": 0.01037556,
      "balance_loss_clip": 1.04907548,
      "balance_loss_mlp": 1.02373993,
      "epoch": 0.35532842326769876,
      "flos": 19785989468160.0,
      "grad_norm": 1.8363308439312656,
      "language_loss": 0.7849791,
      "learning_rate": 2.988736221969144e-06,
      "loss": 0.80649793,
      "num_input_tokens_seen": 127056215,
      "step": 5910,
      "time_per_iteration": 3.9087181091308594
    },
    {
      "auxiliary_loss_clip": 0.01108576,
      "auxiliary_loss_mlp": 0.01041932,
      "balance_loss_clip": 1.04369974,
      "balance_loss_mlp": 1.02553439,
      "epoch": 0.3553885465203668,
      "flos": 17239277237760.0,
      "grad_norm": 1.5506054252026462,
      "language_loss": 0.70843273,
      "learning_rate": 2.98839766262581e-06,
      "loss": 0.72993785,
      "num_input_tokens_seen": 127075825,
      "step": 5911,
      "time_per_iteration": 2.5169689655303955
    },
    {
      "auxiliary_loss_clip": 0.01115483,
      "auxiliary_loss_mlp": 0.01039975,
      "balance_loss_clip": 1.0440197,
      "balance_loss_mlp": 1.02510905,
      "epoch": 0.35544866977303474,
      "flos": 14934350430720.0,
      "grad_norm": 2.2327494856359995,
      "language_loss": 0.86784804,
      "learning_rate": 2.9880590658023366e-06,
      "loss": 0.88940263,
      "num_input_tokens_seen": 127091205,
      "step": 5912,
      "time_per_iteration": 2.4740257263183594
    },
    {
      "auxiliary_loss_clip": 0.01108819,
      "auxiliary_loss_mlp": 0.01034894,
      "balance_loss_clip": 1.04750443,
      "balance_loss_mlp": 1.02110791,
      "epoch": 0.3555087930257027,
      "flos": 19756040503680.0,
      "grad_norm": 4.886429084393904,
      "language_loss": 0.77356434,
      "learning_rate": 2.9877204315115646e-06,
      "loss": 0.79500151,
      "num_input_tokens_seen": 127109210,
      "step": 5913,
      "time_per_iteration": 2.5170209407806396
    },
    {
      "auxiliary_loss_clip": 0.01095072,
      "auxiliary_loss_mlp": 0.01035032,
      "balance_loss_clip": 1.05080402,
      "balance_loss_mlp": 1.0205425,
      "epoch": 0.3555689162783707,
      "flos": 21068252156160.0,
      "grad_norm": 1.300305217949161,
      "language_loss": 0.82614881,
      "learning_rate": 2.9873817597663353e-06,
      "loss": 0.8474499,
      "num_input_tokens_seen": 127128400,
      "step": 5914,
      "time_per_iteration": 2.5719785690307617
    },
    {
      "auxiliary_loss_clip": 0.01129615,
      "auxiliary_loss_mlp": 0.01033783,
      "balance_loss_clip": 1.0464859,
      "balance_loss_mlp": 1.0185957,
      "epoch": 0.35562903953103864,
      "flos": 33069633454080.0,
      "grad_norm": 2.0705083939855324,
      "language_loss": 0.706617,
      "learning_rate": 2.98704305057949e-06,
      "loss": 0.72825092,
      "num_input_tokens_seen": 127149965,
      "step": 5915,
      "time_per_iteration": 2.5745949745178223
    },
    {
      "auxiliary_loss_clip": 0.01117842,
      "auxiliary_loss_mlp": 0.0103844,
      "balance_loss_clip": 1.04573226,
      "balance_loss_mlp": 1.02430749,
      "epoch": 0.3556891627837066,
      "flos": 20557853850240.0,
      "grad_norm": 1.8996638126963372,
      "language_loss": 0.76175886,
      "learning_rate": 2.9867043039638737e-06,
      "loss": 0.78332168,
      "num_input_tokens_seen": 127169865,
      "step": 5916,
      "time_per_iteration": 2.505423069000244
    },
    {
      "auxiliary_loss_clip": 0.01097738,
      "auxiliary_loss_mlp": 0.01037196,
      "balance_loss_clip": 1.04351783,
      "balance_loss_mlp": 1.02298641,
      "epoch": 0.35574928603637457,
      "flos": 20703327932160.0,
      "grad_norm": 1.7476965516960996,
      "language_loss": 0.88649791,
      "learning_rate": 2.986365519932332e-06,
      "loss": 0.90784729,
      "num_input_tokens_seen": 127188075,
      "step": 5917,
      "time_per_iteration": 2.55876088142395
    },
    {
      "auxiliary_loss_clip": 0.01058378,
      "auxiliary_loss_mlp": 0.01042279,
      "balance_loss_clip": 1.04639721,
      "balance_loss_mlp": 1.02462399,
      "epoch": 0.35580940928904253,
      "flos": 15194595444480.0,
      "grad_norm": 1.935987539278264,
      "language_loss": 0.75020069,
      "learning_rate": 2.98602669849771e-06,
      "loss": 0.77120721,
      "num_input_tokens_seen": 127206065,
      "step": 5918,
      "time_per_iteration": 2.612309217453003
    },
    {
      "auxiliary_loss_clip": 0.01043285,
      "auxiliary_loss_mlp": 0.01005142,
      "balance_loss_clip": 1.03003931,
      "balance_loss_mlp": 1.00317526,
      "epoch": 0.3558695325417105,
      "flos": 58639145431680.0,
      "grad_norm": 0.9486154769595739,
      "language_loss": 0.63960767,
      "learning_rate": 2.985687839672857e-06,
      "loss": 0.660092,
      "num_input_tokens_seen": 127257885,
      "step": 5919,
      "time_per_iteration": 2.868290901184082
    },
    {
      "auxiliary_loss_clip": 0.01116442,
      "auxiliary_loss_mlp": 0.01033077,
      "balance_loss_clip": 1.04589784,
      "balance_loss_mlp": 1.01794314,
      "epoch": 0.35592965579437846,
      "flos": 22018233104640.0,
      "grad_norm": 2.2777586381794857,
      "language_loss": 0.7377978,
      "learning_rate": 2.9853489434706223e-06,
      "loss": 0.75929296,
      "num_input_tokens_seen": 127275550,
      "step": 5920,
      "time_per_iteration": 2.51005220413208
    },
    {
      "auxiliary_loss_clip": 0.01088301,
      "auxiliary_loss_mlp": 0.01034474,
      "balance_loss_clip": 1.04487062,
      "balance_loss_mlp": 1.01954854,
      "epoch": 0.35598977904704643,
      "flos": 23367684182400.0,
      "grad_norm": 1.7147366526547945,
      "language_loss": 0.77297139,
      "learning_rate": 2.985010009903857e-06,
      "loss": 0.79419911,
      "num_input_tokens_seen": 127295110,
      "step": 5921,
      "time_per_iteration": 2.5964930057525635
    },
    {
      "auxiliary_loss_clip": 0.01102198,
      "auxiliary_loss_mlp": 0.01038194,
      "balance_loss_clip": 1.04358113,
      "balance_loss_mlp": 1.02374625,
      "epoch": 0.3560499022997144,
      "flos": 17785334770560.0,
      "grad_norm": 1.723932522326925,
      "language_loss": 0.67320609,
      "learning_rate": 2.9846710389854133e-06,
      "loss": 0.69460994,
      "num_input_tokens_seen": 127312865,
      "step": 5922,
      "time_per_iteration": 2.527249813079834
    },
    {
      "auxiliary_loss_clip": 0.01118711,
      "auxiliary_loss_mlp": 0.01036312,
      "balance_loss_clip": 1.05077577,
      "balance_loss_mlp": 1.02111316,
      "epoch": 0.35611002555238236,
      "flos": 20740459616640.0,
      "grad_norm": 2.1678192262330875,
      "language_loss": 0.79424655,
      "learning_rate": 2.9843320307281454e-06,
      "loss": 0.81579685,
      "num_input_tokens_seen": 127331710,
      "step": 5923,
      "time_per_iteration": 2.5060691833496094
    },
    {
      "auxiliary_loss_clip": 0.01114032,
      "auxiliary_loss_mlp": 0.0103813,
      "balance_loss_clip": 1.05080605,
      "balance_loss_mlp": 1.02338338,
      "epoch": 0.3561701488050504,
      "flos": 19462219251840.0,
      "grad_norm": 1.6497729199181856,
      "language_loss": 0.85539162,
      "learning_rate": 2.983992985144908e-06,
      "loss": 0.87691319,
      "num_input_tokens_seen": 127350950,
      "step": 5924,
      "time_per_iteration": 2.5426080226898193
    },
    {
      "auxiliary_loss_clip": 0.01107841,
      "auxiliary_loss_mlp": 0.01037452,
      "balance_loss_clip": 1.04977393,
      "balance_loss_mlp": 1.02160883,
      "epoch": 0.35623027205771834,
      "flos": 30774942023040.0,
      "grad_norm": 2.2827636997021648,
      "language_loss": 0.77633607,
      "learning_rate": 2.9836539022485578e-06,
      "loss": 0.79778898,
      "num_input_tokens_seen": 127369385,
      "step": 5925,
      "time_per_iteration": 2.598454236984253
    },
    {
      "auxiliary_loss_clip": 0.01069496,
      "auxiliary_loss_mlp": 0.01044651,
      "balance_loss_clip": 1.04355383,
      "balance_loss_mlp": 1.02975607,
      "epoch": 0.3562903953103863,
      "flos": 16981079299200.0,
      "grad_norm": 1.8641205229310265,
      "language_loss": 0.75787646,
      "learning_rate": 2.9833147820519535e-06,
      "loss": 0.77901793,
      "num_input_tokens_seen": 127386965,
      "step": 5926,
      "time_per_iteration": 2.5960216522216797
    },
    {
      "auxiliary_loss_clip": 0.01104279,
      "auxiliary_loss_mlp": 0.00838871,
      "balance_loss_clip": 1.04748142,
      "balance_loss_mlp": 1.09982407,
      "epoch": 0.3563505185630543,
      "flos": 23839837482240.0,
      "grad_norm": 1.7908867845668475,
      "language_loss": 0.69476068,
      "learning_rate": 2.9829756245679544e-06,
      "loss": 0.71419215,
      "num_input_tokens_seen": 127406075,
      "step": 5927,
      "time_per_iteration": 2.592653751373291
    },
    {
      "auxiliary_loss_clip": 0.01129811,
      "auxiliary_loss_mlp": 0.01036621,
      "balance_loss_clip": 1.0472225,
      "balance_loss_mlp": 1.02278137,
      "epoch": 0.35641064181572224,
      "flos": 22273450214400.0,
      "grad_norm": 1.9810763922489294,
      "language_loss": 0.79522514,
      "learning_rate": 2.9826364298094212e-06,
      "loss": 0.81688941,
      "num_input_tokens_seen": 127425350,
      "step": 5928,
      "time_per_iteration": 2.4980735778808594
    },
    {
      "auxiliary_loss_clip": 0.01131312,
      "auxiliary_loss_mlp": 0.01036516,
      "balance_loss_clip": 1.04841363,
      "balance_loss_mlp": 1.0220139,
      "epoch": 0.3564707650683902,
      "flos": 23001251587200.0,
      "grad_norm": 1.5959944680572675,
      "language_loss": 0.82120091,
      "learning_rate": 2.982297197789215e-06,
      "loss": 0.84287924,
      "num_input_tokens_seen": 127446335,
      "step": 5929,
      "time_per_iteration": 2.4944825172424316
    },
    {
      "auxiliary_loss_clip": 0.01116979,
      "auxiliary_loss_mlp": 0.0103565,
      "balance_loss_clip": 1.04750919,
      "balance_loss_mlp": 1.02147603,
      "epoch": 0.35653088832105817,
      "flos": 14684268965760.0,
      "grad_norm": 1.591366041767244,
      "language_loss": 0.7028597,
      "learning_rate": 2.981957928520201e-06,
      "loss": 0.72438598,
      "num_input_tokens_seen": 127462795,
      "step": 5930,
      "time_per_iteration": 2.4946279525756836
    },
    {
      "auxiliary_loss_clip": 0.0111807,
      "auxiliary_loss_mlp": 0.01041516,
      "balance_loss_clip": 1.04957342,
      "balance_loss_mlp": 1.02643645,
      "epoch": 0.35659101157372614,
      "flos": 23477068074240.0,
      "grad_norm": 2.0125059655742112,
      "language_loss": 0.68054563,
      "learning_rate": 2.981618622015244e-06,
      "loss": 0.70214152,
      "num_input_tokens_seen": 127482675,
      "step": 5931,
      "time_per_iteration": 2.5273728370666504
    },
    {
      "auxiliary_loss_clip": 0.01121462,
      "auxiliary_loss_mlp": 0.010336,
      "balance_loss_clip": 1.04881334,
      "balance_loss_mlp": 1.01869845,
      "epoch": 0.3566511348263941,
      "flos": 26578672583040.0,
      "grad_norm": 1.5793137997960722,
      "language_loss": 0.67461705,
      "learning_rate": 2.981279278287211e-06,
      "loss": 0.69616765,
      "num_input_tokens_seen": 127502275,
      "step": 5932,
      "time_per_iteration": 2.5972237586975098
    },
    {
      "auxiliary_loss_clip": 0.01080837,
      "auxiliary_loss_mlp": 0.01033702,
      "balance_loss_clip": 1.05182064,
      "balance_loss_mlp": 1.01921225,
      "epoch": 0.35671125807906207,
      "flos": 13115008609920.0,
      "grad_norm": 2.0575435670946547,
      "language_loss": 0.79205954,
      "learning_rate": 2.980939897348969e-06,
      "loss": 0.81320488,
      "num_input_tokens_seen": 127520195,
      "step": 5933,
      "time_per_iteration": 2.6086466312408447
    },
    {
      "auxiliary_loss_clip": 0.01113546,
      "auxiliary_loss_mlp": 0.01048867,
      "balance_loss_clip": 1.04311848,
      "balance_loss_mlp": 1.03195119,
      "epoch": 0.35677138133173003,
      "flos": 33000577557120.0,
      "grad_norm": 1.3926870747702342,
      "language_loss": 0.69856393,
      "learning_rate": 2.980600479213388e-06,
      "loss": 0.72018802,
      "num_input_tokens_seen": 127544495,
      "step": 5934,
      "time_per_iteration": 2.640439510345459
    },
    {
      "auxiliary_loss_clip": 0.0111009,
      "auxiliary_loss_mlp": 0.00830075,
      "balance_loss_clip": 1.04859114,
      "balance_loss_mlp": 1.07722557,
      "epoch": 0.356831504584398,
      "flos": 20777842696320.0,
      "grad_norm": 2.3882184748866293,
      "language_loss": 0.71324217,
      "learning_rate": 2.9802610238933384e-06,
      "loss": 0.73264384,
      "num_input_tokens_seen": 127563810,
      "step": 5935,
      "time_per_iteration": 2.6011602878570557
    },
    {
      "auxiliary_loss_clip": 0.01101052,
      "auxiliary_loss_mlp": 0.01038131,
      "balance_loss_clip": 1.0468781,
      "balance_loss_mlp": 1.02390969,
      "epoch": 0.35689162783706596,
      "flos": 12165566365440.0,
      "grad_norm": 2.5809213160519056,
      "language_loss": 0.78417718,
      "learning_rate": 2.979921531401692e-06,
      "loss": 0.80556905,
      "num_input_tokens_seen": 127579065,
      "step": 5936,
      "time_per_iteration": 2.5282063484191895
    },
    {
      "auxiliary_loss_clip": 0.01119312,
      "auxiliary_loss_mlp": 0.00815233,
      "balance_loss_clip": 1.04761231,
      "balance_loss_mlp": 1.05570483,
      "epoch": 0.356951751089734,
      "flos": 23841489507840.0,
      "grad_norm": 1.4429658904688256,
      "language_loss": 0.64316243,
      "learning_rate": 2.9795820017513242e-06,
      "loss": 0.66250789,
      "num_input_tokens_seen": 127599105,
      "step": 5937,
      "time_per_iteration": 2.577720880508423
    },
    {
      "auxiliary_loss_clip": 0.01134403,
      "auxiliary_loss_mlp": 0.00805491,
      "balance_loss_clip": 1.04921257,
      "balance_loss_mlp": 1.03754735,
      "epoch": 0.35701187434240195,
      "flos": 11722176881280.0,
      "grad_norm": 2.5692444882079695,
      "language_loss": 0.78402746,
      "learning_rate": 2.9792424349551073e-06,
      "loss": 0.80342644,
      "num_input_tokens_seen": 127614940,
      "step": 5938,
      "time_per_iteration": 2.469907283782959
    },
    {
      "auxiliary_loss_clip": 0.01098856,
      "auxiliary_loss_mlp": 0.0104227,
      "balance_loss_clip": 1.0538733,
      "balance_loss_mlp": 1.02801228,
      "epoch": 0.3570719975950699,
      "flos": 24898879100160.0,
      "grad_norm": 1.5137305530285494,
      "language_loss": 0.80483687,
      "learning_rate": 2.9789028310259202e-06,
      "loss": 0.82624805,
      "num_input_tokens_seen": 127634960,
      "step": 5939,
      "time_per_iteration": 2.6085402965545654
    },
    {
      "auxiliary_loss_clip": 0.0111103,
      "auxiliary_loss_mlp": 0.01034921,
      "balance_loss_clip": 1.04574907,
      "balance_loss_mlp": 1.01982903,
      "epoch": 0.3571321208477379,
      "flos": 25994836920960.0,
      "grad_norm": 1.7771710647606638,
      "language_loss": 0.78722101,
      "learning_rate": 2.9785631899766395e-06,
      "loss": 0.80868053,
      "num_input_tokens_seen": 127654545,
      "step": 5940,
      "time_per_iteration": 2.555312395095825
    },
    {
      "auxiliary_loss_clip": 0.01112954,
      "auxiliary_loss_mlp": 0.01033971,
      "balance_loss_clip": 1.05021882,
      "balance_loss_mlp": 1.01883757,
      "epoch": 0.35719224410040584,
      "flos": 14501663199360.0,
      "grad_norm": 1.933200498289488,
      "language_loss": 0.72018665,
      "learning_rate": 2.9782235118201443e-06,
      "loss": 0.74165595,
      "num_input_tokens_seen": 127672320,
      "step": 5941,
      "time_per_iteration": 2.5324909687042236
    },
    {
      "auxiliary_loss_clip": 0.01124933,
      "auxiliary_loss_mlp": 0.01035845,
      "balance_loss_clip": 1.05051076,
      "balance_loss_mlp": 1.02004421,
      "epoch": 0.3572523673530738,
      "flos": 31175453646720.0,
      "grad_norm": 2.12434809799019,
      "language_loss": 0.64263141,
      "learning_rate": 2.9778837965693154e-06,
      "loss": 0.66423917,
      "num_input_tokens_seen": 127693315,
      "step": 5942,
      "time_per_iteration": 2.5858397483825684
    },
    {
      "auxiliary_loss_clip": 0.0111858,
      "auxiliary_loss_mlp": 0.01043422,
      "balance_loss_clip": 1.04726613,
      "balance_loss_mlp": 1.02814543,
      "epoch": 0.3573124906057418,
      "flos": 15851976203520.0,
      "grad_norm": 1.85212196991812,
      "language_loss": 0.73755795,
      "learning_rate": 2.9775440442370354e-06,
      "loss": 0.75917804,
      "num_input_tokens_seen": 127711570,
      "step": 5943,
      "time_per_iteration": 3.895104169845581
    },
    {
      "auxiliary_loss_clip": 0.01052329,
      "auxiliary_loss_mlp": 0.01008355,
      "balance_loss_clip": 1.02331305,
      "balance_loss_mlp": 1.00629294,
      "epoch": 0.35737261385840974,
      "flos": 60822729118080.0,
      "grad_norm": 0.7942548055395825,
      "language_loss": 0.60739923,
      "learning_rate": 2.9772042548361867e-06,
      "loss": 0.6280061,
      "num_input_tokens_seen": 127772475,
      "step": 5944,
      "time_per_iteration": 3.1618263721466064
    },
    {
      "auxiliary_loss_clip": 0.01106056,
      "auxiliary_loss_mlp": 0.01035445,
      "balance_loss_clip": 1.04664624,
      "balance_loss_mlp": 1.02102685,
      "epoch": 0.3574327371110777,
      "flos": 18843765857280.0,
      "grad_norm": 1.684324793783894,
      "language_loss": 0.72055751,
      "learning_rate": 2.976864428379655e-06,
      "loss": 0.74197257,
      "num_input_tokens_seen": 127790940,
      "step": 5945,
      "time_per_iteration": 2.506046772003174
    },
    {
      "auxiliary_loss_clip": 0.01107717,
      "auxiliary_loss_mlp": 0.00796761,
      "balance_loss_clip": 1.04529107,
      "balance_loss_mlp": 1.02169776,
      "epoch": 0.35749286036374567,
      "flos": 23549679417600.0,
      "grad_norm": 1.651420438452578,
      "language_loss": 0.81069994,
      "learning_rate": 2.976524564880326e-06,
      "loss": 0.82974464,
      "num_input_tokens_seen": 127808275,
      "step": 5946,
      "time_per_iteration": 2.546947956085205
    },
    {
      "auxiliary_loss_clip": 0.01134005,
      "auxiliary_loss_mlp": 0.01045799,
      "balance_loss_clip": 1.05035233,
      "balance_loss_mlp": 1.03076649,
      "epoch": 0.35755298361641363,
      "flos": 21105491581440.0,
      "grad_norm": 1.663055461537085,
      "language_loss": 0.6862157,
      "learning_rate": 2.9761846643510882e-06,
      "loss": 0.70801371,
      "num_input_tokens_seen": 127828840,
      "step": 5947,
      "time_per_iteration": 5.221043348312378
    },
    {
      "auxiliary_loss_clip": 0.01103039,
      "auxiliary_loss_mlp": 0.01036612,
      "balance_loss_clip": 1.04591787,
      "balance_loss_mlp": 1.02244985,
      "epoch": 0.3576131068690816,
      "flos": 19245031666560.0,
      "grad_norm": 1.608889370412184,
      "language_loss": 0.75318611,
      "learning_rate": 2.9758447268048297e-06,
      "loss": 0.77458262,
      "num_input_tokens_seen": 127846240,
      "step": 5948,
      "time_per_iteration": 2.5351667404174805
    },
    {
      "auxiliary_loss_clip": 0.01078117,
      "auxiliary_loss_mlp": 0.01043158,
      "balance_loss_clip": 1.04841256,
      "balance_loss_mlp": 1.02860224,
      "epoch": 0.35767323012174956,
      "flos": 28654703971200.0,
      "grad_norm": 2.1862952878685302,
      "language_loss": 0.70593643,
      "learning_rate": 2.9755047522544415e-06,
      "loss": 0.72714919,
      "num_input_tokens_seen": 127866880,
      "step": 5949,
      "time_per_iteration": 4.073309898376465
    },
    {
      "auxiliary_loss_clip": 0.01108156,
      "auxiliary_loss_mlp": 0.01041243,
      "balance_loss_clip": 1.04757714,
      "balance_loss_mlp": 1.02742708,
      "epoch": 0.35773335337441753,
      "flos": 17085363459840.0,
      "grad_norm": 1.762737307740541,
      "language_loss": 0.76782548,
      "learning_rate": 2.9751647407128154e-06,
      "loss": 0.78931946,
      "num_input_tokens_seen": 127883560,
      "step": 5950,
      "time_per_iteration": 2.506887912750244
    },
    {
      "auxiliary_loss_clip": 0.01123682,
      "auxiliary_loss_mlp": 0.01034032,
      "balance_loss_clip": 1.04840076,
      "balance_loss_mlp": 1.01874924,
      "epoch": 0.35779347662708555,
      "flos": 15888605097600.0,
      "grad_norm": 1.6798555282652263,
      "language_loss": 0.72725999,
      "learning_rate": 2.9748246921928445e-06,
      "loss": 0.74883711,
      "num_input_tokens_seen": 127902330,
      "step": 5951,
      "time_per_iteration": 2.490619659423828
    },
    {
      "auxiliary_loss_clip": 0.01126228,
      "auxiliary_loss_mlp": 0.01042133,
      "balance_loss_clip": 1.05055261,
      "balance_loss_mlp": 1.02694535,
      "epoch": 0.3578535998797535,
      "flos": 28658834035200.0,
      "grad_norm": 3.106985764112784,
      "language_loss": 0.69892228,
      "learning_rate": 2.9744846067074236e-06,
      "loss": 0.72060591,
      "num_input_tokens_seen": 127922325,
      "step": 5952,
      "time_per_iteration": 2.550905704498291
    },
    {
      "auxiliary_loss_clip": 0.01079625,
      "auxiliary_loss_mlp": 0.01039683,
      "balance_loss_clip": 1.04491258,
      "balance_loss_mlp": 1.02442431,
      "epoch": 0.3579137231324215,
      "flos": 37852432076160.0,
      "grad_norm": 2.6441960633158024,
      "language_loss": 0.69395846,
      "learning_rate": 2.974144484269449e-06,
      "loss": 0.71515155,
      "num_input_tokens_seen": 127942635,
      "step": 5953,
      "time_per_iteration": 2.743317127227783
    },
    {
      "auxiliary_loss_clip": 0.01109165,
      "auxiliary_loss_mlp": 0.01031774,
      "balance_loss_clip": 1.04885709,
      "balance_loss_mlp": 1.0176959,
      "epoch": 0.35797384638508944,
      "flos": 22346851656960.0,
      "grad_norm": 1.869943660725663,
      "language_loss": 0.66622543,
      "learning_rate": 2.9738043248918175e-06,
      "loss": 0.68763483,
      "num_input_tokens_seen": 127962520,
      "step": 5954,
      "time_per_iteration": 2.557973861694336
    },
    {
      "auxiliary_loss_clip": 0.01108929,
      "auxiliary_loss_mlp": 0.0103941,
      "balance_loss_clip": 1.05043018,
      "balance_loss_mlp": 1.02588558,
      "epoch": 0.3580339696377574,
      "flos": 13589711775360.0,
      "grad_norm": 1.8132767547870885,
      "language_loss": 0.75097382,
      "learning_rate": 2.9734641285874282e-06,
      "loss": 0.77245724,
      "num_input_tokens_seen": 127981180,
      "step": 5955,
      "time_per_iteration": 2.5424773693084717
    },
    {
      "auxiliary_loss_clip": 0.01109619,
      "auxiliary_loss_mlp": 0.01031468,
      "balance_loss_clip": 1.04732335,
      "balance_loss_mlp": 1.01786041,
      "epoch": 0.3580940928904254,
      "flos": 23768231719680.0,
      "grad_norm": 1.5069628464263818,
      "language_loss": 0.76355857,
      "learning_rate": 2.973123895369182e-06,
      "loss": 0.78496945,
      "num_input_tokens_seen": 127999725,
      "step": 5956,
      "time_per_iteration": 2.5322372913360596
    },
    {
      "auxiliary_loss_clip": 0.0112818,
      "auxiliary_loss_mlp": 0.01040332,
      "balance_loss_clip": 1.04806328,
      "balance_loss_mlp": 1.02612805,
      "epoch": 0.35815421614309334,
      "flos": 19463871277440.0,
      "grad_norm": 1.6745704789546216,
      "language_loss": 0.72686601,
      "learning_rate": 2.9727836252499805e-06,
      "loss": 0.74855113,
      "num_input_tokens_seen": 128018885,
      "step": 5957,
      "time_per_iteration": 2.4783859252929688
    },
    {
      "auxiliary_loss_clip": 0.01108578,
      "auxiliary_loss_mlp": 0.01034732,
      "balance_loss_clip": 1.04988599,
      "balance_loss_mlp": 1.02106452,
      "epoch": 0.3582143393957613,
      "flos": 23368186972800.0,
      "grad_norm": 2.1015258464261333,
      "language_loss": 0.71293038,
      "learning_rate": 2.972443318242726e-06,
      "loss": 0.73436344,
      "num_input_tokens_seen": 128037875,
      "step": 5958,
      "time_per_iteration": 2.576857089996338
    },
    {
      "auxiliary_loss_clip": 0.01094885,
      "auxiliary_loss_mlp": 0.01030354,
      "balance_loss_clip": 1.05042398,
      "balance_loss_mlp": 1.01678848,
      "epoch": 0.35827446264842927,
      "flos": 26323275905280.0,
      "grad_norm": 2.112963702113351,
      "language_loss": 0.88615716,
      "learning_rate": 2.972102974360324e-06,
      "loss": 0.90740955,
      "num_input_tokens_seen": 128056045,
      "step": 5959,
      "time_per_iteration": 2.6197562217712402
    },
    {
      "auxiliary_loss_clip": 0.01128393,
      "auxiliary_loss_mlp": 0.01035809,
      "balance_loss_clip": 1.04779959,
      "balance_loss_mlp": 1.02170038,
      "epoch": 0.35833458590109724,
      "flos": 30446610779520.0,
      "grad_norm": 1.6286782247743752,
      "language_loss": 0.58201844,
      "learning_rate": 2.971762593615679e-06,
      "loss": 0.60366046,
      "num_input_tokens_seen": 128077815,
      "step": 5960,
      "time_per_iteration": 2.5444469451904297
    },
    {
      "auxiliary_loss_clip": 0.01130345,
      "auxiliary_loss_mlp": 0.0104042,
      "balance_loss_clip": 1.04821241,
      "balance_loss_mlp": 1.02528048,
      "epoch": 0.3583947091537652,
      "flos": 14829886702080.0,
      "grad_norm": 2.179227495573977,
      "language_loss": 0.76243794,
      "learning_rate": 2.9714221760216993e-06,
      "loss": 0.78414559,
      "num_input_tokens_seen": 128095460,
      "step": 5961,
      "time_per_iteration": 2.4383132457733154
    },
    {
      "auxiliary_loss_clip": 0.01096037,
      "auxiliary_loss_mlp": 0.01031381,
      "balance_loss_clip": 1.04875302,
      "balance_loss_mlp": 1.01729035,
      "epoch": 0.35845483240643317,
      "flos": 34240644743040.0,
      "grad_norm": 1.8495525673537105,
      "language_loss": 0.70254129,
      "learning_rate": 2.971081721591294e-06,
      "loss": 0.7238155,
      "num_input_tokens_seen": 128118605,
      "step": 5962,
      "time_per_iteration": 2.68628191947937
    },
    {
      "auxiliary_loss_clip": 0.01109136,
      "auxiliary_loss_mlp": 0.0103228,
      "balance_loss_clip": 1.04855895,
      "balance_loss_mlp": 1.01987624,
      "epoch": 0.35851495565910113,
      "flos": 20960089326720.0,
      "grad_norm": 1.6595580280926803,
      "language_loss": 0.74467379,
      "learning_rate": 2.9707412303373716e-06,
      "loss": 0.76608795,
      "num_input_tokens_seen": 128139205,
      "step": 5963,
      "time_per_iteration": 2.534900426864624
    },
    {
      "auxiliary_loss_clip": 0.01129915,
      "auxiliary_loss_mlp": 0.01040177,
      "balance_loss_clip": 1.04987526,
      "balance_loss_mlp": 1.02578831,
      "epoch": 0.35857507891176915,
      "flos": 22309863626880.0,
      "grad_norm": 1.5560822474713363,
      "language_loss": 0.78483427,
      "learning_rate": 2.9704007022728447e-06,
      "loss": 0.80653512,
      "num_input_tokens_seen": 128158765,
      "step": 5964,
      "time_per_iteration": 2.5047199726104736
    },
    {
      "auxiliary_loss_clip": 0.01106314,
      "auxiliary_loss_mlp": 0.01038763,
      "balance_loss_clip": 1.05050397,
      "balance_loss_mlp": 1.0238018,
      "epoch": 0.3586352021644371,
      "flos": 23367863750400.0,
      "grad_norm": 1.7920259861153034,
      "language_loss": 0.66476274,
      "learning_rate": 2.970060137410626e-06,
      "loss": 0.68621349,
      "num_input_tokens_seen": 128177850,
      "step": 5965,
      "time_per_iteration": 2.5438201427459717
    },
    {
      "auxiliary_loss_clip": 0.0112784,
      "auxiliary_loss_mlp": 0.00803733,
      "balance_loss_clip": 1.04651546,
      "balance_loss_mlp": 1.03835809,
      "epoch": 0.3586953254171051,
      "flos": 27849227437440.0,
      "grad_norm": 1.6039219532236417,
      "language_loss": 0.78802943,
      "learning_rate": 2.9697195357636294e-06,
      "loss": 0.80734515,
      "num_input_tokens_seen": 128196925,
      "step": 5966,
      "time_per_iteration": 2.5543289184570312
    },
    {
      "auxiliary_loss_clip": 0.01071378,
      "auxiliary_loss_mlp": 0.01036487,
      "balance_loss_clip": 1.04399395,
      "balance_loss_mlp": 1.02124,
      "epoch": 0.35875544866977305,
      "flos": 19500500171520.0,
      "grad_norm": 2.0685592800433783,
      "language_loss": 0.91245162,
      "learning_rate": 2.9693788973447715e-06,
      "loss": 0.93353021,
      "num_input_tokens_seen": 128213955,
      "step": 5967,
      "time_per_iteration": 2.5927343368530273
    },
    {
      "auxiliary_loss_clip": 0.01091079,
      "auxiliary_loss_mlp": 0.01049605,
      "balance_loss_clip": 1.04512167,
      "balance_loss_mlp": 1.03291547,
      "epoch": 0.358815571922441,
      "flos": 21471134077440.0,
      "grad_norm": 2.0134637818794388,
      "language_loss": 0.80468196,
      "learning_rate": 2.9690382221669682e-06,
      "loss": 0.82608879,
      "num_input_tokens_seen": 128232980,
      "step": 5968,
      "time_per_iteration": 2.575441837310791
    },
    {
      "auxiliary_loss_clip": 0.01109604,
      "auxiliary_loss_mlp": 0.0105013,
      "balance_loss_clip": 1.05200207,
      "balance_loss_mlp": 1.03433466,
      "epoch": 0.358875695175109,
      "flos": 21835411856640.0,
      "grad_norm": 2.146694115353564,
      "language_loss": 0.84511769,
      "learning_rate": 2.9686975102431384e-06,
      "loss": 0.86671507,
      "num_input_tokens_seen": 128252795,
      "step": 5969,
      "time_per_iteration": 2.5405983924865723
    },
    {
      "auxiliary_loss_clip": 0.01085553,
      "auxiliary_loss_mlp": 0.01035776,
      "balance_loss_clip": 1.04539156,
      "balance_loss_mlp": 1.0214529,
      "epoch": 0.35893581842777694,
      "flos": 32011633330560.0,
      "grad_norm": 1.822399821351799,
      "language_loss": 0.72436082,
      "learning_rate": 2.968356761586202e-06,
      "loss": 0.74557412,
      "num_input_tokens_seen": 128273115,
      "step": 5970,
      "time_per_iteration": 2.6523404121398926
    },
    {
      "auxiliary_loss_clip": 0.0110633,
      "auxiliary_loss_mlp": 0.01032926,
      "balance_loss_clip": 1.04624701,
      "balance_loss_mlp": 1.01903772,
      "epoch": 0.3589959416804449,
      "flos": 20485817124480.0,
      "grad_norm": 1.8732158741348652,
      "language_loss": 0.79546452,
      "learning_rate": 2.9680159762090805e-06,
      "loss": 0.81685704,
      "num_input_tokens_seen": 128292220,
      "step": 5971,
      "time_per_iteration": 2.519878625869751
    },
    {
      "auxiliary_loss_clip": 0.01087145,
      "auxiliary_loss_mlp": 0.01040368,
      "balance_loss_clip": 1.0437746,
      "balance_loss_mlp": 1.02513289,
      "epoch": 0.3590560649331129,
      "flos": 16180666583040.0,
      "grad_norm": 1.8008861691436284,
      "language_loss": 0.7844969,
      "learning_rate": 2.967675154124696e-06,
      "loss": 0.80577201,
      "num_input_tokens_seen": 128310305,
      "step": 5972,
      "time_per_iteration": 2.5484859943389893
    },
    {
      "auxiliary_loss_clip": 0.01091062,
      "auxiliary_loss_mlp": 0.01035786,
      "balance_loss_clip": 1.04887617,
      "balance_loss_mlp": 1.02118909,
      "epoch": 0.35911618818578084,
      "flos": 20375391738240.0,
      "grad_norm": 1.7911135521298696,
      "language_loss": 0.81204593,
      "learning_rate": 2.9673342953459722e-06,
      "loss": 0.83331442,
      "num_input_tokens_seen": 128328305,
      "step": 5973,
      "time_per_iteration": 2.552591562271118
    },
    {
      "auxiliary_loss_clip": 0.0103082,
      "auxiliary_loss_mlp": 0.01011699,
      "balance_loss_clip": 1.02151942,
      "balance_loss_mlp": 1.0095408,
      "epoch": 0.3591763114384488,
      "flos": 41236691685120.0,
      "grad_norm": 0.9113429665783997,
      "language_loss": 0.56650209,
      "learning_rate": 2.9669933998858355e-06,
      "loss": 0.58692729,
      "num_input_tokens_seen": 128378380,
      "step": 5974,
      "time_per_iteration": 2.9953691959381104
    },
    {
      "auxiliary_loss_clip": 0.01119213,
      "auxiliary_loss_mlp": 0.01039643,
      "balance_loss_clip": 1.04670691,
      "balance_loss_mlp": 1.02593982,
      "epoch": 0.35923643469111677,
      "flos": 18695454600960.0,
      "grad_norm": 1.7152787712184896,
      "language_loss": 0.68930304,
      "learning_rate": 2.9666524677572114e-06,
      "loss": 0.7108916,
      "num_input_tokens_seen": 128394315,
      "step": 5975,
      "time_per_iteration": 2.4960410594940186
    },
    {
      "auxiliary_loss_clip": 0.0112687,
      "auxiliary_loss_mlp": 0.01038117,
      "balance_loss_clip": 1.04644263,
      "balance_loss_mlp": 1.02460432,
      "epoch": 0.35929655794378473,
      "flos": 25009950931200.0,
      "grad_norm": 1.6564275226500058,
      "language_loss": 0.80328655,
      "learning_rate": 2.96631149897303e-06,
      "loss": 0.82493639,
      "num_input_tokens_seen": 128414515,
      "step": 5976,
      "time_per_iteration": 2.542158365249634
    },
    {
      "auxiliary_loss_clip": 0.01067027,
      "auxiliary_loss_mlp": 0.01041434,
      "balance_loss_clip": 1.04111814,
      "balance_loss_mlp": 1.02592516,
      "epoch": 0.35935668119645275,
      "flos": 14975576265600.0,
      "grad_norm": 1.6912765367734666,
      "language_loss": 0.78863472,
      "learning_rate": 2.9659704935462194e-06,
      "loss": 0.80971932,
      "num_input_tokens_seen": 128430615,
      "step": 5977,
      "time_per_iteration": 2.568429470062256
    },
    {
      "auxiliary_loss_clip": 0.01089446,
      "auxiliary_loss_mlp": 0.01039423,
      "balance_loss_clip": 1.04696155,
      "balance_loss_mlp": 1.02589297,
      "epoch": 0.3594168044491207,
      "flos": 21178138838400.0,
      "grad_norm": 1.7619290646588108,
      "language_loss": 0.80130219,
      "learning_rate": 2.9656294514897102e-06,
      "loss": 0.82259083,
      "num_input_tokens_seen": 128449480,
      "step": 5978,
      "time_per_iteration": 2.5663421154022217
    },
    {
      "auxiliary_loss_clip": 0.01130052,
      "auxiliary_loss_mlp": 0.00792957,
      "balance_loss_clip": 1.04811049,
      "balance_loss_mlp": 1.01494205,
      "epoch": 0.3594769277017887,
      "flos": 27672152365440.0,
      "grad_norm": 1.8666987706442564,
      "language_loss": 0.67573881,
      "learning_rate": 2.965288372816436e-06,
      "loss": 0.69496882,
      "num_input_tokens_seen": 128471465,
      "step": 5979,
      "time_per_iteration": 2.5441973209381104
    },
    {
      "auxiliary_loss_clip": 0.01100597,
      "auxiliary_loss_mlp": 0.01037233,
      "balance_loss_clip": 1.0440768,
      "balance_loss_mlp": 1.02235615,
      "epoch": 0.35953705095445665,
      "flos": 23002328995200.0,
      "grad_norm": 2.2184508632961575,
      "language_loss": 0.66860783,
      "learning_rate": 2.9649472575393296e-06,
      "loss": 0.68998617,
      "num_input_tokens_seen": 128490645,
      "step": 5980,
      "time_per_iteration": 2.55603289604187
    },
    {
      "auxiliary_loss_clip": 0.01105323,
      "auxiliary_loss_mlp": 0.01040223,
      "balance_loss_clip": 1.04613113,
      "balance_loss_mlp": 1.02464223,
      "epoch": 0.3595971742071246,
      "flos": 25513992529920.0,
      "grad_norm": 2.091500102247451,
      "language_loss": 0.71031046,
      "learning_rate": 2.964606105671327e-06,
      "loss": 0.73176599,
      "num_input_tokens_seen": 128510225,
      "step": 5981,
      "time_per_iteration": 2.5656609535217285
    },
    {
      "auxiliary_loss_clip": 0.01106431,
      "auxiliary_loss_mlp": 0.01042036,
      "balance_loss_clip": 1.05072498,
      "balance_loss_mlp": 1.02594233,
      "epoch": 0.3596572974597926,
      "flos": 29862559635840.0,
      "grad_norm": 1.6427620930588318,
      "language_loss": 0.71173775,
      "learning_rate": 2.9642649172253635e-06,
      "loss": 0.73322242,
      "num_input_tokens_seen": 128530195,
      "step": 5982,
      "time_per_iteration": 3.9712226390838623
    },
    {
      "auxiliary_loss_clip": 0.01109708,
      "auxiliary_loss_mlp": 0.01048402,
      "balance_loss_clip": 1.04489136,
      "balance_loss_mlp": 1.03334582,
      "epoch": 0.35971742071246054,
      "flos": 23112538899840.0,
      "grad_norm": 1.5844041976940482,
      "language_loss": 0.75634325,
      "learning_rate": 2.9639236922143786e-06,
      "loss": 0.77792442,
      "num_input_tokens_seen": 128549990,
      "step": 5983,
      "time_per_iteration": 2.515801429748535
    },
    {
      "auxiliary_loss_clip": 0.01136459,
      "auxiliary_loss_mlp": 0.01043659,
      "balance_loss_clip": 1.05044293,
      "balance_loss_mlp": 1.02745819,
      "epoch": 0.3597775439651285,
      "flos": 16725359399040.0,
      "grad_norm": 1.6605539248433725,
      "language_loss": 0.76364267,
      "learning_rate": 2.96358243065131e-06,
      "loss": 0.7854439,
      "num_input_tokens_seen": 128567925,
      "step": 5984,
      "time_per_iteration": 2.4650187492370605
    },
    {
      "auxiliary_loss_clip": 0.01117923,
      "auxiliary_loss_mlp": 0.00795274,
      "balance_loss_clip": 1.04816544,
      "balance_loss_mlp": 1.01884317,
      "epoch": 0.3598376672177965,
      "flos": 19719483436800.0,
      "grad_norm": 1.7151553068572867,
      "language_loss": 0.86074632,
      "learning_rate": 2.9632411325490993e-06,
      "loss": 0.87987828,
      "num_input_tokens_seen": 128585655,
      "step": 5985,
      "time_per_iteration": 5.33185338973999
    },
    {
      "auxiliary_loss_clip": 0.01117258,
      "auxiliary_loss_mlp": 0.0104128,
      "balance_loss_clip": 1.04771686,
      "balance_loss_mlp": 1.02504361,
      "epoch": 0.35989779047046444,
      "flos": 17311529445120.0,
      "grad_norm": 1.4029024307346283,
      "language_loss": 0.72453976,
      "learning_rate": 2.9628997979206884e-06,
      "loss": 0.7461251,
      "num_input_tokens_seen": 128604820,
      "step": 5986,
      "time_per_iteration": 2.515920639038086
    },
    {
      "auxiliary_loss_clip": 0.01099703,
      "auxiliary_loss_mlp": 0.01040654,
      "balance_loss_clip": 1.04463291,
      "balance_loss_mlp": 1.02558589,
      "epoch": 0.3599579137231324,
      "flos": 22711237176960.0,
      "grad_norm": 2.056530305797743,
      "language_loss": 0.73960054,
      "learning_rate": 2.9625584267790204e-06,
      "loss": 0.76100415,
      "num_input_tokens_seen": 128623070,
      "step": 5987,
      "time_per_iteration": 2.60262131690979
    },
    {
      "auxiliary_loss_clip": 0.01136935,
      "auxiliary_loss_mlp": 0.01036639,
      "balance_loss_clip": 1.05197752,
      "balance_loss_mlp": 1.02118945,
      "epoch": 0.36001803697580037,
      "flos": 20959873845120.0,
      "grad_norm": 2.0265770174161584,
      "language_loss": 0.69553924,
      "learning_rate": 2.9622170191370404e-06,
      "loss": 0.71727502,
      "num_input_tokens_seen": 128642430,
      "step": 5988,
      "time_per_iteration": 3.848695993423462
    },
    {
      "auxiliary_loss_clip": 0.01124841,
      "auxiliary_loss_mlp": 0.0104116,
      "balance_loss_clip": 1.05010962,
      "balance_loss_mlp": 1.02573478,
      "epoch": 0.36007816022846834,
      "flos": 20485565729280.0,
      "grad_norm": 3.9584616748660078,
      "language_loss": 0.72867572,
      "learning_rate": 2.9618755750076953e-06,
      "loss": 0.75033575,
      "num_input_tokens_seen": 128661285,
      "step": 5989,
      "time_per_iteration": 2.5136044025421143
    },
    {
      "auxiliary_loss_clip": 0.01087466,
      "auxiliary_loss_mlp": 0.01037738,
      "balance_loss_clip": 1.04378819,
      "balance_loss_mlp": 1.02240765,
      "epoch": 0.36013828348113636,
      "flos": 28001237794560.0,
      "grad_norm": 1.5512753552241145,
      "language_loss": 0.80010039,
      "learning_rate": 2.961534094403931e-06,
      "loss": 0.82135248,
      "num_input_tokens_seen": 128682210,
      "step": 5990,
      "time_per_iteration": 2.6140153408050537
    },
    {
      "auxiliary_loss_clip": 0.01121678,
      "auxiliary_loss_mlp": 0.01035383,
      "balance_loss_clip": 1.04902291,
      "balance_loss_mlp": 1.02009451,
      "epoch": 0.3601984067338043,
      "flos": 20082181017600.0,
      "grad_norm": 1.7306091506534582,
      "language_loss": 0.83772665,
      "learning_rate": 2.961192577338698e-06,
      "loss": 0.85929728,
      "num_input_tokens_seen": 128700445,
      "step": 5991,
      "time_per_iteration": 2.5591483116149902
    },
    {
      "auxiliary_loss_clip": 0.01114515,
      "auxiliary_loss_mlp": 0.01042272,
      "balance_loss_clip": 1.05248928,
      "balance_loss_mlp": 1.026703,
      "epoch": 0.3602585299864723,
      "flos": 18617599872000.0,
      "grad_norm": 3.536875879955241,
      "language_loss": 0.75711954,
      "learning_rate": 2.9608510238249463e-06,
      "loss": 0.77868742,
      "num_input_tokens_seen": 128716855,
      "step": 5992,
      "time_per_iteration": 2.5057694911956787
    },
    {
      "auxiliary_loss_clip": 0.01132042,
      "auxiliary_loss_mlp": 0.01038664,
      "balance_loss_clip": 1.04953778,
      "balance_loss_mlp": 1.02282083,
      "epoch": 0.36031865323914025,
      "flos": 19573003774080.0,
      "grad_norm": 1.9265308133832486,
      "language_loss": 0.77588069,
      "learning_rate": 2.960509433875627e-06,
      "loss": 0.79758775,
      "num_input_tokens_seen": 128735835,
      "step": 5993,
      "time_per_iteration": 2.493623971939087
    },
    {
      "auxiliary_loss_clip": 0.01107269,
      "auxiliary_loss_mlp": 0.01049436,
      "balance_loss_clip": 1.04573274,
      "balance_loss_mlp": 1.03203154,
      "epoch": 0.3603787764918082,
      "flos": 17490615678720.0,
      "grad_norm": 2.0472130161696316,
      "language_loss": 0.74881154,
      "learning_rate": 2.9601678075036943e-06,
      "loss": 0.77037859,
      "num_input_tokens_seen": 128752465,
      "step": 5994,
      "time_per_iteration": 2.5470950603485107
    },
    {
      "auxiliary_loss_clip": 0.01095953,
      "auxiliary_loss_mlp": 0.01037361,
      "balance_loss_clip": 1.05219233,
      "balance_loss_mlp": 1.02226257,
      "epoch": 0.3604388997444762,
      "flos": 15523393564800.0,
      "grad_norm": 1.905682320088245,
      "language_loss": 0.68991947,
      "learning_rate": 2.9598261447221024e-06,
      "loss": 0.71125263,
      "num_input_tokens_seen": 128770865,
      "step": 5995,
      "time_per_iteration": 2.5763652324676514
    },
    {
      "auxiliary_loss_clip": 0.0110508,
      "auxiliary_loss_mlp": 0.01042683,
      "balance_loss_clip": 1.0459516,
      "balance_loss_mlp": 1.02623248,
      "epoch": 0.36049902299714415,
      "flos": 17310883000320.0,
      "grad_norm": 1.9305798276998993,
      "language_loss": 0.83027709,
      "learning_rate": 2.9594844455438057e-06,
      "loss": 0.85175472,
      "num_input_tokens_seen": 128789730,
      "step": 5996,
      "time_per_iteration": 2.522080183029175
    },
    {
      "auxiliary_loss_clip": 0.01131943,
      "auxiliary_loss_mlp": 0.01037147,
      "balance_loss_clip": 1.04930615,
      "balance_loss_mlp": 1.02166772,
      "epoch": 0.3605591462498121,
      "flos": 17056025026560.0,
      "grad_norm": 1.6467544033834258,
      "language_loss": 0.73412925,
      "learning_rate": 2.959142709981763e-06,
      "loss": 0.75582016,
      "num_input_tokens_seen": 128806610,
      "step": 5997,
      "time_per_iteration": 2.4631779193878174
    },
    {
      "auxiliary_loss_clip": 0.01126076,
      "auxiliary_loss_mlp": 0.01035183,
      "balance_loss_clip": 1.05029583,
      "balance_loss_mlp": 1.02092528,
      "epoch": 0.3606192695024801,
      "flos": 16836862193280.0,
      "grad_norm": 2.356465110256469,
      "language_loss": 0.69311535,
      "learning_rate": 2.9588009380489337e-06,
      "loss": 0.71472794,
      "num_input_tokens_seen": 128824830,
      "step": 5998,
      "time_per_iteration": 2.492713689804077
    },
    {
      "auxiliary_loss_clip": 0.01082801,
      "auxiliary_loss_mlp": 0.01041155,
      "balance_loss_clip": 1.04466724,
      "balance_loss_mlp": 1.02391696,
      "epoch": 0.36067939275514804,
      "flos": 12129655743360.0,
      "grad_norm": 2.175082646169121,
      "language_loss": 0.76987231,
      "learning_rate": 2.9584591297582758e-06,
      "loss": 0.79111183,
      "num_input_tokens_seen": 128838170,
      "step": 5999,
      "time_per_iteration": 2.5499494075775146
    },
    {
      "auxiliary_loss_clip": 0.01096531,
      "auxiliary_loss_mlp": 0.01042068,
      "balance_loss_clip": 1.04950011,
      "balance_loss_mlp": 1.02746511,
      "epoch": 0.360739516007816,
      "flos": 18041449720320.0,
      "grad_norm": 1.860556870037676,
      "language_loss": 0.78568465,
      "learning_rate": 2.9581172851227516e-06,
      "loss": 0.80707061,
      "num_input_tokens_seen": 128855625,
      "step": 6000,
      "time_per_iteration": 2.5279386043548584
    },
    {
      "auxiliary_loss_clip": 0.01101113,
      "auxiliary_loss_mlp": 0.0103741,
      "balance_loss_clip": 1.05082798,
      "balance_loss_mlp": 1.02217555,
      "epoch": 0.360799639260484,
      "flos": 18549800951040.0,
      "grad_norm": 1.6010085408034473,
      "language_loss": 0.78378236,
      "learning_rate": 2.9577754041553243e-06,
      "loss": 0.80516756,
      "num_input_tokens_seen": 128873540,
      "step": 6001,
      "time_per_iteration": 2.577725648880005
    },
    {
      "auxiliary_loss_clip": 0.01130795,
      "auxiliary_loss_mlp": 0.0079383,
      "balance_loss_clip": 1.04910469,
      "balance_loss_mlp": 1.01483452,
      "epoch": 0.36085976251315194,
      "flos": 19682028529920.0,
      "grad_norm": 2.066788542551071,
      "language_loss": 0.83585441,
      "learning_rate": 2.9574334868689575e-06,
      "loss": 0.85510063,
      "num_input_tokens_seen": 128889925,
      "step": 6002,
      "time_per_iteration": 2.4854626655578613
    },
    {
      "auxiliary_loss_clip": 0.01103822,
      "auxiliary_loss_mlp": 0.01035538,
      "balance_loss_clip": 1.04623854,
      "balance_loss_mlp": 1.02123284,
      "epoch": 0.3609198857658199,
      "flos": 24198943703040.0,
      "grad_norm": 2.2832701920978464,
      "language_loss": 0.91142404,
      "learning_rate": 2.9570915332766165e-06,
      "loss": 0.93281764,
      "num_input_tokens_seen": 128906890,
      "step": 6003,
      "time_per_iteration": 2.550307273864746
    },
    {
      "auxiliary_loss_clip": 0.01031238,
      "auxiliary_loss_mlp": 0.01003173,
      "balance_loss_clip": 1.03934705,
      "balance_loss_mlp": 1.0010035,
      "epoch": 0.3609800090184879,
      "flos": 57115995160320.0,
      "grad_norm": 0.8685928391510435,
      "language_loss": 0.53384614,
      "learning_rate": 2.9567495433912693e-06,
      "loss": 0.55419028,
      "num_input_tokens_seen": 128965940,
      "step": 6004,
      "time_per_iteration": 3.066077947616577
    },
    {
      "auxiliary_loss_clip": 0.01112677,
      "auxiliary_loss_mlp": 0.00791945,
      "balance_loss_clip": 1.04862595,
      "balance_loss_mlp": 1.01112723,
      "epoch": 0.3610401322711559,
      "flos": 20811239366400.0,
      "grad_norm": 1.721171306582842,
      "language_loss": 0.77763391,
      "learning_rate": 2.956407517225883e-06,
      "loss": 0.79668009,
      "num_input_tokens_seen": 128985835,
      "step": 6005,
      "time_per_iteration": 2.5557022094726562
    },
    {
      "auxiliary_loss_clip": 0.01114566,
      "auxiliary_loss_mlp": 0.01047545,
      "balance_loss_clip": 1.04751492,
      "balance_loss_mlp": 1.03217888,
      "epoch": 0.36110025552382385,
      "flos": 13699167494400.0,
      "grad_norm": 2.0635085807579974,
      "language_loss": 0.79108024,
      "learning_rate": 2.956065454793429e-06,
      "loss": 0.81270128,
      "num_input_tokens_seen": 129003120,
      "step": 6006,
      "time_per_iteration": 2.5151889324188232
    },
    {
      "auxiliary_loss_clip": 0.01137226,
      "auxiliary_loss_mlp": 0.0103935,
      "balance_loss_clip": 1.05173552,
      "balance_loss_mlp": 1.022053,
      "epoch": 0.3611603787764918,
      "flos": 22455014486400.0,
      "grad_norm": 1.7596762562214843,
      "language_loss": 0.84508562,
      "learning_rate": 2.955723356106876e-06,
      "loss": 0.86685139,
      "num_input_tokens_seen": 129021645,
      "step": 6007,
      "time_per_iteration": 2.4855101108551025
    },
    {
      "auxiliary_loss_clip": 0.0111788,
      "auxiliary_loss_mlp": 0.01038428,
      "balance_loss_clip": 1.04972982,
      "balance_loss_mlp": 1.02153635,
      "epoch": 0.3612205020291598,
      "flos": 20886651970560.0,
      "grad_norm": 1.9526049356140924,
      "language_loss": 0.72146356,
      "learning_rate": 2.955381221179198e-06,
      "loss": 0.74302661,
      "num_input_tokens_seen": 129038375,
      "step": 6008,
      "time_per_iteration": 2.521165370941162
    },
    {
      "auxiliary_loss_clip": 0.01118094,
      "auxiliary_loss_mlp": 0.01037201,
      "balance_loss_clip": 1.04504991,
      "balance_loss_mlp": 1.02219224,
      "epoch": 0.36128062528182775,
      "flos": 15741981780480.0,
      "grad_norm": 1.8763814119767641,
      "language_loss": 0.83018947,
      "learning_rate": 2.955039050023368e-06,
      "loss": 0.85174245,
      "num_input_tokens_seen": 129056235,
      "step": 6009,
      "time_per_iteration": 2.589008331298828
    },
    {
      "auxiliary_loss_clip": 0.0110336,
      "auxiliary_loss_mlp": 0.01045649,
      "balance_loss_clip": 1.04914606,
      "balance_loss_mlp": 1.03040195,
      "epoch": 0.3613407485344957,
      "flos": 16764502245120.0,
      "grad_norm": 1.8040018740200463,
      "language_loss": 0.76398462,
      "learning_rate": 2.954696842652362e-06,
      "loss": 0.78547472,
      "num_input_tokens_seen": 129072405,
      "step": 6010,
      "time_per_iteration": 2.527660846710205
    },
    {
      "auxiliary_loss_clip": 0.01105414,
      "auxiliary_loss_mlp": 0.0104053,
      "balance_loss_clip": 1.04962349,
      "balance_loss_mlp": 1.02574825,
      "epoch": 0.3614008717871637,
      "flos": 20371189847040.0,
      "grad_norm": 1.5026900150312068,
      "language_loss": 0.8299917,
      "learning_rate": 2.9543545990791554e-06,
      "loss": 0.85145116,
      "num_input_tokens_seen": 129090225,
      "step": 6011,
      "time_per_iteration": 2.562345027923584
    },
    {
      "auxiliary_loss_clip": 0.01139261,
      "auxiliary_loss_mlp": 0.01041161,
      "balance_loss_clip": 1.05095625,
      "balance_loss_mlp": 1.02478123,
      "epoch": 0.36146099503983165,
      "flos": 22776665800320.0,
      "grad_norm": 3.2433870495854227,
      "language_loss": 0.62401962,
      "learning_rate": 2.954012319316727e-06,
      "loss": 0.64582384,
      "num_input_tokens_seen": 129107685,
      "step": 6012,
      "time_per_iteration": 2.4611761569976807
    },
    {
      "auxiliary_loss_clip": 0.01108331,
      "auxiliary_loss_mlp": 0.010409,
      "balance_loss_clip": 1.04890013,
      "balance_loss_mlp": 1.0258199,
      "epoch": 0.3615211182924996,
      "flos": 22996654646400.0,
      "grad_norm": 1.7079969339152121,
      "language_loss": 0.83622134,
      "learning_rate": 2.9536700033780565e-06,
      "loss": 0.85771364,
      "num_input_tokens_seen": 129125315,
      "step": 6013,
      "time_per_iteration": 2.5608584880828857
    },
    {
      "auxiliary_loss_clip": 0.01131872,
      "auxiliary_loss_mlp": 0.01040218,
      "balance_loss_clip": 1.04759896,
      "balance_loss_mlp": 1.02401161,
      "epoch": 0.3615812415451676,
      "flos": 16648079287680.0,
      "grad_norm": 1.6688523653116913,
      "language_loss": 0.91484773,
      "learning_rate": 2.9533276512761228e-06,
      "loss": 0.93656868,
      "num_input_tokens_seen": 129141600,
      "step": 6014,
      "time_per_iteration": 2.430281639099121
    },
    {
      "auxiliary_loss_clip": 0.01129588,
      "auxiliary_loss_mlp": 0.01040714,
      "balance_loss_clip": 1.04605925,
      "balance_loss_mlp": 1.02504969,
      "epoch": 0.36164136479783554,
      "flos": 21320093387520.0,
      "grad_norm": 2.3134108439473002,
      "language_loss": 0.73833084,
      "learning_rate": 2.95298526302391e-06,
      "loss": 0.76003385,
      "num_input_tokens_seen": 129160665,
      "step": 6015,
      "time_per_iteration": 2.4982917308807373
    },
    {
      "auxiliary_loss_clip": 0.01054677,
      "auxiliary_loss_mlp": 0.01045262,
      "balance_loss_clip": 1.04395533,
      "balance_loss_mlp": 1.02741659,
      "epoch": 0.3617014880505035,
      "flos": 24169569356160.0,
      "grad_norm": 1.6534486658128495,
      "language_loss": 0.64854622,
      "learning_rate": 2.9526428386344e-06,
      "loss": 0.66954553,
      "num_input_tokens_seen": 129179220,
      "step": 6016,
      "time_per_iteration": 2.651958465576172
    },
    {
      "auxiliary_loss_clip": 0.01124713,
      "auxiliary_loss_mlp": 0.01040113,
      "balance_loss_clip": 1.04994893,
      "balance_loss_mlp": 1.02307761,
      "epoch": 0.3617616113031715,
      "flos": 39014824101120.0,
      "grad_norm": 1.8047019618565072,
      "language_loss": 0.71663344,
      "learning_rate": 2.9523003781205785e-06,
      "loss": 0.73828167,
      "num_input_tokens_seen": 129200385,
      "step": 6017,
      "time_per_iteration": 2.642540693283081
    },
    {
      "auxiliary_loss_clip": 0.01122275,
      "auxiliary_loss_mlp": 0.01043508,
      "balance_loss_clip": 1.04662895,
      "balance_loss_mlp": 1.02787948,
      "epoch": 0.3618217345558395,
      "flos": 12130840892160.0,
      "grad_norm": 1.801718550190302,
      "language_loss": 0.73705584,
      "learning_rate": 2.9519578814954307e-06,
      "loss": 0.75871366,
      "num_input_tokens_seen": 129217395,
      "step": 6018,
      "time_per_iteration": 2.4781253337860107
    },
    {
      "auxiliary_loss_clip": 0.01088521,
      "auxiliary_loss_mlp": 0.0103668,
      "balance_loss_clip": 1.05030024,
      "balance_loss_mlp": 1.02080142,
      "epoch": 0.36188185780850746,
      "flos": 24935005203840.0,
      "grad_norm": 1.995826852710398,
      "language_loss": 0.69451702,
      "learning_rate": 2.9516153487719448e-06,
      "loss": 0.71576905,
      "num_input_tokens_seen": 129238940,
      "step": 6019,
      "time_per_iteration": 2.611595869064331
    },
    {
      "auxiliary_loss_clip": 0.01113329,
      "auxiliary_loss_mlp": 0.01038067,
      "balance_loss_clip": 1.04730058,
      "balance_loss_mlp": 1.02160454,
      "epoch": 0.3619419810611754,
      "flos": 20958832350720.0,
      "grad_norm": 6.341172316559218,
      "language_loss": 0.76414269,
      "learning_rate": 2.95127277996311e-06,
      "loss": 0.78565669,
      "num_input_tokens_seen": 129258240,
      "step": 6020,
      "time_per_iteration": 2.5248849391937256
    },
    {
      "auxiliary_loss_clip": 0.01128196,
      "auxiliary_loss_mlp": 0.01040535,
      "balance_loss_clip": 1.05107415,
      "balance_loss_mlp": 1.02409613,
      "epoch": 0.3620021043138434,
      "flos": 22528882805760.0,
      "grad_norm": 1.9108090161154,
      "language_loss": 0.73680657,
      "learning_rate": 2.9509301750819156e-06,
      "loss": 0.7584939,
      "num_input_tokens_seen": 129279040,
      "step": 6021,
      "time_per_iteration": 3.899514675140381
    },
    {
      "auxiliary_loss_clip": 0.01095906,
      "auxiliary_loss_mlp": 0.01038029,
      "balance_loss_clip": 1.04940856,
      "balance_loss_mlp": 1.02401614,
      "epoch": 0.36206222756651135,
      "flos": 15596687266560.0,
      "grad_norm": 2.0714076220474538,
      "language_loss": 0.80991471,
      "learning_rate": 2.9505875341413533e-06,
      "loss": 0.83125401,
      "num_input_tokens_seen": 129295415,
      "step": 6022,
      "time_per_iteration": 2.547633647918701
    },
    {
      "auxiliary_loss_clip": 0.01119297,
      "auxiliary_loss_mlp": 0.01039596,
      "balance_loss_clip": 1.04949975,
      "balance_loss_mlp": 1.02507615,
      "epoch": 0.3621223508191793,
      "flos": 23587170238080.0,
      "grad_norm": 1.6581115659754537,
      "language_loss": 0.81454897,
      "learning_rate": 2.950244857154417e-06,
      "loss": 0.83613789,
      "num_input_tokens_seen": 129312620,
      "step": 6023,
      "time_per_iteration": 2.513324022293091
    },
    {
      "auxiliary_loss_clip": 0.01108311,
      "auxiliary_loss_mlp": 0.0103742,
      "balance_loss_clip": 1.04768825,
      "balance_loss_mlp": 1.02187514,
      "epoch": 0.3621824740718473,
      "flos": 22309899540480.0,
      "grad_norm": 1.597737590737222,
      "language_loss": 0.79759592,
      "learning_rate": 2.9499021441341e-06,
      "loss": 0.81905317,
      "num_input_tokens_seen": 129331825,
      "step": 6024,
      "time_per_iteration": 5.382205247879028
    },
    {
      "auxiliary_loss_clip": 0.01096978,
      "auxiliary_loss_mlp": 0.0104278,
      "balance_loss_clip": 1.04573464,
      "balance_loss_mlp": 1.02632928,
      "epoch": 0.36224259732451525,
      "flos": 16763640318720.0,
      "grad_norm": 1.9427386743944617,
      "language_loss": 0.74653208,
      "learning_rate": 2.9495593950933997e-06,
      "loss": 0.76792967,
      "num_input_tokens_seen": 129350400,
      "step": 6025,
      "time_per_iteration": 2.5144035816192627
    },
    {
      "auxiliary_loss_clip": 0.01119483,
      "auxiliary_loss_mlp": 0.00790726,
      "balance_loss_clip": 1.04838967,
      "balance_loss_mlp": 1.01212335,
      "epoch": 0.3623027205771832,
      "flos": 23149742411520.0,
      "grad_norm": 1.6523297486278674,
      "language_loss": 0.72563767,
      "learning_rate": 2.9492166100453107e-06,
      "loss": 0.74473977,
      "num_input_tokens_seen": 129371155,
      "step": 6026,
      "time_per_iteration": 2.5267932415008545
    },
    {
      "auxiliary_loss_clip": 0.01128783,
      "auxiliary_loss_mlp": 0.0104546,
      "balance_loss_clip": 1.05154777,
      "balance_loss_mlp": 1.02983201,
      "epoch": 0.3623628438298512,
      "flos": 28549162834560.0,
      "grad_norm": 2.0215860745139813,
      "language_loss": 0.78965998,
      "learning_rate": 2.948873789002833e-06,
      "loss": 0.81140238,
      "num_input_tokens_seen": 129391230,
      "step": 6027,
      "time_per_iteration": 3.980532646179199
    },
    {
      "auxiliary_loss_clip": 0.01116704,
      "auxiliary_loss_mlp": 0.01039595,
      "balance_loss_clip": 1.04868722,
      "balance_loss_mlp": 1.02302456,
      "epoch": 0.36242296708251914,
      "flos": 25484941405440.0,
      "grad_norm": 1.710193607645512,
      "language_loss": 0.67627615,
      "learning_rate": 2.9485309319789667e-06,
      "loss": 0.69783914,
      "num_input_tokens_seen": 129410065,
      "step": 6028,
      "time_per_iteration": 2.571261167526245
    },
    {
      "auxiliary_loss_clip": 0.01094025,
      "auxiliary_loss_mlp": 0.01036572,
      "balance_loss_clip": 1.04920936,
      "balance_loss_mlp": 1.02154565,
      "epoch": 0.3624830903351871,
      "flos": 16290373697280.0,
      "grad_norm": 1.7136521384275445,
      "language_loss": 0.85163236,
      "learning_rate": 2.9481880389867117e-06,
      "loss": 0.87293828,
      "num_input_tokens_seen": 129428655,
      "step": 6029,
      "time_per_iteration": 2.5734145641326904
    },
    {
      "auxiliary_loss_clip": 0.01095779,
      "auxiliary_loss_mlp": 0.01041319,
      "balance_loss_clip": 1.04746819,
      "balance_loss_mlp": 1.02632213,
      "epoch": 0.36254321358785513,
      "flos": 18296307694080.0,
      "grad_norm": 1.6347458162178703,
      "language_loss": 0.72741926,
      "learning_rate": 2.9478451100390714e-06,
      "loss": 0.74879026,
      "num_input_tokens_seen": 129447845,
      "step": 6030,
      "time_per_iteration": 2.5389468669891357
    },
    {
      "auxiliary_loss_clip": 0.01108675,
      "auxiliary_loss_mlp": 0.01042018,
      "balance_loss_clip": 1.04760242,
      "balance_loss_mlp": 1.02474427,
      "epoch": 0.3626033368405231,
      "flos": 14865294533760.0,
      "grad_norm": 2.1067507796916742,
      "language_loss": 0.74437666,
      "learning_rate": 2.94750214514905e-06,
      "loss": 0.76588356,
      "num_input_tokens_seen": 129463275,
      "step": 6031,
      "time_per_iteration": 2.4884889125823975
    },
    {
      "auxiliary_loss_clip": 0.01088125,
      "auxiliary_loss_mlp": 0.0104254,
      "balance_loss_clip": 1.04448438,
      "balance_loss_mlp": 1.026793,
      "epoch": 0.36266346009319106,
      "flos": 22306595489280.0,
      "grad_norm": 1.9524897222406512,
      "language_loss": 0.73075974,
      "learning_rate": 2.9471591443296516e-06,
      "loss": 0.75206637,
      "num_input_tokens_seen": 129483205,
      "step": 6032,
      "time_per_iteration": 2.5806376934051514
    },
    {
      "auxiliary_loss_clip": 0.01083001,
      "auxiliary_loss_mlp": 0.01044323,
      "balance_loss_clip": 1.04523325,
      "balance_loss_mlp": 1.02927852,
      "epoch": 0.362723583345859,
      "flos": 18222331633920.0,
      "grad_norm": 1.7831734646689246,
      "language_loss": 0.77924532,
      "learning_rate": 2.946816107593884e-06,
      "loss": 0.80051863,
      "num_input_tokens_seen": 129499885,
      "step": 6033,
      "time_per_iteration": 2.5426104068756104
    },
    {
      "auxiliary_loss_clip": 0.01018144,
      "auxiliary_loss_mlp": 0.01015884,
      "balance_loss_clip": 1.03933644,
      "balance_loss_mlp": 1.01364255,
      "epoch": 0.362783706598527,
      "flos": 68499174458880.0,
      "grad_norm": 0.787252805703643,
      "language_loss": 0.64813119,
      "learning_rate": 2.9464730349547547e-06,
      "loss": 0.66847152,
      "num_input_tokens_seen": 129561885,
      "step": 6034,
      "time_per_iteration": 3.234199285507202
    },
    {
      "auxiliary_loss_clip": 0.01114631,
      "auxiliary_loss_mlp": 0.0103987,
      "balance_loss_clip": 1.04740274,
      "balance_loss_mlp": 1.02443266,
      "epoch": 0.36284382985119495,
      "flos": 26576589594240.0,
      "grad_norm": 1.5021956262301723,
      "language_loss": 0.89668798,
      "learning_rate": 2.946129926425273e-06,
      "loss": 0.91823304,
      "num_input_tokens_seen": 129582325,
      "step": 6035,
      "time_per_iteration": 2.550851821899414
    },
    {
      "auxiliary_loss_clip": 0.01106068,
      "auxiliary_loss_mlp": 0.01040541,
      "balance_loss_clip": 1.04616857,
      "balance_loss_mlp": 1.02434087,
      "epoch": 0.3629039531038629,
      "flos": 20156767608960.0,
      "grad_norm": 1.7262617285996955,
      "language_loss": 0.7406894,
      "learning_rate": 2.9457867820184496e-06,
      "loss": 0.76215553,
      "num_input_tokens_seen": 129600350,
      "step": 6036,
      "time_per_iteration": 2.52866792678833
    },
    {
      "auxiliary_loss_clip": 0.01111441,
      "auxiliary_loss_mlp": 0.01034327,
      "balance_loss_clip": 1.04447031,
      "balance_loss_mlp": 1.01912761,
      "epoch": 0.3629640763565309,
      "flos": 18625716345600.0,
      "grad_norm": 1.8352612102065866,
      "language_loss": 0.75705439,
      "learning_rate": 2.945443601747297e-06,
      "loss": 0.77851212,
      "num_input_tokens_seen": 129618425,
      "step": 6037,
      "time_per_iteration": 2.4933624267578125
    },
    {
      "auxiliary_loss_clip": 0.01111583,
      "auxiliary_loss_mlp": 0.01054806,
      "balance_loss_clip": 1.04466653,
      "balance_loss_mlp": 1.03761649,
      "epoch": 0.36302419960919885,
      "flos": 19571459489280.0,
      "grad_norm": 1.5603854449156658,
      "language_loss": 0.78710675,
      "learning_rate": 2.945100385624828e-06,
      "loss": 0.8087706,
      "num_input_tokens_seen": 129636750,
      "step": 6038,
      "time_per_iteration": 2.4962406158447266
    },
    {
      "auxiliary_loss_clip": 0.01040023,
      "auxiliary_loss_mlp": 0.01006832,
      "balance_loss_clip": 1.02537048,
      "balance_loss_mlp": 1.00506783,
      "epoch": 0.3630843228618668,
      "flos": 63797606444160.0,
      "grad_norm": 0.8287535598688474,
      "language_loss": 0.6344533,
      "learning_rate": 2.9447571336640573e-06,
      "loss": 0.65492183,
      "num_input_tokens_seen": 129699030,
      "step": 6039,
      "time_per_iteration": 3.155261516571045
    },
    {
      "auxiliary_loss_clip": 0.01102014,
      "auxiliary_loss_mlp": 0.01044263,
      "balance_loss_clip": 1.04502916,
      "balance_loss_mlp": 1.02794278,
      "epoch": 0.3631444461145348,
      "flos": 21835160461440.0,
      "grad_norm": 2.474446531606057,
      "language_loss": 0.7109338,
      "learning_rate": 2.944413845878002e-06,
      "loss": 0.73239654,
      "num_input_tokens_seen": 129717135,
      "step": 6040,
      "time_per_iteration": 2.5327258110046387
    },
    {
      "auxiliary_loss_clip": 0.0112471,
      "auxiliary_loss_mlp": 0.01039063,
      "balance_loss_clip": 1.04908967,
      "balance_loss_mlp": 1.02370894,
      "epoch": 0.36320456936720275,
      "flos": 21722041555200.0,
      "grad_norm": 1.7116113193266367,
      "language_loss": 0.81228429,
      "learning_rate": 2.9440705222796783e-06,
      "loss": 0.83392203,
      "num_input_tokens_seen": 129735940,
      "step": 6041,
      "time_per_iteration": 2.492807388305664
    },
    {
      "auxiliary_loss_clip": 0.01107398,
      "auxiliary_loss_mlp": 0.01035805,
      "balance_loss_clip": 1.04230332,
      "balance_loss_mlp": 1.0197711,
      "epoch": 0.3632646926198707,
      "flos": 17019072910080.0,
      "grad_norm": 2.1902970298941433,
      "language_loss": 0.83965778,
      "learning_rate": 2.943727162882107e-06,
      "loss": 0.86108977,
      "num_input_tokens_seen": 129752790,
      "step": 6042,
      "time_per_iteration": 2.5199177265167236
    },
    {
      "auxiliary_loss_clip": 0.01113122,
      "auxiliary_loss_mlp": 0.01045505,
      "balance_loss_clip": 1.04703093,
      "balance_loss_mlp": 1.03117049,
      "epoch": 0.36332481587253873,
      "flos": 23331163029120.0,
      "grad_norm": 1.5523653131965562,
      "language_loss": 0.78184271,
      "learning_rate": 2.9433837676983064e-06,
      "loss": 0.80342901,
      "num_input_tokens_seen": 129773655,
      "step": 6043,
      "time_per_iteration": 2.53847336769104
    },
    {
      "auxiliary_loss_clip": 0.01098039,
      "auxiliary_loss_mlp": 0.01038932,
      "balance_loss_clip": 1.04671955,
      "balance_loss_mlp": 1.0232861,
      "epoch": 0.3633849391252067,
      "flos": 10743539857920.0,
      "grad_norm": 2.917747117485441,
      "language_loss": 0.65318924,
      "learning_rate": 2.943040336741298e-06,
      "loss": 0.674559,
      "num_input_tokens_seen": 129791605,
      "step": 6044,
      "time_per_iteration": 2.5502843856811523
    },
    {
      "auxiliary_loss_clip": 0.01103004,
      "auxiliary_loss_mlp": 0.01035534,
      "balance_loss_clip": 1.04720974,
      "balance_loss_mlp": 1.02076983,
      "epoch": 0.36344506237787466,
      "flos": 25849147357440.0,
      "grad_norm": 1.6672479402902334,
      "language_loss": 0.81177461,
      "learning_rate": 2.9426968700241066e-06,
      "loss": 0.83315998,
      "num_input_tokens_seen": 129811075,
      "step": 6045,
      "time_per_iteration": 2.543175220489502
    },
    {
      "auxiliary_loss_clip": 0.01097079,
      "auxiliary_loss_mlp": 0.01039498,
      "balance_loss_clip": 1.0464139,
      "balance_loss_mlp": 1.02378631,
      "epoch": 0.3635051856305426,
      "flos": 30154046503680.0,
      "grad_norm": 13.465663042396093,
      "language_loss": 0.64932823,
      "learning_rate": 2.942353367559755e-06,
      "loss": 0.67069405,
      "num_input_tokens_seen": 129833755,
      "step": 6046,
      "time_per_iteration": 2.6611034870147705
    },
    {
      "auxiliary_loss_clip": 0.01094823,
      "auxiliary_loss_mlp": 0.01038106,
      "balance_loss_clip": 1.04540467,
      "balance_loss_mlp": 1.02360988,
      "epoch": 0.3635653088832106,
      "flos": 22198396746240.0,
      "grad_norm": 1.5016819739742902,
      "language_loss": 0.77839953,
      "learning_rate": 2.9420098293612692e-06,
      "loss": 0.79972881,
      "num_input_tokens_seen": 129854475,
      "step": 6047,
      "time_per_iteration": 2.5873358249664307
    },
    {
      "auxiliary_loss_clip": 0.01124892,
      "auxiliary_loss_mlp": 0.01041456,
      "balance_loss_clip": 1.04404855,
      "balance_loss_mlp": 1.0249691,
      "epoch": 0.36362543213587856,
      "flos": 24787053083520.0,
      "grad_norm": 1.469956359268572,
      "language_loss": 0.79191428,
      "learning_rate": 2.9416662554416767e-06,
      "loss": 0.81357777,
      "num_input_tokens_seen": 129873530,
      "step": 6048,
      "time_per_iteration": 2.5188872814178467
    },
    {
      "auxiliary_loss_clip": 0.01040358,
      "auxiliary_loss_mlp": 0.01006193,
      "balance_loss_clip": 1.02565217,
      "balance_loss_mlp": 1.00408351,
      "epoch": 0.3636855553885465,
      "flos": 62526369231360.0,
      "grad_norm": 0.7629309241054875,
      "language_loss": 0.52642262,
      "learning_rate": 2.9413226458140054e-06,
      "loss": 0.54688811,
      "num_input_tokens_seen": 129940400,
      "step": 6049,
      "time_per_iteration": 3.18770694732666
    },
    {
      "auxiliary_loss_clip": 0.01096147,
      "auxiliary_loss_mlp": 0.01039238,
      "balance_loss_clip": 1.04733598,
      "balance_loss_mlp": 1.02349019,
      "epoch": 0.3637456786412145,
      "flos": 24060652341120.0,
      "grad_norm": 1.883574866921462,
      "language_loss": 0.86228478,
      "learning_rate": 2.9409790004912845e-06,
      "loss": 0.88363862,
      "num_input_tokens_seen": 129958635,
      "step": 6050,
      "time_per_iteration": 2.635986328125
    },
    {
      "auxiliary_loss_clip": 0.01110905,
      "auxiliary_loss_mlp": 0.00793841,
      "balance_loss_clip": 1.04545105,
      "balance_loss_mlp": 1.01881254,
      "epoch": 0.36380580189388245,
      "flos": 16691495852160.0,
      "grad_norm": 1.7856763143130676,
      "language_loss": 0.78172505,
      "learning_rate": 2.940635319486546e-06,
      "loss": 0.80077255,
      "num_input_tokens_seen": 129977685,
      "step": 6051,
      "time_per_iteration": 2.504589319229126
    },
    {
      "auxiliary_loss_clip": 0.01116475,
      "auxiliary_loss_mlp": 0.0103862,
      "balance_loss_clip": 1.04212189,
      "balance_loss_mlp": 1.0237608,
      "epoch": 0.3638659251465504,
      "flos": 25114091437440.0,
      "grad_norm": 1.8315673452217278,
      "language_loss": 0.82433623,
      "learning_rate": 2.940291602812822e-06,
      "loss": 0.84588718,
      "num_input_tokens_seen": 129997530,
      "step": 6052,
      "time_per_iteration": 2.520249605178833
    },
    {
      "auxiliary_loss_clip": 0.01091499,
      "auxiliary_loss_mlp": 0.01036077,
      "balance_loss_clip": 1.04288995,
      "balance_loss_mlp": 1.02184916,
      "epoch": 0.3639260483992184,
      "flos": 23003011353600.0,
      "grad_norm": 1.496378404638616,
      "language_loss": 0.72262114,
      "learning_rate": 2.939947850483145e-06,
      "loss": 0.7438969,
      "num_input_tokens_seen": 130017955,
      "step": 6053,
      "time_per_iteration": 2.5686519145965576
    },
    {
      "auxiliary_loss_clip": 0.01004423,
      "auxiliary_loss_mlp": 0.01005514,
      "balance_loss_clip": 1.02512705,
      "balance_loss_mlp": 1.00323737,
      "epoch": 0.36398617165188635,
      "flos": 70716011160960.0,
      "grad_norm": 0.7717051155076367,
      "language_loss": 0.61222243,
      "learning_rate": 2.9396040625105532e-06,
      "loss": 0.63232177,
      "num_input_tokens_seen": 130074275,
      "step": 6054,
      "time_per_iteration": 3.1961095333099365
    },
    {
      "auxiliary_loss_clip": 0.0110392,
      "auxiliary_loss_mlp": 0.01038181,
      "balance_loss_clip": 1.04573643,
      "balance_loss_mlp": 1.02225471,
      "epoch": 0.3640462949045543,
      "flos": 22235456603520.0,
      "grad_norm": 2.065726572044602,
      "language_loss": 0.75770104,
      "learning_rate": 2.9392602389080802e-06,
      "loss": 0.77912205,
      "num_input_tokens_seen": 130091375,
      "step": 6055,
      "time_per_iteration": 2.5467910766601562
    },
    {
      "auxiliary_loss_clip": 0.0112799,
      "auxiliary_loss_mlp": 0.01040012,
      "balance_loss_clip": 1.04525304,
      "balance_loss_mlp": 1.02484858,
      "epoch": 0.3641064181572223,
      "flos": 21543529939200.0,
      "grad_norm": 1.9147785554283034,
      "language_loss": 0.75472343,
      "learning_rate": 2.938916379688765e-06,
      "loss": 0.77640343,
      "num_input_tokens_seen": 130111595,
      "step": 6056,
      "time_per_iteration": 2.484001636505127
    },
    {
      "auxiliary_loss_clip": 0.01107383,
      "auxiliary_loss_mlp": 0.01039791,
      "balance_loss_clip": 1.04489088,
      "balance_loss_mlp": 1.02419877,
      "epoch": 0.3641665414098903,
      "flos": 22273306560000.0,
      "grad_norm": 1.826966567268114,
      "language_loss": 0.80148941,
      "learning_rate": 2.9385724848656468e-06,
      "loss": 0.82296109,
      "num_input_tokens_seen": 130131440,
      "step": 6057,
      "time_per_iteration": 2.537064790725708
    },
    {
      "auxiliary_loss_clip": 0.01106259,
      "auxiliary_loss_mlp": 0.01040987,
      "balance_loss_clip": 1.04391456,
      "balance_loss_mlp": 1.02494168,
      "epoch": 0.36422666466255826,
      "flos": 28329676778880.0,
      "grad_norm": 2.01456091769444,
      "language_loss": 0.80121446,
      "learning_rate": 2.9382285544517647e-06,
      "loss": 0.82268697,
      "num_input_tokens_seen": 130151375,
      "step": 6058,
      "time_per_iteration": 2.5711193084716797
    },
    {
      "auxiliary_loss_clip": 0.01098697,
      "auxiliary_loss_mlp": 0.00793033,
      "balance_loss_clip": 1.0430814,
      "balance_loss_mlp": 1.01441312,
      "epoch": 0.36428678791522623,
      "flos": 24170503109760.0,
      "grad_norm": 1.6483794857566985,
      "language_loss": 0.84644485,
      "learning_rate": 2.9378845884601636e-06,
      "loss": 0.86536217,
      "num_input_tokens_seen": 130169960,
      "step": 6059,
      "time_per_iteration": 3.924856424331665
    },
    {
      "auxiliary_loss_clip": 0.01087183,
      "auxiliary_loss_mlp": 0.01038061,
      "balance_loss_clip": 1.04311466,
      "balance_loss_mlp": 1.02189624,
      "epoch": 0.3643469111678942,
      "flos": 22528451842560.0,
      "grad_norm": 1.6150583535284802,
      "language_loss": 0.8768636,
      "learning_rate": 2.937540586903884e-06,
      "loss": 0.89811599,
      "num_input_tokens_seen": 130189800,
      "step": 6060,
      "time_per_iteration": 2.562365770339966
    },
    {
      "auxiliary_loss_clip": 0.0111642,
      "auxiliary_loss_mlp": 0.01040432,
      "balance_loss_clip": 1.04550147,
      "balance_loss_mlp": 1.02449405,
      "epoch": 0.36440703442056216,
      "flos": 19426595938560.0,
      "grad_norm": 2.564932396099177,
      "language_loss": 0.67016852,
      "learning_rate": 2.937196549795971e-06,
      "loss": 0.69173706,
      "num_input_tokens_seen": 130206370,
      "step": 6061,
      "time_per_iteration": 2.4822640419006348
    },
    {
      "auxiliary_loss_clip": 0.01110269,
      "auxiliary_loss_mlp": 0.01036598,
      "balance_loss_clip": 1.04770267,
      "balance_loss_mlp": 1.02091002,
      "epoch": 0.3644671576732301,
      "flos": 18040515966720.0,
      "grad_norm": 2.700831389231872,
      "language_loss": 0.75577581,
      "learning_rate": 2.9368524771494718e-06,
      "loss": 0.77724445,
      "num_input_tokens_seen": 130224445,
      "step": 6062,
      "time_per_iteration": 3.891718626022339
    },
    {
      "auxiliary_loss_clip": 0.0109996,
      "auxiliary_loss_mlp": 0.01032633,
      "balance_loss_clip": 1.04464197,
      "balance_loss_mlp": 1.01520491,
      "epoch": 0.3645272809258981,
      "flos": 21542811667200.0,
      "grad_norm": 1.7063669090131988,
      "language_loss": 0.72117579,
      "learning_rate": 2.936508368977432e-06,
      "loss": 0.74250174,
      "num_input_tokens_seen": 130245380,
      "step": 6063,
      "time_per_iteration": 3.9826512336730957
    },
    {
      "auxiliary_loss_clip": 0.01112027,
      "auxiliary_loss_mlp": 0.01044211,
      "balance_loss_clip": 1.04208755,
      "balance_loss_mlp": 1.02848721,
      "epoch": 0.36458740417856605,
      "flos": 22746860490240.0,
      "grad_norm": 1.938577346452718,
      "language_loss": 0.67782688,
      "learning_rate": 2.936164225292901e-06,
      "loss": 0.69938922,
      "num_input_tokens_seen": 130265575,
      "step": 6064,
      "time_per_iteration": 2.4968390464782715
    },
    {
      "auxiliary_loss_clip": 0.01109061,
      "auxiliary_loss_mlp": 0.01049263,
      "balance_loss_clip": 1.0457238,
      "balance_loss_mlp": 1.03356302,
      "epoch": 0.364647527431234,
      "flos": 26140670138880.0,
      "grad_norm": 1.682788671934112,
      "language_loss": 0.74713975,
      "learning_rate": 2.9358200461089297e-06,
      "loss": 0.76872301,
      "num_input_tokens_seen": 130286195,
      "step": 6065,
      "time_per_iteration": 3.950160026550293
    },
    {
      "auxiliary_loss_clip": 0.0110889,
      "auxiliary_loss_mlp": 0.01042763,
      "balance_loss_clip": 1.04672289,
      "balance_loss_mlp": 1.02608538,
      "epoch": 0.364707650683902,
      "flos": 31029907737600.0,
      "grad_norm": 2.752269580475105,
      "language_loss": 0.75211704,
      "learning_rate": 2.9354758314385676e-06,
      "loss": 0.77363354,
      "num_input_tokens_seen": 130306095,
      "step": 6066,
      "time_per_iteration": 2.6268577575683594
    },
    {
      "auxiliary_loss_clip": 0.01117954,
      "auxiliary_loss_mlp": 0.01034824,
      "balance_loss_clip": 1.04449224,
      "balance_loss_mlp": 1.02042317,
      "epoch": 0.36476777393656995,
      "flos": 19572896033280.0,
      "grad_norm": 2.1359486957883456,
      "language_loss": 0.76598907,
      "learning_rate": 2.9351315812948684e-06,
      "loss": 0.78751683,
      "num_input_tokens_seen": 130324685,
      "step": 6067,
      "time_per_iteration": 2.489253520965576
    },
    {
      "auxiliary_loss_clip": 0.0112615,
      "auxiliary_loss_mlp": 0.01039752,
      "balance_loss_clip": 1.04649043,
      "balance_loss_mlp": 1.02653766,
      "epoch": 0.3648278971892379,
      "flos": 17748849530880.0,
      "grad_norm": 1.9603120328968915,
      "language_loss": 0.70918894,
      "learning_rate": 2.934787295690886e-06,
      "loss": 0.7308479,
      "num_input_tokens_seen": 130343855,
      "step": 6068,
      "time_per_iteration": 2.454944133758545
    },
    {
      "auxiliary_loss_clip": 0.01114134,
      "auxiliary_loss_mlp": 0.01037518,
      "balance_loss_clip": 1.04220486,
      "balance_loss_mlp": 1.02262259,
      "epoch": 0.3648880204419059,
      "flos": 17931167988480.0,
      "grad_norm": 1.9875537767721032,
      "language_loss": 0.74069566,
      "learning_rate": 2.9344429746396755e-06,
      "loss": 0.76221216,
      "num_input_tokens_seen": 130362320,
      "step": 6069,
      "time_per_iteration": 2.4577574729919434
    },
    {
      "auxiliary_loss_clip": 0.01108907,
      "auxiliary_loss_mlp": 0.01041682,
      "balance_loss_clip": 1.04535913,
      "balance_loss_mlp": 1.02629244,
      "epoch": 0.3649481436945739,
      "flos": 22638266697600.0,
      "grad_norm": 1.9009548444488127,
      "language_loss": 0.66473973,
      "learning_rate": 2.9340986181542945e-06,
      "loss": 0.68624562,
      "num_input_tokens_seen": 130383165,
      "step": 6070,
      "time_per_iteration": 2.5559580326080322
    },
    {
      "auxiliary_loss_clip": 0.01108932,
      "auxiliary_loss_mlp": 0.01035829,
      "balance_loss_clip": 1.04378128,
      "balance_loss_mlp": 1.02139306,
      "epoch": 0.36500826694724187,
      "flos": 21579656042880.0,
      "grad_norm": 1.872183989302307,
      "language_loss": 0.74425459,
      "learning_rate": 2.9337542262477994e-06,
      "loss": 0.76570225,
      "num_input_tokens_seen": 130402425,
      "step": 6071,
      "time_per_iteration": 2.507399559020996
    },
    {
      "auxiliary_loss_clip": 0.01113699,
      "auxiliary_loss_mlp": 0.01035504,
      "balance_loss_clip": 1.04160333,
      "balance_loss_mlp": 1.02042413,
      "epoch": 0.36506839019990983,
      "flos": 13772533023360.0,
      "grad_norm": 2.139390371167385,
      "language_loss": 0.88506091,
      "learning_rate": 2.9334097989332506e-06,
      "loss": 0.90655291,
      "num_input_tokens_seen": 130419440,
      "step": 6072,
      "time_per_iteration": 2.466857671737671
    },
    {
      "auxiliary_loss_clip": 0.01113469,
      "auxiliary_loss_mlp": 0.01035063,
      "balance_loss_clip": 1.04693174,
      "balance_loss_mlp": 1.0207932,
      "epoch": 0.3651285134525778,
      "flos": 17274972378240.0,
      "grad_norm": 2.6466107161755605,
      "language_loss": 0.72100651,
      "learning_rate": 2.9330653362237094e-06,
      "loss": 0.74249184,
      "num_input_tokens_seen": 130438495,
      "step": 6073,
      "time_per_iteration": 2.4888339042663574
    },
    {
      "auxiliary_loss_clip": 0.01062767,
      "auxiliary_loss_mlp": 0.01039381,
      "balance_loss_clip": 1.05234289,
      "balance_loss_mlp": 1.02314436,
      "epoch": 0.36518863670524576,
      "flos": 21907987286400.0,
      "grad_norm": 1.930718690604759,
      "language_loss": 0.67158616,
      "learning_rate": 2.932720838132236e-06,
      "loss": 0.69260764,
      "num_input_tokens_seen": 130455575,
      "step": 6074,
      "time_per_iteration": 2.617621421813965
    },
    {
      "auxiliary_loss_clip": 0.01094481,
      "auxiliary_loss_mlp": 0.01033242,
      "balance_loss_clip": 1.04447913,
      "balance_loss_mlp": 1.01825118,
      "epoch": 0.3652487599579137,
      "flos": 27122180250240.0,
      "grad_norm": 1.6468493455759396,
      "language_loss": 0.72885096,
      "learning_rate": 2.9323763046718954e-06,
      "loss": 0.75012827,
      "num_input_tokens_seen": 130476385,
      "step": 6075,
      "time_per_iteration": 2.6079723834991455
    },
    {
      "auxiliary_loss_clip": 0.01091713,
      "auxiliary_loss_mlp": 0.01044207,
      "balance_loss_clip": 1.04318392,
      "balance_loss_mlp": 1.027601,
      "epoch": 0.3653088832105817,
      "flos": 19755573626880.0,
      "grad_norm": 2.093723467979191,
      "language_loss": 0.89077723,
      "learning_rate": 2.9320317358557524e-06,
      "loss": 0.91213644,
      "num_input_tokens_seen": 130493630,
      "step": 6076,
      "time_per_iteration": 2.5508856773376465
    },
    {
      "auxiliary_loss_clip": 0.011144,
      "auxiliary_loss_mlp": 0.01038987,
      "balance_loss_clip": 1.04395878,
      "balance_loss_mlp": 1.02340031,
      "epoch": 0.36536900646324966,
      "flos": 13115008609920.0,
      "grad_norm": 1.9196009049641871,
      "language_loss": 0.69566894,
      "learning_rate": 2.931687131696872e-06,
      "loss": 0.71720278,
      "num_input_tokens_seen": 130510735,
      "step": 6077,
      "time_per_iteration": 2.454378604888916
    },
    {
      "auxiliary_loss_clip": 0.01043968,
      "auxiliary_loss_mlp": 0.01005372,
      "balance_loss_clip": 1.01583982,
      "balance_loss_mlp": 1.00376272,
      "epoch": 0.3654291297159176,
      "flos": 71100472383360.0,
      "grad_norm": 0.7612136279177231,
      "language_loss": 0.61825776,
      "learning_rate": 2.9313424922083224e-06,
      "loss": 0.63875115,
      "num_input_tokens_seen": 130577050,
      "step": 6078,
      "time_per_iteration": 3.157278299331665
    },
    {
      "auxiliary_loss_clip": 0.01099927,
      "auxiliary_loss_mlp": 0.0104753,
      "balance_loss_clip": 1.04087925,
      "balance_loss_mlp": 1.03096008,
      "epoch": 0.3654892529685856,
      "flos": 23617478338560.0,
      "grad_norm": 2.1576000060329537,
      "language_loss": 0.78426051,
      "learning_rate": 2.930997817403173e-06,
      "loss": 0.80573511,
      "num_input_tokens_seen": 130593780,
      "step": 6079,
      "time_per_iteration": 2.5255868434906006
    },
    {
      "auxiliary_loss_clip": 0.01121525,
      "auxiliary_loss_mlp": 0.01039465,
      "balance_loss_clip": 1.04751086,
      "balance_loss_mlp": 1.02358651,
      "epoch": 0.36554937622125355,
      "flos": 43470799850880.0,
      "grad_norm": 2.0154887763837888,
      "language_loss": 0.62897575,
      "learning_rate": 2.9306531072944913e-06,
      "loss": 0.65058565,
      "num_input_tokens_seen": 130615510,
      "step": 6080,
      "time_per_iteration": 2.7311177253723145
    },
    {
      "auxiliary_loss_clip": 0.0110001,
      "auxiliary_loss_mlp": 0.01038644,
      "balance_loss_clip": 1.04397547,
      "balance_loss_mlp": 1.02227616,
      "epoch": 0.3656094994739215,
      "flos": 23294641875840.0,
      "grad_norm": 2.2816388599813258,
      "language_loss": 0.6710965,
      "learning_rate": 2.930308361895352e-06,
      "loss": 0.69248307,
      "num_input_tokens_seen": 130635410,
      "step": 6081,
      "time_per_iteration": 2.561708450317383
    },
    {
      "auxiliary_loss_clip": 0.01106086,
      "auxiliary_loss_mlp": 0.0079743,
      "balance_loss_clip": 1.04607272,
      "balance_loss_mlp": 1.02096248,
      "epoch": 0.3656696227265895,
      "flos": 24571984400640.0,
      "grad_norm": 1.858536693097638,
      "language_loss": 0.74903011,
      "learning_rate": 2.9299635812188257e-06,
      "loss": 0.76806527,
      "num_input_tokens_seen": 130657725,
      "step": 6082,
      "time_per_iteration": 2.6148223876953125
    },
    {
      "auxiliary_loss_clip": 0.01075881,
      "auxiliary_loss_mlp": 0.00793764,
      "balance_loss_clip": 1.05218089,
      "balance_loss_mlp": 1.0156951,
      "epoch": 0.3657297459792575,
      "flos": 27928375056000.0,
      "grad_norm": 1.7435726883035343,
      "language_loss": 0.82583416,
      "learning_rate": 2.929618765277987e-06,
      "loss": 0.8445307,
      "num_input_tokens_seen": 130678360,
      "step": 6083,
      "time_per_iteration": 2.6629860401153564
    },
    {
      "auxiliary_loss_clip": 0.01027303,
      "auxiliary_loss_mlp": 0.01006228,
      "balance_loss_clip": 1.01864743,
      "balance_loss_mlp": 1.00452375,
      "epoch": 0.36578986923192547,
      "flos": 67392622126080.0,
      "grad_norm": 0.9284324644293693,
      "language_loss": 0.59299016,
      "learning_rate": 2.9292739140859125e-06,
      "loss": 0.61332548,
      "num_input_tokens_seen": 130742110,
      "step": 6084,
      "time_per_iteration": 3.2181131839752197
    },
    {
      "auxiliary_loss_clip": 0.01085114,
      "auxiliary_loss_mlp": 0.01040993,
      "balance_loss_clip": 1.04278135,
      "balance_loss_mlp": 1.0248636,
      "epoch": 0.36584999248459343,
      "flos": 20227511445120.0,
      "grad_norm": 3.296435953774852,
      "language_loss": 0.72653377,
      "learning_rate": 2.9289290276556767e-06,
      "loss": 0.74779487,
      "num_input_tokens_seen": 130759870,
      "step": 6085,
      "time_per_iteration": 2.539498805999756
    },
    {
      "auxiliary_loss_clip": 0.01094064,
      "auxiliary_loss_mlp": 0.01032965,
      "balance_loss_clip": 1.04931378,
      "balance_loss_mlp": 1.01847506,
      "epoch": 0.3659101157372614,
      "flos": 19062461813760.0,
      "grad_norm": 1.871657058101363,
      "language_loss": 0.78077137,
      "learning_rate": 2.9285841060003604e-06,
      "loss": 0.80204165,
      "num_input_tokens_seen": 130778510,
      "step": 6086,
      "time_per_iteration": 2.590996503829956
    },
    {
      "auxiliary_loss_clip": 0.01112933,
      "auxiliary_loss_mlp": 0.01038213,
      "balance_loss_clip": 1.04846668,
      "balance_loss_mlp": 1.02289474,
      "epoch": 0.36597023898992936,
      "flos": 30810708990720.0,
      "grad_norm": 1.6766818840205613,
      "language_loss": 0.76512969,
      "learning_rate": 2.9282391491330416e-06,
      "loss": 0.78664112,
      "num_input_tokens_seen": 130798535,
      "step": 6087,
      "time_per_iteration": 2.6015231609344482
    },
    {
      "auxiliary_loss_clip": 0.01078051,
      "auxiliary_loss_mlp": 0.01039489,
      "balance_loss_clip": 1.05248952,
      "balance_loss_mlp": 1.02318132,
      "epoch": 0.36603036224259733,
      "flos": 20521799573760.0,
      "grad_norm": 1.9384688504283198,
      "language_loss": 0.70871842,
      "learning_rate": 2.9278941570668002e-06,
      "loss": 0.72989386,
      "num_input_tokens_seen": 130816655,
      "step": 6088,
      "time_per_iteration": 2.5857160091400146
    },
    {
      "auxiliary_loss_clip": 0.01127188,
      "auxiliary_loss_mlp": 0.01036327,
      "balance_loss_clip": 1.04762185,
      "balance_loss_mlp": 1.01955402,
      "epoch": 0.3660904854952653,
      "flos": 38329397798400.0,
      "grad_norm": 1.5557874812146804,
      "language_loss": 0.79729784,
      "learning_rate": 2.92754912981472e-06,
      "loss": 0.81893295,
      "num_input_tokens_seen": 130841225,
      "step": 6089,
      "time_per_iteration": 2.6712942123413086
    },
    {
      "auxiliary_loss_clip": 0.01093846,
      "auxiliary_loss_mlp": 0.01036043,
      "balance_loss_clip": 1.04426503,
      "balance_loss_mlp": 1.02152324,
      "epoch": 0.36615060874793326,
      "flos": 21835555511040.0,
      "grad_norm": 1.9073078871672935,
      "language_loss": 0.71608889,
      "learning_rate": 2.927204067389884e-06,
      "loss": 0.73738778,
      "num_input_tokens_seen": 130861050,
      "step": 6090,
      "time_per_iteration": 2.5433499813079834
    },
    {
      "auxiliary_loss_clip": 0.01096997,
      "auxiliary_loss_mlp": 0.01047141,
      "balance_loss_clip": 1.04760695,
      "balance_loss_mlp": 1.03246593,
      "epoch": 0.3662107320006012,
      "flos": 16581537342720.0,
      "grad_norm": 1.7369872388773808,
      "language_loss": 0.74036288,
      "learning_rate": 2.9268589698053763e-06,
      "loss": 0.76180428,
      "num_input_tokens_seen": 130879775,
      "step": 6091,
      "time_per_iteration": 2.5093183517456055
    },
    {
      "auxiliary_loss_clip": 0.01077293,
      "auxiliary_loss_mlp": 0.0103592,
      "balance_loss_clip": 1.04941869,
      "balance_loss_mlp": 1.02019668,
      "epoch": 0.3662708552532692,
      "flos": 20958365473920.0,
      "grad_norm": 2.0757921258885523,
      "language_loss": 0.72698426,
      "learning_rate": 2.926513837074284e-06,
      "loss": 0.74811637,
      "num_input_tokens_seen": 130898070,
      "step": 6092,
      "time_per_iteration": 2.6026315689086914
    },
    {
      "auxiliary_loss_clip": 0.01118359,
      "auxiliary_loss_mlp": 0.01044163,
      "balance_loss_clip": 1.04406643,
      "balance_loss_mlp": 1.02816463,
      "epoch": 0.36633097850593715,
      "flos": 21902707987200.0,
      "grad_norm": 1.9803768439813776,
      "language_loss": 0.78518808,
      "learning_rate": 2.9261686692096942e-06,
      "loss": 0.8068133,
      "num_input_tokens_seen": 130915250,
      "step": 6093,
      "time_per_iteration": 2.506542444229126
    },
    {
      "auxiliary_loss_clip": 0.01117346,
      "auxiliary_loss_mlp": 0.0104002,
      "balance_loss_clip": 1.0448997,
      "balance_loss_mlp": 1.02501202,
      "epoch": 0.3663911017586051,
      "flos": 32854133808000.0,
      "grad_norm": 1.6762024235929665,
      "language_loss": 0.74190295,
      "learning_rate": 2.925823466224696e-06,
      "loss": 0.76347661,
      "num_input_tokens_seen": 130936995,
      "step": 6094,
      "time_per_iteration": 2.596832513809204
    },
    {
      "auxiliary_loss_clip": 0.01134309,
      "auxiliary_loss_mlp": 0.01050623,
      "balance_loss_clip": 1.04872465,
      "balance_loss_mlp": 1.03435111,
      "epoch": 0.3664512250112731,
      "flos": 27271748482560.0,
      "grad_norm": 1.5466364579683587,
      "language_loss": 0.79537266,
      "learning_rate": 2.9254782281323785e-06,
      "loss": 0.817222,
      "num_input_tokens_seen": 130957970,
      "step": 6095,
      "time_per_iteration": 2.5259475708007812
    },
    {
      "auxiliary_loss_clip": 0.01118217,
      "auxiliary_loss_mlp": 0.0079321,
      "balance_loss_clip": 1.05036068,
      "balance_loss_mlp": 1.01070547,
      "epoch": 0.3665113482639411,
      "flos": 17784436930560.0,
      "grad_norm": 2.0793459336970974,
      "language_loss": 0.73551685,
      "learning_rate": 2.925132954945834e-06,
      "loss": 0.7546311,
      "num_input_tokens_seen": 130974915,
      "step": 6096,
      "time_per_iteration": 2.4809913635253906
    },
    {
      "auxiliary_loss_clip": 0.01094891,
      "auxiliary_loss_mlp": 0.01036343,
      "balance_loss_clip": 1.04402494,
      "balance_loss_mlp": 1.02027345,
      "epoch": 0.36657147151660907,
      "flos": 27854614477440.0,
      "grad_norm": 2.624478588341291,
      "language_loss": 0.67011607,
      "learning_rate": 2.924787646678155e-06,
      "loss": 0.6914283,
      "num_input_tokens_seen": 130995745,
      "step": 6097,
      "time_per_iteration": 2.600778579711914
    },
    {
      "auxiliary_loss_clip": 0.01076159,
      "auxiliary_loss_mlp": 0.01040296,
      "balance_loss_clip": 1.04815149,
      "balance_loss_mlp": 1.02450037,
      "epoch": 0.36663159476927704,
      "flos": 25374013228800.0,
      "grad_norm": 1.40253038604872,
      "language_loss": 0.77797008,
      "learning_rate": 2.9244423033424365e-06,
      "loss": 0.79913461,
      "num_input_tokens_seen": 131015545,
      "step": 6098,
      "time_per_iteration": 4.028882026672363
    },
    {
      "auxiliary_loss_clip": 0.01119002,
      "auxiliary_loss_mlp": 0.01039086,
      "balance_loss_clip": 1.04716396,
      "balance_loss_mlp": 1.02362466,
      "epoch": 0.366691718021945,
      "flos": 21357225072000.0,
      "grad_norm": 2.2749600564227563,
      "language_loss": 0.73552036,
      "learning_rate": 2.9240969249517723e-06,
      "loss": 0.7571013,
      "num_input_tokens_seen": 131033990,
      "step": 6099,
      "time_per_iteration": 2.504903554916382
    },
    {
      "auxiliary_loss_clip": 0.01106505,
      "auxiliary_loss_mlp": 0.01044316,
      "balance_loss_clip": 1.04654765,
      "balance_loss_mlp": 1.03009462,
      "epoch": 0.36675184127461297,
      "flos": 16800376953600.0,
      "grad_norm": 1.9939212998489233,
      "language_loss": 0.84097457,
      "learning_rate": 2.9237515115192602e-06,
      "loss": 0.86248279,
      "num_input_tokens_seen": 131050710,
      "step": 6100,
      "time_per_iteration": 3.865168333053589
    },
    {
      "auxiliary_loss_clip": 0.01101098,
      "auxiliary_loss_mlp": 0.01034076,
      "balance_loss_clip": 1.04634762,
      "balance_loss_mlp": 1.01834023,
      "epoch": 0.36681196452728093,
      "flos": 21906514828800.0,
      "grad_norm": 1.704657061635493,
      "language_loss": 0.70708424,
      "learning_rate": 2.9234060630579992e-06,
      "loss": 0.72843599,
      "num_input_tokens_seen": 131071435,
      "step": 6101,
      "time_per_iteration": 3.9936938285827637
    },
    {
      "auxiliary_loss_clip": 0.01105007,
      "auxiliary_loss_mlp": 0.01050526,
      "balance_loss_clip": 1.04861736,
      "balance_loss_mlp": 1.03364635,
      "epoch": 0.3668720877799489,
      "flos": 17712436118400.0,
      "grad_norm": 2.445095618826411,
      "language_loss": 0.76653767,
      "learning_rate": 2.9230605795810865e-06,
      "loss": 0.78809297,
      "num_input_tokens_seen": 131088775,
      "step": 6102,
      "time_per_iteration": 2.5468952655792236
    },
    {
      "auxiliary_loss_clip": 0.01124421,
      "auxiliary_loss_mlp": 0.01039248,
      "balance_loss_clip": 1.04738736,
      "balance_loss_mlp": 1.02203393,
      "epoch": 0.36693221103261686,
      "flos": 47045455499520.0,
      "grad_norm": 2.811326826953952,
      "language_loss": 0.70174152,
      "learning_rate": 2.922715061101625e-06,
      "loss": 0.72337818,
      "num_input_tokens_seen": 131112800,
      "step": 6103,
      "time_per_iteration": 2.7469322681427
    },
    {
      "auxiliary_loss_clip": 0.01087449,
      "auxiliary_loss_mlp": 0.0103889,
      "balance_loss_clip": 1.04638886,
      "balance_loss_mlp": 1.02311873,
      "epoch": 0.3669923342852848,
      "flos": 15960929132160.0,
      "grad_norm": 1.7040207470236035,
      "language_loss": 0.71408963,
      "learning_rate": 2.922369507632716e-06,
      "loss": 0.73535299,
      "num_input_tokens_seen": 131131150,
      "step": 6104,
      "time_per_iteration": 4.001452684402466
    },
    {
      "auxiliary_loss_clip": 0.0112185,
      "auxiliary_loss_mlp": 0.01035898,
      "balance_loss_clip": 1.04637861,
      "balance_loss_mlp": 1.01961422,
      "epoch": 0.3670524575379528,
      "flos": 19974485064960.0,
      "grad_norm": 2.326823931650879,
      "language_loss": 0.81627047,
      "learning_rate": 2.9220239191874617e-06,
      "loss": 0.83784795,
      "num_input_tokens_seen": 131150365,
      "step": 6105,
      "time_per_iteration": 2.540252923965454
    },
    {
      "auxiliary_loss_clip": 0.01136886,
      "auxiliary_loss_mlp": 0.0104132,
      "balance_loss_clip": 1.04818118,
      "balance_loss_mlp": 1.02484512,
      "epoch": 0.36711258079062076,
      "flos": 25702955003520.0,
      "grad_norm": 1.833168292539064,
      "language_loss": 0.808936,
      "learning_rate": 2.9216782957789692e-06,
      "loss": 0.83071804,
      "num_input_tokens_seen": 131169310,
      "step": 6106,
      "time_per_iteration": 2.5152275562286377
    },
    {
      "auxiliary_loss_clip": 0.01022407,
      "auxiliary_loss_mlp": 0.00776954,
      "balance_loss_clip": 1.01989818,
      "balance_loss_mlp": 1.00891531,
      "epoch": 0.3671727040432887,
      "flos": 60772743342720.0,
      "grad_norm": 0.6889284615548021,
      "language_loss": 0.59231114,
      "learning_rate": 2.9213326374203426e-06,
      "loss": 0.61030471,
      "num_input_tokens_seen": 131232900,
      "step": 6107,
      "time_per_iteration": 3.202441930770874
    },
    {
      "auxiliary_loss_clip": 0.01109968,
      "auxiliary_loss_mlp": 0.01037545,
      "balance_loss_clip": 1.04972935,
      "balance_loss_mlp": 1.02178538,
      "epoch": 0.3672328272959567,
      "flos": 18661303745280.0,
      "grad_norm": 1.6126174885392424,
      "language_loss": 0.74474585,
      "learning_rate": 2.92098694412469e-06,
      "loss": 0.76622093,
      "num_input_tokens_seen": 131250920,
      "step": 6108,
      "time_per_iteration": 2.559065341949463
    },
    {
      "auxiliary_loss_clip": 0.01126866,
      "auxiliary_loss_mlp": 0.01041562,
      "balance_loss_clip": 1.05067766,
      "balance_loss_mlp": 1.02622032,
      "epoch": 0.3672929505486247,
      "flos": 15049049535360.0,
      "grad_norm": 2.9015915564577845,
      "language_loss": 0.73496532,
      "learning_rate": 2.9206412159051213e-06,
      "loss": 0.75664961,
      "num_input_tokens_seen": 131267910,
      "step": 6109,
      "time_per_iteration": 2.4691619873046875
    },
    {
      "auxiliary_loss_clip": 0.0106396,
      "auxiliary_loss_mlp": 0.0104164,
      "balance_loss_clip": 1.04740691,
      "balance_loss_mlp": 1.02587414,
      "epoch": 0.3673530738012927,
      "flos": 20589347099520.0,
      "grad_norm": 1.903176540664286,
      "language_loss": 0.5298025,
      "learning_rate": 2.920295452774744e-06,
      "loss": 0.5508585,
      "num_input_tokens_seen": 131287150,
      "step": 6110,
      "time_per_iteration": 2.6277663707733154
    },
    {
      "auxiliary_loss_clip": 0.01122791,
      "auxiliary_loss_mlp": 0.01039808,
      "balance_loss_clip": 1.04973459,
      "balance_loss_mlp": 1.02279735,
      "epoch": 0.36741319705396064,
      "flos": 21689830033920.0,
      "grad_norm": 1.5042007512305848,
      "language_loss": 0.80457461,
      "learning_rate": 2.919949654746672e-06,
      "loss": 0.82620066,
      "num_input_tokens_seen": 131308225,
      "step": 6111,
      "time_per_iteration": 2.507981777191162
    },
    {
      "auxiliary_loss_clip": 0.01084468,
      "auxiliary_loss_mlp": 0.01045985,
      "balance_loss_clip": 1.04653597,
      "balance_loss_mlp": 1.0303328,
      "epoch": 0.3674733203066286,
      "flos": 29862200499840.0,
      "grad_norm": 1.5817065969633461,
      "language_loss": 0.7239871,
      "learning_rate": 2.9196038218340163e-06,
      "loss": 0.74529165,
      "num_input_tokens_seen": 131332115,
      "step": 6112,
      "time_per_iteration": 2.6650800704956055
    },
    {
      "auxiliary_loss_clip": 0.01124357,
      "auxiliary_loss_mlp": 0.0104989,
      "balance_loss_clip": 1.04994035,
      "balance_loss_mlp": 1.03391623,
      "epoch": 0.36753344355929657,
      "flos": 18257021193600.0,
      "grad_norm": 1.5497026240478489,
      "language_loss": 0.85167468,
      "learning_rate": 2.919257954049892e-06,
      "loss": 0.8734172,
      "num_input_tokens_seen": 131351885,
      "step": 6113,
      "time_per_iteration": 2.5089728832244873
    },
    {
      "auxiliary_loss_clip": 0.01126687,
      "auxiliary_loss_mlp": 0.01040332,
      "balance_loss_clip": 1.04856491,
      "balance_loss_mlp": 1.02389336,
      "epoch": 0.36759356681196453,
      "flos": 25301150490240.0,
      "grad_norm": 1.6326923726119238,
      "language_loss": 0.78214633,
      "learning_rate": 2.918912051407413e-06,
      "loss": 0.80381656,
      "num_input_tokens_seen": 131370245,
      "step": 6114,
      "time_per_iteration": 2.5376169681549072
    },
    {
      "auxiliary_loss_clip": 0.01130001,
      "auxiliary_loss_mlp": 0.01052057,
      "balance_loss_clip": 1.05014563,
      "balance_loss_mlp": 1.03324568,
      "epoch": 0.3676536900646325,
      "flos": 21032952065280.0,
      "grad_norm": 1.5675251942520296,
      "language_loss": 0.66902125,
      "learning_rate": 2.918566113919698e-06,
      "loss": 0.69084179,
      "num_input_tokens_seen": 131388115,
      "step": 6115,
      "time_per_iteration": 2.4861996173858643
    },
    {
      "auxiliary_loss_clip": 0.01103854,
      "auxiliary_loss_mlp": 0.01038856,
      "balance_loss_clip": 1.0450604,
      "balance_loss_mlp": 1.02335334,
      "epoch": 0.36771381331730046,
      "flos": 16288506190080.0,
      "grad_norm": 2.9648017137251914,
      "language_loss": 0.76171589,
      "learning_rate": 2.9182201415998636e-06,
      "loss": 0.78314298,
      "num_input_tokens_seen": 131404595,
      "step": 6116,
      "time_per_iteration": 2.508434772491455
    },
    {
      "auxiliary_loss_clip": 0.01089919,
      "auxiliary_loss_mlp": 0.01044235,
      "balance_loss_clip": 1.05045295,
      "balance_loss_mlp": 1.02843976,
      "epoch": 0.36777393656996843,
      "flos": 22309971367680.0,
      "grad_norm": 2.1194911971804102,
      "language_loss": 0.63123691,
      "learning_rate": 2.9178741344610286e-06,
      "loss": 0.65257847,
      "num_input_tokens_seen": 131423760,
      "step": 6117,
      "time_per_iteration": 2.6013195514678955
    },
    {
      "auxiliary_loss_clip": 0.01107894,
      "auxiliary_loss_mlp": 0.01041373,
      "balance_loss_clip": 1.0507319,
      "balance_loss_mlp": 1.02481472,
      "epoch": 0.3678340598226364,
      "flos": 26834069260800.0,
      "grad_norm": 1.6880812959521245,
      "language_loss": 0.7319963,
      "learning_rate": 2.9175280925163156e-06,
      "loss": 0.75348896,
      "num_input_tokens_seen": 131444955,
      "step": 6118,
      "time_per_iteration": 2.59920334815979
    },
    {
      "auxiliary_loss_clip": 0.01131272,
      "auxiliary_loss_mlp": 0.01047082,
      "balance_loss_clip": 1.05355799,
      "balance_loss_mlp": 1.02978444,
      "epoch": 0.36789418307530436,
      "flos": 21761723105280.0,
      "grad_norm": 1.53855741114017,
      "language_loss": 0.73045731,
      "learning_rate": 2.9171820157788445e-06,
      "loss": 0.75224084,
      "num_input_tokens_seen": 131465720,
      "step": 6119,
      "time_per_iteration": 2.5119521617889404
    },
    {
      "auxiliary_loss_clip": 0.01110214,
      "auxiliary_loss_mlp": 0.01037828,
      "balance_loss_clip": 1.05250955,
      "balance_loss_mlp": 1.02114522,
      "epoch": 0.3679543063279723,
      "flos": 15924192497280.0,
      "grad_norm": 1.861733247793623,
      "language_loss": 0.79959798,
      "learning_rate": 2.9168359042617404e-06,
      "loss": 0.82107842,
      "num_input_tokens_seen": 131483080,
      "step": 6120,
      "time_per_iteration": 2.525136709213257
    },
    {
      "auxiliary_loss_clip": 0.0109022,
      "auxiliary_loss_mlp": 0.01045494,
      "balance_loss_clip": 1.04998565,
      "balance_loss_mlp": 1.02923369,
      "epoch": 0.3680144295806403,
      "flos": 24275541456000.0,
      "grad_norm": 1.8849143042750758,
      "language_loss": 0.64188707,
      "learning_rate": 2.916489757978126e-06,
      "loss": 0.66324425,
      "num_input_tokens_seen": 131502545,
      "step": 6121,
      "time_per_iteration": 2.57275652885437
    },
    {
      "auxiliary_loss_clip": 0.01124923,
      "auxiliary_loss_mlp": 0.01039242,
      "balance_loss_clip": 1.05453372,
      "balance_loss_mlp": 1.02283859,
      "epoch": 0.36807455283330826,
      "flos": 26104148985600.0,
      "grad_norm": 2.2295674156198255,
      "language_loss": 0.71526647,
      "learning_rate": 2.9161435769411286e-06,
      "loss": 0.73690814,
      "num_input_tokens_seen": 131522155,
      "step": 6122,
      "time_per_iteration": 2.5409178733825684
    },
    {
      "auxiliary_loss_clip": 0.01105622,
      "auxiliary_loss_mlp": 0.01044102,
      "balance_loss_clip": 1.05037045,
      "balance_loss_mlp": 1.02697182,
      "epoch": 0.3681346760859763,
      "flos": 24644990793600.0,
      "grad_norm": 1.8540027101138485,
      "language_loss": 0.69202554,
      "learning_rate": 2.915797361163875e-06,
      "loss": 0.71352279,
      "num_input_tokens_seen": 131543865,
      "step": 6123,
      "time_per_iteration": 2.575061082839966
    },
    {
      "auxiliary_loss_clip": 0.01126637,
      "auxiliary_loss_mlp": 0.01043992,
      "balance_loss_clip": 1.0516547,
      "balance_loss_mlp": 1.0267899,
      "epoch": 0.36819479933864424,
      "flos": 23878369797120.0,
      "grad_norm": 2.849257793413385,
      "language_loss": 0.73721933,
      "learning_rate": 2.9154511106594933e-06,
      "loss": 0.75892562,
      "num_input_tokens_seen": 131562155,
      "step": 6124,
      "time_per_iteration": 2.558530569076538
    },
    {
      "auxiliary_loss_clip": 0.0112151,
      "auxiliary_loss_mlp": 0.01044038,
      "balance_loss_clip": 1.05293846,
      "balance_loss_mlp": 1.02678871,
      "epoch": 0.3682549225913122,
      "flos": 25553997302400.0,
      "grad_norm": 2.299019876289324,
      "language_loss": 0.74069089,
      "learning_rate": 2.915104825441114e-06,
      "loss": 0.76234639,
      "num_input_tokens_seen": 131581695,
      "step": 6125,
      "time_per_iteration": 2.5590097904205322
    },
    {
      "auxiliary_loss_clip": 0.01127158,
      "auxiliary_loss_mlp": 0.01049757,
      "balance_loss_clip": 1.05137181,
      "balance_loss_mlp": 1.0325551,
      "epoch": 0.36831504584398017,
      "flos": 16946605221120.0,
      "grad_norm": 1.7862823218916117,
      "language_loss": 0.78219771,
      "learning_rate": 2.9147585055218686e-06,
      "loss": 0.80396688,
      "num_input_tokens_seen": 131599465,
      "step": 6126,
      "time_per_iteration": 2.484788179397583
    },
    {
      "auxiliary_loss_clip": 0.01125966,
      "auxiliary_loss_mlp": 0.01044548,
      "balance_loss_clip": 1.05078197,
      "balance_loss_mlp": 1.0266186,
      "epoch": 0.36837516909664814,
      "flos": 19865065259520.0,
      "grad_norm": 2.5300601756454375,
      "language_loss": 0.65401423,
      "learning_rate": 2.914412150914888e-06,
      "loss": 0.67571932,
      "num_input_tokens_seen": 131618330,
      "step": 6127,
      "time_per_iteration": 2.4821014404296875
    },
    {
      "auxiliary_loss_clip": 0.01116798,
      "auxiliary_loss_mlp": 0.010406,
      "balance_loss_clip": 1.05265021,
      "balance_loss_mlp": 1.02423251,
      "epoch": 0.3684352923493161,
      "flos": 37626984362880.0,
      "grad_norm": 1.7620483621671204,
      "language_loss": 0.70450056,
      "learning_rate": 2.9140657616333074e-06,
      "loss": 0.72607458,
      "num_input_tokens_seen": 131638960,
      "step": 6128,
      "time_per_iteration": 2.6795711517333984
    },
    {
      "auxiliary_loss_clip": 0.01117652,
      "auxiliary_loss_mlp": 0.01045102,
      "balance_loss_clip": 1.05293024,
      "balance_loss_mlp": 1.02897286,
      "epoch": 0.36849541560198407,
      "flos": 14465501182080.0,
      "grad_norm": 1.7183833518881548,
      "language_loss": 0.75252241,
      "learning_rate": 2.9137193376902614e-06,
      "loss": 0.77414995,
      "num_input_tokens_seen": 131657440,
      "step": 6129,
      "time_per_iteration": 2.512605667114258
    },
    {
      "auxiliary_loss_clip": 0.01119593,
      "auxiliary_loss_mlp": 0.01039458,
      "balance_loss_clip": 1.05199516,
      "balance_loss_mlp": 1.02309012,
      "epoch": 0.36855553885465203,
      "flos": 25770753924480.0,
      "grad_norm": 1.585285122339762,
      "language_loss": 0.8442564,
      "learning_rate": 2.9133728790988868e-06,
      "loss": 0.86584687,
      "num_input_tokens_seen": 131678035,
      "step": 6130,
      "time_per_iteration": 2.5723981857299805
    },
    {
      "auxiliary_loss_clip": 0.01031183,
      "auxiliary_loss_mlp": 0.01007887,
      "balance_loss_clip": 1.03440142,
      "balance_loss_mlp": 1.00600374,
      "epoch": 0.36861566210732,
      "flos": 65049417377280.0,
      "grad_norm": 0.8178980750687815,
      "language_loss": 0.60305786,
      "learning_rate": 2.913026385872321e-06,
      "loss": 0.62344861,
      "num_input_tokens_seen": 131742470,
      "step": 6131,
      "time_per_iteration": 3.2082784175872803
    },
    {
      "auxiliary_loss_clip": 0.01098348,
      "auxiliary_loss_mlp": 0.01034192,
      "balance_loss_clip": 1.05278397,
      "balance_loss_mlp": 1.01814628,
      "epoch": 0.36867578535998796,
      "flos": 30954495133440.0,
      "grad_norm": 1.5227882722305346,
      "language_loss": 0.73254573,
      "learning_rate": 2.9126798580237034e-06,
      "loss": 0.75387114,
      "num_input_tokens_seen": 131764570,
      "step": 6132,
      "time_per_iteration": 2.6497342586517334
    },
    {
      "auxiliary_loss_clip": 0.01122661,
      "auxiliary_loss_mlp": 0.01041207,
      "balance_loss_clip": 1.04779208,
      "balance_loss_mlp": 1.02413654,
      "epoch": 0.3687359086126559,
      "flos": 28837956182400.0,
      "grad_norm": 1.5974686555646938,
      "language_loss": 0.74257159,
      "learning_rate": 2.9123332955661736e-06,
      "loss": 0.76421028,
      "num_input_tokens_seen": 131785720,
      "step": 6133,
      "time_per_iteration": 2.5603535175323486
    },
    {
      "auxiliary_loss_clip": 0.01071873,
      "auxiliary_loss_mlp": 0.01050835,
      "balance_loss_clip": 1.04646969,
      "balance_loss_mlp": 1.03269148,
      "epoch": 0.3687960318653239,
      "flos": 21396798881280.0,
      "grad_norm": 1.624825924086567,
      "language_loss": 0.71846235,
      "learning_rate": 2.911986698512874e-06,
      "loss": 0.73968947,
      "num_input_tokens_seen": 131804430,
      "step": 6134,
      "time_per_iteration": 2.6402177810668945
    },
    {
      "auxiliary_loss_clip": 0.01099921,
      "auxiliary_loss_mlp": 0.01037883,
      "balance_loss_clip": 1.04831088,
      "balance_loss_mlp": 1.02165902,
      "epoch": 0.36885615511799186,
      "flos": 20266043760000.0,
      "grad_norm": 1.5045917312043766,
      "language_loss": 0.75128919,
      "learning_rate": 2.9116400668769477e-06,
      "loss": 0.77266729,
      "num_input_tokens_seen": 131822060,
      "step": 6135,
      "time_per_iteration": 2.577648401260376
    },
    {
      "auxiliary_loss_clip": 0.01027316,
      "auxiliary_loss_mlp": 0.01004833,
      "balance_loss_clip": 1.02852416,
      "balance_loss_mlp": 1.00231731,
      "epoch": 0.3689162783706599,
      "flos": 63088836301440.0,
      "grad_norm": 0.8104127227940295,
      "language_loss": 0.58816999,
      "learning_rate": 2.9112934006715376e-06,
      "loss": 0.60849148,
      "num_input_tokens_seen": 131880715,
      "step": 6136,
      "time_per_iteration": 3.112391233444214
    },
    {
      "auxiliary_loss_clip": 0.01107184,
      "auxiliary_loss_mlp": 0.01041645,
      "balance_loss_clip": 1.05224895,
      "balance_loss_mlp": 1.02537298,
      "epoch": 0.36897640162332784,
      "flos": 10961984419200.0,
      "grad_norm": 1.866682844487848,
      "language_loss": 0.78531969,
      "learning_rate": 2.9109466999097918e-06,
      "loss": 0.80680794,
      "num_input_tokens_seen": 131895850,
      "step": 6137,
      "time_per_iteration": 3.9055001735687256
    },
    {
      "auxiliary_loss_clip": 0.01122333,
      "auxiliary_loss_mlp": 0.01043156,
      "balance_loss_clip": 1.04824209,
      "balance_loss_mlp": 1.02701473,
      "epoch": 0.3690365248759958,
      "flos": 20704297599360.0,
      "grad_norm": 1.9926850522133692,
      "language_loss": 0.74251115,
      "learning_rate": 2.9105999646048552e-06,
      "loss": 0.764166,
      "num_input_tokens_seen": 131915775,
      "step": 6138,
      "time_per_iteration": 2.524902582168579
    },
    {
      "auxiliary_loss_clip": 0.01095208,
      "auxiliary_loss_mlp": 0.01042259,
      "balance_loss_clip": 1.0476284,
      "balance_loss_mlp": 1.02580786,
      "epoch": 0.3690966481286638,
      "flos": 31826369957760.0,
      "grad_norm": 1.7278775363486276,
      "language_loss": 0.65140885,
      "learning_rate": 2.9102531947698764e-06,
      "loss": 0.67278349,
      "num_input_tokens_seen": 131935715,
      "step": 6139,
      "time_per_iteration": 4.0203869342803955
    },
    {
      "auxiliary_loss_clip": 0.01097024,
      "auxiliary_loss_mlp": 0.01042575,
      "balance_loss_clip": 1.05215979,
      "balance_loss_mlp": 1.02669644,
      "epoch": 0.36915677138133174,
      "flos": 13114936782720.0,
      "grad_norm": 1.9956853177993548,
      "language_loss": 0.71371067,
      "learning_rate": 2.909906390418006e-06,
      "loss": 0.73510671,
      "num_input_tokens_seen": 131954120,
      "step": 6140,
      "time_per_iteration": 3.9769110679626465
    },
    {
      "auxiliary_loss_clip": 0.010275,
      "auxiliary_loss_mlp": 0.01001973,
      "balance_loss_clip": 1.04225302,
      "balance_loss_mlp": 1.00016141,
      "epoch": 0.3692168946339997,
      "flos": 68686879956480.0,
      "grad_norm": 0.7505637959514583,
      "language_loss": 0.59337437,
      "learning_rate": 2.9095595515623934e-06,
      "loss": 0.61366916,
      "num_input_tokens_seen": 132017485,
      "step": 6141,
      "time_per_iteration": 3.2396280765533447
    },
    {
      "auxiliary_loss_clip": 0.01118507,
      "auxiliary_loss_mlp": 0.01040653,
      "balance_loss_clip": 1.049106,
      "balance_loss_mlp": 1.02491713,
      "epoch": 0.36927701788666767,
      "flos": 22017873968640.0,
      "grad_norm": 1.6767321890717222,
      "language_loss": 0.75071031,
      "learning_rate": 2.909212678216192e-06,
      "loss": 0.77230191,
      "num_input_tokens_seen": 132036760,
      "step": 6142,
      "time_per_iteration": 2.5542895793914795
    },
    {
      "auxiliary_loss_clip": 0.01120082,
      "auxiliary_loss_mlp": 0.01037953,
      "balance_loss_clip": 1.04938674,
      "balance_loss_mlp": 1.02363634,
      "epoch": 0.36933714113933563,
      "flos": 21835591424640.0,
      "grad_norm": 1.8153055345014977,
      "language_loss": 0.77130187,
      "learning_rate": 2.908865770392555e-06,
      "loss": 0.7928822,
      "num_input_tokens_seen": 132056935,
      "step": 6143,
      "time_per_iteration": 3.9607644081115723
    },
    {
      "auxiliary_loss_clip": 0.01117293,
      "auxiliary_loss_mlp": 0.01036106,
      "balance_loss_clip": 1.04770017,
      "balance_loss_mlp": 1.02210498,
      "epoch": 0.3693972643920036,
      "flos": 23691705793920.0,
      "grad_norm": 1.4145135447068977,
      "language_loss": 0.81957954,
      "learning_rate": 2.9085188281046364e-06,
      "loss": 0.84111357,
      "num_input_tokens_seen": 132077285,
      "step": 6144,
      "time_per_iteration": 2.5141496658325195
    },
    {
      "auxiliary_loss_clip": 0.01120902,
      "auxiliary_loss_mlp": 0.01039661,
      "balance_loss_clip": 1.04731321,
      "balance_loss_mlp": 1.02482557,
      "epoch": 0.36945738764467156,
      "flos": 22856747172480.0,
      "grad_norm": 2.3668715566298424,
      "language_loss": 0.78081441,
      "learning_rate": 2.908171851365593e-06,
      "loss": 0.80242002,
      "num_input_tokens_seen": 132095520,
      "step": 6145,
      "time_per_iteration": 2.522660970687866
    },
    {
      "auxiliary_loss_clip": 0.01113343,
      "auxiliary_loss_mlp": 0.01032684,
      "balance_loss_clip": 1.04715729,
      "balance_loss_mlp": 1.01737714,
      "epoch": 0.36951751089733953,
      "flos": 16615939593600.0,
      "grad_norm": 1.6528734925719124,
      "language_loss": 0.76943612,
      "learning_rate": 2.9078248401885815e-06,
      "loss": 0.79089642,
      "num_input_tokens_seen": 132112810,
      "step": 6146,
      "time_per_iteration": 2.481940269470215
    },
    {
      "auxiliary_loss_clip": 0.01105181,
      "auxiliary_loss_mlp": 0.01040696,
      "balance_loss_clip": 1.04799008,
      "balance_loss_mlp": 1.02480531,
      "epoch": 0.3695776341500075,
      "flos": 18914545607040.0,
      "grad_norm": 4.398420387454917,
      "language_loss": 0.80298954,
      "learning_rate": 2.907477794586761e-06,
      "loss": 0.82444835,
      "num_input_tokens_seen": 132131615,
      "step": 6147,
      "time_per_iteration": 2.5308539867401123
    },
    {
      "auxiliary_loss_clip": 0.01098476,
      "auxiliary_loss_mlp": 0.00821718,
      "balance_loss_clip": 1.04763579,
      "balance_loss_mlp": 1.07257986,
      "epoch": 0.36963775740267546,
      "flos": 20808474019200.0,
      "grad_norm": 1.723329282176361,
      "language_loss": 0.8359195,
      "learning_rate": 2.9071307145732926e-06,
      "loss": 0.85512143,
      "num_input_tokens_seen": 132149585,
      "step": 6148,
      "time_per_iteration": 2.5553088188171387
    },
    {
      "auxiliary_loss_clip": 0.01119,
      "auxiliary_loss_mlp": 0.01038145,
      "balance_loss_clip": 1.04968667,
      "balance_loss_mlp": 1.02298117,
      "epoch": 0.3696978806553435,
      "flos": 26061881656320.0,
      "grad_norm": 2.288795123647486,
      "language_loss": 0.7458598,
      "learning_rate": 2.9067836001613357e-06,
      "loss": 0.76743126,
      "num_input_tokens_seen": 132165555,
      "step": 6149,
      "time_per_iteration": 2.531938314437866
    },
    {
      "auxiliary_loss_clip": 0.01132322,
      "auxiliary_loss_mlp": 0.0103992,
      "balance_loss_clip": 1.04774225,
      "balance_loss_mlp": 1.02388644,
      "epoch": 0.36975800390801145,
      "flos": 26833925606400.0,
      "grad_norm": 2.383777960098795,
      "language_loss": 0.71126986,
      "learning_rate": 2.906436451364054e-06,
      "loss": 0.73299229,
      "num_input_tokens_seen": 132185100,
      "step": 6150,
      "time_per_iteration": 2.524066925048828
    },
    {
      "auxiliary_loss_clip": 0.01106454,
      "auxiliary_loss_mlp": 0.01040443,
      "balance_loss_clip": 1.04753923,
      "balance_loss_mlp": 1.02522588,
      "epoch": 0.3698181271606794,
      "flos": 21142623265920.0,
      "grad_norm": 1.8117307459688528,
      "language_loss": 0.81985712,
      "learning_rate": 2.906089268194611e-06,
      "loss": 0.84132612,
      "num_input_tokens_seen": 132203930,
      "step": 6151,
      "time_per_iteration": 2.538400888442993
    },
    {
      "auxiliary_loss_clip": 0.01029227,
      "auxiliary_loss_mlp": 0.01005271,
      "balance_loss_clip": 1.02088761,
      "balance_loss_mlp": 1.00335181,
      "epoch": 0.3698782504133474,
      "flos": 66742639568640.0,
      "grad_norm": 0.8844077944102461,
      "language_loss": 0.63206869,
      "learning_rate": 2.9057420506661726e-06,
      "loss": 0.65241373,
      "num_input_tokens_seen": 132263845,
      "step": 6152,
      "time_per_iteration": 3.2307043075561523
    },
    {
      "auxiliary_loss_clip": 0.01084254,
      "auxiliary_loss_mlp": 0.0104285,
      "balance_loss_clip": 1.04750729,
      "balance_loss_mlp": 1.02732944,
      "epoch": 0.36993837366601534,
      "flos": 24311523905280.0,
      "grad_norm": 2.015452372122809,
      "language_loss": 0.69997603,
      "learning_rate": 2.9053947987919044e-06,
      "loss": 0.72124714,
      "num_input_tokens_seen": 132282350,
      "step": 6153,
      "time_per_iteration": 2.6117208003997803
    },
    {
      "auxiliary_loss_clip": 0.01119298,
      "auxiliary_loss_mlp": 0.01040007,
      "balance_loss_clip": 1.04894376,
      "balance_loss_mlp": 1.02450967,
      "epoch": 0.3699984969186833,
      "flos": 24349194293760.0,
      "grad_norm": 1.7438400191123264,
      "language_loss": 0.72407353,
      "learning_rate": 2.9050475125849755e-06,
      "loss": 0.74566662,
      "num_input_tokens_seen": 132301930,
      "step": 6154,
      "time_per_iteration": 2.5523509979248047
    },
    {
      "auxiliary_loss_clip": 0.01103094,
      "auxiliary_loss_mlp": 0.0103273,
      "balance_loss_clip": 1.04836714,
      "balance_loss_mlp": 1.01861608,
      "epoch": 0.37005862017135127,
      "flos": 19829154637440.0,
      "grad_norm": 1.6464314973772576,
      "language_loss": 0.6855045,
      "learning_rate": 2.9047001920585534e-06,
      "loss": 0.70686269,
      "num_input_tokens_seen": 132320915,
      "step": 6155,
      "time_per_iteration": 2.5585644245147705
    },
    {
      "auxiliary_loss_clip": 0.01118698,
      "auxiliary_loss_mlp": 0.01031832,
      "balance_loss_clip": 1.04801917,
      "balance_loss_mlp": 1.01708579,
      "epoch": 0.37011874342401924,
      "flos": 19573793873280.0,
      "grad_norm": 1.6708482155136135,
      "language_loss": 0.6744163,
      "learning_rate": 2.9043528372258097e-06,
      "loss": 0.6959216,
      "num_input_tokens_seen": 132340415,
      "step": 6156,
      "time_per_iteration": 2.5040578842163086
    },
    {
      "auxiliary_loss_clip": 0.01110524,
      "auxiliary_loss_mlp": 0.01038034,
      "balance_loss_clip": 1.04700351,
      "balance_loss_mlp": 1.02421165,
      "epoch": 0.3701788666766872,
      "flos": 20374350243840.0,
      "grad_norm": 1.9362698794781563,
      "language_loss": 0.81848252,
      "learning_rate": 2.904005448099916e-06,
      "loss": 0.83996809,
      "num_input_tokens_seen": 132358600,
      "step": 6157,
      "time_per_iteration": 2.517601728439331
    },
    {
      "auxiliary_loss_clip": 0.01086716,
      "auxiliary_loss_mlp": 0.01036938,
      "balance_loss_clip": 1.0485661,
      "balance_loss_mlp": 1.02059424,
      "epoch": 0.37023898992935517,
      "flos": 15340931452800.0,
      "grad_norm": 2.626825492492475,
      "language_loss": 0.76345623,
      "learning_rate": 2.9036580246940444e-06,
      "loss": 0.7846927,
      "num_input_tokens_seen": 132373160,
      "step": 6158,
      "time_per_iteration": 2.58760142326355
    },
    {
      "auxiliary_loss_clip": 0.01131595,
      "auxiliary_loss_mlp": 0.01035343,
      "balance_loss_clip": 1.04755771,
      "balance_loss_mlp": 1.0198344,
      "epoch": 0.37029911318202313,
      "flos": 19573937527680.0,
      "grad_norm": 2.2943862564096245,
      "language_loss": 0.68775821,
      "learning_rate": 2.9033105670213708e-06,
      "loss": 0.7094276,
      "num_input_tokens_seen": 132392345,
      "step": 6159,
      "time_per_iteration": 2.468376874923706
    },
    {
      "auxiliary_loss_clip": 0.01102178,
      "auxiliary_loss_mlp": 0.01038151,
      "balance_loss_clip": 1.0471673,
      "balance_loss_mlp": 1.02436399,
      "epoch": 0.3703592364346911,
      "flos": 26213353309440.0,
      "grad_norm": 1.6980922245244061,
      "language_loss": 0.70943904,
      "learning_rate": 2.9029630750950697e-06,
      "loss": 0.73084223,
      "num_input_tokens_seen": 132412620,
      "step": 6160,
      "time_per_iteration": 2.5875155925750732
    },
    {
      "auxiliary_loss_clip": 0.01102664,
      "auxiliary_loss_mlp": 0.01032021,
      "balance_loss_clip": 1.04775286,
      "balance_loss_mlp": 1.01875925,
      "epoch": 0.37041935968735906,
      "flos": 20048317470720.0,
      "grad_norm": 1.722193149307762,
      "language_loss": 0.79082012,
      "learning_rate": 2.9026155489283176e-06,
      "loss": 0.81216693,
      "num_input_tokens_seen": 132431570,
      "step": 6161,
      "time_per_iteration": 2.534054756164551
    },
    {
      "auxiliary_loss_clip": 0.01130329,
      "auxiliary_loss_mlp": 0.01039105,
      "balance_loss_clip": 1.04814982,
      "balance_loss_mlp": 1.02342939,
      "epoch": 0.3704794829400271,
      "flos": 24133802388480.0,
      "grad_norm": 1.85413278882439,
      "language_loss": 0.7959826,
      "learning_rate": 2.902267988534295e-06,
      "loss": 0.81767702,
      "num_input_tokens_seen": 132451525,
      "step": 6162,
      "time_per_iteration": 2.5589780807495117
    },
    {
      "auxiliary_loss_clip": 0.01106328,
      "auxiliary_loss_mlp": 0.00794973,
      "balance_loss_clip": 1.04814398,
      "balance_loss_mlp": 1.01938772,
      "epoch": 0.37053960619269505,
      "flos": 14866874732160.0,
      "grad_norm": 1.893293286849104,
      "language_loss": 0.79322976,
      "learning_rate": 2.9019203939261783e-06,
      "loss": 0.81224275,
      "num_input_tokens_seen": 132469875,
      "step": 6163,
      "time_per_iteration": 2.5130016803741455
    },
    {
      "auxiliary_loss_clip": 0.01116236,
      "auxiliary_loss_mlp": 0.01037411,
      "balance_loss_clip": 1.04762995,
      "balance_loss_mlp": 1.02236688,
      "epoch": 0.370599729445363,
      "flos": 21361498790400.0,
      "grad_norm": 1.5468667580093949,
      "language_loss": 0.67891955,
      "learning_rate": 2.9015727651171507e-06,
      "loss": 0.70045602,
      "num_input_tokens_seen": 132488360,
      "step": 6164,
      "time_per_iteration": 2.500903367996216
    },
    {
      "auxiliary_loss_clip": 0.01104184,
      "auxiliary_loss_mlp": 0.01042532,
      "balance_loss_clip": 1.04926932,
      "balance_loss_mlp": 1.02642679,
      "epoch": 0.370659852698031,
      "flos": 26829041356800.0,
      "grad_norm": 3.0280898960088987,
      "language_loss": 0.83321536,
      "learning_rate": 2.9012251021203935e-06,
      "loss": 0.85468256,
      "num_input_tokens_seen": 132508630,
      "step": 6165,
      "time_per_iteration": 2.5711727142333984
    },
    {
      "auxiliary_loss_clip": 0.01118657,
      "auxiliary_loss_mlp": 0.01040988,
      "balance_loss_clip": 1.05085874,
      "balance_loss_mlp": 1.02406061,
      "epoch": 0.37071997595069894,
      "flos": 19099018880640.0,
      "grad_norm": 5.611673791159961,
      "language_loss": 0.69024491,
      "learning_rate": 2.9008774049490896e-06,
      "loss": 0.71184134,
      "num_input_tokens_seen": 132527465,
      "step": 6166,
      "time_per_iteration": 2.5433568954467773
    },
    {
      "auxiliary_loss_clip": 0.01018998,
      "auxiliary_loss_mlp": 0.0100493,
      "balance_loss_clip": 1.02217937,
      "balance_loss_mlp": 1.00287914,
      "epoch": 0.3707800992033669,
      "flos": 52178384920320.0,
      "grad_norm": 0.7899809670082475,
      "language_loss": 0.56883633,
      "learning_rate": 2.9005296736164244e-06,
      "loss": 0.58907557,
      "num_input_tokens_seen": 132579940,
      "step": 6167,
      "time_per_iteration": 3.0093185901641846
    },
    {
      "auxiliary_loss_clip": 0.01104424,
      "auxiliary_loss_mlp": 0.01034787,
      "balance_loss_clip": 1.04536772,
      "balance_loss_mlp": 1.02047586,
      "epoch": 0.3708402224560349,
      "flos": 19901837808000.0,
      "grad_norm": 1.9196697109395273,
      "language_loss": 0.75189364,
      "learning_rate": 2.900181908135584e-06,
      "loss": 0.77328569,
      "num_input_tokens_seen": 132598390,
      "step": 6168,
      "time_per_iteration": 2.5176985263824463
    },
    {
      "auxiliary_loss_clip": 0.01114721,
      "auxiliary_loss_mlp": 0.00794314,
      "balance_loss_clip": 1.04882908,
      "balance_loss_mlp": 1.01863873,
      "epoch": 0.37090034570870284,
      "flos": 20007630339840.0,
      "grad_norm": 1.6322026154487526,
      "language_loss": 0.7383889,
      "learning_rate": 2.899834108519755e-06,
      "loss": 0.75747931,
      "num_input_tokens_seen": 132616920,
      "step": 6169,
      "time_per_iteration": 2.5108673572540283
    },
    {
      "auxiliary_loss_clip": 0.01128284,
      "auxiliary_loss_mlp": 0.01032862,
      "balance_loss_clip": 1.04925108,
      "balance_loss_mlp": 1.01850915,
      "epoch": 0.3709604689613708,
      "flos": 24134700228480.0,
      "grad_norm": 1.4526614842355259,
      "language_loss": 0.79303074,
      "learning_rate": 2.899486274782127e-06,
      "loss": 0.81464225,
      "num_input_tokens_seen": 132637660,
      "step": 6170,
      "time_per_iteration": 2.5169289112091064
    },
    {
      "auxiliary_loss_clip": 0.01117133,
      "auxiliary_loss_mlp": 0.01037522,
      "balance_loss_clip": 1.05064321,
      "balance_loss_mlp": 1.02219725,
      "epoch": 0.37102059221403877,
      "flos": 23876071326720.0,
      "grad_norm": 1.7631131951199839,
      "language_loss": 0.76149201,
      "learning_rate": 2.8991384069358885e-06,
      "loss": 0.78303856,
      "num_input_tokens_seen": 132657635,
      "step": 6171,
      "time_per_iteration": 2.530855894088745
    },
    {
      "auxiliary_loss_clip": 0.01107081,
      "auxiliary_loss_mlp": 0.01034251,
      "balance_loss_clip": 1.05115438,
      "balance_loss_mlp": 1.01840818,
      "epoch": 0.37108071546670673,
      "flos": 14501268149760.0,
      "grad_norm": 1.924524113688885,
      "language_loss": 0.80882895,
      "learning_rate": 2.898790504994232e-06,
      "loss": 0.83024222,
      "num_input_tokens_seen": 132674455,
      "step": 6172,
      "time_per_iteration": 2.507513999938965
    },
    {
      "auxiliary_loss_clip": 0.01122076,
      "auxiliary_loss_mlp": 0.01034524,
      "balance_loss_clip": 1.04902732,
      "balance_loss_mlp": 1.01852608,
      "epoch": 0.3711408387193747,
      "flos": 34562619279360.0,
      "grad_norm": 1.7862264679881996,
      "language_loss": 0.59549958,
      "learning_rate": 2.89844256897035e-06,
      "loss": 0.61706555,
      "num_input_tokens_seen": 132695140,
      "step": 6173,
      "time_per_iteration": 2.606901168823242
    },
    {
      "auxiliary_loss_clip": 0.01108468,
      "auxiliary_loss_mlp": 0.01034872,
      "balance_loss_clip": 1.0470264,
      "balance_loss_mlp": 1.01905894,
      "epoch": 0.37120096197204266,
      "flos": 17310703432320.0,
      "grad_norm": 1.9394642288181363,
      "language_loss": 0.80719721,
      "learning_rate": 2.898094598877435e-06,
      "loss": 0.82863057,
      "num_input_tokens_seen": 132712470,
      "step": 6174,
      "time_per_iteration": 2.5172033309936523
    },
    {
      "auxiliary_loss_clip": 0.01126777,
      "auxiliary_loss_mlp": 0.01034939,
      "balance_loss_clip": 1.04781544,
      "balance_loss_mlp": 1.020962,
      "epoch": 0.37126108522471063,
      "flos": 30664049760000.0,
      "grad_norm": 1.9898520766266954,
      "language_loss": 0.79815769,
      "learning_rate": 2.8977465947286826e-06,
      "loss": 0.81977487,
      "num_input_tokens_seen": 132732945,
      "step": 6175,
      "time_per_iteration": 3.9395790100097656
    },
    {
      "auxiliary_loss_clip": 0.01120831,
      "auxiliary_loss_mlp": 0.01043015,
      "balance_loss_clip": 1.05058062,
      "balance_loss_mlp": 1.02743483,
      "epoch": 0.37132120847737865,
      "flos": 25155640494720.0,
      "grad_norm": 1.8075899138327054,
      "language_loss": 0.88741648,
      "learning_rate": 2.89739855653729e-06,
      "loss": 0.90905499,
      "num_input_tokens_seen": 132752470,
      "step": 6176,
      "time_per_iteration": 2.530491828918457
    },
    {
      "auxiliary_loss_clip": 0.01120011,
      "auxiliary_loss_mlp": 0.01036782,
      "balance_loss_clip": 1.04930925,
      "balance_loss_mlp": 1.02192879,
      "epoch": 0.3713813317300466,
      "flos": 21213474842880.0,
      "grad_norm": 1.5627903295698438,
      "language_loss": 0.73341531,
      "learning_rate": 2.8970504843164546e-06,
      "loss": 0.75498319,
      "num_input_tokens_seen": 132771485,
      "step": 6177,
      "time_per_iteration": 2.5354979038238525
    },
    {
      "auxiliary_loss_clip": 0.01096408,
      "auxiliary_loss_mlp": 0.010389,
      "balance_loss_clip": 1.04486036,
      "balance_loss_mlp": 1.02364123,
      "epoch": 0.3714414549827146,
      "flos": 21616644072960.0,
      "grad_norm": 1.697243388797151,
      "language_loss": 0.75110203,
      "learning_rate": 2.896702378079374e-06,
      "loss": 0.77245516,
      "num_input_tokens_seen": 132791465,
      "step": 6178,
      "time_per_iteration": 5.406606197357178
    },
    {
      "auxiliary_loss_clip": 0.01070445,
      "auxiliary_loss_mlp": 0.01044133,
      "balance_loss_clip": 1.04715729,
      "balance_loss_mlp": 1.02678812,
      "epoch": 0.37150157823538255,
      "flos": 19972294335360.0,
      "grad_norm": 1.754339233952254,
      "language_loss": 0.71785581,
      "learning_rate": 2.8963542378392502e-06,
      "loss": 0.73900151,
      "num_input_tokens_seen": 132810160,
      "step": 6179,
      "time_per_iteration": 2.6201400756835938
    },
    {
      "auxiliary_loss_clip": 0.01132431,
      "auxiliary_loss_mlp": 0.01041196,
      "balance_loss_clip": 1.0485146,
      "balance_loss_mlp": 1.02468598,
      "epoch": 0.3715617014880505,
      "flos": 24860562266880.0,
      "grad_norm": 2.7936122887318477,
      "language_loss": 0.69970518,
      "learning_rate": 2.896006063609283e-06,
      "loss": 0.72144145,
      "num_input_tokens_seen": 132831265,
      "step": 6180,
      "time_per_iteration": 2.553495168685913
    },
    {
      "auxiliary_loss_clip": 0.01111115,
      "auxiliary_loss_mlp": 0.01032451,
      "balance_loss_clip": 1.04866219,
      "balance_loss_mlp": 1.01780581,
      "epoch": 0.3716218247407185,
      "flos": 20449080489600.0,
      "grad_norm": 1.7078475633671397,
      "language_loss": 0.7835632,
      "learning_rate": 2.8956578554026767e-06,
      "loss": 0.80499887,
      "num_input_tokens_seen": 132850005,
      "step": 6181,
      "time_per_iteration": 3.942554473876953
    },
    {
      "auxiliary_loss_clip": 0.01120653,
      "auxiliary_loss_mlp": 0.01035438,
      "balance_loss_clip": 1.05105305,
      "balance_loss_mlp": 1.0192256,
      "epoch": 0.37168194799338644,
      "flos": 24133479166080.0,
      "grad_norm": 1.7528406209844924,
      "language_loss": 0.78256285,
      "learning_rate": 2.8953096132326343e-06,
      "loss": 0.8041237,
      "num_input_tokens_seen": 132865790,
      "step": 6182,
      "time_per_iteration": 2.506995916366577
    },
    {
      "auxiliary_loss_clip": 0.01030202,
      "auxiliary_loss_mlp": 0.01004693,
      "balance_loss_clip": 1.02361858,
      "balance_loss_mlp": 1.00289273,
      "epoch": 0.3717420712460544,
      "flos": 67408926900480.0,
      "grad_norm": 0.7850100513570387,
      "language_loss": 0.57531583,
      "learning_rate": 2.894961337112362e-06,
      "loss": 0.59566474,
      "num_input_tokens_seen": 132921775,
      "step": 6183,
      "time_per_iteration": 3.1307172775268555
    },
    {
      "auxiliary_loss_clip": 0.01123829,
      "auxiliary_loss_mlp": 0.00798885,
      "balance_loss_clip": 1.04703832,
      "balance_loss_mlp": 1.02290249,
      "epoch": 0.37180219449872237,
      "flos": 22376908362240.0,
      "grad_norm": 1.987292468516446,
      "language_loss": 0.77030551,
      "learning_rate": 2.894613027055066e-06,
      "loss": 0.78953254,
      "num_input_tokens_seen": 132941060,
      "step": 6184,
      "time_per_iteration": 2.5137853622436523
    },
    {
      "auxiliary_loss_clip": 0.01090764,
      "auxiliary_loss_mlp": 0.01039898,
      "balance_loss_clip": 1.04840124,
      "balance_loss_mlp": 1.02487206,
      "epoch": 0.37186231775139034,
      "flos": 21869885934720.0,
      "grad_norm": 2.4799593617328886,
      "language_loss": 0.72667837,
      "learning_rate": 2.894264683073954e-06,
      "loss": 0.74798489,
      "num_input_tokens_seen": 132961850,
      "step": 6185,
      "time_per_iteration": 2.575252056121826
    },
    {
      "auxiliary_loss_clip": 0.0108209,
      "auxiliary_loss_mlp": 0.0103509,
      "balance_loss_clip": 1.04630315,
      "balance_loss_mlp": 1.01918745,
      "epoch": 0.3719224410040583,
      "flos": 22415225195520.0,
      "grad_norm": 1.4290151089854959,
      "language_loss": 0.76732284,
      "learning_rate": 2.8939163051822363e-06,
      "loss": 0.78849465,
      "num_input_tokens_seen": 132981625,
      "step": 6186,
      "time_per_iteration": 2.579238176345825
    },
    {
      "auxiliary_loss_clip": 0.01129203,
      "auxiliary_loss_mlp": 0.01038104,
      "balance_loss_clip": 1.05290973,
      "balance_loss_mlp": 1.02204633,
      "epoch": 0.37198256425672627,
      "flos": 25151223121920.0,
      "grad_norm": 1.7477246713944279,
      "language_loss": 0.834867,
      "learning_rate": 2.8935678933931224e-06,
      "loss": 0.85654014,
      "num_input_tokens_seen": 133001225,
      "step": 6187,
      "time_per_iteration": 2.5495200157165527
    },
    {
      "auxiliary_loss_clip": 0.01118624,
      "auxiliary_loss_mlp": 0.0103659,
      "balance_loss_clip": 1.0459888,
      "balance_loss_mlp": 1.02133131,
      "epoch": 0.37204268750939423,
      "flos": 21138313633920.0,
      "grad_norm": 1.6406345570332361,
      "language_loss": 0.84661615,
      "learning_rate": 2.893219447719824e-06,
      "loss": 0.86816829,
      "num_input_tokens_seen": 133018820,
      "step": 6188,
      "time_per_iteration": 2.4949963092803955
    },
    {
      "auxiliary_loss_clip": 0.01107787,
      "auxiliary_loss_mlp": 0.01036575,
      "balance_loss_clip": 1.05156064,
      "balance_loss_mlp": 1.02068424,
      "epoch": 0.37210281076206225,
      "flos": 21506829217920.0,
      "grad_norm": 1.6342903083261189,
      "language_loss": 0.65319955,
      "learning_rate": 2.8928709681755548e-06,
      "loss": 0.67464316,
      "num_input_tokens_seen": 133040205,
      "step": 6189,
      "time_per_iteration": 2.570903778076172
    },
    {
      "auxiliary_loss_clip": 0.01108993,
      "auxiliary_loss_mlp": 0.01043831,
      "balance_loss_clip": 1.0460602,
      "balance_loss_mlp": 1.02722478,
      "epoch": 0.3721629340147302,
      "flos": 17347835116800.0,
      "grad_norm": 1.6852395240959965,
      "language_loss": 0.84431481,
      "learning_rate": 2.8925224547735293e-06,
      "loss": 0.865843,
      "num_input_tokens_seen": 133058095,
      "step": 6190,
      "time_per_iteration": 2.478821277618408
    },
    {
      "auxiliary_loss_clip": 0.01105737,
      "auxiliary_loss_mlp": 0.01038014,
      "balance_loss_clip": 1.04785359,
      "balance_loss_mlp": 1.02219498,
      "epoch": 0.3722230572673982,
      "flos": 16432400073600.0,
      "grad_norm": 4.988112172548138,
      "language_loss": 0.88325226,
      "learning_rate": 2.8921739075269633e-06,
      "loss": 0.90468979,
      "num_input_tokens_seen": 133071530,
      "step": 6191,
      "time_per_iteration": 2.4879090785980225
    },
    {
      "auxiliary_loss_clip": 0.01087962,
      "auxiliary_loss_mlp": 0.01040563,
      "balance_loss_clip": 1.04809713,
      "balance_loss_mlp": 1.0217514,
      "epoch": 0.37228318052006615,
      "flos": 22674716023680.0,
      "grad_norm": 1.8694452632865397,
      "language_loss": 0.73956245,
      "learning_rate": 2.891825326449073e-06,
      "loss": 0.76084769,
      "num_input_tokens_seen": 133091410,
      "step": 6192,
      "time_per_iteration": 2.620514154434204
    },
    {
      "auxiliary_loss_clip": 0.01131628,
      "auxiliary_loss_mlp": 0.01036427,
      "balance_loss_clip": 1.04907668,
      "balance_loss_mlp": 1.0215435,
      "epoch": 0.3723433037727341,
      "flos": 25265491263360.0,
      "grad_norm": 1.9808759938291214,
      "language_loss": 0.79743671,
      "learning_rate": 2.8914767115530766e-06,
      "loss": 0.81911731,
      "num_input_tokens_seen": 133110365,
      "step": 6193,
      "time_per_iteration": 2.5447306632995605
    },
    {
      "auxiliary_loss_clip": 0.01103783,
      "auxiliary_loss_mlp": 0.01037544,
      "balance_loss_clip": 1.05198598,
      "balance_loss_mlp": 1.02214253,
      "epoch": 0.3724034270254021,
      "flos": 10524664333440.0,
      "grad_norm": 1.7724769803597884,
      "language_loss": 0.84527779,
      "learning_rate": 2.891128062852194e-06,
      "loss": 0.86669111,
      "num_input_tokens_seen": 133128255,
      "step": 6194,
      "time_per_iteration": 2.5395092964172363
    },
    {
      "auxiliary_loss_clip": 0.01108159,
      "auxiliary_loss_mlp": 0.01036757,
      "balance_loss_clip": 1.04802561,
      "balance_loss_mlp": 1.02194512,
      "epoch": 0.37246355027807004,
      "flos": 20266223328000.0,
      "grad_norm": 2.442968745001766,
      "language_loss": 0.76835519,
      "learning_rate": 2.890779380359646e-06,
      "loss": 0.78980434,
      "num_input_tokens_seen": 133143975,
      "step": 6195,
      "time_per_iteration": 2.5289199352264404
    },
    {
      "auxiliary_loss_clip": 0.01111618,
      "auxiliary_loss_mlp": 0.01036352,
      "balance_loss_clip": 1.05036402,
      "balance_loss_mlp": 1.02086723,
      "epoch": 0.372523673530738,
      "flos": 19500571998720.0,
      "grad_norm": 1.4854709748001063,
      "language_loss": 0.79145384,
      "learning_rate": 2.890430664088655e-06,
      "loss": 0.81293356,
      "num_input_tokens_seen": 133162935,
      "step": 6196,
      "time_per_iteration": 2.545846939086914
    },
    {
      "auxiliary_loss_clip": 0.01122084,
      "auxiliary_loss_mlp": 0.0103674,
      "balance_loss_clip": 1.05172873,
      "balance_loss_mlp": 1.0222795,
      "epoch": 0.372583796783406,
      "flos": 16764250849920.0,
      "grad_norm": 2.100293677098559,
      "language_loss": 0.83406287,
      "learning_rate": 2.890081914052443e-06,
      "loss": 0.85565114,
      "num_input_tokens_seen": 133181180,
      "step": 6197,
      "time_per_iteration": 2.4989984035491943
    },
    {
      "auxiliary_loss_clip": 0.01129313,
      "auxiliary_loss_mlp": 0.01038445,
      "balance_loss_clip": 1.04827738,
      "balance_loss_mlp": 1.02229226,
      "epoch": 0.37264392003607394,
      "flos": 22637979388800.0,
      "grad_norm": 1.5257480938126535,
      "language_loss": 0.64560902,
      "learning_rate": 2.889733130264237e-06,
      "loss": 0.66728663,
      "num_input_tokens_seen": 133199615,
      "step": 6198,
      "time_per_iteration": 2.4931671619415283
    },
    {
      "auxiliary_loss_clip": 0.01118013,
      "auxiliary_loss_mlp": 0.01046048,
      "balance_loss_clip": 1.04768467,
      "balance_loss_mlp": 1.03093278,
      "epoch": 0.3727040432887419,
      "flos": 19973120348160.0,
      "grad_norm": 1.4571669166566112,
      "language_loss": 0.73966074,
      "learning_rate": 2.889384312737261e-06,
      "loss": 0.7613014,
      "num_input_tokens_seen": 133219650,
      "step": 6199,
      "time_per_iteration": 2.520815134048462
    },
    {
      "auxiliary_loss_clip": 0.01103371,
      "auxiliary_loss_mlp": 0.0103755,
      "balance_loss_clip": 1.04834831,
      "balance_loss_mlp": 1.02288747,
      "epoch": 0.37276416654140987,
      "flos": 63899122279680.0,
      "grad_norm": 1.9782290599250099,
      "language_loss": 0.80747348,
      "learning_rate": 2.889035461484742e-06,
      "loss": 0.82888269,
      "num_input_tokens_seen": 133245675,
      "step": 6200,
      "time_per_iteration": 2.950160503387451
    },
    {
      "auxiliary_loss_clip": 0.01095082,
      "auxiliary_loss_mlp": 0.01044903,
      "balance_loss_clip": 1.05044675,
      "balance_loss_mlp": 1.02937007,
      "epoch": 0.37282428979407783,
      "flos": 39785970211200.0,
      "grad_norm": 1.8306444564735191,
      "language_loss": 0.60503137,
      "learning_rate": 2.88868657651991e-06,
      "loss": 0.62643123,
      "num_input_tokens_seen": 133266905,
      "step": 6201,
      "time_per_iteration": 2.7049427032470703
    },
    {
      "auxiliary_loss_clip": 0.01120597,
      "auxiliary_loss_mlp": 0.01038731,
      "balance_loss_clip": 1.04923463,
      "balance_loss_mlp": 1.02306104,
      "epoch": 0.37288441304674586,
      "flos": 22709046447360.0,
      "grad_norm": 2.0117258524656547,
      "language_loss": 0.731058,
      "learning_rate": 2.8883376578559934e-06,
      "loss": 0.75265127,
      "num_input_tokens_seen": 133286865,
      "step": 6202,
      "time_per_iteration": 2.5042881965637207
    },
    {
      "auxiliary_loss_clip": 0.01107577,
      "auxiliary_loss_mlp": 0.01037547,
      "balance_loss_clip": 1.04776192,
      "balance_loss_mlp": 1.02189493,
      "epoch": 0.3729445362994138,
      "flos": 18770292587520.0,
      "grad_norm": 65.31828046783563,
      "language_loss": 0.73965305,
      "learning_rate": 2.8879887055062243e-06,
      "loss": 0.76110429,
      "num_input_tokens_seen": 133305295,
      "step": 6203,
      "time_per_iteration": 2.5027801990509033
    },
    {
      "auxiliary_loss_clip": 0.01103091,
      "auxiliary_loss_mlp": 0.01036887,
      "balance_loss_clip": 1.04606581,
      "balance_loss_mlp": 1.0236547,
      "epoch": 0.3730046595520818,
      "flos": 22456199635200.0,
      "grad_norm": 1.7160359928723927,
      "language_loss": 0.81656098,
      "learning_rate": 2.8876397194838353e-06,
      "loss": 0.83796072,
      "num_input_tokens_seen": 133324625,
      "step": 6204,
      "time_per_iteration": 2.544421911239624
    },
    {
      "auxiliary_loss_clip": 0.01119054,
      "auxiliary_loss_mlp": 0.01044697,
      "balance_loss_clip": 1.04687309,
      "balance_loss_mlp": 1.02857959,
      "epoch": 0.37306478280474975,
      "flos": 24316372241280.0,
      "grad_norm": 1.6084665046265005,
      "language_loss": 0.75239861,
      "learning_rate": 2.8872906998020577e-06,
      "loss": 0.77403617,
      "num_input_tokens_seen": 133344625,
      "step": 6205,
      "time_per_iteration": 2.520479440689087
    },
    {
      "auxiliary_loss_clip": 0.0111358,
      "auxiliary_loss_mlp": 0.01038897,
      "balance_loss_clip": 1.04433298,
      "balance_loss_mlp": 1.02294719,
      "epoch": 0.3731249060574177,
      "flos": 15815167741440.0,
      "grad_norm": 1.8812788401078289,
      "language_loss": 0.7762931,
      "learning_rate": 2.886941646474128e-06,
      "loss": 0.79781789,
      "num_input_tokens_seen": 133363605,
      "step": 6206,
      "time_per_iteration": 2.489677667617798
    },
    {
      "auxiliary_loss_clip": 0.01129714,
      "auxiliary_loss_mlp": 0.01041245,
      "balance_loss_clip": 1.04675686,
      "balance_loss_mlp": 1.02500844,
      "epoch": 0.3731850293100857,
      "flos": 19828077229440.0,
      "grad_norm": 2.054049465341689,
      "language_loss": 0.933608,
      "learning_rate": 2.886592559513283e-06,
      "loss": 0.95531756,
      "num_input_tokens_seen": 133379405,
      "step": 6207,
      "time_per_iteration": 2.445352077484131
    },
    {
      "auxiliary_loss_clip": 0.01104449,
      "auxiliary_loss_mlp": 0.01032793,
      "balance_loss_clip": 1.04582262,
      "balance_loss_mlp": 1.01866043,
      "epoch": 0.37324515256275365,
      "flos": 19062354072960.0,
      "grad_norm": 2.142712149543232,
      "language_loss": 0.82495034,
      "learning_rate": 2.886243438932759e-06,
      "loss": 0.84632277,
      "num_input_tokens_seen": 133397585,
      "step": 6208,
      "time_per_iteration": 2.5405585765838623
    },
    {
      "auxiliary_loss_clip": 0.01117668,
      "auxiliary_loss_mlp": 0.01040207,
      "balance_loss_clip": 1.04415798,
      "balance_loss_mlp": 1.02254033,
      "epoch": 0.3733052758154216,
      "flos": 20704333512960.0,
      "grad_norm": 1.7871021478688158,
      "language_loss": 0.73395199,
      "learning_rate": 2.8858942847457953e-06,
      "loss": 0.75553071,
      "num_input_tokens_seen": 133415365,
      "step": 6209,
      "time_per_iteration": 2.486462116241455
    },
    {
      "auxiliary_loss_clip": 0.01091198,
      "auxiliary_loss_mlp": 0.01040027,
      "balance_loss_clip": 1.04810858,
      "balance_loss_mlp": 1.02345657,
      "epoch": 0.3733653990680896,
      "flos": 20193504243840.0,
      "grad_norm": 1.5434100125732917,
      "language_loss": 0.70263684,
      "learning_rate": 2.8855450969656305e-06,
      "loss": 0.72394907,
      "num_input_tokens_seen": 133435700,
      "step": 6210,
      "time_per_iteration": 2.5888261795043945
    },
    {
      "auxiliary_loss_clip": 0.01076959,
      "auxiliary_loss_mlp": 0.01043016,
      "balance_loss_clip": 1.04199803,
      "balance_loss_mlp": 1.02443099,
      "epoch": 0.37342552232075754,
      "flos": 20339660684160.0,
      "grad_norm": 1.8465963146904167,
      "language_loss": 0.7786001,
      "learning_rate": 2.8851958756055073e-06,
      "loss": 0.79979992,
      "num_input_tokens_seen": 133455180,
      "step": 6211,
      "time_per_iteration": 2.570037364959717
    },
    {
      "auxiliary_loss_clip": 0.01119817,
      "auxiliary_loss_mlp": 0.01036586,
      "balance_loss_clip": 1.04642391,
      "balance_loss_mlp": 1.02129149,
      "epoch": 0.3734856455734255,
      "flos": 35517879527040.0,
      "grad_norm": 2.9647646385718556,
      "language_loss": 0.72990268,
      "learning_rate": 2.884846620678668e-06,
      "loss": 0.75146675,
      "num_input_tokens_seen": 133476715,
      "step": 6212,
      "time_per_iteration": 2.6347734928131104
    },
    {
      "auxiliary_loss_clip": 0.01123409,
      "auxiliary_loss_mlp": 0.01046533,
      "balance_loss_clip": 1.05011392,
      "balance_loss_mlp": 1.0297718,
      "epoch": 0.37354576882609347,
      "flos": 21142300043520.0,
      "grad_norm": 2.045015955024548,
      "language_loss": 0.82440293,
      "learning_rate": 2.884497332198356e-06,
      "loss": 0.84610236,
      "num_input_tokens_seen": 133494550,
      "step": 6213,
      "time_per_iteration": 2.4902937412261963
    },
    {
      "auxiliary_loss_clip": 0.01089828,
      "auxiliary_loss_mlp": 0.01049073,
      "balance_loss_clip": 1.04805899,
      "balance_loss_mlp": 1.03126311,
      "epoch": 0.37360589207876144,
      "flos": 21506793304320.0,
      "grad_norm": 2.0570717061013024,
      "language_loss": 0.78883684,
      "learning_rate": 2.8841480101778167e-06,
      "loss": 0.81022584,
      "num_input_tokens_seen": 133512640,
      "step": 6214,
      "time_per_iteration": 3.9945590496063232
    },
    {
      "auxiliary_loss_clip": 0.01106074,
      "auxiliary_loss_mlp": 0.0104574,
      "balance_loss_clip": 1.04477286,
      "balance_loss_mlp": 1.03009999,
      "epoch": 0.37366601533142946,
      "flos": 38435800861440.0,
      "grad_norm": 1.6818093051978518,
      "language_loss": 0.84807032,
      "learning_rate": 2.883798654630296e-06,
      "loss": 0.86958849,
      "num_input_tokens_seen": 133535540,
      "step": 6215,
      "time_per_iteration": 2.683483600616455
    },
    {
      "auxiliary_loss_clip": 0.01098076,
      "auxiliary_loss_mlp": 0.01038863,
      "balance_loss_clip": 1.04696012,
      "balance_loss_mlp": 1.02167273,
      "epoch": 0.3737261385840974,
      "flos": 18441171244800.0,
      "grad_norm": 1.5715619083787566,
      "language_loss": 0.68114662,
      "learning_rate": 2.8834492655690423e-06,
      "loss": 0.70251596,
      "num_input_tokens_seen": 133555795,
      "step": 6216,
      "time_per_iteration": 2.5813605785369873
    },
    {
      "auxiliary_loss_clip": 0.0111467,
      "auxiliary_loss_mlp": 0.01044231,
      "balance_loss_clip": 1.04746246,
      "balance_loss_mlp": 1.02737439,
      "epoch": 0.3737862618367654,
      "flos": 22929861306240.0,
      "grad_norm": 3.069618728397449,
      "language_loss": 0.66158378,
      "learning_rate": 2.883099843007303e-06,
      "loss": 0.68317282,
      "num_input_tokens_seen": 133575905,
      "step": 6217,
      "time_per_iteration": 5.353607892990112
    },
    {
      "auxiliary_loss_clip": 0.01108809,
      "auxiliary_loss_mlp": 0.01041678,
      "balance_loss_clip": 1.04504848,
      "balance_loss_mlp": 1.02515006,
      "epoch": 0.37384638508943335,
      "flos": 15409664127360.0,
      "grad_norm": 1.9000723610627124,
      "language_loss": 0.80299687,
      "learning_rate": 2.88275038695833e-06,
      "loss": 0.82450175,
      "num_input_tokens_seen": 133592585,
      "step": 6218,
      "time_per_iteration": 2.5368142127990723
    },
    {
      "auxiliary_loss_clip": 0.01115797,
      "auxiliary_loss_mlp": 0.0103892,
      "balance_loss_clip": 1.04780459,
      "balance_loss_mlp": 1.02339911,
      "epoch": 0.3739065083421013,
      "flos": 24280820755200.0,
      "grad_norm": 1.363463130915226,
      "language_loss": 0.78892559,
      "learning_rate": 2.8824008974353736e-06,
      "loss": 0.81047279,
      "num_input_tokens_seen": 133615070,
      "step": 6219,
      "time_per_iteration": 2.5432002544403076
    },
    {
      "auxiliary_loss_clip": 0.01106285,
      "auxiliary_loss_mlp": 0.01040331,
      "balance_loss_clip": 1.04747176,
      "balance_loss_mlp": 1.02401137,
      "epoch": 0.3739666315947693,
      "flos": 23002831785600.0,
      "grad_norm": 1.6069880871280953,
      "language_loss": 0.76771057,
      "learning_rate": 2.8820513744516866e-06,
      "loss": 0.78917676,
      "num_input_tokens_seen": 133633490,
      "step": 6220,
      "time_per_iteration": 3.923288583755493
    },
    {
      "auxiliary_loss_clip": 0.01099475,
      "auxiliary_loss_mlp": 0.01042083,
      "balance_loss_clip": 1.04518533,
      "balance_loss_mlp": 1.02594173,
      "epoch": 0.37402675484743725,
      "flos": 19391116279680.0,
      "grad_norm": 1.626671846030633,
      "language_loss": 0.8297711,
      "learning_rate": 2.8817018180205235e-06,
      "loss": 0.85118669,
      "num_input_tokens_seen": 133653425,
      "step": 6221,
      "time_per_iteration": 2.5538079738616943
    },
    {
      "auxiliary_loss_clip": 0.01105206,
      "auxiliary_loss_mlp": 0.01042685,
      "balance_loss_clip": 1.04576468,
      "balance_loss_mlp": 1.02659166,
      "epoch": 0.3740868781001052,
      "flos": 17126158331520.0,
      "grad_norm": 1.6678666592752713,
      "language_loss": 0.76087523,
      "learning_rate": 2.8813522281551387e-06,
      "loss": 0.78235412,
      "num_input_tokens_seen": 133670220,
      "step": 6222,
      "time_per_iteration": 2.4975829124450684
    },
    {
      "auxiliary_loss_clip": 0.01098838,
      "auxiliary_loss_mlp": 0.00796666,
      "balance_loss_clip": 1.04681945,
      "balance_loss_mlp": 1.01915288,
      "epoch": 0.3741470013527732,
      "flos": 20043505048320.0,
      "grad_norm": 1.772820108538877,
      "language_loss": 0.70469058,
      "learning_rate": 2.881002604868789e-06,
      "loss": 0.72364557,
      "num_input_tokens_seen": 133688910,
      "step": 6223,
      "time_per_iteration": 2.565131902694702
    },
    {
      "auxiliary_loss_clip": 0.01100623,
      "auxiliary_loss_mlp": 0.01038299,
      "balance_loss_clip": 1.05081987,
      "balance_loss_mlp": 1.02295661,
      "epoch": 0.37420712460544114,
      "flos": 36897279569280.0,
      "grad_norm": 1.7453910489649678,
      "language_loss": 0.68595231,
      "learning_rate": 2.8806529481747325e-06,
      "loss": 0.70734155,
      "num_input_tokens_seen": 133708690,
      "step": 6224,
      "time_per_iteration": 2.6873717308044434
    },
    {
      "auxiliary_loss_clip": 0.01087909,
      "auxiliary_loss_mlp": 0.01040287,
      "balance_loss_clip": 1.05065513,
      "balance_loss_mlp": 1.02411032,
      "epoch": 0.3742672478581091,
      "flos": 22201198007040.0,
      "grad_norm": 1.9574342831272435,
      "language_loss": 0.70090348,
      "learning_rate": 2.880303258086228e-06,
      "loss": 0.72218549,
      "num_input_tokens_seen": 133728095,
      "step": 6225,
      "time_per_iteration": 2.600332736968994
    },
    {
      "auxiliary_loss_clip": 0.01089229,
      "auxiliary_loss_mlp": 0.01044308,
      "balance_loss_clip": 1.05053711,
      "balance_loss_mlp": 1.02728522,
      "epoch": 0.3743273711107771,
      "flos": 24681547860480.0,
      "grad_norm": 2.014343073937966,
      "language_loss": 0.79092574,
      "learning_rate": 2.879953534616536e-06,
      "loss": 0.8122611,
      "num_input_tokens_seen": 133745590,
      "step": 6226,
      "time_per_iteration": 2.6541037559509277
    },
    {
      "auxiliary_loss_clip": 0.01102335,
      "auxiliary_loss_mlp": 0.01039273,
      "balance_loss_clip": 1.04608703,
      "balance_loss_mlp": 1.02276278,
      "epoch": 0.37438749436344504,
      "flos": 24459619680000.0,
      "grad_norm": 1.7546461621204505,
      "language_loss": 0.67802161,
      "learning_rate": 2.879603777778917e-06,
      "loss": 0.69943774,
      "num_input_tokens_seen": 133766155,
      "step": 6227,
      "time_per_iteration": 2.5752339363098145
    },
    {
      "auxiliary_loss_clip": 0.01094612,
      "auxiliary_loss_mlp": 0.01037112,
      "balance_loss_clip": 1.04520726,
      "balance_loss_mlp": 1.02154374,
      "epoch": 0.374447617616113,
      "flos": 21798747048960.0,
      "grad_norm": 1.7920884735946696,
      "language_loss": 0.82864285,
      "learning_rate": 2.879253987586635e-06,
      "loss": 0.84996015,
      "num_input_tokens_seen": 133783185,
      "step": 6228,
      "time_per_iteration": 2.5617477893829346
    },
    {
      "auxiliary_loss_clip": 0.01086653,
      "auxiliary_loss_mlp": 0.01043546,
      "balance_loss_clip": 1.0460999,
      "balance_loss_mlp": 1.02516413,
      "epoch": 0.374507740868781,
      "flos": 17968191932160.0,
      "grad_norm": 1.583497750276829,
      "language_loss": 0.7457462,
      "learning_rate": 2.8789041640529535e-06,
      "loss": 0.76704818,
      "num_input_tokens_seen": 133800975,
      "step": 6229,
      "time_per_iteration": 2.5277609825134277
    },
    {
      "auxiliary_loss_clip": 0.01091993,
      "auxiliary_loss_mlp": 0.01035731,
      "balance_loss_clip": 1.04557741,
      "balance_loss_mlp": 1.0190711,
      "epoch": 0.374567864121449,
      "flos": 16105828596480.0,
      "grad_norm": 1.766833369613947,
      "language_loss": 0.83750767,
      "learning_rate": 2.8785543071911383e-06,
      "loss": 0.85878491,
      "num_input_tokens_seen": 133818020,
      "step": 6230,
      "time_per_iteration": 2.5546700954437256
    },
    {
      "auxiliary_loss_clip": 0.01119678,
      "auxiliary_loss_mlp": 0.01039324,
      "balance_loss_clip": 1.04836047,
      "balance_loss_mlp": 1.02308178,
      "epoch": 0.37462798737411696,
      "flos": 25773160135680.0,
      "grad_norm": 1.8295478060335288,
      "language_loss": 0.73523349,
      "learning_rate": 2.878204417014456e-06,
      "loss": 0.75682348,
      "num_input_tokens_seen": 133840690,
      "step": 6231,
      "time_per_iteration": 2.569983959197998
    },
    {
      "auxiliary_loss_clip": 0.01117791,
      "auxiliary_loss_mlp": 0.0104465,
      "balance_loss_clip": 1.04973459,
      "balance_loss_mlp": 1.02810335,
      "epoch": 0.3746881106267849,
      "flos": 16654507822080.0,
      "grad_norm": 1.9600159468004925,
      "language_loss": 0.73595011,
      "learning_rate": 2.8778544935361735e-06,
      "loss": 0.75757444,
      "num_input_tokens_seen": 133858350,
      "step": 6232,
      "time_per_iteration": 2.4946014881134033
    },
    {
      "auxiliary_loss_clip": 0.01104005,
      "auxiliary_loss_mlp": 0.01036216,
      "balance_loss_clip": 1.04822969,
      "balance_loss_mlp": 1.01997972,
      "epoch": 0.3747482338794529,
      "flos": 26177981391360.0,
      "grad_norm": 1.5906641874612768,
      "language_loss": 0.76665324,
      "learning_rate": 2.877504536769561e-06,
      "loss": 0.78805548,
      "num_input_tokens_seen": 133879775,
      "step": 6233,
      "time_per_iteration": 2.5760295391082764
    },
    {
      "auxiliary_loss_clip": 0.01112101,
      "auxiliary_loss_mlp": 0.01039156,
      "balance_loss_clip": 1.04833937,
      "balance_loss_mlp": 1.02334929,
      "epoch": 0.37480835713212085,
      "flos": 12021061950720.0,
      "grad_norm": 1.702172753597285,
      "language_loss": 0.6955781,
      "learning_rate": 2.8771545467278883e-06,
      "loss": 0.71709067,
      "num_input_tokens_seen": 133898295,
      "step": 6234,
      "time_per_iteration": 2.503519296646118
    },
    {
      "auxiliary_loss_clip": 0.01118139,
      "auxiliary_loss_mlp": 0.01041922,
      "balance_loss_clip": 1.04639304,
      "balance_loss_mlp": 1.02698493,
      "epoch": 0.3748684803847888,
      "flos": 19679263182720.0,
      "grad_norm": 2.0696472573382305,
      "language_loss": 0.82310772,
      "learning_rate": 2.8768045234244276e-06,
      "loss": 0.84470832,
      "num_input_tokens_seen": 133915230,
      "step": 6235,
      "time_per_iteration": 2.476473331451416
    },
    {
      "auxiliary_loss_clip": 0.01135616,
      "auxiliary_loss_mlp": 0.01034852,
      "balance_loss_clip": 1.05104566,
      "balance_loss_mlp": 1.01991546,
      "epoch": 0.3749286036374568,
      "flos": 20521189042560.0,
      "grad_norm": 1.8616623950776758,
      "language_loss": 0.77790743,
      "learning_rate": 2.8764544668724517e-06,
      "loss": 0.79961216,
      "num_input_tokens_seen": 133934110,
      "step": 6236,
      "time_per_iteration": 2.4650447368621826
    },
    {
      "auxiliary_loss_clip": 0.0111614,
      "auxiliary_loss_mlp": 0.01051829,
      "balance_loss_clip": 1.04723728,
      "balance_loss_mlp": 1.03364944,
      "epoch": 0.37498872689012475,
      "flos": 20704620821760.0,
      "grad_norm": 1.9064991559087816,
      "language_loss": 0.73634315,
      "learning_rate": 2.876104377085234e-06,
      "loss": 0.75802284,
      "num_input_tokens_seen": 133952395,
      "step": 6237,
      "time_per_iteration": 2.4832520484924316
    },
    {
      "auxiliary_loss_clip": 0.01116667,
      "auxiliary_loss_mlp": 0.00793317,
      "balance_loss_clip": 1.04690838,
      "balance_loss_mlp": 1.01313376,
      "epoch": 0.3750488501427927,
      "flos": 21574843620480.0,
      "grad_norm": 2.3171794261531415,
      "language_loss": 0.92827779,
      "learning_rate": 2.8757542540760508e-06,
      "loss": 0.94737768,
      "num_input_tokens_seen": 133969635,
      "step": 6238,
      "time_per_iteration": 2.6572070121765137
    },
    {
      "auxiliary_loss_clip": 0.01134451,
      "auxiliary_loss_mlp": 0.01037387,
      "balance_loss_clip": 1.05003262,
      "balance_loss_mlp": 1.02093649,
      "epoch": 0.3751089733954607,
      "flos": 15923869274880.0,
      "grad_norm": 1.8686878963825586,
      "language_loss": 0.70789278,
      "learning_rate": 2.8754040978581777e-06,
      "loss": 0.72961116,
      "num_input_tokens_seen": 133987215,
      "step": 6239,
      "time_per_iteration": 2.440011978149414
    },
    {
      "auxiliary_loss_clip": 0.01066007,
      "auxiliary_loss_mlp": 0.01033924,
      "balance_loss_clip": 1.04909396,
      "balance_loss_mlp": 1.01823628,
      "epoch": 0.37516909664812864,
      "flos": 36284644177920.0,
      "grad_norm": 1.5485465262125777,
      "language_loss": 0.65509343,
      "learning_rate": 2.875053908444895e-06,
      "loss": 0.67609268,
      "num_input_tokens_seen": 134009250,
      "step": 6240,
      "time_per_iteration": 2.7622854709625244
    },
    {
      "auxiliary_loss_clip": 0.01100543,
      "auxiliary_loss_mlp": 0.00791571,
      "balance_loss_clip": 1.0493803,
      "balance_loss_mlp": 1.01212049,
      "epoch": 0.3752292199007966,
      "flos": 13515915283200.0,
      "grad_norm": 1.6952200620092974,
      "language_loss": 0.76041299,
      "learning_rate": 2.8747036858494795e-06,
      "loss": 0.77933413,
      "num_input_tokens_seen": 134026875,
      "step": 6241,
      "time_per_iteration": 2.5395147800445557
    },
    {
      "auxiliary_loss_clip": 0.01103691,
      "auxiliary_loss_mlp": 0.01043399,
      "balance_loss_clip": 1.04971802,
      "balance_loss_mlp": 1.02612603,
      "epoch": 0.3752893431534646,
      "flos": 27198095644800.0,
      "grad_norm": 1.9018981925988414,
      "language_loss": 0.83181274,
      "learning_rate": 2.874353430085213e-06,
      "loss": 0.85328364,
      "num_input_tokens_seen": 134047185,
      "step": 6242,
      "time_per_iteration": 2.627531051635742
    },
    {
      "auxiliary_loss_clip": 0.01107131,
      "auxiliary_loss_mlp": 0.01039731,
      "balance_loss_clip": 1.05004978,
      "balance_loss_mlp": 1.02490747,
      "epoch": 0.3753494664061326,
      "flos": 30007674581760.0,
      "grad_norm": 2.1023141243268477,
      "language_loss": 0.68689179,
      "learning_rate": 2.8740031411653766e-06,
      "loss": 0.70836037,
      "num_input_tokens_seen": 134067330,
      "step": 6243,
      "time_per_iteration": 2.599716901779175
    },
    {
      "auxiliary_loss_clip": 0.0106202,
      "auxiliary_loss_mlp": 0.00797043,
      "balance_loss_clip": 1.04775429,
      "balance_loss_mlp": 1.01548696,
      "epoch": 0.37540958965880056,
      "flos": 24461954064000.0,
      "grad_norm": 1.744805036867949,
      "language_loss": 0.83688414,
      "learning_rate": 2.8736528191032535e-06,
      "loss": 0.85547471,
      "num_input_tokens_seen": 134085525,
      "step": 6244,
      "time_per_iteration": 2.677353858947754
    },
    {
      "auxiliary_loss_clip": 0.0107392,
      "auxiliary_loss_mlp": 0.01038159,
      "balance_loss_clip": 1.04321206,
      "balance_loss_mlp": 1.02231646,
      "epoch": 0.3754697129114685,
      "flos": 16508387295360.0,
      "grad_norm": 2.7412831991931315,
      "language_loss": 0.83219916,
      "learning_rate": 2.8733024639121277e-06,
      "loss": 0.85331994,
      "num_input_tokens_seen": 134101855,
      "step": 6245,
      "time_per_iteration": 2.5703930854797363
    },
    {
      "auxiliary_loss_clip": 0.01097876,
      "auxiliary_loss_mlp": 0.01042071,
      "balance_loss_clip": 1.04652119,
      "balance_loss_mlp": 1.02556086,
      "epoch": 0.3755298361641365,
      "flos": 19390900798080.0,
      "grad_norm": 2.074386617411249,
      "language_loss": 0.64061356,
      "learning_rate": 2.8729520756052853e-06,
      "loss": 0.66201305,
      "num_input_tokens_seen": 134119360,
      "step": 6246,
      "time_per_iteration": 2.5168678760528564
    },
    {
      "auxiliary_loss_clip": 0.01104083,
      "auxiliary_loss_mlp": 0.01042972,
      "balance_loss_clip": 1.04617643,
      "balance_loss_mlp": 1.02581763,
      "epoch": 0.37558995941680445,
      "flos": 14720395069440.0,
      "grad_norm": 1.7884218680268675,
      "language_loss": 0.74792361,
      "learning_rate": 2.8726016541960124e-06,
      "loss": 0.76939416,
      "num_input_tokens_seen": 134137475,
      "step": 6247,
      "time_per_iteration": 2.522047758102417
    },
    {
      "auxiliary_loss_clip": 0.01120126,
      "auxiliary_loss_mlp": 0.01039532,
      "balance_loss_clip": 1.04701865,
      "balance_loss_mlp": 1.02402294,
      "epoch": 0.3756500826694724,
      "flos": 21689901861120.0,
      "grad_norm": 3.0856778317128897,
      "language_loss": 0.54808867,
      "learning_rate": 2.872251199697598e-06,
      "loss": 0.56968528,
      "num_input_tokens_seen": 134154580,
      "step": 6248,
      "time_per_iteration": 2.485595941543579
    },
    {
      "auxiliary_loss_clip": 0.01115229,
      "auxiliary_loss_mlp": 0.0104264,
      "balance_loss_clip": 1.0488677,
      "balance_loss_mlp": 1.02684438,
      "epoch": 0.3757102059221404,
      "flos": 26505666190080.0,
      "grad_norm": 1.753143259254582,
      "language_loss": 0.84349644,
      "learning_rate": 2.8719007121233297e-06,
      "loss": 0.86507517,
      "num_input_tokens_seen": 134174285,
      "step": 6249,
      "time_per_iteration": 2.55735182762146
    },
    {
      "auxiliary_loss_clip": 0.01100947,
      "auxiliary_loss_mlp": 0.01033177,
      "balance_loss_clip": 1.04589891,
      "balance_loss_mlp": 1.01838338,
      "epoch": 0.37577032917480835,
      "flos": 37338083274240.0,
      "grad_norm": 1.580438824779369,
      "language_loss": 0.6800071,
      "learning_rate": 2.8715501914864993e-06,
      "loss": 0.7013483,
      "num_input_tokens_seen": 134195940,
      "step": 6250,
      "time_per_iteration": 2.66099214553833
    },
    {
      "auxiliary_loss_clip": 0.01108767,
      "auxiliary_loss_mlp": 0.01041538,
      "balance_loss_clip": 1.04586327,
      "balance_loss_mlp": 1.02639866,
      "epoch": 0.3758304524274763,
      "flos": 21908597817600.0,
      "grad_norm": 2.007635406736307,
      "language_loss": 0.77889752,
      "learning_rate": 2.8711996378003987e-06,
      "loss": 0.80040056,
      "num_input_tokens_seen": 134212235,
      "step": 6251,
      "time_per_iteration": 2.528585433959961
    },
    {
      "auxiliary_loss_clip": 0.0111404,
      "auxiliary_loss_mlp": 0.01032749,
      "balance_loss_clip": 1.04653811,
      "balance_loss_mlp": 1.01793134,
      "epoch": 0.3758905756801443,
      "flos": 36569343375360.0,
      "grad_norm": 2.0716536852235605,
      "language_loss": 0.57905209,
      "learning_rate": 2.8708490510783203e-06,
      "loss": 0.60052001,
      "num_input_tokens_seen": 134233810,
      "step": 6252,
      "time_per_iteration": 2.611238479614258
    },
    {
      "auxiliary_loss_clip": 0.01111295,
      "auxiliary_loss_mlp": 0.0104386,
      "balance_loss_clip": 1.04842138,
      "balance_loss_mlp": 1.02744532,
      "epoch": 0.37595069893281224,
      "flos": 24528783317760.0,
      "grad_norm": 1.7232426246631678,
      "language_loss": 0.89264846,
      "learning_rate": 2.8704984313335584e-06,
      "loss": 0.91420007,
      "num_input_tokens_seen": 134252020,
      "step": 6253,
      "time_per_iteration": 4.090145111083984
    },
    {
      "auxiliary_loss_clip": 0.01093515,
      "auxiliary_loss_mlp": 0.01033542,
      "balance_loss_clip": 1.05030608,
      "balance_loss_mlp": 1.01960683,
      "epoch": 0.3760108221854802,
      "flos": 16435021766400.0,
      "grad_norm": 3.9400596879147747,
      "language_loss": 0.76291358,
      "learning_rate": 2.8701477785794097e-06,
      "loss": 0.78418422,
      "num_input_tokens_seen": 134269495,
      "step": 6254,
      "time_per_iteration": 2.5708200931549072
    },
    {
      "auxiliary_loss_clip": 0.01092638,
      "auxiliary_loss_mlp": 0.01042772,
      "balance_loss_clip": 1.04659951,
      "balance_loss_mlp": 1.02662468,
      "epoch": 0.37607094543814823,
      "flos": 13771742924160.0,
      "grad_norm": 2.3121607274902067,
      "language_loss": 0.62340218,
      "learning_rate": 2.869797092829169e-06,
      "loss": 0.64475632,
      "num_input_tokens_seen": 134287035,
      "step": 6255,
      "time_per_iteration": 5.3299009799957275
    },
    {
      "auxiliary_loss_clip": 0.0112189,
      "auxiliary_loss_mlp": 0.01038225,
      "balance_loss_clip": 1.04637432,
      "balance_loss_mlp": 1.02175021,
      "epoch": 0.3761310686908162,
      "flos": 19857918453120.0,
      "grad_norm": 2.4392467552534787,
      "language_loss": 0.73614573,
      "learning_rate": 2.869446374096135e-06,
      "loss": 0.75774688,
      "num_input_tokens_seen": 134304840,
      "step": 6256,
      "time_per_iteration": 2.526726484298706
    },
    {
      "auxiliary_loss_clip": 0.01122089,
      "auxiliary_loss_mlp": 0.01044185,
      "balance_loss_clip": 1.04823148,
      "balance_loss_mlp": 1.02805591,
      "epoch": 0.37619119194348416,
      "flos": 12750802657920.0,
      "grad_norm": 1.7198429314458412,
      "language_loss": 0.70364141,
      "learning_rate": 2.8690956223936088e-06,
      "loss": 0.72530425,
      "num_input_tokens_seen": 134323180,
      "step": 6257,
      "time_per_iteration": 2.5043623447418213
    },
    {
      "auxiliary_loss_clip": 0.01104503,
      "auxiliary_loss_mlp": 0.01035037,
      "balance_loss_clip": 1.04584253,
      "balance_loss_mlp": 1.02045774,
      "epoch": 0.3762513151961521,
      "flos": 17530548624000.0,
      "grad_norm": 1.5685645837260644,
      "language_loss": 0.8428368,
      "learning_rate": 2.868744837734889e-06,
      "loss": 0.86423224,
      "num_input_tokens_seen": 134341390,
      "step": 6258,
      "time_per_iteration": 3.903674840927124
    },
    {
      "auxiliary_loss_clip": 0.01089184,
      "auxiliary_loss_mlp": 0.01038737,
      "balance_loss_clip": 1.04634249,
      "balance_loss_mlp": 1.02480149,
      "epoch": 0.3763114384488201,
      "flos": 23617406511360.0,
      "grad_norm": 1.5041725872815703,
      "language_loss": 0.80803055,
      "learning_rate": 2.868394020133277e-06,
      "loss": 0.82930976,
      "num_input_tokens_seen": 134360425,
      "step": 6259,
      "time_per_iteration": 2.6423726081848145
    },
    {
      "auxiliary_loss_clip": 0.01090628,
      "auxiliary_loss_mlp": 0.01043351,
      "balance_loss_clip": 1.04696751,
      "balance_loss_mlp": 1.02688813,
      "epoch": 0.37637156170148806,
      "flos": 25406978935680.0,
      "grad_norm": 1.9049728775330024,
      "language_loss": 0.71512866,
      "learning_rate": 2.8680431696020783e-06,
      "loss": 0.73646843,
      "num_input_tokens_seen": 134379775,
      "step": 6260,
      "time_per_iteration": 2.652569055557251
    },
    {
      "auxiliary_loss_clip": 0.01106546,
      "auxiliary_loss_mlp": 0.01041718,
      "balance_loss_clip": 1.04931712,
      "balance_loss_mlp": 1.02593446,
      "epoch": 0.376431684954156,
      "flos": 23440906056960.0,
      "grad_norm": 2.05590790519657,
      "language_loss": 0.78053373,
      "learning_rate": 2.867692286154594e-06,
      "loss": 0.80201638,
      "num_input_tokens_seen": 134400315,
      "step": 6261,
      "time_per_iteration": 2.556382656097412
    },
    {
      "auxiliary_loss_clip": 0.01107691,
      "auxiliary_loss_mlp": 0.01047423,
      "balance_loss_clip": 1.04621136,
      "balance_loss_mlp": 1.03013778,
      "epoch": 0.376491808206824,
      "flos": 34204482725760.0,
      "grad_norm": 2.120875862700909,
      "language_loss": 0.80196905,
      "learning_rate": 2.867341369804132e-06,
      "loss": 0.82352018,
      "num_input_tokens_seen": 134422875,
      "step": 6262,
      "time_per_iteration": 2.695587635040283
    },
    {
      "auxiliary_loss_clip": 0.01109861,
      "auxiliary_loss_mlp": 0.01031933,
      "balance_loss_clip": 1.04424179,
      "balance_loss_mlp": 1.01681685,
      "epoch": 0.37655193145949195,
      "flos": 35185669614720.0,
      "grad_norm": 1.8047915016743075,
      "language_loss": 0.80094469,
      "learning_rate": 2.866990420563998e-06,
      "loss": 0.82236266,
      "num_input_tokens_seen": 134443025,
      "step": 6263,
      "time_per_iteration": 2.6193580627441406
    },
    {
      "auxiliary_loss_clip": 0.01132911,
      "auxiliary_loss_mlp": 0.01041241,
      "balance_loss_clip": 1.0491643,
      "balance_loss_mlp": 1.02625036,
      "epoch": 0.3766120547121599,
      "flos": 16761844638720.0,
      "grad_norm": 4.420616055258283,
      "language_loss": 0.79662311,
      "learning_rate": 2.866639438447501e-06,
      "loss": 0.81836468,
      "num_input_tokens_seen": 134460945,
      "step": 6264,
      "time_per_iteration": 2.5011255741119385
    },
    {
      "auxiliary_loss_clip": 0.01128449,
      "auxiliary_loss_mlp": 0.0104237,
      "balance_loss_clip": 1.04519773,
      "balance_loss_mlp": 1.02742743,
      "epoch": 0.3766721779648279,
      "flos": 23550361776000.0,
      "grad_norm": 1.87233426377101,
      "language_loss": 0.73618948,
      "learning_rate": 2.8662884234679497e-06,
      "loss": 0.75789762,
      "num_input_tokens_seen": 134480440,
      "step": 6265,
      "time_per_iteration": 2.5088727474212646
    },
    {
      "auxiliary_loss_clip": 0.01118019,
      "auxiliary_loss_mlp": 0.01037803,
      "balance_loss_clip": 1.04867411,
      "balance_loss_mlp": 1.02399266,
      "epoch": 0.37673230121749585,
      "flos": 29129191655040.0,
      "grad_norm": 1.633533319749515,
      "language_loss": 0.69109797,
      "learning_rate": 2.865937375638654e-06,
      "loss": 0.71265614,
      "num_input_tokens_seen": 134501110,
      "step": 6266,
      "time_per_iteration": 2.5804193019866943
    },
    {
      "auxiliary_loss_clip": 0.01122951,
      "auxiliary_loss_mlp": 0.01042701,
      "balance_loss_clip": 1.04635,
      "balance_loss_mlp": 1.0270251,
      "epoch": 0.3767924244701638,
      "flos": 28146783703680.0,
      "grad_norm": 2.777640661537264,
      "language_loss": 0.63228559,
      "learning_rate": 2.8655862949729264e-06,
      "loss": 0.65394211,
      "num_input_tokens_seen": 134522460,
      "step": 6267,
      "time_per_iteration": 2.567915678024292
    },
    {
      "auxiliary_loss_clip": 0.01038252,
      "auxiliary_loss_mlp": 0.01026883,
      "balance_loss_clip": 1.01935005,
      "balance_loss_mlp": 1.02493978,
      "epoch": 0.37685254772283183,
      "flos": 60797197526400.0,
      "grad_norm": 0.7239239406601489,
      "language_loss": 0.58891565,
      "learning_rate": 2.8652351814840795e-06,
      "loss": 0.60956699,
      "num_input_tokens_seen": 134589545,
      "step": 6268,
      "time_per_iteration": 3.197960615158081
    },
    {
      "auxiliary_loss_clip": 0.01131356,
      "auxiliary_loss_mlp": 0.01042617,
      "balance_loss_clip": 1.04701471,
      "balance_loss_mlp": 1.02665532,
      "epoch": 0.3769126709754998,
      "flos": 26032543223040.0,
      "grad_norm": 1.5727022559127346,
      "language_loss": 0.64881259,
      "learning_rate": 2.8648840351854283e-06,
      "loss": 0.67055225,
      "num_input_tokens_seen": 134610550,
      "step": 6269,
      "time_per_iteration": 2.5015296936035156
    },
    {
      "auxiliary_loss_clip": 0.01096529,
      "auxiliary_loss_mlp": 0.01033376,
      "balance_loss_clip": 1.04855013,
      "balance_loss_mlp": 1.01753354,
      "epoch": 0.37697279422816776,
      "flos": 23579879777280.0,
      "grad_norm": 1.5389491578568593,
      "language_loss": 0.70678443,
      "learning_rate": 2.8645328560902874e-06,
      "loss": 0.72808349,
      "num_input_tokens_seen": 134630485,
      "step": 6270,
      "time_per_iteration": 2.592176914215088
    },
    {
      "auxiliary_loss_clip": 0.0104741,
      "auxiliary_loss_mlp": 0.01002182,
      "balance_loss_clip": 1.01862073,
      "balance_loss_mlp": 1.00044107,
      "epoch": 0.3770329174808357,
      "flos": 64745935367040.0,
      "grad_norm": 0.7116894402980682,
      "language_loss": 0.5617196,
      "learning_rate": 2.8641816442119746e-06,
      "loss": 0.58221549,
      "num_input_tokens_seen": 134693510,
      "step": 6271,
      "time_per_iteration": 3.0768022537231445
    },
    {
      "auxiliary_loss_clip": 0.01118368,
      "auxiliary_loss_mlp": 0.01035094,
      "balance_loss_clip": 1.0470264,
      "balance_loss_mlp": 1.01890504,
      "epoch": 0.3770930407335037,
      "flos": 21835304115840.0,
      "grad_norm": 1.6966604951606268,
      "language_loss": 0.79465914,
      "learning_rate": 2.8638303995638066e-06,
      "loss": 0.81619382,
      "num_input_tokens_seen": 134713115,
      "step": 6272,
      "time_per_iteration": 2.5183780193328857
    },
    {
      "auxiliary_loss_clip": 0.01114201,
      "auxiliary_loss_mlp": 0.01034385,
      "balance_loss_clip": 1.04363155,
      "balance_loss_mlp": 1.01999044,
      "epoch": 0.37715316398617166,
      "flos": 22747901984640.0,
      "grad_norm": 1.4949626471250725,
      "language_loss": 0.74101067,
      "learning_rate": 2.863479122159103e-06,
      "loss": 0.76249659,
      "num_input_tokens_seen": 134732635,
      "step": 6273,
      "time_per_iteration": 2.491856813430786
    },
    {
      "auxiliary_loss_clip": 0.01113017,
      "auxiliary_loss_mlp": 0.01043224,
      "balance_loss_clip": 1.04632235,
      "balance_loss_mlp": 1.0284183,
      "epoch": 0.3772132872388396,
      "flos": 18914581520640.0,
      "grad_norm": 1.4786248227707917,
      "language_loss": 0.71823126,
      "learning_rate": 2.8631278120111858e-06,
      "loss": 0.73979366,
      "num_input_tokens_seen": 134750695,
      "step": 6274,
      "time_per_iteration": 2.4822094440460205
    },
    {
      "auxiliary_loss_clip": 0.01104281,
      "auxiliary_loss_mlp": 0.01039058,
      "balance_loss_clip": 1.04889846,
      "balance_loss_mlp": 1.02458,
      "epoch": 0.3772734104915076,
      "flos": 17346219004800.0,
      "grad_norm": 1.6257731551170356,
      "language_loss": 0.84028113,
      "learning_rate": 2.8627764691333742e-06,
      "loss": 0.86171454,
      "num_input_tokens_seen": 134768935,
      "step": 6275,
      "time_per_iteration": 2.526548385620117
    },
    {
      "auxiliary_loss_clip": 0.01075593,
      "auxiliary_loss_mlp": 0.01036423,
      "balance_loss_clip": 1.04550493,
      "balance_loss_mlp": 1.02298224,
      "epoch": 0.37733353374417555,
      "flos": 32342370785280.0,
      "grad_norm": 1.5078952808177162,
      "language_loss": 0.75282598,
      "learning_rate": 2.8624250935389935e-06,
      "loss": 0.77394611,
      "num_input_tokens_seen": 134791260,
      "step": 6276,
      "time_per_iteration": 2.682257890701294
    },
    {
      "auxiliary_loss_clip": 0.01106965,
      "auxiliary_loss_mlp": 0.0103631,
      "balance_loss_clip": 1.04320621,
      "balance_loss_mlp": 1.02080655,
      "epoch": 0.3773936569968435,
      "flos": 23360681030400.0,
      "grad_norm": 1.8651679243634265,
      "language_loss": 0.85492194,
      "learning_rate": 2.862073685241366e-06,
      "loss": 0.87635469,
      "num_input_tokens_seen": 134808350,
      "step": 6277,
      "time_per_iteration": 2.5136756896972656
    },
    {
      "auxiliary_loss_clip": 0.01118851,
      "auxiliary_loss_mlp": 0.01036512,
      "balance_loss_clip": 1.04914355,
      "balance_loss_mlp": 1.02252293,
      "epoch": 0.3774537802495115,
      "flos": 21466788531840.0,
      "grad_norm": 1.8113932983803622,
      "language_loss": 0.77699912,
      "learning_rate": 2.861722244253818e-06,
      "loss": 0.79855275,
      "num_input_tokens_seen": 134826005,
      "step": 6278,
      "time_per_iteration": 2.496595859527588
    },
    {
      "auxiliary_loss_clip": 0.01103515,
      "auxiliary_loss_mlp": 0.01044338,
      "balance_loss_clip": 1.05167711,
      "balance_loss_mlp": 1.02837563,
      "epoch": 0.37751390350217945,
      "flos": 24973717086720.0,
      "grad_norm": 1.7695666953913747,
      "language_loss": 0.83306634,
      "learning_rate": 2.8613707705896767e-06,
      "loss": 0.85454488,
      "num_input_tokens_seen": 134844995,
      "step": 6279,
      "time_per_iteration": 2.5581116676330566
    },
    {
      "auxiliary_loss_clip": 0.01107659,
      "auxiliary_loss_mlp": 0.01034474,
      "balance_loss_clip": 1.04558599,
      "balance_loss_mlp": 1.02084792,
      "epoch": 0.3775740267548474,
      "flos": 27819098904960.0,
      "grad_norm": 1.837132267407551,
      "language_loss": 0.74380904,
      "learning_rate": 2.861019264262269e-06,
      "loss": 0.76523042,
      "num_input_tokens_seen": 134865285,
      "step": 6280,
      "time_per_iteration": 2.5631895065307617
    },
    {
      "auxiliary_loss_clip": 0.01128281,
      "auxiliary_loss_mlp": 0.0103994,
      "balance_loss_clip": 1.04837251,
      "balance_loss_mlp": 1.0258379,
      "epoch": 0.3776341500075154,
      "flos": 22565224391040.0,
      "grad_norm": 1.4665349653409734,
      "language_loss": 0.76097083,
      "learning_rate": 2.8606677252849242e-06,
      "loss": 0.78265303,
      "num_input_tokens_seen": 134886535,
      "step": 6281,
      "time_per_iteration": 2.47795033454895
    },
    {
      "auxiliary_loss_clip": 0.01099909,
      "auxiliary_loss_mlp": 0.01036022,
      "balance_loss_clip": 1.0458405,
      "balance_loss_mlp": 1.02135885,
      "epoch": 0.3776942732601834,
      "flos": 23077238808960.0,
      "grad_norm": 1.6489685497170306,
      "language_loss": 0.8425836,
      "learning_rate": 2.860316153670974e-06,
      "loss": 0.86394286,
      "num_input_tokens_seen": 134907435,
      "step": 6282,
      "time_per_iteration": 2.5502302646636963
    },
    {
      "auxiliary_loss_clip": 0.01116804,
      "auxiliary_loss_mlp": 0.01034062,
      "balance_loss_clip": 1.04526806,
      "balance_loss_mlp": 1.01939893,
      "epoch": 0.37775439651285136,
      "flos": 21724411852800.0,
      "grad_norm": 1.7004434356725147,
      "language_loss": 0.69931096,
      "learning_rate": 2.8599645494337484e-06,
      "loss": 0.72081959,
      "num_input_tokens_seen": 134925360,
      "step": 6283,
      "time_per_iteration": 2.492694616317749
    },
    {
      "auxiliary_loss_clip": 0.01073362,
      "auxiliary_loss_mlp": 0.01046734,
      "balance_loss_clip": 1.04827213,
      "balance_loss_mlp": 1.02990139,
      "epoch": 0.37781451976551933,
      "flos": 23987753688960.0,
      "grad_norm": 1.6976212999345577,
      "language_loss": 0.75870323,
      "learning_rate": 2.859612912586581e-06,
      "loss": 0.77990413,
      "num_input_tokens_seen": 134944205,
      "step": 6284,
      "time_per_iteration": 2.6398279666900635
    },
    {
      "auxiliary_loss_clip": 0.01136305,
      "auxiliary_loss_mlp": 0.01035205,
      "balance_loss_clip": 1.04945874,
      "balance_loss_mlp": 1.01935041,
      "epoch": 0.3778746430181873,
      "flos": 13727967223680.0,
      "grad_norm": 2.0634885085973393,
      "language_loss": 0.85134172,
      "learning_rate": 2.8592612431428055e-06,
      "loss": 0.87305689,
      "num_input_tokens_seen": 134960255,
      "step": 6285,
      "time_per_iteration": 2.4248037338256836
    },
    {
      "auxiliary_loss_clip": 0.01107277,
      "auxiliary_loss_mlp": 0.01036523,
      "balance_loss_clip": 1.04817677,
      "balance_loss_mlp": 1.02085948,
      "epoch": 0.37793476627085526,
      "flos": 19460495399040.0,
      "grad_norm": 1.7549712303472602,
      "language_loss": 0.84327191,
      "learning_rate": 2.858909541115758e-06,
      "loss": 0.86470997,
      "num_input_tokens_seen": 134978605,
      "step": 6286,
      "time_per_iteration": 2.529057025909424
    },
    {
      "auxiliary_loss_clip": 0.01111017,
      "auxiliary_loss_mlp": 0.0103964,
      "balance_loss_clip": 1.04612172,
      "balance_loss_mlp": 1.02522135,
      "epoch": 0.3779948895235232,
      "flos": 10707018704640.0,
      "grad_norm": 2.3949344392273963,
      "language_loss": 0.81384224,
      "learning_rate": 2.858557806518775e-06,
      "loss": 0.83534884,
      "num_input_tokens_seen": 134995020,
      "step": 6287,
      "time_per_iteration": 2.456221342086792
    },
    {
      "auxiliary_loss_clip": 0.01113242,
      "auxiliary_loss_mlp": 0.01042351,
      "balance_loss_clip": 1.04356039,
      "balance_loss_mlp": 1.02759862,
      "epoch": 0.3780550127761912,
      "flos": 22310007281280.0,
      "grad_norm": 2.234714058809263,
      "language_loss": 0.73391706,
      "learning_rate": 2.8582060393651927e-06,
      "loss": 0.75547296,
      "num_input_tokens_seen": 135012620,
      "step": 6288,
      "time_per_iteration": 2.494032859802246
    },
    {
      "auxiliary_loss_clip": 0.01120431,
      "auxiliary_loss_mlp": 0.0103907,
      "balance_loss_clip": 1.05029762,
      "balance_loss_mlp": 1.02367377,
      "epoch": 0.37811513602885916,
      "flos": 28950644125440.0,
      "grad_norm": 2.666621315894215,
      "language_loss": 0.75349474,
      "learning_rate": 2.857854239668352e-06,
      "loss": 0.77508974,
      "num_input_tokens_seen": 135033365,
      "step": 6289,
      "time_per_iteration": 2.549288034439087
    },
    {
      "auxiliary_loss_clip": 0.01119484,
      "auxiliary_loss_mlp": 0.01042506,
      "balance_loss_clip": 1.04818225,
      "balance_loss_mlp": 1.02797472,
      "epoch": 0.3781752592815271,
      "flos": 23112933949440.0,
      "grad_norm": 1.935743094527195,
      "language_loss": 0.73256671,
      "learning_rate": 2.857502407441593e-06,
      "loss": 0.75418663,
      "num_input_tokens_seen": 135052185,
      "step": 6290,
      "time_per_iteration": 2.5163841247558594
    },
    {
      "auxiliary_loss_clip": 0.01093363,
      "auxiliary_loss_mlp": 0.01042728,
      "balance_loss_clip": 1.04573703,
      "balance_loss_mlp": 1.02588391,
      "epoch": 0.3782353825341951,
      "flos": 19755932762880.0,
      "grad_norm": 2.1175716798700286,
      "language_loss": 0.80137122,
      "learning_rate": 2.8571505426982566e-06,
      "loss": 0.82273209,
      "num_input_tokens_seen": 135070425,
      "step": 6291,
      "time_per_iteration": 3.9359819889068604
    },
    {
      "auxiliary_loss_clip": 0.010959,
      "auxiliary_loss_mlp": 0.01033407,
      "balance_loss_clip": 1.04625881,
      "balance_loss_mlp": 1.01812398,
      "epoch": 0.37829550578686305,
      "flos": 22050839675520.0,
      "grad_norm": 1.9378129918333193,
      "language_loss": 0.76216114,
      "learning_rate": 2.8567986454516854e-06,
      "loss": 0.78345424,
      "num_input_tokens_seen": 135090525,
      "step": 6292,
      "time_per_iteration": 2.5499916076660156
    },
    {
      "auxiliary_loss_clip": 0.01111398,
      "auxiliary_loss_mlp": 0.01045723,
      "balance_loss_clip": 1.04550838,
      "balance_loss_mlp": 1.03002322,
      "epoch": 0.378355629039531,
      "flos": 16470357770880.0,
      "grad_norm": 1.6865831449560085,
      "language_loss": 0.69704771,
      "learning_rate": 2.856446715715224e-06,
      "loss": 0.71861887,
      "num_input_tokens_seen": 135109575,
      "step": 6293,
      "time_per_iteration": 2.4804694652557373
    },
    {
      "auxiliary_loss_clip": 0.01127877,
      "auxiliary_loss_mlp": 0.01039685,
      "balance_loss_clip": 1.04652095,
      "balance_loss_mlp": 1.02432442,
      "epoch": 0.378415752292199,
      "flos": 19974844200960.0,
      "grad_norm": 2.1953425929313033,
      "language_loss": 0.71639282,
      "learning_rate": 2.8560947535022173e-06,
      "loss": 0.73806846,
      "num_input_tokens_seen": 135127000,
      "step": 6294,
      "time_per_iteration": 5.316091299057007
    },
    {
      "auxiliary_loss_clip": 0.01108763,
      "auxiliary_loss_mlp": 0.0103687,
      "balance_loss_clip": 1.04504359,
      "balance_loss_mlp": 1.02096748,
      "epoch": 0.378475875544867,
      "flos": 14647388676480.0,
      "grad_norm": 2.0169620915287796,
      "language_loss": 0.82669044,
      "learning_rate": 2.855742758826011e-06,
      "loss": 0.8481468,
      "num_input_tokens_seen": 135145285,
      "step": 6295,
      "time_per_iteration": 2.4912915229797363
    },
    {
      "auxiliary_loss_clip": 0.01112008,
      "auxiliary_loss_mlp": 0.0103758,
      "balance_loss_clip": 1.04554105,
      "balance_loss_mlp": 1.0228579,
      "epoch": 0.37853599879753497,
      "flos": 26650996617600.0,
      "grad_norm": 1.7833155462626749,
      "language_loss": 0.71388501,
      "learning_rate": 2.8553907316999547e-06,
      "loss": 0.73538089,
      "num_input_tokens_seen": 135165240,
      "step": 6296,
      "time_per_iteration": 2.5307788848876953
    },
    {
      "auxiliary_loss_clip": 0.01129609,
      "auxiliary_loss_mlp": 0.01041777,
      "balance_loss_clip": 1.049582,
      "balance_loss_mlp": 1.02703691,
      "epoch": 0.37859612205020293,
      "flos": 17311960408320.0,
      "grad_norm": 1.7340887818817647,
      "language_loss": 0.76949227,
      "learning_rate": 2.855038672137396e-06,
      "loss": 0.79120612,
      "num_input_tokens_seen": 135184045,
      "step": 6297,
      "time_per_iteration": 3.816436767578125
    },
    {
      "auxiliary_loss_clip": 0.01105482,
      "auxiliary_loss_mlp": 0.01036799,
      "balance_loss_clip": 1.04782891,
      "balance_loss_mlp": 1.02238703,
      "epoch": 0.3786562453028709,
      "flos": 18220392299520.0,
      "grad_norm": 1.8915861456330683,
      "language_loss": 0.79194665,
      "learning_rate": 2.854686580151684e-06,
      "loss": 0.81336945,
      "num_input_tokens_seen": 135202365,
      "step": 6298,
      "time_per_iteration": 2.49326753616333
    },
    {
      "auxiliary_loss_clip": 0.01070566,
      "auxiliary_loss_mlp": 0.0104793,
      "balance_loss_clip": 1.04150486,
      "balance_loss_mlp": 1.0310024,
      "epoch": 0.37871636855553886,
      "flos": 21214875473280.0,
      "grad_norm": 1.6405593902856228,
      "language_loss": 0.84336615,
      "learning_rate": 2.8543344557561722e-06,
      "loss": 0.86455119,
      "num_input_tokens_seen": 135220955,
      "step": 6299,
      "time_per_iteration": 2.569901943206787
    },
    {
      "auxiliary_loss_clip": 0.01093415,
      "auxiliary_loss_mlp": 0.01037315,
      "balance_loss_clip": 1.0419873,
      "balance_loss_mlp": 1.02206767,
      "epoch": 0.3787764918082068,
      "flos": 20952727038720.0,
      "grad_norm": 1.9724564310029635,
      "language_loss": 0.76330405,
      "learning_rate": 2.8539822989642116e-06,
      "loss": 0.78461134,
      "num_input_tokens_seen": 135239715,
      "step": 6300,
      "time_per_iteration": 2.531797170639038
    },
    {
      "auxiliary_loss_clip": 0.01111543,
      "auxiliary_loss_mlp": 0.0103827,
      "balance_loss_clip": 1.04639316,
      "balance_loss_mlp": 1.02133024,
      "epoch": 0.3788366150608748,
      "flos": 17308009912320.0,
      "grad_norm": 1.9532683358153005,
      "language_loss": 0.82976717,
      "learning_rate": 2.8536301097891577e-06,
      "loss": 0.85126531,
      "num_input_tokens_seen": 135257035,
      "step": 6301,
      "time_per_iteration": 2.4886343479156494
    },
    {
      "auxiliary_loss_clip": 0.01112114,
      "auxiliary_loss_mlp": 0.01036237,
      "balance_loss_clip": 1.04647493,
      "balance_loss_mlp": 1.02222395,
      "epoch": 0.37889673831354276,
      "flos": 24311092942080.0,
      "grad_norm": 1.9949576918825387,
      "language_loss": 0.68026334,
      "learning_rate": 2.8532778882443636e-06,
      "loss": 0.70174682,
      "num_input_tokens_seen": 135275720,
      "step": 6302,
      "time_per_iteration": 2.501490831375122
    },
    {
      "auxiliary_loss_clip": 0.01082235,
      "auxiliary_loss_mlp": 0.0104158,
      "balance_loss_clip": 1.04198229,
      "balance_loss_mlp": 1.02677417,
      "epoch": 0.3789568615662107,
      "flos": 26683603188480.0,
      "grad_norm": 1.8379743572737421,
      "language_loss": 0.68425649,
      "learning_rate": 2.8529256343431867e-06,
      "loss": 0.70549464,
      "num_input_tokens_seen": 135294140,
      "step": 6303,
      "time_per_iteration": 2.6166861057281494
    },
    {
      "auxiliary_loss_clip": 0.01126037,
      "auxiliary_loss_mlp": 0.01033583,
      "balance_loss_clip": 1.04558969,
      "balance_loss_mlp": 1.01961756,
      "epoch": 0.3790169848188787,
      "flos": 23585194990080.0,
      "grad_norm": 1.5525068516171425,
      "language_loss": 0.77438104,
      "learning_rate": 2.8525733480989846e-06,
      "loss": 0.79597723,
      "num_input_tokens_seen": 135314845,
      "step": 6304,
      "time_per_iteration": 2.4690423011779785
    },
    {
      "auxiliary_loss_clip": 0.0113629,
      "auxiliary_loss_mlp": 0.01036061,
      "balance_loss_clip": 1.05091834,
      "balance_loss_mlp": 1.02065885,
      "epoch": 0.37907710807154665,
      "flos": 18437436230400.0,
      "grad_norm": 1.840956388400523,
      "language_loss": 0.80538815,
      "learning_rate": 2.8522210295251146e-06,
      "loss": 0.82711166,
      "num_input_tokens_seen": 135333055,
      "step": 6305,
      "time_per_iteration": 2.4449024200439453
    },
    {
      "auxiliary_loss_clip": 0.01042169,
      "auxiliary_loss_mlp": 0.01012063,
      "balance_loss_clip": 1.02367759,
      "balance_loss_mlp": 1.01041794,
      "epoch": 0.3791372313242146,
      "flos": 50107165954560.0,
      "grad_norm": 0.9786051169970005,
      "language_loss": 0.64501238,
      "learning_rate": 2.8518686786349387e-06,
      "loss": 0.66555476,
      "num_input_tokens_seen": 135387865,
      "step": 6306,
      "time_per_iteration": 3.001331329345703
    },
    {
      "auxiliary_loss_clip": 0.01104632,
      "auxiliary_loss_mlp": 0.01053294,
      "balance_loss_clip": 1.0480094,
      "balance_loss_mlp": 1.03654528,
      "epoch": 0.3791973545768826,
      "flos": 24316551809280.0,
      "grad_norm": 1.657933892739309,
      "language_loss": 0.73343682,
      "learning_rate": 2.851516295441817e-06,
      "loss": 0.75501609,
      "num_input_tokens_seen": 135409095,
      "step": 6307,
      "time_per_iteration": 2.5741920471191406
    },
    {
      "auxiliary_loss_clip": 0.01106413,
      "auxiliary_loss_mlp": 0.01038025,
      "balance_loss_clip": 1.04904246,
      "balance_loss_mlp": 1.0226233,
      "epoch": 0.3792574778295506,
      "flos": 21579907438080.0,
      "grad_norm": 1.3406060624191884,
      "language_loss": 0.78220493,
      "learning_rate": 2.851163879959112e-06,
      "loss": 0.80364931,
      "num_input_tokens_seen": 135429585,
      "step": 6308,
      "time_per_iteration": 2.5402448177337646
    },
    {
      "auxiliary_loss_clip": 0.0109138,
      "auxiliary_loss_mlp": 0.01039633,
      "balance_loss_clip": 1.04669261,
      "balance_loss_mlp": 1.02464843,
      "epoch": 0.37931760108221857,
      "flos": 22272731942400.0,
      "grad_norm": 1.9453512664875034,
      "language_loss": 0.72968423,
      "learning_rate": 2.8508114322001876e-06,
      "loss": 0.75099444,
      "num_input_tokens_seen": 135446320,
      "step": 6309,
      "time_per_iteration": 2.5721166133880615
    },
    {
      "auxiliary_loss_clip": 0.01073639,
      "auxiliary_loss_mlp": 0.01043823,
      "balance_loss_clip": 1.04769468,
      "balance_loss_mlp": 1.02824211,
      "epoch": 0.37937772433488653,
      "flos": 19682998197120.0,
      "grad_norm": 1.3618338533149712,
      "language_loss": 0.78296018,
      "learning_rate": 2.8504589521784083e-06,
      "loss": 0.80413479,
      "num_input_tokens_seen": 135465720,
      "step": 6310,
      "time_per_iteration": 2.585340976715088
    },
    {
      "auxiliary_loss_clip": 0.01118459,
      "auxiliary_loss_mlp": 0.00794277,
      "balance_loss_clip": 1.04610276,
      "balance_loss_mlp": 1.01990342,
      "epoch": 0.3794378475875545,
      "flos": 19099378016640.0,
      "grad_norm": 1.7035103523654243,
      "language_loss": 0.76360106,
      "learning_rate": 2.8501064399071403e-06,
      "loss": 0.78272837,
      "num_input_tokens_seen": 135485155,
      "step": 6311,
      "time_per_iteration": 2.5155489444732666
    },
    {
      "auxiliary_loss_clip": 0.01112682,
      "auxiliary_loss_mlp": 0.0103133,
      "balance_loss_clip": 1.05058599,
      "balance_loss_mlp": 1.0173285,
      "epoch": 0.37949797084022246,
      "flos": 20339660684160.0,
      "grad_norm": 1.4954457078019694,
      "language_loss": 0.7033689,
      "learning_rate": 2.8497538953997504e-06,
      "loss": 0.72480905,
      "num_input_tokens_seen": 135502675,
      "step": 6312,
      "time_per_iteration": 2.5423941612243652
    },
    {
      "auxiliary_loss_clip": 0.01025078,
      "auxiliary_loss_mlp": 0.01003501,
      "balance_loss_clip": 1.02749729,
      "balance_loss_mlp": 1.00192738,
      "epoch": 0.37955809409289043,
      "flos": 63972203477760.0,
      "grad_norm": 0.8128010179612137,
      "language_loss": 0.56089389,
      "learning_rate": 2.849401318669608e-06,
      "loss": 0.58117974,
      "num_input_tokens_seen": 135562005,
      "step": 6313,
      "time_per_iteration": 3.1380844116210938
    },
    {
      "auxiliary_loss_clip": 0.01094232,
      "auxiliary_loss_mlp": 0.0104081,
      "balance_loss_clip": 1.04618001,
      "balance_loss_mlp": 1.02609384,
      "epoch": 0.3796182173455584,
      "flos": 31540665179520.0,
      "grad_norm": 1.692945137318909,
      "language_loss": 0.71489549,
      "learning_rate": 2.849048709730083e-06,
      "loss": 0.73624587,
      "num_input_tokens_seen": 135582600,
      "step": 6314,
      "time_per_iteration": 2.6523373126983643
    },
    {
      "auxiliary_loss_clip": 0.01124739,
      "auxiliary_loss_mlp": 0.01039533,
      "balance_loss_clip": 1.05018091,
      "balance_loss_mlp": 1.02407193,
      "epoch": 0.37967834059822636,
      "flos": 12130804978560.0,
      "grad_norm": 2.013223213361535,
      "language_loss": 0.7350834,
      "learning_rate": 2.848696068594545e-06,
      "loss": 0.75672615,
      "num_input_tokens_seen": 135600280,
      "step": 6315,
      "time_per_iteration": 2.4840304851531982
    },
    {
      "auxiliary_loss_clip": 0.01122553,
      "auxiliary_loss_mlp": 0.01042107,
      "balance_loss_clip": 1.04894054,
      "balance_loss_mlp": 1.02751589,
      "epoch": 0.3797384638508943,
      "flos": 39348578298240.0,
      "grad_norm": 2.070269831835568,
      "language_loss": 0.70815194,
      "learning_rate": 2.8483433952763677e-06,
      "loss": 0.72979856,
      "num_input_tokens_seen": 135621560,
      "step": 6316,
      "time_per_iteration": 2.631497621536255
    },
    {
      "auxiliary_loss_clip": 0.01097326,
      "auxiliary_loss_mlp": 0.01032188,
      "balance_loss_clip": 1.04959404,
      "balance_loss_mlp": 1.0186224,
      "epoch": 0.3797985871035623,
      "flos": 34054016653440.0,
      "grad_norm": 1.7368235689825242,
      "language_loss": 0.65033746,
      "learning_rate": 2.847990689788923e-06,
      "loss": 0.67163259,
      "num_input_tokens_seen": 135641745,
      "step": 6317,
      "time_per_iteration": 2.666734218597412
    },
    {
      "auxiliary_loss_clip": 0.01115656,
      "auxiliary_loss_mlp": 0.01034935,
      "balance_loss_clip": 1.04674602,
      "balance_loss_mlp": 1.02156544,
      "epoch": 0.37985871035623026,
      "flos": 23222174186880.0,
      "grad_norm": 2.116810400178657,
      "language_loss": 0.85845613,
      "learning_rate": 2.8476379521455877e-06,
      "loss": 0.87996209,
      "num_input_tokens_seen": 135660650,
      "step": 6318,
      "time_per_iteration": 2.5185465812683105
    },
    {
      "auxiliary_loss_clip": 0.01107025,
      "auxiliary_loss_mlp": 0.01037519,
      "balance_loss_clip": 1.04762673,
      "balance_loss_mlp": 1.02197981,
      "epoch": 0.3799188336088982,
      "flos": 18114958903680.0,
      "grad_norm": 1.8249640820494513,
      "language_loss": 0.75830662,
      "learning_rate": 2.8472851823597354e-06,
      "loss": 0.77975202,
      "num_input_tokens_seen": 135679980,
      "step": 6319,
      "time_per_iteration": 2.5096659660339355
    },
    {
      "auxiliary_loss_clip": 0.01130299,
      "auxiliary_loss_mlp": 0.01036144,
      "balance_loss_clip": 1.04867947,
      "balance_loss_mlp": 1.02209544,
      "epoch": 0.3799789568615662,
      "flos": 21871897096320.0,
      "grad_norm": 1.9231228556726114,
      "language_loss": 0.63970149,
      "learning_rate": 2.846932380444744e-06,
      "loss": 0.66136593,
      "num_input_tokens_seen": 135699400,
      "step": 6320,
      "time_per_iteration": 2.4641120433807373
    },
    {
      "auxiliary_loss_clip": 0.01088843,
      "auxiliary_loss_mlp": 0.01039209,
      "balance_loss_clip": 1.04508543,
      "balance_loss_mlp": 1.02470112,
      "epoch": 0.3800390801142342,
      "flos": 32962943082240.0,
      "grad_norm": 2.0538331486748236,
      "language_loss": 0.71342206,
      "learning_rate": 2.846579546413992e-06,
      "loss": 0.73470259,
      "num_input_tokens_seen": 135723455,
      "step": 6321,
      "time_per_iteration": 2.6682469844818115
    },
    {
      "auxiliary_loss_clip": 0.01093299,
      "auxiliary_loss_mlp": 0.01036557,
      "balance_loss_clip": 1.04529858,
      "balance_loss_mlp": 1.02173984,
      "epoch": 0.38009920336690217,
      "flos": 26907075653760.0,
      "grad_norm": 1.7042915169570947,
      "language_loss": 0.74755621,
      "learning_rate": 2.846226680280859e-06,
      "loss": 0.7688548,
      "num_input_tokens_seen": 135744335,
      "step": 6322,
      "time_per_iteration": 2.606158494949341
    },
    {
      "auxiliary_loss_clip": 0.01117782,
      "auxiliary_loss_mlp": 0.01036415,
      "balance_loss_clip": 1.04635441,
      "balance_loss_mlp": 1.0213114,
      "epoch": 0.38015932661957014,
      "flos": 22488913946880.0,
      "grad_norm": 1.9292761620526755,
      "language_loss": 0.85027391,
      "learning_rate": 2.845873782058725e-06,
      "loss": 0.87181592,
      "num_input_tokens_seen": 135761440,
      "step": 6323,
      "time_per_iteration": 2.483649969100952
    },
    {
      "auxiliary_loss_clip": 0.0110348,
      "auxiliary_loss_mlp": 0.01034562,
      "balance_loss_clip": 1.04330277,
      "balance_loss_mlp": 1.01912403,
      "epoch": 0.3802194498722381,
      "flos": 21980993679360.0,
      "grad_norm": 2.6638962260827825,
      "language_loss": 0.73629892,
      "learning_rate": 2.845520851760973e-06,
      "loss": 0.75767934,
      "num_input_tokens_seen": 135779955,
      "step": 6324,
      "time_per_iteration": 2.522289514541626
    },
    {
      "auxiliary_loss_clip": 0.01100598,
      "auxiliary_loss_mlp": 0.01038874,
      "balance_loss_clip": 1.04796386,
      "balance_loss_mlp": 1.02368081,
      "epoch": 0.38027957312490607,
      "flos": 21324869896320.0,
      "grad_norm": 1.6666539298181506,
      "language_loss": 0.84270024,
      "learning_rate": 2.8451678894009847e-06,
      "loss": 0.86409491,
      "num_input_tokens_seen": 135799840,
      "step": 6325,
      "time_per_iteration": 2.5397698879241943
    },
    {
      "auxiliary_loss_clip": 0.01103476,
      "auxiliary_loss_mlp": 0.01031872,
      "balance_loss_clip": 1.04868793,
      "balance_loss_mlp": 1.01776385,
      "epoch": 0.38033969637757403,
      "flos": 16691244456960.0,
      "grad_norm": 1.6222562906545879,
      "language_loss": 0.79412544,
      "learning_rate": 2.8448148949921465e-06,
      "loss": 0.81547892,
      "num_input_tokens_seen": 135817880,
      "step": 6326,
      "time_per_iteration": 2.501400947570801
    },
    {
      "auxiliary_loss_clip": 0.01117475,
      "auxiliary_loss_mlp": 0.01042191,
      "balance_loss_clip": 1.04808331,
      "balance_loss_mlp": 1.02811193,
      "epoch": 0.380399819630242,
      "flos": 36210847685760.0,
      "grad_norm": 1.7810879360542151,
      "language_loss": 0.72798949,
      "learning_rate": 2.844461868547842e-06,
      "loss": 0.74958616,
      "num_input_tokens_seen": 135838940,
      "step": 6327,
      "time_per_iteration": 2.6304078102111816
    },
    {
      "auxiliary_loss_clip": 0.01127545,
      "auxiliary_loss_mlp": 0.00796115,
      "balance_loss_clip": 1.04769301,
      "balance_loss_mlp": 1.02479756,
      "epoch": 0.38045994288290996,
      "flos": 21288851533440.0,
      "grad_norm": 1.9383777495692633,
      "language_loss": 0.82736421,
      "learning_rate": 2.844108810081459e-06,
      "loss": 0.84660083,
      "num_input_tokens_seen": 135858325,
      "step": 6328,
      "time_per_iteration": 2.491238832473755
    },
    {
      "auxiliary_loss_clip": 0.01116363,
      "auxiliary_loss_mlp": 0.01029724,
      "balance_loss_clip": 1.04594612,
      "balance_loss_mlp": 1.01584792,
      "epoch": 0.38052006613557793,
      "flos": 20922885815040.0,
      "grad_norm": 1.2947157391298052,
      "language_loss": 0.61617184,
      "learning_rate": 2.843755719606385e-06,
      "loss": 0.63763267,
      "num_input_tokens_seen": 135878430,
      "step": 6329,
      "time_per_iteration": 2.493976354598999
    },
    {
      "auxiliary_loss_clip": 0.01110691,
      "auxiliary_loss_mlp": 0.01038331,
      "balance_loss_clip": 1.05132914,
      "balance_loss_mlp": 1.0240314,
      "epoch": 0.3805801893882459,
      "flos": 20990720649600.0,
      "grad_norm": 2.0073395572804493,
      "language_loss": 0.56228602,
      "learning_rate": 2.8434025971360104e-06,
      "loss": 0.58377624,
      "num_input_tokens_seen": 135894755,
      "step": 6330,
      "time_per_iteration": 3.903026819229126
    },
    {
      "auxiliary_loss_clip": 0.01083724,
      "auxiliary_loss_mlp": 0.01034501,
      "balance_loss_clip": 1.05031407,
      "balance_loss_mlp": 1.02129245,
      "epoch": 0.38064031264091386,
      "flos": 25558594243200.0,
      "grad_norm": 1.3720746176492677,
      "language_loss": 0.65748399,
      "learning_rate": 2.8430494426837243e-06,
      "loss": 0.67866623,
      "num_input_tokens_seen": 135918275,
      "step": 6331,
      "time_per_iteration": 2.621837615966797
    },
    {
      "auxiliary_loss_clip": 0.01117081,
      "auxiliary_loss_mlp": 0.01042946,
      "balance_loss_clip": 1.05162823,
      "balance_loss_mlp": 1.02787232,
      "epoch": 0.3807004358935818,
      "flos": 15085857997440.0,
      "grad_norm": 1.5012476474934946,
      "language_loss": 0.76031363,
      "learning_rate": 2.842696256262919e-06,
      "loss": 0.78191388,
      "num_input_tokens_seen": 135937430,
      "step": 6332,
      "time_per_iteration": 3.99216890335083
    },
    {
      "auxiliary_loss_clip": 0.01073796,
      "auxiliary_loss_mlp": 0.00795066,
      "balance_loss_clip": 1.0484314,
      "balance_loss_mlp": 1.01916027,
      "epoch": 0.3807605591462498,
      "flos": 16399398453120.0,
      "grad_norm": 1.703158449001075,
      "language_loss": 0.8230477,
      "learning_rate": 2.842343037886987e-06,
      "loss": 0.84173632,
      "num_input_tokens_seen": 135954210,
      "step": 6333,
      "time_per_iteration": 4.018657684326172
    },
    {
      "auxiliary_loss_clip": 0.0111568,
      "auxiliary_loss_mlp": 0.01030624,
      "balance_loss_clip": 1.04571736,
      "balance_loss_mlp": 1.01695096,
      "epoch": 0.3808206823989178,
      "flos": 29057083102080.0,
      "grad_norm": 1.4346241553768697,
      "language_loss": 0.86159593,
      "learning_rate": 2.8419897875693226e-06,
      "loss": 0.88305897,
      "num_input_tokens_seen": 135974425,
      "step": 6334,
      "time_per_iteration": 2.5841455459594727
    },
    {
      "auxiliary_loss_clip": 0.01119117,
      "auxiliary_loss_mlp": 0.01037235,
      "balance_loss_clip": 1.04527211,
      "balance_loss_mlp": 1.0225842,
      "epoch": 0.3808808056515858,
      "flos": 15705855676800.0,
      "grad_norm": 1.728989168179784,
      "language_loss": 0.79461873,
      "learning_rate": 2.841636505323321e-06,
      "loss": 0.81618226,
      "num_input_tokens_seen": 135991985,
      "step": 6335,
      "time_per_iteration": 2.4574644565582275
    },
    {
      "auxiliary_loss_clip": 0.01117906,
      "auxiliary_loss_mlp": 0.01034915,
      "balance_loss_clip": 1.0464102,
      "balance_loss_mlp": 1.02032995,
      "epoch": 0.38094092890425374,
      "flos": 20704584908160.0,
      "grad_norm": 1.810350578572506,
      "language_loss": 0.73560929,
      "learning_rate": 2.8412831911623795e-06,
      "loss": 0.75713754,
      "num_input_tokens_seen": 136010015,
      "step": 6336,
      "time_per_iteration": 3.863769769668579
    },
    {
      "auxiliary_loss_clip": 0.01111745,
      "auxiliary_loss_mlp": 0.01032754,
      "balance_loss_clip": 1.04689753,
      "balance_loss_mlp": 1.0190866,
      "epoch": 0.3810010521569217,
      "flos": 20667956014080.0,
      "grad_norm": 2.2671224185047714,
      "language_loss": 0.68611503,
      "learning_rate": 2.840929845099894e-06,
      "loss": 0.70756,
      "num_input_tokens_seen": 136028440,
      "step": 6337,
      "time_per_iteration": 2.4967098236083984
    },
    {
      "auxiliary_loss_clip": 0.01105162,
      "auxiliary_loss_mlp": 0.01032759,
      "balance_loss_clip": 1.04364085,
      "balance_loss_mlp": 1.01772642,
      "epoch": 0.38106117540958967,
      "flos": 31827626933760.0,
      "grad_norm": 1.7471261003992924,
      "language_loss": 0.63274467,
      "learning_rate": 2.8405764671492652e-06,
      "loss": 0.6541239,
      "num_input_tokens_seen": 136048360,
      "step": 6338,
      "time_per_iteration": 2.629152297973633
    },
    {
      "auxiliary_loss_clip": 0.01107802,
      "auxiliary_loss_mlp": 0.01041032,
      "balance_loss_clip": 1.04558277,
      "balance_loss_mlp": 1.0254519,
      "epoch": 0.38112129866225763,
      "flos": 16902757693440.0,
      "grad_norm": 1.6669164244577088,
      "language_loss": 0.68992519,
      "learning_rate": 2.8402230573238923e-06,
      "loss": 0.7114135,
      "num_input_tokens_seen": 136065500,
      "step": 6339,
      "time_per_iteration": 2.4987664222717285
    },
    {
      "auxiliary_loss_clip": 0.01104101,
      "auxiliary_loss_mlp": 0.01045232,
      "balance_loss_clip": 1.04830754,
      "balance_loss_mlp": 1.03062332,
      "epoch": 0.3811814219149256,
      "flos": 20887226588160.0,
      "grad_norm": 2.1906189330174612,
      "language_loss": 0.67866683,
      "learning_rate": 2.839869615637177e-06,
      "loss": 0.70016015,
      "num_input_tokens_seen": 136084060,
      "step": 6340,
      "time_per_iteration": 2.5332086086273193
    },
    {
      "auxiliary_loss_clip": 0.0109456,
      "auxiliary_loss_mlp": 0.01037279,
      "balance_loss_clip": 1.04747534,
      "balance_loss_mlp": 1.02233016,
      "epoch": 0.38124154516759357,
      "flos": 16690813493760.0,
      "grad_norm": 2.4035951831324156,
      "language_loss": 0.89299297,
      "learning_rate": 2.839516142102522e-06,
      "loss": 0.91431129,
      "num_input_tokens_seen": 136102310,
      "step": 6341,
      "time_per_iteration": 2.5374531745910645
    },
    {
      "auxiliary_loss_clip": 0.01120275,
      "auxiliary_loss_mlp": 0.01043923,
      "balance_loss_clip": 1.04637682,
      "balance_loss_mlp": 1.02852666,
      "epoch": 0.38130166842026153,
      "flos": 19681956702720.0,
      "grad_norm": 1.5714588810780212,
      "language_loss": 0.75025702,
      "learning_rate": 2.83916263673333e-06,
      "loss": 0.77189898,
      "num_input_tokens_seen": 136120725,
      "step": 6342,
      "time_per_iteration": 2.4857327938079834
    },
    {
      "auxiliary_loss_clip": 0.01105481,
      "auxiliary_loss_mlp": 0.01037153,
      "balance_loss_clip": 1.044891,
      "balance_loss_mlp": 1.02300262,
      "epoch": 0.3813617916729295,
      "flos": 22198432659840.0,
      "grad_norm": 1.7305350387479543,
      "language_loss": 0.8360371,
      "learning_rate": 2.838809099543007e-06,
      "loss": 0.85746348,
      "num_input_tokens_seen": 136139105,
      "step": 6343,
      "time_per_iteration": 2.5172109603881836
    },
    {
      "auxiliary_loss_clip": 0.01072515,
      "auxiliary_loss_mlp": 0.01045337,
      "balance_loss_clip": 1.04667091,
      "balance_loss_mlp": 1.02997136,
      "epoch": 0.38142191492559746,
      "flos": 19096899978240.0,
      "grad_norm": 1.7144557478760796,
      "language_loss": 0.77029335,
      "learning_rate": 2.838455530544959e-06,
      "loss": 0.79147184,
      "num_input_tokens_seen": 136158265,
      "step": 6344,
      "time_per_iteration": 2.604015350341797
    },
    {
      "auxiliary_loss_clip": 0.01093917,
      "auxiliary_loss_mlp": 0.01045229,
      "balance_loss_clip": 1.04493988,
      "balance_loss_mlp": 1.02862298,
      "epoch": 0.3814820381782654,
      "flos": 24097748112000.0,
      "grad_norm": 3.0431561376105067,
      "language_loss": 0.7361986,
      "learning_rate": 2.838101929752593e-06,
      "loss": 0.75759006,
      "num_input_tokens_seen": 136176100,
      "step": 6345,
      "time_per_iteration": 2.5651538372039795
    },
    {
      "auxiliary_loss_clip": 0.01088084,
      "auxiliary_loss_mlp": 0.00795619,
      "balance_loss_clip": 1.04413474,
      "balance_loss_mlp": 1.02345788,
      "epoch": 0.3815421614309334,
      "flos": 15778502933760.0,
      "grad_norm": 1.8084450588779348,
      "language_loss": 0.69712949,
      "learning_rate": 2.8377482971793187e-06,
      "loss": 0.71596646,
      "num_input_tokens_seen": 136195125,
      "step": 6346,
      "time_per_iteration": 2.5558722019195557
    },
    {
      "auxiliary_loss_clip": 0.01118019,
      "auxiliary_loss_mlp": 0.01033786,
      "balance_loss_clip": 1.04623437,
      "balance_loss_mlp": 1.01986814,
      "epoch": 0.38160228468360136,
      "flos": 19899754819200.0,
      "grad_norm": 1.8350031715206616,
      "language_loss": 0.75714093,
      "learning_rate": 2.8373946328385437e-06,
      "loss": 0.77865899,
      "num_input_tokens_seen": 136213885,
      "step": 6347,
      "time_per_iteration": 2.4897735118865967
    },
    {
      "auxiliary_loss_clip": 0.01115721,
      "auxiliary_loss_mlp": 0.01041383,
      "balance_loss_clip": 1.04502559,
      "balance_loss_mlp": 1.02800131,
      "epoch": 0.3816624079362694,
      "flos": 19281050029440.0,
      "grad_norm": 1.5420293332210755,
      "language_loss": 0.74667543,
      "learning_rate": 2.8370409367436813e-06,
      "loss": 0.76824641,
      "num_input_tokens_seen": 136232700,
      "step": 6348,
      "time_per_iteration": 2.489745616912842
    },
    {
      "auxiliary_loss_clip": 0.01100603,
      "auxiliary_loss_mlp": 0.01031436,
      "balance_loss_clip": 1.04628968,
      "balance_loss_mlp": 1.01766753,
      "epoch": 0.38172253118893734,
      "flos": 21177564220800.0,
      "grad_norm": 2.118601453882471,
      "language_loss": 0.87073922,
      "learning_rate": 2.836687208908142e-06,
      "loss": 0.89205962,
      "num_input_tokens_seen": 136248975,
      "step": 6349,
      "time_per_iteration": 2.5078890323638916
    },
    {
      "auxiliary_loss_clip": 0.01113396,
      "auxiliary_loss_mlp": 0.01039322,
      "balance_loss_clip": 1.04738009,
      "balance_loss_mlp": 1.02531528,
      "epoch": 0.3817826544416053,
      "flos": 17529219820800.0,
      "grad_norm": 1.7373272629388106,
      "language_loss": 0.76868415,
      "learning_rate": 2.836333449345341e-06,
      "loss": 0.79021132,
      "num_input_tokens_seen": 136266710,
      "step": 6350,
      "time_per_iteration": 2.4785208702087402
    },
    {
      "auxiliary_loss_clip": 0.01092793,
      "auxiliary_loss_mlp": 0.0103011,
      "balance_loss_clip": 1.04526901,
      "balance_loss_mlp": 1.01451707,
      "epoch": 0.38184277769427327,
      "flos": 16326535714560.0,
      "grad_norm": 2.006814262499669,
      "language_loss": 0.75801581,
      "learning_rate": 2.8359796580686907e-06,
      "loss": 0.77924484,
      "num_input_tokens_seen": 136284445,
      "step": 6351,
      "time_per_iteration": 2.5159828662872314
    },
    {
      "auxiliary_loss_clip": 0.01114761,
      "auxiliary_loss_mlp": 0.01040675,
      "balance_loss_clip": 1.04397559,
      "balance_loss_mlp": 1.02534437,
      "epoch": 0.38190290094694124,
      "flos": 30443450382720.0,
      "grad_norm": 1.6594008193698917,
      "language_loss": 0.74168122,
      "learning_rate": 2.8356258350916085e-06,
      "loss": 0.76323557,
      "num_input_tokens_seen": 136305730,
      "step": 6352,
      "time_per_iteration": 2.568828582763672
    },
    {
      "auxiliary_loss_clip": 0.0109161,
      "auxiliary_loss_mlp": 0.0103401,
      "balance_loss_clip": 1.04612172,
      "balance_loss_mlp": 1.02064633,
      "epoch": 0.3819630241996092,
      "flos": 14209924936320.0,
      "grad_norm": 1.8408255488969296,
      "language_loss": 0.64298224,
      "learning_rate": 2.8352719804275104e-06,
      "loss": 0.66423845,
      "num_input_tokens_seen": 136323850,
      "step": 6353,
      "time_per_iteration": 2.56478214263916
    },
    {
      "auxiliary_loss_clip": 0.01125881,
      "auxiliary_loss_mlp": 0.01034557,
      "balance_loss_clip": 1.0450635,
      "balance_loss_mlp": 1.02084827,
      "epoch": 0.38202314745227717,
      "flos": 25009699536000.0,
      "grad_norm": 1.5482921108665268,
      "language_loss": 0.82974893,
      "learning_rate": 2.834918094089816e-06,
      "loss": 0.85135329,
      "num_input_tokens_seen": 136344880,
      "step": 6354,
      "time_per_iteration": 2.4897360801696777
    },
    {
      "auxiliary_loss_clip": 0.01126808,
      "auxiliary_loss_mlp": 0.01032726,
      "balance_loss_clip": 1.04720318,
      "balance_loss_mlp": 1.01911259,
      "epoch": 0.38208327070494513,
      "flos": 20814507504000.0,
      "grad_norm": 1.620803795496377,
      "language_loss": 0.80179405,
      "learning_rate": 2.834564176091943e-06,
      "loss": 0.82338935,
      "num_input_tokens_seen": 136366060,
      "step": 6355,
      "time_per_iteration": 2.47265362739563
    },
    {
      "auxiliary_loss_clip": 0.01085382,
      "auxiliary_loss_mlp": 0.01042097,
      "balance_loss_clip": 1.04699123,
      "balance_loss_mlp": 1.02747035,
      "epoch": 0.3821433939576131,
      "flos": 22637727993600.0,
      "grad_norm": 1.7256294128417997,
      "language_loss": 0.75121212,
      "learning_rate": 2.8342102264473125e-06,
      "loss": 0.77248693,
      "num_input_tokens_seen": 136385625,
      "step": 6356,
      "time_per_iteration": 2.579561471939087
    },
    {
      "auxiliary_loss_clip": 0.01117539,
      "auxiliary_loss_mlp": 0.00792094,
      "balance_loss_clip": 1.0462625,
      "balance_loss_mlp": 1.01497245,
      "epoch": 0.38220351721028106,
      "flos": 26869872142080.0,
      "grad_norm": 1.8699244707279434,
      "language_loss": 0.81232989,
      "learning_rate": 2.833856245169348e-06,
      "loss": 0.83142626,
      "num_input_tokens_seen": 136405750,
      "step": 6357,
      "time_per_iteration": 2.527331829071045
    },
    {
      "auxiliary_loss_clip": 0.01110157,
      "auxiliary_loss_mlp": 0.01041609,
      "balance_loss_clip": 1.0498569,
      "balance_loss_mlp": 1.02583718,
      "epoch": 0.38226364046294903,
      "flos": 23367468700800.0,
      "grad_norm": 1.7561602764412787,
      "language_loss": 0.77302456,
      "learning_rate": 2.8335022322714695e-06,
      "loss": 0.79454225,
      "num_input_tokens_seen": 136426085,
      "step": 6358,
      "time_per_iteration": 2.530564546585083
    },
    {
      "auxiliary_loss_clip": 0.01107864,
      "auxiliary_loss_mlp": 0.01039992,
      "balance_loss_clip": 1.04442024,
      "balance_loss_mlp": 1.02488852,
      "epoch": 0.382323763715617,
      "flos": 19646225648640.0,
      "grad_norm": 2.182025269785297,
      "language_loss": 0.78510314,
      "learning_rate": 2.8331481877671036e-06,
      "loss": 0.80658174,
      "num_input_tokens_seen": 136442670,
      "step": 6359,
      "time_per_iteration": 2.4887638092041016
    },
    {
      "auxiliary_loss_clip": 0.01063318,
      "auxiliary_loss_mlp": 0.01046756,
      "balance_loss_clip": 1.0430212,
      "balance_loss_mlp": 1.02968562,
      "epoch": 0.38238388696828496,
      "flos": 54124741232640.0,
      "grad_norm": 1.7844643331183885,
      "language_loss": 0.69828594,
      "learning_rate": 2.8327941116696754e-06,
      "loss": 0.71938664,
      "num_input_tokens_seen": 136465730,
      "step": 6360,
      "time_per_iteration": 2.8973679542541504
    },
    {
      "auxiliary_loss_clip": 0.01097921,
      "auxiliary_loss_mlp": 0.01034684,
      "balance_loss_clip": 1.04556656,
      "balance_loss_mlp": 1.0192461,
      "epoch": 0.382444010220953,
      "flos": 24936190352640.0,
      "grad_norm": 1.5966238664745025,
      "language_loss": 0.79015481,
      "learning_rate": 2.83244000399261e-06,
      "loss": 0.81148088,
      "num_input_tokens_seen": 136487215,
      "step": 6361,
      "time_per_iteration": 2.568021059036255
    },
    {
      "auxiliary_loss_clip": 0.01103455,
      "auxiliary_loss_mlp": 0.01037569,
      "balance_loss_clip": 1.04402626,
      "balance_loss_mlp": 1.02366877,
      "epoch": 0.38250413347362094,
      "flos": 42337351209600.0,
      "grad_norm": 1.5130039546850633,
      "language_loss": 0.65356135,
      "learning_rate": 2.832085864749337e-06,
      "loss": 0.67497158,
      "num_input_tokens_seen": 136510365,
      "step": 6362,
      "time_per_iteration": 2.700993061065674
    },
    {
      "auxiliary_loss_clip": 0.01126461,
      "auxiliary_loss_mlp": 0.01035978,
      "balance_loss_clip": 1.04516792,
      "balance_loss_mlp": 1.01991487,
      "epoch": 0.3825642567262889,
      "flos": 16289224462080.0,
      "grad_norm": 1.7438183058183454,
      "language_loss": 0.81665468,
      "learning_rate": 2.8317316939532848e-06,
      "loss": 0.83827907,
      "num_input_tokens_seen": 136527100,
      "step": 6363,
      "time_per_iteration": 2.436859607696533
    },
    {
      "auxiliary_loss_clip": 0.01073779,
      "auxiliary_loss_mlp": 0.01041921,
      "balance_loss_clip": 1.04942608,
      "balance_loss_mlp": 1.02688277,
      "epoch": 0.3826243799789569,
      "flos": 45654778586880.0,
      "grad_norm": 1.6738996421412173,
      "language_loss": 0.58484823,
      "learning_rate": 2.8313774916178825e-06,
      "loss": 0.60600525,
      "num_input_tokens_seen": 136550870,
      "step": 6364,
      "time_per_iteration": 2.7961654663085938
    },
    {
      "auxiliary_loss_clip": 0.01110479,
      "auxiliary_loss_mlp": 0.01037295,
      "balance_loss_clip": 1.04545116,
      "balance_loss_mlp": 1.02228069,
      "epoch": 0.38268450323162484,
      "flos": 25301581453440.0,
      "grad_norm": 1.7900467123771298,
      "language_loss": 0.68722427,
      "learning_rate": 2.8310232577565635e-06,
      "loss": 0.70870203,
      "num_input_tokens_seen": 136569895,
      "step": 6365,
      "time_per_iteration": 2.5532028675079346
    },
    {
      "auxiliary_loss_clip": 0.01113518,
      "auxiliary_loss_mlp": 0.01035949,
      "balance_loss_clip": 1.0451206,
      "balance_loss_mlp": 1.02028513,
      "epoch": 0.3827446264842928,
      "flos": 21836022387840.0,
      "grad_norm": 1.8114283439105476,
      "language_loss": 0.7350021,
      "learning_rate": 2.830668992382758e-06,
      "loss": 0.75649679,
      "num_input_tokens_seen": 136588585,
      "step": 6366,
      "time_per_iteration": 2.488314628601074
    },
    {
      "auxiliary_loss_clip": 0.01107329,
      "auxiliary_loss_mlp": 0.01038246,
      "balance_loss_clip": 1.04560494,
      "balance_loss_mlp": 1.02299929,
      "epoch": 0.38280474973696077,
      "flos": 25734591907200.0,
      "grad_norm": 2.2641980514583686,
      "language_loss": 0.6836859,
      "learning_rate": 2.830314695509902e-06,
      "loss": 0.70514166,
      "num_input_tokens_seen": 136606640,
      "step": 6367,
      "time_per_iteration": 2.571965217590332
    },
    {
      "auxiliary_loss_clip": 0.0111666,
      "auxiliary_loss_mlp": 0.01033853,
      "balance_loss_clip": 1.04687929,
      "balance_loss_mlp": 1.01847494,
      "epoch": 0.38286487298962874,
      "flos": 24895934184960.0,
      "grad_norm": 1.9184713907149546,
      "language_loss": 0.64226711,
      "learning_rate": 2.82996036715143e-06,
      "loss": 0.66377217,
      "num_input_tokens_seen": 136624940,
      "step": 6368,
      "time_per_iteration": 3.9741387367248535
    },
    {
      "auxiliary_loss_clip": 0.01128962,
      "auxiliary_loss_mlp": 0.0103615,
      "balance_loss_clip": 1.04693556,
      "balance_loss_mlp": 1.02070045,
      "epoch": 0.3829249962422967,
      "flos": 28543703967360.0,
      "grad_norm": 1.3524652263270613,
      "language_loss": 0.68305594,
      "learning_rate": 2.8296060073207763e-06,
      "loss": 0.70470709,
      "num_input_tokens_seen": 136645540,
      "step": 6369,
      "time_per_iteration": 2.531676769256592
    },
    {
      "auxiliary_loss_clip": 0.01078835,
      "auxiliary_loss_mlp": 0.01043913,
      "balance_loss_clip": 1.04402494,
      "balance_loss_mlp": 1.02827251,
      "epoch": 0.38298511949496467,
      "flos": 21471205904640.0,
      "grad_norm": 1.8410065133850892,
      "language_loss": 0.786807,
      "learning_rate": 2.8292516160313804e-06,
      "loss": 0.80803442,
      "num_input_tokens_seen": 136664530,
      "step": 6370,
      "time_per_iteration": 2.574394464492798
    },
    {
      "auxiliary_loss_clip": 0.01112429,
      "auxiliary_loss_mlp": 0.01046485,
      "balance_loss_clip": 1.04536867,
      "balance_loss_mlp": 1.03005815,
      "epoch": 0.38304524274763263,
      "flos": 31679998035840.0,
      "grad_norm": 2.6794026258987405,
      "language_loss": 0.6500808,
      "learning_rate": 2.8288971932966805e-06,
      "loss": 0.6716699,
      "num_input_tokens_seen": 136682315,
      "step": 6371,
      "time_per_iteration": 3.9441590309143066
    },
    {
      "auxiliary_loss_clip": 0.01100924,
      "auxiliary_loss_mlp": 0.01037599,
      "balance_loss_clip": 1.04605496,
      "balance_loss_mlp": 1.0216012,
      "epoch": 0.3831053660003006,
      "flos": 25076816098560.0,
      "grad_norm": 2.1885725187002927,
      "language_loss": 0.72838646,
      "learning_rate": 2.8285427391301155e-06,
      "loss": 0.74977171,
      "num_input_tokens_seen": 136701185,
      "step": 6372,
      "time_per_iteration": 3.989328622817993
    },
    {
      "auxiliary_loss_clip": 0.01119035,
      "auxiliary_loss_mlp": 0.01034174,
      "balance_loss_clip": 1.04696989,
      "balance_loss_mlp": 1.01899862,
      "epoch": 0.38316548925296856,
      "flos": 23259018562560.0,
      "grad_norm": 2.0335961211755293,
      "language_loss": 0.85167766,
      "learning_rate": 2.8281882535451266e-06,
      "loss": 0.87320971,
      "num_input_tokens_seen": 136721265,
      "step": 6373,
      "time_per_iteration": 2.516111135482788
    },
    {
      "auxiliary_loss_clip": 0.01079705,
      "auxiliary_loss_mlp": 0.01045943,
      "balance_loss_clip": 1.04332924,
      "balance_loss_mlp": 1.02956963,
      "epoch": 0.3832256125056366,
      "flos": 34423465991040.0,
      "grad_norm": 1.938711263273911,
      "language_loss": 0.75175524,
      "learning_rate": 2.8278337365551567e-06,
      "loss": 0.77301168,
      "num_input_tokens_seen": 136741885,
      "step": 6374,
      "time_per_iteration": 2.6831583976745605
    },
    {
      "auxiliary_loss_clip": 0.01128834,
      "auxiliary_loss_mlp": 0.01036446,
      "balance_loss_clip": 1.05461001,
      "balance_loss_mlp": 1.02201617,
      "epoch": 0.38328573575830455,
      "flos": 21762764599680.0,
      "grad_norm": 2.2257315899505437,
      "language_loss": 0.75724936,
      "learning_rate": 2.8274791881736485e-06,
      "loss": 0.77890217,
      "num_input_tokens_seen": 136760905,
      "step": 6375,
      "time_per_iteration": 3.9062957763671875
    },
    {
      "auxiliary_loss_clip": 0.01117054,
      "auxiliary_loss_mlp": 0.01038314,
      "balance_loss_clip": 1.04728484,
      "balance_loss_mlp": 1.02327561,
      "epoch": 0.3833458590109725,
      "flos": 17380010724480.0,
      "grad_norm": 1.8162715410607537,
      "language_loss": 0.7216894,
      "learning_rate": 2.8271246084140457e-06,
      "loss": 0.7432431,
      "num_input_tokens_seen": 136777240,
      "step": 6376,
      "time_per_iteration": 2.446741819381714
    },
    {
      "auxiliary_loss_clip": 0.01115881,
      "auxiliary_loss_mlp": 0.01037744,
      "balance_loss_clip": 1.04697669,
      "balance_loss_mlp": 1.022282,
      "epoch": 0.3834059822636405,
      "flos": 29424557191680.0,
      "grad_norm": 1.5371758587663247,
      "language_loss": 0.67886591,
      "learning_rate": 2.826769997289796e-06,
      "loss": 0.70040214,
      "num_input_tokens_seen": 136801040,
      "step": 6377,
      "time_per_iteration": 2.5659170150756836
    },
    {
      "auxiliary_loss_clip": 0.01100986,
      "auxiliary_loss_mlp": 0.01041064,
      "balance_loss_clip": 1.04872131,
      "balance_loss_mlp": 1.02530479,
      "epoch": 0.38346610551630844,
      "flos": 21470739027840.0,
      "grad_norm": 2.746369077639069,
      "language_loss": 0.73632789,
      "learning_rate": 2.826415354814344e-06,
      "loss": 0.75774837,
      "num_input_tokens_seen": 136819495,
      "step": 6378,
      "time_per_iteration": 2.5435080528259277
    },
    {
      "auxiliary_loss_clip": 0.01082783,
      "auxiliary_loss_mlp": 0.01034378,
      "balance_loss_clip": 1.04783702,
      "balance_loss_mlp": 1.01997805,
      "epoch": 0.3835262287689764,
      "flos": 27561224188800.0,
      "grad_norm": 1.708181224550276,
      "language_loss": 0.69499195,
      "learning_rate": 2.8260606810011396e-06,
      "loss": 0.71616364,
      "num_input_tokens_seen": 136838840,
      "step": 6379,
      "time_per_iteration": 2.649644613265991
    },
    {
      "auxiliary_loss_clip": 0.01116496,
      "auxiliary_loss_mlp": 0.01037132,
      "balance_loss_clip": 1.04803908,
      "balance_loss_mlp": 1.02224267,
      "epoch": 0.3835863520216444,
      "flos": 15523716787200.0,
      "grad_norm": 2.081648052641886,
      "language_loss": 0.83040178,
      "learning_rate": 2.8257059758636315e-06,
      "loss": 0.85193801,
      "num_input_tokens_seen": 136854425,
      "step": 6380,
      "time_per_iteration": 2.475816249847412
    },
    {
      "auxiliary_loss_clip": 0.01126875,
      "auxiliary_loss_mlp": 0.01034491,
      "balance_loss_clip": 1.04814756,
      "balance_loss_mlp": 1.02027488,
      "epoch": 0.38364647527431234,
      "flos": 21904934630400.0,
      "grad_norm": 1.5878223475344093,
      "language_loss": 0.81285727,
      "learning_rate": 2.8253512394152697e-06,
      "loss": 0.83447093,
      "num_input_tokens_seen": 136874355,
      "step": 6381,
      "time_per_iteration": 2.483790636062622
    },
    {
      "auxiliary_loss_clip": 0.01058466,
      "auxiliary_loss_mlp": 0.01004256,
      "balance_loss_clip": 1.02911222,
      "balance_loss_mlp": 1.00265872,
      "epoch": 0.3837065985269803,
      "flos": 65534927558400.0,
      "grad_norm": 0.7892914286643401,
      "language_loss": 0.60402787,
      "learning_rate": 2.8249964716695068e-06,
      "loss": 0.62465501,
      "num_input_tokens_seen": 136937475,
      "step": 6382,
      "time_per_iteration": 3.074282646179199
    },
    {
      "auxiliary_loss_clip": 0.01128838,
      "auxiliary_loss_mlp": 0.01034878,
      "balance_loss_clip": 1.04534221,
      "balance_loss_mlp": 1.01952386,
      "epoch": 0.38376672177964827,
      "flos": 28256598558720.0,
      "grad_norm": 2.4611504408501124,
      "language_loss": 0.67183125,
      "learning_rate": 2.824641672639794e-06,
      "loss": 0.69346839,
      "num_input_tokens_seen": 136955805,
      "step": 6383,
      "time_per_iteration": 2.5106618404388428
    },
    {
      "auxiliary_loss_clip": 0.01093544,
      "auxiliary_loss_mlp": 0.01036183,
      "balance_loss_clip": 1.0441606,
      "balance_loss_mlp": 1.02139533,
      "epoch": 0.38382684503231623,
      "flos": 20631363033600.0,
      "grad_norm": 1.6752188064510494,
      "language_loss": 0.75306666,
      "learning_rate": 2.824286842339587e-06,
      "loss": 0.77436388,
      "num_input_tokens_seen": 136975240,
      "step": 6384,
      "time_per_iteration": 2.5614356994628906
    },
    {
      "auxiliary_loss_clip": 0.01111901,
      "auxiliary_loss_mlp": 0.01034882,
      "balance_loss_clip": 1.04854298,
      "balance_loss_mlp": 1.02061248,
      "epoch": 0.3838869682849842,
      "flos": 19605825826560.0,
      "grad_norm": 1.3917112743237527,
      "language_loss": 0.76320696,
      "learning_rate": 2.823931980782341e-06,
      "loss": 0.78467476,
      "num_input_tokens_seen": 136994985,
      "step": 6385,
      "time_per_iteration": 2.4885659217834473
    },
    {
      "auxiliary_loss_clip": 0.01044903,
      "auxiliary_loss_mlp": 0.01005726,
      "balance_loss_clip": 1.02653658,
      "balance_loss_mlp": 1.00408077,
      "epoch": 0.38394709153765216,
      "flos": 56556110891520.0,
      "grad_norm": 0.9159290374671689,
      "language_loss": 0.67059577,
      "learning_rate": 2.82357708798151e-06,
      "loss": 0.69110203,
      "num_input_tokens_seen": 137046290,
      "step": 6386,
      "time_per_iteration": 2.9929840564727783
    },
    {
      "auxiliary_loss_clip": 0.01088404,
      "auxiliary_loss_mlp": 0.01037841,
      "balance_loss_clip": 1.04442811,
      "balance_loss_mlp": 1.02351761,
      "epoch": 0.3840072147903202,
      "flos": 15888748752000.0,
      "grad_norm": 1.6690337183295485,
      "language_loss": 0.72564745,
      "learning_rate": 2.8232221639505547e-06,
      "loss": 0.74690986,
      "num_input_tokens_seen": 137064725,
      "step": 6387,
      "time_per_iteration": 2.5338666439056396
    },
    {
      "auxiliary_loss_clip": 0.0112576,
      "auxiliary_loss_mlp": 0.01038222,
      "balance_loss_clip": 1.04753447,
      "balance_loss_mlp": 1.0242269,
      "epoch": 0.38406733804298815,
      "flos": 28218030330240.0,
      "grad_norm": 1.9919553779210797,
      "language_loss": 0.81110233,
      "learning_rate": 2.822867208702932e-06,
      "loss": 0.83274215,
      "num_input_tokens_seen": 137086030,
      "step": 6388,
      "time_per_iteration": 2.531886100769043
    },
    {
      "auxiliary_loss_clip": 0.01092788,
      "auxiliary_loss_mlp": 0.01038434,
      "balance_loss_clip": 1.04276872,
      "balance_loss_mlp": 1.02529097,
      "epoch": 0.3841274612956561,
      "flos": 18223588609920.0,
      "grad_norm": 1.6872158100890784,
      "language_loss": 0.75988734,
      "learning_rate": 2.8225122222521026e-06,
      "loss": 0.78119957,
      "num_input_tokens_seen": 137105400,
      "step": 6389,
      "time_per_iteration": 2.5711817741394043
    },
    {
      "auxiliary_loss_clip": 0.01106041,
      "auxiliary_loss_mlp": 0.01049581,
      "balance_loss_clip": 1.04854417,
      "balance_loss_mlp": 1.03378606,
      "epoch": 0.3841875845483241,
      "flos": 19792884879360.0,
      "grad_norm": 1.621448520199094,
      "language_loss": 0.76619875,
      "learning_rate": 2.8221572046115273e-06,
      "loss": 0.78775495,
      "num_input_tokens_seen": 137124985,
      "step": 6390,
      "time_per_iteration": 2.528108596801758
    },
    {
      "auxiliary_loss_clip": 0.01078395,
      "auxiliary_loss_mlp": 0.0104493,
      "balance_loss_clip": 1.04083443,
      "balance_loss_mlp": 1.02973688,
      "epoch": 0.38424770780099204,
      "flos": 29898829393920.0,
      "grad_norm": 1.7095664629482443,
      "language_loss": 0.69603896,
      "learning_rate": 2.821802155794668e-06,
      "loss": 0.71727222,
      "num_input_tokens_seen": 137146745,
      "step": 6391,
      "time_per_iteration": 2.682647466659546
    },
    {
      "auxiliary_loss_clip": 0.01115273,
      "auxiliary_loss_mlp": 0.01035545,
      "balance_loss_clip": 1.04442716,
      "balance_loss_mlp": 1.02120423,
      "epoch": 0.38430783105366,
      "flos": 20813717404800.0,
      "grad_norm": 1.7733292157718905,
      "language_loss": 0.83981049,
      "learning_rate": 2.8214470758149884e-06,
      "loss": 0.86131859,
      "num_input_tokens_seen": 137163195,
      "step": 6392,
      "time_per_iteration": 2.5066070556640625
    },
    {
      "auxiliary_loss_clip": 0.01111645,
      "auxiliary_loss_mlp": 0.01040523,
      "balance_loss_clip": 1.04268062,
      "balance_loss_mlp": 1.02668846,
      "epoch": 0.384367954306328,
      "flos": 10998577399680.0,
      "grad_norm": 2.3926563810540564,
      "language_loss": 0.60886067,
      "learning_rate": 2.8210919646859536e-06,
      "loss": 0.63038236,
      "num_input_tokens_seen": 137179330,
      "step": 6393,
      "time_per_iteration": 2.463325262069702
    },
    {
      "auxiliary_loss_clip": 0.01098975,
      "auxiliary_loss_mlp": 0.01033872,
      "balance_loss_clip": 1.0439328,
      "balance_loss_mlp": 1.01863742,
      "epoch": 0.38442807755899594,
      "flos": 25338030779520.0,
      "grad_norm": 1.7679377570466532,
      "language_loss": 0.71011388,
      "learning_rate": 2.820736822421029e-06,
      "loss": 0.73144233,
      "num_input_tokens_seen": 137198655,
      "step": 6394,
      "time_per_iteration": 2.593858003616333
    },
    {
      "auxiliary_loss_clip": 0.01116553,
      "auxiliary_loss_mlp": 0.01034707,
      "balance_loss_clip": 1.04613733,
      "balance_loss_mlp": 1.01891172,
      "epoch": 0.3844882008116639,
      "flos": 21069760527360.0,
      "grad_norm": 2.85438184062275,
      "language_loss": 0.8102147,
      "learning_rate": 2.8203816490336822e-06,
      "loss": 0.83172727,
      "num_input_tokens_seen": 137217120,
      "step": 6395,
      "time_per_iteration": 2.5121679306030273
    },
    {
      "auxiliary_loss_clip": 0.01125905,
      "auxiliary_loss_mlp": 0.01043866,
      "balance_loss_clip": 1.05387044,
      "balance_loss_mlp": 1.02995396,
      "epoch": 0.38454832406433187,
      "flos": 17963235855360.0,
      "grad_norm": 1.9148805892792504,
      "language_loss": 0.70467496,
      "learning_rate": 2.8200264445373813e-06,
      "loss": 0.72637272,
      "num_input_tokens_seen": 137234410,
      "step": 6396,
      "time_per_iteration": 2.460824966430664
    },
    {
      "auxiliary_loss_clip": 0.01043402,
      "auxiliary_loss_mlp": 0.01001416,
      "balance_loss_clip": 1.02938461,
      "balance_loss_mlp": 0.99967539,
      "epoch": 0.38460844731699984,
      "flos": 67924999555200.0,
      "grad_norm": 0.8891862899202012,
      "language_loss": 0.59627998,
      "learning_rate": 2.8196712089455954e-06,
      "loss": 0.61672819,
      "num_input_tokens_seen": 137294940,
      "step": 6397,
      "time_per_iteration": 3.1644768714904785
    },
    {
      "auxiliary_loss_clip": 0.01127763,
      "auxiliary_loss_mlp": 0.0102608,
      "balance_loss_clip": 1.04755533,
      "balance_loss_mlp": 1.01174498,
      "epoch": 0.3846685705696678,
      "flos": 25849075530240.0,
      "grad_norm": 1.8033056431073977,
      "language_loss": 0.84768987,
      "learning_rate": 2.819315942271794e-06,
      "loss": 0.86922824,
      "num_input_tokens_seen": 137315035,
      "step": 6398,
      "time_per_iteration": 2.5350446701049805
    },
    {
      "auxiliary_loss_clip": 0.0112694,
      "auxiliary_loss_mlp": 0.01028067,
      "balance_loss_clip": 1.04688048,
      "balance_loss_mlp": 1.01450109,
      "epoch": 0.38472869382233577,
      "flos": 16290194129280.0,
      "grad_norm": 1.9369153608860867,
      "language_loss": 0.79882693,
      "learning_rate": 2.8189606445294515e-06,
      "loss": 0.82037699,
      "num_input_tokens_seen": 137333155,
      "step": 6399,
      "time_per_iteration": 2.450942277908325
    },
    {
      "auxiliary_loss_clip": 0.0112822,
      "auxiliary_loss_mlp": 0.0078985,
      "balance_loss_clip": 1.04645252,
      "balance_loss_mlp": 1.0103178,
      "epoch": 0.38478881707500373,
      "flos": 19353122668800.0,
      "grad_norm": 1.8294283859048401,
      "language_loss": 0.67144096,
      "learning_rate": 2.818605315732038e-06,
      "loss": 0.69062161,
      "num_input_tokens_seen": 137351515,
      "step": 6400,
      "time_per_iteration": 2.456462860107422
    },
    {
      "auxiliary_loss_clip": 0.0110828,
      "auxiliary_loss_mlp": 0.01045962,
      "balance_loss_clip": 1.04651487,
      "balance_loss_mlp": 1.03120399,
      "epoch": 0.38484894032767175,
      "flos": 24860849575680.0,
      "grad_norm": 1.6795876721854746,
      "language_loss": 0.73175955,
      "learning_rate": 2.81824995589303e-06,
      "loss": 0.75330198,
      "num_input_tokens_seen": 137371255,
      "step": 6401,
      "time_per_iteration": 2.544548988342285
    },
    {
      "auxiliary_loss_clip": 0.01092593,
      "auxiliary_loss_mlp": 0.01036682,
      "balance_loss_clip": 1.04587924,
      "balance_loss_mlp": 1.02230573,
      "epoch": 0.3849090635803397,
      "flos": 14501806853760.0,
      "grad_norm": 2.447028324220019,
      "language_loss": 0.72070253,
      "learning_rate": 2.8178945650259012e-06,
      "loss": 0.74199533,
      "num_input_tokens_seen": 137388980,
      "step": 6402,
      "time_per_iteration": 2.540895462036133
    },
    {
      "auxiliary_loss_clip": 0.01122974,
      "auxiliary_loss_mlp": 0.01031177,
      "balance_loss_clip": 1.04454374,
      "balance_loss_mlp": 1.01748633,
      "epoch": 0.3849691868330077,
      "flos": 18515865576960.0,
      "grad_norm": 1.9230008282799038,
      "language_loss": 0.82806671,
      "learning_rate": 2.817539143144128e-06,
      "loss": 0.84960818,
      "num_input_tokens_seen": 137406885,
      "step": 6403,
      "time_per_iteration": 2.4402687549591064
    },
    {
      "auxiliary_loss_clip": 0.01072141,
      "auxiliary_loss_mlp": 0.01035319,
      "balance_loss_clip": 1.04378819,
      "balance_loss_mlp": 1.02001894,
      "epoch": 0.38502931008567565,
      "flos": 21616392677760.0,
      "grad_norm": 1.9048316134222274,
      "language_loss": 0.82962525,
      "learning_rate": 2.817183690261189e-06,
      "loss": 0.8506999,
      "num_input_tokens_seen": 137425535,
      "step": 6404,
      "time_per_iteration": 2.5846331119537354
    },
    {
      "auxiliary_loss_clip": 0.01106865,
      "auxiliary_loss_mlp": 0.01030739,
      "balance_loss_clip": 1.04806674,
      "balance_loss_mlp": 1.01683903,
      "epoch": 0.3850894333383436,
      "flos": 25415346804480.0,
      "grad_norm": 1.6868844228486704,
      "language_loss": 0.69556075,
      "learning_rate": 2.816828206390563e-06,
      "loss": 0.71693677,
      "num_input_tokens_seen": 137447700,
      "step": 6405,
      "time_per_iteration": 2.6025304794311523
    },
    {
      "auxiliary_loss_clip": 0.01096874,
      "auxiliary_loss_mlp": 0.01035062,
      "balance_loss_clip": 1.04440904,
      "balance_loss_mlp": 1.02181792,
      "epoch": 0.3851495565910116,
      "flos": 20227870581120.0,
      "grad_norm": 2.1498616708778346,
      "language_loss": 0.79102367,
      "learning_rate": 2.816472691545729e-06,
      "loss": 0.81234312,
      "num_input_tokens_seen": 137462245,
      "step": 6406,
      "time_per_iteration": 2.505293607711792
    },
    {
      "auxiliary_loss_clip": 0.01115212,
      "auxiliary_loss_mlp": 0.01035923,
      "balance_loss_clip": 1.04609013,
      "balance_loss_mlp": 1.02141571,
      "epoch": 0.38520967984367954,
      "flos": 16508459122560.0,
      "grad_norm": 2.1188889876951467,
      "language_loss": 0.83819681,
      "learning_rate": 2.8161171457401694e-06,
      "loss": 0.85970819,
      "num_input_tokens_seen": 137476455,
      "step": 6407,
      "time_per_iteration": 3.834929943084717
    },
    {
      "auxiliary_loss_clip": 0.01046013,
      "auxiliary_loss_mlp": 0.01009963,
      "balance_loss_clip": 1.02709234,
      "balance_loss_mlp": 1.00821102,
      "epoch": 0.3852698030963475,
      "flos": 61313772971520.0,
      "grad_norm": 0.8448543093117669,
      "language_loss": 0.64932764,
      "learning_rate": 2.815761568987365e-06,
      "loss": 0.66988742,
      "num_input_tokens_seen": 137539845,
      "step": 6408,
      "time_per_iteration": 3.1531708240509033
    },
    {
      "auxiliary_loss_clip": 0.01095217,
      "auxiliary_loss_mlp": 0.01040754,
      "balance_loss_clip": 1.04657722,
      "balance_loss_mlp": 1.02563834,
      "epoch": 0.3853299263490155,
      "flos": 22893016930560.0,
      "grad_norm": 1.4713514769925826,
      "language_loss": 0.7336688,
      "learning_rate": 2.8154059613008e-06,
      "loss": 0.75502855,
      "num_input_tokens_seen": 137559880,
      "step": 6409,
      "time_per_iteration": 2.555018663406372
    },
    {
      "auxiliary_loss_clip": 0.01084014,
      "auxiliary_loss_mlp": 0.01042111,
      "balance_loss_clip": 1.04662347,
      "balance_loss_mlp": 1.02579093,
      "epoch": 0.38539004960168344,
      "flos": 20047491457920.0,
      "grad_norm": 3.1923623249760866,
      "language_loss": 0.70293194,
      "learning_rate": 2.81505032269396e-06,
      "loss": 0.72419322,
      "num_input_tokens_seen": 137578225,
      "step": 6410,
      "time_per_iteration": 3.9310505390167236
    },
    {
      "auxiliary_loss_clip": 0.01019263,
      "auxiliary_loss_mlp": 0.00826258,
      "balance_loss_clip": 1.03001952,
      "balance_loss_mlp": 1.11101174,
      "epoch": 0.3854501728543514,
      "flos": 68730691570560.0,
      "grad_norm": 0.697355969802029,
      "language_loss": 0.60291171,
      "learning_rate": 2.81469465318033e-06,
      "loss": 0.62136698,
      "num_input_tokens_seen": 137645770,
      "step": 6411,
      "time_per_iteration": 4.675171136856079
    },
    {
      "auxiliary_loss_clip": 0.0108134,
      "auxiliary_loss_mlp": 0.01031051,
      "balance_loss_clip": 1.04053092,
      "balance_loss_mlp": 1.01733541,
      "epoch": 0.38551029610701937,
      "flos": 20485027025280.0,
      "grad_norm": 2.634265357709159,
      "language_loss": 0.77659267,
      "learning_rate": 2.814338952773397e-06,
      "loss": 0.7977165,
      "num_input_tokens_seen": 137664090,
      "step": 6412,
      "time_per_iteration": 2.5629477500915527
    },
    {
      "auxiliary_loss_clip": 0.01090303,
      "auxiliary_loss_mlp": 0.01033987,
      "balance_loss_clip": 1.04231524,
      "balance_loss_mlp": 1.0174526,
      "epoch": 0.38557041935968733,
      "flos": 23471788775040.0,
      "grad_norm": 1.8064533287691518,
      "language_loss": 0.78217995,
      "learning_rate": 2.8139832214866493e-06,
      "loss": 0.80342281,
      "num_input_tokens_seen": 137683190,
      "step": 6413,
      "time_per_iteration": 3.9412789344787598
    },
    {
      "auxiliary_loss_clip": 0.01054894,
      "auxiliary_loss_mlp": 0.01005119,
      "balance_loss_clip": 1.02594435,
      "balance_loss_mlp": 1.00322402,
      "epoch": 0.38563054261235535,
      "flos": 63966636869760.0,
      "grad_norm": 0.8091090617658498,
      "language_loss": 0.61364746,
      "learning_rate": 2.813627459333576e-06,
      "loss": 0.6342476,
      "num_input_tokens_seen": 137737315,
      "step": 6414,
      "time_per_iteration": 2.9110121726989746
    },
    {
      "auxiliary_loss_clip": 0.01093371,
      "auxiliary_loss_mlp": 0.01041406,
      "balance_loss_clip": 1.04798675,
      "balance_loss_mlp": 1.02699387,
      "epoch": 0.3856906658650233,
      "flos": 23987789602560.0,
      "grad_norm": 2.050018804744751,
      "language_loss": 0.77890933,
      "learning_rate": 2.8132716663276685e-06,
      "loss": 0.80025709,
      "num_input_tokens_seen": 137753535,
      "step": 6415,
      "time_per_iteration": 2.563208818435669
    },
    {
      "auxiliary_loss_clip": 0.01099374,
      "auxiliary_loss_mlp": 0.01030509,
      "balance_loss_clip": 1.04486275,
      "balance_loss_mlp": 1.01790822,
      "epoch": 0.3857507891176913,
      "flos": 25007436979200.0,
      "grad_norm": 1.6752205860148979,
      "language_loss": 0.79720467,
      "learning_rate": 2.8129158424824173e-06,
      "loss": 0.8185035,
      "num_input_tokens_seen": 137773405,
      "step": 6416,
      "time_per_iteration": 2.5670573711395264
    },
    {
      "auxiliary_loss_clip": 0.01117372,
      "auxiliary_loss_mlp": 0.00826688,
      "balance_loss_clip": 1.04488778,
      "balance_loss_mlp": 1.08185816,
      "epoch": 0.38581091237035925,
      "flos": 21536778182400.0,
      "grad_norm": 2.621507438099194,
      "language_loss": 0.79025185,
      "learning_rate": 2.8125599878113155e-06,
      "loss": 0.80969238,
      "num_input_tokens_seen": 137790810,
      "step": 6417,
      "time_per_iteration": 2.495215892791748
    },
    {
      "auxiliary_loss_clip": 0.01095364,
      "auxiliary_loss_mlp": 0.01034102,
      "balance_loss_clip": 1.04224634,
      "balance_loss_mlp": 1.02094722,
      "epoch": 0.3858710356230272,
      "flos": 17383889393280.0,
      "grad_norm": 2.8377962582679923,
      "language_loss": 0.80012798,
      "learning_rate": 2.8122041023278583e-06,
      "loss": 0.8214227,
      "num_input_tokens_seen": 137810265,
      "step": 6418,
      "time_per_iteration": 2.5195884704589844
    },
    {
      "auxiliary_loss_clip": 0.01094017,
      "auxiliary_loss_mlp": 0.010296,
      "balance_loss_clip": 1.04240453,
      "balance_loss_mlp": 1.01675534,
      "epoch": 0.3859311588756952,
      "flos": 20339588856960.0,
      "grad_norm": 1.891950119925469,
      "language_loss": 0.79742724,
      "learning_rate": 2.8118481860455407e-06,
      "loss": 0.81866336,
      "num_input_tokens_seen": 137828580,
      "step": 6419,
      "time_per_iteration": 2.5282838344573975
    },
    {
      "auxiliary_loss_clip": 0.01099853,
      "auxiliary_loss_mlp": 0.01030153,
      "balance_loss_clip": 1.04338276,
      "balance_loss_mlp": 1.01533544,
      "epoch": 0.38599128212836314,
      "flos": 26321157002880.0,
      "grad_norm": 5.137815322181607,
      "language_loss": 0.67619896,
      "learning_rate": 2.8114922389778573e-06,
      "loss": 0.69749904,
      "num_input_tokens_seen": 137846145,
      "step": 6420,
      "time_per_iteration": 2.5504016876220703
    },
    {
      "auxiliary_loss_clip": 0.01080617,
      "auxiliary_loss_mlp": 0.01040635,
      "balance_loss_clip": 1.0442965,
      "balance_loss_mlp": 1.02607918,
      "epoch": 0.3860514053810311,
      "flos": 13553837066880.0,
      "grad_norm": 1.8462888562357171,
      "language_loss": 0.80995136,
      "learning_rate": 2.8111362611383076e-06,
      "loss": 0.83116388,
      "num_input_tokens_seen": 137863705,
      "step": 6421,
      "time_per_iteration": 2.540515422821045
    },
    {
      "auxiliary_loss_clip": 0.01097788,
      "auxiliary_loss_mlp": 0.01035289,
      "balance_loss_clip": 1.04285395,
      "balance_loss_mlp": 1.02117455,
      "epoch": 0.3861115286336991,
      "flos": 20954271323520.0,
      "grad_norm": 2.2504239316724295,
      "language_loss": 0.72204518,
      "learning_rate": 2.8107802525403886e-06,
      "loss": 0.74337596,
      "num_input_tokens_seen": 137880285,
      "step": 6422,
      "time_per_iteration": 2.4913759231567383
    },
    {
      "auxiliary_loss_clip": 0.01097026,
      "auxiliary_loss_mlp": 0.01036105,
      "balance_loss_clip": 1.04500628,
      "balance_loss_mlp": 1.02296257,
      "epoch": 0.38617165188636704,
      "flos": 16362697731840.0,
      "grad_norm": 1.8788961206036923,
      "language_loss": 0.667135,
      "learning_rate": 2.8104242131976025e-06,
      "loss": 0.68846631,
      "num_input_tokens_seen": 137898335,
      "step": 6423,
      "time_per_iteration": 2.5088589191436768
    },
    {
      "auxiliary_loss_clip": 0.01118285,
      "auxiliary_loss_mlp": 0.01039412,
      "balance_loss_clip": 1.04683709,
      "balance_loss_mlp": 1.02602458,
      "epoch": 0.386231775139035,
      "flos": 34787276893440.0,
      "grad_norm": 2.3321006399855415,
      "language_loss": 0.68705982,
      "learning_rate": 2.810068143123449e-06,
      "loss": 0.70863676,
      "num_input_tokens_seen": 137918605,
      "step": 6424,
      "time_per_iteration": 2.607577085494995
    },
    {
      "auxiliary_loss_clip": 0.01091057,
      "auxiliary_loss_mlp": 0.01034873,
      "balance_loss_clip": 1.04500055,
      "balance_loss_mlp": 1.02098525,
      "epoch": 0.38629189839170297,
      "flos": 21726171619200.0,
      "grad_norm": 1.3686456970652345,
      "language_loss": 0.72535127,
      "learning_rate": 2.809712042331429e-06,
      "loss": 0.74661058,
      "num_input_tokens_seen": 137938245,
      "step": 6425,
      "time_per_iteration": 2.5879650115966797
    },
    {
      "auxiliary_loss_clip": 0.01092373,
      "auxiliary_loss_mlp": 0.00843651,
      "balance_loss_clip": 1.04210281,
      "balance_loss_mlp": 1.10785317,
      "epoch": 0.38635202164437094,
      "flos": 27923634460800.0,
      "grad_norm": 2.3596081162857896,
      "language_loss": 0.80769169,
      "learning_rate": 2.8093559108350484e-06,
      "loss": 0.827052,
      "num_input_tokens_seen": 137956770,
      "step": 6426,
      "time_per_iteration": 2.589916229248047
    },
    {
      "auxiliary_loss_clip": 0.01116973,
      "auxiliary_loss_mlp": 0.01032919,
      "balance_loss_clip": 1.04607701,
      "balance_loss_mlp": 1.01857805,
      "epoch": 0.38641214489703896,
      "flos": 23586631534080.0,
      "grad_norm": 2.475049776675903,
      "language_loss": 0.74423909,
      "learning_rate": 2.80899974864781e-06,
      "loss": 0.76573795,
      "num_input_tokens_seen": 137977040,
      "step": 6427,
      "time_per_iteration": 2.530529022216797
    },
    {
      "auxiliary_loss_clip": 0.01069416,
      "auxiliary_loss_mlp": 0.01046206,
      "balance_loss_clip": 1.04116869,
      "balance_loss_mlp": 1.02990997,
      "epoch": 0.3864722681497069,
      "flos": 12641239198080.0,
      "grad_norm": 2.607437988584307,
      "language_loss": 0.70625383,
      "learning_rate": 2.8086435557832203e-06,
      "loss": 0.72741008,
      "num_input_tokens_seen": 137993545,
      "step": 6428,
      "time_per_iteration": 2.5679309368133545
    },
    {
      "auxiliary_loss_clip": 0.01106689,
      "auxiliary_loss_mlp": 0.010446,
      "balance_loss_clip": 1.04696202,
      "balance_loss_mlp": 1.03084278,
      "epoch": 0.3865323914023749,
      "flos": 17598922162560.0,
      "grad_norm": 2.0686906074719342,
      "language_loss": 0.84260613,
      "learning_rate": 2.8082873322547863e-06,
      "loss": 0.86411899,
      "num_input_tokens_seen": 138010140,
      "step": 6429,
      "time_per_iteration": 2.509223222732544
    },
    {
      "auxiliary_loss_clip": 0.01103519,
      "auxiliary_loss_mlp": 0.01032536,
      "balance_loss_clip": 1.04334426,
      "balance_loss_mlp": 1.01857638,
      "epoch": 0.38659251465504285,
      "flos": 18478949374080.0,
      "grad_norm": 1.8696135737964616,
      "language_loss": 0.80969548,
      "learning_rate": 2.807931078076015e-06,
      "loss": 0.83105612,
      "num_input_tokens_seen": 138028880,
      "step": 6430,
      "time_per_iteration": 2.526221513748169
    },
    {
      "auxiliary_loss_clip": 0.01020528,
      "auxiliary_loss_mlp": 0.01006412,
      "balance_loss_clip": 1.02100384,
      "balance_loss_mlp": 1.00470734,
      "epoch": 0.3866526379077108,
      "flos": 64165726978560.0,
      "grad_norm": 0.7272513900087025,
      "language_loss": 0.58815855,
      "learning_rate": 2.807574793260416e-06,
      "loss": 0.60842794,
      "num_input_tokens_seen": 138098090,
      "step": 6431,
      "time_per_iteration": 3.213054895401001
    },
    {
      "auxiliary_loss_clip": 0.01074426,
      "auxiliary_loss_mlp": 0.01036013,
      "balance_loss_clip": 1.03980923,
      "balance_loss_mlp": 1.02006245,
      "epoch": 0.3867127611603788,
      "flos": 14388292897920.0,
      "grad_norm": 1.8270629578559334,
      "language_loss": 0.79114032,
      "learning_rate": 2.8072184778215004e-06,
      "loss": 0.81224471,
      "num_input_tokens_seen": 138114735,
      "step": 6432,
      "time_per_iteration": 2.595115900039673
    },
    {
      "auxiliary_loss_clip": 0.01114785,
      "auxiliary_loss_mlp": 0.01044955,
      "balance_loss_clip": 1.04202831,
      "balance_loss_mlp": 1.02947009,
      "epoch": 0.38677288441304675,
      "flos": 20010754823040.0,
      "grad_norm": 2.149948489889016,
      "language_loss": 0.80768776,
      "learning_rate": 2.806862131772779e-06,
      "loss": 0.82928514,
      "num_input_tokens_seen": 138130480,
      "step": 6433,
      "time_per_iteration": 2.4791829586029053
    },
    {
      "auxiliary_loss_clip": 0.01100618,
      "auxiliary_loss_mlp": 0.01034886,
      "balance_loss_clip": 1.04633021,
      "balance_loss_mlp": 1.01981807,
      "epoch": 0.3868330076657147,
      "flos": 22236893147520.0,
      "grad_norm": 1.5905732435704893,
      "language_loss": 0.70676589,
      "learning_rate": 2.806505755127765e-06,
      "loss": 0.72812092,
      "num_input_tokens_seen": 138150640,
      "step": 6434,
      "time_per_iteration": 2.5497403144836426
    },
    {
      "auxiliary_loss_clip": 0.01089502,
      "auxiliary_loss_mlp": 0.0104579,
      "balance_loss_clip": 1.04444432,
      "balance_loss_mlp": 1.02862418,
      "epoch": 0.3868931309183827,
      "flos": 16727442387840.0,
      "grad_norm": 1.6536891542224736,
      "language_loss": 0.77389848,
      "learning_rate": 2.806149347899972e-06,
      "loss": 0.79525143,
      "num_input_tokens_seen": 138169700,
      "step": 6435,
      "time_per_iteration": 2.534238815307617
    },
    {
      "auxiliary_loss_clip": 0.01110931,
      "auxiliary_loss_mlp": 0.01032349,
      "balance_loss_clip": 1.04329062,
      "balance_loss_mlp": 1.01800823,
      "epoch": 0.38695325417105064,
      "flos": 22674716023680.0,
      "grad_norm": 1.7212729875705899,
      "language_loss": 0.7963196,
      "learning_rate": 2.805792910102915e-06,
      "loss": 0.81775248,
      "num_input_tokens_seen": 138185835,
      "step": 6436,
      "time_per_iteration": 2.5122249126434326
    },
    {
      "auxiliary_loss_clip": 0.01096522,
      "auxiliary_loss_mlp": 0.01031369,
      "balance_loss_clip": 1.04478908,
      "balance_loss_mlp": 1.01764178,
      "epoch": 0.3870133774237186,
      "flos": 23112036109440.0,
      "grad_norm": 1.6479397156938658,
      "language_loss": 0.76463914,
      "learning_rate": 2.8054364417501093e-06,
      "loss": 0.785918,
      "num_input_tokens_seen": 138204080,
      "step": 6437,
      "time_per_iteration": 2.5441629886627197
    },
    {
      "auxiliary_loss_clip": 0.01104473,
      "auxiliary_loss_mlp": 0.01038849,
      "balance_loss_clip": 1.04617572,
      "balance_loss_mlp": 1.02572989,
      "epoch": 0.3870735006763866,
      "flos": 17675699483520.0,
      "grad_norm": 2.319382720355043,
      "language_loss": 0.81308693,
      "learning_rate": 2.805079942855074e-06,
      "loss": 0.83452016,
      "num_input_tokens_seen": 138220710,
      "step": 6438,
      "time_per_iteration": 2.500417709350586
    },
    {
      "auxiliary_loss_clip": 0.01103677,
      "auxiliary_loss_mlp": 0.00825739,
      "balance_loss_clip": 1.04211783,
      "balance_loss_mlp": 1.07441044,
      "epoch": 0.38713362392905454,
      "flos": 23295791111040.0,
      "grad_norm": 1.3674720515988905,
      "language_loss": 0.75226122,
      "learning_rate": 2.804723413431326e-06,
      "loss": 0.77155542,
      "num_input_tokens_seen": 138241720,
      "step": 6439,
      "time_per_iteration": 2.528979539871216
    },
    {
      "auxiliary_loss_clip": 0.01122976,
      "auxiliary_loss_mlp": 0.01029407,
      "balance_loss_clip": 1.04589605,
      "balance_loss_mlp": 1.01575136,
      "epoch": 0.38719374718172256,
      "flos": 21031192298880.0,
      "grad_norm": 1.5085068929948482,
      "language_loss": 0.73732185,
      "learning_rate": 2.8043668534923855e-06,
      "loss": 0.75884569,
      "num_input_tokens_seen": 138261885,
      "step": 6440,
      "time_per_iteration": 2.481387138366699
    },
    {
      "auxiliary_loss_clip": 0.01116932,
      "auxiliary_loss_mlp": 0.01036042,
      "balance_loss_clip": 1.04523826,
      "balance_loss_mlp": 1.02127779,
      "epoch": 0.3872538704343905,
      "flos": 19609776322560.0,
      "grad_norm": 2.008735136208327,
      "language_loss": 0.82141852,
      "learning_rate": 2.804010263051774e-06,
      "loss": 0.84294826,
      "num_input_tokens_seen": 138280255,
      "step": 6441,
      "time_per_iteration": 2.470254898071289
    },
    {
      "auxiliary_loss_clip": 0.01127464,
      "auxiliary_loss_mlp": 0.01041465,
      "balance_loss_clip": 1.04701149,
      "balance_loss_mlp": 1.0272851,
      "epoch": 0.3873139936870585,
      "flos": 17530045833600.0,
      "grad_norm": 2.4995264736687712,
      "language_loss": 0.81375575,
      "learning_rate": 2.8036536421230118e-06,
      "loss": 0.83544505,
      "num_input_tokens_seen": 138296675,
      "step": 6442,
      "time_per_iteration": 2.4487905502319336
    },
    {
      "auxiliary_loss_clip": 0.0108562,
      "auxiliary_loss_mlp": 0.01035302,
      "balance_loss_clip": 1.04730308,
      "balance_loss_mlp": 1.0205555,
      "epoch": 0.38737411693972645,
      "flos": 17786555832960.0,
      "grad_norm": 1.6953770233927317,
      "language_loss": 0.83627522,
      "learning_rate": 2.803296990719624e-06,
      "loss": 0.85748446,
      "num_input_tokens_seen": 138314985,
      "step": 6443,
      "time_per_iteration": 2.5268115997314453
    },
    {
      "auxiliary_loss_clip": 0.01025715,
      "auxiliary_loss_mlp": 0.01002132,
      "balance_loss_clip": 1.01959956,
      "balance_loss_mlp": 1.00057018,
      "epoch": 0.3874342401923944,
      "flos": 58304637048960.0,
      "grad_norm": 0.764860946417262,
      "language_loss": 0.50276268,
      "learning_rate": 2.8029403088551327e-06,
      "loss": 0.52304113,
      "num_input_tokens_seen": 138373275,
      "step": 6444,
      "time_per_iteration": 3.137730836868286
    },
    {
      "auxiliary_loss_clip": 0.01081012,
      "auxiliary_loss_mlp": 0.00815765,
      "balance_loss_clip": 1.04089999,
      "balance_loss_mlp": 1.05555105,
      "epoch": 0.3874943634450624,
      "flos": 17711933328000.0,
      "grad_norm": 1.7533818152183194,
      "language_loss": 0.78514814,
      "learning_rate": 2.802583596543065e-06,
      "loss": 0.80411595,
      "num_input_tokens_seen": 138391145,
      "step": 6445,
      "time_per_iteration": 2.519932985305786
    },
    {
      "auxiliary_loss_clip": 0.01108652,
      "auxiliary_loss_mlp": 0.01036438,
      "balance_loss_clip": 1.04722869,
      "balance_loss_mlp": 1.02196622,
      "epoch": 0.38755448669773035,
      "flos": 19244852098560.0,
      "grad_norm": 2.1323036684837424,
      "language_loss": 0.81059605,
      "learning_rate": 2.8022268537969474e-06,
      "loss": 0.83204699,
      "num_input_tokens_seen": 138409875,
      "step": 6446,
      "time_per_iteration": 3.885542631149292
    },
    {
      "auxiliary_loss_clip": 0.01101309,
      "auxiliary_loss_mlp": 0.01037506,
      "balance_loss_clip": 1.04401898,
      "balance_loss_mlp": 1.02390981,
      "epoch": 0.3876146099503983,
      "flos": 20594267262720.0,
      "grad_norm": 1.7723757773128392,
      "language_loss": 0.77116525,
      "learning_rate": 2.801870080630306e-06,
      "loss": 0.79255342,
      "num_input_tokens_seen": 138428965,
      "step": 6447,
      "time_per_iteration": 2.538090705871582
    },
    {
      "auxiliary_loss_clip": 0.01102183,
      "auxiliary_loss_mlp": 0.01029247,
      "balance_loss_clip": 1.04315066,
      "balance_loss_mlp": 1.01612806,
      "epoch": 0.3876747332030663,
      "flos": 19281121856640.0,
      "grad_norm": 1.5046075134640446,
      "language_loss": 0.76064706,
      "learning_rate": 2.801513277056671e-06,
      "loss": 0.78196132,
      "num_input_tokens_seen": 138448090,
      "step": 6448,
      "time_per_iteration": 3.904160499572754
    },
    {
      "auxiliary_loss_clip": 0.01095362,
      "auxiliary_loss_mlp": 0.01035652,
      "balance_loss_clip": 1.0449307,
      "balance_loss_mlp": 1.02147222,
      "epoch": 0.38773485645573424,
      "flos": 18945895201920.0,
      "grad_norm": 1.903983117026687,
      "language_loss": 0.76179063,
      "learning_rate": 2.8011564430895725e-06,
      "loss": 0.78310078,
      "num_input_tokens_seen": 138466105,
      "step": 6449,
      "time_per_iteration": 3.9452857971191406
    },
    {
      "auxiliary_loss_clip": 0.0109104,
      "auxiliary_loss_mlp": 0.00800594,
      "balance_loss_clip": 1.04140019,
      "balance_loss_mlp": 1.02999365,
      "epoch": 0.3877949797084022,
      "flos": 23071348978560.0,
      "grad_norm": 1.7083311531223624,
      "language_loss": 0.78086996,
      "learning_rate": 2.800799578742542e-06,
      "loss": 0.79978633,
      "num_input_tokens_seen": 138485160,
      "step": 6450,
      "time_per_iteration": 2.5815467834472656
    },
    {
      "auxiliary_loss_clip": 0.01128808,
      "auxiliary_loss_mlp": 0.01035254,
      "balance_loss_clip": 1.04447317,
      "balance_loss_mlp": 1.02072239,
      "epoch": 0.3878551029610702,
      "flos": 29095543589760.0,
      "grad_norm": 3.0040861983745573,
      "language_loss": 0.78094685,
      "learning_rate": 2.8004426840291106e-06,
      "loss": 0.80258751,
      "num_input_tokens_seen": 138504135,
      "step": 6451,
      "time_per_iteration": 2.5011773109436035
    },
    {
      "auxiliary_loss_clip": 0.01120951,
      "auxiliary_loss_mlp": 0.01026163,
      "balance_loss_clip": 1.0443331,
      "balance_loss_mlp": 1.01289463,
      "epoch": 0.38791522621373814,
      "flos": 20996394998400.0,
      "grad_norm": 1.6754834816634068,
      "language_loss": 0.76596618,
      "learning_rate": 2.800085758962812e-06,
      "loss": 0.78743738,
      "num_input_tokens_seen": 138523955,
      "step": 6452,
      "time_per_iteration": 3.8513684272766113
    },
    {
      "auxiliary_loss_clip": 0.01097488,
      "auxiliary_loss_mlp": 0.01042866,
      "balance_loss_clip": 1.04607129,
      "balance_loss_mlp": 1.02930605,
      "epoch": 0.3879753494664061,
      "flos": 15486836497920.0,
      "grad_norm": 1.5418332183774686,
      "language_loss": 0.79779148,
      "learning_rate": 2.799728803557182e-06,
      "loss": 0.81919497,
      "num_input_tokens_seen": 138541655,
      "step": 6453,
      "time_per_iteration": 2.503101348876953
    },
    {
      "auxiliary_loss_clip": 0.01123483,
      "auxiliary_loss_mlp": 0.01036957,
      "balance_loss_clip": 1.04784894,
      "balance_loss_mlp": 1.02246666,
      "epoch": 0.3880354727190741,
      "flos": 22053964158720.0,
      "grad_norm": 1.8613909243975562,
      "language_loss": 0.71524799,
      "learning_rate": 2.7993718178257555e-06,
      "loss": 0.73685241,
      "num_input_tokens_seen": 138560860,
      "step": 6454,
      "time_per_iteration": 2.5138587951660156
    },
    {
      "auxiliary_loss_clip": 0.01128282,
      "auxiliary_loss_mlp": 0.01035994,
      "balance_loss_clip": 1.04551542,
      "balance_loss_mlp": 1.0214262,
      "epoch": 0.3880955959717421,
      "flos": 20340307128960.0,
      "grad_norm": 1.735057366790028,
      "language_loss": 0.77912402,
      "learning_rate": 2.7990148017820694e-06,
      "loss": 0.80076677,
      "num_input_tokens_seen": 138580200,
      "step": 6455,
      "time_per_iteration": 2.4472837448120117
    },
    {
      "auxiliary_loss_clip": 0.01124393,
      "auxiliary_loss_mlp": 0.01034312,
      "balance_loss_clip": 1.04622817,
      "balance_loss_mlp": 1.02020931,
      "epoch": 0.38815571922441006,
      "flos": 23075407215360.0,
      "grad_norm": 1.4889744514211738,
      "language_loss": 0.75513232,
      "learning_rate": 2.798657755439662e-06,
      "loss": 0.77671939,
      "num_input_tokens_seen": 138598315,
      "step": 6456,
      "time_per_iteration": 2.4791674613952637
    },
    {
      "auxiliary_loss_clip": 0.01056443,
      "auxiliary_loss_mlp": 0.01035683,
      "balance_loss_clip": 1.04675364,
      "balance_loss_mlp": 1.02109766,
      "epoch": 0.388215842477078,
      "flos": 20776944856320.0,
      "grad_norm": 2.213673571912731,
      "language_loss": 0.60642117,
      "learning_rate": 2.7983006788120726e-06,
      "loss": 0.6273424,
      "num_input_tokens_seen": 138615695,
      "step": 6457,
      "time_per_iteration": 2.609815835952759
    },
    {
      "auxiliary_loss_clip": 0.01127587,
      "auxiliary_loss_mlp": 0.01036381,
      "balance_loss_clip": 1.04558468,
      "balance_loss_mlp": 1.0210743,
      "epoch": 0.388275965729746,
      "flos": 20448182649600.0,
      "grad_norm": 2.1542281506287564,
      "language_loss": 0.80208737,
      "learning_rate": 2.797943571912841e-06,
      "loss": 0.82372701,
      "num_input_tokens_seen": 138633180,
      "step": 6458,
      "time_per_iteration": 2.4734554290771484
    },
    {
      "auxiliary_loss_clip": 0.01077192,
      "auxiliary_loss_mlp": 0.01040871,
      "balance_loss_clip": 1.04418659,
      "balance_loss_mlp": 1.02476537,
      "epoch": 0.38833608898241395,
      "flos": 27892392606720.0,
      "grad_norm": 1.7597646954227897,
      "language_loss": 0.81342506,
      "learning_rate": 2.797586434755509e-06,
      "loss": 0.83460569,
      "num_input_tokens_seen": 138654785,
      "step": 6459,
      "time_per_iteration": 2.635648727416992
    },
    {
      "auxiliary_loss_clip": 0.01100715,
      "auxiliary_loss_mlp": 0.01035609,
      "balance_loss_clip": 1.04548264,
      "balance_loss_mlp": 1.02216172,
      "epoch": 0.3883962122350819,
      "flos": 18076390675200.0,
      "grad_norm": 2.160878612451965,
      "language_loss": 0.61459649,
      "learning_rate": 2.7972292673536202e-06,
      "loss": 0.63595963,
      "num_input_tokens_seen": 138673330,
      "step": 6460,
      "time_per_iteration": 2.5126261711120605
    },
    {
      "auxiliary_loss_clip": 0.01112293,
      "auxiliary_loss_mlp": 0.01029407,
      "balance_loss_clip": 1.04562688,
      "balance_loss_mlp": 1.01717639,
      "epoch": 0.3884563354877499,
      "flos": 23622254847360.0,
      "grad_norm": 1.7094844631207189,
      "language_loss": 0.85987031,
      "learning_rate": 2.796872069720717e-06,
      "loss": 0.88128734,
      "num_input_tokens_seen": 138694185,
      "step": 6461,
      "time_per_iteration": 2.5585741996765137
    },
    {
      "auxiliary_loss_clip": 0.01111934,
      "auxiliary_loss_mlp": 0.0103606,
      "balance_loss_clip": 1.04636383,
      "balance_loss_mlp": 1.02229702,
      "epoch": 0.38851645874041785,
      "flos": 27453528236160.0,
      "grad_norm": 3.1576709434995545,
      "language_loss": 0.71633315,
      "learning_rate": 2.7965148418703456e-06,
      "loss": 0.737813,
      "num_input_tokens_seen": 138714625,
      "step": 6462,
      "time_per_iteration": 2.546323776245117
    },
    {
      "auxiliary_loss_clip": 0.01082003,
      "auxiliary_loss_mlp": 0.01039119,
      "balance_loss_clip": 1.04122937,
      "balance_loss_mlp": 1.02346683,
      "epoch": 0.3885765819930858,
      "flos": 25228072270080.0,
      "grad_norm": 2.3346676739206433,
      "language_loss": 0.76121771,
      "learning_rate": 2.796157583816052e-06,
      "loss": 0.78242898,
      "num_input_tokens_seen": 138733585,
      "step": 6463,
      "time_per_iteration": 2.6059985160827637
    },
    {
      "auxiliary_loss_clip": 0.01095339,
      "auxiliary_loss_mlp": 0.01040928,
      "balance_loss_clip": 1.04681075,
      "balance_loss_mlp": 1.02487063,
      "epoch": 0.3886367052457538,
      "flos": 16946605221120.0,
      "grad_norm": 1.9682475406331104,
      "language_loss": 0.7051025,
      "learning_rate": 2.795800295571382e-06,
      "loss": 0.72646517,
      "num_input_tokens_seen": 138752335,
      "step": 6464,
      "time_per_iteration": 2.5421345233917236
    },
    {
      "auxiliary_loss_clip": 0.01101926,
      "auxiliary_loss_mlp": 0.01031564,
      "balance_loss_clip": 1.04567432,
      "balance_loss_mlp": 1.01695502,
      "epoch": 0.38869682849842174,
      "flos": 27154140376320.0,
      "grad_norm": 1.9679199729978532,
      "language_loss": 0.69853705,
      "learning_rate": 2.7954429771498858e-06,
      "loss": 0.719872,
      "num_input_tokens_seen": 138768450,
      "step": 6465,
      "time_per_iteration": 2.568559408187866
    },
    {
      "auxiliary_loss_clip": 0.01092239,
      "auxiliary_loss_mlp": 0.01040036,
      "balance_loss_clip": 1.04632998,
      "balance_loss_mlp": 1.02490854,
      "epoch": 0.3887569517510897,
      "flos": 21063619301760.0,
      "grad_norm": 2.332718092490034,
      "language_loss": 0.77926546,
      "learning_rate": 2.7950856285651117e-06,
      "loss": 0.80058819,
      "num_input_tokens_seen": 138786775,
      "step": 6466,
      "time_per_iteration": 2.560119152069092
    },
    {
      "auxiliary_loss_clip": 0.01089494,
      "auxiliary_loss_mlp": 0.0103751,
      "balance_loss_clip": 1.04366446,
      "balance_loss_mlp": 1.02291858,
      "epoch": 0.38881707500375773,
      "flos": 29497384016640.0,
      "grad_norm": 1.4453514793539861,
      "language_loss": 0.6945613,
      "learning_rate": 2.794728249830611e-06,
      "loss": 0.71583128,
      "num_input_tokens_seen": 138810100,
      "step": 6467,
      "time_per_iteration": 2.630826711654663
    },
    {
      "auxiliary_loss_clip": 0.01090336,
      "auxiliary_loss_mlp": 0.01041421,
      "balance_loss_clip": 1.04450011,
      "balance_loss_mlp": 1.0263114,
      "epoch": 0.3888771982564257,
      "flos": 17488281294720.0,
      "grad_norm": 2.5547947742705253,
      "language_loss": 0.83851385,
      "learning_rate": 2.794370840959936e-06,
      "loss": 0.85983139,
      "num_input_tokens_seen": 138825140,
      "step": 6468,
      "time_per_iteration": 2.523120164871216
    },
    {
      "auxiliary_loss_clip": 0.01108043,
      "auxiliary_loss_mlp": 0.01033603,
      "balance_loss_clip": 1.04663968,
      "balance_loss_mlp": 1.02063346,
      "epoch": 0.38893732150909366,
      "flos": 21942425450880.0,
      "grad_norm": 1.774128131425595,
      "language_loss": 0.8424986,
      "learning_rate": 2.7940134019666383e-06,
      "loss": 0.86391509,
      "num_input_tokens_seen": 138844115,
      "step": 6469,
      "time_per_iteration": 2.6161844730377197
    },
    {
      "auxiliary_loss_clip": 0.01087838,
      "auxiliary_loss_mlp": 0.01040393,
      "balance_loss_clip": 1.04486704,
      "balance_loss_mlp": 1.02450871,
      "epoch": 0.3889974447617616,
      "flos": 24276367468800.0,
      "grad_norm": 1.9106429427235614,
      "language_loss": 0.74792325,
      "learning_rate": 2.793655932864273e-06,
      "loss": 0.76920557,
      "num_input_tokens_seen": 138860860,
      "step": 6470,
      "time_per_iteration": 2.5828757286071777
    },
    {
      "auxiliary_loss_clip": 0.01088232,
      "auxiliary_loss_mlp": 0.00804018,
      "balance_loss_clip": 1.04550672,
      "balance_loss_mlp": 1.0370028,
      "epoch": 0.3890575680144296,
      "flos": 25667116208640.0,
      "grad_norm": 1.6761743616331546,
      "language_loss": 0.74701369,
      "learning_rate": 2.7932984336663953e-06,
      "loss": 0.7659362,
      "num_input_tokens_seen": 138881910,
      "step": 6471,
      "time_per_iteration": 2.634056806564331
    },
    {
      "auxiliary_loss_clip": 0.01078258,
      "auxiliary_loss_mlp": 0.01042269,
      "balance_loss_clip": 1.04491842,
      "balance_loss_mlp": 1.02689099,
      "epoch": 0.38911769126709755,
      "flos": 22855274714880.0,
      "grad_norm": 1.6695312146650534,
      "language_loss": 0.67775333,
      "learning_rate": 2.792940904386562e-06,
      "loss": 0.69895864,
      "num_input_tokens_seen": 138900975,
      "step": 6472,
      "time_per_iteration": 2.594273090362549
    },
    {
      "auxiliary_loss_clip": 0.01103368,
      "auxiliary_loss_mlp": 0.01040574,
      "balance_loss_clip": 1.05138922,
      "balance_loss_mlp": 1.02698433,
      "epoch": 0.3891778145197655,
      "flos": 25447522412160.0,
      "grad_norm": 1.747807806089283,
      "language_loss": 0.7630446,
      "learning_rate": 2.7925833450383293e-06,
      "loss": 0.78448403,
      "num_input_tokens_seen": 138920795,
      "step": 6473,
      "time_per_iteration": 2.615492343902588
    },
    {
      "auxiliary_loss_clip": 0.01109847,
      "auxiliary_loss_mlp": 0.01039906,
      "balance_loss_clip": 1.05172789,
      "balance_loss_mlp": 1.02495754,
      "epoch": 0.3892379377724335,
      "flos": 14027965614720.0,
      "grad_norm": 2.426244317201221,
      "language_loss": 0.70520478,
      "learning_rate": 2.792225755635257e-06,
      "loss": 0.72670233,
      "num_input_tokens_seen": 138938770,
      "step": 6474,
      "time_per_iteration": 2.514801025390625
    },
    {
      "auxiliary_loss_clip": 0.01128446,
      "auxiliary_loss_mlp": 0.01038883,
      "balance_loss_clip": 1.04740047,
      "balance_loss_mlp": 1.0251137,
      "epoch": 0.38929806102510145,
      "flos": 20157449967360.0,
      "grad_norm": 1.5106461438317134,
      "language_loss": 0.68851876,
      "learning_rate": 2.7918681361909046e-06,
      "loss": 0.71019208,
      "num_input_tokens_seen": 138958880,
      "step": 6475,
      "time_per_iteration": 2.4920084476470947
    },
    {
      "auxiliary_loss_clip": 0.01107094,
      "auxiliary_loss_mlp": 0.01052668,
      "balance_loss_clip": 1.04650784,
      "balance_loss_mlp": 1.03551376,
      "epoch": 0.3893581842777694,
      "flos": 22163958581760.0,
      "grad_norm": 1.8907976078309567,
      "language_loss": 0.75904459,
      "learning_rate": 2.7915104867188332e-06,
      "loss": 0.78064221,
      "num_input_tokens_seen": 138977240,
      "step": 6476,
      "time_per_iteration": 2.5170702934265137
    },
    {
      "auxiliary_loss_clip": 0.01043044,
      "auxiliary_loss_mlp": 0.01003089,
      "balance_loss_clip": 1.0333693,
      "balance_loss_mlp": 1.00152743,
      "epoch": 0.3894183075304374,
      "flos": 67301877392640.0,
      "grad_norm": 0.7722506237719168,
      "language_loss": 0.58183157,
      "learning_rate": 2.7911528072326055e-06,
      "loss": 0.6022929,
      "num_input_tokens_seen": 139039035,
      "step": 6477,
      "time_per_iteration": 3.160775661468506
    },
    {
      "auxiliary_loss_clip": 0.01089329,
      "auxiliary_loss_mlp": 0.01034254,
      "balance_loss_clip": 1.04946077,
      "balance_loss_mlp": 1.0181129,
      "epoch": 0.38947843078310534,
      "flos": 18547502480640.0,
      "grad_norm": 1.752639916581999,
      "language_loss": 0.78310955,
      "learning_rate": 2.7907950977457832e-06,
      "loss": 0.80434537,
      "num_input_tokens_seen": 139055560,
      "step": 6478,
      "time_per_iteration": 2.5590474605560303
    },
    {
      "auxiliary_loss_clip": 0.01113235,
      "auxiliary_loss_mlp": 0.01033769,
      "balance_loss_clip": 1.04565907,
      "balance_loss_mlp": 1.01938057,
      "epoch": 0.3895385540357733,
      "flos": 14605875532800.0,
      "grad_norm": 2.302267351062905,
      "language_loss": 0.82982665,
      "learning_rate": 2.7904373582719317e-06,
      "loss": 0.85129672,
      "num_input_tokens_seen": 139071865,
      "step": 6479,
      "time_per_iteration": 2.5025928020477295
    },
    {
      "auxiliary_loss_clip": 0.01129081,
      "auxiliary_loss_mlp": 0.01035035,
      "balance_loss_clip": 1.04950202,
      "balance_loss_mlp": 1.02003217,
      "epoch": 0.38959867728844133,
      "flos": 19975203336960.0,
      "grad_norm": 1.8531599597910755,
      "language_loss": 0.80110466,
      "learning_rate": 2.790079588824617e-06,
      "loss": 0.82274586,
      "num_input_tokens_seen": 139089640,
      "step": 6480,
      "time_per_iteration": 2.4675283432006836
    },
    {
      "auxiliary_loss_clip": 0.01101133,
      "auxiliary_loss_mlp": 0.01032793,
      "balance_loss_clip": 1.0458858,
      "balance_loss_mlp": 1.01864314,
      "epoch": 0.3896588005411093,
      "flos": 22672130244480.0,
      "grad_norm": 1.695870955779693,
      "language_loss": 0.82991642,
      "learning_rate": 2.7897217894174038e-06,
      "loss": 0.85125566,
      "num_input_tokens_seen": 139109365,
      "step": 6481,
      "time_per_iteration": 2.549489974975586
    },
    {
      "auxiliary_loss_clip": 0.01100599,
      "auxiliary_loss_mlp": 0.01034277,
      "balance_loss_clip": 1.04958677,
      "balance_loss_mlp": 1.02066362,
      "epoch": 0.38971892379377726,
      "flos": 20996035862400.0,
      "grad_norm": 1.580801308547055,
      "language_loss": 0.75310647,
      "learning_rate": 2.789363960063863e-06,
      "loss": 0.77445531,
      "num_input_tokens_seen": 139128260,
      "step": 6482,
      "time_per_iteration": 2.5298352241516113
    },
    {
      "auxiliary_loss_clip": 0.01098993,
      "auxiliary_loss_mlp": 0.01035796,
      "balance_loss_clip": 1.04976058,
      "balance_loss_mlp": 1.0219022,
      "epoch": 0.3897790470464452,
      "flos": 22528487756160.0,
      "grad_norm": 1.8238139297719058,
      "language_loss": 0.79219306,
      "learning_rate": 2.78900610077756e-06,
      "loss": 0.81354094,
      "num_input_tokens_seen": 139147315,
      "step": 6483,
      "time_per_iteration": 2.596503973007202
    },
    {
      "auxiliary_loss_clip": 0.0111604,
      "auxiliary_loss_mlp": 0.01030289,
      "balance_loss_clip": 1.04799199,
      "balance_loss_mlp": 1.01479197,
      "epoch": 0.3898391702991132,
      "flos": 26209905603840.0,
      "grad_norm": 1.4533585735454098,
      "language_loss": 0.80041236,
      "learning_rate": 2.788648211572067e-06,
      "loss": 0.82187569,
      "num_input_tokens_seen": 139167270,
      "step": 6484,
      "time_per_iteration": 2.5955512523651123
    },
    {
      "auxiliary_loss_clip": 0.01112282,
      "auxiliary_loss_mlp": 0.01057606,
      "balance_loss_clip": 1.04828477,
      "balance_loss_mlp": 1.04079747,
      "epoch": 0.38989929355178116,
      "flos": 21065558636160.0,
      "grad_norm": 1.6103396017506837,
      "language_loss": 0.77548611,
      "learning_rate": 2.7882902924609557e-06,
      "loss": 0.79718494,
      "num_input_tokens_seen": 139185970,
      "step": 6485,
      "time_per_iteration": 3.8658671379089355
    },
    {
      "auxiliary_loss_clip": 0.01086311,
      "auxiliary_loss_mlp": 0.01038872,
      "balance_loss_clip": 1.0478512,
      "balance_loss_mlp": 1.02360737,
      "epoch": 0.3899594168044491,
      "flos": 25484115392640.0,
      "grad_norm": 2.2792556806055546,
      "language_loss": 0.84804285,
      "learning_rate": 2.7879323434577965e-06,
      "loss": 0.86929464,
      "num_input_tokens_seen": 139203730,
      "step": 6486,
      "time_per_iteration": 2.63047194480896
    },
    {
      "auxiliary_loss_clip": 0.01113583,
      "auxiliary_loss_mlp": 0.0103546,
      "balance_loss_clip": 1.04821718,
      "balance_loss_mlp": 1.02112556,
      "epoch": 0.3900195400571171,
      "flos": 31139363456640.0,
      "grad_norm": 2.43840120546838,
      "language_loss": 0.85338712,
      "learning_rate": 2.7875743645761645e-06,
      "loss": 0.87487757,
      "num_input_tokens_seen": 139222560,
      "step": 6487,
      "time_per_iteration": 3.996201753616333
    },
    {
      "auxiliary_loss_clip": 0.01099628,
      "auxiliary_loss_mlp": 0.01034559,
      "balance_loss_clip": 1.04555285,
      "balance_loss_mlp": 1.01982474,
      "epoch": 0.39007966330978505,
      "flos": 20229917656320.0,
      "grad_norm": 1.5363464239345481,
      "language_loss": 0.72726345,
      "learning_rate": 2.787216355829633e-06,
      "loss": 0.74860531,
      "num_input_tokens_seen": 139242165,
      "step": 6488,
      "time_per_iteration": 3.9912161827087402
    },
    {
      "auxiliary_loss_clip": 0.01095005,
      "auxiliary_loss_mlp": 0.01048057,
      "balance_loss_clip": 1.05022418,
      "balance_loss_mlp": 1.0308907,
      "epoch": 0.390139786562453,
      "flos": 22528739151360.0,
      "grad_norm": 1.946188694621917,
      "language_loss": 0.688541,
      "learning_rate": 2.786858317231779e-06,
      "loss": 0.70997167,
      "num_input_tokens_seen": 139262525,
      "step": 6489,
      "time_per_iteration": 2.5928304195404053
    },
    {
      "auxiliary_loss_clip": 0.01103726,
      "auxiliary_loss_mlp": 0.01038444,
      "balance_loss_clip": 1.04915094,
      "balance_loss_mlp": 1.02438283,
      "epoch": 0.390199909815121,
      "flos": 26432911192320.0,
      "grad_norm": 1.488548938234222,
      "language_loss": 0.80586272,
      "learning_rate": 2.7865002487961788e-06,
      "loss": 0.82728446,
      "num_input_tokens_seen": 139282835,
      "step": 6490,
      "time_per_iteration": 2.5794286727905273
    },
    {
      "auxiliary_loss_clip": 0.01120222,
      "auxiliary_loss_mlp": 0.01037825,
      "balance_loss_clip": 1.04919124,
      "balance_loss_mlp": 1.02309656,
      "epoch": 0.39026003306778895,
      "flos": 17274577328640.0,
      "grad_norm": 1.9041890391731375,
      "language_loss": 0.89497292,
      "learning_rate": 2.7861421505364104e-06,
      "loss": 0.91655338,
      "num_input_tokens_seen": 139299490,
      "step": 6491,
      "time_per_iteration": 3.843534231185913
    },
    {
      "auxiliary_loss_clip": 0.01086671,
      "auxiliary_loss_mlp": 0.01044441,
      "balance_loss_clip": 1.04470718,
      "balance_loss_mlp": 1.02898002,
      "epoch": 0.3903201563204569,
      "flos": 24532841554560.0,
      "grad_norm": 1.7388168573376321,
      "language_loss": 0.78545451,
      "learning_rate": 2.7857840224660523e-06,
      "loss": 0.80676562,
      "num_input_tokens_seen": 139317865,
      "step": 6492,
      "time_per_iteration": 2.593726396560669
    },
    {
      "auxiliary_loss_clip": 0.01104122,
      "auxiliary_loss_mlp": 0.01039043,
      "balance_loss_clip": 1.0449723,
      "balance_loss_mlp": 1.02465487,
      "epoch": 0.39038027957312493,
      "flos": 23767944410880.0,
      "grad_norm": 1.771331957781986,
      "language_loss": 0.73977494,
      "learning_rate": 2.7854258645986857e-06,
      "loss": 0.76120663,
      "num_input_tokens_seen": 139339840,
      "step": 6493,
      "time_per_iteration": 2.5584890842437744
    },
    {
      "auxiliary_loss_clip": 0.01095073,
      "auxiliary_loss_mlp": 0.01038093,
      "balance_loss_clip": 1.04891884,
      "balance_loss_mlp": 1.02340651,
      "epoch": 0.3904404028257929,
      "flos": 14100612871680.0,
      "grad_norm": 2.2268302183507402,
      "language_loss": 0.76042521,
      "learning_rate": 2.7850676769478916e-06,
      "loss": 0.78175688,
      "num_input_tokens_seen": 139357555,
      "step": 6494,
      "time_per_iteration": 2.6278977394104004
    },
    {
      "auxiliary_loss_clip": 0.01124257,
      "auxiliary_loss_mlp": 0.01044904,
      "balance_loss_clip": 1.04912281,
      "balance_loss_mlp": 1.029109,
      "epoch": 0.39050052607846086,
      "flos": 16910048154240.0,
      "grad_norm": 1.896942616763331,
      "language_loss": 0.74498981,
      "learning_rate": 2.7847094595272525e-06,
      "loss": 0.76668143,
      "num_input_tokens_seen": 139374455,
      "step": 6495,
      "time_per_iteration": 2.4797170162200928
    },
    {
      "auxiliary_loss_clip": 0.01132298,
      "auxiliary_loss_mlp": 0.01038809,
      "balance_loss_clip": 1.05101883,
      "balance_loss_mlp": 1.02319217,
      "epoch": 0.39056064933112883,
      "flos": 25915761129600.0,
      "grad_norm": 1.6358953334433695,
      "language_loss": 0.68050885,
      "learning_rate": 2.784351212350352e-06,
      "loss": 0.70221996,
      "num_input_tokens_seen": 139394770,
      "step": 6496,
      "time_per_iteration": 2.527200698852539
    },
    {
      "auxiliary_loss_clip": 0.01033313,
      "auxiliary_loss_mlp": 0.01009611,
      "balance_loss_clip": 1.03580642,
      "balance_loss_mlp": 1.0081389,
      "epoch": 0.3906207725837968,
      "flos": 60028421713920.0,
      "grad_norm": 0.6619433058669779,
      "language_loss": 0.53951764,
      "learning_rate": 2.783992935430775e-06,
      "loss": 0.55994689,
      "num_input_tokens_seen": 139454760,
      "step": 6497,
      "time_per_iteration": 3.2710530757904053
    },
    {
      "auxiliary_loss_clip": 0.01094154,
      "auxiliary_loss_mlp": 0.00790177,
      "balance_loss_clip": 1.05266786,
      "balance_loss_mlp": 1.01068962,
      "epoch": 0.39068089583646476,
      "flos": 21068683119360.0,
      "grad_norm": 2.196080702422757,
      "language_loss": 0.69515872,
      "learning_rate": 2.7836346287821068e-06,
      "loss": 0.71400201,
      "num_input_tokens_seen": 139472645,
      "step": 6498,
      "time_per_iteration": 2.5727479457855225
    },
    {
      "auxiliary_loss_clip": 0.01029207,
      "auxiliary_loss_mlp": 0.0100566,
      "balance_loss_clip": 1.02966237,
      "balance_loss_mlp": 1.00408649,
      "epoch": 0.3907410190891327,
      "flos": 70445677403520.0,
      "grad_norm": 0.7579735807900574,
      "language_loss": 0.51742053,
      "learning_rate": 2.783276292417936e-06,
      "loss": 0.5377692,
      "num_input_tokens_seen": 139536730,
      "step": 6499,
      "time_per_iteration": 3.239419460296631
    },
    {
      "auxiliary_loss_clip": 0.01119521,
      "auxiliary_loss_mlp": 0.01041675,
      "balance_loss_clip": 1.04771924,
      "balance_loss_mlp": 1.02535498,
      "epoch": 0.3908011423418007,
      "flos": 27962454084480.0,
      "grad_norm": 1.591167839935375,
      "language_loss": 0.73884344,
      "learning_rate": 2.7829179263518487e-06,
      "loss": 0.76045543,
      "num_input_tokens_seen": 139557540,
      "step": 6500,
      "time_per_iteration": 2.5714402198791504
    },
    {
      "auxiliary_loss_clip": 0.01121495,
      "auxiliary_loss_mlp": 0.01037048,
      "balance_loss_clip": 1.0518719,
      "balance_loss_mlp": 1.02271867,
      "epoch": 0.39086126559446865,
      "flos": 24462097718400.0,
      "grad_norm": 1.7585663925589514,
      "language_loss": 0.6875813,
      "learning_rate": 2.7825595305974354e-06,
      "loss": 0.70916677,
      "num_input_tokens_seen": 139576875,
      "step": 6501,
      "time_per_iteration": 2.5614917278289795
    },
    {
      "auxiliary_loss_clip": 0.01116175,
      "auxiliary_loss_mlp": 0.01034574,
      "balance_loss_clip": 1.04782939,
      "balance_loss_mlp": 1.02102637,
      "epoch": 0.3909213888471366,
      "flos": 16941541403520.0,
      "grad_norm": 1.8481110564443044,
      "language_loss": 0.79107046,
      "learning_rate": 2.782201105168287e-06,
      "loss": 0.81257802,
      "num_input_tokens_seen": 139594295,
      "step": 6502,
      "time_per_iteration": 2.4810447692871094
    },
    {
      "auxiliary_loss_clip": 0.0110636,
      "auxiliary_loss_mlp": 0.0103626,
      "balance_loss_clip": 1.04994822,
      "balance_loss_mlp": 1.02295589,
      "epoch": 0.3909815120998046,
      "flos": 29278400751360.0,
      "grad_norm": 2.2674713555473294,
      "language_loss": 0.80372298,
      "learning_rate": 2.7818426500779932e-06,
      "loss": 0.82514918,
      "num_input_tokens_seen": 139614080,
      "step": 6503,
      "time_per_iteration": 2.615665912628174
    },
    {
      "auxiliary_loss_clip": 0.01101036,
      "auxiliary_loss_mlp": 0.01031118,
      "balance_loss_clip": 1.04595375,
      "balance_loss_mlp": 1.01796341,
      "epoch": 0.39104163535247255,
      "flos": 18951246328320.0,
      "grad_norm": 1.8548129653679282,
      "language_loss": 0.7147249,
      "learning_rate": 2.7814841653401485e-06,
      "loss": 0.73604643,
      "num_input_tokens_seen": 139632755,
      "step": 6504,
      "time_per_iteration": 2.5145716667175293
    },
    {
      "auxiliary_loss_clip": 0.01125336,
      "auxiliary_loss_mlp": 0.01036441,
      "balance_loss_clip": 1.04541564,
      "balance_loss_mlp": 1.02180207,
      "epoch": 0.3911017586051405,
      "flos": 26323347732480.0,
      "grad_norm": 1.4214077966931933,
      "language_loss": 0.83050662,
      "learning_rate": 2.7811256509683454e-06,
      "loss": 0.85212439,
      "num_input_tokens_seen": 139654205,
      "step": 6505,
      "time_per_iteration": 2.5910468101501465
    },
    {
      "auxiliary_loss_clip": 0.01125664,
      "auxiliary_loss_mlp": 0.01037472,
      "balance_loss_clip": 1.04749238,
      "balance_loss_mlp": 1.02264261,
      "epoch": 0.3911618818578085,
      "flos": 21835770992640.0,
      "grad_norm": 1.8133913656122165,
      "language_loss": 0.70827377,
      "learning_rate": 2.7807671069761797e-06,
      "loss": 0.72990513,
      "num_input_tokens_seen": 139673595,
      "step": 6506,
      "time_per_iteration": 2.4644665718078613
    },
    {
      "auxiliary_loss_clip": 0.01102396,
      "auxiliary_loss_mlp": 0.01040558,
      "balance_loss_clip": 1.04653418,
      "balance_loss_mlp": 1.02668214,
      "epoch": 0.3912220051104765,
      "flos": 16359680989440.0,
      "grad_norm": 1.8994859656300183,
      "language_loss": 0.74931782,
      "learning_rate": 2.7804085333772477e-06,
      "loss": 0.77074736,
      "num_input_tokens_seen": 139690565,
      "step": 6507,
      "time_per_iteration": 2.524256706237793
    },
    {
      "auxiliary_loss_clip": 0.01050121,
      "auxiliary_loss_mlp": 0.01000243,
      "balance_loss_clip": 1.02134609,
      "balance_loss_mlp": 0.99870533,
      "epoch": 0.39128212836314447,
      "flos": 71050986420480.0,
      "grad_norm": 0.7678655831466287,
      "language_loss": 0.5653314,
      "learning_rate": 2.7800499301851446e-06,
      "loss": 0.58583504,
      "num_input_tokens_seen": 139749420,
      "step": 6508,
      "time_per_iteration": 3.2299723625183105
    },
    {
      "auxiliary_loss_clip": 0.01113718,
      "auxiliary_loss_mlp": 0.0104039,
      "balance_loss_clip": 1.04617667,
      "balance_loss_mlp": 1.02669895,
      "epoch": 0.39134225161581243,
      "flos": 20331975173760.0,
      "grad_norm": 1.8487037954179102,
      "language_loss": 0.76441288,
      "learning_rate": 2.779691297413471e-06,
      "loss": 0.785954,
      "num_input_tokens_seen": 139766265,
      "step": 6509,
      "time_per_iteration": 2.5258700847625732
    },
    {
      "auxiliary_loss_clip": 0.01094411,
      "auxiliary_loss_mlp": 0.01040203,
      "balance_loss_clip": 1.04143596,
      "balance_loss_mlp": 1.02389479,
      "epoch": 0.3914023748684804,
      "flos": 17018390551680.0,
      "grad_norm": 2.449716144988444,
      "language_loss": 0.83010173,
      "learning_rate": 2.779332635075825e-06,
      "loss": 0.85144788,
      "num_input_tokens_seen": 139782400,
      "step": 6510,
      "time_per_iteration": 2.5130233764648438
    },
    {
      "auxiliary_loss_clip": 0.01116321,
      "auxiliary_loss_mlp": 0.01036231,
      "balance_loss_clip": 1.04559207,
      "balance_loss_mlp": 1.02180684,
      "epoch": 0.39146249812114836,
      "flos": 18405224709120.0,
      "grad_norm": 1.712606812413767,
      "language_loss": 0.76766217,
      "learning_rate": 2.7789739431858073e-06,
      "loss": 0.78918767,
      "num_input_tokens_seen": 139801435,
      "step": 6511,
      "time_per_iteration": 2.5103158950805664
    },
    {
      "auxiliary_loss_clip": 0.01032997,
      "auxiliary_loss_mlp": 0.01000938,
      "balance_loss_clip": 1.0264858,
      "balance_loss_mlp": 0.99928111,
      "epoch": 0.3915226213738163,
      "flos": 67637355442560.0,
      "grad_norm": 0.7179945269128605,
      "language_loss": 0.57777214,
      "learning_rate": 2.7786152217570196e-06,
      "loss": 0.59811151,
      "num_input_tokens_seen": 139869700,
      "step": 6512,
      "time_per_iteration": 3.1969082355499268
    },
    {
      "auxiliary_loss_clip": 0.01127783,
      "auxiliary_loss_mlp": 0.01033822,
      "balance_loss_clip": 1.04636121,
      "balance_loss_mlp": 1.01828873,
      "epoch": 0.3915827446264843,
      "flos": 26359330181760.0,
      "grad_norm": 1.5778886847800162,
      "language_loss": 0.69012403,
      "learning_rate": 2.7782564708030647e-06,
      "loss": 0.71174008,
      "num_input_tokens_seen": 139890140,
      "step": 6513,
      "time_per_iteration": 2.5405921936035156
    },
    {
      "auxiliary_loss_clip": 0.01088969,
      "auxiliary_loss_mlp": 0.01041526,
      "balance_loss_clip": 1.04830778,
      "balance_loss_mlp": 1.02635098,
      "epoch": 0.39164286787915226,
      "flos": 21943897908480.0,
      "grad_norm": 2.622752036982753,
      "language_loss": 0.75833333,
      "learning_rate": 2.7778976903375464e-06,
      "loss": 0.77963829,
      "num_input_tokens_seen": 139908020,
      "step": 6514,
      "time_per_iteration": 2.5728330612182617
    },
    {
      "auxiliary_loss_clip": 0.01097057,
      "auxiliary_loss_mlp": 0.01041887,
      "balance_loss_clip": 1.04848766,
      "balance_loss_mlp": 1.02823758,
      "epoch": 0.3917029911318202,
      "flos": 16399829416320.0,
      "grad_norm": 1.9146717257782937,
      "language_loss": 0.77287781,
      "learning_rate": 2.7775388803740693e-06,
      "loss": 0.7942673,
      "num_input_tokens_seen": 139926180,
      "step": 6515,
      "time_per_iteration": 2.565471649169922
    },
    {
      "auxiliary_loss_clip": 0.01082277,
      "auxiliary_loss_mlp": 0.01047044,
      "balance_loss_clip": 1.03938937,
      "balance_loss_mlp": 1.03272653,
      "epoch": 0.3917631143844882,
      "flos": 26211701283840.0,
      "grad_norm": 1.3357510502584609,
      "language_loss": 0.79996264,
      "learning_rate": 2.7771800409262406e-06,
      "loss": 0.8212558,
      "num_input_tokens_seen": 139947420,
      "step": 6516,
      "time_per_iteration": 2.5914804935455322
    },
    {
      "auxiliary_loss_clip": 0.01089147,
      "auxiliary_loss_mlp": 0.01035213,
      "balance_loss_clip": 1.04702687,
      "balance_loss_mlp": 1.02053857,
      "epoch": 0.39182323763715615,
      "flos": 18548364407040.0,
      "grad_norm": 1.8963646474710738,
      "language_loss": 0.69793546,
      "learning_rate": 2.7768211720076665e-06,
      "loss": 0.71917909,
      "num_input_tokens_seen": 139965800,
      "step": 6517,
      "time_per_iteration": 2.594360828399658
    },
    {
      "auxiliary_loss_clip": 0.01083654,
      "auxiliary_loss_mlp": 0.01040988,
      "balance_loss_clip": 1.04360485,
      "balance_loss_mlp": 1.02627802,
      "epoch": 0.3918833608898241,
      "flos": 34313543395200.0,
      "grad_norm": 1.6753374280600564,
      "language_loss": 0.72310746,
      "learning_rate": 2.776462273631956e-06,
      "loss": 0.74435389,
      "num_input_tokens_seen": 139988140,
      "step": 6518,
      "time_per_iteration": 2.665189743041992
    },
    {
      "auxiliary_loss_clip": 0.01112441,
      "auxiliary_loss_mlp": 0.01036934,
      "balance_loss_clip": 1.04684412,
      "balance_loss_mlp": 1.02210426,
      "epoch": 0.3919434841424921,
      "flos": 36939582812160.0,
      "grad_norm": 2.7049344028561113,
      "language_loss": 0.61599064,
      "learning_rate": 2.7761033458127177e-06,
      "loss": 0.63748443,
      "num_input_tokens_seen": 140010060,
      "step": 6519,
      "time_per_iteration": 2.677950620651245
    },
    {
      "auxiliary_loss_clip": 0.01134268,
      "auxiliary_loss_mlp": 0.01043602,
      "balance_loss_clip": 1.04913616,
      "balance_loss_mlp": 1.02789068,
      "epoch": 0.3920036073951601,
      "flos": 23508956373120.0,
      "grad_norm": 1.9278475974219398,
      "language_loss": 0.66945064,
      "learning_rate": 2.775744388563563e-06,
      "loss": 0.69122934,
      "num_input_tokens_seen": 140029400,
      "step": 6520,
      "time_per_iteration": 2.5103533267974854
    },
    {
      "auxiliary_loss_clip": 0.01123073,
      "auxiliary_loss_mlp": 0.01037454,
      "balance_loss_clip": 1.04433596,
      "balance_loss_mlp": 1.02305925,
      "epoch": 0.39206373064782807,
      "flos": 18406086635520.0,
      "grad_norm": 1.960270606631519,
      "language_loss": 0.78542721,
      "learning_rate": 2.775385401898104e-06,
      "loss": 0.80703247,
      "num_input_tokens_seen": 140048940,
      "step": 6521,
      "time_per_iteration": 2.480276346206665
    },
    {
      "auxiliary_loss_clip": 0.01118085,
      "auxiliary_loss_mlp": 0.01033948,
      "balance_loss_clip": 1.04567254,
      "balance_loss_mlp": 1.01657927,
      "epoch": 0.39212385390049603,
      "flos": 12313051608960.0,
      "grad_norm": 3.262720618462274,
      "language_loss": 0.70148295,
      "learning_rate": 2.775026385829952e-06,
      "loss": 0.72300327,
      "num_input_tokens_seen": 140066380,
      "step": 6522,
      "time_per_iteration": 2.486363172531128
    },
    {
      "auxiliary_loss_clip": 0.01099447,
      "auxiliary_loss_mlp": 0.01032382,
      "balance_loss_clip": 1.04256392,
      "balance_loss_mlp": 1.01773131,
      "epoch": 0.392183977153164,
      "flos": 19719160214400.0,
      "grad_norm": 2.05340382015813,
      "language_loss": 0.76611352,
      "learning_rate": 2.774667340372722e-06,
      "loss": 0.78743184,
      "num_input_tokens_seen": 140085275,
      "step": 6523,
      "time_per_iteration": 4.015795946121216
    },
    {
      "auxiliary_loss_clip": 0.01104173,
      "auxiliary_loss_mlp": 0.0104086,
      "balance_loss_clip": 1.04404521,
      "balance_loss_mlp": 1.02618575,
      "epoch": 0.39224410040583196,
      "flos": 33144902403840.0,
      "grad_norm": 2.4797829287964017,
      "language_loss": 0.618909,
      "learning_rate": 2.7743082655400293e-06,
      "loss": 0.64035928,
      "num_input_tokens_seen": 140105105,
      "step": 6524,
      "time_per_iteration": 2.6123106479644775
    },
    {
      "auxiliary_loss_clip": 0.01126133,
      "auxiliary_loss_mlp": 0.01035443,
      "balance_loss_clip": 1.04511094,
      "balance_loss_mlp": 1.01998138,
      "epoch": 0.39230422365849993,
      "flos": 27782434097280.0,
      "grad_norm": 2.343147961510626,
      "language_loss": 0.7385394,
      "learning_rate": 2.773949161345489e-06,
      "loss": 0.76015514,
      "num_input_tokens_seen": 140125645,
      "step": 6525,
      "time_per_iteration": 2.549816370010376
    },
    {
      "auxiliary_loss_clip": 0.01101573,
      "auxiliary_loss_mlp": 0.01040002,
      "balance_loss_clip": 1.04183841,
      "balance_loss_mlp": 1.02603042,
      "epoch": 0.3923643469111679,
      "flos": 17931634865280.0,
      "grad_norm": 2.0447693600668853,
      "language_loss": 0.81432021,
      "learning_rate": 2.773590027802719e-06,
      "loss": 0.83573592,
      "num_input_tokens_seen": 140141925,
      "step": 6526,
      "time_per_iteration": 3.847306966781616
    },
    {
      "auxiliary_loss_clip": 0.01112472,
      "auxiliary_loss_mlp": 0.01038072,
      "balance_loss_clip": 1.0431366,
      "balance_loss_mlp": 1.02410126,
      "epoch": 0.39242447016383586,
      "flos": 24059539019520.0,
      "grad_norm": 1.5071774783004153,
      "language_loss": 0.70080006,
      "learning_rate": 2.7732308649253383e-06,
      "loss": 0.72230554,
      "num_input_tokens_seen": 140160965,
      "step": 6527,
      "time_per_iteration": 3.9322421550750732
    },
    {
      "auxiliary_loss_clip": 0.010868,
      "auxiliary_loss_mlp": 0.01033133,
      "balance_loss_clip": 1.04493654,
      "balance_loss_mlp": 1.0186913,
      "epoch": 0.3924845934165038,
      "flos": 10664069016960.0,
      "grad_norm": 2.4529634122330033,
      "language_loss": 0.82066047,
      "learning_rate": 2.772871672726965e-06,
      "loss": 0.84185982,
      "num_input_tokens_seen": 140177780,
      "step": 6528,
      "time_per_iteration": 2.5281364917755127
    },
    {
      "auxiliary_loss_clip": 0.01097169,
      "auxiliary_loss_mlp": 0.01037867,
      "balance_loss_clip": 1.04245925,
      "balance_loss_mlp": 1.02298331,
      "epoch": 0.3925447166691718,
      "flos": 31245910174080.0,
      "grad_norm": 1.6067383373223,
      "language_loss": 0.68480587,
      "learning_rate": 2.7725124512212205e-06,
      "loss": 0.70615625,
      "num_input_tokens_seen": 140201660,
      "step": 6529,
      "time_per_iteration": 4.059704065322876
    },
    {
      "auxiliary_loss_clip": 0.01102352,
      "auxiliary_loss_mlp": 0.0103584,
      "balance_loss_clip": 1.04040909,
      "balance_loss_mlp": 1.02063501,
      "epoch": 0.39260483992183975,
      "flos": 29415040087680.0,
      "grad_norm": 2.25770616302755,
      "language_loss": 0.79929024,
      "learning_rate": 2.7721532004217267e-06,
      "loss": 0.82067215,
      "num_input_tokens_seen": 140218585,
      "step": 6530,
      "time_per_iteration": 2.5790302753448486
    },
    {
      "auxiliary_loss_clip": 0.01111962,
      "auxiliary_loss_mlp": 0.01036436,
      "balance_loss_clip": 1.04183507,
      "balance_loss_mlp": 1.02202344,
      "epoch": 0.3926649631745077,
      "flos": 22857788666880.0,
      "grad_norm": 1.423643310669421,
      "language_loss": 0.7554062,
      "learning_rate": 2.7717939203421063e-06,
      "loss": 0.77689016,
      "num_input_tokens_seen": 140239905,
      "step": 6531,
      "time_per_iteration": 2.5233864784240723
    },
    {
      "auxiliary_loss_clip": 0.01050476,
      "auxiliary_loss_mlp": 0.01003834,
      "balance_loss_clip": 1.02201676,
      "balance_loss_mlp": 1.00217712,
      "epoch": 0.3927250864271757,
      "flos": 63893881872000.0,
      "grad_norm": 0.8310896388553287,
      "language_loss": 0.60358053,
      "learning_rate": 2.7714346109959822e-06,
      "loss": 0.62412363,
      "num_input_tokens_seen": 140293820,
      "step": 6532,
      "time_per_iteration": 2.960930347442627
    },
    {
      "auxiliary_loss_clip": 0.01032531,
      "auxiliary_loss_mlp": 0.01005284,
      "balance_loss_clip": 1.02552342,
      "balance_loss_mlp": 1.00356793,
      "epoch": 0.3927852096798437,
      "flos": 68909741890560.0,
      "grad_norm": 0.7742295183868704,
      "language_loss": 0.55504549,
      "learning_rate": 2.771075272396981e-06,
      "loss": 0.57542366,
      "num_input_tokens_seen": 140360420,
      "step": 6533,
      "time_per_iteration": 3.2317259311676025
    },
    {
      "auxiliary_loss_clip": 0.01111486,
      "auxiliary_loss_mlp": 0.01038896,
      "balance_loss_clip": 1.04632211,
      "balance_loss_mlp": 1.02404261,
      "epoch": 0.39284533293251167,
      "flos": 29715972232320.0,
      "grad_norm": 1.983423524287572,
      "language_loss": 0.75975168,
      "learning_rate": 2.7707159045587284e-06,
      "loss": 0.78125548,
      "num_input_tokens_seen": 140381950,
      "step": 6534,
      "time_per_iteration": 2.609792470932007
    },
    {
      "auxiliary_loss_clip": 0.01113967,
      "auxiliary_loss_mlp": 0.01043836,
      "balance_loss_clip": 1.04352331,
      "balance_loss_mlp": 1.02645552,
      "epoch": 0.39290545618517964,
      "flos": 18552027594240.0,
      "grad_norm": 2.0380364986485735,
      "language_loss": 0.78100467,
      "learning_rate": 2.770356507494851e-06,
      "loss": 0.80258262,
      "num_input_tokens_seen": 140399410,
      "step": 6535,
      "time_per_iteration": 2.4967849254608154
    },
    {
      "auxiliary_loss_clip": 0.01088841,
      "auxiliary_loss_mlp": 0.01029817,
      "balance_loss_clip": 1.04833293,
      "balance_loss_mlp": 1.01594114,
      "epoch": 0.3929655794378476,
      "flos": 26249479413120.0,
      "grad_norm": 1.8878762722402969,
      "language_loss": 0.6833148,
      "learning_rate": 2.769997081218978e-06,
      "loss": 0.70450139,
      "num_input_tokens_seen": 140419055,
      "step": 6536,
      "time_per_iteration": 2.5910658836364746
    },
    {
      "auxiliary_loss_clip": 0.01093825,
      "auxiliary_loss_mlp": 0.01032906,
      "balance_loss_clip": 1.04348433,
      "balance_loss_mlp": 1.01940536,
      "epoch": 0.39302570269051557,
      "flos": 29277933874560.0,
      "grad_norm": 1.8098784897639015,
      "language_loss": 0.69017893,
      "learning_rate": 2.769637625744738e-06,
      "loss": 0.71144623,
      "num_input_tokens_seen": 140438800,
      "step": 6537,
      "time_per_iteration": 2.598487377166748
    },
    {
      "auxiliary_loss_clip": 0.01110906,
      "auxiliary_loss_mlp": 0.01033041,
      "balance_loss_clip": 1.04555035,
      "balance_loss_mlp": 1.01828933,
      "epoch": 0.39308582594318353,
      "flos": 17347440067200.0,
      "grad_norm": 1.9237619915085742,
      "language_loss": 0.78995526,
      "learning_rate": 2.769278141085763e-06,
      "loss": 0.81139475,
      "num_input_tokens_seen": 140456880,
      "step": 6538,
      "time_per_iteration": 2.461862325668335
    },
    {
      "auxiliary_loss_clip": 0.01018406,
      "auxiliary_loss_mlp": 0.01005089,
      "balance_loss_clip": 1.03274381,
      "balance_loss_mlp": 1.00306249,
      "epoch": 0.3931459491958515,
      "flos": 61007094650880.0,
      "grad_norm": 0.8028394820817766,
      "language_loss": 0.61897016,
      "learning_rate": 2.768918627255683e-06,
      "loss": 0.6392051,
      "num_input_tokens_seen": 140507510,
      "step": 6539,
      "time_per_iteration": 2.9905731678009033
    },
    {
      "auxiliary_loss_clip": 0.01101918,
      "auxiliary_loss_mlp": 0.01033936,
      "balance_loss_clip": 1.04968524,
      "balance_loss_mlp": 1.01885617,
      "epoch": 0.39320607244851946,
      "flos": 39016009249920.0,
      "grad_norm": 1.9362877031887438,
      "language_loss": 0.67907995,
      "learning_rate": 2.7685590842681315e-06,
      "loss": 0.7004385,
      "num_input_tokens_seen": 140528740,
      "step": 6540,
      "time_per_iteration": 2.6699414253234863
    },
    {
      "auxiliary_loss_clip": 0.01101049,
      "auxiliary_loss_mlp": 0.0103526,
      "balance_loss_clip": 1.04391897,
      "balance_loss_mlp": 1.02122891,
      "epoch": 0.3932661957011874,
      "flos": 24679752180480.0,
      "grad_norm": 1.708695819762661,
      "language_loss": 0.7223866,
      "learning_rate": 2.7681995121367433e-06,
      "loss": 0.74374968,
      "num_input_tokens_seen": 140547560,
      "step": 6541,
      "time_per_iteration": 2.574753522872925
    },
    {
      "auxiliary_loss_clip": 0.01049207,
      "auxiliary_loss_mlp": 0.01004506,
      "balance_loss_clip": 1.02097559,
      "balance_loss_mlp": 1.00269401,
      "epoch": 0.3933263189538554,
      "flos": 70096552185600.0,
      "grad_norm": 0.8389220469040073,
      "language_loss": 0.6035558,
      "learning_rate": 2.7678399108751516e-06,
      "loss": 0.62409294,
      "num_input_tokens_seen": 140601175,
      "step": 6542,
      "time_per_iteration": 2.9125678539276123
    },
    {
      "auxiliary_loss_clip": 0.01113873,
      "auxiliary_loss_mlp": 0.01036702,
      "balance_loss_clip": 1.0432632,
      "balance_loss_mlp": 1.02279639,
      "epoch": 0.39338644220652336,
      "flos": 22929071207040.0,
      "grad_norm": 1.4779570516063214,
      "language_loss": 0.82212019,
      "learning_rate": 2.7674802804969947e-06,
      "loss": 0.84362596,
      "num_input_tokens_seen": 140622200,
      "step": 6543,
      "time_per_iteration": 2.5446014404296875
    },
    {
      "auxiliary_loss_clip": 0.01097907,
      "auxiliary_loss_mlp": 0.01037998,
      "balance_loss_clip": 1.04045773,
      "balance_loss_mlp": 1.02325153,
      "epoch": 0.3934465654591913,
      "flos": 30848163897600.0,
      "grad_norm": 1.5730113541721435,
      "language_loss": 0.68948078,
      "learning_rate": 2.767120621015908e-06,
      "loss": 0.71083987,
      "num_input_tokens_seen": 140643125,
      "step": 6544,
      "time_per_iteration": 2.5975804328918457
    },
    {
      "auxiliary_loss_clip": 0.01107861,
      "auxiliary_loss_mlp": 0.01038419,
      "balance_loss_clip": 1.04490149,
      "balance_loss_mlp": 1.02310073,
      "epoch": 0.3935066887118593,
      "flos": 29236528471680.0,
      "grad_norm": 2.704837534210723,
      "language_loss": 0.75244486,
      "learning_rate": 2.76676093244553e-06,
      "loss": 0.7739076,
      "num_input_tokens_seen": 140662500,
      "step": 6545,
      "time_per_iteration": 2.6029927730560303
    },
    {
      "auxiliary_loss_clip": 0.01086334,
      "auxiliary_loss_mlp": 0.0103279,
      "balance_loss_clip": 1.0454241,
      "balance_loss_mlp": 1.02014148,
      "epoch": 0.3935668119645273,
      "flos": 19135288638720.0,
      "grad_norm": 1.570697980406247,
      "language_loss": 0.74669802,
      "learning_rate": 2.7664012147995015e-06,
      "loss": 0.76788926,
      "num_input_tokens_seen": 140681960,
      "step": 6546,
      "time_per_iteration": 2.563077449798584
    },
    {
      "auxiliary_loss_clip": 0.01101233,
      "auxiliary_loss_mlp": 0.01033979,
      "balance_loss_clip": 1.04756069,
      "balance_loss_mlp": 1.01876783,
      "epoch": 0.3936269352171953,
      "flos": 18516116972160.0,
      "grad_norm": 1.7126700973427667,
      "language_loss": 0.8162117,
      "learning_rate": 2.7660414680914617e-06,
      "loss": 0.83756381,
      "num_input_tokens_seen": 140699170,
      "step": 6547,
      "time_per_iteration": 2.530360460281372
    },
    {
      "auxiliary_loss_clip": 0.01110186,
      "auxiliary_loss_mlp": 0.00795657,
      "balance_loss_clip": 1.04266548,
      "balance_loss_mlp": 1.01536107,
      "epoch": 0.39368705846986324,
      "flos": 15632813370240.0,
      "grad_norm": 1.7096843779625146,
      "language_loss": 0.840505,
      "learning_rate": 2.7656816923350525e-06,
      "loss": 0.85956347,
      "num_input_tokens_seen": 140714920,
      "step": 6548,
      "time_per_iteration": 2.468116521835327
    },
    {
      "auxiliary_loss_clip": 0.01113171,
      "auxiliary_loss_mlp": 0.00791205,
      "balance_loss_clip": 1.04479074,
      "balance_loss_mlp": 1.01175976,
      "epoch": 0.3937471817225312,
      "flos": 21325839563520.0,
      "grad_norm": 1.5328264746783393,
      "language_loss": 0.72748768,
      "learning_rate": 2.7653218875439174e-06,
      "loss": 0.74653137,
      "num_input_tokens_seen": 140734595,
      "step": 6549,
      "time_per_iteration": 2.5336458683013916
    },
    {
      "auxiliary_loss_clip": 0.01067894,
      "auxiliary_loss_mlp": 0.0104327,
      "balance_loss_clip": 1.04581392,
      "balance_loss_mlp": 1.02672362,
      "epoch": 0.39380730497519917,
      "flos": 20776693461120.0,
      "grad_norm": 1.728756609781234,
      "language_loss": 0.77350819,
      "learning_rate": 2.764962053731699e-06,
      "loss": 0.7946198,
      "num_input_tokens_seen": 140754050,
      "step": 6550,
      "time_per_iteration": 2.635369300842285
    },
    {
      "auxiliary_loss_clip": 0.01103062,
      "auxiliary_loss_mlp": 0.01029657,
      "balance_loss_clip": 1.06046855,
      "balance_loss_mlp": 1.01573312,
      "epoch": 0.39386742822786713,
      "flos": 21609784575360.0,
      "grad_norm": 1.7499595006687143,
      "language_loss": 0.81511652,
      "learning_rate": 2.7646021909120434e-06,
      "loss": 0.83644372,
      "num_input_tokens_seen": 140771440,
      "step": 6551,
      "time_per_iteration": 2.572779417037964
    },
    {
      "auxiliary_loss_clip": 0.01114745,
      "auxiliary_loss_mlp": 0.01041938,
      "balance_loss_clip": 1.04414713,
      "balance_loss_mlp": 1.02710807,
      "epoch": 0.3939275514805351,
      "flos": 12414642249600.0,
      "grad_norm": 2.3453215929744973,
      "language_loss": 0.79836428,
      "learning_rate": 2.764242299098596e-06,
      "loss": 0.81993103,
      "num_input_tokens_seen": 140786715,
      "step": 6552,
      "time_per_iteration": 2.4657928943634033
    },
    {
      "auxiliary_loss_clip": 0.01128403,
      "auxiliary_loss_mlp": 0.01036938,
      "balance_loss_clip": 1.04620075,
      "balance_loss_mlp": 1.02182174,
      "epoch": 0.39398767473320306,
      "flos": 18552027594240.0,
      "grad_norm": 1.6739822262462836,
      "language_loss": 0.7081989,
      "learning_rate": 2.763882378305003e-06,
      "loss": 0.72985232,
      "num_input_tokens_seen": 140804950,
      "step": 6553,
      "time_per_iteration": 2.4562976360321045
    },
    {
      "auxiliary_loss_clip": 0.01113904,
      "auxiliary_loss_mlp": 0.00791218,
      "balance_loss_clip": 1.04650545,
      "balance_loss_mlp": 1.01062727,
      "epoch": 0.39404779798587103,
      "flos": 29308888419840.0,
      "grad_norm": 1.58270591146748,
      "language_loss": 0.64257771,
      "learning_rate": 2.7635224285449144e-06,
      "loss": 0.66162896,
      "num_input_tokens_seen": 140822800,
      "step": 6554,
      "time_per_iteration": 2.562225103378296
    },
    {
      "auxiliary_loss_clip": 0.01111934,
      "auxiliary_loss_mlp": 0.0103973,
      "balance_loss_clip": 1.04912746,
      "balance_loss_mlp": 1.02606916,
      "epoch": 0.394107921238539,
      "flos": 34897055834880.0,
      "grad_norm": 2.3897396205959094,
      "language_loss": 0.7974636,
      "learning_rate": 2.7631624498319796e-06,
      "loss": 0.81898028,
      "num_input_tokens_seen": 140842940,
      "step": 6555,
      "time_per_iteration": 2.635467052459717
    },
    {
      "auxiliary_loss_clip": 0.01101485,
      "auxiliary_loss_mlp": 0.01040731,
      "balance_loss_clip": 1.04772854,
      "balance_loss_mlp": 1.02494729,
      "epoch": 0.39416804449120696,
      "flos": 25081413039360.0,
      "grad_norm": 1.937172891928768,
      "language_loss": 0.71547353,
      "learning_rate": 2.7628024421798473e-06,
      "loss": 0.73689568,
      "num_input_tokens_seen": 140863060,
      "step": 6556,
      "time_per_iteration": 2.553980588912964
    },
    {
      "auxiliary_loss_clip": 0.01125938,
      "auxiliary_loss_mlp": 0.01033758,
      "balance_loss_clip": 1.04393125,
      "balance_loss_mlp": 1.01919675,
      "epoch": 0.3942281677438749,
      "flos": 32306639731200.0,
      "grad_norm": 1.8511526183447458,
      "language_loss": 0.83618891,
      "learning_rate": 2.7624424056021705e-06,
      "loss": 0.85778588,
      "num_input_tokens_seen": 140883795,
      "step": 6557,
      "time_per_iteration": 2.5707714557647705
    },
    {
      "auxiliary_loss_clip": 0.01110106,
      "auxiliary_loss_mlp": 0.01036063,
      "balance_loss_clip": 1.04523647,
      "balance_loss_mlp": 1.02130473,
      "epoch": 0.3942882909965429,
      "flos": 24936621315840.0,
      "grad_norm": 2.4820357961948867,
      "language_loss": 0.80513418,
      "learning_rate": 2.7620823401126004e-06,
      "loss": 0.8265959,
      "num_input_tokens_seen": 140903055,
      "step": 6558,
      "time_per_iteration": 2.5180160999298096
    },
    {
      "auxiliary_loss_clip": 0.01125617,
      "auxiliary_loss_mlp": 0.01040556,
      "balance_loss_clip": 1.04603815,
      "balance_loss_mlp": 1.02644217,
      "epoch": 0.39434841424921085,
      "flos": 11874797769600.0,
      "grad_norm": 1.78445479156159,
      "language_loss": 0.71064246,
      "learning_rate": 2.761722245724792e-06,
      "loss": 0.73230422,
      "num_input_tokens_seen": 140920685,
      "step": 6559,
      "time_per_iteration": 2.445587158203125
    },
    {
      "auxiliary_loss_clip": 0.01108692,
      "auxiliary_loss_mlp": 0.01037462,
      "balance_loss_clip": 1.0456531,
      "balance_loss_mlp": 1.02108836,
      "epoch": 0.3944085375018789,
      "flos": 16361620323840.0,
      "grad_norm": 2.7322800436506496,
      "language_loss": 0.80880839,
      "learning_rate": 2.7613621224524003e-06,
      "loss": 0.83026993,
      "num_input_tokens_seen": 140937320,
      "step": 6560,
      "time_per_iteration": 2.4943389892578125
    },
    {
      "auxiliary_loss_clip": 0.01113483,
      "auxiliary_loss_mlp": 0.0103632,
      "balance_loss_clip": 1.05260634,
      "balance_loss_mlp": 1.02072716,
      "epoch": 0.39446866075454684,
      "flos": 10633365866880.0,
      "grad_norm": 1.9981557721491578,
      "language_loss": 0.83373284,
      "learning_rate": 2.7610019703090803e-06,
      "loss": 0.85523093,
      "num_input_tokens_seen": 140954855,
      "step": 6561,
      "time_per_iteration": 2.5036377906799316
    },
    {
      "auxiliary_loss_clip": 0.01116099,
      "auxiliary_loss_mlp": 0.0104303,
      "balance_loss_clip": 1.04688346,
      "balance_loss_mlp": 1.02835572,
      "epoch": 0.3945287840072148,
      "flos": 18187498419840.0,
      "grad_norm": 3.024338987553006,
      "language_loss": 0.7995165,
      "learning_rate": 2.7606417893084887e-06,
      "loss": 0.8211078,
      "num_input_tokens_seen": 140973250,
      "step": 6562,
      "time_per_iteration": 3.8899197578430176
    },
    {
      "auxiliary_loss_clip": 0.0109792,
      "auxiliary_loss_mlp": 0.01036434,
      "balance_loss_clip": 1.04428542,
      "balance_loss_mlp": 1.02216494,
      "epoch": 0.39458890725988277,
      "flos": 23039891642880.0,
      "grad_norm": 1.749635834600201,
      "language_loss": 0.81152719,
      "learning_rate": 2.7602815794642853e-06,
      "loss": 0.83287072,
      "num_input_tokens_seen": 140993050,
      "step": 6563,
      "time_per_iteration": 2.570053815841675
    },
    {
      "auxiliary_loss_clip": 0.01076978,
      "auxiliary_loss_mlp": 0.01047176,
      "balance_loss_clip": 1.04270768,
      "balance_loss_mlp": 1.02904415,
      "epoch": 0.39464903051255074,
      "flos": 17159052211200.0,
      "grad_norm": 2.190937813483821,
      "language_loss": 0.70371163,
      "learning_rate": 2.759921340790127e-06,
      "loss": 0.72495317,
      "num_input_tokens_seen": 141010815,
      "step": 6564,
      "time_per_iteration": 3.956890821456909
    },
    {
      "auxiliary_loss_clip": 0.01115144,
      "auxiliary_loss_mlp": 0.01037708,
      "balance_loss_clip": 1.0438118,
      "balance_loss_mlp": 1.02277684,
      "epoch": 0.3947091537652187,
      "flos": 15889000147200.0,
      "grad_norm": 2.1480147839194457,
      "language_loss": 0.83046401,
      "learning_rate": 2.759561073299676e-06,
      "loss": 0.85199255,
      "num_input_tokens_seen": 141028720,
      "step": 6565,
      "time_per_iteration": 2.472045421600342
    },
    {
      "auxiliary_loss_clip": 0.01087954,
      "auxiliary_loss_mlp": 0.01044529,
      "balance_loss_clip": 1.04133546,
      "balance_loss_mlp": 1.02793467,
      "epoch": 0.39476927701788667,
      "flos": 18545491319040.0,
      "grad_norm": 1.8136728200934547,
      "language_loss": 0.83558649,
      "learning_rate": 2.7592007770065937e-06,
      "loss": 0.8569113,
      "num_input_tokens_seen": 141046025,
      "step": 6566,
      "time_per_iteration": 3.9070703983306885
    },
    {
      "auxiliary_loss_clip": 0.01132152,
      "auxiliary_loss_mlp": 0.01038013,
      "balance_loss_clip": 1.04662907,
      "balance_loss_mlp": 1.02329624,
      "epoch": 0.39482940027055463,
      "flos": 22275712771200.0,
      "grad_norm": 2.3199407656714714,
      "language_loss": 0.77309573,
      "learning_rate": 2.7588404519245403e-06,
      "loss": 0.79479742,
      "num_input_tokens_seen": 141066865,
      "step": 6567,
      "time_per_iteration": 2.5052387714385986
    },
    {
      "auxiliary_loss_clip": 0.01111325,
      "auxiliary_loss_mlp": 0.01034143,
      "balance_loss_clip": 1.0444901,
      "balance_loss_mlp": 1.01999867,
      "epoch": 0.3948895235232226,
      "flos": 14757634494720.0,
      "grad_norm": 1.9993661265184481,
      "language_loss": 0.80613661,
      "learning_rate": 2.758480098067182e-06,
      "loss": 0.8275913,
      "num_input_tokens_seen": 141084210,
      "step": 6568,
      "time_per_iteration": 3.8558616638183594
    },
    {
      "auxiliary_loss_clip": 0.01093208,
      "auxiliary_loss_mlp": 0.01033372,
      "balance_loss_clip": 1.04424381,
      "balance_loss_mlp": 1.0186733,
      "epoch": 0.39494964677589056,
      "flos": 22565763095040.0,
      "grad_norm": 1.699410380454577,
      "language_loss": 0.84486055,
      "learning_rate": 2.7581197154481816e-06,
      "loss": 0.86612642,
      "num_input_tokens_seen": 141103895,
      "step": 6569,
      "time_per_iteration": 2.5607333183288574
    },
    {
      "auxiliary_loss_clip": 0.01071034,
      "auxiliary_loss_mlp": 0.01036593,
      "balance_loss_clip": 1.04681683,
      "balance_loss_mlp": 1.02203751,
      "epoch": 0.3950097700285585,
      "flos": 22963186149120.0,
      "grad_norm": 1.8535430898870493,
      "language_loss": 0.74630904,
      "learning_rate": 2.7577593040812066e-06,
      "loss": 0.76738536,
      "num_input_tokens_seen": 141124000,
      "step": 6570,
      "time_per_iteration": 2.6543033123016357
    },
    {
      "auxiliary_loss_clip": 0.01099383,
      "auxiliary_loss_mlp": 0.01034956,
      "balance_loss_clip": 1.05071199,
      "balance_loss_mlp": 1.02040088,
      "epoch": 0.3950698932812265,
      "flos": 20595236929920.0,
      "grad_norm": 1.582899525692935,
      "language_loss": 0.79721045,
      "learning_rate": 2.757398863979922e-06,
      "loss": 0.81855381,
      "num_input_tokens_seen": 141142535,
      "step": 6571,
      "time_per_iteration": 2.5591237545013428
    },
    {
      "auxiliary_loss_clip": 0.01100578,
      "auxiliary_loss_mlp": 0.01040897,
      "balance_loss_clip": 1.0495156,
      "balance_loss_mlp": 1.02608538,
      "epoch": 0.39513001653389446,
      "flos": 20375786787840.0,
      "grad_norm": 2.2687270225249376,
      "language_loss": 0.77775174,
      "learning_rate": 2.757038395157997e-06,
      "loss": 0.7991665,
      "num_input_tokens_seen": 141161575,
      "step": 6572,
      "time_per_iteration": 2.5371758937835693
    },
    {
      "auxiliary_loss_clip": 0.01089849,
      "auxiliary_loss_mlp": 0.01038993,
      "balance_loss_clip": 1.04633033,
      "balance_loss_mlp": 1.02412796,
      "epoch": 0.3951901397865625,
      "flos": 26463650256000.0,
      "grad_norm": 1.7983139781578383,
      "language_loss": 0.75037402,
      "learning_rate": 2.7566778976291002e-06,
      "loss": 0.77166241,
      "num_input_tokens_seen": 141181150,
      "step": 6573,
      "time_per_iteration": 2.596883773803711
    },
    {
      "auxiliary_loss_clip": 0.01113084,
      "auxiliary_loss_mlp": 0.0103242,
      "balance_loss_clip": 1.04378939,
      "balance_loss_mlp": 1.01917624,
      "epoch": 0.39525026303923044,
      "flos": 43838345767680.0,
      "grad_norm": 1.572324794709649,
      "language_loss": 0.67958832,
      "learning_rate": 2.7563173714069017e-06,
      "loss": 0.70104337,
      "num_input_tokens_seen": 141206310,
      "step": 6574,
      "time_per_iteration": 2.7271339893341064
    },
    {
      "auxiliary_loss_clip": 0.01066327,
      "auxiliary_loss_mlp": 0.01038054,
      "balance_loss_clip": 1.04167199,
      "balance_loss_mlp": 1.02085173,
      "epoch": 0.3953103862918984,
      "flos": 18040803275520.0,
      "grad_norm": 2.429583157911912,
      "language_loss": 0.72135991,
      "learning_rate": 2.755956816505072e-06,
      "loss": 0.74240375,
      "num_input_tokens_seen": 141223925,
      "step": 6575,
      "time_per_iteration": 2.6008336544036865
    },
    {
      "auxiliary_loss_clip": 0.01111418,
      "auxiliary_loss_mlp": 0.01041134,
      "balance_loss_clip": 1.04874873,
      "balance_loss_mlp": 1.02611351,
      "epoch": 0.3953705095445664,
      "flos": 16976015481600.0,
      "grad_norm": 2.083257067323189,
      "language_loss": 0.73768151,
      "learning_rate": 2.7555962329372845e-06,
      "loss": 0.75920701,
      "num_input_tokens_seen": 141239010,
      "step": 6576,
      "time_per_iteration": 2.4867701530456543
    },
    {
      "auxiliary_loss_clip": 0.01125924,
      "auxiliary_loss_mlp": 0.01037592,
      "balance_loss_clip": 1.04497743,
      "balance_loss_mlp": 1.02448463,
      "epoch": 0.39543063279723434,
      "flos": 17411144837760.0,
      "grad_norm": 2.4136880488074945,
      "language_loss": 0.83418632,
      "learning_rate": 2.7552356207172124e-06,
      "loss": 0.85582149,
      "num_input_tokens_seen": 141252255,
      "step": 6577,
      "time_per_iteration": 2.4168448448181152
    },
    {
      "auxiliary_loss_clip": 0.01103621,
      "auxiliary_loss_mlp": 0.01032801,
      "balance_loss_clip": 1.04427588,
      "balance_loss_mlp": 1.01922894,
      "epoch": 0.3954907560499023,
      "flos": 22784207656320.0,
      "grad_norm": 2.7732685660571748,
      "language_loss": 0.90222919,
      "learning_rate": 2.75487497985853e-06,
      "loss": 0.9235934,
      "num_input_tokens_seen": 141269325,
      "step": 6578,
      "time_per_iteration": 2.5963001251220703
    },
    {
      "auxiliary_loss_clip": 0.01102837,
      "auxiliary_loss_mlp": 0.01035111,
      "balance_loss_clip": 1.04609728,
      "balance_loss_mlp": 1.01905346,
      "epoch": 0.39555087930257027,
      "flos": 21944400698880.0,
      "grad_norm": 1.8165639161378564,
      "language_loss": 0.77991134,
      "learning_rate": 2.7545143103749117e-06,
      "loss": 0.80129081,
      "num_input_tokens_seen": 141288505,
      "step": 6579,
      "time_per_iteration": 2.54209303855896
    },
    {
      "auxiliary_loss_clip": 0.01076232,
      "auxiliary_loss_mlp": 0.01030035,
      "balance_loss_clip": 1.0478344,
      "balance_loss_mlp": 1.01483595,
      "epoch": 0.39561100255523823,
      "flos": 20404622430720.0,
      "grad_norm": 2.0786960025374723,
      "language_loss": 0.68783784,
      "learning_rate": 2.754153612280037e-06,
      "loss": 0.70890045,
      "num_input_tokens_seen": 141303680,
      "step": 6580,
      "time_per_iteration": 2.6481897830963135
    },
    {
      "auxiliary_loss_clip": 0.01113665,
      "auxiliary_loss_mlp": 0.01028537,
      "balance_loss_clip": 1.0456593,
      "balance_loss_mlp": 1.01485729,
      "epoch": 0.3956711258079062,
      "flos": 27964572986880.0,
      "grad_norm": 1.9217436116661628,
      "language_loss": 0.58528984,
      "learning_rate": 2.7537928855875797e-06,
      "loss": 0.60671186,
      "num_input_tokens_seen": 141324090,
      "step": 6581,
      "time_per_iteration": 2.603363513946533
    },
    {
      "auxiliary_loss_clip": 0.01101212,
      "auxiliary_loss_mlp": 0.01045805,
      "balance_loss_clip": 1.04313087,
      "balance_loss_mlp": 1.02952147,
      "epoch": 0.39573124906057416,
      "flos": 14428297670400.0,
      "grad_norm": 1.8135305314487877,
      "language_loss": 0.69298685,
      "learning_rate": 2.7534321303112224e-06,
      "loss": 0.71445704,
      "num_input_tokens_seen": 141342235,
      "step": 6582,
      "time_per_iteration": 2.5629775524139404
    },
    {
      "auxiliary_loss_clip": 0.01127606,
      "auxiliary_loss_mlp": 0.00793416,
      "balance_loss_clip": 1.04615068,
      "balance_loss_mlp": 1.01743054,
      "epoch": 0.39579137231324213,
      "flos": 18733699607040.0,
      "grad_norm": 1.976634325652489,
      "language_loss": 0.759323,
      "learning_rate": 2.753071346464642e-06,
      "loss": 0.77853322,
      "num_input_tokens_seen": 141361195,
      "step": 6583,
      "time_per_iteration": 2.524796962738037
    },
    {
      "auxiliary_loss_clip": 0.01088575,
      "auxiliary_loss_mlp": 0.00797774,
      "balance_loss_clip": 1.04538512,
      "balance_loss_mlp": 1.02646732,
      "epoch": 0.3958514955659101,
      "flos": 17676417755520.0,
      "grad_norm": 1.75031678415596,
      "language_loss": 0.65879476,
      "learning_rate": 2.7527105340615207e-06,
      "loss": 0.67765826,
      "num_input_tokens_seen": 141378275,
      "step": 6584,
      "time_per_iteration": 2.6191680431365967
    },
    {
      "auxiliary_loss_clip": 0.01097822,
      "auxiliary_loss_mlp": 0.0104114,
      "balance_loss_clip": 1.05322647,
      "balance_loss_mlp": 1.02590466,
      "epoch": 0.39591161881857806,
      "flos": 29309103901440.0,
      "grad_norm": 1.9838399360125478,
      "language_loss": 0.72046393,
      "learning_rate": 2.7523496931155413e-06,
      "loss": 0.74185354,
      "num_input_tokens_seen": 141396960,
      "step": 6585,
      "time_per_iteration": 2.6208224296569824
    },
    {
      "auxiliary_loss_clip": 0.01092397,
      "auxiliary_loss_mlp": 0.01030345,
      "balance_loss_clip": 1.04241467,
      "balance_loss_mlp": 1.01628435,
      "epoch": 0.3959717420712461,
      "flos": 25771831332480.0,
      "grad_norm": 1.9189240103135463,
      "language_loss": 0.73699617,
      "learning_rate": 2.7519888236403856e-06,
      "loss": 0.75822365,
      "num_input_tokens_seen": 141417320,
      "step": 6586,
      "time_per_iteration": 2.6023528575897217
    },
    {
      "auxiliary_loss_clip": 0.01107101,
      "auxiliary_loss_mlp": 0.01034424,
      "balance_loss_clip": 1.04649425,
      "balance_loss_mlp": 1.019696,
      "epoch": 0.39603186532391405,
      "flos": 20923783655040.0,
      "grad_norm": 2.1662706459529755,
      "language_loss": 0.71514916,
      "learning_rate": 2.7516279256497382e-06,
      "loss": 0.73656446,
      "num_input_tokens_seen": 141435985,
      "step": 6587,
      "time_per_iteration": 2.53814697265625
    },
    {
      "auxiliary_loss_clip": 0.01014973,
      "auxiliary_loss_mlp": 0.01009685,
      "balance_loss_clip": 1.02351689,
      "balance_loss_mlp": 1.00788546,
      "epoch": 0.396091988576582,
      "flos": 54880986176640.0,
      "grad_norm": 0.9005289410521834,
      "language_loss": 0.61182922,
      "learning_rate": 2.751266999157285e-06,
      "loss": 0.63207579,
      "num_input_tokens_seen": 141486075,
      "step": 6588,
      "time_per_iteration": 3.005932331085205
    },
    {
      "auxiliary_loss_clip": 0.01106546,
      "auxiliary_loss_mlp": 0.0079161,
      "balance_loss_clip": 1.04678178,
      "balance_loss_mlp": 1.01358974,
      "epoch": 0.39615211182925,
      "flos": 20702896968960.0,
      "grad_norm": 1.7862090698581379,
      "language_loss": 0.81270701,
      "learning_rate": 2.7509060441767115e-06,
      "loss": 0.83168852,
      "num_input_tokens_seen": 141505280,
      "step": 6589,
      "time_per_iteration": 2.531843423843384
    },
    {
      "auxiliary_loss_clip": 0.01100859,
      "auxiliary_loss_mlp": 0.01029457,
      "balance_loss_clip": 1.04531682,
      "balance_loss_mlp": 1.0152117,
      "epoch": 0.39621223508191794,
      "flos": 20994312009600.0,
      "grad_norm": 2.3163292881341566,
      "language_loss": 0.6999135,
      "learning_rate": 2.7505450607217057e-06,
      "loss": 0.72121668,
      "num_input_tokens_seen": 141523930,
      "step": 6590,
      "time_per_iteration": 2.5264840126037598
    },
    {
      "auxiliary_loss_clip": 0.01114325,
      "auxiliary_loss_mlp": 0.01047741,
      "balance_loss_clip": 1.0467031,
      "balance_loss_mlp": 1.03229129,
      "epoch": 0.3962723583345859,
      "flos": 23368833417600.0,
      "grad_norm": 1.9498492270278753,
      "language_loss": 0.75504047,
      "learning_rate": 2.750184048805956e-06,
      "loss": 0.7766611,
      "num_input_tokens_seen": 141541320,
      "step": 6591,
      "time_per_iteration": 2.5035622119903564
    },
    {
      "auxiliary_loss_clip": 0.01044753,
      "auxiliary_loss_mlp": 0.01042906,
      "balance_loss_clip": 1.0472331,
      "balance_loss_mlp": 1.02826118,
      "epoch": 0.39633248158725387,
      "flos": 25115599808640.0,
      "grad_norm": 1.7904908142546463,
      "language_loss": 0.78353232,
      "learning_rate": 2.749823008443152e-06,
      "loss": 0.80440891,
      "num_input_tokens_seen": 141561880,
      "step": 6592,
      "time_per_iteration": 2.825254201889038
    },
    {
      "auxiliary_loss_clip": 0.01062211,
      "auxiliary_loss_mlp": 0.01033391,
      "balance_loss_clip": 1.04973197,
      "balance_loss_mlp": 1.01885378,
      "epoch": 0.39639260483992184,
      "flos": 39787622236800.0,
      "grad_norm": 1.8201657228851904,
      "language_loss": 0.69559866,
      "learning_rate": 2.7494619396469843e-06,
      "loss": 0.7165547,
      "num_input_tokens_seen": 141586460,
      "step": 6593,
      "time_per_iteration": 3.0249547958374023
    },
    {
      "auxiliary_loss_clip": 0.0106281,
      "auxiliary_loss_mlp": 0.01037727,
      "balance_loss_clip": 1.04100943,
      "balance_loss_mlp": 1.02254021,
      "epoch": 0.3964527280925898,
      "flos": 17347045017600.0,
      "grad_norm": 1.9540812094946873,
      "language_loss": 0.7753067,
      "learning_rate": 2.7491008424311452e-06,
      "loss": 0.79631209,
      "num_input_tokens_seen": 141605955,
      "step": 6594,
      "time_per_iteration": 2.648566961288452
    },
    {
      "auxiliary_loss_clip": 0.01023848,
      "auxiliary_loss_mlp": 0.0100475,
      "balance_loss_clip": 1.028072,
      "balance_loss_mlp": 1.0028547,
      "epoch": 0.39651285134525777,
      "flos": 71717848369920.0,
      "grad_norm": 0.9477409190783928,
      "language_loss": 0.63049734,
      "learning_rate": 2.7487397168093265e-06,
      "loss": 0.6507833,
      "num_input_tokens_seen": 141673140,
      "step": 6595,
      "time_per_iteration": 3.2045650482177734
    },
    {
      "auxiliary_loss_clip": 0.01094592,
      "auxiliary_loss_mlp": 0.01049683,
      "balance_loss_clip": 1.04693985,
      "balance_loss_mlp": 1.03273129,
      "epoch": 0.39657297459792573,
      "flos": 25775710001280.0,
      "grad_norm": 2.814337477310836,
      "language_loss": 0.63442439,
      "learning_rate": 2.748378562795223e-06,
      "loss": 0.6558671,
      "num_input_tokens_seen": 141692955,
      "step": 6596,
      "time_per_iteration": 2.624256134033203
    },
    {
      "auxiliary_loss_clip": 0.01114888,
      "auxiliary_loss_mlp": 0.01038557,
      "balance_loss_clip": 1.04511738,
      "balance_loss_mlp": 1.02397799,
      "epoch": 0.3966330978505937,
      "flos": 20266115587200.0,
      "grad_norm": 1.8247907389215567,
      "language_loss": 0.78705901,
      "learning_rate": 2.7480173804025293e-06,
      "loss": 0.80859339,
      "num_input_tokens_seen": 141710680,
      "step": 6597,
      "time_per_iteration": 2.528625249862671
    },
    {
      "auxiliary_loss_clip": 0.01097273,
      "auxiliary_loss_mlp": 0.00791487,
      "balance_loss_clip": 1.04583549,
      "balance_loss_mlp": 1.01438141,
      "epoch": 0.39669322110326166,
      "flos": 20631183465600.0,
      "grad_norm": 2.020721703987871,
      "language_loss": 0.66946077,
      "learning_rate": 2.747656169644941e-06,
      "loss": 0.68834841,
      "num_input_tokens_seen": 141729860,
      "step": 6598,
      "time_per_iteration": 2.585181713104248
    },
    {
      "auxiliary_loss_clip": 0.01129011,
      "auxiliary_loss_mlp": 0.0103925,
      "balance_loss_clip": 1.04694188,
      "balance_loss_mlp": 1.02585089,
      "epoch": 0.3967533443559297,
      "flos": 21726063878400.0,
      "grad_norm": 1.811380785757539,
      "language_loss": 0.78609788,
      "learning_rate": 2.747294930536157e-06,
      "loss": 0.80778044,
      "num_input_tokens_seen": 141749060,
      "step": 6599,
      "time_per_iteration": 2.5095973014831543
    },
    {
      "auxiliary_loss_clip": 0.0109857,
      "auxiliary_loss_mlp": 0.01038643,
      "balance_loss_clip": 1.05054116,
      "balance_loss_mlp": 1.02253771,
      "epoch": 0.39681346760859765,
      "flos": 25484151306240.0,
      "grad_norm": 1.9166971724217612,
      "language_loss": 0.729312,
      "learning_rate": 2.7469336630898737e-06,
      "loss": 0.75068414,
      "num_input_tokens_seen": 141769860,
      "step": 6600,
      "time_per_iteration": 4.004424095153809
    },
    {
      "auxiliary_loss_clip": 0.01083259,
      "auxiliary_loss_mlp": 0.01035373,
      "balance_loss_clip": 1.04217076,
      "balance_loss_mlp": 1.02041864,
      "epoch": 0.3968735908612656,
      "flos": 20959586536320.0,
      "grad_norm": 2.254581373045382,
      "language_loss": 0.85992217,
      "learning_rate": 2.746572367319791e-06,
      "loss": 0.88110846,
      "num_input_tokens_seen": 141788465,
      "step": 6601,
      "time_per_iteration": 2.5845186710357666
    },
    {
      "auxiliary_loss_clip": 0.01092844,
      "auxiliary_loss_mlp": 0.01040912,
      "balance_loss_clip": 1.04349339,
      "balance_loss_mlp": 1.02445543,
      "epoch": 0.3969337141139336,
      "flos": 10707090531840.0,
      "grad_norm": 2.0078053802236484,
      "language_loss": 0.70068836,
      "learning_rate": 2.7462110432396095e-06,
      "loss": 0.72202599,
      "num_input_tokens_seen": 141804955,
      "step": 6602,
      "time_per_iteration": 2.5834643840789795
    },
    {
      "auxiliary_loss_clip": 0.01128624,
      "auxiliary_loss_mlp": 0.01041891,
      "balance_loss_clip": 1.04562747,
      "balance_loss_mlp": 1.02698421,
      "epoch": 0.39699383736660154,
      "flos": 17593714690560.0,
      "grad_norm": 2.4532961602339887,
      "language_loss": 0.83546066,
      "learning_rate": 2.7458496908630305e-06,
      "loss": 0.85716581,
      "num_input_tokens_seen": 141820025,
      "step": 6603,
      "time_per_iteration": 4.248618841171265
    },
    {
      "auxiliary_loss_clip": 0.01107549,
      "auxiliary_loss_mlp": 0.01033644,
      "balance_loss_clip": 1.04529059,
      "balance_loss_mlp": 1.01993501,
      "epoch": 0.3970539606192695,
      "flos": 17785945301760.0,
      "grad_norm": 1.6141120448730857,
      "language_loss": 0.72810054,
      "learning_rate": 2.7454883102037563e-06,
      "loss": 0.74951255,
      "num_input_tokens_seen": 141838735,
      "step": 6604,
      "time_per_iteration": 3.9230358600616455
    },
    {
      "auxiliary_loss_clip": 0.01102628,
      "auxiliary_loss_mlp": 0.01035991,
      "balance_loss_clip": 1.04502141,
      "balance_loss_mlp": 1.02143514,
      "epoch": 0.3971140838719375,
      "flos": 24789495208320.0,
      "grad_norm": 1.6406569976852292,
      "language_loss": 0.82818151,
      "learning_rate": 2.745126901275491e-06,
      "loss": 0.84956777,
      "num_input_tokens_seen": 141858090,
      "step": 6605,
      "time_per_iteration": 2.593484401702881
    },
    {
      "auxiliary_loss_clip": 0.01125267,
      "auxiliary_loss_mlp": 0.01029942,
      "balance_loss_clip": 1.04671991,
      "balance_loss_mlp": 1.01719868,
      "epoch": 0.39717420712460544,
      "flos": 24243581329920.0,
      "grad_norm": 1.5940660534516067,
      "language_loss": 0.73561233,
      "learning_rate": 2.7447654640919383e-06,
      "loss": 0.75716448,
      "num_input_tokens_seen": 141877540,
      "step": 6606,
      "time_per_iteration": 3.927874803543091
    },
    {
      "auxiliary_loss_clip": 0.01086922,
      "auxiliary_loss_mlp": 0.01038659,
      "balance_loss_clip": 1.04475546,
      "balance_loss_mlp": 1.02347136,
      "epoch": 0.3972343303772734,
      "flos": 25884698843520.0,
      "grad_norm": 1.85930687831078,
      "language_loss": 0.74577487,
      "learning_rate": 2.744403998666805e-06,
      "loss": 0.76703072,
      "num_input_tokens_seen": 141897315,
      "step": 6607,
      "time_per_iteration": 2.59767746925354
    },
    {
      "auxiliary_loss_clip": 0.01122286,
      "auxiliary_loss_mlp": 0.01035585,
      "balance_loss_clip": 1.04955935,
      "balance_loss_mlp": 1.02135718,
      "epoch": 0.39729445362994137,
      "flos": 45623716300800.0,
      "grad_norm": 1.5498516953767514,
      "language_loss": 0.68004096,
      "learning_rate": 2.744042505013797e-06,
      "loss": 0.70161963,
      "num_input_tokens_seen": 141919580,
      "step": 6608,
      "time_per_iteration": 2.716338634490967
    },
    {
      "auxiliary_loss_clip": 0.01091757,
      "auxiliary_loss_mlp": 0.01045145,
      "balance_loss_clip": 1.04283714,
      "balance_loss_mlp": 1.02780044,
      "epoch": 0.39735457688260933,
      "flos": 20193971120640.0,
      "grad_norm": 2.4867544766959124,
      "language_loss": 0.74320549,
      "learning_rate": 2.7436809831466233e-06,
      "loss": 0.76457453,
      "num_input_tokens_seen": 141937045,
      "step": 6609,
      "time_per_iteration": 2.5321836471557617
    },
    {
      "auxiliary_loss_clip": 0.01111712,
      "auxiliary_loss_mlp": 0.01036801,
      "balance_loss_clip": 1.04804707,
      "balance_loss_mlp": 1.02160144,
      "epoch": 0.3974147001352773,
      "flos": 23331163029120.0,
      "grad_norm": 6.192107592359399,
      "language_loss": 0.71242702,
      "learning_rate": 2.7433194330789927e-06,
      "loss": 0.73391217,
      "num_input_tokens_seen": 141956695,
      "step": 6610,
      "time_per_iteration": 2.553844928741455
    },
    {
      "auxiliary_loss_clip": 0.01105769,
      "auxiliary_loss_mlp": 0.01034005,
      "balance_loss_clip": 1.04215288,
      "balance_loss_mlp": 1.01962876,
      "epoch": 0.39747482338794526,
      "flos": 21688644885120.0,
      "grad_norm": 1.6029606325720518,
      "language_loss": 0.78463888,
      "learning_rate": 2.7429578548246133e-06,
      "loss": 0.80603659,
      "num_input_tokens_seen": 141975935,
      "step": 6611,
      "time_per_iteration": 2.5154647827148438
    },
    {
      "auxiliary_loss_clip": 0.01119294,
      "auxiliary_loss_mlp": 0.01036111,
      "balance_loss_clip": 1.04958093,
      "balance_loss_mlp": 1.02179372,
      "epoch": 0.3975349466406133,
      "flos": 30988717816320.0,
      "grad_norm": 1.8347262980687606,
      "language_loss": 0.79654008,
      "learning_rate": 2.7425962483971985e-06,
      "loss": 0.81809413,
      "num_input_tokens_seen": 141995750,
      "step": 6612,
      "time_per_iteration": 2.6291420459747314
    },
    {
      "auxiliary_loss_clip": 0.01025236,
      "auxiliary_loss_mlp": 0.01004546,
      "balance_loss_clip": 1.0296247,
      "balance_loss_mlp": 1.00285292,
      "epoch": 0.39759506989328125,
      "flos": 63683948833920.0,
      "grad_norm": 0.8464413285376963,
      "language_loss": 0.6498633,
      "learning_rate": 2.742234613810459e-06,
      "loss": 0.67016113,
      "num_input_tokens_seen": 142057655,
      "step": 6613,
      "time_per_iteration": 3.0712215900421143
    },
    {
      "auxiliary_loss_clip": 0.01099541,
      "auxiliary_loss_mlp": 0.01044115,
      "balance_loss_clip": 1.04417574,
      "balance_loss_mlp": 1.02717519,
      "epoch": 0.3976551931459492,
      "flos": 23695835857920.0,
      "grad_norm": 3.30456554549262,
      "language_loss": 0.71247774,
      "learning_rate": 2.741872951078109e-06,
      "loss": 0.73391432,
      "num_input_tokens_seen": 142076020,
      "step": 6614,
      "time_per_iteration": 2.543686866760254
    },
    {
      "auxiliary_loss_clip": 0.0111569,
      "auxiliary_loss_mlp": 0.01035089,
      "balance_loss_clip": 1.04952872,
      "balance_loss_mlp": 1.02104056,
      "epoch": 0.3977153163986172,
      "flos": 15669657745920.0,
      "grad_norm": 1.7697501916538,
      "language_loss": 0.81817067,
      "learning_rate": 2.741511260213862e-06,
      "loss": 0.83967847,
      "num_input_tokens_seen": 142093790,
      "step": 6615,
      "time_per_iteration": 2.4864516258239746
    },
    {
      "auxiliary_loss_clip": 0.01098813,
      "auxiliary_loss_mlp": 0.01029144,
      "balance_loss_clip": 1.04990256,
      "balance_loss_mlp": 1.01591182,
      "epoch": 0.39777543965128515,
      "flos": 14064702249600.0,
      "grad_norm": 2.5078641234160575,
      "language_loss": 0.67732096,
      "learning_rate": 2.741149541231434e-06,
      "loss": 0.69860047,
      "num_input_tokens_seen": 142110545,
      "step": 6616,
      "time_per_iteration": 2.525585412979126
    },
    {
      "auxiliary_loss_clip": 0.01131886,
      "auxiliary_loss_mlp": 0.01037383,
      "balance_loss_clip": 1.04799795,
      "balance_loss_mlp": 1.02299464,
      "epoch": 0.3978355629039531,
      "flos": 23367468700800.0,
      "grad_norm": 2.029452203840985,
      "language_loss": 0.8374964,
      "learning_rate": 2.740787794144541e-06,
      "loss": 0.85918915,
      "num_input_tokens_seen": 142128695,
      "step": 6617,
      "time_per_iteration": 2.466381311416626
    },
    {
      "auxiliary_loss_clip": 0.01126432,
      "auxiliary_loss_mlp": 0.01037081,
      "balance_loss_clip": 1.04967666,
      "balance_loss_mlp": 1.02425408,
      "epoch": 0.3978956861566211,
      "flos": 19062785036160.0,
      "grad_norm": 1.6111116008547097,
      "language_loss": 0.72197276,
      "learning_rate": 2.7404260189669e-06,
      "loss": 0.74360788,
      "num_input_tokens_seen": 142148375,
      "step": 6618,
      "time_per_iteration": 2.5062689781188965
    },
    {
      "auxiliary_loss_clip": 0.01107579,
      "auxiliary_loss_mlp": 0.01034935,
      "balance_loss_clip": 1.04884338,
      "balance_loss_mlp": 1.01967072,
      "epoch": 0.39795580940928904,
      "flos": 30227699341440.0,
      "grad_norm": 1.632886260079407,
      "language_loss": 0.65303481,
      "learning_rate": 2.740064215712231e-06,
      "loss": 0.67445999,
      "num_input_tokens_seen": 142169735,
      "step": 6619,
      "time_per_iteration": 2.6154842376708984
    },
    {
      "auxiliary_loss_clip": 0.01054887,
      "auxiliary_loss_mlp": 0.01009462,
      "balance_loss_clip": 1.02669108,
      "balance_loss_mlp": 1.00813866,
      "epoch": 0.398015932661957,
      "flos": 69847224906240.0,
      "grad_norm": 0.7777127318091575,
      "language_loss": 0.58234823,
      "learning_rate": 2.7397023843942527e-06,
      "loss": 0.6029917,
      "num_input_tokens_seen": 142229520,
      "step": 6620,
      "time_per_iteration": 3.0367910861968994
    },
    {
      "auxiliary_loss_clip": 0.01107987,
      "auxiliary_loss_mlp": 0.01034326,
      "balance_loss_clip": 1.04715371,
      "balance_loss_mlp": 1.0218451,
      "epoch": 0.39807605591462497,
      "flos": 20157773189760.0,
      "grad_norm": 1.6044551552121404,
      "language_loss": 0.79170251,
      "learning_rate": 2.739340525026686e-06,
      "loss": 0.81312567,
      "num_input_tokens_seen": 142247660,
      "step": 6621,
      "time_per_iteration": 2.5394749641418457
    },
    {
      "auxiliary_loss_clip": 0.01108624,
      "auxiliary_loss_mlp": 0.0103226,
      "balance_loss_clip": 1.04898262,
      "balance_loss_mlp": 1.019171,
      "epoch": 0.39813617916729294,
      "flos": 21141761339520.0,
      "grad_norm": 1.802739624070729,
      "language_loss": 0.78021365,
      "learning_rate": 2.738978637623252e-06,
      "loss": 0.80162251,
      "num_input_tokens_seen": 142266990,
      "step": 6622,
      "time_per_iteration": 2.5325136184692383
    },
    {
      "auxiliary_loss_clip": 0.01101331,
      "auxiliary_loss_mlp": 0.01036553,
      "balance_loss_clip": 1.04388988,
      "balance_loss_mlp": 1.02205133,
      "epoch": 0.3981963024199609,
      "flos": 18988485753600.0,
      "grad_norm": 1.637107829984215,
      "language_loss": 0.75154436,
      "learning_rate": 2.738616722197674e-06,
      "loss": 0.77292323,
      "num_input_tokens_seen": 142287170,
      "step": 6623,
      "time_per_iteration": 2.5320024490356445
    },
    {
      "auxiliary_loss_clip": 0.01092659,
      "auxiliary_loss_mlp": 0.01040065,
      "balance_loss_clip": 1.05171227,
      "balance_loss_mlp": 1.02545631,
      "epoch": 0.39825642567262887,
      "flos": 16575108808320.0,
      "grad_norm": 2.0211302244649385,
      "language_loss": 0.79544938,
      "learning_rate": 2.7382547787636766e-06,
      "loss": 0.81677657,
      "num_input_tokens_seen": 142305405,
      "step": 6624,
      "time_per_iteration": 2.5311384201049805
    },
    {
      "auxiliary_loss_clip": 0.0113496,
      "auxiliary_loss_mlp": 0.0104084,
      "balance_loss_clip": 1.05088842,
      "balance_loss_mlp": 1.02547979,
      "epoch": 0.39831654892529683,
      "flos": 22199833290240.0,
      "grad_norm": 2.316400697297772,
      "language_loss": 0.84092838,
      "learning_rate": 2.7378928073349832e-06,
      "loss": 0.8626864,
      "num_input_tokens_seen": 142322710,
      "step": 6625,
      "time_per_iteration": 2.499185800552368
    },
    {
      "auxiliary_loss_clip": 0.01114114,
      "auxiliary_loss_mlp": 0.01039827,
      "balance_loss_clip": 1.05009615,
      "balance_loss_mlp": 1.02539086,
      "epoch": 0.39837667217796485,
      "flos": 10487963612160.0,
      "grad_norm": 1.9888826588300066,
      "language_loss": 0.86251783,
      "learning_rate": 2.737530807925321e-06,
      "loss": 0.88405728,
      "num_input_tokens_seen": 142338535,
      "step": 6626,
      "time_per_iteration": 2.4958102703094482
    },
    {
      "auxiliary_loss_clip": 0.01067081,
      "auxiliary_loss_mlp": 0.00793199,
      "balance_loss_clip": 1.04346299,
      "balance_loss_mlp": 1.01055956,
      "epoch": 0.3984367954306328,
      "flos": 17965282930560.0,
      "grad_norm": 2.1274379717369185,
      "language_loss": 0.83276486,
      "learning_rate": 2.737168780548417e-06,
      "loss": 0.85136765,
      "num_input_tokens_seen": 142354570,
      "step": 6627,
      "time_per_iteration": 2.60463547706604
    },
    {
      "auxiliary_loss_clip": 0.01095945,
      "auxiliary_loss_mlp": 0.00790263,
      "balance_loss_clip": 1.04671407,
      "balance_loss_mlp": 1.01056337,
      "epoch": 0.3984969186833008,
      "flos": 22711057608960.0,
      "grad_norm": 1.6053460685971201,
      "language_loss": 0.82988405,
      "learning_rate": 2.736806725217998e-06,
      "loss": 0.84874612,
      "num_input_tokens_seen": 142374395,
      "step": 6628,
      "time_per_iteration": 2.5752501487731934
    },
    {
      "auxiliary_loss_clip": 0.01090211,
      "auxiliary_loss_mlp": 0.01056971,
      "balance_loss_clip": 1.04778075,
      "balance_loss_mlp": 1.04124713,
      "epoch": 0.39855704193596875,
      "flos": 23405785534080.0,
      "grad_norm": 1.7654908237788431,
      "language_loss": 0.710006,
      "learning_rate": 2.7364446419477945e-06,
      "loss": 0.73147786,
      "num_input_tokens_seen": 142396040,
      "step": 6629,
      "time_per_iteration": 2.587245225906372
    },
    {
      "auxiliary_loss_clip": 0.0109509,
      "auxiliary_loss_mlp": 0.01034134,
      "balance_loss_clip": 1.04905701,
      "balance_loss_mlp": 1.02032912,
      "epoch": 0.3986171651886367,
      "flos": 21251935330560.0,
      "grad_norm": 1.7780248170246171,
      "language_loss": 0.8064906,
      "learning_rate": 2.7360825307515366e-06,
      "loss": 0.82778287,
      "num_input_tokens_seen": 142415495,
      "step": 6630,
      "time_per_iteration": 2.58601713180542
    },
    {
      "auxiliary_loss_clip": 0.01072967,
      "auxiliary_loss_mlp": 0.01031763,
      "balance_loss_clip": 1.04997206,
      "balance_loss_mlp": 1.01732099,
      "epoch": 0.3986772884413047,
      "flos": 12458705258880.0,
      "grad_norm": 1.9129997828801952,
      "language_loss": 0.75246465,
      "learning_rate": 2.7357203916429555e-06,
      "loss": 0.77351195,
      "num_input_tokens_seen": 142431865,
      "step": 6631,
      "time_per_iteration": 2.618156671524048
    },
    {
      "auxiliary_loss_clip": 0.01095693,
      "auxiliary_loss_mlp": 0.0103848,
      "balance_loss_clip": 1.04742503,
      "balance_loss_mlp": 1.02382922,
      "epoch": 0.39873741169397264,
      "flos": 19646117907840.0,
      "grad_norm": 1.7408786044593307,
      "language_loss": 0.71109152,
      "learning_rate": 2.735358224635783e-06,
      "loss": 0.7324332,
      "num_input_tokens_seen": 142450595,
      "step": 6632,
      "time_per_iteration": 2.5608553886413574
    },
    {
      "auxiliary_loss_clip": 0.01063012,
      "auxiliary_loss_mlp": 0.00789224,
      "balance_loss_clip": 1.04863024,
      "balance_loss_mlp": 1.01107264,
      "epoch": 0.3987975349466406,
      "flos": 21684766216320.0,
      "grad_norm": 1.9115777860144054,
      "language_loss": 0.75293374,
      "learning_rate": 2.7349960297437533e-06,
      "loss": 0.77145612,
      "num_input_tokens_seen": 142466650,
      "step": 6633,
      "time_per_iteration": 2.6329824924468994
    },
    {
      "auxiliary_loss_clip": 0.01104516,
      "auxiliary_loss_mlp": 0.01031637,
      "balance_loss_clip": 1.04890907,
      "balance_loss_mlp": 1.01764143,
      "epoch": 0.3988576581993086,
      "flos": 23914064937600.0,
      "grad_norm": 1.8130821436292546,
      "language_loss": 0.81309015,
      "learning_rate": 2.7346338069806e-06,
      "loss": 0.83445168,
      "num_input_tokens_seen": 142486165,
      "step": 6634,
      "time_per_iteration": 2.563260555267334
    },
    {
      "auxiliary_loss_clip": 0.01099629,
      "auxiliary_loss_mlp": 0.01033451,
      "balance_loss_clip": 1.04691696,
      "balance_loss_mlp": 1.01908624,
      "epoch": 0.39891778145197654,
      "flos": 18149899858560.0,
      "grad_norm": 1.930823353288653,
      "language_loss": 0.74485391,
      "learning_rate": 2.7342715563600597e-06,
      "loss": 0.76618475,
      "num_input_tokens_seen": 142505035,
      "step": 6635,
      "time_per_iteration": 2.5654022693634033
    },
    {
      "auxiliary_loss_clip": 0.01102222,
      "auxiliary_loss_mlp": 0.01045563,
      "balance_loss_clip": 1.04787993,
      "balance_loss_mlp": 1.03000665,
      "epoch": 0.3989779047046445,
      "flos": 22595281096320.0,
      "grad_norm": 2.25627492708286,
      "language_loss": 0.66223752,
      "learning_rate": 2.733909277895868e-06,
      "loss": 0.6837154,
      "num_input_tokens_seen": 142521870,
      "step": 6636,
      "time_per_iteration": 2.5659472942352295
    },
    {
      "auxiliary_loss_clip": 0.01117475,
      "auxiliary_loss_mlp": 0.01036413,
      "balance_loss_clip": 1.04848051,
      "balance_loss_mlp": 1.02227485,
      "epoch": 0.39903802795731247,
      "flos": 18077216688000.0,
      "grad_norm": 1.824056196772896,
      "language_loss": 0.8126514,
      "learning_rate": 2.733546971601763e-06,
      "loss": 0.83419025,
      "num_input_tokens_seen": 142540455,
      "step": 6637,
      "time_per_iteration": 2.51161789894104
    },
    {
      "auxiliary_loss_clip": 0.01025974,
      "auxiliary_loss_mlp": 0.01009067,
      "balance_loss_clip": 1.02704072,
      "balance_loss_mlp": 1.00719535,
      "epoch": 0.39909815120998043,
      "flos": 70441367771520.0,
      "grad_norm": 0.721428453646231,
      "language_loss": 0.53183389,
      "learning_rate": 2.733184637491484e-06,
      "loss": 0.55218434,
      "num_input_tokens_seen": 142599665,
      "step": 6638,
      "time_per_iteration": 3.2490475177764893
    },
    {
      "auxiliary_loss_clip": 0.0110783,
      "auxiliary_loss_mlp": 0.00789739,
      "balance_loss_clip": 1.04790819,
      "balance_loss_mlp": 1.01199317,
      "epoch": 0.39915827446264845,
      "flos": 18549262247040.0,
      "grad_norm": 1.5202002055968988,
      "language_loss": 0.75616956,
      "learning_rate": 2.732822275578769e-06,
      "loss": 0.77514529,
      "num_input_tokens_seen": 142618845,
      "step": 6639,
      "time_per_iteration": 2.542187452316284
    },
    {
      "auxiliary_loss_clip": 0.01057393,
      "auxiliary_loss_mlp": 0.010338,
      "balance_loss_clip": 1.0460577,
      "balance_loss_mlp": 1.02069283,
      "epoch": 0.3992183977153164,
      "flos": 29897249195520.0,
      "grad_norm": 1.563687424742328,
      "language_loss": 0.75875157,
      "learning_rate": 2.7324598858773603e-06,
      "loss": 0.77966344,
      "num_input_tokens_seen": 142640885,
      "step": 6640,
      "time_per_iteration": 4.105189323425293
    },
    {
      "auxiliary_loss_clip": 0.01096051,
      "auxiliary_loss_mlp": 0.01033576,
      "balance_loss_clip": 1.04523253,
      "balance_loss_mlp": 1.01966429,
      "epoch": 0.3992785209679844,
      "flos": 22565080736640.0,
      "grad_norm": 2.390954165778915,
      "language_loss": 0.82682025,
      "learning_rate": 2.7320974684009996e-06,
      "loss": 0.84811646,
      "num_input_tokens_seen": 142659340,
      "step": 6641,
      "time_per_iteration": 3.923123359680176
    },
    {
      "auxiliary_loss_clip": 0.01129286,
      "auxiliary_loss_mlp": 0.01031285,
      "balance_loss_clip": 1.04845643,
      "balance_loss_mlp": 1.01745081,
      "epoch": 0.39933864422065235,
      "flos": 19682674974720.0,
      "grad_norm": 2.17654224323603,
      "language_loss": 0.77134776,
      "learning_rate": 2.7317350231634288e-06,
      "loss": 0.79295349,
      "num_input_tokens_seen": 142677085,
      "step": 6642,
      "time_per_iteration": 2.4789516925811768
    },
    {
      "auxiliary_loss_clip": 0.01100885,
      "auxiliary_loss_mlp": 0.01032498,
      "balance_loss_clip": 1.04670358,
      "balance_loss_mlp": 1.01794887,
      "epoch": 0.3993987674733203,
      "flos": 23038491012480.0,
      "grad_norm": 2.0657264196938545,
      "language_loss": 0.72251928,
      "learning_rate": 2.731372550178393e-06,
      "loss": 0.74385315,
      "num_input_tokens_seen": 142694595,
      "step": 6643,
      "time_per_iteration": 4.0063183307647705
    },
    {
      "auxiliary_loss_clip": 0.01119271,
      "auxiliary_loss_mlp": 0.01031487,
      "balance_loss_clip": 1.04842222,
      "balance_loss_mlp": 1.01768851,
      "epoch": 0.3994588907259883,
      "flos": 19390828970880.0,
      "grad_norm": 1.5925859316634123,
      "language_loss": 0.66704696,
      "learning_rate": 2.7310100494596375e-06,
      "loss": 0.68855453,
      "num_input_tokens_seen": 142714175,
      "step": 6644,
      "time_per_iteration": 2.5068705081939697
    },
    {
      "auxiliary_loss_clip": 0.01127084,
      "auxiliary_loss_mlp": 0.01035394,
      "balance_loss_clip": 1.04615211,
      "balance_loss_mlp": 1.0211308,
      "epoch": 0.39951901397865625,
      "flos": 13734395758080.0,
      "grad_norm": 2.30105663831236,
      "language_loss": 0.78360188,
      "learning_rate": 2.730647521020907e-06,
      "loss": 0.80522662,
      "num_input_tokens_seen": 142730955,
      "step": 6645,
      "time_per_iteration": 3.8158793449401855
    },
    {
      "auxiliary_loss_clip": 0.01117545,
      "auxiliary_loss_mlp": 0.01031934,
      "balance_loss_clip": 1.04653192,
      "balance_loss_mlp": 1.01823735,
      "epoch": 0.3995791372313242,
      "flos": 23586451966080.0,
      "grad_norm": 1.556429180877195,
      "language_loss": 0.69869637,
      "learning_rate": 2.73028496487595e-06,
      "loss": 0.72019124,
      "num_input_tokens_seen": 142751200,
      "step": 6646,
      "time_per_iteration": 2.508810520172119
    },
    {
      "auxiliary_loss_clip": 0.01076378,
      "auxiliary_loss_mlp": 0.01035021,
      "balance_loss_clip": 1.04312801,
      "balance_loss_mlp": 1.02056074,
      "epoch": 0.3996392604839922,
      "flos": 21355896268800.0,
      "grad_norm": 1.9722980296152277,
      "language_loss": 0.71553195,
      "learning_rate": 2.729922381038513e-06,
      "loss": 0.73664594,
      "num_input_tokens_seen": 142770170,
      "step": 6647,
      "time_per_iteration": 2.5724337100982666
    },
    {
      "auxiliary_loss_clip": 0.01085415,
      "auxiliary_loss_mlp": 0.01031429,
      "balance_loss_clip": 1.04447556,
      "balance_loss_mlp": 1.0191381,
      "epoch": 0.39969938373666014,
      "flos": 26032255914240.0,
      "grad_norm": 1.5214757912924397,
      "language_loss": 0.74538934,
      "learning_rate": 2.7295597695223463e-06,
      "loss": 0.76655781,
      "num_input_tokens_seen": 142792680,
      "step": 6648,
      "time_per_iteration": 2.6204991340637207
    },
    {
      "auxiliary_loss_clip": 0.0112774,
      "auxiliary_loss_mlp": 0.01028323,
      "balance_loss_clip": 1.04716873,
      "balance_loss_mlp": 1.01410127,
      "epoch": 0.3997595069893281,
      "flos": 20116367786880.0,
      "grad_norm": 2.100429261190177,
      "language_loss": 0.65788472,
      "learning_rate": 2.7291971303412006e-06,
      "loss": 0.67944527,
      "num_input_tokens_seen": 142810510,
      "step": 6649,
      "time_per_iteration": 2.4527840614318848
    },
    {
      "auxiliary_loss_clip": 0.01101084,
      "auxiliary_loss_mlp": 0.01040057,
      "balance_loss_clip": 1.0485698,
      "balance_loss_mlp": 1.0261035,
      "epoch": 0.39981963024199607,
      "flos": 27783403764480.0,
      "grad_norm": 1.6611266173586952,
      "language_loss": 0.75649244,
      "learning_rate": 2.728834463508826e-06,
      "loss": 0.77790391,
      "num_input_tokens_seen": 142832455,
      "step": 6650,
      "time_per_iteration": 2.6117494106292725
    },
    {
      "auxiliary_loss_clip": 0.01127696,
      "auxiliary_loss_mlp": 0.01042065,
      "balance_loss_clip": 1.0472616,
      "balance_loss_mlp": 1.02879727,
      "epoch": 0.39987975349466404,
      "flos": 21944436612480.0,
      "grad_norm": 1.7480518174875843,
      "language_loss": 0.71841586,
      "learning_rate": 2.728471769038975e-06,
      "loss": 0.7401135,
      "num_input_tokens_seen": 142852590,
      "step": 6651,
      "time_per_iteration": 2.481882333755493
    },
    {
      "auxiliary_loss_clip": 0.01128301,
      "auxiliary_loss_mlp": 0.01035813,
      "balance_loss_clip": 1.0468452,
      "balance_loss_mlp": 1.0219605,
      "epoch": 0.39993987674733206,
      "flos": 20704405340160.0,
      "grad_norm": 1.8892781906831801,
      "language_loss": 0.73252457,
      "learning_rate": 2.728109046945403e-06,
      "loss": 0.75416571,
      "num_input_tokens_seen": 142870595,
      "step": 6652,
      "time_per_iteration": 2.4900617599487305
    },
    {
      "auxiliary_loss_clip": 0.01026725,
      "auxiliary_loss_mlp": 0.01003323,
      "balance_loss_clip": 1.02907503,
      "balance_loss_mlp": 1.0018332,
      "epoch": 0.4,
      "flos": 61525429862400.0,
      "grad_norm": 0.8481928237157531,
      "language_loss": 0.60631037,
      "learning_rate": 2.727746297241862e-06,
      "loss": 0.62661088,
      "num_input_tokens_seen": 142925805,
      "step": 6653,
      "time_per_iteration": 3.066547393798828
    },
    {
      "auxiliary_loss_clip": 0.01088531,
      "auxiliary_loss_mlp": 0.01035362,
      "balance_loss_clip": 1.04932237,
      "balance_loss_mlp": 1.02262449,
      "epoch": 0.400060123252668,
      "flos": 14502309644160.0,
      "grad_norm": 2.465504993373776,
      "language_loss": 0.66926008,
      "learning_rate": 2.7273835199421085e-06,
      "loss": 0.69049895,
      "num_input_tokens_seen": 142943145,
      "step": 6654,
      "time_per_iteration": 2.5468766689300537
    },
    {
      "auxiliary_loss_clip": 0.01113883,
      "auxiliary_loss_mlp": 0.01035237,
      "balance_loss_clip": 1.04675174,
      "balance_loss_mlp": 1.02359581,
      "epoch": 0.40012024650533595,
      "flos": 19093308618240.0,
      "grad_norm": 2.854343328058018,
      "language_loss": 0.89911586,
      "learning_rate": 2.7270207150599e-06,
      "loss": 0.92060709,
      "num_input_tokens_seen": 142956925,
      "step": 6655,
      "time_per_iteration": 2.486863136291504
    },
    {
      "auxiliary_loss_clip": 0.01096654,
      "auxiliary_loss_mlp": 0.01031031,
      "balance_loss_clip": 1.04450977,
      "balance_loss_mlp": 1.01835895,
      "epoch": 0.4001803697580039,
      "flos": 29351012094720.0,
      "grad_norm": 1.656922451684095,
      "language_loss": 0.7349906,
      "learning_rate": 2.7266578826089917e-06,
      "loss": 0.75626743,
      "num_input_tokens_seen": 142978040,
      "step": 6656,
      "time_per_iteration": 2.6198902130126953
    },
    {
      "auxiliary_loss_clip": 0.01126906,
      "auxiliary_loss_mlp": 0.01040198,
      "balance_loss_clip": 1.04742432,
      "balance_loss_mlp": 1.02626848,
      "epoch": 0.4002404930106719,
      "flos": 20920048640640.0,
      "grad_norm": 1.680316023502735,
      "language_loss": 0.73478431,
      "learning_rate": 2.726295022603144e-06,
      "loss": 0.7564553,
      "num_input_tokens_seen": 142998390,
      "step": 6657,
      "time_per_iteration": 2.539036273956299
    },
    {
      "auxiliary_loss_clip": 0.01130175,
      "auxiliary_loss_mlp": 0.01034316,
      "balance_loss_clip": 1.05023384,
      "balance_loss_mlp": 1.02029729,
      "epoch": 0.40030061626333985,
      "flos": 28405735827840.0,
      "grad_norm": 1.5045704193256217,
      "language_loss": 0.79794097,
      "learning_rate": 2.725932135056117e-06,
      "loss": 0.81958586,
      "num_input_tokens_seen": 143021505,
      "step": 6658,
      "time_per_iteration": 2.5552303791046143
    },
    {
      "auxiliary_loss_clip": 0.01114553,
      "auxiliary_loss_mlp": 0.01038383,
      "balance_loss_clip": 1.04621208,
      "balance_loss_mlp": 1.02533007,
      "epoch": 0.4003607395160078,
      "flos": 25921615046400.0,
      "grad_norm": 1.9523935863773605,
      "language_loss": 0.77504581,
      "learning_rate": 2.72556921998167e-06,
      "loss": 0.79657519,
      "num_input_tokens_seen": 143041375,
      "step": 6659,
      "time_per_iteration": 2.566817045211792
    },
    {
      "auxiliary_loss_clip": 0.0111742,
      "auxiliary_loss_mlp": 0.01026734,
      "balance_loss_clip": 1.04552305,
      "balance_loss_mlp": 1.01546288,
      "epoch": 0.4004208627686758,
      "flos": 20768648814720.0,
      "grad_norm": 1.7763890227605263,
      "language_loss": 0.72843397,
      "learning_rate": 2.7252062773935662e-06,
      "loss": 0.74987555,
      "num_input_tokens_seen": 143058725,
      "step": 6660,
      "time_per_iteration": 2.476580858230591
    },
    {
      "auxiliary_loss_clip": 0.01098,
      "auxiliary_loss_mlp": 0.01036208,
      "balance_loss_clip": 1.04430938,
      "balance_loss_mlp": 1.02397108,
      "epoch": 0.40048098602134374,
      "flos": 24681224638080.0,
      "grad_norm": 1.738986696764367,
      "language_loss": 0.71276963,
      "learning_rate": 2.7248433073055674e-06,
      "loss": 0.73411167,
      "num_input_tokens_seen": 143076995,
      "step": 6661,
      "time_per_iteration": 2.5720791816711426
    },
    {
      "auxiliary_loss_clip": 0.01129518,
      "auxiliary_loss_mlp": 0.01039377,
      "balance_loss_clip": 1.05025542,
      "balance_loss_mlp": 1.02677667,
      "epoch": 0.4005411092740117,
      "flos": 23185688947200.0,
      "grad_norm": 1.702554894081864,
      "language_loss": 0.75780714,
      "learning_rate": 2.724480309731437e-06,
      "loss": 0.77949607,
      "num_input_tokens_seen": 143096780,
      "step": 6662,
      "time_per_iteration": 2.4837634563446045
    },
    {
      "auxiliary_loss_clip": 0.01111596,
      "auxiliary_loss_mlp": 0.01032609,
      "balance_loss_clip": 1.04482651,
      "balance_loss_mlp": 1.01884019,
      "epoch": 0.4006012325266797,
      "flos": 17522324409600.0,
      "grad_norm": 1.9969985775134622,
      "language_loss": 0.66016865,
      "learning_rate": 2.7241172846849417e-06,
      "loss": 0.68161064,
      "num_input_tokens_seen": 143112590,
      "step": 6663,
      "time_per_iteration": 2.484163284301758
    },
    {
      "auxiliary_loss_clip": 0.01108828,
      "auxiliary_loss_mlp": 0.01041079,
      "balance_loss_clip": 1.04423165,
      "balance_loss_mlp": 1.02710128,
      "epoch": 0.40066135577934764,
      "flos": 19857200181120.0,
      "grad_norm": 2.0221134656796305,
      "language_loss": 0.85540855,
      "learning_rate": 2.7237542321798455e-06,
      "loss": 0.87690759,
      "num_input_tokens_seen": 143130220,
      "step": 6664,
      "time_per_iteration": 2.513077735900879
    },
    {
      "auxiliary_loss_clip": 0.01114753,
      "auxiliary_loss_mlp": 0.01034576,
      "balance_loss_clip": 1.04794908,
      "balance_loss_mlp": 1.02190399,
      "epoch": 0.40072147903201566,
      "flos": 18150007599360.0,
      "grad_norm": 2.124279817821128,
      "language_loss": 0.8484726,
      "learning_rate": 2.723391152229917e-06,
      "loss": 0.86996591,
      "num_input_tokens_seen": 143147160,
      "step": 6665,
      "time_per_iteration": 2.4638185501098633
    },
    {
      "auxiliary_loss_clip": 0.01114628,
      "auxiliary_loss_mlp": 0.0103281,
      "balance_loss_clip": 1.04943013,
      "balance_loss_mlp": 1.01926184,
      "epoch": 0.4007816022846836,
      "flos": 18661267831680.0,
      "grad_norm": 1.9456708311567268,
      "language_loss": 0.78443098,
      "learning_rate": 2.7230280448489236e-06,
      "loss": 0.80590534,
      "num_input_tokens_seen": 143164605,
      "step": 6666,
      "time_per_iteration": 2.4794859886169434
    },
    {
      "auxiliary_loss_clip": 0.01116977,
      "auxiliary_loss_mlp": 0.01040542,
      "balance_loss_clip": 1.04889858,
      "balance_loss_mlp": 1.02686846,
      "epoch": 0.4008417255373516,
      "flos": 25703170485120.0,
      "grad_norm": 1.7880495967244696,
      "language_loss": 0.73444456,
      "learning_rate": 2.7226649100506333e-06,
      "loss": 0.75601977,
      "num_input_tokens_seen": 143183965,
      "step": 6667,
      "time_per_iteration": 2.542250156402588
    },
    {
      "auxiliary_loss_clip": 0.01110169,
      "auxiliary_loss_mlp": 0.01048553,
      "balance_loss_clip": 1.04596782,
      "balance_loss_mlp": 1.03387821,
      "epoch": 0.40090184879001955,
      "flos": 22858614679680.0,
      "grad_norm": 1.6555309638965752,
      "language_loss": 0.75604868,
      "learning_rate": 2.7223017478488183e-06,
      "loss": 0.77763587,
      "num_input_tokens_seen": 143204965,
      "step": 6668,
      "time_per_iteration": 2.5267891883850098
    },
    {
      "auxiliary_loss_clip": 0.01091894,
      "auxiliary_loss_mlp": 0.01035067,
      "balance_loss_clip": 1.04902732,
      "balance_loss_mlp": 1.02160811,
      "epoch": 0.4009619720426875,
      "flos": 29059848449280.0,
      "grad_norm": 1.901452942913847,
      "language_loss": 0.82127237,
      "learning_rate": 2.721938558257248e-06,
      "loss": 0.84254199,
      "num_input_tokens_seen": 143225015,
      "step": 6669,
      "time_per_iteration": 2.632352113723755
    },
    {
      "auxiliary_loss_clip": 0.01031971,
      "auxiliary_loss_mlp": 0.01005789,
      "balance_loss_clip": 1.02105236,
      "balance_loss_mlp": 1.00416756,
      "epoch": 0.4010220952953555,
      "flos": 66059763131520.0,
      "grad_norm": 0.7111326349691449,
      "language_loss": 0.5331794,
      "learning_rate": 2.721575341289695e-06,
      "loss": 0.55355698,
      "num_input_tokens_seen": 143294925,
      "step": 6670,
      "time_per_iteration": 3.269810914993286
    },
    {
      "auxiliary_loss_clip": 0.01079282,
      "auxiliary_loss_mlp": 0.01034661,
      "balance_loss_clip": 1.04974174,
      "balance_loss_mlp": 1.02158356,
      "epoch": 0.40108221854802345,
      "flos": 29642822184960.0,
      "grad_norm": 1.5959015082767698,
      "language_loss": 0.8900888,
      "learning_rate": 2.7212120969599333e-06,
      "loss": 0.9112283,
      "num_input_tokens_seen": 143314170,
      "step": 6671,
      "time_per_iteration": 2.651196002960205
    },
    {
      "auxiliary_loss_clip": 0.01114183,
      "auxiliary_loss_mlp": 0.01033553,
      "balance_loss_clip": 1.04528928,
      "balance_loss_mlp": 1.01948678,
      "epoch": 0.4011423418006914,
      "flos": 19929560129280.0,
      "grad_norm": 1.9201743495944381,
      "language_loss": 0.78964794,
      "learning_rate": 2.720848825281736e-06,
      "loss": 0.81112522,
      "num_input_tokens_seen": 143330050,
      "step": 6672,
      "time_per_iteration": 2.4958817958831787
    },
    {
      "auxiliary_loss_clip": 0.01085274,
      "auxiliary_loss_mlp": 0.01033576,
      "balance_loss_clip": 1.04131913,
      "balance_loss_mlp": 1.01896715,
      "epoch": 0.4012024650533594,
      "flos": 20084299920000.0,
      "grad_norm": 4.39523162939733,
      "language_loss": 0.63255763,
      "learning_rate": 2.72048552626888e-06,
      "loss": 0.65374607,
      "num_input_tokens_seen": 143348650,
      "step": 6673,
      "time_per_iteration": 2.554178476333618
    },
    {
      "auxiliary_loss_clip": 0.01101431,
      "auxiliary_loss_mlp": 0.00795764,
      "balance_loss_clip": 1.04613745,
      "balance_loss_mlp": 1.02548778,
      "epoch": 0.40126258830602735,
      "flos": 21695719864320.0,
      "grad_norm": 1.4396334063754586,
      "language_loss": 0.80240446,
      "learning_rate": 2.7201221999351402e-06,
      "loss": 0.82137644,
      "num_input_tokens_seen": 143370275,
      "step": 6674,
      "time_per_iteration": 2.590885639190674
    },
    {
      "auxiliary_loss_clip": 0.01086393,
      "auxiliary_loss_mlp": 0.01035026,
      "balance_loss_clip": 1.05306005,
      "balance_loss_mlp": 1.02132273,
      "epoch": 0.4013227115586953,
      "flos": 12020379592320.0,
      "grad_norm": 2.152765152486263,
      "language_loss": 0.81931967,
      "learning_rate": 2.719758846294294e-06,
      "loss": 0.84053385,
      "num_input_tokens_seen": 143385390,
      "step": 6675,
      "time_per_iteration": 2.6176342964172363
    },
    {
      "auxiliary_loss_clip": 0.01114885,
      "auxiliary_loss_mlp": 0.01035039,
      "balance_loss_clip": 1.04682112,
      "balance_loss_mlp": 1.02055562,
      "epoch": 0.4013828348113633,
      "flos": 25447522412160.0,
      "grad_norm": 1.746885843783589,
      "language_loss": 0.93411744,
      "learning_rate": 2.71939546536012e-06,
      "loss": 0.95561671,
      "num_input_tokens_seen": 143404215,
      "step": 6676,
      "time_per_iteration": 2.5524468421936035
    },
    {
      "auxiliary_loss_clip": 0.01119761,
      "auxiliary_loss_mlp": 0.01041345,
      "balance_loss_clip": 1.04762864,
      "balance_loss_mlp": 1.02639055,
      "epoch": 0.40144295806403124,
      "flos": 18582946225920.0,
      "grad_norm": 1.8537021691188993,
      "language_loss": 0.79028338,
      "learning_rate": 2.719032057146399e-06,
      "loss": 0.81189442,
      "num_input_tokens_seen": 143422245,
      "step": 6677,
      "time_per_iteration": 2.5016286373138428
    },
    {
      "auxiliary_loss_clip": 0.01107175,
      "auxiliary_loss_mlp": 0.01032262,
      "balance_loss_clip": 1.05084121,
      "balance_loss_mlp": 1.0190413,
      "epoch": 0.4015030813166992,
      "flos": 22930220442240.0,
      "grad_norm": 1.8536075663240672,
      "language_loss": 0.83936155,
      "learning_rate": 2.71866862166691e-06,
      "loss": 0.86075592,
      "num_input_tokens_seen": 143443130,
      "step": 6678,
      "time_per_iteration": 3.911740779876709
    },
    {
      "auxiliary_loss_clip": 0.0112302,
      "auxiliary_loss_mlp": 0.01036531,
      "balance_loss_clip": 1.04641283,
      "balance_loss_mlp": 1.02282774,
      "epoch": 0.4015632045693672,
      "flos": 20595057361920.0,
      "grad_norm": 2.337987728124336,
      "language_loss": 0.63655138,
      "learning_rate": 2.718305158935434e-06,
      "loss": 0.65814692,
      "num_input_tokens_seen": 143461385,
      "step": 6679,
      "time_per_iteration": 2.447779893875122
    },
    {
      "auxiliary_loss_clip": 0.01097038,
      "auxiliary_loss_mlp": 0.01029064,
      "balance_loss_clip": 1.04537344,
      "balance_loss_mlp": 1.01633835,
      "epoch": 0.4016233278220352,
      "flos": 23438930808960.0,
      "grad_norm": 1.4845357228518856,
      "language_loss": 0.78829908,
      "learning_rate": 2.7179416689657554e-06,
      "loss": 0.80956006,
      "num_input_tokens_seen": 143481750,
      "step": 6680,
      "time_per_iteration": 3.912682294845581
    },
    {
      "auxiliary_loss_clip": 0.01094787,
      "auxiliary_loss_mlp": 0.00793894,
      "balance_loss_clip": 1.04841399,
      "balance_loss_mlp": 1.01907921,
      "epoch": 0.40168345107470316,
      "flos": 21431057477760.0,
      "grad_norm": 1.5435336930820476,
      "language_loss": 0.75819141,
      "learning_rate": 2.7175781517716556e-06,
      "loss": 0.77707827,
      "num_input_tokens_seen": 143501540,
      "step": 6681,
      "time_per_iteration": 4.048239469528198
    },
    {
      "auxiliary_loss_clip": 0.01087932,
      "auxiliary_loss_mlp": 0.01029983,
      "balance_loss_clip": 1.04576206,
      "balance_loss_mlp": 1.01685858,
      "epoch": 0.4017435743273711,
      "flos": 22857214049280.0,
      "grad_norm": 1.7314068070559012,
      "language_loss": 0.6397624,
      "learning_rate": 2.7172146073669213e-06,
      "loss": 0.6609416,
      "num_input_tokens_seen": 143520530,
      "step": 6682,
      "time_per_iteration": 2.590996503829956
    },
    {
      "auxiliary_loss_clip": 0.01079986,
      "auxiliary_loss_mlp": 0.01031982,
      "balance_loss_clip": 1.04313564,
      "balance_loss_mlp": 1.01877987,
      "epoch": 0.4018036975800391,
      "flos": 28622312881920.0,
      "grad_norm": 1.7006235455424963,
      "language_loss": 0.72895646,
      "learning_rate": 2.716851035765337e-06,
      "loss": 0.75007612,
      "num_input_tokens_seen": 143540210,
      "step": 6683,
      "time_per_iteration": 4.073796272277832
    },
    {
      "auxiliary_loss_clip": 0.0111102,
      "auxiliary_loss_mlp": 0.01039056,
      "balance_loss_clip": 1.04591417,
      "balance_loss_mlp": 1.02593088,
      "epoch": 0.40186382083270705,
      "flos": 26651212099200.0,
      "grad_norm": 1.5875342191165946,
      "language_loss": 0.73344898,
      "learning_rate": 2.7164874369806896e-06,
      "loss": 0.75494981,
      "num_input_tokens_seen": 143560940,
      "step": 6684,
      "time_per_iteration": 2.556485176086426
    },
    {
      "auxiliary_loss_clip": 0.01043771,
      "auxiliary_loss_mlp": 0.01004956,
      "balance_loss_clip": 1.02355623,
      "balance_loss_mlp": 1.00361466,
      "epoch": 0.401923944085375,
      "flos": 59259969123840.0,
      "grad_norm": 0.8079764037969926,
      "language_loss": 0.60413003,
      "learning_rate": 2.716123811026767e-06,
      "loss": 0.62461734,
      "num_input_tokens_seen": 143624015,
      "step": 6685,
      "time_per_iteration": 3.212909460067749
    },
    {
      "auxiliary_loss_clip": 0.01120327,
      "auxiliary_loss_mlp": 0.01029935,
      "balance_loss_clip": 1.046556,
      "balance_loss_mlp": 1.01602316,
      "epoch": 0.401984067338043,
      "flos": 16982803152000.0,
      "grad_norm": 1.6905805280646442,
      "language_loss": 0.70302176,
      "learning_rate": 2.715760157917357e-06,
      "loss": 0.72452444,
      "num_input_tokens_seen": 143642750,
      "step": 6686,
      "time_per_iteration": 2.490316867828369
    },
    {
      "auxiliary_loss_clip": 0.0110037,
      "auxiliary_loss_mlp": 0.01031434,
      "balance_loss_clip": 1.04465389,
      "balance_loss_mlp": 1.01880956,
      "epoch": 0.40204419059071095,
      "flos": 24972496024320.0,
      "grad_norm": 1.556826139001412,
      "language_loss": 0.74902683,
      "learning_rate": 2.7153964776662504e-06,
      "loss": 0.77034491,
      "num_input_tokens_seen": 143664515,
      "step": 6687,
      "time_per_iteration": 2.5623998641967773
    },
    {
      "auxiliary_loss_clip": 0.01103458,
      "auxiliary_loss_mlp": 0.01030911,
      "balance_loss_clip": 1.0470593,
      "balance_loss_mlp": 1.01720834,
      "epoch": 0.4021043138433789,
      "flos": 23477463123840.0,
      "grad_norm": 2.1519694075367375,
      "language_loss": 0.71244454,
      "learning_rate": 2.7150327702872385e-06,
      "loss": 0.73378825,
      "num_input_tokens_seen": 143683135,
      "step": 6688,
      "time_per_iteration": 2.584134340286255
    },
    {
      "auxiliary_loss_clip": 0.01099979,
      "auxiliary_loss_mlp": 0.01042807,
      "balance_loss_clip": 1.04347336,
      "balance_loss_mlp": 1.02859104,
      "epoch": 0.4021644370960469,
      "flos": 25995806588160.0,
      "grad_norm": 1.6790823630555596,
      "language_loss": 0.64541447,
      "learning_rate": 2.7146690357941112e-06,
      "loss": 0.66684234,
      "num_input_tokens_seen": 143703985,
      "step": 6689,
      "time_per_iteration": 2.590778112411499
    },
    {
      "auxiliary_loss_clip": 0.01115062,
      "auxiliary_loss_mlp": 0.01026854,
      "balance_loss_clip": 1.04477024,
      "balance_loss_mlp": 1.01346636,
      "epoch": 0.40222456034871484,
      "flos": 13587987922560.0,
      "grad_norm": 2.0939155620714787,
      "language_loss": 0.73322725,
      "learning_rate": 2.7143052742006632e-06,
      "loss": 0.7546463,
      "num_input_tokens_seen": 143719245,
      "step": 6690,
      "time_per_iteration": 2.4778342247009277
    },
    {
      "auxiliary_loss_clip": 0.01089535,
      "auxiliary_loss_mlp": 0.01029796,
      "balance_loss_clip": 1.04163349,
      "balance_loss_mlp": 1.01574087,
      "epoch": 0.4022846836013828,
      "flos": 24278019494400.0,
      "grad_norm": 1.5115613565244943,
      "language_loss": 0.74870718,
      "learning_rate": 2.7139414855206872e-06,
      "loss": 0.76990044,
      "num_input_tokens_seen": 143739575,
      "step": 6691,
      "time_per_iteration": 2.5884199142456055
    },
    {
      "auxiliary_loss_clip": 0.01106243,
      "auxiliary_loss_mlp": 0.01034103,
      "balance_loss_clip": 1.04920673,
      "balance_loss_mlp": 1.02095962,
      "epoch": 0.40234480685405083,
      "flos": 20151596050560.0,
      "grad_norm": 1.6671439014002902,
      "language_loss": 0.72618699,
      "learning_rate": 2.7135776697679785e-06,
      "loss": 0.74759048,
      "num_input_tokens_seen": 143758515,
      "step": 6692,
      "time_per_iteration": 2.561613082885742
    },
    {
      "auxiliary_loss_clip": 0.01074646,
      "auxiliary_loss_mlp": 0.01032297,
      "balance_loss_clip": 1.04401493,
      "balance_loss_mlp": 1.01944613,
      "epoch": 0.4024049301067188,
      "flos": 22930220442240.0,
      "grad_norm": 2.080452068441124,
      "language_loss": 0.83805972,
      "learning_rate": 2.7132138269563333e-06,
      "loss": 0.85912919,
      "num_input_tokens_seen": 143776770,
      "step": 6693,
      "time_per_iteration": 2.6005442142486572
    },
    {
      "auxiliary_loss_clip": 0.01084834,
      "auxiliary_loss_mlp": 0.01043643,
      "balance_loss_clip": 1.04639482,
      "balance_loss_mlp": 1.02909911,
      "epoch": 0.40246505335938676,
      "flos": 36028421487360.0,
      "grad_norm": 1.972885657183727,
      "language_loss": 0.71157956,
      "learning_rate": 2.7128499570995483e-06,
      "loss": 0.73286438,
      "num_input_tokens_seen": 143798450,
      "step": 6694,
      "time_per_iteration": 2.7122020721435547
    },
    {
      "auxiliary_loss_clip": 0.01094248,
      "auxiliary_loss_mlp": 0.01033639,
      "balance_loss_clip": 1.04310942,
      "balance_loss_mlp": 1.02008462,
      "epoch": 0.4025251766120547,
      "flos": 20594303176320.0,
      "grad_norm": 2.019754183498444,
      "language_loss": 0.67794913,
      "learning_rate": 2.7124860602114212e-06,
      "loss": 0.69922805,
      "num_input_tokens_seen": 143816995,
      "step": 6695,
      "time_per_iteration": 2.52754545211792
    },
    {
      "auxiliary_loss_clip": 0.01092388,
      "auxiliary_loss_mlp": 0.01035482,
      "balance_loss_clip": 1.04045296,
      "balance_loss_mlp": 1.02143919,
      "epoch": 0.4025852998647227,
      "flos": 64523932381440.0,
      "grad_norm": 2.092151344628605,
      "language_loss": 0.79533088,
      "learning_rate": 2.7121221363057515e-06,
      "loss": 0.8166095,
      "num_input_tokens_seen": 143842090,
      "step": 6696,
      "time_per_iteration": 2.945188045501709
    },
    {
      "auxiliary_loss_clip": 0.01098191,
      "auxiliary_loss_mlp": 0.01048227,
      "balance_loss_clip": 1.0447197,
      "balance_loss_mlp": 1.032336,
      "epoch": 0.40264542311739066,
      "flos": 20886292834560.0,
      "grad_norm": 1.912673719783568,
      "language_loss": 0.71027279,
      "learning_rate": 2.7117581853963393e-06,
      "loss": 0.73173702,
      "num_input_tokens_seen": 143860800,
      "step": 6697,
      "time_per_iteration": 2.576551675796509
    },
    {
      "auxiliary_loss_clip": 0.01112749,
      "auxiliary_loss_mlp": 0.01035592,
      "balance_loss_clip": 1.04578686,
      "balance_loss_mlp": 1.02228236,
      "epoch": 0.4027055463700586,
      "flos": 26250197685120.0,
      "grad_norm": 2.019138799736659,
      "language_loss": 0.6187346,
      "learning_rate": 2.711394207496984e-06,
      "loss": 0.64021802,
      "num_input_tokens_seen": 143878950,
      "step": 6698,
      "time_per_iteration": 2.5355989933013916
    },
    {
      "auxiliary_loss_clip": 0.01114457,
      "auxiliary_loss_mlp": 0.01029172,
      "balance_loss_clip": 1.04649055,
      "balance_loss_mlp": 1.01533175,
      "epoch": 0.4027656696227266,
      "flos": 20631398947200.0,
      "grad_norm": 2.5363923816351828,
      "language_loss": 0.77275556,
      "learning_rate": 2.711030202621491e-06,
      "loss": 0.79419184,
      "num_input_tokens_seen": 143898385,
      "step": 6699,
      "time_per_iteration": 2.5343313217163086
    },
    {
      "auxiliary_loss_clip": 0.01086241,
      "auxiliary_loss_mlp": 0.01027188,
      "balance_loss_clip": 1.04405606,
      "balance_loss_mlp": 1.01451564,
      "epoch": 0.40282579287539455,
      "flos": 22346277039360.0,
      "grad_norm": 2.126319305041707,
      "language_loss": 0.8024683,
      "learning_rate": 2.7106661707836605e-06,
      "loss": 0.82360256,
      "num_input_tokens_seen": 143918795,
      "step": 6700,
      "time_per_iteration": 2.5958404541015625
    },
    {
      "auxiliary_loss_clip": 0.01106192,
      "auxiliary_loss_mlp": 0.01037014,
      "balance_loss_clip": 1.04975724,
      "balance_loss_mlp": 1.02168322,
      "epoch": 0.4028859161280625,
      "flos": 29274988959360.0,
      "grad_norm": 1.8114333728397314,
      "language_loss": 0.75060594,
      "learning_rate": 2.7103021119972977e-06,
      "loss": 0.77203798,
      "num_input_tokens_seen": 143938245,
      "step": 6701,
      "time_per_iteration": 2.606935739517212
    },
    {
      "auxiliary_loss_clip": 0.01092987,
      "auxiliary_loss_mlp": 0.0103394,
      "balance_loss_clip": 1.04410374,
      "balance_loss_mlp": 1.02121449,
      "epoch": 0.4029460393807305,
      "flos": 28622312881920.0,
      "grad_norm": 1.6498034203487644,
      "language_loss": 0.66174555,
      "learning_rate": 2.709938026276208e-06,
      "loss": 0.68301487,
      "num_input_tokens_seen": 143960995,
      "step": 6702,
      "time_per_iteration": 2.600428581237793
    },
    {
      "auxiliary_loss_clip": 0.01097379,
      "auxiliary_loss_mlp": 0.01034683,
      "balance_loss_clip": 1.04463136,
      "balance_loss_mlp": 1.0202291,
      "epoch": 0.40300616263339845,
      "flos": 22601925112320.0,
      "grad_norm": 1.8696773033957998,
      "language_loss": 0.65920657,
      "learning_rate": 2.7095739136341964e-06,
      "loss": 0.68052721,
      "num_input_tokens_seen": 143979910,
      "step": 6703,
      "time_per_iteration": 2.5674004554748535
    },
    {
      "auxiliary_loss_clip": 0.01056396,
      "auxiliary_loss_mlp": 0.01037601,
      "balance_loss_clip": 1.04614019,
      "balance_loss_mlp": 1.02211559,
      "epoch": 0.4030662858860664,
      "flos": 25520313323520.0,
      "grad_norm": 1.933230747477807,
      "language_loss": 0.81902027,
      "learning_rate": 2.709209774085071e-06,
      "loss": 0.83996022,
      "num_input_tokens_seen": 144000095,
      "step": 6704,
      "time_per_iteration": 2.749000072479248
    },
    {
      "auxiliary_loss_clip": 0.01111213,
      "auxiliary_loss_mlp": 0.01032106,
      "balance_loss_clip": 1.05053246,
      "balance_loss_mlp": 1.01855814,
      "epoch": 0.40312640913873443,
      "flos": 23586703361280.0,
      "grad_norm": 1.6333579160463483,
      "language_loss": 0.73098695,
      "learning_rate": 2.7088456076426407e-06,
      "loss": 0.75242013,
      "num_input_tokens_seen": 144019695,
      "step": 6705,
      "time_per_iteration": 2.7992630004882812
    },
    {
      "auxiliary_loss_clip": 0.01109141,
      "auxiliary_loss_mlp": 0.01032208,
      "balance_loss_clip": 1.04561698,
      "balance_loss_mlp": 1.01936889,
      "epoch": 0.4031865323914024,
      "flos": 20011042131840.0,
      "grad_norm": 1.9997398191913804,
      "language_loss": 0.66614842,
      "learning_rate": 2.708481414320713e-06,
      "loss": 0.68756187,
      "num_input_tokens_seen": 144038525,
      "step": 6706,
      "time_per_iteration": 2.5008106231689453
    },
    {
      "auxiliary_loss_clip": 0.01112843,
      "auxiliary_loss_mlp": 0.01038025,
      "balance_loss_clip": 1.04706216,
      "balance_loss_mlp": 1.02420902,
      "epoch": 0.40324665564407036,
      "flos": 21871430219520.0,
      "grad_norm": 1.5240327175239061,
      "language_loss": 0.71257877,
      "learning_rate": 2.7081171941330992e-06,
      "loss": 0.73408747,
      "num_input_tokens_seen": 144059485,
      "step": 6707,
      "time_per_iteration": 2.539362668991089
    },
    {
      "auxiliary_loss_clip": 0.01096289,
      "auxiliary_loss_mlp": 0.01026964,
      "balance_loss_clip": 1.04515123,
      "balance_loss_mlp": 1.01387453,
      "epoch": 0.4033067788967383,
      "flos": 23878728933120.0,
      "grad_norm": 1.5303743920383786,
      "language_loss": 0.79745501,
      "learning_rate": 2.707752947093611e-06,
      "loss": 0.81868756,
      "num_input_tokens_seen": 144080265,
      "step": 6708,
      "time_per_iteration": 2.5736637115478516
    },
    {
      "auxiliary_loss_clip": 0.01074226,
      "auxiliary_loss_mlp": 0.01035686,
      "balance_loss_clip": 1.04180169,
      "balance_loss_mlp": 1.02198887,
      "epoch": 0.4033669021494063,
      "flos": 17419907756160.0,
      "grad_norm": 2.0708209337115684,
      "language_loss": 0.82628655,
      "learning_rate": 2.70738867321606e-06,
      "loss": 0.84738564,
      "num_input_tokens_seen": 144098040,
      "step": 6709,
      "time_per_iteration": 2.5728559494018555
    },
    {
      "auxiliary_loss_clip": 0.01115675,
      "auxiliary_loss_mlp": 0.01035578,
      "balance_loss_clip": 1.05128598,
      "balance_loss_mlp": 1.02145123,
      "epoch": 0.40342702540207426,
      "flos": 29600554855680.0,
      "grad_norm": 1.4957394050170716,
      "language_loss": 0.71136451,
      "learning_rate": 2.70702437251426e-06,
      "loss": 0.73287702,
      "num_input_tokens_seen": 144118265,
      "step": 6710,
      "time_per_iteration": 2.5703036785125732
    },
    {
      "auxiliary_loss_clip": 0.01091098,
      "auxiliary_loss_mlp": 0.01031143,
      "balance_loss_clip": 1.04484558,
      "balance_loss_mlp": 1.01776814,
      "epoch": 0.4034871486547422,
      "flos": 11284605400320.0,
      "grad_norm": 2.477459158824322,
      "language_loss": 0.8527509,
      "learning_rate": 2.7066600450020236e-06,
      "loss": 0.87397337,
      "num_input_tokens_seen": 144133865,
      "step": 6711,
      "time_per_iteration": 2.507282257080078
    },
    {
      "auxiliary_loss_clip": 0.01116941,
      "auxiliary_loss_mlp": 0.01032683,
      "balance_loss_clip": 1.04766989,
      "balance_loss_mlp": 1.01883113,
      "epoch": 0.4035472719074102,
      "flos": 15552839738880.0,
      "grad_norm": 2.0314412210675497,
      "language_loss": 0.7657162,
      "learning_rate": 2.706295690693168e-06,
      "loss": 0.78721249,
      "num_input_tokens_seen": 144150125,
      "step": 6712,
      "time_per_iteration": 2.5194931030273438
    },
    {
      "auxiliary_loss_clip": 0.0110318,
      "auxiliary_loss_mlp": 0.01039378,
      "balance_loss_clip": 1.04748845,
      "balance_loss_mlp": 1.02574682,
      "epoch": 0.40360739516007815,
      "flos": 24674365140480.0,
      "grad_norm": 1.8465563084290522,
      "language_loss": 0.78312701,
      "learning_rate": 2.7059313096015096e-06,
      "loss": 0.80455267,
      "num_input_tokens_seen": 144169295,
      "step": 6713,
      "time_per_iteration": 2.558682918548584
    },
    {
      "auxiliary_loss_clip": 0.01091225,
      "auxiliary_loss_mlp": 0.01036235,
      "balance_loss_clip": 1.04326773,
      "balance_loss_mlp": 1.02098858,
      "epoch": 0.4036675184127461,
      "flos": 17304095329920.0,
      "grad_norm": 2.149655025184392,
      "language_loss": 0.88074416,
      "learning_rate": 2.705566901740865e-06,
      "loss": 0.90201879,
      "num_input_tokens_seen": 144185790,
      "step": 6714,
      "time_per_iteration": 2.5509650707244873
    },
    {
      "auxiliary_loss_clip": 0.01116439,
      "auxiliary_loss_mlp": 0.01036656,
      "balance_loss_clip": 1.04856622,
      "balance_loss_mlp": 1.02342355,
      "epoch": 0.4037276416654141,
      "flos": 19864023765120.0,
      "grad_norm": 1.7265144693806869,
      "language_loss": 0.69112509,
      "learning_rate": 2.7052024671250527e-06,
      "loss": 0.71265602,
      "num_input_tokens_seen": 144205190,
      "step": 6715,
      "time_per_iteration": 2.5454583168029785
    },
    {
      "auxiliary_loss_clip": 0.0108189,
      "auxiliary_loss_mlp": 0.01031611,
      "balance_loss_clip": 1.04356003,
      "balance_loss_mlp": 1.01787806,
      "epoch": 0.40378776491808205,
      "flos": 18296271780480.0,
      "grad_norm": 1.9100919630952946,
      "language_loss": 0.77039754,
      "learning_rate": 2.704838005767892e-06,
      "loss": 0.79153258,
      "num_input_tokens_seen": 144222705,
      "step": 6716,
      "time_per_iteration": 4.118279933929443
    },
    {
      "auxiliary_loss_clip": 0.01078749,
      "auxiliary_loss_mlp": 0.0103177,
      "balance_loss_clip": 1.04607952,
      "balance_loss_mlp": 1.01935434,
      "epoch": 0.40384788817075,
      "flos": 15049372757760.0,
      "grad_norm": 2.216346700890461,
      "language_loss": 0.7642917,
      "learning_rate": 2.7044735176832037e-06,
      "loss": 0.78539687,
      "num_input_tokens_seen": 144239545,
      "step": 6717,
      "time_per_iteration": 2.5754966735839844
    },
    {
      "auxiliary_loss_clip": 0.01037655,
      "auxiliary_loss_mlp": 0.01001156,
      "balance_loss_clip": 1.02711344,
      "balance_loss_mlp": 0.99978542,
      "epoch": 0.40390801142341803,
      "flos": 61929927895680.0,
      "grad_norm": 0.9297642796698974,
      "language_loss": 0.60703176,
      "learning_rate": 2.7041090028848084e-06,
      "loss": 0.62741983,
      "num_input_tokens_seen": 144288145,
      "step": 6718,
      "time_per_iteration": 2.9968924522399902
    },
    {
      "auxiliary_loss_clip": 0.01129054,
      "auxiliary_loss_mlp": 0.01034887,
      "balance_loss_clip": 1.04654026,
      "balance_loss_mlp": 1.02017641,
      "epoch": 0.403968134676086,
      "flos": 22738779930240.0,
      "grad_norm": 3.5512177776227927,
      "language_loss": 0.74935722,
      "learning_rate": 2.7037444613865306e-06,
      "loss": 0.77099663,
      "num_input_tokens_seen": 144302315,
      "step": 6719,
      "time_per_iteration": 3.9547414779663086
    },
    {
      "auxiliary_loss_clip": 0.01118611,
      "auxiliary_loss_mlp": 0.0104188,
      "balance_loss_clip": 1.0483222,
      "balance_loss_mlp": 1.02747345,
      "epoch": 0.40402825792875396,
      "flos": 19784409269760.0,
      "grad_norm": 2.1167355610814655,
      "language_loss": 0.81324291,
      "learning_rate": 2.7033798932021906e-06,
      "loss": 0.83484781,
      "num_input_tokens_seen": 144318990,
      "step": 6720,
      "time_per_iteration": 3.8440959453582764
    },
    {
      "auxiliary_loss_clip": 0.01103371,
      "auxiliary_loss_mlp": 0.01029381,
      "balance_loss_clip": 1.04797411,
      "balance_loss_mlp": 1.0161072,
      "epoch": 0.40408838118142193,
      "flos": 19609273532160.0,
      "grad_norm": 1.9567295308614632,
      "language_loss": 0.77105397,
      "learning_rate": 2.7030152983456153e-06,
      "loss": 0.79238153,
      "num_input_tokens_seen": 144335765,
      "step": 6721,
      "time_per_iteration": 3.928610324859619
    },
    {
      "auxiliary_loss_clip": 0.01091236,
      "auxiliary_loss_mlp": 0.01026126,
      "balance_loss_clip": 1.05040932,
      "balance_loss_mlp": 1.01454473,
      "epoch": 0.4041485044340899,
      "flos": 24426043441920.0,
      "grad_norm": 1.742144270304361,
      "language_loss": 0.7250129,
      "learning_rate": 2.7026506768306304e-06,
      "loss": 0.74618649,
      "num_input_tokens_seen": 144355825,
      "step": 6722,
      "time_per_iteration": 2.5956079959869385
    },
    {
      "auxiliary_loss_clip": 0.01112396,
      "auxiliary_loss_mlp": 0.01028526,
      "balance_loss_clip": 1.04703403,
      "balance_loss_mlp": 1.01606858,
      "epoch": 0.40420862768675786,
      "flos": 16760192613120.0,
      "grad_norm": 1.8434678555712887,
      "language_loss": 0.65662491,
      "learning_rate": 2.7022860286710602e-06,
      "loss": 0.67803419,
      "num_input_tokens_seen": 144374320,
      "step": 6723,
      "time_per_iteration": 2.487516403198242
    },
    {
      "auxiliary_loss_clip": 0.01115401,
      "auxiliary_loss_mlp": 0.01039994,
      "balance_loss_clip": 1.05111742,
      "balance_loss_mlp": 1.02564096,
      "epoch": 0.4042687509394258,
      "flos": 22491571553280.0,
      "grad_norm": 1.49806729908896,
      "language_loss": 0.73690772,
      "learning_rate": 2.701921353880734e-06,
      "loss": 0.75846171,
      "num_input_tokens_seen": 144394325,
      "step": 6724,
      "time_per_iteration": 2.512136936187744
    },
    {
      "auxiliary_loss_clip": 0.01094446,
      "auxiliary_loss_mlp": 0.01034835,
      "balance_loss_clip": 1.0481348,
      "balance_loss_mlp": 1.02170968,
      "epoch": 0.4043288741920938,
      "flos": 30336149479680.0,
      "grad_norm": 1.7345432133911438,
      "language_loss": 0.74707603,
      "learning_rate": 2.7015566524734787e-06,
      "loss": 0.76836884,
      "num_input_tokens_seen": 144412765,
      "step": 6725,
      "time_per_iteration": 2.580963611602783
    },
    {
      "auxiliary_loss_clip": 0.01109255,
      "auxiliary_loss_mlp": 0.01027139,
      "balance_loss_clip": 1.04845643,
      "balance_loss_mlp": 1.01257181,
      "epoch": 0.40438899744476176,
      "flos": 46348321363200.0,
      "grad_norm": 2.644308346963915,
      "language_loss": 0.76618677,
      "learning_rate": 2.701191924463126e-06,
      "loss": 0.78755069,
      "num_input_tokens_seen": 144435400,
      "step": 6726,
      "time_per_iteration": 2.725804567337036
    },
    {
      "auxiliary_loss_clip": 0.01099084,
      "auxiliary_loss_mlp": 0.00792736,
      "balance_loss_clip": 1.04312611,
      "balance_loss_mlp": 1.01369119,
      "epoch": 0.4044491206974297,
      "flos": 13333524998400.0,
      "grad_norm": 1.9159821297985327,
      "language_loss": 0.8176688,
      "learning_rate": 2.7008271698635054e-06,
      "loss": 0.83658695,
      "num_input_tokens_seen": 144452925,
      "step": 6727,
      "time_per_iteration": 2.5089118480682373
    },
    {
      "auxiliary_loss_clip": 0.01122997,
      "auxiliary_loss_mlp": 0.01028049,
      "balance_loss_clip": 1.04507136,
      "balance_loss_mlp": 1.01524568,
      "epoch": 0.4045092439500977,
      "flos": 12093745121280.0,
      "grad_norm": 1.8981940472379273,
      "language_loss": 0.85203177,
      "learning_rate": 2.700462388688447e-06,
      "loss": 0.87354219,
      "num_input_tokens_seen": 144470195,
      "step": 6728,
      "time_per_iteration": 2.508298635482788
    },
    {
      "auxiliary_loss_clip": 0.01090232,
      "auxiliary_loss_mlp": 0.01027345,
      "balance_loss_clip": 1.04597461,
      "balance_loss_mlp": 1.014768,
      "epoch": 0.40456936720276565,
      "flos": 21179683123200.0,
      "grad_norm": 2.006748857296638,
      "language_loss": 0.81981856,
      "learning_rate": 2.700097580951786e-06,
      "loss": 0.84099436,
      "num_input_tokens_seen": 144490320,
      "step": 6729,
      "time_per_iteration": 2.551534414291382
    },
    {
      "auxiliary_loss_clip": 0.0110229,
      "auxiliary_loss_mlp": 0.01035275,
      "balance_loss_clip": 1.04726863,
      "balance_loss_mlp": 1.02197146,
      "epoch": 0.4046294904554336,
      "flos": 23915286000000.0,
      "grad_norm": 2.0931471389866565,
      "language_loss": 0.73000288,
      "learning_rate": 2.6997327466673533e-06,
      "loss": 0.75137854,
      "num_input_tokens_seen": 144508990,
      "step": 6730,
      "time_per_iteration": 2.5615696907043457
    },
    {
      "auxiliary_loss_clip": 0.01109254,
      "auxiliary_loss_mlp": 0.01032428,
      "balance_loss_clip": 1.04385555,
      "balance_loss_mlp": 1.01908827,
      "epoch": 0.4046896137081016,
      "flos": 38071235773440.0,
      "grad_norm": 1.6965617221567244,
      "language_loss": 0.67887986,
      "learning_rate": 2.699367885848985e-06,
      "loss": 0.7002967,
      "num_input_tokens_seen": 144529550,
      "step": 6731,
      "time_per_iteration": 2.6306514739990234
    },
    {
      "auxiliary_loss_clip": 0.01121517,
      "auxiliary_loss_mlp": 0.01031294,
      "balance_loss_clip": 1.04449666,
      "balance_loss_mlp": 1.01928973,
      "epoch": 0.4047497369607696,
      "flos": 23617262856960.0,
      "grad_norm": 1.5635018053618426,
      "language_loss": 0.738428,
      "learning_rate": 2.699002998510517e-06,
      "loss": 0.75995612,
      "num_input_tokens_seen": 144549310,
      "step": 6732,
      "time_per_iteration": 2.49080753326416
    },
    {
      "auxiliary_loss_clip": 0.01093595,
      "auxiliary_loss_mlp": 0.00786568,
      "balance_loss_clip": 1.04447532,
      "balance_loss_mlp": 1.0105648,
      "epoch": 0.40480986021343757,
      "flos": 12823593569280.0,
      "grad_norm": 1.8055656162293465,
      "language_loss": 0.77331543,
      "learning_rate": 2.6986380846657852e-06,
      "loss": 0.79211712,
      "num_input_tokens_seen": 144567430,
      "step": 6733,
      "time_per_iteration": 2.5254604816436768
    },
    {
      "auxiliary_loss_clip": 0.01097295,
      "auxiliary_loss_mlp": 0.01036211,
      "balance_loss_clip": 1.04166567,
      "balance_loss_mlp": 1.02170873,
      "epoch": 0.40486998346610553,
      "flos": 23768770423680.0,
      "grad_norm": 1.8409244543292647,
      "language_loss": 0.76390648,
      "learning_rate": 2.698273144328627e-06,
      "loss": 0.78524154,
      "num_input_tokens_seen": 144585975,
      "step": 6734,
      "time_per_iteration": 2.553057909011841
    },
    {
      "auxiliary_loss_clip": 0.0110519,
      "auxiliary_loss_mlp": 0.01034524,
      "balance_loss_clip": 1.04897547,
      "balance_loss_mlp": 1.02116704,
      "epoch": 0.4049301067187735,
      "flos": 22856818999680.0,
      "grad_norm": 1.9313395418600623,
      "language_loss": 0.64373738,
      "learning_rate": 2.6979081775128805e-06,
      "loss": 0.66513449,
      "num_input_tokens_seen": 144605225,
      "step": 6735,
      "time_per_iteration": 2.5432989597320557
    },
    {
      "auxiliary_loss_clip": 0.010817,
      "auxiliary_loss_mlp": 0.0103774,
      "balance_loss_clip": 1.03999412,
      "balance_loss_mlp": 1.024436,
      "epoch": 0.40499022997144146,
      "flos": 22783992174720.0,
      "grad_norm": 2.0363345303392797,
      "language_loss": 0.82728237,
      "learning_rate": 2.697543184232387e-06,
      "loss": 0.84847677,
      "num_input_tokens_seen": 144624145,
      "step": 6736,
      "time_per_iteration": 2.5772979259490967
    },
    {
      "auxiliary_loss_clip": 0.01093208,
      "auxiliary_loss_mlp": 0.00788069,
      "balance_loss_clip": 1.04388738,
      "balance_loss_mlp": 1.00959826,
      "epoch": 0.4050503532241094,
      "flos": 23039352938880.0,
      "grad_norm": 1.7328782526291835,
      "language_loss": 0.75396949,
      "learning_rate": 2.6971781645009863e-06,
      "loss": 0.77278233,
      "num_input_tokens_seen": 144644470,
      "step": 6737,
      "time_per_iteration": 2.5709688663482666
    },
    {
      "auxiliary_loss_clip": 0.01112422,
      "auxiliary_loss_mlp": 0.01039035,
      "balance_loss_clip": 1.04579854,
      "balance_loss_mlp": 1.02604699,
      "epoch": 0.4051104764767774,
      "flos": 16647756065280.0,
      "grad_norm": 2.3779003906642813,
      "language_loss": 0.71770048,
      "learning_rate": 2.696813118332519e-06,
      "loss": 0.73921502,
      "num_input_tokens_seen": 144661055,
      "step": 6738,
      "time_per_iteration": 2.510666847229004
    },
    {
      "auxiliary_loss_clip": 0.0108557,
      "auxiliary_loss_mlp": 0.01033241,
      "balance_loss_clip": 1.03960752,
      "balance_loss_mlp": 1.02087915,
      "epoch": 0.40517059972944536,
      "flos": 16358962717440.0,
      "grad_norm": 2.033141152557743,
      "language_loss": 0.74733102,
      "learning_rate": 2.696448045740828e-06,
      "loss": 0.76851916,
      "num_input_tokens_seen": 144677935,
      "step": 6739,
      "time_per_iteration": 2.5242350101470947
    },
    {
      "auxiliary_loss_clip": 0.01088354,
      "auxiliary_loss_mlp": 0.0103332,
      "balance_loss_clip": 1.04525959,
      "balance_loss_mlp": 1.01942611,
      "epoch": 0.4052307229821133,
      "flos": 28803374363520.0,
      "grad_norm": 1.8869391037022185,
      "language_loss": 0.74257535,
      "learning_rate": 2.6960829467397576e-06,
      "loss": 0.76379204,
      "num_input_tokens_seen": 144697725,
      "step": 6740,
      "time_per_iteration": 2.6187238693237305
    },
    {
      "auxiliary_loss_clip": 0.0110673,
      "auxiliary_loss_mlp": 0.01031822,
      "balance_loss_clip": 1.04399526,
      "balance_loss_mlp": 1.01899505,
      "epoch": 0.4052908462347813,
      "flos": 21397876289280.0,
      "grad_norm": 1.607266604576023,
      "language_loss": 0.76975703,
      "learning_rate": 2.695717821343153e-06,
      "loss": 0.79114258,
      "num_input_tokens_seen": 144718805,
      "step": 6741,
      "time_per_iteration": 2.55676007270813
    },
    {
      "auxiliary_loss_clip": 0.01126838,
      "auxiliary_loss_mlp": 0.01038336,
      "balance_loss_clip": 1.04665458,
      "balance_loss_mlp": 1.0234586,
      "epoch": 0.40535096948744925,
      "flos": 22419067950720.0,
      "grad_norm": 2.3309666683725405,
      "language_loss": 0.71470547,
      "learning_rate": 2.6953526695648577e-06,
      "loss": 0.73635715,
      "num_input_tokens_seen": 144737105,
      "step": 6742,
      "time_per_iteration": 2.491229772567749
    },
    {
      "auxiliary_loss_clip": 0.01125482,
      "auxiliary_loss_mlp": 0.01031991,
      "balance_loss_clip": 1.04627895,
      "balance_loss_mlp": 1.01817489,
      "epoch": 0.4054110927401172,
      "flos": 17010776868480.0,
      "grad_norm": 2.0732017142561268,
      "language_loss": 0.71839273,
      "learning_rate": 2.6949874914187202e-06,
      "loss": 0.73996747,
      "num_input_tokens_seen": 144751350,
      "step": 6743,
      "time_per_iteration": 2.416482448577881
    },
    {
      "auxiliary_loss_clip": 0.01101797,
      "auxiliary_loss_mlp": 0.01032938,
      "balance_loss_clip": 1.04375541,
      "balance_loss_mlp": 1.01876962,
      "epoch": 0.4054712159927852,
      "flos": 21614848392960.0,
      "grad_norm": 1.962138527337549,
      "language_loss": 0.70713949,
      "learning_rate": 2.694622286918588e-06,
      "loss": 0.7284869,
      "num_input_tokens_seen": 144770030,
      "step": 6744,
      "time_per_iteration": 2.553705930709839
    },
    {
      "auxiliary_loss_clip": 0.01109818,
      "auxiliary_loss_mlp": 0.01035039,
      "balance_loss_clip": 1.04407001,
      "balance_loss_mlp": 1.02281964,
      "epoch": 0.4055313392454532,
      "flos": 25812554376960.0,
      "grad_norm": 1.6239842500609276,
      "language_loss": 0.80172557,
      "learning_rate": 2.6942570560783076e-06,
      "loss": 0.82317412,
      "num_input_tokens_seen": 144790965,
      "step": 6745,
      "time_per_iteration": 2.546596050262451
    },
    {
      "auxiliary_loss_clip": 0.01105699,
      "auxiliary_loss_mlp": 0.01033649,
      "balance_loss_clip": 1.04608583,
      "balance_loss_mlp": 1.01986265,
      "epoch": 0.40559146249812117,
      "flos": 14137098111360.0,
      "grad_norm": 1.834793896836658,
      "language_loss": 0.66692811,
      "learning_rate": 2.693891798911731e-06,
      "loss": 0.68832159,
      "num_input_tokens_seen": 144807755,
      "step": 6746,
      "time_per_iteration": 2.485659122467041
    },
    {
      "auxiliary_loss_clip": 0.01086128,
      "auxiliary_loss_mlp": 0.01029239,
      "balance_loss_clip": 1.04395497,
      "balance_loss_mlp": 1.01647139,
      "epoch": 0.40565158575078913,
      "flos": 41355481962240.0,
      "grad_norm": 1.589996203313907,
      "language_loss": 0.57112443,
      "learning_rate": 2.6935265154327075e-06,
      "loss": 0.59227812,
      "num_input_tokens_seen": 144832405,
      "step": 6747,
      "time_per_iteration": 2.741436243057251
    },
    {
      "auxiliary_loss_clip": 0.01094528,
      "auxiliary_loss_mlp": 0.01039651,
      "balance_loss_clip": 1.04671216,
      "balance_loss_mlp": 1.0270803,
      "epoch": 0.4057117090034571,
      "flos": 28544529980160.0,
      "grad_norm": 1.7276330152873935,
      "language_loss": 0.84409076,
      "learning_rate": 2.693161205655089e-06,
      "loss": 0.86543262,
      "num_input_tokens_seen": 144853890,
      "step": 6748,
      "time_per_iteration": 2.658112049102783
    },
    {
      "auxiliary_loss_clip": 0.01098877,
      "auxiliary_loss_mlp": 0.01041378,
      "balance_loss_clip": 1.04749084,
      "balance_loss_mlp": 1.02743602,
      "epoch": 0.40577183225612506,
      "flos": 18004066640640.0,
      "grad_norm": 2.1749191179579848,
      "language_loss": 0.81296337,
      "learning_rate": 2.6927958695927287e-06,
      "loss": 0.83436596,
      "num_input_tokens_seen": 144871395,
      "step": 6749,
      "time_per_iteration": 2.5143918991088867
    },
    {
      "auxiliary_loss_clip": 0.01110281,
      "auxiliary_loss_mlp": 0.00789194,
      "balance_loss_clip": 1.04438472,
      "balance_loss_mlp": 1.01255584,
      "epoch": 0.40583195550879303,
      "flos": 19536734016000.0,
      "grad_norm": 1.8339487941733779,
      "language_loss": 0.75448799,
      "learning_rate": 2.6924305072594784e-06,
      "loss": 0.7734828,
      "num_input_tokens_seen": 144890975,
      "step": 6750,
      "time_per_iteration": 2.5323739051818848
    },
    {
      "auxiliary_loss_clip": 0.01104872,
      "auxiliary_loss_mlp": 0.01037121,
      "balance_loss_clip": 1.04265809,
      "balance_loss_mlp": 1.02298927,
      "epoch": 0.405892078761461,
      "flos": 22309468577280.0,
      "grad_norm": 2.03104302510843,
      "language_loss": 0.74612653,
      "learning_rate": 2.692065118669195e-06,
      "loss": 0.76754647,
      "num_input_tokens_seen": 144908170,
      "step": 6751,
      "time_per_iteration": 2.5154404640197754
    },
    {
      "auxiliary_loss_clip": 0.01081636,
      "auxiliary_loss_mlp": 0.01039283,
      "balance_loss_clip": 1.04513478,
      "balance_loss_mlp": 1.02341592,
      "epoch": 0.40595220201412896,
      "flos": 25484402701440.0,
      "grad_norm": 1.521370816356625,
      "language_loss": 0.66573215,
      "learning_rate": 2.6916997038357326e-06,
      "loss": 0.68694139,
      "num_input_tokens_seen": 144928020,
      "step": 6752,
      "time_per_iteration": 2.6662120819091797
    },
    {
      "auxiliary_loss_clip": 0.01078377,
      "auxiliary_loss_mlp": 0.01036092,
      "balance_loss_clip": 1.04622817,
      "balance_loss_mlp": 1.02134633,
      "epoch": 0.4060123252667969,
      "flos": 49856004103680.0,
      "grad_norm": 2.1441800518473126,
      "language_loss": 0.70883751,
      "learning_rate": 2.691334262772948e-06,
      "loss": 0.72998226,
      "num_input_tokens_seen": 144951240,
      "step": 6753,
      "time_per_iteration": 2.8603169918060303
    },
    {
      "auxiliary_loss_clip": 0.01099811,
      "auxiliary_loss_mlp": 0.01035218,
      "balance_loss_clip": 1.04219079,
      "balance_loss_mlp": 1.02060258,
      "epoch": 0.4060724485194649,
      "flos": 21135476459520.0,
      "grad_norm": 2.1775197674281714,
      "language_loss": 0.71798944,
      "learning_rate": 2.690968795494699e-06,
      "loss": 0.73933971,
      "num_input_tokens_seen": 144969100,
      "step": 6754,
      "time_per_iteration": 2.513707399368286
    },
    {
      "auxiliary_loss_clip": 0.01093379,
      "auxiliary_loss_mlp": 0.01038131,
      "balance_loss_clip": 1.04248548,
      "balance_loss_mlp": 1.02460051,
      "epoch": 0.40613257177213286,
      "flos": 21758059918080.0,
      "grad_norm": 1.7767669729855629,
      "language_loss": 0.82769632,
      "learning_rate": 2.690603302014844e-06,
      "loss": 0.84901142,
      "num_input_tokens_seen": 144987065,
      "step": 6755,
      "time_per_iteration": 3.9503090381622314
    },
    {
      "auxiliary_loss_clip": 0.01085653,
      "auxiliary_loss_mlp": 0.01037413,
      "balance_loss_clip": 1.04592371,
      "balance_loss_mlp": 1.02338231,
      "epoch": 0.4061926950248008,
      "flos": 25555074710400.0,
      "grad_norm": 1.9390775706915246,
      "language_loss": 0.70986056,
      "learning_rate": 2.6902377823472426e-06,
      "loss": 0.73109126,
      "num_input_tokens_seen": 145007310,
      "step": 6756,
      "time_per_iteration": 2.6233458518981934
    },
    {
      "auxiliary_loss_clip": 0.01063661,
      "auxiliary_loss_mlp": 0.00790099,
      "balance_loss_clip": 1.04155552,
      "balance_loss_mlp": 1.00966692,
      "epoch": 0.4062528182774688,
      "flos": 23695799944320.0,
      "grad_norm": 2.9771929866383897,
      "language_loss": 0.7884202,
      "learning_rate": 2.689872236505755e-06,
      "loss": 0.80695778,
      "num_input_tokens_seen": 145026210,
      "step": 6757,
      "time_per_iteration": 4.1225364208221436
    },
    {
      "auxiliary_loss_clip": 0.01103365,
      "auxiliary_loss_mlp": 0.01029175,
      "balance_loss_clip": 1.04675412,
      "balance_loss_mlp": 1.01563311,
      "epoch": 0.4063129415301368,
      "flos": 21726027964800.0,
      "grad_norm": 1.6309350719740792,
      "language_loss": 0.78776717,
      "learning_rate": 2.6895066645042437e-06,
      "loss": 0.80909258,
      "num_input_tokens_seen": 145045475,
      "step": 6758,
      "time_per_iteration": 2.578040361404419
    },
    {
      "auxiliary_loss_clip": 0.0108644,
      "auxiliary_loss_mlp": 0.01033795,
      "balance_loss_clip": 1.04717946,
      "balance_loss_mlp": 1.02024055,
      "epoch": 0.40637306478280477,
      "flos": 12787575206400.0,
      "grad_norm": 1.9587827870769112,
      "language_loss": 0.88788676,
      "learning_rate": 2.6891410663565703e-06,
      "loss": 0.90908909,
      "num_input_tokens_seen": 145062260,
      "step": 6759,
      "time_per_iteration": 4.040321588516235
    },
    {
      "auxiliary_loss_clip": 0.01097142,
      "auxiliary_loss_mlp": 0.01034848,
      "balance_loss_clip": 1.04584825,
      "balance_loss_mlp": 1.02097845,
      "epoch": 0.40643318803547274,
      "flos": 24024490323840.0,
      "grad_norm": 1.6607598962831327,
      "language_loss": 0.64506137,
      "learning_rate": 2.688775442076598e-06,
      "loss": 0.6663813,
      "num_input_tokens_seen": 145082470,
      "step": 6760,
      "time_per_iteration": 3.9791762828826904
    },
    {
      "auxiliary_loss_clip": 0.01111816,
      "auxiliary_loss_mlp": 0.01032731,
      "balance_loss_clip": 1.04261708,
      "balance_loss_mlp": 1.01861119,
      "epoch": 0.4064933112881407,
      "flos": 25592421876480.0,
      "grad_norm": 1.9105038597210833,
      "language_loss": 0.75233138,
      "learning_rate": 2.688409791678193e-06,
      "loss": 0.77377683,
      "num_input_tokens_seen": 145105685,
      "step": 6761,
      "time_per_iteration": 2.5754306316375732
    },
    {
      "auxiliary_loss_clip": 0.01090834,
      "auxiliary_loss_mlp": 0.01035082,
      "balance_loss_clip": 1.0474577,
      "balance_loss_mlp": 1.02203441,
      "epoch": 0.40655343454080867,
      "flos": 22054323294720.0,
      "grad_norm": 1.4494421181115518,
      "language_loss": 0.70187092,
      "learning_rate": 2.6880441151752185e-06,
      "loss": 0.72313011,
      "num_input_tokens_seen": 145125590,
      "step": 6762,
      "time_per_iteration": 2.5537681579589844
    },
    {
      "auxiliary_loss_clip": 0.0110989,
      "auxiliary_loss_mlp": 0.01034723,
      "balance_loss_clip": 1.04818034,
      "balance_loss_mlp": 1.02181256,
      "epoch": 0.40661355779347663,
      "flos": 26468893641600.0,
      "grad_norm": 1.5531986126226598,
      "language_loss": 0.73318565,
      "learning_rate": 2.6876784125815433e-06,
      "loss": 0.75463182,
      "num_input_tokens_seen": 145146810,
      "step": 6763,
      "time_per_iteration": 2.549959659576416
    },
    {
      "auxiliary_loss_clip": 0.01085367,
      "auxiliary_loss_mlp": 0.01031546,
      "balance_loss_clip": 1.04135203,
      "balance_loss_mlp": 1.0170269,
      "epoch": 0.4066736810461446,
      "flos": 13261129136640.0,
      "grad_norm": 1.62240010954316,
      "language_loss": 0.69232035,
      "learning_rate": 2.687312683911033e-06,
      "loss": 0.71348953,
      "num_input_tokens_seen": 145163130,
      "step": 6764,
      "time_per_iteration": 2.5176656246185303
    },
    {
      "auxiliary_loss_clip": 0.01093138,
      "auxiliary_loss_mlp": 0.01038313,
      "balance_loss_clip": 1.04074407,
      "balance_loss_mlp": 1.02297056,
      "epoch": 0.40673380429881256,
      "flos": 28803625758720.0,
      "grad_norm": 2.0685141023080473,
      "language_loss": 0.91309512,
      "learning_rate": 2.686946929177557e-06,
      "loss": 0.93440962,
      "num_input_tokens_seen": 145181420,
      "step": 6765,
      "time_per_iteration": 2.5855164527893066
    },
    {
      "auxiliary_loss_clip": 0.01111606,
      "auxiliary_loss_mlp": 0.01034931,
      "balance_loss_clip": 1.0433836,
      "balance_loss_mlp": 1.02034521,
      "epoch": 0.4067939275514805,
      "flos": 12495334152960.0,
      "grad_norm": 2.40438238596153,
      "language_loss": 0.78836834,
      "learning_rate": 2.6865811483949855e-06,
      "loss": 0.80983365,
      "num_input_tokens_seen": 145198545,
      "step": 6766,
      "time_per_iteration": 2.491452217102051
    },
    {
      "auxiliary_loss_clip": 0.01124077,
      "auxiliary_loss_mlp": 0.01032513,
      "balance_loss_clip": 1.04334903,
      "balance_loss_mlp": 1.01895857,
      "epoch": 0.4068540508041485,
      "flos": 18770508069120.0,
      "grad_norm": 1.8861793963221483,
      "language_loss": 0.76617962,
      "learning_rate": 2.6862153415771867e-06,
      "loss": 0.78774554,
      "num_input_tokens_seen": 145215835,
      "step": 6767,
      "time_per_iteration": 2.4552407264709473
    },
    {
      "auxiliary_loss_clip": 0.0110969,
      "auxiliary_loss_mlp": 0.01032409,
      "balance_loss_clip": 1.04406357,
      "balance_loss_mlp": 1.01974261,
      "epoch": 0.40691417405681646,
      "flos": 28512821249280.0,
      "grad_norm": 1.8885269377200693,
      "language_loss": 0.77444309,
      "learning_rate": 2.685849508738034e-06,
      "loss": 0.79586411,
      "num_input_tokens_seen": 145236555,
      "step": 6768,
      "time_per_iteration": 2.557558298110962
    },
    {
      "auxiliary_loss_clip": 0.01122129,
      "auxiliary_loss_mlp": 0.01032238,
      "balance_loss_clip": 1.04390907,
      "balance_loss_mlp": 1.01919091,
      "epoch": 0.4069742973094844,
      "flos": 20814040627200.0,
      "grad_norm": 1.9393004801923521,
      "language_loss": 0.86812299,
      "learning_rate": 2.6854836498913995e-06,
      "loss": 0.88966668,
      "num_input_tokens_seen": 145254595,
      "step": 6769,
      "time_per_iteration": 2.469980478286743
    },
    {
      "auxiliary_loss_clip": 0.01095441,
      "auxiliary_loss_mlp": 0.01039219,
      "balance_loss_clip": 1.04509497,
      "balance_loss_mlp": 1.02579582,
      "epoch": 0.4070344205621524,
      "flos": 21470272151040.0,
      "grad_norm": 1.7534827526793941,
      "language_loss": 0.8077876,
      "learning_rate": 2.685117765051156e-06,
      "loss": 0.82913423,
      "num_input_tokens_seen": 145274005,
      "step": 6770,
      "time_per_iteration": 2.516430139541626
    },
    {
      "auxiliary_loss_clip": 0.0112473,
      "auxiliary_loss_mlp": 0.0102955,
      "balance_loss_clip": 1.04366362,
      "balance_loss_mlp": 1.0155375,
      "epoch": 0.4070945438148204,
      "flos": 26830046937600.0,
      "grad_norm": 1.6274716307928103,
      "language_loss": 0.8019802,
      "learning_rate": 2.6847518542311783e-06,
      "loss": 0.82352298,
      "num_input_tokens_seen": 145294850,
      "step": 6771,
      "time_per_iteration": 2.5240402221679688
    },
    {
      "auxiliary_loss_clip": 0.0108845,
      "auxiliary_loss_mlp": 0.0103629,
      "balance_loss_clip": 1.04183793,
      "balance_loss_mlp": 1.02289057,
      "epoch": 0.4071546670674884,
      "flos": 26354158623360.0,
      "grad_norm": 1.4031927181783144,
      "language_loss": 0.7609669,
      "learning_rate": 2.6843859174453417e-06,
      "loss": 0.78221428,
      "num_input_tokens_seen": 145317050,
      "step": 6772,
      "time_per_iteration": 2.6213674545288086
    },
    {
      "auxiliary_loss_clip": 0.01100505,
      "auxiliary_loss_mlp": 0.01038508,
      "balance_loss_clip": 1.04332089,
      "balance_loss_mlp": 1.02422619,
      "epoch": 0.40721479032015634,
      "flos": 17895401020800.0,
      "grad_norm": 1.7235552652366573,
      "language_loss": 0.81552935,
      "learning_rate": 2.6840199547075218e-06,
      "loss": 0.83691949,
      "num_input_tokens_seen": 145334480,
      "step": 6773,
      "time_per_iteration": 2.523390531539917
    },
    {
      "auxiliary_loss_clip": 0.01039919,
      "auxiliary_loss_mlp": 0.01001185,
      "balance_loss_clip": 1.0322634,
      "balance_loss_mlp": 0.99949229,
      "epoch": 0.4072749135728243,
      "flos": 49854570537600.0,
      "grad_norm": 0.83236105649381,
      "language_loss": 0.64414012,
      "learning_rate": 2.683653966031597e-06,
      "loss": 0.66455114,
      "num_input_tokens_seen": 145388695,
      "step": 6774,
      "time_per_iteration": 3.0575952529907227
    },
    {
      "auxiliary_loss_clip": 0.01079444,
      "auxiliary_loss_mlp": 0.01034489,
      "balance_loss_clip": 1.04376674,
      "balance_loss_mlp": 1.02138817,
      "epoch": 0.40733503682549227,
      "flos": 27563630400000.0,
      "grad_norm": 1.7652977917510353,
      "language_loss": 0.72727162,
      "learning_rate": 2.683287951431446e-06,
      "loss": 0.74841094,
      "num_input_tokens_seen": 145408240,
      "step": 6775,
      "time_per_iteration": 2.6476058959960938
    },
    {
      "auxiliary_loss_clip": 0.01105137,
      "auxiliary_loss_mlp": 0.00792784,
      "balance_loss_clip": 1.04783607,
      "balance_loss_mlp": 1.01551116,
      "epoch": 0.40739516007816023,
      "flos": 22126970551680.0,
      "grad_norm": 1.4580746653257373,
      "language_loss": 0.7783798,
      "learning_rate": 2.6829219109209474e-06,
      "loss": 0.79735899,
      "num_input_tokens_seen": 145428395,
      "step": 6776,
      "time_per_iteration": 2.580406665802002
    },
    {
      "auxiliary_loss_clip": 0.01116241,
      "auxiliary_loss_mlp": 0.01041366,
      "balance_loss_clip": 1.04624629,
      "balance_loss_mlp": 1.02708507,
      "epoch": 0.4074552833308282,
      "flos": 23842243693440.0,
      "grad_norm": 2.4426421472706163,
      "language_loss": 0.79424095,
      "learning_rate": 2.682555844513981e-06,
      "loss": 0.81581706,
      "num_input_tokens_seen": 145448290,
      "step": 6777,
      "time_per_iteration": 2.5438525676727295
    },
    {
      "auxiliary_loss_clip": 0.01056421,
      "auxiliary_loss_mlp": 0.01002616,
      "balance_loss_clip": 1.02859378,
      "balance_loss_mlp": 1.00117993,
      "epoch": 0.40751540658349616,
      "flos": 58000008781440.0,
      "grad_norm": 0.6843405677614325,
      "language_loss": 0.53131956,
      "learning_rate": 2.6821897522244286e-06,
      "loss": 0.55190992,
      "num_input_tokens_seen": 145509785,
      "step": 6778,
      "time_per_iteration": 3.108289957046509
    },
    {
      "auxiliary_loss_clip": 0.01125877,
      "auxiliary_loss_mlp": 0.00799633,
      "balance_loss_clip": 1.0468564,
      "balance_loss_mlp": 1.02725291,
      "epoch": 0.40757552983616413,
      "flos": 21214659991680.0,
      "grad_norm": 2.081995608558993,
      "language_loss": 0.82343113,
      "learning_rate": 2.6818236340661718e-06,
      "loss": 0.84268618,
      "num_input_tokens_seen": 145528620,
      "step": 6779,
      "time_per_iteration": 2.5115256309509277
    },
    {
      "auxiliary_loss_clip": 0.01114675,
      "auxiliary_loss_mlp": 0.0103886,
      "balance_loss_clip": 1.04526472,
      "balance_loss_mlp": 1.02520466,
      "epoch": 0.4076356530888321,
      "flos": 26833530556800.0,
      "grad_norm": 1.5959484804484703,
      "language_loss": 0.76289284,
      "learning_rate": 2.6814574900530957e-06,
      "loss": 0.78442812,
      "num_input_tokens_seen": 145547775,
      "step": 6780,
      "time_per_iteration": 2.541353940963745
    },
    {
      "auxiliary_loss_clip": 0.01105502,
      "auxiliary_loss_mlp": 0.01029176,
      "balance_loss_clip": 1.04612505,
      "balance_loss_mlp": 1.01720142,
      "epoch": 0.40769577634150006,
      "flos": 12203021272320.0,
      "grad_norm": 1.9166704183029777,
      "language_loss": 0.65866691,
      "learning_rate": 2.6810913201990827e-06,
      "loss": 0.68001366,
      "num_input_tokens_seen": 145564465,
      "step": 6781,
      "time_per_iteration": 2.4772162437438965
    },
    {
      "auxiliary_loss_clip": 0.01097315,
      "auxiliary_loss_mlp": 0.0103356,
      "balance_loss_clip": 1.04123855,
      "balance_loss_mlp": 1.01980364,
      "epoch": 0.407755899594168,
      "flos": 33655264796160.0,
      "grad_norm": 1.6251405094360731,
      "language_loss": 0.71168756,
      "learning_rate": 2.6807251245180183e-06,
      "loss": 0.73299628,
      "num_input_tokens_seen": 145585965,
      "step": 6782,
      "time_per_iteration": 2.6232407093048096
    },
    {
      "auxiliary_loss_clip": 0.01114304,
      "auxiliary_loss_mlp": 0.01031926,
      "balance_loss_clip": 1.04385304,
      "balance_loss_mlp": 1.01885426,
      "epoch": 0.407816022846836,
      "flos": 20157342226560.0,
      "grad_norm": 1.842325251144414,
      "language_loss": 0.82395387,
      "learning_rate": 2.6803589030237897e-06,
      "loss": 0.84541619,
      "num_input_tokens_seen": 145605000,
      "step": 6783,
      "time_per_iteration": 2.493436098098755
    },
    {
      "auxiliary_loss_clip": 0.01109535,
      "auxiliary_loss_mlp": 0.01035502,
      "balance_loss_clip": 1.04659367,
      "balance_loss_mlp": 1.02181673,
      "epoch": 0.40787614609950396,
      "flos": 21178821196800.0,
      "grad_norm": 1.5243863666915238,
      "language_loss": 0.81154555,
      "learning_rate": 2.679992655730283e-06,
      "loss": 0.83299589,
      "num_input_tokens_seen": 145623740,
      "step": 6784,
      "time_per_iteration": 2.5307202339172363
    },
    {
      "auxiliary_loss_clip": 0.01094729,
      "auxiliary_loss_mlp": 0.01038907,
      "balance_loss_clip": 1.04605031,
      "balance_loss_mlp": 1.02436304,
      "epoch": 0.407936269352172,
      "flos": 20520650338560.0,
      "grad_norm": 1.7274893359850365,
      "language_loss": 0.65807456,
      "learning_rate": 2.679626382651386e-06,
      "loss": 0.67941093,
      "num_input_tokens_seen": 145643515,
      "step": 6785,
      "time_per_iteration": 2.581906795501709
    },
    {
      "auxiliary_loss_clip": 0.01107994,
      "auxiliary_loss_mlp": 0.01034283,
      "balance_loss_clip": 1.04655838,
      "balance_loss_mlp": 1.02111042,
      "epoch": 0.40799639260483994,
      "flos": 20118809911680.0,
      "grad_norm": 2.1681410154160754,
      "language_loss": 0.79699361,
      "learning_rate": 2.679260083800989e-06,
      "loss": 0.81841636,
      "num_input_tokens_seen": 145660890,
      "step": 6786,
      "time_per_iteration": 2.4962656497955322
    },
    {
      "auxiliary_loss_clip": 0.01122157,
      "auxiliary_loss_mlp": 0.01034498,
      "balance_loss_clip": 1.04452169,
      "balance_loss_mlp": 1.02239275,
      "epoch": 0.4080565158575079,
      "flos": 20997328752000.0,
      "grad_norm": 1.6241500255555212,
      "language_loss": 0.81532007,
      "learning_rate": 2.678893759192982e-06,
      "loss": 0.83688664,
      "num_input_tokens_seen": 145680070,
      "step": 6787,
      "time_per_iteration": 2.4843637943267822
    },
    {
      "auxiliary_loss_clip": 0.01112495,
      "auxiliary_loss_mlp": 0.01030257,
      "balance_loss_clip": 1.04711819,
      "balance_loss_mlp": 1.01710236,
      "epoch": 0.40811663911017587,
      "flos": 19317714837120.0,
      "grad_norm": 1.9411784197224986,
      "language_loss": 0.67617333,
      "learning_rate": 2.678527408841255e-06,
      "loss": 0.69760084,
      "num_input_tokens_seen": 145698010,
      "step": 6788,
      "time_per_iteration": 2.500030279159546
    },
    {
      "auxiliary_loss_clip": 0.01095813,
      "auxiliary_loss_mlp": 0.0104233,
      "balance_loss_clip": 1.04167747,
      "balance_loss_mlp": 1.02696443,
      "epoch": 0.40817676236284384,
      "flos": 40625382119040.0,
      "grad_norm": 9.559831749125316,
      "language_loss": 0.66149259,
      "learning_rate": 2.678161032759701e-06,
      "loss": 0.68287408,
      "num_input_tokens_seen": 145722215,
      "step": 6789,
      "time_per_iteration": 2.6981546878814697
    },
    {
      "auxiliary_loss_clip": 0.01076427,
      "auxiliary_loss_mlp": 0.01031993,
      "balance_loss_clip": 1.04110336,
      "balance_loss_mlp": 1.01825428,
      "epoch": 0.4082368856155118,
      "flos": 20522086882560.0,
      "grad_norm": 1.646905535599218,
      "language_loss": 0.6095649,
      "learning_rate": 2.6777946309622123e-06,
      "loss": 0.63064915,
      "num_input_tokens_seen": 145741090,
      "step": 6790,
      "time_per_iteration": 2.578551769256592
    },
    {
      "auxiliary_loss_clip": 0.01105838,
      "auxiliary_loss_mlp": 0.01040567,
      "balance_loss_clip": 1.04481006,
      "balance_loss_mlp": 1.02658403,
      "epoch": 0.40829700886817977,
      "flos": 11427745098240.0,
      "grad_norm": 2.802200415853678,
      "language_loss": 0.69406545,
      "learning_rate": 2.677428203462683e-06,
      "loss": 0.71552956,
      "num_input_tokens_seen": 145754985,
      "step": 6791,
      "time_per_iteration": 2.456068515777588
    },
    {
      "auxiliary_loss_clip": 0.01043774,
      "auxiliary_loss_mlp": 0.01001359,
      "balance_loss_clip": 1.026263,
      "balance_loss_mlp": 1.00005329,
      "epoch": 0.40835713212084773,
      "flos": 67330677121920.0,
      "grad_norm": 0.787686615417677,
      "language_loss": 0.59630609,
      "learning_rate": 2.6770617502750093e-06,
      "loss": 0.61675739,
      "num_input_tokens_seen": 145815260,
      "step": 6792,
      "time_per_iteration": 3.10335111618042
    },
    {
      "auxiliary_loss_clip": 0.01129547,
      "auxiliary_loss_mlp": 0.01036921,
      "balance_loss_clip": 1.04915118,
      "balance_loss_mlp": 1.02288437,
      "epoch": 0.4084172553735157,
      "flos": 21762010414080.0,
      "grad_norm": 2.220334181055106,
      "language_loss": 0.8005563,
      "learning_rate": 2.6766952714130857e-06,
      "loss": 0.82222098,
      "num_input_tokens_seen": 145832665,
      "step": 6793,
      "time_per_iteration": 2.499312400817871
    },
    {
      "auxiliary_loss_clip": 0.01114841,
      "auxiliary_loss_mlp": 0.01033083,
      "balance_loss_clip": 1.04497397,
      "balance_loss_mlp": 1.01919496,
      "epoch": 0.40847737862618366,
      "flos": 27417258478080.0,
      "grad_norm": 2.1006775862584663,
      "language_loss": 0.85182106,
      "learning_rate": 2.6763287668908094e-06,
      "loss": 0.87330031,
      "num_input_tokens_seen": 145850240,
      "step": 6794,
      "time_per_iteration": 3.9080233573913574
    },
    {
      "auxiliary_loss_clip": 0.01093564,
      "auxiliary_loss_mlp": 0.01032582,
      "balance_loss_clip": 1.0453335,
      "balance_loss_mlp": 1.01926661,
      "epoch": 0.4085375018788516,
      "flos": 18587255857920.0,
      "grad_norm": 1.5923176644369232,
      "language_loss": 0.7987864,
      "learning_rate": 2.6759622367220788e-06,
      "loss": 0.82004791,
      "num_input_tokens_seen": 145869545,
      "step": 6795,
      "time_per_iteration": 2.5581021308898926
    },
    {
      "auxiliary_loss_clip": 0.01112214,
      "auxiliary_loss_mlp": 0.01033678,
      "balance_loss_clip": 1.04415727,
      "balance_loss_mlp": 1.01906276,
      "epoch": 0.4085976251315196,
      "flos": 15411783029760.0,
      "grad_norm": 3.1049846236059375,
      "language_loss": 0.69796777,
      "learning_rate": 2.675595680920792e-06,
      "loss": 0.71942663,
      "num_input_tokens_seen": 145884025,
      "step": 6796,
      "time_per_iteration": 2.5323867797851562
    },
    {
      "auxiliary_loss_clip": 0.01106914,
      "auxiliary_loss_mlp": 0.00795973,
      "balance_loss_clip": 1.04184484,
      "balance_loss_mlp": 1.014678,
      "epoch": 0.40865774838418756,
      "flos": 21252222639360.0,
      "grad_norm": 1.6221620314906178,
      "language_loss": 0.77469456,
      "learning_rate": 2.6752290995008498e-06,
      "loss": 0.79372346,
      "num_input_tokens_seen": 145903210,
      "step": 6797,
      "time_per_iteration": 5.385161638259888
    },
    {
      "auxiliary_loss_clip": 0.01112284,
      "auxiliary_loss_mlp": 0.01040024,
      "balance_loss_clip": 1.04173279,
      "balance_loss_mlp": 1.02648234,
      "epoch": 0.4087178716368556,
      "flos": 13772245714560.0,
      "grad_norm": 1.9818854728901645,
      "language_loss": 0.85405278,
      "learning_rate": 2.6748624924761523e-06,
      "loss": 0.8755759,
      "num_input_tokens_seen": 145920985,
      "step": 6798,
      "time_per_iteration": 2.475155830383301
    },
    {
      "auxiliary_loss_clip": 0.01121024,
      "auxiliary_loss_mlp": 0.01032311,
      "balance_loss_clip": 1.04452205,
      "balance_loss_mlp": 1.02052712,
      "epoch": 0.40877799488952354,
      "flos": 23621752056960.0,
      "grad_norm": 1.4741682271874887,
      "language_loss": 0.84287596,
      "learning_rate": 2.674495859860601e-06,
      "loss": 0.86440933,
      "num_input_tokens_seen": 145940350,
      "step": 6799,
      "time_per_iteration": 3.906771659851074
    },
    {
      "auxiliary_loss_clip": 0.01084742,
      "auxiliary_loss_mlp": 0.01043528,
      "balance_loss_clip": 1.04388022,
      "balance_loss_mlp": 1.02807832,
      "epoch": 0.4088381181421915,
      "flos": 20918791664640.0,
      "grad_norm": 2.073906737188535,
      "language_loss": 0.82914329,
      "learning_rate": 2.6741292016681e-06,
      "loss": 0.85042596,
      "num_input_tokens_seen": 145957460,
      "step": 6800,
      "time_per_iteration": 2.5469377040863037
    },
    {
      "auxiliary_loss_clip": 0.01111461,
      "auxiliary_loss_mlp": 0.01038102,
      "balance_loss_clip": 1.04361343,
      "balance_loss_mlp": 1.02401137,
      "epoch": 0.4088982413948595,
      "flos": 13297578462720.0,
      "grad_norm": 1.9413420059262188,
      "language_loss": 0.74830616,
      "learning_rate": 2.6737625179125514e-06,
      "loss": 0.76980174,
      "num_input_tokens_seen": 145975285,
      "step": 6801,
      "time_per_iteration": 2.4857609272003174
    },
    {
      "auxiliary_loss_clip": 0.01113465,
      "auxiliary_loss_mlp": 0.01032503,
      "balance_loss_clip": 1.04201245,
      "balance_loss_mlp": 1.01857281,
      "epoch": 0.40895836464752744,
      "flos": 15267673664640.0,
      "grad_norm": 1.9820038036757794,
      "language_loss": 0.80693758,
      "learning_rate": 2.673395808607861e-06,
      "loss": 0.82839727,
      "num_input_tokens_seen": 145989150,
      "step": 6802,
      "time_per_iteration": 2.435746669769287
    },
    {
      "auxiliary_loss_clip": 0.01112456,
      "auxiliary_loss_mlp": 0.0103981,
      "balance_loss_clip": 1.04710305,
      "balance_loss_mlp": 1.02514112,
      "epoch": 0.4090184879001954,
      "flos": 14501411804160.0,
      "grad_norm": 3.2101755075667726,
      "language_loss": 0.76649463,
      "learning_rate": 2.673029073767934e-06,
      "loss": 0.78801727,
      "num_input_tokens_seen": 146006980,
      "step": 6803,
      "time_per_iteration": 2.5276782512664795
    },
    {
      "auxiliary_loss_clip": 0.01062701,
      "auxiliary_loss_mlp": 0.00793563,
      "balance_loss_clip": 1.0423826,
      "balance_loss_mlp": 1.01671791,
      "epoch": 0.40907861115286337,
      "flos": 13881593692800.0,
      "grad_norm": 1.978222648787648,
      "language_loss": 0.78181839,
      "learning_rate": 2.6726623134066764e-06,
      "loss": 0.80038106,
      "num_input_tokens_seen": 146025125,
      "step": 6804,
      "time_per_iteration": 2.5832555294036865
    },
    {
      "auxiliary_loss_clip": 0.01127294,
      "auxiliary_loss_mlp": 0.01039421,
      "balance_loss_clip": 1.04380643,
      "balance_loss_mlp": 1.02614665,
      "epoch": 0.40913873440553133,
      "flos": 28037615293440.0,
      "grad_norm": 1.927705507054624,
      "language_loss": 0.75525808,
      "learning_rate": 2.672295527537998e-06,
      "loss": 0.77692521,
      "num_input_tokens_seen": 146044990,
      "step": 6805,
      "time_per_iteration": 2.5083510875701904
    },
    {
      "auxiliary_loss_clip": 0.01082036,
      "auxiliary_loss_mlp": 0.0103898,
      "balance_loss_clip": 1.04336238,
      "balance_loss_mlp": 1.02562213,
      "epoch": 0.4091988576581993,
      "flos": 21618188357760.0,
      "grad_norm": 2.616054453999636,
      "language_loss": 0.79236662,
      "learning_rate": 2.671928716175804e-06,
      "loss": 0.81357682,
      "num_input_tokens_seen": 146066045,
      "step": 6806,
      "time_per_iteration": 2.598989248275757
    },
    {
      "auxiliary_loss_clip": 0.01113123,
      "auxiliary_loss_mlp": 0.01029254,
      "balance_loss_clip": 1.0419991,
      "balance_loss_mlp": 1.01550376,
      "epoch": 0.40925898091086726,
      "flos": 25224085860480.0,
      "grad_norm": 2.137445448844716,
      "language_loss": 0.71938109,
      "learning_rate": 2.671561879334007e-06,
      "loss": 0.74080491,
      "num_input_tokens_seen": 146086280,
      "step": 6807,
      "time_per_iteration": 2.54052734375
    },
    {
      "auxiliary_loss_clip": 0.01039489,
      "auxiliary_loss_mlp": 0.01000769,
      "balance_loss_clip": 1.03705525,
      "balance_loss_mlp": 0.99904013,
      "epoch": 0.40931910416353523,
      "flos": 68930568800640.0,
      "grad_norm": 0.8610202469196753,
      "language_loss": 0.58753651,
      "learning_rate": 2.6711950170265155e-06,
      "loss": 0.60793906,
      "num_input_tokens_seen": 146148840,
      "step": 6808,
      "time_per_iteration": 3.219593048095703
    },
    {
      "auxiliary_loss_clip": 0.01102554,
      "auxiliary_loss_mlp": 0.01037232,
      "balance_loss_clip": 1.04235709,
      "balance_loss_mlp": 1.02484632,
      "epoch": 0.4093792274162032,
      "flos": 20189553747840.0,
      "grad_norm": 1.5803397206915966,
      "language_loss": 0.54771984,
      "learning_rate": 2.670828129267242e-06,
      "loss": 0.56911767,
      "num_input_tokens_seen": 146166195,
      "step": 6809,
      "time_per_iteration": 2.556096315383911
    },
    {
      "auxiliary_loss_clip": 0.01099392,
      "auxiliary_loss_mlp": 0.01025969,
      "balance_loss_clip": 1.04130435,
      "balance_loss_mlp": 1.01311862,
      "epoch": 0.40943935066887116,
      "flos": 25228754628480.0,
      "grad_norm": 1.7891714746192517,
      "language_loss": 0.83166957,
      "learning_rate": 2.6704612160700983e-06,
      "loss": 0.85292315,
      "num_input_tokens_seen": 146185045,
      "step": 6810,
      "time_per_iteration": 2.551877498626709
    },
    {
      "auxiliary_loss_clip": 0.01100583,
      "auxiliary_loss_mlp": 0.01049646,
      "balance_loss_clip": 1.045609,
      "balance_loss_mlp": 1.03315938,
      "epoch": 0.4094994739215392,
      "flos": 23255319461760.0,
      "grad_norm": 2.1529406411515444,
      "language_loss": 0.77305675,
      "learning_rate": 2.670094277448999e-06,
      "loss": 0.794559,
      "num_input_tokens_seen": 146204655,
      "step": 6811,
      "time_per_iteration": 2.5452144145965576
    },
    {
      "auxiliary_loss_clip": 0.01123442,
      "auxiliary_loss_mlp": 0.01031266,
      "balance_loss_clip": 1.04289782,
      "balance_loss_mlp": 1.01678181,
      "epoch": 0.40955959717420715,
      "flos": 17382165540480.0,
      "grad_norm": 1.5487855503370782,
      "language_loss": 0.70175147,
      "learning_rate": 2.669727313417857e-06,
      "loss": 0.72329861,
      "num_input_tokens_seen": 146222000,
      "step": 6812,
      "time_per_iteration": 2.4494807720184326
    },
    {
      "auxiliary_loss_clip": 0.01121769,
      "auxiliary_loss_mlp": 0.01038102,
      "balance_loss_clip": 1.04259658,
      "balance_loss_mlp": 1.02424431,
      "epoch": 0.4096197204268751,
      "flos": 25082418620160.0,
      "grad_norm": 1.5196304780112853,
      "language_loss": 0.66595006,
      "learning_rate": 2.6693603239905872e-06,
      "loss": 0.68754888,
      "num_input_tokens_seen": 146242630,
      "step": 6813,
      "time_per_iteration": 2.5121610164642334
    },
    {
      "auxiliary_loss_clip": 0.0110715,
      "auxiliary_loss_mlp": 0.00789541,
      "balance_loss_clip": 1.04417062,
      "balance_loss_mlp": 1.0089097,
      "epoch": 0.4096798436795431,
      "flos": 30586769648640.0,
      "grad_norm": 3.6844110736672557,
      "language_loss": 0.7414341,
      "learning_rate": 2.6689933091811087e-06,
      "loss": 0.76040107,
      "num_input_tokens_seen": 146263070,
      "step": 6814,
      "time_per_iteration": 2.56007981300354
    },
    {
      "auxiliary_loss_clip": 0.01075699,
      "auxiliary_loss_mlp": 0.0103141,
      "balance_loss_clip": 1.04142237,
      "balance_loss_mlp": 1.01798081,
      "epoch": 0.40973996693221104,
      "flos": 24133622820480.0,
      "grad_norm": 2.5041423158762948,
      "language_loss": 0.66443545,
      "learning_rate": 2.6686262690033357e-06,
      "loss": 0.68550658,
      "num_input_tokens_seen": 146282890,
      "step": 6815,
      "time_per_iteration": 2.622347831726074
    },
    {
      "auxiliary_loss_clip": 0.01111054,
      "auxiliary_loss_mlp": 0.01035775,
      "balance_loss_clip": 1.04579067,
      "balance_loss_mlp": 1.02312696,
      "epoch": 0.409800090184879,
      "flos": 23988974751360.0,
      "grad_norm": 1.8516095008226958,
      "language_loss": 0.77184403,
      "learning_rate": 2.668259203471188e-06,
      "loss": 0.79331231,
      "num_input_tokens_seen": 146301755,
      "step": 6816,
      "time_per_iteration": 2.5128870010375977
    },
    {
      "auxiliary_loss_clip": 0.01106723,
      "auxiliary_loss_mlp": 0.01033997,
      "balance_loss_clip": 1.04451704,
      "balance_loss_mlp": 1.02039528,
      "epoch": 0.40986021343754697,
      "flos": 16143678552960.0,
      "grad_norm": 2.2105814782409325,
      "language_loss": 0.81357038,
      "learning_rate": 2.6678921125985843e-06,
      "loss": 0.83497763,
      "num_input_tokens_seen": 146316835,
      "step": 6817,
      "time_per_iteration": 2.534956932067871
    },
    {
      "auxiliary_loss_clip": 0.01103895,
      "auxiliary_loss_mlp": 0.0103461,
      "balance_loss_clip": 1.04257774,
      "balance_loss_mlp": 1.01911891,
      "epoch": 0.40992033669021494,
      "flos": 24790824011520.0,
      "grad_norm": 1.7061832791571057,
      "language_loss": 0.80122232,
      "learning_rate": 2.667524996399444e-06,
      "loss": 0.8226074,
      "num_input_tokens_seen": 146336650,
      "step": 6818,
      "time_per_iteration": 2.576991081237793
    },
    {
      "auxiliary_loss_clip": 0.01095707,
      "auxiliary_loss_mlp": 0.01033932,
      "balance_loss_clip": 1.0426923,
      "balance_loss_mlp": 1.02133763,
      "epoch": 0.4099804599428829,
      "flos": 29641888431360.0,
      "grad_norm": 1.6806689764510547,
      "language_loss": 0.66536307,
      "learning_rate": 2.66715785488769e-06,
      "loss": 0.68665946,
      "num_input_tokens_seen": 146357640,
      "step": 6819,
      "time_per_iteration": 2.6111834049224854
    },
    {
      "auxiliary_loss_clip": 0.01104323,
      "auxiliary_loss_mlp": 0.0104367,
      "balance_loss_clip": 1.04396415,
      "balance_loss_mlp": 1.02762437,
      "epoch": 0.41004058319555087,
      "flos": 24826590979200.0,
      "grad_norm": 1.6414848870951682,
      "language_loss": 0.85289419,
      "learning_rate": 2.6667906880772428e-06,
      "loss": 0.87437403,
      "num_input_tokens_seen": 146379325,
      "step": 6820,
      "time_per_iteration": 2.598809242248535
    },
    {
      "auxiliary_loss_clip": 0.01112333,
      "auxiliary_loss_mlp": 0.01030628,
      "balance_loss_clip": 1.04578853,
      "balance_loss_mlp": 1.01760483,
      "epoch": 0.41010070644821883,
      "flos": 25737464995200.0,
      "grad_norm": 1.8233389300208154,
      "language_loss": 0.71255922,
      "learning_rate": 2.6664234959820256e-06,
      "loss": 0.73398888,
      "num_input_tokens_seen": 146398635,
      "step": 6821,
      "time_per_iteration": 2.5405521392822266
    },
    {
      "auxiliary_loss_clip": 0.01110048,
      "auxiliary_loss_mlp": 0.01032669,
      "balance_loss_clip": 1.04651284,
      "balance_loss_mlp": 1.01956785,
      "epoch": 0.4101608297008868,
      "flos": 22346061557760.0,
      "grad_norm": 1.792796103476639,
      "language_loss": 0.74504697,
      "learning_rate": 2.6660562786159634e-06,
      "loss": 0.76647413,
      "num_input_tokens_seen": 146417585,
      "step": 6822,
      "time_per_iteration": 2.524139642715454
    },
    {
      "auxiliary_loss_clip": 0.01108344,
      "auxiliary_loss_mlp": 0.01031486,
      "balance_loss_clip": 1.04827845,
      "balance_loss_mlp": 1.01836658,
      "epoch": 0.41022095295355476,
      "flos": 21945083057280.0,
      "grad_norm": 1.9860920163740614,
      "language_loss": 0.75527006,
      "learning_rate": 2.6656890359929796e-06,
      "loss": 0.77666831,
      "num_input_tokens_seen": 146437035,
      "step": 6823,
      "time_per_iteration": 2.5477919578552246
    },
    {
      "auxiliary_loss_clip": 0.0108411,
      "auxiliary_loss_mlp": 0.01037442,
      "balance_loss_clip": 1.04988718,
      "balance_loss_mlp": 1.02182627,
      "epoch": 0.4102810762062228,
      "flos": 27450511493760.0,
      "grad_norm": 1.7612051712744778,
      "language_loss": 0.7323904,
      "learning_rate": 2.665321768127001e-06,
      "loss": 0.7536059,
      "num_input_tokens_seen": 146457370,
      "step": 6824,
      "time_per_iteration": 2.656182289123535
    },
    {
      "auxiliary_loss_clip": 0.01094954,
      "auxiliary_loss_mlp": 0.01033747,
      "balance_loss_clip": 1.04316688,
      "balance_loss_mlp": 1.0191915,
      "epoch": 0.41034119945889075,
      "flos": 24499265316480.0,
      "grad_norm": 1.7428451422392197,
      "language_loss": 0.7192505,
      "learning_rate": 2.6649544750319548e-06,
      "loss": 0.74053752,
      "num_input_tokens_seen": 146478105,
      "step": 6825,
      "time_per_iteration": 2.613664388656616
    },
    {
      "auxiliary_loss_clip": 0.01084844,
      "auxiliary_loss_mlp": 0.01037886,
      "balance_loss_clip": 1.04151332,
      "balance_loss_mlp": 1.02491605,
      "epoch": 0.4104013227115587,
      "flos": 24352641999360.0,
      "grad_norm": 2.1563217734495486,
      "language_loss": 0.85209388,
      "learning_rate": 2.664587156721768e-06,
      "loss": 0.87332118,
      "num_input_tokens_seen": 146497835,
      "step": 6826,
      "time_per_iteration": 2.5975167751312256
    },
    {
      "auxiliary_loss_clip": 0.01099989,
      "auxiliary_loss_mlp": 0.00793993,
      "balance_loss_clip": 1.04533291,
      "balance_loss_mlp": 1.01703238,
      "epoch": 0.4104614459642267,
      "flos": 23729340268800.0,
      "grad_norm": 1.6756107297261662,
      "language_loss": 0.66318536,
      "learning_rate": 2.6642198132103696e-06,
      "loss": 0.68212521,
      "num_input_tokens_seen": 146517735,
      "step": 6827,
      "time_per_iteration": 2.5744738578796387
    },
    {
      "auxiliary_loss_clip": 0.01100668,
      "auxiliary_loss_mlp": 0.01025674,
      "balance_loss_clip": 1.04376292,
      "balance_loss_mlp": 1.01281118,
      "epoch": 0.41052156921689464,
      "flos": 22127976132480.0,
      "grad_norm": 3.924311293852761,
      "language_loss": 0.71931303,
      "learning_rate": 2.663852444511689e-06,
      "loss": 0.74057645,
      "num_input_tokens_seen": 146537640,
      "step": 6828,
      "time_per_iteration": 2.5491526126861572
    },
    {
      "auxiliary_loss_clip": 0.01100632,
      "auxiliary_loss_mlp": 0.01037574,
      "balance_loss_clip": 1.04747009,
      "balance_loss_mlp": 1.02294672,
      "epoch": 0.4105816924695626,
      "flos": 20084371747200.0,
      "grad_norm": 1.7355904681060559,
      "language_loss": 0.83521616,
      "learning_rate": 2.6634850506396574e-06,
      "loss": 0.8565982,
      "num_input_tokens_seen": 146554695,
      "step": 6829,
      "time_per_iteration": 2.5204808712005615
    },
    {
      "auxiliary_loss_clip": 0.01110759,
      "auxiliary_loss_mlp": 0.01031591,
      "balance_loss_clip": 1.04394555,
      "balance_loss_mlp": 1.01863921,
      "epoch": 0.4106418157222306,
      "flos": 18076785724800.0,
      "grad_norm": 1.7185138385627095,
      "language_loss": 0.89843929,
      "learning_rate": 2.663117631608206e-06,
      "loss": 0.91986281,
      "num_input_tokens_seen": 146573740,
      "step": 6830,
      "time_per_iteration": 2.5198872089385986
    },
    {
      "auxiliary_loss_clip": 0.01087263,
      "auxiliary_loss_mlp": 0.01026375,
      "balance_loss_clip": 1.04622412,
      "balance_loss_mlp": 1.01277351,
      "epoch": 0.41070193897489854,
      "flos": 21647850013440.0,
      "grad_norm": 1.8207813101767505,
      "language_loss": 0.65579855,
      "learning_rate": 2.662750187431268e-06,
      "loss": 0.67693496,
      "num_input_tokens_seen": 146592885,
      "step": 6831,
      "time_per_iteration": 2.571640729904175
    },
    {
      "auxiliary_loss_clip": 0.01123693,
      "auxiliary_loss_mlp": 0.01033078,
      "balance_loss_clip": 1.04618764,
      "balance_loss_mlp": 1.02037668,
      "epoch": 0.4107620622275665,
      "flos": 26648195356800.0,
      "grad_norm": 1.7369554283845057,
      "language_loss": 0.69637525,
      "learning_rate": 2.662382718122776e-06,
      "loss": 0.71794295,
      "num_input_tokens_seen": 146611995,
      "step": 6832,
      "time_per_iteration": 3.908195972442627
    },
    {
      "auxiliary_loss_clip": 0.01082522,
      "auxiliary_loss_mlp": 0.01032121,
      "balance_loss_clip": 1.04936147,
      "balance_loss_mlp": 1.01956284,
      "epoch": 0.41082218548023447,
      "flos": 18734310138240.0,
      "grad_norm": 2.2055292539818185,
      "language_loss": 0.73723376,
      "learning_rate": 2.662015223696666e-06,
      "loss": 0.75838017,
      "num_input_tokens_seen": 146628045,
      "step": 6833,
      "time_per_iteration": 2.5590522289276123
    },
    {
      "auxiliary_loss_clip": 0.01074915,
      "auxiliary_loss_mlp": 0.01034486,
      "balance_loss_clip": 1.04267335,
      "balance_loss_mlp": 1.01935828,
      "epoch": 0.41088230873290243,
      "flos": 22893771116160.0,
      "grad_norm": 1.5932504281740032,
      "language_loss": 0.72692162,
      "learning_rate": 2.6616477041668713e-06,
      "loss": 0.74801564,
      "num_input_tokens_seen": 146648355,
      "step": 6834,
      "time_per_iteration": 2.5925230979919434
    },
    {
      "auxiliary_loss_clip": 0.01115767,
      "auxiliary_loss_mlp": 0.01038245,
      "balance_loss_clip": 1.04472208,
      "balance_loss_mlp": 1.02424979,
      "epoch": 0.4109424319855704,
      "flos": 24276978000000.0,
      "grad_norm": 1.698842359710795,
      "language_loss": 0.71202642,
      "learning_rate": 2.661280159547329e-06,
      "loss": 0.73356652,
      "num_input_tokens_seen": 146668370,
      "step": 6835,
      "time_per_iteration": 3.9460394382476807
    },
    {
      "auxiliary_loss_clip": 0.01116768,
      "auxiliary_loss_mlp": 0.01033941,
      "balance_loss_clip": 1.04589498,
      "balance_loss_mlp": 1.01917648,
      "epoch": 0.41100255523823837,
      "flos": 12969139478400.0,
      "grad_norm": 2.3589321784292916,
      "language_loss": 0.86749721,
      "learning_rate": 2.660912589851978e-06,
      "loss": 0.88900423,
      "num_input_tokens_seen": 146686665,
      "step": 6836,
      "time_per_iteration": 3.868922710418701
    },
    {
      "auxiliary_loss_clip": 0.01111607,
      "auxiliary_loss_mlp": 0.01029837,
      "balance_loss_clip": 1.04650283,
      "balance_loss_mlp": 1.01687849,
      "epoch": 0.4110626784909064,
      "flos": 23145648261120.0,
      "grad_norm": 2.017286582842602,
      "language_loss": 0.69027245,
      "learning_rate": 2.6605449950947547e-06,
      "loss": 0.71168691,
      "num_input_tokens_seen": 146706570,
      "step": 6837,
      "time_per_iteration": 3.934495687484741
    },
    {
      "auxiliary_loss_clip": 0.01127574,
      "auxiliary_loss_mlp": 0.01033141,
      "balance_loss_clip": 1.04653442,
      "balance_loss_mlp": 1.01869333,
      "epoch": 0.41112280174357435,
      "flos": 22747399194240.0,
      "grad_norm": 2.1667167879120472,
      "language_loss": 0.74920022,
      "learning_rate": 2.660177375289599e-06,
      "loss": 0.77080739,
      "num_input_tokens_seen": 146723425,
      "step": 6838,
      "time_per_iteration": 2.4857592582702637
    },
    {
      "auxiliary_loss_clip": 0.01089753,
      "auxiliary_loss_mlp": 0.01034597,
      "balance_loss_clip": 1.04822397,
      "balance_loss_mlp": 1.02061963,
      "epoch": 0.4111829249962423,
      "flos": 21102403011840.0,
      "grad_norm": 2.333563716313449,
      "language_loss": 0.8221873,
      "learning_rate": 2.659809730450451e-06,
      "loss": 0.84343088,
      "num_input_tokens_seen": 146741640,
      "step": 6839,
      "time_per_iteration": 2.5463900566101074
    },
    {
      "auxiliary_loss_clip": 0.01122073,
      "auxiliary_loss_mlp": 0.01030085,
      "balance_loss_clip": 1.04356182,
      "balance_loss_mlp": 1.01705003,
      "epoch": 0.4112430482489103,
      "flos": 21505787723520.0,
      "grad_norm": 1.9892596638702686,
      "language_loss": 0.80314952,
      "learning_rate": 2.6594420605912523e-06,
      "loss": 0.82467115,
      "num_input_tokens_seen": 146759195,
      "step": 6840,
      "time_per_iteration": 2.4678094387054443
    },
    {
      "auxiliary_loss_clip": 0.01110868,
      "auxiliary_loss_mlp": 0.01030905,
      "balance_loss_clip": 1.0462116,
      "balance_loss_mlp": 1.01845336,
      "epoch": 0.41130317150157825,
      "flos": 19570022945280.0,
      "grad_norm": 1.9255532363099326,
      "language_loss": 0.67520368,
      "learning_rate": 2.6590743657259442e-06,
      "loss": 0.69662136,
      "num_input_tokens_seen": 146774990,
      "step": 6841,
      "time_per_iteration": 2.4816668033599854
    },
    {
      "auxiliary_loss_clip": 0.01047978,
      "auxiliary_loss_mlp": 0.01004488,
      "balance_loss_clip": 1.02940571,
      "balance_loss_mlp": 1.00285482,
      "epoch": 0.4113632947542462,
      "flos": 62383157706240.0,
      "grad_norm": 0.7603981797507843,
      "language_loss": 0.59639072,
      "learning_rate": 2.65870664586847e-06,
      "loss": 0.61691535,
      "num_input_tokens_seen": 146839610,
      "step": 6842,
      "time_per_iteration": 3.179201602935791
    },
    {
      "auxiliary_loss_clip": 0.01104935,
      "auxiliary_loss_mlp": 0.01028371,
      "balance_loss_clip": 1.04531646,
      "balance_loss_mlp": 1.01650929,
      "epoch": 0.4114234180069142,
      "flos": 13918617636480.0,
      "grad_norm": 2.0712292655551052,
      "language_loss": 0.70276594,
      "learning_rate": 2.6583389010327742e-06,
      "loss": 0.72409892,
      "num_input_tokens_seen": 146857360,
      "step": 6843,
      "time_per_iteration": 2.494269847869873
    },
    {
      "auxiliary_loss_clip": 0.01028017,
      "auxiliary_loss_mlp": 0.01008377,
      "balance_loss_clip": 1.03678036,
      "balance_loss_mlp": 1.00676811,
      "epoch": 0.41148354125958214,
      "flos": 64928505219840.0,
      "grad_norm": 0.7242240471852321,
      "language_loss": 0.53604352,
      "learning_rate": 2.6579711312328013e-06,
      "loss": 0.55640751,
      "num_input_tokens_seen": 146917055,
      "step": 6844,
      "time_per_iteration": 3.162383556365967
    },
    {
      "auxiliary_loss_clip": 0.01110926,
      "auxiliary_loss_mlp": 0.01031139,
      "balance_loss_clip": 1.04525852,
      "balance_loss_mlp": 1.01849675,
      "epoch": 0.4115436645122501,
      "flos": 18728779443840.0,
      "grad_norm": 3.9492768586613938,
      "language_loss": 0.66072816,
      "learning_rate": 2.6576033364824967e-06,
      "loss": 0.68214881,
      "num_input_tokens_seen": 146935215,
      "step": 6845,
      "time_per_iteration": 2.485123872756958
    },
    {
      "auxiliary_loss_clip": 0.01123201,
      "auxiliary_loss_mlp": 0.01033014,
      "balance_loss_clip": 1.04721308,
      "balance_loss_mlp": 1.01996684,
      "epoch": 0.41160378776491807,
      "flos": 16252918790400.0,
      "grad_norm": 1.9912357407064933,
      "language_loss": 0.70279056,
      "learning_rate": 2.657235516795808e-06,
      "loss": 0.72435272,
      "num_input_tokens_seen": 146951970,
      "step": 6846,
      "time_per_iteration": 2.4598653316497803
    },
    {
      "auxiliary_loss_clip": 0.01104885,
      "auxiliary_loss_mlp": 0.01035329,
      "balance_loss_clip": 1.04746401,
      "balance_loss_mlp": 1.02154231,
      "epoch": 0.41166391101758604,
      "flos": 27970031854080.0,
      "grad_norm": 1.8455600018088398,
      "language_loss": 0.65078241,
      "learning_rate": 2.6568676721866826e-06,
      "loss": 0.67218459,
      "num_input_tokens_seen": 146975615,
      "step": 6847,
      "time_per_iteration": 2.629786491394043
    },
    {
      "auxiliary_loss_clip": 0.01104074,
      "auxiliary_loss_mlp": 0.01036046,
      "balance_loss_clip": 1.04601502,
      "balance_loss_mlp": 1.02267075,
      "epoch": 0.411724034270254,
      "flos": 34131296764800.0,
      "grad_norm": 1.6154477787203712,
      "language_loss": 0.7042883,
      "learning_rate": 2.656499802669069e-06,
      "loss": 0.72568953,
      "num_input_tokens_seen": 146998855,
      "step": 6848,
      "time_per_iteration": 2.682521104812622
    },
    {
      "auxiliary_loss_clip": 0.01028835,
      "auxiliary_loss_mlp": 0.00848663,
      "balance_loss_clip": 1.02096295,
      "balance_loss_mlp": 1.14628458,
      "epoch": 0.41178415752292197,
      "flos": 67923670752000.0,
      "grad_norm": 0.8921829925355241,
      "language_loss": 0.56203073,
      "learning_rate": 2.6561319082569174e-06,
      "loss": 0.58080572,
      "num_input_tokens_seen": 147062710,
      "step": 6849,
      "time_per_iteration": 3.2218313217163086
    },
    {
      "auxiliary_loss_clip": 0.01102419,
      "auxiliary_loss_mlp": 0.01038409,
      "balance_loss_clip": 1.04372036,
      "balance_loss_mlp": 1.02481949,
      "epoch": 0.41184428077558993,
      "flos": 34313938444800.0,
      "grad_norm": 1.7566024642783675,
      "language_loss": 0.76139331,
      "learning_rate": 2.6557639889641783e-06,
      "loss": 0.78280163,
      "num_input_tokens_seen": 147086075,
      "step": 6850,
      "time_per_iteration": 2.682208776473999
    },
    {
      "auxiliary_loss_clip": 0.01073317,
      "auxiliary_loss_mlp": 0.01030687,
      "balance_loss_clip": 1.04157186,
      "balance_loss_mlp": 1.01805067,
      "epoch": 0.41190440402825795,
      "flos": 35444118948480.0,
      "grad_norm": 1.5472619468632312,
      "language_loss": 0.67980599,
      "learning_rate": 2.6553960448048025e-06,
      "loss": 0.70084596,
      "num_input_tokens_seen": 147107590,
      "step": 6851,
      "time_per_iteration": 2.7286689281463623
    },
    {
      "auxiliary_loss_clip": 0.01088847,
      "auxiliary_loss_mlp": 0.01039864,
      "balance_loss_clip": 1.04607749,
      "balance_loss_mlp": 1.02432525,
      "epoch": 0.4119645272809259,
      "flos": 20849879422080.0,
      "grad_norm": 2.2495081894546582,
      "language_loss": 0.79589081,
      "learning_rate": 2.655028075792743e-06,
      "loss": 0.81717789,
      "num_input_tokens_seen": 147123715,
      "step": 6852,
      "time_per_iteration": 2.5486955642700195
    },
    {
      "auxiliary_loss_clip": 0.01129039,
      "auxiliary_loss_mlp": 0.01035378,
      "balance_loss_clip": 1.04753792,
      "balance_loss_mlp": 1.0209657,
      "epoch": 0.4120246505335939,
      "flos": 27562050201600.0,
      "grad_norm": 2.1397965442679063,
      "language_loss": 0.77571726,
      "learning_rate": 2.6546600819419537e-06,
      "loss": 0.79736137,
      "num_input_tokens_seen": 147144290,
      "step": 6853,
      "time_per_iteration": 2.551239252090454
    },
    {
      "auxiliary_loss_clip": 0.01116806,
      "auxiliary_loss_mlp": 0.01044359,
      "balance_loss_clip": 1.04492426,
      "balance_loss_mlp": 1.02869487,
      "epoch": 0.41208477378626185,
      "flos": 37815444046080.0,
      "grad_norm": 1.6755801738696414,
      "language_loss": 0.65665245,
      "learning_rate": 2.6542920632663883e-06,
      "loss": 0.67826414,
      "num_input_tokens_seen": 147166340,
      "step": 6854,
      "time_per_iteration": 2.6570117473602295
    },
    {
      "auxiliary_loss_clip": 0.01095614,
      "auxiliary_loss_mlp": 0.0103941,
      "balance_loss_clip": 1.04253209,
      "balance_loss_mlp": 1.02501535,
      "epoch": 0.4121448970389298,
      "flos": 23440762402560.0,
      "grad_norm": 1.6990811875700573,
      "language_loss": 0.83648288,
      "learning_rate": 2.6539240197800023e-06,
      "loss": 0.85783309,
      "num_input_tokens_seen": 147184025,
      "step": 6855,
      "time_per_iteration": 2.5852184295654297
    },
    {
      "auxiliary_loss_clip": 0.01105159,
      "auxiliary_loss_mlp": 0.01040965,
      "balance_loss_clip": 1.0444721,
      "balance_loss_mlp": 1.02813804,
      "epoch": 0.4122050202915978,
      "flos": 21325300859520.0,
      "grad_norm": 1.5795301505165584,
      "language_loss": 0.79185343,
      "learning_rate": 2.6535559514967517e-06,
      "loss": 0.81331468,
      "num_input_tokens_seen": 147202730,
      "step": 6856,
      "time_per_iteration": 2.5122902393341064
    },
    {
      "auxiliary_loss_clip": 0.01088115,
      "auxiliary_loss_mlp": 0.01035923,
      "balance_loss_clip": 1.04366636,
      "balance_loss_mlp": 1.0227859,
      "epoch": 0.41226514354426574,
      "flos": 17306286059520.0,
      "grad_norm": 2.4052220252179137,
      "language_loss": 0.80161184,
      "learning_rate": 2.6531878584305935e-06,
      "loss": 0.82285219,
      "num_input_tokens_seen": 147215315,
      "step": 6857,
      "time_per_iteration": 2.5169365406036377
    },
    {
      "auxiliary_loss_clip": 0.01112506,
      "auxiliary_loss_mlp": 0.00836618,
      "balance_loss_clip": 1.04361629,
      "balance_loss_mlp": 1.09867477,
      "epoch": 0.4123252667969337,
      "flos": 17638855107840.0,
      "grad_norm": 1.7040918693998164,
      "language_loss": 0.70627534,
      "learning_rate": 2.6528197405954873e-06,
      "loss": 0.7257666,
      "num_input_tokens_seen": 147233330,
      "step": 6858,
      "time_per_iteration": 2.513631820678711
    },
    {
      "auxiliary_loss_clip": 0.01110818,
      "auxiliary_loss_mlp": 0.01035779,
      "balance_loss_clip": 1.04442716,
      "balance_loss_mlp": 1.02176607,
      "epoch": 0.4123853900496017,
      "flos": 46424811375360.0,
      "grad_norm": 1.4561594251293666,
      "language_loss": 0.59441072,
      "learning_rate": 2.652451598005391e-06,
      "loss": 0.61587667,
      "num_input_tokens_seen": 147257780,
      "step": 6859,
      "time_per_iteration": 2.7461447715759277
    },
    {
      "auxiliary_loss_clip": 0.0112324,
      "auxiliary_loss_mlp": 0.01037139,
      "balance_loss_clip": 1.04316044,
      "balance_loss_mlp": 1.02363849,
      "epoch": 0.41244551330226964,
      "flos": 17675160779520.0,
      "grad_norm": 2.341560675076963,
      "language_loss": 0.73484683,
      "learning_rate": 2.652083430674264e-06,
      "loss": 0.75645065,
      "num_input_tokens_seen": 147276055,
      "step": 6860,
      "time_per_iteration": 2.517282247543335
    },
    {
      "auxiliary_loss_clip": 0.01043564,
      "auxiliary_loss_mlp": 0.01034663,
      "balance_loss_clip": 1.03803754,
      "balance_loss_mlp": 1.02111435,
      "epoch": 0.4125056365549376,
      "flos": 18693730748160.0,
      "grad_norm": 1.6912922910633825,
      "language_loss": 0.7421335,
      "learning_rate": 2.651715238616068e-06,
      "loss": 0.76291579,
      "num_input_tokens_seen": 147293200,
      "step": 6861,
      "time_per_iteration": 2.6245903968811035
    },
    {
      "auxiliary_loss_clip": 0.01099151,
      "auxiliary_loss_mlp": 0.01032094,
      "balance_loss_clip": 1.04199648,
      "balance_loss_mlp": 1.01949382,
      "epoch": 0.41256575980760557,
      "flos": 17895293280000.0,
      "grad_norm": 2.567680720458047,
      "language_loss": 0.79717612,
      "learning_rate": 2.651347021844765e-06,
      "loss": 0.8184886,
      "num_input_tokens_seen": 147310640,
      "step": 6862,
      "time_per_iteration": 2.4875423908233643
    },
    {
      "auxiliary_loss_clip": 0.01097089,
      "auxiliary_loss_mlp": 0.01032981,
      "balance_loss_clip": 1.0434804,
      "balance_loss_mlp": 1.01957035,
      "epoch": 0.41262588306027354,
      "flos": 21981316901760.0,
      "grad_norm": 1.7170241717266996,
      "language_loss": 0.76085299,
      "learning_rate": 2.650978780374318e-06,
      "loss": 0.78215373,
      "num_input_tokens_seen": 147329435,
      "step": 6863,
      "time_per_iteration": 2.5321695804595947
    },
    {
      "auxiliary_loss_clip": 0.01043645,
      "auxiliary_loss_mlp": 0.01008151,
      "balance_loss_clip": 1.02561665,
      "balance_loss_mlp": 1.00648177,
      "epoch": 0.41268600631294156,
      "flos": 53350006740480.0,
      "grad_norm": 0.6976001681477471,
      "language_loss": 0.5270859,
      "learning_rate": 2.650610514218691e-06,
      "loss": 0.54760385,
      "num_input_tokens_seen": 147385805,
      "step": 6864,
      "time_per_iteration": 3.0737171173095703
    },
    {
      "auxiliary_loss_clip": 0.01126081,
      "auxiliary_loss_mlp": 0.01034041,
      "balance_loss_clip": 1.04352593,
      "balance_loss_mlp": 1.01958036,
      "epoch": 0.4127461295656095,
      "flos": 24385356311040.0,
      "grad_norm": 1.7658898685534397,
      "language_loss": 0.72996247,
      "learning_rate": 2.6502422233918468e-06,
      "loss": 0.75156379,
      "num_input_tokens_seen": 147405160,
      "step": 6865,
      "time_per_iteration": 2.506314277648926
    },
    {
      "auxiliary_loss_clip": 0.01050747,
      "auxiliary_loss_mlp": 0.01008981,
      "balance_loss_clip": 1.02269411,
      "balance_loss_mlp": 1.00744367,
      "epoch": 0.4128062528182775,
      "flos": 71705242696320.0,
      "grad_norm": 0.9326349510747677,
      "language_loss": 0.66522324,
      "learning_rate": 2.649873907907753e-06,
      "loss": 0.68582046,
      "num_input_tokens_seen": 147460245,
      "step": 6866,
      "time_per_iteration": 2.9604804515838623
    },
    {
      "auxiliary_loss_clip": 0.0111944,
      "auxiliary_loss_mlp": 0.01031143,
      "balance_loss_clip": 1.0420177,
      "balance_loss_mlp": 1.01818466,
      "epoch": 0.41286637607094545,
      "flos": 17849111368320.0,
      "grad_norm": 2.42395948972015,
      "language_loss": 0.81215298,
      "learning_rate": 2.649505567780375e-06,
      "loss": 0.83365875,
      "num_input_tokens_seen": 147476200,
      "step": 6867,
      "time_per_iteration": 2.4431099891662598
    },
    {
      "auxiliary_loss_clip": 0.01105634,
      "auxiliary_loss_mlp": 0.01031835,
      "balance_loss_clip": 1.0458138,
      "balance_loss_mlp": 1.01785743,
      "epoch": 0.4129264993236134,
      "flos": 25549544016000.0,
      "grad_norm": 2.2593208860925635,
      "language_loss": 0.7712326,
      "learning_rate": 2.6491372030236815e-06,
      "loss": 0.79260725,
      "num_input_tokens_seen": 147494315,
      "step": 6868,
      "time_per_iteration": 2.579080820083618
    },
    {
      "auxiliary_loss_clip": 0.01034617,
      "auxiliary_loss_mlp": 0.01003654,
      "balance_loss_clip": 1.01773357,
      "balance_loss_mlp": 1.00219965,
      "epoch": 0.4129866225762814,
      "flos": 65414446364160.0,
      "grad_norm": 0.8295761864278823,
      "language_loss": 0.57795477,
      "learning_rate": 2.64876881365164e-06,
      "loss": 0.59833741,
      "num_input_tokens_seen": 147543665,
      "step": 6869,
      "time_per_iteration": 2.8623619079589844
    },
    {
      "auxiliary_loss_clip": 0.01110171,
      "auxiliary_loss_mlp": 0.01030627,
      "balance_loss_clip": 1.04470897,
      "balance_loss_mlp": 1.01681638,
      "epoch": 0.41304674582894935,
      "flos": 28876991287680.0,
      "grad_norm": 1.79069044080286,
      "language_loss": 0.75388211,
      "learning_rate": 2.64840039967822e-06,
      "loss": 0.77529013,
      "num_input_tokens_seen": 147564870,
      "step": 6870,
      "time_per_iteration": 2.573387622833252
    },
    {
      "auxiliary_loss_clip": 0.01092789,
      "auxiliary_loss_mlp": 0.01035969,
      "balance_loss_clip": 1.04457641,
      "balance_loss_mlp": 1.02137804,
      "epoch": 0.4131068690816173,
      "flos": 22891975436160.0,
      "grad_norm": 1.782596410138529,
      "language_loss": 0.83561182,
      "learning_rate": 2.6480319611173912e-06,
      "loss": 0.85689938,
      "num_input_tokens_seen": 147584840,
      "step": 6871,
      "time_per_iteration": 4.014943838119507
    },
    {
      "auxiliary_loss_clip": 0.01093322,
      "auxiliary_loss_mlp": 0.01038259,
      "balance_loss_clip": 1.04518151,
      "balance_loss_mlp": 1.0243237,
      "epoch": 0.4131669923342853,
      "flos": 26065185707520.0,
      "grad_norm": 1.8799247628132616,
      "language_loss": 0.68375415,
      "learning_rate": 2.6476634979831263e-06,
      "loss": 0.70506996,
      "num_input_tokens_seen": 147604635,
      "step": 6872,
      "time_per_iteration": 2.613799810409546
    },
    {
      "auxiliary_loss_clip": 0.01107358,
      "auxiliary_loss_mlp": 0.01030945,
      "balance_loss_clip": 1.04393148,
      "balance_loss_mlp": 1.01733088,
      "epoch": 0.41322711558695324,
      "flos": 19244564789760.0,
      "grad_norm": 1.930683615824224,
      "language_loss": 0.75856042,
      "learning_rate": 2.6472950102893964e-06,
      "loss": 0.77994347,
      "num_input_tokens_seen": 147620700,
      "step": 6873,
      "time_per_iteration": 2.507756233215332
    },
    {
      "auxiliary_loss_clip": 0.01100487,
      "auxiliary_loss_mlp": 0.01034354,
      "balance_loss_clip": 1.04543352,
      "balance_loss_mlp": 1.02023411,
      "epoch": 0.4132872388396212,
      "flos": 22674464628480.0,
      "grad_norm": 2.3789708471696915,
      "language_loss": 0.83484882,
      "learning_rate": 2.6469264980501746e-06,
      "loss": 0.85619724,
      "num_input_tokens_seen": 147639490,
      "step": 6874,
      "time_per_iteration": 3.948176145553589
    },
    {
      "auxiliary_loss_clip": 0.01087691,
      "auxiliary_loss_mlp": 0.01032758,
      "balance_loss_clip": 1.04063988,
      "balance_loss_mlp": 1.01833367,
      "epoch": 0.4133473620922892,
      "flos": 20150195420160.0,
      "grad_norm": 1.8888236725794347,
      "language_loss": 0.71631092,
      "learning_rate": 2.646557961279436e-06,
      "loss": 0.73751539,
      "num_input_tokens_seen": 147657205,
      "step": 6875,
      "time_per_iteration": 3.965930223464966
    },
    {
      "auxiliary_loss_clip": 0.01091449,
      "auxiliary_loss_mlp": 0.01035514,
      "balance_loss_clip": 1.04151177,
      "balance_loss_mlp": 1.02225256,
      "epoch": 0.41340748534495714,
      "flos": 24242755317120.0,
      "grad_norm": 1.7486906592704157,
      "language_loss": 0.82589626,
      "learning_rate": 2.646189399991154e-06,
      "loss": 0.84716588,
      "num_input_tokens_seen": 147677005,
      "step": 6876,
      "time_per_iteration": 3.9192960262298584
    },
    {
      "auxiliary_loss_clip": 0.01111288,
      "auxiliary_loss_mlp": 0.01039671,
      "balance_loss_clip": 1.04471803,
      "balance_loss_mlp": 1.02372026,
      "epoch": 0.41346760859762516,
      "flos": 14392171566720.0,
      "grad_norm": 2.339931730187285,
      "language_loss": 0.65881693,
      "learning_rate": 2.6458208141993048e-06,
      "loss": 0.68032658,
      "num_input_tokens_seen": 147693435,
      "step": 6877,
      "time_per_iteration": 2.4733798503875732
    },
    {
      "auxiliary_loss_clip": 0.01109974,
      "auxiliary_loss_mlp": 0.01033887,
      "balance_loss_clip": 1.04372263,
      "balance_loss_mlp": 1.02010632,
      "epoch": 0.4135277318502931,
      "flos": 22492002516480.0,
      "grad_norm": 1.731488260420566,
      "language_loss": 0.76934761,
      "learning_rate": 2.6454522039178668e-06,
      "loss": 0.79078621,
      "num_input_tokens_seen": 147714000,
      "step": 6878,
      "time_per_iteration": 2.5016794204711914
    },
    {
      "auxiliary_loss_clip": 0.01113468,
      "auxiliary_loss_mlp": 0.00822204,
      "balance_loss_clip": 1.04561949,
      "balance_loss_mlp": 1.07058918,
      "epoch": 0.4135878551029611,
      "flos": 22418744728320.0,
      "grad_norm": 1.8305258101647182,
      "language_loss": 0.80112165,
      "learning_rate": 2.6450835691608154e-06,
      "loss": 0.82047838,
      "num_input_tokens_seen": 147731010,
      "step": 6879,
      "time_per_iteration": 2.5050699710845947
    },
    {
      "auxiliary_loss_clip": 0.01122924,
      "auxiliary_loss_mlp": 0.0103613,
      "balance_loss_clip": 1.04474282,
      "balance_loss_mlp": 1.02218223,
      "epoch": 0.41364797835562905,
      "flos": 27053232094080.0,
      "grad_norm": 1.7352500546870113,
      "language_loss": 0.84546053,
      "learning_rate": 2.6447149099421315e-06,
      "loss": 0.86705112,
      "num_input_tokens_seen": 147750880,
      "step": 6880,
      "time_per_iteration": 2.511765480041504
    },
    {
      "auxiliary_loss_clip": 0.01101355,
      "auxiliary_loss_mlp": 0.01027837,
      "balance_loss_clip": 1.04330802,
      "balance_loss_mlp": 1.01408076,
      "epoch": 0.413708101608297,
      "flos": 22967603521920.0,
      "grad_norm": 1.5420282936339662,
      "language_loss": 0.70583749,
      "learning_rate": 2.6443462262757927e-06,
      "loss": 0.72712934,
      "num_input_tokens_seen": 147771360,
      "step": 6881,
      "time_per_iteration": 2.5307204723358154
    },
    {
      "auxiliary_loss_clip": 0.01121507,
      "auxiliary_loss_mlp": 0.01033888,
      "balance_loss_clip": 1.04558444,
      "balance_loss_mlp": 1.02107906,
      "epoch": 0.413768224860965,
      "flos": 13333991875200.0,
      "grad_norm": 1.7579162772761865,
      "language_loss": 0.81531429,
      "learning_rate": 2.6439775181757805e-06,
      "loss": 0.83686829,
      "num_input_tokens_seen": 147787440,
      "step": 6882,
      "time_per_iteration": 2.4370205402374268
    },
    {
      "auxiliary_loss_clip": 0.0110076,
      "auxiliary_loss_mlp": 0.01040632,
      "balance_loss_clip": 1.04251206,
      "balance_loss_mlp": 1.02432466,
      "epoch": 0.41382834811363295,
      "flos": 20813968800000.0,
      "grad_norm": 2.1006013023918166,
      "language_loss": 0.69668627,
      "learning_rate": 2.643608785656077e-06,
      "loss": 0.71810019,
      "num_input_tokens_seen": 147805720,
      "step": 6883,
      "time_per_iteration": 2.527916669845581
    },
    {
      "auxiliary_loss_clip": 0.01113236,
      "auxiliary_loss_mlp": 0.01030275,
      "balance_loss_clip": 1.04398823,
      "balance_loss_mlp": 1.01675677,
      "epoch": 0.4138884713663009,
      "flos": 20667130001280.0,
      "grad_norm": 1.921822025323105,
      "language_loss": 0.75513142,
      "learning_rate": 2.643240028730663e-06,
      "loss": 0.77656651,
      "num_input_tokens_seen": 147824605,
      "step": 6884,
      "time_per_iteration": 2.5019338130950928
    },
    {
      "auxiliary_loss_clip": 0.01096601,
      "auxiliary_loss_mlp": 0.01035224,
      "balance_loss_clip": 1.04711318,
      "balance_loss_mlp": 1.02186084,
      "epoch": 0.4139485946189689,
      "flos": 29056616225280.0,
      "grad_norm": 1.3987079184680344,
      "language_loss": 0.75740409,
      "learning_rate": 2.642871247413523e-06,
      "loss": 0.77872241,
      "num_input_tokens_seen": 147845445,
      "step": 6885,
      "time_per_iteration": 2.6034467220306396
    },
    {
      "auxiliary_loss_clip": 0.01125499,
      "auxiliary_loss_mlp": 0.01038549,
      "balance_loss_clip": 1.04464841,
      "balance_loss_mlp": 1.02481627,
      "epoch": 0.41400871787163684,
      "flos": 24425720219520.0,
      "grad_norm": 1.8016900591452027,
      "language_loss": 0.69951677,
      "learning_rate": 2.6425024417186414e-06,
      "loss": 0.72115725,
      "num_input_tokens_seen": 147865580,
      "step": 6886,
      "time_per_iteration": 2.4895639419555664
    },
    {
      "auxiliary_loss_clip": 0.01123917,
      "auxiliary_loss_mlp": 0.00805974,
      "balance_loss_clip": 1.04437947,
      "balance_loss_mlp": 1.03781629,
      "epoch": 0.4140688411243048,
      "flos": 19464050845440.0,
      "grad_norm": 1.545613378319779,
      "language_loss": 0.75476706,
      "learning_rate": 2.642133611660002e-06,
      "loss": 0.77406597,
      "num_input_tokens_seen": 147885230,
      "step": 6887,
      "time_per_iteration": 2.457165002822876
    },
    {
      "auxiliary_loss_clip": 0.01110131,
      "auxiliary_loss_mlp": 0.01030094,
      "balance_loss_clip": 1.04163003,
      "balance_loss_mlp": 1.01616991,
      "epoch": 0.4141289643769728,
      "flos": 19313656600320.0,
      "grad_norm": 2.1280914559769504,
      "language_loss": 0.70327264,
      "learning_rate": 2.641764757251592e-06,
      "loss": 0.72467488,
      "num_input_tokens_seen": 147903035,
      "step": 6888,
      "time_per_iteration": 2.474846363067627
    },
    {
      "auxiliary_loss_clip": 0.01119953,
      "auxiliary_loss_mlp": 0.01035321,
      "balance_loss_clip": 1.04201293,
      "balance_loss_mlp": 1.02117038,
      "epoch": 0.41418908762964074,
      "flos": 16726903683840.0,
      "grad_norm": 2.0661678781746162,
      "language_loss": 0.75710249,
      "learning_rate": 2.6413958785073976e-06,
      "loss": 0.77865517,
      "num_input_tokens_seen": 147918745,
      "step": 6889,
      "time_per_iteration": 2.426698923110962
    },
    {
      "auxiliary_loss_clip": 0.01092098,
      "auxiliary_loss_mlp": 0.00799513,
      "balance_loss_clip": 1.04875755,
      "balance_loss_mlp": 1.02828777,
      "epoch": 0.41424921088230876,
      "flos": 25296840858240.0,
      "grad_norm": 1.5613438952495111,
      "language_loss": 0.80223608,
      "learning_rate": 2.6410269754414074e-06,
      "loss": 0.82115215,
      "num_input_tokens_seen": 147938265,
      "step": 6890,
      "time_per_iteration": 2.603952407836914
    },
    {
      "auxiliary_loss_clip": 0.01120235,
      "auxiliary_loss_mlp": 0.01037146,
      "balance_loss_clip": 1.04329252,
      "balance_loss_mlp": 1.02314472,
      "epoch": 0.4143093341349767,
      "flos": 20960520289920.0,
      "grad_norm": 1.5619868362425275,
      "language_loss": 0.73876506,
      "learning_rate": 2.6406580480676113e-06,
      "loss": 0.7603389,
      "num_input_tokens_seen": 147957320,
      "step": 6891,
      "time_per_iteration": 2.4755373001098633
    },
    {
      "auxiliary_loss_clip": 0.01082813,
      "auxiliary_loss_mlp": 0.01040406,
      "balance_loss_clip": 1.04322672,
      "balance_loss_mlp": 1.02428269,
      "epoch": 0.4143694573876447,
      "flos": 22017694400640.0,
      "grad_norm": 1.6235253417218969,
      "language_loss": 0.8433007,
      "learning_rate": 2.6402890963999963e-06,
      "loss": 0.86453283,
      "num_input_tokens_seen": 147977045,
      "step": 6892,
      "time_per_iteration": 2.5881948471069336
    },
    {
      "auxiliary_loss_clip": 0.01077008,
      "auxiliary_loss_mlp": 0.00799344,
      "balance_loss_clip": 1.04013932,
      "balance_loss_mlp": 1.02264488,
      "epoch": 0.41442958064031266,
      "flos": 35697396723840.0,
      "grad_norm": 4.418312530801412,
      "language_loss": 0.70536613,
      "learning_rate": 2.6399201204525554e-06,
      "loss": 0.72412968,
      "num_input_tokens_seen": 147996905,
      "step": 6893,
      "time_per_iteration": 2.696990966796875
    },
    {
      "auxiliary_loss_clip": 0.01120416,
      "auxiliary_loss_mlp": 0.01029217,
      "balance_loss_clip": 1.04281068,
      "balance_loss_mlp": 1.01587784,
      "epoch": 0.4144897038929806,
      "flos": 28293766156800.0,
      "grad_norm": 1.4134198980284547,
      "language_loss": 0.73072726,
      "learning_rate": 2.639551120239279e-06,
      "loss": 0.75222361,
      "num_input_tokens_seen": 148017875,
      "step": 6894,
      "time_per_iteration": 2.5443406105041504
    },
    {
      "auxiliary_loss_clip": 0.0111241,
      "auxiliary_loss_mlp": 0.01032425,
      "balance_loss_clip": 1.04221773,
      "balance_loss_mlp": 1.01915717,
      "epoch": 0.4145498271456486,
      "flos": 11648093080320.0,
      "grad_norm": 2.817104185902053,
      "language_loss": 0.63498086,
      "learning_rate": 2.63918209577416e-06,
      "loss": 0.65642923,
      "num_input_tokens_seen": 148032300,
      "step": 6895,
      "time_per_iteration": 2.4575624465942383
    },
    {
      "auxiliary_loss_clip": 0.01079516,
      "auxiliary_loss_mlp": 0.01040907,
      "balance_loss_clip": 1.0429709,
      "balance_loss_mlp": 1.02577388,
      "epoch": 0.41460995039831655,
      "flos": 27235622378880.0,
      "grad_norm": 1.4937700037116637,
      "language_loss": 0.70880163,
      "learning_rate": 2.638813047071192e-06,
      "loss": 0.73000586,
      "num_input_tokens_seen": 148053260,
      "step": 6896,
      "time_per_iteration": 2.6094095706939697
    },
    {
      "auxiliary_loss_clip": 0.01123476,
      "auxiliary_loss_mlp": 0.01040601,
      "balance_loss_clip": 1.04251945,
      "balance_loss_mlp": 1.02498472,
      "epoch": 0.4146700736509845,
      "flos": 25922369232000.0,
      "grad_norm": 1.745138000044116,
      "language_loss": 0.72874725,
      "learning_rate": 2.6384439741443696e-06,
      "loss": 0.75038797,
      "num_input_tokens_seen": 148072965,
      "step": 6897,
      "time_per_iteration": 2.5344982147216797
    },
    {
      "auxiliary_loss_clip": 0.01110051,
      "auxiliary_loss_mlp": 0.01042493,
      "balance_loss_clip": 1.04514027,
      "balance_loss_mlp": 1.02861714,
      "epoch": 0.4147301969036525,
      "flos": 26833243248000.0,
      "grad_norm": 1.8247778571078017,
      "language_loss": 0.84545344,
      "learning_rate": 2.6380748770076873e-06,
      "loss": 0.86697894,
      "num_input_tokens_seen": 148093240,
      "step": 6898,
      "time_per_iteration": 2.5377490520477295
    },
    {
      "auxiliary_loss_clip": 0.010749,
      "auxiliary_loss_mlp": 0.01033005,
      "balance_loss_clip": 1.03971982,
      "balance_loss_mlp": 1.01927233,
      "epoch": 0.41479032015632045,
      "flos": 20298291194880.0,
      "grad_norm": 1.657087210421443,
      "language_loss": 0.74594152,
      "learning_rate": 2.6377057556751416e-06,
      "loss": 0.76702058,
      "num_input_tokens_seen": 148110925,
      "step": 6899,
      "time_per_iteration": 2.5933306217193604
    },
    {
      "auxiliary_loss_clip": 0.01090771,
      "auxiliary_loss_mlp": 0.0103706,
      "balance_loss_clip": 1.04376364,
      "balance_loss_mlp": 1.02155066,
      "epoch": 0.4148504434089884,
      "flos": 25264988472960.0,
      "grad_norm": 1.7893925950247982,
      "language_loss": 0.75787258,
      "learning_rate": 2.6373366101607306e-06,
      "loss": 0.7791509,
      "num_input_tokens_seen": 148130670,
      "step": 6900,
      "time_per_iteration": 2.6006691455841064
    },
    {
      "auxiliary_loss_clip": 0.011074,
      "auxiliary_loss_mlp": 0.01039011,
      "balance_loss_clip": 1.04391813,
      "balance_loss_mlp": 1.02391934,
      "epoch": 0.4149105666616564,
      "flos": 12822300679680.0,
      "grad_norm": 2.1754682086154467,
      "language_loss": 0.80052733,
      "learning_rate": 2.6369674404784503e-06,
      "loss": 0.82199138,
      "num_input_tokens_seen": 148148350,
      "step": 6901,
      "time_per_iteration": 2.486302137374878
    },
    {
      "auxiliary_loss_clip": 0.01082678,
      "auxiliary_loss_mlp": 0.01042254,
      "balance_loss_clip": 1.04026413,
      "balance_loss_mlp": 1.02690005,
      "epoch": 0.41497068991432434,
      "flos": 16763891713920.0,
      "grad_norm": 1.6695877168341542,
      "language_loss": 0.70355242,
      "learning_rate": 2.6365982466423014e-06,
      "loss": 0.72480172,
      "num_input_tokens_seen": 148167550,
      "step": 6902,
      "time_per_iteration": 2.5214319229125977
    },
    {
      "auxiliary_loss_clip": 0.01094603,
      "auxiliary_loss_mlp": 0.00796509,
      "balance_loss_clip": 1.04689479,
      "balance_loss_mlp": 1.02437556,
      "epoch": 0.4150308131669923,
      "flos": 18000906243840.0,
      "grad_norm": 1.6155122145045744,
      "language_loss": 0.83849072,
      "learning_rate": 2.6362290286662834e-06,
      "loss": 0.85740185,
      "num_input_tokens_seen": 148184740,
      "step": 6903,
      "time_per_iteration": 2.515160322189331
    },
    {
      "auxiliary_loss_clip": 0.01128297,
      "auxiliary_loss_mlp": 0.01040419,
      "balance_loss_clip": 1.04547811,
      "balance_loss_mlp": 1.02520776,
      "epoch": 0.41509093641966033,
      "flos": 30044770352640.0,
      "grad_norm": 1.864358375948661,
      "language_loss": 0.6791569,
      "learning_rate": 2.6358597865643968e-06,
      "loss": 0.70084399,
      "num_input_tokens_seen": 148204605,
      "step": 6904,
      "time_per_iteration": 2.5165607929229736
    },
    {
      "auxiliary_loss_clip": 0.0112549,
      "auxiliary_loss_mlp": 0.00793184,
      "balance_loss_clip": 1.04387987,
      "balance_loss_mlp": 1.01789403,
      "epoch": 0.4151510596723283,
      "flos": 24279994742400.0,
      "grad_norm": 1.9199372377495398,
      "language_loss": 0.77346617,
      "learning_rate": 2.635490520350643e-06,
      "loss": 0.79265285,
      "num_input_tokens_seen": 148224675,
      "step": 6905,
      "time_per_iteration": 2.5006649494171143
    },
    {
      "auxiliary_loss_clip": 0.01126654,
      "auxiliary_loss_mlp": 0.01033965,
      "balance_loss_clip": 1.0454253,
      "balance_loss_mlp": 1.01958227,
      "epoch": 0.41521118292499626,
      "flos": 23476206147840.0,
      "grad_norm": 1.8653766223241826,
      "language_loss": 0.68144101,
      "learning_rate": 2.635121230039025e-06,
      "loss": 0.70304722,
      "num_input_tokens_seen": 148243375,
      "step": 6906,
      "time_per_iteration": 2.4892544746398926
    },
    {
      "auxiliary_loss_clip": 0.01102526,
      "auxiliary_loss_mlp": 0.01035422,
      "balance_loss_clip": 1.04413891,
      "balance_loss_mlp": 1.02207685,
      "epoch": 0.4152713061776642,
      "flos": 22125498094080.0,
      "grad_norm": 2.143764362630354,
      "language_loss": 0.67487544,
      "learning_rate": 2.6347519156435467e-06,
      "loss": 0.69625491,
      "num_input_tokens_seen": 148261140,
      "step": 6907,
      "time_per_iteration": 2.526439905166626
    },
    {
      "auxiliary_loss_clip": 0.01092811,
      "auxiliary_loss_mlp": 0.01035015,
      "balance_loss_clip": 1.04461563,
      "balance_loss_mlp": 1.02197921,
      "epoch": 0.4153314294303322,
      "flos": 21251396626560.0,
      "grad_norm": 1.721520187502034,
      "language_loss": 0.77118337,
      "learning_rate": 2.6343825771782123e-06,
      "loss": 0.79246163,
      "num_input_tokens_seen": 148279655,
      "step": 6908,
      "time_per_iteration": 2.561007499694824
    },
    {
      "auxiliary_loss_clip": 0.0103432,
      "auxiliary_loss_mlp": 0.01012766,
      "balance_loss_clip": 1.02551496,
      "balance_loss_mlp": 1.01096618,
      "epoch": 0.41539155268300015,
      "flos": 57920681594880.0,
      "grad_norm": 0.9305522876801535,
      "language_loss": 0.64805162,
      "learning_rate": 2.634013214657026e-06,
      "loss": 0.66852248,
      "num_input_tokens_seen": 148339005,
      "step": 6909,
      "time_per_iteration": 3.128500461578369
    },
    {
      "auxiliary_loss_clip": 0.010939,
      "auxiliary_loss_mlp": 0.01038595,
      "balance_loss_clip": 1.04510522,
      "balance_loss_mlp": 1.02464771,
      "epoch": 0.4154516759356681,
      "flos": 21903677654400.0,
      "grad_norm": 1.4744953430327294,
      "language_loss": 0.87191617,
      "learning_rate": 2.633643828093996e-06,
      "loss": 0.89324105,
      "num_input_tokens_seen": 148358715,
      "step": 6910,
      "time_per_iteration": 3.9596359729766846
    },
    {
      "auxiliary_loss_clip": 0.01041328,
      "auxiliary_loss_mlp": 0.01015311,
      "balance_loss_clip": 1.02311611,
      "balance_loss_mlp": 1.01382124,
      "epoch": 0.4155117991883361,
      "flos": 67833677226240.0,
      "grad_norm": 0.8049100775817685,
      "language_loss": 0.62165642,
      "learning_rate": 2.633274417503128e-06,
      "loss": 0.64222282,
      "num_input_tokens_seen": 148417280,
      "step": 6911,
      "time_per_iteration": 3.0626280307769775
    },
    {
      "auxiliary_loss_clip": 0.01129183,
      "auxiliary_loss_mlp": 0.0103915,
      "balance_loss_clip": 1.04595721,
      "balance_loss_mlp": 1.02473736,
      "epoch": 0.41557192244100405,
      "flos": 14282679934080.0,
      "grad_norm": 4.2071813038520425,
      "language_loss": 0.87356794,
      "learning_rate": 2.6329049828984312e-06,
      "loss": 0.89525121,
      "num_input_tokens_seen": 148432610,
      "step": 6912,
      "time_per_iteration": 3.81512451171875
    },
    {
      "auxiliary_loss_clip": 0.01111267,
      "auxiliary_loss_mlp": 0.01030289,
      "balance_loss_clip": 1.04873657,
      "balance_loss_mlp": 1.01700926,
      "epoch": 0.415632045693672,
      "flos": 24461954064000.0,
      "grad_norm": 2.8867267843255964,
      "language_loss": 0.63191372,
      "learning_rate": 2.632535524293914e-06,
      "loss": 0.65332925,
      "num_input_tokens_seen": 148451510,
      "step": 6913,
      "time_per_iteration": 2.547332525253296
    },
    {
      "auxiliary_loss_clip": 0.01097489,
      "auxiliary_loss_mlp": 0.00792103,
      "balance_loss_clip": 1.0429076,
      "balance_loss_mlp": 1.01695967,
      "epoch": 0.41569216894634,
      "flos": 20115290378880.0,
      "grad_norm": 2.6255682707553665,
      "language_loss": 0.75034308,
      "learning_rate": 2.632166041703586e-06,
      "loss": 0.76923895,
      "num_input_tokens_seen": 148469945,
      "step": 6914,
      "time_per_iteration": 3.8904013633728027
    },
    {
      "auxiliary_loss_clip": 0.01073452,
      "auxiliary_loss_mlp": 0.01041546,
      "balance_loss_clip": 1.04242134,
      "balance_loss_mlp": 1.02638257,
      "epoch": 0.41575229219900794,
      "flos": 23798827128960.0,
      "grad_norm": 3.0273258105101752,
      "language_loss": 0.87738669,
      "learning_rate": 2.631796535141458e-06,
      "loss": 0.89853662,
      "num_input_tokens_seen": 148486655,
      "step": 6915,
      "time_per_iteration": 3.960951089859009
    },
    {
      "auxiliary_loss_clip": 0.01101293,
      "auxiliary_loss_mlp": 0.01043006,
      "balance_loss_clip": 1.04505754,
      "balance_loss_mlp": 1.02868938,
      "epoch": 0.4158124154516759,
      "flos": 23108229267840.0,
      "grad_norm": 2.1041736126068913,
      "language_loss": 0.71252906,
      "learning_rate": 2.6314270046215426e-06,
      "loss": 0.73397207,
      "num_input_tokens_seen": 148505035,
      "step": 6916,
      "time_per_iteration": 2.519998073577881
    },
    {
      "auxiliary_loss_clip": 0.01126722,
      "auxiliary_loss_mlp": 0.01033042,
      "balance_loss_clip": 1.04504514,
      "balance_loss_mlp": 1.01842153,
      "epoch": 0.41587253870434393,
      "flos": 24242970798720.0,
      "grad_norm": 1.5079511011784503,
      "language_loss": 0.71753323,
      "learning_rate": 2.631057450157852e-06,
      "loss": 0.73913091,
      "num_input_tokens_seen": 148525575,
      "step": 6917,
      "time_per_iteration": 2.488110303878784
    },
    {
      "auxiliary_loss_clip": 0.01103585,
      "auxiliary_loss_mlp": 0.0103051,
      "balance_loss_clip": 1.04292619,
      "balance_loss_mlp": 1.0165689,
      "epoch": 0.4159326619570119,
      "flos": 23881602021120.0,
      "grad_norm": 1.5732641281229314,
      "language_loss": 0.80952972,
      "learning_rate": 2.6306878717643988e-06,
      "loss": 0.83087063,
      "num_input_tokens_seen": 148547270,
      "step": 6918,
      "time_per_iteration": 2.552065372467041
    },
    {
      "auxiliary_loss_clip": 0.01114178,
      "auxiliary_loss_mlp": 0.0103586,
      "balance_loss_clip": 1.0491631,
      "balance_loss_mlp": 1.02116108,
      "epoch": 0.41599278520967986,
      "flos": 40626531354240.0,
      "grad_norm": 1.5420121770160584,
      "language_loss": 0.70426035,
      "learning_rate": 2.6303182694551995e-06,
      "loss": 0.72576076,
      "num_input_tokens_seen": 148572100,
      "step": 6919,
      "time_per_iteration": 2.674450159072876
    },
    {
      "auxiliary_loss_clip": 0.0110195,
      "auxiliary_loss_mlp": 0.01035283,
      "balance_loss_clip": 1.04376757,
      "balance_loss_mlp": 1.02039385,
      "epoch": 0.4160529084623478,
      "flos": 18222942165120.0,
      "grad_norm": 1.8389864844528119,
      "language_loss": 0.81145978,
      "learning_rate": 2.6299486432442677e-06,
      "loss": 0.8328321,
      "num_input_tokens_seen": 148591245,
      "step": 6920,
      "time_per_iteration": 2.5236778259277344
    },
    {
      "auxiliary_loss_clip": 0.01102806,
      "auxiliary_loss_mlp": 0.01032314,
      "balance_loss_clip": 1.0440011,
      "balance_loss_mlp": 1.01728725,
      "epoch": 0.4161130317150158,
      "flos": 13661963982720.0,
      "grad_norm": 1.9968737089874469,
      "language_loss": 0.65816391,
      "learning_rate": 2.6295789931456195e-06,
      "loss": 0.67951506,
      "num_input_tokens_seen": 148607980,
      "step": 6921,
      "time_per_iteration": 2.484254837036133
    },
    {
      "auxiliary_loss_clip": 0.01101402,
      "auxiliary_loss_mlp": 0.01037238,
      "balance_loss_clip": 1.04240227,
      "balance_loss_mlp": 1.02294457,
      "epoch": 0.41617315496768376,
      "flos": 16178511767040.0,
      "grad_norm": 2.9649648414612098,
      "language_loss": 0.80679893,
      "learning_rate": 2.629209319173274e-06,
      "loss": 0.82818532,
      "num_input_tokens_seen": 148624490,
      "step": 6922,
      "time_per_iteration": 2.4999377727508545
    },
    {
      "auxiliary_loss_clip": 0.01099887,
      "auxiliary_loss_mlp": 0.01035113,
      "balance_loss_clip": 1.04362857,
      "balance_loss_mlp": 1.02053952,
      "epoch": 0.4162332782203517,
      "flos": 26213317395840.0,
      "grad_norm": 1.703256808317483,
      "language_loss": 0.67282653,
      "learning_rate": 2.628839621341247e-06,
      "loss": 0.6941765,
      "num_input_tokens_seen": 148646490,
      "step": 6923,
      "time_per_iteration": 2.5643982887268066
    },
    {
      "auxiliary_loss_clip": 0.01093541,
      "auxiliary_loss_mlp": 0.01044319,
      "balance_loss_clip": 1.0437429,
      "balance_loss_mlp": 1.02900052,
      "epoch": 0.4162934014730197,
      "flos": 28183987215360.0,
      "grad_norm": 1.8661812988848512,
      "language_loss": 0.76021123,
      "learning_rate": 2.6284698996635593e-06,
      "loss": 0.78158987,
      "num_input_tokens_seen": 148668580,
      "step": 6924,
      "time_per_iteration": 2.5902259349823
    },
    {
      "auxiliary_loss_clip": 0.0112437,
      "auxiliary_loss_mlp": 0.01034779,
      "balance_loss_clip": 1.04298413,
      "balance_loss_mlp": 1.02054548,
      "epoch": 0.41635352472568765,
      "flos": 19865316654720.0,
      "grad_norm": 1.6868022704130128,
      "language_loss": 0.73336565,
      "learning_rate": 2.62810015415423e-06,
      "loss": 0.75495714,
      "num_input_tokens_seen": 148688410,
      "step": 6925,
      "time_per_iteration": 2.4444355964660645
    },
    {
      "auxiliary_loss_clip": 0.01098966,
      "auxiliary_loss_mlp": 0.01032878,
      "balance_loss_clip": 1.0402627,
      "balance_loss_mlp": 1.01947904,
      "epoch": 0.4164136479783556,
      "flos": 14935356011520.0,
      "grad_norm": 2.0324694099439777,
      "language_loss": 0.84183449,
      "learning_rate": 2.6277303848272792e-06,
      "loss": 0.86315292,
      "num_input_tokens_seen": 148704855,
      "step": 6926,
      "time_per_iteration": 2.4940683841705322
    },
    {
      "auxiliary_loss_clip": 0.01090265,
      "auxiliary_loss_mlp": 0.01033759,
      "balance_loss_clip": 1.04507065,
      "balance_loss_mlp": 1.02111673,
      "epoch": 0.4164737712310236,
      "flos": 21757593041280.0,
      "grad_norm": 1.7802143608836776,
      "language_loss": 0.86285698,
      "learning_rate": 2.6273605916967302e-06,
      "loss": 0.88409722,
      "num_input_tokens_seen": 148723065,
      "step": 6927,
      "time_per_iteration": 2.529287815093994
    },
    {
      "auxiliary_loss_clip": 0.01111796,
      "auxiliary_loss_mlp": 0.01038916,
      "balance_loss_clip": 1.04299819,
      "balance_loss_mlp": 1.02408063,
      "epoch": 0.41653389448369155,
      "flos": 20740136394240.0,
      "grad_norm": 2.1932091706759484,
      "language_loss": 0.72163993,
      "learning_rate": 2.626990774776604e-06,
      "loss": 0.74314708,
      "num_input_tokens_seen": 148741780,
      "step": 6928,
      "time_per_iteration": 2.5097031593322754
    },
    {
      "auxiliary_loss_clip": 0.01094496,
      "auxiliary_loss_mlp": 0.01034063,
      "balance_loss_clip": 1.04339266,
      "balance_loss_mlp": 1.02000809,
      "epoch": 0.4165940177363595,
      "flos": 24972891073920.0,
      "grad_norm": 1.8856219313845626,
      "language_loss": 0.78070003,
      "learning_rate": 2.6266209340809254e-06,
      "loss": 0.80198562,
      "num_input_tokens_seen": 148759795,
      "step": 6929,
      "time_per_iteration": 2.5451531410217285
    },
    {
      "auxiliary_loss_clip": 0.01121923,
      "auxiliary_loss_mlp": 0.01032713,
      "balance_loss_clip": 1.04285455,
      "balance_loss_mlp": 1.01902211,
      "epoch": 0.41665414098902753,
      "flos": 20521727746560.0,
      "grad_norm": 1.8243758788810978,
      "language_loss": 0.71122926,
      "learning_rate": 2.6262510696237182e-06,
      "loss": 0.73277557,
      "num_input_tokens_seen": 148778680,
      "step": 6930,
      "time_per_iteration": 2.4697351455688477
    },
    {
      "auxiliary_loss_clip": 0.0109557,
      "auxiliary_loss_mlp": 0.01033431,
      "balance_loss_clip": 1.04315734,
      "balance_loss_mlp": 1.01977551,
      "epoch": 0.4167142642416955,
      "flos": 19682926369920.0,
      "grad_norm": 2.192154911570851,
      "language_loss": 0.81146938,
      "learning_rate": 2.625881181419007e-06,
      "loss": 0.83275938,
      "num_input_tokens_seen": 148796470,
      "step": 6931,
      "time_per_iteration": 2.5105483531951904
    },
    {
      "auxiliary_loss_clip": 0.01070057,
      "auxiliary_loss_mlp": 0.01038411,
      "balance_loss_clip": 1.03872061,
      "balance_loss_mlp": 1.02387965,
      "epoch": 0.41677438749436346,
      "flos": 23763742519680.0,
      "grad_norm": 1.693294371439168,
      "language_loss": 0.78985572,
      "learning_rate": 2.6255112694808193e-06,
      "loss": 0.81094033,
      "num_input_tokens_seen": 148815300,
      "step": 6932,
      "time_per_iteration": 2.6561436653137207
    },
    {
      "auxiliary_loss_clip": 0.0110327,
      "auxiliary_loss_mlp": 0.0079367,
      "balance_loss_clip": 1.04468858,
      "balance_loss_mlp": 1.01456666,
      "epoch": 0.41683451074703143,
      "flos": 30410053712640.0,
      "grad_norm": 1.893006947102337,
      "language_loss": 0.81511551,
      "learning_rate": 2.6251413338231813e-06,
      "loss": 0.83408487,
      "num_input_tokens_seen": 148834315,
      "step": 6933,
      "time_per_iteration": 2.6245996952056885
    },
    {
      "auxiliary_loss_clip": 0.01124809,
      "auxiliary_loss_mlp": 0.01034107,
      "balance_loss_clip": 1.0422039,
      "balance_loss_mlp": 1.01872897,
      "epoch": 0.4168946339996994,
      "flos": 21506757390720.0,
      "grad_norm": 1.776484676391278,
      "language_loss": 0.76844943,
      "learning_rate": 2.624771374460121e-06,
      "loss": 0.79003859,
      "num_input_tokens_seen": 148852420,
      "step": 6934,
      "time_per_iteration": 2.4754505157470703
    },
    {
      "auxiliary_loss_clip": 0.01112361,
      "auxiliary_loss_mlp": 0.01032915,
      "balance_loss_clip": 1.04315639,
      "balance_loss_mlp": 1.01877689,
      "epoch": 0.41695475725236736,
      "flos": 17638675539840.0,
      "grad_norm": 1.8115111655812715,
      "language_loss": 0.67171121,
      "learning_rate": 2.624401391405668e-06,
      "loss": 0.69316393,
      "num_input_tokens_seen": 148869305,
      "step": 6935,
      "time_per_iteration": 2.4536144733428955
    },
    {
      "auxiliary_loss_clip": 0.01098427,
      "auxiliary_loss_mlp": 0.01039249,
      "balance_loss_clip": 1.04711938,
      "balance_loss_mlp": 1.02536726,
      "epoch": 0.4170148805050353,
      "flos": 15668903560320.0,
      "grad_norm": 2.559498736452913,
      "language_loss": 0.73541105,
      "learning_rate": 2.6240313846738513e-06,
      "loss": 0.75678778,
      "num_input_tokens_seen": 148886395,
      "step": 6936,
      "time_per_iteration": 2.4979114532470703
    },
    {
      "auxiliary_loss_clip": 0.01104446,
      "auxiliary_loss_mlp": 0.01035067,
      "balance_loss_clip": 1.04136622,
      "balance_loss_mlp": 1.02126265,
      "epoch": 0.4170750037577033,
      "flos": 15159151699200.0,
      "grad_norm": 1.7716289986615963,
      "language_loss": 0.73347461,
      "learning_rate": 2.6236613542787024e-06,
      "loss": 0.7548697,
      "num_input_tokens_seen": 148905235,
      "step": 6937,
      "time_per_iteration": 2.467724323272705
    },
    {
      "auxiliary_loss_clip": 0.01096507,
      "auxiliary_loss_mlp": 0.0103428,
      "balance_loss_clip": 1.04173493,
      "balance_loss_mlp": 1.02079141,
      "epoch": 0.41713512701037125,
      "flos": 28768289754240.0,
      "grad_norm": 1.480836627043124,
      "language_loss": 0.84399295,
      "learning_rate": 2.6232913002342518e-06,
      "loss": 0.86530083,
      "num_input_tokens_seen": 148928130,
      "step": 6938,
      "time_per_iteration": 2.6135406494140625
    },
    {
      "auxiliary_loss_clip": 0.01105324,
      "auxiliary_loss_mlp": 0.0103553,
      "balance_loss_clip": 1.04726315,
      "balance_loss_mlp": 1.02085543,
      "epoch": 0.4171952502630392,
      "flos": 28256993608320.0,
      "grad_norm": 1.8353499495441778,
      "language_loss": 0.74366438,
      "learning_rate": 2.6229212225545334e-06,
      "loss": 0.76507294,
      "num_input_tokens_seen": 148948790,
      "step": 6939,
      "time_per_iteration": 2.5790164470672607
    },
    {
      "auxiliary_loss_clip": 0.01111685,
      "auxiliary_loss_mlp": 0.01038719,
      "balance_loss_clip": 1.04192686,
      "balance_loss_mlp": 1.02396131,
      "epoch": 0.4172553735157072,
      "flos": 24571697091840.0,
      "grad_norm": 1.5750714725080317,
      "language_loss": 0.75045824,
      "learning_rate": 2.622551121253579e-06,
      "loss": 0.77196223,
      "num_input_tokens_seen": 148967690,
      "step": 6940,
      "time_per_iteration": 2.554197072982788
    },
    {
      "auxiliary_loss_clip": 0.011233,
      "auxiliary_loss_mlp": 0.01038646,
      "balance_loss_clip": 1.04434156,
      "balance_loss_mlp": 1.02560425,
      "epoch": 0.41731549676837515,
      "flos": 27045797978880.0,
      "grad_norm": 1.6804006228458461,
      "language_loss": 0.71419311,
      "learning_rate": 2.622180996345424e-06,
      "loss": 0.73581254,
      "num_input_tokens_seen": 148987150,
      "step": 6941,
      "time_per_iteration": 2.5204460620880127
    },
    {
      "auxiliary_loss_clip": 0.01111186,
      "auxiliary_loss_mlp": 0.01044751,
      "balance_loss_clip": 1.04549289,
      "balance_loss_mlp": 1.03030837,
      "epoch": 0.4173756200210431,
      "flos": 28394063907840.0,
      "grad_norm": 1.9546128114816732,
      "language_loss": 0.73961538,
      "learning_rate": 2.621810847844104e-06,
      "loss": 0.76117474,
      "num_input_tokens_seen": 149004895,
      "step": 6942,
      "time_per_iteration": 2.6186671257019043
    },
    {
      "auxiliary_loss_clip": 0.01086498,
      "auxiliary_loss_mlp": 0.01039954,
      "balance_loss_clip": 1.04080176,
      "balance_loss_mlp": 1.0241704,
      "epoch": 0.41743574327371114,
      "flos": 22521556431360.0,
      "grad_norm": 2.5041437068636365,
      "language_loss": 0.72761106,
      "learning_rate": 2.6214406757636534e-06,
      "loss": 0.74887556,
      "num_input_tokens_seen": 149020970,
      "step": 6943,
      "time_per_iteration": 2.549762725830078
    },
    {
      "auxiliary_loss_clip": 0.01090807,
      "auxiliary_loss_mlp": 0.00794982,
      "balance_loss_clip": 1.04127574,
      "balance_loss_mlp": 1.01885557,
      "epoch": 0.4174958665263791,
      "flos": 30113431200000.0,
      "grad_norm": 1.6342079768451196,
      "language_loss": 0.63768238,
      "learning_rate": 2.621070480118111e-06,
      "loss": 0.65654033,
      "num_input_tokens_seen": 149041795,
      "step": 6944,
      "time_per_iteration": 2.630528211593628
    },
    {
      "auxiliary_loss_clip": 0.01094865,
      "auxiliary_loss_mlp": 0.01036435,
      "balance_loss_clip": 1.04138589,
      "balance_loss_mlp": 1.02184343,
      "epoch": 0.41755598977904707,
      "flos": 25263444188160.0,
      "grad_norm": 1.4292875924874588,
      "language_loss": 0.69998646,
      "learning_rate": 2.620700260921513e-06,
      "loss": 0.72129941,
      "num_input_tokens_seen": 149063700,
      "step": 6945,
      "time_per_iteration": 2.614041566848755
    },
    {
      "auxiliary_loss_clip": 0.01080485,
      "auxiliary_loss_mlp": 0.01050985,
      "balance_loss_clip": 1.03772473,
      "balance_loss_mlp": 1.03315163,
      "epoch": 0.41761611303171503,
      "flos": 19828580019840.0,
      "grad_norm": 2.8055259033223923,
      "language_loss": 0.8062948,
      "learning_rate": 2.620330018187899e-06,
      "loss": 0.82760954,
      "num_input_tokens_seen": 149082410,
      "step": 6946,
      "time_per_iteration": 2.585125207901001
    },
    {
      "auxiliary_loss_clip": 0.01116201,
      "auxiliary_loss_mlp": 0.01034073,
      "balance_loss_clip": 1.04470873,
      "balance_loss_mlp": 1.01982772,
      "epoch": 0.417676236284383,
      "flos": 15523249910400.0,
      "grad_norm": 2.1686195233388657,
      "language_loss": 0.77047092,
      "learning_rate": 2.6199597519313086e-06,
      "loss": 0.79197365,
      "num_input_tokens_seen": 149098745,
      "step": 6947,
      "time_per_iteration": 2.4604761600494385
    },
    {
      "auxiliary_loss_clip": 0.01123668,
      "auxiliary_loss_mlp": 0.01035859,
      "balance_loss_clip": 1.04382682,
      "balance_loss_mlp": 1.02077889,
      "epoch": 0.41773635953705096,
      "flos": 32524473761280.0,
      "grad_norm": 1.5121336451723844,
      "language_loss": 0.71688318,
      "learning_rate": 2.6195894621657825e-06,
      "loss": 0.73847842,
      "num_input_tokens_seen": 149122255,
      "step": 6948,
      "time_per_iteration": 2.580331802368164
    },
    {
      "auxiliary_loss_clip": 0.01107288,
      "auxiliary_loss_mlp": 0.0102911,
      "balance_loss_clip": 1.04142058,
      "balance_loss_mlp": 1.0151273,
      "epoch": 0.4177964827897189,
      "flos": 23440941970560.0,
      "grad_norm": 1.47041347683442,
      "language_loss": 0.77177,
      "learning_rate": 2.619219148905362e-06,
      "loss": 0.79313403,
      "num_input_tokens_seen": 149142845,
      "step": 6949,
      "time_per_iteration": 3.9206690788269043
    },
    {
      "auxiliary_loss_clip": 0.011001,
      "auxiliary_loss_mlp": 0.01033982,
      "balance_loss_clip": 1.04317236,
      "balance_loss_mlp": 1.01909292,
      "epoch": 0.4178566060423869,
      "flos": 22748907565440.0,
      "grad_norm": 1.7206418277445708,
      "language_loss": 0.82161498,
      "learning_rate": 2.6188488121640888e-06,
      "loss": 0.84295583,
      "num_input_tokens_seen": 149163375,
      "step": 6950,
      "time_per_iteration": 2.524818181991577
    },
    {
      "auxiliary_loss_clip": 0.01093164,
      "auxiliary_loss_mlp": 0.00792598,
      "balance_loss_clip": 1.04412961,
      "balance_loss_mlp": 1.01704764,
      "epoch": 0.41791672929505486,
      "flos": 26032794618240.0,
      "grad_norm": 1.2833805198674966,
      "language_loss": 0.75901425,
      "learning_rate": 2.618478451956007e-06,
      "loss": 0.77787185,
      "num_input_tokens_seen": 149185610,
      "step": 6951,
      "time_per_iteration": 3.996936321258545
    },
    {
      "auxiliary_loss_clip": 0.01078356,
      "auxiliary_loss_mlp": 0.0102877,
      "balance_loss_clip": 1.0445621,
      "balance_loss_mlp": 1.01410782,
      "epoch": 0.4179768525477228,
      "flos": 19568694142080.0,
      "grad_norm": 3.464949678195468,
      "language_loss": 0.7324937,
      "learning_rate": 2.61810806829516e-06,
      "loss": 0.75356495,
      "num_input_tokens_seen": 149203990,
      "step": 6952,
      "time_per_iteration": 2.561913251876831
    },
    {
      "auxiliary_loss_clip": 0.01111087,
      "auxiliary_loss_mlp": 0.01033585,
      "balance_loss_clip": 1.04787636,
      "balance_loss_mlp": 1.01983404,
      "epoch": 0.4180369758003908,
      "flos": 17783826399360.0,
      "grad_norm": 2.288798618558008,
      "language_loss": 0.71961844,
      "learning_rate": 2.617737661195593e-06,
      "loss": 0.7410652,
      "num_input_tokens_seen": 149221385,
      "step": 6953,
      "time_per_iteration": 3.8590290546417236
    },
    {
      "auxiliary_loss_clip": 0.0110912,
      "auxiliary_loss_mlp": 0.01033404,
      "balance_loss_clip": 1.04273939,
      "balance_loss_mlp": 1.01847351,
      "epoch": 0.41809709905305875,
      "flos": 20960663944320.0,
      "grad_norm": 1.7641999559919381,
      "language_loss": 0.76169837,
      "learning_rate": 2.617367230671353e-06,
      "loss": 0.78312361,
      "num_input_tokens_seen": 149241175,
      "step": 6954,
      "time_per_iteration": 3.8954226970672607
    },
    {
      "auxiliary_loss_clip": 0.01082141,
      "auxiliary_loss_mlp": 0.01042126,
      "balance_loss_clip": 1.04208469,
      "balance_loss_mlp": 1.02605665,
      "epoch": 0.4181572223057267,
      "flos": 22017622573440.0,
      "grad_norm": 2.1175379458779084,
      "language_loss": 0.84320503,
      "learning_rate": 2.616996776736485e-06,
      "loss": 0.86444771,
      "num_input_tokens_seen": 149259115,
      "step": 6955,
      "time_per_iteration": 2.5775859355926514
    },
    {
      "auxiliary_loss_clip": 0.01111786,
      "auxiliary_loss_mlp": 0.01033201,
      "balance_loss_clip": 1.04369497,
      "balance_loss_mlp": 1.01950431,
      "epoch": 0.4182173455583947,
      "flos": 26245528917120.0,
      "grad_norm": 1.5985109985036736,
      "language_loss": 0.83042669,
      "learning_rate": 2.616626299405037e-06,
      "loss": 0.85187662,
      "num_input_tokens_seen": 149278705,
      "step": 6956,
      "time_per_iteration": 2.5282037258148193
    },
    {
      "auxiliary_loss_clip": 0.01088765,
      "auxiliary_loss_mlp": 0.01037286,
      "balance_loss_clip": 1.04579353,
      "balance_loss_mlp": 1.02211118,
      "epoch": 0.4182774688110627,
      "flos": 14791605782400.0,
      "grad_norm": 4.198351988146465,
      "language_loss": 0.71499705,
      "learning_rate": 2.616255798691059e-06,
      "loss": 0.73625761,
      "num_input_tokens_seen": 149294040,
      "step": 6957,
      "time_per_iteration": 2.5018653869628906
    },
    {
      "auxiliary_loss_clip": 0.01090373,
      "auxiliary_loss_mlp": 0.01038938,
      "balance_loss_clip": 1.04234052,
      "balance_loss_mlp": 1.0259558,
      "epoch": 0.41833759206373067,
      "flos": 20412020632320.0,
      "grad_norm": 2.782988200386959,
      "language_loss": 0.75406015,
      "learning_rate": 2.6158852746085982e-06,
      "loss": 0.77535319,
      "num_input_tokens_seen": 149310385,
      "step": 6958,
      "time_per_iteration": 2.535449981689453
    },
    {
      "auxiliary_loss_clip": 0.01074426,
      "auxiliary_loss_mlp": 0.00790948,
      "balance_loss_clip": 1.03848648,
      "balance_loss_mlp": 1.01177847,
      "epoch": 0.41839771531639863,
      "flos": 23656333875840.0,
      "grad_norm": 1.6622305411751754,
      "language_loss": 0.77178836,
      "learning_rate": 2.6155147271717066e-06,
      "loss": 0.79044211,
      "num_input_tokens_seen": 149328235,
      "step": 6959,
      "time_per_iteration": 2.5820393562316895
    },
    {
      "auxiliary_loss_clip": 0.01082339,
      "auxiliary_loss_mlp": 0.00789895,
      "balance_loss_clip": 1.04118872,
      "balance_loss_mlp": 1.01227415,
      "epoch": 0.4184578385690666,
      "flos": 19754137082880.0,
      "grad_norm": 2.001032584738892,
      "language_loss": 0.76879418,
      "learning_rate": 2.6151441563944347e-06,
      "loss": 0.78751653,
      "num_input_tokens_seen": 149347465,
      "step": 6960,
      "time_per_iteration": 2.570355176925659
    },
    {
      "auxiliary_loss_clip": 0.01090666,
      "auxiliary_loss_mlp": 0.01031417,
      "balance_loss_clip": 1.04411185,
      "balance_loss_mlp": 1.01861954,
      "epoch": 0.41851796182173456,
      "flos": 20193396503040.0,
      "grad_norm": 1.6642645253243589,
      "language_loss": 0.76047957,
      "learning_rate": 2.614773562290835e-06,
      "loss": 0.78170037,
      "num_input_tokens_seen": 149366685,
      "step": 6961,
      "time_per_iteration": 2.517451047897339
    },
    {
      "auxiliary_loss_clip": 0.01023976,
      "auxiliary_loss_mlp": 0.01008544,
      "balance_loss_clip": 1.02967107,
      "balance_loss_mlp": 1.00674391,
      "epoch": 0.41857808507440253,
      "flos": 59018794231680.0,
      "grad_norm": 0.7833144920680323,
      "language_loss": 0.54703987,
      "learning_rate": 2.61440294487496e-06,
      "loss": 0.56736505,
      "num_input_tokens_seen": 149422925,
      "step": 6962,
      "time_per_iteration": 3.095282793045044
    },
    {
      "auxiliary_loss_clip": 0.01109419,
      "auxiliary_loss_mlp": 0.01036067,
      "balance_loss_clip": 1.04438734,
      "balance_loss_mlp": 1.02226841,
      "epoch": 0.4186382083270705,
      "flos": 18478805719680.0,
      "grad_norm": 2.4165113088594743,
      "language_loss": 0.85238779,
      "learning_rate": 2.614032304160864e-06,
      "loss": 0.8738426,
      "num_input_tokens_seen": 149440820,
      "step": 6963,
      "time_per_iteration": 2.477548837661743
    },
    {
      "auxiliary_loss_clip": 0.01096029,
      "auxiliary_loss_mlp": 0.01033145,
      "balance_loss_clip": 1.04380679,
      "balance_loss_mlp": 1.01960909,
      "epoch": 0.41869833157973846,
      "flos": 21578758202880.0,
      "grad_norm": 1.5425291158505081,
      "language_loss": 0.70082188,
      "learning_rate": 2.6136616401626014e-06,
      "loss": 0.72211361,
      "num_input_tokens_seen": 149461060,
      "step": 6964,
      "time_per_iteration": 2.5348896980285645
    },
    {
      "auxiliary_loss_clip": 0.01120682,
      "auxiliary_loss_mlp": 0.01034824,
      "balance_loss_clip": 1.044065,
      "balance_loss_mlp": 1.0214963,
      "epoch": 0.4187584548324064,
      "flos": 35517412650240.0,
      "grad_norm": 1.5660162686770123,
      "language_loss": 0.7134198,
      "learning_rate": 2.6132909528942273e-06,
      "loss": 0.73497486,
      "num_input_tokens_seen": 149483115,
      "step": 6965,
      "time_per_iteration": 2.6114962100982666
    },
    {
      "auxiliary_loss_clip": 0.01071574,
      "auxiliary_loss_mlp": 0.01036573,
      "balance_loss_clip": 1.0394851,
      "balance_loss_mlp": 1.02343011,
      "epoch": 0.4188185780850744,
      "flos": 18655880791680.0,
      "grad_norm": 1.5581665285831399,
      "language_loss": 0.72064012,
      "learning_rate": 2.6129202423697997e-06,
      "loss": 0.74172163,
      "num_input_tokens_seen": 149501495,
      "step": 6966,
      "time_per_iteration": 2.5596861839294434
    },
    {
      "auxiliary_loss_clip": 0.01116619,
      "auxiliary_loss_mlp": 0.01036538,
      "balance_loss_clip": 1.04347372,
      "balance_loss_mlp": 1.02195883,
      "epoch": 0.41887870133774235,
      "flos": 40333428374400.0,
      "grad_norm": 1.9217127421208895,
      "language_loss": 0.71411192,
      "learning_rate": 2.612549508603375e-06,
      "loss": 0.73564351,
      "num_input_tokens_seen": 149523170,
      "step": 6967,
      "time_per_iteration": 2.6552844047546387
    },
    {
      "auxiliary_loss_clip": 0.01042726,
      "auxiliary_loss_mlp": 0.01001547,
      "balance_loss_clip": 1.02402973,
      "balance_loss_mlp": 1.00012255,
      "epoch": 0.4189388245904103,
      "flos": 61371336516480.0,
      "grad_norm": 0.6730346752703964,
      "language_loss": 0.46249914,
      "learning_rate": 2.612178751609011e-06,
      "loss": 0.48294187,
      "num_input_tokens_seen": 149583955,
      "step": 6968,
      "time_per_iteration": 3.088515520095825
    },
    {
      "auxiliary_loss_clip": 0.01113696,
      "auxiliary_loss_mlp": 0.01041323,
      "balance_loss_clip": 1.04185236,
      "balance_loss_mlp": 1.02567053,
      "epoch": 0.4189989478430783,
      "flos": 28215624119040.0,
      "grad_norm": 2.0075571755588446,
      "language_loss": 0.75039572,
      "learning_rate": 2.6118079714007685e-06,
      "loss": 0.77194589,
      "num_input_tokens_seen": 149604440,
      "step": 6969,
      "time_per_iteration": 2.578934907913208
    },
    {
      "auxiliary_loss_clip": 0.01096513,
      "auxiliary_loss_mlp": 0.01035514,
      "balance_loss_clip": 1.04087162,
      "balance_loss_mlp": 1.02324724,
      "epoch": 0.4190590710957463,
      "flos": 24565879088640.0,
      "grad_norm": 1.8173549172145091,
      "language_loss": 0.80765128,
      "learning_rate": 2.611437167992705e-06,
      "loss": 0.82897151,
      "num_input_tokens_seen": 149623745,
      "step": 6970,
      "time_per_iteration": 2.5623490810394287
    },
    {
      "auxiliary_loss_clip": 0.01109747,
      "auxiliary_loss_mlp": 0.01036018,
      "balance_loss_clip": 1.04307866,
      "balance_loss_mlp": 1.02162969,
      "epoch": 0.41911919434841427,
      "flos": 21726027964800.0,
      "grad_norm": 1.7735005611774097,
      "language_loss": 0.82968664,
      "learning_rate": 2.6110663413988835e-06,
      "loss": 0.85114431,
      "num_input_tokens_seen": 149643025,
      "step": 6971,
      "time_per_iteration": 2.489835500717163
    },
    {
      "auxiliary_loss_clip": 0.01095255,
      "auxiliary_loss_mlp": 0.01035934,
      "balance_loss_clip": 1.04397678,
      "balance_loss_mlp": 1.02121735,
      "epoch": 0.41917931760108224,
      "flos": 17601543855360.0,
      "grad_norm": 1.688488430910946,
      "language_loss": 0.75031728,
      "learning_rate": 2.6106954916333648e-06,
      "loss": 0.77162921,
      "num_input_tokens_seen": 149660695,
      "step": 6972,
      "time_per_iteration": 2.486983299255371
    },
    {
      "auxiliary_loss_clip": 0.010942,
      "auxiliary_loss_mlp": 0.0103378,
      "balance_loss_clip": 1.04018617,
      "balance_loss_mlp": 1.02001131,
      "epoch": 0.4192394408537502,
      "flos": 37816701022080.0,
      "grad_norm": 1.6919861510620386,
      "language_loss": 0.72904158,
      "learning_rate": 2.610324618710212e-06,
      "loss": 0.75032133,
      "num_input_tokens_seen": 149682040,
      "step": 6973,
      "time_per_iteration": 2.683706283569336
    },
    {
      "auxiliary_loss_clip": 0.01093353,
      "auxiliary_loss_mlp": 0.01039406,
      "balance_loss_clip": 1.04746461,
      "balance_loss_mlp": 1.02528596,
      "epoch": 0.41929956410641817,
      "flos": 23107726477440.0,
      "grad_norm": 1.6971474998152096,
      "language_loss": 0.7453506,
      "learning_rate": 2.609953722643489e-06,
      "loss": 0.76667821,
      "num_input_tokens_seen": 149700855,
      "step": 6974,
      "time_per_iteration": 2.555488348007202
    },
    {
      "auxiliary_loss_clip": 0.01110434,
      "auxiliary_loss_mlp": 0.01034235,
      "balance_loss_clip": 1.04161453,
      "balance_loss_mlp": 1.02070522,
      "epoch": 0.41935968735908613,
      "flos": 22524537260160.0,
      "grad_norm": 1.7270290497503338,
      "language_loss": 0.72742534,
      "learning_rate": 2.609582803447259e-06,
      "loss": 0.74887204,
      "num_input_tokens_seen": 149717360,
      "step": 6975,
      "time_per_iteration": 2.457200288772583
    },
    {
      "auxiliary_loss_clip": 0.01105409,
      "auxiliary_loss_mlp": 0.01037353,
      "balance_loss_clip": 1.04355121,
      "balance_loss_mlp": 1.02308393,
      "epoch": 0.4194198106117541,
      "flos": 26870446759680.0,
      "grad_norm": 1.5846943497259998,
      "language_loss": 0.80650526,
      "learning_rate": 2.6092118611355885e-06,
      "loss": 0.82793289,
      "num_input_tokens_seen": 149738975,
      "step": 6976,
      "time_per_iteration": 2.524115800857544
    },
    {
      "auxiliary_loss_clip": 0.01095768,
      "auxiliary_loss_mlp": 0.01038116,
      "balance_loss_clip": 1.03763056,
      "balance_loss_mlp": 1.02290535,
      "epoch": 0.41947993386442206,
      "flos": 19902412425600.0,
      "grad_norm": 1.8422888926674093,
      "language_loss": 0.68120319,
      "learning_rate": 2.6088408957225425e-06,
      "loss": 0.70254207,
      "num_input_tokens_seen": 149757055,
      "step": 6977,
      "time_per_iteration": 2.4975345134735107
    },
    {
      "auxiliary_loss_clip": 0.01111489,
      "auxiliary_loss_mlp": 0.01037692,
      "balance_loss_clip": 1.0434972,
      "balance_loss_mlp": 1.02417994,
      "epoch": 0.41954005711709,
      "flos": 17383889393280.0,
      "grad_norm": 2.338707343760537,
      "language_loss": 0.8060019,
      "learning_rate": 2.6084699072221898e-06,
      "loss": 0.82749367,
      "num_input_tokens_seen": 149772885,
      "step": 6978,
      "time_per_iteration": 2.4364173412323
    },
    {
      "auxiliary_loss_clip": 0.01122836,
      "auxiliary_loss_mlp": 0.01037118,
      "balance_loss_clip": 1.04131329,
      "balance_loss_mlp": 1.02264643,
      "epoch": 0.419600180369758,
      "flos": 25003306915200.0,
      "grad_norm": 1.8663935066920778,
      "language_loss": 0.82596147,
      "learning_rate": 2.6080988956485964e-06,
      "loss": 0.847561,
      "num_input_tokens_seen": 149791515,
      "step": 6979,
      "time_per_iteration": 2.4946377277374268
    },
    {
      "auxiliary_loss_clip": 0.01118851,
      "auxiliary_loss_mlp": 0.01035684,
      "balance_loss_clip": 1.04060102,
      "balance_loss_mlp": 1.02214146,
      "epoch": 0.41966030362242596,
      "flos": 17383781652480.0,
      "grad_norm": 3.0591406563818313,
      "language_loss": 0.83501476,
      "learning_rate": 2.6077278610158325e-06,
      "loss": 0.85656011,
      "num_input_tokens_seen": 149807250,
      "step": 6980,
      "time_per_iteration": 2.405071258544922
    },
    {
      "auxiliary_loss_clip": 0.01123053,
      "auxiliary_loss_mlp": 0.01034206,
      "balance_loss_clip": 1.04271507,
      "balance_loss_mlp": 1.02061033,
      "epoch": 0.4197204268750939,
      "flos": 22156165330560.0,
      "grad_norm": 2.4257458328839268,
      "language_loss": 0.78875482,
      "learning_rate": 2.6073568033379665e-06,
      "loss": 0.81032741,
      "num_input_tokens_seen": 149821640,
      "step": 6981,
      "time_per_iteration": 2.4353840351104736
    },
    {
      "auxiliary_loss_clip": 0.01086199,
      "auxiliary_loss_mlp": 0.01032154,
      "balance_loss_clip": 1.04056776,
      "balance_loss_mlp": 1.01869488,
      "epoch": 0.4197805501277619,
      "flos": 22084128604800.0,
      "grad_norm": 1.7496004991136724,
      "language_loss": 0.84303665,
      "learning_rate": 2.6069857226290696e-06,
      "loss": 0.86422014,
      "num_input_tokens_seen": 149840545,
      "step": 6982,
      "time_per_iteration": 2.5416836738586426
    },
    {
      "auxiliary_loss_clip": 0.01117,
      "auxiliary_loss_mlp": 0.01036059,
      "balance_loss_clip": 1.04249036,
      "balance_loss_mlp": 1.02158165,
      "epoch": 0.4198406733804299,
      "flos": 26432192920320.0,
      "grad_norm": 1.9878958739237256,
      "language_loss": 0.56849933,
      "learning_rate": 2.606614618903214e-06,
      "loss": 0.59002995,
      "num_input_tokens_seen": 149860375,
      "step": 6983,
      "time_per_iteration": 2.5312018394470215
    },
    {
      "auxiliary_loss_clip": 0.01109466,
      "auxiliary_loss_mlp": 0.01035413,
      "balance_loss_clip": 1.04274261,
      "balance_loss_mlp": 1.02228236,
      "epoch": 0.4199007966330979,
      "flos": 12531029293440.0,
      "grad_norm": 2.056352007734659,
      "language_loss": 0.82436645,
      "learning_rate": 2.606243492174471e-06,
      "loss": 0.8458153,
      "num_input_tokens_seen": 149877850,
      "step": 6984,
      "time_per_iteration": 2.447140693664551
    },
    {
      "auxiliary_loss_clip": 0.01101743,
      "auxiliary_loss_mlp": 0.01032865,
      "balance_loss_clip": 1.03966606,
      "balance_loss_mlp": 1.01928067,
      "epoch": 0.41996091988576584,
      "flos": 21762944167680.0,
      "grad_norm": 3.6551564368695537,
      "language_loss": 0.79312462,
      "learning_rate": 2.605872342456914e-06,
      "loss": 0.81447071,
      "num_input_tokens_seen": 149896110,
      "step": 6985,
      "time_per_iteration": 2.4955735206604004
    },
    {
      "auxiliary_loss_clip": 0.01123808,
      "auxiliary_loss_mlp": 0.01034165,
      "balance_loss_clip": 1.0411166,
      "balance_loss_mlp": 1.01993144,
      "epoch": 0.4200210431384338,
      "flos": 26541935948160.0,
      "grad_norm": 1.6269455596897593,
      "language_loss": 0.78110814,
      "learning_rate": 2.6055011697646173e-06,
      "loss": 0.80268788,
      "num_input_tokens_seen": 149916495,
      "step": 6986,
      "time_per_iteration": 3.905823230743408
    },
    {
      "auxiliary_loss_clip": 0.01097751,
      "auxiliary_loss_mlp": 0.0102794,
      "balance_loss_clip": 1.04143095,
      "balance_loss_mlp": 1.01541114,
      "epoch": 0.42008116639110177,
      "flos": 26795824254720.0,
      "grad_norm": 1.971381820425643,
      "language_loss": 0.72193265,
      "learning_rate": 2.605129974111655e-06,
      "loss": 0.74318957,
      "num_input_tokens_seen": 149936445,
      "step": 6987,
      "time_per_iteration": 2.548518657684326
    },
    {
      "auxiliary_loss_clip": 0.01098859,
      "auxiliary_loss_mlp": 0.00795641,
      "balance_loss_clip": 1.04161799,
      "balance_loss_mlp": 1.01428938,
      "epoch": 0.42014128964376973,
      "flos": 32087333243520.0,
      "grad_norm": 1.4638129949935732,
      "language_loss": 0.75073153,
      "learning_rate": 2.604758755512104e-06,
      "loss": 0.76967657,
      "num_input_tokens_seen": 149959430,
      "step": 6988,
      "time_per_iteration": 2.615042209625244
    },
    {
      "auxiliary_loss_clip": 0.01112238,
      "auxiliary_loss_mlp": 0.010375,
      "balance_loss_clip": 1.04184413,
      "balance_loss_mlp": 1.02329659,
      "epoch": 0.4202014128964377,
      "flos": 26467133875200.0,
      "grad_norm": 1.8540971695433768,
      "language_loss": 0.74102914,
      "learning_rate": 2.60438751398004e-06,
      "loss": 0.76252657,
      "num_input_tokens_seen": 149980365,
      "step": 6989,
      "time_per_iteration": 3.892676591873169
    },
    {
      "auxiliary_loss_clip": 0.0110363,
      "auxiliary_loss_mlp": 0.01030375,
      "balance_loss_clip": 1.04406548,
      "balance_loss_mlp": 1.0162909,
      "epoch": 0.42026153614910566,
      "flos": 13401216178560.0,
      "grad_norm": 1.946141216382325,
      "language_loss": 0.70664573,
      "learning_rate": 2.6040162495295404e-06,
      "loss": 0.72798574,
      "num_input_tokens_seen": 149997375,
      "step": 6990,
      "time_per_iteration": 2.5177974700927734
    },
    {
      "auxiliary_loss_clip": 0.01039412,
      "auxiliary_loss_mlp": 0.00837294,
      "balance_loss_clip": 1.02923322,
      "balance_loss_mlp": 1.12527084,
      "epoch": 0.42032165940177363,
      "flos": 60250457635200.0,
      "grad_norm": 0.8517478143214952,
      "language_loss": 0.60456693,
      "learning_rate": 2.603644962174685e-06,
      "loss": 0.62333399,
      "num_input_tokens_seen": 150051230,
      "step": 6991,
      "time_per_iteration": 2.9769465923309326
    },
    {
      "auxiliary_loss_clip": 0.01126039,
      "auxiliary_loss_mlp": 0.01038941,
      "balance_loss_clip": 1.04511428,
      "balance_loss_mlp": 1.0244925,
      "epoch": 0.4203817826544416,
      "flos": 24535211852160.0,
      "grad_norm": 1.592413610118192,
      "language_loss": 0.83142668,
      "learning_rate": 2.6032736519295517e-06,
      "loss": 0.85307652,
      "num_input_tokens_seen": 150071135,
      "step": 6992,
      "time_per_iteration": 3.8524773120880127
    },
    {
      "auxiliary_loss_clip": 0.01044119,
      "auxiliary_loss_mlp": 0.01006903,
      "balance_loss_clip": 1.01636052,
      "balance_loss_mlp": 1.00521076,
      "epoch": 0.42044190590710956,
      "flos": 58820781530880.0,
      "grad_norm": 0.8041002601200566,
      "language_loss": 0.65496969,
      "learning_rate": 2.6029023188082217e-06,
      "loss": 0.67547989,
      "num_input_tokens_seen": 150125220,
      "step": 6993,
      "time_per_iteration": 4.385503530502319
    },
    {
      "auxiliary_loss_clip": 0.01127686,
      "auxiliary_loss_mlp": 0.01035387,
      "balance_loss_clip": 1.04343235,
      "balance_loss_mlp": 1.01980615,
      "epoch": 0.4205020291597775,
      "flos": 16436063260800.0,
      "grad_norm": 1.8971275360109305,
      "language_loss": 0.83324945,
      "learning_rate": 2.6025309628247746e-06,
      "loss": 0.85488021,
      "num_input_tokens_seen": 150142300,
      "step": 6994,
      "time_per_iteration": 2.4313316345214844
    },
    {
      "auxiliary_loss_clip": 0.01111691,
      "auxiliary_loss_mlp": 0.00795131,
      "balance_loss_clip": 1.04390705,
      "balance_loss_mlp": 1.02264738,
      "epoch": 0.4205621524124455,
      "flos": 18405655672320.0,
      "grad_norm": 1.5491275975594605,
      "language_loss": 0.78267044,
      "learning_rate": 2.6021595839932934e-06,
      "loss": 0.80173862,
      "num_input_tokens_seen": 150161345,
      "step": 6995,
      "time_per_iteration": 2.4673421382904053
    },
    {
      "auxiliary_loss_clip": 0.01086026,
      "auxiliary_loss_mlp": 0.01030692,
      "balance_loss_clip": 1.04145598,
      "balance_loss_mlp": 1.01739419,
      "epoch": 0.4206222756651135,
      "flos": 25520097841920.0,
      "grad_norm": 1.5142758791513486,
      "language_loss": 0.80236334,
      "learning_rate": 2.60178818232786e-06,
      "loss": 0.82353055,
      "num_input_tokens_seen": 150182420,
      "step": 6996,
      "time_per_iteration": 2.5898354053497314
    },
    {
      "auxiliary_loss_clip": 0.01102471,
      "auxiliary_loss_mlp": 0.00802739,
      "balance_loss_clip": 1.04274917,
      "balance_loss_mlp": 1.03742981,
      "epoch": 0.4206823989177815,
      "flos": 15304338472320.0,
      "grad_norm": 2.1144138920939612,
      "language_loss": 0.75341874,
      "learning_rate": 2.601416757842559e-06,
      "loss": 0.77247083,
      "num_input_tokens_seen": 150200175,
      "step": 6997,
      "time_per_iteration": 2.4930367469787598
    },
    {
      "auxiliary_loss_clip": 0.01120627,
      "auxiliary_loss_mlp": 0.01038064,
      "balance_loss_clip": 1.03995073,
      "balance_loss_mlp": 1.02418196,
      "epoch": 0.42074252217044944,
      "flos": 15554096714880.0,
      "grad_norm": 1.9964550889584023,
      "language_loss": 0.75564432,
      "learning_rate": 2.6010453105514743e-06,
      "loss": 0.77723128,
      "num_input_tokens_seen": 150217100,
      "step": 6998,
      "time_per_iteration": 2.4391465187072754
    },
    {
      "auxiliary_loss_clip": 0.01125954,
      "auxiliary_loss_mlp": 0.01035952,
      "balance_loss_clip": 1.04411089,
      "balance_loss_mlp": 1.02167058,
      "epoch": 0.4208026454231174,
      "flos": 26145877610880.0,
      "grad_norm": 1.7664017630549027,
      "language_loss": 0.75868535,
      "learning_rate": 2.60067384046869e-06,
      "loss": 0.78030443,
      "num_input_tokens_seen": 150239830,
      "step": 6999,
      "time_per_iteration": 2.5153234004974365
    },
    {
      "auxiliary_loss_clip": 0.01075963,
      "auxiliary_loss_mlp": 0.01038823,
      "balance_loss_clip": 1.04111719,
      "balance_loss_mlp": 1.02420831,
      "epoch": 0.42086276867578537,
      "flos": 23550110380800.0,
      "grad_norm": 4.085729936156881,
      "language_loss": 0.64003181,
      "learning_rate": 2.600302347608295e-06,
      "loss": 0.66117972,
      "num_input_tokens_seen": 150260690,
      "step": 7000,
      "time_per_iteration": 2.5757107734680176
    },
    {
      "auxiliary_loss_clip": 0.01086597,
      "auxiliary_loss_mlp": 0.01036669,
      "balance_loss_clip": 1.04553938,
      "balance_loss_mlp": 1.02160072,
      "epoch": 0.42092289192845334,
      "flos": 18113414618880.0,
      "grad_norm": 1.487582695554992,
      "language_loss": 0.7641052,
      "learning_rate": 2.5999308319843743e-06,
      "loss": 0.78533787,
      "num_input_tokens_seen": 150279885,
      "step": 7001,
      "time_per_iteration": 2.5500075817108154
    },
    {
      "auxiliary_loss_clip": 0.01092944,
      "auxiliary_loss_mlp": 0.00802124,
      "balance_loss_clip": 1.04526448,
      "balance_loss_mlp": 1.03571391,
      "epoch": 0.4209830151811213,
      "flos": 20006588845440.0,
      "grad_norm": 1.4567946401476999,
      "language_loss": 0.86736441,
      "learning_rate": 2.5995592936110154e-06,
      "loss": 0.88631499,
      "num_input_tokens_seen": 150297390,
      "step": 7002,
      "time_per_iteration": 2.5626943111419678
    },
    {
      "auxiliary_loss_clip": 0.01084748,
      "auxiliary_loss_mlp": 0.01034753,
      "balance_loss_clip": 1.04329085,
      "balance_loss_mlp": 1.02162206,
      "epoch": 0.42104313843378927,
      "flos": 21978946604160.0,
      "grad_norm": 2.27376547838274,
      "language_loss": 0.6807127,
      "learning_rate": 2.5991877325023096e-06,
      "loss": 0.70190775,
      "num_input_tokens_seen": 150317390,
      "step": 7003,
      "time_per_iteration": 2.543154239654541
    },
    {
      "auxiliary_loss_clip": 0.01125428,
      "auxiliary_loss_mlp": 0.01040035,
      "balance_loss_clip": 1.04376161,
      "balance_loss_mlp": 1.02539587,
      "epoch": 0.42110326168645723,
      "flos": 25443966965760.0,
      "grad_norm": 3.083412188049608,
      "language_loss": 0.77476871,
      "learning_rate": 2.598816148672344e-06,
      "loss": 0.79642338,
      "num_input_tokens_seen": 150337455,
      "step": 7004,
      "time_per_iteration": 2.5100674629211426
    },
    {
      "auxiliary_loss_clip": 0.01118422,
      "auxiliary_loss_mlp": 0.01036475,
      "balance_loss_clip": 1.04269648,
      "balance_loss_mlp": 1.02197361,
      "epoch": 0.4211633849391252,
      "flos": 17822574195840.0,
      "grad_norm": 1.6799881442169071,
      "language_loss": 0.68037724,
      "learning_rate": 2.59844454213521e-06,
      "loss": 0.70192623,
      "num_input_tokens_seen": 150355385,
      "step": 7005,
      "time_per_iteration": 2.4322383403778076
    },
    {
      "auxiliary_loss_clip": 0.01111586,
      "auxiliary_loss_mlp": 0.01034687,
      "balance_loss_clip": 1.04303396,
      "balance_loss_mlp": 1.02103758,
      "epoch": 0.42122350819179316,
      "flos": 16282436791680.0,
      "grad_norm": 2.3828499137170533,
      "language_loss": 0.71933538,
      "learning_rate": 2.5980729129049994e-06,
      "loss": 0.74079812,
      "num_input_tokens_seen": 150371750,
      "step": 7006,
      "time_per_iteration": 2.4813992977142334
    },
    {
      "auxiliary_loss_clip": 0.0112184,
      "auxiliary_loss_mlp": 0.01033454,
      "balance_loss_clip": 1.04175329,
      "balance_loss_mlp": 1.01924467,
      "epoch": 0.4212836314444611,
      "flos": 19645866512640.0,
      "grad_norm": 2.9235464114451264,
      "language_loss": 0.71095145,
      "learning_rate": 2.5977012609958033e-06,
      "loss": 0.73250437,
      "num_input_tokens_seen": 150389955,
      "step": 7007,
      "time_per_iteration": 2.4288744926452637
    },
    {
      "auxiliary_loss_clip": 0.01096691,
      "auxiliary_loss_mlp": 0.00794606,
      "balance_loss_clip": 1.03995943,
      "balance_loss_mlp": 1.02189875,
      "epoch": 0.4213437546971291,
      "flos": 18369026778240.0,
      "grad_norm": 1.7609492528652881,
      "language_loss": 0.82313615,
      "learning_rate": 2.5973295864217166e-06,
      "loss": 0.84204912,
      "num_input_tokens_seen": 150405780,
      "step": 7008,
      "time_per_iteration": 2.5352776050567627
    },
    {
      "auxiliary_loss_clip": 0.01081049,
      "auxiliary_loss_mlp": 0.01038911,
      "balance_loss_clip": 1.04172683,
      "balance_loss_mlp": 1.02462411,
      "epoch": 0.42140387794979706,
      "flos": 27704507541120.0,
      "grad_norm": 1.6603963031220297,
      "language_loss": 0.71658206,
      "learning_rate": 2.596957889196831e-06,
      "loss": 0.73778164,
      "num_input_tokens_seen": 150425615,
      "step": 7009,
      "time_per_iteration": 2.5837604999542236
    },
    {
      "auxiliary_loss_clip": 0.01121318,
      "auxiliary_loss_mlp": 0.01032231,
      "balance_loss_clip": 1.0409776,
      "balance_loss_mlp": 1.01802719,
      "epoch": 0.4214640012024651,
      "flos": 28147071012480.0,
      "grad_norm": 8.56482520856282,
      "language_loss": 0.65433186,
      "learning_rate": 2.596586169335243e-06,
      "loss": 0.67586732,
      "num_input_tokens_seen": 150445765,
      "step": 7010,
      "time_per_iteration": 2.4966347217559814
    },
    {
      "auxiliary_loss_clip": 0.01083258,
      "auxiliary_loss_mlp": 0.01027284,
      "balance_loss_clip": 1.04075432,
      "balance_loss_mlp": 1.01371789,
      "epoch": 0.42152412445513304,
      "flos": 22997265177600.0,
      "grad_norm": 1.9573097546297447,
      "language_loss": 0.72714972,
      "learning_rate": 2.5962144268510477e-06,
      "loss": 0.74825513,
      "num_input_tokens_seen": 150464405,
      "step": 7011,
      "time_per_iteration": 2.5421650409698486
    },
    {
      "auxiliary_loss_clip": 0.01033526,
      "auxiliary_loss_mlp": 0.01001231,
      "balance_loss_clip": 1.01489544,
      "balance_loss_mlp": 0.99962157,
      "epoch": 0.421584247707801,
      "flos": 63749592938880.0,
      "grad_norm": 0.7836872857400162,
      "language_loss": 0.54328597,
      "learning_rate": 2.5958426617583417e-06,
      "loss": 0.56363356,
      "num_input_tokens_seen": 150520430,
      "step": 7012,
      "time_per_iteration": 2.9943134784698486
    },
    {
      "auxiliary_loss_clip": 0.01109142,
      "auxiliary_loss_mlp": 0.01031962,
      "balance_loss_clip": 1.0407716,
      "balance_loss_mlp": 1.01798511,
      "epoch": 0.421644370960469,
      "flos": 24314612474880.0,
      "grad_norm": 1.593901638867922,
      "language_loss": 0.78672719,
      "learning_rate": 2.5954708740712215e-06,
      "loss": 0.80813825,
      "num_input_tokens_seen": 150542610,
      "step": 7013,
      "time_per_iteration": 2.5130465030670166
    },
    {
      "auxiliary_loss_clip": 0.01120087,
      "auxiliary_loss_mlp": 0.01035561,
      "balance_loss_clip": 1.03979373,
      "balance_loss_mlp": 1.02065992,
      "epoch": 0.42170449421313694,
      "flos": 23440690575360.0,
      "grad_norm": 1.7054730141053818,
      "language_loss": 0.80984294,
      "learning_rate": 2.595099063803787e-06,
      "loss": 0.83139944,
      "num_input_tokens_seen": 150560970,
      "step": 7014,
      "time_per_iteration": 2.4900002479553223
    },
    {
      "auxiliary_loss_clip": 0.01108466,
      "auxiliary_loss_mlp": 0.01034277,
      "balance_loss_clip": 1.03981566,
      "balance_loss_mlp": 1.02055001,
      "epoch": 0.4217646174658049,
      "flos": 23695476721920.0,
      "grad_norm": 2.6921600227142477,
      "language_loss": 0.77849889,
      "learning_rate": 2.5947272309701354e-06,
      "loss": 0.79992628,
      "num_input_tokens_seen": 150582615,
      "step": 7015,
      "time_per_iteration": 2.525897741317749
    },
    {
      "auxiliary_loss_clip": 0.01123232,
      "auxiliary_loss_mlp": 0.01037424,
      "balance_loss_clip": 1.04288983,
      "balance_loss_mlp": 1.0229404,
      "epoch": 0.42182474071847287,
      "flos": 24971562270720.0,
      "grad_norm": 1.7465189692441223,
      "language_loss": 0.82268035,
      "learning_rate": 2.594355375584368e-06,
      "loss": 0.84428692,
      "num_input_tokens_seen": 150603640,
      "step": 7016,
      "time_per_iteration": 2.508558988571167
    },
    {
      "auxiliary_loss_clip": 0.01085932,
      "auxiliary_loss_mlp": 0.01033197,
      "balance_loss_clip": 1.04055715,
      "balance_loss_mlp": 1.018677,
      "epoch": 0.42188486397114083,
      "flos": 22856639431680.0,
      "grad_norm": 1.924778520899619,
      "language_loss": 0.68145865,
      "learning_rate": 2.593983497660586e-06,
      "loss": 0.70264989,
      "num_input_tokens_seen": 150622490,
      "step": 7017,
      "time_per_iteration": 2.5409436225891113
    },
    {
      "auxiliary_loss_clip": 0.01031411,
      "auxiliary_loss_mlp": 0.0100418,
      "balance_loss_clip": 1.01367784,
      "balance_loss_mlp": 1.0024519,
      "epoch": 0.4219449872238088,
      "flos": 66975700965120.0,
      "grad_norm": 0.7462697310055326,
      "language_loss": 0.59498364,
      "learning_rate": 2.5936115972128895e-06,
      "loss": 0.61533958,
      "num_input_tokens_seen": 150689545,
      "step": 7018,
      "time_per_iteration": 3.1377434730529785
    },
    {
      "auxiliary_loss_clip": 0.01107205,
      "auxiliary_loss_mlp": 0.01036991,
      "balance_loss_clip": 1.03940046,
      "balance_loss_mlp": 1.02260828,
      "epoch": 0.42200511047647676,
      "flos": 13115367745920.0,
      "grad_norm": 1.9871860115302116,
      "language_loss": 0.75314647,
      "learning_rate": 2.593239674255382e-06,
      "loss": 0.77458841,
      "num_input_tokens_seen": 150707610,
      "step": 7019,
      "time_per_iteration": 2.4950366020202637
    },
    {
      "auxiliary_loss_clip": 0.0110365,
      "auxiliary_loss_mlp": 0.0103437,
      "balance_loss_clip": 1.04106653,
      "balance_loss_mlp": 1.01961148,
      "epoch": 0.42206523372914473,
      "flos": 13991193066240.0,
      "grad_norm": 1.9303302974837744,
      "language_loss": 0.69317132,
      "learning_rate": 2.592867728802166e-06,
      "loss": 0.71455157,
      "num_input_tokens_seen": 150724530,
      "step": 7020,
      "time_per_iteration": 2.489919662475586
    },
    {
      "auxiliary_loss_clip": 0.01096449,
      "auxiliary_loss_mlp": 0.00788022,
      "balance_loss_clip": 1.0414139,
      "balance_loss_mlp": 1.00906694,
      "epoch": 0.4221253569818127,
      "flos": 21942317710080.0,
      "grad_norm": 1.8332652678251198,
      "language_loss": 0.81133378,
      "learning_rate": 2.592495760867347e-06,
      "loss": 0.8301785,
      "num_input_tokens_seen": 150742870,
      "step": 7021,
      "time_per_iteration": 2.52946138381958
    },
    {
      "auxiliary_loss_clip": 0.01054692,
      "auxiliary_loss_mlp": 0.01041952,
      "balance_loss_clip": 1.03792048,
      "balance_loss_mlp": 1.02527475,
      "epoch": 0.42218548023448066,
      "flos": 32192587071360.0,
      "grad_norm": 1.5956273483048007,
      "language_loss": 0.69894695,
      "learning_rate": 2.5921237704650293e-06,
      "loss": 0.71991342,
      "num_input_tokens_seen": 150765500,
      "step": 7022,
      "time_per_iteration": 2.689459800720215
    },
    {
      "auxiliary_loss_clip": 0.0110581,
      "auxiliary_loss_mlp": 0.01027256,
      "balance_loss_clip": 1.04167366,
      "balance_loss_mlp": 1.01588941,
      "epoch": 0.4222456034871487,
      "flos": 30118961894400.0,
      "grad_norm": 1.6207389432016326,
      "language_loss": 0.67661583,
      "learning_rate": 2.5917517576093188e-06,
      "loss": 0.69794655,
      "num_input_tokens_seen": 150784945,
      "step": 7023,
      "time_per_iteration": 2.564887046813965
    },
    {
      "auxiliary_loss_clip": 0.01092414,
      "auxiliary_loss_mlp": 0.0104398,
      "balance_loss_clip": 1.04066443,
      "balance_loss_mlp": 1.02780318,
      "epoch": 0.42230572673981664,
      "flos": 22127904305280.0,
      "grad_norm": 1.6756608384821896,
      "language_loss": 0.69243115,
      "learning_rate": 2.591379722314322e-06,
      "loss": 0.71379513,
      "num_input_tokens_seen": 150803120,
      "step": 7024,
      "time_per_iteration": 2.563551187515259
    },
    {
      "auxiliary_loss_clip": 0.01119943,
      "auxiliary_loss_mlp": 0.01034363,
      "balance_loss_clip": 1.04140472,
      "balance_loss_mlp": 1.02107143,
      "epoch": 0.4223658499924846,
      "flos": 22055077480320.0,
      "grad_norm": 1.6450332544471788,
      "language_loss": 0.76804036,
      "learning_rate": 2.591007664594147e-06,
      "loss": 0.78958338,
      "num_input_tokens_seen": 150823135,
      "step": 7025,
      "time_per_iteration": 3.921006679534912
    },
    {
      "auxiliary_loss_clip": 0.01088488,
      "auxiliary_loss_mlp": 0.01035032,
      "balance_loss_clip": 1.04015827,
      "balance_loss_mlp": 1.02189565,
      "epoch": 0.4224259732451526,
      "flos": 20410727742720.0,
      "grad_norm": 1.6683542186032743,
      "language_loss": 0.79504573,
      "learning_rate": 2.5906355844629024e-06,
      "loss": 0.81628096,
      "num_input_tokens_seen": 150842070,
      "step": 7026,
      "time_per_iteration": 2.5599634647369385
    },
    {
      "auxiliary_loss_clip": 0.01042549,
      "auxiliary_loss_mlp": 0.0100285,
      "balance_loss_clip": 1.01460767,
      "balance_loss_mlp": 1.00108552,
      "epoch": 0.42248609649782054,
      "flos": 62846655828480.0,
      "grad_norm": 0.7465182596330706,
      "language_loss": 0.62032455,
      "learning_rate": 2.5902634819346966e-06,
      "loss": 0.6407786,
      "num_input_tokens_seen": 150907450,
      "step": 7027,
      "time_per_iteration": 3.1564276218414307
    },
    {
      "auxiliary_loss_clip": 0.01116936,
      "auxiliary_loss_mlp": 0.01036496,
      "balance_loss_clip": 1.0409075,
      "balance_loss_mlp": 1.02341866,
      "epoch": 0.4225462197504885,
      "flos": 26249946289920.0,
      "grad_norm": 2.3872782288028733,
      "language_loss": 0.70789182,
      "learning_rate": 2.5898913570236414e-06,
      "loss": 0.72942615,
      "num_input_tokens_seen": 150928040,
      "step": 7028,
      "time_per_iteration": 3.863915205001831
    },
    {
      "auxiliary_loss_clip": 0.01103149,
      "auxiliary_loss_mlp": 0.01037605,
      "balance_loss_clip": 1.04210436,
      "balance_loss_mlp": 1.0236814,
      "epoch": 0.42260634300315647,
      "flos": 20521943228160.0,
      "grad_norm": 1.7250168682658376,
      "language_loss": 0.82116318,
      "learning_rate": 2.589519209743846e-06,
      "loss": 0.84257066,
      "num_input_tokens_seen": 150945760,
      "step": 7029,
      "time_per_iteration": 2.496976375579834
    },
    {
      "auxiliary_loss_clip": 0.01082398,
      "auxiliary_loss_mlp": 0.01039221,
      "balance_loss_clip": 1.04415274,
      "balance_loss_mlp": 1.02429581,
      "epoch": 0.42266646625582444,
      "flos": 24316731377280.0,
      "grad_norm": 1.9299042045927883,
      "language_loss": 0.74598867,
      "learning_rate": 2.589147040109424e-06,
      "loss": 0.76720488,
      "num_input_tokens_seen": 150965665,
      "step": 7030,
      "time_per_iteration": 2.6110968589782715
    },
    {
      "auxiliary_loss_clip": 0.01118719,
      "auxiliary_loss_mlp": 0.01034838,
      "balance_loss_clip": 1.04009414,
      "balance_loss_mlp": 1.02017498,
      "epoch": 0.4227265895084924,
      "flos": 24204151175040.0,
      "grad_norm": 1.9935267024109524,
      "language_loss": 0.86750114,
      "learning_rate": 2.588774848134486e-06,
      "loss": 0.88903666,
      "num_input_tokens_seen": 150982260,
      "step": 7031,
      "time_per_iteration": 3.907400608062744
    },
    {
      "auxiliary_loss_clip": 0.01110462,
      "auxiliary_loss_mlp": 0.01034936,
      "balance_loss_clip": 1.04213285,
      "balance_loss_mlp": 1.01978505,
      "epoch": 0.42278671276116037,
      "flos": 16909760845440.0,
      "grad_norm": 1.845244664076516,
      "language_loss": 0.7320013,
      "learning_rate": 2.5884026338331473e-06,
      "loss": 0.75345528,
      "num_input_tokens_seen": 150999990,
      "step": 7032,
      "time_per_iteration": 3.921236276626587
    },
    {
      "auxiliary_loss_clip": 0.0109206,
      "auxiliary_loss_mlp": 0.01041356,
      "balance_loss_clip": 1.03951287,
      "balance_loss_mlp": 1.02677703,
      "epoch": 0.42284683601382833,
      "flos": 25411073086080.0,
      "grad_norm": 1.5829485731842445,
      "language_loss": 0.70100796,
      "learning_rate": 2.5880303972195222e-06,
      "loss": 0.72234207,
      "num_input_tokens_seen": 151021105,
      "step": 7033,
      "time_per_iteration": 2.558011293411255
    },
    {
      "auxiliary_loss_clip": 0.0109847,
      "auxiliary_loss_mlp": 0.00788591,
      "balance_loss_clip": 1.04182005,
      "balance_loss_mlp": 1.01147199,
      "epoch": 0.4229069592664963,
      "flos": 23040322606080.0,
      "grad_norm": 1.7387664094970163,
      "language_loss": 0.90344977,
      "learning_rate": 2.5876581383077256e-06,
      "loss": 0.92232037,
      "num_input_tokens_seen": 151040665,
      "step": 7034,
      "time_per_iteration": 2.541513442993164
    },
    {
      "auxiliary_loss_clip": 0.01096732,
      "auxiliary_loss_mlp": 0.01035206,
      "balance_loss_clip": 1.04011202,
      "balance_loss_mlp": 1.02219486,
      "epoch": 0.42296708251916426,
      "flos": 26067448264320.0,
      "grad_norm": 1.5088055983426913,
      "language_loss": 0.77187264,
      "learning_rate": 2.5872858571118723e-06,
      "loss": 0.79319203,
      "num_input_tokens_seen": 151061240,
      "step": 7035,
      "time_per_iteration": 2.551865339279175
    },
    {
      "auxiliary_loss_clip": 0.01107698,
      "auxiliary_loss_mlp": 0.01038702,
      "balance_loss_clip": 1.04279757,
      "balance_loss_mlp": 1.02417004,
      "epoch": 0.4230272057718323,
      "flos": 19458376496640.0,
      "grad_norm": 1.7411051809564735,
      "language_loss": 0.82345867,
      "learning_rate": 2.5869135536460817e-06,
      "loss": 0.84492272,
      "num_input_tokens_seen": 151076870,
      "step": 7036,
      "time_per_iteration": 2.4677059650421143
    },
    {
      "auxiliary_loss_clip": 0.01097163,
      "auxiliary_loss_mlp": 0.01034033,
      "balance_loss_clip": 1.04301393,
      "balance_loss_mlp": 1.0204674,
      "epoch": 0.42308732902450025,
      "flos": 22383300983040.0,
      "grad_norm": 1.7405278127379054,
      "language_loss": 0.7047565,
      "learning_rate": 2.58654122792447e-06,
      "loss": 0.7260685,
      "num_input_tokens_seen": 151095110,
      "step": 7037,
      "time_per_iteration": 2.5231099128723145
    },
    {
      "auxiliary_loss_clip": 0.01081118,
      "auxiliary_loss_mlp": 0.00790488,
      "balance_loss_clip": 1.04280174,
      "balance_loss_mlp": 1.01375854,
      "epoch": 0.4231474522771682,
      "flos": 20995425331200.0,
      "grad_norm": 1.5312869060700665,
      "language_loss": 0.77501535,
      "learning_rate": 2.586168879961155e-06,
      "loss": 0.79373145,
      "num_input_tokens_seen": 151114355,
      "step": 7038,
      "time_per_iteration": 2.545872926712036
    },
    {
      "auxiliary_loss_clip": 0.01093844,
      "auxiliary_loss_mlp": 0.01042143,
      "balance_loss_clip": 1.04719996,
      "balance_loss_mlp": 1.02652717,
      "epoch": 0.4232075755298362,
      "flos": 14975863574400.0,
      "grad_norm": 2.685781190410265,
      "language_loss": 0.68071032,
      "learning_rate": 2.585796509770259e-06,
      "loss": 0.70207024,
      "num_input_tokens_seen": 151131505,
      "step": 7039,
      "time_per_iteration": 2.5527939796447754
    },
    {
      "auxiliary_loss_clip": 0.01114997,
      "auxiliary_loss_mlp": 0.01036409,
      "balance_loss_clip": 1.04200912,
      "balance_loss_mlp": 1.02097189,
      "epoch": 0.42326769878250414,
      "flos": 24532661986560.0,
      "grad_norm": 1.58429689968665,
      "language_loss": 0.75884426,
      "learning_rate": 2.5854241173658996e-06,
      "loss": 0.78035825,
      "num_input_tokens_seen": 151151555,
      "step": 7040,
      "time_per_iteration": 2.5146079063415527
    },
    {
      "auxiliary_loss_clip": 0.01113284,
      "auxiliary_loss_mlp": 0.01032714,
      "balance_loss_clip": 1.04413497,
      "balance_loss_mlp": 1.01855755,
      "epoch": 0.4233278220351721,
      "flos": 26870303105280.0,
      "grad_norm": 1.6026896661231305,
      "language_loss": 0.64890045,
      "learning_rate": 2.5850517027621996e-06,
      "loss": 0.67036039,
      "num_input_tokens_seen": 151172385,
      "step": 7041,
      "time_per_iteration": 2.555338144302368
    },
    {
      "auxiliary_loss_clip": 0.01097142,
      "auxiliary_loss_mlp": 0.01032789,
      "balance_loss_clip": 1.04332829,
      "balance_loss_mlp": 1.0177927,
      "epoch": 0.4233879452878401,
      "flos": 42814927463040.0,
      "grad_norm": 1.7860557601866012,
      "language_loss": 0.74348658,
      "learning_rate": 2.5846792659732803e-06,
      "loss": 0.76478589,
      "num_input_tokens_seen": 151194930,
      "step": 7042,
      "time_per_iteration": 2.7212564945220947
    },
    {
      "auxiliary_loss_clip": 0.01108187,
      "auxiliary_loss_mlp": 0.01031474,
      "balance_loss_clip": 1.0431149,
      "balance_loss_mlp": 1.01862335,
      "epoch": 0.42344806854050804,
      "flos": 25229006023680.0,
      "grad_norm": 1.4381619283044869,
      "language_loss": 0.82081324,
      "learning_rate": 2.5843068070132643e-06,
      "loss": 0.84220982,
      "num_input_tokens_seen": 151217905,
      "step": 7043,
      "time_per_iteration": 2.554152727127075
    },
    {
      "auxiliary_loss_clip": 0.01100938,
      "auxiliary_loss_mlp": 0.01040221,
      "balance_loss_clip": 1.04618871,
      "balance_loss_mlp": 1.02509904,
      "epoch": 0.423508191793176,
      "flos": 22778820616320.0,
      "grad_norm": 1.9052524483104685,
      "language_loss": 0.64750731,
      "learning_rate": 2.5839343258962763e-06,
      "loss": 0.66891885,
      "num_input_tokens_seen": 151234580,
      "step": 7044,
      "time_per_iteration": 2.5055863857269287
    },
    {
      "auxiliary_loss_clip": 0.01111929,
      "auxiliary_loss_mlp": 0.01044393,
      "balance_loss_clip": 1.04683697,
      "balance_loss_mlp": 1.02771533,
      "epoch": 0.42356831504584397,
      "flos": 34637493179520.0,
      "grad_norm": 1.7185582339098966,
      "language_loss": 0.75184834,
      "learning_rate": 2.5835618226364393e-06,
      "loss": 0.77341163,
      "num_input_tokens_seen": 151254765,
      "step": 7045,
      "time_per_iteration": 2.6097419261932373
    },
    {
      "auxiliary_loss_clip": 0.01086896,
      "auxiliary_loss_mlp": 0.01047174,
      "balance_loss_clip": 1.04320002,
      "balance_loss_mlp": 1.03184342,
      "epoch": 0.42362843829851193,
      "flos": 17596767346560.0,
      "grad_norm": 1.9691778424247808,
      "language_loss": 0.80431461,
      "learning_rate": 2.5831892972478797e-06,
      "loss": 0.82565534,
      "num_input_tokens_seen": 151269045,
      "step": 7046,
      "time_per_iteration": 2.5086309909820557
    },
    {
      "auxiliary_loss_clip": 0.01046193,
      "auxiliary_loss_mlp": 0.01039601,
      "balance_loss_clip": 1.04344606,
      "balance_loss_mlp": 1.02489018,
      "epoch": 0.4236885615511799,
      "flos": 22565691267840.0,
      "grad_norm": 3.6176522127067003,
      "language_loss": 0.77107334,
      "learning_rate": 2.5828167497447242e-06,
      "loss": 0.79193133,
      "num_input_tokens_seen": 151287530,
      "step": 7047,
      "time_per_iteration": 2.84773588180542
    },
    {
      "auxiliary_loss_clip": 0.01121333,
      "auxiliary_loss_mlp": 0.01035473,
      "balance_loss_clip": 1.04472959,
      "balance_loss_mlp": 1.02182388,
      "epoch": 0.42374868480384786,
      "flos": 26469216864000.0,
      "grad_norm": 2.2768527205339035,
      "language_loss": 0.68171704,
      "learning_rate": 2.582444180141098e-06,
      "loss": 0.70328516,
      "num_input_tokens_seen": 151308905,
      "step": 7048,
      "time_per_iteration": 2.7422633171081543
    },
    {
      "auxiliary_loss_clip": 0.01110229,
      "auxiliary_loss_mlp": 0.01039133,
      "balance_loss_clip": 1.0422622,
      "balance_loss_mlp": 1.02405286,
      "epoch": 0.4238088080565159,
      "flos": 20370220179840.0,
      "grad_norm": 1.7410379554693425,
      "language_loss": 0.77860463,
      "learning_rate": 2.5820715884511307e-06,
      "loss": 0.8000983,
      "num_input_tokens_seen": 151326525,
      "step": 7049,
      "time_per_iteration": 2.471752643585205
    },
    {
      "auxiliary_loss_clip": 0.01116212,
      "auxiliary_loss_mlp": 0.01038661,
      "balance_loss_clip": 1.04669595,
      "balance_loss_mlp": 1.02486849,
      "epoch": 0.42386893130918385,
      "flos": 21172105353600.0,
      "grad_norm": 1.658287175730911,
      "language_loss": 0.82416356,
      "learning_rate": 2.5816989746889504e-06,
      "loss": 0.8457123,
      "num_input_tokens_seen": 151344675,
      "step": 7050,
      "time_per_iteration": 2.5054450035095215
    },
    {
      "auxiliary_loss_clip": 0.01121895,
      "auxiliary_loss_mlp": 0.01033963,
      "balance_loss_clip": 1.04189181,
      "balance_loss_mlp": 1.02017617,
      "epoch": 0.4239290545618518,
      "flos": 17675627656320.0,
      "grad_norm": 2.090773503576969,
      "language_loss": 0.73692894,
      "learning_rate": 2.581326338868687e-06,
      "loss": 0.75848746,
      "num_input_tokens_seen": 151360730,
      "step": 7051,
      "time_per_iteration": 2.4067749977111816
    },
    {
      "auxiliary_loss_clip": 0.01092492,
      "auxiliary_loss_mlp": 0.01032303,
      "balance_loss_clip": 1.04457808,
      "balance_loss_mlp": 1.01901722,
      "epoch": 0.4239891778145198,
      "flos": 24314504734080.0,
      "grad_norm": 1.353913504927804,
      "language_loss": 0.86105639,
      "learning_rate": 2.5809536810044706e-06,
      "loss": 0.88230437,
      "num_input_tokens_seen": 151380445,
      "step": 7052,
      "time_per_iteration": 2.5676839351654053
    },
    {
      "auxiliary_loss_clip": 0.0110062,
      "auxiliary_loss_mlp": 0.01045616,
      "balance_loss_clip": 1.04392684,
      "balance_loss_mlp": 1.03102481,
      "epoch": 0.42404930106718774,
      "flos": 20558428467840.0,
      "grad_norm": 1.3968717581610424,
      "language_loss": 0.72682679,
      "learning_rate": 2.5805810011104323e-06,
      "loss": 0.74828911,
      "num_input_tokens_seen": 151399325,
      "step": 7053,
      "time_per_iteration": 2.5181994438171387
    },
    {
      "auxiliary_loss_clip": 0.01091391,
      "auxiliary_loss_mlp": 0.00790512,
      "balance_loss_clip": 1.04507017,
      "balance_loss_mlp": 1.01264977,
      "epoch": 0.4241094243198557,
      "flos": 22308067946880.0,
      "grad_norm": 1.6157702854212261,
      "language_loss": 0.82432997,
      "learning_rate": 2.580208299200704e-06,
      "loss": 0.84314901,
      "num_input_tokens_seen": 151417240,
      "step": 7054,
      "time_per_iteration": 2.5625245571136475
    },
    {
      "auxiliary_loss_clip": 0.01040327,
      "auxiliary_loss_mlp": 0.01007325,
      "balance_loss_clip": 1.02367318,
      "balance_loss_mlp": 1.00538146,
      "epoch": 0.4241695475725237,
      "flos": 70612445272320.0,
      "grad_norm": 0.7859987729284187,
      "language_loss": 0.60410035,
      "learning_rate": 2.5798355752894183e-06,
      "loss": 0.62457687,
      "num_input_tokens_seen": 151476015,
      "step": 7055,
      "time_per_iteration": 3.0514087677001953
    },
    {
      "auxiliary_loss_clip": 0.01124766,
      "auxiliary_loss_mlp": 0.01040938,
      "balance_loss_clip": 1.04450738,
      "balance_loss_mlp": 1.02600074,
      "epoch": 0.42422967082519164,
      "flos": 14027462824320.0,
      "grad_norm": 2.1851719011765454,
      "language_loss": 0.7677806,
      "learning_rate": 2.5794628293907107e-06,
      "loss": 0.78943765,
      "num_input_tokens_seen": 151492035,
      "step": 7056,
      "time_per_iteration": 2.434020757675171
    },
    {
      "auxiliary_loss_clip": 0.01115721,
      "auxiliary_loss_mlp": 0.01039249,
      "balance_loss_clip": 1.0433923,
      "balance_loss_mlp": 1.02344799,
      "epoch": 0.4242897940778596,
      "flos": 22345522853760.0,
      "grad_norm": 1.8880188697230667,
      "language_loss": 0.84405744,
      "learning_rate": 2.579090061518714e-06,
      "loss": 0.86560714,
      "num_input_tokens_seen": 151508970,
      "step": 7057,
      "time_per_iteration": 2.4808428287506104
    },
    {
      "auxiliary_loss_clip": 0.01087008,
      "auxiliary_loss_mlp": 0.01036114,
      "balance_loss_clip": 1.04494536,
      "balance_loss_mlp": 1.02133179,
      "epoch": 0.42434991733052757,
      "flos": 22595855713920.0,
      "grad_norm": 2.4967374978629633,
      "language_loss": 0.82603312,
      "learning_rate": 2.5787172716875642e-06,
      "loss": 0.84726429,
      "num_input_tokens_seen": 151525295,
      "step": 7058,
      "time_per_iteration": 2.5708515644073486
    },
    {
      "auxiliary_loss_clip": 0.01098802,
      "auxiliary_loss_mlp": 0.00788449,
      "balance_loss_clip": 1.04681826,
      "balance_loss_mlp": 1.01315928,
      "epoch": 0.42441004058319554,
      "flos": 20011437181440.0,
      "grad_norm": 1.6838609578451464,
      "language_loss": 0.80421871,
      "learning_rate": 2.5783444599113973e-06,
      "loss": 0.82309115,
      "num_input_tokens_seen": 151544435,
      "step": 7059,
      "time_per_iteration": 2.5195930004119873
    },
    {
      "auxiliary_loss_clip": 0.01124769,
      "auxiliary_loss_mlp": 0.01037345,
      "balance_loss_clip": 1.04378319,
      "balance_loss_mlp": 1.02191305,
      "epoch": 0.4244701638358635,
      "flos": 11144985235200.0,
      "grad_norm": 2.0413602233076547,
      "language_loss": 0.70270765,
      "learning_rate": 2.57797162620435e-06,
      "loss": 0.72432876,
      "num_input_tokens_seen": 151559520,
      "step": 7060,
      "time_per_iteration": 2.4492642879486084
    },
    {
      "auxiliary_loss_clip": 0.01115279,
      "auxiliary_loss_mlp": 0.01038803,
      "balance_loss_clip": 1.04511213,
      "balance_loss_mlp": 1.02498615,
      "epoch": 0.42453028708853147,
      "flos": 23987753688960.0,
      "grad_norm": 1.6621359157658246,
      "language_loss": 0.76451159,
      "learning_rate": 2.577598770580562e-06,
      "loss": 0.78605241,
      "num_input_tokens_seen": 151579790,
      "step": 7061,
      "time_per_iteration": 2.5197062492370605
    },
    {
      "auxiliary_loss_clip": 0.01117767,
      "auxiliary_loss_mlp": 0.01040767,
      "balance_loss_clip": 1.04680288,
      "balance_loss_mlp": 1.02518058,
      "epoch": 0.42459041034119943,
      "flos": 18406338030720.0,
      "grad_norm": 2.3499794846841,
      "language_loss": 0.72851765,
      "learning_rate": 2.5772258930541693e-06,
      "loss": 0.750103,
      "num_input_tokens_seen": 151598285,
      "step": 7062,
      "time_per_iteration": 2.4814164638519287
    },
    {
      "auxiliary_loss_clip": 0.01098518,
      "auxiliary_loss_mlp": 0.01042669,
      "balance_loss_clip": 1.04355764,
      "balance_loss_mlp": 1.02854824,
      "epoch": 0.42465053359386745,
      "flos": 20958006337920.0,
      "grad_norm": 1.696897952798624,
      "language_loss": 0.66294885,
      "learning_rate": 2.5768529936393137e-06,
      "loss": 0.68436074,
      "num_input_tokens_seen": 151615430,
      "step": 7063,
      "time_per_iteration": 2.495457649230957
    },
    {
      "auxiliary_loss_clip": 0.0108808,
      "auxiliary_loss_mlp": 0.00787092,
      "balance_loss_clip": 1.04106092,
      "balance_loss_mlp": 1.00971353,
      "epoch": 0.4247106568465354,
      "flos": 33106190520960.0,
      "grad_norm": 6.0736057190722095,
      "language_loss": 0.78385466,
      "learning_rate": 2.5764800723501354e-06,
      "loss": 0.8026064,
      "num_input_tokens_seen": 151637030,
      "step": 7064,
      "time_per_iteration": 4.02016544342041
    },
    {
      "auxiliary_loss_clip": 0.01122132,
      "auxiliary_loss_mlp": 0.01036455,
      "balance_loss_clip": 1.04181337,
      "balance_loss_mlp": 1.02221501,
      "epoch": 0.4247707800992034,
      "flos": 20046916840320.0,
      "grad_norm": 1.8067226083134649,
      "language_loss": 0.74775159,
      "learning_rate": 2.5761071292007736e-06,
      "loss": 0.76933742,
      "num_input_tokens_seen": 151655745,
      "step": 7065,
      "time_per_iteration": 2.446256637573242
    },
    {
      "auxiliary_loss_clip": 0.01113483,
      "auxiliary_loss_mlp": 0.01035231,
      "balance_loss_clip": 1.04508638,
      "balance_loss_mlp": 1.02047837,
      "epoch": 0.42483090335187135,
      "flos": 22385132576640.0,
      "grad_norm": 1.305923839861759,
      "language_loss": 0.72343969,
      "learning_rate": 2.5757341642053725e-06,
      "loss": 0.74492681,
      "num_input_tokens_seen": 151678040,
      "step": 7066,
      "time_per_iteration": 2.5459933280944824
    },
    {
      "auxiliary_loss_clip": 0.01087628,
      "auxiliary_loss_mlp": 0.01037353,
      "balance_loss_clip": 1.04357171,
      "balance_loss_mlp": 1.02205825,
      "epoch": 0.4248910266045393,
      "flos": 21356830022400.0,
      "grad_norm": 1.8065899594865165,
      "language_loss": 0.79706848,
      "learning_rate": 2.5753611773780745e-06,
      "loss": 0.81831825,
      "num_input_tokens_seen": 151696410,
      "step": 7067,
      "time_per_iteration": 3.9097342491149902
    },
    {
      "auxiliary_loss_clip": 0.01044395,
      "auxiliary_loss_mlp": 0.01006543,
      "balance_loss_clip": 1.01689756,
      "balance_loss_mlp": 1.00465989,
      "epoch": 0.4249511498572073,
      "flos": 64008114099840.0,
      "grad_norm": 0.9090269517716135,
      "language_loss": 0.63490307,
      "learning_rate": 2.574988168733022e-06,
      "loss": 0.65541244,
      "num_input_tokens_seen": 151756365,
      "step": 7068,
      "time_per_iteration": 2.99078369140625
    },
    {
      "auxiliary_loss_clip": 0.01123654,
      "auxiliary_loss_mlp": 0.01035546,
      "balance_loss_clip": 1.04316211,
      "balance_loss_mlp": 1.02001858,
      "epoch": 0.42501127310987524,
      "flos": 19607046888960.0,
      "grad_norm": 1.5049126963598385,
      "language_loss": 0.7222237,
      "learning_rate": 2.574615138284361e-06,
      "loss": 0.74381566,
      "num_input_tokens_seen": 151775165,
      "step": 7069,
      "time_per_iteration": 2.4509873390197754
    },
    {
      "auxiliary_loss_clip": 0.01123358,
      "auxiliary_loss_mlp": 0.01034542,
      "balance_loss_clip": 1.04268432,
      "balance_loss_mlp": 1.01867473,
      "epoch": 0.4250713963625432,
      "flos": 19462326992640.0,
      "grad_norm": 1.9546087447706204,
      "language_loss": 0.79238623,
      "learning_rate": 2.5742420860462364e-06,
      "loss": 0.81396526,
      "num_input_tokens_seen": 151792620,
      "step": 7070,
      "time_per_iteration": 3.7899694442749023
    },
    {
      "auxiliary_loss_clip": 0.01111104,
      "auxiliary_loss_mlp": 0.01031535,
      "balance_loss_clip": 1.04218507,
      "balance_loss_mlp": 1.01687217,
      "epoch": 0.4251315196152112,
      "flos": 25337707557120.0,
      "grad_norm": 2.055101139780295,
      "language_loss": 0.70003778,
      "learning_rate": 2.573869012032795e-06,
      "loss": 0.72146416,
      "num_input_tokens_seen": 151812850,
      "step": 7071,
      "time_per_iteration": 3.9286630153656006
    },
    {
      "auxiliary_loss_clip": 0.0112051,
      "auxiliary_loss_mlp": 0.01035534,
      "balance_loss_clip": 1.04078484,
      "balance_loss_mlp": 1.02194452,
      "epoch": 0.42519164286787914,
      "flos": 26359186527360.0,
      "grad_norm": 2.429418953200352,
      "language_loss": 0.70702225,
      "learning_rate": 2.5734959162581824e-06,
      "loss": 0.72858274,
      "num_input_tokens_seen": 151831785,
      "step": 7072,
      "time_per_iteration": 2.4975457191467285
    },
    {
      "auxiliary_loss_clip": 0.01087551,
      "auxiliary_loss_mlp": 0.01036267,
      "balance_loss_clip": 1.04201615,
      "balance_loss_mlp": 1.02176476,
      "epoch": 0.4252517661205471,
      "flos": 26031070765440.0,
      "grad_norm": 1.5782417985140955,
      "language_loss": 0.81610107,
      "learning_rate": 2.5731227987365475e-06,
      "loss": 0.83733922,
      "num_input_tokens_seen": 151853885,
      "step": 7073,
      "time_per_iteration": 2.5851187705993652
    },
    {
      "auxiliary_loss_clip": 0.01107558,
      "auxiliary_loss_mlp": 0.01034863,
      "balance_loss_clip": 1.04320014,
      "balance_loss_mlp": 1.02173221,
      "epoch": 0.42531188937321507,
      "flos": 12713635059840.0,
      "grad_norm": 2.31547168305089,
      "language_loss": 0.91184717,
      "learning_rate": 2.5727496594820386e-06,
      "loss": 0.93327141,
      "num_input_tokens_seen": 151871780,
      "step": 7074,
      "time_per_iteration": 2.455430269241333
    },
    {
      "auxiliary_loss_clip": 0.01116257,
      "auxiliary_loss_mlp": 0.00790122,
      "balance_loss_clip": 1.04280257,
      "balance_loss_mlp": 1.01097751,
      "epoch": 0.42537201262588303,
      "flos": 22091670460800.0,
      "grad_norm": 1.635618511304507,
      "language_loss": 0.64261442,
      "learning_rate": 2.572376498508805e-06,
      "loss": 0.66167819,
      "num_input_tokens_seen": 151891600,
      "step": 7075,
      "time_per_iteration": 2.5051214694976807
    },
    {
      "auxiliary_loss_clip": 0.01083047,
      "auxiliary_loss_mlp": 0.01028312,
      "balance_loss_clip": 1.04065859,
      "balance_loss_mlp": 1.01513374,
      "epoch": 0.42543213587855105,
      "flos": 23003119094400.0,
      "grad_norm": 1.5719245344824275,
      "language_loss": 0.73642933,
      "learning_rate": 2.5720033158309973e-06,
      "loss": 0.75754297,
      "num_input_tokens_seen": 151911330,
      "step": 7076,
      "time_per_iteration": 2.562434673309326
    },
    {
      "auxiliary_loss_clip": 0.01097137,
      "auxiliary_loss_mlp": 0.0104239,
      "balance_loss_clip": 1.04017162,
      "balance_loss_mlp": 1.02732205,
      "epoch": 0.425492259131219,
      "flos": 25082454533760.0,
      "grad_norm": 2.0252333586120788,
      "language_loss": 0.79090643,
      "learning_rate": 2.571630111462766e-06,
      "loss": 0.8123017,
      "num_input_tokens_seen": 151930355,
      "step": 7077,
      "time_per_iteration": 2.537177562713623
    },
    {
      "auxiliary_loss_clip": 0.01095039,
      "auxiliary_loss_mlp": 0.01032998,
      "balance_loss_clip": 1.04093552,
      "balance_loss_mlp": 1.02004039,
      "epoch": 0.425552382383887,
      "flos": 22816850140800.0,
      "grad_norm": 2.145074744117491,
      "language_loss": 0.72942531,
      "learning_rate": 2.571256885418265e-06,
      "loss": 0.75070572,
      "num_input_tokens_seen": 151949695,
      "step": 7078,
      "time_per_iteration": 2.5365304946899414
    },
    {
      "auxiliary_loss_clip": 0.01094665,
      "auxiliary_loss_mlp": 0.0103424,
      "balance_loss_clip": 1.04707587,
      "balance_loss_mlp": 1.02140093,
      "epoch": 0.42561250563655495,
      "flos": 13553585671680.0,
      "grad_norm": 1.9916311736525718,
      "language_loss": 0.80102324,
      "learning_rate": 2.5708836377116445e-06,
      "loss": 0.8223123,
      "num_input_tokens_seen": 151967640,
      "step": 7079,
      "time_per_iteration": 2.4939804077148438
    },
    {
      "auxiliary_loss_clip": 0.01111659,
      "auxiliary_loss_mlp": 0.01030311,
      "balance_loss_clip": 1.04563546,
      "balance_loss_mlp": 1.01774049,
      "epoch": 0.4256726288892229,
      "flos": 46978303023360.0,
      "grad_norm": 1.9679222126528748,
      "language_loss": 0.71801943,
      "learning_rate": 2.5705103683570592e-06,
      "loss": 0.73943913,
      "num_input_tokens_seen": 151994020,
      "step": 7080,
      "time_per_iteration": 2.73435378074646
    },
    {
      "auxiliary_loss_clip": 0.01119486,
      "auxiliary_loss_mlp": 0.01032371,
      "balance_loss_clip": 1.04110909,
      "balance_loss_mlp": 1.0191865,
      "epoch": 0.4257327521418909,
      "flos": 23586451966080.0,
      "grad_norm": 3.472488431829096,
      "language_loss": 0.80472827,
      "learning_rate": 2.5701370773686646e-06,
      "loss": 0.82624686,
      "num_input_tokens_seen": 152013415,
      "step": 7081,
      "time_per_iteration": 2.4659922122955322
    },
    {
      "auxiliary_loss_clip": 0.01097215,
      "auxiliary_loss_mlp": 0.01031174,
      "balance_loss_clip": 1.04017711,
      "balance_loss_mlp": 1.01791191,
      "epoch": 0.42579287539455885,
      "flos": 18989994124800.0,
      "grad_norm": 1.6062833608597828,
      "language_loss": 0.81467336,
      "learning_rate": 2.5697637647606138e-06,
      "loss": 0.83595723,
      "num_input_tokens_seen": 152030860,
      "step": 7082,
      "time_per_iteration": 2.5083959102630615
    },
    {
      "auxiliary_loss_clip": 0.01109137,
      "auxiliary_loss_mlp": 0.01036303,
      "balance_loss_clip": 1.04250383,
      "balance_loss_mlp": 1.02276123,
      "epoch": 0.4258529986472268,
      "flos": 25191910252800.0,
      "grad_norm": 1.7677048278392415,
      "language_loss": 0.69581205,
      "learning_rate": 2.569390430547065e-06,
      "loss": 0.71726644,
      "num_input_tokens_seen": 152050395,
      "step": 7083,
      "time_per_iteration": 2.520800828933716
    },
    {
      "auxiliary_loss_clip": 0.01038089,
      "auxiliary_loss_mlp": 0.01001579,
      "balance_loss_clip": 1.01978636,
      "balance_loss_mlp": 0.999982,
      "epoch": 0.4259131218998948,
      "flos": 69968280718080.0,
      "grad_norm": 0.8780290141810734,
      "language_loss": 0.67151326,
      "learning_rate": 2.569017074742173e-06,
      "loss": 0.69190991,
      "num_input_tokens_seen": 152113555,
      "step": 7084,
      "time_per_iteration": 3.195033073425293
    },
    {
      "auxiliary_loss_clip": 0.01107343,
      "auxiliary_loss_mlp": 0.01040154,
      "balance_loss_clip": 1.04236066,
      "balance_loss_mlp": 1.02539611,
      "epoch": 0.42597324515256274,
      "flos": 18004964480640.0,
      "grad_norm": 1.8699861184028506,
      "language_loss": 0.78371215,
      "learning_rate": 2.5686436973600964e-06,
      "loss": 0.80518711,
      "num_input_tokens_seen": 152131575,
      "step": 7085,
      "time_per_iteration": 2.5097341537475586
    },
    {
      "auxiliary_loss_clip": 0.0111941,
      "auxiliary_loss_mlp": 0.01044167,
      "balance_loss_clip": 1.04464507,
      "balance_loss_mlp": 1.02864587,
      "epoch": 0.4260333684052307,
      "flos": 15158792563200.0,
      "grad_norm": 2.1973511056593322,
      "language_loss": 0.76256859,
      "learning_rate": 2.568270298414995e-06,
      "loss": 0.78420436,
      "num_input_tokens_seen": 152149435,
      "step": 7086,
      "time_per_iteration": 2.465534210205078
    },
    {
      "auxiliary_loss_clip": 0.01096453,
      "auxiliary_loss_mlp": 0.01035312,
      "balance_loss_clip": 1.03982604,
      "balance_loss_mlp": 1.02123976,
      "epoch": 0.42609349165789867,
      "flos": 14939342421120.0,
      "grad_norm": 1.8729320279838206,
      "language_loss": 0.80228901,
      "learning_rate": 2.5678968779210255e-06,
      "loss": 0.82360667,
      "num_input_tokens_seen": 152166860,
      "step": 7087,
      "time_per_iteration": 2.4819304943084717
    },
    {
      "auxiliary_loss_clip": 0.01099782,
      "auxiliary_loss_mlp": 0.01031052,
      "balance_loss_clip": 1.04197216,
      "balance_loss_mlp": 1.01657355,
      "epoch": 0.42615361491056664,
      "flos": 23731961961600.0,
      "grad_norm": 1.6155266434852105,
      "language_loss": 0.66055441,
      "learning_rate": 2.5675234358923505e-06,
      "loss": 0.68186271,
      "num_input_tokens_seen": 152187475,
      "step": 7088,
      "time_per_iteration": 2.53523850440979
    },
    {
      "auxiliary_loss_clip": 0.0107733,
      "auxiliary_loss_mlp": 0.0103779,
      "balance_loss_clip": 1.04281831,
      "balance_loss_mlp": 1.02316344,
      "epoch": 0.42621373816323466,
      "flos": 24936441747840.0,
      "grad_norm": 2.159230052434351,
      "language_loss": 0.68153608,
      "learning_rate": 2.56714997234313e-06,
      "loss": 0.70268738,
      "num_input_tokens_seen": 152207235,
      "step": 7089,
      "time_per_iteration": 2.6273391246795654
    },
    {
      "auxiliary_loss_clip": 0.01073972,
      "auxiliary_loss_mlp": 0.01034763,
      "balance_loss_clip": 1.03900361,
      "balance_loss_mlp": 1.02070177,
      "epoch": 0.4262738614159026,
      "flos": 13552975140480.0,
      "grad_norm": 2.0846298702878987,
      "language_loss": 0.72946239,
      "learning_rate": 2.566776487287525e-06,
      "loss": 0.75054967,
      "num_input_tokens_seen": 152224240,
      "step": 7090,
      "time_per_iteration": 2.541851282119751
    },
    {
      "auxiliary_loss_clip": 0.01102356,
      "auxiliary_loss_mlp": 0.01039731,
      "balance_loss_clip": 1.04137409,
      "balance_loss_mlp": 1.0257479,
      "epoch": 0.4263339846685706,
      "flos": 29748794284800.0,
      "grad_norm": 1.9309960364079948,
      "language_loss": 0.75740469,
      "learning_rate": 2.5664029807396994e-06,
      "loss": 0.77882564,
      "num_input_tokens_seen": 152242595,
      "step": 7091,
      "time_per_iteration": 2.581172227859497
    },
    {
      "auxiliary_loss_clip": 0.01067988,
      "auxiliary_loss_mlp": 0.01028689,
      "balance_loss_clip": 1.04044139,
      "balance_loss_mlp": 1.01644027,
      "epoch": 0.42639410792123855,
      "flos": 16834204586880.0,
      "grad_norm": 1.5947324425687468,
      "language_loss": 0.82147139,
      "learning_rate": 2.5660294527138156e-06,
      "loss": 0.84243816,
      "num_input_tokens_seen": 152260840,
      "step": 7092,
      "time_per_iteration": 2.55586314201355
    },
    {
      "auxiliary_loss_clip": 0.0110479,
      "auxiliary_loss_mlp": 0.01039932,
      "balance_loss_clip": 1.04292607,
      "balance_loss_mlp": 1.02590156,
      "epoch": 0.4264542311739065,
      "flos": 28763118195840.0,
      "grad_norm": 1.9357148956136556,
      "language_loss": 0.73959994,
      "learning_rate": 2.565655903224038e-06,
      "loss": 0.76104718,
      "num_input_tokens_seen": 152280580,
      "step": 7093,
      "time_per_iteration": 2.565652847290039
    },
    {
      "auxiliary_loss_clip": 0.01111192,
      "auxiliary_loss_mlp": 0.01032411,
      "balance_loss_clip": 1.04122472,
      "balance_loss_mlp": 1.01816523,
      "epoch": 0.4265143544265745,
      "flos": 24713615727360.0,
      "grad_norm": 3.5015335829004868,
      "language_loss": 0.6971246,
      "learning_rate": 2.565282332284532e-06,
      "loss": 0.71856064,
      "num_input_tokens_seen": 152298455,
      "step": 7094,
      "time_per_iteration": 2.4964139461517334
    },
    {
      "auxiliary_loss_clip": 0.01087608,
      "auxiliary_loss_mlp": 0.01034851,
      "balance_loss_clip": 1.04272807,
      "balance_loss_mlp": 1.02082586,
      "epoch": 0.42657447767924245,
      "flos": 21865971352320.0,
      "grad_norm": 1.5966982751996088,
      "language_loss": 0.81757629,
      "learning_rate": 2.564908739909464e-06,
      "loss": 0.83880091,
      "num_input_tokens_seen": 152316995,
      "step": 7095,
      "time_per_iteration": 2.546546697616577
    },
    {
      "auxiliary_loss_clip": 0.01122847,
      "auxiliary_loss_mlp": 0.01042299,
      "balance_loss_clip": 1.04306161,
      "balance_loss_mlp": 1.02817321,
      "epoch": 0.4266346009319104,
      "flos": 21470236237440.0,
      "grad_norm": 2.324569006395003,
      "language_loss": 0.80066139,
      "learning_rate": 2.5645351261129996e-06,
      "loss": 0.82231283,
      "num_input_tokens_seen": 152334800,
      "step": 7096,
      "time_per_iteration": 2.4624435901641846
    },
    {
      "auxiliary_loss_clip": 0.01116475,
      "auxiliary_loss_mlp": 0.01035747,
      "balance_loss_clip": 1.04503489,
      "balance_loss_mlp": 1.02145362,
      "epoch": 0.4266947241845784,
      "flos": 25519379569920.0,
      "grad_norm": 1.9758945759903384,
      "language_loss": 0.65559256,
      "learning_rate": 2.5641614909093066e-06,
      "loss": 0.67711484,
      "num_input_tokens_seen": 152355175,
      "step": 7097,
      "time_per_iteration": 2.5658419132232666
    },
    {
      "auxiliary_loss_clip": 0.01097219,
      "auxiliary_loss_mlp": 0.0103032,
      "balance_loss_clip": 1.04368246,
      "balance_loss_mlp": 1.01668847,
      "epoch": 0.42675484743724634,
      "flos": 26541217676160.0,
      "grad_norm": 1.8914528928860685,
      "language_loss": 0.74134886,
      "learning_rate": 2.5637878343125535e-06,
      "loss": 0.76262432,
      "num_input_tokens_seen": 152377245,
      "step": 7098,
      "time_per_iteration": 2.5605335235595703
    },
    {
      "auxiliary_loss_clip": 0.01108957,
      "auxiliary_loss_mlp": 0.01030872,
      "balance_loss_clip": 1.04150581,
      "balance_loss_mlp": 1.01761043,
      "epoch": 0.4268149706899143,
      "flos": 23112718467840.0,
      "grad_norm": 1.7248997144902938,
      "language_loss": 0.75221795,
      "learning_rate": 2.5634141563369086e-06,
      "loss": 0.77361625,
      "num_input_tokens_seen": 152396985,
      "step": 7099,
      "time_per_iteration": 2.5321266651153564
    },
    {
      "auxiliary_loss_clip": 0.0110132,
      "auxiliary_loss_mlp": 0.0103901,
      "balance_loss_clip": 1.0418067,
      "balance_loss_mlp": 1.0245862,
      "epoch": 0.4268750939425823,
      "flos": 22706532495360.0,
      "grad_norm": 2.099813860441557,
      "language_loss": 0.82643461,
      "learning_rate": 2.5630404569965432e-06,
      "loss": 0.84783792,
      "num_input_tokens_seen": 152415590,
      "step": 7100,
      "time_per_iteration": 2.5286149978637695
    },
    {
      "auxiliary_loss_clip": 0.01101945,
      "auxiliary_loss_mlp": 0.01030959,
      "balance_loss_clip": 1.04224038,
      "balance_loss_mlp": 1.01757765,
      "epoch": 0.42693521719525024,
      "flos": 25374875155200.0,
      "grad_norm": 1.7806678964552969,
      "language_loss": 0.8222096,
      "learning_rate": 2.562666736305627e-06,
      "loss": 0.8435387,
      "num_input_tokens_seen": 152436735,
      "step": 7101,
      "time_per_iteration": 2.5608038902282715
    },
    {
      "auxiliary_loss_clip": 0.01126285,
      "auxiliary_loss_mlp": 0.01033855,
      "balance_loss_clip": 1.04487562,
      "balance_loss_mlp": 1.01935351,
      "epoch": 0.42699534044791826,
      "flos": 18150689957760.0,
      "grad_norm": 1.8543926593252629,
      "language_loss": 0.7234143,
      "learning_rate": 2.5622929942783314e-06,
      "loss": 0.74501568,
      "num_input_tokens_seen": 152455685,
      "step": 7102,
      "time_per_iteration": 3.832385778427124
    },
    {
      "auxiliary_loss_clip": 0.01110354,
      "auxiliary_loss_mlp": 0.01032876,
      "balance_loss_clip": 1.043136,
      "balance_loss_mlp": 1.01948881,
      "epoch": 0.4270554637005862,
      "flos": 13698413308800.0,
      "grad_norm": 1.7767134499951431,
      "language_loss": 0.82566017,
      "learning_rate": 2.5619192309288297e-06,
      "loss": 0.84709251,
      "num_input_tokens_seen": 152473500,
      "step": 7103,
      "time_per_iteration": 2.4593327045440674
    },
    {
      "auxiliary_loss_clip": 0.01097537,
      "auxiliary_loss_mlp": 0.01045843,
      "balance_loss_clip": 1.03955388,
      "balance_loss_mlp": 1.02911782,
      "epoch": 0.4271155869532542,
      "flos": 17493596507520.0,
      "grad_norm": 1.9885747918755825,
      "language_loss": 0.73549271,
      "learning_rate": 2.561545446271294e-06,
      "loss": 0.75692654,
      "num_input_tokens_seen": 152491320,
      "step": 7104,
      "time_per_iteration": 2.4987401962280273
    },
    {
      "auxiliary_loss_clip": 0.01107817,
      "auxiliary_loss_mlp": 0.0103672,
      "balance_loss_clip": 1.04230475,
      "balance_loss_mlp": 1.02261162,
      "epoch": 0.42717571020592215,
      "flos": 32452293381120.0,
      "grad_norm": 2.246278937984321,
      "language_loss": 0.75245345,
      "learning_rate": 2.5611716403198987e-06,
      "loss": 0.77389878,
      "num_input_tokens_seen": 152511970,
      "step": 7105,
      "time_per_iteration": 3.954803466796875
    },
    {
      "auxiliary_loss_clip": 0.01126078,
      "auxiliary_loss_mlp": 0.01031517,
      "balance_loss_clip": 1.04572642,
      "balance_loss_mlp": 1.01855326,
      "epoch": 0.4272358334585901,
      "flos": 16253062444800.0,
      "grad_norm": 2.1524027328449438,
      "language_loss": 0.76927686,
      "learning_rate": 2.560797813088819e-06,
      "loss": 0.79085284,
      "num_input_tokens_seen": 152530515,
      "step": 7106,
      "time_per_iteration": 2.4423468112945557
    },
    {
      "auxiliary_loss_clip": 0.01100343,
      "auxiliary_loss_mlp": 0.01032191,
      "balance_loss_clip": 1.04261899,
      "balance_loss_mlp": 1.01882124,
      "epoch": 0.4272959567112581,
      "flos": 24200092938240.0,
      "grad_norm": 1.8901180011416496,
      "language_loss": 0.80246234,
      "learning_rate": 2.560423964592229e-06,
      "loss": 0.82378769,
      "num_input_tokens_seen": 152549295,
      "step": 7107,
      "time_per_iteration": 2.5143795013427734
    },
    {
      "auxiliary_loss_clip": 0.01076685,
      "auxiliary_loss_mlp": 0.01037212,
      "balance_loss_clip": 1.04397619,
      "balance_loss_mlp": 1.02301979,
      "epoch": 0.42735607996392605,
      "flos": 27963495578880.0,
      "grad_norm": 1.3439445640469168,
      "language_loss": 0.68042636,
      "learning_rate": 2.5600500948443075e-06,
      "loss": 0.70156533,
      "num_input_tokens_seen": 152570725,
      "step": 7108,
      "time_per_iteration": 2.6578423976898193
    },
    {
      "auxiliary_loss_clip": 0.01103648,
      "auxiliary_loss_mlp": 0.01036918,
      "balance_loss_clip": 1.04338205,
      "balance_loss_mlp": 1.02371001,
      "epoch": 0.427416203216594,
      "flos": 20295597674880.0,
      "grad_norm": 1.7154359516174509,
      "language_loss": 0.712309,
      "learning_rate": 2.5596762038592294e-06,
      "loss": 0.7337147,
      "num_input_tokens_seen": 152588950,
      "step": 7109,
      "time_per_iteration": 3.881704568862915
    },
    {
      "auxiliary_loss_clip": 0.01110112,
      "auxiliary_loss_mlp": 0.01035438,
      "balance_loss_clip": 1.04170704,
      "balance_loss_mlp": 1.01920152,
      "epoch": 0.427476326469262,
      "flos": 26943955943040.0,
      "grad_norm": 6.044346281657514,
      "language_loss": 0.647493,
      "learning_rate": 2.559302291651174e-06,
      "loss": 0.66894847,
      "num_input_tokens_seen": 152608965,
      "step": 7110,
      "time_per_iteration": 3.9058444499969482
    },
    {
      "auxiliary_loss_clip": 0.01124197,
      "auxiliary_loss_mlp": 0.00789898,
      "balance_loss_clip": 1.04377913,
      "balance_loss_mlp": 1.01349866,
      "epoch": 0.42753644972192995,
      "flos": 25702847262720.0,
      "grad_norm": 2.4068797476022916,
      "language_loss": 0.76462936,
      "learning_rate": 2.5589283582343197e-06,
      "loss": 0.78377032,
      "num_input_tokens_seen": 152630220,
      "step": 7111,
      "time_per_iteration": 2.502591609954834
    },
    {
      "auxiliary_loss_clip": 0.01092769,
      "auxiliary_loss_mlp": 0.01029836,
      "balance_loss_clip": 1.04361737,
      "balance_loss_mlp": 1.01548958,
      "epoch": 0.4275965729745979,
      "flos": 18767419499520.0,
      "grad_norm": 1.6874259178605695,
      "language_loss": 0.72963029,
      "learning_rate": 2.558554403622845e-06,
      "loss": 0.75085628,
      "num_input_tokens_seen": 152648835,
      "step": 7112,
      "time_per_iteration": 2.524606943130493
    },
    {
      "auxiliary_loss_clip": 0.01095108,
      "auxiliary_loss_mlp": 0.01039567,
      "balance_loss_clip": 1.04049313,
      "balance_loss_mlp": 1.02612591,
      "epoch": 0.4276566962272659,
      "flos": 23764424878080.0,
      "grad_norm": 1.815185376515291,
      "language_loss": 0.71475184,
      "learning_rate": 2.5581804278309323e-06,
      "loss": 0.73609853,
      "num_input_tokens_seen": 152668375,
      "step": 7113,
      "time_per_iteration": 2.5288755893707275
    },
    {
      "auxiliary_loss_clip": 0.01117542,
      "auxiliary_loss_mlp": 0.01038669,
      "balance_loss_clip": 1.04764462,
      "balance_loss_mlp": 1.02440584,
      "epoch": 0.42771681947993384,
      "flos": 22492505306880.0,
      "grad_norm": 1.7222820990261507,
      "language_loss": 0.62053466,
      "learning_rate": 2.5578064308727617e-06,
      "loss": 0.64209676,
      "num_input_tokens_seen": 152689725,
      "step": 7114,
      "time_per_iteration": 2.5182688236236572
    },
    {
      "auxiliary_loss_clip": 0.01118845,
      "auxiliary_loss_mlp": 0.01041219,
      "balance_loss_clip": 1.04659581,
      "balance_loss_mlp": 1.02494121,
      "epoch": 0.42777694273260186,
      "flos": 25044712318080.0,
      "grad_norm": 2.5162375368993906,
      "language_loss": 0.64796162,
      "learning_rate": 2.5574324127625153e-06,
      "loss": 0.66956222,
      "num_input_tokens_seen": 152709375,
      "step": 7115,
      "time_per_iteration": 2.5169191360473633
    },
    {
      "auxiliary_loss_clip": 0.01097487,
      "auxiliary_loss_mlp": 0.01035211,
      "balance_loss_clip": 1.04078865,
      "balance_loss_mlp": 1.02191877,
      "epoch": 0.4278370659852698,
      "flos": 18661519226880.0,
      "grad_norm": 1.4806147530669242,
      "language_loss": 0.73880363,
      "learning_rate": 2.5570583735143753e-06,
      "loss": 0.76013064,
      "num_input_tokens_seen": 152727510,
      "step": 7116,
      "time_per_iteration": 2.502995729446411
    },
    {
      "auxiliary_loss_clip": 0.01094051,
      "auxiliary_loss_mlp": 0.01037587,
      "balance_loss_clip": 1.03930879,
      "balance_loss_mlp": 1.02468276,
      "epoch": 0.4278971892379378,
      "flos": 27308269635840.0,
      "grad_norm": 1.599206316013892,
      "language_loss": 0.69298828,
      "learning_rate": 2.5566843131425275e-06,
      "loss": 0.71430469,
      "num_input_tokens_seen": 152746670,
      "step": 7117,
      "time_per_iteration": 2.566161870956421
    },
    {
      "auxiliary_loss_clip": 0.01099804,
      "auxiliary_loss_mlp": 0.01038877,
      "balance_loss_clip": 1.04556727,
      "balance_loss_mlp": 1.02510238,
      "epoch": 0.42795731249060576,
      "flos": 12888698970240.0,
      "grad_norm": 2.1870373064494184,
      "language_loss": 0.70131552,
      "learning_rate": 2.5563102316611536e-06,
      "loss": 0.72270238,
      "num_input_tokens_seen": 152760545,
      "step": 7118,
      "time_per_iteration": 2.5052826404571533
    },
    {
      "auxiliary_loss_clip": 0.0108544,
      "auxiliary_loss_mlp": 0.01045679,
      "balance_loss_clip": 1.04267478,
      "balance_loss_mlp": 1.03042006,
      "epoch": 0.4280174357432737,
      "flos": 33401448316800.0,
      "grad_norm": 1.7854188117569785,
      "language_loss": 0.74760103,
      "learning_rate": 2.55593612908444e-06,
      "loss": 0.7689122,
      "num_input_tokens_seen": 152780970,
      "step": 7119,
      "time_per_iteration": 2.657177448272705
    },
    {
      "auxiliary_loss_clip": 0.01060564,
      "auxiliary_loss_mlp": 0.01032496,
      "balance_loss_clip": 1.04095244,
      "balance_loss_mlp": 1.01851881,
      "epoch": 0.4280775589959417,
      "flos": 18259104182400.0,
      "grad_norm": 2.205306109147953,
      "language_loss": 0.7465024,
      "learning_rate": 2.555562005426573e-06,
      "loss": 0.76743305,
      "num_input_tokens_seen": 152798475,
      "step": 7120,
      "time_per_iteration": 2.577143669128418
    },
    {
      "auxiliary_loss_clip": 0.01100396,
      "auxiliary_loss_mlp": 0.00786896,
      "balance_loss_clip": 1.04440749,
      "balance_loss_mlp": 1.00939274,
      "epoch": 0.42813768224860965,
      "flos": 21471277731840.0,
      "grad_norm": 2.4938477814880002,
      "language_loss": 0.77051002,
      "learning_rate": 2.5551878607017385e-06,
      "loss": 0.78938293,
      "num_input_tokens_seen": 152817555,
      "step": 7121,
      "time_per_iteration": 2.542776584625244
    },
    {
      "auxiliary_loss_clip": 0.01101544,
      "auxiliary_loss_mlp": 0.01034834,
      "balance_loss_clip": 1.04453766,
      "balance_loss_mlp": 1.02216172,
      "epoch": 0.4281978055012776,
      "flos": 15669262696320.0,
      "grad_norm": 1.9211332540588129,
      "language_loss": 0.85617977,
      "learning_rate": 2.554813694924126e-06,
      "loss": 0.87754351,
      "num_input_tokens_seen": 152836295,
      "step": 7122,
      "time_per_iteration": 2.508695363998413
    },
    {
      "auxiliary_loss_clip": 0.01070032,
      "auxiliary_loss_mlp": 0.01033916,
      "balance_loss_clip": 1.03951693,
      "balance_loss_mlp": 1.01971269,
      "epoch": 0.4282579287539456,
      "flos": 17712005155200.0,
      "grad_norm": 1.6602051492569085,
      "language_loss": 0.80760944,
      "learning_rate": 2.554439508107921e-06,
      "loss": 0.82864887,
      "num_input_tokens_seen": 152854950,
      "step": 7123,
      "time_per_iteration": 2.601229190826416
    },
    {
      "auxiliary_loss_clip": 0.01083572,
      "auxiliary_loss_mlp": 0.0103561,
      "balance_loss_clip": 1.04765856,
      "balance_loss_mlp": 1.02213979,
      "epoch": 0.42831805200661355,
      "flos": 19281157770240.0,
      "grad_norm": 1.626677010525872,
      "language_loss": 0.80615515,
      "learning_rate": 2.5540653002673153e-06,
      "loss": 0.82734692,
      "num_input_tokens_seen": 152873995,
      "step": 7124,
      "time_per_iteration": 2.5472359657287598
    },
    {
      "auxiliary_loss_clip": 0.01113052,
      "auxiliary_loss_mlp": 0.01038273,
      "balance_loss_clip": 1.04505038,
      "balance_loss_mlp": 1.02438545,
      "epoch": 0.4283781752592815,
      "flos": 19792633484160.0,
      "grad_norm": 1.659274243374459,
      "language_loss": 0.8032515,
      "learning_rate": 2.553691071416498e-06,
      "loss": 0.82476479,
      "num_input_tokens_seen": 152892925,
      "step": 7125,
      "time_per_iteration": 2.475978374481201
    },
    {
      "auxiliary_loss_clip": 0.01121633,
      "auxiliary_loss_mlp": 0.00787718,
      "balance_loss_clip": 1.04536796,
      "balance_loss_mlp": 1.01077163,
      "epoch": 0.4284382985119495,
      "flos": 16508064072960.0,
      "grad_norm": 1.8285167548858474,
      "language_loss": 0.74959713,
      "learning_rate": 2.553316821569659e-06,
      "loss": 0.76869065,
      "num_input_tokens_seen": 152910935,
      "step": 7126,
      "time_per_iteration": 2.442854881286621
    },
    {
      "auxiliary_loss_clip": 0.0111154,
      "auxiliary_loss_mlp": 0.0103346,
      "balance_loss_clip": 1.04463339,
      "balance_loss_mlp": 1.01941705,
      "epoch": 0.42849842176461744,
      "flos": 23330767979520.0,
      "grad_norm": 1.7013304390131359,
      "language_loss": 0.81430924,
      "learning_rate": 2.5529425507409913e-06,
      "loss": 0.83575916,
      "num_input_tokens_seen": 152931030,
      "step": 7127,
      "time_per_iteration": 2.4904518127441406
    },
    {
      "auxiliary_loss_clip": 0.0107768,
      "auxiliary_loss_mlp": 0.01034989,
      "balance_loss_clip": 1.04424167,
      "balance_loss_mlp": 1.02104712,
      "epoch": 0.4285585450172854,
      "flos": 17274433674240.0,
      "grad_norm": 1.8052341028308765,
      "language_loss": 0.76152551,
      "learning_rate": 2.5525682589446867e-06,
      "loss": 0.7826522,
      "num_input_tokens_seen": 152948085,
      "step": 7128,
      "time_per_iteration": 2.560262680053711
    },
    {
      "auxiliary_loss_clip": 0.0107855,
      "auxiliary_loss_mlp": 0.01030312,
      "balance_loss_clip": 1.04392362,
      "balance_loss_mlp": 1.01606679,
      "epoch": 0.42861866826995343,
      "flos": 24279599692800.0,
      "grad_norm": 1.8723746445999567,
      "language_loss": 0.74218053,
      "learning_rate": 2.552193946194937e-06,
      "loss": 0.76326919,
      "num_input_tokens_seen": 152966265,
      "step": 7129,
      "time_per_iteration": 2.5935497283935547
    },
    {
      "auxiliary_loss_clip": 0.01116525,
      "auxiliary_loss_mlp": 0.00788173,
      "balance_loss_clip": 1.04926801,
      "balance_loss_mlp": 1.01015127,
      "epoch": 0.4286787915226214,
      "flos": 24353108876160.0,
      "grad_norm": 1.8165673603095516,
      "language_loss": 0.78034747,
      "learning_rate": 2.5518196125059394e-06,
      "loss": 0.79939449,
      "num_input_tokens_seen": 152986775,
      "step": 7130,
      "time_per_iteration": 2.5303232669830322
    },
    {
      "auxiliary_loss_clip": 0.01105335,
      "auxiliary_loss_mlp": 0.01034821,
      "balance_loss_clip": 1.04727924,
      "balance_loss_mlp": 1.02055192,
      "epoch": 0.42873891477528936,
      "flos": 15449992122240.0,
      "grad_norm": 2.1265040698941378,
      "language_loss": 0.73839617,
      "learning_rate": 2.551445257891886e-06,
      "loss": 0.75979769,
      "num_input_tokens_seen": 153003595,
      "step": 7131,
      "time_per_iteration": 2.4886181354522705
    },
    {
      "auxiliary_loss_clip": 0.01107674,
      "auxiliary_loss_mlp": 0.01036924,
      "balance_loss_clip": 1.04755306,
      "balance_loss_mlp": 1.02246952,
      "epoch": 0.4287990380279573,
      "flos": 17639573379840.0,
      "grad_norm": 2.2220991670556933,
      "language_loss": 0.77415675,
      "learning_rate": 2.551070882366973e-06,
      "loss": 0.79560274,
      "num_input_tokens_seen": 153021960,
      "step": 7132,
      "time_per_iteration": 2.495535373687744
    },
    {
      "auxiliary_loss_clip": 0.01087886,
      "auxiliary_loss_mlp": 0.00789746,
      "balance_loss_clip": 1.04757309,
      "balance_loss_mlp": 1.01407492,
      "epoch": 0.4288591612806253,
      "flos": 27162328677120.0,
      "grad_norm": 1.7216307097934347,
      "language_loss": 0.78607982,
      "learning_rate": 2.550696485945397e-06,
      "loss": 0.80485618,
      "num_input_tokens_seen": 153042110,
      "step": 7133,
      "time_per_iteration": 2.6045432090759277
    },
    {
      "auxiliary_loss_clip": 0.01099091,
      "auxiliary_loss_mlp": 0.0103573,
      "balance_loss_clip": 1.04480028,
      "balance_loss_mlp": 1.0220921,
      "epoch": 0.42891928453329325,
      "flos": 17163182275200.0,
      "grad_norm": 1.7616881695001934,
      "language_loss": 0.7514056,
      "learning_rate": 2.550322068641355e-06,
      "loss": 0.77275383,
      "num_input_tokens_seen": 153058925,
      "step": 7134,
      "time_per_iteration": 2.497849702835083
    },
    {
      "auxiliary_loss_clip": 0.01104864,
      "auxiliary_loss_mlp": 0.01040039,
      "balance_loss_clip": 1.04079866,
      "balance_loss_mlp": 1.0261873,
      "epoch": 0.4289794077859612,
      "flos": 18187031543040.0,
      "grad_norm": 1.8838177514168155,
      "language_loss": 0.83804458,
      "learning_rate": 2.5499476304690455e-06,
      "loss": 0.85949361,
      "num_input_tokens_seen": 153078070,
      "step": 7135,
      "time_per_iteration": 2.457148313522339
    },
    {
      "auxiliary_loss_clip": 0.01054817,
      "auxiliary_loss_mlp": 0.01038818,
      "balance_loss_clip": 1.04291189,
      "balance_loss_mlp": 1.02332091,
      "epoch": 0.4290395310386292,
      "flos": 28256885867520.0,
      "grad_norm": 1.956002903236043,
      "language_loss": 0.75080013,
      "learning_rate": 2.549573171442666e-06,
      "loss": 0.77173638,
      "num_input_tokens_seen": 153096680,
      "step": 7136,
      "time_per_iteration": 2.6529407501220703
    },
    {
      "auxiliary_loss_clip": 0.01108915,
      "auxiliary_loss_mlp": 0.01032643,
      "balance_loss_clip": 1.04487419,
      "balance_loss_mlp": 1.01918483,
      "epoch": 0.42909965429129715,
      "flos": 16216074414720.0,
      "grad_norm": 1.8879144840989626,
      "language_loss": 0.78816801,
      "learning_rate": 2.5491986915764175e-06,
      "loss": 0.80958354,
      "num_input_tokens_seen": 153113305,
      "step": 7137,
      "time_per_iteration": 2.485599994659424
    },
    {
      "auxiliary_loss_clip": 0.01128096,
      "auxiliary_loss_mlp": 0.01034606,
      "balance_loss_clip": 1.04730165,
      "balance_loss_mlp": 1.02031302,
      "epoch": 0.4291597775439651,
      "flos": 23112862122240.0,
      "grad_norm": 1.858696512575758,
      "language_loss": 0.76639199,
      "learning_rate": 2.548824190884499e-06,
      "loss": 0.788019,
      "num_input_tokens_seen": 153132735,
      "step": 7138,
      "time_per_iteration": 2.4957005977630615
    },
    {
      "auxiliary_loss_clip": 0.0103868,
      "auxiliary_loss_mlp": 0.01007225,
      "balance_loss_clip": 1.03116024,
      "balance_loss_mlp": 1.00537705,
      "epoch": 0.4292199007966331,
      "flos": 67546212681600.0,
      "grad_norm": 1.0765855541557439,
      "language_loss": 0.56234348,
      "learning_rate": 2.548449669381113e-06,
      "loss": 0.58280253,
      "num_input_tokens_seen": 153187925,
      "step": 7139,
      "time_per_iteration": 3.0154480934143066
    },
    {
      "auxiliary_loss_clip": 0.01120193,
      "auxiliary_loss_mlp": 0.0078996,
      "balance_loss_clip": 1.04496658,
      "balance_loss_mlp": 1.01529121,
      "epoch": 0.42928002404930105,
      "flos": 22999850956800.0,
      "grad_norm": 1.7105802154765224,
      "language_loss": 0.8082459,
      "learning_rate": 2.5480751270804595e-06,
      "loss": 0.82734746,
      "num_input_tokens_seen": 153206990,
      "step": 7140,
      "time_per_iteration": 2.4906117916107178
    },
    {
      "auxiliary_loss_clip": 0.01112576,
      "auxiliary_loss_mlp": 0.01032992,
      "balance_loss_clip": 1.0448724,
      "balance_loss_mlp": 1.01877081,
      "epoch": 0.429340147301969,
      "flos": 11544922241280.0,
      "grad_norm": 1.9571774577489969,
      "language_loss": 0.82073557,
      "learning_rate": 2.5477005639967424e-06,
      "loss": 0.84219128,
      "num_input_tokens_seen": 153222345,
      "step": 7141,
      "time_per_iteration": 3.8308210372924805
    },
    {
      "auxiliary_loss_clip": 0.01122228,
      "auxiliary_loss_mlp": 0.01038097,
      "balance_loss_clip": 1.04784489,
      "balance_loss_mlp": 1.02315974,
      "epoch": 0.42940027055463703,
      "flos": 25264988472960.0,
      "grad_norm": 1.6991495671669412,
      "language_loss": 0.86448872,
      "learning_rate": 2.547325980144166e-06,
      "loss": 0.88609195,
      "num_input_tokens_seen": 153240570,
      "step": 7142,
      "time_per_iteration": 2.520124912261963
    },
    {
      "auxiliary_loss_clip": 0.01100385,
      "auxiliary_loss_mlp": 0.01030828,
      "balance_loss_clip": 1.04874921,
      "balance_loss_mlp": 1.01753068,
      "epoch": 0.429460393807305,
      "flos": 23805004268160.0,
      "grad_norm": 1.8577434483713888,
      "language_loss": 0.78532672,
      "learning_rate": 2.5469513755369323e-06,
      "loss": 0.80663884,
      "num_input_tokens_seen": 153259575,
      "step": 7143,
      "time_per_iteration": 3.9784836769104004
    },
    {
      "auxiliary_loss_clip": 0.01074613,
      "auxiliary_loss_mlp": 0.01038842,
      "balance_loss_clip": 1.0416522,
      "balance_loss_mlp": 1.02425051,
      "epoch": 0.42952051705997296,
      "flos": 13918294414080.0,
      "grad_norm": 1.9266998326249292,
      "language_loss": 0.7714259,
      "learning_rate": 2.5465767501892484e-06,
      "loss": 0.79256046,
      "num_input_tokens_seen": 153276650,
      "step": 7144,
      "time_per_iteration": 2.5507609844207764
    },
    {
      "auxiliary_loss_clip": 0.01090722,
      "auxiliary_loss_mlp": 0.01029378,
      "balance_loss_clip": 1.04061556,
      "balance_loss_mlp": 1.01506734,
      "epoch": 0.4295806403126409,
      "flos": 26760380509440.0,
      "grad_norm": 1.7250274801367815,
      "language_loss": 0.73375738,
      "learning_rate": 2.54620210411532e-06,
      "loss": 0.75495833,
      "num_input_tokens_seen": 153298025,
      "step": 7145,
      "time_per_iteration": 2.593904495239258
    },
    {
      "auxiliary_loss_clip": 0.0111334,
      "auxiliary_loss_mlp": 0.01035864,
      "balance_loss_clip": 1.04478025,
      "balance_loss_mlp": 1.02149951,
      "epoch": 0.4296407635653089,
      "flos": 20952619297920.0,
      "grad_norm": 2.0973330237123204,
      "language_loss": 0.7935003,
      "learning_rate": 2.545827437329352e-06,
      "loss": 0.81499237,
      "num_input_tokens_seen": 153315775,
      "step": 7146,
      "time_per_iteration": 2.4782676696777344
    },
    {
      "auxiliary_loss_clip": 0.01108198,
      "auxiliary_loss_mlp": 0.010322,
      "balance_loss_clip": 1.0430212,
      "balance_loss_mlp": 1.01970065,
      "epoch": 0.42970088681797686,
      "flos": 15852335339520.0,
      "grad_norm": 3.7010135025747695,
      "language_loss": 0.82933509,
      "learning_rate": 2.5454527498455532e-06,
      "loss": 0.85073906,
      "num_input_tokens_seen": 153332765,
      "step": 7147,
      "time_per_iteration": 3.8305106163024902
    },
    {
      "auxiliary_loss_clip": 0.01111357,
      "auxiliary_loss_mlp": 0.01033832,
      "balance_loss_clip": 1.04784048,
      "balance_loss_mlp": 1.01848936,
      "epoch": 0.4297610100706448,
      "flos": 22382618624640.0,
      "grad_norm": 2.998023342546849,
      "language_loss": 0.87718254,
      "learning_rate": 2.545078041678131e-06,
      "loss": 0.89863443,
      "num_input_tokens_seen": 153350760,
      "step": 7148,
      "time_per_iteration": 3.873980760574341
    },
    {
      "auxiliary_loss_clip": 0.01104989,
      "auxiliary_loss_mlp": 0.0103437,
      "balance_loss_clip": 1.04500723,
      "balance_loss_mlp": 1.02057207,
      "epoch": 0.4298211333233128,
      "flos": 27925681536000.0,
      "grad_norm": 1.5739212943043868,
      "language_loss": 0.77799404,
      "learning_rate": 2.5447033128412957e-06,
      "loss": 0.79938769,
      "num_input_tokens_seen": 153370765,
      "step": 7149,
      "time_per_iteration": 2.567481756210327
    },
    {
      "auxiliary_loss_clip": 0.01080543,
      "auxiliary_loss_mlp": 0.01033198,
      "balance_loss_clip": 1.04116702,
      "balance_loss_mlp": 1.01875019,
      "epoch": 0.42988125657598075,
      "flos": 24425612478720.0,
      "grad_norm": 1.687368369001774,
      "language_loss": 0.79722023,
      "learning_rate": 2.544328563349256e-06,
      "loss": 0.81835765,
      "num_input_tokens_seen": 153390725,
      "step": 7150,
      "time_per_iteration": 2.577098846435547
    },
    {
      "auxiliary_loss_clip": 0.0111714,
      "auxiliary_loss_mlp": 0.0103892,
      "balance_loss_clip": 1.04702568,
      "balance_loss_mlp": 1.02308869,
      "epoch": 0.4299413798286487,
      "flos": 15850180523520.0,
      "grad_norm": 1.5599140326004195,
      "language_loss": 0.74846035,
      "learning_rate": 2.5439537932162222e-06,
      "loss": 0.77002096,
      "num_input_tokens_seen": 153408010,
      "step": 7151,
      "time_per_iteration": 2.4604878425598145
    },
    {
      "auxiliary_loss_clip": 0.01078867,
      "auxiliary_loss_mlp": 0.01034474,
      "balance_loss_clip": 1.04111302,
      "balance_loss_mlp": 1.01970434,
      "epoch": 0.4300015030813167,
      "flos": 22309504490880.0,
      "grad_norm": 1.9132899478263063,
      "language_loss": 0.70264828,
      "learning_rate": 2.543579002456406e-06,
      "loss": 0.7237817,
      "num_input_tokens_seen": 153426865,
      "step": 7152,
      "time_per_iteration": 2.5325629711151123
    },
    {
      "auxiliary_loss_clip": 0.01101138,
      "auxiliary_loss_mlp": 0.01032292,
      "balance_loss_clip": 1.03932214,
      "balance_loss_mlp": 1.01885068,
      "epoch": 0.43006162633398465,
      "flos": 34897666366080.0,
      "grad_norm": 1.836918484590717,
      "language_loss": 0.71012735,
      "learning_rate": 2.54320419108402e-06,
      "loss": 0.73146158,
      "num_input_tokens_seen": 153449410,
      "step": 7153,
      "time_per_iteration": 2.6170499324798584
    },
    {
      "auxiliary_loss_clip": 0.01106755,
      "auxiliary_loss_mlp": 0.01032485,
      "balance_loss_clip": 1.04092097,
      "balance_loss_mlp": 1.01784635,
      "epoch": 0.4301217495866526,
      "flos": 15961575576960.0,
      "grad_norm": 3.304614650439234,
      "language_loss": 0.78413135,
      "learning_rate": 2.542829359113276e-06,
      "loss": 0.80552375,
      "num_input_tokens_seen": 153467910,
      "step": 7154,
      "time_per_iteration": 2.4738783836364746
    },
    {
      "auxiliary_loss_clip": 0.01087922,
      "auxiliary_loss_mlp": 0.01034613,
      "balance_loss_clip": 1.03980148,
      "balance_loss_mlp": 1.02077317,
      "epoch": 0.43018187283932063,
      "flos": 18770364414720.0,
      "grad_norm": 1.5850342352773876,
      "language_loss": 0.7869277,
      "learning_rate": 2.542454506558389e-06,
      "loss": 0.80815303,
      "num_input_tokens_seen": 153487100,
      "step": 7155,
      "time_per_iteration": 2.511211395263672
    },
    {
      "auxiliary_loss_clip": 0.01094572,
      "auxiliary_loss_mlp": 0.01027039,
      "balance_loss_clip": 1.04318476,
      "balance_loss_mlp": 1.01397943,
      "epoch": 0.4302419960919886,
      "flos": 20151703791360.0,
      "grad_norm": 1.8682137433677013,
      "language_loss": 0.88685691,
      "learning_rate": 2.5420796334335723e-06,
      "loss": 0.90807307,
      "num_input_tokens_seen": 153505565,
      "step": 7156,
      "time_per_iteration": 2.547166347503662
    },
    {
      "auxiliary_loss_clip": 0.01124095,
      "auxiliary_loss_mlp": 0.01032233,
      "balance_loss_clip": 1.04390931,
      "balance_loss_mlp": 1.01780868,
      "epoch": 0.43030211934465656,
      "flos": 26432731624320.0,
      "grad_norm": 2.1345554806324767,
      "language_loss": 0.83028519,
      "learning_rate": 2.541704739753042e-06,
      "loss": 0.85184836,
      "num_input_tokens_seen": 153526130,
      "step": 7157,
      "time_per_iteration": 2.491027593612671
    },
    {
      "auxiliary_loss_clip": 0.01127411,
      "auxiliary_loss_mlp": 0.01031808,
      "balance_loss_clip": 1.04642749,
      "balance_loss_mlp": 1.01760471,
      "epoch": 0.43036224259732453,
      "flos": 24389234979840.0,
      "grad_norm": 1.87283904148046,
      "language_loss": 0.71698451,
      "learning_rate": 2.5413298255310132e-06,
      "loss": 0.73857665,
      "num_input_tokens_seen": 153546370,
      "step": 7158,
      "time_per_iteration": 2.4837417602539062
    },
    {
      "auxiliary_loss_clip": 0.01108912,
      "auxiliary_loss_mlp": 0.01031654,
      "balance_loss_clip": 1.0419035,
      "balance_loss_mlp": 1.0183804,
      "epoch": 0.4304223658499925,
      "flos": 17201714590080.0,
      "grad_norm": 1.8570494064800376,
      "language_loss": 0.82598031,
      "learning_rate": 2.5409548907817034e-06,
      "loss": 0.847386,
      "num_input_tokens_seen": 153562800,
      "step": 7159,
      "time_per_iteration": 2.4435482025146484
    },
    {
      "auxiliary_loss_clip": 0.01094612,
      "auxiliary_loss_mlp": 0.01034322,
      "balance_loss_clip": 1.04147649,
      "balance_loss_mlp": 1.02003527,
      "epoch": 0.43048248910266046,
      "flos": 14903000835840.0,
      "grad_norm": 2.250356178031571,
      "language_loss": 0.82969081,
      "learning_rate": 2.54057993551933e-06,
      "loss": 0.85098016,
      "num_input_tokens_seen": 153578395,
      "step": 7160,
      "time_per_iteration": 2.5094876289367676
    },
    {
      "auxiliary_loss_clip": 0.01117857,
      "auxiliary_loss_mlp": 0.01034785,
      "balance_loss_clip": 1.0469799,
      "balance_loss_mlp": 1.01918089,
      "epoch": 0.4305426123553284,
      "flos": 21579835610880.0,
      "grad_norm": 2.120379927808996,
      "language_loss": 0.7697947,
      "learning_rate": 2.5402049597581116e-06,
      "loss": 0.79132116,
      "num_input_tokens_seen": 153596880,
      "step": 7161,
      "time_per_iteration": 2.4977798461914062
    },
    {
      "auxiliary_loss_clip": 0.01109185,
      "auxiliary_loss_mlp": 0.01036649,
      "balance_loss_clip": 1.04250979,
      "balance_loss_mlp": 1.0226717,
      "epoch": 0.4306027356079964,
      "flos": 22601278667520.0,
      "grad_norm": 1.9482475378157686,
      "language_loss": 0.7312752,
      "learning_rate": 2.5398299635122662e-06,
      "loss": 0.75273353,
      "num_input_tokens_seen": 153616570,
      "step": 7162,
      "time_per_iteration": 2.5267441272735596
    },
    {
      "auxiliary_loss_clip": 0.01014414,
      "auxiliary_loss_mlp": 0.00795148,
      "balance_loss_clip": 1.02664149,
      "balance_loss_mlp": 1.05112648,
      "epoch": 0.43066285886066435,
      "flos": 70672091806080.0,
      "grad_norm": 0.8003441778444421,
      "language_loss": 0.59015006,
      "learning_rate": 2.5394549467960147e-06,
      "loss": 0.60824573,
      "num_input_tokens_seen": 153671450,
      "step": 7163,
      "time_per_iteration": 3.0540506839752197
    },
    {
      "auxiliary_loss_clip": 0.01093211,
      "auxiliary_loss_mlp": 0.01034379,
      "balance_loss_clip": 1.04007876,
      "balance_loss_mlp": 1.02022862,
      "epoch": 0.4307229821133323,
      "flos": 26720591218560.0,
      "grad_norm": 1.6421500291672633,
      "language_loss": 0.7923162,
      "learning_rate": 2.5390799096235783e-06,
      "loss": 0.81359208,
      "num_input_tokens_seen": 153691405,
      "step": 7164,
      "time_per_iteration": 2.5857553482055664
    },
    {
      "auxiliary_loss_clip": 0.01124144,
      "auxiliary_loss_mlp": 0.01038097,
      "balance_loss_clip": 1.04244149,
      "balance_loss_mlp": 1.02400661,
      "epoch": 0.4307831053660003,
      "flos": 26177119464960.0,
      "grad_norm": 2.0515474659848483,
      "language_loss": 0.67857838,
      "learning_rate": 2.538704852009177e-06,
      "loss": 0.7002008,
      "num_input_tokens_seen": 153711555,
      "step": 7165,
      "time_per_iteration": 2.5039310455322266
    },
    {
      "auxiliary_loss_clip": 0.01096591,
      "auxiliary_loss_mlp": 0.00798202,
      "balance_loss_clip": 1.04564536,
      "balance_loss_mlp": 1.02970409,
      "epoch": 0.43084322861866825,
      "flos": 18910343715840.0,
      "grad_norm": 2.0743995676077884,
      "language_loss": 0.75120312,
      "learning_rate": 2.538329773967034e-06,
      "loss": 0.77015102,
      "num_input_tokens_seen": 153730095,
      "step": 7166,
      "time_per_iteration": 2.5098583698272705
    },
    {
      "auxiliary_loss_clip": 0.01110317,
      "auxiliary_loss_mlp": 0.0103555,
      "balance_loss_clip": 1.04494476,
      "balance_loss_mlp": 1.02305102,
      "epoch": 0.4309033518713362,
      "flos": 26432911192320.0,
      "grad_norm": 1.6546633075900048,
      "language_loss": 0.71741199,
      "learning_rate": 2.537954675511372e-06,
      "loss": 0.73887068,
      "num_input_tokens_seen": 153749320,
      "step": 7167,
      "time_per_iteration": 2.5271265506744385
    },
    {
      "auxiliary_loss_clip": 0.01092451,
      "auxiliary_loss_mlp": 0.00805582,
      "balance_loss_clip": 1.04476988,
      "balance_loss_mlp": 1.04600835,
      "epoch": 0.43096347512400424,
      "flos": 21213295274880.0,
      "grad_norm": 1.4862017344561447,
      "language_loss": 0.78267533,
      "learning_rate": 2.537579556656414e-06,
      "loss": 0.80165565,
      "num_input_tokens_seen": 153767825,
      "step": 7168,
      "time_per_iteration": 2.527155637741089
    },
    {
      "auxiliary_loss_clip": 0.01101806,
      "auxiliary_loss_mlp": 0.01038104,
      "balance_loss_clip": 1.04676914,
      "balance_loss_mlp": 1.02418673,
      "epoch": 0.4310235983766722,
      "flos": 16540131939840.0,
      "grad_norm": 2.595748750562116,
      "language_loss": 0.82557547,
      "learning_rate": 2.537204417416387e-06,
      "loss": 0.84697461,
      "num_input_tokens_seen": 153785350,
      "step": 7169,
      "time_per_iteration": 2.5173392295837402
    },
    {
      "auxiliary_loss_clip": 0.01033771,
      "auxiliary_loss_mlp": 0.01003736,
      "balance_loss_clip": 1.02829099,
      "balance_loss_mlp": 1.001912,
      "epoch": 0.43108372162934017,
      "flos": 64775704763520.0,
      "grad_norm": 0.6715431690984631,
      "language_loss": 0.6075722,
      "learning_rate": 2.5368292578055132e-06,
      "loss": 0.62794733,
      "num_input_tokens_seen": 153856400,
      "step": 7170,
      "time_per_iteration": 3.2644782066345215
    },
    {
      "auxiliary_loss_clip": 0.01122283,
      "auxiliary_loss_mlp": 0.01031623,
      "balance_loss_clip": 1.04439783,
      "balance_loss_mlp": 1.01880193,
      "epoch": 0.43114384488200813,
      "flos": 13444094039040.0,
      "grad_norm": 1.7640497630991898,
      "language_loss": 0.75683016,
      "learning_rate": 2.536454077838021e-06,
      "loss": 0.77836919,
      "num_input_tokens_seen": 153875230,
      "step": 7171,
      "time_per_iteration": 2.470296859741211
    },
    {
      "auxiliary_loss_clip": 0.01111597,
      "auxiliary_loss_mlp": 0.01034876,
      "balance_loss_clip": 1.04552102,
      "balance_loss_mlp": 1.02180505,
      "epoch": 0.4312039681346761,
      "flos": 26286682924800.0,
      "grad_norm": 1.6202151455132376,
      "language_loss": 0.77508837,
      "learning_rate": 2.5360788775281357e-06,
      "loss": 0.79655302,
      "num_input_tokens_seen": 153894740,
      "step": 7172,
      "time_per_iteration": 2.5180022716522217
    },
    {
      "auxiliary_loss_clip": 0.01097751,
      "auxiliary_loss_mlp": 0.01042857,
      "balance_loss_clip": 1.04301763,
      "balance_loss_mlp": 1.02635837,
      "epoch": 0.43126409138734406,
      "flos": 20376684627840.0,
      "grad_norm": 1.5766275580741786,
      "language_loss": 0.76361495,
      "learning_rate": 2.535703656890086e-06,
      "loss": 0.78502107,
      "num_input_tokens_seen": 153913230,
      "step": 7173,
      "time_per_iteration": 2.517632484436035
    },
    {
      "auxiliary_loss_clip": 0.0112241,
      "auxiliary_loss_mlp": 0.00806646,
      "balance_loss_clip": 1.04486108,
      "balance_loss_mlp": 1.04437518,
      "epoch": 0.431324214640012,
      "flos": 22123091882880.0,
      "grad_norm": 1.4424011181760175,
      "language_loss": 0.77093816,
      "learning_rate": 2.5353284159381e-06,
      "loss": 0.79022872,
      "num_input_tokens_seen": 153933250,
      "step": 7174,
      "time_per_iteration": 2.4716413021087646
    },
    {
      "auxiliary_loss_clip": 0.01124439,
      "auxiliary_loss_mlp": 0.01033607,
      "balance_loss_clip": 1.04388833,
      "balance_loss_mlp": 1.01875937,
      "epoch": 0.43138433789268,
      "flos": 15231008856960.0,
      "grad_norm": 1.5706056318743462,
      "language_loss": 0.8214879,
      "learning_rate": 2.534953154686407e-06,
      "loss": 0.84306836,
      "num_input_tokens_seen": 153951325,
      "step": 7175,
      "time_per_iteration": 2.469491958618164
    },
    {
      "auxiliary_loss_clip": 0.01081358,
      "auxiliary_loss_mlp": 0.01044153,
      "balance_loss_clip": 1.04169416,
      "balance_loss_mlp": 1.02723098,
      "epoch": 0.43144446114534796,
      "flos": 18150294908160.0,
      "grad_norm": 2.067525967973552,
      "language_loss": 0.7475543,
      "learning_rate": 2.5345778731492366e-06,
      "loss": 0.76880938,
      "num_input_tokens_seen": 153966975,
      "step": 7176,
      "time_per_iteration": 2.5579540729522705
    },
    {
      "auxiliary_loss_clip": 0.0111476,
      "auxiliary_loss_mlp": 0.01033392,
      "balance_loss_clip": 1.04381537,
      "balance_loss_mlp": 1.01969481,
      "epoch": 0.4315045843980159,
      "flos": 22929861306240.0,
      "grad_norm": 1.5425117933515733,
      "language_loss": 0.73601735,
      "learning_rate": 2.534202571340819e-06,
      "loss": 0.75749886,
      "num_input_tokens_seen": 153986695,
      "step": 7177,
      "time_per_iteration": 2.525585651397705
    },
    {
      "auxiliary_loss_clip": 0.01110452,
      "auxiliary_loss_mlp": 0.01036827,
      "balance_loss_clip": 1.04464245,
      "balance_loss_mlp": 1.02060223,
      "epoch": 0.4315647076506839,
      "flos": 22126862810880.0,
      "grad_norm": 1.7804819539848253,
      "language_loss": 0.81335664,
      "learning_rate": 2.533827249275387e-06,
      "loss": 0.83482939,
      "num_input_tokens_seen": 154004710,
      "step": 7178,
      "time_per_iteration": 2.5620601177215576
    },
    {
      "auxiliary_loss_clip": 0.01098424,
      "auxiliary_loss_mlp": 0.0103069,
      "balance_loss_clip": 1.04353714,
      "balance_loss_mlp": 1.01758277,
      "epoch": 0.43162483090335185,
      "flos": 26871129118080.0,
      "grad_norm": 1.4989036552721977,
      "language_loss": 0.84284705,
      "learning_rate": 2.5334519069671725e-06,
      "loss": 0.86413819,
      "num_input_tokens_seen": 154024320,
      "step": 7179,
      "time_per_iteration": 2.591660976409912
    },
    {
      "auxiliary_loss_clip": 0.01099284,
      "auxiliary_loss_mlp": 0.01029428,
      "balance_loss_clip": 1.0425458,
      "balance_loss_mlp": 1.01597476,
      "epoch": 0.4316849541560198,
      "flos": 13913122855680.0,
      "grad_norm": 1.7395509246632255,
      "language_loss": 0.75356442,
      "learning_rate": 2.5330765444304075e-06,
      "loss": 0.77485156,
      "num_input_tokens_seen": 154041755,
      "step": 7180,
      "time_per_iteration": 3.897235870361328
    },
    {
      "auxiliary_loss_clip": 0.01097874,
      "auxiliary_loss_mlp": 0.00793993,
      "balance_loss_clip": 1.04047084,
      "balance_loss_mlp": 1.01483476,
      "epoch": 0.4317450774086878,
      "flos": 16435165420800.0,
      "grad_norm": 1.8740179454287824,
      "language_loss": 0.81834078,
      "learning_rate": 2.5327011616793274e-06,
      "loss": 0.83725941,
      "num_input_tokens_seen": 154056775,
      "step": 7181,
      "time_per_iteration": 2.4960196018218994
    },
    {
      "auxiliary_loss_clip": 0.01111081,
      "auxiliary_loss_mlp": 0.01036639,
      "balance_loss_clip": 1.04785466,
      "balance_loss_mlp": 1.0210166,
      "epoch": 0.4318052006613558,
      "flos": 20554980762240.0,
      "grad_norm": 1.7964108726793857,
      "language_loss": 0.88705325,
      "learning_rate": 2.532325758728165e-06,
      "loss": 0.90853047,
      "num_input_tokens_seen": 154075015,
      "step": 7182,
      "time_per_iteration": 3.9089596271514893
    },
    {
      "auxiliary_loss_clip": 0.01111837,
      "auxiliary_loss_mlp": 0.00791525,
      "balance_loss_clip": 1.04494095,
      "balance_loss_mlp": 1.01766801,
      "epoch": 0.43186532391402377,
      "flos": 22820046451200.0,
      "grad_norm": 1.819828211794466,
      "language_loss": 0.75964844,
      "learning_rate": 2.5319503355911566e-06,
      "loss": 0.77868205,
      "num_input_tokens_seen": 154095170,
      "step": 7183,
      "time_per_iteration": 2.503681182861328
    },
    {
      "auxiliary_loss_clip": 0.0111132,
      "auxiliary_loss_mlp": 0.01032347,
      "balance_loss_clip": 1.04361022,
      "balance_loss_mlp": 1.01864958,
      "epoch": 0.43192544716669173,
      "flos": 25556583081600.0,
      "grad_norm": 1.5505146324222656,
      "language_loss": 0.7751857,
      "learning_rate": 2.5315748922825393e-06,
      "loss": 0.79662228,
      "num_input_tokens_seen": 154116895,
      "step": 7184,
      "time_per_iteration": 2.5433454513549805
    },
    {
      "auxiliary_loss_clip": 0.01096725,
      "auxiliary_loss_mlp": 0.01032912,
      "balance_loss_clip": 1.04481542,
      "balance_loss_mlp": 1.01968527,
      "epoch": 0.4319855704193597,
      "flos": 30954674701440.0,
      "grad_norm": 2.090514327419112,
      "language_loss": 0.73300678,
      "learning_rate": 2.5311994288165474e-06,
      "loss": 0.75430316,
      "num_input_tokens_seen": 154138395,
      "step": 7185,
      "time_per_iteration": 2.5992045402526855
    },
    {
      "auxiliary_loss_clip": 0.01108157,
      "auxiliary_loss_mlp": 0.01035203,
      "balance_loss_clip": 1.04424,
      "balance_loss_mlp": 1.02132654,
      "epoch": 0.43204569367202766,
      "flos": 24238732993920.0,
      "grad_norm": 2.5693468084136244,
      "language_loss": 0.75209564,
      "learning_rate": 2.530823945207421e-06,
      "loss": 0.77352923,
      "num_input_tokens_seen": 154156775,
      "step": 7186,
      "time_per_iteration": 3.9048914909362793
    },
    {
      "auxiliary_loss_clip": 0.01090026,
      "auxiliary_loss_mlp": 0.01034374,
      "balance_loss_clip": 1.04436207,
      "balance_loss_mlp": 1.02065897,
      "epoch": 0.43210581692469563,
      "flos": 18406948561920.0,
      "grad_norm": 2.7389551437482753,
      "language_loss": 0.76167178,
      "learning_rate": 2.5304484414693962e-06,
      "loss": 0.78291583,
      "num_input_tokens_seen": 154177500,
      "step": 7187,
      "time_per_iteration": 4.000559091567993
    },
    {
      "auxiliary_loss_clip": 0.01019829,
      "auxiliary_loss_mlp": 0.01000351,
      "balance_loss_clip": 1.02937591,
      "balance_loss_mlp": 0.99861068,
      "epoch": 0.4321659401773636,
      "flos": 49832378910720.0,
      "grad_norm": 0.8579279674889612,
      "language_loss": 0.68133634,
      "learning_rate": 2.530072917616714e-06,
      "loss": 0.70153815,
      "num_input_tokens_seen": 154237110,
      "step": 7188,
      "time_per_iteration": 3.1861555576324463
    },
    {
      "auxiliary_loss_clip": 0.01092769,
      "auxiliary_loss_mlp": 0.01033555,
      "balance_loss_clip": 1.04440522,
      "balance_loss_mlp": 1.02030468,
      "epoch": 0.43222606343003156,
      "flos": 17128564542720.0,
      "grad_norm": 1.9335668276854072,
      "language_loss": 0.77981639,
      "learning_rate": 2.529697373663614e-06,
      "loss": 0.80107963,
      "num_input_tokens_seen": 154253910,
      "step": 7189,
      "time_per_iteration": 2.502847671508789
    },
    {
      "auxiliary_loss_clip": 0.01079544,
      "auxiliary_loss_mlp": 0.01040915,
      "balance_loss_clip": 1.04201579,
      "balance_loss_mlp": 1.02616882,
      "epoch": 0.4322861866826995,
      "flos": 22749949059840.0,
      "grad_norm": 1.9130008486644403,
      "language_loss": 0.71499145,
      "learning_rate": 2.5293218096243364e-06,
      "loss": 0.73619598,
      "num_input_tokens_seen": 154274770,
      "step": 7190,
      "time_per_iteration": 2.6052281856536865
    },
    {
      "auxiliary_loss_clip": 0.01097351,
      "auxiliary_loss_mlp": 0.01031831,
      "balance_loss_clip": 1.04134679,
      "balance_loss_mlp": 1.01852703,
      "epoch": 0.4323463099353675,
      "flos": 27891925729920.0,
      "grad_norm": 1.4172437169133458,
      "language_loss": 0.79497623,
      "learning_rate": 2.5289462255131223e-06,
      "loss": 0.81626809,
      "num_input_tokens_seen": 154295035,
      "step": 7191,
      "time_per_iteration": 2.586731433868408
    },
    {
      "auxiliary_loss_clip": 0.01076263,
      "auxiliary_loss_mlp": 0.01032211,
      "balance_loss_clip": 1.043823,
      "balance_loss_mlp": 1.01934791,
      "epoch": 0.43240643318803546,
      "flos": 21614740652160.0,
      "grad_norm": 1.5758543933735882,
      "language_loss": 0.74813598,
      "learning_rate": 2.5285706213442146e-06,
      "loss": 0.76922071,
      "num_input_tokens_seen": 154314905,
      "step": 7192,
      "time_per_iteration": 2.6035068035125732
    },
    {
      "auxiliary_loss_clip": 0.01079055,
      "auxiliary_loss_mlp": 0.01037027,
      "balance_loss_clip": 1.04351163,
      "balance_loss_mlp": 1.02223349,
      "epoch": 0.4324665564407034,
      "flos": 17558378686080.0,
      "grad_norm": 1.7628846021016014,
      "language_loss": 0.78801572,
      "learning_rate": 2.5281949971318557e-06,
      "loss": 0.80917656,
      "num_input_tokens_seen": 154331740,
      "step": 7193,
      "time_per_iteration": 2.574690580368042
    },
    {
      "auxiliary_loss_clip": 0.01102748,
      "auxiliary_loss_mlp": 0.01039485,
      "balance_loss_clip": 1.04245758,
      "balance_loss_mlp": 1.02568698,
      "epoch": 0.4325266796933714,
      "flos": 18402423448320.0,
      "grad_norm": 1.8062278229436828,
      "language_loss": 0.75939,
      "learning_rate": 2.5278193528902897e-06,
      "loss": 0.78081238,
      "num_input_tokens_seen": 154348740,
      "step": 7194,
      "time_per_iteration": 2.5526301860809326
    },
    {
      "auxiliary_loss_clip": 0.0112613,
      "auxiliary_loss_mlp": 0.01035979,
      "balance_loss_clip": 1.04784608,
      "balance_loss_mlp": 1.02234197,
      "epoch": 0.4325868029460394,
      "flos": 22564793427840.0,
      "grad_norm": 2.0603285712147366,
      "language_loss": 0.59653878,
      "learning_rate": 2.5274436886337613e-06,
      "loss": 0.61815989,
      "num_input_tokens_seen": 154368835,
      "step": 7195,
      "time_per_iteration": 2.497948408126831
    },
    {
      "auxiliary_loss_clip": 0.01102037,
      "auxiliary_loss_mlp": 0.01038858,
      "balance_loss_clip": 1.04349875,
      "balance_loss_mlp": 1.02360535,
      "epoch": 0.43264692619870737,
      "flos": 14605516396800.0,
      "grad_norm": 1.8536712300888893,
      "language_loss": 0.64990151,
      "learning_rate": 2.527068004376515e-06,
      "loss": 0.67131042,
      "num_input_tokens_seen": 154384620,
      "step": 7196,
      "time_per_iteration": 2.5159857273101807
    },
    {
      "auxiliary_loss_clip": 0.01127053,
      "auxiliary_loss_mlp": 0.0103751,
      "balance_loss_clip": 1.04549551,
      "balance_loss_mlp": 1.02337718,
      "epoch": 0.43270704945137534,
      "flos": 21501657659520.0,
      "grad_norm": 2.434843124549809,
      "language_loss": 0.72349226,
      "learning_rate": 2.526692300132797e-06,
      "loss": 0.74513781,
      "num_input_tokens_seen": 154402865,
      "step": 7197,
      "time_per_iteration": 2.466332197189331
    },
    {
      "auxiliary_loss_clip": 0.0111073,
      "auxiliary_loss_mlp": 0.01039162,
      "balance_loss_clip": 1.04610145,
      "balance_loss_mlp": 1.02523851,
      "epoch": 0.4327671727040433,
      "flos": 25155891889920.0,
      "grad_norm": 1.4863479526756327,
      "language_loss": 0.72633946,
      "learning_rate": 2.5263165759168547e-06,
      "loss": 0.74783838,
      "num_input_tokens_seen": 154423625,
      "step": 7198,
      "time_per_iteration": 2.5538575649261475
    },
    {
      "auxiliary_loss_clip": 0.01085396,
      "auxiliary_loss_mlp": 0.01030339,
      "balance_loss_clip": 1.04314554,
      "balance_loss_mlp": 1.01715422,
      "epoch": 0.43282729595671127,
      "flos": 25447163276160.0,
      "grad_norm": 1.3634782753867378,
      "language_loss": 0.80892956,
      "learning_rate": 2.525940831742934e-06,
      "loss": 0.83008695,
      "num_input_tokens_seen": 154444775,
      "step": 7199,
      "time_per_iteration": 2.6172633171081543
    },
    {
      "auxiliary_loss_clip": 0.01104637,
      "auxiliary_loss_mlp": 0.01033394,
      "balance_loss_clip": 1.04460204,
      "balance_loss_mlp": 1.02031112,
      "epoch": 0.43288741920937923,
      "flos": 24126116878080.0,
      "grad_norm": 2.1799005691921174,
      "language_loss": 0.69033992,
      "learning_rate": 2.525565067625286e-06,
      "loss": 0.71172023,
      "num_input_tokens_seen": 154460815,
      "step": 7200,
      "time_per_iteration": 2.5478813648223877
    },
    {
      "auxiliary_loss_clip": 0.01101394,
      "auxiliary_loss_mlp": 0.00787796,
      "balance_loss_clip": 1.04485786,
      "balance_loss_mlp": 1.01040733,
      "epoch": 0.4329475424620472,
      "flos": 19204955066880.0,
      "grad_norm": 1.820368454828695,
      "language_loss": 0.86980492,
      "learning_rate": 2.525189283578157e-06,
      "loss": 0.88869685,
      "num_input_tokens_seen": 154479145,
      "step": 7201,
      "time_per_iteration": 2.6181232929229736
    },
    {
      "auxiliary_loss_clip": 0.01076141,
      "auxiliary_loss_mlp": 0.01036538,
      "balance_loss_clip": 1.04813993,
      "balance_loss_mlp": 1.02094519,
      "epoch": 0.43300766571471516,
      "flos": 22638374438400.0,
      "grad_norm": 1.808363318183575,
      "language_loss": 0.64586151,
      "learning_rate": 2.5248134796157974e-06,
      "loss": 0.66698831,
      "num_input_tokens_seen": 154498905,
      "step": 7202,
      "time_per_iteration": 2.6463847160339355
    },
    {
      "auxiliary_loss_clip": 0.01072861,
      "auxiliary_loss_mlp": 0.01027471,
      "balance_loss_clip": 1.04454947,
      "balance_loss_mlp": 1.0147754,
      "epoch": 0.4330677889673831,
      "flos": 22121080721280.0,
      "grad_norm": 1.7417500350078539,
      "language_loss": 0.81969815,
      "learning_rate": 2.5244376557524586e-06,
      "loss": 0.84070152,
      "num_input_tokens_seen": 154517270,
      "step": 7203,
      "time_per_iteration": 2.6132962703704834
    },
    {
      "auxiliary_loss_clip": 0.01094669,
      "auxiliary_loss_mlp": 0.01043856,
      "balance_loss_clip": 1.04503334,
      "balance_loss_mlp": 1.0296762,
      "epoch": 0.4331279122200511,
      "flos": 23221527742080.0,
      "grad_norm": 1.7763302869295055,
      "language_loss": 0.81001306,
      "learning_rate": 2.5240618120023912e-06,
      "loss": 0.83139825,
      "num_input_tokens_seen": 154535945,
      "step": 7204,
      "time_per_iteration": 2.588550567626953
    },
    {
      "auxiliary_loss_clip": 0.01100733,
      "auxiliary_loss_mlp": 0.01037333,
      "balance_loss_clip": 1.04404449,
      "balance_loss_mlp": 1.02445257,
      "epoch": 0.43318803547271906,
      "flos": 18259750627200.0,
      "grad_norm": 1.9267616720113059,
      "language_loss": 0.73687071,
      "learning_rate": 2.5236859483798468e-06,
      "loss": 0.75825137,
      "num_input_tokens_seen": 154554935,
      "step": 7205,
      "time_per_iteration": 2.506317138671875
    },
    {
      "auxiliary_loss_clip": 0.01122926,
      "auxiliary_loss_mlp": 0.00789484,
      "balance_loss_clip": 1.04672194,
      "balance_loss_mlp": 1.01415563,
      "epoch": 0.433248158725387,
      "flos": 27418407713280.0,
      "grad_norm": 1.6254481429361376,
      "language_loss": 0.75428653,
      "learning_rate": 2.5233100648990803e-06,
      "loss": 0.77341056,
      "num_input_tokens_seen": 154576065,
      "step": 7206,
      "time_per_iteration": 2.5586225986480713
    },
    {
      "auxiliary_loss_clip": 0.01076861,
      "auxiliary_loss_mlp": 0.01033394,
      "balance_loss_clip": 1.0437839,
      "balance_loss_mlp": 1.01990592,
      "epoch": 0.433308281978055,
      "flos": 23218008209280.0,
      "grad_norm": 2.674889081898101,
      "language_loss": 0.79286206,
      "learning_rate": 2.522934161574342e-06,
      "loss": 0.81396461,
      "num_input_tokens_seen": 154595110,
      "step": 7207,
      "time_per_iteration": 2.60380482673645
    },
    {
      "auxiliary_loss_clip": 0.01093676,
      "auxiliary_loss_mlp": 0.01032425,
      "balance_loss_clip": 1.04373813,
      "balance_loss_mlp": 1.01760149,
      "epoch": 0.433368405230723,
      "flos": 15852407166720.0,
      "grad_norm": 1.8908976746113093,
      "language_loss": 0.80900091,
      "learning_rate": 2.5225582384198888e-06,
      "loss": 0.83026195,
      "num_input_tokens_seen": 154612255,
      "step": 7208,
      "time_per_iteration": 2.5419259071350098
    },
    {
      "auxiliary_loss_clip": 0.01104064,
      "auxiliary_loss_mlp": 0.01032313,
      "balance_loss_clip": 1.04598832,
      "balance_loss_mlp": 1.01900959,
      "epoch": 0.433428528483391,
      "flos": 19026084314880.0,
      "grad_norm": 2.0452482392867113,
      "language_loss": 0.69658399,
      "learning_rate": 2.5221822954499744e-06,
      "loss": 0.71794772,
      "num_input_tokens_seen": 154630440,
      "step": 7209,
      "time_per_iteration": 2.519158124923706
    },
    {
      "auxiliary_loss_clip": 0.01110224,
      "auxiliary_loss_mlp": 0.01034751,
      "balance_loss_clip": 1.0434972,
      "balance_loss_mlp": 1.02066052,
      "epoch": 0.43348865173605894,
      "flos": 24718248581760.0,
      "grad_norm": 1.3886188799347143,
      "language_loss": 0.81422794,
      "learning_rate": 2.5218063326788557e-06,
      "loss": 0.83567768,
      "num_input_tokens_seen": 154652515,
      "step": 7210,
      "time_per_iteration": 2.5654988288879395
    },
    {
      "auxiliary_loss_clip": 0.01097538,
      "auxiliary_loss_mlp": 0.01032789,
      "balance_loss_clip": 1.04303992,
      "balance_loss_mlp": 1.02002788,
      "epoch": 0.4335487749887269,
      "flos": 22090664880000.0,
      "grad_norm": 1.7261502580393158,
      "language_loss": 0.81857741,
      "learning_rate": 2.5214303501207885e-06,
      "loss": 0.8398807,
      "num_input_tokens_seen": 154670965,
      "step": 7211,
      "time_per_iteration": 2.519228935241699
    },
    {
      "auxiliary_loss_clip": 0.01110212,
      "auxiliary_loss_mlp": 0.01035225,
      "balance_loss_clip": 1.0418365,
      "balance_loss_mlp": 1.02340531,
      "epoch": 0.43360889824139487,
      "flos": 22382941847040.0,
      "grad_norm": 1.9472703230425426,
      "language_loss": 0.74891973,
      "learning_rate": 2.521054347790029e-06,
      "loss": 0.77037406,
      "num_input_tokens_seen": 154689980,
      "step": 7212,
      "time_per_iteration": 2.508733034133911
    },
    {
      "auxiliary_loss_clip": 0.01103768,
      "auxiliary_loss_mlp": 0.01032382,
      "balance_loss_clip": 1.04431581,
      "balance_loss_mlp": 1.0197463,
      "epoch": 0.43366902149406283,
      "flos": 17528286067200.0,
      "grad_norm": 2.0520229497917013,
      "language_loss": 0.763677,
      "learning_rate": 2.5206783257008375e-06,
      "loss": 0.78503847,
      "num_input_tokens_seen": 154706570,
      "step": 7213,
      "time_per_iteration": 2.519376516342163
    },
    {
      "auxiliary_loss_clip": 0.01112631,
      "auxiliary_loss_mlp": 0.01037261,
      "balance_loss_clip": 1.04516959,
      "balance_loss_mlp": 1.02482164,
      "epoch": 0.4337291447467308,
      "flos": 19022672522880.0,
      "grad_norm": 1.4635139485900466,
      "language_loss": 0.65155429,
      "learning_rate": 2.520302283867471e-06,
      "loss": 0.67305326,
      "num_input_tokens_seen": 154725210,
      "step": 7214,
      "time_per_iteration": 2.507765531539917
    },
    {
      "auxiliary_loss_clip": 0.01095595,
      "auxiliary_loss_mlp": 0.01032202,
      "balance_loss_clip": 1.04115176,
      "balance_loss_mlp": 1.01994717,
      "epoch": 0.43378926799939876,
      "flos": 27234042180480.0,
      "grad_norm": 1.5856144381217636,
      "language_loss": 0.71537769,
      "learning_rate": 2.519926222304191e-06,
      "loss": 0.73665565,
      "num_input_tokens_seen": 154745945,
      "step": 7215,
      "time_per_iteration": 2.57932448387146
    },
    {
      "auxiliary_loss_clip": 0.01094165,
      "auxiliary_loss_mlp": 0.01036555,
      "balance_loss_clip": 1.04285359,
      "balance_loss_mlp": 1.02259004,
      "epoch": 0.43384939125206673,
      "flos": 15961108700160.0,
      "grad_norm": 1.7084298565828318,
      "language_loss": 0.7501989,
      "learning_rate": 2.519550141025255e-06,
      "loss": 0.77150607,
      "num_input_tokens_seen": 154763580,
      "step": 7216,
      "time_per_iteration": 2.5101757049560547
    },
    {
      "auxiliary_loss_clip": 0.01103174,
      "auxiliary_loss_mlp": 0.0104131,
      "balance_loss_clip": 1.04415715,
      "balance_loss_mlp": 1.02611709,
      "epoch": 0.4339095145047347,
      "flos": 21793216354560.0,
      "grad_norm": 2.384456371980344,
      "language_loss": 0.76107067,
      "learning_rate": 2.519174040044927e-06,
      "loss": 0.78251553,
      "num_input_tokens_seen": 154776825,
      "step": 7217,
      "time_per_iteration": 2.507002830505371
    },
    {
      "auxiliary_loss_clip": 0.0108587,
      "auxiliary_loss_mlp": 0.01038506,
      "balance_loss_clip": 1.04082775,
      "balance_loss_mlp": 1.02486801,
      "epoch": 0.43396963775740266,
      "flos": 14209853109120.0,
      "grad_norm": 1.9949445084847832,
      "language_loss": 0.7404508,
      "learning_rate": 2.5187979193774664e-06,
      "loss": 0.76169449,
      "num_input_tokens_seen": 154794025,
      "step": 7218,
      "time_per_iteration": 3.8940656185150146
    },
    {
      "auxiliary_loss_clip": 0.0110471,
      "auxiliary_loss_mlp": 0.01029999,
      "balance_loss_clip": 1.04744363,
      "balance_loss_mlp": 1.01634991,
      "epoch": 0.4340297610100706,
      "flos": 19719052473600.0,
      "grad_norm": 1.5753300192395339,
      "language_loss": 0.68584305,
      "learning_rate": 2.5184217790371367e-06,
      "loss": 0.70719016,
      "num_input_tokens_seen": 154813105,
      "step": 7219,
      "time_per_iteration": 2.53041934967041
    },
    {
      "auxiliary_loss_clip": 0.01093286,
      "auxiliary_loss_mlp": 0.01033601,
      "balance_loss_clip": 1.04342222,
      "balance_loss_mlp": 1.01999974,
      "epoch": 0.4340898842627386,
      "flos": 18953508885120.0,
      "grad_norm": 1.858604914496032,
      "language_loss": 0.77559412,
      "learning_rate": 2.518045619038202e-06,
      "loss": 0.79686302,
      "num_input_tokens_seen": 154833525,
      "step": 7220,
      "time_per_iteration": 3.98419189453125
    },
    {
      "auxiliary_loss_clip": 0.01058319,
      "auxiliary_loss_mlp": 0.01030657,
      "balance_loss_clip": 1.04139566,
      "balance_loss_mlp": 1.01694202,
      "epoch": 0.4341500075154066,
      "flos": 22018304931840.0,
      "grad_norm": 2.18701612705658,
      "language_loss": 0.69821835,
      "learning_rate": 2.5176694393949243e-06,
      "loss": 0.71910816,
      "num_input_tokens_seen": 154853090,
      "step": 7221,
      "time_per_iteration": 2.6250741481781006
    },
    {
      "auxiliary_loss_clip": 0.01113913,
      "auxiliary_loss_mlp": 0.010325,
      "balance_loss_clip": 1.04329634,
      "balance_loss_mlp": 1.01972103,
      "epoch": 0.4342101307680746,
      "flos": 23582465556480.0,
      "grad_norm": 3.208549470630045,
      "language_loss": 0.64773929,
      "learning_rate": 2.51729324012157e-06,
      "loss": 0.6692034,
      "num_input_tokens_seen": 154872055,
      "step": 7222,
      "time_per_iteration": 2.532081365585327
    },
    {
      "auxiliary_loss_clip": 0.01088214,
      "auxiliary_loss_mlp": 0.01031123,
      "balance_loss_clip": 1.04194415,
      "balance_loss_mlp": 1.01697838,
      "epoch": 0.43427025402074254,
      "flos": 17967976450560.0,
      "grad_norm": 5.003645368780293,
      "language_loss": 0.72874624,
      "learning_rate": 2.5169170212324053e-06,
      "loss": 0.74993962,
      "num_input_tokens_seen": 154886645,
      "step": 7223,
      "time_per_iteration": 2.5219576358795166
    },
    {
      "auxiliary_loss_clip": 0.01123806,
      "auxiliary_loss_mlp": 0.01031069,
      "balance_loss_clip": 1.04263806,
      "balance_loss_mlp": 1.01699591,
      "epoch": 0.4343303772734105,
      "flos": 26286395616000.0,
      "grad_norm": 10.947192979562725,
      "language_loss": 0.94228798,
      "learning_rate": 2.516540782741694e-06,
      "loss": 0.96383667,
      "num_input_tokens_seen": 154906775,
      "step": 7224,
      "time_per_iteration": 2.539252519607544
    },
    {
      "auxiliary_loss_clip": 0.01086542,
      "auxiliary_loss_mlp": 0.01040426,
      "balance_loss_clip": 1.04372263,
      "balance_loss_mlp": 1.02651477,
      "epoch": 0.43439050052607847,
      "flos": 26833961520000.0,
      "grad_norm": 2.4043241583048585,
      "language_loss": 0.61016935,
      "learning_rate": 2.5161645246637056e-06,
      "loss": 0.63143909,
      "num_input_tokens_seen": 154926990,
      "step": 7225,
      "time_per_iteration": 4.064651966094971
    },
    {
      "auxiliary_loss_clip": 0.0109511,
      "auxiliary_loss_mlp": 0.00790348,
      "balance_loss_clip": 1.04293156,
      "balance_loss_mlp": 1.01318264,
      "epoch": 0.43445062377874644,
      "flos": 21397660807680.0,
      "grad_norm": 2.187514532586146,
      "language_loss": 0.78010499,
      "learning_rate": 2.5157882470127054e-06,
      "loss": 0.79895961,
      "num_input_tokens_seen": 154946210,
      "step": 7226,
      "time_per_iteration": 2.5720386505126953
    },
    {
      "auxiliary_loss_clip": 0.01109896,
      "auxiliary_loss_mlp": 0.01029652,
      "balance_loss_clip": 1.04391313,
      "balance_loss_mlp": 1.01678312,
      "epoch": 0.4345107470314144,
      "flos": 19901945548800.0,
      "grad_norm": 1.54698783174401,
      "language_loss": 0.84462744,
      "learning_rate": 2.515411949802964e-06,
      "loss": 0.86602294,
      "num_input_tokens_seen": 154964995,
      "step": 7227,
      "time_per_iteration": 2.4900476932525635
    },
    {
      "auxiliary_loss_clip": 0.01107763,
      "auxiliary_loss_mlp": 0.01034924,
      "balance_loss_clip": 1.04310393,
      "balance_loss_mlp": 1.02083981,
      "epoch": 0.43457087028408237,
      "flos": 26432623883520.0,
      "grad_norm": 1.8800967372303998,
      "language_loss": 0.76712102,
      "learning_rate": 2.5150356330487498e-06,
      "loss": 0.78854787,
      "num_input_tokens_seen": 154984775,
      "step": 7228,
      "time_per_iteration": 2.604343891143799
    },
    {
      "auxiliary_loss_clip": 0.01081143,
      "auxiliary_loss_mlp": 0.01036882,
      "balance_loss_clip": 1.04985726,
      "balance_loss_mlp": 1.02293456,
      "epoch": 0.43463099353675033,
      "flos": 31868816855040.0,
      "grad_norm": 1.5915338970076094,
      "language_loss": 0.80441368,
      "learning_rate": 2.5146592967643324e-06,
      "loss": 0.82559395,
      "num_input_tokens_seen": 155008125,
      "step": 7229,
      "time_per_iteration": 2.774876356124878
    },
    {
      "auxiliary_loss_clip": 0.01109515,
      "auxiliary_loss_mlp": 0.01037841,
      "balance_loss_clip": 1.04201961,
      "balance_loss_mlp": 1.02433515,
      "epoch": 0.4346911167894183,
      "flos": 24571266128640.0,
      "grad_norm": 1.8046146559866976,
      "language_loss": 0.82312679,
      "learning_rate": 2.5142829409639834e-06,
      "loss": 0.84460032,
      "num_input_tokens_seen": 155027885,
      "step": 7230,
      "time_per_iteration": 2.612349033355713
    },
    {
      "auxiliary_loss_clip": 0.01110515,
      "auxiliary_loss_mlp": 0.01042211,
      "balance_loss_clip": 1.04521632,
      "balance_loss_mlp": 1.02783442,
      "epoch": 0.43475124004208626,
      "flos": 17090678672640.0,
      "grad_norm": 2.205667708744853,
      "language_loss": 0.77092731,
      "learning_rate": 2.513906565661973e-06,
      "loss": 0.7924546,
      "num_input_tokens_seen": 155043375,
      "step": 7231,
      "time_per_iteration": 2.54496693611145
    },
    {
      "auxiliary_loss_clip": 0.01081378,
      "auxiliary_loss_mlp": 0.01031175,
      "balance_loss_clip": 1.04426992,
      "balance_loss_mlp": 1.01901007,
      "epoch": 0.4348113632947542,
      "flos": 26104615862400.0,
      "grad_norm": 1.4569014854656854,
      "language_loss": 0.68888581,
      "learning_rate": 2.513530170872575e-06,
      "loss": 0.71001136,
      "num_input_tokens_seen": 155062930,
      "step": 7232,
      "time_per_iteration": 2.66925311088562
    },
    {
      "auxiliary_loss_clip": 0.0108838,
      "auxiliary_loss_mlp": 0.01032739,
      "balance_loss_clip": 1.04230499,
      "balance_loss_mlp": 1.01840377,
      "epoch": 0.4348714865474222,
      "flos": 34200496316160.0,
      "grad_norm": 1.6207125911339715,
      "language_loss": 0.72164738,
      "learning_rate": 2.5131537566100605e-06,
      "loss": 0.74285853,
      "num_input_tokens_seen": 155084980,
      "step": 7233,
      "time_per_iteration": 2.7525479793548584
    },
    {
      "auxiliary_loss_clip": 0.01063675,
      "auxiliary_loss_mlp": 0.01036797,
      "balance_loss_clip": 1.04094768,
      "balance_loss_mlp": 1.02156186,
      "epoch": 0.43493160980009016,
      "flos": 31537468869120.0,
      "grad_norm": 1.7189075509451857,
      "language_loss": 0.74246156,
      "learning_rate": 2.5127773228887053e-06,
      "loss": 0.76346624,
      "num_input_tokens_seen": 155107260,
      "step": 7234,
      "time_per_iteration": 2.7633466720581055
    },
    {
      "auxiliary_loss_clip": 0.0110055,
      "auxiliary_loss_mlp": 0.01035951,
      "balance_loss_clip": 1.04332554,
      "balance_loss_mlp": 1.02116871,
      "epoch": 0.4349917330527582,
      "flos": 24061334699520.0,
      "grad_norm": 5.848003252674072,
      "language_loss": 0.59482384,
      "learning_rate": 2.512400869722782e-06,
      "loss": 0.61618888,
      "num_input_tokens_seen": 155126720,
      "step": 7235,
      "time_per_iteration": 2.5894975662231445
    },
    {
      "auxiliary_loss_clip": 0.01060253,
      "auxiliary_loss_mlp": 0.01039881,
      "balance_loss_clip": 1.04084921,
      "balance_loss_mlp": 1.023633,
      "epoch": 0.43505185630542614,
      "flos": 30519329863680.0,
      "grad_norm": 1.6714004043355728,
      "language_loss": 0.77538133,
      "learning_rate": 2.512024397126566e-06,
      "loss": 0.79638267,
      "num_input_tokens_seen": 155148640,
      "step": 7236,
      "time_per_iteration": 2.6797313690185547
    },
    {
      "auxiliary_loss_clip": 0.01119456,
      "auxiliary_loss_mlp": 0.01031595,
      "balance_loss_clip": 1.04346967,
      "balance_loss_mlp": 1.01773715,
      "epoch": 0.4351119795580941,
      "flos": 15735158196480.0,
      "grad_norm": 1.7825589073536603,
      "language_loss": 0.81316006,
      "learning_rate": 2.5116479051143345e-06,
      "loss": 0.83467054,
      "num_input_tokens_seen": 155165870,
      "step": 7237,
      "time_per_iteration": 2.468858003616333
    },
    {
      "auxiliary_loss_clip": 0.01108725,
      "auxiliary_loss_mlp": 0.01035763,
      "balance_loss_clip": 1.04201567,
      "balance_loss_mlp": 1.02188659,
      "epoch": 0.4351721028107621,
      "flos": 18731760272640.0,
      "grad_norm": 1.4130004653838943,
      "language_loss": 0.63141465,
      "learning_rate": 2.5112713937003623e-06,
      "loss": 0.65285957,
      "num_input_tokens_seen": 155185315,
      "step": 7238,
      "time_per_iteration": 2.512594223022461
    },
    {
      "auxiliary_loss_clip": 0.01085322,
      "auxiliary_loss_mlp": 0.00786771,
      "balance_loss_clip": 1.0417273,
      "balance_loss_mlp": 1.00752139,
      "epoch": 0.43523222606343004,
      "flos": 25226887121280.0,
      "grad_norm": 1.5779036020295623,
      "language_loss": 0.85861874,
      "learning_rate": 2.510894862898928e-06,
      "loss": 0.87733972,
      "num_input_tokens_seen": 155205790,
      "step": 7239,
      "time_per_iteration": 2.6338257789611816
    },
    {
      "auxiliary_loss_clip": 0.0110299,
      "auxiliary_loss_mlp": 0.01029941,
      "balance_loss_clip": 1.04554939,
      "balance_loss_mlp": 1.0164113,
      "epoch": 0.435292349316098,
      "flos": 22709190101760.0,
      "grad_norm": 1.9302159132422705,
      "language_loss": 0.7256124,
      "learning_rate": 2.510518312724309e-06,
      "loss": 0.74694169,
      "num_input_tokens_seen": 155226475,
      "step": 7240,
      "time_per_iteration": 2.606358766555786
    },
    {
      "auxiliary_loss_clip": 0.01090677,
      "auxiliary_loss_mlp": 0.0103102,
      "balance_loss_clip": 1.04515338,
      "balance_loss_mlp": 1.0163933,
      "epoch": 0.43535247256876597,
      "flos": 25775889569280.0,
      "grad_norm": 1.8026667005756591,
      "language_loss": 0.8211199,
      "learning_rate": 2.5101417431907842e-06,
      "loss": 0.84233689,
      "num_input_tokens_seen": 155247110,
      "step": 7241,
      "time_per_iteration": 2.6029632091522217
    },
    {
      "auxiliary_loss_clip": 0.0109408,
      "auxiliary_loss_mlp": 0.00786645,
      "balance_loss_clip": 1.04373944,
      "balance_loss_mlp": 1.00742781,
      "epoch": 0.43541259582143393,
      "flos": 17528142412800.0,
      "grad_norm": 2.433334628088347,
      "language_loss": 0.7939961,
      "learning_rate": 2.5097651543126345e-06,
      "loss": 0.81280339,
      "num_input_tokens_seen": 155261335,
      "step": 7242,
      "time_per_iteration": 2.5214080810546875
    },
    {
      "auxiliary_loss_clip": 0.01097931,
      "auxiliary_loss_mlp": 0.01032762,
      "balance_loss_clip": 1.04104292,
      "balance_loss_mlp": 1.01800418,
      "epoch": 0.4354727190741019,
      "flos": 15195205975680.0,
      "grad_norm": 5.902734645839141,
      "language_loss": 0.68018556,
      "learning_rate": 2.509388546104138e-06,
      "loss": 0.70149249,
      "num_input_tokens_seen": 155278510,
      "step": 7243,
      "time_per_iteration": 2.519796133041382
    },
    {
      "auxiliary_loss_clip": 0.01067391,
      "auxiliary_loss_mlp": 0.01028259,
      "balance_loss_clip": 1.04196358,
      "balance_loss_mlp": 1.0154078,
      "epoch": 0.43553284232676986,
      "flos": 16649264436480.0,
      "grad_norm": 1.8763201142544046,
      "language_loss": 0.81636095,
      "learning_rate": 2.5090119185795766e-06,
      "loss": 0.83731741,
      "num_input_tokens_seen": 155296450,
      "step": 7244,
      "time_per_iteration": 2.5679445266723633
    },
    {
      "auxiliary_loss_clip": 0.01063875,
      "auxiliary_loss_mlp": 0.01028406,
      "balance_loss_clip": 1.0415442,
      "balance_loss_mlp": 1.0154475,
      "epoch": 0.43559296557943783,
      "flos": 23400865370880.0,
      "grad_norm": 1.8546071150116974,
      "language_loss": 0.73555183,
      "learning_rate": 2.508635271753234e-06,
      "loss": 0.75647461,
      "num_input_tokens_seen": 155316080,
      "step": 7245,
      "time_per_iteration": 2.6432852745056152
    },
    {
      "auxiliary_loss_clip": 0.01066872,
      "auxiliary_loss_mlp": 0.01036181,
      "balance_loss_clip": 1.0455848,
      "balance_loss_mlp": 1.02277589,
      "epoch": 0.4356530888321058,
      "flos": 22419067950720.0,
      "grad_norm": 1.6613046005343854,
      "language_loss": 0.76942348,
      "learning_rate": 2.508258605639389e-06,
      "loss": 0.79045397,
      "num_input_tokens_seen": 155336765,
      "step": 7246,
      "time_per_iteration": 2.62353777885437
    },
    {
      "auxiliary_loss_clip": 0.01110325,
      "auxiliary_loss_mlp": 0.01036064,
      "balance_loss_clip": 1.04263926,
      "balance_loss_mlp": 1.02215207,
      "epoch": 0.43571321208477376,
      "flos": 21616141282560.0,
      "grad_norm": 1.7381373929755897,
      "language_loss": 0.85437596,
      "learning_rate": 2.5078819202523275e-06,
      "loss": 0.87583983,
      "num_input_tokens_seen": 155356440,
      "step": 7247,
      "time_per_iteration": 2.5292601585388184
    },
    {
      "auxiliary_loss_clip": 0.01122932,
      "auxiliary_loss_mlp": 0.01034083,
      "balance_loss_clip": 1.04459679,
      "balance_loss_mlp": 1.02145314,
      "epoch": 0.4357733353374418,
      "flos": 23987358639360.0,
      "grad_norm": 1.5458919130198876,
      "language_loss": 0.72231913,
      "learning_rate": 2.507505215606333e-06,
      "loss": 0.74388933,
      "num_input_tokens_seen": 155377070,
      "step": 7248,
      "time_per_iteration": 2.497880697250366
    },
    {
      "auxiliary_loss_clip": 0.01110932,
      "auxiliary_loss_mlp": 0.01029607,
      "balance_loss_clip": 1.04502988,
      "balance_loss_mlp": 1.0160116,
      "epoch": 0.43583345859010975,
      "flos": 25264737077760.0,
      "grad_norm": 1.520602260819967,
      "language_loss": 0.87565982,
      "learning_rate": 2.5071284917156893e-06,
      "loss": 0.89706528,
      "num_input_tokens_seen": 155398415,
      "step": 7249,
      "time_per_iteration": 2.5495471954345703
    },
    {
      "auxiliary_loss_clip": 0.01104734,
      "auxiliary_loss_mlp": 0.01036851,
      "balance_loss_clip": 1.04439354,
      "balance_loss_mlp": 1.02391672,
      "epoch": 0.4358935818427777,
      "flos": 23696302734720.0,
      "grad_norm": 1.685638736407713,
      "language_loss": 0.816212,
      "learning_rate": 2.506751748594683e-06,
      "loss": 0.83762789,
      "num_input_tokens_seen": 155415625,
      "step": 7250,
      "time_per_iteration": 2.550302505493164
    },
    {
      "auxiliary_loss_clip": 0.01115696,
      "auxiliary_loss_mlp": 0.01032518,
      "balance_loss_clip": 1.0479089,
      "balance_loss_mlp": 1.01919651,
      "epoch": 0.4359537050954457,
      "flos": 29532827761920.0,
      "grad_norm": 1.887137288889742,
      "language_loss": 0.84538352,
      "learning_rate": 2.5063749862575988e-06,
      "loss": 0.86686569,
      "num_input_tokens_seen": 155435505,
      "step": 7251,
      "time_per_iteration": 2.5762550830841064
    },
    {
      "auxiliary_loss_clip": 0.01100795,
      "auxiliary_loss_mlp": 0.01042239,
      "balance_loss_clip": 1.03984594,
      "balance_loss_mlp": 1.02789164,
      "epoch": 0.43601382834811364,
      "flos": 22711273090560.0,
      "grad_norm": 1.6712002684322527,
      "language_loss": 0.69168866,
      "learning_rate": 2.5059982047187245e-06,
      "loss": 0.71311903,
      "num_input_tokens_seen": 155455425,
      "step": 7252,
      "time_per_iteration": 2.5006253719329834
    },
    {
      "auxiliary_loss_clip": 0.01097068,
      "auxiliary_loss_mlp": 0.01034741,
      "balance_loss_clip": 1.04482543,
      "balance_loss_mlp": 1.01962554,
      "epoch": 0.4360739516007816,
      "flos": 19098731571840.0,
      "grad_norm": 1.6734478385039597,
      "language_loss": 0.83452332,
      "learning_rate": 2.505621403992348e-06,
      "loss": 0.8558414,
      "num_input_tokens_seen": 155474250,
      "step": 7253,
      "time_per_iteration": 2.5249991416931152
    },
    {
      "auxiliary_loss_clip": 0.0111379,
      "auxiliary_loss_mlp": 0.01037383,
      "balance_loss_clip": 1.0484376,
      "balance_loss_mlp": 1.02347755,
      "epoch": 0.43613407485344957,
      "flos": 23404420817280.0,
      "grad_norm": 1.4768437310906288,
      "language_loss": 0.70236647,
      "learning_rate": 2.505244584092757e-06,
      "loss": 0.7238782,
      "num_input_tokens_seen": 155494685,
      "step": 7254,
      "time_per_iteration": 2.5210423469543457
    },
    {
      "auxiliary_loss_clip": 0.01100443,
      "auxiliary_loss_mlp": 0.01032837,
      "balance_loss_clip": 1.04544401,
      "balance_loss_mlp": 1.01984358,
      "epoch": 0.43619419810611754,
      "flos": 22637799820800.0,
      "grad_norm": 1.77377815934272,
      "language_loss": 0.81275082,
      "learning_rate": 2.5048677450342406e-06,
      "loss": 0.83408368,
      "num_input_tokens_seen": 155513040,
      "step": 7255,
      "time_per_iteration": 2.5297908782958984
    },
    {
      "auxiliary_loss_clip": 0.01123111,
      "auxiliary_loss_mlp": 0.0103594,
      "balance_loss_clip": 1.04455829,
      "balance_loss_mlp": 1.02286911,
      "epoch": 0.4362543213587855,
      "flos": 20047958334720.0,
      "grad_norm": 1.687830498793561,
      "language_loss": 0.77567458,
      "learning_rate": 2.504490886831089e-06,
      "loss": 0.79726505,
      "num_input_tokens_seen": 155530100,
      "step": 7256,
      "time_per_iteration": 2.4536232948303223
    },
    {
      "auxiliary_loss_clip": 0.01121378,
      "auxiliary_loss_mlp": 0.01032939,
      "balance_loss_clip": 1.045259,
      "balance_loss_mlp": 1.02013636,
      "epoch": 0.43631444461145347,
      "flos": 21361319222400.0,
      "grad_norm": 1.5985138794526068,
      "language_loss": 0.75924516,
      "learning_rate": 2.5041140094975922e-06,
      "loss": 0.7807883,
      "num_input_tokens_seen": 155549375,
      "step": 7257,
      "time_per_iteration": 3.9023940563201904
    },
    {
      "auxiliary_loss_clip": 0.01111447,
      "auxiliary_loss_mlp": 0.0103793,
      "balance_loss_clip": 1.04384232,
      "balance_loss_mlp": 1.02321959,
      "epoch": 0.43637456786412143,
      "flos": 22418529246720.0,
      "grad_norm": 2.312139632338265,
      "language_loss": 0.72776413,
      "learning_rate": 2.5037371130480417e-06,
      "loss": 0.74925798,
      "num_input_tokens_seen": 155569395,
      "step": 7258,
      "time_per_iteration": 3.927535057067871
    },
    {
      "auxiliary_loss_clip": 0.01100172,
      "auxiliary_loss_mlp": 0.01033658,
      "balance_loss_clip": 1.0451256,
      "balance_loss_mlp": 1.02033019,
      "epoch": 0.4364346911167894,
      "flos": 28548839612160.0,
      "grad_norm": 1.8346638906606565,
      "language_loss": 0.76694906,
      "learning_rate": 2.5033601974967297e-06,
      "loss": 0.7882874,
      "num_input_tokens_seen": 155589090,
      "step": 7259,
      "time_per_iteration": 2.578845500946045
    },
    {
      "auxiliary_loss_clip": 0.01031867,
      "auxiliary_loss_mlp": 0.01003387,
      "balance_loss_clip": 1.03248405,
      "balance_loss_mlp": 1.00170612,
      "epoch": 0.43649481436945736,
      "flos": 62659345380480.0,
      "grad_norm": 0.7456643669606448,
      "language_loss": 0.57006007,
      "learning_rate": 2.5029832628579483e-06,
      "loss": 0.59041262,
      "num_input_tokens_seen": 155648660,
      "step": 7260,
      "time_per_iteration": 3.0990757942199707
    },
    {
      "auxiliary_loss_clip": 0.01100509,
      "auxiliary_loss_mlp": 0.01041276,
      "balance_loss_clip": 1.04117405,
      "balance_loss_mlp": 1.02651823,
      "epoch": 0.4365549376221254,
      "flos": 30592120775040.0,
      "grad_norm": 1.817178369668107,
      "language_loss": 0.71150047,
      "learning_rate": 2.5026063091459907e-06,
      "loss": 0.73291832,
      "num_input_tokens_seen": 155669945,
      "step": 7261,
      "time_per_iteration": 2.6088192462921143
    },
    {
      "auxiliary_loss_clip": 0.01073701,
      "auxiliary_loss_mlp": 0.01050368,
      "balance_loss_clip": 1.04024625,
      "balance_loss_mlp": 1.03521013,
      "epoch": 0.43661506087479335,
      "flos": 17165875795200.0,
      "grad_norm": 2.0033602905366945,
      "language_loss": 0.69336605,
      "learning_rate": 2.5022293363751522e-06,
      "loss": 0.71460676,
      "num_input_tokens_seen": 155688555,
      "step": 7262,
      "time_per_iteration": 2.5762970447540283
    },
    {
      "auxiliary_loss_clip": 0.01066765,
      "auxiliary_loss_mlp": 0.0103067,
      "balance_loss_clip": 1.0453409,
      "balance_loss_mlp": 1.01886189,
      "epoch": 0.4366751841274613,
      "flos": 22047499710720.0,
      "grad_norm": 1.6913406143307854,
      "language_loss": 0.79506409,
      "learning_rate": 2.501852344559726e-06,
      "loss": 0.81603837,
      "num_input_tokens_seen": 155705370,
      "step": 7263,
      "time_per_iteration": 4.011101007461548
    },
    {
      "auxiliary_loss_clip": 0.01085772,
      "auxiliary_loss_mlp": 0.01045001,
      "balance_loss_clip": 1.04642701,
      "balance_loss_mlp": 1.03154778,
      "epoch": 0.4367353073801293,
      "flos": 15997306631040.0,
      "grad_norm": 1.7617497799418937,
      "language_loss": 0.7511878,
      "learning_rate": 2.50147533371401e-06,
      "loss": 0.77249551,
      "num_input_tokens_seen": 155721890,
      "step": 7264,
      "time_per_iteration": 3.9633805751800537
    },
    {
      "auxiliary_loss_clip": 0.0106874,
      "auxiliary_loss_mlp": 0.01033149,
      "balance_loss_clip": 1.04427171,
      "balance_loss_mlp": 1.01911759,
      "epoch": 0.43679543063279724,
      "flos": 38217535868160.0,
      "grad_norm": 2.400409641338185,
      "language_loss": 0.61648768,
      "learning_rate": 2.501098303852298e-06,
      "loss": 0.63750654,
      "num_input_tokens_seen": 155743970,
      "step": 7265,
      "time_per_iteration": 2.7330641746520996
    },
    {
      "auxiliary_loss_clip": 0.01096577,
      "auxiliary_loss_mlp": 0.01027947,
      "balance_loss_clip": 1.04157257,
      "balance_loss_mlp": 1.01507807,
      "epoch": 0.4368555538854652,
      "flos": 15193230727680.0,
      "grad_norm": 1.9545913321836685,
      "language_loss": 0.72570801,
      "learning_rate": 2.5007212549888884e-06,
      "loss": 0.74695319,
      "num_input_tokens_seen": 155761830,
      "step": 7266,
      "time_per_iteration": 2.523548126220703
    },
    {
      "auxiliary_loss_clip": 0.01098852,
      "auxiliary_loss_mlp": 0.01033409,
      "balance_loss_clip": 1.04542255,
      "balance_loss_mlp": 1.02061224,
      "epoch": 0.4369156771381332,
      "flos": 23069086421760.0,
      "grad_norm": 2.558901498717704,
      "language_loss": 0.81844378,
      "learning_rate": 2.5003441871380794e-06,
      "loss": 0.83976638,
      "num_input_tokens_seen": 155779610,
      "step": 7267,
      "time_per_iteration": 2.5329091548919678
    },
    {
      "auxiliary_loss_clip": 0.01117704,
      "auxiliary_loss_mlp": 0.01030942,
      "balance_loss_clip": 1.04261303,
      "balance_loss_mlp": 1.01867557,
      "epoch": 0.43697580039080114,
      "flos": 23441085624960.0,
      "grad_norm": 2.0973947894285203,
      "language_loss": 0.74858427,
      "learning_rate": 2.4999671003141674e-06,
      "loss": 0.77007073,
      "num_input_tokens_seen": 155798765,
      "step": 7268,
      "time_per_iteration": 2.490900754928589
    },
    {
      "auxiliary_loss_clip": 0.01126043,
      "auxiliary_loss_mlp": 0.01037754,
      "balance_loss_clip": 1.04504108,
      "balance_loss_mlp": 1.02358043,
      "epoch": 0.4370359236434691,
      "flos": 18514680428160.0,
      "grad_norm": 2.107106413817305,
      "language_loss": 0.80177653,
      "learning_rate": 2.499589994531454e-06,
      "loss": 0.82341444,
      "num_input_tokens_seen": 155817750,
      "step": 7269,
      "time_per_iteration": 2.4495410919189453
    },
    {
      "auxiliary_loss_clip": 0.01101904,
      "auxiliary_loss_mlp": 0.01035499,
      "balance_loss_clip": 1.04506707,
      "balance_loss_mlp": 1.02247596,
      "epoch": 0.43709604689613707,
      "flos": 23222497409280.0,
      "grad_norm": 1.9840089485260395,
      "language_loss": 0.75407505,
      "learning_rate": 2.499212869804237e-06,
      "loss": 0.7754491,
      "num_input_tokens_seen": 155836490,
      "step": 7270,
      "time_per_iteration": 2.537155866622925
    },
    {
      "auxiliary_loss_clip": 0.01062421,
      "auxiliary_loss_mlp": 0.01040168,
      "balance_loss_clip": 1.03981733,
      "balance_loss_mlp": 1.02467108,
      "epoch": 0.43715617014880503,
      "flos": 23803711378560.0,
      "grad_norm": 1.887500388953251,
      "language_loss": 0.79860747,
      "learning_rate": 2.4988357261468182e-06,
      "loss": 0.81963336,
      "num_input_tokens_seen": 155856225,
      "step": 7271,
      "time_per_iteration": 2.6116034984588623
    },
    {
      "auxiliary_loss_clip": 0.010383,
      "auxiliary_loss_mlp": 0.01005072,
      "balance_loss_clip": 1.02248502,
      "balance_loss_mlp": 1.00331962,
      "epoch": 0.437216293401473,
      "flos": 61941204766080.0,
      "grad_norm": 0.7037976774725125,
      "language_loss": 0.54939276,
      "learning_rate": 2.4984585635734993e-06,
      "loss": 0.56982648,
      "num_input_tokens_seen": 155916770,
      "step": 7272,
      "time_per_iteration": 3.1886825561523438
    },
    {
      "auxiliary_loss_clip": 0.01124651,
      "auxiliary_loss_mlp": 0.01039253,
      "balance_loss_clip": 1.04469633,
      "balance_loss_mlp": 1.02481043,
      "epoch": 0.43727641665414096,
      "flos": 21982250655360.0,
      "grad_norm": 1.657729723776728,
      "language_loss": 0.6988585,
      "learning_rate": 2.498081382098581e-06,
      "loss": 0.72049749,
      "num_input_tokens_seen": 155936490,
      "step": 7273,
      "time_per_iteration": 2.4779486656188965
    },
    {
      "auxiliary_loss_clip": 0.01098901,
      "auxiliary_loss_mlp": 0.01043537,
      "balance_loss_clip": 1.04207301,
      "balance_loss_mlp": 1.0280515,
      "epoch": 0.437336539906809,
      "flos": 39530860842240.0,
      "grad_norm": 1.7662562886006012,
      "language_loss": 0.75557721,
      "learning_rate": 2.497704181736367e-06,
      "loss": 0.77700162,
      "num_input_tokens_seen": 155957595,
      "step": 7274,
      "time_per_iteration": 2.672356367111206
    },
    {
      "auxiliary_loss_clip": 0.01103444,
      "auxiliary_loss_mlp": 0.01025325,
      "balance_loss_clip": 1.04186785,
      "balance_loss_mlp": 1.0135653,
      "epoch": 0.43739666315947695,
      "flos": 17457147181440.0,
      "grad_norm": 1.838885308291702,
      "language_loss": 0.80172539,
      "learning_rate": 2.49732696250116e-06,
      "loss": 0.82301307,
      "num_input_tokens_seen": 155975710,
      "step": 7275,
      "time_per_iteration": 2.493558883666992
    },
    {
      "auxiliary_loss_clip": 0.01100411,
      "auxiliary_loss_mlp": 0.01034172,
      "balance_loss_clip": 1.04695892,
      "balance_loss_mlp": 1.02090383,
      "epoch": 0.4374567864121449,
      "flos": 16358747235840.0,
      "grad_norm": 2.2259481091365645,
      "language_loss": 0.80802965,
      "learning_rate": 2.496949724407266e-06,
      "loss": 0.82937551,
      "num_input_tokens_seen": 155993090,
      "step": 7276,
      "time_per_iteration": 2.5278773307800293
    },
    {
      "auxiliary_loss_clip": 0.01110603,
      "auxiliary_loss_mlp": 0.01029639,
      "balance_loss_clip": 1.04522681,
      "balance_loss_mlp": 1.01557779,
      "epoch": 0.4375169096648129,
      "flos": 30587523834240.0,
      "grad_norm": 2.479621374122403,
      "language_loss": 0.73125517,
      "learning_rate": 2.496572467468988e-06,
      "loss": 0.75265765,
      "num_input_tokens_seen": 156013685,
      "step": 7277,
      "time_per_iteration": 2.598536491394043
    },
    {
      "auxiliary_loss_clip": 0.01101481,
      "auxiliary_loss_mlp": 0.00787412,
      "balance_loss_clip": 1.04628766,
      "balance_loss_mlp": 1.00917768,
      "epoch": 0.43757703291748085,
      "flos": 30555599621760.0,
      "grad_norm": 1.8589230004581698,
      "language_loss": 0.73223746,
      "learning_rate": 2.4961951917006317e-06,
      "loss": 0.75112641,
      "num_input_tokens_seen": 156034300,
      "step": 7278,
      "time_per_iteration": 2.6280500888824463
    },
    {
      "auxiliary_loss_clip": 0.01090068,
      "auxiliary_loss_mlp": 0.01032578,
      "balance_loss_clip": 1.04243207,
      "balance_loss_mlp": 1.0203706,
      "epoch": 0.4376371561701488,
      "flos": 21397373498880.0,
      "grad_norm": 1.6253055856092011,
      "language_loss": 0.66339689,
      "learning_rate": 2.4958178971165046e-06,
      "loss": 0.68462336,
      "num_input_tokens_seen": 156053805,
      "step": 7279,
      "time_per_iteration": 2.550201892852783
    },
    {
      "auxiliary_loss_clip": 0.0112503,
      "auxiliary_loss_mlp": 0.01033004,
      "balance_loss_clip": 1.04413152,
      "balance_loss_mlp": 1.01925349,
      "epoch": 0.4376972794228168,
      "flos": 23404384903680.0,
      "grad_norm": 1.7962008852571945,
      "language_loss": 0.82139683,
      "learning_rate": 2.4954405837309126e-06,
      "loss": 0.84297723,
      "num_input_tokens_seen": 156073295,
      "step": 7280,
      "time_per_iteration": 2.480160713195801
    },
    {
      "auxiliary_loss_clip": 0.01096027,
      "auxiliary_loss_mlp": 0.01031306,
      "balance_loss_clip": 1.04140377,
      "balance_loss_mlp": 1.01856279,
      "epoch": 0.43775740267548474,
      "flos": 22892945103360.0,
      "grad_norm": 1.5435880847155574,
      "language_loss": 0.76742607,
      "learning_rate": 2.4950632515581653e-06,
      "loss": 0.78869939,
      "num_input_tokens_seen": 156094540,
      "step": 7281,
      "time_per_iteration": 2.5410802364349365
    },
    {
      "auxiliary_loss_clip": 0.01094955,
      "auxiliary_loss_mlp": 0.01039375,
      "balance_loss_clip": 1.04076898,
      "balance_loss_mlp": 1.0259521,
      "epoch": 0.4378175259281527,
      "flos": 23294390480640.0,
      "grad_norm": 2.085014865884542,
      "language_loss": 0.7566514,
      "learning_rate": 2.494685900612569e-06,
      "loss": 0.77799469,
      "num_input_tokens_seen": 156114070,
      "step": 7282,
      "time_per_iteration": 2.529700517654419
    },
    {
      "auxiliary_loss_clip": 0.01085106,
      "auxiliary_loss_mlp": 0.01041089,
      "balance_loss_clip": 1.04415417,
      "balance_loss_mlp": 1.02702856,
      "epoch": 0.43787764918082067,
      "flos": 23876897339520.0,
      "grad_norm": 1.6748802268441414,
      "language_loss": 0.84705281,
      "learning_rate": 2.4943085309084333e-06,
      "loss": 0.86831474,
      "num_input_tokens_seen": 156132130,
      "step": 7283,
      "time_per_iteration": 2.5963010787963867
    },
    {
      "auxiliary_loss_clip": 0.0110727,
      "auxiliary_loss_mlp": 0.0103567,
      "balance_loss_clip": 1.04401755,
      "balance_loss_mlp": 1.0214963,
      "epoch": 0.43793777243348864,
      "flos": 23988148738560.0,
      "grad_norm": 1.7193257834672488,
      "language_loss": 0.8056078,
      "learning_rate": 2.49393114246007e-06,
      "loss": 0.82703716,
      "num_input_tokens_seen": 156150820,
      "step": 7284,
      "time_per_iteration": 2.5541956424713135
    },
    {
      "auxiliary_loss_clip": 0.01111387,
      "auxiliary_loss_mlp": 0.01042707,
      "balance_loss_clip": 1.04373908,
      "balance_loss_mlp": 1.02995777,
      "epoch": 0.4379978956861566,
      "flos": 18624064320000.0,
      "grad_norm": 1.4682978882081847,
      "language_loss": 0.79967725,
      "learning_rate": 2.493553735281787e-06,
      "loss": 0.82121819,
      "num_input_tokens_seen": 156170125,
      "step": 7285,
      "time_per_iteration": 2.4881415367126465
    },
    {
      "auxiliary_loss_clip": 0.01108772,
      "auxiliary_loss_mlp": 0.0102905,
      "balance_loss_clip": 1.04188323,
      "balance_loss_mlp": 1.0158,
      "epoch": 0.43805801893882457,
      "flos": 21981388728960.0,
      "grad_norm": 2.039534793323649,
      "language_loss": 0.7510916,
      "learning_rate": 2.493176309387897e-06,
      "loss": 0.77246988,
      "num_input_tokens_seen": 156187320,
      "step": 7286,
      "time_per_iteration": 2.495570659637451
    },
    {
      "auxiliary_loss_clip": 0.01086601,
      "auxiliary_loss_mlp": 0.01028963,
      "balance_loss_clip": 1.0411942,
      "balance_loss_mlp": 1.01488447,
      "epoch": 0.43811814219149253,
      "flos": 26393337383040.0,
      "grad_norm": 1.5211932527350813,
      "language_loss": 0.73524094,
      "learning_rate": 2.492798864792712e-06,
      "loss": 0.75639659,
      "num_input_tokens_seen": 156207455,
      "step": 7287,
      "time_per_iteration": 2.6239006519317627
    },
    {
      "auxiliary_loss_clip": 0.01100832,
      "auxiliary_loss_mlp": 0.01037939,
      "balance_loss_clip": 1.04391932,
      "balance_loss_mlp": 1.02434969,
      "epoch": 0.43817826544416055,
      "flos": 17493309198720.0,
      "grad_norm": 1.7396482701404024,
      "language_loss": 0.8220396,
      "learning_rate": 2.492421401510545e-06,
      "loss": 0.8434273,
      "num_input_tokens_seen": 156226560,
      "step": 7288,
      "time_per_iteration": 2.5081512928009033
    },
    {
      "auxiliary_loss_clip": 0.01085863,
      "auxiliary_loss_mlp": 0.01030177,
      "balance_loss_clip": 1.04114449,
      "balance_loss_mlp": 1.01656342,
      "epoch": 0.4382383886968285,
      "flos": 21581020759680.0,
      "grad_norm": 1.3528146914506565,
      "language_loss": 0.83936065,
      "learning_rate": 2.4920439195557093e-06,
      "loss": 0.86052108,
      "num_input_tokens_seen": 156246740,
      "step": 7289,
      "time_per_iteration": 2.578768730163574
    },
    {
      "auxiliary_loss_clip": 0.0109165,
      "auxiliary_loss_mlp": 0.01049747,
      "balance_loss_clip": 1.03985119,
      "balance_loss_mlp": 1.03379631,
      "epoch": 0.4382985119494965,
      "flos": 27923742201600.0,
      "grad_norm": 1.5559799418764293,
      "language_loss": 0.7812624,
      "learning_rate": 2.4916664189425183e-06,
      "loss": 0.80267644,
      "num_input_tokens_seen": 156266440,
      "step": 7290,
      "time_per_iteration": 2.595935583114624
    },
    {
      "auxiliary_loss_clip": 0.01122612,
      "auxiliary_loss_mlp": 0.0103598,
      "balance_loss_clip": 1.04517508,
      "balance_loss_mlp": 1.02328432,
      "epoch": 0.43835863520216445,
      "flos": 24936836797440.0,
      "grad_norm": 1.719626289184289,
      "language_loss": 0.78072757,
      "learning_rate": 2.491288899685288e-06,
      "loss": 0.80231345,
      "num_input_tokens_seen": 156286900,
      "step": 7291,
      "time_per_iteration": 2.5112600326538086
    },
    {
      "auxiliary_loss_clip": 0.01083989,
      "auxiliary_loss_mlp": 0.01032215,
      "balance_loss_clip": 1.03930771,
      "balance_loss_mlp": 1.0186131,
      "epoch": 0.4384187584548324,
      "flos": 33510293504640.0,
      "grad_norm": 1.6120568289254253,
      "language_loss": 0.64865696,
      "learning_rate": 2.4909113617983325e-06,
      "loss": 0.669819,
      "num_input_tokens_seen": 156307690,
      "step": 7292,
      "time_per_iteration": 2.6372756958007812
    },
    {
      "auxiliary_loss_clip": 0.01106036,
      "auxiliary_loss_mlp": 0.01040141,
      "balance_loss_clip": 1.04044747,
      "balance_loss_mlp": 1.02589548,
      "epoch": 0.4384788817075004,
      "flos": 23951052967680.0,
      "grad_norm": 1.6676056730363993,
      "language_loss": 0.74325472,
      "learning_rate": 2.49053380529597e-06,
      "loss": 0.76471657,
      "num_input_tokens_seen": 156326620,
      "step": 7293,
      "time_per_iteration": 2.521942615509033
    },
    {
      "auxiliary_loss_clip": 0.01093037,
      "auxiliary_loss_mlp": 0.01041417,
      "balance_loss_clip": 1.04162443,
      "balance_loss_mlp": 1.02697444,
      "epoch": 0.43853900496016834,
      "flos": 19098516090240.0,
      "grad_norm": 2.0123886627793897,
      "language_loss": 0.78312457,
      "learning_rate": 2.490156230192516e-06,
      "loss": 0.80446905,
      "num_input_tokens_seen": 156345495,
      "step": 7294,
      "time_per_iteration": 2.524009943008423
    },
    {
      "auxiliary_loss_clip": 0.01082955,
      "auxiliary_loss_mlp": 0.01046811,
      "balance_loss_clip": 1.04227614,
      "balance_loss_mlp": 1.03347743,
      "epoch": 0.4385991282128363,
      "flos": 13225362168960.0,
      "grad_norm": 1.634113811414013,
      "language_loss": 0.73092985,
      "learning_rate": 2.4897786365022883e-06,
      "loss": 0.75222754,
      "num_input_tokens_seen": 156363155,
      "step": 7295,
      "time_per_iteration": 2.560290813446045
    },
    {
      "auxiliary_loss_clip": 0.01084671,
      "auxiliary_loss_mlp": 0.01045887,
      "balance_loss_clip": 1.04488838,
      "balance_loss_mlp": 1.02947187,
      "epoch": 0.4386592514655043,
      "flos": 14319883445760.0,
      "grad_norm": 1.9323782473501545,
      "language_loss": 0.75108927,
      "learning_rate": 2.4894010242396063e-06,
      "loss": 0.7723949,
      "num_input_tokens_seen": 156380940,
      "step": 7296,
      "time_per_iteration": 4.475850343704224
    },
    {
      "auxiliary_loss_clip": 0.01109721,
      "auxiliary_loss_mlp": 0.01034392,
      "balance_loss_clip": 1.04168224,
      "balance_loss_mlp": 1.02099895,
      "epoch": 0.43871937471817224,
      "flos": 22784423137920.0,
      "grad_norm": 1.5040344378435904,
      "language_loss": 0.69268531,
      "learning_rate": 2.4890233934187873e-06,
      "loss": 0.71412647,
      "num_input_tokens_seen": 156400415,
      "step": 7297,
      "time_per_iteration": 3.9088637828826904
    },
    {
      "auxiliary_loss_clip": 0.0110303,
      "auxiliary_loss_mlp": 0.01030831,
      "balance_loss_clip": 1.0407443,
      "balance_loss_mlp": 1.01750314,
      "epoch": 0.4387794979708402,
      "flos": 28072304853120.0,
      "grad_norm": 1.612399057893791,
      "language_loss": 0.70181173,
      "learning_rate": 2.4886457440541535e-06,
      "loss": 0.72315037,
      "num_input_tokens_seen": 156421120,
      "step": 7298,
      "time_per_iteration": 2.5521249771118164
    },
    {
      "auxiliary_loss_clip": 0.01111534,
      "auxiliary_loss_mlp": 0.01030791,
      "balance_loss_clip": 1.04622579,
      "balance_loss_mlp": 1.01725507,
      "epoch": 0.43883962122350817,
      "flos": 26249551240320.0,
      "grad_norm": 2.9047359455227775,
      "language_loss": 0.72251672,
      "learning_rate": 2.4882680761600238e-06,
      "loss": 0.74394,
      "num_input_tokens_seen": 156441535,
      "step": 7299,
      "time_per_iteration": 2.542653799057007
    },
    {
      "auxiliary_loss_clip": 0.01098557,
      "auxiliary_loss_mlp": 0.00793016,
      "balance_loss_clip": 1.04198027,
      "balance_loss_mlp": 1.0152688,
      "epoch": 0.43889974447617613,
      "flos": 25883765089920.0,
      "grad_norm": 1.74336847452392,
      "language_loss": 0.76984966,
      "learning_rate": 2.487890389750719e-06,
      "loss": 0.78876543,
      "num_input_tokens_seen": 156462015,
      "step": 7300,
      "time_per_iteration": 2.5758426189422607
    },
    {
      "auxiliary_loss_clip": 0.01100555,
      "auxiliary_loss_mlp": 0.0103421,
      "balance_loss_clip": 1.04314923,
      "balance_loss_mlp": 1.02014923,
      "epoch": 0.43895986772884416,
      "flos": 25046615738880.0,
      "grad_norm": 1.7160029741199156,
      "language_loss": 0.70551604,
      "learning_rate": 2.4875126848405626e-06,
      "loss": 0.72686374,
      "num_input_tokens_seen": 156482165,
      "step": 7301,
      "time_per_iteration": 3.9254062175750732
    },
    {
      "auxiliary_loss_clip": 0.01076496,
      "auxiliary_loss_mlp": 0.01042117,
      "balance_loss_clip": 1.04203081,
      "balance_loss_mlp": 1.02551103,
      "epoch": 0.4390199909815121,
      "flos": 25994585525760.0,
      "grad_norm": 1.9129672109297557,
      "language_loss": 0.70615268,
      "learning_rate": 2.4871349614438757e-06,
      "loss": 0.72733879,
      "num_input_tokens_seen": 156503170,
      "step": 7302,
      "time_per_iteration": 2.633246660232544
    },
    {
      "auxiliary_loss_clip": 0.01099731,
      "auxiliary_loss_mlp": 0.01033938,
      "balance_loss_clip": 1.04430282,
      "balance_loss_mlp": 1.0210278,
      "epoch": 0.4390801142341801,
      "flos": 29022249888000.0,
      "grad_norm": 1.6276782129548113,
      "language_loss": 0.82104516,
      "learning_rate": 2.486757219574983e-06,
      "loss": 0.84238183,
      "num_input_tokens_seen": 156523005,
      "step": 7303,
      "time_per_iteration": 3.968296766281128
    },
    {
      "auxiliary_loss_clip": 0.01110485,
      "auxiliary_loss_mlp": 0.01040303,
      "balance_loss_clip": 1.04388142,
      "balance_loss_mlp": 1.02490163,
      "epoch": 0.43914023748684805,
      "flos": 33438544087680.0,
      "grad_norm": 1.832046484916962,
      "language_loss": 0.68183291,
      "learning_rate": 2.4863794592482067e-06,
      "loss": 0.70334077,
      "num_input_tokens_seen": 156544440,
      "step": 7304,
      "time_per_iteration": 2.58320951461792
    },
    {
      "auxiliary_loss_clip": 0.01101855,
      "auxiliary_loss_mlp": 0.00790715,
      "balance_loss_clip": 1.04354262,
      "balance_loss_mlp": 1.01553547,
      "epoch": 0.439200360739516,
      "flos": 34531844302080.0,
      "grad_norm": 1.562163311094618,
      "language_loss": 0.78185368,
      "learning_rate": 2.486001680477873e-06,
      "loss": 0.80077946,
      "num_input_tokens_seen": 156565410,
      "step": 7305,
      "time_per_iteration": 2.6304421424865723
    },
    {
      "auxiliary_loss_clip": 0.01098786,
      "auxiliary_loss_mlp": 0.01031546,
      "balance_loss_clip": 1.04474175,
      "balance_loss_mlp": 1.0177002,
      "epoch": 0.439260483992184,
      "flos": 21907843632000.0,
      "grad_norm": 1.9824592696198204,
      "language_loss": 0.68690014,
      "learning_rate": 2.485623883278308e-06,
      "loss": 0.70820343,
      "num_input_tokens_seen": 156584210,
      "step": 7306,
      "time_per_iteration": 2.5036439895629883
    },
    {
      "auxiliary_loss_clip": 0.01082146,
      "auxiliary_loss_mlp": 0.01028948,
      "balance_loss_clip": 1.04121089,
      "balance_loss_mlp": 1.01461899,
      "epoch": 0.43932060724485195,
      "flos": 20996430912000.0,
      "grad_norm": 1.5215665927938415,
      "language_loss": 0.62322366,
      "learning_rate": 2.4852460676638344e-06,
      "loss": 0.64433455,
      "num_input_tokens_seen": 156602730,
      "step": 7307,
      "time_per_iteration": 2.539475202560425
    },
    {
      "auxiliary_loss_clip": 0.01125624,
      "auxiliary_loss_mlp": 0.01033102,
      "balance_loss_clip": 1.04432118,
      "balance_loss_mlp": 1.01953042,
      "epoch": 0.4393807304975199,
      "flos": 17747053850880.0,
      "grad_norm": 2.0718164594003605,
      "language_loss": 0.71858948,
      "learning_rate": 2.4848682336487828e-06,
      "loss": 0.7401768,
      "num_input_tokens_seen": 156619405,
      "step": 7308,
      "time_per_iteration": 2.4014334678649902
    },
    {
      "auxiliary_loss_clip": 0.01106304,
      "auxiliary_loss_mlp": 0.01032865,
      "balance_loss_clip": 1.04377794,
      "balance_loss_mlp": 1.01843476,
      "epoch": 0.4394408537501879,
      "flos": 22528523669760.0,
      "grad_norm": 1.6903482489473085,
      "language_loss": 0.76779723,
      "learning_rate": 2.4844903812474787e-06,
      "loss": 0.78918892,
      "num_input_tokens_seen": 156638165,
      "step": 7309,
      "time_per_iteration": 2.52130126953125
    },
    {
      "auxiliary_loss_clip": 0.01107136,
      "auxiliary_loss_mlp": 0.01029239,
      "balance_loss_clip": 1.04275966,
      "balance_loss_mlp": 1.01618564,
      "epoch": 0.43950097700285584,
      "flos": 23440654661760.0,
      "grad_norm": 1.7338094526054357,
      "language_loss": 0.70763177,
      "learning_rate": 2.484112510474251e-06,
      "loss": 0.72899556,
      "num_input_tokens_seen": 156658845,
      "step": 7310,
      "time_per_iteration": 2.500941038131714
    },
    {
      "auxiliary_loss_clip": 0.01097353,
      "auxiliary_loss_mlp": 0.00791813,
      "balance_loss_clip": 1.04431343,
      "balance_loss_mlp": 1.01585042,
      "epoch": 0.4395611002555238,
      "flos": 23180696956800.0,
      "grad_norm": 3.1050892008620625,
      "language_loss": 0.76220489,
      "learning_rate": 2.483734621343429e-06,
      "loss": 0.78109652,
      "num_input_tokens_seen": 156677275,
      "step": 7311,
      "time_per_iteration": 2.5199697017669678
    },
    {
      "auxiliary_loss_clip": 0.01113482,
      "auxiliary_loss_mlp": 0.01033365,
      "balance_loss_clip": 1.04403877,
      "balance_loss_mlp": 1.02007318,
      "epoch": 0.43962122350819177,
      "flos": 22127365601280.0,
      "grad_norm": 2.750716085253334,
      "language_loss": 0.81371069,
      "learning_rate": 2.483356713869341e-06,
      "loss": 0.83517921,
      "num_input_tokens_seen": 156695815,
      "step": 7312,
      "time_per_iteration": 2.481703281402588
    },
    {
      "auxiliary_loss_clip": 0.01089997,
      "auxiliary_loss_mlp": 0.01030484,
      "balance_loss_clip": 1.04064202,
      "balance_loss_mlp": 1.01725221,
      "epoch": 0.43968134676085974,
      "flos": 17420554200960.0,
      "grad_norm": 3.5717380847507654,
      "language_loss": 0.85412443,
      "learning_rate": 2.482978788066318e-06,
      "loss": 0.87532926,
      "num_input_tokens_seen": 156714385,
      "step": 7313,
      "time_per_iteration": 2.5297653675079346
    },
    {
      "auxiliary_loss_clip": 0.01099571,
      "auxiliary_loss_mlp": 0.010295,
      "balance_loss_clip": 1.04185343,
      "balance_loss_mlp": 1.01567745,
      "epoch": 0.43974147001352776,
      "flos": 18952646958720.0,
      "grad_norm": 1.8014453103123205,
      "language_loss": 0.67811906,
      "learning_rate": 2.4826008439486904e-06,
      "loss": 0.69940972,
      "num_input_tokens_seen": 156732615,
      "step": 7314,
      "time_per_iteration": 2.5030927658081055
    },
    {
      "auxiliary_loss_clip": 0.01104159,
      "auxiliary_loss_mlp": 0.01030509,
      "balance_loss_clip": 1.04398704,
      "balance_loss_mlp": 1.01644814,
      "epoch": 0.4398015932661957,
      "flos": 18953508885120.0,
      "grad_norm": 2.3348147639444816,
      "language_loss": 0.76656681,
      "learning_rate": 2.4822228815307915e-06,
      "loss": 0.7879135,
      "num_input_tokens_seen": 156750920,
      "step": 7315,
      "time_per_iteration": 2.5152525901794434
    },
    {
      "auxiliary_loss_clip": 0.0109834,
      "auxiliary_loss_mlp": 0.01032692,
      "balance_loss_clip": 1.04424989,
      "balance_loss_mlp": 1.01914954,
      "epoch": 0.4398617165188637,
      "flos": 24199913370240.0,
      "grad_norm": 2.086122811746675,
      "language_loss": 0.74020731,
      "learning_rate": 2.4818449008269523e-06,
      "loss": 0.76151764,
      "num_input_tokens_seen": 156768520,
      "step": 7316,
      "time_per_iteration": 2.5329952239990234
    },
    {
      "auxiliary_loss_clip": 0.01087393,
      "auxiliary_loss_mlp": 0.01035921,
      "balance_loss_clip": 1.04759812,
      "balance_loss_mlp": 1.02284336,
      "epoch": 0.43992183977153165,
      "flos": 22236677665920.0,
      "grad_norm": 2.6561423476946335,
      "language_loss": 0.64965689,
      "learning_rate": 2.481466901851506e-06,
      "loss": 0.67089009,
      "num_input_tokens_seen": 156788700,
      "step": 7317,
      "time_per_iteration": 2.5947585105895996
    },
    {
      "auxiliary_loss_clip": 0.01097145,
      "auxiliary_loss_mlp": 0.01037867,
      "balance_loss_clip": 1.04348707,
      "balance_loss_mlp": 1.02453303,
      "epoch": 0.4399819630241996,
      "flos": 18697465762560.0,
      "grad_norm": 2.0040467979956107,
      "language_loss": 0.7995286,
      "learning_rate": 2.4810888846187865e-06,
      "loss": 0.82087874,
      "num_input_tokens_seen": 156806470,
      "step": 7318,
      "time_per_iteration": 2.4920053482055664
    },
    {
      "auxiliary_loss_clip": 0.01081911,
      "auxiliary_loss_mlp": 0.01040482,
      "balance_loss_clip": 1.04068971,
      "balance_loss_mlp": 1.02555692,
      "epoch": 0.4400420862768676,
      "flos": 23879375377920.0,
      "grad_norm": 1.7542271556516045,
      "language_loss": 0.79349953,
      "learning_rate": 2.4807108491431283e-06,
      "loss": 0.81472343,
      "num_input_tokens_seen": 156825895,
      "step": 7319,
      "time_per_iteration": 2.568981170654297
    },
    {
      "auxiliary_loss_clip": 0.011073,
      "auxiliary_loss_mlp": 0.01038426,
      "balance_loss_clip": 1.04273808,
      "balance_loss_mlp": 1.02455008,
      "epoch": 0.44010220952953555,
      "flos": 28037615293440.0,
      "grad_norm": 2.2495257086929734,
      "language_loss": 0.79824746,
      "learning_rate": 2.4803327954388667e-06,
      "loss": 0.81970465,
      "num_input_tokens_seen": 156845990,
      "step": 7320,
      "time_per_iteration": 2.5480270385742188
    },
    {
      "auxiliary_loss_clip": 0.01084848,
      "auxiliary_loss_mlp": 0.01042859,
      "balance_loss_clip": 1.04328895,
      "balance_loss_mlp": 1.02965653,
      "epoch": 0.4401623327822035,
      "flos": 23768985905280.0,
      "grad_norm": 1.7401219076481353,
      "language_loss": 0.69371641,
      "learning_rate": 2.4799547235203376e-06,
      "loss": 0.71499348,
      "num_input_tokens_seen": 156866685,
      "step": 7321,
      "time_per_iteration": 2.5548219680786133
    },
    {
      "auxiliary_loss_clip": 0.01014886,
      "auxiliary_loss_mlp": 0.01018539,
      "balance_loss_clip": 1.02822495,
      "balance_loss_mlp": 1.01687038,
      "epoch": 0.4402224560348715,
      "flos": 70774583264640.0,
      "grad_norm": 0.883113647100991,
      "language_loss": 0.56907403,
      "learning_rate": 2.4795766334018763e-06,
      "loss": 0.58940828,
      "num_input_tokens_seen": 156923450,
      "step": 7322,
      "time_per_iteration": 3.324855089187622
    },
    {
      "auxiliary_loss_clip": 0.0107117,
      "auxiliary_loss_mlp": 0.01040506,
      "balance_loss_clip": 1.03775072,
      "balance_loss_mlp": 1.02678514,
      "epoch": 0.44028257928753944,
      "flos": 22891795868160.0,
      "grad_norm": 1.4123092128131713,
      "language_loss": 0.76160324,
      "learning_rate": 2.479198525097822e-06,
      "loss": 0.78272003,
      "num_input_tokens_seen": 156944795,
      "step": 7323,
      "time_per_iteration": 2.5896964073181152
    },
    {
      "auxiliary_loss_clip": 0.01116577,
      "auxiliary_loss_mlp": 0.01042789,
      "balance_loss_clip": 1.045555,
      "balance_loss_mlp": 1.02899647,
      "epoch": 0.4403427025402074,
      "flos": 17895760156800.0,
      "grad_norm": 1.5210203307415013,
      "language_loss": 0.80623448,
      "learning_rate": 2.478820398622511e-06,
      "loss": 0.82782817,
      "num_input_tokens_seen": 156962755,
      "step": 7324,
      "time_per_iteration": 2.4788978099823
    },
    {
      "auxiliary_loss_clip": 0.01023382,
      "auxiliary_loss_mlp": 0.01005248,
      "balance_loss_clip": 1.02391076,
      "balance_loss_mlp": 1.00338793,
      "epoch": 0.4404028257928754,
      "flos": 69562525708800.0,
      "grad_norm": 0.663576768204068,
      "language_loss": 0.54558647,
      "learning_rate": 2.478442253990283e-06,
      "loss": 0.56587279,
      "num_input_tokens_seen": 157028095,
      "step": 7325,
      "time_per_iteration": 3.1552305221557617
    },
    {
      "auxiliary_loss_clip": 0.01121343,
      "auxiliary_loss_mlp": 0.01028078,
      "balance_loss_clip": 1.04523802,
      "balance_loss_mlp": 1.01594245,
      "epoch": 0.44046294904554334,
      "flos": 20923675914240.0,
      "grad_norm": 1.3848740969027262,
      "language_loss": 0.69718587,
      "learning_rate": 2.4780640912154766e-06,
      "loss": 0.71868008,
      "num_input_tokens_seen": 157048365,
      "step": 7326,
      "time_per_iteration": 2.495586633682251
    },
    {
      "auxiliary_loss_clip": 0.01081628,
      "auxiliary_loss_mlp": 0.0103075,
      "balance_loss_clip": 1.04111886,
      "balance_loss_mlp": 1.01750553,
      "epoch": 0.44052307229821136,
      "flos": 23623475909760.0,
      "grad_norm": 1.6772765867748605,
      "language_loss": 0.76465464,
      "learning_rate": 2.477685910312432e-06,
      "loss": 0.7857784,
      "num_input_tokens_seen": 157069130,
      "step": 7327,
      "time_per_iteration": 2.584575653076172
    },
    {
      "auxiliary_loss_clip": 0.0109579,
      "auxiliary_loss_mlp": 0.01032777,
      "balance_loss_clip": 1.04126573,
      "balance_loss_mlp": 1.01946712,
      "epoch": 0.4405831955508793,
      "flos": 17597665186560.0,
      "grad_norm": 1.8962918938933897,
      "language_loss": 0.8415978,
      "learning_rate": 2.4773077112954897e-06,
      "loss": 0.86288345,
      "num_input_tokens_seen": 157084940,
      "step": 7328,
      "time_per_iteration": 2.489612579345703
    },
    {
      "auxiliary_loss_clip": 0.01097188,
      "auxiliary_loss_mlp": 0.01029125,
      "balance_loss_clip": 1.0418303,
      "balance_loss_mlp": 1.01605964,
      "epoch": 0.4406433188035473,
      "flos": 21463376739840.0,
      "grad_norm": 2.4240167781141895,
      "language_loss": 0.77349484,
      "learning_rate": 2.4769294941789908e-06,
      "loss": 0.79475796,
      "num_input_tokens_seen": 157102770,
      "step": 7329,
      "time_per_iteration": 2.5101242065429688
    },
    {
      "auxiliary_loss_clip": 0.01109307,
      "auxiliary_loss_mlp": 0.01038674,
      "balance_loss_clip": 1.04293323,
      "balance_loss_mlp": 1.02445197,
      "epoch": 0.44070344205621526,
      "flos": 22673566788480.0,
      "grad_norm": 1.5900670990498873,
      "language_loss": 0.73263991,
      "learning_rate": 2.476551258977278e-06,
      "loss": 0.75411963,
      "num_input_tokens_seen": 157122035,
      "step": 7330,
      "time_per_iteration": 2.503915309906006
    },
    {
      "auxiliary_loss_clip": 0.01103224,
      "auxiliary_loss_mlp": 0.01039787,
      "balance_loss_clip": 1.04657865,
      "balance_loss_mlp": 1.02726364,
      "epoch": 0.4407635653088832,
      "flos": 23441193365760.0,
      "grad_norm": 1.9775166347986528,
      "language_loss": 0.74469554,
      "learning_rate": 2.4761730057046936e-06,
      "loss": 0.76612568,
      "num_input_tokens_seen": 157142800,
      "step": 7331,
      "time_per_iteration": 2.5767111778259277
    },
    {
      "auxiliary_loss_clip": 0.01069181,
      "auxiliary_loss_mlp": 0.01030987,
      "balance_loss_clip": 1.04067159,
      "balance_loss_mlp": 1.01786768,
      "epoch": 0.4408236885615512,
      "flos": 24021294013440.0,
      "grad_norm": 1.4761327043926402,
      "language_loss": 0.76282358,
      "learning_rate": 2.475794734375581e-06,
      "loss": 0.78382528,
      "num_input_tokens_seen": 157163295,
      "step": 7332,
      "time_per_iteration": 2.611268997192383
    },
    {
      "auxiliary_loss_clip": 0.01094951,
      "auxiliary_loss_mlp": 0.01037985,
      "balance_loss_clip": 1.04512632,
      "balance_loss_mlp": 1.02548027,
      "epoch": 0.44088381181421915,
      "flos": 12676826597760.0,
      "grad_norm": 1.7762109728756612,
      "language_loss": 0.73912895,
      "learning_rate": 2.475416445004285e-06,
      "loss": 0.76045829,
      "num_input_tokens_seen": 157180890,
      "step": 7333,
      "time_per_iteration": 2.4792332649230957
    },
    {
      "auxiliary_loss_clip": 0.01084746,
      "auxiliary_loss_mlp": 0.01032567,
      "balance_loss_clip": 1.04504991,
      "balance_loss_mlp": 1.01950216,
      "epoch": 0.4409439350668871,
      "flos": 24569865498240.0,
      "grad_norm": 1.6294356654849593,
      "language_loss": 0.79731101,
      "learning_rate": 2.4750381376051493e-06,
      "loss": 0.81848413,
      "num_input_tokens_seen": 157200580,
      "step": 7334,
      "time_per_iteration": 3.951969861984253
    },
    {
      "auxiliary_loss_clip": 0.01094874,
      "auxiliary_loss_mlp": 0.01038546,
      "balance_loss_clip": 1.0414257,
      "balance_loss_mlp": 1.02246523,
      "epoch": 0.4410040583195551,
      "flos": 22668574798080.0,
      "grad_norm": 2.0672676216740014,
      "language_loss": 0.75795656,
      "learning_rate": 2.47465981219252e-06,
      "loss": 0.77929074,
      "num_input_tokens_seen": 157218345,
      "step": 7335,
      "time_per_iteration": 4.149829626083374
    },
    {
      "auxiliary_loss_clip": 0.01105685,
      "auxiliary_loss_mlp": 0.01037094,
      "balance_loss_clip": 1.04501927,
      "balance_loss_mlp": 1.02372479,
      "epoch": 0.44106418157222305,
      "flos": 10852528700160.0,
      "grad_norm": 1.879587574450716,
      "language_loss": 0.72824997,
      "learning_rate": 2.4742814687807423e-06,
      "loss": 0.74967778,
      "num_input_tokens_seen": 157234395,
      "step": 7336,
      "time_per_iteration": 2.513986110687256
    },
    {
      "auxiliary_loss_clip": 0.01113739,
      "auxiliary_loss_mlp": 0.01042333,
      "balance_loss_clip": 1.04164624,
      "balance_loss_mlp": 1.02847493,
      "epoch": 0.441124304824891,
      "flos": 21726710323200.0,
      "grad_norm": 2.0406308707030254,
      "language_loss": 0.63156742,
      "learning_rate": 2.473903107384165e-06,
      "loss": 0.65312815,
      "num_input_tokens_seen": 157254805,
      "step": 7337,
      "time_per_iteration": 2.5007779598236084
    },
    {
      "auxiliary_loss_clip": 0.01029244,
      "auxiliary_loss_mlp": 0.00771692,
      "balance_loss_clip": 1.020679,
      "balance_loss_mlp": 1.0097698,
      "epoch": 0.441184428077559,
      "flos": 63220486625280.0,
      "grad_norm": 0.7489733799119582,
      "language_loss": 0.52717757,
      "learning_rate": 2.473524728017134e-06,
      "loss": 0.54518694,
      "num_input_tokens_seen": 157317870,
      "step": 7338,
      "time_per_iteration": 3.194927215576172
    },
    {
      "auxiliary_loss_clip": 0.01102979,
      "auxiliary_loss_mlp": 0.01042682,
      "balance_loss_clip": 1.04181957,
      "balance_loss_mlp": 1.02750707,
      "epoch": 0.44124455133022694,
      "flos": 21177959270400.0,
      "grad_norm": 1.9488433477430696,
      "language_loss": 0.70537174,
      "learning_rate": 2.473146330693997e-06,
      "loss": 0.7268284,
      "num_input_tokens_seen": 157336505,
      "step": 7339,
      "time_per_iteration": 3.983253240585327
    },
    {
      "auxiliary_loss_clip": 0.01054717,
      "auxiliary_loss_mlp": 0.0104524,
      "balance_loss_clip": 1.04300976,
      "balance_loss_mlp": 1.03150153,
      "epoch": 0.4413046745828949,
      "flos": 17457865453440.0,
      "grad_norm": 1.3853448061024611,
      "language_loss": 0.69350052,
      "learning_rate": 2.472767915429105e-06,
      "loss": 0.71450013,
      "num_input_tokens_seen": 157354995,
      "step": 7340,
      "time_per_iteration": 2.5850021839141846
    },
    {
      "auxiliary_loss_clip": 0.01027026,
      "auxiliary_loss_mlp": 0.01007957,
      "balance_loss_clip": 1.02148509,
      "balance_loss_mlp": 1.00637174,
      "epoch": 0.4413647978355629,
      "flos": 61586153804160.0,
      "grad_norm": 0.903915288173658,
      "language_loss": 0.6401903,
      "learning_rate": 2.4723894822368054e-06,
      "loss": 0.6605401,
      "num_input_tokens_seen": 157404260,
      "step": 7341,
      "time_per_iteration": 4.26634669303894
    },
    {
      "auxiliary_loss_clip": 0.01085196,
      "auxiliary_loss_mlp": 0.01035646,
      "balance_loss_clip": 1.0395925,
      "balance_loss_mlp": 1.02185905,
      "epoch": 0.4414249210882309,
      "flos": 27527001505920.0,
      "grad_norm": 1.987478911152901,
      "language_loss": 0.73558474,
      "learning_rate": 2.47201103113145e-06,
      "loss": 0.75679308,
      "num_input_tokens_seen": 157423045,
      "step": 7342,
      "time_per_iteration": 2.596327543258667
    },
    {
      "auxiliary_loss_clip": 0.01118343,
      "auxiliary_loss_mlp": 0.0103652,
      "balance_loss_clip": 1.0397495,
      "balance_loss_mlp": 1.02245355,
      "epoch": 0.44148504434089886,
      "flos": 23513984277120.0,
      "grad_norm": 1.6995392946371923,
      "language_loss": 0.79435456,
      "learning_rate": 2.4716325621273886e-06,
      "loss": 0.81590319,
      "num_input_tokens_seen": 157441815,
      "step": 7343,
      "time_per_iteration": 2.4992613792419434
    },
    {
      "auxiliary_loss_clip": 0.01084078,
      "auxiliary_loss_mlp": 0.01032114,
      "balance_loss_clip": 1.04259515,
      "balance_loss_mlp": 1.0183332,
      "epoch": 0.4415451675935668,
      "flos": 21580589796480.0,
      "grad_norm": 1.7548972450543137,
      "language_loss": 0.76559412,
      "learning_rate": 2.4712540752389725e-06,
      "loss": 0.78675604,
      "num_input_tokens_seen": 157460470,
      "step": 7344,
      "time_per_iteration": 2.5495591163635254
    },
    {
      "auxiliary_loss_clip": 0.01026551,
      "auxiliary_loss_mlp": 0.01014582,
      "balance_loss_clip": 1.01919246,
      "balance_loss_mlp": 1.0128535,
      "epoch": 0.4416052908462348,
      "flos": 59006368126080.0,
      "grad_norm": 0.7910025494403222,
      "language_loss": 0.63803893,
      "learning_rate": 2.470875570480556e-06,
      "loss": 0.65845031,
      "num_input_tokens_seen": 157512655,
      "step": 7345,
      "time_per_iteration": 2.879918336868286
    },
    {
      "auxiliary_loss_clip": 0.01123826,
      "auxiliary_loss_mlp": 0.01032793,
      "balance_loss_clip": 1.04340053,
      "balance_loss_mlp": 1.01925039,
      "epoch": 0.44166541409890275,
      "flos": 26357642242560.0,
      "grad_norm": 1.9458304046262362,
      "language_loss": 0.85892463,
      "learning_rate": 2.470497047866489e-06,
      "loss": 0.88049078,
      "num_input_tokens_seen": 157533700,
      "step": 7346,
      "time_per_iteration": 2.512547016143799
    },
    {
      "auxiliary_loss_clip": 0.01111426,
      "auxiliary_loss_mlp": 0.01038301,
      "balance_loss_clip": 1.04135537,
      "balance_loss_mlp": 1.02380466,
      "epoch": 0.4417255373515707,
      "flos": 20192678231040.0,
      "grad_norm": 1.7197144635998982,
      "language_loss": 0.80781126,
      "learning_rate": 2.470118507411128e-06,
      "loss": 0.82930857,
      "num_input_tokens_seen": 157551105,
      "step": 7347,
      "time_per_iteration": 2.4806082248687744
    },
    {
      "auxiliary_loss_clip": 0.01096949,
      "auxiliary_loss_mlp": 0.0103153,
      "balance_loss_clip": 1.04523873,
      "balance_loss_mlp": 1.01729655,
      "epoch": 0.4417856606042387,
      "flos": 17887895078400.0,
      "grad_norm": 1.6675034864414466,
      "language_loss": 0.82817107,
      "learning_rate": 2.4697399491288263e-06,
      "loss": 0.84945583,
      "num_input_tokens_seen": 157568285,
      "step": 7348,
      "time_per_iteration": 2.493452548980713
    },
    {
      "auxiliary_loss_clip": 0.01113347,
      "auxiliary_loss_mlp": 0.01036483,
      "balance_loss_clip": 1.04406619,
      "balance_loss_mlp": 1.02273178,
      "epoch": 0.44184578385690665,
      "flos": 27964034282880.0,
      "grad_norm": 1.516526505188605,
      "language_loss": 0.69955873,
      "learning_rate": 2.469361373033938e-06,
      "loss": 0.72105706,
      "num_input_tokens_seen": 157590405,
      "step": 7349,
      "time_per_iteration": 2.5466208457946777
    },
    {
      "auxiliary_loss_clip": 0.01099646,
      "auxiliary_loss_mlp": 0.01035143,
      "balance_loss_clip": 1.0402987,
      "balance_loss_mlp": 1.02074218,
      "epoch": 0.4419059071095746,
      "flos": 23367899664000.0,
      "grad_norm": 1.7730075595979071,
      "language_loss": 0.7432934,
      "learning_rate": 2.468982779140819e-06,
      "loss": 0.76464134,
      "num_input_tokens_seen": 157607420,
      "step": 7350,
      "time_per_iteration": 2.5074541568756104
    },
    {
      "auxiliary_loss_clip": 0.01122072,
      "auxiliary_loss_mlp": 0.0103378,
      "balance_loss_clip": 1.04226518,
      "balance_loss_mlp": 1.02001178,
      "epoch": 0.4419660303622426,
      "flos": 15012169246080.0,
      "grad_norm": 2.1891389855363657,
      "language_loss": 0.81140149,
      "learning_rate": 2.468604167463827e-06,
      "loss": 0.83296001,
      "num_input_tokens_seen": 157624990,
      "step": 7351,
      "time_per_iteration": 2.4371485710144043
    },
    {
      "auxiliary_loss_clip": 0.01080397,
      "auxiliary_loss_mlp": 0.00794901,
      "balance_loss_clip": 1.03796649,
      "balance_loss_mlp": 1.02464104,
      "epoch": 0.44202615361491054,
      "flos": 25371750672000.0,
      "grad_norm": 1.4691185488904712,
      "language_loss": 0.7304672,
      "learning_rate": 2.4682255380173176e-06,
      "loss": 0.74922013,
      "num_input_tokens_seen": 157645300,
      "step": 7352,
      "time_per_iteration": 2.598890542984009
    },
    {
      "auxiliary_loss_clip": 0.01094131,
      "auxiliary_loss_mlp": 0.01027219,
      "balance_loss_clip": 1.04283381,
      "balance_loss_mlp": 1.01365304,
      "epoch": 0.4420862768675785,
      "flos": 24681116897280.0,
      "grad_norm": 2.027823749704245,
      "language_loss": 0.87232047,
      "learning_rate": 2.467846890815649e-06,
      "loss": 0.893534,
      "num_input_tokens_seen": 157664060,
      "step": 7353,
      "time_per_iteration": 2.5594730377197266
    },
    {
      "auxiliary_loss_clip": 0.01122639,
      "auxiliary_loss_mlp": 0.0103584,
      "balance_loss_clip": 1.04311347,
      "balance_loss_mlp": 1.02297163,
      "epoch": 0.44214640012024653,
      "flos": 19528437974400.0,
      "grad_norm": 1.8691485799444854,
      "language_loss": 0.76387608,
      "learning_rate": 2.4674682258731795e-06,
      "loss": 0.78546089,
      "num_input_tokens_seen": 157680905,
      "step": 7354,
      "time_per_iteration": 2.4346916675567627
    },
    {
      "auxiliary_loss_clip": 0.01081284,
      "auxiliary_loss_mlp": 0.01031602,
      "balance_loss_clip": 1.04235959,
      "balance_loss_mlp": 1.01915681,
      "epoch": 0.4422065233729145,
      "flos": 47557434003840.0,
      "grad_norm": 2.7448450973547067,
      "language_loss": 0.6451782,
      "learning_rate": 2.467089543204268e-06,
      "loss": 0.66630709,
      "num_input_tokens_seen": 157701980,
      "step": 7355,
      "time_per_iteration": 2.7877674102783203
    },
    {
      "auxiliary_loss_clip": 0.01123837,
      "auxiliary_loss_mlp": 0.01037196,
      "balance_loss_clip": 1.04200029,
      "balance_loss_mlp": 1.02293825,
      "epoch": 0.44226664662558246,
      "flos": 19281050029440.0,
      "grad_norm": 1.7989927853345915,
      "language_loss": 0.77945948,
      "learning_rate": 2.466710842823274e-06,
      "loss": 0.80106974,
      "num_input_tokens_seen": 157720555,
      "step": 7356,
      "time_per_iteration": 2.4474599361419678
    },
    {
      "auxiliary_loss_clip": 0.0110226,
      "auxiliary_loss_mlp": 0.00788608,
      "balance_loss_clip": 1.04463077,
      "balance_loss_mlp": 1.01282287,
      "epoch": 0.4423267698782504,
      "flos": 17821820010240.0,
      "grad_norm": 1.5763897914923495,
      "language_loss": 0.7725215,
      "learning_rate": 2.4663321247445577e-06,
      "loss": 0.79143018,
      "num_input_tokens_seen": 157739160,
      "step": 7357,
      "time_per_iteration": 2.499589443206787
    },
    {
      "auxiliary_loss_clip": 0.0109905,
      "auxiliary_loss_mlp": 0.01038792,
      "balance_loss_clip": 1.04408932,
      "balance_loss_mlp": 1.02424264,
      "epoch": 0.4423868931309184,
      "flos": 29204424691200.0,
      "grad_norm": 1.6799675589712149,
      "language_loss": 0.73468256,
      "learning_rate": 2.465953388982481e-06,
      "loss": 0.75606096,
      "num_input_tokens_seen": 157760020,
      "step": 7358,
      "time_per_iteration": 2.605456829071045
    },
    {
      "auxiliary_loss_clip": 0.01099128,
      "auxiliary_loss_mlp": 0.01032079,
      "balance_loss_clip": 1.04754293,
      "balance_loss_mlp": 1.01899552,
      "epoch": 0.44244701638358636,
      "flos": 29713135057920.0,
      "grad_norm": 1.701132461754456,
      "language_loss": 0.75652349,
      "learning_rate": 2.465574635551405e-06,
      "loss": 0.77783561,
      "num_input_tokens_seen": 157780435,
      "step": 7359,
      "time_per_iteration": 2.58886981010437
    },
    {
      "auxiliary_loss_clip": 0.01097926,
      "auxiliary_loss_mlp": 0.01031956,
      "balance_loss_clip": 1.04530144,
      "balance_loss_mlp": 1.01766336,
      "epoch": 0.4425071396362543,
      "flos": 22930040874240.0,
      "grad_norm": 1.650021209983342,
      "language_loss": 0.70074397,
      "learning_rate": 2.4651958644656923e-06,
      "loss": 0.7220428,
      "num_input_tokens_seen": 157799420,
      "step": 7360,
      "time_per_iteration": 2.5300168991088867
    },
    {
      "auxiliary_loss_clip": 0.01095398,
      "auxiliary_loss_mlp": 0.01030924,
      "balance_loss_clip": 1.04298425,
      "balance_loss_mlp": 1.01715541,
      "epoch": 0.4425672628889223,
      "flos": 19792346175360.0,
      "grad_norm": 2.295600326958118,
      "language_loss": 0.70199239,
      "learning_rate": 2.4648170757397053e-06,
      "loss": 0.72325552,
      "num_input_tokens_seen": 157817025,
      "step": 7361,
      "time_per_iteration": 2.5028905868530273
    },
    {
      "auxiliary_loss_clip": 0.01098743,
      "auxiliary_loss_mlp": 0.0103586,
      "balance_loss_clip": 1.0415225,
      "balance_loss_mlp": 1.02126312,
      "epoch": 0.44262738614159025,
      "flos": 13662215377920.0,
      "grad_norm": 1.9013795098914195,
      "language_loss": 0.8290031,
      "learning_rate": 2.464438269387809e-06,
      "loss": 0.85034913,
      "num_input_tokens_seen": 157834345,
      "step": 7362,
      "time_per_iteration": 2.49328875541687
    },
    {
      "auxiliary_loss_clip": 0.01096704,
      "auxiliary_loss_mlp": 0.01039972,
      "balance_loss_clip": 1.04851174,
      "balance_loss_mlp": 1.02478456,
      "epoch": 0.4426875093942582,
      "flos": 14210212245120.0,
      "grad_norm": 1.6067379598975573,
      "language_loss": 0.74538147,
      "learning_rate": 2.464059445424366e-06,
      "loss": 0.76674819,
      "num_input_tokens_seen": 157852290,
      "step": 7363,
      "time_per_iteration": 2.523386240005493
    },
    {
      "auxiliary_loss_clip": 0.01013358,
      "auxiliary_loss_mlp": 0.01007973,
      "balance_loss_clip": 1.03492022,
      "balance_loss_mlp": 1.00628042,
      "epoch": 0.4427476326469262,
      "flos": 70117525728000.0,
      "grad_norm": 0.7180701085624108,
      "language_loss": 0.55647713,
      "learning_rate": 2.463680603863743e-06,
      "loss": 0.57669044,
      "num_input_tokens_seen": 157923060,
      "step": 7364,
      "time_per_iteration": 3.2705013751983643
    },
    {
      "auxiliary_loss_clip": 0.01098958,
      "auxiliary_loss_mlp": 0.01035673,
      "balance_loss_clip": 1.04047644,
      "balance_loss_mlp": 1.02224994,
      "epoch": 0.44280775589959415,
      "flos": 25445080287360.0,
      "grad_norm": 1.6661870989897638,
      "language_loss": 0.74262297,
      "learning_rate": 2.463301744720305e-06,
      "loss": 0.76396936,
      "num_input_tokens_seen": 157944110,
      "step": 7365,
      "time_per_iteration": 2.543498992919922
    },
    {
      "auxiliary_loss_clip": 0.01091738,
      "auxiliary_loss_mlp": 0.0103745,
      "balance_loss_clip": 1.04053426,
      "balance_loss_mlp": 1.02364588,
      "epoch": 0.4428678791522621,
      "flos": 22857214049280.0,
      "grad_norm": 1.5999579630264356,
      "language_loss": 0.74163628,
      "learning_rate": 2.4629228680084184e-06,
      "loss": 0.76292813,
      "num_input_tokens_seen": 157964295,
      "step": 7366,
      "time_per_iteration": 2.5248794555664062
    },
    {
      "auxiliary_loss_clip": 0.01101432,
      "auxiliary_loss_mlp": 0.010357,
      "balance_loss_clip": 1.0447402,
      "balance_loss_mlp": 1.02165687,
      "epoch": 0.44292800240493013,
      "flos": 25812446636160.0,
      "grad_norm": 2.343299600941484,
      "language_loss": 0.73612726,
      "learning_rate": 2.46254397374245e-06,
      "loss": 0.7574985,
      "num_input_tokens_seen": 157983970,
      "step": 7367,
      "time_per_iteration": 2.5513699054718018
    },
    {
      "auxiliary_loss_clip": 0.01120895,
      "auxiliary_loss_mlp": 0.01036631,
      "balance_loss_clip": 1.04259396,
      "balance_loss_mlp": 1.0236311,
      "epoch": 0.4429881256575981,
      "flos": 32416885549440.0,
      "grad_norm": 1.3953483112588263,
      "language_loss": 0.73822284,
      "learning_rate": 2.4621650619367677e-06,
      "loss": 0.75979805,
      "num_input_tokens_seen": 158006515,
      "step": 7368,
      "time_per_iteration": 2.545046806335449
    },
    {
      "auxiliary_loss_clip": 0.01092595,
      "auxiliary_loss_mlp": 0.01034556,
      "balance_loss_clip": 1.03986049,
      "balance_loss_mlp": 1.02146721,
      "epoch": 0.44304824891026606,
      "flos": 22163707186560.0,
      "grad_norm": 4.349084119073061,
      "language_loss": 0.79945552,
      "learning_rate": 2.4617861326057403e-06,
      "loss": 0.82072699,
      "num_input_tokens_seen": 158025565,
      "step": 7369,
      "time_per_iteration": 2.5254852771759033
    },
    {
      "auxiliary_loss_clip": 0.01083584,
      "auxiliary_loss_mlp": 0.01032937,
      "balance_loss_clip": 1.04050994,
      "balance_loss_mlp": 1.0200628,
      "epoch": 0.443108372162934,
      "flos": 25338569483520.0,
      "grad_norm": 1.9071217829039973,
      "language_loss": 0.72184193,
      "learning_rate": 2.461407185763737e-06,
      "loss": 0.74300712,
      "num_input_tokens_seen": 158045620,
      "step": 7370,
      "time_per_iteration": 2.589053153991699
    },
    {
      "auxiliary_loss_clip": 0.01119854,
      "auxiliary_loss_mlp": 0.01032076,
      "balance_loss_clip": 1.04219234,
      "balance_loss_mlp": 1.01880765,
      "epoch": 0.443168495415602,
      "flos": 23330947547520.0,
      "grad_norm": 1.9595335993761063,
      "language_loss": 0.70515865,
      "learning_rate": 2.461028221425126e-06,
      "loss": 0.72667789,
      "num_input_tokens_seen": 158063505,
      "step": 7371,
      "time_per_iteration": 2.456566572189331
    },
    {
      "auxiliary_loss_clip": 0.01107272,
      "auxiliary_loss_mlp": 0.01030562,
      "balance_loss_clip": 1.04057467,
      "balance_loss_mlp": 1.01831877,
      "epoch": 0.44322861866826996,
      "flos": 21871502046720.0,
      "grad_norm": 2.149548079615021,
      "language_loss": 0.68605781,
      "learning_rate": 2.4606492396042786e-06,
      "loss": 0.70743614,
      "num_input_tokens_seen": 158080335,
      "step": 7372,
      "time_per_iteration": 2.503039598464966
    },
    {
      "auxiliary_loss_clip": 0.01087745,
      "auxiliary_loss_mlp": 0.01032373,
      "balance_loss_clip": 1.03900802,
      "balance_loss_mlp": 1.01823437,
      "epoch": 0.4432887419209379,
      "flos": 20084407660800.0,
      "grad_norm": 2.4681760221047218,
      "language_loss": 0.8366735,
      "learning_rate": 2.4602702403155664e-06,
      "loss": 0.85787463,
      "num_input_tokens_seen": 158098955,
      "step": 7373,
      "time_per_iteration": 3.9084439277648926
    },
    {
      "auxiliary_loss_clip": 0.0103581,
      "auxiliary_loss_mlp": 0.01002274,
      "balance_loss_clip": 1.02017951,
      "balance_loss_mlp": 1.00048578,
      "epoch": 0.4433488651736059,
      "flos": 70035540935040.0,
      "grad_norm": 0.7534798885330452,
      "language_loss": 0.55265146,
      "learning_rate": 2.4598912235733604e-06,
      "loss": 0.57303226,
      "num_input_tokens_seen": 158164110,
      "step": 7374,
      "time_per_iteration": 4.526760101318359
    },
    {
      "auxiliary_loss_clip": 0.01071479,
      "auxiliary_loss_mlp": 0.01041405,
      "balance_loss_clip": 1.04439151,
      "balance_loss_mlp": 1.02739179,
      "epoch": 0.44340898842627385,
      "flos": 16282472705280.0,
      "grad_norm": 2.3243722632459125,
      "language_loss": 0.82543015,
      "learning_rate": 2.4595121893920327e-06,
      "loss": 0.84655905,
      "num_input_tokens_seen": 158179850,
      "step": 7375,
      "time_per_iteration": 2.538583517074585
    },
    {
      "auxiliary_loss_clip": 0.01122149,
      "auxiliary_loss_mlp": 0.01032179,
      "balance_loss_clip": 1.04297042,
      "balance_loss_mlp": 1.01898253,
      "epoch": 0.4434691116789418,
      "flos": 16611989097600.0,
      "grad_norm": 1.86450824693346,
      "language_loss": 0.83944428,
      "learning_rate": 2.4591331377859578e-06,
      "loss": 0.8609876,
      "num_input_tokens_seen": 158196590,
      "step": 7376,
      "time_per_iteration": 2.4468259811401367
    },
    {
      "auxiliary_loss_clip": 0.0109885,
      "auxiliary_loss_mlp": 0.01032809,
      "balance_loss_clip": 1.04338098,
      "balance_loss_mlp": 1.01997614,
      "epoch": 0.4435292349316098,
      "flos": 19063251912960.0,
      "grad_norm": 1.8186672537306472,
      "language_loss": 0.77104115,
      "learning_rate": 2.4587540687695077e-06,
      "loss": 0.79235774,
      "num_input_tokens_seen": 158216355,
      "step": 7377,
      "time_per_iteration": 2.525489091873169
    },
    {
      "auxiliary_loss_clip": 0.01102739,
      "auxiliary_loss_mlp": 0.01029819,
      "balance_loss_clip": 1.04418039,
      "balance_loss_mlp": 1.01714051,
      "epoch": 0.44358935818427775,
      "flos": 21251324799360.0,
      "grad_norm": 1.7920790697773181,
      "language_loss": 0.75225317,
      "learning_rate": 2.458374982357057e-06,
      "loss": 0.77357876,
      "num_input_tokens_seen": 158235825,
      "step": 7378,
      "time_per_iteration": 3.890754222869873
    },
    {
      "auxiliary_loss_clip": 0.01095166,
      "auxiliary_loss_mlp": 0.01052613,
      "balance_loss_clip": 1.04030919,
      "balance_loss_mlp": 1.03775954,
      "epoch": 0.4436494814369457,
      "flos": 12495298239360.0,
      "grad_norm": 1.7870222650248118,
      "language_loss": 0.68731141,
      "learning_rate": 2.457995878562982e-06,
      "loss": 0.70878923,
      "num_input_tokens_seen": 158254230,
      "step": 7379,
      "time_per_iteration": 3.9493415355682373
    },
    {
      "auxiliary_loss_clip": 0.01058305,
      "auxiliary_loss_mlp": 0.01038986,
      "balance_loss_clip": 1.03815198,
      "balance_loss_mlp": 1.02335763,
      "epoch": 0.44370960468961373,
      "flos": 23659853408640.0,
      "grad_norm": 1.656922489830327,
      "language_loss": 0.73111248,
      "learning_rate": 2.457616757401656e-06,
      "loss": 0.75208533,
      "num_input_tokens_seen": 158273400,
      "step": 7380,
      "time_per_iteration": 2.602708101272583
    },
    {
      "auxiliary_loss_clip": 0.01100841,
      "auxiliary_loss_mlp": 0.01034292,
      "balance_loss_clip": 1.04307294,
      "balance_loss_mlp": 1.02113712,
      "epoch": 0.4437697279422817,
      "flos": 32416849635840.0,
      "grad_norm": 1.5046368819454756,
      "language_loss": 0.64296812,
      "learning_rate": 2.457237618887458e-06,
      "loss": 0.66431952,
      "num_input_tokens_seen": 158296840,
      "step": 7381,
      "time_per_iteration": 2.64060640335083
    },
    {
      "auxiliary_loss_clip": 0.01110219,
      "auxiliary_loss_mlp": 0.01034429,
      "balance_loss_clip": 1.04323483,
      "balance_loss_mlp": 1.02090418,
      "epoch": 0.44382985119494966,
      "flos": 18112875914880.0,
      "grad_norm": 2.073694259336502,
      "language_loss": 0.7970888,
      "learning_rate": 2.456858463034763e-06,
      "loss": 0.81853527,
      "num_input_tokens_seen": 158314935,
      "step": 7382,
      "time_per_iteration": 2.4656431674957275
    },
    {
      "auxiliary_loss_clip": 0.01115497,
      "auxiliary_loss_mlp": 0.01040916,
      "balance_loss_clip": 1.04515433,
      "balance_loss_mlp": 1.0277555,
      "epoch": 0.44388997444761763,
      "flos": 30774151923840.0,
      "grad_norm": 1.6471541169655348,
      "language_loss": 0.65385628,
      "learning_rate": 2.456479289857949e-06,
      "loss": 0.6754204,
      "num_input_tokens_seen": 158334620,
      "step": 7383,
      "time_per_iteration": 2.5567376613616943
    },
    {
      "auxiliary_loss_clip": 0.01100968,
      "auxiliary_loss_mlp": 0.01035326,
      "balance_loss_clip": 1.04497123,
      "balance_loss_mlp": 1.02118218,
      "epoch": 0.4439500977002856,
      "flos": 20339157893760.0,
      "grad_norm": 2.0833132001745103,
      "language_loss": 0.76047927,
      "learning_rate": 2.4561000993713953e-06,
      "loss": 0.78184223,
      "num_input_tokens_seen": 158350550,
      "step": 7384,
      "time_per_iteration": 2.4870405197143555
    },
    {
      "auxiliary_loss_clip": 0.01124874,
      "auxiliary_loss_mlp": 0.01030602,
      "balance_loss_clip": 1.0448823,
      "balance_loss_mlp": 1.01741743,
      "epoch": 0.44401022095295356,
      "flos": 20371225760640.0,
      "grad_norm": 1.8453910164114724,
      "language_loss": 0.81044638,
      "learning_rate": 2.4557208915894796e-06,
      "loss": 0.83200121,
      "num_input_tokens_seen": 158369555,
      "step": 7385,
      "time_per_iteration": 2.4588024616241455
    },
    {
      "auxiliary_loss_clip": 0.0107375,
      "auxiliary_loss_mlp": 0.01034745,
      "balance_loss_clip": 1.038908,
      "balance_loss_mlp": 1.02027917,
      "epoch": 0.4440703442056215,
      "flos": 20230635928320.0,
      "grad_norm": 1.7568600555033005,
      "language_loss": 0.81664991,
      "learning_rate": 2.455341666526582e-06,
      "loss": 0.83773488,
      "num_input_tokens_seen": 158388045,
      "step": 7386,
      "time_per_iteration": 2.565333366394043
    },
    {
      "auxiliary_loss_clip": 0.01081995,
      "auxiliary_loss_mlp": 0.01039313,
      "balance_loss_clip": 1.04043019,
      "balance_loss_mlp": 1.02460265,
      "epoch": 0.4441304674582895,
      "flos": 39494698824960.0,
      "grad_norm": 1.7789878058468669,
      "language_loss": 0.69741619,
      "learning_rate": 2.4549624241970832e-06,
      "loss": 0.71862924,
      "num_input_tokens_seen": 158410115,
      "step": 7387,
      "time_per_iteration": 2.702714443206787
    },
    {
      "auxiliary_loss_clip": 0.01054077,
      "auxiliary_loss_mlp": 0.01038245,
      "balance_loss_clip": 1.04028678,
      "balance_loss_mlp": 1.0243628,
      "epoch": 0.44419059071095746,
      "flos": 14829671220480.0,
      "grad_norm": 1.993022994028245,
      "language_loss": 0.72460485,
      "learning_rate": 2.4545831646153628e-06,
      "loss": 0.7455281,
      "num_input_tokens_seen": 158427765,
      "step": 7388,
      "time_per_iteration": 2.575754404067993
    },
    {
      "auxiliary_loss_clip": 0.01110102,
      "auxiliary_loss_mlp": 0.01032288,
      "balance_loss_clip": 1.04152155,
      "balance_loss_mlp": 1.01860857,
      "epoch": 0.4442507139636254,
      "flos": 22637835734400.0,
      "grad_norm": 1.6065197189641502,
      "language_loss": 0.69025224,
      "learning_rate": 2.4542038877958044e-06,
      "loss": 0.71167612,
      "num_input_tokens_seen": 158446375,
      "step": 7389,
      "time_per_iteration": 2.5158185958862305
    },
    {
      "auxiliary_loss_clip": 0.01110352,
      "auxiliary_loss_mlp": 0.01031833,
      "balance_loss_clip": 1.0422411,
      "balance_loss_mlp": 1.01907194,
      "epoch": 0.4443108372162934,
      "flos": 38290721829120.0,
      "grad_norm": 1.6510287068028187,
      "language_loss": 0.74583483,
      "learning_rate": 2.453824593752788e-06,
      "loss": 0.76725668,
      "num_input_tokens_seen": 158467260,
      "step": 7390,
      "time_per_iteration": 2.651491403579712
    },
    {
      "auxiliary_loss_clip": 0.01102637,
      "auxiliary_loss_mlp": 0.01038086,
      "balance_loss_clip": 1.03918755,
      "balance_loss_mlp": 1.02349496,
      "epoch": 0.44437096046896135,
      "flos": 17748993185280.0,
      "grad_norm": 2.0094964345929798,
      "language_loss": 0.81470346,
      "learning_rate": 2.4534452825006988e-06,
      "loss": 0.83611077,
      "num_input_tokens_seen": 158486720,
      "step": 7391,
      "time_per_iteration": 2.5151851177215576
    },
    {
      "auxiliary_loss_clip": 0.01092809,
      "auxiliary_loss_mlp": 0.01038497,
      "balance_loss_clip": 1.04597723,
      "balance_loss_mlp": 1.02473402,
      "epoch": 0.4444310837216293,
      "flos": 13732348682880.0,
      "grad_norm": 1.6491910526200175,
      "language_loss": 0.73424315,
      "learning_rate": 2.4530659540539185e-06,
      "loss": 0.75555629,
      "num_input_tokens_seen": 158502530,
      "step": 7392,
      "time_per_iteration": 2.5465571880340576
    },
    {
      "auxiliary_loss_clip": 0.01107437,
      "auxiliary_loss_mlp": 0.01029435,
      "balance_loss_clip": 1.04082692,
      "balance_loss_mlp": 1.01664364,
      "epoch": 0.44449120697429734,
      "flos": 25010238240000.0,
      "grad_norm": 1.5639215556325483,
      "language_loss": 0.79372394,
      "learning_rate": 2.4526866084268313e-06,
      "loss": 0.81509268,
      "num_input_tokens_seen": 158522715,
      "step": 7393,
      "time_per_iteration": 2.5820066928863525
    },
    {
      "auxiliary_loss_clip": 0.01113114,
      "auxiliary_loss_mlp": 0.01034856,
      "balance_loss_clip": 1.04216862,
      "balance_loss_mlp": 1.02108729,
      "epoch": 0.4445513302269653,
      "flos": 32671707609600.0,
      "grad_norm": 1.909140796152854,
      "language_loss": 0.80475283,
      "learning_rate": 2.4523072456338226e-06,
      "loss": 0.82623255,
      "num_input_tokens_seen": 158543615,
      "step": 7394,
      "time_per_iteration": 2.564201831817627
    },
    {
      "auxiliary_loss_clip": 0.01093837,
      "auxiliary_loss_mlp": 0.01039095,
      "balance_loss_clip": 1.03972316,
      "balance_loss_mlp": 1.02661431,
      "epoch": 0.44461145347963327,
      "flos": 11655814504320.0,
      "grad_norm": 4.343345611093114,
      "language_loss": 0.7948432,
      "learning_rate": 2.4519278656892785e-06,
      "loss": 0.8161726,
      "num_input_tokens_seen": 158560330,
      "step": 7395,
      "time_per_iteration": 2.490449905395508
    },
    {
      "auxiliary_loss_clip": 0.01094983,
      "auxiliary_loss_mlp": 0.01036902,
      "balance_loss_clip": 1.04043508,
      "balance_loss_mlp": 1.02419996,
      "epoch": 0.44467157673230123,
      "flos": 20886759711360.0,
      "grad_norm": 1.7031303663128772,
      "language_loss": 0.68336976,
      "learning_rate": 2.451548468607584e-06,
      "loss": 0.70468867,
      "num_input_tokens_seen": 158579735,
      "step": 7396,
      "time_per_iteration": 2.559892416000366
    },
    {
      "auxiliary_loss_clip": 0.01106334,
      "auxiliary_loss_mlp": 0.0079152,
      "balance_loss_clip": 1.04057491,
      "balance_loss_mlp": 1.01857686,
      "epoch": 0.4447316999849692,
      "flos": 18546137763840.0,
      "grad_norm": 1.7570315564215924,
      "language_loss": 0.80708921,
      "learning_rate": 2.451169054403126e-06,
      "loss": 0.82606769,
      "num_input_tokens_seen": 158597075,
      "step": 7397,
      "time_per_iteration": 2.50197172164917
    },
    {
      "auxiliary_loss_clip": 0.01107463,
      "auxiliary_loss_mlp": 0.01034873,
      "balance_loss_clip": 1.04048896,
      "balance_loss_mlp": 1.02211142,
      "epoch": 0.44479182323763716,
      "flos": 23769057732480.0,
      "grad_norm": 1.633641888073597,
      "language_loss": 0.67625737,
      "learning_rate": 2.450789623090293e-06,
      "loss": 0.69768071,
      "num_input_tokens_seen": 158616650,
      "step": 7398,
      "time_per_iteration": 2.494442939758301
    },
    {
      "auxiliary_loss_clip": 0.01087317,
      "auxiliary_loss_mlp": 0.01038571,
      "balance_loss_clip": 1.04102325,
      "balance_loss_mlp": 1.02603602,
      "epoch": 0.44485194649030513,
      "flos": 16543831040640.0,
      "grad_norm": 2.3262350985889872,
      "language_loss": 0.69622934,
      "learning_rate": 2.450410174683472e-06,
      "loss": 0.71748817,
      "num_input_tokens_seen": 158634515,
      "step": 7399,
      "time_per_iteration": 2.5390021800994873
    },
    {
      "auxiliary_loss_clip": 0.01086216,
      "auxiliary_loss_mlp": 0.01034776,
      "balance_loss_clip": 1.03936291,
      "balance_loss_mlp": 1.02239609,
      "epoch": 0.4449120697429731,
      "flos": 22600955445120.0,
      "grad_norm": 2.099410451625582,
      "language_loss": 0.72376847,
      "learning_rate": 2.4500307091970514e-06,
      "loss": 0.74497837,
      "num_input_tokens_seen": 158653760,
      "step": 7400,
      "time_per_iteration": 2.543809413909912
    },
    {
      "auxiliary_loss_clip": 0.0107277,
      "auxiliary_loss_mlp": 0.0078921,
      "balance_loss_clip": 1.04336214,
      "balance_loss_mlp": 1.01648426,
      "epoch": 0.44497219299564106,
      "flos": 20004864992640.0,
      "grad_norm": 2.3392910319087727,
      "language_loss": 0.84828854,
      "learning_rate": 2.449651226645422e-06,
      "loss": 0.86690831,
      "num_input_tokens_seen": 158672190,
      "step": 7401,
      "time_per_iteration": 2.604703664779663
    },
    {
      "auxiliary_loss_clip": 0.01091967,
      "auxiliary_loss_mlp": 0.01035204,
      "balance_loss_clip": 1.04000759,
      "balance_loss_mlp": 1.02338433,
      "epoch": 0.445032316248309,
      "flos": 25594253470080.0,
      "grad_norm": 1.545166587646256,
      "language_loss": 0.83198184,
      "learning_rate": 2.449271727042973e-06,
      "loss": 0.85325354,
      "num_input_tokens_seen": 158694115,
      "step": 7402,
      "time_per_iteration": 2.5712177753448486
    },
    {
      "auxiliary_loss_clip": 0.01099046,
      "auxiliary_loss_mlp": 0.01032262,
      "balance_loss_clip": 1.04404616,
      "balance_loss_mlp": 1.01934016,
      "epoch": 0.445092439500977,
      "flos": 21250426959360.0,
      "grad_norm": 1.708147841356983,
      "language_loss": 0.76956093,
      "learning_rate": 2.4488922104040947e-06,
      "loss": 0.790874,
      "num_input_tokens_seen": 158711000,
      "step": 7403,
      "time_per_iteration": 2.5259180068969727
    },
    {
      "auxiliary_loss_clip": 0.01028835,
      "auxiliary_loss_mlp": 0.01007005,
      "balance_loss_clip": 1.02806973,
      "balance_loss_mlp": 1.00540733,
      "epoch": 0.44515256275364495,
      "flos": 57764900309760.0,
      "grad_norm": 0.7431518282666976,
      "language_loss": 0.60000384,
      "learning_rate": 2.4485126767431793e-06,
      "loss": 0.62036222,
      "num_input_tokens_seen": 158769675,
      "step": 7404,
      "time_per_iteration": 3.1163673400878906
    },
    {
      "auxiliary_loss_clip": 0.01095807,
      "auxiliary_loss_mlp": 0.01037454,
      "balance_loss_clip": 1.03983843,
      "balance_loss_mlp": 1.02235627,
      "epoch": 0.4452126860063129,
      "flos": 15596004908160.0,
      "grad_norm": 1.7871058358776286,
      "language_loss": 0.81944239,
      "learning_rate": 2.4481331260746177e-06,
      "loss": 0.84077501,
      "num_input_tokens_seen": 158788215,
      "step": 7405,
      "time_per_iteration": 2.5334198474884033
    },
    {
      "auxiliary_loss_clip": 0.01093631,
      "auxiliary_loss_mlp": 0.01027567,
      "balance_loss_clip": 1.03947926,
      "balance_loss_mlp": 1.01472187,
      "epoch": 0.4452728092589809,
      "flos": 21617398258560.0,
      "grad_norm": 2.2874055860558355,
      "language_loss": 0.74988556,
      "learning_rate": 2.4477535584128036e-06,
      "loss": 0.77109754,
      "num_input_tokens_seen": 158809090,
      "step": 7406,
      "time_per_iteration": 2.5324010848999023
    },
    {
      "auxiliary_loss_clip": 0.01079687,
      "auxiliary_loss_mlp": 0.01030029,
      "balance_loss_clip": 1.03673828,
      "balance_loss_mlp": 1.0174706,
      "epoch": 0.4453329325116489,
      "flos": 29497491757440.0,
      "grad_norm": 1.6667225714028913,
      "language_loss": 0.6522187,
      "learning_rate": 2.447373973772129e-06,
      "loss": 0.67331588,
      "num_input_tokens_seen": 158828320,
      "step": 7407,
      "time_per_iteration": 2.629051685333252
    },
    {
      "auxiliary_loss_clip": 0.01095012,
      "auxiliary_loss_mlp": 0.01032648,
      "balance_loss_clip": 1.04608774,
      "balance_loss_mlp": 1.01952934,
      "epoch": 0.44539305576431687,
      "flos": 21361139654400.0,
      "grad_norm": 1.7194354529599825,
      "language_loss": 0.68226975,
      "learning_rate": 2.4469943721669887e-06,
      "loss": 0.70354635,
      "num_input_tokens_seen": 158847040,
      "step": 7408,
      "time_per_iteration": 2.5417163372039795
    },
    {
      "auxiliary_loss_clip": 0.01117665,
      "auxiliary_loss_mlp": 0.01033904,
      "balance_loss_clip": 1.03917861,
      "balance_loss_mlp": 1.02018929,
      "epoch": 0.44545317901698483,
      "flos": 41427626428800.0,
      "grad_norm": 1.4811674914507897,
      "language_loss": 0.72102529,
      "learning_rate": 2.4466147536117776e-06,
      "loss": 0.74254096,
      "num_input_tokens_seen": 158870490,
      "step": 7409,
      "time_per_iteration": 2.6263303756713867
    },
    {
      "auxiliary_loss_clip": 0.01096067,
      "auxiliary_loss_mlp": 0.01034058,
      "balance_loss_clip": 1.03983068,
      "balance_loss_mlp": 1.02003908,
      "epoch": 0.4455133022696528,
      "flos": 22055005653120.0,
      "grad_norm": 1.8005081749006628,
      "language_loss": 0.65371382,
      "learning_rate": 2.4462351181208895e-06,
      "loss": 0.67501509,
      "num_input_tokens_seen": 158889920,
      "step": 7410,
      "time_per_iteration": 2.515620470046997
    },
    {
      "auxiliary_loss_clip": 0.01096941,
      "auxiliary_loss_mlp": 0.01031289,
      "balance_loss_clip": 1.04130936,
      "balance_loss_mlp": 1.01728225,
      "epoch": 0.44557342552232077,
      "flos": 23476960333440.0,
      "grad_norm": 1.9208992854065592,
      "language_loss": 0.73997593,
      "learning_rate": 2.4458554657087217e-06,
      "loss": 0.76125824,
      "num_input_tokens_seen": 158909580,
      "step": 7411,
      "time_per_iteration": 2.577432632446289
    },
    {
      "auxiliary_loss_clip": 0.01053717,
      "auxiliary_loss_mlp": 0.01029757,
      "balance_loss_clip": 1.04172099,
      "balance_loss_mlp": 1.01707292,
      "epoch": 0.44563354877498873,
      "flos": 19134678107520.0,
      "grad_norm": 1.7759344421222005,
      "language_loss": 0.79032755,
      "learning_rate": 2.4454757963896695e-06,
      "loss": 0.81116229,
      "num_input_tokens_seen": 158924600,
      "step": 7412,
      "time_per_iteration": 3.9864614009857178
    },
    {
      "auxiliary_loss_clip": 0.01099523,
      "auxiliary_loss_mlp": 0.01032412,
      "balance_loss_clip": 1.03955472,
      "balance_loss_mlp": 1.01945972,
      "epoch": 0.4456936720276567,
      "flos": 13621420506240.0,
      "grad_norm": 1.8479450670135789,
      "language_loss": 0.80014396,
      "learning_rate": 2.4450961101781304e-06,
      "loss": 0.82146335,
      "num_input_tokens_seen": 158939345,
      "step": 7413,
      "time_per_iteration": 3.8547914028167725
    },
    {
      "auxiliary_loss_clip": 0.01105613,
      "auxiliary_loss_mlp": 0.0102702,
      "balance_loss_clip": 1.0403136,
      "balance_loss_mlp": 1.0140022,
      "epoch": 0.44575379528032466,
      "flos": 14713715139840.0,
      "grad_norm": 1.8078941267235116,
      "language_loss": 0.76546514,
      "learning_rate": 2.4447164070885026e-06,
      "loss": 0.7867915,
      "num_input_tokens_seen": 158955855,
      "step": 7414,
      "time_per_iteration": 2.4533703327178955
    },
    {
      "auxiliary_loss_clip": 0.01089424,
      "auxiliary_loss_mlp": 0.01038166,
      "balance_loss_clip": 1.03976583,
      "balance_loss_mlp": 1.02405143,
      "epoch": 0.4458139185329926,
      "flos": 24170682677760.0,
      "grad_norm": 1.726835541145688,
      "language_loss": 0.83579743,
      "learning_rate": 2.4443366871351837e-06,
      "loss": 0.85707337,
      "num_input_tokens_seen": 158976315,
      "step": 7415,
      "time_per_iteration": 2.5485339164733887
    },
    {
      "auxiliary_loss_clip": 0.01115156,
      "auxiliary_loss_mlp": 0.01038434,
      "balance_loss_clip": 1.03852189,
      "balance_loss_mlp": 1.0252254,
      "epoch": 0.4458740417856606,
      "flos": 21762225895680.0,
      "grad_norm": 1.6724523862360705,
      "language_loss": 0.84123683,
      "learning_rate": 2.4439569503325732e-06,
      "loss": 0.86277276,
      "num_input_tokens_seen": 158996725,
      "step": 7416,
      "time_per_iteration": 2.4750187397003174
    },
    {
      "auxiliary_loss_clip": 0.01084669,
      "auxiliary_loss_mlp": 0.01030117,
      "balance_loss_clip": 1.04118276,
      "balance_loss_mlp": 1.01701546,
      "epoch": 0.44593416503832856,
      "flos": 21068790860160.0,
      "grad_norm": 4.632257721662155,
      "language_loss": 0.81347209,
      "learning_rate": 2.4435771966950706e-06,
      "loss": 0.83461994,
      "num_input_tokens_seen": 159017255,
      "step": 7417,
      "time_per_iteration": 3.9501397609710693
    },
    {
      "auxiliary_loss_clip": 0.01098307,
      "auxiliary_loss_mlp": 0.01040538,
      "balance_loss_clip": 1.03979218,
      "balance_loss_mlp": 1.02697766,
      "epoch": 0.4459942882909965,
      "flos": 22600488568320.0,
      "grad_norm": 1.9340637062076786,
      "language_loss": 0.81122351,
      "learning_rate": 2.443197426237077e-06,
      "loss": 0.83261192,
      "num_input_tokens_seen": 159035010,
      "step": 7418,
      "time_per_iteration": 3.8855233192443848
    },
    {
      "auxiliary_loss_clip": 0.01107706,
      "auxiliary_loss_mlp": 0.00787007,
      "balance_loss_clip": 1.03898573,
      "balance_loss_mlp": 1.01169086,
      "epoch": 0.4460544115436645,
      "flos": 26505486622080.0,
      "grad_norm": 1.5359206590581067,
      "language_loss": 0.77144349,
      "learning_rate": 2.442817638972991e-06,
      "loss": 0.79039067,
      "num_input_tokens_seen": 159055345,
      "step": 7419,
      "time_per_iteration": 2.51133131980896
    },
    {
      "auxiliary_loss_clip": 0.01085732,
      "auxiliary_loss_mlp": 0.0103486,
      "balance_loss_clip": 1.03909302,
      "balance_loss_mlp": 1.02230108,
      "epoch": 0.4461145347963325,
      "flos": 17604021893760.0,
      "grad_norm": 1.5523187550904443,
      "language_loss": 0.72370851,
      "learning_rate": 2.4424378349172176e-06,
      "loss": 0.74491441,
      "num_input_tokens_seen": 159074225,
      "step": 7420,
      "time_per_iteration": 2.5188379287719727
    },
    {
      "auxiliary_loss_clip": 0.01095908,
      "auxiliary_loss_mlp": 0.01032325,
      "balance_loss_clip": 1.03897285,
      "balance_loss_mlp": 1.01828218,
      "epoch": 0.44617465804900047,
      "flos": 27268193036160.0,
      "grad_norm": 1.6653675043305245,
      "language_loss": 0.7459991,
      "learning_rate": 2.442058014084156e-06,
      "loss": 0.76728141,
      "num_input_tokens_seen": 159095415,
      "step": 7421,
      "time_per_iteration": 2.5413825511932373
    },
    {
      "auxiliary_loss_clip": 0.01053929,
      "auxiliary_loss_mlp": 0.01035933,
      "balance_loss_clip": 1.03758562,
      "balance_loss_mlp": 1.0224508,
      "epoch": 0.44623478130166844,
      "flos": 17786412178560.0,
      "grad_norm": 1.8123598304337882,
      "language_loss": 0.75934547,
      "learning_rate": 2.44167817648821e-06,
      "loss": 0.78024411,
      "num_input_tokens_seen": 159114615,
      "step": 7422,
      "time_per_iteration": 2.586775541305542
    },
    {
      "auxiliary_loss_clip": 0.01118812,
      "auxiliary_loss_mlp": 0.01030441,
      "balance_loss_clip": 1.04219091,
      "balance_loss_mlp": 1.01778126,
      "epoch": 0.4462949045543364,
      "flos": 23003011353600.0,
      "grad_norm": 1.5470905565911495,
      "language_loss": 0.65211606,
      "learning_rate": 2.441298322143784e-06,
      "loss": 0.67360866,
      "num_input_tokens_seen": 159134370,
      "step": 7423,
      "time_per_iteration": 2.4573192596435547
    },
    {
      "auxiliary_loss_clip": 0.01092495,
      "auxiliary_loss_mlp": 0.01031302,
      "balance_loss_clip": 1.04254985,
      "balance_loss_mlp": 1.01957726,
      "epoch": 0.44635502780700437,
      "flos": 17820096157440.0,
      "grad_norm": 1.422198181864022,
      "language_loss": 0.79056275,
      "learning_rate": 2.4409184510652807e-06,
      "loss": 0.81180066,
      "num_input_tokens_seen": 159152540,
      "step": 7424,
      "time_per_iteration": 2.504376173019409
    },
    {
      "auxiliary_loss_clip": 0.01103048,
      "auxiliary_loss_mlp": 0.01030074,
      "balance_loss_clip": 1.04210019,
      "balance_loss_mlp": 1.01786685,
      "epoch": 0.44641515105967233,
      "flos": 26688020561280.0,
      "grad_norm": 1.42210487840812,
      "language_loss": 0.80121565,
      "learning_rate": 2.4405385632671063e-06,
      "loss": 0.8225469,
      "num_input_tokens_seen": 159173425,
      "step": 7425,
      "time_per_iteration": 2.537569046020508
    },
    {
      "auxiliary_loss_clip": 0.01105222,
      "auxiliary_loss_mlp": 0.01028491,
      "balance_loss_clip": 1.04013073,
      "balance_loss_mlp": 1.0164212,
      "epoch": 0.4464752743123403,
      "flos": 18913324544640.0,
      "grad_norm": 1.4848317511871891,
      "language_loss": 0.7725504,
      "learning_rate": 2.4401586587636655e-06,
      "loss": 0.7938875,
      "num_input_tokens_seen": 159191210,
      "step": 7426,
      "time_per_iteration": 2.4929213523864746
    },
    {
      "auxiliary_loss_clip": 0.01089791,
      "auxiliary_loss_mlp": 0.00786154,
      "balance_loss_clip": 1.0381844,
      "balance_loss_mlp": 1.00958741,
      "epoch": 0.44653539756500826,
      "flos": 29570318582400.0,
      "grad_norm": 1.5682584008667182,
      "language_loss": 0.64415956,
      "learning_rate": 2.4397787375693634e-06,
      "loss": 0.66291904,
      "num_input_tokens_seen": 159211755,
      "step": 7427,
      "time_per_iteration": 2.613039016723633
    },
    {
      "auxiliary_loss_clip": 0.0111285,
      "auxiliary_loss_mlp": 0.0103274,
      "balance_loss_clip": 1.04468906,
      "balance_loss_mlp": 1.01978159,
      "epoch": 0.44659552081767623,
      "flos": 21468979261440.0,
      "grad_norm": 2.008674076730728,
      "language_loss": 0.75627744,
      "learning_rate": 2.439398799698608e-06,
      "loss": 0.77773333,
      "num_input_tokens_seen": 159230315,
      "step": 7428,
      "time_per_iteration": 2.490018129348755
    },
    {
      "auxiliary_loss_clip": 0.0108462,
      "auxiliary_loss_mlp": 0.01032958,
      "balance_loss_clip": 1.0385859,
      "balance_loss_mlp": 1.02010155,
      "epoch": 0.4466556440703442,
      "flos": 17931886260480.0,
      "grad_norm": 1.7709711046132655,
      "language_loss": 0.77910864,
      "learning_rate": 2.439018845165806e-06,
      "loss": 0.8002845,
      "num_input_tokens_seen": 159249810,
      "step": 7429,
      "time_per_iteration": 2.5291969776153564
    },
    {
      "auxiliary_loss_clip": 0.01108108,
      "auxiliary_loss_mlp": 0.01031513,
      "balance_loss_clip": 1.04038763,
      "balance_loss_mlp": 1.01760101,
      "epoch": 0.44671576732301216,
      "flos": 21107430915840.0,
      "grad_norm": 2.514145522006151,
      "language_loss": 0.91172755,
      "learning_rate": 2.438638873985366e-06,
      "loss": 0.93312371,
      "num_input_tokens_seen": 159271715,
      "step": 7430,
      "time_per_iteration": 2.537958860397339
    },
    {
      "auxiliary_loss_clip": 0.01097615,
      "auxiliary_loss_mlp": 0.00789099,
      "balance_loss_clip": 1.0413326,
      "balance_loss_mlp": 1.01172304,
      "epoch": 0.4467758905756801,
      "flos": 23508920459520.0,
      "grad_norm": 1.597068142248159,
      "language_loss": 0.79465067,
      "learning_rate": 2.4382588861716954e-06,
      "loss": 0.81351781,
      "num_input_tokens_seen": 159290690,
      "step": 7431,
      "time_per_iteration": 2.534229278564453
    },
    {
      "auxiliary_loss_clip": 0.01099554,
      "auxiliary_loss_mlp": 0.01034894,
      "balance_loss_clip": 1.04109001,
      "balance_loss_mlp": 1.02141738,
      "epoch": 0.4468360138283481,
      "flos": 18734022829440.0,
      "grad_norm": 1.7903861281663358,
      "language_loss": 0.79553199,
      "learning_rate": 2.437878881739204e-06,
      "loss": 0.81687647,
      "num_input_tokens_seen": 159309400,
      "step": 7432,
      "time_per_iteration": 2.509443759918213
    },
    {
      "auxiliary_loss_clip": 0.01083633,
      "auxiliary_loss_mlp": 0.01034475,
      "balance_loss_clip": 1.03936327,
      "balance_loss_mlp": 1.022089,
      "epoch": 0.4468961370810161,
      "flos": 23477139901440.0,
      "grad_norm": 1.7725576508021001,
      "language_loss": 0.76376545,
      "learning_rate": 2.437498860702301e-06,
      "loss": 0.78494656,
      "num_input_tokens_seen": 159327425,
      "step": 7433,
      "time_per_iteration": 2.5415995121002197
    },
    {
      "auxiliary_loss_clip": 0.01098757,
      "auxiliary_loss_mlp": 0.01029928,
      "balance_loss_clip": 1.04148149,
      "balance_loss_mlp": 1.01919317,
      "epoch": 0.4469562603336841,
      "flos": 30075042539520.0,
      "grad_norm": 1.6943480538579203,
      "language_loss": 0.77176708,
      "learning_rate": 2.437118823075398e-06,
      "loss": 0.79305393,
      "num_input_tokens_seen": 159345805,
      "step": 7434,
      "time_per_iteration": 2.5465259552001953
    },
    {
      "auxiliary_loss_clip": 0.01110062,
      "auxiliary_loss_mlp": 0.01028553,
      "balance_loss_clip": 1.04183161,
      "balance_loss_mlp": 1.01555908,
      "epoch": 0.44701638358635204,
      "flos": 22456415116800.0,
      "grad_norm": 2.0620106160321874,
      "language_loss": 0.64582145,
      "learning_rate": 2.436738768872905e-06,
      "loss": 0.6672076,
      "num_input_tokens_seen": 159364595,
      "step": 7435,
      "time_per_iteration": 2.493406295776367
    },
    {
      "auxiliary_loss_clip": 0.01100004,
      "auxiliary_loss_mlp": 0.01029407,
      "balance_loss_clip": 1.04089046,
      "balance_loss_mlp": 1.0159843,
      "epoch": 0.44707650683902,
      "flos": 24057851080320.0,
      "grad_norm": 1.7160011317363102,
      "language_loss": 0.83686835,
      "learning_rate": 2.4363586981092346e-06,
      "loss": 0.8581624,
      "num_input_tokens_seen": 159385265,
      "step": 7436,
      "time_per_iteration": 2.5529539585113525
    },
    {
      "auxiliary_loss_clip": 0.01069151,
      "auxiliary_loss_mlp": 0.01049936,
      "balance_loss_clip": 1.03913295,
      "balance_loss_mlp": 1.03291273,
      "epoch": 0.44713663009168797,
      "flos": 23766938830080.0,
      "grad_norm": 1.7592174959421958,
      "language_loss": 0.79477149,
      "learning_rate": 2.435978610798798e-06,
      "loss": 0.81596237,
      "num_input_tokens_seen": 159405080,
      "step": 7437,
      "time_per_iteration": 2.5765185356140137
    },
    {
      "auxiliary_loss_clip": 0.01074174,
      "auxiliary_loss_mlp": 0.01034003,
      "balance_loss_clip": 1.04423571,
      "balance_loss_mlp": 1.02106929,
      "epoch": 0.44719675334435594,
      "flos": 24499265316480.0,
      "grad_norm": 1.5765958624701573,
      "language_loss": 0.71908504,
      "learning_rate": 2.435598506956009e-06,
      "loss": 0.74016678,
      "num_input_tokens_seen": 159424595,
      "step": 7438,
      "time_per_iteration": 2.614311695098877
    },
    {
      "auxiliary_loss_clip": 0.01083953,
      "auxiliary_loss_mlp": 0.01037594,
      "balance_loss_clip": 1.04163742,
      "balance_loss_mlp": 1.02293086,
      "epoch": 0.4472568765970239,
      "flos": 29781759991680.0,
      "grad_norm": 1.6042391542388204,
      "language_loss": 0.6713571,
      "learning_rate": 2.4352183865952808e-06,
      "loss": 0.69257253,
      "num_input_tokens_seen": 159443865,
      "step": 7439,
      "time_per_iteration": 2.63220477104187
    },
    {
      "auxiliary_loss_clip": 0.01095266,
      "auxiliary_loss_mlp": 0.01039255,
      "balance_loss_clip": 1.03786087,
      "balance_loss_mlp": 1.02393019,
      "epoch": 0.44731699984969187,
      "flos": 24643123286400.0,
      "grad_norm": 1.8831113160536335,
      "language_loss": 0.73817587,
      "learning_rate": 2.4348382497310285e-06,
      "loss": 0.75952113,
      "num_input_tokens_seen": 159464525,
      "step": 7440,
      "time_per_iteration": 2.548387050628662
    },
    {
      "auxiliary_loss_clip": 0.01066938,
      "auxiliary_loss_mlp": 0.01045617,
      "balance_loss_clip": 1.03527236,
      "balance_loss_mlp": 1.03047156,
      "epoch": 0.44737712310235983,
      "flos": 29455691304960.0,
      "grad_norm": 1.680109608425036,
      "language_loss": 0.74019319,
      "learning_rate": 2.4344580963776655e-06,
      "loss": 0.76131868,
      "num_input_tokens_seen": 159486385,
      "step": 7441,
      "time_per_iteration": 2.622896909713745
    },
    {
      "auxiliary_loss_clip": 0.0108788,
      "auxiliary_loss_mlp": 0.01034517,
      "balance_loss_clip": 1.04468584,
      "balance_loss_mlp": 1.02056956,
      "epoch": 0.4474372463550278,
      "flos": 24896832024960.0,
      "grad_norm": 1.8077807839840991,
      "language_loss": 0.74847817,
      "learning_rate": 2.4340779265496082e-06,
      "loss": 0.76970214,
      "num_input_tokens_seen": 159503880,
      "step": 7442,
      "time_per_iteration": 2.565925121307373
    },
    {
      "auxiliary_loss_clip": 0.01122346,
      "auxiliary_loss_mlp": 0.01035802,
      "balance_loss_clip": 1.04035389,
      "balance_loss_mlp": 1.02097213,
      "epoch": 0.44749736960769576,
      "flos": 33181603125120.0,
      "grad_norm": 1.7341627429769624,
      "language_loss": 0.74522817,
      "learning_rate": 2.433697740261273e-06,
      "loss": 0.76680958,
      "num_input_tokens_seen": 159522980,
      "step": 7443,
      "time_per_iteration": 2.5312509536743164
    },
    {
      "auxiliary_loss_clip": 0.01094322,
      "auxiliary_loss_mlp": 0.01033085,
      "balance_loss_clip": 1.03604805,
      "balance_loss_mlp": 1.01800478,
      "epoch": 0.4475574928603637,
      "flos": 21071807602560.0,
      "grad_norm": 1.6553949656962152,
      "language_loss": 0.77939546,
      "learning_rate": 2.4333175375270748e-06,
      "loss": 0.80066943,
      "num_input_tokens_seen": 159543340,
      "step": 7444,
      "time_per_iteration": 2.5193114280700684
    },
    {
      "auxiliary_loss_clip": 0.01100731,
      "auxiliary_loss_mlp": 0.01034158,
      "balance_loss_clip": 1.04064274,
      "balance_loss_mlp": 1.02062154,
      "epoch": 0.4476176161130317,
      "flos": 21862523646720.0,
      "grad_norm": 2.273572040242331,
      "language_loss": 0.84874004,
      "learning_rate": 2.4329373183614333e-06,
      "loss": 0.87008893,
      "num_input_tokens_seen": 159558210,
      "step": 7445,
      "time_per_iteration": 2.458216667175293
    },
    {
      "auxiliary_loss_clip": 0.01083787,
      "auxiliary_loss_mlp": 0.01043496,
      "balance_loss_clip": 1.03954816,
      "balance_loss_mlp": 1.02600765,
      "epoch": 0.4476777393656997,
      "flos": 22528667324160.0,
      "grad_norm": 2.1808979765518313,
      "language_loss": 0.64138019,
      "learning_rate": 2.432557082778765e-06,
      "loss": 0.66265297,
      "num_input_tokens_seen": 159577920,
      "step": 7446,
      "time_per_iteration": 2.550922155380249
    },
    {
      "auxiliary_loss_clip": 0.01036656,
      "auxiliary_loss_mlp": 0.01000138,
      "balance_loss_clip": 1.02048635,
      "balance_loss_mlp": 0.99792045,
      "epoch": 0.4477378626183677,
      "flos": 49017133877760.0,
      "grad_norm": 0.7396427118422974,
      "language_loss": 0.50242054,
      "learning_rate": 2.4321768307934884e-06,
      "loss": 0.52278841,
      "num_input_tokens_seen": 159632295,
      "step": 7447,
      "time_per_iteration": 2.962052583694458
    },
    {
      "auxiliary_loss_clip": 0.01044446,
      "auxiliary_loss_mlp": 0.01002953,
      "balance_loss_clip": 1.01748574,
      "balance_loss_mlp": 1.00084329,
      "epoch": 0.44779798587103564,
      "flos": 56542179392640.0,
      "grad_norm": 0.7769494042661208,
      "language_loss": 0.593665,
      "learning_rate": 2.4317965624200235e-06,
      "loss": 0.61413902,
      "num_input_tokens_seen": 159698435,
      "step": 7448,
      "time_per_iteration": 3.1229686737060547
    },
    {
      "auxiliary_loss_clip": 0.01086581,
      "auxiliary_loss_mlp": 0.01033774,
      "balance_loss_clip": 1.03974962,
      "balance_loss_mlp": 1.02071476,
      "epoch": 0.4478581091237036,
      "flos": 46498536040320.0,
      "grad_norm": 1.646677942202416,
      "language_loss": 0.58944619,
      "learning_rate": 2.431416277672789e-06,
      "loss": 0.61064976,
      "num_input_tokens_seen": 159722150,
      "step": 7449,
      "time_per_iteration": 2.7624213695526123
    },
    {
      "auxiliary_loss_clip": 0.01090197,
      "auxiliary_loss_mlp": 0.01029868,
      "balance_loss_clip": 1.04050636,
      "balance_loss_mlp": 1.01684451,
      "epoch": 0.4479182323763716,
      "flos": 20814363849600.0,
      "grad_norm": 1.7449380054488262,
      "language_loss": 0.79866958,
      "learning_rate": 2.4310359765662065e-06,
      "loss": 0.81987023,
      "num_input_tokens_seen": 159740550,
      "step": 7450,
      "time_per_iteration": 3.9778456687927246
    },
    {
      "auxiliary_loss_clip": 0.01119223,
      "auxiliary_loss_mlp": 0.01042666,
      "balance_loss_clip": 1.04125738,
      "balance_loss_mlp": 1.02927303,
      "epoch": 0.44797835562903954,
      "flos": 14245979212800.0,
      "grad_norm": 2.3212304750461743,
      "language_loss": 0.7958827,
      "learning_rate": 2.430655659114697e-06,
      "loss": 0.8175016,
      "num_input_tokens_seen": 159758245,
      "step": 7451,
      "time_per_iteration": 3.8403098583221436
    },
    {
      "auxiliary_loss_clip": 0.01010712,
      "auxiliary_loss_mlp": 0.01002181,
      "balance_loss_clip": 1.03050435,
      "balance_loss_mlp": 1.00046396,
      "epoch": 0.4480384788817075,
      "flos": 63534560169600.0,
      "grad_norm": 0.8378029958482914,
      "language_loss": 0.62868547,
      "learning_rate": 2.430275325332681e-06,
      "loss": 0.64881438,
      "num_input_tokens_seen": 159826790,
      "step": 7452,
      "time_per_iteration": 3.281829833984375
    },
    {
      "auxiliary_loss_clip": 0.01122056,
      "auxiliary_loss_mlp": 0.01033673,
      "balance_loss_clip": 1.04267943,
      "balance_loss_mlp": 1.01933825,
      "epoch": 0.44809860213437547,
      "flos": 21652626522240.0,
      "grad_norm": 1.8934888103971135,
      "language_loss": 0.62768275,
      "learning_rate": 2.429894975234582e-06,
      "loss": 0.64924002,
      "num_input_tokens_seen": 159845805,
      "step": 7453,
      "time_per_iteration": 2.458674192428589
    },
    {
      "auxiliary_loss_clip": 0.01036726,
      "auxiliary_loss_mlp": 0.01002061,
      "balance_loss_clip": 1.01874316,
      "balance_loss_mlp": 1.00009441,
      "epoch": 0.44815872538704343,
      "flos": 69190634246400.0,
      "grad_norm": 0.7915424986533482,
      "language_loss": 0.57079083,
      "learning_rate": 2.4295146088348224e-06,
      "loss": 0.59117872,
      "num_input_tokens_seen": 159898860,
      "step": 7454,
      "time_per_iteration": 2.9776217937469482
    },
    {
      "auxiliary_loss_clip": 0.01092665,
      "auxiliary_loss_mlp": 0.01040223,
      "balance_loss_clip": 1.03712249,
      "balance_loss_mlp": 1.02592945,
      "epoch": 0.4482188486397114,
      "flos": 12598289510400.0,
      "grad_norm": 2.1182682221340015,
      "language_loss": 0.74760842,
      "learning_rate": 2.4291342261478255e-06,
      "loss": 0.76893723,
      "num_input_tokens_seen": 159911555,
      "step": 7455,
      "time_per_iteration": 2.4634969234466553
    },
    {
      "auxiliary_loss_clip": 0.01096517,
      "auxiliary_loss_mlp": 0.01034734,
      "balance_loss_clip": 1.04014063,
      "balance_loss_mlp": 1.02204418,
      "epoch": 0.44827897189237936,
      "flos": 34058182631040.0,
      "grad_norm": 1.7187053018502862,
      "language_loss": 0.75731248,
      "learning_rate": 2.428753827188016e-06,
      "loss": 0.77862501,
      "num_input_tokens_seen": 159931470,
      "step": 7456,
      "time_per_iteration": 5.374706506729126
    },
    {
      "auxiliary_loss_clip": 0.01122152,
      "auxiliary_loss_mlp": 0.0103624,
      "balance_loss_clip": 1.04475796,
      "balance_loss_mlp": 1.02317476,
      "epoch": 0.44833909514504733,
      "flos": 25147416280320.0,
      "grad_norm": 2.451813000481833,
      "language_loss": 0.76448894,
      "learning_rate": 2.428373411969818e-06,
      "loss": 0.78607297,
      "num_input_tokens_seen": 159946115,
      "step": 7457,
      "time_per_iteration": 2.45881724357605
    },
    {
      "auxiliary_loss_clip": 0.01107207,
      "auxiliary_loss_mlp": 0.01035664,
      "balance_loss_clip": 1.0411675,
      "balance_loss_mlp": 1.02086377,
      "epoch": 0.4483992183977153,
      "flos": 16179984224640.0,
      "grad_norm": 1.9948400991335697,
      "language_loss": 0.68029451,
      "learning_rate": 2.4279929805076576e-06,
      "loss": 0.70172322,
      "num_input_tokens_seen": 159963915,
      "step": 7458,
      "time_per_iteration": 2.464815139770508
    },
    {
      "auxiliary_loss_clip": 0.01083854,
      "auxiliary_loss_mlp": 0.01034659,
      "balance_loss_clip": 1.04222274,
      "balance_loss_mlp": 1.02030063,
      "epoch": 0.44845934165038326,
      "flos": 17746048270080.0,
      "grad_norm": 1.9062169185049425,
      "language_loss": 0.71811426,
      "learning_rate": 2.427612532815961e-06,
      "loss": 0.73929942,
      "num_input_tokens_seen": 159982140,
      "step": 7459,
      "time_per_iteration": 2.5825932025909424
    },
    {
      "auxiliary_loss_clip": 0.01101382,
      "auxiliary_loss_mlp": 0.01035907,
      "balance_loss_clip": 1.03849745,
      "balance_loss_mlp": 1.02139306,
      "epoch": 0.4485194649030513,
      "flos": 21835914647040.0,
      "grad_norm": 1.5429270203410197,
      "language_loss": 0.695104,
      "learning_rate": 2.427232068909154e-06,
      "loss": 0.71647686,
      "num_input_tokens_seen": 160002280,
      "step": 7460,
      "time_per_iteration": 2.4890499114990234
    },
    {
      "auxiliary_loss_clip": 0.01119193,
      "auxiliary_loss_mlp": 0.01035831,
      "balance_loss_clip": 1.04026842,
      "balance_loss_mlp": 1.02229536,
      "epoch": 0.44857958815571924,
      "flos": 20084515401600.0,
      "grad_norm": 1.7863404508744487,
      "language_loss": 0.76794553,
      "learning_rate": 2.4268515888016635e-06,
      "loss": 0.78949577,
      "num_input_tokens_seen": 160020260,
      "step": 7461,
      "time_per_iteration": 2.470203399658203
    },
    {
      "auxiliary_loss_clip": 0.01119589,
      "auxiliary_loss_mlp": 0.01039295,
      "balance_loss_clip": 1.039621,
      "balance_loss_mlp": 1.026003,
      "epoch": 0.4486397114083872,
      "flos": 27053519402880.0,
      "grad_norm": 1.7809043995170295,
      "language_loss": 0.67769396,
      "learning_rate": 2.4264710925079184e-06,
      "loss": 0.69928282,
      "num_input_tokens_seen": 160040240,
      "step": 7462,
      "time_per_iteration": 2.5207231044769287
    },
    {
      "auxiliary_loss_clip": 0.01040115,
      "auxiliary_loss_mlp": 0.01008634,
      "balance_loss_clip": 1.0139997,
      "balance_loss_mlp": 1.00696516,
      "epoch": 0.4486998346610552,
      "flos": 67321195931520.0,
      "grad_norm": 0.7478405465364303,
      "language_loss": 0.54416436,
      "learning_rate": 2.4260905800423462e-06,
      "loss": 0.56465185,
      "num_input_tokens_seen": 160093865,
      "step": 7463,
      "time_per_iteration": 3.1067306995391846
    },
    {
      "auxiliary_loss_clip": 0.01109545,
      "auxiliary_loss_mlp": 0.010308,
      "balance_loss_clip": 1.04103279,
      "balance_loss_mlp": 1.0174005,
      "epoch": 0.44875995791372314,
      "flos": 27636816360960.0,
      "grad_norm": 2.7229290716490095,
      "language_loss": 0.75747603,
      "learning_rate": 2.4257100514193775e-06,
      "loss": 0.77887952,
      "num_input_tokens_seen": 160113590,
      "step": 7464,
      "time_per_iteration": 2.5568926334381104
    },
    {
      "auxiliary_loss_clip": 0.01106957,
      "auxiliary_loss_mlp": 0.01033391,
      "balance_loss_clip": 1.04170692,
      "balance_loss_mlp": 1.02176845,
      "epoch": 0.4488200811663911,
      "flos": 13005947940480.0,
      "grad_norm": 1.7728280577331545,
      "language_loss": 0.74107951,
      "learning_rate": 2.425329506653441e-06,
      "loss": 0.762483,
      "num_input_tokens_seen": 160131795,
      "step": 7465,
      "time_per_iteration": 2.4462995529174805
    },
    {
      "auxiliary_loss_clip": 0.011089,
      "auxiliary_loss_mlp": 0.01042363,
      "balance_loss_clip": 1.0481019,
      "balance_loss_mlp": 1.02694917,
      "epoch": 0.44888020441905907,
      "flos": 27489977562240.0,
      "grad_norm": 2.0001391552874077,
      "language_loss": 0.8027159,
      "learning_rate": 2.424948945758966e-06,
      "loss": 0.82422853,
      "num_input_tokens_seen": 160150635,
      "step": 7466,
      "time_per_iteration": 2.549793243408203
    },
    {
      "auxiliary_loss_clip": 0.01098831,
      "auxiliary_loss_mlp": 0.01034173,
      "balance_loss_clip": 1.0408392,
      "balance_loss_mlp": 1.02033865,
      "epoch": 0.44894032767172704,
      "flos": 18259678800000.0,
      "grad_norm": 2.4774837534927885,
      "language_loss": 0.80069566,
      "learning_rate": 2.4245683687503844e-06,
      "loss": 0.82202572,
      "num_input_tokens_seen": 160168615,
      "step": 7467,
      "time_per_iteration": 2.4681432247161865
    },
    {
      "auxiliary_loss_clip": 0.01068258,
      "auxiliary_loss_mlp": 0.0103307,
      "balance_loss_clip": 1.04233718,
      "balance_loss_mlp": 1.02067196,
      "epoch": 0.449000450924395,
      "flos": 21579835610880.0,
      "grad_norm": 1.7044031511042466,
      "language_loss": 0.75294113,
      "learning_rate": 2.424187775642129e-06,
      "loss": 0.77395451,
      "num_input_tokens_seen": 160187295,
      "step": 7468,
      "time_per_iteration": 2.571585178375244
    },
    {
      "auxiliary_loss_clip": 0.01088927,
      "auxiliary_loss_mlp": 0.01031457,
      "balance_loss_clip": 1.03822994,
      "balance_loss_mlp": 1.0192616,
      "epoch": 0.44906057417706297,
      "flos": 17967904623360.0,
      "grad_norm": 6.109489524242912,
      "language_loss": 0.70895469,
      "learning_rate": 2.4238071664486297e-06,
      "loss": 0.73015857,
      "num_input_tokens_seen": 160205115,
      "step": 7469,
      "time_per_iteration": 2.5202560424804688
    },
    {
      "auxiliary_loss_clip": 0.01106402,
      "auxiliary_loss_mlp": 0.01035682,
      "balance_loss_clip": 1.04301965,
      "balance_loss_mlp": 1.02222979,
      "epoch": 0.44912069742973093,
      "flos": 20047347803520.0,
      "grad_norm": 1.7288823165940888,
      "language_loss": 0.72220266,
      "learning_rate": 2.4234265411843203e-06,
      "loss": 0.7436235,
      "num_input_tokens_seen": 160222580,
      "step": 7470,
      "time_per_iteration": 2.478724479675293
    },
    {
      "auxiliary_loss_clip": 0.01078572,
      "auxiliary_loss_mlp": 0.01036431,
      "balance_loss_clip": 1.03886759,
      "balance_loss_mlp": 1.02201843,
      "epoch": 0.4491808206823989,
      "flos": 21033526682880.0,
      "grad_norm": 2.128938025560052,
      "language_loss": 0.7691859,
      "learning_rate": 2.423045899863634e-06,
      "loss": 0.79033595,
      "num_input_tokens_seen": 160241520,
      "step": 7471,
      "time_per_iteration": 2.5516316890716553
    },
    {
      "auxiliary_loss_clip": 0.01119139,
      "auxiliary_loss_mlp": 0.01036289,
      "balance_loss_clip": 1.04134977,
      "balance_loss_mlp": 1.02337885,
      "epoch": 0.44924094393506686,
      "flos": 22967136645120.0,
      "grad_norm": 1.7221060419447058,
      "language_loss": 0.70381987,
      "learning_rate": 2.4226652425010048e-06,
      "loss": 0.72537416,
      "num_input_tokens_seen": 160261815,
      "step": 7472,
      "time_per_iteration": 2.491719961166382
    },
    {
      "auxiliary_loss_clip": 0.01033797,
      "auxiliary_loss_mlp": 0.01006363,
      "balance_loss_clip": 1.01732326,
      "balance_loss_mlp": 1.00486088,
      "epoch": 0.4493010671877349,
      "flos": 59233467864960.0,
      "grad_norm": 0.7395344164586624,
      "language_loss": 0.61709046,
      "learning_rate": 2.4222845691108676e-06,
      "loss": 0.63749206,
      "num_input_tokens_seen": 160317070,
      "step": 7473,
      "time_per_iteration": 3.0657989978790283
    },
    {
      "auxiliary_loss_clip": 0.01119872,
      "auxiliary_loss_mlp": 0.00788652,
      "balance_loss_clip": 1.04126847,
      "balance_loss_mlp": 1.01353598,
      "epoch": 0.44936119044040285,
      "flos": 18004892653440.0,
      "grad_norm": 4.308204085421883,
      "language_loss": 0.77862155,
      "learning_rate": 2.421903879707657e-06,
      "loss": 0.79770678,
      "num_input_tokens_seen": 160334980,
      "step": 7474,
      "time_per_iteration": 2.442902088165283
    },
    {
      "auxiliary_loss_clip": 0.01076731,
      "auxiliary_loss_mlp": 0.01040008,
      "balance_loss_clip": 1.04098356,
      "balance_loss_mlp": 1.02591181,
      "epoch": 0.4494213136930708,
      "flos": 21251827589760.0,
      "grad_norm": 1.6535241718371663,
      "language_loss": 0.72148156,
      "learning_rate": 2.4215231743058086e-06,
      "loss": 0.74264896,
      "num_input_tokens_seen": 160354500,
      "step": 7475,
      "time_per_iteration": 2.5746469497680664
    },
    {
      "auxiliary_loss_clip": 0.01071706,
      "auxiliary_loss_mlp": 0.0103666,
      "balance_loss_clip": 1.03923965,
      "balance_loss_mlp": 1.02244401,
      "epoch": 0.4494814369457388,
      "flos": 27418695022080.0,
      "grad_norm": 1.8705457211832148,
      "language_loss": 0.77214348,
      "learning_rate": 2.4211424529197594e-06,
      "loss": 0.79322708,
      "num_input_tokens_seen": 160373650,
      "step": 7476,
      "time_per_iteration": 2.6316397190093994
    },
    {
      "auxiliary_loss_clip": 0.0111338,
      "auxiliary_loss_mlp": 0.00791896,
      "balance_loss_clip": 1.04155481,
      "balance_loss_mlp": 1.01589036,
      "epoch": 0.44954156019840674,
      "flos": 22854053652480.0,
      "grad_norm": 2.229506384022198,
      "language_loss": 0.72138512,
      "learning_rate": 2.4207617155639464e-06,
      "loss": 0.74043787,
      "num_input_tokens_seen": 160393430,
      "step": 7477,
      "time_per_iteration": 2.4935190677642822
    },
    {
      "auxiliary_loss_clip": 0.01096664,
      "auxiliary_loss_mlp": 0.01040295,
      "balance_loss_clip": 1.03879094,
      "balance_loss_mlp": 1.02526259,
      "epoch": 0.4496016834510747,
      "flos": 17201570935680.0,
      "grad_norm": 2.4739339032451433,
      "language_loss": 0.68029106,
      "learning_rate": 2.4203809622528062e-06,
      "loss": 0.70166063,
      "num_input_tokens_seen": 160410545,
      "step": 7478,
      "time_per_iteration": 2.488309621810913
    },
    {
      "auxiliary_loss_clip": 0.01092842,
      "auxiliary_loss_mlp": 0.01034262,
      "balance_loss_clip": 1.04334641,
      "balance_loss_mlp": 1.02194726,
      "epoch": 0.4496618067037427,
      "flos": 18916628595840.0,
      "grad_norm": 1.8749234175382625,
      "language_loss": 0.89341116,
      "learning_rate": 2.420000193000779e-06,
      "loss": 0.91468227,
      "num_input_tokens_seen": 160428105,
      "step": 7479,
      "time_per_iteration": 2.5018234252929688
    },
    {
      "auxiliary_loss_clip": 0.01067555,
      "auxiliary_loss_mlp": 0.0103776,
      "balance_loss_clip": 1.04417574,
      "balance_loss_mlp": 1.02324021,
      "epoch": 0.44972192995641064,
      "flos": 21031659175680.0,
      "grad_norm": 1.7576030506875309,
      "language_loss": 0.75431848,
      "learning_rate": 2.419619407822302e-06,
      "loss": 0.77537167,
      "num_input_tokens_seen": 160448815,
      "step": 7480,
      "time_per_iteration": 2.608046531677246
    },
    {
      "auxiliary_loss_clip": 0.01087534,
      "auxiliary_loss_mlp": 0.01034706,
      "balance_loss_clip": 1.04003787,
      "balance_loss_mlp": 1.02046037,
      "epoch": 0.4497820532090786,
      "flos": 20777088510720.0,
      "grad_norm": 2.1611678168009516,
      "language_loss": 0.79406643,
      "learning_rate": 2.419238606731815e-06,
      "loss": 0.8152889,
      "num_input_tokens_seen": 160465940,
      "step": 7481,
      "time_per_iteration": 2.5168232917785645
    },
    {
      "auxiliary_loss_clip": 0.01094823,
      "auxiliary_loss_mlp": 0.01032792,
      "balance_loss_clip": 1.04051697,
      "balance_loss_mlp": 1.01870728,
      "epoch": 0.44984217646174657,
      "flos": 33802606385280.0,
      "grad_norm": 1.6457344749497105,
      "language_loss": 0.68493736,
      "learning_rate": 2.418857789743758e-06,
      "loss": 0.70621353,
      "num_input_tokens_seen": 160486710,
      "step": 7482,
      "time_per_iteration": 2.613288164138794
    },
    {
      "auxiliary_loss_clip": 0.01112768,
      "auxiliary_loss_mlp": 0.01041946,
      "balance_loss_clip": 1.04410338,
      "balance_loss_mlp": 1.02814198,
      "epoch": 0.44990229971441453,
      "flos": 15518365660800.0,
      "grad_norm": 2.328342859930525,
      "language_loss": 0.84898472,
      "learning_rate": 2.418476956872571e-06,
      "loss": 0.87053186,
      "num_input_tokens_seen": 160503405,
      "step": 7483,
      "time_per_iteration": 2.4353723526000977
    },
    {
      "auxiliary_loss_clip": 0.010922,
      "auxiliary_loss_mlp": 0.01039036,
      "balance_loss_clip": 1.0427084,
      "balance_loss_mlp": 1.02442718,
      "epoch": 0.4499624229670825,
      "flos": 29861913191040.0,
      "grad_norm": 1.6508699740926254,
      "language_loss": 0.80455148,
      "learning_rate": 2.4180961081326967e-06,
      "loss": 0.8258639,
      "num_input_tokens_seen": 160525080,
      "step": 7484,
      "time_per_iteration": 2.5738296508789062
    },
    {
      "auxiliary_loss_clip": 0.01071435,
      "auxiliary_loss_mlp": 0.01030682,
      "balance_loss_clip": 1.04156017,
      "balance_loss_mlp": 1.01557827,
      "epoch": 0.45002254621975046,
      "flos": 18513674847360.0,
      "grad_norm": 3.067625360631521,
      "language_loss": 0.7501753,
      "learning_rate": 2.4177152435385754e-06,
      "loss": 0.77119648,
      "num_input_tokens_seen": 160540895,
      "step": 7485,
      "time_per_iteration": 2.5476644039154053
    },
    {
      "auxiliary_loss_clip": 0.01029676,
      "auxiliary_loss_mlp": 0.01003487,
      "balance_loss_clip": 1.02202952,
      "balance_loss_mlp": 1.00182998,
      "epoch": 0.4500826694724185,
      "flos": 70420394229120.0,
      "grad_norm": 0.8104522579117598,
      "language_loss": 0.5866369,
      "learning_rate": 2.4173343631046504e-06,
      "loss": 0.60696852,
      "num_input_tokens_seen": 160598270,
      "step": 7486,
      "time_per_iteration": 3.140064239501953
    },
    {
      "auxiliary_loss_clip": 0.01104297,
      "auxiliary_loss_mlp": 0.01038423,
      "balance_loss_clip": 1.03971446,
      "balance_loss_mlp": 1.02370667,
      "epoch": 0.45014279272508645,
      "flos": 15778897983360.0,
      "grad_norm": 1.9863082933087652,
      "language_loss": 0.83193159,
      "learning_rate": 2.4169534668453654e-06,
      "loss": 0.85335875,
      "num_input_tokens_seen": 160614720,
      "step": 7487,
      "time_per_iteration": 2.433023691177368
    },
    {
      "auxiliary_loss_clip": 0.0111887,
      "auxiliary_loss_mlp": 0.01032526,
      "balance_loss_clip": 1.04187942,
      "balance_loss_mlp": 1.01881707,
      "epoch": 0.4502029159777544,
      "flos": 21799573061760.0,
      "grad_norm": 1.5213904540863725,
      "language_loss": 0.77486223,
      "learning_rate": 2.4165725547751622e-06,
      "loss": 0.79637623,
      "num_input_tokens_seen": 160635170,
      "step": 7488,
      "time_per_iteration": 2.5049641132354736
    },
    {
      "auxiliary_loss_clip": 0.01119658,
      "auxiliary_loss_mlp": 0.01034776,
      "balance_loss_clip": 1.04513001,
      "balance_loss_mlp": 1.02030993,
      "epoch": 0.4502630392304224,
      "flos": 28767966531840.0,
      "grad_norm": 2.060698080381875,
      "language_loss": 0.71967602,
      "learning_rate": 2.4161916269084858e-06,
      "loss": 0.74122036,
      "num_input_tokens_seen": 160654490,
      "step": 7489,
      "time_per_iteration": 3.967219114303589
    },
    {
      "auxiliary_loss_clip": 0.01103419,
      "auxiliary_loss_mlp": 0.01035264,
      "balance_loss_clip": 1.04631448,
      "balance_loss_mlp": 1.01985013,
      "epoch": 0.45032316248309034,
      "flos": 15844182952320.0,
      "grad_norm": 2.1485720870561438,
      "language_loss": 0.69158375,
      "learning_rate": 2.4158106832597817e-06,
      "loss": 0.71297061,
      "num_input_tokens_seen": 160669400,
      "step": 7490,
      "time_per_iteration": 3.9282257556915283
    },
    {
      "auxiliary_loss_clip": 0.01025426,
      "auxiliary_loss_mlp": 0.01009842,
      "balance_loss_clip": 1.02566445,
      "balance_loss_mlp": 1.00826859,
      "epoch": 0.4503832857357583,
      "flos": 57853600945920.0,
      "grad_norm": 0.7702191598155552,
      "language_loss": 0.56740868,
      "learning_rate": 2.415429723843495e-06,
      "loss": 0.58776134,
      "num_input_tokens_seen": 160733820,
      "step": 7491,
      "time_per_iteration": 3.0926125049591064
    },
    {
      "auxiliary_loss_clip": 0.01104112,
      "auxiliary_loss_mlp": 0.01028262,
      "balance_loss_clip": 1.044554,
      "balance_loss_mlp": 1.01510119,
      "epoch": 0.4504434089884263,
      "flos": 23878082488320.0,
      "grad_norm": 1.7178820345370165,
      "language_loss": 0.79525363,
      "learning_rate": 2.4150487486740713e-06,
      "loss": 0.81657743,
      "num_input_tokens_seen": 160753175,
      "step": 7492,
      "time_per_iteration": 2.52463436126709
    },
    {
      "auxiliary_loss_clip": 0.01089897,
      "auxiliary_loss_mlp": 0.00792965,
      "balance_loss_clip": 1.04204929,
      "balance_loss_mlp": 1.01445484,
      "epoch": 0.45050353224109424,
      "flos": 17785083375360.0,
      "grad_norm": 2.3949937946844173,
      "language_loss": 0.9296087,
      "learning_rate": 2.4146677577659573e-06,
      "loss": 0.94843733,
      "num_input_tokens_seen": 160768310,
      "step": 7493,
      "time_per_iteration": 2.514392852783203
    },
    {
      "auxiliary_loss_clip": 0.01034166,
      "auxiliary_loss_mlp": 0.01002672,
      "balance_loss_clip": 1.01782846,
      "balance_loss_mlp": 1.00103903,
      "epoch": 0.4505636554937622,
      "flos": 65063420703360.0,
      "grad_norm": 0.7967236880125421,
      "language_loss": 0.62852103,
      "learning_rate": 2.4142867511336e-06,
      "loss": 0.64888936,
      "num_input_tokens_seen": 160827370,
      "step": 7494,
      "time_per_iteration": 4.554858207702637
    },
    {
      "auxiliary_loss_clip": 0.01119423,
      "auxiliary_loss_mlp": 0.01029227,
      "balance_loss_clip": 1.04248643,
      "balance_loss_mlp": 1.01628661,
      "epoch": 0.45062377874643017,
      "flos": 22200084685440.0,
      "grad_norm": 1.9790247069796794,
      "language_loss": 0.82277077,
      "learning_rate": 2.4139057287914484e-06,
      "loss": 0.84425724,
      "num_input_tokens_seen": 160849140,
      "step": 7495,
      "time_per_iteration": 3.8982043266296387
    },
    {
      "auxiliary_loss_clip": 0.01104936,
      "auxiliary_loss_mlp": 0.0103463,
      "balance_loss_clip": 1.04019189,
      "balance_loss_mlp": 1.01922822,
      "epoch": 0.45068390199909814,
      "flos": 37670293186560.0,
      "grad_norm": 1.8698243006733988,
      "language_loss": 0.85662889,
      "learning_rate": 2.41352469075395e-06,
      "loss": 0.87802452,
      "num_input_tokens_seen": 160871280,
      "step": 7496,
      "time_per_iteration": 2.6323671340942383
    },
    {
      "auxiliary_loss_clip": 0.01121892,
      "auxiliary_loss_mlp": 0.01030328,
      "balance_loss_clip": 1.04297256,
      "balance_loss_mlp": 1.0170238,
      "epoch": 0.4507440252517661,
      "flos": 22302501338880.0,
      "grad_norm": 2.222316084680107,
      "language_loss": 0.76726997,
      "learning_rate": 2.4131436370355534e-06,
      "loss": 0.78879213,
      "num_input_tokens_seen": 160888625,
      "step": 7497,
      "time_per_iteration": 2.4628498554229736
    },
    {
      "auxiliary_loss_clip": 0.01088457,
      "auxiliary_loss_mlp": 0.01032269,
      "balance_loss_clip": 1.03957665,
      "balance_loss_mlp": 1.01873875,
      "epoch": 0.45080414850443407,
      "flos": 13188374138880.0,
      "grad_norm": 1.9030259542056154,
      "language_loss": 0.74904305,
      "learning_rate": 2.4127625676507088e-06,
      "loss": 0.77025032,
      "num_input_tokens_seen": 160907040,
      "step": 7498,
      "time_per_iteration": 2.506969690322876
    },
    {
      "auxiliary_loss_clip": 0.01120582,
      "auxiliary_loss_mlp": 0.01040867,
      "balance_loss_clip": 1.04171968,
      "balance_loss_mlp": 1.02703834,
      "epoch": 0.4508642717571021,
      "flos": 21944939402880.0,
      "grad_norm": 1.9345944622319866,
      "language_loss": 0.70223612,
      "learning_rate": 2.4123814826138663e-06,
      "loss": 0.72385061,
      "num_input_tokens_seen": 160927115,
      "step": 7499,
      "time_per_iteration": 2.4582290649414062
    },
    {
      "auxiliary_loss_clip": 0.01078381,
      "auxiliary_loss_mlp": 0.01033924,
      "balance_loss_clip": 1.04080057,
      "balance_loss_mlp": 1.02008402,
      "epoch": 0.45092439500977005,
      "flos": 23367468700800.0,
      "grad_norm": 1.8952890024017854,
      "language_loss": 0.77068257,
      "learning_rate": 2.412000381939477e-06,
      "loss": 0.79180562,
      "num_input_tokens_seen": 160944405,
      "step": 7500,
      "time_per_iteration": 2.5586631298065186
    },
    {
      "auxiliary_loss_clip": 0.01076981,
      "auxiliary_loss_mlp": 0.01032123,
      "balance_loss_clip": 1.04256213,
      "balance_loss_mlp": 1.01890874,
      "epoch": 0.450984518262438,
      "flos": 20772958446720.0,
      "grad_norm": 1.7559589938866471,
      "language_loss": 0.62458837,
      "learning_rate": 2.411619265641992e-06,
      "loss": 0.64567947,
      "num_input_tokens_seen": 160961345,
      "step": 7501,
      "time_per_iteration": 2.566627264022827
    },
    {
      "auxiliary_loss_clip": 0.01123118,
      "auxiliary_loss_mlp": 0.01032539,
      "balance_loss_clip": 1.04224491,
      "balance_loss_mlp": 1.018031,
      "epoch": 0.451044641515106,
      "flos": 17707372300800.0,
      "grad_norm": 2.019327563735736,
      "language_loss": 0.84837198,
      "learning_rate": 2.411238133735863e-06,
      "loss": 0.86992854,
      "num_input_tokens_seen": 160977330,
      "step": 7502,
      "time_per_iteration": 2.4307401180267334
    },
    {
      "auxiliary_loss_clip": 0.01094173,
      "auxiliary_loss_mlp": 0.0103556,
      "balance_loss_clip": 1.04083872,
      "balance_loss_mlp": 1.02245927,
      "epoch": 0.45110476476777395,
      "flos": 20594698225920.0,
      "grad_norm": 1.4200880764343613,
      "language_loss": 0.79928786,
      "learning_rate": 2.4108569862355418e-06,
      "loss": 0.82058513,
      "num_input_tokens_seen": 160997280,
      "step": 7503,
      "time_per_iteration": 2.52477765083313
    },
    {
      "auxiliary_loss_clip": 0.01101403,
      "auxiliary_loss_mlp": 0.01033581,
      "balance_loss_clip": 1.04441881,
      "balance_loss_mlp": 1.01978815,
      "epoch": 0.4511648880204419,
      "flos": 16034043265920.0,
      "grad_norm": 1.789525204578375,
      "language_loss": 0.81026232,
      "learning_rate": 2.410475823155484e-06,
      "loss": 0.83161217,
      "num_input_tokens_seen": 161014235,
      "step": 7504,
      "time_per_iteration": 2.4716033935546875
    },
    {
      "auxiliary_loss_clip": 0.01086658,
      "auxiliary_loss_mlp": 0.01033315,
      "balance_loss_clip": 1.03950596,
      "balance_loss_mlp": 1.0203985,
      "epoch": 0.4512250112731099,
      "flos": 23978811202560.0,
      "grad_norm": 1.7462433425114166,
      "language_loss": 0.63452345,
      "learning_rate": 2.4100946445101405e-06,
      "loss": 0.65572315,
      "num_input_tokens_seen": 161032360,
      "step": 7505,
      "time_per_iteration": 2.5522115230560303
    },
    {
      "auxiliary_loss_clip": 0.0101216,
      "auxiliary_loss_mlp": 0.01004,
      "balance_loss_clip": 1.02224612,
      "balance_loss_mlp": 1.00242662,
      "epoch": 0.45128513452577784,
      "flos": 71462308037760.0,
      "grad_norm": 0.8389431473825758,
      "language_loss": 0.588947,
      "learning_rate": 2.409713450313968e-06,
      "loss": 0.60910857,
      "num_input_tokens_seen": 161091360,
      "step": 7506,
      "time_per_iteration": 3.214512586593628
    },
    {
      "auxiliary_loss_clip": 0.01070155,
      "auxiliary_loss_mlp": 0.01038292,
      "balance_loss_clip": 1.04112959,
      "balance_loss_mlp": 1.02437425,
      "epoch": 0.4513452577784458,
      "flos": 22090844448000.0,
      "grad_norm": 1.5936984939734624,
      "language_loss": 0.79402089,
      "learning_rate": 2.40933224058142e-06,
      "loss": 0.81510544,
      "num_input_tokens_seen": 161110825,
      "step": 7507,
      "time_per_iteration": 2.5835394859313965
    },
    {
      "auxiliary_loss_clip": 0.0108201,
      "auxiliary_loss_mlp": 0.01031067,
      "balance_loss_clip": 1.04009628,
      "balance_loss_mlp": 1.01608253,
      "epoch": 0.4514053810311138,
      "flos": 24276403382400.0,
      "grad_norm": 1.4778827396728555,
      "language_loss": 0.73775184,
      "learning_rate": 2.4089510153269526e-06,
      "loss": 0.75888264,
      "num_input_tokens_seen": 161130685,
      "step": 7508,
      "time_per_iteration": 2.5691139698028564
    },
    {
      "auxiliary_loss_clip": 0.011081,
      "auxiliary_loss_mlp": 0.0103431,
      "balance_loss_clip": 1.04330087,
      "balance_loss_mlp": 1.02167964,
      "epoch": 0.45146550428378174,
      "flos": 17886781756800.0,
      "grad_norm": 1.9257403990498172,
      "language_loss": 0.79665279,
      "learning_rate": 2.4085697745650217e-06,
      "loss": 0.81807691,
      "num_input_tokens_seen": 161147555,
      "step": 7509,
      "time_per_iteration": 2.4637935161590576
    },
    {
      "auxiliary_loss_clip": 0.01119243,
      "auxiliary_loss_mlp": 0.0103478,
      "balance_loss_clip": 1.04215002,
      "balance_loss_mlp": 1.02162504,
      "epoch": 0.4515256275364497,
      "flos": 24243437675520.0,
      "grad_norm": 1.8399599248982386,
      "language_loss": 0.73125845,
      "learning_rate": 2.4081885183100837e-06,
      "loss": 0.75279868,
      "num_input_tokens_seen": 161166255,
      "step": 7510,
      "time_per_iteration": 2.4808263778686523
    },
    {
      "auxiliary_loss_clip": 0.01120064,
      "auxiliary_loss_mlp": 0.01032065,
      "balance_loss_clip": 1.04056096,
      "balance_loss_mlp": 1.01789093,
      "epoch": 0.45158575078911767,
      "flos": 20631039811200.0,
      "grad_norm": 1.7858241992223973,
      "language_loss": 0.76907313,
      "learning_rate": 2.4078072465765964e-06,
      "loss": 0.79059434,
      "num_input_tokens_seen": 161184720,
      "step": 7511,
      "time_per_iteration": 2.4483795166015625
    },
    {
      "auxiliary_loss_clip": 0.01109253,
      "auxiliary_loss_mlp": 0.01034227,
      "balance_loss_clip": 1.04092622,
      "balance_loss_mlp": 1.02057779,
      "epoch": 0.45164587404178563,
      "flos": 23327751237120.0,
      "grad_norm": 1.4893286749867192,
      "language_loss": 0.78675687,
      "learning_rate": 2.4074259593790174e-06,
      "loss": 0.80819166,
      "num_input_tokens_seen": 161204360,
      "step": 7512,
      "time_per_iteration": 2.4978139400482178
    },
    {
      "auxiliary_loss_clip": 0.01088876,
      "auxiliary_loss_mlp": 0.01036313,
      "balance_loss_clip": 1.04173017,
      "balance_loss_mlp": 1.02194786,
      "epoch": 0.45170599729445365,
      "flos": 23805973935360.0,
      "grad_norm": 2.54205141765288,
      "language_loss": 0.87497962,
      "learning_rate": 2.4070446567318053e-06,
      "loss": 0.89623153,
      "num_input_tokens_seen": 161223575,
      "step": 7513,
      "time_per_iteration": 2.5627903938293457
    },
    {
      "auxiliary_loss_clip": 0.01103741,
      "auxiliary_loss_mlp": 0.01030639,
      "balance_loss_clip": 1.04105401,
      "balance_loss_mlp": 1.01830077,
      "epoch": 0.4517661205471216,
      "flos": 23512942782720.0,
      "grad_norm": 1.5867246989468144,
      "language_loss": 0.67499959,
      "learning_rate": 2.406663338649419e-06,
      "loss": 0.69634336,
      "num_input_tokens_seen": 161243805,
      "step": 7514,
      "time_per_iteration": 2.513930082321167
    },
    {
      "auxiliary_loss_clip": 0.01106967,
      "auxiliary_loss_mlp": 0.01030483,
      "balance_loss_clip": 1.04376233,
      "balance_loss_mlp": 1.01517105,
      "epoch": 0.4518262437997896,
      "flos": 23513948363520.0,
      "grad_norm": 1.74672358084882,
      "language_loss": 0.69302464,
      "learning_rate": 2.406282005146318e-06,
      "loss": 0.71439916,
      "num_input_tokens_seen": 161261450,
      "step": 7515,
      "time_per_iteration": 2.512010097503662
    },
    {
      "auxiliary_loss_clip": 0.01109011,
      "auxiliary_loss_mlp": 0.01040219,
      "balance_loss_clip": 1.04158652,
      "balance_loss_mlp": 1.02571118,
      "epoch": 0.45188636705245755,
      "flos": 14568061489920.0,
      "grad_norm": 2.245857008777817,
      "language_loss": 0.81849134,
      "learning_rate": 2.405900656236963e-06,
      "loss": 0.8399837,
      "num_input_tokens_seen": 161276965,
      "step": 7516,
      "time_per_iteration": 2.4500176906585693
    },
    {
      "auxiliary_loss_clip": 0.01118166,
      "auxiliary_loss_mlp": 0.01035031,
      "balance_loss_clip": 1.04253376,
      "balance_loss_mlp": 1.02116108,
      "epoch": 0.4519464903051255,
      "flos": 19901550499200.0,
      "grad_norm": 1.5550919372812402,
      "language_loss": 0.65483129,
      "learning_rate": 2.4055192919358137e-06,
      "loss": 0.67636323,
      "num_input_tokens_seen": 161295375,
      "step": 7517,
      "time_per_iteration": 2.4442341327667236
    },
    {
      "auxiliary_loss_clip": 0.01084421,
      "auxiliary_loss_mlp": 0.01025559,
      "balance_loss_clip": 1.04027069,
      "balance_loss_mlp": 1.0133636,
      "epoch": 0.4520066135577935,
      "flos": 18844376388480.0,
      "grad_norm": 1.9210714409731768,
      "language_loss": 0.62749964,
      "learning_rate": 2.405137912257333e-06,
      "loss": 0.64859939,
      "num_input_tokens_seen": 161313010,
      "step": 7518,
      "time_per_iteration": 2.535388231277466
    },
    {
      "auxiliary_loss_clip": 0.01107336,
      "auxiliary_loss_mlp": 0.01036308,
      "balance_loss_clip": 1.0416199,
      "balance_loss_mlp": 1.02332616,
      "epoch": 0.45206673681046144,
      "flos": 48214419713280.0,
      "grad_norm": 1.4327375991822382,
      "language_loss": 0.59508288,
      "learning_rate": 2.404756517215982e-06,
      "loss": 0.61651933,
      "num_input_tokens_seen": 161336690,
      "step": 7519,
      "time_per_iteration": 2.71445369720459
    },
    {
      "auxiliary_loss_clip": 0.01112967,
      "auxiliary_loss_mlp": 0.01039751,
      "balance_loss_clip": 1.04477262,
      "balance_loss_mlp": 1.02606535,
      "epoch": 0.4521268600631294,
      "flos": 23842171866240.0,
      "grad_norm": 1.3323757611159335,
      "language_loss": 0.72354436,
      "learning_rate": 2.404375106826223e-06,
      "loss": 0.74507159,
      "num_input_tokens_seen": 161357845,
      "step": 7520,
      "time_per_iteration": 2.5167789459228516
    },
    {
      "auxiliary_loss_clip": 0.01098931,
      "auxiliary_loss_mlp": 0.01036055,
      "balance_loss_clip": 1.04142475,
      "balance_loss_mlp": 1.02310896,
      "epoch": 0.4521869833157974,
      "flos": 18843622202880.0,
      "grad_norm": 1.810369498044126,
      "language_loss": 0.75555933,
      "learning_rate": 2.4039936811025194e-06,
      "loss": 0.77690923,
      "num_input_tokens_seen": 161375160,
      "step": 7521,
      "time_per_iteration": 2.468827724456787
    },
    {
      "auxiliary_loss_clip": 0.0110631,
      "auxiliary_loss_mlp": 0.0103904,
      "balance_loss_clip": 1.04486918,
      "balance_loss_mlp": 1.024616,
      "epoch": 0.45224710656846534,
      "flos": 19788072456960.0,
      "grad_norm": 1.667657404984478,
      "language_loss": 0.675982,
      "learning_rate": 2.4036122400593343e-06,
      "loss": 0.6974355,
      "num_input_tokens_seen": 161393690,
      "step": 7522,
      "time_per_iteration": 2.5115177631378174
    },
    {
      "auxiliary_loss_clip": 0.01106628,
      "auxiliary_loss_mlp": 0.01032874,
      "balance_loss_clip": 1.0411787,
      "balance_loss_mlp": 1.01934421,
      "epoch": 0.4523072298211333,
      "flos": 28256131681920.0,
      "grad_norm": 1.4492898570671542,
      "language_loss": 0.60874474,
      "learning_rate": 2.403230783711134e-06,
      "loss": 0.63013971,
      "num_input_tokens_seen": 161415015,
      "step": 7523,
      "time_per_iteration": 2.5834474563598633
    },
    {
      "auxiliary_loss_clip": 0.01113231,
      "auxiliary_loss_mlp": 0.01035928,
      "balance_loss_clip": 1.0421139,
      "balance_loss_mlp": 1.02142024,
      "epoch": 0.45236735307380127,
      "flos": 11181039511680.0,
      "grad_norm": 1.9161846680095935,
      "language_loss": 0.78269398,
      "learning_rate": 2.4028493120723813e-06,
      "loss": 0.80418557,
      "num_input_tokens_seen": 161432940,
      "step": 7524,
      "time_per_iteration": 2.4899485111236572
    },
    {
      "auxiliary_loss_clip": 0.0107283,
      "auxiliary_loss_mlp": 0.01034809,
      "balance_loss_clip": 1.038746,
      "balance_loss_mlp": 1.02120709,
      "epoch": 0.45242747632646924,
      "flos": 22601386408320.0,
      "grad_norm": 1.7066201955640377,
      "language_loss": 0.63744146,
      "learning_rate": 2.4024678251575417e-06,
      "loss": 0.65851784,
      "num_input_tokens_seen": 161452215,
      "step": 7525,
      "time_per_iteration": 2.57053804397583
    },
    {
      "auxiliary_loss_clip": 0.01108061,
      "auxiliary_loss_mlp": 0.01034486,
      "balance_loss_clip": 1.04107881,
      "balance_loss_mlp": 1.02183247,
      "epoch": 0.45248759957913726,
      "flos": 18256267008000.0,
      "grad_norm": 1.5582623761363177,
      "language_loss": 0.79091644,
      "learning_rate": 2.402086322981083e-06,
      "loss": 0.81234193,
      "num_input_tokens_seen": 161469520,
      "step": 7526,
      "time_per_iteration": 2.458824396133423
    },
    {
      "auxiliary_loss_clip": 0.01092123,
      "auxiliary_loss_mlp": 0.01031904,
      "balance_loss_clip": 1.03953981,
      "balance_loss_mlp": 1.01807523,
      "epoch": 0.4525477228318052,
      "flos": 22450094323200.0,
      "grad_norm": 1.5863929153520704,
      "language_loss": 0.80946767,
      "learning_rate": 2.40170480555747e-06,
      "loss": 0.83070791,
      "num_input_tokens_seen": 161487335,
      "step": 7527,
      "time_per_iteration": 2.587440252304077
    },
    {
      "auxiliary_loss_clip": 0.01091987,
      "auxiliary_loss_mlp": 0.01029398,
      "balance_loss_clip": 1.04374051,
      "balance_loss_mlp": 1.01555192,
      "epoch": 0.4526078460844732,
      "flos": 29644869260160.0,
      "grad_norm": 1.4320675628203965,
      "language_loss": 0.65507007,
      "learning_rate": 2.4013232729011706e-06,
      "loss": 0.67628396,
      "num_input_tokens_seen": 161510095,
      "step": 7528,
      "time_per_iteration": 3.992781162261963
    },
    {
      "auxiliary_loss_clip": 0.01096222,
      "auxiliary_loss_mlp": 0.0103207,
      "balance_loss_clip": 1.04113615,
      "balance_loss_mlp": 1.01856399,
      "epoch": 0.45266796933714115,
      "flos": 23039747988480.0,
      "grad_norm": 1.52337438696075,
      "language_loss": 0.75711334,
      "learning_rate": 2.4009417250266525e-06,
      "loss": 0.77839625,
      "num_input_tokens_seen": 161528725,
      "step": 7529,
      "time_per_iteration": 3.9235610961914062
    },
    {
      "auxiliary_loss_clip": 0.01117994,
      "auxiliary_loss_mlp": 0.01032836,
      "balance_loss_clip": 1.04041648,
      "balance_loss_mlp": 1.01945496,
      "epoch": 0.4527280925898091,
      "flos": 14428405411200.0,
      "grad_norm": 1.857760500987767,
      "language_loss": 0.72947562,
      "learning_rate": 2.400560161948384e-06,
      "loss": 0.75098395,
      "num_input_tokens_seen": 161547195,
      "step": 7530,
      "time_per_iteration": 2.4253456592559814
    },
    {
      "auxiliary_loss_clip": 0.01084435,
      "auxiliary_loss_mlp": 0.01031566,
      "balance_loss_clip": 1.04124618,
      "balance_loss_mlp": 1.01854873,
      "epoch": 0.4527882158424771,
      "flos": 22925515760640.0,
      "grad_norm": 1.6697923639223333,
      "language_loss": 0.75818348,
      "learning_rate": 2.400178583680834e-06,
      "loss": 0.77934349,
      "num_input_tokens_seen": 161565565,
      "step": 7531,
      "time_per_iteration": 2.556476354598999
    },
    {
      "auxiliary_loss_clip": 0.01115238,
      "auxiliary_loss_mlp": 0.01034153,
      "balance_loss_clip": 1.04102397,
      "balance_loss_mlp": 1.02039027,
      "epoch": 0.45284833909514505,
      "flos": 25555326105600.0,
      "grad_norm": 1.4244616415112288,
      "language_loss": 0.66768533,
      "learning_rate": 2.3997969902384717e-06,
      "loss": 0.68917924,
      "num_input_tokens_seen": 161586630,
      "step": 7532,
      "time_per_iteration": 2.491196393966675
    },
    {
      "auxiliary_loss_clip": 0.01105093,
      "auxiliary_loss_mlp": 0.01035699,
      "balance_loss_clip": 1.04056692,
      "balance_loss_mlp": 1.02231169,
      "epoch": 0.452908462347813,
      "flos": 18150007599360.0,
      "grad_norm": 1.9266768190834322,
      "language_loss": 0.78542423,
      "learning_rate": 2.399415381635768e-06,
      "loss": 0.80683208,
      "num_input_tokens_seen": 161603815,
      "step": 7533,
      "time_per_iteration": 3.8915374279022217
    },
    {
      "auxiliary_loss_clip": 0.01091352,
      "auxiliary_loss_mlp": 0.01035265,
      "balance_loss_clip": 1.03835964,
      "balance_loss_mlp": 1.02066791,
      "epoch": 0.452968585600481,
      "flos": 19062749122560.0,
      "grad_norm": 1.725536360241159,
      "language_loss": 0.83097637,
      "learning_rate": 2.3990337578871927e-06,
      "loss": 0.85224253,
      "num_input_tokens_seen": 161622900,
      "step": 7534,
      "time_per_iteration": 4.0086798667907715
    },
    {
      "auxiliary_loss_clip": 0.0109097,
      "auxiliary_loss_mlp": 0.01033708,
      "balance_loss_clip": 1.04400849,
      "balance_loss_mlp": 1.01937866,
      "epoch": 0.45302870885314894,
      "flos": 22051737515520.0,
      "grad_norm": 1.5220336312267768,
      "language_loss": 0.76409733,
      "learning_rate": 2.3986521190072176e-06,
      "loss": 0.78534406,
      "num_input_tokens_seen": 161641700,
      "step": 7535,
      "time_per_iteration": 2.516554832458496
    },
    {
      "auxiliary_loss_clip": 0.01078579,
      "auxiliary_loss_mlp": 0.01034381,
      "balance_loss_clip": 1.04196,
      "balance_loss_mlp": 1.02160788,
      "epoch": 0.4530888321058169,
      "flos": 20376217751040.0,
      "grad_norm": 1.5872489556624223,
      "language_loss": 0.80409974,
      "learning_rate": 2.3982704650103138e-06,
      "loss": 0.82522935,
      "num_input_tokens_seen": 161661955,
      "step": 7536,
      "time_per_iteration": 2.572300434112549
    },
    {
      "auxiliary_loss_clip": 0.01087391,
      "auxiliary_loss_mlp": 0.01034211,
      "balance_loss_clip": 1.03667331,
      "balance_loss_mlp": 1.02067506,
      "epoch": 0.4531489553584849,
      "flos": 14830425406080.0,
      "grad_norm": 1.7209322910412848,
      "language_loss": 0.75941736,
      "learning_rate": 2.3978887959109544e-06,
      "loss": 0.78063339,
      "num_input_tokens_seen": 161679245,
      "step": 7537,
      "time_per_iteration": 2.4924445152282715
    },
    {
      "auxiliary_loss_clip": 0.01108078,
      "auxiliary_loss_mlp": 0.0103189,
      "balance_loss_clip": 1.04024982,
      "balance_loss_mlp": 1.01939058,
      "epoch": 0.45320907861115284,
      "flos": 21944975316480.0,
      "grad_norm": 1.8136449024325656,
      "language_loss": 0.75979364,
      "learning_rate": 2.3975071117236118e-06,
      "loss": 0.78119332,
      "num_input_tokens_seen": 161698795,
      "step": 7538,
      "time_per_iteration": 2.495335817337036
    },
    {
      "auxiliary_loss_clip": 0.01034954,
      "auxiliary_loss_mlp": 0.01009332,
      "balance_loss_clip": 1.01863205,
      "balance_loss_mlp": 1.00791359,
      "epoch": 0.45326920186382086,
      "flos": 66251455038720.0,
      "grad_norm": 0.7896755772819237,
      "language_loss": 0.62403715,
      "learning_rate": 2.3971254124627593e-06,
      "loss": 0.64447999,
      "num_input_tokens_seen": 161761980,
      "step": 7539,
      "time_per_iteration": 3.1319046020507812
    },
    {
      "auxiliary_loss_clip": 0.01119866,
      "auxiliary_loss_mlp": 0.01041231,
      "balance_loss_clip": 1.04283237,
      "balance_loss_mlp": 1.02815986,
      "epoch": 0.4533293251164888,
      "flos": 14684233052160.0,
      "grad_norm": 1.8917120101010776,
      "language_loss": 0.66026628,
      "learning_rate": 2.396743698142872e-06,
      "loss": 0.68187726,
      "num_input_tokens_seen": 161779455,
      "step": 7540,
      "time_per_iteration": 2.4548158645629883
    },
    {
      "auxiliary_loss_clip": 0.01100981,
      "auxiliary_loss_mlp": 0.01040739,
      "balance_loss_clip": 1.04298651,
      "balance_loss_mlp": 1.02508664,
      "epoch": 0.4533894483691568,
      "flos": 22601206840320.0,
      "grad_norm": 1.9469440732431267,
      "language_loss": 0.8499248,
      "learning_rate": 2.396361968778424e-06,
      "loss": 0.87134194,
      "num_input_tokens_seen": 161798980,
      "step": 7541,
      "time_per_iteration": 2.5491039752960205
    },
    {
      "auxiliary_loss_clip": 0.01096254,
      "auxiliary_loss_mlp": 0.01029768,
      "balance_loss_clip": 1.04077625,
      "balance_loss_mlp": 1.01650608,
      "epoch": 0.45344957162182475,
      "flos": 34751617666560.0,
      "grad_norm": 1.7927916555094816,
      "language_loss": 0.76346445,
      "learning_rate": 2.395980224383889e-06,
      "loss": 0.78472465,
      "num_input_tokens_seen": 161819745,
      "step": 7542,
      "time_per_iteration": 2.636831283569336
    },
    {
      "auxiliary_loss_clip": 0.01091845,
      "auxiliary_loss_mlp": 0.01026691,
      "balance_loss_clip": 1.04007053,
      "balance_loss_mlp": 1.01291108,
      "epoch": 0.4535096948744927,
      "flos": 23550218121600.0,
      "grad_norm": 1.6580692177561902,
      "language_loss": 0.80160815,
      "learning_rate": 2.395598464973746e-06,
      "loss": 0.82279348,
      "num_input_tokens_seen": 161838575,
      "step": 7543,
      "time_per_iteration": 2.544360637664795
    },
    {
      "auxiliary_loss_clip": 0.01106536,
      "auxiliary_loss_mlp": 0.00789603,
      "balance_loss_clip": 1.03935826,
      "balance_loss_mlp": 1.01374149,
      "epoch": 0.4535698181271607,
      "flos": 25557552748800.0,
      "grad_norm": 1.610403814559713,
      "language_loss": 0.76227498,
      "learning_rate": 2.395216690562469e-06,
      "loss": 0.78123641,
      "num_input_tokens_seen": 161858590,
      "step": 7544,
      "time_per_iteration": 2.543915271759033
    },
    {
      "auxiliary_loss_clip": 0.01090836,
      "auxiliary_loss_mlp": 0.01035968,
      "balance_loss_clip": 1.04275,
      "balance_loss_mlp": 1.02243745,
      "epoch": 0.45362994137982865,
      "flos": 24864117713280.0,
      "grad_norm": 1.6467215558897073,
      "language_loss": 0.75508571,
      "learning_rate": 2.3948349011645355e-06,
      "loss": 0.77635378,
      "num_input_tokens_seen": 161878390,
      "step": 7545,
      "time_per_iteration": 2.5673301219940186
    },
    {
      "auxiliary_loss_clip": 0.01099725,
      "auxiliary_loss_mlp": 0.01033131,
      "balance_loss_clip": 1.04171968,
      "balance_loss_mlp": 1.01932669,
      "epoch": 0.4536900646324966,
      "flos": 30806794408320.0,
      "grad_norm": 1.8938380501715006,
      "language_loss": 0.72214293,
      "learning_rate": 2.394453096794423e-06,
      "loss": 0.7434715,
      "num_input_tokens_seen": 161898610,
      "step": 7546,
      "time_per_iteration": 2.5985326766967773
    },
    {
      "auxiliary_loss_clip": 0.01100686,
      "auxiliary_loss_mlp": 0.01032289,
      "balance_loss_clip": 1.041049,
      "balance_loss_mlp": 1.01729858,
      "epoch": 0.4537501878851646,
      "flos": 23404313076480.0,
      "grad_norm": 1.5453574207572947,
      "language_loss": 0.7541256,
      "learning_rate": 2.394071277466609e-06,
      "loss": 0.77545536,
      "num_input_tokens_seen": 161918210,
      "step": 7547,
      "time_per_iteration": 2.5293996334075928
    },
    {
      "auxiliary_loss_clip": 0.01111515,
      "auxiliary_loss_mlp": 0.01031868,
      "balance_loss_clip": 1.04169774,
      "balance_loss_mlp": 1.01745534,
      "epoch": 0.45381031113783254,
      "flos": 18149289327360.0,
      "grad_norm": 1.9113224280566248,
      "language_loss": 0.69135523,
      "learning_rate": 2.393689443195573e-06,
      "loss": 0.71278918,
      "num_input_tokens_seen": 161936950,
      "step": 7548,
      "time_per_iteration": 2.471928358078003
    },
    {
      "auxiliary_loss_clip": 0.01119041,
      "auxiliary_loss_mlp": 0.01038828,
      "balance_loss_clip": 1.04054976,
      "balance_loss_mlp": 1.02512455,
      "epoch": 0.4538704343905005,
      "flos": 25336666062720.0,
      "grad_norm": 1.9528313110622653,
      "language_loss": 0.72441006,
      "learning_rate": 2.393307593995794e-06,
      "loss": 0.74598873,
      "num_input_tokens_seen": 161955550,
      "step": 7549,
      "time_per_iteration": 2.5065805912017822
    },
    {
      "auxiliary_loss_clip": 0.01085023,
      "auxiliary_loss_mlp": 0.01027706,
      "balance_loss_clip": 1.0383997,
      "balance_loss_mlp": 1.01484942,
      "epoch": 0.4539305576431685,
      "flos": 28731445378560.0,
      "grad_norm": 1.5510735845768886,
      "language_loss": 0.65360427,
      "learning_rate": 2.392925729881751e-06,
      "loss": 0.67473155,
      "num_input_tokens_seen": 161976760,
      "step": 7550,
      "time_per_iteration": 2.594890594482422
    },
    {
      "auxiliary_loss_clip": 0.01101428,
      "auxiliary_loss_mlp": 0.01031409,
      "balance_loss_clip": 1.04198194,
      "balance_loss_mlp": 1.01880217,
      "epoch": 0.45399068089583644,
      "flos": 22492397566080.0,
      "grad_norm": 1.6528691851977826,
      "language_loss": 0.68808752,
      "learning_rate": 2.3925438508679263e-06,
      "loss": 0.70941591,
      "num_input_tokens_seen": 161996120,
      "step": 7551,
      "time_per_iteration": 2.502305269241333
    },
    {
      "auxiliary_loss_clip": 0.01106462,
      "auxiliary_loss_mlp": 0.01034855,
      "balance_loss_clip": 1.03812587,
      "balance_loss_mlp": 1.01996541,
      "epoch": 0.45405080414850446,
      "flos": 12893403651840.0,
      "grad_norm": 1.9869528637006668,
      "language_loss": 0.79115903,
      "learning_rate": 2.392161956968798e-06,
      "loss": 0.81257224,
      "num_input_tokens_seen": 162011125,
      "step": 7552,
      "time_per_iteration": 2.4385323524475098
    },
    {
      "auxiliary_loss_clip": 0.01032311,
      "auxiliary_loss_mlp": 0.01004622,
      "balance_loss_clip": 1.01695275,
      "balance_loss_mlp": 1.0032568,
      "epoch": 0.4541109274011724,
      "flos": 59766919724160.0,
      "grad_norm": 0.8122332999290258,
      "language_loss": 0.57808006,
      "learning_rate": 2.39178004819885e-06,
      "loss": 0.59844947,
      "num_input_tokens_seen": 162068705,
      "step": 7553,
      "time_per_iteration": 3.0522093772888184
    },
    {
      "auxiliary_loss_clip": 0.01065168,
      "auxiliary_loss_mlp": 0.01036045,
      "balance_loss_clip": 1.04344988,
      "balance_loss_mlp": 1.02260399,
      "epoch": 0.4541710506538404,
      "flos": 28511743841280.0,
      "grad_norm": 1.4331164712803282,
      "language_loss": 0.76836181,
      "learning_rate": 2.3913981245725626e-06,
      "loss": 0.78937399,
      "num_input_tokens_seen": 162089655,
      "step": 7554,
      "time_per_iteration": 2.6535513401031494
    },
    {
      "auxiliary_loss_clip": 0.01099922,
      "auxiliary_loss_mlp": 0.01033105,
      "balance_loss_clip": 1.04087281,
      "balance_loss_mlp": 1.01816773,
      "epoch": 0.45423117390650836,
      "flos": 17675591742720.0,
      "grad_norm": 2.17260200060156,
      "language_loss": 0.76381415,
      "learning_rate": 2.3910161861044194e-06,
      "loss": 0.78514445,
      "num_input_tokens_seen": 162108465,
      "step": 7555,
      "time_per_iteration": 2.5028486251831055
    },
    {
      "auxiliary_loss_clip": 0.01052193,
      "auxiliary_loss_mlp": 0.01033872,
      "balance_loss_clip": 1.03846061,
      "balance_loss_mlp": 1.0204612,
      "epoch": 0.4542912971591763,
      "flos": 28072556248320.0,
      "grad_norm": 1.3031368983599358,
      "language_loss": 0.72527254,
      "learning_rate": 2.390634232808903e-06,
      "loss": 0.74613315,
      "num_input_tokens_seen": 162129910,
      "step": 7556,
      "time_per_iteration": 2.6869678497314453
    },
    {
      "auxiliary_loss_clip": 0.01123979,
      "auxiliary_loss_mlp": 0.01036481,
      "balance_loss_clip": 1.04254627,
      "balance_loss_mlp": 1.02270591,
      "epoch": 0.4543514204118443,
      "flos": 22671771108480.0,
      "grad_norm": 1.975003485957552,
      "language_loss": 0.62864166,
      "learning_rate": 2.3902522647004982e-06,
      "loss": 0.65024632,
      "num_input_tokens_seen": 162148840,
      "step": 7557,
      "time_per_iteration": 2.472999334335327
    },
    {
      "auxiliary_loss_clip": 0.01024851,
      "auxiliary_loss_mlp": 0.01002178,
      "balance_loss_clip": 1.01815712,
      "balance_loss_mlp": 1.0007472,
      "epoch": 0.45441154366451225,
      "flos": 58216549921920.0,
      "grad_norm": 0.6817168858425913,
      "language_loss": 0.57655728,
      "learning_rate": 2.3898702817936875e-06,
      "loss": 0.59682751,
      "num_input_tokens_seen": 162208500,
      "step": 7558,
      "time_per_iteration": 3.0351898670196533
    },
    {
      "auxiliary_loss_clip": 0.01110365,
      "auxiliary_loss_mlp": 0.01039909,
      "balance_loss_clip": 1.04053986,
      "balance_loss_mlp": 1.02473354,
      "epoch": 0.4544716669171802,
      "flos": 16764286763520.0,
      "grad_norm": 2.6652305120113065,
      "language_loss": 0.56044316,
      "learning_rate": 2.3894882841029573e-06,
      "loss": 0.5819459,
      "num_input_tokens_seen": 162224650,
      "step": 7559,
      "time_per_iteration": 2.47643780708313
    },
    {
      "auxiliary_loss_clip": 0.01105864,
      "auxiliary_loss_mlp": 0.00789043,
      "balance_loss_clip": 1.04166746,
      "balance_loss_mlp": 1.01187468,
      "epoch": 0.4545317901698482,
      "flos": 15925233991680.0,
      "grad_norm": 1.77060312511399,
      "language_loss": 0.72154319,
      "learning_rate": 2.389106271642792e-06,
      "loss": 0.74049222,
      "num_input_tokens_seen": 162242930,
      "step": 7560,
      "time_per_iteration": 2.4859554767608643
    },
    {
      "auxiliary_loss_clip": 0.0105071,
      "auxiliary_loss_mlp": 0.01039745,
      "balance_loss_clip": 1.03976691,
      "balance_loss_mlp": 1.02437341,
      "epoch": 0.45459191342251615,
      "flos": 17639752947840.0,
      "grad_norm": 2.689862472272109,
      "language_loss": 0.68986869,
      "learning_rate": 2.3887242444276775e-06,
      "loss": 0.71077323,
      "num_input_tokens_seen": 162261455,
      "step": 7561,
      "time_per_iteration": 2.6903345584869385
    },
    {
      "auxiliary_loss_clip": 0.01095665,
      "auxiliary_loss_mlp": 0.01034862,
      "balance_loss_clip": 1.0406028,
      "balance_loss_mlp": 1.02202296,
      "epoch": 0.4546520366751841,
      "flos": 16176608346240.0,
      "grad_norm": 1.7692013309673016,
      "language_loss": 0.84869254,
      "learning_rate": 2.3883422024721015e-06,
      "loss": 0.86999774,
      "num_input_tokens_seen": 162279725,
      "step": 7562,
      "time_per_iteration": 2.7375617027282715
    },
    {
      "auxiliary_loss_clip": 0.011047,
      "auxiliary_loss_mlp": 0.01034114,
      "balance_loss_clip": 1.03860712,
      "balance_loss_mlp": 1.02082253,
      "epoch": 0.4547121599278521,
      "flos": 19751443562880.0,
      "grad_norm": 1.7870307629486721,
      "language_loss": 0.89485151,
      "learning_rate": 2.38796014579055e-06,
      "loss": 0.91623962,
      "num_input_tokens_seen": 162297865,
      "step": 7563,
      "time_per_iteration": 2.4810233116149902
    },
    {
      "auxiliary_loss_clip": 0.01118039,
      "auxiliary_loss_mlp": 0.00791942,
      "balance_loss_clip": 1.0392766,
      "balance_loss_mlp": 1.0157696,
      "epoch": 0.45477228318052004,
      "flos": 19937461121280.0,
      "grad_norm": 1.9067591967194177,
      "language_loss": 0.7153365,
      "learning_rate": 2.3875780743975097e-06,
      "loss": 0.73443633,
      "num_input_tokens_seen": 162316010,
      "step": 7564,
      "time_per_iteration": 2.448718786239624
    },
    {
      "auxiliary_loss_clip": 0.01109414,
      "auxiliary_loss_mlp": 0.01038265,
      "balance_loss_clip": 1.03885674,
      "balance_loss_mlp": 1.02438927,
      "epoch": 0.454832406433188,
      "flos": 21288312829440.0,
      "grad_norm": 2.0431797494485853,
      "language_loss": 0.68389094,
      "learning_rate": 2.3871959883074713e-06,
      "loss": 0.70536768,
      "num_input_tokens_seen": 162336115,
      "step": 7565,
      "time_per_iteration": 2.5072708129882812
    },
    {
      "auxiliary_loss_clip": 0.01075108,
      "auxiliary_loss_mlp": 0.01034795,
      "balance_loss_clip": 1.03542805,
      "balance_loss_mlp": 1.02078152,
      "epoch": 0.45489252968585603,
      "flos": 24498726612480.0,
      "grad_norm": 1.681564106811362,
      "language_loss": 0.80448914,
      "learning_rate": 2.386813887534922e-06,
      "loss": 0.82558817,
      "num_input_tokens_seen": 162355705,
      "step": 7566,
      "time_per_iteration": 2.5969293117523193
    },
    {
      "auxiliary_loss_clip": 0.01083796,
      "auxiliary_loss_mlp": 0.01034689,
      "balance_loss_clip": 1.04015112,
      "balance_loss_mlp": 1.01960874,
      "epoch": 0.454952652938524,
      "flos": 17092474352640.0,
      "grad_norm": 1.606633612457842,
      "language_loss": 0.73541552,
      "learning_rate": 2.3864317720943508e-06,
      "loss": 0.75660038,
      "num_input_tokens_seen": 162374055,
      "step": 7567,
      "time_per_iteration": 5.267664432525635
    },
    {
      "auxiliary_loss_clip": 0.01084496,
      "auxiliary_loss_mlp": 0.0103801,
      "balance_loss_clip": 1.0390799,
      "balance_loss_mlp": 1.02320433,
      "epoch": 0.45501277619119196,
      "flos": 27630387826560.0,
      "grad_norm": 2.71120563044189,
      "language_loss": 0.80976224,
      "learning_rate": 2.386049642000249e-06,
      "loss": 0.83098733,
      "num_input_tokens_seen": 162393560,
      "step": 7568,
      "time_per_iteration": 2.5954670906066895
    },
    {
      "auxiliary_loss_clip": 0.01113344,
      "auxiliary_loss_mlp": 0.01047838,
      "balance_loss_clip": 1.04025161,
      "balance_loss_mlp": 1.03226352,
      "epoch": 0.4550728994438599,
      "flos": 19974664632960.0,
      "grad_norm": 1.8364789366893806,
      "language_loss": 0.80004013,
      "learning_rate": 2.3856674972671055e-06,
      "loss": 0.82165194,
      "num_input_tokens_seen": 162413170,
      "step": 7569,
      "time_per_iteration": 2.4610133171081543
    },
    {
      "auxiliary_loss_clip": 0.01111416,
      "auxiliary_loss_mlp": 0.01033977,
      "balance_loss_clip": 1.04104984,
      "balance_loss_mlp": 1.01871848,
      "epoch": 0.4551330226965279,
      "flos": 26066873646720.0,
      "grad_norm": 1.5790860316190194,
      "language_loss": 0.75027299,
      "learning_rate": 2.385285337909412e-06,
      "loss": 0.77172697,
      "num_input_tokens_seen": 162434080,
      "step": 7570,
      "time_per_iteration": 2.520038604736328
    },
    {
      "auxiliary_loss_clip": 0.01097931,
      "auxiliary_loss_mlp": 0.01038225,
      "balance_loss_clip": 1.0410794,
      "balance_loss_mlp": 1.02412224,
      "epoch": 0.45519314594919585,
      "flos": 32781091501440.0,
      "grad_norm": 1.8056831411998504,
      "language_loss": 0.74491507,
      "learning_rate": 2.3849031639416596e-06,
      "loss": 0.7662766,
      "num_input_tokens_seen": 162455445,
      "step": 7571,
      "time_per_iteration": 4.003863573074341
    },
    {
      "auxiliary_loss_clip": 0.01103382,
      "auxiliary_loss_mlp": 0.01029787,
      "balance_loss_clip": 1.03892112,
      "balance_loss_mlp": 1.01604199,
      "epoch": 0.4552532692018638,
      "flos": 19172671718400.0,
      "grad_norm": 1.4742333690324358,
      "language_loss": 0.81200504,
      "learning_rate": 2.3845209753783414e-06,
      "loss": 0.83333671,
      "num_input_tokens_seen": 162474940,
      "step": 7572,
      "time_per_iteration": 2.4658901691436768
    },
    {
      "auxiliary_loss_clip": 0.01103799,
      "auxiliary_loss_mlp": 0.01038026,
      "balance_loss_clip": 1.04047716,
      "balance_loss_mlp": 1.02215898,
      "epoch": 0.4553133924545318,
      "flos": 26027156183040.0,
      "grad_norm": 1.9441153247358252,
      "language_loss": 0.72756255,
      "learning_rate": 2.3841387722339486e-06,
      "loss": 0.74898076,
      "num_input_tokens_seen": 162493340,
      "step": 7573,
      "time_per_iteration": 3.9208662509918213
    },
    {
      "auxiliary_loss_clip": 0.01112158,
      "auxiliary_loss_mlp": 0.01038818,
      "balance_loss_clip": 1.04175496,
      "balance_loss_mlp": 1.02211702,
      "epoch": 0.45537351570719975,
      "flos": 30661535808000.0,
      "grad_norm": 1.8539720424639634,
      "language_loss": 0.74764109,
      "learning_rate": 2.3837565545229748e-06,
      "loss": 0.76915085,
      "num_input_tokens_seen": 162514360,
      "step": 7574,
      "time_per_iteration": 2.5490190982818604
    },
    {
      "auxiliary_loss_clip": 0.01108407,
      "auxiliary_loss_mlp": 0.01033563,
      "balance_loss_clip": 1.03923297,
      "balance_loss_mlp": 1.01915097,
      "epoch": 0.4554336389598677,
      "flos": 24353396184960.0,
      "grad_norm": 1.434467350911415,
      "language_loss": 0.71333903,
      "learning_rate": 2.383374322259915e-06,
      "loss": 0.73475873,
      "num_input_tokens_seen": 162535240,
      "step": 7575,
      "time_per_iteration": 2.4957494735717773
    },
    {
      "auxiliary_loss_clip": 0.01094681,
      "auxiliary_loss_mlp": 0.01032105,
      "balance_loss_clip": 1.0382303,
      "balance_loss_mlp": 1.01800251,
      "epoch": 0.4554937622125357,
      "flos": 20557925677440.0,
      "grad_norm": 1.9324751736784862,
      "language_loss": 0.7347014,
      "learning_rate": 2.3829920754592617e-06,
      "loss": 0.75596929,
      "num_input_tokens_seen": 162553880,
      "step": 7576,
      "time_per_iteration": 2.5234806537628174
    },
    {
      "auxiliary_loss_clip": 0.01116268,
      "auxiliary_loss_mlp": 0.01035861,
      "balance_loss_clip": 1.03972268,
      "balance_loss_mlp": 1.02153158,
      "epoch": 0.45555388546520365,
      "flos": 22820764723200.0,
      "grad_norm": 1.6693662114601133,
      "language_loss": 0.66307771,
      "learning_rate": 2.382609814135511e-06,
      "loss": 0.68459892,
      "num_input_tokens_seen": 162574485,
      "step": 7577,
      "time_per_iteration": 2.4518446922302246
    },
    {
      "auxiliary_loss_clip": 0.01094324,
      "auxiliary_loss_mlp": 0.01045008,
      "balance_loss_clip": 1.04298246,
      "balance_loss_mlp": 1.02862883,
      "epoch": 0.4556140087178716,
      "flos": 21725992051200.0,
      "grad_norm": 1.8762253678718777,
      "language_loss": 0.74463916,
      "learning_rate": 2.382227538303157e-06,
      "loss": 0.76603246,
      "num_input_tokens_seen": 162595130,
      "step": 7578,
      "time_per_iteration": 2.538808822631836
    },
    {
      "auxiliary_loss_clip": 0.0106986,
      "auxiliary_loss_mlp": 0.00791618,
      "balance_loss_clip": 1.03989303,
      "balance_loss_mlp": 1.01096559,
      "epoch": 0.45567413197053963,
      "flos": 25994513698560.0,
      "grad_norm": 1.8338235316490974,
      "language_loss": 0.70507818,
      "learning_rate": 2.381845247976697e-06,
      "loss": 0.72369295,
      "num_input_tokens_seen": 162615720,
      "step": 7579,
      "time_per_iteration": 2.6062605381011963
    },
    {
      "auxiliary_loss_clip": 0.01104777,
      "auxiliary_loss_mlp": 0.01033422,
      "balance_loss_clip": 1.03821778,
      "balance_loss_mlp": 1.01955795,
      "epoch": 0.4557342552232076,
      "flos": 21537604195200.0,
      "grad_norm": 1.697454214547009,
      "language_loss": 0.78625524,
      "learning_rate": 2.381462943170627e-06,
      "loss": 0.80763721,
      "num_input_tokens_seen": 162635825,
      "step": 7580,
      "time_per_iteration": 2.5063295364379883
    },
    {
      "auxiliary_loss_clip": 0.01118488,
      "auxiliary_loss_mlp": 0.01031727,
      "balance_loss_clip": 1.04131413,
      "balance_loss_mlp": 1.01711786,
      "epoch": 0.45579437847587556,
      "flos": 40001972647680.0,
      "grad_norm": 2.222231053429561,
      "language_loss": 0.68710983,
      "learning_rate": 2.381080623899444e-06,
      "loss": 0.70861191,
      "num_input_tokens_seen": 162659130,
      "step": 7581,
      "time_per_iteration": 2.623645544052124
    },
    {
      "auxiliary_loss_clip": 0.0110064,
      "auxiliary_loss_mlp": 0.01030509,
      "balance_loss_clip": 1.0360198,
      "balance_loss_mlp": 1.01610899,
      "epoch": 0.4558545017285435,
      "flos": 31138501530240.0,
      "grad_norm": 1.5937889860080712,
      "language_loss": 0.7317872,
      "learning_rate": 2.3806982901776455e-06,
      "loss": 0.75309873,
      "num_input_tokens_seen": 162681665,
      "step": 7582,
      "time_per_iteration": 2.5898470878601074
    },
    {
      "auxiliary_loss_clip": 0.01123027,
      "auxiliary_loss_mlp": 0.01047086,
      "balance_loss_clip": 1.04235196,
      "balance_loss_mlp": 1.03145146,
      "epoch": 0.4559146249812115,
      "flos": 21725776569600.0,
      "grad_norm": 1.9367076076637972,
      "language_loss": 0.72423142,
      "learning_rate": 2.380315942019729e-06,
      "loss": 0.74593258,
      "num_input_tokens_seen": 162702040,
      "step": 7583,
      "time_per_iteration": 2.450486660003662
    },
    {
      "auxiliary_loss_clip": 0.01105483,
      "auxiliary_loss_mlp": 0.01033757,
      "balance_loss_clip": 1.04023194,
      "balance_loss_mlp": 1.0194217,
      "epoch": 0.45597474823387946,
      "flos": 23805973935360.0,
      "grad_norm": 1.766572313546338,
      "language_loss": 0.72733414,
      "learning_rate": 2.379933579440195e-06,
      "loss": 0.74872655,
      "num_input_tokens_seen": 162722375,
      "step": 7584,
      "time_per_iteration": 2.554657459259033
    },
    {
      "auxiliary_loss_clip": 0.01080455,
      "auxiliary_loss_mlp": 0.0103594,
      "balance_loss_clip": 1.03864956,
      "balance_loss_mlp": 1.02105045,
      "epoch": 0.4560348714865474,
      "flos": 31905661230720.0,
      "grad_norm": 1.9903713561869234,
      "language_loss": 0.68426853,
      "learning_rate": 2.379551202453541e-06,
      "loss": 0.70543247,
      "num_input_tokens_seen": 162746095,
      "step": 7585,
      "time_per_iteration": 2.6137399673461914
    },
    {
      "auxiliary_loss_clip": 0.01117344,
      "auxiliary_loss_mlp": 0.01031292,
      "balance_loss_clip": 1.03969336,
      "balance_loss_mlp": 1.01803565,
      "epoch": 0.4560949947392154,
      "flos": 22048828513920.0,
      "grad_norm": 1.4517747517778161,
      "language_loss": 0.76071852,
      "learning_rate": 2.379168811074267e-06,
      "loss": 0.78220487,
      "num_input_tokens_seen": 162766330,
      "step": 7586,
      "time_per_iteration": 2.4805328845977783
    },
    {
      "auxiliary_loss_clip": 0.01098014,
      "auxiliary_loss_mlp": 0.01027486,
      "balance_loss_clip": 1.03953004,
      "balance_loss_mlp": 1.01462364,
      "epoch": 0.45615511799188335,
      "flos": 24571804832640.0,
      "grad_norm": 1.6487776957163107,
      "language_loss": 0.78086066,
      "learning_rate": 2.3787864053168747e-06,
      "loss": 0.80211568,
      "num_input_tokens_seen": 162784755,
      "step": 7587,
      "time_per_iteration": 2.5229883193969727
    },
    {
      "auxiliary_loss_clip": 0.01096809,
      "auxiliary_loss_mlp": 0.01040283,
      "balance_loss_clip": 1.03808415,
      "balance_loss_mlp": 1.02556038,
      "epoch": 0.4562152412445513,
      "flos": 18330709944960.0,
      "grad_norm": 2.3895287283229423,
      "language_loss": 0.6903345,
      "learning_rate": 2.378403985195863e-06,
      "loss": 0.71170545,
      "num_input_tokens_seen": 162803850,
      "step": 7588,
      "time_per_iteration": 2.485286235809326
    },
    {
      "auxiliary_loss_clip": 0.0110323,
      "auxiliary_loss_mlp": 0.01029889,
      "balance_loss_clip": 1.0430268,
      "balance_loss_mlp": 1.01713979,
      "epoch": 0.4562753644972193,
      "flos": 13516525814400.0,
      "grad_norm": 1.7465814289427044,
      "language_loss": 0.79119766,
      "learning_rate": 2.378021550725735e-06,
      "loss": 0.81252885,
      "num_input_tokens_seen": 162820775,
      "step": 7589,
      "time_per_iteration": 2.4393413066864014
    },
    {
      "auxiliary_loss_clip": 0.0110711,
      "auxiliary_loss_mlp": 0.01036011,
      "balance_loss_clip": 1.03968978,
      "balance_loss_mlp": 1.02164078,
      "epoch": 0.45633548774988725,
      "flos": 29639697701760.0,
      "grad_norm": 2.095357071551132,
      "language_loss": 0.61417788,
      "learning_rate": 2.377639101920992e-06,
      "loss": 0.63560903,
      "num_input_tokens_seen": 162839695,
      "step": 7590,
      "time_per_iteration": 2.563778877258301
    },
    {
      "auxiliary_loss_clip": 0.01094144,
      "auxiliary_loss_mlp": 0.01041747,
      "balance_loss_clip": 1.04019403,
      "balance_loss_mlp": 1.02776372,
      "epoch": 0.4563956110025552,
      "flos": 22233409528320.0,
      "grad_norm": 2.24968527411082,
      "language_loss": 0.72622401,
      "learning_rate": 2.377256638796135e-06,
      "loss": 0.74758291,
      "num_input_tokens_seen": 162856095,
      "step": 7591,
      "time_per_iteration": 2.4966702461242676
    },
    {
      "auxiliary_loss_clip": 0.01102012,
      "auxiliary_loss_mlp": 0.0103959,
      "balance_loss_clip": 1.04251075,
      "balance_loss_mlp": 1.02492738,
      "epoch": 0.45645573425522323,
      "flos": 17092043389440.0,
      "grad_norm": 1.7903218773313025,
      "language_loss": 0.76429343,
      "learning_rate": 2.3768741613656695e-06,
      "loss": 0.78570938,
      "num_input_tokens_seen": 162874070,
      "step": 7592,
      "time_per_iteration": 2.526815414428711
    },
    {
      "auxiliary_loss_clip": 0.01092682,
      "auxiliary_loss_mlp": 0.01041989,
      "balance_loss_clip": 1.03756714,
      "balance_loss_mlp": 1.02659929,
      "epoch": 0.4565158575078912,
      "flos": 20332334309760.0,
      "grad_norm": 2.0262548926383723,
      "language_loss": 0.69492704,
      "learning_rate": 2.376491669644098e-06,
      "loss": 0.71627373,
      "num_input_tokens_seen": 162891000,
      "step": 7593,
      "time_per_iteration": 2.5015547275543213
    },
    {
      "auxiliary_loss_clip": 0.01099924,
      "auxiliary_loss_mlp": 0.01029716,
      "balance_loss_clip": 1.03694177,
      "balance_loss_mlp": 1.01770556,
      "epoch": 0.45657598076055916,
      "flos": 23983013093760.0,
      "grad_norm": 1.9634593911032778,
      "language_loss": 0.84365898,
      "learning_rate": 2.3761091636459248e-06,
      "loss": 0.86495543,
      "num_input_tokens_seen": 162910120,
      "step": 7594,
      "time_per_iteration": 2.5480031967163086
    },
    {
      "auxiliary_loss_clip": 0.01033417,
      "auxiliary_loss_mlp": 0.00772001,
      "balance_loss_clip": 1.01620317,
      "balance_loss_mlp": 1.00659323,
      "epoch": 0.45663610401322713,
      "flos": 69364297526400.0,
      "grad_norm": 0.8129282383523677,
      "language_loss": 0.52773058,
      "learning_rate": 2.375726643385654e-06,
      "loss": 0.54578477,
      "num_input_tokens_seen": 162963720,
      "step": 7595,
      "time_per_iteration": 3.119999408721924
    },
    {
      "auxiliary_loss_clip": 0.01087868,
      "auxiliary_loss_mlp": 0.01032857,
      "balance_loss_clip": 1.03932357,
      "balance_loss_mlp": 1.0185039,
      "epoch": 0.4566962272658951,
      "flos": 15149095891200.0,
      "grad_norm": 2.0422655435570922,
      "language_loss": 0.87240541,
      "learning_rate": 2.3753441088777915e-06,
      "loss": 0.89361262,
      "num_input_tokens_seen": 162975760,
      "step": 7596,
      "time_per_iteration": 2.5413429737091064
    },
    {
      "auxiliary_loss_clip": 0.01112869,
      "auxiliary_loss_mlp": 0.01040828,
      "balance_loss_clip": 1.04262829,
      "balance_loss_mlp": 1.02725589,
      "epoch": 0.45675635051856306,
      "flos": 18697465762560.0,
      "grad_norm": 1.6840430860729967,
      "language_loss": 0.77552772,
      "learning_rate": 2.374961560136843e-06,
      "loss": 0.79706478,
      "num_input_tokens_seen": 162994865,
      "step": 7597,
      "time_per_iteration": 2.493898630142212
    },
    {
      "auxiliary_loss_clip": 0.01108369,
      "auxiliary_loss_mlp": 0.01035252,
      "balance_loss_clip": 1.04021645,
      "balance_loss_mlp": 1.0215553,
      "epoch": 0.456816473771231,
      "flos": 19098300608640.0,
      "grad_norm": 2.351601000664415,
      "language_loss": 0.78603959,
      "learning_rate": 2.374578997177314e-06,
      "loss": 0.80747575,
      "num_input_tokens_seen": 163014730,
      "step": 7598,
      "time_per_iteration": 2.5423309803009033
    },
    {
      "auxiliary_loss_clip": 0.01118073,
      "auxiliary_loss_mlp": 0.01029067,
      "balance_loss_clip": 1.04119968,
      "balance_loss_mlp": 1.01639462,
      "epoch": 0.456876597023899,
      "flos": 28950069507840.0,
      "grad_norm": 2.8115651274903604,
      "language_loss": 0.71335077,
      "learning_rate": 2.374196420013712e-06,
      "loss": 0.73482209,
      "num_input_tokens_seen": 163033405,
      "step": 7599,
      "time_per_iteration": 2.520876407623291
    },
    {
      "auxiliary_loss_clip": 0.01084907,
      "auxiliary_loss_mlp": 0.01032506,
      "balance_loss_clip": 1.03826666,
      "balance_loss_mlp": 1.0189935,
      "epoch": 0.45693672027656695,
      "flos": 23289470317440.0,
      "grad_norm": 1.8847014825033102,
      "language_loss": 0.69735563,
      "learning_rate": 2.373813828660544e-06,
      "loss": 0.71852982,
      "num_input_tokens_seen": 163051400,
      "step": 7600,
      "time_per_iteration": 2.580392599105835
    },
    {
      "auxiliary_loss_clip": 0.01058003,
      "auxiliary_loss_mlp": 0.01033901,
      "balance_loss_clip": 1.0433495,
      "balance_loss_mlp": 1.02093685,
      "epoch": 0.4569968435292349,
      "flos": 20558212986240.0,
      "grad_norm": 1.9031234603011618,
      "language_loss": 0.7933526,
      "learning_rate": 2.373431223132319e-06,
      "loss": 0.81427169,
      "num_input_tokens_seen": 163069250,
      "step": 7601,
      "time_per_iteration": 2.6294960975646973
    },
    {
      "auxiliary_loss_clip": 0.01091037,
      "auxiliary_loss_mlp": 0.0103892,
      "balance_loss_clip": 1.03888965,
      "balance_loss_mlp": 1.02590263,
      "epoch": 0.4570569667819029,
      "flos": 41282619223680.0,
      "grad_norm": 1.721885691913234,
      "language_loss": 0.71676493,
      "learning_rate": 2.3730486034435448e-06,
      "loss": 0.73806453,
      "num_input_tokens_seen": 163091755,
      "step": 7602,
      "time_per_iteration": 2.6990182399749756
    },
    {
      "auxiliary_loss_clip": 0.01105011,
      "auxiliary_loss_mlp": 0.01034586,
      "balance_loss_clip": 1.03896666,
      "balance_loss_mlp": 1.01894546,
      "epoch": 0.45711709003457085,
      "flos": 26031573555840.0,
      "grad_norm": 2.252512470955106,
      "language_loss": 0.73047626,
      "learning_rate": 2.372665969608729e-06,
      "loss": 0.75187218,
      "num_input_tokens_seen": 163111600,
      "step": 7603,
      "time_per_iteration": 2.543278217315674
    },
    {
      "auxiliary_loss_clip": 0.01107366,
      "auxiliary_loss_mlp": 0.01033453,
      "balance_loss_clip": 1.04149461,
      "balance_loss_mlp": 1.01888537,
      "epoch": 0.4571772132872388,
      "flos": 22158068751360.0,
      "grad_norm": 1.8718344049941582,
      "language_loss": 0.83037108,
      "learning_rate": 2.372283321642383e-06,
      "loss": 0.85177922,
      "num_input_tokens_seen": 163127350,
      "step": 7604,
      "time_per_iteration": 2.5536022186279297
    },
    {
      "auxiliary_loss_clip": 0.01102427,
      "auxiliary_loss_mlp": 0.01039182,
      "balance_loss_clip": 1.04756379,
      "balance_loss_mlp": 1.02431691,
      "epoch": 0.45723733653990684,
      "flos": 23878872587520.0,
      "grad_norm": 1.996124146096969,
      "language_loss": 0.85733104,
      "learning_rate": 2.371900659559016e-06,
      "loss": 0.87874722,
      "num_input_tokens_seen": 163145855,
      "step": 7605,
      "time_per_iteration": 3.9621567726135254
    },
    {
      "auxiliary_loss_clip": 0.01073793,
      "auxiliary_loss_mlp": 0.01038877,
      "balance_loss_clip": 1.03840542,
      "balance_loss_mlp": 1.02439892,
      "epoch": 0.4572974597925748,
      "flos": 16871803148160.0,
      "grad_norm": 2.4353757228506656,
      "language_loss": 0.73794538,
      "learning_rate": 2.371517983373138e-06,
      "loss": 0.75907207,
      "num_input_tokens_seen": 163163830,
      "step": 7606,
      "time_per_iteration": 3.9344241619110107
    },
    {
      "auxiliary_loss_clip": 0.0108751,
      "auxiliary_loss_mlp": 0.01034559,
      "balance_loss_clip": 1.03937411,
      "balance_loss_mlp": 1.02042079,
      "epoch": 0.45735758304524277,
      "flos": 13771491528960.0,
      "grad_norm": 2.0676747463116576,
      "language_loss": 0.80022287,
      "learning_rate": 2.371135293099262e-06,
      "loss": 0.82144356,
      "num_input_tokens_seen": 163180700,
      "step": 7607,
      "time_per_iteration": 2.505862236022949
    },
    {
      "auxiliary_loss_clip": 0.01091,
      "auxiliary_loss_mlp": 0.0103589,
      "balance_loss_clip": 1.04783189,
      "balance_loss_mlp": 1.0217334,
      "epoch": 0.45741770629791073,
      "flos": 21100750986240.0,
      "grad_norm": 1.7714052568667134,
      "language_loss": 0.80896032,
      "learning_rate": 2.3707525887518982e-06,
      "loss": 0.83022916,
      "num_input_tokens_seen": 163199450,
      "step": 7608,
      "time_per_iteration": 2.601933479309082
    },
    {
      "auxiliary_loss_clip": 0.01098089,
      "auxiliary_loss_mlp": 0.01038844,
      "balance_loss_clip": 1.03991508,
      "balance_loss_mlp": 1.02394307,
      "epoch": 0.4574778295505787,
      "flos": 23112898035840.0,
      "grad_norm": 1.5982228151972964,
      "language_loss": 0.68430746,
      "learning_rate": 2.370369870345559e-06,
      "loss": 0.70567679,
      "num_input_tokens_seen": 163217875,
      "step": 7609,
      "time_per_iteration": 2.5782577991485596
    },
    {
      "auxiliary_loss_clip": 0.01092135,
      "auxiliary_loss_mlp": 0.01042401,
      "balance_loss_clip": 1.04431975,
      "balance_loss_mlp": 1.02806592,
      "epoch": 0.45753795280324666,
      "flos": 24352929308160.0,
      "grad_norm": 1.6947926137591287,
      "language_loss": 0.80750495,
      "learning_rate": 2.369987137894757e-06,
      "loss": 0.82885027,
      "num_input_tokens_seen": 163237430,
      "step": 7610,
      "time_per_iteration": 3.9914653301239014
    },
    {
      "auxiliary_loss_clip": 0.01111081,
      "auxiliary_loss_mlp": 0.01036881,
      "balance_loss_clip": 1.04205537,
      "balance_loss_mlp": 1.02206337,
      "epoch": 0.4575980760559146,
      "flos": 16653789550080.0,
      "grad_norm": 2.1088040958240946,
      "language_loss": 0.82442546,
      "learning_rate": 2.3696043914140057e-06,
      "loss": 0.84590507,
      "num_input_tokens_seen": 163253905,
      "step": 7611,
      "time_per_iteration": 2.5525200366973877
    },
    {
      "auxiliary_loss_clip": 0.01113607,
      "auxiliary_loss_mlp": 0.01030166,
      "balance_loss_clip": 1.04379201,
      "balance_loss_mlp": 1.01568794,
      "epoch": 0.4576581993085826,
      "flos": 35911423912320.0,
      "grad_norm": 1.7814302346055144,
      "language_loss": 0.73668349,
      "learning_rate": 2.369221630917819e-06,
      "loss": 0.75812125,
      "num_input_tokens_seen": 163274285,
      "step": 7612,
      "time_per_iteration": 3.9727871417999268
    },
    {
      "auxiliary_loss_clip": 0.01092458,
      "auxiliary_loss_mlp": 0.01037321,
      "balance_loss_clip": 1.03717506,
      "balance_loss_mlp": 1.022277,
      "epoch": 0.45771832256125056,
      "flos": 20080421251200.0,
      "grad_norm": 1.590652309877682,
      "language_loss": 0.85070229,
      "learning_rate": 2.368838856420711e-06,
      "loss": 0.87200004,
      "num_input_tokens_seen": 163293150,
      "step": 7613,
      "time_per_iteration": 2.5711684226989746
    },
    {
      "auxiliary_loss_clip": 0.01083488,
      "auxiliary_loss_mlp": 0.01032615,
      "balance_loss_clip": 1.03823304,
      "balance_loss_mlp": 1.01875103,
      "epoch": 0.4577784458139185,
      "flos": 10744329957120.0,
      "grad_norm": 2.230470479130553,
      "language_loss": 0.7548058,
      "learning_rate": 2.3684560679371965e-06,
      "loss": 0.77596682,
      "num_input_tokens_seen": 163310065,
      "step": 7614,
      "time_per_iteration": 2.535323143005371
    },
    {
      "auxiliary_loss_clip": 0.01117654,
      "auxiliary_loss_mlp": 0.01030735,
      "balance_loss_clip": 1.0413506,
      "balance_loss_mlp": 1.01758623,
      "epoch": 0.4578385690665865,
      "flos": 21907269014400.0,
      "grad_norm": 1.4667500321312488,
      "language_loss": 0.74462497,
      "learning_rate": 2.368073265481791e-06,
      "loss": 0.76610887,
      "num_input_tokens_seen": 163329415,
      "step": 7615,
      "time_per_iteration": 2.53828501701355
    },
    {
      "auxiliary_loss_clip": 0.01027824,
      "auxiliary_loss_mlp": 0.01011728,
      "balance_loss_clip": 1.02092576,
      "balance_loss_mlp": 1.01021373,
      "epoch": 0.45789869231925445,
      "flos": 64758286667520.0,
      "grad_norm": 0.7859753188367176,
      "language_loss": 0.57693172,
      "learning_rate": 2.3676904490690105e-06,
      "loss": 0.59732723,
      "num_input_tokens_seen": 163385875,
      "step": 7616,
      "time_per_iteration": 3.115879774093628
    },
    {
      "auxiliary_loss_clip": 0.01089357,
      "auxiliary_loss_mlp": 0.0079253,
      "balance_loss_clip": 1.03827643,
      "balance_loss_mlp": 1.01796889,
      "epoch": 0.4579588155719224,
      "flos": 16144001775360.0,
      "grad_norm": 1.661409550548061,
      "language_loss": 0.70831835,
      "learning_rate": 2.3673076187133704e-06,
      "loss": 0.72713721,
      "num_input_tokens_seen": 163405170,
      "step": 7617,
      "time_per_iteration": 2.5494508743286133
    },
    {
      "auxiliary_loss_clip": 0.01121328,
      "auxiliary_loss_mlp": 0.01035157,
      "balance_loss_clip": 1.04333103,
      "balance_loss_mlp": 1.02131724,
      "epoch": 0.45801893882459044,
      "flos": 21395541905280.0,
      "grad_norm": 3.1471515872120253,
      "language_loss": 0.76268744,
      "learning_rate": 2.36692477442939e-06,
      "loss": 0.78425229,
      "num_input_tokens_seen": 163423155,
      "step": 7618,
      "time_per_iteration": 2.5000972747802734
    },
    {
      "auxiliary_loss_clip": 0.0108475,
      "auxiliary_loss_mlp": 0.01040274,
      "balance_loss_clip": 1.04117751,
      "balance_loss_mlp": 1.02689874,
      "epoch": 0.4580790620772584,
      "flos": 19536554448000.0,
      "grad_norm": 1.646988952337748,
      "language_loss": 0.76904356,
      "learning_rate": 2.366541916231585e-06,
      "loss": 0.79029381,
      "num_input_tokens_seen": 163442450,
      "step": 7619,
      "time_per_iteration": 2.6207408905029297
    },
    {
      "auxiliary_loss_clip": 0.01118006,
      "auxiliary_loss_mlp": 0.01037168,
      "balance_loss_clip": 1.0417397,
      "balance_loss_mlp": 1.02474046,
      "epoch": 0.45813918532992637,
      "flos": 16581070465920.0,
      "grad_norm": 1.7959546743044283,
      "language_loss": 0.71756518,
      "learning_rate": 2.366159044134473e-06,
      "loss": 0.73911691,
      "num_input_tokens_seen": 163459810,
      "step": 7620,
      "time_per_iteration": 2.4641294479370117
    },
    {
      "auxiliary_loss_clip": 0.01093194,
      "auxiliary_loss_mlp": 0.01030239,
      "balance_loss_clip": 1.04045558,
      "balance_loss_mlp": 1.01717925,
      "epoch": 0.45819930858259433,
      "flos": 42230301701760.0,
      "grad_norm": 1.55379691382913,
      "language_loss": 0.77805609,
      "learning_rate": 2.3657761581525748e-06,
      "loss": 0.79929042,
      "num_input_tokens_seen": 163482970,
      "step": 7621,
      "time_per_iteration": 2.755755662918091
    },
    {
      "auxiliary_loss_clip": 0.01032376,
      "auxiliary_loss_mlp": 0.01002387,
      "balance_loss_clip": 1.01493573,
      "balance_loss_mlp": 1.00097466,
      "epoch": 0.4582594318352623,
      "flos": 63714795638400.0,
      "grad_norm": 0.7859044406718821,
      "language_loss": 0.65025854,
      "learning_rate": 2.3653932583004063e-06,
      "loss": 0.67060626,
      "num_input_tokens_seen": 163545330,
      "step": 7622,
      "time_per_iteration": 3.1740617752075195
    },
    {
      "auxiliary_loss_clip": 0.0110539,
      "auxiliary_loss_mlp": 0.01031616,
      "balance_loss_clip": 1.04168653,
      "balance_loss_mlp": 1.01737022,
      "epoch": 0.45831955508793026,
      "flos": 26869979882880.0,
      "grad_norm": 1.768788089588182,
      "language_loss": 0.79664278,
      "learning_rate": 2.3650103445924903e-06,
      "loss": 0.81801283,
      "num_input_tokens_seen": 163564620,
      "step": 7623,
      "time_per_iteration": 2.5738155841827393
    },
    {
      "auxiliary_loss_clip": 0.01070845,
      "auxiliary_loss_mlp": 0.01036976,
      "balance_loss_clip": 1.03885829,
      "balance_loss_mlp": 1.02244401,
      "epoch": 0.45837967834059823,
      "flos": 18733951002240.0,
      "grad_norm": 2.5129157285400803,
      "language_loss": 0.70921087,
      "learning_rate": 2.3646274170433452e-06,
      "loss": 0.7302891,
      "num_input_tokens_seen": 163581010,
      "step": 7624,
      "time_per_iteration": 2.6099324226379395
    },
    {
      "auxiliary_loss_clip": 0.01090713,
      "auxiliary_loss_mlp": 0.0103878,
      "balance_loss_clip": 1.03759336,
      "balance_loss_mlp": 1.02426004,
      "epoch": 0.4584398015932662,
      "flos": 21178102924800.0,
      "grad_norm": 1.8986518305475755,
      "language_loss": 0.72960842,
      "learning_rate": 2.364244475667491e-06,
      "loss": 0.75090331,
      "num_input_tokens_seen": 163599955,
      "step": 7625,
      "time_per_iteration": 2.634599447250366
    },
    {
      "auxiliary_loss_clip": 0.01112431,
      "auxiliary_loss_mlp": 0.01039077,
      "balance_loss_clip": 1.04246819,
      "balance_loss_mlp": 1.02556419,
      "epoch": 0.45849992484593416,
      "flos": 19790047704960.0,
      "grad_norm": 2.5742670433707215,
      "language_loss": 0.7777921,
      "learning_rate": 2.363861520479451e-06,
      "loss": 0.79930717,
      "num_input_tokens_seen": 163618545,
      "step": 7626,
      "time_per_iteration": 2.5304484367370605
    },
    {
      "auxiliary_loss_clip": 0.01122124,
      "auxiliary_loss_mlp": 0.01042399,
      "balance_loss_clip": 1.04216182,
      "balance_loss_mlp": 1.02854049,
      "epoch": 0.4585600480986021,
      "flos": 18223265387520.0,
      "grad_norm": 1.730047676486694,
      "language_loss": 0.84948349,
      "learning_rate": 2.3634785514937445e-06,
      "loss": 0.87112868,
      "num_input_tokens_seen": 163636055,
      "step": 7627,
      "time_per_iteration": 2.441471576690674
    },
    {
      "auxiliary_loss_clip": 0.01123868,
      "auxiliary_loss_mlp": 0.01036989,
      "balance_loss_clip": 1.04135537,
      "balance_loss_mlp": 1.02281487,
      "epoch": 0.4586201713512701,
      "flos": 29022213974400.0,
      "grad_norm": 1.5863784464980368,
      "language_loss": 0.69371521,
      "learning_rate": 2.3630955687248953e-06,
      "loss": 0.71532381,
      "num_input_tokens_seen": 163657485,
      "step": 7628,
      "time_per_iteration": 2.5832228660583496
    },
    {
      "auxiliary_loss_clip": 0.0110529,
      "auxiliary_loss_mlp": 0.0103064,
      "balance_loss_clip": 1.03810275,
      "balance_loss_mlp": 1.01687765,
      "epoch": 0.45868029460393805,
      "flos": 23404600385280.0,
      "grad_norm": 1.6224546273552722,
      "language_loss": 0.78075004,
      "learning_rate": 2.3627125721874265e-06,
      "loss": 0.8021093,
      "num_input_tokens_seen": 163676030,
      "step": 7629,
      "time_per_iteration": 2.4919581413269043
    },
    {
      "auxiliary_loss_clip": 0.01103298,
      "auxiliary_loss_mlp": 0.01039635,
      "balance_loss_clip": 1.04149127,
      "balance_loss_mlp": 1.02461517,
      "epoch": 0.458740417856606,
      "flos": 18221972497920.0,
      "grad_norm": 2.1339777208700546,
      "language_loss": 0.79405236,
      "learning_rate": 2.3623295618958595e-06,
      "loss": 0.81548166,
      "num_input_tokens_seen": 163694490,
      "step": 7630,
      "time_per_iteration": 2.596471071243286
    },
    {
      "auxiliary_loss_clip": 0.01100869,
      "auxiliary_loss_mlp": 0.01037962,
      "balance_loss_clip": 1.04167938,
      "balance_loss_mlp": 1.0233829,
      "epoch": 0.458800541109274,
      "flos": 34568760504960.0,
      "grad_norm": 1.7609221671558626,
      "language_loss": 0.71637946,
      "learning_rate": 2.3619465378647198e-06,
      "loss": 0.73776776,
      "num_input_tokens_seen": 163717035,
      "step": 7631,
      "time_per_iteration": 2.623997211456299
    },
    {
      "auxiliary_loss_clip": 0.01084976,
      "auxiliary_loss_mlp": 0.0104287,
      "balance_loss_clip": 1.04098201,
      "balance_loss_mlp": 1.02708673,
      "epoch": 0.458860664361942,
      "flos": 17712112896000.0,
      "grad_norm": 2.144255544167101,
      "language_loss": 0.7130543,
      "learning_rate": 2.361563500108531e-06,
      "loss": 0.73433274,
      "num_input_tokens_seen": 163734525,
      "step": 7632,
      "time_per_iteration": 2.5423994064331055
    },
    {
      "auxiliary_loss_clip": 0.01072048,
      "auxiliary_loss_mlp": 0.00791533,
      "balance_loss_clip": 1.03816247,
      "balance_loss_mlp": 1.01211751,
      "epoch": 0.45892078761460997,
      "flos": 18441889516800.0,
      "grad_norm": 2.7195490602890846,
      "language_loss": 0.69922262,
      "learning_rate": 2.3611804486418178e-06,
      "loss": 0.71785843,
      "num_input_tokens_seen": 163752860,
      "step": 7633,
      "time_per_iteration": 2.57700514793396
    },
    {
      "auxiliary_loss_clip": 0.01108469,
      "auxiliary_loss_mlp": 0.010389,
      "balance_loss_clip": 1.04038715,
      "balance_loss_mlp": 1.0247972,
      "epoch": 0.45898091086727794,
      "flos": 22672956257280.0,
      "grad_norm": 1.5397566883618532,
      "language_loss": 0.81118381,
      "learning_rate": 2.3607973834791062e-06,
      "loss": 0.83265758,
      "num_input_tokens_seen": 163772495,
      "step": 7634,
      "time_per_iteration": 2.5451271533966064
    },
    {
      "auxiliary_loss_clip": 0.01110907,
      "auxiliary_loss_mlp": 0.00793467,
      "balance_loss_clip": 1.04217982,
      "balance_loss_mlp": 1.01617789,
      "epoch": 0.4590410341199459,
      "flos": 21652949744640.0,
      "grad_norm": 1.8397511767835342,
      "language_loss": 0.8146944,
      "learning_rate": 2.3604143046349216e-06,
      "loss": 0.83373815,
      "num_input_tokens_seen": 163791475,
      "step": 7635,
      "time_per_iteration": 2.5213987827301025
    },
    {
      "auxiliary_loss_clip": 0.01096505,
      "auxiliary_loss_mlp": 0.010403,
      "balance_loss_clip": 1.04109836,
      "balance_loss_mlp": 1.02680516,
      "epoch": 0.45910115737261387,
      "flos": 36535372087680.0,
      "grad_norm": 1.8742781608241168,
      "language_loss": 0.64844489,
      "learning_rate": 2.3600312121237905e-06,
      "loss": 0.66981292,
      "num_input_tokens_seen": 163812995,
      "step": 7636,
      "time_per_iteration": 2.681605100631714
    },
    {
      "auxiliary_loss_clip": 0.01104089,
      "auxiliary_loss_mlp": 0.01030441,
      "balance_loss_clip": 1.04196239,
      "balance_loss_mlp": 1.01695275,
      "epoch": 0.45916128062528183,
      "flos": 24419866302720.0,
      "grad_norm": 1.4884766745810083,
      "language_loss": 0.80778164,
      "learning_rate": 2.3596481059602395e-06,
      "loss": 0.82912689,
      "num_input_tokens_seen": 163833945,
      "step": 7637,
      "time_per_iteration": 2.5660412311553955
    },
    {
      "auxiliary_loss_clip": 0.01092463,
      "auxiliary_loss_mlp": 0.01037586,
      "balance_loss_clip": 1.0392096,
      "balance_loss_mlp": 1.02168322,
      "epoch": 0.4592214038779498,
      "flos": 23221958705280.0,
      "grad_norm": 1.4506137287407685,
      "language_loss": 0.75485563,
      "learning_rate": 2.3592649861587965e-06,
      "loss": 0.77615613,
      "num_input_tokens_seen": 163853885,
      "step": 7638,
      "time_per_iteration": 2.558304786682129
    },
    {
      "auxiliary_loss_clip": 0.01106265,
      "auxiliary_loss_mlp": 0.01034468,
      "balance_loss_clip": 1.04026937,
      "balance_loss_mlp": 1.02074742,
      "epoch": 0.45928152713061776,
      "flos": 19172133014400.0,
      "grad_norm": 1.7010931043479793,
      "language_loss": 0.74211645,
      "learning_rate": 2.358881852733989e-06,
      "loss": 0.7635237,
      "num_input_tokens_seen": 163871855,
      "step": 7639,
      "time_per_iteration": 2.509464979171753
    },
    {
      "auxiliary_loss_clip": 0.01121694,
      "auxiliary_loss_mlp": 0.01034818,
      "balance_loss_clip": 1.04186797,
      "balance_loss_mlp": 1.02132952,
      "epoch": 0.4593416503832857,
      "flos": 22414686491520.0,
      "grad_norm": 1.5916016146333272,
      "language_loss": 0.68320984,
      "learning_rate": 2.358498705700346e-06,
      "loss": 0.70477498,
      "num_input_tokens_seen": 163891450,
      "step": 7640,
      "time_per_iteration": 2.5085716247558594
    },
    {
      "auxiliary_loss_clip": 0.01101929,
      "auxiliary_loss_mlp": 0.01036831,
      "balance_loss_clip": 1.03980827,
      "balance_loss_mlp": 1.02272868,
      "epoch": 0.4594017736359537,
      "flos": 18880215183360.0,
      "grad_norm": 1.6915228940210105,
      "language_loss": 0.75646693,
      "learning_rate": 2.3581155450723958e-06,
      "loss": 0.77785456,
      "num_input_tokens_seen": 163909345,
      "step": 7641,
      "time_per_iteration": 2.534411907196045
    },
    {
      "auxiliary_loss_clip": 0.01097747,
      "auxiliary_loss_mlp": 0.01031313,
      "balance_loss_clip": 1.04293299,
      "balance_loss_mlp": 1.01634622,
      "epoch": 0.45946189688862166,
      "flos": 20518567349760.0,
      "grad_norm": 1.6774200501312442,
      "language_loss": 0.74939781,
      "learning_rate": 2.357732370864668e-06,
      "loss": 0.77068841,
      "num_input_tokens_seen": 163926940,
      "step": 7642,
      "time_per_iteration": 2.548884153366089
    },
    {
      "auxiliary_loss_clip": 0.01042152,
      "auxiliary_loss_mlp": 0.01003357,
      "balance_loss_clip": 1.03218675,
      "balance_loss_mlp": 1.00184917,
      "epoch": 0.4595220201412896,
      "flos": 61405990162560.0,
      "grad_norm": 0.8419302966472683,
      "language_loss": 0.58213031,
      "learning_rate": 2.357349183091694e-06,
      "loss": 0.60258532,
      "num_input_tokens_seen": 163977785,
      "step": 7643,
      "time_per_iteration": 4.252629518508911
    },
    {
      "auxiliary_loss_clip": 0.0111339,
      "auxiliary_loss_mlp": 0.01033498,
      "balance_loss_clip": 1.03975976,
      "balance_loss_mlp": 1.01926494,
      "epoch": 0.4595821433939576,
      "flos": 23330947547520.0,
      "grad_norm": 1.5301806325876963,
      "language_loss": 0.93004668,
      "learning_rate": 2.3569659817680016e-06,
      "loss": 0.95151556,
      "num_input_tokens_seen": 163996630,
      "step": 7644,
      "time_per_iteration": 2.534205436706543
    },
    {
      "auxiliary_loss_clip": 0.01108079,
      "auxiliary_loss_mlp": 0.01035623,
      "balance_loss_clip": 1.04201889,
      "balance_loss_mlp": 1.02155066,
      "epoch": 0.4596422666466256,
      "flos": 14282356711680.0,
      "grad_norm": 2.06181017288064,
      "language_loss": 0.82900751,
      "learning_rate": 2.3565827669081243e-06,
      "loss": 0.85044456,
      "num_input_tokens_seen": 164013190,
      "step": 7645,
      "time_per_iteration": 3.869338035583496
    },
    {
      "auxiliary_loss_clip": 0.01015267,
      "auxiliary_loss_mlp": 0.01005511,
      "balance_loss_clip": 1.01651347,
      "balance_loss_mlp": 1.00388992,
      "epoch": 0.4597023898992936,
      "flos": 65727337737600.0,
      "grad_norm": 0.7646749821338306,
      "language_loss": 0.59887481,
      "learning_rate": 2.356199538526593e-06,
      "loss": 0.61908263,
      "num_input_tokens_seen": 164074030,
      "step": 7646,
      "time_per_iteration": 3.106841564178467
    },
    {
      "auxiliary_loss_clip": 0.01105205,
      "auxiliary_loss_mlp": 0.01033821,
      "balance_loss_clip": 1.04034245,
      "balance_loss_mlp": 1.01901531,
      "epoch": 0.45976251315196154,
      "flos": 26907075653760.0,
      "grad_norm": 2.077900568253787,
      "language_loss": 0.72630507,
      "learning_rate": 2.355816296637939e-06,
      "loss": 0.74769533,
      "num_input_tokens_seen": 164095515,
      "step": 7647,
      "time_per_iteration": 2.565035104751587
    },
    {
      "auxiliary_loss_clip": 0.01086973,
      "auxiliary_loss_mlp": 0.01032139,
      "balance_loss_clip": 1.03834581,
      "balance_loss_mlp": 1.01850772,
      "epoch": 0.4598226364046295,
      "flos": 26618066824320.0,
      "grad_norm": 1.698220443095354,
      "language_loss": 0.66848505,
      "learning_rate": 2.3554330412566957e-06,
      "loss": 0.68967617,
      "num_input_tokens_seen": 164117270,
      "step": 7648,
      "time_per_iteration": 2.6183369159698486
    },
    {
      "auxiliary_loss_clip": 0.01109208,
      "auxiliary_loss_mlp": 0.01029101,
      "balance_loss_clip": 1.03942943,
      "balance_loss_mlp": 1.01552308,
      "epoch": 0.45988275965729747,
      "flos": 24387762522240.0,
      "grad_norm": 1.4286224261872482,
      "language_loss": 0.78756607,
      "learning_rate": 2.3550497723973953e-06,
      "loss": 0.80894911,
      "num_input_tokens_seen": 164137850,
      "step": 7649,
      "time_per_iteration": 3.909003734588623
    },
    {
      "auxiliary_loss_clip": 0.01061418,
      "auxiliary_loss_mlp": 0.01035294,
      "balance_loss_clip": 1.04134846,
      "balance_loss_mlp": 1.02091122,
      "epoch": 0.45994288290996543,
      "flos": 24535822383360.0,
      "grad_norm": 1.6865603816500283,
      "language_loss": 0.69116116,
      "learning_rate": 2.3546664900745726e-06,
      "loss": 0.71212828,
      "num_input_tokens_seen": 164157960,
      "step": 7650,
      "time_per_iteration": 2.6528146266937256
    },
    {
      "auxiliary_loss_clip": 0.01113658,
      "auxiliary_loss_mlp": 0.0103579,
      "balance_loss_clip": 1.04118979,
      "balance_loss_mlp": 1.02044725,
      "epoch": 0.4600030061626334,
      "flos": 14830245838080.0,
      "grad_norm": 1.9014868323277268,
      "language_loss": 0.83910304,
      "learning_rate": 2.354283194302761e-06,
      "loss": 0.86059749,
      "num_input_tokens_seen": 164174590,
      "step": 7651,
      "time_per_iteration": 3.8575491905212402
    },
    {
      "auxiliary_loss_clip": 0.0109899,
      "auxiliary_loss_mlp": 0.0079038,
      "balance_loss_clip": 1.041381,
      "balance_loss_mlp": 1.01572144,
      "epoch": 0.46006312941530136,
      "flos": 18113845582080.0,
      "grad_norm": 2.1826951985825187,
      "language_loss": 0.75960147,
      "learning_rate": 2.3538998850964948e-06,
      "loss": 0.77849519,
      "num_input_tokens_seen": 164192935,
      "step": 7652,
      "time_per_iteration": 2.5263516902923584
    },
    {
      "auxiliary_loss_clip": 0.01073694,
      "auxiliary_loss_mlp": 0.01029385,
      "balance_loss_clip": 1.03804576,
      "balance_loss_mlp": 1.0145911,
      "epoch": 0.46012325266796933,
      "flos": 21976468565760.0,
      "grad_norm": 1.7107258164037007,
      "language_loss": 0.76244694,
      "learning_rate": 2.3535165624703097e-06,
      "loss": 0.78347772,
      "num_input_tokens_seen": 164213160,
      "step": 7653,
      "time_per_iteration": 2.6429028511047363
    },
    {
      "auxiliary_loss_clip": 0.01080334,
      "auxiliary_loss_mlp": 0.01037799,
      "balance_loss_clip": 1.04263592,
      "balance_loss_mlp": 1.02138376,
      "epoch": 0.4601833759206373,
      "flos": 15268068714240.0,
      "grad_norm": 2.018157079197094,
      "language_loss": 0.6607368,
      "learning_rate": 2.353133226438741e-06,
      "loss": 0.68191814,
      "num_input_tokens_seen": 164229330,
      "step": 7654,
      "time_per_iteration": 2.553253650665283
    },
    {
      "auxiliary_loss_clip": 0.01095063,
      "auxiliary_loss_mlp": 0.01035969,
      "balance_loss_clip": 1.03785217,
      "balance_loss_mlp": 1.02162218,
      "epoch": 0.46024349917330526,
      "flos": 27088999061760.0,
      "grad_norm": 1.7556587733254012,
      "language_loss": 0.79282081,
      "learning_rate": 2.3527498770163248e-06,
      "loss": 0.81413114,
      "num_input_tokens_seen": 164248240,
      "step": 7655,
      "time_per_iteration": 2.5941550731658936
    },
    {
      "auxiliary_loss_clip": 0.01081913,
      "auxiliary_loss_mlp": 0.01031586,
      "balance_loss_clip": 1.03999507,
      "balance_loss_mlp": 1.01734638,
      "epoch": 0.4603036224259732,
      "flos": 24462923731200.0,
      "grad_norm": 1.5479810482361755,
      "language_loss": 0.67717469,
      "learning_rate": 2.3523665142175985e-06,
      "loss": 0.69830966,
      "num_input_tokens_seen": 164268020,
      "step": 7656,
      "time_per_iteration": 2.6021289825439453
    },
    {
      "auxiliary_loss_clip": 0.0109807,
      "auxiliary_loss_mlp": 0.01031397,
      "balance_loss_clip": 1.03947306,
      "balance_loss_mlp": 1.01744962,
      "epoch": 0.4603637456786412,
      "flos": 28109292883200.0,
      "grad_norm": 1.8122647939833152,
      "language_loss": 0.8100704,
      "learning_rate": 2.351983138057098e-06,
      "loss": 0.83136505,
      "num_input_tokens_seen": 164287305,
      "step": 7657,
      "time_per_iteration": 2.611483573913574
    },
    {
      "auxiliary_loss_clip": 0.01119171,
      "auxiliary_loss_mlp": 0.00788671,
      "balance_loss_clip": 1.03998661,
      "balance_loss_mlp": 1.01264083,
      "epoch": 0.4604238689313092,
      "flos": 24348942898560.0,
      "grad_norm": 2.012227400347502,
      "language_loss": 0.7062875,
      "learning_rate": 2.3515997485493623e-06,
      "loss": 0.725366,
      "num_input_tokens_seen": 164306835,
      "step": 7658,
      "time_per_iteration": 2.5044569969177246
    },
    {
      "auxiliary_loss_clip": 0.01030385,
      "auxiliary_loss_mlp": 0.01004023,
      "balance_loss_clip": 1.01189709,
      "balance_loss_mlp": 1.00210381,
      "epoch": 0.4604839921839772,
      "flos": 53606229431040.0,
      "grad_norm": 0.9541891940241639,
      "language_loss": 0.62080538,
      "learning_rate": 2.351216345708928e-06,
      "loss": 0.64114946,
      "num_input_tokens_seen": 164367095,
      "step": 7659,
      "time_per_iteration": 3.202589988708496
    },
    {
      "auxiliary_loss_clip": 0.01071169,
      "auxiliary_loss_mlp": 0.01041641,
      "balance_loss_clip": 1.03894782,
      "balance_loss_mlp": 1.02561378,
      "epoch": 0.46054411543664514,
      "flos": 31248424126080.0,
      "grad_norm": 1.5576642004348105,
      "language_loss": 0.68752116,
      "learning_rate": 2.350832929550336e-06,
      "loss": 0.70864928,
      "num_input_tokens_seen": 164388895,
      "step": 7660,
      "time_per_iteration": 2.6544415950775146
    },
    {
      "auxiliary_loss_clip": 0.01107435,
      "auxiliary_loss_mlp": 0.01036002,
      "balance_loss_clip": 1.03891981,
      "balance_loss_mlp": 1.02121449,
      "epoch": 0.4606042386893131,
      "flos": 24092863862400.0,
      "grad_norm": 1.7053034197622479,
      "language_loss": 0.77097237,
      "learning_rate": 2.3504495000881227e-06,
      "loss": 0.79240668,
      "num_input_tokens_seen": 164409080,
      "step": 7661,
      "time_per_iteration": 2.5493686199188232
    },
    {
      "auxiliary_loss_clip": 0.01103326,
      "auxiliary_loss_mlp": 0.01042335,
      "balance_loss_clip": 1.03970325,
      "balance_loss_mlp": 1.02660489,
      "epoch": 0.46066436194198107,
      "flos": 26578457101440.0,
      "grad_norm": 1.765423355802958,
      "language_loss": 0.74655795,
      "learning_rate": 2.3500660573368305e-06,
      "loss": 0.76801455,
      "num_input_tokens_seen": 164427585,
      "step": 7662,
      "time_per_iteration": 2.5316247940063477
    },
    {
      "auxiliary_loss_clip": 0.01096165,
      "auxiliary_loss_mlp": 0.01036538,
      "balance_loss_clip": 1.03880656,
      "balance_loss_mlp": 1.02073121,
      "epoch": 0.46072448519464904,
      "flos": 17775602184960.0,
      "grad_norm": 3.071430590193815,
      "language_loss": 0.80206239,
      "learning_rate": 2.349682601310998e-06,
      "loss": 0.82338947,
      "num_input_tokens_seen": 164438455,
      "step": 7663,
      "time_per_iteration": 2.4879274368286133
    },
    {
      "auxiliary_loss_clip": 0.01105761,
      "auxiliary_loss_mlp": 0.01032487,
      "balance_loss_clip": 1.03975666,
      "balance_loss_mlp": 1.01855135,
      "epoch": 0.460784608447317,
      "flos": 15086109392640.0,
      "grad_norm": 1.8812340339377516,
      "language_loss": 0.73201686,
      "learning_rate": 2.3492991320251653e-06,
      "loss": 0.75339931,
      "num_input_tokens_seen": 164456830,
      "step": 7664,
      "time_per_iteration": 2.476667642593384
    },
    {
      "auxiliary_loss_clip": 0.01086414,
      "auxiliary_loss_mlp": 0.01035256,
      "balance_loss_clip": 1.04016805,
      "balance_loss_mlp": 1.02176785,
      "epoch": 0.46084473169998497,
      "flos": 18588261438720.0,
      "grad_norm": 1.763503066088783,
      "language_loss": 0.72367668,
      "learning_rate": 2.3489156494938753e-06,
      "loss": 0.74489331,
      "num_input_tokens_seen": 164475375,
      "step": 7665,
      "time_per_iteration": 2.550098180770874
    },
    {
      "auxiliary_loss_clip": 0.01089384,
      "auxiliary_loss_mlp": 0.01031269,
      "balance_loss_clip": 1.03883672,
      "balance_loss_mlp": 1.01800132,
      "epoch": 0.46090485495265293,
      "flos": 19494789909120.0,
      "grad_norm": 1.6729094451826851,
      "language_loss": 0.78279579,
      "learning_rate": 2.348532153731669e-06,
      "loss": 0.80400229,
      "num_input_tokens_seen": 164492040,
      "step": 7666,
      "time_per_iteration": 2.525930404663086
    },
    {
      "auxiliary_loss_clip": 0.0107826,
      "auxiliary_loss_mlp": 0.01034797,
      "balance_loss_clip": 1.03968883,
      "balance_loss_mlp": 1.01916838,
      "epoch": 0.4609649782053209,
      "flos": 33364927163520.0,
      "grad_norm": 1.3671635653073348,
      "language_loss": 0.7382766,
      "learning_rate": 2.348148644753088e-06,
      "loss": 0.75940716,
      "num_input_tokens_seen": 164513665,
      "step": 7667,
      "time_per_iteration": 2.6547675132751465
    },
    {
      "auxiliary_loss_clip": 0.01073016,
      "auxiliary_loss_mlp": 0.01033071,
      "balance_loss_clip": 1.03946304,
      "balance_loss_mlp": 1.01971912,
      "epoch": 0.46102510145798886,
      "flos": 23769165473280.0,
      "grad_norm": 1.4117675009464312,
      "language_loss": 0.76319575,
      "learning_rate": 2.347765122572676e-06,
      "loss": 0.78425658,
      "num_input_tokens_seen": 164533890,
      "step": 7668,
      "time_per_iteration": 2.6041359901428223
    },
    {
      "auxiliary_loss_clip": 0.01069712,
      "auxiliary_loss_mlp": 0.01030773,
      "balance_loss_clip": 1.04336298,
      "balance_loss_mlp": 1.0178628,
      "epoch": 0.4610852247106568,
      "flos": 23294821443840.0,
      "grad_norm": 1.5044116422787515,
      "language_loss": 0.78235716,
      "learning_rate": 2.347381587204975e-06,
      "loss": 0.80336201,
      "num_input_tokens_seen": 164553815,
      "step": 7669,
      "time_per_iteration": 2.619983673095703
    },
    {
      "auxiliary_loss_clip": 0.01103597,
      "auxiliary_loss_mlp": 0.01037644,
      "balance_loss_clip": 1.0378896,
      "balance_loss_mlp": 1.02172971,
      "epoch": 0.4611453479633248,
      "flos": 25447450584960.0,
      "grad_norm": 2.003135880179364,
      "language_loss": 0.8251074,
      "learning_rate": 2.34699803866453e-06,
      "loss": 0.84651983,
      "num_input_tokens_seen": 164573125,
      "step": 7670,
      "time_per_iteration": 2.5754334926605225
    },
    {
      "auxiliary_loss_clip": 0.01106972,
      "auxiliary_loss_mlp": 0.01033029,
      "balance_loss_clip": 1.03882694,
      "balance_loss_mlp": 1.01949894,
      "epoch": 0.4612054712159928,
      "flos": 21139606523520.0,
      "grad_norm": 1.6971171386104273,
      "language_loss": 0.63679534,
      "learning_rate": 2.3466144769658845e-06,
      "loss": 0.65819538,
      "num_input_tokens_seen": 164592575,
      "step": 7671,
      "time_per_iteration": 2.5200605392456055
    },
    {
      "auxiliary_loss_clip": 0.01024823,
      "auxiliary_loss_mlp": 0.0100358,
      "balance_loss_clip": 1.01737523,
      "balance_loss_mlp": 1.00195873,
      "epoch": 0.4612655944686608,
      "flos": 69959266404480.0,
      "grad_norm": 0.6989707125355941,
      "language_loss": 0.55913389,
      "learning_rate": 2.346230902123583e-06,
      "loss": 0.57941794,
      "num_input_tokens_seen": 164659795,
      "step": 7672,
      "time_per_iteration": 3.2214865684509277
    },
    {
      "auxiliary_loss_clip": 0.01110825,
      "auxiliary_loss_mlp": 0.01035798,
      "balance_loss_clip": 1.04137814,
      "balance_loss_mlp": 1.02210045,
      "epoch": 0.46132571772132874,
      "flos": 16837149502080.0,
      "grad_norm": 1.8526541007839679,
      "language_loss": 0.71173739,
      "learning_rate": 2.3458473141521715e-06,
      "loss": 0.73320365,
      "num_input_tokens_seen": 164678735,
      "step": 7673,
      "time_per_iteration": 2.471381664276123
    },
    {
      "auxiliary_loss_clip": 0.01094855,
      "auxiliary_loss_mlp": 0.01032052,
      "balance_loss_clip": 1.03973222,
      "balance_loss_mlp": 1.01810503,
      "epoch": 0.4613858409739967,
      "flos": 35808935431680.0,
      "grad_norm": 1.7210410752930523,
      "language_loss": 0.70637441,
      "learning_rate": 2.345463713066195e-06,
      "loss": 0.72764355,
      "num_input_tokens_seen": 164700885,
      "step": 7674,
      "time_per_iteration": 2.6591646671295166
    },
    {
      "auxiliary_loss_clip": 0.01094978,
      "auxiliary_loss_mlp": 0.01037482,
      "balance_loss_clip": 1.038414,
      "balance_loss_mlp": 1.02358866,
      "epoch": 0.4614459642266647,
      "flos": 35266756567680.0,
      "grad_norm": 1.5110608931118745,
      "language_loss": 0.65785176,
      "learning_rate": 2.3450800988801996e-06,
      "loss": 0.67917633,
      "num_input_tokens_seen": 164726960,
      "step": 7675,
      "time_per_iteration": 2.671116828918457
    },
    {
      "auxiliary_loss_clip": 0.01041789,
      "auxiliary_loss_mlp": 0.01001623,
      "balance_loss_clip": 1.0143019,
      "balance_loss_mlp": 1.00020409,
      "epoch": 0.46150608747933264,
      "flos": 66704610044160.0,
      "grad_norm": 0.8220956935109288,
      "language_loss": 0.58616233,
      "learning_rate": 2.3446964716087327e-06,
      "loss": 0.60659641,
      "num_input_tokens_seen": 164788525,
      "step": 7676,
      "time_per_iteration": 3.106959104537964
    },
    {
      "auxiliary_loss_clip": 0.01015114,
      "auxiliary_loss_mlp": 0.01000722,
      "balance_loss_clip": 1.01788282,
      "balance_loss_mlp": 0.99926722,
      "epoch": 0.4615662107320006,
      "flos": 55830177025920.0,
      "grad_norm": 0.7888027773719017,
      "language_loss": 0.62663752,
      "learning_rate": 2.344312831266341e-06,
      "loss": 0.64679587,
      "num_input_tokens_seen": 164843525,
      "step": 7677,
      "time_per_iteration": 3.0071446895599365
    },
    {
      "auxiliary_loss_clip": 0.0109576,
      "auxiliary_loss_mlp": 0.01031816,
      "balance_loss_clip": 1.04063618,
      "balance_loss_mlp": 1.01916814,
      "epoch": 0.46162633398466857,
      "flos": 15483245137920.0,
      "grad_norm": 2.2101449459433122,
      "language_loss": 0.76387417,
      "learning_rate": 2.3439291778675718e-06,
      "loss": 0.78514999,
      "num_input_tokens_seen": 164859895,
      "step": 7678,
      "time_per_iteration": 2.5216281414031982
    },
    {
      "auxiliary_loss_clip": 0.01121224,
      "auxiliary_loss_mlp": 0.01031527,
      "balance_loss_clip": 1.04209828,
      "balance_loss_mlp": 1.01750207,
      "epoch": 0.46168645723733653,
      "flos": 20011437181440.0,
      "grad_norm": 1.9514510588442076,
      "language_loss": 0.66871035,
      "learning_rate": 2.343545511426974e-06,
      "loss": 0.69023782,
      "num_input_tokens_seen": 164878030,
      "step": 7679,
      "time_per_iteration": 2.4820895195007324
    },
    {
      "auxiliary_loss_clip": 0.01087137,
      "auxiliary_loss_mlp": 0.01036182,
      "balance_loss_clip": 1.04032731,
      "balance_loss_mlp": 1.02287829,
      "epoch": 0.4617465804900045,
      "flos": 20298542590080.0,
      "grad_norm": 1.8177495470725336,
      "language_loss": 0.70005095,
      "learning_rate": 2.3431618319590963e-06,
      "loss": 0.72128403,
      "num_input_tokens_seen": 164895710,
      "step": 7680,
      "time_per_iteration": 2.564502716064453
    },
    {
      "auxiliary_loss_clip": 0.01128247,
      "auxiliary_loss_mlp": 0.01044718,
      "balance_loss_clip": 1.04595828,
      "balance_loss_mlp": 1.03015614,
      "epoch": 0.46180670374267246,
      "flos": 22346312952960.0,
      "grad_norm": 1.6948636091688467,
      "language_loss": 0.63357991,
      "learning_rate": 2.342778139478487e-06,
      "loss": 0.65530956,
      "num_input_tokens_seen": 164913365,
      "step": 7681,
      "time_per_iteration": 2.461862564086914
    },
    {
      "auxiliary_loss_clip": 0.01105626,
      "auxiliary_loss_mlp": 0.01030681,
      "balance_loss_clip": 1.03897905,
      "balance_loss_mlp": 1.01769948,
      "epoch": 0.46186682699534043,
      "flos": 19895696582400.0,
      "grad_norm": 1.4304829874092857,
      "language_loss": 0.67451274,
      "learning_rate": 2.342394433999697e-06,
      "loss": 0.69587582,
      "num_input_tokens_seen": 164931620,
      "step": 7682,
      "time_per_iteration": 3.886735677719116
    },
    {
      "auxiliary_loss_clip": 0.01081577,
      "auxiliary_loss_mlp": 0.0103836,
      "balance_loss_clip": 1.04018164,
      "balance_loss_mlp": 1.02411473,
      "epoch": 0.4619269502480084,
      "flos": 31503569408640.0,
      "grad_norm": 2.0372989331438967,
      "language_loss": 0.73769724,
      "learning_rate": 2.342010715537275e-06,
      "loss": 0.75889671,
      "num_input_tokens_seen": 164950905,
      "step": 7683,
      "time_per_iteration": 2.623055934906006
    },
    {
      "auxiliary_loss_clip": 0.01119567,
      "auxiliary_loss_mlp": 0.0103395,
      "balance_loss_clip": 1.04215288,
      "balance_loss_mlp": 1.0204854,
      "epoch": 0.46198707350067636,
      "flos": 25009484054400.0,
      "grad_norm": 1.7280956742610316,
      "language_loss": 0.76175058,
      "learning_rate": 2.3416269841057726e-06,
      "loss": 0.78328574,
      "num_input_tokens_seen": 164970950,
      "step": 7684,
      "time_per_iteration": 4.328751087188721
    },
    {
      "auxiliary_loss_clip": 0.01126335,
      "auxiliary_loss_mlp": 0.01039418,
      "balance_loss_clip": 1.04404974,
      "balance_loss_mlp": 1.02515495,
      "epoch": 0.4620471967533444,
      "flos": 18292357198080.0,
      "grad_norm": 1.8526383160449416,
      "language_loss": 0.79703903,
      "learning_rate": 2.3412432397197412e-06,
      "loss": 0.81869656,
      "num_input_tokens_seen": 164989855,
      "step": 7685,
      "time_per_iteration": 2.466585636138916
    },
    {
      "auxiliary_loss_clip": 0.01081243,
      "auxiliary_loss_mlp": 0.01041395,
      "balance_loss_clip": 1.04605615,
      "balance_loss_mlp": 1.02641678,
      "epoch": 0.46210732000601235,
      "flos": 33985104410880.0,
      "grad_norm": 1.6851256628803053,
      "language_loss": 0.66549659,
      "learning_rate": 2.340859482393731e-06,
      "loss": 0.68672293,
      "num_input_tokens_seen": 165012290,
      "step": 7686,
      "time_per_iteration": 2.6781020164489746
    },
    {
      "auxiliary_loss_clip": 0.01100377,
      "auxiliary_loss_mlp": 0.00789326,
      "balance_loss_clip": 1.04238904,
      "balance_loss_mlp": 1.01172185,
      "epoch": 0.4621674432586803,
      "flos": 25009412227200.0,
      "grad_norm": 2.3162017582398384,
      "language_loss": 0.73812568,
      "learning_rate": 2.340475712142296e-06,
      "loss": 0.75702274,
      "num_input_tokens_seen": 165030810,
      "step": 7687,
      "time_per_iteration": 4.028199195861816
    },
    {
      "auxiliary_loss_clip": 0.01056044,
      "auxiliary_loss_mlp": 0.01030527,
      "balance_loss_clip": 1.04256916,
      "balance_loss_mlp": 1.01635885,
      "epoch": 0.4622275665113483,
      "flos": 22014031213440.0,
      "grad_norm": 2.7328188495499823,
      "language_loss": 0.74931192,
      "learning_rate": 2.3400919289799873e-06,
      "loss": 0.7701776,
      "num_input_tokens_seen": 165050205,
      "step": 7688,
      "time_per_iteration": 2.642214059829712
    },
    {
      "auxiliary_loss_clip": 0.01069587,
      "auxiliary_loss_mlp": 0.007884,
      "balance_loss_clip": 1.03822267,
      "balance_loss_mlp": 1.0111419,
      "epoch": 0.46228768976401624,
      "flos": 24058820747520.0,
      "grad_norm": 1.8254642045945535,
      "language_loss": 0.79026461,
      "learning_rate": 2.3397081329213585e-06,
      "loss": 0.80884451,
      "num_input_tokens_seen": 165069370,
      "step": 7689,
      "time_per_iteration": 4.026331424713135
    },
    {
      "auxiliary_loss_clip": 0.01109207,
      "auxiliary_loss_mlp": 0.01036642,
      "balance_loss_clip": 1.03988111,
      "balance_loss_mlp": 1.02172899,
      "epoch": 0.4623478130166842,
      "flos": 26651391667200.0,
      "grad_norm": 2.286629313103284,
      "language_loss": 0.57157046,
      "learning_rate": 2.339324323980964e-06,
      "loss": 0.5930289,
      "num_input_tokens_seen": 165089610,
      "step": 7690,
      "time_per_iteration": 2.5442843437194824
    },
    {
      "auxiliary_loss_clip": 0.01109013,
      "auxiliary_loss_mlp": 0.01034134,
      "balance_loss_clip": 1.04087782,
      "balance_loss_mlp": 1.02018619,
      "epoch": 0.46240793626935217,
      "flos": 20558428467840.0,
      "grad_norm": 1.988708727304612,
      "language_loss": 0.82691473,
      "learning_rate": 2.3389405021733562e-06,
      "loss": 0.84834617,
      "num_input_tokens_seen": 165109050,
      "step": 7691,
      "time_per_iteration": 2.492231845855713
    },
    {
      "auxiliary_loss_clip": 0.0109876,
      "auxiliary_loss_mlp": 0.01027752,
      "balance_loss_clip": 1.04081774,
      "balance_loss_mlp": 1.01453793,
      "epoch": 0.46246805952202014,
      "flos": 22456055980800.0,
      "grad_norm": 1.5312903978158867,
      "language_loss": 0.75474274,
      "learning_rate": 2.338556667513091e-06,
      "loss": 0.77600789,
      "num_input_tokens_seen": 165130130,
      "step": 7692,
      "time_per_iteration": 2.5442233085632324
    },
    {
      "auxiliary_loss_clip": 0.01090252,
      "auxiliary_loss_mlp": 0.01037017,
      "balance_loss_clip": 1.04133844,
      "balance_loss_mlp": 1.02262831,
      "epoch": 0.4625281827746881,
      "flos": 35041308854400.0,
      "grad_norm": 1.6199728044497228,
      "language_loss": 0.74061632,
      "learning_rate": 2.338172820014723e-06,
      "loss": 0.76188904,
      "num_input_tokens_seen": 165152685,
      "step": 7693,
      "time_per_iteration": 2.6615331172943115
    },
    {
      "auxiliary_loss_clip": 0.01079445,
      "auxiliary_loss_mlp": 0.01046493,
      "balance_loss_clip": 1.04113126,
      "balance_loss_mlp": 1.03107977,
      "epoch": 0.46258830602735607,
      "flos": 21068647205760.0,
      "grad_norm": 1.5392312943629898,
      "language_loss": 0.85664201,
      "learning_rate": 2.337788959692808e-06,
      "loss": 0.87790138,
      "num_input_tokens_seen": 165173315,
      "step": 7694,
      "time_per_iteration": 2.5553884506225586
    },
    {
      "auxiliary_loss_clip": 0.01100079,
      "auxiliary_loss_mlp": 0.0104082,
      "balance_loss_clip": 1.04042149,
      "balance_loss_mlp": 1.02743936,
      "epoch": 0.46264842928002403,
      "flos": 26177227205760.0,
      "grad_norm": 1.984011202539224,
      "language_loss": 0.78849781,
      "learning_rate": 2.337405086561902e-06,
      "loss": 0.80990684,
      "num_input_tokens_seen": 165192395,
      "step": 7695,
      "time_per_iteration": 2.555619478225708
    },
    {
      "auxiliary_loss_clip": 0.01104425,
      "auxiliary_loss_mlp": 0.01032235,
      "balance_loss_clip": 1.0392797,
      "balance_loss_mlp": 1.01910377,
      "epoch": 0.462708552532692,
      "flos": 16764214936320.0,
      "grad_norm": 1.7699832113288594,
      "language_loss": 0.72457474,
      "learning_rate": 2.3370212006365606e-06,
      "loss": 0.74594134,
      "num_input_tokens_seen": 165211355,
      "step": 7696,
      "time_per_iteration": 2.470168352127075
    },
    {
      "auxiliary_loss_clip": 0.010982,
      "auxiliary_loss_mlp": 0.0103739,
      "balance_loss_clip": 1.04032063,
      "balance_loss_mlp": 1.0231328,
      "epoch": 0.46276867578535996,
      "flos": 15560453422080.0,
      "grad_norm": 1.5835126638543793,
      "language_loss": 0.69817591,
      "learning_rate": 2.3366373019313423e-06,
      "loss": 0.71953177,
      "num_input_tokens_seen": 165229380,
      "step": 7697,
      "time_per_iteration": 2.501528739929199
    },
    {
      "auxiliary_loss_clip": 0.01119059,
      "auxiliary_loss_mlp": 0.01029787,
      "balance_loss_clip": 1.04218268,
      "balance_loss_mlp": 1.01672149,
      "epoch": 0.462828799038028,
      "flos": 22415404763520.0,
      "grad_norm": 2.3272337162454857,
      "language_loss": 0.84601039,
      "learning_rate": 2.3362533904608025e-06,
      "loss": 0.86749887,
      "num_input_tokens_seen": 165247200,
      "step": 7698,
      "time_per_iteration": 2.4781899452209473
    },
    {
      "auxiliary_loss_clip": 0.0111914,
      "auxiliary_loss_mlp": 0.01033101,
      "balance_loss_clip": 1.04183519,
      "balance_loss_mlp": 1.0199697,
      "epoch": 0.46288892229069595,
      "flos": 21069580959360.0,
      "grad_norm": 2.0092190696938625,
      "language_loss": 0.7102077,
      "learning_rate": 2.335869466239502e-06,
      "loss": 0.7317301,
      "num_input_tokens_seen": 165265825,
      "step": 7699,
      "time_per_iteration": 2.4430227279663086
    },
    {
      "auxiliary_loss_clip": 0.01069208,
      "auxiliary_loss_mlp": 0.01033348,
      "balance_loss_clip": 1.039253,
      "balance_loss_mlp": 1.01895392,
      "epoch": 0.4629490455433639,
      "flos": 23185688947200.0,
      "grad_norm": 1.8784100116108031,
      "language_loss": 0.71567023,
      "learning_rate": 2.335485529281996e-06,
      "loss": 0.73669583,
      "num_input_tokens_seen": 165284380,
      "step": 7700,
      "time_per_iteration": 2.597999334335327
    },
    {
      "auxiliary_loss_clip": 0.01117006,
      "auxiliary_loss_mlp": 0.00786505,
      "balance_loss_clip": 1.03965521,
      "balance_loss_mlp": 1.00879729,
      "epoch": 0.4630091687960319,
      "flos": 18835541642880.0,
      "grad_norm": 2.0507341327455255,
      "language_loss": 0.7250576,
      "learning_rate": 2.3351015796028467e-06,
      "loss": 0.74409264,
      "num_input_tokens_seen": 165300320,
      "step": 7701,
      "time_per_iteration": 2.4569966793060303
    },
    {
      "auxiliary_loss_clip": 0.01077481,
      "auxiliary_loss_mlp": 0.01035126,
      "balance_loss_clip": 1.04007864,
      "balance_loss_mlp": 1.02077353,
      "epoch": 0.46306929204869984,
      "flos": 38907020407680.0,
      "grad_norm": 2.091919920128807,
      "language_loss": 0.64534295,
      "learning_rate": 2.3347176172166114e-06,
      "loss": 0.66646898,
      "num_input_tokens_seen": 165318130,
      "step": 7702,
      "time_per_iteration": 2.6831111907958984
    },
    {
      "auxiliary_loss_clip": 0.01092241,
      "auxiliary_loss_mlp": 0.01027803,
      "balance_loss_clip": 1.03940725,
      "balance_loss_mlp": 1.01445794,
      "epoch": 0.4631294153013678,
      "flos": 19644178573440.0,
      "grad_norm": 2.052567610124491,
      "language_loss": 0.73763227,
      "learning_rate": 2.33433364213785e-06,
      "loss": 0.75883275,
      "num_input_tokens_seen": 165336225,
      "step": 7703,
      "time_per_iteration": 2.528388261795044
    },
    {
      "auxiliary_loss_clip": 0.01098096,
      "auxiliary_loss_mlp": 0.01033677,
      "balance_loss_clip": 1.04039407,
      "balance_loss_mlp": 1.01933014,
      "epoch": 0.4631895385540358,
      "flos": 24608254158720.0,
      "grad_norm": 1.7230614529144024,
      "language_loss": 0.69168186,
      "learning_rate": 2.3339496543811243e-06,
      "loss": 0.71299958,
      "num_input_tokens_seen": 165355005,
      "step": 7704,
      "time_per_iteration": 2.5417864322662354
    },
    {
      "auxiliary_loss_clip": 0.01109415,
      "auxiliary_loss_mlp": 0.01026997,
      "balance_loss_clip": 1.04257524,
      "balance_loss_mlp": 1.01303196,
      "epoch": 0.46324966180670374,
      "flos": 26320115508480.0,
      "grad_norm": 2.8049463022045926,
      "language_loss": 0.81378317,
      "learning_rate": 2.3335656539609934e-06,
      "loss": 0.83514726,
      "num_input_tokens_seen": 165374910,
      "step": 7705,
      "time_per_iteration": 2.5289628505706787
    },
    {
      "auxiliary_loss_clip": 0.01112798,
      "auxiliary_loss_mlp": 0.01031687,
      "balance_loss_clip": 1.04030728,
      "balance_loss_mlp": 1.01836514,
      "epoch": 0.4633097850593717,
      "flos": 19240506552960.0,
      "grad_norm": 1.8889933376802064,
      "language_loss": 0.77424723,
      "learning_rate": 2.3331816408920196e-06,
      "loss": 0.79569209,
      "num_input_tokens_seen": 165392590,
      "step": 7706,
      "time_per_iteration": 2.468928813934326
    },
    {
      "auxiliary_loss_clip": 0.01089299,
      "auxiliary_loss_mlp": 0.01028455,
      "balance_loss_clip": 1.04187787,
      "balance_loss_mlp": 1.01559854,
      "epoch": 0.46336990831203967,
      "flos": 22783166161920.0,
      "grad_norm": 1.9210812452162274,
      "language_loss": 0.70211184,
      "learning_rate": 2.3327976151887654e-06,
      "loss": 0.72328937,
      "num_input_tokens_seen": 165411195,
      "step": 7707,
      "time_per_iteration": 2.516860246658325
    },
    {
      "auxiliary_loss_clip": 0.0109902,
      "auxiliary_loss_mlp": 0.01035662,
      "balance_loss_clip": 1.03833234,
      "balance_loss_mlp": 1.02080822,
      "epoch": 0.46343003156470763,
      "flos": 38210604543360.0,
      "grad_norm": 2.217064240128672,
      "language_loss": 0.60830623,
      "learning_rate": 2.332413576865791e-06,
      "loss": 0.62965304,
      "num_input_tokens_seen": 165430150,
      "step": 7708,
      "time_per_iteration": 2.649106979370117
    },
    {
      "auxiliary_loss_clip": 0.01081182,
      "auxiliary_loss_mlp": 0.01029262,
      "balance_loss_clip": 1.04025376,
      "balance_loss_mlp": 1.01549947,
      "epoch": 0.4634901548173756,
      "flos": 31938555110400.0,
      "grad_norm": 1.9385647275717128,
      "language_loss": 0.77335364,
      "learning_rate": 2.3320295259376614e-06,
      "loss": 0.79445809,
      "num_input_tokens_seen": 165450595,
      "step": 7709,
      "time_per_iteration": 2.63010835647583
    },
    {
      "auxiliary_loss_clip": 0.01121136,
      "auxiliary_loss_mlp": 0.01038412,
      "balance_loss_clip": 1.041605,
      "balance_loss_mlp": 1.02444625,
      "epoch": 0.46355027807004356,
      "flos": 20082540153600.0,
      "grad_norm": 2.2393731375008277,
      "language_loss": 0.76939744,
      "learning_rate": 2.3316454624189385e-06,
      "loss": 0.79099292,
      "num_input_tokens_seen": 165469515,
      "step": 7710,
      "time_per_iteration": 2.457853317260742
    },
    {
      "auxiliary_loss_clip": 0.01111419,
      "auxiliary_loss_mlp": 0.01032317,
      "balance_loss_clip": 1.04102743,
      "balance_loss_mlp": 1.01729631,
      "epoch": 0.4636104013227116,
      "flos": 24061370613120.0,
      "grad_norm": 2.2805322457174664,
      "language_loss": 0.73436117,
      "learning_rate": 2.3312613863241865e-06,
      "loss": 0.75579852,
      "num_input_tokens_seen": 165488125,
      "step": 7711,
      "time_per_iteration": 2.505941867828369
    },
    {
      "auxiliary_loss_clip": 0.0110033,
      "auxiliary_loss_mlp": 0.01044622,
      "balance_loss_clip": 1.04200947,
      "balance_loss_mlp": 1.03055501,
      "epoch": 0.46367052457537955,
      "flos": 23914639555200.0,
      "grad_norm": 1.3153904872717346,
      "language_loss": 0.71470094,
      "learning_rate": 2.33087729766797e-06,
      "loss": 0.73615044,
      "num_input_tokens_seen": 165509225,
      "step": 7712,
      "time_per_iteration": 2.5354371070861816
    },
    {
      "auxiliary_loss_clip": 0.01096787,
      "auxiliary_loss_mlp": 0.01036064,
      "balance_loss_clip": 1.04048181,
      "balance_loss_mlp": 1.01999485,
      "epoch": 0.4637306478280475,
      "flos": 26396533693440.0,
      "grad_norm": 1.826385248557772,
      "language_loss": 0.72965932,
      "learning_rate": 2.3304931964648524e-06,
      "loss": 0.75098789,
      "num_input_tokens_seen": 165529945,
      "step": 7713,
      "time_per_iteration": 2.5611751079559326
    },
    {
      "auxiliary_loss_clip": 0.0108618,
      "auxiliary_loss_mlp": 0.01036593,
      "balance_loss_clip": 1.03890359,
      "balance_loss_mlp": 1.02159643,
      "epoch": 0.4637907710807155,
      "flos": 21980706370560.0,
      "grad_norm": 1.8639041408237584,
      "language_loss": 0.58601058,
      "learning_rate": 2.3301090827294e-06,
      "loss": 0.60723829,
      "num_input_tokens_seen": 165550690,
      "step": 7714,
      "time_per_iteration": 2.5711097717285156
    },
    {
      "auxiliary_loss_clip": 0.01106237,
      "auxiliary_loss_mlp": 0.01027717,
      "balance_loss_clip": 1.03976274,
      "balance_loss_mlp": 1.01429439,
      "epoch": 0.46385089433338345,
      "flos": 12422291846400.0,
      "grad_norm": 2.0394830182093013,
      "language_loss": 0.70224643,
      "learning_rate": 2.3297249564761784e-06,
      "loss": 0.72358596,
      "num_input_tokens_seen": 165567775,
      "step": 7715,
      "time_per_iteration": 2.4826362133026123
    },
    {
      "auxiliary_loss_clip": 0.01125258,
      "auxiliary_loss_mlp": 0.01032213,
      "balance_loss_clip": 1.04203713,
      "balance_loss_mlp": 1.01833689,
      "epoch": 0.4639110175860514,
      "flos": 23915752876800.0,
      "grad_norm": 1.842812977558961,
      "language_loss": 0.68472016,
      "learning_rate": 2.3293408177197527e-06,
      "loss": 0.70629489,
      "num_input_tokens_seen": 165587010,
      "step": 7716,
      "time_per_iteration": 2.4966373443603516
    },
    {
      "auxiliary_loss_clip": 0.01122155,
      "auxiliary_loss_mlp": 0.01028961,
      "balance_loss_clip": 1.04184914,
      "balance_loss_mlp": 1.01429832,
      "epoch": 0.4639711408387194,
      "flos": 25300396304640.0,
      "grad_norm": 1.5544837750882015,
      "language_loss": 0.80950403,
      "learning_rate": 2.328956666474691e-06,
      "loss": 0.83101517,
      "num_input_tokens_seen": 165607850,
      "step": 7717,
      "time_per_iteration": 2.5004947185516357
    },
    {
      "auxiliary_loss_clip": 0.01120239,
      "auxiliary_loss_mlp": 0.01032207,
      "balance_loss_clip": 1.04096198,
      "balance_loss_mlp": 1.01820588,
      "epoch": 0.46403126409138734,
      "flos": 21211822817280.0,
      "grad_norm": 1.7732864542382256,
      "language_loss": 0.73185825,
      "learning_rate": 2.3285725027555593e-06,
      "loss": 0.75338268,
      "num_input_tokens_seen": 165627175,
      "step": 7718,
      "time_per_iteration": 2.4812440872192383
    },
    {
      "auxiliary_loss_clip": 0.01117014,
      "auxiliary_loss_mlp": 0.00788514,
      "balance_loss_clip": 1.03952599,
      "balance_loss_mlp": 1.01359737,
      "epoch": 0.4640913873440553,
      "flos": 35845564325760.0,
      "grad_norm": 1.6892350551723885,
      "language_loss": 0.70344579,
      "learning_rate": 2.3281883265769254e-06,
      "loss": 0.72250116,
      "num_input_tokens_seen": 165648340,
      "step": 7719,
      "time_per_iteration": 2.581648826599121
    },
    {
      "auxiliary_loss_clip": 0.01094536,
      "auxiliary_loss_mlp": 0.01038906,
      "balance_loss_clip": 1.04018426,
      "balance_loss_mlp": 1.02463698,
      "epoch": 0.46415151059672327,
      "flos": 19166207270400.0,
      "grad_norm": 1.7572685751858477,
      "language_loss": 0.86780953,
      "learning_rate": 2.327804137953357e-06,
      "loss": 0.889144,
      "num_input_tokens_seen": 165667195,
      "step": 7720,
      "time_per_iteration": 2.5443360805511475
    },
    {
      "auxiliary_loss_clip": 0.01023437,
      "auxiliary_loss_mlp": 0.01001766,
      "balance_loss_clip": 1.01634669,
      "balance_loss_mlp": 1.00006092,
      "epoch": 0.46421163384939124,
      "flos": 58912750304640.0,
      "grad_norm": 0.7345087865902721,
      "language_loss": 0.55043364,
      "learning_rate": 2.3274199368994226e-06,
      "loss": 0.57068563,
      "num_input_tokens_seen": 165726760,
      "step": 7721,
      "time_per_iteration": 4.473818063735962
    },
    {
      "auxiliary_loss_clip": 0.01097783,
      "auxiliary_loss_mlp": 0.0103775,
      "balance_loss_clip": 1.04168725,
      "balance_loss_mlp": 1.02377295,
      "epoch": 0.4642717571020592,
      "flos": 20157342226560.0,
      "grad_norm": 1.9485585225214987,
      "language_loss": 0.79649895,
      "learning_rate": 2.3270357234296918e-06,
      "loss": 0.81785429,
      "num_input_tokens_seen": 165745005,
      "step": 7722,
      "time_per_iteration": 2.5162031650543213
    },
    {
      "auxiliary_loss_clip": 0.01124674,
      "auxiliary_loss_mlp": 0.01034889,
      "balance_loss_clip": 1.04344118,
      "balance_loss_mlp": 1.02097094,
      "epoch": 0.46433188035472717,
      "flos": 25046184775680.0,
      "grad_norm": 1.5233806165953534,
      "language_loss": 0.7783348,
      "learning_rate": 2.3266514975587332e-06,
      "loss": 0.79993045,
      "num_input_tokens_seen": 165765750,
      "step": 7723,
      "time_per_iteration": 3.8636393547058105
    },
    {
      "auxiliary_loss_clip": 0.01027051,
      "auxiliary_loss_mlp": 0.01035562,
      "balance_loss_clip": 1.04024005,
      "balance_loss_mlp": 1.02228236,
      "epoch": 0.4643920036073952,
      "flos": 28075644817920.0,
      "grad_norm": 1.752273523095809,
      "language_loss": 0.68474835,
      "learning_rate": 2.326267259301118e-06,
      "loss": 0.70537448,
      "num_input_tokens_seen": 165787515,
      "step": 7724,
      "time_per_iteration": 2.8889546394348145
    },
    {
      "auxiliary_loss_clip": 0.01101981,
      "auxiliary_loss_mlp": 0.01032939,
      "balance_loss_clip": 1.041412,
      "balance_loss_mlp": 1.01915216,
      "epoch": 0.46445212686006315,
      "flos": 18369350000640.0,
      "grad_norm": 2.5042636373490317,
      "language_loss": 0.67005563,
      "learning_rate": 2.325883008671415e-06,
      "loss": 0.69140482,
      "num_input_tokens_seen": 165806675,
      "step": 7725,
      "time_per_iteration": 2.68361234664917
    },
    {
      "auxiliary_loss_clip": 0.0110366,
      "auxiliary_loss_mlp": 0.01039991,
      "balance_loss_clip": 1.04111969,
      "balance_loss_mlp": 1.02777767,
      "epoch": 0.4645122501127311,
      "flos": 31721618920320.0,
      "grad_norm": 1.7304440232393894,
      "language_loss": 0.65004623,
      "learning_rate": 2.3254987456841955e-06,
      "loss": 0.67148274,
      "num_input_tokens_seen": 165829835,
      "step": 7726,
      "time_per_iteration": 4.290181398391724
    },
    {
      "auxiliary_loss_clip": 0.01097556,
      "auxiliary_loss_mlp": 0.00789469,
      "balance_loss_clip": 1.04345274,
      "balance_loss_mlp": 1.01277804,
      "epoch": 0.4645723733653991,
      "flos": 23768806337280.0,
      "grad_norm": 1.7559302670064163,
      "language_loss": 0.74857241,
      "learning_rate": 2.3251144703540307e-06,
      "loss": 0.76744264,
      "num_input_tokens_seen": 165849380,
      "step": 7727,
      "time_per_iteration": 2.5930943489074707
    },
    {
      "auxiliary_loss_clip": 0.01096691,
      "auxiliary_loss_mlp": 0.01043322,
      "balance_loss_clip": 1.04132688,
      "balance_loss_mlp": 1.02923203,
      "epoch": 0.46463249661806705,
      "flos": 33145512935040.0,
      "grad_norm": 1.974641867602,
      "language_loss": 0.7852428,
      "learning_rate": 2.3247301826954936e-06,
      "loss": 0.80664295,
      "num_input_tokens_seen": 165868620,
      "step": 7728,
      "time_per_iteration": 4.205172061920166
    },
    {
      "auxiliary_loss_clip": 0.01087579,
      "auxiliary_loss_mlp": 0.0104157,
      "balance_loss_clip": 1.04251158,
      "balance_loss_mlp": 1.02755094,
      "epoch": 0.464692619870735,
      "flos": 18296020385280.0,
      "grad_norm": 2.296302998201104,
      "language_loss": 0.76009619,
      "learning_rate": 2.324345882723155e-06,
      "loss": 0.78138769,
      "num_input_tokens_seen": 165885915,
      "step": 7729,
      "time_per_iteration": 2.5763397216796875
    },
    {
      "auxiliary_loss_clip": 0.01096444,
      "auxiliary_loss_mlp": 0.01048863,
      "balance_loss_clip": 1.04140365,
      "balance_loss_mlp": 1.03322279,
      "epoch": 0.464752743123403,
      "flos": 22638051216000.0,
      "grad_norm": 1.7039507964934366,
      "language_loss": 0.80081117,
      "learning_rate": 2.323961570451588e-06,
      "loss": 0.82226419,
      "num_input_tokens_seen": 165905465,
      "step": 7730,
      "time_per_iteration": 2.584210157394409
    },
    {
      "auxiliary_loss_clip": 0.01119555,
      "auxiliary_loss_mlp": 0.01035517,
      "balance_loss_clip": 1.04214478,
      "balance_loss_mlp": 1.02255309,
      "epoch": 0.46481286637607094,
      "flos": 20412128373120.0,
      "grad_norm": 1.53602550272045,
      "language_loss": 0.7719748,
      "learning_rate": 2.3235772458953655e-06,
      "loss": 0.79352552,
      "num_input_tokens_seen": 165924640,
      "step": 7731,
      "time_per_iteration": 2.4774513244628906
    },
    {
      "auxiliary_loss_clip": 0.01079136,
      "auxiliary_loss_mlp": 0.01030558,
      "balance_loss_clip": 1.04084063,
      "balance_loss_mlp": 1.01711667,
      "epoch": 0.4648729896287389,
      "flos": 34275406129920.0,
      "grad_norm": 4.3548211996039905,
      "language_loss": 0.66017175,
      "learning_rate": 2.323192909069061e-06,
      "loss": 0.68126869,
      "num_input_tokens_seen": 165945765,
      "step": 7732,
      "time_per_iteration": 2.656026840209961
    },
    {
      "auxiliary_loss_clip": 0.01099452,
      "auxiliary_loss_mlp": 0.01043894,
      "balance_loss_clip": 1.04157043,
      "balance_loss_mlp": 1.02849197,
      "epoch": 0.4649331128814069,
      "flos": 21321781326720.0,
      "grad_norm": 2.3000012640744343,
      "language_loss": 0.72725958,
      "learning_rate": 2.32280855998725e-06,
      "loss": 0.74869299,
      "num_input_tokens_seen": 165964025,
      "step": 7733,
      "time_per_iteration": 2.548311471939087
    },
    {
      "auxiliary_loss_clip": 0.01042071,
      "auxiliary_loss_mlp": 0.01005962,
      "balance_loss_clip": 1.01558065,
      "balance_loss_mlp": 1.00429344,
      "epoch": 0.46499323613407484,
      "flos": 58308515717760.0,
      "grad_norm": 1.2256078106869954,
      "language_loss": 0.5191825,
      "learning_rate": 2.3224241986645057e-06,
      "loss": 0.53966278,
      "num_input_tokens_seen": 166021950,
      "step": 7734,
      "time_per_iteration": 3.016254186630249
    },
    {
      "auxiliary_loss_clip": 0.01097316,
      "auxiliary_loss_mlp": 0.01030427,
      "balance_loss_clip": 1.04528999,
      "balance_loss_mlp": 1.01651502,
      "epoch": 0.4650533593867428,
      "flos": 10889660384640.0,
      "grad_norm": 1.9338551794930403,
      "language_loss": 0.75945437,
      "learning_rate": 2.3220398251154035e-06,
      "loss": 0.7807318,
      "num_input_tokens_seen": 166039675,
      "step": 7735,
      "time_per_iteration": 2.52327823638916
    },
    {
      "auxiliary_loss_clip": 0.01075503,
      "auxiliary_loss_mlp": 0.0103847,
      "balance_loss_clip": 1.04125381,
      "balance_loss_mlp": 1.02439713,
      "epoch": 0.46511348263941077,
      "flos": 19974592805760.0,
      "grad_norm": 1.8272204273418973,
      "language_loss": 0.69826007,
      "learning_rate": 2.321655439354519e-06,
      "loss": 0.71939981,
      "num_input_tokens_seen": 166057745,
      "step": 7736,
      "time_per_iteration": 2.5655429363250732
    },
    {
      "auxiliary_loss_clip": 0.01119166,
      "auxiliary_loss_mlp": 0.01031816,
      "balance_loss_clip": 1.04420066,
      "balance_loss_mlp": 1.0194068,
      "epoch": 0.46517360589207873,
      "flos": 19678401256320.0,
      "grad_norm": 1.6239642428290886,
      "language_loss": 0.72159749,
      "learning_rate": 2.321271041396427e-06,
      "loss": 0.74310732,
      "num_input_tokens_seen": 166076440,
      "step": 7737,
      "time_per_iteration": 2.4780867099761963
    },
    {
      "auxiliary_loss_clip": 0.0110596,
      "auxiliary_loss_mlp": 0.0103537,
      "balance_loss_clip": 1.04677153,
      "balance_loss_mlp": 1.02085066,
      "epoch": 0.46523372914474675,
      "flos": 16872665074560.0,
      "grad_norm": 1.9133728843776778,
      "language_loss": 0.83646643,
      "learning_rate": 2.3208866312557065e-06,
      "loss": 0.85787976,
      "num_input_tokens_seen": 166092520,
      "step": 7738,
      "time_per_iteration": 2.4867942333221436
    },
    {
      "auxiliary_loss_clip": 0.01031819,
      "auxiliary_loss_mlp": 0.01002502,
      "balance_loss_clip": 1.01491046,
      "balance_loss_mlp": 1.00092888,
      "epoch": 0.4652938523974147,
      "flos": 53439138339840.0,
      "grad_norm": 0.7611536118798935,
      "language_loss": 0.5783515,
      "learning_rate": 2.320502208946932e-06,
      "loss": 0.59869468,
      "num_input_tokens_seen": 166156285,
      "step": 7739,
      "time_per_iteration": 3.1521754264831543
    },
    {
      "auxiliary_loss_clip": 0.0110212,
      "auxiliary_loss_mlp": 0.01038262,
      "balance_loss_clip": 1.04559946,
      "balance_loss_mlp": 1.02473783,
      "epoch": 0.4653539756500827,
      "flos": 15231296165760.0,
      "grad_norm": 1.838040744933791,
      "language_loss": 0.85096025,
      "learning_rate": 2.3201177744846815e-06,
      "loss": 0.87236404,
      "num_input_tokens_seen": 166173455,
      "step": 7740,
      "time_per_iteration": 2.544041872024536
    },
    {
      "auxiliary_loss_clip": 0.01097522,
      "auxiliary_loss_mlp": 0.01034706,
      "balance_loss_clip": 1.04482722,
      "balance_loss_mlp": 1.02038908,
      "epoch": 0.46541409890275065,
      "flos": 23732249270400.0,
      "grad_norm": 2.144216194559229,
      "language_loss": 0.75812727,
      "learning_rate": 2.3197333278835327e-06,
      "loss": 0.77944958,
      "num_input_tokens_seen": 166194370,
      "step": 7741,
      "time_per_iteration": 2.554856061935425
    },
    {
      "auxiliary_loss_clip": 0.01092479,
      "auxiliary_loss_mlp": 0.01036915,
      "balance_loss_clip": 1.04293716,
      "balance_loss_mlp": 1.02349186,
      "epoch": 0.4654742221554186,
      "flos": 20847329556480.0,
      "grad_norm": 1.9452364256123638,
      "language_loss": 0.81020474,
      "learning_rate": 2.319348869158064e-06,
      "loss": 0.83149868,
      "num_input_tokens_seen": 166213195,
      "step": 7742,
      "time_per_iteration": 2.5594208240509033
    },
    {
      "auxiliary_loss_clip": 0.01100494,
      "auxiliary_loss_mlp": 0.01039096,
      "balance_loss_clip": 1.04219377,
      "balance_loss_mlp": 1.02477944,
      "epoch": 0.4655343454080866,
      "flos": 20704836303360.0,
      "grad_norm": 1.6604379484289034,
      "language_loss": 0.72637701,
      "learning_rate": 2.3189643983228555e-06,
      "loss": 0.74777293,
      "num_input_tokens_seen": 166231350,
      "step": 7743,
      "time_per_iteration": 2.522688627243042
    },
    {
      "auxiliary_loss_clip": 0.01087408,
      "auxiliary_loss_mlp": 0.01029513,
      "balance_loss_clip": 1.04212797,
      "balance_loss_mlp": 1.01545215,
      "epoch": 0.46559446866075455,
      "flos": 18989850470400.0,
      "grad_norm": 2.4220844355991926,
      "language_loss": 0.71052551,
      "learning_rate": 2.318579915392483e-06,
      "loss": 0.73169476,
      "num_input_tokens_seen": 166250530,
      "step": 7744,
      "time_per_iteration": 2.5307161808013916
    },
    {
      "auxiliary_loss_clip": 0.01078532,
      "auxiliary_loss_mlp": 0.01026822,
      "balance_loss_clip": 1.04600739,
      "balance_loss_mlp": 1.01421022,
      "epoch": 0.4656545919134225,
      "flos": 34496364643200.0,
      "grad_norm": 1.498853352547733,
      "language_loss": 0.84978235,
      "learning_rate": 2.31819542038153e-06,
      "loss": 0.87083584,
      "num_input_tokens_seen": 166272545,
      "step": 7745,
      "time_per_iteration": 2.733593225479126
    },
    {
      "auxiliary_loss_clip": 0.0110838,
      "auxiliary_loss_mlp": 0.01035209,
      "balance_loss_clip": 1.04254234,
      "balance_loss_mlp": 1.02133894,
      "epoch": 0.4657147151660905,
      "flos": 24310554238080.0,
      "grad_norm": 1.3545103482103438,
      "language_loss": 0.72861671,
      "learning_rate": 2.317810913304574e-06,
      "loss": 0.75005257,
      "num_input_tokens_seen": 166292135,
      "step": 7746,
      "time_per_iteration": 2.539642095565796
    },
    {
      "auxiliary_loss_clip": 0.01107687,
      "auxiliary_loss_mlp": 0.01035249,
      "balance_loss_clip": 1.04511094,
      "balance_loss_mlp": 1.02202892,
      "epoch": 0.46577483841875844,
      "flos": 58795139220480.0,
      "grad_norm": 1.7019900797669245,
      "language_loss": 0.70105875,
      "learning_rate": 2.3174263941761963e-06,
      "loss": 0.72248816,
      "num_input_tokens_seen": 166316710,
      "step": 7747,
      "time_per_iteration": 2.8398334980010986
    },
    {
      "auxiliary_loss_clip": 0.0107973,
      "auxiliary_loss_mlp": 0.01037722,
      "balance_loss_clip": 1.03959417,
      "balance_loss_mlp": 1.02261806,
      "epoch": 0.4658349616714264,
      "flos": 31321969223040.0,
      "grad_norm": 1.4594100560552439,
      "language_loss": 0.67525494,
      "learning_rate": 2.317041863010978e-06,
      "loss": 0.69642949,
      "num_input_tokens_seen": 166338535,
      "step": 7748,
      "time_per_iteration": 2.661412477493286
    },
    {
      "auxiliary_loss_clip": 0.01085955,
      "auxiliary_loss_mlp": 0.01035706,
      "balance_loss_clip": 1.04428136,
      "balance_loss_mlp": 1.0202086,
      "epoch": 0.46589508492409437,
      "flos": 14860338456960.0,
      "grad_norm": 1.9434076440979862,
      "language_loss": 0.6396181,
      "learning_rate": 2.3166573198235007e-06,
      "loss": 0.66083473,
      "num_input_tokens_seen": 166355540,
      "step": 7749,
      "time_per_iteration": 2.5158324241638184
    },
    {
      "auxiliary_loss_clip": 0.0111475,
      "auxiliary_loss_mlp": 0.01034149,
      "balance_loss_clip": 1.04484582,
      "balance_loss_mlp": 1.01932502,
      "epoch": 0.46595520817676234,
      "flos": 12895989431040.0,
      "grad_norm": 2.1289787477738256,
      "language_loss": 0.74871409,
      "learning_rate": 2.3162727646283456e-06,
      "loss": 0.77020299,
      "num_input_tokens_seen": 166372635,
      "step": 7750,
      "time_per_iteration": 2.5047662258148193
    },
    {
      "auxiliary_loss_clip": 0.01106027,
      "auxiliary_loss_mlp": 0.01028411,
      "balance_loss_clip": 1.0439347,
      "balance_loss_mlp": 1.01398683,
      "epoch": 0.46601533142943036,
      "flos": 32854169721600.0,
      "grad_norm": 2.0242678847766613,
      "language_loss": 0.74018157,
      "learning_rate": 2.3158881974400963e-06,
      "loss": 0.76152599,
      "num_input_tokens_seen": 166393175,
      "step": 7751,
      "time_per_iteration": 2.6138193607330322
    },
    {
      "auxiliary_loss_clip": 0.01090812,
      "auxiliary_loss_mlp": 0.01032046,
      "balance_loss_clip": 1.04297495,
      "balance_loss_mlp": 1.01754451,
      "epoch": 0.4660754546820983,
      "flos": 19967517826560.0,
      "grad_norm": 1.80002853460792,
      "language_loss": 0.73253942,
      "learning_rate": 2.3155036182733345e-06,
      "loss": 0.75376803,
      "num_input_tokens_seen": 166408630,
      "step": 7752,
      "time_per_iteration": 2.598094940185547
    },
    {
      "auxiliary_loss_clip": 0.01095325,
      "auxiliary_loss_mlp": 0.01032063,
      "balance_loss_clip": 1.04391468,
      "balance_loss_mlp": 1.01839614,
      "epoch": 0.4661355779347663,
      "flos": 26688164215680.0,
      "grad_norm": 2.2877430454946346,
      "language_loss": 0.68848234,
      "learning_rate": 2.315119027142644e-06,
      "loss": 0.7097562,
      "num_input_tokens_seen": 166428170,
      "step": 7753,
      "time_per_iteration": 2.575422763824463
    },
    {
      "auxiliary_loss_clip": 0.01094217,
      "auxiliary_loss_mlp": 0.01033142,
      "balance_loss_clip": 1.04422128,
      "balance_loss_mlp": 1.0193789,
      "epoch": 0.46619570118743425,
      "flos": 20959442881920.0,
      "grad_norm": 1.8567069868149921,
      "language_loss": 0.72733343,
      "learning_rate": 2.3147344240626076e-06,
      "loss": 0.74860704,
      "num_input_tokens_seen": 166446705,
      "step": 7754,
      "time_per_iteration": 2.575521230697632
    },
    {
      "auxiliary_loss_clip": 0.01101046,
      "auxiliary_loss_mlp": 0.01028362,
      "balance_loss_clip": 1.04231429,
      "balance_loss_mlp": 1.01423585,
      "epoch": 0.4662558244401022,
      "flos": 24426079355520.0,
      "grad_norm": 1.5780660496469205,
      "language_loss": 0.78762019,
      "learning_rate": 2.3143498090478114e-06,
      "loss": 0.80891424,
      "num_input_tokens_seen": 166466750,
      "step": 7755,
      "time_per_iteration": 2.561568021774292
    },
    {
      "auxiliary_loss_clip": 0.0110499,
      "auxiliary_loss_mlp": 0.01027175,
      "balance_loss_clip": 1.04136801,
      "balance_loss_mlp": 1.01409817,
      "epoch": 0.4663159476927702,
      "flos": 20595452411520.0,
      "grad_norm": 1.7536979059193405,
      "language_loss": 0.72348523,
      "learning_rate": 2.3139651821128382e-06,
      "loss": 0.74480689,
      "num_input_tokens_seen": 166485400,
      "step": 7756,
      "time_per_iteration": 2.532601833343506
    },
    {
      "auxiliary_loss_clip": 0.0110671,
      "auxiliary_loss_mlp": 0.0102983,
      "balance_loss_clip": 1.04089355,
      "balance_loss_mlp": 1.01622224,
      "epoch": 0.46637607094543815,
      "flos": 25661872823040.0,
      "grad_norm": 1.6691703228491073,
      "language_loss": 0.77888852,
      "learning_rate": 2.313580543272274e-06,
      "loss": 0.80025387,
      "num_input_tokens_seen": 166505730,
      "step": 7757,
      "time_per_iteration": 2.5754804611206055
    },
    {
      "auxiliary_loss_clip": 0.01080074,
      "auxiliary_loss_mlp": 0.01030242,
      "balance_loss_clip": 1.03928137,
      "balance_loss_mlp": 1.01724219,
      "epoch": 0.4664361941981061,
      "flos": 24273853516800.0,
      "grad_norm": 1.9725481817779147,
      "language_loss": 0.66091484,
      "learning_rate": 2.313195892540705e-06,
      "loss": 0.68201798,
      "num_input_tokens_seen": 166523770,
      "step": 7758,
      "time_per_iteration": 2.5847082138061523
    },
    {
      "auxiliary_loss_clip": 0.01092305,
      "auxiliary_loss_mlp": 0.01035587,
      "balance_loss_clip": 1.04262066,
      "balance_loss_mlp": 1.0218184,
      "epoch": 0.4664963174507741,
      "flos": 18405871153920.0,
      "grad_norm": 1.7121169313829245,
      "language_loss": 0.74617004,
      "learning_rate": 2.3128112299327147e-06,
      "loss": 0.76744896,
      "num_input_tokens_seen": 166542935,
      "step": 7759,
      "time_per_iteration": 2.508730888366699
    },
    {
      "auxiliary_loss_clip": 0.01098881,
      "auxiliary_loss_mlp": 0.01034406,
      "balance_loss_clip": 1.0432632,
      "balance_loss_mlp": 1.02168643,
      "epoch": 0.46655644070344204,
      "flos": 22455122227200.0,
      "grad_norm": 1.5440393967651211,
      "language_loss": 0.77907872,
      "learning_rate": 2.312426555462893e-06,
      "loss": 0.80041164,
      "num_input_tokens_seen": 166563935,
      "step": 7760,
      "time_per_iteration": 3.9425315856933594
    },
    {
      "auxiliary_loss_clip": 0.01091909,
      "auxiliary_loss_mlp": 0.01028966,
      "balance_loss_clip": 1.03896379,
      "balance_loss_mlp": 1.0159725,
      "epoch": 0.46661656395611,
      "flos": 13808407731840.0,
      "grad_norm": 1.910291063183726,
      "language_loss": 0.74280322,
      "learning_rate": 2.3120418691458237e-06,
      "loss": 0.76401198,
      "num_input_tokens_seen": 166582175,
      "step": 7761,
      "time_per_iteration": 3.89027738571167
    },
    {
      "auxiliary_loss_clip": 0.0110887,
      "auxiliary_loss_mlp": 0.01038883,
      "balance_loss_clip": 1.04518032,
      "balance_loss_mlp": 1.02367759,
      "epoch": 0.466676687208778,
      "flos": 21652159645440.0,
      "grad_norm": 1.740244403852362,
      "language_loss": 0.78757137,
      "learning_rate": 2.3116571709960956e-06,
      "loss": 0.80904889,
      "num_input_tokens_seen": 166601870,
      "step": 7762,
      "time_per_iteration": 2.5022125244140625
    },
    {
      "auxiliary_loss_clip": 0.01028445,
      "auxiliary_loss_mlp": 0.01003018,
      "balance_loss_clip": 1.01245606,
      "balance_loss_mlp": 1.00144434,
      "epoch": 0.46673681046144594,
      "flos": 68534259068160.0,
      "grad_norm": 0.7945625112745611,
      "language_loss": 0.59798241,
      "learning_rate": 2.311272461028297e-06,
      "loss": 0.61829704,
      "num_input_tokens_seen": 166668960,
      "step": 7763,
      "time_per_iteration": 3.1578493118286133
    },
    {
      "auxiliary_loss_clip": 0.01081604,
      "auxiliary_loss_mlp": 0.01036536,
      "balance_loss_clip": 1.04097986,
      "balance_loss_mlp": 1.0211997,
      "epoch": 0.46679693371411396,
      "flos": 15814449469440.0,
      "grad_norm": 2.489887650683663,
      "language_loss": 0.78928566,
      "learning_rate": 2.3108877392570146e-06,
      "loss": 0.81046706,
      "num_input_tokens_seen": 166686110,
      "step": 7764,
      "time_per_iteration": 3.9242348670959473
    },
    {
      "auxiliary_loss_clip": 0.01088922,
      "auxiliary_loss_mlp": 0.01033015,
      "balance_loss_clip": 1.04300296,
      "balance_loss_mlp": 1.02038443,
      "epoch": 0.4668570569667819,
      "flos": 18514572687360.0,
      "grad_norm": 1.8246583984764693,
      "language_loss": 0.72423416,
      "learning_rate": 2.310503005696839e-06,
      "loss": 0.74545348,
      "num_input_tokens_seen": 166703930,
      "step": 7765,
      "time_per_iteration": 2.531252384185791
    },
    {
      "auxiliary_loss_clip": 0.01076088,
      "auxiliary_loss_mlp": 0.01031719,
      "balance_loss_clip": 1.04068065,
      "balance_loss_mlp": 1.01812911,
      "epoch": 0.4669171802194499,
      "flos": 19206643006080.0,
      "grad_norm": 1.9175954286081385,
      "language_loss": 0.77831745,
      "learning_rate": 2.3101182603623576e-06,
      "loss": 0.79939556,
      "num_input_tokens_seen": 166719940,
      "step": 7766,
      "time_per_iteration": 2.5352258682250977
    },
    {
      "auxiliary_loss_clip": 0.01102645,
      "auxiliary_loss_mlp": 0.0103339,
      "balance_loss_clip": 1.03975749,
      "balance_loss_mlp": 1.01993728,
      "epoch": 0.46697730347211786,
      "flos": 12276135406080.0,
      "grad_norm": 3.620814830873186,
      "language_loss": 0.64600128,
      "learning_rate": 2.3097335032681607e-06,
      "loss": 0.66736162,
      "num_input_tokens_seen": 166738285,
      "step": 7767,
      "time_per_iteration": 3.856947422027588
    },
    {
      "auxiliary_loss_clip": 0.01108676,
      "auxiliary_loss_mlp": 0.01037764,
      "balance_loss_clip": 1.04229057,
      "balance_loss_mlp": 1.02434123,
      "epoch": 0.4670374267247858,
      "flos": 23586739274880.0,
      "grad_norm": 1.979717476635691,
      "language_loss": 0.74515891,
      "learning_rate": 2.3093487344288393e-06,
      "loss": 0.76662326,
      "num_input_tokens_seen": 166758170,
      "step": 7768,
      "time_per_iteration": 2.5261070728302
    },
    {
      "auxiliary_loss_clip": 0.01090034,
      "auxiliary_loss_mlp": 0.01026116,
      "balance_loss_clip": 1.04488611,
      "balance_loss_mlp": 1.01314616,
      "epoch": 0.4670975499774538,
      "flos": 15991093578240.0,
      "grad_norm": 1.757373041646167,
      "language_loss": 0.70678699,
      "learning_rate": 2.308963953858982e-06,
      "loss": 0.72794855,
      "num_input_tokens_seen": 166775750,
      "step": 7769,
      "time_per_iteration": 2.5165154933929443
    },
    {
      "auxiliary_loss_clip": 0.01117767,
      "auxiliary_loss_mlp": 0.01031727,
      "balance_loss_clip": 1.04103374,
      "balance_loss_mlp": 1.01890564,
      "epoch": 0.46715767323012175,
      "flos": 15377596260480.0,
      "grad_norm": 1.9186799145266846,
      "language_loss": 0.81349564,
      "learning_rate": 2.3085791615731803e-06,
      "loss": 0.83499062,
      "num_input_tokens_seen": 166791720,
      "step": 7770,
      "time_per_iteration": 2.4689180850982666
    },
    {
      "auxiliary_loss_clip": 0.010363,
      "auxiliary_loss_mlp": 0.01002437,
      "balance_loss_clip": 1.00999415,
      "balance_loss_mlp": 1.00087547,
      "epoch": 0.4672177964827897,
      "flos": 60252217401600.0,
      "grad_norm": 0.8202379013701127,
      "language_loss": 0.55595064,
      "learning_rate": 2.3081943575860265e-06,
      "loss": 0.57633793,
      "num_input_tokens_seen": 166856360,
      "step": 7771,
      "time_per_iteration": 3.073331832885742
    },
    {
      "auxiliary_loss_clip": 0.0110471,
      "auxiliary_loss_mlp": 0.00785829,
      "balance_loss_clip": 1.04168379,
      "balance_loss_mlp": 1.00958645,
      "epoch": 0.4672779197354577,
      "flos": 27636134002560.0,
      "grad_norm": 1.9086491710168587,
      "language_loss": 0.66117942,
      "learning_rate": 2.3078095419121117e-06,
      "loss": 0.68008482,
      "num_input_tokens_seen": 166875925,
      "step": 7772,
      "time_per_iteration": 2.5498743057250977
    },
    {
      "auxiliary_loss_clip": 0.01103612,
      "auxiliary_loss_mlp": 0.01031181,
      "balance_loss_clip": 1.04354715,
      "balance_loss_mlp": 1.01859856,
      "epoch": 0.46733804298812565,
      "flos": 31394257344000.0,
      "grad_norm": 1.8745131584463803,
      "language_loss": 0.6389088,
      "learning_rate": 2.3074247145660283e-06,
      "loss": 0.66025674,
      "num_input_tokens_seen": 166896520,
      "step": 7773,
      "time_per_iteration": 2.6010642051696777
    },
    {
      "auxiliary_loss_clip": 0.01097536,
      "auxiliary_loss_mlp": 0.01034008,
      "balance_loss_clip": 1.04027724,
      "balance_loss_mlp": 1.02015007,
      "epoch": 0.4673981662407936,
      "flos": 19500607912320.0,
      "grad_norm": 1.8147572773654406,
      "language_loss": 0.80300564,
      "learning_rate": 2.3070398755623685e-06,
      "loss": 0.82432109,
      "num_input_tokens_seen": 166915370,
      "step": 7774,
      "time_per_iteration": 2.505866050720215
    },
    {
      "auxiliary_loss_clip": 0.0108598,
      "auxiliary_loss_mlp": 0.01029989,
      "balance_loss_clip": 1.04209292,
      "balance_loss_mlp": 1.01629162,
      "epoch": 0.4674582894934616,
      "flos": 20521835487360.0,
      "grad_norm": 1.6090827505560474,
      "language_loss": 0.77447701,
      "learning_rate": 2.306655024915726e-06,
      "loss": 0.79563665,
      "num_input_tokens_seen": 166934875,
      "step": 7775,
      "time_per_iteration": 2.5865416526794434
    },
    {
      "auxiliary_loss_clip": 0.01085391,
      "auxiliary_loss_mlp": 0.01032283,
      "balance_loss_clip": 1.04042137,
      "balance_loss_mlp": 1.0188179,
      "epoch": 0.46751841274612954,
      "flos": 22090952188800.0,
      "grad_norm": 11.02748829987365,
      "language_loss": 0.69357753,
      "learning_rate": 2.306270162640694e-06,
      "loss": 0.71475422,
      "num_input_tokens_seen": 166954285,
      "step": 7776,
      "time_per_iteration": 2.5515105724334717
    },
    {
      "auxiliary_loss_clip": 0.01109643,
      "auxiliary_loss_mlp": 0.01029245,
      "balance_loss_clip": 1.04358172,
      "balance_loss_mlp": 1.01690102,
      "epoch": 0.46757853599879756,
      "flos": 26980082046720.0,
      "grad_norm": 1.5096878050670963,
      "language_loss": 0.73525393,
      "learning_rate": 2.3058852887518678e-06,
      "loss": 0.75664276,
      "num_input_tokens_seen": 166975975,
      "step": 7777,
      "time_per_iteration": 2.5589287281036377
    },
    {
      "auxiliary_loss_clip": 0.01106795,
      "auxiliary_loss_mlp": 0.01030944,
      "balance_loss_clip": 1.04172242,
      "balance_loss_mlp": 1.01782477,
      "epoch": 0.4676386592514655,
      "flos": 24134053783680.0,
      "grad_norm": 2.243164947834768,
      "language_loss": 0.69572246,
      "learning_rate": 2.3055004032638394e-06,
      "loss": 0.71709985,
      "num_input_tokens_seen": 166996140,
      "step": 7778,
      "time_per_iteration": 2.52089262008667
    },
    {
      "auxiliary_loss_clip": 0.01108111,
      "auxiliary_loss_mlp": 0.01039829,
      "balance_loss_clip": 1.04287267,
      "balance_loss_mlp": 1.02661443,
      "epoch": 0.4676987825041335,
      "flos": 25483720343040.0,
      "grad_norm": 2.1247482545681096,
      "language_loss": 0.73655951,
      "learning_rate": 2.305115506191206e-06,
      "loss": 0.75803888,
      "num_input_tokens_seen": 167016105,
      "step": 7779,
      "time_per_iteration": 2.5840559005737305
    },
    {
      "auxiliary_loss_clip": 0.01072687,
      "auxiliary_loss_mlp": 0.01037005,
      "balance_loss_clip": 1.03887057,
      "balance_loss_mlp": 1.02442884,
      "epoch": 0.46775890575680146,
      "flos": 21945298538880.0,
      "grad_norm": 1.6326438102656085,
      "language_loss": 0.72078359,
      "learning_rate": 2.304730597548562e-06,
      "loss": 0.74188048,
      "num_input_tokens_seen": 167036185,
      "step": 7780,
      "time_per_iteration": 2.6115572452545166
    },
    {
      "auxiliary_loss_clip": 0.01093028,
      "auxiliary_loss_mlp": 0.01048159,
      "balance_loss_clip": 1.04099655,
      "balance_loss_mlp": 1.03211331,
      "epoch": 0.4678190290094694,
      "flos": 25228395492480.0,
      "grad_norm": 1.851462687193211,
      "language_loss": 0.74498624,
      "learning_rate": 2.3043456773505023e-06,
      "loss": 0.76639807,
      "num_input_tokens_seen": 167054515,
      "step": 7781,
      "time_per_iteration": 2.564302444458008
    },
    {
      "auxiliary_loss_clip": 0.01107927,
      "auxiliary_loss_mlp": 0.01034937,
      "balance_loss_clip": 1.04054487,
      "balance_loss_mlp": 1.02113867,
      "epoch": 0.4678791522621374,
      "flos": 32268358811520.0,
      "grad_norm": 1.6428876065416453,
      "language_loss": 0.62826061,
      "learning_rate": 2.3039607456116252e-06,
      "loss": 0.64968932,
      "num_input_tokens_seen": 167077245,
      "step": 7782,
      "time_per_iteration": 2.5739686489105225
    },
    {
      "auxiliary_loss_clip": 0.01097712,
      "auxiliary_loss_mlp": 0.01038862,
      "balance_loss_clip": 1.03938222,
      "balance_loss_mlp": 1.02558231,
      "epoch": 0.46793927551480535,
      "flos": 27046480337280.0,
      "grad_norm": 2.4752652224333445,
      "language_loss": 0.63278008,
      "learning_rate": 2.3035758023465254e-06,
      "loss": 0.65414584,
      "num_input_tokens_seen": 167097235,
      "step": 7783,
      "time_per_iteration": 2.563169002532959
    },
    {
      "auxiliary_loss_clip": 0.01109193,
      "auxiliary_loss_mlp": 0.01036532,
      "balance_loss_clip": 1.04439998,
      "balance_loss_mlp": 1.02201855,
      "epoch": 0.4679993987674733,
      "flos": 17457398576640.0,
      "grad_norm": 2.5885363019199414,
      "language_loss": 0.67707098,
      "learning_rate": 2.303190847569801e-06,
      "loss": 0.69852823,
      "num_input_tokens_seen": 167113155,
      "step": 7784,
      "time_per_iteration": 2.465085744857788
    },
    {
      "auxiliary_loss_clip": 0.0108769,
      "auxiliary_loss_mlp": 0.01027551,
      "balance_loss_clip": 1.04178929,
      "balance_loss_mlp": 1.01518321,
      "epoch": 0.4680595220201413,
      "flos": 17165121609600.0,
      "grad_norm": 2.0317813450862934,
      "language_loss": 0.84221959,
      "learning_rate": 2.3028058812960497e-06,
      "loss": 0.86337203,
      "num_input_tokens_seen": 167131765,
      "step": 7785,
      "time_per_iteration": 2.503460645675659
    },
    {
      "auxiliary_loss_clip": 0.01082189,
      "auxiliary_loss_mlp": 0.01032526,
      "balance_loss_clip": 1.04066288,
      "balance_loss_mlp": 1.01882911,
      "epoch": 0.46811964527280925,
      "flos": 11327591001600.0,
      "grad_norm": 1.8095435468508374,
      "language_loss": 0.77291536,
      "learning_rate": 2.3024209035398678e-06,
      "loss": 0.79406255,
      "num_input_tokens_seen": 167149030,
      "step": 7786,
      "time_per_iteration": 2.531744956970215
    },
    {
      "auxiliary_loss_clip": 0.0110056,
      "auxiliary_loss_mlp": 0.01027167,
      "balance_loss_clip": 1.03814387,
      "balance_loss_mlp": 1.01519227,
      "epoch": 0.4681797685254772,
      "flos": 24278809593600.0,
      "grad_norm": 2.051352693624312,
      "language_loss": 0.7424162,
      "learning_rate": 2.302035914315856e-06,
      "loss": 0.76369345,
      "num_input_tokens_seen": 167167375,
      "step": 7787,
      "time_per_iteration": 2.5101876258850098
    },
    {
      "auxiliary_loss_clip": 0.01089649,
      "auxiliary_loss_mlp": 0.0103599,
      "balance_loss_clip": 1.03978109,
      "balance_loss_mlp": 1.02253091,
      "epoch": 0.4682398917781452,
      "flos": 31650372293760.0,
      "grad_norm": 1.6813381866235253,
      "language_loss": 0.65419388,
      "learning_rate": 2.3016509136386116e-06,
      "loss": 0.67545021,
      "num_input_tokens_seen": 167188065,
      "step": 7788,
      "time_per_iteration": 2.6235709190368652
    },
    {
      "auxiliary_loss_clip": 0.01105487,
      "auxiliary_loss_mlp": 0.01029152,
      "balance_loss_clip": 1.04091871,
      "balance_loss_mlp": 1.01737392,
      "epoch": 0.46830001503081314,
      "flos": 28110765340800.0,
      "grad_norm": 1.7185100266540785,
      "language_loss": 0.64017069,
      "learning_rate": 2.3012659015227343e-06,
      "loss": 0.66151714,
      "num_input_tokens_seen": 167209675,
      "step": 7789,
      "time_per_iteration": 2.5686206817626953
    },
    {
      "auxiliary_loss_clip": 0.01030313,
      "auxiliary_loss_mlp": 0.01001798,
      "balance_loss_clip": 1.01331091,
      "balance_loss_mlp": 1.00040913,
      "epoch": 0.4683601382834811,
      "flos": 57881718316800.0,
      "grad_norm": 0.7071927290681167,
      "language_loss": 0.61891329,
      "learning_rate": 2.300880877982825e-06,
      "loss": 0.63923442,
      "num_input_tokens_seen": 167273940,
      "step": 7790,
      "time_per_iteration": 3.1747379302978516
    },
    {
      "auxiliary_loss_clip": 0.01075716,
      "auxiliary_loss_mlp": 0.01037331,
      "balance_loss_clip": 1.04040253,
      "balance_loss_mlp": 1.02319241,
      "epoch": 0.46842026153614913,
      "flos": 21871933009920.0,
      "grad_norm": 1.5664434720871605,
      "language_loss": 0.79323459,
      "learning_rate": 2.3004958430334808e-06,
      "loss": 0.81436509,
      "num_input_tokens_seen": 167292730,
      "step": 7791,
      "time_per_iteration": 2.575923442840576
    },
    {
      "auxiliary_loss_clip": 0.01107139,
      "auxiliary_loss_mlp": 0.01031444,
      "balance_loss_clip": 1.04109466,
      "balance_loss_mlp": 1.01816988,
      "epoch": 0.4684803847888171,
      "flos": 24900818434560.0,
      "grad_norm": 1.6835025185055241,
      "language_loss": 0.74798787,
      "learning_rate": 2.3001107966893052e-06,
      "loss": 0.76937371,
      "num_input_tokens_seen": 167313460,
      "step": 7792,
      "time_per_iteration": 2.516066312789917
    },
    {
      "auxiliary_loss_clip": 0.01085837,
      "auxiliary_loss_mlp": 0.01036716,
      "balance_loss_clip": 1.03865528,
      "balance_loss_mlp": 1.02282214,
      "epoch": 0.46854050804148506,
      "flos": 26251670142720.0,
      "grad_norm": 1.498163292303882,
      "language_loss": 0.68419969,
      "learning_rate": 2.299725738964898e-06,
      "loss": 0.70542526,
      "num_input_tokens_seen": 167335385,
      "step": 7793,
      "time_per_iteration": 2.596992015838623
    },
    {
      "auxiliary_loss_clip": 0.01105482,
      "auxiliary_loss_mlp": 0.00786867,
      "balance_loss_clip": 1.04161644,
      "balance_loss_mlp": 1.01322246,
      "epoch": 0.468600631294153,
      "flos": 21579799697280.0,
      "grad_norm": 1.599227051108829,
      "language_loss": 0.73983198,
      "learning_rate": 2.2993406698748607e-06,
      "loss": 0.75875551,
      "num_input_tokens_seen": 167353625,
      "step": 7794,
      "time_per_iteration": 2.508617877960205
    },
    {
      "auxiliary_loss_clip": 0.01088755,
      "auxiliary_loss_mlp": 0.01031538,
      "balance_loss_clip": 1.04218686,
      "balance_loss_mlp": 1.01731086,
      "epoch": 0.468660754546821,
      "flos": 25885632597120.0,
      "grad_norm": 1.5223590421742774,
      "language_loss": 0.63093823,
      "learning_rate": 2.2989555894337953e-06,
      "loss": 0.65214121,
      "num_input_tokens_seen": 167374565,
      "step": 7795,
      "time_per_iteration": 2.610881805419922
    },
    {
      "auxiliary_loss_clip": 0.01078247,
      "auxiliary_loss_mlp": 0.01027179,
      "balance_loss_clip": 1.03787279,
      "balance_loss_mlp": 1.01374435,
      "epoch": 0.46872087779948896,
      "flos": 35475001666560.0,
      "grad_norm": 1.7803469037435196,
      "language_loss": 0.68408775,
      "learning_rate": 2.298570497656304e-06,
      "loss": 0.70514202,
      "num_input_tokens_seen": 167395010,
      "step": 7796,
      "time_per_iteration": 2.6751792430877686
    },
    {
      "auxiliary_loss_clip": 0.01119174,
      "auxiliary_loss_mlp": 0.00788559,
      "balance_loss_clip": 1.04212332,
      "balance_loss_mlp": 1.01383114,
      "epoch": 0.4687810010521569,
      "flos": 26396425952640.0,
      "grad_norm": 1.8148000599888063,
      "language_loss": 0.70293629,
      "learning_rate": 2.2981853945569894e-06,
      "loss": 0.72201359,
      "num_input_tokens_seen": 167415285,
      "step": 7797,
      "time_per_iteration": 2.5049824714660645
    },
    {
      "auxiliary_loss_clip": 0.01104133,
      "auxiliary_loss_mlp": 0.01032466,
      "balance_loss_clip": 1.04361916,
      "balance_loss_mlp": 1.01776171,
      "epoch": 0.4688411243048249,
      "flos": 19972761212160.0,
      "grad_norm": 1.9703223898059004,
      "language_loss": 0.6723181,
      "learning_rate": 2.297800280150454e-06,
      "loss": 0.6936841,
      "num_input_tokens_seen": 167432405,
      "step": 7798,
      "time_per_iteration": 2.509965419769287
    },
    {
      "auxiliary_loss_clip": 0.01032918,
      "auxiliary_loss_mlp": 0.01003541,
      "balance_loss_clip": 1.01530409,
      "balance_loss_mlp": 1.00196767,
      "epoch": 0.46890124755749285,
      "flos": 63977015900160.0,
      "grad_norm": 0.9482300437868589,
      "language_loss": 0.64604592,
      "learning_rate": 2.2974151544513033e-06,
      "loss": 0.66641045,
      "num_input_tokens_seen": 167499365,
      "step": 7799,
      "time_per_iteration": 4.639041900634766
    },
    {
      "auxiliary_loss_clip": 0.01087453,
      "auxiliary_loss_mlp": 0.01024358,
      "balance_loss_clip": 1.04077923,
      "balance_loss_mlp": 1.01144791,
      "epoch": 0.4689613708101608,
      "flos": 23768985905280.0,
      "grad_norm": 1.4925565328097643,
      "language_loss": 0.72374094,
      "learning_rate": 2.2970300174741395e-06,
      "loss": 0.7448591,
      "num_input_tokens_seen": 167520390,
      "step": 7800,
      "time_per_iteration": 3.969536542892456
    },
    {
      "auxiliary_loss_clip": 0.01115944,
      "auxiliary_loss_mlp": 0.01030961,
      "balance_loss_clip": 1.0423094,
      "balance_loss_mlp": 1.0189147,
      "epoch": 0.4690214940628288,
      "flos": 24788705109120.0,
      "grad_norm": 2.83580568768042,
      "language_loss": 0.72071826,
      "learning_rate": 2.296644869233568e-06,
      "loss": 0.74218732,
      "num_input_tokens_seen": 167539865,
      "step": 7801,
      "time_per_iteration": 2.489851236343384
    },
    {
      "auxiliary_loss_clip": 0.01089613,
      "auxiliary_loss_mlp": 0.01039249,
      "balance_loss_clip": 1.04181612,
      "balance_loss_mlp": 1.02388334,
      "epoch": 0.46908161731549675,
      "flos": 18077324428800.0,
      "grad_norm": 1.874913106942222,
      "language_loss": 0.62451738,
      "learning_rate": 2.2962597097441936e-06,
      "loss": 0.64580607,
      "num_input_tokens_seen": 167558190,
      "step": 7802,
      "time_per_iteration": 2.5229990482330322
    },
    {
      "auxiliary_loss_clip": 0.01120996,
      "auxiliary_loss_mlp": 0.01036524,
      "balance_loss_clip": 1.04214168,
      "balance_loss_mlp": 1.02333307,
      "epoch": 0.4691417405681647,
      "flos": 25703350053120.0,
      "grad_norm": 1.9867974776887167,
      "language_loss": 0.7339707,
      "learning_rate": 2.2958745390206206e-06,
      "loss": 0.75554591,
      "num_input_tokens_seen": 167577685,
      "step": 7803,
      "time_per_iteration": 3.8851864337921143
    },
    {
      "auxiliary_loss_clip": 0.01094057,
      "auxiliary_loss_mlp": 0.00787738,
      "balance_loss_clip": 1.04124904,
      "balance_loss_mlp": 1.01060212,
      "epoch": 0.46920186382083273,
      "flos": 17457039440640.0,
      "grad_norm": 1.5668638514281588,
      "language_loss": 0.77616018,
      "learning_rate": 2.2954893570774558e-06,
      "loss": 0.79497814,
      "num_input_tokens_seen": 167596390,
      "step": 7804,
      "time_per_iteration": 2.498448133468628
    },
    {
      "auxiliary_loss_clip": 0.01093564,
      "auxiliary_loss_mlp": 0.01027735,
      "balance_loss_clip": 1.04298663,
      "balance_loss_mlp": 1.01500356,
      "epoch": 0.4692619870735007,
      "flos": 20339445202560.0,
      "grad_norm": 1.6935922166858808,
      "language_loss": 0.77065015,
      "learning_rate": 2.295104163929305e-06,
      "loss": 0.7918632,
      "num_input_tokens_seen": 167614980,
      "step": 7805,
      "time_per_iteration": 3.8941218852996826
    },
    {
      "auxiliary_loss_clip": 0.01127043,
      "auxiliary_loss_mlp": 0.01041301,
      "balance_loss_clip": 1.04518533,
      "balance_loss_mlp": 1.02706695,
      "epoch": 0.46932211032616866,
      "flos": 29496558003840.0,
      "grad_norm": 1.6163692188574958,
      "language_loss": 0.82652223,
      "learning_rate": 2.2947189595907742e-06,
      "loss": 0.84820569,
      "num_input_tokens_seen": 167635895,
      "step": 7806,
      "time_per_iteration": 2.5460362434387207
    },
    {
      "auxiliary_loss_clip": 0.01099772,
      "auxiliary_loss_mlp": 0.01037894,
      "balance_loss_clip": 1.04292583,
      "balance_loss_mlp": 1.02376795,
      "epoch": 0.4693822335788366,
      "flos": 36211242735360.0,
      "grad_norm": 1.9314458856024193,
      "language_loss": 0.76935846,
      "learning_rate": 2.294333744076472e-06,
      "loss": 0.79073513,
      "num_input_tokens_seen": 167657440,
      "step": 7807,
      "time_per_iteration": 2.647977352142334
    },
    {
      "auxiliary_loss_clip": 0.01096758,
      "auxiliary_loss_mlp": 0.01033894,
      "balance_loss_clip": 1.04472041,
      "balance_loss_mlp": 1.02008986,
      "epoch": 0.4694423568315046,
      "flos": 20338978325760.0,
      "grad_norm": 2.377820017390032,
      "language_loss": 0.51276386,
      "learning_rate": 2.2939485174010035e-06,
      "loss": 0.53407037,
      "num_input_tokens_seen": 167675025,
      "step": 7808,
      "time_per_iteration": 2.5327816009521484
    },
    {
      "auxiliary_loss_clip": 0.01005202,
      "auxiliary_loss_mlp": 0.01000234,
      "balance_loss_clip": 1.02427244,
      "balance_loss_mlp": 0.99878591,
      "epoch": 0.46950248008417256,
      "flos": 64326353621760.0,
      "grad_norm": 0.7826410186412291,
      "language_loss": 0.5780341,
      "learning_rate": 2.293563279578978e-06,
      "loss": 0.59808844,
      "num_input_tokens_seen": 167729635,
      "step": 7809,
      "time_per_iteration": 3.0488524436950684
    },
    {
      "auxiliary_loss_clip": 0.01085108,
      "auxiliary_loss_mlp": 0.01037032,
      "balance_loss_clip": 1.04521918,
      "balance_loss_mlp": 1.02322721,
      "epoch": 0.4695626033368405,
      "flos": 19200106730880.0,
      "grad_norm": 2.419743901055374,
      "language_loss": 0.71696436,
      "learning_rate": 2.2931780306250045e-06,
      "loss": 0.73818576,
      "num_input_tokens_seen": 167745135,
      "step": 7810,
      "time_per_iteration": 2.594684362411499
    },
    {
      "auxiliary_loss_clip": 0.01112083,
      "auxiliary_loss_mlp": 0.010365,
      "balance_loss_clip": 1.0443666,
      "balance_loss_mlp": 1.02343488,
      "epoch": 0.4696227265895085,
      "flos": 23002436736000.0,
      "grad_norm": 2.830552721719309,
      "language_loss": 0.8116833,
      "learning_rate": 2.29279277055369e-06,
      "loss": 0.8331691,
      "num_input_tokens_seen": 167763875,
      "step": 7811,
      "time_per_iteration": 2.5199778079986572
    },
    {
      "auxiliary_loss_clip": 0.01107054,
      "auxiliary_loss_mlp": 0.01036457,
      "balance_loss_clip": 1.04494643,
      "balance_loss_mlp": 1.02254534,
      "epoch": 0.46968284984217645,
      "flos": 21870855601920.0,
      "grad_norm": 1.6177238215744718,
      "language_loss": 0.8043142,
      "learning_rate": 2.292407499379644e-06,
      "loss": 0.82574934,
      "num_input_tokens_seen": 167784895,
      "step": 7812,
      "time_per_iteration": 2.5386803150177
    },
    {
      "auxiliary_loss_clip": 0.01075692,
      "auxiliary_loss_mlp": 0.01036389,
      "balance_loss_clip": 1.04327106,
      "balance_loss_mlp": 1.02279305,
      "epoch": 0.4697429730948444,
      "flos": 19974987855360.0,
      "grad_norm": 1.9781537206190236,
      "language_loss": 0.74206185,
      "learning_rate": 2.292022217117477e-06,
      "loss": 0.76318264,
      "num_input_tokens_seen": 167803185,
      "step": 7813,
      "time_per_iteration": 2.6149747371673584
    },
    {
      "auxiliary_loss_clip": 0.01095721,
      "auxiliary_loss_mlp": 0.01031782,
      "balance_loss_clip": 1.04365444,
      "balance_loss_mlp": 1.01819801,
      "epoch": 0.4698030963475124,
      "flos": 15156206784000.0,
      "grad_norm": 2.3773481080205885,
      "language_loss": 0.85037363,
      "learning_rate": 2.291636923781798e-06,
      "loss": 0.87164867,
      "num_input_tokens_seen": 167816550,
      "step": 7814,
      "time_per_iteration": 2.4940598011016846
    },
    {
      "auxiliary_loss_clip": 0.01097618,
      "auxiliary_loss_mlp": 0.01036255,
      "balance_loss_clip": 1.04150248,
      "balance_loss_mlp": 1.02376747,
      "epoch": 0.46986321960018035,
      "flos": 15151178880000.0,
      "grad_norm": 1.9373464713664799,
      "language_loss": 0.81707901,
      "learning_rate": 2.291251619387217e-06,
      "loss": 0.83841777,
      "num_input_tokens_seen": 167831845,
      "step": 7815,
      "time_per_iteration": 2.486842632293701
    },
    {
      "auxiliary_loss_clip": 0.01079787,
      "auxiliary_loss_mlp": 0.01037067,
      "balance_loss_clip": 1.04562485,
      "balance_loss_mlp": 1.02241015,
      "epoch": 0.4699233428528483,
      "flos": 23108911626240.0,
      "grad_norm": 2.2724251067961716,
      "language_loss": 0.77227294,
      "learning_rate": 2.2908663039483468e-06,
      "loss": 0.79344147,
      "num_input_tokens_seen": 167850360,
      "step": 7816,
      "time_per_iteration": 2.604668378829956
    },
    {
      "auxiliary_loss_clip": 0.01042301,
      "auxiliary_loss_mlp": 0.01005859,
      "balance_loss_clip": 1.01578307,
      "balance_loss_mlp": 1.00442839,
      "epoch": 0.46998346610551633,
      "flos": 68105558246400.0,
      "grad_norm": 0.8754150850695498,
      "language_loss": 0.59076893,
      "learning_rate": 2.290480977479796e-06,
      "loss": 0.61125052,
      "num_input_tokens_seen": 167908660,
      "step": 7817,
      "time_per_iteration": 3.0702075958251953
    },
    {
      "auxiliary_loss_clip": 0.01098391,
      "auxiliary_loss_mlp": 0.01035148,
      "balance_loss_clip": 1.0446223,
      "balance_loss_mlp": 1.02160013,
      "epoch": 0.4700435893581843,
      "flos": 24129456842880.0,
      "grad_norm": 1.7342451905176604,
      "language_loss": 0.79205859,
      "learning_rate": 2.2900956399961775e-06,
      "loss": 0.81339395,
      "num_input_tokens_seen": 167927905,
      "step": 7818,
      "time_per_iteration": 2.5409815311431885
    },
    {
      "auxiliary_loss_clip": 0.01120619,
      "auxiliary_loss_mlp": 0.01032249,
      "balance_loss_clip": 1.04311085,
      "balance_loss_mlp": 1.01928532,
      "epoch": 0.47010371261085226,
      "flos": 20150518642560.0,
      "grad_norm": 2.3553082068980813,
      "language_loss": 0.83618325,
      "learning_rate": 2.289710291512104e-06,
      "loss": 0.85771197,
      "num_input_tokens_seen": 167945995,
      "step": 7819,
      "time_per_iteration": 2.4858360290527344
    },
    {
      "auxiliary_loss_clip": 0.01093717,
      "auxiliary_loss_mlp": 0.01035902,
      "balance_loss_clip": 1.0424633,
      "balance_loss_mlp": 1.02112031,
      "epoch": 0.47016383586352023,
      "flos": 15122199582720.0,
      "grad_norm": 2.333842257968576,
      "language_loss": 0.76694727,
      "learning_rate": 2.289324932042186e-06,
      "loss": 0.78824353,
      "num_input_tokens_seen": 167963380,
      "step": 7820,
      "time_per_iteration": 2.520315170288086
    },
    {
      "auxiliary_loss_clip": 0.01106047,
      "auxiliary_loss_mlp": 0.01037779,
      "balance_loss_clip": 1.04667997,
      "balance_loss_mlp": 1.02407539,
      "epoch": 0.4702239591161882,
      "flos": 13552975140480.0,
      "grad_norm": 2.0501009044392866,
      "language_loss": 0.74571764,
      "learning_rate": 2.288939561601039e-06,
      "loss": 0.76715589,
      "num_input_tokens_seen": 167981740,
      "step": 7821,
      "time_per_iteration": 2.483879566192627
    },
    {
      "auxiliary_loss_clip": 0.01119841,
      "auxiliary_loss_mlp": 0.01039382,
      "balance_loss_clip": 1.04466391,
      "balance_loss_mlp": 1.0267992,
      "epoch": 0.47028408236885616,
      "flos": 24276511123200.0,
      "grad_norm": 1.8577670756257596,
      "language_loss": 0.8893311,
      "learning_rate": 2.2885541802032746e-06,
      "loss": 0.9109233,
      "num_input_tokens_seen": 167999380,
      "step": 7822,
      "time_per_iteration": 2.497555732727051
    },
    {
      "auxiliary_loss_clip": 0.01104449,
      "auxiliary_loss_mlp": 0.01033828,
      "balance_loss_clip": 1.04547012,
      "balance_loss_mlp": 1.02106667,
      "epoch": 0.4703442056215241,
      "flos": 22856926740480.0,
      "grad_norm": 1.5696363460056841,
      "language_loss": 0.79657233,
      "learning_rate": 2.2881687878635055e-06,
      "loss": 0.81795508,
      "num_input_tokens_seen": 168018395,
      "step": 7823,
      "time_per_iteration": 2.5196025371551514
    },
    {
      "auxiliary_loss_clip": 0.01039711,
      "auxiliary_loss_mlp": 0.01002739,
      "balance_loss_clip": 1.0380075,
      "balance_loss_mlp": 1.00118959,
      "epoch": 0.4704043288741921,
      "flos": 69240227950080.0,
      "grad_norm": 0.688773522865109,
      "language_loss": 0.56644595,
      "learning_rate": 2.2877833845963487e-06,
      "loss": 0.58687049,
      "num_input_tokens_seen": 168084080,
      "step": 7824,
      "time_per_iteration": 3.2227060794830322
    },
    {
      "auxiliary_loss_clip": 0.01097695,
      "auxiliary_loss_mlp": 0.01034318,
      "balance_loss_clip": 1.04258275,
      "balance_loss_mlp": 1.02022123,
      "epoch": 0.47046445212686006,
      "flos": 18041090584320.0,
      "grad_norm": 1.775493884160791,
      "language_loss": 0.81207913,
      "learning_rate": 2.2873979704164157e-06,
      "loss": 0.83339918,
      "num_input_tokens_seen": 168101555,
      "step": 7825,
      "time_per_iteration": 2.541686534881592
    },
    {
      "auxiliary_loss_clip": 0.01102238,
      "auxiliary_loss_mlp": 0.01030419,
      "balance_loss_clip": 1.0446701,
      "balance_loss_mlp": 1.01679945,
      "epoch": 0.470524575379528,
      "flos": 23951448017280.0,
      "grad_norm": 1.5666679305659326,
      "language_loss": 0.66119087,
      "learning_rate": 2.287012545338324e-06,
      "loss": 0.68251741,
      "num_input_tokens_seen": 168121530,
      "step": 7826,
      "time_per_iteration": 2.549367904663086
    },
    {
      "auxiliary_loss_clip": 0.01098056,
      "auxiliary_loss_mlp": 0.01035617,
      "balance_loss_clip": 1.04243529,
      "balance_loss_mlp": 1.02149045,
      "epoch": 0.470584698632196,
      "flos": 18113558273280.0,
      "grad_norm": 1.6692542552206366,
      "language_loss": 0.83901948,
      "learning_rate": 2.2866271093766877e-06,
      "loss": 0.86035627,
      "num_input_tokens_seen": 168140335,
      "step": 7827,
      "time_per_iteration": 2.5409069061279297
    },
    {
      "auxiliary_loss_clip": 0.01024521,
      "auxiliary_loss_mlp": 0.01003383,
      "balance_loss_clip": 1.01735473,
      "balance_loss_mlp": 1.00177407,
      "epoch": 0.47064482188486395,
      "flos": 57251916224640.0,
      "grad_norm": 0.8610311408973074,
      "language_loss": 0.55682701,
      "learning_rate": 2.286241662546122e-06,
      "loss": 0.57710606,
      "num_input_tokens_seen": 168200535,
      "step": 7828,
      "time_per_iteration": 3.111931085586548
    },
    {
      "auxiliary_loss_clip": 0.01118422,
      "auxiliary_loss_mlp": 0.01031249,
      "balance_loss_clip": 1.04340935,
      "balance_loss_mlp": 1.01743269,
      "epoch": 0.4707049451375319,
      "flos": 17895077798400.0,
      "grad_norm": 1.8425627780204024,
      "language_loss": 0.80984652,
      "learning_rate": 2.285856204861245e-06,
      "loss": 0.83134323,
      "num_input_tokens_seen": 168219610,
      "step": 7829,
      "time_per_iteration": 2.471048355102539
    },
    {
      "auxiliary_loss_clip": 0.01120703,
      "auxiliary_loss_mlp": 0.01029969,
      "balance_loss_clip": 1.04541135,
      "balance_loss_mlp": 1.01766062,
      "epoch": 0.47076506839019994,
      "flos": 25232669210880.0,
      "grad_norm": 1.5783653202488326,
      "language_loss": 0.75909483,
      "learning_rate": 2.2854707363366703e-06,
      "loss": 0.78060156,
      "num_input_tokens_seen": 168242505,
      "step": 7830,
      "time_per_iteration": 2.5308735370635986
    },
    {
      "auxiliary_loss_clip": 0.01087555,
      "auxiliary_loss_mlp": 0.01030064,
      "balance_loss_clip": 1.04439831,
      "balance_loss_mlp": 1.01606905,
      "epoch": 0.4708251916428679,
      "flos": 13479681438720.0,
      "grad_norm": 2.0672343326669322,
      "language_loss": 0.78963065,
      "learning_rate": 2.2850852569870177e-06,
      "loss": 0.81080687,
      "num_input_tokens_seen": 168260220,
      "step": 7831,
      "time_per_iteration": 2.5483953952789307
    },
    {
      "auxiliary_loss_clip": 0.01088349,
      "auxiliary_loss_mlp": 0.01042904,
      "balance_loss_clip": 1.04201126,
      "balance_loss_mlp": 1.0249393,
      "epoch": 0.47088531489553587,
      "flos": 30147833450880.0,
      "grad_norm": 1.9436963257224102,
      "language_loss": 0.75794828,
      "learning_rate": 2.2846997668269033e-06,
      "loss": 0.77926075,
      "num_input_tokens_seen": 168277360,
      "step": 7832,
      "time_per_iteration": 2.6287007331848145
    },
    {
      "auxiliary_loss_clip": 0.01095006,
      "auxiliary_loss_mlp": 0.01024493,
      "balance_loss_clip": 1.04665768,
      "balance_loss_mlp": 1.01216054,
      "epoch": 0.47094543814820383,
      "flos": 21798280172160.0,
      "grad_norm": 1.2996970511585297,
      "language_loss": 0.74528241,
      "learning_rate": 2.2843142658709454e-06,
      "loss": 0.76647741,
      "num_input_tokens_seen": 168296605,
      "step": 7833,
      "time_per_iteration": 2.5544190406799316
    },
    {
      "auxiliary_loss_clip": 0.01109942,
      "auxiliary_loss_mlp": 0.01037558,
      "balance_loss_clip": 1.04435849,
      "balance_loss_mlp": 1.0242126,
      "epoch": 0.4710055614008718,
      "flos": 23003011353600.0,
      "grad_norm": 1.7413351217128663,
      "language_loss": 0.7531215,
      "learning_rate": 2.283928754133762e-06,
      "loss": 0.77459651,
      "num_input_tokens_seen": 168316205,
      "step": 7834,
      "time_per_iteration": 2.5212562084198
    },
    {
      "auxiliary_loss_clip": 0.01075123,
      "auxiliary_loss_mlp": 0.0103065,
      "balance_loss_clip": 1.04543972,
      "balance_loss_mlp": 1.01795983,
      "epoch": 0.47106568465353976,
      "flos": 42741346452480.0,
      "grad_norm": 1.4618629374053351,
      "language_loss": 0.66071272,
      "learning_rate": 2.283543231629972e-06,
      "loss": 0.68177044,
      "num_input_tokens_seen": 168338935,
      "step": 7835,
      "time_per_iteration": 2.7880265712738037
    },
    {
      "auxiliary_loss_clip": 0.01034263,
      "auxiliary_loss_mlp": 0.00769413,
      "balance_loss_clip": 1.01788187,
      "balance_loss_mlp": 1.00876665,
      "epoch": 0.4711258079062077,
      "flos": 68554008570240.0,
      "grad_norm": 0.8692285815251268,
      "language_loss": 0.62083316,
      "learning_rate": 2.283157698374194e-06,
      "loss": 0.63887,
      "num_input_tokens_seen": 168392800,
      "step": 7836,
      "time_per_iteration": 3.0945003032684326
    },
    {
      "auxiliary_loss_clip": 0.01087043,
      "auxiliary_loss_mlp": 0.00790056,
      "balance_loss_clip": 1.04354787,
      "balance_loss_mlp": 1.01240993,
      "epoch": 0.4711859311588757,
      "flos": 25446588658560.0,
      "grad_norm": 1.5660742697408774,
      "language_loss": 0.69760656,
      "learning_rate": 2.2827721543810475e-06,
      "loss": 0.71637756,
      "num_input_tokens_seen": 168412940,
      "step": 7837,
      "time_per_iteration": 2.6097195148468018
    },
    {
      "auxiliary_loss_clip": 0.01117851,
      "auxiliary_loss_mlp": 0.01035425,
      "balance_loss_clip": 1.04707813,
      "balance_loss_mlp": 1.02103591,
      "epoch": 0.47124605441154366,
      "flos": 21981891519360.0,
      "grad_norm": 1.710675075871333,
      "language_loss": 0.66048002,
      "learning_rate": 2.282386599665153e-06,
      "loss": 0.6820128,
      "num_input_tokens_seen": 168431995,
      "step": 7838,
      "time_per_iteration": 5.204299211502075
    },
    {
      "auxiliary_loss_clip": 0.01096684,
      "auxiliary_loss_mlp": 0.01033404,
      "balance_loss_clip": 1.04153037,
      "balance_loss_mlp": 1.01870584,
      "epoch": 0.4713061776642116,
      "flos": 25412689198080.0,
      "grad_norm": 1.7531822055245156,
      "language_loss": 0.77314091,
      "learning_rate": 2.2820010342411304e-06,
      "loss": 0.79444182,
      "num_input_tokens_seen": 168454585,
      "step": 7839,
      "time_per_iteration": 2.59617018699646
    },
    {
      "auxiliary_loss_clip": 0.01082936,
      "auxiliary_loss_mlp": 0.01030686,
      "balance_loss_clip": 1.04307461,
      "balance_loss_mlp": 1.01790667,
      "epoch": 0.4713663009168796,
      "flos": 26542259170560.0,
      "grad_norm": 2.5594549056912546,
      "language_loss": 0.72397053,
      "learning_rate": 2.2816154581235993e-06,
      "loss": 0.74510676,
      "num_input_tokens_seen": 168471265,
      "step": 7840,
      "time_per_iteration": 2.580916404724121
    },
    {
      "auxiliary_loss_clip": 0.01094262,
      "auxiliary_loss_mlp": 0.01028446,
      "balance_loss_clip": 1.04370677,
      "balance_loss_mlp": 1.0151186,
      "epoch": 0.47142642416954755,
      "flos": 23623583650560.0,
      "grad_norm": 1.6452167510457716,
      "language_loss": 0.74973953,
      "learning_rate": 2.2812298713271833e-06,
      "loss": 0.77096659,
      "num_input_tokens_seen": 168491360,
      "step": 7841,
      "time_per_iteration": 3.928304672241211
    },
    {
      "auxiliary_loss_clip": 0.01100921,
      "auxiliary_loss_mlp": 0.01029913,
      "balance_loss_clip": 1.04623127,
      "balance_loss_mlp": 1.01679397,
      "epoch": 0.4714865474222155,
      "flos": 22310150935680.0,
      "grad_norm": 1.7741685167163892,
      "language_loss": 0.70594919,
      "learning_rate": 2.280844273866501e-06,
      "loss": 0.72725749,
      "num_input_tokens_seen": 168511335,
      "step": 7842,
      "time_per_iteration": 2.5535922050476074
    },
    {
      "auxiliary_loss_clip": 0.01113939,
      "auxiliary_loss_mlp": 0.0103105,
      "balance_loss_clip": 1.04729843,
      "balance_loss_mlp": 1.01789498,
      "epoch": 0.4715466706748835,
      "flos": 17822430541440.0,
      "grad_norm": 3.5798273298963963,
      "language_loss": 0.78538764,
      "learning_rate": 2.280458665756177e-06,
      "loss": 0.80683756,
      "num_input_tokens_seen": 168529920,
      "step": 7843,
      "time_per_iteration": 3.889812707901001
    },
    {
      "auxiliary_loss_clip": 0.01113656,
      "auxiliary_loss_mlp": 0.0103075,
      "balance_loss_clip": 1.04398429,
      "balance_loss_mlp": 1.01795292,
      "epoch": 0.4716067939275515,
      "flos": 23659530186240.0,
      "grad_norm": 1.6752335315284426,
      "language_loss": 0.74167001,
      "learning_rate": 2.280073047010832e-06,
      "loss": 0.76311409,
      "num_input_tokens_seen": 168550595,
      "step": 7844,
      "time_per_iteration": 2.5311155319213867
    },
    {
      "auxiliary_loss_clip": 0.01096476,
      "auxiliary_loss_mlp": 0.01037635,
      "balance_loss_clip": 1.04676473,
      "balance_loss_mlp": 1.02396202,
      "epoch": 0.47166691718021947,
      "flos": 17930162407680.0,
      "grad_norm": 1.4484723075210906,
      "language_loss": 0.78517878,
      "learning_rate": 2.279687417645088e-06,
      "loss": 0.80651987,
      "num_input_tokens_seen": 168569765,
      "step": 7845,
      "time_per_iteration": 2.5316691398620605
    },
    {
      "auxiliary_loss_clip": 0.01108813,
      "auxiliary_loss_mlp": 0.01035473,
      "balance_loss_clip": 1.04236627,
      "balance_loss_mlp": 1.02271175,
      "epoch": 0.47172704043288743,
      "flos": 26614583205120.0,
      "grad_norm": 1.529334355987423,
      "language_loss": 0.73058677,
      "learning_rate": 2.2793017776735703e-06,
      "loss": 0.7520296,
      "num_input_tokens_seen": 168591525,
      "step": 7846,
      "time_per_iteration": 2.539745807647705
    },
    {
      "auxiliary_loss_clip": 0.01106496,
      "auxiliary_loss_mlp": 0.0103382,
      "balance_loss_clip": 1.04348421,
      "balance_loss_mlp": 1.02039719,
      "epoch": 0.4717871636855554,
      "flos": 27922700707200.0,
      "grad_norm": 1.3156578756186819,
      "language_loss": 0.74193841,
      "learning_rate": 2.2789161271109e-06,
      "loss": 0.76334155,
      "num_input_tokens_seen": 168611235,
      "step": 7847,
      "time_per_iteration": 2.532700300216675
    },
    {
      "auxiliary_loss_clip": 0.01072013,
      "auxiliary_loss_mlp": 0.01034028,
      "balance_loss_clip": 1.04354692,
      "balance_loss_mlp": 1.02073038,
      "epoch": 0.47184728693822336,
      "flos": 14502237816960.0,
      "grad_norm": 1.6682808156659112,
      "language_loss": 0.80529869,
      "learning_rate": 2.278530465971703e-06,
      "loss": 0.82635915,
      "num_input_tokens_seen": 168628710,
      "step": 7848,
      "time_per_iteration": 2.5631067752838135
    },
    {
      "auxiliary_loss_clip": 0.01112697,
      "auxiliary_loss_mlp": 0.01035283,
      "balance_loss_clip": 1.04700828,
      "balance_loss_mlp": 1.02185404,
      "epoch": 0.47190741019089133,
      "flos": 17856545483520.0,
      "grad_norm": 1.9010332329509758,
      "language_loss": 0.70322335,
      "learning_rate": 2.2781447942706032e-06,
      "loss": 0.72470313,
      "num_input_tokens_seen": 168645645,
      "step": 7849,
      "time_per_iteration": 2.461242914199829
    },
    {
      "auxiliary_loss_clip": 0.01094413,
      "auxiliary_loss_mlp": 0.0103482,
      "balance_loss_clip": 1.04442549,
      "balance_loss_mlp": 1.01957917,
      "epoch": 0.4719675334435593,
      "flos": 17895472848000.0,
      "grad_norm": 2.0793600970583075,
      "language_loss": 0.69129956,
      "learning_rate": 2.277759112022224e-06,
      "loss": 0.71259189,
      "num_input_tokens_seen": 168664165,
      "step": 7850,
      "time_per_iteration": 2.5414345264434814
    },
    {
      "auxiliary_loss_clip": 0.01066811,
      "auxiliary_loss_mlp": 0.01029284,
      "balance_loss_clip": 1.04422486,
      "balance_loss_mlp": 1.01527131,
      "epoch": 0.47202765669622726,
      "flos": 20704369426560.0,
      "grad_norm": 1.8475353107094923,
      "language_loss": 0.74707144,
      "learning_rate": 2.2773734192411916e-06,
      "loss": 0.76803237,
      "num_input_tokens_seen": 168681940,
      "step": 7851,
      "time_per_iteration": 2.6205170154571533
    },
    {
      "auxiliary_loss_clip": 0.01059648,
      "auxiliary_loss_mlp": 0.01050332,
      "balance_loss_clip": 1.04149461,
      "balance_loss_mlp": 1.03284407,
      "epoch": 0.4720877799488952,
      "flos": 16360255607040.0,
      "grad_norm": 2.2788699304554303,
      "language_loss": 0.76157784,
      "learning_rate": 2.276987715942132e-06,
      "loss": 0.78267765,
      "num_input_tokens_seen": 168698830,
      "step": 7852,
      "time_per_iteration": 2.5995140075683594
    },
    {
      "auxiliary_loss_clip": 0.01086114,
      "auxiliary_loss_mlp": 0.01030326,
      "balance_loss_clip": 1.04507947,
      "balance_loss_mlp": 1.0163132,
      "epoch": 0.4721479032015632,
      "flos": 20668171495680.0,
      "grad_norm": 1.570091584615071,
      "language_loss": 0.68924516,
      "learning_rate": 2.2766020021396696e-06,
      "loss": 0.71040952,
      "num_input_tokens_seen": 168718305,
      "step": 7853,
      "time_per_iteration": 2.5585246086120605
    },
    {
      "auxiliary_loss_clip": 0.01013121,
      "auxiliary_loss_mlp": 0.01003499,
      "balance_loss_clip": 1.02467847,
      "balance_loss_mlp": 1.00190151,
      "epoch": 0.47220802645423116,
      "flos": 67750438435200.0,
      "grad_norm": 0.7088281950419517,
      "language_loss": 0.50131404,
      "learning_rate": 2.276216277848432e-06,
      "loss": 0.52148032,
      "num_input_tokens_seen": 168782365,
      "step": 7854,
      "time_per_iteration": 3.3216397762298584
    },
    {
      "auxiliary_loss_clip": 0.01114526,
      "auxiliary_loss_mlp": 0.01032482,
      "balance_loss_clip": 1.0461874,
      "balance_loss_mlp": 1.0180217,
      "epoch": 0.4722681497068991,
      "flos": 20921449271040.0,
      "grad_norm": 1.8691712893956496,
      "language_loss": 0.64020348,
      "learning_rate": 2.2758305430830455e-06,
      "loss": 0.66167361,
      "num_input_tokens_seen": 168800485,
      "step": 7855,
      "time_per_iteration": 2.5090510845184326
    },
    {
      "auxiliary_loss_clip": 0.01110962,
      "auxiliary_loss_mlp": 0.01035165,
      "balance_loss_clip": 1.04548383,
      "balance_loss_mlp": 1.02109861,
      "epoch": 0.4723282729595671,
      "flos": 28293083798400.0,
      "grad_norm": 2.198089473204247,
      "language_loss": 0.75841093,
      "learning_rate": 2.2754447978581376e-06,
      "loss": 0.77987218,
      "num_input_tokens_seen": 168818965,
      "step": 7856,
      "time_per_iteration": 2.553051233291626
    },
    {
      "auxiliary_loss_clip": 0.01099619,
      "auxiliary_loss_mlp": 0.01030644,
      "balance_loss_clip": 1.04647315,
      "balance_loss_mlp": 1.0178411,
      "epoch": 0.4723883962122351,
      "flos": 27125053338240.0,
      "grad_norm": 1.8782847456208085,
      "language_loss": 0.74852812,
      "learning_rate": 2.2750590421883347e-06,
      "loss": 0.7698307,
      "num_input_tokens_seen": 168840355,
      "step": 7857,
      "time_per_iteration": 2.5784435272216797
    },
    {
      "auxiliary_loss_clip": 0.0110286,
      "auxiliary_loss_mlp": 0.0104119,
      "balance_loss_clip": 1.04674232,
      "balance_loss_mlp": 1.02820241,
      "epoch": 0.47244851946490307,
      "flos": 31537253387520.0,
      "grad_norm": 1.4448572417124614,
      "language_loss": 0.64798969,
      "learning_rate": 2.2746732760882655e-06,
      "loss": 0.66943026,
      "num_input_tokens_seen": 168861765,
      "step": 7858,
      "time_per_iteration": 2.5971856117248535
    },
    {
      "auxiliary_loss_clip": 0.01109065,
      "auxiliary_loss_mlp": 0.00804305,
      "balance_loss_clip": 1.04224455,
      "balance_loss_mlp": 1.04014659,
      "epoch": 0.47250864271757104,
      "flos": 20886544229760.0,
      "grad_norm": 1.6956906450657268,
      "language_loss": 0.70479804,
      "learning_rate": 2.2742874995725575e-06,
      "loss": 0.72393173,
      "num_input_tokens_seen": 168881310,
      "step": 7859,
      "time_per_iteration": 2.496814727783203
    },
    {
      "auxiliary_loss_clip": 0.01128545,
      "auxiliary_loss_mlp": 0.01037614,
      "balance_loss_clip": 1.04651523,
      "balance_loss_mlp": 1.02404809,
      "epoch": 0.472568765970239,
      "flos": 20522086882560.0,
      "grad_norm": 1.574895694484457,
      "language_loss": 0.61589599,
      "learning_rate": 2.2739017126558413e-06,
      "loss": 0.63755763,
      "num_input_tokens_seen": 168899470,
      "step": 7860,
      "time_per_iteration": 2.445223331451416
    },
    {
      "auxiliary_loss_clip": 0.011032,
      "auxiliary_loss_mlp": 0.01039639,
      "balance_loss_clip": 1.04640949,
      "balance_loss_mlp": 1.0256083,
      "epoch": 0.47262888922290697,
      "flos": 35805200417280.0,
      "grad_norm": 2.013879950426509,
      "language_loss": 0.71912366,
      "learning_rate": 2.2735159153527445e-06,
      "loss": 0.74055201,
      "num_input_tokens_seen": 168921495,
      "step": 7861,
      "time_per_iteration": 2.637028932571411
    },
    {
      "auxiliary_loss_clip": 0.01100782,
      "auxiliary_loss_mlp": 0.01037113,
      "balance_loss_clip": 1.04730034,
      "balance_loss_mlp": 1.02301073,
      "epoch": 0.47268901247557493,
      "flos": 20667740532480.0,
      "grad_norm": 1.8787148958137825,
      "language_loss": 0.85305953,
      "learning_rate": 2.273130107677896e-06,
      "loss": 0.87443846,
      "num_input_tokens_seen": 168940515,
      "step": 7862,
      "time_per_iteration": 2.512751579284668
    },
    {
      "auxiliary_loss_clip": 0.01124052,
      "auxiliary_loss_mlp": 0.01034759,
      "balance_loss_clip": 1.04486907,
      "balance_loss_mlp": 1.02168798,
      "epoch": 0.4727491357282429,
      "flos": 19573291082880.0,
      "grad_norm": 1.985312065509785,
      "language_loss": 0.84358662,
      "learning_rate": 2.272744289645927e-06,
      "loss": 0.86517477,
      "num_input_tokens_seen": 168958340,
      "step": 7863,
      "time_per_iteration": 2.456998586654663
    },
    {
      "auxiliary_loss_clip": 0.01098597,
      "auxiliary_loss_mlp": 0.01036043,
      "balance_loss_clip": 1.04470229,
      "balance_loss_mlp": 1.02313864,
      "epoch": 0.47280925898091086,
      "flos": 18217231902720.0,
      "grad_norm": 2.0777094078912635,
      "language_loss": 0.66088885,
      "learning_rate": 2.272358461271467e-06,
      "loss": 0.68223524,
      "num_input_tokens_seen": 168974850,
      "step": 7864,
      "time_per_iteration": 2.5085079669952393
    },
    {
      "auxiliary_loss_clip": 0.01122901,
      "auxiliary_loss_mlp": 0.01032621,
      "balance_loss_clip": 1.04537392,
      "balance_loss_mlp": 1.01841736,
      "epoch": 0.4728693822335788,
      "flos": 17821820010240.0,
      "grad_norm": 2.1358083086522384,
      "language_loss": 0.65667403,
      "learning_rate": 2.271972622569147e-06,
      "loss": 0.67822921,
      "num_input_tokens_seen": 168992860,
      "step": 7865,
      "time_per_iteration": 2.454620122909546
    },
    {
      "auxiliary_loss_clip": 0.0109619,
      "auxiliary_loss_mlp": 0.00796908,
      "balance_loss_clip": 1.0468657,
      "balance_loss_mlp": 1.02657783,
      "epoch": 0.4729295054862468,
      "flos": 20595057361920.0,
      "grad_norm": 1.677259134499498,
      "language_loss": 0.74465954,
      "learning_rate": 2.2715867735535976e-06,
      "loss": 0.76359046,
      "num_input_tokens_seen": 169010325,
      "step": 7866,
      "time_per_iteration": 2.5188405513763428
    },
    {
      "auxiliary_loss_clip": 0.01124373,
      "auxiliary_loss_mlp": 0.01032116,
      "balance_loss_clip": 1.0450002,
      "balance_loss_mlp": 1.01896739,
      "epoch": 0.47298962873891476,
      "flos": 23368079232000.0,
      "grad_norm": 1.8417876940695144,
      "language_loss": 0.82646149,
      "learning_rate": 2.271200914239451e-06,
      "loss": 0.84802634,
      "num_input_tokens_seen": 169029840,
      "step": 7867,
      "time_per_iteration": 2.4861769676208496
    },
    {
      "auxiliary_loss_clip": 0.01108496,
      "auxiliary_loss_mlp": 0.01030247,
      "balance_loss_clip": 1.04338634,
      "balance_loss_mlp": 1.01710999,
      "epoch": 0.4730497519915827,
      "flos": 22052240305920.0,
      "grad_norm": 1.6854733221413818,
      "language_loss": 0.79784518,
      "learning_rate": 2.2708150446413385e-06,
      "loss": 0.81923264,
      "num_input_tokens_seen": 169049975,
      "step": 7868,
      "time_per_iteration": 2.4869577884674072
    },
    {
      "auxiliary_loss_clip": 0.01047843,
      "auxiliary_loss_mlp": 0.01033451,
      "balance_loss_clip": 1.04658508,
      "balance_loss_mlp": 1.0187881,
      "epoch": 0.4731098752442507,
      "flos": 21069724613760.0,
      "grad_norm": 1.7643923253600329,
      "language_loss": 0.74591804,
      "learning_rate": 2.2704291647738915e-06,
      "loss": 0.76673096,
      "num_input_tokens_seen": 169069540,
      "step": 7869,
      "time_per_iteration": 2.654275894165039
    },
    {
      "auxiliary_loss_clip": 0.0110309,
      "auxiliary_loss_mlp": 0.01042201,
      "balance_loss_clip": 1.04459739,
      "balance_loss_mlp": 1.0270133,
      "epoch": 0.4731699984969187,
      "flos": 22528775064960.0,
      "grad_norm": 1.7137113779126913,
      "language_loss": 0.73861885,
      "learning_rate": 2.2700432746517443e-06,
      "loss": 0.76007175,
      "num_input_tokens_seen": 169089940,
      "step": 7870,
      "time_per_iteration": 2.525596857070923
    },
    {
      "auxiliary_loss_clip": 0.01132272,
      "auxiliary_loss_mlp": 0.01033345,
      "balance_loss_clip": 1.04959464,
      "balance_loss_mlp": 1.01834869,
      "epoch": 0.4732301217495867,
      "flos": 24898124914560.0,
      "grad_norm": 1.9799381224910668,
      "language_loss": 0.81075835,
      "learning_rate": 2.2696573742895292e-06,
      "loss": 0.83241451,
      "num_input_tokens_seen": 169109650,
      "step": 7871,
      "time_per_iteration": 2.5570945739746094
    },
    {
      "auxiliary_loss_clip": 0.01110673,
      "auxiliary_loss_mlp": 0.01035853,
      "balance_loss_clip": 1.0450722,
      "balance_loss_mlp": 1.02224541,
      "epoch": 0.47329024500225464,
      "flos": 22784423137920.0,
      "grad_norm": 1.6501589608011695,
      "language_loss": 0.75898409,
      "learning_rate": 2.269271463701879e-06,
      "loss": 0.78044939,
      "num_input_tokens_seen": 169128990,
      "step": 7872,
      "time_per_iteration": 2.508674383163452
    },
    {
      "auxiliary_loss_clip": 0.010883,
      "auxiliary_loss_mlp": 0.01033499,
      "balance_loss_clip": 1.04230344,
      "balance_loss_mlp": 1.01977229,
      "epoch": 0.4733503682549226,
      "flos": 38695902220800.0,
      "grad_norm": 2.163357922429626,
      "language_loss": 0.6794734,
      "learning_rate": 2.268885542903428e-06,
      "loss": 0.70069134,
      "num_input_tokens_seen": 169154645,
      "step": 7873,
      "time_per_iteration": 2.745468854904175
    },
    {
      "auxiliary_loss_clip": 0.01112973,
      "auxiliary_loss_mlp": 0.01030942,
      "balance_loss_clip": 1.04639578,
      "balance_loss_mlp": 1.01761472,
      "epoch": 0.47341049150759057,
      "flos": 22966849336320.0,
      "grad_norm": 1.7948968175629223,
      "language_loss": 0.7243942,
      "learning_rate": 2.26849961190881e-06,
      "loss": 0.7458334,
      "num_input_tokens_seen": 169174995,
      "step": 7874,
      "time_per_iteration": 2.5028791427612305
    },
    {
      "auxiliary_loss_clip": 0.01104997,
      "auxiliary_loss_mlp": 0.01033651,
      "balance_loss_clip": 1.04604602,
      "balance_loss_mlp": 1.02040732,
      "epoch": 0.47347061476025853,
      "flos": 14538471661440.0,
      "grad_norm": 2.3135753533911116,
      "language_loss": 0.6502344,
      "learning_rate": 2.26811367073266e-06,
      "loss": 0.67162085,
      "num_input_tokens_seen": 169191815,
      "step": 7875,
      "time_per_iteration": 2.491527557373047
    },
    {
      "auxiliary_loss_clip": 0.01077567,
      "auxiliary_loss_mlp": 0.01033667,
      "balance_loss_clip": 1.04841447,
      "balance_loss_mlp": 1.01927257,
      "epoch": 0.4735307380129265,
      "flos": 30263250827520.0,
      "grad_norm": 2.6277602920891683,
      "language_loss": 0.81160057,
      "learning_rate": 2.2677277193896125e-06,
      "loss": 0.83271289,
      "num_input_tokens_seen": 169210430,
      "step": 7876,
      "time_per_iteration": 3.9887137413024902
    },
    {
      "auxiliary_loss_clip": 0.01099623,
      "auxiliary_loss_mlp": 0.01037115,
      "balance_loss_clip": 1.04241133,
      "balance_loss_mlp": 1.02283335,
      "epoch": 0.47359086126559446,
      "flos": 19391044452480.0,
      "grad_norm": 1.6772959476146978,
      "language_loss": 0.79430848,
      "learning_rate": 2.267341757894304e-06,
      "loss": 0.81567585,
      "num_input_tokens_seen": 169229295,
      "step": 7877,
      "time_per_iteration": 3.9000935554504395
    },
    {
      "auxiliary_loss_clip": 0.01110987,
      "auxiliary_loss_mlp": 0.0078808,
      "balance_loss_clip": 1.04309833,
      "balance_loss_mlp": 1.01328754,
      "epoch": 0.47365098451826243,
      "flos": 21939408708480.0,
      "grad_norm": 1.9019161483187086,
      "language_loss": 0.71163207,
      "learning_rate": 2.2669557862613685e-06,
      "loss": 0.73062277,
      "num_input_tokens_seen": 169247855,
      "step": 7878,
      "time_per_iteration": 2.5203518867492676
    },
    {
      "auxiliary_loss_clip": 0.01082708,
      "auxiliary_loss_mlp": 0.01030845,
      "balance_loss_clip": 1.04672337,
      "balance_loss_mlp": 1.01808989,
      "epoch": 0.4737111077709304,
      "flos": 25845053207040.0,
      "grad_norm": 2.1897986426624754,
      "language_loss": 0.75165552,
      "learning_rate": 2.2665698045054425e-06,
      "loss": 0.77279103,
      "num_input_tokens_seen": 169268860,
      "step": 7879,
      "time_per_iteration": 2.6223394870758057
    },
    {
      "auxiliary_loss_clip": 0.01028255,
      "auxiliary_loss_mlp": 0.0100513,
      "balance_loss_clip": 1.02314484,
      "balance_loss_mlp": 1.00363433,
      "epoch": 0.47377123102359836,
      "flos": 67760886314880.0,
      "grad_norm": 0.7380541007532991,
      "language_loss": 0.61276722,
      "learning_rate": 2.266183812641164e-06,
      "loss": 0.63310111,
      "num_input_tokens_seen": 169331855,
      "step": 7880,
      "time_per_iteration": 3.12857985496521
    },
    {
      "auxiliary_loss_clip": 0.01098855,
      "auxiliary_loss_mlp": 0.01034712,
      "balance_loss_clip": 1.04316902,
      "balance_loss_mlp": 1.02041316,
      "epoch": 0.4738313542762663,
      "flos": 24315977191680.0,
      "grad_norm": 3.7712616456423977,
      "language_loss": 0.67881912,
      "learning_rate": 2.2657978106831675e-06,
      "loss": 0.70015478,
      "num_input_tokens_seen": 169352175,
      "step": 7881,
      "time_per_iteration": 3.9322140216827393
    },
    {
      "auxiliary_loss_clip": 0.01055675,
      "auxiliary_loss_mlp": 0.01026281,
      "balance_loss_clip": 1.0423038,
      "balance_loss_mlp": 1.01350725,
      "epoch": 0.4738914775289343,
      "flos": 20705339093760.0,
      "grad_norm": 1.6720640069037578,
      "language_loss": 0.77328652,
      "learning_rate": 2.265411798646092e-06,
      "loss": 0.79410607,
      "num_input_tokens_seen": 169371215,
      "step": 7882,
      "time_per_iteration": 4.012754678726196
    },
    {
      "auxiliary_loss_clip": 0.01108917,
      "auxiliary_loss_mlp": 0.01031878,
      "balance_loss_clip": 1.04257596,
      "balance_loss_mlp": 1.01815093,
      "epoch": 0.4739516007816023,
      "flos": 25446337263360.0,
      "grad_norm": 1.5283700066439676,
      "language_loss": 0.76204574,
      "learning_rate": 2.2650257765445747e-06,
      "loss": 0.7834537,
      "num_input_tokens_seen": 169391745,
      "step": 7883,
      "time_per_iteration": 2.539094924926758
    },
    {
      "auxiliary_loss_clip": 0.01101106,
      "auxiliary_loss_mlp": 0.01031999,
      "balance_loss_clip": 1.04510975,
      "balance_loss_mlp": 1.01939297,
      "epoch": 0.4740117240342703,
      "flos": 19974341410560.0,
      "grad_norm": 1.7409257644805376,
      "language_loss": 0.72069812,
      "learning_rate": 2.2646397443932525e-06,
      "loss": 0.74202919,
      "num_input_tokens_seen": 169409845,
      "step": 7884,
      "time_per_iteration": 2.535902976989746
    },
    {
      "auxiliary_loss_clip": 0.01115072,
      "auxiliary_loss_mlp": 0.01032762,
      "balance_loss_clip": 1.04584384,
      "balance_loss_mlp": 1.01851058,
      "epoch": 0.47407184728693824,
      "flos": 15661146222720.0,
      "grad_norm": 1.9230322521204166,
      "language_loss": 0.81940675,
      "learning_rate": 2.2642537022067655e-06,
      "loss": 0.8408851,
      "num_input_tokens_seen": 169426085,
      "step": 7885,
      "time_per_iteration": 2.466367721557617
    },
    {
      "auxiliary_loss_clip": 0.01092952,
      "auxiliary_loss_mlp": 0.01051328,
      "balance_loss_clip": 1.04568911,
      "balance_loss_mlp": 1.03528845,
      "epoch": 0.4741319705396062,
      "flos": 18588800142720.0,
      "grad_norm": 2.019541119891715,
      "language_loss": 0.7347672,
      "learning_rate": 2.263867649999751e-06,
      "loss": 0.75620997,
      "num_input_tokens_seen": 169444705,
      "step": 7886,
      "time_per_iteration": 2.513636827468872
    },
    {
      "auxiliary_loss_clip": 0.01104759,
      "auxiliary_loss_mlp": 0.01031339,
      "balance_loss_clip": 1.04805231,
      "balance_loss_mlp": 1.01724851,
      "epoch": 0.47419209379227417,
      "flos": 13261093223040.0,
      "grad_norm": 2.011873528410458,
      "language_loss": 0.74302578,
      "learning_rate": 2.263481587786849e-06,
      "loss": 0.76438677,
      "num_input_tokens_seen": 169460850,
      "step": 7887,
      "time_per_iteration": 2.5251471996307373
    },
    {
      "auxiliary_loss_clip": 0.01108833,
      "auxiliary_loss_mlp": 0.01026238,
      "balance_loss_clip": 1.04304671,
      "balance_loss_mlp": 1.01413238,
      "epoch": 0.47425221704494214,
      "flos": 20044043752320.0,
      "grad_norm": 1.8357431524510155,
      "language_loss": 0.7745316,
      "learning_rate": 2.2630955155826993e-06,
      "loss": 0.79588228,
      "num_input_tokens_seen": 169478890,
      "step": 7888,
      "time_per_iteration": 2.528989315032959
    },
    {
      "auxiliary_loss_clip": 0.01112893,
      "auxiliary_loss_mlp": 0.01031638,
      "balance_loss_clip": 1.04553425,
      "balance_loss_mlp": 1.01834595,
      "epoch": 0.4743123402976101,
      "flos": 27271892136960.0,
      "grad_norm": 1.8562524681909491,
      "language_loss": 0.72367698,
      "learning_rate": 2.2627094334019406e-06,
      "loss": 0.74512225,
      "num_input_tokens_seen": 169499690,
      "step": 7889,
      "time_per_iteration": 2.639451503753662
    },
    {
      "auxiliary_loss_clip": 0.01045352,
      "auxiliary_loss_mlp": 0.01002741,
      "balance_loss_clip": 1.01943314,
      "balance_loss_mlp": 1.00135791,
      "epoch": 0.47437246355027807,
      "flos": 55393970261760.0,
      "grad_norm": 0.7114644091171369,
      "language_loss": 0.56075668,
      "learning_rate": 2.262323341259214e-06,
      "loss": 0.58123755,
      "num_input_tokens_seen": 169560475,
      "step": 7890,
      "time_per_iteration": 3.1443679332733154
    },
    {
      "auxiliary_loss_clip": 0.01114261,
      "auxiliary_loss_mlp": 0.01035119,
      "balance_loss_clip": 1.04875112,
      "balance_loss_mlp": 1.02004457,
      "epoch": 0.47443258680294603,
      "flos": 23878477537920.0,
      "grad_norm": 1.8856654805758213,
      "language_loss": 0.6575532,
      "learning_rate": 2.2619372391691605e-06,
      "loss": 0.67904699,
      "num_input_tokens_seen": 169580110,
      "step": 7891,
      "time_per_iteration": 2.516343832015991
    },
    {
      "auxiliary_loss_clip": 0.01127774,
      "auxiliary_loss_mlp": 0.01033178,
      "balance_loss_clip": 1.04667401,
      "balance_loss_mlp": 1.01815152,
      "epoch": 0.474492710055614,
      "flos": 21977761455360.0,
      "grad_norm": 2.1510240608337297,
      "language_loss": 0.69735885,
      "learning_rate": 2.26155112714642e-06,
      "loss": 0.71896839,
      "num_input_tokens_seen": 169597510,
      "step": 7892,
      "time_per_iteration": 2.477475881576538
    },
    {
      "auxiliary_loss_clip": 0.01045705,
      "auxiliary_loss_mlp": 0.01002059,
      "balance_loss_clip": 1.04469252,
      "balance_loss_mlp": 1.00016403,
      "epoch": 0.47455283330828196,
      "flos": 62557180122240.0,
      "grad_norm": 1.5400359516014954,
      "language_loss": 0.58623183,
      "learning_rate": 2.2611650052056355e-06,
      "loss": 0.60670942,
      "num_input_tokens_seen": 169660010,
      "step": 7893,
      "time_per_iteration": 3.2231192588806152
    },
    {
      "auxiliary_loss_clip": 0.0111291,
      "auxiliary_loss_mlp": 0.01032103,
      "balance_loss_clip": 1.04746151,
      "balance_loss_mlp": 1.01941884,
      "epoch": 0.47461295656094993,
      "flos": 12093637380480.0,
      "grad_norm": 2.2568191013430074,
      "language_loss": 0.77532029,
      "learning_rate": 2.2607788733614463e-06,
      "loss": 0.79677033,
      "num_input_tokens_seen": 169678485,
      "step": 7894,
      "time_per_iteration": 2.471271514892578
    },
    {
      "auxiliary_loss_clip": 0.01112303,
      "auxiliary_loss_mlp": 0.01032609,
      "balance_loss_clip": 1.04475832,
      "balance_loss_mlp": 1.01940036,
      "epoch": 0.4746730798136179,
      "flos": 20884568981760.0,
      "grad_norm": 1.74479702897336,
      "language_loss": 0.74589288,
      "learning_rate": 2.260392731628497e-06,
      "loss": 0.76734203,
      "num_input_tokens_seen": 169697335,
      "step": 7895,
      "time_per_iteration": 2.5045104026794434
    },
    {
      "auxiliary_loss_clip": 0.01110755,
      "auxiliary_loss_mlp": 0.01031148,
      "balance_loss_clip": 1.0445559,
      "balance_loss_mlp": 1.01684248,
      "epoch": 0.4747332030662859,
      "flos": 19974808287360.0,
      "grad_norm": 1.8843327432471817,
      "language_loss": 0.82327378,
      "learning_rate": 2.260006580021429e-06,
      "loss": 0.84469277,
      "num_input_tokens_seen": 169715395,
      "step": 7896,
      "time_per_iteration": 2.4799885749816895
    },
    {
      "auxiliary_loss_clip": 0.01111688,
      "auxiliary_loss_mlp": 0.01031695,
      "balance_loss_clip": 1.04618239,
      "balance_loss_mlp": 1.01759887,
      "epoch": 0.4747933263189539,
      "flos": 16034186920320.0,
      "grad_norm": 1.7529518095034993,
      "language_loss": 0.75357842,
      "learning_rate": 2.259620418554886e-06,
      "loss": 0.77501225,
      "num_input_tokens_seen": 169733755,
      "step": 7897,
      "time_per_iteration": 2.4667129516601562
    },
    {
      "auxiliary_loss_clip": 0.01100867,
      "auxiliary_loss_mlp": 0.01033804,
      "balance_loss_clip": 1.04713607,
      "balance_loss_mlp": 1.02019596,
      "epoch": 0.47485344957162184,
      "flos": 13955102876160.0,
      "grad_norm": 2.4089453231735503,
      "language_loss": 0.63765466,
      "learning_rate": 2.25923424724351e-06,
      "loss": 0.65900135,
      "num_input_tokens_seen": 169751390,
      "step": 7898,
      "time_per_iteration": 2.5213942527770996
    },
    {
      "auxiliary_loss_clip": 0.01088989,
      "auxiliary_loss_mlp": 0.01045183,
      "balance_loss_clip": 1.04433858,
      "balance_loss_mlp": 1.02900052,
      "epoch": 0.4749135728242898,
      "flos": 20449080489600.0,
      "grad_norm": 2.086891726909934,
      "language_loss": 0.70067298,
      "learning_rate": 2.258848066101946e-06,
      "loss": 0.72201478,
      "num_input_tokens_seen": 169769500,
      "step": 7899,
      "time_per_iteration": 2.5454118251800537
    },
    {
      "auxiliary_loss_clip": 0.01114291,
      "auxiliary_loss_mlp": 0.01037254,
      "balance_loss_clip": 1.04686189,
      "balance_loss_mlp": 1.023175,
      "epoch": 0.4749736960769578,
      "flos": 28949961767040.0,
      "grad_norm": 2.3551802891398976,
      "language_loss": 0.68554485,
      "learning_rate": 2.258461875144837e-06,
      "loss": 0.70706022,
      "num_input_tokens_seen": 169789215,
      "step": 7900,
      "time_per_iteration": 2.554299831390381
    },
    {
      "auxiliary_loss_clip": 0.01086875,
      "auxiliary_loss_mlp": 0.01038304,
      "balance_loss_clip": 1.04686034,
      "balance_loss_mlp": 1.024369,
      "epoch": 0.47503381932962574,
      "flos": 31938770592000.0,
      "grad_norm": 1.9842125223785956,
      "language_loss": 0.70866525,
      "learning_rate": 2.2580756743868273e-06,
      "loss": 0.72991699,
      "num_input_tokens_seen": 169808825,
      "step": 7901,
      "time_per_iteration": 2.6424949169158936
    },
    {
      "auxiliary_loss_clip": 0.01101185,
      "auxiliary_loss_mlp": 0.01050744,
      "balance_loss_clip": 1.04841745,
      "balance_loss_mlp": 1.03569961,
      "epoch": 0.4750939425822937,
      "flos": 22127257860480.0,
      "grad_norm": 1.8829046912479168,
      "language_loss": 0.73688036,
      "learning_rate": 2.2576894638425636e-06,
      "loss": 0.75839967,
      "num_input_tokens_seen": 169827590,
      "step": 7902,
      "time_per_iteration": 2.564584732055664
    },
    {
      "auxiliary_loss_clip": 0.01084355,
      "auxiliary_loss_mlp": 0.01034364,
      "balance_loss_clip": 1.04687369,
      "balance_loss_mlp": 1.02166867,
      "epoch": 0.47515406583496167,
      "flos": 20850094903680.0,
      "grad_norm": 1.9668977996043955,
      "language_loss": 0.68993938,
      "learning_rate": 2.257303243526688e-06,
      "loss": 0.71112657,
      "num_input_tokens_seen": 169844925,
      "step": 7903,
      "time_per_iteration": 2.548550844192505
    },
    {
      "auxiliary_loss_clip": 0.01098501,
      "auxiliary_loss_mlp": 0.01032627,
      "balance_loss_clip": 1.04421973,
      "balance_loss_mlp": 1.02002096,
      "epoch": 0.47521418908762963,
      "flos": 17524802448000.0,
      "grad_norm": 1.4669599633873274,
      "language_loss": 0.72007382,
      "learning_rate": 2.256917013453848e-06,
      "loss": 0.7413851,
      "num_input_tokens_seen": 169862705,
      "step": 7904,
      "time_per_iteration": 2.5268239974975586
    },
    {
      "auxiliary_loss_clip": 0.0106323,
      "auxiliary_loss_mlp": 0.01041948,
      "balance_loss_clip": 1.04060245,
      "balance_loss_mlp": 1.02758288,
      "epoch": 0.4752743123402976,
      "flos": 20559434048640.0,
      "grad_norm": 1.5730561824044802,
      "language_loss": 0.86303055,
      "learning_rate": 2.25653077363869e-06,
      "loss": 0.88408238,
      "num_input_tokens_seen": 169880155,
      "step": 7905,
      "time_per_iteration": 2.60209321975708
    },
    {
      "auxiliary_loss_clip": 0.01103136,
      "auxiliary_loss_mlp": 0.01033589,
      "balance_loss_clip": 1.04284191,
      "balance_loss_mlp": 1.02098811,
      "epoch": 0.47533443559296557,
      "flos": 26360623071360.0,
      "grad_norm": 1.561531353196979,
      "language_loss": 0.82168734,
      "learning_rate": 2.2561445240958583e-06,
      "loss": 0.84305465,
      "num_input_tokens_seen": 169901525,
      "step": 7906,
      "time_per_iteration": 2.5461008548736572
    },
    {
      "auxiliary_loss_clip": 0.01028043,
      "auxiliary_loss_mlp": 0.01000803,
      "balance_loss_clip": 1.03249574,
      "balance_loss_mlp": 0.99935418,
      "epoch": 0.47539455884563353,
      "flos": 65949660967680.0,
      "grad_norm": 0.6676522658775641,
      "language_loss": 0.58979368,
      "learning_rate": 2.255758264840002e-06,
      "loss": 0.61008215,
      "num_input_tokens_seen": 169970345,
      "step": 7907,
      "time_per_iteration": 3.273958921432495
    },
    {
      "auxiliary_loss_clip": 0.01110473,
      "auxiliary_loss_mlp": 0.01036867,
      "balance_loss_clip": 1.04586864,
      "balance_loss_mlp": 1.02359951,
      "epoch": 0.4754546820983015,
      "flos": 17238128002560.0,
      "grad_norm": 2.1791024539493633,
      "language_loss": 0.81223208,
      "learning_rate": 2.255371995885765e-06,
      "loss": 0.83370554,
      "num_input_tokens_seen": 169986440,
      "step": 7908,
      "time_per_iteration": 2.473233699798584
    },
    {
      "auxiliary_loss_clip": 0.01113691,
      "auxiliary_loss_mlp": 0.01043627,
      "balance_loss_clip": 1.04707849,
      "balance_loss_mlp": 1.02957249,
      "epoch": 0.47551480535096946,
      "flos": 19825886499840.0,
      "grad_norm": 1.6361442303373408,
      "language_loss": 0.73846221,
      "learning_rate": 2.254985717247797e-06,
      "loss": 0.7600354,
      "num_input_tokens_seen": 170005705,
      "step": 7909,
      "time_per_iteration": 2.5353007316589355
    },
    {
      "auxiliary_loss_clip": 0.01097418,
      "auxiliary_loss_mlp": 0.01040114,
      "balance_loss_clip": 1.0461942,
      "balance_loss_mlp": 1.02612472,
      "epoch": 0.4755749286036375,
      "flos": 22163958581760.0,
      "grad_norm": 1.5886671237432652,
      "language_loss": 0.7530821,
      "learning_rate": 2.2545994289407457e-06,
      "loss": 0.7744574,
      "num_input_tokens_seen": 170023415,
      "step": 7910,
      "time_per_iteration": 2.5333316326141357
    },
    {
      "auxiliary_loss_clip": 0.0110793,
      "auxiliary_loss_mlp": 0.01031387,
      "balance_loss_clip": 1.04362202,
      "balance_loss_mlp": 1.01929331,
      "epoch": 0.47563505185630545,
      "flos": 21648280976640.0,
      "grad_norm": 1.9210796249640134,
      "language_loss": 0.78899139,
      "learning_rate": 2.2542131309792577e-06,
      "loss": 0.81038457,
      "num_input_tokens_seen": 170042395,
      "step": 7911,
      "time_per_iteration": 2.516482353210449
    },
    {
      "auxiliary_loss_clip": 0.01095383,
      "auxiliary_loss_mlp": 0.00790299,
      "balance_loss_clip": 1.04274929,
      "balance_loss_mlp": 1.01078188,
      "epoch": 0.4756951751089734,
      "flos": 20628777254400.0,
      "grad_norm": 1.671895736457528,
      "language_loss": 0.76453614,
      "learning_rate": 2.253826823377983e-06,
      "loss": 0.78339297,
      "num_input_tokens_seen": 170061610,
      "step": 7912,
      "time_per_iteration": 2.5568339824676514
    },
    {
      "auxiliary_loss_clip": 0.01121226,
      "auxiliary_loss_mlp": 0.0104083,
      "balance_loss_clip": 1.04505527,
      "balance_loss_mlp": 1.02795553,
      "epoch": 0.4757552983616414,
      "flos": 25848788221440.0,
      "grad_norm": 1.590925139416758,
      "language_loss": 0.74086475,
      "learning_rate": 2.253440506151569e-06,
      "loss": 0.76248533,
      "num_input_tokens_seen": 170083505,
      "step": 7913,
      "time_per_iteration": 2.5217502117156982
    },
    {
      "auxiliary_loss_clip": 0.01104029,
      "auxiliary_loss_mlp": 0.01033695,
      "balance_loss_clip": 1.04732132,
      "balance_loss_mlp": 1.01940775,
      "epoch": 0.47581542161430934,
      "flos": 18223013992320.0,
      "grad_norm": 2.0117762738409053,
      "language_loss": 0.72284842,
      "learning_rate": 2.253054179314666e-06,
      "loss": 0.74422562,
      "num_input_tokens_seen": 170100690,
      "step": 7914,
      "time_per_iteration": 2.484370470046997
    },
    {
      "auxiliary_loss_clip": 0.01102276,
      "auxiliary_loss_mlp": 0.01037785,
      "balance_loss_clip": 1.04976499,
      "balance_loss_mlp": 1.02491081,
      "epoch": 0.4758755448669773,
      "flos": 21579763783680.0,
      "grad_norm": 2.633992041386149,
      "language_loss": 0.64630121,
      "learning_rate": 2.2526678428819227e-06,
      "loss": 0.66770184,
      "num_input_tokens_seen": 170119240,
      "step": 7915,
      "time_per_iteration": 3.911830425262451
    },
    {
      "auxiliary_loss_clip": 0.01118066,
      "auxiliary_loss_mlp": 0.01036492,
      "balance_loss_clip": 1.04475534,
      "balance_loss_mlp": 1.02325964,
      "epoch": 0.47593566811964527,
      "flos": 15231152511360.0,
      "grad_norm": 1.7175023619391758,
      "language_loss": 0.77001595,
      "learning_rate": 2.2522814968679896e-06,
      "loss": 0.79156148,
      "num_input_tokens_seen": 170136450,
      "step": 7916,
      "time_per_iteration": 3.809297561645508
    },
    {
      "auxiliary_loss_clip": 0.01122374,
      "auxiliary_loss_mlp": 0.01033558,
      "balance_loss_clip": 1.04574382,
      "balance_loss_mlp": 1.02098703,
      "epoch": 0.47599579137231324,
      "flos": 21543242630400.0,
      "grad_norm": 2.0782437350111653,
      "language_loss": 0.64476168,
      "learning_rate": 2.2518951412875173e-06,
      "loss": 0.66632098,
      "num_input_tokens_seen": 170155295,
      "step": 7917,
      "time_per_iteration": 2.46280837059021
    },
    {
      "auxiliary_loss_clip": 0.01026135,
      "auxiliary_loss_mlp": 0.01003682,
      "balance_loss_clip": 1.03013968,
      "balance_loss_mlp": 1.00223351,
      "epoch": 0.4760559146249812,
      "flos": 64554602595840.0,
      "grad_norm": 0.8396465506933027,
      "language_loss": 0.65727633,
      "learning_rate": 2.2515087761551557e-06,
      "loss": 0.67757452,
      "num_input_tokens_seen": 170222325,
      "step": 7918,
      "time_per_iteration": 3.1705453395843506
    },
    {
      "auxiliary_loss_clip": 0.0111256,
      "auxiliary_loss_mlp": 0.00787714,
      "balance_loss_clip": 1.04609179,
      "balance_loss_mlp": 1.00953531,
      "epoch": 0.47611603787764917,
      "flos": 22233876405120.0,
      "grad_norm": 1.5987381903912339,
      "language_loss": 0.68757474,
      "learning_rate": 2.2511224014855563e-06,
      "loss": 0.70657742,
      "num_input_tokens_seen": 170241625,
      "step": 7919,
      "time_per_iteration": 3.9664504528045654
    },
    {
      "auxiliary_loss_clip": 0.0110193,
      "auxiliary_loss_mlp": 0.0103276,
      "balance_loss_clip": 1.04592621,
      "balance_loss_mlp": 1.01939034,
      "epoch": 0.47617616113031713,
      "flos": 22780005765120.0,
      "grad_norm": 2.6398904412739266,
      "language_loss": 0.74782836,
      "learning_rate": 2.2507360172933694e-06,
      "loss": 0.76917523,
      "num_input_tokens_seen": 170262470,
      "step": 7920,
      "time_per_iteration": 2.596222162246704
    },
    {
      "auxiliary_loss_clip": 0.01104455,
      "auxiliary_loss_mlp": 0.01030631,
      "balance_loss_clip": 1.04631662,
      "balance_loss_mlp": 1.01638579,
      "epoch": 0.4762362843829851,
      "flos": 24133802388480.0,
      "grad_norm": 1.5695924284316067,
      "language_loss": 0.77119309,
      "learning_rate": 2.2503496235932487e-06,
      "loss": 0.79254395,
      "num_input_tokens_seen": 170283460,
      "step": 7921,
      "time_per_iteration": 3.9479143619537354
    },
    {
      "auxiliary_loss_clip": 0.01103807,
      "auxiliary_loss_mlp": 0.0103944,
      "balance_loss_clip": 1.04562604,
      "balance_loss_mlp": 1.02472925,
      "epoch": 0.47629640763565306,
      "flos": 22452069571200.0,
      "grad_norm": 1.587347881364558,
      "language_loss": 0.78345311,
      "learning_rate": 2.249963220399845e-06,
      "loss": 0.80488563,
      "num_input_tokens_seen": 170304225,
      "step": 7922,
      "time_per_iteration": 2.563905954360962
    },
    {
      "auxiliary_loss_clip": 0.0109222,
      "auxiliary_loss_mlp": 0.0103327,
      "balance_loss_clip": 1.04757655,
      "balance_loss_mlp": 1.01851153,
      "epoch": 0.4763565308883211,
      "flos": 11181398647680.0,
      "grad_norm": 1.7929784272903542,
      "language_loss": 0.72654396,
      "learning_rate": 2.2495768077278104e-06,
      "loss": 0.74779886,
      "num_input_tokens_seen": 170322110,
      "step": 7923,
      "time_per_iteration": 2.533186912536621
    },
    {
      "auxiliary_loss_clip": 0.0109164,
      "auxiliary_loss_mlp": 0.01033432,
      "balance_loss_clip": 1.04366875,
      "balance_loss_mlp": 1.02012873,
      "epoch": 0.47641665414098905,
      "flos": 22382151747840.0,
      "grad_norm": 2.0167660621365147,
      "language_loss": 0.81609678,
      "learning_rate": 2.2491903855917992e-06,
      "loss": 0.83734751,
      "num_input_tokens_seen": 170340700,
      "step": 7924,
      "time_per_iteration": 2.566941738128662
    },
    {
      "auxiliary_loss_clip": 0.01122371,
      "auxiliary_loss_mlp": 0.01035016,
      "balance_loss_clip": 1.05002296,
      "balance_loss_mlp": 1.0203476,
      "epoch": 0.476476777393657,
      "flos": 25046148862080.0,
      "grad_norm": 1.7953145071776975,
      "language_loss": 0.80332279,
      "learning_rate": 2.2488039540064626e-06,
      "loss": 0.82489669,
      "num_input_tokens_seen": 170359780,
      "step": 7925,
      "time_per_iteration": 2.5432941913604736
    },
    {
      "auxiliary_loss_clip": 0.01098328,
      "auxiliary_loss_mlp": 0.01037863,
      "balance_loss_clip": 1.04421604,
      "balance_loss_mlp": 1.02476764,
      "epoch": 0.476536900646325,
      "flos": 27269916888960.0,
      "grad_norm": 1.6033672469144027,
      "language_loss": 0.72047043,
      "learning_rate": 2.2484175129864558e-06,
      "loss": 0.74183226,
      "num_input_tokens_seen": 170381260,
      "step": 7926,
      "time_per_iteration": 2.6178035736083984
    },
    {
      "auxiliary_loss_clip": 0.01117145,
      "auxiliary_loss_mlp": 0.01031365,
      "balance_loss_clip": 1.04817152,
      "balance_loss_mlp": 1.01697636,
      "epoch": 0.47659702389899294,
      "flos": 25301401885440.0,
      "grad_norm": 2.0501690344687744,
      "language_loss": 0.67980039,
      "learning_rate": 2.248031062546432e-06,
      "loss": 0.70128548,
      "num_input_tokens_seen": 170400595,
      "step": 7927,
      "time_per_iteration": 2.5520544052124023
    },
    {
      "auxiliary_loss_clip": 0.01087383,
      "auxiliary_loss_mlp": 0.01029788,
      "balance_loss_clip": 1.0466857,
      "balance_loss_mlp": 1.01704431,
      "epoch": 0.4766571471516609,
      "flos": 25992861672960.0,
      "grad_norm": 1.5605122569709726,
      "language_loss": 0.6821239,
      "learning_rate": 2.247644602701045e-06,
      "loss": 0.70329559,
      "num_input_tokens_seen": 170421110,
      "step": 7928,
      "time_per_iteration": 2.629636764526367
    },
    {
      "auxiliary_loss_clip": 0.01123896,
      "auxiliary_loss_mlp": 0.01033224,
      "balance_loss_clip": 1.04616499,
      "balance_loss_mlp": 1.01990843,
      "epoch": 0.4767172704043289,
      "flos": 16032211672320.0,
      "grad_norm": 2.00702516433815,
      "language_loss": 0.7849443,
      "learning_rate": 2.2472581334649496e-06,
      "loss": 0.80651551,
      "num_input_tokens_seen": 170436700,
      "step": 7929,
      "time_per_iteration": 2.427865505218506
    },
    {
      "auxiliary_loss_clip": 0.01101275,
      "auxiliary_loss_mlp": 0.01033625,
      "balance_loss_clip": 1.04529786,
      "balance_loss_mlp": 1.02041042,
      "epoch": 0.47677739365699684,
      "flos": 39235351651200.0,
      "grad_norm": 1.8095765079488166,
      "language_loss": 0.66343653,
      "learning_rate": 2.2468716548528016e-06,
      "loss": 0.68478549,
      "num_input_tokens_seen": 170459555,
      "step": 7930,
      "time_per_iteration": 2.6742751598358154
    },
    {
      "auxiliary_loss_clip": 0.01115544,
      "auxiliary_loss_mlp": 0.01032584,
      "balance_loss_clip": 1.04621625,
      "balance_loss_mlp": 1.01940536,
      "epoch": 0.4768375169096648,
      "flos": 24717781704960.0,
      "grad_norm": 1.6642785757148895,
      "language_loss": 0.79614997,
      "learning_rate": 2.2464851668792555e-06,
      "loss": 0.8176313,
      "num_input_tokens_seen": 170479175,
      "step": 7931,
      "time_per_iteration": 2.5198466777801514
    },
    {
      "auxiliary_loss_clip": 0.01095633,
      "auxiliary_loss_mlp": 0.01033906,
      "balance_loss_clip": 1.04190636,
      "balance_loss_mlp": 1.01947021,
      "epoch": 0.47689764016233277,
      "flos": 22528667324160.0,
      "grad_norm": 2.1045854461336755,
      "language_loss": 0.76201361,
      "learning_rate": 2.2460986695589678e-06,
      "loss": 0.78330898,
      "num_input_tokens_seen": 170498450,
      "step": 7932,
      "time_per_iteration": 2.5161991119384766
    },
    {
      "auxiliary_loss_clip": 0.01101995,
      "auxiliary_loss_mlp": 0.00789817,
      "balance_loss_clip": 1.04675627,
      "balance_loss_mlp": 1.01639545,
      "epoch": 0.47695776341500074,
      "flos": 15120619384320.0,
      "grad_norm": 1.7107425499347664,
      "language_loss": 0.79247594,
      "learning_rate": 2.245712162906593e-06,
      "loss": 0.8113941,
      "num_input_tokens_seen": 170516255,
      "step": 7933,
      "time_per_iteration": 2.518139123916626
    },
    {
      "auxiliary_loss_clip": 0.01120134,
      "auxiliary_loss_mlp": 0.01040114,
      "balance_loss_clip": 1.04700589,
      "balance_loss_mlp": 1.02446151,
      "epoch": 0.4770178866676687,
      "flos": 14678917839360.0,
      "grad_norm": 8.693257429788233,
      "language_loss": 0.73857754,
      "learning_rate": 2.2453256469367888e-06,
      "loss": 0.76018006,
      "num_input_tokens_seen": 170532705,
      "step": 7934,
      "time_per_iteration": 2.4681923389434814
    },
    {
      "auxiliary_loss_clip": 0.0111536,
      "auxiliary_loss_mlp": 0.01032328,
      "balance_loss_clip": 1.04652548,
      "balance_loss_mlp": 1.01898265,
      "epoch": 0.47707800992033667,
      "flos": 22565583527040.0,
      "grad_norm": 1.7949289412479341,
      "language_loss": 0.79820287,
      "learning_rate": 2.244939121664211e-06,
      "loss": 0.81967974,
      "num_input_tokens_seen": 170551925,
      "step": 7935,
      "time_per_iteration": 2.5077433586120605
    },
    {
      "auxiliary_loss_clip": 0.0109932,
      "auxiliary_loss_mlp": 0.01041884,
      "balance_loss_clip": 1.04692435,
      "balance_loss_mlp": 1.02679193,
      "epoch": 0.4771381331730047,
      "flos": 30918225375360.0,
      "grad_norm": 1.760701819261821,
      "language_loss": 0.71020305,
      "learning_rate": 2.2445525871035177e-06,
      "loss": 0.73161507,
      "num_input_tokens_seen": 170572320,
      "step": 7936,
      "time_per_iteration": 2.626458168029785
    },
    {
      "auxiliary_loss_clip": 0.01127231,
      "auxiliary_loss_mlp": 0.01034535,
      "balance_loss_clip": 1.04676819,
      "balance_loss_mlp": 1.02127862,
      "epoch": 0.47719825642567265,
      "flos": 25738901539200.0,
      "grad_norm": 1.9104246093182893,
      "language_loss": 0.67686248,
      "learning_rate": 2.2441660432693656e-06,
      "loss": 0.69848013,
      "num_input_tokens_seen": 170589470,
      "step": 7937,
      "time_per_iteration": 2.4908578395843506
    },
    {
      "auxiliary_loss_clip": 0.01037597,
      "auxiliary_loss_mlp": 0.0100328,
      "balance_loss_clip": 1.02235472,
      "balance_loss_mlp": 1.00184917,
      "epoch": 0.4772583796783406,
      "flos": 66355128668160.0,
      "grad_norm": 0.7110192493229954,
      "language_loss": 0.56349671,
      "learning_rate": 2.2437794901764128e-06,
      "loss": 0.58390546,
      "num_input_tokens_seen": 170662265,
      "step": 7938,
      "time_per_iteration": 3.2714197635650635
    },
    {
      "auxiliary_loss_clip": 0.01099767,
      "auxiliary_loss_mlp": 0.01050533,
      "balance_loss_clip": 1.04658461,
      "balance_loss_mlp": 1.03404665,
      "epoch": 0.4773185029310086,
      "flos": 22051091070720.0,
      "grad_norm": 1.699524750303965,
      "language_loss": 0.88906789,
      "learning_rate": 2.243392927839317e-06,
      "loss": 0.91057086,
      "num_input_tokens_seen": 170679680,
      "step": 7939,
      "time_per_iteration": 2.5305051803588867
    },
    {
      "auxiliary_loss_clip": 0.01112727,
      "auxiliary_loss_mlp": 0.01037542,
      "balance_loss_clip": 1.04337811,
      "balance_loss_mlp": 1.02433968,
      "epoch": 0.47737862618367655,
      "flos": 16727801523840.0,
      "grad_norm": 1.8968093786999194,
      "language_loss": 0.77089846,
      "learning_rate": 2.2430063562727367e-06,
      "loss": 0.79240119,
      "num_input_tokens_seen": 170697340,
      "step": 7940,
      "time_per_iteration": 2.483955144882202
    },
    {
      "auxiliary_loss_clip": 0.01099375,
      "auxiliary_loss_mlp": 0.01038243,
      "balance_loss_clip": 1.04545951,
      "balance_loss_mlp": 1.02530873,
      "epoch": 0.4774387494363445,
      "flos": 19609453100160.0,
      "grad_norm": 1.7398917826531954,
      "language_loss": 0.84929073,
      "learning_rate": 2.2426197754913322e-06,
      "loss": 0.87066686,
      "num_input_tokens_seen": 170714905,
      "step": 7941,
      "time_per_iteration": 2.516010046005249
    },
    {
      "auxiliary_loss_clip": 0.01107913,
      "auxiliary_loss_mlp": 0.01037107,
      "balance_loss_clip": 1.04793012,
      "balance_loss_mlp": 1.0230881,
      "epoch": 0.4774988726890125,
      "flos": 16653969118080.0,
      "grad_norm": 1.833841966006201,
      "language_loss": 0.76032317,
      "learning_rate": 2.24223318550976e-06,
      "loss": 0.78177339,
      "num_input_tokens_seen": 170731810,
      "step": 7942,
      "time_per_iteration": 2.5157697200775146
    },
    {
      "auxiliary_loss_clip": 0.01115821,
      "auxiliary_loss_mlp": 0.01036617,
      "balance_loss_clip": 1.05051303,
      "balance_loss_mlp": 1.02314651,
      "epoch": 0.47755899594168044,
      "flos": 20485565729280.0,
      "grad_norm": 1.808400357675475,
      "language_loss": 0.64583564,
      "learning_rate": 2.241846586342682e-06,
      "loss": 0.66736001,
      "num_input_tokens_seen": 170750270,
      "step": 7943,
      "time_per_iteration": 2.5122225284576416
    },
    {
      "auxiliary_loss_clip": 0.0109831,
      "auxiliary_loss_mlp": 0.01038951,
      "balance_loss_clip": 1.0486815,
      "balance_loss_mlp": 1.02374005,
      "epoch": 0.4776191191943484,
      "flos": 21652806090240.0,
      "grad_norm": 1.597904530364636,
      "language_loss": 0.73229277,
      "learning_rate": 2.2414599780047577e-06,
      "loss": 0.75366533,
      "num_input_tokens_seen": 170769015,
      "step": 7944,
      "time_per_iteration": 2.582731008529663
    },
    {
      "auxiliary_loss_clip": 0.01116027,
      "auxiliary_loss_mlp": 0.01038434,
      "balance_loss_clip": 1.05278254,
      "balance_loss_mlp": 1.02372324,
      "epoch": 0.4776792424470164,
      "flos": 18770220760320.0,
      "grad_norm": 2.166749323689797,
      "language_loss": 0.67973548,
      "learning_rate": 2.2410733605106456e-06,
      "loss": 0.70128012,
      "num_input_tokens_seen": 170785725,
      "step": 7945,
      "time_per_iteration": 2.498991012573242
    },
    {
      "auxiliary_loss_clip": 0.0108794,
      "auxiliary_loss_mlp": 0.00793568,
      "balance_loss_clip": 1.04279232,
      "balance_loss_mlp": 1.01537585,
      "epoch": 0.47773936569968434,
      "flos": 29715828577920.0,
      "grad_norm": 1.8123795503095148,
      "language_loss": 0.75423908,
      "learning_rate": 2.240686733875009e-06,
      "loss": 0.77305412,
      "num_input_tokens_seen": 170804600,
      "step": 7946,
      "time_per_iteration": 2.642031669616699
    },
    {
      "auxiliary_loss_clip": 0.01110198,
      "auxiliary_loss_mlp": 0.01044016,
      "balance_loss_clip": 1.04856277,
      "balance_loss_mlp": 1.02900743,
      "epoch": 0.4777994889523523,
      "flos": 24791542283520.0,
      "grad_norm": 1.7604116918559216,
      "language_loss": 0.79638869,
      "learning_rate": 2.240300098112506e-06,
      "loss": 0.81793082,
      "num_input_tokens_seen": 170824230,
      "step": 7947,
      "time_per_iteration": 2.5756144523620605
    },
    {
      "auxiliary_loss_clip": 0.01092892,
      "auxiliary_loss_mlp": 0.01037165,
      "balance_loss_clip": 1.04896283,
      "balance_loss_mlp": 1.02346146,
      "epoch": 0.47785961220502027,
      "flos": 17858161595520.0,
      "grad_norm": 1.7826879333591938,
      "language_loss": 0.7381711,
      "learning_rate": 2.2399134532377998e-06,
      "loss": 0.75947165,
      "num_input_tokens_seen": 170843365,
      "step": 7948,
      "time_per_iteration": 2.539172410964966
    },
    {
      "auxiliary_loss_clip": 0.01105724,
      "auxiliary_loss_mlp": 0.01034616,
      "balance_loss_clip": 1.04681802,
      "balance_loss_mlp": 1.01964355,
      "epoch": 0.4779197354576883,
      "flos": 20266546550400.0,
      "grad_norm": 1.484903326231955,
      "language_loss": 0.78211546,
      "learning_rate": 2.2395267992655514e-06,
      "loss": 0.80351883,
      "num_input_tokens_seen": 170863515,
      "step": 7949,
      "time_per_iteration": 2.536595106124878
    },
    {
      "auxiliary_loss_clip": 0.01095368,
      "auxiliary_loss_mlp": 0.01033488,
      "balance_loss_clip": 1.04348207,
      "balance_loss_mlp": 1.01999378,
      "epoch": 0.47797985871035625,
      "flos": 17056599644160.0,
      "grad_norm": 2.0141560676206165,
      "language_loss": 0.73682326,
      "learning_rate": 2.2391401362104227e-06,
      "loss": 0.75811183,
      "num_input_tokens_seen": 170881245,
      "step": 7950,
      "time_per_iteration": 2.5155513286590576
    },
    {
      "auxiliary_loss_clip": 0.01093854,
      "auxiliary_loss_mlp": 0.01044925,
      "balance_loss_clip": 1.04504406,
      "balance_loss_mlp": 1.02916574,
      "epoch": 0.4780399819630242,
      "flos": 31358418549120.0,
      "grad_norm": 1.689160258156553,
      "language_loss": 0.73826838,
      "learning_rate": 2.2387534640870756e-06,
      "loss": 0.75965613,
      "num_input_tokens_seen": 170901285,
      "step": 7951,
      "time_per_iteration": 2.614638328552246
    },
    {
      "auxiliary_loss_clip": 0.01093145,
      "auxiliary_loss_mlp": 0.01035318,
      "balance_loss_clip": 1.04725981,
      "balance_loss_mlp": 1.02050018,
      "epoch": 0.4781001052156922,
      "flos": 24899597372160.0,
      "grad_norm": 1.9087978471958877,
      "language_loss": 0.80701482,
      "learning_rate": 2.238366782910174e-06,
      "loss": 0.82829946,
      "num_input_tokens_seen": 170919740,
      "step": 7952,
      "time_per_iteration": 2.604689836502075
    },
    {
      "auxiliary_loss_clip": 0.01106182,
      "auxiliary_loss_mlp": 0.01041333,
      "balance_loss_clip": 1.04605055,
      "balance_loss_mlp": 1.02644348,
      "epoch": 0.47816022846836015,
      "flos": 18697717157760.0,
      "grad_norm": 1.651025336400082,
      "language_loss": 0.78296095,
      "learning_rate": 2.23798009269438e-06,
      "loss": 0.80443615,
      "num_input_tokens_seen": 170938510,
      "step": 7953,
      "time_per_iteration": 2.533233642578125
    },
    {
      "auxiliary_loss_clip": 0.0111994,
      "auxiliary_loss_mlp": 0.0103718,
      "balance_loss_clip": 1.04947746,
      "balance_loss_mlp": 1.02298796,
      "epoch": 0.4782203517210281,
      "flos": 11977573559040.0,
      "grad_norm": 2.2192884360369445,
      "language_loss": 0.84048921,
      "learning_rate": 2.2375933934543566e-06,
      "loss": 0.86206043,
      "num_input_tokens_seen": 170951170,
      "step": 7954,
      "time_per_iteration": 3.8337700366973877
    },
    {
      "auxiliary_loss_clip": 0.01095608,
      "auxiliary_loss_mlp": 0.01040932,
      "balance_loss_clip": 1.045017,
      "balance_loss_mlp": 1.02649009,
      "epoch": 0.4782804749736961,
      "flos": 20813501923200.0,
      "grad_norm": 1.570816463641743,
      "language_loss": 0.70689481,
      "learning_rate": 2.237206685204768e-06,
      "loss": 0.72826016,
      "num_input_tokens_seen": 170970990,
      "step": 7955,
      "time_per_iteration": 3.910712957382202
    },
    {
      "auxiliary_loss_clip": 0.01109581,
      "auxiliary_loss_mlp": 0.01037219,
      "balance_loss_clip": 1.04923904,
      "balance_loss_mlp": 1.02373636,
      "epoch": 0.47834059822636404,
      "flos": 23840304359040.0,
      "grad_norm": 1.5825527638432113,
      "language_loss": 0.81722689,
      "learning_rate": 2.2368199679602787e-06,
      "loss": 0.83869493,
      "num_input_tokens_seen": 170991215,
      "step": 7956,
      "time_per_iteration": 2.5663464069366455
    },
    {
      "auxiliary_loss_clip": 0.01106545,
      "auxiliary_loss_mlp": 0.01037945,
      "balance_loss_clip": 1.04988432,
      "balance_loss_mlp": 1.02253079,
      "epoch": 0.478400721479032,
      "flos": 22633777497600.0,
      "grad_norm": 1.989595018942994,
      "language_loss": 0.84586465,
      "learning_rate": 2.2364332417355516e-06,
      "loss": 0.86730957,
      "num_input_tokens_seen": 171007325,
      "step": 7957,
      "time_per_iteration": 2.5528128147125244
    },
    {
      "auxiliary_loss_clip": 0.01115839,
      "auxiliary_loss_mlp": 0.01038473,
      "balance_loss_clip": 1.04694676,
      "balance_loss_mlp": 1.02495456,
      "epoch": 0.4784608447317,
      "flos": 19354954262400.0,
      "grad_norm": 1.6605531612423299,
      "language_loss": 0.79764867,
      "learning_rate": 2.2360465065452527e-06,
      "loss": 0.81919181,
      "num_input_tokens_seen": 171025650,
      "step": 7958,
      "time_per_iteration": 3.8734793663024902
    },
    {
      "auxiliary_loss_clip": 0.01086009,
      "auxiliary_loss_mlp": 0.00794047,
      "balance_loss_clip": 1.04212773,
      "balance_loss_mlp": 1.01255167,
      "epoch": 0.47852096798436794,
      "flos": 24021114445440.0,
      "grad_norm": 1.9865551539004418,
      "language_loss": 0.82634449,
      "learning_rate": 2.235659762404047e-06,
      "loss": 0.84514499,
      "num_input_tokens_seen": 171045045,
      "step": 7959,
      "time_per_iteration": 2.586444139480591
    },
    {
      "auxiliary_loss_clip": 0.01089134,
      "auxiliary_loss_mlp": 0.01033418,
      "balance_loss_clip": 1.046875,
      "balance_loss_mlp": 1.02063251,
      "epoch": 0.4785810912370359,
      "flos": 25666433850240.0,
      "grad_norm": 2.590380949950751,
      "language_loss": 0.7345264,
      "learning_rate": 2.235273009326599e-06,
      "loss": 0.75575191,
      "num_input_tokens_seen": 171062910,
      "step": 7960,
      "time_per_iteration": 4.0475547313690186
    },
    {
      "auxiliary_loss_clip": 0.01089162,
      "auxiliary_loss_mlp": 0.01036314,
      "balance_loss_clip": 1.04598641,
      "balance_loss_mlp": 1.02339721,
      "epoch": 0.47864121448970387,
      "flos": 21432134885760.0,
      "grad_norm": 1.9727759984894584,
      "language_loss": 0.76854444,
      "learning_rate": 2.2348862473275745e-06,
      "loss": 0.78979921,
      "num_input_tokens_seen": 171080875,
      "step": 7961,
      "time_per_iteration": 2.5597259998321533
    },
    {
      "auxiliary_loss_clip": 0.01086172,
      "auxiliary_loss_mlp": 0.01030413,
      "balance_loss_clip": 1.04772818,
      "balance_loss_mlp": 1.01707292,
      "epoch": 0.47870133774237184,
      "flos": 16143894034560.0,
      "grad_norm": 1.7352285306360908,
      "language_loss": 0.77921069,
      "learning_rate": 2.2344994764216405e-06,
      "loss": 0.80037653,
      "num_input_tokens_seen": 171099190,
      "step": 7962,
      "time_per_iteration": 2.546804904937744
    },
    {
      "auxiliary_loss_clip": 0.01100811,
      "auxiliary_loss_mlp": 0.01037213,
      "balance_loss_clip": 1.04947925,
      "balance_loss_mlp": 1.02359927,
      "epoch": 0.47876146099503986,
      "flos": 26906788344960.0,
      "grad_norm": 1.6786853142608151,
      "language_loss": 0.6504966,
      "learning_rate": 2.2341126966234635e-06,
      "loss": 0.67187691,
      "num_input_tokens_seen": 171119060,
      "step": 7963,
      "time_per_iteration": 2.594326972961426
    },
    {
      "auxiliary_loss_clip": 0.01115553,
      "auxiliary_loss_mlp": 0.01031533,
      "balance_loss_clip": 1.04705954,
      "balance_loss_mlp": 1.01712084,
      "epoch": 0.4788215842477078,
      "flos": 45332085778560.0,
      "grad_norm": 2.148445581241582,
      "language_loss": 0.77539259,
      "learning_rate": 2.2337259079477083e-06,
      "loss": 0.7968635,
      "num_input_tokens_seen": 171141900,
      "step": 7964,
      "time_per_iteration": 2.7331790924072266
    },
    {
      "auxiliary_loss_clip": 0.0111937,
      "auxiliary_loss_mlp": 0.01033811,
      "balance_loss_clip": 1.0482614,
      "balance_loss_mlp": 1.01817024,
      "epoch": 0.4788817075003758,
      "flos": 22237180456320.0,
      "grad_norm": 1.6952171789585355,
      "language_loss": 0.76253617,
      "learning_rate": 2.233339110409044e-06,
      "loss": 0.78406787,
      "num_input_tokens_seen": 171161045,
      "step": 7965,
      "time_per_iteration": 2.5991547107696533
    },
    {
      "auxiliary_loss_clip": 0.01073149,
      "auxiliary_loss_mlp": 0.0104244,
      "balance_loss_clip": 1.04382575,
      "balance_loss_mlp": 1.02732444,
      "epoch": 0.47894183075304375,
      "flos": 16471183783680.0,
      "grad_norm": 2.715780852327255,
      "language_loss": 0.74698782,
      "learning_rate": 2.232952304022137e-06,
      "loss": 0.76814365,
      "num_input_tokens_seen": 171179675,
      "step": 7966,
      "time_per_iteration": 2.5939178466796875
    },
    {
      "auxiliary_loss_clip": 0.01101344,
      "auxiliary_loss_mlp": 0.01036206,
      "balance_loss_clip": 1.0493474,
      "balance_loss_mlp": 1.02138865,
      "epoch": 0.4790019540057117,
      "flos": 24282688262400.0,
      "grad_norm": 1.6130462295433614,
      "language_loss": 0.73149097,
      "learning_rate": 2.232565488801655e-06,
      "loss": 0.75286651,
      "num_input_tokens_seen": 171201175,
      "step": 7967,
      "time_per_iteration": 2.605095148086548
    },
    {
      "auxiliary_loss_clip": 0.01099671,
      "auxiliary_loss_mlp": 0.01030357,
      "balance_loss_clip": 1.04426003,
      "balance_loss_mlp": 1.01677942,
      "epoch": 0.4790620772583797,
      "flos": 25666469763840.0,
      "grad_norm": 1.8537947714380258,
      "language_loss": 0.79240012,
      "learning_rate": 2.232178664762267e-06,
      "loss": 0.81370044,
      "num_input_tokens_seen": 171221750,
      "step": 7968,
      "time_per_iteration": 2.590744972229004
    },
    {
      "auxiliary_loss_clip": 0.0102481,
      "auxiliary_loss_mlp": 0.01001746,
      "balance_loss_clip": 1.02739024,
      "balance_loss_mlp": 1.00016057,
      "epoch": 0.47912220051104765,
      "flos": 69428077102080.0,
      "grad_norm": 0.7591894852888579,
      "language_loss": 0.6227926,
      "learning_rate": 2.2317918319186408e-06,
      "loss": 0.64305818,
      "num_input_tokens_seen": 171292235,
      "step": 7969,
      "time_per_iteration": 3.2979280948638916
    },
    {
      "auxiliary_loss_clip": 0.01087407,
      "auxiliary_loss_mlp": 0.01031351,
      "balance_loss_clip": 1.04925132,
      "balance_loss_mlp": 1.01807094,
      "epoch": 0.4791823237637156,
      "flos": 24168922911360.0,
      "grad_norm": 1.3992240489168608,
      "language_loss": 0.7716403,
      "learning_rate": 2.2314049902854446e-06,
      "loss": 0.79282784,
      "num_input_tokens_seen": 171312215,
      "step": 7970,
      "time_per_iteration": 2.5915300846099854
    },
    {
      "auxiliary_loss_clip": 0.01110429,
      "auxiliary_loss_mlp": 0.01035666,
      "balance_loss_clip": 1.04323077,
      "balance_loss_mlp": 1.02144432,
      "epoch": 0.4792424470163836,
      "flos": 24751465683840.0,
      "grad_norm": 1.7241747510640428,
      "language_loss": 0.70566279,
      "learning_rate": 2.231018139877349e-06,
      "loss": 0.72712374,
      "num_input_tokens_seen": 171332975,
      "step": 7971,
      "time_per_iteration": 2.552602767944336
    },
    {
      "auxiliary_loss_clip": 0.01079764,
      "auxiliary_loss_mlp": 0.01031007,
      "balance_loss_clip": 1.0461762,
      "balance_loss_mlp": 1.01582003,
      "epoch": 0.47930257026905154,
      "flos": 23257905240960.0,
      "grad_norm": 1.3396476627013838,
      "language_loss": 0.80201268,
      "learning_rate": 2.230631280709021e-06,
      "loss": 0.82312042,
      "num_input_tokens_seen": 171353880,
      "step": 7972,
      "time_per_iteration": 2.6406657695770264
    },
    {
      "auxiliary_loss_clip": 0.01118018,
      "auxiliary_loss_mlp": 0.01031018,
      "balance_loss_clip": 1.0493952,
      "balance_loss_mlp": 1.01668942,
      "epoch": 0.4793626935217195,
      "flos": 14064091718400.0,
      "grad_norm": 1.9637957087103555,
      "language_loss": 0.69868326,
      "learning_rate": 2.2302444127951327e-06,
      "loss": 0.72017366,
      "num_input_tokens_seen": 171370930,
      "step": 7973,
      "time_per_iteration": 2.4983367919921875
    },
    {
      "auxiliary_loss_clip": 0.01117865,
      "auxiliary_loss_mlp": 0.01035348,
      "balance_loss_clip": 1.05195057,
      "balance_loss_mlp": 1.02246189,
      "epoch": 0.4794228167743875,
      "flos": 21798854789760.0,
      "grad_norm": 1.7663354622427891,
      "language_loss": 0.78812718,
      "learning_rate": 2.2298575361503523e-06,
      "loss": 0.80965924,
      "num_input_tokens_seen": 171387575,
      "step": 7974,
      "time_per_iteration": 2.5343334674835205
    },
    {
      "auxiliary_loss_clip": 0.01034345,
      "auxiliary_loss_mlp": 0.0102342,
      "balance_loss_clip": 1.02848315,
      "balance_loss_mlp": 1.02150047,
      "epoch": 0.47948294002705544,
      "flos": 66968805553920.0,
      "grad_norm": 0.7705526526284165,
      "language_loss": 0.54045796,
      "learning_rate": 2.2294706507893517e-06,
      "loss": 0.56103557,
      "num_input_tokens_seen": 171449980,
      "step": 7975,
      "time_per_iteration": 3.1808247566223145
    },
    {
      "auxiliary_loss_clip": 0.01111551,
      "auxiliary_loss_mlp": 0.010396,
      "balance_loss_clip": 1.04666221,
      "balance_loss_mlp": 1.02427626,
      "epoch": 0.47954306327972346,
      "flos": 12422471414400.0,
      "grad_norm": 1.9090714661436667,
      "language_loss": 0.89613652,
      "learning_rate": 2.2290837567268008e-06,
      "loss": 0.91764802,
      "num_input_tokens_seen": 171465290,
      "step": 7976,
      "time_per_iteration": 2.5180437564849854
    },
    {
      "auxiliary_loss_clip": 0.01132828,
      "auxiliary_loss_mlp": 0.01038217,
      "balance_loss_clip": 1.05128086,
      "balance_loss_mlp": 1.0232687,
      "epoch": 0.4796031865323914,
      "flos": 18361951799040.0,
      "grad_norm": 2.042549544397765,
      "language_loss": 0.74030507,
      "learning_rate": 2.2286968539773713e-06,
      "loss": 0.76201552,
      "num_input_tokens_seen": 171481130,
      "step": 7977,
      "time_per_iteration": 2.469179391860962
    },
    {
      "auxiliary_loss_clip": 0.01115423,
      "auxiliary_loss_mlp": 0.00790689,
      "balance_loss_clip": 1.04755938,
      "balance_loss_mlp": 1.01507974,
      "epoch": 0.4796633097850594,
      "flos": 21835088634240.0,
      "grad_norm": 1.5191511809010236,
      "language_loss": 0.78275061,
      "learning_rate": 2.228309942555734e-06,
      "loss": 0.8018117,
      "num_input_tokens_seen": 171501140,
      "step": 7978,
      "time_per_iteration": 2.5458803176879883
    },
    {
      "auxiliary_loss_clip": 0.01104341,
      "auxiliary_loss_mlp": 0.0104026,
      "balance_loss_clip": 1.04708838,
      "balance_loss_mlp": 1.02596724,
      "epoch": 0.47972343303772735,
      "flos": 23437350610560.0,
      "grad_norm": 1.7589351522953791,
      "language_loss": 0.8929801,
      "learning_rate": 2.22792302247656e-06,
      "loss": 0.91442609,
      "num_input_tokens_seen": 171519835,
      "step": 7979,
      "time_per_iteration": 2.560894727706909
    },
    {
      "auxiliary_loss_clip": 0.01119673,
      "auxiliary_loss_mlp": 0.01038595,
      "balance_loss_clip": 1.04921412,
      "balance_loss_mlp": 1.02327657,
      "epoch": 0.4797835562903953,
      "flos": 24899776940160.0,
      "grad_norm": 1.4680783202913248,
      "language_loss": 0.7701292,
      "learning_rate": 2.227536093754523e-06,
      "loss": 0.79171193,
      "num_input_tokens_seen": 171540980,
      "step": 7980,
      "time_per_iteration": 2.5638155937194824
    },
    {
      "auxiliary_loss_clip": 0.01096145,
      "auxiliary_loss_mlp": 0.01039311,
      "balance_loss_clip": 1.04366946,
      "balance_loss_mlp": 1.02381372,
      "epoch": 0.4798436795430633,
      "flos": 35042996793600.0,
      "grad_norm": 1.7711988242157224,
      "language_loss": 0.71917713,
      "learning_rate": 2.227149156404295e-06,
      "loss": 0.74053168,
      "num_input_tokens_seen": 171563600,
      "step": 7981,
      "time_per_iteration": 2.67795467376709
    },
    {
      "auxiliary_loss_clip": 0.01124364,
      "auxiliary_loss_mlp": 0.01030261,
      "balance_loss_clip": 1.04866433,
      "balance_loss_mlp": 1.01709414,
      "epoch": 0.47990380279573125,
      "flos": 20590209025920.0,
      "grad_norm": 2.3529311507642032,
      "language_loss": 0.70075023,
      "learning_rate": 2.2267622104405473e-06,
      "loss": 0.72229648,
      "num_input_tokens_seen": 171580700,
      "step": 7982,
      "time_per_iteration": 2.482614278793335
    },
    {
      "auxiliary_loss_clip": 0.0109771,
      "auxiliary_loss_mlp": 0.01032394,
      "balance_loss_clip": 1.04504037,
      "balance_loss_mlp": 1.02035999,
      "epoch": 0.4799639260483992,
      "flos": 26359402008960.0,
      "grad_norm": 1.6224412701761382,
      "language_loss": 0.70710027,
      "learning_rate": 2.2263752558779544e-06,
      "loss": 0.7284013,
      "num_input_tokens_seen": 171602035,
      "step": 7983,
      "time_per_iteration": 2.5717480182647705
    },
    {
      "auxiliary_loss_clip": 0.01039282,
      "auxiliary_loss_mlp": 0.00775621,
      "balance_loss_clip": 1.0227412,
      "balance_loss_mlp": 1.01854312,
      "epoch": 0.4800240493010672,
      "flos": 70979021521920.0,
      "grad_norm": 0.7945471295294019,
      "language_loss": 0.59447157,
      "learning_rate": 2.2259882927311883e-06,
      "loss": 0.61262059,
      "num_input_tokens_seen": 171659215,
      "step": 7984,
      "time_per_iteration": 3.061324119567871
    },
    {
      "auxiliary_loss_clip": 0.01068264,
      "auxiliary_loss_mlp": 0.0104908,
      "balance_loss_clip": 1.04175973,
      "balance_loss_mlp": 1.03271282,
      "epoch": 0.48008417255373514,
      "flos": 17086656349440.0,
      "grad_norm": 1.6277658017439955,
      "language_loss": 0.66462791,
      "learning_rate": 2.2256013210149247e-06,
      "loss": 0.68580139,
      "num_input_tokens_seen": 171675710,
      "step": 7985,
      "time_per_iteration": 2.5727028846740723
    },
    {
      "auxiliary_loss_clip": 0.01103497,
      "auxiliary_loss_mlp": 0.01037872,
      "balance_loss_clip": 1.04410672,
      "balance_loss_mlp": 1.02342367,
      "epoch": 0.4801442958064031,
      "flos": 15413435055360.0,
      "grad_norm": 2.0236654254725783,
      "language_loss": 0.70275807,
      "learning_rate": 2.225214340743835e-06,
      "loss": 0.72417176,
      "num_input_tokens_seen": 171692510,
      "step": 7986,
      "time_per_iteration": 2.5134100914001465
    },
    {
      "auxiliary_loss_clip": 0.01090282,
      "auxiliary_loss_mlp": 0.01038651,
      "balance_loss_clip": 1.04899478,
      "balance_loss_mlp": 1.0241791,
      "epoch": 0.4802044190590711,
      "flos": 11473747441920.0,
      "grad_norm": 1.8262548037559965,
      "language_loss": 0.78741789,
      "learning_rate": 2.2248273519325956e-06,
      "loss": 0.80870724,
      "num_input_tokens_seen": 171710235,
      "step": 7987,
      "time_per_iteration": 2.5443408489227295
    },
    {
      "auxiliary_loss_clip": 0.01074386,
      "auxiliary_loss_mlp": 0.01042989,
      "balance_loss_clip": 1.04479837,
      "balance_loss_mlp": 1.02895761,
      "epoch": 0.48026454231173904,
      "flos": 20951003185920.0,
      "grad_norm": 1.940270962493836,
      "language_loss": 0.74895096,
      "learning_rate": 2.2244403545958812e-06,
      "loss": 0.77012467,
      "num_input_tokens_seen": 171726715,
      "step": 7988,
      "time_per_iteration": 2.583739995956421
    },
    {
      "auxiliary_loss_clip": 0.0109371,
      "auxiliary_loss_mlp": 0.01034147,
      "balance_loss_clip": 1.04683912,
      "balance_loss_mlp": 1.02054477,
      "epoch": 0.48032466556440706,
      "flos": 20448110822400.0,
      "grad_norm": 1.9571861851088301,
      "language_loss": 0.78846782,
      "learning_rate": 2.224053348748365e-06,
      "loss": 0.80974638,
      "num_input_tokens_seen": 171743605,
      "step": 7989,
      "time_per_iteration": 2.5617904663085938
    },
    {
      "auxiliary_loss_clip": 0.01106184,
      "auxiliary_loss_mlp": 0.01039684,
      "balance_loss_clip": 1.04488981,
      "balance_loss_mlp": 1.0247407,
      "epoch": 0.480384788817075,
      "flos": 37120823861760.0,
      "grad_norm": 1.9483961266067635,
      "language_loss": 0.73559642,
      "learning_rate": 2.223666334404724e-06,
      "loss": 0.75705504,
      "num_input_tokens_seen": 171765445,
      "step": 7990,
      "time_per_iteration": 2.669351577758789
    },
    {
      "auxiliary_loss_clip": 0.01037884,
      "auxiliary_loss_mlp": 0.00771339,
      "balance_loss_clip": 1.02086329,
      "balance_loss_mlp": 1.01046467,
      "epoch": 0.480444912069743,
      "flos": 69552577641600.0,
      "grad_norm": 0.7620089327861589,
      "language_loss": 0.59031105,
      "learning_rate": 2.223279311579633e-06,
      "loss": 0.60840327,
      "num_input_tokens_seen": 171830115,
      "step": 7991,
      "time_per_iteration": 3.2058353424072266
    },
    {
      "auxiliary_loss_clip": 0.01113461,
      "auxiliary_loss_mlp": 0.00789998,
      "balance_loss_clip": 1.04475379,
      "balance_loss_mlp": 1.01277518,
      "epoch": 0.48050503532241096,
      "flos": 29822231640960.0,
      "grad_norm": 1.8491582114963667,
      "language_loss": 0.67547894,
      "learning_rate": 2.222892280287768e-06,
      "loss": 0.69451356,
      "num_input_tokens_seen": 171849135,
      "step": 7992,
      "time_per_iteration": 3.9340479373931885
    },
    {
      "auxiliary_loss_clip": 0.01100241,
      "auxiliary_loss_mlp": 0.01037667,
      "balance_loss_clip": 1.04228377,
      "balance_loss_mlp": 1.02321887,
      "epoch": 0.4805651585750789,
      "flos": 23948539015680.0,
      "grad_norm": 1.6055940471661165,
      "language_loss": 0.76036125,
      "learning_rate": 2.2225052405438056e-06,
      "loss": 0.78174031,
      "num_input_tokens_seen": 171868880,
      "step": 7993,
      "time_per_iteration": 2.5635008811950684
    },
    {
      "auxiliary_loss_clip": 0.01085436,
      "auxiliary_loss_mlp": 0.01036296,
      "balance_loss_clip": 1.04647148,
      "balance_loss_mlp": 1.02256322,
      "epoch": 0.4806252818277469,
      "flos": 25665428269440.0,
      "grad_norm": 1.6382368159501521,
      "language_loss": 0.78782767,
      "learning_rate": 2.222118192362422e-06,
      "loss": 0.80904496,
      "num_input_tokens_seen": 171889455,
      "step": 7994,
      "time_per_iteration": 3.992957830429077
    },
    {
      "auxiliary_loss_clip": 0.01103584,
      "auxiliary_loss_mlp": 0.01033275,
      "balance_loss_clip": 1.04339409,
      "balance_loss_mlp": 1.01938105,
      "epoch": 0.48068540508041485,
      "flos": 13151996640000.0,
      "grad_norm": 1.9498878008857032,
      "language_loss": 0.79708552,
      "learning_rate": 2.2217311357582946e-06,
      "loss": 0.81845409,
      "num_input_tokens_seen": 171906070,
      "step": 7995,
      "time_per_iteration": 2.5124006271362305
    },
    {
      "auxiliary_loss_clip": 0.01077832,
      "auxiliary_loss_mlp": 0.0103577,
      "balance_loss_clip": 1.04481721,
      "balance_loss_mlp": 1.02200699,
      "epoch": 0.4807455283330828,
      "flos": 21176738208000.0,
      "grad_norm": 1.4070564968906234,
      "language_loss": 0.82778364,
      "learning_rate": 2.2213440707461e-06,
      "loss": 0.84891963,
      "num_input_tokens_seen": 171926515,
      "step": 7996,
      "time_per_iteration": 3.989732027053833
    },
    {
      "auxiliary_loss_clip": 0.01055952,
      "auxiliary_loss_mlp": 0.01035714,
      "balance_loss_clip": 1.0447911,
      "balance_loss_mlp": 1.02156997,
      "epoch": 0.4808056515857508,
      "flos": 12275991751680.0,
      "grad_norm": 1.7227742658126144,
      "language_loss": 0.81089115,
      "learning_rate": 2.220956997340516e-06,
      "loss": 0.83180785,
      "num_input_tokens_seen": 171943845,
      "step": 7997,
      "time_per_iteration": 2.635472536087036
    },
    {
      "auxiliary_loss_clip": 0.01074309,
      "auxiliary_loss_mlp": 0.01035421,
      "balance_loss_clip": 1.04094899,
      "balance_loss_mlp": 1.02134848,
      "epoch": 0.48086577483841875,
      "flos": 24826052275200.0,
      "grad_norm": 1.6900927296619095,
      "language_loss": 0.72196531,
      "learning_rate": 2.220569915556221e-06,
      "loss": 0.74306262,
      "num_input_tokens_seen": 171964970,
      "step": 7998,
      "time_per_iteration": 2.6376898288726807
    },
    {
      "auxiliary_loss_clip": 0.01122586,
      "auxiliary_loss_mlp": 0.01037273,
      "balance_loss_clip": 1.04459012,
      "balance_loss_mlp": 1.02318287,
      "epoch": 0.4809258980910867,
      "flos": 24465365856000.0,
      "grad_norm": 1.9081667847185362,
      "language_loss": 0.7096349,
      "learning_rate": 2.220182825407892e-06,
      "loss": 0.73123348,
      "num_input_tokens_seen": 171986340,
      "step": 7999,
      "time_per_iteration": 3.9272279739379883
    },
    {
      "auxiliary_loss_clip": 0.0111552,
      "auxiliary_loss_mlp": 0.01041962,
      "balance_loss_clip": 1.04428041,
      "balance_loss_mlp": 1.02775252,
      "epoch": 0.4809860213437547,
      "flos": 21215952881280.0,
      "grad_norm": 1.6514655308337884,
      "language_loss": 0.71156329,
      "learning_rate": 2.2197957269102083e-06,
      "loss": 0.73313808,
      "num_input_tokens_seen": 172007300,
      "step": 8000,
      "time_per_iteration": 2.525585412979126
    },
    {
      "auxiliary_loss_clip": 0.01112499,
      "auxiliary_loss_mlp": 0.01041162,
      "balance_loss_clip": 1.04567921,
      "balance_loss_mlp": 1.02682757,
      "epoch": 0.48104614459642264,
      "flos": 37632084094080.0,
      "grad_norm": 1.6597023065713772,
      "language_loss": 0.74955338,
      "learning_rate": 2.2194086200778485e-06,
      "loss": 0.77109003,
      "num_input_tokens_seen": 172029585,
      "step": 8001,
      "time_per_iteration": 2.6539931297302246
    },
    {
      "auxiliary_loss_clip": 0.0111224,
      "auxiliary_loss_mlp": 0.01036988,
      "balance_loss_clip": 1.04421353,
      "balance_loss_mlp": 1.02355909,
      "epoch": 0.48110626784909066,
      "flos": 18406122549120.0,
      "grad_norm": 1.685485422686156,
      "language_loss": 0.81354845,
      "learning_rate": 2.219021504925493e-06,
      "loss": 0.83504075,
      "num_input_tokens_seen": 172047495,
      "step": 8002,
      "time_per_iteration": 2.500030994415283
    },
    {
      "auxiliary_loss_clip": 0.01117619,
      "auxiliary_loss_mlp": 0.01037832,
      "balance_loss_clip": 1.04646564,
      "balance_loss_mlp": 1.0231514,
      "epoch": 0.48116639110175863,
      "flos": 28439814856320.0,
      "grad_norm": 2.156423837600068,
      "language_loss": 0.71788907,
      "learning_rate": 2.218634381467819e-06,
      "loss": 0.73944354,
      "num_input_tokens_seen": 172067625,
      "step": 8003,
      "time_per_iteration": 2.559398889541626
    },
    {
      "auxiliary_loss_clip": 0.01106115,
      "auxiliary_loss_mlp": 0.01037714,
      "balance_loss_clip": 1.04408574,
      "balance_loss_mlp": 1.02452326,
      "epoch": 0.4812265143544266,
      "flos": 21725237865600.0,
      "grad_norm": 1.644911104529485,
      "language_loss": 0.82822073,
      "learning_rate": 2.218247249719507e-06,
      "loss": 0.84965897,
      "num_input_tokens_seen": 172087885,
      "step": 8004,
      "time_per_iteration": 2.521289587020874
    },
    {
      "auxiliary_loss_clip": 0.01105942,
      "auxiliary_loss_mlp": 0.01043304,
      "balance_loss_clip": 1.04618657,
      "balance_loss_mlp": 1.02741921,
      "epoch": 0.48128663760709456,
      "flos": 13224679810560.0,
      "grad_norm": 2.1749628497112075,
      "language_loss": 0.77248269,
      "learning_rate": 2.217860109695239e-06,
      "loss": 0.79397511,
      "num_input_tokens_seen": 172105815,
      "step": 8005,
      "time_per_iteration": 2.5096490383148193
    },
    {
      "auxiliary_loss_clip": 0.01106079,
      "auxiliary_loss_mlp": 0.01036391,
      "balance_loss_clip": 1.04356909,
      "balance_loss_mlp": 1.02281928,
      "epoch": 0.4813467608597625,
      "flos": 24243437675520.0,
      "grad_norm": 1.8507650394411095,
      "language_loss": 0.70384717,
      "learning_rate": 2.217472961409692e-06,
      "loss": 0.72527194,
      "num_input_tokens_seen": 172126125,
      "step": 8006,
      "time_per_iteration": 2.528286933898926
    },
    {
      "auxiliary_loss_clip": 0.01094962,
      "auxiliary_loss_mlp": 0.01040489,
      "balance_loss_clip": 1.04188347,
      "balance_loss_mlp": 1.02580249,
      "epoch": 0.4814068841124305,
      "flos": 27480424544640.0,
      "grad_norm": 1.7821661923740897,
      "language_loss": 0.70166117,
      "learning_rate": 2.2170858048775495e-06,
      "loss": 0.72301567,
      "num_input_tokens_seen": 172141945,
      "step": 8007,
      "time_per_iteration": 2.5516135692596436
    },
    {
      "auxiliary_loss_clip": 0.01124542,
      "auxiliary_loss_mlp": 0.01036693,
      "balance_loss_clip": 1.04480386,
      "balance_loss_mlp": 1.02250695,
      "epoch": 0.48146700736509845,
      "flos": 19572896033280.0,
      "grad_norm": 2.2019587951159467,
      "language_loss": 0.71662116,
      "learning_rate": 2.2166986401134914e-06,
      "loss": 0.73823357,
      "num_input_tokens_seen": 172161095,
      "step": 8008,
      "time_per_iteration": 2.4734253883361816
    },
    {
      "auxiliary_loss_clip": 0.01097363,
      "auxiliary_loss_mlp": 0.01042396,
      "balance_loss_clip": 1.04553401,
      "balance_loss_mlp": 1.02735198,
      "epoch": 0.4815271306177664,
      "flos": 20627771673600.0,
      "grad_norm": 2.0053163149361706,
      "language_loss": 0.60602337,
      "learning_rate": 2.216311467132199e-06,
      "loss": 0.62742096,
      "num_input_tokens_seen": 172178750,
      "step": 8009,
      "time_per_iteration": 2.572190523147583
    },
    {
      "auxiliary_loss_clip": 0.01029817,
      "auxiliary_loss_mlp": 0.01006484,
      "balance_loss_clip": 1.02358317,
      "balance_loss_mlp": 1.00511312,
      "epoch": 0.4815872538704344,
      "flos": 67691076232320.0,
      "grad_norm": 0.8631709055799563,
      "language_loss": 0.61366773,
      "learning_rate": 2.2159242859483547e-06,
      "loss": 0.63403082,
      "num_input_tokens_seen": 172240235,
      "step": 8010,
      "time_per_iteration": 3.1639304161071777
    },
    {
      "auxiliary_loss_clip": 0.01116484,
      "auxiliary_loss_mlp": 0.01047555,
      "balance_loss_clip": 1.04741108,
      "balance_loss_mlp": 1.03271973,
      "epoch": 0.48164737712310235,
      "flos": 22820764723200.0,
      "grad_norm": 1.6450113516418963,
      "language_loss": 0.73827612,
      "learning_rate": 2.215537096576639e-06,
      "loss": 0.75991648,
      "num_input_tokens_seen": 172259875,
      "step": 8011,
      "time_per_iteration": 2.5265486240386963
    },
    {
      "auxiliary_loss_clip": 0.01099077,
      "auxiliary_loss_mlp": 0.01036954,
      "balance_loss_clip": 1.04377937,
      "balance_loss_mlp": 1.02381754,
      "epoch": 0.4817075003757703,
      "flos": 23733865382400.0,
      "grad_norm": 1.7998846031039775,
      "language_loss": 0.7934078,
      "learning_rate": 2.2151498990317354e-06,
      "loss": 0.81476814,
      "num_input_tokens_seen": 172280150,
      "step": 8012,
      "time_per_iteration": 2.5880823135375977
    },
    {
      "auxiliary_loss_clip": 0.01089424,
      "auxiliary_loss_mlp": 0.01041885,
      "balance_loss_clip": 1.04727674,
      "balance_loss_mlp": 1.02704906,
      "epoch": 0.4817676236284383,
      "flos": 28182909807360.0,
      "grad_norm": 1.8985004389057338,
      "language_loss": 0.73571336,
      "learning_rate": 2.214762693328326e-06,
      "loss": 0.75702643,
      "num_input_tokens_seen": 172300810,
      "step": 8013,
      "time_per_iteration": 2.6248950958251953
    },
    {
      "auxiliary_loss_clip": 0.01097941,
      "auxiliary_loss_mlp": 0.01028232,
      "balance_loss_clip": 1.04761219,
      "balance_loss_mlp": 1.01520884,
      "epoch": 0.48182774688110624,
      "flos": 17091756080640.0,
      "grad_norm": 1.9895398990617188,
      "language_loss": 0.90716505,
      "learning_rate": 2.214375479481094e-06,
      "loss": 0.92842686,
      "num_input_tokens_seen": 172317930,
      "step": 8014,
      "time_per_iteration": 2.52713680267334
    },
    {
      "auxiliary_loss_clip": 0.01127322,
      "auxiliary_loss_mlp": 0.01038637,
      "balance_loss_clip": 1.04556227,
      "balance_loss_mlp": 1.02394462,
      "epoch": 0.4818878701337742,
      "flos": 12567873669120.0,
      "grad_norm": 2.0312465884024027,
      "language_loss": 0.74541664,
      "learning_rate": 2.213988257504722e-06,
      "loss": 0.76707625,
      "num_input_tokens_seen": 172336340,
      "step": 8015,
      "time_per_iteration": 2.463778018951416
    },
    {
      "auxiliary_loss_clip": 0.01107328,
      "auxiliary_loss_mlp": 0.01038179,
      "balance_loss_clip": 1.04411995,
      "balance_loss_mlp": 1.02359974,
      "epoch": 0.48194799338644223,
      "flos": 24608505553920.0,
      "grad_norm": 1.888295930245216,
      "language_loss": 0.80672365,
      "learning_rate": 2.213601027413894e-06,
      "loss": 0.8281787,
      "num_input_tokens_seen": 172354315,
      "step": 8016,
      "time_per_iteration": 2.5680277347564697
    },
    {
      "auxiliary_loss_clip": 0.0110834,
      "auxiliary_loss_mlp": 0.01032918,
      "balance_loss_clip": 1.04708862,
      "balance_loss_mlp": 1.01945901,
      "epoch": 0.4820081166391102,
      "flos": 21105204272640.0,
      "grad_norm": 1.7199304680982905,
      "language_loss": 0.77431905,
      "learning_rate": 2.2132137892232933e-06,
      "loss": 0.79573166,
      "num_input_tokens_seen": 172372695,
      "step": 8017,
      "time_per_iteration": 2.502817392349243
    },
    {
      "auxiliary_loss_clip": 0.01108557,
      "auxiliary_loss_mlp": 0.01031079,
      "balance_loss_clip": 1.0454185,
      "balance_loss_mlp": 1.01694727,
      "epoch": 0.48206823989177816,
      "flos": 25264593423360.0,
      "grad_norm": 1.8198048844894492,
      "language_loss": 0.80788171,
      "learning_rate": 2.2128265429476043e-06,
      "loss": 0.82927811,
      "num_input_tokens_seen": 172390905,
      "step": 8018,
      "time_per_iteration": 2.5359535217285156
    },
    {
      "auxiliary_loss_clip": 0.01099141,
      "auxiliary_loss_mlp": 0.01031325,
      "balance_loss_clip": 1.04795361,
      "balance_loss_mlp": 1.0181706,
      "epoch": 0.4821283631444461,
      "flos": 24645062620800.0,
      "grad_norm": 1.6358633695959177,
      "language_loss": 0.76281679,
      "learning_rate": 2.2124392886015124e-06,
      "loss": 0.78412151,
      "num_input_tokens_seen": 172412295,
      "step": 8019,
      "time_per_iteration": 2.6041665077209473
    },
    {
      "auxiliary_loss_clip": 0.01088168,
      "auxiliary_loss_mlp": 0.01035981,
      "balance_loss_clip": 1.04522681,
      "balance_loss_mlp": 1.02197397,
      "epoch": 0.4821884863971141,
      "flos": 23952094462080.0,
      "grad_norm": 1.6053213852591102,
      "language_loss": 0.79011923,
      "learning_rate": 2.212052026199701e-06,
      "loss": 0.81136072,
      "num_input_tokens_seen": 172432625,
      "step": 8020,
      "time_per_iteration": 2.6097640991210938
    },
    {
      "auxiliary_loss_clip": 0.01120396,
      "auxiliary_loss_mlp": 0.0103496,
      "balance_loss_clip": 1.04421973,
      "balance_loss_mlp": 1.0212568,
      "epoch": 0.48224860964978206,
      "flos": 17160668323200.0,
      "grad_norm": 1.9303905443822806,
      "language_loss": 0.69620752,
      "learning_rate": 2.211664755756855e-06,
      "loss": 0.7177611,
      "num_input_tokens_seen": 172450010,
      "step": 8021,
      "time_per_iteration": 2.448242664337158
    },
    {
      "auxiliary_loss_clip": 0.01099282,
      "auxiliary_loss_mlp": 0.01033248,
      "balance_loss_clip": 1.04470956,
      "balance_loss_mlp": 1.01832259,
      "epoch": 0.48230873290245,
      "flos": 23075838178560.0,
      "grad_norm": 1.6925421995993188,
      "language_loss": 0.63056177,
      "learning_rate": 2.2112774772876603e-06,
      "loss": 0.65188706,
      "num_input_tokens_seen": 172469080,
      "step": 8022,
      "time_per_iteration": 2.5611445903778076
    },
    {
      "auxiliary_loss_clip": 0.01099057,
      "auxiliary_loss_mlp": 0.0079135,
      "balance_loss_clip": 1.04456615,
      "balance_loss_mlp": 1.01697516,
      "epoch": 0.482368856155118,
      "flos": 19353517718400.0,
      "grad_norm": 2.2580684518982514,
      "language_loss": 0.66276914,
      "learning_rate": 2.2108901908068028e-06,
      "loss": 0.68167317,
      "num_input_tokens_seen": 172484850,
      "step": 8023,
      "time_per_iteration": 2.5222325325012207
    },
    {
      "auxiliary_loss_clip": 0.01049726,
      "auxiliary_loss_mlp": 0.01043639,
      "balance_loss_clip": 1.03846383,
      "balance_loss_mlp": 1.02802277,
      "epoch": 0.48242897940778595,
      "flos": 20078984707200.0,
      "grad_norm": 2.0431999219666577,
      "language_loss": 0.76597983,
      "learning_rate": 2.2105028963289683e-06,
      "loss": 0.78691351,
      "num_input_tokens_seen": 172503525,
      "step": 8024,
      "time_per_iteration": 2.7242565155029297
    },
    {
      "auxiliary_loss_clip": 0.01098628,
      "auxiliary_loss_mlp": 0.0103598,
      "balance_loss_clip": 1.04198337,
      "balance_loss_mlp": 1.02214587,
      "epoch": 0.4824891026604539,
      "flos": 23403989854080.0,
      "grad_norm": 1.5314800273362705,
      "language_loss": 0.75274265,
      "learning_rate": 2.2101155938688423e-06,
      "loss": 0.77408868,
      "num_input_tokens_seen": 172524360,
      "step": 8025,
      "time_per_iteration": 2.5508241653442383
    },
    {
      "auxiliary_loss_clip": 0.01120816,
      "auxiliary_loss_mlp": 0.01032193,
      "balance_loss_clip": 1.04406047,
      "balance_loss_mlp": 1.01915145,
      "epoch": 0.4825492259131219,
      "flos": 20368675895040.0,
      "grad_norm": 2.208461433576477,
      "language_loss": 0.71061063,
      "learning_rate": 2.209728283441112e-06,
      "loss": 0.73214066,
      "num_input_tokens_seen": 172541480,
      "step": 8026,
      "time_per_iteration": 2.4596126079559326
    },
    {
      "auxiliary_loss_clip": 0.01111994,
      "auxiliary_loss_mlp": 0.01042607,
      "balance_loss_clip": 1.04457629,
      "balance_loss_mlp": 1.02685928,
      "epoch": 0.48260934916578985,
      "flos": 14319021519360.0,
      "grad_norm": 2.2022314326436185,
      "language_loss": 0.74541032,
      "learning_rate": 2.209340965060465e-06,
      "loss": 0.76695633,
      "num_input_tokens_seen": 172559005,
      "step": 8027,
      "time_per_iteration": 2.497581720352173
    },
    {
      "auxiliary_loss_clip": 0.01097733,
      "auxiliary_loss_mlp": 0.01035645,
      "balance_loss_clip": 1.04534197,
      "balance_loss_mlp": 1.02230549,
      "epoch": 0.4826694724184578,
      "flos": 22121152548480.0,
      "grad_norm": 1.633610579920971,
      "language_loss": 0.67147934,
      "learning_rate": 2.2089536387415868e-06,
      "loss": 0.6928131,
      "num_input_tokens_seen": 172578435,
      "step": 8028,
      "time_per_iteration": 2.5563721656799316
    },
    {
      "auxiliary_loss_clip": 0.01100639,
      "auxiliary_loss_mlp": 0.01036325,
      "balance_loss_clip": 1.04374409,
      "balance_loss_mlp": 1.02265167,
      "epoch": 0.48272959567112583,
      "flos": 16181169373440.0,
      "grad_norm": 1.664348241645772,
      "language_loss": 0.72752446,
      "learning_rate": 2.2085663044991655e-06,
      "loss": 0.74889404,
      "num_input_tokens_seen": 172596095,
      "step": 8029,
      "time_per_iteration": 2.496027708053589
    },
    {
      "auxiliary_loss_clip": 0.01099044,
      "auxiliary_loss_mlp": 0.01032606,
      "balance_loss_clip": 1.04616046,
      "balance_loss_mlp": 1.01805091,
      "epoch": 0.4827897189237938,
      "flos": 23180445561600.0,
      "grad_norm": 2.2927232451485566,
      "language_loss": 0.84567171,
      "learning_rate": 2.2081789623478896e-06,
      "loss": 0.86698824,
      "num_input_tokens_seen": 172615255,
      "step": 8030,
      "time_per_iteration": 2.5523974895477295
    },
    {
      "auxiliary_loss_clip": 0.01092082,
      "auxiliary_loss_mlp": 0.01035135,
      "balance_loss_clip": 1.04350376,
      "balance_loss_mlp": 1.02149773,
      "epoch": 0.48284984217646176,
      "flos": 21652626522240.0,
      "grad_norm": 2.373108753315014,
      "language_loss": 0.74026269,
      "learning_rate": 2.2077916123024466e-06,
      "loss": 0.76153487,
      "num_input_tokens_seen": 172633185,
      "step": 8031,
      "time_per_iteration": 3.9009711742401123
    },
    {
      "auxiliary_loss_clip": 0.01094642,
      "auxiliary_loss_mlp": 0.0104698,
      "balance_loss_clip": 1.04175663,
      "balance_loss_mlp": 1.03128588,
      "epoch": 0.48290996542912973,
      "flos": 31467443304960.0,
      "grad_norm": 1.6482639868287778,
      "language_loss": 0.71899223,
      "learning_rate": 2.2074042543775245e-06,
      "loss": 0.74040842,
      "num_input_tokens_seen": 172654280,
      "step": 8032,
      "time_per_iteration": 4.063181638717651
    },
    {
      "auxiliary_loss_clip": 0.01103946,
      "auxiliary_loss_mlp": 0.01042394,
      "balance_loss_clip": 1.03987479,
      "balance_loss_mlp": 1.02739704,
      "epoch": 0.4829700886817977,
      "flos": 24461954064000.0,
      "grad_norm": 1.6089281863100537,
      "language_loss": 0.74130857,
      "learning_rate": 2.2070168885878126e-06,
      "loss": 0.76277196,
      "num_input_tokens_seen": 172675545,
      "step": 8033,
      "time_per_iteration": 2.528202772140503
    },
    {
      "auxiliary_loss_clip": 0.01068944,
      "auxiliary_loss_mlp": 0.01034933,
      "balance_loss_clip": 1.04533207,
      "balance_loss_mlp": 1.02115297,
      "epoch": 0.48303021193446566,
      "flos": 25702164904320.0,
      "grad_norm": 1.4784240599143876,
      "language_loss": 0.8342582,
      "learning_rate": 2.2066295149479996e-06,
      "loss": 0.85529703,
      "num_input_tokens_seen": 172696455,
      "step": 8034,
      "time_per_iteration": 2.6595723628997803
    },
    {
      "auxiliary_loss_clip": 0.01085981,
      "auxiliary_loss_mlp": 0.01030286,
      "balance_loss_clip": 1.04177272,
      "balance_loss_mlp": 1.01750088,
      "epoch": 0.4830903351871336,
      "flos": 20085233673600.0,
      "grad_norm": 1.6495252519259012,
      "language_loss": 0.79664314,
      "learning_rate": 2.2062421334727744e-06,
      "loss": 0.81780583,
      "num_input_tokens_seen": 172716720,
      "step": 8035,
      "time_per_iteration": 4.010541200637817
    },
    {
      "auxiliary_loss_clip": 0.01097017,
      "auxiliary_loss_mlp": 0.00790615,
      "balance_loss_clip": 1.04088974,
      "balance_loss_mlp": 1.01177144,
      "epoch": 0.4831504584398016,
      "flos": 39452216014080.0,
      "grad_norm": 2.455228954952055,
      "language_loss": 0.69607526,
      "learning_rate": 2.2058547441768267e-06,
      "loss": 0.71495152,
      "num_input_tokens_seen": 172737435,
      "step": 8036,
      "time_per_iteration": 2.6928060054779053
    },
    {
      "auxiliary_loss_clip": 0.01106709,
      "auxiliary_loss_mlp": 0.01033809,
      "balance_loss_clip": 1.04055691,
      "balance_loss_mlp": 1.02021861,
      "epoch": 0.48321058169246955,
      "flos": 20006588845440.0,
      "grad_norm": 1.7806490882050996,
      "language_loss": 0.72682321,
      "learning_rate": 2.205467347074847e-06,
      "loss": 0.74822843,
      "num_input_tokens_seen": 172755700,
      "step": 8037,
      "time_per_iteration": 2.4832327365875244
    },
    {
      "auxiliary_loss_clip": 0.01075281,
      "auxiliary_loss_mlp": 0.01039803,
      "balance_loss_clip": 1.04387951,
      "balance_loss_mlp": 1.02339935,
      "epoch": 0.4832707049451375,
      "flos": 20741465197440.0,
      "grad_norm": 2.0360338876128137,
      "language_loss": 0.69039655,
      "learning_rate": 2.205079942181525e-06,
      "loss": 0.71154743,
      "num_input_tokens_seen": 172775185,
      "step": 8038,
      "time_per_iteration": 3.977391481399536
    },
    {
      "auxiliary_loss_clip": 0.01085256,
      "auxiliary_loss_mlp": 0.01035087,
      "balance_loss_clip": 1.04075265,
      "balance_loss_mlp": 1.02085328,
      "epoch": 0.4833308281978055,
      "flos": 33145584762240.0,
      "grad_norm": 1.4612275800319958,
      "language_loss": 0.79194593,
      "learning_rate": 2.20469252951155e-06,
      "loss": 0.81314933,
      "num_input_tokens_seen": 172796990,
      "step": 8039,
      "time_per_iteration": 2.6699161529541016
    },
    {
      "auxiliary_loss_clip": 0.01108263,
      "auxiliary_loss_mlp": 0.01029922,
      "balance_loss_clip": 1.04388165,
      "balance_loss_mlp": 1.01682734,
      "epoch": 0.48339095145047345,
      "flos": 19099234362240.0,
      "grad_norm": 1.7034153962295335,
      "language_loss": 0.77699292,
      "learning_rate": 2.2043051090796143e-06,
      "loss": 0.79837477,
      "num_input_tokens_seen": 172814915,
      "step": 8040,
      "time_per_iteration": 2.5116214752197266
    },
    {
      "auxiliary_loss_clip": 0.01107793,
      "auxiliary_loss_mlp": 0.01039403,
      "balance_loss_clip": 1.04019368,
      "balance_loss_mlp": 1.02460909,
      "epoch": 0.4834510747031414,
      "flos": 34459448440320.0,
      "grad_norm": 1.8037629274348885,
      "language_loss": 0.75449324,
      "learning_rate": 2.203917680900409e-06,
      "loss": 0.77596521,
      "num_input_tokens_seen": 172837060,
      "step": 8041,
      "time_per_iteration": 2.6189022064208984
    },
    {
      "auxiliary_loss_clip": 0.01085498,
      "auxiliary_loss_mlp": 0.01037252,
      "balance_loss_clip": 1.04266047,
      "balance_loss_mlp": 1.02299464,
      "epoch": 0.48351119795580944,
      "flos": 27380845065600.0,
      "grad_norm": 1.9666504543158554,
      "language_loss": 0.66742814,
      "learning_rate": 2.203530244988624e-06,
      "loss": 0.68865561,
      "num_input_tokens_seen": 172856545,
      "step": 8042,
      "time_per_iteration": 2.6303889751434326
    },
    {
      "auxiliary_loss_clip": 0.0103154,
      "auxiliary_loss_mlp": 0.01006055,
      "balance_loss_clip": 1.02686262,
      "balance_loss_mlp": 1.00463009,
      "epoch": 0.4835713212084774,
      "flos": 67143941291520.0,
      "grad_norm": 0.691919962628056,
      "language_loss": 0.58555067,
      "learning_rate": 2.2031428013589517e-06,
      "loss": 0.60592663,
      "num_input_tokens_seen": 172923055,
      "step": 8043,
      "time_per_iteration": 3.218369960784912
    },
    {
      "auxiliary_loss_clip": 0.01098506,
      "auxiliary_loss_mlp": 0.01041109,
      "balance_loss_clip": 1.04154694,
      "balance_loss_mlp": 1.02484894,
      "epoch": 0.48363144446114537,
      "flos": 17967473660160.0,
      "grad_norm": 2.091096202790917,
      "language_loss": 0.71887302,
      "learning_rate": 2.2027553500260847e-06,
      "loss": 0.74026918,
      "num_input_tokens_seen": 172940700,
      "step": 8044,
      "time_per_iteration": 2.4962875843048096
    },
    {
      "auxiliary_loss_clip": 0.01079079,
      "auxiliary_loss_mlp": 0.01036219,
      "balance_loss_clip": 1.04973173,
      "balance_loss_mlp": 1.02091289,
      "epoch": 0.48369156771381333,
      "flos": 20593513077120.0,
      "grad_norm": 1.2904734858580333,
      "language_loss": 0.76047087,
      "learning_rate": 2.202367891004714e-06,
      "loss": 0.78162384,
      "num_input_tokens_seen": 172961125,
      "step": 8045,
      "time_per_iteration": 2.6222071647644043
    },
    {
      "auxiliary_loss_clip": 0.01075026,
      "auxiliary_loss_mlp": 0.01034437,
      "balance_loss_clip": 1.04565001,
      "balance_loss_mlp": 1.0208348,
      "epoch": 0.4837516909664813,
      "flos": 22675075159680.0,
      "grad_norm": 1.820455420435124,
      "language_loss": 0.69157314,
      "learning_rate": 2.201980424309533e-06,
      "loss": 0.71266776,
      "num_input_tokens_seen": 172980405,
      "step": 8046,
      "time_per_iteration": 2.613757610321045
    },
    {
      "auxiliary_loss_clip": 0.01120903,
      "auxiliary_loss_mlp": 0.01034789,
      "balance_loss_clip": 1.04292965,
      "balance_loss_mlp": 1.02037692,
      "epoch": 0.48381181421914926,
      "flos": 25518625384320.0,
      "grad_norm": 1.854325052014233,
      "language_loss": 0.82130551,
      "learning_rate": 2.2015929499552337e-06,
      "loss": 0.84286237,
      "num_input_tokens_seen": 172999105,
      "step": 8047,
      "time_per_iteration": 2.5068960189819336
    },
    {
      "auxiliary_loss_clip": 0.010895,
      "auxiliary_loss_mlp": 0.01036576,
      "balance_loss_clip": 1.03982663,
      "balance_loss_mlp": 1.02305186,
      "epoch": 0.4838719374718172,
      "flos": 24207491139840.0,
      "grad_norm": 1.9411473043732483,
      "language_loss": 0.80556458,
      "learning_rate": 2.2012054679565092e-06,
      "loss": 0.82682538,
      "num_input_tokens_seen": 173019935,
      "step": 8048,
      "time_per_iteration": 2.5656442642211914
    },
    {
      "auxiliary_loss_clip": 0.01112461,
      "auxiliary_loss_mlp": 0.01035908,
      "balance_loss_clip": 1.04302192,
      "balance_loss_mlp": 1.0221808,
      "epoch": 0.4839320607244852,
      "flos": 26724577628160.0,
      "grad_norm": 1.735927416868701,
      "language_loss": 0.81283885,
      "learning_rate": 2.200817978328054e-06,
      "loss": 0.83432257,
      "num_input_tokens_seen": 173039700,
      "step": 8049,
      "time_per_iteration": 2.5722410678863525
    },
    {
      "auxiliary_loss_clip": 0.01101002,
      "auxiliary_loss_mlp": 0.01031981,
      "balance_loss_clip": 1.04440165,
      "balance_loss_mlp": 1.01932716,
      "epoch": 0.48399218397715316,
      "flos": 20448900921600.0,
      "grad_norm": 1.624811572387827,
      "language_loss": 0.72720158,
      "learning_rate": 2.2004304810845602e-06,
      "loss": 0.74853146,
      "num_input_tokens_seen": 173059170,
      "step": 8050,
      "time_per_iteration": 2.5246028900146484
    },
    {
      "auxiliary_loss_clip": 0.01038378,
      "auxiliary_loss_mlp": 0.0077353,
      "balance_loss_clip": 1.02284312,
      "balance_loss_mlp": 1.01269627,
      "epoch": 0.4840523072298211,
      "flos": 67180570185600.0,
      "grad_norm": 0.6958646383943926,
      "language_loss": 0.56389439,
      "learning_rate": 2.200042976240723e-06,
      "loss": 0.58201349,
      "num_input_tokens_seen": 173119000,
      "step": 8051,
      "time_per_iteration": 3.1603474617004395
    },
    {
      "auxiliary_loss_clip": 0.0108534,
      "auxiliary_loss_mlp": 0.01032236,
      "balance_loss_clip": 1.04448593,
      "balance_loss_mlp": 1.01821089,
      "epoch": 0.4841124304824891,
      "flos": 22411490181120.0,
      "grad_norm": 2.72459774369678,
      "language_loss": 0.7515325,
      "learning_rate": 2.199655463811236e-06,
      "loss": 0.77270824,
      "num_input_tokens_seen": 173137570,
      "step": 8052,
      "time_per_iteration": 2.5679843425750732
    },
    {
      "auxiliary_loss_clip": 0.01107296,
      "auxiliary_loss_mlp": 0.01029724,
      "balance_loss_clip": 1.04607463,
      "balance_loss_mlp": 1.01681352,
      "epoch": 0.48417255373515705,
      "flos": 13843959217920.0,
      "grad_norm": 2.087045168635755,
      "language_loss": 0.66287655,
      "learning_rate": 2.1992679438107936e-06,
      "loss": 0.68424678,
      "num_input_tokens_seen": 173154355,
      "step": 8053,
      "time_per_iteration": 2.5085248947143555
    },
    {
      "auxiliary_loss_clip": 0.0110683,
      "auxiliary_loss_mlp": 0.01030551,
      "balance_loss_clip": 1.04169178,
      "balance_loss_mlp": 1.01741385,
      "epoch": 0.484232676987825,
      "flos": 31649689935360.0,
      "grad_norm": 1.931049164115971,
      "language_loss": 0.69593501,
      "learning_rate": 2.198880416254091e-06,
      "loss": 0.71730882,
      "num_input_tokens_seen": 173174845,
      "step": 8054,
      "time_per_iteration": 2.5860259532928467
    },
    {
      "auxiliary_loss_clip": 0.01058627,
      "auxiliary_loss_mlp": 0.01032984,
      "balance_loss_clip": 1.03959489,
      "balance_loss_mlp": 1.01935267,
      "epoch": 0.48429280024049304,
      "flos": 24095377814400.0,
      "grad_norm": 1.6836944741831343,
      "language_loss": 0.69572872,
      "learning_rate": 2.1984928811558233e-06,
      "loss": 0.71664488,
      "num_input_tokens_seen": 173195025,
      "step": 8055,
      "time_per_iteration": 2.7091689109802246
    },
    {
      "auxiliary_loss_clip": 0.01113528,
      "auxiliary_loss_mlp": 0.01037876,
      "balance_loss_clip": 1.04603899,
      "balance_loss_mlp": 1.0238328,
      "epoch": 0.484352923493161,
      "flos": 17530081747200.0,
      "grad_norm": 2.3204561315998107,
      "language_loss": 0.63325429,
      "learning_rate": 2.198105338530685e-06,
      "loss": 0.65476829,
      "num_input_tokens_seen": 173213065,
      "step": 8056,
      "time_per_iteration": 2.4866795539855957
    },
    {
      "auxiliary_loss_clip": 0.01107602,
      "auxiliary_loss_mlp": 0.01030609,
      "balance_loss_clip": 1.04094923,
      "balance_loss_mlp": 1.01636326,
      "epoch": 0.48441304674582897,
      "flos": 29166862043520.0,
      "grad_norm": 1.928495605151099,
      "language_loss": 0.67061591,
      "learning_rate": 2.1977177883933726e-06,
      "loss": 0.69199806,
      "num_input_tokens_seen": 173234545,
      "step": 8057,
      "time_per_iteration": 2.5925471782684326
    },
    {
      "auxiliary_loss_clip": 0.0108087,
      "auxiliary_loss_mlp": 0.01039968,
      "balance_loss_clip": 1.03869903,
      "balance_loss_mlp": 1.02450657,
      "epoch": 0.48447316999849693,
      "flos": 15886701676800.0,
      "grad_norm": 1.5810196737291642,
      "language_loss": 0.81700504,
      "learning_rate": 2.1973302307585827e-06,
      "loss": 0.83821338,
      "num_input_tokens_seen": 173252175,
      "step": 8058,
      "time_per_iteration": 2.546215534210205
    },
    {
      "auxiliary_loss_clip": 0.01108649,
      "auxiliary_loss_mlp": 0.01036727,
      "balance_loss_clip": 1.04285967,
      "balance_loss_mlp": 1.02248764,
      "epoch": 0.4845332932511649,
      "flos": 24381405815040.0,
      "grad_norm": 1.5883840125623667,
      "language_loss": 0.79572439,
      "learning_rate": 2.1969426656410097e-06,
      "loss": 0.81717813,
      "num_input_tokens_seen": 173268790,
      "step": 8059,
      "time_per_iteration": 2.5413990020751953
    },
    {
      "auxiliary_loss_clip": 0.01126615,
      "auxiliary_loss_mlp": 0.01039704,
      "balance_loss_clip": 1.04565454,
      "balance_loss_mlp": 1.02492261,
      "epoch": 0.48459341650383286,
      "flos": 37116478316160.0,
      "grad_norm": 2.072628164023448,
      "language_loss": 0.66365272,
      "learning_rate": 2.196555093055352e-06,
      "loss": 0.68531591,
      "num_input_tokens_seen": 173288030,
      "step": 8060,
      "time_per_iteration": 2.6061532497406006
    },
    {
      "auxiliary_loss_clip": 0.01109042,
      "auxiliary_loss_mlp": 0.01037398,
      "balance_loss_clip": 1.04587615,
      "balance_loss_mlp": 1.02321172,
      "epoch": 0.48465353975650083,
      "flos": 22966777509120.0,
      "grad_norm": 1.7953507557679207,
      "language_loss": 0.67175353,
      "learning_rate": 2.1961675130163046e-06,
      "loss": 0.69321799,
      "num_input_tokens_seen": 173305965,
      "step": 8061,
      "time_per_iteration": 2.538194417953491
    },
    {
      "auxiliary_loss_clip": 0.01104356,
      "auxiliary_loss_mlp": 0.01040843,
      "balance_loss_clip": 1.04501653,
      "balance_loss_mlp": 1.02531016,
      "epoch": 0.4847136630091688,
      "flos": 17707695523200.0,
      "grad_norm": 1.9579794918130407,
      "language_loss": 0.82165825,
      "learning_rate": 2.1957799255385653e-06,
      "loss": 0.8431102,
      "num_input_tokens_seen": 173321985,
      "step": 8062,
      "time_per_iteration": 2.5257492065429688
    },
    {
      "auxiliary_loss_clip": 0.0106197,
      "auxiliary_loss_mlp": 0.0103648,
      "balance_loss_clip": 1.04705429,
      "balance_loss_mlp": 1.0229857,
      "epoch": 0.48477378626183676,
      "flos": 22018269018240.0,
      "grad_norm": 1.5645273839193174,
      "language_loss": 0.74499762,
      "learning_rate": 2.1953923306368325e-06,
      "loss": 0.76598209,
      "num_input_tokens_seen": 173341315,
      "step": 8063,
      "time_per_iteration": 2.65763783454895
    },
    {
      "auxiliary_loss_clip": 0.01098927,
      "auxiliary_loss_mlp": 0.01033278,
      "balance_loss_clip": 1.04450142,
      "balance_loss_mlp": 1.01922882,
      "epoch": 0.4848339095145047,
      "flos": 27962956874880.0,
      "grad_norm": 1.6381294335013055,
      "language_loss": 0.79016471,
      "learning_rate": 2.1950047283258023e-06,
      "loss": 0.81148672,
      "num_input_tokens_seen": 173361055,
      "step": 8064,
      "time_per_iteration": 2.6216468811035156
    },
    {
      "auxiliary_loss_clip": 0.01117728,
      "auxiliary_loss_mlp": 0.00791995,
      "balance_loss_clip": 1.04405642,
      "balance_loss_mlp": 1.02030241,
      "epoch": 0.4848940327671727,
      "flos": 21688752625920.0,
      "grad_norm": 1.7414379390169912,
      "language_loss": 0.79106331,
      "learning_rate": 2.194617118620173e-06,
      "loss": 0.81016052,
      "num_input_tokens_seen": 173379255,
      "step": 8065,
      "time_per_iteration": 2.495110511779785
    },
    {
      "auxiliary_loss_clip": 0.0110174,
      "auxiliary_loss_mlp": 0.00789491,
      "balance_loss_clip": 1.04047585,
      "balance_loss_mlp": 1.01603293,
      "epoch": 0.48495415601984065,
      "flos": 20631578515200.0,
      "grad_norm": 1.536726861859149,
      "language_loss": 0.75938404,
      "learning_rate": 2.194229501534644e-06,
      "loss": 0.77829635,
      "num_input_tokens_seen": 173398370,
      "step": 8066,
      "time_per_iteration": 2.5116422176361084
    },
    {
      "auxiliary_loss_clip": 0.01119907,
      "auxiliary_loss_mlp": 0.01033494,
      "balance_loss_clip": 1.0449667,
      "balance_loss_mlp": 1.02063131,
      "epoch": 0.4850142792725086,
      "flos": 25628152930560.0,
      "grad_norm": 1.349258332726653,
      "language_loss": 0.72234511,
      "learning_rate": 2.193841877083912e-06,
      "loss": 0.74387914,
      "num_input_tokens_seen": 173419595,
      "step": 8067,
      "time_per_iteration": 2.5332400798797607
    },
    {
      "auxiliary_loss_clip": 0.0105807,
      "auxiliary_loss_mlp": 0.01032236,
      "balance_loss_clip": 1.05013061,
      "balance_loss_mlp": 1.01884842,
      "epoch": 0.4850744025251766,
      "flos": 13771958405760.0,
      "grad_norm": 2.0045816852838754,
      "language_loss": 0.78928649,
      "learning_rate": 2.1934542452826767e-06,
      "loss": 0.81018949,
      "num_input_tokens_seen": 173435390,
      "step": 8068,
      "time_per_iteration": 2.6218581199645996
    },
    {
      "auxiliary_loss_clip": 0.01091965,
      "auxiliary_loss_mlp": 0.01032422,
      "balance_loss_clip": 1.04138136,
      "balance_loss_mlp": 1.02005458,
      "epoch": 0.4851345257778446,
      "flos": 20261339078400.0,
      "grad_norm": 1.4168946530767803,
      "language_loss": 0.84502017,
      "learning_rate": 2.193066606145638e-06,
      "loss": 0.86626399,
      "num_input_tokens_seen": 173454095,
      "step": 8069,
      "time_per_iteration": 3.9706478118896484
    },
    {
      "auxiliary_loss_clip": 0.01087441,
      "auxiliary_loss_mlp": 0.01030629,
      "balance_loss_clip": 1.04649746,
      "balance_loss_mlp": 1.01798725,
      "epoch": 0.48519464903051257,
      "flos": 27089681420160.0,
      "grad_norm": 1.7905385944213423,
      "language_loss": 0.78190672,
      "learning_rate": 2.192678959687493e-06,
      "loss": 0.80308747,
      "num_input_tokens_seen": 173475300,
      "step": 8070,
      "time_per_iteration": 2.607417345046997
    },
    {
      "auxiliary_loss_clip": 0.01062934,
      "auxiliary_loss_mlp": 0.01031956,
      "balance_loss_clip": 1.04338765,
      "balance_loss_mlp": 1.01821125,
      "epoch": 0.48525477228318054,
      "flos": 17127235739520.0,
      "grad_norm": 1.9883996317401373,
      "language_loss": 0.77856141,
      "learning_rate": 2.192291305922943e-06,
      "loss": 0.7995103,
      "num_input_tokens_seen": 173492005,
      "step": 8071,
      "time_per_iteration": 3.954204559326172
    },
    {
      "auxiliary_loss_clip": 0.01062736,
      "auxiliary_loss_mlp": 0.01031877,
      "balance_loss_clip": 1.0433073,
      "balance_loss_mlp": 1.01823306,
      "epoch": 0.4853148955358485,
      "flos": 28180324028160.0,
      "grad_norm": 2.047495110874624,
      "language_loss": 0.71805716,
      "learning_rate": 2.1919036448666873e-06,
      "loss": 0.73900324,
      "num_input_tokens_seen": 173511995,
      "step": 8072,
      "time_per_iteration": 2.6705756187438965
    },
    {
      "auxiliary_loss_clip": 0.01083582,
      "auxiliary_loss_mlp": 0.01037228,
      "balance_loss_clip": 1.04679,
      "balance_loss_mlp": 1.0232383,
      "epoch": 0.48537501878851647,
      "flos": 17493309198720.0,
      "grad_norm": 2.9145693944352784,
      "language_loss": 0.87731016,
      "learning_rate": 2.1915159765334262e-06,
      "loss": 0.8985182,
      "num_input_tokens_seen": 173530215,
      "step": 8073,
      "time_per_iteration": 2.5898513793945312
    },
    {
      "auxiliary_loss_clip": 0.01080516,
      "auxiliary_loss_mlp": 0.01037457,
      "balance_loss_clip": 1.04322577,
      "balance_loss_mlp": 1.02283621,
      "epoch": 0.48543514204118443,
      "flos": 28584857975040.0,
      "grad_norm": 1.741053796057841,
      "language_loss": 0.60822344,
      "learning_rate": 2.19112830093786e-06,
      "loss": 0.62940317,
      "num_input_tokens_seen": 173550920,
      "step": 8074,
      "time_per_iteration": 4.005762815475464
    },
    {
      "auxiliary_loss_clip": 0.01084992,
      "auxiliary_loss_mlp": 0.0078769,
      "balance_loss_clip": 1.04244399,
      "balance_loss_mlp": 1.00885046,
      "epoch": 0.4854952652938524,
      "flos": 20959981585920.0,
      "grad_norm": 1.5644213884341898,
      "language_loss": 0.7304672,
      "learning_rate": 2.19074061809469e-06,
      "loss": 0.74919403,
      "num_input_tokens_seen": 173569065,
      "step": 8075,
      "time_per_iteration": 2.5772905349731445
    },
    {
      "auxiliary_loss_clip": 0.01115605,
      "auxiliary_loss_mlp": 0.01032692,
      "balance_loss_clip": 1.04364038,
      "balance_loss_mlp": 1.02027035,
      "epoch": 0.48555538854652036,
      "flos": 66529543155840.0,
      "grad_norm": 1.513104823085422,
      "language_loss": 0.81688601,
      "learning_rate": 2.1903529280186163e-06,
      "loss": 0.83836901,
      "num_input_tokens_seen": 173596085,
      "step": 8076,
      "time_per_iteration": 4.3069748878479
    },
    {
      "auxiliary_loss_clip": 0.01100896,
      "auxiliary_loss_mlp": 0.01033121,
      "balance_loss_clip": 1.04442561,
      "balance_loss_mlp": 1.01799345,
      "epoch": 0.4856155117991883,
      "flos": 15924982596480.0,
      "grad_norm": 1.7976956511031954,
      "language_loss": 0.86342371,
      "learning_rate": 2.1899652307243407e-06,
      "loss": 0.88476384,
      "num_input_tokens_seen": 173613900,
      "step": 8077,
      "time_per_iteration": 2.5475127696990967
    },
    {
      "auxiliary_loss_clip": 0.0102125,
      "auxiliary_loss_mlp": 0.01001125,
      "balance_loss_clip": 1.0251441,
      "balance_loss_mlp": 0.99969465,
      "epoch": 0.4856756350518563,
      "flos": 71047395060480.0,
      "grad_norm": 0.9110556763079507,
      "language_loss": 0.58458692,
      "learning_rate": 2.189577526226564e-06,
      "loss": 0.60481066,
      "num_input_tokens_seen": 173671305,
      "step": 8078,
      "time_per_iteration": 3.1366753578186035
    },
    {
      "auxiliary_loss_clip": 0.01123985,
      "auxiliary_loss_mlp": 0.01031958,
      "balance_loss_clip": 1.04551184,
      "balance_loss_mlp": 1.01918459,
      "epoch": 0.48573575830452426,
      "flos": 29825679346560.0,
      "grad_norm": 1.5690117770070544,
      "language_loss": 0.72231007,
      "learning_rate": 2.1891898145399884e-06,
      "loss": 0.74386948,
      "num_input_tokens_seen": 173692070,
      "step": 8079,
      "time_per_iteration": 2.6014459133148193
    },
    {
      "auxiliary_loss_clip": 0.01085168,
      "auxiliary_loss_mlp": 0.01035042,
      "balance_loss_clip": 1.04521048,
      "balance_loss_mlp": 1.02164292,
      "epoch": 0.4857958815571922,
      "flos": 17639501552640.0,
      "grad_norm": 3.0704592043172254,
      "language_loss": 0.79344553,
      "learning_rate": 2.1888020956793172e-06,
      "loss": 0.81464761,
      "num_input_tokens_seen": 173709785,
      "step": 8080,
      "time_per_iteration": 2.55288028717041
    },
    {
      "auxiliary_loss_clip": 0.01096101,
      "auxiliary_loss_mlp": 0.01035649,
      "balance_loss_clip": 1.04326391,
      "balance_loss_mlp": 1.02251792,
      "epoch": 0.4858560048098602,
      "flos": 21105491581440.0,
      "grad_norm": 1.930115923571366,
      "language_loss": 0.83935159,
      "learning_rate": 2.188414369659251e-06,
      "loss": 0.86066908,
      "num_input_tokens_seen": 173728770,
      "step": 8081,
      "time_per_iteration": 2.5439059734344482
    },
    {
      "auxiliary_loss_clip": 0.01106556,
      "auxiliary_loss_mlp": 0.0103689,
      "balance_loss_clip": 1.04207873,
      "balance_loss_mlp": 1.02217984,
      "epoch": 0.4859161280625282,
      "flos": 22090844448000.0,
      "grad_norm": 1.6142836334549737,
      "language_loss": 0.83102417,
      "learning_rate": 2.1880266364944924e-06,
      "loss": 0.85245866,
      "num_input_tokens_seen": 173747355,
      "step": 8082,
      "time_per_iteration": 2.5086960792541504
    },
    {
      "auxiliary_loss_clip": 0.01099711,
      "auxiliary_loss_mlp": 0.01036217,
      "balance_loss_clip": 1.04653335,
      "balance_loss_mlp": 1.02353919,
      "epoch": 0.4859762513151962,
      "flos": 17493452853120.0,
      "grad_norm": 1.9256235275155948,
      "language_loss": 0.87131095,
      "learning_rate": 2.187638896199746e-06,
      "loss": 0.89267021,
      "num_input_tokens_seen": 173764825,
      "step": 8083,
      "time_per_iteration": 2.5098512172698975
    },
    {
      "auxiliary_loss_clip": 0.01073338,
      "auxiliary_loss_mlp": 0.01040413,
      "balance_loss_clip": 1.04475081,
      "balance_loss_mlp": 1.0281229,
      "epoch": 0.48603637456786414,
      "flos": 18004246208640.0,
      "grad_norm": 1.5835787809866984,
      "language_loss": 0.80791634,
      "learning_rate": 2.1872511487897126e-06,
      "loss": 0.82905388,
      "num_input_tokens_seen": 173783215,
      "step": 8084,
      "time_per_iteration": 2.5649662017822266
    },
    {
      "auxiliary_loss_clip": 0.01109753,
      "auxiliary_loss_mlp": 0.01037947,
      "balance_loss_clip": 1.04381752,
      "balance_loss_mlp": 1.02455926,
      "epoch": 0.4860964978205321,
      "flos": 22492038430080.0,
      "grad_norm": 2.262186526282106,
      "language_loss": 0.68337119,
      "learning_rate": 2.186863394279098e-06,
      "loss": 0.70484823,
      "num_input_tokens_seen": 173801905,
      "step": 8085,
      "time_per_iteration": 2.524003505706787
    },
    {
      "auxiliary_loss_clip": 0.01109734,
      "auxiliary_loss_mlp": 0.01035618,
      "balance_loss_clip": 1.04473948,
      "balance_loss_mlp": 1.02256417,
      "epoch": 0.48615662107320007,
      "flos": 23372532518400.0,
      "grad_norm": 1.5424836260046138,
      "language_loss": 0.77530861,
      "learning_rate": 2.1864756326826046e-06,
      "loss": 0.79676211,
      "num_input_tokens_seen": 173824690,
      "step": 8086,
      "time_per_iteration": 2.5818378925323486
    },
    {
      "auxiliary_loss_clip": 0.01119708,
      "auxiliary_loss_mlp": 0.01029038,
      "balance_loss_clip": 1.04351377,
      "balance_loss_mlp": 1.01591873,
      "epoch": 0.48621674432586803,
      "flos": 34418833136640.0,
      "grad_norm": 2.1830885465321646,
      "language_loss": 0.69795281,
      "learning_rate": 2.1860878640149355e-06,
      "loss": 0.71944022,
      "num_input_tokens_seen": 173844450,
      "step": 8087,
      "time_per_iteration": 2.593947649002075
    },
    {
      "auxiliary_loss_clip": 0.01112712,
      "auxiliary_loss_mlp": 0.0103913,
      "balance_loss_clip": 1.04513669,
      "balance_loss_mlp": 1.02510524,
      "epoch": 0.486276867578536,
      "flos": 33107555237760.0,
      "grad_norm": 1.7765760825699368,
      "language_loss": 0.72726607,
      "learning_rate": 2.1857000882907974e-06,
      "loss": 0.74878454,
      "num_input_tokens_seen": 173864975,
      "step": 8088,
      "time_per_iteration": 2.598118782043457
    },
    {
      "auxiliary_loss_clip": 0.01095207,
      "auxiliary_loss_mlp": 0.01035932,
      "balance_loss_clip": 1.03989434,
      "balance_loss_mlp": 1.02282536,
      "epoch": 0.48633699083120396,
      "flos": 21470703114240.0,
      "grad_norm": 1.47348351548496,
      "language_loss": 0.75322795,
      "learning_rate": 2.185312305524892e-06,
      "loss": 0.77453929,
      "num_input_tokens_seen": 173883805,
      "step": 8089,
      "time_per_iteration": 2.5351123809814453
    },
    {
      "auxiliary_loss_clip": 0.01089177,
      "auxiliary_loss_mlp": 0.0103385,
      "balance_loss_clip": 1.0440346,
      "balance_loss_mlp": 1.02014065,
      "epoch": 0.48639711408387193,
      "flos": 20084335833600.0,
      "grad_norm": 2.093287653790774,
      "language_loss": 0.84154546,
      "learning_rate": 2.184924515731926e-06,
      "loss": 0.86277574,
      "num_input_tokens_seen": 173903520,
      "step": 8090,
      "time_per_iteration": 2.5421924591064453
    },
    {
      "auxiliary_loss_clip": 0.01116458,
      "auxiliary_loss_mlp": 0.01029755,
      "balance_loss_clip": 1.0429616,
      "balance_loss_mlp": 1.0172019,
      "epoch": 0.4864572373365399,
      "flos": 20778884190720.0,
      "grad_norm": 1.5122913225045533,
      "language_loss": 0.76171499,
      "learning_rate": 2.1845367189266045e-06,
      "loss": 0.78317714,
      "num_input_tokens_seen": 173924255,
      "step": 8091,
      "time_per_iteration": 2.5014290809631348
    },
    {
      "auxiliary_loss_clip": 0.01111727,
      "auxiliary_loss_mlp": 0.01028693,
      "balance_loss_clip": 1.04444051,
      "balance_loss_mlp": 1.01543045,
      "epoch": 0.48651736058920786,
      "flos": 26025360503040.0,
      "grad_norm": 1.5154144875098614,
      "language_loss": 0.80394107,
      "learning_rate": 2.184148915123631e-06,
      "loss": 0.82534522,
      "num_input_tokens_seen": 173943285,
      "step": 8092,
      "time_per_iteration": 2.5358846187591553
    },
    {
      "auxiliary_loss_clip": 0.01096227,
      "auxiliary_loss_mlp": 0.00793595,
      "balance_loss_clip": 1.04417777,
      "balance_loss_mlp": 1.01579905,
      "epoch": 0.4865774838418758,
      "flos": 20485601642880.0,
      "grad_norm": 1.553319426938984,
      "language_loss": 0.72031498,
      "learning_rate": 2.1837611043377126e-06,
      "loss": 0.73921317,
      "num_input_tokens_seen": 173962205,
      "step": 8093,
      "time_per_iteration": 2.5700132846832275
    },
    {
      "auxiliary_loss_clip": 0.01119111,
      "auxiliary_loss_mlp": 0.01034132,
      "balance_loss_clip": 1.04392397,
      "balance_loss_mlp": 1.02181208,
      "epoch": 0.4866376070945438,
      "flos": 23547704169600.0,
      "grad_norm": 1.892545253406927,
      "language_loss": 0.67960298,
      "learning_rate": 2.1833732865835545e-06,
      "loss": 0.7011354,
      "num_input_tokens_seen": 173980945,
      "step": 8094,
      "time_per_iteration": 2.4802374839782715
    },
    {
      "auxiliary_loss_clip": 0.01104176,
      "auxiliary_loss_mlp": 0.01038143,
      "balance_loss_clip": 1.04653108,
      "balance_loss_mlp": 1.02432108,
      "epoch": 0.4866977303472118,
      "flos": 16690598012160.0,
      "grad_norm": 2.2632429911363126,
      "language_loss": 0.66322744,
      "learning_rate": 2.1829854618758636e-06,
      "loss": 0.68465066,
      "num_input_tokens_seen": 173998860,
      "step": 8095,
      "time_per_iteration": 2.4903366565704346
    },
    {
      "auxiliary_loss_clip": 0.01104543,
      "auxiliary_loss_mlp": 0.01035265,
      "balance_loss_clip": 1.0433352,
      "balance_loss_mlp": 1.02109122,
      "epoch": 0.4867578535998798,
      "flos": 17896011552000.0,
      "grad_norm": 2.0457537628089195,
      "language_loss": 0.78377211,
      "learning_rate": 2.182597630229345e-06,
      "loss": 0.80517018,
      "num_input_tokens_seen": 174016665,
      "step": 8096,
      "time_per_iteration": 2.4708750247955322
    },
    {
      "auxiliary_loss_clip": 0.01093148,
      "auxiliary_loss_mlp": 0.01036048,
      "balance_loss_clip": 1.0427959,
      "balance_loss_mlp": 1.02232671,
      "epoch": 0.48681797685254774,
      "flos": 22637799820800.0,
      "grad_norm": 1.7913719425930572,
      "language_loss": 0.67421496,
      "learning_rate": 2.1822097916587067e-06,
      "loss": 0.69550693,
      "num_input_tokens_seen": 174034800,
      "step": 8097,
      "time_per_iteration": 2.5455687046051025
    },
    {
      "auxiliary_loss_clip": 0.01092317,
      "auxiliary_loss_mlp": 0.01037859,
      "balance_loss_clip": 1.04289055,
      "balance_loss_mlp": 1.02429307,
      "epoch": 0.4868781001052157,
      "flos": 20886077352960.0,
      "grad_norm": 1.4461231272134294,
      "language_loss": 0.71358031,
      "learning_rate": 2.1818219461786543e-06,
      "loss": 0.73488206,
      "num_input_tokens_seen": 174054445,
      "step": 8098,
      "time_per_iteration": 2.5258939266204834
    },
    {
      "auxiliary_loss_clip": 0.01114057,
      "auxiliary_loss_mlp": 0.01034923,
      "balance_loss_clip": 1.04391003,
      "balance_loss_mlp": 1.02085078,
      "epoch": 0.48693822335788367,
      "flos": 41974940937600.0,
      "grad_norm": 2.071861821066187,
      "language_loss": 0.66022909,
      "learning_rate": 2.1814340938038956e-06,
      "loss": 0.68171883,
      "num_input_tokens_seen": 174077890,
      "step": 8099,
      "time_per_iteration": 2.670231342315674
    },
    {
      "auxiliary_loss_clip": 0.01069077,
      "auxiliary_loss_mlp": 0.01037143,
      "balance_loss_clip": 1.04140329,
      "balance_loss_mlp": 1.02392316,
      "epoch": 0.48699834661055164,
      "flos": 24243294021120.0,
      "grad_norm": 1.9499664476428467,
      "language_loss": 0.66903949,
      "learning_rate": 2.181046234549138e-06,
      "loss": 0.69010168,
      "num_input_tokens_seen": 174097460,
      "step": 8100,
      "time_per_iteration": 2.620561122894287
    },
    {
      "auxiliary_loss_clip": 0.01083296,
      "auxiliary_loss_mlp": 0.01031217,
      "balance_loss_clip": 1.04127455,
      "balance_loss_mlp": 1.01884365,
      "epoch": 0.4870584698632196,
      "flos": 25923877603200.0,
      "grad_norm": 1.3640007621508292,
      "language_loss": 0.76923907,
      "learning_rate": 2.180658368429088e-06,
      "loss": 0.79038417,
      "num_input_tokens_seen": 174120775,
      "step": 8101,
      "time_per_iteration": 2.6080002784729004
    },
    {
      "auxiliary_loss_clip": 0.01043951,
      "auxiliary_loss_mlp": 0.01004653,
      "balance_loss_clip": 1.018538,
      "balance_loss_mlp": 1.00310373,
      "epoch": 0.48711859311588757,
      "flos": 70211933648640.0,
      "grad_norm": 0.6753748363785065,
      "language_loss": 0.52264321,
      "learning_rate": 2.1802704954584565e-06,
      "loss": 0.54312921,
      "num_input_tokens_seen": 174189135,
      "step": 8102,
      "time_per_iteration": 3.209597110748291
    },
    {
      "auxiliary_loss_clip": 0.01094413,
      "auxiliary_loss_mlp": 0.01029807,
      "balance_loss_clip": 1.04813147,
      "balance_loss_mlp": 1.01699769,
      "epoch": 0.48717871636855553,
      "flos": 12342964659840.0,
      "grad_norm": 2.129419546935938,
      "language_loss": 0.73577547,
      "learning_rate": 2.1798826156519484e-06,
      "loss": 0.75701767,
      "num_input_tokens_seen": 174203250,
      "step": 8103,
      "time_per_iteration": 2.4841301441192627
    },
    {
      "auxiliary_loss_clip": 0.01112477,
      "auxiliary_loss_mlp": 0.01037246,
      "balance_loss_clip": 1.04553795,
      "balance_loss_mlp": 1.02323341,
      "epoch": 0.4872388396212235,
      "flos": 23477139901440.0,
      "grad_norm": 1.6432199233469378,
      "language_loss": 0.63237584,
      "learning_rate": 2.1794947290242737e-06,
      "loss": 0.65387309,
      "num_input_tokens_seen": 174224145,
      "step": 8104,
      "time_per_iteration": 2.5205342769622803
    },
    {
      "auxiliary_loss_clip": 0.01119659,
      "auxiliary_loss_mlp": 0.01028788,
      "balance_loss_clip": 1.0446806,
      "balance_loss_mlp": 1.01557326,
      "epoch": 0.48729896287389146,
      "flos": 31427582186880.0,
      "grad_norm": 1.5259325128706405,
      "language_loss": 0.69243836,
      "learning_rate": 2.1791068355901413e-06,
      "loss": 0.71392286,
      "num_input_tokens_seen": 174244435,
      "step": 8105,
      "time_per_iteration": 2.5404717922210693
    },
    {
      "auxiliary_loss_clip": 0.01084546,
      "auxiliary_loss_mlp": 0.01028115,
      "balance_loss_clip": 1.04322159,
      "balance_loss_mlp": 1.01571774,
      "epoch": 0.4873590861265594,
      "flos": 19057936700160.0,
      "grad_norm": 1.6274532734657456,
      "language_loss": 0.73808122,
      "learning_rate": 2.178718935364259e-06,
      "loss": 0.75920779,
      "num_input_tokens_seen": 174262710,
      "step": 8106,
      "time_per_iteration": 2.530871868133545
    },
    {
      "auxiliary_loss_clip": 0.01101996,
      "auxiliary_loss_mlp": 0.00791149,
      "balance_loss_clip": 1.04640186,
      "balance_loss_mlp": 1.01616895,
      "epoch": 0.4874192093792274,
      "flos": 24348296453760.0,
      "grad_norm": 2.2555824197212986,
      "language_loss": 0.76555526,
      "learning_rate": 2.1783310283613373e-06,
      "loss": 0.78448677,
      "num_input_tokens_seen": 174281545,
      "step": 8107,
      "time_per_iteration": 3.9273533821105957
    },
    {
      "auxiliary_loss_clip": 0.01069989,
      "auxiliary_loss_mlp": 0.01027448,
      "balance_loss_clip": 1.05000997,
      "balance_loss_mlp": 1.01488936,
      "epoch": 0.4874793326318954,
      "flos": 23112610727040.0,
      "grad_norm": 1.5323598903890174,
      "language_loss": 0.74862814,
      "learning_rate": 2.1779431145960853e-06,
      "loss": 0.76960254,
      "num_input_tokens_seen": 174300290,
      "step": 8108,
      "time_per_iteration": 2.6327316761016846
    },
    {
      "auxiliary_loss_clip": 0.01104897,
      "auxiliary_loss_mlp": 0.01027144,
      "balance_loss_clip": 1.0425694,
      "balance_loss_mlp": 1.01601624,
      "epoch": 0.4875394558845634,
      "flos": 19026156142080.0,
      "grad_norm": 2.032787600270871,
      "language_loss": 0.73974454,
      "learning_rate": 2.177555194083212e-06,
      "loss": 0.76106501,
      "num_input_tokens_seen": 174318490,
      "step": 8109,
      "time_per_iteration": 3.8498401641845703
    },
    {
      "auxiliary_loss_clip": 0.01104728,
      "auxiliary_loss_mlp": 0.01033045,
      "balance_loss_clip": 1.04164934,
      "balance_loss_mlp": 1.01967561,
      "epoch": 0.48759957913723134,
      "flos": 21433607343360.0,
      "grad_norm": 2.173245237349726,
      "language_loss": 0.78550184,
      "learning_rate": 2.177167266837428e-06,
      "loss": 0.80687958,
      "num_input_tokens_seen": 174335505,
      "step": 8110,
      "time_per_iteration": 2.4934215545654297
    },
    {
      "auxiliary_loss_clip": 0.01107773,
      "auxiliary_loss_mlp": 0.01035965,
      "balance_loss_clip": 1.04535818,
      "balance_loss_mlp": 1.02299523,
      "epoch": 0.4876597023898993,
      "flos": 17748669962880.0,
      "grad_norm": 1.8832152702783895,
      "language_loss": 0.72686356,
      "learning_rate": 2.176779332873444e-06,
      "loss": 0.74830091,
      "num_input_tokens_seen": 174353990,
      "step": 8111,
      "time_per_iteration": 2.4821102619171143
    },
    {
      "auxiliary_loss_clip": 0.01106484,
      "auxiliary_loss_mlp": 0.01033979,
      "balance_loss_clip": 1.04365933,
      "balance_loss_mlp": 1.02100897,
      "epoch": 0.4877198256425673,
      "flos": 17019647527680.0,
      "grad_norm": 1.5198546934688855,
      "language_loss": 0.7613883,
      "learning_rate": 2.17639139220597e-06,
      "loss": 0.78279299,
      "num_input_tokens_seen": 174373425,
      "step": 8112,
      "time_per_iteration": 3.9488158226013184
    },
    {
      "auxiliary_loss_clip": 0.01110645,
      "auxiliary_loss_mlp": 0.01036857,
      "balance_loss_clip": 1.04330277,
      "balance_loss_mlp": 1.02374995,
      "epoch": 0.48777994889523524,
      "flos": 22384091082240.0,
      "grad_norm": 1.471544131966931,
      "language_loss": 0.75022453,
      "learning_rate": 2.1760034448497166e-06,
      "loss": 0.77169955,
      "num_input_tokens_seen": 174393070,
      "step": 8113,
      "time_per_iteration": 2.527771472930908
    },
    {
      "auxiliary_loss_clip": 0.01028129,
      "auxiliary_loss_mlp": 0.0076805,
      "balance_loss_clip": 1.02232218,
      "balance_loss_mlp": 1.00289619,
      "epoch": 0.4878400721479032,
      "flos": 61241772159360.0,
      "grad_norm": 0.7765191612344156,
      "language_loss": 0.4885354,
      "learning_rate": 2.1756154908193943e-06,
      "loss": 0.5064972,
      "num_input_tokens_seen": 174446880,
      "step": 8114,
      "time_per_iteration": 3.025965690612793
    },
    {
      "auxiliary_loss_clip": 0.01093277,
      "auxiliary_loss_mlp": 0.01038653,
      "balance_loss_clip": 1.04246044,
      "balance_loss_mlp": 1.02429974,
      "epoch": 0.48790019540057117,
      "flos": 24536612482560.0,
      "grad_norm": 1.3282334923925103,
      "language_loss": 0.76622808,
      "learning_rate": 2.1752275301297155e-06,
      "loss": 0.78754735,
      "num_input_tokens_seen": 174468485,
      "step": 8115,
      "time_per_iteration": 4.069878339767456
    },
    {
      "auxiliary_loss_clip": 0.01101799,
      "auxiliary_loss_mlp": 0.01033808,
      "balance_loss_clip": 1.04689503,
      "balance_loss_mlp": 1.02009273,
      "epoch": 0.48796031865323913,
      "flos": 21833939399040.0,
      "grad_norm": 2.1200828757190924,
      "language_loss": 0.72356194,
      "learning_rate": 2.1748395627953915e-06,
      "loss": 0.74491799,
      "num_input_tokens_seen": 174486360,
      "step": 8116,
      "time_per_iteration": 2.529463291168213
    },
    {
      "auxiliary_loss_clip": 0.01083968,
      "auxiliary_loss_mlp": 0.01032258,
      "balance_loss_clip": 1.04188156,
      "balance_loss_mlp": 1.01946068,
      "epoch": 0.4880204419059071,
      "flos": 18588907883520.0,
      "grad_norm": 1.6563634175350912,
      "language_loss": 0.62728024,
      "learning_rate": 2.1744515888311335e-06,
      "loss": 0.64844251,
      "num_input_tokens_seen": 174505075,
      "step": 8117,
      "time_per_iteration": 2.5554823875427246
    },
    {
      "auxiliary_loss_clip": 0.0108886,
      "auxiliary_loss_mlp": 0.01037181,
      "balance_loss_clip": 1.04035211,
      "balance_loss_mlp": 1.02263713,
      "epoch": 0.48808056515857506,
      "flos": 19172168928000.0,
      "grad_norm": 1.764715300942369,
      "language_loss": 0.79514009,
      "learning_rate": 2.1740636082516533e-06,
      "loss": 0.81640053,
      "num_input_tokens_seen": 174523385,
      "step": 8118,
      "time_per_iteration": 2.5467748641967773
    },
    {
      "auxiliary_loss_clip": 0.0109593,
      "auxiliary_loss_mlp": 0.0103252,
      "balance_loss_clip": 1.04235959,
      "balance_loss_mlp": 1.01958525,
      "epoch": 0.48814068841124303,
      "flos": 20120497850880.0,
      "grad_norm": 1.605964734107816,
      "language_loss": 0.63049293,
      "learning_rate": 2.1736756210716645e-06,
      "loss": 0.65177745,
      "num_input_tokens_seen": 174542200,
      "step": 8119,
      "time_per_iteration": 2.5297837257385254
    },
    {
      "auxiliary_loss_clip": 0.01064031,
      "auxiliary_loss_mlp": 0.00788711,
      "balance_loss_clip": 1.04449069,
      "balance_loss_mlp": 1.01524878,
      "epoch": 0.488200811663911,
      "flos": 22965592360320.0,
      "grad_norm": 1.8106293666514048,
      "language_loss": 0.72157699,
      "learning_rate": 2.173287627305878e-06,
      "loss": 0.74010444,
      "num_input_tokens_seen": 174563620,
      "step": 8120,
      "time_per_iteration": 2.6518826484680176
    },
    {
      "auxiliary_loss_clip": 0.01108767,
      "auxiliary_loss_mlp": 0.01030619,
      "balance_loss_clip": 1.04120183,
      "balance_loss_mlp": 1.01704693,
      "epoch": 0.48826093491657896,
      "flos": 33910697387520.0,
      "grad_norm": 1.9834580113204097,
      "language_loss": 0.63527429,
      "learning_rate": 2.1728996269690075e-06,
      "loss": 0.65666813,
      "num_input_tokens_seen": 174586465,
      "step": 8121,
      "time_per_iteration": 2.604771852493286
    },
    {
      "auxiliary_loss_clip": 0.01107259,
      "auxiliary_loss_mlp": 0.01038214,
      "balance_loss_clip": 1.04153943,
      "balance_loss_mlp": 1.02393293,
      "epoch": 0.488321058169247,
      "flos": 23070307484160.0,
      "grad_norm": 2.049899536914779,
      "language_loss": 0.82906401,
      "learning_rate": 2.1725116200757664e-06,
      "loss": 0.8505187,
      "num_input_tokens_seen": 174604035,
      "step": 8122,
      "time_per_iteration": 2.5064961910247803
    },
    {
      "auxiliary_loss_clip": 0.01108529,
      "auxiliary_loss_mlp": 0.01042636,
      "balance_loss_clip": 1.04306912,
      "balance_loss_mlp": 1.02747846,
      "epoch": 0.48838118142191494,
      "flos": 19317714837120.0,
      "grad_norm": 1.685721105364919,
      "language_loss": 0.84874392,
      "learning_rate": 2.172123606640866e-06,
      "loss": 0.87025559,
      "num_input_tokens_seen": 174621715,
      "step": 8123,
      "time_per_iteration": 2.472954273223877
    },
    {
      "auxiliary_loss_clip": 0.01083099,
      "auxiliary_loss_mlp": 0.01028829,
      "balance_loss_clip": 1.04154336,
      "balance_loss_mlp": 1.01613355,
      "epoch": 0.4884413046745829,
      "flos": 25410678036480.0,
      "grad_norm": 1.4409330014640631,
      "language_loss": 0.85731828,
      "learning_rate": 2.1717355866790227e-06,
      "loss": 0.87843752,
      "num_input_tokens_seen": 174643835,
      "step": 8124,
      "time_per_iteration": 2.589334011077881
    },
    {
      "auxiliary_loss_clip": 0.010965,
      "auxiliary_loss_mlp": 0.01033152,
      "balance_loss_clip": 1.04508674,
      "balance_loss_mlp": 1.02003312,
      "epoch": 0.4885014279272509,
      "flos": 20991546662400.0,
      "grad_norm": 1.8014661625210506,
      "language_loss": 0.7934413,
      "learning_rate": 2.171347560204948e-06,
      "loss": 0.8147378,
      "num_input_tokens_seen": 174660955,
      "step": 8125,
      "time_per_iteration": 2.517944574356079
    },
    {
      "auxiliary_loss_clip": 0.01073206,
      "auxiliary_loss_mlp": 0.01038533,
      "balance_loss_clip": 1.04199886,
      "balance_loss_mlp": 1.024562,
      "epoch": 0.48856155117991884,
      "flos": 13771599269760.0,
      "grad_norm": 1.8015891234723764,
      "language_loss": 0.72931683,
      "learning_rate": 2.170959527233356e-06,
      "loss": 0.75043416,
      "num_input_tokens_seen": 174678270,
      "step": 8126,
      "time_per_iteration": 2.5601820945739746
    },
    {
      "auxiliary_loss_clip": 0.01106437,
      "auxiliary_loss_mlp": 0.01035455,
      "balance_loss_clip": 1.04032207,
      "balance_loss_mlp": 1.02227688,
      "epoch": 0.4886216744325868,
      "flos": 32087764206720.0,
      "grad_norm": 1.6226084213284804,
      "language_loss": 0.68716657,
      "learning_rate": 2.1705714877789633e-06,
      "loss": 0.70858544,
      "num_input_tokens_seen": 174698360,
      "step": 8127,
      "time_per_iteration": 2.603609323501587
    },
    {
      "auxiliary_loss_clip": 0.01119016,
      "auxiliary_loss_mlp": 0.01035766,
      "balance_loss_clip": 1.04029047,
      "balance_loss_mlp": 1.02217019,
      "epoch": 0.48868179768525477,
      "flos": 19610063631360.0,
      "grad_norm": 1.8006941149214921,
      "language_loss": 0.75913417,
      "learning_rate": 2.170183441856481e-06,
      "loss": 0.78068203,
      "num_input_tokens_seen": 174716755,
      "step": 8128,
      "time_per_iteration": 2.4768261909484863
    },
    {
      "auxiliary_loss_clip": 0.01120039,
      "auxiliary_loss_mlp": 0.01031692,
      "balance_loss_clip": 1.04355311,
      "balance_loss_mlp": 1.01926458,
      "epoch": 0.48874192093792274,
      "flos": 21286912199040.0,
      "grad_norm": 1.5527896005823756,
      "language_loss": 0.7595433,
      "learning_rate": 2.1697953894806265e-06,
      "loss": 0.78106058,
      "num_input_tokens_seen": 174735560,
      "step": 8129,
      "time_per_iteration": 2.4578840732574463
    },
    {
      "auxiliary_loss_clip": 0.01106915,
      "auxiliary_loss_mlp": 0.01032148,
      "balance_loss_clip": 1.04175377,
      "balance_loss_mlp": 1.0185225,
      "epoch": 0.4888020441905907,
      "flos": 14173439696640.0,
      "grad_norm": 2.1324097630163967,
      "language_loss": 0.64730191,
      "learning_rate": 2.169407330666114e-06,
      "loss": 0.66869259,
      "num_input_tokens_seen": 174752730,
      "step": 8130,
      "time_per_iteration": 2.4855897426605225
    },
    {
      "auxiliary_loss_clip": 0.01080492,
      "auxiliary_loss_mlp": 0.01026184,
      "balance_loss_clip": 1.03758669,
      "balance_loss_mlp": 1.01320231,
      "epoch": 0.48886216744325867,
      "flos": 24097891766400.0,
      "grad_norm": 2.131177809456845,
      "language_loss": 0.72187507,
      "learning_rate": 2.169019265427658e-06,
      "loss": 0.7429418,
      "num_input_tokens_seen": 174772520,
      "step": 8131,
      "time_per_iteration": 2.581220865249634
    },
    {
      "auxiliary_loss_clip": 0.01110038,
      "auxiliary_loss_mlp": 0.01034483,
      "balance_loss_clip": 1.04270315,
      "balance_loss_mlp": 1.02092242,
      "epoch": 0.48892229069592663,
      "flos": 38431419402240.0,
      "grad_norm": 1.362130759186972,
      "language_loss": 0.69617534,
      "learning_rate": 2.1686311937799745e-06,
      "loss": 0.71762049,
      "num_input_tokens_seen": 174796540,
      "step": 8132,
      "time_per_iteration": 2.67264723777771
    },
    {
      "auxiliary_loss_clip": 0.01110478,
      "auxiliary_loss_mlp": 0.0102957,
      "balance_loss_clip": 1.04317725,
      "balance_loss_mlp": 1.01626658,
      "epoch": 0.4889824139485946,
      "flos": 23843321101440.0,
      "grad_norm": 1.3418582900106162,
      "language_loss": 0.70487022,
      "learning_rate": 2.1682431157377797e-06,
      "loss": 0.72627068,
      "num_input_tokens_seen": 174817840,
      "step": 8133,
      "time_per_iteration": 2.5461366176605225
    },
    {
      "auxiliary_loss_clip": 0.01064592,
      "auxiliary_loss_mlp": 0.01040548,
      "balance_loss_clip": 1.04325628,
      "balance_loss_mlp": 1.02706587,
      "epoch": 0.48904253720126256,
      "flos": 24425827960320.0,
      "grad_norm": 2.001697220511099,
      "language_loss": 0.7036562,
      "learning_rate": 2.1678550313157883e-06,
      "loss": 0.7247076,
      "num_input_tokens_seen": 174837885,
      "step": 8134,
      "time_per_iteration": 2.6415648460388184
    },
    {
      "auxiliary_loss_clip": 0.01079893,
      "auxiliary_loss_mlp": 0.01031956,
      "balance_loss_clip": 1.04276013,
      "balance_loss_mlp": 1.01799083,
      "epoch": 0.4891026604539306,
      "flos": 24170682677760.0,
      "grad_norm": 1.9612867255745032,
      "language_loss": 0.80016589,
      "learning_rate": 2.167466940528718e-06,
      "loss": 0.82128435,
      "num_input_tokens_seen": 174855240,
      "step": 8135,
      "time_per_iteration": 2.6160531044006348
    },
    {
      "auxiliary_loss_clip": 0.01116213,
      "auxiliary_loss_mlp": 0.01034546,
      "balance_loss_clip": 1.04117036,
      "balance_loss_mlp": 1.02286959,
      "epoch": 0.48916278370659855,
      "flos": 21470954509440.0,
      "grad_norm": 1.5139315342196416,
      "language_loss": 0.74057674,
      "learning_rate": 2.1670788433912843e-06,
      "loss": 0.76208436,
      "num_input_tokens_seen": 174875145,
      "step": 8136,
      "time_per_iteration": 2.5476903915405273
    },
    {
      "auxiliary_loss_clip": 0.01093483,
      "auxiliary_loss_mlp": 0.01038322,
      "balance_loss_clip": 1.04297185,
      "balance_loss_mlp": 1.02464867,
      "epoch": 0.4892229069592665,
      "flos": 22309755886080.0,
      "grad_norm": 1.9199294323693348,
      "language_loss": 0.73642403,
      "learning_rate": 2.166690739918204e-06,
      "loss": 0.75774205,
      "num_input_tokens_seen": 174894770,
      "step": 8137,
      "time_per_iteration": 2.583580255508423
    },
    {
      "auxiliary_loss_clip": 0.01048045,
      "auxiliary_loss_mlp": 0.01032757,
      "balance_loss_clip": 1.03976607,
      "balance_loss_mlp": 1.01931643,
      "epoch": 0.4892830302119345,
      "flos": 12786856934400.0,
      "grad_norm": 4.192472822841957,
      "language_loss": 0.75560141,
      "learning_rate": 2.1663026301241944e-06,
      "loss": 0.77640939,
      "num_input_tokens_seen": 174912780,
      "step": 8138,
      "time_per_iteration": 2.6532206535339355
    },
    {
      "auxiliary_loss_clip": 0.01085305,
      "auxiliary_loss_mlp": 0.01031961,
      "balance_loss_clip": 1.04516327,
      "balance_loss_mlp": 1.01955163,
      "epoch": 0.48934315346460244,
      "flos": 20813896972800.0,
      "grad_norm": 1.7596590423628173,
      "language_loss": 0.74386334,
      "learning_rate": 2.165914514023972e-06,
      "loss": 0.76503599,
      "num_input_tokens_seen": 174931250,
      "step": 8139,
      "time_per_iteration": 2.586831569671631
    },
    {
      "auxiliary_loss_clip": 0.01109783,
      "auxiliary_loss_mlp": 0.01032725,
      "balance_loss_clip": 1.04308248,
      "balance_loss_mlp": 1.02063096,
      "epoch": 0.4894032767172704,
      "flos": 19755537713280.0,
      "grad_norm": 1.7945618229847682,
      "language_loss": 0.62678093,
      "learning_rate": 2.165526391632255e-06,
      "loss": 0.648206,
      "num_input_tokens_seen": 174951105,
      "step": 8140,
      "time_per_iteration": 2.5224621295928955
    },
    {
      "auxiliary_loss_clip": 0.01085635,
      "auxiliary_loss_mlp": 0.01036665,
      "balance_loss_clip": 1.04068983,
      "balance_loss_mlp": 1.02271128,
      "epoch": 0.4894633999699384,
      "flos": 17818982835840.0,
      "grad_norm": 5.436897034256438,
      "language_loss": 0.82428455,
      "learning_rate": 2.1651382629637608e-06,
      "loss": 0.84550756,
      "num_input_tokens_seen": 174969120,
      "step": 8141,
      "time_per_iteration": 2.563535213470459
    },
    {
      "auxiliary_loss_clip": 0.0109143,
      "auxiliary_loss_mlp": 0.01033459,
      "balance_loss_clip": 1.04844356,
      "balance_loss_mlp": 1.0198276,
      "epoch": 0.48952352322260634,
      "flos": 25523222325120.0,
      "grad_norm": 1.5747471227180858,
      "language_loss": 0.72389442,
      "learning_rate": 2.1647501280332066e-06,
      "loss": 0.74514329,
      "num_input_tokens_seen": 174991295,
      "step": 8142,
      "time_per_iteration": 2.699516773223877
    },
    {
      "auxiliary_loss_clip": 0.01116483,
      "auxiliary_loss_mlp": 0.01035111,
      "balance_loss_clip": 1.0416038,
      "balance_loss_mlp": 1.02266526,
      "epoch": 0.4895836464752743,
      "flos": 29055502903680.0,
      "grad_norm": 1.6841293516879212,
      "language_loss": 0.6709801,
      "learning_rate": 2.1643619868553105e-06,
      "loss": 0.69249606,
      "num_input_tokens_seen": 175012830,
      "step": 8143,
      "time_per_iteration": 2.559818983078003
    },
    {
      "auxiliary_loss_clip": 0.01107251,
      "auxiliary_loss_mlp": 0.00788635,
      "balance_loss_clip": 1.04220676,
      "balance_loss_mlp": 1.01701093,
      "epoch": 0.48964376972794227,
      "flos": 33546958312320.0,
      "grad_norm": 1.5193816723872193,
      "language_loss": 0.74939322,
      "learning_rate": 2.163973839444793e-06,
      "loss": 0.76835209,
      "num_input_tokens_seen": 175035695,
      "step": 8144,
      "time_per_iteration": 2.669614791870117
    },
    {
      "auxiliary_loss_clip": 0.01094989,
      "auxiliary_loss_mlp": 0.0103335,
      "balance_loss_clip": 1.04217386,
      "balance_loss_mlp": 1.02027845,
      "epoch": 0.48970389298061023,
      "flos": 22054035985920.0,
      "grad_norm": 1.363853225907896,
      "language_loss": 0.75677049,
      "learning_rate": 2.1635856858163695e-06,
      "loss": 0.77805382,
      "num_input_tokens_seen": 175056425,
      "step": 8145,
      "time_per_iteration": 2.5973618030548096
    },
    {
      "auxiliary_loss_clip": 0.01099817,
      "auxiliary_loss_mlp": 0.00788891,
      "balance_loss_clip": 1.041857,
      "balance_loss_mlp": 1.01273489,
      "epoch": 0.4897640162332782,
      "flos": 20084299920000.0,
      "grad_norm": 1.8168382473016025,
      "language_loss": 0.8061372,
      "learning_rate": 2.163197525984761e-06,
      "loss": 0.82502425,
      "num_input_tokens_seen": 175074800,
      "step": 8146,
      "time_per_iteration": 3.8753280639648438
    },
    {
      "auxiliary_loss_clip": 0.01102648,
      "auxiliary_loss_mlp": 0.01030811,
      "balance_loss_clip": 1.03956544,
      "balance_loss_mlp": 1.01799059,
      "epoch": 0.48982413948594616,
      "flos": 23806225330560.0,
      "grad_norm": 7.772188255698242,
      "language_loss": 0.74265862,
      "learning_rate": 2.162809359964687e-06,
      "loss": 0.7639932,
      "num_input_tokens_seen": 175094500,
      "step": 8147,
      "time_per_iteration": 2.5145633220672607
    },
    {
      "auxiliary_loss_clip": 0.01097369,
      "auxiliary_loss_mlp": 0.01031616,
      "balance_loss_clip": 1.04662931,
      "balance_loss_mlp": 1.01835382,
      "epoch": 0.4898842627386142,
      "flos": 17639645207040.0,
      "grad_norm": 2.027271021788659,
      "language_loss": 0.82957882,
      "learning_rate": 2.162421187770864e-06,
      "loss": 0.8508687,
      "num_input_tokens_seen": 175112920,
      "step": 8148,
      "time_per_iteration": 3.86165714263916
    },
    {
      "auxiliary_loss_clip": 0.01084006,
      "auxiliary_loss_mlp": 0.01030095,
      "balance_loss_clip": 1.04030752,
      "balance_loss_mlp": 1.01868081,
      "epoch": 0.48994438599128215,
      "flos": 16617914841600.0,
      "grad_norm": 1.706604264247751,
      "language_loss": 0.73782754,
      "learning_rate": 2.162033009418015e-06,
      "loss": 0.75896847,
      "num_input_tokens_seen": 175129910,
      "step": 8149,
      "time_per_iteration": 2.5396735668182373
    },
    {
      "auxiliary_loss_clip": 0.0112385,
      "auxiliary_loss_mlp": 0.01031942,
      "balance_loss_clip": 1.04451156,
      "balance_loss_mlp": 1.01786327,
      "epoch": 0.4900045092439501,
      "flos": 26614834600320.0,
      "grad_norm": 1.7595967727109412,
      "language_loss": 0.75966281,
      "learning_rate": 2.1616448249208567e-06,
      "loss": 0.78122067,
      "num_input_tokens_seen": 175148705,
      "step": 8150,
      "time_per_iteration": 2.5234949588775635
    },
    {
      "auxiliary_loss_clip": 0.01097136,
      "auxiliary_loss_mlp": 0.0103038,
      "balance_loss_clip": 1.04675341,
      "balance_loss_mlp": 1.01717138,
      "epoch": 0.4900646324966181,
      "flos": 19902125116800.0,
      "grad_norm": 1.8445976789335938,
      "language_loss": 0.72935802,
      "learning_rate": 2.1612566342941106e-06,
      "loss": 0.75063312,
      "num_input_tokens_seen": 175167425,
      "step": 8151,
      "time_per_iteration": 4.036783456802368
    },
    {
      "auxiliary_loss_clip": 0.01017781,
      "auxiliary_loss_mlp": 0.01011436,
      "balance_loss_clip": 1.02117801,
      "balance_loss_mlp": 1.01014888,
      "epoch": 0.49012475574928605,
      "flos": 59189620337280.0,
      "grad_norm": 0.8319905186555866,
      "language_loss": 0.54320681,
      "learning_rate": 2.1608684375524977e-06,
      "loss": 0.56349897,
      "num_input_tokens_seen": 175227985,
      "step": 8152,
      "time_per_iteration": 3.118217945098877
    },
    {
      "auxiliary_loss_clip": 0.01070979,
      "auxiliary_loss_mlp": 0.01034556,
      "balance_loss_clip": 1.04408002,
      "balance_loss_mlp": 1.02143121,
      "epoch": 0.490184879001954,
      "flos": 45259797657600.0,
      "grad_norm": 1.8257337771520936,
      "language_loss": 0.60662997,
      "learning_rate": 2.1604802347107364e-06,
      "loss": 0.62768531,
      "num_input_tokens_seen": 175251895,
      "step": 8153,
      "time_per_iteration": 4.287131309509277
    },
    {
      "auxiliary_loss_clip": 0.01084782,
      "auxiliary_loss_mlp": 0.01035076,
      "balance_loss_clip": 1.04303563,
      "balance_loss_mlp": 1.0217191,
      "epoch": 0.490245002254622,
      "flos": 28002135634560.0,
      "grad_norm": 1.5351365461960136,
      "language_loss": 0.77033615,
      "learning_rate": 2.160092025783549e-06,
      "loss": 0.79153478,
      "num_input_tokens_seen": 175272770,
      "step": 8154,
      "time_per_iteration": 2.6193149089813232
    },
    {
      "auxiliary_loss_clip": 0.01024639,
      "auxiliary_loss_mlp": 0.01002443,
      "balance_loss_clip": 1.01804638,
      "balance_loss_mlp": 1.0010184,
      "epoch": 0.49030512550728994,
      "flos": 58951318533120.0,
      "grad_norm": 0.9862919204668381,
      "language_loss": 0.67072666,
      "learning_rate": 2.1597038107856564e-06,
      "loss": 0.69099748,
      "num_input_tokens_seen": 175336320,
      "step": 8155,
      "time_per_iteration": 3.2242379188537598
    },
    {
      "auxiliary_loss_clip": 0.01119356,
      "auxiliary_loss_mlp": 0.01031473,
      "balance_loss_clip": 1.04346871,
      "balance_loss_mlp": 1.01888502,
      "epoch": 0.4903652487599579,
      "flos": 19791843384960.0,
      "grad_norm": 1.9189194030993046,
      "language_loss": 0.76798302,
      "learning_rate": 2.1593155897317784e-06,
      "loss": 0.7894913,
      "num_input_tokens_seen": 175353540,
      "step": 8156,
      "time_per_iteration": 2.465203285217285
    },
    {
      "auxiliary_loss_clip": 0.01108051,
      "auxiliary_loss_mlp": 0.01036198,
      "balance_loss_clip": 1.04309487,
      "balance_loss_mlp": 1.02356136,
      "epoch": 0.49042537201262587,
      "flos": 21762082241280.0,
      "grad_norm": 2.1078381562573805,
      "language_loss": 0.8385272,
      "learning_rate": 2.1589273626366377e-06,
      "loss": 0.85996974,
      "num_input_tokens_seen": 175370445,
      "step": 8157,
      "time_per_iteration": 2.5290350914001465
    },
    {
      "auxiliary_loss_clip": 0.01109237,
      "auxiliary_loss_mlp": 0.0103457,
      "balance_loss_clip": 1.04240417,
      "balance_loss_mlp": 1.02135527,
      "epoch": 0.49048549526529384,
      "flos": 18953042008320.0,
      "grad_norm": 1.8783022424843117,
      "language_loss": 0.79737127,
      "learning_rate": 2.158539129514956e-06,
      "loss": 0.81880927,
      "num_input_tokens_seen": 175389020,
      "step": 8158,
      "time_per_iteration": 2.4776599407196045
    },
    {
      "auxiliary_loss_clip": 0.01121501,
      "auxiliary_loss_mlp": 0.01031891,
      "balance_loss_clip": 1.04308701,
      "balance_loss_mlp": 1.0186348,
      "epoch": 0.4905456185179618,
      "flos": 26906393295360.0,
      "grad_norm": 1.5501797550037004,
      "language_loss": 0.69289935,
      "learning_rate": 2.158150890381454e-06,
      "loss": 0.71443331,
      "num_input_tokens_seen": 175409545,
      "step": 8159,
      "time_per_iteration": 2.531996250152588
    },
    {
      "auxiliary_loss_clip": 0.01101653,
      "auxiliary_loss_mlp": 0.01041428,
      "balance_loss_clip": 1.04014874,
      "balance_loss_mlp": 1.0269022,
      "epoch": 0.49060574177062977,
      "flos": 20412343854720.0,
      "grad_norm": 2.4179240226182435,
      "language_loss": 0.73228616,
      "learning_rate": 2.157762645250854e-06,
      "loss": 0.75371695,
      "num_input_tokens_seen": 175429335,
      "step": 8160,
      "time_per_iteration": 2.488171339035034
    },
    {
      "auxiliary_loss_clip": 0.01106691,
      "auxiliary_loss_mlp": 0.01041741,
      "balance_loss_clip": 1.04269314,
      "balance_loss_mlp": 1.02822268,
      "epoch": 0.4906658650232978,
      "flos": 17493704248320.0,
      "grad_norm": 2.0219950610111503,
      "language_loss": 0.71496272,
      "learning_rate": 2.1573743941378796e-06,
      "loss": 0.7364471,
      "num_input_tokens_seen": 175446955,
      "step": 8161,
      "time_per_iteration": 2.4966702461242676
    },
    {
      "auxiliary_loss_clip": 0.01074796,
      "auxiliary_loss_mlp": 0.01040495,
      "balance_loss_clip": 1.04266167,
      "balance_loss_mlp": 1.02760291,
      "epoch": 0.49072598827596575,
      "flos": 26614439550720.0,
      "grad_norm": 2.349048493282997,
      "language_loss": 0.68559825,
      "learning_rate": 2.1569861370572517e-06,
      "loss": 0.70675123,
      "num_input_tokens_seen": 175468195,
      "step": 8162,
      "time_per_iteration": 2.6270155906677246
    },
    {
      "auxiliary_loss_clip": 0.01107748,
      "auxiliary_loss_mlp": 0.01037921,
      "balance_loss_clip": 1.04360032,
      "balance_loss_mlp": 1.02327013,
      "epoch": 0.4907861115286337,
      "flos": 20412595249920.0,
      "grad_norm": 2.485934092439231,
      "language_loss": 0.63295412,
      "learning_rate": 2.1565978740236944e-06,
      "loss": 0.65441084,
      "num_input_tokens_seen": 175487455,
      "step": 8163,
      "time_per_iteration": 2.526893138885498
    },
    {
      "auxiliary_loss_clip": 0.01082337,
      "auxiliary_loss_mlp": 0.01035008,
      "balance_loss_clip": 1.04263115,
      "balance_loss_mlp": 1.0220263,
      "epoch": 0.4908462347813017,
      "flos": 14064271286400.0,
      "grad_norm": 2.1156573574833026,
      "language_loss": 0.77520347,
      "learning_rate": 2.1562096050519293e-06,
      "loss": 0.79637688,
      "num_input_tokens_seen": 175504450,
      "step": 8164,
      "time_per_iteration": 2.5376946926116943
    },
    {
      "auxiliary_loss_clip": 0.01102297,
      "auxiliary_loss_mlp": 0.01037919,
      "balance_loss_clip": 1.03904188,
      "balance_loss_mlp": 1.02298248,
      "epoch": 0.49090635803396965,
      "flos": 18735100237440.0,
      "grad_norm": 1.5613918788906072,
      "language_loss": 0.76597595,
      "learning_rate": 2.1558213301566806e-06,
      "loss": 0.78737807,
      "num_input_tokens_seen": 175523600,
      "step": 8165,
      "time_per_iteration": 2.502760171890259
    },
    {
      "auxiliary_loss_clip": 0.01098356,
      "auxiliary_loss_mlp": 0.01033258,
      "balance_loss_clip": 1.04367399,
      "balance_loss_mlp": 1.02036572,
      "epoch": 0.4909664812866376,
      "flos": 20558500295040.0,
      "grad_norm": 1.5894416628421937,
      "language_loss": 0.7737875,
      "learning_rate": 2.1554330493526716e-06,
      "loss": 0.79510367,
      "num_input_tokens_seen": 175542720,
      "step": 8166,
      "time_per_iteration": 2.5327980518341064
    },
    {
      "auxiliary_loss_clip": 0.01035128,
      "auxiliary_loss_mlp": 0.01001411,
      "balance_loss_clip": 1.01836753,
      "balance_loss_mlp": 0.99993914,
      "epoch": 0.4910266045393056,
      "flos": 54684017948160.0,
      "grad_norm": 0.8073798051043309,
      "language_loss": 0.5421083,
      "learning_rate": 2.1550447626546253e-06,
      "loss": 0.56247377,
      "num_input_tokens_seen": 175598640,
      "step": 8167,
      "time_per_iteration": 3.1865673065185547
    },
    {
      "auxiliary_loss_clip": 0.01072112,
      "auxiliary_loss_mlp": 0.01032288,
      "balance_loss_clip": 1.03988433,
      "balance_loss_mlp": 1.01947868,
      "epoch": 0.49108672779197354,
      "flos": 16246454342400.0,
      "grad_norm": 2.083874731765354,
      "language_loss": 0.86155379,
      "learning_rate": 2.1546564700772665e-06,
      "loss": 0.8825978,
      "num_input_tokens_seen": 175615675,
      "step": 8168,
      "time_per_iteration": 2.554624080657959
    },
    {
      "auxiliary_loss_clip": 0.01099465,
      "auxiliary_loss_mlp": 0.01035926,
      "balance_loss_clip": 1.03933775,
      "balance_loss_mlp": 1.02197862,
      "epoch": 0.4911468510446415,
      "flos": 19825419623040.0,
      "grad_norm": 1.6722816474304885,
      "language_loss": 0.73383689,
      "learning_rate": 2.1542681716353193e-06,
      "loss": 0.75519079,
      "num_input_tokens_seen": 175632255,
      "step": 8169,
      "time_per_iteration": 2.512869358062744
    },
    {
      "auxiliary_loss_clip": 0.01104601,
      "auxiliary_loss_mlp": 0.0102872,
      "balance_loss_clip": 1.04063165,
      "balance_loss_mlp": 1.01695979,
      "epoch": 0.4912069742973095,
      "flos": 21212684743680.0,
      "grad_norm": 1.5203929831468295,
      "language_loss": 0.77745038,
      "learning_rate": 2.1538798673435068e-06,
      "loss": 0.79878354,
      "num_input_tokens_seen": 175651625,
      "step": 8170,
      "time_per_iteration": 2.5029845237731934
    },
    {
      "auxiliary_loss_clip": 0.01092141,
      "auxiliary_loss_mlp": 0.01038853,
      "balance_loss_clip": 1.04196894,
      "balance_loss_mlp": 1.02651453,
      "epoch": 0.49126709754997744,
      "flos": 19537129065600.0,
      "grad_norm": 2.3122093064033806,
      "language_loss": 0.75967467,
      "learning_rate": 2.1534915572165545e-06,
      "loss": 0.78098464,
      "num_input_tokens_seen": 175669265,
      "step": 8171,
      "time_per_iteration": 2.5726845264434814
    },
    {
      "auxiliary_loss_clip": 0.01098075,
      "auxiliary_loss_mlp": 0.01035548,
      "balance_loss_clip": 1.04071212,
      "balance_loss_mlp": 1.02275658,
      "epoch": 0.4913272208026454,
      "flos": 12239686080000.0,
      "grad_norm": 1.7433070699281912,
      "language_loss": 0.81155986,
      "learning_rate": 2.1531032412691875e-06,
      "loss": 0.83289611,
      "num_input_tokens_seen": 175686065,
      "step": 8172,
      "time_per_iteration": 2.511493444442749
    },
    {
      "auxiliary_loss_clip": 0.01032847,
      "auxiliary_loss_mlp": 0.01007874,
      "balance_loss_clip": 1.01721597,
      "balance_loss_mlp": 1.00632465,
      "epoch": 0.49138734405531337,
      "flos": 65465871661440.0,
      "grad_norm": 0.6920903811599551,
      "language_loss": 0.53301167,
      "learning_rate": 2.1527149195161295e-06,
      "loss": 0.55341887,
      "num_input_tokens_seen": 175748595,
      "step": 8173,
      "time_per_iteration": 3.091998815536499
    },
    {
      "auxiliary_loss_clip": 0.01110386,
      "auxiliary_loss_mlp": 0.0078913,
      "balance_loss_clip": 1.04217362,
      "balance_loss_mlp": 1.01544785,
      "epoch": 0.4914474673079814,
      "flos": 18439052342400.0,
      "grad_norm": 1.948709681630015,
      "language_loss": 0.62824488,
      "learning_rate": 2.152326591972107e-06,
      "loss": 0.64724004,
      "num_input_tokens_seen": 175766770,
      "step": 8174,
      "time_per_iteration": 2.4670445919036865
    },
    {
      "auxiliary_loss_clip": 0.01080843,
      "auxiliary_loss_mlp": 0.01040713,
      "balance_loss_clip": 1.04117,
      "balance_loss_mlp": 1.02587152,
      "epoch": 0.49150759056064935,
      "flos": 21685053525120.0,
      "grad_norm": 1.8175318427883584,
      "language_loss": 0.68825454,
      "learning_rate": 2.1519382586518445e-06,
      "loss": 0.70947009,
      "num_input_tokens_seen": 175783605,
      "step": 8175,
      "time_per_iteration": 2.54353666305542
    },
    {
      "auxiliary_loss_clip": 0.01106559,
      "auxiliary_loss_mlp": 0.0102869,
      "balance_loss_clip": 1.04069161,
      "balance_loss_mlp": 1.01651239,
      "epoch": 0.4915677138133173,
      "flos": 22382439056640.0,
      "grad_norm": 1.9204395516689636,
      "language_loss": 0.74159741,
      "learning_rate": 2.151549919570068e-06,
      "loss": 0.76294994,
      "num_input_tokens_seen": 175801390,
      "step": 8176,
      "time_per_iteration": 2.485837936401367
    },
    {
      "auxiliary_loss_clip": 0.01108691,
      "auxiliary_loss_mlp": 0.01040028,
      "balance_loss_clip": 1.04147863,
      "balance_loss_mlp": 1.0270406,
      "epoch": 0.4916278370659853,
      "flos": 18402890325120.0,
      "grad_norm": 1.6189458434969857,
      "language_loss": 0.69592416,
      "learning_rate": 2.1511615747415036e-06,
      "loss": 0.7174114,
      "num_input_tokens_seen": 175819830,
      "step": 8177,
      "time_per_iteration": 2.487250328063965
    },
    {
      "auxiliary_loss_clip": 0.01024525,
      "auxiliary_loss_mlp": 0.00777279,
      "balance_loss_clip": 1.01947427,
      "balance_loss_mlp": 1.02545762,
      "epoch": 0.49168796031865325,
      "flos": 66609124715520.0,
      "grad_norm": 0.6872048288824115,
      "language_loss": 0.46260095,
      "learning_rate": 2.150773224180877e-06,
      "loss": 0.48061901,
      "num_input_tokens_seen": 175881765,
      "step": 8178,
      "time_per_iteration": 3.1001181602478027
    },
    {
      "auxiliary_loss_clip": 0.01122505,
      "auxiliary_loss_mlp": 0.01033419,
      "balance_loss_clip": 1.0434767,
      "balance_loss_mlp": 1.01926851,
      "epoch": 0.4917480835713212,
      "flos": 20959335141120.0,
      "grad_norm": 1.801389592271759,
      "language_loss": 0.65958345,
      "learning_rate": 2.1503848679029147e-06,
      "loss": 0.68114269,
      "num_input_tokens_seen": 175901795,
      "step": 8179,
      "time_per_iteration": 2.471752405166626
    },
    {
      "auxiliary_loss_clip": 0.01043284,
      "auxiliary_loss_mlp": 0.01047214,
      "balance_loss_clip": 1.0449059,
      "balance_loss_mlp": 1.0314188,
      "epoch": 0.4918082068239892,
      "flos": 15772900412160.0,
      "grad_norm": 1.8918293400675485,
      "language_loss": 0.70324445,
      "learning_rate": 2.149996505922343e-06,
      "loss": 0.72414947,
      "num_input_tokens_seen": 175917770,
      "step": 8180,
      "time_per_iteration": 2.690161943435669
    },
    {
      "auxiliary_loss_clip": 0.01093835,
      "auxiliary_loss_mlp": 0.01029206,
      "balance_loss_clip": 1.04067683,
      "balance_loss_mlp": 1.01606894,
      "epoch": 0.49186833007665715,
      "flos": 24604806453120.0,
      "grad_norm": 1.7080577910753465,
      "language_loss": 0.84273708,
      "learning_rate": 2.1496081382538895e-06,
      "loss": 0.86396748,
      "num_input_tokens_seen": 175937000,
      "step": 8181,
      "time_per_iteration": 2.715656042098999
    },
    {
      "auxiliary_loss_clip": 0.01115778,
      "auxiliary_loss_mlp": 0.01032121,
      "balance_loss_clip": 1.04210734,
      "balance_loss_mlp": 1.01971149,
      "epoch": 0.4919284533293251,
      "flos": 22090557139200.0,
      "grad_norm": 2.3301684490852135,
      "language_loss": 0.72564888,
      "learning_rate": 2.1492197649122793e-06,
      "loss": 0.74712789,
      "num_input_tokens_seen": 175955170,
      "step": 8182,
      "time_per_iteration": 2.4785163402557373
    },
    {
      "auxiliary_loss_clip": 0.01081984,
      "auxiliary_loss_mlp": 0.01036288,
      "balance_loss_clip": 1.04353404,
      "balance_loss_mlp": 1.02366364,
      "epoch": 0.4919885765819931,
      "flos": 23368043318400.0,
      "grad_norm": 1.7835190969667145,
      "language_loss": 0.72403538,
      "learning_rate": 2.1488313859122412e-06,
      "loss": 0.74521804,
      "num_input_tokens_seen": 175973725,
      "step": 8183,
      "time_per_iteration": 2.565495491027832
    },
    {
      "auxiliary_loss_clip": 0.01061324,
      "auxiliary_loss_mlp": 0.01032347,
      "balance_loss_clip": 1.0381583,
      "balance_loss_mlp": 1.01755333,
      "epoch": 0.49204869983466104,
      "flos": 21360493209600.0,
      "grad_norm": 1.9981745164762756,
      "language_loss": 0.77233094,
      "learning_rate": 2.1484430012685015e-06,
      "loss": 0.79326761,
      "num_input_tokens_seen": 175993885,
      "step": 8184,
      "time_per_iteration": 2.597191572189331
    },
    {
      "auxiliary_loss_clip": 0.01090303,
      "auxiliary_loss_mlp": 0.01030585,
      "balance_loss_clip": 1.04309297,
      "balance_loss_mlp": 1.01785314,
      "epoch": 0.492108823087329,
      "flos": 21142695093120.0,
      "grad_norm": 1.62097660916587,
      "language_loss": 0.70611238,
      "learning_rate": 2.148054610995789e-06,
      "loss": 0.72732127,
      "num_input_tokens_seen": 176014210,
      "step": 8185,
      "time_per_iteration": 4.228662014007568
    },
    {
      "auxiliary_loss_clip": 0.01100955,
      "auxiliary_loss_mlp": 0.01036694,
      "balance_loss_clip": 1.04361618,
      "balance_loss_mlp": 1.02252603,
      "epoch": 0.49216894633999697,
      "flos": 25116605389440.0,
      "grad_norm": 1.6830366602612554,
      "language_loss": 0.74962902,
      "learning_rate": 2.147666215108831e-06,
      "loss": 0.77100551,
      "num_input_tokens_seen": 176033890,
      "step": 8186,
      "time_per_iteration": 2.593404769897461
    },
    {
      "auxiliary_loss_clip": 0.01106909,
      "auxiliary_loss_mlp": 0.01036207,
      "balance_loss_clip": 1.04278398,
      "balance_loss_mlp": 1.02312398,
      "epoch": 0.49222906959266494,
      "flos": 22637943475200.0,
      "grad_norm": 1.833931425884278,
      "language_loss": 0.67632341,
      "learning_rate": 2.1472778136223545e-06,
      "loss": 0.6977545,
      "num_input_tokens_seen": 176052720,
      "step": 8187,
      "time_per_iteration": 3.8550517559051514
    },
    {
      "auxiliary_loss_clip": 0.01073542,
      "auxiliary_loss_mlp": 0.01031649,
      "balance_loss_clip": 1.04005408,
      "balance_loss_mlp": 1.01814842,
      "epoch": 0.49228919284533296,
      "flos": 20410548174720.0,
      "grad_norm": 1.4090276286909484,
      "language_loss": 0.66701066,
      "learning_rate": 2.1468894065510894e-06,
      "loss": 0.68806255,
      "num_input_tokens_seen": 176072545,
      "step": 8188,
      "time_per_iteration": 2.650702953338623
    },
    {
      "auxiliary_loss_clip": 0.01109799,
      "auxiliary_loss_mlp": 0.01025596,
      "balance_loss_clip": 1.04404426,
      "balance_loss_mlp": 1.01369345,
      "epoch": 0.4923493160980009,
      "flos": 27122359818240.0,
      "grad_norm": 1.7132627030585639,
      "language_loss": 0.74939322,
      "learning_rate": 2.1465009939097623e-06,
      "loss": 0.77074718,
      "num_input_tokens_seen": 176091490,
      "step": 8189,
      "time_per_iteration": 3.964816093444824
    },
    {
      "auxiliary_loss_clip": 0.01094647,
      "auxiliary_loss_mlp": 0.01028455,
      "balance_loss_clip": 1.03959036,
      "balance_loss_mlp": 1.01555026,
      "epoch": 0.4924094393506689,
      "flos": 35736683224320.0,
      "grad_norm": 1.5044756666948194,
      "language_loss": 0.64630812,
      "learning_rate": 2.146112575713104e-06,
      "loss": 0.66753912,
      "num_input_tokens_seen": 176113200,
      "step": 8190,
      "time_per_iteration": 2.6320598125457764
    },
    {
      "auxiliary_loss_clip": 0.01116716,
      "auxiliary_loss_mlp": 0.01027166,
      "balance_loss_clip": 1.04219639,
      "balance_loss_mlp": 1.0139699,
      "epoch": 0.49246956260333685,
      "flos": 20412487509120.0,
      "grad_norm": 1.9581182575716516,
      "language_loss": 0.71478641,
      "learning_rate": 2.1457241519758413e-06,
      "loss": 0.73622525,
      "num_input_tokens_seen": 176132485,
      "step": 8191,
      "time_per_iteration": 2.446629285812378
    },
    {
      "auxiliary_loss_clip": 0.01116914,
      "auxiliary_loss_mlp": 0.00808011,
      "balance_loss_clip": 1.04081917,
      "balance_loss_mlp": 1.05142498,
      "epoch": 0.4925296858560048,
      "flos": 38976938231040.0,
      "grad_norm": 1.5021267755938106,
      "language_loss": 0.71638829,
      "learning_rate": 2.1453357227127043e-06,
      "loss": 0.73563755,
      "num_input_tokens_seen": 176155755,
      "step": 8192,
      "time_per_iteration": 4.032135963439941
    },
    {
      "auxiliary_loss_clip": 0.01023999,
      "auxiliary_loss_mlp": 0.01002789,
      "balance_loss_clip": 1.01788235,
      "balance_loss_mlp": 1.00123942,
      "epoch": 0.4925898091086728,
      "flos": 64278917712000.0,
      "grad_norm": 0.7173293710945046,
      "language_loss": 0.52089703,
      "learning_rate": 2.1449472879384224e-06,
      "loss": 0.54116493,
      "num_input_tokens_seen": 176216295,
      "step": 8193,
      "time_per_iteration": 3.1938841342926025
    },
    {
      "auxiliary_loss_clip": 0.01118458,
      "auxiliary_loss_mlp": 0.0103662,
      "balance_loss_clip": 1.04418921,
      "balance_loss_mlp": 1.02375114,
      "epoch": 0.49264993236134075,
      "flos": 23036372110080.0,
      "grad_norm": 1.3849243611771886,
      "language_loss": 0.76958716,
      "learning_rate": 2.1445588476677246e-06,
      "loss": 0.79113793,
      "num_input_tokens_seen": 176235925,
      "step": 8194,
      "time_per_iteration": 2.4860715866088867
    },
    {
      "auxiliary_loss_clip": 0.01091339,
      "auxiliary_loss_mlp": 0.01029505,
      "balance_loss_clip": 1.039729,
      "balance_loss_mlp": 1.01713085,
      "epoch": 0.4927100556140087,
      "flos": 24718212668160.0,
      "grad_norm": 2.170937148500071,
      "language_loss": 0.7013936,
      "learning_rate": 2.144170401915341e-06,
      "loss": 0.72260201,
      "num_input_tokens_seen": 176253865,
      "step": 8195,
      "time_per_iteration": 2.534700870513916
    },
    {
      "auxiliary_loss_clip": 0.01076064,
      "auxiliary_loss_mlp": 0.01026794,
      "balance_loss_clip": 1.04099345,
      "balance_loss_mlp": 1.01446176,
      "epoch": 0.4927701788666767,
      "flos": 23505544581120.0,
      "grad_norm": 1.8678204686243725,
      "language_loss": 0.80775541,
      "learning_rate": 2.143781950696001e-06,
      "loss": 0.82878399,
      "num_input_tokens_seen": 176271525,
      "step": 8196,
      "time_per_iteration": 2.608266592025757
    },
    {
      "auxiliary_loss_clip": 0.01085893,
      "auxiliary_loss_mlp": 0.01035814,
      "balance_loss_clip": 1.03935099,
      "balance_loss_mlp": 1.02165771,
      "epoch": 0.49283030211934464,
      "flos": 22928891639040.0,
      "grad_norm": 1.7068693132646402,
      "language_loss": 0.70260966,
      "learning_rate": 2.1433934940244356e-06,
      "loss": 0.72382665,
      "num_input_tokens_seen": 176290810,
      "step": 8197,
      "time_per_iteration": 2.5475881099700928
    },
    {
      "auxiliary_loss_clip": 0.01104205,
      "auxiliary_loss_mlp": 0.01032423,
      "balance_loss_clip": 1.04301095,
      "balance_loss_mlp": 1.02031732,
      "epoch": 0.4928904253720126,
      "flos": 16873024210560.0,
      "grad_norm": 1.7722197562865314,
      "language_loss": 0.83681887,
      "learning_rate": 2.143005031915374e-06,
      "loss": 0.85818511,
      "num_input_tokens_seen": 176309165,
      "step": 8198,
      "time_per_iteration": 2.4652392864227295
    },
    {
      "auxiliary_loss_clip": 0.01111305,
      "auxiliary_loss_mlp": 0.01035592,
      "balance_loss_clip": 1.04295754,
      "balance_loss_mlp": 1.02214575,
      "epoch": 0.4929505486246806,
      "flos": 14866551509760.0,
      "grad_norm": 1.7837995826813415,
      "language_loss": 0.76211095,
      "learning_rate": 2.1426165643835467e-06,
      "loss": 0.78357995,
      "num_input_tokens_seen": 176324960,
      "step": 8199,
      "time_per_iteration": 2.4514119625091553
    },
    {
      "auxiliary_loss_clip": 0.01100827,
      "auxiliary_loss_mlp": 0.01035113,
      "balance_loss_clip": 1.04276049,
      "balance_loss_mlp": 1.0212431,
      "epoch": 0.49301067187734854,
      "flos": 23842351434240.0,
      "grad_norm": 1.6454780080899907,
      "language_loss": 0.60006422,
      "learning_rate": 2.1422280914436864e-06,
      "loss": 0.6214236,
      "num_input_tokens_seen": 176346195,
      "step": 8200,
      "time_per_iteration": 2.566474676132202
    },
    {
      "auxiliary_loss_clip": 0.01104552,
      "auxiliary_loss_mlp": 0.01035313,
      "balance_loss_clip": 1.04122996,
      "balance_loss_mlp": 1.02284336,
      "epoch": 0.49307079513001656,
      "flos": 22491284244480.0,
      "grad_norm": 1.4222360666984644,
      "language_loss": 0.793917,
      "learning_rate": 2.1418396131105213e-06,
      "loss": 0.81531566,
      "num_input_tokens_seen": 176366735,
      "step": 8201,
      "time_per_iteration": 2.4994237422943115
    },
    {
      "auxiliary_loss_clip": 0.01110423,
      "auxiliary_loss_mlp": 0.01032073,
      "balance_loss_clip": 1.04084098,
      "balance_loss_mlp": 1.01808977,
      "epoch": 0.4931309183826845,
      "flos": 15924587546880.0,
      "grad_norm": 2.100316512426948,
      "language_loss": 0.67547423,
      "learning_rate": 2.141451129398785e-06,
      "loss": 0.69689918,
      "num_input_tokens_seen": 176384475,
      "step": 8202,
      "time_per_iteration": 2.4602534770965576
    },
    {
      "auxiliary_loss_clip": 0.01092109,
      "auxiliary_loss_mlp": 0.01032122,
      "balance_loss_clip": 1.04063725,
      "balance_loss_mlp": 1.01949215,
      "epoch": 0.4931910416353525,
      "flos": 27309059735040.0,
      "grad_norm": 1.978266229650916,
      "language_loss": 0.7520625,
      "learning_rate": 2.1410626403232076e-06,
      "loss": 0.77330482,
      "num_input_tokens_seen": 176402645,
      "step": 8203,
      "time_per_iteration": 2.574143648147583
    },
    {
      "auxiliary_loss_clip": 0.01066459,
      "auxiliary_loss_mlp": 0.01032823,
      "balance_loss_clip": 1.04338944,
      "balance_loss_mlp": 1.02004409,
      "epoch": 0.49325116488802045,
      "flos": 20806139635200.0,
      "grad_norm": 2.3128701371195124,
      "language_loss": 0.80443156,
      "learning_rate": 2.1406741458985197e-06,
      "loss": 0.82542437,
      "num_input_tokens_seen": 176416715,
      "step": 8204,
      "time_per_iteration": 2.5704519748687744
    },
    {
      "auxiliary_loss_clip": 0.01103939,
      "auxiliary_loss_mlp": 0.01033059,
      "balance_loss_clip": 1.04117048,
      "balance_loss_mlp": 1.02121568,
      "epoch": 0.4933112881406884,
      "flos": 19865963099520.0,
      "grad_norm": 1.886996923939048,
      "language_loss": 0.66605663,
      "learning_rate": 2.140285646139455e-06,
      "loss": 0.68742657,
      "num_input_tokens_seen": 176435755,
      "step": 8205,
      "time_per_iteration": 2.4763028621673584
    },
    {
      "auxiliary_loss_clip": 0.01122677,
      "auxiliary_loss_mlp": 0.01035712,
      "balance_loss_clip": 1.04249549,
      "balance_loss_mlp": 1.0212816,
      "epoch": 0.4933714113933564,
      "flos": 21827977741440.0,
      "grad_norm": 1.603523546904001,
      "language_loss": 0.65942907,
      "learning_rate": 2.139897141060744e-06,
      "loss": 0.68101293,
      "num_input_tokens_seen": 176453915,
      "step": 8206,
      "time_per_iteration": 2.4716951847076416
    },
    {
      "auxiliary_loss_clip": 0.01074138,
      "auxiliary_loss_mlp": 0.01038538,
      "balance_loss_clip": 1.04042101,
      "balance_loss_mlp": 1.02302885,
      "epoch": 0.49343153464602435,
      "flos": 27890130049920.0,
      "grad_norm": 1.55371843668325,
      "language_loss": 0.76413816,
      "learning_rate": 2.1395086306771196e-06,
      "loss": 0.78526491,
      "num_input_tokens_seen": 176475175,
      "step": 8207,
      "time_per_iteration": 2.6033709049224854
    },
    {
      "auxiliary_loss_clip": 0.01099875,
      "auxiliary_loss_mlp": 0.01033857,
      "balance_loss_clip": 1.04211402,
      "balance_loss_mlp": 1.01998651,
      "epoch": 0.4934916578986923,
      "flos": 24681080983680.0,
      "grad_norm": 2.141399692839832,
      "language_loss": 0.60236597,
      "learning_rate": 2.1391201150033147e-06,
      "loss": 0.62370324,
      "num_input_tokens_seen": 176494250,
      "step": 8208,
      "time_per_iteration": 2.555893898010254
    },
    {
      "auxiliary_loss_clip": 0.01099271,
      "auxiliary_loss_mlp": 0.01031106,
      "balance_loss_clip": 1.04118562,
      "balance_loss_mlp": 1.01790917,
      "epoch": 0.4935517811513603,
      "flos": 23405139089280.0,
      "grad_norm": 1.955368470513386,
      "language_loss": 0.7840665,
      "learning_rate": 2.1387315940540598e-06,
      "loss": 0.80537021,
      "num_input_tokens_seen": 176513325,
      "step": 8209,
      "time_per_iteration": 2.527578592300415
    },
    {
      "auxiliary_loss_clip": 0.0109113,
      "auxiliary_loss_mlp": 0.00805273,
      "balance_loss_clip": 1.04020905,
      "balance_loss_mlp": 1.03762448,
      "epoch": 0.49361190440402825,
      "flos": 21944508439680.0,
      "grad_norm": 1.9276409194238697,
      "language_loss": 0.78949094,
      "learning_rate": 2.138343067844089e-06,
      "loss": 0.80845493,
      "num_input_tokens_seen": 176532915,
      "step": 8210,
      "time_per_iteration": 2.5510919094085693
    },
    {
      "auxiliary_loss_clip": 0.0110848,
      "auxiliary_loss_mlp": 0.01033892,
      "balance_loss_clip": 1.04342842,
      "balance_loss_mlp": 1.0201894,
      "epoch": 0.4936720276566962,
      "flos": 25115671635840.0,
      "grad_norm": 1.6911658739001645,
      "language_loss": 0.81184864,
      "learning_rate": 2.1379545363881363e-06,
      "loss": 0.8332724,
      "num_input_tokens_seen": 176552775,
      "step": 8211,
      "time_per_iteration": 2.5586726665496826
    },
    {
      "auxiliary_loss_clip": 0.01077549,
      "auxiliary_loss_mlp": 0.01047554,
      "balance_loss_clip": 1.04179156,
      "balance_loss_mlp": 1.03175879,
      "epoch": 0.4937321509093642,
      "flos": 26358935132160.0,
      "grad_norm": 2.3032584658075077,
      "language_loss": 0.9135769,
      "learning_rate": 2.137565999700933e-06,
      "loss": 0.93482792,
      "num_input_tokens_seen": 176572185,
      "step": 8212,
      "time_per_iteration": 2.6023058891296387
    },
    {
      "auxiliary_loss_clip": 0.01076205,
      "auxiliary_loss_mlp": 0.01043381,
      "balance_loss_clip": 1.04086995,
      "balance_loss_mlp": 1.02892709,
      "epoch": 0.49379227416203214,
      "flos": 22961390469120.0,
      "grad_norm": 1.7418930394709895,
      "language_loss": 0.65006077,
      "learning_rate": 2.1371774577972138e-06,
      "loss": 0.6712566,
      "num_input_tokens_seen": 176591490,
      "step": 8213,
      "time_per_iteration": 2.5764567852020264
    },
    {
      "auxiliary_loss_clip": 0.01068894,
      "auxiliary_loss_mlp": 0.00800115,
      "balance_loss_clip": 1.03929889,
      "balance_loss_mlp": 1.03407586,
      "epoch": 0.49385239741470016,
      "flos": 32489101843200.0,
      "grad_norm": 1.7439613070307782,
      "language_loss": 0.75564373,
      "learning_rate": 2.136788910691711e-06,
      "loss": 0.77433378,
      "num_input_tokens_seen": 176612715,
      "step": 8214,
      "time_per_iteration": 2.7135825157165527
    },
    {
      "auxiliary_loss_clip": 0.01119665,
      "auxiliary_loss_mlp": 0.01038105,
      "balance_loss_clip": 1.04347944,
      "balance_loss_mlp": 1.02460456,
      "epoch": 0.4939125206673681,
      "flos": 22492864442880.0,
      "grad_norm": 1.7723785978564133,
      "language_loss": 0.8471694,
      "learning_rate": 2.1364003583991594e-06,
      "loss": 0.86874712,
      "num_input_tokens_seen": 176631950,
      "step": 8215,
      "time_per_iteration": 2.4965803623199463
    },
    {
      "auxiliary_loss_clip": 0.01100935,
      "auxiliary_loss_mlp": 0.01031312,
      "balance_loss_clip": 1.03995395,
      "balance_loss_mlp": 1.01974249,
      "epoch": 0.4939726439200361,
      "flos": 31176351486720.0,
      "grad_norm": 1.5841864041273628,
      "language_loss": 0.83359343,
      "learning_rate": 2.136011800934292e-06,
      "loss": 0.85491592,
      "num_input_tokens_seen": 176653060,
      "step": 8216,
      "time_per_iteration": 2.5763843059539795
    },
    {
      "auxiliary_loss_clip": 0.01089987,
      "auxiliary_loss_mlp": 0.01031398,
      "balance_loss_clip": 1.0424757,
      "balance_loss_mlp": 1.01866615,
      "epoch": 0.49403276717270406,
      "flos": 22674213233280.0,
      "grad_norm": 1.5605228003675826,
      "language_loss": 0.74527383,
      "learning_rate": 2.1356232383118442e-06,
      "loss": 0.76648772,
      "num_input_tokens_seen": 176673895,
      "step": 8217,
      "time_per_iteration": 2.5487473011016846
    },
    {
      "auxiliary_loss_clip": 0.01115838,
      "auxiliary_loss_mlp": 0.00795905,
      "balance_loss_clip": 1.04222,
      "balance_loss_mlp": 1.02696741,
      "epoch": 0.494092890425372,
      "flos": 20741070147840.0,
      "grad_norm": 1.552105665876363,
      "language_loss": 0.78706884,
      "learning_rate": 2.1352346705465494e-06,
      "loss": 0.80618632,
      "num_input_tokens_seen": 176692550,
      "step": 8218,
      "time_per_iteration": 2.4653639793395996
    },
    {
      "auxiliary_loss_clip": 0.01068799,
      "auxiliary_loss_mlp": 0.00790134,
      "balance_loss_clip": 1.04076374,
      "balance_loss_mlp": 1.01710963,
      "epoch": 0.49415301367804,
      "flos": 18369026778240.0,
      "grad_norm": 2.0003490260863708,
      "language_loss": 0.76008332,
      "learning_rate": 2.134846097653142e-06,
      "loss": 0.77867264,
      "num_input_tokens_seen": 176709335,
      "step": 8219,
      "time_per_iteration": 2.5492849349975586
    },
    {
      "auxiliary_loss_clip": 0.01094751,
      "auxiliary_loss_mlp": 0.0103274,
      "balance_loss_clip": 1.04090226,
      "balance_loss_mlp": 1.01978827,
      "epoch": 0.49421313693070795,
      "flos": 17530620451200.0,
      "grad_norm": 2.0574417411977017,
      "language_loss": 0.62444079,
      "learning_rate": 2.134457519646357e-06,
      "loss": 0.64571571,
      "num_input_tokens_seen": 176727715,
      "step": 8220,
      "time_per_iteration": 2.486302137374878
    },
    {
      "auxiliary_loss_clip": 0.0111588,
      "auxiliary_loss_mlp": 0.0103041,
      "balance_loss_clip": 1.04026604,
      "balance_loss_mlp": 1.01790547,
      "epoch": 0.4942732601833759,
      "flos": 20812173120000.0,
      "grad_norm": 1.9065258085619627,
      "language_loss": 0.72104102,
      "learning_rate": 2.1340689365409296e-06,
      "loss": 0.74250388,
      "num_input_tokens_seen": 176747530,
      "step": 8221,
      "time_per_iteration": 2.4592385292053223
    },
    {
      "auxiliary_loss_clip": 0.01082769,
      "auxiliary_loss_mlp": 0.0103572,
      "balance_loss_clip": 1.04552722,
      "balance_loss_mlp": 1.02322054,
      "epoch": 0.4943333834360439,
      "flos": 15048941794560.0,
      "grad_norm": 1.6278659193754357,
      "language_loss": 0.79071248,
      "learning_rate": 2.133680348351595e-06,
      "loss": 0.8118974,
      "num_input_tokens_seen": 176765260,
      "step": 8222,
      "time_per_iteration": 2.5291025638580322
    },
    {
      "auxiliary_loss_clip": 0.01105867,
      "auxiliary_loss_mlp": 0.01038239,
      "balance_loss_clip": 1.04186487,
      "balance_loss_mlp": 1.02523899,
      "epoch": 0.49439350668871185,
      "flos": 16070420764800.0,
      "grad_norm": 2.229720205338952,
      "language_loss": 0.72220767,
      "learning_rate": 2.133291755093088e-06,
      "loss": 0.74364877,
      "num_input_tokens_seen": 176781770,
      "step": 8223,
      "time_per_iteration": 2.44869327545166
    },
    {
      "auxiliary_loss_clip": 0.01108918,
      "auxiliary_loss_mlp": 0.01038011,
      "balance_loss_clip": 1.04276705,
      "balance_loss_mlp": 1.02463543,
      "epoch": 0.4944536299413798,
      "flos": 20880079781760.0,
      "grad_norm": 1.8686341889274716,
      "language_loss": 0.75132382,
      "learning_rate": 2.132903156780144e-06,
      "loss": 0.77279305,
      "num_input_tokens_seen": 176800655,
      "step": 8224,
      "time_per_iteration": 3.846817970275879
    },
    {
      "auxiliary_loss_clip": 0.0110344,
      "auxiliary_loss_mlp": 0.01029758,
      "balance_loss_clip": 1.04538703,
      "balance_loss_mlp": 1.01722884,
      "epoch": 0.4945137531940478,
      "flos": 26608908856320.0,
      "grad_norm": 2.0135140411923,
      "language_loss": 0.63577646,
      "learning_rate": 2.1325145534274997e-06,
      "loss": 0.65710843,
      "num_input_tokens_seen": 176820610,
      "step": 8225,
      "time_per_iteration": 3.9198665618896484
    },
    {
      "auxiliary_loss_clip": 0.01098277,
      "auxiliary_loss_mlp": 0.01033758,
      "balance_loss_clip": 1.04285073,
      "balance_loss_mlp": 1.02075863,
      "epoch": 0.49457387644671574,
      "flos": 23988148738560.0,
      "grad_norm": 2.9966261578870705,
      "language_loss": 0.76168728,
      "learning_rate": 2.1321259450498893e-06,
      "loss": 0.78300762,
      "num_input_tokens_seen": 176840520,
      "step": 8226,
      "time_per_iteration": 2.5569581985473633
    },
    {
      "auxiliary_loss_clip": 0.0111888,
      "auxiliary_loss_mlp": 0.01037322,
      "balance_loss_clip": 1.04131126,
      "balance_loss_mlp": 1.02280235,
      "epoch": 0.49463399969938376,
      "flos": 26976598427520.0,
      "grad_norm": 1.5930431820923325,
      "language_loss": 0.7089982,
      "learning_rate": 2.131737331662051e-06,
      "loss": 0.73056024,
      "num_input_tokens_seen": 176860265,
      "step": 8227,
      "time_per_iteration": 2.498302459716797
    },
    {
      "auxiliary_loss_clip": 0.01099905,
      "auxiliary_loss_mlp": 0.01033971,
      "balance_loss_clip": 1.04320192,
      "balance_loss_mlp": 1.02070892,
      "epoch": 0.49469412295205173,
      "flos": 29681534067840.0,
      "grad_norm": 2.151800891737294,
      "language_loss": 0.71889138,
      "learning_rate": 2.131348713278718e-06,
      "loss": 0.74023014,
      "num_input_tokens_seen": 176882910,
      "step": 8228,
      "time_per_iteration": 4.022500991821289
    },
    {
      "auxiliary_loss_clip": 0.01116246,
      "auxiliary_loss_mlp": 0.01027738,
      "balance_loss_clip": 1.041942,
      "balance_loss_mlp": 1.01506603,
      "epoch": 0.4947542462047197,
      "flos": 24131791226880.0,
      "grad_norm": 1.4343638101702127,
      "language_loss": 0.83861768,
      "learning_rate": 2.1309600899146304e-06,
      "loss": 0.86005753,
      "num_input_tokens_seen": 176903030,
      "step": 8229,
      "time_per_iteration": 2.493532180786133
    },
    {
      "auxiliary_loss_clip": 0.01106864,
      "auxiliary_loss_mlp": 0.01035363,
      "balance_loss_clip": 1.03864026,
      "balance_loss_mlp": 1.02078938,
      "epoch": 0.49481436945738766,
      "flos": 20045049333120.0,
      "grad_norm": 1.9470988235146696,
      "language_loss": 0.7460494,
      "learning_rate": 2.1305714615845227e-06,
      "loss": 0.76747167,
      "num_input_tokens_seen": 176919025,
      "step": 8230,
      "time_per_iteration": 2.491112470626831
    },
    {
      "auxiliary_loss_clip": 0.01102112,
      "auxiliary_loss_mlp": 0.01030215,
      "balance_loss_clip": 1.04261851,
      "balance_loss_mlp": 1.01777506,
      "epoch": 0.4948744927100556,
      "flos": 15669550005120.0,
      "grad_norm": 2.7357266854396025,
      "language_loss": 0.79139841,
      "learning_rate": 2.1301828283031314e-06,
      "loss": 0.81272173,
      "num_input_tokens_seen": 176937945,
      "step": 8231,
      "time_per_iteration": 3.9317104816436768
    },
    {
      "auxiliary_loss_clip": 0.01038907,
      "auxiliary_loss_mlp": 0.01005086,
      "balance_loss_clip": 1.02473545,
      "balance_loss_mlp": 1.00364971,
      "epoch": 0.4949346159627236,
      "flos": 68872071502080.0,
      "grad_norm": 0.7567164058654038,
      "language_loss": 0.60184014,
      "learning_rate": 2.1297941900851944e-06,
      "loss": 0.62228006,
      "num_input_tokens_seen": 177004575,
      "step": 8232,
      "time_per_iteration": 3.226020097732544
    },
    {
      "auxiliary_loss_clip": 0.01098705,
      "auxiliary_loss_mlp": 0.01034017,
      "balance_loss_clip": 1.04238272,
      "balance_loss_mlp": 1.01984274,
      "epoch": 0.49499473921539155,
      "flos": 24790285307520.0,
      "grad_norm": 3.2247624155933683,
      "language_loss": 0.69048202,
      "learning_rate": 2.1294055469454496e-06,
      "loss": 0.71180922,
      "num_input_tokens_seen": 177024155,
      "step": 8233,
      "time_per_iteration": 2.5669918060302734
    },
    {
      "auxiliary_loss_clip": 0.01065076,
      "auxiliary_loss_mlp": 0.01034137,
      "balance_loss_clip": 1.04140592,
      "balance_loss_mlp": 1.01989174,
      "epoch": 0.4950548624680595,
      "flos": 32707905540480.0,
      "grad_norm": 2.1147745658915196,
      "language_loss": 0.66628325,
      "learning_rate": 2.129016898898633e-06,
      "loss": 0.68727541,
      "num_input_tokens_seen": 177046185,
      "step": 8234,
      "time_per_iteration": 2.661149501800537
    },
    {
      "auxiliary_loss_clip": 0.0103074,
      "auxiliary_loss_mlp": 0.01009624,
      "balance_loss_clip": 1.02841949,
      "balance_loss_mlp": 1.00834298,
      "epoch": 0.4951149857207275,
      "flos": 50082173066880.0,
      "grad_norm": 0.7996729967131923,
      "language_loss": 0.58016485,
      "learning_rate": 2.128628245959482e-06,
      "loss": 0.60056847,
      "num_input_tokens_seen": 177099025,
      "step": 8235,
      "time_per_iteration": 3.056988000869751
    },
    {
      "auxiliary_loss_clip": 0.01085857,
      "auxiliary_loss_mlp": 0.01034639,
      "balance_loss_clip": 1.03782916,
      "balance_loss_mlp": 1.02045369,
      "epoch": 0.49517510897339545,
      "flos": 22236785406720.0,
      "grad_norm": 1.6370835549888947,
      "language_loss": 0.76989377,
      "learning_rate": 2.1282395881427355e-06,
      "loss": 0.79109871,
      "num_input_tokens_seen": 177118365,
      "step": 8236,
      "time_per_iteration": 2.55275559425354
    },
    {
      "auxiliary_loss_clip": 0.01077641,
      "auxiliary_loss_mlp": 0.01030353,
      "balance_loss_clip": 1.04453015,
      "balance_loss_mlp": 1.01763356,
      "epoch": 0.4952352322260634,
      "flos": 25374120969600.0,
      "grad_norm": 1.7194510867408521,
      "language_loss": 0.72785258,
      "learning_rate": 2.1278509254631315e-06,
      "loss": 0.74893248,
      "num_input_tokens_seen": 177136415,
      "step": 8237,
      "time_per_iteration": 2.5933873653411865
    },
    {
      "auxiliary_loss_clip": 0.01115543,
      "auxiliary_loss_mlp": 0.0103567,
      "balance_loss_clip": 1.0417732,
      "balance_loss_mlp": 1.0222466,
      "epoch": 0.4952953554787314,
      "flos": 24608721035520.0,
      "grad_norm": 1.9113703578944181,
      "language_loss": 0.75778323,
      "learning_rate": 2.127462257935406e-06,
      "loss": 0.77929538,
      "num_input_tokens_seen": 177155690,
      "step": 8238,
      "time_per_iteration": 2.4981977939605713
    },
    {
      "auxiliary_loss_clip": 0.01075034,
      "auxiliary_loss_mlp": 0.01039101,
      "balance_loss_clip": 1.041363,
      "balance_loss_mlp": 1.02354991,
      "epoch": 0.49535547873139935,
      "flos": 17311278049920.0,
      "grad_norm": 2.1417790602362508,
      "language_loss": 0.74613726,
      "learning_rate": 2.1270735855743008e-06,
      "loss": 0.76727867,
      "num_input_tokens_seen": 177173350,
      "step": 8239,
      "time_per_iteration": 2.5486977100372314
    },
    {
      "auxiliary_loss_clip": 0.01030846,
      "auxiliary_loss_mlp": 0.01035378,
      "balance_loss_clip": 1.0373894,
      "balance_loss_mlp": 1.0202384,
      "epoch": 0.4954156019840673,
      "flos": 20740315962240.0,
      "grad_norm": 2.4496463608147914,
      "language_loss": 0.78849626,
      "learning_rate": 2.126684908394552e-06,
      "loss": 0.80915844,
      "num_input_tokens_seen": 177191115,
      "step": 8240,
      "time_per_iteration": 2.8489527702331543
    },
    {
      "auxiliary_loss_clip": 0.01104155,
      "auxiliary_loss_mlp": 0.01038411,
      "balance_loss_clip": 1.04082799,
      "balance_loss_mlp": 1.02581072,
      "epoch": 0.49547572523673533,
      "flos": 12820684567680.0,
      "grad_norm": 2.114649149884295,
      "language_loss": 0.85849869,
      "learning_rate": 2.126296226410898e-06,
      "loss": 0.87992436,
      "num_input_tokens_seen": 177206155,
      "step": 8241,
      "time_per_iteration": 2.802905797958374
    },
    {
      "auxiliary_loss_clip": 0.01065379,
      "auxiliary_loss_mlp": 0.01029922,
      "balance_loss_clip": 1.04510927,
      "balance_loss_mlp": 1.01695204,
      "epoch": 0.4955358484894033,
      "flos": 15597046402560.0,
      "grad_norm": 3.556286279935567,
      "language_loss": 0.7700932,
      "learning_rate": 2.1259075396380794e-06,
      "loss": 0.7910462,
      "num_input_tokens_seen": 177224815,
      "step": 8242,
      "time_per_iteration": 2.577310562133789
    },
    {
      "auxiliary_loss_clip": 0.01094192,
      "auxiliary_loss_mlp": 0.00788736,
      "balance_loss_clip": 1.04036832,
      "balance_loss_mlp": 1.01332009,
      "epoch": 0.49559597174207126,
      "flos": 26464368528000.0,
      "grad_norm": 1.7650318253653374,
      "language_loss": 0.67253149,
      "learning_rate": 2.125518848090833e-06,
      "loss": 0.69136077,
      "num_input_tokens_seen": 177244490,
      "step": 8243,
      "time_per_iteration": 2.5766868591308594
    },
    {
      "auxiliary_loss_clip": 0.01102861,
      "auxiliary_loss_mlp": 0.010278,
      "balance_loss_clip": 1.04324484,
      "balance_loss_mlp": 1.01559317,
      "epoch": 0.4956560949947392,
      "flos": 23148234040320.0,
      "grad_norm": 1.7879303114443938,
      "language_loss": 0.68124151,
      "learning_rate": 2.125130151783901e-06,
      "loss": 0.70254815,
      "num_input_tokens_seen": 177264340,
      "step": 8244,
      "time_per_iteration": 2.5274124145507812
    },
    {
      "auxiliary_loss_clip": 0.01097277,
      "auxiliary_loss_mlp": 0.0103436,
      "balance_loss_clip": 1.0423367,
      "balance_loss_mlp": 1.01994181,
      "epoch": 0.4957162182474072,
      "flos": 20773461237120.0,
      "grad_norm": 2.0981901438317774,
      "language_loss": 0.75122911,
      "learning_rate": 2.12474145073202e-06,
      "loss": 0.77254546,
      "num_input_tokens_seen": 177283055,
      "step": 8245,
      "time_per_iteration": 2.514234781265259
    },
    {
      "auxiliary_loss_clip": 0.0110625,
      "auxiliary_loss_mlp": 0.01029418,
      "balance_loss_clip": 1.04293156,
      "balance_loss_mlp": 1.01653767,
      "epoch": 0.49577634150007516,
      "flos": 18734202397440.0,
      "grad_norm": 1.749115349064246,
      "language_loss": 0.8133688,
      "learning_rate": 2.1243527449499306e-06,
      "loss": 0.83472544,
      "num_input_tokens_seen": 177301140,
      "step": 8246,
      "time_per_iteration": 2.473923683166504
    },
    {
      "auxiliary_loss_clip": 0.01082053,
      "auxiliary_loss_mlp": 0.01036361,
      "balance_loss_clip": 1.04619312,
      "balance_loss_mlp": 1.02149582,
      "epoch": 0.4958364647527431,
      "flos": 25554176870400.0,
      "grad_norm": 1.6925442231089483,
      "language_loss": 0.83501804,
      "learning_rate": 2.1239640344523733e-06,
      "loss": 0.85620213,
      "num_input_tokens_seen": 177323095,
      "step": 8247,
      "time_per_iteration": 2.58484148979187
    },
    {
      "auxiliary_loss_clip": 0.01087246,
      "auxiliary_loss_mlp": 0.01028547,
      "balance_loss_clip": 1.04121399,
      "balance_loss_mlp": 1.01589286,
      "epoch": 0.4958965880054111,
      "flos": 24425325169920.0,
      "grad_norm": 2.0322335961317433,
      "language_loss": 0.83493853,
      "learning_rate": 2.123575319254087e-06,
      "loss": 0.85609651,
      "num_input_tokens_seen": 177339845,
      "step": 8248,
      "time_per_iteration": 2.5696663856506348
    },
    {
      "auxiliary_loss_clip": 0.01108867,
      "auxiliary_loss_mlp": 0.01027978,
      "balance_loss_clip": 1.04189014,
      "balance_loss_mlp": 1.01417375,
      "epoch": 0.49595671125807905,
      "flos": 25083460114560.0,
      "grad_norm": 2.169301649188192,
      "language_loss": 0.73727107,
      "learning_rate": 2.123186599369812e-06,
      "loss": 0.75863945,
      "num_input_tokens_seen": 177359980,
      "step": 8249,
      "time_per_iteration": 2.529914617538452
    },
    {
      "auxiliary_loss_clip": 0.01098777,
      "auxiliary_loss_mlp": 0.01042292,
      "balance_loss_clip": 1.04141212,
      "balance_loss_mlp": 1.02838051,
      "epoch": 0.496016834510747,
      "flos": 16435883692800.0,
      "grad_norm": 1.6690860858099164,
      "language_loss": 0.75954348,
      "learning_rate": 2.122797874814289e-06,
      "loss": 0.78095424,
      "num_input_tokens_seen": 177378580,
      "step": 8250,
      "time_per_iteration": 2.5000510215759277
    },
    {
      "auxiliary_loss_clip": 0.01120342,
      "auxiliary_loss_mlp": 0.01034071,
      "balance_loss_clip": 1.04261708,
      "balance_loss_mlp": 1.02016485,
      "epoch": 0.496076957763415,
      "flos": 23437925228160.0,
      "grad_norm": 1.8884236459554091,
      "language_loss": 0.70307249,
      "learning_rate": 2.1224091456022585e-06,
      "loss": 0.72461665,
      "num_input_tokens_seen": 177398790,
      "step": 8251,
      "time_per_iteration": 2.4961278438568115
    },
    {
      "auxiliary_loss_clip": 0.01081991,
      "auxiliary_loss_mlp": 0.00787999,
      "balance_loss_clip": 1.04330492,
      "balance_loss_mlp": 1.0115304,
      "epoch": 0.49613708101608295,
      "flos": 16909509450240.0,
      "grad_norm": 2.0339657502628836,
      "language_loss": 0.80320144,
      "learning_rate": 2.122020411748461e-06,
      "loss": 0.82190132,
      "num_input_tokens_seen": 177416515,
      "step": 8252,
      "time_per_iteration": 2.5906784534454346
    },
    {
      "auxiliary_loss_clip": 0.01119823,
      "auxiliary_loss_mlp": 0.01032084,
      "balance_loss_clip": 1.04259717,
      "balance_loss_mlp": 1.01668859,
      "epoch": 0.4961972042687509,
      "flos": 16618094409600.0,
      "grad_norm": 1.733846591994071,
      "language_loss": 0.80749488,
      "learning_rate": 2.1216316732676363e-06,
      "loss": 0.829014,
      "num_input_tokens_seen": 177434425,
      "step": 8253,
      "time_per_iteration": 2.4527575969696045
    },
    {
      "auxiliary_loss_clip": 0.01084869,
      "auxiliary_loss_mlp": 0.0103154,
      "balance_loss_clip": 1.03841114,
      "balance_loss_mlp": 1.018659,
      "epoch": 0.49625732752141893,
      "flos": 28956749437440.0,
      "grad_norm": 1.4359889427262444,
      "language_loss": 0.67382818,
      "learning_rate": 2.1212429301745275e-06,
      "loss": 0.6949923,
      "num_input_tokens_seen": 177459675,
      "step": 8254,
      "time_per_iteration": 2.663331985473633
    },
    {
      "auxiliary_loss_clip": 0.01081584,
      "auxiliary_loss_mlp": 0.01045159,
      "balance_loss_clip": 1.04036856,
      "balance_loss_mlp": 1.02945924,
      "epoch": 0.4963174507740869,
      "flos": 23112359331840.0,
      "grad_norm": 1.999028083230742,
      "language_loss": 0.74119014,
      "learning_rate": 2.1208541824838743e-06,
      "loss": 0.76245761,
      "num_input_tokens_seen": 177478895,
      "step": 8255,
      "time_per_iteration": 2.5629055500030518
    },
    {
      "auxiliary_loss_clip": 0.01088395,
      "auxiliary_loss_mlp": 0.0103663,
      "balance_loss_clip": 1.0406096,
      "balance_loss_mlp": 1.02315342,
      "epoch": 0.49637757402675486,
      "flos": 13917863450880.0,
      "grad_norm": 1.8298861570806362,
      "language_loss": 0.81195819,
      "learning_rate": 2.1204654302104183e-06,
      "loss": 0.83320832,
      "num_input_tokens_seen": 177494920,
      "step": 8256,
      "time_per_iteration": 2.4840705394744873
    },
    {
      "auxiliary_loss_clip": 0.01091504,
      "auxiliary_loss_mlp": 0.01026729,
      "balance_loss_clip": 1.04250872,
      "balance_loss_mlp": 1.01419449,
      "epoch": 0.49643769727942283,
      "flos": 22309001700480.0,
      "grad_norm": 1.4316603479984693,
      "language_loss": 0.80890715,
      "learning_rate": 2.120076673368901e-06,
      "loss": 0.83008957,
      "num_input_tokens_seen": 177515455,
      "step": 8257,
      "time_per_iteration": 2.5427818298339844
    },
    {
      "auxiliary_loss_clip": 0.01121028,
      "auxiliary_loss_mlp": 0.01040869,
      "balance_loss_clip": 1.0410006,
      "balance_loss_mlp": 1.02659369,
      "epoch": 0.4964978205320908,
      "flos": 19500248776320.0,
      "grad_norm": 1.9434309262960543,
      "language_loss": 0.6604377,
      "learning_rate": 2.1196879119740647e-06,
      "loss": 0.68205667,
      "num_input_tokens_seen": 177534040,
      "step": 8258,
      "time_per_iteration": 2.4589219093322754
    },
    {
      "auxiliary_loss_clip": 0.01103835,
      "auxiliary_loss_mlp": 0.01029435,
      "balance_loss_clip": 1.04080963,
      "balance_loss_mlp": 1.01714468,
      "epoch": 0.49655794378475876,
      "flos": 23436524597760.0,
      "grad_norm": 1.504772974661805,
      "language_loss": 0.77591479,
      "learning_rate": 2.1192991460406502e-06,
      "loss": 0.79724747,
      "num_input_tokens_seen": 177554510,
      "step": 8259,
      "time_per_iteration": 2.508404016494751
    },
    {
      "auxiliary_loss_clip": 0.01091971,
      "auxiliary_loss_mlp": 0.01034998,
      "balance_loss_clip": 1.04319739,
      "balance_loss_mlp": 1.02153969,
      "epoch": 0.4966180670374267,
      "flos": 26831124345600.0,
      "grad_norm": 1.4535594909619085,
      "language_loss": 0.78662145,
      "learning_rate": 2.1189103755834e-06,
      "loss": 0.80789113,
      "num_input_tokens_seen": 177575780,
      "step": 8260,
      "time_per_iteration": 2.592573642730713
    },
    {
      "auxiliary_loss_clip": 0.01096943,
      "auxiliary_loss_mlp": 0.01035081,
      "balance_loss_clip": 1.04222417,
      "balance_loss_mlp": 1.02107358,
      "epoch": 0.4966781902900947,
      "flos": 22009326531840.0,
      "grad_norm": 2.699834418626499,
      "language_loss": 0.76301008,
      "learning_rate": 2.1185216006170573e-06,
      "loss": 0.78433037,
      "num_input_tokens_seen": 177588965,
      "step": 8261,
      "time_per_iteration": 2.4924728870391846
    },
    {
      "auxiliary_loss_clip": 0.01068778,
      "auxiliary_loss_mlp": 0.01036501,
      "balance_loss_clip": 1.04025197,
      "balance_loss_mlp": 1.02340627,
      "epoch": 0.49673831354276266,
      "flos": 26213353309440.0,
      "grad_norm": 1.7672741715571316,
      "language_loss": 0.89215088,
      "learning_rate": 2.1181328211563627e-06,
      "loss": 0.91320366,
      "num_input_tokens_seen": 177608425,
      "step": 8262,
      "time_per_iteration": 3.9917151927948
    },
    {
      "auxiliary_loss_clip": 0.01062561,
      "auxiliary_loss_mlp": 0.01030536,
      "balance_loss_clip": 1.03879166,
      "balance_loss_mlp": 1.01782274,
      "epoch": 0.4967984367954306,
      "flos": 23182277155200.0,
      "grad_norm": 1.4595109239297823,
      "language_loss": 0.74021685,
      "learning_rate": 2.11774403721606e-06,
      "loss": 0.76114786,
      "num_input_tokens_seen": 177628240,
      "step": 8263,
      "time_per_iteration": 2.641376256942749
    },
    {
      "auxiliary_loss_clip": 0.01078911,
      "auxiliary_loss_mlp": 0.01036501,
      "balance_loss_clip": 1.04718328,
      "balance_loss_mlp": 1.02121258,
      "epoch": 0.4968585600480986,
      "flos": 19281445079040.0,
      "grad_norm": 1.8564942184666366,
      "language_loss": 0.69077015,
      "learning_rate": 2.1173552488108923e-06,
      "loss": 0.71192425,
      "num_input_tokens_seen": 177645920,
      "step": 8264,
      "time_per_iteration": 4.352736949920654
    },
    {
      "auxiliary_loss_clip": 0.01093772,
      "auxiliary_loss_mlp": 0.01029842,
      "balance_loss_clip": 1.04242849,
      "balance_loss_mlp": 1.01666331,
      "epoch": 0.49691868330076655,
      "flos": 22528703237760.0,
      "grad_norm": 1.5329417271442203,
      "language_loss": 0.64855707,
      "learning_rate": 2.1169664559556007e-06,
      "loss": 0.66979325,
      "num_input_tokens_seen": 177667185,
      "step": 8265,
      "time_per_iteration": 2.572495222091675
    },
    {
      "auxiliary_loss_clip": 0.0102586,
      "auxiliary_loss_mlp": 0.01014397,
      "balance_loss_clip": 1.02034211,
      "balance_loss_mlp": 1.01285958,
      "epoch": 0.4969788065534345,
      "flos": 66577128675840.0,
      "grad_norm": 0.8252788363779046,
      "language_loss": 0.53484285,
      "learning_rate": 2.1165776586649304e-06,
      "loss": 0.55524546,
      "num_input_tokens_seen": 177733020,
      "step": 8266,
      "time_per_iteration": 3.1974520683288574
    },
    {
      "auxiliary_loss_clip": 0.0110382,
      "auxiliary_loss_mlp": 0.01029441,
      "balance_loss_clip": 1.04078603,
      "balance_loss_mlp": 1.0168165,
      "epoch": 0.49703892980610254,
      "flos": 24059503105920.0,
      "grad_norm": 1.6195972177143436,
      "language_loss": 0.79405022,
      "learning_rate": 2.1161888569536223e-06,
      "loss": 0.81538278,
      "num_input_tokens_seen": 177753370,
      "step": 8267,
      "time_per_iteration": 3.937965154647827
    },
    {
      "auxiliary_loss_clip": 0.01095416,
      "auxiliary_loss_mlp": 0.01031467,
      "balance_loss_clip": 1.04329419,
      "balance_loss_mlp": 1.01759052,
      "epoch": 0.4970990530587705,
      "flos": 29126174912640.0,
      "grad_norm": 2.265842549311999,
      "language_loss": 0.74567807,
      "learning_rate": 2.1158000508364223e-06,
      "loss": 0.76694679,
      "num_input_tokens_seen": 177771530,
      "step": 8268,
      "time_per_iteration": 2.600353479385376
    },
    {
      "auxiliary_loss_clip": 0.01106834,
      "auxiliary_loss_mlp": 0.00785669,
      "balance_loss_clip": 1.04018569,
      "balance_loss_mlp": 1.00859439,
      "epoch": 0.49715917631143847,
      "flos": 46026167258880.0,
      "grad_norm": 1.3780759505113545,
      "language_loss": 0.67920923,
      "learning_rate": 2.115411240328073e-06,
      "loss": 0.69813424,
      "num_input_tokens_seen": 177796355,
      "step": 8269,
      "time_per_iteration": 4.095922946929932
    },
    {
      "auxiliary_loss_clip": 0.01090047,
      "auxiliary_loss_mlp": 0.01034709,
      "balance_loss_clip": 1.04288852,
      "balance_loss_mlp": 1.02124429,
      "epoch": 0.49721929956410643,
      "flos": 20191277600640.0,
      "grad_norm": 1.4635929748015757,
      "language_loss": 0.85351253,
      "learning_rate": 2.1150224254433167e-06,
      "loss": 0.87476003,
      "num_input_tokens_seen": 177814300,
      "step": 8270,
      "time_per_iteration": 2.533360004425049
    },
    {
      "auxiliary_loss_clip": 0.01074215,
      "auxiliary_loss_mlp": 0.00788274,
      "balance_loss_clip": 1.0408473,
      "balance_loss_mlp": 1.01257157,
      "epoch": 0.4972794228167744,
      "flos": 21653560275840.0,
      "grad_norm": 1.6923897967273798,
      "language_loss": 0.70878959,
      "learning_rate": 2.114633606196899e-06,
      "loss": 0.72741449,
      "num_input_tokens_seen": 177833615,
      "step": 8271,
      "time_per_iteration": 2.5946452617645264
    },
    {
      "auxiliary_loss_clip": 0.0110408,
      "auxiliary_loss_mlp": 0.01033471,
      "balance_loss_clip": 1.04351509,
      "balance_loss_mlp": 1.01978612,
      "epoch": 0.49733954606944236,
      "flos": 24279743347200.0,
      "grad_norm": 1.6867031908897805,
      "language_loss": 0.78347111,
      "learning_rate": 2.1142447826035635e-06,
      "loss": 0.80484653,
      "num_input_tokens_seen": 177855315,
      "step": 8272,
      "time_per_iteration": 2.518927574157715
    },
    {
      "auxiliary_loss_clip": 0.01089391,
      "auxiliary_loss_mlp": 0.01036757,
      "balance_loss_clip": 1.04304445,
      "balance_loss_mlp": 1.02369201,
      "epoch": 0.4973996693221103,
      "flos": 37852575730560.0,
      "grad_norm": 2.3337748567340157,
      "language_loss": 0.66458786,
      "learning_rate": 2.1138559546780544e-06,
      "loss": 0.68584931,
      "num_input_tokens_seen": 177875590,
      "step": 8273,
      "time_per_iteration": 2.6774890422821045
    },
    {
      "auxiliary_loss_clip": 0.01089566,
      "auxiliary_loss_mlp": 0.01031115,
      "balance_loss_clip": 1.0422008,
      "balance_loss_mlp": 1.01827633,
      "epoch": 0.4974597925747783,
      "flos": 21361426963200.0,
      "grad_norm": 1.6234890371802386,
      "language_loss": 0.78005719,
      "learning_rate": 2.1134671224351163e-06,
      "loss": 0.80126405,
      "num_input_tokens_seen": 177894175,
      "step": 8274,
      "time_per_iteration": 2.541865587234497
    },
    {
      "auxiliary_loss_clip": 0.01085264,
      "auxiliary_loss_mlp": 0.01034474,
      "balance_loss_clip": 1.04453993,
      "balance_loss_mlp": 1.02025282,
      "epoch": 0.49751991582744626,
      "flos": 30738133560960.0,
      "grad_norm": 1.7102443936000828,
      "language_loss": 0.75710166,
      "learning_rate": 2.113078285889493e-06,
      "loss": 0.77829909,
      "num_input_tokens_seen": 177913920,
      "step": 8275,
      "time_per_iteration": 2.6322953701019287
    },
    {
      "auxiliary_loss_clip": 0.01108286,
      "auxiliary_loss_mlp": 0.01035471,
      "balance_loss_clip": 1.04438305,
      "balance_loss_mlp": 1.02020073,
      "epoch": 0.4975800390801142,
      "flos": 14100541044480.0,
      "grad_norm": 1.9323262937798913,
      "language_loss": 0.83768767,
      "learning_rate": 2.1126894450559303e-06,
      "loss": 0.85912526,
      "num_input_tokens_seen": 177930425,
      "step": 8276,
      "time_per_iteration": 2.4601824283599854
    },
    {
      "auxiliary_loss_clip": 0.01112732,
      "auxiliary_loss_mlp": 0.00786133,
      "balance_loss_clip": 1.04028809,
      "balance_loss_mlp": 1.01058018,
      "epoch": 0.4976401623327822,
      "flos": 24207275658240.0,
      "grad_norm": 1.4279676044015661,
      "language_loss": 0.70351017,
      "learning_rate": 2.112300599949172e-06,
      "loss": 0.72249877,
      "num_input_tokens_seen": 177949885,
      "step": 8277,
      "time_per_iteration": 2.4973554611206055
    },
    {
      "auxiliary_loss_clip": 0.01103589,
      "auxiliary_loss_mlp": 0.0103556,
      "balance_loss_clip": 1.04526651,
      "balance_loss_mlp": 1.02201772,
      "epoch": 0.49770028558545015,
      "flos": 21136769349120.0,
      "grad_norm": 2.02878644567125,
      "language_loss": 0.82437861,
      "learning_rate": 2.111911750583964e-06,
      "loss": 0.84577006,
      "num_input_tokens_seen": 177965720,
      "step": 8278,
      "time_per_iteration": 2.4844093322753906
    },
    {
      "auxiliary_loss_clip": 0.01108557,
      "auxiliary_loss_mlp": 0.01037336,
      "balance_loss_clip": 1.04083514,
      "balance_loss_mlp": 1.02426457,
      "epoch": 0.4977604088381181,
      "flos": 16763927627520.0,
      "grad_norm": 2.066400570371332,
      "language_loss": 0.67696869,
      "learning_rate": 2.111522896975052e-06,
      "loss": 0.69842762,
      "num_input_tokens_seen": 177983190,
      "step": 8279,
      "time_per_iteration": 2.4708824157714844
    },
    {
      "auxiliary_loss_clip": 0.0110886,
      "auxiliary_loss_mlp": 0.01035812,
      "balance_loss_clip": 1.04022825,
      "balance_loss_mlp": 1.02154279,
      "epoch": 0.49782053209078614,
      "flos": 15703521292800.0,
      "grad_norm": 2.210821753885307,
      "language_loss": 0.71667689,
      "learning_rate": 2.1111340391371794e-06,
      "loss": 0.73812354,
      "num_input_tokens_seen": 178000155,
      "step": 8280,
      "time_per_iteration": 2.4534268379211426
    },
    {
      "auxiliary_loss_clip": 0.01085837,
      "auxiliary_loss_mlp": 0.01035687,
      "balance_loss_clip": 1.04127645,
      "balance_loss_mlp": 1.02195442,
      "epoch": 0.4978806553434541,
      "flos": 24753692327040.0,
      "grad_norm": 1.6253790667484762,
      "language_loss": 0.64472961,
      "learning_rate": 2.1107451770850936e-06,
      "loss": 0.66594487,
      "num_input_tokens_seen": 178021060,
      "step": 8281,
      "time_per_iteration": 2.580148458480835
    },
    {
      "auxiliary_loss_clip": 0.01112235,
      "auxiliary_loss_mlp": 0.01032598,
      "balance_loss_clip": 1.04339468,
      "balance_loss_mlp": 1.01897228,
      "epoch": 0.49794077859612207,
      "flos": 13115726881920.0,
      "grad_norm": 1.9004071881717206,
      "language_loss": 0.72860074,
      "learning_rate": 2.1103563108335387e-06,
      "loss": 0.75004911,
      "num_input_tokens_seen": 178038180,
      "step": 8282,
      "time_per_iteration": 2.4723243713378906
    },
    {
      "auxiliary_loss_clip": 0.01091799,
      "auxiliary_loss_mlp": 0.01031642,
      "balance_loss_clip": 1.04412222,
      "balance_loss_mlp": 1.02022791,
      "epoch": 0.49800090184879003,
      "flos": 27525133998720.0,
      "grad_norm": 1.4670804018170165,
      "language_loss": 0.72923625,
      "learning_rate": 2.109967440397263e-06,
      "loss": 0.75047064,
      "num_input_tokens_seen": 178057565,
      "step": 8283,
      "time_per_iteration": 2.58441424369812
    },
    {
      "auxiliary_loss_clip": 0.01063508,
      "auxiliary_loss_mlp": 0.01048211,
      "balance_loss_clip": 1.04073596,
      "balance_loss_mlp": 1.03232026,
      "epoch": 0.498061025101458,
      "flos": 19792489829760.0,
      "grad_norm": 1.7031671598654679,
      "language_loss": 0.7834301,
      "learning_rate": 2.1095785657910095e-06,
      "loss": 0.80454731,
      "num_input_tokens_seen": 178076965,
      "step": 8284,
      "time_per_iteration": 2.595801591873169
    },
    {
      "auxiliary_loss_clip": 0.0110144,
      "auxiliary_loss_mlp": 0.01038357,
      "balance_loss_clip": 1.04657888,
      "balance_loss_mlp": 1.02328861,
      "epoch": 0.49812114835412596,
      "flos": 29893909230720.0,
      "grad_norm": 2.182713384667605,
      "language_loss": 0.73730111,
      "learning_rate": 2.109189687029526e-06,
      "loss": 0.75869912,
      "num_input_tokens_seen": 178095105,
      "step": 8285,
      "time_per_iteration": 2.618523597717285
    },
    {
      "auxiliary_loss_clip": 0.01114204,
      "auxiliary_loss_mlp": 0.01033861,
      "balance_loss_clip": 1.04663348,
      "balance_loss_mlp": 1.02028275,
      "epoch": 0.49818127160679393,
      "flos": 23147048891520.0,
      "grad_norm": 1.619073990137544,
      "language_loss": 0.74184692,
      "learning_rate": 2.1088008041275598e-06,
      "loss": 0.76332754,
      "num_input_tokens_seen": 178114505,
      "step": 8286,
      "time_per_iteration": 2.4945077896118164
    },
    {
      "auxiliary_loss_clip": 0.0110234,
      "auxiliary_loss_mlp": 0.01044971,
      "balance_loss_clip": 1.0465827,
      "balance_loss_mlp": 1.03120792,
      "epoch": 0.4982413948594619,
      "flos": 21652806090240.0,
      "grad_norm": 1.6591269687158565,
      "language_loss": 0.85065866,
      "learning_rate": 2.1084119170998545e-06,
      "loss": 0.87213176,
      "num_input_tokens_seen": 178131595,
      "step": 8287,
      "time_per_iteration": 2.5278849601745605
    },
    {
      "auxiliary_loss_clip": 0.01077971,
      "auxiliary_loss_mlp": 0.01027204,
      "balance_loss_clip": 1.04207039,
      "balance_loss_mlp": 1.0136677,
      "epoch": 0.49830151811212986,
      "flos": 32486982940800.0,
      "grad_norm": 1.7161703889111457,
      "language_loss": 0.72539276,
      "learning_rate": 2.108023025961159e-06,
      "loss": 0.74644446,
      "num_input_tokens_seen": 178152055,
      "step": 8288,
      "time_per_iteration": 2.6799209117889404
    },
    {
      "auxiliary_loss_clip": 0.01102303,
      "auxiliary_loss_mlp": 0.01038378,
      "balance_loss_clip": 1.04235888,
      "balance_loss_mlp": 1.02329803,
      "epoch": 0.4983616413647978,
      "flos": 18142358002560.0,
      "grad_norm": 2.805373891426384,
      "language_loss": 0.80313712,
      "learning_rate": 2.10763413072622e-06,
      "loss": 0.82454395,
      "num_input_tokens_seen": 178168150,
      "step": 8289,
      "time_per_iteration": 2.511997938156128
    },
    {
      "auxiliary_loss_clip": 0.01108133,
      "auxiliary_loss_mlp": 0.01034449,
      "balance_loss_clip": 1.04140103,
      "balance_loss_mlp": 1.0211159,
      "epoch": 0.4984217646174658,
      "flos": 19718836992000.0,
      "grad_norm": 2.236405902196735,
      "language_loss": 0.73419756,
      "learning_rate": 2.107245231409784e-06,
      "loss": 0.75562346,
      "num_input_tokens_seen": 178186150,
      "step": 8290,
      "time_per_iteration": 2.5093703269958496
    },
    {
      "auxiliary_loss_clip": 0.01112265,
      "auxiliary_loss_mlp": 0.01035669,
      "balance_loss_clip": 1.04502344,
      "balance_loss_mlp": 1.01955199,
      "epoch": 0.49848188787013376,
      "flos": 24936549488640.0,
      "grad_norm": 1.5531516323638668,
      "language_loss": 0.838732,
      "learning_rate": 2.106856328026598e-06,
      "loss": 0.86021131,
      "num_input_tokens_seen": 178207665,
      "step": 8291,
      "time_per_iteration": 2.5451269149780273
    },
    {
      "auxiliary_loss_clip": 0.01100723,
      "auxiliary_loss_mlp": 0.01047171,
      "balance_loss_clip": 1.04451764,
      "balance_loss_mlp": 1.0304693,
      "epoch": 0.4985420111228017,
      "flos": 22382439056640.0,
      "grad_norm": 1.685407558999026,
      "language_loss": 0.66790563,
      "learning_rate": 2.106467420591409e-06,
      "loss": 0.68938458,
      "num_input_tokens_seen": 178226325,
      "step": 8292,
      "time_per_iteration": 2.5298144817352295
    },
    {
      "auxiliary_loss_clip": 0.01120679,
      "auxiliary_loss_mlp": 0.01033613,
      "balance_loss_clip": 1.04440463,
      "balance_loss_mlp": 1.02060699,
      "epoch": 0.4986021343754697,
      "flos": 16216469464320.0,
      "grad_norm": 1.6234432827980412,
      "language_loss": 0.66481757,
      "learning_rate": 2.106078509118965e-06,
      "loss": 0.68636048,
      "num_input_tokens_seen": 178244960,
      "step": 8293,
      "time_per_iteration": 2.448354721069336
    },
    {
      "auxiliary_loss_clip": 0.01107038,
      "auxiliary_loss_mlp": 0.01030512,
      "balance_loss_clip": 1.04348874,
      "balance_loss_mlp": 1.01693988,
      "epoch": 0.4986622576281377,
      "flos": 23403594804480.0,
      "grad_norm": 1.7356468875953197,
      "language_loss": 0.81552148,
      "learning_rate": 2.1056895936240133e-06,
      "loss": 0.83689696,
      "num_input_tokens_seen": 178265400,
      "step": 8294,
      "time_per_iteration": 2.5063512325286865
    },
    {
      "auxiliary_loss_clip": 0.01111318,
      "auxiliary_loss_mlp": 0.01033695,
      "balance_loss_clip": 1.04295301,
      "balance_loss_mlp": 1.01941395,
      "epoch": 0.49872238088080567,
      "flos": 19974556892160.0,
      "grad_norm": 1.7694458269328155,
      "language_loss": 0.73167044,
      "learning_rate": 2.1053006741213016e-06,
      "loss": 0.75312054,
      "num_input_tokens_seen": 178284535,
      "step": 8295,
      "time_per_iteration": 2.5252130031585693
    },
    {
      "auxiliary_loss_clip": 0.01066127,
      "auxiliary_loss_mlp": 0.01036102,
      "balance_loss_clip": 1.04441404,
      "balance_loss_mlp": 1.02288163,
      "epoch": 0.49878250413347364,
      "flos": 22893016930560.0,
      "grad_norm": 1.8343158368498729,
      "language_loss": 0.67166209,
      "learning_rate": 2.1049117506255775e-06,
      "loss": 0.69268435,
      "num_input_tokens_seen": 178302425,
      "step": 8296,
      "time_per_iteration": 2.612426280975342
    },
    {
      "auxiliary_loss_clip": 0.0110577,
      "auxiliary_loss_mlp": 0.01035974,
      "balance_loss_clip": 1.04502773,
      "balance_loss_mlp": 1.02188945,
      "epoch": 0.4988426273861416,
      "flos": 32598449821440.0,
      "grad_norm": 1.8091622422287084,
      "language_loss": 0.64141661,
      "learning_rate": 2.1045228231515895e-06,
      "loss": 0.66283405,
      "num_input_tokens_seen": 178323065,
      "step": 8297,
      "time_per_iteration": 2.6240246295928955
    },
    {
      "auxiliary_loss_clip": 0.01072491,
      "auxiliary_loss_mlp": 0.0103525,
      "balance_loss_clip": 1.04259777,
      "balance_loss_mlp": 1.0231086,
      "epoch": 0.49890275063880957,
      "flos": 20923604087040.0,
      "grad_norm": 1.7107852096567888,
      "language_loss": 0.69397879,
      "learning_rate": 2.1041338917140857e-06,
      "loss": 0.71505618,
      "num_input_tokens_seen": 178343985,
      "step": 8298,
      "time_per_iteration": 2.6101129055023193
    },
    {
      "auxiliary_loss_clip": 0.01115646,
      "auxiliary_loss_mlp": 0.01034633,
      "balance_loss_clip": 1.04128814,
      "balance_loss_mlp": 1.02160978,
      "epoch": 0.49896287389147753,
      "flos": 18624459369600.0,
      "grad_norm": 1.773800760619909,
      "language_loss": 0.84174359,
      "learning_rate": 2.103744956327814e-06,
      "loss": 0.86324644,
      "num_input_tokens_seen": 178362345,
      "step": 8299,
      "time_per_iteration": 2.442152261734009
    },
    {
      "auxiliary_loss_clip": 0.01089915,
      "auxiliary_loss_mlp": 0.0103724,
      "balance_loss_clip": 1.04108739,
      "balance_loss_mlp": 1.02243388,
      "epoch": 0.4990229971441455,
      "flos": 24826555065600.0,
      "grad_norm": 2.087036896447266,
      "language_loss": 0.69062155,
      "learning_rate": 2.1033560170075234e-06,
      "loss": 0.71189308,
      "num_input_tokens_seen": 178383190,
      "step": 8300,
      "time_per_iteration": 3.952367067337036
    },
    {
      "auxiliary_loss_clip": 0.01049197,
      "auxiliary_loss_mlp": 0.00999463,
      "balance_loss_clip": 1.05027223,
      "balance_loss_mlp": 0.99795455,
      "epoch": 0.49908312039681346,
      "flos": 71384525136000.0,
      "grad_norm": 0.7753404980013431,
      "language_loss": 0.51178277,
      "learning_rate": 2.1029670737679623e-06,
      "loss": 0.53226936,
      "num_input_tokens_seen": 178444250,
      "step": 8301,
      "time_per_iteration": 3.2598118782043457
    },
    {
      "auxiliary_loss_clip": 0.01090249,
      "auxiliary_loss_mlp": 0.01038638,
      "balance_loss_clip": 1.04272902,
      "balance_loss_mlp": 1.02574003,
      "epoch": 0.4991432436494814,
      "flos": 19828651847040.0,
      "grad_norm": 1.7220747278854545,
      "language_loss": 0.84425759,
      "learning_rate": 2.102578126623879e-06,
      "loss": 0.86554646,
      "num_input_tokens_seen": 178463250,
      "step": 8302,
      "time_per_iteration": 3.940723419189453
    },
    {
      "auxiliary_loss_clip": 0.01104913,
      "auxiliary_loss_mlp": 0.01030328,
      "balance_loss_clip": 1.04245794,
      "balance_loss_mlp": 1.01770353,
      "epoch": 0.4992033669021494,
      "flos": 15121912273920.0,
      "grad_norm": 1.8053186726669859,
      "language_loss": 0.69153666,
      "learning_rate": 2.102189175590024e-06,
      "loss": 0.71288902,
      "num_input_tokens_seen": 178481340,
      "step": 8303,
      "time_per_iteration": 2.492274522781372
    },
    {
      "auxiliary_loss_clip": 0.01120545,
      "auxiliary_loss_mlp": 0.0103473,
      "balance_loss_clip": 1.04267526,
      "balance_loss_mlp": 1.0213424,
      "epoch": 0.49926349015481736,
      "flos": 31207952476800.0,
      "grad_norm": 1.6288999303678884,
      "language_loss": 0.72740132,
      "learning_rate": 2.101800220681144e-06,
      "loss": 0.74895406,
      "num_input_tokens_seen": 178501545,
      "step": 8304,
      "time_per_iteration": 2.5359036922454834
    },
    {
      "auxiliary_loss_clip": 0.01108181,
      "auxiliary_loss_mlp": 0.01035318,
      "balance_loss_clip": 1.04286361,
      "balance_loss_mlp": 1.02348065,
      "epoch": 0.4993236134074853,
      "flos": 24900207903360.0,
      "grad_norm": 2.5173084494061615,
      "language_loss": 0.80925357,
      "learning_rate": 2.10141126191199e-06,
      "loss": 0.83068854,
      "num_input_tokens_seen": 178519700,
      "step": 8305,
      "time_per_iteration": 3.968463182449341
    },
    {
      "auxiliary_loss_clip": 0.01023546,
      "auxiliary_loss_mlp": 0.01017304,
      "balance_loss_clip": 1.0263238,
      "balance_loss_mlp": 1.01593351,
      "epoch": 0.4993837366601533,
      "flos": 70420573797120.0,
      "grad_norm": 0.7159508786839746,
      "language_loss": 0.56861955,
      "learning_rate": 2.1010222992973107e-06,
      "loss": 0.589028,
      "num_input_tokens_seen": 178576740,
      "step": 8306,
      "time_per_iteration": 3.2563695907592773
    },
    {
      "auxiliary_loss_clip": 0.01120917,
      "auxiliary_loss_mlp": 0.01038373,
      "balance_loss_clip": 1.04466963,
      "balance_loss_mlp": 1.02397823,
      "epoch": 0.4994438599128213,
      "flos": 15961216440960.0,
      "grad_norm": 1.7746295249646513,
      "language_loss": 0.8258794,
      "learning_rate": 2.1006333328518556e-06,
      "loss": 0.84747231,
      "num_input_tokens_seen": 178594745,
      "step": 8307,
      "time_per_iteration": 2.4751482009887695
    },
    {
      "auxiliary_loss_clip": 0.01118389,
      "auxiliary_loss_mlp": 0.01035619,
      "balance_loss_clip": 1.04283535,
      "balance_loss_mlp": 1.02220178,
      "epoch": 0.4995039831654893,
      "flos": 27928303228800.0,
      "grad_norm": 1.8963119590418955,
      "language_loss": 0.6083709,
      "learning_rate": 2.1002443625903748e-06,
      "loss": 0.62991095,
      "num_input_tokens_seen": 178614110,
      "step": 8308,
      "time_per_iteration": 3.9243569374084473
    },
    {
      "auxiliary_loss_clip": 0.01114351,
      "auxiliary_loss_mlp": 0.01034349,
      "balance_loss_clip": 1.04029083,
      "balance_loss_mlp": 1.02159357,
      "epoch": 0.49956410641815724,
      "flos": 24204797619840.0,
      "grad_norm": 1.5392031226942395,
      "language_loss": 0.74820209,
      "learning_rate": 2.0998553885276168e-06,
      "loss": 0.76968908,
      "num_input_tokens_seen": 178634170,
      "step": 8309,
      "time_per_iteration": 2.479135751724243
    },
    {
      "auxiliary_loss_clip": 0.01096451,
      "auxiliary_loss_mlp": 0.01038101,
      "balance_loss_clip": 1.04047871,
      "balance_loss_mlp": 1.02501774,
      "epoch": 0.4996242296708252,
      "flos": 16180127879040.0,
      "grad_norm": 2.3759373225641776,
      "language_loss": 0.79886603,
      "learning_rate": 2.0994664106783335e-06,
      "loss": 0.82021153,
      "num_input_tokens_seen": 178651775,
      "step": 8310,
      "time_per_iteration": 2.5267388820648193
    },
    {
      "auxiliary_loss_clip": 0.01113241,
      "auxiliary_loss_mlp": 0.01038815,
      "balance_loss_clip": 1.0436089,
      "balance_loss_mlp": 1.02604151,
      "epoch": 0.49968435292349317,
      "flos": 16873527000960.0,
      "grad_norm": 1.4853859276698715,
      "language_loss": 0.70698142,
      "learning_rate": 2.0990774290572735e-06,
      "loss": 0.72850192,
      "num_input_tokens_seen": 178669720,
      "step": 8311,
      "time_per_iteration": 2.4571785926818848
    },
    {
      "auxiliary_loss_clip": 0.01094513,
      "auxiliary_loss_mlp": 0.01038758,
      "balance_loss_clip": 1.04648066,
      "balance_loss_mlp": 1.02627087,
      "epoch": 0.49974447617616113,
      "flos": 14939521989120.0,
      "grad_norm": 1.8909078889077193,
      "language_loss": 0.7707873,
      "learning_rate": 2.098688443679187e-06,
      "loss": 0.79211998,
      "num_input_tokens_seen": 178686765,
      "step": 8312,
      "time_per_iteration": 2.5197155475616455
    },
    {
      "auxiliary_loss_clip": 0.01088037,
      "auxiliary_loss_mlp": 0.01036255,
      "balance_loss_clip": 1.0415833,
      "balance_loss_mlp": 1.02322495,
      "epoch": 0.4998045994288291,
      "flos": 26651535321600.0,
      "grad_norm": 1.8073231747630303,
      "language_loss": 0.84653461,
      "learning_rate": 2.0982994545588256e-06,
      "loss": 0.86777747,
      "num_input_tokens_seen": 178705845,
      "step": 8313,
      "time_per_iteration": 2.6040706634521484
    },
    {
      "auxiliary_loss_clip": 0.01095167,
      "auxiliary_loss_mlp": 0.01034221,
      "balance_loss_clip": 1.04311907,
      "balance_loss_mlp": 1.0204227,
      "epoch": 0.49986472268149706,
      "flos": 20953768533120.0,
      "grad_norm": 1.9232233904885856,
      "language_loss": 0.80765975,
      "learning_rate": 2.097910461710939e-06,
      "loss": 0.82895362,
      "num_input_tokens_seen": 178723410,
      "step": 8314,
      "time_per_iteration": 2.550657272338867
    },
    {
      "auxiliary_loss_clip": 0.01089821,
      "auxiliary_loss_mlp": 0.00790799,
      "balance_loss_clip": 1.04377961,
      "balance_loss_mlp": 1.01382256,
      "epoch": 0.49992484593416503,
      "flos": 22783884433920.0,
      "grad_norm": 1.6976015436860743,
      "language_loss": 0.79403436,
      "learning_rate": 2.0975214651502773e-06,
      "loss": 0.81284052,
      "num_input_tokens_seen": 178743560,
      "step": 8315,
      "time_per_iteration": 2.575409412384033
    },
    {
      "auxiliary_loss_clip": 0.01118394,
      "auxiliary_loss_mlp": 0.01036156,
      "balance_loss_clip": 1.04334402,
      "balance_loss_mlp": 1.02368665,
      "epoch": 0.499984969186833,
      "flos": 46786970252160.0,
      "grad_norm": 1.8001269282836145,
      "language_loss": 0.74541283,
      "learning_rate": 2.0971324648915926e-06,
      "loss": 0.7669583,
      "num_input_tokens_seen": 178767225,
      "step": 8316,
      "time_per_iteration": 2.691697359085083
    },
    {
      "auxiliary_loss_clip": 0.01102825,
      "auxiliary_loss_mlp": 0.01031598,
      "balance_loss_clip": 1.04529154,
      "balance_loss_mlp": 1.01967108,
      "epoch": 0.500045092439501,
      "flos": 25556978131200.0,
      "grad_norm": 1.7634050928814375,
      "language_loss": 0.81316793,
      "learning_rate": 2.0967434609496343e-06,
      "loss": 0.83451211,
      "num_input_tokens_seen": 178786810,
      "step": 8317,
      "time_per_iteration": 2.5452218055725098
    },
    {
      "auxiliary_loss_clip": 0.01096502,
      "auxiliary_loss_mlp": 0.01034825,
      "balance_loss_clip": 1.04122663,
      "balance_loss_mlp": 1.02071631,
      "epoch": 0.5001052156921689,
      "flos": 20704764476160.0,
      "grad_norm": 2.1710839300990163,
      "language_loss": 0.83173037,
      "learning_rate": 2.0963544533391548e-06,
      "loss": 0.85304368,
      "num_input_tokens_seen": 178805660,
      "step": 8318,
      "time_per_iteration": 2.513848304748535
    },
    {
      "auxiliary_loss_clip": 0.01108307,
      "auxiliary_loss_mlp": 0.01035566,
      "balance_loss_clip": 1.04295683,
      "balance_loss_mlp": 1.02296591,
      "epoch": 0.500165338944837,
      "flos": 21251109317760.0,
      "grad_norm": 1.9303645935194018,
      "language_loss": 0.81663924,
      "learning_rate": 2.0959654420749045e-06,
      "loss": 0.83807796,
      "num_input_tokens_seen": 178824780,
      "step": 8319,
      "time_per_iteration": 2.516273021697998
    },
    {
      "auxiliary_loss_clip": 0.01077373,
      "auxiliary_loss_mlp": 0.01031205,
      "balance_loss_clip": 1.04015779,
      "balance_loss_mlp": 1.01819921,
      "epoch": 0.5002254621975049,
      "flos": 27854398995840.0,
      "grad_norm": 1.650030160563755,
      "language_loss": 0.71858674,
      "learning_rate": 2.095576427171635e-06,
      "loss": 0.73967254,
      "num_input_tokens_seen": 178845640,
      "step": 8320,
      "time_per_iteration": 2.601989269256592
    },
    {
      "auxiliary_loss_clip": 0.01092565,
      "auxiliary_loss_mlp": 0.01039989,
      "balance_loss_clip": 1.04522562,
      "balance_loss_mlp": 1.02591038,
      "epoch": 0.5002855854501729,
      "flos": 15551941898880.0,
      "grad_norm": 4.627201082572055,
      "language_loss": 0.76974654,
      "learning_rate": 2.0951874086440978e-06,
      "loss": 0.79107213,
      "num_input_tokens_seen": 178862290,
      "step": 8321,
      "time_per_iteration": 2.5370495319366455
    },
    {
      "auxiliary_loss_clip": 0.01108808,
      "auxiliary_loss_mlp": 0.00790283,
      "balance_loss_clip": 1.04253054,
      "balance_loss_mlp": 1.01375437,
      "epoch": 0.5003457087028408,
      "flos": 16107408794880.0,
      "grad_norm": 1.7035437152457817,
      "language_loss": 0.82761467,
      "learning_rate": 2.0947983865070455e-06,
      "loss": 0.84660554,
      "num_input_tokens_seen": 178879805,
      "step": 8322,
      "time_per_iteration": 2.463927745819092
    },
    {
      "auxiliary_loss_clip": 0.01110869,
      "auxiliary_loss_mlp": 0.01036983,
      "balance_loss_clip": 1.04285789,
      "balance_loss_mlp": 1.02383983,
      "epoch": 0.5004058319555088,
      "flos": 22710518904960.0,
      "grad_norm": 2.78765820846744,
      "language_loss": 0.73320967,
      "learning_rate": 2.094409360775228e-06,
      "loss": 0.75468826,
      "num_input_tokens_seen": 178896985,
      "step": 8323,
      "time_per_iteration": 2.50274395942688
    },
    {
      "auxiliary_loss_clip": 0.01082674,
      "auxiliary_loss_mlp": 0.01035041,
      "balance_loss_clip": 1.04422081,
      "balance_loss_mlp": 1.0217433,
      "epoch": 0.5004659552081767,
      "flos": 30117956313600.0,
      "grad_norm": 1.5467413601669846,
      "language_loss": 0.69575584,
      "learning_rate": 2.0940203314633977e-06,
      "loss": 0.71693289,
      "num_input_tokens_seen": 178920605,
      "step": 8324,
      "time_per_iteration": 2.627406120300293
    },
    {
      "auxiliary_loss_clip": 0.01104864,
      "auxiliary_loss_mlp": 0.00786575,
      "balance_loss_clip": 1.04323494,
      "balance_loss_mlp": 1.00875092,
      "epoch": 0.5005260784608447,
      "flos": 18624710764800.0,
      "grad_norm": 1.9134528655293435,
      "language_loss": 0.72068781,
      "learning_rate": 2.0936312985863077e-06,
      "loss": 0.73960221,
      "num_input_tokens_seen": 178937760,
      "step": 8325,
      "time_per_iteration": 2.511399507522583
    },
    {
      "auxiliary_loss_clip": 0.0108392,
      "auxiliary_loss_mlp": 0.01037578,
      "balance_loss_clip": 1.04034901,
      "balance_loss_mlp": 1.02283156,
      "epoch": 0.5005862017135126,
      "flos": 24859987649280.0,
      "grad_norm": 1.6107122135647332,
      "language_loss": 0.73869491,
      "learning_rate": 2.093242262158709e-06,
      "loss": 0.75990987,
      "num_input_tokens_seen": 178957985,
      "step": 8326,
      "time_per_iteration": 2.5955214500427246
    },
    {
      "auxiliary_loss_clip": 0.01092702,
      "auxiliary_loss_mlp": 0.01033336,
      "balance_loss_clip": 1.0419116,
      "balance_loss_mlp": 1.02042007,
      "epoch": 0.5006463249661807,
      "flos": 18734381965440.0,
      "grad_norm": 1.6999796156529614,
      "language_loss": 0.78296912,
      "learning_rate": 2.0928532221953544e-06,
      "loss": 0.8042295,
      "num_input_tokens_seen": 178977070,
      "step": 8327,
      "time_per_iteration": 2.5630922317504883
    },
    {
      "auxiliary_loss_clip": 0.01121394,
      "auxiliary_loss_mlp": 0.01040196,
      "balance_loss_clip": 1.04362106,
      "balance_loss_mlp": 1.0268805,
      "epoch": 0.5007064482188487,
      "flos": 13042145871360.0,
      "grad_norm": 2.197871095785912,
      "language_loss": 0.87982273,
      "learning_rate": 2.092464178710997e-06,
      "loss": 0.90143859,
      "num_input_tokens_seen": 178994175,
      "step": 8328,
      "time_per_iteration": 2.5174624919891357
    },
    {
      "auxiliary_loss_clip": 0.01086122,
      "auxiliary_loss_mlp": 0.01034344,
      "balance_loss_clip": 1.04282236,
      "balance_loss_mlp": 1.0212965,
      "epoch": 0.5007665714715166,
      "flos": 21288671965440.0,
      "grad_norm": 2.091605870323064,
      "language_loss": 0.74248588,
      "learning_rate": 2.092075131720388e-06,
      "loss": 0.76369053,
      "num_input_tokens_seen": 179013710,
      "step": 8329,
      "time_per_iteration": 2.5970966815948486
    },
    {
      "auxiliary_loss_clip": 0.01118476,
      "auxiliary_loss_mlp": 0.01034034,
      "balance_loss_clip": 1.04362738,
      "balance_loss_mlp": 1.0215826,
      "epoch": 0.5008266947241846,
      "flos": 29754576374400.0,
      "grad_norm": 1.5951068969860314,
      "language_loss": 0.79709327,
      "learning_rate": 2.091686081238281e-06,
      "loss": 0.8186183,
      "num_input_tokens_seen": 179035255,
      "step": 8330,
      "time_per_iteration": 2.550640344619751
    },
    {
      "auxiliary_loss_clip": 0.01040266,
      "auxiliary_loss_mlp": 0.00788151,
      "balance_loss_clip": 1.04207885,
      "balance_loss_mlp": 1.04400337,
      "epoch": 0.5008868179768525,
      "flos": 63557829204480.0,
      "grad_norm": 0.7398171423400741,
      "language_loss": 0.56050718,
      "learning_rate": 2.0912970272794282e-06,
      "loss": 0.57879138,
      "num_input_tokens_seen": 179090915,
      "step": 8331,
      "time_per_iteration": 2.9876041412353516
    },
    {
      "auxiliary_loss_clip": 0.01106812,
      "auxiliary_loss_mlp": 0.01029599,
      "balance_loss_clip": 1.04333484,
      "balance_loss_mlp": 1.01763678,
      "epoch": 0.5009469412295205,
      "flos": 27375637593600.0,
      "grad_norm": 1.904676676204854,
      "language_loss": 0.65000272,
      "learning_rate": 2.0909079698585833e-06,
      "loss": 0.67136687,
      "num_input_tokens_seen": 179109160,
      "step": 8332,
      "time_per_iteration": 2.5699105262756348
    },
    {
      "auxiliary_loss_clip": 0.01114272,
      "auxiliary_loss_mlp": 0.01032045,
      "balance_loss_clip": 1.04036605,
      "balance_loss_mlp": 1.02035081,
      "epoch": 0.5010070644821885,
      "flos": 27378833904000.0,
      "grad_norm": 1.4382196612078442,
      "language_loss": 0.74899262,
      "learning_rate": 2.0905189089904993e-06,
      "loss": 0.77045572,
      "num_input_tokens_seen": 179130610,
      "step": 8333,
      "time_per_iteration": 2.5824947357177734
    },
    {
      "auxiliary_loss_clip": 0.01118958,
      "auxiliary_loss_mlp": 0.01032258,
      "balance_loss_clip": 1.04153872,
      "balance_loss_mlp": 1.01971745,
      "epoch": 0.5010671877348565,
      "flos": 20662748542080.0,
      "grad_norm": 2.212567988923357,
      "language_loss": 0.8039124,
      "learning_rate": 2.090129844689929e-06,
      "loss": 0.82542461,
      "num_input_tokens_seen": 179147860,
      "step": 8334,
      "time_per_iteration": 2.4836337566375732
    },
    {
      "auxiliary_loss_clip": 0.01032896,
      "auxiliary_loss_mlp": 0.0100399,
      "balance_loss_clip": 1.01674795,
      "balance_loss_mlp": 1.00277448,
      "epoch": 0.5011273109875244,
      "flos": 59128645000320.0,
      "grad_norm": 0.9001251420511123,
      "language_loss": 0.62674439,
      "learning_rate": 2.089740776971626e-06,
      "loss": 0.64711332,
      "num_input_tokens_seen": 179210490,
      "step": 8335,
      "time_per_iteration": 3.0748324394226074
    },
    {
      "auxiliary_loss_clip": 0.01103816,
      "auxiliary_loss_mlp": 0.01029961,
      "balance_loss_clip": 1.03994548,
      "balance_loss_mlp": 1.01758718,
      "epoch": 0.5011874342401924,
      "flos": 25336342840320.0,
      "grad_norm": 1.385136279422082,
      "language_loss": 0.79632974,
      "learning_rate": 2.0893517058503435e-06,
      "loss": 0.81766748,
      "num_input_tokens_seen": 179231360,
      "step": 8336,
      "time_per_iteration": 2.5285725593566895
    },
    {
      "auxiliary_loss_clip": 0.01079704,
      "auxiliary_loss_mlp": 0.01031669,
      "balance_loss_clip": 1.04044843,
      "balance_loss_mlp": 1.01831162,
      "epoch": 0.5012475574928603,
      "flos": 20229953569920.0,
      "grad_norm": 1.7242012536977913,
      "language_loss": 0.79654658,
      "learning_rate": 2.088962631340836e-06,
      "loss": 0.81766027,
      "num_input_tokens_seen": 179250625,
      "step": 8337,
      "time_per_iteration": 2.5640838146209717
    },
    {
      "auxiliary_loss_clip": 0.01120688,
      "auxiliary_loss_mlp": 0.01033824,
      "balance_loss_clip": 1.04047585,
      "balance_loss_mlp": 1.02046108,
      "epoch": 0.5013076807455283,
      "flos": 22710123855360.0,
      "grad_norm": 1.9652573496082573,
      "language_loss": 0.79249191,
      "learning_rate": 2.0885735534578555e-06,
      "loss": 0.81403702,
      "num_input_tokens_seen": 179267360,
      "step": 8338,
      "time_per_iteration": 2.4686102867126465
    },
    {
      "auxiliary_loss_clip": 0.01093061,
      "auxiliary_loss_mlp": 0.0102707,
      "balance_loss_clip": 1.04058933,
      "balance_loss_mlp": 1.01414168,
      "epoch": 0.5013678039981962,
      "flos": 24245161528320.0,
      "grad_norm": 1.5582642459852403,
      "language_loss": 0.85071665,
      "learning_rate": 2.0881844722161583e-06,
      "loss": 0.87191796,
      "num_input_tokens_seen": 179289810,
      "step": 8339,
      "time_per_iteration": 3.9518470764160156
    },
    {
      "auxiliary_loss_clip": 0.01104002,
      "auxiliary_loss_mlp": 0.01032442,
      "balance_loss_clip": 1.03941965,
      "balance_loss_mlp": 1.01978767,
      "epoch": 0.5014279272508643,
      "flos": 26176688501760.0,
      "grad_norm": 1.5023565926583609,
      "language_loss": 0.70657265,
      "learning_rate": 2.0877953876304962e-06,
      "loss": 0.7279371,
      "num_input_tokens_seen": 179310620,
      "step": 8340,
      "time_per_iteration": 2.578184127807617
    },
    {
      "auxiliary_loss_clip": 0.01084743,
      "auxiliary_loss_mlp": 0.01041899,
      "balance_loss_clip": 1.04122651,
      "balance_loss_mlp": 1.02672315,
      "epoch": 0.5014880505035323,
      "flos": 21430446946560.0,
      "grad_norm": 1.8615752404837116,
      "language_loss": 0.77822018,
      "learning_rate": 2.0874062997156245e-06,
      "loss": 0.79948652,
      "num_input_tokens_seen": 179329005,
      "step": 8341,
      "time_per_iteration": 3.919377088546753
    },
    {
      "auxiliary_loss_clip": 0.01092707,
      "auxiliary_loss_mlp": 0.01038028,
      "balance_loss_clip": 1.04465115,
      "balance_loss_mlp": 1.02388978,
      "epoch": 0.5015481737562002,
      "flos": 15770745596160.0,
      "grad_norm": 2.5044583866212085,
      "language_loss": 0.89393723,
      "learning_rate": 2.0870172084862975e-06,
      "loss": 0.91524458,
      "num_input_tokens_seen": 179343785,
      "step": 8342,
      "time_per_iteration": 2.526456832885742
    },
    {
      "auxiliary_loss_clip": 0.01092642,
      "auxiliary_loss_mlp": 0.01031849,
      "balance_loss_clip": 1.03812397,
      "balance_loss_mlp": 1.01915288,
      "epoch": 0.5016082970088682,
      "flos": 26830801123200.0,
      "grad_norm": 1.678121041228467,
      "language_loss": 0.76508659,
      "learning_rate": 2.0866281139572682e-06,
      "loss": 0.78633153,
      "num_input_tokens_seen": 179364070,
      "step": 8343,
      "time_per_iteration": 2.5638020038604736
    },
    {
      "auxiliary_loss_clip": 0.01105893,
      "auxiliary_loss_mlp": 0.01029349,
      "balance_loss_clip": 1.04182792,
      "balance_loss_mlp": 1.01725578,
      "epoch": 0.5016684202615361,
      "flos": 21470595373440.0,
      "grad_norm": 1.6715845985726676,
      "language_loss": 0.66906053,
      "learning_rate": 2.086239016143293e-06,
      "loss": 0.690413,
      "num_input_tokens_seen": 179384225,
      "step": 8344,
      "time_per_iteration": 3.8884036540985107
    },
    {
      "auxiliary_loss_clip": 0.01096983,
      "auxiliary_loss_mlp": 0.01032424,
      "balance_loss_clip": 1.04104829,
      "balance_loss_mlp": 1.01996088,
      "epoch": 0.5017285435142042,
      "flos": 26246821806720.0,
      "grad_norm": 2.3370189494265543,
      "language_loss": 0.75373924,
      "learning_rate": 2.0858499150591258e-06,
      "loss": 0.77503324,
      "num_input_tokens_seen": 179402595,
      "step": 8345,
      "time_per_iteration": 2.594977378845215
    },
    {
      "auxiliary_loss_clip": 0.01106592,
      "auxiliary_loss_mlp": 0.0103434,
      "balance_loss_clip": 1.0470376,
      "balance_loss_mlp": 1.02066064,
      "epoch": 0.5017886667668721,
      "flos": 20777555387520.0,
      "grad_norm": 2.019970485225884,
      "language_loss": 0.78519583,
      "learning_rate": 2.0854608107195203e-06,
      "loss": 0.8066051,
      "num_input_tokens_seen": 179419635,
      "step": 8346,
      "time_per_iteration": 2.5184171199798584
    },
    {
      "auxiliary_loss_clip": 0.01094346,
      "auxiliary_loss_mlp": 0.00798231,
      "balance_loss_clip": 1.03872895,
      "balance_loss_mlp": 1.03484702,
      "epoch": 0.5018487900195401,
      "flos": 20156408472960.0,
      "grad_norm": 1.5663734697411769,
      "language_loss": 0.69378477,
      "learning_rate": 2.0850717031392333e-06,
      "loss": 0.7127105,
      "num_input_tokens_seen": 179438770,
      "step": 8347,
      "time_per_iteration": 4.018402814865112
    },
    {
      "auxiliary_loss_clip": 0.0108201,
      "auxiliary_loss_mlp": 0.01036352,
      "balance_loss_clip": 1.0413599,
      "balance_loss_mlp": 1.02323318,
      "epoch": 0.501908913272208,
      "flos": 18150689957760.0,
      "grad_norm": 2.1661089127939976,
      "language_loss": 0.71219385,
      "learning_rate": 2.0846825923330174e-06,
      "loss": 0.7333774,
      "num_input_tokens_seen": 179457475,
      "step": 8348,
      "time_per_iteration": 2.5724430084228516
    },
    {
      "auxiliary_loss_clip": 0.01104929,
      "auxiliary_loss_mlp": 0.01031224,
      "balance_loss_clip": 1.0423857,
      "balance_loss_mlp": 1.01949358,
      "epoch": 0.501969036524876,
      "flos": 23112287504640.0,
      "grad_norm": 1.4330641828254942,
      "language_loss": 0.74088562,
      "learning_rate": 2.0842934783156303e-06,
      "loss": 0.76224709,
      "num_input_tokens_seen": 179478140,
      "step": 8349,
      "time_per_iteration": 2.5157666206359863
    },
    {
      "auxiliary_loss_clip": 0.01106736,
      "auxiliary_loss_mlp": 0.01029767,
      "balance_loss_clip": 1.04048491,
      "balance_loss_mlp": 1.01614761,
      "epoch": 0.5020291597775439,
      "flos": 11363214314880.0,
      "grad_norm": 2.3345324305738875,
      "language_loss": 0.64319181,
      "learning_rate": 2.0839043611018266e-06,
      "loss": 0.66455686,
      "num_input_tokens_seen": 179494325,
      "step": 8350,
      "time_per_iteration": 2.5016677379608154
    },
    {
      "auxiliary_loss_clip": 0.01016,
      "auxiliary_loss_mlp": 0.01010465,
      "balance_loss_clip": 1.01904535,
      "balance_loss_mlp": 1.00895071,
      "epoch": 0.5020892830302119,
      "flos": 64011094928640.0,
      "grad_norm": 0.7782445063299406,
      "language_loss": 0.59862143,
      "learning_rate": 2.0835152407063597e-06,
      "loss": 0.61888611,
      "num_input_tokens_seen": 179553545,
      "step": 8351,
      "time_per_iteration": 3.3042829036712646
    },
    {
      "auxiliary_loss_clip": 0.01092734,
      "auxiliary_loss_mlp": 0.0103472,
      "balance_loss_clip": 1.04183519,
      "balance_loss_mlp": 1.02127349,
      "epoch": 0.5021494062828799,
      "flos": 23732859801600.0,
      "grad_norm": 1.6823765142920162,
      "language_loss": 0.75188833,
      "learning_rate": 2.0831261171439873e-06,
      "loss": 0.77316284,
      "num_input_tokens_seen": 179573645,
      "step": 8352,
      "time_per_iteration": 2.5590567588806152
    },
    {
      "auxiliary_loss_clip": 0.01097948,
      "auxiliary_loss_mlp": 0.01031693,
      "balance_loss_clip": 1.04278874,
      "balance_loss_mlp": 1.01813924,
      "epoch": 0.5022095295355479,
      "flos": 21576747041280.0,
      "grad_norm": 1.6179659374671547,
      "language_loss": 0.71630204,
      "learning_rate": 2.082736990429464e-06,
      "loss": 0.73759848,
      "num_input_tokens_seen": 179591435,
      "step": 8353,
      "time_per_iteration": 2.5196127891540527
    },
    {
      "auxiliary_loss_clip": 0.01112838,
      "auxiliary_loss_mlp": 0.01032982,
      "balance_loss_clip": 1.04568052,
      "balance_loss_mlp": 1.01904619,
      "epoch": 0.5022696527882159,
      "flos": 21397229844480.0,
      "grad_norm": 1.6711307480417537,
      "language_loss": 0.74147379,
      "learning_rate": 2.0823478605775455e-06,
      "loss": 0.762932,
      "num_input_tokens_seen": 179609955,
      "step": 8354,
      "time_per_iteration": 2.5023159980773926
    },
    {
      "auxiliary_loss_clip": 0.01094441,
      "auxiliary_loss_mlp": 0.0103902,
      "balance_loss_clip": 1.04148459,
      "balance_loss_mlp": 1.02526283,
      "epoch": 0.5023297760408838,
      "flos": 27160712565120.0,
      "grad_norm": 1.668388441414166,
      "language_loss": 0.7223568,
      "learning_rate": 2.0819587276029884e-06,
      "loss": 0.74369138,
      "num_input_tokens_seen": 179630875,
      "step": 8355,
      "time_per_iteration": 2.57133150100708
    },
    {
      "auxiliary_loss_clip": 0.01106881,
      "auxiliary_loss_mlp": 0.01035775,
      "balance_loss_clip": 1.04014313,
      "balance_loss_mlp": 1.02195835,
      "epoch": 0.5023898992935518,
      "flos": 26213820186240.0,
      "grad_norm": 4.245687406793476,
      "language_loss": 0.81649113,
      "learning_rate": 2.081569591520548e-06,
      "loss": 0.83791769,
      "num_input_tokens_seen": 179649835,
      "step": 8356,
      "time_per_iteration": 2.5295908451080322
    },
    {
      "auxiliary_loss_clip": 0.01109294,
      "auxiliary_loss_mlp": 0.01033785,
      "balance_loss_clip": 1.0405395,
      "balance_loss_mlp": 1.01908696,
      "epoch": 0.5024500225462197,
      "flos": 13440323111040.0,
      "grad_norm": 2.027064224657736,
      "language_loss": 0.75739896,
      "learning_rate": 2.0811804523449803e-06,
      "loss": 0.77882975,
      "num_input_tokens_seen": 179667605,
      "step": 8357,
      "time_per_iteration": 2.4750051498413086
    },
    {
      "auxiliary_loss_clip": 0.01106908,
      "auxiliary_loss_mlp": 0.01034996,
      "balance_loss_clip": 1.04140472,
      "balance_loss_mlp": 1.02087522,
      "epoch": 0.5025101457988878,
      "flos": 21579584215680.0,
      "grad_norm": 1.776294459394396,
      "language_loss": 0.76360142,
      "learning_rate": 2.0807913100910417e-06,
      "loss": 0.78502047,
      "num_input_tokens_seen": 179686910,
      "step": 8358,
      "time_per_iteration": 2.505016326904297
    },
    {
      "auxiliary_loss_clip": 0.01095827,
      "auxiliary_loss_mlp": 0.01034181,
      "balance_loss_clip": 1.04082155,
      "balance_loss_mlp": 1.02044833,
      "epoch": 0.5025702690515557,
      "flos": 24645134448000.0,
      "grad_norm": 2.006728332983904,
      "language_loss": 0.71918261,
      "learning_rate": 2.0804021647734887e-06,
      "loss": 0.74048269,
      "num_input_tokens_seen": 179706395,
      "step": 8359,
      "time_per_iteration": 2.55633282661438
    },
    {
      "auxiliary_loss_clip": 0.01091156,
      "auxiliary_loss_mlp": 0.01040344,
      "balance_loss_clip": 1.04306269,
      "balance_loss_mlp": 1.02698708,
      "epoch": 0.5026303923042237,
      "flos": 22090162089600.0,
      "grad_norm": 1.587518122587937,
      "language_loss": 0.76975691,
      "learning_rate": 2.080013016407077e-06,
      "loss": 0.79107189,
      "num_input_tokens_seen": 179725735,
      "step": 8360,
      "time_per_iteration": 2.5397491455078125
    },
    {
      "auxiliary_loss_clip": 0.01085468,
      "auxiliary_loss_mlp": 0.01032977,
      "balance_loss_clip": 1.04627287,
      "balance_loss_mlp": 1.02068686,
      "epoch": 0.5026905155568916,
      "flos": 23697200574720.0,
      "grad_norm": 1.9241471861288912,
      "language_loss": 0.76819336,
      "learning_rate": 2.0796238650065645e-06,
      "loss": 0.78937787,
      "num_input_tokens_seen": 179746150,
      "step": 8361,
      "time_per_iteration": 2.582095146179199
    },
    {
      "auxiliary_loss_clip": 0.01087362,
      "auxiliary_loss_mlp": 0.01033917,
      "balance_loss_clip": 1.03914595,
      "balance_loss_mlp": 1.01961792,
      "epoch": 0.5027506388095596,
      "flos": 25812410722560.0,
      "grad_norm": 1.5499227791851073,
      "language_loss": 0.85201824,
      "learning_rate": 2.0792347105867065e-06,
      "loss": 0.87323105,
      "num_input_tokens_seen": 179767550,
      "step": 8362,
      "time_per_iteration": 2.5837697982788086
    },
    {
      "auxiliary_loss_clip": 0.01093362,
      "auxiliary_loss_mlp": 0.01032706,
      "balance_loss_clip": 1.03812337,
      "balance_loss_mlp": 1.02008748,
      "epoch": 0.5028107620622275,
      "flos": 27526606456320.0,
      "grad_norm": 1.5850197923234448,
      "language_loss": 0.78325808,
      "learning_rate": 2.0788455531622605e-06,
      "loss": 0.80451882,
      "num_input_tokens_seen": 179790075,
      "step": 8363,
      "time_per_iteration": 2.5604465007781982
    },
    {
      "auxiliary_loss_clip": 0.01102411,
      "auxiliary_loss_mlp": 0.01029789,
      "balance_loss_clip": 1.04128313,
      "balance_loss_mlp": 1.01683092,
      "epoch": 0.5028708853148955,
      "flos": 24534278098560.0,
      "grad_norm": 2.2138726464268768,
      "language_loss": 0.75600857,
      "learning_rate": 2.0784563927479838e-06,
      "loss": 0.77733058,
      "num_input_tokens_seen": 179806515,
      "step": 8364,
      "time_per_iteration": 2.5202434062957764
    },
    {
      "auxiliary_loss_clip": 0.0111425,
      "auxiliary_loss_mlp": 0.01028072,
      "balance_loss_clip": 1.04075074,
      "balance_loss_mlp": 1.01551974,
      "epoch": 0.5029310085675635,
      "flos": 20813609664000.0,
      "grad_norm": 1.5403258390333454,
      "language_loss": 0.69314456,
      "learning_rate": 2.0780672293586317e-06,
      "loss": 0.71456778,
      "num_input_tokens_seen": 179826450,
      "step": 8365,
      "time_per_iteration": 2.4808809757232666
    },
    {
      "auxiliary_loss_clip": 0.0109873,
      "auxiliary_loss_mlp": 0.01033818,
      "balance_loss_clip": 1.04261589,
      "balance_loss_mlp": 1.01978683,
      "epoch": 0.5029911318202315,
      "flos": 22342470197760.0,
      "grad_norm": 1.6974462876632457,
      "language_loss": 0.73300099,
      "learning_rate": 2.0776780630089635e-06,
      "loss": 0.75432646,
      "num_input_tokens_seen": 179846770,
      "step": 8366,
      "time_per_iteration": 2.554692506790161
    },
    {
      "auxiliary_loss_clip": 0.01105989,
      "auxiliary_loss_mlp": 0.01032146,
      "balance_loss_clip": 1.04349041,
      "balance_loss_mlp": 1.01980782,
      "epoch": 0.5030512550728995,
      "flos": 24352713826560.0,
      "grad_norm": 1.49011278551894,
      "language_loss": 0.78182185,
      "learning_rate": 2.077288893713735e-06,
      "loss": 0.80320323,
      "num_input_tokens_seen": 179866585,
      "step": 8367,
      "time_per_iteration": 2.5213589668273926
    },
    {
      "auxiliary_loss_clip": 0.0110373,
      "auxiliary_loss_mlp": 0.01027098,
      "balance_loss_clip": 1.0389185,
      "balance_loss_mlp": 1.01465893,
      "epoch": 0.5031113783255674,
      "flos": 18259930195200.0,
      "grad_norm": 1.673509604444769,
      "language_loss": 0.69807196,
      "learning_rate": 2.0768997214877035e-06,
      "loss": 0.71938026,
      "num_input_tokens_seen": 179885575,
      "step": 8368,
      "time_per_iteration": 2.4764156341552734
    },
    {
      "auxiliary_loss_clip": 0.01031603,
      "auxiliary_loss_mlp": 0.01011223,
      "balance_loss_clip": 1.01519513,
      "balance_loss_mlp": 1.00973272,
      "epoch": 0.5031715015782354,
      "flos": 57253173200640.0,
      "grad_norm": 0.8572507678480243,
      "language_loss": 0.63373452,
      "learning_rate": 2.0765105463456274e-06,
      "loss": 0.65416276,
      "num_input_tokens_seen": 179939650,
      "step": 8369,
      "time_per_iteration": 3.0477166175842285
    },
    {
      "auxiliary_loss_clip": 0.01100808,
      "auxiliary_loss_mlp": 0.01031055,
      "balance_loss_clip": 1.04158235,
      "balance_loss_mlp": 1.01903296,
      "epoch": 0.5032316248309033,
      "flos": 27527360641920.0,
      "grad_norm": 1.9000713188379037,
      "language_loss": 0.60082871,
      "learning_rate": 2.076121368302263e-06,
      "loss": 0.62214732,
      "num_input_tokens_seen": 179961765,
      "step": 8370,
      "time_per_iteration": 2.559363603591919
    },
    {
      "auxiliary_loss_clip": 0.01069761,
      "auxiliary_loss_mlp": 0.01037001,
      "balance_loss_clip": 1.04024315,
      "balance_loss_mlp": 1.02285695,
      "epoch": 0.5032917480835714,
      "flos": 34495825939200.0,
      "grad_norm": 1.4947196692731288,
      "language_loss": 0.68107289,
      "learning_rate": 2.0757321873723695e-06,
      "loss": 0.70214045,
      "num_input_tokens_seen": 179983015,
      "step": 8371,
      "time_per_iteration": 2.6963963508605957
    },
    {
      "auxiliary_loss_clip": 0.01091771,
      "auxiliary_loss_mlp": 0.01030442,
      "balance_loss_clip": 1.0416894,
      "balance_loss_mlp": 1.01618457,
      "epoch": 0.5033518713362393,
      "flos": 33656773167360.0,
      "grad_norm": 1.6292297431572758,
      "language_loss": 0.67793065,
      "learning_rate": 2.0753430035707042e-06,
      "loss": 0.69915277,
      "num_input_tokens_seen": 180003210,
      "step": 8372,
      "time_per_iteration": 2.6160614490509033
    },
    {
      "auxiliary_loss_clip": 0.01076952,
      "auxiliary_loss_mlp": 0.0103698,
      "balance_loss_clip": 1.0378325,
      "balance_loss_mlp": 1.02118456,
      "epoch": 0.5034119945889073,
      "flos": 28185495586560.0,
      "grad_norm": 4.171915779559252,
      "language_loss": 0.6657064,
      "learning_rate": 2.0749538169120235e-06,
      "loss": 0.68684572,
      "num_input_tokens_seen": 180025530,
      "step": 8373,
      "time_per_iteration": 2.6331064701080322
    },
    {
      "auxiliary_loss_clip": 0.01088916,
      "auxiliary_loss_mlp": 0.010292,
      "balance_loss_clip": 1.0381918,
      "balance_loss_mlp": 1.01678455,
      "epoch": 0.5034721178415752,
      "flos": 21358697529600.0,
      "grad_norm": 1.8738025898459554,
      "language_loss": 0.74379241,
      "learning_rate": 2.0745646274110872e-06,
      "loss": 0.76497352,
      "num_input_tokens_seen": 180043180,
      "step": 8374,
      "time_per_iteration": 2.5294289588928223
    },
    {
      "auxiliary_loss_clip": 0.01093016,
      "auxiliary_loss_mlp": 0.01036604,
      "balance_loss_clip": 1.04265118,
      "balance_loss_mlp": 1.02238274,
      "epoch": 0.5035322410942432,
      "flos": 22674823764480.0,
      "grad_norm": 1.6660186105529953,
      "language_loss": 0.68219459,
      "learning_rate": 2.0741754350826525e-06,
      "loss": 0.70349079,
      "num_input_tokens_seen": 180062905,
      "step": 8375,
      "time_per_iteration": 2.5358998775482178
    },
    {
      "auxiliary_loss_clip": 0.01076513,
      "auxiliary_loss_mlp": 0.01035544,
      "balance_loss_clip": 1.04410505,
      "balance_loss_mlp": 1.021209,
      "epoch": 0.5035923643469111,
      "flos": 19828723674240.0,
      "grad_norm": 1.6995363312710676,
      "language_loss": 0.78610766,
      "learning_rate": 2.0737862399414777e-06,
      "loss": 0.80722827,
      "num_input_tokens_seen": 180082000,
      "step": 8376,
      "time_per_iteration": 2.5749588012695312
    },
    {
      "auxiliary_loss_clip": 0.01108137,
      "auxiliary_loss_mlp": 0.00788239,
      "balance_loss_clip": 1.04034758,
      "balance_loss_mlp": 1.01196957,
      "epoch": 0.5036524875995791,
      "flos": 30514625182080.0,
      "grad_norm": 1.955028721695317,
      "language_loss": 0.59369361,
      "learning_rate": 2.0733970420023213e-06,
      "loss": 0.61265731,
      "num_input_tokens_seen": 180101340,
      "step": 8377,
      "time_per_iteration": 3.8920562267303467
    },
    {
      "auxiliary_loss_clip": 0.01093915,
      "auxiliary_loss_mlp": 0.01038174,
      "balance_loss_clip": 1.0399164,
      "balance_loss_mlp": 1.02498353,
      "epoch": 0.5037126108522471,
      "flos": 14720574637440.0,
      "grad_norm": 1.9590073934771954,
      "language_loss": 0.76073194,
      "learning_rate": 2.0730078412799425e-06,
      "loss": 0.78205287,
      "num_input_tokens_seen": 180119160,
      "step": 8378,
      "time_per_iteration": 2.51405930519104
    },
    {
      "auxiliary_loss_clip": 0.01084516,
      "auxiliary_loss_mlp": 0.01034434,
      "balance_loss_clip": 1.0423764,
      "balance_loss_mlp": 1.02200031,
      "epoch": 0.5037727341049151,
      "flos": 25297702784640.0,
      "grad_norm": 1.9535416930969416,
      "language_loss": 0.7472626,
      "learning_rate": 2.0726186377890985e-06,
      "loss": 0.76845205,
      "num_input_tokens_seen": 180138730,
      "step": 8379,
      "time_per_iteration": 4.035807847976685
    },
    {
      "auxiliary_loss_clip": 0.01101964,
      "auxiliary_loss_mlp": 0.01034991,
      "balance_loss_clip": 1.0419085,
      "balance_loss_mlp": 1.0222652,
      "epoch": 0.5038328573575831,
      "flos": 28541764632960.0,
      "grad_norm": 2.176342160680064,
      "language_loss": 0.66610205,
      "learning_rate": 2.072229431544548e-06,
      "loss": 0.68747157,
      "num_input_tokens_seen": 180158810,
      "step": 8380,
      "time_per_iteration": 2.5718562602996826
    },
    {
      "auxiliary_loss_clip": 0.01065072,
      "auxiliary_loss_mlp": 0.01031638,
      "balance_loss_clip": 1.04503679,
      "balance_loss_mlp": 1.01939559,
      "epoch": 0.503892980610251,
      "flos": 31649869503360.0,
      "grad_norm": 1.9382180392536867,
      "language_loss": 0.63046485,
      "learning_rate": 2.071840222561051e-06,
      "loss": 0.65143198,
      "num_input_tokens_seen": 180179700,
      "step": 8381,
      "time_per_iteration": 2.6786820888519287
    },
    {
      "auxiliary_loss_clip": 0.0108933,
      "auxiliary_loss_mlp": 0.01038228,
      "balance_loss_clip": 1.03652084,
      "balance_loss_mlp": 1.02558541,
      "epoch": 0.503953103862919,
      "flos": 27089358197760.0,
      "grad_norm": 1.5035041232399846,
      "language_loss": 0.67213088,
      "learning_rate": 2.071451010853365e-06,
      "loss": 0.69340646,
      "num_input_tokens_seen": 180199890,
      "step": 8382,
      "time_per_iteration": 2.5657899379730225
    },
    {
      "auxiliary_loss_clip": 0.01102598,
      "auxiliary_loss_mlp": 0.01039453,
      "balance_loss_clip": 1.04116189,
      "balance_loss_mlp": 1.02579153,
      "epoch": 0.5040132271155869,
      "flos": 15632957024640.0,
      "grad_norm": 1.675740406981606,
      "language_loss": 0.62068951,
      "learning_rate": 2.0710617964362506e-06,
      "loss": 0.64211011,
      "num_input_tokens_seen": 180217840,
      "step": 8383,
      "time_per_iteration": 3.9274253845214844
    },
    {
      "auxiliary_loss_clip": 0.01083982,
      "auxiliary_loss_mlp": 0.01028109,
      "balance_loss_clip": 1.04009008,
      "balance_loss_mlp": 1.01612878,
      "epoch": 0.504073350368255,
      "flos": 13590106824960.0,
      "grad_norm": 1.7156675448869185,
      "language_loss": 0.66710418,
      "learning_rate": 2.070672579324465e-06,
      "loss": 0.68822509,
      "num_input_tokens_seen": 180236465,
      "step": 8384,
      "time_per_iteration": 2.5659804344177246
    },
    {
      "auxiliary_loss_clip": 0.01100934,
      "auxiliary_loss_mlp": 0.01038604,
      "balance_loss_clip": 1.04214466,
      "balance_loss_mlp": 1.02656996,
      "epoch": 0.5041334736209229,
      "flos": 29058160510080.0,
      "grad_norm": 1.9073153573329455,
      "language_loss": 0.71046567,
      "learning_rate": 2.0702833595327674e-06,
      "loss": 0.73186111,
      "num_input_tokens_seen": 180258025,
      "step": 8385,
      "time_per_iteration": 2.57018780708313
    },
    {
      "auxiliary_loss_clip": 0.01102447,
      "auxiliary_loss_mlp": 0.01029635,
      "balance_loss_clip": 1.04081357,
      "balance_loss_mlp": 1.01739788,
      "epoch": 0.5041935968735909,
      "flos": 24608361899520.0,
      "grad_norm": 1.8334123881679454,
      "language_loss": 0.83424425,
      "learning_rate": 2.069894137075919e-06,
      "loss": 0.85556507,
      "num_input_tokens_seen": 180277825,
      "step": 8386,
      "time_per_iteration": 3.880866527557373
    },
    {
      "auxiliary_loss_clip": 0.01101565,
      "auxiliary_loss_mlp": 0.0103184,
      "balance_loss_clip": 1.04210579,
      "balance_loss_mlp": 1.01935875,
      "epoch": 0.5042537201262588,
      "flos": 26286934320000.0,
      "grad_norm": 1.8732999463607578,
      "language_loss": 0.66957593,
      "learning_rate": 2.0695049119686766e-06,
      "loss": 0.69090998,
      "num_input_tokens_seen": 180300465,
      "step": 8387,
      "time_per_iteration": 2.5349254608154297
    },
    {
      "auxiliary_loss_clip": 0.01060981,
      "auxiliary_loss_mlp": 0.01033215,
      "balance_loss_clip": 1.03958988,
      "balance_loss_mlp": 1.02068591,
      "epoch": 0.5043138433789268,
      "flos": 22017371178240.0,
      "grad_norm": 1.4571745400429825,
      "language_loss": 0.80275702,
      "learning_rate": 2.0691156842258016e-06,
      "loss": 0.82369906,
      "num_input_tokens_seen": 180321050,
      "step": 8388,
      "time_per_iteration": 2.6253745555877686
    },
    {
      "auxiliary_loss_clip": 0.01104212,
      "auxiliary_loss_mlp": 0.01031644,
      "balance_loss_clip": 1.04124272,
      "balance_loss_mlp": 1.01968706,
      "epoch": 0.5043739666315947,
      "flos": 28767104605440.0,
      "grad_norm": 2.1590236706838164,
      "language_loss": 0.70275986,
      "learning_rate": 2.0687264538620537e-06,
      "loss": 0.72411847,
      "num_input_tokens_seen": 180338870,
      "step": 8389,
      "time_per_iteration": 2.570636034011841
    },
    {
      "auxiliary_loss_clip": 0.01085175,
      "auxiliary_loss_mlp": 0.0103858,
      "balance_loss_clip": 1.04025006,
      "balance_loss_mlp": 1.02628374,
      "epoch": 0.5044340898842627,
      "flos": 27599253713280.0,
      "grad_norm": 1.6404694891923484,
      "language_loss": 0.69485068,
      "learning_rate": 2.068337220892191e-06,
      "loss": 0.71608818,
      "num_input_tokens_seen": 180361285,
      "step": 8390,
      "time_per_iteration": 2.595111846923828
    },
    {
      "auxiliary_loss_clip": 0.01023936,
      "auxiliary_loss_mlp": 0.01008447,
      "balance_loss_clip": 1.01851201,
      "balance_loss_mlp": 1.00714195,
      "epoch": 0.5044942131369307,
      "flos": 67458050749440.0,
      "grad_norm": 0.8548741014068696,
      "language_loss": 0.53004336,
      "learning_rate": 2.067947985330974e-06,
      "loss": 0.55036724,
      "num_input_tokens_seen": 180415170,
      "step": 8391,
      "time_per_iteration": 2.943901300430298
    },
    {
      "auxiliary_loss_clip": 0.01015997,
      "auxiliary_loss_mlp": 0.01011256,
      "balance_loss_clip": 1.02420056,
      "balance_loss_mlp": 1.00983715,
      "epoch": 0.5045543363895987,
      "flos": 58630849390080.0,
      "grad_norm": 0.8584399321609673,
      "language_loss": 0.60789144,
      "learning_rate": 2.0675587471931628e-06,
      "loss": 0.62816393,
      "num_input_tokens_seen": 180468060,
      "step": 8392,
      "time_per_iteration": 2.993121862411499
    },
    {
      "auxiliary_loss_clip": 0.01075417,
      "auxiliary_loss_mlp": 0.01029697,
      "balance_loss_clip": 1.03665912,
      "balance_loss_mlp": 1.01760268,
      "epoch": 0.5046144596422667,
      "flos": 22526620248960.0,
      "grad_norm": 1.612416800045726,
      "language_loss": 0.84507912,
      "learning_rate": 2.067169506493517e-06,
      "loss": 0.86613023,
      "num_input_tokens_seen": 180486610,
      "step": 8393,
      "time_per_iteration": 2.559434413909912
    },
    {
      "auxiliary_loss_clip": 0.01081874,
      "auxiliary_loss_mlp": 0.01028891,
      "balance_loss_clip": 1.03991294,
      "balance_loss_mlp": 1.01669598,
      "epoch": 0.5046745828949346,
      "flos": 27454246508160.0,
      "grad_norm": 2.1378123720809854,
      "language_loss": 0.51084638,
      "learning_rate": 2.0667802632467974e-06,
      "loss": 0.53195399,
      "num_input_tokens_seen": 180508135,
      "step": 8394,
      "time_per_iteration": 2.608208417892456
    },
    {
      "auxiliary_loss_clip": 0.0111493,
      "auxiliary_loss_mlp": 0.01029368,
      "balance_loss_clip": 1.03938246,
      "balance_loss_mlp": 1.01584327,
      "epoch": 0.5047347061476026,
      "flos": 17274541415040.0,
      "grad_norm": 2.3769163853923296,
      "language_loss": 0.7518599,
      "learning_rate": 2.0663910174677627e-06,
      "loss": 0.77330285,
      "num_input_tokens_seen": 180527000,
      "step": 8395,
      "time_per_iteration": 2.4684853553771973
    },
    {
      "auxiliary_loss_clip": 0.01099397,
      "auxiliary_loss_mlp": 0.01029833,
      "balance_loss_clip": 1.04087424,
      "balance_loss_mlp": 1.01736355,
      "epoch": 0.5047948294002705,
      "flos": 16649515831680.0,
      "grad_norm": 1.843748109855953,
      "language_loss": 0.67756963,
      "learning_rate": 2.0660017691711737e-06,
      "loss": 0.69886196,
      "num_input_tokens_seen": 180544715,
      "step": 8396,
      "time_per_iteration": 2.4828054904937744
    },
    {
      "auxiliary_loss_clip": 0.01106409,
      "auxiliary_loss_mlp": 0.01026244,
      "balance_loss_clip": 1.04371953,
      "balance_loss_mlp": 1.01459098,
      "epoch": 0.5048549526529386,
      "flos": 26865706164480.0,
      "grad_norm": 2.1504927008001844,
      "language_loss": 0.78947288,
      "learning_rate": 2.065612518371792e-06,
      "loss": 0.81079936,
      "num_input_tokens_seen": 180565365,
      "step": 8397,
      "time_per_iteration": 2.5937998294830322
    },
    {
      "auxiliary_loss_clip": 0.01071171,
      "auxiliary_loss_mlp": 0.0102644,
      "balance_loss_clip": 1.03864324,
      "balance_loss_mlp": 1.01419091,
      "epoch": 0.5049150759056065,
      "flos": 21833939399040.0,
      "grad_norm": 1.5773040752954472,
      "language_loss": 0.66137278,
      "learning_rate": 2.065223265084376e-06,
      "loss": 0.68234891,
      "num_input_tokens_seen": 180586670,
      "step": 8398,
      "time_per_iteration": 2.635952949523926
    },
    {
      "auxiliary_loss_clip": 0.01103362,
      "auxiliary_loss_mlp": 0.00786081,
      "balance_loss_clip": 1.04233491,
      "balance_loss_mlp": 1.01142883,
      "epoch": 0.5049751991582745,
      "flos": 21685807710720.0,
      "grad_norm": 1.6423942580528361,
      "language_loss": 0.71823138,
      "learning_rate": 2.064834009323688e-06,
      "loss": 0.73712581,
      "num_input_tokens_seen": 180605085,
      "step": 8399,
      "time_per_iteration": 2.496854782104492
    },
    {
      "auxiliary_loss_clip": 0.01088005,
      "auxiliary_loss_mlp": 0.01048203,
      "balance_loss_clip": 1.04220414,
      "balance_loss_mlp": 1.03402901,
      "epoch": 0.5050353224109424,
      "flos": 21359379888000.0,
      "grad_norm": 1.7224260386942996,
      "language_loss": 0.81551576,
      "learning_rate": 2.0644447511044878e-06,
      "loss": 0.83687782,
      "num_input_tokens_seen": 180624370,
      "step": 8400,
      "time_per_iteration": 2.5499682426452637
    },
    {
      "auxiliary_loss_clip": 0.01078583,
      "auxiliary_loss_mlp": 0.01036967,
      "balance_loss_clip": 1.0408442,
      "balance_loss_mlp": 1.02291179,
      "epoch": 0.5050954456636104,
      "flos": 22820082364800.0,
      "grad_norm": 1.9543535402900545,
      "language_loss": 0.78482634,
      "learning_rate": 2.0640554904415362e-06,
      "loss": 0.80598181,
      "num_input_tokens_seen": 180642450,
      "step": 8401,
      "time_per_iteration": 2.571368932723999
    },
    {
      "auxiliary_loss_clip": 0.01118805,
      "auxiliary_loss_mlp": 0.00785619,
      "balance_loss_clip": 1.04163337,
      "balance_loss_mlp": 1.00819874,
      "epoch": 0.5051555689162783,
      "flos": 30448226891520.0,
      "grad_norm": 2.220460810037052,
      "language_loss": 0.69922721,
      "learning_rate": 2.063666227349593e-06,
      "loss": 0.71827137,
      "num_input_tokens_seen": 180665250,
      "step": 8402,
      "time_per_iteration": 2.599173069000244
    },
    {
      "auxiliary_loss_clip": 0.01103932,
      "auxiliary_loss_mlp": 0.00787385,
      "balance_loss_clip": 1.03810143,
      "balance_loss_mlp": 1.01063085,
      "epoch": 0.5052156921689464,
      "flos": 21287953693440.0,
      "grad_norm": 1.6194415505025195,
      "language_loss": 0.69350284,
      "learning_rate": 2.063276961843422e-06,
      "loss": 0.71241599,
      "num_input_tokens_seen": 180687425,
      "step": 8403,
      "time_per_iteration": 2.5743980407714844
    },
    {
      "auxiliary_loss_clip": 0.01103633,
      "auxiliary_loss_mlp": 0.01036017,
      "balance_loss_clip": 1.04155076,
      "balance_loss_mlp": 1.02335143,
      "epoch": 0.5052758154216143,
      "flos": 25081305298560.0,
      "grad_norm": 1.4181534875648099,
      "language_loss": 0.8563683,
      "learning_rate": 2.062887693937781e-06,
      "loss": 0.87776482,
      "num_input_tokens_seen": 180708725,
      "step": 8404,
      "time_per_iteration": 2.530034303665161
    },
    {
      "auxiliary_loss_clip": 0.01082937,
      "auxiliary_loss_mlp": 0.00785783,
      "balance_loss_clip": 1.04170132,
      "balance_loss_mlp": 1.00983238,
      "epoch": 0.5053359386742823,
      "flos": 20885502735360.0,
      "grad_norm": 1.9053414524239496,
      "language_loss": 0.75773805,
      "learning_rate": 2.0624984236474322e-06,
      "loss": 0.77642524,
      "num_input_tokens_seen": 180727990,
      "step": 8405,
      "time_per_iteration": 2.5523202419281006
    },
    {
      "auxiliary_loss_clip": 0.01119013,
      "auxiliary_loss_mlp": 0.01025467,
      "balance_loss_clip": 1.04306197,
      "balance_loss_mlp": 1.01175249,
      "epoch": 0.5053960619269503,
      "flos": 37743335493120.0,
      "grad_norm": 1.6671367491852005,
      "language_loss": 0.73220342,
      "learning_rate": 2.0621091509871378e-06,
      "loss": 0.75364822,
      "num_input_tokens_seen": 180749765,
      "step": 8406,
      "time_per_iteration": 2.601382255554199
    },
    {
      "auxiliary_loss_clip": 0.01078806,
      "auxiliary_loss_mlp": 0.01027112,
      "balance_loss_clip": 1.04232192,
      "balance_loss_mlp": 1.01493478,
      "epoch": 0.5054561851796182,
      "flos": 23513840622720.0,
      "grad_norm": 1.7351113290178766,
      "language_loss": 0.76697052,
      "learning_rate": 2.0617198759716568e-06,
      "loss": 0.78802967,
      "num_input_tokens_seen": 180769580,
      "step": 8407,
      "time_per_iteration": 2.605022668838501
    },
    {
      "auxiliary_loss_clip": 0.01084533,
      "auxiliary_loss_mlp": 0.01029959,
      "balance_loss_clip": 1.03714442,
      "balance_loss_mlp": 1.0178895,
      "epoch": 0.5055163084322862,
      "flos": 30410233280640.0,
      "grad_norm": 1.637995566245657,
      "language_loss": 0.63068825,
      "learning_rate": 2.0613305986157535e-06,
      "loss": 0.65183318,
      "num_input_tokens_seen": 180790295,
      "step": 8408,
      "time_per_iteration": 2.6036229133605957
    },
    {
      "auxiliary_loss_clip": 0.01090863,
      "auxiliary_loss_mlp": 0.01037448,
      "balance_loss_clip": 1.03879428,
      "balance_loss_mlp": 1.0224756,
      "epoch": 0.5055764316849541,
      "flos": 20259651139200.0,
      "grad_norm": 2.32852673126693,
      "language_loss": 0.63454366,
      "learning_rate": 2.0609413189341865e-06,
      "loss": 0.65582675,
      "num_input_tokens_seen": 180807875,
      "step": 8409,
      "time_per_iteration": 2.5086472034454346
    },
    {
      "auxiliary_loss_clip": 0.01093199,
      "auxiliary_loss_mlp": 0.01024442,
      "balance_loss_clip": 1.04343963,
      "balance_loss_mlp": 1.01310492,
      "epoch": 0.5056365549376222,
      "flos": 26070895969920.0,
      "grad_norm": 1.316193627172706,
      "language_loss": 0.70765156,
      "learning_rate": 2.0605520369417193e-06,
      "loss": 0.72882795,
      "num_input_tokens_seen": 180831300,
      "step": 8410,
      "time_per_iteration": 2.5814642906188965
    },
    {
      "auxiliary_loss_clip": 0.01091251,
      "auxiliary_loss_mlp": 0.01034219,
      "balance_loss_clip": 1.04200876,
      "balance_loss_mlp": 1.02104068,
      "epoch": 0.5056966781902901,
      "flos": 19279074781440.0,
      "grad_norm": 1.5752895007860812,
      "language_loss": 0.7915864,
      "learning_rate": 2.060162752653113e-06,
      "loss": 0.81284112,
      "num_input_tokens_seen": 180849055,
      "step": 8411,
      "time_per_iteration": 2.498537063598633
    },
    {
      "auxiliary_loss_clip": 0.01118095,
      "auxiliary_loss_mlp": 0.0103443,
      "balance_loss_clip": 1.04213452,
      "balance_loss_mlp": 1.01987433,
      "epoch": 0.5057568014429581,
      "flos": 21323325611520.0,
      "grad_norm": 2.8093825839201303,
      "language_loss": 0.81804323,
      "learning_rate": 2.0597734660831285e-06,
      "loss": 0.8395685,
      "num_input_tokens_seen": 180867395,
      "step": 8412,
      "time_per_iteration": 2.4487051963806152
    },
    {
      "auxiliary_loss_clip": 0.01098319,
      "auxiliary_loss_mlp": 0.01033441,
      "balance_loss_clip": 1.04329634,
      "balance_loss_mlp": 1.02070975,
      "epoch": 0.505816924695626,
      "flos": 17493596507520.0,
      "grad_norm": 1.7282082041217908,
      "language_loss": 0.80373216,
      "learning_rate": 2.0593841772465283e-06,
      "loss": 0.82504982,
      "num_input_tokens_seen": 180886670,
      "step": 8413,
      "time_per_iteration": 2.5198514461517334
    },
    {
      "auxiliary_loss_clip": 0.01086645,
      "auxiliary_loss_mlp": 0.0078815,
      "balance_loss_clip": 1.04271638,
      "balance_loss_mlp": 1.01414049,
      "epoch": 0.505877047948294,
      "flos": 21142084561920.0,
      "grad_norm": 1.7481565294224306,
      "language_loss": 0.80492842,
      "learning_rate": 2.0589948861580737e-06,
      "loss": 0.82367635,
      "num_input_tokens_seen": 180904645,
      "step": 8414,
      "time_per_iteration": 2.5708065032958984
    },
    {
      "auxiliary_loss_clip": 0.0110375,
      "auxiliary_loss_mlp": 0.01029111,
      "balance_loss_clip": 1.03886151,
      "balance_loss_mlp": 1.01573563,
      "epoch": 0.5059371712009619,
      "flos": 36350036887680.0,
      "grad_norm": 1.938291123186673,
      "language_loss": 0.62169158,
      "learning_rate": 2.058605592832528e-06,
      "loss": 0.64302021,
      "num_input_tokens_seen": 180922340,
      "step": 8415,
      "time_per_iteration": 4.009791851043701
    },
    {
      "auxiliary_loss_clip": 0.01082351,
      "auxiliary_loss_mlp": 0.01030723,
      "balance_loss_clip": 1.0399704,
      "balance_loss_mlp": 1.01780701,
      "epoch": 0.50599729445363,
      "flos": 22673387220480.0,
      "grad_norm": 1.7056369403840987,
      "language_loss": 0.82024163,
      "learning_rate": 2.0582162972846515e-06,
      "loss": 0.84137231,
      "num_input_tokens_seen": 180941350,
      "step": 8416,
      "time_per_iteration": 2.5643579959869385
    },
    {
      "auxiliary_loss_clip": 0.01082369,
      "auxiliary_loss_mlp": 0.0103634,
      "balance_loss_clip": 1.04217124,
      "balance_loss_mlp": 1.02465177,
      "epoch": 0.5060574177062979,
      "flos": 22747866071040.0,
      "grad_norm": 3.7272451435757907,
      "language_loss": 0.79133308,
      "learning_rate": 2.0578269995292078e-06,
      "loss": 0.81252015,
      "num_input_tokens_seen": 180960720,
      "step": 8417,
      "time_per_iteration": 2.5754427909851074
    },
    {
      "auxiliary_loss_clip": 0.01067923,
      "auxiliary_loss_mlp": 0.01033945,
      "balance_loss_clip": 1.04260206,
      "balance_loss_mlp": 1.02111256,
      "epoch": 0.5061175409589659,
      "flos": 21653201139840.0,
      "grad_norm": 8.64729736212539,
      "language_loss": 0.62633532,
      "learning_rate": 2.0574376995809588e-06,
      "loss": 0.64735401,
      "num_input_tokens_seen": 180979725,
      "step": 8418,
      "time_per_iteration": 4.041779518127441
    },
    {
      "auxiliary_loss_clip": 0.01085166,
      "auxiliary_loss_mlp": 0.01032728,
      "balance_loss_clip": 1.0401814,
      "balance_loss_mlp": 1.02010381,
      "epoch": 0.5061776642116339,
      "flos": 21616249023360.0,
      "grad_norm": 2.4718902341180895,
      "language_loss": 0.77418119,
      "learning_rate": 2.0570483974546653e-06,
      "loss": 0.79536009,
      "num_input_tokens_seen": 180998980,
      "step": 8419,
      "time_per_iteration": 2.542443037033081
    },
    {
      "auxiliary_loss_clip": 0.01062278,
      "auxiliary_loss_mlp": 0.01032148,
      "balance_loss_clip": 1.04149306,
      "balance_loss_mlp": 1.01862931,
      "epoch": 0.5062377874643018,
      "flos": 24426294837120.0,
      "grad_norm": 1.9437993632229984,
      "language_loss": 0.76838434,
      "learning_rate": 2.0566590931650917e-06,
      "loss": 0.78932863,
      "num_input_tokens_seen": 181019165,
      "step": 8420,
      "time_per_iteration": 2.6317360401153564
    },
    {
      "auxiliary_loss_clip": 0.01115265,
      "auxiliary_loss_mlp": 0.0103597,
      "balance_loss_clip": 1.03962278,
      "balance_loss_mlp": 1.02266645,
      "epoch": 0.5062979107169698,
      "flos": 22524429519360.0,
      "grad_norm": 1.7443460358000449,
      "language_loss": 0.77435565,
      "learning_rate": 2.056269786726999e-06,
      "loss": 0.79586798,
      "num_input_tokens_seen": 181037110,
      "step": 8421,
      "time_per_iteration": 3.860727310180664
    },
    {
      "auxiliary_loss_clip": 0.01098929,
      "auxiliary_loss_mlp": 0.01034541,
      "balance_loss_clip": 1.03840685,
      "balance_loss_mlp": 1.02141643,
      "epoch": 0.5063580339696377,
      "flos": 24571984400640.0,
      "grad_norm": 1.4799360759487314,
      "language_loss": 0.66523343,
      "learning_rate": 2.0558804781551512e-06,
      "loss": 0.68656814,
      "num_input_tokens_seen": 181057775,
      "step": 8422,
      "time_per_iteration": 2.561185121536255
    },
    {
      "auxiliary_loss_clip": 0.0111579,
      "auxiliary_loss_mlp": 0.01033874,
      "balance_loss_clip": 1.04231596,
      "balance_loss_mlp": 1.02093959,
      "epoch": 0.5064181572223058,
      "flos": 22596143022720.0,
      "grad_norm": 1.551843060205616,
      "language_loss": 0.81736052,
      "learning_rate": 2.05549116746431e-06,
      "loss": 0.83885717,
      "num_input_tokens_seen": 181078260,
      "step": 8423,
      "time_per_iteration": 2.5180530548095703
    },
    {
      "auxiliary_loss_clip": 0.01117767,
      "auxiliary_loss_mlp": 0.00789273,
      "balance_loss_clip": 1.04139352,
      "balance_loss_mlp": 1.01465607,
      "epoch": 0.5064782804749737,
      "flos": 25994944661760.0,
      "grad_norm": 6.583816641917283,
      "language_loss": 0.74760264,
      "learning_rate": 2.055101854669237e-06,
      "loss": 0.76667309,
      "num_input_tokens_seen": 181098755,
      "step": 8424,
      "time_per_iteration": 3.921456813812256
    },
    {
      "auxiliary_loss_clip": 0.01112463,
      "auxiliary_loss_mlp": 0.01035973,
      "balance_loss_clip": 1.03954947,
      "balance_loss_mlp": 1.02309847,
      "epoch": 0.5065384037276417,
      "flos": 28553041503360.0,
      "grad_norm": 1.4076924634346928,
      "language_loss": 0.7120589,
      "learning_rate": 2.0547125397846975e-06,
      "loss": 0.73354328,
      "num_input_tokens_seen": 181121570,
      "step": 8425,
      "time_per_iteration": 2.5431594848632812
    },
    {
      "auxiliary_loss_clip": 0.0107749,
      "auxiliary_loss_mlp": 0.010393,
      "balance_loss_clip": 1.03799415,
      "balance_loss_mlp": 1.02547181,
      "epoch": 0.5065985269803096,
      "flos": 22966023323520.0,
      "grad_norm": 1.8312556042584032,
      "language_loss": 0.78635436,
      "learning_rate": 2.0543232228254524e-06,
      "loss": 0.8075223,
      "num_input_tokens_seen": 181140240,
      "step": 8426,
      "time_per_iteration": 2.554767608642578
    },
    {
      "auxiliary_loss_clip": 0.01105866,
      "auxiliary_loss_mlp": 0.0103987,
      "balance_loss_clip": 1.04233778,
      "balance_loss_mlp": 1.02683997,
      "epoch": 0.5066586502329776,
      "flos": 21608563512960.0,
      "grad_norm": 2.2273907031298945,
      "language_loss": 0.77974129,
      "learning_rate": 2.053933903806265e-06,
      "loss": 0.80119866,
      "num_input_tokens_seen": 181158630,
      "step": 8427,
      "time_per_iteration": 2.497082471847534
    },
    {
      "auxiliary_loss_clip": 0.01110695,
      "auxiliary_loss_mlp": 0.01027892,
      "balance_loss_clip": 1.03863752,
      "balance_loss_mlp": 1.01564884,
      "epoch": 0.5067187734856455,
      "flos": 20339912079360.0,
      "grad_norm": 1.7268489914193934,
      "language_loss": 0.71977621,
      "learning_rate": 2.0535445827418997e-06,
      "loss": 0.74116206,
      "num_input_tokens_seen": 181176405,
      "step": 8428,
      "time_per_iteration": 2.4352338314056396
    },
    {
      "auxiliary_loss_clip": 0.01104993,
      "auxiliary_loss_mlp": 0.00789167,
      "balance_loss_clip": 1.03911841,
      "balance_loss_mlp": 1.0162816,
      "epoch": 0.5067788967383136,
      "flos": 28841080665600.0,
      "grad_norm": 1.6301540872370726,
      "language_loss": 0.82731915,
      "learning_rate": 2.0531552596471168e-06,
      "loss": 0.84626079,
      "num_input_tokens_seen": 181197595,
      "step": 8429,
      "time_per_iteration": 2.556468963623047
    },
    {
      "auxiliary_loss_clip": 0.01087104,
      "auxiliary_loss_mlp": 0.01035316,
      "balance_loss_clip": 1.04222012,
      "balance_loss_mlp": 1.0213809,
      "epoch": 0.5068390199909815,
      "flos": 32450174478720.0,
      "grad_norm": 1.6519146587137663,
      "language_loss": 0.73263961,
      "learning_rate": 2.052765934536682e-06,
      "loss": 0.75386381,
      "num_input_tokens_seen": 181218560,
      "step": 8430,
      "time_per_iteration": 2.6230990886688232
    },
    {
      "auxiliary_loss_clip": 0.01052622,
      "auxiliary_loss_mlp": 0.01035953,
      "balance_loss_clip": 1.035707,
      "balance_loss_mlp": 1.02173114,
      "epoch": 0.5068991432436495,
      "flos": 23146582014720.0,
      "grad_norm": 1.628870406088993,
      "language_loss": 0.76042998,
      "learning_rate": 2.0523766074253575e-06,
      "loss": 0.78131568,
      "num_input_tokens_seen": 181237095,
      "step": 8431,
      "time_per_iteration": 2.6198713779449463
    },
    {
      "auxiliary_loss_clip": 0.01101067,
      "auxiliary_loss_mlp": 0.01034017,
      "balance_loss_clip": 1.04151106,
      "balance_loss_mlp": 1.02107668,
      "epoch": 0.5069592664963174,
      "flos": 19936096404480.0,
      "grad_norm": 1.4580897092131913,
      "language_loss": 0.71959198,
      "learning_rate": 2.0519872783279074e-06,
      "loss": 0.74094284,
      "num_input_tokens_seen": 181255940,
      "step": 8432,
      "time_per_iteration": 2.4795711040496826
    },
    {
      "auxiliary_loss_clip": 0.01000145,
      "auxiliary_loss_mlp": 0.0100751,
      "balance_loss_clip": 1.02003932,
      "balance_loss_mlp": 1.00591242,
      "epoch": 0.5070193897489854,
      "flos": 65793771941760.0,
      "grad_norm": 0.7572006243203351,
      "language_loss": 0.63664293,
      "learning_rate": 2.0515979472590945e-06,
      "loss": 0.65671945,
      "num_input_tokens_seen": 181316945,
      "step": 8433,
      "time_per_iteration": 3.229689836502075
    },
    {
      "auxiliary_loss_clip": 0.01083446,
      "auxiliary_loss_mlp": 0.01041054,
      "balance_loss_clip": 1.04095042,
      "balance_loss_mlp": 1.02800083,
      "epoch": 0.5070795130016534,
      "flos": 17275331514240.0,
      "grad_norm": 2.027464371140594,
      "language_loss": 0.77640438,
      "learning_rate": 2.051208614233681e-06,
      "loss": 0.79764938,
      "num_input_tokens_seen": 181335555,
      "step": 8434,
      "time_per_iteration": 2.537355422973633
    },
    {
      "auxiliary_loss_clip": 0.01094121,
      "auxiliary_loss_mlp": 0.01035145,
      "balance_loss_clip": 1.03935814,
      "balance_loss_mlp": 1.02233553,
      "epoch": 0.5071396362543213,
      "flos": 21069940095360.0,
      "grad_norm": 1.7571671880483892,
      "language_loss": 0.7080127,
      "learning_rate": 2.0508192792664326e-06,
      "loss": 0.72930539,
      "num_input_tokens_seen": 181354580,
      "step": 8435,
      "time_per_iteration": 2.526679515838623
    },
    {
      "auxiliary_loss_clip": 0.01107302,
      "auxiliary_loss_mlp": 0.01036525,
      "balance_loss_clip": 1.04042554,
      "balance_loss_mlp": 1.02295256,
      "epoch": 0.5071997595069894,
      "flos": 23144822248320.0,
      "grad_norm": 1.9047286734412912,
      "language_loss": 0.72185457,
      "learning_rate": 2.050429942372112e-06,
      "loss": 0.74329293,
      "num_input_tokens_seen": 181374320,
      "step": 8436,
      "time_per_iteration": 2.506650686264038
    },
    {
      "auxiliary_loss_clip": 0.01116706,
      "auxiliary_loss_mlp": 0.01036595,
      "balance_loss_clip": 1.04163933,
      "balance_loss_mlp": 1.02274287,
      "epoch": 0.5072598827596573,
      "flos": 22747183712640.0,
      "grad_norm": 1.559618948584808,
      "language_loss": 0.83808744,
      "learning_rate": 2.050040603565483e-06,
      "loss": 0.85962045,
      "num_input_tokens_seen": 181392190,
      "step": 8437,
      "time_per_iteration": 2.489640235900879
    },
    {
      "auxiliary_loss_clip": 0.01100696,
      "auxiliary_loss_mlp": 0.01029433,
      "balance_loss_clip": 1.03826857,
      "balance_loss_mlp": 1.01690459,
      "epoch": 0.5073200060123253,
      "flos": 22566301799040.0,
      "grad_norm": 1.4443720024593156,
      "language_loss": 0.80518192,
      "learning_rate": 2.049651262861309e-06,
      "loss": 0.82648325,
      "num_input_tokens_seen": 181413890,
      "step": 8438,
      "time_per_iteration": 2.53745174407959
    },
    {
      "auxiliary_loss_clip": 0.0107599,
      "auxiliary_loss_mlp": 0.01039016,
      "balance_loss_clip": 1.042364,
      "balance_loss_mlp": 1.02393007,
      "epoch": 0.5073801292649932,
      "flos": 25806341324160.0,
      "grad_norm": 1.8734157588145635,
      "language_loss": 0.79491121,
      "learning_rate": 2.0492619202743543e-06,
      "loss": 0.81606126,
      "num_input_tokens_seen": 181433240,
      "step": 8439,
      "time_per_iteration": 2.6027188301086426
    },
    {
      "auxiliary_loss_clip": 0.01086594,
      "auxiliary_loss_mlp": 0.00789586,
      "balance_loss_clip": 1.04235137,
      "balance_loss_mlp": 1.01187348,
      "epoch": 0.5074402525176612,
      "flos": 25373941401600.0,
      "grad_norm": 1.6145564089304947,
      "language_loss": 0.71176219,
      "learning_rate": 2.048872575819383e-06,
      "loss": 0.730524,
      "num_input_tokens_seen": 181453535,
      "step": 8440,
      "time_per_iteration": 2.5860679149627686
    },
    {
      "auxiliary_loss_clip": 0.0109029,
      "auxiliary_loss_mlp": 0.01032674,
      "balance_loss_clip": 1.03911543,
      "balance_loss_mlp": 1.01965082,
      "epoch": 0.5075003757703291,
      "flos": 26064431521920.0,
      "grad_norm": 1.5976432251225625,
      "language_loss": 0.70990372,
      "learning_rate": 2.048483229511158e-06,
      "loss": 0.7311334,
      "num_input_tokens_seen": 181474195,
      "step": 8441,
      "time_per_iteration": 2.5856683254241943
    },
    {
      "auxiliary_loss_clip": 0.01105978,
      "auxiliary_loss_mlp": 0.00786074,
      "balance_loss_clip": 1.0394392,
      "balance_loss_mlp": 1.00959945,
      "epoch": 0.5075604990229972,
      "flos": 21835447770240.0,
      "grad_norm": 1.5880638316458038,
      "language_loss": 0.63820767,
      "learning_rate": 2.0480938813644445e-06,
      "loss": 0.65712821,
      "num_input_tokens_seen": 181494000,
      "step": 8442,
      "time_per_iteration": 2.509188652038574
    },
    {
      "auxiliary_loss_clip": 0.01061477,
      "auxiliary_loss_mlp": 0.01030075,
      "balance_loss_clip": 1.0414176,
      "balance_loss_mlp": 1.01819551,
      "epoch": 0.5076206222756651,
      "flos": 31978703537280.0,
      "grad_norm": 1.4448797561748918,
      "language_loss": 0.7106123,
      "learning_rate": 2.047704531394006e-06,
      "loss": 0.73152781,
      "num_input_tokens_seen": 181515955,
      "step": 8443,
      "time_per_iteration": 2.675706386566162
    },
    {
      "auxiliary_loss_clip": 0.01035665,
      "auxiliary_loss_mlp": 0.01042641,
      "balance_loss_clip": 1.0385499,
      "balance_loss_mlp": 1.02713215,
      "epoch": 0.5076807455283331,
      "flos": 36904031326080.0,
      "grad_norm": 1.337808592395016,
      "language_loss": 0.62067145,
      "learning_rate": 2.047315179614607e-06,
      "loss": 0.64145446,
      "num_input_tokens_seen": 181540225,
      "step": 8444,
      "time_per_iteration": 2.783217430114746
    },
    {
      "auxiliary_loss_clip": 0.01080601,
      "auxiliary_loss_mlp": 0.01033122,
      "balance_loss_clip": 1.03804421,
      "balance_loss_mlp": 1.02062941,
      "epoch": 0.507740868781001,
      "flos": 29862415981440.0,
      "grad_norm": 1.6606660395372757,
      "language_loss": 0.63766843,
      "learning_rate": 2.046925826041012e-06,
      "loss": 0.65880567,
      "num_input_tokens_seen": 181560125,
      "step": 8445,
      "time_per_iteration": 2.614365339279175
    },
    {
      "auxiliary_loss_clip": 0.01014222,
      "auxiliary_loss_mlp": 0.01001772,
      "balance_loss_clip": 1.01780057,
      "balance_loss_mlp": 1.00048435,
      "epoch": 0.507800992033669,
      "flos": 61918974247680.0,
      "grad_norm": 0.829498800244111,
      "language_loss": 0.61935854,
      "learning_rate": 2.0465364706879845e-06,
      "loss": 0.63951844,
      "num_input_tokens_seen": 181618830,
      "step": 8446,
      "time_per_iteration": 3.172197103500366
    },
    {
      "auxiliary_loss_clip": 0.01075919,
      "auxiliary_loss_mlp": 0.0102812,
      "balance_loss_clip": 1.04047191,
      "balance_loss_mlp": 1.0157702,
      "epoch": 0.507861115286337,
      "flos": 20700490757760.0,
      "grad_norm": 1.708007794034752,
      "language_loss": 0.80623627,
      "learning_rate": 2.04614711357029e-06,
      "loss": 0.82727671,
      "num_input_tokens_seen": 181637120,
      "step": 8447,
      "time_per_iteration": 2.5604207515716553
    },
    {
      "auxiliary_loss_clip": 0.01103641,
      "auxiliary_loss_mlp": 0.01031158,
      "balance_loss_clip": 1.04045546,
      "balance_loss_mlp": 1.01831365,
      "epoch": 0.507921238539005,
      "flos": 30847050576000.0,
      "grad_norm": 1.53814364491918,
      "language_loss": 0.70370436,
      "learning_rate": 2.0457577547026916e-06,
      "loss": 0.72505236,
      "num_input_tokens_seen": 181659965,
      "step": 8448,
      "time_per_iteration": 2.5771970748901367
    },
    {
      "auxiliary_loss_clip": 0.01113468,
      "auxiliary_loss_mlp": 0.00784896,
      "balance_loss_clip": 1.04071712,
      "balance_loss_mlp": 1.01076531,
      "epoch": 0.507981361791673,
      "flos": 35700197984640.0,
      "grad_norm": 1.4912174781121486,
      "language_loss": 0.72009009,
      "learning_rate": 2.045368394099955e-06,
      "loss": 0.73907375,
      "num_input_tokens_seen": 181685290,
      "step": 8449,
      "time_per_iteration": 2.592029333114624
    },
    {
      "auxiliary_loss_clip": 0.01088503,
      "auxiliary_loss_mlp": 0.01028557,
      "balance_loss_clip": 1.03767896,
      "balance_loss_mlp": 1.0164572,
      "epoch": 0.5080414850443409,
      "flos": 27161466750720.0,
      "grad_norm": 1.4206283014566663,
      "language_loss": 0.72786367,
      "learning_rate": 2.044979031776844e-06,
      "loss": 0.74903429,
      "num_input_tokens_seen": 181706080,
      "step": 8450,
      "time_per_iteration": 2.5862224102020264
    },
    {
      "auxiliary_loss_clip": 0.01117151,
      "auxiliary_loss_mlp": 0.01033525,
      "balance_loss_clip": 1.0414958,
      "balance_loss_mlp": 1.02081084,
      "epoch": 0.5081016082970089,
      "flos": 27085192220160.0,
      "grad_norm": 1.7650605983807157,
      "language_loss": 0.76565534,
      "learning_rate": 2.0445896677481234e-06,
      "loss": 0.78716207,
      "num_input_tokens_seen": 181724805,
      "step": 8451,
      "time_per_iteration": 2.4883201122283936
    },
    {
      "auxiliary_loss_clip": 0.01115838,
      "auxiliary_loss_mlp": 0.01034954,
      "balance_loss_clip": 1.04077506,
      "balance_loss_mlp": 1.02258015,
      "epoch": 0.5081617315496768,
      "flos": 22856531690880.0,
      "grad_norm": 1.6642244546573524,
      "language_loss": 0.84962279,
      "learning_rate": 2.044200302028559e-06,
      "loss": 0.8711307,
      "num_input_tokens_seen": 181743725,
      "step": 8452,
      "time_per_iteration": 2.483795166015625
    },
    {
      "auxiliary_loss_clip": 0.01118991,
      "auxiliary_loss_mlp": 0.01034649,
      "balance_loss_clip": 1.04135275,
      "balance_loss_mlp": 1.02131534,
      "epoch": 0.5082218548023448,
      "flos": 16281898087680.0,
      "grad_norm": 2.687465201865233,
      "language_loss": 0.77168679,
      "learning_rate": 2.0438109346329143e-06,
      "loss": 0.79322326,
      "num_input_tokens_seen": 181757720,
      "step": 8453,
      "time_per_iteration": 2.4038193225860596
    },
    {
      "auxiliary_loss_clip": 0.0107959,
      "auxiliary_loss_mlp": 0.01034765,
      "balance_loss_clip": 1.03957427,
      "balance_loss_mlp": 1.02253401,
      "epoch": 0.5082819780550127,
      "flos": 24460768915200.0,
      "grad_norm": 1.71255675120932,
      "language_loss": 0.76246387,
      "learning_rate": 2.0434215655759544e-06,
      "loss": 0.78360742,
      "num_input_tokens_seen": 181778545,
      "step": 8454,
      "time_per_iteration": 3.948051929473877
    },
    {
      "auxiliary_loss_clip": 0.01094485,
      "auxiliary_loss_mlp": 0.01034388,
      "balance_loss_clip": 1.03987026,
      "balance_loss_mlp": 1.02091146,
      "epoch": 0.5083421013076808,
      "flos": 23403271582080.0,
      "grad_norm": 1.7522330201822223,
      "language_loss": 0.89191318,
      "learning_rate": 2.0430321948724446e-06,
      "loss": 0.91320187,
      "num_input_tokens_seen": 181799495,
      "step": 8455,
      "time_per_iteration": 2.5544421672821045
    },
    {
      "auxiliary_loss_clip": 0.01100648,
      "auxiliary_loss_mlp": 0.00786775,
      "balance_loss_clip": 1.04043043,
      "balance_loss_mlp": 1.00923765,
      "epoch": 0.5084022245603487,
      "flos": 23872695448320.0,
      "grad_norm": 1.729811049759438,
      "language_loss": 0.62270033,
      "learning_rate": 2.042642822537149e-06,
      "loss": 0.6415745,
      "num_input_tokens_seen": 181818400,
      "step": 8456,
      "time_per_iteration": 2.5391440391540527
    },
    {
      "auxiliary_loss_clip": 0.01031895,
      "auxiliary_loss_mlp": 0.01000606,
      "balance_loss_clip": 1.01633096,
      "balance_loss_mlp": 0.99928904,
      "epoch": 0.5084623478130167,
      "flos": 62873336655360.0,
      "grad_norm": 0.8328637254632598,
      "language_loss": 0.62473732,
      "learning_rate": 2.0422534485848343e-06,
      "loss": 0.64506233,
      "num_input_tokens_seen": 181875975,
      "step": 8457,
      "time_per_iteration": 4.340438604354858
    },
    {
      "auxiliary_loss_clip": 0.01108196,
      "auxiliary_loss_mlp": 0.01031485,
      "balance_loss_clip": 1.0414288,
      "balance_loss_mlp": 1.01794863,
      "epoch": 0.5085224710656846,
      "flos": 22346133384960.0,
      "grad_norm": 1.6915483785118888,
      "language_loss": 0.67247725,
      "learning_rate": 2.0418640730302644e-06,
      "loss": 0.69387406,
      "num_input_tokens_seen": 181896450,
      "step": 8458,
      "time_per_iteration": 2.509958505630493
    },
    {
      "auxiliary_loss_clip": 0.01102526,
      "auxiliary_loss_mlp": 0.01031608,
      "balance_loss_clip": 1.0378058,
      "balance_loss_mlp": 1.01850045,
      "epoch": 0.5085825943183526,
      "flos": 26066263115520.0,
      "grad_norm": 1.7673153416869285,
      "language_loss": 0.77605283,
      "learning_rate": 2.0414746958882043e-06,
      "loss": 0.79739422,
      "num_input_tokens_seen": 181916770,
      "step": 8459,
      "time_per_iteration": 2.531630039215088
    },
    {
      "auxiliary_loss_clip": 0.01123367,
      "auxiliary_loss_mlp": 0.01035485,
      "balance_loss_clip": 1.04563808,
      "balance_loss_mlp": 1.02196085,
      "epoch": 0.5086427175710206,
      "flos": 17420733768960.0,
      "grad_norm": 1.883996362320534,
      "language_loss": 0.80658954,
      "learning_rate": 2.0410853171734196e-06,
      "loss": 0.82817811,
      "num_input_tokens_seen": 181932710,
      "step": 8460,
      "time_per_iteration": 3.8592710494995117
    },
    {
      "auxiliary_loss_clip": 0.01094328,
      "auxiliary_loss_mlp": 0.01036103,
      "balance_loss_clip": 1.03921711,
      "balance_loss_mlp": 1.02313304,
      "epoch": 0.5087028408236886,
      "flos": 20631758083200.0,
      "grad_norm": 1.6313074723916021,
      "language_loss": 0.68980557,
      "learning_rate": 2.0406959369006754e-06,
      "loss": 0.71110988,
      "num_input_tokens_seen": 181950665,
      "step": 8461,
      "time_per_iteration": 2.5057876110076904
    },
    {
      "auxiliary_loss_clip": 0.01112191,
      "auxiliary_loss_mlp": 0.0103037,
      "balance_loss_clip": 1.03963757,
      "balance_loss_mlp": 1.01745987,
      "epoch": 0.5087629640763566,
      "flos": 25593822506880.0,
      "grad_norm": 1.6392007372773147,
      "language_loss": 0.7583248,
      "learning_rate": 2.0403065550847375e-06,
      "loss": 0.77975041,
      "num_input_tokens_seen": 181971270,
      "step": 8462,
      "time_per_iteration": 2.491183280944824
    },
    {
      "auxiliary_loss_clip": 0.01079014,
      "auxiliary_loss_mlp": 0.01037238,
      "balance_loss_clip": 1.04060483,
      "balance_loss_mlp": 1.0228076,
      "epoch": 0.5088230873290245,
      "flos": 13261631927040.0,
      "grad_norm": 1.978030416566282,
      "language_loss": 0.81222528,
      "learning_rate": 2.0399171717403706e-06,
      "loss": 0.83338779,
      "num_input_tokens_seen": 181988410,
      "step": 8463,
      "time_per_iteration": 2.526637315750122
    },
    {
      "auxiliary_loss_clip": 0.01100473,
      "auxiliary_loss_mlp": 0.01038952,
      "balance_loss_clip": 1.04120779,
      "balance_loss_mlp": 1.02580345,
      "epoch": 0.5088832105816925,
      "flos": 20043469134720.0,
      "grad_norm": 1.7174502814789807,
      "language_loss": 0.76133454,
      "learning_rate": 2.039527786882341e-06,
      "loss": 0.78272879,
      "num_input_tokens_seen": 182006530,
      "step": 8464,
      "time_per_iteration": 3.9028267860412598
    },
    {
      "auxiliary_loss_clip": 0.01030279,
      "auxiliary_loss_mlp": 0.01001937,
      "balance_loss_clip": 1.01459348,
      "balance_loss_mlp": 1.0005722,
      "epoch": 0.5089433338343604,
      "flos": 67422179018880.0,
      "grad_norm": 0.6839108282888562,
      "language_loss": 0.5941779,
      "learning_rate": 2.0391384005254133e-06,
      "loss": 0.61449999,
      "num_input_tokens_seen": 182074240,
      "step": 8465,
      "time_per_iteration": 3.2179949283599854
    },
    {
      "auxiliary_loss_clip": 0.01112183,
      "auxiliary_loss_mlp": 0.01032277,
      "balance_loss_clip": 1.03861642,
      "balance_loss_mlp": 1.01977825,
      "epoch": 0.5090034570870284,
      "flos": 22710339336960.0,
      "grad_norm": 1.8699010143913102,
      "language_loss": 0.80060565,
      "learning_rate": 2.038749012684354e-06,
      "loss": 0.82205033,
      "num_input_tokens_seen": 182093360,
      "step": 8466,
      "time_per_iteration": 2.465264081954956
    },
    {
      "auxiliary_loss_clip": 0.01103136,
      "auxiliary_loss_mlp": 0.01029446,
      "balance_loss_clip": 1.03935778,
      "balance_loss_mlp": 1.01671422,
      "epoch": 0.5090635803396963,
      "flos": 20445812352000.0,
      "grad_norm": 1.5288649731756847,
      "language_loss": 0.78713989,
      "learning_rate": 2.0383596233739286e-06,
      "loss": 0.80846578,
      "num_input_tokens_seen": 182110170,
      "step": 8467,
      "time_per_iteration": 2.4934592247009277
    },
    {
      "auxiliary_loss_clip": 0.01111274,
      "auxiliary_loss_mlp": 0.01029534,
      "balance_loss_clip": 1.03986096,
      "balance_loss_mlp": 1.01694512,
      "epoch": 0.5091237035923644,
      "flos": 23768878164480.0,
      "grad_norm": 1.6737165431161254,
      "language_loss": 0.74663037,
      "learning_rate": 2.0379702326089013e-06,
      "loss": 0.76803845,
      "num_input_tokens_seen": 182129570,
      "step": 8468,
      "time_per_iteration": 2.472665309906006
    },
    {
      "auxiliary_loss_clip": 0.01111538,
      "auxiliary_loss_mlp": 0.0103153,
      "balance_loss_clip": 1.03814864,
      "balance_loss_mlp": 1.01871467,
      "epoch": 0.5091838268450323,
      "flos": 18327908684160.0,
      "grad_norm": 1.7516080704623367,
      "language_loss": 0.77903152,
      "learning_rate": 2.03758084040404e-06,
      "loss": 0.80046225,
      "num_input_tokens_seen": 182147565,
      "step": 8469,
      "time_per_iteration": 2.4365696907043457
    },
    {
      "auxiliary_loss_clip": 0.0109964,
      "auxiliary_loss_mlp": 0.01035044,
      "balance_loss_clip": 1.04064012,
      "balance_loss_mlp": 1.02096486,
      "epoch": 0.5092439500977003,
      "flos": 29057621806080.0,
      "grad_norm": 1.5561405095662946,
      "language_loss": 0.69820547,
      "learning_rate": 2.037191446774109e-06,
      "loss": 0.71955228,
      "num_input_tokens_seen": 182169695,
      "step": 8470,
      "time_per_iteration": 2.5422539710998535
    },
    {
      "auxiliary_loss_clip": 0.01089575,
      "auxiliary_loss_mlp": 0.01042589,
      "balance_loss_clip": 1.03766513,
      "balance_loss_mlp": 1.02772939,
      "epoch": 0.5093040733503682,
      "flos": 13553908894080.0,
      "grad_norm": 2.0002968520619584,
      "language_loss": 0.73626792,
      "learning_rate": 2.0368020517338745e-06,
      "loss": 0.75758952,
      "num_input_tokens_seen": 182186385,
      "step": 8471,
      "time_per_iteration": 2.485443353652954
    },
    {
      "auxiliary_loss_clip": 0.01038483,
      "auxiliary_loss_mlp": 0.01000802,
      "balance_loss_clip": 1.01219559,
      "balance_loss_mlp": 0.99951476,
      "epoch": 0.5093641966030362,
      "flos": 68906617407360.0,
      "grad_norm": 0.7482423278703207,
      "language_loss": 0.5804947,
      "learning_rate": 2.036412655298103e-06,
      "loss": 0.60088754,
      "num_input_tokens_seen": 182247095,
      "step": 8472,
      "time_per_iteration": 3.0671560764312744
    },
    {
      "auxiliary_loss_clip": 0.01064387,
      "auxiliary_loss_mlp": 0.01033355,
      "balance_loss_clip": 1.03913081,
      "balance_loss_mlp": 1.02104032,
      "epoch": 0.5094243198557042,
      "flos": 21580948932480.0,
      "grad_norm": 2.0267896110819326,
      "language_loss": 0.69016147,
      "learning_rate": 2.03602325748156e-06,
      "loss": 0.71113884,
      "num_input_tokens_seen": 182266380,
      "step": 8473,
      "time_per_iteration": 2.5876779556274414
    },
    {
      "auxiliary_loss_clip": 0.01092375,
      "auxiliary_loss_mlp": 0.010308,
      "balance_loss_clip": 1.04048944,
      "balance_loss_mlp": 1.01805103,
      "epoch": 0.5094844431083722,
      "flos": 28840721529600.0,
      "grad_norm": 1.8318567928214506,
      "language_loss": 0.85659617,
      "learning_rate": 2.0356338582990105e-06,
      "loss": 0.87782788,
      "num_input_tokens_seen": 182284685,
      "step": 8474,
      "time_per_iteration": 2.5861573219299316
    },
    {
      "auxiliary_loss_clip": 0.01098306,
      "auxiliary_loss_mlp": 0.01031427,
      "balance_loss_clip": 1.04160357,
      "balance_loss_mlp": 1.0185883,
      "epoch": 0.5095445663610402,
      "flos": 14976114969600.0,
      "grad_norm": 2.1489455909799644,
      "language_loss": 0.64890987,
      "learning_rate": 2.035244457765222e-06,
      "loss": 0.67020726,
      "num_input_tokens_seen": 182301810,
      "step": 8475,
      "time_per_iteration": 2.5175931453704834
    },
    {
      "auxiliary_loss_clip": 0.01097527,
      "auxiliary_loss_mlp": 0.01035871,
      "balance_loss_clip": 1.03992915,
      "balance_loss_mlp": 1.02130413,
      "epoch": 0.5096046896137081,
      "flos": 20777088510720.0,
      "grad_norm": 2.6617836706042057,
      "language_loss": 0.81966078,
      "learning_rate": 2.0348550558949605e-06,
      "loss": 0.84099472,
      "num_input_tokens_seen": 182320285,
      "step": 8476,
      "time_per_iteration": 2.5134925842285156
    },
    {
      "auxiliary_loss_clip": 0.01063856,
      "auxiliary_loss_mlp": 0.01038334,
      "balance_loss_clip": 1.03732347,
      "balance_loss_mlp": 1.02238393,
      "epoch": 0.5096648128663761,
      "flos": 23185078416000.0,
      "grad_norm": 1.9236571564381704,
      "language_loss": 0.80454385,
      "learning_rate": 2.0344656527029917e-06,
      "loss": 0.8255657,
      "num_input_tokens_seen": 182339465,
      "step": 8477,
      "time_per_iteration": 2.6243512630462646
    },
    {
      "auxiliary_loss_clip": 0.01091397,
      "auxiliary_loss_mlp": 0.01028516,
      "balance_loss_clip": 1.04028118,
      "balance_loss_mlp": 1.01356149,
      "epoch": 0.509724936119044,
      "flos": 22309432663680.0,
      "grad_norm": 1.9018708146122782,
      "language_loss": 0.61918348,
      "learning_rate": 2.034076248204082e-06,
      "loss": 0.64038259,
      "num_input_tokens_seen": 182358375,
      "step": 8478,
      "time_per_iteration": 2.51712703704834
    },
    {
      "auxiliary_loss_clip": 0.01099827,
      "auxiliary_loss_mlp": 0.01038606,
      "balance_loss_clip": 1.04236007,
      "balance_loss_mlp": 1.02625585,
      "epoch": 0.509785059371712,
      "flos": 26287077974400.0,
      "grad_norm": 1.7747420593673056,
      "language_loss": 0.66177303,
      "learning_rate": 2.0336868424129968e-06,
      "loss": 0.68315732,
      "num_input_tokens_seen": 182377935,
      "step": 8479,
      "time_per_iteration": 2.5715749263763428
    },
    {
      "auxiliary_loss_clip": 0.01101688,
      "auxiliary_loss_mlp": 0.01031518,
      "balance_loss_clip": 1.04017591,
      "balance_loss_mlp": 1.01907277,
      "epoch": 0.50984518262438,
      "flos": 22964586779520.0,
      "grad_norm": 1.6559938122407516,
      "language_loss": 0.6949656,
      "learning_rate": 2.0332974353445037e-06,
      "loss": 0.71629769,
      "num_input_tokens_seen": 182396440,
      "step": 8480,
      "time_per_iteration": 2.5230281352996826
    },
    {
      "auxiliary_loss_clip": 0.01114796,
      "auxiliary_loss_mlp": 0.01033291,
      "balance_loss_clip": 1.0383774,
      "balance_loss_mlp": 1.02019036,
      "epoch": 0.509905305877048,
      "flos": 26213389223040.0,
      "grad_norm": 2.2008227530417273,
      "language_loss": 0.79506421,
      "learning_rate": 2.0329080270133688e-06,
      "loss": 0.81654513,
      "num_input_tokens_seen": 182415890,
      "step": 8481,
      "time_per_iteration": 2.5043692588806152
    },
    {
      "auxiliary_loss_clip": 0.01096248,
      "auxiliary_loss_mlp": 0.0103316,
      "balance_loss_clip": 1.0368284,
      "balance_loss_mlp": 1.02032709,
      "epoch": 0.5099654291297159,
      "flos": 20340055733760.0,
      "grad_norm": 1.6021707058205554,
      "language_loss": 0.83377618,
      "learning_rate": 2.0325186174343578e-06,
      "loss": 0.85507029,
      "num_input_tokens_seen": 182434235,
      "step": 8482,
      "time_per_iteration": 2.475356340408325
    },
    {
      "auxiliary_loss_clip": 0.01107841,
      "auxiliary_loss_mlp": 0.00788308,
      "balance_loss_clip": 1.03903186,
      "balance_loss_mlp": 1.01259327,
      "epoch": 0.5100255523823839,
      "flos": 29054820545280.0,
      "grad_norm": 6.717126338626177,
      "language_loss": 0.8581934,
      "learning_rate": 2.032129206622238e-06,
      "loss": 0.87715483,
      "num_input_tokens_seen": 182454360,
      "step": 8483,
      "time_per_iteration": 2.5575509071350098
    },
    {
      "auxiliary_loss_clip": 0.01101165,
      "auxiliary_loss_mlp": 0.01032976,
      "balance_loss_clip": 1.03721094,
      "balance_loss_mlp": 1.02063835,
      "epoch": 0.5100856756350518,
      "flos": 22455912326400.0,
      "grad_norm": 2.3297139326039487,
      "language_loss": 0.8309195,
      "learning_rate": 2.031739794591775e-06,
      "loss": 0.85226095,
      "num_input_tokens_seen": 182471940,
      "step": 8484,
      "time_per_iteration": 2.4820194244384766
    },
    {
      "auxiliary_loss_clip": 0.01090087,
      "auxiliary_loss_mlp": 0.01029557,
      "balance_loss_clip": 1.04021943,
      "balance_loss_mlp": 1.01599669,
      "epoch": 0.5101457988877198,
      "flos": 19171055606400.0,
      "grad_norm": 2.338739298438024,
      "language_loss": 0.81667411,
      "learning_rate": 2.031350381357736e-06,
      "loss": 0.8378706,
      "num_input_tokens_seen": 182490685,
      "step": 8485,
      "time_per_iteration": 2.5058841705322266
    },
    {
      "auxiliary_loss_clip": 0.01085618,
      "auxiliary_loss_mlp": 0.01030695,
      "balance_loss_clip": 1.03466165,
      "balance_loss_mlp": 1.01818943,
      "epoch": 0.5102059221403878,
      "flos": 14866371941760.0,
      "grad_norm": 1.9322666227458032,
      "language_loss": 0.74030137,
      "learning_rate": 2.0309609669348874e-06,
      "loss": 0.76146448,
      "num_input_tokens_seen": 182508325,
      "step": 8486,
      "time_per_iteration": 2.490628242492676
    },
    {
      "auxiliary_loss_clip": 0.01072678,
      "auxiliary_loss_mlp": 0.01028349,
      "balance_loss_clip": 1.03834391,
      "balance_loss_mlp": 1.01427603,
      "epoch": 0.5102660453930558,
      "flos": 22961103160320.0,
      "grad_norm": 1.5000149869592727,
      "language_loss": 0.70107067,
      "learning_rate": 2.0305715513379953e-06,
      "loss": 0.72208089,
      "num_input_tokens_seen": 182527020,
      "step": 8487,
      "time_per_iteration": 2.564749002456665
    },
    {
      "auxiliary_loss_clip": 0.01090768,
      "auxiliary_loss_mlp": 0.01031025,
      "balance_loss_clip": 1.038239,
      "balance_loss_mlp": 1.01733959,
      "epoch": 0.5103261686457238,
      "flos": 23149311448320.0,
      "grad_norm": 2.521989436124946,
      "language_loss": 0.7265504,
      "learning_rate": 2.030182134581827e-06,
      "loss": 0.74776828,
      "num_input_tokens_seen": 182543505,
      "step": 8488,
      "time_per_iteration": 2.5093986988067627
    },
    {
      "auxiliary_loss_clip": 0.01073016,
      "auxiliary_loss_mlp": 0.00788837,
      "balance_loss_clip": 1.03661394,
      "balance_loss_mlp": 1.01471996,
      "epoch": 0.5103862918983917,
      "flos": 14319237000960.0,
      "grad_norm": 2.5924326841466256,
      "language_loss": 0.69448793,
      "learning_rate": 2.0297927166811503e-06,
      "loss": 0.71310651,
      "num_input_tokens_seen": 182562250,
      "step": 8489,
      "time_per_iteration": 2.5387628078460693
    },
    {
      "auxiliary_loss_clip": 0.01095016,
      "auxiliary_loss_mlp": 0.0102908,
      "balance_loss_clip": 1.03738999,
      "balance_loss_mlp": 1.0165453,
      "epoch": 0.5104464151510597,
      "flos": 25848536826240.0,
      "grad_norm": 1.8229367280545934,
      "language_loss": 0.72329259,
      "learning_rate": 2.0294032976507297e-06,
      "loss": 0.74453354,
      "num_input_tokens_seen": 182581910,
      "step": 8490,
      "time_per_iteration": 2.5432918071746826
    },
    {
      "auxiliary_loss_clip": 0.0108759,
      "auxiliary_loss_mlp": 0.01027577,
      "balance_loss_clip": 1.04130125,
      "balance_loss_mlp": 1.01561427,
      "epoch": 0.5105065384037276,
      "flos": 21652913831040.0,
      "grad_norm": 1.5132712692514443,
      "language_loss": 0.80656165,
      "learning_rate": 2.0290138775053337e-06,
      "loss": 0.82771337,
      "num_input_tokens_seen": 182601350,
      "step": 8491,
      "time_per_iteration": 2.526655912399292
    },
    {
      "auxiliary_loss_clip": 0.01099633,
      "auxiliary_loss_mlp": 0.0102917,
      "balance_loss_clip": 1.03737497,
      "balance_loss_mlp": 1.01673663,
      "epoch": 0.5105666616563956,
      "flos": 22491571553280.0,
      "grad_norm": 4.285961807182633,
      "language_loss": 0.78407836,
      "learning_rate": 2.028624456259728e-06,
      "loss": 0.80536634,
      "num_input_tokens_seen": 182619660,
      "step": 8492,
      "time_per_iteration": 3.8466880321502686
    },
    {
      "auxiliary_loss_clip": 0.01082895,
      "auxiliary_loss_mlp": 0.01040736,
      "balance_loss_clip": 1.0390631,
      "balance_loss_mlp": 1.0270927,
      "epoch": 0.5106267849090635,
      "flos": 22455768672000.0,
      "grad_norm": 2.3670580246367523,
      "language_loss": 0.77883375,
      "learning_rate": 2.0282350339286804e-06,
      "loss": 0.80007011,
      "num_input_tokens_seen": 182639815,
      "step": 8493,
      "time_per_iteration": 2.54646635055542
    },
    {
      "auxiliary_loss_clip": 0.01072955,
      "auxiliary_loss_mlp": 0.01025611,
      "balance_loss_clip": 1.03940129,
      "balance_loss_mlp": 1.01153219,
      "epoch": 0.5106869081617316,
      "flos": 23547093638400.0,
      "grad_norm": 2.331740091399332,
      "language_loss": 0.83726615,
      "learning_rate": 2.0278456105269574e-06,
      "loss": 0.85825181,
      "num_input_tokens_seen": 182659655,
      "step": 8494,
      "time_per_iteration": 2.5968892574310303
    },
    {
      "auxiliary_loss_clip": 0.01115586,
      "auxiliary_loss_mlp": 0.010336,
      "balance_loss_clip": 1.0398041,
      "balance_loss_mlp": 1.02146435,
      "epoch": 0.5107470314143995,
      "flos": 26792987080320.0,
      "grad_norm": 2.0799626050217106,
      "language_loss": 0.79026008,
      "learning_rate": 2.027456186069326e-06,
      "loss": 0.81175196,
      "num_input_tokens_seen": 182677075,
      "step": 8495,
      "time_per_iteration": 3.869382619857788
    },
    {
      "auxiliary_loss_clip": 0.01080485,
      "auxiliary_loss_mlp": 0.01031055,
      "balance_loss_clip": 1.03772688,
      "balance_loss_mlp": 1.01788282,
      "epoch": 0.5108071546670675,
      "flos": 25739691638400.0,
      "grad_norm": 2.4993039138549844,
      "language_loss": 0.78538018,
      "learning_rate": 2.0270667605705535e-06,
      "loss": 0.80649555,
      "num_input_tokens_seen": 182699625,
      "step": 8496,
      "time_per_iteration": 2.6154603958129883
    },
    {
      "auxiliary_loss_clip": 0.01099764,
      "auxiliary_loss_mlp": 0.01027523,
      "balance_loss_clip": 1.03796947,
      "balance_loss_mlp": 1.01533961,
      "epoch": 0.5108672779197354,
      "flos": 18697537589760.0,
      "grad_norm": 1.9278343638979454,
      "language_loss": 0.78665555,
      "learning_rate": 2.0266773340454066e-06,
      "loss": 0.80792844,
      "num_input_tokens_seen": 182717020,
      "step": 8497,
      "time_per_iteration": 2.4776406288146973
    },
    {
      "auxiliary_loss_clip": 0.01111241,
      "auxiliary_loss_mlp": 0.01031168,
      "balance_loss_clip": 1.03771973,
      "balance_loss_mlp": 1.01863289,
      "epoch": 0.5109274011724034,
      "flos": 26688164215680.0,
      "grad_norm": 1.5942703418103918,
      "language_loss": 0.81735247,
      "learning_rate": 2.0262879065086525e-06,
      "loss": 0.83877659,
      "num_input_tokens_seen": 182736955,
      "step": 8498,
      "time_per_iteration": 2.501861095428467
    },
    {
      "auxiliary_loss_clip": 0.010833,
      "auxiliary_loss_mlp": 0.00786907,
      "balance_loss_clip": 1.0403372,
      "balance_loss_mlp": 1.00919867,
      "epoch": 0.5109875244250714,
      "flos": 22784028088320.0,
      "grad_norm": 1.6919608992134287,
      "language_loss": 0.70547915,
      "learning_rate": 2.0258984779750584e-06,
      "loss": 0.72418123,
      "num_input_tokens_seen": 182757620,
      "step": 8499,
      "time_per_iteration": 4.068626403808594
    },
    {
      "auxiliary_loss_clip": 0.01058926,
      "auxiliary_loss_mlp": 0.01036769,
      "balance_loss_clip": 1.03816581,
      "balance_loss_mlp": 1.02232075,
      "epoch": 0.5110476476777394,
      "flos": 35588515622400.0,
      "grad_norm": 1.452259863659716,
      "language_loss": 0.72447097,
      "learning_rate": 2.0255090484593914e-06,
      "loss": 0.74542797,
      "num_input_tokens_seen": 182780195,
      "step": 8500,
      "time_per_iteration": 2.688023328781128
    },
    {
      "auxiliary_loss_clip": 0.01106402,
      "auxiliary_loss_mlp": 0.01033261,
      "balance_loss_clip": 1.0378406,
      "balance_loss_mlp": 1.01897955,
      "epoch": 0.5111077709304074,
      "flos": 19280798634240.0,
      "grad_norm": 2.7629780418112753,
      "language_loss": 0.63141894,
      "learning_rate": 2.0251196179764183e-06,
      "loss": 0.65281558,
      "num_input_tokens_seen": 182795765,
      "step": 8501,
      "time_per_iteration": 2.463298797607422
    },
    {
      "auxiliary_loss_clip": 0.01112202,
      "auxiliary_loss_mlp": 0.01033182,
      "balance_loss_clip": 1.03539145,
      "balance_loss_mlp": 1.01990843,
      "epoch": 0.5111678941830753,
      "flos": 20668207409280.0,
      "grad_norm": 1.8090197384645332,
      "language_loss": 0.87373227,
      "learning_rate": 2.024730186540907e-06,
      "loss": 0.89518607,
      "num_input_tokens_seen": 182813120,
      "step": 8502,
      "time_per_iteration": 3.8608434200286865
    },
    {
      "auxiliary_loss_clip": 0.01096829,
      "auxiliary_loss_mlp": 0.01032275,
      "balance_loss_clip": 1.03561354,
      "balance_loss_mlp": 1.0204916,
      "epoch": 0.5112280174357433,
      "flos": 26287903987200.0,
      "grad_norm": 4.329037807519145,
      "language_loss": 0.822294,
      "learning_rate": 2.0243407541676253e-06,
      "loss": 0.84358501,
      "num_input_tokens_seen": 182835745,
      "step": 8503,
      "time_per_iteration": 2.5430634021759033
    },
    {
      "auxiliary_loss_clip": 0.01018893,
      "auxiliary_loss_mlp": 0.01004878,
      "balance_loss_clip": 1.0172354,
      "balance_loss_mlp": 1.00352514,
      "epoch": 0.5112881406884112,
      "flos": 59474247707520.0,
      "grad_norm": 0.8571187369412036,
      "language_loss": 0.63892269,
      "learning_rate": 2.023951320871339e-06,
      "loss": 0.65916044,
      "num_input_tokens_seen": 182892540,
      "step": 8504,
      "time_per_iteration": 3.1448419094085693
    },
    {
      "auxiliary_loss_clip": 0.01079456,
      "auxiliary_loss_mlp": 0.00785556,
      "balance_loss_clip": 1.03706396,
      "balance_loss_mlp": 1.00715065,
      "epoch": 0.5113482639410792,
      "flos": 26468857728000.0,
      "grad_norm": 1.8494964572134716,
      "language_loss": 0.8439728,
      "learning_rate": 2.023561886666816e-06,
      "loss": 0.86262298,
      "num_input_tokens_seen": 182911515,
      "step": 8505,
      "time_per_iteration": 2.5878028869628906
    },
    {
      "auxiliary_loss_clip": 0.01101961,
      "auxiliary_loss_mlp": 0.01025019,
      "balance_loss_clip": 1.04045725,
      "balance_loss_mlp": 1.01266897,
      "epoch": 0.5114083871937471,
      "flos": 29895848565120.0,
      "grad_norm": 1.857848781181716,
      "language_loss": 0.74885678,
      "learning_rate": 2.0231724515688246e-06,
      "loss": 0.77012658,
      "num_input_tokens_seen": 182930860,
      "step": 8506,
      "time_per_iteration": 2.5313973426818848
    },
    {
      "auxiliary_loss_clip": 0.01114123,
      "auxiliary_loss_mlp": 0.01034216,
      "balance_loss_clip": 1.0384717,
      "balance_loss_mlp": 1.02012515,
      "epoch": 0.5114685104464152,
      "flos": 24314576561280.0,
      "grad_norm": 2.039076578613514,
      "language_loss": 0.57599789,
      "learning_rate": 2.022783015592131e-06,
      "loss": 0.59748125,
      "num_input_tokens_seen": 182949960,
      "step": 8507,
      "time_per_iteration": 2.465965986251831
    },
    {
      "auxiliary_loss_clip": 0.01106415,
      "auxiliary_loss_mlp": 0.01044944,
      "balance_loss_clip": 1.04207301,
      "balance_loss_mlp": 1.03075171,
      "epoch": 0.5115286336990831,
      "flos": 17019288391680.0,
      "grad_norm": 1.876484982791679,
      "language_loss": 0.8539052,
      "learning_rate": 2.022393578751503e-06,
      "loss": 0.87541878,
      "num_input_tokens_seen": 182968085,
      "step": 8508,
      "time_per_iteration": 2.4681830406188965
    },
    {
      "auxiliary_loss_clip": 0.01083082,
      "auxiliary_loss_mlp": 0.00787266,
      "balance_loss_clip": 1.04010773,
      "balance_loss_mlp": 1.00973368,
      "epoch": 0.5115887569517511,
      "flos": 23659386531840.0,
      "grad_norm": 1.6843393329277068,
      "language_loss": 0.72307301,
      "learning_rate": 2.022004141061709e-06,
      "loss": 0.74177647,
      "num_input_tokens_seen": 182987275,
      "step": 8509,
      "time_per_iteration": 2.540834426879883
    },
    {
      "auxiliary_loss_clip": 0.01111678,
      "auxiliary_loss_mlp": 0.00787104,
      "balance_loss_clip": 1.03836727,
      "balance_loss_mlp": 1.01188612,
      "epoch": 0.511648880204419,
      "flos": 16107193313280.0,
      "grad_norm": 1.8705904004860707,
      "language_loss": 0.75991559,
      "learning_rate": 2.0216147025375153e-06,
      "loss": 0.77890337,
      "num_input_tokens_seen": 183004700,
      "step": 8510,
      "time_per_iteration": 2.4169161319732666
    },
    {
      "auxiliary_loss_clip": 0.01112732,
      "auxiliary_loss_mlp": 0.01034187,
      "balance_loss_clip": 1.03974283,
      "balance_loss_mlp": 1.02196169,
      "epoch": 0.511709003457087,
      "flos": 32634970974720.0,
      "grad_norm": 1.528929768791269,
      "language_loss": 0.70965147,
      "learning_rate": 2.0212252631936907e-06,
      "loss": 0.73112065,
      "num_input_tokens_seen": 183025830,
      "step": 8511,
      "time_per_iteration": 2.529033660888672
    },
    {
      "auxiliary_loss_clip": 0.010885,
      "auxiliary_loss_mlp": 0.01027795,
      "balance_loss_clip": 1.03839064,
      "balance_loss_mlp": 1.01461029,
      "epoch": 0.511769126709755,
      "flos": 21762082241280.0,
      "grad_norm": 2.6962623091539584,
      "language_loss": 0.6648218,
      "learning_rate": 2.020835823045001e-06,
      "loss": 0.68598473,
      "num_input_tokens_seen": 183045140,
      "step": 8512,
      "time_per_iteration": 2.502185821533203
    },
    {
      "auxiliary_loss_clip": 0.01060885,
      "auxiliary_loss_mlp": 0.01036956,
      "balance_loss_clip": 1.03628612,
      "balance_loss_mlp": 1.02197146,
      "epoch": 0.511829249962423,
      "flos": 23915357827200.0,
      "grad_norm": 1.704397315355566,
      "language_loss": 0.66290194,
      "learning_rate": 2.0204463821062146e-06,
      "loss": 0.68388033,
      "num_input_tokens_seen": 183063935,
      "step": 8513,
      "time_per_iteration": 2.614583730697632
    },
    {
      "auxiliary_loss_clip": 0.01072289,
      "auxiliary_loss_mlp": 0.0103108,
      "balance_loss_clip": 1.0384922,
      "balance_loss_mlp": 1.01745391,
      "epoch": 0.511889373215091,
      "flos": 23727005884800.0,
      "grad_norm": 1.8400632377403194,
      "language_loss": 0.69256794,
      "learning_rate": 2.0200569403921e-06,
      "loss": 0.71360165,
      "num_input_tokens_seen": 183084135,
      "step": 8514,
      "time_per_iteration": 2.5788586139678955
    },
    {
      "auxiliary_loss_clip": 0.01111161,
      "auxiliary_loss_mlp": 0.01028245,
      "balance_loss_clip": 1.03791761,
      "balance_loss_mlp": 1.01582956,
      "epoch": 0.5119494964677589,
      "flos": 28111519526400.0,
      "grad_norm": 1.5418333678729363,
      "language_loss": 0.6604147,
      "learning_rate": 2.019667497917424e-06,
      "loss": 0.68180871,
      "num_input_tokens_seen": 183104570,
      "step": 8515,
      "time_per_iteration": 2.50140380859375
    },
    {
      "auxiliary_loss_clip": 0.0110073,
      "auxiliary_loss_mlp": 0.01031181,
      "balance_loss_clip": 1.0379529,
      "balance_loss_mlp": 1.01833582,
      "epoch": 0.5120096197204269,
      "flos": 24973214296320.0,
      "grad_norm": 2.0047258642285004,
      "language_loss": 0.75152266,
      "learning_rate": 2.019278054696955e-06,
      "loss": 0.77284169,
      "num_input_tokens_seen": 183123850,
      "step": 8516,
      "time_per_iteration": 2.5207293033599854
    },
    {
      "auxiliary_loss_clip": 0.01088858,
      "auxiliary_loss_mlp": 0.01037793,
      "balance_loss_clip": 1.04163325,
      "balance_loss_mlp": 1.02391684,
      "epoch": 0.5120697429730948,
      "flos": 17968012364160.0,
      "grad_norm": 1.9310224234446716,
      "language_loss": 0.77854902,
      "learning_rate": 2.0188886107454595e-06,
      "loss": 0.79981554,
      "num_input_tokens_seen": 183141725,
      "step": 8517,
      "time_per_iteration": 2.475924015045166
    },
    {
      "auxiliary_loss_clip": 0.01104853,
      "auxiliary_loss_mlp": 0.01033525,
      "balance_loss_clip": 1.03863788,
      "balance_loss_mlp": 1.01958334,
      "epoch": 0.5121298662257628,
      "flos": 23292343405440.0,
      "grad_norm": 2.6015770295702074,
      "language_loss": 0.73833001,
      "learning_rate": 2.0184991660777063e-06,
      "loss": 0.75971383,
      "num_input_tokens_seen": 183161300,
      "step": 8518,
      "time_per_iteration": 2.542253017425537
    },
    {
      "auxiliary_loss_clip": 0.01100062,
      "auxiliary_loss_mlp": 0.01036792,
      "balance_loss_clip": 1.03933835,
      "balance_loss_mlp": 1.0232265,
      "epoch": 0.5121899894784308,
      "flos": 17311062568320.0,
      "grad_norm": 1.8037453586835346,
      "language_loss": 0.78030556,
      "learning_rate": 2.0181097207084625e-06,
      "loss": 0.80167413,
      "num_input_tokens_seen": 183180495,
      "step": 8519,
      "time_per_iteration": 2.4592535495758057
    },
    {
      "auxiliary_loss_clip": 0.01115858,
      "auxiliary_loss_mlp": 0.01031742,
      "balance_loss_clip": 1.04075921,
      "balance_loss_mlp": 1.01802063,
      "epoch": 0.5122501127310988,
      "flos": 24930085040640.0,
      "grad_norm": 1.5129154003819458,
      "language_loss": 0.79281324,
      "learning_rate": 2.017720274652497e-06,
      "loss": 0.81428921,
      "num_input_tokens_seen": 183200330,
      "step": 8520,
      "time_per_iteration": 2.493725299835205
    },
    {
      "auxiliary_loss_clip": 0.0109773,
      "auxiliary_loss_mlp": 0.01038216,
      "balance_loss_clip": 1.03854382,
      "balance_loss_mlp": 1.02350557,
      "epoch": 0.5123102359837667,
      "flos": 18442859184000.0,
      "grad_norm": 1.6400908849105167,
      "language_loss": 0.81306684,
      "learning_rate": 2.0173308279245765e-06,
      "loss": 0.83442628,
      "num_input_tokens_seen": 183218230,
      "step": 8521,
      "time_per_iteration": 2.4813666343688965
    },
    {
      "auxiliary_loss_clip": 0.01100181,
      "auxiliary_loss_mlp": 0.0102917,
      "balance_loss_clip": 1.0343082,
      "balance_loss_mlp": 1.01547861,
      "epoch": 0.5123703592364347,
      "flos": 26684860164480.0,
      "grad_norm": 1.7728031028366473,
      "language_loss": 0.68200958,
      "learning_rate": 2.0169413805394692e-06,
      "loss": 0.70330298,
      "num_input_tokens_seen": 183236735,
      "step": 8522,
      "time_per_iteration": 2.4941482543945312
    },
    {
      "auxiliary_loss_clip": 0.01089388,
      "auxiliary_loss_mlp": 0.01040951,
      "balance_loss_clip": 1.04107571,
      "balance_loss_mlp": 1.02419055,
      "epoch": 0.5124304824891026,
      "flos": 28803948981120.0,
      "grad_norm": 1.94464710511069,
      "language_loss": 0.61520886,
      "learning_rate": 2.0165519325119433e-06,
      "loss": 0.63651228,
      "num_input_tokens_seen": 183257550,
      "step": 8523,
      "time_per_iteration": 2.590646266937256
    },
    {
      "auxiliary_loss_clip": 0.01078036,
      "auxiliary_loss_mlp": 0.01038553,
      "balance_loss_clip": 1.03704572,
      "balance_loss_mlp": 1.02492774,
      "epoch": 0.5124906057417706,
      "flos": 21761830846080.0,
      "grad_norm": 1.8659488603551109,
      "language_loss": 0.77699399,
      "learning_rate": 2.0161624838567656e-06,
      "loss": 0.79815984,
      "num_input_tokens_seen": 183275515,
      "step": 8524,
      "time_per_iteration": 2.531919479370117
    },
    {
      "auxiliary_loss_clip": 0.01090834,
      "auxiliary_loss_mlp": 0.01034098,
      "balance_loss_clip": 1.03833783,
      "balance_loss_mlp": 1.02115214,
      "epoch": 0.5125507289944387,
      "flos": 18880538405760.0,
      "grad_norm": 1.8235639507165855,
      "language_loss": 0.74825752,
      "learning_rate": 2.015773034588706e-06,
      "loss": 0.76950687,
      "num_input_tokens_seen": 183293880,
      "step": 8525,
      "time_per_iteration": 2.49753999710083
    },
    {
      "auxiliary_loss_clip": 0.01089542,
      "auxiliary_loss_mlp": 0.01040189,
      "balance_loss_clip": 1.03742075,
      "balance_loss_mlp": 1.02464437,
      "epoch": 0.5126108522471066,
      "flos": 35627838036480.0,
      "grad_norm": 1.6084871907089167,
      "language_loss": 0.74656641,
      "learning_rate": 2.015383584722531e-06,
      "loss": 0.76786363,
      "num_input_tokens_seen": 183315860,
      "step": 8526,
      "time_per_iteration": 2.6375234127044678
    },
    {
      "auxiliary_loss_clip": 0.01105222,
      "auxiliary_loss_mlp": 0.01036416,
      "balance_loss_clip": 1.03904283,
      "balance_loss_mlp": 1.02302313,
      "epoch": 0.5126709754997746,
      "flos": 20190918464640.0,
      "grad_norm": 1.633462355052535,
      "language_loss": 0.64857852,
      "learning_rate": 2.0149941342730088e-06,
      "loss": 0.66999495,
      "num_input_tokens_seen": 183335480,
      "step": 8527,
      "time_per_iteration": 2.4771621227264404
    },
    {
      "auxiliary_loss_clip": 0.01090536,
      "auxiliary_loss_mlp": 0.01039145,
      "balance_loss_clip": 1.0409534,
      "balance_loss_mlp": 1.02699161,
      "epoch": 0.5127310987524425,
      "flos": 18588548747520.0,
      "grad_norm": 1.4988698568027683,
      "language_loss": 0.74329489,
      "learning_rate": 2.014604683254908e-06,
      "loss": 0.76459175,
      "num_input_tokens_seen": 183354395,
      "step": 8528,
      "time_per_iteration": 2.528162717819214
    },
    {
      "auxiliary_loss_clip": 0.01100507,
      "auxiliary_loss_mlp": 0.0103178,
      "balance_loss_clip": 1.0365777,
      "balance_loss_mlp": 1.01848817,
      "epoch": 0.5127912220051105,
      "flos": 22454691264000.0,
      "grad_norm": 1.663616603328329,
      "language_loss": 0.83027667,
      "learning_rate": 2.014215231682995e-06,
      "loss": 0.85159957,
      "num_input_tokens_seen": 183372980,
      "step": 8529,
      "time_per_iteration": 2.4947621822357178
    },
    {
      "auxiliary_loss_clip": 0.01064836,
      "auxiliary_loss_mlp": 0.01033797,
      "balance_loss_clip": 1.03843641,
      "balance_loss_mlp": 1.0204215,
      "epoch": 0.5128513452577784,
      "flos": 19093703667840.0,
      "grad_norm": 1.9177734936778659,
      "language_loss": 0.73940903,
      "learning_rate": 2.01382577957204e-06,
      "loss": 0.76039541,
      "num_input_tokens_seen": 183390160,
      "step": 8530,
      "time_per_iteration": 2.565117597579956
    },
    {
      "auxiliary_loss_clip": 0.01011947,
      "auxiliary_loss_mlp": 0.01003082,
      "balance_loss_clip": 1.02039075,
      "balance_loss_mlp": 1.00172317,
      "epoch": 0.5129114685104464,
      "flos": 67892285243520.0,
      "grad_norm": 0.7513837790134419,
      "language_loss": 0.60844225,
      "learning_rate": 2.0134363269368095e-06,
      "loss": 0.62859249,
      "num_input_tokens_seen": 183455280,
      "step": 8531,
      "time_per_iteration": 4.57397985458374
    },
    {
      "auxiliary_loss_clip": 0.01086054,
      "auxiliary_loss_mlp": 0.01034251,
      "balance_loss_clip": 1.03860235,
      "balance_loss_mlp": 1.0192306,
      "epoch": 0.5129715917631144,
      "flos": 20449152316800.0,
      "grad_norm": 1.8214363311986992,
      "language_loss": 0.76958299,
      "learning_rate": 2.0130468737920725e-06,
      "loss": 0.79078609,
      "num_input_tokens_seen": 183473955,
      "step": 8532,
      "time_per_iteration": 2.5325894355773926
    },
    {
      "auxiliary_loss_clip": 0.01092134,
      "auxiliary_loss_mlp": 0.01032867,
      "balance_loss_clip": 1.03907561,
      "balance_loss_mlp": 1.01956367,
      "epoch": 0.5130317150157824,
      "flos": 35116146840960.0,
      "grad_norm": 3.3401705694862573,
      "language_loss": 0.66560805,
      "learning_rate": 2.012657420152597e-06,
      "loss": 0.68685806,
      "num_input_tokens_seen": 183497195,
      "step": 8533,
      "time_per_iteration": 2.6420252323150635
    },
    {
      "auxiliary_loss_clip": 0.01084196,
      "auxiliary_loss_mlp": 0.01035432,
      "balance_loss_clip": 1.03988624,
      "balance_loss_mlp": 1.02159142,
      "epoch": 0.5130918382684503,
      "flos": 19791627903360.0,
      "grad_norm": 1.8812003050440453,
      "language_loss": 0.82081807,
      "learning_rate": 2.01226796603315e-06,
      "loss": 0.84201431,
      "num_input_tokens_seen": 183513675,
      "step": 8534,
      "time_per_iteration": 3.8917222023010254
    },
    {
      "auxiliary_loss_clip": 0.01102644,
      "auxiliary_loss_mlp": 0.01037513,
      "balance_loss_clip": 1.03678107,
      "balance_loss_mlp": 1.02311277,
      "epoch": 0.5131519615211183,
      "flos": 26323096337280.0,
      "grad_norm": 1.445567655969626,
      "language_loss": 0.63663131,
      "learning_rate": 2.0118785114485017e-06,
      "loss": 0.65803289,
      "num_input_tokens_seen": 183535165,
      "step": 8535,
      "time_per_iteration": 2.5275259017944336
    },
    {
      "auxiliary_loss_clip": 0.0110612,
      "auxiliary_loss_mlp": 0.01029183,
      "balance_loss_clip": 1.04111242,
      "balance_loss_mlp": 1.01610565,
      "epoch": 0.5132120847737862,
      "flos": 19171917532800.0,
      "grad_norm": 1.5491483535888144,
      "language_loss": 0.6954205,
      "learning_rate": 2.011489056413418e-06,
      "loss": 0.71677351,
      "num_input_tokens_seen": 183553780,
      "step": 8536,
      "time_per_iteration": 2.476519823074341
    },
    {
      "auxiliary_loss_clip": 0.01104998,
      "auxiliary_loss_mlp": 0.01031496,
      "balance_loss_clip": 1.03939915,
      "balance_loss_mlp": 1.01726198,
      "epoch": 0.5132722080264542,
      "flos": 20230420446720.0,
      "grad_norm": 2.028299354143275,
      "language_loss": 0.71351743,
      "learning_rate": 2.011099600942669e-06,
      "loss": 0.73488235,
      "num_input_tokens_seen": 183572285,
      "step": 8537,
      "time_per_iteration": 3.9539103507995605
    },
    {
      "auxiliary_loss_clip": 0.01070915,
      "auxiliary_loss_mlp": 0.01033102,
      "balance_loss_clip": 1.0367825,
      "balance_loss_mlp": 1.01873136,
      "epoch": 0.5133323312791223,
      "flos": 16469459930880.0,
      "grad_norm": 2.2462021568051833,
      "language_loss": 0.81038719,
      "learning_rate": 2.0107101450510214e-06,
      "loss": 0.8314274,
      "num_input_tokens_seen": 183589330,
      "step": 8538,
      "time_per_iteration": 2.5341107845306396
    },
    {
      "auxiliary_loss_clip": 0.01102783,
      "auxiliary_loss_mlp": 0.01027022,
      "balance_loss_clip": 1.03755832,
      "balance_loss_mlp": 1.0139035,
      "epoch": 0.5133924545317902,
      "flos": 26068094709120.0,
      "grad_norm": 1.7264771425756702,
      "language_loss": 0.78475606,
      "learning_rate": 2.0103206887532437e-06,
      "loss": 0.80605412,
      "num_input_tokens_seen": 183609205,
      "step": 8539,
      "time_per_iteration": 2.515965223312378
    },
    {
      "auxiliary_loss_clip": 0.0109309,
      "auxiliary_loss_mlp": 0.01037779,
      "balance_loss_clip": 1.03835499,
      "balance_loss_mlp": 1.0240221,
      "epoch": 0.5134525777844582,
      "flos": 29131023248640.0,
      "grad_norm": 1.9639788220740237,
      "language_loss": 0.76075006,
      "learning_rate": 2.009931232064105e-06,
      "loss": 0.78205872,
      "num_input_tokens_seen": 183629985,
      "step": 8540,
      "time_per_iteration": 2.570103168487549
    },
    {
      "auxiliary_loss_clip": 0.0106693,
      "auxiliary_loss_mlp": 0.01036062,
      "balance_loss_clip": 1.04174554,
      "balance_loss_mlp": 1.0217092,
      "epoch": 0.5135127010371261,
      "flos": 17454776883840.0,
      "grad_norm": 2.421915761297425,
      "language_loss": 0.74761152,
      "learning_rate": 2.0095417749983724e-06,
      "loss": 0.76864147,
      "num_input_tokens_seen": 183648220,
      "step": 8541,
      "time_per_iteration": 3.973365545272827
    },
    {
      "auxiliary_loss_clip": 0.01061088,
      "auxiliary_loss_mlp": 0.01037689,
      "balance_loss_clip": 1.04061317,
      "balance_loss_mlp": 1.02377772,
      "epoch": 0.5135728242897941,
      "flos": 21944975316480.0,
      "grad_norm": 1.6272618754516988,
      "language_loss": 0.70229125,
      "learning_rate": 2.0091523175708162e-06,
      "loss": 0.723279,
      "num_input_tokens_seen": 183668230,
      "step": 8542,
      "time_per_iteration": 2.614380121231079
    },
    {
      "auxiliary_loss_clip": 0.01095181,
      "auxiliary_loss_mlp": 0.0103425,
      "balance_loss_clip": 1.03928328,
      "balance_loss_mlp": 1.02066624,
      "epoch": 0.513632947542462,
      "flos": 22674859678080.0,
      "grad_norm": 1.9011353756502676,
      "language_loss": 0.79268742,
      "learning_rate": 2.0087628597962023e-06,
      "loss": 0.81398177,
      "num_input_tokens_seen": 183687800,
      "step": 8543,
      "time_per_iteration": 2.5174856185913086
    },
    {
      "auxiliary_loss_clip": 0.01095664,
      "auxiliary_loss_mlp": 0.01037641,
      "balance_loss_clip": 1.04577708,
      "balance_loss_mlp": 1.02335405,
      "epoch": 0.51369307079513,
      "flos": 29457163762560.0,
      "grad_norm": 4.541670995116027,
      "language_loss": 0.67852771,
      "learning_rate": 2.008373401689299e-06,
      "loss": 0.69986069,
      "num_input_tokens_seen": 183709025,
      "step": 8544,
      "time_per_iteration": 2.5607926845550537
    },
    {
      "auxiliary_loss_clip": 0.01083944,
      "auxiliary_loss_mlp": 0.01054567,
      "balance_loss_clip": 1.04018867,
      "balance_loss_mlp": 1.03918314,
      "epoch": 0.513753194047798,
      "flos": 18989347680000.0,
      "grad_norm": 1.922452754399939,
      "language_loss": 0.72351301,
      "learning_rate": 2.0079839432648765e-06,
      "loss": 0.74489808,
      "num_input_tokens_seen": 183725740,
      "step": 8545,
      "time_per_iteration": 2.540252923965454
    },
    {
      "auxiliary_loss_clip": 0.01108523,
      "auxiliary_loss_mlp": 0.01038819,
      "balance_loss_clip": 1.04002571,
      "balance_loss_mlp": 1.02409053,
      "epoch": 0.513813317300466,
      "flos": 17821855923840.0,
      "grad_norm": 1.9614383164009759,
      "language_loss": 0.81687069,
      "learning_rate": 2.0075944845377016e-06,
      "loss": 0.8383441,
      "num_input_tokens_seen": 183743995,
      "step": 8546,
      "time_per_iteration": 2.440058946609497
    },
    {
      "auxiliary_loss_clip": 0.0110423,
      "auxiliary_loss_mlp": 0.01036442,
      "balance_loss_clip": 1.03879285,
      "balance_loss_mlp": 1.02233362,
      "epoch": 0.5138734405531339,
      "flos": 24061191045120.0,
      "grad_norm": 1.7579648262569623,
      "language_loss": 0.73152816,
      "learning_rate": 2.007205025522544e-06,
      "loss": 0.75293493,
      "num_input_tokens_seen": 183764150,
      "step": 8547,
      "time_per_iteration": 2.515700340270996
    },
    {
      "auxiliary_loss_clip": 0.01106029,
      "auxiliary_loss_mlp": 0.01045145,
      "balance_loss_clip": 1.03886485,
      "balance_loss_mlp": 1.03129315,
      "epoch": 0.5139335638058019,
      "flos": 26097253574400.0,
      "grad_norm": 1.651548583529879,
      "language_loss": 0.73236871,
      "learning_rate": 2.0068155662341702e-06,
      "loss": 0.75388044,
      "num_input_tokens_seen": 183783280,
      "step": 8548,
      "time_per_iteration": 2.5228259563446045
    },
    {
      "auxiliary_loss_clip": 0.01089247,
      "auxiliary_loss_mlp": 0.01035427,
      "balance_loss_clip": 1.0380789,
      "balance_loss_mlp": 1.02102065,
      "epoch": 0.5139936870584698,
      "flos": 18917095472640.0,
      "grad_norm": 1.5465203950515087,
      "language_loss": 0.81758618,
      "learning_rate": 2.0064261066873495e-06,
      "loss": 0.83883291,
      "num_input_tokens_seen": 183800725,
      "step": 8549,
      "time_per_iteration": 2.5294625759124756
    },
    {
      "auxiliary_loss_clip": 0.01105525,
      "auxiliary_loss_mlp": 0.01032493,
      "balance_loss_clip": 1.04010272,
      "balance_loss_mlp": 1.02032161,
      "epoch": 0.5140538103111378,
      "flos": 16144001775360.0,
      "grad_norm": 1.9996120092906235,
      "language_loss": 0.72400796,
      "learning_rate": 2.0060366468968504e-06,
      "loss": 0.74538815,
      "num_input_tokens_seen": 183818735,
      "step": 8550,
      "time_per_iteration": 2.464104175567627
    },
    {
      "auxiliary_loss_clip": 0.01105447,
      "auxiliary_loss_mlp": 0.01039684,
      "balance_loss_clip": 1.04095864,
      "balance_loss_mlp": 1.02571893,
      "epoch": 0.5141139335638057,
      "flos": 22420145358720.0,
      "grad_norm": 1.8563793690085533,
      "language_loss": 0.75494033,
      "learning_rate": 2.0056471868774408e-06,
      "loss": 0.77639163,
      "num_input_tokens_seen": 183840015,
      "step": 8551,
      "time_per_iteration": 2.5059967041015625
    },
    {
      "auxiliary_loss_clip": 0.01093192,
      "auxiliary_loss_mlp": 0.01031931,
      "balance_loss_clip": 1.04416585,
      "balance_loss_mlp": 1.01847219,
      "epoch": 0.5141740568164738,
      "flos": 27089645506560.0,
      "grad_norm": 1.7560681170953631,
      "language_loss": 0.69494355,
      "learning_rate": 2.0052577266438897e-06,
      "loss": 0.71619475,
      "num_input_tokens_seen": 183860145,
      "step": 8552,
      "time_per_iteration": 2.5637855529785156
    },
    {
      "auxiliary_loss_clip": 0.0110637,
      "auxiliary_loss_mlp": 0.01036493,
      "balance_loss_clip": 1.03978395,
      "balance_loss_mlp": 1.02205098,
      "epoch": 0.5142341800691418,
      "flos": 24973250209920.0,
      "grad_norm": 1.8659987082348768,
      "language_loss": 0.74622309,
      "learning_rate": 2.004868266210965e-06,
      "loss": 0.76765168,
      "num_input_tokens_seen": 183880540,
      "step": 8553,
      "time_per_iteration": 2.5392589569091797
    },
    {
      "auxiliary_loss_clip": 0.01116562,
      "auxiliary_loss_mlp": 0.01038236,
      "balance_loss_clip": 1.04065275,
      "balance_loss_mlp": 1.02494979,
      "epoch": 0.5142943033218097,
      "flos": 20704513080960.0,
      "grad_norm": 1.8889408439106794,
      "language_loss": 0.67895406,
      "learning_rate": 2.004478805593435e-06,
      "loss": 0.70050204,
      "num_input_tokens_seen": 183900895,
      "step": 8554,
      "time_per_iteration": 2.456629514694214
    },
    {
      "auxiliary_loss_clip": 0.01108686,
      "auxiliary_loss_mlp": 0.01042488,
      "balance_loss_clip": 1.03826857,
      "balance_loss_mlp": 1.02641904,
      "epoch": 0.5143544265744777,
      "flos": 22925479847040.0,
      "grad_norm": 2.2582367517775075,
      "language_loss": 0.7293582,
      "learning_rate": 2.004089344806068e-06,
      "loss": 0.75086993,
      "num_input_tokens_seen": 183920335,
      "step": 8555,
      "time_per_iteration": 2.4943737983703613
    },
    {
      "auxiliary_loss_clip": 0.01082023,
      "auxiliary_loss_mlp": 0.01039064,
      "balance_loss_clip": 1.04231262,
      "balance_loss_mlp": 1.02557528,
      "epoch": 0.5144145498271456,
      "flos": 15921391236480.0,
      "grad_norm": 2.436324916022814,
      "language_loss": 0.74615932,
      "learning_rate": 2.003699883863633e-06,
      "loss": 0.76737022,
      "num_input_tokens_seen": 183936220,
      "step": 8556,
      "time_per_iteration": 2.511085271835327
    },
    {
      "auxiliary_loss_clip": 0.01083344,
      "auxiliary_loss_mlp": 0.01032539,
      "balance_loss_clip": 1.03810716,
      "balance_loss_mlp": 1.01964021,
      "epoch": 0.5144746730798136,
      "flos": 19681238430720.0,
      "grad_norm": 1.9181443825740911,
      "language_loss": 0.86331737,
      "learning_rate": 2.003310422780898e-06,
      "loss": 0.88447618,
      "num_input_tokens_seen": 183953250,
      "step": 8557,
      "time_per_iteration": 2.537074565887451
    },
    {
      "auxiliary_loss_clip": 0.01097555,
      "auxiliary_loss_mlp": 0.01042703,
      "balance_loss_clip": 1.03719175,
      "balance_loss_mlp": 1.02882123,
      "epoch": 0.5145347963324816,
      "flos": 23914711382400.0,
      "grad_norm": 1.6134952408352903,
      "language_loss": 0.89045501,
      "learning_rate": 2.0029209615726307e-06,
      "loss": 0.9118576,
      "num_input_tokens_seen": 183973865,
      "step": 8558,
      "time_per_iteration": 2.4992167949676514
    },
    {
      "auxiliary_loss_clip": 0.01112251,
      "auxiliary_loss_mlp": 0.00787504,
      "balance_loss_clip": 1.03897524,
      "balance_loss_mlp": 1.01240468,
      "epoch": 0.5145949195851496,
      "flos": 18260002022400.0,
      "grad_norm": 1.871781169462433,
      "language_loss": 0.64908433,
      "learning_rate": 2.002531500253602e-06,
      "loss": 0.66808188,
      "num_input_tokens_seen": 183992555,
      "step": 8559,
      "time_per_iteration": 2.4540417194366455
    },
    {
      "auxiliary_loss_clip": 0.01100767,
      "auxiliary_loss_mlp": 0.00787521,
      "balance_loss_clip": 1.04168081,
      "balance_loss_mlp": 1.01117313,
      "epoch": 0.5146550428378175,
      "flos": 26213425136640.0,
      "grad_norm": 1.5867776149050856,
      "language_loss": 0.63171953,
      "learning_rate": 2.002142038838577e-06,
      "loss": 0.6506024,
      "num_input_tokens_seen": 184010825,
      "step": 8560,
      "time_per_iteration": 2.5214316844940186
    },
    {
      "auxiliary_loss_clip": 0.01115182,
      "auxiliary_loss_mlp": 0.01032925,
      "balance_loss_clip": 1.04016554,
      "balance_loss_mlp": 1.01971066,
      "epoch": 0.5147151660904855,
      "flos": 22674177319680.0,
      "grad_norm": 1.6517460657280272,
      "language_loss": 0.70090306,
      "learning_rate": 2.0017525773423265e-06,
      "loss": 0.7223841,
      "num_input_tokens_seen": 184030155,
      "step": 8561,
      "time_per_iteration": 2.466305732727051
    },
    {
      "auxiliary_loss_clip": 0.01092583,
      "auxiliary_loss_mlp": 0.01032801,
      "balance_loss_clip": 1.03860736,
      "balance_loss_mlp": 1.02029586,
      "epoch": 0.5147752893431534,
      "flos": 24972388283520.0,
      "grad_norm": 1.484331517466963,
      "language_loss": 0.66421539,
      "learning_rate": 2.0013631157796177e-06,
      "loss": 0.68546927,
      "num_input_tokens_seen": 184051440,
      "step": 8562,
      "time_per_iteration": 2.5400145053863525
    },
    {
      "auxiliary_loss_clip": 0.01110105,
      "auxiliary_loss_mlp": 0.01033929,
      "balance_loss_clip": 1.04195166,
      "balance_loss_mlp": 1.02031517,
      "epoch": 0.5148354125958214,
      "flos": 22744669760640.0,
      "grad_norm": 1.650623770853723,
      "language_loss": 0.77521873,
      "learning_rate": 2.0009736541652188e-06,
      "loss": 0.79665905,
      "num_input_tokens_seen": 184070205,
      "step": 8563,
      "time_per_iteration": 2.489290237426758
    },
    {
      "auxiliary_loss_clip": 0.01109019,
      "auxiliary_loss_mlp": 0.01036794,
      "balance_loss_clip": 1.03962755,
      "balance_loss_mlp": 1.02154708,
      "epoch": 0.5148955358484893,
      "flos": 23068763199360.0,
      "grad_norm": 1.9590385129349255,
      "language_loss": 0.82628089,
      "learning_rate": 2.0005841925139e-06,
      "loss": 0.84773904,
      "num_input_tokens_seen": 184087345,
      "step": 8564,
      "time_per_iteration": 2.4640767574310303
    },
    {
      "auxiliary_loss_clip": 0.01100596,
      "auxiliary_loss_mlp": 0.01034097,
      "balance_loss_clip": 1.04047012,
      "balance_loss_mlp": 1.01963711,
      "epoch": 0.5149556591011574,
      "flos": 20340127560960.0,
      "grad_norm": 1.6955493044883954,
      "language_loss": 0.729792,
      "learning_rate": 2.0001947308404283e-06,
      "loss": 0.75113893,
      "num_input_tokens_seen": 184107110,
      "step": 8565,
      "time_per_iteration": 2.5263516902923584
    },
    {
      "auxiliary_loss_clip": 0.01110177,
      "auxiliary_loss_mlp": 0.01033668,
      "balance_loss_clip": 1.04088855,
      "balance_loss_mlp": 1.01787281,
      "epoch": 0.5150157823538254,
      "flos": 22638230784000.0,
      "grad_norm": 2.5479449379220553,
      "language_loss": 0.6869483,
      "learning_rate": 1.9998052691595715e-06,
      "loss": 0.70838678,
      "num_input_tokens_seen": 184127105,
      "step": 8566,
      "time_per_iteration": 2.493751049041748
    },
    {
      "auxiliary_loss_clip": 0.0111705,
      "auxiliary_loss_mlp": 0.00789716,
      "balance_loss_clip": 1.03705025,
      "balance_loss_mlp": 1.01374412,
      "epoch": 0.5150759056064933,
      "flos": 26067627832320.0,
      "grad_norm": 1.9293801864144686,
      "language_loss": 0.78126305,
      "learning_rate": 1.9994158074861005e-06,
      "loss": 0.8003307,
      "num_input_tokens_seen": 184148060,
      "step": 8567,
      "time_per_iteration": 2.50801944732666
    },
    {
      "auxiliary_loss_clip": 0.01109856,
      "auxiliary_loss_mlp": 0.01033577,
      "balance_loss_clip": 1.04128003,
      "balance_loss_mlp": 1.01873517,
      "epoch": 0.5151360288591613,
      "flos": 25952641418880.0,
      "grad_norm": 2.005405802350575,
      "language_loss": 0.79084373,
      "learning_rate": 1.9990263458347806e-06,
      "loss": 0.81227803,
      "num_input_tokens_seen": 184166175,
      "step": 8568,
      "time_per_iteration": 2.5143179893493652
    },
    {
      "auxiliary_loss_clip": 0.01092709,
      "auxiliary_loss_mlp": 0.01030602,
      "balance_loss_clip": 1.03774607,
      "balance_loss_mlp": 1.01744187,
      "epoch": 0.5151961521118292,
      "flos": 18507246312960.0,
      "grad_norm": 2.2223144145395315,
      "language_loss": 0.90768683,
      "learning_rate": 1.9986368842203825e-06,
      "loss": 0.92891997,
      "num_input_tokens_seen": 184182600,
      "step": 8569,
      "time_per_iteration": 2.5098636150360107
    },
    {
      "auxiliary_loss_clip": 0.01119809,
      "auxiliary_loss_mlp": 0.01035851,
      "balance_loss_clip": 1.0410316,
      "balance_loss_mlp": 1.02182007,
      "epoch": 0.5152562753644973,
      "flos": 22233696837120.0,
      "grad_norm": 2.4055869379734034,
      "language_loss": 0.76730418,
      "learning_rate": 1.998247422657674e-06,
      "loss": 0.78886074,
      "num_input_tokens_seen": 184202020,
      "step": 8570,
      "time_per_iteration": 4.401336431503296
    },
    {
      "auxiliary_loss_clip": 0.0110532,
      "auxiliary_loss_mlp": 0.0103797,
      "balance_loss_clip": 1.03846729,
      "balance_loss_mlp": 1.02296197,
      "epoch": 0.5153163986171652,
      "flos": 38436555047040.0,
      "grad_norm": 1.9808930605527721,
      "language_loss": 0.73692054,
      "learning_rate": 1.9978579611614227e-06,
      "loss": 0.75835341,
      "num_input_tokens_seen": 184224850,
      "step": 8571,
      "time_per_iteration": 2.6345598697662354
    },
    {
      "auxiliary_loss_clip": 0.0102522,
      "auxiliary_loss_mlp": 0.01003603,
      "balance_loss_clip": 1.01870131,
      "balance_loss_mlp": 1.00211239,
      "epoch": 0.5153765218698332,
      "flos": 66384503015040.0,
      "grad_norm": 0.7770042520564989,
      "language_loss": 0.52896047,
      "learning_rate": 1.9974684997463984e-06,
      "loss": 0.5492487,
      "num_input_tokens_seen": 184288520,
      "step": 8572,
      "time_per_iteration": 3.238548994064331
    },
    {
      "auxiliary_loss_clip": 0.01105502,
      "auxiliary_loss_mlp": 0.01036747,
      "balance_loss_clip": 1.04176402,
      "balance_loss_mlp": 1.02392554,
      "epoch": 0.5154366451225011,
      "flos": 24024669891840.0,
      "grad_norm": 1.7892363458854494,
      "language_loss": 0.75761038,
      "learning_rate": 1.9970790384273687e-06,
      "loss": 0.77903289,
      "num_input_tokens_seen": 184308565,
      "step": 8573,
      "time_per_iteration": 3.9702603816986084
    },
    {
      "auxiliary_loss_clip": 0.01104523,
      "auxiliary_loss_mlp": 0.01029112,
      "balance_loss_clip": 1.03978264,
      "balance_loss_mlp": 1.01516497,
      "epoch": 0.5154967683751691,
      "flos": 23468843859840.0,
      "grad_norm": 1.8610930435920277,
      "language_loss": 0.77387792,
      "learning_rate": 1.996689577219102e-06,
      "loss": 0.7952143,
      "num_input_tokens_seen": 184326795,
      "step": 8574,
      "time_per_iteration": 2.48578143119812
    },
    {
      "auxiliary_loss_clip": 0.01100085,
      "auxiliary_loss_mlp": 0.0103103,
      "balance_loss_clip": 1.04223371,
      "balance_loss_mlp": 1.01820338,
      "epoch": 0.515556891627837,
      "flos": 23805650712960.0,
      "grad_norm": 1.7513908657195885,
      "language_loss": 0.85479772,
      "learning_rate": 1.996300116136367e-06,
      "loss": 0.87610888,
      "num_input_tokens_seen": 184345990,
      "step": 8575,
      "time_per_iteration": 3.9542157649993896
    },
    {
      "auxiliary_loss_clip": 0.01106625,
      "auxiliary_loss_mlp": 0.01036166,
      "balance_loss_clip": 1.03836823,
      "balance_loss_mlp": 1.02271342,
      "epoch": 0.515617014880505,
      "flos": 19828544106240.0,
      "grad_norm": 1.5475710772224927,
      "language_loss": 0.76632905,
      "learning_rate": 1.995910655193932e-06,
      "loss": 0.78775692,
      "num_input_tokens_seen": 184366300,
      "step": 8576,
      "time_per_iteration": 2.5008721351623535
    },
    {
      "auxiliary_loss_clip": 0.01074987,
      "auxiliary_loss_mlp": 0.00790632,
      "balance_loss_clip": 1.04318547,
      "balance_loss_mlp": 1.01209164,
      "epoch": 0.515677138133173,
      "flos": 14245907385600.0,
      "grad_norm": 2.666310295926506,
      "language_loss": 0.75199008,
      "learning_rate": 1.9955211944065654e-06,
      "loss": 0.77064621,
      "num_input_tokens_seen": 184383030,
      "step": 8577,
      "time_per_iteration": 2.5712759494781494
    },
    {
      "auxiliary_loss_clip": 0.01094741,
      "auxiliary_loss_mlp": 0.01047416,
      "balance_loss_clip": 1.0407337,
      "balance_loss_mlp": 1.03115559,
      "epoch": 0.515737261385841,
      "flos": 28289707920000.0,
      "grad_norm": 1.8315170792185766,
      "language_loss": 0.81215495,
      "learning_rate": 1.9951317337890353e-06,
      "loss": 0.8335765,
      "num_input_tokens_seen": 184403410,
      "step": 8578,
      "time_per_iteration": 2.553809642791748
    },
    {
      "auxiliary_loss_clip": 0.01114676,
      "auxiliary_loss_mlp": 0.01030168,
      "balance_loss_clip": 1.04019427,
      "balance_loss_mlp": 1.01690042,
      "epoch": 0.515797384638509,
      "flos": 27891925729920.0,
      "grad_norm": 1.6977384414414518,
      "language_loss": 0.76024151,
      "learning_rate": 1.9947422733561105e-06,
      "loss": 0.78169,
      "num_input_tokens_seen": 184423830,
      "step": 8579,
      "time_per_iteration": 2.499141216278076
    },
    {
      "auxiliary_loss_clip": 0.01082437,
      "auxiliary_loss_mlp": 0.01030039,
      "balance_loss_clip": 1.04215777,
      "balance_loss_mlp": 1.01678884,
      "epoch": 0.5158575078911769,
      "flos": 23040071210880.0,
      "grad_norm": 2.059121257818275,
      "language_loss": 0.79071826,
      "learning_rate": 1.994352813122559e-06,
      "loss": 0.81184304,
      "num_input_tokens_seen": 184445050,
      "step": 8580,
      "time_per_iteration": 3.9380955696105957
    },
    {
      "auxiliary_loss_clip": 0.01085147,
      "auxiliary_loss_mlp": 0.0104807,
      "balance_loss_clip": 1.04545593,
      "balance_loss_mlp": 1.03205454,
      "epoch": 0.5159176311438449,
      "flos": 12641346938880.0,
      "grad_norm": 1.9886861527977173,
      "language_loss": 0.7280103,
      "learning_rate": 1.99396335310315e-06,
      "loss": 0.74934244,
      "num_input_tokens_seen": 184460775,
      "step": 8581,
      "time_per_iteration": 2.4991724491119385
    },
    {
      "auxiliary_loss_clip": 0.01105123,
      "auxiliary_loss_mlp": 0.01032631,
      "balance_loss_clip": 1.04088664,
      "balance_loss_mlp": 1.02003109,
      "epoch": 0.5159777543965128,
      "flos": 15558154951680.0,
      "grad_norm": 2.1893564940004455,
      "language_loss": 0.73795068,
      "learning_rate": 1.9935738933126508e-06,
      "loss": 0.75932819,
      "num_input_tokens_seen": 184477365,
      "step": 8582,
      "time_per_iteration": 2.454841375350952
    },
    {
      "auxiliary_loss_clip": 0.01079283,
      "auxiliary_loss_mlp": 0.01036087,
      "balance_loss_clip": 1.04117155,
      "balance_loss_mlp": 1.02326047,
      "epoch": 0.5160378776491809,
      "flos": 23221671396480.0,
      "grad_norm": 1.6823874019931488,
      "language_loss": 0.65850091,
      "learning_rate": 1.99318443376583e-06,
      "loss": 0.67965466,
      "num_input_tokens_seen": 184497045,
      "step": 8583,
      "time_per_iteration": 2.548562526702881
    },
    {
      "auxiliary_loss_clip": 0.01108923,
      "auxiliary_loss_mlp": 0.01037745,
      "balance_loss_clip": 1.04267263,
      "balance_loss_mlp": 1.02378535,
      "epoch": 0.5160980009018488,
      "flos": 21944616180480.0,
      "grad_norm": 1.4675508228576437,
      "language_loss": 0.75739199,
      "learning_rate": 1.9927949744774568e-06,
      "loss": 0.77885866,
      "num_input_tokens_seen": 184517675,
      "step": 8584,
      "time_per_iteration": 2.4956626892089844
    },
    {
      "auxiliary_loss_clip": 0.01091084,
      "auxiliary_loss_mlp": 0.01045651,
      "balance_loss_clip": 1.04490578,
      "balance_loss_mlp": 1.03181648,
      "epoch": 0.5161581241545168,
      "flos": 22784064001920.0,
      "grad_norm": 3.1448325532382153,
      "language_loss": 0.78881103,
      "learning_rate": 1.9924055154622983e-06,
      "loss": 0.8101784,
      "num_input_tokens_seen": 184537745,
      "step": 8585,
      "time_per_iteration": 2.548659086227417
    },
    {
      "auxiliary_loss_clip": 0.01106024,
      "auxiliary_loss_mlp": 0.01037828,
      "balance_loss_clip": 1.04117298,
      "balance_loss_mlp": 1.02503729,
      "epoch": 0.5162182474071847,
      "flos": 19675384513920.0,
      "grad_norm": 2.2624279082238505,
      "language_loss": 0.81474686,
      "learning_rate": 1.9920160567351238e-06,
      "loss": 0.83618534,
      "num_input_tokens_seen": 184553630,
      "step": 8586,
      "time_per_iteration": 2.450187921524048
    },
    {
      "auxiliary_loss_clip": 0.01094833,
      "auxiliary_loss_mlp": 0.01032462,
      "balance_loss_clip": 1.04062903,
      "balance_loss_mlp": 1.01984978,
      "epoch": 0.5162783706598527,
      "flos": 20046198568320.0,
      "grad_norm": 1.8519621160237456,
      "language_loss": 0.71505368,
      "learning_rate": 1.991626598310701e-06,
      "loss": 0.73632663,
      "num_input_tokens_seen": 184573530,
      "step": 8587,
      "time_per_iteration": 2.5211684703826904
    },
    {
      "auxiliary_loss_clip": 0.01033424,
      "auxiliary_loss_mlp": 0.01005307,
      "balance_loss_clip": 1.01680112,
      "balance_loss_mlp": 1.00400209,
      "epoch": 0.5163384939125206,
      "flos": 69959553713280.0,
      "grad_norm": 0.7337102037370647,
      "language_loss": 0.57840645,
      "learning_rate": 1.9912371402037984e-06,
      "loss": 0.5987938,
      "num_input_tokens_seen": 184637875,
      "step": 8588,
      "time_per_iteration": 3.1115260124206543
    },
    {
      "auxiliary_loss_clip": 0.01101441,
      "auxiliary_loss_mlp": 0.01037255,
      "balance_loss_clip": 1.04240263,
      "balance_loss_mlp": 1.02286673,
      "epoch": 0.5163986171651886,
      "flos": 17417034668160.0,
      "grad_norm": 1.8306779288341224,
      "language_loss": 0.75185496,
      "learning_rate": 1.990847682429185e-06,
      "loss": 0.77324188,
      "num_input_tokens_seen": 184656125,
      "step": 8589,
      "time_per_iteration": 2.508363723754883
    },
    {
      "auxiliary_loss_clip": 0.01109791,
      "auxiliary_loss_mlp": 0.01033641,
      "balance_loss_clip": 1.04191113,
      "balance_loss_mlp": 1.02086151,
      "epoch": 0.5164587404178566,
      "flos": 21322679166720.0,
      "grad_norm": 1.488653688351353,
      "language_loss": 0.6752255,
      "learning_rate": 1.990458225001627e-06,
      "loss": 0.6966598,
      "num_input_tokens_seen": 184675920,
      "step": 8590,
      "time_per_iteration": 2.5060672760009766
    },
    {
      "auxiliary_loss_clip": 0.01031619,
      "auxiliary_loss_mlp": 0.01002712,
      "balance_loss_clip": 1.01607966,
      "balance_loss_mlp": 1.00141859,
      "epoch": 0.5165188636705246,
      "flos": 68057149691520.0,
      "grad_norm": 0.8039291787768882,
      "language_loss": 0.55860424,
      "learning_rate": 1.990068767935895e-06,
      "loss": 0.57894754,
      "num_input_tokens_seen": 184730520,
      "step": 8591,
      "time_per_iteration": 3.027583599090576
    },
    {
      "auxiliary_loss_clip": 0.01089897,
      "auxiliary_loss_mlp": 0.01026033,
      "balance_loss_clip": 1.03914237,
      "balance_loss_mlp": 1.01383185,
      "epoch": 0.5165789869231926,
      "flos": 19385657412480.0,
      "grad_norm": 1.5978202799351757,
      "language_loss": 0.8138454,
      "learning_rate": 1.9896793112467566e-06,
      "loss": 0.83500463,
      "num_input_tokens_seen": 184748340,
      "step": 8592,
      "time_per_iteration": 2.505753755569458
    },
    {
      "auxiliary_loss_clip": 0.01101542,
      "auxiliary_loss_mlp": 0.01029269,
      "balance_loss_clip": 1.04052997,
      "balance_loss_mlp": 1.01665676,
      "epoch": 0.5166391101758605,
      "flos": 20960197067520.0,
      "grad_norm": 1.7479290385938921,
      "language_loss": 0.83339155,
      "learning_rate": 1.989289854948979e-06,
      "loss": 0.85469967,
      "num_input_tokens_seen": 184766615,
      "step": 8593,
      "time_per_iteration": 2.481851577758789
    },
    {
      "auxiliary_loss_clip": 0.01092255,
      "auxiliary_loss_mlp": 0.01036841,
      "balance_loss_clip": 1.04125547,
      "balance_loss_mlp": 1.02272058,
      "epoch": 0.5166992334285285,
      "flos": 29462407148160.0,
      "grad_norm": 1.6274928695054172,
      "language_loss": 0.69300187,
      "learning_rate": 1.9889003990573314e-06,
      "loss": 0.71429282,
      "num_input_tokens_seen": 184788075,
      "step": 8594,
      "time_per_iteration": 2.589886426925659
    },
    {
      "auxiliary_loss_clip": 0.01077325,
      "auxiliary_loss_mlp": 0.01034025,
      "balance_loss_clip": 1.04122496,
      "balance_loss_mlp": 1.02050102,
      "epoch": 0.5167593566811964,
      "flos": 20304360593280.0,
      "grad_norm": 1.5075187437363977,
      "language_loss": 0.77566659,
      "learning_rate": 1.988510943586582e-06,
      "loss": 0.79678011,
      "num_input_tokens_seen": 184808710,
      "step": 8595,
      "time_per_iteration": 2.5821850299835205
    },
    {
      "auxiliary_loss_clip": 0.01117039,
      "auxiliary_loss_mlp": 0.01032609,
      "balance_loss_clip": 1.04245305,
      "balance_loss_mlp": 1.01944256,
      "epoch": 0.5168194799338645,
      "flos": 14611370313600.0,
      "grad_norm": 1.4807585562488366,
      "language_loss": 0.65330303,
      "learning_rate": 1.9881214885514986e-06,
      "loss": 0.67479944,
      "num_input_tokens_seen": 184826475,
      "step": 8596,
      "time_per_iteration": 2.4479734897613525
    },
    {
      "auxiliary_loss_clip": 0.01086051,
      "auxiliary_loss_mlp": 0.0103148,
      "balance_loss_clip": 1.04546177,
      "balance_loss_mlp": 1.01673985,
      "epoch": 0.5168796031865324,
      "flos": 25007257411200.0,
      "grad_norm": 4.6756378246922425,
      "language_loss": 0.75697803,
      "learning_rate": 1.9877320339668492e-06,
      "loss": 0.77815336,
      "num_input_tokens_seen": 184845245,
      "step": 8597,
      "time_per_iteration": 2.575133800506592
    },
    {
      "auxiliary_loss_clip": 0.01115765,
      "auxiliary_loss_mlp": 0.01023144,
      "balance_loss_clip": 1.04075348,
      "balance_loss_mlp": 1.0104121,
      "epoch": 0.5169397264392004,
      "flos": 26939969533440.0,
      "grad_norm": 1.5264818311645665,
      "language_loss": 0.81380886,
      "learning_rate": 1.987342579847403e-06,
      "loss": 0.83519793,
      "num_input_tokens_seen": 184866605,
      "step": 8598,
      "time_per_iteration": 2.495180130004883
    },
    {
      "auxiliary_loss_clip": 0.01069084,
      "auxiliary_loss_mlp": 0.01036324,
      "balance_loss_clip": 1.04006314,
      "balance_loss_mlp": 1.02259731,
      "epoch": 0.5169998496918683,
      "flos": 25407804948480.0,
      "grad_norm": 1.6448997796989744,
      "language_loss": 0.7534405,
      "learning_rate": 1.9869531262079273e-06,
      "loss": 0.77449453,
      "num_input_tokens_seen": 184886945,
      "step": 8599,
      "time_per_iteration": 2.609812021255493
    },
    {
      "auxiliary_loss_clip": 0.01097918,
      "auxiliary_loss_mlp": 0.01033474,
      "balance_loss_clip": 1.04172754,
      "balance_loss_mlp": 1.02040255,
      "epoch": 0.5170599729445363,
      "flos": 24680793674880.0,
      "grad_norm": 2.1230286585075655,
      "language_loss": 0.72361958,
      "learning_rate": 1.9865636730631904e-06,
      "loss": 0.74493355,
      "num_input_tokens_seen": 184905590,
      "step": 8600,
      "time_per_iteration": 2.5428106784820557
    },
    {
      "auxiliary_loss_clip": 0.01083676,
      "auxiliary_loss_mlp": 0.01034362,
      "balance_loss_clip": 1.04336452,
      "balance_loss_mlp": 1.02058148,
      "epoch": 0.5171200961972042,
      "flos": 20994455664000.0,
      "grad_norm": 2.180641977959101,
      "language_loss": 0.74425685,
      "learning_rate": 1.9861742204279602e-06,
      "loss": 0.76543725,
      "num_input_tokens_seen": 184925555,
      "step": 8601,
      "time_per_iteration": 2.566192626953125
    },
    {
      "auxiliary_loss_clip": 0.01105722,
      "auxiliary_loss_mlp": 0.01038287,
      "balance_loss_clip": 1.04074609,
      "balance_loss_mlp": 1.0245955,
      "epoch": 0.5171802194498722,
      "flos": 22745639427840.0,
      "grad_norm": 1.9881379230108247,
      "language_loss": 0.83963472,
      "learning_rate": 1.9857847683170045e-06,
      "loss": 0.86107481,
      "num_input_tokens_seen": 184944490,
      "step": 8602,
      "time_per_iteration": 2.468792200088501
    },
    {
      "auxiliary_loss_clip": 0.01116176,
      "auxiliary_loss_mlp": 0.0103321,
      "balance_loss_clip": 1.0411756,
      "balance_loss_mlp": 1.01985264,
      "epoch": 0.5172403427025402,
      "flos": 28176732668160.0,
      "grad_norm": 1.8942001734722118,
      "language_loss": 0.74595451,
      "learning_rate": 1.9853953167450926e-06,
      "loss": 0.76744837,
      "num_input_tokens_seen": 184963190,
      "step": 8603,
      "time_per_iteration": 2.5143773555755615
    },
    {
      "auxiliary_loss_clip": 0.01096605,
      "auxiliary_loss_mlp": 0.01034888,
      "balance_loss_clip": 1.04209101,
      "balance_loss_mlp": 1.02173889,
      "epoch": 0.5173004659552082,
      "flos": 20337829090560.0,
      "grad_norm": 2.057429506120012,
      "language_loss": 0.72203207,
      "learning_rate": 1.9850058657269915e-06,
      "loss": 0.74334705,
      "num_input_tokens_seen": 184981220,
      "step": 8604,
      "time_per_iteration": 2.5089542865753174
    },
    {
      "auxiliary_loss_clip": 0.01100328,
      "auxiliary_loss_mlp": 0.01033734,
      "balance_loss_clip": 1.04264891,
      "balance_loss_mlp": 1.01988792,
      "epoch": 0.5173605892078762,
      "flos": 19063323740160.0,
      "grad_norm": 2.0697242504782287,
      "language_loss": 0.84679824,
      "learning_rate": 1.984616415277469e-06,
      "loss": 0.86813885,
      "num_input_tokens_seen": 184998810,
      "step": 8605,
      "time_per_iteration": 2.53499174118042
    },
    {
      "auxiliary_loss_clip": 0.01101002,
      "auxiliary_loss_mlp": 0.01026386,
      "balance_loss_clip": 1.04162335,
      "balance_loss_mlp": 1.01367247,
      "epoch": 0.5174207124605441,
      "flos": 27995168396160.0,
      "grad_norm": 1.4840250830994521,
      "language_loss": 0.64516449,
      "learning_rate": 1.984226965411294e-06,
      "loss": 0.66643834,
      "num_input_tokens_seen": 185021185,
      "step": 8606,
      "time_per_iteration": 2.551459789276123
    },
    {
      "auxiliary_loss_clip": 0.01091644,
      "auxiliary_loss_mlp": 0.01033346,
      "balance_loss_clip": 1.04339194,
      "balance_loss_mlp": 1.02017915,
      "epoch": 0.5174808357132121,
      "flos": 19496657416320.0,
      "grad_norm": 1.4977145397267289,
      "language_loss": 0.77585161,
      "learning_rate": 1.983837516143234e-06,
      "loss": 0.79710144,
      "num_input_tokens_seen": 185038465,
      "step": 8607,
      "time_per_iteration": 2.5023813247680664
    },
    {
      "auxiliary_loss_clip": 0.01105354,
      "auxiliary_loss_mlp": 0.01038693,
      "balance_loss_clip": 1.04021573,
      "balance_loss_mlp": 1.02462637,
      "epoch": 0.51754095896588,
      "flos": 22784171742720.0,
      "grad_norm": 1.7001225941145333,
      "language_loss": 0.71555322,
      "learning_rate": 1.983448067488057e-06,
      "loss": 0.73699367,
      "num_input_tokens_seen": 185057340,
      "step": 8608,
      "time_per_iteration": 3.8431813716888428
    },
    {
      "auxiliary_loss_clip": 0.01112579,
      "auxiliary_loss_mlp": 0.01030899,
      "balance_loss_clip": 1.04196024,
      "balance_loss_mlp": 1.01688588,
      "epoch": 0.5176010822185481,
      "flos": 22669257156480.0,
      "grad_norm": 1.7990904336028233,
      "language_loss": 0.8646729,
      "learning_rate": 1.983058619460531e-06,
      "loss": 0.88610768,
      "num_input_tokens_seen": 185074935,
      "step": 8609,
      "time_per_iteration": 2.5052738189697266
    },
    {
      "auxiliary_loss_clip": 0.01102731,
      "auxiliary_loss_mlp": 0.01032207,
      "balance_loss_clip": 1.03873444,
      "balance_loss_mlp": 1.01988101,
      "epoch": 0.517661205471216,
      "flos": 23951196622080.0,
      "grad_norm": 1.6359443929812598,
      "language_loss": 0.73773992,
      "learning_rate": 1.9826691720754237e-06,
      "loss": 0.75908923,
      "num_input_tokens_seen": 185095050,
      "step": 8610,
      "time_per_iteration": 2.5109024047851562
    },
    {
      "auxiliary_loss_clip": 0.01120958,
      "auxiliary_loss_mlp": 0.0103153,
      "balance_loss_clip": 1.04194593,
      "balance_loss_mlp": 1.01689672,
      "epoch": 0.517721328723884,
      "flos": 15596076735360.0,
      "grad_norm": 1.8980612708239204,
      "language_loss": 0.67072654,
      "learning_rate": 1.9822797253475034e-06,
      "loss": 0.69225138,
      "num_input_tokens_seen": 185112275,
      "step": 8611,
      "time_per_iteration": 2.4561238288879395
    },
    {
      "auxiliary_loss_clip": 0.01114688,
      "auxiliary_loss_mlp": 0.01032243,
      "balance_loss_clip": 1.0398258,
      "balance_loss_mlp": 1.01887357,
      "epoch": 0.5177814519765519,
      "flos": 20960197067520.0,
      "grad_norm": 2.113023557949929,
      "language_loss": 0.77132505,
      "learning_rate": 1.9818902792915373e-06,
      "loss": 0.79279435,
      "num_input_tokens_seen": 185132165,
      "step": 8612,
      "time_per_iteration": 3.89871883392334
    },
    {
      "auxiliary_loss_clip": 0.01105385,
      "auxiliary_loss_mlp": 0.01037613,
      "balance_loss_clip": 1.03916693,
      "balance_loss_mlp": 1.02485716,
      "epoch": 0.5178415752292199,
      "flos": 17967832796160.0,
      "grad_norm": 1.9329634531389925,
      "language_loss": 0.8221401,
      "learning_rate": 1.981500833922294e-06,
      "loss": 0.84357005,
      "num_input_tokens_seen": 185151025,
      "step": 8613,
      "time_per_iteration": 2.4571821689605713
    },
    {
      "auxiliary_loss_clip": 0.01118652,
      "auxiliary_loss_mlp": 0.01036757,
      "balance_loss_clip": 1.04296279,
      "balance_loss_mlp": 1.02273762,
      "epoch": 0.5179016984818878,
      "flos": 17821496787840.0,
      "grad_norm": 2.5347526307930255,
      "language_loss": 0.66523302,
      "learning_rate": 1.981111389254541e-06,
      "loss": 0.68678707,
      "num_input_tokens_seen": 185168455,
      "step": 8614,
      "time_per_iteration": 3.8440372943878174
    },
    {
      "auxiliary_loss_clip": 0.01098199,
      "auxiliary_loss_mlp": 0.01033326,
      "balance_loss_clip": 1.04061699,
      "balance_loss_mlp": 1.01952744,
      "epoch": 0.5179618217345558,
      "flos": 17820455293440.0,
      "grad_norm": 1.9245565061443928,
      "language_loss": 0.86345506,
      "learning_rate": 1.9807219453030453e-06,
      "loss": 0.88477027,
      "num_input_tokens_seen": 185184415,
      "step": 8615,
      "time_per_iteration": 2.4730215072631836
    },
    {
      "auxiliary_loss_clip": 0.01103307,
      "auxiliary_loss_mlp": 0.01040736,
      "balance_loss_clip": 1.0403161,
      "balance_loss_mlp": 1.02874982,
      "epoch": 0.5180219449872238,
      "flos": 22522131048960.0,
      "grad_norm": 1.457679491366675,
      "language_loss": 0.80696118,
      "learning_rate": 1.9803325020825763e-06,
      "loss": 0.82840163,
      "num_input_tokens_seen": 185202910,
      "step": 8616,
      "time_per_iteration": 2.5070748329162598
    },
    {
      "auxiliary_loss_clip": 0.01112415,
      "auxiliary_loss_mlp": 0.00791932,
      "balance_loss_clip": 1.04645741,
      "balance_loss_mlp": 1.01826954,
      "epoch": 0.5180820682398918,
      "flos": 23915465568000.0,
      "grad_norm": 1.8703678131432038,
      "language_loss": 0.75376189,
      "learning_rate": 1.9799430596079e-06,
      "loss": 0.77280533,
      "num_input_tokens_seen": 185223085,
      "step": 8617,
      "time_per_iteration": 2.5211021900177
    },
    {
      "auxiliary_loss_clip": 0.01116094,
      "auxiliary_loss_mlp": 0.01040142,
      "balance_loss_clip": 1.04019082,
      "balance_loss_mlp": 1.02600408,
      "epoch": 0.5181421914925598,
      "flos": 16979930064000.0,
      "grad_norm": 1.7310799716943774,
      "language_loss": 0.70150149,
      "learning_rate": 1.979553617893785e-06,
      "loss": 0.72306389,
      "num_input_tokens_seen": 185241295,
      "step": 8618,
      "time_per_iteration": 2.4386653900146484
    },
    {
      "auxiliary_loss_clip": 0.01027701,
      "auxiliary_loss_mlp": 0.01004069,
      "balance_loss_clip": 1.01135397,
      "balance_loss_mlp": 1.00273943,
      "epoch": 0.5182023147452277,
      "flos": 66059870872320.0,
      "grad_norm": 0.9552707444842252,
      "language_loss": 0.67323828,
      "learning_rate": 1.979164176954999e-06,
      "loss": 0.69355601,
      "num_input_tokens_seen": 185298295,
      "step": 8619,
      "time_per_iteration": 4.444633483886719
    },
    {
      "auxiliary_loss_clip": 0.01066565,
      "auxiliary_loss_mlp": 0.01028074,
      "balance_loss_clip": 1.03708029,
      "balance_loss_mlp": 1.01578379,
      "epoch": 0.5182624379978957,
      "flos": 18187749815040.0,
      "grad_norm": 2.0407339789914674,
      "language_loss": 0.79770744,
      "learning_rate": 1.97877473680631e-06,
      "loss": 0.81865382,
      "num_input_tokens_seen": 185317000,
      "step": 8620,
      "time_per_iteration": 2.565626621246338
    },
    {
      "auxiliary_loss_clip": 0.01052473,
      "auxiliary_loss_mlp": 0.00789746,
      "balance_loss_clip": 1.04146719,
      "balance_loss_mlp": 1.01435924,
      "epoch": 0.5183225612505636,
      "flos": 14026708638720.0,
      "grad_norm": 2.144723481257175,
      "language_loss": 0.82410729,
      "learning_rate": 1.9783852974624846e-06,
      "loss": 0.84252942,
      "num_input_tokens_seen": 185331185,
      "step": 8621,
      "time_per_iteration": 2.617462158203125
    },
    {
      "auxiliary_loss_clip": 0.01089281,
      "auxiliary_loss_mlp": 0.01036978,
      "balance_loss_clip": 1.03642046,
      "balance_loss_mlp": 1.02479458,
      "epoch": 0.5183826845032317,
      "flos": 23659781581440.0,
      "grad_norm": 2.6932757791805613,
      "language_loss": 0.65788847,
      "learning_rate": 1.9779958589382905e-06,
      "loss": 0.67915106,
      "num_input_tokens_seen": 185348955,
      "step": 8622,
      "time_per_iteration": 2.542924404144287
    },
    {
      "auxiliary_loss_clip": 0.01097311,
      "auxiliary_loss_mlp": 0.01044226,
      "balance_loss_clip": 1.04015231,
      "balance_loss_mlp": 1.03045094,
      "epoch": 0.5184428077558996,
      "flos": 15888605097600.0,
      "grad_norm": 1.8790793169907474,
      "language_loss": 0.60726023,
      "learning_rate": 1.977606421248497e-06,
      "loss": 0.62867558,
      "num_input_tokens_seen": 185367330,
      "step": 8623,
      "time_per_iteration": 2.5110342502593994
    },
    {
      "auxiliary_loss_clip": 0.01115131,
      "auxiliary_loss_mlp": 0.0103223,
      "balance_loss_clip": 1.03985214,
      "balance_loss_mlp": 1.01982045,
      "epoch": 0.5185029310085676,
      "flos": 21030833162880.0,
      "grad_norm": 1.8068808834345298,
      "language_loss": 0.76377332,
      "learning_rate": 1.9772169844078685e-06,
      "loss": 0.78524697,
      "num_input_tokens_seen": 185385060,
      "step": 8624,
      "time_per_iteration": 2.462740898132324
    },
    {
      "auxiliary_loss_clip": 0.01070055,
      "auxiliary_loss_mlp": 0.01044908,
      "balance_loss_clip": 1.03627169,
      "balance_loss_mlp": 1.03068078,
      "epoch": 0.5185630542612355,
      "flos": 26542690133760.0,
      "grad_norm": 1.9402870740706513,
      "language_loss": 0.71271795,
      "learning_rate": 1.9768275484311756e-06,
      "loss": 0.73386759,
      "num_input_tokens_seen": 185403745,
      "step": 8625,
      "time_per_iteration": 2.602905511856079
    },
    {
      "auxiliary_loss_clip": 0.01092654,
      "auxiliary_loss_mlp": 0.01032411,
      "balance_loss_clip": 1.03752184,
      "balance_loss_mlp": 1.01991749,
      "epoch": 0.5186231775139035,
      "flos": 20668422890880.0,
      "grad_norm": 1.8991767441278926,
      "language_loss": 0.67653024,
      "learning_rate": 1.976438113333184e-06,
      "loss": 0.69778085,
      "num_input_tokens_seen": 185422620,
      "step": 8626,
      "time_per_iteration": 2.5211689472198486
    },
    {
      "auxiliary_loss_clip": 0.01101416,
      "auxiliary_loss_mlp": 0.01031772,
      "balance_loss_clip": 1.03826964,
      "balance_loss_mlp": 1.01927876,
      "epoch": 0.5186833007665714,
      "flos": 20885502735360.0,
      "grad_norm": 2.262655165882253,
      "language_loss": 0.70462686,
      "learning_rate": 1.9760486791286612e-06,
      "loss": 0.7259587,
      "num_input_tokens_seen": 185439380,
      "step": 8627,
      "time_per_iteration": 2.513781785964966
    },
    {
      "auxiliary_loss_clip": 0.01119817,
      "auxiliary_loss_mlp": 0.0079075,
      "balance_loss_clip": 1.04184008,
      "balance_loss_mlp": 1.01597798,
      "epoch": 0.5187434240192395,
      "flos": 20886903365760.0,
      "grad_norm": 1.732148273090294,
      "language_loss": 0.73344553,
      "learning_rate": 1.9756592458323753e-06,
      "loss": 0.75255126,
      "num_input_tokens_seen": 185458830,
      "step": 8628,
      "time_per_iteration": 2.4724137783050537
    },
    {
      "auxiliary_loss_clip": 0.01091875,
      "auxiliary_loss_mlp": 0.01031923,
      "balance_loss_clip": 1.04012024,
      "balance_loss_mlp": 1.01974559,
      "epoch": 0.5188035472719074,
      "flos": 19859929614720.0,
      "grad_norm": 1.7421064604622019,
      "language_loss": 0.77487451,
      "learning_rate": 1.9752698134590927e-06,
      "loss": 0.79611254,
      "num_input_tokens_seen": 185477270,
      "step": 8629,
      "time_per_iteration": 2.5529510974884033
    },
    {
      "auxiliary_loss_clip": 0.011081,
      "auxiliary_loss_mlp": 0.01030999,
      "balance_loss_clip": 1.04129827,
      "balance_loss_mlp": 1.01721287,
      "epoch": 0.5188636705245754,
      "flos": 21138313633920.0,
      "grad_norm": 2.0608653472631775,
      "language_loss": 0.74582458,
      "learning_rate": 1.9748803820235815e-06,
      "loss": 0.76721555,
      "num_input_tokens_seen": 185495795,
      "step": 8630,
      "time_per_iteration": 2.4912517070770264
    },
    {
      "auxiliary_loss_clip": 0.01104471,
      "auxiliary_loss_mlp": 0.01037631,
      "balance_loss_clip": 1.0394485,
      "balance_loss_mlp": 1.02380884,
      "epoch": 0.5189237937772434,
      "flos": 22419786222720.0,
      "grad_norm": 1.6261331927624005,
      "language_loss": 0.80480981,
      "learning_rate": 1.9744909515406093e-06,
      "loss": 0.82623088,
      "num_input_tokens_seen": 185514885,
      "step": 8631,
      "time_per_iteration": 2.515448808670044
    },
    {
      "auxiliary_loss_clip": 0.01105655,
      "auxiliary_loss_mlp": 0.01030781,
      "balance_loss_clip": 1.04018819,
      "balance_loss_mlp": 1.01759684,
      "epoch": 0.5189839170299113,
      "flos": 25446696399360.0,
      "grad_norm": 1.5483085180614025,
      "language_loss": 0.74489039,
      "learning_rate": 1.974101522024942e-06,
      "loss": 0.76625472,
      "num_input_tokens_seen": 185537155,
      "step": 8632,
      "time_per_iteration": 2.5398783683776855
    },
    {
      "auxiliary_loss_clip": 0.01077034,
      "auxiliary_loss_mlp": 0.01030477,
      "balance_loss_clip": 1.03555846,
      "balance_loss_mlp": 1.01798368,
      "epoch": 0.5190440402825793,
      "flos": 18587722734720.0,
      "grad_norm": 3.9692696883533416,
      "language_loss": 0.78825802,
      "learning_rate": 1.9737120934913477e-06,
      "loss": 0.80933315,
      "num_input_tokens_seen": 185555520,
      "step": 8633,
      "time_per_iteration": 2.539912700653076
    },
    {
      "auxiliary_loss_clip": 0.0110449,
      "auxiliary_loss_mlp": 0.01032066,
      "balance_loss_clip": 1.03854156,
      "balance_loss_mlp": 1.01938832,
      "epoch": 0.5191041635352472,
      "flos": 21908633731200.0,
      "grad_norm": 1.6070408338019309,
      "language_loss": 0.80434287,
      "learning_rate": 1.9733226659545936e-06,
      "loss": 0.82570845,
      "num_input_tokens_seen": 185573855,
      "step": 8634,
      "time_per_iteration": 2.558678388595581
    },
    {
      "auxiliary_loss_clip": 0.01110802,
      "auxiliary_loss_mlp": 0.0103714,
      "balance_loss_clip": 1.03786922,
      "balance_loss_mlp": 1.02431917,
      "epoch": 0.5191642867879153,
      "flos": 27527971173120.0,
      "grad_norm": 2.503726346579817,
      "language_loss": 0.68689042,
      "learning_rate": 1.9729332394294467e-06,
      "loss": 0.70836985,
      "num_input_tokens_seen": 185595145,
      "step": 8635,
      "time_per_iteration": 2.518061399459839
    },
    {
      "auxiliary_loss_clip": 0.01091588,
      "auxiliary_loss_mlp": 0.01033248,
      "balance_loss_clip": 1.0377512,
      "balance_loss_mlp": 1.02031398,
      "epoch": 0.5192244100405832,
      "flos": 15705999331200.0,
      "grad_norm": 1.7070705428814292,
      "language_loss": 0.77734834,
      "learning_rate": 1.9725438139306742e-06,
      "loss": 0.79859668,
      "num_input_tokens_seen": 185613320,
      "step": 8636,
      "time_per_iteration": 2.484654188156128
    },
    {
      "auxiliary_loss_clip": 0.01116639,
      "auxiliary_loss_mlp": 0.01031858,
      "balance_loss_clip": 1.04002023,
      "balance_loss_mlp": 1.01842952,
      "epoch": 0.5192845332932512,
      "flos": 12057080313600.0,
      "grad_norm": 2.2779517998232692,
      "language_loss": 0.71702075,
      "learning_rate": 1.9721543894730425e-06,
      "loss": 0.73850572,
      "num_input_tokens_seen": 185630730,
      "step": 8637,
      "time_per_iteration": 2.4275684356689453
    },
    {
      "auxiliary_loss_clip": 0.01077792,
      "auxiliary_loss_mlp": 0.01033174,
      "balance_loss_clip": 1.03835809,
      "balance_loss_mlp": 1.0206871,
      "epoch": 0.5193446565459191,
      "flos": 18953185662720.0,
      "grad_norm": 1.882226768787918,
      "language_loss": 0.75868416,
      "learning_rate": 1.9717649660713194e-06,
      "loss": 0.77979374,
      "num_input_tokens_seen": 185648515,
      "step": 8638,
      "time_per_iteration": 2.519571304321289
    },
    {
      "auxiliary_loss_clip": 0.01081546,
      "auxiliary_loss_mlp": 0.0103252,
      "balance_loss_clip": 1.03955531,
      "balance_loss_mlp": 1.01970518,
      "epoch": 0.5194047797985871,
      "flos": 20374960775040.0,
      "grad_norm": 1.8233617478808053,
      "language_loss": 0.74827063,
      "learning_rate": 1.971375543740272e-06,
      "loss": 0.76941127,
      "num_input_tokens_seen": 185665220,
      "step": 8639,
      "time_per_iteration": 2.5331456661224365
    },
    {
      "auxiliary_loss_clip": 0.01112773,
      "auxiliary_loss_mlp": 0.01027994,
      "balance_loss_clip": 1.03929496,
      "balance_loss_mlp": 1.01505983,
      "epoch": 0.519464903051255,
      "flos": 24353001135360.0,
      "grad_norm": 1.8245252823827376,
      "language_loss": 0.77578926,
      "learning_rate": 1.9709861224946665e-06,
      "loss": 0.79719698,
      "num_input_tokens_seen": 185683750,
      "step": 8640,
      "time_per_iteration": 2.4831273555755615
    },
    {
      "auxiliary_loss_clip": 0.0107576,
      "auxiliary_loss_mlp": 0.01031179,
      "balance_loss_clip": 1.04236495,
      "balance_loss_mlp": 1.01864398,
      "epoch": 0.519525026303923,
      "flos": 14061829161600.0,
      "grad_norm": 2.23274805898455,
      "language_loss": 0.65969908,
      "learning_rate": 1.97059670234927e-06,
      "loss": 0.68076849,
      "num_input_tokens_seen": 185700625,
      "step": 8641,
      "time_per_iteration": 2.511431932449341
    },
    {
      "auxiliary_loss_clip": 0.01112971,
      "auxiliary_loss_mlp": 0.01029658,
      "balance_loss_clip": 1.0400697,
      "balance_loss_mlp": 1.01720643,
      "epoch": 0.519585149556591,
      "flos": 28835873193600.0,
      "grad_norm": 1.8754827005115735,
      "language_loss": 0.76196408,
      "learning_rate": 1.97020728331885e-06,
      "loss": 0.7833904,
      "num_input_tokens_seen": 185721155,
      "step": 8642,
      "time_per_iteration": 2.5133354663848877
    },
    {
      "auxiliary_loss_clip": 0.01113205,
      "auxiliary_loss_mlp": 0.01029929,
      "balance_loss_clip": 1.04036272,
      "balance_loss_mlp": 1.0173825,
      "epoch": 0.519645272809259,
      "flos": 25373007648000.0,
      "grad_norm": 1.5287769252861678,
      "language_loss": 0.82806933,
      "learning_rate": 1.9698178654181726e-06,
      "loss": 0.84950066,
      "num_input_tokens_seen": 185740990,
      "step": 8643,
      "time_per_iteration": 2.5063583850860596
    },
    {
      "auxiliary_loss_clip": 0.01116197,
      "auxiliary_loss_mlp": 0.01040539,
      "balance_loss_clip": 1.03926909,
      "balance_loss_mlp": 1.02707982,
      "epoch": 0.519705396061927,
      "flos": 25372863993600.0,
      "grad_norm": 1.5872518823115098,
      "language_loss": 0.70307505,
      "learning_rate": 1.969428448662004e-06,
      "loss": 0.7246424,
      "num_input_tokens_seen": 185762235,
      "step": 8644,
      "time_per_iteration": 2.493152141571045
    },
    {
      "auxiliary_loss_clip": 0.01102603,
      "auxiliary_loss_mlp": 0.00791586,
      "balance_loss_clip": 1.03869605,
      "balance_loss_mlp": 1.01806927,
      "epoch": 0.5197655193145949,
      "flos": 28476228268800.0,
      "grad_norm": 1.5936948648610312,
      "language_loss": 0.80137736,
      "learning_rate": 1.9690390330651133e-06,
      "loss": 0.82031929,
      "num_input_tokens_seen": 185783415,
      "step": 8645,
      "time_per_iteration": 2.5556018352508545
    },
    {
      "auxiliary_loss_clip": 0.01112383,
      "auxiliary_loss_mlp": 0.01027202,
      "balance_loss_clip": 1.03740573,
      "balance_loss_mlp": 1.01390433,
      "epoch": 0.5198256425672629,
      "flos": 20009138711040.0,
      "grad_norm": 1.689974194669513,
      "language_loss": 0.77839077,
      "learning_rate": 1.968649618642264e-06,
      "loss": 0.79978669,
      "num_input_tokens_seen": 185801345,
      "step": 8646,
      "time_per_iteration": 2.451239585876465
    },
    {
      "auxiliary_loss_clip": 0.01106524,
      "auxiliary_loss_mlp": 0.01035618,
      "balance_loss_clip": 1.04076004,
      "balance_loss_mlp": 1.02287447,
      "epoch": 0.5198857658199308,
      "flos": 19828867328640.0,
      "grad_norm": 1.8369655537570007,
      "language_loss": 0.65590304,
      "learning_rate": 1.9682602054082252e-06,
      "loss": 0.67732447,
      "num_input_tokens_seen": 185820815,
      "step": 8647,
      "time_per_iteration": 3.8405275344848633
    },
    {
      "auxiliary_loss_clip": 0.01115774,
      "auxiliary_loss_mlp": 0.01031931,
      "balance_loss_clip": 1.03884268,
      "balance_loss_mlp": 1.01750708,
      "epoch": 0.5199458890725989,
      "flos": 24461918150400.0,
      "grad_norm": 2.01948628128364,
      "language_loss": 0.71659976,
      "learning_rate": 1.967870793377763e-06,
      "loss": 0.73807681,
      "num_input_tokens_seen": 185841450,
      "step": 8648,
      "time_per_iteration": 2.5002434253692627
    },
    {
      "auxiliary_loss_clip": 0.01095778,
      "auxiliary_loss_mlp": 0.01029353,
      "balance_loss_clip": 1.04074311,
      "balance_loss_mlp": 1.01601398,
      "epoch": 0.5200060123252668,
      "flos": 23404779953280.0,
      "grad_norm": 1.9758222854651328,
      "language_loss": 0.64301622,
      "learning_rate": 1.967481382565642e-06,
      "loss": 0.66426748,
      "num_input_tokens_seen": 185859935,
      "step": 8649,
      "time_per_iteration": 2.542113780975342
    },
    {
      "auxiliary_loss_clip": 0.01092584,
      "auxiliary_loss_mlp": 0.01035441,
      "balance_loss_clip": 1.0396595,
      "balance_loss_mlp": 1.02071285,
      "epoch": 0.5200661355779348,
      "flos": 17201355454080.0,
      "grad_norm": 1.6925558475643878,
      "language_loss": 0.70253748,
      "learning_rate": 1.9670919729866315e-06,
      "loss": 0.72381771,
      "num_input_tokens_seen": 185876795,
      "step": 8650,
      "time_per_iteration": 2.4972283840179443
    },
    {
      "auxiliary_loss_clip": 0.01112512,
      "auxiliary_loss_mlp": 0.01028925,
      "balance_loss_clip": 1.03820491,
      "balance_loss_mlp": 1.01615739,
      "epoch": 0.5201262588306027,
      "flos": 18515075477760.0,
      "grad_norm": 1.777585827141164,
      "language_loss": 0.78004038,
      "learning_rate": 1.966702564655496e-06,
      "loss": 0.80145478,
      "num_input_tokens_seen": 185895570,
      "step": 8651,
      "time_per_iteration": 4.09261679649353
    },
    {
      "auxiliary_loss_clip": 0.01061116,
      "auxiliary_loss_mlp": 0.01036794,
      "balance_loss_clip": 1.04038095,
      "balance_loss_mlp": 1.02291834,
      "epoch": 0.5201863820832707,
      "flos": 18619395552000.0,
      "grad_norm": 2.7212502647484373,
      "language_loss": 0.78867269,
      "learning_rate": 1.966313157587003e-06,
      "loss": 0.80965173,
      "num_input_tokens_seen": 185913700,
      "step": 8652,
      "time_per_iteration": 2.6391310691833496
    },
    {
      "auxiliary_loss_clip": 0.01080732,
      "auxiliary_loss_mlp": 0.01032611,
      "balance_loss_clip": 1.04332149,
      "balance_loss_mlp": 1.01735771,
      "epoch": 0.5202465053359386,
      "flos": 22857142222080.0,
      "grad_norm": 1.7378500904169487,
      "language_loss": 0.70319384,
      "learning_rate": 1.9659237517959187e-06,
      "loss": 0.72432733,
      "num_input_tokens_seen": 185932460,
      "step": 8653,
      "time_per_iteration": 3.9839365482330322
    },
    {
      "auxiliary_loss_clip": 0.01083811,
      "auxiliary_loss_mlp": 0.01040773,
      "balance_loss_clip": 1.03956068,
      "balance_loss_mlp": 1.02678931,
      "epoch": 0.5203066285886067,
      "flos": 21981532383360.0,
      "grad_norm": 2.465497332261721,
      "language_loss": 0.78455937,
      "learning_rate": 1.965534347297008e-06,
      "loss": 0.80580521,
      "num_input_tokens_seen": 185952030,
      "step": 8654,
      "time_per_iteration": 2.5471973419189453
    },
    {
      "auxiliary_loss_clip": 0.01108188,
      "auxiliary_loss_mlp": 0.01037601,
      "balance_loss_clip": 1.04179239,
      "balance_loss_mlp": 1.02391601,
      "epoch": 0.5203667518412746,
      "flos": 20233329448320.0,
      "grad_norm": 2.0324611007248534,
      "language_loss": 0.84102285,
      "learning_rate": 1.9651449441050393e-06,
      "loss": 0.86248076,
      "num_input_tokens_seen": 185973130,
      "step": 8655,
      "time_per_iteration": 2.5385210514068604
    },
    {
      "auxiliary_loss_clip": 0.0110487,
      "auxiliary_loss_mlp": 0.01031588,
      "balance_loss_clip": 1.04070377,
      "balance_loss_mlp": 1.01955414,
      "epoch": 0.5204268750939426,
      "flos": 15705460627200.0,
      "grad_norm": 2.6022067176134818,
      "language_loss": 0.65828109,
      "learning_rate": 1.9647555422347777e-06,
      "loss": 0.67964566,
      "num_input_tokens_seen": 185990200,
      "step": 8656,
      "time_per_iteration": 2.461646556854248
    },
    {
      "auxiliary_loss_clip": 0.01076417,
      "auxiliary_loss_mlp": 0.0104012,
      "balance_loss_clip": 1.04244661,
      "balance_loss_mlp": 1.02675664,
      "epoch": 0.5204869983466105,
      "flos": 27449469999360.0,
      "grad_norm": 1.842141207239616,
      "language_loss": 0.73149371,
      "learning_rate": 1.9643661417009893e-06,
      "loss": 0.75265908,
      "num_input_tokens_seen": 186009880,
      "step": 8657,
      "time_per_iteration": 2.603379249572754
    },
    {
      "auxiliary_loss_clip": 0.01081224,
      "auxiliary_loss_mlp": 0.01031077,
      "balance_loss_clip": 1.038468,
      "balance_loss_mlp": 1.01732051,
      "epoch": 0.5205471215992785,
      "flos": 20595452411520.0,
      "grad_norm": 1.8091173280179478,
      "language_loss": 0.7185241,
      "learning_rate": 1.9639767425184408e-06,
      "loss": 0.73964703,
      "num_input_tokens_seen": 186026680,
      "step": 8658,
      "time_per_iteration": 4.023704290390015
    },
    {
      "auxiliary_loss_clip": 0.0111374,
      "auxiliary_loss_mlp": 0.01031741,
      "balance_loss_clip": 1.03816843,
      "balance_loss_mlp": 1.01890242,
      "epoch": 0.5206072448519465,
      "flos": 22127904305280.0,
      "grad_norm": 1.9619235705195328,
      "language_loss": 0.83147341,
      "learning_rate": 1.963587344701897e-06,
      "loss": 0.85292816,
      "num_input_tokens_seen": 186046920,
      "step": 8659,
      "time_per_iteration": 2.4564363956451416
    },
    {
      "auxiliary_loss_clip": 0.01095368,
      "auxiliary_loss_mlp": 0.01039163,
      "balance_loss_clip": 1.04005694,
      "balance_loss_mlp": 1.02420807,
      "epoch": 0.5206673681046144,
      "flos": 18330422636160.0,
      "grad_norm": 1.9226158220637388,
      "language_loss": 0.75475729,
      "learning_rate": 1.9631979482661253e-06,
      "loss": 0.7761026,
      "num_input_tokens_seen": 186062090,
      "step": 8660,
      "time_per_iteration": 2.490968704223633
    },
    {
      "auxiliary_loss_clip": 0.01114008,
      "auxiliary_loss_mlp": 0.01031966,
      "balance_loss_clip": 1.03960741,
      "balance_loss_mlp": 1.01958585,
      "epoch": 0.5207274913572825,
      "flos": 20230240878720.0,
      "grad_norm": 1.8905497208617705,
      "language_loss": 0.77989209,
      "learning_rate": 1.9628085532258906e-06,
      "loss": 0.80135185,
      "num_input_tokens_seen": 186081135,
      "step": 8661,
      "time_per_iteration": 2.4487507343292236
    },
    {
      "auxiliary_loss_clip": 0.01094765,
      "auxiliary_loss_mlp": 0.01029644,
      "balance_loss_clip": 1.03846431,
      "balance_loss_mlp": 1.01718652,
      "epoch": 0.5207876146099504,
      "flos": 22127042378880.0,
      "grad_norm": 2.0435291822304946,
      "language_loss": 0.70238805,
      "learning_rate": 1.9624191595959603e-06,
      "loss": 0.7236321,
      "num_input_tokens_seen": 186099700,
      "step": 8662,
      "time_per_iteration": 2.5561916828155518
    },
    {
      "auxiliary_loss_clip": 0.01100255,
      "auxiliary_loss_mlp": 0.01028809,
      "balance_loss_clip": 1.03891993,
      "balance_loss_mlp": 1.01496255,
      "epoch": 0.5208477378626184,
      "flos": 23878908501120.0,
      "grad_norm": 1.676964131272476,
      "language_loss": 0.69494367,
      "learning_rate": 1.962029767391098e-06,
      "loss": 0.71623433,
      "num_input_tokens_seen": 186119740,
      "step": 8663,
      "time_per_iteration": 2.5034971237182617
    },
    {
      "auxiliary_loss_clip": 0.01086297,
      "auxiliary_loss_mlp": 0.00790036,
      "balance_loss_clip": 1.03815198,
      "balance_loss_mlp": 1.0140326,
      "epoch": 0.5209078611152863,
      "flos": 20961525870720.0,
      "grad_norm": 2.6085773381028927,
      "language_loss": 0.76811326,
      "learning_rate": 1.961640376626072e-06,
      "loss": 0.78687662,
      "num_input_tokens_seen": 186140645,
      "step": 8664,
      "time_per_iteration": 2.5453169345855713
    },
    {
      "auxiliary_loss_clip": 0.01090071,
      "auxiliary_loss_mlp": 0.01032676,
      "balance_loss_clip": 1.0413065,
      "balance_loss_mlp": 1.0195744,
      "epoch": 0.5209679843679543,
      "flos": 20667740532480.0,
      "grad_norm": 3.8763948052211767,
      "language_loss": 0.76015508,
      "learning_rate": 1.961250987315646e-06,
      "loss": 0.78138262,
      "num_input_tokens_seen": 186160130,
      "step": 8665,
      "time_per_iteration": 2.52441143989563
    },
    {
      "auxiliary_loss_clip": 0.01103812,
      "auxiliary_loss_mlp": 0.01030223,
      "balance_loss_clip": 1.04054928,
      "balance_loss_mlp": 1.01809347,
      "epoch": 0.5210281076206222,
      "flos": 20227295963520.0,
      "grad_norm": 1.9239550526599731,
      "language_loss": 0.72301877,
      "learning_rate": 1.960861599474586e-06,
      "loss": 0.74435914,
      "num_input_tokens_seen": 186179485,
      "step": 8666,
      "time_per_iteration": 2.501018524169922
    },
    {
      "auxiliary_loss_clip": 0.0110119,
      "auxiliary_loss_mlp": 0.01034778,
      "balance_loss_clip": 1.04129028,
      "balance_loss_mlp": 1.01955462,
      "epoch": 0.5210882308732903,
      "flos": 16069989801600.0,
      "grad_norm": 2.3906737167742182,
      "language_loss": 0.68475461,
      "learning_rate": 1.9604722131176592e-06,
      "loss": 0.70611429,
      "num_input_tokens_seen": 186197140,
      "step": 8667,
      "time_per_iteration": 2.507136583328247
    },
    {
      "auxiliary_loss_clip": 0.01070726,
      "auxiliary_loss_mlp": 0.01028167,
      "balance_loss_clip": 1.03863716,
      "balance_loss_mlp": 1.01545882,
      "epoch": 0.5211483541259582,
      "flos": 24825298089600.0,
      "grad_norm": 1.4432150085768403,
      "language_loss": 0.80775458,
      "learning_rate": 1.960082828259629e-06,
      "loss": 0.82874346,
      "num_input_tokens_seen": 186216800,
      "step": 8668,
      "time_per_iteration": 2.637395143508911
    },
    {
      "auxiliary_loss_clip": 0.01094062,
      "auxiliary_loss_mlp": 0.01029462,
      "balance_loss_clip": 1.0389092,
      "balance_loss_mlp": 1.01647377,
      "epoch": 0.5212084773786262,
      "flos": 20370651143040.0,
      "grad_norm": 1.9936162939812752,
      "language_loss": 0.64025676,
      "learning_rate": 1.9596934449152623e-06,
      "loss": 0.66149199,
      "num_input_tokens_seen": 186235320,
      "step": 8669,
      "time_per_iteration": 2.5106399059295654
    },
    {
      "auxiliary_loss_clip": 0.01092902,
      "auxiliary_loss_mlp": 0.00791674,
      "balance_loss_clip": 1.04062152,
      "balance_loss_mlp": 1.01676881,
      "epoch": 0.5212686006312941,
      "flos": 23145468693120.0,
      "grad_norm": 1.5867963858312661,
      "language_loss": 0.66485476,
      "learning_rate": 1.959304063099325e-06,
      "loss": 0.68370044,
      "num_input_tokens_seen": 186254460,
      "step": 8670,
      "time_per_iteration": 2.564114809036255
    },
    {
      "auxiliary_loss_clip": 0.01075817,
      "auxiliary_loss_mlp": 0.01030887,
      "balance_loss_clip": 1.0377295,
      "balance_loss_mlp": 1.01826823,
      "epoch": 0.5213287238839621,
      "flos": 27774030314880.0,
      "grad_norm": 2.0371900041078645,
      "language_loss": 0.75988162,
      "learning_rate": 1.9589146828265806e-06,
      "loss": 0.78094864,
      "num_input_tokens_seen": 186269465,
      "step": 8671,
      "time_per_iteration": 2.577138662338257
    },
    {
      "auxiliary_loss_clip": 0.010869,
      "auxiliary_loss_mlp": 0.01032904,
      "balance_loss_clip": 1.04657304,
      "balance_loss_mlp": 1.01929677,
      "epoch": 0.5213888471366301,
      "flos": 19937676602880.0,
      "grad_norm": 17.105786659557662,
      "language_loss": 0.78165907,
      "learning_rate": 1.958525304111796e-06,
      "loss": 0.8028571,
      "num_input_tokens_seen": 186288660,
      "step": 8672,
      "time_per_iteration": 2.5721752643585205
    },
    {
      "auxiliary_loss_clip": 0.01074656,
      "auxiliary_loss_mlp": 0.01027845,
      "balance_loss_clip": 1.03516054,
      "balance_loss_mlp": 1.01585841,
      "epoch": 0.521448970389298,
      "flos": 16982731324800.0,
      "grad_norm": 2.0719204737226855,
      "language_loss": 0.72157842,
      "learning_rate": 1.958135926969736e-06,
      "loss": 0.74260342,
      "num_input_tokens_seen": 186305760,
      "step": 8673,
      "time_per_iteration": 2.514864444732666
    },
    {
      "auxiliary_loss_clip": 0.01099283,
      "auxiliary_loss_mlp": 0.01031235,
      "balance_loss_clip": 1.03796792,
      "balance_loss_mlp": 1.01799107,
      "epoch": 0.5215090936419661,
      "flos": 18989706816000.0,
      "grad_norm": 1.6930346337464368,
      "language_loss": 0.74821937,
      "learning_rate": 1.957746551415166e-06,
      "loss": 0.76952457,
      "num_input_tokens_seen": 186324135,
      "step": 8674,
      "time_per_iteration": 2.4813055992126465
    },
    {
      "auxiliary_loss_clip": 0.01088927,
      "auxiliary_loss_mlp": 0.01035961,
      "balance_loss_clip": 1.03878188,
      "balance_loss_mlp": 1.0211786,
      "epoch": 0.521569216894634,
      "flos": 16143427157760.0,
      "grad_norm": 1.959735143198719,
      "language_loss": 0.8638131,
      "learning_rate": 1.9573571774628506e-06,
      "loss": 0.88506198,
      "num_input_tokens_seen": 186340205,
      "step": 8675,
      "time_per_iteration": 2.48199462890625
    },
    {
      "auxiliary_loss_clip": 0.01023717,
      "auxiliary_loss_mlp": 0.01001321,
      "balance_loss_clip": 1.02431059,
      "balance_loss_mlp": 0.99999791,
      "epoch": 0.521629340147302,
      "flos": 57579493282560.0,
      "grad_norm": 0.8727889866804611,
      "language_loss": 0.63111597,
      "learning_rate": 1.9569678051275556e-06,
      "loss": 0.65136635,
      "num_input_tokens_seen": 186396940,
      "step": 8676,
      "time_per_iteration": 3.087775945663452
    },
    {
      "auxiliary_loss_clip": 0.01099878,
      "auxiliary_loss_mlp": 0.01027862,
      "balance_loss_clip": 1.03805399,
      "balance_loss_mlp": 1.01529741,
      "epoch": 0.5216894633999699,
      "flos": 26796901662720.0,
      "grad_norm": 1.7445827327926693,
      "language_loss": 0.69051492,
      "learning_rate": 1.956578434424046e-06,
      "loss": 0.71179235,
      "num_input_tokens_seen": 186418680,
      "step": 8677,
      "time_per_iteration": 2.5275986194610596
    },
    {
      "auxiliary_loss_clip": 0.01099798,
      "auxiliary_loss_mlp": 0.0103156,
      "balance_loss_clip": 1.03684926,
      "balance_loss_mlp": 1.01825595,
      "epoch": 0.5217495866526379,
      "flos": 26358719650560.0,
      "grad_norm": 5.657907097070374,
      "language_loss": 0.65031618,
      "learning_rate": 1.956189065367086e-06,
      "loss": 0.67162973,
      "num_input_tokens_seen": 186438265,
      "step": 8678,
      "time_per_iteration": 2.539461135864258
    },
    {
      "auxiliary_loss_clip": 0.01092376,
      "auxiliary_loss_mlp": 0.01032441,
      "balance_loss_clip": 1.03655982,
      "balance_loss_mlp": 1.01836276,
      "epoch": 0.5218097099053058,
      "flos": 23584009841280.0,
      "grad_norm": 2.3486404898146778,
      "language_loss": 0.68526196,
      "learning_rate": 1.9557996979714414e-06,
      "loss": 0.70651019,
      "num_input_tokens_seen": 186456870,
      "step": 8679,
      "time_per_iteration": 2.5256125926971436
    },
    {
      "auxiliary_loss_clip": 0.01115563,
      "auxiliary_loss_mlp": 0.01032283,
      "balance_loss_clip": 1.04051638,
      "balance_loss_mlp": 1.01897919,
      "epoch": 0.5218698331579739,
      "flos": 18077396256000.0,
      "grad_norm": 1.8897415674150861,
      "language_loss": 0.6679877,
      "learning_rate": 1.9554103322518764e-06,
      "loss": 0.68946618,
      "num_input_tokens_seen": 186476425,
      "step": 8680,
      "time_per_iteration": 2.4672603607177734
    },
    {
      "auxiliary_loss_clip": 0.01114402,
      "auxiliary_loss_mlp": 0.01028731,
      "balance_loss_clip": 1.03952765,
      "balance_loss_mlp": 1.0153383,
      "epoch": 0.5219299564106418,
      "flos": 19281121856640.0,
      "grad_norm": 1.9409112336652046,
      "language_loss": 0.83081436,
      "learning_rate": 1.955020968223156e-06,
      "loss": 0.85224569,
      "num_input_tokens_seen": 186492555,
      "step": 8681,
      "time_per_iteration": 2.444559335708618
    },
    {
      "auxiliary_loss_clip": 0.01088146,
      "auxiliary_loss_mlp": 0.01028997,
      "balance_loss_clip": 1.03766418,
      "balance_loss_mlp": 1.01644993,
      "epoch": 0.5219900796633098,
      "flos": 26651355753600.0,
      "grad_norm": 1.7068716944222566,
      "language_loss": 0.77768308,
      "learning_rate": 1.9546316059000454e-06,
      "loss": 0.79885447,
      "num_input_tokens_seen": 186513190,
      "step": 8682,
      "time_per_iteration": 2.5654208660125732
    },
    {
      "auxiliary_loss_clip": 0.01077515,
      "auxiliary_loss_mlp": 0.01039359,
      "balance_loss_clip": 1.0367806,
      "balance_loss_mlp": 1.026407,
      "epoch": 0.5220502029159777,
      "flos": 34312717382400.0,
      "grad_norm": 1.5186975305501975,
      "language_loss": 0.69082403,
      "learning_rate": 1.9542422452973082e-06,
      "loss": 0.71199274,
      "num_input_tokens_seen": 186534830,
      "step": 8683,
      "time_per_iteration": 2.6406726837158203
    },
    {
      "auxiliary_loss_clip": 0.01081916,
      "auxiliary_loss_mlp": 0.01035121,
      "balance_loss_clip": 1.03840232,
      "balance_loss_mlp": 1.02167392,
      "epoch": 0.5221103261686457,
      "flos": 22156488552960.0,
      "grad_norm": 1.624111477684696,
      "language_loss": 0.76172948,
      "learning_rate": 1.9538528864297104e-06,
      "loss": 0.78289986,
      "num_input_tokens_seen": 186554390,
      "step": 8684,
      "time_per_iteration": 2.5561280250549316
    },
    {
      "auxiliary_loss_clip": 0.01093688,
      "auxiliary_loss_mlp": 0.00788077,
      "balance_loss_clip": 1.03638148,
      "balance_loss_mlp": 1.01393366,
      "epoch": 0.5221704494213137,
      "flos": 19208402772480.0,
      "grad_norm": 1.649278347652008,
      "language_loss": 0.75947666,
      "learning_rate": 1.9534635293120153e-06,
      "loss": 0.77829432,
      "num_input_tokens_seen": 186572360,
      "step": 8685,
      "time_per_iteration": 3.879373788833618
    },
    {
      "auxiliary_loss_clip": 0.01090929,
      "auxiliary_loss_mlp": 0.0103485,
      "balance_loss_clip": 1.03983414,
      "balance_loss_mlp": 1.02224922,
      "epoch": 0.5222305726739817,
      "flos": 19354056422400.0,
      "grad_norm": 1.8468944599401247,
      "language_loss": 0.81012225,
      "learning_rate": 1.9530741739589876e-06,
      "loss": 0.83138001,
      "num_input_tokens_seen": 186590655,
      "step": 8686,
      "time_per_iteration": 2.4995908737182617
    },
    {
      "auxiliary_loss_clip": 0.01088291,
      "auxiliary_loss_mlp": 0.01030008,
      "balance_loss_clip": 1.03483987,
      "balance_loss_mlp": 1.01803958,
      "epoch": 0.5222906959266497,
      "flos": 27814789272960.0,
      "grad_norm": 1.843240689994817,
      "language_loss": 0.69451106,
      "learning_rate": 1.9526848203853927e-06,
      "loss": 0.71569407,
      "num_input_tokens_seen": 186610345,
      "step": 8687,
      "time_per_iteration": 2.579038143157959
    },
    {
      "auxiliary_loss_clip": 0.01107926,
      "auxiliary_loss_mlp": 0.01027485,
      "balance_loss_clip": 1.03627372,
      "balance_loss_mlp": 1.015499,
      "epoch": 0.5223508191793176,
      "flos": 12712988615040.0,
      "grad_norm": 2.0685933752717505,
      "language_loss": 0.83317316,
      "learning_rate": 1.9522954686059936e-06,
      "loss": 0.85452724,
      "num_input_tokens_seen": 186624360,
      "step": 8688,
      "time_per_iteration": 2.4149386882781982
    },
    {
      "auxiliary_loss_clip": 0.01099284,
      "auxiliary_loss_mlp": 0.00786259,
      "balance_loss_clip": 1.03697848,
      "balance_loss_mlp": 1.00750661,
      "epoch": 0.5224109424319856,
      "flos": 15632238752640.0,
      "grad_norm": 2.3979638977707998,
      "language_loss": 0.73249626,
      "learning_rate": 1.9519061186355558e-06,
      "loss": 0.75135165,
      "num_input_tokens_seen": 186638680,
      "step": 8689,
      "time_per_iteration": 3.857513189315796
    },
    {
      "auxiliary_loss_clip": 0.01087374,
      "auxiliary_loss_mlp": 0.01026873,
      "balance_loss_clip": 1.03503776,
      "balance_loss_mlp": 1.01392651,
      "epoch": 0.5224710656846535,
      "flos": 15742233175680.0,
      "grad_norm": 1.9199663752926093,
      "language_loss": 0.82369959,
      "learning_rate": 1.9515167704888417e-06,
      "loss": 0.84484208,
      "num_input_tokens_seen": 186655840,
      "step": 8690,
      "time_per_iteration": 2.4782347679138184
    },
    {
      "auxiliary_loss_clip": 0.01083907,
      "auxiliary_loss_mlp": 0.01034509,
      "balance_loss_clip": 1.03830159,
      "balance_loss_mlp": 1.02096689,
      "epoch": 0.5225311889373215,
      "flos": 26030998938240.0,
      "grad_norm": 2.4013411171283674,
      "language_loss": 0.79319954,
      "learning_rate": 1.9511274241806173e-06,
      "loss": 0.81438375,
      "num_input_tokens_seen": 186674150,
      "step": 8691,
      "time_per_iteration": 2.57486629486084
    },
    {
      "auxiliary_loss_clip": 0.0110675,
      "auxiliary_loss_mlp": 0.01031797,
      "balance_loss_clip": 1.03858805,
      "balance_loss_mlp": 1.01758146,
      "epoch": 0.5225913121899894,
      "flos": 18369278173440.0,
      "grad_norm": 2.0922449570523143,
      "language_loss": 0.77166033,
      "learning_rate": 1.950738079725646e-06,
      "loss": 0.79304582,
      "num_input_tokens_seen": 186690675,
      "step": 8692,
      "time_per_iteration": 3.891012668609619
    },
    {
      "auxiliary_loss_clip": 0.01097766,
      "auxiliary_loss_mlp": 0.01031413,
      "balance_loss_clip": 1.03742647,
      "balance_loss_mlp": 1.01959348,
      "epoch": 0.5226514354426575,
      "flos": 29273516501760.0,
      "grad_norm": 1.6522018230322035,
      "language_loss": 0.72648281,
      "learning_rate": 1.950348737138691e-06,
      "loss": 0.7477746,
      "num_input_tokens_seen": 186710380,
      "step": 8693,
      "time_per_iteration": 2.5275096893310547
    },
    {
      "auxiliary_loss_clip": 0.01116571,
      "auxiliary_loss_mlp": 0.01035423,
      "balance_loss_clip": 1.03856099,
      "balance_loss_mlp": 1.02095747,
      "epoch": 0.5227115586953254,
      "flos": 22853299466880.0,
      "grad_norm": 1.9583509776823018,
      "language_loss": 0.81809103,
      "learning_rate": 1.949959396434517e-06,
      "loss": 0.83961099,
      "num_input_tokens_seen": 186729135,
      "step": 8694,
      "time_per_iteration": 2.4680869579315186
    },
    {
      "auxiliary_loss_clip": 0.01010814,
      "auxiliary_loss_mlp": 0.0100502,
      "balance_loss_clip": 1.01996565,
      "balance_loss_mlp": 1.00347078,
      "epoch": 0.5227716819479934,
      "flos": 57474419022720.0,
      "grad_norm": 0.7688733641910941,
      "language_loss": 0.55761546,
      "learning_rate": 1.949570057627888e-06,
      "loss": 0.57777381,
      "num_input_tokens_seen": 186791115,
      "step": 8695,
      "time_per_iteration": 3.1782405376434326
    },
    {
      "auxiliary_loss_clip": 0.01057112,
      "auxiliary_loss_mlp": 0.01035315,
      "balance_loss_clip": 1.03825879,
      "balance_loss_mlp": 1.02232122,
      "epoch": 0.5228318052006613,
      "flos": 13808264077440.0,
      "grad_norm": 2.2608282030909495,
      "language_loss": 0.73245674,
      "learning_rate": 1.9491807207335672e-06,
      "loss": 0.75338095,
      "num_input_tokens_seen": 186808660,
      "step": 8696,
      "time_per_iteration": 4.035473823547363
    },
    {
      "auxiliary_loss_clip": 0.01088869,
      "auxiliary_loss_mlp": 0.01031304,
      "balance_loss_clip": 1.03721988,
      "balance_loss_mlp": 1.01854253,
      "epoch": 0.5228919284533293,
      "flos": 15596184476160.0,
      "grad_norm": 1.5287826612410875,
      "language_loss": 0.71477211,
      "learning_rate": 1.948791385766319e-06,
      "loss": 0.73597383,
      "num_input_tokens_seen": 186825900,
      "step": 8697,
      "time_per_iteration": 2.5421485900878906
    },
    {
      "auxiliary_loss_clip": 0.01085407,
      "auxiliary_loss_mlp": 0.0103197,
      "balance_loss_clip": 1.04171693,
      "balance_loss_mlp": 1.01970959,
      "epoch": 0.5229520517059973,
      "flos": 22491499726080.0,
      "grad_norm": 1.804976649049219,
      "language_loss": 0.81010628,
      "learning_rate": 1.948402052740906e-06,
      "loss": 0.83127999,
      "num_input_tokens_seen": 186843735,
      "step": 8698,
      "time_per_iteration": 2.515723466873169
    },
    {
      "auxiliary_loss_clip": 0.01101743,
      "auxiliary_loss_mlp": 0.01036162,
      "balance_loss_clip": 1.03705716,
      "balance_loss_mlp": 1.02301955,
      "epoch": 0.5230121749586653,
      "flos": 22090880361600.0,
      "grad_norm": 1.6403860049947263,
      "language_loss": 0.74162734,
      "learning_rate": 1.948012721672093e-06,
      "loss": 0.76300633,
      "num_input_tokens_seen": 186862440,
      "step": 8699,
      "time_per_iteration": 2.4957876205444336
    },
    {
      "auxiliary_loss_clip": 0.01103536,
      "auxiliary_loss_mlp": 0.00788205,
      "balance_loss_clip": 1.03603351,
      "balance_loss_mlp": 1.0092628,
      "epoch": 0.5230722982113333,
      "flos": 22127150119680.0,
      "grad_norm": 1.844928870266146,
      "language_loss": 0.7343812,
      "learning_rate": 1.947623392574642e-06,
      "loss": 0.75329852,
      "num_input_tokens_seen": 186880940,
      "step": 8700,
      "time_per_iteration": 2.491260528564453
    },
    {
      "auxiliary_loss_clip": 0.01090994,
      "auxiliary_loss_mlp": 0.01037795,
      "balance_loss_clip": 1.03999245,
      "balance_loss_mlp": 1.02354348,
      "epoch": 0.5231324214640012,
      "flos": 25009268572800.0,
      "grad_norm": 1.70911873312012,
      "language_loss": 0.6664986,
      "learning_rate": 1.947234065463318e-06,
      "loss": 0.68778646,
      "num_input_tokens_seen": 186900785,
      "step": 8701,
      "time_per_iteration": 2.570749282836914
    },
    {
      "auxiliary_loss_clip": 0.0109557,
      "auxiliary_loss_mlp": 0.00788418,
      "balance_loss_clip": 1.03925312,
      "balance_loss_mlp": 1.01158309,
      "epoch": 0.5231925447166692,
      "flos": 25740517651200.0,
      "grad_norm": 2.007791589512374,
      "language_loss": 0.66690004,
      "learning_rate": 1.9468447403528826e-06,
      "loss": 0.68573987,
      "num_input_tokens_seen": 186920895,
      "step": 8702,
      "time_per_iteration": 2.542449474334717
    },
    {
      "auxiliary_loss_clip": 0.01092835,
      "auxiliary_loss_mlp": 0.01035997,
      "balance_loss_clip": 1.03876638,
      "balance_loss_mlp": 1.02218091,
      "epoch": 0.5232526679693371,
      "flos": 21433930565760.0,
      "grad_norm": 1.8321740174622712,
      "language_loss": 0.76433766,
      "learning_rate": 1.946455417258101e-06,
      "loss": 0.78562593,
      "num_input_tokens_seen": 186940605,
      "step": 8703,
      "time_per_iteration": 2.5341320037841797
    },
    {
      "auxiliary_loss_clip": 0.01107759,
      "auxiliary_loss_mlp": 0.01042035,
      "balance_loss_clip": 1.03887486,
      "balance_loss_mlp": 1.02629948,
      "epoch": 0.5233127912220051,
      "flos": 35298393471360.0,
      "grad_norm": 2.0931583108628766,
      "language_loss": 0.77200091,
      "learning_rate": 1.9460660961937348e-06,
      "loss": 0.79349887,
      "num_input_tokens_seen": 186960820,
      "step": 8704,
      "time_per_iteration": 2.5919010639190674
    },
    {
      "auxiliary_loss_clip": 0.01093082,
      "auxiliary_loss_mlp": 0.01041053,
      "balance_loss_clip": 1.04044926,
      "balance_loss_mlp": 1.02816629,
      "epoch": 0.523372914474673,
      "flos": 17051320344960.0,
      "grad_norm": 1.7078220883326165,
      "language_loss": 0.78048301,
      "learning_rate": 1.9456767771745474e-06,
      "loss": 0.80182433,
      "num_input_tokens_seen": 186976240,
      "step": 8705,
      "time_per_iteration": 2.4931838512420654
    },
    {
      "auxiliary_loss_clip": 0.01096726,
      "auxiliary_loss_mlp": 0.01037097,
      "balance_loss_clip": 1.0391897,
      "balance_loss_mlp": 1.02304196,
      "epoch": 0.5234330377273411,
      "flos": 18406302117120.0,
      "grad_norm": 2.3698762430049376,
      "language_loss": 0.69435143,
      "learning_rate": 1.9452874602153027e-06,
      "loss": 0.71568966,
      "num_input_tokens_seen": 186992855,
      "step": 8706,
      "time_per_iteration": 2.4978771209716797
    },
    {
      "auxiliary_loss_clip": 0.01036206,
      "auxiliary_loss_mlp": 0.01001807,
      "balance_loss_clip": 1.0198164,
      "balance_loss_mlp": 1.00045955,
      "epoch": 0.523493160980009,
      "flos": 65850296970240.0,
      "grad_norm": 0.6736277615222989,
      "language_loss": 0.52423972,
      "learning_rate": 1.9448981453307623e-06,
      "loss": 0.54461986,
      "num_input_tokens_seen": 187051205,
      "step": 8707,
      "time_per_iteration": 3.1839168071746826
    },
    {
      "auxiliary_loss_clip": 0.01091507,
      "auxiliary_loss_mlp": 0.01039825,
      "balance_loss_clip": 1.0377363,
      "balance_loss_mlp": 1.02686119,
      "epoch": 0.523553284232677,
      "flos": 21872076664320.0,
      "grad_norm": 1.7655117076606108,
      "language_loss": 0.74599135,
      "learning_rate": 1.9445088325356904e-06,
      "loss": 0.76730466,
      "num_input_tokens_seen": 187070540,
      "step": 8708,
      "time_per_iteration": 2.522873640060425
    },
    {
      "auxiliary_loss_clip": 0.01089592,
      "auxiliary_loss_mlp": 0.01029833,
      "balance_loss_clip": 1.04245305,
      "balance_loss_mlp": 1.01689911,
      "epoch": 0.5236134074853449,
      "flos": 20848191482880.0,
      "grad_norm": 1.5699490454702734,
      "language_loss": 0.77213848,
      "learning_rate": 1.944119521844849e-06,
      "loss": 0.7933327,
      "num_input_tokens_seen": 187089975,
      "step": 8709,
      "time_per_iteration": 2.5431578159332275
    },
    {
      "auxiliary_loss_clip": 0.01068655,
      "auxiliary_loss_mlp": 0.01042045,
      "balance_loss_clip": 1.03533757,
      "balance_loss_mlp": 1.02579033,
      "epoch": 0.5236735307380129,
      "flos": 25520421064320.0,
      "grad_norm": 1.9151045521292223,
      "language_loss": 0.83609593,
      "learning_rate": 1.9437302132730003e-06,
      "loss": 0.85720295,
      "num_input_tokens_seen": 187108775,
      "step": 8710,
      "time_per_iteration": 2.6103403568267822
    },
    {
      "auxiliary_loss_clip": 0.01088739,
      "auxiliary_loss_mlp": 0.01033079,
      "balance_loss_clip": 1.03981674,
      "balance_loss_mlp": 1.01955497,
      "epoch": 0.523733653990681,
      "flos": 23583112001280.0,
      "grad_norm": 2.5695939917736883,
      "language_loss": 0.69107461,
      "learning_rate": 1.943340906834908e-06,
      "loss": 0.71229273,
      "num_input_tokens_seen": 187128830,
      "step": 8711,
      "time_per_iteration": 2.592047691345215
    },
    {
      "auxiliary_loss_clip": 0.01102208,
      "auxiliary_loss_mlp": 0.01034814,
      "balance_loss_clip": 1.03765416,
      "balance_loss_mlp": 1.0216887,
      "epoch": 0.5237937772433489,
      "flos": 21106245767040.0,
      "grad_norm": 2.0123222747919955,
      "language_loss": 0.83267754,
      "learning_rate": 1.9429516025453345e-06,
      "loss": 0.85404783,
      "num_input_tokens_seen": 187149570,
      "step": 8712,
      "time_per_iteration": 2.488284111022949
    },
    {
      "auxiliary_loss_clip": 0.01117119,
      "auxiliary_loss_mlp": 0.01038516,
      "balance_loss_clip": 1.03876901,
      "balance_loss_mlp": 1.02411592,
      "epoch": 0.5238539004960169,
      "flos": 19172887200000.0,
      "grad_norm": 1.8564009908708676,
      "language_loss": 0.69867665,
      "learning_rate": 1.9425623004190415e-06,
      "loss": 0.72023302,
      "num_input_tokens_seen": 187170575,
      "step": 8713,
      "time_per_iteration": 2.4852495193481445
    },
    {
      "auxiliary_loss_clip": 0.01078974,
      "auxiliary_loss_mlp": 0.01041201,
      "balance_loss_clip": 1.03527677,
      "balance_loss_mlp": 1.02442789,
      "epoch": 0.5239140237486848,
      "flos": 17888218300800.0,
      "grad_norm": 2.752794153358208,
      "language_loss": 0.76825464,
      "learning_rate": 1.9421730004707925e-06,
      "loss": 0.78945637,
      "num_input_tokens_seen": 187187190,
      "step": 8714,
      "time_per_iteration": 2.5047988891601562
    },
    {
      "auxiliary_loss_clip": 0.01079565,
      "auxiliary_loss_mlp": 0.01038349,
      "balance_loss_clip": 1.03809917,
      "balance_loss_mlp": 1.02167177,
      "epoch": 0.5239741470013528,
      "flos": 17930413802880.0,
      "grad_norm": 1.8740416002762232,
      "language_loss": 0.75704336,
      "learning_rate": 1.9417837027153483e-06,
      "loss": 0.7782225,
      "num_input_tokens_seen": 187204350,
      "step": 8715,
      "time_per_iteration": 2.549891710281372
    },
    {
      "auxiliary_loss_clip": 0.01089722,
      "auxiliary_loss_mlp": 0.01029783,
      "balance_loss_clip": 1.03718758,
      "balance_loss_mlp": 1.0165385,
      "epoch": 0.5240342702540207,
      "flos": 30993386584320.0,
      "grad_norm": 1.5661019729460361,
      "language_loss": 0.71173388,
      "learning_rate": 1.9413944071674723e-06,
      "loss": 0.73292887,
      "num_input_tokens_seen": 187225605,
      "step": 8716,
      "time_per_iteration": 2.587007522583008
    },
    {
      "auxiliary_loss_clip": 0.01114639,
      "auxiliary_loss_mlp": 0.01036077,
      "balance_loss_clip": 1.03913188,
      "balance_loss_mlp": 1.02388155,
      "epoch": 0.5240943935066887,
      "flos": 25005066681600.0,
      "grad_norm": 2.3391550836856854,
      "language_loss": 0.86920726,
      "learning_rate": 1.941005113841926e-06,
      "loss": 0.89071441,
      "num_input_tokens_seen": 187241335,
      "step": 8717,
      "time_per_iteration": 2.4550609588623047
    },
    {
      "auxiliary_loss_clip": 0.01101924,
      "auxiliary_loss_mlp": 0.01034714,
      "balance_loss_clip": 1.04214692,
      "balance_loss_mlp": 1.02126169,
      "epoch": 0.5241545167593566,
      "flos": 23659099223040.0,
      "grad_norm": 1.9312547466147425,
      "language_loss": 0.61441904,
      "learning_rate": 1.9406158227534723e-06,
      "loss": 0.63578546,
      "num_input_tokens_seen": 187259925,
      "step": 8718,
      "time_per_iteration": 2.513948917388916
    },
    {
      "auxiliary_loss_clip": 0.01089298,
      "auxiliary_loss_mlp": 0.01036237,
      "balance_loss_clip": 1.04176033,
      "balance_loss_mlp": 1.02199745,
      "epoch": 0.5242146400120247,
      "flos": 23400398494080.0,
      "grad_norm": 1.6803650926277252,
      "language_loss": 0.71983242,
      "learning_rate": 1.940226533916872e-06,
      "loss": 0.74108779,
      "num_input_tokens_seen": 187279035,
      "step": 8719,
      "time_per_iteration": 2.560619354248047
    },
    {
      "auxiliary_loss_clip": 0.01098694,
      "auxiliary_loss_mlp": 0.01027124,
      "balance_loss_clip": 1.03732991,
      "balance_loss_mlp": 1.01532853,
      "epoch": 0.5242747632646926,
      "flos": 17749065012480.0,
      "grad_norm": 2.1605978537716424,
      "language_loss": 0.73222649,
      "learning_rate": 1.9398372473468877e-06,
      "loss": 0.75348473,
      "num_input_tokens_seen": 187297555,
      "step": 8720,
      "time_per_iteration": 2.4765515327453613
    },
    {
      "auxiliary_loss_clip": 0.01102629,
      "auxiliary_loss_mlp": 0.01032963,
      "balance_loss_clip": 1.03714252,
      "balance_loss_mlp": 1.01958764,
      "epoch": 0.5243348865173606,
      "flos": 32597731549440.0,
      "grad_norm": 1.6013333602451,
      "language_loss": 0.70355785,
      "learning_rate": 1.939447963058281e-06,
      "loss": 0.72491378,
      "num_input_tokens_seen": 187320265,
      "step": 8721,
      "time_per_iteration": 2.5593059062957764
    },
    {
      "auxiliary_loss_clip": 0.0105838,
      "auxiliary_loss_mlp": 0.0103393,
      "balance_loss_clip": 1.03512335,
      "balance_loss_mlp": 1.02016115,
      "epoch": 0.5243950097700285,
      "flos": 25484115392640.0,
      "grad_norm": 1.754915162271086,
      "language_loss": 0.86436117,
      "learning_rate": 1.939058681065813e-06,
      "loss": 0.8852843,
      "num_input_tokens_seen": 187338045,
      "step": 8722,
      "time_per_iteration": 2.58982253074646
    },
    {
      "auxiliary_loss_clip": 0.01112868,
      "auxiliary_loss_mlp": 0.0102969,
      "balance_loss_clip": 1.03897548,
      "balance_loss_mlp": 1.01555729,
      "epoch": 0.5244551330226965,
      "flos": 15268391936640.0,
      "grad_norm": 1.7893624686833824,
      "language_loss": 0.79821312,
      "learning_rate": 1.938669401384247e-06,
      "loss": 0.81963867,
      "num_input_tokens_seen": 187356040,
      "step": 8723,
      "time_per_iteration": 2.4377899169921875
    },
    {
      "auxiliary_loss_clip": 0.01105652,
      "auxiliary_loss_mlp": 0.01041454,
      "balance_loss_clip": 1.04452538,
      "balance_loss_mlp": 1.02679682,
      "epoch": 0.5245152562753645,
      "flos": 22237108629120.0,
      "grad_norm": 2.304215940609295,
      "language_loss": 0.74813628,
      "learning_rate": 1.9382801240283426e-06,
      "loss": 0.76960731,
      "num_input_tokens_seen": 187374185,
      "step": 8724,
      "time_per_iteration": 3.865424394607544
    },
    {
      "auxiliary_loss_clip": 0.01119721,
      "auxiliary_loss_mlp": 0.01033141,
      "balance_loss_clip": 1.03965557,
      "balance_loss_mlp": 1.01779842,
      "epoch": 0.5245753795280325,
      "flos": 29426460612480.0,
      "grad_norm": 1.8266430287742712,
      "language_loss": 0.7056551,
      "learning_rate": 1.9378908490128625e-06,
      "loss": 0.7271837,
      "num_input_tokens_seen": 187396640,
      "step": 8725,
      "time_per_iteration": 2.5272607803344727
    },
    {
      "auxiliary_loss_clip": 0.01021145,
      "auxiliary_loss_mlp": 0.01013254,
      "balance_loss_clip": 1.02527869,
      "balance_loss_mlp": 1.01144171,
      "epoch": 0.5246355027807005,
      "flos": 58834392785280.0,
      "grad_norm": 0.7587613476094566,
      "language_loss": 0.55685878,
      "learning_rate": 1.937501576352568e-06,
      "loss": 0.5772028,
      "num_input_tokens_seen": 187455945,
      "step": 8726,
      "time_per_iteration": 3.165379762649536
    },
    {
      "auxiliary_loss_clip": 0.0102999,
      "auxiliary_loss_mlp": 0.01008475,
      "balance_loss_clip": 1.031129,
      "balance_loss_mlp": 1.00641251,
      "epoch": 0.5246956260333684,
      "flos": 64526592965760.0,
      "grad_norm": 0.8004615717218573,
      "language_loss": 0.58393478,
      "learning_rate": 1.937112306062219e-06,
      "loss": 0.60431945,
      "num_input_tokens_seen": 187519975,
      "step": 8727,
      "time_per_iteration": 3.1170053482055664
    },
    {
      "auxiliary_loss_clip": 0.01108464,
      "auxiliary_loss_mlp": 0.01031549,
      "balance_loss_clip": 1.0397141,
      "balance_loss_mlp": 1.01727414,
      "epoch": 0.5247557492860364,
      "flos": 24533631653760.0,
      "grad_norm": 1.6500189483872674,
      "language_loss": 0.70486724,
      "learning_rate": 1.9367230381565786e-06,
      "loss": 0.72626734,
      "num_input_tokens_seen": 187541775,
      "step": 8728,
      "time_per_iteration": 3.8922770023345947
    },
    {
      "auxiliary_loss_clip": 0.01102032,
      "auxiliary_loss_mlp": 0.01028116,
      "balance_loss_clip": 1.03779984,
      "balance_loss_mlp": 1.01569438,
      "epoch": 0.5248158725387043,
      "flos": 18806131382400.0,
      "grad_norm": 1.472238882268332,
      "language_loss": 0.69734502,
      "learning_rate": 1.9363337726504062e-06,
      "loss": 0.71864647,
      "num_input_tokens_seen": 187560425,
      "step": 8729,
      "time_per_iteration": 2.5030877590179443
    },
    {
      "auxiliary_loss_clip": 0.01078935,
      "auxiliary_loss_mlp": 0.01031255,
      "balance_loss_clip": 1.04031134,
      "balance_loss_mlp": 1.01780224,
      "epoch": 0.5248759957913723,
      "flos": 20955851521920.0,
      "grad_norm": 1.773979241842481,
      "language_loss": 0.83741182,
      "learning_rate": 1.935944509558464e-06,
      "loss": 0.85851371,
      "num_input_tokens_seen": 187579930,
      "step": 8730,
      "time_per_iteration": 2.5479469299316406
    },
    {
      "auxiliary_loss_clip": 0.0107961,
      "auxiliary_loss_mlp": 0.01032536,
      "balance_loss_clip": 1.03739667,
      "balance_loss_mlp": 1.01914883,
      "epoch": 0.5249361190440403,
      "flos": 18660980522880.0,
      "grad_norm": 15.874104710957344,
      "language_loss": 0.79655683,
      "learning_rate": 1.9355552488955125e-06,
      "loss": 0.81767821,
      "num_input_tokens_seen": 187595365,
      "step": 8731,
      "time_per_iteration": 3.8872251510620117
    },
    {
      "auxiliary_loss_clip": 0.01097067,
      "auxiliary_loss_mlp": 0.01029303,
      "balance_loss_clip": 1.03607988,
      "balance_loss_mlp": 1.01585007,
      "epoch": 0.5249962422967083,
      "flos": 24863327614080.0,
      "grad_norm": 1.679343717844904,
      "language_loss": 0.83085221,
      "learning_rate": 1.935165990676312e-06,
      "loss": 0.85211599,
      "num_input_tokens_seen": 187614715,
      "step": 8732,
      "time_per_iteration": 2.5287296772003174
    },
    {
      "auxiliary_loss_clip": 0.0110179,
      "auxiliary_loss_mlp": 0.01033179,
      "balance_loss_clip": 1.03795397,
      "balance_loss_mlp": 1.02053738,
      "epoch": 0.5250563655493762,
      "flos": 15262681674240.0,
      "grad_norm": 1.6752122474050988,
      "language_loss": 0.77650261,
      "learning_rate": 1.9347767349156237e-06,
      "loss": 0.79785228,
      "num_input_tokens_seen": 187630745,
      "step": 8733,
      "time_per_iteration": 2.432487964630127
    },
    {
      "auxiliary_loss_clip": 0.01118262,
      "auxiliary_loss_mlp": 0.01038246,
      "balance_loss_clip": 1.04037309,
      "balance_loss_mlp": 1.02437568,
      "epoch": 0.5251164888020442,
      "flos": 18625177641600.0,
      "grad_norm": 2.4242306782419245,
      "language_loss": 0.81816769,
      "learning_rate": 1.934387481628208e-06,
      "loss": 0.83973283,
      "num_input_tokens_seen": 187648200,
      "step": 8734,
      "time_per_iteration": 2.428380250930786
    },
    {
      "auxiliary_loss_clip": 0.01086979,
      "auxiliary_loss_mlp": 0.01031016,
      "balance_loss_clip": 1.04016435,
      "balance_loss_mlp": 1.01767647,
      "epoch": 0.5251766120547121,
      "flos": 29710764760320.0,
      "grad_norm": 1.431980013944061,
      "language_loss": 0.76728791,
      "learning_rate": 1.933998230828826e-06,
      "loss": 0.78846788,
      "num_input_tokens_seen": 187669205,
      "step": 8735,
      "time_per_iteration": 3.9099881649017334
    },
    {
      "auxiliary_loss_clip": 0.01103534,
      "auxiliary_loss_mlp": 0.0103034,
      "balance_loss_clip": 1.03952146,
      "balance_loss_mlp": 1.01832414,
      "epoch": 0.5252367353073801,
      "flos": 23440295525760.0,
      "grad_norm": 1.90823147654049,
      "language_loss": 0.80534697,
      "learning_rate": 1.9336089825322376e-06,
      "loss": 0.82668573,
      "num_input_tokens_seen": 187690890,
      "step": 8736,
      "time_per_iteration": 2.521735191345215
    },
    {
      "auxiliary_loss_clip": 0.01116154,
      "auxiliary_loss_mlp": 0.01035315,
      "balance_loss_clip": 1.03986359,
      "balance_loss_mlp": 1.02193356,
      "epoch": 0.5252968585600482,
      "flos": 30810708990720.0,
      "grad_norm": 2.39095008799196,
      "language_loss": 0.69893038,
      "learning_rate": 1.9332197367532033e-06,
      "loss": 0.72044516,
      "num_input_tokens_seen": 187713045,
      "step": 8737,
      "time_per_iteration": 2.529759407043457
    },
    {
      "auxiliary_loss_clip": 0.01091121,
      "auxiliary_loss_mlp": 0.01031976,
      "balance_loss_clip": 1.03723693,
      "balance_loss_mlp": 1.01913714,
      "epoch": 0.5253569818127161,
      "flos": 20628274464000.0,
      "grad_norm": 1.54763237067798,
      "language_loss": 0.7729758,
      "learning_rate": 1.9328304935064833e-06,
      "loss": 0.79420674,
      "num_input_tokens_seen": 187733640,
      "step": 8738,
      "time_per_iteration": 2.5520503520965576
    },
    {
      "auxiliary_loss_clip": 0.01025362,
      "auxiliary_loss_mlp": 0.00775365,
      "balance_loss_clip": 1.01867914,
      "balance_loss_mlp": 1.01960647,
      "epoch": 0.5254171050653841,
      "flos": 63428695810560.0,
      "grad_norm": 0.7453024967392763,
      "language_loss": 0.54465032,
      "learning_rate": 1.932441252806837e-06,
      "loss": 0.56265759,
      "num_input_tokens_seen": 187792930,
      "step": 8739,
      "time_per_iteration": 3.113673448562622
    },
    {
      "auxiliary_loss_clip": 0.0108667,
      "auxiliary_loss_mlp": 0.0103158,
      "balance_loss_clip": 1.03758585,
      "balance_loss_mlp": 1.01904488,
      "epoch": 0.525477228318052,
      "flos": 34670782108800.0,
      "grad_norm": 1.6506604187133807,
      "language_loss": 0.84365582,
      "learning_rate": 1.9320520146690263e-06,
      "loss": 0.8648383,
      "num_input_tokens_seen": 187812495,
      "step": 8740,
      "time_per_iteration": 2.629868745803833
    },
    {
      "auxiliary_loss_clip": 0.01098376,
      "auxiliary_loss_mlp": 0.00789305,
      "balance_loss_clip": 1.03589976,
      "balance_loss_mlp": 1.01393938,
      "epoch": 0.52553735157072,
      "flos": 17930844766080.0,
      "grad_norm": 1.9490823115248268,
      "language_loss": 0.69834214,
      "learning_rate": 1.9316627791078093e-06,
      "loss": 0.71721888,
      "num_input_tokens_seen": 187829685,
      "step": 8741,
      "time_per_iteration": 2.4583370685577393
    },
    {
      "auxiliary_loss_clip": 0.01094397,
      "auxiliary_loss_mlp": 0.01032928,
      "balance_loss_clip": 1.03901696,
      "balance_loss_mlp": 1.01961875,
      "epoch": 0.5255974748233879,
      "flos": 9940864584960.0,
      "grad_norm": 1.8550474595218813,
      "language_loss": 0.66424233,
      "learning_rate": 1.931273546137947e-06,
      "loss": 0.68551564,
      "num_input_tokens_seen": 187846495,
      "step": 8742,
      "time_per_iteration": 2.539607048034668
    },
    {
      "auxiliary_loss_clip": 0.01080288,
      "auxiliary_loss_mlp": 0.01040521,
      "balance_loss_clip": 1.03840637,
      "balance_loss_mlp": 1.02488041,
      "epoch": 0.5256575980760559,
      "flos": 16868427269760.0,
      "grad_norm": 2.2847074711141926,
      "language_loss": 0.6357438,
      "learning_rate": 1.9308843157741983e-06,
      "loss": 0.6569519,
      "num_input_tokens_seen": 187862010,
      "step": 8743,
      "time_per_iteration": 2.506277084350586
    },
    {
      "auxiliary_loss_clip": 0.0103348,
      "auxiliary_loss_mlp": 0.01006182,
      "balance_loss_clip": 1.01725364,
      "balance_loss_mlp": 1.00473964,
      "epoch": 0.5257177213287239,
      "flos": 62386210362240.0,
      "grad_norm": 0.7854302653966595,
      "language_loss": 0.54115582,
      "learning_rate": 1.930495088031323e-06,
      "loss": 0.56155246,
      "num_input_tokens_seen": 187922730,
      "step": 8744,
      "time_per_iteration": 3.2051408290863037
    },
    {
      "auxiliary_loss_clip": 0.01098493,
      "auxiliary_loss_mlp": 0.01034768,
      "balance_loss_clip": 1.04423082,
      "balance_loss_mlp": 1.02048063,
      "epoch": 0.5257778445813919,
      "flos": 20776908942720.0,
      "grad_norm": 2.4300920843025735,
      "language_loss": 0.75861394,
      "learning_rate": 1.9301058629240814e-06,
      "loss": 0.77994657,
      "num_input_tokens_seen": 187940160,
      "step": 8745,
      "time_per_iteration": 2.525797128677368
    },
    {
      "auxiliary_loss_clip": 0.01102098,
      "auxiliary_loss_mlp": 0.01033682,
      "balance_loss_clip": 1.03777289,
      "balance_loss_mlp": 1.02114165,
      "epoch": 0.5258379678340598,
      "flos": 17018606033280.0,
      "grad_norm": 1.8451093705348274,
      "language_loss": 0.81816769,
      "learning_rate": 1.9297166404672324e-06,
      "loss": 0.83952546,
      "num_input_tokens_seen": 187958625,
      "step": 8746,
      "time_per_iteration": 2.4902639389038086
    },
    {
      "auxiliary_loss_clip": 0.01100951,
      "auxiliary_loss_mlp": 0.01032961,
      "balance_loss_clip": 1.03706956,
      "balance_loss_mlp": 1.01931751,
      "epoch": 0.5258980910867278,
      "flos": 21068754946560.0,
      "grad_norm": 1.8976963714524646,
      "language_loss": 0.7579363,
      "learning_rate": 1.9293274206755353e-06,
      "loss": 0.77927542,
      "num_input_tokens_seen": 187977575,
      "step": 8747,
      "time_per_iteration": 2.5029735565185547
    },
    {
      "auxiliary_loss_clip": 0.01048828,
      "auxiliary_loss_mlp": 0.01032271,
      "balance_loss_clip": 1.03603172,
      "balance_loss_mlp": 1.01900923,
      "epoch": 0.5259582143393957,
      "flos": 18004461690240.0,
      "grad_norm": 1.9619672488026192,
      "language_loss": 0.830715,
      "learning_rate": 1.9289382035637505e-06,
      "loss": 0.85152602,
      "num_input_tokens_seen": 187996650,
      "step": 8748,
      "time_per_iteration": 2.6143271923065186
    },
    {
      "auxiliary_loss_clip": 0.010925,
      "auxiliary_loss_mlp": 0.01032891,
      "balance_loss_clip": 1.03604746,
      "balance_loss_mlp": 1.01839495,
      "epoch": 0.5260183375920637,
      "flos": 22783848520320.0,
      "grad_norm": 1.9865028714177118,
      "language_loss": 0.80437124,
      "learning_rate": 1.9285489891466345e-06,
      "loss": 0.82562506,
      "num_input_tokens_seen": 188013510,
      "step": 8749,
      "time_per_iteration": 2.5303120613098145
    },
    {
      "auxiliary_loss_clip": 0.0110192,
      "auxiliary_loss_mlp": 0.01035656,
      "balance_loss_clip": 1.04081452,
      "balance_loss_mlp": 1.02232218,
      "epoch": 0.5260784608447318,
      "flos": 27052406081280.0,
      "grad_norm": 1.9137734935401607,
      "language_loss": 0.72589874,
      "learning_rate": 1.9281597774389487e-06,
      "loss": 0.74727452,
      "num_input_tokens_seen": 188032085,
      "step": 8750,
      "time_per_iteration": 2.5260841846466064
    },
    {
      "auxiliary_loss_clip": 0.01090016,
      "auxiliary_loss_mlp": 0.0103127,
      "balance_loss_clip": 1.03455818,
      "balance_loss_mlp": 1.01832414,
      "epoch": 0.5261385840973997,
      "flos": 20662820369280.0,
      "grad_norm": 1.513532697986344,
      "language_loss": 0.76519632,
      "learning_rate": 1.9277705684554517e-06,
      "loss": 0.7864092,
      "num_input_tokens_seen": 188050590,
      "step": 8751,
      "time_per_iteration": 2.5140843391418457
    },
    {
      "auxiliary_loss_clip": 0.01110582,
      "auxiliary_loss_mlp": 0.01037628,
      "balance_loss_clip": 1.03830051,
      "balance_loss_mlp": 1.02496815,
      "epoch": 0.5261987073500677,
      "flos": 23622649896960.0,
      "grad_norm": 1.37578034543914,
      "language_loss": 0.75937521,
      "learning_rate": 1.927381362210902e-06,
      "loss": 0.78085732,
      "num_input_tokens_seen": 188071620,
      "step": 8752,
      "time_per_iteration": 2.473507881164551
    },
    {
      "auxiliary_loss_clip": 0.01104805,
      "auxiliary_loss_mlp": 0.0103585,
      "balance_loss_clip": 1.03832579,
      "balance_loss_mlp": 1.02160466,
      "epoch": 0.5262588306027356,
      "flos": 27636241743360.0,
      "grad_norm": 1.4313284955959715,
      "language_loss": 0.67802441,
      "learning_rate": 1.926992158720058e-06,
      "loss": 0.69943094,
      "num_input_tokens_seen": 188091740,
      "step": 8753,
      "time_per_iteration": 2.5171830654144287
    },
    {
      "auxiliary_loss_clip": 0.01102388,
      "auxiliary_loss_mlp": 0.01034519,
      "balance_loss_clip": 1.03840876,
      "balance_loss_mlp": 1.02174544,
      "epoch": 0.5263189538554036,
      "flos": 21759711943680.0,
      "grad_norm": 1.4857943818650503,
      "language_loss": 0.83956176,
      "learning_rate": 1.9266029579976785e-06,
      "loss": 0.8609308,
      "num_input_tokens_seen": 188111165,
      "step": 8754,
      "time_per_iteration": 2.4751389026641846
    },
    {
      "auxiliary_loss_clip": 0.0110078,
      "auxiliary_loss_mlp": 0.01032398,
      "balance_loss_clip": 1.03789735,
      "balance_loss_mlp": 1.01862383,
      "epoch": 0.5263790771080715,
      "flos": 14276359140480.0,
      "grad_norm": 2.091364656377627,
      "language_loss": 0.87477779,
      "learning_rate": 1.926213760058522e-06,
      "loss": 0.8961097,
      "num_input_tokens_seen": 188127825,
      "step": 8755,
      "time_per_iteration": 2.4508447647094727
    },
    {
      "auxiliary_loss_clip": 0.01015278,
      "auxiliary_loss_mlp": 0.01016607,
      "balance_loss_clip": 1.02509058,
      "balance_loss_mlp": 1.01528955,
      "epoch": 0.5264392003607395,
      "flos": 65806413528960.0,
      "grad_norm": 0.7252144031363951,
      "language_loss": 0.58847034,
      "learning_rate": 1.9258245649173477e-06,
      "loss": 0.60878921,
      "num_input_tokens_seen": 188194050,
      "step": 8756,
      "time_per_iteration": 3.2170541286468506
    },
    {
      "auxiliary_loss_clip": 0.01083571,
      "auxiliary_loss_mlp": 0.0102772,
      "balance_loss_clip": 1.03610897,
      "balance_loss_mlp": 1.01441038,
      "epoch": 0.5264993236134075,
      "flos": 21032413361280.0,
      "grad_norm": 1.6580312490106202,
      "language_loss": 0.7033093,
      "learning_rate": 1.925435372588913e-06,
      "loss": 0.72442222,
      "num_input_tokens_seen": 188212565,
      "step": 8757,
      "time_per_iteration": 2.525977373123169
    },
    {
      "auxiliary_loss_clip": 0.01101005,
      "auxiliary_loss_mlp": 0.01031387,
      "balance_loss_clip": 1.03686535,
      "balance_loss_mlp": 1.01815498,
      "epoch": 0.5265594468660755,
      "flos": 16618202150400.0,
      "grad_norm": 2.1628097329276317,
      "language_loss": 0.87727249,
      "learning_rate": 1.9250461830879768e-06,
      "loss": 0.89859641,
      "num_input_tokens_seen": 188229505,
      "step": 8758,
      "time_per_iteration": 2.447530508041382
    },
    {
      "auxiliary_loss_clip": 0.0105304,
      "auxiliary_loss_mlp": 0.010353,
      "balance_loss_clip": 1.03562498,
      "balance_loss_mlp": 1.02101827,
      "epoch": 0.5266195701187434,
      "flos": 24134125610880.0,
      "grad_norm": 1.558003131295313,
      "language_loss": 0.75944114,
      "learning_rate": 1.9246569964292965e-06,
      "loss": 0.78032458,
      "num_input_tokens_seen": 188250395,
      "step": 8759,
      "time_per_iteration": 2.6150898933410645
    },
    {
      "auxiliary_loss_clip": 0.01088366,
      "auxiliary_loss_mlp": 0.01025233,
      "balance_loss_clip": 1.03640616,
      "balance_loss_mlp": 1.01219761,
      "epoch": 0.5266796933714114,
      "flos": 15844111125120.0,
      "grad_norm": 2.20014332190006,
      "language_loss": 0.71758115,
      "learning_rate": 1.9242678126276307e-06,
      "loss": 0.73871708,
      "num_input_tokens_seen": 188266785,
      "step": 8760,
      "time_per_iteration": 2.4882659912109375
    },
    {
      "auxiliary_loss_clip": 0.01093168,
      "auxiliary_loss_mlp": 0.01032629,
      "balance_loss_clip": 1.04041111,
      "balance_loss_mlp": 1.0185864,
      "epoch": 0.5267398166240793,
      "flos": 20951434149120.0,
      "grad_norm": 2.5770813794599916,
      "language_loss": 0.75720441,
      "learning_rate": 1.923878631697736e-06,
      "loss": 0.77846241,
      "num_input_tokens_seen": 188282525,
      "step": 8761,
      "time_per_iteration": 2.5288515090942383
    },
    {
      "auxiliary_loss_clip": 0.01097739,
      "auxiliary_loss_mlp": 0.00799474,
      "balance_loss_clip": 1.0361172,
      "balance_loss_mlp": 1.03104079,
      "epoch": 0.5267999398767473,
      "flos": 20996394998400.0,
      "grad_norm": 1.6357661820363198,
      "language_loss": 0.71001863,
      "learning_rate": 1.923489453654373e-06,
      "loss": 0.72899067,
      "num_input_tokens_seen": 188301395,
      "step": 8762,
      "time_per_iteration": 2.4837486743927
    },
    {
      "auxiliary_loss_clip": 0.01020802,
      "auxiliary_loss_mlp": 0.01003086,
      "balance_loss_clip": 1.02040768,
      "balance_loss_mlp": 1.00191772,
      "epoch": 0.5268600631294152,
      "flos": 66849401767680.0,
      "grad_norm": 0.9282053988036777,
      "language_loss": 0.65442872,
      "learning_rate": 1.9231002785122963e-06,
      "loss": 0.6746676,
      "num_input_tokens_seen": 188357665,
      "step": 8763,
      "time_per_iteration": 4.388975143432617
    },
    {
      "auxiliary_loss_clip": 0.01099311,
      "auxiliary_loss_mlp": 0.01025882,
      "balance_loss_clip": 1.0356431,
      "balance_loss_mlp": 1.01282275,
      "epoch": 0.5269201863820833,
      "flos": 17165552572800.0,
      "grad_norm": 2.328471188473617,
      "language_loss": 0.7106244,
      "learning_rate": 1.922711106286265e-06,
      "loss": 0.73187625,
      "num_input_tokens_seen": 188376935,
      "step": 8764,
      "time_per_iteration": 2.4630911350250244
    },
    {
      "auxiliary_loss_clip": 0.01072168,
      "auxiliary_loss_mlp": 0.01033618,
      "balance_loss_clip": 1.0352385,
      "balance_loss_mlp": 1.01860321,
      "epoch": 0.5269803096347513,
      "flos": 20522589672960.0,
      "grad_norm": 1.6848563622492858,
      "language_loss": 0.74117392,
      "learning_rate": 1.9223219369910368e-06,
      "loss": 0.76223177,
      "num_input_tokens_seen": 188394995,
      "step": 8765,
      "time_per_iteration": 2.530003070831299
    },
    {
      "auxiliary_loss_clip": 0.01090848,
      "auxiliary_loss_mlp": 0.01031658,
      "balance_loss_clip": 1.03428078,
      "balance_loss_mlp": 1.0173645,
      "epoch": 0.5270404328874192,
      "flos": 27230989524480.0,
      "grad_norm": 1.4736679451753976,
      "language_loss": 0.85367787,
      "learning_rate": 1.9219327706413677e-06,
      "loss": 0.87490296,
      "num_input_tokens_seen": 188415475,
      "step": 8766,
      "time_per_iteration": 3.929028272628784
    },
    {
      "auxiliary_loss_clip": 0.01116803,
      "auxiliary_loss_mlp": 0.01034115,
      "balance_loss_clip": 1.04054379,
      "balance_loss_mlp": 1.01983964,
      "epoch": 0.5271005561400872,
      "flos": 23110491824640.0,
      "grad_norm": 2.1549988913092943,
      "language_loss": 0.79211247,
      "learning_rate": 1.921543607252017e-06,
      "loss": 0.81362164,
      "num_input_tokens_seen": 188435665,
      "step": 8767,
      "time_per_iteration": 2.4735889434814453
    },
    {
      "auxiliary_loss_clip": 0.01104873,
      "auxiliary_loss_mlp": 0.01031467,
      "balance_loss_clip": 1.03779876,
      "balance_loss_mlp": 1.01709056,
      "epoch": 0.5271606793927551,
      "flos": 22564793427840.0,
      "grad_norm": 1.8582806303516148,
      "language_loss": 0.73821193,
      "learning_rate": 1.9211544468377394e-06,
      "loss": 0.75957531,
      "num_input_tokens_seen": 188455405,
      "step": 8768,
      "time_per_iteration": 2.5003201961517334
    },
    {
      "auxiliary_loss_clip": 0.01085458,
      "auxiliary_loss_mlp": 0.01041004,
      "balance_loss_clip": 1.0357368,
      "balance_loss_mlp": 1.02784872,
      "epoch": 0.5272208026454231,
      "flos": 18764259102720.0,
      "grad_norm": 2.2739189662658954,
      "language_loss": 0.74598145,
      "learning_rate": 1.9207652894132933e-06,
      "loss": 0.76724613,
      "num_input_tokens_seen": 188472940,
      "step": 8769,
      "time_per_iteration": 3.878206253051758
    },
    {
      "auxiliary_loss_clip": 0.01075737,
      "auxiliary_loss_mlp": 0.01030189,
      "balance_loss_clip": 1.0349617,
      "balance_loss_mlp": 1.01696324,
      "epoch": 0.5272809258980911,
      "flos": 20412164286720.0,
      "grad_norm": 1.7114294198484459,
      "language_loss": 0.74189192,
      "learning_rate": 1.920376134993436e-06,
      "loss": 0.7629512,
      "num_input_tokens_seen": 188493035,
      "step": 8770,
      "time_per_iteration": 2.53420352935791
    },
    {
      "auxiliary_loss_clip": 0.01113597,
      "auxiliary_loss_mlp": 0.01031135,
      "balance_loss_clip": 1.03863168,
      "balance_loss_mlp": 1.01783741,
      "epoch": 0.5273410491507591,
      "flos": 28256742213120.0,
      "grad_norm": 1.9888282488461249,
      "language_loss": 0.67910552,
      "learning_rate": 1.9199869835929224e-06,
      "loss": 0.70055288,
      "num_input_tokens_seen": 188513860,
      "step": 8771,
      "time_per_iteration": 2.506368637084961
    },
    {
      "auxiliary_loss_clip": 0.01099512,
      "auxiliary_loss_mlp": 0.01038363,
      "balance_loss_clip": 1.03726149,
      "balance_loss_mlp": 1.02508342,
      "epoch": 0.527401172403427,
      "flos": 22455158140800.0,
      "grad_norm": 1.8280771152763378,
      "language_loss": 0.7672298,
      "learning_rate": 1.9195978352265115e-06,
      "loss": 0.78860855,
      "num_input_tokens_seen": 188533345,
      "step": 8772,
      "time_per_iteration": 2.479888916015625
    },
    {
      "auxiliary_loss_clip": 0.01098336,
      "auxiliary_loss_mlp": 0.01039468,
      "balance_loss_clip": 1.0369966,
      "balance_loss_mlp": 1.0252223,
      "epoch": 0.527461295656095,
      "flos": 21031084558080.0,
      "grad_norm": 2.4591070126743797,
      "language_loss": 0.6626935,
      "learning_rate": 1.9192086899089585e-06,
      "loss": 0.68407154,
      "num_input_tokens_seen": 188551550,
      "step": 8773,
      "time_per_iteration": 3.850606679916382
    },
    {
      "auxiliary_loss_clip": 0.01081411,
      "auxiliary_loss_mlp": 0.01037131,
      "balance_loss_clip": 1.03733945,
      "balance_loss_mlp": 1.02506709,
      "epoch": 0.5275214189087629,
      "flos": 26322018929280.0,
      "grad_norm": 1.7402422253504257,
      "language_loss": 0.85879868,
      "learning_rate": 1.91881954765502e-06,
      "loss": 0.87998414,
      "num_input_tokens_seen": 188571615,
      "step": 8774,
      "time_per_iteration": 2.5890674591064453
    },
    {
      "auxiliary_loss_clip": 0.01081116,
      "auxiliary_loss_mlp": 0.01031287,
      "balance_loss_clip": 1.03658891,
      "balance_loss_mlp": 1.0184598,
      "epoch": 0.5275815421614309,
      "flos": 20047024581120.0,
      "grad_norm": 1.5031323413406452,
      "language_loss": 0.80095267,
      "learning_rate": 1.9184304084794523e-06,
      "loss": 0.82207668,
      "num_input_tokens_seen": 188591965,
      "step": 8775,
      "time_per_iteration": 2.5047855377197266
    },
    {
      "auxiliary_loss_clip": 0.01087304,
      "auxiliary_loss_mlp": 0.01035428,
      "balance_loss_clip": 1.03458476,
      "balance_loss_mlp": 1.02202332,
      "epoch": 0.5276416654140988,
      "flos": 21432206712960.0,
      "grad_norm": 1.7424391304097946,
      "language_loss": 0.83600855,
      "learning_rate": 1.918041272397012e-06,
      "loss": 0.85723585,
      "num_input_tokens_seen": 188610675,
      "step": 8776,
      "time_per_iteration": 2.5054194927215576
    },
    {
      "auxiliary_loss_clip": 0.01086547,
      "auxiliary_loss_mlp": 0.01030136,
      "balance_loss_clip": 1.0421344,
      "balance_loss_mlp": 1.01652265,
      "epoch": 0.5277017886667669,
      "flos": 17165085696000.0,
      "grad_norm": 1.839343081238076,
      "language_loss": 0.67329097,
      "learning_rate": 1.9176521394224547e-06,
      "loss": 0.69445777,
      "num_input_tokens_seen": 188628235,
      "step": 8777,
      "time_per_iteration": 2.48655104637146
    },
    {
      "auxiliary_loss_clip": 0.01092961,
      "auxiliary_loss_mlp": 0.01037147,
      "balance_loss_clip": 1.03936529,
      "balance_loss_mlp": 1.02461267,
      "epoch": 0.5277619119194349,
      "flos": 20448146736000.0,
      "grad_norm": 1.51519378765227,
      "language_loss": 0.82578886,
      "learning_rate": 1.9172630095705358e-06,
      "loss": 0.84708989,
      "num_input_tokens_seen": 188648925,
      "step": 8778,
      "time_per_iteration": 2.523207664489746
    },
    {
      "auxiliary_loss_clip": 0.01103771,
      "auxiliary_loss_mlp": 0.0103582,
      "balance_loss_clip": 1.03966248,
      "balance_loss_mlp": 1.0219084,
      "epoch": 0.5278220351721028,
      "flos": 24061083304320.0,
      "grad_norm": 1.9086379584411584,
      "language_loss": 0.79408193,
      "learning_rate": 1.916873882856013e-06,
      "loss": 0.81547785,
      "num_input_tokens_seen": 188668125,
      "step": 8779,
      "time_per_iteration": 2.499063014984131
    },
    {
      "auxiliary_loss_clip": 0.01096369,
      "auxiliary_loss_mlp": 0.01032062,
      "balance_loss_clip": 1.03559589,
      "balance_loss_mlp": 1.02033758,
      "epoch": 0.5278821584247708,
      "flos": 24642907804800.0,
      "grad_norm": 2.84478715134165,
      "language_loss": 0.76659805,
      "learning_rate": 1.9164847592936406e-06,
      "loss": 0.78788233,
      "num_input_tokens_seen": 188684410,
      "step": 8780,
      "time_per_iteration": 2.5104031562805176
    },
    {
      "auxiliary_loss_clip": 0.01085147,
      "auxiliary_loss_mlp": 0.01029641,
      "balance_loss_clip": 1.03968799,
      "balance_loss_mlp": 1.01595581,
      "epoch": 0.5279422816774387,
      "flos": 35408244240000.0,
      "grad_norm": 1.5504035762518522,
      "language_loss": 0.69428086,
      "learning_rate": 1.916095638898174e-06,
      "loss": 0.71542871,
      "num_input_tokens_seen": 188706130,
      "step": 8781,
      "time_per_iteration": 2.643890142440796
    },
    {
      "auxiliary_loss_clip": 0.01097896,
      "auxiliary_loss_mlp": 0.0102834,
      "balance_loss_clip": 1.03702712,
      "balance_loss_mlp": 1.01668167,
      "epoch": 0.5280024049301068,
      "flos": 22967028904320.0,
      "grad_norm": 1.597486996682263,
      "language_loss": 0.72057271,
      "learning_rate": 1.9157065216843696e-06,
      "loss": 0.74183506,
      "num_input_tokens_seen": 188725030,
      "step": 8782,
      "time_per_iteration": 2.510983467102051
    },
    {
      "auxiliary_loss_clip": 0.01086,
      "auxiliary_loss_mlp": 0.01027642,
      "balance_loss_clip": 1.03913343,
      "balance_loss_mlp": 1.01461256,
      "epoch": 0.5280625281827747,
      "flos": 21507619317120.0,
      "grad_norm": 1.8543426203779696,
      "language_loss": 0.68641186,
      "learning_rate": 1.915317407666982e-06,
      "loss": 0.70754826,
      "num_input_tokens_seen": 188744325,
      "step": 8783,
      "time_per_iteration": 2.5230393409729004
    },
    {
      "auxiliary_loss_clip": 0.01110435,
      "auxiliary_loss_mlp": 0.01040192,
      "balance_loss_clip": 1.03967094,
      "balance_loss_mlp": 1.02476656,
      "epoch": 0.5281226514354427,
      "flos": 31208167958400.0,
      "grad_norm": 1.9930674524024772,
      "language_loss": 0.69487369,
      "learning_rate": 1.9149282968607674e-06,
      "loss": 0.71637988,
      "num_input_tokens_seen": 188765100,
      "step": 8784,
      "time_per_iteration": 2.5743744373321533
    },
    {
      "auxiliary_loss_clip": 0.01116069,
      "auxiliary_loss_mlp": 0.01030107,
      "balance_loss_clip": 1.03734517,
      "balance_loss_mlp": 1.01619542,
      "epoch": 0.5281827746881106,
      "flos": 25077821679360.0,
      "grad_norm": 2.211887463883251,
      "language_loss": 0.75191069,
      "learning_rate": 1.91453918928048e-06,
      "loss": 0.77337247,
      "num_input_tokens_seen": 188783995,
      "step": 8785,
      "time_per_iteration": 2.4650795459747314
    },
    {
      "auxiliary_loss_clip": 0.01104204,
      "auxiliary_loss_mlp": 0.0103318,
      "balance_loss_clip": 1.04009986,
      "balance_loss_mlp": 1.01934588,
      "epoch": 0.5282428979407786,
      "flos": 20631255292800.0,
      "grad_norm": 1.7297466688498235,
      "language_loss": 0.83484375,
      "learning_rate": 1.9141500849408745e-06,
      "loss": 0.85621762,
      "num_input_tokens_seen": 188803120,
      "step": 8786,
      "time_per_iteration": 2.4867002964019775
    },
    {
      "auxiliary_loss_clip": 0.01073853,
      "auxiliary_loss_mlp": 0.01026898,
      "balance_loss_clip": 1.03518224,
      "balance_loss_mlp": 1.01532888,
      "epoch": 0.5283030211934465,
      "flos": 22419391173120.0,
      "grad_norm": 2.2262280595857824,
      "language_loss": 0.82892692,
      "learning_rate": 1.9137609838567076e-06,
      "loss": 0.84993446,
      "num_input_tokens_seen": 188820960,
      "step": 8787,
      "time_per_iteration": 2.5288782119750977
    },
    {
      "auxiliary_loss_clip": 0.01060071,
      "auxiliary_loss_mlp": 0.01031263,
      "balance_loss_clip": 1.03881454,
      "balance_loss_mlp": 1.01950264,
      "epoch": 0.5283631444461145,
      "flos": 23615467176960.0,
      "grad_norm": 1.8242972321961186,
      "language_loss": 0.83473521,
      "learning_rate": 1.9133718860427316e-06,
      "loss": 0.85564852,
      "num_input_tokens_seen": 188837165,
      "step": 8788,
      "time_per_iteration": 2.5809125900268555
    },
    {
      "auxiliary_loss_clip": 0.01078238,
      "auxiliary_loss_mlp": 0.01040578,
      "balance_loss_clip": 1.04147923,
      "balance_loss_mlp": 1.02698851,
      "epoch": 0.5284232676987825,
      "flos": 32671994918400.0,
      "grad_norm": 3.7437486723179845,
      "language_loss": 0.7515614,
      "learning_rate": 1.9129827915137027e-06,
      "loss": 0.7727496,
      "num_input_tokens_seen": 188858555,
      "step": 8789,
      "time_per_iteration": 2.6311721801757812
    },
    {
      "auxiliary_loss_clip": 0.01103336,
      "auxiliary_loss_mlp": 0.01033487,
      "balance_loss_clip": 1.03890955,
      "balance_loss_mlp": 1.0206542,
      "epoch": 0.5284833909514505,
      "flos": 26760919213440.0,
      "grad_norm": 1.8307351550993258,
      "language_loss": 0.69937754,
      "learning_rate": 1.9125937002843754e-06,
      "loss": 0.7207458,
      "num_input_tokens_seen": 188879050,
      "step": 8790,
      "time_per_iteration": 2.536754608154297
    },
    {
      "auxiliary_loss_clip": 0.01109516,
      "auxiliary_loss_mlp": 0.01026304,
      "balance_loss_clip": 1.03735662,
      "balance_loss_mlp": 1.01402557,
      "epoch": 0.5285435142041185,
      "flos": 22090700793600.0,
      "grad_norm": 1.5749588783401367,
      "language_loss": 0.79040545,
      "learning_rate": 1.9122046123695036e-06,
      "loss": 0.81176364,
      "num_input_tokens_seen": 188898885,
      "step": 8791,
      "time_per_iteration": 2.479295492172241
    },
    {
      "auxiliary_loss_clip": 0.01066234,
      "auxiliary_loss_mlp": 0.01027761,
      "balance_loss_clip": 1.04116201,
      "balance_loss_mlp": 1.01482677,
      "epoch": 0.5286036374567864,
      "flos": 20375463565440.0,
      "grad_norm": 2.0122397834961947,
      "language_loss": 0.66251415,
      "learning_rate": 1.9118155277838423e-06,
      "loss": 0.68345416,
      "num_input_tokens_seen": 188917225,
      "step": 8792,
      "time_per_iteration": 2.5758509635925293
    },
    {
      "auxiliary_loss_clip": 0.01083508,
      "auxiliary_loss_mlp": 0.01041334,
      "balance_loss_clip": 1.03471971,
      "balance_loss_mlp": 1.0281074,
      "epoch": 0.5286637607094544,
      "flos": 24352175122560.0,
      "grad_norm": 1.959065383644262,
      "language_loss": 0.79847378,
      "learning_rate": 1.9114264465421443e-06,
      "loss": 0.81972218,
      "num_input_tokens_seen": 188936120,
      "step": 8793,
      "time_per_iteration": 2.5378692150115967
    },
    {
      "auxiliary_loss_clip": 0.01112714,
      "auxiliary_loss_mlp": 0.01040285,
      "balance_loss_clip": 1.03875935,
      "balance_loss_mlp": 1.02697515,
      "epoch": 0.5287238839621223,
      "flos": 17271165536640.0,
      "grad_norm": 2.011786586860156,
      "language_loss": 0.85092175,
      "learning_rate": 1.9110373686591645e-06,
      "loss": 0.87245166,
      "num_input_tokens_seen": 188953405,
      "step": 8794,
      "time_per_iteration": 2.4437456130981445
    },
    {
      "auxiliary_loss_clip": 0.01093681,
      "auxiliary_loss_mlp": 0.01036395,
      "balance_loss_clip": 1.04109693,
      "balance_loss_mlp": 1.0221076,
      "epoch": 0.5287840072147904,
      "flos": 17566890209280.0,
      "grad_norm": 2.6654457031694587,
      "language_loss": 0.68050051,
      "learning_rate": 1.9106482941496564e-06,
      "loss": 0.70180124,
      "num_input_tokens_seen": 188971150,
      "step": 8795,
      "time_per_iteration": 2.4985647201538086
    },
    {
      "auxiliary_loss_clip": 0.01088587,
      "auxiliary_loss_mlp": 0.01031375,
      "balance_loss_clip": 1.03899491,
      "balance_loss_mlp": 1.01879263,
      "epoch": 0.5288441304674583,
      "flos": 18552099421440.0,
      "grad_norm": 1.8338784703551305,
      "language_loss": 0.80688709,
      "learning_rate": 1.910259223028374e-06,
      "loss": 0.82808667,
      "num_input_tokens_seen": 188989550,
      "step": 8796,
      "time_per_iteration": 2.5006144046783447
    },
    {
      "auxiliary_loss_clip": 0.0107164,
      "auxiliary_loss_mlp": 0.01043675,
      "balance_loss_clip": 1.03526008,
      "balance_loss_mlp": 1.02792752,
      "epoch": 0.5289042537201263,
      "flos": 20814507504000.0,
      "grad_norm": 1.7429985262340641,
      "language_loss": 0.69441444,
      "learning_rate": 1.909870155310071e-06,
      "loss": 0.71556759,
      "num_input_tokens_seen": 189008795,
      "step": 8797,
      "time_per_iteration": 2.530583381652832
    },
    {
      "auxiliary_loss_clip": 0.0109465,
      "auxiliary_loss_mlp": 0.01036467,
      "balance_loss_clip": 1.03958988,
      "balance_loss_mlp": 1.02408719,
      "epoch": 0.5289643769727942,
      "flos": 15735265937280.0,
      "grad_norm": 1.5172498369020773,
      "language_loss": 0.82386565,
      "learning_rate": 1.9094810910095005e-06,
      "loss": 0.84517688,
      "num_input_tokens_seen": 189025540,
      "step": 8798,
      "time_per_iteration": 2.4581429958343506
    },
    {
      "auxiliary_loss_clip": 0.01088611,
      "auxiliary_loss_mlp": 0.00798745,
      "balance_loss_clip": 1.03587651,
      "balance_loss_mlp": 1.02617455,
      "epoch": 0.5290245002254622,
      "flos": 19537308633600.0,
      "grad_norm": 1.9671604210827982,
      "language_loss": 0.70817506,
      "learning_rate": 1.9090920301414166e-06,
      "loss": 0.72704864,
      "num_input_tokens_seen": 189044885,
      "step": 8799,
      "time_per_iteration": 2.49458384513855
    },
    {
      "auxiliary_loss_clip": 0.01099038,
      "auxiliary_loss_mlp": 0.01035159,
      "balance_loss_clip": 1.03981745,
      "balance_loss_mlp": 1.02289796,
      "epoch": 0.5290846234781301,
      "flos": 15815131827840.0,
      "grad_norm": 1.9078503403709233,
      "language_loss": 0.68971741,
      "learning_rate": 1.9087029727205716e-06,
      "loss": 0.71105939,
      "num_input_tokens_seen": 189061280,
      "step": 8800,
      "time_per_iteration": 2.460244655609131
    },
    {
      "auxiliary_loss_clip": 0.0101171,
      "auxiliary_loss_mlp": 0.01008655,
      "balance_loss_clip": 1.01485229,
      "balance_loss_mlp": 1.00731945,
      "epoch": 0.5291447467307981,
      "flos": 70057624821120.0,
      "grad_norm": 1.0095888968722198,
      "language_loss": 0.57046759,
      "learning_rate": 1.9083139187617193e-06,
      "loss": 0.5906713,
      "num_input_tokens_seen": 189114775,
      "step": 8801,
      "time_per_iteration": 4.424926042556763
    },
    {
      "auxiliary_loss_clip": 0.01097415,
      "auxiliary_loss_mlp": 0.01035249,
      "balance_loss_clip": 1.04018021,
      "balance_loss_mlp": 1.02245808,
      "epoch": 0.529204869983466,
      "flos": 28364186770560.0,
      "grad_norm": 1.505397784479068,
      "language_loss": 0.63628584,
      "learning_rate": 1.9079248682796123e-06,
      "loss": 0.65761244,
      "num_input_tokens_seen": 189134700,
      "step": 8802,
      "time_per_iteration": 2.567531108856201
    },
    {
      "auxiliary_loss_clip": 0.01092525,
      "auxiliary_loss_mlp": 0.01027263,
      "balance_loss_clip": 1.04003406,
      "balance_loss_mlp": 1.01413834,
      "epoch": 0.5292649932361341,
      "flos": 33758830684800.0,
      "grad_norm": 1.5434640690065624,
      "language_loss": 0.68880737,
      "learning_rate": 1.907535821289003e-06,
      "loss": 0.71000528,
      "num_input_tokens_seen": 189155365,
      "step": 8803,
      "time_per_iteration": 2.634021043777466
    },
    {
      "auxiliary_loss_clip": 0.01096368,
      "auxiliary_loss_mlp": 0.00790468,
      "balance_loss_clip": 1.0362643,
      "balance_loss_mlp": 1.01746297,
      "epoch": 0.5293251164888021,
      "flos": 20447679859200.0,
      "grad_norm": 1.6597999254506213,
      "language_loss": 0.7643441,
      "learning_rate": 1.9071467778046458e-06,
      "loss": 0.78321242,
      "num_input_tokens_seen": 189173885,
      "step": 8804,
      "time_per_iteration": 2.4745709896087646
    },
    {
      "auxiliary_loss_clip": 0.01031976,
      "auxiliary_loss_mlp": 0.01003489,
      "balance_loss_clip": 1.01580024,
      "balance_loss_mlp": 1.00225496,
      "epoch": 0.52938523974147,
      "flos": 66545312204160.0,
      "grad_norm": 0.7526647173236996,
      "language_loss": 0.5297249,
      "learning_rate": 1.906757737841291e-06,
      "loss": 0.55007958,
      "num_input_tokens_seen": 189236515,
      "step": 8805,
      "time_per_iteration": 4.5177929401397705
    },
    {
      "auxiliary_loss_clip": 0.01030523,
      "auxiliary_loss_mlp": 0.01006132,
      "balance_loss_clip": 1.01418293,
      "balance_loss_mlp": 1.00459373,
      "epoch": 0.529445362994138,
      "flos": 67151734542720.0,
      "grad_norm": 0.7368576678339734,
      "language_loss": 0.63772273,
      "learning_rate": 1.906368701413693e-06,
      "loss": 0.65808928,
      "num_input_tokens_seen": 189300500,
      "step": 8806,
      "time_per_iteration": 3.100186586380005
    },
    {
      "auxiliary_loss_clip": 0.01104209,
      "auxiliary_loss_mlp": 0.01030991,
      "balance_loss_clip": 1.03837538,
      "balance_loss_mlp": 1.01811016,
      "epoch": 0.5295054862468059,
      "flos": 17749316407680.0,
      "grad_norm": 1.6385880970395554,
      "language_loss": 0.7227385,
      "learning_rate": 1.9059796685366026e-06,
      "loss": 0.74409056,
      "num_input_tokens_seen": 189319745,
      "step": 8807,
      "time_per_iteration": 2.466193675994873
    },
    {
      "auxiliary_loss_clip": 0.01080501,
      "auxiliary_loss_mlp": 0.01029892,
      "balance_loss_clip": 1.03908706,
      "balance_loss_mlp": 1.01819766,
      "epoch": 0.529565609499474,
      "flos": 11397401084160.0,
      "grad_norm": 2.2615758654520786,
      "language_loss": 0.69373488,
      "learning_rate": 1.9055906392247723e-06,
      "loss": 0.71483874,
      "num_input_tokens_seen": 189334550,
      "step": 8808,
      "time_per_iteration": 3.892692804336548
    },
    {
      "auxiliary_loss_clip": 0.01100043,
      "auxiliary_loss_mlp": 0.01031469,
      "balance_loss_clip": 1.03770971,
      "balance_loss_mlp": 1.01949441,
      "epoch": 0.5296257327521419,
      "flos": 17196363463680.0,
      "grad_norm": 2.026108909974527,
      "language_loss": 0.86613929,
      "learning_rate": 1.9052016134929554e-06,
      "loss": 0.88745439,
      "num_input_tokens_seen": 189351735,
      "step": 8809,
      "time_per_iteration": 2.4426333904266357
    },
    {
      "auxiliary_loss_clip": 0.01105817,
      "auxiliary_loss_mlp": 0.01032646,
      "balance_loss_clip": 1.03905237,
      "balance_loss_mlp": 1.01865697,
      "epoch": 0.5296858560048099,
      "flos": 39964086777600.0,
      "grad_norm": 1.8689752856074047,
      "language_loss": 0.64191604,
      "learning_rate": 1.9048125913559016e-06,
      "loss": 0.66330075,
      "num_input_tokens_seen": 189373105,
      "step": 8810,
      "time_per_iteration": 2.6315553188323975
    },
    {
      "auxiliary_loss_clip": 0.01109029,
      "auxiliary_loss_mlp": 0.01030302,
      "balance_loss_clip": 1.03765941,
      "balance_loss_mlp": 1.01749945,
      "epoch": 0.5297459792574778,
      "flos": 20961418129920.0,
      "grad_norm": 1.60926233169571,
      "language_loss": 0.68098927,
      "learning_rate": 1.9044235728283646e-06,
      "loss": 0.70238256,
      "num_input_tokens_seen": 189394615,
      "step": 8811,
      "time_per_iteration": 2.4903066158294678
    },
    {
      "auxiliary_loss_clip": 0.0100176,
      "auxiliary_loss_mlp": 0.01006006,
      "balance_loss_clip": 1.01468039,
      "balance_loss_mlp": 1.00457561,
      "epoch": 0.5298061025101458,
      "flos": 66523620389760.0,
      "grad_norm": 0.6646458687708908,
      "language_loss": 0.53352982,
      "learning_rate": 1.9040345579250953e-06,
      "loss": 0.55360746,
      "num_input_tokens_seen": 189459750,
      "step": 8812,
      "time_per_iteration": 4.6251256465911865
    },
    {
      "auxiliary_loss_clip": 0.0102073,
      "auxiliary_loss_mlp": 0.01007061,
      "balance_loss_clip": 1.0154233,
      "balance_loss_mlp": 1.00551116,
      "epoch": 0.5298662257628137,
      "flos": 67662994775040.0,
      "grad_norm": 0.7250707842165657,
      "language_loss": 0.56333899,
      "learning_rate": 1.9036455466608453e-06,
      "loss": 0.58361697,
      "num_input_tokens_seen": 189527540,
      "step": 8813,
      "time_per_iteration": 3.178156614303589
    },
    {
      "auxiliary_loss_clip": 0.01060476,
      "auxiliary_loss_mlp": 0.01031507,
      "balance_loss_clip": 1.03429401,
      "balance_loss_mlp": 1.01925218,
      "epoch": 0.5299263490154817,
      "flos": 19646405216640.0,
      "grad_norm": 1.5607167064876728,
      "language_loss": 0.81642097,
      "learning_rate": 1.9032565390503657e-06,
      "loss": 0.83734083,
      "num_input_tokens_seen": 189546900,
      "step": 8814,
      "time_per_iteration": 2.544072151184082
    },
    {
      "auxiliary_loss_clip": 0.01117883,
      "auxiliary_loss_mlp": 0.01028198,
      "balance_loss_clip": 1.0414778,
      "balance_loss_mlp": 1.0158186,
      "epoch": 0.5299864722681497,
      "flos": 22055005653120.0,
      "grad_norm": 1.541376360201913,
      "language_loss": 0.8487488,
      "learning_rate": 1.9028675351084076e-06,
      "loss": 0.87020963,
      "num_input_tokens_seen": 189566490,
      "step": 8815,
      "time_per_iteration": 2.4473845958709717
    },
    {
      "auxiliary_loss_clip": 0.01108933,
      "auxiliary_loss_mlp": 0.01030556,
      "balance_loss_clip": 1.03812003,
      "balance_loss_mlp": 1.01870692,
      "epoch": 0.5300465955208177,
      "flos": 21763698353280.0,
      "grad_norm": 2.052153190873808,
      "language_loss": 0.6629948,
      "learning_rate": 1.9024785348497225e-06,
      "loss": 0.68438971,
      "num_input_tokens_seen": 189585580,
      "step": 8816,
      "time_per_iteration": 2.4568278789520264
    },
    {
      "auxiliary_loss_clip": 0.01086469,
      "auxiliary_loss_mlp": 0.01032495,
      "balance_loss_clip": 1.03479123,
      "balance_loss_mlp": 1.01998353,
      "epoch": 0.5301067187734857,
      "flos": 42996491735040.0,
      "grad_norm": 1.6192045599106437,
      "language_loss": 0.72049272,
      "learning_rate": 1.9020895382890611e-06,
      "loss": 0.74168229,
      "num_input_tokens_seen": 189608485,
      "step": 8817,
      "time_per_iteration": 2.7027342319488525
    },
    {
      "auxiliary_loss_clip": 0.01083771,
      "auxiliary_loss_mlp": 0.01034017,
      "balance_loss_clip": 1.03864694,
      "balance_loss_mlp": 1.01944351,
      "epoch": 0.5301668420261536,
      "flos": 20554298403840.0,
      "grad_norm": 1.6949087051134715,
      "language_loss": 0.6534133,
      "learning_rate": 1.9017005454411743e-06,
      "loss": 0.67459118,
      "num_input_tokens_seen": 189627815,
      "step": 8818,
      "time_per_iteration": 2.549274444580078
    },
    {
      "auxiliary_loss_clip": 0.01063469,
      "auxiliary_loss_mlp": 0.01029937,
      "balance_loss_clip": 1.03526986,
      "balance_loss_mlp": 1.0158112,
      "epoch": 0.5302269652788216,
      "flos": 17486665182720.0,
      "grad_norm": 1.9988119199409549,
      "language_loss": 0.75040686,
      "learning_rate": 1.9013115563208126e-06,
      "loss": 0.77134085,
      "num_input_tokens_seen": 189644850,
      "step": 8819,
      "time_per_iteration": 2.549574613571167
    },
    {
      "auxiliary_loss_clip": 0.01078472,
      "auxiliary_loss_mlp": 0.01042435,
      "balance_loss_clip": 1.0372076,
      "balance_loss_mlp": 1.02724767,
      "epoch": 0.5302870885314895,
      "flos": 14574202715520.0,
      "grad_norm": 1.7223021518799049,
      "language_loss": 0.82230282,
      "learning_rate": 1.9009225709427267e-06,
      "loss": 0.84351188,
      "num_input_tokens_seen": 189660945,
      "step": 8820,
      "time_per_iteration": 2.4999117851257324
    },
    {
      "auxiliary_loss_clip": 0.0109066,
      "auxiliary_loss_mlp": 0.0102789,
      "balance_loss_clip": 1.03676426,
      "balance_loss_mlp": 1.0159694,
      "epoch": 0.5303472117841576,
      "flos": 23438032968960.0,
      "grad_norm": 1.3701693847326617,
      "language_loss": 0.72499669,
      "learning_rate": 1.9005335893216667e-06,
      "loss": 0.7461822,
      "num_input_tokens_seen": 189680425,
      "step": 8821,
      "time_per_iteration": 2.5279476642608643
    },
    {
      "auxiliary_loss_clip": 0.01089517,
      "auxiliary_loss_mlp": 0.01029088,
      "balance_loss_clip": 1.03791213,
      "balance_loss_mlp": 1.01773906,
      "epoch": 0.5304073350368255,
      "flos": 22709010533760.0,
      "grad_norm": 1.388243600782498,
      "language_loss": 0.74370039,
      "learning_rate": 1.9001446114723824e-06,
      "loss": 0.76488644,
      "num_input_tokens_seen": 189700375,
      "step": 8822,
      "time_per_iteration": 2.51324200630188
    },
    {
      "auxiliary_loss_clip": 0.01077314,
      "auxiliary_loss_mlp": 0.010316,
      "balance_loss_clip": 1.0345695,
      "balance_loss_mlp": 1.01812339,
      "epoch": 0.5304674582894935,
      "flos": 27928554624000.0,
      "grad_norm": 1.6591147642005715,
      "language_loss": 0.67588174,
      "learning_rate": 1.8997556374096257e-06,
      "loss": 0.69697094,
      "num_input_tokens_seen": 189721225,
      "step": 8823,
      "time_per_iteration": 2.6033661365509033
    },
    {
      "auxiliary_loss_clip": 0.01114512,
      "auxiliary_loss_mlp": 0.01035561,
      "balance_loss_clip": 1.03810596,
      "balance_loss_mlp": 1.02057099,
      "epoch": 0.5305275815421614,
      "flos": 21250642440960.0,
      "grad_norm": 1.5888902252179407,
      "language_loss": 0.6928134,
      "learning_rate": 1.8993666671481444e-06,
      "loss": 0.71431422,
      "num_input_tokens_seen": 189740170,
      "step": 8824,
      "time_per_iteration": 2.451096773147583
    },
    {
      "auxiliary_loss_clip": 0.01085273,
      "auxiliary_loss_mlp": 0.0078509,
      "balance_loss_clip": 1.03660226,
      "balance_loss_mlp": 1.00846243,
      "epoch": 0.5305877047948294,
      "flos": 17603088140160.0,
      "grad_norm": 1.8468785738780347,
      "language_loss": 0.76105809,
      "learning_rate": 1.898977700702689e-06,
      "loss": 0.77976173,
      "num_input_tokens_seen": 189757890,
      "step": 8825,
      "time_per_iteration": 2.4923934936523438
    },
    {
      "auxiliary_loss_clip": 0.01037063,
      "auxiliary_loss_mlp": 0.01039817,
      "balance_loss_clip": 1.03609324,
      "balance_loss_mlp": 1.0268054,
      "epoch": 0.5306478280474973,
      "flos": 15195493284480.0,
      "grad_norm": 1.810733362323992,
      "language_loss": 0.85685831,
      "learning_rate": 1.8985887380880103e-06,
      "loss": 0.87762713,
      "num_input_tokens_seen": 189775390,
      "step": 8826,
      "time_per_iteration": 2.6935248374938965
    },
    {
      "auxiliary_loss_clip": 0.01108904,
      "auxiliary_loss_mlp": 0.0103208,
      "balance_loss_clip": 1.03781676,
      "balance_loss_mlp": 1.01948571,
      "epoch": 0.5307079513001653,
      "flos": 15341218761600.0,
      "grad_norm": 1.4344760742001632,
      "language_loss": 0.64074767,
      "learning_rate": 1.8981997793188558e-06,
      "loss": 0.66215754,
      "num_input_tokens_seen": 189793975,
      "step": 8827,
      "time_per_iteration": 2.5506958961486816
    },
    {
      "auxiliary_loss_clip": 0.01092027,
      "auxiliary_loss_mlp": 0.01036651,
      "balance_loss_clip": 1.03888774,
      "balance_loss_mlp": 1.02267981,
      "epoch": 0.5307680745528333,
      "flos": 43544452688640.0,
      "grad_norm": 1.5165095006892586,
      "language_loss": 0.6009624,
      "learning_rate": 1.8978108244099762e-06,
      "loss": 0.62224919,
      "num_input_tokens_seen": 189817870,
      "step": 8828,
      "time_per_iteration": 2.7139461040496826
    },
    {
      "auxiliary_loss_clip": 0.01104112,
      "auxiliary_loss_mlp": 0.01034345,
      "balance_loss_clip": 1.03852534,
      "balance_loss_mlp": 1.02090454,
      "epoch": 0.5308281978055013,
      "flos": 20048928001920.0,
      "grad_norm": 1.8206481503508534,
      "language_loss": 0.81830728,
      "learning_rate": 1.8974218733761208e-06,
      "loss": 0.83969182,
      "num_input_tokens_seen": 189837905,
      "step": 8829,
      "time_per_iteration": 2.475341796875
    },
    {
      "auxiliary_loss_clip": 0.0109483,
      "auxiliary_loss_mlp": 0.0103445,
      "balance_loss_clip": 1.03921843,
      "balance_loss_mlp": 1.02185559,
      "epoch": 0.5308883210581693,
      "flos": 20703938463360.0,
      "grad_norm": 1.5178068612446072,
      "language_loss": 0.78223705,
      "learning_rate": 1.8970329262320375e-06,
      "loss": 0.80352986,
      "num_input_tokens_seen": 189856970,
      "step": 8830,
      "time_per_iteration": 2.494155168533325
    },
    {
      "auxiliary_loss_clip": 0.0109943,
      "auxiliary_loss_mlp": 0.01030385,
      "balance_loss_clip": 1.0378654,
      "balance_loss_mlp": 1.0178982,
      "epoch": 0.5309484443108372,
      "flos": 14355506759040.0,
      "grad_norm": 1.9570317922740803,
      "language_loss": 0.80517089,
      "learning_rate": 1.8966439829924768e-06,
      "loss": 0.82646906,
      "num_input_tokens_seen": 189872830,
      "step": 8831,
      "time_per_iteration": 2.431706428527832
    },
    {
      "auxiliary_loss_clip": 0.01095836,
      "auxiliary_loss_mlp": 0.01033891,
      "balance_loss_clip": 1.03578544,
      "balance_loss_mlp": 1.02115941,
      "epoch": 0.5310085675635052,
      "flos": 20010503427840.0,
      "grad_norm": 1.7306962771579704,
      "language_loss": 0.73024249,
      "learning_rate": 1.896255043672186e-06,
      "loss": 0.75153977,
      "num_input_tokens_seen": 189891635,
      "step": 8832,
      "time_per_iteration": 2.46694016456604
    },
    {
      "auxiliary_loss_clip": 0.01079498,
      "auxiliary_loss_mlp": 0.01033476,
      "balance_loss_clip": 1.03745055,
      "balance_loss_mlp": 1.02007651,
      "epoch": 0.5310686908161731,
      "flos": 22127293774080.0,
      "grad_norm": 2.5453984931172355,
      "language_loss": 0.75989699,
      "learning_rate": 1.8958661082859143e-06,
      "loss": 0.78102678,
      "num_input_tokens_seen": 189909050,
      "step": 8833,
      "time_per_iteration": 2.529231071472168
    },
    {
      "auxiliary_loss_clip": 0.01077317,
      "auxiliary_loss_mlp": 0.01029805,
      "balance_loss_clip": 1.03443646,
      "balance_loss_mlp": 1.01670957,
      "epoch": 0.5311288140688412,
      "flos": 24717889445760.0,
      "grad_norm": 1.8849927235719406,
      "language_loss": 0.73522967,
      "learning_rate": 1.8954771768484103e-06,
      "loss": 0.75630087,
      "num_input_tokens_seen": 189927405,
      "step": 8834,
      "time_per_iteration": 2.5676491260528564
    },
    {
      "auxiliary_loss_clip": 0.01115219,
      "auxiliary_loss_mlp": 0.01039172,
      "balance_loss_clip": 1.03777611,
      "balance_loss_mlp": 1.02546263,
      "epoch": 0.5311889373215091,
      "flos": 24097712198400.0,
      "grad_norm": 1.7697506996630872,
      "language_loss": 0.77522957,
      "learning_rate": 1.8950882493744226e-06,
      "loss": 0.79677343,
      "num_input_tokens_seen": 189947740,
      "step": 8835,
      "time_per_iteration": 2.472043752670288
    },
    {
      "auxiliary_loss_clip": 0.01085646,
      "auxiliary_loss_mlp": 0.01038329,
      "balance_loss_clip": 1.03710377,
      "balance_loss_mlp": 1.02481627,
      "epoch": 0.5312490605741771,
      "flos": 22017012042240.0,
      "grad_norm": 1.6297169140076533,
      "language_loss": 0.72666991,
      "learning_rate": 1.8946993258786985e-06,
      "loss": 0.74790967,
      "num_input_tokens_seen": 189966495,
      "step": 8836,
      "time_per_iteration": 2.528639316558838
    },
    {
      "auxiliary_loss_clip": 0.01089143,
      "auxiliary_loss_mlp": 0.01034296,
      "balance_loss_clip": 1.03573227,
      "balance_loss_mlp": 1.02083778,
      "epoch": 0.531309183826845,
      "flos": 19390541662080.0,
      "grad_norm": 1.6820242543645831,
      "language_loss": 0.80414808,
      "learning_rate": 1.894310406375987e-06,
      "loss": 0.82538247,
      "num_input_tokens_seen": 189985325,
      "step": 8837,
      "time_per_iteration": 2.481680393218994
    },
    {
      "auxiliary_loss_clip": 0.01097307,
      "auxiliary_loss_mlp": 0.01028349,
      "balance_loss_clip": 1.04013777,
      "balance_loss_mlp": 1.01520014,
      "epoch": 0.531369307079513,
      "flos": 20190056538240.0,
      "grad_norm": 1.8454643863168314,
      "language_loss": 0.85842234,
      "learning_rate": 1.893921490881035e-06,
      "loss": 0.87967891,
      "num_input_tokens_seen": 190003290,
      "step": 8838,
      "time_per_iteration": 2.4948747158050537
    },
    {
      "auxiliary_loss_clip": 0.01086514,
      "auxiliary_loss_mlp": 0.01032278,
      "balance_loss_clip": 1.03731215,
      "balance_loss_mlp": 1.02058935,
      "epoch": 0.5314294303321809,
      "flos": 18880143356160.0,
      "grad_norm": 1.589437835727425,
      "language_loss": 0.7282114,
      "learning_rate": 1.8935325794085906e-06,
      "loss": 0.7493993,
      "num_input_tokens_seen": 190023260,
      "step": 8839,
      "time_per_iteration": 2.5506763458251953
    },
    {
      "auxiliary_loss_clip": 0.01089451,
      "auxiliary_loss_mlp": 0.01033648,
      "balance_loss_clip": 1.03523993,
      "balance_loss_mlp": 1.02104795,
      "epoch": 0.531489553584849,
      "flos": 23040035297280.0,
      "grad_norm": 2.3672513792432643,
      "language_loss": 0.76840651,
      "learning_rate": 1.8931436719734023e-06,
      "loss": 0.78963757,
      "num_input_tokens_seen": 190042035,
      "step": 8840,
      "time_per_iteration": 3.9193923473358154
    },
    {
      "auxiliary_loss_clip": 0.01075765,
      "auxiliary_loss_mlp": 0.01032175,
      "balance_loss_clip": 1.03995538,
      "balance_loss_mlp": 1.0185616,
      "epoch": 0.5315496768375169,
      "flos": 19790478668160.0,
      "grad_norm": 2.114153460744273,
      "language_loss": 0.77165067,
      "learning_rate": 1.892754768590216e-06,
      "loss": 0.79273009,
      "num_input_tokens_seen": 190057545,
      "step": 8841,
      "time_per_iteration": 2.4978842735290527
    },
    {
      "auxiliary_loss_clip": 0.01023043,
      "auxiliary_loss_mlp": 0.01011148,
      "balance_loss_clip": 1.01688302,
      "balance_loss_mlp": 1.00971115,
      "epoch": 0.5316098000901849,
      "flos": 71023228185600.0,
      "grad_norm": 0.6964527354883857,
      "language_loss": 0.56793249,
      "learning_rate": 1.8923658692737793e-06,
      "loss": 0.58827436,
      "num_input_tokens_seen": 190123800,
      "step": 8842,
      "time_per_iteration": 3.2480628490448
    },
    {
      "auxiliary_loss_clip": 0.0108984,
      "auxiliary_loss_mlp": 0.01038474,
      "balance_loss_clip": 1.03701329,
      "balance_loss_mlp": 1.02423465,
      "epoch": 0.5316699233428529,
      "flos": 16435560470400.0,
      "grad_norm": 1.7767282875113812,
      "language_loss": 0.73261827,
      "learning_rate": 1.8919769740388407e-06,
      "loss": 0.75390142,
      "num_input_tokens_seen": 190141625,
      "step": 8843,
      "time_per_iteration": 2.5629043579101562
    },
    {
      "auxiliary_loss_clip": 0.01021872,
      "auxiliary_loss_mlp": 0.01008534,
      "balance_loss_clip": 1.01825929,
      "balance_loss_mlp": 1.00718737,
      "epoch": 0.5317300465955208,
      "flos": 67420814302080.0,
      "grad_norm": 0.884266741422586,
      "language_loss": 0.61104798,
      "learning_rate": 1.891588082900145e-06,
      "loss": 0.63135207,
      "num_input_tokens_seen": 190198110,
      "step": 8844,
      "time_per_iteration": 4.508849143981934
    },
    {
      "auxiliary_loss_clip": 0.01029252,
      "auxiliary_loss_mlp": 0.01004391,
      "balance_loss_clip": 1.01285982,
      "balance_loss_mlp": 1.00306785,
      "epoch": 0.5317901698481888,
      "flos": 59508075340800.0,
      "grad_norm": 0.9214549893239615,
      "language_loss": 0.62220001,
      "learning_rate": 1.8911991958724411e-06,
      "loss": 0.64253652,
      "num_input_tokens_seen": 190259950,
      "step": 8845,
      "time_per_iteration": 3.0963025093078613
    },
    {
      "auxiliary_loss_clip": 0.0107924,
      "auxiliary_loss_mlp": 0.01036637,
      "balance_loss_clip": 1.03666067,
      "balance_loss_mlp": 1.02254677,
      "epoch": 0.5318502931008567,
      "flos": 19129219240320.0,
      "grad_norm": 1.9460863601061407,
      "language_loss": 0.75377858,
      "learning_rate": 1.890810312970474e-06,
      "loss": 0.77493727,
      "num_input_tokens_seen": 190278265,
      "step": 8846,
      "time_per_iteration": 3.9222922325134277
    },
    {
      "auxiliary_loss_clip": 0.01100045,
      "auxiliary_loss_mlp": 0.01030283,
      "balance_loss_clip": 1.03783047,
      "balance_loss_mlp": 1.01860678,
      "epoch": 0.5319104163535248,
      "flos": 24681045070080.0,
      "grad_norm": 2.3332897113657602,
      "language_loss": 0.75595665,
      "learning_rate": 1.8904214342089903e-06,
      "loss": 0.77725989,
      "num_input_tokens_seen": 190298400,
      "step": 8847,
      "time_per_iteration": 2.5144827365875244
    },
    {
      "auxiliary_loss_clip": 0.01088401,
      "auxiliary_loss_mlp": 0.01033667,
      "balance_loss_clip": 1.03498816,
      "balance_loss_mlp": 1.02192438,
      "epoch": 0.5319705396061927,
      "flos": 19385513758080.0,
      "grad_norm": 1.6118528524542113,
      "language_loss": 0.87481391,
      "learning_rate": 1.8900325596027378e-06,
      "loss": 0.8960346,
      "num_input_tokens_seen": 190316235,
      "step": 8848,
      "time_per_iteration": 2.500892162322998
    },
    {
      "auxiliary_loss_clip": 0.01073238,
      "auxiliary_loss_mlp": 0.01042044,
      "balance_loss_clip": 1.03651643,
      "balance_loss_mlp": 1.02747679,
      "epoch": 0.5320306628588607,
      "flos": 18259319664000.0,
      "grad_norm": 1.9194592834757513,
      "language_loss": 0.74341285,
      "learning_rate": 1.8896436891664609e-06,
      "loss": 0.76456559,
      "num_input_tokens_seen": 190335060,
      "step": 8849,
      "time_per_iteration": 2.507690906524658
    },
    {
      "auxiliary_loss_clip": 0.01098214,
      "auxiliary_loss_mlp": 0.01030079,
      "balance_loss_clip": 1.03563452,
      "balance_loss_mlp": 1.01648879,
      "epoch": 0.5320907861115286,
      "flos": 23732321097600.0,
      "grad_norm": 2.0803518343236513,
      "language_loss": 0.80034107,
      "learning_rate": 1.8892548229149066e-06,
      "loss": 0.82162398,
      "num_input_tokens_seen": 190353265,
      "step": 8850,
      "time_per_iteration": 3.8599796295166016
    },
    {
      "auxiliary_loss_clip": 0.01109002,
      "auxiliary_loss_mlp": 0.01031979,
      "balance_loss_clip": 1.03503299,
      "balance_loss_mlp": 1.01934302,
      "epoch": 0.5321509093641966,
      "flos": 34495251321600.0,
      "grad_norm": 1.452499425185094,
      "language_loss": 0.5480473,
      "learning_rate": 1.888865960862821e-06,
      "loss": 0.56945717,
      "num_input_tokens_seen": 190376575,
      "step": 8851,
      "time_per_iteration": 2.5714170932769775
    },
    {
      "auxiliary_loss_clip": 0.01100574,
      "auxiliary_loss_mlp": 0.01031272,
      "balance_loss_clip": 1.03755212,
      "balance_loss_mlp": 1.0185349,
      "epoch": 0.5322110326168645,
      "flos": 20010934391040.0,
      "grad_norm": 2.305627235795431,
      "language_loss": 0.68762118,
      "learning_rate": 1.8884771030249484e-06,
      "loss": 0.70893961,
      "num_input_tokens_seen": 190395185,
      "step": 8852,
      "time_per_iteration": 2.4640114307403564
    },
    {
      "auxiliary_loss_clip": 0.01025032,
      "auxiliary_loss_mlp": 0.00770314,
      "balance_loss_clip": 1.0187819,
      "balance_loss_mlp": 1.0128727,
      "epoch": 0.5322711558695326,
      "flos": 64631164435200.0,
      "grad_norm": 0.8017353068623295,
      "language_loss": 0.62855095,
      "learning_rate": 1.8880882494160357e-06,
      "loss": 0.6465044,
      "num_input_tokens_seen": 190452595,
      "step": 8853,
      "time_per_iteration": 3.0840625762939453
    },
    {
      "auxiliary_loss_clip": 0.01102932,
      "auxiliary_loss_mlp": 0.01030862,
      "balance_loss_clip": 1.03599966,
      "balance_loss_mlp": 1.017892,
      "epoch": 0.5323312791222005,
      "flos": 14939342421120.0,
      "grad_norm": 2.7146717863509395,
      "language_loss": 0.7978189,
      "learning_rate": 1.8876994000508278e-06,
      "loss": 0.81915677,
      "num_input_tokens_seen": 190469140,
      "step": 8854,
      "time_per_iteration": 2.446993589401245
    },
    {
      "auxiliary_loss_clip": 0.01085087,
      "auxiliary_loss_mlp": 0.01030757,
      "balance_loss_clip": 1.03972244,
      "balance_loss_mlp": 1.01891923,
      "epoch": 0.5323914023748685,
      "flos": 23440834229760.0,
      "grad_norm": 1.6034142014212835,
      "language_loss": 0.72991025,
      "learning_rate": 1.8873105549440698e-06,
      "loss": 0.75106871,
      "num_input_tokens_seen": 190489015,
      "step": 8855,
      "time_per_iteration": 2.5314016342163086
    },
    {
      "auxiliary_loss_clip": 0.01084784,
      "auxiliary_loss_mlp": 0.00787678,
      "balance_loss_clip": 1.0345422,
      "balance_loss_mlp": 1.01163268,
      "epoch": 0.5324515256275365,
      "flos": 26286180134400.0,
      "grad_norm": 1.9676317023358667,
      "language_loss": 0.65120047,
      "learning_rate": 1.886921714110507e-06,
      "loss": 0.66992515,
      "num_input_tokens_seen": 190508065,
      "step": 8856,
      "time_per_iteration": 2.54663348197937
    },
    {
      "auxiliary_loss_clip": 0.01093704,
      "auxiliary_loss_mlp": 0.01041999,
      "balance_loss_clip": 1.04278493,
      "balance_loss_mlp": 1.02801585,
      "epoch": 0.5325116488802044,
      "flos": 26870913636480.0,
      "grad_norm": 1.7394483018327733,
      "language_loss": 0.77508807,
      "learning_rate": 1.8865328775648842e-06,
      "loss": 0.79644513,
      "num_input_tokens_seen": 190527045,
      "step": 8857,
      "time_per_iteration": 2.5581252574920654
    },
    {
      "auxiliary_loss_clip": 0.01078226,
      "auxiliary_loss_mlp": 0.0103389,
      "balance_loss_clip": 1.03519368,
      "balance_loss_mlp": 1.02083063,
      "epoch": 0.5325717721328724,
      "flos": 25884734757120.0,
      "grad_norm": 3.300974127399562,
      "language_loss": 0.71253586,
      "learning_rate": 1.8861440453219456e-06,
      "loss": 0.733657,
      "num_input_tokens_seen": 190544075,
      "step": 8858,
      "time_per_iteration": 2.570976734161377
    },
    {
      "auxiliary_loss_clip": 0.011003,
      "auxiliary_loss_mlp": 0.01035948,
      "balance_loss_clip": 1.03818583,
      "balance_loss_mlp": 1.02226269,
      "epoch": 0.5326318953855403,
      "flos": 21799321666560.0,
      "grad_norm": 1.8036911421776591,
      "language_loss": 0.69247031,
      "learning_rate": 1.8857552173964367e-06,
      "loss": 0.7138328,
      "num_input_tokens_seen": 190566030,
      "step": 8859,
      "time_per_iteration": 2.5476136207580566
    },
    {
      "auxiliary_loss_clip": 0.01097535,
      "auxiliary_loss_mlp": 0.01027081,
      "balance_loss_clip": 1.03798366,
      "balance_loss_mlp": 1.01573253,
      "epoch": 0.5326920186382084,
      "flos": 20922921728640.0,
      "grad_norm": 1.5253963137840456,
      "language_loss": 0.69729698,
      "learning_rate": 1.8853663938031013e-06,
      "loss": 0.71854317,
      "num_input_tokens_seen": 190585605,
      "step": 8860,
      "time_per_iteration": 2.4793801307678223
    },
    {
      "auxiliary_loss_clip": 0.01090447,
      "auxiliary_loss_mlp": 0.01033406,
      "balance_loss_clip": 1.03846955,
      "balance_loss_mlp": 1.02121651,
      "epoch": 0.5327521418908763,
      "flos": 21433427775360.0,
      "grad_norm": 1.8416910874366852,
      "language_loss": 0.78061098,
      "learning_rate": 1.884977574556683e-06,
      "loss": 0.80184948,
      "num_input_tokens_seen": 190604625,
      "step": 8861,
      "time_per_iteration": 2.5385754108428955
    },
    {
      "auxiliary_loss_clip": 0.01068515,
      "auxiliary_loss_mlp": 0.01038013,
      "balance_loss_clip": 1.03563857,
      "balance_loss_mlp": 1.02446485,
      "epoch": 0.5328122651435443,
      "flos": 21760250647680.0,
      "grad_norm": 1.6168232085497585,
      "language_loss": 0.85558975,
      "learning_rate": 1.8845887596719279e-06,
      "loss": 0.87665498,
      "num_input_tokens_seen": 190625060,
      "step": 8862,
      "time_per_iteration": 2.6035940647125244
    },
    {
      "auxiliary_loss_clip": 0.01089523,
      "auxiliary_loss_mlp": 0.01035891,
      "balance_loss_clip": 1.03618598,
      "balance_loss_mlp": 1.0208708,
      "epoch": 0.5328723883962122,
      "flos": 18296487262080.0,
      "grad_norm": 1.9317830966729932,
      "language_loss": 0.61976731,
      "learning_rate": 1.8841999491635778e-06,
      "loss": 0.64102149,
      "num_input_tokens_seen": 190643150,
      "step": 8863,
      "time_per_iteration": 2.501035213470459
    },
    {
      "auxiliary_loss_clip": 0.01084806,
      "auxiliary_loss_mlp": 0.01033481,
      "balance_loss_clip": 1.03646755,
      "balance_loss_mlp": 1.02075517,
      "epoch": 0.5329325116488802,
      "flos": 25374911068800.0,
      "grad_norm": 2.1153957919630813,
      "language_loss": 0.73506457,
      "learning_rate": 1.883811143046377e-06,
      "loss": 0.7562474,
      "num_input_tokens_seen": 190662725,
      "step": 8864,
      "time_per_iteration": 2.559576988220215
    },
    {
      "auxiliary_loss_clip": 0.01109811,
      "auxiliary_loss_mlp": 0.01033179,
      "balance_loss_clip": 1.03644633,
      "balance_loss_mlp": 1.02090037,
      "epoch": 0.5329926349015481,
      "flos": 25592098654080.0,
      "grad_norm": 1.8087329027638728,
      "language_loss": 0.64423448,
      "learning_rate": 1.8834223413350702e-06,
      "loss": 0.66566437,
      "num_input_tokens_seen": 190683680,
      "step": 8865,
      "time_per_iteration": 2.5059685707092285
    },
    {
      "auxiliary_loss_clip": 0.01103147,
      "auxiliary_loss_mlp": 0.01028963,
      "balance_loss_clip": 1.03757572,
      "balance_loss_mlp": 1.01585007,
      "epoch": 0.5330527581542162,
      "flos": 22889605138560.0,
      "grad_norm": 1.7431726859571242,
      "language_loss": 0.78397131,
      "learning_rate": 1.8830335440443989e-06,
      "loss": 0.80529237,
      "num_input_tokens_seen": 190703350,
      "step": 8866,
      "time_per_iteration": 2.4838199615478516
    },
    {
      "auxiliary_loss_clip": 0.01100124,
      "auxiliary_loss_mlp": 0.01030063,
      "balance_loss_clip": 1.0372231,
      "balance_loss_mlp": 1.01787341,
      "epoch": 0.5331128814068841,
      "flos": 16026752805120.0,
      "grad_norm": 1.8639575183351227,
      "language_loss": 0.73488742,
      "learning_rate": 1.882644751189108e-06,
      "loss": 0.75618929,
      "num_input_tokens_seen": 190721170,
      "step": 8867,
      "time_per_iteration": 2.47721529006958
    },
    {
      "auxiliary_loss_clip": 0.01091277,
      "auxiliary_loss_mlp": 0.01035932,
      "balance_loss_clip": 1.03729224,
      "balance_loss_mlp": 1.02189505,
      "epoch": 0.5331730046595521,
      "flos": 39344699629440.0,
      "grad_norm": 1.5235289337311557,
      "language_loss": 0.71963638,
      "learning_rate": 1.88225596278394e-06,
      "loss": 0.7409085,
      "num_input_tokens_seen": 190743795,
      "step": 8868,
      "time_per_iteration": 2.6715803146362305
    },
    {
      "auxiliary_loss_clip": 0.01079845,
      "auxiliary_loss_mlp": 0.0103175,
      "balance_loss_clip": 1.03589606,
      "balance_loss_mlp": 1.01900065,
      "epoch": 0.5332331279122201,
      "flos": 24024382583040.0,
      "grad_norm": 1.6987140598097066,
      "language_loss": 0.78299797,
      "learning_rate": 1.881867178843637e-06,
      "loss": 0.80411392,
      "num_input_tokens_seen": 190761560,
      "step": 8869,
      "time_per_iteration": 2.55598521232605
    },
    {
      "auxiliary_loss_clip": 0.01106015,
      "auxiliary_loss_mlp": 0.01034442,
      "balance_loss_clip": 1.03921533,
      "balance_loss_mlp": 1.02135849,
      "epoch": 0.533293251164888,
      "flos": 17129318728320.0,
      "grad_norm": 2.2120467209435355,
      "language_loss": 0.76083565,
      "learning_rate": 1.8814783993829434e-06,
      "loss": 0.78224027,
      "num_input_tokens_seen": 190778875,
      "step": 8870,
      "time_per_iteration": 2.4554052352905273
    },
    {
      "auxiliary_loss_clip": 0.01096725,
      "auxiliary_loss_mlp": 0.01036731,
      "balance_loss_clip": 1.04203832,
      "balance_loss_mlp": 1.02241373,
      "epoch": 0.533353374417556,
      "flos": 22126360020480.0,
      "grad_norm": 2.2588376931270484,
      "language_loss": 0.75658548,
      "learning_rate": 1.8810896244165997e-06,
      "loss": 0.77792013,
      "num_input_tokens_seen": 190799830,
      "step": 8871,
      "time_per_iteration": 2.550447940826416
    },
    {
      "auxiliary_loss_clip": 0.01090881,
      "auxiliary_loss_mlp": 0.01029798,
      "balance_loss_clip": 1.03847063,
      "balance_loss_mlp": 1.01709008,
      "epoch": 0.533413497670224,
      "flos": 15011091838080.0,
      "grad_norm": 1.8282517877914424,
      "language_loss": 0.71990031,
      "learning_rate": 1.8807008539593498e-06,
      "loss": 0.74110711,
      "num_input_tokens_seen": 190817155,
      "step": 8872,
      "time_per_iteration": 2.4777228832244873
    },
    {
      "auxiliary_loss_clip": 0.01096523,
      "auxiliary_loss_mlp": 0.01040674,
      "balance_loss_clip": 1.04248929,
      "balance_loss_mlp": 1.02679777,
      "epoch": 0.533473620922892,
      "flos": 19609955890560.0,
      "grad_norm": 1.7363853137474092,
      "language_loss": 0.64858961,
      "learning_rate": 1.880312088025936e-06,
      "loss": 0.66996157,
      "num_input_tokens_seen": 190835240,
      "step": 8873,
      "time_per_iteration": 2.5047240257263184
    },
    {
      "auxiliary_loss_clip": 0.01089234,
      "auxiliary_loss_mlp": 0.01040777,
      "balance_loss_clip": 1.03890991,
      "balance_loss_mlp": 1.02805746,
      "epoch": 0.5335337441755599,
      "flos": 14282644020480.0,
      "grad_norm": 2.230796992790766,
      "language_loss": 0.79716361,
      "learning_rate": 1.879923326631099e-06,
      "loss": 0.81846368,
      "num_input_tokens_seen": 190851620,
      "step": 8874,
      "time_per_iteration": 2.460770845413208
    },
    {
      "auxiliary_loss_clip": 0.01101512,
      "auxiliary_loss_mlp": 0.01030072,
      "balance_loss_clip": 1.03743339,
      "balance_loss_mlp": 1.01690519,
      "epoch": 0.5335938674282279,
      "flos": 20814830726400.0,
      "grad_norm": 1.797949795920948,
      "language_loss": 0.69499087,
      "learning_rate": 1.879534569789582e-06,
      "loss": 0.71630675,
      "num_input_tokens_seen": 190870545,
      "step": 8875,
      "time_per_iteration": 2.4655468463897705
    },
    {
      "auxiliary_loss_clip": 0.01040372,
      "auxiliary_loss_mlp": 0.01008783,
      "balance_loss_clip": 1.01463079,
      "balance_loss_mlp": 1.00745392,
      "epoch": 0.5336539906808958,
      "flos": 71396448451200.0,
      "grad_norm": 0.7295463589188508,
      "language_loss": 0.59673262,
      "learning_rate": 1.879145817516126e-06,
      "loss": 0.61722422,
      "num_input_tokens_seen": 190931995,
      "step": 8876,
      "time_per_iteration": 3.1551454067230225
    },
    {
      "auxiliary_loss_clip": 0.01100858,
      "auxiliary_loss_mlp": 0.01036225,
      "balance_loss_clip": 1.03767931,
      "balance_loss_mlp": 1.02388048,
      "epoch": 0.5337141139335638,
      "flos": 20152996680960.0,
      "grad_norm": 1.7616443016570276,
      "language_loss": 0.74893641,
      "learning_rate": 1.8787570698254727e-06,
      "loss": 0.77030718,
      "num_input_tokens_seen": 190949890,
      "step": 8877,
      "time_per_iteration": 2.466925859451294
    },
    {
      "auxiliary_loss_clip": 0.01027719,
      "auxiliary_loss_mlp": 0.01006753,
      "balance_loss_clip": 1.01745057,
      "balance_loss_mlp": 1.00536382,
      "epoch": 0.5337742371862317,
      "flos": 67728387484800.0,
      "grad_norm": 0.7557983227766831,
      "language_loss": 0.57166171,
      "learning_rate": 1.8783683267323629e-06,
      "loss": 0.59200644,
      "num_input_tokens_seen": 191008480,
      "step": 8878,
      "time_per_iteration": 4.393060684204102
    },
    {
      "auxiliary_loss_clip": 0.01115291,
      "auxiliary_loss_mlp": 0.01032935,
      "balance_loss_clip": 1.03868067,
      "balance_loss_mlp": 1.01957726,
      "epoch": 0.5338343604388998,
      "flos": 25008909436800.0,
      "grad_norm": 1.4166382808807116,
      "language_loss": 0.72675014,
      "learning_rate": 1.8779795882515395e-06,
      "loss": 0.74823242,
      "num_input_tokens_seen": 191028995,
      "step": 8879,
      "time_per_iteration": 2.4932034015655518
    },
    {
      "auxiliary_loss_clip": 0.01116124,
      "auxiliary_loss_mlp": 0.01032356,
      "balance_loss_clip": 1.04079676,
      "balance_loss_mlp": 1.01928496,
      "epoch": 0.5338944836915677,
      "flos": 17601256546560.0,
      "grad_norm": 2.7273697000534267,
      "language_loss": 0.83437538,
      "learning_rate": 1.8775908543977416e-06,
      "loss": 0.85586023,
      "num_input_tokens_seen": 191045285,
      "step": 8880,
      "time_per_iteration": 2.4517805576324463
    },
    {
      "auxiliary_loss_clip": 0.01048233,
      "auxiliary_loss_mlp": 0.01033523,
      "balance_loss_clip": 1.04017198,
      "balance_loss_mlp": 1.02097654,
      "epoch": 0.5339546069442357,
      "flos": 21724124544000.0,
      "grad_norm": 1.4839499407428323,
      "language_loss": 0.79164863,
      "learning_rate": 1.8772021251857107e-06,
      "loss": 0.8124662,
      "num_input_tokens_seen": 191066105,
      "step": 8881,
      "time_per_iteration": 2.6266725063323975
    },
    {
      "auxiliary_loss_clip": 0.01018255,
      "auxiliary_loss_mlp": 0.0100028,
      "balance_loss_clip": 1.01925111,
      "balance_loss_mlp": 0.99889147,
      "epoch": 0.5340147301969036,
      "flos": 69723583315200.0,
      "grad_norm": 0.7876625850245238,
      "language_loss": 0.59219396,
      "learning_rate": 1.8768134006301882e-06,
      "loss": 0.61237931,
      "num_input_tokens_seen": 191126315,
      "step": 8882,
      "time_per_iteration": 4.386980295181274
    },
    {
      "auxiliary_loss_clip": 0.01017834,
      "auxiliary_loss_mlp": 0.01004975,
      "balance_loss_clip": 1.01094866,
      "balance_loss_mlp": 1.00361562,
      "epoch": 0.5340748534495716,
      "flos": 63880701580800.0,
      "grad_norm": 0.8822989567967736,
      "language_loss": 0.63754034,
      "learning_rate": 1.876424680745913e-06,
      "loss": 0.65776849,
      "num_input_tokens_seen": 191174240,
      "step": 8883,
      "time_per_iteration": 2.942675828933716
    },
    {
      "auxiliary_loss_clip": 0.01070417,
      "auxiliary_loss_mlp": 0.01030394,
      "balance_loss_clip": 1.03633618,
      "balance_loss_mlp": 1.0168699,
      "epoch": 0.5341349767022396,
      "flos": 28694313694080.0,
      "grad_norm": 2.255749921925642,
      "language_loss": 0.81706226,
      "learning_rate": 1.8760359655476272e-06,
      "loss": 0.83807027,
      "num_input_tokens_seen": 191193335,
      "step": 8884,
      "time_per_iteration": 4.074026346206665
    },
    {
      "auxiliary_loss_clip": 0.01077506,
      "auxiliary_loss_mlp": 0.01033267,
      "balance_loss_clip": 1.03473306,
      "balance_loss_mlp": 1.01964712,
      "epoch": 0.5341950999549075,
      "flos": 16289691338880.0,
      "grad_norm": 1.7164014047768612,
      "language_loss": 0.71803379,
      "learning_rate": 1.8756472550500695e-06,
      "loss": 0.73914146,
      "num_input_tokens_seen": 191210900,
      "step": 8885,
      "time_per_iteration": 2.485318899154663
    },
    {
      "auxiliary_loss_clip": 0.01093728,
      "auxiliary_loss_mlp": 0.01033108,
      "balance_loss_clip": 1.03687954,
      "balance_loss_mlp": 1.01938748,
      "epoch": 0.5342552232075756,
      "flos": 14355650413440.0,
      "grad_norm": 1.9035852968634204,
      "language_loss": 0.78585005,
      "learning_rate": 1.87525854926798e-06,
      "loss": 0.80711842,
      "num_input_tokens_seen": 191226730,
      "step": 8886,
      "time_per_iteration": 2.507058620452881
    },
    {
      "auxiliary_loss_clip": 0.01075181,
      "auxiliary_loss_mlp": 0.00804615,
      "balance_loss_clip": 1.03742695,
      "balance_loss_mlp": 1.04334879,
      "epoch": 0.5343153464602435,
      "flos": 30297976300800.0,
      "grad_norm": 1.39343387785755,
      "language_loss": 0.74625349,
      "learning_rate": 1.8748698482160996e-06,
      "loss": 0.76505142,
      "num_input_tokens_seen": 191250435,
      "step": 8887,
      "time_per_iteration": 2.6480603218078613
    },
    {
      "auxiliary_loss_clip": 0.01085944,
      "auxiliary_loss_mlp": 0.01029564,
      "balance_loss_clip": 1.03524911,
      "balance_loss_mlp": 1.0169276,
      "epoch": 0.5343754697129115,
      "flos": 15596292216960.0,
      "grad_norm": 2.2899171369686306,
      "language_loss": 0.69406605,
      "learning_rate": 1.8744811519091663e-06,
      "loss": 0.71522117,
      "num_input_tokens_seen": 191268315,
      "step": 8888,
      "time_per_iteration": 2.4928061962127686
    },
    {
      "auxiliary_loss_clip": 0.0110891,
      "auxiliary_loss_mlp": 0.0103624,
      "balance_loss_clip": 1.03843808,
      "balance_loss_mlp": 1.02288246,
      "epoch": 0.5344355929655794,
      "flos": 16909617191040.0,
      "grad_norm": 3.4626087682820854,
      "language_loss": 0.77534282,
      "learning_rate": 1.8740924603619208e-06,
      "loss": 0.79679435,
      "num_input_tokens_seen": 191287000,
      "step": 8889,
      "time_per_iteration": 3.863157033920288
    },
    {
      "auxiliary_loss_clip": 0.01112028,
      "auxiliary_loss_mlp": 0.01036966,
      "balance_loss_clip": 1.03865254,
      "balance_loss_mlp": 1.0237155,
      "epoch": 0.5344957162182474,
      "flos": 16798186224000.0,
      "grad_norm": 1.9650683719511481,
      "language_loss": 0.69346285,
      "learning_rate": 1.873703773589102e-06,
      "loss": 0.71495277,
      "num_input_tokens_seen": 191304565,
      "step": 8890,
      "time_per_iteration": 2.4177639484405518
    },
    {
      "auxiliary_loss_clip": 0.01116397,
      "auxiliary_loss_mlp": 0.01038386,
      "balance_loss_clip": 1.03845286,
      "balance_loss_mlp": 1.02395535,
      "epoch": 0.5345558394709153,
      "flos": 12705590413440.0,
      "grad_norm": 3.135280956408226,
      "language_loss": 0.76970375,
      "learning_rate": 1.8733150916054483e-06,
      "loss": 0.79125154,
      "num_input_tokens_seen": 191318300,
      "step": 8891,
      "time_per_iteration": 2.386655330657959
    },
    {
      "auxiliary_loss_clip": 0.01093729,
      "auxiliary_loss_mlp": 0.01037799,
      "balance_loss_clip": 1.03569019,
      "balance_loss_mlp": 1.02445388,
      "epoch": 0.5346159627235834,
      "flos": 22455050400000.0,
      "grad_norm": 1.6681495886216764,
      "language_loss": 0.73889321,
      "learning_rate": 1.872926414425699e-06,
      "loss": 0.76020855,
      "num_input_tokens_seen": 191337925,
      "step": 8892,
      "time_per_iteration": 2.5085411071777344
    },
    {
      "auxiliary_loss_clip": 0.01092789,
      "auxiliary_loss_mlp": 0.01030449,
      "balance_loss_clip": 1.0367012,
      "balance_loss_mlp": 1.01797962,
      "epoch": 0.5346760859762513,
      "flos": 22415763899520.0,
      "grad_norm": 1.5279530923065847,
      "language_loss": 0.87617207,
      "learning_rate": 1.8725377420645932e-06,
      "loss": 0.89740443,
      "num_input_tokens_seen": 191357120,
      "step": 8893,
      "time_per_iteration": 2.5016818046569824
    },
    {
      "auxiliary_loss_clip": 0.01107833,
      "auxiliary_loss_mlp": 0.01030459,
      "balance_loss_clip": 1.03651381,
      "balance_loss_mlp": 1.01875293,
      "epoch": 0.5347362092289193,
      "flos": 22816131868800.0,
      "grad_norm": 1.7576140555522919,
      "language_loss": 0.7268635,
      "learning_rate": 1.872149074536869e-06,
      "loss": 0.74824643,
      "num_input_tokens_seen": 191375395,
      "step": 8894,
      "time_per_iteration": 2.4415650367736816
    },
    {
      "auxiliary_loss_clip": 0.01098356,
      "auxiliary_loss_mlp": 0.01028788,
      "balance_loss_clip": 1.03535032,
      "balance_loss_mlp": 1.01590705,
      "epoch": 0.5347963324815872,
      "flos": 23219480666880.0,
      "grad_norm": 1.6736654905903399,
      "language_loss": 0.74563682,
      "learning_rate": 1.8717604118572648e-06,
      "loss": 0.76690823,
      "num_input_tokens_seen": 191395595,
      "step": 8895,
      "time_per_iteration": 2.4711191654205322
    },
    {
      "auxiliary_loss_clip": 0.01079011,
      "auxiliary_loss_mlp": 0.01034528,
      "balance_loss_clip": 1.03552985,
      "balance_loss_mlp": 1.02138567,
      "epoch": 0.5348564557342552,
      "flos": 22601350494720.0,
      "grad_norm": 1.670656835252365,
      "language_loss": 0.76953089,
      "learning_rate": 1.8713717540405178e-06,
      "loss": 0.79066622,
      "num_input_tokens_seen": 191413730,
      "step": 8896,
      "time_per_iteration": 2.5109450817108154
    },
    {
      "auxiliary_loss_clip": 0.01084855,
      "auxiliary_loss_mlp": 0.01029306,
      "balance_loss_clip": 1.0397799,
      "balance_loss_mlp": 1.01608586,
      "epoch": 0.5349165789869232,
      "flos": 18002378701440.0,
      "grad_norm": 1.7839552913938588,
      "language_loss": 0.78747451,
      "learning_rate": 1.8709831011013676e-06,
      "loss": 0.80861616,
      "num_input_tokens_seen": 191432400,
      "step": 8897,
      "time_per_iteration": 2.491626024246216
    },
    {
      "auxiliary_loss_clip": 0.01100499,
      "auxiliary_loss_mlp": 0.01030194,
      "balance_loss_clip": 1.03750265,
      "balance_loss_mlp": 1.0171349,
      "epoch": 0.5349767022395912,
      "flos": 17159770483200.0,
      "grad_norm": 1.9061309050400779,
      "language_loss": 0.75872719,
      "learning_rate": 1.8705944530545509e-06,
      "loss": 0.78003418,
      "num_input_tokens_seen": 191448855,
      "step": 8898,
      "time_per_iteration": 2.445260524749756
    },
    {
      "auxiliary_loss_clip": 0.01026779,
      "auxiliary_loss_mlp": 0.01006708,
      "balance_loss_clip": 1.01092434,
      "balance_loss_mlp": 1.00555217,
      "epoch": 0.5350368254922592,
      "flos": 70992058158720.0,
      "grad_norm": 1.094603363731391,
      "language_loss": 0.57986009,
      "learning_rate": 1.8702058099148052e-06,
      "loss": 0.60019499,
      "num_input_tokens_seen": 191519690,
      "step": 8899,
      "time_per_iteration": 3.2612180709838867
    },
    {
      "auxiliary_loss_clip": 0.01086621,
      "auxiliary_loss_mlp": 0.01027841,
      "balance_loss_clip": 1.03631639,
      "balance_loss_mlp": 1.01577699,
      "epoch": 0.5350969487449271,
      "flos": 27417833095680.0,
      "grad_norm": 1.704524789822995,
      "language_loss": 0.69910288,
      "learning_rate": 1.869817171696868e-06,
      "loss": 0.72024751,
      "num_input_tokens_seen": 191539380,
      "step": 8900,
      "time_per_iteration": 2.5682921409606934
    },
    {
      "auxiliary_loss_clip": 0.01090842,
      "auxiliary_loss_mlp": 0.01030027,
      "balance_loss_clip": 1.03661263,
      "balance_loss_mlp": 1.01713467,
      "epoch": 0.5351570719975951,
      "flos": 19316134638720.0,
      "grad_norm": 1.7186931972719326,
      "language_loss": 0.71464384,
      "learning_rate": 1.8694285384154777e-06,
      "loss": 0.73585254,
      "num_input_tokens_seen": 191557400,
      "step": 8901,
      "time_per_iteration": 2.5040676593780518
    },
    {
      "auxiliary_loss_clip": 0.01078472,
      "auxiliary_loss_mlp": 0.01028715,
      "balance_loss_clip": 1.03499532,
      "balance_loss_mlp": 1.01540565,
      "epoch": 0.535217195250263,
      "flos": 19828580019840.0,
      "grad_norm": 1.9175726107131226,
      "language_loss": 0.77480114,
      "learning_rate": 1.8690399100853699e-06,
      "loss": 0.79587305,
      "num_input_tokens_seen": 191575860,
      "step": 8902,
      "time_per_iteration": 2.533740282058716
    },
    {
      "auxiliary_loss_clip": 0.01078443,
      "auxiliary_loss_mlp": 0.01034326,
      "balance_loss_clip": 1.03658271,
      "balance_loss_mlp": 1.02166653,
      "epoch": 0.535277318502931,
      "flos": 22127868391680.0,
      "grad_norm": 1.551114097450694,
      "language_loss": 0.69892013,
      "learning_rate": 1.868651286721281e-06,
      "loss": 0.72004777,
      "num_input_tokens_seen": 191595775,
      "step": 8903,
      "time_per_iteration": 2.534207820892334
    },
    {
      "auxiliary_loss_clip": 0.01102388,
      "auxiliary_loss_mlp": 0.00786981,
      "balance_loss_clip": 1.03719354,
      "balance_loss_mlp": 1.01051903,
      "epoch": 0.5353374417555989,
      "flos": 25045897466880.0,
      "grad_norm": 1.4992524107329255,
      "language_loss": 0.72362369,
      "learning_rate": 1.86826266833795e-06,
      "loss": 0.74251741,
      "num_input_tokens_seen": 191617785,
      "step": 8904,
      "time_per_iteration": 2.523803234100342
    },
    {
      "auxiliary_loss_clip": 0.01089668,
      "auxiliary_loss_mlp": 0.01035265,
      "balance_loss_clip": 1.04012585,
      "balance_loss_mlp": 1.02200258,
      "epoch": 0.535397565008267,
      "flos": 19388710068480.0,
      "grad_norm": 1.7563683257851448,
      "language_loss": 0.73378301,
      "learning_rate": 1.8678740549501103e-06,
      "loss": 0.7550323,
      "num_input_tokens_seen": 191636900,
      "step": 8905,
      "time_per_iteration": 2.5006916522979736
    },
    {
      "auxiliary_loss_clip": 0.01095148,
      "auxiliary_loss_mlp": 0.01032679,
      "balance_loss_clip": 1.03652143,
      "balance_loss_mlp": 1.0216403,
      "epoch": 0.5354576882609349,
      "flos": 21471205904640.0,
      "grad_norm": 1.7235856618897354,
      "language_loss": 0.83773911,
      "learning_rate": 1.8674854465725005e-06,
      "loss": 0.85901737,
      "num_input_tokens_seen": 191656720,
      "step": 8906,
      "time_per_iteration": 2.4851813316345215
    },
    {
      "auxiliary_loss_clip": 0.01102375,
      "auxiliary_loss_mlp": 0.00786295,
      "balance_loss_clip": 1.03774607,
      "balance_loss_mlp": 1.00998139,
      "epoch": 0.5355178115136029,
      "flos": 20777519473920.0,
      "grad_norm": 2.0021584411697786,
      "language_loss": 0.73802584,
      "learning_rate": 1.8670968432198563e-06,
      "loss": 0.75691259,
      "num_input_tokens_seen": 191674445,
      "step": 8907,
      "time_per_iteration": 2.487630844116211
    },
    {
      "auxiliary_loss_clip": 0.01095163,
      "auxiliary_loss_mlp": 0.01032942,
      "balance_loss_clip": 1.03531194,
      "balance_loss_mlp": 1.01875019,
      "epoch": 0.5355779347662708,
      "flos": 23514020190720.0,
      "grad_norm": 1.7719611686255785,
      "language_loss": 0.76476085,
      "learning_rate": 1.866708244906912e-06,
      "loss": 0.78604186,
      "num_input_tokens_seen": 191695000,
      "step": 8908,
      "time_per_iteration": 2.4954416751861572
    },
    {
      "auxiliary_loss_clip": 0.01084163,
      "auxiliary_loss_mlp": 0.00787081,
      "balance_loss_clip": 1.03497124,
      "balance_loss_mlp": 1.00915468,
      "epoch": 0.5356380580189388,
      "flos": 20303211358080.0,
      "grad_norm": 1.9714029672802387,
      "language_loss": 0.7414428,
      "learning_rate": 1.8663196516484055e-06,
      "loss": 0.7601552,
      "num_input_tokens_seen": 191713295,
      "step": 8909,
      "time_per_iteration": 2.5071089267730713
    },
    {
      "auxiliary_loss_clip": 0.01078037,
      "auxiliary_loss_mlp": 0.0103401,
      "balance_loss_clip": 1.03829002,
      "balance_loss_mlp": 1.02163029,
      "epoch": 0.5356981812716068,
      "flos": 21361642444800.0,
      "grad_norm": 11.121529690526431,
      "language_loss": 0.84191227,
      "learning_rate": 1.8659310634590702e-06,
      "loss": 0.8630327,
      "num_input_tokens_seen": 191732725,
      "step": 8910,
      "time_per_iteration": 2.543142318725586
    },
    {
      "auxiliary_loss_clip": 0.01091764,
      "auxiliary_loss_mlp": 0.01027181,
      "balance_loss_clip": 1.0372839,
      "balance_loss_mlp": 1.01377547,
      "epoch": 0.5357583045242748,
      "flos": 23111246010240.0,
      "grad_norm": 1.5760762893812508,
      "language_loss": 0.81649995,
      "learning_rate": 1.8655424803536427e-06,
      "loss": 0.8376894,
      "num_input_tokens_seen": 191753765,
      "step": 8911,
      "time_per_iteration": 2.5574283599853516
    },
    {
      "auxiliary_loss_clip": 0.01072073,
      "auxiliary_loss_mlp": 0.01039355,
      "balance_loss_clip": 1.03520608,
      "balance_loss_mlp": 1.0267725,
      "epoch": 0.5358184277769428,
      "flos": 21141761339520.0,
      "grad_norm": 1.6406136273204073,
      "language_loss": 0.68994147,
      "learning_rate": 1.8651539023468585e-06,
      "loss": 0.7110557,
      "num_input_tokens_seen": 191773560,
      "step": 8912,
      "time_per_iteration": 2.535024404525757
    },
    {
      "auxiliary_loss_clip": 0.01088705,
      "auxiliary_loss_mlp": 0.01035572,
      "balance_loss_clip": 1.03659451,
      "balance_loss_mlp": 1.02258408,
      "epoch": 0.5358785510296107,
      "flos": 16282400878080.0,
      "grad_norm": 2.178268828648602,
      "language_loss": 0.71479493,
      "learning_rate": 1.8647653294534509e-06,
      "loss": 0.73603773,
      "num_input_tokens_seen": 191791255,
      "step": 8913,
      "time_per_iteration": 2.4686648845672607
    },
    {
      "auxiliary_loss_clip": 0.01083739,
      "auxiliary_loss_mlp": 0.01036224,
      "balance_loss_clip": 1.03838789,
      "balance_loss_mlp": 1.02324235,
      "epoch": 0.5359386742822787,
      "flos": 16976877408000.0,
      "grad_norm": 1.6163191546918634,
      "language_loss": 0.72359121,
      "learning_rate": 1.864376761688156e-06,
      "loss": 0.74479079,
      "num_input_tokens_seen": 191809325,
      "step": 8914,
      "time_per_iteration": 2.5103631019592285
    },
    {
      "auxiliary_loss_clip": 0.01099529,
      "auxiliary_loss_mlp": 0.01034842,
      "balance_loss_clip": 1.04048657,
      "balance_loss_mlp": 1.02001882,
      "epoch": 0.5359987975349466,
      "flos": 20812927305600.0,
      "grad_norm": 1.7922389264544722,
      "language_loss": 0.70658588,
      "learning_rate": 1.8639881990657079e-06,
      "loss": 0.72792959,
      "num_input_tokens_seen": 191829795,
      "step": 8915,
      "time_per_iteration": 2.550107002258301
    },
    {
      "auxiliary_loss_clip": 0.01084687,
      "auxiliary_loss_mlp": 0.01037855,
      "balance_loss_clip": 1.03698206,
      "balance_loss_mlp": 1.02413952,
      "epoch": 0.5360589207876146,
      "flos": 22199941031040.0,
      "grad_norm": 1.7221890696256141,
      "language_loss": 0.75130999,
      "learning_rate": 1.8635996416008408e-06,
      "loss": 0.77253544,
      "num_input_tokens_seen": 191850840,
      "step": 8916,
      "time_per_iteration": 2.5234532356262207
    },
    {
      "auxiliary_loss_clip": 0.0107029,
      "auxiliary_loss_mlp": 0.00785513,
      "balance_loss_clip": 1.03715968,
      "balance_loss_mlp": 1.00818968,
      "epoch": 0.5361190440402825,
      "flos": 31394365084800.0,
      "grad_norm": 2.3221406271133045,
      "language_loss": 0.72271752,
      "learning_rate": 1.863211089308289e-06,
      "loss": 0.74127561,
      "num_input_tokens_seen": 191869520,
      "step": 8917,
      "time_per_iteration": 2.6759450435638428
    },
    {
      "auxiliary_loss_clip": 0.01088855,
      "auxiliary_loss_mlp": 0.0103816,
      "balance_loss_clip": 1.03693211,
      "balance_loss_mlp": 1.02473116,
      "epoch": 0.5361791672929506,
      "flos": 16069882060800.0,
      "grad_norm": 1.981894412743698,
      "language_loss": 0.71170914,
      "learning_rate": 1.8628225422027865e-06,
      "loss": 0.7329793,
      "num_input_tokens_seen": 191887240,
      "step": 8918,
      "time_per_iteration": 3.9495198726654053
    },
    {
      "auxiliary_loss_clip": 0.01091629,
      "auxiliary_loss_mlp": 0.01034494,
      "balance_loss_clip": 1.03959763,
      "balance_loss_mlp": 1.02194178,
      "epoch": 0.5362392905456185,
      "flos": 20740926493440.0,
      "grad_norm": 1.710703532081005,
      "language_loss": 0.75058365,
      "learning_rate": 1.862434000299067e-06,
      "loss": 0.77184486,
      "num_input_tokens_seen": 191905690,
      "step": 8919,
      "time_per_iteration": 2.532991886138916
    },
    {
      "auxiliary_loss_clip": 0.0108537,
      "auxiliary_loss_mlp": 0.01029967,
      "balance_loss_clip": 1.04037786,
      "balance_loss_mlp": 1.01765823,
      "epoch": 0.5362994137982865,
      "flos": 17340077779200.0,
      "grad_norm": 2.237403090290034,
      "language_loss": 0.71441686,
      "learning_rate": 1.862045463611864e-06,
      "loss": 0.73557019,
      "num_input_tokens_seen": 191920725,
      "step": 8920,
      "time_per_iteration": 2.48478102684021
    },
    {
      "auxiliary_loss_clip": 0.01095063,
      "auxiliary_loss_mlp": 0.01035092,
      "balance_loss_clip": 1.03514338,
      "balance_loss_mlp": 1.02069736,
      "epoch": 0.5363595370509544,
      "flos": 42813957795840.0,
      "grad_norm": 1.7922836310646222,
      "language_loss": 0.68832541,
      "learning_rate": 1.8616569321559105e-06,
      "loss": 0.70962697,
      "num_input_tokens_seen": 191944645,
      "step": 8921,
      "time_per_iteration": 4.064924001693726
    },
    {
      "auxiliary_loss_clip": 0.01099985,
      "auxiliary_loss_mlp": 0.01030729,
      "balance_loss_clip": 1.03775787,
      "balance_loss_mlp": 1.01771116,
      "epoch": 0.5364196603036224,
      "flos": 19171953446400.0,
      "grad_norm": 2.1305176704732234,
      "language_loss": 0.81924653,
      "learning_rate": 1.86126840594594e-06,
      "loss": 0.84055364,
      "num_input_tokens_seen": 191962265,
      "step": 8922,
      "time_per_iteration": 2.4761569499969482
    },
    {
      "auxiliary_loss_clip": 0.01101068,
      "auxiliary_loss_mlp": 0.010267,
      "balance_loss_clip": 1.03643286,
      "balance_loss_mlp": 1.01461208,
      "epoch": 0.5364797835562904,
      "flos": 17931060247680.0,
      "grad_norm": 1.9525577281782869,
      "language_loss": 0.76908201,
      "learning_rate": 1.860879884996686e-06,
      "loss": 0.79035968,
      "num_input_tokens_seen": 191978850,
      "step": 8923,
      "time_per_iteration": 3.8821146488189697
    },
    {
      "auxiliary_loss_clip": 0.010887,
      "auxiliary_loss_mlp": 0.0103472,
      "balance_loss_clip": 1.03849816,
      "balance_loss_mlp": 1.02098143,
      "epoch": 0.5365399068089584,
      "flos": 30228058477440.0,
      "grad_norm": 1.3904465728104847,
      "language_loss": 0.70315063,
      "learning_rate": 1.8604913693228804e-06,
      "loss": 0.72438478,
      "num_input_tokens_seen": 192002000,
      "step": 8924,
      "time_per_iteration": 2.585658311843872
    },
    {
      "auxiliary_loss_clip": 0.01082502,
      "auxiliary_loss_mlp": 0.01032079,
      "balance_loss_clip": 1.03798163,
      "balance_loss_mlp": 1.01800585,
      "epoch": 0.5366000300616264,
      "flos": 24891696380160.0,
      "grad_norm": 1.8689126593599472,
      "language_loss": 0.87042952,
      "learning_rate": 1.8601028589392558e-06,
      "loss": 0.89157534,
      "num_input_tokens_seen": 192019100,
      "step": 8925,
      "time_per_iteration": 2.5954110622406006
    },
    {
      "auxiliary_loss_clip": 0.01112053,
      "auxiliary_loss_mlp": 0.01033105,
      "balance_loss_clip": 1.03596377,
      "balance_loss_mlp": 1.02000391,
      "epoch": 0.5366601533142943,
      "flos": 29826649013760.0,
      "grad_norm": 1.7866558047221521,
      "language_loss": 0.77989936,
      "learning_rate": 1.8597143538605455e-06,
      "loss": 0.80135095,
      "num_input_tokens_seen": 192041660,
      "step": 8926,
      "time_per_iteration": 2.524592399597168
    },
    {
      "auxiliary_loss_clip": 0.01077005,
      "auxiliary_loss_mlp": 0.01027443,
      "balance_loss_clip": 1.04053354,
      "balance_loss_mlp": 1.01607013,
      "epoch": 0.5367202765669623,
      "flos": 27199352620800.0,
      "grad_norm": 1.425367494953728,
      "language_loss": 0.66990697,
      "learning_rate": 1.85932585410148e-06,
      "loss": 0.69095147,
      "num_input_tokens_seen": 192063540,
      "step": 8927,
      "time_per_iteration": 2.6046059131622314
    },
    {
      "auxiliary_loss_clip": 0.0110115,
      "auxiliary_loss_mlp": 0.01028776,
      "balance_loss_clip": 1.03526187,
      "balance_loss_mlp": 1.01570511,
      "epoch": 0.5367803998196302,
      "flos": 20229953569920.0,
      "grad_norm": 1.8071236903875367,
      "language_loss": 0.73617846,
      "learning_rate": 1.8589373596767929e-06,
      "loss": 0.75747776,
      "num_input_tokens_seen": 192081760,
      "step": 8928,
      "time_per_iteration": 3.858457565307617
    },
    {
      "auxiliary_loss_clip": 0.01087394,
      "auxiliary_loss_mlp": 0.01028725,
      "balance_loss_clip": 1.03588283,
      "balance_loss_mlp": 1.01641083,
      "epoch": 0.5368405230722982,
      "flos": 32154629374080.0,
      "grad_norm": 1.757375477986784,
      "language_loss": 0.6329931,
      "learning_rate": 1.8585488706012154e-06,
      "loss": 0.6541543,
      "num_input_tokens_seen": 192101620,
      "step": 8929,
      "time_per_iteration": 2.6298763751983643
    },
    {
      "auxiliary_loss_clip": 0.01100704,
      "auxiliary_loss_mlp": 0.01029564,
      "balance_loss_clip": 1.03688645,
      "balance_loss_mlp": 1.01741052,
      "epoch": 0.5369006463249661,
      "flos": 26247935128320.0,
      "grad_norm": 1.7773420048192499,
      "language_loss": 0.66246402,
      "learning_rate": 1.8581603868894781e-06,
      "loss": 0.68376672,
      "num_input_tokens_seen": 192121805,
      "step": 8930,
      "time_per_iteration": 2.553950309753418
    },
    {
      "auxiliary_loss_clip": 0.01063294,
      "auxiliary_loss_mlp": 0.01029219,
      "balance_loss_clip": 1.03907895,
      "balance_loss_mlp": 1.01625443,
      "epoch": 0.5369607695776342,
      "flos": 26211306234240.0,
      "grad_norm": 1.5126561385389086,
      "language_loss": 0.67342651,
      "learning_rate": 1.8577719085563136e-06,
      "loss": 0.69435161,
      "num_input_tokens_seen": 192141765,
      "step": 8931,
      "time_per_iteration": 2.626486301422119
    },
    {
      "auxiliary_loss_clip": 0.01062837,
      "auxiliary_loss_mlp": 0.01033273,
      "balance_loss_clip": 1.03529286,
      "balance_loss_mlp": 1.01915264,
      "epoch": 0.5370208928303021,
      "flos": 25009017177600.0,
      "grad_norm": 1.7147059236887303,
      "language_loss": 0.75593829,
      "learning_rate": 1.8573834356164525e-06,
      "loss": 0.7768994,
      "num_input_tokens_seen": 192161560,
      "step": 8932,
      "time_per_iteration": 2.622781753540039
    },
    {
      "auxiliary_loss_clip": 0.01086127,
      "auxiliary_loss_mlp": 0.01036937,
      "balance_loss_clip": 1.03999209,
      "balance_loss_mlp": 1.02280521,
      "epoch": 0.5370810160829701,
      "flos": 31792147274880.0,
      "grad_norm": 2.1614863671277074,
      "language_loss": 0.65837419,
      "learning_rate": 1.8569949680846261e-06,
      "loss": 0.67960489,
      "num_input_tokens_seen": 192180190,
      "step": 8933,
      "time_per_iteration": 2.6252777576446533
    },
    {
      "auxiliary_loss_clip": 0.01096774,
      "auxiliary_loss_mlp": 0.00787517,
      "balance_loss_clip": 1.04176331,
      "balance_loss_mlp": 1.01287794,
      "epoch": 0.537141139335638,
      "flos": 23842602829440.0,
      "grad_norm": 1.623774217188273,
      "language_loss": 0.8308084,
      "learning_rate": 1.856606505975565e-06,
      "loss": 0.84965134,
      "num_input_tokens_seen": 192198855,
      "step": 8934,
      "time_per_iteration": 2.532641649246216
    },
    {
      "auxiliary_loss_clip": 0.01074536,
      "auxiliary_loss_mlp": 0.01033699,
      "balance_loss_clip": 1.0359751,
      "balance_loss_mlp": 1.01999605,
      "epoch": 0.537201262588306,
      "flos": 18508826511360.0,
      "grad_norm": 1.7227510801660562,
      "language_loss": 0.79553771,
      "learning_rate": 1.856218049303999e-06,
      "loss": 0.81662011,
      "num_input_tokens_seen": 192216555,
      "step": 8935,
      "time_per_iteration": 2.5425782203674316
    },
    {
      "auxiliary_loss_clip": 0.01100116,
      "auxiliary_loss_mlp": 0.01040439,
      "balance_loss_clip": 1.03709674,
      "balance_loss_mlp": 1.02776098,
      "epoch": 0.537261385840974,
      "flos": 25662950231040.0,
      "grad_norm": 1.7740954502005102,
      "language_loss": 0.8393265,
      "learning_rate": 1.855829598084659e-06,
      "loss": 0.86073202,
      "num_input_tokens_seen": 192236910,
      "step": 8936,
      "time_per_iteration": 2.5221385955810547
    },
    {
      "auxiliary_loss_clip": 0.01078764,
      "auxiliary_loss_mlp": 0.01030945,
      "balance_loss_clip": 1.03781295,
      "balance_loss_mlp": 1.01848817,
      "epoch": 0.537321509093642,
      "flos": 40735017406080.0,
      "grad_norm": 1.4968376646867219,
      "language_loss": 0.72594219,
      "learning_rate": 1.8554411523322754e-06,
      "loss": 0.74703932,
      "num_input_tokens_seen": 192260790,
      "step": 8937,
      "time_per_iteration": 2.733687162399292
    },
    {
      "auxiliary_loss_clip": 0.01086074,
      "auxiliary_loss_mlp": 0.01036449,
      "balance_loss_clip": 1.03420961,
      "balance_loss_mlp": 1.02166748,
      "epoch": 0.53738163234631,
      "flos": 17238487138560.0,
      "grad_norm": 2.9523899072059896,
      "language_loss": 0.81532574,
      "learning_rate": 1.8550527120615778e-06,
      "loss": 0.83655095,
      "num_input_tokens_seen": 192277230,
      "step": 8938,
      "time_per_iteration": 2.5704145431518555
    },
    {
      "auxiliary_loss_clip": 0.0111633,
      "auxiliary_loss_mlp": 0.01032924,
      "balance_loss_clip": 1.03881454,
      "balance_loss_mlp": 1.02070522,
      "epoch": 0.5374417555989779,
      "flos": 12821977457280.0,
      "grad_norm": 2.581111830647663,
      "language_loss": 0.80771393,
      "learning_rate": 1.8546642772872957e-06,
      "loss": 0.82920647,
      "num_input_tokens_seen": 192292840,
      "step": 8939,
      "time_per_iteration": 2.4279019832611084
    },
    {
      "auxiliary_loss_clip": 0.0100938,
      "auxiliary_loss_mlp": 0.0100379,
      "balance_loss_clip": 1.01319385,
      "balance_loss_mlp": 1.00222874,
      "epoch": 0.5375018788516459,
      "flos": 67256018703360.0,
      "grad_norm": 0.8217244339751164,
      "language_loss": 0.524607,
      "learning_rate": 1.8542758480241589e-06,
      "loss": 0.54473871,
      "num_input_tokens_seen": 192358240,
      "step": 8940,
      "time_per_iteration": 3.160722255706787
    },
    {
      "auxiliary_loss_clip": 0.01072641,
      "auxiliary_loss_mlp": 0.01027153,
      "balance_loss_clip": 1.03677011,
      "balance_loss_mlp": 1.01455832,
      "epoch": 0.5375620021043138,
      "flos": 18114168804480.0,
      "grad_norm": 2.1731706115368032,
      "language_loss": 0.71649551,
      "learning_rate": 1.8538874242868965e-06,
      "loss": 0.73749346,
      "num_input_tokens_seen": 192377370,
      "step": 8941,
      "time_per_iteration": 2.5893988609313965
    },
    {
      "auxiliary_loss_clip": 0.0108598,
      "auxiliary_loss_mlp": 0.01028551,
      "balance_loss_clip": 1.03660941,
      "balance_loss_mlp": 1.01600993,
      "epoch": 0.5376221253569818,
      "flos": 23149383275520.0,
      "grad_norm": 1.6273562073356052,
      "language_loss": 0.79512596,
      "learning_rate": 1.853499006090237e-06,
      "loss": 0.81627119,
      "num_input_tokens_seen": 192396450,
      "step": 8942,
      "time_per_iteration": 2.558892250061035
    },
    {
      "auxiliary_loss_clip": 0.01115638,
      "auxiliary_loss_mlp": 0.010362,
      "balance_loss_clip": 1.03928542,
      "balance_loss_mlp": 1.02289057,
      "epoch": 0.5376822486096497,
      "flos": 29972302663680.0,
      "grad_norm": 1.9206287849499029,
      "language_loss": 0.6984508,
      "learning_rate": 1.853110593448911e-06,
      "loss": 0.71996915,
      "num_input_tokens_seen": 192417390,
      "step": 8943,
      "time_per_iteration": 2.5439293384552
    },
    {
      "auxiliary_loss_clip": 0.01026858,
      "auxiliary_loss_mlp": 0.01002389,
      "balance_loss_clip": 1.01146913,
      "balance_loss_mlp": 1.00103045,
      "epoch": 0.5377423718623178,
      "flos": 54168950874240.0,
      "grad_norm": 0.7956089372104103,
      "language_loss": 0.59615719,
      "learning_rate": 1.852722186377645e-06,
      "loss": 0.61644965,
      "num_input_tokens_seen": 192478060,
      "step": 8944,
      "time_per_iteration": 3.1171228885650635
    },
    {
      "auxiliary_loss_clip": 0.01069861,
      "auxiliary_loss_mlp": 0.01032711,
      "balance_loss_clip": 1.04065657,
      "balance_loss_mlp": 1.01824462,
      "epoch": 0.5378024951149857,
      "flos": 23257079228160.0,
      "grad_norm": 1.9195335770468946,
      "language_loss": 0.7720226,
      "learning_rate": 1.852333784891169e-06,
      "loss": 0.79304832,
      "num_input_tokens_seen": 192495985,
      "step": 8945,
      "time_per_iteration": 2.5904903411865234
    },
    {
      "auxiliary_loss_clip": 0.01102723,
      "auxiliary_loss_mlp": 0.01033022,
      "balance_loss_clip": 1.0365026,
      "balance_loss_mlp": 1.02037346,
      "epoch": 0.5378626183676537,
      "flos": 24024095274240.0,
      "grad_norm": 1.6852921736213922,
      "language_loss": 0.68432856,
      "learning_rate": 1.8519453890042112e-06,
      "loss": 0.70568597,
      "num_input_tokens_seen": 192515445,
      "step": 8946,
      "time_per_iteration": 2.5133869647979736
    },
    {
      "auxiliary_loss_clip": 0.01068995,
      "auxiliary_loss_mlp": 0.01036557,
      "balance_loss_clip": 1.04022777,
      "balance_loss_mlp": 1.02364707,
      "epoch": 0.5379227416203216,
      "flos": 27161789973120.0,
      "grad_norm": 1.550835129867525,
      "language_loss": 0.76972997,
      "learning_rate": 1.851556998731498e-06,
      "loss": 0.79078549,
      "num_input_tokens_seen": 192536530,
      "step": 8947,
      "time_per_iteration": 2.627469778060913
    },
    {
      "auxiliary_loss_clip": 0.01101992,
      "auxiliary_loss_mlp": 0.01031235,
      "balance_loss_clip": 1.03842509,
      "balance_loss_mlp": 1.01878965,
      "epoch": 0.5379828648729896,
      "flos": 24681619687680.0,
      "grad_norm": 1.5815137505445422,
      "language_loss": 0.60249668,
      "learning_rate": 1.8511686140877592e-06,
      "loss": 0.62382895,
      "num_input_tokens_seen": 192556075,
      "step": 8948,
      "time_per_iteration": 2.554457426071167
    },
    {
      "auxiliary_loss_clip": 0.01075094,
      "auxiliary_loss_mlp": 0.01032037,
      "balance_loss_clip": 1.03831923,
      "balance_loss_mlp": 1.01931107,
      "epoch": 0.5380429881256577,
      "flos": 22523280284160.0,
      "grad_norm": 1.857736213339742,
      "language_loss": 0.79527497,
      "learning_rate": 1.8507802350877205e-06,
      "loss": 0.81634629,
      "num_input_tokens_seen": 192575535,
      "step": 8949,
      "time_per_iteration": 2.564959764480591
    },
    {
      "auxiliary_loss_clip": 0.01071397,
      "auxiliary_loss_mlp": 0.01041463,
      "balance_loss_clip": 1.03570783,
      "balance_loss_mlp": 1.02657342,
      "epoch": 0.5381031113783256,
      "flos": 26979543342720.0,
      "grad_norm": 1.6886043704165576,
      "language_loss": 0.77723074,
      "learning_rate": 1.850391861746111e-06,
      "loss": 0.79835927,
      "num_input_tokens_seen": 192594490,
      "step": 8950,
      "time_per_iteration": 2.597773790359497
    },
    {
      "auxiliary_loss_clip": 0.0109415,
      "auxiliary_loss_mlp": 0.01029314,
      "balance_loss_clip": 1.03973591,
      "balance_loss_mlp": 1.01767325,
      "epoch": 0.5381632346309936,
      "flos": 24754087376640.0,
      "grad_norm": 1.6693721306578446,
      "language_loss": 0.72677785,
      "learning_rate": 1.8500034940776573e-06,
      "loss": 0.74801248,
      "num_input_tokens_seen": 192615650,
      "step": 8951,
      "time_per_iteration": 2.5851144790649414
    },
    {
      "auxiliary_loss_clip": 0.0111338,
      "auxiliary_loss_mlp": 0.00788854,
      "balance_loss_clip": 1.03822589,
      "balance_loss_mlp": 1.01364625,
      "epoch": 0.5382233578836615,
      "flos": 15560058372480.0,
      "grad_norm": 2.0379043118044056,
      "language_loss": 0.74970996,
      "learning_rate": 1.849615132097085e-06,
      "loss": 0.76873231,
      "num_input_tokens_seen": 192633840,
      "step": 8952,
      "time_per_iteration": 2.4450502395629883
    },
    {
      "auxiliary_loss_clip": 0.01091225,
      "auxiliary_loss_mlp": 0.01029464,
      "balance_loss_clip": 1.04296994,
      "balance_loss_mlp": 1.01632094,
      "epoch": 0.5382834811363295,
      "flos": 25084501608960.0,
      "grad_norm": 1.7912685242630695,
      "language_loss": 0.79508352,
      "learning_rate": 1.8492267758191228e-06,
      "loss": 0.81629038,
      "num_input_tokens_seen": 192655890,
      "step": 8953,
      "time_per_iteration": 2.5817384719848633
    },
    {
      "auxiliary_loss_clip": 0.01076072,
      "auxiliary_loss_mlp": 0.01032894,
      "balance_loss_clip": 1.0358057,
      "balance_loss_mlp": 1.01847577,
      "epoch": 0.5383436043889974,
      "flos": 13297901685120.0,
      "grad_norm": 1.9080768860268316,
      "language_loss": 0.80637932,
      "learning_rate": 1.8488384252584964e-06,
      "loss": 0.82746893,
      "num_input_tokens_seen": 192673025,
      "step": 8954,
      "time_per_iteration": 2.5152320861816406
    },
    {
      "auxiliary_loss_clip": 0.01114978,
      "auxiliary_loss_mlp": 0.01032032,
      "balance_loss_clip": 1.04075968,
      "balance_loss_mlp": 1.01862144,
      "epoch": 0.5384037276416654,
      "flos": 23039388852480.0,
      "grad_norm": 1.847672985210783,
      "language_loss": 0.76295984,
      "learning_rate": 1.8484500804299318e-06,
      "loss": 0.78442997,
      "num_input_tokens_seen": 192692190,
      "step": 8955,
      "time_per_iteration": 2.487030029296875
    },
    {
      "auxiliary_loss_clip": 0.01088854,
      "auxiliary_loss_mlp": 0.01035328,
      "balance_loss_clip": 1.03847671,
      "balance_loss_mlp": 1.02229285,
      "epoch": 0.5384638508943334,
      "flos": 20631147552000.0,
      "grad_norm": 1.6215195242212326,
      "language_loss": 0.78587997,
      "learning_rate": 1.8480617413481557e-06,
      "loss": 0.80712181,
      "num_input_tokens_seen": 192710380,
      "step": 8956,
      "time_per_iteration": 3.9153616428375244
    },
    {
      "auxiliary_loss_clip": 0.01020436,
      "auxiliary_loss_mlp": 0.01001077,
      "balance_loss_clip": 1.01385522,
      "balance_loss_mlp": 0.99981892,
      "epoch": 0.5385239741470014,
      "flos": 66737683491840.0,
      "grad_norm": 0.8608245383155181,
      "language_loss": 0.63458884,
      "learning_rate": 1.8476734080278932e-06,
      "loss": 0.65480399,
      "num_input_tokens_seen": 192768995,
      "step": 8957,
      "time_per_iteration": 3.069251537322998
    },
    {
      "auxiliary_loss_clip": 0.01004868,
      "auxiliary_loss_mlp": 0.01001626,
      "balance_loss_clip": 1.02332425,
      "balance_loss_mlp": 1.00021935,
      "epoch": 0.5385840973996693,
      "flos": 64716058229760.0,
      "grad_norm": 0.7004818267151707,
      "language_loss": 0.51620746,
      "learning_rate": 1.8472850804838705e-06,
      "loss": 0.53627241,
      "num_input_tokens_seen": 192825585,
      "step": 8958,
      "time_per_iteration": 3.2046284675598145
    },
    {
      "auxiliary_loss_clip": 0.01107496,
      "auxiliary_loss_mlp": 0.01030644,
      "balance_loss_clip": 1.04311562,
      "balance_loss_mlp": 1.01651788,
      "epoch": 0.5386442206523373,
      "flos": 26141783460480.0,
      "grad_norm": 2.7832927916946906,
      "language_loss": 0.77180469,
      "learning_rate": 1.8468967587308128e-06,
      "loss": 0.79318607,
      "num_input_tokens_seen": 192847335,
      "step": 8959,
      "time_per_iteration": 3.943035840988159
    },
    {
      "auxiliary_loss_clip": 0.0106982,
      "auxiliary_loss_mlp": 0.01026804,
      "balance_loss_clip": 1.04172206,
      "balance_loss_mlp": 1.01399469,
      "epoch": 0.5387043439050052,
      "flos": 18251849635200.0,
      "grad_norm": 1.9450873835299438,
      "language_loss": 0.83387446,
      "learning_rate": 1.8465084427834455e-06,
      "loss": 0.85484076,
      "num_input_tokens_seen": 192862205,
      "step": 8960,
      "time_per_iteration": 2.5648298263549805
    },
    {
      "auxiliary_loss_clip": 0.0110176,
      "auxiliary_loss_mlp": 0.01032006,
      "balance_loss_clip": 1.03888476,
      "balance_loss_mlp": 1.01944757,
      "epoch": 0.5387644671576732,
      "flos": 29788296266880.0,
      "grad_norm": 1.5138238534573276,
      "language_loss": 0.78271234,
      "learning_rate": 1.8461201326564933e-06,
      "loss": 0.80404997,
      "num_input_tokens_seen": 192883695,
      "step": 8961,
      "time_per_iteration": 2.5684401988983154
    },
    {
      "auxiliary_loss_clip": 0.01079737,
      "auxiliary_loss_mlp": 0.01031935,
      "balance_loss_clip": 1.03799152,
      "balance_loss_mlp": 1.01956749,
      "epoch": 0.5388245904103413,
      "flos": 22374466237440.0,
      "grad_norm": 1.6529682027253239,
      "language_loss": 0.84323907,
      "learning_rate": 1.845731828364681e-06,
      "loss": 0.8643558,
      "num_input_tokens_seen": 192900190,
      "step": 8962,
      "time_per_iteration": 2.539980173110962
    },
    {
      "auxiliary_loss_clip": 0.01014657,
      "auxiliary_loss_mlp": 0.01004993,
      "balance_loss_clip": 1.01684284,
      "balance_loss_mlp": 1.00387287,
      "epoch": 0.5388847136630092,
      "flos": 69807794751360.0,
      "grad_norm": 0.7302228009693614,
      "language_loss": 0.54185164,
      "learning_rate": 1.8453435299227333e-06,
      "loss": 0.56204808,
      "num_input_tokens_seen": 192958675,
      "step": 8963,
      "time_per_iteration": 4.494645118713379
    },
    {
      "auxiliary_loss_clip": 0.01016213,
      "auxiliary_loss_mlp": 0.01005063,
      "balance_loss_clip": 1.01070714,
      "balance_loss_mlp": 1.00386524,
      "epoch": 0.5389448369156772,
      "flos": 69822303845760.0,
      "grad_norm": 0.8032540861171646,
      "language_loss": 0.63396126,
      "learning_rate": 1.8449552373453744e-06,
      "loss": 0.65417403,
      "num_input_tokens_seen": 193033135,
      "step": 8964,
      "time_per_iteration": 3.189657688140869
    },
    {
      "auxiliary_loss_clip": 0.01054799,
      "auxiliary_loss_mlp": 0.01027984,
      "balance_loss_clip": 1.03819358,
      "balance_loss_mlp": 1.01501417,
      "epoch": 0.5390049601683451,
      "flos": 31722444933120.0,
      "grad_norm": 1.5310846577892065,
      "language_loss": 0.70003909,
      "learning_rate": 1.8445669506473287e-06,
      "loss": 0.72086692,
      "num_input_tokens_seen": 193055570,
      "step": 8965,
      "time_per_iteration": 2.7072057723999023
    },
    {
      "auxiliary_loss_clip": 0.01086453,
      "auxiliary_loss_mlp": 0.00788173,
      "balance_loss_clip": 1.03756237,
      "balance_loss_mlp": 1.01005554,
      "epoch": 0.5390650834210131,
      "flos": 18113486446080.0,
      "grad_norm": 2.0430118499214935,
      "language_loss": 0.82317597,
      "learning_rate": 1.8441786698433192e-06,
      "loss": 0.84192222,
      "num_input_tokens_seen": 193073120,
      "step": 8966,
      "time_per_iteration": 2.5024240016937256
    },
    {
      "auxiliary_loss_clip": 0.01112712,
      "auxiliary_loss_mlp": 0.01030779,
      "balance_loss_clip": 1.03995252,
      "balance_loss_mlp": 1.01779699,
      "epoch": 0.539125206673681,
      "flos": 17416711445760.0,
      "grad_norm": 3.2694762089253273,
      "language_loss": 0.72555935,
      "learning_rate": 1.8437903949480706e-06,
      "loss": 0.74699432,
      "num_input_tokens_seen": 193090105,
      "step": 8967,
      "time_per_iteration": 3.8562052249908447
    },
    {
      "auxiliary_loss_clip": 0.01086516,
      "auxiliary_loss_mlp": 0.01028724,
      "balance_loss_clip": 1.03353548,
      "balance_loss_mlp": 1.01681542,
      "epoch": 0.539185329926349,
      "flos": 22198935450240.0,
      "grad_norm": 2.4174754686683264,
      "language_loss": 0.81935549,
      "learning_rate": 1.8434021259763065e-06,
      "loss": 0.84050786,
      "num_input_tokens_seen": 193109325,
      "step": 8968,
      "time_per_iteration": 2.5234062671661377
    },
    {
      "auxiliary_loss_clip": 0.01083209,
      "auxiliary_loss_mlp": 0.01030566,
      "balance_loss_clip": 1.03888309,
      "balance_loss_mlp": 1.01700568,
      "epoch": 0.539245453179017,
      "flos": 21434397442560.0,
      "grad_norm": 1.4498448325163416,
      "language_loss": 0.73872691,
      "learning_rate": 1.8430138629427484e-06,
      "loss": 0.75986463,
      "num_input_tokens_seen": 193130595,
      "step": 8969,
      "time_per_iteration": 2.5730488300323486
    },
    {
      "auxiliary_loss_clip": 0.01078993,
      "auxiliary_loss_mlp": 0.00787802,
      "balance_loss_clip": 1.03503621,
      "balance_loss_mlp": 1.01004243,
      "epoch": 0.539305576431685,
      "flos": 20735000749440.0,
      "grad_norm": 1.8559793787074723,
      "language_loss": 0.82414252,
      "learning_rate": 1.8426256058621205e-06,
      "loss": 0.84281051,
      "num_input_tokens_seen": 193148930,
      "step": 8970,
      "time_per_iteration": 2.5401108264923096
    },
    {
      "auxiliary_loss_clip": 0.01087227,
      "auxiliary_loss_mlp": 0.01026178,
      "balance_loss_clip": 1.03720665,
      "balance_loss_mlp": 1.01387584,
      "epoch": 0.5393656996843529,
      "flos": 30920452018560.0,
      "grad_norm": 1.3538298827353374,
      "language_loss": 0.75321865,
      "learning_rate": 1.842237354749146e-06,
      "loss": 0.77435267,
      "num_input_tokens_seen": 193170140,
      "step": 8971,
      "time_per_iteration": 2.599991798400879
    },
    {
      "auxiliary_loss_clip": 0.0102668,
      "auxiliary_loss_mlp": 0.01022115,
      "balance_loss_clip": 1.01067686,
      "balance_loss_mlp": 1.02049339,
      "epoch": 0.5394258229370209,
      "flos": 50317781351040.0,
      "grad_norm": 0.8980780205792912,
      "language_loss": 0.60345495,
      "learning_rate": 1.8418491096185465e-06,
      "loss": 0.62394285,
      "num_input_tokens_seen": 193227235,
      "step": 8972,
      "time_per_iteration": 3.108747959136963
    },
    {
      "auxiliary_loss_clip": 0.01102131,
      "auxiliary_loss_mlp": 0.01039127,
      "balance_loss_clip": 1.03776753,
      "balance_loss_mlp": 1.02593076,
      "epoch": 0.5394859461896888,
      "flos": 25411935012480.0,
      "grad_norm": 1.389896566125859,
      "language_loss": 0.78473043,
      "learning_rate": 1.841460870485045e-06,
      "loss": 0.80614305,
      "num_input_tokens_seen": 193248435,
      "step": 8973,
      "time_per_iteration": 2.5342893600463867
    },
    {
      "auxiliary_loss_clip": 0.0110821,
      "auxiliary_loss_mlp": 0.0103565,
      "balance_loss_clip": 1.03819692,
      "balance_loss_mlp": 1.02122617,
      "epoch": 0.5395460694423568,
      "flos": 25478476957440.0,
      "grad_norm": 2.037819006284775,
      "language_loss": 0.7387737,
      "learning_rate": 1.8410726373633623e-06,
      "loss": 0.7602123,
      "num_input_tokens_seen": 193267490,
      "step": 8974,
      "time_per_iteration": 2.525157928466797
    },
    {
      "auxiliary_loss_clip": 0.01034237,
      "auxiliary_loss_mlp": 0.01005025,
      "balance_loss_clip": 1.00874305,
      "balance_loss_mlp": 1.00373173,
      "epoch": 0.5396061926950249,
      "flos": 53249493507840.0,
      "grad_norm": 0.7354506955113285,
      "language_loss": 0.51044875,
      "learning_rate": 1.8406844102682215e-06,
      "loss": 0.53084135,
      "num_input_tokens_seen": 193326050,
      "step": 8975,
      "time_per_iteration": 3.0601558685302734
    },
    {
      "auxiliary_loss_clip": 0.01100065,
      "auxiliary_loss_mlp": 0.0103608,
      "balance_loss_clip": 1.03755164,
      "balance_loss_mlp": 1.02275801,
      "epoch": 0.5396663159476928,
      "flos": 26725080418560.0,
      "grad_norm": 1.4713408842978375,
      "language_loss": 0.72080517,
      "learning_rate": 1.840296189214344e-06,
      "loss": 0.74216658,
      "num_input_tokens_seen": 193348785,
      "step": 8976,
      "time_per_iteration": 2.56437611579895
    },
    {
      "auxiliary_loss_clip": 0.0109804,
      "auxiliary_loss_mlp": 0.00790177,
      "balance_loss_clip": 1.03591943,
      "balance_loss_mlp": 1.01650476,
      "epoch": 0.5397264392003608,
      "flos": 23253380127360.0,
      "grad_norm": 1.5898124373749882,
      "language_loss": 0.69634992,
      "learning_rate": 1.8399079742164509e-06,
      "loss": 0.71523207,
      "num_input_tokens_seen": 193367080,
      "step": 8977,
      "time_per_iteration": 2.518348217010498
    },
    {
      "auxiliary_loss_clip": 0.01046584,
      "auxiliary_loss_mlp": 0.01033166,
      "balance_loss_clip": 1.0418191,
      "balance_loss_mlp": 1.02044618,
      "epoch": 0.5397865624530287,
      "flos": 18294188791680.0,
      "grad_norm": 1.6349730671587264,
      "language_loss": 0.7273187,
      "learning_rate": 1.8395197652892636e-06,
      "loss": 0.74811614,
      "num_input_tokens_seen": 193383715,
      "step": 8978,
      "time_per_iteration": 2.6040353775024414
    },
    {
      "auxiliary_loss_clip": 0.01085298,
      "auxiliary_loss_mlp": 0.01035912,
      "balance_loss_clip": 1.04603219,
      "balance_loss_mlp": 1.02139211,
      "epoch": 0.5398466857056967,
      "flos": 15297514888320.0,
      "grad_norm": 2.116793839666026,
      "language_loss": 0.74471521,
      "learning_rate": 1.8391315624475028e-06,
      "loss": 0.76592731,
      "num_input_tokens_seen": 193400560,
      "step": 8979,
      "time_per_iteration": 2.5447163581848145
    },
    {
      "auxiliary_loss_clip": 0.01062467,
      "auxiliary_loss_mlp": 0.01044959,
      "balance_loss_clip": 1.03942513,
      "balance_loss_mlp": 1.0299449,
      "epoch": 0.5399068089583646,
      "flos": 17821748183040.0,
      "grad_norm": 2.019653663028868,
      "language_loss": 0.76644987,
      "learning_rate": 1.8387433657058892e-06,
      "loss": 0.7875241,
      "num_input_tokens_seen": 193418680,
      "step": 8980,
      "time_per_iteration": 2.597350597381592
    },
    {
      "auxiliary_loss_clip": 0.0111169,
      "auxiliary_loss_mlp": 0.01032152,
      "balance_loss_clip": 1.03664529,
      "balance_loss_mlp": 1.0199033,
      "epoch": 0.5399669322110326,
      "flos": 27381635164800.0,
      "grad_norm": 1.7517860604137983,
      "language_loss": 0.81925678,
      "learning_rate": 1.8383551750791431e-06,
      "loss": 0.84069514,
      "num_input_tokens_seen": 193439310,
      "step": 8981,
      "time_per_iteration": 2.5185766220092773
    },
    {
      "auxiliary_loss_clip": 0.01103233,
      "auxiliary_loss_mlp": 0.01032914,
      "balance_loss_clip": 1.03661942,
      "balance_loss_mlp": 1.01953864,
      "epoch": 0.5400270554637006,
      "flos": 20449116403200.0,
      "grad_norm": 1.8610331352040506,
      "language_loss": 0.66786838,
      "learning_rate": 1.8379669905819857e-06,
      "loss": 0.68922985,
      "num_input_tokens_seen": 193458115,
      "step": 8982,
      "time_per_iteration": 2.489445209503174
    },
    {
      "auxiliary_loss_clip": 0.0107952,
      "auxiliary_loss_mlp": 0.00786378,
      "balance_loss_clip": 1.04083228,
      "balance_loss_mlp": 1.01105237,
      "epoch": 0.5400871787163686,
      "flos": 21689578638720.0,
      "grad_norm": 1.516081190680007,
      "language_loss": 0.82631624,
      "learning_rate": 1.8375788122291358e-06,
      "loss": 0.84497523,
      "num_input_tokens_seen": 193477365,
      "step": 8983,
      "time_per_iteration": 2.5980191230773926
    },
    {
      "auxiliary_loss_clip": 0.01063665,
      "auxiliary_loss_mlp": 0.01038051,
      "balance_loss_clip": 1.03467512,
      "balance_loss_mlp": 1.02483034,
      "epoch": 0.5401473019690365,
      "flos": 19204739585280.0,
      "grad_norm": 2.0185699101337375,
      "language_loss": 0.70384634,
      "learning_rate": 1.8371906400353138e-06,
      "loss": 0.72486347,
      "num_input_tokens_seen": 193495595,
      "step": 8984,
      "time_per_iteration": 2.561084270477295
    },
    {
      "auxiliary_loss_clip": 0.01117203,
      "auxiliary_loss_mlp": 0.01032341,
      "balance_loss_clip": 1.04021561,
      "balance_loss_mlp": 1.01883459,
      "epoch": 0.5402074252217045,
      "flos": 20627376624000.0,
      "grad_norm": 1.6983648034322454,
      "language_loss": 0.8010385,
      "learning_rate": 1.8368024740152386e-06,
      "loss": 0.82253391,
      "num_input_tokens_seen": 193514035,
      "step": 8985,
      "time_per_iteration": 2.474837303161621
    },
    {
      "auxiliary_loss_clip": 0.01067908,
      "auxiliary_loss_mlp": 0.0102828,
      "balance_loss_clip": 1.03586555,
      "balance_loss_mlp": 1.01517916,
      "epoch": 0.5402675484743724,
      "flos": 24973465691520.0,
      "grad_norm": 1.476038245299173,
      "language_loss": 0.78848535,
      "learning_rate": 1.83641431418363e-06,
      "loss": 0.80944729,
      "num_input_tokens_seen": 193535445,
      "step": 8986,
      "time_per_iteration": 2.598332643508911
    },
    {
      "auxiliary_loss_clip": 0.01095431,
      "auxiliary_loss_mlp": 0.01035798,
      "balance_loss_clip": 1.03484893,
      "balance_loss_mlp": 1.0224824,
      "epoch": 0.5403276717270404,
      "flos": 19459022941440.0,
      "grad_norm": 1.665564339634343,
      "language_loss": 0.76875085,
      "learning_rate": 1.8360261605552075e-06,
      "loss": 0.79006314,
      "num_input_tokens_seen": 193554780,
      "step": 8987,
      "time_per_iteration": 2.5021634101867676
    },
    {
      "auxiliary_loss_clip": 0.01087319,
      "auxiliary_loss_mlp": 0.01032048,
      "balance_loss_clip": 1.03954387,
      "balance_loss_mlp": 1.01957834,
      "epoch": 0.5403877949797083,
      "flos": 18442140912000.0,
      "grad_norm": 1.806909544124405,
      "language_loss": 0.70866454,
      "learning_rate": 1.8356380131446887e-06,
      "loss": 0.72985822,
      "num_input_tokens_seen": 193573580,
      "step": 8988,
      "time_per_iteration": 2.5186872482299805
    },
    {
      "auxiliary_loss_clip": 0.01067522,
      "auxiliary_loss_mlp": 0.01036822,
      "balance_loss_clip": 1.04112542,
      "balance_loss_mlp": 1.02333939,
      "epoch": 0.5404479182323764,
      "flos": 28292868316800.0,
      "grad_norm": 2.3624582471215256,
      "language_loss": 0.67279983,
      "learning_rate": 1.8352498719667934e-06,
      "loss": 0.69384325,
      "num_input_tokens_seen": 193590490,
      "step": 8989,
      "time_per_iteration": 2.670578956604004
    },
    {
      "auxiliary_loss_clip": 0.0109984,
      "auxiliary_loss_mlp": 0.01042541,
      "balance_loss_clip": 1.03620303,
      "balance_loss_mlp": 1.02913547,
      "epoch": 0.5405080414850444,
      "flos": 23367325046400.0,
      "grad_norm": 1.5372061292660113,
      "language_loss": 0.77815855,
      "learning_rate": 1.8348617370362399e-06,
      "loss": 0.7995823,
      "num_input_tokens_seen": 193609900,
      "step": 8990,
      "time_per_iteration": 2.5062739849090576
    },
    {
      "auxiliary_loss_clip": 0.01096009,
      "auxiliary_loss_mlp": 0.01027886,
      "balance_loss_clip": 1.03445649,
      "balance_loss_mlp": 1.0161978,
      "epoch": 0.5405681647377123,
      "flos": 21106425335040.0,
      "grad_norm": 1.6943494911565302,
      "language_loss": 0.69358146,
      "learning_rate": 1.834473608367745e-06,
      "loss": 0.71482038,
      "num_input_tokens_seen": 193629775,
      "step": 8991,
      "time_per_iteration": 2.4924204349517822
    },
    {
      "auxiliary_loss_clip": 0.0105727,
      "auxiliary_loss_mlp": 0.01037648,
      "balance_loss_clip": 1.0374999,
      "balance_loss_mlp": 1.02374196,
      "epoch": 0.5406282879903803,
      "flos": 20449188230400.0,
      "grad_norm": 1.734297729509414,
      "language_loss": 0.7639643,
      "learning_rate": 1.8340854859760277e-06,
      "loss": 0.78491354,
      "num_input_tokens_seen": 193648070,
      "step": 8992,
      "time_per_iteration": 2.6205179691314697
    },
    {
      "auxiliary_loss_clip": 0.01086617,
      "auxiliary_loss_mlp": 0.01039318,
      "balance_loss_clip": 1.03492975,
      "balance_loss_mlp": 1.02512634,
      "epoch": 0.5406884112430482,
      "flos": 14209493973120.0,
      "grad_norm": 2.8342517261033646,
      "language_loss": 0.7647841,
      "learning_rate": 1.8336973698758056e-06,
      "loss": 0.78604352,
      "num_input_tokens_seen": 193665060,
      "step": 8993,
      "time_per_iteration": 2.5117013454437256
    },
    {
      "auxiliary_loss_clip": 0.01097053,
      "auxiliary_loss_mlp": 0.01033006,
      "balance_loss_clip": 1.03522456,
      "balance_loss_mlp": 1.02035832,
      "epoch": 0.5407485344957162,
      "flos": 23875568536320.0,
      "grad_norm": 1.6092732413014166,
      "language_loss": 0.70641994,
      "learning_rate": 1.8333092600817959e-06,
      "loss": 0.72772062,
      "num_input_tokens_seen": 193683620,
      "step": 8994,
      "time_per_iteration": 3.8526992797851562
    },
    {
      "auxiliary_loss_clip": 0.01098969,
      "auxiliary_loss_mlp": 0.01031094,
      "balance_loss_clip": 1.03746009,
      "balance_loss_mlp": 1.01695561,
      "epoch": 0.5408086577483842,
      "flos": 23148485435520.0,
      "grad_norm": 1.887956651489494,
      "language_loss": 0.74918735,
      "learning_rate": 1.8329211566087157e-06,
      "loss": 0.77048802,
      "num_input_tokens_seen": 193702990,
      "step": 8995,
      "time_per_iteration": 2.503073215484619
    },
    {
      "auxiliary_loss_clip": 0.01096647,
      "auxiliary_loss_mlp": 0.0103194,
      "balance_loss_clip": 1.03555882,
      "balance_loss_mlp": 1.019894,
      "epoch": 0.5408687810010522,
      "flos": 18771046773120.0,
      "grad_norm": 1.9124453428725674,
      "language_loss": 0.73430395,
      "learning_rate": 1.832533059471282e-06,
      "loss": 0.75558984,
      "num_input_tokens_seen": 193721785,
      "step": 8996,
      "time_per_iteration": 2.481541633605957
    },
    {
      "auxiliary_loss_clip": 0.01065822,
      "auxiliary_loss_mlp": 0.01033418,
      "balance_loss_clip": 1.03668904,
      "balance_loss_mlp": 1.02109778,
      "epoch": 0.5409289042537201,
      "flos": 13881557779200.0,
      "grad_norm": 5.093146298279868,
      "language_loss": 0.72760212,
      "learning_rate": 1.8321449686842115e-06,
      "loss": 0.74859452,
      "num_input_tokens_seen": 193740315,
      "step": 8997,
      "time_per_iteration": 2.5911202430725098
    },
    {
      "auxiliary_loss_clip": 0.01111267,
      "auxiliary_loss_mlp": 0.01031842,
      "balance_loss_clip": 1.03773379,
      "balance_loss_mlp": 1.01890802,
      "epoch": 0.5409890275063881,
      "flos": 14465357527680.0,
      "grad_norm": 2.205689543848071,
      "language_loss": 0.71490097,
      "learning_rate": 1.8317568842622207e-06,
      "loss": 0.73633206,
      "num_input_tokens_seen": 193757580,
      "step": 8998,
      "time_per_iteration": 3.8622753620147705
    },
    {
      "auxiliary_loss_clip": 0.01077485,
      "auxiliary_loss_mlp": 0.01037496,
      "balance_loss_clip": 1.03660285,
      "balance_loss_mlp": 1.02493167,
      "epoch": 0.541049150759056,
      "flos": 48977449349760.0,
      "grad_norm": 1.5353029117854466,
      "language_loss": 0.70503473,
      "learning_rate": 1.8313688062200256e-06,
      "loss": 0.72618449,
      "num_input_tokens_seen": 193780965,
      "step": 8999,
      "time_per_iteration": 2.7941699028015137
    },
    {
      "auxiliary_loss_clip": 0.01087633,
      "auxiliary_loss_mlp": 0.01031613,
      "balance_loss_clip": 1.04038405,
      "balance_loss_mlp": 1.01855409,
      "epoch": 0.541109274011724,
      "flos": 18147601388160.0,
      "grad_norm": 2.296375129540677,
      "language_loss": 0.80612016,
      "learning_rate": 1.8309807345723422e-06,
      "loss": 0.82731259,
      "num_input_tokens_seen": 193797855,
      "step": 9000,
      "time_per_iteration": 2.5019028186798096
    },
    {
      "auxiliary_loss_clip": 0.01064081,
      "auxiliary_loss_mlp": 0.01032041,
      "balance_loss_clip": 1.03634465,
      "balance_loss_mlp": 1.01900506,
      "epoch": 0.541169397264392,
      "flos": 20522553759360.0,
      "grad_norm": 1.9877262960966249,
      "language_loss": 0.72775459,
      "learning_rate": 1.8305926693338863e-06,
      "loss": 0.74871582,
      "num_input_tokens_seen": 193817375,
      "step": 9001,
      "time_per_iteration": 3.960714340209961
    },
    {
      "auxiliary_loss_clip": 0.01079827,
      "auxiliary_loss_mlp": 0.01030612,
      "balance_loss_clip": 1.03557348,
      "balance_loss_mlp": 1.01694453,
      "epoch": 0.54122952051706,
      "flos": 20044043752320.0,
      "grad_norm": 2.328365229761933,
      "language_loss": 0.85296762,
      "learning_rate": 1.8302046105193734e-06,
      "loss": 0.87407196,
      "num_input_tokens_seen": 193832205,
      "step": 9002,
      "time_per_iteration": 2.533674716949463
    },
    {
      "auxiliary_loss_clip": 0.01063926,
      "auxiliary_loss_mlp": 0.01027372,
      "balance_loss_clip": 1.03707337,
      "balance_loss_mlp": 1.01615477,
      "epoch": 0.541289643769728,
      "flos": 19062246332160.0,
      "grad_norm": 1.7180419713044937,
      "language_loss": 0.77926314,
      "learning_rate": 1.8298165581435183e-06,
      "loss": 0.80017614,
      "num_input_tokens_seen": 193849830,
      "step": 9003,
      "time_per_iteration": 2.571176767349243
    },
    {
      "auxiliary_loss_clip": 0.01099996,
      "auxiliary_loss_mlp": 0.01030045,
      "balance_loss_clip": 1.03824914,
      "balance_loss_mlp": 1.0170455,
      "epoch": 0.5413497670223959,
      "flos": 22382295402240.0,
      "grad_norm": 1.8245218035359125,
      "language_loss": 0.69655013,
      "learning_rate": 1.8294285122210372e-06,
      "loss": 0.71785051,
      "num_input_tokens_seen": 193869945,
      "step": 9004,
      "time_per_iteration": 2.5167720317840576
    },
    {
      "auxiliary_loss_clip": 0.01026332,
      "auxiliary_loss_mlp": 0.0100922,
      "balance_loss_clip": 1.01177895,
      "balance_loss_mlp": 1.00788534,
      "epoch": 0.5414098902750639,
      "flos": 70031734093440.0,
      "grad_norm": 0.9744414470977771,
      "language_loss": 0.59128577,
      "learning_rate": 1.8290404727666434e-06,
      "loss": 0.61164129,
      "num_input_tokens_seen": 193930860,
      "step": 9005,
      "time_per_iteration": 4.561850070953369
    },
    {
      "auxiliary_loss_clip": 0.0111435,
      "auxiliary_loss_mlp": 0.00787489,
      "balance_loss_clip": 1.03940248,
      "balance_loss_mlp": 1.01237082,
      "epoch": 0.5414700135277318,
      "flos": 21798962530560.0,
      "grad_norm": 1.829621112750986,
      "language_loss": 0.77954936,
      "learning_rate": 1.8286524397950517e-06,
      "loss": 0.79856777,
      "num_input_tokens_seen": 193949075,
      "step": 9006,
      "time_per_iteration": 2.490750551223755
    },
    {
      "auxiliary_loss_clip": 0.01088761,
      "auxiliary_loss_mlp": 0.01031195,
      "balance_loss_clip": 1.03726792,
      "balance_loss_mlp": 1.0201267,
      "epoch": 0.5415301367803999,
      "flos": 16907929251840.0,
      "grad_norm": 1.7304624427593889,
      "language_loss": 0.8300035,
      "learning_rate": 1.8282644133209777e-06,
      "loss": 0.85120308,
      "num_input_tokens_seen": 193967630,
      "step": 9007,
      "time_per_iteration": 2.5465402603149414
    },
    {
      "auxiliary_loss_clip": 0.01099296,
      "auxiliary_loss_mlp": 0.01029115,
      "balance_loss_clip": 1.03999591,
      "balance_loss_mlp": 1.01652646,
      "epoch": 0.5415902600330678,
      "flos": 25704176065920.0,
      "grad_norm": 2.164805792944478,
      "language_loss": 0.67006731,
      "learning_rate": 1.8278763933591334e-06,
      "loss": 0.69135141,
      "num_input_tokens_seen": 193988730,
      "step": 9008,
      "time_per_iteration": 2.5700058937072754
    },
    {
      "auxiliary_loss_clip": 0.01117051,
      "auxiliary_loss_mlp": 0.01030141,
      "balance_loss_clip": 1.03925014,
      "balance_loss_mlp": 1.01607442,
      "epoch": 0.5416503832857358,
      "flos": 19208151377280.0,
      "grad_norm": 1.96650324802902,
      "language_loss": 0.73738062,
      "learning_rate": 1.827488379924234e-06,
      "loss": 0.75885254,
      "num_input_tokens_seen": 194005160,
      "step": 9009,
      "time_per_iteration": 2.4481890201568604
    },
    {
      "auxiliary_loss_clip": 0.01066669,
      "auxiliary_loss_mlp": 0.01037681,
      "balance_loss_clip": 1.04091513,
      "balance_loss_mlp": 1.02398372,
      "epoch": 0.5417105065384037,
      "flos": 12713706887040.0,
      "grad_norm": 2.481950048039745,
      "language_loss": 0.87103724,
      "learning_rate": 1.8271003730309923e-06,
      "loss": 0.89208078,
      "num_input_tokens_seen": 194021700,
      "step": 9010,
      "time_per_iteration": 2.567962884902954
    },
    {
      "auxiliary_loss_clip": 0.0111018,
      "auxiliary_loss_mlp": 0.01032929,
      "balance_loss_clip": 1.03703988,
      "balance_loss_mlp": 1.02060246,
      "epoch": 0.5417706297910717,
      "flos": 30335933998080.0,
      "grad_norm": 2.0290939267670964,
      "language_loss": 0.6419155,
      "learning_rate": 1.826712372694122e-06,
      "loss": 0.66334653,
      "num_input_tokens_seen": 194042620,
      "step": 9011,
      "time_per_iteration": 2.5366978645324707
    },
    {
      "auxiliary_loss_clip": 0.01101676,
      "auxiliary_loss_mlp": 0.01032005,
      "balance_loss_clip": 1.03946483,
      "balance_loss_mlp": 1.01969087,
      "epoch": 0.5418307530437396,
      "flos": 29020992912000.0,
      "grad_norm": 2.4660140687050025,
      "language_loss": 0.79273152,
      "learning_rate": 1.8263243789283362e-06,
      "loss": 0.81406832,
      "num_input_tokens_seen": 194061800,
      "step": 9012,
      "time_per_iteration": 2.5396668910980225
    },
    {
      "auxiliary_loss_clip": 0.01110597,
      "auxiliary_loss_mlp": 0.0102844,
      "balance_loss_clip": 1.03703058,
      "balance_loss_mlp": 1.01566112,
      "epoch": 0.5418908762964076,
      "flos": 16873455173760.0,
      "grad_norm": 1.9026068004204453,
      "language_loss": 0.74460232,
      "learning_rate": 1.8259363917483466e-06,
      "loss": 0.76599264,
      "num_input_tokens_seen": 194079890,
      "step": 9013,
      "time_per_iteration": 2.440497636795044
    },
    {
      "auxiliary_loss_clip": 0.010836,
      "auxiliary_loss_mlp": 0.01030591,
      "balance_loss_clip": 1.03814483,
      "balance_loss_mlp": 1.01754344,
      "epoch": 0.5419509995490756,
      "flos": 18949702043520.0,
      "grad_norm": 1.797670527425618,
      "language_loss": 0.72167557,
      "learning_rate": 1.8255484111688667e-06,
      "loss": 0.74281752,
      "num_input_tokens_seen": 194097625,
      "step": 9014,
      "time_per_iteration": 2.540179967880249
    },
    {
      "auxiliary_loss_clip": 0.01091472,
      "auxiliary_loss_mlp": 0.01032746,
      "balance_loss_clip": 1.03869987,
      "balance_loss_mlp": 1.01978159,
      "epoch": 0.5420111228017436,
      "flos": 18077719478400.0,
      "grad_norm": 1.5851409787372044,
      "language_loss": 0.80587494,
      "learning_rate": 1.8251604372046085e-06,
      "loss": 0.82711709,
      "num_input_tokens_seen": 194116055,
      "step": 9015,
      "time_per_iteration": 2.4949209690093994
    },
    {
      "auxiliary_loss_clip": 0.01105458,
      "auxiliary_loss_mlp": 0.01037514,
      "balance_loss_clip": 1.0389322,
      "balance_loss_mlp": 1.02438939,
      "epoch": 0.5420712460544116,
      "flos": 19061779455360.0,
      "grad_norm": 2.1876972861777606,
      "language_loss": 0.81075823,
      "learning_rate": 1.8247724698702843e-06,
      "loss": 0.83218789,
      "num_input_tokens_seen": 194130365,
      "step": 9016,
      "time_per_iteration": 2.496673345565796
    },
    {
      "auxiliary_loss_clip": 0.01112133,
      "auxiliary_loss_mlp": 0.01028427,
      "balance_loss_clip": 1.03839231,
      "balance_loss_mlp": 1.01629138,
      "epoch": 0.5421313693070795,
      "flos": 18187103370240.0,
      "grad_norm": 1.5799604000639118,
      "language_loss": 0.81194806,
      "learning_rate": 1.8243845091806053e-06,
      "loss": 0.8333537,
      "num_input_tokens_seen": 194148975,
      "step": 9017,
      "time_per_iteration": 2.4462225437164307
    },
    {
      "auxiliary_loss_clip": 0.01110453,
      "auxiliary_loss_mlp": 0.01033291,
      "balance_loss_clip": 1.0386529,
      "balance_loss_mlp": 1.02053595,
      "epoch": 0.5421914925597475,
      "flos": 13005947940480.0,
      "grad_norm": 1.626466044312829,
      "language_loss": 0.77270234,
      "learning_rate": 1.8239965551502837e-06,
      "loss": 0.79413986,
      "num_input_tokens_seen": 194167185,
      "step": 9018,
      "time_per_iteration": 2.445436477661133
    },
    {
      "auxiliary_loss_clip": 0.0111323,
      "auxiliary_loss_mlp": 0.01034975,
      "balance_loss_clip": 1.03649855,
      "balance_loss_mlp": 1.02199888,
      "epoch": 0.5422516158124154,
      "flos": 46758457831680.0,
      "grad_norm": 1.4598934170250157,
      "language_loss": 0.66739309,
      "learning_rate": 1.8236086077940303e-06,
      "loss": 0.6888752,
      "num_input_tokens_seen": 194192840,
      "step": 9019,
      "time_per_iteration": 2.716231346130371
    },
    {
      "auxiliary_loss_clip": 0.0109325,
      "auxiliary_loss_mlp": 0.01028046,
      "balance_loss_clip": 1.03508842,
      "balance_loss_mlp": 1.01579106,
      "epoch": 0.5423117390650835,
      "flos": 31758642864000.0,
      "grad_norm": 2.0181835289150447,
      "language_loss": 0.69360793,
      "learning_rate": 1.8232206671265555e-06,
      "loss": 0.71482092,
      "num_input_tokens_seen": 194213150,
      "step": 9020,
      "time_per_iteration": 2.5642106533050537
    },
    {
      "auxiliary_loss_clip": 0.01074971,
      "auxiliary_loss_mlp": 0.01034782,
      "balance_loss_clip": 1.03558731,
      "balance_loss_mlp": 1.02238464,
      "epoch": 0.5423718623177514,
      "flos": 27201974313600.0,
      "grad_norm": 1.439624582948833,
      "language_loss": 0.80503327,
      "learning_rate": 1.8228327331625717e-06,
      "loss": 0.82613081,
      "num_input_tokens_seen": 194234665,
      "step": 9021,
      "time_per_iteration": 2.6191725730895996
    },
    {
      "auxiliary_loss_clip": 0.01064879,
      "auxiliary_loss_mlp": 0.01038306,
      "balance_loss_clip": 1.03731728,
      "balance_loss_mlp": 1.0250442,
      "epoch": 0.5424319855704194,
      "flos": 23546447193600.0,
      "grad_norm": 1.6671805105907298,
      "language_loss": 0.78964758,
      "learning_rate": 1.822444805916788e-06,
      "loss": 0.81067944,
      "num_input_tokens_seen": 194253790,
      "step": 9022,
      "time_per_iteration": 2.5976905822753906
    },
    {
      "auxiliary_loss_clip": 0.01082774,
      "auxiliary_loss_mlp": 0.00787268,
      "balance_loss_clip": 1.03949833,
      "balance_loss_mlp": 1.00947332,
      "epoch": 0.5424921088230873,
      "flos": 26615624699520.0,
      "grad_norm": 1.5419661356733698,
      "language_loss": 0.81814778,
      "learning_rate": 1.822056885403915e-06,
      "loss": 0.83684826,
      "num_input_tokens_seen": 194274950,
      "step": 9023,
      "time_per_iteration": 2.5814690589904785
    },
    {
      "auxiliary_loss_clip": 0.01098211,
      "auxiliary_loss_mlp": 0.01028731,
      "balance_loss_clip": 1.0401324,
      "balance_loss_mlp": 1.01627421,
      "epoch": 0.5425522320757553,
      "flos": 23586811102080.0,
      "grad_norm": 3.2072712072118685,
      "language_loss": 0.71036136,
      "learning_rate": 1.8216689716386627e-06,
      "loss": 0.7316308,
      "num_input_tokens_seen": 194296155,
      "step": 9024,
      "time_per_iteration": 2.526933431625366
    },
    {
      "auxiliary_loss_clip": 0.01100473,
      "auxiliary_loss_mlp": 0.01029757,
      "balance_loss_clip": 1.03571081,
      "balance_loss_mlp": 1.01740086,
      "epoch": 0.5426123553284232,
      "flos": 30592264429440.0,
      "grad_norm": 1.6955890752835932,
      "language_loss": 0.65148836,
      "learning_rate": 1.8212810646357405e-06,
      "loss": 0.67279065,
      "num_input_tokens_seen": 194318025,
      "step": 9025,
      "time_per_iteration": 2.567335844039917
    },
    {
      "auxiliary_loss_clip": 0.010888,
      "auxiliary_loss_mlp": 0.00785393,
      "balance_loss_clip": 1.04107428,
      "balance_loss_mlp": 1.00912964,
      "epoch": 0.5426724785810912,
      "flos": 12495118671360.0,
      "grad_norm": 1.8892355541168753,
      "language_loss": 0.73817015,
      "learning_rate": 1.8208931644098591e-06,
      "loss": 0.75691211,
      "num_input_tokens_seen": 194336150,
      "step": 9026,
      "time_per_iteration": 2.5372366905212402
    },
    {
      "auxiliary_loss_clip": 0.01088316,
      "auxiliary_loss_mlp": 0.0104074,
      "balance_loss_clip": 1.03508484,
      "balance_loss_mlp": 1.02624416,
      "epoch": 0.5427326018337592,
      "flos": 26064611089920.0,
      "grad_norm": 1.6224016608846457,
      "language_loss": 0.78815281,
      "learning_rate": 1.8205052709757265e-06,
      "loss": 0.80944335,
      "num_input_tokens_seen": 194355980,
      "step": 9027,
      "time_per_iteration": 2.5523788928985596
    },
    {
      "auxiliary_loss_clip": 0.0101166,
      "auxiliary_loss_mlp": 0.01000587,
      "balance_loss_clip": 1.02074337,
      "balance_loss_mlp": 0.99937135,
      "epoch": 0.5427927250864272,
      "flos": 65984745576960.0,
      "grad_norm": 0.7448524556632176,
      "language_loss": 0.56509846,
      "learning_rate": 1.8201173843480515e-06,
      "loss": 0.58522093,
      "num_input_tokens_seen": 194422660,
      "step": 9028,
      "time_per_iteration": 3.2044901847839355
    },
    {
      "auxiliary_loss_clip": 0.01070536,
      "auxiliary_loss_mlp": 0.01028012,
      "balance_loss_clip": 1.04061913,
      "balance_loss_mlp": 1.01450586,
      "epoch": 0.5428528483390952,
      "flos": 19975382904960.0,
      "grad_norm": 2.9108428548857646,
      "language_loss": 0.77966297,
      "learning_rate": 1.8197295045415442e-06,
      "loss": 0.80064845,
      "num_input_tokens_seen": 194438545,
      "step": 9029,
      "time_per_iteration": 2.590405225753784
    },
    {
      "auxiliary_loss_clip": 0.01068568,
      "auxiliary_loss_mlp": 0.01028015,
      "balance_loss_clip": 1.03545821,
      "balance_loss_mlp": 1.01465774,
      "epoch": 0.5429129715917631,
      "flos": 21832323287040.0,
      "grad_norm": 1.4987413113833492,
      "language_loss": 0.83415121,
      "learning_rate": 1.8193416315709112e-06,
      "loss": 0.85511708,
      "num_input_tokens_seen": 194458060,
      "step": 9030,
      "time_per_iteration": 2.6201014518737793
    },
    {
      "auxiliary_loss_clip": 0.01111003,
      "auxiliary_loss_mlp": 0.01032043,
      "balance_loss_clip": 1.03846836,
      "balance_loss_mlp": 1.01998472,
      "epoch": 0.5429730948444311,
      "flos": 27782685492480.0,
      "grad_norm": 1.7448337417438335,
      "language_loss": 0.7505796,
      "learning_rate": 1.8189537654508623e-06,
      "loss": 0.77201003,
      "num_input_tokens_seen": 194477405,
      "step": 9031,
      "time_per_iteration": 2.5539753437042236
    },
    {
      "auxiliary_loss_clip": 0.01093674,
      "auxiliary_loss_mlp": 0.01035064,
      "balance_loss_clip": 1.03664088,
      "balance_loss_mlp": 1.0224452,
      "epoch": 0.543033218097099,
      "flos": 26760452336640.0,
      "grad_norm": 1.649333596332213,
      "language_loss": 0.85120881,
      "learning_rate": 1.8185659061961045e-06,
      "loss": 0.87249625,
      "num_input_tokens_seen": 194497085,
      "step": 9032,
      "time_per_iteration": 2.531602382659912
    },
    {
      "auxiliary_loss_clip": 0.01094836,
      "auxiliary_loss_mlp": 0.01035454,
      "balance_loss_clip": 1.03813934,
      "balance_loss_mlp": 1.02276397,
      "epoch": 0.5430933413497671,
      "flos": 22675254727680.0,
      "grad_norm": 1.6691807673468801,
      "language_loss": 0.73818803,
      "learning_rate": 1.8181780538213457e-06,
      "loss": 0.75949097,
      "num_input_tokens_seen": 194516785,
      "step": 9033,
      "time_per_iteration": 4.019034147262573
    },
    {
      "auxiliary_loss_clip": 0.01076812,
      "auxiliary_loss_mlp": 0.01039901,
      "balance_loss_clip": 1.04054165,
      "balance_loss_mlp": 1.02657318,
      "epoch": 0.543153464602435,
      "flos": 24607499973120.0,
      "grad_norm": 2.177149551833534,
      "language_loss": 0.75962365,
      "learning_rate": 1.8177902083412935e-06,
      "loss": 0.78079081,
      "num_input_tokens_seen": 194536475,
      "step": 9034,
      "time_per_iteration": 2.608349323272705
    },
    {
      "auxiliary_loss_clip": 0.01073814,
      "auxiliary_loss_mlp": 0.01033606,
      "balance_loss_clip": 1.04040384,
      "balance_loss_mlp": 1.02194715,
      "epoch": 0.543213587855103,
      "flos": 19025725178880.0,
      "grad_norm": 1.7314573153893202,
      "language_loss": 0.84264791,
      "learning_rate": 1.817402369770655e-06,
      "loss": 0.86372221,
      "num_input_tokens_seen": 194554495,
      "step": 9035,
      "time_per_iteration": 2.597259521484375
    },
    {
      "auxiliary_loss_clip": 0.01009304,
      "auxiliary_loss_mlp": 0.0100764,
      "balance_loss_clip": 1.01642275,
      "balance_loss_mlp": 1.00638235,
      "epoch": 0.5432737111077709,
      "flos": 65686435125120.0,
      "grad_norm": 0.7232411195063265,
      "language_loss": 0.55878568,
      "learning_rate": 1.8170145381241364e-06,
      "loss": 0.57895517,
      "num_input_tokens_seen": 194617620,
      "step": 9036,
      "time_per_iteration": 3.1327948570251465
    },
    {
      "auxiliary_loss_clip": 0.01057429,
      "auxiliary_loss_mlp": 0.01033254,
      "balance_loss_clip": 1.0377841,
      "balance_loss_mlp": 1.02063608,
      "epoch": 0.5433338343604389,
      "flos": 22091670460800.0,
      "grad_norm": 1.6156970382036415,
      "language_loss": 0.75243801,
      "learning_rate": 1.8166267134164451e-06,
      "loss": 0.77334487,
      "num_input_tokens_seen": 194637690,
      "step": 9037,
      "time_per_iteration": 3.9420387744903564
    },
    {
      "auxiliary_loss_clip": 0.01091466,
      "auxiliary_loss_mlp": 0.01035157,
      "balance_loss_clip": 1.04248655,
      "balance_loss_mlp": 1.02242565,
      "epoch": 0.5433939576131068,
      "flos": 34672649616000.0,
      "grad_norm": 1.846956549857097,
      "language_loss": 0.66237342,
      "learning_rate": 1.8162388956622875e-06,
      "loss": 0.68363965,
      "num_input_tokens_seen": 194659520,
      "step": 9038,
      "time_per_iteration": 2.6579747200012207
    },
    {
      "auxiliary_loss_clip": 0.01097383,
      "auxiliary_loss_mlp": 0.01029744,
      "balance_loss_clip": 1.03508306,
      "balance_loss_mlp": 1.01845479,
      "epoch": 0.5434540808657748,
      "flos": 20303355012480.0,
      "grad_norm": 1.8885040058958604,
      "language_loss": 0.78409743,
      "learning_rate": 1.8158510848763692e-06,
      "loss": 0.80536872,
      "num_input_tokens_seen": 194677645,
      "step": 9039,
      "time_per_iteration": 2.491894483566284
    },
    {
      "auxiliary_loss_clip": 0.01078009,
      "auxiliary_loss_mlp": 0.01035669,
      "balance_loss_clip": 1.03773236,
      "balance_loss_mlp": 1.02285385,
      "epoch": 0.5435142041184428,
      "flos": 23112790295040.0,
      "grad_norm": 1.7568204511025118,
      "language_loss": 0.76658249,
      "learning_rate": 1.8154632810733962e-06,
      "loss": 0.78771925,
      "num_input_tokens_seen": 194697400,
      "step": 9040,
      "time_per_iteration": 3.959312677383423
    },
    {
      "auxiliary_loss_clip": 0.01018391,
      "auxiliary_loss_mlp": 0.01005324,
      "balance_loss_clip": 1.01505589,
      "balance_loss_mlp": 1.00409627,
      "epoch": 0.5435743273711108,
      "flos": 64012746954240.0,
      "grad_norm": 0.6675623494882927,
      "language_loss": 0.52433085,
      "learning_rate": 1.815075484268074e-06,
      "loss": 0.54456806,
      "num_input_tokens_seen": 194761205,
      "step": 9041,
      "time_per_iteration": 3.1104893684387207
    },
    {
      "auxiliary_loss_clip": 0.01087213,
      "auxiliary_loss_mlp": 0.01037125,
      "balance_loss_clip": 1.03828955,
      "balance_loss_mlp": 1.02455389,
      "epoch": 0.5436344506237788,
      "flos": 25118903859840.0,
      "grad_norm": 1.582556449652193,
      "language_loss": 0.76215678,
      "learning_rate": 1.8146876944751078e-06,
      "loss": 0.78340018,
      "num_input_tokens_seen": 194782445,
      "step": 9042,
      "time_per_iteration": 2.550203323364258
    },
    {
      "auxiliary_loss_clip": 0.01074291,
      "auxiliary_loss_mlp": 0.01028118,
      "balance_loss_clip": 1.0361079,
      "balance_loss_mlp": 1.01644766,
      "epoch": 0.5436945738764467,
      "flos": 19572967860480.0,
      "grad_norm": 1.6194063047556018,
      "language_loss": 0.67279786,
      "learning_rate": 1.8142999117092033e-06,
      "loss": 0.69382197,
      "num_input_tokens_seen": 194800325,
      "step": 9043,
      "time_per_iteration": 2.5432426929473877
    },
    {
      "auxiliary_loss_clip": 0.01073419,
      "auxiliary_loss_mlp": 0.01027417,
      "balance_loss_clip": 1.03991759,
      "balance_loss_mlp": 1.01541281,
      "epoch": 0.5437546971291147,
      "flos": 21142515525120.0,
      "grad_norm": 1.7008843979854196,
      "language_loss": 0.84020931,
      "learning_rate": 1.8139121359850644e-06,
      "loss": 0.86121762,
      "num_input_tokens_seen": 194818675,
      "step": 9044,
      "time_per_iteration": 4.035784721374512
    },
    {
      "auxiliary_loss_clip": 0.01115294,
      "auxiliary_loss_mlp": 0.01027595,
      "balance_loss_clip": 1.03845739,
      "balance_loss_mlp": 1.01467872,
      "epoch": 0.5438148203817826,
      "flos": 25118688378240.0,
      "grad_norm": 1.6251611139407796,
      "language_loss": 0.62207472,
      "learning_rate": 1.8135243673173956e-06,
      "loss": 0.64350361,
      "num_input_tokens_seen": 194836595,
      "step": 9045,
      "time_per_iteration": 2.480785369873047
    },
    {
      "auxiliary_loss_clip": 0.01112877,
      "auxiliary_loss_mlp": 0.01032363,
      "balance_loss_clip": 1.03929281,
      "balance_loss_mlp": 1.02018631,
      "epoch": 0.5438749436344507,
      "flos": 23002939526400.0,
      "grad_norm": 1.506588028997212,
      "language_loss": 0.70023298,
      "learning_rate": 1.8131366057209023e-06,
      "loss": 0.72168541,
      "num_input_tokens_seen": 194857520,
      "step": 9046,
      "time_per_iteration": 2.4960861206054688
    },
    {
      "auxiliary_loss_clip": 0.01107879,
      "auxiliary_loss_mlp": 0.01029237,
      "balance_loss_clip": 1.03652823,
      "balance_loss_mlp": 1.01769757,
      "epoch": 0.5439350668871186,
      "flos": 15487016065920.0,
      "grad_norm": 1.6363992819873512,
      "language_loss": 0.76999432,
      "learning_rate": 1.8127488512102868e-06,
      "loss": 0.7913655,
      "num_input_tokens_seen": 194876020,
      "step": 9047,
      "time_per_iteration": 2.4601383209228516
    },
    {
      "auxiliary_loss_clip": 0.01086822,
      "auxiliary_loss_mlp": 0.01039641,
      "balance_loss_clip": 1.03717864,
      "balance_loss_mlp": 1.02594948,
      "epoch": 0.5439951901397866,
      "flos": 17238415311360.0,
      "grad_norm": 1.7753024210787685,
      "language_loss": 0.73034161,
      "learning_rate": 1.8123611038002547e-06,
      "loss": 0.75160623,
      "num_input_tokens_seen": 194894650,
      "step": 9048,
      "time_per_iteration": 2.4942758083343506
    },
    {
      "auxiliary_loss_clip": 0.01064068,
      "auxiliary_loss_mlp": 0.01039026,
      "balance_loss_clip": 1.03591108,
      "balance_loss_mlp": 1.02441108,
      "epoch": 0.5440553133924545,
      "flos": 18661016436480.0,
      "grad_norm": 2.1115039067885095,
      "language_loss": 0.93189508,
      "learning_rate": 1.8119733635055076e-06,
      "loss": 0.95292598,
      "num_input_tokens_seen": 194911935,
      "step": 9049,
      "time_per_iteration": 2.5587680339813232
    },
    {
      "auxiliary_loss_clip": 0.0109937,
      "auxiliary_loss_mlp": 0.01030508,
      "balance_loss_clip": 1.03740788,
      "balance_loss_mlp": 1.01909339,
      "epoch": 0.5441154366451225,
      "flos": 27122934435840.0,
      "grad_norm": 1.779089241638626,
      "language_loss": 0.74020743,
      "learning_rate": 1.8115856303407492e-06,
      "loss": 0.76150626,
      "num_input_tokens_seen": 194931620,
      "step": 9050,
      "time_per_iteration": 2.522371530532837
    },
    {
      "auxiliary_loss_clip": 0.01102681,
      "auxiliary_loss_mlp": 0.01027546,
      "balance_loss_clip": 1.03944254,
      "balance_loss_mlp": 1.0156126,
      "epoch": 0.5441755598977904,
      "flos": 25993867253760.0,
      "grad_norm": 1.811214799253383,
      "language_loss": 0.66952026,
      "learning_rate": 1.8111979043206832e-06,
      "loss": 0.69082248,
      "num_input_tokens_seen": 194952560,
      "step": 9051,
      "time_per_iteration": 2.526942253112793
    },
    {
      "auxiliary_loss_clip": 0.01073427,
      "auxiliary_loss_mlp": 0.01031999,
      "balance_loss_clip": 1.03525507,
      "balance_loss_mlp": 1.02013171,
      "epoch": 0.5442356831504584,
      "flos": 32380041173760.0,
      "grad_norm": 1.7337497313243895,
      "language_loss": 0.67624438,
      "learning_rate": 1.810810185460011e-06,
      "loss": 0.69729871,
      "num_input_tokens_seen": 194973915,
      "step": 9052,
      "time_per_iteration": 2.628087043762207
    },
    {
      "auxiliary_loss_clip": 0.01112482,
      "auxiliary_loss_mlp": 0.01033291,
      "balance_loss_clip": 1.03801537,
      "balance_loss_mlp": 1.02061915,
      "epoch": 0.5442958064031264,
      "flos": 24164290056960.0,
      "grad_norm": 1.8082724514407518,
      "language_loss": 0.9288038,
      "learning_rate": 1.810422473773436e-06,
      "loss": 0.95026153,
      "num_input_tokens_seen": 194990170,
      "step": 9053,
      "time_per_iteration": 2.4998860359191895
    },
    {
      "auxiliary_loss_clip": 0.0109381,
      "auxiliary_loss_mlp": 0.01034754,
      "balance_loss_clip": 1.03827691,
      "balance_loss_mlp": 1.02208853,
      "epoch": 0.5443559296557944,
      "flos": 18764690065920.0,
      "grad_norm": 8.638178631778084,
      "language_loss": 0.83121467,
      "learning_rate": 1.8100347692756595e-06,
      "loss": 0.85250038,
      "num_input_tokens_seen": 195006395,
      "step": 9054,
      "time_per_iteration": 2.5083248615264893
    },
    {
      "auxiliary_loss_clip": 0.01090456,
      "auxiliary_loss_mlp": 0.01031982,
      "balance_loss_clip": 1.0386256,
      "balance_loss_mlp": 1.01908398,
      "epoch": 0.5444160529084624,
      "flos": 22632556435200.0,
      "grad_norm": 1.9685317471784334,
      "language_loss": 0.68473017,
      "learning_rate": 1.8096470719813836e-06,
      "loss": 0.70595455,
      "num_input_tokens_seen": 195025080,
      "step": 9055,
      "time_per_iteration": 2.5200605392456055
    },
    {
      "auxiliary_loss_clip": 0.0100702,
      "auxiliary_loss_mlp": 0.01004894,
      "balance_loss_clip": 1.01095891,
      "balance_loss_mlp": 1.00358891,
      "epoch": 0.5444761761611303,
      "flos": 69671909600640.0,
      "grad_norm": 0.725917733276565,
      "language_loss": 0.57704818,
      "learning_rate": 1.80925938190531e-06,
      "loss": 0.59716731,
      "num_input_tokens_seen": 195085725,
      "step": 9056,
      "time_per_iteration": 3.1435201168060303
    },
    {
      "auxiliary_loss_clip": 0.01076636,
      "auxiliary_loss_mlp": 0.01031076,
      "balance_loss_clip": 1.04070544,
      "balance_loss_mlp": 1.01867795,
      "epoch": 0.5445362994137983,
      "flos": 14278442129280.0,
      "grad_norm": 1.8917457824076311,
      "language_loss": 0.69350058,
      "learning_rate": 1.8088716990621395e-06,
      "loss": 0.71457767,
      "num_input_tokens_seen": 195102585,
      "step": 9057,
      "time_per_iteration": 2.520031690597534
    },
    {
      "auxiliary_loss_clip": 0.01099148,
      "auxiliary_loss_mlp": 0.01031347,
      "balance_loss_clip": 1.03764343,
      "balance_loss_mlp": 1.01890731,
      "epoch": 0.5445964226664662,
      "flos": 28986195611520.0,
      "grad_norm": 2.155498460602353,
      "language_loss": 0.75044382,
      "learning_rate": 1.8084840234665738e-06,
      "loss": 0.77174878,
      "num_input_tokens_seen": 195120055,
      "step": 9058,
      "time_per_iteration": 2.553114891052246
    },
    {
      "auxiliary_loss_clip": 0.0101676,
      "auxiliary_loss_mlp": 0.01009968,
      "balance_loss_clip": 1.02650714,
      "balance_loss_mlp": 1.00858498,
      "epoch": 0.5446565459191343,
      "flos": 68620230270720.0,
      "grad_norm": 0.8048550723931328,
      "language_loss": 0.62723708,
      "learning_rate": 1.808096355133312e-06,
      "loss": 0.64750433,
      "num_input_tokens_seen": 195181045,
      "step": 9059,
      "time_per_iteration": 3.257617473602295
    },
    {
      "auxiliary_loss_clip": 0.0109825,
      "auxiliary_loss_mlp": 0.0103244,
      "balance_loss_clip": 1.03770018,
      "balance_loss_mlp": 1.02048886,
      "epoch": 0.5447166691718022,
      "flos": 16216469464320.0,
      "grad_norm": 1.7657786315333013,
      "language_loss": 0.7965287,
      "learning_rate": 1.8077086940770572e-06,
      "loss": 0.81783557,
      "num_input_tokens_seen": 195198840,
      "step": 9060,
      "time_per_iteration": 2.473949432373047
    },
    {
      "auxiliary_loss_clip": 0.01101269,
      "auxiliary_loss_mlp": 0.01033198,
      "balance_loss_clip": 1.03676617,
      "balance_loss_mlp": 1.02067566,
      "epoch": 0.5447767924244702,
      "flos": 25849039616640.0,
      "grad_norm": 1.5700885608539095,
      "language_loss": 0.7956872,
      "learning_rate": 1.8073210403125072e-06,
      "loss": 0.81703186,
      "num_input_tokens_seen": 195218720,
      "step": 9061,
      "time_per_iteration": 2.528481960296631
    },
    {
      "auxiliary_loss_clip": 0.01097665,
      "auxiliary_loss_mlp": 0.01027185,
      "balance_loss_clip": 1.03706169,
      "balance_loss_mlp": 1.01525819,
      "epoch": 0.5448369156771381,
      "flos": 19677718897920.0,
      "grad_norm": 1.6945780076340198,
      "language_loss": 0.87041277,
      "learning_rate": 1.8069333938543627e-06,
      "loss": 0.89166129,
      "num_input_tokens_seen": 195235770,
      "step": 9062,
      "time_per_iteration": 2.4849300384521484
    },
    {
      "auxiliary_loss_clip": 0.01091394,
      "auxiliary_loss_mlp": 0.01032756,
      "balance_loss_clip": 1.03740311,
      "balance_loss_mlp": 1.01781321,
      "epoch": 0.5448970389298061,
      "flos": 19281804215040.0,
      "grad_norm": 1.71200442947359,
      "language_loss": 0.82353669,
      "learning_rate": 1.8065457547173233e-06,
      "loss": 0.84477818,
      "num_input_tokens_seen": 195254870,
      "step": 9063,
      "time_per_iteration": 2.5113160610198975
    },
    {
      "auxiliary_loss_clip": 0.01109921,
      "auxiliary_loss_mlp": 0.01029919,
      "balance_loss_clip": 1.0360496,
      "balance_loss_mlp": 1.01675189,
      "epoch": 0.544957162182474,
      "flos": 20991690316800.0,
      "grad_norm": 1.720453524271078,
      "language_loss": 0.63998973,
      "learning_rate": 1.8061581229160878e-06,
      "loss": 0.66138816,
      "num_input_tokens_seen": 195273390,
      "step": 9064,
      "time_per_iteration": 2.4803221225738525
    },
    {
      "auxiliary_loss_clip": 0.01111928,
      "auxiliary_loss_mlp": 0.01032882,
      "balance_loss_clip": 1.03676772,
      "balance_loss_mlp": 1.01988244,
      "epoch": 0.545017285435142,
      "flos": 25374587846400.0,
      "grad_norm": 1.837266156051042,
      "language_loss": 0.79722536,
      "learning_rate": 1.8057704984653566e-06,
      "loss": 0.81867349,
      "num_input_tokens_seen": 195295635,
      "step": 9065,
      "time_per_iteration": 2.5089051723480225
    },
    {
      "auxiliary_loss_clip": 0.01073946,
      "auxiliary_loss_mlp": 0.01028961,
      "balance_loss_clip": 1.03595042,
      "balance_loss_mlp": 1.01814842,
      "epoch": 0.54507740868781,
      "flos": 19134749934720.0,
      "grad_norm": 2.1605137285326292,
      "language_loss": 0.78284389,
      "learning_rate": 1.805382881379827e-06,
      "loss": 0.803873,
      "num_input_tokens_seen": 195312545,
      "step": 9066,
      "time_per_iteration": 2.5532727241516113
    },
    {
      "auxiliary_loss_clip": 0.01104484,
      "auxiliary_loss_mlp": 0.01031982,
      "balance_loss_clip": 1.03657341,
      "balance_loss_mlp": 1.01893413,
      "epoch": 0.545137531940478,
      "flos": 26249802635520.0,
      "grad_norm": 1.776655665004028,
      "language_loss": 0.75673556,
      "learning_rate": 1.8049952716741975e-06,
      "loss": 0.77810019,
      "num_input_tokens_seen": 195332955,
      "step": 9067,
      "time_per_iteration": 2.5351815223693848
    },
    {
      "auxiliary_loss_clip": 0.01081281,
      "auxiliary_loss_mlp": 0.01034262,
      "balance_loss_clip": 1.04158652,
      "balance_loss_mlp": 1.01923537,
      "epoch": 0.545197655193146,
      "flos": 37555629995520.0,
      "grad_norm": 2.12782606443098,
      "language_loss": 0.63100278,
      "learning_rate": 1.8046076693631682e-06,
      "loss": 0.65215826,
      "num_input_tokens_seen": 195355930,
      "step": 9068,
      "time_per_iteration": 2.7010996341705322
    },
    {
      "auxiliary_loss_clip": 0.01077001,
      "auxiliary_loss_mlp": 0.01035297,
      "balance_loss_clip": 1.04055059,
      "balance_loss_mlp": 1.02307236,
      "epoch": 0.5452577784458139,
      "flos": 26031250333440.0,
      "grad_norm": 1.5083720289963207,
      "language_loss": 0.72082126,
      "learning_rate": 1.8042200744614343e-06,
      "loss": 0.74194419,
      "num_input_tokens_seen": 195376445,
      "step": 9069,
      "time_per_iteration": 2.5882680416107178
    },
    {
      "auxiliary_loss_clip": 0.01108921,
      "auxiliary_loss_mlp": 0.01025729,
      "balance_loss_clip": 1.03889608,
      "balance_loss_mlp": 1.01461256,
      "epoch": 0.5453179016984819,
      "flos": 17639034675840.0,
      "grad_norm": 1.6896226298518662,
      "language_loss": 0.73785514,
      "learning_rate": 1.8038324869836957e-06,
      "loss": 0.75920165,
      "num_input_tokens_seen": 195393725,
      "step": 9070,
      "time_per_iteration": 2.45109486579895
    },
    {
      "auxiliary_loss_clip": 0.01098038,
      "auxiliary_loss_mlp": 0.01028184,
      "balance_loss_clip": 1.04171908,
      "balance_loss_mlp": 1.01594758,
      "epoch": 0.5453780249511498,
      "flos": 23216679406080.0,
      "grad_norm": 1.9704054612314836,
      "language_loss": 0.60602748,
      "learning_rate": 1.8034449069446489e-06,
      "loss": 0.62728965,
      "num_input_tokens_seen": 195411380,
      "step": 9071,
      "time_per_iteration": 3.912318229675293
    },
    {
      "auxiliary_loss_clip": 0.01033132,
      "auxiliary_loss_mlp": 0.01004421,
      "balance_loss_clip": 1.00809467,
      "balance_loss_mlp": 1.00327063,
      "epoch": 0.5454381482038179,
      "flos": 68696504801280.0,
      "grad_norm": 0.7015668975774855,
      "language_loss": 0.57133102,
      "learning_rate": 1.80305733435899e-06,
      "loss": 0.59170663,
      "num_input_tokens_seen": 195482015,
      "step": 9072,
      "time_per_iteration": 3.173682689666748
    },
    {
      "auxiliary_loss_clip": 0.01081162,
      "auxiliary_loss_mlp": 0.01033772,
      "balance_loss_clip": 1.0347743,
      "balance_loss_mlp": 1.02060509,
      "epoch": 0.5454982714564858,
      "flos": 13260626346240.0,
      "grad_norm": 1.655506888137964,
      "language_loss": 0.70027244,
      "learning_rate": 1.8026697692414174e-06,
      "loss": 0.72142178,
      "num_input_tokens_seen": 195500440,
      "step": 9073,
      "time_per_iteration": 2.527085304260254
    },
    {
      "auxiliary_loss_clip": 0.01087558,
      "auxiliary_loss_mlp": 0.01030307,
      "balance_loss_clip": 1.03680432,
      "balance_loss_mlp": 1.01874352,
      "epoch": 0.5455583947091538,
      "flos": 21835878733440.0,
      "grad_norm": 2.767510351363698,
      "language_loss": 0.71530104,
      "learning_rate": 1.802282211606627e-06,
      "loss": 0.7364797,
      "num_input_tokens_seen": 195520860,
      "step": 9074,
      "time_per_iteration": 2.5396437644958496
    },
    {
      "auxiliary_loss_clip": 0.01096946,
      "auxiliary_loss_mlp": 0.01038335,
      "balance_loss_clip": 1.03580821,
      "balance_loss_mlp": 1.02647316,
      "epoch": 0.5456185179618217,
      "flos": 17817438551040.0,
      "grad_norm": 3.732793104052465,
      "language_loss": 0.68383515,
      "learning_rate": 1.8018946614693148e-06,
      "loss": 0.70518792,
      "num_input_tokens_seen": 195538615,
      "step": 9075,
      "time_per_iteration": 2.484367847442627
    },
    {
      "auxiliary_loss_clip": 0.01097827,
      "auxiliary_loss_mlp": 0.01030398,
      "balance_loss_clip": 1.03685379,
      "balance_loss_mlp": 1.01943028,
      "epoch": 0.5456786412144897,
      "flos": 21069401391360.0,
      "grad_norm": 1.6545992355371972,
      "language_loss": 0.81085992,
      "learning_rate": 1.8015071188441768e-06,
      "loss": 0.83214217,
      "num_input_tokens_seen": 195557460,
      "step": 9076,
      "time_per_iteration": 3.837646007537842
    },
    {
      "auxiliary_loss_clip": 0.011008,
      "auxiliary_loss_mlp": 0.01029899,
      "balance_loss_clip": 1.03743911,
      "balance_loss_mlp": 1.01769781,
      "epoch": 0.5457387644671576,
      "flos": 23294965098240.0,
      "grad_norm": 1.7367395153774532,
      "language_loss": 0.80189443,
      "learning_rate": 1.8011195837459089e-06,
      "loss": 0.82320142,
      "num_input_tokens_seen": 195577985,
      "step": 9077,
      "time_per_iteration": 2.5006985664367676
    },
    {
      "auxiliary_loss_clip": 0.01101663,
      "auxiliary_loss_mlp": 0.01030754,
      "balance_loss_clip": 1.03645563,
      "balance_loss_mlp": 1.0189817,
      "epoch": 0.5457988877198257,
      "flos": 21617039122560.0,
      "grad_norm": 1.8416388927923766,
      "language_loss": 0.67578876,
      "learning_rate": 1.8007320561892064e-06,
      "loss": 0.69711298,
      "num_input_tokens_seen": 195597620,
      "step": 9078,
      "time_per_iteration": 2.526109218597412
    },
    {
      "auxiliary_loss_clip": 0.01103343,
      "auxiliary_loss_mlp": 0.01033005,
      "balance_loss_clip": 1.03761232,
      "balance_loss_mlp": 1.02015388,
      "epoch": 0.5458590109724936,
      "flos": 23762485543680.0,
      "grad_norm": 1.7697433132597218,
      "language_loss": 0.81552184,
      "learning_rate": 1.800344536188764e-06,
      "loss": 0.83688533,
      "num_input_tokens_seen": 195615910,
      "step": 9079,
      "time_per_iteration": 3.893725872039795
    },
    {
      "auxiliary_loss_clip": 0.01116028,
      "auxiliary_loss_mlp": 0.01031469,
      "balance_loss_clip": 1.03879833,
      "balance_loss_mlp": 1.0177722,
      "epoch": 0.5459191342251616,
      "flos": 24424283675520.0,
      "grad_norm": 1.5955631509530896,
      "language_loss": 0.75598013,
      "learning_rate": 1.799957023759277e-06,
      "loss": 0.77745509,
      "num_input_tokens_seen": 195635620,
      "step": 9080,
      "time_per_iteration": 2.4910190105438232
    },
    {
      "auxiliary_loss_clip": 0.01075057,
      "auxiliary_loss_mlp": 0.01032124,
      "balance_loss_clip": 1.03797495,
      "balance_loss_mlp": 1.01907086,
      "epoch": 0.5459792574778296,
      "flos": 23623009032960.0,
      "grad_norm": 2.4386449623215505,
      "language_loss": 0.83679944,
      "learning_rate": 1.7995695189154392e-06,
      "loss": 0.85787123,
      "num_input_tokens_seen": 195652495,
      "step": 9081,
      "time_per_iteration": 2.5730223655700684
    },
    {
      "auxiliary_loss_clip": 0.01115211,
      "auxiliary_loss_mlp": 0.01030308,
      "balance_loss_clip": 1.03886223,
      "balance_loss_mlp": 1.01730824,
      "epoch": 0.5460393807304975,
      "flos": 19135540033920.0,
      "grad_norm": 1.6754877599203015,
      "language_loss": 0.70361197,
      "learning_rate": 1.7991820216719461e-06,
      "loss": 0.72506714,
      "num_input_tokens_seen": 195671965,
      "step": 9082,
      "time_per_iteration": 2.4609599113464355
    },
    {
      "auxiliary_loss_clip": 0.01107181,
      "auxiliary_loss_mlp": 0.01025787,
      "balance_loss_clip": 1.03598082,
      "balance_loss_mlp": 1.01358604,
      "epoch": 0.5460995039831655,
      "flos": 35918534805120.0,
      "grad_norm": 1.607551248324354,
      "language_loss": 0.66585368,
      "learning_rate": 1.7987945320434906e-06,
      "loss": 0.68718338,
      "num_input_tokens_seen": 195694725,
      "step": 9083,
      "time_per_iteration": 3.99273943901062
    },
    {
      "auxiliary_loss_clip": 0.01088033,
      "auxiliary_loss_mlp": 0.01027548,
      "balance_loss_clip": 1.04325294,
      "balance_loss_mlp": 1.01595497,
      "epoch": 0.5461596272358334,
      "flos": 26759231274240.0,
      "grad_norm": 1.6296280274723032,
      "language_loss": 0.78946543,
      "learning_rate": 1.798407050044766e-06,
      "loss": 0.81062132,
      "num_input_tokens_seen": 195714090,
      "step": 9084,
      "time_per_iteration": 2.571420907974243
    },
    {
      "auxiliary_loss_clip": 0.01104108,
      "auxiliary_loss_mlp": 0.01029718,
      "balance_loss_clip": 1.04017723,
      "balance_loss_mlp": 1.01782095,
      "epoch": 0.5462197504885015,
      "flos": 20886580143360.0,
      "grad_norm": 1.728308910760671,
      "language_loss": 0.7531783,
      "learning_rate": 1.7980195756904675e-06,
      "loss": 0.77451652,
      "num_input_tokens_seen": 195733585,
      "step": 9085,
      "time_per_iteration": 2.507934093475342
    },
    {
      "auxiliary_loss_clip": 0.01088993,
      "auxiliary_loss_mlp": 0.01029241,
      "balance_loss_clip": 1.03604543,
      "balance_loss_mlp": 1.01679015,
      "epoch": 0.5462798737411694,
      "flos": 25804976607360.0,
      "grad_norm": 2.5779307180765554,
      "language_loss": 0.74743629,
      "learning_rate": 1.7976321089952857e-06,
      "loss": 0.76861864,
      "num_input_tokens_seen": 195752820,
      "step": 9086,
      "time_per_iteration": 2.5553877353668213
    },
    {
      "auxiliary_loss_clip": 0.01099068,
      "auxiliary_loss_mlp": 0.0102914,
      "balance_loss_clip": 1.03991938,
      "balance_loss_mlp": 1.0166111,
      "epoch": 0.5463399969938374,
      "flos": 25775027642880.0,
      "grad_norm": 1.558258750295256,
      "language_loss": 0.7683959,
      "learning_rate": 1.7972446499739155e-06,
      "loss": 0.78967798,
      "num_input_tokens_seen": 195773740,
      "step": 9087,
      "time_per_iteration": 2.548981189727783
    },
    {
      "auxiliary_loss_clip": 0.01104792,
      "auxiliary_loss_mlp": 0.01038875,
      "balance_loss_clip": 1.03964388,
      "balance_loss_mlp": 1.02513599,
      "epoch": 0.5464001202465053,
      "flos": 18843298980480.0,
      "grad_norm": 1.6925034344298395,
      "language_loss": 0.77573633,
      "learning_rate": 1.7968571986410484e-06,
      "loss": 0.79717302,
      "num_input_tokens_seen": 195792125,
      "step": 9088,
      "time_per_iteration": 2.4666686058044434
    },
    {
      "auxiliary_loss_clip": 0.00990068,
      "auxiliary_loss_mlp": 0.01001345,
      "balance_loss_clip": 1.02576923,
      "balance_loss_mlp": 1.00003326,
      "epoch": 0.5464602434991733,
      "flos": 69049541623680.0,
      "grad_norm": 1.0024909420216783,
      "language_loss": 0.57715523,
      "learning_rate": 1.7964697550113758e-06,
      "loss": 0.59706938,
      "num_input_tokens_seen": 195854935,
      "step": 9089,
      "time_per_iteration": 3.314000368118286
    },
    {
      "auxiliary_loss_clip": 0.01077028,
      "auxiliary_loss_mlp": 0.01033225,
      "balance_loss_clip": 1.03647375,
      "balance_loss_mlp": 1.01991498,
      "epoch": 0.5465203667518412,
      "flos": 27560039040000.0,
      "grad_norm": 1.8946113531166529,
      "language_loss": 0.76899409,
      "learning_rate": 1.7960823190995918e-06,
      "loss": 0.79009664,
      "num_input_tokens_seen": 195874715,
      "step": 9090,
      "time_per_iteration": 2.8511955738067627
    },
    {
      "auxiliary_loss_clip": 0.01097919,
      "auxiliary_loss_mlp": 0.01034327,
      "balance_loss_clip": 1.03663933,
      "balance_loss_mlp": 1.0203495,
      "epoch": 0.5465804900045093,
      "flos": 21210206705280.0,
      "grad_norm": 1.89859236567917,
      "language_loss": 0.73386395,
      "learning_rate": 1.7956948909203855e-06,
      "loss": 0.75518638,
      "num_input_tokens_seen": 195892610,
      "step": 9091,
      "time_per_iteration": 2.535501480102539
    },
    {
      "auxiliary_loss_clip": 0.0109239,
      "auxiliary_loss_mlp": 0.01033449,
      "balance_loss_clip": 1.03855824,
      "balance_loss_mlp": 1.02080703,
      "epoch": 0.5466406132571772,
      "flos": 22488949860480.0,
      "grad_norm": 2.2432237533814745,
      "language_loss": 0.77975792,
      "learning_rate": 1.7953074704884498e-06,
      "loss": 0.80101633,
      "num_input_tokens_seen": 195911085,
      "step": 9092,
      "time_per_iteration": 2.524381160736084
    },
    {
      "auxiliary_loss_clip": 0.01114126,
      "auxiliary_loss_mlp": 0.01029631,
      "balance_loss_clip": 1.03911471,
      "balance_loss_mlp": 1.01678646,
      "epoch": 0.5467007365098452,
      "flos": 17675843137920.0,
      "grad_norm": 2.5560811533909646,
      "language_loss": 0.74815691,
      "learning_rate": 1.794920057818476e-06,
      "loss": 0.76959455,
      "num_input_tokens_seen": 195929845,
      "step": 9093,
      "time_per_iteration": 2.4311556816101074
    },
    {
      "auxiliary_loss_clip": 0.01102788,
      "auxiliary_loss_mlp": 0.01034996,
      "balance_loss_clip": 1.03676414,
      "balance_loss_mlp": 1.02067304,
      "epoch": 0.5467608597625132,
      "flos": 15698852524800.0,
      "grad_norm": 4.632504497341471,
      "language_loss": 0.69262433,
      "learning_rate": 1.7945326529251533e-06,
      "loss": 0.71400213,
      "num_input_tokens_seen": 195946350,
      "step": 9094,
      "time_per_iteration": 2.457397699356079
    },
    {
      "auxiliary_loss_clip": 0.01090582,
      "auxiliary_loss_mlp": 0.01029615,
      "balance_loss_clip": 1.0390532,
      "balance_loss_mlp": 1.01824832,
      "epoch": 0.5468209830151811,
      "flos": 24312816794880.0,
      "grad_norm": 3.5106181196736723,
      "language_loss": 0.67821211,
      "learning_rate": 1.7941452558231731e-06,
      "loss": 0.69941413,
      "num_input_tokens_seen": 195959840,
      "step": 9095,
      "time_per_iteration": 2.5091392993927
    },
    {
      "auxiliary_loss_clip": 0.01080213,
      "auxiliary_loss_mlp": 0.01031795,
      "balance_loss_clip": 1.04425609,
      "balance_loss_mlp": 1.01983821,
      "epoch": 0.5468811062678491,
      "flos": 29166323339520.0,
      "grad_norm": 1.5026796906758875,
      "language_loss": 0.66413796,
      "learning_rate": 1.7937578665272256e-06,
      "loss": 0.68525803,
      "num_input_tokens_seen": 195981125,
      "step": 9096,
      "time_per_iteration": 2.6219098567962646
    },
    {
      "auxiliary_loss_clip": 0.01010715,
      "auxiliary_loss_mlp": 0.01003331,
      "balance_loss_clip": 1.01650739,
      "balance_loss_mlp": 1.0020256,
      "epoch": 0.546941229520517,
      "flos": 67867037982720.0,
      "grad_norm": 0.7377101599122364,
      "language_loss": 0.57526577,
      "learning_rate": 1.7933704850520007e-06,
      "loss": 0.59540617,
      "num_input_tokens_seen": 196038880,
      "step": 9097,
      "time_per_iteration": 3.251030921936035
    },
    {
      "auxiliary_loss_clip": 0.01040347,
      "auxiliary_loss_mlp": 0.01004357,
      "balance_loss_clip": 1.02954769,
      "balance_loss_mlp": 1.00291431,
      "epoch": 0.5470013527731851,
      "flos": 58270306625280.0,
      "grad_norm": 0.9562805267724301,
      "language_loss": 0.64785743,
      "learning_rate": 1.7929831114121868e-06,
      "loss": 0.66830444,
      "num_input_tokens_seen": 196099215,
      "step": 9098,
      "time_per_iteration": 3.0809147357940674
    },
    {
      "auxiliary_loss_clip": 0.01105056,
      "auxiliary_loss_mlp": 0.01036154,
      "balance_loss_clip": 1.03881073,
      "balance_loss_mlp": 1.02352405,
      "epoch": 0.547061476025853,
      "flos": 22965915582720.0,
      "grad_norm": 1.7879110172769614,
      "language_loss": 0.73245716,
      "learning_rate": 1.7925957456224753e-06,
      "loss": 0.75386918,
      "num_input_tokens_seen": 196120370,
      "step": 9099,
      "time_per_iteration": 2.523378610610962
    },
    {
      "auxiliary_loss_clip": 0.01086199,
      "auxiliary_loss_mlp": 0.01029651,
      "balance_loss_clip": 1.0370183,
      "balance_loss_mlp": 1.01882052,
      "epoch": 0.547121599278521,
      "flos": 29968244426880.0,
      "grad_norm": 1.9865718273102215,
      "language_loss": 0.72587025,
      "learning_rate": 1.7922083876975537e-06,
      "loss": 0.74702871,
      "num_input_tokens_seen": 196139075,
      "step": 9100,
      "time_per_iteration": 2.5985329151153564
    },
    {
      "auxiliary_loss_clip": 0.01098067,
      "auxiliary_loss_mlp": 0.00786434,
      "balance_loss_clip": 1.04210234,
      "balance_loss_mlp": 1.01205826,
      "epoch": 0.5471817225311889,
      "flos": 36535443914880.0,
      "grad_norm": 1.8013064486118635,
      "language_loss": 0.67988044,
      "learning_rate": 1.7918210376521102e-06,
      "loss": 0.69872546,
      "num_input_tokens_seen": 196159990,
      "step": 9101,
      "time_per_iteration": 2.642435073852539
    },
    {
      "auxiliary_loss_clip": 0.0110972,
      "auxiliary_loss_mlp": 0.01027901,
      "balance_loss_clip": 1.03669512,
      "balance_loss_mlp": 1.0157299,
      "epoch": 0.5472418457838569,
      "flos": 25775243124480.0,
      "grad_norm": 3.139420862311445,
      "language_loss": 0.77886522,
      "learning_rate": 1.7914336955008343e-06,
      "loss": 0.80024147,
      "num_input_tokens_seen": 196180570,
      "step": 9102,
      "time_per_iteration": 2.508272647857666
    },
    {
      "auxiliary_loss_clip": 0.010769,
      "auxiliary_loss_mlp": 0.01031768,
      "balance_loss_clip": 1.03672791,
      "balance_loss_mlp": 1.01903653,
      "epoch": 0.5473019690365248,
      "flos": 27887687925120.0,
      "grad_norm": 1.556925876717397,
      "language_loss": 0.72307342,
      "learning_rate": 1.791046361258413e-06,
      "loss": 0.74416006,
      "num_input_tokens_seen": 196200300,
      "step": 9103,
      "time_per_iteration": 2.5973784923553467
    },
    {
      "auxiliary_loss_clip": 0.01086247,
      "auxiliary_loss_mlp": 0.01027999,
      "balance_loss_clip": 1.03858757,
      "balance_loss_mlp": 1.01597059,
      "epoch": 0.5473620922891929,
      "flos": 57631490219520.0,
      "grad_norm": 1.292142755368098,
      "language_loss": 0.65295285,
      "learning_rate": 1.7906590349395356e-06,
      "loss": 0.67409533,
      "num_input_tokens_seen": 196228525,
      "step": 9104,
      "time_per_iteration": 2.8923470973968506
    },
    {
      "auxiliary_loss_clip": 0.0109803,
      "auxiliary_loss_mlp": 0.0102948,
      "balance_loss_clip": 1.03963041,
      "balance_loss_mlp": 1.01650369,
      "epoch": 0.5474222155418608,
      "flos": 19354056422400.0,
      "grad_norm": 3.1867569241726272,
      "language_loss": 0.81474429,
      "learning_rate": 1.790271716558888e-06,
      "loss": 0.8360194,
      "num_input_tokens_seen": 196247690,
      "step": 9105,
      "time_per_iteration": 2.485377788543701
    },
    {
      "auxiliary_loss_clip": 0.01108327,
      "auxiliary_loss_mlp": 0.01027042,
      "balance_loss_clip": 1.03625906,
      "balance_loss_mlp": 1.01566982,
      "epoch": 0.5474823387945288,
      "flos": 25120448144640.0,
      "grad_norm": 1.4552425807266383,
      "language_loss": 0.8021878,
      "learning_rate": 1.7898844061311575e-06,
      "loss": 0.8235414,
      "num_input_tokens_seen": 196268555,
      "step": 9106,
      "time_per_iteration": 2.5222840309143066
    },
    {
      "auxiliary_loss_clip": 0.01100481,
      "auxiliary_loss_mlp": 0.0103184,
      "balance_loss_clip": 1.0372858,
      "balance_loss_mlp": 1.02020562,
      "epoch": 0.5475424620471967,
      "flos": 18004174381440.0,
      "grad_norm": 1.7849866643417949,
      "language_loss": 0.69666106,
      "learning_rate": 1.7894971036710322e-06,
      "loss": 0.71798426,
      "num_input_tokens_seen": 196285585,
      "step": 9107,
      "time_per_iteration": 2.4588370323181152
    },
    {
      "auxiliary_loss_clip": 0.01101865,
      "auxiliary_loss_mlp": 0.01028987,
      "balance_loss_clip": 1.03543878,
      "balance_loss_mlp": 1.01682186,
      "epoch": 0.5476025852998647,
      "flos": 22309324922880.0,
      "grad_norm": 1.6988179954113647,
      "language_loss": 0.63101363,
      "learning_rate": 1.789109809193197e-06,
      "loss": 0.65232205,
      "num_input_tokens_seen": 196305085,
      "step": 9108,
      "time_per_iteration": 2.4863169193267822
    },
    {
      "auxiliary_loss_clip": 0.01107632,
      "auxiliary_loss_mlp": 0.01027908,
      "balance_loss_clip": 1.03548121,
      "balance_loss_mlp": 1.01653576,
      "epoch": 0.5476627085525327,
      "flos": 20120497850880.0,
      "grad_norm": 1.6618998228949053,
      "language_loss": 0.75062859,
      "learning_rate": 1.7887225227123396e-06,
      "loss": 0.77198404,
      "num_input_tokens_seen": 196323945,
      "step": 9109,
      "time_per_iteration": 3.852219343185425
    },
    {
      "auxiliary_loss_clip": 0.01089465,
      "auxiliary_loss_mlp": 0.01031192,
      "balance_loss_clip": 1.03771591,
      "balance_loss_mlp": 1.01852608,
      "epoch": 0.5477228318052006,
      "flos": 17712579772800.0,
      "grad_norm": 1.988479209665081,
      "language_loss": 0.77645838,
      "learning_rate": 1.7883352442431457e-06,
      "loss": 0.797665,
      "num_input_tokens_seen": 196342200,
      "step": 9110,
      "time_per_iteration": 2.497479200363159
    },
    {
      "auxiliary_loss_clip": 0.01094623,
      "auxiliary_loss_mlp": 0.01030159,
      "balance_loss_clip": 1.03592563,
      "balance_loss_mlp": 1.01904297,
      "epoch": 0.5477829550578687,
      "flos": 25848895962240.0,
      "grad_norm": 1.4745311331084272,
      "language_loss": 0.71200025,
      "learning_rate": 1.7879479738002993e-06,
      "loss": 0.73324811,
      "num_input_tokens_seen": 196362940,
      "step": 9111,
      "time_per_iteration": 2.5354013442993164
    },
    {
      "auxiliary_loss_clip": 0.010992,
      "auxiliary_loss_mlp": 0.01034125,
      "balance_loss_clip": 1.03639507,
      "balance_loss_mlp": 1.02201366,
      "epoch": 0.5478430783105366,
      "flos": 23039676161280.0,
      "grad_norm": 1.5779325335810526,
      "language_loss": 0.71172303,
      "learning_rate": 1.7875607113984876e-06,
      "loss": 0.73305631,
      "num_input_tokens_seen": 196383070,
      "step": 9112,
      "time_per_iteration": 2.4987633228302
    },
    {
      "auxiliary_loss_clip": 0.0106972,
      "auxiliary_loss_mlp": 0.01028087,
      "balance_loss_clip": 1.04203367,
      "balance_loss_mlp": 1.01595712,
      "epoch": 0.5479032015632046,
      "flos": 16071210864000.0,
      "grad_norm": 2.201731698176773,
      "language_loss": 0.88258183,
      "learning_rate": 1.7871734570523953e-06,
      "loss": 0.90355986,
      "num_input_tokens_seen": 196398485,
      "step": 9113,
      "time_per_iteration": 2.5496573448181152
    },
    {
      "auxiliary_loss_clip": 0.01061169,
      "auxiliary_loss_mlp": 0.01030443,
      "balance_loss_clip": 1.03865457,
      "balance_loss_mlp": 1.01780725,
      "epoch": 0.5479633248158725,
      "flos": 24278701852800.0,
      "grad_norm": 1.7496574520780568,
      "language_loss": 0.72767347,
      "learning_rate": 1.7867862107767067e-06,
      "loss": 0.74858958,
      "num_input_tokens_seen": 196417725,
      "step": 9114,
      "time_per_iteration": 2.633417844772339
    },
    {
      "auxiliary_loss_clip": 0.01084555,
      "auxiliary_loss_mlp": 0.00786272,
      "balance_loss_clip": 1.03569877,
      "balance_loss_mlp": 1.01198792,
      "epoch": 0.5480234480685405,
      "flos": 26358216860160.0,
      "grad_norm": 1.6240438191481192,
      "language_loss": 0.72103494,
      "learning_rate": 1.7863989725861066e-06,
      "loss": 0.73974323,
      "num_input_tokens_seen": 196437840,
      "step": 9115,
      "time_per_iteration": 3.8979034423828125
    },
    {
      "auxiliary_loss_clip": 0.01075949,
      "auxiliary_loss_mlp": 0.00789128,
      "balance_loss_clip": 1.0352149,
      "balance_loss_mlp": 1.01042271,
      "epoch": 0.5480835713212084,
      "flos": 22055077480320.0,
      "grad_norm": 1.630516091270635,
      "language_loss": 0.72263342,
      "learning_rate": 1.7860117424952781e-06,
      "loss": 0.74128413,
      "num_input_tokens_seen": 196457300,
      "step": 9116,
      "time_per_iteration": 2.5679996013641357
    },
    {
      "auxiliary_loss_clip": 0.01088092,
      "auxiliary_loss_mlp": 0.01033118,
      "balance_loss_clip": 1.04064107,
      "balance_loss_mlp": 1.02082181,
      "epoch": 0.5481436945738765,
      "flos": 25301042749440.0,
      "grad_norm": 1.7591338751320926,
      "language_loss": 0.76449931,
      "learning_rate": 1.7856245205189063e-06,
      "loss": 0.78571141,
      "num_input_tokens_seen": 196476720,
      "step": 9117,
      "time_per_iteration": 2.5712339878082275
    },
    {
      "auxiliary_loss_clip": 0.0106678,
      "auxiliary_loss_mlp": 0.01026374,
      "balance_loss_clip": 1.03618455,
      "balance_loss_mlp": 1.01461434,
      "epoch": 0.5482038178265444,
      "flos": 33580857772800.0,
      "grad_norm": 1.7219317281507678,
      "language_loss": 0.62723464,
      "learning_rate": 1.785237306671674e-06,
      "loss": 0.64816618,
      "num_input_tokens_seen": 196496765,
      "step": 9118,
      "time_per_iteration": 4.037683010101318
    },
    {
      "auxiliary_loss_clip": 0.01114479,
      "auxiliary_loss_mlp": 0.01028156,
      "balance_loss_clip": 1.0398767,
      "balance_loss_mlp": 1.01498902,
      "epoch": 0.5482639410792124,
      "flos": 19026192055680.0,
      "grad_norm": 1.9021889402381704,
      "language_loss": 0.79078496,
      "learning_rate": 1.7848501009682646e-06,
      "loss": 0.81221128,
      "num_input_tokens_seen": 196516220,
      "step": 9119,
      "time_per_iteration": 2.4755282402038574
    },
    {
      "auxiliary_loss_clip": 0.01086226,
      "auxiliary_loss_mlp": 0.00783113,
      "balance_loss_clip": 1.04251456,
      "balance_loss_mlp": 1.00897145,
      "epoch": 0.5483240643318803,
      "flos": 25410318900480.0,
      "grad_norm": 1.952237986788984,
      "language_loss": 0.82475519,
      "learning_rate": 1.7844629034233604e-06,
      "loss": 0.84344864,
      "num_input_tokens_seen": 196533860,
      "step": 9120,
      "time_per_iteration": 2.570537567138672
    },
    {
      "auxiliary_loss_clip": 0.01078023,
      "auxiliary_loss_mlp": 0.01032149,
      "balance_loss_clip": 1.0377388,
      "balance_loss_mlp": 1.0195483,
      "epoch": 0.5483841875845483,
      "flos": 21466896272640.0,
      "grad_norm": 3.214409804996618,
      "language_loss": 0.80613893,
      "learning_rate": 1.7840757140516455e-06,
      "loss": 0.82724071,
      "num_input_tokens_seen": 196551305,
      "step": 9121,
      "time_per_iteration": 4.053008794784546
    },
    {
      "auxiliary_loss_clip": 0.0106819,
      "auxiliary_loss_mlp": 0.0103326,
      "balance_loss_clip": 1.03818846,
      "balance_loss_mlp": 1.02015328,
      "epoch": 0.5484443108372163,
      "flos": 24747263792640.0,
      "grad_norm": 1.720306074841834,
      "language_loss": 0.60944057,
      "learning_rate": 1.7836885328678008e-06,
      "loss": 0.63045508,
      "num_input_tokens_seen": 196569420,
      "step": 9122,
      "time_per_iteration": 2.6311779022216797
    },
    {
      "auxiliary_loss_clip": 0.01090232,
      "auxiliary_loss_mlp": 0.01032653,
      "balance_loss_clip": 1.03752279,
      "balance_loss_mlp": 1.02196038,
      "epoch": 0.5485044340898843,
      "flos": 25375377945600.0,
      "grad_norm": 1.7375714350613267,
      "language_loss": 0.71759999,
      "learning_rate": 1.7833013598865084e-06,
      "loss": 0.7388289,
      "num_input_tokens_seen": 196590610,
      "step": 9123,
      "time_per_iteration": 2.5749564170837402
    },
    {
      "auxiliary_loss_clip": 0.0110989,
      "auxiliary_loss_mlp": 0.01028374,
      "balance_loss_clip": 1.0371964,
      "balance_loss_mlp": 1.01691175,
      "epoch": 0.5485645573425523,
      "flos": 12641167370880.0,
      "grad_norm": 2.9465734215715575,
      "language_loss": 0.83138907,
      "learning_rate": 1.7829141951224505e-06,
      "loss": 0.85277164,
      "num_input_tokens_seen": 196606495,
      "step": 9124,
      "time_per_iteration": 2.4362239837646484
    },
    {
      "auxiliary_loss_clip": 0.01087859,
      "auxiliary_loss_mlp": 0.01029698,
      "balance_loss_clip": 1.04143906,
      "balance_loss_mlp": 1.01805115,
      "epoch": 0.5486246805952202,
      "flos": 28329425383680.0,
      "grad_norm": 1.5270193027120782,
      "language_loss": 0.80174333,
      "learning_rate": 1.7825270385903075e-06,
      "loss": 0.82291889,
      "num_input_tokens_seen": 196626365,
      "step": 9125,
      "time_per_iteration": 2.6047277450561523
    },
    {
      "auxiliary_loss_clip": 0.01100805,
      "auxiliary_loss_mlp": 0.01030935,
      "balance_loss_clip": 1.03758359,
      "balance_loss_mlp": 1.01884103,
      "epoch": 0.5486848038478882,
      "flos": 16800017817600.0,
      "grad_norm": 2.016757199992043,
      "language_loss": 0.74316865,
      "learning_rate": 1.7821398903047617e-06,
      "loss": 0.76448607,
      "num_input_tokens_seen": 196644465,
      "step": 9126,
      "time_per_iteration": 2.4601564407348633
    },
    {
      "auxiliary_loss_clip": 0.01099153,
      "auxiliary_loss_mlp": 0.01032595,
      "balance_loss_clip": 1.03665328,
      "balance_loss_mlp": 1.0192014,
      "epoch": 0.5487449271005561,
      "flos": 17236224581760.0,
      "grad_norm": 2.2354961476678583,
      "language_loss": 0.66772574,
      "learning_rate": 1.7817527502804928e-06,
      "loss": 0.68904322,
      "num_input_tokens_seen": 196659160,
      "step": 9127,
      "time_per_iteration": 2.470860004425049
    },
    {
      "auxiliary_loss_clip": 0.01078002,
      "auxiliary_loss_mlp": 0.0103292,
      "balance_loss_clip": 1.03862715,
      "balance_loss_mlp": 1.01978862,
      "epoch": 0.5488050503532241,
      "flos": 17340867878400.0,
      "grad_norm": 1.749484320955835,
      "language_loss": 0.83326924,
      "learning_rate": 1.781365618532181e-06,
      "loss": 0.85437846,
      "num_input_tokens_seen": 196677410,
      "step": 9128,
      "time_per_iteration": 2.560222625732422
    },
    {
      "auxiliary_loss_clip": 0.01071128,
      "auxiliary_loss_mlp": 0.01029936,
      "balance_loss_clip": 1.0409596,
      "balance_loss_mlp": 1.01756835,
      "epoch": 0.548865173605892,
      "flos": 17239169496960.0,
      "grad_norm": 2.1339805237501666,
      "language_loss": 0.74420315,
      "learning_rate": 1.7809784950745078e-06,
      "loss": 0.76521379,
      "num_input_tokens_seen": 196696765,
      "step": 9129,
      "time_per_iteration": 2.591024875640869
    },
    {
      "auxiliary_loss_clip": 0.01073561,
      "auxiliary_loss_mlp": 0.01030011,
      "balance_loss_clip": 1.03966689,
      "balance_loss_mlp": 1.01664734,
      "epoch": 0.5489252968585601,
      "flos": 17456716218240.0,
      "grad_norm": 2.98776978728015,
      "language_loss": 0.6269443,
      "learning_rate": 1.7805913799221511e-06,
      "loss": 0.64798003,
      "num_input_tokens_seen": 196714895,
      "step": 9130,
      "time_per_iteration": 2.529120683670044
    },
    {
      "auxiliary_loss_clip": 0.01112856,
      "auxiliary_loss_mlp": 0.00786545,
      "balance_loss_clip": 1.03765893,
      "balance_loss_mlp": 1.01075208,
      "epoch": 0.548985420111228,
      "flos": 26323383646080.0,
      "grad_norm": 1.7818070292069321,
      "language_loss": 0.62715018,
      "learning_rate": 1.7802042730897915e-06,
      "loss": 0.64614415,
      "num_input_tokens_seen": 196735510,
      "step": 9131,
      "time_per_iteration": 2.5174167156219482
    },
    {
      "auxiliary_loss_clip": 0.01101985,
      "auxiliary_loss_mlp": 0.01032821,
      "balance_loss_clip": 1.03668141,
      "balance_loss_mlp": 1.01893258,
      "epoch": 0.549045543363896,
      "flos": 18693730748160.0,
      "grad_norm": 2.164578285563406,
      "language_loss": 0.7472018,
      "learning_rate": 1.7798171745921084e-06,
      "loss": 0.76854986,
      "num_input_tokens_seen": 196752855,
      "step": 9132,
      "time_per_iteration": 2.4597842693328857
    },
    {
      "auxiliary_loss_clip": 0.01097992,
      "auxiliary_loss_mlp": 0.01026874,
      "balance_loss_clip": 1.03401458,
      "balance_loss_mlp": 1.0151732,
      "epoch": 0.5491056666165639,
      "flos": 24717386655360.0,
      "grad_norm": 1.555481337925201,
      "language_loss": 0.8123821,
      "learning_rate": 1.7794300844437795e-06,
      "loss": 0.83363074,
      "num_input_tokens_seen": 196772230,
      "step": 9133,
      "time_per_iteration": 2.5265960693359375
    },
    {
      "auxiliary_loss_clip": 0.01086135,
      "auxiliary_loss_mlp": 0.00784822,
      "balance_loss_clip": 1.03546023,
      "balance_loss_mlp": 1.00997472,
      "epoch": 0.5491657898692319,
      "flos": 21576926609280.0,
      "grad_norm": 2.032429761792675,
      "language_loss": 0.7009443,
      "learning_rate": 1.7790430026594841e-06,
      "loss": 0.71965384,
      "num_input_tokens_seen": 196790405,
      "step": 9134,
      "time_per_iteration": 2.509330987930298
    },
    {
      "auxiliary_loss_clip": 0.01083527,
      "auxiliary_loss_mlp": 0.01031705,
      "balance_loss_clip": 1.03638935,
      "balance_loss_mlp": 1.0198679,
      "epoch": 0.5492259131219,
      "flos": 50476432746240.0,
      "grad_norm": 2.3434157482377524,
      "language_loss": 0.61746079,
      "learning_rate": 1.7786559292539004e-06,
      "loss": 0.6386131,
      "num_input_tokens_seen": 196813785,
      "step": 9135,
      "time_per_iteration": 2.8207311630249023
    },
    {
      "auxiliary_loss_clip": 0.01101433,
      "auxiliary_loss_mlp": 0.0102837,
      "balance_loss_clip": 1.03707528,
      "balance_loss_mlp": 1.01509643,
      "epoch": 0.5492860363745679,
      "flos": 25119262995840.0,
      "grad_norm": 1.9605175986547425,
      "language_loss": 0.7257266,
      "learning_rate": 1.7782688642417058e-06,
      "loss": 0.74702466,
      "num_input_tokens_seen": 196834390,
      "step": 9136,
      "time_per_iteration": 2.5177831649780273
    },
    {
      "auxiliary_loss_clip": 0.01052819,
      "auxiliary_loss_mlp": 0.01033173,
      "balance_loss_clip": 1.04112053,
      "balance_loss_mlp": 1.0194577,
      "epoch": 0.5493461596272359,
      "flos": 22633777497600.0,
      "grad_norm": 2.1261136285034543,
      "language_loss": 0.67374837,
      "learning_rate": 1.7778818076375781e-06,
      "loss": 0.69460833,
      "num_input_tokens_seen": 196853290,
      "step": 9137,
      "time_per_iteration": 2.640841245651245
    },
    {
      "auxiliary_loss_clip": 0.01034908,
      "auxiliary_loss_mlp": 0.01002308,
      "balance_loss_clip": 1.01946437,
      "balance_loss_mlp": 1.00115192,
      "epoch": 0.5494062828799038,
      "flos": 66151800754560.0,
      "grad_norm": 0.740875386583104,
      "language_loss": 0.65232223,
      "learning_rate": 1.7774947594561947e-06,
      "loss": 0.67269439,
      "num_input_tokens_seen": 196913120,
      "step": 9138,
      "time_per_iteration": 3.1284584999084473
    },
    {
      "auxiliary_loss_clip": 0.01100065,
      "auxiliary_loss_mlp": 0.0102861,
      "balance_loss_clip": 1.03799915,
      "balance_loss_mlp": 1.01606345,
      "epoch": 0.5494664061325718,
      "flos": 21105958458240.0,
      "grad_norm": 2.3287675391998355,
      "language_loss": 0.74886763,
      "learning_rate": 1.7771077197122321e-06,
      "loss": 0.77015442,
      "num_input_tokens_seen": 196931530,
      "step": 9139,
      "time_per_iteration": 2.500924825668335
    },
    {
      "auxiliary_loss_clip": 0.01100264,
      "auxiliary_loss_mlp": 0.01026412,
      "balance_loss_clip": 1.0370723,
      "balance_loss_mlp": 1.01418126,
      "epoch": 0.5495265293852397,
      "flos": 14392566616320.0,
      "grad_norm": 1.7920903742566054,
      "language_loss": 0.71093786,
      "learning_rate": 1.7767206884203672e-06,
      "loss": 0.73220462,
      "num_input_tokens_seen": 196949430,
      "step": 9140,
      "time_per_iteration": 2.4827287197113037
    },
    {
      "auxiliary_loss_clip": 0.01083584,
      "auxiliary_loss_mlp": 0.01033285,
      "balance_loss_clip": 1.03614902,
      "balance_loss_mlp": 1.02045166,
      "epoch": 0.5495866526379077,
      "flos": 25549148966400.0,
      "grad_norm": 1.6577703854096209,
      "language_loss": 0.76387066,
      "learning_rate": 1.7763336655952762e-06,
      "loss": 0.78503931,
      "num_input_tokens_seen": 196968265,
      "step": 9141,
      "time_per_iteration": 2.5839693546295166
    },
    {
      "auxiliary_loss_clip": 0.01074439,
      "auxiliary_loss_mlp": 0.01028393,
      "balance_loss_clip": 1.03862727,
      "balance_loss_mlp": 1.01675797,
      "epoch": 0.5496467758905756,
      "flos": 21317256213120.0,
      "grad_norm": 1.9563277771403476,
      "language_loss": 0.75034821,
      "learning_rate": 1.7759466512516346e-06,
      "loss": 0.77137649,
      "num_input_tokens_seen": 196984930,
      "step": 9142,
      "time_per_iteration": 2.5676512718200684
    },
    {
      "auxiliary_loss_clip": 0.01095419,
      "auxiliary_loss_mlp": 0.01030739,
      "balance_loss_clip": 1.03950453,
      "balance_loss_mlp": 1.01739311,
      "epoch": 0.5497068991432437,
      "flos": 22233086305920.0,
      "grad_norm": 1.9379801582030713,
      "language_loss": 0.76630896,
      "learning_rate": 1.7755596454041192e-06,
      "loss": 0.78757054,
      "num_input_tokens_seen": 197002320,
      "step": 9143,
      "time_per_iteration": 2.544015407562256
    },
    {
      "auxiliary_loss_clip": 0.01081911,
      "auxiliary_loss_mlp": 0.01033129,
      "balance_loss_clip": 1.03471541,
      "balance_loss_mlp": 1.02033782,
      "epoch": 0.5497670223959116,
      "flos": 18479093028480.0,
      "grad_norm": 4.099580023777893,
      "language_loss": 0.80330348,
      "learning_rate": 1.7751726480674044e-06,
      "loss": 0.82445389,
      "num_input_tokens_seen": 197020825,
      "step": 9144,
      "time_per_iteration": 2.489915370941162
    },
    {
      "auxiliary_loss_clip": 0.01101648,
      "auxiliary_loss_mlp": 0.01027099,
      "balance_loss_clip": 1.03811932,
      "balance_loss_mlp": 1.01513064,
      "epoch": 0.5498271456485796,
      "flos": 29205107049600.0,
      "grad_norm": 1.8185254524841954,
      "language_loss": 0.71269548,
      "learning_rate": 1.7747856592561645e-06,
      "loss": 0.73398298,
      "num_input_tokens_seen": 197040450,
      "step": 9145,
      "time_per_iteration": 2.560147523880005
    },
    {
      "auxiliary_loss_clip": 0.01100681,
      "auxiliary_loss_mlp": 0.01027319,
      "balance_loss_clip": 1.03704405,
      "balance_loss_mlp": 1.01570237,
      "epoch": 0.5498872689012475,
      "flos": 34824372664320.0,
      "grad_norm": 1.760744688100453,
      "language_loss": 0.7038452,
      "learning_rate": 1.774398678985076e-06,
      "loss": 0.72512519,
      "num_input_tokens_seen": 197063930,
      "step": 9146,
      "time_per_iteration": 2.6060314178466797
    },
    {
      "auxiliary_loss_clip": 0.01084148,
      "auxiliary_loss_mlp": 0.0102818,
      "balance_loss_clip": 1.03428125,
      "balance_loss_mlp": 1.01687932,
      "epoch": 0.5499473921539155,
      "flos": 25921938268800.0,
      "grad_norm": 1.824634066142312,
      "language_loss": 0.64010239,
      "learning_rate": 1.7740117072688113e-06,
      "loss": 0.66122568,
      "num_input_tokens_seen": 197082660,
      "step": 9147,
      "time_per_iteration": 3.919830083847046
    },
    {
      "auxiliary_loss_clip": 0.01111746,
      "auxiliary_loss_mlp": 0.01031343,
      "balance_loss_clip": 1.03857374,
      "balance_loss_mlp": 1.01932681,
      "epoch": 0.5500075154065835,
      "flos": 22273701609600.0,
      "grad_norm": 2.0296121353461873,
      "language_loss": 0.8074317,
      "learning_rate": 1.7736247441220458e-06,
      "loss": 0.82886255,
      "num_input_tokens_seen": 197100675,
      "step": 9148,
      "time_per_iteration": 2.473344087600708
    },
    {
      "auxiliary_loss_clip": 0.01089832,
      "auxiliary_loss_mlp": 0.0103612,
      "balance_loss_clip": 1.03807521,
      "balance_loss_mlp": 1.02421677,
      "epoch": 0.5500676386592515,
      "flos": 28037507552640.0,
      "grad_norm": 1.7689018046858889,
      "language_loss": 0.79149961,
      "learning_rate": 1.773237789559453e-06,
      "loss": 0.81275916,
      "num_input_tokens_seen": 197121320,
      "step": 9149,
      "time_per_iteration": 2.5751101970672607
    },
    {
      "auxiliary_loss_clip": 0.01075527,
      "auxiliary_loss_mlp": 0.01027371,
      "balance_loss_clip": 1.03975415,
      "balance_loss_mlp": 1.0151639,
      "epoch": 0.5501277619119195,
      "flos": 23914819123200.0,
      "grad_norm": 2.0292330639041136,
      "language_loss": 0.71273017,
      "learning_rate": 1.7728508435957052e-06,
      "loss": 0.73375916,
      "num_input_tokens_seen": 197138965,
      "step": 9150,
      "time_per_iteration": 2.559328317642212
    },
    {
      "auxiliary_loss_clip": 0.01090551,
      "auxiliary_loss_mlp": 0.01027186,
      "balance_loss_clip": 1.03465724,
      "balance_loss_mlp": 1.01414502,
      "epoch": 0.5501878851645874,
      "flos": 20923783655040.0,
      "grad_norm": 2.1607337095503167,
      "language_loss": 0.74963742,
      "learning_rate": 1.772463906245477e-06,
      "loss": 0.77081478,
      "num_input_tokens_seen": 197156460,
      "step": 9151,
      "time_per_iteration": 2.5070278644561768
    },
    {
      "auxiliary_loss_clip": 0.0108567,
      "auxiliary_loss_mlp": 0.01026133,
      "balance_loss_clip": 1.03946686,
      "balance_loss_mlp": 1.01476669,
      "epoch": 0.5502480084172554,
      "flos": 20665298407680.0,
      "grad_norm": 2.319543398417275,
      "language_loss": 0.76117802,
      "learning_rate": 1.7720769775234394e-06,
      "loss": 0.78229606,
      "num_input_tokens_seen": 197175140,
      "step": 9152,
      "time_per_iteration": 2.5324277877807617
    },
    {
      "auxiliary_loss_clip": 0.0108517,
      "auxiliary_loss_mlp": 0.0103224,
      "balance_loss_clip": 1.03517222,
      "balance_loss_mlp": 1.02073014,
      "epoch": 0.5503081316699233,
      "flos": 26432552056320.0,
      "grad_norm": 2.486708010974758,
      "language_loss": 0.82483357,
      "learning_rate": 1.7716900574442662e-06,
      "loss": 0.8460077,
      "num_input_tokens_seen": 197194345,
      "step": 9153,
      "time_per_iteration": 3.968703031539917
    },
    {
      "auxiliary_loss_clip": 0.01097479,
      "auxiliary_loss_mlp": 0.0102813,
      "balance_loss_clip": 1.03706956,
      "balance_loss_mlp": 1.01588094,
      "epoch": 0.5503682549225913,
      "flos": 30629144718720.0,
      "grad_norm": 1.815863827952672,
      "language_loss": 0.74294901,
      "learning_rate": 1.7713031460226294e-06,
      "loss": 0.7642051,
      "num_input_tokens_seen": 197215535,
      "step": 9154,
      "time_per_iteration": 2.558206081390381
    },
    {
      "auxiliary_loss_clip": 0.01093477,
      "auxiliary_loss_mlp": 0.01032964,
      "balance_loss_clip": 1.03595996,
      "balance_loss_mlp": 1.0196718,
      "epoch": 0.5504283781752592,
      "flos": 22565439872640.0,
      "grad_norm": 1.6870966290325398,
      "language_loss": 0.727575,
      "learning_rate": 1.770916243273199e-06,
      "loss": 0.74883944,
      "num_input_tokens_seen": 197234945,
      "step": 9155,
      "time_per_iteration": 2.5434789657592773
    },
    {
      "auxiliary_loss_clip": 0.01028537,
      "auxiliary_loss_mlp": 0.01006263,
      "balance_loss_clip": 1.01329494,
      "balance_loss_mlp": 1.00500488,
      "epoch": 0.5504885014279273,
      "flos": 67901009270400.0,
      "grad_norm": 0.7559758096156651,
      "language_loss": 0.55377138,
      "learning_rate": 1.7705293492106483e-06,
      "loss": 0.57411933,
      "num_input_tokens_seen": 197302285,
      "step": 9156,
      "time_per_iteration": 3.228513717651367
    },
    {
      "auxiliary_loss_clip": 0.01098842,
      "auxiliary_loss_mlp": 0.01032227,
      "balance_loss_clip": 1.03650784,
      "balance_loss_mlp": 1.02044892,
      "epoch": 0.5505486246805952,
      "flos": 22450058409600.0,
      "grad_norm": 1.7116578908025295,
      "language_loss": 0.82887399,
      "learning_rate": 1.7701424638496475e-06,
      "loss": 0.85018468,
      "num_input_tokens_seen": 197321575,
      "step": 9157,
      "time_per_iteration": 4.127148151397705
    },
    {
      "auxiliary_loss_clip": 0.01115824,
      "auxiliary_loss_mlp": 0.01031216,
      "balance_loss_clip": 1.03888226,
      "balance_loss_mlp": 1.01785862,
      "epoch": 0.5506087479332632,
      "flos": 26906896085760.0,
      "grad_norm": 3.8994073894539625,
      "language_loss": 0.76044083,
      "learning_rate": 1.7697555872048677e-06,
      "loss": 0.78191119,
      "num_input_tokens_seen": 197340255,
      "step": 9158,
      "time_per_iteration": 2.5094587802886963
    },
    {
      "auxiliary_loss_clip": 0.01068013,
      "auxiliary_loss_mlp": 0.01026332,
      "balance_loss_clip": 1.03470111,
      "balance_loss_mlp": 1.01466155,
      "epoch": 0.5506688711859311,
      "flos": 22930256355840.0,
      "grad_norm": 1.8901565007315186,
      "language_loss": 0.69496763,
      "learning_rate": 1.769368719290979e-06,
      "loss": 0.71591109,
      "num_input_tokens_seen": 197360360,
      "step": 9159,
      "time_per_iteration": 2.62975811958313
    },
    {
      "auxiliary_loss_clip": 0.01065368,
      "auxiliary_loss_mlp": 0.00786873,
      "balance_loss_clip": 1.03515744,
      "balance_loss_mlp": 1.0104928,
      "epoch": 0.5507289944385991,
      "flos": 29606408772480.0,
      "grad_norm": 2.1811235040565333,
      "language_loss": 0.67977452,
      "learning_rate": 1.7689818601226516e-06,
      "loss": 0.6982969,
      "num_input_tokens_seen": 197381905,
      "step": 9160,
      "time_per_iteration": 4.135736703872681
    },
    {
      "auxiliary_loss_clip": 0.01108592,
      "auxiliary_loss_mlp": 0.01027575,
      "balance_loss_clip": 1.0374161,
      "balance_loss_mlp": 1.01608956,
      "epoch": 0.5507891176912671,
      "flos": 15334431091200.0,
      "grad_norm": 1.895495249725565,
      "language_loss": 0.71360892,
      "learning_rate": 1.7685950097145552e-06,
      "loss": 0.73497057,
      "num_input_tokens_seen": 197398555,
      "step": 9161,
      "time_per_iteration": 2.4529480934143066
    },
    {
      "auxiliary_loss_clip": 0.01099521,
      "auxiliary_loss_mlp": 0.01035047,
      "balance_loss_clip": 1.03763628,
      "balance_loss_mlp": 1.02306628,
      "epoch": 0.5508492409439351,
      "flos": 26578313447040.0,
      "grad_norm": 1.73705114642714,
      "language_loss": 0.69468468,
      "learning_rate": 1.768208168081359e-06,
      "loss": 0.71603036,
      "num_input_tokens_seen": 197419630,
      "step": 9162,
      "time_per_iteration": 2.531546115875244
    },
    {
      "auxiliary_loss_clip": 0.01108823,
      "auxiliary_loss_mlp": 0.01032896,
      "balance_loss_clip": 1.03627169,
      "balance_loss_mlp": 1.02101684,
      "epoch": 0.5509093641966031,
      "flos": 25443428261760.0,
      "grad_norm": 1.6804727551668313,
      "language_loss": 0.85837239,
      "learning_rate": 1.767821335237733e-06,
      "loss": 0.87978959,
      "num_input_tokens_seen": 197438480,
      "step": 9163,
      "time_per_iteration": 2.486949920654297
    },
    {
      "auxiliary_loss_clip": 0.01079671,
      "auxiliary_loss_mlp": 0.01029241,
      "balance_loss_clip": 1.03816891,
      "balance_loss_mlp": 1.01718926,
      "epoch": 0.550969487449271,
      "flos": 18698543170560.0,
      "grad_norm": 1.7238811900965705,
      "language_loss": 0.80695719,
      "learning_rate": 1.7674345111983441e-06,
      "loss": 0.82804632,
      "num_input_tokens_seen": 197456755,
      "step": 9164,
      "time_per_iteration": 2.5517404079437256
    },
    {
      "auxiliary_loss_clip": 0.01093426,
      "auxiliary_loss_mlp": 0.01025818,
      "balance_loss_clip": 1.0404768,
      "balance_loss_mlp": 1.01342618,
      "epoch": 0.551029610701939,
      "flos": 22708723224960.0,
      "grad_norm": 2.25742835822819,
      "language_loss": 0.73406327,
      "learning_rate": 1.767047695977863e-06,
      "loss": 0.7552557,
      "num_input_tokens_seen": 197475530,
      "step": 9165,
      "time_per_iteration": 2.557344675064087
    },
    {
      "auxiliary_loss_clip": 0.01092023,
      "auxiliary_loss_mlp": 0.01031013,
      "balance_loss_clip": 1.03421283,
      "balance_loss_mlp": 1.01835263,
      "epoch": 0.5510897339546069,
      "flos": 12420496166400.0,
      "grad_norm": 2.151564440507071,
      "language_loss": 0.7892493,
      "learning_rate": 1.7666608895909563e-06,
      "loss": 0.81047964,
      "num_input_tokens_seen": 197490835,
      "step": 9166,
      "time_per_iteration": 2.480553388595581
    },
    {
      "auxiliary_loss_clip": 0.01079158,
      "auxiliary_loss_mlp": 0.01025847,
      "balance_loss_clip": 1.03541327,
      "balance_loss_mlp": 1.01371157,
      "epoch": 0.5511498572072749,
      "flos": 18770579896320.0,
      "grad_norm": 2.216688604592005,
      "language_loss": 0.76407039,
      "learning_rate": 1.7662740920522913e-06,
      "loss": 0.78512049,
      "num_input_tokens_seen": 197508770,
      "step": 9167,
      "time_per_iteration": 2.5474066734313965
    },
    {
      "auxiliary_loss_clip": 0.01098354,
      "auxiliary_loss_mlp": 0.01027353,
      "balance_loss_clip": 1.03647351,
      "balance_loss_mlp": 1.0146811,
      "epoch": 0.5512099804599428,
      "flos": 19573326996480.0,
      "grad_norm": 2.207078315857769,
      "language_loss": 0.80298448,
      "learning_rate": 1.7658873033765374e-06,
      "loss": 0.82424152,
      "num_input_tokens_seen": 197527340,
      "step": 9168,
      "time_per_iteration": 2.477987289428711
    },
    {
      "auxiliary_loss_clip": 0.01101744,
      "auxiliary_loss_mlp": 0.01033624,
      "balance_loss_clip": 1.03748608,
      "balance_loss_mlp": 1.0216316,
      "epoch": 0.5512701037126109,
      "flos": 26245600744320.0,
      "grad_norm": 1.7487305953637575,
      "language_loss": 0.69116163,
      "learning_rate": 1.7655005235783591e-06,
      "loss": 0.71251535,
      "num_input_tokens_seen": 197547280,
      "step": 9169,
      "time_per_iteration": 2.5438039302825928
    },
    {
      "auxiliary_loss_clip": 0.01092903,
      "auxiliary_loss_mlp": 0.01023757,
      "balance_loss_clip": 1.03344059,
      "balance_loss_mlp": 1.01267052,
      "epoch": 0.5513302269652788,
      "flos": 21945406279680.0,
      "grad_norm": 2.223150093606078,
      "language_loss": 0.85670787,
      "learning_rate": 1.7651137526724251e-06,
      "loss": 0.87787449,
      "num_input_tokens_seen": 197565045,
      "step": 9170,
      "time_per_iteration": 2.4929566383361816
    },
    {
      "auxiliary_loss_clip": 0.01025906,
      "auxiliary_loss_mlp": 0.01006178,
      "balance_loss_clip": 1.02775908,
      "balance_loss_mlp": 1.0045805,
      "epoch": 0.5513903502179468,
      "flos": 68235948616320.0,
      "grad_norm": 0.7844996675351996,
      "language_loss": 0.59902245,
      "learning_rate": 1.7647269906734017e-06,
      "loss": 0.61934328,
      "num_input_tokens_seen": 197625005,
      "step": 9171,
      "time_per_iteration": 3.170624017715454
    },
    {
      "auxiliary_loss_clip": 0.01074488,
      "auxiliary_loss_mlp": 0.01033903,
      "balance_loss_clip": 1.03444338,
      "balance_loss_mlp": 1.02138019,
      "epoch": 0.5514504734706147,
      "flos": 18734238311040.0,
      "grad_norm": 1.438863379179184,
      "language_loss": 0.70347774,
      "learning_rate": 1.7643402375959533e-06,
      "loss": 0.72456169,
      "num_input_tokens_seen": 197645050,
      "step": 9172,
      "time_per_iteration": 2.5470964908599854
    },
    {
      "auxiliary_loss_clip": 0.0110833,
      "auxiliary_loss_mlp": 0.01032653,
      "balance_loss_clip": 1.0356425,
      "balance_loss_mlp": 1.02043366,
      "epoch": 0.5515105967232827,
      "flos": 22270972176000.0,
      "grad_norm": 3.0715803079433814,
      "language_loss": 0.76043999,
      "learning_rate": 1.7639534934547474e-06,
      "loss": 0.7818498,
      "num_input_tokens_seen": 197663910,
      "step": 9173,
      "time_per_iteration": 2.489335536956787
    },
    {
      "auxiliary_loss_clip": 0.010726,
      "auxiliary_loss_mlp": 0.01029931,
      "balance_loss_clip": 1.03572893,
      "balance_loss_mlp": 1.01759887,
      "epoch": 0.5515707199759508,
      "flos": 22557682535040.0,
      "grad_norm": 4.629307836348251,
      "language_loss": 0.75141406,
      "learning_rate": 1.7635667582644484e-06,
      "loss": 0.77243936,
      "num_input_tokens_seen": 197681580,
      "step": 9174,
      "time_per_iteration": 2.5552353858947754
    },
    {
      "auxiliary_loss_clip": 0.01089819,
      "auxiliary_loss_mlp": 0.01031743,
      "balance_loss_clip": 1.03563261,
      "balance_loss_mlp": 1.01952362,
      "epoch": 0.5516308432286187,
      "flos": 28291072636800.0,
      "grad_norm": 1.7588985192034923,
      "language_loss": 0.72754312,
      "learning_rate": 1.7631800320397217e-06,
      "loss": 0.74875867,
      "num_input_tokens_seen": 197702095,
      "step": 9175,
      "time_per_iteration": 2.579268217086792
    },
    {
      "auxiliary_loss_clip": 0.01099585,
      "auxiliary_loss_mlp": 0.01032682,
      "balance_loss_clip": 1.0371691,
      "balance_loss_mlp": 1.02079654,
      "epoch": 0.5516909664812867,
      "flos": 18764474584320.0,
      "grad_norm": 1.759549723090395,
      "language_loss": 0.69160211,
      "learning_rate": 1.7627933147952318e-06,
      "loss": 0.71292484,
      "num_input_tokens_seen": 197720720,
      "step": 9176,
      "time_per_iteration": 2.4819114208221436
    },
    {
      "auxiliary_loss_clip": 0.01098268,
      "auxiliary_loss_mlp": 0.01030195,
      "balance_loss_clip": 1.03603411,
      "balance_loss_mlp": 1.01827407,
      "epoch": 0.5517510897339546,
      "flos": 27740346336000.0,
      "grad_norm": 1.7950917465037164,
      "language_loss": 0.71046102,
      "learning_rate": 1.7624066065456435e-06,
      "loss": 0.73174566,
      "num_input_tokens_seen": 197741820,
      "step": 9177,
      "time_per_iteration": 2.5527186393737793
    },
    {
      "auxiliary_loss_clip": 0.01100365,
      "auxiliary_loss_mlp": 0.01028986,
      "balance_loss_clip": 1.0382092,
      "balance_loss_mlp": 1.01703501,
      "epoch": 0.5518112129866226,
      "flos": 18404470523520.0,
      "grad_norm": 1.522700806431929,
      "language_loss": 0.80141199,
      "learning_rate": 1.7620199073056204e-06,
      "loss": 0.82270551,
      "num_input_tokens_seen": 197759160,
      "step": 9178,
      "time_per_iteration": 2.473494291305542
    },
    {
      "auxiliary_loss_clip": 0.0105906,
      "auxiliary_loss_mlp": 0.01042042,
      "balance_loss_clip": 1.03941488,
      "balance_loss_mlp": 1.02830958,
      "epoch": 0.5518713362392905,
      "flos": 25082670015360.0,
      "grad_norm": 1.958444900132511,
      "language_loss": 0.74800748,
      "learning_rate": 1.761633217089826e-06,
      "loss": 0.76901847,
      "num_input_tokens_seen": 197779760,
      "step": 9179,
      "time_per_iteration": 2.6429245471954346
    },
    {
      "auxiliary_loss_clip": 0.01100626,
      "auxiliary_loss_mlp": 0.01035327,
      "balance_loss_clip": 1.03754091,
      "balance_loss_mlp": 1.02354264,
      "epoch": 0.5519314594919585,
      "flos": 36538999361280.0,
      "grad_norm": 1.941967734829439,
      "language_loss": 0.70055342,
      "learning_rate": 1.761246535912924e-06,
      "loss": 0.72191298,
      "num_input_tokens_seen": 197801545,
      "step": 9180,
      "time_per_iteration": 2.6099414825439453
    },
    {
      "auxiliary_loss_clip": 0.01097539,
      "auxiliary_loss_mlp": 0.01041077,
      "balance_loss_clip": 1.03737533,
      "balance_loss_mlp": 1.0280714,
      "epoch": 0.5519915827446265,
      "flos": 20448613612800.0,
      "grad_norm": 1.8775381149689536,
      "language_loss": 0.66910481,
      "learning_rate": 1.7608598637895776e-06,
      "loss": 0.69049102,
      "num_input_tokens_seen": 197820760,
      "step": 9181,
      "time_per_iteration": 2.484083652496338
    },
    {
      "auxiliary_loss_clip": 0.01112912,
      "auxiliary_loss_mlp": 0.01031289,
      "balance_loss_clip": 1.03732347,
      "balance_loss_mlp": 1.01865244,
      "epoch": 0.5520517059972945,
      "flos": 23768052151680.0,
      "grad_norm": 1.8881515107107425,
      "language_loss": 0.79315281,
      "learning_rate": 1.7604732007344486e-06,
      "loss": 0.81459486,
      "num_input_tokens_seen": 197840195,
      "step": 9182,
      "time_per_iteration": 2.461601734161377
    },
    {
      "auxiliary_loss_clip": 0.01076978,
      "auxiliary_loss_mlp": 0.01029311,
      "balance_loss_clip": 1.03972721,
      "balance_loss_mlp": 1.01741982,
      "epoch": 0.5521118292499624,
      "flos": 22196457411840.0,
      "grad_norm": 1.9807399050987984,
      "language_loss": 0.8253656,
      "learning_rate": 1.7600865467622003e-06,
      "loss": 0.84642851,
      "num_input_tokens_seen": 197859475,
      "step": 9183,
      "time_per_iteration": 2.577390432357788
    },
    {
      "auxiliary_loss_clip": 0.01087516,
      "auxiliary_loss_mlp": 0.01027296,
      "balance_loss_clip": 1.03545189,
      "balance_loss_mlp": 1.01515436,
      "epoch": 0.5521719525026304,
      "flos": 23583291569280.0,
      "grad_norm": 1.3691207812979886,
      "language_loss": 0.67124712,
      "learning_rate": 1.7596999018874936e-06,
      "loss": 0.69239521,
      "num_input_tokens_seen": 197879395,
      "step": 9184,
      "time_per_iteration": 2.5361216068267822
    },
    {
      "auxiliary_loss_clip": 0.01099133,
      "auxiliary_loss_mlp": 0.01027031,
      "balance_loss_clip": 1.03682065,
      "balance_loss_mlp": 1.01372719,
      "epoch": 0.5522320757552983,
      "flos": 26137617482880.0,
      "grad_norm": 1.5643196818495981,
      "language_loss": 0.76349288,
      "learning_rate": 1.7593132661249917e-06,
      "loss": 0.78475451,
      "num_input_tokens_seen": 197900815,
      "step": 9185,
      "time_per_iteration": 2.5326552391052246
    },
    {
      "auxiliary_loss_clip": 0.01074204,
      "auxiliary_loss_mlp": 0.01036946,
      "balance_loss_clip": 1.03681254,
      "balance_loss_mlp": 1.02404797,
      "epoch": 0.5522921990079663,
      "flos": 24676160820480.0,
      "grad_norm": 1.656327483033755,
      "language_loss": 0.73878682,
      "learning_rate": 1.7589266394893536e-06,
      "loss": 0.75989836,
      "num_input_tokens_seen": 197918985,
      "step": 9186,
      "time_per_iteration": 3.9809212684631348
    },
    {
      "auxiliary_loss_clip": 0.0108285,
      "auxiliary_loss_mlp": 0.01039482,
      "balance_loss_clip": 1.03874826,
      "balance_loss_mlp": 1.0276866,
      "epoch": 0.5523523222606344,
      "flos": 22748153379840.0,
      "grad_norm": 4.473836625042513,
      "language_loss": 0.66705084,
      "learning_rate": 1.7585400219952421e-06,
      "loss": 0.68827415,
      "num_input_tokens_seen": 197937725,
      "step": 9187,
      "time_per_iteration": 2.562554359436035
    },
    {
      "auxiliary_loss_clip": 0.01089216,
      "auxiliary_loss_mlp": 0.01028824,
      "balance_loss_clip": 1.03812981,
      "balance_loss_mlp": 1.01703405,
      "epoch": 0.5524124455133023,
      "flos": 19755825022080.0,
      "grad_norm": 4.89989645539051,
      "language_loss": 0.77515376,
      "learning_rate": 1.758153413657318e-06,
      "loss": 0.79633415,
      "num_input_tokens_seen": 197955635,
      "step": 9188,
      "time_per_iteration": 2.511859655380249
    },
    {
      "auxiliary_loss_clip": 0.01085198,
      "auxiliary_loss_mlp": 0.01031317,
      "balance_loss_clip": 1.03534079,
      "balance_loss_mlp": 1.01836526,
      "epoch": 0.5524725687659703,
      "flos": 23294821443840.0,
      "grad_norm": 1.731031988092773,
      "language_loss": 0.81078517,
      "learning_rate": 1.7577668144902394e-06,
      "loss": 0.83195031,
      "num_input_tokens_seen": 197974490,
      "step": 9189,
      "time_per_iteration": 2.5119194984436035
    },
    {
      "auxiliary_loss_clip": 0.01096036,
      "auxiliary_loss_mlp": 0.00786532,
      "balance_loss_clip": 1.03920031,
      "balance_loss_mlp": 1.01278746,
      "epoch": 0.5525326920186382,
      "flos": 24862178378880.0,
      "grad_norm": 1.548034912015803,
      "language_loss": 0.76317739,
      "learning_rate": 1.7573802245086684e-06,
      "loss": 0.78200305,
      "num_input_tokens_seen": 197995735,
      "step": 9190,
      "time_per_iteration": 2.5424177646636963
    },
    {
      "auxiliary_loss_clip": 0.01113975,
      "auxiliary_loss_mlp": 0.0103527,
      "balance_loss_clip": 1.03742433,
      "balance_loss_mlp": 1.02150106,
      "epoch": 0.5525928152713062,
      "flos": 13735580906880.0,
      "grad_norm": 2.3840229457497153,
      "language_loss": 0.78950357,
      "learning_rate": 1.7569936437272627e-06,
      "loss": 0.81099606,
      "num_input_tokens_seen": 198009685,
      "step": 9191,
      "time_per_iteration": 2.4241771697998047
    },
    {
      "auxiliary_loss_clip": 0.0104746,
      "auxiliary_loss_mlp": 0.0103459,
      "balance_loss_clip": 1.04087448,
      "balance_loss_mlp": 1.02188194,
      "epoch": 0.5526529385239741,
      "flos": 13071592045440.0,
      "grad_norm": 2.0273667058435345,
      "language_loss": 0.68775868,
      "learning_rate": 1.7566070721606829e-06,
      "loss": 0.70857918,
      "num_input_tokens_seen": 198026845,
      "step": 9192,
      "time_per_iteration": 2.592529773712158
    },
    {
      "auxiliary_loss_clip": 0.01097099,
      "auxiliary_loss_mlp": 0.01029051,
      "balance_loss_clip": 1.0364418,
      "balance_loss_mlp": 1.0181495,
      "epoch": 0.5527130617766421,
      "flos": 23148377694720.0,
      "grad_norm": 1.6134222717304798,
      "language_loss": 0.77370894,
      "learning_rate": 1.756220509823588e-06,
      "loss": 0.79497045,
      "num_input_tokens_seen": 198045275,
      "step": 9193,
      "time_per_iteration": 3.849308967590332
    },
    {
      "auxiliary_loss_clip": 0.01074553,
      "auxiliary_loss_mlp": 0.01032571,
      "balance_loss_clip": 1.03490388,
      "balance_loss_mlp": 1.02011919,
      "epoch": 0.55277318502931,
      "flos": 21285547482240.0,
      "grad_norm": 1.549976468829453,
      "language_loss": 0.78372353,
      "learning_rate": 1.7558339567306344e-06,
      "loss": 0.80479479,
      "num_input_tokens_seen": 198065760,
      "step": 9194,
      "time_per_iteration": 2.5448360443115234
    },
    {
      "auxiliary_loss_clip": 0.01083978,
      "auxiliary_loss_mlp": 0.01030934,
      "balance_loss_clip": 1.03762555,
      "balance_loss_mlp": 1.01803517,
      "epoch": 0.5528333082819781,
      "flos": 38324549462400.0,
      "grad_norm": 2.241780925080222,
      "language_loss": 0.69620025,
      "learning_rate": 1.7554474128964825e-06,
      "loss": 0.71734935,
      "num_input_tokens_seen": 198087595,
      "step": 9195,
      "time_per_iteration": 4.048158168792725
    },
    {
      "auxiliary_loss_clip": 0.01096879,
      "auxiliary_loss_mlp": 0.01032627,
      "balance_loss_clip": 1.03946328,
      "balance_loss_mlp": 1.01920366,
      "epoch": 0.552893431534646,
      "flos": 13553621585280.0,
      "grad_norm": 2.3022190335135977,
      "language_loss": 0.7398985,
      "learning_rate": 1.7550608783357887e-06,
      "loss": 0.76119357,
      "num_input_tokens_seen": 198104620,
      "step": 9196,
      "time_per_iteration": 2.483397960662842
    },
    {
      "auxiliary_loss_clip": 0.01094451,
      "auxiliary_loss_mlp": 0.01032259,
      "balance_loss_clip": 1.03725684,
      "balance_loss_mlp": 1.01969433,
      "epoch": 0.552953554787314,
      "flos": 21939408708480.0,
      "grad_norm": 1.5651130025717144,
      "language_loss": 0.76774412,
      "learning_rate": 1.7546743530632115e-06,
      "loss": 0.78901124,
      "num_input_tokens_seen": 198123565,
      "step": 9197,
      "time_per_iteration": 2.484361410140991
    },
    {
      "auxiliary_loss_clip": 0.01087518,
      "auxiliary_loss_mlp": 0.01026839,
      "balance_loss_clip": 1.03640032,
      "balance_loss_mlp": 1.01523972,
      "epoch": 0.5530136780399819,
      "flos": 43658002558080.0,
      "grad_norm": 1.628978080815795,
      "language_loss": 0.76238602,
      "learning_rate": 1.754287837093407e-06,
      "loss": 0.78352964,
      "num_input_tokens_seen": 198148270,
      "step": 9198,
      "time_per_iteration": 2.73512601852417
    },
    {
      "auxiliary_loss_clip": 0.0110815,
      "auxiliary_loss_mlp": 0.01025125,
      "balance_loss_clip": 1.03572226,
      "balance_loss_mlp": 1.01363301,
      "epoch": 0.5530738012926499,
      "flos": 25045502417280.0,
      "grad_norm": 1.4706444323606764,
      "language_loss": 0.79431868,
      "learning_rate": 1.7539013304410327e-06,
      "loss": 0.81565142,
      "num_input_tokens_seen": 198168810,
      "step": 9199,
      "time_per_iteration": 3.9382777214050293
    },
    {
      "auxiliary_loss_clip": 0.01071513,
      "auxiliary_loss_mlp": 0.01033742,
      "balance_loss_clip": 1.03774548,
      "balance_loss_mlp": 1.02050972,
      "epoch": 0.553133924545318,
      "flos": 16472081623680.0,
      "grad_norm": 1.8433772572409615,
      "language_loss": 0.64030671,
      "learning_rate": 1.7535148331207443e-06,
      "loss": 0.66135925,
      "num_input_tokens_seen": 198186200,
      "step": 9200,
      "time_per_iteration": 2.521204948425293
    },
    {
      "auxiliary_loss_clip": 0.01092547,
      "auxiliary_loss_mlp": 0.01027662,
      "balance_loss_clip": 1.03884149,
      "balance_loss_mlp": 1.01372004,
      "epoch": 0.5531940477979859,
      "flos": 24606207083520.0,
      "grad_norm": 1.9050925246039874,
      "language_loss": 0.66281343,
      "learning_rate": 1.7531283451471978e-06,
      "loss": 0.68401545,
      "num_input_tokens_seen": 198207050,
      "step": 9201,
      "time_per_iteration": 2.5391438007354736
    },
    {
      "auxiliary_loss_clip": 0.01099361,
      "auxiliary_loss_mlp": 0.0103532,
      "balance_loss_clip": 1.03848493,
      "balance_loss_mlp": 1.02215385,
      "epoch": 0.5532541710506539,
      "flos": 22159577122560.0,
      "grad_norm": 2.068933342127184,
      "language_loss": 0.6073693,
      "learning_rate": 1.7527418665350502e-06,
      "loss": 0.62871611,
      "num_input_tokens_seen": 198224565,
      "step": 9202,
      "time_per_iteration": 2.4795942306518555
    },
    {
      "auxiliary_loss_clip": 0.01098169,
      "auxiliary_loss_mlp": 0.00785787,
      "balance_loss_clip": 1.03810966,
      "balance_loss_mlp": 1.01087368,
      "epoch": 0.5533142943033218,
      "flos": 21397265758080.0,
      "grad_norm": 1.7308005648452462,
      "language_loss": 0.6456387,
      "learning_rate": 1.7523553972989548e-06,
      "loss": 0.6644783,
      "num_input_tokens_seen": 198244790,
      "step": 9203,
      "time_per_iteration": 2.5000290870666504
    },
    {
      "auxiliary_loss_clip": 0.01098529,
      "auxiliary_loss_mlp": 0.01029641,
      "balance_loss_clip": 1.03688419,
      "balance_loss_mlp": 1.01777339,
      "epoch": 0.5533744175559898,
      "flos": 23550541344000.0,
      "grad_norm": 2.0225689797249826,
      "language_loss": 0.63412708,
      "learning_rate": 1.7519689374535683e-06,
      "loss": 0.65540874,
      "num_input_tokens_seen": 198264375,
      "step": 9204,
      "time_per_iteration": 2.5506935119628906
    },
    {
      "auxiliary_loss_clip": 0.01096509,
      "auxiliary_loss_mlp": 0.01025634,
      "balance_loss_clip": 1.03670585,
      "balance_loss_mlp": 1.01412439,
      "epoch": 0.5534345408086577,
      "flos": 24061514267520.0,
      "grad_norm": 1.4923392811623066,
      "language_loss": 0.77339876,
      "learning_rate": 1.7515824870135445e-06,
      "loss": 0.79462016,
      "num_input_tokens_seen": 198283895,
      "step": 9205,
      "time_per_iteration": 2.5277438163757324
    },
    {
      "auxiliary_loss_clip": 0.01060611,
      "auxiliary_loss_mlp": 0.01035673,
      "balance_loss_clip": 1.03573966,
      "balance_loss_mlp": 1.02306056,
      "epoch": 0.5534946640613257,
      "flos": 33771831408000.0,
      "grad_norm": 1.4455693743846116,
      "language_loss": 0.72512615,
      "learning_rate": 1.751196045993537e-06,
      "loss": 0.74608898,
      "num_input_tokens_seen": 198310035,
      "step": 9206,
      "time_per_iteration": 2.7150843143463135
    },
    {
      "auxiliary_loss_clip": 0.01064224,
      "auxiliary_loss_mlp": 0.01031903,
      "balance_loss_clip": 1.03492141,
      "balance_loss_mlp": 1.01953542,
      "epoch": 0.5535547873139937,
      "flos": 15159223526400.0,
      "grad_norm": 2.0217546338335213,
      "language_loss": 0.75455046,
      "learning_rate": 1.7508096144082012e-06,
      "loss": 0.77551174,
      "num_input_tokens_seen": 198327810,
      "step": 9207,
      "time_per_iteration": 2.5404369831085205
    },
    {
      "auxiliary_loss_clip": 0.01079858,
      "auxiliary_loss_mlp": 0.01031932,
      "balance_loss_clip": 1.03802347,
      "balance_loss_mlp": 1.01948071,
      "epoch": 0.5536149105666617,
      "flos": 16980863817600.0,
      "grad_norm": 2.4183595719787694,
      "language_loss": 0.61700737,
      "learning_rate": 1.750423192272189e-06,
      "loss": 0.6381253,
      "num_input_tokens_seen": 198343150,
      "step": 9208,
      "time_per_iteration": 2.4944663047790527
    },
    {
      "auxiliary_loss_clip": 0.01112705,
      "auxiliary_loss_mlp": 0.01032597,
      "balance_loss_clip": 1.03887272,
      "balance_loss_mlp": 1.02068794,
      "epoch": 0.5536750338193296,
      "flos": 18149935772160.0,
      "grad_norm": 2.0926900898326872,
      "language_loss": 0.64433318,
      "learning_rate": 1.7500367796001547e-06,
      "loss": 0.66578621,
      "num_input_tokens_seen": 198360925,
      "step": 9209,
      "time_per_iteration": 2.4638783931732178
    },
    {
      "auxiliary_loss_clip": 0.01075837,
      "auxiliary_loss_mlp": 0.01033201,
      "balance_loss_clip": 1.03729177,
      "balance_loss_mlp": 1.01952219,
      "epoch": 0.5537351570719976,
      "flos": 22747794243840.0,
      "grad_norm": 2.004884717263606,
      "language_loss": 0.82772017,
      "learning_rate": 1.7496503764067513e-06,
      "loss": 0.84881061,
      "num_input_tokens_seen": 198379265,
      "step": 9210,
      "time_per_iteration": 2.5391480922698975
    },
    {
      "auxiliary_loss_clip": 0.0108742,
      "auxiliary_loss_mlp": 0.01028596,
      "balance_loss_clip": 1.03505921,
      "balance_loss_mlp": 1.01711667,
      "epoch": 0.5537952803246655,
      "flos": 26356026130560.0,
      "grad_norm": 2.039373873493051,
      "language_loss": 0.72720921,
      "learning_rate": 1.74926398270663e-06,
      "loss": 0.74836934,
      "num_input_tokens_seen": 198399490,
      "step": 9211,
      "time_per_iteration": 2.5721867084503174
    },
    {
      "auxiliary_loss_clip": 0.01082191,
      "auxiliary_loss_mlp": 0.01032535,
      "balance_loss_clip": 1.03763032,
      "balance_loss_mlp": 1.01871872,
      "epoch": 0.5538554035773335,
      "flos": 18037427397120.0,
      "grad_norm": 1.8696746983669041,
      "language_loss": 0.66595,
      "learning_rate": 1.7488775985144437e-06,
      "loss": 0.68709725,
      "num_input_tokens_seen": 198419110,
      "step": 9212,
      "time_per_iteration": 2.5349819660186768
    },
    {
      "auxiliary_loss_clip": 0.01079438,
      "auxiliary_loss_mlp": 0.01028115,
      "balance_loss_clip": 1.03466356,
      "balance_loss_mlp": 1.01408374,
      "epoch": 0.5539155268300014,
      "flos": 31686247002240.0,
      "grad_norm": 1.4324252264799842,
      "language_loss": 0.51715839,
      "learning_rate": 1.7484912238448443e-06,
      "loss": 0.53823388,
      "num_input_tokens_seen": 198441360,
      "step": 9213,
      "time_per_iteration": 2.6196541786193848
    },
    {
      "auxiliary_loss_clip": 0.01084608,
      "auxiliary_loss_mlp": 0.01035347,
      "balance_loss_clip": 1.04235852,
      "balance_loss_mlp": 1.02248406,
      "epoch": 0.5539756500826695,
      "flos": 15193769431680.0,
      "grad_norm": 1.8636886443264649,
      "language_loss": 0.85641068,
      "learning_rate": 1.7481048587124827e-06,
      "loss": 0.87761021,
      "num_input_tokens_seen": 198459835,
      "step": 9214,
      "time_per_iteration": 2.5426511764526367
    },
    {
      "auxiliary_loss_clip": 0.01101033,
      "auxiliary_loss_mlp": 0.01027522,
      "balance_loss_clip": 1.03871202,
      "balance_loss_mlp": 1.01545835,
      "epoch": 0.5540357733353375,
      "flos": 26353117128960.0,
      "grad_norm": 1.5922579801598906,
      "language_loss": 0.69933975,
      "learning_rate": 1.7477185031320108e-06,
      "loss": 0.72062528,
      "num_input_tokens_seen": 198478955,
      "step": 9215,
      "time_per_iteration": 2.502953290939331
    },
    {
      "auxiliary_loss_clip": 0.01088703,
      "auxiliary_loss_mlp": 0.01031955,
      "balance_loss_clip": 1.04404366,
      "balance_loss_mlp": 1.01903844,
      "epoch": 0.5540958965880054,
      "flos": 21323684747520.0,
      "grad_norm": 1.6079157501598225,
      "language_loss": 0.73063791,
      "learning_rate": 1.7473321571180773e-06,
      "loss": 0.75184447,
      "num_input_tokens_seen": 198499030,
      "step": 9216,
      "time_per_iteration": 2.5603458881378174
    },
    {
      "auxiliary_loss_clip": 0.01087715,
      "auxiliary_loss_mlp": 0.01032016,
      "balance_loss_clip": 1.03921628,
      "balance_loss_mlp": 1.01943338,
      "epoch": 0.5541560198406734,
      "flos": 25666828899840.0,
      "grad_norm": 1.8843916401385659,
      "language_loss": 0.7158742,
      "learning_rate": 1.7469458206853345e-06,
      "loss": 0.73707145,
      "num_input_tokens_seen": 198520265,
      "step": 9217,
      "time_per_iteration": 2.5675241947174072
    },
    {
      "auxiliary_loss_clip": 0.01092956,
      "auxiliary_loss_mlp": 0.0102795,
      "balance_loss_clip": 1.03819513,
      "balance_loss_mlp": 1.01590335,
      "epoch": 0.5542161430933413,
      "flos": 21939624190080.0,
      "grad_norm": 1.6876168922932369,
      "language_loss": 0.7828809,
      "learning_rate": 1.7465594938484315e-06,
      "loss": 0.80408996,
      "num_input_tokens_seen": 198539645,
      "step": 9218,
      "time_per_iteration": 2.4953677654266357
    },
    {
      "auxiliary_loss_clip": 0.0107282,
      "auxiliary_loss_mlp": 0.01031362,
      "balance_loss_clip": 1.03889871,
      "balance_loss_mlp": 1.01717627,
      "epoch": 0.5542762663460093,
      "flos": 19571459489280.0,
      "grad_norm": 1.5336018196184211,
      "language_loss": 0.72026777,
      "learning_rate": 1.7461731766220176e-06,
      "loss": 0.74130964,
      "num_input_tokens_seen": 198558710,
      "step": 9219,
      "time_per_iteration": 2.5532217025756836
    },
    {
      "auxiliary_loss_clip": 0.01101989,
      "auxiliary_loss_mlp": 0.01039675,
      "balance_loss_clip": 1.04388523,
      "balance_loss_mlp": 1.02644277,
      "epoch": 0.5543363895986773,
      "flos": 19499063627520.0,
      "grad_norm": 1.5884246347454203,
      "language_loss": 0.71396136,
      "learning_rate": 1.7457868690207426e-06,
      "loss": 0.73537797,
      "num_input_tokens_seen": 198577050,
      "step": 9220,
      "time_per_iteration": 2.496856927871704
    },
    {
      "auxiliary_loss_clip": 0.01110228,
      "auxiliary_loss_mlp": 0.01024064,
      "balance_loss_clip": 1.03937697,
      "balance_loss_mlp": 1.01284659,
      "epoch": 0.5543965128513453,
      "flos": 22635609091200.0,
      "grad_norm": 1.6781320916828435,
      "language_loss": 0.7919178,
      "learning_rate": 1.7454005710592547e-06,
      "loss": 0.81326067,
      "num_input_tokens_seen": 198595290,
      "step": 9221,
      "time_per_iteration": 2.456348419189453
    },
    {
      "auxiliary_loss_clip": 0.0108488,
      "auxiliary_loss_mlp": 0.01030959,
      "balance_loss_clip": 1.03992224,
      "balance_loss_mlp": 1.0182631,
      "epoch": 0.5544566361040132,
      "flos": 25989952671360.0,
      "grad_norm": 1.8844998607941885,
      "language_loss": 0.83779639,
      "learning_rate": 1.7450142827522027e-06,
      "loss": 0.85895479,
      "num_input_tokens_seen": 198614110,
      "step": 9222,
      "time_per_iteration": 2.58412504196167
    },
    {
      "auxiliary_loss_clip": 0.010885,
      "auxiliary_loss_mlp": 0.00789209,
      "balance_loss_clip": 1.04125893,
      "balance_loss_mlp": 1.01626444,
      "epoch": 0.5545167593566812,
      "flos": 28257568225920.0,
      "grad_norm": 2.454960872768613,
      "language_loss": 0.75720191,
      "learning_rate": 1.7446280041142344e-06,
      "loss": 0.77597904,
      "num_input_tokens_seen": 198633880,
      "step": 9223,
      "time_per_iteration": 2.6636600494384766
    },
    {
      "auxiliary_loss_clip": 0.01084014,
      "auxiliary_loss_mlp": 0.01031846,
      "balance_loss_clip": 1.03745711,
      "balance_loss_mlp": 1.01845324,
      "epoch": 0.5545768826093491,
      "flos": 28476551491200.0,
      "grad_norm": 1.729386844148969,
      "language_loss": 0.81540793,
      "learning_rate": 1.7442417351599986e-06,
      "loss": 0.83656657,
      "num_input_tokens_seen": 198653505,
      "step": 9224,
      "time_per_iteration": 3.961113452911377
    },
    {
      "auxiliary_loss_clip": 0.01099412,
      "auxiliary_loss_mlp": 0.010383,
      "balance_loss_clip": 1.041448,
      "balance_loss_mlp": 1.02575362,
      "epoch": 0.5546370058620171,
      "flos": 18478051534080.0,
      "grad_norm": 2.3427705376369072,
      "language_loss": 0.57385319,
      "learning_rate": 1.743855475904141e-06,
      "loss": 0.5952304,
      "num_input_tokens_seen": 198671890,
      "step": 9225,
      "time_per_iteration": 2.4746692180633545
    },
    {
      "auxiliary_loss_clip": 0.01103413,
      "auxiliary_loss_mlp": 0.01035663,
      "balance_loss_clip": 1.03905272,
      "balance_loss_mlp": 1.02282381,
      "epoch": 0.554697129114685,
      "flos": 22930507751040.0,
      "grad_norm": 1.5754252047546868,
      "language_loss": 0.67601967,
      "learning_rate": 1.7434692263613098e-06,
      "loss": 0.6974104,
      "num_input_tokens_seen": 198691995,
      "step": 9226,
      "time_per_iteration": 2.529801368713379
    },
    {
      "auxiliary_loss_clip": 0.0107589,
      "auxiliary_loss_mlp": 0.01034848,
      "balance_loss_clip": 1.03553343,
      "balance_loss_mlp": 1.02216959,
      "epoch": 0.5547572523673531,
      "flos": 21797166850560.0,
      "grad_norm": 1.4452483372481744,
      "language_loss": 0.7447269,
      "learning_rate": 1.7430829865461518e-06,
      "loss": 0.76583433,
      "num_input_tokens_seen": 198712440,
      "step": 9227,
      "time_per_iteration": 2.543351888656616
    },
    {
      "auxiliary_loss_clip": 0.0108192,
      "auxiliary_loss_mlp": 0.0103139,
      "balance_loss_clip": 1.04099202,
      "balance_loss_mlp": 1.01774037,
      "epoch": 0.5548173756200211,
      "flos": 22342829333760.0,
      "grad_norm": 1.5549632243988263,
      "language_loss": 0.73412877,
      "learning_rate": 1.7426967564733118e-06,
      "loss": 0.7552619,
      "num_input_tokens_seen": 198731515,
      "step": 9228,
      "time_per_iteration": 2.5747568607330322
    },
    {
      "auxiliary_loss_clip": 0.01113663,
      "auxiliary_loss_mlp": 0.01031409,
      "balance_loss_clip": 1.04031098,
      "balance_loss_mlp": 1.01914263,
      "epoch": 0.554877498872689,
      "flos": 17858736213120.0,
      "grad_norm": 1.7079664703445376,
      "language_loss": 0.75590831,
      "learning_rate": 1.7423105361574373e-06,
      "loss": 0.77735907,
      "num_input_tokens_seen": 198749750,
      "step": 9229,
      "time_per_iteration": 2.4259326457977295
    },
    {
      "auxiliary_loss_clip": 0.01103036,
      "auxiliary_loss_mlp": 0.00787872,
      "balance_loss_clip": 1.04107666,
      "balance_loss_mlp": 1.01292682,
      "epoch": 0.554937622125357,
      "flos": 17238343484160.0,
      "grad_norm": 1.392236585985139,
      "language_loss": 0.68602949,
      "learning_rate": 1.741924325613172e-06,
      "loss": 0.70493865,
      "num_input_tokens_seen": 198768320,
      "step": 9230,
      "time_per_iteration": 2.484239101409912
    },
    {
      "auxiliary_loss_clip": 0.01066671,
      "auxiliary_loss_mlp": 0.0103211,
      "balance_loss_clip": 1.03852582,
      "balance_loss_mlp": 1.01848459,
      "epoch": 0.5549977453780249,
      "flos": 25368087484800.0,
      "grad_norm": 2.4816291506755324,
      "language_loss": 0.67803216,
      "learning_rate": 1.741538124855163e-06,
      "loss": 0.69901991,
      "num_input_tokens_seen": 198787230,
      "step": 9231,
      "time_per_iteration": 3.9619076251983643
    },
    {
      "auxiliary_loss_clip": 0.01116943,
      "auxiliary_loss_mlp": 0.01030176,
      "balance_loss_clip": 1.0403595,
      "balance_loss_mlp": 1.01626396,
      "epoch": 0.555057868630693,
      "flos": 25079114568960.0,
      "grad_norm": 1.9456566159495485,
      "language_loss": 0.78373992,
      "learning_rate": 1.7411519338980548e-06,
      "loss": 0.80521107,
      "num_input_tokens_seen": 198806720,
      "step": 9232,
      "time_per_iteration": 2.4839911460876465
    },
    {
      "auxiliary_loss_clip": 0.01070521,
      "auxiliary_loss_mlp": 0.01030729,
      "balance_loss_clip": 1.03725934,
      "balance_loss_mlp": 1.01886761,
      "epoch": 0.5551179918833609,
      "flos": 26104220812800.0,
      "grad_norm": 1.6714053460051095,
      "language_loss": 0.82636631,
      "learning_rate": 1.7407657527564898e-06,
      "loss": 0.84737879,
      "num_input_tokens_seen": 198826235,
      "step": 9233,
      "time_per_iteration": 2.595015048980713
    },
    {
      "auxiliary_loss_clip": 0.0109895,
      "auxiliary_loss_mlp": 0.01034063,
      "balance_loss_clip": 1.0364517,
      "balance_loss_mlp": 1.02114701,
      "epoch": 0.5551781151360289,
      "flos": 19384759572480.0,
      "grad_norm": 2.570636962837449,
      "language_loss": 0.74818951,
      "learning_rate": 1.7403795814451142e-06,
      "loss": 0.76951969,
      "num_input_tokens_seen": 198842655,
      "step": 9234,
      "time_per_iteration": 3.8458969593048096
    },
    {
      "auxiliary_loss_clip": 0.01085751,
      "auxiliary_loss_mlp": 0.01025802,
      "balance_loss_clip": 1.03575516,
      "balance_loss_mlp": 1.01410782,
      "epoch": 0.5552382383886968,
      "flos": 21725956137600.0,
      "grad_norm": 2.4216947749386666,
      "language_loss": 0.64437759,
      "learning_rate": 1.7399934199785706e-06,
      "loss": 0.66549313,
      "num_input_tokens_seen": 198861210,
      "step": 9235,
      "time_per_iteration": 2.515397071838379
    },
    {
      "auxiliary_loss_clip": 0.0106001,
      "auxiliary_loss_mlp": 0.01029439,
      "balance_loss_clip": 1.03462648,
      "balance_loss_mlp": 1.01630831,
      "epoch": 0.5552983616413648,
      "flos": 14356189117440.0,
      "grad_norm": 1.8377859500445055,
      "language_loss": 0.67924416,
      "learning_rate": 1.7396072683715029e-06,
      "loss": 0.70013869,
      "num_input_tokens_seen": 198880045,
      "step": 9236,
      "time_per_iteration": 2.559159517288208
    },
    {
      "auxiliary_loss_clip": 0.01106953,
      "auxiliary_loss_mlp": 0.01024875,
      "balance_loss_clip": 1.03716922,
      "balance_loss_mlp": 1.01251328,
      "epoch": 0.5553584848940327,
      "flos": 25478548784640.0,
      "grad_norm": 2.607257502786614,
      "language_loss": 0.86658078,
      "learning_rate": 1.7392211266385536e-06,
      "loss": 0.8878991,
      "num_input_tokens_seen": 198900210,
      "step": 9237,
      "time_per_iteration": 2.4844696521759033
    },
    {
      "auxiliary_loss_clip": 0.01098543,
      "auxiliary_loss_mlp": 0.01030161,
      "balance_loss_clip": 1.03731704,
      "balance_loss_mlp": 1.0177933,
      "epoch": 0.5554186081467007,
      "flos": 22163850840960.0,
      "grad_norm": 2.1201127491948086,
      "language_loss": 0.73213482,
      "learning_rate": 1.7388349947943652e-06,
      "loss": 0.7534219,
      "num_input_tokens_seen": 198919055,
      "step": 9238,
      "time_per_iteration": 3.860952854156494
    },
    {
      "auxiliary_loss_clip": 0.01098973,
      "auxiliary_loss_mlp": 0.01031919,
      "balance_loss_clip": 1.03707111,
      "balance_loss_mlp": 1.01879978,
      "epoch": 0.5554787313993687,
      "flos": 49746656125440.0,
      "grad_norm": 1.5986924904293494,
      "language_loss": 0.78359497,
      "learning_rate": 1.73844887285358e-06,
      "loss": 0.80490386,
      "num_input_tokens_seen": 198943505,
      "step": 9239,
      "time_per_iteration": 2.7362375259399414
    },
    {
      "auxiliary_loss_clip": 0.01091596,
      "auxiliary_loss_mlp": 0.01026957,
      "balance_loss_clip": 1.03996217,
      "balance_loss_mlp": 1.01471376,
      "epoch": 0.5555388546520367,
      "flos": 22127365601280.0,
      "grad_norm": 1.5797567868410776,
      "language_loss": 0.79935038,
      "learning_rate": 1.7380627608308393e-06,
      "loss": 0.82053584,
      "num_input_tokens_seen": 198963590,
      "step": 9240,
      "time_per_iteration": 2.5304806232452393
    },
    {
      "auxiliary_loss_clip": 0.010884,
      "auxiliary_loss_mlp": 0.01030705,
      "balance_loss_clip": 1.03815746,
      "balance_loss_mlp": 1.01851535,
      "epoch": 0.5555989779047047,
      "flos": 24682122478080.0,
      "grad_norm": 1.6418642776552745,
      "language_loss": 0.65455174,
      "learning_rate": 1.737676658740786e-06,
      "loss": 0.67574281,
      "num_input_tokens_seen": 198982680,
      "step": 9241,
      "time_per_iteration": 2.541163206100464
    },
    {
      "auxiliary_loss_clip": 0.01101333,
      "auxiliary_loss_mlp": 0.00785443,
      "balance_loss_clip": 1.03913236,
      "balance_loss_mlp": 1.00852275,
      "epoch": 0.5556591011573726,
      "flos": 16106510954880.0,
      "grad_norm": 1.9522455324501808,
      "language_loss": 0.72838306,
      "learning_rate": 1.7372905665980594e-06,
      "loss": 0.74725085,
      "num_input_tokens_seen": 199000185,
      "step": 9242,
      "time_per_iteration": 2.5102603435516357
    },
    {
      "auxiliary_loss_clip": 0.01094133,
      "auxiliary_loss_mlp": 0.0102875,
      "balance_loss_clip": 1.03840399,
      "balance_loss_mlp": 1.01502872,
      "epoch": 0.5557192244100406,
      "flos": 12933695733120.0,
      "grad_norm": 1.7030309253366334,
      "language_loss": 0.64064467,
      "learning_rate": 1.7369044844173012e-06,
      "loss": 0.6618734,
      "num_input_tokens_seen": 199018380,
      "step": 9243,
      "time_per_iteration": 2.515479803085327
    },
    {
      "auxiliary_loss_clip": 0.01085555,
      "auxiliary_loss_mlp": 0.00784378,
      "balance_loss_clip": 1.04136014,
      "balance_loss_mlp": 1.00763202,
      "epoch": 0.5557793476627085,
      "flos": 23111712887040.0,
      "grad_norm": 1.7897037553694843,
      "language_loss": 0.75145388,
      "learning_rate": 1.7365184122131509e-06,
      "loss": 0.77015316,
      "num_input_tokens_seen": 199037115,
      "step": 9244,
      "time_per_iteration": 2.554372549057007
    },
    {
      "auxiliary_loss_clip": 0.01083768,
      "auxiliary_loss_mlp": 0.01028383,
      "balance_loss_clip": 1.03616667,
      "balance_loss_mlp": 1.01679039,
      "epoch": 0.5558394709153766,
      "flos": 21428040735360.0,
      "grad_norm": 2.1110454118428024,
      "language_loss": 0.75248206,
      "learning_rate": 1.7361323500002486e-06,
      "loss": 0.77360356,
      "num_input_tokens_seen": 199053375,
      "step": 9245,
      "time_per_iteration": 2.544123649597168
    },
    {
      "auxiliary_loss_clip": 0.01097791,
      "auxiliary_loss_mlp": 0.01031413,
      "balance_loss_clip": 1.04095864,
      "balance_loss_mlp": 1.01776338,
      "epoch": 0.5558995941680445,
      "flos": 25078324469760.0,
      "grad_norm": 2.2690338716604055,
      "language_loss": 0.79468119,
      "learning_rate": 1.7357462977932348e-06,
      "loss": 0.81597322,
      "num_input_tokens_seen": 199070930,
      "step": 9246,
      "time_per_iteration": 2.5477354526519775
    },
    {
      "auxiliary_loss_clip": 0.01111665,
      "auxiliary_loss_mlp": 0.0103233,
      "balance_loss_clip": 1.03897953,
      "balance_loss_mlp": 1.01997995,
      "epoch": 0.5559597174207125,
      "flos": 20011149872640.0,
      "grad_norm": 1.7955239648050072,
      "language_loss": 0.73729455,
      "learning_rate": 1.7353602556067471e-06,
      "loss": 0.75873446,
      "num_input_tokens_seen": 199088675,
      "step": 9247,
      "time_per_iteration": 2.4549548625946045
    },
    {
      "auxiliary_loss_clip": 0.0108993,
      "auxiliary_loss_mlp": 0.01032757,
      "balance_loss_clip": 1.04747367,
      "balance_loss_mlp": 1.01933992,
      "epoch": 0.5560198406733804,
      "flos": 16835677044480.0,
      "grad_norm": 3.3036057233172698,
      "language_loss": 0.75539708,
      "learning_rate": 1.7349742234554254e-06,
      "loss": 0.77662396,
      "num_input_tokens_seen": 199103075,
      "step": 9248,
      "time_per_iteration": 2.4962079524993896
    },
    {
      "auxiliary_loss_clip": 0.01007114,
      "auxiliary_loss_mlp": 0.01009328,
      "balance_loss_clip": 1.02209187,
      "balance_loss_mlp": 1.00769448,
      "epoch": 0.5560799639260484,
      "flos": 70697051758080.0,
      "grad_norm": 0.8645574125346883,
      "language_loss": 0.59445953,
      "learning_rate": 1.7345882013539081e-06,
      "loss": 0.61462402,
      "num_input_tokens_seen": 199160325,
      "step": 9249,
      "time_per_iteration": 3.2806990146636963
    },
    {
      "auxiliary_loss_clip": 0.01108776,
      "auxiliary_loss_mlp": 0.01027543,
      "balance_loss_clip": 1.0365293,
      "balance_loss_mlp": 1.01473963,
      "epoch": 0.5561400871787163,
      "flos": 23148593176320.0,
      "grad_norm": 2.1369669708512524,
      "language_loss": 0.79854095,
      "learning_rate": 1.734202189316832e-06,
      "loss": 0.81990415,
      "num_input_tokens_seen": 199179760,
      "step": 9250,
      "time_per_iteration": 2.4590866565704346
    },
    {
      "auxiliary_loss_clip": 0.01091024,
      "auxiliary_loss_mlp": 0.01030969,
      "balance_loss_clip": 1.03848529,
      "balance_loss_mlp": 1.01745033,
      "epoch": 0.5562002104313843,
      "flos": 17566423332480.0,
      "grad_norm": 3.3083841367021454,
      "language_loss": 0.68958294,
      "learning_rate": 1.733816187358836e-06,
      "loss": 0.71080279,
      "num_input_tokens_seen": 199196695,
      "step": 9251,
      "time_per_iteration": 2.488954782485962
    },
    {
      "auxiliary_loss_clip": 0.01100085,
      "auxiliary_loss_mlp": 0.01032027,
      "balance_loss_clip": 1.03843224,
      "balance_loss_mlp": 1.01951051,
      "epoch": 0.5562603336840523,
      "flos": 25045430590080.0,
      "grad_norm": 1.5191850042723756,
      "language_loss": 0.75476271,
      "learning_rate": 1.7334301954945569e-06,
      "loss": 0.77608383,
      "num_input_tokens_seen": 199217845,
      "step": 9252,
      "time_per_iteration": 2.5424764156341553
    },
    {
      "auxiliary_loss_clip": 0.01098183,
      "auxiliary_loss_mlp": 0.01038468,
      "balance_loss_clip": 1.03679466,
      "balance_loss_mlp": 1.02427006,
      "epoch": 0.5563204569367203,
      "flos": 29059022436480.0,
      "grad_norm": 1.4535720262549348,
      "language_loss": 0.7262845,
      "learning_rate": 1.7330442137386313e-06,
      "loss": 0.74765104,
      "num_input_tokens_seen": 199239250,
      "step": 9253,
      "time_per_iteration": 2.56081485748291
    },
    {
      "auxiliary_loss_clip": 0.01083012,
      "auxiliary_loss_mlp": 0.01030001,
      "balance_loss_clip": 1.03931928,
      "balance_loss_mlp": 1.01789546,
      "epoch": 0.5563805801893883,
      "flos": 22090449398400.0,
      "grad_norm": 6.315474985076652,
      "language_loss": 0.83348107,
      "learning_rate": 1.7326582421056965e-06,
      "loss": 0.85461122,
      "num_input_tokens_seen": 199258320,
      "step": 9254,
      "time_per_iteration": 2.544121026992798
    },
    {
      "auxiliary_loss_clip": 0.01027132,
      "auxiliary_loss_mlp": 0.01003389,
      "balance_loss_clip": 1.02202535,
      "balance_loss_mlp": 1.00204742,
      "epoch": 0.5564407034420562,
      "flos": 58636128689280.0,
      "grad_norm": 0.863284816924183,
      "language_loss": 0.64806914,
      "learning_rate": 1.732272280610387e-06,
      "loss": 0.6683743,
      "num_input_tokens_seen": 199314840,
      "step": 9255,
      "time_per_iteration": 2.9900217056274414
    },
    {
      "auxiliary_loss_clip": 0.01103088,
      "auxiliary_loss_mlp": 0.01033401,
      "balance_loss_clip": 1.04153883,
      "balance_loss_mlp": 1.02152228,
      "epoch": 0.5565008266947242,
      "flos": 23112323418240.0,
      "grad_norm": 1.8659135847542294,
      "language_loss": 0.69254965,
      "learning_rate": 1.7318863292673399e-06,
      "loss": 0.71391451,
      "num_input_tokens_seen": 199335405,
      "step": 9256,
      "time_per_iteration": 2.502626895904541
    },
    {
      "auxiliary_loss_clip": 0.01074155,
      "auxiliary_loss_mlp": 0.01028687,
      "balance_loss_clip": 1.03710318,
      "balance_loss_mlp": 1.01768374,
      "epoch": 0.5565609499473921,
      "flos": 21578399066880.0,
      "grad_norm": 1.6561048640173295,
      "language_loss": 0.75781006,
      "learning_rate": 1.73150038809119e-06,
      "loss": 0.7788384,
      "num_input_tokens_seen": 199354345,
      "step": 9257,
      "time_per_iteration": 2.5551674365997314
    },
    {
      "auxiliary_loss_clip": 0.01063328,
      "auxiliary_loss_mlp": 0.01033384,
      "balance_loss_clip": 1.03519392,
      "balance_loss_mlp": 1.02099192,
      "epoch": 0.5566210732000602,
      "flos": 18369637309440.0,
      "grad_norm": 1.9506178880043559,
      "language_loss": 0.61418414,
      "learning_rate": 1.7311144570965724e-06,
      "loss": 0.63515127,
      "num_input_tokens_seen": 199372250,
      "step": 9258,
      "time_per_iteration": 2.5554490089416504
    },
    {
      "auxiliary_loss_clip": 0.01076318,
      "auxiliary_loss_mlp": 0.0103137,
      "balance_loss_clip": 1.03725195,
      "balance_loss_mlp": 1.0173986,
      "epoch": 0.5566811964527281,
      "flos": 25703350053120.0,
      "grad_norm": 1.715468866147634,
      "language_loss": 0.79034162,
      "learning_rate": 1.7307285362981215e-06,
      "loss": 0.81141853,
      "num_input_tokens_seen": 199392815,
      "step": 9259,
      "time_per_iteration": 2.5837697982788086
    },
    {
      "auxiliary_loss_clip": 0.01081378,
      "auxiliary_loss_mlp": 0.01031596,
      "balance_loss_clip": 1.03923571,
      "balance_loss_mlp": 1.01847744,
      "epoch": 0.5567413197053961,
      "flos": 26943991856640.0,
      "grad_norm": 2.276803107768523,
      "language_loss": 0.8167277,
      "learning_rate": 1.7303426257104712e-06,
      "loss": 0.83785748,
      "num_input_tokens_seen": 199412375,
      "step": 9260,
      "time_per_iteration": 2.5871708393096924
    },
    {
      "auxiliary_loss_clip": 0.01109911,
      "auxiliary_loss_mlp": 0.01037219,
      "balance_loss_clip": 1.03747904,
      "balance_loss_mlp": 1.02433896,
      "epoch": 0.556801442958064,
      "flos": 20850597694080.0,
      "grad_norm": 1.5841920486549808,
      "language_loss": 0.68875349,
      "learning_rate": 1.729956725348256e-06,
      "loss": 0.71022481,
      "num_input_tokens_seen": 199431490,
      "step": 9261,
      "time_per_iteration": 2.4769859313964844
    },
    {
      "auxiliary_loss_clip": 0.01012354,
      "auxiliary_loss_mlp": 0.0100232,
      "balance_loss_clip": 1.01759315,
      "balance_loss_mlp": 1.0009371,
      "epoch": 0.556861566210732,
      "flos": 70498213044480.0,
      "grad_norm": 0.7310752990794981,
      "language_loss": 0.6110152,
      "learning_rate": 1.729570835226108e-06,
      "loss": 0.63116193,
      "num_input_tokens_seen": 199495855,
      "step": 9262,
      "time_per_iteration": 3.1564443111419678
    },
    {
      "auxiliary_loss_clip": 0.0110256,
      "auxiliary_loss_mlp": 0.01033327,
      "balance_loss_clip": 1.03790021,
      "balance_loss_mlp": 1.02101254,
      "epoch": 0.5569216894633999,
      "flos": 25337276593920.0,
      "grad_norm": 1.536958729900559,
      "language_loss": 0.6440993,
      "learning_rate": 1.7291849553586622e-06,
      "loss": 0.6654582,
      "num_input_tokens_seen": 199515870,
      "step": 9263,
      "time_per_iteration": 3.9006993770599365
    },
    {
      "auxiliary_loss_clip": 0.01088302,
      "auxiliary_loss_mlp": 0.01031746,
      "balance_loss_clip": 1.03710258,
      "balance_loss_mlp": 1.01881146,
      "epoch": 0.556981812716068,
      "flos": 22638733574400.0,
      "grad_norm": 1.8950747552676095,
      "language_loss": 0.73224789,
      "learning_rate": 1.7287990857605497e-06,
      "loss": 0.75344837,
      "num_input_tokens_seen": 199535745,
      "step": 9264,
      "time_per_iteration": 2.5430479049682617
    },
    {
      "auxiliary_loss_clip": 0.01083097,
      "auxiliary_loss_mlp": 0.01028906,
      "balance_loss_clip": 1.03940511,
      "balance_loss_mlp": 1.01657999,
      "epoch": 0.5570419359687359,
      "flos": 11035852738560.0,
      "grad_norm": 1.8429761803584588,
      "language_loss": 0.76437515,
      "learning_rate": 1.7284132264464022e-06,
      "loss": 0.78549516,
      "num_input_tokens_seen": 199554035,
      "step": 9265,
      "time_per_iteration": 2.5609729290008545
    },
    {
      "auxiliary_loss_clip": 0.01083715,
      "auxiliary_loss_mlp": 0.01029671,
      "balance_loss_clip": 1.04055107,
      "balance_loss_mlp": 1.01880479,
      "epoch": 0.5571020592214039,
      "flos": 22823135020800.0,
      "grad_norm": 1.296277318398208,
      "language_loss": 0.70678025,
      "learning_rate": 1.7280273774308536e-06,
      "loss": 0.72791415,
      "num_input_tokens_seen": 199576120,
      "step": 9266,
      "time_per_iteration": 2.5526628494262695
    },
    {
      "auxiliary_loss_clip": 0.01085415,
      "auxiliary_loss_mlp": 0.01032251,
      "balance_loss_clip": 1.03604507,
      "balance_loss_mlp": 1.02006817,
      "epoch": 0.5571621824740719,
      "flos": 22927778317440.0,
      "grad_norm": 1.9024782108254799,
      "language_loss": 0.67872924,
      "learning_rate": 1.727641538728533e-06,
      "loss": 0.69990587,
      "num_input_tokens_seen": 199593780,
      "step": 9267,
      "time_per_iteration": 2.5226635932922363
    },
    {
      "auxiliary_loss_clip": 0.01097601,
      "auxiliary_loss_mlp": 0.01037509,
      "balance_loss_clip": 1.03778744,
      "balance_loss_mlp": 1.02626181,
      "epoch": 0.5572223057267398,
      "flos": 22966705681920.0,
      "grad_norm": 2.132692049370671,
      "language_loss": 0.74250698,
      "learning_rate": 1.7272557103540736e-06,
      "loss": 0.76385808,
      "num_input_tokens_seen": 199613220,
      "step": 9268,
      "time_per_iteration": 2.502830743789673
    },
    {
      "auxiliary_loss_clip": 0.01093548,
      "auxiliary_loss_mlp": 0.00784376,
      "balance_loss_clip": 1.03892636,
      "balance_loss_mlp": 1.00959504,
      "epoch": 0.5572824289794078,
      "flos": 20960053413120.0,
      "grad_norm": 1.8846523052749902,
      "language_loss": 0.75372112,
      "learning_rate": 1.726869892322104e-06,
      "loss": 0.7725004,
      "num_input_tokens_seen": 199632085,
      "step": 9269,
      "time_per_iteration": 2.5183188915252686
    },
    {
      "auxiliary_loss_clip": 0.01074248,
      "auxiliary_loss_mlp": 0.01034971,
      "balance_loss_clip": 1.035537,
      "balance_loss_mlp": 1.02220976,
      "epoch": 0.5573425522320757,
      "flos": 25042413847680.0,
      "grad_norm": 10.65934501529905,
      "language_loss": 0.8284595,
      "learning_rate": 1.726484084647256e-06,
      "loss": 0.84955174,
      "num_input_tokens_seen": 199649295,
      "step": 9270,
      "time_per_iteration": 3.9425487518310547
    },
    {
      "auxiliary_loss_clip": 0.01075268,
      "auxiliary_loss_mlp": 0.01033916,
      "balance_loss_clip": 1.03774118,
      "balance_loss_mlp": 1.02107692,
      "epoch": 0.5574026754847438,
      "flos": 23659637927040.0,
      "grad_norm": 2.1630774781823874,
      "language_loss": 0.79608065,
      "learning_rate": 1.7260982873441591e-06,
      "loss": 0.81717253,
      "num_input_tokens_seen": 199668870,
      "step": 9271,
      "time_per_iteration": 2.600764513015747
    },
    {
      "auxiliary_loss_clip": 0.01089546,
      "auxiliary_loss_mlp": 0.01028246,
      "balance_loss_clip": 1.03716731,
      "balance_loss_mlp": 1.01576471,
      "epoch": 0.5574627987374117,
      "flos": 24782240661120.0,
      "grad_norm": 1.7868484966728213,
      "language_loss": 0.90281993,
      "learning_rate": 1.725712500427442e-06,
      "loss": 0.92399788,
      "num_input_tokens_seen": 199684870,
      "step": 9272,
      "time_per_iteration": 2.524334669113159
    },
    {
      "auxiliary_loss_clip": 0.01076301,
      "auxiliary_loss_mlp": 0.01029245,
      "balance_loss_clip": 1.03709054,
      "balance_loss_mlp": 1.01690078,
      "epoch": 0.5575229219900797,
      "flos": 21834944979840.0,
      "grad_norm": 2.2198055549630262,
      "language_loss": 0.8413626,
      "learning_rate": 1.7253267239117347e-06,
      "loss": 0.862418,
      "num_input_tokens_seen": 199701975,
      "step": 9273,
      "time_per_iteration": 3.9291181564331055
    },
    {
      "auxiliary_loss_clip": 0.01102316,
      "auxiliary_loss_mlp": 0.010364,
      "balance_loss_clip": 1.03910935,
      "balance_loss_mlp": 1.02295899,
      "epoch": 0.5575830452427476,
      "flos": 27815148408960.0,
      "grad_norm": 2.257955534193018,
      "language_loss": 0.74082726,
      "learning_rate": 1.7249409578116655e-06,
      "loss": 0.76221436,
      "num_input_tokens_seen": 199721865,
      "step": 9274,
      "time_per_iteration": 2.534677505493164
    },
    {
      "auxiliary_loss_clip": 0.01091298,
      "auxiliary_loss_mlp": 0.01034066,
      "balance_loss_clip": 1.04075003,
      "balance_loss_mlp": 1.01977277,
      "epoch": 0.5576431684954156,
      "flos": 17812805696640.0,
      "grad_norm": 3.0099792473739604,
      "language_loss": 0.77474451,
      "learning_rate": 1.7245552021418629e-06,
      "loss": 0.79599822,
      "num_input_tokens_seen": 199736455,
      "step": 9275,
      "time_per_iteration": 2.514914035797119
    },
    {
      "auxiliary_loss_clip": 0.01086796,
      "auxiliary_loss_mlp": 0.01029485,
      "balance_loss_clip": 1.04049301,
      "balance_loss_mlp": 1.01711702,
      "epoch": 0.5577032917480835,
      "flos": 15486872411520.0,
      "grad_norm": 1.722481453024456,
      "language_loss": 0.74795926,
      "learning_rate": 1.7241694569169546e-06,
      "loss": 0.76912212,
      "num_input_tokens_seen": 199753125,
      "step": 9276,
      "time_per_iteration": 2.4797325134277344
    },
    {
      "auxiliary_loss_clip": 0.01088014,
      "auxiliary_loss_mlp": 0.01034936,
      "balance_loss_clip": 1.03590655,
      "balance_loss_mlp": 1.02236557,
      "epoch": 0.5577634150007516,
      "flos": 21579763783680.0,
      "grad_norm": 1.7235889009155767,
      "language_loss": 0.75455391,
      "learning_rate": 1.7237837221515678e-06,
      "loss": 0.77578342,
      "num_input_tokens_seen": 199771365,
      "step": 9277,
      "time_per_iteration": 3.9213991165161133
    },
    {
      "auxiliary_loss_clip": 0.01106342,
      "auxiliary_loss_mlp": 0.01031913,
      "balance_loss_clip": 1.03589463,
      "balance_loss_mlp": 1.02005219,
      "epoch": 0.5578235382534195,
      "flos": 21139750177920.0,
      "grad_norm": 1.6654597259938995,
      "language_loss": 0.7135663,
      "learning_rate": 1.7233979978603304e-06,
      "loss": 0.73494887,
      "num_input_tokens_seen": 199790035,
      "step": 9278,
      "time_per_iteration": 2.4813168048858643
    },
    {
      "auxiliary_loss_clip": 0.01076452,
      "auxiliary_loss_mlp": 0.01033407,
      "balance_loss_clip": 1.04011393,
      "balance_loss_mlp": 1.02012682,
      "epoch": 0.5578836615060875,
      "flos": 26505199313280.0,
      "grad_norm": 1.942565106206994,
      "language_loss": 0.75457817,
      "learning_rate": 1.723012284057868e-06,
      "loss": 0.77567679,
      "num_input_tokens_seen": 199811125,
      "step": 9279,
      "time_per_iteration": 2.621128559112549
    },
    {
      "auxiliary_loss_clip": 0.01084726,
      "auxiliary_loss_mlp": 0.01027221,
      "balance_loss_clip": 1.03383732,
      "balance_loss_mlp": 1.01465642,
      "epoch": 0.5579437847587555,
      "flos": 20153786780160.0,
      "grad_norm": 1.5344578561257778,
      "language_loss": 0.67429984,
      "learning_rate": 1.7226265807588082e-06,
      "loss": 0.69541931,
      "num_input_tokens_seen": 199829915,
      "step": 9280,
      "time_per_iteration": 2.5274171829223633
    },
    {
      "auxiliary_loss_clip": 0.01097536,
      "auxiliary_loss_mlp": 0.01034868,
      "balance_loss_clip": 1.03527164,
      "balance_loss_mlp": 1.02207041,
      "epoch": 0.5580039080114234,
      "flos": 26102281478400.0,
      "grad_norm": 1.5339149355155148,
      "language_loss": 0.73460191,
      "learning_rate": 1.7222408879777763e-06,
      "loss": 0.75592601,
      "num_input_tokens_seen": 199850670,
      "step": 9281,
      "time_per_iteration": 2.5548722743988037
    },
    {
      "auxiliary_loss_clip": 0.01077825,
      "auxiliary_loss_mlp": 0.00786272,
      "balance_loss_clip": 1.03608155,
      "balance_loss_mlp": 1.00974751,
      "epoch": 0.5580640312640914,
      "flos": 13771671096960.0,
      "grad_norm": 4.047925833270152,
      "language_loss": 0.74718153,
      "learning_rate": 1.7218552057293974e-06,
      "loss": 0.76582253,
      "num_input_tokens_seen": 199867645,
      "step": 9282,
      "time_per_iteration": 2.5195586681365967
    },
    {
      "auxiliary_loss_clip": 0.0104781,
      "auxiliary_loss_mlp": 0.01028052,
      "balance_loss_clip": 1.03452599,
      "balance_loss_mlp": 1.01514149,
      "epoch": 0.5581241545167593,
      "flos": 17675986792320.0,
      "grad_norm": 1.8039225396648588,
      "language_loss": 0.66199958,
      "learning_rate": 1.721469534028297e-06,
      "loss": 0.68275821,
      "num_input_tokens_seen": 199886320,
      "step": 9283,
      "time_per_iteration": 2.611476421356201
    },
    {
      "auxiliary_loss_clip": 0.01075334,
      "auxiliary_loss_mlp": 0.01028203,
      "balance_loss_clip": 1.03736901,
      "balance_loss_mlp": 1.01681268,
      "epoch": 0.5581842777694274,
      "flos": 19569161018880.0,
      "grad_norm": 1.7907966232406831,
      "language_loss": 0.82992196,
      "learning_rate": 1.7210838728890994e-06,
      "loss": 0.85095739,
      "num_input_tokens_seen": 199904895,
      "step": 9284,
      "time_per_iteration": 2.538071632385254
    },
    {
      "auxiliary_loss_clip": 0.01088708,
      "auxiliary_loss_mlp": 0.01030789,
      "balance_loss_clip": 1.03762078,
      "balance_loss_mlp": 1.01840281,
      "epoch": 0.5582444010220953,
      "flos": 20595165102720.0,
      "grad_norm": 2.831500349052182,
      "language_loss": 0.85048562,
      "learning_rate": 1.7206982223264304e-06,
      "loss": 0.87168062,
      "num_input_tokens_seen": 199921090,
      "step": 9285,
      "time_per_iteration": 2.531393051147461
    },
    {
      "auxiliary_loss_clip": 0.01086254,
      "auxiliary_loss_mlp": 0.01031245,
      "balance_loss_clip": 1.0395143,
      "balance_loss_mlp": 1.01937127,
      "epoch": 0.5583045242747633,
      "flos": 19135504120320.0,
      "grad_norm": 3.0833788721320388,
      "language_loss": 0.73827493,
      "learning_rate": 1.720312582354912e-06,
      "loss": 0.7594499,
      "num_input_tokens_seen": 199939925,
      "step": 9286,
      "time_per_iteration": 2.528172254562378
    },
    {
      "auxiliary_loss_clip": 0.0111022,
      "auxiliary_loss_mlp": 0.01030184,
      "balance_loss_clip": 1.03732789,
      "balance_loss_mlp": 1.01804852,
      "epoch": 0.5583646475274312,
      "flos": 27454569730560.0,
      "grad_norm": 1.6421252882918709,
      "language_loss": 0.74206048,
      "learning_rate": 1.7199269529891684e-06,
      "loss": 0.76346451,
      "num_input_tokens_seen": 199960015,
      "step": 9287,
      "time_per_iteration": 2.5323171615600586
    },
    {
      "auxiliary_loss_clip": 0.01076446,
      "auxiliary_loss_mlp": 0.01032983,
      "balance_loss_clip": 1.03623533,
      "balance_loss_mlp": 1.01917291,
      "epoch": 0.5584247707800992,
      "flos": 23653784010240.0,
      "grad_norm": 1.5950398165563044,
      "language_loss": 0.75009918,
      "learning_rate": 1.7195413342438233e-06,
      "loss": 0.77119344,
      "num_input_tokens_seen": 199980505,
      "step": 9288,
      "time_per_iteration": 2.5779786109924316
    },
    {
      "auxiliary_loss_clip": 0.01090862,
      "auxiliary_loss_mlp": 0.01036754,
      "balance_loss_clip": 1.03878117,
      "balance_loss_mlp": 1.02319956,
      "epoch": 0.5584848940327671,
      "flos": 13698880185600.0,
      "grad_norm": 2.6200809035972528,
      "language_loss": 0.77538633,
      "learning_rate": 1.7191557261334984e-06,
      "loss": 0.79666245,
      "num_input_tokens_seen": 199999020,
      "step": 9289,
      "time_per_iteration": 2.5084128379821777
    },
    {
      "auxiliary_loss_clip": 0.01084042,
      "auxiliary_loss_mlp": 0.01032878,
      "balance_loss_clip": 1.03829741,
      "balance_loss_mlp": 1.01974702,
      "epoch": 0.5585450172854352,
      "flos": 27016208150400.0,
      "grad_norm": 1.8442385577077245,
      "language_loss": 0.61099917,
      "learning_rate": 1.718770128672817e-06,
      "loss": 0.63216835,
      "num_input_tokens_seen": 200019020,
      "step": 9290,
      "time_per_iteration": 2.584376573562622
    },
    {
      "auxiliary_loss_clip": 0.01064133,
      "auxiliary_loss_mlp": 0.01031855,
      "balance_loss_clip": 1.03695512,
      "balance_loss_mlp": 1.01891482,
      "epoch": 0.5586051405381031,
      "flos": 23185653033600.0,
      "grad_norm": 2.177509334865663,
      "language_loss": 0.67519283,
      "learning_rate": 1.7183845418764e-06,
      "loss": 0.69615275,
      "num_input_tokens_seen": 200038110,
      "step": 9291,
      "time_per_iteration": 2.613873243331909
    },
    {
      "auxiliary_loss_clip": 0.01081186,
      "auxiliary_loss_mlp": 0.01036134,
      "balance_loss_clip": 1.0370661,
      "balance_loss_mlp": 1.02299106,
      "epoch": 0.5586652637907711,
      "flos": 20775544225920.0,
      "grad_norm": 2.057270516585976,
      "language_loss": 0.84080648,
      "learning_rate": 1.7179989657588698e-06,
      "loss": 0.86197972,
      "num_input_tokens_seen": 200056210,
      "step": 9292,
      "time_per_iteration": 2.5488624572753906
    },
    {
      "auxiliary_loss_clip": 0.01082483,
      "auxiliary_loss_mlp": 0.01040436,
      "balance_loss_clip": 1.03684318,
      "balance_loss_mlp": 1.02787685,
      "epoch": 0.5587253870434391,
      "flos": 28219897837440.0,
      "grad_norm": 1.9812537329909878,
      "language_loss": 0.73783636,
      "learning_rate": 1.7176134003348476e-06,
      "loss": 0.75906563,
      "num_input_tokens_seen": 200075620,
      "step": 9293,
      "time_per_iteration": 2.565099000930786
    },
    {
      "auxiliary_loss_clip": 0.01085241,
      "auxiliary_loss_mlp": 0.01031714,
      "balance_loss_clip": 1.03964758,
      "balance_loss_mlp": 1.01985919,
      "epoch": 0.558785510296107,
      "flos": 26615732440320.0,
      "grad_norm": 1.7155395037241226,
      "language_loss": 0.72868657,
      "learning_rate": 1.7172278456189523e-06,
      "loss": 0.74985611,
      "num_input_tokens_seen": 200095945,
      "step": 9294,
      "time_per_iteration": 2.5708858966827393
    },
    {
      "auxiliary_loss_clip": 0.01088022,
      "auxiliary_loss_mlp": 0.00784022,
      "balance_loss_clip": 1.03776479,
      "balance_loss_mlp": 1.00640321,
      "epoch": 0.558845633548775,
      "flos": 20156767608960.0,
      "grad_norm": 2.115245257803104,
      "language_loss": 0.68389428,
      "learning_rate": 1.716842301625806e-06,
      "loss": 0.70261478,
      "num_input_tokens_seen": 200114185,
      "step": 9295,
      "time_per_iteration": 2.5001771450042725
    },
    {
      "auxiliary_loss_clip": 0.01111829,
      "auxiliary_loss_mlp": 0.01035651,
      "balance_loss_clip": 1.03927064,
      "balance_loss_mlp": 1.02317619,
      "epoch": 0.5589057568014429,
      "flos": 24350774492160.0,
      "grad_norm": 1.4579667802869498,
      "language_loss": 0.80545008,
      "learning_rate": 1.7164567683700281e-06,
      "loss": 0.82692486,
      "num_input_tokens_seen": 200135030,
      "step": 9296,
      "time_per_iteration": 2.493241548538208
    },
    {
      "auxiliary_loss_clip": 0.01100381,
      "auxiliary_loss_mlp": 0.01031323,
      "balance_loss_clip": 1.0391562,
      "balance_loss_mlp": 1.01910448,
      "epoch": 0.558965880054111,
      "flos": 21105168359040.0,
      "grad_norm": 1.5628938698633716,
      "language_loss": 0.65356159,
      "learning_rate": 1.7160712458662379e-06,
      "loss": 0.6748786,
      "num_input_tokens_seen": 200154290,
      "step": 9297,
      "time_per_iteration": 2.485689401626587
    },
    {
      "auxiliary_loss_clip": 0.01079895,
      "auxiliary_loss_mlp": 0.01038539,
      "balance_loss_clip": 1.03919578,
      "balance_loss_mlp": 1.02547407,
      "epoch": 0.5590260033067789,
      "flos": 18436071513600.0,
      "grad_norm": 1.7148020006594242,
      "language_loss": 0.75172293,
      "learning_rate": 1.7156857341290544e-06,
      "loss": 0.77290732,
      "num_input_tokens_seen": 200171555,
      "step": 9298,
      "time_per_iteration": 2.5237314701080322
    },
    {
      "auxiliary_loss_clip": 0.01029385,
      "auxiliary_loss_mlp": 0.01020868,
      "balance_loss_clip": 1.02465034,
      "balance_loss_mlp": 1.01938391,
      "epoch": 0.5590861265594469,
      "flos": 70577432490240.0,
      "grad_norm": 0.6889700247642193,
      "language_loss": 0.52480936,
      "learning_rate": 1.7153002331730967e-06,
      "loss": 0.54531187,
      "num_input_tokens_seen": 200237010,
      "step": 9299,
      "time_per_iteration": 3.1776397228240967
    },
    {
      "auxiliary_loss_clip": 0.01094724,
      "auxiliary_loss_mlp": 0.01029909,
      "balance_loss_clip": 1.036973,
      "balance_loss_mlp": 1.01793456,
      "epoch": 0.5591462498121148,
      "flos": 30664408896000.0,
      "grad_norm": 2.081541097560223,
      "language_loss": 0.68651921,
      "learning_rate": 1.7149147430129824e-06,
      "loss": 0.70776552,
      "num_input_tokens_seen": 200260820,
      "step": 9300,
      "time_per_iteration": 2.6008825302124023
    },
    {
      "auxiliary_loss_clip": 0.01060439,
      "auxiliary_loss_mlp": 0.01048505,
      "balance_loss_clip": 1.03524494,
      "balance_loss_mlp": 1.0330677,
      "epoch": 0.5592063730647828,
      "flos": 18150438562560.0,
      "grad_norm": 1.7535674574501428,
      "language_loss": 0.81873721,
      "learning_rate": 1.7145292636633293e-06,
      "loss": 0.83982658,
      "num_input_tokens_seen": 200278035,
      "step": 9301,
      "time_per_iteration": 3.9318954944610596
    },
    {
      "auxiliary_loss_clip": 0.01108909,
      "auxiliary_loss_mlp": 0.01028666,
      "balance_loss_clip": 1.03641939,
      "balance_loss_mlp": 1.01595795,
      "epoch": 0.5592664963174507,
      "flos": 24060400945920.0,
      "grad_norm": 1.8715383153947094,
      "language_loss": 0.67560685,
      "learning_rate": 1.714143795138756e-06,
      "loss": 0.69698262,
      "num_input_tokens_seen": 200297255,
      "step": 9302,
      "time_per_iteration": 2.4877126216888428
    },
    {
      "auxiliary_loss_clip": 0.01079143,
      "auxiliary_loss_mlp": 0.01026456,
      "balance_loss_clip": 1.03685999,
      "balance_loss_mlp": 1.01280057,
      "epoch": 0.5593266195701188,
      "flos": 19827897661440.0,
      "grad_norm": 1.6409799682839519,
      "language_loss": 0.71072584,
      "learning_rate": 1.713758337453878e-06,
      "loss": 0.73178184,
      "num_input_tokens_seen": 200317505,
      "step": 9303,
      "time_per_iteration": 2.5849597454071045
    },
    {
      "auxiliary_loss_clip": 0.0105093,
      "auxiliary_loss_mlp": 0.01037208,
      "balance_loss_clip": 1.04224896,
      "balance_loss_mlp": 1.02343917,
      "epoch": 0.5593867428227867,
      "flos": 25300755440640.0,
      "grad_norm": 1.6142673580676925,
      "language_loss": 0.72790986,
      "learning_rate": 1.7133728906233124e-06,
      "loss": 0.74879122,
      "num_input_tokens_seen": 200338350,
      "step": 9304,
      "time_per_iteration": 2.6511220932006836
    },
    {
      "auxiliary_loss_clip": 0.01099198,
      "auxiliary_loss_mlp": 0.01026887,
      "balance_loss_clip": 1.03736722,
      "balance_loss_mlp": 1.01496601,
      "epoch": 0.5594468660754547,
      "flos": 12933013374720.0,
      "grad_norm": 1.874079009152534,
      "language_loss": 0.77963239,
      "learning_rate": 1.7129874546616763e-06,
      "loss": 0.80089325,
      "num_input_tokens_seen": 200353965,
      "step": 9305,
      "time_per_iteration": 2.4741809368133545
    },
    {
      "auxiliary_loss_clip": 0.01065702,
      "auxiliary_loss_mlp": 0.01028429,
      "balance_loss_clip": 1.04197383,
      "balance_loss_mlp": 1.01681852,
      "epoch": 0.5595069893281227,
      "flos": 19062713208960.0,
      "grad_norm": 1.676513155394174,
      "language_loss": 0.69372118,
      "learning_rate": 1.7126020295835836e-06,
      "loss": 0.71466255,
      "num_input_tokens_seen": 200373595,
      "step": 9306,
      "time_per_iteration": 2.5882318019866943
    },
    {
      "auxiliary_loss_clip": 0.01032163,
      "auxiliary_loss_mlp": 0.01006556,
      "balance_loss_clip": 1.03052533,
      "balance_loss_mlp": 1.00542974,
      "epoch": 0.5595671125807906,
      "flos": 70273375862400.0,
      "grad_norm": 0.9517494553785063,
      "language_loss": 0.60333657,
      "learning_rate": 1.7122166154036518e-06,
      "loss": 0.62372375,
      "num_input_tokens_seen": 200429155,
      "step": 9307,
      "time_per_iteration": 3.2533724308013916
    },
    {
      "auxiliary_loss_clip": 0.01101653,
      "auxiliary_loss_mlp": 0.0103418,
      "balance_loss_clip": 1.03889966,
      "balance_loss_mlp": 1.02252722,
      "epoch": 0.5596272358334586,
      "flos": 20665513889280.0,
      "grad_norm": 1.662121201733906,
      "language_loss": 0.74144435,
      "learning_rate": 1.7118312121364943e-06,
      "loss": 0.76280272,
      "num_input_tokens_seen": 200448290,
      "step": 9308,
      "time_per_iteration": 3.8773386478424072
    },
    {
      "auxiliary_loss_clip": 0.01045679,
      "auxiliary_loss_mlp": 0.01034167,
      "balance_loss_clip": 1.03381205,
      "balance_loss_mlp": 1.02021313,
      "epoch": 0.5596873590861265,
      "flos": 25041013217280.0,
      "grad_norm": 1.8882836832514818,
      "language_loss": 0.69511026,
      "learning_rate": 1.7114458197967257e-06,
      "loss": 0.71590877,
      "num_input_tokens_seen": 200466555,
      "step": 9309,
      "time_per_iteration": 2.6559715270996094
    },
    {
      "auxiliary_loss_clip": 0.01090759,
      "auxiliary_loss_mlp": 0.01033054,
      "balance_loss_clip": 1.04302597,
      "balance_loss_mlp": 1.01877916,
      "epoch": 0.5597474823387946,
      "flos": 25958387594880.0,
      "grad_norm": 1.9948411755617754,
      "language_loss": 0.75446773,
      "learning_rate": 1.7110604383989613e-06,
      "loss": 0.77570587,
      "num_input_tokens_seen": 200485980,
      "step": 9310,
      "time_per_iteration": 2.58152174949646
    },
    {
      "auxiliary_loss_clip": 0.01103228,
      "auxiliary_loss_mlp": 0.01031969,
      "balance_loss_clip": 1.03931928,
      "balance_loss_mlp": 1.01886201,
      "epoch": 0.5598076055914625,
      "flos": 26177442687360.0,
      "grad_norm": 2.541352849066168,
      "language_loss": 0.69325423,
      "learning_rate": 1.7106750679578133e-06,
      "loss": 0.71460629,
      "num_input_tokens_seen": 200504555,
      "step": 9311,
      "time_per_iteration": 2.52984619140625
    },
    {
      "auxiliary_loss_clip": 0.01098545,
      "auxiliary_loss_mlp": 0.01030531,
      "balance_loss_clip": 1.03770733,
      "balance_loss_mlp": 1.01866984,
      "epoch": 0.5598677288441305,
      "flos": 11655778590720.0,
      "grad_norm": 1.8758990424242923,
      "language_loss": 0.72236401,
      "learning_rate": 1.7102897084878962e-06,
      "loss": 0.74365479,
      "num_input_tokens_seen": 200522700,
      "step": 9312,
      "time_per_iteration": 3.8581783771514893
    },
    {
      "auxiliary_loss_clip": 0.01076877,
      "auxiliary_loss_mlp": 0.01031402,
      "balance_loss_clip": 1.04378629,
      "balance_loss_mlp": 1.01862895,
      "epoch": 0.5599278520967984,
      "flos": 22966597941120.0,
      "grad_norm": 1.7961384951975135,
      "language_loss": 0.8943485,
      "learning_rate": 1.709904360003822e-06,
      "loss": 0.91543126,
      "num_input_tokens_seen": 200541910,
      "step": 9313,
      "time_per_iteration": 2.5529215335845947
    },
    {
      "auxiliary_loss_clip": 0.01075227,
      "auxiliary_loss_mlp": 0.0103836,
      "balance_loss_clip": 1.03923512,
      "balance_loss_mlp": 1.02487743,
      "epoch": 0.5599879753494664,
      "flos": 21215557831680.0,
      "grad_norm": 1.4886928521994263,
      "language_loss": 0.7788862,
      "learning_rate": 1.709519022520204e-06,
      "loss": 0.80002207,
      "num_input_tokens_seen": 200562600,
      "step": 9314,
      "time_per_iteration": 2.591083526611328
    },
    {
      "auxiliary_loss_clip": 0.01076787,
      "auxiliary_loss_mlp": 0.01028373,
      "balance_loss_clip": 1.04045606,
      "balance_loss_mlp": 1.01643372,
      "epoch": 0.5600480986021343,
      "flos": 31903219105920.0,
      "grad_norm": 2.105222841137118,
      "language_loss": 0.70438743,
      "learning_rate": 1.7091336960516537e-06,
      "loss": 0.72543907,
      "num_input_tokens_seen": 200584795,
      "step": 9315,
      "time_per_iteration": 4.026345491409302
    },
    {
      "auxiliary_loss_clip": 0.01092855,
      "auxiliary_loss_mlp": 0.0103451,
      "balance_loss_clip": 1.03894186,
      "balance_loss_mlp": 1.02163553,
      "epoch": 0.5601082218548024,
      "flos": 28476048700800.0,
      "grad_norm": 2.211467999961726,
      "language_loss": 0.66429168,
      "learning_rate": 1.7087483806127824e-06,
      "loss": 0.68556535,
      "num_input_tokens_seen": 200606945,
      "step": 9316,
      "time_per_iteration": 2.569499969482422
    },
    {
      "auxiliary_loss_clip": 0.01076533,
      "auxiliary_loss_mlp": 0.01033773,
      "balance_loss_clip": 1.03805971,
      "balance_loss_mlp": 1.01884258,
      "epoch": 0.5601683451074703,
      "flos": 24097173494400.0,
      "grad_norm": 2.0741026604657997,
      "language_loss": 0.86383349,
      "learning_rate": 1.7083630762182022e-06,
      "loss": 0.88493657,
      "num_input_tokens_seen": 200626340,
      "step": 9317,
      "time_per_iteration": 2.570748805999756
    },
    {
      "auxiliary_loss_clip": 0.01102515,
      "auxiliary_loss_mlp": 0.0103889,
      "balance_loss_clip": 1.03849125,
      "balance_loss_mlp": 1.02474022,
      "epoch": 0.5602284683601383,
      "flos": 26356205698560.0,
      "grad_norm": 1.7243457283227637,
      "language_loss": 0.77314579,
      "learning_rate": 1.7079777828825233e-06,
      "loss": 0.79455984,
      "num_input_tokens_seen": 200644520,
      "step": 9318,
      "time_per_iteration": 2.5484185218811035
    },
    {
      "auxiliary_loss_clip": 0.0109766,
      "auxiliary_loss_mlp": 0.01038962,
      "balance_loss_clip": 1.03782666,
      "balance_loss_mlp": 1.02745795,
      "epoch": 0.5602885916128063,
      "flos": 24496392228480.0,
      "grad_norm": 1.7943379673439557,
      "language_loss": 0.76230872,
      "learning_rate": 1.7075925006203558e-06,
      "loss": 0.78367496,
      "num_input_tokens_seen": 200664845,
      "step": 9319,
      "time_per_iteration": 2.56019926071167
    },
    {
      "auxiliary_loss_clip": 0.01099232,
      "auxiliary_loss_mlp": 0.0103246,
      "balance_loss_clip": 1.03906703,
      "balance_loss_mlp": 1.02097988,
      "epoch": 0.5603487148654742,
      "flos": 27345006270720.0,
      "grad_norm": 1.5092001707115357,
      "language_loss": 0.85344785,
      "learning_rate": 1.7072072294463101e-06,
      "loss": 0.87476474,
      "num_input_tokens_seen": 200686535,
      "step": 9320,
      "time_per_iteration": 2.5362508296966553
    },
    {
      "auxiliary_loss_clip": 0.01041524,
      "auxiliary_loss_mlp": 0.01008643,
      "balance_loss_clip": 1.02621961,
      "balance_loss_mlp": 1.00734413,
      "epoch": 0.5604088381181422,
      "flos": 54087756180480.0,
      "grad_norm": 0.7458539124914149,
      "language_loss": 0.5254609,
      "learning_rate": 1.706821969374996e-06,
      "loss": 0.54596257,
      "num_input_tokens_seen": 200736965,
      "step": 9321,
      "time_per_iteration": 2.924009084701538
    },
    {
      "auxiliary_loss_clip": 0.01091628,
      "auxiliary_loss_mlp": 0.01032986,
      "balance_loss_clip": 1.04160309,
      "balance_loss_mlp": 1.02081442,
      "epoch": 0.5604689613708101,
      "flos": 22236390357120.0,
      "grad_norm": 1.3940571223144915,
      "language_loss": 0.74636042,
      "learning_rate": 1.7064367204210216e-06,
      "loss": 0.76760662,
      "num_input_tokens_seen": 200757420,
      "step": 9322,
      "time_per_iteration": 2.5459938049316406
    },
    {
      "auxiliary_loss_clip": 0.01111726,
      "auxiliary_loss_mlp": 0.01032005,
      "balance_loss_clip": 1.03866124,
      "balance_loss_mlp": 1.01843858,
      "epoch": 0.5605290846234782,
      "flos": 35297782940160.0,
      "grad_norm": 1.54102256411904,
      "language_loss": 0.73883295,
      "learning_rate": 1.7060514825989963e-06,
      "loss": 0.76027024,
      "num_input_tokens_seen": 200779520,
      "step": 9323,
      "time_per_iteration": 2.5787808895111084
    },
    {
      "auxiliary_loss_clip": 0.01090983,
      "auxiliary_loss_mlp": 0.01030303,
      "balance_loss_clip": 1.04297519,
      "balance_loss_mlp": 1.01696992,
      "epoch": 0.5605892078761461,
      "flos": 20263314326400.0,
      "grad_norm": 1.876011712378322,
      "language_loss": 0.61724555,
      "learning_rate": 1.7056662559235286e-06,
      "loss": 0.63845837,
      "num_input_tokens_seen": 200799485,
      "step": 9324,
      "time_per_iteration": 2.535074472427368
    },
    {
      "auxiliary_loss_clip": 0.01071107,
      "auxiliary_loss_mlp": 0.01031797,
      "balance_loss_clip": 1.0370568,
      "balance_loss_mlp": 1.01840377,
      "epoch": 0.5606493311288141,
      "flos": 17308333134720.0,
      "grad_norm": 1.8382714956922233,
      "language_loss": 0.87792736,
      "learning_rate": 1.705281040409226e-06,
      "loss": 0.89895636,
      "num_input_tokens_seen": 200817540,
      "step": 9325,
      "time_per_iteration": 2.560593366622925
    },
    {
      "auxiliary_loss_clip": 0.01089722,
      "auxiliary_loss_mlp": 0.01029751,
      "balance_loss_clip": 1.03756285,
      "balance_loss_mlp": 1.01653099,
      "epoch": 0.560709454381482,
      "flos": 21652985658240.0,
      "grad_norm": 1.6562763509642444,
      "language_loss": 0.74071145,
      "learning_rate": 1.7048958360706952e-06,
      "loss": 0.76190615,
      "num_input_tokens_seen": 200838380,
      "step": 9326,
      "time_per_iteration": 2.5099828243255615
    },
    {
      "auxiliary_loss_clip": 0.01091892,
      "auxiliary_loss_mlp": 0.01029884,
      "balance_loss_clip": 1.03790474,
      "balance_loss_mlp": 1.01628172,
      "epoch": 0.56076957763415,
      "flos": 20303355012480.0,
      "grad_norm": 1.9911858593542273,
      "language_loss": 0.78044271,
      "learning_rate": 1.7045106429225447e-06,
      "loss": 0.80166054,
      "num_input_tokens_seen": 200855640,
      "step": 9327,
      "time_per_iteration": 2.510442018508911
    },
    {
      "auxiliary_loss_clip": 0.01097905,
      "auxiliary_loss_mlp": 0.01029824,
      "balance_loss_clip": 1.04228163,
      "balance_loss_mlp": 1.01640153,
      "epoch": 0.5608297008868179,
      "flos": 25045897466880.0,
      "grad_norm": 1.4493306223439848,
      "language_loss": 0.78121883,
      "learning_rate": 1.7041254609793795e-06,
      "loss": 0.80249619,
      "num_input_tokens_seen": 200876585,
      "step": 9328,
      "time_per_iteration": 2.5305118560791016
    },
    {
      "auxiliary_loss_clip": 0.01110225,
      "auxiliary_loss_mlp": 0.01029719,
      "balance_loss_clip": 1.0378145,
      "balance_loss_mlp": 1.01698709,
      "epoch": 0.560889824139486,
      "flos": 19866825025920.0,
      "grad_norm": 1.5745731126480518,
      "language_loss": 0.73454034,
      "learning_rate": 1.7037402902558066e-06,
      "loss": 0.75593972,
      "num_input_tokens_seen": 200898175,
      "step": 9329,
      "time_per_iteration": 2.4786343574523926
    },
    {
      "auxiliary_loss_clip": 0.01090978,
      "auxiliary_loss_mlp": 0.00789981,
      "balance_loss_clip": 1.03804386,
      "balance_loss_mlp": 1.01403785,
      "epoch": 0.5609499473921539,
      "flos": 22929394429440.0,
      "grad_norm": 1.5866826951013648,
      "language_loss": 0.83143938,
      "learning_rate": 1.7033551307664324e-06,
      "loss": 0.85024893,
      "num_input_tokens_seen": 200917515,
      "step": 9330,
      "time_per_iteration": 2.5354905128479004
    },
    {
      "auxiliary_loss_clip": 0.01042815,
      "auxiliary_loss_mlp": 0.01003857,
      "balance_loss_clip": 1.01775885,
      "balance_loss_mlp": 1.00229573,
      "epoch": 0.5610100706448219,
      "flos": 53035825455360.0,
      "grad_norm": 0.7142013859279007,
      "language_loss": 0.57880241,
      "learning_rate": 1.7029699825258603e-06,
      "loss": 0.59926909,
      "num_input_tokens_seen": 200978615,
      "step": 9331,
      "time_per_iteration": 3.087099313735962
    },
    {
      "auxiliary_loss_clip": 0.0106843,
      "auxiliary_loss_mlp": 0.01027695,
      "balance_loss_clip": 1.03785789,
      "balance_loss_mlp": 1.01510596,
      "epoch": 0.5610701938974898,
      "flos": 21834944979840.0,
      "grad_norm": 1.879815463438497,
      "language_loss": 0.8186813,
      "learning_rate": 1.7025848455486971e-06,
      "loss": 0.83964252,
      "num_input_tokens_seen": 200997745,
      "step": 9332,
      "time_per_iteration": 2.597041606903076
    },
    {
      "auxiliary_loss_clip": 0.01100431,
      "auxiliary_loss_mlp": 0.01035171,
      "balance_loss_clip": 1.03918087,
      "balance_loss_mlp": 1.02059186,
      "epoch": 0.5611303171501578,
      "flos": 17457183095040.0,
      "grad_norm": 2.0775373317690713,
      "language_loss": 0.8164537,
      "learning_rate": 1.7021997198495454e-06,
      "loss": 0.83780968,
      "num_input_tokens_seen": 201016370,
      "step": 9333,
      "time_per_iteration": 2.4655468463897705
    },
    {
      "auxiliary_loss_clip": 0.01111977,
      "auxiliary_loss_mlp": 0.01023601,
      "balance_loss_clip": 1.03824282,
      "balance_loss_mlp": 1.01164436,
      "epoch": 0.5611904404028258,
      "flos": 22637799820800.0,
      "grad_norm": 1.8453909505917674,
      "language_loss": 0.72838748,
      "learning_rate": 1.7018146054430108e-06,
      "loss": 0.74974322,
      "num_input_tokens_seen": 201034310,
      "step": 9334,
      "time_per_iteration": 2.473379373550415
    },
    {
      "auxiliary_loss_clip": 0.01096113,
      "auxiliary_loss_mlp": 0.01034211,
      "balance_loss_clip": 1.0409447,
      "balance_loss_mlp": 1.02095532,
      "epoch": 0.5612505636554938,
      "flos": 14316327999360.0,
      "grad_norm": 1.8849672621860833,
      "language_loss": 0.71426183,
      "learning_rate": 1.7014295023436961e-06,
      "loss": 0.73556507,
      "num_input_tokens_seen": 201052030,
      "step": 9335,
      "time_per_iteration": 2.4770567417144775
    },
    {
      "auxiliary_loss_clip": 0.01090733,
      "auxiliary_loss_mlp": 0.01028536,
      "balance_loss_clip": 1.03755212,
      "balance_loss_mlp": 1.01572657,
      "epoch": 0.5613106869081618,
      "flos": 16508279554560.0,
      "grad_norm": 1.8156922276970768,
      "language_loss": 0.76860809,
      "learning_rate": 1.701044410566205e-06,
      "loss": 0.78980076,
      "num_input_tokens_seen": 201068445,
      "step": 9336,
      "time_per_iteration": 2.5070266723632812
    },
    {
      "auxiliary_loss_clip": 0.01101448,
      "auxiliary_loss_mlp": 0.01032146,
      "balance_loss_clip": 1.03978574,
      "balance_loss_mlp": 1.01964128,
      "epoch": 0.5613708101608297,
      "flos": 24058569352320.0,
      "grad_norm": 2.2375545258636174,
      "language_loss": 0.64473438,
      "learning_rate": 1.7006593301251393e-06,
      "loss": 0.66607034,
      "num_input_tokens_seen": 201082140,
      "step": 9337,
      "time_per_iteration": 2.4856998920440674
    },
    {
      "auxiliary_loss_clip": 0.01024051,
      "auxiliary_loss_mlp": 0.01001836,
      "balance_loss_clip": 1.01985121,
      "balance_loss_mlp": 1.00033951,
      "epoch": 0.5614309334134977,
      "flos": 64905735997440.0,
      "grad_norm": 0.8868348179993262,
      "language_loss": 0.62574863,
      "learning_rate": 1.700274261035102e-06,
      "loss": 0.64600754,
      "num_input_tokens_seen": 201137245,
      "step": 9338,
      "time_per_iteration": 3.0840704441070557
    },
    {
      "auxiliary_loss_clip": 0.01082776,
      "auxiliary_loss_mlp": 0.01031496,
      "balance_loss_clip": 1.03797936,
      "balance_loss_mlp": 1.01868713,
      "epoch": 0.5614910566661656,
      "flos": 32919849740160.0,
      "grad_norm": 3.279197656943634,
      "language_loss": 0.65949774,
      "learning_rate": 1.6998892033106946e-06,
      "loss": 0.68064046,
      "num_input_tokens_seen": 201157270,
      "step": 9339,
      "time_per_iteration": 2.6490259170532227
    },
    {
      "auxiliary_loss_clip": 0.01098018,
      "auxiliary_loss_mlp": 0.01039473,
      "balance_loss_clip": 1.03838491,
      "balance_loss_mlp": 1.02504897,
      "epoch": 0.5615511799188336,
      "flos": 18588871969920.0,
      "grad_norm": 2.303453153399956,
      "language_loss": 0.70590496,
      "learning_rate": 1.6995041569665184e-06,
      "loss": 0.7272799,
      "num_input_tokens_seen": 201174530,
      "step": 9340,
      "time_per_iteration": 3.883044958114624
    },
    {
      "auxiliary_loss_clip": 0.01071681,
      "auxiliary_loss_mlp": 0.01028288,
      "balance_loss_clip": 1.04391074,
      "balance_loss_mlp": 1.01590264,
      "epoch": 0.5616113031715015,
      "flos": 22820010537600.0,
      "grad_norm": 1.638559826412057,
      "language_loss": 0.76829135,
      "learning_rate": 1.6991191220171756e-06,
      "loss": 0.78929102,
      "num_input_tokens_seen": 201194905,
      "step": 9341,
      "time_per_iteration": 2.601888656616211
    },
    {
      "auxiliary_loss_clip": 0.01064228,
      "auxiliary_loss_mlp": 0.01031881,
      "balance_loss_clip": 1.0363977,
      "balance_loss_mlp": 1.01804662,
      "epoch": 0.5616714264241696,
      "flos": 22345702421760.0,
      "grad_norm": 1.6571228345986255,
      "language_loss": 0.79375285,
      "learning_rate": 1.6987340984772653e-06,
      "loss": 0.81471395,
      "num_input_tokens_seen": 201213715,
      "step": 9342,
      "time_per_iteration": 2.568110704421997
    },
    {
      "auxiliary_loss_clip": 0.01085196,
      "auxiliary_loss_mlp": 0.01030522,
      "balance_loss_clip": 1.03859007,
      "balance_loss_mlp": 1.01674139,
      "epoch": 0.5617315496768375,
      "flos": 18807783408000.0,
      "grad_norm": 2.0344669972784053,
      "language_loss": 0.76482189,
      "learning_rate": 1.6983490863613882e-06,
      "loss": 0.78597909,
      "num_input_tokens_seen": 201231415,
      "step": 9343,
      "time_per_iteration": 2.5452048778533936
    },
    {
      "auxiliary_loss_clip": 0.0107111,
      "auxiliary_loss_mlp": 0.0103515,
      "balance_loss_clip": 1.04025972,
      "balance_loss_mlp": 1.02182865,
      "epoch": 0.5617916729295055,
      "flos": 18369314087040.0,
      "grad_norm": 1.8572183807417713,
      "language_loss": 0.68752396,
      "learning_rate": 1.6979640856841442e-06,
      "loss": 0.70858657,
      "num_input_tokens_seen": 201249625,
      "step": 9344,
      "time_per_iteration": 2.564502239227295
    },
    {
      "auxiliary_loss_clip": 0.01113618,
      "auxiliary_loss_mlp": 0.01034877,
      "balance_loss_clip": 1.03929281,
      "balance_loss_mlp": 1.02095318,
      "epoch": 0.5618517961821734,
      "flos": 28179964892160.0,
      "grad_norm": 2.100795304990162,
      "language_loss": 0.66206849,
      "learning_rate": 1.6975790964601318e-06,
      "loss": 0.68355346,
      "num_input_tokens_seen": 201271205,
      "step": 9345,
      "time_per_iteration": 2.5180368423461914
    },
    {
      "auxiliary_loss_clip": 0.01092563,
      "auxiliary_loss_mlp": 0.01030571,
      "balance_loss_clip": 1.04067087,
      "balance_loss_mlp": 1.0179702,
      "epoch": 0.5619119194348414,
      "flos": 15486872411520.0,
      "grad_norm": 1.9679433082372297,
      "language_loss": 0.87439847,
      "learning_rate": 1.6971941187039512e-06,
      "loss": 0.89562976,
      "num_input_tokens_seen": 201287700,
      "step": 9346,
      "time_per_iteration": 2.476707935333252
    },
    {
      "auxiliary_loss_clip": 0.01093545,
      "auxiliary_loss_mlp": 0.01031666,
      "balance_loss_clip": 1.03859186,
      "balance_loss_mlp": 1.01818347,
      "epoch": 0.5619720426875094,
      "flos": 29128652951040.0,
      "grad_norm": 2.215882990486965,
      "language_loss": 0.5927698,
      "learning_rate": 1.6968091524301993e-06,
      "loss": 0.6140219,
      "num_input_tokens_seen": 201307530,
      "step": 9347,
      "time_per_iteration": 3.933993101119995
    },
    {
      "auxiliary_loss_clip": 0.01104372,
      "auxiliary_loss_mlp": 0.01031903,
      "balance_loss_clip": 1.03979254,
      "balance_loss_mlp": 1.01818812,
      "epoch": 0.5620321659401774,
      "flos": 18003743418240.0,
      "grad_norm": 2.304931331570922,
      "language_loss": 0.69189036,
      "learning_rate": 1.6964241976534745e-06,
      "loss": 0.71325308,
      "num_input_tokens_seen": 201326210,
      "step": 9348,
      "time_per_iteration": 2.451066493988037
    },
    {
      "auxiliary_loss_clip": 0.01068484,
      "auxiliary_loss_mlp": 0.01029134,
      "balance_loss_clip": 1.04046798,
      "balance_loss_mlp": 1.01471531,
      "epoch": 0.5620922891928454,
      "flos": 20594518657920.0,
      "grad_norm": 1.7982387438774174,
      "language_loss": 0.79603934,
      "learning_rate": 1.6960392543883754e-06,
      "loss": 0.81701553,
      "num_input_tokens_seen": 201346120,
      "step": 9349,
      "time_per_iteration": 2.5846493244171143
    },
    {
      "auxiliary_loss_clip": 0.01062185,
      "auxiliary_loss_mlp": 0.01031226,
      "balance_loss_clip": 1.03984249,
      "balance_loss_mlp": 1.01785052,
      "epoch": 0.5621524124455133,
      "flos": 26287006147200.0,
      "grad_norm": 2.016058167229315,
      "language_loss": 0.66746902,
      "learning_rate": 1.6956543226494975e-06,
      "loss": 0.68840313,
      "num_input_tokens_seen": 201365700,
      "step": 9350,
      "time_per_iteration": 4.067665338516235
    },
    {
      "auxiliary_loss_clip": 0.01073501,
      "auxiliary_loss_mlp": 0.01038237,
      "balance_loss_clip": 1.04419971,
      "balance_loss_mlp": 1.02432501,
      "epoch": 0.5622125356981813,
      "flos": 12750299867520.0,
      "grad_norm": 2.0500690679256444,
      "language_loss": 0.78735316,
      "learning_rate": 1.6952694024514381e-06,
      "loss": 0.80847055,
      "num_input_tokens_seen": 201382795,
      "step": 9351,
      "time_per_iteration": 2.5577759742736816
    },
    {
      "auxiliary_loss_clip": 0.01093227,
      "auxiliary_loss_mlp": 0.00789144,
      "balance_loss_clip": 1.0379926,
      "balance_loss_mlp": 1.01164389,
      "epoch": 0.5622726589508492,
      "flos": 23805327490560.0,
      "grad_norm": 1.6073788900393462,
      "language_loss": 0.59213662,
      "learning_rate": 1.6948844938087945e-06,
      "loss": 0.61096025,
      "num_input_tokens_seen": 201402780,
      "step": 9352,
      "time_per_iteration": 2.5616302490234375
    },
    {
      "auxiliary_loss_clip": 0.01099843,
      "auxiliary_loss_mlp": 0.01032175,
      "balance_loss_clip": 1.03955722,
      "balance_loss_mlp": 1.01988459,
      "epoch": 0.5623327822035172,
      "flos": 24718212668160.0,
      "grad_norm": 1.3064064455625075,
      "language_loss": 0.72121274,
      "learning_rate": 1.6944995967361604e-06,
      "loss": 0.74253297,
      "num_input_tokens_seen": 201424140,
      "step": 9353,
      "time_per_iteration": 2.549245834350586
    },
    {
      "auxiliary_loss_clip": 0.01091485,
      "auxiliary_loss_mlp": 0.01027659,
      "balance_loss_clip": 1.03886986,
      "balance_loss_mlp": 1.01443863,
      "epoch": 0.5623929054561851,
      "flos": 14019274523520.0,
      "grad_norm": 2.5594044568081027,
      "language_loss": 0.76142114,
      "learning_rate": 1.6941147112481327e-06,
      "loss": 0.78261262,
      "num_input_tokens_seen": 201439645,
      "step": 9354,
      "time_per_iteration": 3.862574338912964
    },
    {
      "auxiliary_loss_clip": 0.01082287,
      "auxiliary_loss_mlp": 0.01034386,
      "balance_loss_clip": 1.03830123,
      "balance_loss_mlp": 1.02163625,
      "epoch": 0.5624530287088532,
      "flos": 20704405340160.0,
      "grad_norm": 1.9631790305060648,
      "language_loss": 0.72727287,
      "learning_rate": 1.6937298373593056e-06,
      "loss": 0.74843955,
      "num_input_tokens_seen": 201459970,
      "step": 9355,
      "time_per_iteration": 2.565553903579712
    },
    {
      "auxiliary_loss_clip": 0.01100663,
      "auxiliary_loss_mlp": 0.01030477,
      "balance_loss_clip": 1.04190707,
      "balance_loss_mlp": 1.01770973,
      "epoch": 0.5625131519615211,
      "flos": 21470918595840.0,
      "grad_norm": 1.5587871446400752,
      "language_loss": 0.73628783,
      "learning_rate": 1.693344975084274e-06,
      "loss": 0.75759917,
      "num_input_tokens_seen": 201480055,
      "step": 9356,
      "time_per_iteration": 2.501264810562134
    },
    {
      "auxiliary_loss_clip": 0.01111598,
      "auxiliary_loss_mlp": 0.01034447,
      "balance_loss_clip": 1.03874242,
      "balance_loss_mlp": 1.02170348,
      "epoch": 0.5625732752141891,
      "flos": 18698004466560.0,
      "grad_norm": 1.807961351399199,
      "language_loss": 0.83234429,
      "learning_rate": 1.6929601244376318e-06,
      "loss": 0.85380483,
      "num_input_tokens_seen": 201497645,
      "step": 9357,
      "time_per_iteration": 2.454166889190674
    },
    {
      "auxiliary_loss_clip": 0.01100978,
      "auxiliary_loss_mlp": 0.01031851,
      "balance_loss_clip": 1.03882813,
      "balance_loss_mlp": 1.01958418,
      "epoch": 0.562633398466857,
      "flos": 16216900427520.0,
      "grad_norm": 1.9660649183852326,
      "language_loss": 0.72328264,
      "learning_rate": 1.6925752854339722e-06,
      "loss": 0.74461091,
      "num_input_tokens_seen": 201515455,
      "step": 9358,
      "time_per_iteration": 2.4750874042510986
    },
    {
      "auxiliary_loss_clip": 0.01110108,
      "auxiliary_loss_mlp": 0.01041243,
      "balance_loss_clip": 1.03749764,
      "balance_loss_mlp": 1.02810621,
      "epoch": 0.562693521719525,
      "flos": 22491930689280.0,
      "grad_norm": 1.6024787990662106,
      "language_loss": 0.77394766,
      "learning_rate": 1.6921904580878885e-06,
      "loss": 0.79546118,
      "num_input_tokens_seen": 201534500,
      "step": 9359,
      "time_per_iteration": 2.4744019508361816
    },
    {
      "auxiliary_loss_clip": 0.01090111,
      "auxiliary_loss_mlp": 0.01029637,
      "balance_loss_clip": 1.03858745,
      "balance_loss_mlp": 1.0175674,
      "epoch": 0.562753644972193,
      "flos": 25331171281920.0,
      "grad_norm": 1.9846689235623096,
      "language_loss": 0.70311201,
      "learning_rate": 1.6918056424139736e-06,
      "loss": 0.72430944,
      "num_input_tokens_seen": 201553280,
      "step": 9360,
      "time_per_iteration": 2.5659847259521484
    },
    {
      "auxiliary_loss_clip": 0.01006198,
      "auxiliary_loss_mlp": 0.01004226,
      "balance_loss_clip": 1.02138138,
      "balance_loss_mlp": 1.00301039,
      "epoch": 0.562813768224861,
      "flos": 67392622126080.0,
      "grad_norm": 0.7739893344841449,
      "language_loss": 0.55600226,
      "learning_rate": 1.6914208384268197e-06,
      "loss": 0.57610649,
      "num_input_tokens_seen": 201610030,
      "step": 9361,
      "time_per_iteration": 3.1143572330474854
    },
    {
      "auxiliary_loss_clip": 0.01088486,
      "auxiliary_loss_mlp": 0.01032795,
      "balance_loss_clip": 1.04013407,
      "balance_loss_mlp": 1.02079654,
      "epoch": 0.562873891477529,
      "flos": 23331163029120.0,
      "grad_norm": 1.488539213613743,
      "language_loss": 0.81743741,
      "learning_rate": 1.691036046141018e-06,
      "loss": 0.83865023,
      "num_input_tokens_seen": 201628370,
      "step": 9362,
      "time_per_iteration": 2.5324044227600098
    },
    {
      "auxiliary_loss_clip": 0.01077023,
      "auxiliary_loss_mlp": 0.00784802,
      "balance_loss_clip": 1.03751123,
      "balance_loss_mlp": 1.00905442,
      "epoch": 0.5629340147301969,
      "flos": 38472824805120.0,
      "grad_norm": 1.6304881807604725,
      "language_loss": 0.74809235,
      "learning_rate": 1.6906512655711614e-06,
      "loss": 0.76671058,
      "num_input_tokens_seen": 201649790,
      "step": 9363,
      "time_per_iteration": 2.703882932662964
    },
    {
      "auxiliary_loss_clip": 0.01103976,
      "auxiliary_loss_mlp": 0.01033823,
      "balance_loss_clip": 1.03822982,
      "balance_loss_mlp": 1.02025056,
      "epoch": 0.5629941379828649,
      "flos": 29242023252480.0,
      "grad_norm": 1.710527994379688,
      "language_loss": 0.83184212,
      "learning_rate": 1.690266496731839e-06,
      "loss": 0.85322011,
      "num_input_tokens_seen": 201669175,
      "step": 9364,
      "time_per_iteration": 2.554075002670288
    },
    {
      "auxiliary_loss_clip": 0.01079624,
      "auxiliary_loss_mlp": 0.0103297,
      "balance_loss_clip": 1.03799689,
      "balance_loss_mlp": 1.02106071,
      "epoch": 0.5630542612355328,
      "flos": 19420885676160.0,
      "grad_norm": 2.0800502094097832,
      "language_loss": 0.64866316,
      "learning_rate": 1.689881739637642e-06,
      "loss": 0.66978908,
      "num_input_tokens_seen": 201687000,
      "step": 9365,
      "time_per_iteration": 2.525122880935669
    },
    {
      "auxiliary_loss_clip": 0.01091095,
      "auxiliary_loss_mlp": 0.01034414,
      "balance_loss_clip": 1.03925359,
      "balance_loss_mlp": 1.02105093,
      "epoch": 0.5631143844882008,
      "flos": 22266303408000.0,
      "grad_norm": 3.453108107726314,
      "language_loss": 0.81807137,
      "learning_rate": 1.6894969943031611e-06,
      "loss": 0.8393265,
      "num_input_tokens_seen": 201703335,
      "step": 9366,
      "time_per_iteration": 2.5021493434906006
    },
    {
      "auxiliary_loss_clip": 0.01109566,
      "auxiliary_loss_mlp": 0.01027982,
      "balance_loss_clip": 1.03839612,
      "balance_loss_mlp": 1.01622832,
      "epoch": 0.5631745077408687,
      "flos": 22965305051520.0,
      "grad_norm": 1.5515847825215352,
      "language_loss": 0.73240149,
      "learning_rate": 1.6891122607429845e-06,
      "loss": 0.75377691,
      "num_input_tokens_seen": 201723495,
      "step": 9367,
      "time_per_iteration": 2.487976312637329
    },
    {
      "auxiliary_loss_clip": 0.01021211,
      "auxiliary_loss_mlp": 0.01007049,
      "balance_loss_clip": 1.01759744,
      "balance_loss_mlp": 1.00563645,
      "epoch": 0.5632346309935368,
      "flos": 65080515576960.0,
      "grad_norm": 0.6263592657517756,
      "language_loss": 0.53490776,
      "learning_rate": 1.6887275389717028e-06,
      "loss": 0.55519038,
      "num_input_tokens_seen": 201792615,
      "step": 9368,
      "time_per_iteration": 3.229602813720703
    },
    {
      "auxiliary_loss_clip": 0.01113177,
      "auxiliary_loss_mlp": 0.01033855,
      "balance_loss_clip": 1.04071581,
      "balance_loss_mlp": 1.02124834,
      "epoch": 0.5632947542462047,
      "flos": 23002903612800.0,
      "grad_norm": 1.8234014189515582,
      "language_loss": 0.69244248,
      "learning_rate": 1.6883428290039046e-06,
      "loss": 0.71391279,
      "num_input_tokens_seen": 201812520,
      "step": 9369,
      "time_per_iteration": 2.480339765548706
    },
    {
      "auxiliary_loss_clip": 0.01075581,
      "auxiliary_loss_mlp": 0.01034334,
      "balance_loss_clip": 1.0343945,
      "balance_loss_mlp": 1.02131617,
      "epoch": 0.5633548774988727,
      "flos": 30482593228800.0,
      "grad_norm": 1.825745429638711,
      "language_loss": 0.76096964,
      "learning_rate": 1.6879581308541763e-06,
      "loss": 0.78206879,
      "num_input_tokens_seen": 201834185,
      "step": 9370,
      "time_per_iteration": 2.6140341758728027
    },
    {
      "auxiliary_loss_clip": 0.01095187,
      "auxiliary_loss_mlp": 0.01032704,
      "balance_loss_clip": 1.03973377,
      "balance_loss_mlp": 1.01877999,
      "epoch": 0.5634150007515406,
      "flos": 18515039564160.0,
      "grad_norm": 2.5909808535340755,
      "language_loss": 0.75996423,
      "learning_rate": 1.687573444537108e-06,
      "loss": 0.78124309,
      "num_input_tokens_seen": 201851305,
      "step": 9371,
      "time_per_iteration": 2.530202865600586
    },
    {
      "auxiliary_loss_clip": 0.01098329,
      "auxiliary_loss_mlp": 0.01034136,
      "balance_loss_clip": 1.03762794,
      "balance_loss_mlp": 1.0225842,
      "epoch": 0.5634751240042086,
      "flos": 19244672530560.0,
      "grad_norm": 1.8071363711463713,
      "language_loss": 0.76317263,
      "learning_rate": 1.687188770067285e-06,
      "loss": 0.78449726,
      "num_input_tokens_seen": 201870350,
      "step": 9372,
      "time_per_iteration": 2.4666340351104736
    },
    {
      "auxiliary_loss_clip": 0.01084013,
      "auxiliary_loss_mlp": 0.01029356,
      "balance_loss_clip": 1.03692687,
      "balance_loss_mlp": 1.01659465,
      "epoch": 0.5635352472568766,
      "flos": 12020630987520.0,
      "grad_norm": 2.2339747876177327,
      "language_loss": 0.72041327,
      "learning_rate": 1.6868041074592956e-06,
      "loss": 0.74154699,
      "num_input_tokens_seen": 201886800,
      "step": 9373,
      "time_per_iteration": 2.500312566757202
    },
    {
      "auxiliary_loss_clip": 0.01079994,
      "auxiliary_loss_mlp": 0.01029511,
      "balance_loss_clip": 1.03900743,
      "balance_loss_mlp": 1.01548052,
      "epoch": 0.5635953705095446,
      "flos": 21871645701120.0,
      "grad_norm": 2.169217842665235,
      "language_loss": 0.82631791,
      "learning_rate": 1.6864194567277264e-06,
      "loss": 0.847413,
      "num_input_tokens_seen": 201904730,
      "step": 9374,
      "time_per_iteration": 2.559629201889038
    },
    {
      "auxiliary_loss_clip": 0.01096293,
      "auxiliary_loss_mlp": 0.01026069,
      "balance_loss_clip": 1.03486478,
      "balance_loss_mlp": 1.01389194,
      "epoch": 0.5636554937622126,
      "flos": 27126166659840.0,
      "grad_norm": 1.8846048261226038,
      "language_loss": 0.66332394,
      "learning_rate": 1.6860348178871618e-06,
      "loss": 0.68454754,
      "num_input_tokens_seen": 201924850,
      "step": 9375,
      "time_per_iteration": 2.5452280044555664
    },
    {
      "auxiliary_loss_clip": 0.01076355,
      "auxiliary_loss_mlp": 0.00785803,
      "balance_loss_clip": 1.03947616,
      "balance_loss_mlp": 1.01079333,
      "epoch": 0.5637156170148805,
      "flos": 12926405272320.0,
      "grad_norm": 2.3234475342398073,
      "language_loss": 0.81036025,
      "learning_rate": 1.6856501909521889e-06,
      "loss": 0.82898188,
      "num_input_tokens_seen": 201939500,
      "step": 9376,
      "time_per_iteration": 2.535027265548706
    },
    {
      "auxiliary_loss_clip": 0.01092978,
      "auxiliary_loss_mlp": 0.0103248,
      "balance_loss_clip": 1.03780413,
      "balance_loss_mlp": 1.01926529,
      "epoch": 0.5637757402675485,
      "flos": 45551033130240.0,
      "grad_norm": 2.890655240608413,
      "language_loss": 0.69575053,
      "learning_rate": 1.6852655759373925e-06,
      "loss": 0.71700513,
      "num_input_tokens_seen": 201963000,
      "step": 9377,
      "time_per_iteration": 2.7293825149536133
    },
    {
      "auxiliary_loss_clip": 0.01074432,
      "auxiliary_loss_mlp": 0.01029122,
      "balance_loss_clip": 1.03781152,
      "balance_loss_mlp": 1.01694465,
      "epoch": 0.5638358635202164,
      "flos": 20886041439360.0,
      "grad_norm": 1.3939822536844346,
      "language_loss": 0.74550712,
      "learning_rate": 1.6848809728573565e-06,
      "loss": 0.76654267,
      "num_input_tokens_seen": 201983145,
      "step": 9378,
      "time_per_iteration": 3.932915687561035
    },
    {
      "auxiliary_loss_clip": 0.01116313,
      "auxiliary_loss_mlp": 0.01030499,
      "balance_loss_clip": 1.03789353,
      "balance_loss_mlp": 1.01710629,
      "epoch": 0.5638959867728844,
      "flos": 18806562345600.0,
      "grad_norm": 2.634038597833133,
      "language_loss": 0.81880987,
      "learning_rate": 1.6844963817266656e-06,
      "loss": 0.84027803,
      "num_input_tokens_seen": 202000335,
      "step": 9379,
      "time_per_iteration": 2.470479726791382
    },
    {
      "auxiliary_loss_clip": 0.01087478,
      "auxiliary_loss_mlp": 0.01032588,
      "balance_loss_clip": 1.03610086,
      "balance_loss_mlp": 1.01983869,
      "epoch": 0.5639561100255523,
      "flos": 27490336698240.0,
      "grad_norm": 2.004599446517043,
      "language_loss": 0.72453254,
      "learning_rate": 1.6841118025599042e-06,
      "loss": 0.74573326,
      "num_input_tokens_seen": 202018275,
      "step": 9380,
      "time_per_iteration": 2.5774669647216797
    },
    {
      "auxiliary_loss_clip": 0.01076245,
      "auxiliary_loss_mlp": 0.01033224,
      "balance_loss_clip": 1.03908956,
      "balance_loss_mlp": 1.01909125,
      "epoch": 0.5640162332782204,
      "flos": 18076570243200.0,
      "grad_norm": 2.1803062428701487,
      "language_loss": 0.74085033,
      "learning_rate": 1.6837272353716542e-06,
      "loss": 0.76194501,
      "num_input_tokens_seen": 202034330,
      "step": 9381,
      "time_per_iteration": 2.563936948776245
    },
    {
      "auxiliary_loss_clip": 0.01063693,
      "auxiliary_loss_mlp": 0.0103206,
      "balance_loss_clip": 1.03514504,
      "balance_loss_mlp": 1.01887584,
      "epoch": 0.5640763565308883,
      "flos": 20884856290560.0,
      "grad_norm": 1.9192443092313876,
      "language_loss": 0.71914607,
      "learning_rate": 1.683342680176499e-06,
      "loss": 0.7401036,
      "num_input_tokens_seen": 202053100,
      "step": 9382,
      "time_per_iteration": 2.5952281951904297
    },
    {
      "auxiliary_loss_clip": 0.0104088,
      "auxiliary_loss_mlp": 0.01003718,
      "balance_loss_clip": 1.01596165,
      "balance_loss_mlp": 1.00258541,
      "epoch": 0.5641364797835563,
      "flos": 64447912224000.0,
      "grad_norm": 0.7348883059868159,
      "language_loss": 0.5439325,
      "learning_rate": 1.682958136989022e-06,
      "loss": 0.5643785,
      "num_input_tokens_seen": 202120125,
      "step": 9383,
      "time_per_iteration": 3.1972410678863525
    },
    {
      "auxiliary_loss_clip": 0.01106318,
      "auxiliary_loss_mlp": 0.01025779,
      "balance_loss_clip": 1.03903937,
      "balance_loss_mlp": 1.01225507,
      "epoch": 0.5641966030362242,
      "flos": 18660944609280.0,
      "grad_norm": 1.8086907017111076,
      "language_loss": 0.70529813,
      "learning_rate": 1.6825736058238033e-06,
      "loss": 0.72661906,
      "num_input_tokens_seen": 202138030,
      "step": 9384,
      "time_per_iteration": 2.4926223754882812
    },
    {
      "auxiliary_loss_clip": 0.01088484,
      "auxiliary_loss_mlp": 0.01031245,
      "balance_loss_clip": 1.03737712,
      "balance_loss_mlp": 1.01758409,
      "epoch": 0.5642567262888922,
      "flos": 22492325738880.0,
      "grad_norm": 1.9390452495247288,
      "language_loss": 0.75992256,
      "learning_rate": 1.6821890866954263e-06,
      "loss": 0.78111982,
      "num_input_tokens_seen": 202155580,
      "step": 9385,
      "time_per_iteration": 2.538809061050415
    },
    {
      "auxiliary_loss_clip": 0.01097528,
      "auxiliary_loss_mlp": 0.01031364,
      "balance_loss_clip": 1.03587413,
      "balance_loss_mlp": 1.01850152,
      "epoch": 0.5643168495415603,
      "flos": 13003972692480.0,
      "grad_norm": 1.8727301872221094,
      "language_loss": 0.82192922,
      "learning_rate": 1.6818045796184703e-06,
      "loss": 0.84321809,
      "num_input_tokens_seen": 202170365,
      "step": 9386,
      "time_per_iteration": 3.8266384601593018
    },
    {
      "auxiliary_loss_clip": 0.01109576,
      "auxiliary_loss_mlp": 0.01033686,
      "balance_loss_clip": 1.04208469,
      "balance_loss_mlp": 1.01995897,
      "epoch": 0.5643769727942282,
      "flos": 18588297352320.0,
      "grad_norm": 1.8946487539682784,
      "language_loss": 0.69898617,
      "learning_rate": 1.681420084607516e-06,
      "loss": 0.72041881,
      "num_input_tokens_seen": 202189095,
      "step": 9387,
      "time_per_iteration": 2.4678027629852295
    },
    {
      "auxiliary_loss_clip": 0.01103875,
      "auxiliary_loss_mlp": 0.01031184,
      "balance_loss_clip": 1.0389564,
      "balance_loss_mlp": 1.01866698,
      "epoch": 0.5644370960468962,
      "flos": 33806269572480.0,
      "grad_norm": 1.6670007568389555,
      "language_loss": 0.74526334,
      "learning_rate": 1.6810356016771452e-06,
      "loss": 0.76661396,
      "num_input_tokens_seen": 202213500,
      "step": 9388,
      "time_per_iteration": 2.626058578491211
    },
    {
      "auxiliary_loss_clip": 0.01096248,
      "auxiliary_loss_mlp": 0.01030777,
      "balance_loss_clip": 1.03724921,
      "balance_loss_mlp": 1.0195055,
      "epoch": 0.5644972192995641,
      "flos": 21214911386880.0,
      "grad_norm": 1.6060694701436016,
      "language_loss": 0.82418311,
      "learning_rate": 1.6806511308419353e-06,
      "loss": 0.84545338,
      "num_input_tokens_seen": 202231920,
      "step": 9389,
      "time_per_iteration": 3.9387266635894775
    },
    {
      "auxiliary_loss_clip": 0.01078195,
      "auxiliary_loss_mlp": 0.01031208,
      "balance_loss_clip": 1.03638661,
      "balance_loss_mlp": 1.01712966,
      "epoch": 0.5645573425522321,
      "flos": 18587722734720.0,
      "grad_norm": 2.1402179283191947,
      "language_loss": 0.63467348,
      "learning_rate": 1.680266672116467e-06,
      "loss": 0.6557675,
      "num_input_tokens_seen": 202247600,
      "step": 9390,
      "time_per_iteration": 2.530947685241699
    },
    {
      "auxiliary_loss_clip": 0.01090565,
      "auxiliary_loss_mlp": 0.01028264,
      "balance_loss_clip": 1.04131854,
      "balance_loss_mlp": 1.01739788,
      "epoch": 0.5646174658049,
      "flos": 18113809668480.0,
      "grad_norm": 1.698357523672355,
      "language_loss": 0.92235613,
      "learning_rate": 1.6798822255153192e-06,
      "loss": 0.94354439,
      "num_input_tokens_seen": 202265350,
      "step": 9391,
      "time_per_iteration": 2.5351412296295166
    },
    {
      "auxiliary_loss_clip": 0.0110622,
      "auxiliary_loss_mlp": 0.01033582,
      "balance_loss_clip": 1.03977036,
      "balance_loss_mlp": 1.02008104,
      "epoch": 0.564677589057568,
      "flos": 28329964087680.0,
      "grad_norm": 2.506781699755024,
      "language_loss": 0.59922493,
      "learning_rate": 1.6794977910530684e-06,
      "loss": 0.62062299,
      "num_input_tokens_seen": 202284285,
      "step": 9392,
      "time_per_iteration": 2.5862443447113037
    },
    {
      "auxiliary_loss_clip": 0.01066851,
      "auxiliary_loss_mlp": 0.01028316,
      "balance_loss_clip": 1.03738284,
      "balance_loss_mlp": 1.01454687,
      "epoch": 0.564737712310236,
      "flos": 22163743100160.0,
      "grad_norm": 2.235141862550795,
      "language_loss": 0.8123157,
      "learning_rate": 1.6791133687442937e-06,
      "loss": 0.83326733,
      "num_input_tokens_seen": 202303450,
      "step": 9393,
      "time_per_iteration": 3.9696593284606934
    },
    {
      "auxiliary_loss_clip": 0.0108975,
      "auxiliary_loss_mlp": 0.01027999,
      "balance_loss_clip": 1.03925169,
      "balance_loss_mlp": 1.01563156,
      "epoch": 0.564797835562904,
      "flos": 20959011918720.0,
      "grad_norm": 1.6813817087844667,
      "language_loss": 0.87495577,
      "learning_rate": 1.6787289586035725e-06,
      "loss": 0.8961333,
      "num_input_tokens_seen": 202322315,
      "step": 9394,
      "time_per_iteration": 2.5262956619262695
    },
    {
      "auxiliary_loss_clip": 0.01101296,
      "auxiliary_loss_mlp": 0.01031042,
      "balance_loss_clip": 1.03990197,
      "balance_loss_mlp": 1.01884079,
      "epoch": 0.5648579588155719,
      "flos": 17420302805760.0,
      "grad_norm": 1.8154715138474475,
      "language_loss": 0.84764075,
      "learning_rate": 1.6783445606454814e-06,
      "loss": 0.8689642,
      "num_input_tokens_seen": 202339905,
      "step": 9395,
      "time_per_iteration": 2.4802451133728027
    },
    {
      "auxiliary_loss_clip": 0.01033372,
      "auxiliary_loss_mlp": 0.01006163,
      "balance_loss_clip": 1.0189352,
      "balance_loss_mlp": 1.00503635,
      "epoch": 0.5649180820682399,
      "flos": 69929568835200.0,
      "grad_norm": 0.8120434171049438,
      "language_loss": 0.58315897,
      "learning_rate": 1.677960174884597e-06,
      "loss": 0.60355425,
      "num_input_tokens_seen": 202397320,
      "step": 9396,
      "time_per_iteration": 3.101322889328003
    },
    {
      "auxiliary_loss_clip": 0.01092862,
      "auxiliary_loss_mlp": 0.01027955,
      "balance_loss_clip": 1.03923392,
      "balance_loss_mlp": 1.01577806,
      "epoch": 0.5649782053209078,
      "flos": 24973070641920.0,
      "grad_norm": 2.1540038890366477,
      "language_loss": 0.69734752,
      "learning_rate": 1.6775758013354943e-06,
      "loss": 0.71855569,
      "num_input_tokens_seen": 202416865,
      "step": 9397,
      "time_per_iteration": 2.5832560062408447
    },
    {
      "auxiliary_loss_clip": 0.01080149,
      "auxiliary_loss_mlp": 0.01032477,
      "balance_loss_clip": 1.04243374,
      "balance_loss_mlp": 1.02063322,
      "epoch": 0.5650383285735758,
      "flos": 21726602582400.0,
      "grad_norm": 1.888326422181968,
      "language_loss": 0.67385888,
      "learning_rate": 1.67719144001275e-06,
      "loss": 0.69498515,
      "num_input_tokens_seen": 202436210,
      "step": 9398,
      "time_per_iteration": 2.598712205886841
    },
    {
      "auxiliary_loss_clip": 0.01025956,
      "auxiliary_loss_mlp": 0.01000848,
      "balance_loss_clip": 1.01994824,
      "balance_loss_mlp": 0.99962026,
      "epoch": 0.5650984518262439,
      "flos": 65904484636800.0,
      "grad_norm": 0.8333174124160037,
      "language_loss": 0.58152932,
      "learning_rate": 1.6768070909309386e-06,
      "loss": 0.60179734,
      "num_input_tokens_seen": 202492925,
      "step": 9399,
      "time_per_iteration": 3.1023948192596436
    },
    {
      "auxiliary_loss_clip": 0.01069043,
      "auxiliary_loss_mlp": 0.01036029,
      "balance_loss_clip": 1.03508413,
      "balance_loss_mlp": 1.0211997,
      "epoch": 0.5651585750789118,
      "flos": 21032592929280.0,
      "grad_norm": 1.8590439226234374,
      "language_loss": 0.73216635,
      "learning_rate": 1.6764227541046347e-06,
      "loss": 0.7532171,
      "num_input_tokens_seen": 202511905,
      "step": 9400,
      "time_per_iteration": 2.5995612144470215
    },
    {
      "auxiliary_loss_clip": 0.01085246,
      "auxiliary_loss_mlp": 0.01032467,
      "balance_loss_clip": 1.0402267,
      "balance_loss_mlp": 1.01838279,
      "epoch": 0.5652186983315798,
      "flos": 18551919853440.0,
      "grad_norm": 1.9318558765516585,
      "language_loss": 0.61068463,
      "learning_rate": 1.676038429548412e-06,
      "loss": 0.63186181,
      "num_input_tokens_seen": 202529815,
      "step": 9401,
      "time_per_iteration": 2.5785653591156006
    },
    {
      "auxiliary_loss_clip": 0.01080643,
      "auxiliary_loss_mlp": 0.01028243,
      "balance_loss_clip": 1.03759658,
      "balance_loss_mlp": 1.01608348,
      "epoch": 0.5652788215842477,
      "flos": 18478662065280.0,
      "grad_norm": 1.8467456792970585,
      "language_loss": 0.81327617,
      "learning_rate": 1.6756541172768453e-06,
      "loss": 0.83436501,
      "num_input_tokens_seen": 202547710,
      "step": 9402,
      "time_per_iteration": 2.533280849456787
    },
    {
      "auxiliary_loss_clip": 0.010595,
      "auxiliary_loss_mlp": 0.01042671,
      "balance_loss_clip": 1.03430557,
      "balance_loss_mlp": 1.02911067,
      "epoch": 0.5653389448369157,
      "flos": 30044052080640.0,
      "grad_norm": 1.4848761393433714,
      "language_loss": 0.77594614,
      "learning_rate": 1.6752698173045068e-06,
      "loss": 0.7969678,
      "num_input_tokens_seen": 202568835,
      "step": 9403,
      "time_per_iteration": 2.6822876930236816
    },
    {
      "auxiliary_loss_clip": 0.01063241,
      "auxiliary_loss_mlp": 0.0104845,
      "balance_loss_clip": 1.03883731,
      "balance_loss_mlp": 1.03394222,
      "epoch": 0.5653990680895836,
      "flos": 16727550128640.0,
      "grad_norm": 1.64258243294836,
      "language_loss": 0.69148064,
      "learning_rate": 1.6748855296459685e-06,
      "loss": 0.71259749,
      "num_input_tokens_seen": 202587385,
      "step": 9404,
      "time_per_iteration": 2.5738272666931152
    },
    {
      "auxiliary_loss_clip": 0.01083386,
      "auxiliary_loss_mlp": 0.01030198,
      "balance_loss_clip": 1.0379405,
      "balance_loss_mlp": 1.01867676,
      "epoch": 0.5654591913422516,
      "flos": 14538256179840.0,
      "grad_norm": 1.7369339932280703,
      "language_loss": 0.6721428,
      "learning_rate": 1.6745012543158045e-06,
      "loss": 0.69327861,
      "num_input_tokens_seen": 202604815,
      "step": 9405,
      "time_per_iteration": 2.515717029571533
    },
    {
      "auxiliary_loss_clip": 0.01081215,
      "auxiliary_loss_mlp": 0.01031419,
      "balance_loss_clip": 1.03832436,
      "balance_loss_mlp": 1.01945019,
      "epoch": 0.5655193145949196,
      "flos": 26209905603840.0,
      "grad_norm": 1.7241675671362502,
      "language_loss": 0.7436446,
      "learning_rate": 1.6741169913285852e-06,
      "loss": 0.76477093,
      "num_input_tokens_seen": 202623775,
      "step": 9406,
      "time_per_iteration": 2.5901730060577393
    },
    {
      "auxiliary_loss_clip": 0.01062365,
      "auxiliary_loss_mlp": 0.01044241,
      "balance_loss_clip": 1.03849387,
      "balance_loss_mlp": 1.02903008,
      "epoch": 0.5655794378475876,
      "flos": 25046579825280.0,
      "grad_norm": 1.7474346239927856,
      "language_loss": 0.79793251,
      "learning_rate": 1.673732740698882e-06,
      "loss": 0.81899858,
      "num_input_tokens_seen": 202643375,
      "step": 9407,
      "time_per_iteration": 2.654787063598633
    },
    {
      "auxiliary_loss_clip": 0.01073196,
      "auxiliary_loss_mlp": 0.01037575,
      "balance_loss_clip": 1.03745532,
      "balance_loss_mlp": 1.0241226,
      "epoch": 0.5656395611002555,
      "flos": 31032852652800.0,
      "grad_norm": 1.5671701454727633,
      "language_loss": 0.70797014,
      "learning_rate": 1.6733485024412666e-06,
      "loss": 0.72907782,
      "num_input_tokens_seen": 202668400,
      "step": 9408,
      "time_per_iteration": 2.706003189086914
    },
    {
      "auxiliary_loss_clip": 0.01063863,
      "auxiliary_loss_mlp": 0.0103425,
      "balance_loss_clip": 1.04368174,
      "balance_loss_mlp": 1.02167296,
      "epoch": 0.5656996843529235,
      "flos": 20229522606720.0,
      "grad_norm": 2.1246550367021406,
      "language_loss": 0.81068408,
      "learning_rate": 1.672964276570308e-06,
      "loss": 0.83166522,
      "num_input_tokens_seen": 202685125,
      "step": 9409,
      "time_per_iteration": 2.608349084854126
    },
    {
      "auxiliary_loss_clip": 0.01075747,
      "auxiliary_loss_mlp": 0.01029962,
      "balance_loss_clip": 1.03635073,
      "balance_loss_mlp": 1.0173254,
      "epoch": 0.5657598076055914,
      "flos": 20996251344000.0,
      "grad_norm": 1.5779337799422444,
      "language_loss": 0.7848115,
      "learning_rate": 1.6725800631005776e-06,
      "loss": 0.80586857,
      "num_input_tokens_seen": 202703830,
      "step": 9410,
      "time_per_iteration": 2.580579996109009
    },
    {
      "auxiliary_loss_clip": 0.01110768,
      "auxiliary_loss_mlp": 0.01033301,
      "balance_loss_clip": 1.03760624,
      "balance_loss_mlp": 1.02123713,
      "epoch": 0.5658199308582594,
      "flos": 11545999649280.0,
      "grad_norm": 2.361251989439924,
      "language_loss": 0.83583331,
      "learning_rate": 1.6721958620466432e-06,
      "loss": 0.85727394,
      "num_input_tokens_seen": 202719835,
      "step": 9411,
      "time_per_iteration": 2.4839742183685303
    },
    {
      "auxiliary_loss_clip": 0.01105789,
      "auxiliary_loss_mlp": 0.01029918,
      "balance_loss_clip": 1.0400666,
      "balance_loss_mlp": 1.01675749,
      "epoch": 0.5658800541109275,
      "flos": 14172146807040.0,
      "grad_norm": 3.0903679575153706,
      "language_loss": 0.67093891,
      "learning_rate": 1.6718116734230749e-06,
      "loss": 0.69229603,
      "num_input_tokens_seen": 202736795,
      "step": 9412,
      "time_per_iteration": 2.4792423248291016
    },
    {
      "auxiliary_loss_clip": 0.01097257,
      "auxiliary_loss_mlp": 0.010279,
      "balance_loss_clip": 1.03848267,
      "balance_loss_mlp": 1.01712906,
      "epoch": 0.5659401773635954,
      "flos": 27305073325440.0,
      "grad_norm": 1.583659610135019,
      "language_loss": 0.58127111,
      "learning_rate": 1.6714274972444413e-06,
      "loss": 0.60252273,
      "num_input_tokens_seen": 202756900,
      "step": 9413,
      "time_per_iteration": 2.5727343559265137
    },
    {
      "auxiliary_loss_clip": 0.01032565,
      "auxiliary_loss_mlp": 0.01035711,
      "balance_loss_clip": 1.03311539,
      "balance_loss_mlp": 1.02232337,
      "epoch": 0.5660003006162634,
      "flos": 16728196573440.0,
      "grad_norm": 1.724829667822143,
      "language_loss": 0.69248837,
      "learning_rate": 1.6710433335253092e-06,
      "loss": 0.71317112,
      "num_input_tokens_seen": 202775145,
      "step": 9414,
      "time_per_iteration": 2.668618679046631
    },
    {
      "auxiliary_loss_clip": 0.01039584,
      "auxiliary_loss_mlp": 0.01028347,
      "balance_loss_clip": 1.03917158,
      "balance_loss_mlp": 1.01689112,
      "epoch": 0.5660604238689313,
      "flos": 21653452535040.0,
      "grad_norm": 1.4720773893730983,
      "language_loss": 0.78094149,
      "learning_rate": 1.670659182280247e-06,
      "loss": 0.80162078,
      "num_input_tokens_seen": 202794505,
      "step": 9415,
      "time_per_iteration": 2.933361291885376
    },
    {
      "auxiliary_loss_clip": 0.01030256,
      "auxiliary_loss_mlp": 0.01002366,
      "balance_loss_clip": 1.02621818,
      "balance_loss_mlp": 1.00137091,
      "epoch": 0.5661205471215993,
      "flos": 68824022083200.0,
      "grad_norm": 0.6851654552910538,
      "language_loss": 0.49239028,
      "learning_rate": 1.670275043523822e-06,
      "loss": 0.51271647,
      "num_input_tokens_seen": 202858580,
      "step": 9416,
      "time_per_iteration": 3.2647109031677246
    },
    {
      "auxiliary_loss_clip": 0.01100318,
      "auxiliary_loss_mlp": 0.00785275,
      "balance_loss_clip": 1.03867745,
      "balance_loss_mlp": 1.00980496,
      "epoch": 0.5661806703742672,
      "flos": 28621774177920.0,
      "grad_norm": 1.781036512704545,
      "language_loss": 0.62796152,
      "learning_rate": 1.6698909172706e-06,
      "loss": 0.64681745,
      "num_input_tokens_seen": 202878565,
      "step": 9417,
      "time_per_iteration": 4.056067705154419
    },
    {
      "auxiliary_loss_clip": 0.0109187,
      "auxiliary_loss_mlp": 0.01026912,
      "balance_loss_clip": 1.03865957,
      "balance_loss_mlp": 1.0144608,
      "epoch": 0.5662407936269352,
      "flos": 21397948116480.0,
      "grad_norm": 1.816116606688781,
      "language_loss": 0.69057071,
      "learning_rate": 1.6695068035351479e-06,
      "loss": 0.71175849,
      "num_input_tokens_seen": 202897350,
      "step": 9418,
      "time_per_iteration": 2.544419050216675
    },
    {
      "auxiliary_loss_clip": 0.01098202,
      "auxiliary_loss_mlp": 0.01032025,
      "balance_loss_clip": 1.03552067,
      "balance_loss_mlp": 1.01804137,
      "epoch": 0.5663009168796032,
      "flos": 25660005315840.0,
      "grad_norm": 2.353023749423087,
      "language_loss": 0.6496402,
      "learning_rate": 1.6691227023320304e-06,
      "loss": 0.67094249,
      "num_input_tokens_seen": 202916745,
      "step": 9419,
      "time_per_iteration": 2.54282546043396
    },
    {
      "auxiliary_loss_clip": 0.00983641,
      "auxiliary_loss_mlp": 0.010054,
      "balance_loss_clip": 1.02711225,
      "balance_loss_mlp": 1.00399327,
      "epoch": 0.5663610401322712,
      "flos": 67930458422400.0,
      "grad_norm": 0.7393120236411663,
      "language_loss": 0.59719312,
      "learning_rate": 1.6687386136758135e-06,
      "loss": 0.61708355,
      "num_input_tokens_seen": 202982375,
      "step": 9420,
      "time_per_iteration": 3.463653564453125
    },
    {
      "auxiliary_loss_clip": 0.01088325,
      "auxiliary_loss_mlp": 0.00784966,
      "balance_loss_clip": 1.03516269,
      "balance_loss_mlp": 1.00989616,
      "epoch": 0.5664211633849391,
      "flos": 24609367480320.0,
      "grad_norm": 1.8085022441401146,
      "language_loss": 0.74569303,
      "learning_rate": 1.6683545375810618e-06,
      "loss": 0.76442587,
      "num_input_tokens_seen": 203002430,
      "step": 9421,
      "time_per_iteration": 2.815002679824829
    },
    {
      "auxiliary_loss_clip": 0.01078119,
      "auxiliary_loss_mlp": 0.01033662,
      "balance_loss_clip": 1.03704166,
      "balance_loss_mlp": 1.02069223,
      "epoch": 0.5664812866376071,
      "flos": 11648811352320.0,
      "grad_norm": 2.096999602224946,
      "language_loss": 0.72561389,
      "learning_rate": 1.6679704740623389e-06,
      "loss": 0.74673176,
      "num_input_tokens_seen": 203019425,
      "step": 9422,
      "time_per_iteration": 2.505920171737671
    },
    {
      "auxiliary_loss_clip": 0.01097897,
      "auxiliary_loss_mlp": 0.01035055,
      "balance_loss_clip": 1.03904164,
      "balance_loss_mlp": 1.02351534,
      "epoch": 0.566541409890275,
      "flos": 24643985212800.0,
      "grad_norm": 2.0138486613147832,
      "language_loss": 0.81824577,
      "learning_rate": 1.6675864231342085e-06,
      "loss": 0.83957529,
      "num_input_tokens_seen": 203039035,
      "step": 9423,
      "time_per_iteration": 2.5557520389556885
    },
    {
      "auxiliary_loss_clip": 0.01090302,
      "auxiliary_loss_mlp": 0.01034161,
      "balance_loss_clip": 1.03805995,
      "balance_loss_mlp": 1.02187681,
      "epoch": 0.566601533142943,
      "flos": 22270577126400.0,
      "grad_norm": 1.529964361287081,
      "language_loss": 0.80773771,
      "learning_rate": 1.6672023848112353e-06,
      "loss": 0.82898235,
      "num_input_tokens_seen": 203059320,
      "step": 9424,
      "time_per_iteration": 4.007587432861328
    },
    {
      "auxiliary_loss_clip": 0.01116572,
      "auxiliary_loss_mlp": 0.00787162,
      "balance_loss_clip": 1.04077363,
      "balance_loss_mlp": 1.01129794,
      "epoch": 0.5666616563956111,
      "flos": 29971656218880.0,
      "grad_norm": 1.8408710483586221,
      "language_loss": 0.78992331,
      "learning_rate": 1.6668183591079805e-06,
      "loss": 0.80896062,
      "num_input_tokens_seen": 203078490,
      "step": 9425,
      "time_per_iteration": 2.5645346641540527
    },
    {
      "auxiliary_loss_clip": 0.01084399,
      "auxiliary_loss_mlp": 0.01033393,
      "balance_loss_clip": 1.04023814,
      "balance_loss_mlp": 1.02083397,
      "epoch": 0.566721779648279,
      "flos": 17781456101760.0,
      "grad_norm": 1.8594613756821663,
      "language_loss": 0.59248328,
      "learning_rate": 1.6664343460390064e-06,
      "loss": 0.61366117,
      "num_input_tokens_seen": 203096065,
      "step": 9426,
      "time_per_iteration": 2.4931955337524414
    },
    {
      "auxiliary_loss_clip": 0.01104383,
      "auxiliary_loss_mlp": 0.01031612,
      "balance_loss_clip": 1.03974771,
      "balance_loss_mlp": 1.01938677,
      "epoch": 0.566781902900947,
      "flos": 21033490769280.0,
      "grad_norm": 1.7592573971758756,
      "language_loss": 0.82018334,
      "learning_rate": 1.6660503456188764e-06,
      "loss": 0.84154332,
      "num_input_tokens_seen": 203115270,
      "step": 9427,
      "time_per_iteration": 4.71933126449585
    },
    {
      "auxiliary_loss_clip": 0.01110346,
      "auxiliary_loss_mlp": 0.01032167,
      "balance_loss_clip": 1.04027009,
      "balance_loss_mlp": 1.01978755,
      "epoch": 0.5668420261536149,
      "flos": 23148593176320.0,
      "grad_norm": 2.569835256152958,
      "language_loss": 0.86509335,
      "learning_rate": 1.6656663578621498e-06,
      "loss": 0.88651848,
      "num_input_tokens_seen": 203134290,
      "step": 9428,
      "time_per_iteration": 2.5907247066497803
    },
    {
      "auxiliary_loss_clip": 0.01093258,
      "auxiliary_loss_mlp": 0.01035893,
      "balance_loss_clip": 1.03980136,
      "balance_loss_mlp": 1.02258348,
      "epoch": 0.5669021494062829,
      "flos": 22601601889920.0,
      "grad_norm": 2.1620256535281994,
      "language_loss": 0.73413855,
      "learning_rate": 1.6652823827833886e-06,
      "loss": 0.75543004,
      "num_input_tokens_seen": 203152935,
      "step": 9429,
      "time_per_iteration": 2.6047463417053223
    },
    {
      "auxiliary_loss_clip": 0.01091,
      "auxiliary_loss_mlp": 0.00786316,
      "balance_loss_clip": 1.0375818,
      "balance_loss_mlp": 1.00933361,
      "epoch": 0.5669622726589508,
      "flos": 17381231786880.0,
      "grad_norm": 1.6780703664676782,
      "language_loss": 0.7547344,
      "learning_rate": 1.6648984203971538e-06,
      "loss": 0.77350754,
      "num_input_tokens_seen": 203170110,
      "step": 9430,
      "time_per_iteration": 2.5320253372192383
    },
    {
      "auxiliary_loss_clip": 0.01112957,
      "auxiliary_loss_mlp": 0.01029904,
      "balance_loss_clip": 1.03937054,
      "balance_loss_mlp": 1.01711845,
      "epoch": 0.5670223959116188,
      "flos": 18763253521920.0,
      "grad_norm": 2.0890869452883902,
      "language_loss": 0.73027909,
      "learning_rate": 1.6645144707180032e-06,
      "loss": 0.75170767,
      "num_input_tokens_seen": 203188825,
      "step": 9431,
      "time_per_iteration": 3.854848861694336
    },
    {
      "auxiliary_loss_clip": 0.01062869,
      "auxiliary_loss_mlp": 0.01026345,
      "balance_loss_clip": 1.03701806,
      "balance_loss_mlp": 1.01512742,
      "epoch": 0.5670825191642868,
      "flos": 13553334276480.0,
      "grad_norm": 1.710588049403222,
      "language_loss": 0.73312503,
      "learning_rate": 1.6641305337604984e-06,
      "loss": 0.75401717,
      "num_input_tokens_seen": 203206860,
      "step": 9432,
      "time_per_iteration": 2.565703868865967
    },
    {
      "auxiliary_loss_clip": 0.0106692,
      "auxiliary_loss_mlp": 0.01031564,
      "balance_loss_clip": 1.03643513,
      "balance_loss_mlp": 1.01936889,
      "epoch": 0.5671426424169548,
      "flos": 22054035985920.0,
      "grad_norm": 1.504580750906571,
      "language_loss": 0.77700514,
      "learning_rate": 1.663746609539197e-06,
      "loss": 0.79798996,
      "num_input_tokens_seen": 203225625,
      "step": 9433,
      "time_per_iteration": 2.612353563308716
    },
    {
      "auxiliary_loss_clip": 0.01116609,
      "auxiliary_loss_mlp": 0.01035837,
      "balance_loss_clip": 1.03993678,
      "balance_loss_mlp": 1.02098393,
      "epoch": 0.5672027656696227,
      "flos": 21323972056320.0,
      "grad_norm": 2.2646590216597997,
      "language_loss": 0.63676155,
      "learning_rate": 1.6633626980686582e-06,
      "loss": 0.65828598,
      "num_input_tokens_seen": 203242920,
      "step": 9434,
      "time_per_iteration": 2.4881749153137207
    },
    {
      "auxiliary_loss_clip": 0.01098462,
      "auxiliary_loss_mlp": 0.0102465,
      "balance_loss_clip": 1.03732324,
      "balance_loss_mlp": 1.01260996,
      "epoch": 0.5672628889222907,
      "flos": 23514056104320.0,
      "grad_norm": 1.6937135340170024,
      "language_loss": 0.66725165,
      "learning_rate": 1.6629787993634399e-06,
      "loss": 0.68848276,
      "num_input_tokens_seen": 203261995,
      "step": 9435,
      "time_per_iteration": 2.563514471054077
    },
    {
      "auxiliary_loss_clip": 0.01085727,
      "auxiliary_loss_mlp": 0.00784651,
      "balance_loss_clip": 1.0356406,
      "balance_loss_mlp": 1.00972307,
      "epoch": 0.5673230121749586,
      "flos": 27121928855040.0,
      "grad_norm": 1.5282927868031748,
      "language_loss": 0.71763206,
      "learning_rate": 1.6625949134380984e-06,
      "loss": 0.73633587,
      "num_input_tokens_seen": 203280670,
      "step": 9436,
      "time_per_iteration": 2.5691051483154297
    },
    {
      "auxiliary_loss_clip": 0.01113266,
      "auxiliary_loss_mlp": 0.01029383,
      "balance_loss_clip": 1.03857851,
      "balance_loss_mlp": 1.01644313,
      "epoch": 0.5673831354276266,
      "flos": 31141985149440.0,
      "grad_norm": 1.5229693757185925,
      "language_loss": 0.74188656,
      "learning_rate": 1.6622110403071921e-06,
      "loss": 0.76331306,
      "num_input_tokens_seen": 203304800,
      "step": 9437,
      "time_per_iteration": 2.5707600116729736
    },
    {
      "auxiliary_loss_clip": 0.01106628,
      "auxiliary_loss_mlp": 0.01031466,
      "balance_loss_clip": 1.04255581,
      "balance_loss_mlp": 1.01792395,
      "epoch": 0.5674432586802945,
      "flos": 27673193859840.0,
      "grad_norm": 1.7392912999870531,
      "language_loss": 0.61368614,
      "learning_rate": 1.661827179985277e-06,
      "loss": 0.63506705,
      "num_input_tokens_seen": 203324060,
      "step": 9438,
      "time_per_iteration": 2.5404555797576904
    },
    {
      "auxiliary_loss_clip": 0.01091258,
      "auxiliary_loss_mlp": 0.0102871,
      "balance_loss_clip": 1.03732967,
      "balance_loss_mlp": 1.01585364,
      "epoch": 0.5675033819329626,
      "flos": 26615157822720.0,
      "grad_norm": 1.5045269495473483,
      "language_loss": 0.75023627,
      "learning_rate": 1.661443332486909e-06,
      "loss": 0.77143598,
      "num_input_tokens_seen": 203344360,
      "step": 9439,
      "time_per_iteration": 2.584634304046631
    },
    {
      "auxiliary_loss_clip": 0.01090335,
      "auxiliary_loss_mlp": 0.01029376,
      "balance_loss_clip": 1.04464197,
      "balance_loss_mlp": 1.01566684,
      "epoch": 0.5675635051856306,
      "flos": 19098372435840.0,
      "grad_norm": 1.9179427770924722,
      "language_loss": 0.84025669,
      "learning_rate": 1.6610594978266438e-06,
      "loss": 0.86145377,
      "num_input_tokens_seen": 203362115,
      "step": 9440,
      "time_per_iteration": 2.508071184158325
    },
    {
      "auxiliary_loss_clip": 0.01081371,
      "auxiliary_loss_mlp": 0.010334,
      "balance_loss_clip": 1.03790474,
      "balance_loss_mlp": 1.01960206,
      "epoch": 0.5676236284382985,
      "flos": 17566315591680.0,
      "grad_norm": 2.0220374187629355,
      "language_loss": 0.74992871,
      "learning_rate": 1.6606756760190365e-06,
      "loss": 0.77107644,
      "num_input_tokens_seen": 203380550,
      "step": 9441,
      "time_per_iteration": 2.5473110675811768
    },
    {
      "auxiliary_loss_clip": 0.01063378,
      "auxiliary_loss_mlp": 0.01037608,
      "balance_loss_clip": 1.03607392,
      "balance_loss_mlp": 1.02425611,
      "epoch": 0.5676837516909665,
      "flos": 15954069634560.0,
      "grad_norm": 1.7791149116592928,
      "language_loss": 0.83313572,
      "learning_rate": 1.6602918670786413e-06,
      "loss": 0.85414553,
      "num_input_tokens_seen": 203396590,
      "step": 9442,
      "time_per_iteration": 2.543396472930908
    },
    {
      "auxiliary_loss_clip": 0.01076146,
      "auxiliary_loss_mlp": 0.0102843,
      "balance_loss_clip": 1.03780532,
      "balance_loss_mlp": 1.01625848,
      "epoch": 0.5677438749436344,
      "flos": 18295912644480.0,
      "grad_norm": 1.827091324933523,
      "language_loss": 0.74132037,
      "learning_rate": 1.6599080710200126e-06,
      "loss": 0.76236618,
      "num_input_tokens_seen": 203414280,
      "step": 9443,
      "time_per_iteration": 2.5442094802856445
    },
    {
      "auxiliary_loss_clip": 0.01092193,
      "auxiliary_loss_mlp": 0.01030371,
      "balance_loss_clip": 1.03975224,
      "balance_loss_mlp": 1.01769304,
      "epoch": 0.5678039981963025,
      "flos": 17931311642880.0,
      "grad_norm": 2.0645873240618826,
      "language_loss": 0.77802598,
      "learning_rate": 1.6595242878577046e-06,
      "loss": 0.79925168,
      "num_input_tokens_seen": 203433280,
      "step": 9444,
      "time_per_iteration": 2.4990310668945312
    },
    {
      "auxiliary_loss_clip": 0.01074801,
      "auxiliary_loss_mlp": 0.0103886,
      "balance_loss_clip": 1.03873658,
      "balance_loss_mlp": 1.02617657,
      "epoch": 0.5678641214489704,
      "flos": 19316350120320.0,
      "grad_norm": 1.7144500893703216,
      "language_loss": 0.80599928,
      "learning_rate": 1.6591405176062687e-06,
      "loss": 0.82713592,
      "num_input_tokens_seen": 203449935,
      "step": 9445,
      "time_per_iteration": 2.556428909301758
    },
    {
      "auxiliary_loss_clip": 0.01109444,
      "auxiliary_loss_mlp": 0.01027659,
      "balance_loss_clip": 1.03605199,
      "balance_loss_mlp": 1.01514792,
      "epoch": 0.5679242447016384,
      "flos": 27751084502400.0,
      "grad_norm": 1.7663567069093988,
      "language_loss": 0.71003497,
      "learning_rate": 1.658756760280259e-06,
      "loss": 0.73140597,
      "num_input_tokens_seen": 203473025,
      "step": 9446,
      "time_per_iteration": 2.5397725105285645
    },
    {
      "auxiliary_loss_clip": 0.0107626,
      "auxiliary_loss_mlp": 0.01031974,
      "balance_loss_clip": 1.03566146,
      "balance_loss_mlp": 1.01879573,
      "epoch": 0.5679843679543063,
      "flos": 23769093646080.0,
      "grad_norm": 2.467073147527505,
      "language_loss": 0.73228145,
      "learning_rate": 1.6583730158942276e-06,
      "loss": 0.75336379,
      "num_input_tokens_seen": 203492895,
      "step": 9447,
      "time_per_iteration": 2.5875515937805176
    },
    {
      "auxiliary_loss_clip": 0.01090615,
      "auxiliary_loss_mlp": 0.01033431,
      "balance_loss_clip": 1.03765309,
      "balance_loss_mlp": 1.02071118,
      "epoch": 0.5680444912069743,
      "flos": 25591883172480.0,
      "grad_norm": 1.9268829056879138,
      "language_loss": 0.74979532,
      "learning_rate": 1.657989284462725e-06,
      "loss": 0.77103579,
      "num_input_tokens_seen": 203513710,
      "step": 9448,
      "time_per_iteration": 2.565929889678955
    },
    {
      "auxiliary_loss_clip": 0.01070782,
      "auxiliary_loss_mlp": 0.01044614,
      "balance_loss_clip": 1.03884685,
      "balance_loss_mlp": 1.02953434,
      "epoch": 0.5681046144596422,
      "flos": 23695799944320.0,
      "grad_norm": 2.486389283656631,
      "language_loss": 0.760584,
      "learning_rate": 1.6576055660003038e-06,
      "loss": 0.78173792,
      "num_input_tokens_seen": 203531630,
      "step": 9449,
      "time_per_iteration": 2.626469135284424
    },
    {
      "auxiliary_loss_clip": 0.0108577,
      "auxiliary_loss_mlp": 0.01039751,
      "balance_loss_clip": 1.03516948,
      "balance_loss_mlp": 1.02529049,
      "epoch": 0.5681647377123102,
      "flos": 28000770917760.0,
      "grad_norm": 1.632940063822298,
      "language_loss": 0.75036335,
      "learning_rate": 1.6572218605215128e-06,
      "loss": 0.77161849,
      "num_input_tokens_seen": 203551885,
      "step": 9450,
      "time_per_iteration": 2.5862135887145996
    },
    {
      "auxiliary_loss_clip": 0.01095114,
      "auxiliary_loss_mlp": 0.01034721,
      "balance_loss_clip": 1.03866935,
      "balance_loss_mlp": 1.02158999,
      "epoch": 0.5682248609649782,
      "flos": 22747758330240.0,
      "grad_norm": 1.7075379180463586,
      "language_loss": 0.66782826,
      "learning_rate": 1.6568381680409038e-06,
      "loss": 0.68912661,
      "num_input_tokens_seen": 203572250,
      "step": 9451,
      "time_per_iteration": 2.552676200866699
    },
    {
      "auxiliary_loss_clip": 0.01091081,
      "auxiliary_loss_mlp": 0.01032935,
      "balance_loss_clip": 1.03900123,
      "balance_loss_mlp": 1.01767564,
      "epoch": 0.5682849842176462,
      "flos": 21288600138240.0,
      "grad_norm": 2.591651548254325,
      "language_loss": 0.71823847,
      "learning_rate": 1.656454488573026e-06,
      "loss": 0.73947865,
      "num_input_tokens_seen": 203590605,
      "step": 9452,
      "time_per_iteration": 2.5411341190338135
    },
    {
      "auxiliary_loss_clip": 0.0106842,
      "auxiliary_loss_mlp": 0.01031153,
      "balance_loss_clip": 1.03632891,
      "balance_loss_mlp": 1.01897609,
      "epoch": 0.5683451074703142,
      "flos": 21141689512320.0,
      "grad_norm": 1.408418980924042,
      "language_loss": 0.70049727,
      "learning_rate": 1.656070822132428e-06,
      "loss": 0.72149301,
      "num_input_tokens_seen": 203610080,
      "step": 9453,
      "time_per_iteration": 2.568737506866455
    },
    {
      "auxiliary_loss_clip": 0.01070748,
      "auxiliary_loss_mlp": 0.00784549,
      "balance_loss_clip": 1.0384481,
      "balance_loss_mlp": 1.01025033,
      "epoch": 0.5684052307229821,
      "flos": 22344481359360.0,
      "grad_norm": 1.6873294828023975,
      "language_loss": 0.70155478,
      "learning_rate": 1.6556871687336592e-06,
      "loss": 0.72010773,
      "num_input_tokens_seen": 203630060,
      "step": 9454,
      "time_per_iteration": 2.6179285049438477
    },
    {
      "auxiliary_loss_clip": 0.01087093,
      "auxiliary_loss_mlp": 0.01031171,
      "balance_loss_clip": 1.03575301,
      "balance_loss_mlp": 1.01926827,
      "epoch": 0.5684653539756501,
      "flos": 21798639308160.0,
      "grad_norm": 2.127529145871682,
      "language_loss": 0.60131872,
      "learning_rate": 1.6553035283912671e-06,
      "loss": 0.62250137,
      "num_input_tokens_seen": 203649065,
      "step": 9455,
      "time_per_iteration": 2.5475995540618896
    },
    {
      "auxiliary_loss_clip": 0.01072543,
      "auxiliary_loss_mlp": 0.01029429,
      "balance_loss_clip": 1.04012346,
      "balance_loss_mlp": 1.01646531,
      "epoch": 0.568525477228318,
      "flos": 22999635475200.0,
      "grad_norm": 1.8054933074336006,
      "language_loss": 0.73351681,
      "learning_rate": 1.6549199011198e-06,
      "loss": 0.75453651,
      "num_input_tokens_seen": 203667545,
      "step": 9456,
      "time_per_iteration": 3.944758892059326
    },
    {
      "auxiliary_loss_clip": 0.01090795,
      "auxiliary_loss_mlp": 0.01030062,
      "balance_loss_clip": 1.03727818,
      "balance_loss_mlp": 1.0184269,
      "epoch": 0.568585600480986,
      "flos": 21392489249280.0,
      "grad_norm": 1.5783313473980736,
      "language_loss": 0.7643187,
      "learning_rate": 1.6545362869338048e-06,
      "loss": 0.78552735,
      "num_input_tokens_seen": 203686025,
      "step": 9457,
      "time_per_iteration": 2.5339467525482178
    },
    {
      "auxiliary_loss_clip": 0.01101802,
      "auxiliary_loss_mlp": 0.01033221,
      "balance_loss_clip": 1.037799,
      "balance_loss_mlp": 1.01950622,
      "epoch": 0.568645723733654,
      "flos": 30007351359360.0,
      "grad_norm": 1.7771198828565364,
      "language_loss": 0.66468215,
      "learning_rate": 1.6541526858478285e-06,
      "loss": 0.6860323,
      "num_input_tokens_seen": 203705540,
      "step": 9458,
      "time_per_iteration": 2.5632290840148926
    },
    {
      "auxiliary_loss_clip": 0.01101051,
      "auxiliary_loss_mlp": 0.01027337,
      "balance_loss_clip": 1.03715098,
      "balance_loss_mlp": 1.01404488,
      "epoch": 0.568705846986322,
      "flos": 20412667077120.0,
      "grad_norm": 2.090457539303114,
      "language_loss": 0.68473279,
      "learning_rate": 1.6537690978764167e-06,
      "loss": 0.70601666,
      "num_input_tokens_seen": 203723670,
      "step": 9459,
      "time_per_iteration": 2.5082497596740723
    },
    {
      "auxiliary_loss_clip": 0.01084902,
      "auxiliary_loss_mlp": 0.01031621,
      "balance_loss_clip": 1.03917813,
      "balance_loss_mlp": 1.01847839,
      "epoch": 0.5687659702389899,
      "flos": 17456752131840.0,
      "grad_norm": 2.1690679099230734,
      "language_loss": 0.76732177,
      "learning_rate": 1.6533855230341155e-06,
      "loss": 0.78848696,
      "num_input_tokens_seen": 203739705,
      "step": 9460,
      "time_per_iteration": 2.5181405544281006
    },
    {
      "auxiliary_loss_clip": 0.01055513,
      "auxiliary_loss_mlp": 0.01031193,
      "balance_loss_clip": 1.03686357,
      "balance_loss_mlp": 1.01860452,
      "epoch": 0.5688260934916579,
      "flos": 25406081095680.0,
      "grad_norm": 1.673635462474603,
      "language_loss": 0.71674144,
      "learning_rate": 1.65300196133547e-06,
      "loss": 0.73760849,
      "num_input_tokens_seen": 203759000,
      "step": 9461,
      "time_per_iteration": 2.676396608352661
    },
    {
      "auxiliary_loss_clip": 0.01100836,
      "auxiliary_loss_mlp": 0.01030499,
      "balance_loss_clip": 1.03674638,
      "balance_loss_mlp": 1.01722503,
      "epoch": 0.5688862167443258,
      "flos": 21608024808960.0,
      "grad_norm": 2.1587466833064477,
      "language_loss": 0.73147863,
      "learning_rate": 1.6526184127950249e-06,
      "loss": 0.752792,
      "num_input_tokens_seen": 203774295,
      "step": 9462,
      "time_per_iteration": 2.464184284210205
    },
    {
      "auxiliary_loss_clip": 0.01098623,
      "auxiliary_loss_mlp": 0.01027506,
      "balance_loss_clip": 1.03695846,
      "balance_loss_mlp": 1.0157994,
      "epoch": 0.5689463399969938,
      "flos": 22418996123520.0,
      "grad_norm": 2.0342704236367255,
      "language_loss": 0.72991014,
      "learning_rate": 1.6522348774273246e-06,
      "loss": 0.75117147,
      "num_input_tokens_seen": 203792710,
      "step": 9463,
      "time_per_iteration": 3.860083818435669
    },
    {
      "auxiliary_loss_clip": 0.01099876,
      "auxiliary_loss_mlp": 0.01032294,
      "balance_loss_clip": 1.03736341,
      "balance_loss_mlp": 1.01937807,
      "epoch": 0.5690064632496618,
      "flos": 18296810484480.0,
      "grad_norm": 1.7213839344008175,
      "language_loss": 0.7408691,
      "learning_rate": 1.6518513552469123e-06,
      "loss": 0.76219082,
      "num_input_tokens_seen": 203811645,
      "step": 9464,
      "time_per_iteration": 2.469008207321167
    },
    {
      "auxiliary_loss_clip": 0.01103381,
      "auxiliary_loss_mlp": 0.00786281,
      "balance_loss_clip": 1.03821933,
      "balance_loss_mlp": 1.00935578,
      "epoch": 0.5690665865023298,
      "flos": 21579260993280.0,
      "grad_norm": 1.6464597927038038,
      "language_loss": 0.84073055,
      "learning_rate": 1.6514678462683312e-06,
      "loss": 0.85962713,
      "num_input_tokens_seen": 203830040,
      "step": 9465,
      "time_per_iteration": 2.521355390548706
    },
    {
      "auxiliary_loss_clip": 0.0108818,
      "auxiliary_loss_mlp": 0.01029175,
      "balance_loss_clip": 1.03558838,
      "balance_loss_mlp": 1.01678944,
      "epoch": 0.5691267097549978,
      "flos": 24421446501120.0,
      "grad_norm": 1.61043418137058,
      "language_loss": 0.72300458,
      "learning_rate": 1.651084350506125e-06,
      "loss": 0.74417818,
      "num_input_tokens_seen": 203851245,
      "step": 9466,
      "time_per_iteration": 3.9728055000305176
    },
    {
      "auxiliary_loss_clip": 0.01020244,
      "auxiliary_loss_mlp": 0.01007428,
      "balance_loss_clip": 1.02555883,
      "balance_loss_mlp": 1.0062356,
      "epoch": 0.5691868330076657,
      "flos": 61657906199040.0,
      "grad_norm": 0.7186943324151508,
      "language_loss": 0.55366957,
      "learning_rate": 1.6507008679748343e-06,
      "loss": 0.57394636,
      "num_input_tokens_seen": 203916400,
      "step": 9467,
      "time_per_iteration": 3.2471742630004883
    },
    {
      "auxiliary_loss_clip": 0.01096585,
      "auxiliary_loss_mlp": 0.01033974,
      "balance_loss_clip": 1.03627002,
      "balance_loss_mlp": 1.01959753,
      "epoch": 0.5692469562603337,
      "flos": 21325193118720.0,
      "grad_norm": 2.083688115989741,
      "language_loss": 0.63707232,
      "learning_rate": 1.6503173986890023e-06,
      "loss": 0.65837789,
      "num_input_tokens_seen": 203935870,
      "step": 9468,
      "time_per_iteration": 2.5051703453063965
    },
    {
      "auxiliary_loss_clip": 0.0106662,
      "auxiliary_loss_mlp": 0.01034962,
      "balance_loss_clip": 1.04002345,
      "balance_loss_mlp": 1.0203706,
      "epoch": 0.5693070795130016,
      "flos": 23367899664000.0,
      "grad_norm": 1.7097172990780025,
      "language_loss": 0.79330862,
      "learning_rate": 1.64993394266317e-06,
      "loss": 0.81432438,
      "num_input_tokens_seen": 203954950,
      "step": 9469,
      "time_per_iteration": 2.6063969135284424
    },
    {
      "auxiliary_loss_clip": 0.01074256,
      "auxiliary_loss_mlp": 0.01041037,
      "balance_loss_clip": 1.03514552,
      "balance_loss_mlp": 1.02636826,
      "epoch": 0.5693672027656697,
      "flos": 18697250280960.0,
      "grad_norm": 2.0801613930733276,
      "language_loss": 0.69057661,
      "learning_rate": 1.6495504999118769e-06,
      "loss": 0.71172953,
      "num_input_tokens_seen": 203972715,
      "step": 9470,
      "time_per_iteration": 3.8974034786224365
    },
    {
      "auxiliary_loss_clip": 0.01087557,
      "auxiliary_loss_mlp": 0.01033091,
      "balance_loss_clip": 1.04060674,
      "balance_loss_mlp": 1.01995432,
      "epoch": 0.5694273260183376,
      "flos": 20449188230400.0,
      "grad_norm": 1.6127834131511065,
      "language_loss": 0.74745941,
      "learning_rate": 1.6491670704496644e-06,
      "loss": 0.76866591,
      "num_input_tokens_seen": 203990775,
      "step": 9471,
      "time_per_iteration": 2.530219078063965
    },
    {
      "auxiliary_loss_clip": 0.01069863,
      "auxiliary_loss_mlp": 0.01035513,
      "balance_loss_clip": 1.03857756,
      "balance_loss_mlp": 1.02144027,
      "epoch": 0.5694874492710056,
      "flos": 17603195880960.0,
      "grad_norm": 1.8550542983181562,
      "language_loss": 0.57301068,
      "learning_rate": 1.6487836542910716e-06,
      "loss": 0.59406447,
      "num_input_tokens_seen": 204008845,
      "step": 9472,
      "time_per_iteration": 2.570781946182251
    },
    {
      "auxiliary_loss_clip": 0.01074903,
      "auxiliary_loss_mlp": 0.01027963,
      "balance_loss_clip": 1.03609252,
      "balance_loss_mlp": 1.01506495,
      "epoch": 0.5695475725236735,
      "flos": 13370836250880.0,
      "grad_norm": 1.796439276664439,
      "language_loss": 0.73667431,
      "learning_rate": 1.648400251450638e-06,
      "loss": 0.75770295,
      "num_input_tokens_seen": 204023755,
      "step": 9473,
      "time_per_iteration": 2.5423378944396973
    },
    {
      "auxiliary_loss_clip": 0.01020307,
      "auxiliary_loss_mlp": 0.01005212,
      "balance_loss_clip": 1.02489138,
      "balance_loss_mlp": 1.00396013,
      "epoch": 0.5696076957763415,
      "flos": 68174398661760.0,
      "grad_norm": 0.6567724433936057,
      "language_loss": 0.57614493,
      "learning_rate": 1.6480168619429023e-06,
      "loss": 0.59640014,
      "num_input_tokens_seen": 204091255,
      "step": 9474,
      "time_per_iteration": 3.2049617767333984
    },
    {
      "auxiliary_loss_clip": 0.01098285,
      "auxiliary_loss_mlp": 0.01029717,
      "balance_loss_clip": 1.03776395,
      "balance_loss_mlp": 1.01587653,
      "epoch": 0.5696678190290094,
      "flos": 33838301525760.0,
      "grad_norm": 1.7969436915155912,
      "language_loss": 0.53792381,
      "learning_rate": 1.6476334857824017e-06,
      "loss": 0.5592038,
      "num_input_tokens_seen": 204113285,
      "step": 9475,
      "time_per_iteration": 2.6284215450286865
    },
    {
      "auxiliary_loss_clip": 0.01113868,
      "auxiliary_loss_mlp": 0.01035213,
      "balance_loss_clip": 1.03971708,
      "balance_loss_mlp": 1.02190375,
      "epoch": 0.5697279422816774,
      "flos": 26356600748160.0,
      "grad_norm": 1.5480221995339087,
      "language_loss": 0.79586828,
      "learning_rate": 1.647250122983675e-06,
      "loss": 0.81735909,
      "num_input_tokens_seen": 204133045,
      "step": 9476,
      "time_per_iteration": 2.482653856277466
    },
    {
      "auxiliary_loss_clip": 0.01092047,
      "auxiliary_loss_mlp": 0.01033411,
      "balance_loss_clip": 1.04360485,
      "balance_loss_mlp": 1.02069688,
      "epoch": 0.5697880655343454,
      "flos": 22930507751040.0,
      "grad_norm": 2.3836782465357693,
      "language_loss": 0.66246665,
      "learning_rate": 1.6468667735612592e-06,
      "loss": 0.68372118,
      "num_input_tokens_seen": 204152590,
      "step": 9477,
      "time_per_iteration": 2.5630757808685303
    },
    {
      "auxiliary_loss_clip": 0.01078296,
      "auxiliary_loss_mlp": 0.01029073,
      "balance_loss_clip": 1.03743505,
      "balance_loss_mlp": 1.01606703,
      "epoch": 0.5698481887870134,
      "flos": 26761314263040.0,
      "grad_norm": 1.714710410538919,
      "language_loss": 0.70881665,
      "learning_rate": 1.6464834375296906e-06,
      "loss": 0.72989029,
      "num_input_tokens_seen": 204171815,
      "step": 9478,
      "time_per_iteration": 2.5945138931274414
    },
    {
      "auxiliary_loss_clip": 0.01079669,
      "auxiliary_loss_mlp": 0.01024578,
      "balance_loss_clip": 1.03807688,
      "balance_loss_mlp": 1.01318753,
      "epoch": 0.5699083120396814,
      "flos": 15742269089280.0,
      "grad_norm": 1.566459101634958,
      "language_loss": 0.69044161,
      "learning_rate": 1.6461001149035055e-06,
      "loss": 0.71148407,
      "num_input_tokens_seen": 204188535,
      "step": 9479,
      "time_per_iteration": 2.5524508953094482
    },
    {
      "auxiliary_loss_clip": 0.01069883,
      "auxiliary_loss_mlp": 0.01029678,
      "balance_loss_clip": 1.03769422,
      "balance_loss_mlp": 1.01777482,
      "epoch": 0.5699684352923493,
      "flos": 19537272720000.0,
      "grad_norm": 1.385054220675366,
      "language_loss": 0.71479565,
      "learning_rate": 1.6457168056972392e-06,
      "loss": 0.73579121,
      "num_input_tokens_seen": 204208365,
      "step": 9480,
      "time_per_iteration": 2.5460689067840576
    },
    {
      "auxiliary_loss_clip": 0.01082401,
      "auxiliary_loss_mlp": 0.00785777,
      "balance_loss_clip": 1.03889263,
      "balance_loss_mlp": 1.00795007,
      "epoch": 0.5700285585450173,
      "flos": 16253349753600.0,
      "grad_norm": 4.027757456211928,
      "language_loss": 0.72264338,
      "learning_rate": 1.6453335099254276e-06,
      "loss": 0.74132514,
      "num_input_tokens_seen": 204226560,
      "step": 9481,
      "time_per_iteration": 2.559252977371216
    },
    {
      "auxiliary_loss_clip": 0.01102826,
      "auxiliary_loss_mlp": 0.01035159,
      "balance_loss_clip": 1.0395689,
      "balance_loss_mlp": 1.02197456,
      "epoch": 0.5700886817976852,
      "flos": 19864993432320.0,
      "grad_norm": 1.6869627984187903,
      "language_loss": 0.78536201,
      "learning_rate": 1.6449502276026041e-06,
      "loss": 0.80674189,
      "num_input_tokens_seen": 204245410,
      "step": 9482,
      "time_per_iteration": 2.4823200702667236
    },
    {
      "auxiliary_loss_clip": 0.01087469,
      "auxiliary_loss_mlp": 0.01028214,
      "balance_loss_clip": 1.03766799,
      "balance_loss_mlp": 1.01589954,
      "epoch": 0.5701488050503533,
      "flos": 23841704989440.0,
      "grad_norm": 1.5037004707024952,
      "language_loss": 0.77988756,
      "learning_rate": 1.6445669587433043e-06,
      "loss": 0.8010444,
      "num_input_tokens_seen": 204264840,
      "step": 9483,
      "time_per_iteration": 2.5463674068450928
    },
    {
      "auxiliary_loss_clip": 0.01088439,
      "auxiliary_loss_mlp": 0.01035682,
      "balance_loss_clip": 1.03827941,
      "balance_loss_mlp": 1.023278,
      "epoch": 0.5702089283030212,
      "flos": 23659673840640.0,
      "grad_norm": 1.6172058110188359,
      "language_loss": 0.80973971,
      "learning_rate": 1.6441837033620612e-06,
      "loss": 0.8309809,
      "num_input_tokens_seen": 204284335,
      "step": 9484,
      "time_per_iteration": 2.5354485511779785
    },
    {
      "auxiliary_loss_clip": 0.01110895,
      "auxiliary_loss_mlp": 0.00786989,
      "balance_loss_clip": 1.03757882,
      "balance_loss_mlp": 1.0116365,
      "epoch": 0.5702690515556892,
      "flos": 27891171544320.0,
      "grad_norm": 2.8551639768353785,
      "language_loss": 0.60914063,
      "learning_rate": 1.6438004614734073e-06,
      "loss": 0.62811947,
      "num_input_tokens_seen": 204302590,
      "step": 9485,
      "time_per_iteration": 2.5341286659240723
    },
    {
      "auxiliary_loss_clip": 0.01099886,
      "auxiliary_loss_mlp": 0.01031667,
      "balance_loss_clip": 1.0382452,
      "balance_loss_mlp": 1.01859605,
      "epoch": 0.5703291748083571,
      "flos": 24023951619840.0,
      "grad_norm": 2.1961525566407714,
      "language_loss": 0.65175825,
      "learning_rate": 1.6434172330918757e-06,
      "loss": 0.67307377,
      "num_input_tokens_seen": 204323055,
      "step": 9486,
      "time_per_iteration": 2.505262851715088
    },
    {
      "auxiliary_loss_clip": 0.01029934,
      "auxiliary_loss_mlp": 0.01020796,
      "balance_loss_clip": 1.02298546,
      "balance_loss_mlp": 1.01963377,
      "epoch": 0.5703892980610251,
      "flos": 57023382919680.0,
      "grad_norm": 0.6696989982409408,
      "language_loss": 0.47989193,
      "learning_rate": 1.6430340182319978e-06,
      "loss": 0.50039923,
      "num_input_tokens_seen": 204386160,
      "step": 9487,
      "time_per_iteration": 3.252068519592285
    },
    {
      "auxiliary_loss_clip": 0.01076471,
      "auxiliary_loss_mlp": 0.00785945,
      "balance_loss_clip": 1.0373621,
      "balance_loss_mlp": 1.01003265,
      "epoch": 0.570449421313693,
      "flos": 24351025887360.0,
      "grad_norm": 1.7061662664711827,
      "language_loss": 0.8588841,
      "learning_rate": 1.6426508169083067e-06,
      "loss": 0.87750828,
      "num_input_tokens_seen": 204406315,
      "step": 9488,
      "time_per_iteration": 2.575137138366699
    },
    {
      "auxiliary_loss_clip": 0.01077628,
      "auxiliary_loss_mlp": 0.01034673,
      "balance_loss_clip": 1.0379169,
      "balance_loss_mlp": 1.02144718,
      "epoch": 0.570509544566361,
      "flos": 24828566227200.0,
      "grad_norm": 9.053896645043254,
      "language_loss": 0.79399848,
      "learning_rate": 1.6422676291353314e-06,
      "loss": 0.81512153,
      "num_input_tokens_seen": 204427645,
      "step": 9489,
      "time_per_iteration": 2.613002061843872
    },
    {
      "auxiliary_loss_clip": 0.01090082,
      "auxiliary_loss_mlp": 0.01031092,
      "balance_loss_clip": 1.03820372,
      "balance_loss_mlp": 1.01949286,
      "epoch": 0.570569667819029,
      "flos": 21397301671680.0,
      "grad_norm": 1.800161456174711,
      "language_loss": 0.7021488,
      "learning_rate": 1.641884454927604e-06,
      "loss": 0.72336054,
      "num_input_tokens_seen": 204445910,
      "step": 9490,
      "time_per_iteration": 2.5316083431243896
    },
    {
      "auxiliary_loss_clip": 0.01078642,
      "auxiliary_loss_mlp": 0.01033057,
      "balance_loss_clip": 1.0384903,
      "balance_loss_mlp": 1.02069473,
      "epoch": 0.570629791071697,
      "flos": 23216751233280.0,
      "grad_norm": 1.5502703082637892,
      "language_loss": 0.76322162,
      "learning_rate": 1.6415012942996548e-06,
      "loss": 0.78433859,
      "num_input_tokens_seen": 204464680,
      "step": 9491,
      "time_per_iteration": 2.5991523265838623
    },
    {
      "auxiliary_loss_clip": 0.01016325,
      "auxiliary_loss_mlp": 0.00798094,
      "balance_loss_clip": 1.0230701,
      "balance_loss_mlp": 1.06960046,
      "epoch": 0.570689914324365,
      "flos": 65284666525440.0,
      "grad_norm": 0.8034113437131128,
      "language_loss": 0.5734486,
      "learning_rate": 1.641118147266011e-06,
      "loss": 0.59159279,
      "num_input_tokens_seen": 204525580,
      "step": 9492,
      "time_per_iteration": 3.1516993045806885
    },
    {
      "auxiliary_loss_clip": 0.01086539,
      "auxiliary_loss_mlp": 0.00786372,
      "balance_loss_clip": 1.04094994,
      "balance_loss_mlp": 1.012802,
      "epoch": 0.5707500375770329,
      "flos": 21141904993920.0,
      "grad_norm": 1.8700851173636142,
      "language_loss": 0.71845615,
      "learning_rate": 1.6407350138412035e-06,
      "loss": 0.73718536,
      "num_input_tokens_seen": 204541320,
      "step": 9493,
      "time_per_iteration": 2.5684752464294434
    },
    {
      "auxiliary_loss_clip": 0.01115669,
      "auxiliary_loss_mlp": 0.01031945,
      "balance_loss_clip": 1.04044378,
      "balance_loss_mlp": 1.01914227,
      "epoch": 0.5708101608297009,
      "flos": 20812747737600.0,
      "grad_norm": 1.7727857290087572,
      "language_loss": 0.78454,
      "learning_rate": 1.6403518940397606e-06,
      "loss": 0.80601609,
      "num_input_tokens_seen": 204560275,
      "step": 9494,
      "time_per_iteration": 3.8410916328430176
    },
    {
      "auxiliary_loss_clip": 0.011148,
      "auxiliary_loss_mlp": 0.01038202,
      "balance_loss_clip": 1.03880918,
      "balance_loss_mlp": 1.02446353,
      "epoch": 0.5708702840823688,
      "flos": 25812338895360.0,
      "grad_norm": 2.579301509376585,
      "language_loss": 0.80801511,
      "learning_rate": 1.6399687878762096e-06,
      "loss": 0.82954514,
      "num_input_tokens_seen": 204579430,
      "step": 9495,
      "time_per_iteration": 2.513277053833008
    },
    {
      "auxiliary_loss_clip": 0.01079157,
      "auxiliary_loss_mlp": 0.0104062,
      "balance_loss_clip": 1.038077,
      "balance_loss_mlp": 1.02490783,
      "epoch": 0.5709304073350369,
      "flos": 23651916503040.0,
      "grad_norm": 2.14819061834764,
      "language_loss": 0.65971398,
      "learning_rate": 1.6395856953650784e-06,
      "loss": 0.68091172,
      "num_input_tokens_seen": 204597710,
      "step": 9496,
      "time_per_iteration": 2.5825417041778564
    },
    {
      "auxiliary_loss_clip": 0.01115401,
      "auxiliary_loss_mlp": 0.01037288,
      "balance_loss_clip": 1.04022908,
      "balance_loss_mlp": 1.0235852,
      "epoch": 0.5709905305877048,
      "flos": 16107552449280.0,
      "grad_norm": 2.1603963563875297,
      "language_loss": 0.69408298,
      "learning_rate": 1.6392026165208938e-06,
      "loss": 0.71560991,
      "num_input_tokens_seen": 204616140,
      "step": 9497,
      "time_per_iteration": 2.456901788711548
    },
    {
      "auxiliary_loss_clip": 0.01097515,
      "auxiliary_loss_mlp": 0.00797183,
      "balance_loss_clip": 1.03938162,
      "balance_loss_mlp": 1.02839005,
      "epoch": 0.5710506538403728,
      "flos": 24750819239040.0,
      "grad_norm": 1.856306117199049,
      "language_loss": 0.8126272,
      "learning_rate": 1.638819551358182e-06,
      "loss": 0.83157414,
      "num_input_tokens_seen": 204636470,
      "step": 9498,
      "time_per_iteration": 2.519699811935425
    },
    {
      "auxiliary_loss_clip": 0.01114113,
      "auxiliary_loss_mlp": 0.0103438,
      "balance_loss_clip": 1.04001558,
      "balance_loss_mlp": 1.02022386,
      "epoch": 0.5711107770930407,
      "flos": 21982250655360.0,
      "grad_norm": 3.0399086489373386,
      "language_loss": 0.6640833,
      "learning_rate": 1.638436499891469e-06,
      "loss": 0.68556827,
      "num_input_tokens_seen": 204656640,
      "step": 9499,
      "time_per_iteration": 2.4740893840789795
    },
    {
      "auxiliary_loss_clip": 0.01083928,
      "auxiliary_loss_mlp": 0.01035625,
      "balance_loss_clip": 1.04174042,
      "balance_loss_mlp": 1.02272081,
      "epoch": 0.5711709003457087,
      "flos": 19574009354880.0,
      "grad_norm": 1.6450544273909862,
      "language_loss": 0.71376085,
      "learning_rate": 1.6380534621352805e-06,
      "loss": 0.73495638,
      "num_input_tokens_seen": 204675475,
      "step": 9500,
      "time_per_iteration": 2.5214314460754395
    },
    {
      "auxiliary_loss_clip": 0.01086509,
      "auxiliary_loss_mlp": 0.01036318,
      "balance_loss_clip": 1.03938651,
      "balance_loss_mlp": 1.02260911,
      "epoch": 0.5712310235983766,
      "flos": 24242683489920.0,
      "grad_norm": 1.7203672939306964,
      "language_loss": 0.7567029,
      "learning_rate": 1.6376704381041407e-06,
      "loss": 0.77793121,
      "num_input_tokens_seen": 204695385,
      "step": 9501,
      "time_per_iteration": 3.9852135181427
    },
    {
      "auxiliary_loss_clip": 0.01093509,
      "auxiliary_loss_mlp": 0.01034624,
      "balance_loss_clip": 1.03907716,
      "balance_loss_mlp": 1.02219605,
      "epoch": 0.5712911468510447,
      "flos": 20996143603200.0,
      "grad_norm": 1.6642240527826175,
      "language_loss": 0.7493223,
      "learning_rate": 1.6372874278125742e-06,
      "loss": 0.77060366,
      "num_input_tokens_seen": 204714730,
      "step": 9502,
      "time_per_iteration": 2.512223720550537
    },
    {
      "auxiliary_loss_clip": 0.01086754,
      "auxiliary_loss_mlp": 0.01026872,
      "balance_loss_clip": 1.04020965,
      "balance_loss_mlp": 1.01454008,
      "epoch": 0.5713512701037126,
      "flos": 18916987731840.0,
      "grad_norm": 1.818740889834848,
      "language_loss": 0.82179761,
      "learning_rate": 1.636904431275105e-06,
      "loss": 0.84293383,
      "num_input_tokens_seen": 204735025,
      "step": 9503,
      "time_per_iteration": 2.5661613941192627
    },
    {
      "auxiliary_loss_clip": 0.01077436,
      "auxiliary_loss_mlp": 0.01031894,
      "balance_loss_clip": 1.03812718,
      "balance_loss_mlp": 1.02017009,
      "epoch": 0.5714113933563806,
      "flos": 17413443308160.0,
      "grad_norm": 2.209171273062203,
      "language_loss": 0.85906088,
      "learning_rate": 1.6365214485062553e-06,
      "loss": 0.88015425,
      "num_input_tokens_seen": 204751365,
      "step": 9504,
      "time_per_iteration": 3.9453797340393066
    },
    {
      "auxiliary_loss_clip": 0.0107023,
      "auxiliary_loss_mlp": 0.01032079,
      "balance_loss_clip": 1.04096985,
      "balance_loss_mlp": 1.01927614,
      "epoch": 0.5714715166090486,
      "flos": 20193360589440.0,
      "grad_norm": 1.7724107493948658,
      "language_loss": 0.75532007,
      "learning_rate": 1.6361384795205496e-06,
      "loss": 0.77634317,
      "num_input_tokens_seen": 204768980,
      "step": 9505,
      "time_per_iteration": 2.5709176063537598
    },
    {
      "auxiliary_loss_clip": 0.0111023,
      "auxiliary_loss_mlp": 0.01032708,
      "balance_loss_clip": 1.03836262,
      "balance_loss_mlp": 1.02120447,
      "epoch": 0.5715316398617165,
      "flos": 18551668458240.0,
      "grad_norm": 1.4318363449775744,
      "language_loss": 0.81548148,
      "learning_rate": 1.635755524332509e-06,
      "loss": 0.83691084,
      "num_input_tokens_seen": 204788110,
      "step": 9506,
      "time_per_iteration": 2.4674127101898193
    },
    {
      "auxiliary_loss_clip": 0.01072317,
      "auxiliary_loss_mlp": 0.00802353,
      "balance_loss_clip": 1.03636277,
      "balance_loss_mlp": 1.04051137,
      "epoch": 0.5715917631143845,
      "flos": 18478195188480.0,
      "grad_norm": 1.896509368230024,
      "language_loss": 0.77432179,
      "learning_rate": 1.6353725829566552e-06,
      "loss": 0.79306853,
      "num_input_tokens_seen": 204807240,
      "step": 9507,
      "time_per_iteration": 2.5694875717163086
    },
    {
      "auxiliary_loss_clip": 0.01092036,
      "auxiliary_loss_mlp": 0.01036311,
      "balance_loss_clip": 1.03982234,
      "balance_loss_mlp": 1.02262592,
      "epoch": 0.5716518863670524,
      "flos": 24020037037440.0,
      "grad_norm": 1.4251877427544266,
      "language_loss": 0.68302393,
      "learning_rate": 1.63498965540751e-06,
      "loss": 0.70430744,
      "num_input_tokens_seen": 204826415,
      "step": 9508,
      "time_per_iteration": 2.550493001937866
    },
    {
      "auxiliary_loss_clip": 0.01111714,
      "auxiliary_loss_mlp": 0.01030543,
      "balance_loss_clip": 1.03777885,
      "balance_loss_mlp": 1.01777518,
      "epoch": 0.5717120096197205,
      "flos": 17819485626240.0,
      "grad_norm": 2.013557444144772,
      "language_loss": 0.79321647,
      "learning_rate": 1.634606741699593e-06,
      "loss": 0.81463903,
      "num_input_tokens_seen": 204844305,
      "step": 9509,
      "time_per_iteration": 3.924833059310913
    },
    {
      "auxiliary_loss_clip": 0.01094957,
      "auxiliary_loss_mlp": 0.01033618,
      "balance_loss_clip": 1.03725672,
      "balance_loss_mlp": 1.02118492,
      "epoch": 0.5717721328723884,
      "flos": 21866043179520.0,
      "grad_norm": 1.8201422809928949,
      "language_loss": 0.72045708,
      "learning_rate": 1.6342238418474255e-06,
      "loss": 0.74174279,
      "num_input_tokens_seen": 204861765,
      "step": 9510,
      "time_per_iteration": 2.4938831329345703
    },
    {
      "auxiliary_loss_clip": 0.01088147,
      "auxiliary_loss_mlp": 0.01024815,
      "balance_loss_clip": 1.03722477,
      "balance_loss_mlp": 1.0132463,
      "epoch": 0.5718322561250564,
      "flos": 28437624126720.0,
      "grad_norm": 1.4977403682916084,
      "language_loss": 0.69609344,
      "learning_rate": 1.6338409558655264e-06,
      "loss": 0.71722305,
      "num_input_tokens_seen": 204882505,
      "step": 9511,
      "time_per_iteration": 2.5952227115631104
    },
    {
      "auxiliary_loss_clip": 0.01090976,
      "auxiliary_loss_mlp": 0.01034036,
      "balance_loss_clip": 1.04039609,
      "balance_loss_mlp": 1.02191842,
      "epoch": 0.5718923793777243,
      "flos": 13551825905280.0,
      "grad_norm": 2.1236188820867494,
      "language_loss": 0.61620033,
      "learning_rate": 1.6334580837684152e-06,
      "loss": 0.63745046,
      "num_input_tokens_seen": 204899830,
      "step": 9512,
      "time_per_iteration": 2.5052785873413086
    },
    {
      "auxiliary_loss_clip": 0.01086958,
      "auxiliary_loss_mlp": 0.01025913,
      "balance_loss_clip": 1.03767371,
      "balance_loss_mlp": 1.01407504,
      "epoch": 0.5719525026303923,
      "flos": 17822035491840.0,
      "grad_norm": 2.247684883837382,
      "language_loss": 0.7584554,
      "learning_rate": 1.6330752255706104e-06,
      "loss": 0.77958411,
      "num_input_tokens_seen": 204918100,
      "step": 9513,
      "time_per_iteration": 2.5619709491729736
    },
    {
      "auxiliary_loss_clip": 0.01043548,
      "auxiliary_loss_mlp": 0.01004773,
      "balance_loss_clip": 1.02763939,
      "balance_loss_mlp": 1.00355721,
      "epoch": 0.5720126258830602,
      "flos": 61298042814720.0,
      "grad_norm": 0.8927797788488366,
      "language_loss": 0.66823095,
      "learning_rate": 1.6326923812866288e-06,
      "loss": 0.68871415,
      "num_input_tokens_seen": 204972925,
      "step": 9514,
      "time_per_iteration": 3.097731590270996
    },
    {
      "auxiliary_loss_clip": 0.01107083,
      "auxiliary_loss_mlp": 0.01038213,
      "balance_loss_clip": 1.04164267,
      "balance_loss_mlp": 1.02489722,
      "epoch": 0.5720727491357283,
      "flos": 23988040997760.0,
      "grad_norm": 2.0966821645380884,
      "language_loss": 0.81354249,
      "learning_rate": 1.63230955093099e-06,
      "loss": 0.83499551,
      "num_input_tokens_seen": 204990910,
      "step": 9515,
      "time_per_iteration": 2.524500608444214
    },
    {
      "auxiliary_loss_clip": 0.01090439,
      "auxiliary_loss_mlp": 0.01028255,
      "balance_loss_clip": 1.03541255,
      "balance_loss_mlp": 1.01575601,
      "epoch": 0.5721328723883962,
      "flos": 23405426398080.0,
      "grad_norm": 1.6819670190450924,
      "language_loss": 0.8565501,
      "learning_rate": 1.6319267345182092e-06,
      "loss": 0.8777371,
      "num_input_tokens_seen": 205010500,
      "step": 9516,
      "time_per_iteration": 2.5218660831451416
    },
    {
      "auxiliary_loss_clip": 0.01079447,
      "auxiliary_loss_mlp": 0.01028215,
      "balance_loss_clip": 1.03960609,
      "balance_loss_mlp": 1.01551962,
      "epoch": 0.5721929956410642,
      "flos": 18804910320000.0,
      "grad_norm": 1.8982365547288318,
      "language_loss": 0.87668842,
      "learning_rate": 1.6315439320628038e-06,
      "loss": 0.89776504,
      "num_input_tokens_seen": 205028560,
      "step": 9517,
      "time_per_iteration": 2.5268032550811768
    },
    {
      "auxiliary_loss_clip": 0.01062084,
      "auxiliary_loss_mlp": 0.01026157,
      "balance_loss_clip": 1.03693318,
      "balance_loss_mlp": 1.01369381,
      "epoch": 0.5722531188937322,
      "flos": 27196659100800.0,
      "grad_norm": 1.7285483319456438,
      "language_loss": 0.85271597,
      "learning_rate": 1.6311611435792893e-06,
      "loss": 0.8735984,
      "num_input_tokens_seen": 205048650,
      "step": 9518,
      "time_per_iteration": 2.62981915473938
    },
    {
      "auxiliary_loss_clip": 0.01096559,
      "auxiliary_loss_mlp": 0.01027152,
      "balance_loss_clip": 1.03863716,
      "balance_loss_mlp": 1.01575518,
      "epoch": 0.5723132421464001,
      "flos": 15195672852480.0,
      "grad_norm": 1.7223778037151694,
      "language_loss": 0.78654826,
      "learning_rate": 1.6307783690821812e-06,
      "loss": 0.80778533,
      "num_input_tokens_seen": 205066480,
      "step": 9519,
      "time_per_iteration": 2.499634265899658
    },
    {
      "auxiliary_loss_clip": 0.01108861,
      "auxiliary_loss_mlp": 0.01027137,
      "balance_loss_clip": 1.03815114,
      "balance_loss_mlp": 1.01516831,
      "epoch": 0.5723733653990681,
      "flos": 27599433281280.0,
      "grad_norm": 1.7617794677634286,
      "language_loss": 0.82896066,
      "learning_rate": 1.6303956085859944e-06,
      "loss": 0.85032058,
      "num_input_tokens_seen": 205087475,
      "step": 9520,
      "time_per_iteration": 2.5477538108825684
    },
    {
      "auxiliary_loss_clip": 0.01091186,
      "auxiliary_loss_mlp": 0.01037051,
      "balance_loss_clip": 1.03956175,
      "balance_loss_mlp": 1.02417064,
      "epoch": 0.572433488651736,
      "flos": 18222870337920.0,
      "grad_norm": 2.1708767714932202,
      "language_loss": 0.72239113,
      "learning_rate": 1.630012862105243e-06,
      "loss": 0.74367344,
      "num_input_tokens_seen": 205106495,
      "step": 9521,
      "time_per_iteration": 2.495607852935791
    },
    {
      "auxiliary_loss_clip": 0.01109241,
      "auxiliary_loss_mlp": 0.00797837,
      "balance_loss_clip": 1.03825366,
      "balance_loss_mlp": 1.03448629,
      "epoch": 0.5724936119044041,
      "flos": 31249106484480.0,
      "grad_norm": 1.8031782177854123,
      "language_loss": 0.78143996,
      "learning_rate": 1.6296301296544415e-06,
      "loss": 0.80051076,
      "num_input_tokens_seen": 205128285,
      "step": 9522,
      "time_per_iteration": 2.54917573928833
    },
    {
      "auxiliary_loss_clip": 0.01082542,
      "auxiliary_loss_mlp": 0.010259,
      "balance_loss_clip": 1.03938484,
      "balance_loss_mlp": 1.01508784,
      "epoch": 0.572553735157072,
      "flos": 19202189719680.0,
      "grad_norm": 1.7727585244931998,
      "language_loss": 0.71645367,
      "learning_rate": 1.629247411248102e-06,
      "loss": 0.7375381,
      "num_input_tokens_seen": 205146595,
      "step": 9523,
      "time_per_iteration": 2.5215957164764404
    },
    {
      "auxiliary_loss_clip": 0.01086657,
      "auxiliary_loss_mlp": 0.01025677,
      "balance_loss_clip": 1.03703606,
      "balance_loss_mlp": 1.01430416,
      "epoch": 0.57261385840974,
      "flos": 21214911386880.0,
      "grad_norm": 1.8085329766047524,
      "language_loss": 0.70161301,
      "learning_rate": 1.628864706900738e-06,
      "loss": 0.72273636,
      "num_input_tokens_seen": 205164295,
      "step": 9524,
      "time_per_iteration": 2.532341718673706
    },
    {
      "auxiliary_loss_clip": 0.01101731,
      "auxiliary_loss_mlp": 0.01026577,
      "balance_loss_clip": 1.0407536,
      "balance_loss_mlp": 1.01506162,
      "epoch": 0.5726739816624079,
      "flos": 33984529793280.0,
      "grad_norm": 1.6096829344576329,
      "language_loss": 0.65060532,
      "learning_rate": 1.6284820166268615e-06,
      "loss": 0.67188841,
      "num_input_tokens_seen": 205185380,
      "step": 9525,
      "time_per_iteration": 2.594330310821533
    },
    {
      "auxiliary_loss_clip": 0.01083082,
      "auxiliary_loss_mlp": 0.01025402,
      "balance_loss_clip": 1.03504348,
      "balance_loss_mlp": 1.01423824,
      "epoch": 0.5727341049150759,
      "flos": 24275972419200.0,
      "grad_norm": 1.6166338521519648,
      "language_loss": 0.72365582,
      "learning_rate": 1.628099340440984e-06,
      "loss": 0.74474066,
      "num_input_tokens_seen": 205204895,
      "step": 9526,
      "time_per_iteration": 2.5427823066711426
    },
    {
      "auxiliary_loss_clip": 0.01097089,
      "auxiliary_loss_mlp": 0.01030829,
      "balance_loss_clip": 1.03804088,
      "balance_loss_mlp": 1.01923609,
      "epoch": 0.5727942281677438,
      "flos": 28400564269440.0,
      "grad_norm": 1.6831639797099494,
      "language_loss": 0.79868984,
      "learning_rate": 1.6277166783576176e-06,
      "loss": 0.819969,
      "num_input_tokens_seen": 205223440,
      "step": 9527,
      "time_per_iteration": 2.551806688308716
    },
    {
      "auxiliary_loss_clip": 0.01097417,
      "auxiliary_loss_mlp": 0.01032638,
      "balance_loss_clip": 1.03760695,
      "balance_loss_mlp": 1.01979947,
      "epoch": 0.5728543514204119,
      "flos": 19536769929600.0,
      "grad_norm": 1.6517106805543817,
      "language_loss": 0.72146332,
      "learning_rate": 1.6273340303912713e-06,
      "loss": 0.74276388,
      "num_input_tokens_seen": 205242800,
      "step": 9528,
      "time_per_iteration": 2.492645025253296
    },
    {
      "auxiliary_loss_clip": 0.01106807,
      "auxiliary_loss_mlp": 0.01032753,
      "balance_loss_clip": 1.03640521,
      "balance_loss_mlp": 1.02024162,
      "epoch": 0.5729144746730798,
      "flos": 21506757390720.0,
      "grad_norm": 1.9677750259052675,
      "language_loss": 0.86092019,
      "learning_rate": 1.6269513965564557e-06,
      "loss": 0.88231575,
      "num_input_tokens_seen": 205259465,
      "step": 9529,
      "time_per_iteration": 2.477224826812744
    },
    {
      "auxiliary_loss_clip": 0.01033155,
      "auxiliary_loss_mlp": 0.01001732,
      "balance_loss_clip": 1.02842808,
      "balance_loss_mlp": 1.0005753,
      "epoch": 0.5729745979257478,
      "flos": 58681628242560.0,
      "grad_norm": 0.7630517016599336,
      "language_loss": 0.5611006,
      "learning_rate": 1.6265687768676813e-06,
      "loss": 0.58144951,
      "num_input_tokens_seen": 205314100,
      "step": 9530,
      "time_per_iteration": 3.0021424293518066
    },
    {
      "auxiliary_loss_clip": 0.01092804,
      "auxiliary_loss_mlp": 0.01027348,
      "balance_loss_clip": 1.0413326,
      "balance_loss_mlp": 1.01554012,
      "epoch": 0.5730347211784158,
      "flos": 18552099421440.0,
      "grad_norm": 1.778605134057174,
      "language_loss": 0.66374612,
      "learning_rate": 1.6261861713394553e-06,
      "loss": 0.68494761,
      "num_input_tokens_seen": 205333420,
      "step": 9531,
      "time_per_iteration": 2.5145435333251953
    },
    {
      "auxiliary_loss_clip": 0.0109431,
      "auxiliary_loss_mlp": 0.01038,
      "balance_loss_clip": 1.03529692,
      "balance_loss_mlp": 1.02353382,
      "epoch": 0.5730948444310837,
      "flos": 38031482396160.0,
      "grad_norm": 2.150210760011325,
      "language_loss": 0.75336659,
      "learning_rate": 1.6258035799862876e-06,
      "loss": 0.77468973,
      "num_input_tokens_seen": 205350995,
      "step": 9532,
      "time_per_iteration": 2.6048686504364014
    },
    {
      "auxiliary_loss_clip": 0.01108589,
      "auxiliary_loss_mlp": 0.01027962,
      "balance_loss_clip": 1.0378356,
      "balance_loss_mlp": 1.01571977,
      "epoch": 0.5731549676837517,
      "flos": 25227066689280.0,
      "grad_norm": 1.33373848369291,
      "language_loss": 0.78671956,
      "learning_rate": 1.625421002822686e-06,
      "loss": 0.80808502,
      "num_input_tokens_seen": 205372675,
      "step": 9533,
      "time_per_iteration": 3.8980746269226074
    },
    {
      "auxiliary_loss_clip": 0.01098039,
      "auxiliary_loss_mlp": 0.01026549,
      "balance_loss_clip": 1.03936124,
      "balance_loss_mlp": 1.01529002,
      "epoch": 0.5732150909364196,
      "flos": 23368222886400.0,
      "grad_norm": 1.5860104457102466,
      "language_loss": 0.85742164,
      "learning_rate": 1.6250384398631574e-06,
      "loss": 0.87866759,
      "num_input_tokens_seen": 205392590,
      "step": 9534,
      "time_per_iteration": 2.5236244201660156
    },
    {
      "auxiliary_loss_clip": 0.01089552,
      "auxiliary_loss_mlp": 0.01030832,
      "balance_loss_clip": 1.03856468,
      "balance_loss_mlp": 1.01801145,
      "epoch": 0.5732752141890877,
      "flos": 23079357711360.0,
      "grad_norm": 1.6351604502384107,
      "language_loss": 0.75038755,
      "learning_rate": 1.6246558911222085e-06,
      "loss": 0.77159131,
      "num_input_tokens_seen": 205414885,
      "step": 9535,
      "time_per_iteration": 2.592090606689453
    },
    {
      "auxiliary_loss_clip": 0.01093295,
      "auxiliary_loss_mlp": 0.01031837,
      "balance_loss_clip": 1.03809488,
      "balance_loss_mlp": 1.01915288,
      "epoch": 0.5733353374417556,
      "flos": 24352282863360.0,
      "grad_norm": 1.6562353931487452,
      "language_loss": 0.70850229,
      "learning_rate": 1.624273356614346e-06,
      "loss": 0.72975361,
      "num_input_tokens_seen": 205434440,
      "step": 9536,
      "time_per_iteration": 2.567009925842285
    },
    {
      "auxiliary_loss_clip": 0.01067973,
      "auxiliary_loss_mlp": 0.0103559,
      "balance_loss_clip": 1.03442121,
      "balance_loss_mlp": 1.02162492,
      "epoch": 0.5733954606944236,
      "flos": 27198849830400.0,
      "grad_norm": 1.8241620583052136,
      "language_loss": 0.69811648,
      "learning_rate": 1.6238908363540755e-06,
      "loss": 0.71915209,
      "num_input_tokens_seen": 205454225,
      "step": 9537,
      "time_per_iteration": 2.606750726699829
    },
    {
      "auxiliary_loss_clip": 0.01110012,
      "auxiliary_loss_mlp": 0.01032646,
      "balance_loss_clip": 1.03860426,
      "balance_loss_mlp": 1.02011752,
      "epoch": 0.5734555839470915,
      "flos": 28765129357440.0,
      "grad_norm": 1.8480729092766628,
      "language_loss": 0.62788153,
      "learning_rate": 1.623508330355902e-06,
      "loss": 0.64930809,
      "num_input_tokens_seen": 205474750,
      "step": 9538,
      "time_per_iteration": 2.5225141048431396
    },
    {
      "auxiliary_loss_clip": 0.01097506,
      "auxiliary_loss_mlp": 0.01032649,
      "balance_loss_clip": 1.03817534,
      "balance_loss_mlp": 1.02024531,
      "epoch": 0.5735157071997595,
      "flos": 22966813422720.0,
      "grad_norm": 1.7368566504944476,
      "language_loss": 0.8321678,
      "learning_rate": 1.6231258386343306e-06,
      "loss": 0.85346937,
      "num_input_tokens_seen": 205495495,
      "step": 9539,
      "time_per_iteration": 2.540273427963257
    },
    {
      "auxiliary_loss_clip": 0.01073486,
      "auxiliary_loss_mlp": 0.01034016,
      "balance_loss_clip": 1.03858471,
      "balance_loss_mlp": 1.02130246,
      "epoch": 0.5735758304524274,
      "flos": 18989455420800.0,
      "grad_norm": 1.9165004901199791,
      "language_loss": 0.73458493,
      "learning_rate": 1.6227433612038647e-06,
      "loss": 0.75565988,
      "num_input_tokens_seen": 205510070,
      "step": 9540,
      "time_per_iteration": 3.978902578353882
    },
    {
      "auxiliary_loss_clip": 0.01094561,
      "auxiliary_loss_mlp": 0.00791357,
      "balance_loss_clip": 1.03992331,
      "balance_loss_mlp": 1.02316213,
      "epoch": 0.5736359537050955,
      "flos": 28397942576640.0,
      "grad_norm": 1.8703763574794876,
      "language_loss": 0.80427146,
      "learning_rate": 1.6223608980790089e-06,
      "loss": 0.82313067,
      "num_input_tokens_seen": 205530190,
      "step": 9541,
      "time_per_iteration": 2.5595176219940186
    },
    {
      "auxiliary_loss_clip": 0.01091424,
      "auxiliary_loss_mlp": 0.01033802,
      "balance_loss_clip": 1.03888714,
      "balance_loss_mlp": 1.02083206,
      "epoch": 0.5736960769577634,
      "flos": 15627210848640.0,
      "grad_norm": 2.510515481482136,
      "language_loss": 0.64581394,
      "learning_rate": 1.6219784492742654e-06,
      "loss": 0.66706622,
      "num_input_tokens_seen": 205547380,
      "step": 9542,
      "time_per_iteration": 2.511321783065796
    },
    {
      "auxiliary_loss_clip": 0.01087974,
      "auxiliary_loss_mlp": 0.01031043,
      "balance_loss_clip": 1.03642023,
      "balance_loss_mlp": 1.01929522,
      "epoch": 0.5737562002104314,
      "flos": 18003994813440.0,
      "grad_norm": 2.3422373565128427,
      "language_loss": 0.82524121,
      "learning_rate": 1.6215960148041365e-06,
      "loss": 0.84643137,
      "num_input_tokens_seen": 205566540,
      "step": 9543,
      "time_per_iteration": 3.890683174133301
    },
    {
      "auxiliary_loss_clip": 0.01074786,
      "auxiliary_loss_mlp": 0.01032018,
      "balance_loss_clip": 1.04066706,
      "balance_loss_mlp": 1.01816535,
      "epoch": 0.5738163234630994,
      "flos": 20698192287360.0,
      "grad_norm": 2.0063288820610983,
      "language_loss": 0.73534238,
      "learning_rate": 1.6212135946831257e-06,
      "loss": 0.75641048,
      "num_input_tokens_seen": 205584200,
      "step": 9544,
      "time_per_iteration": 2.5836920738220215
    },
    {
      "auxiliary_loss_clip": 0.01060159,
      "auxiliary_loss_mlp": 0.01034549,
      "balance_loss_clip": 1.03603303,
      "balance_loss_mlp": 1.02042317,
      "epoch": 0.5738764467157673,
      "flos": 23149311448320.0,
      "grad_norm": 1.611398487346533,
      "language_loss": 0.76189983,
      "learning_rate": 1.620831188925733e-06,
      "loss": 0.78284693,
      "num_input_tokens_seen": 205604675,
      "step": 9545,
      "time_per_iteration": 2.6100571155548096
    },
    {
      "auxiliary_loss_clip": 0.0108825,
      "auxiliary_loss_mlp": 0.01033616,
      "balance_loss_clip": 1.0384413,
      "balance_loss_mlp": 1.0210638,
      "epoch": 0.5739365699684353,
      "flos": 29492930730240.0,
      "grad_norm": 2.1403588151374495,
      "language_loss": 0.56709427,
      "learning_rate": 1.620448797546459e-06,
      "loss": 0.58831292,
      "num_input_tokens_seen": 205624680,
      "step": 9546,
      "time_per_iteration": 2.593029260635376
    },
    {
      "auxiliary_loss_clip": 0.01083136,
      "auxiliary_loss_mlp": 0.01034081,
      "balance_loss_clip": 1.034724,
      "balance_loss_mlp": 1.02137363,
      "epoch": 0.5739966932211032,
      "flos": 14027247342720.0,
      "grad_norm": 2.1381204111938703,
      "language_loss": 0.76421195,
      "learning_rate": 1.6200664205598055e-06,
      "loss": 0.78538418,
      "num_input_tokens_seen": 205641950,
      "step": 9547,
      "time_per_iteration": 2.507086992263794
    },
    {
      "auxiliary_loss_clip": 0.01097711,
      "auxiliary_loss_mlp": 0.01034414,
      "balance_loss_clip": 1.03584099,
      "balance_loss_mlp": 1.02099085,
      "epoch": 0.5740568164737713,
      "flos": 19062030850560.0,
      "grad_norm": 1.896861626753832,
      "language_loss": 0.7394681,
      "learning_rate": 1.6196840579802704e-06,
      "loss": 0.76078933,
      "num_input_tokens_seen": 205660130,
      "step": 9548,
      "time_per_iteration": 3.8896377086639404
    },
    {
      "auxiliary_loss_clip": 0.01080759,
      "auxiliary_loss_mlp": 0.01040283,
      "balance_loss_clip": 1.03564858,
      "balance_loss_mlp": 1.02707505,
      "epoch": 0.5741169397264392,
      "flos": 22127832478080.0,
      "grad_norm": 3.1162462266455933,
      "language_loss": 0.69195485,
      "learning_rate": 1.619301709822355e-06,
      "loss": 0.71316522,
      "num_input_tokens_seen": 205678895,
      "step": 9549,
      "time_per_iteration": 2.5695111751556396
    },
    {
      "auxiliary_loss_clip": 0.01070365,
      "auxiliary_loss_mlp": 0.01028518,
      "balance_loss_clip": 1.04490864,
      "balance_loss_mlp": 1.0167284,
      "epoch": 0.5741770629791072,
      "flos": 24936836797440.0,
      "grad_norm": 1.543047171172197,
      "language_loss": 0.79638988,
      "learning_rate": 1.6189193761005564e-06,
      "loss": 0.81737864,
      "num_input_tokens_seen": 205698450,
      "step": 9550,
      "time_per_iteration": 2.617417573928833
    },
    {
      "auxiliary_loss_clip": 0.01084074,
      "auxiliary_loss_mlp": 0.01040287,
      "balance_loss_clip": 1.03985322,
      "balance_loss_mlp": 1.02609527,
      "epoch": 0.5742371862317751,
      "flos": 18801462614400.0,
      "grad_norm": 1.9132755642380894,
      "language_loss": 0.68039703,
      "learning_rate": 1.6185370568293727e-06,
      "loss": 0.70164061,
      "num_input_tokens_seen": 205714870,
      "step": 9551,
      "time_per_iteration": 2.5182394981384277
    },
    {
      "auxiliary_loss_clip": 0.01076337,
      "auxiliary_loss_mlp": 0.01033148,
      "balance_loss_clip": 1.04079342,
      "balance_loss_mlp": 1.02067876,
      "epoch": 0.5742973094844431,
      "flos": 24460661174400.0,
      "grad_norm": 1.6332408902198359,
      "language_loss": 0.71926105,
      "learning_rate": 1.6181547520233031e-06,
      "loss": 0.74035591,
      "num_input_tokens_seen": 205736045,
      "step": 9552,
      "time_per_iteration": 2.6009294986724854
    },
    {
      "auxiliary_loss_clip": 0.01096652,
      "auxiliary_loss_mlp": 0.01031157,
      "balance_loss_clip": 1.04038477,
      "balance_loss_mlp": 1.01824081,
      "epoch": 0.574357432737111,
      "flos": 21652770176640.0,
      "grad_norm": 2.2882464147097545,
      "language_loss": 0.79886734,
      "learning_rate": 1.617772461696843e-06,
      "loss": 0.82014537,
      "num_input_tokens_seen": 205754445,
      "step": 9553,
      "time_per_iteration": 2.5065300464630127
    },
    {
      "auxiliary_loss_clip": 0.01101528,
      "auxiliary_loss_mlp": 0.01029829,
      "balance_loss_clip": 1.03676379,
      "balance_loss_mlp": 1.01701379,
      "epoch": 0.5744175559897791,
      "flos": 16544728880640.0,
      "grad_norm": 2.191253223324492,
      "language_loss": 0.83503735,
      "learning_rate": 1.6173901858644895e-06,
      "loss": 0.8563509,
      "num_input_tokens_seen": 205770595,
      "step": 9554,
      "time_per_iteration": 2.491607904434204
    },
    {
      "auxiliary_loss_clip": 0.01103781,
      "auxiliary_loss_mlp": 0.00789142,
      "balance_loss_clip": 1.03803802,
      "balance_loss_mlp": 1.01378846,
      "epoch": 0.574477679242447,
      "flos": 24207598880640.0,
      "grad_norm": 1.3239887581741785,
      "language_loss": 0.70904803,
      "learning_rate": 1.6170079245407385e-06,
      "loss": 0.72797728,
      "num_input_tokens_seen": 205791935,
      "step": 9555,
      "time_per_iteration": 2.5555660724639893
    },
    {
      "auxiliary_loss_clip": 0.0108746,
      "auxiliary_loss_mlp": 0.01027006,
      "balance_loss_clip": 1.03735936,
      "balance_loss_mlp": 1.01377928,
      "epoch": 0.574537802495115,
      "flos": 14903000835840.0,
      "grad_norm": 2.1372378962410665,
      "language_loss": 0.72535646,
      "learning_rate": 1.6166256777400853e-06,
      "loss": 0.74650109,
      "num_input_tokens_seen": 205807260,
      "step": 9556,
      "time_per_iteration": 2.4919373989105225
    },
    {
      "auxiliary_loss_clip": 0.01101088,
      "auxiliary_loss_mlp": 0.01031495,
      "balance_loss_clip": 1.03875434,
      "balance_loss_mlp": 1.01873934,
      "epoch": 0.5745979257477829,
      "flos": 24934969290240.0,
      "grad_norm": 1.5507228973860683,
      "language_loss": 0.7394014,
      "learning_rate": 1.6162434454770248e-06,
      "loss": 0.76072729,
      "num_input_tokens_seen": 205826885,
      "step": 9557,
      "time_per_iteration": 2.5309243202209473
    },
    {
      "auxiliary_loss_clip": 0.01101422,
      "auxiliary_loss_mlp": 0.01033608,
      "balance_loss_clip": 1.03812909,
      "balance_loss_mlp": 1.0213176,
      "epoch": 0.5746580490004509,
      "flos": 17235757704960.0,
      "grad_norm": 1.511129056785785,
      "language_loss": 0.67315263,
      "learning_rate": 1.6158612277660514e-06,
      "loss": 0.69450295,
      "num_input_tokens_seen": 205844630,
      "step": 9558,
      "time_per_iteration": 2.47845721244812
    },
    {
      "auxiliary_loss_clip": 0.01085446,
      "auxiliary_loss_mlp": 0.01049259,
      "balance_loss_clip": 1.03800106,
      "balance_loss_mlp": 1.03150845,
      "epoch": 0.5747181722531189,
      "flos": 13187871348480.0,
      "grad_norm": 1.9897538431465565,
      "language_loss": 0.70817542,
      "learning_rate": 1.615479024621659e-06,
      "loss": 0.72952253,
      "num_input_tokens_seen": 205860960,
      "step": 9559,
      "time_per_iteration": 2.483955144882202
    },
    {
      "auxiliary_loss_clip": 0.01090683,
      "auxiliary_loss_mlp": 0.00786203,
      "balance_loss_clip": 1.04041719,
      "balance_loss_mlp": 1.01151371,
      "epoch": 0.5747782955057869,
      "flos": 22963006581120.0,
      "grad_norm": 1.611522133177141,
      "language_loss": 0.7929498,
      "learning_rate": 1.6150968360583398e-06,
      "loss": 0.8117187,
      "num_input_tokens_seen": 205880675,
      "step": 9560,
      "time_per_iteration": 2.576812267303467
    },
    {
      "auxiliary_loss_clip": 0.01045484,
      "auxiliary_loss_mlp": 0.01030483,
      "balance_loss_clip": 1.03717983,
      "balance_loss_mlp": 1.01734078,
      "epoch": 0.5748384187584549,
      "flos": 23403235668480.0,
      "grad_norm": 3.0295522807320423,
      "language_loss": 0.64246464,
      "learning_rate": 1.614714662090588e-06,
      "loss": 0.66322428,
      "num_input_tokens_seen": 205900050,
      "step": 9561,
      "time_per_iteration": 2.643965244293213
    },
    {
      "auxiliary_loss_clip": 0.01109067,
      "auxiliary_loss_mlp": 0.01040413,
      "balance_loss_clip": 1.04039204,
      "balance_loss_mlp": 1.02645993,
      "epoch": 0.5748985420111228,
      "flos": 17785514338560.0,
      "grad_norm": 1.850556924646519,
      "language_loss": 0.71675783,
      "learning_rate": 1.6143325027328945e-06,
      "loss": 0.73825264,
      "num_input_tokens_seen": 205918855,
      "step": 9562,
      "time_per_iteration": 2.50701642036438
    },
    {
      "auxiliary_loss_clip": 0.01063053,
      "auxiliary_loss_mlp": 0.01035412,
      "balance_loss_clip": 1.0414021,
      "balance_loss_mlp": 1.02338362,
      "epoch": 0.5749586652637908,
      "flos": 19866250408320.0,
      "grad_norm": 1.4968871289394081,
      "language_loss": 0.84102875,
      "learning_rate": 1.613950357999751e-06,
      "loss": 0.86201334,
      "num_input_tokens_seen": 205936970,
      "step": 9563,
      "time_per_iteration": 2.5932610034942627
    },
    {
      "auxiliary_loss_clip": 0.01064101,
      "auxiliary_loss_mlp": 0.01036531,
      "balance_loss_clip": 1.03741908,
      "balance_loss_mlp": 1.02323937,
      "epoch": 0.5750187885164587,
      "flos": 21287235421440.0,
      "grad_norm": 1.8234061124145455,
      "language_loss": 0.56915355,
      "learning_rate": 1.6135682279056488e-06,
      "loss": 0.59015989,
      "num_input_tokens_seen": 205954630,
      "step": 9564,
      "time_per_iteration": 2.6312255859375
    },
    {
      "auxiliary_loss_clip": 0.01082275,
      "auxiliary_loss_mlp": 0.01036811,
      "balance_loss_clip": 1.03487062,
      "balance_loss_mlp": 1.02307844,
      "epoch": 0.5750789117691267,
      "flos": 18804658924800.0,
      "grad_norm": 1.762261285838027,
      "language_loss": 0.76083028,
      "learning_rate": 1.613186112465078e-06,
      "loss": 0.78202111,
      "num_input_tokens_seen": 205971510,
      "step": 9565,
      "time_per_iteration": 2.5008556842803955
    },
    {
      "auxiliary_loss_clip": 0.01013251,
      "auxiliary_loss_mlp": 0.01005743,
      "balance_loss_clip": 1.02556431,
      "balance_loss_mlp": 1.004354,
      "epoch": 0.5751390350217946,
      "flos": 70663224124800.0,
      "grad_norm": 0.7416702879767103,
      "language_loss": 0.60841537,
      "learning_rate": 1.6128040116925287e-06,
      "loss": 0.62860537,
      "num_input_tokens_seen": 206035125,
      "step": 9566,
      "time_per_iteration": 3.2971808910369873
    },
    {
      "auxiliary_loss_clip": 0.01085744,
      "auxiliary_loss_mlp": 0.01030164,
      "balance_loss_clip": 1.03921092,
      "balance_loss_mlp": 1.01796889,
      "epoch": 0.5751991582744627,
      "flos": 14246338348800.0,
      "grad_norm": 1.7064161138435505,
      "language_loss": 0.75460857,
      "learning_rate": 1.6124219256024901e-06,
      "loss": 0.77576768,
      "num_input_tokens_seen": 206052075,
      "step": 9567,
      "time_per_iteration": 2.557204008102417
    },
    {
      "auxiliary_loss_clip": 0.01099941,
      "auxiliary_loss_mlp": 0.01032585,
      "balance_loss_clip": 1.03807843,
      "balance_loss_mlp": 1.01970434,
      "epoch": 0.5752592815271306,
      "flos": 18328160079360.0,
      "grad_norm": 1.4999982983897755,
      "language_loss": 0.74601126,
      "learning_rate": 1.6120398542094504e-06,
      "loss": 0.76733649,
      "num_input_tokens_seen": 206069970,
      "step": 9568,
      "time_per_iteration": 2.516064405441284
    },
    {
      "auxiliary_loss_clip": 0.01113469,
      "auxiliary_loss_mlp": 0.01030367,
      "balance_loss_clip": 1.03943384,
      "balance_loss_mlp": 1.01774919,
      "epoch": 0.5753194047797986,
      "flos": 20922742160640.0,
      "grad_norm": 1.6085225091542386,
      "language_loss": 0.7107749,
      "learning_rate": 1.6116577975278994e-06,
      "loss": 0.73221332,
      "num_input_tokens_seen": 206088950,
      "step": 9569,
      "time_per_iteration": 2.586909532546997
    },
    {
      "auxiliary_loss_clip": 0.0110229,
      "auxiliary_loss_mlp": 0.01041371,
      "balance_loss_clip": 1.03902614,
      "balance_loss_mlp": 1.02807927,
      "epoch": 0.5753795280324665,
      "flos": 19281804215040.0,
      "grad_norm": 2.363356065798469,
      "language_loss": 0.55784857,
      "learning_rate": 1.6112757555723223e-06,
      "loss": 0.5792852,
      "num_input_tokens_seen": 206107780,
      "step": 9570,
      "time_per_iteration": 2.4974775314331055
    },
    {
      "auxiliary_loss_clip": 0.01110224,
      "auxiliary_loss_mlp": 0.01035424,
      "balance_loss_clip": 1.03806901,
      "balance_loss_mlp": 1.02342534,
      "epoch": 0.5754396512851345,
      "flos": 21652877917440.0,
      "grad_norm": 1.6164740166745846,
      "language_loss": 0.64654905,
      "learning_rate": 1.6108937283572082e-06,
      "loss": 0.66800547,
      "num_input_tokens_seen": 206127445,
      "step": 9571,
      "time_per_iteration": 3.870680093765259
    },
    {
      "auxiliary_loss_clip": 0.01102108,
      "auxiliary_loss_mlp": 0.01032571,
      "balance_loss_clip": 1.03762579,
      "balance_loss_mlp": 1.01991665,
      "epoch": 0.5754997745378025,
      "flos": 51021700179840.0,
      "grad_norm": 1.5836331463291646,
      "language_loss": 0.66833389,
      "learning_rate": 1.6105117158970434e-06,
      "loss": 0.6896807,
      "num_input_tokens_seen": 206152005,
      "step": 9572,
      "time_per_iteration": 2.7728612422943115
    },
    {
      "auxiliary_loss_clip": 0.01088745,
      "auxiliary_loss_mlp": 0.01033571,
      "balance_loss_clip": 1.03803968,
      "balance_loss_mlp": 1.0209825,
      "epoch": 0.5755598977904705,
      "flos": 22856890826880.0,
      "grad_norm": 2.519247914246552,
      "language_loss": 0.72137564,
      "learning_rate": 1.6101297182063123e-06,
      "loss": 0.74259883,
      "num_input_tokens_seen": 206169875,
      "step": 9573,
      "time_per_iteration": 2.536163806915283
    },
    {
      "auxiliary_loss_clip": 0.01108362,
      "auxiliary_loss_mlp": 0.01029622,
      "balance_loss_clip": 1.03982091,
      "balance_loss_mlp": 1.01814842,
      "epoch": 0.5756200210431385,
      "flos": 38472824805120.0,
      "grad_norm": 1.7326197357493094,
      "language_loss": 0.76557893,
      "learning_rate": 1.6097477352995022e-06,
      "loss": 0.78695875,
      "num_input_tokens_seen": 206192635,
      "step": 9574,
      "time_per_iteration": 2.5983545780181885
    },
    {
      "auxiliary_loss_clip": 0.01061326,
      "auxiliary_loss_mlp": 0.0103832,
      "balance_loss_clip": 1.03499579,
      "balance_loss_mlp": 1.02400911,
      "epoch": 0.5756801442958064,
      "flos": 23910006700800.0,
      "grad_norm": 2.77648120227331,
      "language_loss": 0.6636247,
      "learning_rate": 1.6093657671910968e-06,
      "loss": 0.68462121,
      "num_input_tokens_seen": 206211485,
      "step": 9575,
      "time_per_iteration": 2.6172397136688232
    },
    {
      "auxiliary_loss_clip": 0.01088042,
      "auxiliary_loss_mlp": 0.01030814,
      "balance_loss_clip": 1.0380342,
      "balance_loss_mlp": 1.01893449,
      "epoch": 0.5757402675484744,
      "flos": 21105276099840.0,
      "grad_norm": 1.5558898645892454,
      "language_loss": 0.79777193,
      "learning_rate": 1.6089838138955804e-06,
      "loss": 0.81896049,
      "num_input_tokens_seen": 206231740,
      "step": 9576,
      "time_per_iteration": 2.535367727279663
    },
    {
      "auxiliary_loss_clip": 0.01088145,
      "auxiliary_loss_mlp": 0.01026765,
      "balance_loss_clip": 1.03839636,
      "balance_loss_mlp": 1.01539254,
      "epoch": 0.5758003908011423,
      "flos": 20559110826240.0,
      "grad_norm": 1.6535958508242505,
      "language_loss": 0.693295,
      "learning_rate": 1.6086018754274372e-06,
      "loss": 0.7144441,
      "num_input_tokens_seen": 206250975,
      "step": 9577,
      "time_per_iteration": 2.526874303817749
    },
    {
      "auxiliary_loss_clip": 0.01102996,
      "auxiliary_loss_mlp": 0.01034222,
      "balance_loss_clip": 1.03848052,
      "balance_loss_mlp": 1.02199745,
      "epoch": 0.5758605140538103,
      "flos": 16473015377280.0,
      "grad_norm": 1.6495591472385231,
      "language_loss": 0.66861677,
      "learning_rate": 1.6082199518011504e-06,
      "loss": 0.68998897,
      "num_input_tokens_seen": 206268800,
      "step": 9578,
      "time_per_iteration": 2.470482587814331
    },
    {
      "auxiliary_loss_clip": 0.01089002,
      "auxiliary_loss_mlp": 0.01026933,
      "balance_loss_clip": 1.03784275,
      "balance_loss_mlp": 1.01522708,
      "epoch": 0.5759206373064782,
      "flos": 21287558643840.0,
      "grad_norm": 1.6553460605654928,
      "language_loss": 0.72552121,
      "learning_rate": 1.6078380430312016e-06,
      "loss": 0.74668056,
      "num_input_tokens_seen": 206287190,
      "step": 9579,
      "time_per_iteration": 3.8875722885131836
    },
    {
      "auxiliary_loss_clip": 0.01095066,
      "auxiliary_loss_mlp": 0.0103113,
      "balance_loss_clip": 1.04029238,
      "balance_loss_mlp": 1.01738548,
      "epoch": 0.5759807605591463,
      "flos": 26067879227520.0,
      "grad_norm": 2.318208160183359,
      "language_loss": 0.63953638,
      "learning_rate": 1.6074561491320742e-06,
      "loss": 0.66079837,
      "num_input_tokens_seen": 206307020,
      "step": 9580,
      "time_per_iteration": 2.5736711025238037
    },
    {
      "auxiliary_loss_clip": 0.01083778,
      "auxiliary_loss_mlp": 0.01032191,
      "balance_loss_clip": 1.03510988,
      "balance_loss_mlp": 1.01860678,
      "epoch": 0.5760408838118142,
      "flos": 18873068376960.0,
      "grad_norm": 1.8295650273476893,
      "language_loss": 0.85351741,
      "learning_rate": 1.6070742701182486e-06,
      "loss": 0.87467706,
      "num_input_tokens_seen": 206324095,
      "step": 9581,
      "time_per_iteration": 2.509777784347534
    },
    {
      "auxiliary_loss_clip": 0.01119343,
      "auxiliary_loss_mlp": 0.01041945,
      "balance_loss_clip": 1.04206812,
      "balance_loss_mlp": 1.02830756,
      "epoch": 0.5761010070644822,
      "flos": 15378134964480.0,
      "grad_norm": 3.2848204941500736,
      "language_loss": 0.66850209,
      "learning_rate": 1.6066924060042057e-06,
      "loss": 0.69011497,
      "num_input_tokens_seen": 206343210,
      "step": 9582,
      "time_per_iteration": 3.871877431869507
    },
    {
      "auxiliary_loss_clip": 0.0103245,
      "auxiliary_loss_mlp": 0.01003135,
      "balance_loss_clip": 1.02614343,
      "balance_loss_mlp": 1.00197911,
      "epoch": 0.5761611303171501,
      "flos": 71471932882560.0,
      "grad_norm": 0.6714170727724333,
      "language_loss": 0.5725106,
      "learning_rate": 1.6063105568044271e-06,
      "loss": 0.59286642,
      "num_input_tokens_seen": 206415935,
      "step": 9583,
      "time_per_iteration": 3.314527988433838
    },
    {
      "auxiliary_loss_clip": 0.01085306,
      "auxiliary_loss_mlp": 0.01028457,
      "balance_loss_clip": 1.03902435,
      "balance_loss_mlp": 1.01620865,
      "epoch": 0.5762212535698181,
      "flos": 16246167033600.0,
      "grad_norm": 1.7737378394620331,
      "language_loss": 0.82577968,
      "learning_rate": 1.6059287225333912e-06,
      "loss": 0.84691727,
      "num_input_tokens_seen": 206431900,
      "step": 9584,
      "time_per_iteration": 2.503521203994751
    },
    {
      "auxiliary_loss_clip": 0.01047451,
      "auxiliary_loss_mlp": 0.01003377,
      "balance_loss_clip": 1.02265346,
      "balance_loss_mlp": 1.00222707,
      "epoch": 0.5762813768224861,
      "flos": 70185504216960.0,
      "grad_norm": 0.6332332230034082,
      "language_loss": 0.49561346,
      "learning_rate": 1.6055469032055773e-06,
      "loss": 0.51612175,
      "num_input_tokens_seen": 206501200,
      "step": 9585,
      "time_per_iteration": 3.142387628555298
    },
    {
      "auxiliary_loss_clip": 0.01083443,
      "auxiliary_loss_mlp": 0.01026817,
      "balance_loss_clip": 1.03562474,
      "balance_loss_mlp": 1.01515877,
      "epoch": 0.5763415000751541,
      "flos": 20518028645760.0,
      "grad_norm": 1.6885465176305137,
      "language_loss": 0.84566736,
      "learning_rate": 1.605165098835465e-06,
      "loss": 0.86676997,
      "num_input_tokens_seen": 206520575,
      "step": 9586,
      "time_per_iteration": 2.5307562351226807
    },
    {
      "auxiliary_loss_clip": 0.01098699,
      "auxiliary_loss_mlp": 0.01030525,
      "balance_loss_clip": 1.04122639,
      "balance_loss_mlp": 1.01747131,
      "epoch": 0.5764016233278221,
      "flos": 15815526877440.0,
      "grad_norm": 1.7563100313009583,
      "language_loss": 0.79709017,
      "learning_rate": 1.6047833094375308e-06,
      "loss": 0.8183825,
      "num_input_tokens_seen": 206538060,
      "step": 9587,
      "time_per_iteration": 3.8719122409820557
    },
    {
      "auxiliary_loss_clip": 0.01080611,
      "auxiliary_loss_mlp": 0.01034322,
      "balance_loss_clip": 1.03551686,
      "balance_loss_mlp": 1.0208571,
      "epoch": 0.57646174658049,
      "flos": 20772312001920.0,
      "grad_norm": 1.803565523749224,
      "language_loss": 0.66185069,
      "learning_rate": 1.6044015350262542e-06,
      "loss": 0.68300003,
      "num_input_tokens_seen": 206557320,
      "step": 9588,
      "time_per_iteration": 2.5080082416534424
    },
    {
      "auxiliary_loss_clip": 0.01084625,
      "auxiliary_loss_mlp": 0.01035045,
      "balance_loss_clip": 1.03616285,
      "balance_loss_mlp": 1.02128196,
      "epoch": 0.576521869833158,
      "flos": 23549930812800.0,
      "grad_norm": 1.9641677735767007,
      "language_loss": 0.78744984,
      "learning_rate": 1.6040197756161104e-06,
      "loss": 0.80864656,
      "num_input_tokens_seen": 206575780,
      "step": 9589,
      "time_per_iteration": 2.5306313037872314
    },
    {
      "auxiliary_loss_clip": 0.01106804,
      "auxiliary_loss_mlp": 0.01026467,
      "balance_loss_clip": 1.03625321,
      "balance_loss_mlp": 1.01465368,
      "epoch": 0.5765819930858259,
      "flos": 20266582464000.0,
      "grad_norm": 1.9597697625278394,
      "language_loss": 0.79119581,
      "learning_rate": 1.6036380312215762e-06,
      "loss": 0.81252855,
      "num_input_tokens_seen": 206594100,
      "step": 9590,
      "time_per_iteration": 2.4647674560546875
    },
    {
      "auxiliary_loss_clip": 0.0105259,
      "auxiliary_loss_mlp": 0.00784155,
      "balance_loss_clip": 1.03951335,
      "balance_loss_mlp": 1.0084399,
      "epoch": 0.5766421163384939,
      "flos": 23148772744320.0,
      "grad_norm": 1.6967300517918502,
      "language_loss": 0.62544715,
      "learning_rate": 1.6032563018571283e-06,
      "loss": 0.64381462,
      "num_input_tokens_seen": 206613325,
      "step": 9591,
      "time_per_iteration": 2.6545443534851074
    },
    {
      "auxiliary_loss_clip": 0.01112376,
      "auxiliary_loss_mlp": 0.00789919,
      "balance_loss_clip": 1.03913271,
      "balance_loss_mlp": 1.01460314,
      "epoch": 0.5767022395911618,
      "flos": 25848895962240.0,
      "grad_norm": 1.6436047468632307,
      "language_loss": 0.77798533,
      "learning_rate": 1.6028745875372406e-06,
      "loss": 0.79700828,
      "num_input_tokens_seen": 206634265,
      "step": 9592,
      "time_per_iteration": 2.5054421424865723
    },
    {
      "auxiliary_loss_clip": 0.01002907,
      "auxiliary_loss_mlp": 0.01009004,
      "balance_loss_clip": 1.0270834,
      "balance_loss_mlp": 1.00735891,
      "epoch": 0.5767623628438299,
      "flos": 68293299657600.0,
      "grad_norm": 0.7332544973540513,
      "language_loss": 0.59644496,
      "learning_rate": 1.6024928882763885e-06,
      "loss": 0.6165641,
      "num_input_tokens_seen": 206696990,
      "step": 9593,
      "time_per_iteration": 3.35587477684021
    },
    {
      "auxiliary_loss_clip": 0.01103429,
      "auxiliary_loss_mlp": 0.01045106,
      "balance_loss_clip": 1.03781521,
      "balance_loss_mlp": 1.03147435,
      "epoch": 0.5768224860964978,
      "flos": 30188448754560.0,
      "grad_norm": 1.5638496278570684,
      "language_loss": 0.71214902,
      "learning_rate": 1.6021112040890463e-06,
      "loss": 0.73363435,
      "num_input_tokens_seen": 206717815,
      "step": 9594,
      "time_per_iteration": 2.58227276802063
    },
    {
      "auxiliary_loss_clip": 0.01077126,
      "auxiliary_loss_mlp": 0.01030767,
      "balance_loss_clip": 1.03752565,
      "balance_loss_mlp": 1.01966286,
      "epoch": 0.5768826093491658,
      "flos": 17895041884800.0,
      "grad_norm": 1.7154469790847517,
      "language_loss": 0.70838284,
      "learning_rate": 1.6017295349896863e-06,
      "loss": 0.72946179,
      "num_input_tokens_seen": 206735985,
      "step": 9595,
      "time_per_iteration": 2.5414834022521973
    },
    {
      "auxiliary_loss_clip": 0.01111601,
      "auxiliary_loss_mlp": 0.01029945,
      "balance_loss_clip": 1.03897047,
      "balance_loss_mlp": 1.01726127,
      "epoch": 0.5769427326018337,
      "flos": 17457183095040.0,
      "grad_norm": 2.293411051975021,
      "language_loss": 0.70005077,
      "learning_rate": 1.6013478809927828e-06,
      "loss": 0.7214663,
      "num_input_tokens_seen": 206753370,
      "step": 9596,
      "time_per_iteration": 2.4497220516204834
    },
    {
      "auxiliary_loss_clip": 0.01090222,
      "auxiliary_loss_mlp": 0.01037031,
      "balance_loss_clip": 1.04040253,
      "balance_loss_mlp": 1.022493,
      "epoch": 0.5770028558545017,
      "flos": 39421728345600.0,
      "grad_norm": 4.511767085649352,
      "language_loss": 0.67250514,
      "learning_rate": 1.6009662421128074e-06,
      "loss": 0.69377768,
      "num_input_tokens_seen": 206777645,
      "step": 9597,
      "time_per_iteration": 2.679090976715088
    },
    {
      "auxiliary_loss_clip": 0.01089452,
      "auxiliary_loss_mlp": 0.0103314,
      "balance_loss_clip": 1.03847456,
      "balance_loss_mlp": 1.02099228,
      "epoch": 0.5770629791071697,
      "flos": 21536383132800.0,
      "grad_norm": 1.7757863049632339,
      "language_loss": 0.81311107,
      "learning_rate": 1.6005846183642323e-06,
      "loss": 0.83433688,
      "num_input_tokens_seen": 206794865,
      "step": 9598,
      "time_per_iteration": 2.546495199203491
    },
    {
      "auxiliary_loss_clip": 0.01068511,
      "auxiliary_loss_mlp": 0.01044445,
      "balance_loss_clip": 1.03362823,
      "balance_loss_mlp": 1.02923369,
      "epoch": 0.5771231023598377,
      "flos": 20886795624960.0,
      "grad_norm": 1.4492360448841175,
      "language_loss": 0.7266261,
      "learning_rate": 1.6002030097615277e-06,
      "loss": 0.74775565,
      "num_input_tokens_seen": 206814095,
      "step": 9599,
      "time_per_iteration": 2.5570120811462402
    },
    {
      "auxiliary_loss_clip": 0.01108364,
      "auxiliary_loss_mlp": 0.01032859,
      "balance_loss_clip": 1.03841829,
      "balance_loss_mlp": 1.02113485,
      "epoch": 0.5771832256125057,
      "flos": 18077216688000.0,
      "grad_norm": 33.99850999422814,
      "language_loss": 0.77838981,
      "learning_rate": 1.5998214163191663e-06,
      "loss": 0.79980201,
      "num_input_tokens_seen": 206832245,
      "step": 9600,
      "time_per_iteration": 2.451108455657959
    },
    {
      "auxiliary_loss_clip": 0.01106277,
      "auxiliary_loss_mlp": 0.00787037,
      "balance_loss_clip": 1.04156065,
      "balance_loss_mlp": 1.00862527,
      "epoch": 0.5772433488651736,
      "flos": 26359078786560.0,
      "grad_norm": 3.651324503435479,
      "language_loss": 0.72531855,
      "learning_rate": 1.5994398380516163e-06,
      "loss": 0.74425173,
      "num_input_tokens_seen": 206851535,
      "step": 9601,
      "time_per_iteration": 2.5447776317596436
    },
    {
      "auxiliary_loss_clip": 0.01054195,
      "auxiliary_loss_mlp": 0.01036181,
      "balance_loss_clip": 1.03777623,
      "balance_loss_mlp": 1.02346158,
      "epoch": 0.5773034721178416,
      "flos": 19680987035520.0,
      "grad_norm": 1.9632505663559048,
      "language_loss": 0.6870755,
      "learning_rate": 1.599058274973348e-06,
      "loss": 0.70797926,
      "num_input_tokens_seen": 206870595,
      "step": 9602,
      "time_per_iteration": 2.6276421546936035
    },
    {
      "auxiliary_loss_clip": 0.01081105,
      "auxiliary_loss_mlp": 0.01039592,
      "balance_loss_clip": 1.03620386,
      "balance_loss_mlp": 1.02668142,
      "epoch": 0.5773635953705095,
      "flos": 25082885496960.0,
      "grad_norm": 1.456868008043915,
      "language_loss": 0.72893697,
      "learning_rate": 1.5986767270988297e-06,
      "loss": 0.75014395,
      "num_input_tokens_seen": 206892320,
      "step": 9603,
      "time_per_iteration": 2.551957607269287
    },
    {
      "auxiliary_loss_clip": 0.01098239,
      "auxiliary_loss_mlp": 0.01031097,
      "balance_loss_clip": 1.03829718,
      "balance_loss_mlp": 1.01883674,
      "epoch": 0.5774237186231775,
      "flos": 21032987978880.0,
      "grad_norm": 1.637205433736546,
      "language_loss": 0.76486218,
      "learning_rate": 1.5982951944425298e-06,
      "loss": 0.78615558,
      "num_input_tokens_seen": 206912485,
      "step": 9604,
      "time_per_iteration": 2.5517444610595703
    },
    {
      "auxiliary_loss_clip": 0.01078623,
      "auxiliary_loss_mlp": 0.0103517,
      "balance_loss_clip": 1.03819728,
      "balance_loss_mlp": 1.02109146,
      "epoch": 0.5774838418758454,
      "flos": 15231727128960.0,
      "grad_norm": 1.5914744355961994,
      "language_loss": 0.83448887,
      "learning_rate": 1.5979136770189174e-06,
      "loss": 0.85562682,
      "num_input_tokens_seen": 206929100,
      "step": 9605,
      "time_per_iteration": 2.5371854305267334
    },
    {
      "auxiliary_loss_clip": 0.01087309,
      "auxiliary_loss_mlp": 0.01033881,
      "balance_loss_clip": 1.04086876,
      "balance_loss_mlp": 1.01918876,
      "epoch": 0.5775439651285135,
      "flos": 23582609210880.0,
      "grad_norm": 1.6257056546838746,
      "language_loss": 0.78168225,
      "learning_rate": 1.5975321748424581e-06,
      "loss": 0.80289418,
      "num_input_tokens_seen": 206947020,
      "step": 9606,
      "time_per_iteration": 2.572598695755005
    },
    {
      "auxiliary_loss_clip": 0.01110487,
      "auxiliary_loss_mlp": 0.01039323,
      "balance_loss_clip": 1.037907,
      "balance_loss_mlp": 1.02723527,
      "epoch": 0.5776040883811814,
      "flos": 18040515966720.0,
      "grad_norm": 1.7001614488418964,
      "language_loss": 0.73802781,
      "learning_rate": 1.597150687927619e-06,
      "loss": 0.7595259,
      "num_input_tokens_seen": 206964065,
      "step": 9607,
      "time_per_iteration": 2.438049077987671
    },
    {
      "auxiliary_loss_clip": 0.01072783,
      "auxiliary_loss_mlp": 0.01034188,
      "balance_loss_clip": 1.04144335,
      "balance_loss_mlp": 1.02112854,
      "epoch": 0.5776642116338494,
      "flos": 18624638937600.0,
      "grad_norm": 1.621530558822664,
      "language_loss": 0.69327891,
      "learning_rate": 1.5967692162888664e-06,
      "loss": 0.71434867,
      "num_input_tokens_seen": 206981940,
      "step": 9608,
      "time_per_iteration": 2.536414384841919
    },
    {
      "auxiliary_loss_clip": 0.01079845,
      "auxiliary_loss_mlp": 0.01039604,
      "balance_loss_clip": 1.03722501,
      "balance_loss_mlp": 1.02625895,
      "epoch": 0.5777243348865173,
      "flos": 28402539517440.0,
      "grad_norm": 1.9686946279303033,
      "language_loss": 0.76697063,
      "learning_rate": 1.596387759940665e-06,
      "loss": 0.78816509,
      "num_input_tokens_seen": 207002365,
      "step": 9609,
      "time_per_iteration": 2.6088550090789795
    },
    {
      "auxiliary_loss_clip": 0.01079729,
      "auxiliary_loss_mlp": 0.01033535,
      "balance_loss_clip": 1.03666258,
      "balance_loss_mlp": 1.02106571,
      "epoch": 0.5777844581391853,
      "flos": 24024705805440.0,
      "grad_norm": 1.6330184431273478,
      "language_loss": 0.77313435,
      "learning_rate": 1.5960063188974808e-06,
      "loss": 0.79426706,
      "num_input_tokens_seen": 207021195,
      "step": 9610,
      "time_per_iteration": 3.98441481590271
    },
    {
      "auxiliary_loss_clip": 0.01076136,
      "auxiliary_loss_mlp": 0.01031014,
      "balance_loss_clip": 1.03542721,
      "balance_loss_mlp": 1.01751316,
      "epoch": 0.5778445813918534,
      "flos": 17777361951360.0,
      "grad_norm": 2.220790911156004,
      "language_loss": 0.69114554,
      "learning_rate": 1.5956248931737777e-06,
      "loss": 0.71221697,
      "num_input_tokens_seen": 207037465,
      "step": 9611,
      "time_per_iteration": 2.5201942920684814
    },
    {
      "auxiliary_loss_clip": 0.01097783,
      "auxiliary_loss_mlp": 0.01029928,
      "balance_loss_clip": 1.03599942,
      "balance_loss_mlp": 1.01685715,
      "epoch": 0.5779047046445213,
      "flos": 22233194046720.0,
      "grad_norm": 1.8832902346325437,
      "language_loss": 0.83548117,
      "learning_rate": 1.5952434827840185e-06,
      "loss": 0.85675824,
      "num_input_tokens_seen": 207054230,
      "step": 9612,
      "time_per_iteration": 2.481104850769043
    },
    {
      "auxiliary_loss_clip": 0.01111463,
      "auxiliary_loss_mlp": 0.0103062,
      "balance_loss_clip": 1.03877521,
      "balance_loss_mlp": 1.01763272,
      "epoch": 0.5779648278971893,
      "flos": 21434361528960.0,
      "grad_norm": 1.5968871680160088,
      "language_loss": 0.79444182,
      "learning_rate": 1.594862087742667e-06,
      "loss": 0.81586266,
      "num_input_tokens_seen": 207073150,
      "step": 9613,
      "time_per_iteration": 2.4771249294281006
    },
    {
      "auxiliary_loss_clip": 0.01096945,
      "auxiliary_loss_mlp": 0.01035991,
      "balance_loss_clip": 1.03515565,
      "balance_loss_mlp": 1.02439821,
      "epoch": 0.5780249511498572,
      "flos": 19026120228480.0,
      "grad_norm": 1.964052428129157,
      "language_loss": 0.77522904,
      "learning_rate": 1.5944807080641863e-06,
      "loss": 0.79655844,
      "num_input_tokens_seen": 207090375,
      "step": 9614,
      "time_per_iteration": 2.4691452980041504
    },
    {
      "auxiliary_loss_clip": 0.01078888,
      "auxiliary_loss_mlp": 0.01035095,
      "balance_loss_clip": 1.03708076,
      "balance_loss_mlp": 1.02248836,
      "epoch": 0.5780850744025252,
      "flos": 12124663752960.0,
      "grad_norm": 2.6195108466702175,
      "language_loss": 0.80592608,
      "learning_rate": 1.5940993437630375e-06,
      "loss": 0.82706589,
      "num_input_tokens_seen": 207106030,
      "step": 9615,
      "time_per_iteration": 2.52592396736145
    },
    {
      "auxiliary_loss_clip": 0.01098756,
      "auxiliary_loss_mlp": 0.01032849,
      "balance_loss_clip": 1.03605521,
      "balance_loss_mlp": 1.01991487,
      "epoch": 0.5781451976551931,
      "flos": 25044425009280.0,
      "grad_norm": 1.62400853154449,
      "language_loss": 0.6682651,
      "learning_rate": 1.5937179948536825e-06,
      "loss": 0.68958116,
      "num_input_tokens_seen": 207125435,
      "step": 9616,
      "time_per_iteration": 2.516204833984375
    },
    {
      "auxiliary_loss_clip": 0.01099607,
      "auxiliary_loss_mlp": 0.01030558,
      "balance_loss_clip": 1.03791416,
      "balance_loss_mlp": 1.018399,
      "epoch": 0.5782053209078611,
      "flos": 19245606284160.0,
      "grad_norm": 1.6368851459517004,
      "language_loss": 0.77510601,
      "learning_rate": 1.5933366613505812e-06,
      "loss": 0.79640758,
      "num_input_tokens_seen": 207145095,
      "step": 9617,
      "time_per_iteration": 3.8840296268463135
    },
    {
      "auxiliary_loss_clip": 0.0108614,
      "auxiliary_loss_mlp": 0.01032421,
      "balance_loss_clip": 1.03626132,
      "balance_loss_mlp": 1.0189209,
      "epoch": 0.578265444160529,
      "flos": 25993831340160.0,
      "grad_norm": 1.487659265962285,
      "language_loss": 0.74895203,
      "learning_rate": 1.5929553432681947e-06,
      "loss": 0.77013755,
      "num_input_tokens_seen": 207166045,
      "step": 9618,
      "time_per_iteration": 2.6071295738220215
    },
    {
      "auxiliary_loss_clip": 0.0111016,
      "auxiliary_loss_mlp": 0.01030254,
      "balance_loss_clip": 1.03822553,
      "balance_loss_mlp": 1.01807642,
      "epoch": 0.5783255674131971,
      "flos": 21798603394560.0,
      "grad_norm": 1.5122850014564808,
      "language_loss": 0.81224424,
      "learning_rate": 1.5925740406209826e-06,
      "loss": 0.83364838,
      "num_input_tokens_seen": 207185290,
      "step": 9619,
      "time_per_iteration": 2.4854722023010254
    },
    {
      "auxiliary_loss_clip": 0.01090868,
      "auxiliary_loss_mlp": 0.01031663,
      "balance_loss_clip": 1.03743315,
      "balance_loss_mlp": 1.01924777,
      "epoch": 0.578385690665865,
      "flos": 24789746603520.0,
      "grad_norm": 1.749753163619696,
      "language_loss": 0.72332227,
      "learning_rate": 1.5921927534234039e-06,
      "loss": 0.74454761,
      "num_input_tokens_seen": 207205505,
      "step": 9620,
      "time_per_iteration": 3.9409501552581787
    },
    {
      "auxiliary_loss_clip": 0.01091337,
      "auxiliary_loss_mlp": 0.01029432,
      "balance_loss_clip": 1.03855944,
      "balance_loss_mlp": 1.01671791,
      "epoch": 0.578445813918533,
      "flos": 21212864311680.0,
      "grad_norm": 1.677801573517177,
      "language_loss": 0.77153313,
      "learning_rate": 1.591811481689916e-06,
      "loss": 0.79274088,
      "num_input_tokens_seen": 207225315,
      "step": 9621,
      "time_per_iteration": 2.5292625427246094
    },
    {
      "auxiliary_loss_clip": 0.01051488,
      "auxiliary_loss_mlp": 0.01040284,
      "balance_loss_clip": 1.03520775,
      "balance_loss_mlp": 1.02531695,
      "epoch": 0.5785059371712009,
      "flos": 25046795306880.0,
      "grad_norm": 1.4256362301776235,
      "language_loss": 0.70454353,
      "learning_rate": 1.5914302254349787e-06,
      "loss": 0.72546124,
      "num_input_tokens_seen": 207247690,
      "step": 9622,
      "time_per_iteration": 2.6778180599212646
    },
    {
      "auxiliary_loss_clip": 0.01022642,
      "auxiliary_loss_mlp": 0.0101059,
      "balance_loss_clip": 1.01941776,
      "balance_loss_mlp": 1.00925529,
      "epoch": 0.5785660604238689,
      "flos": 70843172284800.0,
      "grad_norm": 0.7716990871375858,
      "language_loss": 0.55968612,
      "learning_rate": 1.5910489846730476e-06,
      "loss": 0.58001846,
      "num_input_tokens_seen": 207301735,
      "step": 9623,
      "time_per_iteration": 3.1689460277557373
    },
    {
      "auxiliary_loss_clip": 0.01075807,
      "auxiliary_loss_mlp": 0.01035289,
      "balance_loss_clip": 1.03954959,
      "balance_loss_mlp": 1.0218842,
      "epoch": 0.578626183676537,
      "flos": 31649977244160.0,
      "grad_norm": 1.7579473328603779,
      "language_loss": 0.71149564,
      "learning_rate": 1.5906677594185799e-06,
      "loss": 0.73260659,
      "num_input_tokens_seen": 207321240,
      "step": 9624,
      "time_per_iteration": 2.6557812690734863
    },
    {
      "auxiliary_loss_clip": 0.01077434,
      "auxiliary_loss_mlp": 0.01035714,
      "balance_loss_clip": 1.03694963,
      "balance_loss_mlp": 1.02247608,
      "epoch": 0.5786863069292049,
      "flos": 21865181253120.0,
      "grad_norm": 1.9196384659944488,
      "language_loss": 0.82300937,
      "learning_rate": 1.5902865496860322e-06,
      "loss": 0.84414083,
      "num_input_tokens_seen": 207339540,
      "step": 9625,
      "time_per_iteration": 4.006484746932983
    },
    {
      "auxiliary_loss_clip": 0.01108705,
      "auxiliary_loss_mlp": 0.01034307,
      "balance_loss_clip": 1.03711653,
      "balance_loss_mlp": 1.02035403,
      "epoch": 0.5787464301818729,
      "flos": 23364954748800.0,
      "grad_norm": 1.4302865888315015,
      "language_loss": 0.7019484,
      "learning_rate": 1.5899053554898591e-06,
      "loss": 0.72337854,
      "num_input_tokens_seen": 207360470,
      "step": 9626,
      "time_per_iteration": 2.5032944679260254
    },
    {
      "auxiliary_loss_clip": 0.01086055,
      "auxiliary_loss_mlp": 0.01032099,
      "balance_loss_clip": 1.03581452,
      "balance_loss_mlp": 1.02001071,
      "epoch": 0.5788065534345408,
      "flos": 30004011394560.0,
      "grad_norm": 1.4808204878485407,
      "language_loss": 0.71533668,
      "learning_rate": 1.5895241768445166e-06,
      "loss": 0.73651826,
      "num_input_tokens_seen": 207383080,
      "step": 9627,
      "time_per_iteration": 2.610231399536133
    },
    {
      "auxiliary_loss_clip": 0.01098718,
      "auxiliary_loss_mlp": 0.01024868,
      "balance_loss_clip": 1.03699946,
      "balance_loss_mlp": 1.01254725,
      "epoch": 0.5788666766872088,
      "flos": 24527849564160.0,
      "grad_norm": 2.0279055591584787,
      "language_loss": 0.83884835,
      "learning_rate": 1.589143013764458e-06,
      "loss": 0.86008424,
      "num_input_tokens_seen": 207401000,
      "step": 9628,
      "time_per_iteration": 2.52591872215271
    },
    {
      "auxiliary_loss_clip": 0.01089572,
      "auxiliary_loss_mlp": 0.01028582,
      "balance_loss_clip": 1.03638864,
      "balance_loss_mlp": 1.0153017,
      "epoch": 0.5789267999398767,
      "flos": 23732823888000.0,
      "grad_norm": 1.5397687382546326,
      "language_loss": 0.72260225,
      "learning_rate": 1.5887618662641376e-06,
      "loss": 0.74378377,
      "num_input_tokens_seen": 207419230,
      "step": 9629,
      "time_per_iteration": 2.559020757675171
    },
    {
      "auxiliary_loss_clip": 0.01089128,
      "auxiliary_loss_mlp": 0.01032604,
      "balance_loss_clip": 1.03952777,
      "balance_loss_mlp": 1.01937807,
      "epoch": 0.5789869231925447,
      "flos": 21135045496320.0,
      "grad_norm": 2.111620354855419,
      "language_loss": 0.74529856,
      "learning_rate": 1.5883807343580087e-06,
      "loss": 0.76651585,
      "num_input_tokens_seen": 207437615,
      "step": 9630,
      "time_per_iteration": 2.5147297382354736
    },
    {
      "auxiliary_loss_clip": 0.01077298,
      "auxiliary_loss_mlp": 0.00787696,
      "balance_loss_clip": 1.03749299,
      "balance_loss_mlp": 1.01085997,
      "epoch": 0.5790470464452127,
      "flos": 21209632087680.0,
      "grad_norm": 1.570739191210158,
      "language_loss": 0.78826874,
      "learning_rate": 1.587999618060523e-06,
      "loss": 0.80691868,
      "num_input_tokens_seen": 207457270,
      "step": 9631,
      "time_per_iteration": 2.5936620235443115
    },
    {
      "auxiliary_loss_clip": 0.01111485,
      "auxiliary_loss_mlp": 0.01026315,
      "balance_loss_clip": 1.03834999,
      "balance_loss_mlp": 1.01352978,
      "epoch": 0.5791071696978807,
      "flos": 23404384903680.0,
      "grad_norm": 1.8068787515011382,
      "language_loss": 0.74892652,
      "learning_rate": 1.5876185173861333e-06,
      "loss": 0.7703045,
      "num_input_tokens_seen": 207477890,
      "step": 9632,
      "time_per_iteration": 2.480193853378296
    },
    {
      "auxiliary_loss_clip": 0.01091008,
      "auxiliary_loss_mlp": 0.01026111,
      "balance_loss_clip": 1.03899837,
      "balance_loss_mlp": 1.0124675,
      "epoch": 0.5791672929505486,
      "flos": 24206521472640.0,
      "grad_norm": 2.012962602706486,
      "language_loss": 0.79316151,
      "learning_rate": 1.5872374323492915e-06,
      "loss": 0.81433266,
      "num_input_tokens_seen": 207497670,
      "step": 9633,
      "time_per_iteration": 2.5568864345550537
    },
    {
      "auxiliary_loss_clip": 0.01082563,
      "auxiliary_loss_mlp": 0.01037819,
      "balance_loss_clip": 1.03884637,
      "balance_loss_mlp": 1.0240978,
      "epoch": 0.5792274162032166,
      "flos": 24348871071360.0,
      "grad_norm": 1.7563188477883869,
      "language_loss": 0.77693903,
      "learning_rate": 1.5868563629644464e-06,
      "loss": 0.79814291,
      "num_input_tokens_seen": 207516105,
      "step": 9634,
      "time_per_iteration": 2.5786101818084717
    },
    {
      "auxiliary_loss_clip": 0.01093814,
      "auxiliary_loss_mlp": 0.01031412,
      "balance_loss_clip": 1.03762794,
      "balance_loss_mlp": 1.01805449,
      "epoch": 0.5792875394558845,
      "flos": 20449403712000.0,
      "grad_norm": 2.1912193904873005,
      "language_loss": 0.63574183,
      "learning_rate": 1.5864753092460502e-06,
      "loss": 0.6569941,
      "num_input_tokens_seen": 207533685,
      "step": 9635,
      "time_per_iteration": 2.5347909927368164
    },
    {
      "auxiliary_loss_clip": 0.01084408,
      "auxiliary_loss_mlp": 0.01034394,
      "balance_loss_clip": 1.03752232,
      "balance_loss_mlp": 1.02185369,
      "epoch": 0.5793476627085525,
      "flos": 24060329118720.0,
      "grad_norm": 1.416642595016995,
      "language_loss": 0.77277547,
      "learning_rate": 1.5860942712085516e-06,
      "loss": 0.79396349,
      "num_input_tokens_seen": 207552840,
      "step": 9636,
      "time_per_iteration": 2.5575127601623535
    },
    {
      "auxiliary_loss_clip": 0.0108202,
      "auxiliary_loss_mlp": 0.01027875,
      "balance_loss_clip": 1.03496385,
      "balance_loss_mlp": 1.01582313,
      "epoch": 0.5794077859612206,
      "flos": 22054287381120.0,
      "grad_norm": 1.9333052842627727,
      "language_loss": 0.68403339,
      "learning_rate": 1.5857132488663998e-06,
      "loss": 0.70513237,
      "num_input_tokens_seen": 207572095,
      "step": 9637,
      "time_per_iteration": 2.542168617248535
    },
    {
      "auxiliary_loss_clip": 0.01069495,
      "auxiliary_loss_mlp": 0.01033839,
      "balance_loss_clip": 1.03677177,
      "balance_loss_mlp": 1.02080941,
      "epoch": 0.5794679092138885,
      "flos": 11434855991040.0,
      "grad_norm": 5.6608969495046,
      "language_loss": 0.72325277,
      "learning_rate": 1.585332242234043e-06,
      "loss": 0.74428618,
      "num_input_tokens_seen": 207587495,
      "step": 9638,
      "time_per_iteration": 2.544402837753296
    },
    {
      "auxiliary_loss_clip": 0.01102764,
      "auxiliary_loss_mlp": 0.0103167,
      "balance_loss_clip": 1.04162478,
      "balance_loss_mlp": 1.01979721,
      "epoch": 0.5795280324665565,
      "flos": 18880215183360.0,
      "grad_norm": 1.6026392766461843,
      "language_loss": 0.72239196,
      "learning_rate": 1.5849512513259291e-06,
      "loss": 0.74373633,
      "num_input_tokens_seen": 207606795,
      "step": 9639,
      "time_per_iteration": 2.4948065280914307
    },
    {
      "auxiliary_loss_clip": 0.01088875,
      "auxiliary_loss_mlp": 0.01032392,
      "balance_loss_clip": 1.03790855,
      "balance_loss_mlp": 1.01978016,
      "epoch": 0.5795881557192244,
      "flos": 13005947940480.0,
      "grad_norm": 9.70793302030323,
      "language_loss": 0.69621074,
      "learning_rate": 1.5845702761565054e-06,
      "loss": 0.71742344,
      "num_input_tokens_seen": 207623620,
      "step": 9640,
      "time_per_iteration": 2.5191168785095215
    },
    {
      "auxiliary_loss_clip": 0.01091223,
      "auxiliary_loss_mlp": 0.01038218,
      "balance_loss_clip": 1.03970647,
      "balance_loss_mlp": 1.02366829,
      "epoch": 0.5796482789718924,
      "flos": 19932397303680.0,
      "grad_norm": 2.4049987717189265,
      "language_loss": 0.77881765,
      "learning_rate": 1.5841893167402183e-06,
      "loss": 0.80011207,
      "num_input_tokens_seen": 207639380,
      "step": 9641,
      "time_per_iteration": 2.5128629207611084
    },
    {
      "auxiliary_loss_clip": 0.01111348,
      "auxiliary_loss_mlp": 0.01034317,
      "balance_loss_clip": 1.03949213,
      "balance_loss_mlp": 1.02190733,
      "epoch": 0.5797084022245603,
      "flos": 21650794928640.0,
      "grad_norm": 1.7042051271300673,
      "language_loss": 0.73442292,
      "learning_rate": 1.5838083730915143e-06,
      "loss": 0.75587958,
      "num_input_tokens_seen": 207657915,
      "step": 9642,
      "time_per_iteration": 2.480365753173828
    },
    {
      "auxiliary_loss_clip": 0.01092534,
      "auxiliary_loss_mlp": 0.01031976,
      "balance_loss_clip": 1.03840709,
      "balance_loss_mlp": 1.01903617,
      "epoch": 0.5797685254772283,
      "flos": 26031573555840.0,
      "grad_norm": 3.7456062141671684,
      "language_loss": 0.73700696,
      "learning_rate": 1.5834274452248378e-06,
      "loss": 0.75825202,
      "num_input_tokens_seen": 207678620,
      "step": 9643,
      "time_per_iteration": 2.5942113399505615
    },
    {
      "auxiliary_loss_clip": 0.0111393,
      "auxiliary_loss_mlp": 0.01029845,
      "balance_loss_clip": 1.03935742,
      "balance_loss_mlp": 1.01707125,
      "epoch": 0.5798286487298963,
      "flos": 22705167778560.0,
      "grad_norm": 2.1441141952492466,
      "language_loss": 0.67371035,
      "learning_rate": 1.5830465331546352e-06,
      "loss": 0.69514811,
      "num_input_tokens_seen": 207696980,
      "step": 9644,
      "time_per_iteration": 2.4940803050994873
    },
    {
      "auxiliary_loss_clip": 0.01107337,
      "auxiliary_loss_mlp": 0.01031855,
      "balance_loss_clip": 1.04122233,
      "balance_loss_mlp": 1.01870656,
      "epoch": 0.5798887719825643,
      "flos": 23148988225920.0,
      "grad_norm": 2.1534512485486723,
      "language_loss": 0.8538307,
      "learning_rate": 1.5826656368953496e-06,
      "loss": 0.87522262,
      "num_input_tokens_seen": 207714065,
      "step": 9645,
      "time_per_iteration": 2.502521514892578
    },
    {
      "auxiliary_loss_clip": 0.01113835,
      "auxiliary_loss_mlp": 0.01034153,
      "balance_loss_clip": 1.04054999,
      "balance_loss_mlp": 1.02183914,
      "epoch": 0.5799488952352322,
      "flos": 24426043441920.0,
      "grad_norm": 2.0125740626461313,
      "language_loss": 0.75746059,
      "learning_rate": 1.5822847564614244e-06,
      "loss": 0.77894044,
      "num_input_tokens_seen": 207734720,
      "step": 9646,
      "time_per_iteration": 2.490468978881836
    },
    {
      "auxiliary_loss_clip": 0.01094095,
      "auxiliary_loss_mlp": 0.01033538,
      "balance_loss_clip": 1.04113126,
      "balance_loss_mlp": 1.01970983,
      "epoch": 0.5800090184879002,
      "flos": 38395903829760.0,
      "grad_norm": 1.7952208947652493,
      "language_loss": 0.59184659,
      "learning_rate": 1.5819038918673038e-06,
      "loss": 0.61312288,
      "num_input_tokens_seen": 207755435,
      "step": 9647,
      "time_per_iteration": 2.6860110759735107
    },
    {
      "auxiliary_loss_clip": 0.01070926,
      "auxiliary_loss_mlp": 0.01044541,
      "balance_loss_clip": 1.03808463,
      "balance_loss_mlp": 1.02869225,
      "epoch": 0.5800691417405681,
      "flos": 19784840232960.0,
      "grad_norm": 1.4926907155484304,
      "language_loss": 0.84346664,
      "learning_rate": 1.5815230431274288e-06,
      "loss": 0.86462122,
      "num_input_tokens_seen": 207773570,
      "step": 9648,
      "time_per_iteration": 2.5891292095184326
    },
    {
      "auxiliary_loss_clip": 0.01033565,
      "auxiliary_loss_mlp": 0.01003602,
      "balance_loss_clip": 1.018767,
      "balance_loss_mlp": 1.00248134,
      "epoch": 0.5801292649932361,
      "flos": 70314565783680.0,
      "grad_norm": 0.8368925951623254,
      "language_loss": 0.63002115,
      "learning_rate": 1.581142210256242e-06,
      "loss": 0.65039277,
      "num_input_tokens_seen": 207830095,
      "step": 9649,
      "time_per_iteration": 4.470088958740234
    },
    {
      "auxiliary_loss_clip": 0.01072201,
      "auxiliary_loss_mlp": 0.01036685,
      "balance_loss_clip": 1.03441966,
      "balance_loss_mlp": 1.02248716,
      "epoch": 0.5801893882459042,
      "flos": 18734812928640.0,
      "grad_norm": 1.9888139046756772,
      "language_loss": 0.8212713,
      "learning_rate": 1.5807613932681857e-06,
      "loss": 0.84236014,
      "num_input_tokens_seen": 207848555,
      "step": 9650,
      "time_per_iteration": 2.545161247253418
    },
    {
      "auxiliary_loss_clip": 0.01082369,
      "auxiliary_loss_mlp": 0.01034652,
      "balance_loss_clip": 1.04259562,
      "balance_loss_mlp": 1.0218128,
      "epoch": 0.5802495114985721,
      "flos": 15596507698560.0,
      "grad_norm": 2.900755113185745,
      "language_loss": 0.77960712,
      "learning_rate": 1.580380592177698e-06,
      "loss": 0.80077732,
      "num_input_tokens_seen": 207867060,
      "step": 9651,
      "time_per_iteration": 2.577195882797241
    },
    {
      "auxiliary_loss_clip": 0.01095074,
      "auxiliary_loss_mlp": 0.01040219,
      "balance_loss_clip": 1.0403657,
      "balance_loss_mlp": 1.02672482,
      "epoch": 0.5803096347512401,
      "flos": 18255405081600.0,
      "grad_norm": 2.0172398935791156,
      "language_loss": 0.74172199,
      "learning_rate": 1.5799998069992213e-06,
      "loss": 0.76307487,
      "num_input_tokens_seen": 207884520,
      "step": 9652,
      "time_per_iteration": 2.5115127563476562
    },
    {
      "auxiliary_loss_clip": 0.01093659,
      "auxiliary_loss_mlp": 0.01033447,
      "balance_loss_clip": 1.03852367,
      "balance_loss_mlp": 1.01991069,
      "epoch": 0.580369758003908,
      "flos": 22893160584960.0,
      "grad_norm": 2.1188365206042996,
      "language_loss": 0.76770031,
      "learning_rate": 1.579619037747193e-06,
      "loss": 0.78897142,
      "num_input_tokens_seen": 207905370,
      "step": 9653,
      "time_per_iteration": 2.577439785003662
    },
    {
      "auxiliary_loss_clip": 0.01113134,
      "auxiliary_loss_mlp": 0.01034671,
      "balance_loss_clip": 1.03919435,
      "balance_loss_mlp": 1.02066982,
      "epoch": 0.580429881256576,
      "flos": 18697681244160.0,
      "grad_norm": 3.1285110799308873,
      "language_loss": 0.74182719,
      "learning_rate": 1.5792382844360534e-06,
      "loss": 0.76330525,
      "num_input_tokens_seen": 207923790,
      "step": 9654,
      "time_per_iteration": 2.4604549407958984
    },
    {
      "auxiliary_loss_clip": 0.01048238,
      "auxiliary_loss_mlp": 0.01034319,
      "balance_loss_clip": 1.04043448,
      "balance_loss_mlp": 1.0219456,
      "epoch": 0.5804900045092439,
      "flos": 24681978823680.0,
      "grad_norm": 1.865068539020379,
      "language_loss": 0.70127845,
      "learning_rate": 1.5788575470802408e-06,
      "loss": 0.72210407,
      "num_input_tokens_seen": 207942335,
      "step": 9655,
      "time_per_iteration": 2.6621060371398926
    },
    {
      "auxiliary_loss_clip": 0.01116148,
      "auxiliary_loss_mlp": 0.01036488,
      "balance_loss_clip": 1.03818643,
      "balance_loss_mlp": 1.02329707,
      "epoch": 0.580550127761912,
      "flos": 23112790295040.0,
      "grad_norm": 1.9393951827402272,
      "language_loss": 0.6920054,
      "learning_rate": 1.5784768256941915e-06,
      "loss": 0.71353173,
      "num_input_tokens_seen": 207961975,
      "step": 9656,
      "time_per_iteration": 3.8509809970855713
    },
    {
      "auxiliary_loss_clip": 0.01095543,
      "auxiliary_loss_mlp": 0.010313,
      "balance_loss_clip": 1.03799713,
      "balance_loss_mlp": 1.01908755,
      "epoch": 0.5806102510145799,
      "flos": 18475681236480.0,
      "grad_norm": 1.6292756304836526,
      "language_loss": 0.71927881,
      "learning_rate": 1.5780961202923433e-06,
      "loss": 0.74054724,
      "num_input_tokens_seen": 207979520,
      "step": 9657,
      "time_per_iteration": 2.4866409301757812
    },
    {
      "auxiliary_loss_clip": 0.01103426,
      "auxiliary_loss_mlp": 0.01033571,
      "balance_loss_clip": 1.03936303,
      "balance_loss_mlp": 1.0198915,
      "epoch": 0.5806703742672479,
      "flos": 23915645136000.0,
      "grad_norm": 1.9859342223916547,
      "language_loss": 0.70817888,
      "learning_rate": 1.5777154308891328e-06,
      "loss": 0.72954881,
      "num_input_tokens_seen": 207998375,
      "step": 9658,
      "time_per_iteration": 2.561589002609253
    },
    {
      "auxiliary_loss_clip": 0.01029212,
      "auxiliary_loss_mlp": 0.01005754,
      "balance_loss_clip": 1.01462007,
      "balance_loss_mlp": 1.00457406,
      "epoch": 0.5807304975199158,
      "flos": 66311999412480.0,
      "grad_norm": 0.6521085049930855,
      "language_loss": 0.5355624,
      "learning_rate": 1.5773347574989953e-06,
      "loss": 0.55591214,
      "num_input_tokens_seen": 208060605,
      "step": 9659,
      "time_per_iteration": 4.5142662525177
    },
    {
      "auxiliary_loss_clip": 0.01104818,
      "auxiliary_loss_mlp": 0.0104022,
      "balance_loss_clip": 1.04019094,
      "balance_loss_mlp": 1.02672529,
      "epoch": 0.5807906207725838,
      "flos": 31722444933120.0,
      "grad_norm": 2.011363366137313,
      "language_loss": 0.62232268,
      "learning_rate": 1.576954100136366e-06,
      "loss": 0.64377308,
      "num_input_tokens_seen": 208080320,
      "step": 9660,
      "time_per_iteration": 2.5697338581085205
    },
    {
      "auxiliary_loss_clip": 0.01099887,
      "auxiliary_loss_mlp": 0.01035638,
      "balance_loss_clip": 1.03493285,
      "balance_loss_mlp": 1.02226257,
      "epoch": 0.5808507440252517,
      "flos": 23801161512960.0,
      "grad_norm": 1.4287554787905368,
      "language_loss": 0.65328401,
      "learning_rate": 1.5765734588156797e-06,
      "loss": 0.67463923,
      "num_input_tokens_seen": 208099305,
      "step": 9661,
      "time_per_iteration": 2.5479750633239746
    },
    {
      "auxiliary_loss_clip": 0.01059065,
      "auxiliary_loss_mlp": 0.01028042,
      "balance_loss_clip": 1.03673077,
      "balance_loss_mlp": 1.01663911,
      "epoch": 0.5809108672779197,
      "flos": 13698449222400.0,
      "grad_norm": 1.5159816296374204,
      "language_loss": 0.74355787,
      "learning_rate": 1.5761928335513704e-06,
      "loss": 0.76442897,
      "num_input_tokens_seen": 208116960,
      "step": 9662,
      "time_per_iteration": 2.6193628311157227
    },
    {
      "auxiliary_loss_clip": 0.01039112,
      "auxiliary_loss_mlp": 0.00999901,
      "balance_loss_clip": 1.01458621,
      "balance_loss_mlp": 0.99867862,
      "epoch": 0.5809709905305876,
      "flos": 69134866381440.0,
      "grad_norm": 0.8750482968604066,
      "language_loss": 0.58327734,
      "learning_rate": 1.5758122243578709e-06,
      "loss": 0.60366744,
      "num_input_tokens_seen": 208182190,
      "step": 9663,
      "time_per_iteration": 3.1766486167907715
    },
    {
      "auxiliary_loss_clip": 0.01090779,
      "auxiliary_loss_mlp": 0.01028796,
      "balance_loss_clip": 1.03814459,
      "balance_loss_mlp": 1.01643968,
      "epoch": 0.5810311137832557,
      "flos": 19827538525440.0,
      "grad_norm": 2.10974465886977,
      "language_loss": 0.81897408,
      "learning_rate": 1.5754316312496152e-06,
      "loss": 0.84016979,
      "num_input_tokens_seen": 208197015,
      "step": 9664,
      "time_per_iteration": 3.943603754043579
    },
    {
      "auxiliary_loss_clip": 0.01089299,
      "auxiliary_loss_mlp": 0.00787229,
      "balance_loss_clip": 1.03435671,
      "balance_loss_mlp": 1.00810921,
      "epoch": 0.5810912370359237,
      "flos": 29238503719680.0,
      "grad_norm": 1.6760001109891884,
      "language_loss": 0.81738728,
      "learning_rate": 1.5750510542410337e-06,
      "loss": 0.83615255,
      "num_input_tokens_seen": 208215795,
      "step": 9665,
      "time_per_iteration": 2.6040873527526855
    },
    {
      "auxiliary_loss_clip": 0.01093465,
      "auxiliary_loss_mlp": 0.01033162,
      "balance_loss_clip": 1.03939962,
      "balance_loss_mlp": 1.01868391,
      "epoch": 0.5811513602885916,
      "flos": 22785572373120.0,
      "grad_norm": 1.5774222070234176,
      "language_loss": 0.8126744,
      "learning_rate": 1.5746704933465599e-06,
      "loss": 0.83394074,
      "num_input_tokens_seen": 208234655,
      "step": 9666,
      "time_per_iteration": 2.575927734375
    },
    {
      "auxiliary_loss_clip": 0.01097827,
      "auxiliary_loss_mlp": 0.01036237,
      "balance_loss_clip": 1.04124784,
      "balance_loss_mlp": 1.02433944,
      "epoch": 0.5812114835412596,
      "flos": 18734346051840.0,
      "grad_norm": 2.7661009622485686,
      "language_loss": 0.8019889,
      "learning_rate": 1.5742899485806227e-06,
      "loss": 0.82332957,
      "num_input_tokens_seen": 208251300,
      "step": 9667,
      "time_per_iteration": 2.489210367202759
    },
    {
      "auxiliary_loss_clip": 0.01106145,
      "auxiliary_loss_mlp": 0.0103396,
      "balance_loss_clip": 1.03772783,
      "balance_loss_mlp": 1.02032232,
      "epoch": 0.5812716067939275,
      "flos": 26431295080320.0,
      "grad_norm": 1.613827804355796,
      "language_loss": 0.78760469,
      "learning_rate": 1.573909419957653e-06,
      "loss": 0.80900574,
      "num_input_tokens_seen": 208272685,
      "step": 9668,
      "time_per_iteration": 2.576481342315674
    },
    {
      "auxiliary_loss_clip": 0.01090393,
      "auxiliary_loss_mlp": 0.01029848,
      "balance_loss_clip": 1.03876138,
      "balance_loss_mlp": 1.01785016,
      "epoch": 0.5813317300465956,
      "flos": 43397865285120.0,
      "grad_norm": 1.9330845303390036,
      "language_loss": 0.6447556,
      "learning_rate": 1.5735289074920819e-06,
      "loss": 0.66595793,
      "num_input_tokens_seen": 208294315,
      "step": 9669,
      "time_per_iteration": 2.7305614948272705
    },
    {
      "auxiliary_loss_clip": 0.01062096,
      "auxiliary_loss_mlp": 0.01033377,
      "balance_loss_clip": 1.03899097,
      "balance_loss_mlp": 1.02034712,
      "epoch": 0.5813918532992635,
      "flos": 24785472885120.0,
      "grad_norm": 1.5323161940272176,
      "language_loss": 0.73099363,
      "learning_rate": 1.5731484111983363e-06,
      "loss": 0.75194836,
      "num_input_tokens_seen": 208315610,
      "step": 9670,
      "time_per_iteration": 2.6420578956604004
    },
    {
      "auxiliary_loss_clip": 0.01081181,
      "auxiliary_loss_mlp": 0.01041479,
      "balance_loss_clip": 1.03766251,
      "balance_loss_mlp": 1.02876544,
      "epoch": 0.5814519765519315,
      "flos": 22857357703680.0,
      "grad_norm": 1.9984474454413688,
      "language_loss": 0.78904748,
      "learning_rate": 1.5727679310908464e-06,
      "loss": 0.810274,
      "num_input_tokens_seen": 208334725,
      "step": 9671,
      "time_per_iteration": 2.5628437995910645
    },
    {
      "auxiliary_loss_clip": 0.0106911,
      "auxiliary_loss_mlp": 0.01032912,
      "balance_loss_clip": 1.03729868,
      "balance_loss_mlp": 1.01895297,
      "epoch": 0.5815120998045994,
      "flos": 24060831909120.0,
      "grad_norm": 2.081288871488614,
      "language_loss": 0.61383343,
      "learning_rate": 1.5723874671840399e-06,
      "loss": 0.6348536,
      "num_input_tokens_seen": 208353825,
      "step": 9672,
      "time_per_iteration": 2.6273372173309326
    },
    {
      "auxiliary_loss_clip": 0.01066832,
      "auxiliary_loss_mlp": 0.01040701,
      "balance_loss_clip": 1.03763509,
      "balance_loss_mlp": 1.02736163,
      "epoch": 0.5815722230572674,
      "flos": 24279491952000.0,
      "grad_norm": 1.6354908146370508,
      "language_loss": 0.81136405,
      "learning_rate": 1.572007019492342e-06,
      "loss": 0.83243942,
      "num_input_tokens_seen": 208374160,
      "step": 9673,
      "time_per_iteration": 2.6280763149261475
    },
    {
      "auxiliary_loss_clip": 0.01081723,
      "auxiliary_loss_mlp": 0.01035208,
      "balance_loss_clip": 1.04130912,
      "balance_loss_mlp": 1.02135026,
      "epoch": 0.5816323463099353,
      "flos": 22200371994240.0,
      "grad_norm": 1.9092854713163045,
      "language_loss": 0.88017917,
      "learning_rate": 1.5716265880301817e-06,
      "loss": 0.90134847,
      "num_input_tokens_seen": 208392105,
      "step": 9674,
      "time_per_iteration": 2.558434247970581
    },
    {
      "auxiliary_loss_clip": 0.01111774,
      "auxiliary_loss_mlp": 0.0078829,
      "balance_loss_clip": 1.03817892,
      "balance_loss_mlp": 1.01314139,
      "epoch": 0.5816924695626033,
      "flos": 24134448833280.0,
      "grad_norm": 1.528446293420111,
      "language_loss": 0.79174829,
      "learning_rate": 1.571246172811984e-06,
      "loss": 0.81074893,
      "num_input_tokens_seen": 208411755,
      "step": 9675,
      "time_per_iteration": 2.5101263523101807
    },
    {
      "auxiliary_loss_clip": 0.01100106,
      "auxiliary_loss_mlp": 0.01033619,
      "balance_loss_clip": 1.0388186,
      "balance_loss_mlp": 1.02006531,
      "epoch": 0.5817525928152713,
      "flos": 21324223451520.0,
      "grad_norm": 2.1326498867658046,
      "language_loss": 0.70043015,
      "learning_rate": 1.5708657738521748e-06,
      "loss": 0.72176743,
      "num_input_tokens_seen": 208429995,
      "step": 9676,
      "time_per_iteration": 2.485180377960205
    },
    {
      "auxiliary_loss_clip": 0.01059926,
      "auxiliary_loss_mlp": 0.01028928,
      "balance_loss_clip": 1.03895807,
      "balance_loss_mlp": 1.01570141,
      "epoch": 0.5818127160679393,
      "flos": 26934510666240.0,
      "grad_norm": 2.1910144496441735,
      "language_loss": 0.63410819,
      "learning_rate": 1.5704853911651779e-06,
      "loss": 0.65499669,
      "num_input_tokens_seen": 208443655,
      "step": 9677,
      "time_per_iteration": 2.646524667739868
    },
    {
      "auxiliary_loss_clip": 0.01023462,
      "auxiliary_loss_mlp": 0.01001475,
      "balance_loss_clip": 1.02554035,
      "balance_loss_mlp": 1.00030637,
      "epoch": 0.5818728393206073,
      "flos": 63918626342400.0,
      "grad_norm": 0.8016440887987955,
      "language_loss": 0.54147637,
      "learning_rate": 1.5701050247654182e-06,
      "loss": 0.56172574,
      "num_input_tokens_seen": 208498405,
      "step": 9678,
      "time_per_iteration": 3.207714080810547
    },
    {
      "auxiliary_loss_clip": 0.01025672,
      "auxiliary_loss_mlp": 0.01007996,
      "balance_loss_clip": 1.02118897,
      "balance_loss_mlp": 1.00641024,
      "epoch": 0.5819329625732752,
      "flos": 64954108638720.0,
      "grad_norm": 0.7440782150046145,
      "language_loss": 0.56239396,
      "learning_rate": 1.569724674667319e-06,
      "loss": 0.58273065,
      "num_input_tokens_seen": 208559075,
      "step": 9679,
      "time_per_iteration": 3.0381968021392822
    },
    {
      "auxiliary_loss_clip": 0.01109972,
      "auxiliary_loss_mlp": 0.01030784,
      "balance_loss_clip": 1.03743422,
      "balance_loss_mlp": 1.01917338,
      "epoch": 0.5819930858259432,
      "flos": 21215270522880.0,
      "grad_norm": 1.737877709267846,
      "language_loss": 0.65344119,
      "learning_rate": 1.5693443408853032e-06,
      "loss": 0.67484879,
      "num_input_tokens_seen": 208577770,
      "step": 9680,
      "time_per_iteration": 2.499100923538208
    },
    {
      "auxiliary_loss_clip": 0.0108846,
      "auxiliary_loss_mlp": 0.01027264,
      "balance_loss_clip": 1.0373317,
      "balance_loss_mlp": 1.01548588,
      "epoch": 0.5820532090786111,
      "flos": 19458520151040.0,
      "grad_norm": 1.977328511850355,
      "language_loss": 0.83486295,
      "learning_rate": 1.5689640234337933e-06,
      "loss": 0.85602015,
      "num_input_tokens_seen": 208595110,
      "step": 9681,
      "time_per_iteration": 2.534601926803589
    },
    {
      "auxiliary_loss_clip": 0.01110927,
      "auxiliary_loss_mlp": 0.01028284,
      "balance_loss_clip": 1.03819335,
      "balance_loss_mlp": 1.01600552,
      "epoch": 0.5821133323312792,
      "flos": 17712615686400.0,
      "grad_norm": 1.8546241476533565,
      "language_loss": 0.75860071,
      "learning_rate": 1.5685837223272109e-06,
      "loss": 0.77999282,
      "num_input_tokens_seen": 208612080,
      "step": 9682,
      "time_per_iteration": 2.4762351512908936
    },
    {
      "auxiliary_loss_clip": 0.01056264,
      "auxiliary_loss_mlp": 0.01035487,
      "balance_loss_clip": 1.03525233,
      "balance_loss_mlp": 1.02121806,
      "epoch": 0.5821734555839471,
      "flos": 24571804832640.0,
      "grad_norm": 1.8968213965619711,
      "language_loss": 0.74988681,
      "learning_rate": 1.568203437579977e-06,
      "loss": 0.77080429,
      "num_input_tokens_seen": 208630235,
      "step": 9683,
      "time_per_iteration": 2.6503167152404785
    },
    {
      "auxiliary_loss_clip": 0.0108717,
      "auxiliary_loss_mlp": 0.01030351,
      "balance_loss_clip": 1.03854156,
      "balance_loss_mlp": 1.01766098,
      "epoch": 0.5822335788366151,
      "flos": 22382259488640.0,
      "grad_norm": 1.720758563175461,
      "language_loss": 0.74009919,
      "learning_rate": 1.5678231692065116e-06,
      "loss": 0.7612744,
      "num_input_tokens_seen": 208647925,
      "step": 9684,
      "time_per_iteration": 2.5448365211486816
    },
    {
      "auxiliary_loss_clip": 0.01093407,
      "auxiliary_loss_mlp": 0.01038203,
      "balance_loss_clip": 1.03911161,
      "balance_loss_mlp": 1.02488124,
      "epoch": 0.582293702089283,
      "flos": 26722494639360.0,
      "grad_norm": 2.4591071958590414,
      "language_loss": 0.78578782,
      "learning_rate": 1.5674429172212348e-06,
      "loss": 0.80710393,
      "num_input_tokens_seen": 208666180,
      "step": 9685,
      "time_per_iteration": 2.5617973804473877
    },
    {
      "auxiliary_loss_clip": 0.01111417,
      "auxiliary_loss_mlp": 0.01033656,
      "balance_loss_clip": 1.03905463,
      "balance_loss_mlp": 1.02087688,
      "epoch": 0.582353825341951,
      "flos": 17348661129600.0,
      "grad_norm": 1.8532190127315478,
      "language_loss": 0.75009811,
      "learning_rate": 1.5670626816385667e-06,
      "loss": 0.77154881,
      "num_input_tokens_seen": 208684240,
      "step": 9686,
      "time_per_iteration": 2.444061756134033
    },
    {
      "auxiliary_loss_clip": 0.0102923,
      "auxiliary_loss_mlp": 0.01021969,
      "balance_loss_clip": 1.01453114,
      "balance_loss_mlp": 1.02086663,
      "epoch": 0.5824139485946189,
      "flos": 55473261534720.0,
      "grad_norm": 0.8257090074653991,
      "language_loss": 0.57389104,
      "learning_rate": 1.5666824624729244e-06,
      "loss": 0.59440303,
      "num_input_tokens_seen": 208736090,
      "step": 9687,
      "time_per_iteration": 4.245602607727051
    },
    {
      "auxiliary_loss_clip": 0.01072114,
      "auxiliary_loss_mlp": 0.01030604,
      "balance_loss_clip": 1.03698444,
      "balance_loss_mlp": 1.0161916,
      "epoch": 0.582474071847287,
      "flos": 20303031790080.0,
      "grad_norm": 3.375118318481554,
      "language_loss": 0.70135355,
      "learning_rate": 1.566302259738727e-06,
      "loss": 0.72238076,
      "num_input_tokens_seen": 208754600,
      "step": 9688,
      "time_per_iteration": 2.5675840377807617
    },
    {
      "auxiliary_loss_clip": 0.01101622,
      "auxiliary_loss_mlp": 0.01031143,
      "balance_loss_clip": 1.03738344,
      "balance_loss_mlp": 1.01889467,
      "epoch": 0.5825341950999549,
      "flos": 23878010661120.0,
      "grad_norm": 2.0451778245297745,
      "language_loss": 0.6478132,
      "learning_rate": 1.5659220734503918e-06,
      "loss": 0.66914082,
      "num_input_tokens_seen": 208773140,
      "step": 9689,
      "time_per_iteration": 2.509417772293091
    },
    {
      "auxiliary_loss_clip": 0.01090404,
      "auxiliary_loss_mlp": 0.0078717,
      "balance_loss_clip": 1.03882194,
      "balance_loss_mlp": 1.01166737,
      "epoch": 0.5825943183526229,
      "flos": 23113041690240.0,
      "grad_norm": 1.851121758499308,
      "language_loss": 0.73272884,
      "learning_rate": 1.5655419036223341e-06,
      "loss": 0.75150454,
      "num_input_tokens_seen": 208793410,
      "step": 9690,
      "time_per_iteration": 2.553427219390869
    },
    {
      "auxiliary_loss_clip": 0.01089254,
      "auxiliary_loss_mlp": 0.01038313,
      "balance_loss_clip": 1.0366919,
      "balance_loss_mlp": 1.02388895,
      "epoch": 0.5826544416052909,
      "flos": 22857429530880.0,
      "grad_norm": 1.692964391838581,
      "language_loss": 0.75841761,
      "learning_rate": 1.5651617502689717e-06,
      "loss": 0.77969325,
      "num_input_tokens_seen": 208811920,
      "step": 9691,
      "time_per_iteration": 2.5359082221984863
    },
    {
      "auxiliary_loss_clip": 0.01101382,
      "auxiliary_loss_mlp": 0.01029231,
      "balance_loss_clip": 1.03681457,
      "balance_loss_mlp": 1.01633239,
      "epoch": 0.5827145648579588,
      "flos": 31501845555840.0,
      "grad_norm": 1.7634569892114038,
      "language_loss": 0.8062728,
      "learning_rate": 1.5647816134047184e-06,
      "loss": 0.8275789,
      "num_input_tokens_seen": 208834720,
      "step": 9692,
      "time_per_iteration": 2.572535991668701
    },
    {
      "auxiliary_loss_clip": 0.01028018,
      "auxiliary_loss_mlp": 0.01009247,
      "balance_loss_clip": 1.01335144,
      "balance_loss_mlp": 1.00808454,
      "epoch": 0.5827746881106268,
      "flos": 69811817074560.0,
      "grad_norm": 0.7589954772475821,
      "language_loss": 0.56916434,
      "learning_rate": 1.5644014930439907e-06,
      "loss": 0.58953696,
      "num_input_tokens_seen": 208898415,
      "step": 9693,
      "time_per_iteration": 3.0765902996063232
    },
    {
      "auxiliary_loss_clip": 0.01096518,
      "auxiliary_loss_mlp": 0.00786404,
      "balance_loss_clip": 1.03533256,
      "balance_loss_mlp": 1.01185715,
      "epoch": 0.5828348113632947,
      "flos": 23112395245440.0,
      "grad_norm": 1.6007685806698053,
      "language_loss": 0.79004842,
      "learning_rate": 1.5640213892012025e-06,
      "loss": 0.80887765,
      "num_input_tokens_seen": 208919045,
      "step": 9694,
      "time_per_iteration": 3.8807618618011475
    },
    {
      "auxiliary_loss_clip": 0.01078851,
      "auxiliary_loss_mlp": 0.01035754,
      "balance_loss_clip": 1.03770614,
      "balance_loss_mlp": 1.02243245,
      "epoch": 0.5828949346159628,
      "flos": 21873082245120.0,
      "grad_norm": 1.5250935060362156,
      "language_loss": 0.76176476,
      "learning_rate": 1.5636413018907656e-06,
      "loss": 0.78291082,
      "num_input_tokens_seen": 208939375,
      "step": 9695,
      "time_per_iteration": 2.5749073028564453
    },
    {
      "auxiliary_loss_clip": 0.01027179,
      "auxiliary_loss_mlp": 0.01003175,
      "balance_loss_clip": 1.01258159,
      "balance_loss_mlp": 1.00198925,
      "epoch": 0.5829550578686307,
      "flos": 65962553950080.0,
      "grad_norm": 0.7722824934367444,
      "language_loss": 0.55043417,
      "learning_rate": 1.563261231127095e-06,
      "loss": 0.57073772,
      "num_input_tokens_seen": 209004760,
      "step": 9696,
      "time_per_iteration": 3.2099177837371826
    },
    {
      "auxiliary_loss_clip": 0.01075708,
      "auxiliary_loss_mlp": 0.01029282,
      "balance_loss_clip": 1.03986526,
      "balance_loss_mlp": 1.01700878,
      "epoch": 0.5830151811212987,
      "flos": 16289799079680.0,
      "grad_norm": 1.8591119985358442,
      "language_loss": 0.76430261,
      "learning_rate": 1.5628811769246021e-06,
      "loss": 0.78535247,
      "num_input_tokens_seen": 209022930,
      "step": 9697,
      "time_per_iteration": 2.5392792224884033
    },
    {
      "auxiliary_loss_clip": 0.01113123,
      "auxiliary_loss_mlp": 0.01032122,
      "balance_loss_clip": 1.03758454,
      "balance_loss_mlp": 1.0188067,
      "epoch": 0.5830753043739666,
      "flos": 24168851084160.0,
      "grad_norm": 1.699767819917473,
      "language_loss": 0.77386546,
      "learning_rate": 1.5625011392976991e-06,
      "loss": 0.79531789,
      "num_input_tokens_seen": 209043740,
      "step": 9698,
      "time_per_iteration": 3.921567916870117
    },
    {
      "auxiliary_loss_clip": 0.01071583,
      "auxiliary_loss_mlp": 0.01035593,
      "balance_loss_clip": 1.03712368,
      "balance_loss_mlp": 1.02222395,
      "epoch": 0.5831354276266346,
      "flos": 27059050097280.0,
      "grad_norm": 1.6084596605716308,
      "language_loss": 0.83392954,
      "learning_rate": 1.5621211182607966e-06,
      "loss": 0.85500133,
      "num_input_tokens_seen": 209068885,
      "step": 9699,
      "time_per_iteration": 2.712510347366333
    },
    {
      "auxiliary_loss_clip": 0.01087192,
      "auxiliary_loss_mlp": 0.01033532,
      "balance_loss_clip": 1.03641319,
      "balance_loss_mlp": 1.01931024,
      "epoch": 0.5831955508793025,
      "flos": 23623475909760.0,
      "grad_norm": 2.0176714780087073,
      "language_loss": 0.66298932,
      "learning_rate": 1.561741113828305e-06,
      "loss": 0.68419659,
      "num_input_tokens_seen": 209087340,
      "step": 9700,
      "time_per_iteration": 2.5855231285095215
    },
    {
      "auxiliary_loss_clip": 0.0110083,
      "auxiliary_loss_mlp": 0.0103192,
      "balance_loss_clip": 1.03688288,
      "balance_loss_mlp": 1.01903987,
      "epoch": 0.5832556741319705,
      "flos": 24973250209920.0,
      "grad_norm": 1.5919089363235128,
      "language_loss": 0.71417332,
      "learning_rate": 1.5613611260146344e-06,
      "loss": 0.73550075,
      "num_input_tokens_seen": 209108840,
      "step": 9701,
      "time_per_iteration": 2.552030563354492
    },
    {
      "auxiliary_loss_clip": 0.0108613,
      "auxiliary_loss_mlp": 0.01031428,
      "balance_loss_clip": 1.03619194,
      "balance_loss_mlp": 1.01907206,
      "epoch": 0.5833157973846385,
      "flos": 23221563655680.0,
      "grad_norm": 1.783521519410721,
      "language_loss": 0.85411823,
      "learning_rate": 1.5609811548341936e-06,
      "loss": 0.87529385,
      "num_input_tokens_seen": 209127985,
      "step": 9702,
      "time_per_iteration": 4.058727502822876
    },
    {
      "auxiliary_loss_clip": 0.01093036,
      "auxiliary_loss_mlp": 0.0103002,
      "balance_loss_clip": 1.03486216,
      "balance_loss_mlp": 1.01831973,
      "epoch": 0.5833759206373065,
      "flos": 21977941023360.0,
      "grad_norm": 1.5216422300525767,
      "language_loss": 0.77711606,
      "learning_rate": 1.560601200301392e-06,
      "loss": 0.79834664,
      "num_input_tokens_seen": 209146885,
      "step": 9703,
      "time_per_iteration": 2.5280017852783203
    },
    {
      "auxiliary_loss_clip": 0.01112103,
      "auxiliary_loss_mlp": 0.01029884,
      "balance_loss_clip": 1.03744507,
      "balance_loss_mlp": 1.01686001,
      "epoch": 0.5834360438899745,
      "flos": 21762405463680.0,
      "grad_norm": 4.0664772205101105,
      "language_loss": 0.71134675,
      "learning_rate": 1.5602212624306366e-06,
      "loss": 0.73276663,
      "num_input_tokens_seen": 209166130,
      "step": 9704,
      "time_per_iteration": 2.4972457885742188
    },
    {
      "auxiliary_loss_clip": 0.01088403,
      "auxiliary_loss_mlp": 0.01031322,
      "balance_loss_clip": 1.04265368,
      "balance_loss_mlp": 1.01949692,
      "epoch": 0.5834961671426424,
      "flos": 15992566035840.0,
      "grad_norm": 1.729744324478098,
      "language_loss": 0.81618983,
      "learning_rate": 1.559841341236335e-06,
      "loss": 0.83738708,
      "num_input_tokens_seen": 209183350,
      "step": 9705,
      "time_per_iteration": 2.5404393672943115
    },
    {
      "auxiliary_loss_clip": 0.0106489,
      "auxiliary_loss_mlp": 0.01030431,
      "balance_loss_clip": 1.03608203,
      "balance_loss_mlp": 1.01775277,
      "epoch": 0.5835562903953104,
      "flos": 22818322598400.0,
      "grad_norm": 2.083095865815948,
      "language_loss": 0.8053838,
      "learning_rate": 1.5594614367328937e-06,
      "loss": 0.82633698,
      "num_input_tokens_seen": 209203945,
      "step": 9706,
      "time_per_iteration": 2.585801839828491
    },
    {
      "auxiliary_loss_clip": 0.01097794,
      "auxiliary_loss_mlp": 0.0103242,
      "balance_loss_clip": 1.03742778,
      "balance_loss_mlp": 1.01940238,
      "epoch": 0.5836164136479783,
      "flos": 48468056624640.0,
      "grad_norm": 2.477424665590362,
      "language_loss": 0.75106394,
      "learning_rate": 1.5590815489347187e-06,
      "loss": 0.77236605,
      "num_input_tokens_seen": 209227080,
      "step": 9707,
      "time_per_iteration": 2.732780933380127
    },
    {
      "auxiliary_loss_clip": 0.01079921,
      "auxiliary_loss_mlp": 0.01031397,
      "balance_loss_clip": 1.03913832,
      "balance_loss_mlp": 1.01921368,
      "epoch": 0.5836765369006464,
      "flos": 26905998245760.0,
      "grad_norm": 1.6486625773042023,
      "language_loss": 0.81566226,
      "learning_rate": 1.5587016778562163e-06,
      "loss": 0.83677542,
      "num_input_tokens_seen": 209248170,
      "step": 9708,
      "time_per_iteration": 2.6179630756378174
    },
    {
      "auxiliary_loss_clip": 0.0109623,
      "auxiliary_loss_mlp": 0.01031008,
      "balance_loss_clip": 1.04087853,
      "balance_loss_mlp": 1.01850891,
      "epoch": 0.5837366601533143,
      "flos": 20084048524800.0,
      "grad_norm": 1.5024775154525496,
      "language_loss": 0.78799313,
      "learning_rate": 1.5583218235117896e-06,
      "loss": 0.80926549,
      "num_input_tokens_seen": 209267730,
      "step": 9709,
      "time_per_iteration": 2.526174306869507
    },
    {
      "auxiliary_loss_clip": 0.01017574,
      "auxiliary_loss_mlp": 0.01005849,
      "balance_loss_clip": 1.0118804,
      "balance_loss_mlp": 1.00463891,
      "epoch": 0.5837967834059823,
      "flos": 65363885971200.0,
      "grad_norm": 0.7618155826030154,
      "language_loss": 0.56576705,
      "learning_rate": 1.557941985915844e-06,
      "loss": 0.58600128,
      "num_input_tokens_seen": 209332510,
      "step": 9710,
      "time_per_iteration": 3.146564245223999
    },
    {
      "auxiliary_loss_clip": 0.01077113,
      "auxiliary_loss_mlp": 0.01028621,
      "balance_loss_clip": 1.0378933,
      "balance_loss_mlp": 1.01740968,
      "epoch": 0.5838569066586502,
      "flos": 25338641310720.0,
      "grad_norm": 1.5541538853730885,
      "language_loss": 0.6531918,
      "learning_rate": 1.5575621650827833e-06,
      "loss": 0.67424905,
      "num_input_tokens_seen": 209353355,
      "step": 9711,
      "time_per_iteration": 2.6623027324676514
    },
    {
      "auxiliary_loss_clip": 0.01113468,
      "auxiliary_loss_mlp": 0.01034006,
      "balance_loss_clip": 1.03689349,
      "balance_loss_mlp": 1.01981401,
      "epoch": 0.5839170299113182,
      "flos": 22229243550720.0,
      "grad_norm": 1.773640185748023,
      "language_loss": 0.78762579,
      "learning_rate": 1.5571823610270085e-06,
      "loss": 0.80910051,
      "num_input_tokens_seen": 209370960,
      "step": 9712,
      "time_per_iteration": 2.488757610321045
    },
    {
      "auxiliary_loss_clip": 0.01072236,
      "auxiliary_loss_mlp": 0.00786924,
      "balance_loss_clip": 1.03372228,
      "balance_loss_mlp": 1.00890756,
      "epoch": 0.5839771531639861,
      "flos": 22200012858240.0,
      "grad_norm": 1.504094943550871,
      "language_loss": 0.7316227,
      "learning_rate": 1.5568025737629234e-06,
      "loss": 0.75021422,
      "num_input_tokens_seen": 209390955,
      "step": 9713,
      "time_per_iteration": 2.6179497241973877
    },
    {
      "auxiliary_loss_clip": 0.01091146,
      "auxiliary_loss_mlp": 0.01031076,
      "balance_loss_clip": 1.03654754,
      "balance_loss_mlp": 1.0171169,
      "epoch": 0.5840372764166541,
      "flos": 22419355259520.0,
      "grad_norm": 1.7942594156588376,
      "language_loss": 0.69045496,
      "learning_rate": 1.5564228033049292e-06,
      "loss": 0.71167719,
      "num_input_tokens_seen": 209410260,
      "step": 9714,
      "time_per_iteration": 2.540234327316284
    },
    {
      "auxiliary_loss_clip": 0.01109877,
      "auxiliary_loss_mlp": 0.01030724,
      "balance_loss_clip": 1.03588855,
      "balance_loss_mlp": 1.01775432,
      "epoch": 0.5840973996693221,
      "flos": 19828256797440.0,
      "grad_norm": 1.8622988785938341,
      "language_loss": 0.8018434,
      "learning_rate": 1.5560430496674268e-06,
      "loss": 0.8232494,
      "num_input_tokens_seen": 209429920,
      "step": 9715,
      "time_per_iteration": 2.478945255279541
    },
    {
      "auxiliary_loss_clip": 0.01088122,
      "auxiliary_loss_mlp": 0.01033787,
      "balance_loss_clip": 1.03425694,
      "balance_loss_mlp": 1.02055442,
      "epoch": 0.5841575229219901,
      "flos": 21142982401920.0,
      "grad_norm": 2.224598048742477,
      "language_loss": 0.72166967,
      "learning_rate": 1.5556633128648167e-06,
      "loss": 0.74288875,
      "num_input_tokens_seen": 209449470,
      "step": 9716,
      "time_per_iteration": 2.552072763442993
    },
    {
      "auxiliary_loss_clip": 0.01083544,
      "auxiliary_loss_mlp": 0.01033659,
      "balance_loss_clip": 1.03606355,
      "balance_loss_mlp": 1.02146339,
      "epoch": 0.5842176461746581,
      "flos": 24640322025600.0,
      "grad_norm": 1.7686244979918377,
      "language_loss": 0.75020474,
      "learning_rate": 1.5552835929114976e-06,
      "loss": 0.77137673,
      "num_input_tokens_seen": 209467695,
      "step": 9717,
      "time_per_iteration": 2.5849974155426025
    },
    {
      "auxiliary_loss_clip": 0.01098651,
      "auxiliary_loss_mlp": 0.01035637,
      "balance_loss_clip": 1.03567743,
      "balance_loss_mlp": 1.02243447,
      "epoch": 0.584277769427326,
      "flos": 19131158574720.0,
      "grad_norm": 2.0323020161954384,
      "language_loss": 0.8008523,
      "learning_rate": 1.5549038898218697e-06,
      "loss": 0.82219517,
      "num_input_tokens_seen": 209484250,
      "step": 9718,
      "time_per_iteration": 2.4527716636657715
    },
    {
      "auxiliary_loss_clip": 0.01086716,
      "auxiliary_loss_mlp": 0.01031093,
      "balance_loss_clip": 1.03591502,
      "balance_loss_mlp": 1.01776576,
      "epoch": 0.584337892679994,
      "flos": 22675111073280.0,
      "grad_norm": 1.7083170752983856,
      "language_loss": 0.67511809,
      "learning_rate": 1.5545242036103306e-06,
      "loss": 0.69629616,
      "num_input_tokens_seen": 209502830,
      "step": 9719,
      "time_per_iteration": 2.54215931892395
    },
    {
      "auxiliary_loss_clip": 0.01111362,
      "auxiliary_loss_mlp": 0.01032261,
      "balance_loss_clip": 1.03730285,
      "balance_loss_mlp": 1.01952386,
      "epoch": 0.5843980159326619,
      "flos": 31284083352960.0,
      "grad_norm": 2.1832606661477483,
      "language_loss": 0.76061189,
      "learning_rate": 1.5541445342912786e-06,
      "loss": 0.78204823,
      "num_input_tokens_seen": 209525995,
      "step": 9720,
      "time_per_iteration": 2.532802104949951
    },
    {
      "auxiliary_loss_clip": 0.01076505,
      "auxiliary_loss_mlp": 0.01036644,
      "balance_loss_clip": 1.03520632,
      "balance_loss_mlp": 1.02469313,
      "epoch": 0.58445813918533,
      "flos": 22748117466240.0,
      "grad_norm": 1.4785674678532308,
      "language_loss": 0.82928276,
      "learning_rate": 1.5537648818791105e-06,
      "loss": 0.85041416,
      "num_input_tokens_seen": 209545895,
      "step": 9721,
      "time_per_iteration": 2.55588960647583
    },
    {
      "auxiliary_loss_clip": 0.01039165,
      "auxiliary_loss_mlp": 0.01001236,
      "balance_loss_clip": 1.0144273,
      "balance_loss_mlp": 1.00006223,
      "epoch": 0.5845182624379979,
      "flos": 60686556658560.0,
      "grad_norm": 0.9332007813588825,
      "language_loss": 0.71330398,
      "learning_rate": 1.5533852463882226e-06,
      "loss": 0.73370802,
      "num_input_tokens_seen": 209602315,
      "step": 9722,
      "time_per_iteration": 3.117319107055664
    },
    {
      "auxiliary_loss_clip": 0.01092433,
      "auxiliary_loss_mlp": 0.01038294,
      "balance_loss_clip": 1.03577447,
      "balance_loss_mlp": 1.0257889,
      "epoch": 0.5845783856906659,
      "flos": 16362446336640.0,
      "grad_norm": 2.0645246566155526,
      "language_loss": 0.89334166,
      "learning_rate": 1.5530056278330113e-06,
      "loss": 0.91464895,
      "num_input_tokens_seen": 209617615,
      "step": 9723,
      "time_per_iteration": 2.4645912647247314
    },
    {
      "auxiliary_loss_clip": 0.01088404,
      "auxiliary_loss_mlp": 0.01028894,
      "balance_loss_clip": 1.04311967,
      "balance_loss_mlp": 1.0171454,
      "epoch": 0.5846385089433338,
      "flos": 20083402080000.0,
      "grad_norm": 1.4432273107401823,
      "language_loss": 0.68348706,
      "learning_rate": 1.5526260262278709e-06,
      "loss": 0.70466,
      "num_input_tokens_seen": 209637005,
      "step": 9724,
      "time_per_iteration": 2.544124126434326
    },
    {
      "auxiliary_loss_clip": 0.01102932,
      "auxiliary_loss_mlp": 0.01036582,
      "balance_loss_clip": 1.039294,
      "balance_loss_mlp": 1.02332044,
      "epoch": 0.5846986321960018,
      "flos": 17311062568320.0,
      "grad_norm": 1.807815661131636,
      "language_loss": 0.86239684,
      "learning_rate": 1.552246441587197e-06,
      "loss": 0.88379204,
      "num_input_tokens_seen": 209653170,
      "step": 9725,
      "time_per_iteration": 2.4518752098083496
    },
    {
      "auxiliary_loss_clip": 0.01097181,
      "auxiliary_loss_mlp": 0.01040424,
      "balance_loss_clip": 1.04152524,
      "balance_loss_mlp": 1.02744782,
      "epoch": 0.5847587554486697,
      "flos": 17197907748480.0,
      "grad_norm": 2.0087510121496983,
      "language_loss": 0.8304562,
      "learning_rate": 1.5518668739253821e-06,
      "loss": 0.85183227,
      "num_input_tokens_seen": 209671275,
      "step": 9726,
      "time_per_iteration": 3.8872323036193848
    },
    {
      "auxiliary_loss_clip": 0.01058426,
      "auxiliary_loss_mlp": 0.00787076,
      "balance_loss_clip": 1.04306662,
      "balance_loss_mlp": 1.01010334,
      "epoch": 0.5848188787013378,
      "flos": 24529106540160.0,
      "grad_norm": 2.0851336476345996,
      "language_loss": 0.6673792,
      "learning_rate": 1.5514873232568206e-06,
      "loss": 0.68583423,
      "num_input_tokens_seen": 209690380,
      "step": 9727,
      "time_per_iteration": 2.6648616790771484
    },
    {
      "auxiliary_loss_clip": 0.01077277,
      "auxiliary_loss_mlp": 0.0104738,
      "balance_loss_clip": 1.03815508,
      "balance_loss_mlp": 1.03309834,
      "epoch": 0.5848790019540057,
      "flos": 20628382204800.0,
      "grad_norm": 1.8939032787781607,
      "language_loss": 0.8117739,
      "learning_rate": 1.5511077895959055e-06,
      "loss": 0.83302045,
      "num_input_tokens_seen": 209708845,
      "step": 9728,
      "time_per_iteration": 2.5723607540130615
    },
    {
      "auxiliary_loss_clip": 0.01096825,
      "auxiliary_loss_mlp": 0.01036652,
      "balance_loss_clip": 1.03702021,
      "balance_loss_mlp": 1.02479053,
      "epoch": 0.5849391252066737,
      "flos": 22418852469120.0,
      "grad_norm": 1.8883601993267758,
      "language_loss": 0.77684081,
      "learning_rate": 1.550728272957027e-06,
      "loss": 0.79817557,
      "num_input_tokens_seen": 209729000,
      "step": 9729,
      "time_per_iteration": 2.5113677978515625
    },
    {
      "auxiliary_loss_clip": 0.01094553,
      "auxiliary_loss_mlp": 0.01035389,
      "balance_loss_clip": 1.03544831,
      "balance_loss_mlp": 1.02184737,
      "epoch": 0.5849992484593417,
      "flos": 25410929431680.0,
      "grad_norm": 1.7184935229225724,
      "language_loss": 0.70226097,
      "learning_rate": 1.5503487733545782e-06,
      "loss": 0.72356039,
      "num_input_tokens_seen": 209747435,
      "step": 9730,
      "time_per_iteration": 2.5356411933898926
    },
    {
      "auxiliary_loss_clip": 0.01116016,
      "auxiliary_loss_mlp": 0.01037856,
      "balance_loss_clip": 1.03973603,
      "balance_loss_mlp": 1.02378917,
      "epoch": 0.5850593717120096,
      "flos": 21065163586560.0,
      "grad_norm": 1.8104590166265047,
      "language_loss": 0.7845158,
      "learning_rate": 1.5499692908029482e-06,
      "loss": 0.80605453,
      "num_input_tokens_seen": 209764910,
      "step": 9731,
      "time_per_iteration": 2.4685041904449463
    },
    {
      "auxiliary_loss_clip": 0.01097721,
      "auxiliary_loss_mlp": 0.01046286,
      "balance_loss_clip": 1.03801632,
      "balance_loss_mlp": 1.03206468,
      "epoch": 0.5851194949646776,
      "flos": 25301545539840.0,
      "grad_norm": 2.0684086050749126,
      "language_loss": 0.71009183,
      "learning_rate": 1.549589825316528e-06,
      "loss": 0.73153192,
      "num_input_tokens_seen": 209786115,
      "step": 9732,
      "time_per_iteration": 2.526305675506592
    },
    {
      "auxiliary_loss_clip": 0.01059388,
      "auxiliary_loss_mlp": 0.01039272,
      "balance_loss_clip": 1.03481865,
      "balance_loss_mlp": 1.02426374,
      "epoch": 0.5851796182173455,
      "flos": 23587242065280.0,
      "grad_norm": 1.7510051863735043,
      "language_loss": 0.52563792,
      "learning_rate": 1.5492103769097075e-06,
      "loss": 0.54662454,
      "num_input_tokens_seen": 209806095,
      "step": 9733,
      "time_per_iteration": 3.988006353378296
    },
    {
      "auxiliary_loss_clip": 0.01095774,
      "auxiliary_loss_mlp": 0.0103647,
      "balance_loss_clip": 1.03661466,
      "balance_loss_mlp": 1.02297521,
      "epoch": 0.5852397414700136,
      "flos": 24822712310400.0,
      "grad_norm": 2.1778207085029457,
      "language_loss": 0.87252879,
      "learning_rate": 1.5488309455968739e-06,
      "loss": 0.89385128,
      "num_input_tokens_seen": 209823650,
      "step": 9734,
      "time_per_iteration": 2.511223316192627
    },
    {
      "auxiliary_loss_clip": 0.01084268,
      "auxiliary_loss_mlp": 0.01032859,
      "balance_loss_clip": 1.03744793,
      "balance_loss_mlp": 1.02099812,
      "epoch": 0.5852998647226815,
      "flos": 19937784343680.0,
      "grad_norm": 1.5963916871040333,
      "language_loss": 0.72229505,
      "learning_rate": 1.5484515313924163e-06,
      "loss": 0.74346626,
      "num_input_tokens_seen": 209843220,
      "step": 9735,
      "time_per_iteration": 2.5299553871154785
    },
    {
      "auxiliary_loss_clip": 0.01100232,
      "auxiliary_loss_mlp": 0.01040078,
      "balance_loss_clip": 1.03700614,
      "balance_loss_mlp": 1.02609456,
      "epoch": 0.5853599879753495,
      "flos": 16720367408640.0,
      "grad_norm": 2.8952653470644756,
      "language_loss": 0.74215198,
      "learning_rate": 1.5480721343107217e-06,
      "loss": 0.76355511,
      "num_input_tokens_seen": 209854880,
      "step": 9736,
      "time_per_iteration": 2.433213949203491
    },
    {
      "auxiliary_loss_clip": 0.01077161,
      "auxiliary_loss_mlp": 0.01036161,
      "balance_loss_clip": 1.03607893,
      "balance_loss_mlp": 1.02233911,
      "epoch": 0.5854201112280174,
      "flos": 44456583680640.0,
      "grad_norm": 1.942817040201886,
      "language_loss": 0.70867777,
      "learning_rate": 1.5476927543661772e-06,
      "loss": 0.72981101,
      "num_input_tokens_seen": 209877870,
      "step": 9737,
      "time_per_iteration": 4.199691295623779
    },
    {
      "auxiliary_loss_clip": 0.01072462,
      "auxiliary_loss_mlp": 0.01039381,
      "balance_loss_clip": 1.03706181,
      "balance_loss_mlp": 1.02620268,
      "epoch": 0.5854802344806854,
      "flos": 20339193807360.0,
      "grad_norm": 1.7636976658255459,
      "language_loss": 0.81906962,
      "learning_rate": 1.547313391573169e-06,
      "loss": 0.84018809,
      "num_input_tokens_seen": 209896690,
      "step": 9738,
      "time_per_iteration": 2.589310884475708
    },
    {
      "auxiliary_loss_clip": 0.01115642,
      "auxiliary_loss_mlp": 0.00787831,
      "balance_loss_clip": 1.03914273,
      "balance_loss_mlp": 1.01140237,
      "epoch": 0.5855403577333533,
      "flos": 20921054221440.0,
      "grad_norm": 1.6307167669259224,
      "language_loss": 0.68397492,
      "learning_rate": 1.546934045946082e-06,
      "loss": 0.70300967,
      "num_input_tokens_seen": 209914640,
      "step": 9739,
      "time_per_iteration": 2.452491283416748
    },
    {
      "auxiliary_loss_clip": 0.01111522,
      "auxiliary_loss_mlp": 0.01027029,
      "balance_loss_clip": 1.03640151,
      "balance_loss_mlp": 1.0139401,
      "epoch": 0.5856004809860214,
      "flos": 20448649526400.0,
      "grad_norm": 2.526446551073146,
      "language_loss": 0.58765775,
      "learning_rate": 1.5465547174993017e-06,
      "loss": 0.60904324,
      "num_input_tokens_seen": 209933375,
      "step": 9740,
      "time_per_iteration": 2.4622416496276855
    },
    {
      "auxiliary_loss_clip": 0.01084705,
      "auxiliary_loss_mlp": 0.01029606,
      "balance_loss_clip": 1.03778756,
      "balance_loss_mlp": 1.01737523,
      "epoch": 0.5856606042386893,
      "flos": 19640766781440.0,
      "grad_norm": 2.1259023268483572,
      "language_loss": 0.75180453,
      "learning_rate": 1.5461754062472113e-06,
      "loss": 0.77294767,
      "num_input_tokens_seen": 209952055,
      "step": 9741,
      "time_per_iteration": 3.94122576713562
    },
    {
      "auxiliary_loss_clip": 0.01079799,
      "auxiliary_loss_mlp": 0.01032272,
      "balance_loss_clip": 1.03870523,
      "balance_loss_mlp": 1.01943946,
      "epoch": 0.5857207274913573,
      "flos": 21686166846720.0,
      "grad_norm": 1.584600684819687,
      "language_loss": 0.75671571,
      "learning_rate": 1.5457961122041959e-06,
      "loss": 0.77783644,
      "num_input_tokens_seen": 209971190,
      "step": 9742,
      "time_per_iteration": 2.579705238342285
    },
    {
      "auxiliary_loss_clip": 0.01087098,
      "auxiliary_loss_mlp": 0.0102857,
      "balance_loss_clip": 1.0374918,
      "balance_loss_mlp": 1.01650572,
      "epoch": 0.5857808507440253,
      "flos": 23182708118400.0,
      "grad_norm": 1.6859565108814716,
      "language_loss": 0.75093699,
      "learning_rate": 1.5454168353846369e-06,
      "loss": 0.77209365,
      "num_input_tokens_seen": 209990695,
      "step": 9743,
      "time_per_iteration": 2.5167527198791504
    },
    {
      "auxiliary_loss_clip": 0.01087332,
      "auxiliary_loss_mlp": 0.01030164,
      "balance_loss_clip": 1.03858042,
      "balance_loss_mlp": 1.01833892,
      "epoch": 0.5858409739966932,
      "flos": 27235299156480.0,
      "grad_norm": 1.6268001699396888,
      "language_loss": 0.81011724,
      "learning_rate": 1.5450375758029172e-06,
      "loss": 0.83129221,
      "num_input_tokens_seen": 210010210,
      "step": 9744,
      "time_per_iteration": 2.583547592163086
    },
    {
      "auxiliary_loss_clip": 0.01087243,
      "auxiliary_loss_mlp": 0.01030565,
      "balance_loss_clip": 1.03879786,
      "balance_loss_mlp": 1.01807177,
      "epoch": 0.5859010972493612,
      "flos": 27855512317440.0,
      "grad_norm": 1.6942867990049928,
      "language_loss": 0.71709675,
      "learning_rate": 1.5446583334734183e-06,
      "loss": 0.73827481,
      "num_input_tokens_seen": 210030030,
      "step": 9745,
      "time_per_iteration": 2.5846216678619385
    },
    {
      "auxiliary_loss_clip": 0.01019295,
      "auxiliary_loss_mlp": 0.01000799,
      "balance_loss_clip": 1.01718855,
      "balance_loss_mlp": 0.99955952,
      "epoch": 0.5859612205020291,
      "flos": 70007064428160.0,
      "grad_norm": 0.7253393624645269,
      "language_loss": 0.53302419,
      "learning_rate": 1.5442791084105204e-06,
      "loss": 0.55322516,
      "num_input_tokens_seen": 210094840,
      "step": 9746,
      "time_per_iteration": 3.2194530963897705
    },
    {
      "auxiliary_loss_clip": 0.01094164,
      "auxiliary_loss_mlp": 0.01032769,
      "balance_loss_clip": 1.0390842,
      "balance_loss_mlp": 1.01908338,
      "epoch": 0.5860213437546972,
      "flos": 24056019486720.0,
      "grad_norm": 2.124330420668836,
      "language_loss": 0.73340702,
      "learning_rate": 1.5438999006286054e-06,
      "loss": 0.75467634,
      "num_input_tokens_seen": 210114660,
      "step": 9747,
      "time_per_iteration": 2.5679404735565186
    },
    {
      "auxiliary_loss_clip": 0.01086067,
      "auxiliary_loss_mlp": 0.0103675,
      "balance_loss_clip": 1.03708172,
      "balance_loss_mlp": 1.02136016,
      "epoch": 0.5860814670073651,
      "flos": 18947583141120.0,
      "grad_norm": 3.3900216214184953,
      "language_loss": 0.8151229,
      "learning_rate": 1.543520710142051e-06,
      "loss": 0.83635104,
      "num_input_tokens_seen": 210132770,
      "step": 9748,
      "time_per_iteration": 2.4953346252441406
    },
    {
      "auxiliary_loss_clip": 0.01101803,
      "auxiliary_loss_mlp": 0.01030876,
      "balance_loss_clip": 1.03795373,
      "balance_loss_mlp": 1.01799512,
      "epoch": 0.5861415902600331,
      "flos": 22561848512640.0,
      "grad_norm": 1.758573184427131,
      "language_loss": 0.71919,
      "learning_rate": 1.5431415369652375e-06,
      "loss": 0.74051678,
      "num_input_tokens_seen": 210151895,
      "step": 9749,
      "time_per_iteration": 2.5414841175079346
    },
    {
      "auxiliary_loss_clip": 0.01089843,
      "auxiliary_loss_mlp": 0.01027837,
      "balance_loss_clip": 1.03953218,
      "balance_loss_mlp": 1.01506364,
      "epoch": 0.586201713512701,
      "flos": 14392027912320.0,
      "grad_norm": 2.2561656675407726,
      "language_loss": 0.75129086,
      "learning_rate": 1.5427623811125428e-06,
      "loss": 0.77246761,
      "num_input_tokens_seen": 210168040,
      "step": 9750,
      "time_per_iteration": 2.5102453231811523
    },
    {
      "auxiliary_loss_clip": 0.01078935,
      "auxiliary_loss_mlp": 0.01033065,
      "balance_loss_clip": 1.04317117,
      "balance_loss_mlp": 1.0199579,
      "epoch": 0.586261836765369,
      "flos": 19498560837120.0,
      "grad_norm": 1.8068652651517798,
      "language_loss": 0.71003479,
      "learning_rate": 1.542383242598344e-06,
      "loss": 0.7311548,
      "num_input_tokens_seen": 210187720,
      "step": 9751,
      "time_per_iteration": 2.5922772884368896
    },
    {
      "auxiliary_loss_clip": 0.0111616,
      "auxiliary_loss_mlp": 0.01035471,
      "balance_loss_clip": 1.03947616,
      "balance_loss_mlp": 1.02116585,
      "epoch": 0.5863219600180369,
      "flos": 20701819560960.0,
      "grad_norm": 1.7696133906560123,
      "language_loss": 0.74616861,
      "learning_rate": 1.5420041214370184e-06,
      "loss": 0.76768494,
      "num_input_tokens_seen": 210206080,
      "step": 9752,
      "time_per_iteration": 2.4440674781799316
    },
    {
      "auxiliary_loss_clip": 0.01099928,
      "auxiliary_loss_mlp": 0.010285,
      "balance_loss_clip": 1.0385654,
      "balance_loss_mlp": 1.01572132,
      "epoch": 0.586382083270705,
      "flos": 19792130693760.0,
      "grad_norm": 1.7425574926982417,
      "language_loss": 0.77049434,
      "learning_rate": 1.541625017642943e-06,
      "loss": 0.79177862,
      "num_input_tokens_seen": 210225660,
      "step": 9753,
      "time_per_iteration": 2.50671648979187
    },
    {
      "auxiliary_loss_clip": 0.0110907,
      "auxiliary_loss_mlp": 0.01026102,
      "balance_loss_clip": 1.03890264,
      "balance_loss_mlp": 1.01407409,
      "epoch": 0.5864422065233729,
      "flos": 16500558130560.0,
      "grad_norm": 1.8013959514570013,
      "language_loss": 0.71102989,
      "learning_rate": 1.5412459312304927e-06,
      "loss": 0.73238158,
      "num_input_tokens_seen": 210242725,
      "step": 9754,
      "time_per_iteration": 2.450523853302002
    },
    {
      "auxiliary_loss_clip": 0.01087892,
      "auxiliary_loss_mlp": 0.01030881,
      "balance_loss_clip": 1.03663969,
      "balance_loss_mlp": 1.01708865,
      "epoch": 0.5865023297760409,
      "flos": 20413277608320.0,
      "grad_norm": 2.434358315089651,
      "language_loss": 0.72120059,
      "learning_rate": 1.540866862214043e-06,
      "loss": 0.74238837,
      "num_input_tokens_seen": 210263225,
      "step": 9755,
      "time_per_iteration": 2.5571441650390625
    },
    {
      "auxiliary_loss_clip": 0.01014294,
      "auxiliary_loss_mlp": 0.01005688,
      "balance_loss_clip": 1.01985657,
      "balance_loss_mlp": 1.00439453,
      "epoch": 0.5865624530287089,
      "flos": 63350769254400.0,
      "grad_norm": 0.7416104160457917,
      "language_loss": 0.56947958,
      "learning_rate": 1.540487810607967e-06,
      "loss": 0.58967936,
      "num_input_tokens_seen": 210322310,
      "step": 9756,
      "time_per_iteration": 3.178391456604004
    },
    {
      "auxiliary_loss_clip": 0.0111047,
      "auxiliary_loss_mlp": 0.01033578,
      "balance_loss_clip": 1.03848875,
      "balance_loss_mlp": 1.02173471,
      "epoch": 0.5866225762813768,
      "flos": 27016279977600.0,
      "grad_norm": 2.0634787975902715,
      "language_loss": 0.76056182,
      "learning_rate": 1.5401087764266396e-06,
      "loss": 0.78200227,
      "num_input_tokens_seen": 210340845,
      "step": 9757,
      "time_per_iteration": 2.5442492961883545
    },
    {
      "auxiliary_loss_clip": 0.01022971,
      "auxiliary_loss_mlp": 0.01004466,
      "balance_loss_clip": 1.01776791,
      "balance_loss_mlp": 1.00320852,
      "epoch": 0.5866826995340448,
      "flos": 72987038507520.0,
      "grad_norm": 0.8455578931380895,
      "language_loss": 0.60415035,
      "learning_rate": 1.5397297596844337e-06,
      "loss": 0.6244247,
      "num_input_tokens_seen": 210397815,
      "step": 9758,
      "time_per_iteration": 3.119826316833496
    },
    {
      "auxiliary_loss_clip": 0.01118532,
      "auxiliary_loss_mlp": 0.01029564,
      "balance_loss_clip": 1.04075909,
      "balance_loss_mlp": 1.0159502,
      "epoch": 0.5867428227867127,
      "flos": 21285727050240.0,
      "grad_norm": 2.599047411874615,
      "language_loss": 0.72203374,
      "learning_rate": 1.5393507603957212e-06,
      "loss": 0.74351466,
      "num_input_tokens_seen": 210413900,
      "step": 9759,
      "time_per_iteration": 2.4718644618988037
    },
    {
      "auxiliary_loss_clip": 0.0108858,
      "auxiliary_loss_mlp": 0.01033183,
      "balance_loss_clip": 1.03727615,
      "balance_loss_mlp": 1.02099991,
      "epoch": 0.5868029460393808,
      "flos": 33468852188160.0,
      "grad_norm": 1.5756043382474239,
      "language_loss": 0.73370838,
      "learning_rate": 1.5389717785748742e-06,
      "loss": 0.75492603,
      "num_input_tokens_seen": 210434110,
      "step": 9760,
      "time_per_iteration": 2.63358211517334
    },
    {
      "auxiliary_loss_clip": 0.01099196,
      "auxiliary_loss_mlp": 0.01032232,
      "balance_loss_clip": 1.03842068,
      "balance_loss_mlp": 1.01911926,
      "epoch": 0.5868630692920487,
      "flos": 17889475276800.0,
      "grad_norm": 1.7408274139291435,
      "language_loss": 0.72576219,
      "learning_rate": 1.5385928142362637e-06,
      "loss": 0.74707639,
      "num_input_tokens_seen": 210451685,
      "step": 9761,
      "time_per_iteration": 2.4819915294647217
    },
    {
      "auxiliary_loss_clip": 0.01096412,
      "auxiliary_loss_mlp": 0.0103288,
      "balance_loss_clip": 1.03928137,
      "balance_loss_mlp": 1.01872969,
      "epoch": 0.5869231925447167,
      "flos": 21035035054080.0,
      "grad_norm": 1.743486170683403,
      "language_loss": 0.75101382,
      "learning_rate": 1.5382138673942597e-06,
      "loss": 0.77230674,
      "num_input_tokens_seen": 210470825,
      "step": 9762,
      "time_per_iteration": 2.5373435020446777
    },
    {
      "auxiliary_loss_clip": 0.01075535,
      "auxiliary_loss_mlp": 0.01034681,
      "balance_loss_clip": 1.03870296,
      "balance_loss_mlp": 1.02079272,
      "epoch": 0.5869833157973846,
      "flos": 74738219293440.0,
      "grad_norm": 1.295391831116469,
      "language_loss": 0.72367644,
      "learning_rate": 1.5378349380632317e-06,
      "loss": 0.74477863,
      "num_input_tokens_seen": 210500075,
      "step": 9763,
      "time_per_iteration": 2.9667675495147705
    },
    {
      "auxiliary_loss_clip": 0.01099154,
      "auxiliary_loss_mlp": 0.01034563,
      "balance_loss_clip": 1.0377363,
      "balance_loss_mlp": 1.02224255,
      "epoch": 0.5870434390500526,
      "flos": 17638998762240.0,
      "grad_norm": 1.5739301011450604,
      "language_loss": 0.79865897,
      "learning_rate": 1.53745602625755e-06,
      "loss": 0.81999612,
      "num_input_tokens_seen": 210518150,
      "step": 9764,
      "time_per_iteration": 3.8340649604797363
    },
    {
      "auxiliary_loss_clip": 0.01086227,
      "auxiliary_loss_mlp": 0.01034175,
      "balance_loss_clip": 1.0416683,
      "balance_loss_mlp": 1.02172387,
      "epoch": 0.5871035623027205,
      "flos": 21506146859520.0,
      "grad_norm": 1.9507442618381798,
      "language_loss": 0.79024267,
      "learning_rate": 1.5370771319915819e-06,
      "loss": 0.81144667,
      "num_input_tokens_seen": 210537760,
      "step": 9765,
      "time_per_iteration": 2.5186378955841064
    },
    {
      "auxiliary_loss_clip": 0.01081511,
      "auxiliary_loss_mlp": 0.0103874,
      "balance_loss_clip": 1.03905666,
      "balance_loss_mlp": 1.02531707,
      "epoch": 0.5871636855553886,
      "flos": 13551861818880.0,
      "grad_norm": 1.8416580105716132,
      "language_loss": 0.83941203,
      "learning_rate": 1.5366982552796947e-06,
      "loss": 0.86061454,
      "num_input_tokens_seen": 210555515,
      "step": 9766,
      "time_per_iteration": 2.5129966735839844
    },
    {
      "auxiliary_loss_clip": 0.01105947,
      "auxiliary_loss_mlp": 0.01036324,
      "balance_loss_clip": 1.04003024,
      "balance_loss_mlp": 1.02361083,
      "epoch": 0.5872238088080565,
      "flos": 26212922346240.0,
      "grad_norm": 1.7282659316141296,
      "language_loss": 0.69688237,
      "learning_rate": 1.536319396136257e-06,
      "loss": 0.71830511,
      "num_input_tokens_seen": 210575000,
      "step": 9767,
      "time_per_iteration": 2.549257755279541
    },
    {
      "auxiliary_loss_clip": 0.01097425,
      "auxiliary_loss_mlp": 0.00787526,
      "balance_loss_clip": 1.03678215,
      "balance_loss_mlp": 1.00764179,
      "epoch": 0.5872839320607245,
      "flos": 30665198995200.0,
      "grad_norm": 2.1039210906789227,
      "language_loss": 0.63602257,
      "learning_rate": 1.5359405545756336e-06,
      "loss": 0.65487206,
      "num_input_tokens_seen": 210595185,
      "step": 9768,
      "time_per_iteration": 2.5859522819519043
    },
    {
      "auxiliary_loss_clip": 0.01043246,
      "auxiliary_loss_mlp": 0.00782503,
      "balance_loss_clip": 1.01825178,
      "balance_loss_mlp": 1.0361836,
      "epoch": 0.5873440553133924,
      "flos": 60303570871680.0,
      "grad_norm": 0.723566838371461,
      "language_loss": 0.53938699,
      "learning_rate": 1.5355617306121914e-06,
      "loss": 0.55764449,
      "num_input_tokens_seen": 210653210,
      "step": 9769,
      "time_per_iteration": 3.0935494899749756
    },
    {
      "auxiliary_loss_clip": 0.01080416,
      "auxiliary_loss_mlp": 0.01036693,
      "balance_loss_clip": 1.03841472,
      "balance_loss_mlp": 1.02374053,
      "epoch": 0.5874041785660604,
      "flos": 21539292134400.0,
      "grad_norm": 1.4141090337780486,
      "language_loss": 0.70547748,
      "learning_rate": 1.5351829242602945e-06,
      "loss": 0.72664857,
      "num_input_tokens_seen": 210673750,
      "step": 9770,
      "time_per_iteration": 2.581315040588379
    },
    {
      "auxiliary_loss_clip": 0.0106611,
      "auxiliary_loss_mlp": 0.01033904,
      "balance_loss_clip": 1.03683901,
      "balance_loss_mlp": 1.02077293,
      "epoch": 0.5874643018187284,
      "flos": 24388947671040.0,
      "grad_norm": 1.8064983361768363,
      "language_loss": 0.67889071,
      "learning_rate": 1.5348041355343077e-06,
      "loss": 0.69989085,
      "num_input_tokens_seen": 210692960,
      "step": 9771,
      "time_per_iteration": 2.605313777923584
    },
    {
      "auxiliary_loss_clip": 0.01067209,
      "auxiliary_loss_mlp": 0.0103773,
      "balance_loss_clip": 1.03580904,
      "balance_loss_mlp": 1.02188754,
      "epoch": 0.5875244250713964,
      "flos": 28147717457280.0,
      "grad_norm": 1.7052973001700296,
      "language_loss": 0.65857887,
      "learning_rate": 1.5344253644485954e-06,
      "loss": 0.67962825,
      "num_input_tokens_seen": 210714040,
      "step": 9772,
      "time_per_iteration": 4.01865291595459
    },
    {
      "auxiliary_loss_clip": 0.01118571,
      "auxiliary_loss_mlp": 0.01044037,
      "balance_loss_clip": 1.04177642,
      "balance_loss_mlp": 1.0293324,
      "epoch": 0.5875845483240644,
      "flos": 25812410722560.0,
      "grad_norm": 1.6964720499189807,
      "language_loss": 0.74325776,
      "learning_rate": 1.534046611017519e-06,
      "loss": 0.76488388,
      "num_input_tokens_seen": 210733710,
      "step": 9773,
      "time_per_iteration": 2.521066188812256
    },
    {
      "auxiliary_loss_clip": 0.01080907,
      "auxiliary_loss_mlp": 0.01041614,
      "balance_loss_clip": 1.0390569,
      "balance_loss_mlp": 1.02748156,
      "epoch": 0.5876446715767323,
      "flos": 26906572863360.0,
      "grad_norm": 2.079384471812669,
      "language_loss": 0.53757101,
      "learning_rate": 1.5336678752554421e-06,
      "loss": 0.55879623,
      "num_input_tokens_seen": 210753580,
      "step": 9774,
      "time_per_iteration": 2.600245475769043
    },
    {
      "auxiliary_loss_clip": 0.01105871,
      "auxiliary_loss_mlp": 0.01035641,
      "balance_loss_clip": 1.04047668,
      "balance_loss_mlp": 1.02205133,
      "epoch": 0.5877047948294003,
      "flos": 36684832579200.0,
      "grad_norm": 2.250352830420552,
      "language_loss": 0.64837056,
      "learning_rate": 1.5332891571767264e-06,
      "loss": 0.66978568,
      "num_input_tokens_seen": 210773495,
      "step": 9775,
      "time_per_iteration": 4.013902902603149
    },
    {
      "auxiliary_loss_clip": 0.01099395,
      "auxiliary_loss_mlp": 0.01031805,
      "balance_loss_clip": 1.04133022,
      "balance_loss_mlp": 1.01880491,
      "epoch": 0.5877649180820682,
      "flos": 26724721282560.0,
      "grad_norm": 1.6246862364354295,
      "language_loss": 0.73812592,
      "learning_rate": 1.5329104567957326e-06,
      "loss": 0.75943792,
      "num_input_tokens_seen": 210793645,
      "step": 9776,
      "time_per_iteration": 2.5575692653656006
    },
    {
      "auxiliary_loss_clip": 0.01111875,
      "auxiliary_loss_mlp": 0.01032593,
      "balance_loss_clip": 1.03809237,
      "balance_loss_mlp": 1.0201596,
      "epoch": 0.5878250413347362,
      "flos": 21032197879680.0,
      "grad_norm": 2.018040131483488,
      "language_loss": 0.74001086,
      "learning_rate": 1.532531774126821e-06,
      "loss": 0.76145554,
      "num_input_tokens_seen": 210813415,
      "step": 9777,
      "time_per_iteration": 2.4730143547058105
    },
    {
      "auxiliary_loss_clip": 0.01076816,
      "auxiliary_loss_mlp": 0.01034888,
      "balance_loss_clip": 1.038064,
      "balance_loss_mlp": 1.02340198,
      "epoch": 0.5878851645874041,
      "flos": 25484259047040.0,
      "grad_norm": 1.4407696526715994,
      "language_loss": 0.74290252,
      "learning_rate": 1.5321531091843512e-06,
      "loss": 0.76401955,
      "num_input_tokens_seen": 210833850,
      "step": 9778,
      "time_per_iteration": 2.60758376121521
    },
    {
      "auxiliary_loss_clip": 0.01077382,
      "auxiliary_loss_mlp": 0.01034811,
      "balance_loss_clip": 1.03715062,
      "balance_loss_mlp": 1.02050543,
      "epoch": 0.5879452878400722,
      "flos": 23769129559680.0,
      "grad_norm": 1.8398665779029335,
      "language_loss": 0.70236313,
      "learning_rate": 1.5317744619826824e-06,
      "loss": 0.72348511,
      "num_input_tokens_seen": 210853115,
      "step": 9779,
      "time_per_iteration": 2.5499143600463867
    },
    {
      "auxiliary_loss_clip": 0.01114743,
      "auxiliary_loss_mlp": 0.00793182,
      "balance_loss_clip": 1.03845167,
      "balance_loss_mlp": 1.02141297,
      "epoch": 0.5880054110927401,
      "flos": 17824513530240.0,
      "grad_norm": 1.9558306794631963,
      "language_loss": 0.6662885,
      "learning_rate": 1.5313958325361727e-06,
      "loss": 0.6853677,
      "num_input_tokens_seen": 210872090,
      "step": 9780,
      "time_per_iteration": 3.889853000640869
    },
    {
      "auxiliary_loss_clip": 0.01087679,
      "auxiliary_loss_mlp": 0.01035633,
      "balance_loss_clip": 1.04505301,
      "balance_loss_mlp": 1.02198994,
      "epoch": 0.5880655343454081,
      "flos": 19463404400640.0,
      "grad_norm": 2.7006532753907564,
      "language_loss": 0.72241211,
      "learning_rate": 1.5310172208591807e-06,
      "loss": 0.74364519,
      "num_input_tokens_seen": 210888490,
      "step": 9781,
      "time_per_iteration": 2.5106546878814697
    },
    {
      "auxiliary_loss_clip": 0.01092238,
      "auxiliary_loss_mlp": 0.00791532,
      "balance_loss_clip": 1.03898883,
      "balance_loss_mlp": 1.02023554,
      "epoch": 0.588125657598076,
      "flos": 21397588980480.0,
      "grad_norm": 1.3534517984080048,
      "language_loss": 0.70462739,
      "learning_rate": 1.5306386269660622e-06,
      "loss": 0.72346509,
      "num_input_tokens_seen": 210908220,
      "step": 9782,
      "time_per_iteration": 2.5486207008361816
    },
    {
      "auxiliary_loss_clip": 0.01099559,
      "auxiliary_loss_mlp": 0.01039959,
      "balance_loss_clip": 1.03660679,
      "balance_loss_mlp": 1.02572536,
      "epoch": 0.588185780850744,
      "flos": 16034653797120.0,
      "grad_norm": 2.161754714537969,
      "language_loss": 0.70372021,
      "learning_rate": 1.5302600508711741e-06,
      "loss": 0.72511542,
      "num_input_tokens_seen": 210923945,
      "step": 9783,
      "time_per_iteration": 2.4505860805511475
    },
    {
      "auxiliary_loss_clip": 0.01083231,
      "auxiliary_loss_mlp": 0.01034092,
      "balance_loss_clip": 1.03991699,
      "balance_loss_mlp": 1.02077055,
      "epoch": 0.588245904103412,
      "flos": 23728226947200.0,
      "grad_norm": 1.7518219762691998,
      "language_loss": 0.68820012,
      "learning_rate": 1.5298814925888719e-06,
      "loss": 0.70937335,
      "num_input_tokens_seen": 210941955,
      "step": 9784,
      "time_per_iteration": 2.5787034034729004
    },
    {
      "auxiliary_loss_clip": 0.01068515,
      "auxiliary_loss_mlp": 0.01032514,
      "balance_loss_clip": 1.03915787,
      "balance_loss_mlp": 1.01985955,
      "epoch": 0.58830602735608,
      "flos": 33802534558080.0,
      "grad_norm": 1.8369422479704325,
      "language_loss": 0.69522029,
      "learning_rate": 1.5295029521335102e-06,
      "loss": 0.71623063,
      "num_input_tokens_seen": 210963105,
      "step": 9785,
      "time_per_iteration": 2.697991371154785
    },
    {
      "auxiliary_loss_clip": 0.01099112,
      "auxiliary_loss_mlp": 0.01027398,
      "balance_loss_clip": 1.03823066,
      "balance_loss_mlp": 1.01535821,
      "epoch": 0.588366150608748,
      "flos": 17090714586240.0,
      "grad_norm": 1.8887212292080808,
      "language_loss": 0.77096963,
      "learning_rate": 1.5291244295194448e-06,
      "loss": 0.79223472,
      "num_input_tokens_seen": 210978720,
      "step": 9786,
      "time_per_iteration": 2.472264051437378
    },
    {
      "auxiliary_loss_clip": 0.01098721,
      "auxiliary_loss_mlp": 0.01035546,
      "balance_loss_clip": 1.04028976,
      "balance_loss_mlp": 1.02224243,
      "epoch": 0.5884262738614159,
      "flos": 22127186033280.0,
      "grad_norm": 1.5004041176995888,
      "language_loss": 0.79224455,
      "learning_rate": 1.5287459247610276e-06,
      "loss": 0.81358725,
      "num_input_tokens_seen": 210998750,
      "step": 9787,
      "time_per_iteration": 2.5327038764953613
    },
    {
      "auxiliary_loss_clip": 0.01074992,
      "auxiliary_loss_mlp": 0.01030696,
      "balance_loss_clip": 1.03912926,
      "balance_loss_mlp": 1.01884675,
      "epoch": 0.5884863971140839,
      "flos": 21031838743680.0,
      "grad_norm": 1.5344081525962847,
      "language_loss": 0.66256833,
      "learning_rate": 1.5283674378726116e-06,
      "loss": 0.68362516,
      "num_input_tokens_seen": 211017550,
      "step": 9788,
      "time_per_iteration": 2.556525468826294
    },
    {
      "auxiliary_loss_clip": 0.01082947,
      "auxiliary_loss_mlp": 0.01037752,
      "balance_loss_clip": 1.03685665,
      "balance_loss_mlp": 1.02304161,
      "epoch": 0.5885465203667518,
      "flos": 23805112008960.0,
      "grad_norm": 2.0761374385340434,
      "language_loss": 0.80231154,
      "learning_rate": 1.5279889688685506e-06,
      "loss": 0.82351851,
      "num_input_tokens_seen": 211034135,
      "step": 9789,
      "time_per_iteration": 2.531259536743164
    },
    {
      "auxiliary_loss_clip": 0.01079249,
      "auxiliary_loss_mlp": 0.00786765,
      "balance_loss_clip": 1.03453588,
      "balance_loss_mlp": 1.01000154,
      "epoch": 0.5886066436194198,
      "flos": 18880574319360.0,
      "grad_norm": 1.771163247482478,
      "language_loss": 0.70251375,
      "learning_rate": 1.5276105177631944e-06,
      "loss": 0.721174,
      "num_input_tokens_seen": 211053850,
      "step": 9790,
      "time_per_iteration": 2.534313917160034
    },
    {
      "auxiliary_loss_clip": 0.0107608,
      "auxiliary_loss_mlp": 0.01032007,
      "balance_loss_clip": 1.040277,
      "balance_loss_mlp": 1.01922154,
      "epoch": 0.5886667668720877,
      "flos": 24790141653120.0,
      "grad_norm": 8.454696958345949,
      "language_loss": 0.8373248,
      "learning_rate": 1.527232084570895e-06,
      "loss": 0.85840559,
      "num_input_tokens_seen": 211072165,
      "step": 9791,
      "time_per_iteration": 2.5829927921295166
    },
    {
      "auxiliary_loss_clip": 0.01100541,
      "auxiliary_loss_mlp": 0.01047077,
      "balance_loss_clip": 1.03928971,
      "balance_loss_mlp": 1.03283167,
      "epoch": 0.5887268901247558,
      "flos": 21614381516160.0,
      "grad_norm": 1.5533064488736121,
      "language_loss": 0.76606023,
      "learning_rate": 1.5268536693060026e-06,
      "loss": 0.7875365,
      "num_input_tokens_seen": 211089630,
      "step": 9792,
      "time_per_iteration": 2.520554542541504
    },
    {
      "auxiliary_loss_clip": 0.01056414,
      "auxiliary_loss_mlp": 0.01043274,
      "balance_loss_clip": 1.0333004,
      "balance_loss_mlp": 1.0278008,
      "epoch": 0.5887870133774237,
      "flos": 20481722974080.0,
      "grad_norm": 1.8218083200241963,
      "language_loss": 0.69043422,
      "learning_rate": 1.5264752719828662e-06,
      "loss": 0.71143103,
      "num_input_tokens_seen": 211106120,
      "step": 9793,
      "time_per_iteration": 2.5907654762268066
    },
    {
      "auxiliary_loss_clip": 0.01109965,
      "auxiliary_loss_mlp": 0.01033782,
      "balance_loss_clip": 1.03868628,
      "balance_loss_mlp": 1.02062774,
      "epoch": 0.5888471366300917,
      "flos": 19206283870080.0,
      "grad_norm": 1.7369088137908923,
      "language_loss": 0.59978366,
      "learning_rate": 1.5260968926158353e-06,
      "loss": 0.62122118,
      "num_input_tokens_seen": 211122450,
      "step": 9794,
      "time_per_iteration": 2.449765205383301
    },
    {
      "auxiliary_loss_clip": 0.01081435,
      "auxiliary_loss_mlp": 0.01033024,
      "balance_loss_clip": 1.03863013,
      "balance_loss_mlp": 1.01942217,
      "epoch": 0.5889072598827596,
      "flos": 19972904866560.0,
      "grad_norm": 1.8171701324838332,
      "language_loss": 0.64642334,
      "learning_rate": 1.525718531219257e-06,
      "loss": 0.66756797,
      "num_input_tokens_seen": 211141765,
      "step": 9795,
      "time_per_iteration": 2.583070755004883
    },
    {
      "auxiliary_loss_clip": 0.01080582,
      "auxiliary_loss_mlp": 0.01033878,
      "balance_loss_clip": 1.03665304,
      "balance_loss_mlp": 1.02210045,
      "epoch": 0.5889673831354276,
      "flos": 20741249715840.0,
      "grad_norm": 1.8290381374896718,
      "language_loss": 0.74590302,
      "learning_rate": 1.5253401878074801e-06,
      "loss": 0.76704764,
      "num_input_tokens_seen": 211160475,
      "step": 9796,
      "time_per_iteration": 2.5452826023101807
    },
    {
      "auxiliary_loss_clip": 0.01087929,
      "auxiliary_loss_mlp": 0.01029938,
      "balance_loss_clip": 1.03788877,
      "balance_loss_mlp": 1.01737356,
      "epoch": 0.5890275063880956,
      "flos": 25300935008640.0,
      "grad_norm": 1.485693062621769,
      "language_loss": 0.82924139,
      "learning_rate": 1.5249618623948507e-06,
      "loss": 0.85042012,
      "num_input_tokens_seen": 211180480,
      "step": 9797,
      "time_per_iteration": 2.559847354888916
    },
    {
      "auxiliary_loss_clip": 0.01084673,
      "auxiliary_loss_mlp": 0.01029867,
      "balance_loss_clip": 1.0354116,
      "balance_loss_mlp": 1.01720071,
      "epoch": 0.5890876296407636,
      "flos": 11765377964160.0,
      "grad_norm": 1.670108079573868,
      "language_loss": 0.79379642,
      "learning_rate": 1.5245835549957152e-06,
      "loss": 0.81494188,
      "num_input_tokens_seen": 211198000,
      "step": 9798,
      "time_per_iteration": 2.5193421840667725
    },
    {
      "auxiliary_loss_clip": 0.01108556,
      "auxiliary_loss_mlp": 0.01033422,
      "balance_loss_clip": 1.03775144,
      "balance_loss_mlp": 1.02147102,
      "epoch": 0.5891477528934316,
      "flos": 13589460380160.0,
      "grad_norm": 2.618494668340049,
      "language_loss": 0.73831493,
      "learning_rate": 1.5242052656244186e-06,
      "loss": 0.75973469,
      "num_input_tokens_seen": 211214765,
      "step": 9799,
      "time_per_iteration": 2.42594313621521
    },
    {
      "auxiliary_loss_clip": 0.01080204,
      "auxiliary_loss_mlp": 0.01033764,
      "balance_loss_clip": 1.03908372,
      "balance_loss_mlp": 1.01944721,
      "epoch": 0.5892078761460995,
      "flos": 15049193189760.0,
      "grad_norm": 2.000297352632636,
      "language_loss": 0.75438011,
      "learning_rate": 1.5238269942953064e-06,
      "loss": 0.77551979,
      "num_input_tokens_seen": 211232335,
      "step": 9800,
      "time_per_iteration": 2.5530693531036377
    },
    {
      "auxiliary_loss_clip": 0.01063226,
      "auxiliary_loss_mlp": 0.01044626,
      "balance_loss_clip": 1.03567576,
      "balance_loss_mlp": 1.03047574,
      "epoch": 0.5892679993987675,
      "flos": 15778215624960.0,
      "grad_norm": 2.0618376103104312,
      "language_loss": 0.78760672,
      "learning_rate": 1.523448741022722e-06,
      "loss": 0.8086853,
      "num_input_tokens_seen": 211249985,
      "step": 9801,
      "time_per_iteration": 2.59871768951416
    },
    {
      "auxiliary_loss_clip": 0.01079812,
      "auxiliary_loss_mlp": 0.01033732,
      "balance_loss_clip": 1.04100442,
      "balance_loss_mlp": 1.02005327,
      "epoch": 0.5893281226514354,
      "flos": 25265203954560.0,
      "grad_norm": 2.050347508255539,
      "language_loss": 0.66018283,
      "learning_rate": 1.5230705058210088e-06,
      "loss": 0.68131828,
      "num_input_tokens_seen": 211268425,
      "step": 9802,
      "time_per_iteration": 2.582305908203125
    },
    {
      "auxiliary_loss_clip": 0.0109933,
      "auxiliary_loss_mlp": 0.01030834,
      "balance_loss_clip": 1.03833342,
      "balance_loss_mlp": 1.01806617,
      "epoch": 0.5893882459041034,
      "flos": 19458232842240.0,
      "grad_norm": 1.8212430810938476,
      "language_loss": 0.7818222,
      "learning_rate": 1.5226922887045108e-06,
      "loss": 0.80312383,
      "num_input_tokens_seen": 211286680,
      "step": 9803,
      "time_per_iteration": 3.8443236351013184
    },
    {
      "auxiliary_loss_clip": 0.01101908,
      "auxiliary_loss_mlp": 0.01036838,
      "balance_loss_clip": 1.0387677,
      "balance_loss_mlp": 1.02437496,
      "epoch": 0.5894483691567713,
      "flos": 20634056553600.0,
      "grad_norm": 1.8522288904853887,
      "language_loss": 0.72998148,
      "learning_rate": 1.5223140896875686e-06,
      "loss": 0.75136894,
      "num_input_tokens_seen": 211307700,
      "step": 9804,
      "time_per_iteration": 2.521259307861328
    },
    {
      "auxiliary_loss_clip": 0.01088233,
      "auxiliary_loss_mlp": 0.01034433,
      "balance_loss_clip": 1.04182434,
      "balance_loss_mlp": 1.02138555,
      "epoch": 0.5895084924094394,
      "flos": 17778223877760.0,
      "grad_norm": 1.605320292980396,
      "language_loss": 0.74651647,
      "learning_rate": 1.5219359087845234e-06,
      "loss": 0.76774317,
      "num_input_tokens_seen": 211324835,
      "step": 9805,
      "time_per_iteration": 2.5242626667022705
    },
    {
      "auxiliary_loss_clip": 0.01109022,
      "auxiliary_loss_mlp": 0.00787177,
      "balance_loss_clip": 1.04005837,
      "balance_loss_mlp": 1.00952387,
      "epoch": 0.5895686156621073,
      "flos": 20121072468480.0,
      "grad_norm": 1.717204136845522,
      "language_loss": 0.77869302,
      "learning_rate": 1.5215577460097174e-06,
      "loss": 0.79765499,
      "num_input_tokens_seen": 211344130,
      "step": 9806,
      "time_per_iteration": 2.494640350341797
    },
    {
      "auxiliary_loss_clip": 0.01111259,
      "auxiliary_loss_mlp": 0.01031034,
      "balance_loss_clip": 1.03814399,
      "balance_loss_mlp": 1.01787281,
      "epoch": 0.5896287389147753,
      "flos": 20850058990080.0,
      "grad_norm": 1.9855212292299347,
      "language_loss": 0.76451027,
      "learning_rate": 1.5211796013774887e-06,
      "loss": 0.78593314,
      "num_input_tokens_seen": 211362915,
      "step": 9807,
      "time_per_iteration": 2.487518310546875
    },
    {
      "auxiliary_loss_clip": 0.01108258,
      "auxiliary_loss_mlp": 0.01028626,
      "balance_loss_clip": 1.04268312,
      "balance_loss_mlp": 1.01534653,
      "epoch": 0.5896888621674432,
      "flos": 14537897043840.0,
      "grad_norm": 1.7748492418605226,
      "language_loss": 0.74086779,
      "learning_rate": 1.5208014749021786e-06,
      "loss": 0.7622366,
      "num_input_tokens_seen": 211380700,
      "step": 9808,
      "time_per_iteration": 2.487553119659424
    },
    {
      "auxiliary_loss_clip": 0.01068213,
      "auxiliary_loss_mlp": 0.01032901,
      "balance_loss_clip": 1.04380679,
      "balance_loss_mlp": 1.01799417,
      "epoch": 0.5897489854201112,
      "flos": 20886759711360.0,
      "grad_norm": 2.1271699493166487,
      "language_loss": 0.72336638,
      "learning_rate": 1.5204233665981236e-06,
      "loss": 0.74437749,
      "num_input_tokens_seen": 211400095,
      "step": 9809,
      "time_per_iteration": 2.62705659866333
    },
    {
      "auxiliary_loss_clip": 0.01093689,
      "auxiliary_loss_mlp": 0.01034995,
      "balance_loss_clip": 1.04045033,
      "balance_loss_mlp": 1.02195358,
      "epoch": 0.5898091086727792,
      "flos": 20011149872640.0,
      "grad_norm": 1.9768195398791444,
      "language_loss": 0.82200551,
      "learning_rate": 1.5200452764796627e-06,
      "loss": 0.8432923,
      "num_input_tokens_seen": 211417810,
      "step": 9810,
      "time_per_iteration": 2.508047342300415
    },
    {
      "auxiliary_loss_clip": 0.01102053,
      "auxiliary_loss_mlp": 0.01028912,
      "balance_loss_clip": 1.03986835,
      "balance_loss_mlp": 1.01663351,
      "epoch": 0.5898692319254472,
      "flos": 16253242012800.0,
      "grad_norm": 1.6721642447318088,
      "language_loss": 0.81253982,
      "learning_rate": 1.5196672045611336e-06,
      "loss": 0.83384949,
      "num_input_tokens_seen": 211436020,
      "step": 9811,
      "time_per_iteration": 3.883561849594116
    },
    {
      "auxiliary_loss_clip": 0.01104352,
      "auxiliary_loss_mlp": 0.01031075,
      "balance_loss_clip": 1.03946757,
      "balance_loss_mlp": 1.01710987,
      "epoch": 0.5899293551781152,
      "flos": 20448541785600.0,
      "grad_norm": 1.6624653670722664,
      "language_loss": 0.76807928,
      "learning_rate": 1.5192891508568715e-06,
      "loss": 0.78943348,
      "num_input_tokens_seen": 211454335,
      "step": 9812,
      "time_per_iteration": 2.513965368270874
    },
    {
      "auxiliary_loss_clip": 0.01078846,
      "auxiliary_loss_mlp": 0.01031703,
      "balance_loss_clip": 1.04029775,
      "balance_loss_mlp": 1.01998448,
      "epoch": 0.5899894784307831,
      "flos": 13881701433600.0,
      "grad_norm": 2.0878018853786977,
      "language_loss": 0.70123708,
      "learning_rate": 1.5189111153812133e-06,
      "loss": 0.72234261,
      "num_input_tokens_seen": 211472775,
      "step": 9813,
      "time_per_iteration": 2.536134958267212
    },
    {
      "auxiliary_loss_clip": 0.01087342,
      "auxiliary_loss_mlp": 0.01034334,
      "balance_loss_clip": 1.04080617,
      "balance_loss_mlp": 1.02157807,
      "epoch": 0.5900496016834511,
      "flos": 20083797129600.0,
      "grad_norm": 2.2328678252728116,
      "language_loss": 0.72156942,
      "learning_rate": 1.518533098148494e-06,
      "loss": 0.74278623,
      "num_input_tokens_seen": 211492195,
      "step": 9814,
      "time_per_iteration": 3.934065580368042
    },
    {
      "auxiliary_loss_clip": 0.01093714,
      "auxiliary_loss_mlp": 0.01030344,
      "balance_loss_clip": 1.0410459,
      "balance_loss_mlp": 1.01751137,
      "epoch": 0.590109724936119,
      "flos": 20259148348800.0,
      "grad_norm": 2.1763705642038085,
      "language_loss": 0.78292441,
      "learning_rate": 1.5181550991730476e-06,
      "loss": 0.80416489,
      "num_input_tokens_seen": 211510220,
      "step": 9815,
      "time_per_iteration": 2.5309579372406006
    },
    {
      "auxiliary_loss_clip": 0.01087764,
      "auxiliary_loss_mlp": 0.00788168,
      "balance_loss_clip": 1.03954411,
      "balance_loss_mlp": 1.01195502,
      "epoch": 0.590169848188787,
      "flos": 24235069806720.0,
      "grad_norm": 3.684914932130535,
      "language_loss": 0.75931346,
      "learning_rate": 1.5177771184692083e-06,
      "loss": 0.77807271,
      "num_input_tokens_seen": 211526260,
      "step": 9816,
      "time_per_iteration": 2.5609230995178223
    },
    {
      "auxiliary_loss_clip": 0.01113324,
      "auxiliary_loss_mlp": 0.01037378,
      "balance_loss_clip": 1.04132223,
      "balance_loss_mlp": 1.02461696,
      "epoch": 0.590229971441455,
      "flos": 17784724239360.0,
      "grad_norm": 1.896639751456738,
      "language_loss": 0.81112707,
      "learning_rate": 1.517399156051309e-06,
      "loss": 0.83263409,
      "num_input_tokens_seen": 211542890,
      "step": 9817,
      "time_per_iteration": 2.438495397567749
    },
    {
      "auxiliary_loss_clip": 0.01068123,
      "auxiliary_loss_mlp": 0.01037646,
      "balance_loss_clip": 1.03910494,
      "balance_loss_mlp": 1.02399039,
      "epoch": 0.590290094694123,
      "flos": 22236893147520.0,
      "grad_norm": 1.725273257630142,
      "language_loss": 0.76314938,
      "learning_rate": 1.517021211933682e-06,
      "loss": 0.78420699,
      "num_input_tokens_seen": 211562685,
      "step": 9818,
      "time_per_iteration": 4.051042079925537
    },
    {
      "auxiliary_loss_clip": 0.01076578,
      "auxiliary_loss_mlp": 0.01032048,
      "balance_loss_clip": 1.04300404,
      "balance_loss_mlp": 1.0199008,
      "epoch": 0.5903502179467909,
      "flos": 19098623831040.0,
      "grad_norm": 1.9294361333421874,
      "language_loss": 0.66574407,
      "learning_rate": 1.5166432861306592e-06,
      "loss": 0.68683034,
      "num_input_tokens_seen": 211579960,
      "step": 9819,
      "time_per_iteration": 2.5461270809173584
    },
    {
      "auxiliary_loss_clip": 0.01114073,
      "auxiliary_loss_mlp": 0.01032228,
      "balance_loss_clip": 1.04092884,
      "balance_loss_mlp": 1.01943135,
      "epoch": 0.5904103411994589,
      "flos": 24235500769920.0,
      "grad_norm": 1.637266323299819,
      "language_loss": 0.78202331,
      "learning_rate": 1.5162653786565714e-06,
      "loss": 0.80348635,
      "num_input_tokens_seen": 211599310,
      "step": 9820,
      "time_per_iteration": 2.4735982418060303
    },
    {
      "auxiliary_loss_clip": 0.01011357,
      "auxiliary_loss_mlp": 0.01009019,
      "balance_loss_clip": 1.02297664,
      "balance_loss_mlp": 1.0077734,
      "epoch": 0.5904704644521268,
      "flos": 64876613045760.0,
      "grad_norm": 1.0410484768421304,
      "language_loss": 0.65130115,
      "learning_rate": 1.5158874895257487e-06,
      "loss": 0.67150486,
      "num_input_tokens_seen": 211658790,
      "step": 9821,
      "time_per_iteration": 3.1371514797210693
    },
    {
      "auxiliary_loss_clip": 0.01072334,
      "auxiliary_loss_mlp": 0.01037201,
      "balance_loss_clip": 1.03776062,
      "balance_loss_mlp": 1.02491653,
      "epoch": 0.5905305877047948,
      "flos": 19609991804160.0,
      "grad_norm": 1.866637047790167,
      "language_loss": 0.61203998,
      "learning_rate": 1.515509618752521e-06,
      "loss": 0.63313532,
      "num_input_tokens_seen": 211677240,
      "step": 9822,
      "time_per_iteration": 2.5514883995056152
    },
    {
      "auxiliary_loss_clip": 0.01114618,
      "auxiliary_loss_mlp": 0.01039342,
      "balance_loss_clip": 1.04027772,
      "balance_loss_mlp": 1.02646101,
      "epoch": 0.5905907109574628,
      "flos": 18989634988800.0,
      "grad_norm": 1.9228268526462933,
      "language_loss": 0.82235456,
      "learning_rate": 1.5151317663512173e-06,
      "loss": 0.84389412,
      "num_input_tokens_seen": 211695485,
      "step": 9823,
      "time_per_iteration": 2.4734208583831787
    },
    {
      "auxiliary_loss_clip": 0.01092112,
      "auxiliary_loss_mlp": 0.01029321,
      "balance_loss_clip": 1.04015994,
      "balance_loss_mlp": 1.01647043,
      "epoch": 0.5906508342101308,
      "flos": 22200407907840.0,
      "grad_norm": 2.003158329502571,
      "language_loss": 0.72716427,
      "learning_rate": 1.514753932336165e-06,
      "loss": 0.74837852,
      "num_input_tokens_seen": 211713090,
      "step": 9824,
      "time_per_iteration": 2.534318447113037
    },
    {
      "auxiliary_loss_clip": 0.01089893,
      "auxiliary_loss_mlp": 0.00789678,
      "balance_loss_clip": 1.04036725,
      "balance_loss_mlp": 1.01239514,
      "epoch": 0.5907109574627988,
      "flos": 20886687884160.0,
      "grad_norm": 2.0895637166885024,
      "language_loss": 0.83182895,
      "learning_rate": 1.514376116721693e-06,
      "loss": 0.85062468,
      "num_input_tokens_seen": 211732510,
      "step": 9825,
      "time_per_iteration": 2.5419535636901855
    },
    {
      "auxiliary_loss_clip": 0.01097235,
      "auxiliary_loss_mlp": 0.01032831,
      "balance_loss_clip": 1.03819513,
      "balance_loss_mlp": 1.0218817,
      "epoch": 0.5907710807154667,
      "flos": 21506649649920.0,
      "grad_norm": 1.7481737745251156,
      "language_loss": 0.7643137,
      "learning_rate": 1.5139983195221272e-06,
      "loss": 0.78561437,
      "num_input_tokens_seen": 211748695,
      "step": 9826,
      "time_per_iteration": 2.4918174743652344
    },
    {
      "auxiliary_loss_clip": 0.0108668,
      "auxiliary_loss_mlp": 0.01026646,
      "balance_loss_clip": 1.037992,
      "balance_loss_mlp": 1.01436126,
      "epoch": 0.5908312039681347,
      "flos": 22018376759040.0,
      "grad_norm": 1.7093622234729342,
      "language_loss": 0.72092712,
      "learning_rate": 1.513620540751793e-06,
      "loss": 0.74206036,
      "num_input_tokens_seen": 211768545,
      "step": 9827,
      "time_per_iteration": 2.523790121078491
    },
    {
      "auxiliary_loss_clip": 0.01067216,
      "auxiliary_loss_mlp": 0.01027879,
      "balance_loss_clip": 1.03720737,
      "balance_loss_mlp": 1.01629162,
      "epoch": 0.5908913272208026,
      "flos": 18479523991680.0,
      "grad_norm": 1.673243890709185,
      "language_loss": 0.79903722,
      "learning_rate": 1.5132427804250178e-06,
      "loss": 0.81998813,
      "num_input_tokens_seen": 211786665,
      "step": 9828,
      "time_per_iteration": 2.5636303424835205
    },
    {
      "auxiliary_loss_clip": 0.01070159,
      "auxiliary_loss_mlp": 0.0103292,
      "balance_loss_clip": 1.04254913,
      "balance_loss_mlp": 1.01953912,
      "epoch": 0.5909514504734706,
      "flos": 12312189682560.0,
      "grad_norm": 2.19545181808018,
      "language_loss": 0.88184428,
      "learning_rate": 1.5128650385561241e-06,
      "loss": 0.90287513,
      "num_input_tokens_seen": 211801215,
      "step": 9829,
      "time_per_iteration": 2.5871691703796387
    },
    {
      "auxiliary_loss_clip": 0.01028918,
      "auxiliary_loss_mlp": 0.01001604,
      "balance_loss_clip": 1.02365088,
      "balance_loss_mlp": 1.0003705,
      "epoch": 0.5910115737261386,
      "flos": 70213262451840.0,
      "grad_norm": 0.756472002514238,
      "language_loss": 0.57840121,
      "learning_rate": 1.5124873151594376e-06,
      "loss": 0.59870636,
      "num_input_tokens_seen": 211857005,
      "step": 9830,
      "time_per_iteration": 3.0987348556518555
    },
    {
      "auxiliary_loss_clip": 0.01108899,
      "auxiliary_loss_mlp": 0.00787866,
      "balance_loss_clip": 1.04092383,
      "balance_loss_mlp": 1.01049888,
      "epoch": 0.5910716969788066,
      "flos": 22017766227840.0,
      "grad_norm": 2.1670634946625964,
      "language_loss": 0.76108062,
      "learning_rate": 1.5121096102492812e-06,
      "loss": 0.78004825,
      "num_input_tokens_seen": 211876675,
      "step": 9831,
      "time_per_iteration": 2.5106663703918457
    },
    {
      "auxiliary_loss_clip": 0.01083917,
      "auxiliary_loss_mlp": 0.01028926,
      "balance_loss_clip": 1.04151165,
      "balance_loss_mlp": 1.01596808,
      "epoch": 0.5911318202314745,
      "flos": 21251648021760.0,
      "grad_norm": 1.784911574620458,
      "language_loss": 0.77619344,
      "learning_rate": 1.5117319238399767e-06,
      "loss": 0.79732186,
      "num_input_tokens_seen": 211895725,
      "step": 9832,
      "time_per_iteration": 2.550844192504883
    },
    {
      "auxiliary_loss_clip": 0.01099598,
      "auxiliary_loss_mlp": 0.01026135,
      "balance_loss_clip": 1.038517,
      "balance_loss_mlp": 1.01344514,
      "epoch": 0.5911919434841425,
      "flos": 17821604528640.0,
      "grad_norm": 1.7770388404304318,
      "language_loss": 0.8345865,
      "learning_rate": 1.511354255945847e-06,
      "loss": 0.85584378,
      "num_input_tokens_seen": 211913860,
      "step": 9833,
      "time_per_iteration": 2.500671863555908
    },
    {
      "auxiliary_loss_clip": 0.0110316,
      "auxiliary_loss_mlp": 0.01029018,
      "balance_loss_clip": 1.03980684,
      "balance_loss_mlp": 1.01597643,
      "epoch": 0.5912520667368104,
      "flos": 20374781207040.0,
      "grad_norm": 1.4785805711954314,
      "language_loss": 0.74187219,
      "learning_rate": 1.5109766065812123e-06,
      "loss": 0.76319396,
      "num_input_tokens_seen": 211932880,
      "step": 9834,
      "time_per_iteration": 2.512303590774536
    },
    {
      "auxiliary_loss_clip": 0.01112082,
      "auxiliary_loss_mlp": 0.01031195,
      "balance_loss_clip": 1.03837514,
      "balance_loss_mlp": 1.01863647,
      "epoch": 0.5913121899894784,
      "flos": 17930557457280.0,
      "grad_norm": 3.2239431820696236,
      "language_loss": 0.77622437,
      "learning_rate": 1.5105989757603942e-06,
      "loss": 0.79765713,
      "num_input_tokens_seen": 211948625,
      "step": 9835,
      "time_per_iteration": 2.444979429244995
    },
    {
      "auxiliary_loss_clip": 0.01090603,
      "auxiliary_loss_mlp": 0.01031517,
      "balance_loss_clip": 1.03800499,
      "balance_loss_mlp": 1.01859426,
      "epoch": 0.5913723132421465,
      "flos": 22126934638080.0,
      "grad_norm": 1.9103873684530082,
      "language_loss": 0.74125898,
      "learning_rate": 1.5102213634977117e-06,
      "loss": 0.76248014,
      "num_input_tokens_seen": 211965355,
      "step": 9836,
      "time_per_iteration": 2.5275633335113525
    },
    {
      "auxiliary_loss_clip": 0.01077196,
      "auxiliary_loss_mlp": 0.010319,
      "balance_loss_clip": 1.03770971,
      "balance_loss_mlp": 1.01794076,
      "epoch": 0.5914324364948144,
      "flos": 15697918771200.0,
      "grad_norm": 1.9865068878752574,
      "language_loss": 0.82167792,
      "learning_rate": 1.5098437698074841e-06,
      "loss": 0.84276891,
      "num_input_tokens_seen": 211982245,
      "step": 9837,
      "time_per_iteration": 2.5246269702911377
    },
    {
      "auxiliary_loss_clip": 0.01074107,
      "auxiliary_loss_mlp": 0.01032113,
      "balance_loss_clip": 1.03695917,
      "balance_loss_mlp": 1.0177784,
      "epoch": 0.5914925597474824,
      "flos": 22747327367040.0,
      "grad_norm": 1.6639652050668934,
      "language_loss": 0.79549718,
      "learning_rate": 1.5094661947040304e-06,
      "loss": 0.81655937,
      "num_input_tokens_seen": 212000250,
      "step": 9838,
      "time_per_iteration": 2.574160099029541
    },
    {
      "auxiliary_loss_clip": 0.01066743,
      "auxiliary_loss_mlp": 0.01035253,
      "balance_loss_clip": 1.0425539,
      "balance_loss_mlp": 1.02272391,
      "epoch": 0.5915526830001503,
      "flos": 18292788161280.0,
      "grad_norm": 1.9185663045283423,
      "language_loss": 0.69497162,
      "learning_rate": 1.5090886382016673e-06,
      "loss": 0.71599162,
      "num_input_tokens_seen": 212017505,
      "step": 9839,
      "time_per_iteration": 2.569204330444336
    },
    {
      "auxiliary_loss_clip": 0.01089936,
      "auxiliary_loss_mlp": 0.0103793,
      "balance_loss_clip": 1.04100585,
      "balance_loss_mlp": 1.02504313,
      "epoch": 0.5916128062528183,
      "flos": 17019072910080.0,
      "grad_norm": 2.084320085397688,
      "language_loss": 0.65990424,
      "learning_rate": 1.5087111003147124e-06,
      "loss": 0.68118286,
      "num_input_tokens_seen": 212034595,
      "step": 9840,
      "time_per_iteration": 2.5272700786590576
    },
    {
      "auxiliary_loss_clip": 0.01087699,
      "auxiliary_loss_mlp": 0.0103131,
      "balance_loss_clip": 1.03908157,
      "balance_loss_mlp": 1.01790547,
      "epoch": 0.5916729295054862,
      "flos": 24754231031040.0,
      "grad_norm": 1.8946270754325716,
      "language_loss": 0.81893617,
      "learning_rate": 1.5083335810574813e-06,
      "loss": 0.84012628,
      "num_input_tokens_seen": 212055775,
      "step": 9841,
      "time_per_iteration": 3.930525302886963
    },
    {
      "auxiliary_loss_clip": 0.01086308,
      "auxiliary_loss_mlp": 0.01031802,
      "balance_loss_clip": 1.04003787,
      "balance_loss_mlp": 1.01933837,
      "epoch": 0.5917330527581542,
      "flos": 15958199698560.0,
      "grad_norm": 1.5354615079987628,
      "language_loss": 0.6906454,
      "learning_rate": 1.507956080444291e-06,
      "loss": 0.71182644,
      "num_input_tokens_seen": 212074000,
      "step": 9842,
      "time_per_iteration": 2.532405376434326
    },
    {
      "auxiliary_loss_clip": 0.0108932,
      "auxiliary_loss_mlp": 0.010304,
      "balance_loss_clip": 1.03821778,
      "balance_loss_mlp": 1.01729906,
      "epoch": 0.5917931760108222,
      "flos": 23800730549760.0,
      "grad_norm": 1.8376882155106844,
      "language_loss": 0.82905746,
      "learning_rate": 1.5075785984894549e-06,
      "loss": 0.85025465,
      "num_input_tokens_seen": 212091415,
      "step": 9843,
      "time_per_iteration": 2.5572309494018555
    },
    {
      "auxiliary_loss_clip": 0.01086739,
      "auxiliary_loss_mlp": 0.01031407,
      "balance_loss_clip": 1.03759265,
      "balance_loss_mlp": 1.01757288,
      "epoch": 0.5918532992634902,
      "flos": 23249609199360.0,
      "grad_norm": 2.5863809374432662,
      "language_loss": 0.81397957,
      "learning_rate": 1.5072011352072875e-06,
      "loss": 0.83516103,
      "num_input_tokens_seen": 212105255,
      "step": 9844,
      "time_per_iteration": 2.496699810028076
    },
    {
      "auxiliary_loss_clip": 0.01073007,
      "auxiliary_loss_mlp": 0.01028995,
      "balance_loss_clip": 1.04379535,
      "balance_loss_mlp": 1.01615047,
      "epoch": 0.5919134225161581,
      "flos": 19499853726720.0,
      "grad_norm": 1.8759841586584602,
      "language_loss": 0.74528074,
      "learning_rate": 1.5068236906121032e-06,
      "loss": 0.7663008,
      "num_input_tokens_seen": 212122765,
      "step": 9845,
      "time_per_iteration": 2.576406478881836
    },
    {
      "auxiliary_loss_clip": 0.01071963,
      "auxiliary_loss_mlp": 0.01032875,
      "balance_loss_clip": 1.0367198,
      "balance_loss_mlp": 1.01756835,
      "epoch": 0.5919735457688261,
      "flos": 38800940567040.0,
      "grad_norm": 1.9421099722789443,
      "language_loss": 0.63590753,
      "learning_rate": 1.506446264718213e-06,
      "loss": 0.6569559,
      "num_input_tokens_seen": 212143960,
      "step": 9846,
      "time_per_iteration": 2.711371421813965
    },
    {
      "auxiliary_loss_clip": 0.0107157,
      "auxiliary_loss_mlp": 0.00786157,
      "balance_loss_clip": 1.03969145,
      "balance_loss_mlp": 1.01349831,
      "epoch": 0.592033669021494,
      "flos": 22163994495360.0,
      "grad_norm": 1.9881070420526656,
      "language_loss": 0.76501119,
      "learning_rate": 1.506068857539931e-06,
      "loss": 0.78358847,
      "num_input_tokens_seen": 212162005,
      "step": 9847,
      "time_per_iteration": 2.585439682006836
    },
    {
      "auxiliary_loss_clip": 0.01088007,
      "auxiliary_loss_mlp": 0.01029809,
      "balance_loss_clip": 1.03804719,
      "balance_loss_mlp": 1.01657724,
      "epoch": 0.592093792274162,
      "flos": 22710985781760.0,
      "grad_norm": 1.6555255614634647,
      "language_loss": 0.62148297,
      "learning_rate": 1.5056914690915667e-06,
      "loss": 0.64266121,
      "num_input_tokens_seen": 212181635,
      "step": 9848,
      "time_per_iteration": 2.5455000400543213
    },
    {
      "auxiliary_loss_clip": 0.01102427,
      "auxiliary_loss_mlp": 0.01034017,
      "balance_loss_clip": 1.03977418,
      "balance_loss_mlp": 1.02164268,
      "epoch": 0.59215391552683,
      "flos": 22528954632960.0,
      "grad_norm": 1.7745316391621855,
      "language_loss": 0.76170218,
      "learning_rate": 1.5053140993874312e-06,
      "loss": 0.78306657,
      "num_input_tokens_seen": 212201615,
      "step": 9849,
      "time_per_iteration": 2.5041754245758057
    },
    {
      "auxiliary_loss_clip": 0.01086643,
      "auxiliary_loss_mlp": 0.0103282,
      "balance_loss_clip": 1.03692365,
      "balance_loss_mlp": 1.01948667,
      "epoch": 0.592214038779498,
      "flos": 24499013921280.0,
      "grad_norm": 1.7556690663919403,
      "language_loss": 0.75196922,
      "learning_rate": 1.5049367484418353e-06,
      "loss": 0.77316386,
      "num_input_tokens_seen": 212219355,
      "step": 9850,
      "time_per_iteration": 3.974691867828369
    },
    {
      "auxiliary_loss_clip": 0.01076381,
      "auxiliary_loss_mlp": 0.01031843,
      "balance_loss_clip": 1.03938556,
      "balance_loss_mlp": 1.01824689,
      "epoch": 0.592274162032166,
      "flos": 21831353619840.0,
      "grad_norm": 1.945095142186983,
      "language_loss": 0.75400156,
      "learning_rate": 1.5045594162690868e-06,
      "loss": 0.77508378,
      "num_input_tokens_seen": 212236710,
      "step": 9851,
      "time_per_iteration": 2.5520565509796143
    },
    {
      "auxiliary_loss_clip": 0.01091127,
      "auxiliary_loss_mlp": 0.01031894,
      "balance_loss_clip": 1.03992486,
      "balance_loss_mlp": 1.01939535,
      "epoch": 0.5923342852848339,
      "flos": 24608146417920.0,
      "grad_norm": 1.8919764684737679,
      "language_loss": 0.7040534,
      "learning_rate": 1.5041821028834954e-06,
      "loss": 0.72528362,
      "num_input_tokens_seen": 212256195,
      "step": 9852,
      "time_per_iteration": 2.548759937286377
    },
    {
      "auxiliary_loss_clip": 0.01095862,
      "auxiliary_loss_mlp": 0.0078804,
      "balance_loss_clip": 1.04011524,
      "balance_loss_mlp": 1.01216006,
      "epoch": 0.5923944085375019,
      "flos": 19938143479680.0,
      "grad_norm": 1.6090319990815676,
      "language_loss": 0.80295765,
      "learning_rate": 1.5038048082993685e-06,
      "loss": 0.82179666,
      "num_input_tokens_seen": 212274085,
      "step": 9853,
      "time_per_iteration": 3.914362907409668
    },
    {
      "auxiliary_loss_clip": 0.01086923,
      "auxiliary_loss_mlp": 0.01026305,
      "balance_loss_clip": 1.03888106,
      "balance_loss_mlp": 1.0150274,
      "epoch": 0.5924545317901698,
      "flos": 28658510812800.0,
      "grad_norm": 2.352007838498656,
      "language_loss": 0.67241514,
      "learning_rate": 1.5034275325310124e-06,
      "loss": 0.69354743,
      "num_input_tokens_seen": 212295530,
      "step": 9854,
      "time_per_iteration": 2.59879994392395
    },
    {
      "auxiliary_loss_clip": 0.0108378,
      "auxiliary_loss_mlp": 0.01028277,
      "balance_loss_clip": 1.03945851,
      "balance_loss_mlp": 1.01578355,
      "epoch": 0.5925146550428378,
      "flos": 19864885691520.0,
      "grad_norm": 1.7138799087444703,
      "language_loss": 0.88820267,
      "learning_rate": 1.5030502755927344e-06,
      "loss": 0.90932322,
      "num_input_tokens_seen": 212313770,
      "step": 9855,
      "time_per_iteration": 2.5400307178497314
    },
    {
      "auxiliary_loss_clip": 0.01101517,
      "auxiliary_loss_mlp": 0.01029643,
      "balance_loss_clip": 1.04023695,
      "balance_loss_mlp": 1.01815104,
      "epoch": 0.5925747782955058,
      "flos": 15122989681920.0,
      "grad_norm": 1.6914566048389879,
      "language_loss": 0.87011135,
      "learning_rate": 1.5026730374988397e-06,
      "loss": 0.89142299,
      "num_input_tokens_seen": 212331525,
      "step": 9856,
      "time_per_iteration": 3.8931527137756348
    },
    {
      "auxiliary_loss_clip": 0.01100305,
      "auxiliary_loss_mlp": 0.0103244,
      "balance_loss_clip": 1.03776622,
      "balance_loss_mlp": 1.02076912,
      "epoch": 0.5926349015481738,
      "flos": 18405440190720.0,
      "grad_norm": 2.1509412297441606,
      "language_loss": 0.77796149,
      "learning_rate": 1.5022958182636332e-06,
      "loss": 0.79928887,
      "num_input_tokens_seen": 212347295,
      "step": 9857,
      "time_per_iteration": 2.460188388824463
    },
    {
      "auxiliary_loss_clip": 0.01070509,
      "auxiliary_loss_mlp": 0.010456,
      "balance_loss_clip": 1.03666258,
      "balance_loss_mlp": 1.03054404,
      "epoch": 0.5926950248008417,
      "flos": 23111138269440.0,
      "grad_norm": 2.0611348538119443,
      "language_loss": 0.64408517,
      "learning_rate": 1.501918617901419e-06,
      "loss": 0.66524625,
      "num_input_tokens_seen": 212365750,
      "step": 9858,
      "time_per_iteration": 2.548243522644043
    },
    {
      "auxiliary_loss_clip": 0.01099923,
      "auxiliary_loss_mlp": 0.01027326,
      "balance_loss_clip": 1.04045558,
      "balance_loss_mlp": 1.01499999,
      "epoch": 0.5927551480535097,
      "flos": 28033916192640.0,
      "grad_norm": 1.861983520034219,
      "language_loss": 0.76944762,
      "learning_rate": 1.501541436426501e-06,
      "loss": 0.79072011,
      "num_input_tokens_seen": 212385300,
      "step": 9859,
      "time_per_iteration": 2.5366523265838623
    },
    {
      "auxiliary_loss_clip": 0.01070226,
      "auxiliary_loss_mlp": 0.0078616,
      "balance_loss_clip": 1.04022706,
      "balance_loss_mlp": 1.00947225,
      "epoch": 0.5928152713061776,
      "flos": 21798675221760.0,
      "grad_norm": 1.9034874480074724,
      "language_loss": 0.75506556,
      "learning_rate": 1.5011642738531818e-06,
      "loss": 0.77362943,
      "num_input_tokens_seen": 212402140,
      "step": 9860,
      "time_per_iteration": 2.6008763313293457
    },
    {
      "auxiliary_loss_clip": 0.01077186,
      "auxiliary_loss_mlp": 0.01032292,
      "balance_loss_clip": 1.0417496,
      "balance_loss_mlp": 1.02084148,
      "epoch": 0.5928753945588456,
      "flos": 24316839118080.0,
      "grad_norm": 1.5388682379169103,
      "language_loss": 0.76042235,
      "learning_rate": 1.500787130195763e-06,
      "loss": 0.78151715,
      "num_input_tokens_seen": 212421790,
      "step": 9861,
      "time_per_iteration": 2.5807487964630127
    },
    {
      "auxiliary_loss_clip": 0.01074882,
      "auxiliary_loss_mlp": 0.0102921,
      "balance_loss_clip": 1.03693175,
      "balance_loss_mlp": 1.01796889,
      "epoch": 0.5929355178115137,
      "flos": 26464619923200.0,
      "grad_norm": 1.6613216357323606,
      "language_loss": 0.70428526,
      "learning_rate": 1.5004100054685465e-06,
      "loss": 0.72532618,
      "num_input_tokens_seen": 212442115,
      "step": 9862,
      "time_per_iteration": 2.6039986610412598
    },
    {
      "auxiliary_loss_clip": 0.01057782,
      "auxiliary_loss_mlp": 0.01035022,
      "balance_loss_clip": 1.03697205,
      "balance_loss_mlp": 1.02276707,
      "epoch": 0.5929956410641816,
      "flos": 24965995662720.0,
      "grad_norm": 1.7937472333377105,
      "language_loss": 0.77674246,
      "learning_rate": 1.500032899685832e-06,
      "loss": 0.79767048,
      "num_input_tokens_seen": 212459535,
      "step": 9863,
      "time_per_iteration": 2.6328370571136475
    },
    {
      "auxiliary_loss_clip": 0.0108896,
      "auxiliary_loss_mlp": 0.01036079,
      "balance_loss_clip": 1.04194498,
      "balance_loss_mlp": 1.02306151,
      "epoch": 0.5930557643168496,
      "flos": 26208325405440.0,
      "grad_norm": 1.7081637176838518,
      "language_loss": 0.704476,
      "learning_rate": 1.499655812861921e-06,
      "loss": 0.72572637,
      "num_input_tokens_seen": 212479385,
      "step": 9864,
      "time_per_iteration": 2.597909688949585
    },
    {
      "auxiliary_loss_clip": 0.01086546,
      "auxiliary_loss_mlp": 0.01032078,
      "balance_loss_clip": 1.03879309,
      "balance_loss_mlp": 1.01880407,
      "epoch": 0.5931158875695175,
      "flos": 27854937699840.0,
      "grad_norm": 1.582104930575943,
      "language_loss": 0.67374563,
      "learning_rate": 1.4992787450111112e-06,
      "loss": 0.69493186,
      "num_input_tokens_seen": 212500060,
      "step": 9865,
      "time_per_iteration": 2.595360040664673
    },
    {
      "auxiliary_loss_clip": 0.01094316,
      "auxiliary_loss_mlp": 0.01035074,
      "balance_loss_clip": 1.03960347,
      "balance_loss_mlp": 1.02137673,
      "epoch": 0.5931760108221855,
      "flos": 15413650536960.0,
      "grad_norm": 2.0892524101652827,
      "language_loss": 0.7844094,
      "learning_rate": 1.4989016961477015e-06,
      "loss": 0.80570334,
      "num_input_tokens_seen": 212518590,
      "step": 9866,
      "time_per_iteration": 2.5086817741394043
    },
    {
      "auxiliary_loss_clip": 0.01087272,
      "auxiliary_loss_mlp": 0.01026949,
      "balance_loss_clip": 1.0428443,
      "balance_loss_mlp": 1.01518297,
      "epoch": 0.5932361340748534,
      "flos": 30188520581760.0,
      "grad_norm": 2.3946422043220816,
      "language_loss": 0.72184849,
      "learning_rate": 1.4985246662859903e-06,
      "loss": 0.74299073,
      "num_input_tokens_seen": 212538190,
      "step": 9867,
      "time_per_iteration": 2.6354801654815674
    },
    {
      "auxiliary_loss_clip": 0.01090008,
      "auxiliary_loss_mlp": 0.01032237,
      "balance_loss_clip": 1.04020524,
      "balance_loss_mlp": 1.01884961,
      "epoch": 0.5932962573275214,
      "flos": 20157557708160.0,
      "grad_norm": 1.533993730977218,
      "language_loss": 0.66655385,
      "learning_rate": 1.4981476554402732e-06,
      "loss": 0.68777627,
      "num_input_tokens_seen": 212557820,
      "step": 9868,
      "time_per_iteration": 2.5394532680511475
    },
    {
      "auxiliary_loss_clip": 0.01055969,
      "auxiliary_loss_mlp": 0.00786531,
      "balance_loss_clip": 1.03622556,
      "balance_loss_mlp": 1.01135468,
      "epoch": 0.5933563805801894,
      "flos": 25445906300160.0,
      "grad_norm": 1.485757783492498,
      "language_loss": 0.75397348,
      "learning_rate": 1.4977706636248478e-06,
      "loss": 0.77239847,
      "num_input_tokens_seen": 212577645,
      "step": 9869,
      "time_per_iteration": 2.665998935699463
    },
    {
      "auxiliary_loss_clip": 0.0107226,
      "auxiliary_loss_mlp": 0.01036952,
      "balance_loss_clip": 1.04438031,
      "balance_loss_mlp": 1.02335644,
      "epoch": 0.5934165038328574,
      "flos": 59995740337920.0,
      "grad_norm": 1.695929421402375,
      "language_loss": 0.74059808,
      "learning_rate": 1.4973936908540091e-06,
      "loss": 0.76169026,
      "num_input_tokens_seen": 212603430,
      "step": 9870,
      "time_per_iteration": 2.9447684288024902
    },
    {
      "auxiliary_loss_clip": 0.01071845,
      "auxiliary_loss_mlp": 0.01030196,
      "balance_loss_clip": 1.04079032,
      "balance_loss_mlp": 1.01772022,
      "epoch": 0.5934766270855253,
      "flos": 24420548661120.0,
      "grad_norm": 2.1091478710565825,
      "language_loss": 0.72341073,
      "learning_rate": 1.4970167371420517e-06,
      "loss": 0.74443114,
      "num_input_tokens_seen": 212620730,
      "step": 9871,
      "time_per_iteration": 2.6282167434692383
    },
    {
      "auxiliary_loss_clip": 0.01081017,
      "auxiliary_loss_mlp": 0.01034872,
      "balance_loss_clip": 1.03962374,
      "balance_loss_mlp": 1.02146745,
      "epoch": 0.5935367503381933,
      "flos": 23513158264320.0,
      "grad_norm": 1.8491600721977461,
      "language_loss": 0.74199682,
      "learning_rate": 1.496639802503271e-06,
      "loss": 0.7631557,
      "num_input_tokens_seen": 212639745,
      "step": 9872,
      "time_per_iteration": 2.592555522918701
    },
    {
      "auxiliary_loss_clip": 0.01105774,
      "auxiliary_loss_mlp": 0.01033745,
      "balance_loss_clip": 1.03941345,
      "balance_loss_mlp": 1.01982129,
      "epoch": 0.5935968735908612,
      "flos": 18948337326720.0,
      "grad_norm": 2.139377049560551,
      "language_loss": 0.79569316,
      "learning_rate": 1.4962628869519583e-06,
      "loss": 0.81708825,
      "num_input_tokens_seen": 212655915,
      "step": 9873,
      "time_per_iteration": 2.4909873008728027
    },
    {
      "auxiliary_loss_clip": 0.01102829,
      "auxiliary_loss_mlp": 0.01032602,
      "balance_loss_clip": 1.04025936,
      "balance_loss_mlp": 1.01997149,
      "epoch": 0.5936569968435292,
      "flos": 25483433034240.0,
      "grad_norm": 1.5986979303218471,
      "language_loss": 0.84910524,
      "learning_rate": 1.4958859905024078e-06,
      "loss": 0.87045956,
      "num_input_tokens_seen": 212676115,
      "step": 9874,
      "time_per_iteration": 2.542112112045288
    },
    {
      "auxiliary_loss_clip": 0.01029309,
      "auxiliary_loss_mlp": 0.01004002,
      "balance_loss_clip": 1.02346742,
      "balance_loss_mlp": 1.00266063,
      "epoch": 0.5937171200961973,
      "flos": 66378361789440.0,
      "grad_norm": 0.7141555613170256,
      "language_loss": 0.60065114,
      "learning_rate": 1.4955091131689115e-06,
      "loss": 0.62098432,
      "num_input_tokens_seen": 212737560,
      "step": 9875,
      "time_per_iteration": 3.216447353363037
    },
    {
      "auxiliary_loss_clip": 0.01091187,
      "auxiliary_loss_mlp": 0.01033532,
      "balance_loss_clip": 1.03756762,
      "balance_loss_mlp": 1.01908958,
      "epoch": 0.5937772433488652,
      "flos": 14903467712640.0,
      "grad_norm": 2.2212223089953795,
      "language_loss": 0.77613425,
      "learning_rate": 1.4951322549657594e-06,
      "loss": 0.7973814,
      "num_input_tokens_seen": 212755365,
      "step": 9876,
      "time_per_iteration": 2.501455307006836
    },
    {
      "auxiliary_loss_clip": 0.01096131,
      "auxiliary_loss_mlp": 0.01029737,
      "balance_loss_clip": 1.03669047,
      "balance_loss_mlp": 1.01772714,
      "epoch": 0.5938373666015332,
      "flos": 22561489376640.0,
      "grad_norm": 1.4844602311851767,
      "language_loss": 0.75932467,
      "learning_rate": 1.494755415907243e-06,
      "loss": 0.78058332,
      "num_input_tokens_seen": 212773875,
      "step": 9877,
      "time_per_iteration": 2.5205228328704834
    },
    {
      "auxiliary_loss_clip": 0.01100634,
      "auxiliary_loss_mlp": 0.01031299,
      "balance_loss_clip": 1.03684437,
      "balance_loss_mlp": 1.01781678,
      "epoch": 0.5938974898542011,
      "flos": 18440883936000.0,
      "grad_norm": 2.240245576893391,
      "language_loss": 0.81115097,
      "learning_rate": 1.4943785960076522e-06,
      "loss": 0.8324703,
      "num_input_tokens_seen": 212790590,
      "step": 9878,
      "time_per_iteration": 2.467648506164551
    },
    {
      "auxiliary_loss_clip": 0.0108592,
      "auxiliary_loss_mlp": 0.00789427,
      "balance_loss_clip": 1.03731871,
      "balance_loss_mlp": 1.01361036,
      "epoch": 0.5939576131068691,
      "flos": 45586728270720.0,
      "grad_norm": 2.013554050118306,
      "language_loss": 0.7108115,
      "learning_rate": 1.4940017952812754e-06,
      "loss": 0.72956491,
      "num_input_tokens_seen": 212812265,
      "step": 9879,
      "time_per_iteration": 2.748717784881592
    },
    {
      "auxiliary_loss_clip": 0.01099443,
      "auxiliary_loss_mlp": 0.01032817,
      "balance_loss_clip": 1.03882217,
      "balance_loss_mlp": 1.02052689,
      "epoch": 0.594017736359537,
      "flos": 23587708942080.0,
      "grad_norm": 1.6633596186124735,
      "language_loss": 0.57777631,
      "learning_rate": 1.493625013742401e-06,
      "loss": 0.59909892,
      "num_input_tokens_seen": 212831915,
      "step": 9880,
      "time_per_iteration": 3.878962755203247
    },
    {
      "auxiliary_loss_clip": 0.01101919,
      "auxiliary_loss_mlp": 0.01038005,
      "balance_loss_clip": 1.03806078,
      "balance_loss_mlp": 1.02470684,
      "epoch": 0.594077859612205,
      "flos": 29457235589760.0,
      "grad_norm": 1.7697321674140203,
      "language_loss": 0.77656424,
      "learning_rate": 1.4932482514053177e-06,
      "loss": 0.7979635,
      "num_input_tokens_seen": 212851350,
      "step": 9881,
      "time_per_iteration": 2.5421557426452637
    },
    {
      "auxiliary_loss_clip": 0.01098524,
      "auxiliary_loss_mlp": 0.0102823,
      "balance_loss_clip": 1.03649139,
      "balance_loss_mlp": 1.01602292,
      "epoch": 0.594137982864873,
      "flos": 16800089644800.0,
      "grad_norm": 2.1422483564596257,
      "language_loss": 0.82581383,
      "learning_rate": 1.4928715082843112e-06,
      "loss": 0.84708142,
      "num_input_tokens_seen": 212867995,
      "step": 9882,
      "time_per_iteration": 2.466517686843872
    },
    {
      "auxiliary_loss_clip": 0.01100199,
      "auxiliary_loss_mlp": 0.01035577,
      "balance_loss_clip": 1.03823781,
      "balance_loss_mlp": 1.02385902,
      "epoch": 0.594198106117541,
      "flos": 12750263953920.0,
      "grad_norm": 2.4012796297569277,
      "language_loss": 0.79718667,
      "learning_rate": 1.492494784393667e-06,
      "loss": 0.81854445,
      "num_input_tokens_seen": 212885220,
      "step": 9883,
      "time_per_iteration": 2.4641411304473877
    },
    {
      "auxiliary_loss_clip": 0.01085098,
      "auxiliary_loss_mlp": 0.0078619,
      "balance_loss_clip": 1.04476762,
      "balance_loss_mlp": 1.01141906,
      "epoch": 0.5942582293702089,
      "flos": 20996538652800.0,
      "grad_norm": 2.134529982944676,
      "language_loss": 0.74623209,
      "learning_rate": 1.4921180797476725e-06,
      "loss": 0.76494491,
      "num_input_tokens_seen": 212903195,
      "step": 9884,
      "time_per_iteration": 2.5674521923065186
    },
    {
      "auxiliary_loss_clip": 0.01114208,
      "auxiliary_loss_mlp": 0.01032101,
      "balance_loss_clip": 1.04095745,
      "balance_loss_mlp": 1.01915443,
      "epoch": 0.5943183526228769,
      "flos": 28291431772800.0,
      "grad_norm": 2.340225712451861,
      "language_loss": 0.66142064,
      "learning_rate": 1.4917413943606106e-06,
      "loss": 0.68288374,
      "num_input_tokens_seen": 212923340,
      "step": 9885,
      "time_per_iteration": 2.5157217979431152
    },
    {
      "auxiliary_loss_clip": 0.01089189,
      "auxiliary_loss_mlp": 0.01039127,
      "balance_loss_clip": 1.03976357,
      "balance_loss_mlp": 1.02637124,
      "epoch": 0.5943784758755448,
      "flos": 26614619118720.0,
      "grad_norm": 2.4255074821586575,
      "language_loss": 0.77457261,
      "learning_rate": 1.4913647282467667e-06,
      "loss": 0.79585576,
      "num_input_tokens_seen": 212942755,
      "step": 9886,
      "time_per_iteration": 2.573957920074463
    },
    {
      "auxiliary_loss_clip": 0.0103233,
      "auxiliary_loss_mlp": 0.0100115,
      "balance_loss_clip": 1.022367,
      "balance_loss_mlp": 0.99993438,
      "epoch": 0.5944385991282128,
      "flos": 64190935347840.0,
      "grad_norm": 0.862837093346675,
      "language_loss": 0.64530551,
      "learning_rate": 1.490988081420423e-06,
      "loss": 0.66564029,
      "num_input_tokens_seen": 212999355,
      "step": 9887,
      "time_per_iteration": 2.991119861602783
    },
    {
      "auxiliary_loss_clip": 0.01096838,
      "auxiliary_loss_mlp": 0.01030106,
      "balance_loss_clip": 1.03936362,
      "balance_loss_mlp": 1.01738667,
      "epoch": 0.5944987223808808,
      "flos": 19571998193280.0,
      "grad_norm": 2.018216627000778,
      "language_loss": 0.69329154,
      "learning_rate": 1.4906114538958615e-06,
      "loss": 0.71456099,
      "num_input_tokens_seen": 213018570,
      "step": 9888,
      "time_per_iteration": 2.5026822090148926
    },
    {
      "auxiliary_loss_clip": 0.01084951,
      "auxiliary_loss_mlp": 0.010316,
      "balance_loss_clip": 1.03680241,
      "balance_loss_mlp": 1.0177002,
      "epoch": 0.5945588456335488,
      "flos": 26177586341760.0,
      "grad_norm": 1.5614394370071636,
      "language_loss": 0.79486835,
      "learning_rate": 1.490234845687366e-06,
      "loss": 0.81603384,
      "num_input_tokens_seen": 213037735,
      "step": 9889,
      "time_per_iteration": 3.958974599838257
    },
    {
      "auxiliary_loss_clip": 0.01076651,
      "auxiliary_loss_mlp": 0.01031999,
      "balance_loss_clip": 1.03798008,
      "balance_loss_mlp": 1.01938057,
      "epoch": 0.5946189688862168,
      "flos": 20446494710400.0,
      "grad_norm": 2.0143273525859646,
      "language_loss": 0.70347524,
      "learning_rate": 1.4898582568092154e-06,
      "loss": 0.72456175,
      "num_input_tokens_seen": 213057160,
      "step": 9890,
      "time_per_iteration": 2.544884443283081
    },
    {
      "auxiliary_loss_clip": 0.01078601,
      "auxiliary_loss_mlp": 0.01032309,
      "balance_loss_clip": 1.03856206,
      "balance_loss_mlp": 1.01922548,
      "epoch": 0.5946790921388847,
      "flos": 13437521850240.0,
      "grad_norm": 1.9948031566568865,
      "language_loss": 0.69487274,
      "learning_rate": 1.489481687275691e-06,
      "loss": 0.71598184,
      "num_input_tokens_seen": 213073630,
      "step": 9891,
      "time_per_iteration": 3.916905641555786
    },
    {
      "auxiliary_loss_clip": 0.01098576,
      "auxiliary_loss_mlp": 0.01034386,
      "balance_loss_clip": 1.0378921,
      "balance_loss_mlp": 1.02174997,
      "epoch": 0.5947392153915527,
      "flos": 20412272027520.0,
      "grad_norm": 1.79919145934123,
      "language_loss": 0.53321147,
      "learning_rate": 1.4891051371010726e-06,
      "loss": 0.55454111,
      "num_input_tokens_seen": 213092450,
      "step": 9892,
      "time_per_iteration": 2.4892539978027344
    },
    {
      "auxiliary_loss_clip": 0.0101252,
      "auxiliary_loss_mlp": 0.01001987,
      "balance_loss_clip": 1.02158308,
      "balance_loss_mlp": 1.00059783,
      "epoch": 0.5947993386442206,
      "flos": 65619138994560.0,
      "grad_norm": 0.6648049247026984,
      "language_loss": 0.54517436,
      "learning_rate": 1.4887286062996375e-06,
      "loss": 0.56531942,
      "num_input_tokens_seen": 213155465,
      "step": 9893,
      "time_per_iteration": 3.221587896347046
    },
    {
      "auxiliary_loss_clip": 0.01072481,
      "auxiliary_loss_mlp": 0.01028502,
      "balance_loss_clip": 1.0386219,
      "balance_loss_mlp": 1.0169028,
      "epoch": 0.5948594618968887,
      "flos": 23183103168000.0,
      "grad_norm": 1.6150938406979223,
      "language_loss": 0.74806881,
      "learning_rate": 1.4883520948856658e-06,
      "loss": 0.76907855,
      "num_input_tokens_seen": 213174875,
      "step": 9894,
      "time_per_iteration": 4.0031208992004395
    },
    {
      "auxiliary_loss_clip": 0.01078028,
      "auxiliary_loss_mlp": 0.01027931,
      "balance_loss_clip": 1.03744316,
      "balance_loss_mlp": 1.01592112,
      "epoch": 0.5949195851495566,
      "flos": 13626771632640.0,
      "grad_norm": 1.8328721191747943,
      "language_loss": 0.7779544,
      "learning_rate": 1.487975602873434e-06,
      "loss": 0.79901403,
      "num_input_tokens_seen": 213192695,
      "step": 9895,
      "time_per_iteration": 2.536914110183716
    },
    {
      "auxiliary_loss_clip": 0.01061158,
      "auxiliary_loss_mlp": 0.01031304,
      "balance_loss_clip": 1.03524113,
      "balance_loss_mlp": 1.01796484,
      "epoch": 0.5949797084022246,
      "flos": 19751012599680.0,
      "grad_norm": 1.8102334658804082,
      "language_loss": 0.79003006,
      "learning_rate": 1.4875991302772182e-06,
      "loss": 0.81095469,
      "num_input_tokens_seen": 213211195,
      "step": 9896,
      "time_per_iteration": 2.6095926761627197
    },
    {
      "auxiliary_loss_clip": 0.0110129,
      "auxiliary_loss_mlp": 0.01033005,
      "balance_loss_clip": 1.03842342,
      "balance_loss_mlp": 1.02022529,
      "epoch": 0.5950398316548925,
      "flos": 25773878407680.0,
      "grad_norm": 1.4755857071709906,
      "language_loss": 0.83230621,
      "learning_rate": 1.4872226771112954e-06,
      "loss": 0.85364914,
      "num_input_tokens_seen": 213231975,
      "step": 9897,
      "time_per_iteration": 2.5507583618164062
    },
    {
      "auxiliary_loss_clip": 0.01094842,
      "auxiliary_loss_mlp": 0.01034288,
      "balance_loss_clip": 1.04020929,
      "balance_loss_mlp": 1.02184904,
      "epoch": 0.5950999549075605,
      "flos": 23039029716480.0,
      "grad_norm": 1.8631106953418137,
      "language_loss": 0.70362848,
      "learning_rate": 1.486846243389939e-06,
      "loss": 0.7249198,
      "num_input_tokens_seen": 213249760,
      "step": 9898,
      "time_per_iteration": 2.539904832839966
    },
    {
      "auxiliary_loss_clip": 0.01101505,
      "auxiliary_loss_mlp": 0.01042889,
      "balance_loss_clip": 1.03783953,
      "balance_loss_mlp": 1.02703393,
      "epoch": 0.5951600781602284,
      "flos": 32446367637120.0,
      "grad_norm": 2.237914418290006,
      "language_loss": 0.64122039,
      "learning_rate": 1.4864698291274251e-06,
      "loss": 0.66266435,
      "num_input_tokens_seen": 213269890,
      "step": 9899,
      "time_per_iteration": 2.578726291656494
    },
    {
      "auxiliary_loss_clip": 0.01109309,
      "auxiliary_loss_mlp": 0.01026389,
      "balance_loss_clip": 1.03830779,
      "balance_loss_mlp": 1.01532078,
      "epoch": 0.5952202014128964,
      "flos": 23800874204160.0,
      "grad_norm": 1.6573328855171923,
      "language_loss": 0.72140336,
      "learning_rate": 1.4860934343380267e-06,
      "loss": 0.74276036,
      "num_input_tokens_seen": 213289400,
      "step": 9900,
      "time_per_iteration": 2.5060765743255615
    },
    {
      "auxiliary_loss_clip": 0.01108556,
      "auxiliary_loss_mlp": 0.01029682,
      "balance_loss_clip": 1.03844059,
      "balance_loss_mlp": 1.01740289,
      "epoch": 0.5952803246655644,
      "flos": 22492182084480.0,
      "grad_norm": 1.6924728637463484,
      "language_loss": 0.84392416,
      "learning_rate": 1.4857170590360169e-06,
      "loss": 0.8653065,
      "num_input_tokens_seen": 213308040,
      "step": 9901,
      "time_per_iteration": 2.4726016521453857
    },
    {
      "auxiliary_loss_clip": 0.01005666,
      "auxiliary_loss_mlp": 0.01007597,
      "balance_loss_clip": 1.02270305,
      "balance_loss_mlp": 1.00636292,
      "epoch": 0.5953404479182324,
      "flos": 51234688851840.0,
      "grad_norm": 0.8024002967358176,
      "language_loss": 0.58276147,
      "learning_rate": 1.4853407032356674e-06,
      "loss": 0.60289413,
      "num_input_tokens_seen": 213358585,
      "step": 9902,
      "time_per_iteration": 3.09171724319458
    },
    {
      "auxiliary_loss_clip": 0.01054401,
      "auxiliary_loss_mlp": 0.01032851,
      "balance_loss_clip": 1.03985071,
      "balance_loss_mlp": 1.02018487,
      "epoch": 0.5954005711709004,
      "flos": 23112682554240.0,
      "grad_norm": 1.677296382406316,
      "language_loss": 0.77128893,
      "learning_rate": 1.4849643669512503e-06,
      "loss": 0.79216152,
      "num_input_tokens_seen": 213379585,
      "step": 9903,
      "time_per_iteration": 2.887528657913208
    },
    {
      "auxiliary_loss_clip": 0.01075971,
      "auxiliary_loss_mlp": 0.0103487,
      "balance_loss_clip": 1.04032373,
      "balance_loss_mlp": 1.02293682,
      "epoch": 0.5954606944235683,
      "flos": 35954732736000.0,
      "grad_norm": 2.715804357459143,
      "language_loss": 0.78005838,
      "learning_rate": 1.4845880501970362e-06,
      "loss": 0.80116677,
      "num_input_tokens_seen": 213401465,
      "step": 9904,
      "time_per_iteration": 2.6910061836242676
    },
    {
      "auxiliary_loss_clip": 0.011003,
      "auxiliary_loss_mlp": 0.01034715,
      "balance_loss_clip": 1.04118919,
      "balance_loss_mlp": 1.02167392,
      "epoch": 0.5955208176762363,
      "flos": 30443665864320.0,
      "grad_norm": 1.599259808304985,
      "language_loss": 0.72802532,
      "learning_rate": 1.4842117529872942e-06,
      "loss": 0.74937546,
      "num_input_tokens_seen": 213422720,
      "step": 9905,
      "time_per_iteration": 2.582089900970459
    },
    {
      "auxiliary_loss_clip": 0.01101946,
      "auxiliary_loss_mlp": 0.01031396,
      "balance_loss_clip": 1.03781736,
      "balance_loss_mlp": 1.01852167,
      "epoch": 0.5955809409289042,
      "flos": 17640112083840.0,
      "grad_norm": 1.8894650969618925,
      "language_loss": 0.69754016,
      "learning_rate": 1.483835475336295e-06,
      "loss": 0.71887362,
      "num_input_tokens_seen": 213439480,
      "step": 9906,
      "time_per_iteration": 2.4769270420074463
    },
    {
      "auxiliary_loss_clip": 0.01101669,
      "auxiliary_loss_mlp": 0.01036825,
      "balance_loss_clip": 1.04016018,
      "balance_loss_mlp": 1.02411699,
      "epoch": 0.5956410641815723,
      "flos": 24279887001600.0,
      "grad_norm": 2.0475935065787176,
      "language_loss": 0.75090694,
      "learning_rate": 1.4834592172583057e-06,
      "loss": 0.77229184,
      "num_input_tokens_seen": 213458895,
      "step": 9907,
      "time_per_iteration": 2.5450284481048584
    },
    {
      "auxiliary_loss_clip": 0.01086305,
      "auxiliary_loss_mlp": 0.01034422,
      "balance_loss_clip": 1.04040527,
      "balance_loss_mlp": 1.02199471,
      "epoch": 0.5957011874342402,
      "flos": 35734277013120.0,
      "grad_norm": 1.4449155807552163,
      "language_loss": 0.66762429,
      "learning_rate": 1.483082978767595e-06,
      "loss": 0.68883157,
      "num_input_tokens_seen": 213481730,
      "step": 9908,
      "time_per_iteration": 2.63934588432312
    },
    {
      "auxiliary_loss_clip": 0.01039039,
      "auxiliary_loss_mlp": 0.01031933,
      "balance_loss_clip": 1.03553379,
      "balance_loss_mlp": 1.02002418,
      "epoch": 0.5957613106869082,
      "flos": 21245004005760.0,
      "grad_norm": 2.03279110852491,
      "language_loss": 0.76309848,
      "learning_rate": 1.4827067598784298e-06,
      "loss": 0.78380817,
      "num_input_tokens_seen": 213497225,
      "step": 9909,
      "time_per_iteration": 2.656296730041504
    },
    {
      "auxiliary_loss_clip": 0.01040471,
      "auxiliary_loss_mlp": 0.01007415,
      "balance_loss_clip": 1.01593447,
      "balance_loss_mlp": 1.00614572,
      "epoch": 0.5958214339395761,
      "flos": 65940969876480.0,
      "grad_norm": 0.9275788849983656,
      "language_loss": 0.73391551,
      "learning_rate": 1.4823305606050753e-06,
      "loss": 0.75439435,
      "num_input_tokens_seen": 213556890,
      "step": 9910,
      "time_per_iteration": 3.1315133571624756
    },
    {
      "auxiliary_loss_clip": 0.0108864,
      "auxiliary_loss_mlp": 0.01034284,
      "balance_loss_clip": 1.04212213,
      "balance_loss_mlp": 1.02173114,
      "epoch": 0.5958815571922441,
      "flos": 23218690567680.0,
      "grad_norm": 1.6124512825927848,
      "language_loss": 0.69472575,
      "learning_rate": 1.481954380961799e-06,
      "loss": 0.71595502,
      "num_input_tokens_seen": 213575800,
      "step": 9911,
      "time_per_iteration": 2.5535411834716797
    },
    {
      "auxiliary_loss_clip": 0.01109977,
      "auxiliary_loss_mlp": 0.01036141,
      "balance_loss_clip": 1.04162765,
      "balance_loss_mlp": 1.0226469,
      "epoch": 0.595941680444912,
      "flos": 16538623568640.0,
      "grad_norm": 1.9116322958736558,
      "language_loss": 0.65578735,
      "learning_rate": 1.4815782209628631e-06,
      "loss": 0.67724854,
      "num_input_tokens_seen": 213592740,
      "step": 9912,
      "time_per_iteration": 2.451472759246826
    },
    {
      "auxiliary_loss_clip": 0.01081522,
      "auxiliary_loss_mlp": 0.01034869,
      "balance_loss_clip": 1.03756559,
      "balance_loss_mlp": 1.02162516,
      "epoch": 0.59600180369758,
      "flos": 27818883423360.0,
      "grad_norm": 2.0174755107934064,
      "language_loss": 0.73462057,
      "learning_rate": 1.4812020806225337e-06,
      "loss": 0.75578445,
      "num_input_tokens_seen": 213611970,
      "step": 9913,
      "time_per_iteration": 2.600966215133667
    },
    {
      "auxiliary_loss_clip": 0.01086639,
      "auxiliary_loss_mlp": 0.00786152,
      "balance_loss_clip": 1.04114628,
      "balance_loss_mlp": 1.00907373,
      "epoch": 0.596061926950248,
      "flos": 29491566013440.0,
      "grad_norm": 2.0391472080313835,
      "language_loss": 0.80441689,
      "learning_rate": 1.4808259599550738e-06,
      "loss": 0.82314479,
      "num_input_tokens_seen": 213632230,
      "step": 9914,
      "time_per_iteration": 2.6229355335235596
    },
    {
      "auxiliary_loss_clip": 0.01077418,
      "auxiliary_loss_mlp": 0.01034165,
      "balance_loss_clip": 1.03807545,
      "balance_loss_mlp": 1.02171922,
      "epoch": 0.596122050202916,
      "flos": 16836790366080.0,
      "grad_norm": 1.860059397535033,
      "language_loss": 0.67537236,
      "learning_rate": 1.4804498589747448e-06,
      "loss": 0.69648826,
      "num_input_tokens_seen": 213649645,
      "step": 9915,
      "time_per_iteration": 2.5574097633361816
    },
    {
      "auxiliary_loss_clip": 0.01087435,
      "auxiliary_loss_mlp": 0.01033603,
      "balance_loss_clip": 1.04017234,
      "balance_loss_mlp": 1.02143192,
      "epoch": 0.596182173455584,
      "flos": 20996646393600.0,
      "grad_norm": 1.6779341832809922,
      "language_loss": 0.78899252,
      "learning_rate": 1.4800737776958095e-06,
      "loss": 0.81020296,
      "num_input_tokens_seen": 213668850,
      "step": 9916,
      "time_per_iteration": 2.5337300300598145
    },
    {
      "auxiliary_loss_clip": 0.01090272,
      "auxiliary_loss_mlp": 0.01030851,
      "balance_loss_clip": 1.03732061,
      "balance_loss_mlp": 1.01813126,
      "epoch": 0.5962422967082519,
      "flos": 16065680169600.0,
      "grad_norm": 2.112220237370678,
      "language_loss": 0.82733524,
      "learning_rate": 1.4796977161325286e-06,
      "loss": 0.8485465,
      "num_input_tokens_seen": 213685695,
      "step": 9917,
      "time_per_iteration": 2.5230658054351807
    },
    {
      "auxiliary_loss_clip": 0.01084245,
      "auxiliary_loss_mlp": 0.01036425,
      "balance_loss_clip": 1.03763437,
      "balance_loss_mlp": 1.02403355,
      "epoch": 0.5963024199609199,
      "flos": 12166966995840.0,
      "grad_norm": 1.998308255896773,
      "language_loss": 0.77536386,
      "learning_rate": 1.4793216742991625e-06,
      "loss": 0.79657054,
      "num_input_tokens_seen": 213703515,
      "step": 9918,
      "time_per_iteration": 2.508625030517578
    },
    {
      "auxiliary_loss_clip": 0.011028,
      "auxiliary_loss_mlp": 0.01037278,
      "balance_loss_clip": 1.04080439,
      "balance_loss_mlp": 1.02462947,
      "epoch": 0.5963625432135878,
      "flos": 28074280101120.0,
      "grad_norm": 1.4941230028775716,
      "language_loss": 0.78969336,
      "learning_rate": 1.4789456522099707e-06,
      "loss": 0.81109411,
      "num_input_tokens_seen": 213724170,
      "step": 9919,
      "time_per_iteration": 3.9412906169891357
    },
    {
      "auxiliary_loss_clip": 0.01087666,
      "auxiliary_loss_mlp": 0.01032964,
      "balance_loss_clip": 1.04023504,
      "balance_loss_mlp": 1.01963091,
      "epoch": 0.5964226664662559,
      "flos": 19860324664320.0,
      "grad_norm": 6.158667021537544,
      "language_loss": 0.77535331,
      "learning_rate": 1.4785696498792122e-06,
      "loss": 0.79655957,
      "num_input_tokens_seen": 213740620,
      "step": 9920,
      "time_per_iteration": 2.521289587020874
    },
    {
      "auxiliary_loss_clip": 0.01100342,
      "auxiliary_loss_mlp": 0.01036639,
      "balance_loss_clip": 1.04024756,
      "balance_loss_mlp": 1.02221429,
      "epoch": 0.5964827897189238,
      "flos": 12932618325120.0,
      "grad_norm": 2.357372067891289,
      "language_loss": 0.82598889,
      "learning_rate": 1.4781936673211446e-06,
      "loss": 0.8473587,
      "num_input_tokens_seen": 213755390,
      "step": 9921,
      "time_per_iteration": 2.4739091396331787
    },
    {
      "auxiliary_loss_clip": 0.01100088,
      "auxiliary_loss_mlp": 0.01031538,
      "balance_loss_clip": 1.04002142,
      "balance_loss_mlp": 1.01759648,
      "epoch": 0.5965429129715918,
      "flos": 18150797698560.0,
      "grad_norm": 1.8921866960932194,
      "language_loss": 0.80833578,
      "learning_rate": 1.4778177045500252e-06,
      "loss": 0.82965207,
      "num_input_tokens_seen": 213773225,
      "step": 9922,
      "time_per_iteration": 2.4793074131011963
    },
    {
      "auxiliary_loss_clip": 0.01102302,
      "auxiliary_loss_mlp": 0.00784794,
      "balance_loss_clip": 1.03800893,
      "balance_loss_mlp": 1.00918078,
      "epoch": 0.5966030362242597,
      "flos": 21763231476480.0,
      "grad_norm": 2.342034338988647,
      "language_loss": 0.77173853,
      "learning_rate": 1.477441761580111e-06,
      "loss": 0.79060948,
      "num_input_tokens_seen": 213791860,
      "step": 9923,
      "time_per_iteration": 2.5016345977783203
    },
    {
      "auxiliary_loss_clip": 0.01099735,
      "auxiliary_loss_mlp": 0.01036581,
      "balance_loss_clip": 1.04068804,
      "balance_loss_mlp": 1.02154887,
      "epoch": 0.5966631594769277,
      "flos": 18807208790400.0,
      "grad_norm": 1.958172843801533,
      "language_loss": 0.76033211,
      "learning_rate": 1.4770658384256573e-06,
      "loss": 0.78169525,
      "num_input_tokens_seen": 213809455,
      "step": 9924,
      "time_per_iteration": 2.5159738063812256
    },
    {
      "auxiliary_loss_clip": 0.01093692,
      "auxiliary_loss_mlp": 0.01038972,
      "balance_loss_clip": 1.03634071,
      "balance_loss_mlp": 1.02405918,
      "epoch": 0.5967232827295956,
      "flos": 14064163545600.0,
      "grad_norm": 1.905418099156943,
      "language_loss": 0.66658163,
      "learning_rate": 1.4766899351009204e-06,
      "loss": 0.68790829,
      "num_input_tokens_seen": 213826615,
      "step": 9925,
      "time_per_iteration": 2.4532430171966553
    },
    {
      "auxiliary_loss_clip": 0.01085397,
      "auxiliary_loss_mlp": 0.01038931,
      "balance_loss_clip": 1.04368818,
      "balance_loss_mlp": 1.02529347,
      "epoch": 0.5967834059822636,
      "flos": 17238235743360.0,
      "grad_norm": 2.082044141284348,
      "language_loss": 0.71400452,
      "learning_rate": 1.4763140516201528e-06,
      "loss": 0.73524779,
      "num_input_tokens_seen": 213844495,
      "step": 9926,
      "time_per_iteration": 2.5140082836151123
    },
    {
      "auxiliary_loss_clip": 0.01067154,
      "auxiliary_loss_mlp": 0.00788388,
      "balance_loss_clip": 1.03603327,
      "balance_loss_mlp": 1.01054525,
      "epoch": 0.5968435292349316,
      "flos": 42520244284800.0,
      "grad_norm": 1.8901173497320498,
      "language_loss": 0.70230258,
      "learning_rate": 1.4759381879976088e-06,
      "loss": 0.72085798,
      "num_input_tokens_seen": 213869125,
      "step": 9927,
      "time_per_iteration": 2.775291919708252
    },
    {
      "auxiliary_loss_clip": 0.01071971,
      "auxiliary_loss_mlp": 0.01030473,
      "balance_loss_clip": 1.04107928,
      "balance_loss_mlp": 1.01650143,
      "epoch": 0.5969036524875996,
      "flos": 37630898945280.0,
      "grad_norm": 1.8761927242970284,
      "language_loss": 0.6366837,
      "learning_rate": 1.4755623442475415e-06,
      "loss": 0.65770817,
      "num_input_tokens_seen": 213891115,
      "step": 9928,
      "time_per_iteration": 4.160938262939453
    },
    {
      "auxiliary_loss_clip": 0.01109795,
      "auxiliary_loss_mlp": 0.01031217,
      "balance_loss_clip": 1.03759372,
      "balance_loss_mlp": 1.01892602,
      "epoch": 0.5969637757402676,
      "flos": 23148377694720.0,
      "grad_norm": 1.6192415440609704,
      "language_loss": 0.69579244,
      "learning_rate": 1.4751865203842022e-06,
      "loss": 0.71720254,
      "num_input_tokens_seen": 213911925,
      "step": 9929,
      "time_per_iteration": 3.9009790420532227
    },
    {
      "auxiliary_loss_clip": 0.01065301,
      "auxiliary_loss_mlp": 0.01030544,
      "balance_loss_clip": 1.03899539,
      "balance_loss_mlp": 1.01898623,
      "epoch": 0.5970238989929355,
      "flos": 24020934877440.0,
      "grad_norm": 1.7669403123104326,
      "language_loss": 0.76367021,
      "learning_rate": 1.4748107164218431e-06,
      "loss": 0.78462863,
      "num_input_tokens_seen": 213930715,
      "step": 9930,
      "time_per_iteration": 2.596670627593994
    },
    {
      "auxiliary_loss_clip": 0.01091422,
      "auxiliary_loss_mlp": 0.01033737,
      "balance_loss_clip": 1.04268742,
      "balance_loss_mlp": 1.01958084,
      "epoch": 0.5970840222456035,
      "flos": 19426883247360.0,
      "grad_norm": 1.5495078475348034,
      "language_loss": 0.6887964,
      "learning_rate": 1.4744349323747146e-06,
      "loss": 0.71004808,
      "num_input_tokens_seen": 213950015,
      "step": 9931,
      "time_per_iteration": 2.5316030979156494
    },
    {
      "auxiliary_loss_clip": 0.01031462,
      "auxiliary_loss_mlp": 0.01002353,
      "balance_loss_clip": 1.0212822,
      "balance_loss_mlp": 1.00084484,
      "epoch": 0.5971441454982714,
      "flos": 62976615235200.0,
      "grad_norm": 0.8549358942710722,
      "language_loss": 0.6422137,
      "learning_rate": 1.474059168257065e-06,
      "loss": 0.66255188,
      "num_input_tokens_seen": 214003330,
      "step": 9932,
      "time_per_iteration": 3.0554797649383545
    },
    {
      "auxiliary_loss_clip": 0.01081903,
      "auxiliary_loss_mlp": 0.01029666,
      "balance_loss_clip": 1.03865552,
      "balance_loss_mlp": 1.01628447,
      "epoch": 0.5972042687509395,
      "flos": 20266223328000.0,
      "grad_norm": 2.030760897076716,
      "language_loss": 0.73987758,
      "learning_rate": 1.4736834240831454e-06,
      "loss": 0.76099324,
      "num_input_tokens_seen": 214021680,
      "step": 9933,
      "time_per_iteration": 3.9462296962738037
    },
    {
      "auxiliary_loss_clip": 0.01042052,
      "auxiliary_loss_mlp": 0.01002729,
      "balance_loss_clip": 1.03416657,
      "balance_loss_mlp": 1.00120318,
      "epoch": 0.5972643920036074,
      "flos": 71652383832960.0,
      "grad_norm": 0.6685761530391885,
      "language_loss": 0.52001274,
      "learning_rate": 1.473307699867203e-06,
      "loss": 0.54046053,
      "num_input_tokens_seen": 214090265,
      "step": 9934,
      "time_per_iteration": 3.2040419578552246
    },
    {
      "auxiliary_loss_clip": 0.01043251,
      "auxiliary_loss_mlp": 0.01003202,
      "balance_loss_clip": 1.01859117,
      "balance_loss_mlp": 1.00195599,
      "epoch": 0.5973245152562754,
      "flos": 56892702263040.0,
      "grad_norm": 0.8303508399286358,
      "language_loss": 0.54184294,
      "learning_rate": 1.4729319956234849e-06,
      "loss": 0.56230748,
      "num_input_tokens_seen": 214146375,
      "step": 9935,
      "time_per_iteration": 3.0248618125915527
    },
    {
      "auxiliary_loss_clip": 0.01093962,
      "auxiliary_loss_mlp": 0.01035018,
      "balance_loss_clip": 1.04049563,
      "balance_loss_mlp": 1.02114224,
      "epoch": 0.5973846385089433,
      "flos": 24164361884160.0,
      "grad_norm": 1.5739976533121374,
      "language_loss": 0.659971,
      "learning_rate": 1.4725563113662394e-06,
      "loss": 0.68126082,
      "num_input_tokens_seen": 214165340,
      "step": 9936,
      "time_per_iteration": 2.555187463760376
    },
    {
      "auxiliary_loss_clip": 0.01063084,
      "auxiliary_loss_mlp": 0.01036489,
      "balance_loss_clip": 1.04163027,
      "balance_loss_mlp": 1.0235424,
      "epoch": 0.5974447617616113,
      "flos": 17670599752320.0,
      "grad_norm": 2.4972349090020196,
      "language_loss": 0.67417616,
      "learning_rate": 1.4721806471097103e-06,
      "loss": 0.69517189,
      "num_input_tokens_seen": 214181360,
      "step": 9937,
      "time_per_iteration": 2.578324556350708
    },
    {
      "auxiliary_loss_clip": 0.01104724,
      "auxiliary_loss_mlp": 0.01031505,
      "balance_loss_clip": 1.03940046,
      "balance_loss_mlp": 1.01831436,
      "epoch": 0.5975048850142792,
      "flos": 22892514140160.0,
      "grad_norm": 2.3822933670878768,
      "language_loss": 0.77450418,
      "learning_rate": 1.4718050028681442e-06,
      "loss": 0.79586637,
      "num_input_tokens_seen": 214198525,
      "step": 9938,
      "time_per_iteration": -0.15279340744018555
    },
    {
      "auxiliary_loss_clip": 0.01101715,
      "auxiliary_loss_mlp": 0.01030984,
      "balance_loss_clip": 1.03743696,
      "balance_loss_mlp": 1.01753688,
      "epoch": 0.5975650082669473,
      "flos": 24353108876160.0,
      "grad_norm": 1.639244316691942,
      "language_loss": 0.75531,
      "learning_rate": 1.4714293786557855e-06,
      "loss": 0.77663696,
      "num_input_tokens_seen": 214218710,
      "step": 9939,
      "time_per_iteration": 2.553797483444214
    },
    {
      "auxiliary_loss_clip": 0.01074603,
      "auxiliary_loss_mlp": 0.01034866,
      "balance_loss_clip": 1.04097939,
      "balance_loss_mlp": 1.01957774,
      "epoch": 0.5976251315196152,
      "flos": 20923352691840.0,
      "grad_norm": 2.1198116975492236,
      "language_loss": 0.67961419,
      "learning_rate": 1.471053774486878e-06,
      "loss": 0.70070887,
      "num_input_tokens_seen": 214237800,
      "step": 9940,
      "time_per_iteration": 2.6156911849975586
    },
    {
      "auxiliary_loss_clip": 0.01086864,
      "auxiliary_loss_mlp": 0.01033649,
      "balance_loss_clip": 1.04089689,
      "balance_loss_mlp": 1.02180552,
      "epoch": 0.5976852547722832,
      "flos": 35844594658560.0,
      "grad_norm": 1.3610750392125175,
      "language_loss": 0.7028901,
      "learning_rate": 1.470678190375664e-06,
      "loss": 0.72409523,
      "num_input_tokens_seen": 214260355,
      "step": 9941,
      "time_per_iteration": 2.688088893890381
    },
    {
      "auxiliary_loss_clip": 0.01090483,
      "auxiliary_loss_mlp": 0.01032781,
      "balance_loss_clip": 1.03843665,
      "balance_loss_mlp": 1.01941228,
      "epoch": 0.5977453780249512,
      "flos": 12855948744960.0,
      "grad_norm": 2.0442114160901745,
      "language_loss": 0.77498066,
      "learning_rate": 1.470302626336386e-06,
      "loss": 0.79621327,
      "num_input_tokens_seen": 214277120,
      "step": 9942,
      "time_per_iteration": 2.5859720706939697
    },
    {
      "auxiliary_loss_clip": 0.0106455,
      "auxiliary_loss_mlp": 0.01038434,
      "balance_loss_clip": 1.03912997,
      "balance_loss_mlp": 1.02556562,
      "epoch": 0.5978055012776191,
      "flos": 20959155573120.0,
      "grad_norm": 1.968505086226769,
      "language_loss": 0.75784367,
      "learning_rate": 1.4699270823832857e-06,
      "loss": 0.7788735,
      "num_input_tokens_seen": 214295300,
      "step": 9943,
      "time_per_iteration": 2.607978105545044
    },
    {
      "auxiliary_loss_clip": 0.01053578,
      "auxiliary_loss_mlp": 0.01033183,
      "balance_loss_clip": 1.03855121,
      "balance_loss_mlp": 1.02142906,
      "epoch": 0.5978656245302871,
      "flos": 34058003063040.0,
      "grad_norm": 1.8145597272272949,
      "language_loss": 0.62346232,
      "learning_rate": 1.4695515585306032e-06,
      "loss": 0.64432991,
      "num_input_tokens_seen": 214317050,
      "step": 9944,
      "time_per_iteration": 2.76664662361145
    },
    {
      "auxiliary_loss_clip": 0.01091241,
      "auxiliary_loss_mlp": 0.01032702,
      "balance_loss_clip": 1.03920972,
      "balance_loss_mlp": 1.01992893,
      "epoch": 0.597925747782955,
      "flos": 37373275624320.0,
      "grad_norm": 1.6320610528755304,
      "language_loss": 0.72717929,
      "learning_rate": 1.4691760547925795e-06,
      "loss": 0.74841875,
      "num_input_tokens_seen": 214337470,
      "step": 9945,
      "time_per_iteration": 2.6749441623687744
    },
    {
      "auxiliary_loss_clip": 0.01062879,
      "auxiliary_loss_mlp": 0.01034458,
      "balance_loss_clip": 1.0400672,
      "balance_loss_mlp": 1.02118385,
      "epoch": 0.5979858710356231,
      "flos": 25374803328000.0,
      "grad_norm": 1.9688571116306892,
      "language_loss": 0.66930079,
      "learning_rate": 1.4688005711834522e-06,
      "loss": 0.69027412,
      "num_input_tokens_seen": 214357975,
      "step": 9946,
      "time_per_iteration": 2.658874034881592
    },
    {
      "auxiliary_loss_clip": 0.01099908,
      "auxiliary_loss_mlp": 0.01042404,
      "balance_loss_clip": 1.03763676,
      "balance_loss_mlp": 1.02775884,
      "epoch": 0.598045994288291,
      "flos": 13698413308800.0,
      "grad_norm": 2.8121075167084593,
      "language_loss": 0.88837194,
      "learning_rate": 1.468425107717461e-06,
      "loss": 0.90979505,
      "num_input_tokens_seen": 214374125,
      "step": 9947,
      "time_per_iteration": 2.4754021167755127
    },
    {
      "auxiliary_loss_clip": 0.01106567,
      "auxiliary_loss_mlp": 0.0103345,
      "balance_loss_clip": 1.03714359,
      "balance_loss_mlp": 1.02244759,
      "epoch": 0.598106117540959,
      "flos": 21981352815360.0,
      "grad_norm": 1.7296805620945566,
      "language_loss": 0.72271764,
      "learning_rate": 1.4680496644088432e-06,
      "loss": 0.7441178,
      "num_input_tokens_seen": 214393395,
      "step": 9948,
      "time_per_iteration": 2.477919816970825
    },
    {
      "auxiliary_loss_clip": 0.01090151,
      "auxiliary_loss_mlp": 0.01032647,
      "balance_loss_clip": 1.03858924,
      "balance_loss_mlp": 1.01865768,
      "epoch": 0.5981662407936269,
      "flos": 20559362221440.0,
      "grad_norm": 1.8877963961587216,
      "language_loss": 0.89256656,
      "learning_rate": 1.4676742412718347e-06,
      "loss": 0.91379452,
      "num_input_tokens_seen": 214411550,
      "step": 9949,
      "time_per_iteration": 2.5178425312042236
    },
    {
      "auxiliary_loss_clip": 0.01098404,
      "auxiliary_loss_mlp": 0.01029049,
      "balance_loss_clip": 1.03819585,
      "balance_loss_mlp": 1.01755762,
      "epoch": 0.5982263640462949,
      "flos": 14063840323200.0,
      "grad_norm": 2.025294997686301,
      "language_loss": 0.70611203,
      "learning_rate": 1.467298838320673e-06,
      "loss": 0.72738659,
      "num_input_tokens_seen": 214429780,
      "step": 9950,
      "time_per_iteration": 2.458994150161743
    },
    {
      "auxiliary_loss_clip": 0.01100612,
      "auxiliary_loss_mlp": 0.01033299,
      "balance_loss_clip": 1.03830361,
      "balance_loss_mlp": 1.0205493,
      "epoch": 0.5982864872989628,
      "flos": 17707228646400.0,
      "grad_norm": 1.5658436154703812,
      "language_loss": 0.78309256,
      "learning_rate": 1.4669234555695921e-06,
      "loss": 0.80443162,
      "num_input_tokens_seen": 214447775,
      "step": 9951,
      "time_per_iteration": 2.482403516769409
    },
    {
      "auxiliary_loss_clip": 0.01090014,
      "auxiliary_loss_mlp": 0.01035477,
      "balance_loss_clip": 1.03847587,
      "balance_loss_mlp": 1.02188134,
      "epoch": 0.5983466105516309,
      "flos": 16764789553920.0,
      "grad_norm": 1.6459489315392206,
      "language_loss": 0.74011648,
      "learning_rate": 1.4665480930328275e-06,
      "loss": 0.76137137,
      "num_input_tokens_seen": 214467245,
      "step": 9952,
      "time_per_iteration": 2.5072500705718994
    },
    {
      "auxiliary_loss_clip": 0.01090399,
      "auxiliary_loss_mlp": 0.00790936,
      "balance_loss_clip": 1.03710818,
      "balance_loss_mlp": 1.01688266,
      "epoch": 0.5984067338042988,
      "flos": 20042714949120.0,
      "grad_norm": 3.179770037265985,
      "language_loss": 0.79040891,
      "learning_rate": 1.466172750724613e-06,
      "loss": 0.80922222,
      "num_input_tokens_seen": 214484385,
      "step": 9953,
      "time_per_iteration": 2.516258955001831
    },
    {
      "auxiliary_loss_clip": 0.01078635,
      "auxiliary_loss_mlp": 0.01030863,
      "balance_loss_clip": 1.0368048,
      "balance_loss_mlp": 1.01883459,
      "epoch": 0.5984668570569668,
      "flos": 26319900026880.0,
      "grad_norm": 1.4079486782394104,
      "language_loss": 0.69663417,
      "learning_rate": 1.4657974286591807e-06,
      "loss": 0.71772909,
      "num_input_tokens_seen": 214503465,
      "step": 9954,
      "time_per_iteration": 2.620380401611328
    },
    {
      "auxiliary_loss_clip": 0.0108761,
      "auxiliary_loss_mlp": 0.01029632,
      "balance_loss_clip": 1.03767979,
      "balance_loss_mlp": 1.01731133,
      "epoch": 0.5985269803096348,
      "flos": 20593728558720.0,
      "grad_norm": 1.7690164240785402,
      "language_loss": 0.73001599,
      "learning_rate": 1.4654221268507637e-06,
      "loss": 0.75118846,
      "num_input_tokens_seen": 214520725,
      "step": 9955,
      "time_per_iteration": 2.51125168800354
    },
    {
      "auxiliary_loss_clip": 0.01110852,
      "auxiliary_loss_mlp": 0.01029638,
      "balance_loss_clip": 1.0372901,
      "balance_loss_mlp": 1.01684666,
      "epoch": 0.5985871035623027,
      "flos": 26865382942080.0,
      "grad_norm": 1.7835559076894665,
      "language_loss": 0.68579543,
      "learning_rate": 1.4650468453135934e-06,
      "loss": 0.70720035,
      "num_input_tokens_seen": 214540675,
      "step": 9956,
      "time_per_iteration": 3.9317078590393066
    },
    {
      "auxiliary_loss_clip": 0.01114278,
      "auxiliary_loss_mlp": 0.01033938,
      "balance_loss_clip": 1.04028368,
      "balance_loss_mlp": 1.02096188,
      "epoch": 0.5986472268149707,
      "flos": 19609704495360.0,
      "grad_norm": 1.9673797068581835,
      "language_loss": 0.74095362,
      "learning_rate": 1.4646715840618999e-06,
      "loss": 0.76243579,
      "num_input_tokens_seen": 214559910,
      "step": 9957,
      "time_per_iteration": 2.4396958351135254
    },
    {
      "auxiliary_loss_clip": 0.01073535,
      "auxiliary_loss_mlp": 0.01028771,
      "balance_loss_clip": 1.03662837,
      "balance_loss_mlp": 1.01635551,
      "epoch": 0.5987073500676386,
      "flos": 21794616984960.0,
      "grad_norm": 1.7390693782133018,
      "language_loss": 0.84983194,
      "learning_rate": 1.4642963431099138e-06,
      "loss": 0.87085497,
      "num_input_tokens_seen": 214575960,
      "step": 9958,
      "time_per_iteration": 2.5433621406555176
    },
    {
      "auxiliary_loss_clip": 0.01080942,
      "auxiliary_loss_mlp": 0.00788761,
      "balance_loss_clip": 1.03764927,
      "balance_loss_mlp": 1.01539588,
      "epoch": 0.5987674733203067,
      "flos": 24314361079680.0,
      "grad_norm": 1.7129298842779497,
      "language_loss": 0.66301304,
      "learning_rate": 1.463921122471864e-06,
      "loss": 0.68171012,
      "num_input_tokens_seen": 214594230,
      "step": 9959,
      "time_per_iteration": 2.57487416267395
    },
    {
      "auxiliary_loss_clip": 0.01101359,
      "auxiliary_loss_mlp": 0.01029859,
      "balance_loss_clip": 1.03855109,
      "balance_loss_mlp": 1.01731253,
      "epoch": 0.5988275965729746,
      "flos": 21320201128320.0,
      "grad_norm": 1.699800112394332,
      "language_loss": 0.83356196,
      "learning_rate": 1.4635459221619796e-06,
      "loss": 0.85487413,
      "num_input_tokens_seen": 214613130,
      "step": 9960,
      "time_per_iteration": 2.5008766651153564
    },
    {
      "auxiliary_loss_clip": 0.01090775,
      "auxiliary_loss_mlp": 0.01026026,
      "balance_loss_clip": 1.03624046,
      "balance_loss_mlp": 1.01406932,
      "epoch": 0.5988877198256426,
      "flos": 25118041933440.0,
      "grad_norm": 1.6932379922753558,
      "language_loss": 0.79163313,
      "learning_rate": 1.4631707421944868e-06,
      "loss": 0.81280118,
      "num_input_tokens_seen": 214634470,
      "step": 9961,
      "time_per_iteration": 2.5360066890716553
    },
    {
      "auxiliary_loss_clip": 0.01109622,
      "auxiliary_loss_mlp": 0.0102993,
      "balance_loss_clip": 1.03743315,
      "balance_loss_mlp": 1.01742542,
      "epoch": 0.5989478430783105,
      "flos": 26429104350720.0,
      "grad_norm": 1.6161378047906223,
      "language_loss": 0.67028213,
      "learning_rate": 1.4627955825836136e-06,
      "loss": 0.69167769,
      "num_input_tokens_seen": 214654030,
      "step": 9962,
      "time_per_iteration": 2.5018529891967773
    },
    {
      "auxiliary_loss_clip": 0.0109492,
      "auxiliary_loss_mlp": 0.01036387,
      "balance_loss_clip": 1.03582644,
      "balance_loss_mlp": 1.02237964,
      "epoch": 0.5990079663309785,
      "flos": 25778439434880.0,
      "grad_norm": 2.4311265134033238,
      "language_loss": 0.74284446,
      "learning_rate": 1.4624204433435857e-06,
      "loss": 0.76415753,
      "num_input_tokens_seen": 214676985,
      "step": 9963,
      "time_per_iteration": 2.5437021255493164
    },
    {
      "auxiliary_loss_clip": 0.010987,
      "auxiliary_loss_mlp": 0.01031198,
      "balance_loss_clip": 1.03762984,
      "balance_loss_mlp": 1.01888371,
      "epoch": 0.5990680895836464,
      "flos": 36831779118720.0,
      "grad_norm": 1.9373227927078158,
      "language_loss": 0.68197942,
      "learning_rate": 1.4620453244886281e-06,
      "loss": 0.70327842,
      "num_input_tokens_seen": 214700105,
      "step": 9964,
      "time_per_iteration": 2.629603862762451
    },
    {
      "auxiliary_loss_clip": 0.01075555,
      "auxiliary_loss_mlp": 0.01029123,
      "balance_loss_clip": 1.03792787,
      "balance_loss_mlp": 1.01572418,
      "epoch": 0.5991282128363145,
      "flos": 24133550993280.0,
      "grad_norm": 2.140079384373744,
      "language_loss": 0.76875317,
      "learning_rate": 1.4616702260329662e-06,
      "loss": 0.78979993,
      "num_input_tokens_seen": 214717885,
      "step": 9965,
      "time_per_iteration": 2.54160475730896
    },
    {
      "auxiliary_loss_clip": 0.01096892,
      "auxiliary_loss_mlp": 0.01028845,
      "balance_loss_clip": 1.04016089,
      "balance_loss_mlp": 1.0164299,
      "epoch": 0.5991883360889824,
      "flos": 10304064956160.0,
      "grad_norm": 2.853214738541968,
      "language_loss": 0.77355063,
      "learning_rate": 1.4612951479908229e-06,
      "loss": 0.79480797,
      "num_input_tokens_seen": 214733680,
      "step": 9966,
      "time_per_iteration": 3.840423107147217
    },
    {
      "auxiliary_loss_clip": 0.01073903,
      "auxiliary_loss_mlp": 0.01027319,
      "balance_loss_clip": 1.03661287,
      "balance_loss_mlp": 1.01572585,
      "epoch": 0.5992484593416504,
      "flos": 23951196622080.0,
      "grad_norm": 1.5184891482317633,
      "language_loss": 0.7356196,
      "learning_rate": 1.460920090376422e-06,
      "loss": 0.75663185,
      "num_input_tokens_seen": 214753285,
      "step": 9967,
      "time_per_iteration": 2.5566506385803223
    },
    {
      "auxiliary_loss_clip": 0.01104107,
      "auxiliary_loss_mlp": 0.01033026,
      "balance_loss_clip": 1.03896761,
      "balance_loss_mlp": 1.02005625,
      "epoch": 0.5993085825943184,
      "flos": 11944105061760.0,
      "grad_norm": 2.1874404383089,
      "language_loss": 0.68645012,
      "learning_rate": 1.4605450532039847e-06,
      "loss": 0.70782149,
      "num_input_tokens_seen": 214767810,
      "step": 9968,
      "time_per_iteration": 3.831204652786255
    },
    {
      "auxiliary_loss_clip": 0.01099824,
      "auxiliary_loss_mlp": 0.01033307,
      "balance_loss_clip": 1.03700709,
      "balance_loss_mlp": 1.01981235,
      "epoch": 0.5993687058469863,
      "flos": 19026838500480.0,
      "grad_norm": 1.6247381169512767,
      "language_loss": 0.79320014,
      "learning_rate": 1.4601700364877334e-06,
      "loss": 0.81453145,
      "num_input_tokens_seen": 214786040,
      "step": 9969,
      "time_per_iteration": 2.4817569255828857
    },
    {
      "auxiliary_loss_clip": 0.01099553,
      "auxiliary_loss_mlp": 0.01030315,
      "balance_loss_clip": 1.03625309,
      "balance_loss_mlp": 1.01775646,
      "epoch": 0.5994288290996543,
      "flos": 14282967242880.0,
      "grad_norm": 3.259396347304868,
      "language_loss": 0.81121689,
      "learning_rate": 1.4597950402418889e-06,
      "loss": 0.8325156,
      "num_input_tokens_seen": 214803110,
      "step": 9970,
      "time_per_iteration": 2.464970350265503
    },
    {
      "auxiliary_loss_clip": 0.01060044,
      "auxiliary_loss_mlp": 0.01039992,
      "balance_loss_clip": 1.03763771,
      "balance_loss_mlp": 1.0249002,
      "epoch": 0.5994889523523222,
      "flos": 19206643006080.0,
      "grad_norm": 1.88287261595569,
      "language_loss": 0.62268108,
      "learning_rate": 1.4594200644806697e-06,
      "loss": 0.64368141,
      "num_input_tokens_seen": 214819945,
      "step": 9971,
      "time_per_iteration": 4.010250091552734
    },
    {
      "auxiliary_loss_clip": 0.01107305,
      "auxiliary_loss_mlp": 0.01032226,
      "balance_loss_clip": 1.03753161,
      "balance_loss_mlp": 1.02057898,
      "epoch": 0.5995490756049903,
      "flos": 28037040675840.0,
      "grad_norm": 1.619864367766327,
      "language_loss": 0.7910223,
      "learning_rate": 1.4590451092182962e-06,
      "loss": 0.81241763,
      "num_input_tokens_seen": 214838810,
      "step": 9972,
      "time_per_iteration": 2.5014054775238037
    },
    {
      "auxiliary_loss_clip": 0.01070036,
      "auxiliary_loss_mlp": 0.01033673,
      "balance_loss_clip": 1.03677845,
      "balance_loss_mlp": 1.02044678,
      "epoch": 0.5996091988576582,
      "flos": 29052953038080.0,
      "grad_norm": 2.3433235600783804,
      "language_loss": 0.76146877,
      "learning_rate": 1.4586701744689864e-06,
      "loss": 0.78250587,
      "num_input_tokens_seen": 214857040,
      "step": 9973,
      "time_per_iteration": 2.6118710041046143
    },
    {
      "auxiliary_loss_clip": 0.01074863,
      "auxiliary_loss_mlp": 0.01030518,
      "balance_loss_clip": 1.03901875,
      "balance_loss_mlp": 1.01800656,
      "epoch": 0.5996693221103262,
      "flos": 20813968800000.0,
      "grad_norm": 2.1030472768023993,
      "language_loss": 0.65293252,
      "learning_rate": 1.4582952602469578e-06,
      "loss": 0.67398632,
      "num_input_tokens_seen": 214873375,
      "step": 9974,
      "time_per_iteration": 2.558683395385742
    },
    {
      "auxiliary_loss_clip": 0.01099061,
      "auxiliary_loss_mlp": 0.01030271,
      "balance_loss_clip": 1.03643036,
      "balance_loss_mlp": 1.01751614,
      "epoch": 0.5997294453629941,
      "flos": 23768914078080.0,
      "grad_norm": 1.4002754057744817,
      "language_loss": 0.74702406,
      "learning_rate": 1.457920366566428e-06,
      "loss": 0.76831734,
      "num_input_tokens_seen": 214893900,
      "step": 9975,
      "time_per_iteration": 2.52449893951416
    },
    {
      "auxiliary_loss_clip": 0.01110682,
      "auxiliary_loss_mlp": 0.01028474,
      "balance_loss_clip": 1.03785825,
      "balance_loss_mlp": 1.01533723,
      "epoch": 0.5997895686156621,
      "flos": 20960017499520.0,
      "grad_norm": 1.8546596572625194,
      "language_loss": 0.77098382,
      "learning_rate": 1.457545493441611e-06,
      "loss": 0.79237545,
      "num_input_tokens_seen": 214912110,
      "step": 9976,
      "time_per_iteration": 2.4777016639709473
    },
    {
      "auxiliary_loss_clip": 0.01090892,
      "auxiliary_loss_mlp": 0.01038506,
      "balance_loss_clip": 1.03621447,
      "balance_loss_mlp": 1.02519083,
      "epoch": 0.59984969186833,
      "flos": 28365443746560.0,
      "grad_norm": 2.394676883213683,
      "language_loss": 0.74494195,
      "learning_rate": 1.4571706408867237e-06,
      "loss": 0.76623595,
      "num_input_tokens_seen": 214930140,
      "step": 9977,
      "time_per_iteration": 2.56866192817688
    },
    {
      "auxiliary_loss_clip": 0.01075392,
      "auxiliary_loss_mlp": 0.01031854,
      "balance_loss_clip": 1.03867257,
      "balance_loss_mlp": 1.01899171,
      "epoch": 0.5999098151209981,
      "flos": 22565906749440.0,
      "grad_norm": 2.122529937081405,
      "language_loss": 0.69012928,
      "learning_rate": 1.4567958089159802e-06,
      "loss": 0.71120179,
      "num_input_tokens_seen": 214949200,
      "step": 9978,
      "time_per_iteration": 2.576420783996582
    },
    {
      "auxiliary_loss_clip": 0.01118716,
      "auxiliary_loss_mlp": 0.01032802,
      "balance_loss_clip": 1.04264665,
      "balance_loss_mlp": 1.01955771,
      "epoch": 0.599969938373666,
      "flos": 18768712389120.0,
      "grad_norm": 1.930847031543128,
      "language_loss": 0.80789244,
      "learning_rate": 1.456420997543594e-06,
      "loss": 0.82940757,
      "num_input_tokens_seen": 214965775,
      "step": 9979,
      "time_per_iteration": 2.437883138656616
    },
    {
      "auxiliary_loss_clip": 0.01105676,
      "auxiliary_loss_mlp": 0.0103138,
      "balance_loss_clip": 1.03632534,
      "balance_loss_mlp": 1.01920891,
      "epoch": 0.600030061626334,
      "flos": 11327231865600.0,
      "grad_norm": 1.7943686882502095,
      "language_loss": 0.69894212,
      "learning_rate": 1.4560462067837782e-06,
      "loss": 0.72031271,
      "num_input_tokens_seen": 214982480,
      "step": 9980,
      "time_per_iteration": 2.443035125732422
    },
    {
      "auxiliary_loss_clip": 0.01100059,
      "auxiliary_loss_mlp": 0.01031282,
      "balance_loss_clip": 1.03742337,
      "balance_loss_mlp": 1.01756048,
      "epoch": 0.600090184879002,
      "flos": 16578664254720.0,
      "grad_norm": 2.5929201752022677,
      "language_loss": 0.68188822,
      "learning_rate": 1.4556714366507445e-06,
      "loss": 0.70320165,
      "num_input_tokens_seen": 214998110,
      "step": 9981,
      "time_per_iteration": 2.466353178024292
    },
    {
      "auxiliary_loss_clip": 0.01099796,
      "auxiliary_loss_mlp": 0.01031603,
      "balance_loss_clip": 1.03912926,
      "balance_loss_mlp": 1.02048683,
      "epoch": 0.6001503081316699,
      "flos": 23618627573760.0,
      "grad_norm": 3.8311568841443346,
      "language_loss": 0.78344584,
      "learning_rate": 1.4552966871587048e-06,
      "loss": 0.80475974,
      "num_input_tokens_seen": 215017995,
      "step": 9982,
      "time_per_iteration": 2.5512075424194336
    },
    {
      "auxiliary_loss_clip": 0.01060003,
      "auxiliary_loss_mlp": 0.01040685,
      "balance_loss_clip": 1.03550911,
      "balance_loss_mlp": 1.02544999,
      "epoch": 0.6002104313843379,
      "flos": 20667668705280.0,
      "grad_norm": 1.7094990159988908,
      "language_loss": 0.73001802,
      "learning_rate": 1.4549219583218686e-06,
      "loss": 0.7510249,
      "num_input_tokens_seen": 215038285,
      "step": 9983,
      "time_per_iteration": 2.5976078510284424
    },
    {
      "auxiliary_loss_clip": 0.01070241,
      "auxiliary_loss_mlp": 0.01030999,
      "balance_loss_clip": 1.03625476,
      "balance_loss_mlp": 1.0180769,
      "epoch": 0.6002705546370058,
      "flos": 22455229968000.0,
      "grad_norm": 3.580002728486087,
      "language_loss": 0.78129643,
      "learning_rate": 1.454547250154447e-06,
      "loss": 0.80230886,
      "num_input_tokens_seen": 215057825,
      "step": 9984,
      "time_per_iteration": 2.5612192153930664
    },
    {
      "auxiliary_loss_clip": 0.01099272,
      "auxiliary_loss_mlp": 0.01031012,
      "balance_loss_clip": 1.03801036,
      "balance_loss_mlp": 1.0189836,
      "epoch": 0.6003306778896739,
      "flos": 25191982080000.0,
      "grad_norm": 1.8639196756054945,
      "language_loss": 0.83415568,
      "learning_rate": 1.4541725626706485e-06,
      "loss": 0.8554585,
      "num_input_tokens_seen": 215077790,
      "step": 9985,
      "time_per_iteration": 2.6163461208343506
    },
    {
      "auxiliary_loss_clip": 0.01099048,
      "auxiliary_loss_mlp": 0.01033067,
      "balance_loss_clip": 1.03778839,
      "balance_loss_mlp": 1.02109194,
      "epoch": 0.6003908011423418,
      "flos": 26687733252480.0,
      "grad_norm": 1.9832185370832152,
      "language_loss": 0.70600045,
      "learning_rate": 1.4537978958846809e-06,
      "loss": 0.72732162,
      "num_input_tokens_seen": 215097650,
      "step": 9986,
      "time_per_iteration": 2.5708351135253906
    },
    {
      "auxiliary_loss_clip": 0.01113712,
      "auxiliary_loss_mlp": 0.00787891,
      "balance_loss_clip": 1.04021382,
      "balance_loss_mlp": 1.01209807,
      "epoch": 0.6004509243950098,
      "flos": 22565080736640.0,
      "grad_norm": 1.5663550536344832,
      "language_loss": 0.71615899,
      "learning_rate": 1.4534232498107514e-06,
      "loss": 0.73517501,
      "num_input_tokens_seen": 215118235,
      "step": 9987,
      "time_per_iteration": 2.5417897701263428
    },
    {
      "auxiliary_loss_clip": 0.01087486,
      "auxiliary_loss_mlp": 0.01030792,
      "balance_loss_clip": 1.03754115,
      "balance_loss_mlp": 1.01894212,
      "epoch": 0.6005110476476777,
      "flos": 19719303868800.0,
      "grad_norm": 1.7036796877513132,
      "language_loss": 0.84506053,
      "learning_rate": 1.4530486244630673e-06,
      "loss": 0.86624324,
      "num_input_tokens_seen": 215136755,
      "step": 9988,
      "time_per_iteration": 2.5509607791900635
    },
    {
      "auxiliary_loss_clip": 0.0110018,
      "auxiliary_loss_mlp": 0.01036326,
      "balance_loss_clip": 1.03813744,
      "balance_loss_mlp": 1.02334452,
      "epoch": 0.6005711709003457,
      "flos": 17712543859200.0,
      "grad_norm": 8.228756840956002,
      "language_loss": 0.65192425,
      "learning_rate": 1.4526740198558346e-06,
      "loss": 0.67328936,
      "num_input_tokens_seen": 215155225,
      "step": 9989,
      "time_per_iteration": 2.4805426597595215
    },
    {
      "auxiliary_loss_clip": 0.01096014,
      "auxiliary_loss_mlp": 0.01035707,
      "balance_loss_clip": 1.03547645,
      "balance_loss_mlp": 1.02437568,
      "epoch": 0.6006312941530136,
      "flos": 18514464946560.0,
      "grad_norm": 1.6056280009443284,
      "language_loss": 0.80855834,
      "learning_rate": 1.452299436003257e-06,
      "loss": 0.82987553,
      "num_input_tokens_seen": 215174815,
      "step": 9990,
      "time_per_iteration": 2.531773090362549
    },
    {
      "auxiliary_loss_clip": 0.01075435,
      "auxiliary_loss_mlp": 0.0103509,
      "balance_loss_clip": 1.04266191,
      "balance_loss_mlp": 1.02238786,
      "epoch": 0.6006914174056817,
      "flos": 21390837223680.0,
      "grad_norm": 1.7997512067192623,
      "language_loss": 0.83117062,
      "learning_rate": 1.4519248729195403e-06,
      "loss": 0.85227585,
      "num_input_tokens_seen": 215192045,
      "step": 9991,
      "time_per_iteration": 2.5668325424194336
    },
    {
      "auxiliary_loss_clip": 0.01059375,
      "auxiliary_loss_mlp": 0.01031658,
      "balance_loss_clip": 1.03750944,
      "balance_loss_mlp": 1.01810443,
      "epoch": 0.6007515406583496,
      "flos": 12750515349120.0,
      "grad_norm": 1.934758158684221,
      "language_loss": 0.82606888,
      "learning_rate": 1.4515503306188878e-06,
      "loss": 0.84697926,
      "num_input_tokens_seen": 215209885,
      "step": 9992,
      "time_per_iteration": 2.6903655529022217
    },
    {
      "auxiliary_loss_clip": 0.01086856,
      "auxiliary_loss_mlp": 0.00787445,
      "balance_loss_clip": 1.03711379,
      "balance_loss_mlp": 1.01446509,
      "epoch": 0.6008116639110176,
      "flos": 19206894401280.0,
      "grad_norm": 1.9120570550179155,
      "language_loss": 0.66055644,
      "learning_rate": 1.4511758091155008e-06,
      "loss": 0.67929947,
      "num_input_tokens_seen": 215228150,
      "step": 9993,
      "time_per_iteration": 2.741182327270508
    },
    {
      "auxiliary_loss_clip": 0.01073553,
      "auxiliary_loss_mlp": 0.01028204,
      "balance_loss_clip": 1.03909647,
      "balance_loss_mlp": 1.01515102,
      "epoch": 0.6008717871636855,
      "flos": 17055342668160.0,
      "grad_norm": 2.8458277610948555,
      "language_loss": 0.80971175,
      "learning_rate": 1.4508013084235826e-06,
      "loss": 0.83072937,
      "num_input_tokens_seen": 215243755,
      "step": 9994,
      "time_per_iteration": 2.6786487102508545
    },
    {
      "auxiliary_loss_clip": 0.01066765,
      "auxiliary_loss_mlp": 0.01027861,
      "balance_loss_clip": 1.03585184,
      "balance_loss_mlp": 1.01611304,
      "epoch": 0.6009319104163535,
      "flos": 20298686244480.0,
      "grad_norm": 1.8170581974707647,
      "language_loss": 0.72317708,
      "learning_rate": 1.4504268285573337e-06,
      "loss": 0.74412334,
      "num_input_tokens_seen": 215262130,
      "step": 9995,
      "time_per_iteration": 3.984469175338745
    },
    {
      "auxiliary_loss_clip": 0.01089105,
      "auxiliary_loss_mlp": 0.0103242,
      "balance_loss_clip": 1.03649473,
      "balance_loss_mlp": 1.01970673,
      "epoch": 0.6009920336690215,
      "flos": 21836776573440.0,
      "grad_norm": 1.7409343669513493,
      "language_loss": 0.81288779,
      "learning_rate": 1.4500523695309546e-06,
      "loss": 0.83410305,
      "num_input_tokens_seen": 215281785,
      "step": 9996,
      "time_per_iteration": 2.6140377521514893
    },
    {
      "auxiliary_loss_clip": 0.01049171,
      "auxiliary_loss_mlp": 0.01043779,
      "balance_loss_clip": 1.033584,
      "balance_loss_mlp": 1.02959275,
      "epoch": 0.6010521569216895,
      "flos": 22596107109120.0,
      "grad_norm": 10.34631279393732,
      "language_loss": 0.7867161,
      "learning_rate": 1.4496779313586447e-06,
      "loss": 0.80764556,
      "num_input_tokens_seen": 215297550,
      "step": 9997,
      "time_per_iteration": 2.8118796348571777
    },
    {
      "auxiliary_loss_clip": 0.01103018,
      "auxiliary_loss_mlp": 0.01028977,
      "balance_loss_clip": 1.03900909,
      "balance_loss_mlp": 1.01580465,
      "epoch": 0.6011122801743575,
      "flos": 19171702051200.0,
      "grad_norm": 1.642982254397811,
      "language_loss": 0.72720617,
      "learning_rate": 1.4493035140546028e-06,
      "loss": 0.7485261,
      "num_input_tokens_seen": 215316360,
      "step": 9998,
      "time_per_iteration": 2.697793960571289
    },
    {
      "auxiliary_loss_clip": 0.01080905,
      "auxiliary_loss_mlp": 0.01027578,
      "balance_loss_clip": 1.04057693,
      "balance_loss_mlp": 1.01567507,
      "epoch": 0.6011724034270254,
      "flos": 25010022758400.0,
      "grad_norm": 1.486840100577563,
      "language_loss": 0.72455883,
      "learning_rate": 1.448929117633027e-06,
      "loss": 0.74564362,
      "num_input_tokens_seen": 215336405,
      "step": 9999,
      "time_per_iteration": 2.7487761974334717
    },
    {
      "auxiliary_loss_clip": 0.01065872,
      "auxiliary_loss_mlp": 0.01034368,
      "balance_loss_clip": 1.03843439,
      "balance_loss_mlp": 1.02114201,
      "epoch": 0.6012325266796934,
      "flos": 21797669640960.0,
      "grad_norm": 1.4672808716871348,
      "language_loss": 0.78525561,
      "learning_rate": 1.4485547421081142e-06,
      "loss": 0.80625802,
      "num_input_tokens_seen": 215356590,
      "step": 10000,
      "time_per_iteration": 2.6902832984924316
    },
    {
      "auxiliary_loss_clip": 0.01116198,
      "auxiliary_loss_mlp": 0.01031271,
      "balance_loss_clip": 1.0403235,
      "balance_loss_mlp": 1.0178721,
      "epoch": 0.6012926499323613,
      "flos": 19573003774080.0,
      "grad_norm": 2.5382312407458762,
      "language_loss": 0.77747631,
      "learning_rate": 1.4481803874940608e-06,
      "loss": 0.79895103,
      "num_input_tokens_seen": 215374295,
      "step": 10001,
      "time_per_iteration": 2.506680965423584
    },
    {
      "auxiliary_loss_clip": 0.0110164,
      "auxiliary_loss_mlp": 0.0102915,
      "balance_loss_clip": 1.03822732,
      "balance_loss_mlp": 1.01619208,
      "epoch": 0.6013527731850293,
      "flos": 34860786076800.0,
      "grad_norm": 1.8184031201503985,
      "language_loss": 0.58914065,
      "learning_rate": 1.4478060538050624e-06,
      "loss": 0.61044854,
      "num_input_tokens_seen": 215394535,
      "step": 10002,
      "time_per_iteration": 2.691248655319214
    },
    {
      "auxiliary_loss_clip": 0.0109344,
      "auxiliary_loss_mlp": 0.01034637,
      "balance_loss_clip": 1.03983879,
      "balance_loss_mlp": 1.02042079,
      "epoch": 0.6014128964376972,
      "flos": 23291948355840.0,
      "grad_norm": 1.5107903509049625,
      "language_loss": 0.77614057,
      "learning_rate": 1.447431741055314e-06,
      "loss": 0.79742134,
      "num_input_tokens_seen": 215414355,
      "step": 10003,
      "time_per_iteration": 2.5917367935180664
    },
    {
      "auxiliary_loss_clip": 0.0111465,
      "auxiliary_loss_mlp": 0.01031541,
      "balance_loss_clip": 1.0406009,
      "balance_loss_mlp": 1.01877987,
      "epoch": 0.6014730196903653,
      "flos": 24820916630400.0,
      "grad_norm": 8.886514689833119,
      "language_loss": 0.77389681,
      "learning_rate": 1.4470574492590091e-06,
      "loss": 0.79535878,
      "num_input_tokens_seen": 215428280,
      "step": 10004,
      "time_per_iteration": 3.924069881439209
    },
    {
      "auxiliary_loss_clip": 0.01098082,
      "auxiliary_loss_mlp": 0.01029029,
      "balance_loss_clip": 1.03743792,
      "balance_loss_mlp": 1.01664352,
      "epoch": 0.6015331429430332,
      "flos": 23112359331840.0,
      "grad_norm": 1.4582750139819272,
      "language_loss": 0.72383267,
      "learning_rate": 1.4466831784303408e-06,
      "loss": 0.74510384,
      "num_input_tokens_seen": 215448970,
      "step": 10005,
      "time_per_iteration": 2.5578627586364746
    },
    {
      "auxiliary_loss_clip": 0.01109255,
      "auxiliary_loss_mlp": 0.01027058,
      "balance_loss_clip": 1.03986657,
      "balance_loss_mlp": 1.0146904,
      "epoch": 0.6015932661957012,
      "flos": 19201363706880.0,
      "grad_norm": 2.1097385541079317,
      "language_loss": 0.74524677,
      "learning_rate": 1.4463089285835026e-06,
      "loss": 0.76660991,
      "num_input_tokens_seen": 215465260,
      "step": 10006,
      "time_per_iteration": 3.9293792247772217
    },
    {
      "auxiliary_loss_clip": 0.0108655,
      "auxiliary_loss_mlp": 0.01031022,
      "balance_loss_clip": 1.04013157,
      "balance_loss_mlp": 1.01849294,
      "epoch": 0.6016533894483691,
      "flos": 18113630100480.0,
      "grad_norm": 1.8268246506023238,
      "language_loss": 0.73822033,
      "learning_rate": 1.445934699732685e-06,
      "loss": 0.75939608,
      "num_input_tokens_seen": 215482725,
      "step": 10007,
      "time_per_iteration": 2.5542588233947754
    },
    {
      "auxiliary_loss_clip": 0.01087849,
      "auxiliary_loss_mlp": 0.01025782,
      "balance_loss_clip": 1.03711653,
      "balance_loss_mlp": 1.01421225,
      "epoch": 0.6017135127010371,
      "flos": 16216900427520.0,
      "grad_norm": 1.78721342428747,
      "language_loss": 0.70173252,
      "learning_rate": 1.4455604918920785e-06,
      "loss": 0.7228688,
      "num_input_tokens_seen": 215500420,
      "step": 10008,
      "time_per_iteration": 2.530632257461548
    },
    {
      "auxiliary_loss_clip": 0.0110045,
      "auxiliary_loss_mlp": 0.0102661,
      "balance_loss_clip": 1.03837061,
      "balance_loss_mlp": 1.01474833,
      "epoch": 0.6017736359537051,
      "flos": 23444246021760.0,
      "grad_norm": 1.7146552045949488,
      "language_loss": 0.76314962,
      "learning_rate": 1.4451863050758748e-06,
      "loss": 0.78442025,
      "num_input_tokens_seen": 215522260,
      "step": 10009,
      "time_per_iteration": 3.926481246948242
    },
    {
      "auxiliary_loss_clip": 0.01085722,
      "auxiliary_loss_mlp": 0.00789428,
      "balance_loss_clip": 1.03799129,
      "balance_loss_mlp": 1.0190345,
      "epoch": 0.601833759206373,
      "flos": 23514056104320.0,
      "grad_norm": 2.84948881476015,
      "language_loss": 0.74167722,
      "learning_rate": 1.4448121392982608e-06,
      "loss": 0.76042873,
      "num_input_tokens_seen": 215541715,
      "step": 10010,
      "time_per_iteration": 2.5889763832092285
    },
    {
      "auxiliary_loss_clip": 0.01035235,
      "auxiliary_loss_mlp": 0.01004011,
      "balance_loss_clip": 1.01998329,
      "balance_loss_mlp": 1.00284863,
      "epoch": 0.6018938824590411,
      "flos": 63991668648960.0,
      "grad_norm": 0.8025599094925673,
      "language_loss": 0.55105728,
      "learning_rate": 1.4444379945734268e-06,
      "loss": 0.57144976,
      "num_input_tokens_seen": 215603020,
      "step": 10011,
      "time_per_iteration": 3.181299924850464
    },
    {
      "auxiliary_loss_clip": 0.01102279,
      "auxiliary_loss_mlp": 0.01036126,
      "balance_loss_clip": 1.03917122,
      "balance_loss_mlp": 1.02390707,
      "epoch": 0.601954005711709,
      "flos": 34640007131520.0,
      "grad_norm": 1.3793109772808514,
      "language_loss": 0.61939204,
      "learning_rate": 1.44406387091556e-06,
      "loss": 0.64077616,
      "num_input_tokens_seen": 215625115,
      "step": 10012,
      "time_per_iteration": 2.6178345680236816
    },
    {
      "auxiliary_loss_clip": 0.01071802,
      "auxiliary_loss_mlp": 0.01026072,
      "balance_loss_clip": 1.03939795,
      "balance_loss_mlp": 1.01448512,
      "epoch": 0.602014128964377,
      "flos": 19427062815360.0,
      "grad_norm": 1.6741393497811643,
      "language_loss": 0.75156295,
      "learning_rate": 1.4436897683388462e-06,
      "loss": 0.77254164,
      "num_input_tokens_seen": 215643730,
      "step": 10013,
      "time_per_iteration": 2.5505754947662354
    },
    {
      "auxiliary_loss_clip": 0.0110563,
      "auxiliary_loss_mlp": 0.01030736,
      "balance_loss_clip": 1.03795075,
      "balance_loss_mlp": 1.01970887,
      "epoch": 0.6020742522170449,
      "flos": 28329389470080.0,
      "grad_norm": 1.6220764217896915,
      "language_loss": 0.81536901,
      "learning_rate": 1.4433156868574732e-06,
      "loss": 0.83673269,
      "num_input_tokens_seen": 215664425,
      "step": 10014,
      "time_per_iteration": 2.5219874382019043
    },
    {
      "auxiliary_loss_clip": 0.01083955,
      "auxiliary_loss_mlp": 0.0102623,
      "balance_loss_clip": 1.03634024,
      "balance_loss_mlp": 1.01432085,
      "epoch": 0.6021343754697129,
      "flos": 22747040058240.0,
      "grad_norm": 1.3948756444303017,
      "language_loss": 0.72337914,
      "learning_rate": 1.442941626485624e-06,
      "loss": 0.74448103,
      "num_input_tokens_seen": 215684280,
      "step": 10015,
      "time_per_iteration": 2.5843329429626465
    },
    {
      "auxiliary_loss_clip": 0.01020776,
      "auxiliary_loss_mlp": 0.01007277,
      "balance_loss_clip": 1.01716185,
      "balance_loss_mlp": 1.00602531,
      "epoch": 0.6021944987223808,
      "flos": 65752007402880.0,
      "grad_norm": 0.8416662579324772,
      "language_loss": 0.5483712,
      "learning_rate": 1.4425675872374848e-06,
      "loss": 0.56865168,
      "num_input_tokens_seen": 215739780,
      "step": 10016,
      "time_per_iteration": 3.023139476776123
    },
    {
      "auxiliary_loss_clip": 0.01086657,
      "auxiliary_loss_mlp": 0.01030684,
      "balance_loss_clip": 1.03799367,
      "balance_loss_mlp": 1.01856041,
      "epoch": 0.6022546219750489,
      "flos": 16105182151680.0,
      "grad_norm": 1.6989356408620242,
      "language_loss": 0.82890427,
      "learning_rate": 1.4421935691272381e-06,
      "loss": 0.85007763,
      "num_input_tokens_seen": 215757885,
      "step": 10017,
      "time_per_iteration": 2.5541739463806152
    },
    {
      "auxiliary_loss_clip": 0.01088339,
      "auxiliary_loss_mlp": 0.01031514,
      "balance_loss_clip": 1.04317355,
      "balance_loss_mlp": 1.01959944,
      "epoch": 0.6023147452277168,
      "flos": 25512555985920.0,
      "grad_norm": 2.5493988119164457,
      "language_loss": 0.83910334,
      "learning_rate": 1.4418195721690677e-06,
      "loss": 0.86030191,
      "num_input_tokens_seen": 215776415,
      "step": 10018,
      "time_per_iteration": 2.5687685012817383
    },
    {
      "auxiliary_loss_clip": 0.01087911,
      "auxiliary_loss_mlp": 0.01034725,
      "balance_loss_clip": 1.03670549,
      "balance_loss_mlp": 1.02173078,
      "epoch": 0.6023748684803848,
      "flos": 22636075968000.0,
      "grad_norm": 2.035362453600421,
      "language_loss": 0.7856679,
      "learning_rate": 1.4414455963771549e-06,
      "loss": 0.80689418,
      "num_input_tokens_seen": 215794865,
      "step": 10019,
      "time_per_iteration": 2.5579895973205566
    },
    {
      "auxiliary_loss_clip": 0.01072989,
      "auxiliary_loss_mlp": 0.0078647,
      "balance_loss_clip": 1.04076838,
      "balance_loss_mlp": 1.01004028,
      "epoch": 0.6024349917330527,
      "flos": 26210444307840.0,
      "grad_norm": 1.4361806427390396,
      "language_loss": 0.73806286,
      "learning_rate": 1.441071641765681e-06,
      "loss": 0.75665748,
      "num_input_tokens_seen": 215816840,
      "step": 10020,
      "time_per_iteration": 2.642089366912842
    },
    {
      "auxiliary_loss_clip": 0.01090328,
      "auxiliary_loss_mlp": 0.01032133,
      "balance_loss_clip": 1.0381645,
      "balance_loss_mlp": 1.01978922,
      "epoch": 0.6024951149857207,
      "flos": 21251755762560.0,
      "grad_norm": 1.477871585532041,
      "language_loss": 0.64007169,
      "learning_rate": 1.4406977083488264e-06,
      "loss": 0.66129625,
      "num_input_tokens_seen": 215836100,
      "step": 10021,
      "time_per_iteration": 2.5580034255981445
    },
    {
      "auxiliary_loss_clip": 0.01098366,
      "auxiliary_loss_mlp": 0.01032265,
      "balance_loss_clip": 1.03771758,
      "balance_loss_mlp": 1.01953924,
      "epoch": 0.6025552382383887,
      "flos": 26943453152640.0,
      "grad_norm": 1.387329587016392,
      "language_loss": 0.80269909,
      "learning_rate": 1.4403237961407704e-06,
      "loss": 0.82400542,
      "num_input_tokens_seen": 215858480,
      "step": 10022,
      "time_per_iteration": 2.5585343837738037
    },
    {
      "auxiliary_loss_clip": 0.01106971,
      "auxiliary_loss_mlp": 0.0103117,
      "balance_loss_clip": 1.04113483,
      "balance_loss_mlp": 1.01903415,
      "epoch": 0.6026153614910567,
      "flos": 31684379495040.0,
      "grad_norm": 3.8272796718036983,
      "language_loss": 0.66632277,
      "learning_rate": 1.439949905155693e-06,
      "loss": 0.68770415,
      "num_input_tokens_seen": 215879950,
      "step": 10023,
      "time_per_iteration": 2.5940968990325928
    },
    {
      "auxiliary_loss_clip": 0.01100837,
      "auxiliary_loss_mlp": 0.01032589,
      "balance_loss_clip": 1.0367192,
      "balance_loss_mlp": 1.01994133,
      "epoch": 0.6026754847437247,
      "flos": 29312731175040.0,
      "grad_norm": 1.8587340628416016,
      "language_loss": 0.73924708,
      "learning_rate": 1.4395760354077707e-06,
      "loss": 0.76058137,
      "num_input_tokens_seen": 215899830,
      "step": 10024,
      "time_per_iteration": 2.574982166290283
    },
    {
      "auxiliary_loss_clip": 0.01100314,
      "auxiliary_loss_mlp": 0.01031888,
      "balance_loss_clip": 1.04014325,
      "balance_loss_mlp": 1.01930511,
      "epoch": 0.6027356079963926,
      "flos": 23586775188480.0,
      "grad_norm": 1.780291278895535,
      "language_loss": 0.72702587,
      "learning_rate": 1.4392021869111815e-06,
      "loss": 0.74834788,
      "num_input_tokens_seen": 215920440,
      "step": 10025,
      "time_per_iteration": 2.5485243797302246
    },
    {
      "auxiliary_loss_clip": 0.01115867,
      "auxiliary_loss_mlp": 0.01035967,
      "balance_loss_clip": 1.03976631,
      "balance_loss_mlp": 1.02255642,
      "epoch": 0.6027957312490606,
      "flos": 20813753318400.0,
      "grad_norm": 2.278380399845514,
      "language_loss": 0.67450213,
      "learning_rate": 1.4388283596801016e-06,
      "loss": 0.69602048,
      "num_input_tokens_seen": 215940535,
      "step": 10026,
      "time_per_iteration": 2.522057294845581
    },
    {
      "auxiliary_loss_clip": 0.01104737,
      "auxiliary_loss_mlp": 0.01030633,
      "balance_loss_clip": 1.03670835,
      "balance_loss_mlp": 1.01895642,
      "epoch": 0.6028558545017285,
      "flos": 19935773182080.0,
      "grad_norm": 1.8826848187594503,
      "language_loss": 0.80337417,
      "learning_rate": 1.4384545537287061e-06,
      "loss": 0.82472783,
      "num_input_tokens_seen": 215958045,
      "step": 10027,
      "time_per_iteration": 2.491328477859497
    },
    {
      "auxiliary_loss_clip": 0.01082059,
      "auxiliary_loss_mlp": 0.01029844,
      "balance_loss_clip": 1.04274571,
      "balance_loss_mlp": 1.01727355,
      "epoch": 0.6029159777543965,
      "flos": 22820836550400.0,
      "grad_norm": 1.899605678064577,
      "language_loss": 0.71217263,
      "learning_rate": 1.438080769071171e-06,
      "loss": 0.73329163,
      "num_input_tokens_seen": 215977330,
      "step": 10028,
      "time_per_iteration": 2.5809173583984375
    },
    {
      "auxiliary_loss_clip": 0.0107371,
      "auxiliary_loss_mlp": 0.01035639,
      "balance_loss_clip": 1.0388236,
      "balance_loss_mlp": 1.02302027,
      "epoch": 0.6029761010070644,
      "flos": 23587242065280.0,
      "grad_norm": 1.7290199032334437,
      "language_loss": 0.84897351,
      "learning_rate": 1.437707005721669e-06,
      "loss": 0.870067,
      "num_input_tokens_seen": 215997865,
      "step": 10029,
      "time_per_iteration": 2.6202635765075684
    },
    {
      "auxiliary_loss_clip": 0.01088868,
      "auxiliary_loss_mlp": 0.01029704,
      "balance_loss_clip": 1.03800082,
      "balance_loss_mlp": 1.01800966,
      "epoch": 0.6030362242597325,
      "flos": 13662430859520.0,
      "grad_norm": 1.7696920129827098,
      "language_loss": 0.79993951,
      "learning_rate": 1.437333263694373e-06,
      "loss": 0.82112527,
      "num_input_tokens_seen": 216016230,
      "step": 10030,
      "time_per_iteration": 2.513556718826294
    },
    {
      "auxiliary_loss_clip": 0.01052144,
      "auxiliary_loss_mlp": 0.01037801,
      "balance_loss_clip": 1.03636169,
      "balance_loss_mlp": 1.02405632,
      "epoch": 0.6030963475124004,
      "flos": 24422883045120.0,
      "grad_norm": 1.9813475529165967,
      "language_loss": 0.71288133,
      "learning_rate": 1.4369595430034572e-06,
      "loss": 0.73378074,
      "num_input_tokens_seen": 216035785,
      "step": 10031,
      "time_per_iteration": 2.669219970703125
    },
    {
      "auxiliary_loss_clip": 0.01069588,
      "auxiliary_loss_mlp": 0.01033822,
      "balance_loss_clip": 1.03760839,
      "balance_loss_mlp": 1.02010083,
      "epoch": 0.6031564707650684,
      "flos": 29644043247360.0,
      "grad_norm": 1.5132774954035395,
      "language_loss": 0.73141527,
      "learning_rate": 1.4365858436630912e-06,
      "loss": 0.75244939,
      "num_input_tokens_seen": 216059555,
      "step": 10032,
      "time_per_iteration": 2.6674206256866455
    },
    {
      "auxiliary_loss_clip": 0.01093796,
      "auxiliary_loss_mlp": 0.01031666,
      "balance_loss_clip": 1.04071283,
      "balance_loss_mlp": 1.01852334,
      "epoch": 0.6032165940177363,
      "flos": 16618776768000.0,
      "grad_norm": 2.0525892666054197,
      "language_loss": 0.68470132,
      "learning_rate": 1.4362121656874465e-06,
      "loss": 0.70595598,
      "num_input_tokens_seen": 216077235,
      "step": 10033,
      "time_per_iteration": 2.539726972579956
    },
    {
      "auxiliary_loss_clip": 0.01085261,
      "auxiliary_loss_mlp": 0.01032135,
      "balance_loss_clip": 1.03887606,
      "balance_loss_mlp": 1.01887321,
      "epoch": 0.6032767172704043,
      "flos": 17488173553920.0,
      "grad_norm": 2.134185519419214,
      "language_loss": 0.75730848,
      "learning_rate": 1.4358385090906934e-06,
      "loss": 0.77848244,
      "num_input_tokens_seen": 216094985,
      "step": 10034,
      "time_per_iteration": 3.9015564918518066
    },
    {
      "auxiliary_loss_clip": 0.01088412,
      "auxiliary_loss_mlp": 0.01031156,
      "balance_loss_clip": 1.0383203,
      "balance_loss_mlp": 1.01845431,
      "epoch": 0.6033368405230723,
      "flos": 26832955939200.0,
      "grad_norm": 2.489865183325225,
      "language_loss": 0.74558115,
      "learning_rate": 1.4354648738870004e-06,
      "loss": 0.7667768,
      "num_input_tokens_seen": 216115905,
      "step": 10035,
      "time_per_iteration": 2.6159863471984863
    },
    {
      "auxiliary_loss_clip": 0.01082399,
      "auxiliary_loss_mlp": 0.01024622,
      "balance_loss_clip": 1.03924513,
      "balance_loss_mlp": 1.01292145,
      "epoch": 0.6033969637757403,
      "flos": 16909904499840.0,
      "grad_norm": 1.8108608768309096,
      "language_loss": 0.86822319,
      "learning_rate": 1.435091260090536e-06,
      "loss": 0.88929343,
      "num_input_tokens_seen": 216132420,
      "step": 10036,
      "time_per_iteration": 2.5128564834594727
    },
    {
      "auxiliary_loss_clip": 0.01074886,
      "auxiliary_loss_mlp": 0.01029626,
      "balance_loss_clip": 1.0412364,
      "balance_loss_mlp": 1.01705503,
      "epoch": 0.6034570870284083,
      "flos": 22930076787840.0,
      "grad_norm": 1.846408447273741,
      "language_loss": 0.70473069,
      "learning_rate": 1.4347176677154676e-06,
      "loss": 0.72577584,
      "num_input_tokens_seen": 216149800,
      "step": 10037,
      "time_per_iteration": 2.5700318813323975
    },
    {
      "auxiliary_loss_clip": 0.01096522,
      "auxiliary_loss_mlp": 0.01033579,
      "balance_loss_clip": 1.03896713,
      "balance_loss_mlp": 1.01970911,
      "epoch": 0.6035172102810762,
      "flos": 23366319465600.0,
      "grad_norm": 1.6500345768370672,
      "language_loss": 0.84729314,
      "learning_rate": 1.4343440967759616e-06,
      "loss": 0.86859417,
      "num_input_tokens_seen": 216168200,
      "step": 10038,
      "time_per_iteration": 2.5120959281921387
    },
    {
      "auxiliary_loss_clip": 0.01093517,
      "auxiliary_loss_mlp": 0.01031391,
      "balance_loss_clip": 1.0379976,
      "balance_loss_mlp": 1.01883864,
      "epoch": 0.6035773335337442,
      "flos": 20887082933760.0,
      "grad_norm": 1.9736761768548143,
      "language_loss": 0.7612592,
      "learning_rate": 1.4339705472861846e-06,
      "loss": 0.78250819,
      "num_input_tokens_seen": 216187105,
      "step": 10039,
      "time_per_iteration": 2.552495002746582
    },
    {
      "auxiliary_loss_clip": 0.01099093,
      "auxiliary_loss_mlp": 0.01029786,
      "balance_loss_clip": 1.03828049,
      "balance_loss_mlp": 1.01819897,
      "epoch": 0.6036374567864121,
      "flos": 24936298093440.0,
      "grad_norm": 2.2320380375193873,
      "language_loss": 0.71332216,
      "learning_rate": 1.433597019260301e-06,
      "loss": 0.73461097,
      "num_input_tokens_seen": 216205440,
      "step": 10040,
      "time_per_iteration": 2.5314314365386963
    },
    {
      "auxiliary_loss_clip": 0.01104374,
      "auxiliary_loss_mlp": 0.01030455,
      "balance_loss_clip": 1.04112959,
      "balance_loss_mlp": 1.01626277,
      "epoch": 0.6036975800390801,
      "flos": 23148269953920.0,
      "grad_norm": 2.1767315235121205,
      "language_loss": 0.78115016,
      "learning_rate": 1.433223512712475e-06,
      "loss": 0.80249846,
      "num_input_tokens_seen": 216223130,
      "step": 10041,
      "time_per_iteration": 2.515191078186035
    },
    {
      "auxiliary_loss_clip": 0.01091591,
      "auxiliary_loss_mlp": 0.01030151,
      "balance_loss_clip": 1.04043984,
      "balance_loss_mlp": 1.01795566,
      "epoch": 0.603757703291748,
      "flos": 18660729127680.0,
      "grad_norm": 1.660766860055552,
      "language_loss": 0.76066607,
      "learning_rate": 1.4328500276568704e-06,
      "loss": 0.78188354,
      "num_input_tokens_seen": 216240260,
      "step": 10042,
      "time_per_iteration": 2.5008740425109863
    },
    {
      "auxiliary_loss_clip": 0.0106493,
      "auxiliary_loss_mlp": 0.01024528,
      "balance_loss_clip": 1.03642654,
      "balance_loss_mlp": 1.01273203,
      "epoch": 0.6038178265444161,
      "flos": 19682603147520.0,
      "grad_norm": 2.1005259480863203,
      "language_loss": 0.85002667,
      "learning_rate": 1.4324765641076498e-06,
      "loss": 0.87092125,
      "num_input_tokens_seen": 216258510,
      "step": 10043,
      "time_per_iteration": 3.965954065322876
    },
    {
      "auxiliary_loss_clip": 0.01075978,
      "auxiliary_loss_mlp": 0.01037561,
      "balance_loss_clip": 1.03671026,
      "balance_loss_mlp": 1.02416217,
      "epoch": 0.603877949797084,
      "flos": 22638230784000.0,
      "grad_norm": 1.944853373709849,
      "language_loss": 0.69597578,
      "learning_rate": 1.432103122078974e-06,
      "loss": 0.71711123,
      "num_input_tokens_seen": 216277550,
      "step": 10044,
      "time_per_iteration": 3.9723825454711914
    },
    {
      "auxiliary_loss_clip": 0.01103901,
      "auxiliary_loss_mlp": 0.01027208,
      "balance_loss_clip": 1.04527903,
      "balance_loss_mlp": 1.01386905,
      "epoch": 0.603938073049752,
      "flos": 25447881548160.0,
      "grad_norm": 2.42866345554089,
      "language_loss": 0.78087759,
      "learning_rate": 1.4317297015850057e-06,
      "loss": 0.80218863,
      "num_input_tokens_seen": 216296690,
      "step": 10045,
      "time_per_iteration": 2.5499019622802734
    },
    {
      "auxiliary_loss_clip": 0.01061632,
      "auxiliary_loss_mlp": 0.01031419,
      "balance_loss_clip": 1.03772235,
      "balance_loss_mlp": 1.01880646,
      "epoch": 0.6039981963024199,
      "flos": 22340135813760.0,
      "grad_norm": 1.7276766180751577,
      "language_loss": 0.77439433,
      "learning_rate": 1.4313563026399036e-06,
      "loss": 0.7953248,
      "num_input_tokens_seen": 216316110,
      "step": 10046,
      "time_per_iteration": 2.6359434127807617
    },
    {
      "auxiliary_loss_clip": 0.01057289,
      "auxiliary_loss_mlp": 0.01036129,
      "balance_loss_clip": 1.03355813,
      "balance_loss_mlp": 1.02301645,
      "epoch": 0.6040583195550879,
      "flos": 20703148364160.0,
      "grad_norm": 1.530203271510399,
      "language_loss": 0.86968774,
      "learning_rate": 1.430982925257827e-06,
      "loss": 0.89062196,
      "num_input_tokens_seen": 216333855,
      "step": 10047,
      "time_per_iteration": 2.581923007965088
    },
    {
      "auxiliary_loss_clip": 0.01103544,
      "auxiliary_loss_mlp": 0.0102462,
      "balance_loss_clip": 1.04317439,
      "balance_loss_mlp": 1.01305699,
      "epoch": 0.604118442807756,
      "flos": 27163118776320.0,
      "grad_norm": 1.5130385978010459,
      "language_loss": 0.75436294,
      "learning_rate": 1.4306095694529358e-06,
      "loss": 0.7756446,
      "num_input_tokens_seen": 216354890,
      "step": 10048,
      "time_per_iteration": 3.9316518306732178
    },
    {
      "auxiliary_loss_clip": 0.01110704,
      "auxiliary_loss_mlp": 0.01035817,
      "balance_loss_clip": 1.04134667,
      "balance_loss_mlp": 1.02115417,
      "epoch": 0.6041785660604239,
      "flos": 30881524654080.0,
      "grad_norm": 5.647231011256132,
      "language_loss": 0.66300952,
      "learning_rate": 1.430236235239386e-06,
      "loss": 0.68447471,
      "num_input_tokens_seen": 216376055,
      "step": 10049,
      "time_per_iteration": 2.5822269916534424
    },
    {
      "auxiliary_loss_clip": 0.01085332,
      "auxiliary_loss_mlp": 0.0103781,
      "balance_loss_clip": 1.0371691,
      "balance_loss_mlp": 1.02421403,
      "epoch": 0.6042386893130919,
      "flos": 19938215306880.0,
      "grad_norm": 1.4933750124994163,
      "language_loss": 0.66821849,
      "learning_rate": 1.429862922631336e-06,
      "loss": 0.68944991,
      "num_input_tokens_seen": 216396295,
      "step": 10050,
      "time_per_iteration": 2.521122455596924
    },
    {
      "auxiliary_loss_clip": 0.01079505,
      "auxiliary_loss_mlp": 0.01030597,
      "balance_loss_clip": 1.03853583,
      "balance_loss_mlp": 1.01819277,
      "epoch": 0.6042988125657598,
      "flos": 32415915882240.0,
      "grad_norm": 1.8554980457976378,
      "language_loss": 0.69577515,
      "learning_rate": 1.4294896316429408e-06,
      "loss": 0.71687621,
      "num_input_tokens_seen": 216416605,
      "step": 10051,
      "time_per_iteration": 2.6521992683410645
    },
    {
      "auxiliary_loss_clip": 0.01097246,
      "auxiliary_loss_mlp": 0.01031722,
      "balance_loss_clip": 1.03660965,
      "balance_loss_mlp": 1.01974773,
      "epoch": 0.6043589358184278,
      "flos": 17420805596160.0,
      "grad_norm": 2.3333945917612886,
      "language_loss": 0.6423952,
      "learning_rate": 1.4291163622883553e-06,
      "loss": 0.6636849,
      "num_input_tokens_seen": 216435130,
      "step": 10052,
      "time_per_iteration": 2.465782403945923
    },
    {
      "auxiliary_loss_clip": 0.01088852,
      "auxiliary_loss_mlp": 0.01028161,
      "balance_loss_clip": 1.03805947,
      "balance_loss_mlp": 1.01535237,
      "epoch": 0.6044190590710957,
      "flos": 27672834723840.0,
      "grad_norm": 1.9917833193835313,
      "language_loss": 0.68856055,
      "learning_rate": 1.4287431145817358e-06,
      "loss": 0.70973068,
      "num_input_tokens_seen": 216455640,
      "step": 10053,
      "time_per_iteration": 2.5840976238250732
    },
    {
      "auxiliary_loss_clip": 0.01028069,
      "auxiliary_loss_mlp": 0.0100337,
      "balance_loss_clip": 1.02368617,
      "balance_loss_mlp": 1.00227344,
      "epoch": 0.6044791823237637,
      "flos": 65316267515520.0,
      "grad_norm": 0.8018026378644575,
      "language_loss": 0.60426438,
      "learning_rate": 1.4283698885372336e-06,
      "loss": 0.62457883,
      "num_input_tokens_seen": 216518130,
      "step": 10054,
      "time_per_iteration": 3.2507219314575195
    },
    {
      "auxiliary_loss_clip": 0.01056424,
      "auxiliary_loss_mlp": 0.01030655,
      "balance_loss_clip": 1.04382837,
      "balance_loss_mlp": 1.01806021,
      "epoch": 0.6045393055764317,
      "flos": 24492369905280.0,
      "grad_norm": 1.5951397016679112,
      "language_loss": 0.85608095,
      "learning_rate": 1.4279966841690027e-06,
      "loss": 0.87695169,
      "num_input_tokens_seen": 216536845,
      "step": 10055,
      "time_per_iteration": 2.655701160430908
    },
    {
      "auxiliary_loss_clip": 0.01091049,
      "auxiliary_loss_mlp": 0.01040739,
      "balance_loss_clip": 1.04066801,
      "balance_loss_mlp": 1.02654696,
      "epoch": 0.6045994288290997,
      "flos": 19054345340160.0,
      "grad_norm": 2.809967490897323,
      "language_loss": 0.73294115,
      "learning_rate": 1.4276235014911952e-06,
      "loss": 0.75425899,
      "num_input_tokens_seen": 216551860,
      "step": 10056,
      "time_per_iteration": 2.477033853530884
    },
    {
      "auxiliary_loss_clip": 0.01074125,
      "auxiliary_loss_mlp": 0.01036273,
      "balance_loss_clip": 1.04142225,
      "balance_loss_mlp": 1.02298737,
      "epoch": 0.6046595520817676,
      "flos": 26576697335040.0,
      "grad_norm": 1.5265776167068597,
      "language_loss": 0.80492866,
      "learning_rate": 1.4272503405179616e-06,
      "loss": 0.82603264,
      "num_input_tokens_seen": 216574775,
      "step": 10057,
      "time_per_iteration": 2.622055768966675
    },
    {
      "auxiliary_loss_clip": 0.01109472,
      "auxiliary_loss_mlp": 0.00787098,
      "balance_loss_clip": 1.03863764,
      "balance_loss_mlp": 1.01336122,
      "epoch": 0.6047196753344356,
      "flos": 13582277660160.0,
      "grad_norm": 2.203609390783533,
      "language_loss": 0.74722803,
      "learning_rate": 1.4268772012634527e-06,
      "loss": 0.76619375,
      "num_input_tokens_seen": 216590100,
      "step": 10058,
      "time_per_iteration": 2.441200017929077
    },
    {
      "auxiliary_loss_clip": 0.01096729,
      "auxiliary_loss_mlp": 0.01027782,
      "balance_loss_clip": 1.0372988,
      "balance_loss_mlp": 1.01575351,
      "epoch": 0.6047797985871035,
      "flos": 25520456977920.0,
      "grad_norm": 2.1534508456887695,
      "language_loss": 0.71329868,
      "learning_rate": 1.4265040837418176e-06,
      "loss": 0.7345438,
      "num_input_tokens_seen": 216610145,
      "step": 10059,
      "time_per_iteration": 2.543438673019409
    },
    {
      "auxiliary_loss_clip": 0.01086643,
      "auxiliary_loss_mlp": 0.01028579,
      "balance_loss_clip": 1.03793144,
      "balance_loss_mlp": 1.01639533,
      "epoch": 0.6048399218397715,
      "flos": 20520147548160.0,
      "grad_norm": 2.424597540042457,
      "language_loss": 0.76132894,
      "learning_rate": 1.4261309879672054e-06,
      "loss": 0.78248119,
      "num_input_tokens_seen": 216630625,
      "step": 10060,
      "time_per_iteration": 2.532600164413452
    },
    {
      "auxiliary_loss_clip": 0.01098083,
      "auxiliary_loss_mlp": 0.01034049,
      "balance_loss_clip": 1.036502,
      "balance_loss_mlp": 1.02206254,
      "epoch": 0.6049000450924396,
      "flos": 20408788408320.0,
      "grad_norm": 2.2244763773548577,
      "language_loss": 0.73717588,
      "learning_rate": 1.4257579139537628e-06,
      "loss": 0.75849724,
      "num_input_tokens_seen": 216649255,
      "step": 10061,
      "time_per_iteration": 2.515824794769287
    },
    {
      "auxiliary_loss_clip": 0.01069205,
      "auxiliary_loss_mlp": 0.00787569,
      "balance_loss_clip": 1.03727555,
      "balance_loss_mlp": 1.01358938,
      "epoch": 0.6049601683451075,
      "flos": 20741357456640.0,
      "grad_norm": 1.9155354692803155,
      "language_loss": 0.67381537,
      "learning_rate": 1.425384861715639e-06,
      "loss": 0.69238311,
      "num_input_tokens_seen": 216668100,
      "step": 10062,
      "time_per_iteration": 2.587156295776367
    },
    {
      "auxiliary_loss_clip": 0.01093396,
      "auxiliary_loss_mlp": 0.0103161,
      "balance_loss_clip": 1.03691483,
      "balance_loss_mlp": 1.01912308,
      "epoch": 0.6050202915977755,
      "flos": 20083114771200.0,
      "grad_norm": 2.0934210712138834,
      "language_loss": 0.71068633,
      "learning_rate": 1.425011831266978e-06,
      "loss": 0.73193645,
      "num_input_tokens_seen": 216686125,
      "step": 10063,
      "time_per_iteration": 2.496239423751831
    },
    {
      "auxiliary_loss_clip": 0.01107002,
      "auxiliary_loss_mlp": 0.01034767,
      "balance_loss_clip": 1.03790951,
      "balance_loss_mlp": 1.02281022,
      "epoch": 0.6050804148504434,
      "flos": 15960821391360.0,
      "grad_norm": 1.7509912637358285,
      "language_loss": 0.84598935,
      "learning_rate": 1.424638822621926e-06,
      "loss": 0.86740708,
      "num_input_tokens_seen": 216704265,
      "step": 10064,
      "time_per_iteration": 2.4439849853515625
    },
    {
      "auxiliary_loss_clip": 0.01098897,
      "auxiliary_loss_mlp": 0.01030894,
      "balance_loss_clip": 1.03793502,
      "balance_loss_mlp": 1.01864564,
      "epoch": 0.6051405381031114,
      "flos": 17456644391040.0,
      "grad_norm": 2.4685377207523262,
      "language_loss": 0.7937507,
      "learning_rate": 1.4242658357946278e-06,
      "loss": 0.81504858,
      "num_input_tokens_seen": 216721765,
      "step": 10065,
      "time_per_iteration": 2.4791414737701416
    },
    {
      "auxiliary_loss_clip": 0.0106653,
      "auxiliary_loss_mlp": 0.01033147,
      "balance_loss_clip": 1.04216015,
      "balance_loss_mlp": 1.01889563,
      "epoch": 0.6052006613557793,
      "flos": 11400130517760.0,
      "grad_norm": 1.9631116759794434,
      "language_loss": 0.78413916,
      "learning_rate": 1.423892870799226e-06,
      "loss": 0.80513591,
      "num_input_tokens_seen": 216738295,
      "step": 10066,
      "time_per_iteration": 2.562825918197632
    },
    {
      "auxiliary_loss_clip": 0.01060878,
      "auxiliary_loss_mlp": 0.0103147,
      "balance_loss_clip": 1.03819776,
      "balance_loss_mlp": 1.01878607,
      "epoch": 0.6052607846084473,
      "flos": 24750998807040.0,
      "grad_norm": 2.1079384813004705,
      "language_loss": 0.72724432,
      "learning_rate": 1.4235199276498655e-06,
      "loss": 0.74816787,
      "num_input_tokens_seen": 216759875,
      "step": 10067,
      "time_per_iteration": 2.6640305519104004
    },
    {
      "auxiliary_loss_clip": 0.01089594,
      "auxiliary_loss_mlp": 0.0078549,
      "balance_loss_clip": 1.03948879,
      "balance_loss_mlp": 1.01107931,
      "epoch": 0.6053209078611153,
      "flos": 20741141975040.0,
      "grad_norm": 1.346722421466001,
      "language_loss": 0.6892972,
      "learning_rate": 1.4231470063606863e-06,
      "loss": 0.70804799,
      "num_input_tokens_seen": 216780705,
      "step": 10068,
      "time_per_iteration": 2.581026077270508
    },
    {
      "auxiliary_loss_clip": 0.01096887,
      "auxiliary_loss_mlp": 0.01033899,
      "balance_loss_clip": 1.03938198,
      "balance_loss_mlp": 1.02137578,
      "epoch": 0.6053810311137833,
      "flos": 18953149749120.0,
      "grad_norm": 4.8685391118757,
      "language_loss": 0.87230909,
      "learning_rate": 1.4227741069458303e-06,
      "loss": 0.89361691,
      "num_input_tokens_seen": 216797625,
      "step": 10069,
      "time_per_iteration": 2.506182909011841
    },
    {
      "auxiliary_loss_clip": 0.01076567,
      "auxiliary_loss_mlp": 0.01024266,
      "balance_loss_clip": 1.03692257,
      "balance_loss_mlp": 1.01288128,
      "epoch": 0.6054411543664512,
      "flos": 23951124794880.0,
      "grad_norm": 1.526687099184133,
      "language_loss": 0.82826591,
      "learning_rate": 1.4224012294194387e-06,
      "loss": 0.84927428,
      "num_input_tokens_seen": 216817610,
      "step": 10070,
      "time_per_iteration": 2.5885255336761475
    },
    {
      "auxiliary_loss_clip": 0.01092829,
      "auxiliary_loss_mlp": 0.01031249,
      "balance_loss_clip": 1.03813362,
      "balance_loss_mlp": 1.01892841,
      "epoch": 0.6055012776191192,
      "flos": 20593979953920.0,
      "grad_norm": 1.512132514023569,
      "language_loss": 0.85890388,
      "learning_rate": 1.4220283737956496e-06,
      "loss": 0.88014466,
      "num_input_tokens_seen": 216836835,
      "step": 10071,
      "time_per_iteration": 2.5930097103118896
    },
    {
      "auxiliary_loss_clip": 0.01101909,
      "auxiliary_loss_mlp": 0.01032398,
      "balance_loss_clip": 1.03912544,
      "balance_loss_mlp": 1.0189867,
      "epoch": 0.6055614008717871,
      "flos": 30298191782400.0,
      "grad_norm": 1.615428585967105,
      "language_loss": 0.77037328,
      "learning_rate": 1.421655540088603e-06,
      "loss": 0.79171634,
      "num_input_tokens_seen": 216856760,
      "step": 10072,
      "time_per_iteration": 4.005670785903931
    },
    {
      "auxiliary_loss_clip": 0.01090166,
      "auxiliary_loss_mlp": 0.0102862,
      "balance_loss_clip": 1.03636837,
      "balance_loss_mlp": 1.01536441,
      "epoch": 0.6056215241244551,
      "flos": 27125017424640.0,
      "grad_norm": 2.209314346321389,
      "language_loss": 0.74752963,
      "learning_rate": 1.4212827283124367e-06,
      "loss": 0.76871747,
      "num_input_tokens_seen": 216878795,
      "step": 10073,
      "time_per_iteration": 2.6189732551574707
    },
    {
      "auxiliary_loss_clip": 0.01015094,
      "auxiliary_loss_mlp": 0.0100186,
      "balance_loss_clip": 1.02935171,
      "balance_loss_mlp": 1.00048351,
      "epoch": 0.6056816473771232,
      "flos": 56007323925120.0,
      "grad_norm": 0.7547475344909664,
      "language_loss": 0.55225217,
      "learning_rate": 1.4209099384812863e-06,
      "loss": 0.57242173,
      "num_input_tokens_seen": 216937800,
      "step": 10074,
      "time_per_iteration": 3.2505900859832764
    },
    {
      "auxiliary_loss_clip": 0.01073018,
      "auxiliary_loss_mlp": 0.01038857,
      "balance_loss_clip": 1.04232025,
      "balance_loss_mlp": 1.02480233,
      "epoch": 0.6057417706297911,
      "flos": 23549499849600.0,
      "grad_norm": 1.8094936446011878,
      "language_loss": 0.81689227,
      "learning_rate": 1.4205371706092894e-06,
      "loss": 0.83801103,
      "num_input_tokens_seen": 216955280,
      "step": 10075,
      "time_per_iteration": 2.6116485595703125
    },
    {
      "auxiliary_loss_clip": 0.01101141,
      "auxiliary_loss_mlp": 0.0102348,
      "balance_loss_clip": 1.03826797,
      "balance_loss_mlp": 1.01123726,
      "epoch": 0.6058018938824591,
      "flos": 27744296832000.0,
      "grad_norm": 2.124959838017282,
      "language_loss": 0.78273004,
      "learning_rate": 1.4201644247105813e-06,
      "loss": 0.80397618,
      "num_input_tokens_seen": 216976950,
      "step": 10076,
      "time_per_iteration": 2.5469682216644287
    },
    {
      "auxiliary_loss_clip": 0.01100008,
      "auxiliary_loss_mlp": 0.01030333,
      "balance_loss_clip": 1.03713512,
      "balance_loss_mlp": 1.01774442,
      "epoch": 0.605862017135127,
      "flos": 22783381643520.0,
      "grad_norm": 1.6994052790445462,
      "language_loss": 0.72594124,
      "learning_rate": 1.4197917007992964e-06,
      "loss": 0.74724466,
      "num_input_tokens_seen": 216996945,
      "step": 10077,
      "time_per_iteration": 2.5557994842529297
    },
    {
      "auxiliary_loss_clip": 0.01111495,
      "auxiliary_loss_mlp": 0.01029074,
      "balance_loss_clip": 1.03958499,
      "balance_loss_mlp": 1.01670003,
      "epoch": 0.605922140387795,
      "flos": 21215019127680.0,
      "grad_norm": 36.37244585435063,
      "language_loss": 0.55466735,
      "learning_rate": 1.4194189988895682e-06,
      "loss": 0.57607305,
      "num_input_tokens_seen": 217016580,
      "step": 10078,
      "time_per_iteration": 2.482950210571289
    },
    {
      "auxiliary_loss_clip": 0.01066806,
      "auxiliary_loss_mlp": 0.0103017,
      "balance_loss_clip": 1.03607512,
      "balance_loss_mlp": 1.01759994,
      "epoch": 0.6059822636404629,
      "flos": 27268372604160.0,
      "grad_norm": 1.532521720993168,
      "language_loss": 0.70199001,
      "learning_rate": 1.4190463189955297e-06,
      "loss": 0.72295982,
      "num_input_tokens_seen": 217037300,
      "step": 10079,
      "time_per_iteration": 2.659311294555664
    },
    {
      "auxiliary_loss_clip": 0.01086417,
      "auxiliary_loss_mlp": 0.01030417,
      "balance_loss_clip": 1.03749657,
      "balance_loss_mlp": 1.01885331,
      "epoch": 0.606042386893131,
      "flos": 20631327120000.0,
      "grad_norm": 1.8889697135097516,
      "language_loss": 0.62466854,
      "learning_rate": 1.4186736611313131e-06,
      "loss": 0.64583683,
      "num_input_tokens_seen": 217055805,
      "step": 10080,
      "time_per_iteration": 2.537776231765747
    },
    {
      "auxiliary_loss_clip": 0.01087986,
      "auxiliary_loss_mlp": 0.01030295,
      "balance_loss_clip": 1.03703547,
      "balance_loss_mlp": 1.01755154,
      "epoch": 0.6061025101457989,
      "flos": 23002293081600.0,
      "grad_norm": 2.0287072227362857,
      "language_loss": 0.7139802,
      "learning_rate": 1.4183010253110492e-06,
      "loss": 0.73516303,
      "num_input_tokens_seen": 217074175,
      "step": 10081,
      "time_per_iteration": 2.578606128692627
    },
    {
      "auxiliary_loss_clip": 0.01088144,
      "auxiliary_loss_mlp": 0.01026355,
      "balance_loss_clip": 1.04286683,
      "balance_loss_mlp": 1.01405859,
      "epoch": 0.6061626333984669,
      "flos": 29898937134720.0,
      "grad_norm": 1.771134490365611,
      "language_loss": 0.68795866,
      "learning_rate": 1.4179284115488691e-06,
      "loss": 0.70910364,
      "num_input_tokens_seen": 217095695,
      "step": 10082,
      "time_per_iteration": 3.9480671882629395
    },
    {
      "auxiliary_loss_clip": 0.01110603,
      "auxiliary_loss_mlp": 0.01029862,
      "balance_loss_clip": 1.03985119,
      "balance_loss_mlp": 1.01795876,
      "epoch": 0.6062227566511348,
      "flos": 25009196745600.0,
      "grad_norm": 1.3659074724538873,
      "language_loss": 0.65917313,
      "learning_rate": 1.4175558198589015e-06,
      "loss": 0.68057787,
      "num_input_tokens_seen": 217116260,
      "step": 10083,
      "time_per_iteration": 3.8909735679626465
    },
    {
      "auxiliary_loss_clip": 0.01098368,
      "auxiliary_loss_mlp": 0.01027993,
      "balance_loss_clip": 1.03775549,
      "balance_loss_mlp": 1.01572025,
      "epoch": 0.6062828799038028,
      "flos": 19463943104640.0,
      "grad_norm": 2.427560007929621,
      "language_loss": 0.73923266,
      "learning_rate": 1.4171832502552764e-06,
      "loss": 0.76049626,
      "num_input_tokens_seen": 217134465,
      "step": 10084,
      "time_per_iteration": 2.4919352531433105
    },
    {
      "auxiliary_loss_clip": 0.01086371,
      "auxiliary_loss_mlp": 0.01031852,
      "balance_loss_clip": 1.03908849,
      "balance_loss_mlp": 1.01924586,
      "epoch": 0.6063430031564707,
      "flos": 13589568120960.0,
      "grad_norm": 3.1147119012383615,
      "language_loss": 0.72302085,
      "learning_rate": 1.4168107027521204e-06,
      "loss": 0.74420309,
      "num_input_tokens_seen": 217149920,
      "step": 10085,
      "time_per_iteration": 2.5389692783355713
    },
    {
      "auxiliary_loss_clip": 0.0110912,
      "auxiliary_loss_mlp": 0.01029878,
      "balance_loss_clip": 1.0387938,
      "balance_loss_mlp": 1.01844001,
      "epoch": 0.6064031264091387,
      "flos": 23255499029760.0,
      "grad_norm": 2.28883799147687,
      "language_loss": 0.76265967,
      "learning_rate": 1.4164381773635605e-06,
      "loss": 0.78404963,
      "num_input_tokens_seen": 217168165,
      "step": 10086,
      "time_per_iteration": 2.5144565105438232
    },
    {
      "auxiliary_loss_clip": 0.01078396,
      "auxiliary_loss_mlp": 0.01029951,
      "balance_loss_clip": 1.03846371,
      "balance_loss_mlp": 1.0174458,
      "epoch": 0.6064632496618068,
      "flos": 22458462192000.0,
      "grad_norm": 1.3638182156046654,
      "language_loss": 0.73000646,
      "learning_rate": 1.4160656741037246e-06,
      "loss": 0.75108999,
      "num_input_tokens_seen": 217190070,
      "step": 10087,
      "time_per_iteration": 3.9673447608947754
    },
    {
      "auxiliary_loss_clip": 0.01095183,
      "auxiliary_loss_mlp": 0.01024976,
      "balance_loss_clip": 1.03723156,
      "balance_loss_mlp": 1.01480782,
      "epoch": 0.6065233729144747,
      "flos": 25118652464640.0,
      "grad_norm": 1.7020182777110142,
      "language_loss": 0.84054983,
      "learning_rate": 1.4156931929867355e-06,
      "loss": 0.86175144,
      "num_input_tokens_seen": 217209370,
      "step": 10088,
      "time_per_iteration": 2.530547618865967
    },
    {
      "auxiliary_loss_clip": 0.01057305,
      "auxiliary_loss_mlp": 0.00785011,
      "balance_loss_clip": 1.03657377,
      "balance_loss_mlp": 1.0093832,
      "epoch": 0.6065834961671427,
      "flos": 23477355383040.0,
      "grad_norm": 2.1312560347616247,
      "language_loss": 0.71591777,
      "learning_rate": 1.4153207340267201e-06,
      "loss": 0.73434097,
      "num_input_tokens_seen": 217226990,
      "step": 10089,
      "time_per_iteration": 2.6258692741394043
    },
    {
      "auxiliary_loss_clip": 0.01098715,
      "auxiliary_loss_mlp": 0.01034237,
      "balance_loss_clip": 1.03908098,
      "balance_loss_mlp": 1.02309644,
      "epoch": 0.6066436194198106,
      "flos": 17019396132480.0,
      "grad_norm": 2.1155703923655618,
      "language_loss": 0.82850921,
      "learning_rate": 1.4149482972378009e-06,
      "loss": 0.84983873,
      "num_input_tokens_seen": 217244585,
      "step": 10090,
      "time_per_iteration": 2.5076911449432373
    },
    {
      "auxiliary_loss_clip": 0.01081239,
      "auxiliary_loss_mlp": 0.01039936,
      "balance_loss_clip": 1.04204226,
      "balance_loss_mlp": 1.02659106,
      "epoch": 0.6067037426724786,
      "flos": 18514752255360.0,
      "grad_norm": 2.816886186304469,
      "language_loss": 0.75057954,
      "learning_rate": 1.4145758826341e-06,
      "loss": 0.77179134,
      "num_input_tokens_seen": 217263435,
      "step": 10091,
      "time_per_iteration": 2.5508944988250732
    },
    {
      "auxiliary_loss_clip": 0.01105864,
      "auxiliary_loss_mlp": 0.01035511,
      "balance_loss_clip": 1.03675258,
      "balance_loss_mlp": 1.02388263,
      "epoch": 0.6067638659251465,
      "flos": 22345989730560.0,
      "grad_norm": 1.5527249444712599,
      "language_loss": 0.79681635,
      "learning_rate": 1.4142034902297415e-06,
      "loss": 0.81823015,
      "num_input_tokens_seen": 217283725,
      "step": 10092,
      "time_per_iteration": 2.514124870300293
    },
    {
      "auxiliary_loss_clip": 0.01087898,
      "auxiliary_loss_mlp": 0.01029619,
      "balance_loss_clip": 1.03711772,
      "balance_loss_mlp": 1.01695275,
      "epoch": 0.6068239891778145,
      "flos": 12451019748480.0,
      "grad_norm": 1.796857633109894,
      "language_loss": 0.76177251,
      "learning_rate": 1.4138311200388444e-06,
      "loss": 0.78294766,
      "num_input_tokens_seen": 217301120,
      "step": 10093,
      "time_per_iteration": 2.504765510559082
    },
    {
      "auxiliary_loss_clip": 0.01082817,
      "auxiliary_loss_mlp": 0.01032849,
      "balance_loss_clip": 1.03776622,
      "balance_loss_mlp": 1.02035642,
      "epoch": 0.6068841124304825,
      "flos": 23185868515200.0,
      "grad_norm": 1.7667628140435616,
      "language_loss": 0.87489903,
      "learning_rate": 1.4134587720755304e-06,
      "loss": 0.8960557,
      "num_input_tokens_seen": 217319585,
      "step": 10094,
      "time_per_iteration": 2.539463996887207
    },
    {
      "auxiliary_loss_clip": 0.01099245,
      "auxiliary_loss_mlp": 0.01024953,
      "balance_loss_clip": 1.03777254,
      "balance_loss_mlp": 1.01241803,
      "epoch": 0.6069442356831505,
      "flos": 18587902302720.0,
      "grad_norm": 1.6380361602165818,
      "language_loss": 0.72406709,
      "learning_rate": 1.413086446353919e-06,
      "loss": 0.74530911,
      "num_input_tokens_seen": 217338880,
      "step": 10095,
      "time_per_iteration": 2.507211446762085
    },
    {
      "auxiliary_loss_clip": 0.01083323,
      "auxiliary_loss_mlp": 0.01027,
      "balance_loss_clip": 1.0356288,
      "balance_loss_mlp": 1.01557398,
      "epoch": 0.6070043589358184,
      "flos": 20960340721920.0,
      "grad_norm": 2.252348080494498,
      "language_loss": 0.76800525,
      "learning_rate": 1.4127141428881273e-06,
      "loss": 0.78910851,
      "num_input_tokens_seen": 217357480,
      "step": 10096,
      "time_per_iteration": 2.553959846496582
    },
    {
      "auxiliary_loss_clip": 0.01109714,
      "auxiliary_loss_mlp": 0.01034686,
      "balance_loss_clip": 1.0382272,
      "balance_loss_mlp": 1.02275896,
      "epoch": 0.6070644821884864,
      "flos": 11692443398400.0,
      "grad_norm": 1.8216081439368703,
      "language_loss": 0.79915786,
      "learning_rate": 1.4123418616922749e-06,
      "loss": 0.82060188,
      "num_input_tokens_seen": 217374575,
      "step": 10097,
      "time_per_iteration": 2.4473273754119873
    },
    {
      "auxiliary_loss_clip": 0.01086015,
      "auxiliary_loss_mlp": 0.01030596,
      "balance_loss_clip": 1.04188621,
      "balance_loss_mlp": 1.01876497,
      "epoch": 0.6071246054411543,
      "flos": 19310568030720.0,
      "grad_norm": 1.5438980391260921,
      "language_loss": 0.67334974,
      "learning_rate": 1.411969602780478e-06,
      "loss": 0.69451582,
      "num_input_tokens_seen": 217392950,
      "step": 10098,
      "time_per_iteration": 2.5609679222106934
    },
    {
      "auxiliary_loss_clip": 0.01107615,
      "auxiliary_loss_mlp": 0.01025373,
      "balance_loss_clip": 1.03776169,
      "balance_loss_mlp": 1.01399493,
      "epoch": 0.6071847286938223,
      "flos": 17749029098880.0,
      "grad_norm": 2.1809886427641145,
      "language_loss": 0.80454355,
      "learning_rate": 1.4115973661668523e-06,
      "loss": 0.82587337,
      "num_input_tokens_seen": 217412145,
      "step": 10099,
      "time_per_iteration": 2.463916778564453
    },
    {
      "auxiliary_loss_clip": 0.0107941,
      "auxiliary_loss_mlp": 0.01034261,
      "balance_loss_clip": 1.0378598,
      "balance_loss_mlp": 1.02145839,
      "epoch": 0.6072448519464904,
      "flos": 22637512512000.0,
      "grad_norm": 4.2772732445089785,
      "language_loss": 0.70693994,
      "learning_rate": 1.4112251518655133e-06,
      "loss": 0.72807664,
      "num_input_tokens_seen": 217432080,
      "step": 10100,
      "time_per_iteration": 2.5603649616241455
    },
    {
      "auxiliary_loss_clip": 0.01075409,
      "auxiliary_loss_mlp": 0.01032691,
      "balance_loss_clip": 1.03980482,
      "balance_loss_mlp": 1.01961398,
      "epoch": 0.6073049751991583,
      "flos": 19537308633600.0,
      "grad_norm": 2.1576441637890666,
      "language_loss": 0.70571029,
      "learning_rate": 1.4108529598905764e-06,
      "loss": 0.72679132,
      "num_input_tokens_seen": 217450945,
      "step": 10101,
      "time_per_iteration": 2.552314519882202
    },
    {
      "auxiliary_loss_clip": 0.01080797,
      "auxiliary_loss_mlp": 0.01025996,
      "balance_loss_clip": 1.03707147,
      "balance_loss_mlp": 1.01400352,
      "epoch": 0.6073650984518263,
      "flos": 28294233033600.0,
      "grad_norm": 1.9419885592159174,
      "language_loss": 0.69043058,
      "learning_rate": 1.410480790256154e-06,
      "loss": 0.7114985,
      "num_input_tokens_seen": 217473105,
      "step": 10102,
      "time_per_iteration": 2.5911049842834473
    },
    {
      "auxiliary_loss_clip": 0.01110628,
      "auxiliary_loss_mlp": 0.01029899,
      "balance_loss_clip": 1.03876746,
      "balance_loss_mlp": 1.01833558,
      "epoch": 0.6074252217044942,
      "flos": 25664422688640.0,
      "grad_norm": 2.134431057031402,
      "language_loss": 0.73848087,
      "learning_rate": 1.4101086429763589e-06,
      "loss": 0.75988615,
      "num_input_tokens_seen": 217491780,
      "step": 10103,
      "time_per_iteration": 2.4777464866638184
    },
    {
      "auxiliary_loss_clip": 0.01084011,
      "auxiliary_loss_mlp": 0.01038861,
      "balance_loss_clip": 1.04460454,
      "balance_loss_mlp": 1.02581358,
      "epoch": 0.6074853449571622,
      "flos": 22857106308480.0,
      "grad_norm": 1.6861131771687243,
      "language_loss": 0.76911712,
      "learning_rate": 1.4097365180653032e-06,
      "loss": 0.79034585,
      "num_input_tokens_seen": 217510605,
      "step": 10104,
      "time_per_iteration": 2.5619850158691406
    },
    {
      "auxiliary_loss_clip": 0.01024129,
      "auxiliary_loss_mlp": 0.01000516,
      "balance_loss_clip": 1.02892244,
      "balance_loss_mlp": 0.99924099,
      "epoch": 0.6075454682098301,
      "flos": 67111406547840.0,
      "grad_norm": 0.7067489793114724,
      "language_loss": 0.55994225,
      "learning_rate": 1.4093644155370977e-06,
      "loss": 0.58018875,
      "num_input_tokens_seen": 217574815,
      "step": 10105,
      "time_per_iteration": 3.1713335514068604
    },
    {
      "auxiliary_loss_clip": 0.01041295,
      "auxiliary_loss_mlp": 0.01001101,
      "balance_loss_clip": 1.02708519,
      "balance_loss_mlp": 0.99978393,
      "epoch": 0.6076055914624982,
      "flos": 70712024751360.0,
      "grad_norm": 0.7589759624522114,
      "language_loss": 0.56881475,
      "learning_rate": 1.4089923354058533e-06,
      "loss": 0.58923876,
      "num_input_tokens_seen": 217632375,
      "step": 10106,
      "time_per_iteration": 3.062274694442749
    },
    {
      "auxiliary_loss_clip": 0.01057769,
      "auxiliary_loss_mlp": 0.01030084,
      "balance_loss_clip": 1.03403449,
      "balance_loss_mlp": 1.01830578,
      "epoch": 0.6076657147151661,
      "flos": 28364545906560.0,
      "grad_norm": 1.5231891464913365,
      "language_loss": 0.69024849,
      "learning_rate": 1.4086202776856784e-06,
      "loss": 0.71112704,
      "num_input_tokens_seen": 217653055,
      "step": 10107,
      "time_per_iteration": 2.6602609157562256
    },
    {
      "auxiliary_loss_clip": 0.01098742,
      "auxiliary_loss_mlp": 0.01026549,
      "balance_loss_clip": 1.03704607,
      "balance_loss_mlp": 1.01463389,
      "epoch": 0.6077258379678341,
      "flos": 15049767807360.0,
      "grad_norm": 2.383265547185116,
      "language_loss": 0.80982554,
      "learning_rate": 1.4082482423906815e-06,
      "loss": 0.83107841,
      "num_input_tokens_seen": 217671520,
      "step": 10108,
      "time_per_iteration": 2.479660749435425
    },
    {
      "auxiliary_loss_clip": 0.01086565,
      "auxiliary_loss_mlp": 0.01031314,
      "balance_loss_clip": 1.03756785,
      "balance_loss_mlp": 1.0180167,
      "epoch": 0.607785961220502,
      "flos": 36167251553280.0,
      "grad_norm": 2.247146226624824,
      "language_loss": 0.71179348,
      "learning_rate": 1.4078762295349714e-06,
      "loss": 0.73297226,
      "num_input_tokens_seen": 217691880,
      "step": 10109,
      "time_per_iteration": 2.6645328998565674
    },
    {
      "auxiliary_loss_clip": 0.01082993,
      "auxiliary_loss_mlp": 0.01030889,
      "balance_loss_clip": 1.03634346,
      "balance_loss_mlp": 1.01995742,
      "epoch": 0.60784608447317,
      "flos": 22524249951360.0,
      "grad_norm": 1.615307358151303,
      "language_loss": 0.80181587,
      "learning_rate": 1.407504239132653e-06,
      "loss": 0.82295465,
      "num_input_tokens_seen": 217710530,
      "step": 10110,
      "time_per_iteration": 2.539525270462036
    },
    {
      "auxiliary_loss_clip": 0.010835,
      "auxiliary_loss_mlp": 0.01030211,
      "balance_loss_clip": 1.03693557,
      "balance_loss_mlp": 1.01781344,
      "epoch": 0.6079062077258379,
      "flos": 23841166285440.0,
      "grad_norm": 2.376654815883698,
      "language_loss": 0.70474136,
      "learning_rate": 1.4071322711978338e-06,
      "loss": 0.72587848,
      "num_input_tokens_seen": 217728650,
      "step": 10111,
      "time_per_iteration": 4.0014402866363525
    },
    {
      "auxiliary_loss_clip": 0.01078302,
      "auxiliary_loss_mlp": 0.01029566,
      "balance_loss_clip": 1.03875852,
      "balance_loss_mlp": 1.01744282,
      "epoch": 0.6079663309785059,
      "flos": 23367037737600.0,
      "grad_norm": 2.0117154825809367,
      "language_loss": 0.65347505,
      "learning_rate": 1.4067603257446186e-06,
      "loss": 0.67455375,
      "num_input_tokens_seen": 217747135,
      "step": 10112,
      "time_per_iteration": 2.5879979133605957
    },
    {
      "auxiliary_loss_clip": 0.01040861,
      "auxiliary_loss_mlp": 0.01004656,
      "balance_loss_clip": 1.02544117,
      "balance_loss_mlp": 1.00338066,
      "epoch": 0.6080264542311739,
      "flos": 71382873110400.0,
      "grad_norm": 0.6485341283140896,
      "language_loss": 0.49555272,
      "learning_rate": 1.4063884027871105e-06,
      "loss": 0.5160079,
      "num_input_tokens_seen": 217811860,
      "step": 10113,
      "time_per_iteration": 3.143420934677124
    },
    {
      "auxiliary_loss_clip": 0.01040704,
      "auxiliary_loss_mlp": 0.01003105,
      "balance_loss_clip": 1.02533448,
      "balance_loss_mlp": 1.0018295,
      "epoch": 0.6080865774838419,
      "flos": 66529833442560.0,
      "grad_norm": 0.839224920970825,
      "language_loss": 0.56921518,
      "learning_rate": 1.4060165023394147e-06,
      "loss": 0.58965325,
      "num_input_tokens_seen": 217866510,
      "step": 10114,
      "time_per_iteration": 3.030075788497925
    },
    {
      "auxiliary_loss_clip": 0.01114013,
      "auxiliary_loss_mlp": 0.0102681,
      "balance_loss_clip": 1.04011869,
      "balance_loss_mlp": 1.01368535,
      "epoch": 0.6081467007365099,
      "flos": 19207935895680.0,
      "grad_norm": 1.9186102390174857,
      "language_loss": 0.70383108,
      "learning_rate": 1.4056446244156317e-06,
      "loss": 0.72523928,
      "num_input_tokens_seen": 217885650,
      "step": 10115,
      "time_per_iteration": 2.4729413986206055
    },
    {
      "auxiliary_loss_clip": 0.01071014,
      "auxiliary_loss_mlp": 0.01027665,
      "balance_loss_clip": 1.03522587,
      "balance_loss_mlp": 1.01520813,
      "epoch": 0.6082068239891778,
      "flos": 24167737762560.0,
      "grad_norm": 1.8493641633281328,
      "language_loss": 0.72481185,
      "learning_rate": 1.4052727690298642e-06,
      "loss": 0.74579865,
      "num_input_tokens_seen": 217905300,
      "step": 10116,
      "time_per_iteration": 2.5942163467407227
    },
    {
      "auxiliary_loss_clip": 0.01088772,
      "auxiliary_loss_mlp": 0.01040124,
      "balance_loss_clip": 1.04132521,
      "balance_loss_mlp": 1.02659941,
      "epoch": 0.6082669472418458,
      "flos": 37413316310400.0,
      "grad_norm": 1.531703364635155,
      "language_loss": 0.53677702,
      "learning_rate": 1.4049009361962138e-06,
      "loss": 0.55806595,
      "num_input_tokens_seen": 217927845,
      "step": 10117,
      "time_per_iteration": 2.662885904312134
    },
    {
      "auxiliary_loss_clip": 0.01089284,
      "auxiliary_loss_mlp": 0.01026941,
      "balance_loss_clip": 1.0377779,
      "balance_loss_mlp": 1.01549745,
      "epoch": 0.6083270704945137,
      "flos": 15085534775040.0,
      "grad_norm": 1.785879291536565,
      "language_loss": 0.7034502,
      "learning_rate": 1.4045291259287786e-06,
      "loss": 0.72461247,
      "num_input_tokens_seen": 217946145,
      "step": 10118,
      "time_per_iteration": 2.5185561180114746
    },
    {
      "auxiliary_loss_clip": 0.01055036,
      "auxiliary_loss_mlp": 0.0103151,
      "balance_loss_clip": 1.03682446,
      "balance_loss_mlp": 1.01933312,
      "epoch": 0.6083871937471818,
      "flos": 20668458804480.0,
      "grad_norm": 1.7941337197016602,
      "language_loss": 0.74591386,
      "learning_rate": 1.4041573382416588e-06,
      "loss": 0.76677936,
      "num_input_tokens_seen": 217965190,
      "step": 10119,
      "time_per_iteration": 2.613905906677246
    },
    {
      "auxiliary_loss_clip": 0.01095943,
      "auxiliary_loss_mlp": 0.01030554,
      "balance_loss_clip": 1.03750825,
      "balance_loss_mlp": 1.01896644,
      "epoch": 0.6084473169998497,
      "flos": 21506901045120.0,
      "grad_norm": 2.069858026655161,
      "language_loss": 0.67005479,
      "learning_rate": 1.4037855731489525e-06,
      "loss": 0.69131982,
      "num_input_tokens_seen": 217983625,
      "step": 10120,
      "time_per_iteration": 2.4982779026031494
    },
    {
      "auxiliary_loss_clip": 0.01102397,
      "auxiliary_loss_mlp": 0.01035246,
      "balance_loss_clip": 1.03989029,
      "balance_loss_mlp": 1.02286005,
      "epoch": 0.6085074402525177,
      "flos": 26870051710080.0,
      "grad_norm": 1.696661199381485,
      "language_loss": 0.74752969,
      "learning_rate": 1.4034138306647571e-06,
      "loss": 0.76890612,
      "num_input_tokens_seen": 218006005,
      "step": 10121,
      "time_per_iteration": 5.2744951248168945
    },
    {
      "auxiliary_loss_clip": 0.01098779,
      "auxiliary_loss_mlp": 0.01025497,
      "balance_loss_clip": 1.03736043,
      "balance_loss_mlp": 1.01403534,
      "epoch": 0.6085675635051856,
      "flos": 10889839952640.0,
      "grad_norm": 1.9442267479161475,
      "language_loss": 0.80375433,
      "learning_rate": 1.4030421108031685e-06,
      "loss": 0.82499713,
      "num_input_tokens_seen": 218024195,
      "step": 10122,
      "time_per_iteration": 2.4725024700164795
    },
    {
      "auxiliary_loss_clip": 0.0109483,
      "auxiliary_loss_mlp": 0.01031382,
      "balance_loss_clip": 1.03785467,
      "balance_loss_mlp": 1.01873422,
      "epoch": 0.6086276867578536,
      "flos": 34862186707200.0,
      "grad_norm": 1.6542715319695749,
      "language_loss": 0.55798459,
      "learning_rate": 1.402670413578284e-06,
      "loss": 0.57924676,
      "num_input_tokens_seen": 218047190,
      "step": 10123,
      "time_per_iteration": 2.631117582321167
    },
    {
      "auxiliary_loss_clip": 0.01100034,
      "auxiliary_loss_mlp": 0.01033493,
      "balance_loss_clip": 1.04006767,
      "balance_loss_mlp": 1.02113152,
      "epoch": 0.6086878100105215,
      "flos": 20047706939520.0,
      "grad_norm": 1.9790485303433387,
      "language_loss": 0.73741937,
      "learning_rate": 1.4022987390041965e-06,
      "loss": 0.75875461,
      "num_input_tokens_seen": 218065945,
      "step": 10124,
      "time_per_iteration": 2.4969887733459473
    },
    {
      "auxiliary_loss_clip": 0.01085164,
      "auxiliary_loss_mlp": 0.01033136,
      "balance_loss_clip": 1.03588128,
      "balance_loss_mlp": 1.0203805,
      "epoch": 0.6087479332631895,
      "flos": 18332469711360.0,
      "grad_norm": 2.01110865161991,
      "language_loss": 0.65510881,
      "learning_rate": 1.4019270870950006e-06,
      "loss": 0.67629176,
      "num_input_tokens_seen": 218085285,
      "step": 10125,
      "time_per_iteration": 3.8776230812072754
    },
    {
      "auxiliary_loss_clip": 0.01109377,
      "auxiliary_loss_mlp": 0.01032711,
      "balance_loss_clip": 1.03950143,
      "balance_loss_mlp": 1.0205518,
      "epoch": 0.6088080565158575,
      "flos": 24493411399680.0,
      "grad_norm": 2.03139563504538,
      "language_loss": 0.76294327,
      "learning_rate": 1.40155545786479e-06,
      "loss": 0.7843641,
      "num_input_tokens_seen": 218104735,
      "step": 10126,
      "time_per_iteration": 2.4875481128692627
    },
    {
      "auxiliary_loss_clip": 0.01077787,
      "auxiliary_loss_mlp": 0.01030588,
      "balance_loss_clip": 1.04146051,
      "balance_loss_mlp": 1.01752257,
      "epoch": 0.6088681797685255,
      "flos": 10269016260480.0,
      "grad_norm": 2.858209441634897,
      "language_loss": 0.71822596,
      "learning_rate": 1.4011838513276558e-06,
      "loss": 0.73930967,
      "num_input_tokens_seen": 218121855,
      "step": 10127,
      "time_per_iteration": 2.531590700149536
    },
    {
      "auxiliary_loss_clip": 0.01114685,
      "auxiliary_loss_mlp": 0.01028935,
      "balance_loss_clip": 1.04054165,
      "balance_loss_mlp": 1.01579213,
      "epoch": 0.6089283030211935,
      "flos": 21973703218560.0,
      "grad_norm": 2.0917846231890986,
      "language_loss": 0.7219485,
      "learning_rate": 1.400812267497691e-06,
      "loss": 0.7433846,
      "num_input_tokens_seen": 218137325,
      "step": 10128,
      "time_per_iteration": 2.4722726345062256
    },
    {
      "auxiliary_loss_clip": 0.01066513,
      "auxiliary_loss_mlp": 0.01030115,
      "balance_loss_clip": 1.03676772,
      "balance_loss_mlp": 1.01831293,
      "epoch": 0.6089884262738614,
      "flos": 17785191116160.0,
      "grad_norm": 2.235975837790555,
      "language_loss": 0.73396081,
      "learning_rate": 1.4004407063889842e-06,
      "loss": 0.75492704,
      "num_input_tokens_seen": 218155530,
      "step": 10129,
      "time_per_iteration": 2.568052291870117
    },
    {
      "auxiliary_loss_clip": 0.01108865,
      "auxiliary_loss_mlp": 0.0102975,
      "balance_loss_clip": 1.03810024,
      "balance_loss_mlp": 1.01753092,
      "epoch": 0.6090485495265294,
      "flos": 36910423946880.0,
      "grad_norm": 1.5310407447365664,
      "language_loss": 0.65575331,
      "learning_rate": 1.400069168015626e-06,
      "loss": 0.67713946,
      "num_input_tokens_seen": 218182535,
      "step": 10130,
      "time_per_iteration": 2.668473482131958
    },
    {
      "auxiliary_loss_clip": 0.010827,
      "auxiliary_loss_mlp": 0.01026036,
      "balance_loss_clip": 1.04056454,
      "balance_loss_mlp": 1.01517034,
      "epoch": 0.6091086727791973,
      "flos": 19899036547200.0,
      "grad_norm": 1.7854178451113443,
      "language_loss": 0.7704854,
      "learning_rate": 1.3996976523917054e-06,
      "loss": 0.79157281,
      "num_input_tokens_seen": 218201740,
      "step": 10131,
      "time_per_iteration": 2.509843349456787
    },
    {
      "auxiliary_loss_clip": 0.01079239,
      "auxiliary_loss_mlp": 0.01030195,
      "balance_loss_clip": 1.03994274,
      "balance_loss_mlp": 1.01928115,
      "epoch": 0.6091687960318654,
      "flos": 22163635359360.0,
      "grad_norm": 1.8119145125513794,
      "language_loss": 0.77362418,
      "learning_rate": 1.3993261595313093e-06,
      "loss": 0.79471856,
      "num_input_tokens_seen": 218219800,
      "step": 10132,
      "time_per_iteration": 2.585703134536743
    },
    {
      "auxiliary_loss_clip": 0.01105255,
      "auxiliary_loss_mlp": 0.01029992,
      "balance_loss_clip": 1.03796816,
      "balance_loss_mlp": 1.01913166,
      "epoch": 0.6092289192845333,
      "flos": 21465280160640.0,
      "grad_norm": 1.794455669509033,
      "language_loss": 0.75940758,
      "learning_rate": 1.3989546894485261e-06,
      "loss": 0.78076005,
      "num_input_tokens_seen": 218237585,
      "step": 10133,
      "time_per_iteration": 2.497647762298584
    },
    {
      "auxiliary_loss_clip": 0.01095238,
      "auxiliary_loss_mlp": 0.01028711,
      "balance_loss_clip": 1.03561091,
      "balance_loss_mlp": 1.01631343,
      "epoch": 0.6092890425372013,
      "flos": 28694924225280.0,
      "grad_norm": 1.873089198292273,
      "language_loss": 0.63727963,
      "learning_rate": 1.3985832421574414e-06,
      "loss": 0.65851915,
      "num_input_tokens_seen": 218258700,
      "step": 10134,
      "time_per_iteration": 2.5632131099700928
    },
    {
      "auxiliary_loss_clip": 0.01084006,
      "auxiliary_loss_mlp": 0.01030072,
      "balance_loss_clip": 1.03696895,
      "balance_loss_mlp": 1.01835966,
      "epoch": 0.6093491657898692,
      "flos": 20813178700800.0,
      "grad_norm": 1.7168237757893283,
      "language_loss": 0.78609681,
      "learning_rate": 1.3982118176721397e-06,
      "loss": 0.80723763,
      "num_input_tokens_seen": 218275655,
      "step": 10135,
      "time_per_iteration": 2.507474422454834
    },
    {
      "auxiliary_loss_clip": 0.01083452,
      "auxiliary_loss_mlp": 0.01026831,
      "balance_loss_clip": 1.03629935,
      "balance_loss_mlp": 1.01584041,
      "epoch": 0.6094092890425372,
      "flos": 25446983708160.0,
      "grad_norm": 2.363236335467373,
      "language_loss": 0.72389984,
      "learning_rate": 1.3978404160067069e-06,
      "loss": 0.74500263,
      "num_input_tokens_seen": 218295720,
      "step": 10136,
      "time_per_iteration": 2.5551064014434814
    },
    {
      "auxiliary_loss_clip": 0.01110666,
      "auxiliary_loss_mlp": 0.01030796,
      "balance_loss_clip": 1.03961921,
      "balance_loss_mlp": 1.01860738,
      "epoch": 0.6094694122952051,
      "flos": 35621265847680.0,
      "grad_norm": 1.7981181628078036,
      "language_loss": 0.74066174,
      "learning_rate": 1.3974690371752253e-06,
      "loss": 0.76207638,
      "num_input_tokens_seen": 218316745,
      "step": 10137,
      "time_per_iteration": 2.6114141941070557
    },
    {
      "auxiliary_loss_clip": 0.01095888,
      "auxiliary_loss_mlp": 0.01038823,
      "balance_loss_clip": 1.03672469,
      "balance_loss_mlp": 1.02513814,
      "epoch": 0.6095295355478731,
      "flos": 24456962073600.0,
      "grad_norm": 1.72962104763926,
      "language_loss": 0.80443662,
      "learning_rate": 1.3970976811917785e-06,
      "loss": 0.82578373,
      "num_input_tokens_seen": 218335385,
      "step": 10138,
      "time_per_iteration": 2.5112786293029785
    },
    {
      "auxiliary_loss_clip": 0.01080951,
      "auxiliary_loss_mlp": 0.01027386,
      "balance_loss_clip": 1.04020405,
      "balance_loss_mlp": 1.01644874,
      "epoch": 0.6095896588005411,
      "flos": 15633208419840.0,
      "grad_norm": 1.5899562880518696,
      "language_loss": 0.80836701,
      "learning_rate": 1.3967263480704481e-06,
      "loss": 0.82945037,
      "num_input_tokens_seen": 218353320,
      "step": 10139,
      "time_per_iteration": 2.5003786087036133
    },
    {
      "auxiliary_loss_clip": 0.01080651,
      "auxiliary_loss_mlp": 0.01032044,
      "balance_loss_clip": 1.03951144,
      "balance_loss_mlp": 1.01945555,
      "epoch": 0.6096497820532091,
      "flos": 15550577182080.0,
      "grad_norm": 2.1010908805724373,
      "language_loss": 0.8361665,
      "learning_rate": 1.396355037825315e-06,
      "loss": 0.85729343,
      "num_input_tokens_seen": 218365620,
      "step": 10140,
      "time_per_iteration": 2.510923385620117
    },
    {
      "auxiliary_loss_clip": 0.01099789,
      "auxiliary_loss_mlp": 0.01028179,
      "balance_loss_clip": 1.03891838,
      "balance_loss_mlp": 1.01618648,
      "epoch": 0.6097099053058771,
      "flos": 24204474397440.0,
      "grad_norm": 1.917319986730665,
      "language_loss": 0.75691509,
      "learning_rate": 1.3959837504704592e-06,
      "loss": 0.77819479,
      "num_input_tokens_seen": 218383785,
      "step": 10141,
      "time_per_iteration": 2.5229432582855225
    },
    {
      "auxiliary_loss_clip": 0.01079984,
      "auxiliary_loss_mlp": 0.0103105,
      "balance_loss_clip": 1.03546834,
      "balance_loss_mlp": 1.01809192,
      "epoch": 0.609770028558545,
      "flos": 19570238426880.0,
      "grad_norm": 1.9392365331154207,
      "language_loss": 0.76060402,
      "learning_rate": 1.3956124860199603e-06,
      "loss": 0.78171438,
      "num_input_tokens_seen": 218399055,
      "step": 10142,
      "time_per_iteration": 2.504120349884033
    },
    {
      "auxiliary_loss_clip": 0.01109195,
      "auxiliary_loss_mlp": 0.01029922,
      "balance_loss_clip": 1.03872252,
      "balance_loss_mlp": 1.01730323,
      "epoch": 0.609830151811213,
      "flos": 23949185460480.0,
      "grad_norm": 1.7393159210216467,
      "language_loss": 0.76719296,
      "learning_rate": 1.3952412444878964e-06,
      "loss": 0.78858411,
      "num_input_tokens_seen": 218419120,
      "step": 10143,
      "time_per_iteration": 2.469773292541504
    },
    {
      "auxiliary_loss_clip": 0.01094025,
      "auxiliary_loss_mlp": 0.01036689,
      "balance_loss_clip": 1.03656435,
      "balance_loss_mlp": 1.02279496,
      "epoch": 0.6098902750638809,
      "flos": 16179732829440.0,
      "grad_norm": 1.7772425180792084,
      "language_loss": 0.75079644,
      "learning_rate": 1.3948700258883448e-06,
      "loss": 0.77210361,
      "num_input_tokens_seen": 218435290,
      "step": 10144,
      "time_per_iteration": 2.4947926998138428
    },
    {
      "auxiliary_loss_clip": 0.01084778,
      "auxiliary_loss_mlp": 0.01027049,
      "balance_loss_clip": 1.04163742,
      "balance_loss_mlp": 1.01508081,
      "epoch": 0.609950398316549,
      "flos": 44526393763200.0,
      "grad_norm": 1.8300954599103036,
      "language_loss": 0.72733283,
      "learning_rate": 1.394498830235383e-06,
      "loss": 0.74845105,
      "num_input_tokens_seen": 218457880,
      "step": 10145,
      "time_per_iteration": 2.7409489154815674
    },
    {
      "auxiliary_loss_clip": 0.01084523,
      "auxiliary_loss_mlp": 0.01029535,
      "balance_loss_clip": 1.03455305,
      "balance_loss_mlp": 1.01781654,
      "epoch": 0.6100105215692169,
      "flos": 23221743223680.0,
      "grad_norm": 1.812367853459384,
      "language_loss": 0.69748074,
      "learning_rate": 1.3941276575430862e-06,
      "loss": 0.71862125,
      "num_input_tokens_seen": 218475930,
      "step": 10146,
      "time_per_iteration": 2.539449691772461
    },
    {
      "auxiliary_loss_clip": 0.01065625,
      "auxiliary_loss_mlp": 0.00784931,
      "balance_loss_clip": 1.04261255,
      "balance_loss_mlp": 1.01222348,
      "epoch": 0.6100706448218849,
      "flos": 15012564295680.0,
      "grad_norm": 2.0080601145439916,
      "language_loss": 0.77256197,
      "learning_rate": 1.3937565078255289e-06,
      "loss": 0.79106748,
      "num_input_tokens_seen": 218493675,
      "step": 10147,
      "time_per_iteration": 2.572662591934204
    },
    {
      "auxiliary_loss_clip": 0.01085018,
      "auxiliary_loss_mlp": 0.01023657,
      "balance_loss_clip": 1.03516579,
      "balance_loss_mlp": 1.01211798,
      "epoch": 0.6101307680745528,
      "flos": 19639976682240.0,
      "grad_norm": 1.86110956229174,
      "language_loss": 0.78293884,
      "learning_rate": 1.393385381096786e-06,
      "loss": 0.80402559,
      "num_input_tokens_seen": 218511780,
      "step": 10148,
      "time_per_iteration": 2.5289101600646973
    },
    {
      "auxiliary_loss_clip": 0.01077175,
      "auxiliary_loss_mlp": 0.01031712,
      "balance_loss_clip": 1.03503716,
      "balance_loss_mlp": 1.01773477,
      "epoch": 0.6101908913272208,
      "flos": 29935566028800.0,
      "grad_norm": 2.3635902014251835,
      "language_loss": 0.54202837,
      "learning_rate": 1.39301427737093e-06,
      "loss": 0.56311721,
      "num_input_tokens_seen": 218531850,
      "step": 10149,
      "time_per_iteration": 2.619762420654297
    },
    {
      "auxiliary_loss_clip": 0.01083485,
      "auxiliary_loss_mlp": 0.01026576,
      "balance_loss_clip": 1.03855777,
      "balance_loss_mlp": 1.01503634,
      "epoch": 0.6102510145798887,
      "flos": 21798639308160.0,
      "grad_norm": 1.8681107549946896,
      "language_loss": 0.80730009,
      "learning_rate": 1.3926431966620333e-06,
      "loss": 0.82840073,
      "num_input_tokens_seen": 218551245,
      "step": 10150,
      "time_per_iteration": 3.9417476654052734
    },
    {
      "auxiliary_loss_clip": 0.01084342,
      "auxiliary_loss_mlp": 0.01032942,
      "balance_loss_clip": 1.03987634,
      "balance_loss_mlp": 1.02071667,
      "epoch": 0.6103111378325567,
      "flos": 20706129192960.0,
      "grad_norm": 1.4856071233612973,
      "language_loss": 0.68915397,
      "learning_rate": 1.3922721389841684e-06,
      "loss": 0.71032679,
      "num_input_tokens_seen": 218571365,
      "step": 10151,
      "time_per_iteration": 2.561948776245117
    },
    {
      "auxiliary_loss_clip": 0.01106503,
      "auxiliary_loss_mlp": 0.01026207,
      "balance_loss_clip": 1.03653932,
      "balance_loss_mlp": 1.01563323,
      "epoch": 0.6103712610852247,
      "flos": 29381643417600.0,
      "grad_norm": 1.7423662520513388,
      "language_loss": 0.70889014,
      "learning_rate": 1.3919011043514036e-06,
      "loss": 0.73021722,
      "num_input_tokens_seen": 218588315,
      "step": 10152,
      "time_per_iteration": 2.5450708866119385
    },
    {
      "auxiliary_loss_clip": 0.01080719,
      "auxiliary_loss_mlp": 0.01031472,
      "balance_loss_clip": 1.03937674,
      "balance_loss_mlp": 1.01929474,
      "epoch": 0.6104313843378927,
      "flos": 20813035046400.0,
      "grad_norm": 1.782877085597693,
      "language_loss": 0.78318816,
      "learning_rate": 1.391530092777811e-06,
      "loss": 0.80431008,
      "num_input_tokens_seen": 218605940,
      "step": 10153,
      "time_per_iteration": 2.5597496032714844
    },
    {
      "auxiliary_loss_clip": 0.01085942,
      "auxiliary_loss_mlp": 0.01030169,
      "balance_loss_clip": 1.03910351,
      "balance_loss_mlp": 1.01776481,
      "epoch": 0.6104915075905607,
      "flos": 26578457101440.0,
      "grad_norm": 2.514899705875235,
      "language_loss": 0.79401416,
      "learning_rate": 1.3911591042774573e-06,
      "loss": 0.81517529,
      "num_input_tokens_seen": 218626100,
      "step": 10154,
      "time_per_iteration": 2.594357490539551
    },
    {
      "auxiliary_loss_clip": 0.01095669,
      "auxiliary_loss_mlp": 0.01027386,
      "balance_loss_clip": 1.03743958,
      "balance_loss_mlp": 1.01632333,
      "epoch": 0.6105516308432286,
      "flos": 23915788790400.0,
      "grad_norm": 1.5875644640180018,
      "language_loss": 0.70466548,
      "learning_rate": 1.3907881388644116e-06,
      "loss": 0.72589606,
      "num_input_tokens_seen": 218645060,
      "step": 10155,
      "time_per_iteration": 2.5733187198638916
    },
    {
      "auxiliary_loss_clip": 0.01099141,
      "auxiliary_loss_mlp": 0.01032359,
      "balance_loss_clip": 1.0399605,
      "balance_loss_mlp": 1.01947212,
      "epoch": 0.6106117540958966,
      "flos": 31577365900800.0,
      "grad_norm": 1.5362458098545764,
      "language_loss": 0.71467257,
      "learning_rate": 1.3904171965527413e-06,
      "loss": 0.7359876,
      "num_input_tokens_seen": 218667690,
      "step": 10156,
      "time_per_iteration": 2.6149215698242188
    },
    {
      "auxiliary_loss_clip": 0.0108415,
      "auxiliary_loss_mlp": 0.01029759,
      "balance_loss_clip": 1.03883815,
      "balance_loss_mlp": 1.01814198,
      "epoch": 0.6106718773485645,
      "flos": 19608160210560.0,
      "grad_norm": 1.8206241603048816,
      "language_loss": 0.67108792,
      "learning_rate": 1.3900462773565114e-06,
      "loss": 0.69222701,
      "num_input_tokens_seen": 218687505,
      "step": 10157,
      "time_per_iteration": 2.526320219039917
    },
    {
      "auxiliary_loss_clip": 0.01067985,
      "auxiliary_loss_mlp": 0.01024081,
      "balance_loss_clip": 1.03313506,
      "balance_loss_mlp": 1.0124526,
      "epoch": 0.6107320006012326,
      "flos": 17123895774720.0,
      "grad_norm": 2.5933821155179406,
      "language_loss": 0.71980464,
      "learning_rate": 1.3896753812897877e-06,
      "loss": 0.74072534,
      "num_input_tokens_seen": 218705315,
      "step": 10158,
      "time_per_iteration": 2.542780637741089
    },
    {
      "auxiliary_loss_clip": 0.01099099,
      "auxiliary_loss_mlp": 0.01037961,
      "balance_loss_clip": 1.04322124,
      "balance_loss_mlp": 1.02592087,
      "epoch": 0.6107921238539005,
      "flos": 30148228500480.0,
      "grad_norm": 1.6006702799657258,
      "language_loss": 0.69394654,
      "learning_rate": 1.389304508366635e-06,
      "loss": 0.71531713,
      "num_input_tokens_seen": 218725735,
      "step": 10159,
      "time_per_iteration": 2.584451675415039
    },
    {
      "auxiliary_loss_clip": 0.01110662,
      "auxiliary_loss_mlp": 0.01028962,
      "balance_loss_clip": 1.03923595,
      "balance_loss_mlp": 1.01648664,
      "epoch": 0.6108522471065685,
      "flos": 18440273404800.0,
      "grad_norm": 1.9284641850138533,
      "language_loss": 0.78986657,
      "learning_rate": 1.3889336586011167e-06,
      "loss": 0.81126279,
      "num_input_tokens_seen": 218743215,
      "step": 10160,
      "time_per_iteration": 5.160766124725342
    },
    {
      "auxiliary_loss_clip": 0.01039791,
      "auxiliary_loss_mlp": 0.01001075,
      "balance_loss_clip": 1.02613711,
      "balance_loss_mlp": 0.99968666,
      "epoch": 0.6109123703592364,
      "flos": 64135454791680.0,
      "grad_norm": 0.8181351945715727,
      "language_loss": 0.61448401,
      "learning_rate": 1.388562832007295e-06,
      "loss": 0.6348927,
      "num_input_tokens_seen": 218806440,
      "step": 10161,
      "time_per_iteration": 3.247194528579712
    },
    {
      "auxiliary_loss_clip": 0.01089724,
      "auxiliary_loss_mlp": 0.00784919,
      "balance_loss_clip": 1.03910327,
      "balance_loss_mlp": 1.01073456,
      "epoch": 0.6109724936119044,
      "flos": 20667848273280.0,
      "grad_norm": 1.7856100134031274,
      "language_loss": 0.7660346,
      "learning_rate": 1.3881920285992324e-06,
      "loss": 0.78478098,
      "num_input_tokens_seen": 218825720,
      "step": 10162,
      "time_per_iteration": 2.5445828437805176
    },
    {
      "auxiliary_loss_clip": 0.01108449,
      "auxiliary_loss_mlp": 0.01030689,
      "balance_loss_clip": 1.03837812,
      "balance_loss_mlp": 1.01878023,
      "epoch": 0.6110326168645723,
      "flos": 31351882273920.0,
      "grad_norm": 1.7572601513815718,
      "language_loss": 0.71885288,
      "learning_rate": 1.3878212483909888e-06,
      "loss": 0.74024427,
      "num_input_tokens_seen": 218847735,
      "step": 10163,
      "time_per_iteration": 2.58492112159729
    },
    {
      "auxiliary_loss_clip": 0.01104489,
      "auxiliary_loss_mlp": 0.0102415,
      "balance_loss_clip": 1.03618336,
      "balance_loss_mlp": 1.01287913,
      "epoch": 0.6110927401172404,
      "flos": 25003378742400.0,
      "grad_norm": 2.2297757113873313,
      "language_loss": 0.60236794,
      "learning_rate": 1.387450491396625e-06,
      "loss": 0.62365437,
      "num_input_tokens_seen": 218866585,
      "step": 10164,
      "time_per_iteration": 3.8871047496795654
    },
    {
      "auxiliary_loss_clip": 0.0109438,
      "auxiliary_loss_mlp": 0.01033188,
      "balance_loss_clip": 1.04022074,
      "balance_loss_mlp": 1.02130914,
      "epoch": 0.6111528633699083,
      "flos": 26248078782720.0,
      "grad_norm": 1.8873474024065287,
      "language_loss": 0.75990093,
      "learning_rate": 1.3870797576302003e-06,
      "loss": 0.78117657,
      "num_input_tokens_seen": 218885560,
      "step": 10165,
      "time_per_iteration": 2.532177448272705
    },
    {
      "auxiliary_loss_clip": 0.01086034,
      "auxiliary_loss_mlp": 0.01025146,
      "balance_loss_clip": 1.03884554,
      "balance_loss_mlp": 1.0134995,
      "epoch": 0.6112129866225763,
      "flos": 22382474970240.0,
      "grad_norm": 1.5810355111808556,
      "language_loss": 0.79265761,
      "learning_rate": 1.3867090471057719e-06,
      "loss": 0.81376946,
      "num_input_tokens_seen": 218905055,
      "step": 10166,
      "time_per_iteration": 2.5334417819976807
    },
    {
      "auxiliary_loss_clip": 0.01086589,
      "auxiliary_loss_mlp": 0.01029244,
      "balance_loss_clip": 1.04019165,
      "balance_loss_mlp": 1.01701355,
      "epoch": 0.6112731098752443,
      "flos": 25227892702080.0,
      "grad_norm": 1.9142568512646563,
      "language_loss": 0.67655414,
      "learning_rate": 1.3863383598373987e-06,
      "loss": 0.69771248,
      "num_input_tokens_seen": 218924030,
      "step": 10167,
      "time_per_iteration": 2.5724308490753174
    },
    {
      "auxiliary_loss_clip": 0.01106213,
      "auxiliary_loss_mlp": 0.01029995,
      "balance_loss_clip": 1.03771079,
      "balance_loss_mlp": 1.01895666,
      "epoch": 0.6113332331279122,
      "flos": 22893160584960.0,
      "grad_norm": 1.7961700419833686,
      "language_loss": 0.79167277,
      "learning_rate": 1.3859676958391364e-06,
      "loss": 0.81303483,
      "num_input_tokens_seen": 218943750,
      "step": 10168,
      "time_per_iteration": 2.475111484527588
    },
    {
      "auxiliary_loss_clip": 0.01115829,
      "auxiliary_loss_mlp": 0.01035394,
      "balance_loss_clip": 1.03994715,
      "balance_loss_mlp": 1.02222157,
      "epoch": 0.6113933563805802,
      "flos": 18620329305600.0,
      "grad_norm": 3.1879555409815588,
      "language_loss": 0.86169767,
      "learning_rate": 1.3855970551250398e-06,
      "loss": 0.88320994,
      "num_input_tokens_seen": 218957585,
      "step": 10169,
      "time_per_iteration": 2.4860074520111084
    },
    {
      "auxiliary_loss_clip": 0.01105278,
      "auxiliary_loss_mlp": 0.01028232,
      "balance_loss_clip": 1.03640556,
      "balance_loss_mlp": 1.01732993,
      "epoch": 0.6114534796332481,
      "flos": 41866275317760.0,
      "grad_norm": 1.6698845254163028,
      "language_loss": 0.79124212,
      "learning_rate": 1.3852264377091652e-06,
      "loss": 0.81257725,
      "num_input_tokens_seen": 218980025,
      "step": 10170,
      "time_per_iteration": 2.6383047103881836
    },
    {
      "auxiliary_loss_clip": 0.0109204,
      "auxiliary_loss_mlp": 0.01036064,
      "balance_loss_clip": 1.03851867,
      "balance_loss_mlp": 1.02276063,
      "epoch": 0.6115136028859162,
      "flos": 21908454163200.0,
      "grad_norm": 2.367535272214396,
      "language_loss": 0.6862554,
      "learning_rate": 1.3848558436055651e-06,
      "loss": 0.70753646,
      "num_input_tokens_seen": 218998200,
      "step": 10171,
      "time_per_iteration": 2.5391898155212402
    },
    {
      "auxiliary_loss_clip": 0.01078992,
      "auxiliary_loss_mlp": 0.01032516,
      "balance_loss_clip": 1.03605032,
      "balance_loss_mlp": 1.01899791,
      "epoch": 0.6115737261385841,
      "flos": 28804846821120.0,
      "grad_norm": 1.606489435806958,
      "language_loss": 0.79047585,
      "learning_rate": 1.3844852728282934e-06,
      "loss": 0.81159091,
      "num_input_tokens_seen": 219017910,
      "step": 10172,
      "time_per_iteration": 2.601757526397705
    },
    {
      "auxiliary_loss_clip": 0.01079576,
      "auxiliary_loss_mlp": 0.01032326,
      "balance_loss_clip": 1.03974771,
      "balance_loss_mlp": 1.01977921,
      "epoch": 0.6116338493912521,
      "flos": 21251468453760.0,
      "grad_norm": 1.9051435859480492,
      "language_loss": 0.66701716,
      "learning_rate": 1.3841147253914022e-06,
      "loss": 0.68813622,
      "num_input_tokens_seen": 219037730,
      "step": 10173,
      "time_per_iteration": 2.5703351497650146
    },
    {
      "auxiliary_loss_clip": 0.01085336,
      "auxiliary_loss_mlp": 0.0103252,
      "balance_loss_clip": 1.03827262,
      "balance_loss_mlp": 1.01953828,
      "epoch": 0.61169397264392,
      "flos": 17530189488000.0,
      "grad_norm": 1.8851407338856854,
      "language_loss": 0.55363429,
      "learning_rate": 1.3837442013089416e-06,
      "loss": 0.57481283,
      "num_input_tokens_seen": 219056755,
      "step": 10174,
      "time_per_iteration": 2.487077474594116
    },
    {
      "auxiliary_loss_clip": 0.01091291,
      "auxiliary_loss_mlp": 0.01033021,
      "balance_loss_clip": 1.0400902,
      "balance_loss_mlp": 1.02059317,
      "epoch": 0.611754095896588,
      "flos": 23951555758080.0,
      "grad_norm": 2.0585531417243024,
      "language_loss": 0.66235387,
      "learning_rate": 1.3833737005949628e-06,
      "loss": 0.68359697,
      "num_input_tokens_seen": 219076985,
      "step": 10175,
      "time_per_iteration": 2.5436248779296875
    },
    {
      "auxiliary_loss_clip": 0.01092964,
      "auxiliary_loss_mlp": 0.00784987,
      "balance_loss_clip": 1.03435385,
      "balance_loss_mlp": 1.01135993,
      "epoch": 0.6118142191492559,
      "flos": 25994872834560.0,
      "grad_norm": 2.1660646922432507,
      "language_loss": 0.82745576,
      "learning_rate": 1.3830032232635154e-06,
      "loss": 0.84623528,
      "num_input_tokens_seen": 219096050,
      "step": 10176,
      "time_per_iteration": 2.5258448123931885
    },
    {
      "auxiliary_loss_clip": 0.01091179,
      "auxiliary_loss_mlp": 0.01032469,
      "balance_loss_clip": 1.03945029,
      "balance_loss_mlp": 1.01963639,
      "epoch": 0.611874342401924,
      "flos": 24603190341120.0,
      "grad_norm": 1.9120050131464255,
      "language_loss": 0.77478898,
      "learning_rate": 1.3826327693286474e-06,
      "loss": 0.79602551,
      "num_input_tokens_seen": 219112665,
      "step": 10177,
      "time_per_iteration": 2.5447707176208496
    },
    {
      "auxiliary_loss_clip": 0.01094922,
      "auxiliary_loss_mlp": 0.00787892,
      "balance_loss_clip": 1.03725624,
      "balance_loss_mlp": 1.0143013,
      "epoch": 0.6119344656545919,
      "flos": 15887132640000.0,
      "grad_norm": 2.1458630443362723,
      "language_loss": 0.75486624,
      "learning_rate": 1.3822623388044065e-06,
      "loss": 0.7736944,
      "num_input_tokens_seen": 219129120,
      "step": 10178,
      "time_per_iteration": 2.4664957523345947
    },
    {
      "auxiliary_loss_clip": 0.01084039,
      "auxiliary_loss_mlp": 0.01040886,
      "balance_loss_clip": 1.03765059,
      "balance_loss_mlp": 1.02677155,
      "epoch": 0.6119945889072599,
      "flos": 21652877917440.0,
      "grad_norm": 2.9479238990543837,
      "language_loss": 0.66898644,
      "learning_rate": 1.3818919317048402e-06,
      "loss": 0.69023561,
      "num_input_tokens_seen": 219148950,
      "step": 10179,
      "time_per_iteration": 2.5287182331085205
    },
    {
      "auxiliary_loss_clip": 0.01091012,
      "auxiliary_loss_mlp": 0.01038202,
      "balance_loss_clip": 1.04047966,
      "balance_loss_mlp": 1.02623916,
      "epoch": 0.6120547121599279,
      "flos": 13772533023360.0,
      "grad_norm": 1.7656587862876623,
      "language_loss": 0.84259897,
      "learning_rate": 1.3815215480439933e-06,
      "loss": 0.86389112,
      "num_input_tokens_seen": 219165585,
      "step": 10180,
      "time_per_iteration": 2.4879043102264404
    },
    {
      "auxiliary_loss_clip": 0.01108867,
      "auxiliary_loss_mlp": 0.01027426,
      "balance_loss_clip": 1.03928947,
      "balance_loss_mlp": 1.01486194,
      "epoch": 0.6121148354125958,
      "flos": 20079164275200.0,
      "grad_norm": 1.6146685470597626,
      "language_loss": 0.77675188,
      "learning_rate": 1.3811511878359113e-06,
      "loss": 0.79811484,
      "num_input_tokens_seen": 219183280,
      "step": 10181,
      "time_per_iteration": 2.4676477909088135
    },
    {
      "auxiliary_loss_clip": 0.01108362,
      "auxiliary_loss_mlp": 0.01027401,
      "balance_loss_clip": 1.0372467,
      "balance_loss_mlp": 1.01569438,
      "epoch": 0.6121749586652638,
      "flos": 13471313569920.0,
      "grad_norm": 1.9976146753898771,
      "language_loss": 0.80407619,
      "learning_rate": 1.3807808510946384e-06,
      "loss": 0.82543385,
      "num_input_tokens_seen": 219197200,
      "step": 10182,
      "time_per_iteration": 2.410322904586792
    },
    {
      "auxiliary_loss_clip": 0.01070267,
      "auxiliary_loss_mlp": 0.01029706,
      "balance_loss_clip": 1.03342152,
      "balance_loss_mlp": 1.01888752,
      "epoch": 0.6122350819179317,
      "flos": 20120533764480.0,
      "grad_norm": 2.6237994279498236,
      "language_loss": 0.83146781,
      "learning_rate": 1.3804105378342177e-06,
      "loss": 0.85246748,
      "num_input_tokens_seen": 219216825,
      "step": 10183,
      "time_per_iteration": 2.548956871032715
    },
    {
      "auxiliary_loss_clip": 0.01040661,
      "auxiliary_loss_mlp": 0.01001611,
      "balance_loss_clip": 1.03084183,
      "balance_loss_mlp": 1.0001868,
      "epoch": 0.6122952051705998,
      "flos": 65429242767360.0,
      "grad_norm": 0.7022768072656922,
      "language_loss": 0.62885171,
      "learning_rate": 1.3800402480686914e-06,
      "loss": 0.64927441,
      "num_input_tokens_seen": 219283795,
      "step": 10184,
      "time_per_iteration": 3.2226603031158447
    },
    {
      "auxiliary_loss_clip": 0.01099949,
      "auxiliary_loss_mlp": 0.010286,
      "balance_loss_clip": 1.03837574,
      "balance_loss_mlp": 1.01743042,
      "epoch": 0.6123553284232677,
      "flos": 20376253664640.0,
      "grad_norm": 1.8147870050876351,
      "language_loss": 0.82094753,
      "learning_rate": 1.379669981812101e-06,
      "loss": 0.84223306,
      "num_input_tokens_seen": 219302385,
      "step": 10185,
      "time_per_iteration": 2.489056348800659
    },
    {
      "auxiliary_loss_clip": 0.01086258,
      "auxiliary_loss_mlp": 0.01034421,
      "balance_loss_clip": 1.03816962,
      "balance_loss_mlp": 1.02179098,
      "epoch": 0.6124154516759357,
      "flos": 23987645948160.0,
      "grad_norm": 3.5252668067394377,
      "language_loss": 0.74537563,
      "learning_rate": 1.3792997390784868e-06,
      "loss": 0.76658237,
      "num_input_tokens_seen": 219319765,
      "step": 10186,
      "time_per_iteration": 2.578298330307007
    },
    {
      "auxiliary_loss_clip": 0.01093134,
      "auxiliary_loss_mlp": 0.01026952,
      "balance_loss_clip": 1.03544974,
      "balance_loss_mlp": 1.01580584,
      "epoch": 0.6124755749286036,
      "flos": 21468799693440.0,
      "grad_norm": 1.6015830423530744,
      "language_loss": 0.78132993,
      "learning_rate": 1.3789295198818895e-06,
      "loss": 0.80253077,
      "num_input_tokens_seen": 219337440,
      "step": 10187,
      "time_per_iteration": 2.502084732055664
    },
    {
      "auxiliary_loss_clip": 0.01105797,
      "auxiliary_loss_mlp": 0.01028408,
      "balance_loss_clip": 1.03572822,
      "balance_loss_mlp": 1.01636779,
      "epoch": 0.6125356981812716,
      "flos": 23879195809920.0,
      "grad_norm": 1.76111000557454,
      "language_loss": 0.83186466,
      "learning_rate": 1.3785593242363462e-06,
      "loss": 0.85320669,
      "num_input_tokens_seen": 219357525,
      "step": 10188,
      "time_per_iteration": 2.500108480453491
    },
    {
      "auxiliary_loss_clip": 0.01075087,
      "auxiliary_loss_mlp": 0.01027012,
      "balance_loss_clip": 1.03529668,
      "balance_loss_mlp": 1.01556146,
      "epoch": 0.6125958214339395,
      "flos": 14425604150400.0,
      "grad_norm": 1.8096866681312471,
      "language_loss": 0.75693601,
      "learning_rate": 1.378189152155896e-06,
      "loss": 0.77795696,
      "num_input_tokens_seen": 219374855,
      "step": 10189,
      "time_per_iteration": 3.9984209537506104
    },
    {
      "auxiliary_loss_clip": 0.01094062,
      "auxiliary_loss_mlp": 0.01031518,
      "balance_loss_clip": 1.03628778,
      "balance_loss_mlp": 1.01901889,
      "epoch": 0.6126559446866076,
      "flos": 23259090389760.0,
      "grad_norm": 1.6778316556019974,
      "language_loss": 0.74275112,
      "learning_rate": 1.3778190036545758e-06,
      "loss": 0.76400685,
      "num_input_tokens_seen": 219394740,
      "step": 10190,
      "time_per_iteration": 2.51216459274292
    },
    {
      "auxiliary_loss_clip": 0.01099358,
      "auxiliary_loss_mlp": 0.01029618,
      "balance_loss_clip": 1.03943455,
      "balance_loss_mlp": 1.01706481,
      "epoch": 0.6127160679392755,
      "flos": 26864808324480.0,
      "grad_norm": 1.8199000982029145,
      "language_loss": 0.68294638,
      "learning_rate": 1.3774488787464207e-06,
      "loss": 0.70423615,
      "num_input_tokens_seen": 219413755,
      "step": 10191,
      "time_per_iteration": 2.560763120651245
    },
    {
      "auxiliary_loss_clip": 0.01095655,
      "auxiliary_loss_mlp": 0.0103257,
      "balance_loss_clip": 1.03646564,
      "balance_loss_mlp": 1.01974344,
      "epoch": 0.6127761911919435,
      "flos": 26396425952640.0,
      "grad_norm": 2.979927296657686,
      "language_loss": 0.73893303,
      "learning_rate": 1.377078777445467e-06,
      "loss": 0.76021528,
      "num_input_tokens_seen": 219433560,
      "step": 10192,
      "time_per_iteration": 2.534044027328491
    },
    {
      "auxiliary_loss_clip": 0.01076914,
      "auxiliary_loss_mlp": 0.01029556,
      "balance_loss_clip": 1.03871632,
      "balance_loss_mlp": 1.01784396,
      "epoch": 0.6128363144446115,
      "flos": 22634747164800.0,
      "grad_norm": 1.8328183827410847,
      "language_loss": 0.83676761,
      "learning_rate": 1.3767086997657478e-06,
      "loss": 0.85783231,
      "num_input_tokens_seen": 219452640,
      "step": 10193,
      "time_per_iteration": 2.5681216716766357
    },
    {
      "auxiliary_loss_clip": 0.01075755,
      "auxiliary_loss_mlp": 0.01030134,
      "balance_loss_clip": 1.03667557,
      "balance_loss_mlp": 1.01802862,
      "epoch": 0.6128964376972794,
      "flos": 26759051706240.0,
      "grad_norm": 2.266010589479677,
      "language_loss": 0.69968528,
      "learning_rate": 1.3763386457212979e-06,
      "loss": 0.72074413,
      "num_input_tokens_seen": 219468585,
      "step": 10194,
      "time_per_iteration": 2.5876457691192627
    },
    {
      "auxiliary_loss_clip": 0.01024636,
      "auxiliary_loss_mlp": 0.01001403,
      "balance_loss_clip": 1.03119636,
      "balance_loss_mlp": 0.99993652,
      "epoch": 0.6129565609499474,
      "flos": 65567929178880.0,
      "grad_norm": 0.8235849902142223,
      "language_loss": 0.58698481,
      "learning_rate": 1.375968615326149e-06,
      "loss": 0.60724521,
      "num_input_tokens_seen": 219523015,
      "step": 10195,
      "time_per_iteration": 2.9530692100524902
    },
    {
      "auxiliary_loss_clip": 0.01089044,
      "auxiliary_loss_mlp": 0.01033119,
      "balance_loss_clip": 1.03925681,
      "balance_loss_mlp": 1.02026248,
      "epoch": 0.6130166842026153,
      "flos": 16362087200640.0,
      "grad_norm": 2.0283780970817014,
      "language_loss": 0.69745886,
      "learning_rate": 1.3755986085943324e-06,
      "loss": 0.71868044,
      "num_input_tokens_seen": 219539980,
      "step": 10196,
      "time_per_iteration": 2.4988269805908203
    },
    {
      "auxiliary_loss_clip": 0.01083896,
      "auxiliary_loss_mlp": 0.01036206,
      "balance_loss_clip": 1.03658199,
      "balance_loss_mlp": 1.02445221,
      "epoch": 0.6130768074552834,
      "flos": 23652455207040.0,
      "grad_norm": 1.8623167749058633,
      "language_loss": 0.71213996,
      "learning_rate": 1.3752286255398788e-06,
      "loss": 0.73334098,
      "num_input_tokens_seen": 219556980,
      "step": 10197,
      "time_per_iteration": 2.5619845390319824
    },
    {
      "auxiliary_loss_clip": 0.01096203,
      "auxiliary_loss_mlp": 0.01043961,
      "balance_loss_clip": 1.03658664,
      "balance_loss_mlp": 1.03021014,
      "epoch": 0.6131369307079513,
      "flos": 20047455544320.0,
      "grad_norm": 1.9462460614312307,
      "language_loss": 0.78875613,
      "learning_rate": 1.3748586661768191e-06,
      "loss": 0.81015778,
      "num_input_tokens_seen": 219576410,
      "step": 10198,
      "time_per_iteration": 3.835993528366089
    },
    {
      "auxiliary_loss_clip": 0.01079207,
      "auxiliary_loss_mlp": 0.01027939,
      "balance_loss_clip": 1.03741086,
      "balance_loss_mlp": 1.01554728,
      "epoch": 0.6131970539606193,
      "flos": 22672166158080.0,
      "grad_norm": 1.4314566593996108,
      "language_loss": 0.74471033,
      "learning_rate": 1.374488730519181e-06,
      "loss": 0.76578182,
      "num_input_tokens_seen": 219597180,
      "step": 10199,
      "time_per_iteration": 3.9720675945281982
    },
    {
      "auxiliary_loss_clip": 0.01086606,
      "auxiliary_loss_mlp": 0.01037243,
      "balance_loss_clip": 1.03985882,
      "balance_loss_mlp": 1.02434492,
      "epoch": 0.6132571772132872,
      "flos": 26870913636480.0,
      "grad_norm": 1.7137009607826121,
      "language_loss": 0.61709464,
      "learning_rate": 1.374118818580993e-06,
      "loss": 0.63833314,
      "num_input_tokens_seen": 219617630,
      "step": 10200,
      "time_per_iteration": 2.5755362510681152
    },
    {
      "auxiliary_loss_clip": 0.01082855,
      "auxiliary_loss_mlp": 0.01032493,
      "balance_loss_clip": 1.03654432,
      "balance_loss_mlp": 1.02041769,
      "epoch": 0.6133173004659552,
      "flos": 22892657794560.0,
      "grad_norm": 1.8980646399552337,
      "language_loss": 0.68832231,
      "learning_rate": 1.3737489303762822e-06,
      "loss": 0.70947582,
      "num_input_tokens_seen": 219637025,
      "step": 10201,
      "time_per_iteration": 2.5203874111175537
    },
    {
      "auxiliary_loss_clip": 0.01083913,
      "auxiliary_loss_mlp": 0.01027391,
      "balance_loss_clip": 1.03465509,
      "balance_loss_mlp": 1.01517773,
      "epoch": 0.6133774237186231,
      "flos": 20485098852480.0,
      "grad_norm": 1.8709508096096885,
      "language_loss": 0.83417147,
      "learning_rate": 1.3733790659190746e-06,
      "loss": 0.85528451,
      "num_input_tokens_seen": 219656625,
      "step": 10202,
      "time_per_iteration": 2.523024797439575
    },
    {
      "auxiliary_loss_clip": 0.01050576,
      "auxiliary_loss_mlp": 0.00999771,
      "balance_loss_clip": 1.02662802,
      "balance_loss_mlp": 0.99825668,
      "epoch": 0.6134375469712912,
      "flos": 69413065217280.0,
      "grad_norm": 0.9154062420573623,
      "language_loss": 0.67095113,
      "learning_rate": 1.3730092252233953e-06,
      "loss": 0.69145465,
      "num_input_tokens_seen": 219718090,
      "step": 10203,
      "time_per_iteration": 4.557872533798218
    },
    {
      "auxiliary_loss_clip": 0.01098851,
      "auxiliary_loss_mlp": 0.0103018,
      "balance_loss_clip": 1.03732657,
      "balance_loss_mlp": 1.01803279,
      "epoch": 0.6134976702239591,
      "flos": 41281541815680.0,
      "grad_norm": 1.7769324382872933,
      "language_loss": 0.61393714,
      "learning_rate": 1.37263940830327e-06,
      "loss": 0.63522744,
      "num_input_tokens_seen": 219740100,
      "step": 10204,
      "time_per_iteration": 2.666447401046753
    },
    {
      "auxiliary_loss_clip": 0.01071889,
      "auxiliary_loss_mlp": 0.01031362,
      "balance_loss_clip": 1.03573871,
      "balance_loss_mlp": 1.01917291,
      "epoch": 0.6135577934766271,
      "flos": 22346600261760.0,
      "grad_norm": 2.3897787938049118,
      "language_loss": 0.72596067,
      "learning_rate": 1.3722696151727204e-06,
      "loss": 0.74699318,
      "num_input_tokens_seen": 219761225,
      "step": 10205,
      "time_per_iteration": 2.592449426651001
    },
    {
      "auxiliary_loss_clip": 0.01095865,
      "auxiliary_loss_mlp": 0.01023205,
      "balance_loss_clip": 1.03816569,
      "balance_loss_mlp": 1.01073587,
      "epoch": 0.6136179167292951,
      "flos": 23728155120000.0,
      "grad_norm": 1.8067155011382532,
      "language_loss": 0.76360196,
      "learning_rate": 1.3718998458457701e-06,
      "loss": 0.78479266,
      "num_input_tokens_seen": 219780085,
      "step": 10206,
      "time_per_iteration": 2.5326292514801025
    },
    {
      "auxiliary_loss_clip": 0.01074129,
      "auxiliary_loss_mlp": 0.01026323,
      "balance_loss_clip": 1.03786755,
      "balance_loss_mlp": 1.01322186,
      "epoch": 0.613678039981963,
      "flos": 26024678144640.0,
      "grad_norm": 1.8978621964155893,
      "language_loss": 0.75261116,
      "learning_rate": 1.3715301003364407e-06,
      "loss": 0.77361566,
      "num_input_tokens_seen": 219797895,
      "step": 10207,
      "time_per_iteration": 2.6151251792907715
    },
    {
      "auxiliary_loss_clip": 0.01097819,
      "auxiliary_loss_mlp": 0.0103179,
      "balance_loss_clip": 1.03820753,
      "balance_loss_mlp": 1.02004838,
      "epoch": 0.613738163234631,
      "flos": 9859957200000.0,
      "grad_norm": 2.555077534421386,
      "language_loss": 0.8252461,
      "learning_rate": 1.3711603786587525e-06,
      "loss": 0.84654218,
      "num_input_tokens_seen": 219811295,
      "step": 10208,
      "time_per_iteration": 2.4949758052825928
    },
    {
      "auxiliary_loss_clip": 0.01092763,
      "auxiliary_loss_mlp": 0.01032406,
      "balance_loss_clip": 1.04013252,
      "balance_loss_mlp": 1.01945424,
      "epoch": 0.613798286487299,
      "flos": 33182070001920.0,
      "grad_norm": 1.7685576060596173,
      "language_loss": 0.7251513,
      "learning_rate": 1.3707906808267265e-06,
      "loss": 0.74640298,
      "num_input_tokens_seen": 219832735,
      "step": 10209,
      "time_per_iteration": 2.6282427310943604
    },
    {
      "auxiliary_loss_clip": 0.0110913,
      "auxiliary_loss_mlp": 0.01033454,
      "balance_loss_clip": 1.0397408,
      "balance_loss_mlp": 1.02165842,
      "epoch": 0.613858409739967,
      "flos": 25627901535360.0,
      "grad_norm": 1.6759304400704642,
      "language_loss": 0.7409184,
      "learning_rate": 1.37042100685438e-06,
      "loss": 0.76234424,
      "num_input_tokens_seen": 219852755,
      "step": 10210,
      "time_per_iteration": 2.5209672451019287
    },
    {
      "auxiliary_loss_clip": 0.01024328,
      "auxiliary_loss_mlp": 0.01000136,
      "balance_loss_clip": 1.02747774,
      "balance_loss_mlp": 0.99862796,
      "epoch": 0.6139185329926349,
      "flos": 67192313932800.0,
      "grad_norm": 0.8858236280253011,
      "language_loss": 0.64987516,
      "learning_rate": 1.3700513567557325e-06,
      "loss": 0.67011988,
      "num_input_tokens_seen": 219922785,
      "step": 10211,
      "time_per_iteration": 3.3196659088134766
    },
    {
      "auxiliary_loss_clip": 0.01086298,
      "auxiliary_loss_mlp": 0.00785925,
      "balance_loss_clip": 1.03694797,
      "balance_loss_mlp": 1.00914359,
      "epoch": 0.6139786562453029,
      "flos": 21543637680000.0,
      "grad_norm": 1.6198379044707127,
      "language_loss": 0.75674862,
      "learning_rate": 1.369681730544801e-06,
      "loss": 0.77547085,
      "num_input_tokens_seen": 219942215,
      "step": 10212,
      "time_per_iteration": 2.5614516735076904
    },
    {
      "auxiliary_loss_clip": 0.01083389,
      "auxiliary_loss_mlp": 0.01038145,
      "balance_loss_clip": 1.03661454,
      "balance_loss_mlp": 1.02431655,
      "epoch": 0.6140387794979708,
      "flos": 26068489758720.0,
      "grad_norm": 1.7322671967399428,
      "language_loss": 0.7396251,
      "learning_rate": 1.3693121282356009e-06,
      "loss": 0.76084042,
      "num_input_tokens_seen": 219963830,
      "step": 10213,
      "time_per_iteration": 2.563505172729492
    },
    {
      "auxiliary_loss_clip": 0.01091717,
      "auxiliary_loss_mlp": 0.01033704,
      "balance_loss_clip": 1.03780079,
      "balance_loss_mlp": 1.02031684,
      "epoch": 0.6140989027506388,
      "flos": 23694614795520.0,
      "grad_norm": 1.4971214823645598,
      "language_loss": 0.730308,
      "learning_rate": 1.3689425498421483e-06,
      "loss": 0.75156212,
      "num_input_tokens_seen": 219983815,
      "step": 10214,
      "time_per_iteration": 2.5438082218170166
    },
    {
      "auxiliary_loss_clip": 0.01110875,
      "auxiliary_loss_mlp": 0.01028778,
      "balance_loss_clip": 1.03800249,
      "balance_loss_mlp": 1.01543283,
      "epoch": 0.6141590260033067,
      "flos": 22231721589120.0,
      "grad_norm": 4.450962756596791,
      "language_loss": 0.74865997,
      "learning_rate": 1.3685729953784572e-06,
      "loss": 0.77005649,
      "num_input_tokens_seen": 220003165,
      "step": 10215,
      "time_per_iteration": 2.4792234897613525
    },
    {
      "auxiliary_loss_clip": 0.01095603,
      "auxiliary_loss_mlp": 0.01028119,
      "balance_loss_clip": 1.03800476,
      "balance_loss_mlp": 1.01578116,
      "epoch": 0.6142191492559748,
      "flos": 23871653953920.0,
      "grad_norm": 1.8587732938396342,
      "language_loss": 0.78437495,
      "learning_rate": 1.368203464858542e-06,
      "loss": 0.80561221,
      "num_input_tokens_seen": 220021015,
      "step": 10216,
      "time_per_iteration": 2.4932825565338135
    },
    {
      "auxiliary_loss_clip": 0.01109399,
      "auxiliary_loss_mlp": 0.0103382,
      "balance_loss_clip": 1.03902853,
      "balance_loss_mlp": 1.02040315,
      "epoch": 0.6142792725086427,
      "flos": 15042513260160.0,
      "grad_norm": 2.318602889699367,
      "language_loss": 0.80001408,
      "learning_rate": 1.3678339582964147e-06,
      "loss": 0.8214463,
      "num_input_tokens_seen": 220035780,
      "step": 10217,
      "time_per_iteration": 2.423921823501587
    },
    {
      "auxiliary_loss_clip": 0.01089122,
      "auxiliary_loss_mlp": 0.01026092,
      "balance_loss_clip": 1.03718984,
      "balance_loss_mlp": 1.01382577,
      "epoch": 0.6143393957613107,
      "flos": 23330947547520.0,
      "grad_norm": 2.834210476196226,
      "language_loss": 0.78120613,
      "learning_rate": 1.3674644757060865e-06,
      "loss": 0.80235827,
      "num_input_tokens_seen": 220054280,
      "step": 10218,
      "time_per_iteration": 2.5289647579193115
    },
    {
      "auxiliary_loss_clip": 0.01099492,
      "auxiliary_loss_mlp": 0.0103255,
      "balance_loss_clip": 1.03815651,
      "balance_loss_mlp": 1.02034843,
      "epoch": 0.6143995190139786,
      "flos": 20117086058880.0,
      "grad_norm": 1.5673785903940236,
      "language_loss": 0.81909072,
      "learning_rate": 1.367095017101569e-06,
      "loss": 0.84041107,
      "num_input_tokens_seen": 220074120,
      "step": 10219,
      "time_per_iteration": 2.4970695972442627
    },
    {
      "auxiliary_loss_clip": 0.01096317,
      "auxiliary_loss_mlp": 0.01031347,
      "balance_loss_clip": 1.03866565,
      "balance_loss_mlp": 1.01834726,
      "epoch": 0.6144596422666466,
      "flos": 42303559489920.0,
      "grad_norm": 1.93357609024482,
      "language_loss": 0.66760385,
      "learning_rate": 1.3667255824968717e-06,
      "loss": 0.68888044,
      "num_input_tokens_seen": 220096320,
      "step": 10220,
      "time_per_iteration": 2.6759612560272217
    },
    {
      "auxiliary_loss_clip": 0.01097559,
      "auxiliary_loss_mlp": 0.01027433,
      "balance_loss_clip": 1.03649032,
      "balance_loss_mlp": 1.01581633,
      "epoch": 0.6145197655193146,
      "flos": 21573622558080.0,
      "grad_norm": 2.1283990410274702,
      "language_loss": 0.71633416,
      "learning_rate": 1.3663561719060041e-06,
      "loss": 0.73758405,
      "num_input_tokens_seen": 220114850,
      "step": 10221,
      "time_per_iteration": 2.494719982147217
    },
    {
      "auxiliary_loss_clip": 0.01063045,
      "auxiliary_loss_mlp": 0.01028043,
      "balance_loss_clip": 1.03575158,
      "balance_loss_mlp": 1.01618218,
      "epoch": 0.6145798887719826,
      "flos": 21471098163840.0,
      "grad_norm": 3.7930726611775216,
      "language_loss": 0.79521847,
      "learning_rate": 1.3659867853429735e-06,
      "loss": 0.81612939,
      "num_input_tokens_seen": 220133395,
      "step": 10222,
      "time_per_iteration": 2.57665753364563
    },
    {
      "auxiliary_loss_clip": 0.01086133,
      "auxiliary_loss_mlp": 0.01037146,
      "balance_loss_clip": 1.03818393,
      "balance_loss_mlp": 1.02349615,
      "epoch": 0.6146400120246506,
      "flos": 20777016683520.0,
      "grad_norm": 1.9537653220063085,
      "language_loss": 0.76319802,
      "learning_rate": 1.365617422821788e-06,
      "loss": 0.78443074,
      "num_input_tokens_seen": 220152790,
      "step": 10223,
      "time_per_iteration": 2.5164098739624023
    },
    {
      "auxiliary_loss_clip": 0.01086211,
      "auxiliary_loss_mlp": 0.01030261,
      "balance_loss_clip": 1.03788996,
      "balance_loss_mlp": 1.01796508,
      "epoch": 0.6147001352773185,
      "flos": 13881306384000.0,
      "grad_norm": 2.493494849162463,
      "language_loss": 0.78241336,
      "learning_rate": 1.3652480843564535e-06,
      "loss": 0.80357814,
      "num_input_tokens_seen": 220169535,
      "step": 10224,
      "time_per_iteration": 2.4829282760620117
    },
    {
      "auxiliary_loss_clip": 0.01072507,
      "auxiliary_loss_mlp": 0.0102725,
      "balance_loss_clip": 1.0352149,
      "balance_loss_mlp": 1.0163064,
      "epoch": 0.6147602585299865,
      "flos": 56641791807360.0,
      "grad_norm": 1.1559311676050361,
      "language_loss": 0.66325742,
      "learning_rate": 1.3648787699609746e-06,
      "loss": 0.684255,
      "num_input_tokens_seen": 220195305,
      "step": 10225,
      "time_per_iteration": 2.8859102725982666
    },
    {
      "auxiliary_loss_clip": 0.01099133,
      "auxiliary_loss_mlp": 0.00786981,
      "balance_loss_clip": 1.04110968,
      "balance_loss_mlp": 1.01052558,
      "epoch": 0.6148203817826544,
      "flos": 32817217605120.0,
      "grad_norm": 2.496833825111756,
      "language_loss": 0.63203716,
      "learning_rate": 1.364509479649357e-06,
      "loss": 0.65089834,
      "num_input_tokens_seen": 220215040,
      "step": 10226,
      "time_per_iteration": 2.6053338050842285
    },
    {
      "auxiliary_loss_clip": 0.01084718,
      "auxiliary_loss_mlp": 0.01033174,
      "balance_loss_clip": 1.03708267,
      "balance_loss_mlp": 1.01979876,
      "epoch": 0.6148805050353224,
      "flos": 18332038748160.0,
      "grad_norm": 1.8368155433849405,
      "language_loss": 0.75898898,
      "learning_rate": 1.3641402134356037e-06,
      "loss": 0.78016782,
      "num_input_tokens_seen": 220234205,
      "step": 10227,
      "time_per_iteration": 2.554015874862671
    },
    {
      "auxiliary_loss_clip": 0.0105252,
      "auxiliary_loss_mlp": 0.01036876,
      "balance_loss_clip": 1.03829098,
      "balance_loss_mlp": 1.02161133,
      "epoch": 0.6149406282879903,
      "flos": 14063983977600.0,
      "grad_norm": 2.004585210787863,
      "language_loss": 0.61337602,
      "learning_rate": 1.3637709713337164e-06,
      "loss": 0.63426995,
      "num_input_tokens_seen": 220252730,
      "step": 10228,
      "time_per_iteration": 3.9705421924591064
    },
    {
      "auxiliary_loss_clip": 0.01086942,
      "auxiliary_loss_mlp": 0.01032551,
      "balance_loss_clip": 1.03793836,
      "balance_loss_mlp": 1.02023041,
      "epoch": 0.6150007515406584,
      "flos": 25190186400000.0,
      "grad_norm": 1.3152393509810452,
      "language_loss": 0.74257982,
      "learning_rate": 1.3634017533576985e-06,
      "loss": 0.76377475,
      "num_input_tokens_seen": 220273345,
      "step": 10229,
      "time_per_iteration": 2.5854804515838623
    },
    {
      "auxiliary_loss_clip": 0.01110701,
      "auxiliary_loss_mlp": 0.01036374,
      "balance_loss_clip": 1.03990602,
      "balance_loss_mlp": 1.02388668,
      "epoch": 0.6150608747933263,
      "flos": 21945262625280.0,
      "grad_norm": 3.324939528301568,
      "language_loss": 0.77909106,
      "learning_rate": 1.3630325595215493e-06,
      "loss": 0.80056179,
      "num_input_tokens_seen": 220293845,
      "step": 10230,
      "time_per_iteration": 2.4635536670684814
    },
    {
      "auxiliary_loss_clip": 0.0108274,
      "auxiliary_loss_mlp": 0.01029674,
      "balance_loss_clip": 1.03993201,
      "balance_loss_mlp": 1.01746118,
      "epoch": 0.6151209980459943,
      "flos": 30117453523200.0,
      "grad_norm": 1.6797997199124084,
      "language_loss": 0.73174465,
      "learning_rate": 1.36266338983927e-06,
      "loss": 0.75286883,
      "num_input_tokens_seen": 220316070,
      "step": 10231,
      "time_per_iteration": 2.595430850982666
    },
    {
      "auxiliary_loss_clip": 0.01089274,
      "auxiliary_loss_mlp": 0.01031708,
      "balance_loss_clip": 1.03798175,
      "balance_loss_mlp": 1.02025807,
      "epoch": 0.6151811212986622,
      "flos": 30008356940160.0,
      "grad_norm": 1.8893920630954821,
      "language_loss": 0.69852841,
      "learning_rate": 1.362294244324858e-06,
      "loss": 0.71973825,
      "num_input_tokens_seen": 220335695,
      "step": 10232,
      "time_per_iteration": 2.585747241973877
    },
    {
      "auxiliary_loss_clip": 0.01094224,
      "auxiliary_loss_mlp": 0.00784915,
      "balance_loss_clip": 1.03705537,
      "balance_loss_mlp": 1.00893879,
      "epoch": 0.6152412445513302,
      "flos": 18872888808960.0,
      "grad_norm": 1.9680161414636268,
      "language_loss": 0.91989201,
      "learning_rate": 1.3619251229923126e-06,
      "loss": 0.93868339,
      "num_input_tokens_seen": 220353720,
      "step": 10233,
      "time_per_iteration": 2.478266477584839
    },
    {
      "auxiliary_loss_clip": 0.01079566,
      "auxiliary_loss_mlp": 0.01034115,
      "balance_loss_clip": 1.04080832,
      "balance_loss_mlp": 1.02266526,
      "epoch": 0.6153013678039982,
      "flos": 25703601448320.0,
      "grad_norm": 1.7495715166050214,
      "language_loss": 0.7149142,
      "learning_rate": 1.3615560258556306e-06,
      "loss": 0.73605096,
      "num_input_tokens_seen": 220372515,
      "step": 10234,
      "time_per_iteration": 2.558286666870117
    },
    {
      "auxiliary_loss_clip": 0.01097236,
      "auxiliary_loss_mlp": 0.00786267,
      "balance_loss_clip": 1.03533542,
      "balance_loss_mlp": 1.01059198,
      "epoch": 0.6153614910566662,
      "flos": 28510271383680.0,
      "grad_norm": 2.200766251748644,
      "language_loss": 0.66462857,
      "learning_rate": 1.3611869529288077e-06,
      "loss": 0.68346357,
      "num_input_tokens_seen": 220393490,
      "step": 10235,
      "time_per_iteration": 2.54298734664917
    },
    {
      "auxiliary_loss_clip": 0.01102499,
      "auxiliary_loss_mlp": 0.0103039,
      "balance_loss_clip": 1.03842068,
      "balance_loss_mlp": 1.017712,
      "epoch": 0.6154216143093342,
      "flos": 23549787158400.0,
      "grad_norm": 1.739940392355627,
      "language_loss": 0.81289065,
      "learning_rate": 1.3608179042258398e-06,
      "loss": 0.83421957,
      "num_input_tokens_seen": 220412855,
      "step": 10236,
      "time_per_iteration": 2.5148465633392334
    },
    {
      "auxiliary_loss_clip": 0.01111338,
      "auxiliary_loss_mlp": 0.01029271,
      "balance_loss_clip": 1.03766549,
      "balance_loss_mlp": 1.01661694,
      "epoch": 0.6154817375620021,
      "flos": 22748081552640.0,
      "grad_norm": 1.4262432839947412,
      "language_loss": 0.80673063,
      "learning_rate": 1.360448879760721e-06,
      "loss": 0.82813668,
      "num_input_tokens_seen": 220433440,
      "step": 10237,
      "time_per_iteration": 3.8444180488586426
    },
    {
      "auxiliary_loss_clip": 0.01094292,
      "auxiliary_loss_mlp": 0.01040143,
      "balance_loss_clip": 1.03935122,
      "balance_loss_mlp": 1.0276264,
      "epoch": 0.6155418608146701,
      "flos": 27162975121920.0,
      "grad_norm": 1.6382636584281152,
      "language_loss": 0.76049191,
      "learning_rate": 1.3600798795474449e-06,
      "loss": 0.78183627,
      "num_input_tokens_seen": 220453445,
      "step": 10238,
      "time_per_iteration": 3.959036350250244
    },
    {
      "auxiliary_loss_clip": 0.01009162,
      "auxiliary_loss_mlp": 0.01013501,
      "balance_loss_clip": 1.03454757,
      "balance_loss_mlp": 1.01179624,
      "epoch": 0.615601984067338,
      "flos": 68811165014400.0,
      "grad_norm": 0.7688438613844096,
      "language_loss": 0.57610494,
      "learning_rate": 1.3597109036000036e-06,
      "loss": 0.59633154,
      "num_input_tokens_seen": 220509730,
      "step": 10239,
      "time_per_iteration": 3.2117981910705566
    },
    {
      "auxiliary_loss_clip": 0.0108801,
      "auxiliary_loss_mlp": 0.01033565,
      "balance_loss_clip": 1.03454494,
      "balance_loss_mlp": 1.02047002,
      "epoch": 0.615662107320006,
      "flos": 15517144598400.0,
      "grad_norm": 1.8078174938220897,
      "language_loss": 0.77351052,
      "learning_rate": 1.3593419519323892e-06,
      "loss": 0.79472625,
      "num_input_tokens_seen": 220527295,
      "step": 10240,
      "time_per_iteration": 2.51479434967041
    },
    {
      "auxiliary_loss_clip": 0.01110679,
      "auxiliary_loss_mlp": 0.01030448,
      "balance_loss_clip": 1.03921294,
      "balance_loss_mlp": 1.0178833,
      "epoch": 0.615722230572674,
      "flos": 21063691128960.0,
      "grad_norm": 2.3083528019141166,
      "language_loss": 0.72301877,
      "learning_rate": 1.3589730245585922e-06,
      "loss": 0.74443007,
      "num_input_tokens_seen": 220542730,
      "step": 10241,
      "time_per_iteration": 3.8369054794311523
    },
    {
      "auxiliary_loss_clip": 0.01105561,
      "auxiliary_loss_mlp": 0.01025442,
      "balance_loss_clip": 1.03683424,
      "balance_loss_mlp": 1.0138911,
      "epoch": 0.615782353825342,
      "flos": 23256791919360.0,
      "grad_norm": 14.319940167568046,
      "language_loss": 0.71880233,
      "learning_rate": 1.3586041214926018e-06,
      "loss": 0.74011242,
      "num_input_tokens_seen": 220562995,
      "step": 10242,
      "time_per_iteration": 2.481170177459717
    },
    {
      "auxiliary_loss_clip": 0.01098335,
      "auxiliary_loss_mlp": 0.01028903,
      "balance_loss_clip": 1.0377363,
      "balance_loss_mlp": 1.01713133,
      "epoch": 0.6158424770780099,
      "flos": 21103911383040.0,
      "grad_norm": 2.170928545685879,
      "language_loss": 0.72744012,
      "learning_rate": 1.3582352427484086e-06,
      "loss": 0.74871254,
      "num_input_tokens_seen": 220581775,
      "step": 10243,
      "time_per_iteration": 2.475595235824585
    },
    {
      "auxiliary_loss_clip": 0.01034927,
      "auxiliary_loss_mlp": 0.01003229,
      "balance_loss_clip": 1.02173257,
      "balance_loss_mlp": 1.0019654,
      "epoch": 0.6159026003306779,
      "flos": 70333276769280.0,
      "grad_norm": 0.7592068581646891,
      "language_loss": 0.56892526,
      "learning_rate": 1.3578663883399984e-06,
      "loss": 0.58930683,
      "num_input_tokens_seen": 220646395,
      "step": 10244,
      "time_per_iteration": 3.1413724422454834
    },
    {
      "auxiliary_loss_clip": 0.01107209,
      "auxiliary_loss_mlp": 0.01029501,
      "balance_loss_clip": 1.03656387,
      "balance_loss_mlp": 1.01681161,
      "epoch": 0.6159627235833458,
      "flos": 33874355802240.0,
      "grad_norm": 1.5500352338073242,
      "language_loss": 0.63436699,
      "learning_rate": 1.3574975582813593e-06,
      "loss": 0.65573406,
      "num_input_tokens_seen": 220668335,
      "step": 10245,
      "time_per_iteration": 2.563464879989624
    },
    {
      "auxiliary_loss_clip": 0.0105564,
      "auxiliary_loss_mlp": 0.01032317,
      "balance_loss_clip": 1.0324384,
      "balance_loss_mlp": 1.02016997,
      "epoch": 0.6160228468360138,
      "flos": 26575440359040.0,
      "grad_norm": 1.7476371647555244,
      "language_loss": 0.79063284,
      "learning_rate": 1.3571287525864771e-06,
      "loss": 0.81151241,
      "num_input_tokens_seen": 220688915,
      "step": 10246,
      "time_per_iteration": 2.608859062194824
    },
    {
      "auxiliary_loss_clip": 0.01077149,
      "auxiliary_loss_mlp": 0.00788216,
      "balance_loss_clip": 1.04020476,
      "balance_loss_mlp": 1.01119256,
      "epoch": 0.6160829700886818,
      "flos": 17193274894080.0,
      "grad_norm": 2.947957169372509,
      "language_loss": 0.87565172,
      "learning_rate": 1.3567599712693368e-06,
      "loss": 0.89430535,
      "num_input_tokens_seen": 220703465,
      "step": 10247,
      "time_per_iteration": 2.5552761554718018
    },
    {
      "auxiliary_loss_clip": 0.01043612,
      "auxiliary_loss_mlp": 0.01036031,
      "balance_loss_clip": 1.03960216,
      "balance_loss_mlp": 1.02254796,
      "epoch": 0.6161430933413498,
      "flos": 23623547736960.0,
      "grad_norm": 1.5998113456242489,
      "language_loss": 0.79863495,
      "learning_rate": 1.3563912143439235e-06,
      "loss": 0.81943142,
      "num_input_tokens_seen": 220722090,
      "step": 10248,
      "time_per_iteration": 2.6468286514282227
    },
    {
      "auxiliary_loss_clip": 0.01063432,
      "auxiliary_loss_mlp": 0.01032661,
      "balance_loss_clip": 1.03630614,
      "balance_loss_mlp": 1.02045357,
      "epoch": 0.6162032165940178,
      "flos": 23002436736000.0,
      "grad_norm": 1.799877011203152,
      "language_loss": 0.86936712,
      "learning_rate": 1.3560224818242191e-06,
      "loss": 0.89032805,
      "num_input_tokens_seen": 220741075,
      "step": 10249,
      "time_per_iteration": 2.5907704830169678
    },
    {
      "auxiliary_loss_clip": 0.01107475,
      "auxiliary_loss_mlp": 0.01025248,
      "balance_loss_clip": 1.03703952,
      "balance_loss_mlp": 1.01244521,
      "epoch": 0.6162633398466857,
      "flos": 39421979740800.0,
      "grad_norm": 2.273218177838173,
      "language_loss": 0.69177878,
      "learning_rate": 1.3556537737242072e-06,
      "loss": 0.71310604,
      "num_input_tokens_seen": 220763395,
      "step": 10250,
      "time_per_iteration": 2.6226603984832764
    },
    {
      "auxiliary_loss_clip": 0.0107604,
      "auxiliary_loss_mlp": 0.01026132,
      "balance_loss_clip": 1.03333259,
      "balance_loss_mlp": 1.01482522,
      "epoch": 0.6163234630993537,
      "flos": 19244672530560.0,
      "grad_norm": 2.267348926726471,
      "language_loss": 0.74363124,
      "learning_rate": 1.3552850900578692e-06,
      "loss": 0.76465297,
      "num_input_tokens_seen": 220780640,
      "step": 10251,
      "time_per_iteration": 2.509047031402588
    },
    {
      "auxiliary_loss_clip": 0.01091166,
      "auxiliary_loss_mlp": 0.01031399,
      "balance_loss_clip": 1.03396702,
      "balance_loss_mlp": 1.01780295,
      "epoch": 0.6163835863520216,
      "flos": 15961791058560.0,
      "grad_norm": 2.0825681110960406,
      "language_loss": 0.68289703,
      "learning_rate": 1.3549164308391844e-06,
      "loss": 0.70412266,
      "num_input_tokens_seen": 220797960,
      "step": 10252,
      "time_per_iteration": 2.4604580402374268
    },
    {
      "auxiliary_loss_clip": 0.00989968,
      "auxiliary_loss_mlp": 0.01004572,
      "balance_loss_clip": 1.03122044,
      "balance_loss_mlp": 1.00326717,
      "epoch": 0.6164437096046896,
      "flos": 68103834393600.0,
      "grad_norm": 0.9136717854383188,
      "language_loss": 0.57880998,
      "learning_rate": 1.3545477960821333e-06,
      "loss": 0.59875536,
      "num_input_tokens_seen": 220856930,
      "step": 10253,
      "time_per_iteration": 3.4326703548431396
    },
    {
      "auxiliary_loss_clip": 0.01086396,
      "auxiliary_loss_mlp": 0.01030106,
      "balance_loss_clip": 1.03560495,
      "balance_loss_mlp": 1.01773763,
      "epoch": 0.6165038328573575,
      "flos": 21361211481600.0,
      "grad_norm": 1.461358401726161,
      "language_loss": 0.79552978,
      "learning_rate": 1.3541791858006946e-06,
      "loss": 0.81669474,
      "num_input_tokens_seen": 220877595,
      "step": 10254,
      "time_per_iteration": 2.8940834999084473
    },
    {
      "auxiliary_loss_clip": 0.01085464,
      "auxiliary_loss_mlp": 0.01030813,
      "balance_loss_clip": 1.0382266,
      "balance_loss_mlp": 1.0187844,
      "epoch": 0.6165639561100256,
      "flos": 21101972048640.0,
      "grad_norm": 1.689340356196069,
      "language_loss": 0.80574417,
      "learning_rate": 1.353810600008846e-06,
      "loss": 0.82690692,
      "num_input_tokens_seen": 220896880,
      "step": 10255,
      "time_per_iteration": 2.551485061645508
    },
    {
      "auxiliary_loss_clip": 0.01088236,
      "auxiliary_loss_mlp": 0.010319,
      "balance_loss_clip": 1.03644359,
      "balance_loss_mlp": 1.01964569,
      "epoch": 0.6166240793626935,
      "flos": 25338533569920.0,
      "grad_norm": 2.0296515596003637,
      "language_loss": 0.65380895,
      "learning_rate": 1.3534420387205646e-06,
      "loss": 0.67501032,
      "num_input_tokens_seen": 220916425,
      "step": 10256,
      "time_per_iteration": 2.558652877807617
    },
    {
      "auxiliary_loss_clip": 0.01094125,
      "auxiliary_loss_mlp": 0.01026196,
      "balance_loss_clip": 1.03669965,
      "balance_loss_mlp": 1.01522863,
      "epoch": 0.6166842026153615,
      "flos": 19682639061120.0,
      "grad_norm": 1.5967727580458746,
      "language_loss": 0.72283745,
      "learning_rate": 1.353073501949825e-06,
      "loss": 0.74404073,
      "num_input_tokens_seen": 220935050,
      "step": 10257,
      "time_per_iteration": 2.499439001083374
    },
    {
      "auxiliary_loss_clip": 0.01086563,
      "auxiliary_loss_mlp": 0.01030204,
      "balance_loss_clip": 1.03654647,
      "balance_loss_mlp": 1.01778221,
      "epoch": 0.6167443258680294,
      "flos": 19318361281920.0,
      "grad_norm": 1.6120629274208997,
      "language_loss": 0.71972466,
      "learning_rate": 1.3527049897106034e-06,
      "loss": 0.74089229,
      "num_input_tokens_seen": 220953085,
      "step": 10258,
      "time_per_iteration": 2.510300874710083
    },
    {
      "auxiliary_loss_clip": 0.0108065,
      "auxiliary_loss_mlp": 0.01037214,
      "balance_loss_clip": 1.03400958,
      "balance_loss_mlp": 1.02370739,
      "epoch": 0.6168044491206974,
      "flos": 25265239868160.0,
      "grad_norm": 3.1279935596064,
      "language_loss": 0.63617224,
      "learning_rate": 1.3523365020168735e-06,
      "loss": 0.6573509,
      "num_input_tokens_seen": 220969050,
      "step": 10259,
      "time_per_iteration": 2.5416276454925537
    },
    {
      "auxiliary_loss_clip": 0.01071367,
      "auxiliary_loss_mlp": 0.01029676,
      "balance_loss_clip": 1.03679192,
      "balance_loss_mlp": 1.01696277,
      "epoch": 0.6168645723733654,
      "flos": 13219903301760.0,
      "grad_norm": 1.9535502989265747,
      "language_loss": 0.71337712,
      "learning_rate": 1.3519680388826084e-06,
      "loss": 0.73438758,
      "num_input_tokens_seen": 220985825,
      "step": 10260,
      "time_per_iteration": 2.5156314373016357
    },
    {
      "auxiliary_loss_clip": 0.01103344,
      "auxiliary_loss_mlp": 0.01034953,
      "balance_loss_clip": 1.04115415,
      "balance_loss_mlp": 1.02125573,
      "epoch": 0.6169246956260334,
      "flos": 26652038112000.0,
      "grad_norm": 2.038283578406827,
      "language_loss": 0.68817651,
      "learning_rate": 1.3515996003217803e-06,
      "loss": 0.7095595,
      "num_input_tokens_seen": 221004465,
      "step": 10261,
      "time_per_iteration": 2.5381405353546143
    },
    {
      "auxiliary_loss_clip": 0.01070509,
      "auxiliary_loss_mlp": 0.01039098,
      "balance_loss_clip": 1.03544199,
      "balance_loss_mlp": 1.027946,
      "epoch": 0.6169848188787014,
      "flos": 23148413608320.0,
      "grad_norm": 1.8179393463184357,
      "language_loss": 0.71136773,
      "learning_rate": 1.3512311863483602e-06,
      "loss": 0.73246378,
      "num_input_tokens_seen": 221023260,
      "step": 10262,
      "time_per_iteration": 2.5414698123931885
    },
    {
      "auxiliary_loss_clip": 0.01087335,
      "auxiliary_loss_mlp": 0.01033536,
      "balance_loss_clip": 1.03633368,
      "balance_loss_mlp": 1.0212512,
      "epoch": 0.6170449421313693,
      "flos": 23331917214720.0,
      "grad_norm": 1.8264791575709374,
      "language_loss": 0.69918311,
      "learning_rate": 1.3508627969763188e-06,
      "loss": 0.72039181,
      "num_input_tokens_seen": 221043090,
      "step": 10263,
      "time_per_iteration": 2.5504252910614014
    },
    {
      "auxiliary_loss_clip": 0.01047187,
      "auxiliary_loss_mlp": 0.01028447,
      "balance_loss_clip": 1.03797102,
      "balance_loss_mlp": 1.01612651,
      "epoch": 0.6171050653840373,
      "flos": 15851617067520.0,
      "grad_norm": 3.32835224168281,
      "language_loss": 0.76334941,
      "learning_rate": 1.3504944322196244e-06,
      "loss": 0.78410578,
      "num_input_tokens_seen": 221061435,
      "step": 10264,
      "time_per_iteration": 2.5779831409454346
    },
    {
      "auxiliary_loss_clip": 0.01106245,
      "auxiliary_loss_mlp": 0.01029297,
      "balance_loss_clip": 1.03624678,
      "balance_loss_mlp": 1.01686394,
      "epoch": 0.6171651886367052,
      "flos": 20045516209920.0,
      "grad_norm": 5.066960682632096,
      "language_loss": 0.85550296,
      "learning_rate": 1.350126092092247e-06,
      "loss": 0.87685847,
      "num_input_tokens_seen": 221078705,
      "step": 10265,
      "time_per_iteration": 2.480560541152954
    },
    {
      "auxiliary_loss_clip": 0.01053405,
      "auxiliary_loss_mlp": 0.01030875,
      "balance_loss_clip": 1.03426051,
      "balance_loss_mlp": 1.01853704,
      "epoch": 0.6172253118893732,
      "flos": 26432695710720.0,
      "grad_norm": 1.9867782983933089,
      "language_loss": 0.64988726,
      "learning_rate": 1.349757776608153e-06,
      "loss": 0.67072999,
      "num_input_tokens_seen": 221099245,
      "step": 10266,
      "time_per_iteration": 4.344377756118774
    },
    {
      "auxiliary_loss_clip": 0.01069706,
      "auxiliary_loss_mlp": 0.01028356,
      "balance_loss_clip": 1.0333643,
      "balance_loss_mlp": 1.01644087,
      "epoch": 0.6172854351420412,
      "flos": 22632879657600.0,
      "grad_norm": 1.6163246522770094,
      "language_loss": 0.75515342,
      "learning_rate": 1.3493894857813094e-06,
      "loss": 0.77613401,
      "num_input_tokens_seen": 221116930,
      "step": 10267,
      "time_per_iteration": 2.554121971130371
    },
    {
      "auxiliary_loss_clip": 0.01078712,
      "auxiliary_loss_mlp": 0.01026404,
      "balance_loss_clip": 1.03581953,
      "balance_loss_mlp": 1.01338005,
      "epoch": 0.6173455583947092,
      "flos": 21212936138880.0,
      "grad_norm": 1.7166279770541952,
      "language_loss": 0.75002134,
      "learning_rate": 1.3490212196256818e-06,
      "loss": 0.77107257,
      "num_input_tokens_seen": 221137660,
      "step": 10268,
      "time_per_iteration": 2.587272882461548
    },
    {
      "auxiliary_loss_clip": 0.01087673,
      "auxiliary_loss_mlp": 0.01027021,
      "balance_loss_clip": 1.03616226,
      "balance_loss_mlp": 1.01489711,
      "epoch": 0.6174056816473771,
      "flos": 19500284689920.0,
      "grad_norm": 1.6121790360367356,
      "language_loss": 0.75653851,
      "learning_rate": 1.3486529781552342e-06,
      "loss": 0.77768546,
      "num_input_tokens_seen": 221156225,
      "step": 10269,
      "time_per_iteration": 2.5345916748046875
    },
    {
      "auxiliary_loss_clip": 0.01104588,
      "auxiliary_loss_mlp": 0.01027478,
      "balance_loss_clip": 1.03485847,
      "balance_loss_mlp": 1.01605737,
      "epoch": 0.6174658049000451,
      "flos": 15997342544640.0,
      "grad_norm": 2.254513479315107,
      "language_loss": 0.77144992,
      "learning_rate": 1.3482847613839318e-06,
      "loss": 0.79277062,
      "num_input_tokens_seen": 221173820,
      "step": 10270,
      "time_per_iteration": 2.437910318374634
    },
    {
      "auxiliary_loss_clip": 0.01086445,
      "auxiliary_loss_mlp": 0.01025662,
      "balance_loss_clip": 1.03582597,
      "balance_loss_mlp": 1.01327586,
      "epoch": 0.617525928152713,
      "flos": 21903893136000.0,
      "grad_norm": 1.7875085568229236,
      "language_loss": 0.8252691,
      "learning_rate": 1.347916569325736e-06,
      "loss": 0.84639013,
      "num_input_tokens_seen": 221191815,
      "step": 10271,
      "time_per_iteration": 2.5061938762664795
    },
    {
      "auxiliary_loss_clip": 0.0110873,
      "auxiliary_loss_mlp": 0.00784467,
      "balance_loss_clip": 1.03745782,
      "balance_loss_mlp": 1.0106523,
      "epoch": 0.617586051405381,
      "flos": 21105958458240.0,
      "grad_norm": 1.555063770090248,
      "language_loss": 0.77043736,
      "learning_rate": 1.3475484019946093e-06,
      "loss": 0.78936929,
      "num_input_tokens_seen": 221211205,
      "step": 10272,
      "time_per_iteration": 2.481411933898926
    },
    {
      "auxiliary_loss_clip": 0.01016631,
      "auxiliary_loss_mlp": 0.01001503,
      "balance_loss_clip": 1.0223248,
      "balance_loss_mlp": 1.00011384,
      "epoch": 0.617646174658049,
      "flos": 58610776665600.0,
      "grad_norm": 0.8011639776995017,
      "language_loss": 0.59090126,
      "learning_rate": 1.347180259404513e-06,
      "loss": 0.61108255,
      "num_input_tokens_seen": 221268430,
      "step": 10273,
      "time_per_iteration": 3.0405077934265137
    },
    {
      "auxiliary_loss_clip": 0.01078612,
      "auxiliary_loss_mlp": 0.01036429,
      "balance_loss_clip": 1.03396821,
      "balance_loss_mlp": 1.02209997,
      "epoch": 0.617706297910717,
      "flos": 13878684691200.0,
      "grad_norm": 2.4027133598899586,
      "language_loss": 0.72935045,
      "learning_rate": 1.3468121415694059e-06,
      "loss": 0.75050086,
      "num_input_tokens_seen": 221281930,
      "step": 10274,
      "time_per_iteration": 2.483344316482544
    },
    {
      "auxiliary_loss_clip": 0.01096752,
      "auxiliary_loss_mlp": 0.00785104,
      "balance_loss_clip": 1.036479,
      "balance_loss_mlp": 1.01038933,
      "epoch": 0.617766421163385,
      "flos": 19208438686080.0,
      "grad_norm": 1.8340669538607355,
      "language_loss": 0.77427703,
      "learning_rate": 1.3464440485032484e-06,
      "loss": 0.79309559,
      "num_input_tokens_seen": 221301605,
      "step": 10275,
      "time_per_iteration": 2.4744439125061035
    },
    {
      "auxiliary_loss_clip": 0.01066913,
      "auxiliary_loss_mlp": 0.01026109,
      "balance_loss_clip": 1.03537679,
      "balance_loss_mlp": 1.01476634,
      "epoch": 0.6178265444160529,
      "flos": 22565978576640.0,
      "grad_norm": 1.6823389299960438,
      "language_loss": 0.79322952,
      "learning_rate": 1.346075980219998e-06,
      "loss": 0.81415975,
      "num_input_tokens_seen": 221320105,
      "step": 10276,
      "time_per_iteration": 5.547476053237915
    },
    {
      "auxiliary_loss_clip": 0.01053461,
      "auxiliary_loss_mlp": 0.01036649,
      "balance_loss_clip": 1.03901362,
      "balance_loss_mlp": 1.02347684,
      "epoch": 0.6178866676687209,
      "flos": 11984289402240.0,
      "grad_norm": 1.9270541958919831,
      "language_loss": 0.81211567,
      "learning_rate": 1.345707936733612e-06,
      "loss": 0.83301681,
      "num_input_tokens_seen": 221335915,
      "step": 10277,
      "time_per_iteration": 2.5963456630706787
    },
    {
      "auxiliary_loss_clip": 0.01079933,
      "auxiliary_loss_mlp": 0.01026735,
      "balance_loss_clip": 1.03674221,
      "balance_loss_mlp": 1.01346087,
      "epoch": 0.6179467909213888,
      "flos": 20991510748800.0,
      "grad_norm": 1.7193063455956077,
      "language_loss": 0.81607622,
      "learning_rate": 1.3453399180580466e-06,
      "loss": 0.83714288,
      "num_input_tokens_seen": 221353965,
      "step": 10278,
      "time_per_iteration": 2.5844919681549072
    },
    {
      "auxiliary_loss_clip": 0.01062582,
      "auxiliary_loss_mlp": 0.00785535,
      "balance_loss_clip": 1.03555202,
      "balance_loss_mlp": 1.01073933,
      "epoch": 0.6180069141740568,
      "flos": 25338102606720.0,
      "grad_norm": 1.5301185914636373,
      "language_loss": 0.74026775,
      "learning_rate": 1.3449719242072567e-06,
      "loss": 0.75874889,
      "num_input_tokens_seen": 221374080,
      "step": 10279,
      "time_per_iteration": 4.073084831237793
    },
    {
      "auxiliary_loss_clip": 0.01090273,
      "auxiliary_loss_mlp": 0.01028244,
      "balance_loss_clip": 1.03362036,
      "balance_loss_mlp": 1.01634109,
      "epoch": 0.6180670374267248,
      "flos": 19645722858240.0,
      "grad_norm": 1.488923194662771,
      "language_loss": 0.70864463,
      "learning_rate": 1.3446039551951975e-06,
      "loss": 0.72982979,
      "num_input_tokens_seen": 221392910,
      "step": 10280,
      "time_per_iteration": 2.514727830886841
    },
    {
      "auxiliary_loss_clip": 0.01108591,
      "auxiliary_loss_mlp": 0.01031955,
      "balance_loss_clip": 1.03755963,
      "balance_loss_mlp": 1.01996887,
      "epoch": 0.6181271606793928,
      "flos": 19464876858240.0,
      "grad_norm": 1.6093896808902897,
      "language_loss": 0.72689259,
      "learning_rate": 1.3442360110358215e-06,
      "loss": 0.74829811,
      "num_input_tokens_seen": 221410990,
      "step": 10281,
      "time_per_iteration": 2.4567606449127197
    },
    {
      "auxiliary_loss_clip": 0.01084615,
      "auxiliary_loss_mlp": 0.01027857,
      "balance_loss_clip": 1.03775978,
      "balance_loss_mlp": 1.01691985,
      "epoch": 0.6181872839320607,
      "flos": 25594289383680.0,
      "grad_norm": 1.6278077466563692,
      "language_loss": 0.76938921,
      "learning_rate": 1.3438680917430827e-06,
      "loss": 0.79051393,
      "num_input_tokens_seen": 221431020,
      "step": 10282,
      "time_per_iteration": 2.5994722843170166
    },
    {
      "auxiliary_loss_clip": 0.01082563,
      "auxiliary_loss_mlp": 0.01034803,
      "balance_loss_clip": 1.03503931,
      "balance_loss_mlp": 1.01906705,
      "epoch": 0.6182474071847287,
      "flos": 25551806572800.0,
      "grad_norm": 1.6542664626260082,
      "language_loss": 0.69322509,
      "learning_rate": 1.343500197330931e-06,
      "loss": 0.71439874,
      "num_input_tokens_seen": 221453235,
      "step": 10283,
      "time_per_iteration": 2.5605452060699463
    },
    {
      "auxiliary_loss_clip": 0.01102248,
      "auxiliary_loss_mlp": 0.01026941,
      "balance_loss_clip": 1.03720725,
      "balance_loss_mlp": 1.01413751,
      "epoch": 0.6183075304373966,
      "flos": 22123738327680.0,
      "grad_norm": 1.6156015153088243,
      "language_loss": 0.75176501,
      "learning_rate": 1.3431323278133176e-06,
      "loss": 0.77305686,
      "num_input_tokens_seen": 221472560,
      "step": 10284,
      "time_per_iteration": 2.5085856914520264
    },
    {
      "auxiliary_loss_clip": 0.01091694,
      "auxiliary_loss_mlp": 0.01030913,
      "balance_loss_clip": 1.04089141,
      "balance_loss_mlp": 1.0191772,
      "epoch": 0.6183676536900646,
      "flos": 22455589104000.0,
      "grad_norm": 1.4856857004221031,
      "language_loss": 0.75640005,
      "learning_rate": 1.3427644832041922e-06,
      "loss": 0.77762616,
      "num_input_tokens_seen": 221492835,
      "step": 10285,
      "time_per_iteration": 2.509333372116089
    },
    {
      "auxiliary_loss_clip": 0.01073046,
      "auxiliary_loss_mlp": 0.01031708,
      "balance_loss_clip": 1.03572083,
      "balance_loss_mlp": 1.01970983,
      "epoch": 0.6184277769427327,
      "flos": 23364128736000.0,
      "grad_norm": 1.5022663645266483,
      "language_loss": 0.72952276,
      "learning_rate": 1.342396663517503e-06,
      "loss": 0.7505703,
      "num_input_tokens_seen": 221511870,
      "step": 10286,
      "time_per_iteration": 2.5647313594818115
    },
    {
      "auxiliary_loss_clip": 0.01105529,
      "auxiliary_loss_mlp": 0.0102884,
      "balance_loss_clip": 1.03666496,
      "balance_loss_mlp": 1.01718175,
      "epoch": 0.6184879001954006,
      "flos": 22711057608960.0,
      "grad_norm": 2.1744758048938744,
      "language_loss": 0.75755203,
      "learning_rate": 1.342028868767199e-06,
      "loss": 0.77889574,
      "num_input_tokens_seen": 221529915,
      "step": 10287,
      "time_per_iteration": 2.453536033630371
    },
    {
      "auxiliary_loss_clip": 0.01069371,
      "auxiliary_loss_mlp": 0.01034871,
      "balance_loss_clip": 1.03658044,
      "balance_loss_mlp": 1.02319479,
      "epoch": 0.6185480234480686,
      "flos": 23841920471040.0,
      "grad_norm": 1.7165971344134126,
      "language_loss": 0.73420835,
      "learning_rate": 1.3416610989672262e-06,
      "loss": 0.75525069,
      "num_input_tokens_seen": 221549745,
      "step": 10288,
      "time_per_iteration": 2.5722339153289795
    },
    {
      "auxiliary_loss_clip": 0.01091018,
      "auxiliary_loss_mlp": 0.01029263,
      "balance_loss_clip": 1.03441691,
      "balance_loss_mlp": 1.01799178,
      "epoch": 0.6186081467007365,
      "flos": 45477595774080.0,
      "grad_norm": 1.5838363096575276,
      "language_loss": 0.72793931,
      "learning_rate": 1.3412933541315296e-06,
      "loss": 0.74914211,
      "num_input_tokens_seen": 221572455,
      "step": 10289,
      "time_per_iteration": 2.6831247806549072
    },
    {
      "auxiliary_loss_clip": 0.01081792,
      "auxiliary_loss_mlp": 0.01032198,
      "balance_loss_clip": 1.03572559,
      "balance_loss_mlp": 1.01931763,
      "epoch": 0.6186682699534045,
      "flos": 23550864566400.0,
      "grad_norm": 1.498537633321843,
      "language_loss": 0.79339182,
      "learning_rate": 1.340925634274056e-06,
      "loss": 0.81453168,
      "num_input_tokens_seen": 221591325,
      "step": 10290,
      "time_per_iteration": 2.553004264831543
    },
    {
      "auxiliary_loss_clip": 0.01099041,
      "auxiliary_loss_mlp": 0.01033184,
      "balance_loss_clip": 1.03823638,
      "balance_loss_mlp": 1.02105451,
      "epoch": 0.6187283932060724,
      "flos": 25774201630080.0,
      "grad_norm": 1.5833518444468346,
      "language_loss": 0.81344706,
      "learning_rate": 1.3405579394087475e-06,
      "loss": 0.83476937,
      "num_input_tokens_seen": 221611640,
      "step": 10291,
      "time_per_iteration": 2.52955961227417
    },
    {
      "auxiliary_loss_clip": 0.01108247,
      "auxiliary_loss_mlp": 0.0103369,
      "balance_loss_clip": 1.03807414,
      "balance_loss_mlp": 1.02170944,
      "epoch": 0.6187885164587404,
      "flos": 25265203954560.0,
      "grad_norm": 1.692201678811394,
      "language_loss": 0.77624643,
      "learning_rate": 1.3401902695495487e-06,
      "loss": 0.79766577,
      "num_input_tokens_seen": 221631225,
      "step": 10292,
      "time_per_iteration": 2.4941632747650146
    },
    {
      "auxiliary_loss_clip": 0.01082694,
      "auxiliary_loss_mlp": 0.01040013,
      "balance_loss_clip": 1.03734505,
      "balance_loss_mlp": 1.02541006,
      "epoch": 0.6188486397114084,
      "flos": 26250772302720.0,
      "grad_norm": 1.794377095878869,
      "language_loss": 0.73138344,
      "learning_rate": 1.339822624710401e-06,
      "loss": 0.7526105,
      "num_input_tokens_seen": 221651035,
      "step": 10293,
      "time_per_iteration": 2.571413278579712
    },
    {
      "auxiliary_loss_clip": 0.01076871,
      "auxiliary_loss_mlp": 0.00785003,
      "balance_loss_clip": 1.03779554,
      "balance_loss_mlp": 1.01049256,
      "epoch": 0.6189087629640764,
      "flos": 20923388605440.0,
      "grad_norm": 1.5083177373802097,
      "language_loss": 0.82837451,
      "learning_rate": 1.3394550049052454e-06,
      "loss": 0.84699327,
      "num_input_tokens_seen": 221671300,
      "step": 10294,
      "time_per_iteration": 2.5740389823913574
    },
    {
      "auxiliary_loss_clip": 0.0108895,
      "auxiliary_loss_mlp": 0.01031108,
      "balance_loss_clip": 1.03736806,
      "balance_loss_mlp": 1.01942575,
      "epoch": 0.6189688862167443,
      "flos": 14829814874880.0,
      "grad_norm": 2.1384213592833876,
      "language_loss": 0.70723808,
      "learning_rate": 1.3390874101480225e-06,
      "loss": 0.72843874,
      "num_input_tokens_seen": 221687320,
      "step": 10295,
      "time_per_iteration": 2.515923261642456
    },
    {
      "auxiliary_loss_clip": 0.01108449,
      "auxiliary_loss_mlp": 0.01031569,
      "balance_loss_clip": 1.03921723,
      "balance_loss_mlp": 1.01958835,
      "epoch": 0.6190290094694123,
      "flos": 24285058560000.0,
      "grad_norm": 1.7178641077291836,
      "language_loss": 0.70616865,
      "learning_rate": 1.3387198404526705e-06,
      "loss": 0.72756886,
      "num_input_tokens_seen": 221710175,
      "step": 10296,
      "time_per_iteration": 2.53440260887146
    },
    {
      "auxiliary_loss_clip": 0.01075297,
      "auxiliary_loss_mlp": 0.01033247,
      "balance_loss_clip": 1.03913009,
      "balance_loss_mlp": 1.019907,
      "epoch": 0.6190891327220802,
      "flos": 22529457423360.0,
      "grad_norm": 1.7830074676652925,
      "language_loss": 0.71781969,
      "learning_rate": 1.3383522958331287e-06,
      "loss": 0.73890513,
      "num_input_tokens_seen": 221728145,
      "step": 10297,
      "time_per_iteration": 2.5861740112304688
    },
    {
      "auxiliary_loss_clip": 0.01044815,
      "auxiliary_loss_mlp": 0.0100358,
      "balance_loss_clip": 1.0209558,
      "balance_loss_mlp": 1.00233996,
      "epoch": 0.6191492559747482,
      "flos": 67729357152000.0,
      "grad_norm": 0.8816755927391389,
      "language_loss": 0.64136255,
      "learning_rate": 1.3379847763033345e-06,
      "loss": 0.66184652,
      "num_input_tokens_seen": 221786100,
      "step": 10298,
      "time_per_iteration": 2.9709692001342773
    },
    {
      "auxiliary_loss_clip": 0.01109861,
      "auxiliary_loss_mlp": 0.01031404,
      "balance_loss_clip": 1.03855181,
      "balance_loss_mlp": 1.01944184,
      "epoch": 0.6192093792274163,
      "flos": 22346672088960.0,
      "grad_norm": 1.9823745420851533,
      "language_loss": 0.73649418,
      "learning_rate": 1.3376172818772236e-06,
      "loss": 0.75790679,
      "num_input_tokens_seen": 221806450,
      "step": 10299,
      "time_per_iteration": 2.5974173545837402
    },
    {
      "auxiliary_loss_clip": 0.01101892,
      "auxiliary_loss_mlp": 0.01028699,
      "balance_loss_clip": 1.03896368,
      "balance_loss_mlp": 1.01682615,
      "epoch": 0.6192695024800842,
      "flos": 13553944807680.0,
      "grad_norm": 2.5566539974626514,
      "language_loss": 0.687989,
      "learning_rate": 1.337249812568732e-06,
      "loss": 0.70929492,
      "num_input_tokens_seen": 221823330,
      "step": 10300,
      "time_per_iteration": 2.480738401412964
    },
    {
      "auxiliary_loss_clip": 0.01098053,
      "auxiliary_loss_mlp": 0.00785341,
      "balance_loss_clip": 1.04128623,
      "balance_loss_mlp": 1.01004052,
      "epoch": 0.6193296257327522,
      "flos": 17415310815360.0,
      "grad_norm": 1.7742560864353165,
      "language_loss": 0.66922784,
      "learning_rate": 1.3368823683917939e-06,
      "loss": 0.68806183,
      "num_input_tokens_seen": 221839360,
      "step": 10301,
      "time_per_iteration": 2.4901843070983887
    },
    {
      "auxiliary_loss_clip": 0.01068234,
      "auxiliary_loss_mlp": 0.01028135,
      "balance_loss_clip": 1.0390507,
      "balance_loss_mlp": 1.01672029,
      "epoch": 0.6193897489854201,
      "flos": 31101118450560.0,
      "grad_norm": 1.529122654058849,
      "language_loss": 0.72958273,
      "learning_rate": 1.3365149493603424e-06,
      "loss": 0.7505464,
      "num_input_tokens_seen": 221859465,
      "step": 10302,
      "time_per_iteration": 2.6799395084381104
    },
    {
      "auxiliary_loss_clip": 0.01082328,
      "auxiliary_loss_mlp": 0.01027953,
      "balance_loss_clip": 1.03887951,
      "balance_loss_mlp": 1.01532888,
      "epoch": 0.6194498722380881,
      "flos": 19134031662720.0,
      "grad_norm": 1.7015760030566522,
      "language_loss": 0.80575347,
      "learning_rate": 1.3361475554883107e-06,
      "loss": 0.82685632,
      "num_input_tokens_seen": 221878555,
      "step": 10303,
      "time_per_iteration": 2.5071494579315186
    },
    {
      "auxiliary_loss_clip": 0.01111572,
      "auxiliary_loss_mlp": 0.01030429,
      "balance_loss_clip": 1.03826904,
      "balance_loss_mlp": 1.01704216,
      "epoch": 0.619509995490756,
      "flos": 21835088634240.0,
      "grad_norm": 1.6482641788859587,
      "language_loss": 0.76572907,
      "learning_rate": 1.3357801867896307e-06,
      "loss": 0.78714907,
      "num_input_tokens_seen": 221898790,
      "step": 10304,
      "time_per_iteration": 2.464911460876465
    },
    {
      "auxiliary_loss_clip": 0.01079939,
      "auxiliary_loss_mlp": 0.01031805,
      "balance_loss_clip": 1.04153776,
      "balance_loss_mlp": 1.01932335,
      "epoch": 0.619570118743424,
      "flos": 23806548552960.0,
      "grad_norm": 2.2077018144314087,
      "language_loss": 0.76596689,
      "learning_rate": 1.3354128432782324e-06,
      "loss": 0.78708434,
      "num_input_tokens_seen": 221918875,
      "step": 10305,
      "time_per_iteration": 3.916080951690674
    },
    {
      "auxiliary_loss_clip": 0.01105065,
      "auxiliary_loss_mlp": 0.01033138,
      "balance_loss_clip": 1.04022217,
      "balance_loss_mlp": 1.01931012,
      "epoch": 0.619630241996092,
      "flos": 21101612912640.0,
      "grad_norm": 1.6726846660843253,
      "language_loss": 0.7924422,
      "learning_rate": 1.335045524968045e-06,
      "loss": 0.81382418,
      "num_input_tokens_seen": 221937895,
      "step": 10306,
      "time_per_iteration": 2.530425548553467
    },
    {
      "auxiliary_loss_clip": 0.01052084,
      "auxiliary_loss_mlp": 0.01028661,
      "balance_loss_clip": 1.04066885,
      "balance_loss_mlp": 1.01737797,
      "epoch": 0.61969036524876,
      "flos": 27308269635840.0,
      "grad_norm": 1.557628871275947,
      "language_loss": 0.8012374,
      "learning_rate": 1.3346782318729988e-06,
      "loss": 0.82204485,
      "num_input_tokens_seen": 221955920,
      "step": 10307,
      "time_per_iteration": 2.6432037353515625
    },
    {
      "auxiliary_loss_clip": 0.01021573,
      "auxiliary_loss_mlp": 0.01003084,
      "balance_loss_clip": 1.02691841,
      "balance_loss_mlp": 1.00165904,
      "epoch": 0.6197504885014279,
      "flos": 51648955384320.0,
      "grad_norm": 0.8092209614688123,
      "language_loss": 0.59383929,
      "learning_rate": 1.3343109640070203e-06,
      "loss": 0.61408579,
      "num_input_tokens_seen": 222011405,
      "step": 10308,
      "time_per_iteration": 3.1812753677368164
    },
    {
      "auxiliary_loss_clip": 0.01082104,
      "auxiliary_loss_mlp": 0.0102583,
      "balance_loss_clip": 1.04003859,
      "balance_loss_mlp": 1.0154705,
      "epoch": 0.6198106117540959,
      "flos": 30557107992960.0,
      "grad_norm": 1.7162497927668046,
      "language_loss": 0.67710119,
      "learning_rate": 1.333943721384037e-06,
      "loss": 0.6981805,
      "num_input_tokens_seen": 222034545,
      "step": 10309,
      "time_per_iteration": 2.6088027954101562
    },
    {
      "auxiliary_loss_clip": 0.01079463,
      "auxiliary_loss_mlp": 0.0102943,
      "balance_loss_clip": 1.03533792,
      "balance_loss_mlp": 1.01749158,
      "epoch": 0.6198707350067638,
      "flos": 18909733184640.0,
      "grad_norm": 1.5576319839388018,
      "language_loss": 0.71857953,
      "learning_rate": 1.3335765040179746e-06,
      "loss": 0.73966843,
      "num_input_tokens_seen": 222052690,
      "step": 10310,
      "time_per_iteration": 2.5423073768615723
    },
    {
      "auxiliary_loss_clip": 0.01093323,
      "auxiliary_loss_mlp": 0.01032611,
      "balance_loss_clip": 1.04148149,
      "balance_loss_mlp": 1.01904464,
      "epoch": 0.6199308582594318,
      "flos": 21433858738560.0,
      "grad_norm": 2.4594977468660293,
      "language_loss": 0.78862923,
      "learning_rate": 1.3332093119227573e-06,
      "loss": 0.8098886,
      "num_input_tokens_seen": 222069095,
      "step": 10311,
      "time_per_iteration": 2.525202989578247
    },
    {
      "auxiliary_loss_clip": 0.01073605,
      "auxiliary_loss_mlp": 0.01031523,
      "balance_loss_clip": 1.03583932,
      "balance_loss_mlp": 1.01941156,
      "epoch": 0.6199909815120999,
      "flos": 18407379525120.0,
      "grad_norm": 1.75126879216098,
      "language_loss": 0.72338039,
      "learning_rate": 1.3328421451123105e-06,
      "loss": 0.74443167,
      "num_input_tokens_seen": 222087360,
      "step": 10312,
      "time_per_iteration": 2.559826135635376
    },
    {
      "auxiliary_loss_clip": 0.01060855,
      "auxiliary_loss_mlp": 0.01028366,
      "balance_loss_clip": 1.03760147,
      "balance_loss_mlp": 1.01633203,
      "epoch": 0.6200511047647678,
      "flos": 21466860359040.0,
      "grad_norm": 1.7678603314887196,
      "language_loss": 0.71853507,
      "learning_rate": 1.3324750036005557e-06,
      "loss": 0.73942727,
      "num_input_tokens_seen": 222106130,
      "step": 10313,
      "time_per_iteration": 2.570801019668579
    },
    {
      "auxiliary_loss_clip": 0.01103209,
      "auxiliary_loss_mlp": 0.01030186,
      "balance_loss_clip": 1.03893971,
      "balance_loss_mlp": 1.01722169,
      "epoch": 0.6201112280174358,
      "flos": 18215903099520.0,
      "grad_norm": 2.0664286698498553,
      "language_loss": 0.78303385,
      "learning_rate": 1.332107887401416e-06,
      "loss": 0.80436784,
      "num_input_tokens_seen": 222123125,
      "step": 10314,
      "time_per_iteration": 3.8403100967407227
    },
    {
      "auxiliary_loss_clip": 0.01096006,
      "auxiliary_loss_mlp": 0.01032835,
      "balance_loss_clip": 1.03559232,
      "balance_loss_mlp": 1.02052689,
      "epoch": 0.6201713512701037,
      "flos": 20011185786240.0,
      "grad_norm": 1.807697478550938,
      "language_loss": 0.78333187,
      "learning_rate": 1.331740796528812e-06,
      "loss": 0.80462027,
      "num_input_tokens_seen": 222140655,
      "step": 10315,
      "time_per_iteration": 3.851771354675293
    },
    {
      "auxiliary_loss_clip": 0.01077511,
      "auxiliary_loss_mlp": 0.01035927,
      "balance_loss_clip": 1.04183197,
      "balance_loss_mlp": 1.02366674,
      "epoch": 0.6202314745227717,
      "flos": 22487692884480.0,
      "grad_norm": 1.8403004707279857,
      "language_loss": 0.75833726,
      "learning_rate": 1.3313737309966641e-06,
      "loss": 0.77947164,
      "num_input_tokens_seen": 222160450,
      "step": 10316,
      "time_per_iteration": 2.598686933517456
    },
    {
      "auxiliary_loss_clip": 0.01109253,
      "auxiliary_loss_mlp": 0.01028945,
      "balance_loss_clip": 1.0361917,
      "balance_loss_mlp": 1.01667225,
      "epoch": 0.6202915977754396,
      "flos": 26828682220800.0,
      "grad_norm": 1.7797223190782672,
      "language_loss": 0.77644354,
      "learning_rate": 1.3310066908188915e-06,
      "loss": 0.79782557,
      "num_input_tokens_seen": 222179170,
      "step": 10317,
      "time_per_iteration": 2.5139169692993164
    },
    {
      "auxiliary_loss_clip": 0.01028359,
      "auxiliary_loss_mlp": 0.01000336,
      "balance_loss_clip": 1.02421284,
      "balance_loss_mlp": 0.99906081,
      "epoch": 0.6203517210281076,
      "flos": 62742694890240.0,
      "grad_norm": 0.6956812243488443,
      "language_loss": 0.59051585,
      "learning_rate": 1.3306396760094122e-06,
      "loss": 0.61080277,
      "num_input_tokens_seen": 222242660,
      "step": 10318,
      "time_per_iteration": 4.550368547439575
    },
    {
      "auxiliary_loss_clip": 0.01083668,
      "auxiliary_loss_mlp": 0.01034023,
      "balance_loss_clip": 1.04110122,
      "balance_loss_mlp": 1.02097535,
      "epoch": 0.6204118442807756,
      "flos": 23404277162880.0,
      "grad_norm": 2.790269011975817,
      "language_loss": 0.7773385,
      "learning_rate": 1.330272686582143e-06,
      "loss": 0.79851538,
      "num_input_tokens_seen": 222262170,
      "step": 10319,
      "time_per_iteration": 2.539809465408325
    },
    {
      "auxiliary_loss_clip": 0.01088697,
      "auxiliary_loss_mlp": 0.01028265,
      "balance_loss_clip": 1.03839445,
      "balance_loss_mlp": 1.0164572,
      "epoch": 0.6204719675334436,
      "flos": 20193647898240.0,
      "grad_norm": 1.7207014502681377,
      "language_loss": 0.66218185,
      "learning_rate": 1.3299057225510013e-06,
      "loss": 0.68335152,
      "num_input_tokens_seen": 222280375,
      "step": 10320,
      "time_per_iteration": 2.538475751876831
    },
    {
      "auxiliary_loss_clip": 0.01068927,
      "auxiliary_loss_mlp": 0.01028818,
      "balance_loss_clip": 1.03717184,
      "balance_loss_mlp": 1.01701617,
      "epoch": 0.6205320907861115,
      "flos": 13188050916480.0,
      "grad_norm": 1.711420935965839,
      "language_loss": 0.76076519,
      "learning_rate": 1.3295387839299013e-06,
      "loss": 0.78174269,
      "num_input_tokens_seen": 222297325,
      "step": 10321,
      "time_per_iteration": 2.525261640548706
    },
    {
      "auxiliary_loss_clip": 0.01081773,
      "auxiliary_loss_mlp": 0.01022758,
      "balance_loss_clip": 1.03942227,
      "balance_loss_mlp": 1.01132631,
      "epoch": 0.6205922140387795,
      "flos": 20668386977280.0,
      "grad_norm": 2.1564900165092955,
      "language_loss": 0.73388302,
      "learning_rate": 1.329171870732758e-06,
      "loss": 0.75492841,
      "num_input_tokens_seen": 222317095,
      "step": 10322,
      "time_per_iteration": 2.52860426902771
    },
    {
      "auxiliary_loss_clip": 0.01071381,
      "auxiliary_loss_mlp": 0.01025753,
      "balance_loss_clip": 1.03729033,
      "balance_loss_mlp": 1.01399326,
      "epoch": 0.6206523372914474,
      "flos": 23877831093120.0,
      "grad_norm": 1.776981103225701,
      "language_loss": 0.72640198,
      "learning_rate": 1.3288049829734845e-06,
      "loss": 0.74737334,
      "num_input_tokens_seen": 222337055,
      "step": 10323,
      "time_per_iteration": 2.565063238143921
    },
    {
      "auxiliary_loss_clip": 0.01105665,
      "auxiliary_loss_mlp": 0.01028668,
      "balance_loss_clip": 1.03933084,
      "balance_loss_mlp": 1.01592481,
      "epoch": 0.6207124605441154,
      "flos": 13406603218560.0,
      "grad_norm": 2.3102252525514615,
      "language_loss": 0.586932,
      "learning_rate": 1.3284381206659933e-06,
      "loss": 0.60827535,
      "num_input_tokens_seen": 222354515,
      "step": 10324,
      "time_per_iteration": 2.4527344703674316
    },
    {
      "auxiliary_loss_clip": 0.01071201,
      "auxiliary_loss_mlp": 0.01037858,
      "balance_loss_clip": 1.04039443,
      "balance_loss_mlp": 1.02391016,
      "epoch": 0.6207725837967835,
      "flos": 18916341287040.0,
      "grad_norm": 2.104094587113255,
      "language_loss": 0.76805454,
      "learning_rate": 1.3280712838241956e-06,
      "loss": 0.78914511,
      "num_input_tokens_seen": 222372755,
      "step": 10325,
      "time_per_iteration": 2.5205447673797607
    },
    {
      "auxiliary_loss_clip": 0.01101399,
      "auxiliary_loss_mlp": 0.01027308,
      "balance_loss_clip": 1.03905392,
      "balance_loss_mlp": 1.01425457,
      "epoch": 0.6208327070494514,
      "flos": 23980211832960.0,
      "grad_norm": 2.371576641875241,
      "language_loss": 0.72539979,
      "learning_rate": 1.327704472462003e-06,
      "loss": 0.74668694,
      "num_input_tokens_seen": 222391380,
      "step": 10326,
      "time_per_iteration": 2.501467227935791
    },
    {
      "auxiliary_loss_clip": 0.01102767,
      "auxiliary_loss_mlp": 0.01037325,
      "balance_loss_clip": 1.04048181,
      "balance_loss_mlp": 1.02509975,
      "epoch": 0.6208928303021194,
      "flos": 22820405587200.0,
      "grad_norm": 2.59792764824432,
      "language_loss": 0.73704427,
      "learning_rate": 1.3273376865933234e-06,
      "loss": 0.75844526,
      "num_input_tokens_seen": 222411165,
      "step": 10327,
      "time_per_iteration": 2.5073843002319336
    },
    {
      "auxiliary_loss_clip": 0.01088193,
      "auxiliary_loss_mlp": 0.01034855,
      "balance_loss_clip": 1.04096103,
      "balance_loss_mlp": 1.02153933,
      "epoch": 0.6209529535547873,
      "flos": 17564519911680.0,
      "grad_norm": 2.5240700304934047,
      "language_loss": 0.80002755,
      "learning_rate": 1.326970926232066e-06,
      "loss": 0.82125807,
      "num_input_tokens_seen": 222428110,
      "step": 10328,
      "time_per_iteration": 2.4913713932037354
    },
    {
      "auxiliary_loss_clip": 0.01077828,
      "auxiliary_loss_mlp": 0.01037384,
      "balance_loss_clip": 1.0369004,
      "balance_loss_mlp": 1.02487898,
      "epoch": 0.6210130768074553,
      "flos": 22011912311040.0,
      "grad_norm": 1.6867383608971656,
      "language_loss": 0.78065646,
      "learning_rate": 1.3266041913921396e-06,
      "loss": 0.8018086,
      "num_input_tokens_seen": 222446385,
      "step": 10329,
      "time_per_iteration": 2.5424201488494873
    },
    {
      "auxiliary_loss_clip": 0.01027342,
      "auxiliary_loss_mlp": 0.01006194,
      "balance_loss_clip": 1.01567531,
      "balance_loss_mlp": 1.00486517,
      "epoch": 0.6210732000601232,
      "flos": 63676873854720.0,
      "grad_norm": 0.8307831784421537,
      "language_loss": 0.62188965,
      "learning_rate": 1.3262374820874484e-06,
      "loss": 0.64222497,
      "num_input_tokens_seen": 222502150,
      "step": 10330,
      "time_per_iteration": 3.069375991821289
    },
    {
      "auxiliary_loss_clip": 0.011051,
      "auxiliary_loss_mlp": 0.01035233,
      "balance_loss_clip": 1.03961062,
      "balance_loss_mlp": 1.02148867,
      "epoch": 0.6211333233127913,
      "flos": 24243365848320.0,
      "grad_norm": 2.0695077976396643,
      "language_loss": 0.77176577,
      "learning_rate": 1.3258707983319002e-06,
      "loss": 0.79316908,
      "num_input_tokens_seen": 222519880,
      "step": 10331,
      "time_per_iteration": 2.502328634262085
    },
    {
      "auxiliary_loss_clip": 0.01114872,
      "auxiliary_loss_mlp": 0.01035757,
      "balance_loss_clip": 1.04056787,
      "balance_loss_mlp": 1.02265573,
      "epoch": 0.6211934465654592,
      "flos": 16943803960320.0,
      "grad_norm": 1.8742113176044293,
      "language_loss": 0.67774278,
      "learning_rate": 1.3255041401393992e-06,
      "loss": 0.69924909,
      "num_input_tokens_seen": 222538545,
      "step": 10332,
      "time_per_iteration": 2.464479684829712
    },
    {
      "auxiliary_loss_clip": 0.0107873,
      "auxiliary_loss_mlp": 0.0103072,
      "balance_loss_clip": 1.03871882,
      "balance_loss_mlp": 1.01891816,
      "epoch": 0.6212535698181272,
      "flos": 15267386355840.0,
      "grad_norm": 1.4222642310937574,
      "language_loss": 0.76207268,
      "learning_rate": 1.3251375075238476e-06,
      "loss": 0.78316718,
      "num_input_tokens_seen": 222556935,
      "step": 10333,
      "time_per_iteration": 2.5225982666015625
    },
    {
      "auxiliary_loss_clip": 0.01085125,
      "auxiliary_loss_mlp": 0.01029752,
      "balance_loss_clip": 1.04086828,
      "balance_loss_mlp": 1.01789665,
      "epoch": 0.6213136930707951,
      "flos": 13443950384640.0,
      "grad_norm": 2.185863346171761,
      "language_loss": 0.69652915,
      "learning_rate": 1.3247709004991507e-06,
      "loss": 0.71767795,
      "num_input_tokens_seen": 222574035,
      "step": 10334,
      "time_per_iteration": 2.5230274200439453
    },
    {
      "auxiliary_loss_clip": 0.01088844,
      "auxiliary_loss_mlp": 0.00784995,
      "balance_loss_clip": 1.04090142,
      "balance_loss_mlp": 1.0100708,
      "epoch": 0.6213738163234631,
      "flos": 18111223889280.0,
      "grad_norm": 1.9573043984485898,
      "language_loss": 0.70273542,
      "learning_rate": 1.3244043190792078e-06,
      "loss": 0.72147381,
      "num_input_tokens_seen": 222592290,
      "step": 10335,
      "time_per_iteration": 2.51918363571167
    },
    {
      "auxiliary_loss_clip": 0.01059882,
      "auxiliary_loss_mlp": 0.01033342,
      "balance_loss_clip": 1.03663707,
      "balance_loss_mlp": 1.02041936,
      "epoch": 0.621433939576131,
      "flos": 25337348421120.0,
      "grad_norm": 1.4624269149397995,
      "language_loss": 0.80290651,
      "learning_rate": 1.3240377632779213e-06,
      "loss": 0.82383871,
      "num_input_tokens_seen": 222612805,
      "step": 10336,
      "time_per_iteration": 2.615630865097046
    },
    {
      "auxiliary_loss_clip": 0.0110888,
      "auxiliary_loss_mlp": 0.01031597,
      "balance_loss_clip": 1.0396378,
      "balance_loss_mlp": 1.01919365,
      "epoch": 0.621494062828799,
      "flos": 22565619440640.0,
      "grad_norm": 1.7862466225823384,
      "language_loss": 0.73111564,
      "learning_rate": 1.3236712331091907e-06,
      "loss": 0.75252044,
      "num_input_tokens_seen": 222632260,
      "step": 10337,
      "time_per_iteration": 2.468074321746826
    },
    {
      "auxiliary_loss_clip": 0.01112796,
      "auxiliary_loss_mlp": 0.01036184,
      "balance_loss_clip": 1.03913784,
      "balance_loss_mlp": 1.02311921,
      "epoch": 0.621554186081467,
      "flos": 27417976750080.0,
      "grad_norm": 1.9997229042462794,
      "language_loss": 0.63068038,
      "learning_rate": 1.3233047285869145e-06,
      "loss": 0.65217018,
      "num_input_tokens_seen": 222653570,
      "step": 10338,
      "time_per_iteration": 2.516507148742676
    },
    {
      "auxiliary_loss_clip": 0.01099956,
      "auxiliary_loss_mlp": 0.01034586,
      "balance_loss_clip": 1.04040265,
      "balance_loss_mlp": 1.02195001,
      "epoch": 0.621614309334135,
      "flos": 22346815743360.0,
      "grad_norm": 1.6080909449961953,
      "language_loss": 0.71445632,
      "learning_rate": 1.322938249724991e-06,
      "loss": 0.7358017,
      "num_input_tokens_seen": 222672480,
      "step": 10339,
      "time_per_iteration": 2.486429452896118
    },
    {
      "auxiliary_loss_clip": 0.0106019,
      "auxiliary_loss_mlp": 0.01036279,
      "balance_loss_clip": 1.03747368,
      "balance_loss_mlp": 1.02235532,
      "epoch": 0.621674432586803,
      "flos": 19281229597440.0,
      "grad_norm": 1.6758144968377342,
      "language_loss": 0.69712049,
      "learning_rate": 1.3225717965373166e-06,
      "loss": 0.71808517,
      "num_input_tokens_seen": 222691200,
      "step": 10340,
      "time_per_iteration": 2.5642330646514893
    },
    {
      "auxiliary_loss_clip": 0.01065288,
      "auxiliary_loss_mlp": 0.01027713,
      "balance_loss_clip": 1.03641653,
      "balance_loss_mlp": 1.01572096,
      "epoch": 0.6217345558394709,
      "flos": 21609533180160.0,
      "grad_norm": 1.9192231681758047,
      "language_loss": 0.68937415,
      "learning_rate": 1.322205369037788e-06,
      "loss": 0.71030414,
      "num_input_tokens_seen": 222709975,
      "step": 10341,
      "time_per_iteration": 2.6003472805023193
    },
    {
      "auxiliary_loss_clip": 0.01099741,
      "auxiliary_loss_mlp": 0.01031007,
      "balance_loss_clip": 1.03924048,
      "balance_loss_mlp": 1.0170356,
      "epoch": 0.6217946790921389,
      "flos": 18004102554240.0,
      "grad_norm": 1.7398247492128267,
      "language_loss": 0.80931234,
      "learning_rate": 1.321838967240299e-06,
      "loss": 0.83061981,
      "num_input_tokens_seen": 222729005,
      "step": 10342,
      "time_per_iteration": 2.4681191444396973
    },
    {
      "auxiliary_loss_clip": 0.01027816,
      "auxiliary_loss_mlp": 0.0100228,
      "balance_loss_clip": 1.02269125,
      "balance_loss_mlp": 1.00106454,
      "epoch": 0.6218548023448068,
      "flos": 61973631768960.0,
      "grad_norm": 1.8906652110824225,
      "language_loss": 0.57344019,
      "learning_rate": 1.3214725911587452e-06,
      "loss": 0.59374118,
      "num_input_tokens_seen": 222786090,
      "step": 10343,
      "time_per_iteration": 4.383916139602661
    },
    {
      "auxiliary_loss_clip": 0.0106897,
      "auxiliary_loss_mlp": 0.01025635,
      "balance_loss_clip": 1.03554857,
      "balance_loss_mlp": 1.01416671,
      "epoch": 0.6219149255974749,
      "flos": 25739152934400.0,
      "grad_norm": 1.827995475032794,
      "language_loss": 0.73243034,
      "learning_rate": 1.3211062408070184e-06,
      "loss": 0.75337642,
      "num_input_tokens_seen": 222806100,
      "step": 10344,
      "time_per_iteration": 2.5724427700042725
    },
    {
      "auxiliary_loss_clip": 0.01102498,
      "auxiliary_loss_mlp": 0.01041809,
      "balance_loss_clip": 1.04142594,
      "balance_loss_mlp": 1.0301801,
      "epoch": 0.6219750488501428,
      "flos": 25411073086080.0,
      "grad_norm": 1.7640645888649056,
      "language_loss": 0.59894419,
      "learning_rate": 1.3207399161990105e-06,
      "loss": 0.62038732,
      "num_input_tokens_seen": 222826575,
      "step": 10345,
      "time_per_iteration": 2.531609296798706
    },
    {
      "auxiliary_loss_clip": 0.01048528,
      "auxiliary_loss_mlp": 0.01036025,
      "balance_loss_clip": 1.03690147,
      "balance_loss_mlp": 1.02268577,
      "epoch": 0.6220351721028108,
      "flos": 20047383717120.0,
      "grad_norm": 1.9353705140344668,
      "language_loss": 0.77904141,
      "learning_rate": 1.320373617348614e-06,
      "loss": 0.79988688,
      "num_input_tokens_seen": 222845285,
      "step": 10346,
      "time_per_iteration": 2.603142261505127
    },
    {
      "auxiliary_loss_clip": 0.01078848,
      "auxiliary_loss_mlp": 0.01033387,
      "balance_loss_clip": 1.03928995,
      "balance_loss_mlp": 1.02019036,
      "epoch": 0.6220952953554787,
      "flos": 27488397363840.0,
      "grad_norm": 1.7123622155346108,
      "language_loss": 0.71470809,
      "learning_rate": 1.3200073442697171e-06,
      "loss": 0.73583049,
      "num_input_tokens_seen": 222864575,
      "step": 10347,
      "time_per_iteration": 2.7059385776519775
    },
    {
      "auxiliary_loss_clip": 0.0109624,
      "auxiliary_loss_mlp": 0.01030858,
      "balance_loss_clip": 1.0373286,
      "balance_loss_mlp": 1.01856172,
      "epoch": 0.6221554186081467,
      "flos": 19207612673280.0,
      "grad_norm": 1.8654247065584086,
      "language_loss": 0.71920455,
      "learning_rate": 1.3196410969762108e-06,
      "loss": 0.74047554,
      "num_input_tokens_seen": 222884420,
      "step": 10348,
      "time_per_iteration": 2.498401403427124
    },
    {
      "auxiliary_loss_clip": 0.01021966,
      "auxiliary_loss_mlp": 0.01001482,
      "balance_loss_clip": 1.03225851,
      "balance_loss_mlp": 1.00022411,
      "epoch": 0.6222155418608146,
      "flos": 62950939989120.0,
      "grad_norm": 0.811725789409365,
      "language_loss": 0.54196525,
      "learning_rate": 1.3192748754819815e-06,
      "loss": 0.56219971,
      "num_input_tokens_seen": 222944690,
      "step": 10349,
      "time_per_iteration": 3.1721293926239014
    },
    {
      "auxiliary_loss_clip": 0.01075881,
      "auxiliary_loss_mlp": 0.01026513,
      "balance_loss_clip": 1.04153109,
      "balance_loss_mlp": 1.01450253,
      "epoch": 0.6222756651134826,
      "flos": 22601099099520.0,
      "grad_norm": 1.82479656803921,
      "language_loss": 0.69939268,
      "learning_rate": 1.3189086798009173e-06,
      "loss": 0.72041661,
      "num_input_tokens_seen": 222962990,
      "step": 10350,
      "time_per_iteration": 2.57356595993042
    },
    {
      "auxiliary_loss_clip": 0.01111037,
      "auxiliary_loss_mlp": 0.01034267,
      "balance_loss_clip": 1.03903794,
      "balance_loss_mlp": 1.02170825,
      "epoch": 0.6223357883661506,
      "flos": 21142228216320.0,
      "grad_norm": 2.03964845280813,
      "language_loss": 0.56645346,
      "learning_rate": 1.3185425099469046e-06,
      "loss": 0.58790648,
      "num_input_tokens_seen": 222980715,
      "step": 10351,
      "time_per_iteration": 2.4460716247558594
    },
    {
      "auxiliary_loss_clip": 0.01035462,
      "auxiliary_loss_mlp": 0.01001648,
      "balance_loss_clip": 1.02616692,
      "balance_loss_mlp": 1.00040233,
      "epoch": 0.6223959116188186,
      "flos": 63765071700480.0,
      "grad_norm": 0.8033526009642041,
      "language_loss": 0.61123174,
      "learning_rate": 1.3181763659338276e-06,
      "loss": 0.63160288,
      "num_input_tokens_seen": 223040685,
      "step": 10352,
      "time_per_iteration": 3.0476980209350586
    },
    {
      "auxiliary_loss_clip": 0.01106781,
      "auxiliary_loss_mlp": 0.01030706,
      "balance_loss_clip": 1.03785253,
      "balance_loss_mlp": 1.01859462,
      "epoch": 0.6224560348714866,
      "flos": 22565727181440.0,
      "grad_norm": 2.0346916955962193,
      "language_loss": 0.81658596,
      "learning_rate": 1.3178102477755714e-06,
      "loss": 0.83796084,
      "num_input_tokens_seen": 223059000,
      "step": 10353,
      "time_per_iteration": 3.892784595489502
    },
    {
      "auxiliary_loss_clip": 0.0109372,
      "auxiliary_loss_mlp": 0.01030166,
      "balance_loss_clip": 1.03770709,
      "balance_loss_mlp": 1.01884675,
      "epoch": 0.6225161581241545,
      "flos": 24097748112000.0,
      "grad_norm": 1.471011403383393,
      "language_loss": 0.75709689,
      "learning_rate": 1.3174441554860195e-06,
      "loss": 0.77833569,
      "num_input_tokens_seen": 223079345,
      "step": 10354,
      "time_per_iteration": 2.5224435329437256
    },
    {
      "auxiliary_loss_clip": 0.01072381,
      "auxiliary_loss_mlp": 0.01030493,
      "balance_loss_clip": 1.03809977,
      "balance_loss_mlp": 1.01835728,
      "epoch": 0.6225762813768225,
      "flos": 20443513881600.0,
      "grad_norm": 1.5260825061607357,
      "language_loss": 0.78664702,
      "learning_rate": 1.3170780890790528e-06,
      "loss": 0.80767578,
      "num_input_tokens_seen": 223097880,
      "step": 10355,
      "time_per_iteration": 2.554327964782715
    },
    {
      "auxiliary_loss_clip": 0.01099284,
      "auxiliary_loss_mlp": 0.01034064,
      "balance_loss_clip": 1.03988457,
      "balance_loss_mlp": 1.02217841,
      "epoch": 0.6226364046294904,
      "flos": 27198131558400.0,
      "grad_norm": 2.047577098888076,
      "language_loss": 0.78286564,
      "learning_rate": 1.3167120485685538e-06,
      "loss": 0.8041991,
      "num_input_tokens_seen": 223118185,
      "step": 10356,
      "time_per_iteration": 2.535832166671753
    },
    {
      "auxiliary_loss_clip": 0.01093008,
      "auxiliary_loss_mlp": 0.00784927,
      "balance_loss_clip": 1.03905332,
      "balance_loss_mlp": 1.00820613,
      "epoch": 0.6226965278821585,
      "flos": 20445776438400.0,
      "grad_norm": 2.1942806145280764,
      "language_loss": 0.67501903,
      "learning_rate": 1.3163460339684024e-06,
      "loss": 0.69379842,
      "num_input_tokens_seen": 223137600,
      "step": 10357,
      "time_per_iteration": 3.95259165763855
    },
    {
      "auxiliary_loss_clip": 0.01092412,
      "auxiliary_loss_mlp": 0.01031835,
      "balance_loss_clip": 1.03930557,
      "balance_loss_mlp": 1.01811981,
      "epoch": 0.6227566511348264,
      "flos": 22162737519360.0,
      "grad_norm": 3.4018562099545018,
      "language_loss": 0.76039255,
      "learning_rate": 1.3159800452924778e-06,
      "loss": 0.78163499,
      "num_input_tokens_seen": 223154360,
      "step": 10358,
      "time_per_iteration": 2.5029590129852295
    },
    {
      "auxiliary_loss_clip": 0.01086061,
      "auxiliary_loss_mlp": 0.01031804,
      "balance_loss_clip": 1.03723025,
      "balance_loss_mlp": 1.01966226,
      "epoch": 0.6228167743874944,
      "flos": 18040875102720.0,
      "grad_norm": 3.1089172565217766,
      "language_loss": 0.82582486,
      "learning_rate": 1.3156140825546588e-06,
      "loss": 0.84700352,
      "num_input_tokens_seen": 223172255,
      "step": 10359,
      "time_per_iteration": 2.5493814945220947
    },
    {
      "auxiliary_loss_clip": 0.01080427,
      "auxiliary_loss_mlp": 0.01046967,
      "balance_loss_clip": 1.03541434,
      "balance_loss_mlp": 1.03285813,
      "epoch": 0.6228768976401623,
      "flos": 17742851959680.0,
      "grad_norm": 5.604555152594505,
      "language_loss": 0.73099411,
      "learning_rate": 1.315248145768822e-06,
      "loss": 0.75226796,
      "num_input_tokens_seen": 223186965,
      "step": 10360,
      "time_per_iteration": 2.473407745361328
    },
    {
      "auxiliary_loss_clip": 0.01097597,
      "auxiliary_loss_mlp": 0.01036051,
      "balance_loss_clip": 1.03653371,
      "balance_loss_mlp": 1.02354574,
      "epoch": 0.6229370208928303,
      "flos": 17894934144000.0,
      "grad_norm": 2.001196274797408,
      "language_loss": 0.77576435,
      "learning_rate": 1.3148822349488442e-06,
      "loss": 0.79710084,
      "num_input_tokens_seen": 223206045,
      "step": 10361,
      "time_per_iteration": 2.4644112586975098
    },
    {
      "auxiliary_loss_clip": 0.01069927,
      "auxiliary_loss_mlp": 0.0103138,
      "balance_loss_clip": 1.03875542,
      "balance_loss_mlp": 1.01941133,
      "epoch": 0.6229971441454982,
      "flos": 17347763289600.0,
      "grad_norm": 1.5414892471466635,
      "language_loss": 0.67892194,
      "learning_rate": 1.3145163501086005e-06,
      "loss": 0.69993496,
      "num_input_tokens_seen": 223224820,
      "step": 10362,
      "time_per_iteration": 2.5147337913513184
    },
    {
      "auxiliary_loss_clip": 0.01089181,
      "auxiliary_loss_mlp": 0.01031704,
      "balance_loss_clip": 1.0372901,
      "balance_loss_mlp": 1.01916337,
      "epoch": 0.6230572673981662,
      "flos": 29241376807680.0,
      "grad_norm": 2.092862548291175,
      "language_loss": 0.67383564,
      "learning_rate": 1.3141504912619658e-06,
      "loss": 0.69504452,
      "num_input_tokens_seen": 223243205,
      "step": 10363,
      "time_per_iteration": 2.5770883560180664
    },
    {
      "auxiliary_loss_clip": 0.01070685,
      "auxiliary_loss_mlp": 0.01033479,
      "balance_loss_clip": 1.03938055,
      "balance_loss_mlp": 1.02033055,
      "epoch": 0.6231173906508342,
      "flos": 16325961096960.0,
      "grad_norm": 2.250168678401797,
      "language_loss": 0.86724919,
      "learning_rate": 1.3137846584228127e-06,
      "loss": 0.88829088,
      "num_input_tokens_seen": 223261370,
      "step": 10364,
      "time_per_iteration": 2.5342624187469482
    },
    {
      "auxiliary_loss_clip": 0.01023784,
      "auxiliary_loss_mlp": 0.01000774,
      "balance_loss_clip": 1.02004421,
      "balance_loss_mlp": 0.99938548,
      "epoch": 0.6231775139035022,
      "flos": 68702032517760.0,
      "grad_norm": 0.8808761643034688,
      "language_loss": 0.60814065,
      "learning_rate": 1.313418851605015e-06,
      "loss": 0.62838626,
      "num_input_tokens_seen": 223315050,
      "step": 10365,
      "time_per_iteration": 3.1386258602142334
    },
    {
      "auxiliary_loss_clip": 0.01076642,
      "auxiliary_loss_mlp": 0.00786899,
      "balance_loss_clip": 1.03966212,
      "balance_loss_mlp": 1.0105809,
      "epoch": 0.6232376371561702,
      "flos": 19821038163840.0,
      "grad_norm": 7.625894810710482,
      "language_loss": 0.74899304,
      "learning_rate": 1.3130530708224427e-06,
      "loss": 0.76762843,
      "num_input_tokens_seen": 223332130,
      "step": 10366,
      "time_per_iteration": 2.541701316833496
    },
    {
      "auxiliary_loss_clip": 0.01101756,
      "auxiliary_loss_mlp": 0.01040647,
      "balance_loss_clip": 1.04080284,
      "balance_loss_mlp": 1.0274806,
      "epoch": 0.6232977604088381,
      "flos": 23258264376960.0,
      "grad_norm": 1.9781159198130742,
      "language_loss": 0.7611028,
      "learning_rate": 1.3126873160889665e-06,
      "loss": 0.78252685,
      "num_input_tokens_seen": 223351605,
      "step": 10367,
      "time_per_iteration": 2.5061776638031006
    },
    {
      "auxiliary_loss_clip": 0.01096769,
      "auxiliary_loss_mlp": 0.01037181,
      "balance_loss_clip": 1.03838134,
      "balance_loss_mlp": 1.02496839,
      "epoch": 0.6233578836615061,
      "flos": 21106425335040.0,
      "grad_norm": 1.537283045540266,
      "language_loss": 0.78581578,
      "learning_rate": 1.312321587418457e-06,
      "loss": 0.80715531,
      "num_input_tokens_seen": 223372090,
      "step": 10368,
      "time_per_iteration": 2.496858835220337
    },
    {
      "auxiliary_loss_clip": 0.0104641,
      "auxiliary_loss_mlp": 0.01036249,
      "balance_loss_clip": 1.04031181,
      "balance_loss_mlp": 1.02340984,
      "epoch": 0.623418006914174,
      "flos": 23769416868480.0,
      "grad_norm": 1.895976559733545,
      "language_loss": 0.68021685,
      "learning_rate": 1.3119558848247811e-06,
      "loss": 0.70104349,
      "num_input_tokens_seen": 223390110,
      "step": 10369,
      "time_per_iteration": 2.6398022174835205
    },
    {
      "auxiliary_loss_clip": 0.01112052,
      "auxiliary_loss_mlp": 0.01035761,
      "balance_loss_clip": 1.04059601,
      "balance_loss_mlp": 1.0227617,
      "epoch": 0.6234781301668421,
      "flos": 17890480857600.0,
      "grad_norm": 2.5836196270929586,
      "language_loss": 0.88088644,
      "learning_rate": 1.3115902083218072e-06,
      "loss": 0.90236461,
      "num_input_tokens_seen": 223404205,
      "step": 10370,
      "time_per_iteration": 2.3988420963287354
    },
    {
      "auxiliary_loss_clip": 0.01106747,
      "auxiliary_loss_mlp": 0.01030359,
      "balance_loss_clip": 1.03732979,
      "balance_loss_mlp": 1.01847959,
      "epoch": 0.62353825341951,
      "flos": 26175503352960.0,
      "grad_norm": 1.505710372062568,
      "language_loss": 0.65805304,
      "learning_rate": 1.311224557923402e-06,
      "loss": 0.67942405,
      "num_input_tokens_seen": 223424855,
      "step": 10371,
      "time_per_iteration": 2.5088272094726562
    },
    {
      "auxiliary_loss_clip": 0.01092385,
      "auxiliary_loss_mlp": 0.0102714,
      "balance_loss_clip": 1.03725612,
      "balance_loss_mlp": 1.01713872,
      "epoch": 0.623598376672178,
      "flos": 31139902160640.0,
      "grad_norm": 1.3053524897179611,
      "language_loss": 0.77621478,
      "learning_rate": 1.3108589336434298e-06,
      "loss": 0.79741013,
      "num_input_tokens_seen": 223447225,
      "step": 10372,
      "time_per_iteration": 2.5837314128875732
    },
    {
      "auxiliary_loss_clip": 0.01096435,
      "auxiliary_loss_mlp": 0.01031537,
      "balance_loss_clip": 1.03546214,
      "balance_loss_mlp": 1.01871586,
      "epoch": 0.6236584999248459,
      "flos": 23730202195200.0,
      "grad_norm": 1.5828462597840935,
      "language_loss": 0.77391291,
      "learning_rate": 1.3104933354957568e-06,
      "loss": 0.79519272,
      "num_input_tokens_seen": 223467520,
      "step": 10373,
      "time_per_iteration": 2.5462214946746826
    },
    {
      "auxiliary_loss_clip": 0.01092891,
      "auxiliary_loss_mlp": 0.01027112,
      "balance_loss_clip": 1.03700876,
      "balance_loss_mlp": 1.01590073,
      "epoch": 0.6237186231775139,
      "flos": 21762764599680.0,
      "grad_norm": 2.4228806480173706,
      "language_loss": 0.69275683,
      "learning_rate": 1.3101277634942448e-06,
      "loss": 0.71395683,
      "num_input_tokens_seen": 223488130,
      "step": 10374,
      "time_per_iteration": 2.4801623821258545
    },
    {
      "auxiliary_loss_clip": 0.01092384,
      "auxiliary_loss_mlp": 0.01029258,
      "balance_loss_clip": 1.039446,
      "balance_loss_mlp": 1.01715851,
      "epoch": 0.6237787464301818,
      "flos": 14939486075520.0,
      "grad_norm": 1.7237660611709427,
      "language_loss": 0.7708413,
      "learning_rate": 1.3097622176527577e-06,
      "loss": 0.79205775,
      "num_input_tokens_seen": 223505105,
      "step": 10375,
      "time_per_iteration": 2.5024116039276123
    },
    {
      "auxiliary_loss_clip": 0.01083034,
      "auxiliary_loss_mlp": 0.01026957,
      "balance_loss_clip": 1.03987169,
      "balance_loss_mlp": 1.01519108,
      "epoch": 0.6238388696828499,
      "flos": 35590311302400.0,
      "grad_norm": 1.4721036058764214,
      "language_loss": 0.70102376,
      "learning_rate": 1.3093966979851566e-06,
      "loss": 0.72212368,
      "num_input_tokens_seen": 223528065,
      "step": 10376,
      "time_per_iteration": 2.641611099243164
    },
    {
      "auxiliary_loss_clip": 0.01079238,
      "auxiliary_loss_mlp": 0.01030921,
      "balance_loss_clip": 1.0372684,
      "balance_loss_mlp": 1.01767731,
      "epoch": 0.6238989929355178,
      "flos": 23623511823360.0,
      "grad_norm": 1.65726625716149,
      "language_loss": 0.76454961,
      "learning_rate": 1.309031204505301e-06,
      "loss": 0.78565121,
      "num_input_tokens_seen": 223547305,
      "step": 10377,
      "time_per_iteration": 2.587254285812378
    },
    {
      "auxiliary_loss_clip": 0.01087826,
      "auxiliary_loss_mlp": 0.0102827,
      "balance_loss_clip": 1.03940511,
      "balance_loss_mlp": 1.01745141,
      "epoch": 0.6239591161881858,
      "flos": 22087468569600.0,
      "grad_norm": 2.0477407822732894,
      "language_loss": 0.68186331,
      "learning_rate": 1.308665737227052e-06,
      "loss": 0.70302427,
      "num_input_tokens_seen": 223567205,
      "step": 10378,
      "time_per_iteration": 2.518216609954834
    },
    {
      "auxiliary_loss_clip": 0.01084619,
      "auxiliary_loss_mlp": 0.01026748,
      "balance_loss_clip": 1.03818178,
      "balance_loss_mlp": 1.01473141,
      "epoch": 0.6240192394408538,
      "flos": 24535930124160.0,
      "grad_norm": 1.7917867412069672,
      "language_loss": 0.76723635,
      "learning_rate": 1.3083002961642675e-06,
      "loss": 0.78835005,
      "num_input_tokens_seen": 223586560,
      "step": 10379,
      "time_per_iteration": 2.5792078971862793
    },
    {
      "auxiliary_loss_clip": 0.01085049,
      "auxiliary_loss_mlp": 0.01028258,
      "balance_loss_clip": 1.04044986,
      "balance_loss_mlp": 1.01665318,
      "epoch": 0.6240793626935217,
      "flos": 27931930502400.0,
      "grad_norm": 1.4348673256370892,
      "language_loss": 0.79410815,
      "learning_rate": 1.3079348813308051e-06,
      "loss": 0.81524122,
      "num_input_tokens_seen": 223610595,
      "step": 10380,
      "time_per_iteration": 2.6024346351623535
    },
    {
      "auxiliary_loss_clip": 0.01093035,
      "auxiliary_loss_mlp": 0.01029581,
      "balance_loss_clip": 1.04064012,
      "balance_loss_mlp": 1.01884627,
      "epoch": 0.6241394859461897,
      "flos": 22892514140160.0,
      "grad_norm": 1.5306960978845066,
      "language_loss": 0.79842067,
      "learning_rate": 1.3075694927405207e-06,
      "loss": 0.81964684,
      "num_input_tokens_seen": 223630230,
      "step": 10381,
      "time_per_iteration": 2.5170445442199707
    },
    {
      "auxiliary_loss_clip": 0.0108241,
      "auxiliary_loss_mlp": 0.01032437,
      "balance_loss_clip": 1.03534675,
      "balance_loss_mlp": 1.0192883,
      "epoch": 0.6241996091988576,
      "flos": 12750766744320.0,
      "grad_norm": 2.161369883552236,
      "language_loss": 0.7489233,
      "learning_rate": 1.3072041304072718e-06,
      "loss": 0.77007174,
      "num_input_tokens_seen": 223648360,
      "step": 10382,
      "time_per_iteration": 3.867119550704956
    },
    {
      "auxiliary_loss_clip": 0.01095523,
      "auxiliary_loss_mlp": 0.01024456,
      "balance_loss_clip": 1.03789592,
      "balance_loss_mlp": 1.01317883,
      "epoch": 0.6242597324515257,
      "flos": 25851302173440.0,
      "grad_norm": 1.6190919635976357,
      "language_loss": 0.78386092,
      "learning_rate": 1.306838794344911e-06,
      "loss": 0.80506068,
      "num_input_tokens_seen": 223671255,
      "step": 10383,
      "time_per_iteration": 2.5709927082061768
    },
    {
      "auxiliary_loss_clip": 0.01073515,
      "auxiliary_loss_mlp": 0.01028624,
      "balance_loss_clip": 1.03564739,
      "balance_loss_mlp": 1.01720381,
      "epoch": 0.6243198557041936,
      "flos": 19937712516480.0,
      "grad_norm": 1.9323041271273642,
      "language_loss": 0.75544882,
      "learning_rate": 1.3064734845672925e-06,
      "loss": 0.77647024,
      "num_input_tokens_seen": 223689860,
      "step": 10384,
      "time_per_iteration": 2.5589845180511475
    },
    {
      "auxiliary_loss_clip": 0.01084772,
      "auxiliary_loss_mlp": 0.01037185,
      "balance_loss_clip": 1.03560615,
      "balance_loss_mlp": 1.02313042,
      "epoch": 0.6243799789568616,
      "flos": 18406194376320.0,
      "grad_norm": 2.2162107632293377,
      "language_loss": 0.66755736,
      "learning_rate": 1.3061082010882694e-06,
      "loss": 0.68877685,
      "num_input_tokens_seen": 223707835,
      "step": 10385,
      "time_per_iteration": 2.5081567764282227
    },
    {
      "auxiliary_loss_clip": 0.01034468,
      "auxiliary_loss_mlp": 0.01008308,
      "balance_loss_clip": 1.03001821,
      "balance_loss_mlp": 1.0070858,
      "epoch": 0.6244401022095295,
      "flos": 66027587523840.0,
      "grad_norm": 0.753854596023269,
      "language_loss": 0.6196686,
      "learning_rate": 1.305742943921692e-06,
      "loss": 0.64009643,
      "num_input_tokens_seen": 223771875,
      "step": 10386,
      "time_per_iteration": 3.1580936908721924
    },
    {
      "auxiliary_loss_clip": 0.01096665,
      "auxiliary_loss_mlp": 0.01031631,
      "balance_loss_clip": 1.03673744,
      "balance_loss_mlp": 1.01936424,
      "epoch": 0.6245002254621975,
      "flos": 24571266128640.0,
      "grad_norm": 2.8958572725957485,
      "language_loss": 0.71930933,
      "learning_rate": 1.3053777130814128e-06,
      "loss": 0.74059224,
      "num_input_tokens_seen": 223788895,
      "step": 10387,
      "time_per_iteration": 2.50054931640625
    },
    {
      "auxiliary_loss_clip": 0.01102334,
      "auxiliary_loss_mlp": 0.01036718,
      "balance_loss_clip": 1.03838325,
      "balance_loss_mlp": 1.02301502,
      "epoch": 0.6245603487148654,
      "flos": 29168837291520.0,
      "grad_norm": 2.9658874948828493,
      "language_loss": 0.65422416,
      "learning_rate": 1.3050125085812798e-06,
      "loss": 0.67561471,
      "num_input_tokens_seen": 223810385,
      "step": 10388,
      "time_per_iteration": 2.5476651191711426
    },
    {
      "auxiliary_loss_clip": 0.01070668,
      "auxiliary_loss_mlp": 0.01027808,
      "balance_loss_clip": 1.03812194,
      "balance_loss_mlp": 1.01654899,
      "epoch": 0.6246204719675335,
      "flos": 14790097411200.0,
      "grad_norm": 1.7528521602825702,
      "language_loss": 0.79301149,
      "learning_rate": 1.3046473304351417e-06,
      "loss": 0.81399626,
      "num_input_tokens_seen": 223826040,
      "step": 10389,
      "time_per_iteration": 2.509653091430664
    },
    {
      "auxiliary_loss_clip": 0.01083142,
      "auxiliary_loss_mlp": 0.01032612,
      "balance_loss_clip": 1.03612423,
      "balance_loss_mlp": 1.0200417,
      "epoch": 0.6246805952202014,
      "flos": 12493538472960.0,
      "grad_norm": 1.8561598384607805,
      "language_loss": 0.60762113,
      "learning_rate": 1.3042821786568475e-06,
      "loss": 0.6287787,
      "num_input_tokens_seen": 223842300,
      "step": 10390,
      "time_per_iteration": 2.4881911277770996
    },
    {
      "auxiliary_loss_clip": 0.01088382,
      "auxiliary_loss_mlp": 0.01032524,
      "balance_loss_clip": 1.03770399,
      "balance_loss_mlp": 1.02013266,
      "epoch": 0.6247407184728694,
      "flos": 12786677366400.0,
      "grad_norm": 1.9640398734552889,
      "language_loss": 0.77128243,
      "learning_rate": 1.3039170532602416e-06,
      "loss": 0.79249144,
      "num_input_tokens_seen": 223858320,
      "step": 10391,
      "time_per_iteration": 2.4894776344299316
    },
    {
      "auxiliary_loss_clip": 0.01087652,
      "auxiliary_loss_mlp": 0.01028608,
      "balance_loss_clip": 1.03794158,
      "balance_loss_mlp": 1.01593614,
      "epoch": 0.6248008417255374,
      "flos": 40629188960640.0,
      "grad_norm": 1.5299198989858573,
      "language_loss": 0.6466549,
      "learning_rate": 1.3035519542591718e-06,
      "loss": 0.66781747,
      "num_input_tokens_seen": 223883545,
      "step": 10392,
      "time_per_iteration": 5.53208327293396
    },
    {
      "auxiliary_loss_clip": 0.01089913,
      "auxiliary_loss_mlp": 0.01029168,
      "balance_loss_clip": 1.03911757,
      "balance_loss_mlp": 1.01685345,
      "epoch": 0.6248609649782053,
      "flos": 19902017376000.0,
      "grad_norm": 1.7866888538317303,
      "language_loss": 0.76745772,
      "learning_rate": 1.3031868816674819e-06,
      "loss": 0.78864855,
      "num_input_tokens_seen": 223901445,
      "step": 10393,
      "time_per_iteration": 2.520374059677124
    },
    {
      "auxiliary_loss_clip": 0.0108032,
      "auxiliary_loss_mlp": 0.00784953,
      "balance_loss_clip": 1.03901064,
      "balance_loss_mlp": 1.00790405,
      "epoch": 0.6249210882308733,
      "flos": 19682746801920.0,
      "grad_norm": 1.771515239873333,
      "language_loss": 0.8265655,
      "learning_rate": 1.3028218354990142e-06,
      "loss": 0.84521818,
      "num_input_tokens_seen": 223920170,
      "step": 10394,
      "time_per_iteration": 2.5773651599884033
    },
    {
      "auxiliary_loss_clip": 0.01091014,
      "auxiliary_loss_mlp": 0.01032526,
      "balance_loss_clip": 1.03950429,
      "balance_loss_mlp": 1.01928186,
      "epoch": 0.6249812114835412,
      "flos": 13990726189440.0,
      "grad_norm": 1.755840881824861,
      "language_loss": 0.74949431,
      "learning_rate": 1.3024568157676128e-06,
      "loss": 0.77072972,
      "num_input_tokens_seen": 223936495,
      "step": 10395,
      "time_per_iteration": 3.9070889949798584
    },
    {
      "auxiliary_loss_clip": 0.01089312,
      "auxiliary_loss_mlp": 0.01031167,
      "balance_loss_clip": 1.03601873,
      "balance_loss_mlp": 1.0184232,
      "epoch": 0.6250413347362093,
      "flos": 14530031965440.0,
      "grad_norm": 2.652138439941464,
      "language_loss": 0.72603959,
      "learning_rate": 1.302091822487119e-06,
      "loss": 0.74724436,
      "num_input_tokens_seen": 223950070,
      "step": 10396,
      "time_per_iteration": 2.4720776081085205
    },
    {
      "auxiliary_loss_clip": 0.01078763,
      "auxiliary_loss_mlp": 0.01034238,
      "balance_loss_clip": 1.043872,
      "balance_loss_mlp": 1.02221608,
      "epoch": 0.6251014579888772,
      "flos": 22963006581120.0,
      "grad_norm": 11.242142536766837,
      "language_loss": 0.75799924,
      "learning_rate": 1.3017268556713732e-06,
      "loss": 0.77912927,
      "num_input_tokens_seen": 223970065,
      "step": 10397,
      "time_per_iteration": 2.5743565559387207
    },
    {
      "auxiliary_loss_clip": 0.01082056,
      "auxiliary_loss_mlp": 0.0103089,
      "balance_loss_clip": 1.03473949,
      "balance_loss_mlp": 1.01837873,
      "epoch": 0.6251615812415452,
      "flos": 28111232217600.0,
      "grad_norm": 2.4450501163243086,
      "language_loss": 0.75257838,
      "learning_rate": 1.3013619153342154e-06,
      "loss": 0.77370787,
      "num_input_tokens_seen": 223990315,
      "step": 10398,
      "time_per_iteration": 2.56254243850708
    },
    {
      "auxiliary_loss_clip": 0.01110794,
      "auxiliary_loss_mlp": 0.01029827,
      "balance_loss_clip": 1.03671312,
      "balance_loss_mlp": 1.01629722,
      "epoch": 0.6252217044942131,
      "flos": 26724469887360.0,
      "grad_norm": 1.7573489953865604,
      "language_loss": 0.74180114,
      "learning_rate": 1.300997001489483e-06,
      "loss": 0.76320732,
      "num_input_tokens_seen": 224009960,
      "step": 10399,
      "time_per_iteration": 2.4849932193756104
    },
    {
      "auxiliary_loss_clip": 0.01072698,
      "auxiliary_loss_mlp": 0.01032018,
      "balance_loss_clip": 1.03841877,
      "balance_loss_mlp": 1.02025211,
      "epoch": 0.6252818277468811,
      "flos": 20006768413440.0,
      "grad_norm": 1.551611372192619,
      "language_loss": 0.74387175,
      "learning_rate": 1.3006321141510147e-06,
      "loss": 0.76491886,
      "num_input_tokens_seen": 224028870,
      "step": 10400,
      "time_per_iteration": 2.5610952377319336
    },
    {
      "auxiliary_loss_clip": 0.01023217,
      "auxiliary_loss_mlp": 0.01003463,
      "balance_loss_clip": 1.02647913,
      "balance_loss_mlp": 1.00210404,
      "epoch": 0.625341950999549,
      "flos": 59278285059840.0,
      "grad_norm": 0.8429702007772868,
      "language_loss": 0.56505823,
      "learning_rate": 1.3002672533326465e-06,
      "loss": 0.585325,
      "num_input_tokens_seen": 224094140,
      "step": 10401,
      "time_per_iteration": 3.249868392944336
    },
    {
      "auxiliary_loss_clip": 0.01099944,
      "auxiliary_loss_mlp": 0.01030144,
      "balance_loss_clip": 1.03767502,
      "balance_loss_mlp": 1.01737118,
      "epoch": 0.625402074252217,
      "flos": 20157090831360.0,
      "grad_norm": 2.2794690756638296,
      "language_loss": 0.82826817,
      "learning_rate": 1.2999024190482146e-06,
      "loss": 0.84956908,
      "num_input_tokens_seen": 224113235,
      "step": 10402,
      "time_per_iteration": 2.482293128967285
    },
    {
      "auxiliary_loss_clip": 0.01037223,
      "auxiliary_loss_mlp": 0.01031548,
      "balance_loss_clip": 1.03736842,
      "balance_loss_mlp": 1.01960289,
      "epoch": 0.625462197504885,
      "flos": 29132531619840.0,
      "grad_norm": 2.564860651947877,
      "language_loss": 0.69261849,
      "learning_rate": 1.2995376113115527e-06,
      "loss": 0.71330619,
      "num_input_tokens_seen": 224134530,
      "step": 10403,
      "time_per_iteration": 2.7535903453826904
    },
    {
      "auxiliary_loss_clip": 0.01075038,
      "auxiliary_loss_mlp": 0.01032263,
      "balance_loss_clip": 1.03579271,
      "balance_loss_mlp": 1.01786876,
      "epoch": 0.625522320757553,
      "flos": 26104436294400.0,
      "grad_norm": 1.5550691738195606,
      "language_loss": 0.71659315,
      "learning_rate": 1.2991728301364954e-06,
      "loss": 0.73766619,
      "num_input_tokens_seen": 224154170,
      "step": 10404,
      "time_per_iteration": 2.7630813121795654
    },
    {
      "auxiliary_loss_clip": 0.01068122,
      "auxiliary_loss_mlp": 0.01031655,
      "balance_loss_clip": 1.0386225,
      "balance_loss_mlp": 1.0191499,
      "epoch": 0.625582444010221,
      "flos": 20630967984000.0,
      "grad_norm": 1.7783686166310555,
      "language_loss": 0.69349581,
      "learning_rate": 1.2988080755368742e-06,
      "loss": 0.71449363,
      "num_input_tokens_seen": 224172730,
      "step": 10405,
      "time_per_iteration": 2.5756609439849854
    },
    {
      "auxiliary_loss_clip": 0.01082225,
      "auxiliary_loss_mlp": 0.01032083,
      "balance_loss_clip": 1.03801572,
      "balance_loss_mlp": 1.02007818,
      "epoch": 0.6256425672628889,
      "flos": 20521512264960.0,
      "grad_norm": 1.566962298246778,
      "language_loss": 0.78846246,
      "learning_rate": 1.2984433475265207e-06,
      "loss": 0.8096056,
      "num_input_tokens_seen": 224192620,
      "step": 10406,
      "time_per_iteration": 2.528078079223633
    },
    {
      "auxiliary_loss_clip": 0.01079428,
      "auxiliary_loss_mlp": 0.01033924,
      "balance_loss_clip": 1.0381856,
      "balance_loss_mlp": 1.02164531,
      "epoch": 0.6257026905155569,
      "flos": 29529200488320.0,
      "grad_norm": 1.8508674870017676,
      "language_loss": 0.69063848,
      "learning_rate": 1.2980786461192666e-06,
      "loss": 0.71177202,
      "num_input_tokens_seen": 224214660,
      "step": 10407,
      "time_per_iteration": 2.6205174922943115
    },
    {
      "auxiliary_loss_clip": 0.01096013,
      "auxiliary_loss_mlp": 0.00783655,
      "balance_loss_clip": 1.03869891,
      "balance_loss_mlp": 1.009426,
      "epoch": 0.6257628137682248,
      "flos": 24024885373440.0,
      "grad_norm": 1.629642239136061,
      "language_loss": 0.85067582,
      "learning_rate": 1.2977139713289398e-06,
      "loss": 0.8694725,
      "num_input_tokens_seen": 224234170,
      "step": 10408,
      "time_per_iteration": 2.5103774070739746
    },
    {
      "auxiliary_loss_clip": 0.01085378,
      "auxiliary_loss_mlp": 0.00787292,
      "balance_loss_clip": 1.03668761,
      "balance_loss_mlp": 1.01578641,
      "epoch": 0.6258229370208929,
      "flos": 20850956830080.0,
      "grad_norm": 1.6257219481484102,
      "language_loss": 0.79731441,
      "learning_rate": 1.2973493231693699e-06,
      "loss": 0.81604111,
      "num_input_tokens_seen": 224253115,
      "step": 10409,
      "time_per_iteration": 2.540896415710449
    },
    {
      "auxiliary_loss_clip": 0.01086161,
      "auxiliary_loss_mlp": 0.0103257,
      "balance_loss_clip": 1.03662527,
      "balance_loss_mlp": 1.02032137,
      "epoch": 0.6258830602735608,
      "flos": 22231542021120.0,
      "grad_norm": 2.811501506232355,
      "language_loss": 0.69803405,
      "learning_rate": 1.2969847016543845e-06,
      "loss": 0.71922135,
      "num_input_tokens_seen": 224271375,
      "step": 10410,
      "time_per_iteration": 2.5354232788085938
    },
    {
      "auxiliary_loss_clip": 0.01061325,
      "auxiliary_loss_mlp": 0.01024336,
      "balance_loss_clip": 1.03703201,
      "balance_loss_mlp": 1.01270103,
      "epoch": 0.6259431835262288,
      "flos": 25076887925760.0,
      "grad_norm": 1.8128142464813768,
      "language_loss": 0.67567825,
      "learning_rate": 1.2966201067978086e-06,
      "loss": 0.69653487,
      "num_input_tokens_seen": 224290315,
      "step": 10411,
      "time_per_iteration": 2.6177170276641846
    },
    {
      "auxiliary_loss_clip": 0.01062388,
      "auxiliary_loss_mlp": 0.01030892,
      "balance_loss_clip": 1.03773892,
      "balance_loss_mlp": 1.01920366,
      "epoch": 0.6260033067788967,
      "flos": 28252288926720.0,
      "grad_norm": 6.259631450100867,
      "language_loss": 0.69355655,
      "learning_rate": 1.2962555386134702e-06,
      "loss": 0.7144894,
      "num_input_tokens_seen": 224310545,
      "step": 10412,
      "time_per_iteration": 2.6401069164276123
    },
    {
      "auxiliary_loss_clip": 0.01076364,
      "auxiliary_loss_mlp": 0.01038343,
      "balance_loss_clip": 1.03599548,
      "balance_loss_mlp": 1.02546906,
      "epoch": 0.6260634300315647,
      "flos": 23367432787200.0,
      "grad_norm": 1.494274797240139,
      "language_loss": 0.69671708,
      "learning_rate": 1.2958909971151908e-06,
      "loss": 0.71786416,
      "num_input_tokens_seen": 224331115,
      "step": 10413,
      "time_per_iteration": 2.580209732055664
    },
    {
      "auxiliary_loss_clip": 0.01076207,
      "auxiliary_loss_mlp": 0.01033572,
      "balance_loss_clip": 1.03482604,
      "balance_loss_mlp": 1.01977921,
      "epoch": 0.6261235532842326,
      "flos": 18035308494720.0,
      "grad_norm": 2.4251597160614313,
      "language_loss": 0.81072265,
      "learning_rate": 1.295526482316796e-06,
      "loss": 0.83182049,
      "num_input_tokens_seen": 224347525,
      "step": 10414,
      "time_per_iteration": 2.4999582767486572
    },
    {
      "auxiliary_loss_clip": 0.0109864,
      "auxiliary_loss_mlp": 0.01033936,
      "balance_loss_clip": 1.04002595,
      "balance_loss_mlp": 1.02203941,
      "epoch": 0.6261836765369007,
      "flos": 22011265866240.0,
      "grad_norm": 1.6899991779662074,
      "language_loss": 0.74586475,
      "learning_rate": 1.2951619942321083e-06,
      "loss": 0.76719046,
      "num_input_tokens_seen": 224367045,
      "step": 10415,
      "time_per_iteration": 2.512517213821411
    },
    {
      "auxiliary_loss_clip": 0.01055623,
      "auxiliary_loss_mlp": 0.01028742,
      "balance_loss_clip": 1.03598559,
      "balance_loss_mlp": 1.0164634,
      "epoch": 0.6262437997895686,
      "flos": 24936010784640.0,
      "grad_norm": 1.5513972733717871,
      "language_loss": 0.74768817,
      "learning_rate": 1.2947975328749472e-06,
      "loss": 0.76853192,
      "num_input_tokens_seen": 224388860,
      "step": 10416,
      "time_per_iteration": 2.6437771320343018
    },
    {
      "auxiliary_loss_clip": 0.01081191,
      "auxiliary_loss_mlp": 0.01028181,
      "balance_loss_clip": 1.04057097,
      "balance_loss_mlp": 1.01689196,
      "epoch": 0.6263039230422366,
      "flos": 31608428186880.0,
      "grad_norm": 1.6596885946855786,
      "language_loss": 0.84374994,
      "learning_rate": 1.2944330982591352e-06,
      "loss": 0.86484361,
      "num_input_tokens_seen": 224409645,
      "step": 10417,
      "time_per_iteration": 2.6428351402282715
    },
    {
      "auxiliary_loss_clip": 0.01098519,
      "auxiliary_loss_mlp": 0.01031505,
      "balance_loss_clip": 1.03787374,
      "balance_loss_mlp": 1.01870775,
      "epoch": 0.6263640462949046,
      "flos": 17639465639040.0,
      "grad_norm": 3.1389044322443667,
      "language_loss": 0.57318801,
      "learning_rate": 1.2940686903984904e-06,
      "loss": 0.59448826,
      "num_input_tokens_seen": 224428530,
      "step": 10418,
      "time_per_iteration": 2.4687135219573975
    },
    {
      "auxiliary_loss_clip": 0.01102639,
      "auxiliary_loss_mlp": 0.0103628,
      "balance_loss_clip": 1.03803611,
      "balance_loss_mlp": 1.02283931,
      "epoch": 0.6264241695475725,
      "flos": 19974951941760.0,
      "grad_norm": 1.8368170140089326,
      "language_loss": 0.84816325,
      "learning_rate": 1.2937043093068316e-06,
      "loss": 0.86955237,
      "num_input_tokens_seen": 224447175,
      "step": 10419,
      "time_per_iteration": 2.489936351776123
    },
    {
      "auxiliary_loss_clip": 0.01111679,
      "auxiliary_loss_mlp": 0.01029631,
      "balance_loss_clip": 1.04020083,
      "balance_loss_mlp": 1.01777017,
      "epoch": 0.6264842928002405,
      "flos": 27344323912320.0,
      "grad_norm": 1.4376554861546764,
      "language_loss": 0.64747703,
      "learning_rate": 1.2933399549979762e-06,
      "loss": 0.66889012,
      "num_input_tokens_seen": 224469445,
      "step": 10420,
      "time_per_iteration": 3.8574912548065186
    },
    {
      "auxiliary_loss_clip": 0.01070009,
      "auxiliary_loss_mlp": 0.01032825,
      "balance_loss_clip": 1.03726792,
      "balance_loss_mlp": 1.01945519,
      "epoch": 0.6265444160529084,
      "flos": 22997265177600.0,
      "grad_norm": 2.004632189187469,
      "language_loss": 0.86015034,
      "learning_rate": 1.292975627485741e-06,
      "loss": 0.88117862,
      "num_input_tokens_seen": 224486590,
      "step": 10421,
      "time_per_iteration": 2.548362970352173
    },
    {
      "auxiliary_loss_clip": 0.01072644,
      "auxiliary_loss_mlp": 0.01031616,
      "balance_loss_clip": 1.03743339,
      "balance_loss_mlp": 1.0198741,
      "epoch": 0.6266045393055765,
      "flos": 19938323047680.0,
      "grad_norm": 2.094769272788569,
      "language_loss": 0.79985642,
      "learning_rate": 1.2926113267839403e-06,
      "loss": 0.82089901,
      "num_input_tokens_seen": 224502795,
      "step": 10422,
      "time_per_iteration": 2.5369112491607666
    },
    {
      "auxiliary_loss_clip": 0.01095666,
      "auxiliary_loss_mlp": 0.01025654,
      "balance_loss_clip": 1.03545022,
      "balance_loss_mlp": 1.01291609,
      "epoch": 0.6266646625582444,
      "flos": 24389091325440.0,
      "grad_norm": 1.9877085515847919,
      "language_loss": 0.7444396,
      "learning_rate": 1.292247052906389e-06,
      "loss": 0.76565278,
      "num_input_tokens_seen": 224522300,
      "step": 10423,
      "time_per_iteration": 2.5348217487335205
    },
    {
      "auxiliary_loss_clip": 0.01106551,
      "auxiliary_loss_mlp": 0.01027551,
      "balance_loss_clip": 1.03696132,
      "balance_loss_mlp": 1.01588047,
      "epoch": 0.6267247858109124,
      "flos": 14683802088960.0,
      "grad_norm": 2.7841963141948907,
      "language_loss": 0.77676523,
      "learning_rate": 1.2918828058669004e-06,
      "loss": 0.79810631,
      "num_input_tokens_seen": 224538260,
      "step": 10424,
      "time_per_iteration": 2.422861099243164
    },
    {
      "auxiliary_loss_clip": 0.01106234,
      "auxiliary_loss_mlp": 0.01032231,
      "balance_loss_clip": 1.03700781,
      "balance_loss_mlp": 1.01921892,
      "epoch": 0.6267849090635803,
      "flos": 24929977299840.0,
      "grad_norm": 1.8180902902581721,
      "language_loss": 0.69285065,
      "learning_rate": 1.2915185856792868e-06,
      "loss": 0.71423531,
      "num_input_tokens_seen": 224559155,
      "step": 10425,
      "time_per_iteration": 2.479146957397461
    },
    {
      "auxiliary_loss_clip": 0.01082064,
      "auxiliary_loss_mlp": 0.01030429,
      "balance_loss_clip": 1.03597403,
      "balance_loss_mlp": 1.01942599,
      "epoch": 0.6268450323162483,
      "flos": 25337851211520.0,
      "grad_norm": 1.611407812190046,
      "language_loss": 0.74569792,
      "learning_rate": 1.2911543923573598e-06,
      "loss": 0.76682281,
      "num_input_tokens_seen": 224578660,
      "step": 10426,
      "time_per_iteration": 2.555128812789917
    },
    {
      "auxiliary_loss_clip": 0.01098036,
      "auxiliary_loss_mlp": 0.00786314,
      "balance_loss_clip": 1.0368886,
      "balance_loss_mlp": 1.01085114,
      "epoch": 0.6269051555689162,
      "flos": 26177299032960.0,
      "grad_norm": 1.425805946791349,
      "language_loss": 0.80349433,
      "learning_rate": 1.290790225914929e-06,
      "loss": 0.82233787,
      "num_input_tokens_seen": 224599080,
      "step": 10427,
      "time_per_iteration": 2.553365468978882
    },
    {
      "auxiliary_loss_clip": 0.01076682,
      "auxiliary_loss_mlp": 0.01031944,
      "balance_loss_clip": 1.03792226,
      "balance_loss_mlp": 1.01962972,
      "epoch": 0.6269652788215843,
      "flos": 18256877539200.0,
      "grad_norm": 1.8475751978887245,
      "language_loss": 0.68124717,
      "learning_rate": 1.2904260863658034e-06,
      "loss": 0.70233345,
      "num_input_tokens_seen": 224614225,
      "step": 10428,
      "time_per_iteration": 2.507397174835205
    },
    {
      "auxiliary_loss_clip": 0.01067205,
      "auxiliary_loss_mlp": 0.01047759,
      "balance_loss_clip": 1.0371592,
      "balance_loss_mlp": 1.03386509,
      "epoch": 0.6270254020742522,
      "flos": 11765413877760.0,
      "grad_norm": 1.8234825614288965,
      "language_loss": 0.71407247,
      "learning_rate": 1.2900619737237928e-06,
      "loss": 0.7352221,
      "num_input_tokens_seen": 224632365,
      "step": 10429,
      "time_per_iteration": 2.5222201347351074
    },
    {
      "auxiliary_loss_clip": 0.01102614,
      "auxiliary_loss_mlp": 0.01034309,
      "balance_loss_clip": 1.0397048,
      "balance_loss_mlp": 1.02147579,
      "epoch": 0.6270855253269202,
      "flos": 23475631530240.0,
      "grad_norm": 1.743923439680762,
      "language_loss": 0.79471499,
      "learning_rate": 1.2896978880027023e-06,
      "loss": 0.81608421,
      "num_input_tokens_seen": 224651125,
      "step": 10430,
      "time_per_iteration": 3.9124057292938232
    },
    {
      "auxiliary_loss_clip": 0.01049955,
      "auxiliary_loss_mlp": 0.01004965,
      "balance_loss_clip": 1.02635908,
      "balance_loss_mlp": 1.00360036,
      "epoch": 0.6271456485795882,
      "flos": 70064520232320.0,
      "grad_norm": 0.7686165039481807,
      "language_loss": 0.59174538,
      "learning_rate": 1.2893338292163393e-06,
      "loss": 0.61229455,
      "num_input_tokens_seen": 224716115,
      "step": 10431,
      "time_per_iteration": 4.584550857543945
    },
    {
      "auxiliary_loss_clip": 0.01031006,
      "auxiliary_loss_mlp": 0.0100516,
      "balance_loss_clip": 1.02803683,
      "balance_loss_mlp": 1.00378323,
      "epoch": 0.6272057718322561,
      "flos": 65156718280320.0,
      "grad_norm": 0.8744097845670681,
      "language_loss": 0.63741279,
      "learning_rate": 1.2889697973785095e-06,
      "loss": 0.65777445,
      "num_input_tokens_seen": 224782930,
      "step": 10432,
      "time_per_iteration": 3.1971993446350098
    },
    {
      "auxiliary_loss_clip": 0.01084216,
      "auxiliary_loss_mlp": 0.01030319,
      "balance_loss_clip": 1.03753877,
      "balance_loss_mlp": 1.0195545,
      "epoch": 0.6272658950849241,
      "flos": 24389342720640.0,
      "grad_norm": 3.99249862163295,
      "language_loss": 0.6458472,
      "learning_rate": 1.2886057925030153e-06,
      "loss": 0.66699255,
      "num_input_tokens_seen": 224802010,
      "step": 10433,
      "time_per_iteration": 3.944110870361328
    },
    {
      "auxiliary_loss_clip": 0.01102605,
      "auxiliary_loss_mlp": 0.01031805,
      "balance_loss_clip": 1.03975785,
      "balance_loss_mlp": 1.01906776,
      "epoch": 0.627326018337592,
      "flos": 17966001202560.0,
      "grad_norm": 2.750243860524733,
      "language_loss": 0.61885941,
      "learning_rate": 1.2882418146036612e-06,
      "loss": 0.64020354,
      "num_input_tokens_seen": 224818875,
      "step": 10434,
      "time_per_iteration": 2.494591236114502
    },
    {
      "auxiliary_loss_clip": 0.01069389,
      "auxiliary_loss_mlp": 0.01025425,
      "balance_loss_clip": 1.03280759,
      "balance_loss_mlp": 1.01368928,
      "epoch": 0.6273861415902601,
      "flos": 20230097224320.0,
      "grad_norm": 1.6102089799984283,
      "language_loss": 0.84554607,
      "learning_rate": 1.2878778636942484e-06,
      "loss": 0.86649418,
      "num_input_tokens_seen": 224837790,
      "step": 10435,
      "time_per_iteration": 2.560455799102783
    },
    {
      "auxiliary_loss_clip": 0.01049343,
      "auxiliary_loss_mlp": 0.01002958,
      "balance_loss_clip": 1.02559829,
      "balance_loss_mlp": 1.00161076,
      "epoch": 0.627446264842928,
      "flos": 64953210798720.0,
      "grad_norm": 0.7287984806956278,
      "language_loss": 0.61533028,
      "learning_rate": 1.2875139397885786e-06,
      "loss": 0.63585329,
      "num_input_tokens_seen": 224899685,
      "step": 10436,
      "time_per_iteration": 3.0884416103363037
    },
    {
      "auxiliary_loss_clip": 0.01088045,
      "auxiliary_loss_mlp": 0.01037397,
      "balance_loss_clip": 1.04009748,
      "balance_loss_mlp": 1.0241468,
      "epoch": 0.627506388095596,
      "flos": 23584261236480.0,
      "grad_norm": 1.4383569300774937,
      "language_loss": 0.77256131,
      "learning_rate": 1.2871500429004523e-06,
      "loss": 0.79381573,
      "num_input_tokens_seen": 224918650,
      "step": 10437,
      "time_per_iteration": 2.5357086658477783
    },
    {
      "auxiliary_loss_clip": 0.0103856,
      "auxiliary_loss_mlp": 0.01002528,
      "balance_loss_clip": 1.02459991,
      "balance_loss_mlp": 1.00127625,
      "epoch": 0.6275665113482639,
      "flos": 67583631674880.0,
      "grad_norm": 0.8055904083249047,
      "language_loss": 0.54314667,
      "learning_rate": 1.2867861730436667e-06,
      "loss": 0.56355751,
      "num_input_tokens_seen": 224981575,
      "step": 10438,
      "time_per_iteration": 3.0434789657592773
    },
    {
      "auxiliary_loss_clip": 0.01057016,
      "auxiliary_loss_mlp": 0.0103828,
      "balance_loss_clip": 1.0337944,
      "balance_loss_mlp": 1.02538204,
      "epoch": 0.6276266346009319,
      "flos": 27636924101760.0,
      "grad_norm": 1.8268920442601184,
      "language_loss": 0.84416324,
      "learning_rate": 1.2864223302320214e-06,
      "loss": 0.86511624,
      "num_input_tokens_seen": 225000820,
      "step": 10439,
      "time_per_iteration": 2.6678171157836914
    },
    {
      "auxiliary_loss_clip": 0.01076378,
      "auxiliary_loss_mlp": 0.01040074,
      "balance_loss_clip": 1.03852093,
      "balance_loss_mlp": 1.02671027,
      "epoch": 0.6276867578535998,
      "flos": 22746142218240.0,
      "grad_norm": 2.508422080702262,
      "language_loss": 0.80865014,
      "learning_rate": 1.2860585144793128e-06,
      "loss": 0.82981455,
      "num_input_tokens_seen": 225017585,
      "step": 10440,
      "time_per_iteration": 2.555166244506836
    },
    {
      "auxiliary_loss_clip": 0.01055044,
      "auxiliary_loss_mlp": 0.01030056,
      "balance_loss_clip": 1.03748941,
      "balance_loss_mlp": 1.01893353,
      "epoch": 0.6277468811062679,
      "flos": 24644200694400.0,
      "grad_norm": 1.7098780243676146,
      "language_loss": 0.74343956,
      "learning_rate": 1.285694725799337e-06,
      "loss": 0.76429057,
      "num_input_tokens_seen": 225039085,
      "step": 10441,
      "time_per_iteration": 2.6546590328216553
    },
    {
      "auxiliary_loss_clip": 0.01091845,
      "auxiliary_loss_mlp": 0.01028767,
      "balance_loss_clip": 1.03727984,
      "balance_loss_mlp": 1.01571918,
      "epoch": 0.6278070043589358,
      "flos": 19678975873920.0,
      "grad_norm": 1.8383259840663142,
      "language_loss": 0.72230935,
      "learning_rate": 1.2853309642058884e-06,
      "loss": 0.74351549,
      "num_input_tokens_seen": 225058105,
      "step": 10442,
      "time_per_iteration": 2.509080171585083
    },
    {
      "auxiliary_loss_clip": 0.01075829,
      "auxiliary_loss_mlp": 0.01030231,
      "balance_loss_clip": 1.03729677,
      "balance_loss_mlp": 1.01780891,
      "epoch": 0.6278671276116038,
      "flos": 22121834906880.0,
      "grad_norm": 1.5202285859416114,
      "language_loss": 0.71443617,
      "learning_rate": 1.284967229712762e-06,
      "loss": 0.73549676,
      "num_input_tokens_seen": 225077605,
      "step": 10443,
      "time_per_iteration": 2.588299512863159
    },
    {
      "auxiliary_loss_clip": 0.01108643,
      "auxiliary_loss_mlp": 0.01029084,
      "balance_loss_clip": 1.03854513,
      "balance_loss_mlp": 1.01713276,
      "epoch": 0.6279272508642717,
      "flos": 23038562839680.0,
      "grad_norm": 2.14392193710865,
      "language_loss": 0.73154235,
      "learning_rate": 1.2846035223337492e-06,
      "loss": 0.75291967,
      "num_input_tokens_seen": 225097775,
      "step": 10444,
      "time_per_iteration": 2.492387294769287
    },
    {
      "auxiliary_loss_clip": 0.01071985,
      "auxiliary_loss_mlp": 0.01028078,
      "balance_loss_clip": 1.03842688,
      "balance_loss_mlp": 1.01588273,
      "epoch": 0.6279873741169397,
      "flos": 19824090819840.0,
      "grad_norm": 1.7962479354736811,
      "language_loss": 0.72045946,
      "learning_rate": 1.2842398420826423e-06,
      "loss": 0.74146008,
      "num_input_tokens_seen": 225115585,
      "step": 10445,
      "time_per_iteration": 2.553072929382324
    },
    {
      "auxiliary_loss_clip": 0.01094436,
      "auxiliary_loss_mlp": 0.01027553,
      "balance_loss_clip": 1.03948474,
      "balance_loss_mlp": 1.01539373,
      "epoch": 0.6280474973696077,
      "flos": 23915393740800.0,
      "grad_norm": 1.531478507658272,
      "language_loss": 0.69112146,
      "learning_rate": 1.2838761889732331e-06,
      "loss": 0.71234137,
      "num_input_tokens_seen": 225135575,
      "step": 10446,
      "time_per_iteration": 2.511483907699585
    },
    {
      "auxiliary_loss_clip": 0.01068177,
      "auxiliary_loss_mlp": 0.01029336,
      "balance_loss_clip": 1.03662562,
      "balance_loss_mlp": 1.01545405,
      "epoch": 0.6281076206222757,
      "flos": 17967976450560.0,
      "grad_norm": 1.776753116634411,
      "language_loss": 0.73856747,
      "learning_rate": 1.2835125630193102e-06,
      "loss": 0.7595427,
      "num_input_tokens_seen": 225154230,
      "step": 10447,
      "time_per_iteration": 2.5679781436920166
    },
    {
      "auxiliary_loss_clip": 0.01043792,
      "auxiliary_loss_mlp": 0.01006207,
      "balance_loss_clip": 1.02967763,
      "balance_loss_mlp": 1.00492597,
      "epoch": 0.6281677438749437,
      "flos": 66778370622720.0,
      "grad_norm": 0.6869506431524492,
      "language_loss": 0.5236814,
      "learning_rate": 1.2831489642346626e-06,
      "loss": 0.54418141,
      "num_input_tokens_seen": 225213650,
      "step": 10448,
      "time_per_iteration": 2.990557909011841
    },
    {
      "auxiliary_loss_clip": 0.01085086,
      "auxiliary_loss_mlp": 0.01048298,
      "balance_loss_clip": 1.0377562,
      "balance_loss_mlp": 1.03411841,
      "epoch": 0.6282278671276116,
      "flos": 11656173640320.0,
      "grad_norm": 1.8944972927434274,
      "language_loss": 0.90761006,
      "learning_rate": 1.282785392633079e-06,
      "loss": 0.92894393,
      "num_input_tokens_seen": 225230135,
      "step": 10449,
      "time_per_iteration": 2.5036983489990234
    },
    {
      "auxiliary_loss_clip": 0.01106798,
      "auxiliary_loss_mlp": 0.01028375,
      "balance_loss_clip": 1.03751111,
      "balance_loss_mlp": 1.01682949,
      "epoch": 0.6282879903802796,
      "flos": 42741597847680.0,
      "grad_norm": 1.4834722397891185,
      "language_loss": 0.60122532,
      "learning_rate": 1.2824218482283438e-06,
      "loss": 0.62257707,
      "num_input_tokens_seen": 225253520,
      "step": 10450,
      "time_per_iteration": 2.639711618423462
    },
    {
      "auxiliary_loss_clip": 0.01081561,
      "auxiliary_loss_mlp": 0.01028382,
      "balance_loss_clip": 1.0379591,
      "balance_loss_mlp": 1.01666951,
      "epoch": 0.6283481136329475,
      "flos": 20009210538240.0,
      "grad_norm": 1.588551342234234,
      "language_loss": 0.76779187,
      "learning_rate": 1.2820583310342452e-06,
      "loss": 0.78889132,
      "num_input_tokens_seen": 225272460,
      "step": 10451,
      "time_per_iteration": 2.532128095626831
    },
    {
      "auxiliary_loss_clip": 0.01083765,
      "auxiliary_loss_mlp": 0.01031223,
      "balance_loss_clip": 1.03846264,
      "balance_loss_mlp": 1.01854515,
      "epoch": 0.6284082368856155,
      "flos": 21904431840000.0,
      "grad_norm": 2.194690266465946,
      "language_loss": 0.77720118,
      "learning_rate": 1.281694841064566e-06,
      "loss": 0.79835111,
      "num_input_tokens_seen": 225291700,
      "step": 10452,
      "time_per_iteration": 2.523129940032959
    },
    {
      "auxiliary_loss_clip": 0.01078399,
      "auxiliary_loss_mlp": 0.01030085,
      "balance_loss_clip": 1.03804016,
      "balance_loss_mlp": 1.01728773,
      "epoch": 0.6284683601382834,
      "flos": 25484187219840.0,
      "grad_norm": 1.6950617626260656,
      "language_loss": 0.72531033,
      "learning_rate": 1.2813313783330904e-06,
      "loss": 0.74639511,
      "num_input_tokens_seen": 225311470,
      "step": 10453,
      "time_per_iteration": 2.6004416942596436
    },
    {
      "auxiliary_loss_clip": 0.01050795,
      "auxiliary_loss_mlp": 0.01034909,
      "balance_loss_clip": 1.03386927,
      "balance_loss_mlp": 1.01974547,
      "epoch": 0.6285284833909515,
      "flos": 16538695395840.0,
      "grad_norm": 1.6545657219540946,
      "language_loss": 0.80570495,
      "learning_rate": 1.2809679428536013e-06,
      "loss": 0.82656199,
      "num_input_tokens_seen": 225328385,
      "step": 10454,
      "time_per_iteration": 2.5904788970947266
    },
    {
      "auxiliary_loss_clip": 0.01074603,
      "auxiliary_loss_mlp": 0.01029196,
      "balance_loss_clip": 1.03764808,
      "balance_loss_mlp": 1.01815748,
      "epoch": 0.6285886066436194,
      "flos": 22820692896000.0,
      "grad_norm": 2.07112487949402,
      "language_loss": 0.81907523,
      "learning_rate": 1.2806045346398792e-06,
      "loss": 0.84011316,
      "num_input_tokens_seen": 225348415,
      "step": 10455,
      "time_per_iteration": 2.575608730316162
    },
    {
      "auxiliary_loss_clip": 0.01060475,
      "auxiliary_loss_mlp": 0.00784683,
      "balance_loss_clip": 1.03584111,
      "balance_loss_mlp": 1.00549078,
      "epoch": 0.6286487298962874,
      "flos": 24715734629760.0,
      "grad_norm": 1.541619301668581,
      "language_loss": 0.81592834,
      "learning_rate": 1.280241153705706e-06,
      "loss": 0.83437991,
      "num_input_tokens_seen": 225367740,
      "step": 10456,
      "time_per_iteration": 2.6370105743408203
    },
    {
      "auxiliary_loss_clip": 0.01091321,
      "auxiliary_loss_mlp": 0.01029678,
      "balance_loss_clip": 1.0441339,
      "balance_loss_mlp": 1.01692295,
      "epoch": 0.6287088531489553,
      "flos": 20740818752640.0,
      "grad_norm": 1.5018675779804156,
      "language_loss": 0.72194254,
      "learning_rate": 1.27987780006486e-06,
      "loss": 0.74315262,
      "num_input_tokens_seen": 225388405,
      "step": 10457,
      "time_per_iteration": 2.5565288066864014
    },
    {
      "auxiliary_loss_clip": 0.01101149,
      "auxiliary_loss_mlp": 0.01028119,
      "balance_loss_clip": 1.03623724,
      "balance_loss_mlp": 1.01517272,
      "epoch": 0.6287689764016233,
      "flos": 23070630706560.0,
      "grad_norm": 1.8241817018671518,
      "language_loss": 0.79583764,
      "learning_rate": 1.2795144737311202e-06,
      "loss": 0.81713033,
      "num_input_tokens_seen": 225408360,
      "step": 10458,
      "time_per_iteration": 2.489292621612549
    },
    {
      "auxiliary_loss_clip": 0.01099832,
      "auxiliary_loss_mlp": 0.01032788,
      "balance_loss_clip": 1.03849244,
      "balance_loss_mlp": 1.02071261,
      "epoch": 0.6288290996542913,
      "flos": 32233669251840.0,
      "grad_norm": 1.4902561451532517,
      "language_loss": 0.61248761,
      "learning_rate": 1.2791511747182635e-06,
      "loss": 0.63381386,
      "num_input_tokens_seen": 225431310,
      "step": 10459,
      "time_per_iteration": 3.933231830596924
    },
    {
      "auxiliary_loss_clip": 0.01083965,
      "auxiliary_loss_mlp": 0.01029117,
      "balance_loss_clip": 1.03682375,
      "balance_loss_mlp": 1.01722038,
      "epoch": 0.6288892229069593,
      "flos": 24641327606400.0,
      "grad_norm": 1.800175076774128,
      "language_loss": 0.78969765,
      "learning_rate": 1.2787879030400666e-06,
      "loss": 0.81082845,
      "num_input_tokens_seen": 225450385,
      "step": 10460,
      "time_per_iteration": 2.5650625228881836
    },
    {
      "auxiliary_loss_clip": 0.01067188,
      "auxiliary_loss_mlp": 0.01027537,
      "balance_loss_clip": 1.03648257,
      "balance_loss_mlp": 1.01518083,
      "epoch": 0.6289493461596273,
      "flos": 17858341163520.0,
      "grad_norm": 1.752754339269406,
      "language_loss": 0.74070787,
      "learning_rate": 1.2784246587103047e-06,
      "loss": 0.76165515,
      "num_input_tokens_seen": 225467325,
      "step": 10461,
      "time_per_iteration": 2.5390443801879883
    },
    {
      "auxiliary_loss_clip": 0.01084882,
      "auxiliary_loss_mlp": 0.01034855,
      "balance_loss_clip": 1.03583491,
      "balance_loss_mlp": 1.02319658,
      "epoch": 0.6290094694122952,
      "flos": 22345379199360.0,
      "grad_norm": 1.6507179811252104,
      "language_loss": 0.70163506,
      "learning_rate": 1.2780614417427523e-06,
      "loss": 0.72283244,
      "num_input_tokens_seen": 225487370,
      "step": 10462,
      "time_per_iteration": 2.543658971786499
    },
    {
      "auxiliary_loss_clip": 0.01103109,
      "auxiliary_loss_mlp": 0.01028842,
      "balance_loss_clip": 1.03808475,
      "balance_loss_mlp": 1.01845336,
      "epoch": 0.6290695926649632,
      "flos": 28402431776640.0,
      "grad_norm": 1.841764030354768,
      "language_loss": 0.72116363,
      "learning_rate": 1.2776982521511821e-06,
      "loss": 0.74248308,
      "num_input_tokens_seen": 225506915,
      "step": 10463,
      "time_per_iteration": 2.508042335510254
    },
    {
      "auxiliary_loss_clip": 0.01087609,
      "auxiliary_loss_mlp": 0.01032292,
      "balance_loss_clip": 1.04092717,
      "balance_loss_mlp": 1.02065718,
      "epoch": 0.6291297159176311,
      "flos": 21505464501120.0,
      "grad_norm": 1.6381239241135095,
      "language_loss": 0.72186029,
      "learning_rate": 1.2773350899493665e-06,
      "loss": 0.74305928,
      "num_input_tokens_seen": 225525670,
      "step": 10464,
      "time_per_iteration": 2.5293281078338623
    },
    {
      "auxiliary_loss_clip": 0.01085454,
      "auxiliary_loss_mlp": 0.01028171,
      "balance_loss_clip": 1.03850806,
      "balance_loss_mlp": 1.01698923,
      "epoch": 0.6291898391702991,
      "flos": 12203308581120.0,
      "grad_norm": 1.6177134344889372,
      "language_loss": 0.69458163,
      "learning_rate": 1.2769719551510768e-06,
      "loss": 0.71571791,
      "num_input_tokens_seen": 225542235,
      "step": 10465,
      "time_per_iteration": 2.492640972137451
    },
    {
      "auxiliary_loss_clip": 0.010391,
      "auxiliary_loss_mlp": 0.0100376,
      "balance_loss_clip": 1.02578509,
      "balance_loss_mlp": 1.00250244,
      "epoch": 0.629249962422967,
      "flos": 69299479434240.0,
      "grad_norm": 0.7009259653390438,
      "language_loss": 0.59794044,
      "learning_rate": 1.2766088477700832e-06,
      "loss": 0.61836898,
      "num_input_tokens_seen": 225607185,
      "step": 10466,
      "time_per_iteration": 3.272179365158081
    },
    {
      "auxiliary_loss_clip": 0.01066533,
      "auxiliary_loss_mlp": 0.01029548,
      "balance_loss_clip": 1.03397846,
      "balance_loss_mlp": 1.01871216,
      "epoch": 0.6293100856756351,
      "flos": 40077888042240.0,
      "grad_norm": 2.393849595449034,
      "language_loss": 0.64921367,
      "learning_rate": 1.276245767820154e-06,
      "loss": 0.67017448,
      "num_input_tokens_seen": 225628785,
      "step": 10467,
      "time_per_iteration": 2.7018699645996094
    },
    {
      "auxiliary_loss_clip": 0.01019268,
      "auxiliary_loss_mlp": 0.01002856,
      "balance_loss_clip": 1.01706111,
      "balance_loss_mlp": 1.00156808,
      "epoch": 0.629370208928303,
      "flos": 67501108177920.0,
      "grad_norm": 0.7942453267081774,
      "language_loss": 0.56899703,
      "learning_rate": 1.2758827153150586e-06,
      "loss": 0.58921826,
      "num_input_tokens_seen": 225678980,
      "step": 10468,
      "time_per_iteration": 2.908304452896118
    },
    {
      "auxiliary_loss_clip": 0.01008467,
      "auxiliary_loss_mlp": 0.01000136,
      "balance_loss_clip": 1.02712178,
      "balance_loss_mlp": 0.99870509,
      "epoch": 0.629430332180971,
      "flos": 60660450449280.0,
      "grad_norm": 0.7405193155279937,
      "language_loss": 0.58006585,
      "learning_rate": 1.2755196902685626e-06,
      "loss": 0.6001519,
      "num_input_tokens_seen": 225740295,
      "step": 10469,
      "time_per_iteration": 5.97453761100769
    },
    {
      "auxiliary_loss_clip": 0.01036168,
      "auxiliary_loss_mlp": 0.00999944,
      "balance_loss_clip": 1.02890873,
      "balance_loss_mlp": 0.99828738,
      "epoch": 0.6294904554336389,
      "flos": 66869764778880.0,
      "grad_norm": 0.6767443153658657,
      "language_loss": 0.52137339,
      "learning_rate": 1.2751566926944329e-06,
      "loss": 0.54173458,
      "num_input_tokens_seen": 225805615,
      "step": 10470,
      "time_per_iteration": 3.2685842514038086
    },
    {
      "auxiliary_loss_clip": 0.01093847,
      "auxiliary_loss_mlp": 0.0103005,
      "balance_loss_clip": 1.04072762,
      "balance_loss_mlp": 1.01885629,
      "epoch": 0.6295505786863069,
      "flos": 42522794150400.0,
      "grad_norm": 1.9510836308884505,
      "language_loss": 0.74669957,
      "learning_rate": 1.2747937226064342e-06,
      "loss": 0.76793861,
      "num_input_tokens_seen": 225826585,
      "step": 10471,
      "time_per_iteration": 2.686530590057373
    },
    {
      "auxiliary_loss_clip": 0.01080681,
      "auxiliary_loss_mlp": 0.01030393,
      "balance_loss_clip": 1.04161811,
      "balance_loss_mlp": 1.01884758,
      "epoch": 0.629610701938975,
      "flos": 17384140788480.0,
      "grad_norm": 1.7565582357091873,
      "language_loss": 0.62086272,
      "learning_rate": 1.2744307800183297e-06,
      "loss": 0.6419735,
      "num_input_tokens_seen": 225844095,
      "step": 10472,
      "time_per_iteration": 3.914329767227173
    },
    {
      "auxiliary_loss_clip": 0.01111821,
      "auxiliary_loss_mlp": 0.01030811,
      "balance_loss_clip": 1.03961885,
      "balance_loss_mlp": 1.01852083,
      "epoch": 0.6296708251916429,
      "flos": 24242934885120.0,
      "grad_norm": 1.6115408607331574,
      "language_loss": 0.69474196,
      "learning_rate": 1.2740678649438828e-06,
      "loss": 0.71616828,
      "num_input_tokens_seen": 225864310,
      "step": 10473,
      "time_per_iteration": 2.497291088104248
    },
    {
      "auxiliary_loss_clip": 0.0108226,
      "auxiliary_loss_mlp": 0.01029698,
      "balance_loss_clip": 1.03530264,
      "balance_loss_mlp": 1.01836729,
      "epoch": 0.6297309484443109,
      "flos": 19278536077440.0,
      "grad_norm": 1.5506833814762324,
      "language_loss": 0.7438013,
      "learning_rate": 1.2737049773968554e-06,
      "loss": 0.76492083,
      "num_input_tokens_seen": 225883830,
      "step": 10474,
      "time_per_iteration": 2.5240895748138428
    },
    {
      "auxiliary_loss_clip": 0.01084032,
      "auxiliary_loss_mlp": 0.00784835,
      "balance_loss_clip": 1.036659,
      "balance_loss_mlp": 1.01253259,
      "epoch": 0.6297910716969788,
      "flos": 30662685043200.0,
      "grad_norm": 1.4450556041237104,
      "language_loss": 0.66291058,
      "learning_rate": 1.2733421173910081e-06,
      "loss": 0.68159926,
      "num_input_tokens_seen": 225905755,
      "step": 10475,
      "time_per_iteration": 2.6339662075042725
    },
    {
      "auxiliary_loss_clip": 0.01057661,
      "auxiliary_loss_mlp": 0.01029973,
      "balance_loss_clip": 1.03506255,
      "balance_loss_mlp": 1.01895189,
      "epoch": 0.6298511949496468,
      "flos": 14423018371200.0,
      "grad_norm": 1.7687442924172339,
      "language_loss": 0.9007566,
      "learning_rate": 1.272979284940101e-06,
      "loss": 0.92163295,
      "num_input_tokens_seen": 225922155,
      "step": 10476,
      "time_per_iteration": 2.537750005722046
    },
    {
      "auxiliary_loss_clip": 0.01105513,
      "auxiliary_loss_mlp": 0.01036464,
      "balance_loss_clip": 1.03718662,
      "balance_loss_mlp": 1.02496672,
      "epoch": 0.6299113182023147,
      "flos": 23514163845120.0,
      "grad_norm": 1.7832388501654397,
      "language_loss": 0.74788046,
      "learning_rate": 1.2726164800578913e-06,
      "loss": 0.76930022,
      "num_input_tokens_seen": 225941060,
      "step": 10477,
      "time_per_iteration": 2.4685287475585938
    },
    {
      "auxiliary_loss_clip": 0.01093125,
      "auxiliary_loss_mlp": 0.01029171,
      "balance_loss_clip": 1.03766918,
      "balance_loss_mlp": 1.01692247,
      "epoch": 0.6299714414549827,
      "flos": 22674500542080.0,
      "grad_norm": 1.6778968843349085,
      "language_loss": 0.70425665,
      "learning_rate": 1.272253702758138e-06,
      "loss": 0.72547966,
      "num_input_tokens_seen": 225960870,
      "step": 10478,
      "time_per_iteration": 2.4794669151306152
    },
    {
      "auxiliary_loss_clip": 0.01101266,
      "auxiliary_loss_mlp": 0.01029021,
      "balance_loss_clip": 1.03814077,
      "balance_loss_mlp": 1.01621771,
      "epoch": 0.6300315647076506,
      "flos": 14501735026560.0,
      "grad_norm": 2.7074303051437973,
      "language_loss": 0.67291868,
      "learning_rate": 1.2718909530545974e-06,
      "loss": 0.69422162,
      "num_input_tokens_seen": 225977895,
      "step": 10479,
      "time_per_iteration": 2.4521307945251465
    },
    {
      "auxiliary_loss_clip": 0.0108355,
      "auxiliary_loss_mlp": 0.00785358,
      "balance_loss_clip": 1.0402422,
      "balance_loss_mlp": 1.01043844,
      "epoch": 0.6300916879603187,
      "flos": 21871681614720.0,
      "grad_norm": 1.8990660902812975,
      "language_loss": 0.73871839,
      "learning_rate": 1.2715282309610245e-06,
      "loss": 0.75740749,
      "num_input_tokens_seen": 225997835,
      "step": 10480,
      "time_per_iteration": 2.520901679992676
    },
    {
      "auxiliary_loss_clip": 0.01095591,
      "auxiliary_loss_mlp": 0.01034103,
      "balance_loss_clip": 1.036443,
      "balance_loss_mlp": 1.02116251,
      "epoch": 0.6301518112129866,
      "flos": 21834047139840.0,
      "grad_norm": 1.9568995595898655,
      "language_loss": 0.78821027,
      "learning_rate": 1.2711655364911744e-06,
      "loss": 0.80950725,
      "num_input_tokens_seen": 226017620,
      "step": 10481,
      "time_per_iteration": 2.484386920928955
    },
    {
      "auxiliary_loss_clip": 0.01032231,
      "auxiliary_loss_mlp": 0.01010949,
      "balance_loss_clip": 1.02637422,
      "balance_loss_mlp": 1.00939965,
      "epoch": 0.6302119344656546,
      "flos": 44334237957120.0,
      "grad_norm": 0.8954171672310948,
      "language_loss": 0.6186353,
      "learning_rate": 1.2708028696588e-06,
      "loss": 0.63906717,
      "num_input_tokens_seen": 226068755,
      "step": 10482,
      "time_per_iteration": 2.8795056343078613
    },
    {
      "auxiliary_loss_clip": 0.01101226,
      "auxiliary_loss_mlp": 0.01032438,
      "balance_loss_clip": 1.03761482,
      "balance_loss_mlp": 1.01893759,
      "epoch": 0.6302720577183225,
      "flos": 11217919800960.0,
      "grad_norm": 2.035149665709166,
      "language_loss": 0.82634151,
      "learning_rate": 1.2704402304776541e-06,
      "loss": 0.84767812,
      "num_input_tokens_seen": 226084395,
      "step": 10483,
      "time_per_iteration": 2.448918581008911
    },
    {
      "auxiliary_loss_clip": 0.01086029,
      "auxiliary_loss_mlp": 0.0103189,
      "balance_loss_clip": 1.03589058,
      "balance_loss_mlp": 1.02026749,
      "epoch": 0.6303321809709905,
      "flos": 27964932122880.0,
      "grad_norm": 1.4582160143364844,
      "language_loss": 0.72521245,
      "learning_rate": 1.270077618961487e-06,
      "loss": 0.74639165,
      "num_input_tokens_seen": 226105890,
      "step": 10484,
      "time_per_iteration": 2.530829906463623
    },
    {
      "auxiliary_loss_clip": 0.01076054,
      "auxiliary_loss_mlp": 0.01028038,
      "balance_loss_clip": 1.03542852,
      "balance_loss_mlp": 1.0160991,
      "epoch": 0.6303923042236586,
      "flos": 28220759763840.0,
      "grad_norm": 2.1379980548812685,
      "language_loss": 0.74501657,
      "learning_rate": 1.2697150351240506e-06,
      "loss": 0.76605749,
      "num_input_tokens_seen": 226126760,
      "step": 10485,
      "time_per_iteration": 2.630833148956299
    },
    {
      "auxiliary_loss_clip": 0.01083941,
      "auxiliary_loss_mlp": 0.00784366,
      "balance_loss_clip": 1.03724194,
      "balance_loss_mlp": 1.00880313,
      "epoch": 0.6304524274763265,
      "flos": 27631034271360.0,
      "grad_norm": 1.9531258039914436,
      "language_loss": 0.81068009,
      "learning_rate": 1.269352478979093e-06,
      "loss": 0.82936317,
      "num_input_tokens_seen": 226147315,
      "step": 10486,
      "time_per_iteration": 2.5729544162750244
    },
    {
      "auxiliary_loss_clip": 0.01082417,
      "auxiliary_loss_mlp": 0.01034659,
      "balance_loss_clip": 1.03560126,
      "balance_loss_mlp": 1.02316141,
      "epoch": 0.6305125507289945,
      "flos": 17311313963520.0,
      "grad_norm": 1.6826916855888017,
      "language_loss": 0.63334012,
      "learning_rate": 1.2689899505403628e-06,
      "loss": 0.65451086,
      "num_input_tokens_seen": 226165935,
      "step": 10487,
      "time_per_iteration": 2.514842987060547
    },
    {
      "auxiliary_loss_clip": 0.01108009,
      "auxiliary_loss_mlp": 0.01038306,
      "balance_loss_clip": 1.03787124,
      "balance_loss_mlp": 1.02642083,
      "epoch": 0.6305726739816624,
      "flos": 25808280658560.0,
      "grad_norm": 1.5504982762983408,
      "language_loss": 0.6718967,
      "learning_rate": 1.2686274498216065e-06,
      "loss": 0.69335985,
      "num_input_tokens_seen": 226186890,
      "step": 10488,
      "time_per_iteration": 2.498486280441284
    },
    {
      "auxiliary_loss_clip": 0.0108743,
      "auxiliary_loss_mlp": 0.01031539,
      "balance_loss_clip": 1.03654432,
      "balance_loss_mlp": 1.01993954,
      "epoch": 0.6306327972343304,
      "flos": 21797454159360.0,
      "grad_norm": 2.094565465839246,
      "language_loss": 0.6705246,
      "learning_rate": 1.2682649768365706e-06,
      "loss": 0.69171429,
      "num_input_tokens_seen": 226206710,
      "step": 10489,
      "time_per_iteration": 2.5409953594207764
    },
    {
      "auxiliary_loss_clip": 0.01072758,
      "auxiliary_loss_mlp": 0.01041532,
      "balance_loss_clip": 1.03659379,
      "balance_loss_mlp": 1.02753639,
      "epoch": 0.6306929204869983,
      "flos": 20777375819520.0,
      "grad_norm": 1.7948449882716373,
      "language_loss": 0.69217938,
      "learning_rate": 1.2679025315990007e-06,
      "loss": 0.71332228,
      "num_input_tokens_seen": 226225565,
      "step": 10490,
      "time_per_iteration": 2.5488736629486084
    },
    {
      "auxiliary_loss_clip": 0.01086427,
      "auxiliary_loss_mlp": 0.01035201,
      "balance_loss_clip": 1.0366075,
      "balance_loss_mlp": 1.02272558,
      "epoch": 0.6307530437396663,
      "flos": 23654214973440.0,
      "grad_norm": 1.825427942427149,
      "language_loss": 0.78223962,
      "learning_rate": 1.2675401141226393e-06,
      "loss": 0.80345595,
      "num_input_tokens_seen": 226243680,
      "step": 10491,
      "time_per_iteration": 2.5528132915496826
    },
    {
      "auxiliary_loss_clip": 0.01083466,
      "auxiliary_loss_mlp": 0.01035587,
      "balance_loss_clip": 1.03730237,
      "balance_loss_mlp": 1.02329659,
      "epoch": 0.6308131669923343,
      "flos": 24719002767360.0,
      "grad_norm": 1.8944825292091856,
      "language_loss": 0.55792022,
      "learning_rate": 1.2671777244212308e-06,
      "loss": 0.57911074,
      "num_input_tokens_seen": 226264345,
      "step": 10492,
      "time_per_iteration": 2.5597493648529053
    },
    {
      "auxiliary_loss_clip": 0.01110738,
      "auxiliary_loss_mlp": 0.0104064,
      "balance_loss_clip": 1.03852737,
      "balance_loss_mlp": 1.0273962,
      "epoch": 0.6308732902450023,
      "flos": 22565403959040.0,
      "grad_norm": 1.8615324605616275,
      "language_loss": 0.64181936,
      "learning_rate": 1.2668153625085168e-06,
      "loss": 0.66333318,
      "num_input_tokens_seen": 226283165,
      "step": 10493,
      "time_per_iteration": 2.48433518409729
    },
    {
      "auxiliary_loss_clip": 0.01073759,
      "auxiliary_loss_mlp": 0.01029784,
      "balance_loss_clip": 1.03834248,
      "balance_loss_mlp": 1.01732635,
      "epoch": 0.6309334134976702,
      "flos": 24644200694400.0,
      "grad_norm": 1.4202608420929337,
      "language_loss": 0.82673806,
      "learning_rate": 1.2664530283982367e-06,
      "loss": 0.84777343,
      "num_input_tokens_seen": 226304080,
      "step": 10494,
      "time_per_iteration": 2.594668388366699
    },
    {
      "auxiliary_loss_clip": 0.01091924,
      "auxiliary_loss_mlp": 0.01033698,
      "balance_loss_clip": 1.03738141,
      "balance_loss_mlp": 1.02078748,
      "epoch": 0.6309935367503382,
      "flos": 41427949651200.0,
      "grad_norm": 3.434646394729339,
      "language_loss": 0.79658103,
      "learning_rate": 1.2660907221041317e-06,
      "loss": 0.8178373,
      "num_input_tokens_seen": 226325925,
      "step": 10495,
      "time_per_iteration": 2.6996254920959473
    },
    {
      "auxiliary_loss_clip": 0.01081919,
      "auxiliary_loss_mlp": 0.01033486,
      "balance_loss_clip": 1.03627288,
      "balance_loss_mlp": 1.02038503,
      "epoch": 0.6310536600030061,
      "flos": 15118931445120.0,
      "grad_norm": 1.7743230445697442,
      "language_loss": 0.7063098,
      "learning_rate": 1.2657284436399403e-06,
      "loss": 0.72746384,
      "num_input_tokens_seen": 226344190,
      "step": 10496,
      "time_per_iteration": 2.5031557083129883
    },
    {
      "auxiliary_loss_clip": 0.01090348,
      "auxiliary_loss_mlp": 0.01036718,
      "balance_loss_clip": 1.03776622,
      "balance_loss_mlp": 1.02354538,
      "epoch": 0.6311137832556741,
      "flos": 15231619388160.0,
      "grad_norm": 1.9867693697785116,
      "language_loss": 0.79791808,
      "learning_rate": 1.2653661930193997e-06,
      "loss": 0.81918871,
      "num_input_tokens_seen": 226361520,
      "step": 10497,
      "time_per_iteration": 2.5074574947357178
    },
    {
      "auxiliary_loss_clip": 0.0107574,
      "auxiliary_loss_mlp": 0.01032038,
      "balance_loss_clip": 1.03615975,
      "balance_loss_mlp": 1.02045727,
      "epoch": 0.6311739065083422,
      "flos": 22018664067840.0,
      "grad_norm": 1.9590378822756374,
      "language_loss": 0.7390185,
      "learning_rate": 1.265003970256247e-06,
      "loss": 0.76009625,
      "num_input_tokens_seen": 226381920,
      "step": 10498,
      "time_per_iteration": 3.965893268585205
    },
    {
      "auxiliary_loss_clip": 0.01098145,
      "auxiliary_loss_mlp": 0.01028761,
      "balance_loss_clip": 1.03659105,
      "balance_loss_mlp": 1.01632166,
      "epoch": 0.6312340297610101,
      "flos": 22710770300160.0,
      "grad_norm": 2.1780558938648276,
      "language_loss": 0.69721425,
      "learning_rate": 1.264641775364217e-06,
      "loss": 0.71848333,
      "num_input_tokens_seen": 226400035,
      "step": 10499,
      "time_per_iteration": 2.488330125808716
    },
    {
      "auxiliary_loss_clip": 0.01097316,
      "auxiliary_loss_mlp": 0.01040215,
      "balance_loss_clip": 1.03871763,
      "balance_loss_mlp": 1.02818084,
      "epoch": 0.6312941530136781,
      "flos": 24280102483200.0,
      "grad_norm": 1.69885770270483,
      "language_loss": 0.7015202,
      "learning_rate": 1.2642796083570448e-06,
      "loss": 0.7228955,
      "num_input_tokens_seen": 226418280,
      "step": 10500,
      "time_per_iteration": 2.520646810531616
    },
    {
      "auxiliary_loss_clip": 0.01110193,
      "auxiliary_loss_mlp": 0.01031497,
      "balance_loss_clip": 1.0394671,
      "balance_loss_mlp": 1.01985013,
      "epoch": 0.631354276266346,
      "flos": 21725956137600.0,
      "grad_norm": 1.7522789144702988,
      "language_loss": 0.7435782,
      "learning_rate": 1.2639174692484634e-06,
      "loss": 0.7649951,
      "num_input_tokens_seen": 226436650,
      "step": 10501,
      "time_per_iteration": 2.4520750045776367
    },
    {
      "auxiliary_loss_clip": 0.01096283,
      "auxiliary_loss_mlp": 0.00783534,
      "balance_loss_clip": 1.03789592,
      "balance_loss_mlp": 1.00925899,
      "epoch": 0.631414399519014,
      "flos": 24025100855040.0,
      "grad_norm": 1.6276588518667332,
      "language_loss": 0.75542927,
      "learning_rate": 1.2635553580522053e-06,
      "loss": 0.77422738,
      "num_input_tokens_seen": 226456275,
      "step": 10502,
      "time_per_iteration": 2.5318431854248047
    },
    {
      "auxiliary_loss_clip": 0.01101556,
      "auxiliary_loss_mlp": 0.01044399,
      "balance_loss_clip": 1.0386728,
      "balance_loss_mlp": 1.03218639,
      "epoch": 0.6314745227716819,
      "flos": 24315797623680.0,
      "grad_norm": 3.800702862327018,
      "language_loss": 0.85384303,
      "learning_rate": 1.2631932747820022e-06,
      "loss": 0.87530255,
      "num_input_tokens_seen": 226473610,
      "step": 10503,
      "time_per_iteration": 2.506032705307007
    },
    {
      "auxiliary_loss_clip": 0.01087914,
      "auxiliary_loss_mlp": 0.0103367,
      "balance_loss_clip": 1.04024482,
      "balance_loss_mlp": 1.02131426,
      "epoch": 0.6315346460243499,
      "flos": 23366391292800.0,
      "grad_norm": 1.9024358108121282,
      "language_loss": 0.86522186,
      "learning_rate": 1.2628312194515838e-06,
      "loss": 0.88643765,
      "num_input_tokens_seen": 226493665,
      "step": 10504,
      "time_per_iteration": 2.5335652828216553
    },
    {
      "auxiliary_loss_clip": 0.01078504,
      "auxiliary_loss_mlp": 0.01035085,
      "balance_loss_clip": 1.04135215,
      "balance_loss_mlp": 1.02256835,
      "epoch": 0.6315947692770179,
      "flos": 20260333497600.0,
      "grad_norm": 1.6384216761153434,
      "language_loss": 0.76190734,
      "learning_rate": 1.2624691920746793e-06,
      "loss": 0.78304321,
      "num_input_tokens_seen": 226511625,
      "step": 10505,
      "time_per_iteration": 2.5487937927246094
    },
    {
      "auxiliary_loss_clip": 0.01068355,
      "auxiliary_loss_mlp": 0.01032616,
      "balance_loss_clip": 1.03893781,
      "balance_loss_mlp": 1.01916361,
      "epoch": 0.6316548925296859,
      "flos": 25265850399360.0,
      "grad_norm": 1.8319588736677788,
      "language_loss": 0.81721222,
      "learning_rate": 1.2621071926650166e-06,
      "loss": 0.83822191,
      "num_input_tokens_seen": 226530085,
      "step": 10506,
      "time_per_iteration": 2.6095211505889893
    },
    {
      "auxiliary_loss_clip": 0.011114,
      "auxiliary_loss_mlp": 0.01036511,
      "balance_loss_clip": 1.0398252,
      "balance_loss_mlp": 1.02427411,
      "epoch": 0.6317150157823538,
      "flos": 22930579578240.0,
      "grad_norm": 2.01629539375655,
      "language_loss": 0.74459285,
      "learning_rate": 1.2617452212363238e-06,
      "loss": 0.76607203,
      "num_input_tokens_seen": 226548115,
      "step": 10507,
      "time_per_iteration": 3.8870036602020264
    },
    {
      "auxiliary_loss_clip": 0.01089849,
      "auxiliary_loss_mlp": 0.01036765,
      "balance_loss_clip": 1.04218125,
      "balance_loss_mlp": 1.0238905,
      "epoch": 0.6317751390350218,
      "flos": 22527051212160.0,
      "grad_norm": 1.6501918638687654,
      "language_loss": 0.67755139,
      "learning_rate": 1.2613832778023258e-06,
      "loss": 0.69881749,
      "num_input_tokens_seen": 226567955,
      "step": 10508,
      "time_per_iteration": 4.017300605773926
    },
    {
      "auxiliary_loss_clip": 0.01075927,
      "auxiliary_loss_mlp": 0.01032185,
      "balance_loss_clip": 1.03583074,
      "balance_loss_mlp": 1.01983523,
      "epoch": 0.6318352622876897,
      "flos": 23294749616640.0,
      "grad_norm": 1.610016961259701,
      "language_loss": 0.70742261,
      "learning_rate": 1.2610213623767478e-06,
      "loss": 0.72850376,
      "num_input_tokens_seen": 226588205,
      "step": 10509,
      "time_per_iteration": 2.6113953590393066
    },
    {
      "auxiliary_loss_clip": 0.01092941,
      "auxiliary_loss_mlp": 0.01026058,
      "balance_loss_clip": 1.03823936,
      "balance_loss_mlp": 1.01464987,
      "epoch": 0.6318953855403577,
      "flos": 20704082117760.0,
      "grad_norm": 1.5948657287673125,
      "language_loss": 0.79462647,
      "learning_rate": 1.2606594749733143e-06,
      "loss": 0.81581646,
      "num_input_tokens_seen": 226606965,
      "step": 10510,
      "time_per_iteration": 2.5095455646514893
    },
    {
      "auxiliary_loss_clip": 0.01066681,
      "auxiliary_loss_mlp": 0.00785347,
      "balance_loss_clip": 1.03814363,
      "balance_loss_mlp": 1.0115844,
      "epoch": 0.6319555087930258,
      "flos": 22820046451200.0,
      "grad_norm": 1.5847971526489208,
      "language_loss": 0.70468187,
      "learning_rate": 1.2602976156057469e-06,
      "loss": 0.72320211,
      "num_input_tokens_seen": 226627845,
      "step": 10511,
      "time_per_iteration": 4.087428092956543
    },
    {
      "auxiliary_loss_clip": 0.01105327,
      "auxiliary_loss_mlp": 0.01031349,
      "balance_loss_clip": 1.03692687,
      "balance_loss_mlp": 1.01991105,
      "epoch": 0.6320156320456937,
      "flos": 19970929618560.0,
      "grad_norm": 1.8580616848983822,
      "language_loss": 0.79993069,
      "learning_rate": 1.2599357842877684e-06,
      "loss": 0.82129741,
      "num_input_tokens_seen": 226645855,
      "step": 10512,
      "time_per_iteration": 2.464200019836426
    },
    {
      "auxiliary_loss_clip": 0.01100752,
      "auxiliary_loss_mlp": 0.0103292,
      "balance_loss_clip": 1.04056525,
      "balance_loss_mlp": 1.02010477,
      "epoch": 0.6320757552983617,
      "flos": 27013406889600.0,
      "grad_norm": 1.9764285329037476,
      "language_loss": 0.70472342,
      "learning_rate": 1.2595739810330994e-06,
      "loss": 0.72606009,
      "num_input_tokens_seen": 226665375,
      "step": 10513,
      "time_per_iteration": 2.5635552406311035
    },
    {
      "auxiliary_loss_clip": 0.01102045,
      "auxiliary_loss_mlp": 0.01030321,
      "balance_loss_clip": 1.03888941,
      "balance_loss_mlp": 1.01708317,
      "epoch": 0.6321358785510296,
      "flos": 23695943598720.0,
      "grad_norm": 1.716797906945892,
      "language_loss": 0.66222686,
      "learning_rate": 1.259212205855459e-06,
      "loss": 0.68355048,
      "num_input_tokens_seen": 226685270,
      "step": 10514,
      "time_per_iteration": 2.5186305046081543
    },
    {
      "auxiliary_loss_clip": 0.01074081,
      "auxiliary_loss_mlp": 0.01028489,
      "balance_loss_clip": 1.03432751,
      "balance_loss_mlp": 1.01656771,
      "epoch": 0.6321960018036976,
      "flos": 25995231970560.0,
      "grad_norm": 1.971696350095617,
      "language_loss": 0.74315715,
      "learning_rate": 1.2588504587685663e-06,
      "loss": 0.76418281,
      "num_input_tokens_seen": 226705325,
      "step": 10515,
      "time_per_iteration": 2.618481397628784
    },
    {
      "auxiliary_loss_clip": 0.01082854,
      "auxiliary_loss_mlp": 0.01026245,
      "balance_loss_clip": 1.04078698,
      "balance_loss_mlp": 1.01483631,
      "epoch": 0.6322561250563655,
      "flos": 22821016118400.0,
      "grad_norm": 1.7586305557149635,
      "language_loss": 0.90038139,
      "learning_rate": 1.2584887397861379e-06,
      "loss": 0.92147237,
      "num_input_tokens_seen": 226723815,
      "step": 10516,
      "time_per_iteration": 2.5367228984832764
    },
    {
      "auxiliary_loss_clip": 0.01117405,
      "auxiliary_loss_mlp": 0.01031848,
      "balance_loss_clip": 1.04213536,
      "balance_loss_mlp": 1.01790667,
      "epoch": 0.6323162483090335,
      "flos": 18988413926400.0,
      "grad_norm": 1.7411984915827756,
      "language_loss": 0.81769997,
      "learning_rate": 1.2581270489218911e-06,
      "loss": 0.83919251,
      "num_input_tokens_seen": 226741550,
      "step": 10517,
      "time_per_iteration": 2.4627575874328613
    },
    {
      "auxiliary_loss_clip": 0.01053949,
      "auxiliary_loss_mlp": 0.01037518,
      "balance_loss_clip": 1.03771651,
      "balance_loss_mlp": 1.02563858,
      "epoch": 0.6323763715617015,
      "flos": 19865173000320.0,
      "grad_norm": 1.7297809544068898,
      "language_loss": 0.7778542,
      "learning_rate": 1.257765386189541e-06,
      "loss": 0.79876888,
      "num_input_tokens_seen": 226761115,
      "step": 10518,
      "time_per_iteration": 2.6264665126800537
    },
    {
      "auxiliary_loss_clip": 0.01092704,
      "auxiliary_loss_mlp": 0.01030383,
      "balance_loss_clip": 1.03918958,
      "balance_loss_mlp": 1.01821136,
      "epoch": 0.6324364948143695,
      "flos": 22782699285120.0,
      "grad_norm": 1.6727278190339019,
      "language_loss": 0.85325098,
      "learning_rate": 1.2574037516028018e-06,
      "loss": 0.87448186,
      "num_input_tokens_seen": 226782225,
      "step": 10519,
      "time_per_iteration": 2.5523741245269775
    },
    {
      "auxiliary_loss_clip": 0.01083174,
      "auxiliary_loss_mlp": 0.01035671,
      "balance_loss_clip": 1.03699279,
      "balance_loss_mlp": 1.02389896,
      "epoch": 0.6324966180670374,
      "flos": 22235923480320.0,
      "grad_norm": 1.4426770262871134,
      "language_loss": 0.71873581,
      "learning_rate": 1.2570421451753867e-06,
      "loss": 0.73992419,
      "num_input_tokens_seen": 226802375,
      "step": 10520,
      "time_per_iteration": 2.5580546855926514
    },
    {
      "auxiliary_loss_clip": 0.01095247,
      "auxiliary_loss_mlp": 0.01032168,
      "balance_loss_clip": 1.03676105,
      "balance_loss_mlp": 1.02065825,
      "epoch": 0.6325567413197054,
      "flos": 21689183589120.0,
      "grad_norm": 1.6578344010073633,
      "language_loss": 0.71758783,
      "learning_rate": 1.2566805669210081e-06,
      "loss": 0.73886198,
      "num_input_tokens_seen": 226822165,
      "step": 10521,
      "time_per_iteration": 2.487914800643921
    },
    {
      "auxiliary_loss_clip": 0.01071764,
      "auxiliary_loss_mlp": 0.01036686,
      "balance_loss_clip": 1.03828335,
      "balance_loss_mlp": 1.02196383,
      "epoch": 0.6326168645723733,
      "flos": 19937137898880.0,
      "grad_norm": 1.6807197210836269,
      "language_loss": 0.72257626,
      "learning_rate": 1.256319016853377e-06,
      "loss": 0.74366081,
      "num_input_tokens_seen": 226841645,
      "step": 10522,
      "time_per_iteration": 2.568253755569458
    },
    {
      "auxiliary_loss_clip": 0.01066753,
      "auxiliary_loss_mlp": 0.01029309,
      "balance_loss_clip": 1.04293251,
      "balance_loss_mlp": 1.01737583,
      "epoch": 0.6326769878250413,
      "flos": 20230348619520.0,
      "grad_norm": 1.9482538328279864,
      "language_loss": 0.81763285,
      "learning_rate": 1.2559574949862023e-06,
      "loss": 0.83859342,
      "num_input_tokens_seen": 226860355,
      "step": 10523,
      "time_per_iteration": 2.5915725231170654
    },
    {
      "auxiliary_loss_clip": 0.01096528,
      "auxiliary_loss_mlp": 0.01027074,
      "balance_loss_clip": 1.03789043,
      "balance_loss_mlp": 1.01523089,
      "epoch": 0.6327371110777094,
      "flos": 20775759707520.0,
      "grad_norm": 1.980881709990781,
      "language_loss": 0.73682535,
      "learning_rate": 1.255596001333195e-06,
      "loss": 0.75806141,
      "num_input_tokens_seen": 226878390,
      "step": 10524,
      "time_per_iteration": 2.503743886947632
    },
    {
      "auxiliary_loss_clip": 0.01096141,
      "auxiliary_loss_mlp": 0.01034498,
      "balance_loss_clip": 1.04060507,
      "balance_loss_mlp": 1.02089047,
      "epoch": 0.6327972343303773,
      "flos": 30336544529280.0,
      "grad_norm": 2.937902524645182,
      "language_loss": 0.84871769,
      "learning_rate": 1.2552345359080615e-06,
      "loss": 0.87002409,
      "num_input_tokens_seen": 226898420,
      "step": 10525,
      "time_per_iteration": 2.588752269744873
    },
    {
      "auxiliary_loss_clip": 0.01078106,
      "auxiliary_loss_mlp": 0.01028034,
      "balance_loss_clip": 1.03365541,
      "balance_loss_mlp": 1.01505256,
      "epoch": 0.6328573575830453,
      "flos": 17092258871040.0,
      "grad_norm": 1.5877539651621269,
      "language_loss": 0.66856855,
      "learning_rate": 1.2548730987245093e-06,
      "loss": 0.68962991,
      "num_input_tokens_seen": 226916305,
      "step": 10526,
      "time_per_iteration": 2.5043728351593018
    },
    {
      "auxiliary_loss_clip": 0.01104247,
      "auxiliary_loss_mlp": 0.01034052,
      "balance_loss_clip": 1.04123306,
      "balance_loss_mlp": 1.02062964,
      "epoch": 0.6329174808357132,
      "flos": 25047154442880.0,
      "grad_norm": 1.5695055229957688,
      "language_loss": 0.73526955,
      "learning_rate": 1.254511689796244e-06,
      "loss": 0.75665253,
      "num_input_tokens_seen": 226937705,
      "step": 10527,
      "time_per_iteration": 2.5358808040618896
    },
    {
      "auxiliary_loss_clip": 0.01096332,
      "auxiliary_loss_mlp": 0.01031171,
      "balance_loss_clip": 1.03936195,
      "balance_loss_mlp": 1.01972675,
      "epoch": 0.6329776040883812,
      "flos": 16836826279680.0,
      "grad_norm": 2.628399565810566,
      "language_loss": 0.71817124,
      "learning_rate": 1.2541503091369693e-06,
      "loss": 0.73944628,
      "num_input_tokens_seen": 226954880,
      "step": 10528,
      "time_per_iteration": 2.59552264213562
    },
    {
      "auxiliary_loss_clip": 0.01094437,
      "auxiliary_loss_mlp": 0.01028836,
      "balance_loss_clip": 1.03611851,
      "balance_loss_mlp": 1.01602137,
      "epoch": 0.6330377273410491,
      "flos": 13516705382400.0,
      "grad_norm": 1.939395114410684,
      "language_loss": 0.66771805,
      "learning_rate": 1.2537889567603905e-06,
      "loss": 0.68895084,
      "num_input_tokens_seen": 226972595,
      "step": 10529,
      "time_per_iteration": 2.456913948059082
    },
    {
      "auxiliary_loss_clip": 0.01100628,
      "auxiliary_loss_mlp": 0.01029875,
      "balance_loss_clip": 1.03807127,
      "balance_loss_mlp": 1.01672602,
      "epoch": 0.6330978505937171,
      "flos": 21538825257600.0,
      "grad_norm": 2.3215196514771894,
      "language_loss": 0.75099337,
      "learning_rate": 1.2534276326802092e-06,
      "loss": 0.7722984,
      "num_input_tokens_seen": 226991910,
      "step": 10530,
      "time_per_iteration": 2.5166869163513184
    },
    {
      "auxiliary_loss_clip": 0.01100665,
      "auxiliary_loss_mlp": 0.00787797,
      "balance_loss_clip": 1.04024553,
      "balance_loss_mlp": 1.01569068,
      "epoch": 0.6331579738463851,
      "flos": 25009484054400.0,
      "grad_norm": 1.5957438975326417,
      "language_loss": 0.73999476,
      "learning_rate": 1.2530663369101259e-06,
      "loss": 0.75887942,
      "num_input_tokens_seen": 227010175,
      "step": 10531,
      "time_per_iteration": 2.5931880474090576
    },
    {
      "auxiliary_loss_clip": 0.01068481,
      "auxiliary_loss_mlp": 0.01028153,
      "balance_loss_clip": 1.0345068,
      "balance_loss_mlp": 1.01580334,
      "epoch": 0.6332180970990531,
      "flos": 14976007228800.0,
      "grad_norm": 2.66832208196273,
      "language_loss": 0.79366064,
      "learning_rate": 1.2527050694638432e-06,
      "loss": 0.81462693,
      "num_input_tokens_seen": 227025540,
      "step": 10532,
      "time_per_iteration": 2.524585247039795
    },
    {
      "auxiliary_loss_clip": 0.0109445,
      "auxiliary_loss_mlp": 0.01024925,
      "balance_loss_clip": 1.0368259,
      "balance_loss_mlp": 1.01436889,
      "epoch": 0.633278220351721,
      "flos": 22706963458560.0,
      "grad_norm": 1.5063211651932675,
      "language_loss": 0.74266708,
      "learning_rate": 1.2523438303550582e-06,
      "loss": 0.76386082,
      "num_input_tokens_seen": 227045520,
      "step": 10533,
      "time_per_iteration": 2.531118869781494
    },
    {
      "auxiliary_loss_clip": 0.01088628,
      "auxiliary_loss_mlp": 0.01036176,
      "balance_loss_clip": 1.04040611,
      "balance_loss_mlp": 1.02297938,
      "epoch": 0.633338343604389,
      "flos": 12602922364800.0,
      "grad_norm": 2.4538112656450113,
      "language_loss": 0.76876557,
      "learning_rate": 1.2519826195974706e-06,
      "loss": 0.79001367,
      "num_input_tokens_seen": 227059420,
      "step": 10534,
      "time_per_iteration": 2.489245891571045
    },
    {
      "auxiliary_loss_clip": 0.01075489,
      "auxiliary_loss_mlp": 0.01040224,
      "balance_loss_clip": 1.0376786,
      "balance_loss_mlp": 1.02655053,
      "epoch": 0.6333984668570569,
      "flos": 25960111447680.0,
      "grad_norm": 1.513563332197795,
      "language_loss": 0.85622549,
      "learning_rate": 1.251621437204777e-06,
      "loss": 0.87738258,
      "num_input_tokens_seen": 227081310,
      "step": 10535,
      "time_per_iteration": 2.6177361011505127
    },
    {
      "auxiliary_loss_clip": 0.01100602,
      "auxiliary_loss_mlp": 0.01031115,
      "balance_loss_clip": 1.03882122,
      "balance_loss_mlp": 1.01838899,
      "epoch": 0.6334585901097249,
      "flos": 23659242877440.0,
      "grad_norm": 1.9602123185922353,
      "language_loss": 0.76321566,
      "learning_rate": 1.2512602831906733e-06,
      "loss": 0.78453285,
      "num_input_tokens_seen": 227100365,
      "step": 10536,
      "time_per_iteration": 2.526759147644043
    },
    {
      "auxiliary_loss_clip": 0.01094546,
      "auxiliary_loss_mlp": 0.01030403,
      "balance_loss_clip": 1.03875113,
      "balance_loss_mlp": 1.0177604,
      "epoch": 0.633518713362393,
      "flos": 28760496503040.0,
      "grad_norm": 1.9444093213721925,
      "language_loss": 0.59998196,
      "learning_rate": 1.250899157568855e-06,
      "loss": 0.62123144,
      "num_input_tokens_seen": 227119680,
      "step": 10537,
      "time_per_iteration": 3.973539352416992
    },
    {
      "auxiliary_loss_clip": 0.01023145,
      "auxiliary_loss_mlp": 0.01001659,
      "balance_loss_clip": 1.03101969,
      "balance_loss_mlp": 1.00019884,
      "epoch": 0.6335788366150609,
      "flos": 70420322401920.0,
      "grad_norm": 0.7826699306019819,
      "language_loss": 0.52401054,
      "learning_rate": 1.2505380603530155e-06,
      "loss": 0.54425859,
      "num_input_tokens_seen": 227184465,
      "step": 10538,
      "time_per_iteration": 3.271909475326538
    },
    {
      "auxiliary_loss_clip": 0.0108849,
      "auxiliary_loss_mlp": 0.01029852,
      "balance_loss_clip": 1.03758752,
      "balance_loss_mlp": 1.0167985,
      "epoch": 0.6336389598677289,
      "flos": 23732069702400.0,
      "grad_norm": 1.785509499663211,
      "language_loss": 0.83286905,
      "learning_rate": 1.250176991556848e-06,
      "loss": 0.85405242,
      "num_input_tokens_seen": 227202185,
      "step": 10539,
      "time_per_iteration": 2.5774717330932617
    },
    {
      "auxiliary_loss_clip": 0.01082994,
      "auxiliary_loss_mlp": 0.01030872,
      "balance_loss_clip": 1.03720069,
      "balance_loss_mlp": 1.01722229,
      "epoch": 0.6336990831203968,
      "flos": 29276676898560.0,
      "grad_norm": 1.74746656512703,
      "language_loss": 0.86958325,
      "learning_rate": 1.2498159511940438e-06,
      "loss": 0.89072186,
      "num_input_tokens_seen": 227222020,
      "step": 10540,
      "time_per_iteration": 2.561495304107666
    },
    {
      "auxiliary_loss_clip": 0.01083062,
      "auxiliary_loss_mlp": 0.01029738,
      "balance_loss_clip": 1.03696024,
      "balance_loss_mlp": 1.01899767,
      "epoch": 0.6337592063730648,
      "flos": 29096836479360.0,
      "grad_norm": 1.7199646866032428,
      "language_loss": 0.72492909,
      "learning_rate": 1.2494549392782943e-06,
      "loss": 0.74605703,
      "num_input_tokens_seen": 227240885,
      "step": 10541,
      "time_per_iteration": 2.609320878982544
    },
    {
      "auxiliary_loss_clip": 0.01101306,
      "auxiliary_loss_mlp": 0.01032092,
      "balance_loss_clip": 1.03793716,
      "balance_loss_mlp": 1.01880002,
      "epoch": 0.6338193296257327,
      "flos": 34706477249280.0,
      "grad_norm": 2.241530031887131,
      "language_loss": 0.84868586,
      "learning_rate": 1.2490939558232887e-06,
      "loss": 0.87001985,
      "num_input_tokens_seen": 227257880,
      "step": 10542,
      "time_per_iteration": 2.5793919563293457
    },
    {
      "auxiliary_loss_clip": 0.01098041,
      "auxiliary_loss_mlp": 0.01027509,
      "balance_loss_clip": 1.03885901,
      "balance_loss_mlp": 1.01450884,
      "epoch": 0.6338794528784008,
      "flos": 16687581269760.0,
      "grad_norm": 1.6306502789302466,
      "language_loss": 0.77579951,
      "learning_rate": 1.2487330008427153e-06,
      "loss": 0.79705495,
      "num_input_tokens_seen": 227274840,
      "step": 10543,
      "time_per_iteration": 2.5054731369018555
    },
    {
      "auxiliary_loss_clip": 0.01055056,
      "auxiliary_loss_mlp": 0.01031077,
      "balance_loss_clip": 1.03563094,
      "balance_loss_mlp": 1.0194366,
      "epoch": 0.6339395761310687,
      "flos": 22346600261760.0,
      "grad_norm": 1.6385980734790173,
      "language_loss": 0.73702031,
      "learning_rate": 1.2483720743502618e-06,
      "loss": 0.75788164,
      "num_input_tokens_seen": 227294835,
      "step": 10544,
      "time_per_iteration": 2.5859904289245605
    },
    {
      "auxiliary_loss_clip": 0.01082034,
      "auxiliary_loss_mlp": 0.0103345,
      "balance_loss_clip": 1.03790951,
      "balance_loss_mlp": 1.02080226,
      "epoch": 0.6339996993837367,
      "flos": 18551812112640.0,
      "grad_norm": 1.9337740477926275,
      "language_loss": 0.6832059,
      "learning_rate": 1.2480111763596144e-06,
      "loss": 0.70436072,
      "num_input_tokens_seen": 227314935,
      "step": 10545,
      "time_per_iteration": 2.6033241748809814
    },
    {
      "auxiliary_loss_clip": 0.01084084,
      "auxiliary_loss_mlp": 0.01034082,
      "balance_loss_clip": 1.03618336,
      "balance_loss_mlp": 1.02166057,
      "epoch": 0.6340598226364046,
      "flos": 12969498614400.0,
      "grad_norm": 2.2348258232641505,
      "language_loss": 0.7162692,
      "learning_rate": 1.2476503068844592e-06,
      "loss": 0.73745084,
      "num_input_tokens_seen": 227332905,
      "step": 10546,
      "time_per_iteration": 3.8842544555664062
    },
    {
      "auxiliary_loss_clip": 0.01091283,
      "auxiliary_loss_mlp": 0.0102749,
      "balance_loss_clip": 1.03733659,
      "balance_loss_mlp": 1.01613498,
      "epoch": 0.6341199458890726,
      "flos": 26687984647680.0,
      "grad_norm": 1.357550578734522,
      "language_loss": 0.77875954,
      "learning_rate": 1.2472894659384792e-06,
      "loss": 0.79994726,
      "num_input_tokens_seen": 227354915,
      "step": 10547,
      "time_per_iteration": 3.960388422012329
    },
    {
      "auxiliary_loss_clip": 0.01065754,
      "auxiliary_loss_mlp": 0.01032103,
      "balance_loss_clip": 1.03406513,
      "balance_loss_mlp": 1.0198487,
      "epoch": 0.6341800691417405,
      "flos": 18734274224640.0,
      "grad_norm": 1.7761098996400293,
      "language_loss": 0.63565892,
      "learning_rate": 1.2469286535353578e-06,
      "loss": 0.65663749,
      "num_input_tokens_seen": 227372990,
      "step": 10548,
      "time_per_iteration": 2.558922290802002
    },
    {
      "auxiliary_loss_clip": 0.01082999,
      "auxiliary_loss_mlp": 0.01031295,
      "balance_loss_clip": 1.03667426,
      "balance_loss_mlp": 1.01923144,
      "epoch": 0.6342401923944085,
      "flos": 26249443499520.0,
      "grad_norm": 1.955875792108172,
      "language_loss": 0.61780417,
      "learning_rate": 1.2465678696887785e-06,
      "loss": 0.63894713,
      "num_input_tokens_seen": 227393270,
      "step": 10549,
      "time_per_iteration": 2.6088898181915283
    },
    {
      "auxiliary_loss_clip": 0.01058167,
      "auxiliary_loss_mlp": 0.01028106,
      "balance_loss_clip": 1.03806317,
      "balance_loss_mlp": 1.01706696,
      "epoch": 0.6343003156470765,
      "flos": 24680937329280.0,
      "grad_norm": 1.5351086264604867,
      "language_loss": 0.73456025,
      "learning_rate": 1.2462071144124197e-06,
      "loss": 0.75542301,
      "num_input_tokens_seen": 227413630,
      "step": 10550,
      "time_per_iteration": 3.9695441722869873
    },
    {
      "auxiliary_loss_clip": 0.0100266,
      "auxiliary_loss_mlp": 0.01001434,
      "balance_loss_clip": 1.01397312,
      "balance_loss_mlp": 0.99997979,
      "epoch": 0.6343604388997445,
      "flos": 69805352626560.0,
      "grad_norm": 0.6879621401315671,
      "language_loss": 0.57727134,
      "learning_rate": 1.2458463877199638e-06,
      "loss": 0.59731233,
      "num_input_tokens_seen": 227476630,
      "step": 10551,
      "time_per_iteration": 3.1948065757751465
    },
    {
      "auxiliary_loss_clip": 0.01072036,
      "auxiliary_loss_mlp": 0.01026089,
      "balance_loss_clip": 1.03960013,
      "balance_loss_mlp": 1.0151633,
      "epoch": 0.6344205621524125,
      "flos": 21982430223360.0,
      "grad_norm": 1.69756485178634,
      "language_loss": 0.67153227,
      "learning_rate": 1.2454856896250881e-06,
      "loss": 0.69251347,
      "num_input_tokens_seen": 227496060,
      "step": 10552,
      "time_per_iteration": 2.5714244842529297
    },
    {
      "auxiliary_loss_clip": 0.01072104,
      "auxiliary_loss_mlp": 0.01029449,
      "balance_loss_clip": 1.03612089,
      "balance_loss_mlp": 1.01658618,
      "epoch": 0.6344806854050804,
      "flos": 20448865008000.0,
      "grad_norm": 1.7273900690083654,
      "language_loss": 0.81999224,
      "learning_rate": 1.24512502014147e-06,
      "loss": 0.84100783,
      "num_input_tokens_seen": 227513440,
      "step": 10553,
      "time_per_iteration": 2.5873239040374756
    },
    {
      "auxiliary_loss_clip": 0.01099567,
      "auxiliary_loss_mlp": 0.01031168,
      "balance_loss_clip": 1.03832245,
      "balance_loss_mlp": 1.01934874,
      "epoch": 0.6345408086577484,
      "flos": 40510611187200.0,
      "grad_norm": 2.0535091885207724,
      "language_loss": 0.55393964,
      "learning_rate": 1.2447643792827879e-06,
      "loss": 0.57524699,
      "num_input_tokens_seen": 227535395,
      "step": 10554,
      "time_per_iteration": 2.66570782661438
    },
    {
      "auxiliary_loss_clip": 0.01090598,
      "auxiliary_loss_mlp": 0.01028703,
      "balance_loss_clip": 1.03940797,
      "balance_loss_mlp": 1.01651394,
      "epoch": 0.6346009319104163,
      "flos": 21361319222400.0,
      "grad_norm": 2.0507047877694307,
      "language_loss": 0.70662886,
      "learning_rate": 1.2444037670627153e-06,
      "loss": 0.72782189,
      "num_input_tokens_seen": 227554545,
      "step": 10555,
      "time_per_iteration": 2.5427963733673096
    },
    {
      "auxiliary_loss_clip": 0.0102667,
      "auxiliary_loss_mlp": 0.01001924,
      "balance_loss_clip": 1.02273798,
      "balance_loss_mlp": 1.00069571,
      "epoch": 0.6346610551630844,
      "flos": 71365419100800.0,
      "grad_norm": 0.768927993630889,
      "language_loss": 0.55314529,
      "learning_rate": 1.2440431834949276e-06,
      "loss": 0.57343125,
      "num_input_tokens_seen": 227608575,
      "step": 10556,
      "time_per_iteration": 3.045254945755005
    },
    {
      "auxiliary_loss_clip": 0.01089012,
      "auxiliary_loss_mlp": 0.01031473,
      "balance_loss_clip": 1.03687024,
      "balance_loss_mlp": 1.01731133,
      "epoch": 0.6347211784157523,
      "flos": 25411504049280.0,
      "grad_norm": 1.8054348455867575,
      "language_loss": 0.68206024,
      "learning_rate": 1.2436826285930985e-06,
      "loss": 0.70326513,
      "num_input_tokens_seen": 227628175,
      "step": 10557,
      "time_per_iteration": 2.5723249912261963
    },
    {
      "auxiliary_loss_clip": 0.01077479,
      "auxiliary_loss_mlp": 0.01034883,
      "balance_loss_clip": 1.03510571,
      "balance_loss_mlp": 1.02255046,
      "epoch": 0.6347813016684203,
      "flos": 15742735966080.0,
      "grad_norm": 1.7233972761198761,
      "language_loss": 0.70164657,
      "learning_rate": 1.2433221023709002e-06,
      "loss": 0.72277015,
      "num_input_tokens_seen": 227645330,
      "step": 10558,
      "time_per_iteration": 2.5047454833984375
    },
    {
      "auxiliary_loss_clip": 0.0108616,
      "auxiliary_loss_mlp": 0.0102853,
      "balance_loss_clip": 1.03999698,
      "balance_loss_mlp": 1.01613224,
      "epoch": 0.6348414249210882,
      "flos": 21464777370240.0,
      "grad_norm": 1.5896154627905474,
      "language_loss": 0.78300655,
      "learning_rate": 1.2429616048420031e-06,
      "loss": 0.80415344,
      "num_input_tokens_seen": 227665250,
      "step": 10559,
      "time_per_iteration": 2.549590826034546
    },
    {
      "auxiliary_loss_clip": 0.01084934,
      "auxiliary_loss_mlp": 0.01037077,
      "balance_loss_clip": 1.03663993,
      "balance_loss_mlp": 1.02411306,
      "epoch": 0.6349015481737562,
      "flos": 21653057485440.0,
      "grad_norm": 1.9576938440010785,
      "language_loss": 0.67729032,
      "learning_rate": 1.242601136020078e-06,
      "loss": 0.69851041,
      "num_input_tokens_seen": 227685070,
      "step": 10560,
      "time_per_iteration": 2.549433708190918
    },
    {
      "auxiliary_loss_clip": 0.0108257,
      "auxiliary_loss_mlp": 0.01039993,
      "balance_loss_clip": 1.03630352,
      "balance_loss_mlp": 1.02605128,
      "epoch": 0.6349616714264241,
      "flos": 22194984954240.0,
      "grad_norm": 1.8616636705562717,
      "language_loss": 0.76985544,
      "learning_rate": 1.2422406959187939e-06,
      "loss": 0.79108107,
      "num_input_tokens_seen": 227704430,
      "step": 10561,
      "time_per_iteration": 2.569769859313965
    },
    {
      "auxiliary_loss_clip": 0.01087622,
      "auxiliary_loss_mlp": 0.01030268,
      "balance_loss_clip": 1.03796577,
      "balance_loss_mlp": 1.01819801,
      "epoch": 0.6350217946790921,
      "flos": 25410354814080.0,
      "grad_norm": 1.904674523468591,
      "language_loss": 0.72441125,
      "learning_rate": 1.2418802845518178e-06,
      "loss": 0.74559015,
      "num_input_tokens_seen": 227724920,
      "step": 10562,
      "time_per_iteration": 2.576662302017212
    },
    {
      "auxiliary_loss_clip": 0.01097431,
      "auxiliary_loss_mlp": 0.01032684,
      "balance_loss_clip": 1.03995991,
      "balance_loss_mlp": 1.01952946,
      "epoch": 0.63508191793176,
      "flos": 19718944732800.0,
      "grad_norm": 2.8593023211441033,
      "language_loss": 0.8088994,
      "learning_rate": 1.2415199019328185e-06,
      "loss": 0.83020055,
      "num_input_tokens_seen": 227743400,
      "step": 10563,
      "time_per_iteration": 2.4870877265930176
    },
    {
      "auxiliary_loss_clip": 0.01082045,
      "auxiliary_loss_mlp": 0.0103485,
      "balance_loss_clip": 1.04034185,
      "balance_loss_mlp": 1.02276254,
      "epoch": 0.6351420411844281,
      "flos": 18186923802240.0,
      "grad_norm": 2.243216889151289,
      "language_loss": 0.80948257,
      "learning_rate": 1.2411595480754597e-06,
      "loss": 0.8306514,
      "num_input_tokens_seen": 227759990,
      "step": 10564,
      "time_per_iteration": 2.5609982013702393
    },
    {
      "auxiliary_loss_clip": 0.01081514,
      "auxiliary_loss_mlp": 0.01038817,
      "balance_loss_clip": 1.03971624,
      "balance_loss_mlp": 1.02494144,
      "epoch": 0.6352021644370961,
      "flos": 33726511422720.0,
      "grad_norm": 1.55146090560981,
      "language_loss": 0.7262218,
      "learning_rate": 1.240799222993407e-06,
      "loss": 0.7474252,
      "num_input_tokens_seen": 227780835,
      "step": 10565,
      "time_per_iteration": 2.639836072921753
    },
    {
      "auxiliary_loss_clip": 0.01094597,
      "auxiliary_loss_mlp": 0.01031296,
      "balance_loss_clip": 1.03827131,
      "balance_loss_mlp": 1.01763475,
      "epoch": 0.635262287689764,
      "flos": 20374781207040.0,
      "grad_norm": 1.9175968216730581,
      "language_loss": 0.68946874,
      "learning_rate": 1.240438926700324e-06,
      "loss": 0.71072769,
      "num_input_tokens_seen": 227798580,
      "step": 10566,
      "time_per_iteration": 2.4870307445526123
    },
    {
      "auxiliary_loss_clip": 0.01095158,
      "auxiliary_loss_mlp": 0.01032377,
      "balance_loss_clip": 1.03828096,
      "balance_loss_mlp": 1.02089095,
      "epoch": 0.635322410942432,
      "flos": 27525421307520.0,
      "grad_norm": 1.5963396740231894,
      "language_loss": 0.69727767,
      "learning_rate": 1.2400786592098725e-06,
      "loss": 0.71855307,
      "num_input_tokens_seen": 227819210,
      "step": 10567,
      "time_per_iteration": 2.535710573196411
    },
    {
      "auxiliary_loss_clip": 0.01095542,
      "auxiliary_loss_mlp": 0.01028201,
      "balance_loss_clip": 1.04300332,
      "balance_loss_mlp": 1.01695323,
      "epoch": 0.6353825341950999,
      "flos": 21543601766400.0,
      "grad_norm": 1.8626197517743353,
      "language_loss": 0.84522688,
      "learning_rate": 1.2397184205357154e-06,
      "loss": 0.86646426,
      "num_input_tokens_seen": 227838340,
      "step": 10568,
      "time_per_iteration": 2.5074243545532227
    },
    {
      "auxiliary_loss_clip": 0.01058869,
      "auxiliary_loss_mlp": 0.01039135,
      "balance_loss_clip": 1.04156578,
      "balance_loss_mlp": 1.0260638,
      "epoch": 0.635442657447768,
      "flos": 31759756185600.0,
      "grad_norm": 1.7697464809189671,
      "language_loss": 0.84041762,
      "learning_rate": 1.2393582106915113e-06,
      "loss": 0.86139762,
      "num_input_tokens_seen": 227859170,
      "step": 10569,
      "time_per_iteration": 2.6775002479553223
    },
    {
      "auxiliary_loss_clip": 0.01097443,
      "auxiliary_loss_mlp": 0.0102956,
      "balance_loss_clip": 1.03916085,
      "balance_loss_mlp": 1.01692367,
      "epoch": 0.6355027807004359,
      "flos": 19828831415040.0,
      "grad_norm": 1.6430141818593216,
      "language_loss": 0.69300056,
      "learning_rate": 1.2389980296909198e-06,
      "loss": 0.71427059,
      "num_input_tokens_seen": 227878545,
      "step": 10570,
      "time_per_iteration": 2.5021378993988037
    },
    {
      "auxiliary_loss_clip": 0.01099807,
      "auxiliary_loss_mlp": 0.01031427,
      "balance_loss_clip": 1.03807199,
      "balance_loss_mlp": 1.01876664,
      "epoch": 0.6355629039531039,
      "flos": 30372383324160.0,
      "grad_norm": 1.6486580787950484,
      "language_loss": 0.65871722,
      "learning_rate": 1.2386378775476e-06,
      "loss": 0.68002951,
      "num_input_tokens_seen": 227898875,
      "step": 10571,
      "time_per_iteration": 2.5564448833465576
    },
    {
      "auxiliary_loss_clip": 0.01104579,
      "auxiliary_loss_mlp": 0.01027409,
      "balance_loss_clip": 1.04042172,
      "balance_loss_mlp": 1.01494551,
      "epoch": 0.6356230272057718,
      "flos": 17932065828480.0,
      "grad_norm": 1.7713311403265748,
      "language_loss": 0.71233553,
      "learning_rate": 1.2382777542752074e-06,
      "loss": 0.73365545,
      "num_input_tokens_seen": 227917130,
      "step": 10572,
      "time_per_iteration": 2.4843251705169678
    },
    {
      "auxiliary_loss_clip": 0.01075187,
      "auxiliary_loss_mlp": 0.01030305,
      "balance_loss_clip": 1.0363152,
      "balance_loss_mlp": 1.01859879,
      "epoch": 0.6356831504584398,
      "flos": 25375844822400.0,
      "grad_norm": 1.4083202598842577,
      "language_loss": 0.81385899,
      "learning_rate": 1.2379176598873992e-06,
      "loss": 0.83491391,
      "num_input_tokens_seen": 227939550,
      "step": 10573,
      "time_per_iteration": 2.6022627353668213
    },
    {
      "auxiliary_loss_clip": 0.01091274,
      "auxiliary_loss_mlp": 0.01030503,
      "balance_loss_clip": 1.0388453,
      "balance_loss_mlp": 1.01841497,
      "epoch": 0.6357432737111077,
      "flos": 46500331720320.0,
      "grad_norm": 1.6141386554762005,
      "language_loss": 0.69002855,
      "learning_rate": 1.2375575943978303e-06,
      "loss": 0.71124625,
      "num_input_tokens_seen": 227962200,
      "step": 10574,
      "time_per_iteration": 2.738844633102417
    },
    {
      "auxiliary_loss_clip": 0.01109528,
      "auxiliary_loss_mlp": 0.01027475,
      "balance_loss_clip": 1.03956556,
      "balance_loss_mlp": 1.01494622,
      "epoch": 0.6358033969637757,
      "flos": 17274361847040.0,
      "grad_norm": 2.3325537115895014,
      "language_loss": 0.8660686,
      "learning_rate": 1.2371975578201525e-06,
      "loss": 0.88743865,
      "num_input_tokens_seen": 227979270,
      "step": 10575,
      "time_per_iteration": 3.8155055046081543
    },
    {
      "auxiliary_loss_clip": 0.01109239,
      "auxiliary_loss_mlp": 0.01030693,
      "balance_loss_clip": 1.03939342,
      "balance_loss_mlp": 1.01878428,
      "epoch": 0.6358635202164437,
      "flos": 27125520215040.0,
      "grad_norm": 2.0030041306584563,
      "language_loss": 0.72213882,
      "learning_rate": 1.2368375501680204e-06,
      "loss": 0.74353814,
      "num_input_tokens_seen": 228000550,
      "step": 10576,
      "time_per_iteration": 2.4968883991241455
    },
    {
      "auxiliary_loss_clip": 0.01089871,
      "auxiliary_loss_mlp": 0.01032989,
      "balance_loss_clip": 1.03925872,
      "balance_loss_mlp": 1.02000165,
      "epoch": 0.6359236434691117,
      "flos": 27525205825920.0,
      "grad_norm": 1.5600574988079687,
      "language_loss": 0.69240618,
      "learning_rate": 1.236477571455085e-06,
      "loss": 0.71363485,
      "num_input_tokens_seen": 228022005,
      "step": 10577,
      "time_per_iteration": 2.5839946269989014
    },
    {
      "auxiliary_loss_clip": 0.01067328,
      "auxiliary_loss_mlp": 0.01032321,
      "balance_loss_clip": 1.03625107,
      "balance_loss_mlp": 1.01995277,
      "epoch": 0.6359837667217797,
      "flos": 39348290989440.0,
      "grad_norm": 1.7317600226230725,
      "language_loss": 0.72062105,
      "learning_rate": 1.2361176216949964e-06,
      "loss": 0.74161756,
      "num_input_tokens_seen": 228043770,
      "step": 10578,
      "time_per_iteration": 2.719874143600464
    },
    {
      "auxiliary_loss_clip": 0.01021625,
      "auxiliary_loss_mlp": 0.00770629,
      "balance_loss_clip": 1.02529371,
      "balance_loss_mlp": 1.01208818,
      "epoch": 0.6360438899744476,
      "flos": 56413797206400.0,
      "grad_norm": 0.7026955199567267,
      "language_loss": 0.54541039,
      "learning_rate": 1.2357577009014044e-06,
      "loss": 0.56333292,
      "num_input_tokens_seen": 228104985,
      "step": 10579,
      "time_per_iteration": 3.253638982772827
    },
    {
      "auxiliary_loss_clip": 0.01086378,
      "auxiliary_loss_mlp": 0.01029923,
      "balance_loss_clip": 1.0377028,
      "balance_loss_mlp": 1.01717365,
      "epoch": 0.6361040132271156,
      "flos": 24973106555520.0,
      "grad_norm": 2.396740357738586,
      "language_loss": 0.77945763,
      "learning_rate": 1.2353978090879568e-06,
      "loss": 0.80062068,
      "num_input_tokens_seen": 228125620,
      "step": 10580,
      "time_per_iteration": 2.5658862590789795
    },
    {
      "auxiliary_loss_clip": 0.0108074,
      "auxiliary_loss_mlp": 0.00783536,
      "balance_loss_clip": 1.03921103,
      "balance_loss_mlp": 1.00873637,
      "epoch": 0.6361641364797835,
      "flos": 23259198130560.0,
      "grad_norm": 2.4370329345016284,
      "language_loss": 0.66725898,
      "learning_rate": 1.235037946268301e-06,
      "loss": 0.68590176,
      "num_input_tokens_seen": 228143495,
      "step": 10581,
      "time_per_iteration": 2.5811774730682373
    },
    {
      "auxiliary_loss_clip": 0.01095828,
      "auxiliary_loss_mlp": 0.01029557,
      "balance_loss_clip": 1.04072952,
      "balance_loss_mlp": 1.01765358,
      "epoch": 0.6362242597324516,
      "flos": 25994513698560.0,
      "grad_norm": 1.4179251893570455,
      "language_loss": 0.68543196,
      "learning_rate": 1.2346781124560828e-06,
      "loss": 0.70668578,
      "num_input_tokens_seen": 228166500,
      "step": 10582,
      "time_per_iteration": 2.5556700229644775
    },
    {
      "auxiliary_loss_clip": 0.0108419,
      "auxiliary_loss_mlp": 0.01036269,
      "balance_loss_clip": 1.04179013,
      "balance_loss_mlp": 1.0239011,
      "epoch": 0.6362843829851195,
      "flos": 25703242312320.0,
      "grad_norm": 2.3198661430020158,
      "language_loss": 0.84813017,
      "learning_rate": 1.2343183076649473e-06,
      "loss": 0.86933476,
      "num_input_tokens_seen": 228185325,
      "step": 10583,
      "time_per_iteration": 2.573737621307373
    },
    {
      "auxiliary_loss_clip": 0.01088817,
      "auxiliary_loss_mlp": 0.01027765,
      "balance_loss_clip": 1.04400635,
      "balance_loss_mlp": 1.01596332,
      "epoch": 0.6363445062377875,
      "flos": 20522912895360.0,
      "grad_norm": 1.5038935965841294,
      "language_loss": 0.75519216,
      "learning_rate": 1.233958531908538e-06,
      "loss": 0.77635801,
      "num_input_tokens_seen": 228204050,
      "step": 10584,
      "time_per_iteration": 2.5386343002319336
    },
    {
      "auxiliary_loss_clip": 0.01094055,
      "auxiliary_loss_mlp": 0.01036339,
      "balance_loss_clip": 1.03997135,
      "balance_loss_mlp": 1.02245712,
      "epoch": 0.6364046294904554,
      "flos": 19463799450240.0,
      "grad_norm": 1.8629655573226724,
      "language_loss": 0.72640276,
      "learning_rate": 1.2335987852004985e-06,
      "loss": 0.74770677,
      "num_input_tokens_seen": 228222430,
      "step": 10585,
      "time_per_iteration": 5.3067967891693115
    },
    {
      "auxiliary_loss_clip": 0.01076809,
      "auxiliary_loss_mlp": 0.01029179,
      "balance_loss_clip": 1.04106331,
      "balance_loss_mlp": 1.017717,
      "epoch": 0.6364647527431234,
      "flos": 20995892208000.0,
      "grad_norm": 1.7410569703313215,
      "language_loss": 0.82380742,
      "learning_rate": 1.2332390675544697e-06,
      "loss": 0.84486729,
      "num_input_tokens_seen": 228241925,
      "step": 10586,
      "time_per_iteration": 2.5842554569244385
    },
    {
      "auxiliary_loss_clip": 0.01097977,
      "auxiliary_loss_mlp": 0.01026002,
      "balance_loss_clip": 1.03952742,
      "balance_loss_mlp": 1.01419473,
      "epoch": 0.6365248759957913,
      "flos": 25770789838080.0,
      "grad_norm": 1.6122861917773275,
      "language_loss": 0.7263,
      "learning_rate": 1.2328793789840918e-06,
      "loss": 0.74753976,
      "num_input_tokens_seen": 228262535,
      "step": 10587,
      "time_per_iteration": 2.5400798320770264
    },
    {
      "auxiliary_loss_clip": 0.01086407,
      "auxiliary_loss_mlp": 0.01025559,
      "balance_loss_clip": 1.03835928,
      "balance_loss_mlp": 1.01397169,
      "epoch": 0.6365849992484593,
      "flos": 22455589104000.0,
      "grad_norm": 1.9977663847203795,
      "language_loss": 0.77079427,
      "learning_rate": 1.2325197195030058e-06,
      "loss": 0.79191393,
      "num_input_tokens_seen": 228281340,
      "step": 10588,
      "time_per_iteration": 3.921621561050415
    },
    {
      "auxiliary_loss_clip": 0.01061936,
      "auxiliary_loss_mlp": 0.01027897,
      "balance_loss_clip": 1.03755939,
      "balance_loss_mlp": 1.0152545,
      "epoch": 0.6366451225011273,
      "flos": 19025689265280.0,
      "grad_norm": 1.432642160747503,
      "language_loss": 0.80006671,
      "learning_rate": 1.2321600891248478e-06,
      "loss": 0.82096499,
      "num_input_tokens_seen": 228300865,
      "step": 10589,
      "time_per_iteration": 2.6001532077789307
    },
    {
      "auxiliary_loss_clip": 0.01088129,
      "auxiliary_loss_mlp": 0.01031001,
      "balance_loss_clip": 1.03787804,
      "balance_loss_mlp": 1.01867509,
      "epoch": 0.6367052457537953,
      "flos": 25228395492480.0,
      "grad_norm": 2.014756581317245,
      "language_loss": 0.67392558,
      "learning_rate": 1.231800487863257e-06,
      "loss": 0.69511688,
      "num_input_tokens_seen": 228320815,
      "step": 10590,
      "time_per_iteration": 2.5307834148406982
    },
    {
      "auxiliary_loss_clip": 0.0110423,
      "auxiliary_loss_mlp": 0.01034873,
      "balance_loss_clip": 1.03927469,
      "balance_loss_mlp": 1.02252293,
      "epoch": 0.6367653690064633,
      "flos": 19208438686080.0,
      "grad_norm": 1.6268985843307984,
      "language_loss": 0.79291576,
      "learning_rate": 1.2314409157318685e-06,
      "loss": 0.8143068,
      "num_input_tokens_seen": 228339065,
      "step": 10591,
      "time_per_iteration": 2.4899609088897705
    },
    {
      "auxiliary_loss_clip": 0.01087565,
      "auxiliary_loss_mlp": 0.01029456,
      "balance_loss_clip": 1.04006362,
      "balance_loss_mlp": 1.01813149,
      "epoch": 0.6368254922591312,
      "flos": 23546806329600.0,
      "grad_norm": 1.476222640370243,
      "language_loss": 0.8902781,
      "learning_rate": 1.231081372744317e-06,
      "loss": 0.91144836,
      "num_input_tokens_seen": 228359210,
      "step": 10592,
      "time_per_iteration": 2.5395405292510986
    },
    {
      "auxiliary_loss_clip": 0.01095085,
      "auxiliary_loss_mlp": 0.01028206,
      "balance_loss_clip": 1.03732872,
      "balance_loss_mlp": 1.01722145,
      "epoch": 0.6368856155117992,
      "flos": 26467313443200.0,
      "grad_norm": 1.3845321305252674,
      "language_loss": 0.68453342,
      "learning_rate": 1.2307218589142376e-06,
      "loss": 0.70576632,
      "num_input_tokens_seen": 228379630,
      "step": 10593,
      "time_per_iteration": 2.5410890579223633
    },
    {
      "auxiliary_loss_clip": 0.01059228,
      "auxiliary_loss_mlp": 0.01041815,
      "balance_loss_clip": 1.03358293,
      "balance_loss_mlp": 1.02866042,
      "epoch": 0.6369457387644671,
      "flos": 33692432394240.0,
      "grad_norm": 1.6788798902175317,
      "language_loss": 0.63272697,
      "learning_rate": 1.2303623742552618e-06,
      "loss": 0.65373731,
      "num_input_tokens_seen": 228401410,
      "step": 10594,
      "time_per_iteration": 2.687072992324829
    },
    {
      "auxiliary_loss_clip": 0.01031247,
      "auxiliary_loss_mlp": 0.01003319,
      "balance_loss_clip": 1.02186799,
      "balance_loss_mlp": 1.00213242,
      "epoch": 0.6370058620171352,
      "flos": 70908600908160.0,
      "grad_norm": 0.7706815545726196,
      "language_loss": 0.54639351,
      "learning_rate": 1.230002918781022e-06,
      "loss": 0.5667392,
      "num_input_tokens_seen": 228470335,
      "step": 10595,
      "time_per_iteration": 3.2205002307891846
    },
    {
      "auxiliary_loss_clip": 0.01113028,
      "auxiliary_loss_mlp": 0.01035685,
      "balance_loss_clip": 1.03995776,
      "balance_loss_mlp": 1.02271485,
      "epoch": 0.6370659852698031,
      "flos": 21141940907520.0,
      "grad_norm": 1.716430089546519,
      "language_loss": 0.66389728,
      "learning_rate": 1.2296434925051493e-06,
      "loss": 0.68538439,
      "num_input_tokens_seen": 228490765,
      "step": 10596,
      "time_per_iteration": 2.493497371673584
    },
    {
      "auxiliary_loss_clip": 0.01094543,
      "auxiliary_loss_mlp": 0.01036079,
      "balance_loss_clip": 1.038589,
      "balance_loss_mlp": 1.02345443,
      "epoch": 0.6371261085224711,
      "flos": 20193288762240.0,
      "grad_norm": 2.533463305799403,
      "language_loss": 0.79094434,
      "learning_rate": 1.2292840954412718e-06,
      "loss": 0.81225055,
      "num_input_tokens_seen": 228509700,
      "step": 10597,
      "time_per_iteration": 2.533220052719116
    },
    {
      "auxiliary_loss_clip": 0.01100456,
      "auxiliary_loss_mlp": 0.01032748,
      "balance_loss_clip": 1.04082847,
      "balance_loss_mlp": 1.02132785,
      "epoch": 0.637186231775139,
      "flos": 19683536901120.0,
      "grad_norm": 1.6471371615898471,
      "language_loss": 0.74772608,
      "learning_rate": 1.2289247276030189e-06,
      "loss": 0.76905811,
      "num_input_tokens_seen": 228529050,
      "step": 10598,
      "time_per_iteration": 2.504966974258423
    },
    {
      "auxiliary_loss_clip": 0.01077132,
      "auxiliary_loss_mlp": 0.007904,
      "balance_loss_clip": 1.03781414,
      "balance_loss_mlp": 1.02102137,
      "epoch": 0.637246355027807,
      "flos": 13071196995840.0,
      "grad_norm": 1.8015614543218852,
      "language_loss": 0.68382251,
      "learning_rate": 1.2285653890040176e-06,
      "loss": 0.70249784,
      "num_input_tokens_seen": 228544665,
      "step": 10599,
      "time_per_iteration": 2.5411181449890137
    },
    {
      "auxiliary_loss_clip": 0.01080667,
      "auxiliary_loss_mlp": 0.01032036,
      "balance_loss_clip": 1.03808689,
      "balance_loss_mlp": 1.01914954,
      "epoch": 0.6373064782804749,
      "flos": 18222654856320.0,
      "grad_norm": 1.9271204030144624,
      "language_loss": 0.80587375,
      "learning_rate": 1.2282060796578942e-06,
      "loss": 0.8270008,
      "num_input_tokens_seen": 228562060,
      "step": 10600,
      "time_per_iteration": 2.5319392681121826
    },
    {
      "auxiliary_loss_clip": 0.01097512,
      "auxiliary_loss_mlp": 0.01032623,
      "balance_loss_clip": 1.03736496,
      "balance_loss_mlp": 1.02058268,
      "epoch": 0.637366601533143,
      "flos": 24498475217280.0,
      "grad_norm": 1.5201785343886824,
      "language_loss": 0.79992706,
      "learning_rate": 1.2278467995782732e-06,
      "loss": 0.82122838,
      "num_input_tokens_seen": 228582550,
      "step": 10601,
      "time_per_iteration": 2.537689447402954
    },
    {
      "auxiliary_loss_clip": 0.0107755,
      "auxiliary_loss_mlp": 0.01026697,
      "balance_loss_clip": 1.03894663,
      "balance_loss_mlp": 1.01497912,
      "epoch": 0.6374267247858109,
      "flos": 26359042872960.0,
      "grad_norm": 1.806291525451451,
      "language_loss": 0.66894019,
      "learning_rate": 1.2274875487787797e-06,
      "loss": 0.68998265,
      "num_input_tokens_seen": 228604960,
      "step": 10602,
      "time_per_iteration": 2.6020562648773193
    },
    {
      "auxiliary_loss_clip": 0.01035194,
      "auxiliary_loss_mlp": 0.01027661,
      "balance_loss_clip": 1.03490067,
      "balance_loss_mlp": 1.01494145,
      "epoch": 0.6374868480384789,
      "flos": 20371728551040.0,
      "grad_norm": 1.5914099239666384,
      "language_loss": 0.79837358,
      "learning_rate": 1.2271283272730354e-06,
      "loss": 0.81900203,
      "num_input_tokens_seen": 228622195,
      "step": 10603,
      "time_per_iteration": 2.6560206413269043
    },
    {
      "auxiliary_loss_clip": 0.0106639,
      "auxiliary_loss_mlp": 0.00791115,
      "balance_loss_clip": 1.03710616,
      "balance_loss_mlp": 1.02125776,
      "epoch": 0.6375469712911469,
      "flos": 20996251344000.0,
      "grad_norm": 1.874311847492079,
      "language_loss": 0.77045524,
      "learning_rate": 1.2267691350746621e-06,
      "loss": 0.78903031,
      "num_input_tokens_seen": 228639735,
      "step": 10604,
      "time_per_iteration": 2.5924839973449707
    },
    {
      "auxiliary_loss_clip": 0.01091034,
      "auxiliary_loss_mlp": 0.0102782,
      "balance_loss_clip": 1.03786027,
      "balance_loss_mlp": 1.01539803,
      "epoch": 0.6376070945438148,
      "flos": 19715748422400.0,
      "grad_norm": 1.7642229448911142,
      "language_loss": 0.76682401,
      "learning_rate": 1.226409972197281e-06,
      "loss": 0.78801262,
      "num_input_tokens_seen": 228658195,
      "step": 10605,
      "time_per_iteration": 2.539275884628296
    },
    {
      "auxiliary_loss_clip": 0.01049274,
      "auxiliary_loss_mlp": 0.01034901,
      "balance_loss_clip": 1.03460908,
      "balance_loss_mlp": 1.0202266,
      "epoch": 0.6376672177964828,
      "flos": 21506757390720.0,
      "grad_norm": 1.6943622817758432,
      "language_loss": 0.65842664,
      "learning_rate": 1.2260508386545106e-06,
      "loss": 0.67926836,
      "num_input_tokens_seen": 228677415,
      "step": 10606,
      "time_per_iteration": 2.6188032627105713
    },
    {
      "auxiliary_loss_clip": 0.01082432,
      "auxiliary_loss_mlp": 0.01030893,
      "balance_loss_clip": 1.03741598,
      "balance_loss_mlp": 1.01977038,
      "epoch": 0.6377273410491507,
      "flos": 18843873598080.0,
      "grad_norm": 1.5060057860549643,
      "language_loss": 0.74787706,
      "learning_rate": 1.225691734459971e-06,
      "loss": 0.76901031,
      "num_input_tokens_seen": 228696450,
      "step": 10607,
      "time_per_iteration": 2.5601186752319336
    },
    {
      "auxiliary_loss_clip": 0.01087619,
      "auxiliary_loss_mlp": 0.01031525,
      "balance_loss_clip": 1.04268575,
      "balance_loss_mlp": 1.01981235,
      "epoch": 0.6377874643018188,
      "flos": 53062970181120.0,
      "grad_norm": 1.625230585795342,
      "language_loss": 0.65903187,
      "learning_rate": 1.225332659627278e-06,
      "loss": 0.68022335,
      "num_input_tokens_seen": 228721600,
      "step": 10608,
      "time_per_iteration": 2.835698127746582
    },
    {
      "auxiliary_loss_clip": 0.00993806,
      "auxiliary_loss_mlp": 0.01003597,
      "balance_loss_clip": 1.02922952,
      "balance_loss_mlp": 1.00214314,
      "epoch": 0.6378475875544867,
      "flos": 65135026465920.0,
      "grad_norm": 0.7259006701989857,
      "language_loss": 0.51886308,
      "learning_rate": 1.2249736141700475e-06,
      "loss": 0.53883708,
      "num_input_tokens_seen": 228784535,
      "step": 10609,
      "time_per_iteration": 3.326773166656494
    },
    {
      "auxiliary_loss_clip": 0.01094753,
      "auxiliary_loss_mlp": 0.01023027,
      "balance_loss_clip": 1.03626096,
      "balance_loss_mlp": 1.01241696,
      "epoch": 0.6379077108071547,
      "flos": 23002759958400.0,
      "grad_norm": 1.6244175203052962,
      "language_loss": 0.74367881,
      "learning_rate": 1.2246145981018965e-06,
      "loss": 0.76485658,
      "num_input_tokens_seen": 228804110,
      "step": 10610,
      "time_per_iteration": 2.8030338287353516
    },
    {
      "auxiliary_loss_clip": 0.01027526,
      "auxiliary_loss_mlp": 0.01001043,
      "balance_loss_clip": 1.02578855,
      "balance_loss_mlp": 0.9997372,
      "epoch": 0.6379678340598226,
      "flos": 67601947610880.0,
      "grad_norm": 0.8444297028765864,
      "language_loss": 0.63132167,
      "learning_rate": 1.2242556114364364e-06,
      "loss": 0.65160728,
      "num_input_tokens_seen": 228867705,
      "step": 10611,
      "time_per_iteration": 3.192202568054199
    },
    {
      "auxiliary_loss_clip": 0.01096088,
      "auxiliary_loss_mlp": 0.01030956,
      "balance_loss_clip": 1.04088545,
      "balance_loss_mlp": 1.01874256,
      "epoch": 0.6380279573124906,
      "flos": 29680061610240.0,
      "grad_norm": 2.1266035582521443,
      "language_loss": 0.72293448,
      "learning_rate": 1.223896654187282e-06,
      "loss": 0.74420488,
      "num_input_tokens_seen": 228889215,
      "step": 10612,
      "time_per_iteration": 2.558593988418579
    },
    {
      "auxiliary_loss_clip": 0.01028617,
      "auxiliary_loss_mlp": 0.01000551,
      "balance_loss_clip": 1.02391768,
      "balance_loss_mlp": 0.99934673,
      "epoch": 0.6380880805651585,
      "flos": 66484046580480.0,
      "grad_norm": 0.7098322955574932,
      "language_loss": 0.57913148,
      "learning_rate": 1.2235377263680446e-06,
      "loss": 0.59942317,
      "num_input_tokens_seen": 228948465,
      "step": 10613,
      "time_per_iteration": 4.628596067428589
    },
    {
      "auxiliary_loss_clip": 0.01065626,
      "auxiliary_loss_mlp": 0.0103147,
      "balance_loss_clip": 1.03560281,
      "balance_loss_mlp": 1.01849449,
      "epoch": 0.6381482038178266,
      "flos": 23914998691200.0,
      "grad_norm": 1.697003726146515,
      "language_loss": 0.75300199,
      "learning_rate": 1.2231788279923334e-06,
      "loss": 0.77397299,
      "num_input_tokens_seen": 228967955,
      "step": 10614,
      "time_per_iteration": 2.6400740146636963
    },
    {
      "auxiliary_loss_clip": 0.0108878,
      "auxiliary_loss_mlp": 0.00789274,
      "balance_loss_clip": 1.03912497,
      "balance_loss_mlp": 1.01940835,
      "epoch": 0.6382083270704945,
      "flos": 24243042625920.0,
      "grad_norm": 1.9305483265645065,
      "language_loss": 0.80068582,
      "learning_rate": 1.2228199590737599e-06,
      "loss": 0.81946635,
      "num_input_tokens_seen": 228985495,
      "step": 10615,
      "time_per_iteration": 2.5855653285980225
    },
    {
      "auxiliary_loss_clip": 0.01023924,
      "auxiliary_loss_mlp": 0.01001207,
      "balance_loss_clip": 1.02456212,
      "balance_loss_mlp": 1.00005686,
      "epoch": 0.6382684503231625,
      "flos": 70775552931840.0,
      "grad_norm": 0.6609482387874969,
      "language_loss": 0.55633688,
      "learning_rate": 1.2224611196259305e-06,
      "loss": 0.57658815,
      "num_input_tokens_seen": 229052995,
      "step": 10616,
      "time_per_iteration": 3.21382999420166
    },
    {
      "auxiliary_loss_clip": 0.01081936,
      "auxiliary_loss_mlp": 0.01032061,
      "balance_loss_clip": 1.03583002,
      "balance_loss_mlp": 1.01957989,
      "epoch": 0.6383285735758305,
      "flos": 16544836621440.0,
      "grad_norm": 7.199249743760886,
      "language_loss": 0.84164095,
      "learning_rate": 1.2221023096624538e-06,
      "loss": 0.86278093,
      "num_input_tokens_seen": 229071030,
      "step": 10617,
      "time_per_iteration": 2.546757221221924
    },
    {
      "auxiliary_loss_clip": 0.01097971,
      "auxiliary_loss_mlp": 0.01038625,
      "balance_loss_clip": 1.03711343,
      "balance_loss_mlp": 1.02587509,
      "epoch": 0.6383886968284984,
      "flos": 14427651225600.0,
      "grad_norm": 1.7834522254103407,
      "language_loss": 0.87007916,
      "learning_rate": 1.221743529196936e-06,
      "loss": 0.89144516,
      "num_input_tokens_seen": 229088275,
      "step": 10618,
      "time_per_iteration": 2.479863166809082
    },
    {
      "auxiliary_loss_clip": 0.01057729,
      "auxiliary_loss_mlp": 0.01034686,
      "balance_loss_clip": 1.03724813,
      "balance_loss_mlp": 1.0240643,
      "epoch": 0.6384488200811664,
      "flos": 17929659617280.0,
      "grad_norm": 1.7048161234910315,
      "language_loss": 0.73478442,
      "learning_rate": 1.2213847782429806e-06,
      "loss": 0.75570858,
      "num_input_tokens_seen": 229105190,
      "step": 10619,
      "time_per_iteration": 2.6371538639068604
    },
    {
      "auxiliary_loss_clip": 0.01087406,
      "auxiliary_loss_mlp": 0.01036796,
      "balance_loss_clip": 1.03632832,
      "balance_loss_mlp": 1.02292585,
      "epoch": 0.6385089433338343,
      "flos": 18515578268160.0,
      "grad_norm": 1.8394048520634028,
      "language_loss": 0.76501107,
      "learning_rate": 1.221026056814193e-06,
      "loss": 0.78625315,
      "num_input_tokens_seen": 229122290,
      "step": 10620,
      "time_per_iteration": 2.513417959213257
    },
    {
      "auxiliary_loss_clip": 0.01085036,
      "auxiliary_loss_mlp": 0.01033367,
      "balance_loss_clip": 1.03768075,
      "balance_loss_mlp": 1.0214994,
      "epoch": 0.6385690665865024,
      "flos": 24753620499840.0,
      "grad_norm": 3.4623536299118802,
      "language_loss": 0.70625973,
      "learning_rate": 1.2206673649241752e-06,
      "loss": 0.72744375,
      "num_input_tokens_seen": 229141620,
      "step": 10621,
      "time_per_iteration": 2.5744690895080566
    },
    {
      "auxiliary_loss_clip": 0.01078944,
      "auxiliary_loss_mlp": 0.01027197,
      "balance_loss_clip": 1.03351283,
      "balance_loss_mlp": 1.01633143,
      "epoch": 0.6386291898391703,
      "flos": 20120569678080.0,
      "grad_norm": 2.4354088198756507,
      "language_loss": 0.77532756,
      "learning_rate": 1.220308702586529e-06,
      "loss": 0.79638892,
      "num_input_tokens_seen": 229161570,
      "step": 10622,
      "time_per_iteration": 2.5386087894439697
    },
    {
      "auxiliary_loss_clip": 0.01069983,
      "auxiliary_loss_mlp": 0.01031072,
      "balance_loss_clip": 1.03689754,
      "balance_loss_mlp": 1.01947868,
      "epoch": 0.6386893130918383,
      "flos": 16867278034560.0,
      "grad_norm": 1.982349775054232,
      "language_loss": 0.74486357,
      "learning_rate": 1.2199500698148546e-06,
      "loss": 0.76587415,
      "num_input_tokens_seen": 229178465,
      "step": 10623,
      "time_per_iteration": 5.300457715988159
    },
    {
      "auxiliary_loss_clip": 0.01082646,
      "auxiliary_loss_mlp": 0.01029706,
      "balance_loss_clip": 1.03433108,
      "balance_loss_mlp": 1.01910865,
      "epoch": 0.6387494363445062,
      "flos": 22966274718720.0,
      "grad_norm": 1.4123451840168164,
      "language_loss": 0.76775539,
      "learning_rate": 1.2195914666227527e-06,
      "loss": 0.78887892,
      "num_input_tokens_seen": 229198975,
      "step": 10624,
      "time_per_iteration": 2.5468196868896484
    },
    {
      "auxiliary_loss_clip": 0.01046308,
      "auxiliary_loss_mlp": 0.01038088,
      "balance_loss_clip": 1.03643274,
      "balance_loss_mlp": 1.02594066,
      "epoch": 0.6388095595971742,
      "flos": 22857716839680.0,
      "grad_norm": 1.8563133958409395,
      "language_loss": 0.80021334,
      "learning_rate": 1.21923289302382e-06,
      "loss": 0.82105732,
      "num_input_tokens_seen": 229218825,
      "step": 10625,
      "time_per_iteration": 2.6852328777313232
    },
    {
      "auxiliary_loss_clip": 0.01086704,
      "auxiliary_loss_mlp": 0.01036439,
      "balance_loss_clip": 1.0389812,
      "balance_loss_mlp": 1.0244348,
      "epoch": 0.6388696828498421,
      "flos": 17311529445120.0,
      "grad_norm": 1.9805391918811175,
      "language_loss": 0.73081028,
      "learning_rate": 1.218874349031654e-06,
      "loss": 0.7520417,
      "num_input_tokens_seen": 229236060,
      "step": 10626,
      "time_per_iteration": 3.8844687938690186
    },
    {
      "auxiliary_loss_clip": 0.01087223,
      "auxiliary_loss_mlp": 0.01031883,
      "balance_loss_clip": 1.03671002,
      "balance_loss_mlp": 1.01958609,
      "epoch": 0.6389298061025102,
      "flos": 17128636369920.0,
      "grad_norm": 1.7459912572064038,
      "language_loss": 0.73049319,
      "learning_rate": 1.2185158346598517e-06,
      "loss": 0.75168431,
      "num_input_tokens_seen": 229255160,
      "step": 10627,
      "time_per_iteration": 2.5216500759124756
    },
    {
      "auxiliary_loss_clip": 0.01084389,
      "auxiliary_loss_mlp": 0.01032766,
      "balance_loss_clip": 1.0376972,
      "balance_loss_mlp": 1.01897335,
      "epoch": 0.6389899293551781,
      "flos": 27710971989120.0,
      "grad_norm": 1.7498744173717078,
      "language_loss": 0.66715705,
      "learning_rate": 1.2181573499220064e-06,
      "loss": 0.68832862,
      "num_input_tokens_seen": 229278705,
      "step": 10628,
      "time_per_iteration": 2.6234233379364014
    },
    {
      "auxiliary_loss_clip": 0.01103771,
      "auxiliary_loss_mlp": 0.01027333,
      "balance_loss_clip": 1.03736091,
      "balance_loss_mlp": 1.01625872,
      "epoch": 0.6390500526078461,
      "flos": 21215701486080.0,
      "grad_norm": 1.8088700466854297,
      "language_loss": 0.68440163,
      "learning_rate": 1.2177988948317135e-06,
      "loss": 0.70571268,
      "num_input_tokens_seen": 229299990,
      "step": 10629,
      "time_per_iteration": 2.5285661220550537
    },
    {
      "auxiliary_loss_clip": 0.01073556,
      "auxiliary_loss_mlp": 0.01038788,
      "balance_loss_clip": 1.03982925,
      "balance_loss_mlp": 1.02466726,
      "epoch": 0.6391101758605141,
      "flos": 21581056673280.0,
      "grad_norm": 1.525273910218152,
      "language_loss": 0.75580662,
      "learning_rate": 1.2174404694025646e-06,
      "loss": 0.77693009,
      "num_input_tokens_seen": 229319230,
      "step": 10630,
      "time_per_iteration": 2.585092067718506
    },
    {
      "auxiliary_loss_clip": 0.01084004,
      "auxiliary_loss_mlp": 0.01034175,
      "balance_loss_clip": 1.03681135,
      "balance_loss_mlp": 1.02271867,
      "epoch": 0.639170299113182,
      "flos": 19900473091200.0,
      "grad_norm": 1.68123864272701,
      "language_loss": 0.70551634,
      "learning_rate": 1.2170820736481511e-06,
      "loss": 0.72669816,
      "num_input_tokens_seen": 229338600,
      "step": 10631,
      "time_per_iteration": 2.553377628326416
    },
    {
      "auxiliary_loss_clip": 0.01023401,
      "auxiliary_loss_mlp": 0.01003214,
      "balance_loss_clip": 1.02293193,
      "balance_loss_mlp": 1.0016408,
      "epoch": 0.63923042236585,
      "flos": 69877604833920.0,
      "grad_norm": 0.7669052218146686,
      "language_loss": 0.63028049,
      "learning_rate": 1.2167237075820646e-06,
      "loss": 0.65054661,
      "num_input_tokens_seen": 229402420,
      "step": 10632,
      "time_per_iteration": 3.1966795921325684
    },
    {
      "auxiliary_loss_clip": 0.01084947,
      "auxiliary_loss_mlp": 0.01034015,
      "balance_loss_clip": 1.03842092,
      "balance_loss_mlp": 1.02209973,
      "epoch": 0.639290545618518,
      "flos": 22674823764480.0,
      "grad_norm": 1.9294946291815969,
      "language_loss": 0.66229236,
      "learning_rate": 1.216365371217893e-06,
      "loss": 0.68348199,
      "num_input_tokens_seen": 229419185,
      "step": 10633,
      "time_per_iteration": 2.552764415740967
    },
    {
      "auxiliary_loss_clip": 0.01041353,
      "auxiliary_loss_mlp": 0.01032866,
      "balance_loss_clip": 1.03705561,
      "balance_loss_mlp": 1.0213387,
      "epoch": 0.639350668871186,
      "flos": 19829190551040.0,
      "grad_norm": 1.9667574030115986,
      "language_loss": 0.82007074,
      "learning_rate": 1.216007064569225e-06,
      "loss": 0.84081304,
      "num_input_tokens_seen": 229436735,
      "step": 10634,
      "time_per_iteration": 2.6581647396087646
    },
    {
      "auxiliary_loss_clip": 0.01084513,
      "auxiliary_loss_mlp": 0.01033768,
      "balance_loss_clip": 1.03957093,
      "balance_loss_mlp": 1.02049422,
      "epoch": 0.6394107921238539,
      "flos": 20553328736640.0,
      "grad_norm": 1.5998754826832327,
      "language_loss": 0.75050831,
      "learning_rate": 1.2156487876496483e-06,
      "loss": 0.77169108,
      "num_input_tokens_seen": 229455595,
      "step": 10635,
      "time_per_iteration": 2.5308456420898438
    },
    {
      "auxiliary_loss_clip": 0.0109347,
      "auxiliary_loss_mlp": 0.01031985,
      "balance_loss_clip": 1.03821421,
      "balance_loss_mlp": 1.01972461,
      "epoch": 0.6394709153765219,
      "flos": 25774991729280.0,
      "grad_norm": 1.6398251517255955,
      "language_loss": 0.71796834,
      "learning_rate": 1.2152905404727475e-06,
      "loss": 0.73922288,
      "num_input_tokens_seen": 229476230,
      "step": 10636,
      "time_per_iteration": 2.5443918704986572
    },
    {
      "auxiliary_loss_clip": 0.0108799,
      "auxiliary_loss_mlp": 0.01036748,
      "balance_loss_clip": 1.03783464,
      "balance_loss_mlp": 1.02408767,
      "epoch": 0.6395310386291898,
      "flos": 17530153574400.0,
      "grad_norm": 2.3152411707440383,
      "language_loss": 0.73855585,
      "learning_rate": 1.2149323230521085e-06,
      "loss": 0.75980318,
      "num_input_tokens_seen": 229494300,
      "step": 10637,
      "time_per_iteration": 2.5376334190368652
    },
    {
      "auxiliary_loss_clip": 0.01093353,
      "auxiliary_loss_mlp": 0.0103101,
      "balance_loss_clip": 1.03733158,
      "balance_loss_mlp": 1.01874948,
      "epoch": 0.6395911618818578,
      "flos": 18588225525120.0,
      "grad_norm": 1.7574470293202038,
      "language_loss": 0.7759431,
      "learning_rate": 1.2145741354013143e-06,
      "loss": 0.79718673,
      "num_input_tokens_seen": 229512985,
      "step": 10638,
      "time_per_iteration": 2.4902961254119873
    },
    {
      "auxiliary_loss_clip": 0.01085546,
      "auxiliary_loss_mlp": 0.01033191,
      "balance_loss_clip": 1.03722572,
      "balance_loss_mlp": 1.02097845,
      "epoch": 0.6396512851345257,
      "flos": 28366557068160.0,
      "grad_norm": 1.6472630730060214,
      "language_loss": 0.8168813,
      "learning_rate": 1.2142159775339478e-06,
      "loss": 0.8380686,
      "num_input_tokens_seen": 229534270,
      "step": 10639,
      "time_per_iteration": 2.623180627822876
    },
    {
      "auxiliary_loss_clip": 0.01024798,
      "auxiliary_loss_mlp": 0.01008083,
      "balance_loss_clip": 1.02058208,
      "balance_loss_mlp": 1.0069741,
      "epoch": 0.6397114083871938,
      "flos": 70724307202560.0,
      "grad_norm": 0.8097543018368745,
      "language_loss": 0.59057498,
      "learning_rate": 1.21385784946359e-06,
      "loss": 0.6109038,
      "num_input_tokens_seen": 229596455,
      "step": 10640,
      "time_per_iteration": 3.094064950942993
    },
    {
      "auxiliary_loss_clip": 0.01077691,
      "auxiliary_loss_mlp": 0.01029952,
      "balance_loss_clip": 1.03505087,
      "balance_loss_mlp": 1.01860905,
      "epoch": 0.6397715316398617,
      "flos": 18142537570560.0,
      "grad_norm": 2.0004483854090194,
      "language_loss": 0.78249389,
      "learning_rate": 1.2134997512038215e-06,
      "loss": 0.80357039,
      "num_input_tokens_seen": 229612860,
      "step": 10641,
      "time_per_iteration": 2.55521559715271
    },
    {
      "auxiliary_loss_clip": 0.01064215,
      "auxiliary_loss_mlp": 0.01034078,
      "balance_loss_clip": 1.03840911,
      "balance_loss_mlp": 1.02188921,
      "epoch": 0.6398316548925297,
      "flos": 25739512070400.0,
      "grad_norm": 1.7605405144010378,
      "language_loss": 0.63275725,
      "learning_rate": 1.2131416827682209e-06,
      "loss": 0.65374017,
      "num_input_tokens_seen": 229633960,
      "step": 10642,
      "time_per_iteration": 2.6367416381835938
    },
    {
      "auxiliary_loss_clip": 0.01023784,
      "auxiliary_loss_mlp": 0.01009325,
      "balance_loss_clip": 1.01989901,
      "balance_loss_mlp": 1.00781715,
      "epoch": 0.6398917781451977,
      "flos": 71214234756480.0,
      "grad_norm": 0.9186626590139303,
      "language_loss": 0.55989778,
      "learning_rate": 1.2127836441703667e-06,
      "loss": 0.58022892,
      "num_input_tokens_seen": 229686730,
      "step": 10643,
      "time_per_iteration": 3.0957283973693848
    },
    {
      "auxiliary_loss_clip": 0.01076696,
      "auxiliary_loss_mlp": 0.01025561,
      "balance_loss_clip": 1.03767824,
      "balance_loss_mlp": 1.01366985,
      "epoch": 0.6399519013978656,
      "flos": 20521835487360.0,
      "grad_norm": 1.869268047431947,
      "language_loss": 0.76367986,
      "learning_rate": 1.2124256354238358e-06,
      "loss": 0.78470242,
      "num_input_tokens_seen": 229704800,
      "step": 10644,
      "time_per_iteration": 2.589582920074463
    },
    {
      "auxiliary_loss_clip": 0.01074414,
      "auxiliary_loss_mlp": 0.0103113,
      "balance_loss_clip": 1.03845167,
      "balance_loss_mlp": 1.01866651,
      "epoch": 0.6400120246505336,
      "flos": 24460840742400.0,
      "grad_norm": 1.5506676687480787,
      "language_loss": 0.82183152,
      "learning_rate": 1.212067656542203e-06,
      "loss": 0.84288692,
      "num_input_tokens_seen": 229725265,
      "step": 10645,
      "time_per_iteration": 2.626453161239624
    },
    {
      "auxiliary_loss_clip": 0.01101633,
      "auxiliary_loss_mlp": 0.01037156,
      "balance_loss_clip": 1.03813922,
      "balance_loss_mlp": 1.02339339,
      "epoch": 0.6400721479032015,
      "flos": 28366090191360.0,
      "grad_norm": 1.8248219168910742,
      "language_loss": 0.73576993,
      "learning_rate": 1.2117097075390447e-06,
      "loss": 0.7571578,
      "num_input_tokens_seen": 229744840,
      "step": 10646,
      "time_per_iteration": 2.5459043979644775
    },
    {
      "auxiliary_loss_clip": 0.01070311,
      "auxiliary_loss_mlp": 0.01033661,
      "balance_loss_clip": 1.03601599,
      "balance_loss_mlp": 1.02047634,
      "epoch": 0.6401322711558696,
      "flos": 17816540711040.0,
      "grad_norm": 3.0329530619297964,
      "language_loss": 0.79693675,
      "learning_rate": 1.2113517884279327e-06,
      "loss": 0.81797642,
      "num_input_tokens_seen": 229759095,
      "step": 10647,
      "time_per_iteration": 2.561713933944702
    },
    {
      "auxiliary_loss_clip": 0.01063798,
      "auxiliary_loss_mlp": 0.01029492,
      "balance_loss_clip": 1.03699231,
      "balance_loss_mlp": 1.01746941,
      "epoch": 0.6401923944085375,
      "flos": 26030855283840.0,
      "grad_norm": 1.4714823235803258,
      "language_loss": 0.75659335,
      "learning_rate": 1.2109938992224399e-06,
      "loss": 0.77752626,
      "num_input_tokens_seen": 229777750,
      "step": 10648,
      "time_per_iteration": 2.618218183517456
    },
    {
      "auxiliary_loss_clip": 0.01079695,
      "auxiliary_loss_mlp": 0.01031251,
      "balance_loss_clip": 1.03484452,
      "balance_loss_mlp": 1.01912701,
      "epoch": 0.6402525176612055,
      "flos": 23586451966080.0,
      "grad_norm": 1.8215906032668436,
      "language_loss": 0.78631699,
      "learning_rate": 1.210636039936138e-06,
      "loss": 0.80742645,
      "num_input_tokens_seen": 229796785,
      "step": 10649,
      "time_per_iteration": 2.5896358489990234
    },
    {
      "auxiliary_loss_clip": 0.0105315,
      "auxiliary_loss_mlp": 0.01034399,
      "balance_loss_clip": 1.03871858,
      "balance_loss_mlp": 1.02163172,
      "epoch": 0.6403126409138734,
      "flos": 18041413806720.0,
      "grad_norm": 2.0627578325323275,
      "language_loss": 0.75343543,
      "learning_rate": 1.2102782105825956e-06,
      "loss": 0.77431095,
      "num_input_tokens_seen": 229815425,
      "step": 10650,
      "time_per_iteration": 2.6042821407318115
    },
    {
      "auxiliary_loss_clip": 0.01106651,
      "auxiliary_loss_mlp": 0.01033172,
      "balance_loss_clip": 1.03691292,
      "balance_loss_mlp": 1.02023828,
      "epoch": 0.6403727641665414,
      "flos": 21979485308160.0,
      "grad_norm": 2.124133180054986,
      "language_loss": 0.70639223,
      "learning_rate": 1.2099204111753833e-06,
      "loss": 0.72779042,
      "num_input_tokens_seen": 229834545,
      "step": 10651,
      "time_per_iteration": 2.506472587585449
    },
    {
      "auxiliary_loss_clip": 0.01076281,
      "auxiliary_loss_mlp": 0.01035259,
      "balance_loss_clip": 1.03668284,
      "balance_loss_mlp": 1.02267003,
      "epoch": 0.6404328874192093,
      "flos": 24895539135360.0,
      "grad_norm": 2.3233289959957935,
      "language_loss": 0.63792026,
      "learning_rate": 1.2095626417280684e-06,
      "loss": 0.65903568,
      "num_input_tokens_seen": 229849175,
      "step": 10652,
      "time_per_iteration": 3.994550943374634
    },
    {
      "auxiliary_loss_clip": 0.01086032,
      "auxiliary_loss_mlp": 0.01026127,
      "balance_loss_clip": 1.03702414,
      "balance_loss_mlp": 1.01421189,
      "epoch": 0.6404930106718774,
      "flos": 17597198309760.0,
      "grad_norm": 1.945934460867734,
      "language_loss": 0.79144204,
      "learning_rate": 1.2092049022542168e-06,
      "loss": 0.8125636,
      "num_input_tokens_seen": 229865400,
      "step": 10653,
      "time_per_iteration": 2.5494399070739746
    },
    {
      "auxiliary_loss_clip": 0.0108468,
      "auxiliary_loss_mlp": 0.01053337,
      "balance_loss_clip": 1.03476858,
      "balance_loss_mlp": 1.0374285,
      "epoch": 0.6405531339245453,
      "flos": 20157880930560.0,
      "grad_norm": 2.506373753271721,
      "language_loss": 0.70548904,
      "learning_rate": 1.2088471927673952e-06,
      "loss": 0.72686911,
      "num_input_tokens_seen": 229882945,
      "step": 10654,
      "time_per_iteration": 2.540357828140259
    },
    {
      "auxiliary_loss_clip": 0.0110224,
      "auxiliary_loss_mlp": 0.01038472,
      "balance_loss_clip": 1.03888631,
      "balance_loss_mlp": 1.02534103,
      "epoch": 0.6406132571772133,
      "flos": 21942281796480.0,
      "grad_norm": 1.9879002018190768,
      "language_loss": 0.72897911,
      "learning_rate": 1.2084895132811666e-06,
      "loss": 0.75038624,
      "num_input_tokens_seen": 229901590,
      "step": 10655,
      "time_per_iteration": 2.5071895122528076
    },
    {
      "auxiliary_loss_clip": 0.01075262,
      "auxiliary_loss_mlp": 0.01031901,
      "balance_loss_clip": 1.03950953,
      "balance_loss_mlp": 1.01967013,
      "epoch": 0.6406733804298813,
      "flos": 28768002445440.0,
      "grad_norm": 1.6574084248589094,
      "language_loss": 0.82783759,
      "learning_rate": 1.2081318638090952e-06,
      "loss": 0.8489092,
      "num_input_tokens_seen": 229922535,
      "step": 10656,
      "time_per_iteration": 2.632880687713623
    },
    {
      "auxiliary_loss_clip": 0.01059402,
      "auxiliary_loss_mlp": 0.01034412,
      "balance_loss_clip": 1.03604412,
      "balance_loss_mlp": 1.02277684,
      "epoch": 0.6407335036825492,
      "flos": 17457183095040.0,
      "grad_norm": 2.5307208687193583,
      "language_loss": 0.72402805,
      "learning_rate": 1.2077742443647433e-06,
      "loss": 0.74496621,
      "num_input_tokens_seen": 229939575,
      "step": 10657,
      "time_per_iteration": 2.5616750717163086
    },
    {
      "auxiliary_loss_clip": 0.01076301,
      "auxiliary_loss_mlp": 0.01034233,
      "balance_loss_clip": 1.03701544,
      "balance_loss_mlp": 1.02262223,
      "epoch": 0.6407936269352172,
      "flos": 22125282612480.0,
      "grad_norm": 1.6143709492864202,
      "language_loss": 0.77109498,
      "learning_rate": 1.2074166549616707e-06,
      "loss": 0.79220033,
      "num_input_tokens_seen": 229958840,
      "step": 10658,
      "time_per_iteration": 2.5812625885009766
    },
    {
      "auxiliary_loss_clip": 0.01109909,
      "auxiliary_loss_mlp": 0.01033437,
      "balance_loss_clip": 1.03756893,
      "balance_loss_mlp": 1.02066398,
      "epoch": 0.6408537501878852,
      "flos": 23110635479040.0,
      "grad_norm": 1.6040513429097012,
      "language_loss": 0.76382488,
      "learning_rate": 1.2070590956134386e-06,
      "loss": 0.78525835,
      "num_input_tokens_seen": 229979680,
      "step": 10659,
      "time_per_iteration": 2.5037031173706055
    },
    {
      "auxiliary_loss_clip": 0.01097619,
      "auxiliary_loss_mlp": 0.01031433,
      "balance_loss_clip": 1.03777409,
      "balance_loss_mlp": 1.01901102,
      "epoch": 0.6409138734405532,
      "flos": 16472440759680.0,
      "grad_norm": 1.7913422737024063,
      "language_loss": 0.78062034,
      "learning_rate": 1.2067015663336046e-06,
      "loss": 0.80191088,
      "num_input_tokens_seen": 229996830,
      "step": 10660,
      "time_per_iteration": 2.483149766921997
    },
    {
      "auxiliary_loss_clip": 0.01087589,
      "auxiliary_loss_mlp": 0.010357,
      "balance_loss_clip": 1.0388366,
      "balance_loss_mlp": 1.02194905,
      "epoch": 0.6409739966932211,
      "flos": 22777922776320.0,
      "grad_norm": 1.757879581550092,
      "language_loss": 0.68496054,
      "learning_rate": 1.206344067135727e-06,
      "loss": 0.70619345,
      "num_input_tokens_seen": 230015115,
      "step": 10661,
      "time_per_iteration": 2.5265753269195557
    },
    {
      "auxiliary_loss_clip": 0.01106617,
      "auxiliary_loss_mlp": 0.01033158,
      "balance_loss_clip": 1.03826094,
      "balance_loss_mlp": 1.02225637,
      "epoch": 0.6410341199458891,
      "flos": 25152049134720.0,
      "grad_norm": 1.5420823415939664,
      "language_loss": 0.75890774,
      "learning_rate": 1.205986598033362e-06,
      "loss": 0.7803055,
      "num_input_tokens_seen": 230035515,
      "step": 10662,
      "time_per_iteration": 3.891908645629883
    },
    {
      "auxiliary_loss_clip": 0.01092904,
      "auxiliary_loss_mlp": 0.01035063,
      "balance_loss_clip": 1.03514111,
      "balance_loss_mlp": 1.02218819,
      "epoch": 0.641094243198557,
      "flos": 27046193028480.0,
      "grad_norm": 1.9166916891724162,
      "language_loss": 0.70033383,
      "learning_rate": 1.2056291590400644e-06,
      "loss": 0.72161353,
      "num_input_tokens_seen": 230054355,
      "step": 10663,
      "time_per_iteration": 2.5537500381469727
    },
    {
      "auxiliary_loss_clip": 0.01075068,
      "auxiliary_loss_mlp": 0.0103879,
      "balance_loss_clip": 1.03740311,
      "balance_loss_mlp": 1.0250566,
      "epoch": 0.641154366451225,
      "flos": 25374551932800.0,
      "grad_norm": 1.9311504788833653,
      "language_loss": 0.68331772,
      "learning_rate": 1.205271750169389e-06,
      "loss": 0.70445633,
      "num_input_tokens_seen": 230074605,
      "step": 10664,
      "time_per_iteration": 2.615206241607666
    },
    {
      "auxiliary_loss_clip": 0.01083814,
      "auxiliary_loss_mlp": 0.01030767,
      "balance_loss_clip": 1.03548193,
      "balance_loss_mlp": 1.01928163,
      "epoch": 0.6412144897038929,
      "flos": 25153342024320.0,
      "grad_norm": 1.8779500932300415,
      "language_loss": 0.66696513,
      "learning_rate": 1.2049143714348881e-06,
      "loss": 0.68811095,
      "num_input_tokens_seen": 230093820,
      "step": 10665,
      "time_per_iteration": 3.9181010723114014
    },
    {
      "auxiliary_loss_clip": 0.01096171,
      "auxiliary_loss_mlp": 0.01027476,
      "balance_loss_clip": 1.036533,
      "balance_loss_mlp": 1.01583552,
      "epoch": 0.641274612956561,
      "flos": 23440762402560.0,
      "grad_norm": 1.605038628058924,
      "language_loss": 0.645751,
      "learning_rate": 1.2045570228501145e-06,
      "loss": 0.66698748,
      "num_input_tokens_seen": 230114285,
      "step": 10666,
      "time_per_iteration": 2.514486312866211
    },
    {
      "auxiliary_loss_clip": 0.01098931,
      "auxiliary_loss_mlp": 0.01032719,
      "balance_loss_clip": 1.03799176,
      "balance_loss_mlp": 1.02086365,
      "epoch": 0.6413347362092289,
      "flos": 19427493778560.0,
      "grad_norm": 1.5483041282076118,
      "language_loss": 0.71107274,
      "learning_rate": 1.2041997044286176e-06,
      "loss": 0.73238927,
      "num_input_tokens_seen": 230132760,
      "step": 10667,
      "time_per_iteration": 2.500885009765625
    },
    {
      "auxiliary_loss_clip": 0.01063031,
      "auxiliary_loss_mlp": 0.00789424,
      "balance_loss_clip": 1.03703284,
      "balance_loss_mlp": 1.0124836,
      "epoch": 0.6413948594618969,
      "flos": 17196578945280.0,
      "grad_norm": 2.1429649200452947,
      "language_loss": 0.7761206,
      "learning_rate": 1.2038424161839484e-06,
      "loss": 0.79464513,
      "num_input_tokens_seen": 230149690,
      "step": 10668,
      "time_per_iteration": 2.56838321685791
    },
    {
      "auxiliary_loss_clip": 0.01100436,
      "auxiliary_loss_mlp": 0.01034341,
      "balance_loss_clip": 1.03936613,
      "balance_loss_mlp": 1.02212846,
      "epoch": 0.6414549827145648,
      "flos": 22269787027200.0,
      "grad_norm": 1.515874035966659,
      "language_loss": 0.67534292,
      "learning_rate": 1.2034851581296544e-06,
      "loss": 0.69669074,
      "num_input_tokens_seen": 230166950,
      "step": 10669,
      "time_per_iteration": 2.5067434310913086
    },
    {
      "auxiliary_loss_clip": 0.01107529,
      "auxiliary_loss_mlp": 0.01036429,
      "balance_loss_clip": 1.04118359,
      "balance_loss_mlp": 1.02344155,
      "epoch": 0.6415151059672328,
      "flos": 19640192163840.0,
      "grad_norm": 2.0794713233847877,
      "language_loss": 0.78583986,
      "learning_rate": 1.2031279302792825e-06,
      "loss": 0.80727947,
      "num_input_tokens_seen": 230184785,
      "step": 10670,
      "time_per_iteration": 2.4940426349639893
    },
    {
      "auxiliary_loss_clip": 0.01075647,
      "auxiliary_loss_mlp": 0.01028839,
      "balance_loss_clip": 1.03982055,
      "balance_loss_mlp": 1.01607776,
      "epoch": 0.6415752292199008,
      "flos": 14865833237760.0,
      "grad_norm": 2.5087119477926425,
      "language_loss": 0.88529766,
      "learning_rate": 1.20277073264638e-06,
      "loss": 0.90634251,
      "num_input_tokens_seen": 230201385,
      "step": 10671,
      "time_per_iteration": 2.5395915508270264
    },
    {
      "auxiliary_loss_clip": 0.01097744,
      "auxiliary_loss_mlp": 0.0102375,
      "balance_loss_clip": 1.03886104,
      "balance_loss_mlp": 1.01241875,
      "epoch": 0.6416353524725688,
      "flos": 13735580906880.0,
      "grad_norm": 1.482995887741132,
      "language_loss": 0.68903661,
      "learning_rate": 1.2024135652444907e-06,
      "loss": 0.71025157,
      "num_input_tokens_seen": 230220380,
      "step": 10672,
      "time_per_iteration": 2.4820749759674072
    },
    {
      "auxiliary_loss_clip": 0.01103532,
      "auxiliary_loss_mlp": 0.01033143,
      "balance_loss_clip": 1.03869033,
      "balance_loss_mlp": 1.01925528,
      "epoch": 0.6416954757252368,
      "flos": 24534924543360.0,
      "grad_norm": 2.1619931151472205,
      "language_loss": 0.74549836,
      "learning_rate": 1.2020564280871593e-06,
      "loss": 0.76686513,
      "num_input_tokens_seen": 230239845,
      "step": 10673,
      "time_per_iteration": 2.523625373840332
    },
    {
      "auxiliary_loss_clip": 0.01073728,
      "auxiliary_loss_mlp": 0.01036104,
      "balance_loss_clip": 1.03991282,
      "balance_loss_mlp": 1.02327693,
      "epoch": 0.6417555989779047,
      "flos": 27710002321920.0,
      "grad_norm": 1.5930787829842543,
      "language_loss": 0.69138294,
      "learning_rate": 1.2016993211879283e-06,
      "loss": 0.71248126,
      "num_input_tokens_seen": 230262420,
      "step": 10674,
      "time_per_iteration": 2.6199512481689453
    },
    {
      "auxiliary_loss_clip": 0.01113578,
      "auxiliary_loss_mlp": 0.01034777,
      "balance_loss_clip": 1.03821135,
      "balance_loss_mlp": 1.02205193,
      "epoch": 0.6418157222305727,
      "flos": 20556632787840.0,
      "grad_norm": 1.8267320230228428,
      "language_loss": 0.66165471,
      "learning_rate": 1.201342244560338e-06,
      "loss": 0.68313825,
      "num_input_tokens_seen": 230279950,
      "step": 10675,
      "time_per_iteration": 2.4864213466644287
    },
    {
      "auxiliary_loss_clip": 0.01111771,
      "auxiliary_loss_mlp": 0.01036542,
      "balance_loss_clip": 1.04024839,
      "balance_loss_mlp": 1.02460313,
      "epoch": 0.6418758454832406,
      "flos": 22601530062720.0,
      "grad_norm": 1.7515998993860813,
      "language_loss": 0.6624611,
      "learning_rate": 1.2009851982179307e-06,
      "loss": 0.68394423,
      "num_input_tokens_seen": 230299705,
      "step": 10676,
      "time_per_iteration": 2.503316640853882
    },
    {
      "auxiliary_loss_clip": 0.0111234,
      "auxiliary_loss_mlp": 0.01028498,
      "balance_loss_clip": 1.04013729,
      "balance_loss_mlp": 1.01450264,
      "epoch": 0.6419359687359086,
      "flos": 27375098889600.0,
      "grad_norm": 2.025926092179471,
      "language_loss": 0.75690615,
      "learning_rate": 1.2006281821742446e-06,
      "loss": 0.77831453,
      "num_input_tokens_seen": 230320030,
      "step": 10677,
      "time_per_iteration": 2.514413356781006
    },
    {
      "auxiliary_loss_clip": 0.01026168,
      "auxiliary_loss_mlp": 0.01003983,
      "balance_loss_clip": 1.02320063,
      "balance_loss_mlp": 1.00276744,
      "epoch": 0.6419960919885765,
      "flos": 67251924552960.0,
      "grad_norm": 0.7674776320867205,
      "language_loss": 0.60715884,
      "learning_rate": 1.200271196442818e-06,
      "loss": 0.6274603,
      "num_input_tokens_seen": 230381495,
      "step": 10678,
      "time_per_iteration": 3.224430561065674
    },
    {
      "auxiliary_loss_clip": 0.0109653,
      "auxiliary_loss_mlp": 0.01035258,
      "balance_loss_clip": 1.03887713,
      "balance_loss_mlp": 1.02336085,
      "epoch": 0.6420562152412446,
      "flos": 19901873721600.0,
      "grad_norm": 1.6599348145572048,
      "language_loss": 0.67208672,
      "learning_rate": 1.1999142410371875e-06,
      "loss": 0.69340456,
      "num_input_tokens_seen": 230401385,
      "step": 10679,
      "time_per_iteration": 2.5160202980041504
    },
    {
      "auxiliary_loss_clip": 0.01098628,
      "auxiliary_loss_mlp": 0.01039237,
      "balance_loss_clip": 1.03889012,
      "balance_loss_mlp": 1.02445543,
      "epoch": 0.6421163384939125,
      "flos": 24790177566720.0,
      "grad_norm": 1.6922279659592054,
      "language_loss": 0.72859657,
      "learning_rate": 1.1995573159708897e-06,
      "loss": 0.74997526,
      "num_input_tokens_seen": 230421340,
      "step": 10680,
      "time_per_iteration": 2.5372424125671387
    },
    {
      "auxiliary_loss_clip": 0.01071857,
      "auxiliary_loss_mlp": 0.01026686,
      "balance_loss_clip": 1.03520751,
      "balance_loss_mlp": 1.01578999,
      "epoch": 0.6421764617465805,
      "flos": 25592816926080.0,
      "grad_norm": 1.7693621300510207,
      "language_loss": 0.6754694,
      "learning_rate": 1.1992004212574582e-06,
      "loss": 0.69645476,
      "num_input_tokens_seen": 230441270,
      "step": 10681,
      "time_per_iteration": 2.595010995864868
    },
    {
      "auxiliary_loss_clip": 0.01105926,
      "auxiliary_loss_mlp": 0.01027745,
      "balance_loss_clip": 1.03621423,
      "balance_loss_mlp": 1.01591969,
      "epoch": 0.6422365849992484,
      "flos": 14134727813760.0,
      "grad_norm": 1.9759917465024635,
      "language_loss": 0.74773932,
      "learning_rate": 1.198843556910427e-06,
      "loss": 0.76907599,
      "num_input_tokens_seen": 230457455,
      "step": 10682,
      "time_per_iteration": 2.446984052658081
    },
    {
      "auxiliary_loss_clip": 0.01045584,
      "auxiliary_loss_mlp": 0.01027991,
      "balance_loss_clip": 1.03712058,
      "balance_loss_mlp": 1.016976,
      "epoch": 0.6422967082519164,
      "flos": 22383911514240.0,
      "grad_norm": 1.4183250818430206,
      "language_loss": 0.7905221,
      "learning_rate": 1.1984867229433287e-06,
      "loss": 0.81125778,
      "num_input_tokens_seen": 230478955,
      "step": 10683,
      "time_per_iteration": 2.6508278846740723
    },
    {
      "auxiliary_loss_clip": 0.01110467,
      "auxiliary_loss_mlp": 0.01034158,
      "balance_loss_clip": 1.03834367,
      "balance_loss_mlp": 1.02148557,
      "epoch": 0.6423568315045844,
      "flos": 14647927380480.0,
      "grad_norm": 1.820414586406883,
      "language_loss": 0.67033017,
      "learning_rate": 1.1981299193696941e-06,
      "loss": 0.69177639,
      "num_input_tokens_seen": 230496425,
      "step": 10684,
      "time_per_iteration": 2.4523346424102783
    },
    {
      "auxiliary_loss_clip": 0.01098402,
      "auxiliary_loss_mlp": 0.01033185,
      "balance_loss_clip": 1.03757918,
      "balance_loss_mlp": 1.02088892,
      "epoch": 0.6424169547572524,
      "flos": 26833925606400.0,
      "grad_norm": 2.096095937857753,
      "language_loss": 0.71103621,
      "learning_rate": 1.1977731462030533e-06,
      "loss": 0.73235202,
      "num_input_tokens_seen": 230516245,
      "step": 10685,
      "time_per_iteration": 2.5548653602600098
    },
    {
      "auxiliary_loss_clip": 0.01071244,
      "auxiliary_loss_mlp": 0.01032009,
      "balance_loss_clip": 1.03491533,
      "balance_loss_mlp": 1.02060032,
      "epoch": 0.6424770780099204,
      "flos": 22707430335360.0,
      "grad_norm": 1.529714359533987,
      "language_loss": 0.75596321,
      "learning_rate": 1.197416403456935e-06,
      "loss": 0.77699578,
      "num_input_tokens_seen": 230534745,
      "step": 10686,
      "time_per_iteration": 2.574174642562866
    },
    {
      "auxiliary_loss_clip": 0.01077799,
      "auxiliary_loss_mlp": 0.01030736,
      "balance_loss_clip": 1.04172373,
      "balance_loss_mlp": 1.01783204,
      "epoch": 0.6425372012625883,
      "flos": 28469512425600.0,
      "grad_norm": 2.9893543470126636,
      "language_loss": 0.68566996,
      "learning_rate": 1.197059691144867e-06,
      "loss": 0.70675528,
      "num_input_tokens_seen": 230555895,
      "step": 10687,
      "time_per_iteration": 2.6129000186920166
    },
    {
      "auxiliary_loss_clip": 0.01083928,
      "auxiliary_loss_mlp": 0.0103331,
      "balance_loss_clip": 1.03662455,
      "balance_loss_mlp": 1.02087033,
      "epoch": 0.6425973245152563,
      "flos": 29351694453120.0,
      "grad_norm": 1.9504893445404612,
      "language_loss": 0.66426373,
      "learning_rate": 1.1967030092803767e-06,
      "loss": 0.68543607,
      "num_input_tokens_seen": 230577460,
      "step": 10688,
      "time_per_iteration": 2.606288194656372
    },
    {
      "auxiliary_loss_clip": 0.01107812,
      "auxiliary_loss_mlp": 0.01032259,
      "balance_loss_clip": 1.03751671,
      "balance_loss_mlp": 1.01989651,
      "epoch": 0.6426574477679242,
      "flos": 16430388912000.0,
      "grad_norm": 1.9398742556351503,
      "language_loss": 0.73157191,
      "learning_rate": 1.1963463578769876e-06,
      "loss": 0.75297266,
      "num_input_tokens_seen": 230595030,
      "step": 10689,
      "time_per_iteration": 2.448148012161255
    },
    {
      "auxiliary_loss_clip": 0.01097194,
      "auxiliary_loss_mlp": 0.01031155,
      "balance_loss_clip": 1.03837478,
      "balance_loss_mlp": 1.01992571,
      "epoch": 0.6427175710205922,
      "flos": 21835914647040.0,
      "grad_norm": 2.0876584893358805,
      "language_loss": 0.7145409,
      "learning_rate": 1.195989736948226e-06,
      "loss": 0.73582435,
      "num_input_tokens_seen": 230615135,
      "step": 10690,
      "time_per_iteration": 2.510194778442383
    },
    {
      "auxiliary_loss_clip": 0.01079313,
      "auxiliary_loss_mlp": 0.01030675,
      "balance_loss_clip": 1.03566325,
      "balance_loss_mlp": 1.01806879,
      "epoch": 0.6427776942732601,
      "flos": 17786627660160.0,
      "grad_norm": 1.8016449168418145,
      "language_loss": 0.77499115,
      "learning_rate": 1.1956331465076143e-06,
      "loss": 0.79609108,
      "num_input_tokens_seen": 230631965,
      "step": 10691,
      "time_per_iteration": 3.928469181060791
    },
    {
      "auxiliary_loss_clip": 0.01089317,
      "auxiliary_loss_mlp": 0.01030347,
      "balance_loss_clip": 1.03904366,
      "balance_loss_mlp": 1.01830685,
      "epoch": 0.6428378175259282,
      "flos": 15085893911040.0,
      "grad_norm": 2.0544852232811754,
      "language_loss": 0.74434912,
      "learning_rate": 1.1952765865686738e-06,
      "loss": 0.76554573,
      "num_input_tokens_seen": 230649565,
      "step": 10692,
      "time_per_iteration": 2.5260448455810547
    },
    {
      "auxiliary_loss_clip": 0.01097025,
      "auxiliary_loss_mlp": 0.01033411,
      "balance_loss_clip": 1.03669238,
      "balance_loss_mlp": 1.02143645,
      "epoch": 0.6428979407785961,
      "flos": 23841776816640.0,
      "grad_norm": 2.0732558200951052,
      "language_loss": 0.61448371,
      "learning_rate": 1.1949200571449263e-06,
      "loss": 0.63578808,
      "num_input_tokens_seen": 230669265,
      "step": 10693,
      "time_per_iteration": 2.5173304080963135
    },
    {
      "auxiliary_loss_clip": 0.01073114,
      "auxiliary_loss_mlp": 0.01025598,
      "balance_loss_clip": 1.03783488,
      "balance_loss_mlp": 1.01325965,
      "epoch": 0.6429580640312641,
      "flos": 32926852892160.0,
      "grad_norm": 1.5807021092004159,
      "language_loss": 0.5953294,
      "learning_rate": 1.1945635582498903e-06,
      "loss": 0.61631656,
      "num_input_tokens_seen": 230690575,
      "step": 10694,
      "time_per_iteration": 2.63952374458313
    },
    {
      "auxiliary_loss_clip": 0.01089077,
      "auxiliary_loss_mlp": 0.0103271,
      "balance_loss_clip": 1.03687632,
      "balance_loss_mlp": 1.02082491,
      "epoch": 0.643018187283932,
      "flos": 21068359896960.0,
      "grad_norm": 1.460974579309108,
      "language_loss": 0.79723209,
      "learning_rate": 1.1942070898970853e-06,
      "loss": 0.81844997,
      "num_input_tokens_seen": 230709420,
      "step": 10695,
      "time_per_iteration": 2.5102667808532715
    },
    {
      "auxiliary_loss_clip": 0.01109631,
      "auxiliary_loss_mlp": 0.01039429,
      "balance_loss_clip": 1.03777957,
      "balance_loss_mlp": 1.026739,
      "epoch": 0.6430783105366,
      "flos": 26724649455360.0,
      "grad_norm": 1.7222336275599037,
      "language_loss": 0.73721987,
      "learning_rate": 1.1938506521000285e-06,
      "loss": 0.7587105,
      "num_input_tokens_seen": 230729350,
      "step": 10696,
      "time_per_iteration": 2.5229713916778564
    },
    {
      "auxiliary_loss_clip": 0.01072787,
      "auxiliary_loss_mlp": 0.01028325,
      "balance_loss_clip": 1.03618991,
      "balance_loss_mlp": 1.01615405,
      "epoch": 0.643138433789268,
      "flos": 23696841438720.0,
      "grad_norm": 1.6906056994983796,
      "language_loss": 0.75699425,
      "learning_rate": 1.1934942448722347e-06,
      "loss": 0.77800536,
      "num_input_tokens_seen": 230749220,
      "step": 10697,
      "time_per_iteration": 2.5625908374786377
    },
    {
      "auxiliary_loss_clip": 0.01082497,
      "auxiliary_loss_mlp": 0.01031461,
      "balance_loss_clip": 1.03581583,
      "balance_loss_mlp": 1.01986766,
      "epoch": 0.643198557041936,
      "flos": 34202184255360.0,
      "grad_norm": 1.8433623439356899,
      "language_loss": 0.65914893,
      "learning_rate": 1.1931378682272208e-06,
      "loss": 0.68028855,
      "num_input_tokens_seen": 230770245,
      "step": 10698,
      "time_per_iteration": 2.6539034843444824
    },
    {
      "auxiliary_loss_clip": 0.01043119,
      "auxiliary_loss_mlp": 0.0100344,
      "balance_loss_clip": 1.01932836,
      "balance_loss_mlp": 1.00206935,
      "epoch": 0.643258680294604,
      "flos": 67626473621760.0,
      "grad_norm": 1.2503049063321305,
      "language_loss": 0.63498163,
      "learning_rate": 1.1927815221784996e-06,
      "loss": 0.65544713,
      "num_input_tokens_seen": 230837030,
      "step": 10699,
      "time_per_iteration": 3.0581884384155273
    },
    {
      "auxiliary_loss_clip": 0.01096487,
      "auxiliary_loss_mlp": 0.01024817,
      "balance_loss_clip": 1.03864038,
      "balance_loss_mlp": 1.01382637,
      "epoch": 0.6433188035472719,
      "flos": 25185984508800.0,
      "grad_norm": 1.6248772242465586,
      "language_loss": 0.69215727,
      "learning_rate": 1.1924252067395838e-06,
      "loss": 0.71337032,
      "num_input_tokens_seen": 230856845,
      "step": 10700,
      "time_per_iteration": 5.349958419799805
    },
    {
      "auxiliary_loss_clip": 0.01107929,
      "auxiliary_loss_mlp": 0.01024682,
      "balance_loss_clip": 1.03784144,
      "balance_loss_mlp": 1.01314282,
      "epoch": 0.6433789267999399,
      "flos": 24973573432320.0,
      "grad_norm": 1.740074667730253,
      "language_loss": 0.73258615,
      "learning_rate": 1.1920689219239855e-06,
      "loss": 0.75391227,
      "num_input_tokens_seen": 230878785,
      "step": 10701,
      "time_per_iteration": 2.5069432258605957
    },
    {
      "auxiliary_loss_clip": 0.01098587,
      "auxiliary_loss_mlp": 0.01031206,
      "balance_loss_clip": 1.03620863,
      "balance_loss_mlp": 1.01763403,
      "epoch": 0.6434390500526078,
      "flos": 17566028282880.0,
      "grad_norm": 1.9822277230578806,
      "language_loss": 0.81884062,
      "learning_rate": 1.1917126677452144e-06,
      "loss": 0.84013855,
      "num_input_tokens_seen": 230895445,
      "step": 10702,
      "time_per_iteration": 2.485189437866211
    },
    {
      "auxiliary_loss_clip": 0.01077349,
      "auxiliary_loss_mlp": 0.01035912,
      "balance_loss_clip": 1.03441417,
      "balance_loss_mlp": 1.02397299,
      "epoch": 0.6434991733052758,
      "flos": 20843594542080.0,
      "grad_norm": 1.892178901806213,
      "language_loss": 0.74422443,
      "learning_rate": 1.1913564442167798e-06,
      "loss": 0.76535702,
      "num_input_tokens_seen": 230911375,
      "step": 10703,
      "time_per_iteration": 3.90875244140625
    },
    {
      "auxiliary_loss_clip": 0.00985179,
      "auxiliary_loss_mlp": 0.01003421,
      "balance_loss_clip": 1.01602709,
      "balance_loss_mlp": 1.00199056,
      "epoch": 0.6435592965579437,
      "flos": 66094596345600.0,
      "grad_norm": 0.655118093932809,
      "language_loss": 0.54591918,
      "learning_rate": 1.1910002513521898e-06,
      "loss": 0.5658052,
      "num_input_tokens_seen": 230975990,
      "step": 10704,
      "time_per_iteration": 3.2133750915527344
    },
    {
      "auxiliary_loss_clip": 0.01067165,
      "auxiliary_loss_mlp": 0.0102325,
      "balance_loss_clip": 1.03984773,
      "balance_loss_mlp": 1.01282549,
      "epoch": 0.6436194198106118,
      "flos": 23768842250880.0,
      "grad_norm": 1.5656041012221376,
      "language_loss": 0.76886475,
      "learning_rate": 1.1906440891649519e-06,
      "loss": 0.78976882,
      "num_input_tokens_seen": 230997110,
      "step": 10705,
      "time_per_iteration": 2.5902888774871826
    },
    {
      "auxiliary_loss_clip": 0.01072254,
      "auxiliary_loss_mlp": 0.01032448,
      "balance_loss_clip": 1.03522885,
      "balance_loss_mlp": 1.02078331,
      "epoch": 0.6436795430632797,
      "flos": 20230312705920.0,
      "grad_norm": 1.7628866599601776,
      "language_loss": 0.79084647,
      "learning_rate": 1.1902879576685708e-06,
      "loss": 0.81189346,
      "num_input_tokens_seen": 231015590,
      "step": 10706,
      "time_per_iteration": 2.53989315032959
    },
    {
      "auxiliary_loss_clip": 0.01067623,
      "auxiliary_loss_mlp": 0.01032826,
      "balance_loss_clip": 1.033602,
      "balance_loss_mlp": 1.02001119,
      "epoch": 0.6437396663159477,
      "flos": 20301846641280.0,
      "grad_norm": 2.09196144795845,
      "language_loss": 0.79975772,
      "learning_rate": 1.1899318568765518e-06,
      "loss": 0.82076222,
      "num_input_tokens_seen": 231033800,
      "step": 10707,
      "time_per_iteration": 2.5570809841156006
    },
    {
      "auxiliary_loss_clip": 0.01095537,
      "auxiliary_loss_mlp": 0.01031402,
      "balance_loss_clip": 1.03617036,
      "balance_loss_mlp": 1.01977921,
      "epoch": 0.6437997895686156,
      "flos": 23878585278720.0,
      "grad_norm": 2.430675151707551,
      "language_loss": 0.85490155,
      "learning_rate": 1.1895757868023978e-06,
      "loss": 0.87617087,
      "num_input_tokens_seen": 231053160,
      "step": 10708,
      "time_per_iteration": 2.507781505584717
    },
    {
      "auxiliary_loss_clip": 0.01063857,
      "auxiliary_loss_mlp": 0.01039504,
      "balance_loss_clip": 1.03857517,
      "balance_loss_mlp": 1.02596784,
      "epoch": 0.6438599128212836,
      "flos": 18989275852800.0,
      "grad_norm": 2.310358254164252,
      "language_loss": 0.65404481,
      "learning_rate": 1.1892197474596106e-06,
      "loss": 0.67507851,
      "num_input_tokens_seen": 231069470,
      "step": 10709,
      "time_per_iteration": 2.595135450363159
    },
    {
      "auxiliary_loss_clip": 0.01106285,
      "auxiliary_loss_mlp": 0.01028894,
      "balance_loss_clip": 1.03671169,
      "balance_loss_mlp": 1.01774192,
      "epoch": 0.6439200360739517,
      "flos": 24096347481600.0,
      "grad_norm": 1.8010250619603243,
      "language_loss": 0.80538428,
      "learning_rate": 1.1888637388616929e-06,
      "loss": 0.82673609,
      "num_input_tokens_seen": 231088205,
      "step": 10710,
      "time_per_iteration": 2.4793200492858887
    },
    {
      "auxiliary_loss_clip": 0.01095147,
      "auxiliary_loss_mlp": 0.01029469,
      "balance_loss_clip": 1.03561211,
      "balance_loss_mlp": 1.01789427,
      "epoch": 0.6439801593266196,
      "flos": 31902141697920.0,
      "grad_norm": 1.6801219724805279,
      "language_loss": 0.65769738,
      "learning_rate": 1.1885077610221425e-06,
      "loss": 0.67894357,
      "num_input_tokens_seen": 231107850,
      "step": 10711,
      "time_per_iteration": 2.5856051445007324
    },
    {
      "auxiliary_loss_clip": 0.01077833,
      "auxiliary_loss_mlp": 0.01032019,
      "balance_loss_clip": 1.04254329,
      "balance_loss_mlp": 1.02016997,
      "epoch": 0.6440402825792876,
      "flos": 27125879351040.0,
      "grad_norm": 1.5874691447264566,
      "language_loss": 0.78942645,
      "learning_rate": 1.1881518139544597e-06,
      "loss": 0.810525,
      "num_input_tokens_seen": 231127200,
      "step": 10712,
      "time_per_iteration": 2.592106819152832
    },
    {
      "auxiliary_loss_clip": 0.01095215,
      "auxiliary_loss_mlp": 0.01035634,
      "balance_loss_clip": 1.03543472,
      "balance_loss_mlp": 1.02328992,
      "epoch": 0.6441004058319555,
      "flos": 20667704618880.0,
      "grad_norm": 1.6502827891128635,
      "language_loss": 0.82609928,
      "learning_rate": 1.1877958976721417e-06,
      "loss": 0.84740776,
      "num_input_tokens_seen": 231146360,
      "step": 10713,
      "time_per_iteration": 2.492992639541626
    },
    {
      "auxiliary_loss_clip": 0.01105242,
      "auxiliary_loss_mlp": 0.01036179,
      "balance_loss_clip": 1.03849769,
      "balance_loss_mlp": 1.02529478,
      "epoch": 0.6441605290846235,
      "flos": 26026006947840.0,
      "grad_norm": 1.4415022433147315,
      "language_loss": 0.78526127,
      "learning_rate": 1.187440012188684e-06,
      "loss": 0.80667549,
      "num_input_tokens_seen": 231168350,
      "step": 10714,
      "time_per_iteration": 2.5010437965393066
    },
    {
      "auxiliary_loss_clip": 0.01073708,
      "auxiliary_loss_mlp": 0.01031551,
      "balance_loss_clip": 1.03567171,
      "balance_loss_mlp": 1.02064371,
      "epoch": 0.6442206523372914,
      "flos": 24899489631360.0,
      "grad_norm": 1.429510608203221,
      "language_loss": 0.81507838,
      "learning_rate": 1.187084157517583e-06,
      "loss": 0.83613098,
      "num_input_tokens_seen": 231188385,
      "step": 10715,
      "time_per_iteration": 2.597356081008911
    },
    {
      "auxiliary_loss_clip": 0.01077905,
      "auxiliary_loss_mlp": 0.01032225,
      "balance_loss_clip": 1.03378403,
      "balance_loss_mlp": 1.02011895,
      "epoch": 0.6442807755899594,
      "flos": 25156322853120.0,
      "grad_norm": 2.391603752972217,
      "language_loss": 0.81440055,
      "learning_rate": 1.186728333672332e-06,
      "loss": 0.83550191,
      "num_input_tokens_seen": 231209880,
      "step": 10716,
      "time_per_iteration": 2.567639112472534
    },
    {
      "auxiliary_loss_clip": 0.01076389,
      "auxiliary_loss_mlp": 0.01037475,
      "balance_loss_clip": 1.03721952,
      "balance_loss_mlp": 1.02370667,
      "epoch": 0.6443408988426274,
      "flos": 27344503480320.0,
      "grad_norm": 1.8761185512391365,
      "language_loss": 0.77988088,
      "learning_rate": 1.186372540666424e-06,
      "loss": 0.80101955,
      "num_input_tokens_seen": 231230765,
      "step": 10717,
      "time_per_iteration": 2.610337495803833
    },
    {
      "auxiliary_loss_clip": 0.01104986,
      "auxiliary_loss_mlp": 0.01030663,
      "balance_loss_clip": 1.0378381,
      "balance_loss_mlp": 1.01920652,
      "epoch": 0.6444010220952954,
      "flos": 27928339142400.0,
      "grad_norm": 1.6964793201411565,
      "language_loss": 0.68676591,
      "learning_rate": 1.1860167785133513e-06,
      "loss": 0.70812237,
      "num_input_tokens_seen": 231252350,
      "step": 10718,
      "time_per_iteration": 2.526170253753662
    },
    {
      "auxiliary_loss_clip": 0.01032978,
      "auxiliary_loss_mlp": 0.01008761,
      "balance_loss_clip": 1.01939046,
      "balance_loss_mlp": 1.00758076,
      "epoch": 0.6444611453479633,
      "flos": 71215024855680.0,
      "grad_norm": 0.772576237681771,
      "language_loss": 0.49658263,
      "learning_rate": 1.185661047226603e-06,
      "loss": 0.51699996,
      "num_input_tokens_seen": 231313865,
      "step": 10719,
      "time_per_iteration": 3.2545406818389893
    },
    {
      "auxiliary_loss_clip": 0.01111479,
      "auxiliary_loss_mlp": 0.01041004,
      "balance_loss_clip": 1.0397985,
      "balance_loss_mlp": 1.0286057,
      "epoch": 0.6445212686006313,
      "flos": 22705131864960.0,
      "grad_norm": 1.8224569597914426,
      "language_loss": 0.78261185,
      "learning_rate": 1.18530534681967e-06,
      "loss": 0.80413663,
      "num_input_tokens_seen": 231331710,
      "step": 10720,
      "time_per_iteration": 2.464001178741455
    },
    {
      "auxiliary_loss_clip": 0.010867,
      "auxiliary_loss_mlp": 0.01035724,
      "balance_loss_clip": 1.03712773,
      "balance_loss_mlp": 1.02273667,
      "epoch": 0.6445813918532992,
      "flos": 21178821196800.0,
      "grad_norm": 1.7049677385774804,
      "language_loss": 0.76939267,
      "learning_rate": 1.18494967730604e-06,
      "loss": 0.79061693,
      "num_input_tokens_seen": 231350705,
      "step": 10721,
      "time_per_iteration": 2.5521111488342285
    },
    {
      "auxiliary_loss_clip": 0.0107624,
      "auxiliary_loss_mlp": 0.01033391,
      "balance_loss_clip": 1.03623903,
      "balance_loss_mlp": 1.02096903,
      "epoch": 0.6446415151059672,
      "flos": 25191910252800.0,
      "grad_norm": 1.8867814526378333,
      "language_loss": 0.7289089,
      "learning_rate": 1.1845940386991995e-06,
      "loss": 0.75000525,
      "num_input_tokens_seen": 231369550,
      "step": 10722,
      "time_per_iteration": 2.5758020877838135
    },
    {
      "auxiliary_loss_clip": 0.01107628,
      "auxiliary_loss_mlp": 0.01032106,
      "balance_loss_clip": 1.03904021,
      "balance_loss_mlp": 1.0204947,
      "epoch": 0.6447016383586353,
      "flos": 25302227898240.0,
      "grad_norm": 1.7175229646101169,
      "language_loss": 0.77943432,
      "learning_rate": 1.184238431012635e-06,
      "loss": 0.80083168,
      "num_input_tokens_seen": 231389285,
      "step": 10723,
      "time_per_iteration": 2.508270502090454
    },
    {
      "auxiliary_loss_clip": 0.01099986,
      "auxiliary_loss_mlp": 0.01033358,
      "balance_loss_clip": 1.03770494,
      "balance_loss_mlp": 1.02068007,
      "epoch": 0.6447617616113032,
      "flos": 27703142824320.0,
      "grad_norm": 1.8833660437019557,
      "language_loss": 0.5864929,
      "learning_rate": 1.1838828542598312e-06,
      "loss": 0.60782635,
      "num_input_tokens_seen": 231408820,
      "step": 10724,
      "time_per_iteration": 2.541466474533081
    },
    {
      "auxiliary_loss_clip": 0.01094911,
      "auxiliary_loss_mlp": 0.01032358,
      "balance_loss_clip": 1.03866494,
      "balance_loss_mlp": 1.02152836,
      "epoch": 0.6448218848639712,
      "flos": 23039101543680.0,
      "grad_norm": 1.857538111681203,
      "language_loss": 0.83943677,
      "learning_rate": 1.183527308454271e-06,
      "loss": 0.86070943,
      "num_input_tokens_seen": 231428100,
      "step": 10725,
      "time_per_iteration": 2.5026473999023438
    },
    {
      "auxiliary_loss_clip": 0.01083616,
      "auxiliary_loss_mlp": 0.01037141,
      "balance_loss_clip": 1.03489971,
      "balance_loss_mlp": 1.02459383,
      "epoch": 0.6448820081166391,
      "flos": 24496104919680.0,
      "grad_norm": 1.8793116716435447,
      "language_loss": 0.82239032,
      "learning_rate": 1.1831717936094368e-06,
      "loss": 0.84359789,
      "num_input_tokens_seen": 231445810,
      "step": 10726,
      "time_per_iteration": 2.538616895675659
    },
    {
      "auxiliary_loss_clip": 0.01094501,
      "auxiliary_loss_mlp": 0.01036534,
      "balance_loss_clip": 1.03625321,
      "balance_loss_mlp": 1.02374887,
      "epoch": 0.6449421313693071,
      "flos": 22419283432320.0,
      "grad_norm": 1.7217019998434009,
      "language_loss": 0.81072003,
      "learning_rate": 1.1828163097388108e-06,
      "loss": 0.83203042,
      "num_input_tokens_seen": 231463570,
      "step": 10727,
      "time_per_iteration": 2.49383282661438
    },
    {
      "auxiliary_loss_clip": 0.01103252,
      "auxiliary_loss_mlp": 0.01030496,
      "balance_loss_clip": 1.03792632,
      "balance_loss_mlp": 1.0178833,
      "epoch": 0.645002254621975,
      "flos": 20225715765120.0,
      "grad_norm": 2.4965734116377942,
      "language_loss": 0.791372,
      "learning_rate": 1.1824608568558717e-06,
      "loss": 0.81270945,
      "num_input_tokens_seen": 231482155,
      "step": 10728,
      "time_per_iteration": 2.4856793880462646
    },
    {
      "auxiliary_loss_clip": 0.010232,
      "auxiliary_loss_mlp": 0.01036034,
      "balance_loss_clip": 1.03572488,
      "balance_loss_mlp": 1.02210474,
      "epoch": 0.645062377874643,
      "flos": 27855440490240.0,
      "grad_norm": 1.6412189808494695,
      "language_loss": 0.74160379,
      "learning_rate": 1.1821054349740988e-06,
      "loss": 0.76219618,
      "num_input_tokens_seen": 231502465,
      "step": 10729,
      "time_per_iteration": 2.84660267829895
    },
    {
      "auxiliary_loss_clip": 0.01075382,
      "auxiliary_loss_mlp": 0.01034042,
      "balance_loss_clip": 1.04068279,
      "balance_loss_mlp": 1.02145994,
      "epoch": 0.645122501127311,
      "flos": 25301509626240.0,
      "grad_norm": 1.8341320947960646,
      "language_loss": 0.66785014,
      "learning_rate": 1.1817500441069706e-06,
      "loss": 0.6889444,
      "num_input_tokens_seen": 231522740,
      "step": 10730,
      "time_per_iteration": 4.323291540145874
    },
    {
      "auxiliary_loss_clip": 0.01045341,
      "auxiliary_loss_mlp": 0.01031662,
      "balance_loss_clip": 1.03647614,
      "balance_loss_mlp": 1.01815557,
      "epoch": 0.645182624379979,
      "flos": 18807352444800.0,
      "grad_norm": 1.6242795209044594,
      "language_loss": 0.63822514,
      "learning_rate": 1.1813946842679614e-06,
      "loss": 0.65899521,
      "num_input_tokens_seen": 231542050,
      "step": 10731,
      "time_per_iteration": 2.6461358070373535
    },
    {
      "auxiliary_loss_clip": 0.01105792,
      "auxiliary_loss_mlp": 0.01033371,
      "balance_loss_clip": 1.03678143,
      "balance_loss_mlp": 1.02167082,
      "epoch": 0.6452427476326469,
      "flos": 18332182402560.0,
      "grad_norm": 1.6292537941411842,
      "language_loss": 0.67954171,
      "learning_rate": 1.1810393554705492e-06,
      "loss": 0.70093334,
      "num_input_tokens_seen": 231560380,
      "step": 10732,
      "time_per_iteration": 2.4734530448913574
    },
    {
      "auxiliary_loss_clip": 0.01096116,
      "auxiliary_loss_mlp": 0.01035244,
      "balance_loss_clip": 1.03679085,
      "balance_loss_mlp": 1.02326918,
      "epoch": 0.6453028708853149,
      "flos": 22784746360320.0,
      "grad_norm": 1.710046091057466,
      "language_loss": 0.75712347,
      "learning_rate": 1.1806840577282055e-06,
      "loss": 0.77843708,
      "num_input_tokens_seen": 231580810,
      "step": 10733,
      "time_per_iteration": 2.5369744300842285
    },
    {
      "auxiliary_loss_clip": 0.01098785,
      "auxiliary_loss_mlp": 0.01039892,
      "balance_loss_clip": 1.03820229,
      "balance_loss_mlp": 1.02666605,
      "epoch": 0.6453629941379828,
      "flos": 23945989150080.0,
      "grad_norm": 1.984953096963954,
      "language_loss": 0.6647867,
      "learning_rate": 1.1803287910544048e-06,
      "loss": 0.68617356,
      "num_input_tokens_seen": 231600585,
      "step": 10734,
      "time_per_iteration": 2.542736530303955
    },
    {
      "auxiliary_loss_clip": 0.01107706,
      "auxiliary_loss_mlp": 0.01037093,
      "balance_loss_clip": 1.04120278,
      "balance_loss_mlp": 1.02532673,
      "epoch": 0.6454231173906508,
      "flos": 17676381841920.0,
      "grad_norm": 2.344353347254017,
      "language_loss": 0.73335731,
      "learning_rate": 1.1799735554626191e-06,
      "loss": 0.75480533,
      "num_input_tokens_seen": 231618765,
      "step": 10735,
      "time_per_iteration": 2.4669041633605957
    },
    {
      "auxiliary_loss_clip": 0.01046055,
      "auxiliary_loss_mlp": 0.00787337,
      "balance_loss_clip": 1.03531909,
      "balance_loss_mlp": 1.01391506,
      "epoch": 0.6454832406433189,
      "flos": 23292774368640.0,
      "grad_norm": 1.769851821522435,
      "language_loss": 0.74800116,
      "learning_rate": 1.1796183509663176e-06,
      "loss": 0.76633501,
      "num_input_tokens_seen": 231638525,
      "step": 10736,
      "time_per_iteration": 2.6443064212799072
    },
    {
      "auxiliary_loss_clip": 0.01102691,
      "auxiliary_loss_mlp": 0.01030444,
      "balance_loss_clip": 1.04011691,
      "balance_loss_mlp": 1.01754546,
      "epoch": 0.6455433638959868,
      "flos": 20157198572160.0,
      "grad_norm": 2.3168225654948875,
      "language_loss": 0.70609915,
      "learning_rate": 1.1792631775789708e-06,
      "loss": 0.72743058,
      "num_input_tokens_seen": 231656785,
      "step": 10737,
      "time_per_iteration": 2.497847080230713
    },
    {
      "auxiliary_loss_clip": 0.01032626,
      "auxiliary_loss_mlp": 0.01003183,
      "balance_loss_clip": 1.01770639,
      "balance_loss_mlp": 1.00198483,
      "epoch": 0.6456034871486548,
      "flos": 66532922012160.0,
      "grad_norm": 0.7813204639379189,
      "language_loss": 0.58500129,
      "learning_rate": 1.1789080353140464e-06,
      "loss": 0.60535932,
      "num_input_tokens_seen": 231719075,
      "step": 10738,
      "time_per_iteration": 3.1762871742248535
    },
    {
      "auxiliary_loss_clip": 0.01073087,
      "auxiliary_loss_mlp": 0.01027247,
      "balance_loss_clip": 1.03997445,
      "balance_loss_mlp": 1.01525474,
      "epoch": 0.6456636104013227,
      "flos": 24206090509440.0,
      "grad_norm": 1.8449617508384977,
      "language_loss": 0.74379617,
      "learning_rate": 1.1785529241850118e-06,
      "loss": 0.76479948,
      "num_input_tokens_seen": 231737810,
      "step": 10739,
      "time_per_iteration": 4.017168045043945
    },
    {
      "auxiliary_loss_clip": 0.01089754,
      "auxiliary_loss_mlp": 0.00787377,
      "balance_loss_clip": 1.03843188,
      "balance_loss_mlp": 1.01397634,
      "epoch": 0.6457237336539907,
      "flos": 23624086440960.0,
      "grad_norm": 1.734705617852928,
      "language_loss": 0.71410245,
      "learning_rate": 1.1781978442053324e-06,
      "loss": 0.7328738,
      "num_input_tokens_seen": 231756140,
      "step": 10740,
      "time_per_iteration": 2.55741810798645
    },
    {
      "auxiliary_loss_clip": 0.01027809,
      "auxiliary_loss_mlp": 0.00999753,
      "balance_loss_clip": 1.02273357,
      "balance_loss_mlp": 0.99845999,
      "epoch": 0.6457838569066586,
      "flos": 65846023251840.0,
      "grad_norm": 0.6628763083831545,
      "language_loss": 0.55314875,
      "learning_rate": 1.1778427953884733e-06,
      "loss": 0.57342434,
      "num_input_tokens_seen": 231823665,
      "step": 10741,
      "time_per_iteration": 4.818443059921265
    },
    {
      "auxiliary_loss_clip": 0.01107349,
      "auxiliary_loss_mlp": 0.0103369,
      "balance_loss_clip": 1.03827596,
      "balance_loss_mlp": 1.02201939,
      "epoch": 0.6458439801593266,
      "flos": 22381972179840.0,
      "grad_norm": 1.8635666008698664,
      "language_loss": 0.80573636,
      "learning_rate": 1.1774877777478977e-06,
      "loss": 0.82714677,
      "num_input_tokens_seen": 231844500,
      "step": 10742,
      "time_per_iteration": 2.52644681930542
    },
    {
      "auxiliary_loss_clip": 0.01083242,
      "auxiliary_loss_mlp": 0.01028499,
      "balance_loss_clip": 1.03633404,
      "balance_loss_mlp": 1.01678717,
      "epoch": 0.6459041034119946,
      "flos": 24789243813120.0,
      "grad_norm": 2.4806653089138946,
      "language_loss": 0.81827974,
      "learning_rate": 1.1771327912970678e-06,
      "loss": 0.83939719,
      "num_input_tokens_seen": 231864510,
      "step": 10743,
      "time_per_iteration": 2.5649142265319824
    },
    {
      "auxiliary_loss_clip": 0.01084996,
      "auxiliary_loss_mlp": 0.01029572,
      "balance_loss_clip": 1.04103112,
      "balance_loss_mlp": 1.01785362,
      "epoch": 0.6459642266646626,
      "flos": 18325358818560.0,
      "grad_norm": 1.9457046909617843,
      "language_loss": 0.72075367,
      "learning_rate": 1.1767778360494453e-06,
      "loss": 0.74189937,
      "num_input_tokens_seen": 231881555,
      "step": 10744,
      "time_per_iteration": 2.5138468742370605
    },
    {
      "auxiliary_loss_clip": 0.01108293,
      "auxiliary_loss_mlp": 0.01030506,
      "balance_loss_clip": 1.03772449,
      "balance_loss_mlp": 1.01893651,
      "epoch": 0.6460243499173305,
      "flos": 43581368891520.0,
      "grad_norm": 1.7399079530100137,
      "language_loss": 0.66602194,
      "learning_rate": 1.1764229120184896e-06,
      "loss": 0.68740988,
      "num_input_tokens_seen": 231905945,
      "step": 10745,
      "time_per_iteration": 2.6695830821990967
    },
    {
      "auxiliary_loss_clip": 0.01095568,
      "auxiliary_loss_mlp": 0.01035373,
      "balance_loss_clip": 1.03650928,
      "balance_loss_mlp": 1.02243829,
      "epoch": 0.6460844731699985,
      "flos": 19244026085760.0,
      "grad_norm": 2.4615505988355024,
      "language_loss": 0.73407769,
      "learning_rate": 1.1760680192176597e-06,
      "loss": 0.75538701,
      "num_input_tokens_seen": 231922535,
      "step": 10746,
      "time_per_iteration": 2.487253427505493
    },
    {
      "auxiliary_loss_clip": 0.01099982,
      "auxiliary_loss_mlp": 0.01033294,
      "balance_loss_clip": 1.0389322,
      "balance_loss_mlp": 1.02168298,
      "epoch": 0.6461445964226664,
      "flos": 27453348668160.0,
      "grad_norm": 1.460959762464095,
      "language_loss": 0.66358662,
      "learning_rate": 1.175713157660413e-06,
      "loss": 0.68491936,
      "num_input_tokens_seen": 231944800,
      "step": 10747,
      "time_per_iteration": 2.543606996536255
    },
    {
      "auxiliary_loss_clip": 0.01086158,
      "auxiliary_loss_mlp": 0.01036792,
      "balance_loss_clip": 1.040797,
      "balance_loss_mlp": 1.02482319,
      "epoch": 0.6462047196753344,
      "flos": 20295489934080.0,
      "grad_norm": 1.6227589634632975,
      "language_loss": 0.67213959,
      "learning_rate": 1.1753583273602056e-06,
      "loss": 0.69336903,
      "num_input_tokens_seen": 231962970,
      "step": 10748,
      "time_per_iteration": 2.530475616455078
    },
    {
      "auxiliary_loss_clip": 0.01113162,
      "auxiliary_loss_mlp": 0.01040421,
      "balance_loss_clip": 1.03946543,
      "balance_loss_mlp": 1.02708769,
      "epoch": 0.6462648429280025,
      "flos": 22018340845440.0,
      "grad_norm": 1.7253829082692396,
      "language_loss": 0.75941634,
      "learning_rate": 1.1750035283304937e-06,
      "loss": 0.78095222,
      "num_input_tokens_seen": 231981195,
      "step": 10749,
      "time_per_iteration": 2.4684014320373535
    },
    {
      "auxiliary_loss_clip": 0.0107207,
      "auxiliary_loss_mlp": 0.01039995,
      "balance_loss_clip": 1.03483081,
      "balance_loss_mlp": 1.02628577,
      "epoch": 0.6463249661806704,
      "flos": 27781141207680.0,
      "grad_norm": 1.5287751419560436,
      "language_loss": 0.7697928,
      "learning_rate": 1.17464876058473e-06,
      "loss": 0.7909134,
      "num_input_tokens_seen": 232001735,
      "step": 10750,
      "time_per_iteration": 2.599959373474121
    },
    {
      "auxiliary_loss_clip": 0.01093723,
      "auxiliary_loss_mlp": 0.01033293,
      "balance_loss_clip": 1.03934991,
      "balance_loss_mlp": 1.01911867,
      "epoch": 0.6463850894333384,
      "flos": 22050588280320.0,
      "grad_norm": 1.982609519196789,
      "language_loss": 0.68403465,
      "learning_rate": 1.1742940241363683e-06,
      "loss": 0.70530486,
      "num_input_tokens_seen": 232019830,
      "step": 10751,
      "time_per_iteration": 2.5396275520324707
    },
    {
      "auxiliary_loss_clip": 0.01090121,
      "auxiliary_loss_mlp": 0.01033009,
      "balance_loss_clip": 1.03819633,
      "balance_loss_mlp": 1.02038455,
      "epoch": 0.6464452126860063,
      "flos": 21106245767040.0,
      "grad_norm": 1.9948023528023426,
      "language_loss": 0.71233034,
      "learning_rate": 1.1739393189988604e-06,
      "loss": 0.73356169,
      "num_input_tokens_seen": 232039625,
      "step": 10752,
      "time_per_iteration": 2.5323193073272705
    },
    {
      "auxiliary_loss_clip": 0.01079024,
      "auxiliary_loss_mlp": 0.01037798,
      "balance_loss_clip": 1.03662348,
      "balance_loss_mlp": 1.02335572,
      "epoch": 0.6465053359386743,
      "flos": 16028045694720.0,
      "grad_norm": 4.659649573876688,
      "language_loss": 0.78227007,
      "learning_rate": 1.1735846451856554e-06,
      "loss": 0.80343831,
      "num_input_tokens_seen": 232055855,
      "step": 10753,
      "time_per_iteration": 2.511215925216675
    },
    {
      "auxiliary_loss_clip": 0.01109301,
      "auxiliary_loss_mlp": 0.01043237,
      "balance_loss_clip": 1.03921962,
      "balance_loss_mlp": 1.03084528,
      "epoch": 0.6465654591913422,
      "flos": 23398674641280.0,
      "grad_norm": 1.7292863076245988,
      "language_loss": 0.8475771,
      "learning_rate": 1.1732300027102041e-06,
      "loss": 0.86910248,
      "num_input_tokens_seen": 232073475,
      "step": 10754,
      "time_per_iteration": 2.51213002204895
    },
    {
      "auxiliary_loss_clip": 0.01084695,
      "auxiliary_loss_mlp": 0.01038073,
      "balance_loss_clip": 1.03652668,
      "balance_loss_mlp": 1.02545476,
      "epoch": 0.6466255824440102,
      "flos": 15377273038080.0,
      "grad_norm": 2.2507102663776375,
      "language_loss": 0.59696364,
      "learning_rate": 1.1728753915859541e-06,
      "loss": 0.61819136,
      "num_input_tokens_seen": 232091090,
      "step": 10755,
      "time_per_iteration": 2.5140185356140137
    },
    {
      "auxiliary_loss_clip": 0.01073443,
      "auxiliary_loss_mlp": 0.01033761,
      "balance_loss_clip": 1.03756607,
      "balance_loss_mlp": 1.02066028,
      "epoch": 0.6466857056966782,
      "flos": 16252846963200.0,
      "grad_norm": 2.1763052449966884,
      "language_loss": 0.67887831,
      "learning_rate": 1.1725208118263518e-06,
      "loss": 0.69995034,
      "num_input_tokens_seen": 232107320,
      "step": 10756,
      "time_per_iteration": 2.519484043121338
    },
    {
      "auxiliary_loss_clip": 0.01071164,
      "auxiliary_loss_mlp": 0.0104248,
      "balance_loss_clip": 1.03862834,
      "balance_loss_mlp": 1.02786446,
      "epoch": 0.6467458289493462,
      "flos": 21178246579200.0,
      "grad_norm": 2.6495682462564654,
      "language_loss": 0.73768914,
      "learning_rate": 1.172166263444844e-06,
      "loss": 0.75882554,
      "num_input_tokens_seen": 232123930,
      "step": 10757,
      "time_per_iteration": 2.5767314434051514
    },
    {
      "auxiliary_loss_clip": 0.01058943,
      "auxiliary_loss_mlp": 0.01030682,
      "balance_loss_clip": 1.04149616,
      "balance_loss_mlp": 1.01830792,
      "epoch": 0.6468059522020141,
      "flos": 17968299672960.0,
      "grad_norm": 1.9890584897278025,
      "language_loss": 0.74637449,
      "learning_rate": 1.1718117464548734e-06,
      "loss": 0.7672708,
      "num_input_tokens_seen": 232142905,
      "step": 10758,
      "time_per_iteration": 2.591008186340332
    },
    {
      "auxiliary_loss_clip": 0.01074277,
      "auxiliary_loss_mlp": 0.01032116,
      "balance_loss_clip": 1.04017925,
      "balance_loss_mlp": 1.01912189,
      "epoch": 0.6468660754546821,
      "flos": 17890157635200.0,
      "grad_norm": 1.5665831968537518,
      "language_loss": 0.67982197,
      "learning_rate": 1.1714572608698845e-06,
      "loss": 0.70088589,
      "num_input_tokens_seen": 232162230,
      "step": 10759,
      "time_per_iteration": 2.5653076171875
    },
    {
      "auxiliary_loss_clip": 0.01076027,
      "auxiliary_loss_mlp": 0.01034024,
      "balance_loss_clip": 1.0356946,
      "balance_loss_mlp": 1.0200702,
      "epoch": 0.64692619870735,
      "flos": 22600991358720.0,
      "grad_norm": 2.6767163976055564,
      "language_loss": 0.75280559,
      "learning_rate": 1.1711028067033197e-06,
      "loss": 0.77390611,
      "num_input_tokens_seen": 232182700,
      "step": 10760,
      "time_per_iteration": 2.577657699584961
    },
    {
      "auxiliary_loss_clip": 0.01081889,
      "auxiliary_loss_mlp": 0.01034462,
      "balance_loss_clip": 1.03400064,
      "balance_loss_mlp": 1.02096128,
      "epoch": 0.646986321960018,
      "flos": 49600786993920.0,
      "grad_norm": 2.1457542250923383,
      "language_loss": 0.6537292,
      "learning_rate": 1.1707483839686194e-06,
      "loss": 0.67489266,
      "num_input_tokens_seen": 232208235,
      "step": 10761,
      "time_per_iteration": 2.793678045272827
    },
    {
      "auxiliary_loss_clip": 0.01064524,
      "auxiliary_loss_mlp": 0.01032682,
      "balance_loss_clip": 1.03952003,
      "balance_loss_mlp": 1.01956296,
      "epoch": 0.6470464452126861,
      "flos": 21908454163200.0,
      "grad_norm": 2.691367679127516,
      "language_loss": 0.69507748,
      "learning_rate": 1.1703939926792235e-06,
      "loss": 0.71604955,
      "num_input_tokens_seen": 232228720,
      "step": 10762,
      "time_per_iteration": 2.6064181327819824
    },
    {
      "auxiliary_loss_clip": 0.01113191,
      "auxiliary_loss_mlp": 0.01032917,
      "balance_loss_clip": 1.03908443,
      "balance_loss_mlp": 1.02010202,
      "epoch": 0.647106568465354,
      "flos": 18106124158080.0,
      "grad_norm": 2.1550699648300107,
      "language_loss": 0.82997417,
      "learning_rate": 1.1700396328485705e-06,
      "loss": 0.85143524,
      "num_input_tokens_seen": 232244655,
      "step": 10763,
      "time_per_iteration": 2.4468226432800293
    },
    {
      "auxiliary_loss_clip": 0.01041708,
      "auxiliary_loss_mlp": 0.01003134,
      "balance_loss_clip": 1.01763368,
      "balance_loss_mlp": 1.00188279,
      "epoch": 0.647166691718022,
      "flos": 69480038125440.0,
      "grad_norm": 0.7178166113170772,
      "language_loss": 0.5779472,
      "learning_rate": 1.1696853044900978e-06,
      "loss": 0.59839565,
      "num_input_tokens_seen": 232308685,
      "step": 10764,
      "time_per_iteration": 3.246119499206543
    },
    {
      "auxiliary_loss_clip": 0.01074275,
      "auxiliary_loss_mlp": 0.01034693,
      "balance_loss_clip": 1.03737903,
      "balance_loss_mlp": 1.02218151,
      "epoch": 0.6472268149706899,
      "flos": 34095170661120.0,
      "grad_norm": 1.7882582856362779,
      "language_loss": 0.60231578,
      "learning_rate": 1.1693310076172413e-06,
      "loss": 0.62340546,
      "num_input_tokens_seen": 232327520,
      "step": 10765,
      "time_per_iteration": 2.664862632751465
    },
    {
      "auxiliary_loss_clip": 0.01108362,
      "auxiliary_loss_mlp": 0.01028249,
      "balance_loss_clip": 1.03808594,
      "balance_loss_mlp": 1.01583946,
      "epoch": 0.6472869382233579,
      "flos": 28111232217600.0,
      "grad_norm": 2.601342048524849,
      "language_loss": 0.62936431,
      "learning_rate": 1.168976742243437e-06,
      "loss": 0.65073037,
      "num_input_tokens_seen": 232349025,
      "step": 10766,
      "time_per_iteration": 2.5155725479125977
    },
    {
      "auxiliary_loss_clip": 0.01083722,
      "auxiliary_loss_mlp": 0.01034892,
      "balance_loss_clip": 1.03673732,
      "balance_loss_mlp": 1.02087307,
      "epoch": 0.6473470614760258,
      "flos": 22492146170880.0,
      "grad_norm": 1.7035968320643127,
      "language_loss": 0.75789464,
      "learning_rate": 1.1686225083821174e-06,
      "loss": 0.77908075,
      "num_input_tokens_seen": 232367835,
      "step": 10767,
      "time_per_iteration": 2.544235944747925
    },
    {
      "auxiliary_loss_clip": 0.01100124,
      "auxiliary_loss_mlp": 0.01032247,
      "balance_loss_clip": 1.03880954,
      "balance_loss_mlp": 1.01970649,
      "epoch": 0.6474071847286939,
      "flos": 14538938538240.0,
      "grad_norm": 1.9178373725294062,
      "language_loss": 0.77569056,
      "learning_rate": 1.1682683060467153e-06,
      "loss": 0.79701436,
      "num_input_tokens_seen": 232385840,
      "step": 10768,
      "time_per_iteration": 2.468111753463745
    },
    {
      "auxiliary_loss_clip": 0.01060561,
      "auxiliary_loss_mlp": 0.01030847,
      "balance_loss_clip": 1.03844023,
      "balance_loss_mlp": 1.01776338,
      "epoch": 0.6474673079813618,
      "flos": 24098214988800.0,
      "grad_norm": 1.6734653304041587,
      "language_loss": 0.71524322,
      "learning_rate": 1.167914135250663e-06,
      "loss": 0.7361573,
      "num_input_tokens_seen": 232406205,
      "step": 10769,
      "time_per_iteration": 4.009365558624268
    },
    {
      "auxiliary_loss_clip": 0.01107484,
      "auxiliary_loss_mlp": 0.01033151,
      "balance_loss_clip": 1.03935421,
      "balance_loss_mlp": 1.02070558,
      "epoch": 0.6475274312340298,
      "flos": 14976186796800.0,
      "grad_norm": 1.9675535972396452,
      "language_loss": 0.7214489,
      "learning_rate": 1.1675599960073895e-06,
      "loss": 0.74285531,
      "num_input_tokens_seen": 232424995,
      "step": 10770,
      "time_per_iteration": 2.452566623687744
    },
    {
      "auxiliary_loss_clip": 0.01070731,
      "auxiliary_loss_mlp": 0.0103096,
      "balance_loss_clip": 1.03674281,
      "balance_loss_mlp": 1.01728714,
      "epoch": 0.6475875544866977,
      "flos": 25045322849280.0,
      "grad_norm": 1.7926108212100622,
      "language_loss": 0.7352742,
      "learning_rate": 1.167205888330325e-06,
      "loss": 0.75629115,
      "num_input_tokens_seen": 232445870,
      "step": 10771,
      "time_per_iteration": 2.7192792892456055
    },
    {
      "auxiliary_loss_clip": 0.01072603,
      "auxiliary_loss_mlp": 0.01034167,
      "balance_loss_clip": 1.03630257,
      "balance_loss_mlp": 1.02071381,
      "epoch": 0.6476476777393657,
      "flos": 16472153450880.0,
      "grad_norm": 2.149520472917965,
      "language_loss": 0.741907,
      "learning_rate": 1.1668518122328958e-06,
      "loss": 0.76297474,
      "num_input_tokens_seen": 232464285,
      "step": 10772,
      "time_per_iteration": 2.5347797870635986
    },
    {
      "auxiliary_loss_clip": 0.01084278,
      "auxiliary_loss_mlp": 0.01030312,
      "balance_loss_clip": 1.03940511,
      "balance_loss_mlp": 1.01914763,
      "epoch": 0.6477078009920336,
      "flos": 25812267068160.0,
      "grad_norm": 1.4925691973003317,
      "language_loss": 0.82980812,
      "learning_rate": 1.1664977677285305e-06,
      "loss": 0.85095406,
      "num_input_tokens_seen": 232485815,
      "step": 10773,
      "time_per_iteration": 2.6020233631134033
    },
    {
      "auxiliary_loss_clip": 0.01095559,
      "auxiliary_loss_mlp": 0.00785626,
      "balance_loss_clip": 1.03757477,
      "balance_loss_mlp": 1.01295435,
      "epoch": 0.6477679242447016,
      "flos": 17676130446720.0,
      "grad_norm": 1.492190015844518,
      "language_loss": 0.7888521,
      "learning_rate": 1.1661437548306524e-06,
      "loss": 0.80766398,
      "num_input_tokens_seen": 232504875,
      "step": 10774,
      "time_per_iteration": 2.4853177070617676
    },
    {
      "auxiliary_loss_clip": 0.01099789,
      "auxiliary_loss_mlp": 0.01037461,
      "balance_loss_clip": 1.03829122,
      "balance_loss_mlp": 1.02493179,
      "epoch": 0.6478280474973696,
      "flos": 21032305620480.0,
      "grad_norm": 2.3486306343146244,
      "language_loss": 0.69258976,
      "learning_rate": 1.1657897735526867e-06,
      "loss": 0.71396232,
      "num_input_tokens_seen": 232521945,
      "step": 10775,
      "time_per_iteration": 2.508967638015747
    },
    {
      "auxiliary_loss_clip": 0.01076953,
      "auxiliary_loss_mlp": 0.01037181,
      "balance_loss_clip": 1.04035926,
      "balance_loss_mlp": 1.02459884,
      "epoch": 0.6478881707500376,
      "flos": 21616931381760.0,
      "grad_norm": 1.9964279025709317,
      "language_loss": 0.6570127,
      "learning_rate": 1.1654358239080574e-06,
      "loss": 0.67815405,
      "num_input_tokens_seen": 232541500,
      "step": 10776,
      "time_per_iteration": 2.5700719356536865
    },
    {
      "auxiliary_loss_clip": 0.01080352,
      "auxiliary_loss_mlp": 0.01036913,
      "balance_loss_clip": 1.03493881,
      "balance_loss_mlp": 1.02377045,
      "epoch": 0.6479482940027056,
      "flos": 18442571875200.0,
      "grad_norm": 2.378883810921812,
      "language_loss": 0.78561085,
      "learning_rate": 1.1650819059101839e-06,
      "loss": 0.80678356,
      "num_input_tokens_seen": 232559720,
      "step": 10777,
      "time_per_iteration": 3.9503233432769775
    },
    {
      "auxiliary_loss_clip": 0.01097566,
      "auxiliary_loss_mlp": 0.01031277,
      "balance_loss_clip": 1.04234338,
      "balance_loss_mlp": 1.01873028,
      "epoch": 0.6480084172553735,
      "flos": 22164066322560.0,
      "grad_norm": 1.8539494197163027,
      "language_loss": 0.73406875,
      "learning_rate": 1.1647280195724896e-06,
      "loss": 0.75535715,
      "num_input_tokens_seen": 232579370,
      "step": 10778,
      "time_per_iteration": 4.317025899887085
    },
    {
      "auxiliary_loss_clip": 0.01091898,
      "auxiliary_loss_mlp": 0.01030295,
      "balance_loss_clip": 1.03476238,
      "balance_loss_mlp": 1.01854706,
      "epoch": 0.6480685405080415,
      "flos": 24316228586880.0,
      "grad_norm": 1.4645261642371676,
      "language_loss": 0.77937901,
      "learning_rate": 1.1643741649083923e-06,
      "loss": 0.80060095,
      "num_input_tokens_seen": 232600495,
      "step": 10779,
      "time_per_iteration": 2.5587470531463623
    },
    {
      "auxiliary_loss_clip": 0.01030093,
      "auxiliary_loss_mlp": 0.01000108,
      "balance_loss_clip": 1.02036703,
      "balance_loss_mlp": 0.99889803,
      "epoch": 0.6481286637607094,
      "flos": 59891207760000.0,
      "grad_norm": 0.7192576263371642,
      "language_loss": 0.59413499,
      "learning_rate": 1.1640203419313095e-06,
      "loss": 0.61443698,
      "num_input_tokens_seen": 232663165,
      "step": 10780,
      "time_per_iteration": 4.460371971130371
    },
    {
      "auxiliary_loss_clip": 0.01020616,
      "auxiliary_loss_mlp": 0.010298,
      "balance_loss_clip": 1.03719664,
      "balance_loss_mlp": 1.01755738,
      "epoch": 0.6481887870133775,
      "flos": 25484187219840.0,
      "grad_norm": 1.9390187131851686,
      "language_loss": 0.79268229,
      "learning_rate": 1.1636665506546599e-06,
      "loss": 0.81318653,
      "num_input_tokens_seen": 232683385,
      "step": 10781,
      "time_per_iteration": 2.9183521270751953
    },
    {
      "auxiliary_loss_clip": 0.01113693,
      "auxiliary_loss_mlp": 0.01036435,
      "balance_loss_clip": 1.04026341,
      "balance_loss_mlp": 1.02173603,
      "epoch": 0.6482489102660454,
      "flos": 19930206574080.0,
      "grad_norm": 2.141203443765911,
      "language_loss": 0.7890631,
      "learning_rate": 1.1633127910918578e-06,
      "loss": 0.8105644,
      "num_input_tokens_seen": 232699095,
      "step": 10782,
      "time_per_iteration": 2.594644784927368
    },
    {
      "auxiliary_loss_clip": 0.01101141,
      "auxiliary_loss_mlp": 0.00784812,
      "balance_loss_clip": 1.03910315,
      "balance_loss_mlp": 1.00894487,
      "epoch": 0.6483090335187134,
      "flos": 26979471515520.0,
      "grad_norm": 2.1197321037722388,
      "language_loss": 0.6439662,
      "learning_rate": 1.1629590632563187e-06,
      "loss": 0.6628257,
      "num_input_tokens_seen": 232717920,
      "step": 10783,
      "time_per_iteration": 2.5316011905670166
    },
    {
      "auxiliary_loss_clip": 0.01112955,
      "auxiliary_loss_mlp": 0.01034064,
      "balance_loss_clip": 1.03939986,
      "balance_loss_mlp": 1.02026546,
      "epoch": 0.6483691567713813,
      "flos": 25077965333760.0,
      "grad_norm": 1.7445508451507303,
      "language_loss": 0.8865329,
      "learning_rate": 1.1626053671614561e-06,
      "loss": 0.90800309,
      "num_input_tokens_seen": 232737605,
      "step": 10784,
      "time_per_iteration": 2.5041744709014893
    },
    {
      "auxiliary_loss_clip": 0.01085048,
      "auxiliary_loss_mlp": 0.0102645,
      "balance_loss_clip": 1.04044604,
      "balance_loss_mlp": 1.01337898,
      "epoch": 0.6484292800240493,
      "flos": 16105972250880.0,
      "grad_norm": 2.2517179017266353,
      "language_loss": 0.730865,
      "learning_rate": 1.1622517028206815e-06,
      "loss": 0.75197995,
      "num_input_tokens_seen": 232755110,
      "step": 10785,
      "time_per_iteration": 2.4927921295166016
    },
    {
      "auxiliary_loss_clip": 0.01071694,
      "auxiliary_loss_mlp": 0.01030423,
      "balance_loss_clip": 1.03565121,
      "balance_loss_mlp": 1.01844835,
      "epoch": 0.6484894032767172,
      "flos": 28840398307200.0,
      "grad_norm": 1.4504981767277063,
      "language_loss": 0.69174457,
      "learning_rate": 1.1618980702474071e-06,
      "loss": 0.71276569,
      "num_input_tokens_seen": 232779040,
      "step": 10786,
      "time_per_iteration": 2.637023687362671
    },
    {
      "auxiliary_loss_clip": 0.0107124,
      "auxiliary_loss_mlp": 0.01028789,
      "balance_loss_clip": 1.03702283,
      "balance_loss_mlp": 1.01659942,
      "epoch": 0.6485495265293852,
      "flos": 30227052896640.0,
      "grad_norm": 1.955527208248782,
      "language_loss": 0.71080935,
      "learning_rate": 1.161544469455041e-06,
      "loss": 0.73180962,
      "num_input_tokens_seen": 232800515,
      "step": 10787,
      "time_per_iteration": 2.614147186279297
    },
    {
      "auxiliary_loss_clip": 0.01113598,
      "auxiliary_loss_mlp": 0.0103362,
      "balance_loss_clip": 1.03914118,
      "balance_loss_mlp": 1.02028656,
      "epoch": 0.6486096497820532,
      "flos": 20082181017600.0,
      "grad_norm": 2.055583677718124,
      "language_loss": 0.84486711,
      "learning_rate": 1.1611909004569934e-06,
      "loss": 0.86633927,
      "num_input_tokens_seen": 232818450,
      "step": 10788,
      "time_per_iteration": 2.4580438137054443
    },
    {
      "auxiliary_loss_clip": 0.01072576,
      "auxiliary_loss_mlp": 0.01030618,
      "balance_loss_clip": 1.03824687,
      "balance_loss_mlp": 1.01737952,
      "epoch": 0.6486697730347212,
      "flos": 17129067333120.0,
      "grad_norm": 1.8839890860574426,
      "language_loss": 0.77342075,
      "learning_rate": 1.1608373632666708e-06,
      "loss": 0.79445267,
      "num_input_tokens_seen": 232834785,
      "step": 10789,
      "time_per_iteration": 2.5165505409240723
    },
    {
      "auxiliary_loss_clip": 0.01093536,
      "auxiliary_loss_mlp": 0.01029313,
      "balance_loss_clip": 1.03598702,
      "balance_loss_mlp": 1.01721287,
      "epoch": 0.6487298962873892,
      "flos": 38911940570880.0,
      "grad_norm": 1.6244439796516261,
      "language_loss": 0.75932133,
      "learning_rate": 1.160483857897479e-06,
      "loss": 0.78054976,
      "num_input_tokens_seen": 232856050,
      "step": 10790,
      "time_per_iteration": 2.6491663455963135
    },
    {
      "auxiliary_loss_clip": 0.01108302,
      "auxiliary_loss_mlp": 0.01032588,
      "balance_loss_clip": 1.03946781,
      "balance_loss_mlp": 1.02140069,
      "epoch": 0.6487900195400571,
      "flos": 11947840076160.0,
      "grad_norm": 2.0093216122545896,
      "language_loss": 0.60459471,
      "learning_rate": 1.160130384362823e-06,
      "loss": 0.62600356,
      "num_input_tokens_seen": 232873945,
      "step": 10791,
      "time_per_iteration": 2.446176052093506
    },
    {
      "auxiliary_loss_clip": 0.01072077,
      "auxiliary_loss_mlp": 0.01029806,
      "balance_loss_clip": 1.03958428,
      "balance_loss_mlp": 1.0174973,
      "epoch": 0.6488501427927251,
      "flos": 22344445445760.0,
      "grad_norm": 1.9670572843143324,
      "language_loss": 0.85917628,
      "learning_rate": 1.1597769426761082e-06,
      "loss": 0.88019502,
      "num_input_tokens_seen": 232892160,
      "step": 10792,
      "time_per_iteration": 2.5741450786590576
    },
    {
      "auxiliary_loss_clip": 0.01087515,
      "auxiliary_loss_mlp": 0.01035014,
      "balance_loss_clip": 1.03758466,
      "balance_loss_mlp": 1.02231252,
      "epoch": 0.648910266045393,
      "flos": 22236282616320.0,
      "grad_norm": 2.0601792562184547,
      "language_loss": 0.77837694,
      "learning_rate": 1.159423532850735e-06,
      "loss": 0.79960227,
      "num_input_tokens_seen": 232911725,
      "step": 10793,
      "time_per_iteration": 2.5487234592437744
    },
    {
      "auxiliary_loss_clip": 0.0107859,
      "auxiliary_loss_mlp": 0.01027874,
      "balance_loss_clip": 1.0392127,
      "balance_loss_mlp": 1.01485062,
      "epoch": 0.6489703892980611,
      "flos": 25301258231040.0,
      "grad_norm": 1.8377240530924162,
      "language_loss": 0.74552679,
      "learning_rate": 1.1590701549001055e-06,
      "loss": 0.76659143,
      "num_input_tokens_seen": 232929085,
      "step": 10794,
      "time_per_iteration": 2.588306427001953
    },
    {
      "auxiliary_loss_clip": 0.01096046,
      "auxiliary_loss_mlp": 0.00784646,
      "balance_loss_clip": 1.03560793,
      "balance_loss_mlp": 1.00858915,
      "epoch": 0.649030512550729,
      "flos": 24571912573440.0,
      "grad_norm": 2.3336258793750413,
      "language_loss": 0.70063579,
      "learning_rate": 1.158716808837621e-06,
      "loss": 0.71944273,
      "num_input_tokens_seen": 232949455,
      "step": 10795,
      "time_per_iteration": 2.5507302284240723
    },
    {
      "auxiliary_loss_clip": 0.01089269,
      "auxiliary_loss_mlp": 0.01034702,
      "balance_loss_clip": 1.03774738,
      "balance_loss_mlp": 1.02191687,
      "epoch": 0.649090635803397,
      "flos": 26244702904320.0,
      "grad_norm": 1.7307885102877514,
      "language_loss": 0.53935683,
      "learning_rate": 1.158363494676679e-06,
      "loss": 0.56059659,
      "num_input_tokens_seen": 232969445,
      "step": 10796,
      "time_per_iteration": 2.571962833404541
    },
    {
      "auxiliary_loss_clip": 0.0109816,
      "auxiliary_loss_mlp": 0.01027203,
      "balance_loss_clip": 1.03845549,
      "balance_loss_mlp": 1.01566398,
      "epoch": 0.6491507590560649,
      "flos": 24937375501440.0,
      "grad_norm": 1.5254690241461684,
      "language_loss": 0.77848792,
      "learning_rate": 1.1580102124306775e-06,
      "loss": 0.79974163,
      "num_input_tokens_seen": 232988900,
      "step": 10797,
      "time_per_iteration": 2.5275557041168213
    },
    {
      "auxiliary_loss_clip": 0.0106141,
      "auxiliary_loss_mlp": 0.0102776,
      "balance_loss_clip": 1.04046857,
      "balance_loss_mlp": 1.01644707,
      "epoch": 0.6492108823087329,
      "flos": 19499781899520.0,
      "grad_norm": 2.01200120791438,
      "language_loss": 0.70773047,
      "learning_rate": 1.1576569621130134e-06,
      "loss": 0.72862214,
      "num_input_tokens_seen": 233005060,
      "step": 10798,
      "time_per_iteration": 2.613248586654663
    },
    {
      "auxiliary_loss_clip": 0.01065558,
      "auxiliary_loss_mlp": 0.01027845,
      "balance_loss_clip": 1.03709698,
      "balance_loss_mlp": 1.01646614,
      "epoch": 0.6492710055614008,
      "flos": 19719303868800.0,
      "grad_norm": 1.7369758207834491,
      "language_loss": 0.76312768,
      "learning_rate": 1.1573037437370811e-06,
      "loss": 0.78406179,
      "num_input_tokens_seen": 233023375,
      "step": 10799,
      "time_per_iteration": 2.5846493244171143
    },
    {
      "auxiliary_loss_clip": 0.01096481,
      "auxiliary_loss_mlp": 0.01036077,
      "balance_loss_clip": 1.03801656,
      "balance_loss_mlp": 1.02311933,
      "epoch": 0.6493311288140688,
      "flos": 24317018686080.0,
      "grad_norm": 1.7850094599071913,
      "language_loss": 0.71932316,
      "learning_rate": 1.1569505573162755e-06,
      "loss": 0.74064875,
      "num_input_tokens_seen": 233043130,
      "step": 10800,
      "time_per_iteration": 2.525892734527588
    },
    {
      "auxiliary_loss_clip": 0.01036204,
      "auxiliary_loss_mlp": 0.01012844,
      "balance_loss_clip": 1.02189076,
      "balance_loss_mlp": 1.01123524,
      "epoch": 0.6493912520667368,
      "flos": 70934635290240.0,
      "grad_norm": 1.3964363959007293,
      "language_loss": 0.60226667,
      "learning_rate": 1.1565974028639897e-06,
      "loss": 0.62275708,
      "num_input_tokens_seen": 233110560,
      "step": 10801,
      "time_per_iteration": 3.2199690341949463
    },
    {
      "auxiliary_loss_clip": 0.01100035,
      "auxiliary_loss_mlp": 0.01039297,
      "balance_loss_clip": 1.03923988,
      "balance_loss_mlp": 1.02536178,
      "epoch": 0.6494513753194048,
      "flos": 25337779384320.0,
      "grad_norm": 2.606048263019423,
      "language_loss": 0.78505588,
      "learning_rate": 1.156244280393614e-06,
      "loss": 0.80644923,
      "num_input_tokens_seen": 233130080,
      "step": 10802,
      "time_per_iteration": 2.5339558124542236
    },
    {
      "auxiliary_loss_clip": 0.01108572,
      "auxiliary_loss_mlp": 0.01039378,
      "balance_loss_clip": 1.03718543,
      "balance_loss_mlp": 1.02625942,
      "epoch": 0.6495114985720728,
      "flos": 24681978823680.0,
      "grad_norm": 1.628258856748789,
      "language_loss": 0.7490561,
      "learning_rate": 1.155891189918541e-06,
      "loss": 0.77053565,
      "num_input_tokens_seen": 233150235,
      "step": 10803,
      "time_per_iteration": 2.503297805786133
    },
    {
      "auxiliary_loss_clip": 0.01047514,
      "auxiliary_loss_mlp": 0.01033403,
      "balance_loss_clip": 1.03505397,
      "balance_loss_mlp": 1.02104068,
      "epoch": 0.6495716218247407,
      "flos": 23651162317440.0,
      "grad_norm": 3.233980426963653,
      "language_loss": 0.70581353,
      "learning_rate": 1.1555381314521578e-06,
      "loss": 0.72662276,
      "num_input_tokens_seen": 233166710,
      "step": 10804,
      "time_per_iteration": 2.6159756183624268
    },
    {
      "auxiliary_loss_clip": 0.01095566,
      "auxiliary_loss_mlp": 0.01029845,
      "balance_loss_clip": 1.04000044,
      "balance_loss_mlp": 1.01717353,
      "epoch": 0.6496317450774087,
      "flos": 22346169298560.0,
      "grad_norm": 1.7186501701784938,
      "language_loss": 0.7280727,
      "learning_rate": 1.1551851050078537e-06,
      "loss": 0.74932683,
      "num_input_tokens_seen": 233185445,
      "step": 10805,
      "time_per_iteration": 2.508727788925171
    },
    {
      "auxiliary_loss_clip": 0.01082426,
      "auxiliary_loss_mlp": 0.01029346,
      "balance_loss_clip": 1.03741968,
      "balance_loss_mlp": 1.01754451,
      "epoch": 0.6496918683300766,
      "flos": 30518647505280.0,
      "grad_norm": 2.472781373991973,
      "language_loss": 0.66088378,
      "learning_rate": 1.1548321105990155e-06,
      "loss": 0.68200147,
      "num_input_tokens_seen": 233205805,
      "step": 10806,
      "time_per_iteration": 2.593019485473633
    },
    {
      "auxiliary_loss_clip": 0.01087898,
      "auxiliary_loss_mlp": 0.00784409,
      "balance_loss_clip": 1.03662145,
      "balance_loss_mlp": 1.00834572,
      "epoch": 0.6497519915827447,
      "flos": 12458992567680.0,
      "grad_norm": 2.444582300884639,
      "language_loss": 0.79103303,
      "learning_rate": 1.1544791482390275e-06,
      "loss": 0.80975616,
      "num_input_tokens_seen": 233224215,
      "step": 10807,
      "time_per_iteration": 2.508929491043091
    },
    {
      "auxiliary_loss_clip": 0.01024422,
      "auxiliary_loss_mlp": 0.01007793,
      "balance_loss_clip": 1.01980305,
      "balance_loss_mlp": 1.00626683,
      "epoch": 0.6498121148354126,
      "flos": 69093748287360.0,
      "grad_norm": 0.7795136053045956,
      "language_loss": 0.58864272,
      "learning_rate": 1.1541262179412745e-06,
      "loss": 0.6089648,
      "num_input_tokens_seen": 233294440,
      "step": 10808,
      "time_per_iteration": 4.63586688041687
    },
    {
      "auxiliary_loss_clip": 0.01088008,
      "auxiliary_loss_mlp": 0.01028692,
      "balance_loss_clip": 1.03988504,
      "balance_loss_mlp": 1.01629984,
      "epoch": 0.6498722380880806,
      "flos": 36897135914880.0,
      "grad_norm": 1.8207848275677214,
      "language_loss": 0.62903976,
      "learning_rate": 1.1537733197191415e-06,
      "loss": 0.6502068,
      "num_input_tokens_seen": 233316125,
      "step": 10809,
      "time_per_iteration": 2.6601736545562744
    },
    {
      "auxiliary_loss_clip": 0.01096209,
      "auxiliary_loss_mlp": 0.00783379,
      "balance_loss_clip": 1.03831279,
      "balance_loss_mlp": 1.00902462,
      "epoch": 0.6499323613407485,
      "flos": 29017760688000.0,
      "grad_norm": 1.7898135189730002,
      "language_loss": 0.814785,
      "learning_rate": 1.153420453586008e-06,
      "loss": 0.83358085,
      "num_input_tokens_seen": 233336140,
      "step": 10810,
      "time_per_iteration": 2.588871955871582
    },
    {
      "auxiliary_loss_clip": 0.01074225,
      "auxiliary_loss_mlp": 0.0103531,
      "balance_loss_clip": 1.0373497,
      "balance_loss_mlp": 1.02410424,
      "epoch": 0.6499924845934165,
      "flos": 20119240874880.0,
      "grad_norm": 1.5827835015202556,
      "language_loss": 0.71862185,
      "learning_rate": 1.1530676195552561e-06,
      "loss": 0.73971725,
      "num_input_tokens_seen": 233356095,
      "step": 10811,
      "time_per_iteration": 2.6089158058166504
    },
    {
      "auxiliary_loss_clip": 0.0105107,
      "auxiliary_loss_mlp": 0.01029996,
      "balance_loss_clip": 1.04375672,
      "balance_loss_mlp": 1.01810455,
      "epoch": 0.6500526078460844,
      "flos": 24421338760320.0,
      "grad_norm": 1.7525204743255034,
      "language_loss": 0.77620107,
      "learning_rate": 1.1527148176402649e-06,
      "loss": 0.79701173,
      "num_input_tokens_seen": 233376830,
      "step": 10812,
      "time_per_iteration": 2.6667354106903076
    },
    {
      "auxiliary_loss_clip": 0.01094441,
      "auxiliary_loss_mlp": 0.01033217,
      "balance_loss_clip": 1.03827488,
      "balance_loss_mlp": 1.02031267,
      "epoch": 0.6501127310987524,
      "flos": 23331019374720.0,
      "grad_norm": 1.890557051759369,
      "language_loss": 0.85433686,
      "learning_rate": 1.152362047854413e-06,
      "loss": 0.87561345,
      "num_input_tokens_seen": 233395275,
      "step": 10813,
      "time_per_iteration": 2.5099005699157715
    },
    {
      "auxiliary_loss_clip": 0.01067381,
      "auxiliary_loss_mlp": 0.01035483,
      "balance_loss_clip": 1.0348376,
      "balance_loss_mlp": 1.02188146,
      "epoch": 0.6501728543514204,
      "flos": 18697824898560.0,
      "grad_norm": 2.2667465136846943,
      "language_loss": 0.79894125,
      "learning_rate": 1.1520093102110764e-06,
      "loss": 0.81996989,
      "num_input_tokens_seen": 233413345,
      "step": 10814,
      "time_per_iteration": 2.5397024154663086
    },
    {
      "auxiliary_loss_clip": 0.01062621,
      "auxiliary_loss_mlp": 0.00786453,
      "balance_loss_clip": 1.03896356,
      "balance_loss_mlp": 1.00645459,
      "epoch": 0.6502329776040884,
      "flos": 44199858199680.0,
      "grad_norm": 1.5161634010550664,
      "language_loss": 0.65540689,
      "learning_rate": 1.1516566047236328e-06,
      "loss": 0.67389762,
      "num_input_tokens_seen": 233436105,
      "step": 10815,
      "time_per_iteration": 2.784630298614502
    },
    {
      "auxiliary_loss_clip": 0.01113056,
      "auxiliary_loss_mlp": 0.0103452,
      "balance_loss_clip": 1.03862357,
      "balance_loss_mlp": 1.01941073,
      "epoch": 0.6502931008567564,
      "flos": 14574741419520.0,
      "grad_norm": 1.8653314857749332,
      "language_loss": 0.75280809,
      "learning_rate": 1.1513039314054546e-06,
      "loss": 0.77428377,
      "num_input_tokens_seen": 233452320,
      "step": 10816,
      "time_per_iteration": 3.8703465461730957
    },
    {
      "auxiliary_loss_clip": 0.01082646,
      "auxiliary_loss_mlp": 0.01032519,
      "balance_loss_clip": 1.03866863,
      "balance_loss_mlp": 1.02021015,
      "epoch": 0.6503532241094243,
      "flos": 21395003201280.0,
      "grad_norm": 1.6818816060432134,
      "language_loss": 0.72664881,
      "learning_rate": 1.1509512902699174e-06,
      "loss": 0.74780047,
      "num_input_tokens_seen": 233469920,
      "step": 10817,
      "time_per_iteration": 4.233865737915039
    },
    {
      "auxiliary_loss_clip": 0.01067492,
      "auxiliary_loss_mlp": 0.0104173,
      "balance_loss_clip": 1.03493023,
      "balance_loss_mlp": 1.02741921,
      "epoch": 0.6504133473620923,
      "flos": 74740840986240.0,
      "grad_norm": 1.465798716362421,
      "language_loss": 0.72209191,
      "learning_rate": 1.1505986813303916e-06,
      "loss": 0.74318409,
      "num_input_tokens_seen": 233499780,
      "step": 10818,
      "time_per_iteration": 4.36546516418457
    },
    {
      "auxiliary_loss_clip": 0.01080152,
      "auxiliary_loss_mlp": 0.01029584,
      "balance_loss_clip": 1.03738284,
      "balance_loss_mlp": 1.01678741,
      "epoch": 0.6504734706147602,
      "flos": 19713270384000.0,
      "grad_norm": 1.8389009891727541,
      "language_loss": 0.65284097,
      "learning_rate": 1.150246104600249e-06,
      "loss": 0.67393839,
      "num_input_tokens_seen": 233518235,
      "step": 10819,
      "time_per_iteration": 2.5477025508880615
    },
    {
      "auxiliary_loss_clip": 0.01075335,
      "auxiliary_loss_mlp": 0.01035021,
      "balance_loss_clip": 1.03602982,
      "balance_loss_mlp": 1.0218606,
      "epoch": 0.6505335938674283,
      "flos": 25556870390400.0,
      "grad_norm": 1.8708765659572038,
      "language_loss": 0.83945853,
      "learning_rate": 1.14989356009286e-06,
      "loss": 0.86056209,
      "num_input_tokens_seen": 233535215,
      "step": 10820,
      "time_per_iteration": 2.6009325981140137
    },
    {
      "auxiliary_loss_clip": 0.01102907,
      "auxiliary_loss_mlp": 0.01031039,
      "balance_loss_clip": 1.03822231,
      "balance_loss_mlp": 1.01763439,
      "epoch": 0.6505937171200962,
      "flos": 17821424960640.0,
      "grad_norm": 2.0361158991282338,
      "language_loss": 0.78221142,
      "learning_rate": 1.1495410478215914e-06,
      "loss": 0.80355084,
      "num_input_tokens_seen": 233552775,
      "step": 10821,
      "time_per_iteration": 2.469614267349243
    },
    {
      "auxiliary_loss_clip": 0.01076003,
      "auxiliary_loss_mlp": 0.01027571,
      "balance_loss_clip": 1.03945732,
      "balance_loss_mlp": 1.01641262,
      "epoch": 0.6506538403727642,
      "flos": 20668135582080.0,
      "grad_norm": 1.528245507586087,
      "language_loss": 0.80212843,
      "learning_rate": 1.1491885677998126e-06,
      "loss": 0.82316417,
      "num_input_tokens_seen": 233572080,
      "step": 10822,
      "time_per_iteration": 2.570295810699463
    },
    {
      "auxiliary_loss_clip": 0.01076058,
      "auxiliary_loss_mlp": 0.01026325,
      "balance_loss_clip": 1.037027,
      "balance_loss_mlp": 1.01476789,
      "epoch": 0.6507139636254321,
      "flos": 11721422695680.0,
      "grad_norm": 1.8140927505067028,
      "language_loss": 0.87250751,
      "learning_rate": 1.1488361200408883e-06,
      "loss": 0.89353132,
      "num_input_tokens_seen": 233589155,
      "step": 10823,
      "time_per_iteration": 2.5395734310150146
    },
    {
      "auxiliary_loss_clip": 0.01108799,
      "auxiliary_loss_mlp": 0.01032307,
      "balance_loss_clip": 1.03718066,
      "balance_loss_mlp": 1.01967657,
      "epoch": 0.6507740868781001,
      "flos": 26761745226240.0,
      "grad_norm": 1.7479068227122119,
      "language_loss": 0.66900325,
      "learning_rate": 1.148483704558183e-06,
      "loss": 0.69041425,
      "num_input_tokens_seen": 233608180,
      "step": 10824,
      "time_per_iteration": 2.533613443374634
    },
    {
      "auxiliary_loss_clip": 0.01089043,
      "auxiliary_loss_mlp": 0.01031949,
      "balance_loss_clip": 1.03653038,
      "balance_loss_mlp": 1.01918149,
      "epoch": 0.650834210130768,
      "flos": 16471722487680.0,
      "grad_norm": 2.797305843123813,
      "language_loss": 0.87735647,
      "learning_rate": 1.1481313213650607e-06,
      "loss": 0.89856637,
      "num_input_tokens_seen": 233625750,
      "step": 10825,
      "time_per_iteration": 2.494980812072754
    },
    {
      "auxiliary_loss_clip": 0.01090448,
      "auxiliary_loss_mlp": 0.01031004,
      "balance_loss_clip": 1.03720713,
      "balance_loss_mlp": 1.01662171,
      "epoch": 0.650894333383436,
      "flos": 17128672283520.0,
      "grad_norm": 2.244732216280115,
      "language_loss": 0.73206115,
      "learning_rate": 1.147778970474885e-06,
      "loss": 0.75327575,
      "num_input_tokens_seen": 233644235,
      "step": 10826,
      "time_per_iteration": 2.502307891845703
    },
    {
      "auxiliary_loss_clip": 0.01101314,
      "auxiliary_loss_mlp": 0.01030526,
      "balance_loss_clip": 1.03999078,
      "balance_loss_mlp": 1.01863539,
      "epoch": 0.650954456636104,
      "flos": 18734238311040.0,
      "grad_norm": 2.082191236876549,
      "language_loss": 0.68971097,
      "learning_rate": 1.1474266519010157e-06,
      "loss": 0.71102935,
      "num_input_tokens_seen": 233662845,
      "step": 10827,
      "time_per_iteration": 2.461458444595337
    },
    {
      "auxiliary_loss_clip": 0.01087093,
      "auxiliary_loss_mlp": 0.01030947,
      "balance_loss_clip": 1.03717375,
      "balance_loss_mlp": 1.01939607,
      "epoch": 0.651014579888772,
      "flos": 24528244613760.0,
      "grad_norm": 1.956977286298672,
      "language_loss": 0.76755702,
      "learning_rate": 1.1470743656568136e-06,
      "loss": 0.78873742,
      "num_input_tokens_seen": 233681990,
      "step": 10828,
      "time_per_iteration": 2.538895845413208
    },
    {
      "auxiliary_loss_clip": 0.01098203,
      "auxiliary_loss_mlp": 0.01027382,
      "balance_loss_clip": 1.03946972,
      "balance_loss_mlp": 1.01598549,
      "epoch": 0.65107470314144,
      "flos": 24061083304320.0,
      "grad_norm": 1.845372111498304,
      "language_loss": 0.8926639,
      "learning_rate": 1.1467221117556362e-06,
      "loss": 0.91391969,
      "num_input_tokens_seen": 233698930,
      "step": 10829,
      "time_per_iteration": 2.49698805809021
    },
    {
      "auxiliary_loss_clip": 0.01040847,
      "auxiliary_loss_mlp": 0.01003437,
      "balance_loss_clip": 1.01726055,
      "balance_loss_mlp": 1.00222158,
      "epoch": 0.6511348263941079,
      "flos": 72480734352000.0,
      "grad_norm": 0.6468861240069984,
      "language_loss": 0.55441332,
      "learning_rate": 1.1463698902108428e-06,
      "loss": 0.57485616,
      "num_input_tokens_seen": 233769825,
      "step": 10830,
      "time_per_iteration": 3.224823474884033
    },
    {
      "auxiliary_loss_clip": 0.01076734,
      "auxiliary_loss_mlp": 0.01030084,
      "balance_loss_clip": 1.03698134,
      "balance_loss_mlp": 1.01686358,
      "epoch": 0.6511949496467759,
      "flos": 23367684182400.0,
      "grad_norm": 1.8468766852587188,
      "language_loss": 0.74901825,
      "learning_rate": 1.1460177010357878e-06,
      "loss": 0.77008641,
      "num_input_tokens_seen": 233787095,
      "step": 10831,
      "time_per_iteration": 2.559882402420044
    },
    {
      "auxiliary_loss_clip": 0.01024253,
      "auxiliary_loss_mlp": 0.01000358,
      "balance_loss_clip": 1.02023482,
      "balance_loss_mlp": 0.99905294,
      "epoch": 0.6512550728994438,
      "flos": 67333191073920.0,
      "grad_norm": 0.6433499206561896,
      "language_loss": 0.51024532,
      "learning_rate": 1.145665544243828e-06,
      "loss": 0.53049141,
      "num_input_tokens_seen": 233853050,
      "step": 10832,
      "time_per_iteration": 3.2183475494384766
    },
    {
      "auxiliary_loss_clip": 0.01091328,
      "auxiliary_loss_mlp": 0.01033865,
      "balance_loss_clip": 1.03718877,
      "balance_loss_mlp": 1.02052534,
      "epoch": 0.6513151961521119,
      "flos": 21141689512320.0,
      "grad_norm": 2.2675436416416983,
      "language_loss": 0.82866722,
      "learning_rate": 1.145313419848316e-06,
      "loss": 0.84991914,
      "num_input_tokens_seen": 233871385,
      "step": 10833,
      "time_per_iteration": 2.539659261703491
    },
    {
      "auxiliary_loss_clip": 0.01089093,
      "auxiliary_loss_mlp": 0.01036349,
      "balance_loss_clip": 1.03957152,
      "balance_loss_mlp": 1.02384984,
      "epoch": 0.6513753194047798,
      "flos": 15158828476800.0,
      "grad_norm": 2.0150143147862507,
      "language_loss": 0.83618259,
      "learning_rate": 1.1449613278626049e-06,
      "loss": 0.85743701,
      "num_input_tokens_seen": 233888175,
      "step": 10834,
      "time_per_iteration": 2.500420570373535
    },
    {
      "auxiliary_loss_clip": 0.01098123,
      "auxiliary_loss_mlp": 0.01035858,
      "balance_loss_clip": 1.03775632,
      "balance_loss_mlp": 1.02295923,
      "epoch": 0.6514354426574478,
      "flos": 30226621933440.0,
      "grad_norm": 1.506117679789368,
      "language_loss": 0.77330196,
      "learning_rate": 1.1446092683000455e-06,
      "loss": 0.79464179,
      "num_input_tokens_seen": 233911470,
      "step": 10835,
      "time_per_iteration": 2.563380479812622
    },
    {
      "auxiliary_loss_clip": 0.01085391,
      "auxiliary_loss_mlp": 0.01035904,
      "balance_loss_clip": 1.040411,
      "balance_loss_mlp": 1.02320218,
      "epoch": 0.6514955659101157,
      "flos": 24205587719040.0,
      "grad_norm": 1.4675255036248633,
      "language_loss": 0.77480686,
      "learning_rate": 1.1442572411739882e-06,
      "loss": 0.79601979,
      "num_input_tokens_seen": 233932135,
      "step": 10836,
      "time_per_iteration": 2.577540636062622
    },
    {
      "auxiliary_loss_clip": 0.0107192,
      "auxiliary_loss_mlp": 0.01031715,
      "balance_loss_clip": 1.03894758,
      "balance_loss_mlp": 1.01941276,
      "epoch": 0.6515556891627837,
      "flos": 12377761960320.0,
      "grad_norm": 2.3385840011820282,
      "language_loss": 0.82533896,
      "learning_rate": 1.143905246497783e-06,
      "loss": 0.84637529,
      "num_input_tokens_seen": 233947880,
      "step": 10837,
      "time_per_iteration": 2.5280227661132812
    },
    {
      "auxiliary_loss_clip": 0.01070959,
      "auxiliary_loss_mlp": 0.01034056,
      "balance_loss_clip": 1.03985631,
      "balance_loss_mlp": 1.0203588,
      "epoch": 0.6516158124154516,
      "flos": 49601217957120.0,
      "grad_norm": 1.7852471160976675,
      "language_loss": 0.58364999,
      "learning_rate": 1.1435532842847758e-06,
      "loss": 0.60470009,
      "num_input_tokens_seen": 233971475,
      "step": 10838,
      "time_per_iteration": 2.794718027114868
    },
    {
      "auxiliary_loss_clip": 0.01040663,
      "auxiliary_loss_mlp": 0.01006199,
      "balance_loss_clip": 1.0169487,
      "balance_loss_mlp": 1.00512064,
      "epoch": 0.6516759356681197,
      "flos": 59702748076800.0,
      "grad_norm": 0.7274251215956774,
      "language_loss": 0.60958636,
      "learning_rate": 1.1432013545483147e-06,
      "loss": 0.63005495,
      "num_input_tokens_seen": 234030690,
      "step": 10839,
      "time_per_iteration": 3.1185834407806396
    },
    {
      "auxiliary_loss_clip": 0.01085601,
      "auxiliary_loss_mlp": 0.01027683,
      "balance_loss_clip": 1.03666115,
      "balance_loss_mlp": 1.01624513,
      "epoch": 0.6517360589207876,
      "flos": 37450807130880.0,
      "grad_norm": 1.6739563623864795,
      "language_loss": 0.67641318,
      "learning_rate": 1.1428494573017439e-06,
      "loss": 0.69754606,
      "num_input_tokens_seen": 234052470,
      "step": 10840,
      "time_per_iteration": 2.64815092086792
    },
    {
      "auxiliary_loss_clip": 0.01064177,
      "auxiliary_loss_mlp": 0.01030037,
      "balance_loss_clip": 1.03569424,
      "balance_loss_mlp": 1.01843202,
      "epoch": 0.6517961821734556,
      "flos": 25374911068800.0,
      "grad_norm": 1.9474680950646095,
      "language_loss": 0.73262733,
      "learning_rate": 1.1424975925584071e-06,
      "loss": 0.75356948,
      "num_input_tokens_seen": 234071495,
      "step": 10841,
      "time_per_iteration": 2.6112797260284424
    },
    {
      "auxiliary_loss_clip": 0.01110645,
      "auxiliary_loss_mlp": 0.01033798,
      "balance_loss_clip": 1.0379777,
      "balance_loss_mlp": 1.02126884,
      "epoch": 0.6518563054261236,
      "flos": 28766996864640.0,
      "grad_norm": 1.4773492185626065,
      "language_loss": 0.62787426,
      "learning_rate": 1.142145760331648e-06,
      "loss": 0.6493187,
      "num_input_tokens_seen": 234092325,
      "step": 10842,
      "time_per_iteration": 2.5213606357574463
    },
    {
      "auxiliary_loss_clip": 0.01035363,
      "auxiliary_loss_mlp": 0.01002822,
      "balance_loss_clip": 1.02118802,
      "balance_loss_mlp": 1.00182664,
      "epoch": 0.6519164286787915,
      "flos": 68924750797440.0,
      "grad_norm": 0.8463672212232929,
      "language_loss": 0.56147408,
      "learning_rate": 1.141793960634807e-06,
      "loss": 0.58185595,
      "num_input_tokens_seen": 234148005,
      "step": 10843,
      "time_per_iteration": 3.0202598571777344
    },
    {
      "auxiliary_loss_clip": 0.01101656,
      "auxiliary_loss_mlp": 0.01038961,
      "balance_loss_clip": 1.03770614,
      "balance_loss_mlp": 1.02552009,
      "epoch": 0.6519765519314595,
      "flos": 20441933683200.0,
      "grad_norm": 1.6773474343015606,
      "language_loss": 0.828035,
      "learning_rate": 1.1414421934812253e-06,
      "loss": 0.84944117,
      "num_input_tokens_seen": 234164280,
      "step": 10844,
      "time_per_iteration": 2.4833710193634033
    },
    {
      "auxiliary_loss_clip": 0.01098385,
      "auxiliary_loss_mlp": 0.01029329,
      "balance_loss_clip": 1.03728831,
      "balance_loss_mlp": 1.01649618,
      "epoch": 0.6520366751841274,
      "flos": 28402970480640.0,
      "grad_norm": 2.077900733313284,
      "language_loss": 0.59636086,
      "learning_rate": 1.1410904588842421e-06,
      "loss": 0.61763799,
      "num_input_tokens_seen": 234185090,
      "step": 10845,
      "time_per_iteration": 2.548968553543091
    },
    {
      "auxiliary_loss_clip": 0.01099984,
      "auxiliary_loss_mlp": 0.01029579,
      "balance_loss_clip": 1.03880823,
      "balance_loss_mlp": 1.01677001,
      "epoch": 0.6520967984367955,
      "flos": 22273414300800.0,
      "grad_norm": 2.091054401889517,
      "language_loss": 0.7937665,
      "learning_rate": 1.140738756857194e-06,
      "loss": 0.81506211,
      "num_input_tokens_seen": 234204050,
      "step": 10846,
      "time_per_iteration": 3.8867971897125244
    },
    {
      "auxiliary_loss_clip": 0.01034825,
      "auxiliary_loss_mlp": 0.01001706,
      "balance_loss_clip": 1.02056575,
      "balance_loss_mlp": 1.00061476,
      "epoch": 0.6521569216894634,
      "flos": 68917140092160.0,
      "grad_norm": 0.7449381766435149,
      "language_loss": 0.6020146,
      "learning_rate": 1.1403870874134192e-06,
      "loss": 0.6223799,
      "num_input_tokens_seen": 234269790,
      "step": 10847,
      "time_per_iteration": 3.2183175086975098
    },
    {
      "auxiliary_loss_clip": 0.01115772,
      "auxiliary_loss_mlp": 0.01041456,
      "balance_loss_clip": 1.04113114,
      "balance_loss_mlp": 1.02837324,
      "epoch": 0.6522170449421314,
      "flos": 29130520458240.0,
      "grad_norm": 1.8120169668174,
      "language_loss": 0.80988997,
      "learning_rate": 1.1400354505662514e-06,
      "loss": 0.83146226,
      "num_input_tokens_seen": 234290135,
      "step": 10848,
      "time_per_iteration": 2.544066905975342
    },
    {
      "auxiliary_loss_clip": 0.01088277,
      "auxiliary_loss_mlp": 0.01035117,
      "balance_loss_clip": 1.03760076,
      "balance_loss_mlp": 1.02301168,
      "epoch": 0.6522771681947993,
      "flos": 26651930371200.0,
      "grad_norm": 3.6354759927776548,
      "language_loss": 0.7523973,
      "learning_rate": 1.1396838463290263e-06,
      "loss": 0.77363127,
      "num_input_tokens_seen": 234309535,
      "step": 10849,
      "time_per_iteration": 2.5597898960113525
    },
    {
      "auxiliary_loss_clip": 0.01059093,
      "auxiliary_loss_mlp": 0.01035362,
      "balance_loss_clip": 1.0357666,
      "balance_loss_mlp": 1.02317894,
      "epoch": 0.6523372914474673,
      "flos": 25739763465600.0,
      "grad_norm": 1.416184795344904,
      "language_loss": 0.68250722,
      "learning_rate": 1.1393322747150752e-06,
      "loss": 0.70345175,
      "num_input_tokens_seen": 234328755,
      "step": 10850,
      "time_per_iteration": 2.6185190677642822
    },
    {
      "auxiliary_loss_clip": 0.01084319,
      "auxiliary_loss_mlp": 0.00784098,
      "balance_loss_clip": 1.03772867,
      "balance_loss_mlp": 1.00883222,
      "epoch": 0.6523974147001352,
      "flos": 24827345164800.0,
      "grad_norm": 1.7948936087414975,
      "language_loss": 0.66825646,
      "learning_rate": 1.1389807357377313e-06,
      "loss": 0.68694067,
      "num_input_tokens_seen": 234348655,
      "step": 10851,
      "time_per_iteration": 2.5662763118743896
    },
    {
      "auxiliary_loss_clip": 0.01089198,
      "auxiliary_loss_mlp": 0.01030607,
      "balance_loss_clip": 1.04136252,
      "balance_loss_mlp": 1.01853704,
      "epoch": 0.6524575379528033,
      "flos": 26317637470080.0,
      "grad_norm": 2.6535981320643844,
      "language_loss": 0.73673689,
      "learning_rate": 1.1386292294103235e-06,
      "loss": 0.75793493,
      "num_input_tokens_seen": 234367445,
      "step": 10852,
      "time_per_iteration": 2.552741289138794
    },
    {
      "auxiliary_loss_clip": 0.01086573,
      "auxiliary_loss_mlp": 0.01031536,
      "balance_loss_clip": 1.03940094,
      "balance_loss_mlp": 1.01766562,
      "epoch": 0.6525176612054712,
      "flos": 19494143464320.0,
      "grad_norm": 1.7751625330413754,
      "language_loss": 0.66367018,
      "learning_rate": 1.1382777557461812e-06,
      "loss": 0.68485129,
      "num_input_tokens_seen": 234384825,
      "step": 10853,
      "time_per_iteration": 2.514789581298828
    },
    {
      "auxiliary_loss_clip": 0.01011277,
      "auxiliary_loss_mlp": 0.0100139,
      "balance_loss_clip": 1.02000296,
      "balance_loss_mlp": 1.00000668,
      "epoch": 0.6525777844581392,
      "flos": 71706894721920.0,
      "grad_norm": 0.7223150624548311,
      "language_loss": 0.63072127,
      "learning_rate": 1.137926314758634e-06,
      "loss": 0.65084791,
      "num_input_tokens_seen": 234450630,
      "step": 10854,
      "time_per_iteration": 4.630734205245972
    },
    {
      "auxiliary_loss_clip": 0.01095443,
      "auxiliary_loss_mlp": 0.0104241,
      "balance_loss_clip": 1.03637886,
      "balance_loss_mlp": 1.02713907,
      "epoch": 0.6526379077108072,
      "flos": 26653115520000.0,
      "grad_norm": 1.6978022973130333,
      "language_loss": 0.77782345,
      "learning_rate": 1.1375749064610072e-06,
      "loss": 0.79920197,
      "num_input_tokens_seen": 234473505,
      "step": 10855,
      "time_per_iteration": 2.583150863647461
    },
    {
      "auxiliary_loss_clip": 0.01069276,
      "auxiliary_loss_mlp": 0.01029336,
      "balance_loss_clip": 1.03323829,
      "balance_loss_mlp": 1.01720095,
      "epoch": 0.6526980309634751,
      "flos": 22820369673600.0,
      "grad_norm": 2.257752124681326,
      "language_loss": 0.79010236,
      "learning_rate": 1.1372235308666256e-06,
      "loss": 0.81108856,
      "num_input_tokens_seen": 234492485,
      "step": 10856,
      "time_per_iteration": 3.9577157497406006
    },
    {
      "auxiliary_loss_clip": 0.01110234,
      "auxiliary_loss_mlp": 0.01030873,
      "balance_loss_clip": 1.0382998,
      "balance_loss_mlp": 1.01691365,
      "epoch": 0.6527581542161431,
      "flos": 28365048696960.0,
      "grad_norm": 1.504921792299299,
      "language_loss": 0.7373727,
      "learning_rate": 1.136872187988815e-06,
      "loss": 0.75878382,
      "num_input_tokens_seen": 234512645,
      "step": 10857,
      "time_per_iteration": 3.8830676078796387
    },
    {
      "auxiliary_loss_clip": 0.01088845,
      "auxiliary_loss_mlp": 0.01032874,
      "balance_loss_clip": 1.03803396,
      "balance_loss_mlp": 1.02070236,
      "epoch": 0.652818277468811,
      "flos": 18369206346240.0,
      "grad_norm": 1.9997293574171637,
      "language_loss": 0.62626219,
      "learning_rate": 1.1365208778408965e-06,
      "loss": 0.6474793,
      "num_input_tokens_seen": 234529310,
      "step": 10858,
      "time_per_iteration": 2.543156147003174
    },
    {
      "auxiliary_loss_clip": 0.01105718,
      "auxiliary_loss_mlp": 0.01032892,
      "balance_loss_clip": 1.03653944,
      "balance_loss_mlp": 1.02067947,
      "epoch": 0.6528784007214791,
      "flos": 18036170421120.0,
      "grad_norm": 2.118610447873736,
      "language_loss": 0.78523827,
      "learning_rate": 1.1361696004361939e-06,
      "loss": 0.80662441,
      "num_input_tokens_seen": 234546685,
      "step": 10859,
      "time_per_iteration": 2.439804792404175
    },
    {
      "auxiliary_loss_clip": 0.01100464,
      "auxiliary_loss_mlp": 0.01031562,
      "balance_loss_clip": 1.03716373,
      "balance_loss_mlp": 1.01834214,
      "epoch": 0.652938523974147,
      "flos": 22382008093440.0,
      "grad_norm": 1.496511111290527,
      "language_loss": 0.67768341,
      "learning_rate": 1.1358183557880256e-06,
      "loss": 0.6990037,
      "num_input_tokens_seen": 234566255,
      "step": 10860,
      "time_per_iteration": 2.5201399326324463
    },
    {
      "auxiliary_loss_clip": 0.01101014,
      "auxiliary_loss_mlp": 0.01030996,
      "balance_loss_clip": 1.04219747,
      "balance_loss_mlp": 1.01795506,
      "epoch": 0.652998647226815,
      "flos": 16764035368320.0,
      "grad_norm": 2.0296975956061822,
      "language_loss": 0.66773653,
      "learning_rate": 1.135467143909712e-06,
      "loss": 0.68905663,
      "num_input_tokens_seen": 234585405,
      "step": 10861,
      "time_per_iteration": 2.470863103866577
    },
    {
      "auxiliary_loss_clip": 0.01089393,
      "auxiliary_loss_mlp": 0.01032265,
      "balance_loss_clip": 1.03825879,
      "balance_loss_mlp": 1.0185796,
      "epoch": 0.6530587704794829,
      "flos": 35772522019200.0,
      "grad_norm": 1.7064096510872873,
      "language_loss": 0.65419567,
      "learning_rate": 1.135115964814572e-06,
      "loss": 0.67541224,
      "num_input_tokens_seen": 234608095,
      "step": 10862,
      "time_per_iteration": 2.674995183944702
    },
    {
      "auxiliary_loss_clip": 0.01087565,
      "auxiliary_loss_mlp": 0.01032565,
      "balance_loss_clip": 1.03752661,
      "balance_loss_mlp": 1.0202384,
      "epoch": 0.6531188937321509,
      "flos": 19316134638720.0,
      "grad_norm": 1.8515995145069566,
      "language_loss": 0.76960623,
      "learning_rate": 1.13476481851592e-06,
      "loss": 0.7908076,
      "num_input_tokens_seen": 234627335,
      "step": 10863,
      "time_per_iteration": 2.509773015975952
    },
    {
      "auxiliary_loss_clip": 0.01088656,
      "auxiliary_loss_mlp": 0.01031279,
      "balance_loss_clip": 1.03767371,
      "balance_loss_mlp": 1.01945353,
      "epoch": 0.6531790169848188,
      "flos": 22893771116160.0,
      "grad_norm": 1.7485438131262032,
      "language_loss": 0.7496413,
      "learning_rate": 1.1344137050270739e-06,
      "loss": 0.77084064,
      "num_input_tokens_seen": 234646540,
      "step": 10864,
      "time_per_iteration": 2.5541629791259766
    },
    {
      "auxiliary_loss_clip": 0.01096195,
      "auxiliary_loss_mlp": 0.01035596,
      "balance_loss_clip": 1.03746128,
      "balance_loss_mlp": 1.0236752,
      "epoch": 0.6532391402374869,
      "flos": 29563530912000.0,
      "grad_norm": 2.0607523566270265,
      "language_loss": 0.86101133,
      "learning_rate": 1.1340626243613458e-06,
      "loss": 0.88232923,
      "num_input_tokens_seen": 234665470,
      "step": 10865,
      "time_per_iteration": 2.5562562942504883
    },
    {
      "auxiliary_loss_clip": 0.01082487,
      "auxiliary_loss_mlp": 0.00784578,
      "balance_loss_clip": 1.03818643,
      "balance_loss_mlp": 1.0074079,
      "epoch": 0.6532992634901548,
      "flos": 23105463920640.0,
      "grad_norm": 1.613110110397674,
      "language_loss": 0.81552613,
      "learning_rate": 1.133711576532051e-06,
      "loss": 0.83419681,
      "num_input_tokens_seen": 234683955,
      "step": 10866,
      "time_per_iteration": 2.5847620964050293
    },
    {
      "auxiliary_loss_clip": 0.01087972,
      "auxiliary_loss_mlp": 0.01029404,
      "balance_loss_clip": 1.03894162,
      "balance_loss_mlp": 1.017501,
      "epoch": 0.6533593867428228,
      "flos": 26067340523520.0,
      "grad_norm": 1.4194325761346736,
      "language_loss": 0.82335746,
      "learning_rate": 1.1333605615524995e-06,
      "loss": 0.84453118,
      "num_input_tokens_seen": 234704595,
      "step": 10867,
      "time_per_iteration": 2.5538136959075928
    },
    {
      "auxiliary_loss_clip": 0.010893,
      "auxiliary_loss_mlp": 0.01026039,
      "balance_loss_clip": 1.03794134,
      "balance_loss_mlp": 1.01400495,
      "epoch": 0.6534195099954908,
      "flos": 21212469262080.0,
      "grad_norm": 1.769531282493091,
      "language_loss": 0.81221932,
      "learning_rate": 1.1330095794360016e-06,
      "loss": 0.83337271,
      "num_input_tokens_seen": 234724090,
      "step": 10868,
      "time_per_iteration": 2.5559310913085938
    },
    {
      "auxiliary_loss_clip": 0.01080513,
      "auxiliary_loss_mlp": 0.01029419,
      "balance_loss_clip": 1.03910804,
      "balance_loss_mlp": 1.01614511,
      "epoch": 0.6534796332481587,
      "flos": 19646584784640.0,
      "grad_norm": 1.8119193745419222,
      "language_loss": 0.79635876,
      "learning_rate": 1.1326586301958675e-06,
      "loss": 0.81745803,
      "num_input_tokens_seen": 234742560,
      "step": 10869,
      "time_per_iteration": 2.5338783264160156
    },
    {
      "auxiliary_loss_clip": 0.0110191,
      "auxiliary_loss_mlp": 0.01036466,
      "balance_loss_clip": 1.04027116,
      "balance_loss_mlp": 1.02400899,
      "epoch": 0.6535397565008267,
      "flos": 24022479162240.0,
      "grad_norm": 1.6470435219842774,
      "language_loss": 0.71647787,
      "learning_rate": 1.1323077138454063e-06,
      "loss": 0.73786157,
      "num_input_tokens_seen": 234762315,
      "step": 10870,
      "time_per_iteration": 2.5453672409057617
    },
    {
      "auxiliary_loss_clip": 0.01072374,
      "auxiliary_loss_mlp": 0.01034361,
      "balance_loss_clip": 1.04071116,
      "balance_loss_mlp": 1.02201152,
      "epoch": 0.6535998797534947,
      "flos": 24602759377920.0,
      "grad_norm": 2.0343108444804243,
      "language_loss": 0.74712098,
      "learning_rate": 1.1319568303979221e-06,
      "loss": 0.7681883,
      "num_input_tokens_seen": 234781300,
      "step": 10871,
      "time_per_iteration": 2.5975208282470703
    },
    {
      "auxiliary_loss_clip": 0.01096596,
      "auxiliary_loss_mlp": 0.00782041,
      "balance_loss_clip": 1.03843784,
      "balance_loss_mlp": 1.0076139,
      "epoch": 0.6536600030061627,
      "flos": 23364164649600.0,
      "grad_norm": 1.875299406284731,
      "language_loss": 0.55750573,
      "learning_rate": 1.1316059798667227e-06,
      "loss": 0.57629216,
      "num_input_tokens_seen": 234801040,
      "step": 10872,
      "time_per_iteration": 2.568873167037964
    },
    {
      "auxiliary_loss_clip": 0.01084774,
      "auxiliary_loss_mlp": 0.01030444,
      "balance_loss_clip": 1.03760636,
      "balance_loss_mlp": 1.01895797,
      "epoch": 0.6537201262588306,
      "flos": 23878477537920.0,
      "grad_norm": 2.0157718828644065,
      "language_loss": 0.74593991,
      "learning_rate": 1.1312551622651112e-06,
      "loss": 0.76709205,
      "num_input_tokens_seen": 234821415,
      "step": 10873,
      "time_per_iteration": 2.549917221069336
    },
    {
      "auxiliary_loss_clip": 0.01099346,
      "auxiliary_loss_mlp": 0.01032145,
      "balance_loss_clip": 1.0393014,
      "balance_loss_mlp": 1.02011669,
      "epoch": 0.6537802495114986,
      "flos": 24354760901760.0,
      "grad_norm": 2.8793260969181955,
      "language_loss": 0.75525743,
      "learning_rate": 1.1309043776063917e-06,
      "loss": 0.77657235,
      "num_input_tokens_seen": 234843795,
      "step": 10874,
      "time_per_iteration": 2.5836093425750732
    },
    {
      "auxiliary_loss_clip": 0.01066368,
      "auxiliary_loss_mlp": 0.01033984,
      "balance_loss_clip": 1.03696704,
      "balance_loss_mlp": 1.02151453,
      "epoch": 0.6538403727641665,
      "flos": 27996892248960.0,
      "grad_norm": 1.5187063072012412,
      "language_loss": 0.81436902,
      "learning_rate": 1.1305536259038642e-06,
      "loss": 0.83537251,
      "num_input_tokens_seen": 234862350,
      "step": 10875,
      "time_per_iteration": 2.607084035873413
    },
    {
      "auxiliary_loss_clip": 0.01109289,
      "auxiliary_loss_mlp": 0.01033868,
      "balance_loss_clip": 1.03787315,
      "balance_loss_mlp": 1.02200723,
      "epoch": 0.6539004960168345,
      "flos": 27563594486400.0,
      "grad_norm": 1.5632735539068399,
      "language_loss": 0.69671249,
      "learning_rate": 1.1302029071708314e-06,
      "loss": 0.71814412,
      "num_input_tokens_seen": 234881790,
      "step": 10876,
      "time_per_iteration": 2.519221782684326
    },
    {
      "auxiliary_loss_clip": 0.01022337,
      "auxiliary_loss_mlp": 0.01032827,
      "balance_loss_clip": 1.03706181,
      "balance_loss_mlp": 1.02051878,
      "epoch": 0.6539606192695024,
      "flos": 14530067879040.0,
      "grad_norm": 2.095460281106697,
      "language_loss": 0.795169,
      "learning_rate": 1.1298522214205908e-06,
      "loss": 0.81572068,
      "num_input_tokens_seen": 234897775,
      "step": 10877,
      "time_per_iteration": 2.6558961868286133
    },
    {
      "auxiliary_loss_clip": 0.01087244,
      "auxiliary_loss_mlp": 0.007837,
      "balance_loss_clip": 1.03653634,
      "balance_loss_mlp": 1.00687265,
      "epoch": 0.6540207425221705,
      "flos": 21616356764160.0,
      "grad_norm": 2.4313923327090117,
      "language_loss": 0.80020893,
      "learning_rate": 1.1295015686664408e-06,
      "loss": 0.81891835,
      "num_input_tokens_seen": 234918395,
      "step": 10878,
      "time_per_iteration": 2.5630314350128174
    },
    {
      "auxiliary_loss_clip": 0.01087121,
      "auxiliary_loss_mlp": 0.01030421,
      "balance_loss_clip": 1.03700638,
      "balance_loss_mlp": 1.01731443,
      "epoch": 0.6540808657748384,
      "flos": 17668983640320.0,
      "grad_norm": 1.841258597107234,
      "language_loss": 0.84446859,
      "learning_rate": 1.1291509489216797e-06,
      "loss": 0.86564398,
      "num_input_tokens_seen": 234936260,
      "step": 10879,
      "time_per_iteration": 2.5075719356536865
    },
    {
      "auxiliary_loss_clip": 0.0108959,
      "auxiliary_loss_mlp": 0.01030572,
      "balance_loss_clip": 1.0381403,
      "balance_loss_mlp": 1.01776934,
      "epoch": 0.6541409890275064,
      "flos": 14538292093440.0,
      "grad_norm": 2.4825455538715664,
      "language_loss": 0.71814543,
      "learning_rate": 1.128800362199601e-06,
      "loss": 0.7393471,
      "num_input_tokens_seen": 234952110,
      "step": 10880,
      "time_per_iteration": 2.5109384059906006
    },
    {
      "auxiliary_loss_clip": 0.01068131,
      "auxiliary_loss_mlp": 0.01035287,
      "balance_loss_clip": 1.03466284,
      "balance_loss_mlp": 1.02333605,
      "epoch": 0.6542011122801744,
      "flos": 17165301177600.0,
      "grad_norm": 1.847362264682207,
      "language_loss": 0.84179729,
      "learning_rate": 1.1284498085135005e-06,
      "loss": 0.86283147,
      "num_input_tokens_seen": 234970810,
      "step": 10881,
      "time_per_iteration": 2.534040927886963
    },
    {
      "auxiliary_loss_clip": 0.01075729,
      "auxiliary_loss_mlp": 0.01034845,
      "balance_loss_clip": 1.03714895,
      "balance_loss_mlp": 1.02121353,
      "epoch": 0.6542612355328423,
      "flos": 18186600579840.0,
      "grad_norm": 4.525366677783798,
      "language_loss": 0.77792585,
      "learning_rate": 1.1280992878766699e-06,
      "loss": 0.79903162,
      "num_input_tokens_seen": 234989565,
      "step": 10882,
      "time_per_iteration": 2.5615649223327637
    },
    {
      "auxiliary_loss_clip": 0.01113417,
      "auxiliary_loss_mlp": 0.0103057,
      "balance_loss_clip": 1.04128957,
      "balance_loss_mlp": 1.01773167,
      "epoch": 0.6543213587855103,
      "flos": 19792453916160.0,
      "grad_norm": 2.3371318049137706,
      "language_loss": 0.81885612,
      "learning_rate": 1.1277488003024024e-06,
      "loss": 0.84029591,
      "num_input_tokens_seen": 235007955,
      "step": 10883,
      "time_per_iteration": 2.4713690280914307
    },
    {
      "auxiliary_loss_clip": 0.01067735,
      "auxiliary_loss_mlp": 0.0103696,
      "balance_loss_clip": 1.03847122,
      "balance_loss_mlp": 1.0242939,
      "epoch": 0.6543814820381783,
      "flos": 21105096531840.0,
      "grad_norm": 2.7435056228217864,
      "language_loss": 0.85496563,
      "learning_rate": 1.127398345803988e-06,
      "loss": 0.87601256,
      "num_input_tokens_seen": 235024860,
      "step": 10884,
      "time_per_iteration": 2.5894830226898193
    },
    {
      "auxiliary_loss_clip": 0.01087766,
      "auxiliary_loss_mlp": 0.01033576,
      "balance_loss_clip": 1.03778529,
      "balance_loss_mlp": 1.0211308,
      "epoch": 0.6544416052908463,
      "flos": 20194042947840.0,
      "grad_norm": 3.9736559598563046,
      "language_loss": 0.8019731,
      "learning_rate": 1.127047924394715e-06,
      "loss": 0.82318646,
      "num_input_tokens_seen": 235043815,
      "step": 10885,
      "time_per_iteration": 3.912710189819336
    },
    {
      "auxiliary_loss_clip": 0.01071218,
      "auxiliary_loss_mlp": 0.0102659,
      "balance_loss_clip": 1.03616595,
      "balance_loss_mlp": 1.01410866,
      "epoch": 0.6545017285435142,
      "flos": 23368258800000.0,
      "grad_norm": 1.9799649291500878,
      "language_loss": 0.72243744,
      "learning_rate": 1.1266975360878722e-06,
      "loss": 0.74341553,
      "num_input_tokens_seen": 235062985,
      "step": 10886,
      "time_per_iteration": 2.6095807552337646
    },
    {
      "auxiliary_loss_clip": 0.0109142,
      "auxiliary_loss_mlp": 0.01029772,
      "balance_loss_clip": 1.03834307,
      "balance_loss_mlp": 1.01833439,
      "epoch": 0.6545618517961822,
      "flos": 19134714021120.0,
      "grad_norm": 1.9253947887259737,
      "language_loss": 0.78439051,
      "learning_rate": 1.1263471808967468e-06,
      "loss": 0.80560243,
      "num_input_tokens_seen": 235081670,
      "step": 10887,
      "time_per_iteration": 2.4991934299468994
    },
    {
      "auxiliary_loss_clip": 0.01081157,
      "auxiliary_loss_mlp": 0.01028871,
      "balance_loss_clip": 1.03582907,
      "balance_loss_mlp": 1.01706386,
      "epoch": 0.6546219750488501,
      "flos": 14938624149120.0,
      "grad_norm": 1.8295015868568918,
      "language_loss": 0.78982043,
      "learning_rate": 1.1259968588346234e-06,
      "loss": 0.81092072,
      "num_input_tokens_seen": 235098510,
      "step": 10888,
      "time_per_iteration": 2.5154595375061035
    },
    {
      "auxiliary_loss_clip": 0.01094807,
      "auxiliary_loss_mlp": 0.01027046,
      "balance_loss_clip": 1.03706431,
      "balance_loss_mlp": 1.01581025,
      "epoch": 0.6546820983015181,
      "flos": 36320518886400.0,
      "grad_norm": 1.6388845895555322,
      "language_loss": 0.66443855,
      "learning_rate": 1.1256465699147874e-06,
      "loss": 0.68565708,
      "num_input_tokens_seen": 235119990,
      "step": 10889,
      "time_per_iteration": 2.635481595993042
    },
    {
      "auxiliary_loss_clip": 0.01082975,
      "auxiliary_loss_mlp": 0.01037286,
      "balance_loss_clip": 1.0362339,
      "balance_loss_mlp": 1.0228734,
      "epoch": 0.654742221554186,
      "flos": 20411446014720.0,
      "grad_norm": 1.4411008137715386,
      "language_loss": 0.79942608,
      "learning_rate": 1.1252963141505203e-06,
      "loss": 0.8206287,
      "num_input_tokens_seen": 235139255,
      "step": 10890,
      "time_per_iteration": 2.535886287689209
    },
    {
      "auxiliary_loss_clip": 0.01096971,
      "auxiliary_loss_mlp": 0.00785566,
      "balance_loss_clip": 1.03591442,
      "balance_loss_mlp": 1.01205635,
      "epoch": 0.6548023448068541,
      "flos": 24863650836480.0,
      "grad_norm": 2.1220157167283005,
      "language_loss": 0.65587497,
      "learning_rate": 1.1249460915551052e-06,
      "loss": 0.67470032,
      "num_input_tokens_seen": 235158455,
      "step": 10891,
      "time_per_iteration": 2.531151056289673
    },
    {
      "auxiliary_loss_clip": 0.01095059,
      "auxiliary_loss_mlp": 0.01031746,
      "balance_loss_clip": 1.03685737,
      "balance_loss_mlp": 1.01994455,
      "epoch": 0.654862468059522,
      "flos": 21427573858560.0,
      "grad_norm": 1.7752708378056687,
      "language_loss": 0.79383391,
      "learning_rate": 1.1245959021418214e-06,
      "loss": 0.81510198,
      "num_input_tokens_seen": 235177350,
      "step": 10892,
      "time_per_iteration": 2.506728410720825
    },
    {
      "auxiliary_loss_clip": 0.01103002,
      "auxiliary_loss_mlp": 0.01032236,
      "balance_loss_clip": 1.04070961,
      "balance_loss_mlp": 1.0207864,
      "epoch": 0.65492259131219,
      "flos": 26577846570240.0,
      "grad_norm": 1.9735708056841559,
      "language_loss": 0.78078663,
      "learning_rate": 1.1242457459239497e-06,
      "loss": 0.80213904,
      "num_input_tokens_seen": 235196435,
      "step": 10893,
      "time_per_iteration": 3.959026336669922
    },
    {
      "auxiliary_loss_clip": 0.01113494,
      "auxiliary_loss_mlp": 0.01029418,
      "balance_loss_clip": 1.03963482,
      "balance_loss_mlp": 1.01593578,
      "epoch": 0.6549827145648579,
      "flos": 21501334437120.0,
      "grad_norm": 1.6143446364421794,
      "language_loss": 0.70238882,
      "learning_rate": 1.123895622914766e-06,
      "loss": 0.72381794,
      "num_input_tokens_seen": 235215430,
      "step": 10894,
      "time_per_iteration": 2.4856536388397217
    },
    {
      "auxiliary_loss_clip": 0.01100924,
      "auxiliary_loss_mlp": 0.01032113,
      "balance_loss_clip": 1.0370717,
      "balance_loss_mlp": 1.01997137,
      "epoch": 0.6550428378175259,
      "flos": 22594275515520.0,
      "grad_norm": 2.7203556887317437,
      "language_loss": 0.62210202,
      "learning_rate": 1.123545533127549e-06,
      "loss": 0.64343238,
      "num_input_tokens_seen": 235232015,
      "step": 10895,
      "time_per_iteration": 3.8678109645843506
    },
    {
      "auxiliary_loss_clip": 0.01096384,
      "auxiliary_loss_mlp": 0.01034398,
      "balance_loss_clip": 1.03733349,
      "balance_loss_mlp": 1.02252483,
      "epoch": 0.655102961070194,
      "flos": 12823809050880.0,
      "grad_norm": 2.0154067341578075,
      "language_loss": 0.79223192,
      "learning_rate": 1.1231954765755722e-06,
      "loss": 0.81353974,
      "num_input_tokens_seen": 235248115,
      "step": 10896,
      "time_per_iteration": 3.919327735900879
    },
    {
      "auxiliary_loss_clip": 0.01086508,
      "auxiliary_loss_mlp": 0.0103315,
      "balance_loss_clip": 1.03857374,
      "balance_loss_mlp": 1.02164006,
      "epoch": 0.6551630843228619,
      "flos": 24791075406720.0,
      "grad_norm": 1.4040748850926725,
      "language_loss": 0.70801938,
      "learning_rate": 1.1228454532721111e-06,
      "loss": 0.72921598,
      "num_input_tokens_seen": 235270785,
      "step": 10897,
      "time_per_iteration": 2.583308696746826
    },
    {
      "auxiliary_loss_clip": 0.01110685,
      "auxiliary_loss_mlp": 0.01032442,
      "balance_loss_clip": 1.03797257,
      "balance_loss_mlp": 1.02001405,
      "epoch": 0.6552232075755299,
      "flos": 16724461559040.0,
      "grad_norm": 1.5446789999960828,
      "language_loss": 0.75449693,
      "learning_rate": 1.1224954632304391e-06,
      "loss": 0.77592826,
      "num_input_tokens_seen": 235287905,
      "step": 10898,
      "time_per_iteration": 2.4441328048706055
    },
    {
      "auxiliary_loss_clip": 0.01088076,
      "auxiliary_loss_mlp": 0.01033366,
      "balance_loss_clip": 1.03904319,
      "balance_loss_mlp": 1.02148664,
      "epoch": 0.6552833308281978,
      "flos": 22016473338240.0,
      "grad_norm": 2.607113137539881,
      "language_loss": 0.7370007,
      "learning_rate": 1.122145506463827e-06,
      "loss": 0.75821513,
      "num_input_tokens_seen": 235305525,
      "step": 10899,
      "time_per_iteration": 2.522364616394043
    },
    {
      "auxiliary_loss_clip": 0.01088621,
      "auxiliary_loss_mlp": 0.01028141,
      "balance_loss_clip": 1.03916264,
      "balance_loss_mlp": 1.01621389,
      "epoch": 0.6553434540808658,
      "flos": 24863399441280.0,
      "grad_norm": 1.8367856406331626,
      "language_loss": 0.55525243,
      "learning_rate": 1.1217955829855443e-06,
      "loss": 0.57642007,
      "num_input_tokens_seen": 235324415,
      "step": 10900,
      "time_per_iteration": 2.5799546241760254
    },
    {
      "auxiliary_loss_clip": 0.0109805,
      "auxiliary_loss_mlp": 0.01034605,
      "balance_loss_clip": 1.03920555,
      "balance_loss_mlp": 1.02137923,
      "epoch": 0.6554035773335337,
      "flos": 23221060865280.0,
      "grad_norm": 1.7363179209225317,
      "language_loss": 0.76883984,
      "learning_rate": 1.1214456928088622e-06,
      "loss": 0.79016644,
      "num_input_tokens_seen": 235341595,
      "step": 10901,
      "time_per_iteration": 2.492652416229248
    },
    {
      "auxiliary_loss_clip": 0.01107665,
      "auxiliary_loss_mlp": 0.01029335,
      "balance_loss_clip": 1.03799772,
      "balance_loss_mlp": 1.01632953,
      "epoch": 0.6554637005862017,
      "flos": 22783597125120.0,
      "grad_norm": 1.803819093287761,
      "language_loss": 0.73036075,
      "learning_rate": 1.1210958359470463e-06,
      "loss": 0.7517308,
      "num_input_tokens_seen": 235361700,
      "step": 10902,
      "time_per_iteration": 2.4898486137390137
    },
    {
      "auxiliary_loss_clip": 0.01108466,
      "auxiliary_loss_mlp": 0.01031355,
      "balance_loss_clip": 1.03959239,
      "balance_loss_mlp": 1.01951742,
      "epoch": 0.6555238238388696,
      "flos": 21507224267520.0,
      "grad_norm": 1.5487699655676994,
      "language_loss": 0.67837745,
      "learning_rate": 1.1207460124133645e-06,
      "loss": 0.6997757,
      "num_input_tokens_seen": 235382065,
      "step": 10903,
      "time_per_iteration": 2.4815502166748047
    },
    {
      "auxiliary_loss_clip": 0.01089874,
      "auxiliary_loss_mlp": 0.00786394,
      "balance_loss_clip": 1.03664851,
      "balance_loss_mlp": 1.01119137,
      "epoch": 0.6555839470915377,
      "flos": 30519473518080.0,
      "grad_norm": 1.6766263631888907,
      "language_loss": 0.6687727,
      "learning_rate": 1.1203962222210832e-06,
      "loss": 0.68753541,
      "num_input_tokens_seen": 235402130,
      "step": 10904,
      "time_per_iteration": 2.6318843364715576
    },
    {
      "auxiliary_loss_clip": 0.01101284,
      "auxiliary_loss_mlp": 0.01040065,
      "balance_loss_clip": 1.03723419,
      "balance_loss_mlp": 1.02643991,
      "epoch": 0.6556440703442056,
      "flos": 24642943718400.0,
      "grad_norm": 1.7050090131214304,
      "language_loss": 0.90187824,
      "learning_rate": 1.120046465383464e-06,
      "loss": 0.92329174,
      "num_input_tokens_seen": 235420435,
      "step": 10905,
      "time_per_iteration": 2.5309221744537354
    },
    {
      "auxiliary_loss_clip": 0.01096587,
      "auxiliary_loss_mlp": 0.01033692,
      "balance_loss_clip": 1.0370605,
      "balance_loss_mlp": 1.02168131,
      "epoch": 0.6557041935968736,
      "flos": 23732464752000.0,
      "grad_norm": 1.741994651852035,
      "language_loss": 0.75895923,
      "learning_rate": 1.1196967419137721e-06,
      "loss": 0.78026205,
      "num_input_tokens_seen": 235439960,
      "step": 10906,
      "time_per_iteration": 2.5316450595855713
    },
    {
      "auxiliary_loss_clip": 0.01114595,
      "auxiliary_loss_mlp": 0.01040804,
      "balance_loss_clip": 1.04115486,
      "balance_loss_mlp": 1.02823925,
      "epoch": 0.6557643168495415,
      "flos": 11102753819520.0,
      "grad_norm": 2.4926634801994014,
      "language_loss": 0.74042159,
      "learning_rate": 1.119347051825267e-06,
      "loss": 0.76197553,
      "num_input_tokens_seen": 235457495,
      "step": 10907,
      "time_per_iteration": 2.4513485431671143
    },
    {
      "auxiliary_loss_clip": 0.01069272,
      "auxiliary_loss_mlp": 0.01030498,
      "balance_loss_clip": 1.03824818,
      "balance_loss_mlp": 1.01720047,
      "epoch": 0.6558244401022095,
      "flos": 30191034533760.0,
      "grad_norm": 1.4563662230411119,
      "language_loss": 0.72106874,
      "learning_rate": 1.118997395131211e-06,
      "loss": 0.74206644,
      "num_input_tokens_seen": 235479525,
      "step": 10908,
      "time_per_iteration": 2.651982307434082
    },
    {
      "auxiliary_loss_clip": 0.01111297,
      "auxiliary_loss_mlp": 0.01034454,
      "balance_loss_clip": 1.03956842,
      "balance_loss_mlp": 1.02095902,
      "epoch": 0.6558845633548775,
      "flos": 17931060247680.0,
      "grad_norm": 2.3400954174110056,
      "language_loss": 0.81519341,
      "learning_rate": 1.118647771844861e-06,
      "loss": 0.83665097,
      "num_input_tokens_seen": 235496305,
      "step": 10909,
      "time_per_iteration": 2.4321510791778564
    },
    {
      "auxiliary_loss_clip": 0.01111168,
      "auxiliary_loss_mlp": 0.01037052,
      "balance_loss_clip": 1.03928018,
      "balance_loss_mlp": 1.02397442,
      "epoch": 0.6559446866075455,
      "flos": 21904144531200.0,
      "grad_norm": 2.1365357843882196,
      "language_loss": 0.6432696,
      "learning_rate": 1.1182981819794767e-06,
      "loss": 0.66475177,
      "num_input_tokens_seen": 235512545,
      "step": 10910,
      "time_per_iteration": 2.4814348220825195
    },
    {
      "auxiliary_loss_clip": 0.01088948,
      "auxiliary_loss_mlp": 0.01035976,
      "balance_loss_clip": 1.03747988,
      "balance_loss_mlp": 1.02165258,
      "epoch": 0.6560048098602135,
      "flos": 14127976056960.0,
      "grad_norm": 2.7659221474624665,
      "language_loss": 0.75299627,
      "learning_rate": 1.117948625548313e-06,
      "loss": 0.7742455,
      "num_input_tokens_seen": 235526045,
      "step": 10911,
      "time_per_iteration": 2.4738376140594482
    },
    {
      "auxiliary_loss_clip": 0.01105677,
      "auxiliary_loss_mlp": 0.01029545,
      "balance_loss_clip": 1.03766096,
      "balance_loss_mlp": 1.01815462,
      "epoch": 0.6560649331128814,
      "flos": 18807567926400.0,
      "grad_norm": 1.5200852465908343,
      "language_loss": 0.75296658,
      "learning_rate": 1.1175991025646265e-06,
      "loss": 0.77431881,
      "num_input_tokens_seen": 235545285,
      "step": 10912,
      "time_per_iteration": 2.514439821243286
    },
    {
      "auxiliary_loss_clip": 0.01076924,
      "auxiliary_loss_mlp": 0.00787216,
      "balance_loss_clip": 1.0415169,
      "balance_loss_mlp": 1.00960577,
      "epoch": 0.6561250563655494,
      "flos": 17053618815360.0,
      "grad_norm": 1.618503188776137,
      "language_loss": 0.77449143,
      "learning_rate": 1.1172496130416697e-06,
      "loss": 0.79313278,
      "num_input_tokens_seen": 235563150,
      "step": 10913,
      "time_per_iteration": 2.5515522956848145
    },
    {
      "auxiliary_loss_clip": 0.01078895,
      "auxiliary_loss_mlp": 0.01030943,
      "balance_loss_clip": 1.03542709,
      "balance_loss_mlp": 1.01988077,
      "epoch": 0.6561851796182173,
      "flos": 22637656166400.0,
      "grad_norm": 1.6684927938194365,
      "language_loss": 0.71227038,
      "learning_rate": 1.1169001569926961e-06,
      "loss": 0.73336875,
      "num_input_tokens_seen": 235582535,
      "step": 10914,
      "time_per_iteration": 2.531829357147217
    },
    {
      "auxiliary_loss_clip": 0.01078783,
      "auxiliary_loss_mlp": 0.01031537,
      "balance_loss_clip": 1.03844404,
      "balance_loss_mlp": 1.01913905,
      "epoch": 0.6562453028708853,
      "flos": 19239213663360.0,
      "grad_norm": 1.7070527763054648,
      "language_loss": 0.7402603,
      "learning_rate": 1.116550734430958e-06,
      "loss": 0.76136351,
      "num_input_tokens_seen": 235601490,
      "step": 10915,
      "time_per_iteration": 2.5312304496765137
    },
    {
      "auxiliary_loss_clip": 0.01067605,
      "auxiliary_loss_mlp": 0.01030573,
      "balance_loss_clip": 1.0364567,
      "balance_loss_mlp": 1.01738238,
      "epoch": 0.6563054261235532,
      "flos": 23801305167360.0,
      "grad_norm": 1.5556434291678525,
      "language_loss": 0.79454994,
      "learning_rate": 1.1162013453697042e-06,
      "loss": 0.81553173,
      "num_input_tokens_seen": 235619165,
      "step": 10916,
      "time_per_iteration": 2.5691590309143066
    },
    {
      "auxiliary_loss_clip": 0.01087859,
      "auxiliary_loss_mlp": 0.01036089,
      "balance_loss_clip": 1.03686178,
      "balance_loss_mlp": 1.0242393,
      "epoch": 0.6563655493762213,
      "flos": 19240039676160.0,
      "grad_norm": 1.8932857043353615,
      "language_loss": 0.76368535,
      "learning_rate": 1.1158519898221831e-06,
      "loss": 0.78492486,
      "num_input_tokens_seen": 235637115,
      "step": 10917,
      "time_per_iteration": 2.4964098930358887
    },
    {
      "auxiliary_loss_clip": 0.01108189,
      "auxiliary_loss_mlp": 0.00785681,
      "balance_loss_clip": 1.03817725,
      "balance_loss_mlp": 1.01213837,
      "epoch": 0.6564256726288892,
      "flos": 25556439427200.0,
      "grad_norm": 2.058053847025102,
      "language_loss": 0.70270252,
      "learning_rate": 1.1155026678016445e-06,
      "loss": 0.72164118,
      "num_input_tokens_seen": 235656330,
      "step": 10918,
      "time_per_iteration": 2.5073118209838867
    },
    {
      "auxiliary_loss_clip": 0.01073582,
      "auxiliary_loss_mlp": 0.01034107,
      "balance_loss_clip": 1.03773141,
      "balance_loss_mlp": 1.02238274,
      "epoch": 0.6564857958815572,
      "flos": 22200623389440.0,
      "grad_norm": 1.717615659092299,
      "language_loss": 0.76589179,
      "learning_rate": 1.115153379321332e-06,
      "loss": 0.78696871,
      "num_input_tokens_seen": 235674510,
      "step": 10919,
      "time_per_iteration": 2.5614089965820312
    },
    {
      "auxiliary_loss_clip": 0.01033647,
      "auxiliary_loss_mlp": 0.00769247,
      "balance_loss_clip": 1.01979208,
      "balance_loss_mlp": 1.010167,
      "epoch": 0.6565459191342251,
      "flos": 58123144604160.0,
      "grad_norm": 0.8189887772813239,
      "language_loss": 0.53056657,
      "learning_rate": 1.1148041243944931e-06,
      "loss": 0.54859543,
      "num_input_tokens_seen": 235735050,
      "step": 10920,
      "time_per_iteration": 3.1434290409088135
    },
    {
      "auxiliary_loss_clip": 0.0109809,
      "auxiliary_loss_mlp": 0.01029961,
      "balance_loss_clip": 1.03791666,
      "balance_loss_mlp": 1.01754498,
      "epoch": 0.6566060423868931,
      "flos": 30809631582720.0,
      "grad_norm": 1.7048351079846842,
      "language_loss": 0.65675789,
      "learning_rate": 1.1144549030343697e-06,
      "loss": 0.67803842,
      "num_input_tokens_seen": 235757545,
      "step": 10921,
      "time_per_iteration": 2.5793631076812744
    },
    {
      "auxiliary_loss_clip": 0.01083497,
      "auxiliary_loss_mlp": 0.01040231,
      "balance_loss_clip": 1.0362246,
      "balance_loss_mlp": 1.02535915,
      "epoch": 0.6566661656395612,
      "flos": 23367432787200.0,
      "grad_norm": 1.9742450489124095,
      "language_loss": 0.81682491,
      "learning_rate": 1.114105715254205e-06,
      "loss": 0.83806217,
      "num_input_tokens_seen": 235777265,
      "step": 10922,
      "time_per_iteration": 2.5383758544921875
    },
    {
      "auxiliary_loss_clip": 0.01061919,
      "auxiliary_loss_mlp": 0.00786362,
      "balance_loss_clip": 1.0395447,
      "balance_loss_mlp": 1.00730646,
      "epoch": 0.6567262888922291,
      "flos": 25735597488000.0,
      "grad_norm": 2.6797044190952404,
      "language_loss": 0.71179086,
      "learning_rate": 1.1137565610672414e-06,
      "loss": 0.73027366,
      "num_input_tokens_seen": 235796565,
      "step": 10923,
      "time_per_iteration": 2.6199796199798584
    },
    {
      "auxiliary_loss_clip": 0.01074722,
      "auxiliary_loss_mlp": 0.01031441,
      "balance_loss_clip": 1.03949857,
      "balance_loss_mlp": 1.01933503,
      "epoch": 0.6567864121448971,
      "flos": 17123716206720.0,
      "grad_norm": 2.1300040269042864,
      "language_loss": 0.81046969,
      "learning_rate": 1.1134074404867169e-06,
      "loss": 0.83153129,
      "num_input_tokens_seen": 235814805,
      "step": 10924,
      "time_per_iteration": 3.9857194423675537
    },
    {
      "auxiliary_loss_clip": 0.01094173,
      "auxiliary_loss_mlp": 0.01030409,
      "balance_loss_clip": 1.03714895,
      "balance_loss_mlp": 1.01877975,
      "epoch": 0.656846535397565,
      "flos": 22419319345920.0,
      "grad_norm": 1.5318290311240728,
      "language_loss": 0.72245705,
      "learning_rate": 1.1130583535258717e-06,
      "loss": 0.74370289,
      "num_input_tokens_seen": 235833405,
      "step": 10925,
      "time_per_iteration": 2.5227954387664795
    },
    {
      "auxiliary_loss_clip": 0.01096496,
      "auxiliary_loss_mlp": 0.01028031,
      "balance_loss_clip": 1.03675818,
      "balance_loss_mlp": 1.0155493,
      "epoch": 0.656906658650233,
      "flos": 17704535126400.0,
      "grad_norm": 2.1032170221013704,
      "language_loss": 0.72379535,
      "learning_rate": 1.112709300197942e-06,
      "loss": 0.74504066,
      "num_input_tokens_seen": 235848530,
      "step": 10926,
      "time_per_iteration": 2.454610824584961
    },
    {
      "auxiliary_loss_clip": 0.01072578,
      "auxiliary_loss_mlp": 0.01032557,
      "balance_loss_clip": 1.03776121,
      "balance_loss_mlp": 1.01945615,
      "epoch": 0.6569667819029009,
      "flos": 21175158009600.0,
      "grad_norm": 1.6901319476694325,
      "language_loss": 0.72691488,
      "learning_rate": 1.1123602805161656e-06,
      "loss": 0.74796629,
      "num_input_tokens_seen": 235867225,
      "step": 10927,
      "time_per_iteration": 2.5982627868652344
    },
    {
      "auxiliary_loss_clip": 0.0101282,
      "auxiliary_loss_mlp": 0.01002606,
      "balance_loss_clip": 1.0197196,
      "balance_loss_mlp": 1.00136662,
      "epoch": 0.6570269051555689,
      "flos": 68761897511040.0,
      "grad_norm": 0.7425789246489861,
      "language_loss": 0.64459002,
      "learning_rate": 1.112011294493775e-06,
      "loss": 0.66474426,
      "num_input_tokens_seen": 235932925,
      "step": 10928,
      "time_per_iteration": 3.1639184951782227
    },
    {
      "auxiliary_loss_clip": 0.01097391,
      "auxiliary_loss_mlp": 0.01031812,
      "balance_loss_clip": 1.0370152,
      "balance_loss_mlp": 1.01933718,
      "epoch": 0.6570870284082369,
      "flos": 26319289495680.0,
      "grad_norm": 1.8127148322452193,
      "language_loss": 0.778157,
      "learning_rate": 1.1116623421440063e-06,
      "loss": 0.79944897,
      "num_input_tokens_seen": 235952680,
      "step": 10929,
      "time_per_iteration": 2.5535881519317627
    },
    {
      "auxiliary_loss_clip": 0.01076061,
      "auxiliary_loss_mlp": 0.01032997,
      "balance_loss_clip": 1.03775978,
      "balance_loss_mlp": 1.02052212,
      "epoch": 0.6571471516609049,
      "flos": 26174749167360.0,
      "grad_norm": 1.8692325043586049,
      "language_loss": 0.65370613,
      "learning_rate": 1.1113134234800895e-06,
      "loss": 0.6747967,
      "num_input_tokens_seen": 235972075,
      "step": 10930,
      "time_per_iteration": 2.599954605102539
    },
    {
      "auxiliary_loss_clip": 0.01062329,
      "auxiliary_loss_mlp": 0.01030535,
      "balance_loss_clip": 1.03582382,
      "balance_loss_mlp": 1.01740372,
      "epoch": 0.6572072749135728,
      "flos": 20376253664640.0,
      "grad_norm": 1.540139726926366,
      "language_loss": 0.70585549,
      "learning_rate": 1.110964538515258e-06,
      "loss": 0.72678411,
      "num_input_tokens_seen": 235990340,
      "step": 10931,
      "time_per_iteration": 2.588287591934204
    },
    {
      "auxiliary_loss_clip": 0.01066136,
      "auxiliary_loss_mlp": 0.01033907,
      "balance_loss_clip": 1.04021871,
      "balance_loss_mlp": 1.02209318,
      "epoch": 0.6572673981662408,
      "flos": 17128744110720.0,
      "grad_norm": 1.9249751670693578,
      "language_loss": 0.68712294,
      "learning_rate": 1.1106156872627393e-06,
      "loss": 0.70812333,
      "num_input_tokens_seen": 236007470,
      "step": 10932,
      "time_per_iteration": 3.9705886840820312
    },
    {
      "auxiliary_loss_clip": 0.01083894,
      "auxiliary_loss_mlp": 0.00784373,
      "balance_loss_clip": 1.03421426,
      "balance_loss_mlp": 1.00894582,
      "epoch": 0.6573275214189087,
      "flos": 41275113281280.0,
      "grad_norm": 1.7262791622039473,
      "language_loss": 0.80009705,
      "learning_rate": 1.1102668697357626e-06,
      "loss": 0.81877971,
      "num_input_tokens_seen": 236029030,
      "step": 10933,
      "time_per_iteration": 2.6912357807159424
    },
    {
      "auxiliary_loss_clip": 0.01062026,
      "auxiliary_loss_mlp": 0.0103121,
      "balance_loss_clip": 1.03975034,
      "balance_loss_mlp": 1.01804948,
      "epoch": 0.6573876446715767,
      "flos": 22890143842560.0,
      "grad_norm": 1.577479343975404,
      "language_loss": 0.7354871,
      "learning_rate": 1.1099180859475571e-06,
      "loss": 0.75641948,
      "num_input_tokens_seen": 236047160,
      "step": 10934,
      "time_per_iteration": 2.59609317779541
    },
    {
      "auxiliary_loss_clip": 0.01092295,
      "auxiliary_loss_mlp": 0.01042411,
      "balance_loss_clip": 1.03622735,
      "balance_loss_mlp": 1.02836251,
      "epoch": 0.6574477679242448,
      "flos": 44018150273280.0,
      "grad_norm": 1.5900709566484772,
      "language_loss": 0.76386952,
      "learning_rate": 1.1095693359113454e-06,
      "loss": 0.78521657,
      "num_input_tokens_seen": 236069215,
      "step": 10935,
      "time_per_iteration": 4.12696099281311
    },
    {
      "auxiliary_loss_clip": 0.01071261,
      "auxiliary_loss_mlp": 0.01040486,
      "balance_loss_clip": 1.03728843,
      "balance_loss_mlp": 1.02656865,
      "epoch": 0.6575078911769127,
      "flos": 24571517523840.0,
      "grad_norm": 1.7686229124657173,
      "language_loss": 0.78033817,
      "learning_rate": 1.1092206196403538e-06,
      "loss": 0.80145562,
      "num_input_tokens_seen": 236088335,
      "step": 10936,
      "time_per_iteration": 2.5832505226135254
    },
    {
      "auxiliary_loss_clip": 0.01069666,
      "auxiliary_loss_mlp": 0.01028636,
      "balance_loss_clip": 1.03751242,
      "balance_loss_mlp": 1.01660132,
      "epoch": 0.6575680144295807,
      "flos": 20924035050240.0,
      "grad_norm": 1.840250693405559,
      "language_loss": 0.69199264,
      "learning_rate": 1.1088719371478056e-06,
      "loss": 0.71297562,
      "num_input_tokens_seen": 236108540,
      "step": 10937,
      "time_per_iteration": 2.595842123031616
    },
    {
      "auxiliary_loss_clip": 0.01083317,
      "auxiliary_loss_mlp": 0.01033101,
      "balance_loss_clip": 1.0394367,
      "balance_loss_mlp": 1.02059031,
      "epoch": 0.6576281376822486,
      "flos": 10925642833920.0,
      "grad_norm": 2.211503383760117,
      "language_loss": 0.69073159,
      "learning_rate": 1.1085232884469236e-06,
      "loss": 0.7118957,
      "num_input_tokens_seen": 236124495,
      "step": 10938,
      "time_per_iteration": 2.5007216930389404
    },
    {
      "auxiliary_loss_clip": 0.01082918,
      "auxiliary_loss_mlp": 0.01032266,
      "balance_loss_clip": 1.03909218,
      "balance_loss_mlp": 1.01921892,
      "epoch": 0.6576882609349166,
      "flos": 19281552819840.0,
      "grad_norm": 1.997515160961871,
      "language_loss": 0.71224135,
      "learning_rate": 1.108174673550927e-06,
      "loss": 0.73339319,
      "num_input_tokens_seen": 236142550,
      "step": 10939,
      "time_per_iteration": 2.52908992767334
    },
    {
      "auxiliary_loss_clip": 0.01088779,
      "auxiliary_loss_mlp": 0.0078919,
      "balance_loss_clip": 1.04016244,
      "balance_loss_mlp": 1.01674581,
      "epoch": 0.6577483841875845,
      "flos": 20220544206720.0,
      "grad_norm": 1.9809672850798477,
      "language_loss": 0.77577031,
      "learning_rate": 1.107826092473037e-06,
      "loss": 0.79455006,
      "num_input_tokens_seen": 236156620,
      "step": 10940,
      "time_per_iteration": 2.5060229301452637
    },
    {
      "auxiliary_loss_clip": 0.01067396,
      "auxiliary_loss_mlp": 0.01031791,
      "balance_loss_clip": 1.03545451,
      "balance_loss_mlp": 1.01865435,
      "epoch": 0.6578085074402525,
      "flos": 34751078962560.0,
      "grad_norm": 1.8593915106700138,
      "language_loss": 0.68431807,
      "learning_rate": 1.107477545226471e-06,
      "loss": 0.70530993,
      "num_input_tokens_seen": 236177095,
      "step": 10941,
      "time_per_iteration": 2.7031853199005127
    },
    {
      "auxiliary_loss_clip": 0.01093888,
      "auxiliary_loss_mlp": 0.00785908,
      "balance_loss_clip": 1.03533781,
      "balance_loss_mlp": 1.01122284,
      "epoch": 0.6578686306929205,
      "flos": 23470998675840.0,
      "grad_norm": 1.9365051030933063,
      "language_loss": 0.68396783,
      "learning_rate": 1.1071290318244448e-06,
      "loss": 0.70276582,
      "num_input_tokens_seen": 236194695,
      "step": 10942,
      "time_per_iteration": 2.5266947746276855
    },
    {
      "auxiliary_loss_clip": 0.01079372,
      "auxiliary_loss_mlp": 0.01036726,
      "balance_loss_clip": 1.03797984,
      "balance_loss_mlp": 1.02235544,
      "epoch": 0.6579287539455885,
      "flos": 18077073033600.0,
      "grad_norm": 1.78271150059956,
      "language_loss": 0.71440423,
      "learning_rate": 1.1067805522801753e-06,
      "loss": 0.73556513,
      "num_input_tokens_seen": 236213885,
      "step": 10943,
      "time_per_iteration": 2.539088726043701
    },
    {
      "auxiliary_loss_clip": 0.01072207,
      "auxiliary_loss_mlp": 0.01033719,
      "balance_loss_clip": 1.03580832,
      "balance_loss_mlp": 1.02075434,
      "epoch": 0.6579888771982564,
      "flos": 28661383900800.0,
      "grad_norm": 1.6602153534335515,
      "language_loss": 0.5902738,
      "learning_rate": 1.1064321066068778e-06,
      "loss": 0.61133307,
      "num_input_tokens_seen": 236237315,
      "step": 10944,
      "time_per_iteration": 2.6308932304382324
    },
    {
      "auxiliary_loss_clip": 0.01102569,
      "auxiliary_loss_mlp": 0.01034697,
      "balance_loss_clip": 1.03838277,
      "balance_loss_mlp": 1.02163219,
      "epoch": 0.6580490004509244,
      "flos": 25046543911680.0,
      "grad_norm": 1.5134433290499227,
      "language_loss": 0.72436273,
      "learning_rate": 1.1060836948177646e-06,
      "loss": 0.74573541,
      "num_input_tokens_seen": 236256345,
      "step": 10945,
      "time_per_iteration": 2.5506606101989746
    },
    {
      "auxiliary_loss_clip": 0.01087112,
      "auxiliary_loss_mlp": 0.01026005,
      "balance_loss_clip": 1.03854847,
      "balance_loss_mlp": 1.01456702,
      "epoch": 0.6581091237035923,
      "flos": 43508793461760.0,
      "grad_norm": 1.4777038696822193,
      "language_loss": 0.70577073,
      "learning_rate": 1.105735316926046e-06,
      "loss": 0.72690189,
      "num_input_tokens_seen": 236281890,
      "step": 10946,
      "time_per_iteration": 2.727125406265259
    },
    {
      "auxiliary_loss_clip": 0.01098843,
      "auxiliary_loss_mlp": 0.01032271,
      "balance_loss_clip": 1.03856599,
      "balance_loss_mlp": 1.01964712,
      "epoch": 0.6581692469562603,
      "flos": 22415404763520.0,
      "grad_norm": 2.0722928857646834,
      "language_loss": 0.81608349,
      "learning_rate": 1.105386972944934e-06,
      "loss": 0.8373946,
      "num_input_tokens_seen": 236298370,
      "step": 10947,
      "time_per_iteration": 2.5034046173095703
    },
    {
      "auxiliary_loss_clip": 0.01054511,
      "auxiliary_loss_mlp": 0.00786713,
      "balance_loss_clip": 1.03518009,
      "balance_loss_mlp": 1.00965214,
      "epoch": 0.6582293702089284,
      "flos": 24859772167680.0,
      "grad_norm": 1.5638099324036803,
      "language_loss": 0.77447677,
      "learning_rate": 1.1050386628876385e-06,
      "loss": 0.792889,
      "num_input_tokens_seen": 236317380,
      "step": 10948,
      "time_per_iteration": 2.620025157928467
    },
    {
      "auxiliary_loss_clip": 0.01099287,
      "auxiliary_loss_mlp": 0.01028059,
      "balance_loss_clip": 1.03937781,
      "balance_loss_mlp": 1.0159955,
      "epoch": 0.6582894934615963,
      "flos": 23039676161280.0,
      "grad_norm": 1.634848404571932,
      "language_loss": 0.78899598,
      "learning_rate": 1.1046903867673655e-06,
      "loss": 0.81026942,
      "num_input_tokens_seen": 236336210,
      "step": 10949,
      "time_per_iteration": 2.5271012783050537
    },
    {
      "auxiliary_loss_clip": 0.01032543,
      "auxiliary_loss_mlp": 0.01001961,
      "balance_loss_clip": 1.01848471,
      "balance_loss_mlp": 1.00080478,
      "epoch": 0.6583496167142643,
      "flos": 72551980978560.0,
      "grad_norm": 0.7377619949118958,
      "language_loss": 0.61851352,
      "learning_rate": 1.104342144597323e-06,
      "loss": 0.63885856,
      "num_input_tokens_seen": 236403090,
      "step": 10950,
      "time_per_iteration": 3.187147617340088
    },
    {
      "auxiliary_loss_clip": 0.01096108,
      "auxiliary_loss_mlp": 0.01034052,
      "balance_loss_clip": 1.03713667,
      "balance_loss_mlp": 1.02284098,
      "epoch": 0.6584097399669322,
      "flos": 13078846592640.0,
      "grad_norm": 2.149396029143752,
      "language_loss": 0.66765654,
      "learning_rate": 1.1039939363907178e-06,
      "loss": 0.68895811,
      "num_input_tokens_seen": 236420475,
      "step": 10951,
      "time_per_iteration": 2.4940176010131836
    },
    {
      "auxiliary_loss_clip": 0.01096812,
      "auxiliary_loss_mlp": 0.01031196,
      "balance_loss_clip": 1.03847194,
      "balance_loss_mlp": 1.01919723,
      "epoch": 0.6584698632196002,
      "flos": 28693164458880.0,
      "grad_norm": 1.3849748252602445,
      "language_loss": 0.76362801,
      "learning_rate": 1.1036457621607504e-06,
      "loss": 0.78490806,
      "num_input_tokens_seen": 236441915,
      "step": 10952,
      "time_per_iteration": 2.5679867267608643
    },
    {
      "auxiliary_loss_clip": 0.01107508,
      "auxiliary_loss_mlp": 0.01028151,
      "balance_loss_clip": 1.03843665,
      "balance_loss_mlp": 1.01596212,
      "epoch": 0.6585299864722681,
      "flos": 14319272914560.0,
      "grad_norm": 1.7050995072644481,
      "language_loss": 0.73508406,
      "learning_rate": 1.1032976219206257e-06,
      "loss": 0.75644064,
      "num_input_tokens_seen": 236460340,
      "step": 10953,
      "time_per_iteration": 2.4455037117004395
    },
    {
      "auxiliary_loss_clip": 0.01077469,
      "auxiliary_loss_mlp": 0.01036487,
      "balance_loss_clip": 1.03841233,
      "balance_loss_mlp": 1.02426815,
      "epoch": 0.6585901097249361,
      "flos": 26797907243520.0,
      "grad_norm": 2.0280863441706964,
      "language_loss": 0.78416699,
      "learning_rate": 1.102949515683546e-06,
      "loss": 0.80530655,
      "num_input_tokens_seen": 236478280,
      "step": 10954,
      "time_per_iteration": 2.594600200653076
    },
    {
      "auxiliary_loss_clip": 0.01086415,
      "auxiliary_loss_mlp": 0.01037053,
      "balance_loss_clip": 1.03639007,
      "balance_loss_mlp": 1.02402949,
      "epoch": 0.658650232977604,
      "flos": 18733124989440.0,
      "grad_norm": 2.2078788133817437,
      "language_loss": 0.6941635,
      "learning_rate": 1.1026014434627096e-06,
      "loss": 0.71539819,
      "num_input_tokens_seen": 236493225,
      "step": 10955,
      "time_per_iteration": 2.506420135498047
    },
    {
      "auxiliary_loss_clip": 0.01078707,
      "auxiliary_loss_mlp": 0.01032871,
      "balance_loss_clip": 1.03514743,
      "balance_loss_mlp": 1.02122378,
      "epoch": 0.6587103562302721,
      "flos": 24753440931840.0,
      "grad_norm": 2.0094008943933144,
      "language_loss": 0.80655712,
      "learning_rate": 1.1022534052713172e-06,
      "loss": 0.8276729,
      "num_input_tokens_seen": 236514420,
      "step": 10956,
      "time_per_iteration": 2.539672374725342
    },
    {
      "auxiliary_loss_clip": 0.01095897,
      "auxiliary_loss_mlp": 0.01035682,
      "balance_loss_clip": 1.03782034,
      "balance_loss_mlp": 1.02274764,
      "epoch": 0.65877047948294,
      "flos": 22346133384960.0,
      "grad_norm": 3.2955298818046432,
      "language_loss": 0.81297946,
      "learning_rate": 1.1019054011225648e-06,
      "loss": 0.83429521,
      "num_input_tokens_seen": 236532785,
      "step": 10957,
      "time_per_iteration": 2.5078375339508057
    },
    {
      "auxiliary_loss_clip": 0.01085278,
      "auxiliary_loss_mlp": 0.01032202,
      "balance_loss_clip": 1.03803444,
      "balance_loss_mlp": 1.02090049,
      "epoch": 0.658830602735608,
      "flos": 45180542298240.0,
      "grad_norm": 1.700017406455416,
      "language_loss": 0.75681484,
      "learning_rate": 1.1015574310296506e-06,
      "loss": 0.77798969,
      "num_input_tokens_seen": 236553330,
      "step": 10958,
      "time_per_iteration": 2.7244529724121094
    },
    {
      "auxiliary_loss_clip": 0.01066663,
      "auxiliary_loss_mlp": 0.01041742,
      "balance_loss_clip": 1.03698802,
      "balance_loss_mlp": 1.02769351,
      "epoch": 0.6588907259882759,
      "flos": 19901622326400.0,
      "grad_norm": 1.6301790002542313,
      "language_loss": 0.75097644,
      "learning_rate": 1.1012094950057678e-06,
      "loss": 0.77206051,
      "num_input_tokens_seen": 236572960,
      "step": 10959,
      "time_per_iteration": 2.553173303604126
    },
    {
      "auxiliary_loss_clip": 0.01099147,
      "auxiliary_loss_mlp": 0.0102815,
      "balance_loss_clip": 1.03878355,
      "balance_loss_mlp": 1.01620483,
      "epoch": 0.6589508492409439,
      "flos": 24133766474880.0,
      "grad_norm": 1.4609028550256833,
      "language_loss": 0.65063661,
      "learning_rate": 1.1008615930641107e-06,
      "loss": 0.67190957,
      "num_input_tokens_seen": 236594090,
      "step": 10960,
      "time_per_iteration": 2.518650531768799
    },
    {
      "auxiliary_loss_clip": 0.011138,
      "auxiliary_loss_mlp": 0.01034029,
      "balance_loss_clip": 1.03985524,
      "balance_loss_mlp": 1.02090454,
      "epoch": 0.659010972493612,
      "flos": 18222906251520.0,
      "grad_norm": 2.0655555520332776,
      "language_loss": 0.81823611,
      "learning_rate": 1.1005137252178734e-06,
      "loss": 0.83971441,
      "num_input_tokens_seen": 236610190,
      "step": 10961,
      "time_per_iteration": 2.4724037647247314
    },
    {
      "auxiliary_loss_clip": 0.0107201,
      "auxiliary_loss_mlp": 0.01028116,
      "balance_loss_clip": 1.0380106,
      "balance_loss_mlp": 1.01555717,
      "epoch": 0.6590710957462799,
      "flos": 27600007898880.0,
      "grad_norm": 1.646902415267453,
      "language_loss": 0.73463553,
      "learning_rate": 1.1001658914802453e-06,
      "loss": 0.75563681,
      "num_input_tokens_seen": 236631575,
      "step": 10962,
      "time_per_iteration": 3.992591142654419
    },
    {
      "auxiliary_loss_clip": 0.01092617,
      "auxiliary_loss_mlp": 0.01033288,
      "balance_loss_clip": 1.03636169,
      "balance_loss_mlp": 1.02105141,
      "epoch": 0.6591312189989479,
      "flos": 20302959962880.0,
      "grad_norm": 1.9481164769281438,
      "language_loss": 0.79391396,
      "learning_rate": 1.0998180918644165e-06,
      "loss": 0.81517303,
      "num_input_tokens_seen": 236649815,
      "step": 10963,
      "time_per_iteration": 2.4816064834594727
    },
    {
      "auxiliary_loss_clip": 0.0106407,
      "auxiliary_loss_mlp": 0.00783774,
      "balance_loss_clip": 1.03766823,
      "balance_loss_mlp": 1.00918508,
      "epoch": 0.6591913422516158,
      "flos": 12312943868160.0,
      "grad_norm": 1.703441493194395,
      "language_loss": 0.78184438,
      "learning_rate": 1.0994703263835754e-06,
      "loss": 0.80032277,
      "num_input_tokens_seen": 236668335,
      "step": 10964,
      "time_per_iteration": 2.5914885997772217
    },
    {
      "auxiliary_loss_clip": 0.010701,
      "auxiliary_loss_mlp": 0.01036323,
      "balance_loss_clip": 1.03348804,
      "balance_loss_mlp": 1.02393115,
      "epoch": 0.6592514655042838,
      "flos": 25884591102720.0,
      "grad_norm": 1.632598999353539,
      "language_loss": 0.74086773,
      "learning_rate": 1.0991225950509106e-06,
      "loss": 0.7619319,
      "num_input_tokens_seen": 236688945,
      "step": 10965,
      "time_per_iteration": 2.5860815048217773
    },
    {
      "auxiliary_loss_clip": 0.01077072,
      "auxiliary_loss_mlp": 0.01039815,
      "balance_loss_clip": 1.03657007,
      "balance_loss_mlp": 1.0260818,
      "epoch": 0.6593115887569517,
      "flos": 14063624841600.0,
      "grad_norm": 2.0081459045722636,
      "language_loss": 0.7359612,
      "learning_rate": 1.0987748978796067e-06,
      "loss": 0.75713003,
      "num_input_tokens_seen": 236707055,
      "step": 10966,
      "time_per_iteration": 2.5395796298980713
    },
    {
      "auxiliary_loss_clip": 0.01098246,
      "auxiliary_loss_mlp": 0.01030968,
      "balance_loss_clip": 1.03718257,
      "balance_loss_mlp": 1.018296,
      "epoch": 0.6593717120096197,
      "flos": 24717925359360.0,
      "grad_norm": 1.5918891907391046,
      "language_loss": 0.76943457,
      "learning_rate": 1.0984272348828487e-06,
      "loss": 0.79072672,
      "num_input_tokens_seen": 236725900,
      "step": 10967,
      "time_per_iteration": 2.531558036804199
    },
    {
      "auxiliary_loss_clip": 0.01034609,
      "auxiliary_loss_mlp": 0.01002903,
      "balance_loss_clip": 1.02028227,
      "balance_loss_mlp": 1.0016278,
      "epoch": 0.6594318352622877,
      "flos": 55558083502080.0,
      "grad_norm": 0.6972305191951089,
      "language_loss": 0.48431277,
      "learning_rate": 1.0980796060738221e-06,
      "loss": 0.50468791,
      "num_input_tokens_seen": 236788415,
      "step": 10968,
      "time_per_iteration": 3.092557191848755
    },
    {
      "auxiliary_loss_clip": 0.01062771,
      "auxiliary_loss_mlp": 0.01039081,
      "balance_loss_clip": 1.03428519,
      "balance_loss_mlp": 1.02474022,
      "epoch": 0.6594919585149557,
      "flos": 17456931699840.0,
      "grad_norm": 1.79877549657663,
      "language_loss": 0.7931813,
      "learning_rate": 1.0977320114657058e-06,
      "loss": 0.81419981,
      "num_input_tokens_seen": 236805155,
      "step": 10969,
      "time_per_iteration": 2.597667694091797
    },
    {
      "auxiliary_loss_clip": 0.01098388,
      "auxiliary_loss_mlp": 0.01030483,
      "balance_loss_clip": 1.03732538,
      "balance_loss_mlp": 1.01818681,
      "epoch": 0.6595520817676236,
      "flos": 18223229473920.0,
      "grad_norm": 1.9422096989428075,
      "language_loss": 0.65215999,
      "learning_rate": 1.0973844510716817e-06,
      "loss": 0.67344868,
      "num_input_tokens_seen": 236824360,
      "step": 10970,
      "time_per_iteration": 3.9430458545684814
    },
    {
      "auxiliary_loss_clip": 0.01094226,
      "auxiliary_loss_mlp": 0.01028264,
      "balance_loss_clip": 1.03527927,
      "balance_loss_mlp": 1.01595521,
      "epoch": 0.6596122050202916,
      "flos": 22199761463040.0,
      "grad_norm": 1.5786732364178528,
      "language_loss": 0.76502132,
      "learning_rate": 1.0970369249049308e-06,
      "loss": 0.7862463,
      "num_input_tokens_seen": 236844640,
      "step": 10971,
      "time_per_iteration": 2.514813184738159
    },
    {
      "auxiliary_loss_clip": 0.01043378,
      "auxiliary_loss_mlp": 0.01044852,
      "balance_loss_clip": 1.0335182,
      "balance_loss_mlp": 1.03017163,
      "epoch": 0.6596723282729595,
      "flos": 14173834746240.0,
      "grad_norm": 3.5379035853757523,
      "language_loss": 0.70206261,
      "learning_rate": 1.096689432978629e-06,
      "loss": 0.72294486,
      "num_input_tokens_seen": 236861160,
      "step": 10972,
      "time_per_iteration": 2.5806121826171875
    },
    {
      "auxiliary_loss_clip": 0.01096231,
      "auxiliary_loss_mlp": 0.01025676,
      "balance_loss_clip": 1.03761888,
      "balance_loss_mlp": 1.01261687,
      "epoch": 0.6597324515256275,
      "flos": 30553193410560.0,
      "grad_norm": 1.594671924616411,
      "language_loss": 0.55900061,
      "learning_rate": 1.0963419753059556e-06,
      "loss": 0.58021963,
      "num_input_tokens_seen": 236880465,
      "step": 10973,
      "time_per_iteration": 4.008740186691284
    },
    {
      "auxiliary_loss_clip": 0.01091342,
      "auxiliary_loss_mlp": 0.01042363,
      "balance_loss_clip": 1.03891706,
      "balance_loss_mlp": 1.03013206,
      "epoch": 0.6597925747782956,
      "flos": 17639860688640.0,
      "grad_norm": 1.892908870824964,
      "language_loss": 0.78921151,
      "learning_rate": 1.0959945519000839e-06,
      "loss": 0.81054848,
      "num_input_tokens_seen": 236897730,
      "step": 10974,
      "time_per_iteration": 2.502997398376465
    },
    {
      "auxiliary_loss_clip": 0.01094204,
      "auxiliary_loss_mlp": 0.01036404,
      "balance_loss_clip": 1.03799474,
      "balance_loss_mlp": 1.0244056,
      "epoch": 0.6598526980309635,
      "flos": 22819112697600.0,
      "grad_norm": 2.1299921944402223,
      "language_loss": 0.68948513,
      "learning_rate": 1.0956471627741906e-06,
      "loss": 0.71079123,
      "num_input_tokens_seen": 236917300,
      "step": 10975,
      "time_per_iteration": 2.5102145671844482
    },
    {
      "auxiliary_loss_clip": 0.01098361,
      "auxiliary_loss_mlp": 0.01027822,
      "balance_loss_clip": 1.03691709,
      "balance_loss_mlp": 1.01613367,
      "epoch": 0.6599128212836315,
      "flos": 21068036674560.0,
      "grad_norm": 1.611104817529319,
      "language_loss": 0.70939487,
      "learning_rate": 1.0952998079414464e-06,
      "loss": 0.73065674,
      "num_input_tokens_seen": 236935590,
      "step": 10976,
      "time_per_iteration": 2.5006821155548096
    },
    {
      "auxiliary_loss_clip": 0.01078949,
      "auxiliary_loss_mlp": 0.01034011,
      "balance_loss_clip": 1.03654623,
      "balance_loss_mlp": 1.02143455,
      "epoch": 0.6599729445362994,
      "flos": 22163527618560.0,
      "grad_norm": 1.9683526658730834,
      "language_loss": 0.67455268,
      "learning_rate": 1.0949524874150243e-06,
      "loss": 0.69568229,
      "num_input_tokens_seen": 236952830,
      "step": 10977,
      "time_per_iteration": 2.516997814178467
    },
    {
      "auxiliary_loss_clip": 0.01075723,
      "auxiliary_loss_mlp": 0.01033768,
      "balance_loss_clip": 1.03831661,
      "balance_loss_mlp": 1.01985645,
      "epoch": 0.6600330677889674,
      "flos": 18150079426560.0,
      "grad_norm": 2.5127972780395926,
      "language_loss": 0.81452751,
      "learning_rate": 1.0946052012080952e-06,
      "loss": 0.83562243,
      "num_input_tokens_seen": 236971930,
      "step": 10978,
      "time_per_iteration": 2.5371835231781006
    },
    {
      "auxiliary_loss_clip": 0.01080499,
      "auxiliary_loss_mlp": 0.01036969,
      "balance_loss_clip": 1.03906441,
      "balance_loss_mlp": 1.02377236,
      "epoch": 0.6600931910416353,
      "flos": 18150115340160.0,
      "grad_norm": 2.1973208368420263,
      "language_loss": 0.67483437,
      "learning_rate": 1.0942579493338278e-06,
      "loss": 0.69600898,
      "num_input_tokens_seen": 236989920,
      "step": 10979,
      "time_per_iteration": 2.510268449783325
    },
    {
      "auxiliary_loss_clip": 0.01078525,
      "auxiliary_loss_mlp": 0.01030196,
      "balance_loss_clip": 1.03651333,
      "balance_loss_mlp": 1.01692748,
      "epoch": 0.6601533142943034,
      "flos": 17420733768960.0,
      "grad_norm": 2.309782370268834,
      "language_loss": 0.73387527,
      "learning_rate": 1.0939107318053889e-06,
      "loss": 0.75496256,
      "num_input_tokens_seen": 237006570,
      "step": 10980,
      "time_per_iteration": 2.5269112586975098
    },
    {
      "auxiliary_loss_clip": 0.01068587,
      "auxiliary_loss_mlp": 0.01029493,
      "balance_loss_clip": 1.03577161,
      "balance_loss_mlp": 1.01847219,
      "epoch": 0.6602134375469713,
      "flos": 28219574615040.0,
      "grad_norm": 1.539564115176228,
      "language_loss": 0.72533095,
      "learning_rate": 1.0935635486359459e-06,
      "loss": 0.74631172,
      "num_input_tokens_seen": 237028415,
      "step": 10981,
      "time_per_iteration": 2.6143224239349365
    },
    {
      "auxiliary_loss_clip": 0.01066544,
      "auxiliary_loss_mlp": 0.01034617,
      "balance_loss_clip": 1.04045057,
      "balance_loss_mlp": 1.02226067,
      "epoch": 0.6602735607996393,
      "flos": 29418056830080.0,
      "grad_norm": 2.029004689529742,
      "language_loss": 0.68557507,
      "learning_rate": 1.0932163998386647e-06,
      "loss": 0.70658672,
      "num_input_tokens_seen": 237046595,
      "step": 10982,
      "time_per_iteration": 2.671036720275879
    },
    {
      "auxiliary_loss_clip": 0.01099073,
      "auxiliary_loss_mlp": 0.01032544,
      "balance_loss_clip": 1.039433,
      "balance_loss_mlp": 1.02007449,
      "epoch": 0.6603336840523072,
      "flos": 18588045957120.0,
      "grad_norm": 1.535531817469894,
      "language_loss": 0.69561803,
      "learning_rate": 1.0928692854267075e-06,
      "loss": 0.71693426,
      "num_input_tokens_seen": 237066150,
      "step": 10983,
      "time_per_iteration": 2.4844815731048584
    },
    {
      "auxiliary_loss_clip": 0.01099264,
      "auxiliary_loss_mlp": 0.0103055,
      "balance_loss_clip": 1.03642571,
      "balance_loss_mlp": 1.0176934,
      "epoch": 0.6603938073049752,
      "flos": 33254860913280.0,
      "grad_norm": 1.7446803785531146,
      "language_loss": 0.70136803,
      "learning_rate": 1.092522205413239e-06,
      "loss": 0.72266614,
      "num_input_tokens_seen": 237087060,
      "step": 10984,
      "time_per_iteration": 2.595243215560913
    },
    {
      "auxiliary_loss_clip": 0.01074969,
      "auxiliary_loss_mlp": 0.01033822,
      "balance_loss_clip": 1.03625524,
      "balance_loss_mlp": 1.02173424,
      "epoch": 0.6604539305576431,
      "flos": 17384284442880.0,
      "grad_norm": 1.6302502547189304,
      "language_loss": 0.8385613,
      "learning_rate": 1.0921751598114193e-06,
      "loss": 0.85964918,
      "num_input_tokens_seen": 237103825,
      "step": 10985,
      "time_per_iteration": 2.531421184539795
    },
    {
      "auxiliary_loss_clip": 0.01099241,
      "auxiliary_loss_mlp": 0.01032226,
      "balance_loss_clip": 1.03824401,
      "balance_loss_mlp": 1.01939893,
      "epoch": 0.6605140538103111,
      "flos": 21251145231360.0,
      "grad_norm": 2.035015959733069,
      "language_loss": 0.74134773,
      "learning_rate": 1.0918281486344077e-06,
      "loss": 0.76266241,
      "num_input_tokens_seen": 237121740,
      "step": 10986,
      "time_per_iteration": 2.5107126235961914
    },
    {
      "auxiliary_loss_clip": 0.01098411,
      "auxiliary_loss_mlp": 0.01031647,
      "balance_loss_clip": 1.03912783,
      "balance_loss_mlp": 1.01910043,
      "epoch": 0.6605741770629792,
      "flos": 13881701433600.0,
      "grad_norm": 1.8256307821213793,
      "language_loss": 0.79115045,
      "learning_rate": 1.0914811718953636e-06,
      "loss": 0.812451,
      "num_input_tokens_seen": 237139565,
      "step": 10987,
      "time_per_iteration": 2.4756245613098145
    },
    {
      "auxiliary_loss_clip": 0.01017669,
      "auxiliary_loss_mlp": 0.01002607,
      "balance_loss_clip": 1.01724267,
      "balance_loss_mlp": 1.00105715,
      "epoch": 0.6606343003156471,
      "flos": 69316215171840.0,
      "grad_norm": 0.8249888511591914,
      "language_loss": 0.54088306,
      "learning_rate": 1.0911342296074454e-06,
      "loss": 0.56108594,
      "num_input_tokens_seen": 237201055,
      "step": 10988,
      "time_per_iteration": 3.2313458919525146
    },
    {
      "auxiliary_loss_clip": 0.01050108,
      "auxiliary_loss_mlp": 0.0103833,
      "balance_loss_clip": 1.03544879,
      "balance_loss_mlp": 1.0265702,
      "epoch": 0.6606944235683151,
      "flos": 27272394927360.0,
      "grad_norm": 1.58414147408499,
      "language_loss": 0.77414286,
      "learning_rate": 1.0907873217838077e-06,
      "loss": 0.79502726,
      "num_input_tokens_seen": 237221805,
      "step": 10989,
      "time_per_iteration": 2.672549247741699
    },
    {
      "auxiliary_loss_clip": 0.01086907,
      "auxiliary_loss_mlp": 0.01035778,
      "balance_loss_clip": 1.03882933,
      "balance_loss_mlp": 1.0236845,
      "epoch": 0.660754546820983,
      "flos": 13772820332160.0,
      "grad_norm": 3.8991220155516277,
      "language_loss": 0.77416521,
      "learning_rate": 1.0904404484376064e-06,
      "loss": 0.7953921,
      "num_input_tokens_seen": 237238270,
      "step": 10990,
      "time_per_iteration": 2.5019192695617676
    },
    {
      "auxiliary_loss_clip": 0.01111253,
      "auxiliary_loss_mlp": 0.01031841,
      "balance_loss_clip": 1.03877926,
      "balance_loss_mlp": 1.01921678,
      "epoch": 0.660814670073651,
      "flos": 15705209232000.0,
      "grad_norm": 1.890721713600211,
      "language_loss": 0.60847914,
      "learning_rate": 1.0900936095819937e-06,
      "loss": 0.62991005,
      "num_input_tokens_seen": 237255400,
      "step": 10991,
      "time_per_iteration": 2.426070213317871
    },
    {
      "auxiliary_loss_clip": 0.01085545,
      "auxiliary_loss_mlp": 0.01039386,
      "balance_loss_clip": 1.03754497,
      "balance_loss_mlp": 1.02638626,
      "epoch": 0.6608747933263189,
      "flos": 20850023076480.0,
      "grad_norm": 2.351906829554656,
      "language_loss": 0.68451601,
      "learning_rate": 1.0897468052301234e-06,
      "loss": 0.70576537,
      "num_input_tokens_seen": 237273105,
      "step": 10992,
      "time_per_iteration": 2.526390552520752
    },
    {
      "auxiliary_loss_clip": 0.01097595,
      "auxiliary_loss_mlp": 0.01031316,
      "balance_loss_clip": 1.03999448,
      "balance_loss_mlp": 1.01879311,
      "epoch": 0.660934916578987,
      "flos": 20632117219200.0,
      "grad_norm": 1.692606113014454,
      "language_loss": 0.87810957,
      "learning_rate": 1.0894000353951444e-06,
      "loss": 0.89939868,
      "num_input_tokens_seen": 237292650,
      "step": 10993,
      "time_per_iteration": 2.509467840194702
    },
    {
      "auxiliary_loss_clip": 0.01106376,
      "auxiliary_loss_mlp": 0.01033804,
      "balance_loss_clip": 1.03956771,
      "balance_loss_mlp": 1.01949263,
      "epoch": 0.6609950398316549,
      "flos": 25113588647040.0,
      "grad_norm": 1.586754515859154,
      "language_loss": 0.6688354,
      "learning_rate": 1.0890533000902078e-06,
      "loss": 0.69023716,
      "num_input_tokens_seen": 237312865,
      "step": 10994,
      "time_per_iteration": 2.5339748859405518
    },
    {
      "auxiliary_loss_clip": 0.01075443,
      "auxiliary_loss_mlp": 0.01036637,
      "balance_loss_clip": 1.03810287,
      "balance_loss_mlp": 1.02291608,
      "epoch": 0.6610551630843229,
      "flos": 18661196004480.0,
      "grad_norm": 1.6643150612945448,
      "language_loss": 0.76312196,
      "learning_rate": 1.0887065993284626e-06,
      "loss": 0.78424281,
      "num_input_tokens_seen": 237331210,
      "step": 10995,
      "time_per_iteration": 2.6006290912628174
    },
    {
      "auxiliary_loss_clip": 0.01088144,
      "auxiliary_loss_mlp": 0.01028232,
      "balance_loss_clip": 1.03749371,
      "balance_loss_mlp": 1.01662111,
      "epoch": 0.6611152863369908,
      "flos": 23258192549760.0,
      "grad_norm": 1.8493694949902377,
      "language_loss": 0.74405807,
      "learning_rate": 1.088359933123053e-06,
      "loss": 0.76522183,
      "num_input_tokens_seen": 237349455,
      "step": 10996,
      "time_per_iteration": 2.5358595848083496
    },
    {
      "auxiliary_loss_clip": 0.01110505,
      "auxiliary_loss_mlp": 0.01035082,
      "balance_loss_clip": 1.04009736,
      "balance_loss_mlp": 1.02275598,
      "epoch": 0.6611754095896588,
      "flos": 22159720776960.0,
      "grad_norm": 1.6543055538088762,
      "language_loss": 0.68897629,
      "learning_rate": 1.088013301487126e-06,
      "loss": 0.71043217,
      "num_input_tokens_seen": 237367100,
      "step": 10997,
      "time_per_iteration": 2.463078498840332
    },
    {
      "auxiliary_loss_clip": 0.01089983,
      "auxiliary_loss_mlp": 0.0103382,
      "balance_loss_clip": 1.03695679,
      "balance_loss_mlp": 1.02201223,
      "epoch": 0.6612355328423267,
      "flos": 13991228979840.0,
      "grad_norm": 2.4279057244130757,
      "language_loss": 0.68274689,
      "learning_rate": 1.0876667044338269e-06,
      "loss": 0.70398498,
      "num_input_tokens_seen": 237384840,
      "step": 10998,
      "time_per_iteration": 2.513188600540161
    },
    {
      "auxiliary_loss_clip": 0.01033112,
      "auxiliary_loss_mlp": 0.01001796,
      "balance_loss_clip": 1.01902902,
      "balance_loss_mlp": 1.00069368,
      "epoch": 0.6612956560949947,
      "flos": 61453716359040.0,
      "grad_norm": 0.6504061895015423,
      "language_loss": 0.51181149,
      "learning_rate": 1.087320141976297e-06,
      "loss": 0.53216058,
      "num_input_tokens_seen": 237443355,
      "step": 10999,
      "time_per_iteration": 3.0710206031799316
    },
    {
      "auxiliary_loss_clip": 0.01112354,
      "auxiliary_loss_mlp": 0.00784465,
      "balance_loss_clip": 1.03849626,
      "balance_loss_mlp": 1.00969279,
      "epoch": 0.6613557793476627,
      "flos": 21616644072960.0,
      "grad_norm": 2.3532159853796895,
      "language_loss": 0.70549917,
      "learning_rate": 1.086973614127679e-06,
      "loss": 0.72446728,
      "num_input_tokens_seen": 237459205,
      "step": 11000,
      "time_per_iteration": 2.4705307483673096
    },
    {
      "auxiliary_loss_clip": 0.01084651,
      "auxiliary_loss_mlp": 0.01036257,
      "balance_loss_clip": 1.03694868,
      "balance_loss_mlp": 1.02486074,
      "epoch": 0.6614159026003307,
      "flos": 34020117192960.0,
      "grad_norm": 1.4885447331513868,
      "language_loss": 0.65230799,
      "learning_rate": 1.0866271209011133e-06,
      "loss": 0.67351711,
      "num_input_tokens_seen": 237483580,
      "step": 11001,
      "time_per_iteration": 4.037245035171509
    },
    {
      "auxiliary_loss_clip": 0.01107428,
      "auxiliary_loss_mlp": 0.01030343,
      "balance_loss_clip": 1.03702545,
      "balance_loss_mlp": 1.01818371,
      "epoch": 0.6614760258529987,
      "flos": 24097281235200.0,
      "grad_norm": 1.8165935099746786,
      "language_loss": 0.72964859,
      "learning_rate": 1.086280662309739e-06,
      "loss": 0.75102627,
      "num_input_tokens_seen": 237502860,
      "step": 11002,
      "time_per_iteration": 2.4819676876068115
    },
    {
      "auxiliary_loss_clip": 0.01092983,
      "auxiliary_loss_mlp": 0.01034802,
      "balance_loss_clip": 1.0361222,
      "balance_loss_mlp": 1.02177238,
      "epoch": 0.6615361491056666,
      "flos": 14903790935040.0,
      "grad_norm": 2.25590901041263,
      "language_loss": 0.78587663,
      "learning_rate": 1.0859342383666928e-06,
      "loss": 0.80715454,
      "num_input_tokens_seen": 237521030,
      "step": 11003,
      "time_per_iteration": 2.5022425651550293
    },
    {
      "auxiliary_loss_clip": 0.01099959,
      "auxiliary_loss_mlp": 0.01035469,
      "balance_loss_clip": 1.03900003,
      "balance_loss_mlp": 1.0221833,
      "epoch": 0.6615962723583346,
      "flos": 15304877176320.0,
      "grad_norm": 1.859887957516743,
      "language_loss": 0.68861866,
      "learning_rate": 1.0855878490851119e-06,
      "loss": 0.70997298,
      "num_input_tokens_seen": 237539585,
      "step": 11004,
      "time_per_iteration": 2.502073287963867
    },
    {
      "auxiliary_loss_clip": 0.01100187,
      "auxiliary_loss_mlp": 0.01037841,
      "balance_loss_clip": 1.03811502,
      "balance_loss_mlp": 1.02441776,
      "epoch": 0.6616563956110025,
      "flos": 18732586285440.0,
      "grad_norm": 2.132041449999643,
      "language_loss": 0.69262278,
      "learning_rate": 1.085241494478132e-06,
      "loss": 0.71400309,
      "num_input_tokens_seen": 237557655,
      "step": 11005,
      "time_per_iteration": 2.4770278930664062
    },
    {
      "auxiliary_loss_clip": 0.01086684,
      "auxiliary_loss_mlp": 0.01028522,
      "balance_loss_clip": 1.03623819,
      "balance_loss_mlp": 1.01651168,
      "epoch": 0.6617165188636706,
      "flos": 24495063425280.0,
      "grad_norm": 1.620151244122595,
      "language_loss": 0.78177696,
      "learning_rate": 1.0848951745588855e-06,
      "loss": 0.80292898,
      "num_input_tokens_seen": 237577000,
      "step": 11006,
      "time_per_iteration": 2.5543947219848633
    },
    {
      "auxiliary_loss_clip": 0.01096718,
      "auxiliary_loss_mlp": 0.01031442,
      "balance_loss_clip": 1.03731656,
      "balance_loss_mlp": 1.01916933,
      "epoch": 0.6617766421163385,
      "flos": 22379673709440.0,
      "grad_norm": 1.436141257845966,
      "language_loss": 0.76022017,
      "learning_rate": 1.0845488893405068e-06,
      "loss": 0.78150177,
      "num_input_tokens_seen": 237597960,
      "step": 11007,
      "time_per_iteration": 2.5301594734191895
    },
    {
      "auxiliary_loss_clip": 0.01095554,
      "auxiliary_loss_mlp": 0.01027678,
      "balance_loss_clip": 1.03835046,
      "balance_loss_mlp": 1.01569152,
      "epoch": 0.6618367653690065,
      "flos": 20850418126080.0,
      "grad_norm": 1.6843715653485822,
      "language_loss": 0.78426909,
      "learning_rate": 1.0842026388361248e-06,
      "loss": 0.8055014,
      "num_input_tokens_seen": 237616385,
      "step": 11008,
      "time_per_iteration": 2.5143046379089355
    },
    {
      "auxiliary_loss_clip": 0.01112487,
      "auxiliary_loss_mlp": 0.01032887,
      "balance_loss_clip": 1.0380621,
      "balance_loss_mlp": 1.01972055,
      "epoch": 0.6618968886216744,
      "flos": 17712328377600.0,
      "grad_norm": 2.35417988748547,
      "language_loss": 0.81828797,
      "learning_rate": 1.0838564230588715e-06,
      "loss": 0.83974171,
      "num_input_tokens_seen": 237634930,
      "step": 11009,
      "time_per_iteration": 3.8380331993103027
    },
    {
      "auxiliary_loss_clip": 0.01024452,
      "auxiliary_loss_mlp": 0.01005524,
      "balance_loss_clip": 1.02895141,
      "balance_loss_mlp": 1.00392616,
      "epoch": 0.6619570118743424,
      "flos": 67035347498880.0,
      "grad_norm": 0.9750546269005811,
      "language_loss": 0.67333871,
      "learning_rate": 1.0835102420218735e-06,
      "loss": 0.6936385,
      "num_input_tokens_seen": 237693175,
      "step": 11010,
      "time_per_iteration": 3.09132719039917
    },
    {
      "auxiliary_loss_clip": 0.01097992,
      "auxiliary_loss_mlp": 0.01033668,
      "balance_loss_clip": 1.03670752,
      "balance_loss_mlp": 1.02033472,
      "epoch": 0.6620171351270103,
      "flos": 18660908695680.0,
      "grad_norm": 1.545539040673936,
      "language_loss": 0.71293044,
      "learning_rate": 1.0831640957382593e-06,
      "loss": 0.73424703,
      "num_input_tokens_seen": 237713160,
      "step": 11011,
      "time_per_iteration": 3.877072334289551
    },
    {
      "auxiliary_loss_clip": 0.01099008,
      "auxiliary_loss_mlp": 0.0103112,
      "balance_loss_clip": 1.04056048,
      "balance_loss_mlp": 1.01999807,
      "epoch": 0.6620772583796783,
      "flos": 24170503109760.0,
      "grad_norm": 1.5140888882876735,
      "language_loss": 0.7239567,
      "learning_rate": 1.0828179842211557e-06,
      "loss": 0.74525791,
      "num_input_tokens_seen": 237733600,
      "step": 11012,
      "time_per_iteration": 2.5689635276794434
    },
    {
      "auxiliary_loss_clip": 0.01091247,
      "auxiliary_loss_mlp": 0.01030157,
      "balance_loss_clip": 1.03635955,
      "balance_loss_mlp": 1.01941633,
      "epoch": 0.6621373816323463,
      "flos": 23623547736960.0,
      "grad_norm": 1.5496519389432861,
      "language_loss": 0.79415476,
      "learning_rate": 1.0824719074836845e-06,
      "loss": 0.81536877,
      "num_input_tokens_seen": 237752135,
      "step": 11013,
      "time_per_iteration": 2.5208241939544678
    },
    {
      "auxiliary_loss_clip": 0.01085436,
      "auxiliary_loss_mlp": 0.01030174,
      "balance_loss_clip": 1.03705502,
      "balance_loss_mlp": 1.01847315,
      "epoch": 0.6621975048850143,
      "flos": 18442212739200.0,
      "grad_norm": 2.580843846316289,
      "language_loss": 0.70719492,
      "learning_rate": 1.082125865538971e-06,
      "loss": 0.72835106,
      "num_input_tokens_seen": 237770735,
      "step": 11014,
      "time_per_iteration": 2.5338551998138428
    },
    {
      "auxiliary_loss_clip": 0.01078431,
      "auxiliary_loss_mlp": 0.00781704,
      "balance_loss_clip": 1.04077458,
      "balance_loss_mlp": 1.00886154,
      "epoch": 0.6622576281376823,
      "flos": 14063876236800.0,
      "grad_norm": 1.7434265747445956,
      "language_loss": 0.77246261,
      "learning_rate": 1.081779858400137e-06,
      "loss": 0.7910639,
      "num_input_tokens_seen": 237789005,
      "step": 11015,
      "time_per_iteration": 2.538248300552368
    },
    {
      "auxiliary_loss_clip": 0.01096627,
      "auxiliary_loss_mlp": 0.00783727,
      "balance_loss_clip": 1.03737009,
      "balance_loss_mlp": 1.00835776,
      "epoch": 0.6623177513903502,
      "flos": 17018965169280.0,
      "grad_norm": 1.738101777870775,
      "language_loss": 0.81962752,
      "learning_rate": 1.0814338860803021e-06,
      "loss": 0.83843106,
      "num_input_tokens_seen": 237807740,
      "step": 11016,
      "time_per_iteration": 2.5041842460632324
    },
    {
      "auxiliary_loss_clip": 0.01092416,
      "auxiliary_loss_mlp": 0.01033888,
      "balance_loss_clip": 1.03647399,
      "balance_loss_mlp": 1.02165079,
      "epoch": 0.6623778746430182,
      "flos": 17271021882240.0,
      "grad_norm": 1.986037931733407,
      "language_loss": 0.69949079,
      "learning_rate": 1.0810879485925864e-06,
      "loss": 0.72075379,
      "num_input_tokens_seen": 237826340,
      "step": 11017,
      "time_per_iteration": 2.519240379333496
    },
    {
      "auxiliary_loss_clip": 0.01078409,
      "auxiliary_loss_mlp": 0.01038691,
      "balance_loss_clip": 1.03462887,
      "balance_loss_mlp": 1.02642465,
      "epoch": 0.6624379978956861,
      "flos": 48792688767360.0,
      "grad_norm": 1.7244389799511692,
      "language_loss": 0.77387106,
      "learning_rate": 1.0807420459501084e-06,
      "loss": 0.79504204,
      "num_input_tokens_seen": 237848305,
      "step": 11018,
      "time_per_iteration": 2.7897450923919678
    },
    {
      "auxiliary_loss_clip": 0.01087543,
      "auxiliary_loss_mlp": 0.01041037,
      "balance_loss_clip": 1.03831017,
      "balance_loss_mlp": 1.02868092,
      "epoch": 0.6624981211483542,
      "flos": 18952431477120.0,
      "grad_norm": 2.200270648644647,
      "language_loss": 0.83262616,
      "learning_rate": 1.0803961781659841e-06,
      "loss": 0.853912,
      "num_input_tokens_seen": 237867020,
      "step": 11019,
      "time_per_iteration": 2.530020236968994
    },
    {
      "auxiliary_loss_clip": 0.01090986,
      "auxiliary_loss_mlp": 0.00786273,
      "balance_loss_clip": 1.0362401,
      "balance_loss_mlp": 1.009444,
      "epoch": 0.6625582444010221,
      "flos": 23256576437760.0,
      "grad_norm": 1.578391139759986,
      "language_loss": 0.71927679,
      "learning_rate": 1.080050345253328e-06,
      "loss": 0.73804933,
      "num_input_tokens_seen": 237886710,
      "step": 11020,
      "time_per_iteration": 2.504007339477539
    },
    {
      "auxiliary_loss_clip": 0.0108644,
      "auxiliary_loss_mlp": 0.01029977,
      "balance_loss_clip": 1.03569579,
      "balance_loss_mlp": 1.01604128,
      "epoch": 0.6626183676536901,
      "flos": 21394823633280.0,
      "grad_norm": 1.8676620977001952,
      "language_loss": 0.72544914,
      "learning_rate": 1.0797045472252554e-06,
      "loss": 0.74661332,
      "num_input_tokens_seen": 237904795,
      "step": 11021,
      "time_per_iteration": 2.541767120361328
    },
    {
      "auxiliary_loss_clip": 0.01083561,
      "auxiliary_loss_mlp": 0.01032259,
      "balance_loss_clip": 1.03970146,
      "balance_loss_mlp": 1.01971245,
      "epoch": 0.662678490906358,
      "flos": 14571293713920.0,
      "grad_norm": 2.0392094865709107,
      "language_loss": 0.83303922,
      "learning_rate": 1.0793587840948793e-06,
      "loss": 0.85419738,
      "num_input_tokens_seen": 237921320,
      "step": 11022,
      "time_per_iteration": 2.5115621089935303
    },
    {
      "auxiliary_loss_clip": 0.01091535,
      "auxiliary_loss_mlp": 0.0103249,
      "balance_loss_clip": 1.03603125,
      "balance_loss_mlp": 1.01925802,
      "epoch": 0.662738614159026,
      "flos": 15992350554240.0,
      "grad_norm": 2.2559057639958957,
      "language_loss": 0.72427523,
      "learning_rate": 1.0790130558753099e-06,
      "loss": 0.74551547,
      "num_input_tokens_seen": 237933525,
      "step": 11023,
      "time_per_iteration": 2.4849395751953125
    },
    {
      "auxiliary_loss_clip": 0.01071707,
      "auxiliary_loss_mlp": 0.0103186,
      "balance_loss_clip": 1.03268158,
      "balance_loss_mlp": 1.02000427,
      "epoch": 0.6627987374116939,
      "flos": 19536338966400.0,
      "grad_norm": 1.7089693147990586,
      "language_loss": 0.74997479,
      "learning_rate": 1.0786673625796574e-06,
      "loss": 0.77101046,
      "num_input_tokens_seen": 237953395,
      "step": 11024,
      "time_per_iteration": 2.554080009460449
    },
    {
      "auxiliary_loss_clip": 0.0107883,
      "auxiliary_loss_mlp": 0.01033115,
      "balance_loss_clip": 1.03908741,
      "balance_loss_mlp": 1.02029371,
      "epoch": 0.662858860664362,
      "flos": 15702838934400.0,
      "grad_norm": 2.9515079458586246,
      "language_loss": 0.69405419,
      "learning_rate": 1.0783217042210306e-06,
      "loss": 0.71517366,
      "num_input_tokens_seen": 237971445,
      "step": 11025,
      "time_per_iteration": 2.5423531532287598
    },
    {
      "auxiliary_loss_clip": 0.01109895,
      "auxiliary_loss_mlp": 0.01033453,
      "balance_loss_clip": 1.03895485,
      "balance_loss_mlp": 1.02103114,
      "epoch": 0.6629189839170299,
      "flos": 20154289570560.0,
      "grad_norm": 1.5456591667701003,
      "language_loss": 0.78641146,
      "learning_rate": 1.0779760808125379e-06,
      "loss": 0.80784494,
      "num_input_tokens_seen": 237989965,
      "step": 11026,
      "time_per_iteration": 2.4615278244018555
    },
    {
      "auxiliary_loss_clip": 0.01097427,
      "auxiliary_loss_mlp": 0.01032279,
      "balance_loss_clip": 1.03814387,
      "balance_loss_mlp": 1.02080536,
      "epoch": 0.6629791071696979,
      "flos": 20915415786240.0,
      "grad_norm": 1.7348636602386804,
      "language_loss": 0.75763988,
      "learning_rate": 1.0776304923672842e-06,
      "loss": 0.77893698,
      "num_input_tokens_seen": 238006820,
      "step": 11027,
      "time_per_iteration": 2.508775234222412
    },
    {
      "auxiliary_loss_clip": 0.01083159,
      "auxiliary_loss_mlp": 0.01037634,
      "balance_loss_clip": 1.03754854,
      "balance_loss_mlp": 1.02429414,
      "epoch": 0.6630392304223659,
      "flos": 20846898593280.0,
      "grad_norm": 2.4881292857626076,
      "language_loss": 0.70050722,
      "learning_rate": 1.0772849388983742e-06,
      "loss": 0.72171509,
      "num_input_tokens_seen": 238022560,
      "step": 11028,
      "time_per_iteration": 2.508983612060547
    },
    {
      "auxiliary_loss_clip": 0.01096318,
      "auxiliary_loss_mlp": 0.01035633,
      "balance_loss_clip": 1.03727007,
      "balance_loss_mlp": 1.02498734,
      "epoch": 0.6630993536750338,
      "flos": 20995820380800.0,
      "grad_norm": 2.036728257625756,
      "language_loss": 0.79152519,
      "learning_rate": 1.0769394204189138e-06,
      "loss": 0.81284463,
      "num_input_tokens_seen": 238041895,
      "step": 11029,
      "time_per_iteration": 2.5053725242614746
    },
    {
      "auxiliary_loss_clip": 0.01109523,
      "auxiliary_loss_mlp": 0.01034906,
      "balance_loss_clip": 1.03683639,
      "balance_loss_mlp": 1.02207947,
      "epoch": 0.6631594769277018,
      "flos": 18259032355200.0,
      "grad_norm": 2.0924422070046576,
      "language_loss": 0.76275468,
      "learning_rate": 1.0765939369420012e-06,
      "loss": 0.784199,
      "num_input_tokens_seen": 238060445,
      "step": 11030,
      "time_per_iteration": 2.427781581878662
    },
    {
      "auxiliary_loss_clip": 0.0110572,
      "auxiliary_loss_mlp": 0.0103281,
      "balance_loss_clip": 1.04026532,
      "balance_loss_mlp": 1.01993585,
      "epoch": 0.6632196001803697,
      "flos": 17820491207040.0,
      "grad_norm": 2.484757292142552,
      "language_loss": 0.74786896,
      "learning_rate": 1.0762484884807391e-06,
      "loss": 0.76925427,
      "num_input_tokens_seen": 238077080,
      "step": 11031,
      "time_per_iteration": 2.4635274410247803
    },
    {
      "auxiliary_loss_clip": 0.01099601,
      "auxiliary_loss_mlp": 0.01035647,
      "balance_loss_clip": 1.03786683,
      "balance_loss_mlp": 1.02336836,
      "epoch": 0.6632797234330378,
      "flos": 12670182581760.0,
      "grad_norm": 3.4373314916005393,
      "language_loss": 0.75015366,
      "learning_rate": 1.075903075048228e-06,
      "loss": 0.77150607,
      "num_input_tokens_seen": 238091045,
      "step": 11032,
      "time_per_iteration": 2.4324967861175537
    },
    {
      "auxiliary_loss_clip": 0.01065898,
      "auxiliary_loss_mlp": 0.01033041,
      "balance_loss_clip": 1.03565323,
      "balance_loss_mlp": 1.02117372,
      "epoch": 0.6633398466857057,
      "flos": 23584728113280.0,
      "grad_norm": 1.664783140786834,
      "language_loss": 0.80556256,
      "learning_rate": 1.0755576966575635e-06,
      "loss": 0.82655191,
      "num_input_tokens_seen": 238110220,
      "step": 11033,
      "time_per_iteration": 2.5921905040740967
    },
    {
      "auxiliary_loss_clip": 0.01087237,
      "auxiliary_loss_mlp": 0.01030017,
      "balance_loss_clip": 1.03588367,
      "balance_loss_mlp": 1.01797116,
      "epoch": 0.6633999699383737,
      "flos": 20631686256000.0,
      "grad_norm": 1.7960508557916515,
      "language_loss": 0.80459327,
      "learning_rate": 1.0752123533218451e-06,
      "loss": 0.82576585,
      "num_input_tokens_seen": 238130400,
      "step": 11034,
      "time_per_iteration": 2.5333621501922607
    },
    {
      "auxiliary_loss_clip": 0.01094604,
      "auxiliary_loss_mlp": 0.01028273,
      "balance_loss_clip": 1.04068053,
      "balance_loss_mlp": 1.01717472,
      "epoch": 0.6634600931910416,
      "flos": 21797095023360.0,
      "grad_norm": 1.5659032669100699,
      "language_loss": 0.75768536,
      "learning_rate": 1.074867045054166e-06,
      "loss": 0.77891409,
      "num_input_tokens_seen": 238148165,
      "step": 11035,
      "time_per_iteration": 2.515777349472046
    },
    {
      "auxiliary_loss_clip": 0.01075631,
      "auxiliary_loss_mlp": 0.0102893,
      "balance_loss_clip": 1.03583193,
      "balance_loss_mlp": 1.01657438,
      "epoch": 0.6635202164437096,
      "flos": 18732873594240.0,
      "grad_norm": 1.8389129377232585,
      "language_loss": 0.83159602,
      "learning_rate": 1.074521771867622e-06,
      "loss": 0.85264158,
      "num_input_tokens_seen": 238166360,
      "step": 11036,
      "time_per_iteration": 2.5286686420440674
    },
    {
      "auxiliary_loss_clip": 0.01041688,
      "auxiliary_loss_mlp": 0.0101001,
      "balance_loss_clip": 1.0179472,
      "balance_loss_mlp": 1.00891924,
      "epoch": 0.6635803396963775,
      "flos": 60222771227520.0,
      "grad_norm": 0.7784713802701722,
      "language_loss": 0.52358103,
      "learning_rate": 1.0741765337753044e-06,
      "loss": 0.54409802,
      "num_input_tokens_seen": 238227630,
      "step": 11037,
      "time_per_iteration": 3.066641092300415
    },
    {
      "auxiliary_loss_clip": 0.01061705,
      "auxiliary_loss_mlp": 0.01036086,
      "balance_loss_clip": 1.03859568,
      "balance_loss_mlp": 1.0231874,
      "epoch": 0.6636404629490456,
      "flos": 29167041611520.0,
      "grad_norm": 1.626875357304339,
      "language_loss": 0.78969097,
      "learning_rate": 1.0738313307903052e-06,
      "loss": 0.81066883,
      "num_input_tokens_seen": 238248435,
      "step": 11038,
      "time_per_iteration": 2.6506268978118896
    },
    {
      "auxiliary_loss_clip": 0.01076779,
      "auxiliary_loss_mlp": 0.01039656,
      "balance_loss_clip": 1.03708959,
      "balance_loss_mlp": 1.02632225,
      "epoch": 0.6637005862017135,
      "flos": 38907702766080.0,
      "grad_norm": 2.178367291989814,
      "language_loss": 0.64003748,
      "learning_rate": 1.073486162925716e-06,
      "loss": 0.66120183,
      "num_input_tokens_seen": 238268755,
      "step": 11039,
      "time_per_iteration": 2.708106279373169
    },
    {
      "auxiliary_loss_clip": 0.01077766,
      "auxiliary_loss_mlp": 0.01027431,
      "balance_loss_clip": 1.03969169,
      "balance_loss_mlp": 1.01515269,
      "epoch": 0.6637607094543815,
      "flos": 22783345729920.0,
      "grad_norm": 1.4294960898227393,
      "language_loss": 0.63908958,
      "learning_rate": 1.0731410301946237e-06,
      "loss": 0.66014159,
      "num_input_tokens_seen": 238290120,
      "step": 11040,
      "time_per_iteration": 3.9671413898468018
    },
    {
      "auxiliary_loss_clip": 0.01069837,
      "auxiliary_loss_mlp": 0.01039659,
      "balance_loss_clip": 1.03451109,
      "balance_loss_mlp": 1.02609932,
      "epoch": 0.6638208327070495,
      "flos": 18114096977280.0,
      "grad_norm": 2.5208647684566343,
      "language_loss": 0.71539247,
      "learning_rate": 1.0727959326101161e-06,
      "loss": 0.73648739,
      "num_input_tokens_seen": 238309290,
      "step": 11041,
      "time_per_iteration": 2.5506534576416016
    },
    {
      "auxiliary_loss_clip": 0.01093377,
      "auxiliary_loss_mlp": 0.0104439,
      "balance_loss_clip": 1.03661895,
      "balance_loss_mlp": 1.02985275,
      "epoch": 0.6638809559597174,
      "flos": 29424880414080.0,
      "grad_norm": 2.1595034504289075,
      "language_loss": 0.62019277,
      "learning_rate": 1.0724508701852806e-06,
      "loss": 0.64157045,
      "num_input_tokens_seen": 238327280,
      "step": 11042,
      "time_per_iteration": 2.5637094974517822
    },
    {
      "auxiliary_loss_clip": 0.01100338,
      "auxiliary_loss_mlp": 0.0103117,
      "balance_loss_clip": 1.03690696,
      "balance_loss_mlp": 1.01724625,
      "epoch": 0.6639410792123854,
      "flos": 28072699902720.0,
      "grad_norm": 1.7953385432722393,
      "language_loss": 0.68618917,
      "learning_rate": 1.0721058429331998e-06,
      "loss": 0.70750421,
      "num_input_tokens_seen": 238346330,
      "step": 11043,
      "time_per_iteration": 2.5357120037078857
    },
    {
      "auxiliary_loss_clip": 0.01095319,
      "auxiliary_loss_mlp": 0.01027001,
      "balance_loss_clip": 1.0392592,
      "balance_loss_mlp": 1.0164752,
      "epoch": 0.6640012024650533,
      "flos": 25556367600000.0,
      "grad_norm": 1.663911642267484,
      "language_loss": 0.83953547,
      "learning_rate": 1.0717608508669587e-06,
      "loss": 0.86075866,
      "num_input_tokens_seen": 238364650,
      "step": 11044,
      "time_per_iteration": 2.5406494140625
    },
    {
      "auxiliary_loss_clip": 0.01070382,
      "auxiliary_loss_mlp": 0.01030778,
      "balance_loss_clip": 1.03660154,
      "balance_loss_mlp": 1.01787353,
      "epoch": 0.6640613257177214,
      "flos": 14866946559360.0,
      "grad_norm": 2.271812081657629,
      "language_loss": 0.69502258,
      "learning_rate": 1.0714158939996392e-06,
      "loss": 0.71603417,
      "num_input_tokens_seen": 238381630,
      "step": 11045,
      "time_per_iteration": 2.541288375854492
    },
    {
      "auxiliary_loss_clip": 0.01099832,
      "auxiliary_loss_mlp": 0.01028612,
      "balance_loss_clip": 1.03987265,
      "balance_loss_mlp": 1.01607108,
      "epoch": 0.6641214489703893,
      "flos": 23221096778880.0,
      "grad_norm": 1.5296019298084123,
      "language_loss": 0.64230257,
      "learning_rate": 1.0710709723443235e-06,
      "loss": 0.66358709,
      "num_input_tokens_seen": 238402595,
      "step": 11046,
      "time_per_iteration": 2.5263280868530273
    },
    {
      "auxiliary_loss_clip": 0.01078289,
      "auxiliary_loss_mlp": 0.01029634,
      "balance_loss_clip": 1.03760195,
      "balance_loss_mlp": 1.01658607,
      "epoch": 0.6641815722230573,
      "flos": 37742617221120.0,
      "grad_norm": 1.4711934492700298,
      "language_loss": 0.71332347,
      "learning_rate": 1.070726085914088e-06,
      "loss": 0.73440266,
      "num_input_tokens_seen": 238426860,
      "step": 11047,
      "time_per_iteration": 4.136700391769409
    },
    {
      "auxiliary_loss_clip": 0.01051542,
      "auxiliary_loss_mlp": 0.01034909,
      "balance_loss_clip": 1.04150629,
      "balance_loss_mlp": 1.02262414,
      "epoch": 0.6642416954757252,
      "flos": 17931132074880.0,
      "grad_norm": 1.9708782645392695,
      "language_loss": 0.77218825,
      "learning_rate": 1.0703812347220126e-06,
      "loss": 0.79305273,
      "num_input_tokens_seen": 238443990,
      "step": 11048,
      "time_per_iteration": 4.022372007369995
    },
    {
      "auxiliary_loss_clip": 0.01017733,
      "auxiliary_loss_mlp": 0.01005796,
      "balance_loss_clip": 1.02388501,
      "balance_loss_mlp": 1.00467575,
      "epoch": 0.6643018187283932,
      "flos": 51995384104320.0,
      "grad_norm": 0.7538568889327667,
      "language_loss": 0.55069649,
      "learning_rate": 1.0700364187811745e-06,
      "loss": 0.57093179,
      "num_input_tokens_seen": 238503045,
      "step": 11049,
      "time_per_iteration": 3.184821605682373
    },
    {
      "auxiliary_loss_clip": 0.01098711,
      "auxiliary_loss_mlp": 0.01026587,
      "balance_loss_clip": 1.03860414,
      "balance_loss_mlp": 1.01542914,
      "epoch": 0.6643619419810611,
      "flos": 30226657847040.0,
      "grad_norm": 1.6827277619146914,
      "language_loss": 0.6425401,
      "learning_rate": 1.069691638104648e-06,
      "loss": 0.66379309,
      "num_input_tokens_seen": 238527320,
      "step": 11050,
      "time_per_iteration": 2.5646986961364746
    },
    {
      "auxiliary_loss_clip": 0.01107558,
      "auxiliary_loss_mlp": 0.01027664,
      "balance_loss_clip": 1.03929281,
      "balance_loss_mlp": 1.01625538,
      "epoch": 0.6644220652337292,
      "flos": 22966131064320.0,
      "grad_norm": 1.9107587604432172,
      "language_loss": 0.78804231,
      "learning_rate": 1.0693468927055085e-06,
      "loss": 0.80939454,
      "num_input_tokens_seen": 238546030,
      "step": 11051,
      "time_per_iteration": 3.8733208179473877
    },
    {
      "auxiliary_loss_clip": 0.01087765,
      "auxiliary_loss_mlp": 0.01032768,
      "balance_loss_clip": 1.04077959,
      "balance_loss_mlp": 1.02090096,
      "epoch": 0.6644821884863971,
      "flos": 21142228216320.0,
      "grad_norm": 1.6879672573824887,
      "language_loss": 0.85401672,
      "learning_rate": 1.0690021825968276e-06,
      "loss": 0.87522209,
      "num_input_tokens_seen": 238564175,
      "step": 11052,
      "time_per_iteration": 2.525153160095215
    },
    {
      "auxiliary_loss_clip": 0.01062033,
      "auxiliary_loss_mlp": 0.01034936,
      "balance_loss_clip": 1.03534222,
      "balance_loss_mlp": 1.02130437,
      "epoch": 0.6645423117390651,
      "flos": 20192821885440.0,
      "grad_norm": 2.2837350206064007,
      "language_loss": 0.74819767,
      "learning_rate": 1.0686575077916776e-06,
      "loss": 0.76916742,
      "num_input_tokens_seen": 238581010,
      "step": 11053,
      "time_per_iteration": 2.5611822605133057
    },
    {
      "auxiliary_loss_clip": 0.01086221,
      "auxiliary_loss_mlp": 0.01026306,
      "balance_loss_clip": 1.03843653,
      "balance_loss_mlp": 1.01485026,
      "epoch": 0.6646024349917331,
      "flos": 24351959640960.0,
      "grad_norm": 1.5354267960615073,
      "language_loss": 0.79526347,
      "learning_rate": 1.0683128683031278e-06,
      "loss": 0.81638873,
      "num_input_tokens_seen": 238601365,
      "step": 11054,
      "time_per_iteration": 2.547175407409668
    },
    {
      "auxiliary_loss_clip": 0.01061832,
      "auxiliary_loss_mlp": 0.01027393,
      "balance_loss_clip": 1.03673625,
      "balance_loss_mlp": 1.01603222,
      "epoch": 0.664662558244401,
      "flos": 18806706000000.0,
      "grad_norm": 1.5082398458964315,
      "language_loss": 0.73926282,
      "learning_rate": 1.0679682641442472e-06,
      "loss": 0.76015508,
      "num_input_tokens_seen": 238619850,
      "step": 11055,
      "time_per_iteration": 2.5629475116729736
    },
    {
      "auxiliary_loss_clip": 0.01074713,
      "auxiliary_loss_mlp": 0.01037022,
      "balance_loss_clip": 1.03642583,
      "balance_loss_mlp": 1.02378964,
      "epoch": 0.664722681497069,
      "flos": 18952790613120.0,
      "grad_norm": 1.7578777973609383,
      "language_loss": 0.73113978,
      "learning_rate": 1.0676236953281042e-06,
      "loss": 0.75225711,
      "num_input_tokens_seen": 238637635,
      "step": 11056,
      "time_per_iteration": 2.5518975257873535
    },
    {
      "auxiliary_loss_clip": 0.01065975,
      "auxiliary_loss_mlp": 0.01029609,
      "balance_loss_clip": 1.03763461,
      "balance_loss_mlp": 1.0172236,
      "epoch": 0.6647828047497369,
      "flos": 19571279921280.0,
      "grad_norm": 1.7264258080387986,
      "language_loss": 0.69815093,
      "learning_rate": 1.0672791618677641e-06,
      "loss": 0.71910679,
      "num_input_tokens_seen": 238656200,
      "step": 11057,
      "time_per_iteration": 2.587862014770508
    },
    {
      "auxiliary_loss_clip": 0.01098117,
      "auxiliary_loss_mlp": 0.01030395,
      "balance_loss_clip": 1.03796077,
      "balance_loss_mlp": 1.01761603,
      "epoch": 0.664842928002405,
      "flos": 23149455102720.0,
      "grad_norm": 1.8080603369582222,
      "language_loss": 0.80389333,
      "learning_rate": 1.066934663776291e-06,
      "loss": 0.82517844,
      "num_input_tokens_seen": 238675005,
      "step": 11058,
      "time_per_iteration": 2.5208160877227783
    },
    {
      "auxiliary_loss_clip": 0.01017268,
      "auxiliary_loss_mlp": 0.01000923,
      "balance_loss_clip": 1.02322757,
      "balance_loss_mlp": 0.9997074,
      "epoch": 0.6649030512550729,
      "flos": 65244913148160.0,
      "grad_norm": 0.8122905214309895,
      "language_loss": 0.62603319,
      "learning_rate": 1.0665902010667496e-06,
      "loss": 0.64621508,
      "num_input_tokens_seen": 238731425,
      "step": 11059,
      "time_per_iteration": 3.0559463500976562
    },
    {
      "auxiliary_loss_clip": 0.01096188,
      "auxiliary_loss_mlp": 0.01034315,
      "balance_loss_clip": 1.03824413,
      "balance_loss_mlp": 1.02291262,
      "epoch": 0.6649631745077409,
      "flos": 20194797133440.0,
      "grad_norm": 1.4885684902407454,
      "language_loss": 0.78889346,
      "learning_rate": 1.0662457737522008e-06,
      "loss": 0.81019849,
      "num_input_tokens_seen": 238752020,
      "step": 11060,
      "time_per_iteration": 2.5351386070251465
    },
    {
      "auxiliary_loss_clip": 0.01074799,
      "auxiliary_loss_mlp": 0.01032418,
      "balance_loss_clip": 1.03679013,
      "balance_loss_mlp": 1.01965082,
      "epoch": 0.6650232977604088,
      "flos": 17238558965760.0,
      "grad_norm": 3.0884884290643306,
      "language_loss": 0.78827119,
      "learning_rate": 1.0659013818457055e-06,
      "loss": 0.80934334,
      "num_input_tokens_seen": 238769665,
      "step": 11061,
      "time_per_iteration": 2.53368878364563
    },
    {
      "auxiliary_loss_clip": 0.01085812,
      "auxiliary_loss_mlp": 0.01027401,
      "balance_loss_clip": 1.03967547,
      "balance_loss_mlp": 1.01592684,
      "epoch": 0.6650834210130768,
      "flos": 10006867825920.0,
      "grad_norm": 2.1040815979927614,
      "language_loss": 0.57230723,
      "learning_rate": 1.0655570253603243e-06,
      "loss": 0.59343934,
      "num_input_tokens_seen": 238782180,
      "step": 11062,
      "time_per_iteration": 2.4920263290405273
    },
    {
      "auxiliary_loss_clip": 0.01097108,
      "auxiliary_loss_mlp": 0.01030739,
      "balance_loss_clip": 1.03653121,
      "balance_loss_mlp": 1.01552808,
      "epoch": 0.6651435442657447,
      "flos": 10452088903680.0,
      "grad_norm": 1.7387067973288919,
      "language_loss": 0.76328224,
      "learning_rate": 1.0652127043091144e-06,
      "loss": 0.78456068,
      "num_input_tokens_seen": 238800315,
      "step": 11063,
      "time_per_iteration": 2.4727320671081543
    },
    {
      "auxiliary_loss_clip": 0.01050048,
      "auxiliary_loss_mlp": 0.01039573,
      "balance_loss_clip": 1.03824139,
      "balance_loss_mlp": 1.02701473,
      "epoch": 0.6652036675184128,
      "flos": 22344229964160.0,
      "grad_norm": 1.3284605367771636,
      "language_loss": 0.70547807,
      "learning_rate": 1.0648684187051316e-06,
      "loss": 0.72637433,
      "num_input_tokens_seen": 238822250,
      "step": 11064,
      "time_per_iteration": 2.6677913665771484
    },
    {
      "auxiliary_loss_clip": 0.01042667,
      "auxiliary_loss_mlp": 0.01003371,
      "balance_loss_clip": 1.018852,
      "balance_loss_mlp": 1.00215495,
      "epoch": 0.6652637907710807,
      "flos": 52909633998720.0,
      "grad_norm": 0.8461319635708201,
      "language_loss": 0.63093621,
      "learning_rate": 1.0645241685614322e-06,
      "loss": 0.65139657,
      "num_input_tokens_seen": 238877190,
      "step": 11065,
      "time_per_iteration": 3.047184705734253
    },
    {
      "auxiliary_loss_clip": 0.0109385,
      "auxiliary_loss_mlp": 0.01033788,
      "balance_loss_clip": 1.03674412,
      "balance_loss_mlp": 1.02041245,
      "epoch": 0.6653239140237487,
      "flos": 23104637907840.0,
      "grad_norm": 1.910241746635521,
      "language_loss": 0.62350047,
      "learning_rate": 1.0641799538910708e-06,
      "loss": 0.64477682,
      "num_input_tokens_seen": 238896010,
      "step": 11066,
      "time_per_iteration": 2.505260467529297
    },
    {
      "auxiliary_loss_clip": 0.01071102,
      "auxiliary_loss_mlp": 0.01035685,
      "balance_loss_clip": 1.034127,
      "balance_loss_mlp": 1.02080214,
      "epoch": 0.6653840372764167,
      "flos": 25959393175680.0,
      "grad_norm": 1.549308035522599,
      "language_loss": 0.70160532,
      "learning_rate": 1.0638357747070985e-06,
      "loss": 0.72267318,
      "num_input_tokens_seen": 238918990,
      "step": 11067,
      "time_per_iteration": 2.6121559143066406
    },
    {
      "auxiliary_loss_clip": 0.01029928,
      "auxiliary_loss_mlp": 0.01010654,
      "balance_loss_clip": 1.02550006,
      "balance_loss_mlp": 1.00906897,
      "epoch": 0.6654441605290846,
      "flos": 66041985899520.0,
      "grad_norm": 0.9272667256793777,
      "language_loss": 0.72155225,
      "learning_rate": 1.0634916310225684e-06,
      "loss": 0.74195802,
      "num_input_tokens_seen": 238975735,
      "step": 11068,
      "time_per_iteration": 3.097623586654663
    },
    {
      "auxiliary_loss_clip": 0.01014911,
      "auxiliary_loss_mlp": 0.01003074,
      "balance_loss_clip": 1.01997912,
      "balance_loss_mlp": 1.00184572,
      "epoch": 0.6655042837817526,
      "flos": 65196112521600.0,
      "grad_norm": 0.7059489001146961,
      "language_loss": 0.57854605,
      "learning_rate": 1.0631475228505285e-06,
      "loss": 0.59872591,
      "num_input_tokens_seen": 239042360,
      "step": 11069,
      "time_per_iteration": 3.2743537425994873
    },
    {
      "auxiliary_loss_clip": 0.01031491,
      "auxiliary_loss_mlp": 0.01002015,
      "balance_loss_clip": 1.01893866,
      "balance_loss_mlp": 1.00082862,
      "epoch": 0.6655644070344205,
      "flos": 69008746752000.0,
      "grad_norm": 0.7448897092041924,
      "language_loss": 0.63509452,
      "learning_rate": 1.062803450204029e-06,
      "loss": 0.6554296,
      "num_input_tokens_seen": 239109410,
      "step": 11070,
      "time_per_iteration": 3.1486968994140625
    },
    {
      "auxiliary_loss_clip": 0.0110611,
      "auxiliary_loss_mlp": 0.01024661,
      "balance_loss_clip": 1.03591967,
      "balance_loss_mlp": 1.0125258,
      "epoch": 0.6656245302870886,
      "flos": 36315562809600.0,
      "grad_norm": 2.326293851220407,
      "language_loss": 0.58568168,
      "learning_rate": 1.062459413096116e-06,
      "loss": 0.60698938,
      "num_input_tokens_seen": 239135345,
      "step": 11071,
      "time_per_iteration": 2.6182057857513428
    },
    {
      "auxiliary_loss_clip": 0.01101404,
      "auxiliary_loss_mlp": 0.01026133,
      "balance_loss_clip": 1.04060376,
      "balance_loss_mlp": 1.01470709,
      "epoch": 0.6656846535397565,
      "flos": 21794832466560.0,
      "grad_norm": 1.8650627524177608,
      "language_loss": 0.72921616,
      "learning_rate": 1.0621154115398364e-06,
      "loss": 0.7504915,
      "num_input_tokens_seen": 239154340,
      "step": 11072,
      "time_per_iteration": 2.5002121925354004
    },
    {
      "auxiliary_loss_clip": 0.01098142,
      "auxiliary_loss_mlp": 0.01029295,
      "balance_loss_clip": 1.03989959,
      "balance_loss_mlp": 1.01642013,
      "epoch": 0.6657447767924245,
      "flos": 37487615592960.0,
      "grad_norm": 1.7615903543181435,
      "language_loss": 0.70327383,
      "learning_rate": 1.0617714455482353e-06,
      "loss": 0.72454822,
      "num_input_tokens_seen": 239177815,
      "step": 11073,
      "time_per_iteration": 2.6457648277282715
    },
    {
      "auxiliary_loss_clip": 0.01078691,
      "auxiliary_loss_mlp": 0.0102946,
      "balance_loss_clip": 1.03865123,
      "balance_loss_mlp": 1.01669288,
      "epoch": 0.6658049000450924,
      "flos": 16837688206080.0,
      "grad_norm": 2.325918437425113,
      "language_loss": 0.5627259,
      "learning_rate": 1.061427515134354e-06,
      "loss": 0.58380747,
      "num_input_tokens_seen": 239195735,
      "step": 11074,
      "time_per_iteration": 2.539134979248047
    },
    {
      "auxiliary_loss_clip": 0.01110839,
      "auxiliary_loss_mlp": 0.00786026,
      "balance_loss_clip": 1.04087174,
      "balance_loss_mlp": 1.0130589,
      "epoch": 0.6658650232977604,
      "flos": 33510975863040.0,
      "grad_norm": 1.5143850287098277,
      "language_loss": 0.72126311,
      "learning_rate": 1.061083620311235e-06,
      "loss": 0.74023175,
      "num_input_tokens_seen": 239217535,
      "step": 11075,
      "time_per_iteration": 2.583421468734741
    },
    {
      "auxiliary_loss_clip": 0.01093605,
      "auxiliary_loss_mlp": 0.01029748,
      "balance_loss_clip": 1.03642213,
      "balance_loss_mlp": 1.0184288,
      "epoch": 0.6659251465504283,
      "flos": 37706311549440.0,
      "grad_norm": 1.4088291285426227,
      "language_loss": 0.65885389,
      "learning_rate": 1.0607397610919202e-06,
      "loss": 0.68008739,
      "num_input_tokens_seen": 239241975,
      "step": 11076,
      "time_per_iteration": 2.6276655197143555
    },
    {
      "auxiliary_loss_clip": 0.01085386,
      "auxiliary_loss_mlp": 0.0103102,
      "balance_loss_clip": 1.03583133,
      "balance_loss_mlp": 1.01777601,
      "epoch": 0.6659852698030964,
      "flos": 24893420232960.0,
      "grad_norm": 2.2159325474363185,
      "language_loss": 0.75381356,
      "learning_rate": 1.0603959374894468e-06,
      "loss": 0.77497762,
      "num_input_tokens_seen": 239262025,
      "step": 11077,
      "time_per_iteration": 2.557206392288208
    },
    {
      "auxiliary_loss_clip": 0.01085709,
      "auxiliary_loss_mlp": 0.01027927,
      "balance_loss_clip": 1.03738892,
      "balance_loss_mlp": 1.01607215,
      "epoch": 0.6660453930557643,
      "flos": 24352821567360.0,
      "grad_norm": 1.5968885040484282,
      "language_loss": 0.66666043,
      "learning_rate": 1.0600521495168538e-06,
      "loss": 0.68779683,
      "num_input_tokens_seen": 239282775,
      "step": 11078,
      "time_per_iteration": 2.5548598766326904
    },
    {
      "auxiliary_loss_clip": 0.01109768,
      "auxiliary_loss_mlp": 0.01029689,
      "balance_loss_clip": 1.03803039,
      "balance_loss_mlp": 1.0168916,
      "epoch": 0.6661055163084323,
      "flos": 10597814380800.0,
      "grad_norm": 1.850906655953811,
      "language_loss": 0.69527125,
      "learning_rate": 1.0597083971871783e-06,
      "loss": 0.71666586,
      "num_input_tokens_seen": 239299775,
      "step": 11079,
      "time_per_iteration": 3.8391008377075195
    },
    {
      "auxiliary_loss_clip": 0.01084683,
      "auxiliary_loss_mlp": 0.01026764,
      "balance_loss_clip": 1.03612328,
      "balance_loss_mlp": 1.01481938,
      "epoch": 0.6661656395611003,
      "flos": 24057491944320.0,
      "grad_norm": 1.4989782896625081,
      "language_loss": 0.8039813,
      "learning_rate": 1.0593646805134544e-06,
      "loss": 0.82509577,
      "num_input_tokens_seen": 239319660,
      "step": 11080,
      "time_per_iteration": 2.5860471725463867
    },
    {
      "auxiliary_loss_clip": 0.01072455,
      "auxiliary_loss_mlp": 0.01026845,
      "balance_loss_clip": 1.03668118,
      "balance_loss_mlp": 1.01608622,
      "epoch": 0.6662257628137682,
      "flos": 23036192542080.0,
      "grad_norm": 1.72490820734142,
      "language_loss": 0.78342462,
      "learning_rate": 1.0590209995087157e-06,
      "loss": 0.80441761,
      "num_input_tokens_seen": 239339215,
      "step": 11081,
      "time_per_iteration": 2.5940327644348145
    },
    {
      "auxiliary_loss_clip": 0.01071467,
      "auxiliary_loss_mlp": 0.01035053,
      "balance_loss_clip": 1.03891361,
      "balance_loss_mlp": 1.02035475,
      "epoch": 0.6662858860664362,
      "flos": 24754446512640.0,
      "grad_norm": 2.251565942930042,
      "language_loss": 0.80400741,
      "learning_rate": 1.0586773541859946e-06,
      "loss": 0.82507259,
      "num_input_tokens_seen": 239358545,
      "step": 11082,
      "time_per_iteration": 2.5947303771972656
    },
    {
      "auxiliary_loss_clip": 0.01077123,
      "auxiliary_loss_mlp": 0.01030396,
      "balance_loss_clip": 1.03947341,
      "balance_loss_mlp": 1.01901186,
      "epoch": 0.6663460093191041,
      "flos": 20009066883840.0,
      "grad_norm": 1.5600028838695947,
      "language_loss": 0.84018499,
      "learning_rate": 1.0583337445583234e-06,
      "loss": 0.86126018,
      "num_input_tokens_seen": 239376665,
      "step": 11083,
      "time_per_iteration": 2.6363441944122314
    },
    {
      "auxiliary_loss_clip": 0.01077011,
      "auxiliary_loss_mlp": 0.01030233,
      "balance_loss_clip": 1.0433321,
      "balance_loss_mlp": 1.01742387,
      "epoch": 0.6664061325717722,
      "flos": 17821389047040.0,
      "grad_norm": 2.426734344852017,
      "language_loss": 0.85635322,
      "learning_rate": 1.057990170638731e-06,
      "loss": 0.87742567,
      "num_input_tokens_seen": 239394345,
      "step": 11084,
      "time_per_iteration": 2.5958895683288574
    },
    {
      "auxiliary_loss_clip": 0.01091009,
      "auxiliary_loss_mlp": 0.01029227,
      "balance_loss_clip": 1.03904784,
      "balance_loss_mlp": 1.01581621,
      "epoch": 0.6664662558244401,
      "flos": 18076893465600.0,
      "grad_norm": 2.3770810926584423,
      "language_loss": 0.73197442,
      "learning_rate": 1.0576466324402452e-06,
      "loss": 0.75317675,
      "num_input_tokens_seen": 239410605,
      "step": 11085,
      "time_per_iteration": 2.534353256225586
    },
    {
      "auxiliary_loss_clip": 0.01085899,
      "auxiliary_loss_mlp": 0.01030976,
      "balance_loss_clip": 1.03704047,
      "balance_loss_mlp": 1.01878667,
      "epoch": 0.6665263790771081,
      "flos": 21574197175680.0,
      "grad_norm": 1.8523175316190241,
      "language_loss": 0.80661798,
      "learning_rate": 1.057303129975894e-06,
      "loss": 0.82778674,
      "num_input_tokens_seen": 239427155,
      "step": 11086,
      "time_per_iteration": 3.9244210720062256
    },
    {
      "auxiliary_loss_clip": 0.01082954,
      "auxiliary_loss_mlp": 0.0103486,
      "balance_loss_clip": 1.03553081,
      "balance_loss_mlp": 1.02252233,
      "epoch": 0.666586502329776,
      "flos": 24206629213440.0,
      "grad_norm": 1.9839264829117593,
      "language_loss": 0.75059915,
      "learning_rate": 1.056959663258702e-06,
      "loss": 0.77177727,
      "num_input_tokens_seen": 239445510,
      "step": 11087,
      "time_per_iteration": 3.9119529724121094
    },
    {
      "auxiliary_loss_clip": 0.01096081,
      "auxiliary_loss_mlp": 0.01029192,
      "balance_loss_clip": 1.03673339,
      "balance_loss_mlp": 1.01684153,
      "epoch": 0.666646625582444,
      "flos": 22200515648640.0,
      "grad_norm": 1.69722909199316,
      "language_loss": 0.6493963,
      "learning_rate": 1.0566162323016939e-06,
      "loss": 0.67064899,
      "num_input_tokens_seen": 239464805,
      "step": 11088,
      "time_per_iteration": 2.5081703662872314
    },
    {
      "auxiliary_loss_clip": 0.01096813,
      "auxiliary_loss_mlp": 0.01027546,
      "balance_loss_clip": 1.03666997,
      "balance_loss_mlp": 1.01424813,
      "epoch": 0.6667067488351119,
      "flos": 18259930195200.0,
      "grad_norm": 1.8340074928453254,
      "language_loss": 0.63883829,
      "learning_rate": 1.0562728371178928e-06,
      "loss": 0.66008186,
      "num_input_tokens_seen": 239483890,
      "step": 11089,
      "time_per_iteration": 3.9034316539764404
    },
    {
      "auxiliary_loss_clip": 0.01107341,
      "auxiliary_loss_mlp": 0.01030363,
      "balance_loss_clip": 1.03779531,
      "balance_loss_mlp": 1.01869845,
      "epoch": 0.66676687208778,
      "flos": 17236547804160.0,
      "grad_norm": 2.19155075225952,
      "language_loss": 0.80883372,
      "learning_rate": 1.0559294777203221e-06,
      "loss": 0.8302108,
      "num_input_tokens_seen": 239500080,
      "step": 11090,
      "time_per_iteration": 2.430241823196411
    },
    {
      "auxiliary_loss_clip": 0.01088615,
      "auxiliary_loss_mlp": 0.01029589,
      "balance_loss_clip": 1.03666914,
      "balance_loss_mlp": 1.0170784,
      "epoch": 0.6668269953404479,
      "flos": 19752197748480.0,
      "grad_norm": 2.0814703782893886,
      "language_loss": 0.77926624,
      "learning_rate": 1.0555861541219984e-06,
      "loss": 0.8004483,
      "num_input_tokens_seen": 239517335,
      "step": 11091,
      "time_per_iteration": 2.498779058456421
    },
    {
      "auxiliary_loss_clip": 0.01107187,
      "auxiliary_loss_mlp": 0.01033002,
      "balance_loss_clip": 1.03669667,
      "balance_loss_mlp": 1.02038991,
      "epoch": 0.6668871185931159,
      "flos": 20558428467840.0,
      "grad_norm": 1.852239748468378,
      "language_loss": 0.79544377,
      "learning_rate": 1.0552428663359425e-06,
      "loss": 0.81684566,
      "num_input_tokens_seen": 239536240,
      "step": 11092,
      "time_per_iteration": 2.4648191928863525
    },
    {
      "auxiliary_loss_clip": 0.01018619,
      "auxiliary_loss_mlp": 0.01004018,
      "balance_loss_clip": 1.02090716,
      "balance_loss_mlp": 1.00262916,
      "epoch": 0.6669472418457839,
      "flos": 58088167735680.0,
      "grad_norm": 0.7547449259343832,
      "language_loss": 0.57736588,
      "learning_rate": 1.0548996143751724e-06,
      "loss": 0.59759223,
      "num_input_tokens_seen": 239598000,
      "step": 11093,
      "time_per_iteration": 3.1754696369171143
    },
    {
      "auxiliary_loss_clip": 0.01107076,
      "auxiliary_loss_mlp": 0.01030807,
      "balance_loss_clip": 1.03759003,
      "balance_loss_mlp": 1.01833153,
      "epoch": 0.6670073650984518,
      "flos": 26065113880320.0,
      "grad_norm": 1.5235225362191867,
      "language_loss": 0.76402473,
      "learning_rate": 1.054556398252703e-06,
      "loss": 0.78540367,
      "num_input_tokens_seen": 239617650,
      "step": 11094,
      "time_per_iteration": 2.5357391834259033
    },
    {
      "auxiliary_loss_clip": 0.01109025,
      "auxiliary_loss_mlp": 0.01033487,
      "balance_loss_clip": 1.03803003,
      "balance_loss_mlp": 1.01992083,
      "epoch": 0.6670674883511198,
      "flos": 32416849635840.0,
      "grad_norm": 2.0519537655987357,
      "language_loss": 0.73284245,
      "learning_rate": 1.05421321798155e-06,
      "loss": 0.75426763,
      "num_input_tokens_seen": 239639825,
      "step": 11095,
      "time_per_iteration": 2.547170877456665
    },
    {
      "auxiliary_loss_clip": 0.01097626,
      "auxiliary_loss_mlp": 0.0103473,
      "balance_loss_clip": 1.03813982,
      "balance_loss_mlp": 1.02246368,
      "epoch": 0.6671276116037878,
      "flos": 18037786533120.0,
      "grad_norm": 2.158558505714864,
      "language_loss": 0.73227227,
      "learning_rate": 1.053870073574727e-06,
      "loss": 0.75359583,
      "num_input_tokens_seen": 239656300,
      "step": 11096,
      "time_per_iteration": 2.5008816719055176
    },
    {
      "auxiliary_loss_clip": 0.01066157,
      "auxiliary_loss_mlp": 0.01031927,
      "balance_loss_clip": 1.03745914,
      "balance_loss_mlp": 1.01955879,
      "epoch": 0.6671877348564558,
      "flos": 23767046570880.0,
      "grad_norm": 1.8789209083310365,
      "language_loss": 0.6456641,
      "learning_rate": 1.0535269650452456e-06,
      "loss": 0.66664493,
      "num_input_tokens_seen": 239676655,
      "step": 11097,
      "time_per_iteration": 2.567314624786377
    },
    {
      "auxiliary_loss_clip": 0.01096721,
      "auxiliary_loss_mlp": 0.01034365,
      "balance_loss_clip": 1.0366118,
      "balance_loss_mlp": 1.02208662,
      "epoch": 0.6672478581091237,
      "flos": 20918360701440.0,
      "grad_norm": 2.0577321443636003,
      "language_loss": 0.75588703,
      "learning_rate": 1.0531838924061158e-06,
      "loss": 0.77719796,
      "num_input_tokens_seen": 239695430,
      "step": 11098,
      "time_per_iteration": 2.5067646503448486
    },
    {
      "auxiliary_loss_clip": 0.01110228,
      "auxiliary_loss_mlp": 0.01028932,
      "balance_loss_clip": 1.03892994,
      "balance_loss_mlp": 1.01721382,
      "epoch": 0.6673079813617917,
      "flos": 27855799626240.0,
      "grad_norm": 1.5330941094557409,
      "language_loss": 0.74252093,
      "learning_rate": 1.0528408556703476e-06,
      "loss": 0.76391256,
      "num_input_tokens_seen": 239717070,
      "step": 11099,
      "time_per_iteration": 2.5015766620635986
    },
    {
      "auxiliary_loss_clip": 0.01095709,
      "auxiliary_loss_mlp": 0.01032527,
      "balance_loss_clip": 1.03664899,
      "balance_loss_mlp": 1.02033794,
      "epoch": 0.6673681046144596,
      "flos": 21616859554560.0,
      "grad_norm": 2.0439684907817783,
      "language_loss": 0.78070068,
      "learning_rate": 1.0524978548509502e-06,
      "loss": 0.80198306,
      "num_input_tokens_seen": 239737105,
      "step": 11100,
      "time_per_iteration": 2.518545389175415
    },
    {
      "auxiliary_loss_clip": 0.01106245,
      "auxiliary_loss_mlp": 0.01038531,
      "balance_loss_clip": 1.03745496,
      "balance_loss_mlp": 1.02649665,
      "epoch": 0.6674282278671276,
      "flos": 20889884194560.0,
      "grad_norm": 2.4292090260575248,
      "language_loss": 0.6000495,
      "learning_rate": 1.0521548899609288e-06,
      "loss": 0.62149727,
      "num_input_tokens_seen": 239757835,
      "step": 11101,
      "time_per_iteration": 2.5018982887268066
    },
    {
      "auxiliary_loss_clip": 0.01092704,
      "auxiliary_loss_mlp": 0.0103263,
      "balance_loss_clip": 1.03760481,
      "balance_loss_mlp": 1.01976132,
      "epoch": 0.6674883511197955,
      "flos": 23624194181760.0,
      "grad_norm": 1.9021233140857503,
      "language_loss": 0.71461767,
      "learning_rate": 1.0518119610132884e-06,
      "loss": 0.73587102,
      "num_input_tokens_seen": 239775425,
      "step": 11102,
      "time_per_iteration": 2.5586724281311035
    },
    {
      "auxiliary_loss_clip": 0.01098808,
      "auxiliary_loss_mlp": 0.01030107,
      "balance_loss_clip": 1.03743768,
      "balance_loss_mlp": 1.01751864,
      "epoch": 0.6675484743724636,
      "flos": 19609668581760.0,
      "grad_norm": 1.7679899271873711,
      "language_loss": 0.84191072,
      "learning_rate": 1.051469068021034e-06,
      "loss": 0.86319995,
      "num_input_tokens_seen": 239794605,
      "step": 11103,
      "time_per_iteration": 2.4873523712158203
    },
    {
      "auxiliary_loss_clip": 0.0108739,
      "auxiliary_loss_mlp": 0.01026331,
      "balance_loss_clip": 1.0367136,
      "balance_loss_mlp": 1.01423693,
      "epoch": 0.6676085976251315,
      "flos": 14319452482560.0,
      "grad_norm": 1.9992358746769094,
      "language_loss": 0.78408754,
      "learning_rate": 1.0511262109971668e-06,
      "loss": 0.80522472,
      "num_input_tokens_seen": 239812135,
      "step": 11104,
      "time_per_iteration": 2.5060653686523438
    },
    {
      "auxiliary_loss_clip": 0.0106754,
      "auxiliary_loss_mlp": 0.01030663,
      "balance_loss_clip": 1.04049158,
      "balance_loss_mlp": 1.01853347,
      "epoch": 0.6676687208777995,
      "flos": 38104596529920.0,
      "grad_norm": 1.6015175838503426,
      "language_loss": 0.58029187,
      "learning_rate": 1.0507833899546889e-06,
      "loss": 0.60127389,
      "num_input_tokens_seen": 239835845,
      "step": 11105,
      "time_per_iteration": 2.7381420135498047
    },
    {
      "auxiliary_loss_clip": 0.01101919,
      "auxiliary_loss_mlp": 0.01036095,
      "balance_loss_clip": 1.03843141,
      "balance_loss_mlp": 1.02258813,
      "epoch": 0.6677288441304675,
      "flos": 23981576549760.0,
      "grad_norm": 1.7268760904705092,
      "language_loss": 0.73059368,
      "learning_rate": 1.0504406049066e-06,
      "loss": 0.75197375,
      "num_input_tokens_seen": 239853820,
      "step": 11106,
      "time_per_iteration": 2.5066568851470947
    },
    {
      "auxiliary_loss_clip": 0.01108837,
      "auxiliary_loss_mlp": 0.0103147,
      "balance_loss_clip": 1.03845429,
      "balance_loss_mlp": 1.01892936,
      "epoch": 0.6677889673831354,
      "flos": 24170682677760.0,
      "grad_norm": 1.6293079266028472,
      "language_loss": 0.7678858,
      "learning_rate": 1.0500978558659e-06,
      "loss": 0.78928888,
      "num_input_tokens_seen": 239873365,
      "step": 11107,
      "time_per_iteration": 2.513930559158325
    },
    {
      "auxiliary_loss_clip": 0.0108366,
      "auxiliary_loss_mlp": 0.01029065,
      "balance_loss_clip": 1.03562307,
      "balance_loss_mlp": 1.01665592,
      "epoch": 0.6678490906358034,
      "flos": 22309648145280.0,
      "grad_norm": 2.1193948039654833,
      "language_loss": 0.89753294,
      "learning_rate": 1.049755142845583e-06,
      "loss": 0.91866016,
      "num_input_tokens_seen": 239891215,
      "step": 11108,
      "time_per_iteration": 2.533521890640259
    },
    {
      "auxiliary_loss_clip": 0.01075058,
      "auxiliary_loss_mlp": 0.01022584,
      "balance_loss_clip": 1.0388515,
      "balance_loss_mlp": 1.01208735,
      "epoch": 0.6679092138884714,
      "flos": 36898752026880.0,
      "grad_norm": 1.4242363988399898,
      "language_loss": 0.82517898,
      "learning_rate": 1.049412465858646e-06,
      "loss": 0.84615535,
      "num_input_tokens_seen": 239913490,
      "step": 11109,
      "time_per_iteration": 2.6941442489624023
    },
    {
      "auxiliary_loss_clip": 0.01081675,
      "auxiliary_loss_mlp": 0.01031842,
      "balance_loss_clip": 1.03705513,
      "balance_loss_mlp": 1.01865733,
      "epoch": 0.6679693371411394,
      "flos": 18150294908160.0,
      "grad_norm": 2.116081089081043,
      "language_loss": 0.69058001,
      "learning_rate": 1.0490698249180847e-06,
      "loss": 0.71171516,
      "num_input_tokens_seen": 239931565,
      "step": 11110,
      "time_per_iteration": 2.605433225631714
    },
    {
      "auxiliary_loss_clip": 0.01082559,
      "auxiliary_loss_mlp": 0.01030629,
      "balance_loss_clip": 1.04038692,
      "balance_loss_mlp": 1.01716423,
      "epoch": 0.6680294603938073,
      "flos": 27198167472000.0,
      "grad_norm": 1.4191915130798025,
      "language_loss": 0.73227501,
      "learning_rate": 1.04872722003689e-06,
      "loss": 0.75340688,
      "num_input_tokens_seen": 239952395,
      "step": 11111,
      "time_per_iteration": 2.594424247741699
    },
    {
      "auxiliary_loss_clip": 0.01106634,
      "auxiliary_loss_mlp": 0.01029921,
      "balance_loss_clip": 1.03738177,
      "balance_loss_mlp": 1.0181911,
      "epoch": 0.6680895836464753,
      "flos": 21725309692800.0,
      "grad_norm": 1.9700593284828976,
      "language_loss": 0.64723182,
      "learning_rate": 1.0483846512280553e-06,
      "loss": 0.6685974,
      "num_input_tokens_seen": 239968910,
      "step": 11112,
      "time_per_iteration": 2.466550588607788
    },
    {
      "auxiliary_loss_clip": 0.01082668,
      "auxiliary_loss_mlp": 0.01029341,
      "balance_loss_clip": 1.03659797,
      "balance_loss_mlp": 1.01683021,
      "epoch": 0.6681497068991432,
      "flos": 19646477043840.0,
      "grad_norm": 2.1642911536883163,
      "language_loss": 0.63146424,
      "learning_rate": 1.048042118504569e-06,
      "loss": 0.65258431,
      "num_input_tokens_seen": 239987680,
      "step": 11113,
      "time_per_iteration": 2.516191005706787
    },
    {
      "auxiliary_loss_clip": 0.01065058,
      "auxiliary_loss_mlp": 0.01031405,
      "balance_loss_clip": 1.03824902,
      "balance_loss_mlp": 1.01986599,
      "epoch": 0.6682098301518112,
      "flos": 17419153570560.0,
      "grad_norm": 1.7717385213509131,
      "language_loss": 0.65932858,
      "learning_rate": 1.047699621879422e-06,
      "loss": 0.6802932,
      "num_input_tokens_seen": 240005790,
      "step": 11114,
      "time_per_iteration": 2.608201026916504
    },
    {
      "auxiliary_loss_clip": 0.01098472,
      "auxiliary_loss_mlp": 0.0103635,
      "balance_loss_clip": 1.03740478,
      "balance_loss_mlp": 1.02400577,
      "epoch": 0.6682699534044791,
      "flos": 22599016110720.0,
      "grad_norm": 1.5397630694577742,
      "language_loss": 0.78444195,
      "learning_rate": 1.0473571613655998e-06,
      "loss": 0.80579019,
      "num_input_tokens_seen": 240025895,
      "step": 11115,
      "time_per_iteration": 2.5170388221740723
    },
    {
      "auxiliary_loss_clip": 0.01056736,
      "auxiliary_loss_mlp": 0.00786172,
      "balance_loss_clip": 1.03332734,
      "balance_loss_mlp": 1.00995076,
      "epoch": 0.6683300766571472,
      "flos": 24863686750080.0,
      "grad_norm": 1.7204507478249427,
      "language_loss": 0.79509777,
      "learning_rate": 1.0470147369760896e-06,
      "loss": 0.81352687,
      "num_input_tokens_seen": 240044880,
      "step": 11116,
      "time_per_iteration": 2.635437250137329
    },
    {
      "auxiliary_loss_clip": 0.01078927,
      "auxiliary_loss_mlp": 0.01032375,
      "balance_loss_clip": 1.03794909,
      "balance_loss_mlp": 1.01902342,
      "epoch": 0.6683901999098151,
      "flos": 27126633536640.0,
      "grad_norm": 1.654898173189274,
      "language_loss": 0.79437375,
      "learning_rate": 1.0466723487238768e-06,
      "loss": 0.81548679,
      "num_input_tokens_seen": 240065785,
      "step": 11117,
      "time_per_iteration": 2.614931106567383
    },
    {
      "auxiliary_loss_clip": 0.01062651,
      "auxiliary_loss_mlp": 0.01031023,
      "balance_loss_clip": 1.03810024,
      "balance_loss_mlp": 1.01727247,
      "epoch": 0.6684503231624831,
      "flos": 20739023072640.0,
      "grad_norm": 3.5624378159656036,
      "language_loss": 0.65671599,
      "learning_rate": 1.0463299966219441e-06,
      "loss": 0.67765278,
      "num_input_tokens_seen": 240085130,
      "step": 11118,
      "time_per_iteration": 3.97074294090271
    },
    {
      "auxiliary_loss_clip": 0.01086293,
      "auxiliary_loss_mlp": 0.01027758,
      "balance_loss_clip": 1.03691721,
      "balance_loss_mlp": 1.01620054,
      "epoch": 0.668510446415151,
      "flos": 21762189982080.0,
      "grad_norm": 1.5060460234290567,
      "language_loss": 0.69316787,
      "learning_rate": 1.0459876806832727e-06,
      "loss": 0.71430838,
      "num_input_tokens_seen": 240105495,
      "step": 11119,
      "time_per_iteration": 2.642090082168579
    },
    {
      "auxiliary_loss_clip": 0.01082906,
      "auxiliary_loss_mlp": 0.0103454,
      "balance_loss_clip": 1.03614581,
      "balance_loss_mlp": 1.02185643,
      "epoch": 0.668570569667819,
      "flos": 30191250015360.0,
      "grad_norm": 1.7088413246480656,
      "language_loss": 0.6716072,
      "learning_rate": 1.0456454009208448e-06,
      "loss": 0.69278169,
      "num_input_tokens_seen": 240125455,
      "step": 11120,
      "time_per_iteration": 2.6110098361968994
    },
    {
      "auxiliary_loss_clip": 0.01077055,
      "auxiliary_loss_mlp": 0.01030823,
      "balance_loss_clip": 1.03615975,
      "balance_loss_mlp": 1.01772237,
      "epoch": 0.668630692920487,
      "flos": 24170646764160.0,
      "grad_norm": 1.9286932427393801,
      "language_loss": 0.7203868,
      "learning_rate": 1.045303157347638e-06,
      "loss": 0.74146557,
      "num_input_tokens_seen": 240143870,
      "step": 11121,
      "time_per_iteration": 2.589672327041626
    },
    {
      "auxiliary_loss_clip": 0.01087448,
      "auxiliary_loss_mlp": 0.01036959,
      "balance_loss_clip": 1.03637934,
      "balance_loss_mlp": 1.02428746,
      "epoch": 0.668690816173155,
      "flos": 17457147181440.0,
      "grad_norm": 2.592174320613198,
      "language_loss": 0.69477034,
      "learning_rate": 1.0449609499766316e-06,
      "loss": 0.71601439,
      "num_input_tokens_seen": 240161020,
      "step": 11122,
      "time_per_iteration": 2.504709482192993
    },
    {
      "auxiliary_loss_clip": 0.01050187,
      "auxiliary_loss_mlp": 0.00788142,
      "balance_loss_clip": 1.03421211,
      "balance_loss_mlp": 1.01366878,
      "epoch": 0.668750939425823,
      "flos": 25005102595200.0,
      "grad_norm": 2.153701545684978,
      "language_loss": 0.71475536,
      "learning_rate": 1.0446187788208015e-06,
      "loss": 0.73313862,
      "num_input_tokens_seen": 240179820,
      "step": 11123,
      "time_per_iteration": 2.652235746383667
    },
    {
      "auxiliary_loss_clip": 0.01092825,
      "auxiliary_loss_mlp": 0.01033687,
      "balance_loss_clip": 1.04074097,
      "balance_loss_mlp": 1.02096784,
      "epoch": 0.6688110626784909,
      "flos": 24096778444800.0,
      "grad_norm": 1.854922254012912,
      "language_loss": 0.79357159,
      "learning_rate": 1.0442766438931244e-06,
      "loss": 0.81483674,
      "num_input_tokens_seen": 240200130,
      "step": 11124,
      "time_per_iteration": 2.5575473308563232
    },
    {
      "auxiliary_loss_clip": 0.01086132,
      "auxiliary_loss_mlp": 0.01039241,
      "balance_loss_clip": 1.04114175,
      "balance_loss_mlp": 1.02678323,
      "epoch": 0.6688711859311589,
      "flos": 21759532375680.0,
      "grad_norm": 1.7887270666401778,
      "language_loss": 0.74375939,
      "learning_rate": 1.0439345452065716e-06,
      "loss": 0.7650131,
      "num_input_tokens_seen": 240217945,
      "step": 11125,
      "time_per_iteration": 3.960653066635132
    },
    {
      "auxiliary_loss_clip": 0.01082429,
      "auxiliary_loss_mlp": 0.01033705,
      "balance_loss_clip": 1.04057789,
      "balance_loss_mlp": 1.02084816,
      "epoch": 0.6689313091838268,
      "flos": 22929645824640.0,
      "grad_norm": 1.7680983525976157,
      "language_loss": 0.6638881,
      "learning_rate": 1.043592482774116e-06,
      "loss": 0.68504941,
      "num_input_tokens_seen": 240237220,
      "step": 11126,
      "time_per_iteration": 3.9651849269866943
    },
    {
      "auxiliary_loss_clip": 0.01090915,
      "auxiliary_loss_mlp": 0.01027097,
      "balance_loss_clip": 1.03428817,
      "balance_loss_mlp": 1.01454997,
      "epoch": 0.6689914324364948,
      "flos": 20886149180160.0,
      "grad_norm": 1.79171889965827,
      "language_loss": 0.70812583,
      "learning_rate": 1.0432504566087305e-06,
      "loss": 0.72930592,
      "num_input_tokens_seen": 240256000,
      "step": 11127,
      "time_per_iteration": 2.5190625190734863
    },
    {
      "auxiliary_loss_clip": 0.01090194,
      "auxiliary_loss_mlp": 0.01032342,
      "balance_loss_clip": 1.03724146,
      "balance_loss_mlp": 1.01904416,
      "epoch": 0.6690515556891627,
      "flos": 22748225207040.0,
      "grad_norm": 1.8406556428845449,
      "language_loss": 0.80339694,
      "learning_rate": 1.0429084667233827e-06,
      "loss": 0.82462233,
      "num_input_tokens_seen": 240275845,
      "step": 11128,
      "time_per_iteration": 3.926719903945923
    },
    {
      "auxiliary_loss_clip": 0.01110719,
      "auxiliary_loss_mlp": 0.01031969,
      "balance_loss_clip": 1.03837144,
      "balance_loss_mlp": 1.01971459,
      "epoch": 0.6691116789418308,
      "flos": 23331450337920.0,
      "grad_norm": 1.8637335035113767,
      "language_loss": 0.81007028,
      "learning_rate": 1.0425665131310427e-06,
      "loss": 0.83149713,
      "num_input_tokens_seen": 240294095,
      "step": 11129,
      "time_per_iteration": 2.5060718059539795
    },
    {
      "auxiliary_loss_clip": 0.0108876,
      "auxiliary_loss_mlp": 0.01033604,
      "balance_loss_clip": 1.03524077,
      "balance_loss_mlp": 1.02204108,
      "epoch": 0.6691718021944987,
      "flos": 32447014081920.0,
      "grad_norm": 1.5203831596751347,
      "language_loss": 0.70229268,
      "learning_rate": 1.0422245958446762e-06,
      "loss": 0.72351635,
      "num_input_tokens_seen": 240313460,
      "step": 11130,
      "time_per_iteration": 2.587913751602173
    },
    {
      "auxiliary_loss_clip": 0.01084579,
      "auxiliary_loss_mlp": 0.0103518,
      "balance_loss_clip": 1.03711021,
      "balance_loss_mlp": 1.023265,
      "epoch": 0.6692319254471667,
      "flos": 23731602825600.0,
      "grad_norm": 1.7452408168586244,
      "language_loss": 0.70099163,
      "learning_rate": 1.0418827148772486e-06,
      "loss": 0.72218925,
      "num_input_tokens_seen": 240333540,
      "step": 11131,
      "time_per_iteration": 2.5660438537597656
    },
    {
      "auxiliary_loss_clip": 0.01099497,
      "auxiliary_loss_mlp": 0.01031695,
      "balance_loss_clip": 1.03750181,
      "balance_loss_mlp": 1.01809371,
      "epoch": 0.6692920486998346,
      "flos": 14427902620800.0,
      "grad_norm": 2.3591237404501735,
      "language_loss": 0.65898609,
      "learning_rate": 1.0415408702417243e-06,
      "loss": 0.68029797,
      "num_input_tokens_seen": 240350085,
      "step": 11132,
      "time_per_iteration": 2.4764678478240967
    },
    {
      "auxiliary_loss_clip": 0.0109867,
      "auxiliary_loss_mlp": 0.01033902,
      "balance_loss_clip": 1.03604078,
      "balance_loss_mlp": 1.02032971,
      "epoch": 0.6693521719525026,
      "flos": 21507475662720.0,
      "grad_norm": 1.7313026276371768,
      "language_loss": 0.74369109,
      "learning_rate": 1.0411990619510661e-06,
      "loss": 0.76501679,
      "num_input_tokens_seen": 240370015,
      "step": 11133,
      "time_per_iteration": 2.511470317840576
    },
    {
      "auxiliary_loss_clip": 0.01104507,
      "auxiliary_loss_mlp": 0.01034093,
      "balance_loss_clip": 1.04059196,
      "balance_loss_mlp": 1.02056313,
      "epoch": 0.6694122952051706,
      "flos": 25406943022080.0,
      "grad_norm": 1.7945153149203978,
      "language_loss": 0.66560668,
      "learning_rate": 1.0408572900182363e-06,
      "loss": 0.68699265,
      "num_input_tokens_seen": 240390770,
      "step": 11134,
      "time_per_iteration": 2.5532233715057373
    },
    {
      "auxiliary_loss_clip": 0.01103437,
      "auxiliary_loss_mlp": 0.01034132,
      "balance_loss_clip": 1.04024124,
      "balance_loss_mlp": 1.01973701,
      "epoch": 0.6694724184578386,
      "flos": 25661729168640.0,
      "grad_norm": 1.9922334018781445,
      "language_loss": 0.77483219,
      "learning_rate": 1.0405155544561943e-06,
      "loss": 0.7962079,
      "num_input_tokens_seen": 240409590,
      "step": 11135,
      "time_per_iteration": 2.5475106239318848
    },
    {
      "auxiliary_loss_clip": 0.01097916,
      "auxiliary_loss_mlp": 0.0102759,
      "balance_loss_clip": 1.0381211,
      "balance_loss_mlp": 1.01518035,
      "epoch": 0.6695325417105066,
      "flos": 17709311635200.0,
      "grad_norm": 1.6280874162001817,
      "language_loss": 0.74363661,
      "learning_rate": 1.040173855277898e-06,
      "loss": 0.76489168,
      "num_input_tokens_seen": 240428180,
      "step": 11136,
      "time_per_iteration": 2.4975924491882324
    },
    {
      "auxiliary_loss_clip": 0.01103267,
      "auxiliary_loss_mlp": 0.01031744,
      "balance_loss_clip": 1.04021764,
      "balance_loss_mlp": 1.01864886,
      "epoch": 0.6695926649631745,
      "flos": 24460050643200.0,
      "grad_norm": 1.6470798393037906,
      "language_loss": 0.62345713,
      "learning_rate": 1.0398321924963061e-06,
      "loss": 0.64480722,
      "num_input_tokens_seen": 240447815,
      "step": 11137,
      "time_per_iteration": 2.535275459289551
    },
    {
      "auxiliary_loss_clip": 0.01108315,
      "auxiliary_loss_mlp": 0.01030459,
      "balance_loss_clip": 1.03794074,
      "balance_loss_mlp": 1.01772773,
      "epoch": 0.6696527882158425,
      "flos": 24280138396800.0,
      "grad_norm": 1.7154699088642178,
      "language_loss": 0.65805918,
      "learning_rate": 1.0394905661243724e-06,
      "loss": 0.67944688,
      "num_input_tokens_seen": 240468635,
      "step": 11138,
      "time_per_iteration": 2.4959747791290283
    },
    {
      "auxiliary_loss_clip": 0.01075254,
      "auxiliary_loss_mlp": 0.0103293,
      "balance_loss_clip": 1.03577018,
      "balance_loss_mlp": 1.02035308,
      "epoch": 0.6697129114685104,
      "flos": 23002759958400.0,
      "grad_norm": 1.6244088113888004,
      "language_loss": 0.72562897,
      "learning_rate": 1.039148976175053e-06,
      "loss": 0.74671084,
      "num_input_tokens_seen": 240488550,
      "step": 11139,
      "time_per_iteration": 2.5799169540405273
    },
    {
      "auxiliary_loss_clip": 0.01069867,
      "auxiliary_loss_mlp": 0.01030705,
      "balance_loss_clip": 1.03786778,
      "balance_loss_mlp": 1.01923132,
      "epoch": 0.6697730347211784,
      "flos": 22638123043200.0,
      "grad_norm": 1.9194515136399346,
      "language_loss": 0.70211923,
      "learning_rate": 1.0388074226613016e-06,
      "loss": 0.72312498,
      "num_input_tokens_seen": 240508330,
      "step": 11140,
      "time_per_iteration": 2.6189382076263428
    },
    {
      "auxiliary_loss_clip": 0.01099209,
      "auxiliary_loss_mlp": 0.01027442,
      "balance_loss_clip": 1.03550708,
      "balance_loss_mlp": 1.01453161,
      "epoch": 0.6698331579738463,
      "flos": 28877242682880.0,
      "grad_norm": 1.9749737472382,
      "language_loss": 0.75714004,
      "learning_rate": 1.0384659055960691e-06,
      "loss": 0.77840656,
      "num_input_tokens_seen": 240528470,
      "step": 11141,
      "time_per_iteration": 2.5399861335754395
    },
    {
      "auxiliary_loss_clip": 0.01099553,
      "auxiliary_loss_mlp": 0.01032607,
      "balance_loss_clip": 1.03879547,
      "balance_loss_mlp": 1.01982749,
      "epoch": 0.6698932812265144,
      "flos": 24207096090240.0,
      "grad_norm": 1.7954899902479762,
      "language_loss": 0.81750733,
      "learning_rate": 1.0381244249923052e-06,
      "loss": 0.83882892,
      "num_input_tokens_seen": 240547815,
      "step": 11142,
      "time_per_iteration": 2.5337319374084473
    },
    {
      "auxiliary_loss_clip": 0.01061848,
      "auxiliary_loss_mlp": 0.01028272,
      "balance_loss_clip": 1.03777277,
      "balance_loss_mlp": 1.01543319,
      "epoch": 0.6699534044791823,
      "flos": 22090269830400.0,
      "grad_norm": 1.5050005388668242,
      "language_loss": 0.69980276,
      "learning_rate": 1.037782980862959e-06,
      "loss": 0.72070396,
      "num_input_tokens_seen": 240567765,
      "step": 11143,
      "time_per_iteration": 2.594655752182007
    },
    {
      "auxiliary_loss_clip": 0.01068609,
      "auxiliary_loss_mlp": 0.00787856,
      "balance_loss_clip": 1.03652573,
      "balance_loss_mlp": 1.01380205,
      "epoch": 0.6700135277318503,
      "flos": 25192377129600.0,
      "grad_norm": 1.4494566555493236,
      "language_loss": 0.70115811,
      "learning_rate": 1.0374415732209796e-06,
      "loss": 0.71972275,
      "num_input_tokens_seen": 240590750,
      "step": 11144,
      "time_per_iteration": 2.6190598011016846
    },
    {
      "auxiliary_loss_clip": 0.01084731,
      "auxiliary_loss_mlp": 0.01032379,
      "balance_loss_clip": 1.03690815,
      "balance_loss_mlp": 1.01941538,
      "epoch": 0.6700736509845182,
      "flos": 23440187784960.0,
      "grad_norm": 1.7250605210360679,
      "language_loss": 0.74402547,
      "learning_rate": 1.0371002020793114e-06,
      "loss": 0.76519656,
      "num_input_tokens_seen": 240608875,
      "step": 11145,
      "time_per_iteration": 2.546212673187256
    },
    {
      "auxiliary_loss_clip": 0.01088137,
      "auxiliary_loss_mlp": 0.01029175,
      "balance_loss_clip": 1.03650618,
      "balance_loss_mlp": 1.01631868,
      "epoch": 0.6701337742371862,
      "flos": 24389953251840.0,
      "grad_norm": 1.5400994755234974,
      "language_loss": 0.70848072,
      "learning_rate": 1.0367588674509008e-06,
      "loss": 0.72965389,
      "num_input_tokens_seen": 240628565,
      "step": 11146,
      "time_per_iteration": 2.559746026992798
    },
    {
      "auxiliary_loss_clip": 0.01103769,
      "auxiliary_loss_mlp": 0.00784546,
      "balance_loss_clip": 1.03633714,
      "balance_loss_mlp": 1.01254535,
      "epoch": 0.6701938974898543,
      "flos": 14793652857600.0,
      "grad_norm": 2.2118845348916896,
      "language_loss": 0.78795528,
      "learning_rate": 1.0364175693486905e-06,
      "loss": 0.80683845,
      "num_input_tokens_seen": 240646325,
      "step": 11147,
      "time_per_iteration": 2.4373953342437744
    },
    {
      "auxiliary_loss_clip": 0.01099769,
      "auxiliary_loss_mlp": 0.00786169,
      "balance_loss_clip": 1.03881836,
      "balance_loss_mlp": 1.01124632,
      "epoch": 0.6702540207425222,
      "flos": 20154002261760.0,
      "grad_norm": 1.669961269119579,
      "language_loss": 0.70442593,
      "learning_rate": 1.0360763077856218e-06,
      "loss": 0.72328532,
      "num_input_tokens_seen": 240666145,
      "step": 11148,
      "time_per_iteration": 2.5195634365081787
    },
    {
      "auxiliary_loss_clip": 0.01089226,
      "auxiliary_loss_mlp": 0.01037473,
      "balance_loss_clip": 1.03607345,
      "balance_loss_mlp": 1.02494454,
      "epoch": 0.6703141439951902,
      "flos": 21214157201280.0,
      "grad_norm": 1.6725129498581126,
      "language_loss": 0.70427907,
      "learning_rate": 1.035735082774636e-06,
      "loss": 0.72554606,
      "num_input_tokens_seen": 240685570,
      "step": 11149,
      "time_per_iteration": 2.5429177284240723
    },
    {
      "auxiliary_loss_clip": 0.0108917,
      "auxiliary_loss_mlp": 0.01031564,
      "balance_loss_clip": 1.037588,
      "balance_loss_mlp": 1.01982152,
      "epoch": 0.6703742672478581,
      "flos": 23112538899840.0,
      "grad_norm": 2.117238624376736,
      "language_loss": 0.73890287,
      "learning_rate": 1.0353938943286727e-06,
      "loss": 0.7601102,
      "num_input_tokens_seen": 240706945,
      "step": 11150,
      "time_per_iteration": 2.600705146789551
    },
    {
      "auxiliary_loss_clip": 0.01099246,
      "auxiliary_loss_mlp": 0.01031402,
      "balance_loss_clip": 1.04018855,
      "balance_loss_mlp": 1.01926064,
      "epoch": 0.6704343905005261,
      "flos": 22528918719360.0,
      "grad_norm": 1.809909937030258,
      "language_loss": 0.78517437,
      "learning_rate": 1.035052742460671e-06,
      "loss": 0.80648088,
      "num_input_tokens_seen": 240727990,
      "step": 11151,
      "time_per_iteration": 2.535999298095703
    },
    {
      "auxiliary_loss_clip": 0.01010281,
      "auxiliary_loss_mlp": 0.01002854,
      "balance_loss_clip": 1.0277344,
      "balance_loss_mlp": 1.00166786,
      "epoch": 0.670494513753194,
      "flos": 64793158773120.0,
      "grad_norm": 0.821624527455594,
      "language_loss": 0.55449015,
      "learning_rate": 1.0347116271835643e-06,
      "loss": 0.5746215,
      "num_input_tokens_seen": 240790380,
      "step": 11152,
      "time_per_iteration": 3.3057069778442383
    },
    {
      "auxiliary_loss_clip": 0.01083942,
      "auxiliary_loss_mlp": 0.01035123,
      "balance_loss_clip": 1.03544688,
      "balance_loss_mlp": 1.0226357,
      "epoch": 0.670554637005862,
      "flos": 23511506238720.0,
      "grad_norm": 1.7718278081132208,
      "language_loss": 0.80944306,
      "learning_rate": 1.0343705485102896e-06,
      "loss": 0.8306337,
      "num_input_tokens_seen": 240811545,
      "step": 11153,
      "time_per_iteration": 2.570666551589966
    },
    {
      "auxiliary_loss_clip": 0.0107097,
      "auxiliary_loss_mlp": 0.007849,
      "balance_loss_clip": 1.03694868,
      "balance_loss_mlp": 1.00856972,
      "epoch": 0.67061476025853,
      "flos": 19463404400640.0,
      "grad_norm": 1.4533893613368856,
      "language_loss": 0.76416385,
      "learning_rate": 1.0340295064537814e-06,
      "loss": 0.78272259,
      "num_input_tokens_seen": 240831380,
      "step": 11154,
      "time_per_iteration": 2.573660135269165
    },
    {
      "auxiliary_loss_clip": 0.01090807,
      "auxiliary_loss_mlp": 0.01035316,
      "balance_loss_clip": 1.03838861,
      "balance_loss_mlp": 1.02220321,
      "epoch": 0.670674883511198,
      "flos": 20519967980160.0,
      "grad_norm": 1.5533702824568059,
      "language_loss": 0.76374054,
      "learning_rate": 1.0336885010269702e-06,
      "loss": 0.78500175,
      "num_input_tokens_seen": 240851855,
      "step": 11155,
      "time_per_iteration": 2.5386762619018555
    },
    {
      "auxiliary_loss_clip": 0.01111646,
      "auxiliary_loss_mlp": 0.01036989,
      "balance_loss_clip": 1.04007804,
      "balance_loss_mlp": 1.024418,
      "epoch": 0.6707350067638659,
      "flos": 25483971738240.0,
      "grad_norm": 2.0616804940236624,
      "language_loss": 0.81810671,
      "learning_rate": 1.0333475322427878e-06,
      "loss": 0.83959305,
      "num_input_tokens_seen": 240869980,
      "step": 11156,
      "time_per_iteration": 2.491286277770996
    },
    {
      "auxiliary_loss_clip": 0.01107788,
      "auxiliary_loss_mlp": 0.01029687,
      "balance_loss_clip": 1.03756714,
      "balance_loss_mlp": 1.01793313,
      "epoch": 0.6707951300165339,
      "flos": 22273450214400.0,
      "grad_norm": 2.452254137979995,
      "language_loss": 0.7468152,
      "learning_rate": 1.033006600114165e-06,
      "loss": 0.76819003,
      "num_input_tokens_seen": 240888680,
      "step": 11157,
      "time_per_iteration": 3.949521780014038
    },
    {
      "auxiliary_loss_clip": 0.01102319,
      "auxiliary_loss_mlp": 0.010354,
      "balance_loss_clip": 1.03972936,
      "balance_loss_mlp": 1.02232265,
      "epoch": 0.6708552532692018,
      "flos": 23984593292160.0,
      "grad_norm": 1.5780989958744807,
      "language_loss": 0.74441206,
      "learning_rate": 1.0326657046540282e-06,
      "loss": 0.76578927,
      "num_input_tokens_seen": 240909050,
      "step": 11158,
      "time_per_iteration": 2.5311756134033203
    },
    {
      "auxiliary_loss_clip": 0.01113754,
      "auxiliary_loss_mlp": 0.01032582,
      "balance_loss_clip": 1.04026949,
      "balance_loss_mlp": 1.0197196,
      "epoch": 0.6709153765218698,
      "flos": 24937519155840.0,
      "grad_norm": 1.629136663987161,
      "language_loss": 0.81752515,
      "learning_rate": 1.0323248458753044e-06,
      "loss": 0.83898854,
      "num_input_tokens_seen": 240930035,
      "step": 11159,
      "time_per_iteration": 2.5144901275634766
    },
    {
      "auxiliary_loss_clip": 0.0108846,
      "auxiliary_loss_mlp": 0.01031277,
      "balance_loss_clip": 1.03810954,
      "balance_loss_mlp": 1.01912379,
      "epoch": 0.6709754997745379,
      "flos": 17530225401600.0,
      "grad_norm": 1.690002783568212,
      "language_loss": 0.76928449,
      "learning_rate": 1.0319840237909193e-06,
      "loss": 0.79048181,
      "num_input_tokens_seen": 240948895,
      "step": 11160,
      "time_per_iteration": 2.541985273361206
    },
    {
      "auxiliary_loss_clip": 0.01082295,
      "auxiliary_loss_mlp": 0.01029282,
      "balance_loss_clip": 1.03641176,
      "balance_loss_mlp": 1.01755214,
      "epoch": 0.6710356230272058,
      "flos": 22090880361600.0,
      "grad_norm": 2.1949819872839424,
      "language_loss": 0.73530984,
      "learning_rate": 1.0316432384137978e-06,
      "loss": 0.75642562,
      "num_input_tokens_seen": 240967770,
      "step": 11161,
      "time_per_iteration": 2.546900987625122
    },
    {
      "auxiliary_loss_clip": 0.01089946,
      "auxiliary_loss_mlp": 0.01037883,
      "balance_loss_clip": 1.03725076,
      "balance_loss_mlp": 1.02448392,
      "epoch": 0.6710957462798738,
      "flos": 24206449645440.0,
      "grad_norm": 1.7251392150168565,
      "language_loss": 0.68347704,
      "learning_rate": 1.0313024897568618e-06,
      "loss": 0.70475537,
      "num_input_tokens_seen": 240988985,
      "step": 11162,
      "time_per_iteration": 2.573518753051758
    },
    {
      "auxiliary_loss_clip": 0.01084979,
      "auxiliary_loss_mlp": 0.01037204,
      "balance_loss_clip": 1.03518832,
      "balance_loss_mlp": 1.0250144,
      "epoch": 0.6711558695325417,
      "flos": 19093955063040.0,
      "grad_norm": 1.7983279135850545,
      "language_loss": 0.70118272,
      "learning_rate": 1.030961777833032e-06,
      "loss": 0.72240454,
      "num_input_tokens_seen": 241005455,
      "step": 11163,
      "time_per_iteration": 2.50148868560791
    },
    {
      "auxiliary_loss_clip": 0.01108389,
      "auxiliary_loss_mlp": 0.01030085,
      "balance_loss_clip": 1.04009414,
      "balance_loss_mlp": 1.0182122,
      "epoch": 0.6712159927852097,
      "flos": 25557875971200.0,
      "grad_norm": 1.7311372854804425,
      "language_loss": 0.75485975,
      "learning_rate": 1.0306211026552291e-06,
      "loss": 0.77624452,
      "num_input_tokens_seen": 241026175,
      "step": 11164,
      "time_per_iteration": 4.046183347702026
    },
    {
      "auxiliary_loss_clip": 0.01108915,
      "auxiliary_loss_mlp": 0.01033658,
      "balance_loss_clip": 1.03876221,
      "balance_loss_mlp": 1.02166545,
      "epoch": 0.6712761160378776,
      "flos": 22228812587520.0,
      "grad_norm": 2.891102490826458,
      "language_loss": 0.64820254,
      "learning_rate": 1.0302804642363704e-06,
      "loss": 0.66962826,
      "num_input_tokens_seen": 241044040,
      "step": 11165,
      "time_per_iteration": 3.8313703536987305
    },
    {
      "auxiliary_loss_clip": 0.0110735,
      "auxiliary_loss_mlp": 0.01031737,
      "balance_loss_clip": 1.03859735,
      "balance_loss_mlp": 1.01962554,
      "epoch": 0.6713362392905456,
      "flos": 22455517276800.0,
      "grad_norm": 1.8073188172542927,
      "language_loss": 0.71737725,
      "learning_rate": 1.0299398625893738e-06,
      "loss": 0.7387681,
      "num_input_tokens_seen": 241063615,
      "step": 11166,
      "time_per_iteration": 2.4835660457611084
    },
    {
      "auxiliary_loss_clip": 0.01107072,
      "auxiliary_loss_mlp": 0.0102923,
      "balance_loss_clip": 1.03940558,
      "balance_loss_mlp": 1.01815522,
      "epoch": 0.6713963625432136,
      "flos": 25630200005760.0,
      "grad_norm": 1.832026511197579,
      "language_loss": 0.77227235,
      "learning_rate": 1.0295992977271546e-06,
      "loss": 0.79363537,
      "num_input_tokens_seen": 241082520,
      "step": 11167,
      "time_per_iteration": 3.973959445953369
    },
    {
      "auxiliary_loss_clip": 0.01096591,
      "auxiliary_loss_mlp": 0.01036114,
      "balance_loss_clip": 1.03569508,
      "balance_loss_mlp": 1.02395487,
      "epoch": 0.6714564857958816,
      "flos": 35006475640320.0,
      "grad_norm": 1.7576613324708539,
      "language_loss": 0.69122016,
      "learning_rate": 1.029258769662629e-06,
      "loss": 0.71254718,
      "num_input_tokens_seen": 241103505,
      "step": 11168,
      "time_per_iteration": 2.6422994136810303
    },
    {
      "auxiliary_loss_clip": 0.01079879,
      "auxiliary_loss_mlp": 0.01040955,
      "balance_loss_clip": 1.03839016,
      "balance_loss_mlp": 1.02774119,
      "epoch": 0.6715166090485495,
      "flos": 26279931168000.0,
      "grad_norm": 2.07282971909186,
      "language_loss": 0.73275113,
      "learning_rate": 1.0289182784087068e-06,
      "loss": 0.75395942,
      "num_input_tokens_seen": 241122885,
      "step": 11169,
      "time_per_iteration": 2.6094915866851807
    },
    {
      "auxiliary_loss_clip": 0.01102115,
      "auxiliary_loss_mlp": 0.01034457,
      "balance_loss_clip": 1.03890109,
      "balance_loss_mlp": 1.02139151,
      "epoch": 0.6715767323012175,
      "flos": 15924156583680.0,
      "grad_norm": 2.0675575403293234,
      "language_loss": 0.76189041,
      "learning_rate": 1.0285778239783005e-06,
      "loss": 0.78325611,
      "num_input_tokens_seen": 241140865,
      "step": 11170,
      "time_per_iteration": 2.482563018798828
    },
    {
      "auxiliary_loss_clip": 0.01089138,
      "auxiliary_loss_mlp": 0.01028099,
      "balance_loss_clip": 1.03741002,
      "balance_loss_mlp": 1.01529014,
      "epoch": 0.6716368555538854,
      "flos": 17491441691520.0,
      "grad_norm": 2.003326158799613,
      "language_loss": 0.74310929,
      "learning_rate": 1.0282374063843212e-06,
      "loss": 0.76428163,
      "num_input_tokens_seen": 241158225,
      "step": 11171,
      "time_per_iteration": 2.535719871520996
    },
    {
      "auxiliary_loss_clip": 0.01069851,
      "auxiliary_loss_mlp": 0.01041668,
      "balance_loss_clip": 1.03620553,
      "balance_loss_mlp": 1.027601,
      "epoch": 0.6716969788065534,
      "flos": 16761521416320.0,
      "grad_norm": 1.5155629790740002,
      "language_loss": 0.86344361,
      "learning_rate": 1.0278970256396762e-06,
      "loss": 0.8845588,
      "num_input_tokens_seen": 241175215,
      "step": 11172,
      "time_per_iteration": 2.5671091079711914
    },
    {
      "auxiliary_loss_clip": 0.01094401,
      "auxiliary_loss_mlp": 0.01033697,
      "balance_loss_clip": 1.03556895,
      "balance_loss_mlp": 1.02127504,
      "epoch": 0.6717571020592215,
      "flos": 22709800632960.0,
      "grad_norm": 1.655590981767189,
      "language_loss": 0.63643694,
      "learning_rate": 1.0275566817572733e-06,
      "loss": 0.65771782,
      "num_input_tokens_seen": 241195250,
      "step": 11173,
      "time_per_iteration": 2.5429654121398926
    },
    {
      "auxiliary_loss_clip": 0.01104758,
      "auxiliary_loss_mlp": 0.01035595,
      "balance_loss_clip": 1.03741825,
      "balance_loss_mlp": 1.02170753,
      "epoch": 0.6718172253118894,
      "flos": 18734094656640.0,
      "grad_norm": 2.294228763400917,
      "language_loss": 0.71856713,
      "learning_rate": 1.02721637475002e-06,
      "loss": 0.73997062,
      "num_input_tokens_seen": 241210720,
      "step": 11174,
      "time_per_iteration": 2.4690868854522705
    },
    {
      "auxiliary_loss_clip": 0.0107672,
      "auxiliary_loss_mlp": 0.01026719,
      "balance_loss_clip": 1.03739667,
      "balance_loss_mlp": 1.01505446,
      "epoch": 0.6718773485645574,
      "flos": 15632526061440.0,
      "grad_norm": 1.9065860058608657,
      "language_loss": 0.69085479,
      "learning_rate": 1.0268761046308178e-06,
      "loss": 0.71188927,
      "num_input_tokens_seen": 241227395,
      "step": 11175,
      "time_per_iteration": 2.550529718399048
    },
    {
      "auxiliary_loss_clip": 0.01087032,
      "auxiliary_loss_mlp": 0.01034059,
      "balance_loss_clip": 1.03828883,
      "balance_loss_mlp": 1.02229929,
      "epoch": 0.6719374718172253,
      "flos": 19354774694400.0,
      "grad_norm": 3.1871664522555108,
      "language_loss": 0.73600304,
      "learning_rate": 1.0265358714125714e-06,
      "loss": 0.75721395,
      "num_input_tokens_seen": 241246355,
      "step": 11176,
      "time_per_iteration": 2.521174192428589
    },
    {
      "auxiliary_loss_clip": 0.01086752,
      "auxiliary_loss_mlp": 0.01030983,
      "balance_loss_clip": 1.03646684,
      "balance_loss_mlp": 1.01799536,
      "epoch": 0.6719975950698933,
      "flos": 21981316901760.0,
      "grad_norm": 1.6028528203342292,
      "language_loss": 0.73358011,
      "learning_rate": 1.026195675108182e-06,
      "loss": 0.75475746,
      "num_input_tokens_seen": 241264180,
      "step": 11177,
      "time_per_iteration": 2.5350260734558105
    },
    {
      "auxiliary_loss_clip": 0.0110925,
      "auxiliary_loss_mlp": 0.01031259,
      "balance_loss_clip": 1.03796065,
      "balance_loss_mlp": 1.01836669,
      "epoch": 0.6720577183225612,
      "flos": 25228072270080.0,
      "grad_norm": 2.1925361146129725,
      "language_loss": 0.76340181,
      "learning_rate": 1.025855515730551e-06,
      "loss": 0.78480685,
      "num_input_tokens_seen": 241282245,
      "step": 11178,
      "time_per_iteration": 2.5118415355682373
    },
    {
      "auxiliary_loss_clip": 0.01101523,
      "auxiliary_loss_mlp": 0.01033525,
      "balance_loss_clip": 1.04026687,
      "balance_loss_mlp": 1.02184832,
      "epoch": 0.6721178415752292,
      "flos": 16945886949120.0,
      "grad_norm": 1.7084218941072153,
      "language_loss": 0.70352155,
      "learning_rate": 1.0255153932925766e-06,
      "loss": 0.72487199,
      "num_input_tokens_seen": 241300745,
      "step": 11179,
      "time_per_iteration": 2.4907569885253906
    },
    {
      "auxiliary_loss_clip": 0.01063802,
      "auxiliary_loss_mlp": 0.01027477,
      "balance_loss_clip": 1.03762841,
      "balance_loss_mlp": 1.01594996,
      "epoch": 0.6721779648278972,
      "flos": 21541375123200.0,
      "grad_norm": 1.7765959475979738,
      "language_loss": 0.73905778,
      "learning_rate": 1.0251753078071557e-06,
      "loss": 0.75997061,
      "num_input_tokens_seen": 241319320,
      "step": 11180,
      "time_per_iteration": 2.6004762649536133
    },
    {
      "auxiliary_loss_clip": 0.01088491,
      "auxiliary_loss_mlp": 0.01029591,
      "balance_loss_clip": 1.03749859,
      "balance_loss_mlp": 1.01752067,
      "epoch": 0.6722380880805652,
      "flos": 22605444645120.0,
      "grad_norm": 1.390113056655007,
      "language_loss": 0.75353789,
      "learning_rate": 1.0248352592871848e-06,
      "loss": 0.7747187,
      "num_input_tokens_seen": 241342225,
      "step": 11181,
      "time_per_iteration": 2.5657315254211426
    },
    {
      "auxiliary_loss_clip": 0.01088386,
      "auxiliary_loss_mlp": 0.01026605,
      "balance_loss_clip": 1.03632402,
      "balance_loss_mlp": 1.01505351,
      "epoch": 0.6722982113332331,
      "flos": 15925269905280.0,
      "grad_norm": 2.1507330573896506,
      "language_loss": 0.7449019,
      "learning_rate": 1.0244952477455585e-06,
      "loss": 0.76605177,
      "num_input_tokens_seen": 241358240,
      "step": 11182,
      "time_per_iteration": 2.4705393314361572
    },
    {
      "auxiliary_loss_clip": 0.01093469,
      "auxiliary_loss_mlp": 0.01030924,
      "balance_loss_clip": 1.03639483,
      "balance_loss_mlp": 1.01946795,
      "epoch": 0.6723583345859011,
      "flos": 20596170683520.0,
      "grad_norm": 2.140643625195713,
      "language_loss": 0.69452417,
      "learning_rate": 1.0241552731951699e-06,
      "loss": 0.7157681,
      "num_input_tokens_seen": 241378420,
      "step": 11183,
      "time_per_iteration": 2.518209934234619
    },
    {
      "auxiliary_loss_clip": 0.01061801,
      "auxiliary_loss_mlp": 0.01032915,
      "balance_loss_clip": 1.03899431,
      "balance_loss_mlp": 1.02020192,
      "epoch": 0.672418457838569,
      "flos": 21725848396800.0,
      "grad_norm": 1.6451151439031801,
      "language_loss": 0.77900374,
      "learning_rate": 1.0238153356489112e-06,
      "loss": 0.79995084,
      "num_input_tokens_seen": 241397185,
      "step": 11184,
      "time_per_iteration": 2.590519905090332
    },
    {
      "auxiliary_loss_clip": 0.01091958,
      "auxiliary_loss_mlp": 0.00789843,
      "balance_loss_clip": 1.04020548,
      "balance_loss_mlp": 1.01178253,
      "epoch": 0.672478581091237,
      "flos": 21470379891840.0,
      "grad_norm": 1.9979396566701118,
      "language_loss": 0.66266799,
      "learning_rate": 1.0234754351196743e-06,
      "loss": 0.68148601,
      "num_input_tokens_seen": 241415785,
      "step": 11185,
      "time_per_iteration": 2.534670829772949
    },
    {
      "auxiliary_loss_clip": 0.01073227,
      "auxiliary_loss_mlp": 0.01032925,
      "balance_loss_clip": 1.0356158,
      "balance_loss_mlp": 1.01966286,
      "epoch": 0.6725387043439051,
      "flos": 30846763267200.0,
      "grad_norm": 1.6997577867553462,
      "language_loss": 0.80412382,
      "learning_rate": 1.023135571620345e-06,
      "loss": 0.82518536,
      "num_input_tokens_seen": 241437390,
      "step": 11186,
      "time_per_iteration": 2.617849349975586
    },
    {
      "auxiliary_loss_clip": 0.0109758,
      "auxiliary_loss_mlp": 0.0103274,
      "balance_loss_clip": 1.03992486,
      "balance_loss_mlp": 1.02151608,
      "epoch": 0.672598827596573,
      "flos": 24055947659520.0,
      "grad_norm": 1.434669984834181,
      "language_loss": 0.79968005,
      "learning_rate": 1.022795745163813e-06,
      "loss": 0.82098323,
      "num_input_tokens_seen": 241458085,
      "step": 11187,
      "time_per_iteration": 2.5392544269561768
    },
    {
      "auxiliary_loss_clip": 0.01076437,
      "auxiliary_loss_mlp": 0.01030686,
      "balance_loss_clip": 1.04419088,
      "balance_loss_mlp": 1.0173707,
      "epoch": 0.672658950849241,
      "flos": 21871861182720.0,
      "grad_norm": 1.7464528252004903,
      "language_loss": 0.70637846,
      "learning_rate": 1.022455955762965e-06,
      "loss": 0.72744966,
      "num_input_tokens_seen": 241476880,
      "step": 11188,
      "time_per_iteration": 2.593278646469116
    },
    {
      "auxiliary_loss_clip": 0.01044573,
      "auxiliary_loss_mlp": 0.01030197,
      "balance_loss_clip": 1.03801847,
      "balance_loss_mlp": 1.0186162,
      "epoch": 0.6727190741019089,
      "flos": 23222102359680.0,
      "grad_norm": 1.6492729140266218,
      "language_loss": 0.75602478,
      "learning_rate": 1.0221162034306842e-06,
      "loss": 0.7767725,
      "num_input_tokens_seen": 241496535,
      "step": 11189,
      "time_per_iteration": 2.880229949951172
    },
    {
      "auxiliary_loss_clip": 0.01111935,
      "auxiliary_loss_mlp": 0.01030487,
      "balance_loss_clip": 1.03764439,
      "balance_loss_mlp": 1.01632464,
      "epoch": 0.6727791973545769,
      "flos": 15778610674560.0,
      "grad_norm": 4.000289212610063,
      "language_loss": 0.75086403,
      "learning_rate": 1.0217764881798562e-06,
      "loss": 0.7722882,
      "num_input_tokens_seen": 241513465,
      "step": 11190,
      "time_per_iteration": 2.9061520099639893
    },
    {
      "auxiliary_loss_clip": 0.01053457,
      "auxiliary_loss_mlp": 0.01033777,
      "balance_loss_clip": 1.03577077,
      "balance_loss_mlp": 1.02028871,
      "epoch": 0.6728393206072448,
      "flos": 21249852341760.0,
      "grad_norm": 2.0448653418114193,
      "language_loss": 0.76895911,
      "learning_rate": 1.0214368100233612e-06,
      "loss": 0.7898314,
      "num_input_tokens_seen": 241534125,
      "step": 11191,
      "time_per_iteration": 2.6471197605133057
    },
    {
      "auxiliary_loss_clip": 0.01106829,
      "auxiliary_loss_mlp": 0.01033093,
      "balance_loss_clip": 1.03835869,
      "balance_loss_mlp": 1.02099955,
      "epoch": 0.6728994438599128,
      "flos": 32123279779200.0,
      "grad_norm": 1.9418401640012881,
      "language_loss": 0.86514127,
      "learning_rate": 1.0210971689740802e-06,
      "loss": 0.88654047,
      "num_input_tokens_seen": 241556340,
      "step": 11192,
      "time_per_iteration": 2.539520740509033
    },
    {
      "auxiliary_loss_clip": 0.010992,
      "auxiliary_loss_mlp": 0.0103389,
      "balance_loss_clip": 1.03783131,
      "balance_loss_mlp": 1.02059817,
      "epoch": 0.6729595671125808,
      "flos": 23112359331840.0,
      "grad_norm": 2.047186374413518,
      "language_loss": 0.75897503,
      "learning_rate": 1.0207575650448923e-06,
      "loss": 0.78030598,
      "num_input_tokens_seen": 241575185,
      "step": 11193,
      "time_per_iteration": 2.514896869659424
    },
    {
      "auxiliary_loss_clip": 0.01075128,
      "auxiliary_loss_mlp": 0.010297,
      "balance_loss_clip": 1.04095149,
      "balance_loss_mlp": 1.01752877,
      "epoch": 0.6730196903652488,
      "flos": 14611406227200.0,
      "grad_norm": 1.8418265841298234,
      "language_loss": 0.79120541,
      "learning_rate": 1.0204179982486758e-06,
      "loss": 0.81225371,
      "num_input_tokens_seen": 241592970,
      "step": 11194,
      "time_per_iteration": 2.5387256145477295
    },
    {
      "auxiliary_loss_clip": 0.01098367,
      "auxiliary_loss_mlp": 0.01027926,
      "balance_loss_clip": 1.03702855,
      "balance_loss_mlp": 1.01619554,
      "epoch": 0.6730798136179167,
      "flos": 21105922544640.0,
      "grad_norm": 1.8973532738663774,
      "language_loss": 0.89677441,
      "learning_rate": 1.0200784685983075e-06,
      "loss": 0.9180373,
      "num_input_tokens_seen": 241610245,
      "step": 11195,
      "time_per_iteration": 3.9281036853790283
    },
    {
      "auxiliary_loss_clip": 0.01098807,
      "auxiliary_loss_mlp": 0.01030775,
      "balance_loss_clip": 1.03831482,
      "balance_loss_mlp": 1.01892591,
      "epoch": 0.6731399368705847,
      "flos": 28986267438720.0,
      "grad_norm": 1.667219489195078,
      "language_loss": 0.72454709,
      "learning_rate": 1.019738976106662e-06,
      "loss": 0.74584293,
      "num_input_tokens_seen": 241630350,
      "step": 11196,
      "time_per_iteration": 2.579883575439453
    },
    {
      "auxiliary_loss_clip": 0.00993809,
      "auxiliary_loss_mlp": 0.01002569,
      "balance_loss_clip": 1.02653849,
      "balance_loss_mlp": 1.00146055,
      "epoch": 0.6732000601232526,
      "flos": 64743708723840.0,
      "grad_norm": 0.775854963192077,
      "language_loss": 0.56544685,
      "learning_rate": 1.0193995207866123e-06,
      "loss": 0.58541059,
      "num_input_tokens_seen": 241692380,
      "step": 11197,
      "time_per_iteration": 3.1687936782836914
    },
    {
      "auxiliary_loss_clip": 0.01086007,
      "auxiliary_loss_mlp": 0.0102803,
      "balance_loss_clip": 1.03831542,
      "balance_loss_mlp": 1.0164609,
      "epoch": 0.6732601833759206,
      "flos": 17201642762880.0,
      "grad_norm": 2.033903051069441,
      "language_loss": 0.75428402,
      "learning_rate": 1.0190601026510312e-06,
      "loss": 0.77542436,
      "num_input_tokens_seen": 241710430,
      "step": 11198,
      "time_per_iteration": 2.541346311569214
    },
    {
      "auxiliary_loss_clip": 0.01098282,
      "auxiliary_loss_mlp": 0.01033094,
      "balance_loss_clip": 1.0363934,
      "balance_loss_mlp": 1.01990926,
      "epoch": 0.6733203066285887,
      "flos": 18658861620480.0,
      "grad_norm": 4.624830922006331,
      "language_loss": 0.81814289,
      "learning_rate": 1.0187207217127892e-06,
      "loss": 0.83945668,
      "num_input_tokens_seen": 241724775,
      "step": 11199,
      "time_per_iteration": 2.4509060382843018
    },
    {
      "auxiliary_loss_clip": 0.0105772,
      "auxiliary_loss_mlp": 0.01033749,
      "balance_loss_clip": 1.03935146,
      "balance_loss_mlp": 1.02043951,
      "epoch": 0.6733804298812566,
      "flos": 35809330481280.0,
      "grad_norm": 1.9259791812715914,
      "language_loss": 0.71202707,
      "learning_rate": 1.0183813779847552e-06,
      "loss": 0.73294175,
      "num_input_tokens_seen": 241744440,
      "step": 11200,
      "time_per_iteration": 2.714695930480957
    },
    {
      "auxiliary_loss_clip": 0.0111199,
      "auxiliary_loss_mlp": 0.0103467,
      "balance_loss_clip": 1.04087305,
      "balance_loss_mlp": 1.02252269,
      "epoch": 0.6734405531339246,
      "flos": 61638833099520.0,
      "grad_norm": 1.6328925533307903,
      "language_loss": 0.64701486,
      "learning_rate": 1.0180420714797987e-06,
      "loss": 0.66848141,
      "num_input_tokens_seen": 241771705,
      "step": 11201,
      "time_per_iteration": 2.833658456802368
    },
    {
      "auxiliary_loss_clip": 0.01088358,
      "auxiliary_loss_mlp": 0.01032795,
      "balance_loss_clip": 1.03687835,
      "balance_loss_mlp": 1.02041554,
      "epoch": 0.6735006763865925,
      "flos": 20522338277760.0,
      "grad_norm": 1.579770557438319,
      "language_loss": 0.63168699,
      "learning_rate": 1.0177028022107856e-06,
      "loss": 0.65289849,
      "num_input_tokens_seen": 241790830,
      "step": 11202,
      "time_per_iteration": 3.9499635696411133
    },
    {
      "auxiliary_loss_clip": 0.01109855,
      "auxiliary_loss_mlp": 0.01027862,
      "balance_loss_clip": 1.03839087,
      "balance_loss_mlp": 1.01599455,
      "epoch": 0.6735607996392605,
      "flos": 13918869031680.0,
      "grad_norm": 1.7293273690747857,
      "language_loss": 0.74743617,
      "learning_rate": 1.0173635701905796e-06,
      "loss": 0.76881337,
      "num_input_tokens_seen": 241808165,
      "step": 11203,
      "time_per_iteration": 4.1642467975616455
    },
    {
      "auxiliary_loss_clip": 0.01092979,
      "auxiliary_loss_mlp": 0.01032343,
      "balance_loss_clip": 1.03863263,
      "balance_loss_mlp": 1.01810956,
      "epoch": 0.6736209228919284,
      "flos": 18807244704000.0,
      "grad_norm": 1.6105073564340877,
      "language_loss": 0.67384911,
      "learning_rate": 1.0170243754320456e-06,
      "loss": 0.69510233,
      "num_input_tokens_seen": 241826925,
      "step": 11204,
      "time_per_iteration": 2.5578625202178955
    },
    {
      "auxiliary_loss_clip": 0.01105027,
      "auxiliary_loss_mlp": 0.01033115,
      "balance_loss_clip": 1.04087293,
      "balance_loss_mlp": 1.01978779,
      "epoch": 0.6736810461445965,
      "flos": 20373129181440.0,
      "grad_norm": 1.5526410561168842,
      "language_loss": 0.73993373,
      "learning_rate": 1.0166852179480465e-06,
      "loss": 0.76131511,
      "num_input_tokens_seen": 241845525,
      "step": 11205,
      "time_per_iteration": 2.5272927284240723
    },
    {
      "auxiliary_loss_clip": 0.01106327,
      "auxiliary_loss_mlp": 0.01032327,
      "balance_loss_clip": 1.03770375,
      "balance_loss_mlp": 1.02038181,
      "epoch": 0.6737411693972644,
      "flos": 30007530927360.0,
      "grad_norm": 1.6012801619297645,
      "language_loss": 0.71229994,
      "learning_rate": 1.0163460977514416e-06,
      "loss": 0.73368645,
      "num_input_tokens_seen": 241866815,
      "step": 11206,
      "time_per_iteration": 4.160272836685181
    },
    {
      "auxiliary_loss_clip": 0.01077571,
      "auxiliary_loss_mlp": 0.00788087,
      "balance_loss_clip": 1.04172087,
      "balance_loss_mlp": 1.01261926,
      "epoch": 0.6738012926499324,
      "flos": 25447342844160.0,
      "grad_norm": 2.6293643485688274,
      "language_loss": 0.67332584,
      "learning_rate": 1.016007014855092e-06,
      "loss": 0.69198245,
      "num_input_tokens_seen": 241887050,
      "step": 11207,
      "time_per_iteration": 2.630535364151001
    },
    {
      "auxiliary_loss_clip": 0.01057752,
      "auxiliary_loss_mlp": 0.01033185,
      "balance_loss_clip": 1.03516972,
      "balance_loss_mlp": 1.02081728,
      "epoch": 0.6738614159026003,
      "flos": 20776873029120.0,
      "grad_norm": 2.0532750887548135,
      "language_loss": 0.73619837,
      "learning_rate": 1.0156679692718553e-06,
      "loss": 0.75710773,
      "num_input_tokens_seen": 241904280,
      "step": 11208,
      "time_per_iteration": 2.5767581462860107
    },
    {
      "auxiliary_loss_clip": 0.01094087,
      "auxiliary_loss_mlp": 0.01040581,
      "balance_loss_clip": 1.0358007,
      "balance_loss_mlp": 1.02476788,
      "epoch": 0.6739215391552683,
      "flos": 19566898462080.0,
      "grad_norm": 1.8552214927405155,
      "language_loss": 0.75483179,
      "learning_rate": 1.0153289610145867e-06,
      "loss": 0.77617854,
      "num_input_tokens_seen": 241919190,
      "step": 11209,
      "time_per_iteration": 2.4877548217773438
    },
    {
      "auxiliary_loss_clip": 0.01069849,
      "auxiliary_loss_mlp": 0.01027791,
      "balance_loss_clip": 1.0386343,
      "balance_loss_mlp": 1.01668704,
      "epoch": 0.6739816624079362,
      "flos": 24388193485440.0,
      "grad_norm": 1.7746572513346497,
      "language_loss": 0.66730881,
      "learning_rate": 1.0149899900961428e-06,
      "loss": 0.68828523,
      "num_input_tokens_seen": 241940525,
      "step": 11210,
      "time_per_iteration": 2.6432743072509766
    },
    {
      "auxiliary_loss_clip": 0.01103598,
      "auxiliary_loss_mlp": 0.01028597,
      "balance_loss_clip": 1.03594232,
      "balance_loss_mlp": 1.01734376,
      "epoch": 0.6740417856606042,
      "flos": 22528164533760.0,
      "grad_norm": 2.1679953652701145,
      "language_loss": 0.80335391,
      "learning_rate": 1.014651056529377e-06,
      "loss": 0.82467592,
      "num_input_tokens_seen": 241959290,
      "step": 11211,
      "time_per_iteration": 2.5106260776519775
    },
    {
      "auxiliary_loss_clip": 0.01057759,
      "auxiliary_loss_mlp": 0.01027955,
      "balance_loss_clip": 1.03544784,
      "balance_loss_mlp": 1.01611114,
      "epoch": 0.6741019089132723,
      "flos": 25775458606080.0,
      "grad_norm": 1.3414673890790885,
      "language_loss": 0.76675403,
      "learning_rate": 1.014312160327143e-06,
      "loss": 0.78761113,
      "num_input_tokens_seen": 241980715,
      "step": 11212,
      "time_per_iteration": 2.634998083114624
    },
    {
      "auxiliary_loss_clip": 0.01069544,
      "auxiliary_loss_mlp": 0.00789201,
      "balance_loss_clip": 1.03742933,
      "balance_loss_mlp": 1.01390529,
      "epoch": 0.6741620321659402,
      "flos": 21105671149440.0,
      "grad_norm": 1.5627067309024232,
      "language_loss": 0.78160793,
      "learning_rate": 1.0139733015022905e-06,
      "loss": 0.8001954,
      "num_input_tokens_seen": 241999985,
      "step": 11213,
      "time_per_iteration": 2.575955629348755
    },
    {
      "auxiliary_loss_clip": 0.01067286,
      "auxiliary_loss_mlp": 0.01031546,
      "balance_loss_clip": 1.04294205,
      "balance_loss_mlp": 1.01846242,
      "epoch": 0.6742221554186082,
      "flos": 20740423703040.0,
      "grad_norm": 1.888988724255275,
      "language_loss": 0.67686701,
      "learning_rate": 1.0136344800676685e-06,
      "loss": 0.69785523,
      "num_input_tokens_seen": 242018990,
      "step": 11214,
      "time_per_iteration": 2.5863049030303955
    },
    {
      "auxiliary_loss_clip": 0.01108122,
      "auxiliary_loss_mlp": 0.00785562,
      "balance_loss_clip": 1.03770804,
      "balance_loss_mlp": 1.01265097,
      "epoch": 0.6742822786712761,
      "flos": 37774146384000.0,
      "grad_norm": 1.8826555093447497,
      "language_loss": 0.72790879,
      "learning_rate": 1.0132956960361263e-06,
      "loss": 0.7468456,
      "num_input_tokens_seen": 242039340,
      "step": 11215,
      "time_per_iteration": 2.618163824081421
    },
    {
      "auxiliary_loss_clip": 0.01099039,
      "auxiliary_loss_mlp": 0.00783671,
      "balance_loss_clip": 1.03703034,
      "balance_loss_mlp": 1.00937867,
      "epoch": 0.6743424019239441,
      "flos": 37263891732480.0,
      "grad_norm": 1.7781193511213345,
      "language_loss": 0.66751736,
      "learning_rate": 1.0129569494205096e-06,
      "loss": 0.6863445,
      "num_input_tokens_seen": 242062215,
      "step": 11216,
      "time_per_iteration": 2.6357555389404297
    },
    {
      "auxiliary_loss_clip": 0.01040756,
      "auxiliary_loss_mlp": 0.01001271,
      "balance_loss_clip": 1.01703167,
      "balance_loss_mlp": 1.0001564,
      "epoch": 0.674402525176612,
      "flos": 65997746300160.0,
      "grad_norm": 0.6726541678347203,
      "language_loss": 0.56246388,
      "learning_rate": 1.0126182402336646e-06,
      "loss": 0.58288413,
      "num_input_tokens_seen": 242131130,
      "step": 11217,
      "time_per_iteration": 3.2011559009552
    },
    {
      "auxiliary_loss_clip": 0.01095697,
      "auxiliary_loss_mlp": 0.0103033,
      "balance_loss_clip": 1.03691816,
      "balance_loss_mlp": 1.01786625,
      "epoch": 0.67446264842928,
      "flos": 26461208131200.0,
      "grad_norm": 2.006648604304057,
      "language_loss": 0.74572873,
      "learning_rate": 1.0122795684884363e-06,
      "loss": 0.76698899,
      "num_input_tokens_seen": 242149720,
      "step": 11218,
      "time_per_iteration": 2.519341468811035
    },
    {
      "auxiliary_loss_clip": 0.01076575,
      "auxiliary_loss_mlp": 0.01045191,
      "balance_loss_clip": 1.03749061,
      "balance_loss_mlp": 1.03101707,
      "epoch": 0.674522771681948,
      "flos": 23732392924800.0,
      "grad_norm": 1.6833081610275735,
      "language_loss": 0.66056311,
      "learning_rate": 1.0119409341976639e-06,
      "loss": 0.68178082,
      "num_input_tokens_seen": 242168875,
      "step": 11219,
      "time_per_iteration": 2.5805864334106445
    },
    {
      "auxiliary_loss_clip": 0.01068136,
      "auxiliary_loss_mlp": 0.01044542,
      "balance_loss_clip": 1.03388619,
      "balance_loss_mlp": 1.02952158,
      "epoch": 0.674582894934616,
      "flos": 24754338771840.0,
      "grad_norm": 1.6769712672294836,
      "language_loss": 0.74949503,
      "learning_rate": 1.0116023373741904e-06,
      "loss": 0.77062178,
      "num_input_tokens_seen": 242188465,
      "step": 11220,
      "time_per_iteration": 2.5654799938201904
    },
    {
      "auxiliary_loss_clip": 0.01093699,
      "auxiliary_loss_mlp": 0.01031658,
      "balance_loss_clip": 1.03836036,
      "balance_loss_mlp": 1.01894999,
      "epoch": 0.6746430181872839,
      "flos": 24826626892800.0,
      "grad_norm": 1.684097534818142,
      "language_loss": 0.70311868,
      "learning_rate": 1.0112637780308554e-06,
      "loss": 0.72437227,
      "num_input_tokens_seen": 242208675,
      "step": 11221,
      "time_per_iteration": 2.531670570373535
    },
    {
      "auxiliary_loss_clip": 0.01074835,
      "auxiliary_loss_mlp": 0.01026447,
      "balance_loss_clip": 1.03820896,
      "balance_loss_mlp": 1.01497889,
      "epoch": 0.6747031414399519,
      "flos": 16873491087360.0,
      "grad_norm": 1.8396711416520009,
      "language_loss": 0.58010447,
      "learning_rate": 1.010925256180498e-06,
      "loss": 0.60111725,
      "num_input_tokens_seen": 242227440,
      "step": 11222,
      "time_per_iteration": 2.539881944656372
    },
    {
      "auxiliary_loss_clip": 0.01095852,
      "auxiliary_loss_mlp": 0.0103165,
      "balance_loss_clip": 1.03996491,
      "balance_loss_mlp": 1.01926422,
      "epoch": 0.6747632646926198,
      "flos": 22784925928320.0,
      "grad_norm": 1.6803421286531328,
      "language_loss": 0.76848686,
      "learning_rate": 1.0105867718359528e-06,
      "loss": 0.7897619,
      "num_input_tokens_seen": 242245240,
      "step": 11223,
      "time_per_iteration": 2.524216890335083
    },
    {
      "auxiliary_loss_clip": 0.01101226,
      "auxiliary_loss_mlp": 0.01030111,
      "balance_loss_clip": 1.03852725,
      "balance_loss_mlp": 1.01780891,
      "epoch": 0.6748233879452878,
      "flos": 20046090827520.0,
      "grad_norm": 1.6622883997081865,
      "language_loss": 0.75132096,
      "learning_rate": 1.0102483250100574e-06,
      "loss": 0.77263439,
      "num_input_tokens_seen": 242263435,
      "step": 11224,
      "time_per_iteration": 2.493838310241699
    },
    {
      "auxiliary_loss_clip": 0.010615,
      "auxiliary_loss_mlp": 0.01025783,
      "balance_loss_clip": 1.03717279,
      "balance_loss_mlp": 1.01547194,
      "epoch": 0.6748835111979558,
      "flos": 23002831785600.0,
      "grad_norm": 1.5439312517237889,
      "language_loss": 0.6325618,
      "learning_rate": 1.0099099157156445e-06,
      "loss": 0.65343463,
      "num_input_tokens_seen": 242282765,
      "step": 11225,
      "time_per_iteration": 2.5922675132751465
    },
    {
      "auxiliary_loss_clip": 0.01103167,
      "auxiliary_loss_mlp": 0.00781354,
      "balance_loss_clip": 1.03706169,
      "balance_loss_mlp": 1.00684762,
      "epoch": 0.6749436344506238,
      "flos": 12197311009920.0,
      "grad_norm": 1.8251466787149069,
      "language_loss": 0.64401162,
      "learning_rate": 1.0095715439655462e-06,
      "loss": 0.66285682,
      "num_input_tokens_seen": 242298980,
      "step": 11226,
      "time_per_iteration": 2.4293322563171387
    },
    {
      "auxiliary_loss_clip": 0.01098274,
      "auxiliary_loss_mlp": 0.01029576,
      "balance_loss_clip": 1.03761756,
      "balance_loss_mlp": 1.01703525,
      "epoch": 0.6750037577032918,
      "flos": 11873720361600.0,
      "grad_norm": 2.197825593419333,
      "language_loss": 0.71788269,
      "learning_rate": 1.0092332097725945e-06,
      "loss": 0.73916119,
      "num_input_tokens_seen": 242315420,
      "step": 11227,
      "time_per_iteration": 2.471691846847534
    },
    {
      "auxiliary_loss_clip": 0.01080051,
      "auxiliary_loss_mlp": 0.01032925,
      "balance_loss_clip": 1.03549123,
      "balance_loss_mlp": 1.01935267,
      "epoch": 0.6750638809559597,
      "flos": 17019611614080.0,
      "grad_norm": 3.047396895402804,
      "language_loss": 0.70984793,
      "learning_rate": 1.0088949131496183e-06,
      "loss": 0.73097771,
      "num_input_tokens_seen": 242332805,
      "step": 11228,
      "time_per_iteration": 2.5047240257263184
    },
    {
      "auxiliary_loss_clip": 0.01024254,
      "auxiliary_loss_mlp": 0.01003042,
      "balance_loss_clip": 1.02404177,
      "balance_loss_mlp": 1.00190973,
      "epoch": 0.6751240042086277,
      "flos": 70951011891840.0,
      "grad_norm": 0.7554039387089345,
      "language_loss": 0.53265351,
      "learning_rate": 1.0085566541094482e-06,
      "loss": 0.55292648,
      "num_input_tokens_seen": 242396160,
      "step": 11229,
      "time_per_iteration": 3.2080209255218506
    },
    {
      "auxiliary_loss_clip": 0.01095109,
      "auxiliary_loss_mlp": 0.01025725,
      "balance_loss_clip": 1.03643477,
      "balance_loss_mlp": 1.0145191,
      "epoch": 0.6751841274612956,
      "flos": 22675146986880.0,
      "grad_norm": 1.699454475147527,
      "language_loss": 0.80108643,
      "learning_rate": 1.0082184326649072e-06,
      "loss": 0.82229471,
      "num_input_tokens_seen": 242414660,
      "step": 11230,
      "time_per_iteration": 2.5285768508911133
    },
    {
      "auxiliary_loss_clip": 0.01076092,
      "auxiliary_loss_mlp": 0.01031441,
      "balance_loss_clip": 1.0347693,
      "balance_loss_mlp": 1.0200808,
      "epoch": 0.6752442507139637,
      "flos": 21288636051840.0,
      "grad_norm": 1.4917026545505254,
      "language_loss": 0.6578297,
      "learning_rate": 1.0078802488288228e-06,
      "loss": 0.67890501,
      "num_input_tokens_seen": 242434225,
      "step": 11231,
      "time_per_iteration": 2.537249803543091
    },
    {
      "auxiliary_loss_clip": 0.01068856,
      "auxiliary_loss_mlp": 0.01035599,
      "balance_loss_clip": 1.03798318,
      "balance_loss_mlp": 1.02221203,
      "epoch": 0.6753043739666316,
      "flos": 28256921781120.0,
      "grad_norm": 1.8353877667159697,
      "language_loss": 0.66488421,
      "learning_rate": 1.0075421026140198e-06,
      "loss": 0.68592876,
      "num_input_tokens_seen": 242454355,
      "step": 11232,
      "time_per_iteration": 2.6465675830841064
    },
    {
      "auxiliary_loss_clip": 0.01072187,
      "auxiliary_loss_mlp": 0.01028129,
      "balance_loss_clip": 1.03568149,
      "balance_loss_mlp": 1.01588058,
      "epoch": 0.6753644972192996,
      "flos": 21360349555200.0,
      "grad_norm": 1.8952136019121177,
      "language_loss": 0.72473449,
      "learning_rate": 1.0072039940333188e-06,
      "loss": 0.74573767,
      "num_input_tokens_seen": 242474935,
      "step": 11233,
      "time_per_iteration": 2.5725321769714355
    },
    {
      "auxiliary_loss_clip": 0.01097944,
      "auxiliary_loss_mlp": 0.01031698,
      "balance_loss_clip": 1.03702009,
      "balance_loss_mlp": 1.01965809,
      "epoch": 0.6754246204719675,
      "flos": 26541971861760.0,
      "grad_norm": 2.2188958137715957,
      "language_loss": 0.77114558,
      "learning_rate": 1.0068659230995418e-06,
      "loss": 0.79244196,
      "num_input_tokens_seen": 242495530,
      "step": 11234,
      "time_per_iteration": 3.921712875366211
    },
    {
      "auxiliary_loss_clip": 0.01106449,
      "auxiliary_loss_mlp": 0.01032155,
      "balance_loss_clip": 1.03728724,
      "balance_loss_mlp": 1.01972139,
      "epoch": 0.6754847437246355,
      "flos": 25556690822400.0,
      "grad_norm": 1.5365311895682032,
      "language_loss": 0.75671494,
      "learning_rate": 1.0065278898255101e-06,
      "loss": 0.77810097,
      "num_input_tokens_seen": 242514550,
      "step": 11235,
      "time_per_iteration": 2.497377634048462
    },
    {
      "auxiliary_loss_clip": 0.01031891,
      "auxiliary_loss_mlp": 0.01002204,
      "balance_loss_clip": 1.01776671,
      "balance_loss_mlp": 1.00104797,
      "epoch": 0.6755448669773034,
      "flos": 59513318726400.0,
      "grad_norm": 0.7804714761872624,
      "language_loss": 0.51341504,
      "learning_rate": 1.0061898942240387e-06,
      "loss": 0.53375596,
      "num_input_tokens_seen": 242569200,
      "step": 11236,
      "time_per_iteration": 3.0710935592651367
    },
    {
      "auxiliary_loss_clip": 0.01074998,
      "auxiliary_loss_mlp": 0.01027878,
      "balance_loss_clip": 1.03601551,
      "balance_loss_mlp": 1.0141151,
      "epoch": 0.6756049902299714,
      "flos": 23294534135040.0,
      "grad_norm": 1.8812417926536926,
      "language_loss": 0.75262332,
      "learning_rate": 1.0058519363079464e-06,
      "loss": 0.77365208,
      "num_input_tokens_seen": 242586950,
      "step": 11237,
      "time_per_iteration": 2.5880773067474365
    },
    {
      "auxiliary_loss_clip": 0.01083009,
      "auxiliary_loss_mlp": 0.01041087,
      "balance_loss_clip": 1.03868961,
      "balance_loss_mlp": 1.02909994,
      "epoch": 0.6756651134826394,
      "flos": 31575426566400.0,
      "grad_norm": 1.6963738923401783,
      "language_loss": 0.77451974,
      "learning_rate": 1.0055140160900482e-06,
      "loss": 0.79576063,
      "num_input_tokens_seen": 242607380,
      "step": 11238,
      "time_per_iteration": 2.6080729961395264
    },
    {
      "auxiliary_loss_clip": 0.01096538,
      "auxiliary_loss_mlp": 0.01034803,
      "balance_loss_clip": 1.03722048,
      "balance_loss_mlp": 1.0215826,
      "epoch": 0.6757252367353074,
      "flos": 27272287186560.0,
      "grad_norm": 1.8825860175814748,
      "language_loss": 0.66638398,
      "learning_rate": 1.0051761335831587e-06,
      "loss": 0.68769735,
      "num_input_tokens_seen": 242628025,
      "step": 11239,
      "time_per_iteration": 2.580176830291748
    },
    {
      "auxiliary_loss_clip": 0.01083033,
      "auxiliary_loss_mlp": 0.01028796,
      "balance_loss_clip": 1.0383172,
      "balance_loss_mlp": 1.01654768,
      "epoch": 0.6757853599879754,
      "flos": 16830900535680.0,
      "grad_norm": 1.8279893691017335,
      "language_loss": 0.82793653,
      "learning_rate": 1.0048382888000898e-06,
      "loss": 0.84905487,
      "num_input_tokens_seen": 242643825,
      "step": 11240,
      "time_per_iteration": 2.515573024749756
    },
    {
      "auxiliary_loss_clip": 0.01083185,
      "auxiliary_loss_mlp": 0.01033271,
      "balance_loss_clip": 1.0386827,
      "balance_loss_mlp": 1.01850116,
      "epoch": 0.6758454832406433,
      "flos": 23220055284480.0,
      "grad_norm": 1.9926900253676532,
      "language_loss": 0.74531102,
      "learning_rate": 1.0045004817536525e-06,
      "loss": 0.76647556,
      "num_input_tokens_seen": 242661820,
      "step": 11241,
      "time_per_iteration": 3.920835018157959
    },
    {
      "auxiliary_loss_clip": 0.01066617,
      "auxiliary_loss_mlp": 0.01034307,
      "balance_loss_clip": 1.03940296,
      "balance_loss_mlp": 1.0219692,
      "epoch": 0.6759056064933113,
      "flos": 16289547684480.0,
      "grad_norm": 2.0795740830007174,
      "language_loss": 0.79858404,
      "learning_rate": 1.0041627124566572e-06,
      "loss": 0.81959331,
      "num_input_tokens_seen": 242679890,
      "step": 11242,
      "time_per_iteration": 4.099387168884277
    },
    {
      "auxiliary_loss_clip": 0.0109138,
      "auxiliary_loss_mlp": 0.0103711,
      "balance_loss_clip": 1.03500128,
      "balance_loss_mlp": 1.02411628,
      "epoch": 0.6759657297459792,
      "flos": 25922297404800.0,
      "grad_norm": 1.8088287677359127,
      "language_loss": 0.73073232,
      "learning_rate": 1.0038249809219109e-06,
      "loss": 0.7520172,
      "num_input_tokens_seen": 242699495,
      "step": 11243,
      "time_per_iteration": 2.5249197483062744
    },
    {
      "auxiliary_loss_clip": 0.01097196,
      "auxiliary_loss_mlp": 0.01035652,
      "balance_loss_clip": 1.03774071,
      "balance_loss_mlp": 1.02360046,
      "epoch": 0.6760258529986473,
      "flos": 23000820624000.0,
      "grad_norm": 1.5683994600089841,
      "language_loss": 0.72768569,
      "learning_rate": 1.003487287162221e-06,
      "loss": 0.74901414,
      "num_input_tokens_seen": 242719500,
      "step": 11244,
      "time_per_iteration": 3.8905811309814453
    },
    {
      "auxiliary_loss_clip": 0.01109765,
      "auxiliary_loss_mlp": 0.01037245,
      "balance_loss_clip": 1.0386796,
      "balance_loss_mlp": 1.0249424,
      "epoch": 0.6760859762513152,
      "flos": 20959335141120.0,
      "grad_norm": 1.971878203645556,
      "language_loss": 0.85613072,
      "learning_rate": 1.003149631190393e-06,
      "loss": 0.87760085,
      "num_input_tokens_seen": 242738325,
      "step": 11245,
      "time_per_iteration": 2.462027072906494
    },
    {
      "auxiliary_loss_clip": 0.01111663,
      "auxiliary_loss_mlp": 0.00785207,
      "balance_loss_clip": 1.03794718,
      "balance_loss_mlp": 1.0105319,
      "epoch": 0.6761460995039832,
      "flos": 23622937205760.0,
      "grad_norm": 1.7724238345245786,
      "language_loss": 0.73446345,
      "learning_rate": 1.0028120130192327e-06,
      "loss": 0.75343215,
      "num_input_tokens_seen": 242756620,
      "step": 11246,
      "time_per_iteration": 2.4964733123779297
    },
    {
      "auxiliary_loss_clip": 0.01093931,
      "auxiliary_loss_mlp": 0.01024137,
      "balance_loss_clip": 1.03896332,
      "balance_loss_mlp": 1.01258612,
      "epoch": 0.6762062227566511,
      "flos": 20770875457920.0,
      "grad_norm": 1.7517141650394952,
      "language_loss": 0.87737894,
      "learning_rate": 1.002474432661539e-06,
      "loss": 0.89855963,
      "num_input_tokens_seen": 242774505,
      "step": 11247,
      "time_per_iteration": 2.4913222789764404
    },
    {
      "auxiliary_loss_clip": 0.01022719,
      "auxiliary_loss_mlp": 0.00999919,
      "balance_loss_clip": 1.02173889,
      "balance_loss_mlp": 0.99860191,
      "epoch": 0.6762663460093191,
      "flos": 52818099166080.0,
      "grad_norm": 0.827180242164474,
      "language_loss": 0.54015899,
      "learning_rate": 1.002136890130115e-06,
      "loss": 0.56038541,
      "num_input_tokens_seen": 242828645,
      "step": 11248,
      "time_per_iteration": 3.166027069091797
    },
    {
      "auxiliary_loss_clip": 0.01053103,
      "auxiliary_loss_mlp": 0.01026683,
      "balance_loss_clip": 1.03786385,
      "balance_loss_mlp": 1.01540577,
      "epoch": 0.676326469261987,
      "flos": 23696302734720.0,
      "grad_norm": 1.4603574843310045,
      "language_loss": 0.73628283,
      "learning_rate": 1.001799385437761e-06,
      "loss": 0.75708067,
      "num_input_tokens_seen": 242850100,
      "step": 11249,
      "time_per_iteration": 2.705786943435669
    },
    {
      "auxiliary_loss_clip": 0.01097202,
      "auxiliary_loss_mlp": 0.01035333,
      "balance_loss_clip": 1.03576481,
      "balance_loss_mlp": 1.02255416,
      "epoch": 0.676386592514655,
      "flos": 14063732582400.0,
      "grad_norm": 2.3171966899326395,
      "language_loss": 0.74269789,
      "learning_rate": 1.0014619185972732e-06,
      "loss": 0.76402324,
      "num_input_tokens_seen": 242867775,
      "step": 11250,
      "time_per_iteration": 2.4809460639953613
    },
    {
      "auxiliary_loss_clip": 0.01107524,
      "auxiliary_loss_mlp": 0.01027296,
      "balance_loss_clip": 1.03709388,
      "balance_loss_mlp": 1.01509547,
      "epoch": 0.676446715767323,
      "flos": 20412236113920.0,
      "grad_norm": 1.9202803592555908,
      "language_loss": 0.74984312,
      "learning_rate": 1.0011244896214497e-06,
      "loss": 0.7711913,
      "num_input_tokens_seen": 242886865,
      "step": 11251,
      "time_per_iteration": 2.4777307510375977
    },
    {
      "auxiliary_loss_clip": 0.01079672,
      "auxiliary_loss_mlp": 0.01027935,
      "balance_loss_clip": 1.03933764,
      "balance_loss_mlp": 1.01590085,
      "epoch": 0.676506839019991,
      "flos": 21288241002240.0,
      "grad_norm": 1.7684979797640559,
      "language_loss": 0.7011463,
      "learning_rate": 1.0007870985230873e-06,
      "loss": 0.72222239,
      "num_input_tokens_seen": 242906705,
      "step": 11252,
      "time_per_iteration": 2.5613412857055664
    },
    {
      "auxiliary_loss_clip": 0.01066767,
      "auxiliary_loss_mlp": 0.01030242,
      "balance_loss_clip": 1.03815317,
      "balance_loss_mlp": 1.01848757,
      "epoch": 0.676566962272659,
      "flos": 29932477459200.0,
      "grad_norm": 2.036885969286238,
      "language_loss": 0.66977358,
      "learning_rate": 1.0004497453149765e-06,
      "loss": 0.69074357,
      "num_input_tokens_seen": 242925215,
      "step": 11253,
      "time_per_iteration": 2.663843870162964
    },
    {
      "auxiliary_loss_clip": 0.01068672,
      "auxiliary_loss_mlp": 0.00788535,
      "balance_loss_clip": 1.03625953,
      "balance_loss_mlp": 1.01045847,
      "epoch": 0.6766270855253269,
      "flos": 17931203902080.0,
      "grad_norm": 1.6113550405494432,
      "language_loss": 0.77346969,
      "learning_rate": 1.0001124300099115e-06,
      "loss": 0.79204178,
      "num_input_tokens_seen": 242944750,
      "step": 11254,
      "time_per_iteration": 2.567986488342285
    },
    {
      "auxiliary_loss_clip": 0.01096643,
      "auxiliary_loss_mlp": 0.01028438,
      "balance_loss_clip": 1.0365442,
      "balance_loss_mlp": 1.01601648,
      "epoch": 0.6766872087779949,
      "flos": 23104853389440.0,
      "grad_norm": 2.360344964153428,
      "language_loss": 0.72230399,
      "learning_rate": 9.997751526206835e-07,
      "loss": 0.74355477,
      "num_input_tokens_seen": 242963860,
      "step": 11255,
      "time_per_iteration": 2.520810604095459
    },
    {
      "auxiliary_loss_clip": 0.0105826,
      "auxiliary_loss_mlp": 0.00783077,
      "balance_loss_clip": 1.03344357,
      "balance_loss_mlp": 1.00534487,
      "epoch": 0.6767473320306628,
      "flos": 26213137827840.0,
      "grad_norm": 1.989044825054772,
      "language_loss": 0.75307876,
      "learning_rate": 9.994379131600828e-07,
      "loss": 0.77149218,
      "num_input_tokens_seen": 242983050,
      "step": 11256,
      "time_per_iteration": 2.6196107864379883
    },
    {
      "auxiliary_loss_clip": 0.01099093,
      "auxiliary_loss_mlp": 0.01032004,
      "balance_loss_clip": 1.03954887,
      "balance_loss_mlp": 1.01974297,
      "epoch": 0.6768074552833309,
      "flos": 18368739469440.0,
      "grad_norm": 2.4110915340058345,
      "language_loss": 0.65320396,
      "learning_rate": 9.991007116408965e-07,
      "loss": 0.67451495,
      "num_input_tokens_seen": 243001125,
      "step": 11257,
      "time_per_iteration": 2.4969475269317627
    },
    {
      "auxiliary_loss_clip": 0.0105978,
      "auxiliary_loss_mlp": 0.01029458,
      "balance_loss_clip": 1.03739333,
      "balance_loss_mlp": 1.01819921,
      "epoch": 0.6768675785359988,
      "flos": 23039927556480.0,
      "grad_norm": 1.561676436815102,
      "language_loss": 0.75489664,
      "learning_rate": 9.987635480759109e-07,
      "loss": 0.77578902,
      "num_input_tokens_seen": 243021865,
      "step": 11258,
      "time_per_iteration": 2.5986268520355225
    },
    {
      "auxiliary_loss_clip": 0.01083282,
      "auxiliary_loss_mlp": 0.01032981,
      "balance_loss_clip": 1.03721976,
      "balance_loss_mlp": 1.02185917,
      "epoch": 0.6769277017886668,
      "flos": 33036524092800.0,
      "grad_norm": 1.7374839245482832,
      "language_loss": 0.6713357,
      "learning_rate": 9.984264224779127e-07,
      "loss": 0.69249833,
      "num_input_tokens_seen": 243042970,
      "step": 11259,
      "time_per_iteration": 2.635932445526123
    },
    {
      "auxiliary_loss_clip": 0.01087147,
      "auxiliary_loss_mlp": 0.01034055,
      "balance_loss_clip": 1.03711748,
      "balance_loss_mlp": 1.02165174,
      "epoch": 0.6769878250413347,
      "flos": 20848406964480.0,
      "grad_norm": 2.3103119216870254,
      "language_loss": 0.85854387,
      "learning_rate": 9.980893348596839e-07,
      "loss": 0.87975585,
      "num_input_tokens_seen": 243058470,
      "step": 11260,
      "time_per_iteration": 2.5188682079315186
    },
    {
      "auxiliary_loss_clip": 0.01082256,
      "auxiliary_loss_mlp": 0.01038412,
      "balance_loss_clip": 1.03511262,
      "balance_loss_mlp": 1.02519727,
      "epoch": 0.6770479482940027,
      "flos": 15595968994560.0,
      "grad_norm": 2.4979688850736275,
      "language_loss": 0.77731162,
      "learning_rate": 9.977522852340081e-07,
      "loss": 0.7985183,
      "num_input_tokens_seen": 243076630,
      "step": 11261,
      "time_per_iteration": 2.5241751670837402
    },
    {
      "auxiliary_loss_clip": 0.01082731,
      "auxiliary_loss_mlp": 0.01033641,
      "balance_loss_clip": 1.03349948,
      "balance_loss_mlp": 1.02192903,
      "epoch": 0.6771080715466706,
      "flos": 18621011664000.0,
      "grad_norm": 1.6817148993457622,
      "language_loss": 0.87913311,
      "learning_rate": 9.97415273613666e-07,
      "loss": 0.90029681,
      "num_input_tokens_seen": 243092260,
      "step": 11262,
      "time_per_iteration": 2.496824264526367
    },
    {
      "auxiliary_loss_clip": 0.01085594,
      "auxiliary_loss_mlp": 0.01030522,
      "balance_loss_clip": 1.0404706,
      "balance_loss_mlp": 1.01820183,
      "epoch": 0.6771681947993387,
      "flos": 12495441893760.0,
      "grad_norm": 2.8954232470334356,
      "language_loss": 0.74345362,
      "learning_rate": 9.97078300011439e-07,
      "loss": 0.76461482,
      "num_input_tokens_seen": 243109405,
      "step": 11263,
      "time_per_iteration": 2.51190447807312
    },
    {
      "auxiliary_loss_clip": 0.01101446,
      "auxiliary_loss_mlp": 0.01034895,
      "balance_loss_clip": 1.03728461,
      "balance_loss_mlp": 1.02133501,
      "epoch": 0.6772283180520066,
      "flos": 22236964974720.0,
      "grad_norm": 2.460272594087343,
      "language_loss": 0.67923069,
      "learning_rate": 9.967413644401016e-07,
      "loss": 0.70059407,
      "num_input_tokens_seen": 243128135,
      "step": 11264,
      "time_per_iteration": 2.4950995445251465
    },
    {
      "auxiliary_loss_clip": 0.01084964,
      "auxiliary_loss_mlp": 0.01031281,
      "balance_loss_clip": 1.03830028,
      "balance_loss_mlp": 1.01872218,
      "epoch": 0.6772884413046746,
      "flos": 16143139848960.0,
      "grad_norm": 1.9464328432111242,
      "language_loss": 0.73068166,
      "learning_rate": 9.964044669124324e-07,
      "loss": 0.75184417,
      "num_input_tokens_seen": 243146785,
      "step": 11265,
      "time_per_iteration": 2.5188872814178467
    },
    {
      "auxiliary_loss_clip": 0.01067693,
      "auxiliary_loss_mlp": 0.01035907,
      "balance_loss_clip": 1.03675711,
      "balance_loss_mlp": 1.02288306,
      "epoch": 0.6773485645573426,
      "flos": 19135755515520.0,
      "grad_norm": 4.948670490078943,
      "language_loss": 0.61745793,
      "learning_rate": 9.96067607441207e-07,
      "loss": 0.6384939,
      "num_input_tokens_seen": 243165275,
      "step": 11266,
      "time_per_iteration": 2.5451176166534424
    },
    {
      "auxiliary_loss_clip": 0.01078076,
      "auxiliary_loss_mlp": 0.01033687,
      "balance_loss_clip": 1.03779149,
      "balance_loss_mlp": 1.02128363,
      "epoch": 0.6774086878100105,
      "flos": 14136918543360.0,
      "grad_norm": 2.1612913539893635,
      "language_loss": 0.70582259,
      "learning_rate": 9.957307860391976e-07,
      "loss": 0.72694021,
      "num_input_tokens_seen": 243182845,
      "step": 11267,
      "time_per_iteration": 2.5539705753326416
    },
    {
      "auxiliary_loss_clip": 0.01107455,
      "auxiliary_loss_mlp": 0.01031406,
      "balance_loss_clip": 1.03654552,
      "balance_loss_mlp": 1.01895428,
      "epoch": 0.6774688110626785,
      "flos": 22197067943040.0,
      "grad_norm": 2.2445925100737596,
      "language_loss": 0.71771944,
      "learning_rate": 9.953940027191785e-07,
      "loss": 0.73910803,
      "num_input_tokens_seen": 243201475,
      "step": 11268,
      "time_per_iteration": 2.4651012420654297
    },
    {
      "auxiliary_loss_clip": 0.01087331,
      "auxiliary_loss_mlp": 0.01037103,
      "balance_loss_clip": 1.03935814,
      "balance_loss_mlp": 1.02393007,
      "epoch": 0.6775289343153464,
      "flos": 23039963470080.0,
      "grad_norm": 1.6259732999444814,
      "language_loss": 0.76856041,
      "learning_rate": 9.950572574939194e-07,
      "loss": 0.78980476,
      "num_input_tokens_seen": 243221850,
      "step": 11269,
      "time_per_iteration": 2.5684173107147217
    },
    {
      "auxiliary_loss_clip": 0.01077042,
      "auxiliary_loss_mlp": 0.01038206,
      "balance_loss_clip": 1.03705502,
      "balance_loss_mlp": 1.02501512,
      "epoch": 0.6775890575680145,
      "flos": 18293506433280.0,
      "grad_norm": 1.863150776143917,
      "language_loss": 0.74140954,
      "learning_rate": 9.94720550376189e-07,
      "loss": 0.76256198,
      "num_input_tokens_seen": 243239855,
      "step": 11270,
      "time_per_iteration": 2.5385191440582275
    },
    {
      "auxiliary_loss_clip": 0.01062043,
      "auxiliary_loss_mlp": 0.01037934,
      "balance_loss_clip": 1.04216564,
      "balance_loss_mlp": 1.02490413,
      "epoch": 0.6776491808206824,
      "flos": 25336450581120.0,
      "grad_norm": 1.7226179730345514,
      "language_loss": 0.73134023,
      "learning_rate": 9.94383881378756e-07,
      "loss": 0.75233996,
      "num_input_tokens_seen": 243260085,
      "step": 11271,
      "time_per_iteration": 2.64546537399292
    },
    {
      "auxiliary_loss_clip": 0.01108033,
      "auxiliary_loss_mlp": 0.01036523,
      "balance_loss_clip": 1.03797531,
      "balance_loss_mlp": 1.02462053,
      "epoch": 0.6777093040733504,
      "flos": 26028233591040.0,
      "grad_norm": 1.665639811972565,
      "language_loss": 0.67906159,
      "learning_rate": 9.94047250514387e-07,
      "loss": 0.70050716,
      "num_input_tokens_seen": 243280065,
      "step": 11272,
      "time_per_iteration": 3.9303812980651855
    },
    {
      "auxiliary_loss_clip": 0.0110138,
      "auxiliary_loss_mlp": 0.01035731,
      "balance_loss_clip": 1.0423739,
      "balance_loss_mlp": 1.02158046,
      "epoch": 0.6777694273260183,
      "flos": 18003599763840.0,
      "grad_norm": 1.8388325812392945,
      "language_loss": 0.73865318,
      "learning_rate": 9.937106577958481e-07,
      "loss": 0.76002431,
      "num_input_tokens_seen": 243297775,
      "step": 11273,
      "time_per_iteration": 2.4871222972869873
    },
    {
      "auxiliary_loss_clip": 0.01098013,
      "auxiliary_loss_mlp": 0.01042299,
      "balance_loss_clip": 1.03939009,
      "balance_loss_mlp": 1.02987742,
      "epoch": 0.6778295505786863,
      "flos": 23441085624960.0,
      "grad_norm": 1.9395464682107304,
      "language_loss": 0.70598471,
      "learning_rate": 9.933741032359015e-07,
      "loss": 0.72738785,
      "num_input_tokens_seen": 243315760,
      "step": 11274,
      "time_per_iteration": 2.5556960105895996
    },
    {
      "auxiliary_loss_clip": 0.0111118,
      "auxiliary_loss_mlp": 0.01033175,
      "balance_loss_clip": 1.038715,
      "balance_loss_mlp": 1.02058089,
      "epoch": 0.6778896738313542,
      "flos": 19098408349440.0,
      "grad_norm": 1.658397391296205,
      "language_loss": 0.65436578,
      "learning_rate": 9.930375868473093e-07,
      "loss": 0.67580932,
      "num_input_tokens_seen": 243335715,
      "step": 11275,
      "time_per_iteration": 2.508690118789673
    },
    {
      "auxiliary_loss_clip": 0.01100248,
      "auxiliary_loss_mlp": 0.01032043,
      "balance_loss_clip": 1.03983307,
      "balance_loss_mlp": 1.02070642,
      "epoch": 0.6779497970840223,
      "flos": 26103933504000.0,
      "grad_norm": 2.5968849791762154,
      "language_loss": 0.72783303,
      "learning_rate": 9.927011086428335e-07,
      "loss": 0.74915588,
      "num_input_tokens_seen": 243356935,
      "step": 11276,
      "time_per_iteration": 2.5464112758636475
    },
    {
      "auxiliary_loss_clip": 0.01083235,
      "auxiliary_loss_mlp": 0.00783898,
      "balance_loss_clip": 1.03691387,
      "balance_loss_mlp": 1.00857639,
      "epoch": 0.6780099203366902,
      "flos": 19719232041600.0,
      "grad_norm": 1.7086017109518157,
      "language_loss": 0.76773846,
      "learning_rate": 9.923646686352317e-07,
      "loss": 0.7864098,
      "num_input_tokens_seen": 243375625,
      "step": 11277,
      "time_per_iteration": 2.5250627994537354
    },
    {
      "auxiliary_loss_clip": 0.01088594,
      "auxiliary_loss_mlp": 0.0103047,
      "balance_loss_clip": 1.03731298,
      "balance_loss_mlp": 1.01781034,
      "epoch": 0.6780700435893582,
      "flos": 18214538382720.0,
      "grad_norm": 2.6549513351788536,
      "language_loss": 0.83685207,
      "learning_rate": 9.920282668372627e-07,
      "loss": 0.85804272,
      "num_input_tokens_seen": 243390195,
      "step": 11278,
      "time_per_iteration": 2.4845290184020996
    },
    {
      "auxiliary_loss_clip": 0.01077936,
      "auxiliary_loss_mlp": 0.00781669,
      "balance_loss_clip": 1.03755689,
      "balance_loss_mlp": 1.00728488,
      "epoch": 0.6781301668420262,
      "flos": 25376239872000.0,
      "grad_norm": 1.6048465874229025,
      "language_loss": 0.70756793,
      "learning_rate": 9.916919032616844e-07,
      "loss": 0.72616398,
      "num_input_tokens_seen": 243411690,
      "step": 11279,
      "time_per_iteration": 3.9578559398651123
    },
    {
      "auxiliary_loss_clip": 0.01101721,
      "auxiliary_loss_mlp": 0.01032261,
      "balance_loss_clip": 1.04064119,
      "balance_loss_mlp": 1.01906419,
      "epoch": 0.6781902900946941,
      "flos": 24020432087040.0,
      "grad_norm": 2.1978858686198333,
      "language_loss": 0.73653269,
      "learning_rate": 9.913555779212485e-07,
      "loss": 0.75787246,
      "num_input_tokens_seen": 243430280,
      "step": 11280,
      "time_per_iteration": 2.5262610912323
    },
    {
      "auxiliary_loss_clip": 0.01098812,
      "auxiliary_loss_mlp": 0.01031105,
      "balance_loss_clip": 1.03679895,
      "balance_loss_mlp": 1.01798582,
      "epoch": 0.6782504133473621,
      "flos": 19646764352640.0,
      "grad_norm": 1.7627693475441908,
      "language_loss": 0.69765925,
      "learning_rate": 9.910192908287104e-07,
      "loss": 0.71895838,
      "num_input_tokens_seen": 243448690,
      "step": 11281,
      "time_per_iteration": 3.85888671875
    },
    {
      "auxiliary_loss_clip": 0.01105408,
      "auxiliary_loss_mlp": 0.0102916,
      "balance_loss_clip": 1.03735423,
      "balance_loss_mlp": 1.01753139,
      "epoch": 0.67831053660003,
      "flos": 24932742647040.0,
      "grad_norm": 1.5115776096943059,
      "language_loss": 0.63709664,
      "learning_rate": 9.906830419968217e-07,
      "loss": 0.65844238,
      "num_input_tokens_seen": 243470695,
      "step": 11282,
      "time_per_iteration": 2.547520399093628
    },
    {
      "auxiliary_loss_clip": 0.01073788,
      "auxiliary_loss_mlp": 0.01045916,
      "balance_loss_clip": 1.03701115,
      "balance_loss_mlp": 1.02998936,
      "epoch": 0.6783706598526981,
      "flos": 31208383440000.0,
      "grad_norm": 1.5575533717162613,
      "language_loss": 0.7416662,
      "learning_rate": 9.90346831438334e-07,
      "loss": 0.76286322,
      "num_input_tokens_seen": 243493345,
      "step": 11283,
      "time_per_iteration": 4.0113184452056885
    },
    {
      "auxiliary_loss_clip": 0.01099059,
      "auxiliary_loss_mlp": 0.01027761,
      "balance_loss_clip": 1.03813148,
      "balance_loss_mlp": 1.01580453,
      "epoch": 0.678430783105366,
      "flos": 35441317687680.0,
      "grad_norm": 1.5241142442773663,
      "language_loss": 0.57079846,
      "learning_rate": 9.900106591659948e-07,
      "loss": 0.59206665,
      "num_input_tokens_seen": 243515670,
      "step": 11284,
      "time_per_iteration": 2.627899169921875
    },
    {
      "auxiliary_loss_clip": 0.0108468,
      "auxiliary_loss_mlp": 0.01028351,
      "balance_loss_clip": 1.03672862,
      "balance_loss_mlp": 1.01635873,
      "epoch": 0.678490906358034,
      "flos": 14428800460800.0,
      "grad_norm": 2.264045487179026,
      "language_loss": 0.75050914,
      "learning_rate": 9.896745251925535e-07,
      "loss": 0.77163941,
      "num_input_tokens_seen": 243533625,
      "step": 11285,
      "time_per_iteration": 2.496394634246826
    },
    {
      "auxiliary_loss_clip": 0.01107334,
      "auxiliary_loss_mlp": 0.01029987,
      "balance_loss_clip": 1.03958416,
      "balance_loss_mlp": 1.01804769,
      "epoch": 0.6785510296107019,
      "flos": 24311236596480.0,
      "grad_norm": 1.5404605155249622,
      "language_loss": 0.66311276,
      "learning_rate": 9.893384295307557e-07,
      "loss": 0.68448597,
      "num_input_tokens_seen": 243553040,
      "step": 11286,
      "time_per_iteration": 2.5142822265625
    },
    {
      "auxiliary_loss_clip": 0.01085951,
      "auxiliary_loss_mlp": 0.01028972,
      "balance_loss_clip": 1.03610098,
      "balance_loss_mlp": 1.01659179,
      "epoch": 0.6786111528633699,
      "flos": 26977244872320.0,
      "grad_norm": 2.353817619532873,
      "language_loss": 0.5237444,
      "learning_rate": 9.890023721933447e-07,
      "loss": 0.54489362,
      "num_input_tokens_seen": 243572590,
      "step": 11287,
      "time_per_iteration": 2.5641238689422607
    },
    {
      "auxiliary_loss_clip": 0.01064997,
      "auxiliary_loss_mlp": 0.01034058,
      "balance_loss_clip": 1.0366056,
      "balance_loss_mlp": 1.02114117,
      "epoch": 0.6786712761160378,
      "flos": 24317557390080.0,
      "grad_norm": 1.632726162135347,
      "language_loss": 0.77089953,
      "learning_rate": 9.886663531930655e-07,
      "loss": 0.79189008,
      "num_input_tokens_seen": 243594140,
      "step": 11288,
      "time_per_iteration": 2.599918842315674
    },
    {
      "auxiliary_loss_clip": 0.01101803,
      "auxiliary_loss_mlp": 0.01031559,
      "balance_loss_clip": 1.04069889,
      "balance_loss_mlp": 1.01964998,
      "epoch": 0.6787313993687059,
      "flos": 22930435923840.0,
      "grad_norm": 2.6085445459847323,
      "language_loss": 0.73474169,
      "learning_rate": 9.883303725426593e-07,
      "loss": 0.75607538,
      "num_input_tokens_seen": 243615170,
      "step": 11289,
      "time_per_iteration": 2.528892755508423
    },
    {
      "auxiliary_loss_clip": 0.0111037,
      "auxiliary_loss_mlp": 0.0103123,
      "balance_loss_clip": 1.03953648,
      "balance_loss_mlp": 1.0189395,
      "epoch": 0.6787915226213738,
      "flos": 26868435598080.0,
      "grad_norm": 2.633804296201977,
      "language_loss": 0.8003493,
      "learning_rate": 9.879944302548682e-07,
      "loss": 0.8217653,
      "num_input_tokens_seen": 243635675,
      "step": 11290,
      "time_per_iteration": 2.5173683166503906
    },
    {
      "auxiliary_loss_clip": 0.0109449,
      "auxiliary_loss_mlp": 0.01030085,
      "balance_loss_clip": 1.0380677,
      "balance_loss_mlp": 1.01853991,
      "epoch": 0.6788516458740418,
      "flos": 20008851402240.0,
      "grad_norm": 1.4910515264103732,
      "language_loss": 0.75296432,
      "learning_rate": 9.87658526342428e-07,
      "loss": 0.77421004,
      "num_input_tokens_seen": 243654950,
      "step": 11291,
      "time_per_iteration": 2.5094945430755615
    },
    {
      "auxiliary_loss_clip": 0.01087394,
      "auxiliary_loss_mlp": 0.00783182,
      "balance_loss_clip": 1.03688264,
      "balance_loss_mlp": 1.00812459,
      "epoch": 0.6789117691267098,
      "flos": 28727099832960.0,
      "grad_norm": 1.729904044414138,
      "language_loss": 0.75527298,
      "learning_rate": 9.873226608180785e-07,
      "loss": 0.77397871,
      "num_input_tokens_seen": 243674970,
      "step": 11292,
      "time_per_iteration": 2.5904951095581055
    },
    {
      "auxiliary_loss_clip": 0.0106963,
      "auxiliary_loss_mlp": 0.01031719,
      "balance_loss_clip": 1.03809905,
      "balance_loss_mlp": 1.0191009,
      "epoch": 0.6789718923793777,
      "flos": 23403451150080.0,
      "grad_norm": 1.8272803045658337,
      "language_loss": 0.84175199,
      "learning_rate": 9.869868336945556e-07,
      "loss": 0.86276549,
      "num_input_tokens_seen": 243693440,
      "step": 11293,
      "time_per_iteration": 2.5738327503204346
    },
    {
      "auxiliary_loss_clip": 0.01115488,
      "auxiliary_loss_mlp": 0.01038931,
      "balance_loss_clip": 1.04129934,
      "balance_loss_mlp": 1.02562761,
      "epoch": 0.6790320156320457,
      "flos": 20448865008000.0,
      "grad_norm": 2.377145510561933,
      "language_loss": 0.7906661,
      "learning_rate": 9.866510449845929e-07,
      "loss": 0.81221032,
      "num_input_tokens_seen": 243710055,
      "step": 11294,
      "time_per_iteration": 2.488168239593506
    },
    {
      "auxiliary_loss_clip": 0.0108498,
      "auxiliary_loss_mlp": 0.01028211,
      "balance_loss_clip": 1.03740191,
      "balance_loss_mlp": 1.0166955,
      "epoch": 0.6790921388847136,
      "flos": 24167199058560.0,
      "grad_norm": 1.6404121332622654,
      "language_loss": 0.78962719,
      "learning_rate": 9.86315294700924e-07,
      "loss": 0.81075919,
      "num_input_tokens_seen": 243728635,
      "step": 11295,
      "time_per_iteration": 2.5469017028808594
    },
    {
      "auxiliary_loss_clip": 0.01080085,
      "auxiliary_loss_mlp": 0.01029027,
      "balance_loss_clip": 1.03869069,
      "balance_loss_mlp": 1.01853025,
      "epoch": 0.6791522621373817,
      "flos": 21908095027200.0,
      "grad_norm": 1.73703517013498,
      "language_loss": 0.71212947,
      "learning_rate": 9.859795828562823e-07,
      "loss": 0.73322058,
      "num_input_tokens_seen": 243748330,
      "step": 11296,
      "time_per_iteration": 2.5319128036499023
    },
    {
      "auxiliary_loss_clip": 0.01096353,
      "auxiliary_loss_mlp": 0.01029331,
      "balance_loss_clip": 1.03779042,
      "balance_loss_mlp": 1.01744008,
      "epoch": 0.6792123853900496,
      "flos": 24826519152000.0,
      "grad_norm": 1.5398975463998565,
      "language_loss": 0.70905346,
      "learning_rate": 9.856439094633949e-07,
      "loss": 0.73031026,
      "num_input_tokens_seen": 243769380,
      "step": 11297,
      "time_per_iteration": 2.535752058029175
    },
    {
      "auxiliary_loss_clip": 0.01081402,
      "auxiliary_loss_mlp": 0.01031478,
      "balance_loss_clip": 1.03778899,
      "balance_loss_mlp": 1.01838255,
      "epoch": 0.6792725086427176,
      "flos": 17566279678080.0,
      "grad_norm": 2.137310019739984,
      "language_loss": 0.66094947,
      "learning_rate": 9.853082745349918e-07,
      "loss": 0.68207824,
      "num_input_tokens_seen": 243785510,
      "step": 11298,
      "time_per_iteration": 2.5172343254089355
    },
    {
      "auxiliary_loss_clip": 0.01101602,
      "auxiliary_loss_mlp": 0.01027485,
      "balance_loss_clip": 1.04063594,
      "balance_loss_mlp": 1.01586175,
      "epoch": 0.6793326318953855,
      "flos": 26941837040640.0,
      "grad_norm": 2.148893226096472,
      "language_loss": 0.71899366,
      "learning_rate": 9.84972678083801e-07,
      "loss": 0.7402845,
      "num_input_tokens_seen": 243805545,
      "step": 11299,
      "time_per_iteration": 2.536752462387085
    },
    {
      "auxiliary_loss_clip": 0.01110755,
      "auxiliary_loss_mlp": 0.01031713,
      "balance_loss_clip": 1.04001009,
      "balance_loss_mlp": 1.01889205,
      "epoch": 0.6793927551480535,
      "flos": 24318275662080.0,
      "grad_norm": 1.353178559706865,
      "language_loss": 0.77259094,
      "learning_rate": 9.846371201225488e-07,
      "loss": 0.79401559,
      "num_input_tokens_seen": 243825185,
      "step": 11300,
      "time_per_iteration": 2.5052132606506348
    },
    {
      "auxiliary_loss_clip": 0.0109676,
      "auxiliary_loss_mlp": 0.01030214,
      "balance_loss_clip": 1.03802788,
      "balance_loss_mlp": 1.0179112,
      "epoch": 0.6794528784007214,
      "flos": 11436615757440.0,
      "grad_norm": 1.7830687395494682,
      "language_loss": 0.62994963,
      "learning_rate": 9.843016006639577e-07,
      "loss": 0.65121937,
      "num_input_tokens_seen": 243841600,
      "step": 11301,
      "time_per_iteration": 2.4741768836975098
    },
    {
      "auxiliary_loss_clip": 0.01096666,
      "auxiliary_loss_mlp": 0.01027887,
      "balance_loss_clip": 1.03781223,
      "balance_loss_mlp": 1.01597798,
      "epoch": 0.6795130016533895,
      "flos": 25229688382080.0,
      "grad_norm": 1.7008111347600436,
      "language_loss": 0.8289358,
      "learning_rate": 9.839661197207525e-07,
      "loss": 0.85018134,
      "num_input_tokens_seen": 243862250,
      "step": 11302,
      "time_per_iteration": 2.547041893005371
    },
    {
      "auxiliary_loss_clip": 0.01100924,
      "auxiliary_loss_mlp": 0.01032871,
      "balance_loss_clip": 1.03853989,
      "balance_loss_mlp": 1.0206995,
      "epoch": 0.6795731249060574,
      "flos": 18296415434880.0,
      "grad_norm": 1.8492822343041018,
      "language_loss": 0.69587982,
      "learning_rate": 9.83630677305654e-07,
      "loss": 0.7172178,
      "num_input_tokens_seen": 243880560,
      "step": 11303,
      "time_per_iteration": 2.4684371948242188
    },
    {
      "auxiliary_loss_clip": 0.01079833,
      "auxiliary_loss_mlp": 0.01030364,
      "balance_loss_clip": 1.03808653,
      "balance_loss_mlp": 1.01817465,
      "epoch": 0.6796332481587254,
      "flos": 20300374183680.0,
      "grad_norm": 1.9099078014163595,
      "language_loss": 0.69987142,
      "learning_rate": 9.832952734313813e-07,
      "loss": 0.72097337,
      "num_input_tokens_seen": 243900635,
      "step": 11304,
      "time_per_iteration": 2.5655274391174316
    },
    {
      "auxiliary_loss_clip": 0.01099787,
      "auxiliary_loss_mlp": 0.01031914,
      "balance_loss_clip": 1.04047132,
      "balance_loss_mlp": 1.0191704,
      "epoch": 0.6796933714113934,
      "flos": 23586847015680.0,
      "grad_norm": 1.85155893041327,
      "language_loss": 0.72529292,
      "learning_rate": 9.829599081106536e-07,
      "loss": 0.74660993,
      "num_input_tokens_seen": 243920160,
      "step": 11305,
      "time_per_iteration": 2.5372602939605713
    },
    {
      "auxiliary_loss_clip": 0.01087784,
      "auxiliary_loss_mlp": 0.01028124,
      "balance_loss_clip": 1.03835177,
      "balance_loss_mlp": 1.01537418,
      "epoch": 0.6797534946640613,
      "flos": 27119917693440.0,
      "grad_norm": 2.2138651303726786,
      "language_loss": 0.65991563,
      "learning_rate": 9.826245813561882e-07,
      "loss": 0.68107462,
      "num_input_tokens_seen": 243939015,
      "step": 11306,
      "time_per_iteration": 2.5896072387695312
    },
    {
      "auxiliary_loss_clip": 0.01083843,
      "auxiliary_loss_mlp": 0.01028852,
      "balance_loss_clip": 1.0373795,
      "balance_loss_mlp": 1.0157268,
      "epoch": 0.6798136179167293,
      "flos": 22127437428480.0,
      "grad_norm": 1.6580266369311318,
      "language_loss": 0.80147004,
      "learning_rate": 9.822892931807021e-07,
      "loss": 0.82259703,
      "num_input_tokens_seen": 243958470,
      "step": 11307,
      "time_per_iteration": 2.553403615951538
    },
    {
      "auxiliary_loss_clip": 0.01083212,
      "auxiliary_loss_mlp": 0.0103153,
      "balance_loss_clip": 1.03764081,
      "balance_loss_mlp": 1.0182023,
      "epoch": 0.6798737411693972,
      "flos": 17488640430720.0,
      "grad_norm": 1.5159396369257339,
      "language_loss": 0.89017034,
      "learning_rate": 9.819540435969066e-07,
      "loss": 0.91131777,
      "num_input_tokens_seen": 243975450,
      "step": 11308,
      "time_per_iteration": 2.49739670753479
    },
    {
      "auxiliary_loss_clip": 0.01074635,
      "auxiliary_loss_mlp": 0.01036185,
      "balance_loss_clip": 1.0372324,
      "balance_loss_mlp": 1.02182078,
      "epoch": 0.6799338644220653,
      "flos": 22892262744960.0,
      "grad_norm": 1.8680959500455288,
      "language_loss": 0.71436697,
      "learning_rate": 9.816188326175154e-07,
      "loss": 0.73547518,
      "num_input_tokens_seen": 243994355,
      "step": 11309,
      "time_per_iteration": 2.58166241645813
    },
    {
      "auxiliary_loss_clip": 0.01079032,
      "auxiliary_loss_mlp": 0.01037473,
      "balance_loss_clip": 1.03870201,
      "balance_loss_mlp": 1.02503943,
      "epoch": 0.6799939876747332,
      "flos": 23180409648000.0,
      "grad_norm": 1.6627840320014546,
      "language_loss": 0.84366608,
      "learning_rate": 9.812836602552411e-07,
      "loss": 0.86483121,
      "num_input_tokens_seen": 244011620,
      "step": 11310,
      "time_per_iteration": 2.5842363834381104
    },
    {
      "auxiliary_loss_clip": 0.01082058,
      "auxiliary_loss_mlp": 0.01027157,
      "balance_loss_clip": 1.04048085,
      "balance_loss_mlp": 1.01548624,
      "epoch": 0.6800541109274012,
      "flos": 19499925553920.0,
      "grad_norm": 2.3346566499196224,
      "language_loss": 0.82924855,
      "learning_rate": 9.80948526522792e-07,
      "loss": 0.85034072,
      "num_input_tokens_seen": 244029925,
      "step": 11311,
      "time_per_iteration": 3.947524070739746
    },
    {
      "auxiliary_loss_clip": 0.01069683,
      "auxiliary_loss_mlp": 0.01030629,
      "balance_loss_clip": 1.03477478,
      "balance_loss_mlp": 1.01635337,
      "epoch": 0.6801142341800691,
      "flos": 22277652105600.0,
      "grad_norm": 1.9532074566577264,
      "language_loss": 0.7633853,
      "learning_rate": 9.806134314328767e-07,
      "loss": 0.78438842,
      "num_input_tokens_seen": 244051225,
      "step": 11312,
      "time_per_iteration": 2.623706579208374
    },
    {
      "auxiliary_loss_clip": 0.01040385,
      "auxiliary_loss_mlp": 0.01002701,
      "balance_loss_clip": 1.01683259,
      "balance_loss_mlp": 1.00162244,
      "epoch": 0.6801743574327371,
      "flos": 68714817759360.0,
      "grad_norm": 0.6726405090063666,
      "language_loss": 0.57192826,
      "learning_rate": 9.802783749982038e-07,
      "loss": 0.59235913,
      "num_input_tokens_seen": 244115930,
      "step": 11313,
      "time_per_iteration": 3.1924264430999756
    },
    {
      "auxiliary_loss_clip": 0.01098393,
      "auxiliary_loss_mlp": 0.01028376,
      "balance_loss_clip": 1.03727269,
      "balance_loss_mlp": 1.01570952,
      "epoch": 0.680234480685405,
      "flos": 29460467813760.0,
      "grad_norm": 1.7953557966967988,
      "language_loss": 0.68546128,
      "learning_rate": 9.799433572314754e-07,
      "loss": 0.70672899,
      "num_input_tokens_seen": 244137320,
      "step": 11314,
      "time_per_iteration": 2.578650712966919
    },
    {
      "auxiliary_loss_clip": 0.01095249,
      "auxiliary_loss_mlp": 0.01030387,
      "balance_loss_clip": 1.0357101,
      "balance_loss_mlp": 1.01903844,
      "epoch": 0.6802946039380731,
      "flos": 15916866122880.0,
      "grad_norm": 1.7862059790901734,
      "language_loss": 0.81197375,
      "learning_rate": 9.796083781453972e-07,
      "loss": 0.83323014,
      "num_input_tokens_seen": 244152755,
      "step": 11315,
      "time_per_iteration": 2.468475341796875
    },
    {
      "auxiliary_loss_clip": 0.010682,
      "auxiliary_loss_mlp": 0.01030343,
      "balance_loss_clip": 1.03781319,
      "balance_loss_mlp": 1.0175159,
      "epoch": 0.680354727190741,
      "flos": 22018664067840.0,
      "grad_norm": 1.819915167524122,
      "language_loss": 0.70187336,
      "learning_rate": 9.792734377526718e-07,
      "loss": 0.72285879,
      "num_input_tokens_seen": 244171480,
      "step": 11316,
      "time_per_iteration": 2.606863021850586
    },
    {
      "auxiliary_loss_clip": 0.01097202,
      "auxiliary_loss_mlp": 0.01029152,
      "balance_loss_clip": 1.0389272,
      "balance_loss_mlp": 1.01730275,
      "epoch": 0.680414850443409,
      "flos": 18441494467200.0,
      "grad_norm": 2.3346384910493065,
      "language_loss": 0.66890556,
      "learning_rate": 9.789385360660003e-07,
      "loss": 0.6901691,
      "num_input_tokens_seen": 244187920,
      "step": 11317,
      "time_per_iteration": 2.468327045440674
    },
    {
      "auxiliary_loss_clip": 0.01100303,
      "auxiliary_loss_mlp": 0.01041996,
      "balance_loss_clip": 1.04138422,
      "balance_loss_mlp": 1.02968729,
      "epoch": 0.680474973696077,
      "flos": 26358611909760.0,
      "grad_norm": 1.4470042030747343,
      "language_loss": 0.75017303,
      "learning_rate": 9.78603673098082e-07,
      "loss": 0.77159607,
      "num_input_tokens_seen": 244209565,
      "step": 11318,
      "time_per_iteration": 3.980802536010742
    },
    {
      "auxiliary_loss_clip": 0.01078326,
      "auxiliary_loss_mlp": 0.0102978,
      "balance_loss_clip": 1.03511858,
      "balance_loss_mlp": 1.01764417,
      "epoch": 0.6805350969487449,
      "flos": 18333116156160.0,
      "grad_norm": 1.752481762407116,
      "language_loss": 0.67976326,
      "learning_rate": 9.782688488616143e-07,
      "loss": 0.70084429,
      "num_input_tokens_seen": 244228015,
      "step": 11319,
      "time_per_iteration": 3.930938720703125
    },
    {
      "auxiliary_loss_clip": 0.01066684,
      "auxiliary_loss_mlp": 0.00783765,
      "balance_loss_clip": 1.03812683,
      "balance_loss_mlp": 1.01020217,
      "epoch": 0.6805952202014129,
      "flos": 19937497034880.0,
      "grad_norm": 1.5719463127360238,
      "language_loss": 0.76486695,
      "learning_rate": 9.779340633692945e-07,
      "loss": 0.78337145,
      "num_input_tokens_seen": 244245615,
      "step": 11320,
      "time_per_iteration": 2.617776393890381
    },
    {
      "auxiliary_loss_clip": 0.01082389,
      "auxiliary_loss_mlp": 0.01033109,
      "balance_loss_clip": 1.03925347,
      "balance_loss_mlp": 1.02062809,
      "epoch": 0.6806553434540809,
      "flos": 25224301342080.0,
      "grad_norm": 1.767672852683966,
      "language_loss": 0.74750793,
      "learning_rate": 9.77599316633817e-07,
      "loss": 0.76866293,
      "num_input_tokens_seen": 244263625,
      "step": 11321,
      "time_per_iteration": 2.5507652759552
    },
    {
      "auxiliary_loss_clip": 0.01089101,
      "auxiliary_loss_mlp": 0.0103497,
      "balance_loss_clip": 1.03983974,
      "balance_loss_mlp": 1.0228461,
      "epoch": 0.6807154667067489,
      "flos": 17785586165760.0,
      "grad_norm": 1.8064561006328692,
      "language_loss": 0.73274028,
      "learning_rate": 9.772646086678758e-07,
      "loss": 0.75398099,
      "num_input_tokens_seen": 244282745,
      "step": 11322,
      "time_per_iteration": 3.9771735668182373
    },
    {
      "auxiliary_loss_clip": 0.01058416,
      "auxiliary_loss_mlp": 0.00784286,
      "balance_loss_clip": 1.03563952,
      "balance_loss_mlp": 1.00806046,
      "epoch": 0.6807755899594168,
      "flos": 22199905117440.0,
      "grad_norm": 1.6378377296218878,
      "language_loss": 0.78434718,
      "learning_rate": 9.769299394841638e-07,
      "loss": 0.80277425,
      "num_input_tokens_seen": 244303770,
      "step": 11323,
      "time_per_iteration": 2.6198973655700684
    },
    {
      "auxiliary_loss_clip": 0.01004988,
      "auxiliary_loss_mlp": 0.01000746,
      "balance_loss_clip": 1.01535845,
      "balance_loss_mlp": 0.99944705,
      "epoch": 0.6808357132120848,
      "flos": 68631073200000.0,
      "grad_norm": 0.7797215423739108,
      "language_loss": 0.57184803,
      "learning_rate": 9.765953090953714e-07,
      "loss": 0.59190536,
      "num_input_tokens_seen": 244355910,
      "step": 11324,
      "time_per_iteration": 2.9474036693573
    },
    {
      "auxiliary_loss_clip": 0.01090824,
      "auxiliary_loss_mlp": 0.01036381,
      "balance_loss_clip": 1.03919363,
      "balance_loss_mlp": 1.02360153,
      "epoch": 0.6808958364647527,
      "flos": 23843357015040.0,
      "grad_norm": 1.8917357527354923,
      "language_loss": 0.67793894,
      "learning_rate": 9.76260717514186e-07,
      "loss": 0.699211,
      "num_input_tokens_seen": 244376610,
      "step": 11325,
      "time_per_iteration": 2.584002733230591
    },
    {
      "auxiliary_loss_clip": 0.01100804,
      "auxiliary_loss_mlp": 0.01031193,
      "balance_loss_clip": 1.03773487,
      "balance_loss_mlp": 1.0175494,
      "epoch": 0.6809559597174207,
      "flos": 17711717846400.0,
      "grad_norm": 2.2234321127830112,
      "language_loss": 0.70260054,
      "learning_rate": 9.759261647532974e-07,
      "loss": 0.72392046,
      "num_input_tokens_seen": 244393000,
      "step": 11326,
      "time_per_iteration": 2.468193769454956
    },
    {
      "auxiliary_loss_clip": 0.01109435,
      "auxiliary_loss_mlp": 0.01030247,
      "balance_loss_clip": 1.03811026,
      "balance_loss_mlp": 1.01783752,
      "epoch": 0.6810160829700886,
      "flos": 22491894775680.0,
      "grad_norm": 1.7641944298454715,
      "language_loss": 0.72747242,
      "learning_rate": 9.75591650825392e-07,
      "loss": 0.7488693,
      "num_input_tokens_seen": 244409515,
      "step": 11327,
      "time_per_iteration": 2.470832586288452
    },
    {
      "auxiliary_loss_clip": 0.01095245,
      "auxiliary_loss_mlp": 0.01027005,
      "balance_loss_clip": 1.03701615,
      "balance_loss_mlp": 1.01457763,
      "epoch": 0.6810762062227567,
      "flos": 16832875783680.0,
      "grad_norm": 2.2110267127136884,
      "language_loss": 0.77548981,
      "learning_rate": 9.752571757431526e-07,
      "loss": 0.79671234,
      "num_input_tokens_seen": 244427165,
      "step": 11328,
      "time_per_iteration": 2.4907774925231934
    },
    {
      "auxiliary_loss_clip": 0.0110952,
      "auxiliary_loss_mlp": 0.01030277,
      "balance_loss_clip": 1.03842664,
      "balance_loss_mlp": 1.0177182,
      "epoch": 0.6811363294754246,
      "flos": 12714676554240.0,
      "grad_norm": 1.7972976167893433,
      "language_loss": 0.64053321,
      "learning_rate": 9.74922739519265e-07,
      "loss": 0.66193122,
      "num_input_tokens_seen": 244445705,
      "step": 11329,
      "time_per_iteration": 2.463643789291382
    },
    {
      "auxiliary_loss_clip": 0.01054594,
      "auxiliary_loss_mlp": 0.0078534,
      "balance_loss_clip": 1.03984284,
      "balance_loss_mlp": 1.01065469,
      "epoch": 0.6811964527280926,
      "flos": 17711969241600.0,
      "grad_norm": 2.32060578602773,
      "language_loss": 0.79223466,
      "learning_rate": 9.745883421664096e-07,
      "loss": 0.8106339,
      "num_input_tokens_seen": 244460415,
      "step": 11330,
      "time_per_iteration": 2.5927722454071045
    },
    {
      "auxiliary_loss_clip": 0.01097879,
      "auxiliary_loss_mlp": 0.01029578,
      "balance_loss_clip": 1.03905749,
      "balance_loss_mlp": 1.01736474,
      "epoch": 0.6812565759807605,
      "flos": 24863471268480.0,
      "grad_norm": 1.8904605851726493,
      "language_loss": 0.63908291,
      "learning_rate": 9.742539836972665e-07,
      "loss": 0.66035748,
      "num_input_tokens_seen": 244480555,
      "step": 11331,
      "time_per_iteration": 2.545167922973633
    },
    {
      "auxiliary_loss_clip": 0.01062222,
      "auxiliary_loss_mlp": 0.01036265,
      "balance_loss_clip": 1.03728533,
      "balance_loss_mlp": 1.02191257,
      "epoch": 0.6813166992334285,
      "flos": 17166019449600.0,
      "grad_norm": 1.5712565503147384,
      "language_loss": 0.72481,
      "learning_rate": 9.739196641245148e-07,
      "loss": 0.74579489,
      "num_input_tokens_seen": 244498540,
      "step": 11332,
      "time_per_iteration": 2.5672414302825928
    },
    {
      "auxiliary_loss_clip": 0.01099028,
      "auxiliary_loss_mlp": 0.01032209,
      "balance_loss_clip": 1.0414741,
      "balance_loss_mlp": 1.01950085,
      "epoch": 0.6813768224860965,
      "flos": 18843550375680.0,
      "grad_norm": 2.198148990591329,
      "language_loss": 0.74718726,
      "learning_rate": 9.735853834608326e-07,
      "loss": 0.76849961,
      "num_input_tokens_seen": 244517015,
      "step": 11333,
      "time_per_iteration": 2.4991281032562256
    },
    {
      "auxiliary_loss_clip": 0.01101331,
      "auxiliary_loss_mlp": 0.01029407,
      "balance_loss_clip": 1.03947449,
      "balance_loss_mlp": 1.0164485,
      "epoch": 0.6814369457387645,
      "flos": 24532733813760.0,
      "grad_norm": 1.5281602425151177,
      "language_loss": 0.72173041,
      "learning_rate": 9.732511417188963e-07,
      "loss": 0.74303782,
      "num_input_tokens_seen": 244537450,
      "step": 11334,
      "time_per_iteration": 2.5361852645874023
    },
    {
      "auxiliary_loss_clip": 0.01093795,
      "auxiliary_loss_mlp": 0.01031294,
      "balance_loss_clip": 1.04000282,
      "balance_loss_mlp": 1.01941502,
      "epoch": 0.6814970689914325,
      "flos": 18222978078720.0,
      "grad_norm": 1.8316761165276925,
      "language_loss": 0.86018687,
      "learning_rate": 9.729169389113791e-07,
      "loss": 0.88143778,
      "num_input_tokens_seen": 244555640,
      "step": 11335,
      "time_per_iteration": 2.495326519012451
    },
    {
      "auxiliary_loss_clip": 0.01090138,
      "auxiliary_loss_mlp": 0.01030014,
      "balance_loss_clip": 1.03561282,
      "balance_loss_mlp": 1.01862955,
      "epoch": 0.6815571922441004,
      "flos": 25228790542080.0,
      "grad_norm": 1.7131706427574538,
      "language_loss": 0.82225895,
      "learning_rate": 9.725827750509542e-07,
      "loss": 0.8434605,
      "num_input_tokens_seen": 244574005,
      "step": 11336,
      "time_per_iteration": 2.5380184650421143
    },
    {
      "auxiliary_loss_clip": 0.01065252,
      "auxiliary_loss_mlp": 0.01029359,
      "balance_loss_clip": 1.03850031,
      "balance_loss_mlp": 1.01767099,
      "epoch": 0.6816173154967684,
      "flos": 19456078026240.0,
      "grad_norm": 1.918348983684643,
      "language_loss": 0.81621361,
      "learning_rate": 9.72248650150294e-07,
      "loss": 0.83715975,
      "num_input_tokens_seen": 244591395,
      "step": 11337,
      "time_per_iteration": 2.564608097076416
    },
    {
      "auxiliary_loss_clip": 0.0106218,
      "auxiliary_loss_mlp": 0.01028783,
      "balance_loss_clip": 1.03797662,
      "balance_loss_mlp": 1.01690936,
      "epoch": 0.6816774387494363,
      "flos": 17931455297280.0,
      "grad_norm": 1.6496382397559601,
      "language_loss": 0.7265414,
      "learning_rate": 9.719145642220673e-07,
      "loss": 0.74745095,
      "num_input_tokens_seen": 244610400,
      "step": 11338,
      "time_per_iteration": 2.583285331726074
    },
    {
      "auxiliary_loss_clip": 0.01065471,
      "auxiliary_loss_mlp": 0.01037713,
      "balance_loss_clip": 1.03513169,
      "balance_loss_mlp": 1.02502298,
      "epoch": 0.6817375620021043,
      "flos": 22233014478720.0,
      "grad_norm": 1.4971786125462128,
      "language_loss": 0.77644396,
      "learning_rate": 9.715805172789435e-07,
      "loss": 0.79747576,
      "num_input_tokens_seen": 244630400,
      "step": 11339,
      "time_per_iteration": 2.5944504737854004
    },
    {
      "auxiliary_loss_clip": 0.01075877,
      "auxiliary_loss_mlp": 0.01032186,
      "balance_loss_clip": 1.03746629,
      "balance_loss_mlp": 1.01984739,
      "epoch": 0.6817976852547722,
      "flos": 25374408278400.0,
      "grad_norm": 2.107595286540718,
      "language_loss": 0.70708972,
      "learning_rate": 9.712465093335901e-07,
      "loss": 0.72817028,
      "num_input_tokens_seen": 244649155,
      "step": 11340,
      "time_per_iteration": 2.603073835372925
    },
    {
      "auxiliary_loss_clip": 0.01088816,
      "auxiliary_loss_mlp": 0.01033854,
      "balance_loss_clip": 1.03917527,
      "balance_loss_mlp": 1.02212334,
      "epoch": 0.6818578085074403,
      "flos": 22265764704000.0,
      "grad_norm": 3.0919727721284187,
      "language_loss": 0.83444768,
      "learning_rate": 9.709125403986722e-07,
      "loss": 0.85567439,
      "num_input_tokens_seen": 244665470,
      "step": 11341,
      "time_per_iteration": 2.5407416820526123
    },
    {
      "auxiliary_loss_clip": 0.01075115,
      "auxiliary_loss_mlp": 0.01034773,
      "balance_loss_clip": 1.03949237,
      "balance_loss_mlp": 1.02148724,
      "epoch": 0.6819179317601082,
      "flos": 19318145800320.0,
      "grad_norm": 1.7484711111410332,
      "language_loss": 0.68129337,
      "learning_rate": 9.705786104868531e-07,
      "loss": 0.70239222,
      "num_input_tokens_seen": 244684390,
      "step": 11342,
      "time_per_iteration": 2.5858495235443115
    },
    {
      "auxiliary_loss_clip": 0.01061678,
      "auxiliary_loss_mlp": 0.01025698,
      "balance_loss_clip": 1.03961182,
      "balance_loss_mlp": 1.01389647,
      "epoch": 0.6819780550127762,
      "flos": 21104126864640.0,
      "grad_norm": 1.6705947512339279,
      "language_loss": 0.75156009,
      "learning_rate": 9.702447196107963e-07,
      "loss": 0.77243388,
      "num_input_tokens_seen": 244703370,
      "step": 11343,
      "time_per_iteration": 2.594759225845337
    },
    {
      "auxiliary_loss_clip": 0.01069195,
      "auxiliary_loss_mlp": 0.01040065,
      "balance_loss_clip": 1.03855479,
      "balance_loss_mlp": 1.02624261,
      "epoch": 0.6820381782654441,
      "flos": 29716403195520.0,
      "grad_norm": 1.5452535094416928,
      "language_loss": 0.79770076,
      "learning_rate": 9.699108677831639e-07,
      "loss": 0.81879342,
      "num_input_tokens_seen": 244723325,
      "step": 11344,
      "time_per_iteration": 2.6296212673187256
    },
    {
      "auxiliary_loss_clip": 0.01072494,
      "auxiliary_loss_mlp": 0.01034267,
      "balance_loss_clip": 1.03777766,
      "balance_loss_mlp": 1.02190471,
      "epoch": 0.6820983015181121,
      "flos": 29242130993280.0,
      "grad_norm": 2.079376972446784,
      "language_loss": 0.6657576,
      "learning_rate": 9.695770550166136e-07,
      "loss": 0.68682516,
      "num_input_tokens_seen": 244745650,
      "step": 11345,
      "time_per_iteration": 2.6608028411865234
    },
    {
      "auxiliary_loss_clip": 0.01088922,
      "auxiliary_loss_mlp": 0.01034172,
      "balance_loss_clip": 1.03829432,
      "balance_loss_mlp": 1.02187562,
      "epoch": 0.6821584247707801,
      "flos": 18871775487360.0,
      "grad_norm": 2.2982905693772193,
      "language_loss": 0.65017307,
      "learning_rate": 9.692432813238054e-07,
      "loss": 0.671404,
      "num_input_tokens_seen": 244760270,
      "step": 11346,
      "time_per_iteration": 2.5151944160461426
    },
    {
      "auxiliary_loss_clip": 0.01046362,
      "auxiliary_loss_mlp": 0.00788847,
      "balance_loss_clip": 1.03304911,
      "balance_loss_mlp": 1.01023161,
      "epoch": 0.6822185480234481,
      "flos": 21324582587520.0,
      "grad_norm": 1.6266741612945521,
      "language_loss": 0.78512639,
      "learning_rate": 9.689095467173952e-07,
      "loss": 0.80347848,
      "num_input_tokens_seen": 244779565,
      "step": 11347,
      "time_per_iteration": 2.6327075958251953
    },
    {
      "auxiliary_loss_clip": 0.01034309,
      "auxiliary_loss_mlp": 0.01007326,
      "balance_loss_clip": 1.02003956,
      "balance_loss_mlp": 1.00626516,
      "epoch": 0.6822786712761161,
      "flos": 63488306430720.0,
      "grad_norm": 0.7972561555383708,
      "language_loss": 0.52531981,
      "learning_rate": 9.685758512100378e-07,
      "loss": 0.54573619,
      "num_input_tokens_seen": 244838480,
      "step": 11348,
      "time_per_iteration": 3.0992941856384277
    },
    {
      "auxiliary_loss_clip": 0.01106358,
      "auxiliary_loss_mlp": 0.01033273,
      "balance_loss_clip": 1.03746831,
      "balance_loss_mlp": 1.02195454,
      "epoch": 0.682338794528784,
      "flos": 21068934514560.0,
      "grad_norm": 1.7735176642726107,
      "language_loss": 0.79889691,
      "learning_rate": 9.682421948143873e-07,
      "loss": 0.82029319,
      "num_input_tokens_seen": 244855265,
      "step": 11349,
      "time_per_iteration": 2.4684245586395264
    },
    {
      "auxiliary_loss_clip": 0.01103746,
      "auxiliary_loss_mlp": 0.01028953,
      "balance_loss_clip": 1.03963041,
      "balance_loss_mlp": 1.01445687,
      "epoch": 0.682398917781452,
      "flos": 36283243547520.0,
      "grad_norm": 1.738595501565156,
      "language_loss": 0.74083048,
      "learning_rate": 9.67908577543096e-07,
      "loss": 0.76215744,
      "num_input_tokens_seen": 244875555,
      "step": 11350,
      "time_per_iteration": 4.031501054763794
    },
    {
      "auxiliary_loss_clip": 0.01109196,
      "auxiliary_loss_mlp": 0.01030694,
      "balance_loss_clip": 1.03981709,
      "balance_loss_mlp": 1.01803982,
      "epoch": 0.6824590410341199,
      "flos": 24859197550080.0,
      "grad_norm": 2.6012320625598915,
      "language_loss": 0.79583561,
      "learning_rate": 9.675749994088161e-07,
      "loss": 0.81723452,
      "num_input_tokens_seen": 244895270,
      "step": 11351,
      "time_per_iteration": 2.503495931625366
    },
    {
      "auxiliary_loss_clip": 0.01095326,
      "auxiliary_loss_mlp": 0.01032304,
      "balance_loss_clip": 1.037503,
      "balance_loss_mlp": 1.02043688,
      "epoch": 0.6825191642867879,
      "flos": 22452392793600.0,
      "grad_norm": 1.742169009480748,
      "language_loss": 0.7348603,
      "learning_rate": 9.672414604241954e-07,
      "loss": 0.75613666,
      "num_input_tokens_seen": 244914535,
      "step": 11352,
      "time_per_iteration": 2.515620231628418
    },
    {
      "auxiliary_loss_clip": 0.01064416,
      "auxiliary_loss_mlp": 0.01036509,
      "balance_loss_clip": 1.03673625,
      "balance_loss_mlp": 1.02307463,
      "epoch": 0.6825792875394558,
      "flos": 29424377623680.0,
      "grad_norm": 1.4123191336238536,
      "language_loss": 0.79971373,
      "learning_rate": 9.669079606018814e-07,
      "loss": 0.820723,
      "num_input_tokens_seen": 244936095,
      "step": 11353,
      "time_per_iteration": 2.6720385551452637
    },
    {
      "auxiliary_loss_clip": 0.01099518,
      "auxiliary_loss_mlp": 0.01025603,
      "balance_loss_clip": 1.03843141,
      "balance_loss_mlp": 1.01349711,
      "epoch": 0.6826394107921239,
      "flos": 18770974945920.0,
      "grad_norm": 1.8151097952738282,
      "language_loss": 0.78500664,
      "learning_rate": 9.665744999545218e-07,
      "loss": 0.80625784,
      "num_input_tokens_seen": 244955290,
      "step": 11354,
      "time_per_iteration": 2.519303321838379
    },
    {
      "auxiliary_loss_clip": 0.01048615,
      "auxiliary_loss_mlp": 0.01025831,
      "balance_loss_clip": 1.03794158,
      "balance_loss_mlp": 1.01429737,
      "epoch": 0.6826995340447918,
      "flos": 16617591619200.0,
      "grad_norm": 3.093361099137183,
      "language_loss": 0.62438875,
      "learning_rate": 9.662410784947599e-07,
      "loss": 0.6451332,
      "num_input_tokens_seen": 244972935,
      "step": 11355,
      "time_per_iteration": 2.6056294441223145
    },
    {
      "auxiliary_loss_clip": 0.01056004,
      "auxiliary_loss_mlp": 0.01031237,
      "balance_loss_clip": 1.03393865,
      "balance_loss_mlp": 1.01874971,
      "epoch": 0.6827596572974598,
      "flos": 20848299223680.0,
      "grad_norm": 2.297275313866363,
      "language_loss": 0.82370067,
      "learning_rate": 9.659076962352398e-07,
      "loss": 0.84457302,
      "num_input_tokens_seen": 244989440,
      "step": 11356,
      "time_per_iteration": 2.599545478820801
    },
    {
      "auxiliary_loss_clip": 0.01090238,
      "auxiliary_loss_mlp": 0.01029568,
      "balance_loss_clip": 1.03895962,
      "balance_loss_mlp": 1.01721799,
      "epoch": 0.6828197805501277,
      "flos": 22748081552640.0,
      "grad_norm": 1.8048899478883387,
      "language_loss": 0.78528112,
      "learning_rate": 9.655743531886052e-07,
      "loss": 0.80647922,
      "num_input_tokens_seen": 245007830,
      "step": 11357,
      "time_per_iteration": 3.951115131378174
    },
    {
      "auxiliary_loss_clip": 0.01026009,
      "auxiliary_loss_mlp": 0.01015501,
      "balance_loss_clip": 1.02150726,
      "balance_loss_mlp": 1.01396322,
      "epoch": 0.6828799038027957,
      "flos": 71646565829760.0,
      "grad_norm": 0.8253304017308701,
      "language_loss": 0.59631574,
      "learning_rate": 9.65241049367493e-07,
      "loss": 0.61673081,
      "num_input_tokens_seen": 245070720,
      "step": 11358,
      "time_per_iteration": 4.559441804885864
    },
    {
      "auxiliary_loss_clip": 0.01073793,
      "auxiliary_loss_mlp": 0.01043988,
      "balance_loss_clip": 1.03598976,
      "balance_loss_mlp": 1.02833617,
      "epoch": 0.6829400270554637,
      "flos": 19829154637440.0,
      "grad_norm": 1.6460912612489027,
      "language_loss": 0.78148782,
      "learning_rate": 9.64907784784544e-07,
      "loss": 0.80266565,
      "num_input_tokens_seen": 245089070,
      "step": 11359,
      "time_per_iteration": 2.547093152999878
    },
    {
      "auxiliary_loss_clip": 0.01097999,
      "auxiliary_loss_mlp": 0.01035025,
      "balance_loss_clip": 1.03795624,
      "balance_loss_mlp": 1.02295542,
      "epoch": 0.6830001503081317,
      "flos": 21980634543360.0,
      "grad_norm": 1.9944399875935386,
      "language_loss": 0.81711262,
      "learning_rate": 9.645745594523958e-07,
      "loss": 0.83844286,
      "num_input_tokens_seen": 245106500,
      "step": 11360,
      "time_per_iteration": 2.5164954662323
    },
    {
      "auxiliary_loss_clip": 0.01098589,
      "auxiliary_loss_mlp": 0.01041626,
      "balance_loss_clip": 1.04038978,
      "balance_loss_mlp": 1.0273509,
      "epoch": 0.6830602735607997,
      "flos": 24316767290880.0,
      "grad_norm": 1.7543633563900936,
      "language_loss": 0.75226271,
      "learning_rate": 9.642413733836844e-07,
      "loss": 0.77366495,
      "num_input_tokens_seen": 245125260,
      "step": 11361,
      "time_per_iteration": 3.8951685428619385
    },
    {
      "auxiliary_loss_clip": 0.01026905,
      "auxiliary_loss_mlp": 0.01001758,
      "balance_loss_clip": 1.02968693,
      "balance_loss_mlp": 1.00010133,
      "epoch": 0.6831203968134676,
      "flos": 57690062323200.0,
      "grad_norm": 0.8736122754526313,
      "language_loss": 0.59712279,
      "learning_rate": 9.639082265910437e-07,
      "loss": 0.61740941,
      "num_input_tokens_seen": 245188730,
      "step": 11362,
      "time_per_iteration": 3.2457966804504395
    },
    {
      "auxiliary_loss_clip": 0.01084027,
      "auxiliary_loss_mlp": 0.01031305,
      "balance_loss_clip": 1.03762841,
      "balance_loss_mlp": 1.01829326,
      "epoch": 0.6831805200661356,
      "flos": 14388436552320.0,
      "grad_norm": 2.013118795762214,
      "language_loss": 0.75559527,
      "learning_rate": 9.635751190871074e-07,
      "loss": 0.77674854,
      "num_input_tokens_seen": 245205065,
      "step": 11363,
      "time_per_iteration": 2.5156185626983643
    },
    {
      "auxiliary_loss_clip": 0.01085886,
      "auxiliary_loss_mlp": 0.01037594,
      "balance_loss_clip": 1.03690422,
      "balance_loss_mlp": 1.02457058,
      "epoch": 0.6832406433188035,
      "flos": 22820297846400.0,
      "grad_norm": 2.3034784906298116,
      "language_loss": 0.88843584,
      "learning_rate": 9.632420508845063e-07,
      "loss": 0.90967065,
      "num_input_tokens_seen": 245224265,
      "step": 11364,
      "time_per_iteration": 2.6027309894561768
    },
    {
      "auxiliary_loss_clip": 0.01085488,
      "auxiliary_loss_mlp": 0.01030007,
      "balance_loss_clip": 1.03705812,
      "balance_loss_mlp": 1.01824105,
      "epoch": 0.6833007665714715,
      "flos": 17561718650880.0,
      "grad_norm": 1.829993237633201,
      "language_loss": 0.87884098,
      "learning_rate": 9.629090219958697e-07,
      "loss": 0.89999592,
      "num_input_tokens_seen": 245243360,
      "step": 11365,
      "time_per_iteration": 2.5539305210113525
    },
    {
      "auxiliary_loss_clip": 0.010746,
      "auxiliary_loss_mlp": 0.0104208,
      "balance_loss_clip": 1.03959465,
      "balance_loss_mlp": 1.02779245,
      "epoch": 0.6833608898241395,
      "flos": 22445928345600.0,
      "grad_norm": 2.160290711893958,
      "language_loss": 0.80838954,
      "learning_rate": 9.625760324338272e-07,
      "loss": 0.82955635,
      "num_input_tokens_seen": 245256350,
      "step": 11366,
      "time_per_iteration": 2.59409499168396
    },
    {
      "auxiliary_loss_clip": 0.01087494,
      "auxiliary_loss_mlp": 0.010276,
      "balance_loss_clip": 1.0372858,
      "balance_loss_mlp": 1.01573873,
      "epoch": 0.6834210130768075,
      "flos": 24534637234560.0,
      "grad_norm": 1.3894193812855193,
      "language_loss": 0.765728,
      "learning_rate": 9.622430822110062e-07,
      "loss": 0.78687888,
      "num_input_tokens_seen": 245277575,
      "step": 11367,
      "time_per_iteration": 2.580702543258667
    },
    {
      "auxiliary_loss_clip": 0.01083542,
      "auxiliary_loss_mlp": 0.01034899,
      "balance_loss_clip": 1.04022753,
      "balance_loss_mlp": 1.02219093,
      "epoch": 0.6834811363294754,
      "flos": 20047132321920.0,
      "grad_norm": 1.4337130130469145,
      "language_loss": 0.68759072,
      "learning_rate": 9.619101713400312e-07,
      "loss": 0.70877516,
      "num_input_tokens_seen": 245296615,
      "step": 11368,
      "time_per_iteration": 2.5280230045318604
    },
    {
      "auxiliary_loss_clip": 0.01072724,
      "auxiliary_loss_mlp": 0.01034831,
      "balance_loss_clip": 1.03652501,
      "balance_loss_mlp": 1.02256465,
      "epoch": 0.6835412595821434,
      "flos": 24790752184320.0,
      "grad_norm": 1.7418089827258016,
      "language_loss": 0.73126447,
      "learning_rate": 9.615772998335261e-07,
      "loss": 0.75234008,
      "num_input_tokens_seen": 245316275,
      "step": 11369,
      "time_per_iteration": 2.6257810592651367
    },
    {
      "auxiliary_loss_clip": 0.01094768,
      "auxiliary_loss_mlp": 0.01029793,
      "balance_loss_clip": 1.03916466,
      "balance_loss_mlp": 1.01758575,
      "epoch": 0.6836013828348113,
      "flos": 19500356517120.0,
      "grad_norm": 1.9916695327197576,
      "language_loss": 0.79240727,
      "learning_rate": 9.612444677041138e-07,
      "loss": 0.81365281,
      "num_input_tokens_seen": 245334595,
      "step": 11370,
      "time_per_iteration": 2.512848377227783
    },
    {
      "auxiliary_loss_clip": 0.01033342,
      "auxiliary_loss_mlp": 0.01000816,
      "balance_loss_clip": 1.01909912,
      "balance_loss_mlp": 0.99958223,
      "epoch": 0.6836615060874793,
      "flos": 58363999251840.0,
      "grad_norm": 0.7455881973849126,
      "language_loss": 0.59805,
      "learning_rate": 9.609116749644162e-07,
      "loss": 0.61839163,
      "num_input_tokens_seen": 245389750,
      "step": 11371,
      "time_per_iteration": 3.042795181274414
    },
    {
      "auxiliary_loss_clip": 0.01080399,
      "auxiliary_loss_mlp": 0.0102586,
      "balance_loss_clip": 1.03813672,
      "balance_loss_mlp": 1.01443934,
      "epoch": 0.6837216293401474,
      "flos": 12166895168640.0,
      "grad_norm": 2.2867942593909842,
      "language_loss": 0.63596565,
      "learning_rate": 9.605789216270511e-07,
      "loss": 0.6570282,
      "num_input_tokens_seen": 245407530,
      "step": 11372,
      "time_per_iteration": 2.5345940589904785
    },
    {
      "auxiliary_loss_clip": 0.01096055,
      "auxiliary_loss_mlp": 0.0102583,
      "balance_loss_clip": 1.04106116,
      "balance_loss_mlp": 1.01366496,
      "epoch": 0.6837817525928153,
      "flos": 22127581082880.0,
      "grad_norm": 1.5448013075167855,
      "language_loss": 0.71649814,
      "learning_rate": 9.602462077046375e-07,
      "loss": 0.73771703,
      "num_input_tokens_seen": 245427000,
      "step": 11373,
      "time_per_iteration": 2.5215413570404053
    },
    {
      "auxiliary_loss_clip": 0.01016624,
      "auxiliary_loss_mlp": 0.01002736,
      "balance_loss_clip": 1.02275836,
      "balance_loss_mlp": 1.00150192,
      "epoch": 0.6838418758454833,
      "flos": 65005928985600.0,
      "grad_norm": 1.181953799033402,
      "language_loss": 0.56667852,
      "learning_rate": 9.599135332097935e-07,
      "loss": 0.5868721,
      "num_input_tokens_seen": 245491620,
      "step": 11374,
      "time_per_iteration": 3.328202247619629
    },
    {
      "auxiliary_loss_clip": 0.01100091,
      "auxiliary_loss_mlp": 0.01024921,
      "balance_loss_clip": 1.03909492,
      "balance_loss_mlp": 1.01192713,
      "epoch": 0.6839019990981512,
      "flos": 21030833162880.0,
      "grad_norm": 1.5320337329056335,
      "language_loss": 0.73462367,
      "learning_rate": 9.595808981551312e-07,
      "loss": 0.7558738,
      "num_input_tokens_seen": 245511285,
      "step": 11375,
      "time_per_iteration": 2.540921449661255
    },
    {
      "auxiliary_loss_clip": 0.01087643,
      "auxiliary_loss_mlp": 0.01029436,
      "balance_loss_clip": 1.03876781,
      "balance_loss_mlp": 1.01765227,
      "epoch": 0.6839621223508192,
      "flos": 24935543907840.0,
      "grad_norm": 1.7868558021118364,
      "language_loss": 0.7068814,
      "learning_rate": 9.592483025532651e-07,
      "loss": 0.72805214,
      "num_input_tokens_seen": 245532910,
      "step": 11376,
      "time_per_iteration": 2.6083621978759766
    },
    {
      "auxiliary_loss_clip": 0.01111528,
      "auxiliary_loss_mlp": 0.01031669,
      "balance_loss_clip": 1.03941512,
      "balance_loss_mlp": 1.01911676,
      "epoch": 0.6840222456034871,
      "flos": 26358827391360.0,
      "grad_norm": 2.024560998065436,
      "language_loss": 0.74704027,
      "learning_rate": 9.58915746416808e-07,
      "loss": 0.76847225,
      "num_input_tokens_seen": 245550540,
      "step": 11377,
      "time_per_iteration": 2.512059211730957
    },
    {
      "auxiliary_loss_clip": 0.0102507,
      "auxiliary_loss_mlp": 0.01001488,
      "balance_loss_clip": 1.01996422,
      "balance_loss_mlp": 1.00016451,
      "epoch": 0.6840823688561551,
      "flos": 65988336936960.0,
      "grad_norm": 0.7432249475106598,
      "language_loss": 0.56819725,
      "learning_rate": 9.585832297583707e-07,
      "loss": 0.58846283,
      "num_input_tokens_seen": 245619570,
      "step": 11378,
      "time_per_iteration": 3.2503654956817627
    },
    {
      "auxiliary_loss_clip": 0.01110055,
      "auxiliary_loss_mlp": 0.01033149,
      "balance_loss_clip": 1.03868163,
      "balance_loss_mlp": 1.02085805,
      "epoch": 0.684142492108823,
      "flos": 21397588980480.0,
      "grad_norm": 1.829940400253596,
      "language_loss": 0.7836839,
      "learning_rate": 9.58250752590561e-07,
      "loss": 0.80511594,
      "num_input_tokens_seen": 245637980,
      "step": 11379,
      "time_per_iteration": 2.462859869003296
    },
    {
      "auxiliary_loss_clip": 0.01104389,
      "auxiliary_loss_mlp": 0.01026321,
      "balance_loss_clip": 1.03928351,
      "balance_loss_mlp": 1.01545465,
      "epoch": 0.6842026153614911,
      "flos": 18801426700800.0,
      "grad_norm": 1.7543472145137915,
      "language_loss": 0.68883812,
      "learning_rate": 9.57918314925988e-07,
      "loss": 0.71014524,
      "num_input_tokens_seen": 245655690,
      "step": 11380,
      "time_per_iteration": 2.4678092002868652
    },
    {
      "auxiliary_loss_clip": 0.01081955,
      "auxiliary_loss_mlp": 0.01032349,
      "balance_loss_clip": 1.03509045,
      "balance_loss_mlp": 1.02004695,
      "epoch": 0.684262738614159,
      "flos": 19646405216640.0,
      "grad_norm": 1.8137124533758013,
      "language_loss": 0.78094637,
      "learning_rate": 9.575859167772568e-07,
      "loss": 0.80208939,
      "num_input_tokens_seen": 245671525,
      "step": 11381,
      "time_per_iteration": 2.517038345336914
    },
    {
      "auxiliary_loss_clip": 0.01030891,
      "auxiliary_loss_mlp": 0.01000374,
      "balance_loss_clip": 1.01793683,
      "balance_loss_mlp": 0.99906886,
      "epoch": 0.684322861866827,
      "flos": 62354462739840.0,
      "grad_norm": 0.9011887082202911,
      "language_loss": 0.671359,
      "learning_rate": 9.572535581569713e-07,
      "loss": 0.69167167,
      "num_input_tokens_seen": 245724115,
      "step": 11382,
      "time_per_iteration": 2.966581344604492
    },
    {
      "auxiliary_loss_clip": 0.01034154,
      "auxiliary_loss_mlp": 0.01002146,
      "balance_loss_clip": 1.02082193,
      "balance_loss_mlp": 1.00106132,
      "epoch": 0.6843829851194949,
      "flos": 65805048812160.0,
      "grad_norm": 0.8162837053982014,
      "language_loss": 0.58098197,
      "learning_rate": 9.569212390777356e-07,
      "loss": 0.60134494,
      "num_input_tokens_seen": 245789245,
      "step": 11383,
      "time_per_iteration": 3.1445367336273193
    },
    {
      "auxiliary_loss_clip": 0.01057001,
      "auxiliary_loss_mlp": 0.01029528,
      "balance_loss_clip": 1.03648496,
      "balance_loss_mlp": 1.01663613,
      "epoch": 0.6844431083721629,
      "flos": 27855153181440.0,
      "grad_norm": 1.7851111497915635,
      "language_loss": 0.79878062,
      "learning_rate": 9.565889595521517e-07,
      "loss": 0.81964594,
      "num_input_tokens_seen": 245812420,
      "step": 11384,
      "time_per_iteration": 2.646496295928955
    },
    {
      "auxiliary_loss_clip": 0.0109685,
      "auxiliary_loss_mlp": 0.01033558,
      "balance_loss_clip": 1.03658664,
      "balance_loss_mlp": 1.02141654,
      "epoch": 0.684503231624831,
      "flos": 18255010032000.0,
      "grad_norm": 2.0098304209651845,
      "language_loss": 0.77383798,
      "learning_rate": 9.562567195928187e-07,
      "loss": 0.79514205,
      "num_input_tokens_seen": 245829135,
      "step": 11385,
      "time_per_iteration": 2.463932991027832
    },
    {
      "auxiliary_loss_clip": 0.01076556,
      "auxiliary_loss_mlp": 0.01035219,
      "balance_loss_clip": 1.03820896,
      "balance_loss_mlp": 1.02159929,
      "epoch": 0.6845633548774989,
      "flos": 17639681120640.0,
      "grad_norm": 2.1078684767727665,
      "language_loss": 0.84392172,
      "learning_rate": 9.55924519212335e-07,
      "loss": 0.86503947,
      "num_input_tokens_seen": 245847140,
      "step": 11386,
      "time_per_iteration": 2.5560219287872314
    },
    {
      "auxiliary_loss_clip": 0.01098611,
      "auxiliary_loss_mlp": 0.01038926,
      "balance_loss_clip": 1.03967607,
      "balance_loss_mlp": 1.02738023,
      "epoch": 0.6846234781301669,
      "flos": 20807576179200.0,
      "grad_norm": 1.9916059577906498,
      "language_loss": 0.83388048,
      "learning_rate": 9.555923584232984e-07,
      "loss": 0.85525584,
      "num_input_tokens_seen": 245862855,
      "step": 11387,
      "time_per_iteration": 2.489928722381592
    },
    {
      "auxiliary_loss_clip": 0.01089618,
      "auxiliary_loss_mlp": 0.01029691,
      "balance_loss_clip": 1.03529549,
      "balance_loss_mlp": 1.0175674,
      "epoch": 0.6846836013828348,
      "flos": 36101176485120.0,
      "grad_norm": 1.8323405784119928,
      "language_loss": 0.72747052,
      "learning_rate": 9.552602372383047e-07,
      "loss": 0.7486636,
      "num_input_tokens_seen": 245885415,
      "step": 11388,
      "time_per_iteration": 2.6355738639831543
    },
    {
      "auxiliary_loss_clip": 0.01090608,
      "auxiliary_loss_mlp": 0.01023893,
      "balance_loss_clip": 1.03869319,
      "balance_loss_mlp": 1.01250851,
      "epoch": 0.6847437246355028,
      "flos": 43142468607360.0,
      "grad_norm": 1.775987397619916,
      "language_loss": 0.62148428,
      "learning_rate": 9.549281556699469e-07,
      "loss": 0.64262927,
      "num_input_tokens_seen": 245906285,
      "step": 11389,
      "time_per_iteration": 4.121963977813721
    },
    {
      "auxiliary_loss_clip": 0.01019581,
      "auxiliary_loss_mlp": 0.01004076,
      "balance_loss_clip": 1.01684594,
      "balance_loss_mlp": 1.00279999,
      "epoch": 0.6848038478881707,
      "flos": 71663729552640.0,
      "grad_norm": 0.7275726256563138,
      "language_loss": 0.55985212,
      "learning_rate": 9.54596113730818e-07,
      "loss": 0.58008873,
      "num_input_tokens_seen": 245967620,
      "step": 11390,
      "time_per_iteration": 3.2232506275177
    },
    {
      "auxiliary_loss_clip": 0.01072564,
      "auxiliary_loss_mlp": 0.00784662,
      "balance_loss_clip": 1.03897607,
      "balance_loss_mlp": 1.01127028,
      "epoch": 0.6848639711408387,
      "flos": 19937820257280.0,
      "grad_norm": 2.0468234556533,
      "language_loss": 0.8801896,
      "learning_rate": 9.542641114335109e-07,
      "loss": 0.89876187,
      "num_input_tokens_seen": 245985075,
      "step": 11391,
      "time_per_iteration": 2.5837225914001465
    },
    {
      "auxiliary_loss_clip": 0.01065401,
      "auxiliary_loss_mlp": 0.01032579,
      "balance_loss_clip": 1.03754306,
      "balance_loss_mlp": 1.02037835,
      "epoch": 0.6849240943935067,
      "flos": 26867501844480.0,
      "grad_norm": 1.6152753449798716,
      "language_loss": 0.79395598,
      "learning_rate": 9.539321487906117e-07,
      "loss": 0.8149358,
      "num_input_tokens_seen": 246003560,
      "step": 11392,
      "time_per_iteration": 2.6332597732543945
    },
    {
      "auxiliary_loss_clip": 0.01083457,
      "auxiliary_loss_mlp": 0.01029598,
      "balance_loss_clip": 1.0365659,
      "balance_loss_mlp": 1.01779032,
      "epoch": 0.6849842176461747,
      "flos": 13735365425280.0,
      "grad_norm": 2.35059530763193,
      "language_loss": 0.70904553,
      "learning_rate": 9.536002258147104e-07,
      "loss": 0.73017609,
      "num_input_tokens_seen": 246019600,
      "step": 11393,
      "time_per_iteration": 2.502363443374634
    },
    {
      "auxiliary_loss_clip": 0.01068315,
      "auxiliary_loss_mlp": 0.01029452,
      "balance_loss_clip": 1.03562844,
      "balance_loss_mlp": 1.01657748,
      "epoch": 0.6850443408988426,
      "flos": 24973070641920.0,
      "grad_norm": 1.5830588230547482,
      "language_loss": 0.64385206,
      "learning_rate": 9.532683425183936e-07,
      "loss": 0.66482973,
      "num_input_tokens_seen": 246038920,
      "step": 11394,
      "time_per_iteration": 2.636575698852539
    },
    {
      "auxiliary_loss_clip": 0.01081125,
      "auxiliary_loss_mlp": 0.00786992,
      "balance_loss_clip": 1.03684044,
      "balance_loss_mlp": 1.01286519,
      "epoch": 0.6851044641515106,
      "flos": 27744225004800.0,
      "grad_norm": 2.940442165797074,
      "language_loss": 0.80673003,
      "learning_rate": 9.529364989142468e-07,
      "loss": 0.82541126,
      "num_input_tokens_seen": 246060490,
      "step": 11395,
      "time_per_iteration": 3.9998552799224854
    },
    {
      "auxiliary_loss_clip": 0.01076742,
      "auxiliary_loss_mlp": 0.01030285,
      "balance_loss_clip": 1.03838277,
      "balance_loss_mlp": 1.01743436,
      "epoch": 0.6851645874041785,
      "flos": 24351061800960.0,
      "grad_norm": 1.8595842335394746,
      "language_loss": 0.73059464,
      "learning_rate": 9.526046950148527e-07,
      "loss": 0.75166488,
      "num_input_tokens_seen": 246081465,
      "step": 11396,
      "time_per_iteration": 4.009026050567627
    },
    {
      "auxiliary_loss_clip": 0.01078479,
      "auxiliary_loss_mlp": 0.01029572,
      "balance_loss_clip": 1.03799772,
      "balance_loss_mlp": 1.01641679,
      "epoch": 0.6852247106568465,
      "flos": 15077849264640.0,
      "grad_norm": 2.269817428417604,
      "language_loss": 0.78728217,
      "learning_rate": 9.522729308327931e-07,
      "loss": 0.80836272,
      "num_input_tokens_seen": 246096110,
      "step": 11397,
      "time_per_iteration": 2.5394840240478516
    },
    {
      "auxiliary_loss_clip": 0.01031734,
      "auxiliary_loss_mlp": 0.01031483,
      "balance_loss_clip": 1.0340718,
      "balance_loss_mlp": 1.01889491,
      "epoch": 0.6852848339095146,
      "flos": 18770005278720.0,
      "grad_norm": 1.7676110547601929,
      "language_loss": 0.71246088,
      "learning_rate": 9.519412063806493e-07,
      "loss": 0.73309302,
      "num_input_tokens_seen": 246114785,
      "step": 11398,
      "time_per_iteration": 2.6638059616088867
    },
    {
      "auxiliary_loss_clip": 0.01060211,
      "auxiliary_loss_mlp": 0.01028504,
      "balance_loss_clip": 1.03939176,
      "balance_loss_mlp": 1.01724505,
      "epoch": 0.6853449571621825,
      "flos": 27854363082240.0,
      "grad_norm": 1.617733669740887,
      "language_loss": 0.7059083,
      "learning_rate": 9.516095216709996e-07,
      "loss": 0.72679543,
      "num_input_tokens_seen": 246136375,
      "step": 11399,
      "time_per_iteration": 2.667738199234009
    },
    {
      "auxiliary_loss_clip": 0.01093752,
      "auxiliary_loss_mlp": 0.01032957,
      "balance_loss_clip": 1.03763115,
      "balance_loss_mlp": 1.02054763,
      "epoch": 0.6854050804148505,
      "flos": 18150510389760.0,
      "grad_norm": 1.5863679957489545,
      "language_loss": 0.70315838,
      "learning_rate": 9.512778767164217e-07,
      "loss": 0.72442555,
      "num_input_tokens_seen": 246155090,
      "step": 11400,
      "time_per_iteration": 4.011098384857178
    },
    {
      "auxiliary_loss_clip": 0.01071444,
      "auxiliary_loss_mlp": 0.01033995,
      "balance_loss_clip": 1.04054964,
      "balance_loss_mlp": 1.01917708,
      "epoch": 0.6854652036675184,
      "flos": 16326212492160.0,
      "grad_norm": 1.9814019869484805,
      "language_loss": 0.7803458,
      "learning_rate": 9.509462715294927e-07,
      "loss": 0.80140018,
      "num_input_tokens_seen": 246172645,
      "step": 11401,
      "time_per_iteration": 2.6040945053100586
    },
    {
      "auxiliary_loss_clip": 0.01106112,
      "auxiliary_loss_mlp": 0.01034656,
      "balance_loss_clip": 1.03719592,
      "balance_loss_mlp": 1.02287817,
      "epoch": 0.6855253269201864,
      "flos": 14940814878720.0,
      "grad_norm": 1.7749016626328709,
      "language_loss": 0.7542643,
      "learning_rate": 9.50614706122786e-07,
      "loss": 0.77567196,
      "num_input_tokens_seen": 246189055,
      "step": 11402,
      "time_per_iteration": 2.458620309829712
    },
    {
      "auxiliary_loss_clip": 0.01093512,
      "auxiliary_loss_mlp": 0.01034954,
      "balance_loss_clip": 1.03650677,
      "balance_loss_mlp": 1.02157283,
      "epoch": 0.6855854501728543,
      "flos": 23037736826880.0,
      "grad_norm": 1.6999621881828242,
      "language_loss": 0.72720325,
      "learning_rate": 9.502831805088742e-07,
      "loss": 0.74848795,
      "num_input_tokens_seen": 246207990,
      "step": 11403,
      "time_per_iteration": 2.520113229751587
    },
    {
      "auxiliary_loss_clip": 0.01105597,
      "auxiliary_loss_mlp": 0.01032573,
      "balance_loss_clip": 1.03764164,
      "balance_loss_mlp": 1.02079546,
      "epoch": 0.6856455734255223,
      "flos": 13253623194240.0,
      "grad_norm": 2.536807438699413,
      "language_loss": 0.81293035,
      "learning_rate": 9.499516947003294e-07,
      "loss": 0.83431196,
      "num_input_tokens_seen": 246221595,
      "step": 11404,
      "time_per_iteration": 2.4568464756011963
    },
    {
      "auxiliary_loss_clip": 0.01081239,
      "auxiliary_loss_mlp": 0.01033805,
      "balance_loss_clip": 1.03802299,
      "balance_loss_mlp": 1.02190208,
      "epoch": 0.6857056966781903,
      "flos": 23333461499520.0,
      "grad_norm": 1.4723023974445661,
      "language_loss": 0.77854437,
      "learning_rate": 9.496202487097222e-07,
      "loss": 0.79969478,
      "num_input_tokens_seen": 246242970,
      "step": 11405,
      "time_per_iteration": 2.578655481338501
    },
    {
      "auxiliary_loss_clip": 0.01033966,
      "auxiliary_loss_mlp": 0.01000903,
      "balance_loss_clip": 1.02004254,
      "balance_loss_mlp": 0.99963301,
      "epoch": 0.6857658199308583,
      "flos": 61852647784320.0,
      "grad_norm": 0.7898547458298292,
      "language_loss": 0.61001587,
      "learning_rate": 9.492888425496199e-07,
      "loss": 0.6303646,
      "num_input_tokens_seen": 246300405,
      "step": 11406,
      "time_per_iteration": 3.1653146743774414
    },
    {
      "auxiliary_loss_clip": 0.01075611,
      "auxiliary_loss_mlp": 0.01033986,
      "balance_loss_clip": 1.03691149,
      "balance_loss_mlp": 1.02046216,
      "epoch": 0.6858259431835262,
      "flos": 16654543735680.0,
      "grad_norm": 1.8719105601890103,
      "language_loss": 0.77124083,
      "learning_rate": 9.489574762325907e-07,
      "loss": 0.79233682,
      "num_input_tokens_seen": 246318780,
      "step": 11407,
      "time_per_iteration": 2.56085467338562
    },
    {
      "auxiliary_loss_clip": 0.01086802,
      "auxiliary_loss_mlp": 0.01033378,
      "balance_loss_clip": 1.03794515,
      "balance_loss_mlp": 1.02019405,
      "epoch": 0.6858860664361942,
      "flos": 21872974504320.0,
      "grad_norm": 2.225081083741925,
      "language_loss": 0.71079099,
      "learning_rate": 9.486261497711991e-07,
      "loss": 0.73199284,
      "num_input_tokens_seen": 246339405,
      "step": 11408,
      "time_per_iteration": 2.640377998352051
    },
    {
      "auxiliary_loss_clip": 0.01100329,
      "auxiliary_loss_mlp": 0.01028856,
      "balance_loss_clip": 1.03923047,
      "balance_loss_mlp": 1.01620746,
      "epoch": 0.6859461896888621,
      "flos": 15267637751040.0,
      "grad_norm": 1.7751810859494537,
      "language_loss": 0.70194495,
      "learning_rate": 9.482948631780087e-07,
      "loss": 0.72323674,
      "num_input_tokens_seen": 246357055,
      "step": 11409,
      "time_per_iteration": 2.55047345161438
    },
    {
      "auxiliary_loss_clip": 0.0106277,
      "auxiliary_loss_mlp": 0.01029288,
      "balance_loss_clip": 1.03840387,
      "balance_loss_mlp": 1.01808286,
      "epoch": 0.6860063129415301,
      "flos": 18620293392000.0,
      "grad_norm": 1.7040722363644618,
      "language_loss": 0.7829141,
      "learning_rate": 9.479636164655825e-07,
      "loss": 0.80383474,
      "num_input_tokens_seen": 246374050,
      "step": 11410,
      "time_per_iteration": 2.6055891513824463
    },
    {
      "auxiliary_loss_clip": 0.01101767,
      "auxiliary_loss_mlp": 0.01037662,
      "balance_loss_clip": 1.03859377,
      "balance_loss_mlp": 1.02454865,
      "epoch": 0.6860664361941982,
      "flos": 23951376190080.0,
      "grad_norm": 1.9950751465373529,
      "language_loss": 0.71431071,
      "learning_rate": 9.476324096464821e-07,
      "loss": 0.73570502,
      "num_input_tokens_seen": 246392910,
      "step": 11411,
      "time_per_iteration": 2.558617115020752
    },
    {
      "auxiliary_loss_clip": 0.01056707,
      "auxiliary_loss_mlp": 0.01034377,
      "balance_loss_clip": 1.03479791,
      "balance_loss_mlp": 1.01996493,
      "epoch": 0.6861265594468661,
      "flos": 20407782827520.0,
      "grad_norm": 2.064182975961849,
      "language_loss": 0.70096773,
      "learning_rate": 9.473012427332654e-07,
      "loss": 0.72187859,
      "num_input_tokens_seen": 246411540,
      "step": 11412,
      "time_per_iteration": 2.682128667831421
    },
    {
      "auxiliary_loss_clip": 0.01110037,
      "auxiliary_loss_mlp": 0.01033626,
      "balance_loss_clip": 1.03911066,
      "balance_loss_mlp": 1.02067947,
      "epoch": 0.6861866826995341,
      "flos": 11428571111040.0,
      "grad_norm": 5.795879312976152,
      "language_loss": 0.71915996,
      "learning_rate": 9.469701157384919e-07,
      "loss": 0.74059659,
      "num_input_tokens_seen": 246423295,
      "step": 11413,
      "time_per_iteration": 2.458012819290161
    },
    {
      "auxiliary_loss_clip": 0.01097792,
      "auxiliary_loss_mlp": 0.01030015,
      "balance_loss_clip": 1.03783751,
      "balance_loss_mlp": 1.01827264,
      "epoch": 0.686246805952202,
      "flos": 15997593939840.0,
      "grad_norm": 1.7930772299805322,
      "language_loss": 0.73785454,
      "learning_rate": 9.466390286747164e-07,
      "loss": 0.75913262,
      "num_input_tokens_seen": 246441045,
      "step": 11414,
      "time_per_iteration": 2.4923696517944336
    },
    {
      "auxiliary_loss_clip": 0.01086172,
      "auxiliary_loss_mlp": 0.01033301,
      "balance_loss_clip": 1.04000449,
      "balance_loss_mlp": 1.02067029,
      "epoch": 0.68630692920487,
      "flos": 19826712512640.0,
      "grad_norm": 2.7877135287137427,
      "language_loss": 0.86407208,
      "learning_rate": 9.46307981554495e-07,
      "loss": 0.88526678,
      "num_input_tokens_seen": 246456905,
      "step": 11415,
      "time_per_iteration": 2.538278341293335
    },
    {
      "auxiliary_loss_clip": 0.01100452,
      "auxiliary_loss_mlp": 0.01033772,
      "balance_loss_clip": 1.03888345,
      "balance_loss_mlp": 1.02102852,
      "epoch": 0.6863670524575379,
      "flos": 26286216048000.0,
      "grad_norm": 1.9061608587414378,
      "language_loss": 0.67228067,
      "learning_rate": 9.459769743903801e-07,
      "loss": 0.69362295,
      "num_input_tokens_seen": 246477545,
      "step": 11416,
      "time_per_iteration": 2.553169012069702
    },
    {
      "auxiliary_loss_clip": 0.01083689,
      "auxiliary_loss_mlp": 0.0103632,
      "balance_loss_clip": 1.03695393,
      "balance_loss_mlp": 1.02345753,
      "epoch": 0.686427175710206,
      "flos": 19173138595200.0,
      "grad_norm": 1.3836163866428348,
      "language_loss": 0.75890708,
      "learning_rate": 9.456460071949237e-07,
      "loss": 0.78010714,
      "num_input_tokens_seen": 246496705,
      "step": 11417,
      "time_per_iteration": 2.5242466926574707
    },
    {
      "auxiliary_loss_clip": 0.01083122,
      "auxiliary_loss_mlp": 0.01035265,
      "balance_loss_clip": 1.03656077,
      "balance_loss_mlp": 1.02145481,
      "epoch": 0.6864872989628739,
      "flos": 18916628595840.0,
      "grad_norm": 1.7836321650040894,
      "language_loss": 0.76978844,
      "learning_rate": 9.45315079980678e-07,
      "loss": 0.79097229,
      "num_input_tokens_seen": 246514860,
      "step": 11418,
      "time_per_iteration": 2.524371862411499
    },
    {
      "auxiliary_loss_clip": 0.01062194,
      "auxiliary_loss_mlp": 0.01030069,
      "balance_loss_clip": 1.03855324,
      "balance_loss_mlp": 1.01848185,
      "epoch": 0.6865474222155419,
      "flos": 25956196865280.0,
      "grad_norm": 1.7082498959988437,
      "language_loss": 0.76234138,
      "learning_rate": 9.449841927601887e-07,
      "loss": 0.78326404,
      "num_input_tokens_seen": 246536145,
      "step": 11419,
      "time_per_iteration": 2.6247496604919434
    },
    {
      "auxiliary_loss_clip": 0.01107955,
      "auxiliary_loss_mlp": 0.010373,
      "balance_loss_clip": 1.03886867,
      "balance_loss_mlp": 1.02554655,
      "epoch": 0.6866075454682098,
      "flos": 18478087447680.0,
      "grad_norm": 1.6983114077629229,
      "language_loss": 0.71529216,
      "learning_rate": 9.446533455460044e-07,
      "loss": 0.73674476,
      "num_input_tokens_seen": 246553265,
      "step": 11420,
      "time_per_iteration": 2.4778966903686523
    },
    {
      "auxiliary_loss_clip": 0.01068991,
      "auxiliary_loss_mlp": 0.01031503,
      "balance_loss_clip": 1.03464687,
      "balance_loss_mlp": 1.01856935,
      "epoch": 0.6866676687208778,
      "flos": 34239998298240.0,
      "grad_norm": 2.193146275849193,
      "language_loss": 0.74469376,
      "learning_rate": 9.443225383506712e-07,
      "loss": 0.76569867,
      "num_input_tokens_seen": 246575130,
      "step": 11421,
      "time_per_iteration": 2.66667103767395
    },
    {
      "auxiliary_loss_clip": 0.01092991,
      "auxiliary_loss_mlp": 0.0103167,
      "balance_loss_clip": 1.03733706,
      "balance_loss_mlp": 1.01999342,
      "epoch": 0.6867277919735457,
      "flos": 21721754246400.0,
      "grad_norm": 1.777018691523352,
      "language_loss": 0.7735886,
      "learning_rate": 9.439917711867338e-07,
      "loss": 0.79483521,
      "num_input_tokens_seen": 246593095,
      "step": 11422,
      "time_per_iteration": 2.5080108642578125
    },
    {
      "auxiliary_loss_clip": 0.01100764,
      "auxiliary_loss_mlp": 0.01036988,
      "balance_loss_clip": 1.03911531,
      "balance_loss_mlp": 1.02383876,
      "epoch": 0.6867879152262137,
      "flos": 24097999507200.0,
      "grad_norm": 1.7994108828596531,
      "language_loss": 0.76967907,
      "learning_rate": 9.436610440667334e-07,
      "loss": 0.79105651,
      "num_input_tokens_seen": 246612165,
      "step": 11423,
      "time_per_iteration": 2.5217063426971436
    },
    {
      "auxiliary_loss_clip": 0.01078917,
      "auxiliary_loss_mlp": 0.01031786,
      "balance_loss_clip": 1.03899741,
      "balance_loss_mlp": 1.01953125,
      "epoch": 0.6868480384788818,
      "flos": 21615818060160.0,
      "grad_norm": 1.524977571288621,
      "language_loss": 0.73016912,
      "learning_rate": 9.433303570032129e-07,
      "loss": 0.75127614,
      "num_input_tokens_seen": 246632065,
      "step": 11424,
      "time_per_iteration": 2.6006581783294678
    },
    {
      "auxiliary_loss_clip": 0.01087787,
      "auxiliary_loss_mlp": 0.0102901,
      "balance_loss_clip": 1.03919303,
      "balance_loss_mlp": 1.01761305,
      "epoch": 0.6869081617315497,
      "flos": 26286144220800.0,
      "grad_norm": 1.9642174724006458,
      "language_loss": 0.64932722,
      "learning_rate": 9.429997100087112e-07,
      "loss": 0.67049521,
      "num_input_tokens_seen": 246651245,
      "step": 11425,
      "time_per_iteration": 2.569391965866089
    },
    {
      "auxiliary_loss_clip": 0.01071264,
      "auxiliary_loss_mlp": 0.01028106,
      "balance_loss_clip": 1.04019499,
      "balance_loss_mlp": 1.01622033,
      "epoch": 0.6869682849842177,
      "flos": 21105096531840.0,
      "grad_norm": 1.4142545289190422,
      "language_loss": 0.71945751,
      "learning_rate": 9.426691030957657e-07,
      "loss": 0.74045122,
      "num_input_tokens_seen": 246672225,
      "step": 11426,
      "time_per_iteration": 2.5821170806884766
    },
    {
      "auxiliary_loss_clip": 0.01057033,
      "auxiliary_loss_mlp": 0.01030521,
      "balance_loss_clip": 1.03673804,
      "balance_loss_mlp": 1.0175333,
      "epoch": 0.6870284082368856,
      "flos": 17092653920640.0,
      "grad_norm": 2.0454233430095763,
      "language_loss": 0.84965444,
      "learning_rate": 9.423385362769136e-07,
      "loss": 0.87052995,
      "num_input_tokens_seen": 246688385,
      "step": 11427,
      "time_per_iteration": 2.5578720569610596
    },
    {
      "auxiliary_loss_clip": 0.01097558,
      "auxiliary_loss_mlp": 0.01033786,
      "balance_loss_clip": 1.04009426,
      "balance_loss_mlp": 1.02188277,
      "epoch": 0.6870885314895536,
      "flos": 27308090067840.0,
      "grad_norm": 1.445908983231136,
      "language_loss": 0.76231742,
      "learning_rate": 9.420080095646909e-07,
      "loss": 0.78363085,
      "num_input_tokens_seen": 246710730,
      "step": 11428,
      "time_per_iteration": 3.9582669734954834
    },
    {
      "auxiliary_loss_clip": 0.01075587,
      "auxiliary_loss_mlp": 0.01034424,
      "balance_loss_clip": 1.03871548,
      "balance_loss_mlp": 1.02173972,
      "epoch": 0.6871486547422215,
      "flos": 20814543417600.0,
      "grad_norm": 1.9062096245793856,
      "language_loss": 0.73327291,
      "learning_rate": 9.4167752297163e-07,
      "loss": 0.75437301,
      "num_input_tokens_seen": 246730350,
      "step": 11429,
      "time_per_iteration": 2.5858376026153564
    },
    {
      "auxiliary_loss_clip": 0.01087148,
      "auxiliary_loss_mlp": 0.01024392,
      "balance_loss_clip": 1.03928888,
      "balance_loss_mlp": 1.01250052,
      "epoch": 0.6872087779948896,
      "flos": 30154118330880.0,
      "grad_norm": 1.662589406254346,
      "language_loss": 0.83055973,
      "learning_rate": 9.413470765102643e-07,
      "loss": 0.85167515,
      "num_input_tokens_seen": 246751700,
      "step": 11430,
      "time_per_iteration": 2.598278522491455
    },
    {
      "auxiliary_loss_clip": 0.01098133,
      "auxiliary_loss_mlp": 0.01032419,
      "balance_loss_clip": 1.03750777,
      "balance_loss_mlp": 1.0205934,
      "epoch": 0.6872689012475575,
      "flos": 20704584908160.0,
      "grad_norm": 2.0104257300578623,
      "language_loss": 0.70134217,
      "learning_rate": 9.410166701931225e-07,
      "loss": 0.72264767,
      "num_input_tokens_seen": 246769860,
      "step": 11431,
      "time_per_iteration": 2.526407480239868
    },
    {
      "auxiliary_loss_clip": 0.0108811,
      "auxiliary_loss_mlp": 0.00787454,
      "balance_loss_clip": 1.03745556,
      "balance_loss_mlp": 1.01696467,
      "epoch": 0.6873290245002255,
      "flos": 25520852027520.0,
      "grad_norm": 1.824118707970035,
      "language_loss": 0.79818785,
      "learning_rate": 9.406863040327355e-07,
      "loss": 0.81694353,
      "num_input_tokens_seen": 246789905,
      "step": 11432,
      "time_per_iteration": 2.570183038711548
    },
    {
      "auxiliary_loss_clip": 0.01083503,
      "auxiliary_loss_mlp": 0.01026461,
      "balance_loss_clip": 1.0376724,
      "balance_loss_mlp": 1.01515985,
      "epoch": 0.6873891477528934,
      "flos": 25191479289600.0,
      "grad_norm": 1.579242325458712,
      "language_loss": 0.67906314,
      "learning_rate": 9.403559780416295e-07,
      "loss": 0.70016277,
      "num_input_tokens_seen": 246808815,
      "step": 11433,
      "time_per_iteration": 2.5659139156341553
    },
    {
      "auxiliary_loss_clip": 0.01100434,
      "auxiliary_loss_mlp": 0.01036137,
      "balance_loss_clip": 1.04089642,
      "balance_loss_mlp": 1.0239954,
      "epoch": 0.6874492710055614,
      "flos": 35152380685440.0,
      "grad_norm": 1.95080507916852,
      "language_loss": 0.73071945,
      "learning_rate": 9.400256922323309e-07,
      "loss": 0.75208509,
      "num_input_tokens_seen": 246829775,
      "step": 11434,
      "time_per_iteration": 5.479167699813843
    },
    {
      "auxiliary_loss_clip": 0.01072833,
      "auxiliary_loss_mlp": 0.0102573,
      "balance_loss_clip": 1.03958762,
      "balance_loss_mlp": 1.01397049,
      "epoch": 0.6875093942582293,
      "flos": 17822215059840.0,
      "grad_norm": 1.5768000847545227,
      "language_loss": 0.80319405,
      "learning_rate": 9.396954466173657e-07,
      "loss": 0.82417971,
      "num_input_tokens_seen": 246848045,
      "step": 11435,
      "time_per_iteration": 2.544827699661255
    },
    {
      "auxiliary_loss_clip": 0.01108312,
      "auxiliary_loss_mlp": 0.01030579,
      "balance_loss_clip": 1.03763103,
      "balance_loss_mlp": 1.01811004,
      "epoch": 0.6875695175108973,
      "flos": 20704548994560.0,
      "grad_norm": 2.0596846662916657,
      "language_loss": 0.81089199,
      "learning_rate": 9.393652412092538e-07,
      "loss": 0.83228093,
      "num_input_tokens_seen": 246866095,
      "step": 11436,
      "time_per_iteration": 2.4881951808929443
    },
    {
      "auxiliary_loss_clip": 0.01063329,
      "auxiliary_loss_mlp": 0.01034274,
      "balance_loss_clip": 1.03495896,
      "balance_loss_mlp": 1.02287722,
      "epoch": 0.6876296407635654,
      "flos": 25374013228800.0,
      "grad_norm": 1.7494839947163467,
      "language_loss": 0.82151163,
      "learning_rate": 9.390350760205183e-07,
      "loss": 0.84248769,
      "num_input_tokens_seen": 246883975,
      "step": 11437,
      "time_per_iteration": 2.586728096008301
    },
    {
      "auxiliary_loss_clip": 0.01090755,
      "auxiliary_loss_mlp": 0.0103588,
      "balance_loss_clip": 1.03755188,
      "balance_loss_mlp": 1.02252257,
      "epoch": 0.6876897640162333,
      "flos": 23222317841280.0,
      "grad_norm": 2.6163446464936597,
      "language_loss": 0.78465873,
      "learning_rate": 9.387049510636793e-07,
      "loss": 0.80592507,
      "num_input_tokens_seen": 246901560,
      "step": 11438,
      "time_per_iteration": 3.9343626499176025
    },
    {
      "auxiliary_loss_clip": 0.01103305,
      "auxiliary_loss_mlp": 0.01031001,
      "balance_loss_clip": 1.03793025,
      "balance_loss_mlp": 1.01960468,
      "epoch": 0.6877498872689013,
      "flos": 27124335066240.0,
      "grad_norm": 1.6861873561798344,
      "language_loss": 0.7256217,
      "learning_rate": 9.383748663512554e-07,
      "loss": 0.74696481,
      "num_input_tokens_seen": 246922655,
      "step": 11439,
      "time_per_iteration": 2.5388436317443848
    },
    {
      "auxiliary_loss_clip": 0.01095734,
      "auxiliary_loss_mlp": 0.01024314,
      "balance_loss_clip": 1.03848076,
      "balance_loss_mlp": 1.01272082,
      "epoch": 0.6878100105215692,
      "flos": 11581658876160.0,
      "grad_norm": 2.8440063880928834,
      "language_loss": 0.75213927,
      "learning_rate": 9.380448218957623e-07,
      "loss": 0.77333975,
      "num_input_tokens_seen": 246940100,
      "step": 11440,
      "time_per_iteration": 2.4714014530181885
    },
    {
      "auxiliary_loss_clip": 0.01062046,
      "auxiliary_loss_mlp": 0.0103617,
      "balance_loss_clip": 1.03529191,
      "balance_loss_mlp": 1.02452874,
      "epoch": 0.6878701337742372,
      "flos": 20303175444480.0,
      "grad_norm": 1.6414942417203626,
      "language_loss": 0.71753949,
      "learning_rate": 9.377148177097167e-07,
      "loss": 0.73852164,
      "num_input_tokens_seen": 246958545,
      "step": 11441,
      "time_per_iteration": 2.571171760559082
    },
    {
      "auxiliary_loss_clip": 0.01072869,
      "auxiliary_loss_mlp": 0.01038817,
      "balance_loss_clip": 1.03888178,
      "balance_loss_mlp": 1.02421975,
      "epoch": 0.6879302570269051,
      "flos": 13840080549120.0,
      "grad_norm": 1.6860979342043239,
      "language_loss": 0.67043531,
      "learning_rate": 9.373848538056317e-07,
      "loss": 0.69155216,
      "num_input_tokens_seen": 246974805,
      "step": 11442,
      "time_per_iteration": 2.564851999282837
    },
    {
      "auxiliary_loss_clip": 0.01094096,
      "auxiliary_loss_mlp": 0.01030363,
      "balance_loss_clip": 1.03994882,
      "balance_loss_mlp": 1.01913345,
      "epoch": 0.6879903802795732,
      "flos": 21324654414720.0,
      "grad_norm": 1.8966687516498697,
      "language_loss": 0.69875681,
      "learning_rate": 9.370549301960189e-07,
      "loss": 0.72000146,
      "num_input_tokens_seen": 246992505,
      "step": 11443,
      "time_per_iteration": 2.528290033340454
    },
    {
      "auxiliary_loss_clip": 0.01091565,
      "auxiliary_loss_mlp": 0.01031857,
      "balance_loss_clip": 1.04010141,
      "balance_loss_mlp": 1.01885724,
      "epoch": 0.6880505035322411,
      "flos": 25152049134720.0,
      "grad_norm": 1.4390044685152852,
      "language_loss": 0.76349401,
      "learning_rate": 9.367250468933893e-07,
      "loss": 0.78472823,
      "num_input_tokens_seen": 247013370,
      "step": 11444,
      "time_per_iteration": 2.5917370319366455
    },
    {
      "auxiliary_loss_clip": 0.01105579,
      "auxiliary_loss_mlp": 0.01028441,
      "balance_loss_clip": 1.03835213,
      "balance_loss_mlp": 1.01742601,
      "epoch": 0.6881106267849091,
      "flos": 23215530170880.0,
      "grad_norm": 2.353068624582353,
      "language_loss": 0.77054465,
      "learning_rate": 9.363952039102536e-07,
      "loss": 0.79188484,
      "num_input_tokens_seen": 247029855,
      "step": 11445,
      "time_per_iteration": 2.477593421936035
    },
    {
      "auxiliary_loss_clip": 0.01033652,
      "auxiliary_loss_mlp": 0.01009225,
      "balance_loss_clip": 1.01920426,
      "balance_loss_mlp": 1.00802124,
      "epoch": 0.688170750037577,
      "flos": 48484397312640.0,
      "grad_norm": 0.8159958908223361,
      "language_loss": 0.58371258,
      "learning_rate": 9.360654012591183e-07,
      "loss": 0.60414135,
      "num_input_tokens_seen": 247085030,
      "step": 11446,
      "time_per_iteration": 3.138810634613037
    },
    {
      "auxiliary_loss_clip": 0.01096145,
      "auxiliary_loss_mlp": 0.01028908,
      "balance_loss_clip": 1.03529334,
      "balance_loss_mlp": 1.01658213,
      "epoch": 0.688230873290245,
      "flos": 22783633038720.0,
      "grad_norm": 1.4991421741977153,
      "language_loss": 0.75608331,
      "learning_rate": 9.357356389524886e-07,
      "loss": 0.77733386,
      "num_input_tokens_seen": 247104840,
      "step": 11447,
      "time_per_iteration": 2.563877582550049
    },
    {
      "auxiliary_loss_clip": 0.01088315,
      "auxiliary_loss_mlp": 0.01033648,
      "balance_loss_clip": 1.03789973,
      "balance_loss_mlp": 1.02148247,
      "epoch": 0.6882909965429129,
      "flos": 22455660931200.0,
      "grad_norm": 2.11551184388984,
      "language_loss": 0.73368597,
      "learning_rate": 9.354059170028705e-07,
      "loss": 0.75490558,
      "num_input_tokens_seen": 247121905,
      "step": 11448,
      "time_per_iteration": 2.5566985607147217
    },
    {
      "auxiliary_loss_clip": 0.01096727,
      "auxiliary_loss_mlp": 0.01038275,
      "balance_loss_clip": 1.03699541,
      "balance_loss_mlp": 1.02445889,
      "epoch": 0.688351119795581,
      "flos": 26214143408640.0,
      "grad_norm": 2.06113313338257,
      "language_loss": 0.74604321,
      "learning_rate": 9.350762354227673e-07,
      "loss": 0.76739323,
      "num_input_tokens_seen": 247142375,
      "step": 11449,
      "time_per_iteration": 2.5545544624328613
    },
    {
      "auxiliary_loss_clip": 0.01105536,
      "auxiliary_loss_mlp": 0.0103598,
      "balance_loss_clip": 1.03778148,
      "balance_loss_mlp": 1.02476811,
      "epoch": 0.6884112430482489,
      "flos": 22565260304640.0,
      "grad_norm": 1.7323562048981278,
      "language_loss": 0.7006005,
      "learning_rate": 9.34746594224679e-07,
      "loss": 0.72201574,
      "num_input_tokens_seen": 247161095,
      "step": 11450,
      "time_per_iteration": 2.4812815189361572
    },
    {
      "auxiliary_loss_clip": 0.01079645,
      "auxiliary_loss_mlp": 0.01033478,
      "balance_loss_clip": 1.04131794,
      "balance_loss_mlp": 1.02066302,
      "epoch": 0.6884713663009169,
      "flos": 17341047446400.0,
      "grad_norm": 1.9053156360853918,
      "language_loss": 0.76227933,
      "learning_rate": 9.344169934211068e-07,
      "loss": 0.78341061,
      "num_input_tokens_seen": 247178565,
      "step": 11451,
      "time_per_iteration": 2.564218759536743
    },
    {
      "auxiliary_loss_clip": 0.01095961,
      "auxiliary_loss_mlp": 0.01027679,
      "balance_loss_clip": 1.04069829,
      "balance_loss_mlp": 1.01589513,
      "epoch": 0.6885314895535849,
      "flos": 26470832976000.0,
      "grad_norm": 1.347276968038282,
      "language_loss": 0.69318259,
      "learning_rate": 9.340874330245505e-07,
      "loss": 0.71441901,
      "num_input_tokens_seen": 247202345,
      "step": 11452,
      "time_per_iteration": 2.5687992572784424
    },
    {
      "auxiliary_loss_clip": 0.01108843,
      "auxiliary_loss_mlp": 0.01033018,
      "balance_loss_clip": 1.03970218,
      "balance_loss_mlp": 1.01935613,
      "epoch": 0.6885916128062528,
      "flos": 20521548178560.0,
      "grad_norm": 1.648134840434037,
      "language_loss": 0.71994674,
      "learning_rate": 9.337579130475042e-07,
      "loss": 0.74136537,
      "num_input_tokens_seen": 247219240,
      "step": 11453,
      "time_per_iteration": 2.4747517108917236
    },
    {
      "auxiliary_loss_clip": 0.01035385,
      "auxiliary_loss_mlp": 0.00768007,
      "balance_loss_clip": 1.02127814,
      "balance_loss_mlp": 1.00858808,
      "epoch": 0.6886517360589208,
      "flos": 70715795679360.0,
      "grad_norm": 0.784916887746031,
      "language_loss": 0.50684935,
      "learning_rate": 9.334284335024644e-07,
      "loss": 0.52488333,
      "num_input_tokens_seen": 247272010,
      "step": 11454,
      "time_per_iteration": 2.9891676902770996
    },
    {
      "auxiliary_loss_clip": 0.01095286,
      "auxiliary_loss_mlp": 0.01031577,
      "balance_loss_clip": 1.03927326,
      "balance_loss_mlp": 1.01999569,
      "epoch": 0.6887118593115887,
      "flos": 17893533513600.0,
      "grad_norm": 1.83434119553909,
      "language_loss": 0.75101876,
      "learning_rate": 9.330989944019263e-07,
      "loss": 0.77228737,
      "num_input_tokens_seen": 247290630,
      "step": 11455,
      "time_per_iteration": 2.5094077587127686
    },
    {
      "auxiliary_loss_clip": 0.01086757,
      "auxiliary_loss_mlp": 0.0103474,
      "balance_loss_clip": 1.03523481,
      "balance_loss_mlp": 1.02131057,
      "epoch": 0.6887719825642568,
      "flos": 17453017117440.0,
      "grad_norm": 2.985845080615853,
      "language_loss": 0.72799158,
      "learning_rate": 9.327695957583803e-07,
      "loss": 0.74920654,
      "num_input_tokens_seen": 247304800,
      "step": 11456,
      "time_per_iteration": 2.503765821456909
    },
    {
      "auxiliary_loss_clip": 0.01084597,
      "auxiliary_loss_mlp": 0.01036065,
      "balance_loss_clip": 1.03876019,
      "balance_loss_mlp": 1.02443588,
      "epoch": 0.6888321058169247,
      "flos": 23070199743360.0,
      "grad_norm": 1.5929865752865773,
      "language_loss": 0.80839884,
      "learning_rate": 9.32440237584319e-07,
      "loss": 0.8296054,
      "num_input_tokens_seen": 247323450,
      "step": 11457,
      "time_per_iteration": 2.5366861820220947
    },
    {
      "auxiliary_loss_clip": 0.01101335,
      "auxiliary_loss_mlp": 0.00786439,
      "balance_loss_clip": 1.04080307,
      "balance_loss_mlp": 1.01491094,
      "epoch": 0.6888922290695927,
      "flos": 23368833417600.0,
      "grad_norm": 1.5730700375361752,
      "language_loss": 0.76262349,
      "learning_rate": 9.321109198922301e-07,
      "loss": 0.78150117,
      "num_input_tokens_seen": 247343845,
      "step": 11458,
      "time_per_iteration": 2.543712854385376
    },
    {
      "auxiliary_loss_clip": 0.01108561,
      "auxiliary_loss_mlp": 0.01032465,
      "balance_loss_clip": 1.03920674,
      "balance_loss_mlp": 1.02136064,
      "epoch": 0.6889523523222606,
      "flos": 17631636474240.0,
      "grad_norm": 2.8012430757822546,
      "language_loss": 0.68008316,
      "learning_rate": 9.31781642694603e-07,
      "loss": 0.70149338,
      "num_input_tokens_seen": 247356650,
      "step": 11459,
      "time_per_iteration": 2.4367411136627197
    },
    {
      "auxiliary_loss_clip": 0.0106729,
      "auxiliary_loss_mlp": 0.01031597,
      "balance_loss_clip": 1.03747118,
      "balance_loss_mlp": 1.0205524,
      "epoch": 0.6890124755749286,
      "flos": 25228144097280.0,
      "grad_norm": 1.5504796504736178,
      "language_loss": 0.68578011,
      "learning_rate": 9.314524060039221e-07,
      "loss": 0.70676893,
      "num_input_tokens_seen": 247377340,
      "step": 11460,
      "time_per_iteration": 2.617654800415039
    },
    {
      "auxiliary_loss_clip": 0.01078716,
      "auxiliary_loss_mlp": 0.01028734,
      "balance_loss_clip": 1.03686678,
      "balance_loss_mlp": 1.01571012,
      "epoch": 0.6890725988275965,
      "flos": 20230240878720.0,
      "grad_norm": 1.719543490378219,
      "language_loss": 0.76796889,
      "learning_rate": 9.311232098326731e-07,
      "loss": 0.78904343,
      "num_input_tokens_seen": 247395805,
      "step": 11461,
      "time_per_iteration": 2.581594705581665
    },
    {
      "auxiliary_loss_clip": 0.01085678,
      "auxiliary_loss_mlp": 0.01034437,
      "balance_loss_clip": 1.03688693,
      "balance_loss_mlp": 1.02212918,
      "epoch": 0.6891327220802645,
      "flos": 14535311264640.0,
      "grad_norm": 1.626075294528417,
      "language_loss": 0.6937964,
      "learning_rate": 9.307940541933401e-07,
      "loss": 0.71499753,
      "num_input_tokens_seen": 247413165,
      "step": 11462,
      "time_per_iteration": 2.5369086265563965
    },
    {
      "auxiliary_loss_clip": 0.01098153,
      "auxiliary_loss_mlp": 0.01027771,
      "balance_loss_clip": 1.03927135,
      "balance_loss_mlp": 1.01575506,
      "epoch": 0.6891928453329325,
      "flos": 21139139646720.0,
      "grad_norm": 1.5177456877688893,
      "language_loss": 0.8743223,
      "learning_rate": 9.304649390984034e-07,
      "loss": 0.89558148,
      "num_input_tokens_seen": 247433140,
      "step": 11463,
      "time_per_iteration": 2.531630039215088
    },
    {
      "auxiliary_loss_clip": 0.01056129,
      "auxiliary_loss_mlp": 0.0102571,
      "balance_loss_clip": 1.03694463,
      "balance_loss_mlp": 1.01515388,
      "epoch": 0.6892529685856005,
      "flos": 17858520731520.0,
      "grad_norm": 1.5516166483067624,
      "language_loss": 0.68284923,
      "learning_rate": 9.301358645603428e-07,
      "loss": 0.70366764,
      "num_input_tokens_seen": 247451265,
      "step": 11464,
      "time_per_iteration": 2.620295524597168
    },
    {
      "auxiliary_loss_clip": 0.01098104,
      "auxiliary_loss_mlp": 0.01033107,
      "balance_loss_clip": 1.03891134,
      "balance_loss_mlp": 1.02078032,
      "epoch": 0.6893130918382685,
      "flos": 29934811843200.0,
      "grad_norm": 2.0434623909115106,
      "language_loss": 0.65135026,
      "learning_rate": 9.298068305916373e-07,
      "loss": 0.67266238,
      "num_input_tokens_seen": 247471645,
      "step": 11465,
      "time_per_iteration": 2.573216438293457
    },
    {
      "auxiliary_loss_clip": 0.01099285,
      "auxiliary_loss_mlp": 0.01033309,
      "balance_loss_clip": 1.03826201,
      "balance_loss_mlp": 1.02178788,
      "epoch": 0.6893732150909364,
      "flos": 24388516707840.0,
      "grad_norm": 1.4113955974712737,
      "language_loss": 0.72682059,
      "learning_rate": 9.294778372047649e-07,
      "loss": 0.74814653,
      "num_input_tokens_seen": 247491170,
      "step": 11466,
      "time_per_iteration": 2.564704418182373
    },
    {
      "auxiliary_loss_clip": 0.01109742,
      "auxiliary_loss_mlp": 0.01030109,
      "balance_loss_clip": 1.03976667,
      "balance_loss_mlp": 1.0185163,
      "epoch": 0.6894333383436044,
      "flos": 16982874979200.0,
      "grad_norm": 1.619239188073845,
      "language_loss": 0.71228349,
      "learning_rate": 9.291488844121995e-07,
      "loss": 0.73368204,
      "num_input_tokens_seen": 247509005,
      "step": 11467,
      "time_per_iteration": 3.8964946269989014
    },
    {
      "auxiliary_loss_clip": 0.01088983,
      "auxiliary_loss_mlp": 0.01031787,
      "balance_loss_clip": 1.03825259,
      "balance_loss_mlp": 1.01835203,
      "epoch": 0.6894934615962723,
      "flos": 18985540838400.0,
      "grad_norm": 1.8735746515695006,
      "language_loss": 0.80700499,
      "learning_rate": 9.288199722264156e-07,
      "loss": 0.82821268,
      "num_input_tokens_seen": 247527050,
      "step": 11468,
      "time_per_iteration": 2.530848741531372
    },
    {
      "auxiliary_loss_clip": 0.011108,
      "auxiliary_loss_mlp": 0.01035027,
      "balance_loss_clip": 1.03990638,
      "balance_loss_mlp": 1.02287912,
      "epoch": 0.6895535848489404,
      "flos": 34531664734080.0,
      "grad_norm": 1.4993232749910217,
      "language_loss": 0.66019428,
      "learning_rate": 9.284911006598875e-07,
      "loss": 0.68165249,
      "num_input_tokens_seen": 247547765,
      "step": 11469,
      "time_per_iteration": 2.5818252563476562
    },
    {
      "auxiliary_loss_clip": 0.01032964,
      "auxiliary_loss_mlp": 0.01000552,
      "balance_loss_clip": 1.01902914,
      "balance_loss_mlp": 0.99922925,
      "epoch": 0.6896137081016083,
      "flos": 50075852273280.0,
      "grad_norm": 0.8020540980364241,
      "language_loss": 0.55178946,
      "learning_rate": 9.281622697250824e-07,
      "loss": 0.57212466,
      "num_input_tokens_seen": 247603515,
      "step": 11470,
      "time_per_iteration": 3.0206611156463623
    },
    {
      "auxiliary_loss_clip": 0.01093128,
      "auxiliary_loss_mlp": 0.01030418,
      "balance_loss_clip": 1.03805709,
      "balance_loss_mlp": 1.02043438,
      "epoch": 0.6896738313542763,
      "flos": 19938215306880.0,
      "grad_norm": 1.8052213816998894,
      "language_loss": 0.78427488,
      "learning_rate": 9.278334794344715e-07,
      "loss": 0.80551034,
      "num_input_tokens_seen": 247622110,
      "step": 11471,
      "time_per_iteration": 2.5166096687316895
    },
    {
      "auxiliary_loss_clip": 0.01084252,
      "auxiliary_loss_mlp": 0.01029672,
      "balance_loss_clip": 1.03623652,
      "balance_loss_mlp": 1.01750708,
      "epoch": 0.6897339546069442,
      "flos": 21725489260800.0,
      "grad_norm": 2.3083258065226606,
      "language_loss": 0.78290164,
      "learning_rate": 9.275047298005232e-07,
      "loss": 0.80404091,
      "num_input_tokens_seen": 247641905,
      "step": 11472,
      "time_per_iteration": 2.5432848930358887
    },
    {
      "auxiliary_loss_clip": 0.01085399,
      "auxiliary_loss_mlp": 0.01029241,
      "balance_loss_clip": 1.03692961,
      "balance_loss_mlp": 1.01779723,
      "epoch": 0.6897940778596122,
      "flos": 19826497031040.0,
      "grad_norm": 1.5446035072598734,
      "language_loss": 0.76133192,
      "learning_rate": 9.271760208357024e-07,
      "loss": 0.78247833,
      "num_input_tokens_seen": 247660945,
      "step": 11473,
      "time_per_iteration": 5.38184118270874
    },
    {
      "auxiliary_loss_clip": 0.01067694,
      "auxiliary_loss_mlp": 0.01043327,
      "balance_loss_clip": 1.03552461,
      "balance_loss_mlp": 1.02788973,
      "epoch": 0.6898542011122801,
      "flos": 17310056987520.0,
      "grad_norm": 2.169301010265059,
      "language_loss": 0.75581002,
      "learning_rate": 9.268473525524751e-07,
      "loss": 0.7769202,
      "num_input_tokens_seen": 247678395,
      "step": 11474,
      "time_per_iteration": 2.56606388092041
    },
    {
      "auxiliary_loss_clip": 0.01057735,
      "auxiliary_loss_mlp": 0.01028637,
      "balance_loss_clip": 1.03975761,
      "balance_loss_mlp": 1.0169661,
      "epoch": 0.6899143243649482,
      "flos": 24754051463040.0,
      "grad_norm": 1.5968351408096138,
      "language_loss": 0.74305242,
      "learning_rate": 9.26518724963303e-07,
      "loss": 0.76391613,
      "num_input_tokens_seen": 247698380,
      "step": 11475,
      "time_per_iteration": 2.6367831230163574
    },
    {
      "auxiliary_loss_clip": 0.01076173,
      "auxiliary_loss_mlp": 0.01029732,
      "balance_loss_clip": 1.03456497,
      "balance_loss_mlp": 1.01714337,
      "epoch": 0.6899744476176161,
      "flos": 17234536642560.0,
      "grad_norm": 2.2658212986996524,
      "language_loss": 0.88487005,
      "learning_rate": 9.261901380806491e-07,
      "loss": 0.90592903,
      "num_input_tokens_seen": 247716370,
      "step": 11476,
      "time_per_iteration": 2.5360498428344727
    },
    {
      "auxiliary_loss_clip": 0.01106591,
      "auxiliary_loss_mlp": 0.01031922,
      "balance_loss_clip": 1.03791142,
      "balance_loss_mlp": 1.02009022,
      "epoch": 0.6900345708702841,
      "flos": 25410678036480.0,
      "grad_norm": 1.350912154783926,
      "language_loss": 0.70155728,
      "learning_rate": 9.258615919169724e-07,
      "loss": 0.72294247,
      "num_input_tokens_seen": 247737335,
      "step": 11477,
      "time_per_iteration": 3.9897637367248535
    },
    {
      "auxiliary_loss_clip": 0.01101457,
      "auxiliary_loss_mlp": 0.01037143,
      "balance_loss_clip": 1.03915668,
      "balance_loss_mlp": 1.02430379,
      "epoch": 0.6900946941229521,
      "flos": 23434190213760.0,
      "grad_norm": 2.0402804301591315,
      "language_loss": 0.68282974,
      "learning_rate": 9.255330864847313e-07,
      "loss": 0.70421576,
      "num_input_tokens_seen": 247756680,
      "step": 11478,
      "time_per_iteration": 2.5227434635162354
    },
    {
      "auxiliary_loss_clip": 0.01099613,
      "auxiliary_loss_mlp": 0.01031825,
      "balance_loss_clip": 1.04012632,
      "balance_loss_mlp": 1.01992822,
      "epoch": 0.69015481737562,
      "flos": 17820096157440.0,
      "grad_norm": 1.9277083020336874,
      "language_loss": 0.76375258,
      "learning_rate": 9.252046217963843e-07,
      "loss": 0.78506696,
      "num_input_tokens_seen": 247774265,
      "step": 11479,
      "time_per_iteration": 2.4884185791015625
    },
    {
      "auxiliary_loss_clip": 0.01098143,
      "auxiliary_loss_mlp": 0.01028102,
      "balance_loss_clip": 1.03714812,
      "balance_loss_mlp": 1.01502442,
      "epoch": 0.690214940628288,
      "flos": 17456500736640.0,
      "grad_norm": 1.7504323783780453,
      "language_loss": 0.78582793,
      "learning_rate": 9.248761978643856e-07,
      "loss": 0.80709034,
      "num_input_tokens_seen": 247792395,
      "step": 11480,
      "time_per_iteration": 2.493065357208252
    },
    {
      "auxiliary_loss_clip": 0.01069474,
      "auxiliary_loss_mlp": 0.0102965,
      "balance_loss_clip": 1.03823185,
      "balance_loss_mlp": 1.01706147,
      "epoch": 0.6902750638809559,
      "flos": 29566691308800.0,
      "grad_norm": 1.7087624689519594,
      "language_loss": 0.75639343,
      "learning_rate": 9.245478147011885e-07,
      "loss": 0.7773847,
      "num_input_tokens_seen": 247811985,
      "step": 11481,
      "time_per_iteration": 2.64522123336792
    },
    {
      "auxiliary_loss_clip": 0.01071827,
      "auxiliary_loss_mlp": 0.01026442,
      "balance_loss_clip": 1.0383749,
      "balance_loss_mlp": 1.0142473,
      "epoch": 0.690335187133624,
      "flos": 25557121785600.0,
      "grad_norm": 1.6449690653431126,
      "language_loss": 0.69518256,
      "learning_rate": 9.24219472319246e-07,
      "loss": 0.71616518,
      "num_input_tokens_seen": 247831880,
      "step": 11482,
      "time_per_iteration": 2.599090576171875
    },
    {
      "auxiliary_loss_clip": 0.01107959,
      "auxiliary_loss_mlp": 0.01026531,
      "balance_loss_clip": 1.03889692,
      "balance_loss_mlp": 1.01492596,
      "epoch": 0.6903953103862919,
      "flos": 22488447070080.0,
      "grad_norm": 2.2324430710717484,
      "language_loss": 0.83057004,
      "learning_rate": 9.238911707310096e-07,
      "loss": 0.85191488,
      "num_input_tokens_seen": 247851170,
      "step": 11483,
      "time_per_iteration": 2.499938726425171
    },
    {
      "auxiliary_loss_clip": 0.01109281,
      "auxiliary_loss_mlp": 0.01026648,
      "balance_loss_clip": 1.03849447,
      "balance_loss_mlp": 1.01506126,
      "epoch": 0.6904554336389599,
      "flos": 26100521712000.0,
      "grad_norm": 1.9324833726294723,
      "language_loss": 0.65330982,
      "learning_rate": 9.235629099489273e-07,
      "loss": 0.67466915,
      "num_input_tokens_seen": 247868950,
      "step": 11484,
      "time_per_iteration": 2.510108709335327
    },
    {
      "auxiliary_loss_clip": 0.01074964,
      "auxiliary_loss_mlp": 0.01038129,
      "balance_loss_clip": 1.03615117,
      "balance_loss_mlp": 1.02524233,
      "epoch": 0.6905155568916278,
      "flos": 31171754545920.0,
      "grad_norm": 1.5246888137201462,
      "language_loss": 0.73463404,
      "learning_rate": 9.232346899854479e-07,
      "loss": 0.7557649,
      "num_input_tokens_seen": 247889805,
      "step": 11485,
      "time_per_iteration": 2.6824488639831543
    },
    {
      "auxiliary_loss_clip": 0.01096005,
      "auxiliary_loss_mlp": 0.00786125,
      "balance_loss_clip": 1.04018223,
      "balance_loss_mlp": 1.01482916,
      "epoch": 0.6905756801442958,
      "flos": 17639681120640.0,
      "grad_norm": 1.8083237321300252,
      "language_loss": 0.85021484,
      "learning_rate": 9.22906510853017e-07,
      "loss": 0.86903608,
      "num_input_tokens_seen": 247908585,
      "step": 11486,
      "time_per_iteration": 2.5284829139709473
    },
    {
      "auxiliary_loss_clip": 0.01047712,
      "auxiliary_loss_mlp": 0.01032887,
      "balance_loss_clip": 1.03719711,
      "balance_loss_mlp": 1.02093041,
      "epoch": 0.6906358033969637,
      "flos": 22343691260160.0,
      "grad_norm": 1.4919931089704883,
      "language_loss": 0.72719777,
      "learning_rate": 9.225783725640786e-07,
      "loss": 0.74800378,
      "num_input_tokens_seen": 247928480,
      "step": 11487,
      "time_per_iteration": 2.677279233932495
    },
    {
      "auxiliary_loss_clip": 0.01025064,
      "auxiliary_loss_mlp": 0.01007333,
      "balance_loss_clip": 1.02179337,
      "balance_loss_mlp": 1.00610542,
      "epoch": 0.6906959266496318,
      "flos": 69747789081600.0,
      "grad_norm": 0.8997483272382313,
      "language_loss": 0.66656947,
      "learning_rate": 9.222502751310759e-07,
      "loss": 0.68689346,
      "num_input_tokens_seen": 247988855,
      "step": 11488,
      "time_per_iteration": 3.141378879547119
    },
    {
      "auxiliary_loss_clip": 0.01085145,
      "auxiliary_loss_mlp": 0.01029974,
      "balance_loss_clip": 1.03876352,
      "balance_loss_mlp": 1.01665831,
      "epoch": 0.6907560499022997,
      "flos": 21434253788160.0,
      "grad_norm": 1.7555176107071115,
      "language_loss": 0.74565029,
      "learning_rate": 9.219222185664519e-07,
      "loss": 0.76680148,
      "num_input_tokens_seen": 248007685,
      "step": 11489,
      "time_per_iteration": 2.5758278369903564
    },
    {
      "auxiliary_loss_clip": 0.0109822,
      "auxiliary_loss_mlp": 0.01035625,
      "balance_loss_clip": 1.03867936,
      "balance_loss_mlp": 1.02264345,
      "epoch": 0.6908161731549677,
      "flos": 14392207480320.0,
      "grad_norm": 1.9774004828026746,
      "language_loss": 0.62383074,
      "learning_rate": 9.215942028826445e-07,
      "loss": 0.64516914,
      "num_input_tokens_seen": 248025145,
      "step": 11490,
      "time_per_iteration": 2.491811752319336
    },
    {
      "auxiliary_loss_clip": 0.01085356,
      "auxiliary_loss_mlp": 0.01027434,
      "balance_loss_clip": 1.03801596,
      "balance_loss_mlp": 1.01579905,
      "epoch": 0.6908762964076357,
      "flos": 20010970304640.0,
      "grad_norm": 1.6320383516224128,
      "language_loss": 0.72620809,
      "learning_rate": 9.212662280920937e-07,
      "loss": 0.74733597,
      "num_input_tokens_seen": 248043750,
      "step": 11491,
      "time_per_iteration": 2.537949800491333
    },
    {
      "auxiliary_loss_clip": 0.01082294,
      "auxiliary_loss_mlp": 0.00784719,
      "balance_loss_clip": 1.03577888,
      "balance_loss_mlp": 1.01184583,
      "epoch": 0.6909364196603036,
      "flos": 28769079853440.0,
      "grad_norm": 1.3337280727263037,
      "language_loss": 0.70180809,
      "learning_rate": 9.20938294207235e-07,
      "loss": 0.7204783,
      "num_input_tokens_seen": 248065765,
      "step": 11492,
      "time_per_iteration": 2.62387752532959
    },
    {
      "auxiliary_loss_clip": 0.01069976,
      "auxiliary_loss_mlp": 0.01030664,
      "balance_loss_clip": 1.03867435,
      "balance_loss_mlp": 1.01834333,
      "epoch": 0.6909965429129716,
      "flos": 22528128620160.0,
      "grad_norm": 1.781710254864284,
      "language_loss": 0.7487042,
      "learning_rate": 9.206104012405049e-07,
      "loss": 0.76971066,
      "num_input_tokens_seen": 248083810,
      "step": 11493,
      "time_per_iteration": 2.6055917739868164
    },
    {
      "auxiliary_loss_clip": 0.0110836,
      "auxiliary_loss_mlp": 0.01026955,
      "balance_loss_clip": 1.03904271,
      "balance_loss_mlp": 1.01478982,
      "epoch": 0.6910566661656395,
      "flos": 18405942981120.0,
      "grad_norm": 1.7515749065651012,
      "language_loss": 0.74460083,
      "learning_rate": 9.20282549204336e-07,
      "loss": 0.76595396,
      "num_input_tokens_seen": 248103185,
      "step": 11494,
      "time_per_iteration": 2.4650652408599854
    },
    {
      "auxiliary_loss_clip": 0.0108188,
      "auxiliary_loss_mlp": 0.01029112,
      "balance_loss_clip": 1.03782964,
      "balance_loss_mlp": 1.01735795,
      "epoch": 0.6911167894183076,
      "flos": 30773972355840.0,
      "grad_norm": 1.6060273319418366,
      "language_loss": 0.68092096,
      "learning_rate": 9.19954738111161e-07,
      "loss": 0.7020309,
      "num_input_tokens_seen": 248125665,
      "step": 11495,
      "time_per_iteration": 2.602530002593994
    },
    {
      "auxiliary_loss_clip": 0.01086733,
      "auxiliary_loss_mlp": 0.01028045,
      "balance_loss_clip": 1.03678966,
      "balance_loss_mlp": 1.01601052,
      "epoch": 0.6911769126709755,
      "flos": 13735724561280.0,
      "grad_norm": 1.8540005339578358,
      "language_loss": 0.74233568,
      "learning_rate": 9.196269679734119e-07,
      "loss": 0.76348341,
      "num_input_tokens_seen": 248142545,
      "step": 11496,
      "time_per_iteration": 2.5139498710632324
    },
    {
      "auxiliary_loss_clip": 0.01073489,
      "auxiliary_loss_mlp": 0.01029773,
      "balance_loss_clip": 1.03654218,
      "balance_loss_mlp": 1.01822186,
      "epoch": 0.6912370359236435,
      "flos": 17566854295680.0,
      "grad_norm": 1.6634619754334028,
      "language_loss": 0.80280024,
      "learning_rate": 9.19299238803515e-07,
      "loss": 0.82383287,
      "num_input_tokens_seen": 248160225,
      "step": 11497,
      "time_per_iteration": 2.595149517059326
    },
    {
      "auxiliary_loss_clip": 0.01070507,
      "auxiliary_loss_mlp": 0.01032623,
      "balance_loss_clip": 1.03659368,
      "balance_loss_mlp": 1.02026153,
      "epoch": 0.6912971591763114,
      "flos": 22090772620800.0,
      "grad_norm": 1.5666459166352902,
      "language_loss": 0.80494916,
      "learning_rate": 9.189715506138993e-07,
      "loss": 0.82598054,
      "num_input_tokens_seen": 248180430,
      "step": 11498,
      "time_per_iteration": 2.593575954437256
    },
    {
      "auxiliary_loss_clip": 0.01094559,
      "auxiliary_loss_mlp": 0.01030863,
      "balance_loss_clip": 1.03701878,
      "balance_loss_mlp": 1.01856613,
      "epoch": 0.6913572824289794,
      "flos": 29971476650880.0,
      "grad_norm": 1.6360571499419487,
      "language_loss": 0.85776275,
      "learning_rate": 9.186439034169915e-07,
      "loss": 0.879017,
      "num_input_tokens_seen": 248202365,
      "step": 11499,
      "time_per_iteration": 2.614140033721924
    },
    {
      "auxiliary_loss_clip": 0.01083387,
      "auxiliary_loss_mlp": 0.00786161,
      "balance_loss_clip": 1.04025352,
      "balance_loss_mlp": 1.01319003,
      "epoch": 0.6914174056816473,
      "flos": 20448936835200.0,
      "grad_norm": 1.6576742317174742,
      "language_loss": 0.75895166,
      "learning_rate": 9.183162972252145e-07,
      "loss": 0.77764714,
      "num_input_tokens_seen": 248221750,
      "step": 11500,
      "time_per_iteration": 2.542088031768799
    },
    {
      "auxiliary_loss_clip": 0.01050551,
      "auxiliary_loss_mlp": 0.0103942,
      "balance_loss_clip": 1.03364122,
      "balance_loss_mlp": 1.02534711,
      "epoch": 0.6914775289343154,
      "flos": 21282530739840.0,
      "grad_norm": 1.9332684901417738,
      "language_loss": 0.77126652,
      "learning_rate": 9.179887320509921e-07,
      "loss": 0.79216623,
      "num_input_tokens_seen": 248239535,
      "step": 11501,
      "time_per_iteration": 2.6261849403381348
    },
    {
      "auxiliary_loss_clip": 0.01094944,
      "auxiliary_loss_mlp": 0.01036617,
      "balance_loss_clip": 1.03756201,
      "balance_loss_mlp": 1.02380168,
      "epoch": 0.6915376521869833,
      "flos": 23878118401920.0,
      "grad_norm": 1.7147280278722756,
      "language_loss": 0.73538828,
      "learning_rate": 9.176612079067458e-07,
      "loss": 0.75670385,
      "num_input_tokens_seen": 248259055,
      "step": 11502,
      "time_per_iteration": 2.5268783569335938
    },
    {
      "auxiliary_loss_clip": 0.01039332,
      "auxiliary_loss_mlp": 0.01038969,
      "balance_loss_clip": 1.03506398,
      "balance_loss_mlp": 1.02438414,
      "epoch": 0.6915977754396513,
      "flos": 11510268595200.0,
      "grad_norm": 1.7539512755822142,
      "language_loss": 0.7345463,
      "learning_rate": 9.173337248048953e-07,
      "loss": 0.75532937,
      "num_input_tokens_seen": 248276765,
      "step": 11503,
      "time_per_iteration": 2.6470236778259277
    },
    {
      "auxiliary_loss_clip": 0.01093247,
      "auxiliary_loss_mlp": 0.01034107,
      "balance_loss_clip": 1.03653204,
      "balance_loss_mlp": 1.02107167,
      "epoch": 0.6916578986923193,
      "flos": 22601278667520.0,
      "grad_norm": 1.6932377836640133,
      "language_loss": 0.77073497,
      "learning_rate": 9.170062827578575e-07,
      "loss": 0.79200852,
      "num_input_tokens_seen": 248295310,
      "step": 11504,
      "time_per_iteration": 2.5143537521362305
    },
    {
      "auxiliary_loss_clip": 0.01065735,
      "auxiliary_loss_mlp": 0.01029651,
      "balance_loss_clip": 1.03413653,
      "balance_loss_mlp": 1.01576948,
      "epoch": 0.6917180219449872,
      "flos": 23477355383040.0,
      "grad_norm": 1.7519993769180897,
      "language_loss": 0.73731923,
      "learning_rate": 9.166788817780499e-07,
      "loss": 0.75827312,
      "num_input_tokens_seen": 248315230,
      "step": 11505,
      "time_per_iteration": 2.593613862991333
    },
    {
      "auxiliary_loss_clip": 0.01048753,
      "auxiliary_loss_mlp": 0.00783879,
      "balance_loss_clip": 1.03420949,
      "balance_loss_mlp": 1.00903058,
      "epoch": 0.6917781451976552,
      "flos": 23732536579200.0,
      "grad_norm": 1.8026245567771824,
      "language_loss": 0.87746751,
      "learning_rate": 9.163515218778886e-07,
      "loss": 0.8957938,
      "num_input_tokens_seen": 248332980,
      "step": 11506,
      "time_per_iteration": 4.057895183563232
    },
    {
      "auxiliary_loss_clip": 0.01081745,
      "auxiliary_loss_mlp": 0.01026309,
      "balance_loss_clip": 1.03855765,
      "balance_loss_mlp": 1.01452518,
      "epoch": 0.6918382684503231,
      "flos": 31466760946560.0,
      "grad_norm": 1.9780133252791468,
      "language_loss": 0.69949859,
      "learning_rate": 9.160242030697856e-07,
      "loss": 0.72057915,
      "num_input_tokens_seen": 248352865,
      "step": 11507,
      "time_per_iteration": 2.597958564758301
    },
    {
      "auxiliary_loss_clip": 0.01090087,
      "auxiliary_loss_mlp": 0.01034416,
      "balance_loss_clip": 1.03736949,
      "balance_loss_mlp": 1.0217973,
      "epoch": 0.6918983917029912,
      "flos": 21650471706240.0,
      "grad_norm": 1.880003189312143,
      "language_loss": 0.76902592,
      "learning_rate": 9.156969253661538e-07,
      "loss": 0.79027092,
      "num_input_tokens_seen": 248371125,
      "step": 11508,
      "time_per_iteration": 2.535619020462036
    },
    {
      "auxiliary_loss_clip": 0.01091399,
      "auxiliary_loss_mlp": 0.01027637,
      "balance_loss_clip": 1.03742683,
      "balance_loss_mlp": 1.01637757,
      "epoch": 0.6919585149556591,
      "flos": 25550082720000.0,
      "grad_norm": 1.6071026767318808,
      "language_loss": 0.75224471,
      "learning_rate": 9.153696887794027e-07,
      "loss": 0.77343506,
      "num_input_tokens_seen": 248390455,
      "step": 11509,
      "time_per_iteration": 2.532285213470459
    },
    {
      "auxiliary_loss_clip": 0.01058346,
      "auxiliary_loss_mlp": 0.01031165,
      "balance_loss_clip": 1.04072499,
      "balance_loss_mlp": 1.01964355,
      "epoch": 0.6920186382083271,
      "flos": 23659781581440.0,
      "grad_norm": 1.6680142996265412,
      "language_loss": 0.63952225,
      "learning_rate": 9.150424933219425e-07,
      "loss": 0.66041744,
      "num_input_tokens_seen": 248411305,
      "step": 11510,
      "time_per_iteration": 2.6466453075408936
    },
    {
      "auxiliary_loss_clip": 0.0107867,
      "auxiliary_loss_mlp": 0.01032557,
      "balance_loss_clip": 1.03913891,
      "balance_loss_mlp": 1.01890159,
      "epoch": 0.692078761460995,
      "flos": 19061959023360.0,
      "grad_norm": 2.0849651324754226,
      "language_loss": 0.75363463,
      "learning_rate": 9.147153390061788e-07,
      "loss": 0.77474689,
      "num_input_tokens_seen": 248430190,
      "step": 11511,
      "time_per_iteration": 2.5598580837249756
    },
    {
      "auxiliary_loss_clip": 0.01076386,
      "auxiliary_loss_mlp": 0.01027077,
      "balance_loss_clip": 1.03788114,
      "balance_loss_mlp": 1.0158236,
      "epoch": 0.692138884713663,
      "flos": 29023291382400.0,
      "grad_norm": 1.5764586035969197,
      "language_loss": 0.62677753,
      "learning_rate": 9.143882258445184e-07,
      "loss": 0.64781219,
      "num_input_tokens_seen": 248450830,
      "step": 11512,
      "time_per_iteration": 4.0212719440460205
    },
    {
      "auxiliary_loss_clip": 0.01076562,
      "auxiliary_loss_mlp": 0.01031216,
      "balance_loss_clip": 1.03709912,
      "balance_loss_mlp": 1.01877666,
      "epoch": 0.6921990079663309,
      "flos": 14757849976320.0,
      "grad_norm": 1.8368351142741068,
      "language_loss": 0.83071059,
      "learning_rate": 9.140611538493666e-07,
      "loss": 0.8517884,
      "num_input_tokens_seen": 248468585,
      "step": 11513,
      "time_per_iteration": 2.534562110900879
    },
    {
      "auxiliary_loss_clip": 0.01052205,
      "auxiliary_loss_mlp": 0.01030107,
      "balance_loss_clip": 1.03649378,
      "balance_loss_mlp": 1.0188117,
      "epoch": 0.692259131218999,
      "flos": 23841848643840.0,
      "grad_norm": 2.2776837305804483,
      "language_loss": 0.78419566,
      "learning_rate": 9.137341230331233e-07,
      "loss": 0.80501878,
      "num_input_tokens_seen": 248490535,
      "step": 11514,
      "time_per_iteration": 2.6435012817382812
    },
    {
      "auxiliary_loss_clip": 0.01062341,
      "auxiliary_loss_mlp": 0.01030621,
      "balance_loss_clip": 1.03468084,
      "balance_loss_mlp": 1.01869404,
      "epoch": 0.6923192544716669,
      "flos": 19135073157120.0,
      "grad_norm": 2.148732784216196,
      "language_loss": 0.74760127,
      "learning_rate": 9.134071334081907e-07,
      "loss": 0.76853085,
      "num_input_tokens_seen": 248508575,
      "step": 11515,
      "time_per_iteration": 2.5892364978790283
    },
    {
      "auxiliary_loss_clip": 0.01061557,
      "auxiliary_loss_mlp": 0.01031373,
      "balance_loss_clip": 1.03609776,
      "balance_loss_mlp": 1.01927888,
      "epoch": 0.6923793777243349,
      "flos": 28074639237120.0,
      "grad_norm": 2.154242519472449,
      "language_loss": 0.53908622,
      "learning_rate": 9.130801849869694e-07,
      "loss": 0.56001556,
      "num_input_tokens_seen": 248527025,
      "step": 11516,
      "time_per_iteration": 4.123652935028076
    },
    {
      "auxiliary_loss_clip": 0.01092494,
      "auxiliary_loss_mlp": 0.01034493,
      "balance_loss_clip": 1.03654683,
      "balance_loss_mlp": 1.02220249,
      "epoch": 0.6924395009770029,
      "flos": 16581250033920.0,
      "grad_norm": 1.964842651787121,
      "language_loss": 0.73093152,
      "learning_rate": 9.127532777818557e-07,
      "loss": 0.75220144,
      "num_input_tokens_seen": 248544275,
      "step": 11517,
      "time_per_iteration": 2.4755313396453857
    },
    {
      "auxiliary_loss_clip": 0.01109496,
      "auxiliary_loss_mlp": 0.0103376,
      "balance_loss_clip": 1.03839397,
      "balance_loss_mlp": 1.02141595,
      "epoch": 0.6924996242296708,
      "flos": 16655297921280.0,
      "grad_norm": 1.7210671489724525,
      "language_loss": 0.7633394,
      "learning_rate": 9.124264118052465e-07,
      "loss": 0.78477198,
      "num_input_tokens_seen": 248561870,
      "step": 11518,
      "time_per_iteration": 2.460892677307129
    },
    {
      "auxiliary_loss_clip": 0.01102218,
      "auxiliary_loss_mlp": 0.01033521,
      "balance_loss_clip": 1.0394969,
      "balance_loss_mlp": 1.02003217,
      "epoch": 0.6925597474823388,
      "flos": 34754167532160.0,
      "grad_norm": 1.3275083687108058,
      "language_loss": 0.64251351,
      "learning_rate": 9.120995870695376e-07,
      "loss": 0.66387093,
      "num_input_tokens_seen": 248588190,
      "step": 11519,
      "time_per_iteration": 2.648158550262451
    },
    {
      "auxiliary_loss_clip": 0.01076527,
      "auxiliary_loss_mlp": 0.0103226,
      "balance_loss_clip": 1.03673315,
      "balance_loss_mlp": 1.01984429,
      "epoch": 0.6926198707350067,
      "flos": 21871717528320.0,
      "grad_norm": 1.732336126451384,
      "language_loss": 0.62479293,
      "learning_rate": 9.117728035871212e-07,
      "loss": 0.64588082,
      "num_input_tokens_seen": 248606460,
      "step": 11520,
      "time_per_iteration": 2.539802074432373
    },
    {
      "auxiliary_loss_clip": 0.01072514,
      "auxiliary_loss_mlp": 0.01034323,
      "balance_loss_clip": 1.03715491,
      "balance_loss_mlp": 1.0202328,
      "epoch": 0.6926799939876748,
      "flos": 13006271162880.0,
      "grad_norm": 2.089066612918736,
      "language_loss": 0.78004092,
      "learning_rate": 9.114460613703887e-07,
      "loss": 0.80110937,
      "num_input_tokens_seen": 248623715,
      "step": 11521,
      "time_per_iteration": 2.5447330474853516
    },
    {
      "auxiliary_loss_clip": 0.0109783,
      "auxiliary_loss_mlp": 0.01038239,
      "balance_loss_clip": 1.03801024,
      "balance_loss_mlp": 1.0234983,
      "epoch": 0.6927401172403427,
      "flos": 16761234107520.0,
      "grad_norm": 2.12506696948796,
      "language_loss": 0.81809753,
      "learning_rate": 9.111193604317304e-07,
      "loss": 0.83945823,
      "num_input_tokens_seen": 248640575,
      "step": 11522,
      "time_per_iteration": 2.4877171516418457
    },
    {
      "auxiliary_loss_clip": 0.01093234,
      "auxiliary_loss_mlp": 0.01029576,
      "balance_loss_clip": 1.04046619,
      "balance_loss_mlp": 1.01786947,
      "epoch": 0.6928002404930107,
      "flos": 25705648523520.0,
      "grad_norm": 1.3680752820117759,
      "language_loss": 0.76819265,
      "learning_rate": 9.107927007835361e-07,
      "loss": 0.78942072,
      "num_input_tokens_seen": 248663535,
      "step": 11523,
      "time_per_iteration": 2.5879862308502197
    },
    {
      "auxiliary_loss_clip": 0.01076897,
      "auxiliary_loss_mlp": 0.01033836,
      "balance_loss_clip": 1.03781271,
      "balance_loss_mlp": 1.02240419,
      "epoch": 0.6928603637456786,
      "flos": 18588261438720.0,
      "grad_norm": 1.6828766233132744,
      "language_loss": 0.6805048,
      "learning_rate": 9.104660824381915e-07,
      "loss": 0.70161211,
      "num_input_tokens_seen": 248681125,
      "step": 11524,
      "time_per_iteration": 2.5742835998535156
    },
    {
      "auxiliary_loss_clip": 0.01074975,
      "auxiliary_loss_mlp": 0.01029699,
      "balance_loss_clip": 1.03701997,
      "balance_loss_mlp": 1.0172776,
      "epoch": 0.6929204869983466,
      "flos": 22200874784640.0,
      "grad_norm": 1.566347873841233,
      "language_loss": 0.64769542,
      "learning_rate": 9.101395054080815e-07,
      "loss": 0.66874218,
      "num_input_tokens_seen": 248700555,
      "step": 11525,
      "time_per_iteration": 2.57802677154541
    },
    {
      "auxiliary_loss_clip": 0.0106811,
      "auxiliary_loss_mlp": 0.01036019,
      "balance_loss_clip": 1.03706932,
      "balance_loss_mlp": 1.0238539,
      "epoch": 0.6929806102510145,
      "flos": 17894754576000.0,
      "grad_norm": 2.1958125975303218,
      "language_loss": 0.70275247,
      "learning_rate": 9.098129697055907e-07,
      "loss": 0.72379375,
      "num_input_tokens_seen": 248716095,
      "step": 11526,
      "time_per_iteration": 2.5674993991851807
    },
    {
      "auxiliary_loss_clip": 0.01083358,
      "auxiliary_loss_mlp": 0.01028052,
      "balance_loss_clip": 1.03500605,
      "balance_loss_mlp": 1.01619065,
      "epoch": 0.6930407335036826,
      "flos": 19755178577280.0,
      "grad_norm": 1.503721439238539,
      "language_loss": 0.76036912,
      "learning_rate": 9.094864753431022e-07,
      "loss": 0.78148317,
      "num_input_tokens_seen": 248735330,
      "step": 11527,
      "time_per_iteration": 2.51962947845459
    },
    {
      "auxiliary_loss_clip": 0.01080973,
      "auxiliary_loss_mlp": 0.01033222,
      "balance_loss_clip": 1.03536916,
      "balance_loss_mlp": 1.0208776,
      "epoch": 0.6931008567563505,
      "flos": 21544248211200.0,
      "grad_norm": 1.4546127968777582,
      "language_loss": 0.79153526,
      "learning_rate": 9.091600223329952e-07,
      "loss": 0.8126772,
      "num_input_tokens_seen": 248754530,
      "step": 11528,
      "time_per_iteration": 2.5327298641204834
    },
    {
      "auxiliary_loss_clip": 0.01093706,
      "auxiliary_loss_mlp": 0.01030535,
      "balance_loss_clip": 1.03776288,
      "balance_loss_mlp": 1.01863217,
      "epoch": 0.6931609800090185,
      "flos": 26250018117120.0,
      "grad_norm": 1.4104602640715094,
      "language_loss": 0.76036894,
      "learning_rate": 9.088336106876491e-07,
      "loss": 0.78161132,
      "num_input_tokens_seen": 248775825,
      "step": 11529,
      "time_per_iteration": 2.539283275604248
    },
    {
      "auxiliary_loss_clip": 0.01106062,
      "auxiliary_loss_mlp": 0.00782841,
      "balance_loss_clip": 1.03866982,
      "balance_loss_mlp": 1.0085938,
      "epoch": 0.6932211032616865,
      "flos": 32343376366080.0,
      "grad_norm": 6.982333751388168,
      "language_loss": 0.71928287,
      "learning_rate": 9.085072404194436e-07,
      "loss": 0.73817194,
      "num_input_tokens_seen": 248796180,
      "step": 11530,
      "time_per_iteration": 2.5521209239959717
    },
    {
      "auxiliary_loss_clip": 0.01093341,
      "auxiliary_loss_mlp": 0.01033172,
      "balance_loss_clip": 1.04072857,
      "balance_loss_mlp": 1.01924264,
      "epoch": 0.6932812265143544,
      "flos": 22049079909120.0,
      "grad_norm": 1.9515254673676938,
      "language_loss": 0.78686053,
      "learning_rate": 9.081809115407513e-07,
      "loss": 0.80812573,
      "num_input_tokens_seen": 248814735,
      "step": 11531,
      "time_per_iteration": 2.5200512409210205
    },
    {
      "auxiliary_loss_clip": 0.01092199,
      "auxiliary_loss_mlp": 0.01034945,
      "balance_loss_clip": 1.03893888,
      "balance_loss_mlp": 1.0242753,
      "epoch": 0.6933413497670224,
      "flos": 26256626219520.0,
      "grad_norm": 1.4085295446256927,
      "language_loss": 0.69592738,
      "learning_rate": 9.078546240639484e-07,
      "loss": 0.71719885,
      "num_input_tokens_seen": 248839140,
      "step": 11532,
      "time_per_iteration": 2.58559513092041
    },
    {
      "auxiliary_loss_clip": 0.01084519,
      "auxiliary_loss_mlp": 0.01031373,
      "balance_loss_clip": 1.03865266,
      "balance_loss_mlp": 1.01858234,
      "epoch": 0.6934014730196904,
      "flos": 19573003774080.0,
      "grad_norm": 1.554673290477012,
      "language_loss": 0.67017949,
      "learning_rate": 9.075283780014082e-07,
      "loss": 0.69133836,
      "num_input_tokens_seen": 248858300,
      "step": 11533,
      "time_per_iteration": 2.5110485553741455
    },
    {
      "auxiliary_loss_clip": 0.01085987,
      "auxiliary_loss_mlp": 0.01030007,
      "balance_loss_clip": 1.04090405,
      "balance_loss_mlp": 1.01711404,
      "epoch": 0.6934615962723584,
      "flos": 22119249127680.0,
      "grad_norm": 2.977107955291198,
      "language_loss": 0.59072703,
      "learning_rate": 9.072021733655007e-07,
      "loss": 0.61188698,
      "num_input_tokens_seen": 248876310,
      "step": 11534,
      "time_per_iteration": 2.52600359916687
    },
    {
      "auxiliary_loss_clip": 0.01073416,
      "auxiliary_loss_mlp": 0.0103144,
      "balance_loss_clip": 1.03744996,
      "balance_loss_mlp": 1.01847553,
      "epoch": 0.6935217195250263,
      "flos": 21360816432000.0,
      "grad_norm": 2.0324317405649754,
      "language_loss": 0.70779502,
      "learning_rate": 9.068760101685971e-07,
      "loss": 0.72884357,
      "num_input_tokens_seen": 248895650,
      "step": 11535,
      "time_per_iteration": 2.5498600006103516
    },
    {
      "auxiliary_loss_clip": 0.01021595,
      "auxiliary_loss_mlp": 0.01001887,
      "balance_loss_clip": 1.0176003,
      "balance_loss_mlp": 1.00069451,
      "epoch": 0.6935818427776943,
      "flos": 64063813115520.0,
      "grad_norm": 0.7241843174833122,
      "language_loss": 0.59044254,
      "learning_rate": 9.065498884230638e-07,
      "loss": 0.6106773,
      "num_input_tokens_seen": 248963920,
      "step": 11536,
      "time_per_iteration": 3.255309581756592
    },
    {
      "auxiliary_loss_clip": 0.01096492,
      "auxiliary_loss_mlp": 0.00784736,
      "balance_loss_clip": 1.03987789,
      "balance_loss_mlp": 1.01075459,
      "epoch": 0.6936419660303622,
      "flos": 20302564913280.0,
      "grad_norm": 1.7529541804176498,
      "language_loss": 0.72382271,
      "learning_rate": 9.062238081412692e-07,
      "loss": 0.74263501,
      "num_input_tokens_seen": 248983380,
      "step": 11537,
      "time_per_iteration": 2.5206539630889893
    },
    {
      "auxiliary_loss_clip": 0.01031541,
      "auxiliary_loss_mlp": 0.00766933,
      "balance_loss_clip": 1.01679254,
      "balance_loss_mlp": 1.00851095,
      "epoch": 0.6937020892830302,
      "flos": 67182581347200.0,
      "grad_norm": 0.7528622964845968,
      "language_loss": 0.55615675,
      "learning_rate": 9.058977693355767e-07,
      "loss": 0.5741415,
      "num_input_tokens_seen": 249044680,
      "step": 11538,
      "time_per_iteration": 3.119452714920044
    },
    {
      "auxiliary_loss_clip": 0.01093215,
      "auxiliary_loss_mlp": 0.01028778,
      "balance_loss_clip": 1.0377512,
      "balance_loss_mlp": 1.01763177,
      "epoch": 0.6937622125356981,
      "flos": 23878190229120.0,
      "grad_norm": 1.5262308818576538,
      "language_loss": 0.77768296,
      "learning_rate": 9.055717720183505e-07,
      "loss": 0.79890287,
      "num_input_tokens_seen": 249061060,
      "step": 11539,
      "time_per_iteration": 2.5113086700439453
    },
    {
      "auxiliary_loss_clip": 0.01084979,
      "auxiliary_loss_mlp": 0.01027042,
      "balance_loss_clip": 1.03773904,
      "balance_loss_mlp": 1.0156281,
      "epoch": 0.6938223357883662,
      "flos": 28730619365760.0,
      "grad_norm": 1.5878029127671371,
      "language_loss": 0.63956314,
      "learning_rate": 9.05245816201953e-07,
      "loss": 0.66068333,
      "num_input_tokens_seen": 249081430,
      "step": 11540,
      "time_per_iteration": 2.5757639408111572
    },
    {
      "auxiliary_loss_clip": 0.01071618,
      "auxiliary_loss_mlp": 0.01032531,
      "balance_loss_clip": 1.03737032,
      "balance_loss_mlp": 1.02079487,
      "epoch": 0.6938824590410341,
      "flos": 28655027193600.0,
      "grad_norm": 1.4489586272488857,
      "language_loss": 0.86628324,
      "learning_rate": 9.049199018987437e-07,
      "loss": 0.88732469,
      "num_input_tokens_seen": 249103020,
      "step": 11541,
      "time_per_iteration": 2.6268303394317627
    },
    {
      "auxiliary_loss_clip": 0.01109956,
      "auxiliary_loss_mlp": 0.00783026,
      "balance_loss_clip": 1.03911757,
      "balance_loss_mlp": 1.00755906,
      "epoch": 0.6939425822937021,
      "flos": 18983062800000.0,
      "grad_norm": 1.766015316804286,
      "language_loss": 0.841021,
      "learning_rate": 9.04594029121081e-07,
      "loss": 0.8599509,
      "num_input_tokens_seen": 249120810,
      "step": 11542,
      "time_per_iteration": 2.4829230308532715
    },
    {
      "auxiliary_loss_clip": 0.01096475,
      "auxiliary_loss_mlp": 0.01030197,
      "balance_loss_clip": 1.03753901,
      "balance_loss_mlp": 1.01699436,
      "epoch": 0.6940027055463701,
      "flos": 23075838178560.0,
      "grad_norm": 1.8431457194930394,
      "language_loss": 0.75162065,
      "learning_rate": 9.04268197881323e-07,
      "loss": 0.77288735,
      "num_input_tokens_seen": 249138050,
      "step": 11543,
      "time_per_iteration": 2.5082197189331055
    },
    {
      "auxiliary_loss_clip": 0.01085049,
      "auxiliary_loss_mlp": 0.01031878,
      "balance_loss_clip": 1.03581154,
      "balance_loss_mlp": 1.0199337,
      "epoch": 0.694062828799038,
      "flos": 18186564666240.0,
      "grad_norm": 1.6546455438981238,
      "language_loss": 0.76035595,
      "learning_rate": 9.039424081918241e-07,
      "loss": 0.78152525,
      "num_input_tokens_seen": 249155570,
      "step": 11544,
      "time_per_iteration": 2.518556594848633
    },
    {
      "auxiliary_loss_clip": 0.0105658,
      "auxiliary_loss_mlp": 0.01038495,
      "balance_loss_clip": 1.03617811,
      "balance_loss_mlp": 1.02508414,
      "epoch": 0.694122952051706,
      "flos": 17821532701440.0,
      "grad_norm": 1.7018896824648146,
      "language_loss": 0.70961523,
      "learning_rate": 9.036166600649388e-07,
      "loss": 0.73056597,
      "num_input_tokens_seen": 249172960,
      "step": 11545,
      "time_per_iteration": 3.967756748199463
    },
    {
      "auxiliary_loss_clip": 0.01093874,
      "auxiliary_loss_mlp": 0.01028782,
      "balance_loss_clip": 1.03851128,
      "balance_loss_mlp": 1.01778531,
      "epoch": 0.694183075304374,
      "flos": 21215306436480.0,
      "grad_norm": 1.8222998799391918,
      "language_loss": 0.79489124,
      "learning_rate": 9.0329095351302e-07,
      "loss": 0.81611782,
      "num_input_tokens_seen": 249192450,
      "step": 11546,
      "time_per_iteration": 2.4853761196136475
    },
    {
      "auxiliary_loss_clip": 0.01073808,
      "auxiliary_loss_mlp": 0.01028575,
      "balance_loss_clip": 1.03572989,
      "balance_loss_mlp": 1.01664805,
      "epoch": 0.694243198557042,
      "flos": 24060508686720.0,
      "grad_norm": 1.3841970929035312,
      "language_loss": 0.78597176,
      "learning_rate": 9.029652885484194e-07,
      "loss": 0.80699551,
      "num_input_tokens_seen": 249214320,
      "step": 11547,
      "time_per_iteration": 2.5980873107910156
    },
    {
      "auxiliary_loss_clip": 0.01085946,
      "auxiliary_loss_mlp": 0.00783726,
      "balance_loss_clip": 1.04119778,
      "balance_loss_mlp": 1.00895524,
      "epoch": 0.6943033218097099,
      "flos": 21141869080320.0,
      "grad_norm": 1.9956712134191052,
      "language_loss": 0.80624211,
      "learning_rate": 9.026396651834834e-07,
      "loss": 0.82493889,
      "num_input_tokens_seen": 249230925,
      "step": 11548,
      "time_per_iteration": 2.5110929012298584
    },
    {
      "auxiliary_loss_clip": 0.01040014,
      "auxiliary_loss_mlp": 0.00767949,
      "balance_loss_clip": 1.01642752,
      "balance_loss_mlp": 1.00862718,
      "epoch": 0.6943634450623779,
      "flos": 57812015975040.0,
      "grad_norm": 0.693627695318321,
      "language_loss": 0.53716421,
      "learning_rate": 9.023140834305613e-07,
      "loss": 0.55524385,
      "num_input_tokens_seen": 249293975,
      "step": 11549,
      "time_per_iteration": 3.0799429416656494
    },
    {
      "auxiliary_loss_clip": 0.01090059,
      "auxiliary_loss_mlp": 0.0103254,
      "balance_loss_clip": 1.03443229,
      "balance_loss_mlp": 1.02002859,
      "epoch": 0.6944235683150458,
      "flos": 30590684231040.0,
      "grad_norm": 1.3328138422694822,
      "language_loss": 0.73501015,
      "learning_rate": 9.01988543302e-07,
      "loss": 0.75623608,
      "num_input_tokens_seen": 249315285,
      "step": 11550,
      "time_per_iteration": 3.981412410736084
    },
    {
      "auxiliary_loss_clip": 0.0108495,
      "auxiliary_loss_mlp": 0.01030778,
      "balance_loss_clip": 1.04017138,
      "balance_loss_mlp": 1.01856518,
      "epoch": 0.6944836915677138,
      "flos": 19719447523200.0,
      "grad_norm": 1.8918275700491303,
      "language_loss": 0.74127704,
      "learning_rate": 9.016630448101425e-07,
      "loss": 0.76243436,
      "num_input_tokens_seen": 249333505,
      "step": 11551,
      "time_per_iteration": 2.5003764629364014
    },
    {
      "auxiliary_loss_clip": 0.01110293,
      "auxiliary_loss_mlp": 0.01033751,
      "balance_loss_clip": 1.03992248,
      "balance_loss_mlp": 1.02130008,
      "epoch": 0.6945438148203817,
      "flos": 24863579009280.0,
      "grad_norm": 1.480116436208681,
      "language_loss": 0.84529865,
      "learning_rate": 9.01337587967333e-07,
      "loss": 0.86673903,
      "num_input_tokens_seen": 249354180,
      "step": 11552,
      "time_per_iteration": 2.510984182357788
    },
    {
      "auxiliary_loss_clip": 0.01107875,
      "auxiliary_loss_mlp": 0.01035585,
      "balance_loss_clip": 1.03833151,
      "balance_loss_mlp": 1.02324045,
      "epoch": 0.6946039380730498,
      "flos": 33326646243840.0,
      "grad_norm": 1.456649241300002,
      "language_loss": 0.67661858,
      "learning_rate": 9.010121727859117e-07,
      "loss": 0.69805324,
      "num_input_tokens_seen": 249377035,
      "step": 11553,
      "time_per_iteration": 2.56299090385437
    },
    {
      "auxiliary_loss_clip": 0.01091482,
      "auxiliary_loss_mlp": 0.01030902,
      "balance_loss_clip": 1.03878379,
      "balance_loss_mlp": 1.01799822,
      "epoch": 0.6946640613257177,
      "flos": 20850956830080.0,
      "grad_norm": 1.5602166679508476,
      "language_loss": 0.79334867,
      "learning_rate": 9.006867992782195e-07,
      "loss": 0.81457257,
      "num_input_tokens_seen": 249396155,
      "step": 11554,
      "time_per_iteration": 2.522529125213623
    },
    {
      "auxiliary_loss_clip": 0.01099624,
      "auxiliary_loss_mlp": 0.01027459,
      "balance_loss_clip": 1.03815126,
      "balance_loss_mlp": 1.01539516,
      "epoch": 0.6947241845783857,
      "flos": 19354846521600.0,
      "grad_norm": 1.8916059669058414,
      "language_loss": 0.72587615,
      "learning_rate": 9.003614674565934e-07,
      "loss": 0.74714702,
      "num_input_tokens_seen": 249414555,
      "step": 11555,
      "time_per_iteration": 3.9766833782196045
    },
    {
      "auxiliary_loss_clip": 0.01071789,
      "auxiliary_loss_mlp": 0.01028035,
      "balance_loss_clip": 1.03719246,
      "balance_loss_mlp": 1.01648962,
      "epoch": 0.6947843078310536,
      "flos": 27120240915840.0,
      "grad_norm": 1.6388796685923996,
      "language_loss": 0.78356063,
      "learning_rate": 9.000361773333705e-07,
      "loss": 0.80455887,
      "num_input_tokens_seen": 249433570,
      "step": 11556,
      "time_per_iteration": 2.601313829421997
    },
    {
      "auxiliary_loss_clip": 0.01048784,
      "auxiliary_loss_mlp": 0.01036342,
      "balance_loss_clip": 1.03508961,
      "balance_loss_mlp": 1.02427769,
      "epoch": 0.6948444310837216,
      "flos": 28585109370240.0,
      "grad_norm": 1.9986140100262613,
      "language_loss": 0.60326982,
      "learning_rate": 8.997109289208869e-07,
      "loss": 0.62412107,
      "num_input_tokens_seen": 249453735,
      "step": 11557,
      "time_per_iteration": 2.6899044513702393
    },
    {
      "auxiliary_loss_clip": 0.01081671,
      "auxiliary_loss_mlp": 0.01039409,
      "balance_loss_clip": 1.03883576,
      "balance_loss_mlp": 1.02695775,
      "epoch": 0.6949045543363896,
      "flos": 15669262696320.0,
      "grad_norm": 1.9502714952820943,
      "language_loss": 0.85429251,
      "learning_rate": 8.993857222314752e-07,
      "loss": 0.8755033,
      "num_input_tokens_seen": 249470805,
      "step": 11558,
      "time_per_iteration": 2.5189859867095947
    },
    {
      "auxiliary_loss_clip": 0.01099455,
      "auxiliary_loss_mlp": 0.01032285,
      "balance_loss_clip": 1.0372684,
      "balance_loss_mlp": 1.01921415,
      "epoch": 0.6949646775890576,
      "flos": 23259413612160.0,
      "grad_norm": 1.6264404527466336,
      "language_loss": 0.70318854,
      "learning_rate": 8.990605572774664e-07,
      "loss": 0.7245059,
      "num_input_tokens_seen": 249491150,
      "step": 11559,
      "time_per_iteration": 2.5682179927825928
    },
    {
      "auxiliary_loss_clip": 0.01072851,
      "auxiliary_loss_mlp": 0.01031149,
      "balance_loss_clip": 1.03717589,
      "balance_loss_mlp": 1.01955581,
      "epoch": 0.6950248008417256,
      "flos": 22382546797440.0,
      "grad_norm": 1.502713768165792,
      "language_loss": 0.78759617,
      "learning_rate": 8.987354340711921e-07,
      "loss": 0.80863613,
      "num_input_tokens_seen": 249511560,
      "step": 11560,
      "time_per_iteration": 2.5688703060150146
    },
    {
      "auxiliary_loss_clip": 0.01083954,
      "auxiliary_loss_mlp": 0.0103489,
      "balance_loss_clip": 1.03869629,
      "balance_loss_mlp": 1.02327943,
      "epoch": 0.6950849240943935,
      "flos": 23477355383040.0,
      "grad_norm": 1.5939421772320905,
      "language_loss": 0.77200353,
      "learning_rate": 8.9841035262498e-07,
      "loss": 0.79319191,
      "num_input_tokens_seen": 249531910,
      "step": 11561,
      "time_per_iteration": 2.5506646633148193
    },
    {
      "auxiliary_loss_clip": 0.01105125,
      "auxiliary_loss_mlp": 0.0103237,
      "balance_loss_clip": 1.03588986,
      "balance_loss_mlp": 1.01931643,
      "epoch": 0.6951450473470615,
      "flos": 17420554200960.0,
      "grad_norm": 1.9766727316345976,
      "language_loss": 0.78556663,
      "learning_rate": 8.980853129511577e-07,
      "loss": 0.80694163,
      "num_input_tokens_seen": 249550300,
      "step": 11562,
      "time_per_iteration": 2.434783697128296
    },
    {
      "auxiliary_loss_clip": 0.01098796,
      "auxiliary_loss_mlp": 0.01032259,
      "balance_loss_clip": 1.03601718,
      "balance_loss_mlp": 1.01979017,
      "epoch": 0.6952051705997294,
      "flos": 20485745297280.0,
      "grad_norm": 2.4021135688952255,
      "language_loss": 0.69444847,
      "learning_rate": 8.977603150620515e-07,
      "loss": 0.71575904,
      "num_input_tokens_seen": 249567740,
      "step": 11563,
      "time_per_iteration": 2.5006840229034424
    },
    {
      "auxiliary_loss_clip": 0.0109366,
      "auxiliary_loss_mlp": 0.01026818,
      "balance_loss_clip": 1.037328,
      "balance_loss_mlp": 1.01570773,
      "epoch": 0.6952652938523974,
      "flos": 13989541040640.0,
      "grad_norm": 2.1051244499605395,
      "language_loss": 0.73340476,
      "learning_rate": 8.974353589699846e-07,
      "loss": 0.75460958,
      "num_input_tokens_seen": 249582700,
      "step": 11564,
      "time_per_iteration": 2.4636423587799072
    },
    {
      "auxiliary_loss_clip": 0.0108898,
      "auxiliary_loss_mlp": 0.01035111,
      "balance_loss_clip": 1.04445517,
      "balance_loss_mlp": 1.02017379,
      "epoch": 0.6953254171050653,
      "flos": 30953956429440.0,
      "grad_norm": 1.9197100866343377,
      "language_loss": 0.71899599,
      "learning_rate": 8.971104446872785e-07,
      "loss": 0.74023688,
      "num_input_tokens_seen": 249602920,
      "step": 11565,
      "time_per_iteration": 2.6331241130828857
    },
    {
      "auxiliary_loss_clip": 0.01022016,
      "auxiliary_loss_mlp": 0.01001067,
      "balance_loss_clip": 1.0166868,
      "balance_loss_mlp": 0.99992907,
      "epoch": 0.6953855403577334,
      "flos": 61670257499520.0,
      "grad_norm": 0.9210811614800977,
      "language_loss": 0.58498913,
      "learning_rate": 8.96785572226255e-07,
      "loss": 0.60521996,
      "num_input_tokens_seen": 249660400,
      "step": 11566,
      "time_per_iteration": 2.984909772872925
    },
    {
      "auxiliary_loss_clip": 0.01072804,
      "auxiliary_loss_mlp": 0.01031733,
      "balance_loss_clip": 1.0366025,
      "balance_loss_mlp": 1.0192641,
      "epoch": 0.6954456636104013,
      "flos": 23039029716480.0,
      "grad_norm": 1.942226661943873,
      "language_loss": 0.73774338,
      "learning_rate": 8.964607415992338e-07,
      "loss": 0.7587887,
      "num_input_tokens_seen": 249679335,
      "step": 11567,
      "time_per_iteration": 2.576402425765991
    },
    {
      "auxiliary_loss_clip": 0.01077205,
      "auxiliary_loss_mlp": 0.01034757,
      "balance_loss_clip": 1.03492141,
      "balance_loss_mlp": 1.02175689,
      "epoch": 0.6955057868630693,
      "flos": 23918518224000.0,
      "grad_norm": 1.2375882145837898,
      "language_loss": 0.7700156,
      "learning_rate": 8.961359528185313e-07,
      "loss": 0.79113525,
      "num_input_tokens_seen": 249701805,
      "step": 11568,
      "time_per_iteration": 2.571838140487671
    },
    {
      "auxiliary_loss_clip": 0.01100896,
      "auxiliary_loss_mlp": 0.0103356,
      "balance_loss_clip": 1.0418129,
      "balance_loss_mlp": 1.02191293,
      "epoch": 0.6955659101157372,
      "flos": 22594634651520.0,
      "grad_norm": 1.6531010744445864,
      "language_loss": 0.72638625,
      "learning_rate": 8.958112058964649e-07,
      "loss": 0.74773085,
      "num_input_tokens_seen": 249720550,
      "step": 11569,
      "time_per_iteration": 2.506361961364746
    },
    {
      "auxiliary_loss_clip": 0.01085143,
      "auxiliary_loss_mlp": 0.0103026,
      "balance_loss_clip": 1.03961325,
      "balance_loss_mlp": 1.01775467,
      "epoch": 0.6956260333684052,
      "flos": 24572523104640.0,
      "grad_norm": 1.9059175120258178,
      "language_loss": 0.77295828,
      "learning_rate": 8.954865008453471e-07,
      "loss": 0.79411232,
      "num_input_tokens_seen": 249740325,
      "step": 11570,
      "time_per_iteration": 2.5600571632385254
    },
    {
      "auxiliary_loss_clip": 0.01098457,
      "auxiliary_loss_mlp": 0.01030787,
      "balance_loss_clip": 1.03717446,
      "balance_loss_mlp": 1.01830602,
      "epoch": 0.6956861566210732,
      "flos": 25846058787840.0,
      "grad_norm": 1.7048480305460711,
      "language_loss": 0.74396443,
      "learning_rate": 8.95161837677493e-07,
      "loss": 0.76525694,
      "num_input_tokens_seen": 249760570,
      "step": 11571,
      "time_per_iteration": 2.537386417388916
    },
    {
      "auxiliary_loss_clip": 0.01092564,
      "auxiliary_loss_mlp": 0.01033948,
      "balance_loss_clip": 1.03711987,
      "balance_loss_mlp": 1.02169919,
      "epoch": 0.6957462798737412,
      "flos": 15301393557120.0,
      "grad_norm": 1.8242420376776582,
      "language_loss": 0.74985242,
      "learning_rate": 8.948372164052118e-07,
      "loss": 0.77111757,
      "num_input_tokens_seen": 249778290,
      "step": 11572,
      "time_per_iteration": 2.4695069789886475
    },
    {
      "auxiliary_loss_clip": 0.01084639,
      "auxiliary_loss_mlp": 0.01026366,
      "balance_loss_clip": 1.03598559,
      "balance_loss_mlp": 1.01436734,
      "epoch": 0.6958064031264092,
      "flos": 36246830135040.0,
      "grad_norm": 1.716612678357441,
      "language_loss": 0.70222354,
      "learning_rate": 8.94512637040814e-07,
      "loss": 0.72333366,
      "num_input_tokens_seen": 249800925,
      "step": 11573,
      "time_per_iteration": 2.6503615379333496
    },
    {
      "auxiliary_loss_clip": 0.01090087,
      "auxiliary_loss_mlp": 0.01034883,
      "balance_loss_clip": 1.04286349,
      "balance_loss_mlp": 1.02230668,
      "epoch": 0.6958665263790771,
      "flos": 19208725994880.0,
      "grad_norm": 1.729779742476007,
      "language_loss": 0.74797636,
      "learning_rate": 8.941880995966095e-07,
      "loss": 0.76922613,
      "num_input_tokens_seen": 249820500,
      "step": 11574,
      "time_per_iteration": 2.5210659503936768
    },
    {
      "auxiliary_loss_clip": 0.01074309,
      "auxiliary_loss_mlp": 0.01032494,
      "balance_loss_clip": 1.03454471,
      "balance_loss_mlp": 1.02037668,
      "epoch": 0.6959266496317451,
      "flos": 21795838047360.0,
      "grad_norm": 1.6066623123268264,
      "language_loss": 0.74561888,
      "learning_rate": 8.938636040849014e-07,
      "loss": 0.76668692,
      "num_input_tokens_seen": 249839845,
      "step": 11575,
      "time_per_iteration": 2.549417734146118
    },
    {
      "auxiliary_loss_clip": 0.01099038,
      "auxiliary_loss_mlp": 0.01031629,
      "balance_loss_clip": 1.03831816,
      "balance_loss_mlp": 1.01856947,
      "epoch": 0.695986772884413,
      "flos": 20558248899840.0,
      "grad_norm": 1.8538208113472099,
      "language_loss": 0.78650939,
      "learning_rate": 8.935391505179966e-07,
      "loss": 0.80781603,
      "num_input_tokens_seen": 249857400,
      "step": 11576,
      "time_per_iteration": 2.48110032081604
    },
    {
      "auxiliary_loss_clip": 0.01066454,
      "auxiliary_loss_mlp": 0.01032171,
      "balance_loss_clip": 1.036865,
      "balance_loss_mlp": 1.02029741,
      "epoch": 0.696046896137081,
      "flos": 14936217937920.0,
      "grad_norm": 2.1154820446153177,
      "language_loss": 0.56747109,
      "learning_rate": 8.932147389081985e-07,
      "loss": 0.58845735,
      "num_input_tokens_seen": 249871645,
      "step": 11577,
      "time_per_iteration": 2.546812057495117
    },
    {
      "auxiliary_loss_clip": 0.01028256,
      "auxiliary_loss_mlp": 0.01025362,
      "balance_loss_clip": 1.03237867,
      "balance_loss_mlp": 1.01451969,
      "epoch": 0.696107019389749,
      "flos": 30740216549760.0,
      "grad_norm": 1.297726010261835,
      "language_loss": 0.76719797,
      "learning_rate": 8.928903692678081e-07,
      "loss": 0.78773415,
      "num_input_tokens_seen": 249894215,
      "step": 11578,
      "time_per_iteration": 2.885678291320801
    },
    {
      "auxiliary_loss_clip": 0.01074987,
      "auxiliary_loss_mlp": 0.01035105,
      "balance_loss_clip": 1.03776205,
      "balance_loss_mlp": 1.02268338,
      "epoch": 0.696167142642417,
      "flos": 20776729374720.0,
      "grad_norm": 1.9584420358393861,
      "language_loss": 0.79627872,
      "learning_rate": 8.925660416091254e-07,
      "loss": 0.81737959,
      "num_input_tokens_seen": 249912850,
      "step": 11579,
      "time_per_iteration": 2.7108285427093506
    },
    {
      "auxiliary_loss_clip": 0.01067143,
      "auxiliary_loss_mlp": 0.0102794,
      "balance_loss_clip": 1.03409719,
      "balance_loss_mlp": 1.01533961,
      "epoch": 0.6962272658950849,
      "flos": 22565152563840.0,
      "grad_norm": 1.7330315618683318,
      "language_loss": 0.73211539,
      "learning_rate": 8.922417559444502e-07,
      "loss": 0.75306618,
      "num_input_tokens_seen": 249932650,
      "step": 11580,
      "time_per_iteration": 2.577279567718506
    },
    {
      "auxiliary_loss_clip": 0.01090181,
      "auxiliary_loss_mlp": 0.01032286,
      "balance_loss_clip": 1.0374825,
      "balance_loss_mlp": 1.01890469,
      "epoch": 0.6962873891477529,
      "flos": 22200156512640.0,
      "grad_norm": 2.0967796874827416,
      "language_loss": 0.65886384,
      "learning_rate": 8.919175122860787e-07,
      "loss": 0.68008846,
      "num_input_tokens_seen": 249951205,
      "step": 11581,
      "time_per_iteration": 2.5310490131378174
    },
    {
      "auxiliary_loss_clip": 0.01110171,
      "auxiliary_loss_mlp": 0.01030606,
      "balance_loss_clip": 1.03931451,
      "balance_loss_mlp": 1.01875687,
      "epoch": 0.6963475124004208,
      "flos": 12489695717760.0,
      "grad_norm": 2.0552883552307475,
      "language_loss": 0.76492858,
      "learning_rate": 8.915933106463056e-07,
      "loss": 0.78633636,
      "num_input_tokens_seen": 249967045,
      "step": 11582,
      "time_per_iteration": 2.4396276473999023
    },
    {
      "auxiliary_loss_clip": 0.01083505,
      "auxiliary_loss_mlp": 0.01029433,
      "balance_loss_clip": 1.03555989,
      "balance_loss_mlp": 1.01787579,
      "epoch": 0.6964076356530888,
      "flos": 17165085696000.0,
      "grad_norm": 1.887196774370862,
      "language_loss": 0.70021725,
      "learning_rate": 8.91269151037425e-07,
      "loss": 0.72134668,
      "num_input_tokens_seen": 249984565,
      "step": 11583,
      "time_per_iteration": 2.5314180850982666
    },
    {
      "auxiliary_loss_clip": 0.01078262,
      "auxiliary_loss_mlp": 0.01034699,
      "balance_loss_clip": 1.03958964,
      "balance_loss_mlp": 1.02225995,
      "epoch": 0.6964677589057569,
      "flos": 19937317466880.0,
      "grad_norm": 1.6615103163418148,
      "language_loss": 0.82128561,
      "learning_rate": 8.909450334717301e-07,
      "loss": 0.84241521,
      "num_input_tokens_seen": 250004235,
      "step": 11584,
      "time_per_iteration": 3.9758424758911133
    },
    {
      "auxiliary_loss_clip": 0.01054117,
      "auxiliary_loss_mlp": 0.01036788,
      "balance_loss_clip": 1.03500772,
      "balance_loss_mlp": 1.02348447,
      "epoch": 0.6965278821584248,
      "flos": 22784064001920.0,
      "grad_norm": 2.1972421121543135,
      "language_loss": 0.79556167,
      "learning_rate": 8.906209579615107e-07,
      "loss": 0.81647068,
      "num_input_tokens_seen": 250017645,
      "step": 11585,
      "time_per_iteration": 2.6209864616394043
    },
    {
      "auxiliary_loss_clip": 0.01105688,
      "auxiliary_loss_mlp": 0.01028836,
      "balance_loss_clip": 1.03845549,
      "balance_loss_mlp": 1.01750565,
      "epoch": 0.6965880054110928,
      "flos": 20047563285120.0,
      "grad_norm": 1.6149166186203592,
      "language_loss": 0.77747899,
      "learning_rate": 8.90296924519055e-07,
      "loss": 0.79882419,
      "num_input_tokens_seen": 250037640,
      "step": 11586,
      "time_per_iteration": 2.464665174484253
    },
    {
      "auxiliary_loss_clip": 0.01092575,
      "auxiliary_loss_mlp": 0.01031165,
      "balance_loss_clip": 1.03671098,
      "balance_loss_mlp": 1.01974487,
      "epoch": 0.6966481286637607,
      "flos": 21908238681600.0,
      "grad_norm": 1.6118720034053964,
      "language_loss": 0.78739834,
      "learning_rate": 8.899729331566519e-07,
      "loss": 0.80863577,
      "num_input_tokens_seen": 250056490,
      "step": 11587,
      "time_per_iteration": 2.485600233078003
    },
    {
      "auxiliary_loss_clip": 0.01084437,
      "auxiliary_loss_mlp": 0.01034282,
      "balance_loss_clip": 1.03889596,
      "balance_loss_mlp": 1.02200902,
      "epoch": 0.6967082519164287,
      "flos": 15633172506240.0,
      "grad_norm": 1.871355851597638,
      "language_loss": 0.72749972,
      "learning_rate": 8.896489838865857e-07,
      "loss": 0.74868691,
      "num_input_tokens_seen": 250074285,
      "step": 11588,
      "time_per_iteration": 4.004549026489258
    },
    {
      "auxiliary_loss_clip": 0.01081705,
      "auxiliary_loss_mlp": 0.01026241,
      "balance_loss_clip": 1.03766048,
      "balance_loss_mlp": 1.01523161,
      "epoch": 0.6967683751690966,
      "flos": 24024598064640.0,
      "grad_norm": 1.7589821407936792,
      "language_loss": 0.75029171,
      "learning_rate": 8.893250767211413e-07,
      "loss": 0.77137113,
      "num_input_tokens_seen": 250093350,
      "step": 11589,
      "time_per_iteration": 4.230475187301636
    },
    {
      "auxiliary_loss_clip": 0.01087643,
      "auxiliary_loss_mlp": 0.01033031,
      "balance_loss_clip": 1.03818691,
      "balance_loss_mlp": 1.02108061,
      "epoch": 0.6968284984217646,
      "flos": 31024700265600.0,
      "grad_norm": 1.9310478595577625,
      "language_loss": 0.63802624,
      "learning_rate": 8.890012116726012e-07,
      "loss": 0.65923297,
      "num_input_tokens_seen": 250114170,
      "step": 11590,
      "time_per_iteration": 2.6473636627197266
    },
    {
      "auxiliary_loss_clip": 0.0100175,
      "auxiliary_loss_mlp": 0.01006028,
      "balance_loss_clip": 1.02569675,
      "balance_loss_mlp": 1.00447798,
      "epoch": 0.6968886216744326,
      "flos": 67622990002560.0,
      "grad_norm": 0.755309384811812,
      "language_loss": 0.61233872,
      "learning_rate": 8.88677388753248e-07,
      "loss": 0.63241649,
      "num_input_tokens_seen": 250178250,
      "step": 11591,
      "time_per_iteration": 3.332343578338623
    },
    {
      "auxiliary_loss_clip": 0.01061291,
      "auxiliary_loss_mlp": 0.00784282,
      "balance_loss_clip": 1.04065645,
      "balance_loss_mlp": 1.01007962,
      "epoch": 0.6969487449271006,
      "flos": 24863686750080.0,
      "grad_norm": 1.4878111979042927,
      "language_loss": 0.69142789,
      "learning_rate": 8.883536079753582e-07,
      "loss": 0.70988357,
      "num_input_tokens_seen": 250198420,
      "step": 11592,
      "time_per_iteration": 2.696342945098877
    },
    {
      "auxiliary_loss_clip": 0.01074144,
      "auxiliary_loss_mlp": 0.0103007,
      "balance_loss_clip": 1.03810918,
      "balance_loss_mlp": 1.01860237,
      "epoch": 0.6970088681797685,
      "flos": 28767858791040.0,
      "grad_norm": 2.1586066381260367,
      "language_loss": 0.61897862,
      "learning_rate": 8.880298693512109e-07,
      "loss": 0.64002073,
      "num_input_tokens_seen": 250220650,
      "step": 11593,
      "time_per_iteration": 4.039926052093506
    },
    {
      "auxiliary_loss_clip": 0.01082229,
      "auxiliary_loss_mlp": 0.01026232,
      "balance_loss_clip": 1.03727734,
      "balance_loss_mlp": 1.01521659,
      "epoch": 0.6970689914324365,
      "flos": 27308556944640.0,
      "grad_norm": 1.6113703907142711,
      "language_loss": 0.54399586,
      "learning_rate": 8.877061728930832e-07,
      "loss": 0.56508052,
      "num_input_tokens_seen": 250241750,
      "step": 11594,
      "time_per_iteration": 2.5867154598236084
    },
    {
      "auxiliary_loss_clip": 0.01096877,
      "auxiliary_loss_mlp": 0.01029105,
      "balance_loss_clip": 1.03802431,
      "balance_loss_mlp": 1.01773202,
      "epoch": 0.6971291146851044,
      "flos": 19136258305920.0,
      "grad_norm": 1.9689699017760949,
      "language_loss": 0.76842809,
      "learning_rate": 8.87382518613248e-07,
      "loss": 0.78968793,
      "num_input_tokens_seen": 250259445,
      "step": 11595,
      "time_per_iteration": 2.4944686889648438
    },
    {
      "auxiliary_loss_clip": 0.01088432,
      "auxiliary_loss_mlp": 0.00784121,
      "balance_loss_clip": 1.04163313,
      "balance_loss_mlp": 1.00978994,
      "epoch": 0.6971892379377724,
      "flos": 14610508387200.0,
      "grad_norm": 2.360032110718637,
      "language_loss": 0.71422029,
      "learning_rate": 8.870589065239793e-07,
      "loss": 0.7329458,
      "num_input_tokens_seen": 250275640,
      "step": 11596,
      "time_per_iteration": 2.513915538787842
    },
    {
      "auxiliary_loss_clip": 0.01111856,
      "auxiliary_loss_mlp": 0.01032576,
      "balance_loss_clip": 1.04144597,
      "balance_loss_mlp": 1.02019596,
      "epoch": 0.6972493611904405,
      "flos": 22307457415680.0,
      "grad_norm": 1.614574847703069,
      "language_loss": 0.76117587,
      "learning_rate": 8.867353366375492e-07,
      "loss": 0.78262019,
      "num_input_tokens_seen": 250296435,
      "step": 11597,
      "time_per_iteration": 2.495518684387207
    },
    {
      "auxiliary_loss_clip": 0.01098346,
      "auxiliary_loss_mlp": 0.01035012,
      "balance_loss_clip": 1.03905272,
      "balance_loss_mlp": 1.02273369,
      "epoch": 0.6973094844431084,
      "flos": 17420374632960.0,
      "grad_norm": 1.7916977176158162,
      "language_loss": 0.75067943,
      "learning_rate": 8.864118089662267e-07,
      "loss": 0.77201301,
      "num_input_tokens_seen": 250314035,
      "step": 11598,
      "time_per_iteration": 2.4761149883270264
    },
    {
      "auxiliary_loss_clip": 0.01092342,
      "auxiliary_loss_mlp": 0.01035458,
      "balance_loss_clip": 1.03887093,
      "balance_loss_mlp": 1.0220468,
      "epoch": 0.6973696076957764,
      "flos": 27235370983680.0,
      "grad_norm": 2.004033220693446,
      "language_loss": 0.8987689,
      "learning_rate": 8.860883235222791e-07,
      "loss": 0.92004699,
      "num_input_tokens_seen": 250332995,
      "step": 11599,
      "time_per_iteration": 2.5988433361053467
    },
    {
      "auxiliary_loss_clip": 0.01103858,
      "auxiliary_loss_mlp": 0.01038963,
      "balance_loss_clip": 1.0388124,
      "balance_loss_mlp": 1.02531934,
      "epoch": 0.6974297309484443,
      "flos": 22018089450240.0,
      "grad_norm": 2.0039679646590995,
      "language_loss": 0.69956863,
      "learning_rate": 8.85764880317974e-07,
      "loss": 0.72099686,
      "num_input_tokens_seen": 250352120,
      "step": 11600,
      "time_per_iteration": 2.5185418128967285
    },
    {
      "auxiliary_loss_clip": 0.01067837,
      "auxiliary_loss_mlp": 0.01033322,
      "balance_loss_clip": 1.03533983,
      "balance_loss_mlp": 1.02106142,
      "epoch": 0.6974898542011123,
      "flos": 28366449327360.0,
      "grad_norm": 1.8179634263588829,
      "language_loss": 0.76815355,
      "learning_rate": 8.854414793655771e-07,
      "loss": 0.78916514,
      "num_input_tokens_seen": 250371705,
      "step": 11601,
      "time_per_iteration": 2.648249864578247
    },
    {
      "auxiliary_loss_clip": 0.01090894,
      "auxiliary_loss_mlp": 0.00781846,
      "balance_loss_clip": 1.0363965,
      "balance_loss_mlp": 1.00815308,
      "epoch": 0.6975499774537802,
      "flos": 15232050351360.0,
      "grad_norm": 1.7901274720395772,
      "language_loss": 0.72351599,
      "learning_rate": 8.851181206773508e-07,
      "loss": 0.74224341,
      "num_input_tokens_seen": 250390485,
      "step": 11602,
      "time_per_iteration": 2.4818081855773926
    },
    {
      "auxiliary_loss_clip": 0.01087237,
      "auxiliary_loss_mlp": 0.00783236,
      "balance_loss_clip": 1.0370779,
      "balance_loss_mlp": 1.0086174,
      "epoch": 0.6976101007064482,
      "flos": 22157422306560.0,
      "grad_norm": 1.9797595866147253,
      "language_loss": 0.76548946,
      "learning_rate": 8.847948042655567e-07,
      "loss": 0.78419423,
      "num_input_tokens_seen": 250407020,
      "step": 11603,
      "time_per_iteration": 2.5223965644836426
    },
    {
      "auxiliary_loss_clip": 0.01061024,
      "auxiliary_loss_mlp": 0.01032046,
      "balance_loss_clip": 1.03567362,
      "balance_loss_mlp": 1.01982713,
      "epoch": 0.6976702239591162,
      "flos": 22273522041600.0,
      "grad_norm": 1.4872085981213494,
      "language_loss": 0.6234889,
      "learning_rate": 8.844715301424557e-07,
      "loss": 0.64441955,
      "num_input_tokens_seen": 250425880,
      "step": 11604,
      "time_per_iteration": 2.587207078933716
    },
    {
      "auxiliary_loss_clip": 0.0110062,
      "auxiliary_loss_mlp": 0.01032298,
      "balance_loss_clip": 1.03902102,
      "balance_loss_mlp": 1.01891124,
      "epoch": 0.6977303472117842,
      "flos": 25848608653440.0,
      "grad_norm": 2.70760858262613,
      "language_loss": 0.81257117,
      "learning_rate": 8.841482983203057e-07,
      "loss": 0.83390039,
      "num_input_tokens_seen": 250442925,
      "step": 11605,
      "time_per_iteration": 2.523991346359253
    },
    {
      "auxiliary_loss_clip": 0.01096993,
      "auxiliary_loss_mlp": 0.01034336,
      "balance_loss_clip": 1.03790116,
      "balance_loss_mlp": 1.02259374,
      "epoch": 0.6977904704644521,
      "flos": 20959586536320.0,
      "grad_norm": 1.827302283519282,
      "language_loss": 0.703269,
      "learning_rate": 8.838251088113638e-07,
      "loss": 0.72458231,
      "num_input_tokens_seen": 250461220,
      "step": 11606,
      "time_per_iteration": 2.489455461502075
    },
    {
      "auxiliary_loss_clip": 0.01088456,
      "auxiliary_loss_mlp": 0.01032003,
      "balance_loss_clip": 1.03850365,
      "balance_loss_mlp": 1.01974845,
      "epoch": 0.6978505937171201,
      "flos": 22055041566720.0,
      "grad_norm": 1.9199621148293204,
      "language_loss": 0.82697958,
      "learning_rate": 8.835019616278856e-07,
      "loss": 0.84818423,
      "num_input_tokens_seen": 250480975,
      "step": 11607,
      "time_per_iteration": 2.5460731983184814
    },
    {
      "auxiliary_loss_clip": 0.01089748,
      "auxiliary_loss_mlp": 0.01033513,
      "balance_loss_clip": 1.04107738,
      "balance_loss_mlp": 1.02070999,
      "epoch": 0.697910716969788,
      "flos": 20043720529920.0,
      "grad_norm": 1.9845516772369873,
      "language_loss": 0.7886852,
      "learning_rate": 8.831788567821265e-07,
      "loss": 0.80991787,
      "num_input_tokens_seen": 250497980,
      "step": 11608,
      "time_per_iteration": 2.5225706100463867
    },
    {
      "auxiliary_loss_clip": 0.01087328,
      "auxiliary_loss_mlp": 0.01030459,
      "balance_loss_clip": 1.03504729,
      "balance_loss_mlp": 1.0183531,
      "epoch": 0.697970840222456,
      "flos": 15888245961600.0,
      "grad_norm": 1.9927316823391503,
      "language_loss": 0.89995676,
      "learning_rate": 8.828557942863357e-07,
      "loss": 0.92113459,
      "num_input_tokens_seen": 250511910,
      "step": 11609,
      "time_per_iteration": 2.4873039722442627
    },
    {
      "auxiliary_loss_clip": 0.01074088,
      "auxiliary_loss_mlp": 0.01029503,
      "balance_loss_clip": 1.03751016,
      "balance_loss_mlp": 1.0172956,
      "epoch": 0.698030963475124,
      "flos": 21215629658880.0,
      "grad_norm": 1.7984799746843998,
      "language_loss": 0.6438781,
      "learning_rate": 8.82532774152765e-07,
      "loss": 0.66491401,
      "num_input_tokens_seen": 250531090,
      "step": 11610,
      "time_per_iteration": 2.5700743198394775
    },
    {
      "auxiliary_loss_clip": 0.01074583,
      "auxiliary_loss_mlp": 0.01032269,
      "balance_loss_clip": 1.03653002,
      "balance_loss_mlp": 1.02085459,
      "epoch": 0.698091086727792,
      "flos": 33759728524800.0,
      "grad_norm": 1.7783151681090366,
      "language_loss": 0.8446793,
      "learning_rate": 8.822097963936643e-07,
      "loss": 0.86574781,
      "num_input_tokens_seen": 250551565,
      "step": 11611,
      "time_per_iteration": 2.6626999378204346
    },
    {
      "auxiliary_loss_clip": 0.01099727,
      "auxiliary_loss_mlp": 0.01035132,
      "balance_loss_clip": 1.03826952,
      "balance_loss_mlp": 1.02264535,
      "epoch": 0.69815120998046,
      "flos": 15887850912000.0,
      "grad_norm": 1.97151326710403,
      "language_loss": 0.708148,
      "learning_rate": 8.818868610212793e-07,
      "loss": 0.7294966,
      "num_input_tokens_seen": 250569625,
      "step": 11612,
      "time_per_iteration": 2.478031635284424
    },
    {
      "auxiliary_loss_clip": 0.01089094,
      "auxiliary_loss_mlp": 0.01036026,
      "balance_loss_clip": 1.03435469,
      "balance_loss_mlp": 1.02365208,
      "epoch": 0.6982113332331279,
      "flos": 18947044437120.0,
      "grad_norm": 1.7458639871337391,
      "language_loss": 0.80888253,
      "learning_rate": 8.815639680478573e-07,
      "loss": 0.83013374,
      "num_input_tokens_seen": 250586960,
      "step": 11613,
      "time_per_iteration": 2.50899076461792
    },
    {
      "auxiliary_loss_clip": 0.01096035,
      "auxiliary_loss_mlp": 0.01031793,
      "balance_loss_clip": 1.03709435,
      "balance_loss_mlp": 1.020504,
      "epoch": 0.6982714564857959,
      "flos": 24389594115840.0,
      "grad_norm": 2.0393271133890734,
      "language_loss": 0.75293213,
      "learning_rate": 8.812411174856411e-07,
      "loss": 0.77421045,
      "num_input_tokens_seen": 250605080,
      "step": 11614,
      "time_per_iteration": 2.5243375301361084
    },
    {
      "auxiliary_loss_clip": 0.01043805,
      "auxiliary_loss_mlp": 0.0103306,
      "balance_loss_clip": 1.03833961,
      "balance_loss_mlp": 1.02082944,
      "epoch": 0.6983315797384638,
      "flos": 20083725302400.0,
      "grad_norm": 2.4297123627585506,
      "language_loss": 0.77038515,
      "learning_rate": 8.809183093468746e-07,
      "loss": 0.79115379,
      "num_input_tokens_seen": 250623965,
      "step": 11615,
      "time_per_iteration": 2.6455461978912354
    },
    {
      "auxiliary_loss_clip": 0.01077781,
      "auxiliary_loss_mlp": 0.01028352,
      "balance_loss_clip": 1.03699136,
      "balance_loss_mlp": 1.01662147,
      "epoch": 0.6983917029911318,
      "flos": 13512431664000.0,
      "grad_norm": 2.0366886690334507,
      "language_loss": 0.73017669,
      "learning_rate": 8.80595543643797e-07,
      "loss": 0.75123811,
      "num_input_tokens_seen": 250640675,
      "step": 11616,
      "time_per_iteration": 2.4994442462921143
    },
    {
      "auxiliary_loss_clip": 0.01108978,
      "auxiliary_loss_mlp": 0.01035114,
      "balance_loss_clip": 1.04022694,
      "balance_loss_mlp": 1.02301383,
      "epoch": 0.6984518262437998,
      "flos": 22018412672640.0,
      "grad_norm": 1.5659592354012795,
      "language_loss": 0.84289134,
      "learning_rate": 8.802728203886487e-07,
      "loss": 0.86433226,
      "num_input_tokens_seen": 250660295,
      "step": 11617,
      "time_per_iteration": 2.5050692558288574
    },
    {
      "auxiliary_loss_clip": 0.01070137,
      "auxiliary_loss_mlp": 0.01040555,
      "balance_loss_clip": 1.03556848,
      "balance_loss_mlp": 1.02777004,
      "epoch": 0.6985119494964678,
      "flos": 18770615809920.0,
      "grad_norm": 2.334105077529399,
      "language_loss": 0.59075326,
      "learning_rate": 8.799501395936682e-07,
      "loss": 0.61186016,
      "num_input_tokens_seen": 250678155,
      "step": 11618,
      "time_per_iteration": 2.559399127960205
    },
    {
      "auxiliary_loss_clip": 0.01084384,
      "auxiliary_loss_mlp": 0.01034218,
      "balance_loss_clip": 1.03866911,
      "balance_loss_mlp": 1.02270818,
      "epoch": 0.6985720727491357,
      "flos": 22382834106240.0,
      "grad_norm": 1.744411002999002,
      "language_loss": 0.83011043,
      "learning_rate": 8.796275012710903e-07,
      "loss": 0.85129642,
      "num_input_tokens_seen": 250697230,
      "step": 11619,
      "time_per_iteration": 2.5530714988708496
    },
    {
      "auxiliary_loss_clip": 0.01091857,
      "auxiliary_loss_mlp": 0.01030743,
      "balance_loss_clip": 1.03619719,
      "balance_loss_mlp": 1.02008021,
      "epoch": 0.6986321960018037,
      "flos": 39567884785920.0,
      "grad_norm": 4.131222918843272,
      "language_loss": 0.67416215,
      "learning_rate": 8.793049054331494e-07,
      "loss": 0.69538814,
      "num_input_tokens_seen": 250719865,
      "step": 11620,
      "time_per_iteration": 2.6566836833953857
    },
    {
      "auxiliary_loss_clip": 0.01056215,
      "auxiliary_loss_mlp": 0.01032177,
      "balance_loss_clip": 1.03419876,
      "balance_loss_mlp": 1.01934993,
      "epoch": 0.6986923192544716,
      "flos": 17967725055360.0,
      "grad_norm": 3.386079483907327,
      "language_loss": 0.72860217,
      "learning_rate": 8.789823520920794e-07,
      "loss": 0.74948603,
      "num_input_tokens_seen": 250736565,
      "step": 11621,
      "time_per_iteration": 2.6057660579681396
    },
    {
      "auxiliary_loss_clip": 0.010628,
      "auxiliary_loss_mlp": 0.0103637,
      "balance_loss_clip": 1.03512406,
      "balance_loss_mlp": 1.02367437,
      "epoch": 0.6987524425071396,
      "flos": 25594325297280.0,
      "grad_norm": 1.635852918600558,
      "language_loss": 0.68306255,
      "learning_rate": 8.7865984126011e-07,
      "loss": 0.70405424,
      "num_input_tokens_seen": 250757235,
      "step": 11622,
      "time_per_iteration": 4.02208948135376
    },
    {
      "auxiliary_loss_clip": 0.01039789,
      "auxiliary_loss_mlp": 0.01025178,
      "balance_loss_clip": 1.03490174,
      "balance_loss_mlp": 1.01380539,
      "epoch": 0.6988125657598077,
      "flos": 17530081747200.0,
      "grad_norm": 1.8856548991659343,
      "language_loss": 0.6310088,
      "learning_rate": 8.783373729494721e-07,
      "loss": 0.65165842,
      "num_input_tokens_seen": 250775585,
      "step": 11623,
      "time_per_iteration": 2.5960323810577393
    },
    {
      "auxiliary_loss_clip": 0.01110971,
      "auxiliary_loss_mlp": 0.01029229,
      "balance_loss_clip": 1.03726506,
      "balance_loss_mlp": 1.01662874,
      "epoch": 0.6988726890124756,
      "flos": 39165721136640.0,
      "grad_norm": 1.9906199048573243,
      "language_loss": 0.60603845,
      "learning_rate": 8.780149471723932e-07,
      "loss": 0.62744045,
      "num_input_tokens_seen": 250795725,
      "step": 11624,
      "time_per_iteration": 2.6303582191467285
    },
    {
      "auxiliary_loss_clip": 0.01097084,
      "auxiliary_loss_mlp": 0.01040616,
      "balance_loss_clip": 1.03489172,
      "balance_loss_mlp": 1.02775919,
      "epoch": 0.6989328122651436,
      "flos": 20193468330240.0,
      "grad_norm": 1.6049260550650024,
      "language_loss": 0.78129572,
      "learning_rate": 8.776925639411017e-07,
      "loss": 0.80267268,
      "num_input_tokens_seen": 250814555,
      "step": 11625,
      "time_per_iteration": 2.4975905418395996
    },
    {
      "auxiliary_loss_clip": 0.01071339,
      "auxiliary_loss_mlp": 0.01034528,
      "balance_loss_clip": 1.03620565,
      "balance_loss_mlp": 1.02314305,
      "epoch": 0.6989929355178115,
      "flos": 21834873152640.0,
      "grad_norm": 1.8297947250522166,
      "language_loss": 0.65934289,
      "learning_rate": 8.773702232678188e-07,
      "loss": 0.68040156,
      "num_input_tokens_seen": 250833105,
      "step": 11626,
      "time_per_iteration": 2.571974515914917
    },
    {
      "auxiliary_loss_clip": 0.01089188,
      "auxiliary_loss_mlp": 0.00784795,
      "balance_loss_clip": 1.03971827,
      "balance_loss_mlp": 1.01098335,
      "epoch": 0.6990530587704795,
      "flos": 26322880855680.0,
      "grad_norm": 1.6576301174758237,
      "language_loss": 0.69948518,
      "learning_rate": 8.770479251647697e-07,
      "loss": 0.71822494,
      "num_input_tokens_seen": 250852570,
      "step": 11627,
      "time_per_iteration": 3.964367151260376
    },
    {
      "auxiliary_loss_clip": 0.01105442,
      "auxiliary_loss_mlp": 0.01029193,
      "balance_loss_clip": 1.03888559,
      "balance_loss_mlp": 1.01857734,
      "epoch": 0.6991131820231474,
      "flos": 19828975069440.0,
      "grad_norm": 1.6714802433606066,
      "language_loss": 0.62302589,
      "learning_rate": 8.767256696441768e-07,
      "loss": 0.64437222,
      "num_input_tokens_seen": 250870500,
      "step": 11628,
      "time_per_iteration": 3.8639156818389893
    },
    {
      "auxiliary_loss_clip": 0.01098835,
      "auxiliary_loss_mlp": 0.01038672,
      "balance_loss_clip": 1.03770101,
      "balance_loss_mlp": 1.02633405,
      "epoch": 0.6991733052758154,
      "flos": 33984817102080.0,
      "grad_norm": 2.075455389954821,
      "language_loss": 0.68541098,
      "learning_rate": 8.764034567182581e-07,
      "loss": 0.70678598,
      "num_input_tokens_seen": 250892745,
      "step": 11629,
      "time_per_iteration": 2.602429151535034
    },
    {
      "auxiliary_loss_clip": 0.01106748,
      "auxiliary_loss_mlp": 0.01032233,
      "balance_loss_clip": 1.03843009,
      "balance_loss_mlp": 1.01967442,
      "epoch": 0.6992334285284834,
      "flos": 15633136592640.0,
      "grad_norm": 1.6571688404362803,
      "language_loss": 0.72301197,
      "learning_rate": 8.760812863992337e-07,
      "loss": 0.74440175,
      "num_input_tokens_seen": 250910225,
      "step": 11630,
      "time_per_iteration": 2.4353671073913574
    },
    {
      "auxiliary_loss_clip": 0.01107934,
      "auxiliary_loss_mlp": 0.01036637,
      "balance_loss_clip": 1.03946304,
      "balance_loss_mlp": 1.02487659,
      "epoch": 0.6992935517811514,
      "flos": 21726279360000.0,
      "grad_norm": 1.6046770057801882,
      "language_loss": 0.7403295,
      "learning_rate": 8.757591586993196e-07,
      "loss": 0.76177526,
      "num_input_tokens_seen": 250929715,
      "step": 11631,
      "time_per_iteration": 2.4698269367218018
    },
    {
      "auxiliary_loss_clip": 0.01102011,
      "auxiliary_loss_mlp": 0.01033465,
      "balance_loss_clip": 1.04084122,
      "balance_loss_mlp": 1.02035189,
      "epoch": 0.6993536750338193,
      "flos": 20115254465280.0,
      "grad_norm": 2.0372618385698607,
      "language_loss": 0.89238816,
      "learning_rate": 8.7543707363073e-07,
      "loss": 0.9137429,
      "num_input_tokens_seen": 250944230,
      "step": 11632,
      "time_per_iteration": 3.8528854846954346
    },
    {
      "auxiliary_loss_clip": 0.01090129,
      "auxiliary_loss_mlp": 0.01039243,
      "balance_loss_clip": 1.04033828,
      "balance_loss_mlp": 1.02744675,
      "epoch": 0.6994137982864873,
      "flos": 22010547594240.0,
      "grad_norm": 1.8449444165268079,
      "language_loss": 0.80279648,
      "learning_rate": 8.751150312056792e-07,
      "loss": 0.82409018,
      "num_input_tokens_seen": 250961865,
      "step": 11633,
      "time_per_iteration": 2.5351340770721436
    },
    {
      "auxiliary_loss_clip": 0.01112415,
      "auxiliary_loss_mlp": 0.01033316,
      "balance_loss_clip": 1.03859079,
      "balance_loss_mlp": 1.02010202,
      "epoch": 0.6994739215391552,
      "flos": 25519020433920.0,
      "grad_norm": 2.0023965752864195,
      "language_loss": 0.67186904,
      "learning_rate": 8.747930314363794e-07,
      "loss": 0.69332635,
      "num_input_tokens_seen": 250982025,
      "step": 11634,
      "time_per_iteration": 2.5170748233795166
    },
    {
      "auxiliary_loss_clip": 0.01017975,
      "auxiliary_loss_mlp": 0.01003663,
      "balance_loss_clip": 1.02216136,
      "balance_loss_mlp": 1.00244069,
      "epoch": 0.6995340447918232,
      "flos": 59128357691520.0,
      "grad_norm": 0.6824868185738696,
      "language_loss": 0.53154755,
      "learning_rate": 8.744710743350412e-07,
      "loss": 0.55176395,
      "num_input_tokens_seen": 251046900,
      "step": 11635,
      "time_per_iteration": 3.3090882301330566
    },
    {
      "auxiliary_loss_clip": 0.01081746,
      "auxiliary_loss_mlp": 0.01032081,
      "balance_loss_clip": 1.03831732,
      "balance_loss_mlp": 1.01969481,
      "epoch": 0.6995941680444913,
      "flos": 17967832796160.0,
      "grad_norm": 1.543247371669746,
      "language_loss": 0.82134199,
      "learning_rate": 8.741491599138726e-07,
      "loss": 0.8424803,
      "num_input_tokens_seen": 251065050,
      "step": 11636,
      "time_per_iteration": 2.525515079498291
    },
    {
      "auxiliary_loss_clip": 0.01109311,
      "auxiliary_loss_mlp": 0.01029057,
      "balance_loss_clip": 1.03839111,
      "balance_loss_mlp": 1.01721358,
      "epoch": 0.6996542912971592,
      "flos": 21980095839360.0,
      "grad_norm": 1.9641883971246108,
      "language_loss": 0.8291384,
      "learning_rate": 8.738272881850801e-07,
      "loss": 0.85052204,
      "num_input_tokens_seen": 251083355,
      "step": 11637,
      "time_per_iteration": 2.4900054931640625
    },
    {
      "auxiliary_loss_clip": 0.01058112,
      "auxiliary_loss_mlp": 0.01034517,
      "balance_loss_clip": 1.03711534,
      "balance_loss_mlp": 1.02257836,
      "epoch": 0.6997144145498272,
      "flos": 11686158518400.0,
      "grad_norm": 1.9820674585982874,
      "language_loss": 0.67660356,
      "learning_rate": 8.735054591608704e-07,
      "loss": 0.69752985,
      "num_input_tokens_seen": 251096420,
      "step": 11638,
      "time_per_iteration": 2.55896258354187
    },
    {
      "auxiliary_loss_clip": 0.01103546,
      "auxiliary_loss_mlp": 0.01036508,
      "balance_loss_clip": 1.03947139,
      "balance_loss_mlp": 1.02331161,
      "epoch": 0.6997745378024951,
      "flos": 29607162958080.0,
      "grad_norm": 1.8776661858203898,
      "language_loss": 0.78070623,
      "learning_rate": 8.731836728534459e-07,
      "loss": 0.8021068,
      "num_input_tokens_seen": 251115410,
      "step": 11639,
      "time_per_iteration": 2.5807676315307617
    },
    {
      "auxiliary_loss_clip": 0.01084789,
      "auxiliary_loss_mlp": 0.01040051,
      "balance_loss_clip": 1.03858697,
      "balance_loss_mlp": 1.02697408,
      "epoch": 0.6998346610551631,
      "flos": 20886616056960.0,
      "grad_norm": 2.441823037389954,
      "language_loss": 0.82474792,
      "learning_rate": 8.728619292750093e-07,
      "loss": 0.84599632,
      "num_input_tokens_seen": 251133530,
      "step": 11640,
      "time_per_iteration": 2.5432631969451904
    },
    {
      "auxiliary_loss_clip": 0.01074823,
      "auxiliary_loss_mlp": 0.01031993,
      "balance_loss_clip": 1.03623557,
      "balance_loss_mlp": 1.02018583,
      "epoch": 0.699894784307831,
      "flos": 27163046949120.0,
      "grad_norm": 2.118550322597399,
      "language_loss": 0.7536903,
      "learning_rate": 8.725402284377619e-07,
      "loss": 0.77475846,
      "num_input_tokens_seen": 251153985,
      "step": 11641,
      "time_per_iteration": 2.618746042251587
    },
    {
      "auxiliary_loss_clip": 0.01085111,
      "auxiliary_loss_mlp": 0.01026572,
      "balance_loss_clip": 1.03744984,
      "balance_loss_mlp": 1.01359606,
      "epoch": 0.699954907560499,
      "flos": 20923640000640.0,
      "grad_norm": 1.9298636216157266,
      "language_loss": 0.77513671,
      "learning_rate": 8.722185703539022e-07,
      "loss": 0.79625356,
      "num_input_tokens_seen": 251173225,
      "step": 11642,
      "time_per_iteration": 2.5321662425994873
    },
    {
      "auxiliary_loss_clip": 0.01104789,
      "auxiliary_loss_mlp": 0.01034561,
      "balance_loss_clip": 1.03960967,
      "balance_loss_mlp": 1.02089918,
      "epoch": 0.700015030813167,
      "flos": 28657792540800.0,
      "grad_norm": 2.2469556344545776,
      "language_loss": 0.74993038,
      "learning_rate": 8.718969550356266e-07,
      "loss": 0.77132386,
      "num_input_tokens_seen": 251192485,
      "step": 11643,
      "time_per_iteration": 2.5436928272247314
    },
    {
      "auxiliary_loss_clip": 0.0107878,
      "auxiliary_loss_mlp": 0.01027858,
      "balance_loss_clip": 1.04129004,
      "balance_loss_mlp": 1.01531088,
      "epoch": 0.700075154065835,
      "flos": 29205286617600.0,
      "grad_norm": 1.5583622312659202,
      "language_loss": 0.60165536,
      "learning_rate": 8.715753824951315e-07,
      "loss": 0.62272179,
      "num_input_tokens_seen": 251214965,
      "step": 11644,
      "time_per_iteration": 2.6340270042419434
    },
    {
      "auxiliary_loss_clip": 0.01096185,
      "auxiliary_loss_mlp": 0.01028727,
      "balance_loss_clip": 1.0373342,
      "balance_loss_mlp": 1.0169189,
      "epoch": 0.7001352773185029,
      "flos": 23112431159040.0,
      "grad_norm": 1.6306492229985121,
      "language_loss": 0.81637901,
      "learning_rate": 8.712538527446119e-07,
      "loss": 0.83762813,
      "num_input_tokens_seen": 251234500,
      "step": 11645,
      "time_per_iteration": 2.4953136444091797
    },
    {
      "auxiliary_loss_clip": 0.01097437,
      "auxiliary_loss_mlp": 0.01030538,
      "balance_loss_clip": 1.03786159,
      "balance_loss_mlp": 1.0184617,
      "epoch": 0.7001954005711709,
      "flos": 21322858734720.0,
      "grad_norm": 2.384354512229564,
      "language_loss": 0.68038946,
      "learning_rate": 8.709323657962584e-07,
      "loss": 0.70166922,
      "num_input_tokens_seen": 251254360,
      "step": 11646,
      "time_per_iteration": 2.583320379257202
    },
    {
      "auxiliary_loss_clip": 0.01095032,
      "auxiliary_loss_mlp": 0.01042035,
      "balance_loss_clip": 1.03914225,
      "balance_loss_mlp": 1.02916622,
      "epoch": 0.7002555238238388,
      "flos": 24535822383360.0,
      "grad_norm": 1.5193886731794732,
      "language_loss": 0.71137947,
      "learning_rate": 8.706109216622635e-07,
      "loss": 0.73275012,
      "num_input_tokens_seen": 251274790,
      "step": 11647,
      "time_per_iteration": 2.6119205951690674
    },
    {
      "auxiliary_loss_clip": 0.01098602,
      "auxiliary_loss_mlp": 0.0103472,
      "balance_loss_clip": 1.03871202,
      "balance_loss_mlp": 1.02173173,
      "epoch": 0.7003156470765068,
      "flos": 39056552726400.0,
      "grad_norm": 2.7755211511756297,
      "language_loss": 0.71841234,
      "learning_rate": 8.702895203548155e-07,
      "loss": 0.73974562,
      "num_input_tokens_seen": 251296275,
      "step": 11648,
      "time_per_iteration": 2.7204642295837402
    },
    {
      "auxiliary_loss_clip": 0.01054779,
      "auxiliary_loss_mlp": 0.01036175,
      "balance_loss_clip": 1.0332917,
      "balance_loss_mlp": 1.02332377,
      "epoch": 0.7003757703291749,
      "flos": 28804092635520.0,
      "grad_norm": 1.9746677341631502,
      "language_loss": 0.77012253,
      "learning_rate": 8.699681618861014e-07,
      "loss": 0.79103208,
      "num_input_tokens_seen": 251317375,
      "step": 11649,
      "time_per_iteration": 2.651092290878296
    },
    {
      "auxiliary_loss_clip": 0.01085065,
      "auxiliary_loss_mlp": 0.01030477,
      "balance_loss_clip": 1.03721607,
      "balance_loss_mlp": 1.01865149,
      "epoch": 0.7004358935818428,
      "flos": 15953854152960.0,
      "grad_norm": 1.6312078217325998,
      "language_loss": 0.7837671,
      "learning_rate": 8.69646846268308e-07,
      "loss": 0.80492258,
      "num_input_tokens_seen": 251333570,
      "step": 11650,
      "time_per_iteration": 2.5105443000793457
    },
    {
      "auxiliary_loss_clip": 0.01085424,
      "auxiliary_loss_mlp": 0.01033539,
      "balance_loss_clip": 1.03742421,
      "balance_loss_mlp": 1.02167726,
      "epoch": 0.7004960168345108,
      "flos": 20411984718720.0,
      "grad_norm": 2.090413522621314,
      "language_loss": 0.78522408,
      "learning_rate": 8.693255735136194e-07,
      "loss": 0.80641365,
      "num_input_tokens_seen": 251351070,
      "step": 11651,
      "time_per_iteration": 2.5129969120025635
    },
    {
      "auxiliary_loss_clip": 0.01074062,
      "auxiliary_loss_mlp": 0.01042029,
      "balance_loss_clip": 1.03705049,
      "balance_loss_mlp": 1.02880239,
      "epoch": 0.7005561400871787,
      "flos": 17347547808000.0,
      "grad_norm": 1.699463331329315,
      "language_loss": 0.69598293,
      "learning_rate": 8.690043436342198e-07,
      "loss": 0.71714383,
      "num_input_tokens_seen": 251370005,
      "step": 11652,
      "time_per_iteration": 2.561394214630127
    },
    {
      "auxiliary_loss_clip": 0.01099391,
      "auxiliary_loss_mlp": 0.01029342,
      "balance_loss_clip": 1.03844905,
      "balance_loss_mlp": 1.01701581,
      "epoch": 0.7006162633398467,
      "flos": 25302120157440.0,
      "grad_norm": 1.3816660927355262,
      "language_loss": 0.74408352,
      "learning_rate": 8.686831566422874e-07,
      "loss": 0.76537085,
      "num_input_tokens_seen": 251391210,
      "step": 11653,
      "time_per_iteration": 2.526205062866211
    },
    {
      "auxiliary_loss_clip": 0.01089243,
      "auxiliary_loss_mlp": 0.01031125,
      "balance_loss_clip": 1.03834093,
      "balance_loss_mlp": 1.01783299,
      "epoch": 0.7006763865925146,
      "flos": 20668997508480.0,
      "grad_norm": 1.8296665532719785,
      "language_loss": 0.70239884,
      "learning_rate": 8.68362012550003e-07,
      "loss": 0.72360253,
      "num_input_tokens_seen": 251411505,
      "step": 11654,
      "time_per_iteration": 2.527838945388794
    },
    {
      "auxiliary_loss_clip": 0.01062811,
      "auxiliary_loss_mlp": 0.01033056,
      "balance_loss_clip": 1.0373013,
      "balance_loss_mlp": 1.01972246,
      "epoch": 0.7007365098451827,
      "flos": 20046449963520.0,
      "grad_norm": 2.508125494193808,
      "language_loss": 0.72440171,
      "learning_rate": 8.680409113695453e-07,
      "loss": 0.74536043,
      "num_input_tokens_seen": 251428975,
      "step": 11655,
      "time_per_iteration": 2.5665245056152344
    },
    {
      "auxiliary_loss_clip": 0.0110636,
      "auxiliary_loss_mlp": 0.01039345,
      "balance_loss_clip": 1.04095531,
      "balance_loss_mlp": 1.02532601,
      "epoch": 0.7007966330978506,
      "flos": 20777375819520.0,
      "grad_norm": 2.177742683359031,
      "language_loss": 0.70460486,
      "learning_rate": 8.677198531130889e-07,
      "loss": 0.72606194,
      "num_input_tokens_seen": 251446940,
      "step": 11656,
      "time_per_iteration": 2.507297992706299
    },
    {
      "auxiliary_loss_clip": 0.01061795,
      "auxiliary_loss_mlp": 0.01030597,
      "balance_loss_clip": 1.03479767,
      "balance_loss_mlp": 1.01907587,
      "epoch": 0.7008567563505186,
      "flos": 29638189330560.0,
      "grad_norm": 1.4807869480691012,
      "language_loss": 0.77707064,
      "learning_rate": 8.673988377928092e-07,
      "loss": 0.79799461,
      "num_input_tokens_seen": 251466205,
      "step": 11657,
      "time_per_iteration": 2.6491570472717285
    },
    {
      "auxiliary_loss_clip": 0.01114083,
      "auxiliary_loss_mlp": 0.01038219,
      "balance_loss_clip": 1.04003406,
      "balance_loss_mlp": 1.02424729,
      "epoch": 0.7009168796031865,
      "flos": 17092007475840.0,
      "grad_norm": 2.0914776754255757,
      "language_loss": 0.77994049,
      "learning_rate": 8.670778654208797e-07,
      "loss": 0.80146348,
      "num_input_tokens_seen": 251484820,
      "step": 11658,
      "time_per_iteration": 2.483335018157959
    },
    {
      "auxiliary_loss_clip": 0.01080669,
      "auxiliary_loss_mlp": 0.01031049,
      "balance_loss_clip": 1.03424621,
      "balance_loss_mlp": 1.01876473,
      "epoch": 0.7009770028558545,
      "flos": 20448972748800.0,
      "grad_norm": 2.11980527845079,
      "language_loss": 0.83149624,
      "learning_rate": 8.667569360094713e-07,
      "loss": 0.85261345,
      "num_input_tokens_seen": 251502670,
      "step": 11659,
      "time_per_iteration": 2.5163156986236572
    },
    {
      "auxiliary_loss_clip": 0.01073173,
      "auxiliary_loss_mlp": 0.01027319,
      "balance_loss_clip": 1.03734148,
      "balance_loss_mlp": 1.01537406,
      "epoch": 0.7010371261085224,
      "flos": 19245139407360.0,
      "grad_norm": 1.9684346535858375,
      "language_loss": 0.69493812,
      "learning_rate": 8.664360495707526e-07,
      "loss": 0.7159431,
      "num_input_tokens_seen": 251521630,
      "step": 11660,
      "time_per_iteration": 2.6095449924468994
    },
    {
      "auxiliary_loss_clip": 0.0111057,
      "auxiliary_loss_mlp": 0.01033319,
      "balance_loss_clip": 1.03764462,
      "balance_loss_mlp": 1.0200454,
      "epoch": 0.7010972493611904,
      "flos": 22127581082880.0,
      "grad_norm": 2.273328612861046,
      "language_loss": 0.8073343,
      "learning_rate": 8.661152061168924e-07,
      "loss": 0.8287732,
      "num_input_tokens_seen": 251540105,
      "step": 11661,
      "time_per_iteration": 3.8412692546844482
    },
    {
      "auxiliary_loss_clip": 0.01095854,
      "auxiliary_loss_mlp": 0.01031741,
      "balance_loss_clip": 1.03607786,
      "balance_loss_mlp": 1.02002239,
      "epoch": 0.7011573726138585,
      "flos": 31391132860800.0,
      "grad_norm": 1.5535228028985364,
      "language_loss": 0.78975725,
      "learning_rate": 8.657944056600579e-07,
      "loss": 0.81103319,
      "num_input_tokens_seen": 251560530,
      "step": 11662,
      "time_per_iteration": 2.55916166305542
    },
    {
      "auxiliary_loss_clip": 0.0109587,
      "auxiliary_loss_mlp": 0.01033724,
      "balance_loss_clip": 1.03703821,
      "balance_loss_mlp": 1.01949024,
      "epoch": 0.7012174958665264,
      "flos": 18150582216960.0,
      "grad_norm": 2.0019490148212906,
      "language_loss": 0.83651692,
      "learning_rate": 8.654736482124134e-07,
      "loss": 0.85781288,
      "num_input_tokens_seen": 251577930,
      "step": 11663,
      "time_per_iteration": 2.486645460128784
    },
    {
      "auxiliary_loss_clip": 0.01026993,
      "auxiliary_loss_mlp": 0.00999185,
      "balance_loss_clip": 1.01534092,
      "balance_loss_mlp": 0.99792749,
      "epoch": 0.7012776191191944,
      "flos": 60651256567680.0,
      "grad_norm": 0.8249261101150267,
      "language_loss": 0.53831387,
      "learning_rate": 8.651529337861209e-07,
      "loss": 0.55857563,
      "num_input_tokens_seen": 251638820,
      "step": 11664,
      "time_per_iteration": 3.086350202560425
    },
    {
      "auxiliary_loss_clip": 0.01088866,
      "auxiliary_loss_mlp": 0.01032512,
      "balance_loss_clip": 1.03787446,
      "balance_loss_mlp": 1.01917219,
      "epoch": 0.7013377423718623,
      "flos": 27198598435200.0,
      "grad_norm": 1.908641428435087,
      "language_loss": 0.78739572,
      "learning_rate": 8.64832262393344e-07,
      "loss": 0.80860949,
      "num_input_tokens_seen": 251658070,
      "step": 11665,
      "time_per_iteration": 3.9786338806152344
    },
    {
      "auxiliary_loss_clip": 0.01093263,
      "auxiliary_loss_mlp": 0.01031295,
      "balance_loss_clip": 1.0356977,
      "balance_loss_mlp": 1.01892674,
      "epoch": 0.7013978656245303,
      "flos": 16543543731840.0,
      "grad_norm": 2.069274616445226,
      "language_loss": 0.76798058,
      "learning_rate": 8.645116340462404e-07,
      "loss": 0.78922617,
      "num_input_tokens_seen": 251671575,
      "step": 11666,
      "time_per_iteration": 3.823756694793701
    },
    {
      "auxiliary_loss_clip": 0.01095496,
      "auxiliary_loss_mlp": 0.01031778,
      "balance_loss_clip": 1.03868997,
      "balance_loss_mlp": 1.01955938,
      "epoch": 0.7014579888771982,
      "flos": 23143780753920.0,
      "grad_norm": 1.7658689542971027,
      "language_loss": 0.80991936,
      "learning_rate": 8.641910487569695e-07,
      "loss": 0.83119214,
      "num_input_tokens_seen": 251689350,
      "step": 11667,
      "time_per_iteration": 2.5267486572265625
    },
    {
      "auxiliary_loss_clip": 0.01074155,
      "auxiliary_loss_mlp": 0.01037938,
      "balance_loss_clip": 1.03484488,
      "balance_loss_mlp": 1.02529597,
      "epoch": 0.7015181121298663,
      "flos": 25082095397760.0,
      "grad_norm": 2.2202330773476597,
      "language_loss": 0.65010554,
      "learning_rate": 8.638705065376879e-07,
      "loss": 0.6712265,
      "num_input_tokens_seen": 251704635,
      "step": 11668,
      "time_per_iteration": 2.592855930328369
    },
    {
      "auxiliary_loss_clip": 0.01089247,
      "auxiliary_loss_mlp": 0.01026861,
      "balance_loss_clip": 1.03752637,
      "balance_loss_mlp": 1.01414132,
      "epoch": 0.7015782353825342,
      "flos": 23327894891520.0,
      "grad_norm": 1.8397464349538428,
      "language_loss": 0.76577854,
      "learning_rate": 8.635500074005519e-07,
      "loss": 0.78693962,
      "num_input_tokens_seen": 251723035,
      "step": 11669,
      "time_per_iteration": 2.5408968925476074
    },
    {
      "auxiliary_loss_clip": 0.01021863,
      "auxiliary_loss_mlp": 0.01002902,
      "balance_loss_clip": 1.01728368,
      "balance_loss_mlp": 1.00166798,
      "epoch": 0.7016383586352022,
      "flos": 70397161107840.0,
      "grad_norm": 0.6949909298262044,
      "language_loss": 0.54486209,
      "learning_rate": 8.632295513577122e-07,
      "loss": 0.56510973,
      "num_input_tokens_seen": 251791630,
      "step": 11670,
      "time_per_iteration": 3.255687952041626
    },
    {
      "auxiliary_loss_clip": 0.01082778,
      "auxiliary_loss_mlp": 0.01033905,
      "balance_loss_clip": 1.03899777,
      "balance_loss_mlp": 1.02170396,
      "epoch": 0.7016984818878701,
      "flos": 19792274348160.0,
      "grad_norm": 1.6434150961952654,
      "language_loss": 0.81707668,
      "learning_rate": 8.629091384213218e-07,
      "loss": 0.83824348,
      "num_input_tokens_seen": 251809840,
      "step": 11671,
      "time_per_iteration": 4.001113176345825
    },
    {
      "auxiliary_loss_clip": 0.01103205,
      "auxiliary_loss_mlp": 0.01036079,
      "balance_loss_clip": 1.04144382,
      "balance_loss_mlp": 1.02291822,
      "epoch": 0.7017586051405381,
      "flos": 12896923184640.0,
      "grad_norm": 1.9030743572461828,
      "language_loss": 0.75307971,
      "learning_rate": 8.625887686035313e-07,
      "loss": 0.77447259,
      "num_input_tokens_seen": 251827550,
      "step": 11672,
      "time_per_iteration": 2.480471134185791
    },
    {
      "auxiliary_loss_clip": 0.01091748,
      "auxiliary_loss_mlp": 0.01033033,
      "balance_loss_clip": 1.03535175,
      "balance_loss_mlp": 1.02026534,
      "epoch": 0.701818728393206,
      "flos": 18332828847360.0,
      "grad_norm": 1.5541167687171387,
      "language_loss": 0.8667959,
      "learning_rate": 8.622684419164883e-07,
      "loss": 0.88804376,
      "num_input_tokens_seen": 251844880,
      "step": 11673,
      "time_per_iteration": 2.4832005500793457
    },
    {
      "auxiliary_loss_clip": 0.01092337,
      "auxiliary_loss_mlp": 0.01027328,
      "balance_loss_clip": 1.0356307,
      "balance_loss_mlp": 1.01497769,
      "epoch": 0.701878851645874,
      "flos": 17384212615680.0,
      "grad_norm": 1.8475501416767877,
      "language_loss": 0.73217845,
      "learning_rate": 8.619481583723399e-07,
      "loss": 0.75337505,
      "num_input_tokens_seen": 251861025,
      "step": 11674,
      "time_per_iteration": 2.4956910610198975
    },
    {
      "auxiliary_loss_clip": 0.01099386,
      "auxiliary_loss_mlp": 0.00783174,
      "balance_loss_clip": 1.04171383,
      "balance_loss_mlp": 1.01021194,
      "epoch": 0.701938974898542,
      "flos": 23915501481600.0,
      "grad_norm": 1.6197329247009413,
      "language_loss": 0.71774125,
      "learning_rate": 8.616279179832329e-07,
      "loss": 0.73656684,
      "num_input_tokens_seen": 251880175,
      "step": 11675,
      "time_per_iteration": 2.5394179821014404
    },
    {
      "auxiliary_loss_clip": 0.01072048,
      "auxiliary_loss_mlp": 0.01026924,
      "balance_loss_clip": 1.03800023,
      "balance_loss_mlp": 1.01421034,
      "epoch": 0.70199909815121,
      "flos": 21795586652160.0,
      "grad_norm": 7.93405941555717,
      "language_loss": 0.51435649,
      "learning_rate": 8.613077207613078e-07,
      "loss": 0.53534621,
      "num_input_tokens_seen": 251899005,
      "step": 11676,
      "time_per_iteration": 2.561394453048706
    },
    {
      "auxiliary_loss_clip": 0.01015733,
      "auxiliary_loss_mlp": 0.00762028,
      "balance_loss_clip": 1.01460743,
      "balance_loss_mlp": 0.99878138,
      "epoch": 0.702059221403878,
      "flos": 71715047109120.0,
      "grad_norm": 0.7270107810398855,
      "language_loss": 0.59218633,
      "learning_rate": 8.609875667187079e-07,
      "loss": 0.60996389,
      "num_input_tokens_seen": 251966790,
      "step": 11677,
      "time_per_iteration": 3.2434308528900146
    },
    {
      "auxiliary_loss_clip": 0.01095461,
      "auxiliary_loss_mlp": 0.01033374,
      "balance_loss_clip": 1.03669035,
      "balance_loss_mlp": 1.01965332,
      "epoch": 0.7021193446565459,
      "flos": 28111052649600.0,
      "grad_norm": 1.9263400276737335,
      "language_loss": 0.62514853,
      "learning_rate": 8.606674558675737e-07,
      "loss": 0.64643687,
      "num_input_tokens_seen": 251989315,
      "step": 11678,
      "time_per_iteration": 2.590395927429199
    },
    {
      "auxiliary_loss_clip": 0.01107546,
      "auxiliary_loss_mlp": 0.0103387,
      "balance_loss_clip": 1.03823817,
      "balance_loss_mlp": 1.02162766,
      "epoch": 0.7021794679092139,
      "flos": 22924905229440.0,
      "grad_norm": 7.773691393850617,
      "language_loss": 0.79587078,
      "learning_rate": 8.603473882200444e-07,
      "loss": 0.81728494,
      "num_input_tokens_seen": 252006620,
      "step": 11679,
      "time_per_iteration": 2.4781200885772705
    },
    {
      "auxiliary_loss_clip": 0.01081602,
      "auxiliary_loss_mlp": 0.01040635,
      "balance_loss_clip": 1.03721821,
      "balance_loss_mlp": 1.02743888,
      "epoch": 0.7022395911618818,
      "flos": 18077827219200.0,
      "grad_norm": 2.071870755534729,
      "language_loss": 0.70324391,
      "learning_rate": 8.600273637882567e-07,
      "loss": 0.72446632,
      "num_input_tokens_seen": 252024570,
      "step": 11680,
      "time_per_iteration": 2.516818046569824
    },
    {
      "auxiliary_loss_clip": 0.01070358,
      "auxiliary_loss_mlp": 0.01034743,
      "balance_loss_clip": 1.0345521,
      "balance_loss_mlp": 1.02066422,
      "epoch": 0.7022997144145499,
      "flos": 16034294661120.0,
      "grad_norm": 1.624028475717048,
      "language_loss": 0.74861097,
      "learning_rate": 8.597073825843446e-07,
      "loss": 0.76966202,
      "num_input_tokens_seen": 252042775,
      "step": 11681,
      "time_per_iteration": 2.5234322547912598
    },
    {
      "auxiliary_loss_clip": 0.01089602,
      "auxiliary_loss_mlp": 0.01031362,
      "balance_loss_clip": 1.0375675,
      "balance_loss_mlp": 1.01980436,
      "epoch": 0.7023598376672178,
      "flos": 26468678160000.0,
      "grad_norm": 1.4866793340277638,
      "language_loss": 0.76763314,
      "learning_rate": 8.593874446204434e-07,
      "loss": 0.7888428,
      "num_input_tokens_seen": 252063690,
      "step": 11682,
      "time_per_iteration": 2.5828285217285156
    },
    {
      "auxiliary_loss_clip": 0.01076574,
      "auxiliary_loss_mlp": 0.00784565,
      "balance_loss_clip": 1.03744698,
      "balance_loss_mlp": 1.00867653,
      "epoch": 0.7024199609198858,
      "flos": 17055917285760.0,
      "grad_norm": 1.973403372435661,
      "language_loss": 0.73600554,
      "learning_rate": 8.590675499086841e-07,
      "loss": 0.75461692,
      "num_input_tokens_seen": 252080335,
      "step": 11683,
      "time_per_iteration": 2.53600811958313
    },
    {
      "auxiliary_loss_clip": 0.01076331,
      "auxiliary_loss_mlp": 0.01030963,
      "balance_loss_clip": 1.04144514,
      "balance_loss_mlp": 1.01821363,
      "epoch": 0.7024800841725537,
      "flos": 25849039616640.0,
      "grad_norm": 1.9916999784303735,
      "language_loss": 0.71512711,
      "learning_rate": 8.587476984611976e-07,
      "loss": 0.73620003,
      "num_input_tokens_seen": 252101075,
      "step": 11684,
      "time_per_iteration": 2.617619514465332
    },
    {
      "auxiliary_loss_clip": 0.01100817,
      "auxiliary_loss_mlp": 0.01033217,
      "balance_loss_clip": 1.03923774,
      "balance_loss_mlp": 1.02007461,
      "epoch": 0.7025402074252217,
      "flos": 23513014609920.0,
      "grad_norm": 1.8944183372659253,
      "language_loss": 0.72068226,
      "learning_rate": 8.584278902901128e-07,
      "loss": 0.74202263,
      "num_input_tokens_seen": 252120510,
      "step": 11685,
      "time_per_iteration": 2.562440872192383
    },
    {
      "auxiliary_loss_clip": 0.01095737,
      "auxiliary_loss_mlp": 0.01027164,
      "balance_loss_clip": 1.03666508,
      "balance_loss_mlp": 1.0153327,
      "epoch": 0.7026003306778896,
      "flos": 20150985519360.0,
      "grad_norm": 1.6239189378195842,
      "language_loss": 0.84483153,
      "learning_rate": 8.581081254075582e-07,
      "loss": 0.8660605,
      "num_input_tokens_seen": 252137590,
      "step": 11686,
      "time_per_iteration": 2.5414278507232666
    },
    {
      "auxiliary_loss_clip": 0.01026341,
      "auxiliary_loss_mlp": 0.0100339,
      "balance_loss_clip": 1.01655436,
      "balance_loss_mlp": 1.00209641,
      "epoch": 0.7026604539305576,
      "flos": 64772400712320.0,
      "grad_norm": 0.9773861699920862,
      "language_loss": 0.69951403,
      "learning_rate": 8.577884038256566e-07,
      "loss": 0.71981132,
      "num_input_tokens_seen": 252199830,
      "step": 11687,
      "time_per_iteration": 3.255021572113037
    },
    {
      "auxiliary_loss_clip": 0.01074159,
      "auxiliary_loss_mlp": 0.01028636,
      "balance_loss_clip": 1.03629589,
      "balance_loss_mlp": 1.0155766,
      "epoch": 0.7027205771832256,
      "flos": 21871466133120.0,
      "grad_norm": 2.0138503158810908,
      "language_loss": 0.7737844,
      "learning_rate": 8.574687255565329e-07,
      "loss": 0.79481232,
      "num_input_tokens_seen": 252217200,
      "step": 11688,
      "time_per_iteration": 2.581228017807007
    },
    {
      "auxiliary_loss_clip": 0.01107949,
      "auxiliary_loss_mlp": 0.01033187,
      "balance_loss_clip": 1.03732574,
      "balance_loss_mlp": 1.02083063,
      "epoch": 0.7027807004358936,
      "flos": 23367791923200.0,
      "grad_norm": 2.0928053511229026,
      "language_loss": 0.68636906,
      "learning_rate": 8.571490906123107e-07,
      "loss": 0.70778036,
      "num_input_tokens_seen": 252236105,
      "step": 11689,
      "time_per_iteration": 2.4807448387145996
    },
    {
      "auxiliary_loss_clip": 0.01088128,
      "auxiliary_loss_mlp": 0.01037083,
      "balance_loss_clip": 1.03803945,
      "balance_loss_mlp": 1.02376103,
      "epoch": 0.7028408236885616,
      "flos": 15304266645120.0,
      "grad_norm": 1.8761650720900458,
      "language_loss": 0.79821563,
      "learning_rate": 8.568294990051086e-07,
      "loss": 0.81946778,
      "num_input_tokens_seen": 252253315,
      "step": 11690,
      "time_per_iteration": 2.540001392364502
    },
    {
      "auxiliary_loss_clip": 0.01108758,
      "auxiliary_loss_mlp": 0.01035851,
      "balance_loss_clip": 1.03877854,
      "balance_loss_mlp": 1.02338195,
      "epoch": 0.7029009469412295,
      "flos": 22018197191040.0,
      "grad_norm": 1.6497035030885268,
      "language_loss": 0.76026654,
      "learning_rate": 8.56509950747047e-07,
      "loss": 0.78171265,
      "num_input_tokens_seen": 252272765,
      "step": 11691,
      "time_per_iteration": 2.463104486465454
    },
    {
      "auxiliary_loss_clip": 0.0108517,
      "auxiliary_loss_mlp": 0.01028651,
      "balance_loss_clip": 1.03843546,
      "balance_loss_mlp": 1.01619363,
      "epoch": 0.7029610701938975,
      "flos": 21835519597440.0,
      "grad_norm": 1.9719776900687946,
      "language_loss": 0.81780684,
      "learning_rate": 8.561904458502429e-07,
      "loss": 0.83894509,
      "num_input_tokens_seen": 252290510,
      "step": 11692,
      "time_per_iteration": 2.553443670272827
    },
    {
      "auxiliary_loss_clip": 0.01083435,
      "auxiliary_loss_mlp": 0.01028637,
      "balance_loss_clip": 1.038625,
      "balance_loss_mlp": 1.01615584,
      "epoch": 0.7030211934465654,
      "flos": 19135647774720.0,
      "grad_norm": 2.33299099524304,
      "language_loss": 0.76412767,
      "learning_rate": 8.558709843268111e-07,
      "loss": 0.7852484,
      "num_input_tokens_seen": 252309365,
      "step": 11693,
      "time_per_iteration": 2.5329227447509766
    },
    {
      "auxiliary_loss_clip": 0.01082942,
      "auxiliary_loss_mlp": 0.01034407,
      "balance_loss_clip": 1.03969002,
      "balance_loss_mlp": 1.02198577,
      "epoch": 0.7030813166992335,
      "flos": 38546010766080.0,
      "grad_norm": 1.451107548712621,
      "language_loss": 0.68589211,
      "learning_rate": 8.55551566188866e-07,
      "loss": 0.70706558,
      "num_input_tokens_seen": 252333010,
      "step": 11694,
      "time_per_iteration": 2.671213388442993
    },
    {
      "auxiliary_loss_clip": 0.01108932,
      "auxiliary_loss_mlp": 0.01029666,
      "balance_loss_clip": 1.03763795,
      "balance_loss_mlp": 1.01708341,
      "epoch": 0.7031414399519014,
      "flos": 14720897859840.0,
      "grad_norm": 1.9852122729820922,
      "language_loss": 0.75640815,
      "learning_rate": 8.552321914485203e-07,
      "loss": 0.77779412,
      "num_input_tokens_seen": 252351330,
      "step": 11695,
      "time_per_iteration": 2.4433882236480713
    },
    {
      "auxiliary_loss_clip": 0.01086093,
      "auxiliary_loss_mlp": 0.01034236,
      "balance_loss_clip": 1.03988075,
      "balance_loss_mlp": 1.02103329,
      "epoch": 0.7032015632045694,
      "flos": 14027247342720.0,
      "grad_norm": 1.9603002272708652,
      "language_loss": 0.73376334,
      "learning_rate": 8.549128601178852e-07,
      "loss": 0.75496668,
      "num_input_tokens_seen": 252369580,
      "step": 11696,
      "time_per_iteration": 2.5174925327301025
    },
    {
      "auxiliary_loss_clip": 0.01091487,
      "auxiliary_loss_mlp": 0.01031183,
      "balance_loss_clip": 1.0377785,
      "balance_loss_mlp": 1.0180527,
      "epoch": 0.7032616864572373,
      "flos": 27637175496960.0,
      "grad_norm": 1.5598199007359645,
      "language_loss": 0.75203633,
      "learning_rate": 8.545935722090693e-07,
      "loss": 0.77326298,
      "num_input_tokens_seen": 252390525,
      "step": 11697,
      "time_per_iteration": 2.6158759593963623
    },
    {
      "auxiliary_loss_clip": 0.01066269,
      "auxiliary_loss_mlp": 0.01039776,
      "balance_loss_clip": 1.03967977,
      "balance_loss_mlp": 1.02488685,
      "epoch": 0.7033218097099053,
      "flos": 17967294092160.0,
      "grad_norm": 1.827019582886223,
      "language_loss": 0.80849636,
      "learning_rate": 8.542743277341793e-07,
      "loss": 0.82955682,
      "num_input_tokens_seen": 252407470,
      "step": 11698,
      "time_per_iteration": 2.592144012451172
    },
    {
      "auxiliary_loss_clip": 0.01081467,
      "auxiliary_loss_mlp": 0.01038221,
      "balance_loss_clip": 1.03769171,
      "balance_loss_mlp": 1.02419066,
      "epoch": 0.7033819329625732,
      "flos": 19501721233920.0,
      "grad_norm": 1.3880417450218534,
      "language_loss": 0.8449133,
      "learning_rate": 8.539551267053222e-07,
      "loss": 0.86611021,
      "num_input_tokens_seen": 252427025,
      "step": 11699,
      "time_per_iteration": 3.9582715034484863
    },
    {
      "auxiliary_loss_clip": 0.01094181,
      "auxiliary_loss_mlp": 0.01031779,
      "balance_loss_clip": 1.03830743,
      "balance_loss_mlp": 1.01815975,
      "epoch": 0.7034420562152413,
      "flos": 23987645948160.0,
      "grad_norm": 1.9729835915935523,
      "language_loss": 0.78932315,
      "learning_rate": 8.53635969134601e-07,
      "loss": 0.81058282,
      "num_input_tokens_seen": 252445410,
      "step": 11700,
      "time_per_iteration": 2.5399725437164307
    },
    {
      "auxiliary_loss_clip": 0.01098447,
      "auxiliary_loss_mlp": 0.0102865,
      "balance_loss_clip": 1.03803396,
      "balance_loss_mlp": 1.01548386,
      "epoch": 0.7035021794679092,
      "flos": 35043427756800.0,
      "grad_norm": 1.6551720302184092,
      "language_loss": 0.7440753,
      "learning_rate": 8.533168550341186e-07,
      "loss": 0.76534629,
      "num_input_tokens_seen": 252463905,
      "step": 11701,
      "time_per_iteration": 2.6103675365448
    },
    {
      "auxiliary_loss_clip": 0.01103597,
      "auxiliary_loss_mlp": 0.01032213,
      "balance_loss_clip": 1.03967512,
      "balance_loss_mlp": 1.01829004,
      "epoch": 0.7035623027205772,
      "flos": 10997428164480.0,
      "grad_norm": 5.2130018392353,
      "language_loss": 0.83875316,
      "learning_rate": 8.529977844159769e-07,
      "loss": 0.8601113,
      "num_input_tokens_seen": 252478655,
      "step": 11702,
      "time_per_iteration": 2.452625274658203
    },
    {
      "auxiliary_loss_clip": 0.01109592,
      "auxiliary_loss_mlp": 0.01039368,
      "balance_loss_clip": 1.03842723,
      "balance_loss_mlp": 1.02629089,
      "epoch": 0.7036224259732452,
      "flos": 23623727304960.0,
      "grad_norm": 1.9073911807117558,
      "language_loss": 0.61312211,
      "learning_rate": 8.526787572922738e-07,
      "loss": 0.63461173,
      "num_input_tokens_seen": 252498740,
      "step": 11703,
      "time_per_iteration": 2.5002970695495605
    },
    {
      "auxiliary_loss_clip": 0.01107788,
      "auxiliary_loss_mlp": 0.01030847,
      "balance_loss_clip": 1.03621066,
      "balance_loss_mlp": 1.01782322,
      "epoch": 0.7036825492259131,
      "flos": 31686175175040.0,
      "grad_norm": 2.036441512749634,
      "language_loss": 0.61324269,
      "learning_rate": 8.523597736751067e-07,
      "loss": 0.63462901,
      "num_input_tokens_seen": 252517800,
      "step": 11704,
      "time_per_iteration": 3.936185836791992
    },
    {
      "auxiliary_loss_clip": 0.01089586,
      "auxiliary_loss_mlp": 0.01030082,
      "balance_loss_clip": 1.03697801,
      "balance_loss_mlp": 1.0186851,
      "epoch": 0.7037426724785811,
      "flos": 30192866127360.0,
      "grad_norm": 1.7626100065574815,
      "language_loss": 0.70869005,
      "learning_rate": 8.520408335765719e-07,
      "loss": 0.72988665,
      "num_input_tokens_seen": 252539620,
      "step": 11705,
      "time_per_iteration": 3.9462175369262695
    },
    {
      "auxiliary_loss_clip": 0.01096185,
      "auxiliary_loss_mlp": 0.0103185,
      "balance_loss_clip": 1.03791797,
      "balance_loss_mlp": 1.01966059,
      "epoch": 0.703802795731249,
      "flos": 24311523905280.0,
      "grad_norm": 1.8838781764608052,
      "language_loss": 0.6192103,
      "learning_rate": 8.517219370087645e-07,
      "loss": 0.64049065,
      "num_input_tokens_seen": 252557300,
      "step": 11706,
      "time_per_iteration": 2.537309408187866
    },
    {
      "auxiliary_loss_clip": 0.01101443,
      "auxiliary_loss_mlp": 0.0103074,
      "balance_loss_clip": 1.03947043,
      "balance_loss_mlp": 1.01806164,
      "epoch": 0.7038629189839171,
      "flos": 22528954632960.0,
      "grad_norm": 2.0897908155901432,
      "language_loss": 0.68060553,
      "learning_rate": 8.514030839837756e-07,
      "loss": 0.70192736,
      "num_input_tokens_seen": 252576715,
      "step": 11707,
      "time_per_iteration": 2.495288848876953
    },
    {
      "auxiliary_loss_clip": 0.01105921,
      "auxiliary_loss_mlp": 0.01028658,
      "balance_loss_clip": 1.03768575,
      "balance_loss_mlp": 1.01680267,
      "epoch": 0.703923042236585,
      "flos": 26250484993920.0,
      "grad_norm": 1.9709128433853522,
      "language_loss": 0.76417005,
      "learning_rate": 8.510842745136974e-07,
      "loss": 0.78551579,
      "num_input_tokens_seen": 252596190,
      "step": 11708,
      "time_per_iteration": 2.5281362533569336
    },
    {
      "auxiliary_loss_clip": 0.01087273,
      "auxiliary_loss_mlp": 0.01030137,
      "balance_loss_clip": 1.03888965,
      "balance_loss_mlp": 1.01783419,
      "epoch": 0.703983165489253,
      "flos": 19390254353280.0,
      "grad_norm": 1.7007020195476077,
      "language_loss": 0.72190034,
      "learning_rate": 8.50765508610619e-07,
      "loss": 0.74307442,
      "num_input_tokens_seen": 252613410,
      "step": 11709,
      "time_per_iteration": 3.92002534866333
    },
    {
      "auxiliary_loss_clip": 0.01094312,
      "auxiliary_loss_mlp": 0.01030088,
      "balance_loss_clip": 1.03575802,
      "balance_loss_mlp": 1.01814353,
      "epoch": 0.7040432887419209,
      "flos": 16683630773760.0,
      "grad_norm": 2.049738401494441,
      "language_loss": 0.78564483,
      "learning_rate": 8.504467862866267e-07,
      "loss": 0.80688882,
      "num_input_tokens_seen": 252629150,
      "step": 11710,
      "time_per_iteration": 2.469845771789551
    },
    {
      "auxiliary_loss_clip": 0.01100555,
      "auxiliary_loss_mlp": 0.01035073,
      "balance_loss_clip": 1.03912258,
      "balance_loss_mlp": 1.02189994,
      "epoch": 0.7041034119945889,
      "flos": 21141402203520.0,
      "grad_norm": 1.6663176282873677,
      "language_loss": 0.77239931,
      "learning_rate": 8.501281075538076e-07,
      "loss": 0.79375559,
      "num_input_tokens_seen": 252648225,
      "step": 11711,
      "time_per_iteration": 2.5155739784240723
    },
    {
      "auxiliary_loss_clip": 0.01073004,
      "auxiliary_loss_mlp": 0.01033657,
      "balance_loss_clip": 1.03606319,
      "balance_loss_mlp": 1.02232051,
      "epoch": 0.7041635352472568,
      "flos": 16910299549440.0,
      "grad_norm": 3.5009015602312306,
      "language_loss": 0.73658901,
      "learning_rate": 8.498094724242457e-07,
      "loss": 0.75765562,
      "num_input_tokens_seen": 252665380,
      "step": 11712,
      "time_per_iteration": 2.5495121479034424
    },
    {
      "auxiliary_loss_clip": 0.01006631,
      "auxiliary_loss_mlp": 0.01003699,
      "balance_loss_clip": 1.01611829,
      "balance_loss_mlp": 1.00241113,
      "epoch": 0.7042236584999249,
      "flos": 71681219475840.0,
      "grad_norm": 0.9275569670744976,
      "language_loss": 0.64700055,
      "learning_rate": 8.494908809100247e-07,
      "loss": 0.66710389,
      "num_input_tokens_seen": 252727950,
      "step": 11713,
      "time_per_iteration": 3.2103207111358643
    },
    {
      "auxiliary_loss_clip": 0.01093392,
      "auxiliary_loss_mlp": 0.01029812,
      "balance_loss_clip": 1.03576803,
      "balance_loss_mlp": 1.0179621,
      "epoch": 0.7042837817525928,
      "flos": 28658187590400.0,
      "grad_norm": 2.2262515812528036,
      "language_loss": 0.72744685,
      "learning_rate": 8.49172333023225e-07,
      "loss": 0.74867892,
      "num_input_tokens_seen": 252746770,
      "step": 11714,
      "time_per_iteration": 2.566638708114624
    },
    {
      "auxiliary_loss_clip": 0.01081928,
      "auxiliary_loss_mlp": 0.0078598,
      "balance_loss_clip": 1.03789282,
      "balance_loss_mlp": 1.01079726,
      "epoch": 0.7043439050052608,
      "flos": 19753562465280.0,
      "grad_norm": 1.7394690244270359,
      "language_loss": 0.79933918,
      "learning_rate": 8.488538287759248e-07,
      "loss": 0.8180182,
      "num_input_tokens_seen": 252765610,
      "step": 11715,
      "time_per_iteration": 2.5400609970092773
    },
    {
      "auxiliary_loss_clip": 0.01079747,
      "auxiliary_loss_mlp": 0.01034229,
      "balance_loss_clip": 1.03748035,
      "balance_loss_mlp": 1.02124715,
      "epoch": 0.7044040282579288,
      "flos": 11538529620480.0,
      "grad_norm": 2.4840332909708844,
      "language_loss": 0.71195287,
      "learning_rate": 8.485353681802037e-07,
      "loss": 0.73309267,
      "num_input_tokens_seen": 252781610,
      "step": 11716,
      "time_per_iteration": 2.53765869140625
    },
    {
      "auxiliary_loss_clip": 0.01075572,
      "auxiliary_loss_mlp": 0.0103227,
      "balance_loss_clip": 1.04001689,
      "balance_loss_mlp": 1.01978314,
      "epoch": 0.7044641515105967,
      "flos": 33656126722560.0,
      "grad_norm": 1.8183806938266367,
      "language_loss": 0.66098332,
      "learning_rate": 8.482169512481358e-07,
      "loss": 0.68206179,
      "num_input_tokens_seen": 252800600,
      "step": 11717,
      "time_per_iteration": 2.690755844116211
    },
    {
      "auxiliary_loss_clip": 0.01108715,
      "auxiliary_loss_mlp": 0.01032406,
      "balance_loss_clip": 1.03860378,
      "balance_loss_mlp": 1.02054453,
      "epoch": 0.7045242747632647,
      "flos": 26723859356160.0,
      "grad_norm": 1.4860496105974812,
      "language_loss": 0.74098718,
      "learning_rate": 8.478985779917967e-07,
      "loss": 0.7623983,
      "num_input_tokens_seen": 252822310,
      "step": 11718,
      "time_per_iteration": 2.507378339767456
    },
    {
      "auxiliary_loss_clip": 0.01095746,
      "auxiliary_loss_mlp": 0.01031432,
      "balance_loss_clip": 1.03792727,
      "balance_loss_mlp": 1.01970792,
      "epoch": 0.7045843980159326,
      "flos": 26797655848320.0,
      "grad_norm": 1.940278222231708,
      "language_loss": 0.79780775,
      "learning_rate": 8.475802484232606e-07,
      "loss": 0.81907958,
      "num_input_tokens_seen": 252842355,
      "step": 11719,
      "time_per_iteration": 2.53147292137146
    },
    {
      "auxiliary_loss_clip": 0.01098667,
      "auxiliary_loss_mlp": 0.01038546,
      "balance_loss_clip": 1.03910041,
      "balance_loss_mlp": 1.02576637,
      "epoch": 0.7046445212686007,
      "flos": 41574824363520.0,
      "grad_norm": 1.7784401798791458,
      "language_loss": 0.65803373,
      "learning_rate": 8.472619625545951e-07,
      "loss": 0.67940587,
      "num_input_tokens_seen": 252866785,
      "step": 11720,
      "time_per_iteration": 2.6626977920532227
    },
    {
      "auxiliary_loss_clip": 0.01090455,
      "auxiliary_loss_mlp": 0.0103214,
      "balance_loss_clip": 1.03945065,
      "balance_loss_mlp": 1.01939011,
      "epoch": 0.7047046445212686,
      "flos": 15560166113280.0,
      "grad_norm": 2.593472882636684,
      "language_loss": 0.80224824,
      "learning_rate": 8.46943720397872e-07,
      "loss": 0.82347417,
      "num_input_tokens_seen": 252881870,
      "step": 11721,
      "time_per_iteration": 2.4858405590057373
    },
    {
      "auxiliary_loss_clip": 0.01014677,
      "auxiliary_loss_mlp": 0.01003405,
      "balance_loss_clip": 1.02508867,
      "balance_loss_mlp": 1.00205755,
      "epoch": 0.7047647677739366,
      "flos": 70410269571840.0,
      "grad_norm": 0.7634690391024583,
      "language_loss": 0.64797199,
      "learning_rate": 8.466255219651582e-07,
      "loss": 0.66815281,
      "num_input_tokens_seen": 252951300,
      "step": 11722,
      "time_per_iteration": 3.277519464492798
    },
    {
      "auxiliary_loss_clip": 0.01089366,
      "auxiliary_loss_mlp": 0.01031564,
      "balance_loss_clip": 1.03956354,
      "balance_loss_mlp": 1.01966119,
      "epoch": 0.7048248910266045,
      "flos": 23660032976640.0,
      "grad_norm": 1.6053386007587653,
      "language_loss": 0.65612215,
      "learning_rate": 8.463073672685211e-07,
      "loss": 0.67733145,
      "num_input_tokens_seen": 252971400,
      "step": 11723,
      "time_per_iteration": 2.54923677444458
    },
    {
      "auxiliary_loss_clip": 0.01080984,
      "auxiliary_loss_mlp": 0.01030259,
      "balance_loss_clip": 1.03960061,
      "balance_loss_mlp": 1.01739645,
      "epoch": 0.7048850142792725,
      "flos": 21397158017280.0,
      "grad_norm": 1.9030844166031275,
      "language_loss": 0.80619681,
      "learning_rate": 8.459892563200235e-07,
      "loss": 0.82730925,
      "num_input_tokens_seen": 252989475,
      "step": 11724,
      "time_per_iteration": 2.5637145042419434
    },
    {
      "auxiliary_loss_clip": 0.01096083,
      "auxiliary_loss_mlp": 0.01036562,
      "balance_loss_clip": 1.03602719,
      "balance_loss_mlp": 1.02429521,
      "epoch": 0.7049451375319404,
      "flos": 21648101408640.0,
      "grad_norm": 1.5905229290765486,
      "language_loss": 0.73085213,
      "learning_rate": 8.456711891317296e-07,
      "loss": 0.75217855,
      "num_input_tokens_seen": 253007220,
      "step": 11725,
      "time_per_iteration": 2.51859188079834
    },
    {
      "auxiliary_loss_clip": 0.01064955,
      "auxiliary_loss_mlp": 0.01029019,
      "balance_loss_clip": 1.03554225,
      "balance_loss_mlp": 1.01511919,
      "epoch": 0.7050052607846085,
      "flos": 14866802904960.0,
      "grad_norm": 2.09625082182208,
      "language_loss": 0.78043294,
      "learning_rate": 8.453531657156998e-07,
      "loss": 0.80137277,
      "num_input_tokens_seen": 253025410,
      "step": 11726,
      "time_per_iteration": 2.564790725708008
    },
    {
      "auxiliary_loss_clip": 0.01087936,
      "auxiliary_loss_mlp": 0.01031773,
      "balance_loss_clip": 1.0375284,
      "balance_loss_mlp": 1.01978111,
      "epoch": 0.7050653840372764,
      "flos": 19241763528960.0,
      "grad_norm": 1.7348314044740911,
      "language_loss": 0.70631778,
      "learning_rate": 8.450351860839931e-07,
      "loss": 0.72751486,
      "num_input_tokens_seen": 253043305,
      "step": 11727,
      "time_per_iteration": 2.516083240509033
    },
    {
      "auxiliary_loss_clip": 0.01100409,
      "auxiliary_loss_mlp": 0.00781516,
      "balance_loss_clip": 1.03520465,
      "balance_loss_mlp": 1.00799775,
      "epoch": 0.7051255072899444,
      "flos": 27780422935680.0,
      "grad_norm": 1.645166540566181,
      "language_loss": 0.68970931,
      "learning_rate": 8.44717250248668e-07,
      "loss": 0.70852852,
      "num_input_tokens_seen": 253062790,
      "step": 11728,
      "time_per_iteration": 2.5329418182373047
    },
    {
      "auxiliary_loss_clip": 0.01075751,
      "auxiliary_loss_mlp": 0.0078375,
      "balance_loss_clip": 1.0375483,
      "balance_loss_mlp": 1.00812995,
      "epoch": 0.7051856305426124,
      "flos": 27892033470720.0,
      "grad_norm": 1.886471842304787,
      "language_loss": 0.73166412,
      "learning_rate": 8.443993582217803e-07,
      "loss": 0.75025916,
      "num_input_tokens_seen": 253082055,
      "step": 11729,
      "time_per_iteration": 2.6186745166778564
    },
    {
      "auxiliary_loss_clip": 0.01095193,
      "auxiliary_loss_mlp": 0.01034549,
      "balance_loss_clip": 1.03856039,
      "balance_loss_mlp": 1.02088773,
      "epoch": 0.7052457537952803,
      "flos": 25043563082880.0,
      "grad_norm": 1.678978843799601,
      "language_loss": 0.78334993,
      "learning_rate": 8.440815100153862e-07,
      "loss": 0.80464733,
      "num_input_tokens_seen": 253102575,
      "step": 11730,
      "time_per_iteration": 2.5716500282287598
    },
    {
      "auxiliary_loss_clip": 0.01107593,
      "auxiliary_loss_mlp": 0.010339,
      "balance_loss_clip": 1.03676224,
      "balance_loss_mlp": 1.02141285,
      "epoch": 0.7053058770479483,
      "flos": 21871717528320.0,
      "grad_norm": 2.100382514042608,
      "language_loss": 0.62752306,
      "learning_rate": 8.437637056415359e-07,
      "loss": 0.64893794,
      "num_input_tokens_seen": 253121290,
      "step": 11731,
      "time_per_iteration": 2.479445695877075
    },
    {
      "auxiliary_loss_clip": 0.01059838,
      "auxiliary_loss_mlp": 0.01029828,
      "balance_loss_clip": 1.03720498,
      "balance_loss_mlp": 1.0165062,
      "epoch": 0.7053660003006162,
      "flos": 16398716094720.0,
      "grad_norm": 2.1499882554655665,
      "language_loss": 0.7428602,
      "learning_rate": 8.434459451122815e-07,
      "loss": 0.76375687,
      "num_input_tokens_seen": 253139720,
      "step": 11732,
      "time_per_iteration": 2.580887794494629
    },
    {
      "auxiliary_loss_clip": 0.01095905,
      "auxiliary_loss_mlp": 0.01028627,
      "balance_loss_clip": 1.03817809,
      "balance_loss_mlp": 1.01681876,
      "epoch": 0.7054261235532843,
      "flos": 22711560399360.0,
      "grad_norm": 1.572719301647992,
      "language_loss": 0.71157181,
      "learning_rate": 8.431282284396735e-07,
      "loss": 0.73281717,
      "num_input_tokens_seen": 253160250,
      "step": 11733,
      "time_per_iteration": 2.5111806392669678
    },
    {
      "auxiliary_loss_clip": 0.01069874,
      "auxiliary_loss_mlp": 0.01032028,
      "balance_loss_clip": 1.03671587,
      "balance_loss_mlp": 1.01994014,
      "epoch": 0.7054862468059522,
      "flos": 13589711775360.0,
      "grad_norm": 2.0239998012713856,
      "language_loss": 0.73311627,
      "learning_rate": 8.428105556357583e-07,
      "loss": 0.75413525,
      "num_input_tokens_seen": 253178710,
      "step": 11734,
      "time_per_iteration": 2.566929578781128
    },
    {
      "auxiliary_loss_clip": 0.01074013,
      "auxiliary_loss_mlp": 0.01044834,
      "balance_loss_clip": 1.03641403,
      "balance_loss_mlp": 1.03048158,
      "epoch": 0.7055463700586202,
      "flos": 15880704105600.0,
      "grad_norm": 2.3707147072257184,
      "language_loss": 0.6883027,
      "learning_rate": 8.424929267125829e-07,
      "loss": 0.70949113,
      "num_input_tokens_seen": 253194805,
      "step": 11735,
      "time_per_iteration": 2.520636558532715
    },
    {
      "auxiliary_loss_clip": 0.01081642,
      "auxiliary_loss_mlp": 0.01036385,
      "balance_loss_clip": 1.03612125,
      "balance_loss_mlp": 1.02184129,
      "epoch": 0.7056064933112881,
      "flos": 23076161400960.0,
      "grad_norm": 2.099644841805034,
      "language_loss": 0.72407782,
      "learning_rate": 8.421753416821933e-07,
      "loss": 0.74525809,
      "num_input_tokens_seen": 253213895,
      "step": 11736,
      "time_per_iteration": 2.5788166522979736
    },
    {
      "auxiliary_loss_clip": 0.01084986,
      "auxiliary_loss_mlp": 0.01025215,
      "balance_loss_clip": 1.03614426,
      "balance_loss_mlp": 1.01371121,
      "epoch": 0.7056666165639561,
      "flos": 24057168721920.0,
      "grad_norm": 1.6443687621179437,
      "language_loss": 0.69059753,
      "learning_rate": 8.41857800556629e-07,
      "loss": 0.71169949,
      "num_input_tokens_seen": 253231620,
      "step": 11737,
      "time_per_iteration": 4.0410637855529785
    },
    {
      "auxiliary_loss_clip": 0.01074524,
      "auxiliary_loss_mlp": 0.01032512,
      "balance_loss_clip": 1.03916621,
      "balance_loss_mlp": 1.01990008,
      "epoch": 0.705726739816624,
      "flos": 17493237371520.0,
      "grad_norm": 2.4056486582375793,
      "language_loss": 0.67776322,
      "learning_rate": 8.415403033479332e-07,
      "loss": 0.69883358,
      "num_input_tokens_seen": 253249590,
      "step": 11738,
      "time_per_iteration": 2.5653176307678223
    },
    {
      "auxiliary_loss_clip": 0.01108116,
      "auxiliary_loss_mlp": 0.0103425,
      "balance_loss_clip": 1.03806901,
      "balance_loss_mlp": 1.02125001,
      "epoch": 0.7057868630692921,
      "flos": 51350426472960.0,
      "grad_norm": 1.6822044444000621,
      "language_loss": 0.74834973,
      "learning_rate": 8.41222850068145e-07,
      "loss": 0.76977336,
      "num_input_tokens_seen": 253273870,
      "step": 11739,
      "time_per_iteration": 2.74135422706604
    },
    {
      "auxiliary_loss_clip": 0.01079474,
      "auxiliary_loss_mlp": 0.00784056,
      "balance_loss_clip": 1.03554964,
      "balance_loss_mlp": 1.00883055,
      "epoch": 0.70584698632196,
      "flos": 26102963836800.0,
      "grad_norm": 2.133805715702363,
      "language_loss": 0.71574008,
      "learning_rate": 8.409054407293032e-07,
      "loss": 0.73437542,
      "num_input_tokens_seen": 253293720,
      "step": 11740,
      "time_per_iteration": 2.5713467597961426
    },
    {
      "auxiliary_loss_clip": 0.01075772,
      "auxiliary_loss_mlp": 0.01026328,
      "balance_loss_clip": 1.03785539,
      "balance_loss_mlp": 1.01503849,
      "epoch": 0.705907109574628,
      "flos": 21543134889600.0,
      "grad_norm": 1.6867974234924925,
      "language_loss": 0.81837296,
      "learning_rate": 8.405880753434434e-07,
      "loss": 0.83939397,
      "num_input_tokens_seen": 253313700,
      "step": 11741,
      "time_per_iteration": 2.558472156524658
    },
    {
      "auxiliary_loss_clip": 0.01083992,
      "auxiliary_loss_mlp": 0.0103098,
      "balance_loss_clip": 1.03586745,
      "balance_loss_mlp": 1.01877344,
      "epoch": 0.705967232827296,
      "flos": 22710842127360.0,
      "grad_norm": 2.6628635799343794,
      "language_loss": 0.7812252,
      "learning_rate": 8.402707539225993e-07,
      "loss": 0.80237496,
      "num_input_tokens_seen": 253332425,
      "step": 11742,
      "time_per_iteration": 2.5442867279052734
    },
    {
      "auxiliary_loss_clip": 0.01110491,
      "auxiliary_loss_mlp": 0.01031524,
      "balance_loss_clip": 1.03789842,
      "balance_loss_mlp": 1.01853657,
      "epoch": 0.7060273560799639,
      "flos": 28691225124480.0,
      "grad_norm": 1.5722572089306288,
      "language_loss": 0.64324683,
      "learning_rate": 8.39953476478805e-07,
      "loss": 0.66466701,
      "num_input_tokens_seen": 253353620,
      "step": 11743,
      "time_per_iteration": 3.9510343074798584
    },
    {
      "auxiliary_loss_clip": 0.01081626,
      "auxiliary_loss_mlp": 0.01030667,
      "balance_loss_clip": 1.03437364,
      "balance_loss_mlp": 1.0174284,
      "epoch": 0.7060874793326319,
      "flos": 15706178899200.0,
      "grad_norm": 23.628967158589028,
      "language_loss": 0.65466368,
      "learning_rate": 8.396362430240902e-07,
      "loss": 0.67578661,
      "num_input_tokens_seen": 253370930,
      "step": 11744,
      "time_per_iteration": 2.5341358184814453
    },
    {
      "auxiliary_loss_clip": 0.01094277,
      "auxiliary_loss_mlp": 0.01029577,
      "balance_loss_clip": 1.03605366,
      "balance_loss_mlp": 1.0175606,
      "epoch": 0.7061476025852998,
      "flos": 21506757390720.0,
      "grad_norm": 1.8873769020853095,
      "language_loss": 0.63876891,
      "learning_rate": 8.393190535704857e-07,
      "loss": 0.66000742,
      "num_input_tokens_seen": 253389810,
      "step": 11745,
      "time_per_iteration": 2.508784055709839
    },
    {
      "auxiliary_loss_clip": 0.01074613,
      "auxiliary_loss_mlp": 0.01034852,
      "balance_loss_clip": 1.03648329,
      "balance_loss_mlp": 1.02226973,
      "epoch": 0.7062077258379679,
      "flos": 28181832399360.0,
      "grad_norm": 1.884017002957585,
      "language_loss": 0.71712196,
      "learning_rate": 8.390019081300188e-07,
      "loss": 0.73821664,
      "num_input_tokens_seen": 253408685,
      "step": 11746,
      "time_per_iteration": 2.613478183746338
    },
    {
      "auxiliary_loss_clip": 0.01053227,
      "auxiliary_loss_mlp": 0.01030466,
      "balance_loss_clip": 1.03926897,
      "balance_loss_mlp": 1.01847911,
      "epoch": 0.7062678490906358,
      "flos": 27853680723840.0,
      "grad_norm": 1.7902915060111233,
      "language_loss": 0.79313517,
      "learning_rate": 8.386848067147175e-07,
      "loss": 0.81397212,
      "num_input_tokens_seen": 253429685,
      "step": 11747,
      "time_per_iteration": 2.6816179752349854
    },
    {
      "auxiliary_loss_clip": 0.01093136,
      "auxiliary_loss_mlp": 0.01030117,
      "balance_loss_clip": 1.03552139,
      "balance_loss_mlp": 1.01861334,
      "epoch": 0.7063279723433038,
      "flos": 23184862934400.0,
      "grad_norm": 1.8581009844236147,
      "language_loss": 0.65167099,
      "learning_rate": 8.383677493366031e-07,
      "loss": 0.67290354,
      "num_input_tokens_seen": 253448260,
      "step": 11748,
      "time_per_iteration": 3.88802433013916
    },
    {
      "auxiliary_loss_clip": 0.01064998,
      "auxiliary_loss_mlp": 0.01034784,
      "balance_loss_clip": 1.03788066,
      "balance_loss_mlp": 1.02226686,
      "epoch": 0.7063880955959717,
      "flos": 20188655907840.0,
      "grad_norm": 2.843370998842864,
      "language_loss": 0.79533505,
      "learning_rate": 8.380507360077003e-07,
      "loss": 0.81633282,
      "num_input_tokens_seen": 253467725,
      "step": 11749,
      "time_per_iteration": 2.59216570854187
    },
    {
      "auxiliary_loss_clip": 0.01038315,
      "auxiliary_loss_mlp": 0.01004633,
      "balance_loss_clip": 1.0147779,
      "balance_loss_mlp": 1.0034951,
      "epoch": 0.7064482188486397,
      "flos": 63668182763520.0,
      "grad_norm": 0.7908088262345453,
      "language_loss": 0.54062223,
      "learning_rate": 8.377337667400304e-07,
      "loss": 0.56105173,
      "num_input_tokens_seen": 253526940,
      "step": 11750,
      "time_per_iteration": 3.053283452987671
    },
    {
      "auxiliary_loss_clip": 0.01085283,
      "auxiliary_loss_mlp": 0.01033455,
      "balance_loss_clip": 1.03646743,
      "balance_loss_mlp": 1.02056873,
      "epoch": 0.7065083421013076,
      "flos": 25191227894400.0,
      "grad_norm": 1.8886910707641753,
      "language_loss": 0.78826952,
      "learning_rate": 8.37416841545612e-07,
      "loss": 0.80945694,
      "num_input_tokens_seen": 253546160,
      "step": 11751,
      "time_per_iteration": 2.5615971088409424
    },
    {
      "auxiliary_loss_clip": 0.01069679,
      "auxiliary_loss_mlp": 0.01031199,
      "balance_loss_clip": 1.03490472,
      "balance_loss_mlp": 1.01984441,
      "epoch": 0.7065684653539757,
      "flos": 22893699288960.0,
      "grad_norm": 2.140202270237752,
      "language_loss": 0.68199325,
      "learning_rate": 8.370999604364634e-07,
      "loss": 0.7030021,
      "num_input_tokens_seen": 253565505,
      "step": 11752,
      "time_per_iteration": 2.6132891178131104
    },
    {
      "auxiliary_loss_clip": 0.01051679,
      "auxiliary_loss_mlp": 0.007845,
      "balance_loss_clip": 1.03809214,
      "balance_loss_mlp": 1.01100874,
      "epoch": 0.7066285886066436,
      "flos": 23550254035200.0,
      "grad_norm": 1.8909861484710473,
      "language_loss": 0.76385832,
      "learning_rate": 8.367831234246025e-07,
      "loss": 0.78222013,
      "num_input_tokens_seen": 253585125,
      "step": 11753,
      "time_per_iteration": 2.7270426750183105
    },
    {
      "auxiliary_loss_clip": 0.01073342,
      "auxiliary_loss_mlp": 0.00785004,
      "balance_loss_clip": 1.0347265,
      "balance_loss_mlp": 1.01207876,
      "epoch": 0.7066887118593116,
      "flos": 21069293650560.0,
      "grad_norm": 1.4932652205863637,
      "language_loss": 0.70989978,
      "learning_rate": 8.364663305220405e-07,
      "loss": 0.7284832,
      "num_input_tokens_seen": 253604815,
      "step": 11754,
      "time_per_iteration": 2.643204689025879
    },
    {
      "auxiliary_loss_clip": 0.0107381,
      "auxiliary_loss_mlp": 0.0103763,
      "balance_loss_clip": 1.03607035,
      "balance_loss_mlp": 1.02372456,
      "epoch": 0.7067488351119796,
      "flos": 21176307244800.0,
      "grad_norm": 1.7785395898334453,
      "language_loss": 0.89349473,
      "learning_rate": 8.361495817407919e-07,
      "loss": 0.91460913,
      "num_input_tokens_seen": 253622855,
      "step": 11755,
      "time_per_iteration": 2.5736193656921387
    },
    {
      "auxiliary_loss_clip": 0.01082476,
      "auxiliary_loss_mlp": 0.00782447,
      "balance_loss_clip": 1.0361644,
      "balance_loss_mlp": 1.00743508,
      "epoch": 0.7068089583646475,
      "flos": 20449224144000.0,
      "grad_norm": 1.5269057761188012,
      "language_loss": 0.79689479,
      "learning_rate": 8.358328770928678e-07,
      "loss": 0.81554401,
      "num_input_tokens_seen": 253642760,
      "step": 11756,
      "time_per_iteration": 2.585432767868042
    },
    {
      "auxiliary_loss_clip": 0.0100527,
      "auxiliary_loss_mlp": 0.01004182,
      "balance_loss_clip": 1.02044058,
      "balance_loss_mlp": 1.00290668,
      "epoch": 0.7068690816173155,
      "flos": 59109179829120.0,
      "grad_norm": 0.8574023488957492,
      "language_loss": 0.60406053,
      "learning_rate": 8.355162165902785e-07,
      "loss": 0.62415504,
      "num_input_tokens_seen": 253695685,
      "step": 11757,
      "time_per_iteration": 2.9820353984832764
    },
    {
      "auxiliary_loss_clip": 0.0107363,
      "auxiliary_loss_mlp": 0.01033049,
      "balance_loss_clip": 1.03798866,
      "balance_loss_mlp": 1.02071655,
      "epoch": 0.7069292048699835,
      "flos": 16251554073600.0,
      "grad_norm": 1.8213451879188125,
      "language_loss": 0.80529422,
      "learning_rate": 8.351996002450307e-07,
      "loss": 0.826361,
      "num_input_tokens_seen": 253713305,
      "step": 11758,
      "time_per_iteration": 2.5374510288238525
    },
    {
      "auxiliary_loss_clip": 0.01072002,
      "auxiliary_loss_mlp": 0.00785076,
      "balance_loss_clip": 1.03570902,
      "balance_loss_mlp": 1.011199,
      "epoch": 0.7069893281226515,
      "flos": 41172768455040.0,
      "grad_norm": 1.730482523339282,
      "language_loss": 0.77776068,
      "learning_rate": 8.348830280691304e-07,
      "loss": 0.79633147,
      "num_input_tokens_seen": 253736100,
      "step": 11759,
      "time_per_iteration": 2.7453253269195557
    },
    {
      "auxiliary_loss_clip": 0.01096337,
      "auxiliary_loss_mlp": 0.01029086,
      "balance_loss_clip": 1.03594518,
      "balance_loss_mlp": 1.01679528,
      "epoch": 0.7070494513753194,
      "flos": 24207275658240.0,
      "grad_norm": 2.681303411552605,
      "language_loss": 0.68027902,
      "learning_rate": 8.34566500074583e-07,
      "loss": 0.70153326,
      "num_input_tokens_seen": 253757350,
      "step": 11760,
      "time_per_iteration": 2.540365219116211
    },
    {
      "auxiliary_loss_clip": 0.01079297,
      "auxiliary_loss_mlp": 0.01032333,
      "balance_loss_clip": 1.04006934,
      "balance_loss_mlp": 1.02058494,
      "epoch": 0.7071095746279874,
      "flos": 20185675079040.0,
      "grad_norm": 1.8235027358405613,
      "language_loss": 0.80528826,
      "learning_rate": 8.342500162733899e-07,
      "loss": 0.82640457,
      "num_input_tokens_seen": 253772855,
      "step": 11761,
      "time_per_iteration": 2.5380539894104004
    },
    {
      "auxiliary_loss_clip": 0.01082145,
      "auxiliary_loss_mlp": 0.01038786,
      "balance_loss_clip": 1.03504276,
      "balance_loss_mlp": 1.02412307,
      "epoch": 0.7071696978806553,
      "flos": 18183045133440.0,
      "grad_norm": 2.3139277819693453,
      "language_loss": 0.75076562,
      "learning_rate": 8.33933576677553e-07,
      "loss": 0.77197492,
      "num_input_tokens_seen": 253790360,
      "step": 11762,
      "time_per_iteration": 2.5037145614624023
    },
    {
      "auxiliary_loss_clip": 0.01083745,
      "auxiliary_loss_mlp": 0.01033295,
      "balance_loss_clip": 1.03664887,
      "balance_loss_mlp": 1.02190483,
      "epoch": 0.7072298211333233,
      "flos": 24131719399680.0,
      "grad_norm": 1.6690803003598065,
      "language_loss": 0.76951879,
      "learning_rate": 8.336171812990724e-07,
      "loss": 0.79068917,
      "num_input_tokens_seen": 253810585,
      "step": 11763,
      "time_per_iteration": 2.5493457317352295
    },
    {
      "auxiliary_loss_clip": 0.01075866,
      "auxiliary_loss_mlp": 0.00785916,
      "balance_loss_clip": 1.03565979,
      "balance_loss_mlp": 1.01069307,
      "epoch": 0.7072899443859912,
      "flos": 27198418867200.0,
      "grad_norm": 2.2169997222184334,
      "language_loss": 0.78759825,
      "learning_rate": 8.333008301499453e-07,
      "loss": 0.80621612,
      "num_input_tokens_seen": 253829080,
      "step": 11764,
      "time_per_iteration": 2.592029094696045
    },
    {
      "auxiliary_loss_clip": 0.01061989,
      "auxiliary_loss_mlp": 0.01032588,
      "balance_loss_clip": 1.03582239,
      "balance_loss_mlp": 1.01939774,
      "epoch": 0.7073500676386593,
      "flos": 16435596384000.0,
      "grad_norm": 1.5838489283731232,
      "language_loss": 0.79637909,
      "learning_rate": 8.32984523242167e-07,
      "loss": 0.81732488,
      "num_input_tokens_seen": 253846780,
      "step": 11765,
      "time_per_iteration": 2.652137279510498
    },
    {
      "auxiliary_loss_clip": 0.01104192,
      "auxiliary_loss_mlp": 0.01024895,
      "balance_loss_clip": 1.03707469,
      "balance_loss_mlp": 1.01370692,
      "epoch": 0.7074101908913272,
      "flos": 27673732563840.0,
      "grad_norm": 1.6262590107812513,
      "language_loss": 0.68390685,
      "learning_rate": 8.326682605877324e-07,
      "loss": 0.70519769,
      "num_input_tokens_seen": 253867075,
      "step": 11766,
      "time_per_iteration": 2.5330007076263428
    },
    {
      "auxiliary_loss_clip": 0.01085234,
      "auxiliary_loss_mlp": 0.01037371,
      "balance_loss_clip": 1.03614569,
      "balance_loss_mlp": 1.02407885,
      "epoch": 0.7074703141439952,
      "flos": 22238078296320.0,
      "grad_norm": 1.8637518776668605,
      "language_loss": 0.64006382,
      "learning_rate": 8.323520421986352e-07,
      "loss": 0.66128993,
      "num_input_tokens_seen": 253885790,
      "step": 11767,
      "time_per_iteration": 2.5549161434173584
    },
    {
      "auxiliary_loss_clip": 0.01094544,
      "auxiliary_loss_mlp": 0.01024941,
      "balance_loss_clip": 1.03521633,
      "balance_loss_mlp": 1.01260853,
      "epoch": 0.7075304373966632,
      "flos": 29643217234560.0,
      "grad_norm": 1.9748919106062652,
      "language_loss": 0.52758294,
      "learning_rate": 8.320358680868646e-07,
      "loss": 0.54877782,
      "num_input_tokens_seen": 253907070,
      "step": 11768,
      "time_per_iteration": 2.557574987411499
    },
    {
      "auxiliary_loss_clip": 0.01083107,
      "auxiliary_loss_mlp": 0.00783384,
      "balance_loss_clip": 1.03648615,
      "balance_loss_mlp": 1.00769377,
      "epoch": 0.7075905606493311,
      "flos": 19755214490880.0,
      "grad_norm": 1.8294972946890684,
      "language_loss": 0.75557309,
      "learning_rate": 8.317197382644119e-07,
      "loss": 0.77423799,
      "num_input_tokens_seen": 253927290,
      "step": 11769,
      "time_per_iteration": 2.557758092880249
    },
    {
      "auxiliary_loss_clip": 0.01021914,
      "auxiliary_loss_mlp": 0.01007352,
      "balance_loss_clip": 1.01784515,
      "balance_loss_mlp": 1.00601125,
      "epoch": 0.7076506839019991,
      "flos": 65716132694400.0,
      "grad_norm": 0.8495030966364372,
      "language_loss": 0.62002724,
      "learning_rate": 8.314036527432637e-07,
      "loss": 0.64031994,
      "num_input_tokens_seen": 253983440,
      "step": 11770,
      "time_per_iteration": 3.073887348175049
    },
    {
      "auxiliary_loss_clip": 0.01073644,
      "auxiliary_loss_mlp": 0.01033228,
      "balance_loss_clip": 1.03441691,
      "balance_loss_mlp": 1.0205977,
      "epoch": 0.707710807154667,
      "flos": 23765286804480.0,
      "grad_norm": 1.7094072526073503,
      "language_loss": 0.76054037,
      "learning_rate": 8.310876115354055e-07,
      "loss": 0.78160906,
      "num_input_tokens_seen": 254003825,
      "step": 11771,
      "time_per_iteration": 2.5860376358032227
    },
    {
      "auxiliary_loss_clip": 0.0109128,
      "auxiliary_loss_mlp": 0.01028431,
      "balance_loss_clip": 1.03465319,
      "balance_loss_mlp": 1.01721907,
      "epoch": 0.7077709304073351,
      "flos": 21251360712960.0,
      "grad_norm": 1.6088683127666719,
      "language_loss": 0.71301168,
      "learning_rate": 8.307716146528221e-07,
      "loss": 0.73420882,
      "num_input_tokens_seen": 254023345,
      "step": 11772,
      "time_per_iteration": 2.496386766433716
    },
    {
      "auxiliary_loss_clip": 0.01066219,
      "auxiliary_loss_mlp": 0.01027038,
      "balance_loss_clip": 1.03573573,
      "balance_loss_mlp": 1.01420522,
      "epoch": 0.707831053660003,
      "flos": 20740746925440.0,
      "grad_norm": 2.1593412016459355,
      "language_loss": 0.6966899,
      "learning_rate": 8.30455662107496e-07,
      "loss": 0.71762252,
      "num_input_tokens_seen": 254041815,
      "step": 11773,
      "time_per_iteration": 2.5877084732055664
    },
    {
      "auxiliary_loss_clip": 0.01096354,
      "auxiliary_loss_mlp": 0.01034478,
      "balance_loss_clip": 1.03592932,
      "balance_loss_mlp": 1.02252698,
      "epoch": 0.707891176912671,
      "flos": 21980993679360.0,
      "grad_norm": 1.4667405218579366,
      "language_loss": 0.70309919,
      "learning_rate": 8.301397539114095e-07,
      "loss": 0.72440749,
      "num_input_tokens_seen": 254062065,
      "step": 11774,
      "time_per_iteration": 2.528050661087036
    },
    {
      "auxiliary_loss_clip": 0.01082864,
      "auxiliary_loss_mlp": 0.01027584,
      "balance_loss_clip": 1.03866959,
      "balance_loss_mlp": 1.01574612,
      "epoch": 0.7079513001653389,
      "flos": 21068970428160.0,
      "grad_norm": 1.4512599272490367,
      "language_loss": 0.74325287,
      "learning_rate": 8.298238900765407e-07,
      "loss": 0.76435733,
      "num_input_tokens_seen": 254080605,
      "step": 11775,
      "time_per_iteration": 2.541327953338623
    },
    {
      "auxiliary_loss_clip": 0.01077382,
      "auxiliary_loss_mlp": 0.00783649,
      "balance_loss_clip": 1.03799796,
      "balance_loss_mlp": 1.00809169,
      "epoch": 0.7080114234180069,
      "flos": 18040659621120.0,
      "grad_norm": 1.6262815112444449,
      "language_loss": 0.8675217,
      "learning_rate": 8.295080706148665e-07,
      "loss": 0.88613206,
      "num_input_tokens_seen": 254098710,
      "step": 11776,
      "time_per_iteration": 3.9424960613250732
    },
    {
      "auxiliary_loss_clip": 0.01087989,
      "auxiliary_loss_mlp": 0.01033326,
      "balance_loss_clip": 1.034235,
      "balance_loss_mlp": 1.02172065,
      "epoch": 0.7080715466706748,
      "flos": 15122271409920.0,
      "grad_norm": 1.4666147076005813,
      "language_loss": 0.74810511,
      "learning_rate": 8.291922955383641e-07,
      "loss": 0.76931828,
      "num_input_tokens_seen": 254117200,
      "step": 11777,
      "time_per_iteration": 2.4756884574890137
    },
    {
      "auxiliary_loss_clip": 0.01090502,
      "auxiliary_loss_mlp": 0.01034049,
      "balance_loss_clip": 1.04117811,
      "balance_loss_mlp": 1.02126408,
      "epoch": 0.7081316699233429,
      "flos": 14422802889600.0,
      "grad_norm": 2.5244873900628697,
      "language_loss": 0.81826872,
      "learning_rate": 8.288765648590066e-07,
      "loss": 0.83951414,
      "num_input_tokens_seen": 254132115,
      "step": 11778,
      "time_per_iteration": 2.5068929195404053
    },
    {
      "auxiliary_loss_clip": 0.01078624,
      "auxiliary_loss_mlp": 0.01031977,
      "balance_loss_clip": 1.03473139,
      "balance_loss_mlp": 1.02104592,
      "epoch": 0.7081917931760108,
      "flos": 23222389668480.0,
      "grad_norm": 1.5017814693634735,
      "language_loss": 0.84945858,
      "learning_rate": 8.285608785887673e-07,
      "loss": 0.87056464,
      "num_input_tokens_seen": 254152285,
      "step": 11779,
      "time_per_iteration": 2.535895824432373
    },
    {
      "auxiliary_loss_clip": 0.01082955,
      "auxiliary_loss_mlp": 0.01033405,
      "balance_loss_clip": 1.03695011,
      "balance_loss_mlp": 1.02127576,
      "epoch": 0.7082519164286788,
      "flos": 39308429871360.0,
      "grad_norm": 2.2331249118925895,
      "language_loss": 0.71925896,
      "learning_rate": 8.28245236739618e-07,
      "loss": 0.74042261,
      "num_input_tokens_seen": 254172805,
      "step": 11780,
      "time_per_iteration": 2.6861252784729004
    },
    {
      "auxiliary_loss_clip": 0.01064549,
      "auxiliary_loss_mlp": 0.01028315,
      "balance_loss_clip": 1.0365622,
      "balance_loss_mlp": 1.01675165,
      "epoch": 0.7083120396813467,
      "flos": 21651154064640.0,
      "grad_norm": 1.4513691112671563,
      "language_loss": 0.73076469,
      "learning_rate": 8.279296393235256e-07,
      "loss": 0.75169325,
      "num_input_tokens_seen": 254191890,
      "step": 11781,
      "time_per_iteration": 2.581162929534912
    },
    {
      "auxiliary_loss_clip": 0.01093519,
      "auxiliary_loss_mlp": 0.01029457,
      "balance_loss_clip": 1.03640628,
      "balance_loss_mlp": 1.01784563,
      "epoch": 0.7083721629340147,
      "flos": 17567033863680.0,
      "grad_norm": 1.5744545022046663,
      "language_loss": 0.77138209,
      "learning_rate": 8.276140863524585e-07,
      "loss": 0.79261178,
      "num_input_tokens_seen": 254210150,
      "step": 11782,
      "time_per_iteration": 5.248617649078369
    },
    {
      "auxiliary_loss_clip": 0.01080739,
      "auxiliary_loss_mlp": 0.01029301,
      "balance_loss_clip": 1.03518724,
      "balance_loss_mlp": 1.01834559,
      "epoch": 0.7084322861866827,
      "flos": 29350509304320.0,
      "grad_norm": 1.4298268694176417,
      "language_loss": 0.69667482,
      "learning_rate": 8.272985778383828e-07,
      "loss": 0.71777511,
      "num_input_tokens_seen": 254233015,
      "step": 11783,
      "time_per_iteration": 2.5956382751464844
    },
    {
      "auxiliary_loss_clip": 0.01069598,
      "auxiliary_loss_mlp": 0.01028866,
      "balance_loss_clip": 1.03777146,
      "balance_loss_mlp": 1.01669478,
      "epoch": 0.7084924094393507,
      "flos": 20194294343040.0,
      "grad_norm": 1.5843600085134995,
      "language_loss": 0.79068208,
      "learning_rate": 8.269831137932632e-07,
      "loss": 0.81166667,
      "num_input_tokens_seen": 254251345,
      "step": 11784,
      "time_per_iteration": 2.5822486877441406
    },
    {
      "auxiliary_loss_clip": 0.01105168,
      "auxiliary_loss_mlp": 0.01029712,
      "balance_loss_clip": 1.03690481,
      "balance_loss_mlp": 1.01790476,
      "epoch": 0.7085525326920187,
      "flos": 23477211728640.0,
      "grad_norm": 1.7391549636840924,
      "language_loss": 0.77021039,
      "learning_rate": 8.266676942290609e-07,
      "loss": 0.79155916,
      "num_input_tokens_seen": 254269905,
      "step": 11785,
      "time_per_iteration": 2.5559744834899902
    },
    {
      "auxiliary_loss_clip": 0.01080647,
      "auxiliary_loss_mlp": 0.0103144,
      "balance_loss_clip": 1.03644383,
      "balance_loss_mlp": 1.01901209,
      "epoch": 0.7086126559446866,
      "flos": 25958818558080.0,
      "grad_norm": 1.5380593344271858,
      "language_loss": 0.78051555,
      "learning_rate": 8.26352319157738e-07,
      "loss": 0.80163634,
      "num_input_tokens_seen": 254289990,
      "step": 11786,
      "time_per_iteration": 2.570871591567993
    },
    {
      "auxiliary_loss_clip": 0.01106305,
      "auxiliary_loss_mlp": 0.01028955,
      "balance_loss_clip": 1.03630805,
      "balance_loss_mlp": 1.01662898,
      "epoch": 0.7086727791973546,
      "flos": 26724793109760.0,
      "grad_norm": 2.9181617068764836,
      "language_loss": 0.78732586,
      "learning_rate": 8.260369885912526e-07,
      "loss": 0.80867839,
      "num_input_tokens_seen": 254309085,
      "step": 11787,
      "time_per_iteration": 3.886140823364258
    },
    {
      "auxiliary_loss_clip": 0.01096342,
      "auxiliary_loss_mlp": 0.01027359,
      "balance_loss_clip": 1.03754616,
      "balance_loss_mlp": 1.01572466,
      "epoch": 0.7087329024500225,
      "flos": 21683365585920.0,
      "grad_norm": 1.8951355388829545,
      "language_loss": 0.76805693,
      "learning_rate": 8.257217025415615e-07,
      "loss": 0.78929394,
      "num_input_tokens_seen": 254327045,
      "step": 11788,
      "time_per_iteration": 2.5009024143218994
    },
    {
      "auxiliary_loss_clip": 0.01070801,
      "auxiliary_loss_mlp": 0.0103852,
      "balance_loss_clip": 1.03607392,
      "balance_loss_mlp": 1.02303433,
      "epoch": 0.7087930257026905,
      "flos": 17931060247680.0,
      "grad_norm": 2.42268286824252,
      "language_loss": 0.68139791,
      "learning_rate": 8.254064610206212e-07,
      "loss": 0.70249116,
      "num_input_tokens_seen": 254344585,
      "step": 11789,
      "time_per_iteration": 2.5427772998809814
    },
    {
      "auxiliary_loss_clip": 0.01051101,
      "auxiliary_loss_mlp": 0.01028895,
      "balance_loss_clip": 1.03739977,
      "balance_loss_mlp": 1.01679528,
      "epoch": 0.7088531489553584,
      "flos": 18911528864640.0,
      "grad_norm": 1.5526868780707694,
      "language_loss": 0.77620441,
      "learning_rate": 8.250912640403858e-07,
      "loss": 0.79700434,
      "num_input_tokens_seen": 254362470,
      "step": 11790,
      "time_per_iteration": 2.6356847286224365
    },
    {
      "auxiliary_loss_clip": 0.01087626,
      "auxiliary_loss_mlp": 0.01027092,
      "balance_loss_clip": 1.03649497,
      "balance_loss_mlp": 1.01427078,
      "epoch": 0.7089132722080265,
      "flos": 27380880979200.0,
      "grad_norm": 1.6718298735321997,
      "language_loss": 0.70836681,
      "learning_rate": 8.247761116128085e-07,
      "loss": 0.729514,
      "num_input_tokens_seen": 254383190,
      "step": 11791,
      "time_per_iteration": 2.5816049575805664
    },
    {
      "auxiliary_loss_clip": 0.01096124,
      "auxiliary_loss_mlp": 0.01033513,
      "balance_loss_clip": 1.03756809,
      "balance_loss_mlp": 1.02128756,
      "epoch": 0.7089733954606944,
      "flos": 22162917087360.0,
      "grad_norm": 1.8999428669554808,
      "language_loss": 0.81973851,
      "learning_rate": 8.244610037498376e-07,
      "loss": 0.84103489,
      "num_input_tokens_seen": 254403115,
      "step": 11792,
      "time_per_iteration": 2.5651743412017822
    },
    {
      "auxiliary_loss_clip": 0.01073021,
      "auxiliary_loss_mlp": 0.01028474,
      "balance_loss_clip": 1.03711462,
      "balance_loss_mlp": 1.01649308,
      "epoch": 0.7090335187133624,
      "flos": 24425827960320.0,
      "grad_norm": 2.5508410370732313,
      "language_loss": 0.6502139,
      "learning_rate": 8.241459404634232e-07,
      "loss": 0.67122889,
      "num_input_tokens_seen": 254421875,
      "step": 11793,
      "time_per_iteration": 2.593559741973877
    },
    {
      "auxiliary_loss_clip": 0.01090906,
      "auxiliary_loss_mlp": 0.01031697,
      "balance_loss_clip": 1.03546977,
      "balance_loss_mlp": 1.01872087,
      "epoch": 0.7090936419660303,
      "flos": 21835232288640.0,
      "grad_norm": 2.497946969333162,
      "language_loss": 0.70430225,
      "learning_rate": 8.238309217655133e-07,
      "loss": 0.72552824,
      "num_input_tokens_seen": 254440765,
      "step": 11794,
      "time_per_iteration": 2.5262818336486816
    },
    {
      "auxiliary_loss_clip": 0.01088519,
      "auxiliary_loss_mlp": 0.01028834,
      "balance_loss_clip": 1.04117095,
      "balance_loss_mlp": 1.01780736,
      "epoch": 0.7091537652186983,
      "flos": 20082360585600.0,
      "grad_norm": 1.869664078333111,
      "language_loss": 0.75351489,
      "learning_rate": 8.23515947668052e-07,
      "loss": 0.77468848,
      "num_input_tokens_seen": 254459480,
      "step": 11795,
      "time_per_iteration": 2.5282771587371826
    },
    {
      "auxiliary_loss_clip": 0.01070049,
      "auxiliary_loss_mlp": 0.01030747,
      "balance_loss_clip": 1.03758407,
      "balance_loss_mlp": 1.01929724,
      "epoch": 0.7092138884713663,
      "flos": 13151565676800.0,
      "grad_norm": 2.709712184618782,
      "language_loss": 0.75185621,
      "learning_rate": 8.232010181829838e-07,
      "loss": 0.77286416,
      "num_input_tokens_seen": 254473985,
      "step": 11796,
      "time_per_iteration": 2.5336527824401855
    },
    {
      "auxiliary_loss_clip": 0.0109888,
      "auxiliary_loss_mlp": 0.01040536,
      "balance_loss_clip": 1.03872907,
      "balance_loss_mlp": 1.02514577,
      "epoch": 0.7092740117240343,
      "flos": 21645982506240.0,
      "grad_norm": 1.5567548151617336,
      "language_loss": 0.74274707,
      "learning_rate": 8.228861333222523e-07,
      "loss": 0.76414126,
      "num_input_tokens_seen": 254492135,
      "step": 11797,
      "time_per_iteration": 2.500070333480835
    },
    {
      "auxiliary_loss_clip": 0.0106308,
      "auxiliary_loss_mlp": 0.01031671,
      "balance_loss_clip": 1.03985226,
      "balance_loss_mlp": 1.01958323,
      "epoch": 0.7093341349767023,
      "flos": 21032521102080.0,
      "grad_norm": 1.595048094160187,
      "language_loss": 0.79609418,
      "learning_rate": 8.225712930977953e-07,
      "loss": 0.81704175,
      "num_input_tokens_seen": 254512865,
      "step": 11798,
      "time_per_iteration": 2.6384851932525635
    },
    {
      "auxiliary_loss_clip": 0.01076029,
      "auxiliary_loss_mlp": 0.01037743,
      "balance_loss_clip": 1.0350647,
      "balance_loss_mlp": 1.02492833,
      "epoch": 0.7093942582293702,
      "flos": 22017658487040.0,
      "grad_norm": 1.7185255966250672,
      "language_loss": 0.66528994,
      "learning_rate": 8.222564975215529e-07,
      "loss": 0.68642765,
      "num_input_tokens_seen": 254532605,
      "step": 11799,
      "time_per_iteration": 2.5415475368499756
    },
    {
      "auxiliary_loss_clip": 0.0110682,
      "auxiliary_loss_mlp": 0.01027262,
      "balance_loss_clip": 1.03691745,
      "balance_loss_mlp": 1.01465523,
      "epoch": 0.7094543814820382,
      "flos": 27235586465280.0,
      "grad_norm": 1.6727735975361675,
      "language_loss": 0.8149184,
      "learning_rate": 8.219417466054622e-07,
      "loss": 0.83625925,
      "num_input_tokens_seen": 254553780,
      "step": 11800,
      "time_per_iteration": 2.5287108421325684
    },
    {
      "auxiliary_loss_clip": 0.0108172,
      "auxiliary_loss_mlp": 0.01029222,
      "balance_loss_clip": 1.03490257,
      "balance_loss_mlp": 1.01817727,
      "epoch": 0.7095145047347061,
      "flos": 12089148180480.0,
      "grad_norm": 1.779697027967369,
      "language_loss": 0.86577427,
      "learning_rate": 8.21627040361459e-07,
      "loss": 0.88688368,
      "num_input_tokens_seen": 254567510,
      "step": 11801,
      "time_per_iteration": 2.480809450149536
    },
    {
      "auxiliary_loss_clip": 0.01106804,
      "auxiliary_loss_mlp": 0.01033733,
      "balance_loss_clip": 1.0369041,
      "balance_loss_mlp": 1.0220803,
      "epoch": 0.7095746279873741,
      "flos": 19383789905280.0,
      "grad_norm": 1.874459133629234,
      "language_loss": 0.75855446,
      "learning_rate": 8.213123788014758e-07,
      "loss": 0.7799598,
      "num_input_tokens_seen": 254585565,
      "step": 11802,
      "time_per_iteration": 2.46742582321167
    },
    {
      "auxiliary_loss_clip": 0.01093842,
      "auxiliary_loss_mlp": 0.01042488,
      "balance_loss_clip": 1.03737068,
      "balance_loss_mlp": 1.03023946,
      "epoch": 0.709634751240042,
      "flos": 21360600950400.0,
      "grad_norm": 1.983154845812974,
      "language_loss": 0.81480432,
      "learning_rate": 8.209977619374462e-07,
      "loss": 0.83616769,
      "num_input_tokens_seen": 254603465,
      "step": 11803,
      "time_per_iteration": 2.4947409629821777
    },
    {
      "auxiliary_loss_clip": 0.01107663,
      "auxiliary_loss_mlp": 0.01029458,
      "balance_loss_clip": 1.03627753,
      "balance_loss_mlp": 1.01635647,
      "epoch": 0.7096948744927101,
      "flos": 13917037438080.0,
      "grad_norm": 2.4515657432602866,
      "language_loss": 0.67676228,
      "learning_rate": 8.206831897812995e-07,
      "loss": 0.69813347,
      "num_input_tokens_seen": 254620500,
      "step": 11804,
      "time_per_iteration": 2.4461069107055664
    },
    {
      "auxiliary_loss_clip": 0.01093427,
      "auxiliary_loss_mlp": 0.01026395,
      "balance_loss_clip": 1.03755891,
      "balance_loss_mlp": 1.01544011,
      "epoch": 0.709754997745378,
      "flos": 30298335436800.0,
      "grad_norm": 1.897949680464105,
      "language_loss": 0.77906889,
      "learning_rate": 8.203686623449637e-07,
      "loss": 0.8002671,
      "num_input_tokens_seen": 254638565,
      "step": 11805,
      "time_per_iteration": 2.5589823722839355
    },
    {
      "auxiliary_loss_clip": 0.01084555,
      "auxiliary_loss_mlp": 0.00785677,
      "balance_loss_clip": 1.03506815,
      "balance_loss_mlp": 1.01206911,
      "epoch": 0.709815120998046,
      "flos": 18515147304960.0,
      "grad_norm": 2.4471816686872505,
      "language_loss": 0.78567314,
      "learning_rate": 8.200541796403667e-07,
      "loss": 0.80437547,
      "num_input_tokens_seen": 254657505,
      "step": 11806,
      "time_per_iteration": 2.5408620834350586
    },
    {
      "auxiliary_loss_clip": 0.01076941,
      "auxiliary_loss_mlp": 0.01034891,
      "balance_loss_clip": 1.03621745,
      "balance_loss_mlp": 1.02205253,
      "epoch": 0.7098752442507139,
      "flos": 22272588288000.0,
      "grad_norm": 2.177272877487699,
      "language_loss": 0.56475133,
      "learning_rate": 8.197397416794332e-07,
      "loss": 0.58586967,
      "num_input_tokens_seen": 254674730,
      "step": 11807,
      "time_per_iteration": 2.522702932357788
    },
    {
      "auxiliary_loss_clip": 0.01109393,
      "auxiliary_loss_mlp": 0.01035938,
      "balance_loss_clip": 1.03614485,
      "balance_loss_mlp": 1.02402294,
      "epoch": 0.7099353675033819,
      "flos": 19275447507840.0,
      "grad_norm": 2.5389195494465717,
      "language_loss": 0.68648064,
      "learning_rate": 8.194253484740882e-07,
      "loss": 0.70793396,
      "num_input_tokens_seen": 254691665,
      "step": 11808,
      "time_per_iteration": 2.468324661254883
    },
    {
      "auxiliary_loss_clip": 0.01093992,
      "auxiliary_loss_mlp": 0.01029173,
      "balance_loss_clip": 1.03763556,
      "balance_loss_mlp": 1.01772344,
      "epoch": 0.70999549075605,
      "flos": 21908525990400.0,
      "grad_norm": 2.6876997439282464,
      "language_loss": 0.71508497,
      "learning_rate": 8.191110000362513e-07,
      "loss": 0.73631662,
      "num_input_tokens_seen": 254711610,
      "step": 11809,
      "time_per_iteration": 2.4962143898010254
    },
    {
      "auxiliary_loss_clip": 0.01036507,
      "auxiliary_loss_mlp": 0.01001069,
      "balance_loss_clip": 1.01284981,
      "balance_loss_mlp": 0.99986535,
      "epoch": 0.7100556140087179,
      "flos": 70456053456000.0,
      "grad_norm": 0.7483510495460608,
      "language_loss": 0.5945394,
      "learning_rate": 8.187966963778435e-07,
      "loss": 0.61491513,
      "num_input_tokens_seen": 254772615,
      "step": 11810,
      "time_per_iteration": 3.169534683227539
    },
    {
      "auxiliary_loss_clip": 0.01039605,
      "auxiliary_loss_mlp": 0.01043761,
      "balance_loss_clip": 1.0354991,
      "balance_loss_mlp": 1.02996242,
      "epoch": 0.7101157372613859,
      "flos": 23039568420480.0,
      "grad_norm": 1.5628443904461748,
      "language_loss": 0.74144542,
      "learning_rate": 8.18482437510784e-07,
      "loss": 0.76227903,
      "num_input_tokens_seen": 254791375,
      "step": 11811,
      "time_per_iteration": 2.6607277393341064
    },
    {
      "auxiliary_loss_clip": 0.01073759,
      "auxiliary_loss_mlp": 0.01027984,
      "balance_loss_clip": 1.03779316,
      "balance_loss_mlp": 1.01615262,
      "epoch": 0.7101758605140538,
      "flos": 23185329811200.0,
      "grad_norm": 1.8342083935579412,
      "language_loss": 0.83482438,
      "learning_rate": 8.181682234469882e-07,
      "loss": 0.85584188,
      "num_input_tokens_seen": 254809300,
      "step": 11812,
      "time_per_iteration": 2.5695459842681885
    },
    {
      "auxiliary_loss_clip": 0.01109164,
      "auxiliary_loss_mlp": 0.0103029,
      "balance_loss_clip": 1.03775239,
      "balance_loss_mlp": 1.01718855,
      "epoch": 0.7102359837667218,
      "flos": 23696123166720.0,
      "grad_norm": 2.045256692905346,
      "language_loss": 0.7014035,
      "learning_rate": 8.178540541983716e-07,
      "loss": 0.72279805,
      "num_input_tokens_seen": 254829325,
      "step": 11813,
      "time_per_iteration": 2.5160768032073975
    },
    {
      "auxiliary_loss_clip": 0.01103267,
      "auxiliary_loss_mlp": 0.01024456,
      "balance_loss_clip": 1.03558064,
      "balance_loss_mlp": 1.01321435,
      "epoch": 0.7102961070193897,
      "flos": 19391116279680.0,
      "grad_norm": 1.7146182990175622,
      "language_loss": 0.81607974,
      "learning_rate": 8.175399297768495e-07,
      "loss": 0.83735693,
      "num_input_tokens_seen": 254847690,
      "step": 11814,
      "time_per_iteration": 2.4950411319732666
    },
    {
      "auxiliary_loss_clip": 0.01104892,
      "auxiliary_loss_mlp": 0.01028509,
      "balance_loss_clip": 1.03693318,
      "balance_loss_mlp": 1.0165168,
      "epoch": 0.7103562302720577,
      "flos": 21507511576320.0,
      "grad_norm": 1.7360616123628676,
      "language_loss": 0.76018602,
      "learning_rate": 8.172258501943301e-07,
      "loss": 0.78152001,
      "num_input_tokens_seen": 254865960,
      "step": 11815,
      "time_per_iteration": 3.8388211727142334
    },
    {
      "auxiliary_loss_clip": 0.01064508,
      "auxiliary_loss_mlp": 0.01033115,
      "balance_loss_clip": 1.03775096,
      "balance_loss_mlp": 1.02106917,
      "epoch": 0.7104163535247257,
      "flos": 14535059869440.0,
      "grad_norm": 1.8152900184516274,
      "language_loss": 0.78410369,
      "learning_rate": 8.16911815462725e-07,
      "loss": 0.80507994,
      "num_input_tokens_seen": 254882815,
      "step": 11816,
      "time_per_iteration": 2.5539920330047607
    },
    {
      "auxiliary_loss_clip": 0.01080519,
      "auxiliary_loss_mlp": 0.0103175,
      "balance_loss_clip": 1.03673339,
      "balance_loss_mlp": 1.01962662,
      "epoch": 0.7104764767773937,
      "flos": 11400310085760.0,
      "grad_norm": 1.774224886354809,
      "language_loss": 0.86314046,
      "learning_rate": 8.165978255939426e-07,
      "loss": 0.8842631,
      "num_input_tokens_seen": 254898705,
      "step": 11817,
      "time_per_iteration": 2.5200963020324707
    },
    {
      "auxiliary_loss_clip": 0.01057021,
      "auxiliary_loss_mlp": 0.01027428,
      "balance_loss_clip": 1.03649855,
      "balance_loss_mlp": 1.01615119,
      "epoch": 0.7105366000300616,
      "flos": 11690432236800.0,
      "grad_norm": 3.029342511072785,
      "language_loss": 0.84540069,
      "learning_rate": 8.162838805998897e-07,
      "loss": 0.86624521,
      "num_input_tokens_seen": 254913665,
      "step": 11818,
      "time_per_iteration": 2.545146942138672
    },
    {
      "auxiliary_loss_clip": 0.01105989,
      "auxiliary_loss_mlp": 0.01029801,
      "balance_loss_clip": 1.03595114,
      "balance_loss_mlp": 1.01715291,
      "epoch": 0.7105967232827296,
      "flos": 19354020508800.0,
      "grad_norm": 2.0980833699804404,
      "language_loss": 0.75530052,
      "learning_rate": 8.159699804924709e-07,
      "loss": 0.77665842,
      "num_input_tokens_seen": 254932140,
      "step": 11819,
      "time_per_iteration": 2.464923143386841
    },
    {
      "auxiliary_loss_clip": 0.01063071,
      "auxiliary_loss_mlp": 0.01031666,
      "balance_loss_clip": 1.03433299,
      "balance_loss_mlp": 1.01755786,
      "epoch": 0.7106568465353975,
      "flos": 22930400010240.0,
      "grad_norm": 1.6720868736826446,
      "language_loss": 0.7086072,
      "learning_rate": 8.156561252835883e-07,
      "loss": 0.72955453,
      "num_input_tokens_seen": 254951580,
      "step": 11820,
      "time_per_iteration": 3.9713776111602783
    },
    {
      "auxiliary_loss_clip": 0.01094728,
      "auxiliary_loss_mlp": 0.01026727,
      "balance_loss_clip": 1.03714085,
      "balance_loss_mlp": 1.01519907,
      "epoch": 0.7107169697880655,
      "flos": 19099665325440.0,
      "grad_norm": 1.8393880804519012,
      "language_loss": 0.75470376,
      "learning_rate": 8.153423149851449e-07,
      "loss": 0.7759183,
      "num_input_tokens_seen": 254969425,
      "step": 11821,
      "time_per_iteration": 3.8560032844543457
    },
    {
      "auxiliary_loss_clip": 0.01005554,
      "auxiliary_loss_mlp": 0.01000658,
      "balance_loss_clip": 1.02657866,
      "balance_loss_mlp": 0.99923342,
      "epoch": 0.7107770930407336,
      "flos": 63638054231040.0,
      "grad_norm": 0.7624554414622751,
      "language_loss": 0.55088925,
      "learning_rate": 8.150285496090388e-07,
      "loss": 0.57095134,
      "num_input_tokens_seen": 255032680,
      "step": 11822,
      "time_per_iteration": 3.2205963134765625
    },
    {
      "auxiliary_loss_clip": 0.01090015,
      "auxiliary_loss_mlp": 0.01027214,
      "balance_loss_clip": 1.03647017,
      "balance_loss_mlp": 1.01507294,
      "epoch": 0.7108372162934015,
      "flos": 22054466949120.0,
      "grad_norm": 2.0107513586868864,
      "language_loss": 0.59872448,
      "learning_rate": 8.147148291671688e-07,
      "loss": 0.61989677,
      "num_input_tokens_seen": 255054400,
      "step": 11823,
      "time_per_iteration": 2.5446133613586426
    },
    {
      "auxiliary_loss_clip": 0.01094459,
      "auxiliary_loss_mlp": 0.01030182,
      "balance_loss_clip": 1.03660488,
      "balance_loss_mlp": 1.01879203,
      "epoch": 0.7108973395460695,
      "flos": 19135144984320.0,
      "grad_norm": 2.480542282973698,
      "language_loss": 0.71025974,
      "learning_rate": 8.144011536714322e-07,
      "loss": 0.73150617,
      "num_input_tokens_seen": 255072785,
      "step": 11824,
      "time_per_iteration": 2.485163688659668
    },
    {
      "auxiliary_loss_clip": 0.01073271,
      "auxiliary_loss_mlp": 0.00785804,
      "balance_loss_clip": 1.03370345,
      "balance_loss_mlp": 1.00998855,
      "epoch": 0.7109574627987374,
      "flos": 17894431353600.0,
      "grad_norm": 1.6959991386346989,
      "language_loss": 0.7262789,
      "learning_rate": 8.140875231337223e-07,
      "loss": 0.74486959,
      "num_input_tokens_seen": 255091820,
      "step": 11825,
      "time_per_iteration": 2.5357978343963623
    },
    {
      "auxiliary_loss_clip": 0.0108435,
      "auxiliary_loss_mlp": 0.01031824,
      "balance_loss_clip": 1.03866196,
      "balance_loss_mlp": 1.01993287,
      "epoch": 0.7110175860514054,
      "flos": 28979623422720.0,
      "grad_norm": 1.7127081326772016,
      "language_loss": 0.79757535,
      "learning_rate": 8.137739375659321e-07,
      "loss": 0.81873703,
      "num_input_tokens_seen": 255111720,
      "step": 11826,
      "time_per_iteration": 4.0725390911102295
    },
    {
      "auxiliary_loss_clip": 0.0109324,
      "auxiliary_loss_mlp": 0.01030281,
      "balance_loss_clip": 1.03673542,
      "balance_loss_mlp": 1.01910472,
      "epoch": 0.7110777093040733,
      "flos": 26173312623360.0,
      "grad_norm": 1.5462702664498276,
      "language_loss": 0.83024204,
      "learning_rate": 8.134603969799527e-07,
      "loss": 0.85147727,
      "num_input_tokens_seen": 255133495,
      "step": 11827,
      "time_per_iteration": 2.559511661529541
    },
    {
      "auxiliary_loss_clip": 0.01074178,
      "auxiliary_loss_mlp": 0.01031989,
      "balance_loss_clip": 1.03546691,
      "balance_loss_mlp": 1.01910865,
      "epoch": 0.7111378325567413,
      "flos": 26869943969280.0,
      "grad_norm": 1.5420453511531356,
      "language_loss": 0.62318075,
      "learning_rate": 8.131469013876748e-07,
      "loss": 0.64424241,
      "num_input_tokens_seen": 255156880,
      "step": 11828,
      "time_per_iteration": 2.609049081802368
    },
    {
      "auxiliary_loss_clip": 0.01104588,
      "auxiliary_loss_mlp": 0.01030224,
      "balance_loss_clip": 1.03608727,
      "balance_loss_mlp": 1.01801646,
      "epoch": 0.7111979558094093,
      "flos": 27271820309760.0,
      "grad_norm": 1.6134600151247893,
      "language_loss": 0.7210629,
      "learning_rate": 8.128334508009846e-07,
      "loss": 0.74241096,
      "num_input_tokens_seen": 255178920,
      "step": 11829,
      "time_per_iteration": 2.512749433517456
    },
    {
      "auxiliary_loss_clip": 0.0110395,
      "auxiliary_loss_mlp": 0.01030233,
      "balance_loss_clip": 1.03586268,
      "balance_loss_mlp": 1.01892567,
      "epoch": 0.7112580790620773,
      "flos": 25046938961280.0,
      "grad_norm": 2.2646881047557756,
      "language_loss": 0.80222052,
      "learning_rate": 8.125200452317697e-07,
      "loss": 0.82356232,
      "num_input_tokens_seen": 255198095,
      "step": 11830,
      "time_per_iteration": 2.496009111404419
    },
    {
      "auxiliary_loss_clip": 0.01092872,
      "auxiliary_loss_mlp": 0.01034766,
      "balance_loss_clip": 1.03572667,
      "balance_loss_mlp": 1.02273822,
      "epoch": 0.7113182023147452,
      "flos": 21646628951040.0,
      "grad_norm": 1.874368782057851,
      "language_loss": 0.84069848,
      "learning_rate": 8.122066846919138e-07,
      "loss": 0.86197484,
      "num_input_tokens_seen": 255215860,
      "step": 11831,
      "time_per_iteration": 2.5103979110717773
    },
    {
      "auxiliary_loss_clip": 0.01084066,
      "auxiliary_loss_mlp": 0.01028375,
      "balance_loss_clip": 1.03553617,
      "balance_loss_mlp": 1.01665688,
      "epoch": 0.7113783255674132,
      "flos": 20996287257600.0,
      "grad_norm": 2.4048252376882253,
      "language_loss": 0.7734983,
      "learning_rate": 8.118933691932985e-07,
      "loss": 0.79462272,
      "num_input_tokens_seen": 255235425,
      "step": 11832,
      "time_per_iteration": 2.522078514099121
    },
    {
      "auxiliary_loss_clip": 0.01026809,
      "auxiliary_loss_mlp": 0.01002407,
      "balance_loss_clip": 1.01400328,
      "balance_loss_mlp": 1.00123906,
      "epoch": 0.7114384488200811,
      "flos": 66771080161920.0,
      "grad_norm": 0.7473647088820893,
      "language_loss": 0.56620133,
      "learning_rate": 8.115800987478059e-07,
      "loss": 0.58649349,
      "num_input_tokens_seen": 255291680,
      "step": 11833,
      "time_per_iteration": 3.0380194187164307
    },
    {
      "auxiliary_loss_clip": 0.0105636,
      "auxiliary_loss_mlp": 0.01032075,
      "balance_loss_clip": 1.03495419,
      "balance_loss_mlp": 1.02041054,
      "epoch": 0.7114985720727491,
      "flos": 25010058672000.0,
      "grad_norm": 1.8646495189767185,
      "language_loss": 0.70823872,
      "learning_rate": 8.11266873367315e-07,
      "loss": 0.72912312,
      "num_input_tokens_seen": 255313880,
      "step": 11834,
      "time_per_iteration": 2.6403307914733887
    },
    {
      "auxiliary_loss_clip": 0.01107827,
      "auxiliary_loss_mlp": 0.01031694,
      "balance_loss_clip": 1.03704464,
      "balance_loss_mlp": 1.01949906,
      "epoch": 0.7115586953254172,
      "flos": 21470128496640.0,
      "grad_norm": 1.861218453935047,
      "language_loss": 0.79392874,
      "learning_rate": 8.10953693063704e-07,
      "loss": 0.81532401,
      "num_input_tokens_seen": 255332390,
      "step": 11835,
      "time_per_iteration": 2.4613046646118164
    },
    {
      "auxiliary_loss_clip": 0.01092018,
      "auxiliary_loss_mlp": 0.01029997,
      "balance_loss_clip": 1.0357089,
      "balance_loss_mlp": 1.01895785,
      "epoch": 0.7116188185780851,
      "flos": 28622600190720.0,
      "grad_norm": 1.4523913423713999,
      "language_loss": 0.75679755,
      "learning_rate": 8.10640557848848e-07,
      "loss": 0.7780177,
      "num_input_tokens_seen": 255354025,
      "step": 11836,
      "time_per_iteration": 2.571061611175537
    },
    {
      "auxiliary_loss_clip": 0.01034412,
      "auxiliary_loss_mlp": 0.01029725,
      "balance_loss_clip": 1.03492141,
      "balance_loss_mlp": 1.01766098,
      "epoch": 0.7116789418307531,
      "flos": 25293608634240.0,
      "grad_norm": 1.6158199173897911,
      "language_loss": 0.70127606,
      "learning_rate": 8.103274677346208e-07,
      "loss": 0.72191739,
      "num_input_tokens_seen": 255371400,
      "step": 11837,
      "time_per_iteration": 2.6663031578063965
    },
    {
      "auxiliary_loss_clip": 0.01098959,
      "auxiliary_loss_mlp": 0.0103757,
      "balance_loss_clip": 1.03818679,
      "balance_loss_mlp": 1.02466011,
      "epoch": 0.711739065083421,
      "flos": 25557301353600.0,
      "grad_norm": 12.682156190045957,
      "language_loss": 0.61593491,
      "learning_rate": 8.100144227328958e-07,
      "loss": 0.63730025,
      "num_input_tokens_seen": 255390710,
      "step": 11838,
      "time_per_iteration": 2.543238401412964
    },
    {
      "auxiliary_loss_clip": 0.01094955,
      "auxiliary_loss_mlp": 0.01029894,
      "balance_loss_clip": 1.03737485,
      "balance_loss_mlp": 1.01805031,
      "epoch": 0.711799188336089,
      "flos": 26140993361280.0,
      "grad_norm": 2.22765909212795,
      "language_loss": 0.68396914,
      "learning_rate": 8.097014228555426e-07,
      "loss": 0.7052176,
      "num_input_tokens_seen": 255408790,
      "step": 11839,
      "time_per_iteration": 2.5329294204711914
    },
    {
      "auxiliary_loss_clip": 0.0110666,
      "auxiliary_loss_mlp": 0.01032533,
      "balance_loss_clip": 1.03773117,
      "balance_loss_mlp": 1.02114224,
      "epoch": 0.7118593115887569,
      "flos": 21140648017920.0,
      "grad_norm": 1.933846206407439,
      "language_loss": 0.84085697,
      "learning_rate": 8.093884681144305e-07,
      "loss": 0.8622489,
      "num_input_tokens_seen": 255426280,
      "step": 11840,
      "time_per_iteration": 2.4671452045440674
    },
    {
      "auxiliary_loss_clip": 0.01083932,
      "auxiliary_loss_mlp": 0.01029337,
      "balance_loss_clip": 1.03714299,
      "balance_loss_mlp": 1.01730323,
      "epoch": 0.711919434841425,
      "flos": 14975684006400.0,
      "grad_norm": 1.866741654406252,
      "language_loss": 0.7684772,
      "learning_rate": 8.090755585214277e-07,
      "loss": 0.78960991,
      "num_input_tokens_seen": 255442935,
      "step": 11841,
      "time_per_iteration": 2.4922056198120117
    },
    {
      "auxiliary_loss_clip": 0.01089062,
      "auxiliary_loss_mlp": 0.01033742,
      "balance_loss_clip": 1.03655291,
      "balance_loss_mlp": 1.0213865,
      "epoch": 0.7119795580940929,
      "flos": 16508997826560.0,
      "grad_norm": 2.00249006511256,
      "language_loss": 0.75072569,
      "learning_rate": 8.087626940883994e-07,
      "loss": 0.7719537,
      "num_input_tokens_seen": 255460925,
      "step": 11842,
      "time_per_iteration": 2.5219979286193848
    },
    {
      "auxiliary_loss_clip": 0.01031893,
      "auxiliary_loss_mlp": 0.01004313,
      "balance_loss_clip": 1.02496755,
      "balance_loss_mlp": 1.0026679,
      "epoch": 0.7120396813467609,
      "flos": 66570736055040.0,
      "grad_norm": 0.7894660375181115,
      "language_loss": 0.61636162,
      "learning_rate": 8.084498748272082e-07,
      "loss": 0.6367237,
      "num_input_tokens_seen": 255521360,
      "step": 11843,
      "time_per_iteration": 3.072843551635742
    },
    {
      "auxiliary_loss_clip": 0.01104189,
      "auxiliary_loss_mlp": 0.01026271,
      "balance_loss_clip": 1.03663766,
      "balance_loss_mlp": 1.01468945,
      "epoch": 0.7120998045994288,
      "flos": 26432731624320.0,
      "grad_norm": 1.7311346008422905,
      "language_loss": 0.80057162,
      "learning_rate": 8.081371007497171e-07,
      "loss": 0.82187617,
      "num_input_tokens_seen": 255541435,
      "step": 11844,
      "time_per_iteration": 2.513951063156128
    },
    {
      "auxiliary_loss_clip": 0.01054041,
      "auxiliary_loss_mlp": 0.01029306,
      "balance_loss_clip": 1.0336287,
      "balance_loss_mlp": 1.01640785,
      "epoch": 0.7121599278520968,
      "flos": 16427982700800.0,
      "grad_norm": 2.303612782063063,
      "language_loss": 0.78724492,
      "learning_rate": 8.078243718677873e-07,
      "loss": 0.80807841,
      "num_input_tokens_seen": 255558505,
      "step": 11845,
      "time_per_iteration": 2.5456557273864746
    },
    {
      "auxiliary_loss_clip": 0.01087584,
      "auxiliary_loss_mlp": 0.01033699,
      "balance_loss_clip": 1.0350672,
      "balance_loss_mlp": 1.02158749,
      "epoch": 0.7122200511047647,
      "flos": 28949889939840.0,
      "grad_norm": 1.8720681526840515,
      "language_loss": 0.7750122,
      "learning_rate": 8.075116881932762e-07,
      "loss": 0.79622501,
      "num_input_tokens_seen": 255577815,
      "step": 11846,
      "time_per_iteration": 2.5881800651550293
    },
    {
      "auxiliary_loss_clip": 0.01095822,
      "auxiliary_loss_mlp": 0.01031201,
      "balance_loss_clip": 1.03732419,
      "balance_loss_mlp": 1.01884484,
      "epoch": 0.7122801743574327,
      "flos": 16471866142080.0,
      "grad_norm": 1.8885479218310322,
      "language_loss": 0.58531195,
      "learning_rate": 8.071990497380421e-07,
      "loss": 0.60658216,
      "num_input_tokens_seen": 255595885,
      "step": 11847,
      "time_per_iteration": 2.4930670261383057
    },
    {
      "auxiliary_loss_clip": 0.01090091,
      "auxiliary_loss_mlp": 0.00784673,
      "balance_loss_clip": 1.03572762,
      "balance_loss_mlp": 1.0121789,
      "epoch": 0.7123402976101008,
      "flos": 20631039811200.0,
      "grad_norm": 1.543318032956219,
      "language_loss": 0.71290123,
      "learning_rate": 8.068864565139395e-07,
      "loss": 0.7316488,
      "num_input_tokens_seen": 255616750,
      "step": 11848,
      "time_per_iteration": 2.5611441135406494
    },
    {
      "auxiliary_loss_clip": 0.01026213,
      "auxiliary_loss_mlp": 0.0100091,
      "balance_loss_clip": 1.01165271,
      "balance_loss_mlp": 0.9998129,
      "epoch": 0.7124004208627687,
      "flos": 62325734837760.0,
      "grad_norm": 0.8457736018533026,
      "language_loss": 0.62973934,
      "learning_rate": 8.065739085328211e-07,
      "loss": 0.65001059,
      "num_input_tokens_seen": 255677900,
      "step": 11849,
      "time_per_iteration": 3.0652496814727783
    },
    {
      "auxiliary_loss_clip": 0.01078561,
      "auxiliary_loss_mlp": 0.01033119,
      "balance_loss_clip": 1.03540254,
      "balance_loss_mlp": 1.02107918,
      "epoch": 0.7124605441154367,
      "flos": 39675975788160.0,
      "grad_norm": 1.4900401305365265,
      "language_loss": 0.64076561,
      "learning_rate": 8.0626140580654e-07,
      "loss": 0.6618824,
      "num_input_tokens_seen": 255699140,
      "step": 11850,
      "time_per_iteration": 2.6940503120422363
    },
    {
      "auxiliary_loss_clip": 0.01095926,
      "auxiliary_loss_mlp": 0.01029669,
      "balance_loss_clip": 1.03738248,
      "balance_loss_mlp": 1.01761723,
      "epoch": 0.7125206673681046,
      "flos": 28181868312960.0,
      "grad_norm": 1.5865269787958802,
      "language_loss": 0.70058656,
      "learning_rate": 8.05948948346946e-07,
      "loss": 0.72184247,
      "num_input_tokens_seen": 255719640,
      "step": 11851,
      "time_per_iteration": 2.5767931938171387
    },
    {
      "auxiliary_loss_clip": 0.01094576,
      "auxiliary_loss_mlp": 0.01033952,
      "balance_loss_clip": 1.03691983,
      "balance_loss_mlp": 1.02331805,
      "epoch": 0.7125807906207726,
      "flos": 26176939896960.0,
      "grad_norm": 1.447539260732319,
      "language_loss": 0.83216786,
      "learning_rate": 8.056365361658882e-07,
      "loss": 0.85345322,
      "num_input_tokens_seen": 255740450,
      "step": 11852,
      "time_per_iteration": 2.5376980304718018
    },
    {
      "auxiliary_loss_clip": 0.01094221,
      "auxiliary_loss_mlp": 0.00784218,
      "balance_loss_clip": 1.03471673,
      "balance_loss_mlp": 1.00823605,
      "epoch": 0.7126409138734405,
      "flos": 17157328358400.0,
      "grad_norm": 2.247147444011058,
      "language_loss": 0.73505586,
      "learning_rate": 8.053241692752126e-07,
      "loss": 0.75384027,
      "num_input_tokens_seen": 255758070,
      "step": 11853,
      "time_per_iteration": 2.4619317054748535
    },
    {
      "auxiliary_loss_clip": 0.01064247,
      "auxiliary_loss_mlp": 0.01034341,
      "balance_loss_clip": 1.03320742,
      "balance_loss_mlp": 1.02280164,
      "epoch": 0.7127010371261085,
      "flos": 18769933451520.0,
      "grad_norm": 2.0096426133941754,
      "language_loss": 0.92386901,
      "learning_rate": 8.050118476867635e-07,
      "loss": 0.94485486,
      "num_input_tokens_seen": 255775685,
      "step": 11854,
      "time_per_iteration": 3.9513845443725586
    },
    {
      "auxiliary_loss_clip": 0.01092261,
      "auxiliary_loss_mlp": 0.01029316,
      "balance_loss_clip": 1.03584743,
      "balance_loss_mlp": 1.01803911,
      "epoch": 0.7127611603787765,
      "flos": 20376433232640.0,
      "grad_norm": 1.8176076335420885,
      "language_loss": 0.79688942,
      "learning_rate": 8.046995714123856e-07,
      "loss": 0.81810522,
      "num_input_tokens_seen": 255794750,
      "step": 11855,
      "time_per_iteration": 2.499260663986206
    },
    {
      "auxiliary_loss_clip": 0.01057293,
      "auxiliary_loss_mlp": 0.010327,
      "balance_loss_clip": 1.03393352,
      "balance_loss_mlp": 1.01972425,
      "epoch": 0.7128212836314445,
      "flos": 20449008662400.0,
      "grad_norm": 1.8737781497438626,
      "language_loss": 0.72817171,
      "learning_rate": 8.043873404639192e-07,
      "loss": 0.74907172,
      "num_input_tokens_seen": 255813325,
      "step": 11856,
      "time_per_iteration": 2.576819658279419
    },
    {
      "auxiliary_loss_clip": 0.01096066,
      "auxiliary_loss_mlp": 0.01029329,
      "balance_loss_clip": 1.03684235,
      "balance_loss_mlp": 1.01756287,
      "epoch": 0.7128814068841124,
      "flos": 23440834229760.0,
      "grad_norm": 1.6067167456492044,
      "language_loss": 0.69833457,
      "learning_rate": 8.040751548532046e-07,
      "loss": 0.71958852,
      "num_input_tokens_seen": 255832470,
      "step": 11857,
      "time_per_iteration": 2.524803638458252
    },
    {
      "auxiliary_loss_clip": 0.01091206,
      "auxiliary_loss_mlp": 0.01027345,
      "balance_loss_clip": 1.03372598,
      "balance_loss_mlp": 1.01538277,
      "epoch": 0.7129415301367804,
      "flos": 18222942165120.0,
      "grad_norm": 2.691411406228035,
      "language_loss": 0.84989041,
      "learning_rate": 8.03763014592081e-07,
      "loss": 0.87107587,
      "num_input_tokens_seen": 255849740,
      "step": 11858,
      "time_per_iteration": 2.4561057090759277
    },
    {
      "auxiliary_loss_clip": 0.01111246,
      "auxiliary_loss_mlp": 0.01031432,
      "balance_loss_clip": 1.03905129,
      "balance_loss_mlp": 1.0192368,
      "epoch": 0.7130016533894483,
      "flos": 15523896355200.0,
      "grad_norm": 1.694803481926178,
      "language_loss": 0.80377996,
      "learning_rate": 8.034509196923829e-07,
      "loss": 0.82520676,
      "num_input_tokens_seen": 255866975,
      "step": 11859,
      "time_per_iteration": 3.843292236328125
    },
    {
      "auxiliary_loss_clip": 0.01080829,
      "auxiliary_loss_mlp": 0.01030176,
      "balance_loss_clip": 1.03532851,
      "balance_loss_mlp": 1.01836848,
      "epoch": 0.7130617766421163,
      "flos": 57115668960000.0,
      "grad_norm": 1.1831192337888181,
      "language_loss": 0.68725646,
      "learning_rate": 8.031388701659456e-07,
      "loss": 0.70836651,
      "num_input_tokens_seen": 255892915,
      "step": 11860,
      "time_per_iteration": 4.219373464584351
    },
    {
      "auxiliary_loss_clip": 0.01092856,
      "auxiliary_loss_mlp": 0.01030615,
      "balance_loss_clip": 1.03511167,
      "balance_loss_mlp": 1.01783574,
      "epoch": 0.7131218998947844,
      "flos": 19788252024960.0,
      "grad_norm": 1.9842520926471592,
      "language_loss": 0.64796263,
      "learning_rate": 8.028268660246023e-07,
      "loss": 0.66919732,
      "num_input_tokens_seen": 255911480,
      "step": 11861,
      "time_per_iteration": 2.4956915378570557
    },
    {
      "auxiliary_loss_clip": 0.01089276,
      "auxiliary_loss_mlp": 0.0102885,
      "balance_loss_clip": 1.03841305,
      "balance_loss_mlp": 1.01632655,
      "epoch": 0.7131820231474523,
      "flos": 26651894457600.0,
      "grad_norm": 1.949185867684346,
      "language_loss": 0.67191565,
      "learning_rate": 8.025149072801849e-07,
      "loss": 0.69309688,
      "num_input_tokens_seen": 255931140,
      "step": 11862,
      "time_per_iteration": 2.5770859718322754
    },
    {
      "auxiliary_loss_clip": 0.01079411,
      "auxiliary_loss_mlp": 0.01035825,
      "balance_loss_clip": 1.03638506,
      "balance_loss_mlp": 1.0248282,
      "epoch": 0.7132421464001203,
      "flos": 29205609840000.0,
      "grad_norm": 1.8990135900784915,
      "language_loss": 0.66636598,
      "learning_rate": 8.022029939445214e-07,
      "loss": 0.6875183,
      "num_input_tokens_seen": 255951665,
      "step": 11863,
      "time_per_iteration": 2.5878512859344482
    },
    {
      "auxiliary_loss_clip": 0.0106513,
      "auxiliary_loss_mlp": 0.01040531,
      "balance_loss_clip": 1.03751254,
      "balance_loss_mlp": 1.02701211,
      "epoch": 0.7133022696527882,
      "flos": 23073611535360.0,
      "grad_norm": 2.3910207723614754,
      "language_loss": 0.65615326,
      "learning_rate": 8.018911260294414e-07,
      "loss": 0.67720985,
      "num_input_tokens_seen": 255970055,
      "step": 11864,
      "time_per_iteration": 2.5974202156066895
    },
    {
      "auxiliary_loss_clip": 0.01097885,
      "auxiliary_loss_mlp": 0.010326,
      "balance_loss_clip": 1.03635299,
      "balance_loss_mlp": 1.02013707,
      "epoch": 0.7133623929054562,
      "flos": 17457111267840.0,
      "grad_norm": 2.0563613836936465,
      "language_loss": 0.85554898,
      "learning_rate": 8.015793035467697e-07,
      "loss": 0.87685382,
      "num_input_tokens_seen": 255987720,
      "step": 11865,
      "time_per_iteration": 3.9420690536499023
    },
    {
      "auxiliary_loss_clip": 0.01066798,
      "auxiliary_loss_mlp": 0.01031966,
      "balance_loss_clip": 1.03292727,
      "balance_loss_mlp": 1.01873946,
      "epoch": 0.7134225161581241,
      "flos": 19536554448000.0,
      "grad_norm": 2.0721396734448825,
      "language_loss": 0.74694479,
      "learning_rate": 8.012675265083304e-07,
      "loss": 0.76793242,
      "num_input_tokens_seen": 256005490,
      "step": 11866,
      "time_per_iteration": 2.534437656402588
    },
    {
      "auxiliary_loss_clip": 0.01077016,
      "auxiliary_loss_mlp": 0.01030303,
      "balance_loss_clip": 1.03772712,
      "balance_loss_mlp": 1.0176909,
      "epoch": 0.7134826394107922,
      "flos": 26250089944320.0,
      "grad_norm": 3.4130435227312184,
      "language_loss": 0.7049039,
      "learning_rate": 8.009557949259464e-07,
      "loss": 0.72597706,
      "num_input_tokens_seen": 256026030,
      "step": 11867,
      "time_per_iteration": 2.592672348022461
    },
    {
      "auxiliary_loss_clip": 0.01092016,
      "auxiliary_loss_mlp": 0.01027192,
      "balance_loss_clip": 1.0363214,
      "balance_loss_mlp": 1.01621854,
      "epoch": 0.7135427626634601,
      "flos": 15815311395840.0,
      "grad_norm": 4.665538270106113,
      "language_loss": 0.72006929,
      "learning_rate": 8.006441088114397e-07,
      "loss": 0.74126142,
      "num_input_tokens_seen": 256043680,
      "step": 11868,
      "time_per_iteration": 2.4610776901245117
    },
    {
      "auxiliary_loss_clip": 0.01058804,
      "auxiliary_loss_mlp": 0.01030942,
      "balance_loss_clip": 1.03747511,
      "balance_loss_mlp": 1.01667285,
      "epoch": 0.7136028859161281,
      "flos": 18223409041920.0,
      "grad_norm": 2.2766069515922123,
      "language_loss": 0.66072744,
      "learning_rate": 8.003324681766286e-07,
      "loss": 0.68162495,
      "num_input_tokens_seen": 256059705,
      "step": 11869,
      "time_per_iteration": 2.5713844299316406
    },
    {
      "auxiliary_loss_clip": 0.01079261,
      "auxiliary_loss_mlp": 0.01023946,
      "balance_loss_clip": 1.0330894,
      "balance_loss_mlp": 1.01216269,
      "epoch": 0.713663009168796,
      "flos": 24314827956480.0,
      "grad_norm": 1.6070053236537025,
      "language_loss": 0.77950412,
      "learning_rate": 8.000208730333298e-07,
      "loss": 0.80053616,
      "num_input_tokens_seen": 256079785,
      "step": 11870,
      "time_per_iteration": 2.574803352355957
    },
    {
      "auxiliary_loss_clip": 0.01056852,
      "auxiliary_loss_mlp": 0.01031798,
      "balance_loss_clip": 1.03575695,
      "balance_loss_mlp": 1.01929331,
      "epoch": 0.713723132421464,
      "flos": 26538488242560.0,
      "grad_norm": 1.6566413940146572,
      "language_loss": 0.8109538,
      "learning_rate": 7.997093233933597e-07,
      "loss": 0.83184028,
      "num_input_tokens_seen": 256099000,
      "step": 11871,
      "time_per_iteration": 2.632986307144165
    },
    {
      "auxiliary_loss_clip": 0.01073702,
      "auxiliary_loss_mlp": 0.01036256,
      "balance_loss_clip": 1.03506637,
      "balance_loss_mlp": 1.02367902,
      "epoch": 0.7137832556741319,
      "flos": 19865675790720.0,
      "grad_norm": 1.6219934602651103,
      "language_loss": 0.78686517,
      "learning_rate": 7.993978192685331e-07,
      "loss": 0.8079648,
      "num_input_tokens_seen": 256117985,
      "step": 11872,
      "time_per_iteration": 2.5503997802734375
    },
    {
      "auxiliary_loss_clip": 0.01095935,
      "auxiliary_loss_mlp": 0.0102694,
      "balance_loss_clip": 1.03575158,
      "balance_loss_mlp": 1.01429737,
      "epoch": 0.7138433789267999,
      "flos": 21688932193920.0,
      "grad_norm": 2.447171832310374,
      "language_loss": 0.83954954,
      "learning_rate": 7.990863606706606e-07,
      "loss": 0.86077827,
      "num_input_tokens_seen": 256134350,
      "step": 11873,
      "time_per_iteration": 2.5009546279907227
    },
    {
      "auxiliary_loss_clip": 0.01065996,
      "auxiliary_loss_mlp": 0.01031953,
      "balance_loss_clip": 1.03228652,
      "balance_loss_mlp": 1.02116418,
      "epoch": 0.713903502179468,
      "flos": 17602729004160.0,
      "grad_norm": 1.9116131856114693,
      "language_loss": 0.85802627,
      "learning_rate": 7.987749476115539e-07,
      "loss": 0.87900573,
      "num_input_tokens_seen": 256150610,
      "step": 11874,
      "time_per_iteration": 2.5309700965881348
    },
    {
      "auxiliary_loss_clip": 0.01094896,
      "auxiliary_loss_mlp": 0.01029862,
      "balance_loss_clip": 1.03430212,
      "balance_loss_mlp": 1.01755941,
      "epoch": 0.7139636254321359,
      "flos": 18040336398720.0,
      "grad_norm": 1.832195227066355,
      "language_loss": 0.83300328,
      "learning_rate": 7.984635801030228e-07,
      "loss": 0.85425079,
      "num_input_tokens_seen": 256168620,
      "step": 11875,
      "time_per_iteration": 2.4892821311950684
    },
    {
      "auxiliary_loss_clip": 0.01085866,
      "auxiliary_loss_mlp": 0.01040109,
      "balance_loss_clip": 1.03569484,
      "balance_loss_mlp": 1.02513015,
      "epoch": 0.7140237486848039,
      "flos": 23331127115520.0,
      "grad_norm": 1.913690450430302,
      "language_loss": 0.69006503,
      "learning_rate": 7.981522581568721e-07,
      "loss": 0.71132481,
      "num_input_tokens_seen": 256186700,
      "step": 11876,
      "time_per_iteration": 2.5290708541870117
    },
    {
      "auxiliary_loss_clip": 0.01108248,
      "auxiliary_loss_mlp": 0.01036059,
      "balance_loss_clip": 1.03723371,
      "balance_loss_mlp": 1.02377415,
      "epoch": 0.7140838719374718,
      "flos": 16837077674880.0,
      "grad_norm": 1.914859379315155,
      "language_loss": 0.77591717,
      "learning_rate": 7.978409817849079e-07,
      "loss": 0.79736018,
      "num_input_tokens_seen": 256205390,
      "step": 11877,
      "time_per_iteration": 2.4598898887634277
    },
    {
      "auxiliary_loss_clip": 0.01093202,
      "auxiliary_loss_mlp": 0.01030669,
      "balance_loss_clip": 1.03598809,
      "balance_loss_mlp": 1.01986909,
      "epoch": 0.7141439951901398,
      "flos": 21142012734720.0,
      "grad_norm": 2.2572516141584784,
      "language_loss": 0.69643414,
      "learning_rate": 7.97529750998934e-07,
      "loss": 0.71767288,
      "num_input_tokens_seen": 256224575,
      "step": 11878,
      "time_per_iteration": 2.491856575012207
    },
    {
      "auxiliary_loss_clip": 0.01069122,
      "auxiliary_loss_mlp": 0.01031759,
      "balance_loss_clip": 1.03823733,
      "balance_loss_mlp": 1.02105379,
      "epoch": 0.7142041184428077,
      "flos": 24717709877760.0,
      "grad_norm": 2.0550238121408735,
      "language_loss": 0.67554271,
      "learning_rate": 7.972185658107535e-07,
      "loss": 0.69655156,
      "num_input_tokens_seen": 256242130,
      "step": 11879,
      "time_per_iteration": 2.6008684635162354
    },
    {
      "auxiliary_loss_clip": 0.01054414,
      "auxiliary_loss_mlp": 0.0103855,
      "balance_loss_clip": 1.0356257,
      "balance_loss_mlp": 1.02488256,
      "epoch": 0.7142642416954758,
      "flos": 21908202768000.0,
      "grad_norm": 1.7002622285179256,
      "language_loss": 0.69038773,
      "learning_rate": 7.969074262321646e-07,
      "loss": 0.71131736,
      "num_input_tokens_seen": 256261920,
      "step": 11880,
      "time_per_iteration": 2.5862131118774414
    },
    {
      "auxiliary_loss_clip": 0.01079027,
      "auxiliary_loss_mlp": 0.0103795,
      "balance_loss_clip": 1.03326225,
      "balance_loss_mlp": 1.02536178,
      "epoch": 0.7143243649481437,
      "flos": 20805636844800.0,
      "grad_norm": 2.0631736125291624,
      "language_loss": 0.80819881,
      "learning_rate": 7.965963322749674e-07,
      "loss": 0.82936859,
      "num_input_tokens_seen": 256277970,
      "step": 11881,
      "time_per_iteration": 2.528226613998413
    },
    {
      "auxiliary_loss_clip": 0.01073091,
      "auxiliary_loss_mlp": 0.01030609,
      "balance_loss_clip": 1.03494263,
      "balance_loss_mlp": 1.01909375,
      "epoch": 0.7143844882008117,
      "flos": 27235011847680.0,
      "grad_norm": 1.7965439155116674,
      "language_loss": 0.63635337,
      "learning_rate": 7.962852839509579e-07,
      "loss": 0.65739036,
      "num_input_tokens_seen": 256298205,
      "step": 11882,
      "time_per_iteration": 2.581442356109619
    },
    {
      "auxiliary_loss_clip": 0.01108301,
      "auxiliary_loss_mlp": 0.0103197,
      "balance_loss_clip": 1.03696358,
      "balance_loss_mlp": 1.0202632,
      "epoch": 0.7144446114534796,
      "flos": 17929623703680.0,
      "grad_norm": 1.8266771586439765,
      "language_loss": 0.68850267,
      "learning_rate": 7.959742812719304e-07,
      "loss": 0.70990539,
      "num_input_tokens_seen": 256316685,
      "step": 11883,
      "time_per_iteration": 2.4749042987823486
    },
    {
      "auxiliary_loss_clip": 0.01094861,
      "auxiliary_loss_mlp": 0.0103434,
      "balance_loss_clip": 1.03595853,
      "balance_loss_mlp": 1.0211314,
      "epoch": 0.7145047347061476,
      "flos": 20740962407040.0,
      "grad_norm": 2.0740513718359,
      "language_loss": 0.77720529,
      "learning_rate": 7.956633242496788e-07,
      "loss": 0.79849732,
      "num_input_tokens_seen": 256334205,
      "step": 11884,
      "time_per_iteration": 2.484001874923706
    },
    {
      "auxiliary_loss_clip": 0.01101532,
      "auxiliary_loss_mlp": 0.01032463,
      "balance_loss_clip": 1.03768301,
      "balance_loss_mlp": 1.0182296,
      "epoch": 0.7145648579588155,
      "flos": 21178605715200.0,
      "grad_norm": 1.990302923190699,
      "language_loss": 0.73748326,
      "learning_rate": 7.953524128959954e-07,
      "loss": 0.75882322,
      "num_input_tokens_seen": 256353340,
      "step": 11885,
      "time_per_iteration": 2.492763042449951
    },
    {
      "auxiliary_loss_clip": 0.0101581,
      "auxiliary_loss_mlp": 0.01002063,
      "balance_loss_clip": 1.01223159,
      "balance_loss_mlp": 1.00094819,
      "epoch": 0.7146249812114835,
      "flos": 64784539509120.0,
      "grad_norm": 0.8919310366882793,
      "language_loss": 0.66400403,
      "learning_rate": 7.95041547222669e-07,
      "loss": 0.68418276,
      "num_input_tokens_seen": 256411550,
      "step": 11886,
      "time_per_iteration": 3.0921013355255127
    },
    {
      "auxiliary_loss_clip": 0.01064035,
      "auxiliary_loss_mlp": 0.0102448,
      "balance_loss_clip": 1.0365144,
      "balance_loss_mlp": 1.01279795,
      "epoch": 0.7146851044641516,
      "flos": 18113881495680.0,
      "grad_norm": 1.7895413634252861,
      "language_loss": 0.7466644,
      "learning_rate": 7.947307272414874e-07,
      "loss": 0.76754957,
      "num_input_tokens_seen": 256430360,
      "step": 11887,
      "time_per_iteration": 2.578949213027954
    },
    {
      "auxiliary_loss_clip": 0.01094653,
      "auxiliary_loss_mlp": 0.01026772,
      "balance_loss_clip": 1.03624165,
      "balance_loss_mlp": 1.01498199,
      "epoch": 0.7147452277168195,
      "flos": 19243846517760.0,
      "grad_norm": 1.5797823712738683,
      "language_loss": 0.71481609,
      "learning_rate": 7.944199529642372e-07,
      "loss": 0.73603034,
      "num_input_tokens_seen": 256449750,
      "step": 11888,
      "time_per_iteration": 2.4761970043182373
    },
    {
      "auxiliary_loss_clip": 0.01090796,
      "auxiliary_loss_mlp": 0.01034195,
      "balance_loss_clip": 1.03321862,
      "balance_loss_mlp": 1.02160621,
      "epoch": 0.7148053509694875,
      "flos": 23764712186880.0,
      "grad_norm": 1.8447439981754719,
      "language_loss": 0.84303319,
      "learning_rate": 7.941092244027041e-07,
      "loss": 0.86428314,
      "num_input_tokens_seen": 256467330,
      "step": 11889,
      "time_per_iteration": 2.523054361343384
    },
    {
      "auxiliary_loss_clip": 0.01057389,
      "auxiliary_loss_mlp": 0.01023255,
      "balance_loss_clip": 1.03502536,
      "balance_loss_mlp": 1.01141715,
      "epoch": 0.7148654742221554,
      "flos": 22485322586880.0,
      "grad_norm": 1.8364966417839728,
      "language_loss": 0.75516546,
      "learning_rate": 7.937985415686695e-07,
      "loss": 0.77597183,
      "num_input_tokens_seen": 256485705,
      "step": 11890,
      "time_per_iteration": 2.614504337310791
    },
    {
      "auxiliary_loss_clip": 0.01065813,
      "auxiliary_loss_mlp": 0.01030742,
      "balance_loss_clip": 1.03362501,
      "balance_loss_mlp": 1.01907182,
      "epoch": 0.7149255974748234,
      "flos": 24679213476480.0,
      "grad_norm": 1.535683416527105,
      "language_loss": 0.74108684,
      "learning_rate": 7.934879044739147e-07,
      "loss": 0.76205242,
      "num_input_tokens_seen": 256504755,
      "step": 11891,
      "time_per_iteration": 2.602437973022461
    },
    {
      "auxiliary_loss_clip": 0.01063185,
      "auxiliary_loss_mlp": 0.01032125,
      "balance_loss_clip": 1.03751791,
      "balance_loss_mlp": 1.02013874,
      "epoch": 0.7149857207274913,
      "flos": 18405583845120.0,
      "grad_norm": 1.8894274865163574,
      "language_loss": 0.67594242,
      "learning_rate": 7.931773131302211e-07,
      "loss": 0.69689548,
      "num_input_tokens_seen": 256523670,
      "step": 11892,
      "time_per_iteration": 2.5746917724609375
    },
    {
      "auxiliary_loss_clip": 0.01071957,
      "auxiliary_loss_mlp": 0.01031614,
      "balance_loss_clip": 1.03711224,
      "balance_loss_mlp": 1.0182035,
      "epoch": 0.7150458439801594,
      "flos": 24969515195520.0,
      "grad_norm": 2.0549812982612794,
      "language_loss": 0.74275911,
      "learning_rate": 7.928667675493632e-07,
      "loss": 0.76379478,
      "num_input_tokens_seen": 256542225,
      "step": 11893,
      "time_per_iteration": 3.954145908355713
    },
    {
      "auxiliary_loss_clip": 0.0111004,
      "auxiliary_loss_mlp": 0.01029979,
      "balance_loss_clip": 1.03733444,
      "balance_loss_mlp": 1.01714611,
      "epoch": 0.7151059672328273,
      "flos": 16690777580160.0,
      "grad_norm": 2.2244024088954855,
      "language_loss": 0.66541582,
      "learning_rate": 7.925562677431185e-07,
      "loss": 0.68681604,
      "num_input_tokens_seen": 256560730,
      "step": 11894,
      "time_per_iteration": 2.4796295166015625
    },
    {
      "auxiliary_loss_clip": 0.01072546,
      "auxiliary_loss_mlp": 0.01030943,
      "balance_loss_clip": 1.03764808,
      "balance_loss_mlp": 1.0192188,
      "epoch": 0.7151660904854953,
      "flos": 27271820309760.0,
      "grad_norm": 1.780738127185801,
      "language_loss": 0.7779547,
      "learning_rate": 7.922458137232613e-07,
      "loss": 0.79898959,
      "num_input_tokens_seen": 256580505,
      "step": 11895,
      "time_per_iteration": 2.6147029399871826
    },
    {
      "auxiliary_loss_clip": 0.0109602,
      "auxiliary_loss_mlp": 0.01032251,
      "balance_loss_clip": 1.03577352,
      "balance_loss_mlp": 1.01910257,
      "epoch": 0.7152262137381632,
      "flos": 18332254229760.0,
      "grad_norm": 1.7696653270613458,
      "language_loss": 0.69489056,
      "learning_rate": 7.919354055015643e-07,
      "loss": 0.71617329,
      "num_input_tokens_seen": 256597330,
      "step": 11896,
      "time_per_iteration": 2.481764554977417
    },
    {
      "auxiliary_loss_clip": 0.0108372,
      "auxiliary_loss_mlp": 0.01043568,
      "balance_loss_clip": 1.03410304,
      "balance_loss_mlp": 1.03040719,
      "epoch": 0.7152863369908312,
      "flos": 21799285752960.0,
      "grad_norm": 1.7559405298288877,
      "language_loss": 0.86517429,
      "learning_rate": 7.91625043089798e-07,
      "loss": 0.88644719,
      "num_input_tokens_seen": 256616030,
      "step": 11897,
      "time_per_iteration": 2.5246143341064453
    },
    {
      "auxiliary_loss_clip": 0.0108311,
      "auxiliary_loss_mlp": 0.01031906,
      "balance_loss_clip": 1.03514028,
      "balance_loss_mlp": 1.02007425,
      "epoch": 0.7153464602434991,
      "flos": 22158427887360.0,
      "grad_norm": 2.184864142292303,
      "language_loss": 0.78063875,
      "learning_rate": 7.913147264997304e-07,
      "loss": 0.80178893,
      "num_input_tokens_seen": 256635570,
      "step": 11898,
      "time_per_iteration": 3.9438400268554688
    },
    {
      "auxiliary_loss_clip": 0.01084592,
      "auxiliary_loss_mlp": 0.01029775,
      "balance_loss_clip": 1.03584361,
      "balance_loss_mlp": 1.01650119,
      "epoch": 0.7154065834961671,
      "flos": 24716057852160.0,
      "grad_norm": 1.8065302332067816,
      "language_loss": 0.72755057,
      "learning_rate": 7.910044557431302e-07,
      "loss": 0.74869424,
      "num_input_tokens_seen": 256655290,
      "step": 11899,
      "time_per_iteration": 3.915525197982788
    },
    {
      "auxiliary_loss_clip": 0.01095671,
      "auxiliary_loss_mlp": 0.01034876,
      "balance_loss_clip": 1.03631747,
      "balance_loss_mlp": 1.0218997,
      "epoch": 0.7154667067488351,
      "flos": 22601494149120.0,
      "grad_norm": 2.0182848483493445,
      "language_loss": 0.75700736,
      "learning_rate": 7.906942308317614e-07,
      "loss": 0.77831286,
      "num_input_tokens_seen": 256671605,
      "step": 11900,
      "time_per_iteration": 2.495344877243042
    },
    {
      "auxiliary_loss_clip": 0.01096633,
      "auxiliary_loss_mlp": 0.01030882,
      "balance_loss_clip": 1.03725648,
      "balance_loss_mlp": 1.01905012,
      "epoch": 0.7155268300015031,
      "flos": 18771154513920.0,
      "grad_norm": 1.855831343502479,
      "language_loss": 0.80745786,
      "learning_rate": 7.903840517773886e-07,
      "loss": 0.82873297,
      "num_input_tokens_seen": 256689680,
      "step": 11901,
      "time_per_iteration": 2.4897193908691406
    },
    {
      "auxiliary_loss_clip": 0.01075414,
      "auxiliary_loss_mlp": 0.0103534,
      "balance_loss_clip": 1.03567767,
      "balance_loss_mlp": 1.02233982,
      "epoch": 0.7155869532541711,
      "flos": 18296343607680.0,
      "grad_norm": 1.6869585099267845,
      "language_loss": 0.81097108,
      "learning_rate": 7.900739185917744e-07,
      "loss": 0.8320787,
      "num_input_tokens_seen": 256707760,
      "step": 11902,
      "time_per_iteration": 2.5315418243408203
    },
    {
      "auxiliary_loss_clip": 0.01064875,
      "auxiliary_loss_mlp": 0.01028391,
      "balance_loss_clip": 1.03545594,
      "balance_loss_mlp": 1.01625574,
      "epoch": 0.715647076506839,
      "flos": 11980805783040.0,
      "grad_norm": 2.1659083018660046,
      "language_loss": 0.68158323,
      "learning_rate": 7.897638312866785e-07,
      "loss": 0.70251596,
      "num_input_tokens_seen": 256724150,
      "step": 11903,
      "time_per_iteration": 2.543220043182373
    },
    {
      "auxiliary_loss_clip": 0.01066087,
      "auxiliary_loss_mlp": 0.01028473,
      "balance_loss_clip": 1.03474689,
      "balance_loss_mlp": 1.01713622,
      "epoch": 0.715707199759507,
      "flos": 18951641377920.0,
      "grad_norm": 2.086185697869419,
      "language_loss": 0.75846875,
      "learning_rate": 7.894537898738589e-07,
      "loss": 0.77941442,
      "num_input_tokens_seen": 256742780,
      "step": 11904,
      "time_per_iteration": 3.9726293087005615
    },
    {
      "auxiliary_loss_clip": 0.01083958,
      "auxiliary_loss_mlp": 0.0103717,
      "balance_loss_clip": 1.03550279,
      "balance_loss_mlp": 1.02415848,
      "epoch": 0.7157673230121749,
      "flos": 15304410299520.0,
      "grad_norm": 1.8797725905315583,
      "language_loss": 0.72079659,
      "learning_rate": 7.891437943650727e-07,
      "loss": 0.74200791,
      "num_input_tokens_seen": 256761355,
      "step": 11905,
      "time_per_iteration": 2.5121965408325195
    },
    {
      "auxiliary_loss_clip": 0.010706,
      "auxiliary_loss_mlp": 0.01030552,
      "balance_loss_clip": 1.0352273,
      "balance_loss_mlp": 1.01890564,
      "epoch": 0.715827446264843,
      "flos": 23221850964480.0,
      "grad_norm": 1.5288461622524643,
      "language_loss": 0.78202069,
      "learning_rate": 7.88833844772076e-07,
      "loss": 0.80303228,
      "num_input_tokens_seen": 256781335,
      "step": 11906,
      "time_per_iteration": 2.589752435684204
    },
    {
      "auxiliary_loss_clip": 0.01017114,
      "auxiliary_loss_mlp": 0.0100149,
      "balance_loss_clip": 1.01495051,
      "balance_loss_mlp": 1.00045335,
      "epoch": 0.7158875695175109,
      "flos": 60975421833600.0,
      "grad_norm": 0.7356091262984021,
      "language_loss": 0.55318624,
      "learning_rate": 7.885239411066205e-07,
      "loss": 0.57337219,
      "num_input_tokens_seen": 256838890,
      "step": 11907,
      "time_per_iteration": 3.05844783782959
    },
    {
      "auxiliary_loss_clip": 0.01088971,
      "auxiliary_loss_mlp": 0.01036521,
      "balance_loss_clip": 1.03285444,
      "balance_loss_mlp": 1.02376008,
      "epoch": 0.7159476927701789,
      "flos": 17128780024320.0,
      "grad_norm": 1.761849679914604,
      "language_loss": 0.69494903,
      "learning_rate": 7.882140833804593e-07,
      "loss": 0.71620393,
      "num_input_tokens_seen": 256858145,
      "step": 11908,
      "time_per_iteration": 2.499251365661621
    },
    {
      "auxiliary_loss_clip": 0.01063117,
      "auxiliary_loss_mlp": 0.01032358,
      "balance_loss_clip": 1.03309298,
      "balance_loss_mlp": 1.01854181,
      "epoch": 0.7160078160228468,
      "flos": 22490601886080.0,
      "grad_norm": 1.6657865461485866,
      "language_loss": 0.71263754,
      "learning_rate": 7.879042716053415e-07,
      "loss": 0.73359239,
      "num_input_tokens_seen": 256878545,
      "step": 11909,
      "time_per_iteration": 2.5751895904541016
    },
    {
      "auxiliary_loss_clip": 0.01095261,
      "auxiliary_loss_mlp": 0.01028239,
      "balance_loss_clip": 1.03943181,
      "balance_loss_mlp": 1.01643777,
      "epoch": 0.7160679392755148,
      "flos": 30590935626240.0,
      "grad_norm": 1.452694043151886,
      "language_loss": 0.7526508,
      "learning_rate": 7.875945057930144e-07,
      "loss": 0.77388579,
      "num_input_tokens_seen": 256899920,
      "step": 11910,
      "time_per_iteration": 2.5689001083374023
    },
    {
      "auxiliary_loss_clip": 0.01080746,
      "auxiliary_loss_mlp": 0.01030019,
      "balance_loss_clip": 1.03654635,
      "balance_loss_mlp": 1.01920056,
      "epoch": 0.7161280625281827,
      "flos": 21323648833920.0,
      "grad_norm": 2.289138568065848,
      "language_loss": 0.7694726,
      "learning_rate": 7.872847859552251e-07,
      "loss": 0.79058027,
      "num_input_tokens_seen": 256918460,
      "step": 11911,
      "time_per_iteration": 2.521012306213379
    },
    {
      "auxiliary_loss_clip": 0.01071177,
      "auxiliary_loss_mlp": 0.0103926,
      "balance_loss_clip": 1.03565681,
      "balance_loss_mlp": 1.02482414,
      "epoch": 0.7161881857808508,
      "flos": 61860078921600.0,
      "grad_norm": 1.96050940371369,
      "language_loss": 0.58668876,
      "learning_rate": 7.869751121037192e-07,
      "loss": 0.60779309,
      "num_input_tokens_seen": 256942015,
      "step": 11912,
      "time_per_iteration": 2.9391982555389404
    },
    {
      "auxiliary_loss_clip": 0.01095743,
      "auxiliary_loss_mlp": 0.01034948,
      "balance_loss_clip": 1.03753757,
      "balance_loss_mlp": 1.02249646,
      "epoch": 0.7162483090335187,
      "flos": 20812101292800.0,
      "grad_norm": 1.8204868442911681,
      "language_loss": 0.78144741,
      "learning_rate": 7.866654842502376e-07,
      "loss": 0.8027544,
      "num_input_tokens_seen": 256961065,
      "step": 11913,
      "time_per_iteration": 2.499769687652588
    },
    {
      "auxiliary_loss_clip": 0.01081474,
      "auxiliary_loss_mlp": 0.01027649,
      "balance_loss_clip": 1.03529704,
      "balance_loss_mlp": 1.01678944,
      "epoch": 0.7163084322861867,
      "flos": 24097532630400.0,
      "grad_norm": 1.6600341929572915,
      "language_loss": 0.74030364,
      "learning_rate": 7.863559024065234e-07,
      "loss": 0.76139492,
      "num_input_tokens_seen": 256982165,
      "step": 11914,
      "time_per_iteration": 2.575164556503296
    },
    {
      "auxiliary_loss_clip": 0.0107013,
      "auxiliary_loss_mlp": 0.0103154,
      "balance_loss_clip": 1.03499794,
      "balance_loss_mlp": 1.01939893,
      "epoch": 0.7163685555388547,
      "flos": 20080888128000.0,
      "grad_norm": 1.8111072853554178,
      "language_loss": 0.74018669,
      "learning_rate": 7.860463665843143e-07,
      "loss": 0.76120341,
      "num_input_tokens_seen": 256999825,
      "step": 11915,
      "time_per_iteration": 2.5423030853271484
    },
    {
      "auxiliary_loss_clip": 0.01105887,
      "auxiliary_loss_mlp": 0.01028124,
      "balance_loss_clip": 1.03528428,
      "balance_loss_mlp": 1.01636422,
      "epoch": 0.7164286787915226,
      "flos": 17456967613440.0,
      "grad_norm": 2.532321004061479,
      "language_loss": 0.81086397,
      "learning_rate": 7.85736876795349e-07,
      "loss": 0.83220398,
      "num_input_tokens_seen": 257017450,
      "step": 11916,
      "time_per_iteration": 2.459557056427002
    },
    {
      "auxiliary_loss_clip": 0.01031688,
      "auxiliary_loss_mlp": 0.01028929,
      "balance_loss_clip": 1.03626668,
      "balance_loss_mlp": 1.01691329,
      "epoch": 0.7164888020441906,
      "flos": 19718908819200.0,
      "grad_norm": 1.8201878594680159,
      "language_loss": 0.68475401,
      "learning_rate": 7.854274330513626e-07,
      "loss": 0.70536017,
      "num_input_tokens_seen": 257035465,
      "step": 11917,
      "time_per_iteration": 2.6407663822174072
    },
    {
      "auxiliary_loss_clip": 0.01082482,
      "auxiliary_loss_mlp": 0.01027083,
      "balance_loss_clip": 1.03440392,
      "balance_loss_mlp": 1.0145005,
      "epoch": 0.7165489252968585,
      "flos": 21470523546240.0,
      "grad_norm": 1.6352184773263274,
      "language_loss": 0.76114947,
      "learning_rate": 7.851180353640896e-07,
      "loss": 0.78224516,
      "num_input_tokens_seen": 257053750,
      "step": 11918,
      "time_per_iteration": 2.5341713428497314
    },
    {
      "auxiliary_loss_clip": 0.01015531,
      "auxiliary_loss_mlp": 0.01002811,
      "balance_loss_clip": 1.01218748,
      "balance_loss_mlp": 1.00179768,
      "epoch": 0.7166090485495266,
      "flos": 69928060464000.0,
      "grad_norm": 0.6264180327898039,
      "language_loss": 0.53935421,
      "learning_rate": 7.848086837452639e-07,
      "loss": 0.55953759,
      "num_input_tokens_seen": 257121215,
      "step": 11919,
      "time_per_iteration": 3.1593003273010254
    },
    {
      "auxiliary_loss_clip": 0.01086793,
      "auxiliary_loss_mlp": 0.01032523,
      "balance_loss_clip": 1.03838396,
      "balance_loss_mlp": 1.02109051,
      "epoch": 0.7166691718021945,
      "flos": 27343892949120.0,
      "grad_norm": 2.023797999723082,
      "language_loss": 0.68727535,
      "learning_rate": 7.844993782066132e-07,
      "loss": 0.70846856,
      "num_input_tokens_seen": 257143370,
      "step": 11920,
      "time_per_iteration": 2.569462537765503
    },
    {
      "auxiliary_loss_clip": 0.01086441,
      "auxiliary_loss_mlp": 0.01035859,
      "balance_loss_clip": 1.03462398,
      "balance_loss_mlp": 1.02350247,
      "epoch": 0.7167292950548625,
      "flos": 30408868563840.0,
      "grad_norm": 2.008898626318707,
      "language_loss": 0.74751419,
      "learning_rate": 7.841901187598678e-07,
      "loss": 0.7687372,
      "num_input_tokens_seen": 257162160,
      "step": 11921,
      "time_per_iteration": 2.5921475887298584
    },
    {
      "auxiliary_loss_clip": 0.01080461,
      "auxiliary_loss_mlp": 0.01033541,
      "balance_loss_clip": 1.03792822,
      "balance_loss_mlp": 1.01931977,
      "epoch": 0.7167894183075304,
      "flos": 14571257800320.0,
      "grad_norm": 2.2047923633807276,
      "language_loss": 0.75202829,
      "learning_rate": 7.83880905416755e-07,
      "loss": 0.77316833,
      "num_input_tokens_seen": 257179300,
      "step": 11922,
      "time_per_iteration": 2.5345499515533447
    },
    {
      "auxiliary_loss_clip": 0.01018658,
      "auxiliary_loss_mlp": 0.01001639,
      "balance_loss_clip": 1.013906,
      "balance_loss_mlp": 1.00050008,
      "epoch": 0.7168495415601984,
      "flos": 64110674407680.0,
      "grad_norm": 0.7566472549204487,
      "language_loss": 0.55167407,
      "learning_rate": 7.83571738189001e-07,
      "loss": 0.571877,
      "num_input_tokens_seen": 257235470,
      "step": 11923,
      "time_per_iteration": 2.9644393920898438
    },
    {
      "auxiliary_loss_clip": 0.01074464,
      "auxiliary_loss_mlp": 0.01034733,
      "balance_loss_clip": 1.03556681,
      "balance_loss_mlp": 1.02249002,
      "epoch": 0.7169096648128663,
      "flos": 24681440119680.0,
      "grad_norm": 1.4626887072055013,
      "language_loss": 0.77142984,
      "learning_rate": 7.832626170883279e-07,
      "loss": 0.79252177,
      "num_input_tokens_seen": 257255850,
      "step": 11924,
      "time_per_iteration": 2.575505495071411
    },
    {
      "auxiliary_loss_clip": 0.01073734,
      "auxiliary_loss_mlp": 0.01029994,
      "balance_loss_clip": 1.03700185,
      "balance_loss_mlp": 1.01868749,
      "epoch": 0.7169697880655344,
      "flos": 20667525050880.0,
      "grad_norm": 1.7291360149024588,
      "language_loss": 0.68022263,
      "learning_rate": 7.829535421264588e-07,
      "loss": 0.70125997,
      "num_input_tokens_seen": 257275425,
      "step": 11925,
      "time_per_iteration": 2.5726964473724365
    },
    {
      "auxiliary_loss_clip": 0.01074099,
      "auxiliary_loss_mlp": 0.01028597,
      "balance_loss_clip": 1.03459764,
      "balance_loss_mlp": 1.0176121,
      "epoch": 0.7170299113182023,
      "flos": 21032700670080.0,
      "grad_norm": 1.9775866566987244,
      "language_loss": 0.77234817,
      "learning_rate": 7.826445133151133e-07,
      "loss": 0.79337513,
      "num_input_tokens_seen": 257295740,
      "step": 11926,
      "time_per_iteration": 2.574920892715454
    },
    {
      "auxiliary_loss_clip": 0.01094547,
      "auxiliary_loss_mlp": 0.00783269,
      "balance_loss_clip": 1.03597498,
      "balance_loss_mlp": 1.00650382,
      "epoch": 0.7170900345708703,
      "flos": 22893304239360.0,
      "grad_norm": 2.0433559573548212,
      "language_loss": 0.7709229,
      "learning_rate": 7.823355306660093e-07,
      "loss": 0.7897011,
      "num_input_tokens_seen": 257315970,
      "step": 11927,
      "time_per_iteration": 2.5493104457855225
    },
    {
      "auxiliary_loss_clip": 0.01095649,
      "auxiliary_loss_mlp": 0.01029021,
      "balance_loss_clip": 1.03817272,
      "balance_loss_mlp": 1.01665914,
      "epoch": 0.7171501578235383,
      "flos": 15518688883200.0,
      "grad_norm": 1.6086582198331,
      "language_loss": 0.69142753,
      "learning_rate": 7.820265941908642e-07,
      "loss": 0.71267426,
      "num_input_tokens_seen": 257334230,
      "step": 11928,
      "time_per_iteration": 2.501865863800049
    },
    {
      "auxiliary_loss_clip": 0.01055872,
      "auxiliary_loss_mlp": 0.01034698,
      "balance_loss_clip": 1.0337249,
      "balance_loss_mlp": 1.02308655,
      "epoch": 0.7172102810762062,
      "flos": 26104292640000.0,
      "grad_norm": 1.7555819902799286,
      "language_loss": 0.65029305,
      "learning_rate": 7.817177039013931e-07,
      "loss": 0.67119873,
      "num_input_tokens_seen": 257352145,
      "step": 11929,
      "time_per_iteration": 2.6298696994781494
    },
    {
      "auxiliary_loss_clip": 0.01082374,
      "auxiliary_loss_mlp": 0.0102927,
      "balance_loss_clip": 1.03439498,
      "balance_loss_mlp": 1.01658034,
      "epoch": 0.7172704043288742,
      "flos": 21506649649920.0,
      "grad_norm": 3.2089638946706063,
      "language_loss": 0.70039117,
      "learning_rate": 7.81408859809308e-07,
      "loss": 0.72150767,
      "num_input_tokens_seen": 257371460,
      "step": 11930,
      "time_per_iteration": 2.5198023319244385
    },
    {
      "auxiliary_loss_clip": 0.01071388,
      "auxiliary_loss_mlp": 0.01030712,
      "balance_loss_clip": 1.03288949,
      "balance_loss_mlp": 1.01873779,
      "epoch": 0.7173305275815421,
      "flos": 18770939032320.0,
      "grad_norm": 2.7680510460882433,
      "language_loss": 0.8053897,
      "learning_rate": 7.811000619263219e-07,
      "loss": 0.82641077,
      "num_input_tokens_seen": 257390800,
      "step": 11931,
      "time_per_iteration": 3.936819076538086
    },
    {
      "auxiliary_loss_clip": 0.01091791,
      "auxiliary_loss_mlp": 0.01027576,
      "balance_loss_clip": 1.03471351,
      "balance_loss_mlp": 1.01675797,
      "epoch": 0.7173906508342102,
      "flos": 16179876483840.0,
      "grad_norm": 2.080520246818637,
      "language_loss": 0.78385794,
      "learning_rate": 7.80791310264143e-07,
      "loss": 0.80505162,
      "num_input_tokens_seen": 257407495,
      "step": 11932,
      "time_per_iteration": 2.4642670154571533
    },
    {
      "auxiliary_loss_clip": 0.01088343,
      "auxiliary_loss_mlp": 0.01028244,
      "balance_loss_clip": 1.03453326,
      "balance_loss_mlp": 1.01635909,
      "epoch": 0.7174507740868781,
      "flos": 26613864933120.0,
      "grad_norm": 1.4173414863590335,
      "language_loss": 0.74876332,
      "learning_rate": 7.804826048344803e-07,
      "loss": 0.76992917,
      "num_input_tokens_seen": 257429675,
      "step": 11933,
      "time_per_iteration": 2.535809278488159
    },
    {
      "auxiliary_loss_clip": 0.01111608,
      "auxiliary_loss_mlp": 0.01036912,
      "balance_loss_clip": 1.03723192,
      "balance_loss_mlp": 1.02234423,
      "epoch": 0.7175108973395461,
      "flos": 18432911116800.0,
      "grad_norm": 2.4870275276948006,
      "language_loss": 0.68987137,
      "learning_rate": 7.801739456490388e-07,
      "loss": 0.71135664,
      "num_input_tokens_seen": 257442765,
      "step": 11934,
      "time_per_iteration": 2.4210188388824463
    },
    {
      "auxiliary_loss_clip": 0.01096084,
      "auxiliary_loss_mlp": 0.0103114,
      "balance_loss_clip": 1.0358572,
      "balance_loss_mlp": 1.01911783,
      "epoch": 0.717571020592214,
      "flos": 23914962777600.0,
      "grad_norm": 5.497518538818899,
      "language_loss": 0.86571723,
      "learning_rate": 7.798653327195237e-07,
      "loss": 0.88698947,
      "num_input_tokens_seen": 257459310,
      "step": 11935,
      "time_per_iteration": 2.5081276893615723
    },
    {
      "auxiliary_loss_clip": 0.01072066,
      "auxiliary_loss_mlp": 0.01031464,
      "balance_loss_clip": 1.03450167,
      "balance_loss_mlp": 1.01880431,
      "epoch": 0.717631143844882,
      "flos": 38256930109440.0,
      "grad_norm": 1.3725472031103507,
      "language_loss": 0.73713285,
      "learning_rate": 7.795567660576388e-07,
      "loss": 0.75816816,
      "num_input_tokens_seen": 257484750,
      "step": 11936,
      "time_per_iteration": 4.097309589385986
    },
    {
      "auxiliary_loss_clip": 0.01033596,
      "auxiliary_loss_mlp": 0.010013,
      "balance_loss_clip": 1.00995016,
      "balance_loss_mlp": 1.00019729,
      "epoch": 0.7176912670975499,
      "flos": 65515896328320.0,
      "grad_norm": 0.7666390825232345,
      "language_loss": 0.5592922,
      "learning_rate": 7.79248245675082e-07,
      "loss": 0.57964122,
      "num_input_tokens_seen": 257543110,
      "step": 11937,
      "time_per_iteration": 4.443563938140869
    },
    {
      "auxiliary_loss_clip": 0.01098663,
      "auxiliary_loss_mlp": 0.01033259,
      "balance_loss_clip": 1.03763294,
      "balance_loss_mlp": 1.01959193,
      "epoch": 0.717751390350218,
      "flos": 31281066610560.0,
      "grad_norm": 2.271688120780419,
      "language_loss": 0.54557323,
      "learning_rate": 7.789397715835542e-07,
      "loss": 0.56689239,
      "num_input_tokens_seen": 257567410,
      "step": 11938,
      "time_per_iteration": 2.576124429702759
    },
    {
      "auxiliary_loss_clip": 0.01091387,
      "auxiliary_loss_mlp": 0.01025214,
      "balance_loss_clip": 1.03489923,
      "balance_loss_mlp": 1.01377034,
      "epoch": 0.7178115136028859,
      "flos": 19859031774720.0,
      "grad_norm": 1.5892759121295195,
      "language_loss": 0.7644465,
      "learning_rate": 7.786313437947527e-07,
      "loss": 0.78561252,
      "num_input_tokens_seen": 257586270,
      "step": 11939,
      "time_per_iteration": 2.5021820068359375
    },
    {
      "auxiliary_loss_clip": 0.01016396,
      "auxiliary_loss_mlp": 0.01000559,
      "balance_loss_clip": 1.01326609,
      "balance_loss_mlp": 0.99925405,
      "epoch": 0.7178716368555539,
      "flos": 64348655967360.0,
      "grad_norm": 0.7549039688059097,
      "language_loss": 0.61367393,
      "learning_rate": 7.783229623203738e-07,
      "loss": 0.63384348,
      "num_input_tokens_seen": 257647415,
      "step": 11940,
      "time_per_iteration": 3.084197759628296
    },
    {
      "auxiliary_loss_clip": 0.01067936,
      "auxiliary_loss_mlp": 0.01028819,
      "balance_loss_clip": 1.03505886,
      "balance_loss_mlp": 1.01734543,
      "epoch": 0.7179317601082219,
      "flos": 26762607152640.0,
      "grad_norm": 1.5919596117727126,
      "language_loss": 0.58824259,
      "learning_rate": 7.780146271721097e-07,
      "loss": 0.60921013,
      "num_input_tokens_seen": 257669795,
      "step": 11941,
      "time_per_iteration": 2.6211977005004883
    },
    {
      "auxiliary_loss_clip": 0.01083134,
      "auxiliary_loss_mlp": 0.01028395,
      "balance_loss_clip": 1.0357852,
      "balance_loss_mlp": 1.0166049,
      "epoch": 0.7179918833608898,
      "flos": 23513804709120.0,
      "grad_norm": 2.0447078731059753,
      "language_loss": 0.7928949,
      "learning_rate": 7.777063383616543e-07,
      "loss": 0.81401014,
      "num_input_tokens_seen": 257687415,
      "step": 11942,
      "time_per_iteration": 4.0934977531433105
    },
    {
      "auxiliary_loss_clip": 0.01095392,
      "auxiliary_loss_mlp": 0.01039686,
      "balance_loss_clip": 1.03605509,
      "balance_loss_mlp": 1.02699018,
      "epoch": 0.7180520066135578,
      "flos": 17165588486400.0,
      "grad_norm": 1.8302745906112567,
      "language_loss": 0.66460538,
      "learning_rate": 7.773980959006968e-07,
      "loss": 0.68595618,
      "num_input_tokens_seen": 257706215,
      "step": 11943,
      "time_per_iteration": 2.4753940105438232
    },
    {
      "auxiliary_loss_clip": 0.01103664,
      "auxiliary_loss_mlp": 0.01028658,
      "balance_loss_clip": 1.03550446,
      "balance_loss_mlp": 1.01656985,
      "epoch": 0.7181121298662257,
      "flos": 17566638814080.0,
      "grad_norm": 1.7739858808267672,
      "language_loss": 0.7895906,
      "learning_rate": 7.770898998009254e-07,
      "loss": 0.8109138,
      "num_input_tokens_seen": 257724740,
      "step": 11944,
      "time_per_iteration": 2.4623310565948486
    },
    {
      "auxiliary_loss_clip": 0.01089342,
      "auxiliary_loss_mlp": 0.00785549,
      "balance_loss_clip": 1.0379163,
      "balance_loss_mlp": 1.01027071,
      "epoch": 0.7181722531188938,
      "flos": 11947660508160.0,
      "grad_norm": 2.036852574355583,
      "language_loss": 0.63525385,
      "learning_rate": 7.767817500740277e-07,
      "loss": 0.65400279,
      "num_input_tokens_seen": 257742060,
      "step": 11945,
      "time_per_iteration": 2.5050928592681885
    },
    {
      "auxiliary_loss_clip": 0.01023159,
      "auxiliary_loss_mlp": 0.01001693,
      "balance_loss_clip": 1.01315904,
      "balance_loss_mlp": 1.00036383,
      "epoch": 0.7182323763715617,
      "flos": 65503649790720.0,
      "grad_norm": 0.698944392168418,
      "language_loss": 0.51087195,
      "learning_rate": 7.76473646731689e-07,
      "loss": 0.53112048,
      "num_input_tokens_seen": 257802250,
      "step": 11946,
      "time_per_iteration": 3.0180633068084717
    },
    {
      "auxiliary_loss_clip": 0.01075111,
      "auxiliary_loss_mlp": 0.01035964,
      "balance_loss_clip": 1.03638554,
      "balance_loss_mlp": 1.02191496,
      "epoch": 0.7182924996242297,
      "flos": 20630932070400.0,
      "grad_norm": 1.6275578141030953,
      "language_loss": 0.7479611,
      "learning_rate": 7.761655897855925e-07,
      "loss": 0.76907182,
      "num_input_tokens_seen": 257821155,
      "step": 11947,
      "time_per_iteration": 2.546682596206665
    },
    {
      "auxiliary_loss_clip": 0.01060375,
      "auxiliary_loss_mlp": 0.00783189,
      "balance_loss_clip": 1.03242838,
      "balance_loss_mlp": 1.00876665,
      "epoch": 0.7183526228768976,
      "flos": 16216433550720.0,
      "grad_norm": 1.4652449489507813,
      "language_loss": 0.72349209,
      "learning_rate": 7.758575792474187e-07,
      "loss": 0.74192774,
      "num_input_tokens_seen": 257839905,
      "step": 11948,
      "time_per_iteration": 2.5727756023406982
    },
    {
      "auxiliary_loss_clip": 0.01087136,
      "auxiliary_loss_mlp": 0.01042832,
      "balance_loss_clip": 1.0364722,
      "balance_loss_mlp": 1.02890825,
      "epoch": 0.7184127461295656,
      "flos": 22232655342720.0,
      "grad_norm": 1.5105876666756677,
      "language_loss": 0.71491194,
      "learning_rate": 7.755496151288483e-07,
      "loss": 0.73621166,
      "num_input_tokens_seen": 257860055,
      "step": 11949,
      "time_per_iteration": 2.529378652572632
    },
    {
      "auxiliary_loss_clip": 0.01104122,
      "auxiliary_loss_mlp": 0.00783372,
      "balance_loss_clip": 1.03601885,
      "balance_loss_mlp": 1.00932324,
      "epoch": 0.7184728693822335,
      "flos": 27344503480320.0,
      "grad_norm": 2.0280337450901578,
      "language_loss": 0.76404715,
      "learning_rate": 7.752416974415598e-07,
      "loss": 0.78292221,
      "num_input_tokens_seen": 257879315,
      "step": 11950,
      "time_per_iteration": 2.514943838119507
    },
    {
      "auxiliary_loss_clip": 0.01109008,
      "auxiliary_loss_mlp": 0.01031684,
      "balance_loss_clip": 1.03807354,
      "balance_loss_mlp": 1.0187434,
      "epoch": 0.7185329926349016,
      "flos": 16508530949760.0,
      "grad_norm": 4.127421701277714,
      "language_loss": 0.67468417,
      "learning_rate": 7.749338261972282e-07,
      "loss": 0.69609106,
      "num_input_tokens_seen": 257896570,
      "step": 11951,
      "time_per_iteration": 2.436288595199585
    },
    {
      "auxiliary_loss_clip": 0.01088086,
      "auxiliary_loss_mlp": 0.01037747,
      "balance_loss_clip": 1.0376842,
      "balance_loss_mlp": 1.02410376,
      "epoch": 0.7185931158875695,
      "flos": 23951052967680.0,
      "grad_norm": 1.7098013833672452,
      "language_loss": 0.78060764,
      "learning_rate": 7.746260014075286e-07,
      "loss": 0.80186599,
      "num_input_tokens_seen": 257916855,
      "step": 11952,
      "time_per_iteration": 2.561812162399292
    },
    {
      "auxiliary_loss_clip": 0.01101387,
      "auxiliary_loss_mlp": 0.01032717,
      "balance_loss_clip": 1.03797626,
      "balance_loss_mlp": 1.01994944,
      "epoch": 0.7186532391402375,
      "flos": 26542007775360.0,
      "grad_norm": 1.9238958633197314,
      "language_loss": 0.74790251,
      "learning_rate": 7.743182230841352e-07,
      "loss": 0.76924354,
      "num_input_tokens_seen": 257937140,
      "step": 11953,
      "time_per_iteration": 2.5227158069610596
    },
    {
      "auxiliary_loss_clip": 0.0109612,
      "auxiliary_loss_mlp": 0.01028249,
      "balance_loss_clip": 1.03425157,
      "balance_loss_mlp": 1.0156548,
      "epoch": 0.7187133623929055,
      "flos": 22383049587840.0,
      "grad_norm": 1.7677469913819763,
      "language_loss": 0.72929847,
      "learning_rate": 7.740104912387164e-07,
      "loss": 0.75054216,
      "num_input_tokens_seen": 257956785,
      "step": 11954,
      "time_per_iteration": 2.504838705062866
    },
    {
      "auxiliary_loss_clip": 0.01087634,
      "auxiliary_loss_mlp": 0.0103513,
      "balance_loss_clip": 1.03904963,
      "balance_loss_mlp": 1.02281022,
      "epoch": 0.7187734856455734,
      "flos": 15779580341760.0,
      "grad_norm": 1.744630523984479,
      "language_loss": 0.74574912,
      "learning_rate": 7.737028058829425e-07,
      "loss": 0.76697671,
      "num_input_tokens_seen": 257975455,
      "step": 11955,
      "time_per_iteration": 2.5056939125061035
    },
    {
      "auxiliary_loss_clip": 0.01073534,
      "auxiliary_loss_mlp": 0.01030926,
      "balance_loss_clip": 1.03618169,
      "balance_loss_mlp": 1.01850998,
      "epoch": 0.7188336088982414,
      "flos": 31759612531200.0,
      "grad_norm": 1.5533836704141994,
      "language_loss": 0.73488235,
      "learning_rate": 7.733951670284817e-07,
      "loss": 0.75592697,
      "num_input_tokens_seen": 257996850,
      "step": 11956,
      "time_per_iteration": 2.6486711502075195
    },
    {
      "auxiliary_loss_clip": 0.01023921,
      "auxiliary_loss_mlp": 0.01036927,
      "balance_loss_clip": 1.03080606,
      "balance_loss_mlp": 1.02239561,
      "epoch": 0.7188937321509093,
      "flos": 21465208333440.0,
      "grad_norm": 1.6516988276465094,
      "language_loss": 0.70909607,
      "learning_rate": 7.730875746869987e-07,
      "loss": 0.7297045,
      "num_input_tokens_seen": 258016145,
      "step": 11957,
      "time_per_iteration": 2.692265510559082
    },
    {
      "auxiliary_loss_clip": 0.01058149,
      "auxiliary_loss_mlp": 0.01035928,
      "balance_loss_clip": 1.03375614,
      "balance_loss_mlp": 1.02246952,
      "epoch": 0.7189538554035774,
      "flos": 27271497087360.0,
      "grad_norm": 1.8997440984303826,
      "language_loss": 0.7322818,
      "learning_rate": 7.727800288701582e-07,
      "loss": 0.75322258,
      "num_input_tokens_seen": 258035420,
      "step": 11958,
      "time_per_iteration": 2.6634421348571777
    },
    {
      "auxiliary_loss_clip": 0.01092577,
      "auxiliary_loss_mlp": 0.01040512,
      "balance_loss_clip": 1.03554463,
      "balance_loss_mlp": 1.02738082,
      "epoch": 0.7190139786562453,
      "flos": 21580625710080.0,
      "grad_norm": 1.5645846052535453,
      "language_loss": 0.83934444,
      "learning_rate": 7.724725295896215e-07,
      "loss": 0.86067533,
      "num_input_tokens_seen": 258053520,
      "step": 11959,
      "time_per_iteration": 2.5061912536621094
    },
    {
      "auxiliary_loss_clip": 0.01112787,
      "auxiliary_loss_mlp": 0.01029964,
      "balance_loss_clip": 1.03966784,
      "balance_loss_mlp": 1.01658893,
      "epoch": 0.7190741019089133,
      "flos": 26721237663360.0,
      "grad_norm": 1.5769344893515211,
      "language_loss": 0.81988418,
      "learning_rate": 7.7216507685705e-07,
      "loss": 0.84131163,
      "num_input_tokens_seen": 258073020,
      "step": 11960,
      "time_per_iteration": 2.502078056335449
    },
    {
      "auxiliary_loss_clip": 0.01080888,
      "auxiliary_loss_mlp": 0.01035717,
      "balance_loss_clip": 1.03750098,
      "balance_loss_mlp": 1.02279496,
      "epoch": 0.7191342251615812,
      "flos": 26104759516800.0,
      "grad_norm": 1.6282973881142366,
      "language_loss": 0.78032422,
      "learning_rate": 7.718576706841013e-07,
      "loss": 0.80149019,
      "num_input_tokens_seen": 258093155,
      "step": 11961,
      "time_per_iteration": 2.5713517665863037
    },
    {
      "auxiliary_loss_clip": 0.01089553,
      "auxiliary_loss_mlp": 0.0102982,
      "balance_loss_clip": 1.03739953,
      "balance_loss_mlp": 1.01836967,
      "epoch": 0.7191943484142492,
      "flos": 22967028904320.0,
      "grad_norm": 1.344376399097129,
      "language_loss": 0.75017726,
      "learning_rate": 7.715503110824326e-07,
      "loss": 0.77137107,
      "num_input_tokens_seen": 258113905,
      "step": 11962,
      "time_per_iteration": 2.5273945331573486
    },
    {
      "auxiliary_loss_clip": 0.01096078,
      "auxiliary_loss_mlp": 0.01031183,
      "balance_loss_clip": 1.03594208,
      "balance_loss_mlp": 1.01756918,
      "epoch": 0.7192544716669171,
      "flos": 22565332131840.0,
      "grad_norm": 2.100076080534987,
      "language_loss": 0.75496048,
      "learning_rate": 7.712429980637001e-07,
      "loss": 0.77623308,
      "num_input_tokens_seen": 258132820,
      "step": 11963,
      "time_per_iteration": 2.5283877849578857
    },
    {
      "auxiliary_loss_clip": 0.01078361,
      "auxiliary_loss_mlp": 0.01037671,
      "balance_loss_clip": 1.03812051,
      "balance_loss_mlp": 1.02446294,
      "epoch": 0.7193145949195852,
      "flos": 18982200873600.0,
      "grad_norm": 2.2933348881577182,
      "language_loss": 0.80852747,
      "learning_rate": 7.709357316395564e-07,
      "loss": 0.82968783,
      "num_input_tokens_seen": 258148055,
      "step": 11964,
      "time_per_iteration": 2.556147336959839
    },
    {
      "auxiliary_loss_clip": 0.01095073,
      "auxiliary_loss_mlp": 0.01031987,
      "balance_loss_clip": 1.03599191,
      "balance_loss_mlp": 1.01940441,
      "epoch": 0.7193747181722531,
      "flos": 18004246208640.0,
      "grad_norm": 1.8639765448097971,
      "language_loss": 0.74940085,
      "learning_rate": 7.70628511821652e-07,
      "loss": 0.77067143,
      "num_input_tokens_seen": 258165995,
      "step": 11965,
      "time_per_iteration": 2.476600170135498
    },
    {
      "auxiliary_loss_clip": 0.01081713,
      "auxiliary_loss_mlp": 0.01032802,
      "balance_loss_clip": 1.037673,
      "balance_loss_mlp": 1.01983798,
      "epoch": 0.7194348414249211,
      "flos": 24389414547840.0,
      "grad_norm": 1.925133362027158,
      "language_loss": 0.77662426,
      "learning_rate": 7.703213386216377e-07,
      "loss": 0.79776943,
      "num_input_tokens_seen": 258186165,
      "step": 11966,
      "time_per_iteration": 2.5590975284576416
    },
    {
      "auxiliary_loss_clip": 0.01082997,
      "auxiliary_loss_mlp": 0.0103501,
      "balance_loss_clip": 1.03593802,
      "balance_loss_mlp": 1.02129447,
      "epoch": 0.7194949646775891,
      "flos": 22163455791360.0,
      "grad_norm": 1.8400521531104652,
      "language_loss": 0.72888118,
      "learning_rate": 7.700142120511619e-07,
      "loss": 0.75006121,
      "num_input_tokens_seen": 258204595,
      "step": 11967,
      "time_per_iteration": 2.5277252197265625
    },
    {
      "auxiliary_loss_clip": 0.01080502,
      "auxiliary_loss_mlp": 0.01028038,
      "balance_loss_clip": 1.03939724,
      "balance_loss_mlp": 1.01711869,
      "epoch": 0.719555087930257,
      "flos": 20266366982400.0,
      "grad_norm": 1.9907374907859772,
      "language_loss": 0.8167311,
      "learning_rate": 7.6970713212187e-07,
      "loss": 0.83781654,
      "num_input_tokens_seen": 258223110,
      "step": 11968,
      "time_per_iteration": 2.5159270763397217
    },
    {
      "auxiliary_loss_clip": 0.01083858,
      "auxiliary_loss_mlp": 0.01028345,
      "balance_loss_clip": 1.03673291,
      "balance_loss_mlp": 1.01616788,
      "epoch": 0.719615211182925,
      "flos": 24716309247360.0,
      "grad_norm": 2.3722066231361922,
      "language_loss": 0.76219612,
      "learning_rate": 7.69400098845407e-07,
      "loss": 0.78331816,
      "num_input_tokens_seen": 258242660,
      "step": 11969,
      "time_per_iteration": 3.935575485229492
    },
    {
      "auxiliary_loss_clip": 0.01066707,
      "auxiliary_loss_mlp": 0.01031633,
      "balance_loss_clip": 1.03221524,
      "balance_loss_mlp": 1.01812685,
      "epoch": 0.719675334435593,
      "flos": 20009641501440.0,
      "grad_norm": 2.8591538340745632,
      "language_loss": 0.70530272,
      "learning_rate": 7.69093112233417e-07,
      "loss": 0.72628617,
      "num_input_tokens_seen": 258261850,
      "step": 11970,
      "time_per_iteration": 2.5201539993286133
    },
    {
      "auxiliary_loss_clip": 0.01017196,
      "auxiliary_loss_mlp": 0.01002668,
      "balance_loss_clip": 1.01321507,
      "balance_loss_mlp": 1.00159538,
      "epoch": 0.719735457688261,
      "flos": 44199861177600.0,
      "grad_norm": 0.9178475229041617,
      "language_loss": 0.60893989,
      "learning_rate": 7.68786172297538e-07,
      "loss": 0.62913859,
      "num_input_tokens_seen": 258312570,
      "step": 11971,
      "time_per_iteration": 3.0158333778381348
    },
    {
      "auxiliary_loss_clip": 0.01112753,
      "auxiliary_loss_mlp": 0.0103447,
      "balance_loss_clip": 1.0379045,
      "balance_loss_mlp": 1.02103555,
      "epoch": 0.7197955809409289,
      "flos": 16802890905600.0,
      "grad_norm": 1.9598894525282011,
      "language_loss": 0.80275059,
      "learning_rate": 7.684792790494105e-07,
      "loss": 0.8242228,
      "num_input_tokens_seen": 258331600,
      "step": 11972,
      "time_per_iteration": 2.477065324783325
    },
    {
      "auxiliary_loss_clip": 0.01088517,
      "auxiliary_loss_mlp": 0.01036442,
      "balance_loss_clip": 1.03619552,
      "balance_loss_mlp": 1.0236032,
      "epoch": 0.7198557041935969,
      "flos": 24535391420160.0,
      "grad_norm": 1.5136043354465274,
      "language_loss": 0.75390255,
      "learning_rate": 7.681724325006733e-07,
      "loss": 0.77515221,
      "num_input_tokens_seen": 258351785,
      "step": 11973,
      "time_per_iteration": 2.5443708896636963
    },
    {
      "auxiliary_loss_clip": 0.01000675,
      "auxiliary_loss_mlp": 0.01001318,
      "balance_loss_clip": 1.01600599,
      "balance_loss_mlp": 1.00006068,
      "epoch": 0.7199158274462648,
      "flos": 70710839602560.0,
      "grad_norm": 0.8523903830877986,
      "language_loss": 0.57264346,
      "learning_rate": 7.6786563266296e-07,
      "loss": 0.59266341,
      "num_input_tokens_seen": 258404035,
      "step": 11974,
      "time_per_iteration": 4.425535440444946
    },
    {
      "auxiliary_loss_clip": 0.01080529,
      "auxiliary_loss_mlp": 0.0103166,
      "balance_loss_clip": 1.03489065,
      "balance_loss_mlp": 1.01887488,
      "epoch": 0.7199759506989328,
      "flos": 29347995352320.0,
      "grad_norm": 2.108964309678405,
      "language_loss": 0.62172365,
      "learning_rate": 7.675588795479062e-07,
      "loss": 0.64284551,
      "num_input_tokens_seen": 258424850,
      "step": 11975,
      "time_per_iteration": 2.5879931449890137
    },
    {
      "auxiliary_loss_clip": 0.01093522,
      "auxiliary_loss_mlp": 0.01028939,
      "balance_loss_clip": 1.03317308,
      "balance_loss_mlp": 1.01698792,
      "epoch": 0.7200360739516007,
      "flos": 24640465680000.0,
      "grad_norm": 1.819074630101109,
      "language_loss": 0.67485356,
      "learning_rate": 7.672521731671425e-07,
      "loss": 0.69607818,
      "num_input_tokens_seen": 258445485,
      "step": 11976,
      "time_per_iteration": 3.925523281097412
    },
    {
      "auxiliary_loss_clip": 0.01076926,
      "auxiliary_loss_mlp": 0.01029167,
      "balance_loss_clip": 1.03639579,
      "balance_loss_mlp": 1.01729941,
      "epoch": 0.7200961972042688,
      "flos": 20812855478400.0,
      "grad_norm": 1.744707228433334,
      "language_loss": 0.66999346,
      "learning_rate": 7.669455135323004e-07,
      "loss": 0.69105434,
      "num_input_tokens_seen": 258464505,
      "step": 11977,
      "time_per_iteration": 2.5600762367248535
    },
    {
      "auxiliary_loss_clip": 0.01085726,
      "auxiliary_loss_mlp": 0.01033241,
      "balance_loss_clip": 1.035079,
      "balance_loss_mlp": 1.02092052,
      "epoch": 0.7201563204569367,
      "flos": 31245910174080.0,
      "grad_norm": 1.7173216383780043,
      "language_loss": 0.75373375,
      "learning_rate": 7.666389006550074e-07,
      "loss": 0.77492344,
      "num_input_tokens_seen": 258487190,
      "step": 11978,
      "time_per_iteration": 2.5942816734313965
    },
    {
      "auxiliary_loss_clip": 0.01104049,
      "auxiliary_loss_mlp": 0.0102955,
      "balance_loss_clip": 1.03595078,
      "balance_loss_mlp": 1.01661026,
      "epoch": 0.7202164437096047,
      "flos": 26651391667200.0,
      "grad_norm": 2.235893175719401,
      "language_loss": 0.78588408,
      "learning_rate": 7.663323345468908e-07,
      "loss": 0.8072201,
      "num_input_tokens_seen": 258503790,
      "step": 11979,
      "time_per_iteration": 2.518784761428833
    },
    {
      "auxiliary_loss_clip": 0.01094984,
      "auxiliary_loss_mlp": 0.01027738,
      "balance_loss_clip": 1.0360558,
      "balance_loss_mlp": 1.01507258,
      "epoch": 0.7202765669622727,
      "flos": 25959608657280.0,
      "grad_norm": 1.8854244847471613,
      "language_loss": 0.64800066,
      "learning_rate": 7.660258152195767e-07,
      "loss": 0.6692279,
      "num_input_tokens_seen": 258527335,
      "step": 11980,
      "time_per_iteration": 4.045369386672974
    },
    {
      "auxiliary_loss_clip": 0.01099664,
      "auxiliary_loss_mlp": 0.01036817,
      "balance_loss_clip": 1.03794646,
      "balance_loss_mlp": 1.02297068,
      "epoch": 0.7203366902149406,
      "flos": 28512354372480.0,
      "grad_norm": 1.9268248034762696,
      "language_loss": 0.66886318,
      "learning_rate": 7.657193426846871e-07,
      "loss": 0.69022793,
      "num_input_tokens_seen": 258546690,
      "step": 11981,
      "time_per_iteration": 2.5467936992645264
    },
    {
      "auxiliary_loss_clip": 0.01083847,
      "auxiliary_loss_mlp": 0.01033136,
      "balance_loss_clip": 1.03635216,
      "balance_loss_mlp": 1.02019036,
      "epoch": 0.7203968134676086,
      "flos": 21106030285440.0,
      "grad_norm": 1.7268608056602643,
      "language_loss": 0.74096441,
      "learning_rate": 7.65412916953843e-07,
      "loss": 0.76213431,
      "num_input_tokens_seen": 258566340,
      "step": 11982,
      "time_per_iteration": 2.5383734703063965
    },
    {
      "auxiliary_loss_clip": 0.01078343,
      "auxiliary_loss_mlp": 0.00784917,
      "balance_loss_clip": 1.03620768,
      "balance_loss_mlp": 1.00889111,
      "epoch": 0.7204569367202766,
      "flos": 18332146488960.0,
      "grad_norm": 1.7903193345995545,
      "language_loss": 0.6587494,
      "learning_rate": 7.65106538038665e-07,
      "loss": 0.67738199,
      "num_input_tokens_seen": 258584455,
      "step": 11983,
      "time_per_iteration": 2.521113157272339
    },
    {
      "auxiliary_loss_clip": 0.01077321,
      "auxiliary_loss_mlp": 0.01031413,
      "balance_loss_clip": 1.03817928,
      "balance_loss_mlp": 1.01897907,
      "epoch": 0.7205170599729446,
      "flos": 23255103980160.0,
      "grad_norm": 1.6679414728822142,
      "language_loss": 0.66225243,
      "learning_rate": 7.648002059507715e-07,
      "loss": 0.68333983,
      "num_input_tokens_seen": 258604725,
      "step": 11984,
      "time_per_iteration": 2.5684456825256348
    },
    {
      "auxiliary_loss_clip": 0.01098886,
      "auxiliary_loss_mlp": 0.01034207,
      "balance_loss_clip": 1.03855991,
      "balance_loss_mlp": 1.02065301,
      "epoch": 0.7205771832256125,
      "flos": 20120892900480.0,
      "grad_norm": 1.5776603370450002,
      "language_loss": 0.7374866,
      "learning_rate": 7.644939207017771e-07,
      "loss": 0.75881755,
      "num_input_tokens_seen": 258622885,
      "step": 11985,
      "time_per_iteration": 2.4892077445983887
    },
    {
      "auxiliary_loss_clip": 0.01095526,
      "auxiliary_loss_mlp": 0.01032051,
      "balance_loss_clip": 1.03703034,
      "balance_loss_mlp": 1.02022505,
      "epoch": 0.7206373064782805,
      "flos": 27703250565120.0,
      "grad_norm": 1.7072336282473266,
      "language_loss": 0.62416381,
      "learning_rate": 7.641876823032977e-07,
      "loss": 0.64543957,
      "num_input_tokens_seen": 258644305,
      "step": 11986,
      "time_per_iteration": 2.554241895675659
    },
    {
      "auxiliary_loss_clip": 0.01089283,
      "auxiliary_loss_mlp": 0.01034542,
      "balance_loss_clip": 1.03890109,
      "balance_loss_mlp": 1.02101159,
      "epoch": 0.7206974297309484,
      "flos": 17968156018560.0,
      "grad_norm": 1.7891264296187577,
      "language_loss": 0.72496486,
      "learning_rate": 7.638814907669455e-07,
      "loss": 0.74620312,
      "num_input_tokens_seen": 258661775,
      "step": 11987,
      "time_per_iteration": 2.5046520233154297
    },
    {
      "auxiliary_loss_clip": 0.01086952,
      "auxiliary_loss_mlp": 0.01033432,
      "balance_loss_clip": 1.03658283,
      "balance_loss_mlp": 1.02068269,
      "epoch": 0.7207575529836164,
      "flos": 16983162288000.0,
      "grad_norm": 1.7345410516408515,
      "language_loss": 0.78755867,
      "learning_rate": 7.635753461043301e-07,
      "loss": 0.80876255,
      "num_input_tokens_seen": 258679830,
      "step": 11988,
      "time_per_iteration": 2.5454559326171875
    },
    {
      "auxiliary_loss_clip": 0.0110594,
      "auxiliary_loss_mlp": 0.01028152,
      "balance_loss_clip": 1.03609252,
      "balance_loss_mlp": 1.01613617,
      "epoch": 0.7208176762362843,
      "flos": 18727594295040.0,
      "grad_norm": 1.7864472258961135,
      "language_loss": 0.78671622,
      "learning_rate": 7.632692483270618e-07,
      "loss": 0.80805719,
      "num_input_tokens_seen": 258697415,
      "step": 11989,
      "time_per_iteration": 2.4553842544555664
    },
    {
      "auxiliary_loss_clip": 0.01103535,
      "auxiliary_loss_mlp": 0.01032147,
      "balance_loss_clip": 1.03552556,
      "balance_loss_mlp": 1.02014863,
      "epoch": 0.7208777994889524,
      "flos": 18734489706240.0,
      "grad_norm": 1.827959631832912,
      "language_loss": 0.82320738,
      "learning_rate": 7.629631974467481e-07,
      "loss": 0.8445642,
      "num_input_tokens_seen": 258716755,
      "step": 11990,
      "time_per_iteration": 2.4371187686920166
    },
    {
      "auxiliary_loss_clip": 0.01083967,
      "auxiliary_loss_mlp": 0.01036502,
      "balance_loss_clip": 1.03551769,
      "balance_loss_mlp": 1.02472973,
      "epoch": 0.7209379227416203,
      "flos": 14793437376000.0,
      "grad_norm": 2.052091838304321,
      "language_loss": 0.76172066,
      "learning_rate": 7.626571934749931e-07,
      "loss": 0.78292537,
      "num_input_tokens_seen": 258733270,
      "step": 11991,
      "time_per_iteration": 2.4821038246154785
    },
    {
      "auxiliary_loss_clip": 0.01063648,
      "auxiliary_loss_mlp": 0.0102788,
      "balance_loss_clip": 1.03486955,
      "balance_loss_mlp": 1.01530933,
      "epoch": 0.7209980459942883,
      "flos": 29636860527360.0,
      "grad_norm": 1.542962209759498,
      "language_loss": 0.7266674,
      "learning_rate": 7.623512364234022e-07,
      "loss": 0.74758267,
      "num_input_tokens_seen": 258755270,
      "step": 11992,
      "time_per_iteration": 2.6267826557159424
    },
    {
      "auxiliary_loss_clip": 0.01097742,
      "auxiliary_loss_mlp": 0.0103026,
      "balance_loss_clip": 1.0365088,
      "balance_loss_mlp": 1.01822579,
      "epoch": 0.7210581692469563,
      "flos": 23477175815040.0,
      "grad_norm": 1.459461514357293,
      "language_loss": 0.66498494,
      "learning_rate": 7.620453263035755e-07,
      "loss": 0.68626499,
      "num_input_tokens_seen": 258775340,
      "step": 11993,
      "time_per_iteration": 2.513005256652832
    },
    {
      "auxiliary_loss_clip": 0.01096224,
      "auxiliary_loss_mlp": 0.01029127,
      "balance_loss_clip": 1.03503299,
      "balance_loss_mlp": 1.01721835,
      "epoch": 0.7211182924996242,
      "flos": 26099839353600.0,
      "grad_norm": 1.8025434650429608,
      "language_loss": 0.6585182,
      "learning_rate": 7.61739463127115e-07,
      "loss": 0.67977166,
      "num_input_tokens_seen": 258794580,
      "step": 11994,
      "time_per_iteration": 2.524806261062622
    },
    {
      "auxiliary_loss_clip": 0.01097571,
      "auxiliary_loss_mlp": 0.01035138,
      "balance_loss_clip": 1.03643489,
      "balance_loss_mlp": 1.02159548,
      "epoch": 0.7211784157522922,
      "flos": 17712076982400.0,
      "grad_norm": 2.0592358953569536,
      "language_loss": 0.67078084,
      "learning_rate": 7.614336469056172e-07,
      "loss": 0.69210792,
      "num_input_tokens_seen": 258812330,
      "step": 11995,
      "time_per_iteration": 2.4566006660461426
    },
    {
      "auxiliary_loss_clip": 0.01080525,
      "auxiliary_loss_mlp": 0.01030873,
      "balance_loss_clip": 1.03520775,
      "balance_loss_mlp": 1.01771784,
      "epoch": 0.7212385390049602,
      "flos": 24423637230720.0,
      "grad_norm": 1.7696630963596875,
      "language_loss": 0.79476064,
      "learning_rate": 7.6112787765068e-07,
      "loss": 0.81587464,
      "num_input_tokens_seen": 258831770,
      "step": 11996,
      "time_per_iteration": 2.566114664077759
    },
    {
      "auxiliary_loss_clip": 0.01107747,
      "auxiliary_loss_mlp": 0.01032484,
      "balance_loss_clip": 1.03763366,
      "balance_loss_mlp": 1.02042043,
      "epoch": 0.7212986622576282,
      "flos": 28147250580480.0,
      "grad_norm": 3.147617406938141,
      "language_loss": 0.81704319,
      "learning_rate": 7.60822155373899e-07,
      "loss": 0.83844554,
      "num_input_tokens_seen": 258849090,
      "step": 11997,
      "time_per_iteration": 2.534407615661621
    },
    {
      "auxiliary_loss_clip": 0.01108767,
      "auxiliary_loss_mlp": 0.01032345,
      "balance_loss_clip": 1.0365963,
      "balance_loss_mlp": 1.0195415,
      "epoch": 0.7213587855102961,
      "flos": 21835770992640.0,
      "grad_norm": 2.310638742655128,
      "language_loss": 0.6672585,
      "learning_rate": 7.605164800868646e-07,
      "loss": 0.68866962,
      "num_input_tokens_seen": 258868230,
      "step": 11998,
      "time_per_iteration": 2.464881420135498
    },
    {
      "auxiliary_loss_clip": 0.01107303,
      "auxiliary_loss_mlp": 0.01031357,
      "balance_loss_clip": 1.03766561,
      "balance_loss_mlp": 1.02005041,
      "epoch": 0.7214189087629641,
      "flos": 14611549881600.0,
      "grad_norm": 2.011737402008611,
      "language_loss": 0.72461486,
      "learning_rate": 7.602108518011696e-07,
      "loss": 0.74600148,
      "num_input_tokens_seen": 258885525,
      "step": 11999,
      "time_per_iteration": 2.4400832653045654
    },
    {
      "auxiliary_loss_clip": 0.01087117,
      "auxiliary_loss_mlp": 0.01025985,
      "balance_loss_clip": 1.03618598,
      "balance_loss_mlp": 1.01310432,
      "epoch": 0.721479032015632,
      "flos": 19390864884480.0,
      "grad_norm": 2.4655986123336113,
      "language_loss": 0.83106911,
      "learning_rate": 7.599052705284039e-07,
      "loss": 0.85220015,
      "num_input_tokens_seen": 258903245,
      "step": 12000,
      "time_per_iteration": 2.5015060901641846
    },
    {
      "auxiliary_loss_clip": 0.01097878,
      "auxiliary_loss_mlp": 0.01035213,
      "balance_loss_clip": 1.0381391,
      "balance_loss_mlp": 1.02260637,
      "epoch": 0.7215391552683,
      "flos": 18512884748160.0,
      "grad_norm": 1.8239456493197628,
      "language_loss": 0.77221334,
      "learning_rate": 7.59599736280154e-07,
      "loss": 0.79354429,
      "num_input_tokens_seen": 258921245,
      "step": 12001,
      "time_per_iteration": 2.4782397747039795
    },
    {
      "auxiliary_loss_clip": 0.01091122,
      "auxiliary_loss_mlp": 0.01038009,
      "balance_loss_clip": 1.03736472,
      "balance_loss_mlp": 1.025581,
      "epoch": 0.721599278520968,
      "flos": 23258731253760.0,
      "grad_norm": 1.8001923668220672,
      "language_loss": 0.81426156,
      "learning_rate": 7.592942490680066e-07,
      "loss": 0.83555293,
      "num_input_tokens_seen": 258939425,
      "step": 12002,
      "time_per_iteration": 2.511728286743164
    },
    {
      "auxiliary_loss_clip": 0.01099182,
      "auxiliary_loss_mlp": 0.01029968,
      "balance_loss_clip": 1.03849268,
      "balance_loss_mlp": 1.01708198,
      "epoch": 0.721659401773636,
      "flos": 39199045979520.0,
      "grad_norm": 1.980066303326202,
      "language_loss": 0.62417209,
      "learning_rate": 7.589888089035462e-07,
      "loss": 0.64546359,
      "num_input_tokens_seen": 258960710,
      "step": 12003,
      "time_per_iteration": 2.6388163566589355
    },
    {
      "auxiliary_loss_clip": 0.01107954,
      "auxiliary_loss_mlp": 0.01031313,
      "balance_loss_clip": 1.03727698,
      "balance_loss_mlp": 1.01825905,
      "epoch": 0.7217195250263039,
      "flos": 14939917038720.0,
      "grad_norm": 2.239024754848087,
      "language_loss": 0.68578589,
      "learning_rate": 7.586834157983544e-07,
      "loss": 0.70717853,
      "num_input_tokens_seen": 258978475,
      "step": 12004,
      "time_per_iteration": 2.4650685787200928
    },
    {
      "auxiliary_loss_clip": 0.01019369,
      "auxiliary_loss_mlp": 0.01004285,
      "balance_loss_clip": 1.02252102,
      "balance_loss_mlp": 1.00286663,
      "epoch": 0.7217796482789719,
      "flos": 70869206666880.0,
      "grad_norm": 0.8582544965383146,
      "language_loss": 0.54174662,
      "learning_rate": 7.583780697640112e-07,
      "loss": 0.56198323,
      "num_input_tokens_seen": 259037520,
      "step": 12005,
      "time_per_iteration": 3.0646817684173584
    },
    {
      "auxiliary_loss_clip": 0.01075798,
      "auxiliary_loss_mlp": 0.01034497,
      "balance_loss_clip": 1.0373733,
      "balance_loss_mlp": 1.02150953,
      "epoch": 0.7218397715316398,
      "flos": 37451525402880.0,
      "grad_norm": 1.4733300486680136,
      "language_loss": 0.63298815,
      "learning_rate": 7.580727708120962e-07,
      "loss": 0.65409112,
      "num_input_tokens_seen": 259061325,
      "step": 12006,
      "time_per_iteration": 2.6908769607543945
    },
    {
      "auxiliary_loss_clip": 0.01080681,
      "auxiliary_loss_mlp": 0.01033462,
      "balance_loss_clip": 1.03509653,
      "balance_loss_mlp": 1.02143431,
      "epoch": 0.7218998947843078,
      "flos": 22710662559360.0,
      "grad_norm": 1.6369783112978062,
      "language_loss": 0.919523,
      "learning_rate": 7.577675189541865e-07,
      "loss": 0.94066441,
      "num_input_tokens_seen": 259078135,
      "step": 12007,
      "time_per_iteration": 2.5146210193634033
    },
    {
      "auxiliary_loss_clip": 0.0107023,
      "auxiliary_loss_mlp": 0.01033641,
      "balance_loss_clip": 1.03320217,
      "balance_loss_mlp": 1.01913309,
      "epoch": 0.7219600180369758,
      "flos": 12167182477440.0,
      "grad_norm": 1.701079361968365,
      "language_loss": 0.6390093,
      "learning_rate": 7.574623142018568e-07,
      "loss": 0.66004807,
      "num_input_tokens_seen": 259095910,
      "step": 12008,
      "time_per_iteration": 3.9234368801116943
    },
    {
      "auxiliary_loss_clip": 0.01098823,
      "auxiliary_loss_mlp": 0.01036323,
      "balance_loss_clip": 1.03745031,
      "balance_loss_mlp": 1.02314425,
      "epoch": 0.7220201412896438,
      "flos": 22596573985920.0,
      "grad_norm": 2.2332447735548877,
      "language_loss": 0.78396165,
      "learning_rate": 7.57157156566681e-07,
      "loss": 0.80531311,
      "num_input_tokens_seen": 259114225,
      "step": 12009,
      "time_per_iteration": 2.511363983154297
    },
    {
      "auxiliary_loss_clip": 0.01097408,
      "auxiliary_loss_mlp": 0.01037882,
      "balance_loss_clip": 1.03786349,
      "balance_loss_mlp": 1.02431035,
      "epoch": 0.7220802645423118,
      "flos": 26718651884160.0,
      "grad_norm": 1.700950145887698,
      "language_loss": 0.63799286,
      "learning_rate": 7.568520460602297e-07,
      "loss": 0.65934575,
      "num_input_tokens_seen": 259134660,
      "step": 12010,
      "time_per_iteration": 2.5277822017669678
    },
    {
      "auxiliary_loss_clip": 0.01106897,
      "auxiliary_loss_mlp": 0.0102896,
      "balance_loss_clip": 1.03661275,
      "balance_loss_mlp": 1.01633024,
      "epoch": 0.7221403877949797,
      "flos": 24420548661120.0,
      "grad_norm": 1.8626682606796008,
      "language_loss": 0.77600878,
      "learning_rate": 7.565469826940742e-07,
      "loss": 0.79736733,
      "num_input_tokens_seen": 259153300,
      "step": 12011,
      "time_per_iteration": 2.4944875240325928
    },
    {
      "auxiliary_loss_clip": 0.01090813,
      "auxiliary_loss_mlp": 0.01035923,
      "balance_loss_clip": 1.03575528,
      "balance_loss_mlp": 1.02413368,
      "epoch": 0.7222005110476477,
      "flos": 23514379326720.0,
      "grad_norm": 1.622613441040394,
      "language_loss": 0.78984195,
      "learning_rate": 7.56241966479781e-07,
      "loss": 0.8111093,
      "num_input_tokens_seen": 259172115,
      "step": 12012,
      "time_per_iteration": 2.496013879776001
    },
    {
      "auxiliary_loss_clip": 0.01085982,
      "auxiliary_loss_mlp": 0.01027085,
      "balance_loss_clip": 1.0372982,
      "balance_loss_mlp": 1.01553392,
      "epoch": 0.7222606343003156,
      "flos": 23112538899840.0,
      "grad_norm": 1.753571793027519,
      "language_loss": 0.75383413,
      "learning_rate": 7.559369974289171e-07,
      "loss": 0.77496481,
      "num_input_tokens_seen": 259191345,
      "step": 12013,
      "time_per_iteration": 3.9288089275360107
    },
    {
      "auxiliary_loss_clip": 0.01105736,
      "auxiliary_loss_mlp": 0.01026423,
      "balance_loss_clip": 1.03695226,
      "balance_loss_mlp": 1.01448441,
      "epoch": 0.7223207575529836,
      "flos": 24351169541760.0,
      "grad_norm": 1.5153039809778643,
      "language_loss": 0.75723654,
      "learning_rate": 7.556320755530484e-07,
      "loss": 0.77855814,
      "num_input_tokens_seen": 259211700,
      "step": 12014,
      "time_per_iteration": 3.8815391063690186
    },
    {
      "auxiliary_loss_clip": 0.01096531,
      "auxiliary_loss_mlp": 0.01032295,
      "balance_loss_clip": 1.03440797,
      "balance_loss_mlp": 1.01985586,
      "epoch": 0.7223808808056515,
      "flos": 28330179569280.0,
      "grad_norm": 1.5699895809204507,
      "language_loss": 0.86429709,
      "learning_rate": 7.553272008637346e-07,
      "loss": 0.88558531,
      "num_input_tokens_seen": 259233825,
      "step": 12015,
      "time_per_iteration": 2.5353620052337646
    },
    {
      "auxiliary_loss_clip": 0.0109457,
      "auxiliary_loss_mlp": 0.01034589,
      "balance_loss_clip": 1.03699851,
      "balance_loss_mlp": 1.02272129,
      "epoch": 0.7224410040583196,
      "flos": 21069437304960.0,
      "grad_norm": 1.884004524586934,
      "language_loss": 0.78138977,
      "learning_rate": 7.55022373372538e-07,
      "loss": 0.80268139,
      "num_input_tokens_seen": 259253055,
      "step": 12016,
      "time_per_iteration": 2.4846980571746826
    },
    {
      "auxiliary_loss_clip": 0.01063125,
      "auxiliary_loss_mlp": 0.01043581,
      "balance_loss_clip": 1.03347993,
      "balance_loss_mlp": 1.0292697,
      "epoch": 0.7225011273109875,
      "flos": 26795429205120.0,
      "grad_norm": 1.5678902462553646,
      "language_loss": 0.77870649,
      "learning_rate": 7.547175930910186e-07,
      "loss": 0.79977351,
      "num_input_tokens_seen": 259273420,
      "step": 12017,
      "time_per_iteration": 2.5885603427886963
    },
    {
      "auxiliary_loss_clip": 0.01103695,
      "auxiliary_loss_mlp": 0.01027548,
      "balance_loss_clip": 1.03679359,
      "balance_loss_mlp": 1.0162586,
      "epoch": 0.7225612505636555,
      "flos": 23583578878080.0,
      "grad_norm": 2.333669782076671,
      "language_loss": 0.7376796,
      "learning_rate": 7.54412860030732e-07,
      "loss": 0.75899208,
      "num_input_tokens_seen": 259291000,
      "step": 12018,
      "time_per_iteration": 2.4856722354888916
    },
    {
      "auxiliary_loss_clip": 0.01074458,
      "auxiliary_loss_mlp": 0.01027646,
      "balance_loss_clip": 1.03902197,
      "balance_loss_mlp": 1.01694131,
      "epoch": 0.7226213738163234,
      "flos": 20777627214720.0,
      "grad_norm": 1.7235592131281006,
      "language_loss": 0.77772021,
      "learning_rate": 7.541081742032347e-07,
      "loss": 0.79874122,
      "num_input_tokens_seen": 259312390,
      "step": 12019,
      "time_per_iteration": 4.036248445510864
    },
    {
      "auxiliary_loss_clip": 0.01079685,
      "auxiliary_loss_mlp": 0.01026771,
      "balance_loss_clip": 1.03650761,
      "balance_loss_mlp": 1.01461768,
      "epoch": 0.7226814970689914,
      "flos": 32635832901120.0,
      "grad_norm": 1.6867685183578407,
      "language_loss": 0.74288005,
      "learning_rate": 7.53803535620081e-07,
      "loss": 0.76394463,
      "num_input_tokens_seen": 259332645,
      "step": 12020,
      "time_per_iteration": 2.6293344497680664
    },
    {
      "auxiliary_loss_clip": 0.01098923,
      "auxiliary_loss_mlp": 0.010308,
      "balance_loss_clip": 1.0364778,
      "balance_loss_mlp": 1.0194875,
      "epoch": 0.7227416203216595,
      "flos": 22454368041600.0,
      "grad_norm": 1.937676675698137,
      "language_loss": 0.77237892,
      "learning_rate": 7.534989442928219e-07,
      "loss": 0.79367614,
      "num_input_tokens_seen": 259353810,
      "step": 12021,
      "time_per_iteration": 2.4995036125183105
    },
    {
      "auxiliary_loss_clip": 0.01070137,
      "auxiliary_loss_mlp": 0.01033653,
      "balance_loss_clip": 1.03585839,
      "balance_loss_mlp": 1.02160704,
      "epoch": 0.7228017435743274,
      "flos": 21652303299840.0,
      "grad_norm": 2.5279874129766986,
      "language_loss": 0.68424916,
      "learning_rate": 7.531944002330073e-07,
      "loss": 0.7052871,
      "num_input_tokens_seen": 259372460,
      "step": 12022,
      "time_per_iteration": 2.608020544052124
    },
    {
      "auxiliary_loss_clip": 0.01096684,
      "auxiliary_loss_mlp": 0.01029501,
      "balance_loss_clip": 1.03570461,
      "balance_loss_mlp": 1.01686454,
      "epoch": 0.7228618668269954,
      "flos": 29533474206720.0,
      "grad_norm": 1.8817701882617144,
      "language_loss": 0.6948241,
      "learning_rate": 7.528899034521858e-07,
      "loss": 0.71608603,
      "num_input_tokens_seen": 259393275,
      "step": 12023,
      "time_per_iteration": 2.5548338890075684
    },
    {
      "auxiliary_loss_clip": 0.01077351,
      "auxiliary_loss_mlp": 0.01030063,
      "balance_loss_clip": 1.03201461,
      "balance_loss_mlp": 1.01730752,
      "epoch": 0.7229219900796633,
      "flos": 27453815544960.0,
      "grad_norm": 1.6617959609199873,
      "language_loss": 0.71049076,
      "learning_rate": 7.525854539619052e-07,
      "loss": 0.73156488,
      "num_input_tokens_seen": 259416205,
      "step": 12024,
      "time_per_iteration": 2.5706751346588135
    },
    {
      "auxiliary_loss_clip": 0.0107279,
      "auxiliary_loss_mlp": 0.01031386,
      "balance_loss_clip": 1.0356729,
      "balance_loss_mlp": 1.01975107,
      "epoch": 0.7229821133323313,
      "flos": 16289368116480.0,
      "grad_norm": 1.761326845041343,
      "language_loss": 0.75981498,
      "learning_rate": 7.522810517737089e-07,
      "loss": 0.78085673,
      "num_input_tokens_seen": 259433115,
      "step": 12025,
      "time_per_iteration": 2.525991439819336
    },
    {
      "auxiliary_loss_clip": 0.01093724,
      "auxiliary_loss_mlp": 0.01028903,
      "balance_loss_clip": 1.03583622,
      "balance_loss_mlp": 1.01711369,
      "epoch": 0.7230422365849992,
      "flos": 20412343854720.0,
      "grad_norm": 1.9006716741424727,
      "language_loss": 0.76709855,
      "learning_rate": 7.519766968991395e-07,
      "loss": 0.78832483,
      "num_input_tokens_seen": 259450475,
      "step": 12026,
      "time_per_iteration": 2.5133049488067627
    },
    {
      "auxiliary_loss_clip": 0.01095282,
      "auxiliary_loss_mlp": 0.01036215,
      "balance_loss_clip": 1.03621578,
      "balance_loss_mlp": 1.02425265,
      "epoch": 0.7231023598376672,
      "flos": 25593499284480.0,
      "grad_norm": 2.0467044266848164,
      "language_loss": 0.67610765,
      "learning_rate": 7.516723893497388e-07,
      "loss": 0.69742262,
      "num_input_tokens_seen": 259469355,
      "step": 12027,
      "time_per_iteration": 2.5275237560272217
    },
    {
      "auxiliary_loss_clip": 0.01056664,
      "auxiliary_loss_mlp": 0.010309,
      "balance_loss_clip": 1.03832984,
      "balance_loss_mlp": 1.01818049,
      "epoch": 0.7231624830903352,
      "flos": 25149607009920.0,
      "grad_norm": 1.8912338525285748,
      "language_loss": 0.78774494,
      "learning_rate": 7.513681291370469e-07,
      "loss": 0.80862057,
      "num_input_tokens_seen": 259486565,
      "step": 12028,
      "time_per_iteration": 2.6056907176971436
    },
    {
      "auxiliary_loss_clip": 0.01072549,
      "auxiliary_loss_mlp": 0.01026304,
      "balance_loss_clip": 1.0337292,
      "balance_loss_mlp": 1.013394,
      "epoch": 0.7232226063430032,
      "flos": 21725740656000.0,
      "grad_norm": 1.7703722098966417,
      "language_loss": 0.82011998,
      "learning_rate": 7.510639162726e-07,
      "loss": 0.8411085,
      "num_input_tokens_seen": 259505070,
      "step": 12029,
      "time_per_iteration": 2.564087152481079
    },
    {
      "auxiliary_loss_clip": 0.0101507,
      "auxiliary_loss_mlp": 0.01003054,
      "balance_loss_clip": 1.01466393,
      "balance_loss_mlp": 1.00179601,
      "epoch": 0.7232827295956711,
      "flos": 68436798491520.0,
      "grad_norm": 0.8130069444961493,
      "language_loss": 0.61730456,
      "learning_rate": 7.507597507679347e-07,
      "loss": 0.63748574,
      "num_input_tokens_seen": 259569135,
      "step": 12030,
      "time_per_iteration": 3.185398578643799
    },
    {
      "auxiliary_loss_clip": 0.01087985,
      "auxiliary_loss_mlp": 0.01031014,
      "balance_loss_clip": 1.03377557,
      "balance_loss_mlp": 1.01743603,
      "epoch": 0.7233428528483391,
      "flos": 20192642317440.0,
      "grad_norm": 1.6767182833603806,
      "language_loss": 0.77793306,
      "learning_rate": 7.504556326345859e-07,
      "loss": 0.79912305,
      "num_input_tokens_seen": 259587035,
      "step": 12031,
      "time_per_iteration": 2.4862210750579834
    },
    {
      "auxiliary_loss_clip": 0.01096666,
      "auxiliary_loss_mlp": 0.01028414,
      "balance_loss_clip": 1.03638232,
      "balance_loss_mlp": 1.01638007,
      "epoch": 0.723402976101007,
      "flos": 23949472769280.0,
      "grad_norm": 2.015650162426516,
      "language_loss": 0.81679213,
      "learning_rate": 7.501515618840834e-07,
      "loss": 0.83804286,
      "num_input_tokens_seen": 259606140,
      "step": 12032,
      "time_per_iteration": 2.5163557529449463
    },
    {
      "auxiliary_loss_clip": 0.01069283,
      "auxiliary_loss_mlp": 0.01034441,
      "balance_loss_clip": 1.03423166,
      "balance_loss_mlp": 1.02126265,
      "epoch": 0.723463099353675,
      "flos": 20813394182400.0,
      "grad_norm": 1.7396139573261402,
      "language_loss": 0.75269216,
      "learning_rate": 7.498475385279592e-07,
      "loss": 0.77372944,
      "num_input_tokens_seen": 259624275,
      "step": 12033,
      "time_per_iteration": 2.559782028198242
    },
    {
      "auxiliary_loss_clip": 0.01068344,
      "auxiliary_loss_mlp": 0.01026201,
      "balance_loss_clip": 1.03266847,
      "balance_loss_mlp": 1.01488256,
      "epoch": 0.723523222606343,
      "flos": 19098013299840.0,
      "grad_norm": 1.563194924121279,
      "language_loss": 0.74913895,
      "learning_rate": 7.495435625777423e-07,
      "loss": 0.77008438,
      "num_input_tokens_seen": 259643465,
      "step": 12034,
      "time_per_iteration": 2.53656268119812
    },
    {
      "auxiliary_loss_clip": 0.01080924,
      "auxiliary_loss_mlp": 0.01027445,
      "balance_loss_clip": 1.03433323,
      "balance_loss_mlp": 1.01664472,
      "epoch": 0.723583345859011,
      "flos": 26506994993280.0,
      "grad_norm": 1.7359936945536365,
      "language_loss": 0.8062402,
      "learning_rate": 7.492396340449578e-07,
      "loss": 0.82732391,
      "num_input_tokens_seen": 259662500,
      "step": 12035,
      "time_per_iteration": 2.591151714324951
    },
    {
      "auxiliary_loss_clip": 0.01055741,
      "auxiliary_loss_mlp": 0.01030585,
      "balance_loss_clip": 1.03806329,
      "balance_loss_mlp": 1.01862192,
      "epoch": 0.723643469111679,
      "flos": 16033863697920.0,
      "grad_norm": 1.8392335778011457,
      "language_loss": 0.61007047,
      "learning_rate": 7.489357529411326e-07,
      "loss": 0.63093376,
      "num_input_tokens_seen": 259680140,
      "step": 12036,
      "time_per_iteration": 2.585636615753174
    },
    {
      "auxiliary_loss_clip": 0.01091633,
      "auxiliary_loss_mlp": 0.01029561,
      "balance_loss_clip": 1.03542757,
      "balance_loss_mlp": 1.01883769,
      "epoch": 0.7237035923643469,
      "flos": 21945549934080.0,
      "grad_norm": 5.773644310912796,
      "language_loss": 0.67612338,
      "learning_rate": 7.486319192777883e-07,
      "loss": 0.69733536,
      "num_input_tokens_seen": 259700160,
      "step": 12037,
      "time_per_iteration": 2.513305187225342
    },
    {
      "auxiliary_loss_clip": 0.01105141,
      "auxiliary_loss_mlp": 0.01033339,
      "balance_loss_clip": 1.0364213,
      "balance_loss_mlp": 1.02138281,
      "epoch": 0.7237637156170149,
      "flos": 23583112001280.0,
      "grad_norm": 1.7881570557848763,
      "language_loss": 0.72066736,
      "learning_rate": 7.483281330664479e-07,
      "loss": 0.74205214,
      "num_input_tokens_seen": 259720525,
      "step": 12038,
      "time_per_iteration": 2.48521089553833
    },
    {
      "auxiliary_loss_clip": 0.01106585,
      "auxiliary_loss_mlp": 0.010331,
      "balance_loss_clip": 1.03717208,
      "balance_loss_mlp": 1.01985562,
      "epoch": 0.7238238388696828,
      "flos": 20594698225920.0,
      "grad_norm": 1.8347899964493977,
      "language_loss": 0.72325647,
      "learning_rate": 7.480243943186293e-07,
      "loss": 0.74465334,
      "num_input_tokens_seen": 259738680,
      "step": 12039,
      "time_per_iteration": 2.4861931800842285
    },
    {
      "auxiliary_loss_clip": 0.01107249,
      "auxiliary_loss_mlp": 0.01031444,
      "balance_loss_clip": 1.03725815,
      "balance_loss_mlp": 1.02025652,
      "epoch": 0.7238839621223508,
      "flos": 24207024263040.0,
      "grad_norm": 1.8494753458420639,
      "language_loss": 0.76147449,
      "learning_rate": 7.477207030458513e-07,
      "loss": 0.78286141,
      "num_input_tokens_seen": 259758790,
      "step": 12040,
      "time_per_iteration": 2.489377975463867
    },
    {
      "auxiliary_loss_clip": 0.01073174,
      "auxiliary_loss_mlp": 0.01033173,
      "balance_loss_clip": 1.03523135,
      "balance_loss_mlp": 1.02029252,
      "epoch": 0.7239440853750188,
      "flos": 14209745368320.0,
      "grad_norm": 1.561388328065423,
      "language_loss": 0.76634896,
      "learning_rate": 7.474170592596301e-07,
      "loss": 0.78741241,
      "num_input_tokens_seen": 259777370,
      "step": 12041,
      "time_per_iteration": 2.5473968982696533
    },
    {
      "auxiliary_loss_clip": 0.01095947,
      "auxiliary_loss_mlp": 0.01025372,
      "balance_loss_clip": 1.03461051,
      "balance_loss_mlp": 1.01388657,
      "epoch": 0.7240042086276868,
      "flos": 21614812479360.0,
      "grad_norm": 2.138919121835004,
      "language_loss": 0.63423288,
      "learning_rate": 7.471134629714797e-07,
      "loss": 0.65544605,
      "num_input_tokens_seen": 259794665,
      "step": 12042,
      "time_per_iteration": 2.532137632369995
    },
    {
      "auxiliary_loss_clip": 0.01071893,
      "auxiliary_loss_mlp": 0.01029544,
      "balance_loss_clip": 1.03962529,
      "balance_loss_mlp": 1.0170033,
      "epoch": 0.7240643318803547,
      "flos": 23331450337920.0,
      "grad_norm": 3.5352282732388236,
      "language_loss": 0.83733058,
      "learning_rate": 7.468099141929116e-07,
      "loss": 0.85834491,
      "num_input_tokens_seen": 259811110,
      "step": 12043,
      "time_per_iteration": 2.593433141708374
    },
    {
      "auxiliary_loss_clip": 0.01074017,
      "auxiliary_loss_mlp": 0.01029312,
      "balance_loss_clip": 1.03610802,
      "balance_loss_mlp": 1.01698542,
      "epoch": 0.7241244551330227,
      "flos": 24024849459840.0,
      "grad_norm": 1.616934480297615,
      "language_loss": 0.64475393,
      "learning_rate": 7.465064129354379e-07,
      "loss": 0.66578722,
      "num_input_tokens_seen": 259831080,
      "step": 12044,
      "time_per_iteration": 2.5667264461517334
    },
    {
      "auxiliary_loss_clip": 0.01108824,
      "auxiliary_loss_mlp": 0.01033325,
      "balance_loss_clip": 1.03960264,
      "balance_loss_mlp": 1.02086782,
      "epoch": 0.7241845783856906,
      "flos": 18730323728640.0,
      "grad_norm": 1.5034712126132521,
      "language_loss": 0.81702274,
      "learning_rate": 7.462029592105658e-07,
      "loss": 0.83844423,
      "num_input_tokens_seen": 259850135,
      "step": 12045,
      "time_per_iteration": 2.4614691734313965
    },
    {
      "auxiliary_loss_clip": 0.01104987,
      "auxiliary_loss_mlp": 0.01031726,
      "balance_loss_clip": 1.03690577,
      "balance_loss_mlp": 1.01990628,
      "epoch": 0.7242447016383586,
      "flos": 19498668577920.0,
      "grad_norm": 1.5437592165485716,
      "language_loss": 0.71926022,
      "learning_rate": 7.458995530298034e-07,
      "loss": 0.74062729,
      "num_input_tokens_seen": 259868185,
      "step": 12046,
      "time_per_iteration": 2.428363561630249
    },
    {
      "auxiliary_loss_clip": 0.01070242,
      "auxiliary_loss_mlp": 0.010324,
      "balance_loss_clip": 1.03301752,
      "balance_loss_mlp": 1.01935828,
      "epoch": 0.7243048248910267,
      "flos": 22163491704960.0,
      "grad_norm": 1.7175394474273817,
      "language_loss": 0.71314514,
      "learning_rate": 7.455961944046553e-07,
      "loss": 0.73417151,
      "num_input_tokens_seen": 259887055,
      "step": 12047,
      "time_per_iteration": 3.9650988578796387
    },
    {
      "auxiliary_loss_clip": 0.01089607,
      "auxiliary_loss_mlp": 0.01032789,
      "balance_loss_clip": 1.03833389,
      "balance_loss_mlp": 1.01980138,
      "epoch": 0.7243649481436946,
      "flos": 27672762896640.0,
      "grad_norm": 1.6455949899176083,
      "language_loss": 0.6974414,
      "learning_rate": 7.45292883346627e-07,
      "loss": 0.71866536,
      "num_input_tokens_seen": 259908295,
      "step": 12048,
      "time_per_iteration": 2.5880589485168457
    },
    {
      "auxiliary_loss_clip": 0.01018369,
      "auxiliary_loss_mlp": 0.01017048,
      "balance_loss_clip": 1.01493549,
      "balance_loss_mlp": 1.0157311,
      "epoch": 0.7244250713963626,
      "flos": 63244545759360.0,
      "grad_norm": 0.8301141688313411,
      "language_loss": 0.53733033,
      "learning_rate": 7.449896198672168e-07,
      "loss": 0.55768454,
      "num_input_tokens_seen": 259968475,
      "step": 12049,
      "time_per_iteration": 3.1318917274475098
    },
    {
      "auxiliary_loss_clip": 0.01091656,
      "auxiliary_loss_mlp": 0.01030192,
      "balance_loss_clip": 1.03678107,
      "balance_loss_mlp": 1.01595831,
      "epoch": 0.7244851946490305,
      "flos": 17967114524160.0,
      "grad_norm": 2.54345288859689,
      "language_loss": 0.5977515,
      "learning_rate": 7.446864039779258e-07,
      "loss": 0.61896992,
      "num_input_tokens_seen": 259984865,
      "step": 12050,
      "time_per_iteration": 2.4973537921905518
    },
    {
      "auxiliary_loss_clip": 0.00998821,
      "auxiliary_loss_mlp": 0.01003926,
      "balance_loss_clip": 1.01465797,
      "balance_loss_mlp": 1.00290704,
      "epoch": 0.7245453179016985,
      "flos": 70943649603840.0,
      "grad_norm": 0.7248307499298045,
      "language_loss": 0.53253198,
      "learning_rate": 7.443832356902528e-07,
      "loss": 0.55255949,
      "num_input_tokens_seen": 260046735,
      "step": 12051,
      "time_per_iteration": 3.188436269760132
    },
    {
      "auxiliary_loss_clip": 0.01092954,
      "auxiliary_loss_mlp": 0.0103045,
      "balance_loss_clip": 1.03552222,
      "balance_loss_mlp": 1.0192802,
      "epoch": 0.7246054411543664,
      "flos": 24568464867840.0,
      "grad_norm": 1.5463501637460988,
      "language_loss": 0.72028095,
      "learning_rate": 7.440801150156927e-07,
      "loss": 0.74151498,
      "num_input_tokens_seen": 260067950,
      "step": 12052,
      "time_per_iteration": 2.5369439125061035
    },
    {
      "auxiliary_loss_clip": 0.01092402,
      "auxiliary_loss_mlp": 0.01030849,
      "balance_loss_clip": 1.0350529,
      "balance_loss_mlp": 1.01771235,
      "epoch": 0.7246655644070344,
      "flos": 32338312548480.0,
      "grad_norm": 1.8208137207200468,
      "language_loss": 0.74431336,
      "learning_rate": 7.437770419657415e-07,
      "loss": 0.76554596,
      "num_input_tokens_seen": 260087730,
      "step": 12053,
      "time_per_iteration": 5.335455417633057
    },
    {
      "auxiliary_loss_clip": 0.01065874,
      "auxiliary_loss_mlp": 0.01029771,
      "balance_loss_clip": 1.03450203,
      "balance_loss_mlp": 1.01701605,
      "epoch": 0.7247256876597024,
      "flos": 21872471713920.0,
      "grad_norm": 1.862456790147135,
      "language_loss": 0.78572923,
      "learning_rate": 7.434740165518898e-07,
      "loss": 0.80668569,
      "num_input_tokens_seen": 260107760,
      "step": 12054,
      "time_per_iteration": 2.5593061447143555
    },
    {
      "auxiliary_loss_clip": 0.01072642,
      "auxiliary_loss_mlp": 0.01034667,
      "balance_loss_clip": 1.03686976,
      "balance_loss_mlp": 1.02242374,
      "epoch": 0.7247858109123704,
      "flos": 16213093585920.0,
      "grad_norm": 2.567759240883679,
      "language_loss": 0.68252182,
      "learning_rate": 7.431710387856301e-07,
      "loss": 0.70359492,
      "num_input_tokens_seen": 260123660,
      "step": 12055,
      "time_per_iteration": 2.544769287109375
    },
    {
      "auxiliary_loss_clip": 0.01072342,
      "auxiliary_loss_mlp": 0.01033629,
      "balance_loss_clip": 1.03556049,
      "balance_loss_mlp": 1.02230453,
      "epoch": 0.7248459341650383,
      "flos": 20850705434880.0,
      "grad_norm": 1.6318691601811386,
      "language_loss": 0.7387743,
      "learning_rate": 7.428681086784496e-07,
      "loss": 0.75983399,
      "num_input_tokens_seen": 260142690,
      "step": 12056,
      "time_per_iteration": 2.54860782623291
    },
    {
      "auxiliary_loss_clip": 0.01102147,
      "auxiliary_loss_mlp": 0.01024554,
      "balance_loss_clip": 1.03538585,
      "balance_loss_mlp": 1.01273489,
      "epoch": 0.7249060574177063,
      "flos": 25921794614400.0,
      "grad_norm": 1.5826393719871827,
      "language_loss": 0.70736146,
      "learning_rate": 7.425652262418368e-07,
      "loss": 0.7286284,
      "num_input_tokens_seen": 260162590,
      "step": 12057,
      "time_per_iteration": 2.54713773727417
    },
    {
      "auxiliary_loss_clip": 0.01062142,
      "auxiliary_loss_mlp": 0.01042156,
      "balance_loss_clip": 1.03575027,
      "balance_loss_mlp": 1.02809572,
      "epoch": 0.7249661806703742,
      "flos": 17345536646400.0,
      "grad_norm": 1.7124489422546483,
      "language_loss": 0.6259014,
      "learning_rate": 7.42262391487277e-07,
      "loss": 0.6469444,
      "num_input_tokens_seen": 260181065,
      "step": 12058,
      "time_per_iteration": 3.9628541469573975
    },
    {
      "auxiliary_loss_clip": 0.01065656,
      "auxiliary_loss_mlp": 0.01030065,
      "balance_loss_clip": 1.03413725,
      "balance_loss_mlp": 1.01705289,
      "epoch": 0.7250263039230422,
      "flos": 19574153009280.0,
      "grad_norm": 1.870493903940984,
      "language_loss": 0.74835473,
      "learning_rate": 7.419596044262535e-07,
      "loss": 0.76931196,
      "num_input_tokens_seen": 260200330,
      "step": 12059,
      "time_per_iteration": 2.603731632232666
    },
    {
      "auxiliary_loss_clip": 0.01092774,
      "auxiliary_loss_mlp": 0.01034223,
      "balance_loss_clip": 1.03543174,
      "balance_loss_mlp": 1.02312481,
      "epoch": 0.7250864271757103,
      "flos": 21976648133760.0,
      "grad_norm": 1.7309711390955966,
      "language_loss": 0.79603863,
      "learning_rate": 7.416568650702472e-07,
      "loss": 0.8173086,
      "num_input_tokens_seen": 260219975,
      "step": 12060,
      "time_per_iteration": 2.5388925075531006
    },
    {
      "auxiliary_loss_clip": 0.01095629,
      "auxiliary_loss_mlp": 0.0102721,
      "balance_loss_clip": 1.03683019,
      "balance_loss_mlp": 1.01454389,
      "epoch": 0.7251465504283782,
      "flos": 25012608537600.0,
      "grad_norm": 1.7643136579579441,
      "language_loss": 0.76437026,
      "learning_rate": 7.413541734307393e-07,
      "loss": 0.78559864,
      "num_input_tokens_seen": 260242025,
      "step": 12061,
      "time_per_iteration": 2.5268454551696777
    },
    {
      "auxiliary_loss_clip": 0.01102949,
      "auxiliary_loss_mlp": 0.00784429,
      "balance_loss_clip": 1.03619266,
      "balance_loss_mlp": 1.01295125,
      "epoch": 0.7252066736810462,
      "flos": 16690131135360.0,
      "grad_norm": 1.8093876542213756,
      "language_loss": 0.8109585,
      "learning_rate": 7.410515295192068e-07,
      "loss": 0.82983232,
      "num_input_tokens_seen": 260260015,
      "step": 12062,
      "time_per_iteration": 2.4611501693725586
    },
    {
      "auxiliary_loss_clip": 0.01059375,
      "auxiliary_loss_mlp": 0.01031209,
      "balance_loss_clip": 1.03579783,
      "balance_loss_mlp": 1.01683199,
      "epoch": 0.7252667969337141,
      "flos": 25703026830720.0,
      "grad_norm": 2.031485697906808,
      "language_loss": 0.69039035,
      "learning_rate": 7.407489333471262e-07,
      "loss": 0.7112962,
      "num_input_tokens_seen": 260278635,
      "step": 12063,
      "time_per_iteration": 2.625467538833618
    },
    {
      "auxiliary_loss_clip": 0.01069853,
      "auxiliary_loss_mlp": 0.01031412,
      "balance_loss_clip": 1.03425527,
      "balance_loss_mlp": 1.01947927,
      "epoch": 0.7253269201863821,
      "flos": 18259930195200.0,
      "grad_norm": 1.4214818276206647,
      "language_loss": 0.70028841,
      "learning_rate": 7.40446384925973e-07,
      "loss": 0.72130102,
      "num_input_tokens_seen": 260298510,
      "step": 12064,
      "time_per_iteration": 2.5612504482269287
    },
    {
      "auxiliary_loss_clip": 0.0108436,
      "auxiliary_loss_mlp": 0.0102763,
      "balance_loss_clip": 1.0359565,
      "balance_loss_mlp": 1.0154531,
      "epoch": 0.72538704343905,
      "flos": 20411805150720.0,
      "grad_norm": 1.7934062175647467,
      "language_loss": 0.90316552,
      "learning_rate": 7.401438842672192e-07,
      "loss": 0.92428547,
      "num_input_tokens_seen": 260317405,
      "step": 12065,
      "time_per_iteration": 2.5122909545898438
    },
    {
      "auxiliary_loss_clip": 0.01026617,
      "auxiliary_loss_mlp": 0.01002208,
      "balance_loss_clip": 1.01271915,
      "balance_loss_mlp": 1.00094438,
      "epoch": 0.725447166691718,
      "flos": 70151209706880.0,
      "grad_norm": 0.6514073900311427,
      "language_loss": 0.56107795,
      "learning_rate": 7.398414313823349e-07,
      "loss": 0.58136624,
      "num_input_tokens_seen": 260388085,
      "step": 12066,
      "time_per_iteration": 3.2351391315460205
    },
    {
      "auxiliary_loss_clip": 0.01057319,
      "auxiliary_loss_mlp": 0.0102951,
      "balance_loss_clip": 1.03594136,
      "balance_loss_mlp": 1.01796508,
      "epoch": 0.725507289944386,
      "flos": 27052334254080.0,
      "grad_norm": 1.9448730747002732,
      "language_loss": 0.76818871,
      "learning_rate": 7.395390262827897e-07,
      "loss": 0.78905696,
      "num_input_tokens_seen": 260406165,
      "step": 12067,
      "time_per_iteration": 2.6431632041931152
    },
    {
      "auxiliary_loss_clip": 0.01018336,
      "auxiliary_loss_mlp": 0.0100215,
      "balance_loss_clip": 1.01543272,
      "balance_loss_mlp": 1.00086808,
      "epoch": 0.725567413197054,
      "flos": 62921924778240.0,
      "grad_norm": 0.7294498130827991,
      "language_loss": 0.57062531,
      "learning_rate": 7.392366689800515e-07,
      "loss": 0.59083021,
      "num_input_tokens_seen": 260461365,
      "step": 12068,
      "time_per_iteration": 3.0213775634765625
    },
    {
      "auxiliary_loss_clip": 0.009979,
      "auxiliary_loss_mlp": 0.0100484,
      "balance_loss_clip": 1.01375771,
      "balance_loss_mlp": 1.00337946,
      "epoch": 0.7256275364497219,
      "flos": 60295957188480.0,
      "grad_norm": 0.6613310731478995,
      "language_loss": 0.55451965,
      "learning_rate": 7.389343594855848e-07,
      "loss": 0.57454705,
      "num_input_tokens_seen": 260523795,
      "step": 12069,
      "time_per_iteration": 3.221919059753418
    },
    {
      "auxiliary_loss_clip": 0.01070812,
      "auxiliary_loss_mlp": 0.01026827,
      "balance_loss_clip": 1.03617215,
      "balance_loss_mlp": 1.01608014,
      "epoch": 0.7256876597023899,
      "flos": 24498511130880.0,
      "grad_norm": 2.078612941503024,
      "language_loss": 0.79943287,
      "learning_rate": 7.38632097810854e-07,
      "loss": 0.8204093,
      "num_input_tokens_seen": 260544765,
      "step": 12070,
      "time_per_iteration": 2.588759183883667
    },
    {
      "auxiliary_loss_clip": 0.01078647,
      "auxiliary_loss_mlp": 0.01033976,
      "balance_loss_clip": 1.03508735,
      "balance_loss_mlp": 1.0222342,
      "epoch": 0.7257477829550578,
      "flos": 24352749740160.0,
      "grad_norm": 1.7428715643978419,
      "language_loss": 0.72020352,
      "learning_rate": 7.383298839673197e-07,
      "loss": 0.74132967,
      "num_input_tokens_seen": 260564340,
      "step": 12071,
      "time_per_iteration": 2.5696303844451904
    },
    {
      "auxiliary_loss_clip": 0.01104781,
      "auxiliary_loss_mlp": 0.01034634,
      "balance_loss_clip": 1.03749824,
      "balance_loss_mlp": 1.02328527,
      "epoch": 0.7258079062077258,
      "flos": 17202217380480.0,
      "grad_norm": 1.9036849757727847,
      "language_loss": 0.70424938,
      "learning_rate": 7.380277179664436e-07,
      "loss": 0.72564352,
      "num_input_tokens_seen": 260582565,
      "step": 12072,
      "time_per_iteration": 2.4668962955474854
    },
    {
      "auxiliary_loss_clip": 0.01070394,
      "auxiliary_loss_mlp": 0.01029567,
      "balance_loss_clip": 1.03344893,
      "balance_loss_mlp": 1.01630473,
      "epoch": 0.7258680294603939,
      "flos": 21580338401280.0,
      "grad_norm": 1.9044703883633056,
      "language_loss": 0.78595936,
      "learning_rate": 7.377255998196821e-07,
      "loss": 0.80695903,
      "num_input_tokens_seen": 260601700,
      "step": 12073,
      "time_per_iteration": 2.5717856884002686
    },
    {
      "auxiliary_loss_clip": 0.01083957,
      "auxiliary_loss_mlp": 0.01028966,
      "balance_loss_clip": 1.03698766,
      "balance_loss_mlp": 1.01711631,
      "epoch": 0.7259281527130618,
      "flos": 34855399036800.0,
      "grad_norm": 1.473466774468194,
      "language_loss": 0.70346648,
      "learning_rate": 7.374235295384923e-07,
      "loss": 0.72459567,
      "num_input_tokens_seen": 260623040,
      "step": 12074,
      "time_per_iteration": 2.677253484725952
    },
    {
      "auxiliary_loss_clip": 0.0108701,
      "auxiliary_loss_mlp": 0.01027795,
      "balance_loss_clip": 1.0359273,
      "balance_loss_mlp": 1.01485467,
      "epoch": 0.7259882759657298,
      "flos": 25404644551680.0,
      "grad_norm": 1.7400756138624678,
      "language_loss": 0.74372625,
      "learning_rate": 7.371215071343302e-07,
      "loss": 0.76487434,
      "num_input_tokens_seen": 260642735,
      "step": 12075,
      "time_per_iteration": 2.5911202430725098
    },
    {
      "auxiliary_loss_clip": 0.01094969,
      "auxiliary_loss_mlp": 0.01030425,
      "balance_loss_clip": 1.03616822,
      "balance_loss_mlp": 1.01811635,
      "epoch": 0.7260483992183977,
      "flos": 62953630531200.0,
      "grad_norm": 1.4471211753394115,
      "language_loss": 0.63434672,
      "learning_rate": 7.368195326186458e-07,
      "loss": 0.65560061,
      "num_input_tokens_seen": 260669935,
      "step": 12076,
      "time_per_iteration": 2.896605968475342
    },
    {
      "auxiliary_loss_clip": 0.01068898,
      "auxiliary_loss_mlp": 0.01028761,
      "balance_loss_clip": 1.03445029,
      "balance_loss_mlp": 1.01648235,
      "epoch": 0.7261085224710657,
      "flos": 26467528924800.0,
      "grad_norm": 1.779625291235308,
      "language_loss": 0.78761232,
      "learning_rate": 7.365176060028912e-07,
      "loss": 0.80858892,
      "num_input_tokens_seen": 260689605,
      "step": 12077,
      "time_per_iteration": 2.656214475631714
    },
    {
      "auxiliary_loss_clip": 0.01034142,
      "auxiliary_loss_mlp": 0.00761894,
      "balance_loss_clip": 1.01068807,
      "balance_loss_mlp": 0.99981654,
      "epoch": 0.7261686457237336,
      "flos": 66772732187520.0,
      "grad_norm": 0.8839711671155598,
      "language_loss": 0.65010571,
      "learning_rate": 7.362157272985163e-07,
      "loss": 0.66806608,
      "num_input_tokens_seen": 260748265,
      "step": 12078,
      "time_per_iteration": 3.087881565093994
    },
    {
      "auxiliary_loss_clip": 0.01025205,
      "auxiliary_loss_mlp": 0.01000259,
      "balance_loss_clip": 1.01211166,
      "balance_loss_mlp": 0.99908525,
      "epoch": 0.7262287689764017,
      "flos": 69999594399360.0,
      "grad_norm": 0.7254744502548687,
      "language_loss": 0.59272003,
      "learning_rate": 7.359138965169671e-07,
      "loss": 0.61297476,
      "num_input_tokens_seen": 260816715,
      "step": 12079,
      "time_per_iteration": 3.243366241455078
    },
    {
      "auxiliary_loss_clip": 0.01060475,
      "auxiliary_loss_mlp": 0.01031249,
      "balance_loss_clip": 1.03429389,
      "balance_loss_mlp": 1.01864231,
      "epoch": 0.7262888922290696,
      "flos": 23805435231360.0,
      "grad_norm": 1.9454027931627436,
      "language_loss": 0.64561987,
      "learning_rate": 7.356121136696895e-07,
      "loss": 0.66653711,
      "num_input_tokens_seen": 260836765,
      "step": 12080,
      "time_per_iteration": 2.6117634773254395
    },
    {
      "auxiliary_loss_clip": 0.01062258,
      "auxiliary_loss_mlp": 0.01026702,
      "balance_loss_clip": 1.03513932,
      "balance_loss_mlp": 1.01346993,
      "epoch": 0.7263490154817376,
      "flos": 19500320603520.0,
      "grad_norm": 2.4216944007441272,
      "language_loss": 0.69773746,
      "learning_rate": 7.35310378768128e-07,
      "loss": 0.71862704,
      "num_input_tokens_seen": 260854610,
      "step": 12081,
      "time_per_iteration": 2.568429470062256
    },
    {
      "auxiliary_loss_clip": 0.0111064,
      "auxiliary_loss_mlp": 0.01028066,
      "balance_loss_clip": 1.03892088,
      "balance_loss_mlp": 1.01583529,
      "epoch": 0.7264091387344055,
      "flos": 16286243633280.0,
      "grad_norm": 1.8282647643934549,
      "language_loss": 0.81226361,
      "learning_rate": 7.350086918237237e-07,
      "loss": 0.83365071,
      "num_input_tokens_seen": 260871620,
      "step": 12082,
      "time_per_iteration": 2.4535439014434814
    },
    {
      "auxiliary_loss_clip": 0.0109942,
      "auxiliary_loss_mlp": 0.01035395,
      "balance_loss_clip": 1.03548002,
      "balance_loss_mlp": 1.02175188,
      "epoch": 0.7264692619870735,
      "flos": 24352031468160.0,
      "grad_norm": 1.8091799150093841,
      "language_loss": 0.77245378,
      "learning_rate": 7.347070528479158e-07,
      "loss": 0.7938019,
      "num_input_tokens_seen": 260890490,
      "step": 12083,
      "time_per_iteration": 2.505014181137085
    },
    {
      "auxiliary_loss_clip": 0.01109663,
      "auxiliary_loss_mlp": 0.01030074,
      "balance_loss_clip": 1.03907013,
      "balance_loss_mlp": 1.01762843,
      "epoch": 0.7265293852397414,
      "flos": 25119478477440.0,
      "grad_norm": 1.739594900242687,
      "language_loss": 0.73149818,
      "learning_rate": 7.344054618521433e-07,
      "loss": 0.75289547,
      "num_input_tokens_seen": 260909700,
      "step": 12084,
      "time_per_iteration": 2.5074477195739746
    },
    {
      "auxiliary_loss_clip": 0.01110321,
      "auxiliary_loss_mlp": 0.01031249,
      "balance_loss_clip": 1.0383575,
      "balance_loss_mlp": 1.01876807,
      "epoch": 0.7265895084924094,
      "flos": 22638230784000.0,
      "grad_norm": 1.6987177193557668,
      "language_loss": 0.77904421,
      "learning_rate": 7.34103918847843e-07,
      "loss": 0.80045992,
      "num_input_tokens_seen": 260929090,
      "step": 12085,
      "time_per_iteration": 2.4601125717163086
    },
    {
      "auxiliary_loss_clip": 0.01096509,
      "auxiliary_loss_mlp": 0.01036336,
      "balance_loss_clip": 1.03654861,
      "balance_loss_mlp": 1.02434349,
      "epoch": 0.7266496317450775,
      "flos": 23368222886400.0,
      "grad_norm": 1.7056139911658998,
      "language_loss": 0.72291797,
      "learning_rate": 7.338024238464493e-07,
      "loss": 0.74424648,
      "num_input_tokens_seen": 260946615,
      "step": 12086,
      "time_per_iteration": 3.8972010612487793
    },
    {
      "auxiliary_loss_clip": 0.01070111,
      "auxiliary_loss_mlp": 0.01037722,
      "balance_loss_clip": 1.03504872,
      "balance_loss_mlp": 1.02384639,
      "epoch": 0.7267097549977454,
      "flos": 28074603323520.0,
      "grad_norm": 1.786568369976785,
      "language_loss": 0.69613171,
      "learning_rate": 7.335009768593938e-07,
      "loss": 0.71721005,
      "num_input_tokens_seen": 260968515,
      "step": 12087,
      "time_per_iteration": 2.6122074127197266
    },
    {
      "auxiliary_loss_clip": 0.01111247,
      "auxiliary_loss_mlp": 0.01034559,
      "balance_loss_clip": 1.03990328,
      "balance_loss_mlp": 1.02170849,
      "epoch": 0.7267698782504134,
      "flos": 22195523658240.0,
      "grad_norm": 1.6591495146513835,
      "language_loss": 0.7868706,
      "learning_rate": 7.331995778981088e-07,
      "loss": 0.80832863,
      "num_input_tokens_seen": 260986790,
      "step": 12088,
      "time_per_iteration": 2.4600605964660645
    },
    {
      "auxiliary_loss_clip": 0.0109193,
      "auxiliary_loss_mlp": 0.01037103,
      "balance_loss_clip": 1.03561759,
      "balance_loss_mlp": 1.02497375,
      "epoch": 0.7268300015030813,
      "flos": 18514859996160.0,
      "grad_norm": 1.750284055885843,
      "language_loss": 0.73755956,
      "learning_rate": 7.328982269740221e-07,
      "loss": 0.75884986,
      "num_input_tokens_seen": 261004925,
      "step": 12089,
      "time_per_iteration": 2.4675662517547607
    },
    {
      "auxiliary_loss_clip": 0.01085884,
      "auxiliary_loss_mlp": 0.01037074,
      "balance_loss_clip": 1.03726053,
      "balance_loss_mlp": 1.0243063,
      "epoch": 0.7268901247557493,
      "flos": 23986029836160.0,
      "grad_norm": 1.6839338805212147,
      "language_loss": 0.71141034,
      "learning_rate": 7.325969240985616e-07,
      "loss": 0.73263991,
      "num_input_tokens_seen": 261023895,
      "step": 12090,
      "time_per_iteration": 2.5432631969451904
    },
    {
      "auxiliary_loss_clip": 0.01053722,
      "auxiliary_loss_mlp": 0.01030365,
      "balance_loss_clip": 1.03311217,
      "balance_loss_mlp": 1.01719272,
      "epoch": 0.7269502480084172,
      "flos": 32088087429120.0,
      "grad_norm": 2.023584258104744,
      "language_loss": 0.77505851,
      "learning_rate": 7.322956692831528e-07,
      "loss": 0.79589939,
      "num_input_tokens_seen": 261045445,
      "step": 12091,
      "time_per_iteration": 2.6982319355010986
    },
    {
      "auxiliary_loss_clip": 0.01089219,
      "auxiliary_loss_mlp": 0.0078541,
      "balance_loss_clip": 1.03398526,
      "balance_loss_mlp": 1.01048934,
      "epoch": 0.7270103712610853,
      "flos": 19062785036160.0,
      "grad_norm": 1.985390735686807,
      "language_loss": 0.71077895,
      "learning_rate": 7.319944625392205e-07,
      "loss": 0.72952521,
      "num_input_tokens_seen": 261064275,
      "step": 12092,
      "time_per_iteration": 5.292281150817871
    },
    {
      "auxiliary_loss_clip": 0.01096932,
      "auxiliary_loss_mlp": 0.01030834,
      "balance_loss_clip": 1.03749704,
      "balance_loss_mlp": 1.01812005,
      "epoch": 0.7270704945137532,
      "flos": 34532921710080.0,
      "grad_norm": 2.4519523245384667,
      "language_loss": 0.61069274,
      "learning_rate": 7.31693303878184e-07,
      "loss": 0.63197041,
      "num_input_tokens_seen": 261083310,
      "step": 12093,
      "time_per_iteration": 2.599919080734253
    },
    {
      "auxiliary_loss_clip": 0.01083774,
      "auxiliary_loss_mlp": 0.01032849,
      "balance_loss_clip": 1.03720212,
      "balance_loss_mlp": 1.02082646,
      "epoch": 0.7271306177664212,
      "flos": 21507583403520.0,
      "grad_norm": 1.5421839824071106,
      "language_loss": 0.75120366,
      "learning_rate": 7.313921933114644e-07,
      "loss": 0.77236998,
      "num_input_tokens_seen": 261103460,
      "step": 12094,
      "time_per_iteration": 2.5601837635040283
    },
    {
      "auxiliary_loss_clip": 0.01071814,
      "auxiliary_loss_mlp": 0.01029567,
      "balance_loss_clip": 1.03479648,
      "balance_loss_mlp": 1.01774156,
      "epoch": 0.7271907410190891,
      "flos": 22272444633600.0,
      "grad_norm": 1.9504807486963558,
      "language_loss": 0.8508302,
      "learning_rate": 7.310911308504808e-07,
      "loss": 0.87184405,
      "num_input_tokens_seen": 261121375,
      "step": 12095,
      "time_per_iteration": 2.568286418914795
    },
    {
      "auxiliary_loss_clip": 0.01093457,
      "auxiliary_loss_mlp": 0.01036108,
      "balance_loss_clip": 1.03504193,
      "balance_loss_mlp": 1.02281046,
      "epoch": 0.7272508642717571,
      "flos": 22893124671360.0,
      "grad_norm": 1.6534090229212621,
      "language_loss": 0.77680135,
      "learning_rate": 7.307901165066479e-07,
      "loss": 0.79809701,
      "num_input_tokens_seen": 261141105,
      "step": 12096,
      "time_per_iteration": 2.546279191970825
    },
    {
      "auxiliary_loss_clip": 0.01107624,
      "auxiliary_loss_mlp": 0.01035108,
      "balance_loss_clip": 1.03816819,
      "balance_loss_mlp": 1.02298999,
      "epoch": 0.727310987524425,
      "flos": 11655886331520.0,
      "grad_norm": 2.183348696327486,
      "language_loss": 0.71996117,
      "learning_rate": 7.30489150291381e-07,
      "loss": 0.7413885,
      "num_input_tokens_seen": 261159255,
      "step": 12097,
      "time_per_iteration": 3.9706411361694336
    },
    {
      "auxiliary_loss_clip": 0.01097316,
      "auxiliary_loss_mlp": 0.00785194,
      "balance_loss_clip": 1.03751874,
      "balance_loss_mlp": 1.00964057,
      "epoch": 0.727371110777093,
      "flos": 24535319592960.0,
      "grad_norm": 1.7909135868610941,
      "language_loss": 0.765881,
      "learning_rate": 7.301882322160935e-07,
      "loss": 0.78470612,
      "num_input_tokens_seen": 261177960,
      "step": 12098,
      "time_per_iteration": 2.5226917266845703
    },
    {
      "auxiliary_loss_clip": 0.0108499,
      "auxiliary_loss_mlp": 0.01028352,
      "balance_loss_clip": 1.03374183,
      "balance_loss_mlp": 1.01561451,
      "epoch": 0.7274312340297611,
      "flos": 74739835405440.0,
      "grad_norm": 1.5921925280659814,
      "language_loss": 0.67428392,
      "learning_rate": 7.298873622921952e-07,
      "loss": 0.6954174,
      "num_input_tokens_seen": 261205660,
      "step": 12099,
      "time_per_iteration": 2.9600038528442383
    },
    {
      "auxiliary_loss_clip": 0.01094425,
      "auxiliary_loss_mlp": 0.01037034,
      "balance_loss_clip": 1.03392446,
      "balance_loss_mlp": 1.02126312,
      "epoch": 0.727491357282429,
      "flos": 22342865247360.0,
      "grad_norm": 1.8426247410635586,
      "language_loss": 0.72587132,
      "learning_rate": 7.29586540531095e-07,
      "loss": 0.74718589,
      "num_input_tokens_seen": 261225185,
      "step": 12100,
      "time_per_iteration": 2.4981589317321777
    },
    {
      "auxiliary_loss_clip": 0.0109872,
      "auxiliary_loss_mlp": 0.01034599,
      "balance_loss_clip": 1.03840923,
      "balance_loss_mlp": 1.02257693,
      "epoch": 0.727551480535097,
      "flos": 23297550877440.0,
      "grad_norm": 1.325732827677211,
      "language_loss": 0.74673867,
      "learning_rate": 7.292857669442005e-07,
      "loss": 0.76807189,
      "num_input_tokens_seen": 261247965,
      "step": 12101,
      "time_per_iteration": 2.5331127643585205
    },
    {
      "auxiliary_loss_clip": 0.01065417,
      "auxiliary_loss_mlp": 0.01028608,
      "balance_loss_clip": 1.03697205,
      "balance_loss_mlp": 1.01682448,
      "epoch": 0.7276116037877649,
      "flos": 21470559459840.0,
      "grad_norm": 1.7615616465733526,
      "language_loss": 0.82319009,
      "learning_rate": 7.289850415429177e-07,
      "loss": 0.8441304,
      "num_input_tokens_seen": 261267585,
      "step": 12102,
      "time_per_iteration": 2.5522501468658447
    },
    {
      "auxiliary_loss_clip": 0.01095292,
      "auxiliary_loss_mlp": 0.01031517,
      "balance_loss_clip": 1.0367589,
      "balance_loss_mlp": 1.01952469,
      "epoch": 0.7276717270404329,
      "flos": 21464059098240.0,
      "grad_norm": 2.356109842589698,
      "language_loss": 0.81421649,
      "learning_rate": 7.286843643386495e-07,
      "loss": 0.8354845,
      "num_input_tokens_seen": 261285200,
      "step": 12103,
      "time_per_iteration": 2.4886813163757324
    },
    {
      "auxiliary_loss_clip": 0.01085406,
      "auxiliary_loss_mlp": 0.01024279,
      "balance_loss_clip": 1.03722668,
      "balance_loss_mlp": 1.01192331,
      "epoch": 0.7277318502931008,
      "flos": 16837221329280.0,
      "grad_norm": 1.5948474048661339,
      "language_loss": 0.66325122,
      "learning_rate": 7.283837353427968e-07,
      "loss": 0.68434805,
      "num_input_tokens_seen": 261303645,
      "step": 12104,
      "time_per_iteration": 2.4910850524902344
    },
    {
      "auxiliary_loss_clip": 0.01074002,
      "auxiliary_loss_mlp": 0.01030787,
      "balance_loss_clip": 1.03629148,
      "balance_loss_mlp": 1.01879406,
      "epoch": 0.7277919735457689,
      "flos": 33400550476800.0,
      "grad_norm": 1.6939034048727415,
      "language_loss": 0.65817595,
      "learning_rate": 7.280831545667611e-07,
      "loss": 0.6792239,
      "num_input_tokens_seen": 261323265,
      "step": 12105,
      "time_per_iteration": 2.6548991203308105
    },
    {
      "auxiliary_loss_clip": 0.01107874,
      "auxiliary_loss_mlp": 0.01031937,
      "balance_loss_clip": 1.03852963,
      "balance_loss_mlp": 1.01953304,
      "epoch": 0.7278520967984368,
      "flos": 19206499351680.0,
      "grad_norm": 2.40070127107402,
      "language_loss": 0.75826406,
      "learning_rate": 7.27782622021939e-07,
      "loss": 0.77966213,
      "num_input_tokens_seen": 261339745,
      "step": 12106,
      "time_per_iteration": 2.4392592906951904
    },
    {
      "auxiliary_loss_clip": 0.0109763,
      "auxiliary_loss_mlp": 0.01032175,
      "balance_loss_clip": 1.03637004,
      "balance_loss_mlp": 1.01919925,
      "epoch": 0.7279122200511048,
      "flos": 34094667870720.0,
      "grad_norm": 2.210961204880018,
      "language_loss": 0.70262212,
      "learning_rate": 7.274821377197273e-07,
      "loss": 0.72392011,
      "num_input_tokens_seen": 261359310,
      "step": 12107,
      "time_per_iteration": 2.5871362686157227
    },
    {
      "auxiliary_loss_clip": 0.01087768,
      "auxiliary_loss_mlp": 0.01031879,
      "balance_loss_clip": 1.03368664,
      "balance_loss_mlp": 1.01961279,
      "epoch": 0.7279723433037727,
      "flos": 54599049348480.0,
      "grad_norm": 1.6505647132033816,
      "language_loss": 0.75248361,
      "learning_rate": 7.271817016715205e-07,
      "loss": 0.77368009,
      "num_input_tokens_seen": 261384640,
      "step": 12108,
      "time_per_iteration": 2.8128466606140137
    },
    {
      "auxiliary_loss_clip": 0.01106494,
      "auxiliary_loss_mlp": 0.01032013,
      "balance_loss_clip": 1.03681302,
      "balance_loss_mlp": 1.01952529,
      "epoch": 0.7280324665564407,
      "flos": 36137482156800.0,
      "grad_norm": 1.5067594216334428,
      "language_loss": 0.66851044,
      "learning_rate": 7.268813138887124e-07,
      "loss": 0.68989551,
      "num_input_tokens_seen": 261405290,
      "step": 12109,
      "time_per_iteration": 2.6085293292999268
    },
    {
      "auxiliary_loss_clip": 0.01068849,
      "auxiliary_loss_mlp": 0.01029309,
      "balance_loss_clip": 1.03734112,
      "balance_loss_mlp": 1.01682806,
      "epoch": 0.7280925898091086,
      "flos": 11618539165440.0,
      "grad_norm": 2.152472722496652,
      "language_loss": 0.6315912,
      "learning_rate": 7.265809743826912e-07,
      "loss": 0.65257275,
      "num_input_tokens_seen": 261419710,
      "step": 12110,
      "time_per_iteration": 2.5594074726104736
    },
    {
      "auxiliary_loss_clip": 0.01073521,
      "auxiliary_loss_mlp": 0.0102916,
      "balance_loss_clip": 1.0340215,
      "balance_loss_mlp": 1.01623178,
      "epoch": 0.7281527130617766,
      "flos": 34277094069120.0,
      "grad_norm": 2.133492304068536,
      "language_loss": 0.58019507,
      "learning_rate": 7.26280683164847e-07,
      "loss": 0.60122186,
      "num_input_tokens_seen": 261442385,
      "step": 12111,
      "time_per_iteration": 2.69405198097229
    },
    {
      "auxiliary_loss_clip": 0.01058253,
      "auxiliary_loss_mlp": 0.01033407,
      "balance_loss_clip": 1.03869963,
      "balance_loss_mlp": 1.02107465,
      "epoch": 0.7282128363144446,
      "flos": 13918043018880.0,
      "grad_norm": 2.130579804929628,
      "language_loss": 0.74094379,
      "learning_rate": 7.259804402465677e-07,
      "loss": 0.76186043,
      "num_input_tokens_seen": 261459805,
      "step": 12112,
      "time_per_iteration": 2.6144602298736572
    },
    {
      "auxiliary_loss_clip": 0.01095518,
      "auxiliary_loss_mlp": 0.01032037,
      "balance_loss_clip": 1.03666496,
      "balance_loss_mlp": 1.02026498,
      "epoch": 0.7282729595671126,
      "flos": 20777627214720.0,
      "grad_norm": 1.7875389044051893,
      "language_loss": 0.66821688,
      "learning_rate": 7.25680245639237e-07,
      "loss": 0.6894924,
      "num_input_tokens_seen": 261477175,
      "step": 12113,
      "time_per_iteration": 2.542703151702881
    },
    {
      "auxiliary_loss_clip": 0.01066408,
      "auxiliary_loss_mlp": 0.01030429,
      "balance_loss_clip": 1.03428376,
      "balance_loss_mlp": 1.0180198,
      "epoch": 0.7283330828197806,
      "flos": 16325422392960.0,
      "grad_norm": 1.7569735550944434,
      "language_loss": 0.73304653,
      "learning_rate": 7.253800993542399e-07,
      "loss": 0.75401491,
      "num_input_tokens_seen": 261494990,
      "step": 12114,
      "time_per_iteration": 2.5143110752105713
    },
    {
      "auxiliary_loss_clip": 0.010846,
      "auxiliary_loss_mlp": 0.01029392,
      "balance_loss_clip": 1.03680158,
      "balance_loss_mlp": 1.01733398,
      "epoch": 0.7283932060724485,
      "flos": 27490193043840.0,
      "grad_norm": 3.2340551869252363,
      "language_loss": 0.68026924,
      "learning_rate": 7.250800014029564e-07,
      "loss": 0.70140916,
      "num_input_tokens_seen": 261514445,
      "step": 12115,
      "time_per_iteration": 2.562067747116089
    },
    {
      "auxiliary_loss_clip": 0.01108366,
      "auxiliary_loss_mlp": 0.0102848,
      "balance_loss_clip": 1.03693914,
      "balance_loss_mlp": 1.01602888,
      "epoch": 0.7284533293251165,
      "flos": 18367877543040.0,
      "grad_norm": 1.5589159961447048,
      "language_loss": 0.59752607,
      "learning_rate": 7.247799517967674e-07,
      "loss": 0.61889458,
      "num_input_tokens_seen": 261533565,
      "step": 12116,
      "time_per_iteration": 2.465832471847534
    },
    {
      "auxiliary_loss_clip": 0.01093857,
      "auxiliary_loss_mlp": 0.01029133,
      "balance_loss_clip": 1.03625727,
      "balance_loss_mlp": 1.01720572,
      "epoch": 0.7285134525777844,
      "flos": 21725525174400.0,
      "grad_norm": 2.1464667573619987,
      "language_loss": 0.72947896,
      "learning_rate": 7.2447995054705e-07,
      "loss": 0.75070882,
      "num_input_tokens_seen": 261553795,
      "step": 12117,
      "time_per_iteration": 2.5057084560394287
    },
    {
      "auxiliary_loss_clip": 0.01090971,
      "auxiliary_loss_mlp": 0.01031989,
      "balance_loss_clip": 1.03504086,
      "balance_loss_mlp": 1.01926303,
      "epoch": 0.7285735758304525,
      "flos": 20741357456640.0,
      "grad_norm": 1.9156027623374166,
      "language_loss": 0.69684726,
      "learning_rate": 7.241799976651807e-07,
      "loss": 0.71807683,
      "num_input_tokens_seen": 261572565,
      "step": 12118,
      "time_per_iteration": 2.499309778213501
    },
    {
      "auxiliary_loss_clip": 0.01049055,
      "auxiliary_loss_mlp": 0.01033897,
      "balance_loss_clip": 1.03242564,
      "balance_loss_mlp": 1.02233982,
      "epoch": 0.7286336990831204,
      "flos": 17310954827520.0,
      "grad_norm": 1.5895184201469303,
      "language_loss": 0.84561068,
      "learning_rate": 7.238800931625346e-07,
      "loss": 0.86644018,
      "num_input_tokens_seen": 261590910,
      "step": 12119,
      "time_per_iteration": 2.579619884490967
    },
    {
      "auxiliary_loss_clip": 0.01105595,
      "auxiliary_loss_mlp": 0.01029089,
      "balance_loss_clip": 1.03642988,
      "balance_loss_mlp": 1.01673877,
      "epoch": 0.7286938223357884,
      "flos": 19787390098560.0,
      "grad_norm": 2.06661196787075,
      "language_loss": 0.81058902,
      "learning_rate": 7.235802370504831e-07,
      "loss": 0.83193588,
      "num_input_tokens_seen": 261606005,
      "step": 12120,
      "time_per_iteration": 2.4286489486694336
    },
    {
      "auxiliary_loss_clip": 0.01067924,
      "auxiliary_loss_mlp": 0.01036705,
      "balance_loss_clip": 1.03572428,
      "balance_loss_mlp": 1.02384186,
      "epoch": 0.7287539455884563,
      "flos": 15340859625600.0,
      "grad_norm": 2.1353127034616324,
      "language_loss": 0.78884685,
      "learning_rate": 7.232804293403963e-07,
      "loss": 0.80989313,
      "num_input_tokens_seen": 261622305,
      "step": 12121,
      "time_per_iteration": 2.5597317218780518
    },
    {
      "auxiliary_loss_clip": 0.0110532,
      "auxiliary_loss_mlp": 0.01029836,
      "balance_loss_clip": 1.03410387,
      "balance_loss_mlp": 1.0170331,
      "epoch": 0.7288140688411243,
      "flos": 25192484870400.0,
      "grad_norm": 1.7137812353684079,
      "language_loss": 0.69354594,
      "learning_rate": 7.229806700436441e-07,
      "loss": 0.71489751,
      "num_input_tokens_seen": 261642465,
      "step": 12122,
      "time_per_iteration": 2.4917633533477783
    },
    {
      "auxiliary_loss_clip": 0.01057641,
      "auxiliary_loss_mlp": 0.01029208,
      "balance_loss_clip": 1.0318979,
      "balance_loss_mlp": 1.01764441,
      "epoch": 0.7288741920937922,
      "flos": 23984162328960.0,
      "grad_norm": 2.434902994526673,
      "language_loss": 0.87207139,
      "learning_rate": 7.226809591715923e-07,
      "loss": 0.89293993,
      "num_input_tokens_seen": 261661420,
      "step": 12123,
      "time_per_iteration": 2.6010611057281494
    },
    {
      "auxiliary_loss_clip": 0.01071349,
      "auxiliary_loss_mlp": 0.01026116,
      "balance_loss_clip": 1.03383207,
      "balance_loss_mlp": 1.0141356,
      "epoch": 0.7289343153464602,
      "flos": 22744921155840.0,
      "grad_norm": 1.872124618955195,
      "language_loss": 0.82915974,
      "learning_rate": 7.223812967356065e-07,
      "loss": 0.85013437,
      "num_input_tokens_seen": 261680865,
      "step": 12124,
      "time_per_iteration": 3.999394655227661
    },
    {
      "auxiliary_loss_clip": 0.01084653,
      "auxiliary_loss_mlp": 0.0102914,
      "balance_loss_clip": 1.03655601,
      "balance_loss_mlp": 1.01744568,
      "epoch": 0.7289944385991282,
      "flos": 24900028335360.0,
      "grad_norm": 1.8367033457742714,
      "language_loss": 0.67231667,
      "learning_rate": 7.220816827470499e-07,
      "loss": 0.69345462,
      "num_input_tokens_seen": 261701455,
      "step": 12125,
      "time_per_iteration": 2.5594820976257324
    },
    {
      "auxiliary_loss_clip": 0.01097662,
      "auxiliary_loss_mlp": 0.01035607,
      "balance_loss_clip": 1.03495657,
      "balance_loss_mlp": 1.0222199,
      "epoch": 0.7290545618517962,
      "flos": 22967064817920.0,
      "grad_norm": 1.7685340277093167,
      "language_loss": 0.75128347,
      "learning_rate": 7.217821172172855e-07,
      "loss": 0.77261609,
      "num_input_tokens_seen": 261721260,
      "step": 12126,
      "time_per_iteration": 2.503328323364258
    },
    {
      "auxiliary_loss_clip": 0.01012956,
      "auxiliary_loss_mlp": 0.01003667,
      "balance_loss_clip": 1.0101037,
      "balance_loss_mlp": 1.00257003,
      "epoch": 0.7291146851044642,
      "flos": 61901523216000.0,
      "grad_norm": 0.8287363126752945,
      "language_loss": 0.58662975,
      "learning_rate": 7.2148260015767e-07,
      "loss": 0.60679591,
      "num_input_tokens_seen": 261779370,
      "step": 12127,
      "time_per_iteration": 3.0622060298919678
    },
    {
      "auxiliary_loss_clip": 0.01070926,
      "auxiliary_loss_mlp": 0.01025979,
      "balance_loss_clip": 1.03445911,
      "balance_loss_mlp": 1.01458263,
      "epoch": 0.7291748083571321,
      "flos": 23330947547520.0,
      "grad_norm": 9.868314490620364,
      "language_loss": 0.68703234,
      "learning_rate": 7.21183131579562e-07,
      "loss": 0.70800138,
      "num_input_tokens_seen": 261798050,
      "step": 12128,
      "time_per_iteration": 2.5697333812713623
    },
    {
      "auxiliary_loss_clip": 0.01079772,
      "auxiliary_loss_mlp": 0.01034387,
      "balance_loss_clip": 1.03502595,
      "balance_loss_mlp": 1.02125049,
      "epoch": 0.7292349316098001,
      "flos": 28330000001280.0,
      "grad_norm": 1.9859632205981075,
      "language_loss": 0.65702462,
      "learning_rate": 7.20883711494319e-07,
      "loss": 0.67816615,
      "num_input_tokens_seen": 261817660,
      "step": 12129,
      "time_per_iteration": 2.572272300720215
    },
    {
      "auxiliary_loss_clip": 0.01102996,
      "auxiliary_loss_mlp": 0.01024404,
      "balance_loss_clip": 1.03634334,
      "balance_loss_mlp": 1.01248872,
      "epoch": 0.729295054862468,
      "flos": 24132222190080.0,
      "grad_norm": 2.495590145825102,
      "language_loss": 0.74374104,
      "learning_rate": 7.205843399132927e-07,
      "loss": 0.76501507,
      "num_input_tokens_seen": 261837935,
      "step": 12130,
      "time_per_iteration": 3.8999977111816406
    },
    {
      "auxiliary_loss_clip": 0.0108059,
      "auxiliary_loss_mlp": 0.01030048,
      "balance_loss_clip": 1.03467011,
      "balance_loss_mlp": 1.01753092,
      "epoch": 0.7293551781151361,
      "flos": 22816239609600.0,
      "grad_norm": 1.6413930632407545,
      "language_loss": 0.69903195,
      "learning_rate": 7.202850168478374e-07,
      "loss": 0.72013831,
      "num_input_tokens_seen": 261857575,
      "step": 12131,
      "time_per_iteration": 3.940873384475708
    },
    {
      "auxiliary_loss_clip": 0.01070646,
      "auxiliary_loss_mlp": 0.0102917,
      "balance_loss_clip": 1.03668511,
      "balance_loss_mlp": 1.01775026,
      "epoch": 0.729415301367804,
      "flos": 22126683242880.0,
      "grad_norm": 1.6469513284956319,
      "language_loss": 0.77522135,
      "learning_rate": 7.199857423093025e-07,
      "loss": 0.79621947,
      "num_input_tokens_seen": 261877265,
      "step": 12132,
      "time_per_iteration": 2.5600850582122803
    },
    {
      "auxiliary_loss_clip": 0.01094654,
      "auxiliary_loss_mlp": 0.01033928,
      "balance_loss_clip": 1.03567886,
      "balance_loss_mlp": 1.0223105,
      "epoch": 0.729475424620472,
      "flos": 12349608675840.0,
      "grad_norm": 2.2503498914172875,
      "language_loss": 0.79077697,
      "learning_rate": 7.196865163090358e-07,
      "loss": 0.81206286,
      "num_input_tokens_seen": 261893695,
      "step": 12133,
      "time_per_iteration": 2.4739465713500977
    },
    {
      "auxiliary_loss_clip": 0.01057137,
      "auxiliary_loss_mlp": 0.01028639,
      "balance_loss_clip": 1.03379488,
      "balance_loss_mlp": 1.01603889,
      "epoch": 0.7295355478731399,
      "flos": 22195308176640.0,
      "grad_norm": 1.8054816923496675,
      "language_loss": 0.72126245,
      "learning_rate": 7.193873388583846e-07,
      "loss": 0.74212027,
      "num_input_tokens_seen": 261911825,
      "step": 12134,
      "time_per_iteration": 2.590374708175659
    },
    {
      "auxiliary_loss_clip": 0.01086924,
      "auxiliary_loss_mlp": 0.01035337,
      "balance_loss_clip": 1.03775597,
      "balance_loss_mlp": 1.02274847,
      "epoch": 0.7295956711258079,
      "flos": 23222030532480.0,
      "grad_norm": 1.6303757249806112,
      "language_loss": 0.71407837,
      "learning_rate": 7.190882099686939e-07,
      "loss": 0.73530102,
      "num_input_tokens_seen": 261931190,
      "step": 12135,
      "time_per_iteration": 2.5872268676757812
    },
    {
      "auxiliary_loss_clip": 0.01073048,
      "auxiliary_loss_mlp": 0.01033688,
      "balance_loss_clip": 1.03635979,
      "balance_loss_mlp": 1.02118886,
      "epoch": 0.7296557943784758,
      "flos": 31869104163840.0,
      "grad_norm": 2.080626534780749,
      "language_loss": 0.62233913,
      "learning_rate": 7.187891296513075e-07,
      "loss": 0.64340639,
      "num_input_tokens_seen": 261951240,
      "step": 12136,
      "time_per_iteration": 4.03627347946167
    },
    {
      "auxiliary_loss_clip": 0.01087992,
      "auxiliary_loss_mlp": 0.00784563,
      "balance_loss_clip": 1.03473139,
      "balance_loss_mlp": 1.01282203,
      "epoch": 0.7297159176311439,
      "flos": 26651714889600.0,
      "grad_norm": 1.8292242131094316,
      "language_loss": 0.74551702,
      "learning_rate": 7.184900979175654e-07,
      "loss": 0.76424247,
      "num_input_tokens_seen": 261971605,
      "step": 12137,
      "time_per_iteration": 2.564791202545166
    },
    {
      "auxiliary_loss_clip": 0.01095748,
      "auxiliary_loss_mlp": 0.00785852,
      "balance_loss_clip": 1.03766274,
      "balance_loss_mlp": 1.01235557,
      "epoch": 0.7297760408838118,
      "flos": 24749562263040.0,
      "grad_norm": 1.5217106503866549,
      "language_loss": 0.74206787,
      "learning_rate": 7.181911147788069e-07,
      "loss": 0.76088387,
      "num_input_tokens_seen": 261990830,
      "step": 12138,
      "time_per_iteration": 2.5630712509155273
    },
    {
      "auxiliary_loss_clip": 0.0106901,
      "auxiliary_loss_mlp": 0.01027764,
      "balance_loss_clip": 1.03547955,
      "balance_loss_mlp": 1.01639771,
      "epoch": 0.7298361641364798,
      "flos": 18073768982400.0,
      "grad_norm": 2.058195805372477,
      "language_loss": 0.7172122,
      "learning_rate": 7.178921802463702e-07,
      "loss": 0.73817992,
      "num_input_tokens_seen": 262008190,
      "step": 12139,
      "time_per_iteration": 2.5305986404418945
    },
    {
      "auxiliary_loss_clip": 0.01090863,
      "auxiliary_loss_mlp": 0.01025113,
      "balance_loss_clip": 1.03631091,
      "balance_loss_mlp": 1.01409197,
      "epoch": 0.7298962873891478,
      "flos": 29895597169920.0,
      "grad_norm": 1.4676610942274326,
      "language_loss": 0.73236519,
      "learning_rate": 7.175932943315898e-07,
      "loss": 0.75352502,
      "num_input_tokens_seen": 262030460,
      "step": 12140,
      "time_per_iteration": 2.574385643005371
    },
    {
      "auxiliary_loss_clip": 0.0108027,
      "auxiliary_loss_mlp": 0.01028968,
      "balance_loss_clip": 1.03666496,
      "balance_loss_mlp": 1.01657605,
      "epoch": 0.7299564106418157,
      "flos": 32266096254720.0,
      "grad_norm": 1.771165640291079,
      "language_loss": 0.55632985,
      "learning_rate": 7.172944570458003e-07,
      "loss": 0.57742226,
      "num_input_tokens_seen": 262050830,
      "step": 12141,
      "time_per_iteration": 2.601027250289917
    },
    {
      "auxiliary_loss_clip": 0.01067363,
      "auxiliary_loss_mlp": 0.01028747,
      "balance_loss_clip": 1.03368926,
      "balance_loss_mlp": 1.01741612,
      "epoch": 0.7300165338944837,
      "flos": 22930292269440.0,
      "grad_norm": 1.4405938316127724,
      "language_loss": 0.72589171,
      "learning_rate": 7.169956684003342e-07,
      "loss": 0.74685282,
      "num_input_tokens_seen": 262071245,
      "step": 12142,
      "time_per_iteration": 2.6035315990448
    },
    {
      "auxiliary_loss_clip": 0.0110383,
      "auxiliary_loss_mlp": 0.01030758,
      "balance_loss_clip": 1.03628922,
      "balance_loss_mlp": 1.01973701,
      "epoch": 0.7300766571471516,
      "flos": 19828795501440.0,
      "grad_norm": 1.798745397451092,
      "language_loss": 0.73460209,
      "learning_rate": 7.16696928406521e-07,
      "loss": 0.75594795,
      "num_input_tokens_seen": 262087525,
      "step": 12143,
      "time_per_iteration": 2.4336977005004883
    },
    {
      "auxiliary_loss_clip": 0.01066358,
      "auxiliary_loss_mlp": 0.0103052,
      "balance_loss_clip": 1.03430152,
      "balance_loss_mlp": 1.01863468,
      "epoch": 0.7301367803998197,
      "flos": 24347829576960.0,
      "grad_norm": 1.9831028861358369,
      "language_loss": 0.66752279,
      "learning_rate": 7.163982370756882e-07,
      "loss": 0.68849152,
      "num_input_tokens_seen": 262107355,
      "step": 12144,
      "time_per_iteration": 2.578795909881592
    },
    {
      "auxiliary_loss_clip": 0.01079888,
      "auxiliary_loss_mlp": 0.01029324,
      "balance_loss_clip": 1.03616571,
      "balance_loss_mlp": 1.01779604,
      "epoch": 0.7301969036524876,
      "flos": 15304518040320.0,
      "grad_norm": 1.6753059808641482,
      "language_loss": 0.79220361,
      "learning_rate": 7.160995944191627e-07,
      "loss": 0.81329572,
      "num_input_tokens_seen": 262125645,
      "step": 12145,
      "time_per_iteration": 2.49168062210083
    },
    {
      "auxiliary_loss_clip": 0.01064341,
      "auxiliary_loss_mlp": 0.01029634,
      "balance_loss_clip": 1.03362203,
      "balance_loss_mlp": 1.01758254,
      "epoch": 0.7302570269051556,
      "flos": 23507268433920.0,
      "grad_norm": 1.9013261310962988,
      "language_loss": 0.91275251,
      "learning_rate": 7.158010004482702e-07,
      "loss": 0.93369222,
      "num_input_tokens_seen": 262144075,
      "step": 12146,
      "time_per_iteration": 2.566948175430298
    },
    {
      "auxiliary_loss_clip": 0.01101664,
      "auxiliary_loss_mlp": 0.01026799,
      "balance_loss_clip": 1.03589916,
      "balance_loss_mlp": 1.01581371,
      "epoch": 0.7303171501578235,
      "flos": 20523056549760.0,
      "grad_norm": 1.7362189369761787,
      "language_loss": 0.62057298,
      "learning_rate": 7.155024551743316e-07,
      "loss": 0.64185762,
      "num_input_tokens_seen": 262165940,
      "step": 12147,
      "time_per_iteration": 2.48987078666687
    },
    {
      "auxiliary_loss_clip": 0.0110726,
      "auxiliary_loss_mlp": 0.0103415,
      "balance_loss_clip": 1.03776598,
      "balance_loss_mlp": 1.02162671,
      "epoch": 0.7303772734104915,
      "flos": 18332613365760.0,
      "grad_norm": 1.8189118680423935,
      "language_loss": 0.75371218,
      "learning_rate": 7.152039586086693e-07,
      "loss": 0.77512622,
      "num_input_tokens_seen": 262184520,
      "step": 12148,
      "time_per_iteration": 2.448821783065796
    },
    {
      "auxiliary_loss_clip": 0.01014714,
      "auxiliary_loss_mlp": 0.00763043,
      "balance_loss_clip": 1.01300943,
      "balance_loss_mlp": 1.00126159,
      "epoch": 0.7304373966631594,
      "flos": 60654776100480.0,
      "grad_norm": 2.0807285473069195,
      "language_loss": 0.56692851,
      "learning_rate": 7.149055107626017e-07,
      "loss": 0.58470607,
      "num_input_tokens_seen": 262247070,
      "step": 12149,
      "time_per_iteration": 3.121549367904663
    },
    {
      "auxiliary_loss_clip": 0.01084145,
      "auxiliary_loss_mlp": 0.01032407,
      "balance_loss_clip": 1.03458107,
      "balance_loss_mlp": 1.02012825,
      "epoch": 0.7304975199158275,
      "flos": 19828077229440.0,
      "grad_norm": 1.864375819260993,
      "language_loss": 0.7362631,
      "learning_rate": 7.146071116474451e-07,
      "loss": 0.75742865,
      "num_input_tokens_seen": 262266605,
      "step": 12150,
      "time_per_iteration": 2.5382537841796875
    },
    {
      "auxiliary_loss_clip": 0.01106947,
      "auxiliary_loss_mlp": 0.01030004,
      "balance_loss_clip": 1.0364027,
      "balance_loss_mlp": 1.01770771,
      "epoch": 0.7305576431684954,
      "flos": 13223997452160.0,
      "grad_norm": 2.127109901952234,
      "language_loss": 0.83812892,
      "learning_rate": 7.143087612745158e-07,
      "loss": 0.8594985,
      "num_input_tokens_seen": 262283880,
      "step": 12151,
      "time_per_iteration": 2.461709976196289
    },
    {
      "auxiliary_loss_clip": 0.01073895,
      "auxiliary_loss_mlp": 0.0103285,
      "balance_loss_clip": 1.0343411,
      "balance_loss_mlp": 1.0199995,
      "epoch": 0.7306177664211634,
      "flos": 24060472773120.0,
      "grad_norm": 1.9053741579471506,
      "language_loss": 0.78414309,
      "learning_rate": 7.14010459655127e-07,
      "loss": 0.80521053,
      "num_input_tokens_seen": 262304155,
      "step": 12152,
      "time_per_iteration": 2.5980355739593506
    },
    {
      "auxiliary_loss_clip": 0.01071062,
      "auxiliary_loss_mlp": 0.01030978,
      "balance_loss_clip": 1.03856063,
      "balance_loss_mlp": 1.01914096,
      "epoch": 0.7306778896738314,
      "flos": 27089106802560.0,
      "grad_norm": 1.5602149627885848,
      "language_loss": 0.79708254,
      "learning_rate": 7.137122068005919e-07,
      "loss": 0.81810296,
      "num_input_tokens_seen": 262325660,
      "step": 12153,
      "time_per_iteration": 2.6157665252685547
    },
    {
      "auxiliary_loss_clip": 0.01096761,
      "auxiliary_loss_mlp": 0.01031994,
      "balance_loss_clip": 1.03674555,
      "balance_loss_mlp": 1.01972139,
      "epoch": 0.7307380129264993,
      "flos": 16690669839360.0,
      "grad_norm": 1.617713780939764,
      "language_loss": 0.67402518,
      "learning_rate": 7.134140027222173e-07,
      "loss": 0.69531274,
      "num_input_tokens_seen": 262344075,
      "step": 12154,
      "time_per_iteration": 2.52443790435791
    },
    {
      "auxiliary_loss_clip": 0.0106592,
      "auxiliary_loss_mlp": 0.01030977,
      "balance_loss_clip": 1.0369277,
      "balance_loss_mlp": 1.01906204,
      "epoch": 0.7307981361791673,
      "flos": 21725740656000.0,
      "grad_norm": 1.7662641630879095,
      "language_loss": 0.65752101,
      "learning_rate": 7.131158474313128e-07,
      "loss": 0.67848998,
      "num_input_tokens_seen": 262363305,
      "step": 12155,
      "time_per_iteration": 2.637516975402832
    },
    {
      "auxiliary_loss_clip": 0.01080885,
      "auxiliary_loss_mlp": 0.01025961,
      "balance_loss_clip": 1.03422606,
      "balance_loss_mlp": 1.0144279,
      "epoch": 0.7308582594318352,
      "flos": 18040659621120.0,
      "grad_norm": 1.7213057498786966,
      "language_loss": 0.81634414,
      "learning_rate": 7.128177409391851e-07,
      "loss": 0.8374126,
      "num_input_tokens_seen": 262380730,
      "step": 12156,
      "time_per_iteration": 2.500887155532837
    },
    {
      "auxiliary_loss_clip": 0.01064914,
      "auxiliary_loss_mlp": 0.01033524,
      "balance_loss_clip": 1.03322053,
      "balance_loss_mlp": 1.02193069,
      "epoch": 0.7309183826845033,
      "flos": 13844964798720.0,
      "grad_norm": 2.1564948654742357,
      "language_loss": 0.75436789,
      "learning_rate": 7.125196832571367e-07,
      "loss": 0.7753523,
      "num_input_tokens_seen": 262395480,
      "step": 12157,
      "time_per_iteration": 2.5256032943725586
    },
    {
      "auxiliary_loss_clip": 0.01088919,
      "auxiliary_loss_mlp": 0.01026278,
      "balance_loss_clip": 1.03489876,
      "balance_loss_mlp": 1.01565599,
      "epoch": 0.7309785059371712,
      "flos": 17019216564480.0,
      "grad_norm": 2.1149091433231426,
      "language_loss": 0.72738671,
      "learning_rate": 7.122216743964713e-07,
      "loss": 0.74853867,
      "num_input_tokens_seen": 262413340,
      "step": 12158,
      "time_per_iteration": 2.472226858139038
    },
    {
      "auxiliary_loss_clip": 0.01083329,
      "auxiliary_loss_mlp": 0.01032648,
      "balance_loss_clip": 1.03660393,
      "balance_loss_mlp": 1.02095342,
      "epoch": 0.7310386291898392,
      "flos": 26502398052480.0,
      "grad_norm": 1.5339658510692826,
      "language_loss": 0.85443014,
      "learning_rate": 7.119237143684896e-07,
      "loss": 0.87558991,
      "num_input_tokens_seen": 262433455,
      "step": 12159,
      "time_per_iteration": 2.5926854610443115
    },
    {
      "auxiliary_loss_clip": 0.01087547,
      "auxiliary_loss_mlp": 0.01031791,
      "balance_loss_clip": 1.03634381,
      "balance_loss_mlp": 1.01879716,
      "epoch": 0.7310987524425071,
      "flos": 16945922862720.0,
      "grad_norm": 2.1719780067199097,
      "language_loss": 0.73492551,
      "learning_rate": 7.116258031844895e-07,
      "loss": 0.75611883,
      "num_input_tokens_seen": 262450335,
      "step": 12160,
      "time_per_iteration": 2.505530595779419
    },
    {
      "auxiliary_loss_clip": 0.01095841,
      "auxiliary_loss_mlp": 0.01029986,
      "balance_loss_clip": 1.03621459,
      "balance_loss_mlp": 1.01768947,
      "epoch": 0.7311588756951751,
      "flos": 13845288021120.0,
      "grad_norm": 4.041954196852807,
      "language_loss": 0.72731662,
      "learning_rate": 7.113279408557675e-07,
      "loss": 0.74857485,
      "num_input_tokens_seen": 262468240,
      "step": 12161,
      "time_per_iteration": 2.4886934757232666
    },
    {
      "auxiliary_loss_clip": 0.0107576,
      "auxiliary_loss_mlp": 0.007856,
      "balance_loss_clip": 1.03442049,
      "balance_loss_mlp": 1.01074433,
      "epoch": 0.731218998947843,
      "flos": 28767894704640.0,
      "grad_norm": 1.658365968196064,
      "language_loss": 0.69311839,
      "learning_rate": 7.110301273936192e-07,
      "loss": 0.71173197,
      "num_input_tokens_seen": 262487045,
      "step": 12162,
      "time_per_iteration": 2.610635280609131
    },
    {
      "auxiliary_loss_clip": 0.01096674,
      "auxiliary_loss_mlp": 0.0102893,
      "balance_loss_clip": 1.03742051,
      "balance_loss_mlp": 1.0165143,
      "epoch": 0.7312791222005111,
      "flos": 27088783580160.0,
      "grad_norm": 1.7200663810451444,
      "language_loss": 0.66680902,
      "learning_rate": 7.107323628093382e-07,
      "loss": 0.68806505,
      "num_input_tokens_seen": 262504855,
      "step": 12163,
      "time_per_iteration": 3.9545681476593018
    },
    {
      "auxiliary_loss_clip": 0.01083334,
      "auxiliary_loss_mlp": 0.01028973,
      "balance_loss_clip": 1.0348618,
      "balance_loss_mlp": 1.01665318,
      "epoch": 0.731339245453179,
      "flos": 20924035050240.0,
      "grad_norm": 1.4195676560214838,
      "language_loss": 0.68352103,
      "learning_rate": 7.104346471142153e-07,
      "loss": 0.70464408,
      "num_input_tokens_seen": 262524920,
      "step": 12164,
      "time_per_iteration": 2.535433053970337
    },
    {
      "auxiliary_loss_clip": 0.01056443,
      "auxiliary_loss_mlp": 0.01032429,
      "balance_loss_clip": 1.03634596,
      "balance_loss_mlp": 1.02106237,
      "epoch": 0.731399368705847,
      "flos": 23075694524160.0,
      "grad_norm": 2.000782192096785,
      "language_loss": 0.73065889,
      "learning_rate": 7.101369803195391e-07,
      "loss": 0.75154757,
      "num_input_tokens_seen": 262545725,
      "step": 12165,
      "time_per_iteration": 2.611187696456909
    },
    {
      "auxiliary_loss_clip": 0.01096552,
      "auxiliary_loss_mlp": 0.01032837,
      "balance_loss_clip": 1.03569484,
      "balance_loss_mlp": 1.02050447,
      "epoch": 0.731459491958515,
      "flos": 23582681038080.0,
      "grad_norm": 1.8724166401356932,
      "language_loss": 0.76712817,
      "learning_rate": 7.098393624365988e-07,
      "loss": 0.78842211,
      "num_input_tokens_seen": 262565480,
      "step": 12166,
      "time_per_iteration": 2.5444936752319336
    },
    {
      "auxiliary_loss_clip": 0.010849,
      "auxiliary_loss_mlp": 0.01028588,
      "balance_loss_clip": 1.03746367,
      "balance_loss_mlp": 1.01689386,
      "epoch": 0.7315196152111829,
      "flos": 22379278659840.0,
      "grad_norm": 1.8078748632185937,
      "language_loss": 0.79643393,
      "learning_rate": 7.095417934766781e-07,
      "loss": 0.81756878,
      "num_input_tokens_seen": 262584145,
      "step": 12167,
      "time_per_iteration": 2.556309461593628
    },
    {
      "auxiliary_loss_clip": 0.01091775,
      "auxiliary_loss_mlp": 0.0103838,
      "balance_loss_clip": 1.0349052,
      "balance_loss_mlp": 1.0269593,
      "epoch": 0.7315797384638509,
      "flos": 26177047637760.0,
      "grad_norm": 1.598248579297874,
      "language_loss": 0.76937073,
      "learning_rate": 7.092442734510622e-07,
      "loss": 0.7906723,
      "num_input_tokens_seen": 262604045,
      "step": 12168,
      "time_per_iteration": 2.5405566692352295
    },
    {
      "auxiliary_loss_clip": 0.0109053,
      "auxiliary_loss_mlp": 0.01041971,
      "balance_loss_clip": 1.03346539,
      "balance_loss_mlp": 1.02709937,
      "epoch": 0.7316398617165188,
      "flos": 21506326427520.0,
      "grad_norm": 1.5925881837266163,
      "language_loss": 0.81629133,
      "learning_rate": 7.089468023710326e-07,
      "loss": 0.83761632,
      "num_input_tokens_seen": 262624540,
      "step": 12169,
      "time_per_iteration": 3.923776626586914
    },
    {
      "auxiliary_loss_clip": 0.01092199,
      "auxiliary_loss_mlp": 0.01035162,
      "balance_loss_clip": 1.03636134,
      "balance_loss_mlp": 1.02246046,
      "epoch": 0.7316999849691869,
      "flos": 30482557315200.0,
      "grad_norm": 1.7275232719825073,
      "language_loss": 0.6984548,
      "learning_rate": 7.08649380247871e-07,
      "loss": 0.71972841,
      "num_input_tokens_seen": 262644545,
      "step": 12170,
      "time_per_iteration": 4.0130650997161865
    },
    {
      "auxiliary_loss_clip": 0.01103708,
      "auxiliary_loss_mlp": 0.01030996,
      "balance_loss_clip": 1.03580594,
      "balance_loss_mlp": 1.01844919,
      "epoch": 0.7317601082218548,
      "flos": 21543781334400.0,
      "grad_norm": 2.2563000803045616,
      "language_loss": 0.70006883,
      "learning_rate": 7.083520070928533e-07,
      "loss": 0.72141588,
      "num_input_tokens_seen": 262662570,
      "step": 12171,
      "time_per_iteration": 2.4790258407592773
    },
    {
      "auxiliary_loss_clip": 0.01105116,
      "auxiliary_loss_mlp": 0.01034007,
      "balance_loss_clip": 1.036991,
      "balance_loss_mlp": 1.02204442,
      "epoch": 0.7318202314745228,
      "flos": 33251592775680.0,
      "grad_norm": 1.4798358863123173,
      "language_loss": 0.65642416,
      "learning_rate": 7.080546829172564e-07,
      "loss": 0.67781544,
      "num_input_tokens_seen": 262683245,
      "step": 12172,
      "time_per_iteration": 2.545663356781006
    },
    {
      "auxiliary_loss_clip": 0.01107223,
      "auxiliary_loss_mlp": 0.01028272,
      "balance_loss_clip": 1.03812122,
      "balance_loss_mlp": 1.01650012,
      "epoch": 0.7318803547271907,
      "flos": 20157054917760.0,
      "grad_norm": 2.2924174950720673,
      "language_loss": 0.6126498,
      "learning_rate": 7.077574077323564e-07,
      "loss": 0.63400477,
      "num_input_tokens_seen": 262701585,
      "step": 12173,
      "time_per_iteration": 2.4946556091308594
    },
    {
      "auxiliary_loss_clip": 0.01051495,
      "auxiliary_loss_mlp": 0.01027729,
      "balance_loss_clip": 1.03314364,
      "balance_loss_mlp": 1.01648128,
      "epoch": 0.7319404779798587,
      "flos": 20558536208640.0,
      "grad_norm": 2.003536136149973,
      "language_loss": 0.74096155,
      "learning_rate": 7.074601815494243e-07,
      "loss": 0.7617538,
      "num_input_tokens_seen": 262719295,
      "step": 12174,
      "time_per_iteration": 4.071456670761108
    },
    {
      "auxiliary_loss_clip": 0.011015,
      "auxiliary_loss_mlp": 0.0102565,
      "balance_loss_clip": 1.03538501,
      "balance_loss_mlp": 1.01435494,
      "epoch": 0.7320006012325266,
      "flos": 28695391102080.0,
      "grad_norm": 1.5822214535534571,
      "language_loss": 0.80860305,
      "learning_rate": 7.071630043797317e-07,
      "loss": 0.82987452,
      "num_input_tokens_seen": 262739995,
      "step": 12175,
      "time_per_iteration": 2.5340375900268555
    },
    {
      "auxiliary_loss_clip": 0.01082489,
      "auxiliary_loss_mlp": 0.01026196,
      "balance_loss_clip": 1.03391075,
      "balance_loss_mlp": 1.01486468,
      "epoch": 0.7320607244851947,
      "flos": 16362697731840.0,
      "grad_norm": 2.1877810998487446,
      "language_loss": 0.77081192,
      "learning_rate": 7.068658762345488e-07,
      "loss": 0.79189873,
      "num_input_tokens_seen": 262757680,
      "step": 12176,
      "time_per_iteration": 2.5229880809783936
    },
    {
      "auxiliary_loss_clip": 0.01093376,
      "auxiliary_loss_mlp": 0.01031002,
      "balance_loss_clip": 1.03711486,
      "balance_loss_mlp": 1.01982582,
      "epoch": 0.7321208477378626,
      "flos": 20955097336320.0,
      "grad_norm": 1.785670685780079,
      "language_loss": 0.76603872,
      "learning_rate": 7.065687971251399e-07,
      "loss": 0.78728253,
      "num_input_tokens_seen": 262776990,
      "step": 12177,
      "time_per_iteration": 2.5359249114990234
    },
    {
      "auxiliary_loss_clip": 0.01069404,
      "auxiliary_loss_mlp": 0.01036241,
      "balance_loss_clip": 1.03280807,
      "balance_loss_mlp": 1.02498209,
      "epoch": 0.7321809709905306,
      "flos": 13845072539520.0,
      "grad_norm": 3.2277009465400472,
      "language_loss": 0.74902439,
      "learning_rate": 7.06271767062772e-07,
      "loss": 0.7700808,
      "num_input_tokens_seen": 262795440,
      "step": 12178,
      "time_per_iteration": 2.5238635540008545
    },
    {
      "auxiliary_loss_clip": 0.0108552,
      "auxiliary_loss_mlp": 0.01032065,
      "balance_loss_clip": 1.03340244,
      "balance_loss_mlp": 1.02009642,
      "epoch": 0.7322410942431986,
      "flos": 26979938392320.0,
      "grad_norm": 1.826059033437472,
      "language_loss": 0.82664537,
      "learning_rate": 7.059747860587084e-07,
      "loss": 0.84782118,
      "num_input_tokens_seen": 262816385,
      "step": 12179,
      "time_per_iteration": 2.561383008956909
    },
    {
      "auxiliary_loss_clip": 0.01076566,
      "auxiliary_loss_mlp": 0.01038721,
      "balance_loss_clip": 1.03493857,
      "balance_loss_mlp": 1.02650774,
      "epoch": 0.7323012174958665,
      "flos": 17639717034240.0,
      "grad_norm": 1.645567605811073,
      "language_loss": 0.74722755,
      "learning_rate": 7.056778541242115e-07,
      "loss": 0.7683804,
      "num_input_tokens_seen": 262834955,
      "step": 12180,
      "time_per_iteration": 2.479093551635742
    },
    {
      "auxiliary_loss_clip": 0.01094636,
      "auxiliary_loss_mlp": 0.00784351,
      "balance_loss_clip": 1.03318477,
      "balance_loss_mlp": 1.00963068,
      "epoch": 0.7323613407485345,
      "flos": 32342765834880.0,
      "grad_norm": 1.9447181447400566,
      "language_loss": 0.79317242,
      "learning_rate": 7.053809712705396e-07,
      "loss": 0.81196231,
      "num_input_tokens_seen": 262853555,
      "step": 12181,
      "time_per_iteration": 2.5624895095825195
    },
    {
      "auxiliary_loss_clip": 0.01096512,
      "auxiliary_loss_mlp": 0.00783416,
      "balance_loss_clip": 1.03621399,
      "balance_loss_mlp": 1.00844932,
      "epoch": 0.7324214640012024,
      "flos": 18362777811840.0,
      "grad_norm": 1.7281617960017766,
      "language_loss": 0.71418881,
      "learning_rate": 7.050841375089506e-07,
      "loss": 0.73298812,
      "num_input_tokens_seen": 262870975,
      "step": 12182,
      "time_per_iteration": 2.4794068336486816
    },
    {
      "auxiliary_loss_clip": 0.01107688,
      "auxiliary_loss_mlp": 0.0103414,
      "balance_loss_clip": 1.03815758,
      "balance_loss_mlp": 1.02242732,
      "epoch": 0.7324815872538705,
      "flos": 30812289189120.0,
      "grad_norm": 1.477633362737075,
      "language_loss": 0.7107693,
      "learning_rate": 7.047873528507015e-07,
      "loss": 0.73218757,
      "num_input_tokens_seen": 262892635,
      "step": 12183,
      "time_per_iteration": 2.540525436401367
    },
    {
      "auxiliary_loss_clip": 0.01093975,
      "auxiliary_loss_mlp": 0.01032074,
      "balance_loss_clip": 1.03820705,
      "balance_loss_mlp": 1.01931262,
      "epoch": 0.7325417105065384,
      "flos": 21505069451520.0,
      "grad_norm": 1.9627729767132935,
      "language_loss": 0.72779489,
      "learning_rate": 7.04490617307045e-07,
      "loss": 0.74905533,
      "num_input_tokens_seen": 262910725,
      "step": 12184,
      "time_per_iteration": 2.523193836212158
    },
    {
      "auxiliary_loss_clip": 0.0101514,
      "auxiliary_loss_mlp": 0.01003336,
      "balance_loss_clip": 1.01120138,
      "balance_loss_mlp": 1.00225759,
      "epoch": 0.7326018337592064,
      "flos": 67257742556160.0,
      "grad_norm": 1.339435748316049,
      "language_loss": 0.65098244,
      "learning_rate": 7.041939308892344e-07,
      "loss": 0.67116719,
      "num_input_tokens_seen": 262974150,
      "step": 12185,
      "time_per_iteration": 3.1242823600769043
    },
    {
      "auxiliary_loss_clip": 0.01104429,
      "auxiliary_loss_mlp": 0.01029341,
      "balance_loss_clip": 1.03400588,
      "balance_loss_mlp": 1.01682985,
      "epoch": 0.7326619570118743,
      "flos": 22857070394880.0,
      "grad_norm": 2.0846602095860245,
      "language_loss": 0.8041386,
      "learning_rate": 7.038972936085197e-07,
      "loss": 0.82547629,
      "num_input_tokens_seen": 262993370,
      "step": 12186,
      "time_per_iteration": 2.490663766860962
    },
    {
      "auxiliary_loss_clip": 0.01095516,
      "auxiliary_loss_mlp": 0.01030224,
      "balance_loss_clip": 1.03551579,
      "balance_loss_mlp": 1.01712859,
      "epoch": 0.7327220802645423,
      "flos": 23327499841920.0,
      "grad_norm": 1.6573777505686993,
      "language_loss": 0.73288512,
      "learning_rate": 7.036007054761508e-07,
      "loss": 0.75414258,
      "num_input_tokens_seen": 263012665,
      "step": 12187,
      "time_per_iteration": 2.5017013549804688
    },
    {
      "auxiliary_loss_clip": 0.01107925,
      "auxiliary_loss_mlp": 0.0103238,
      "balance_loss_clip": 1.03768647,
      "balance_loss_mlp": 1.02054274,
      "epoch": 0.7327822035172102,
      "flos": 23180661043200.0,
      "grad_norm": 1.7425932039670355,
      "language_loss": 0.8897692,
      "learning_rate": 7.033041665033716e-07,
      "loss": 0.91117227,
      "num_input_tokens_seen": 263031475,
      "step": 12188,
      "time_per_iteration": 2.503183603286743
    },
    {
      "auxiliary_loss_clip": 0.01063339,
      "auxiliary_loss_mlp": 0.01033187,
      "balance_loss_clip": 1.0333935,
      "balance_loss_mlp": 1.0200913,
      "epoch": 0.7328423267698783,
      "flos": 21066600130560.0,
      "grad_norm": 1.8639320661878513,
      "language_loss": 0.74465799,
      "learning_rate": 7.030076767014284e-07,
      "loss": 0.76562327,
      "num_input_tokens_seen": 263051445,
      "step": 12189,
      "time_per_iteration": 2.592329263687134
    },
    {
      "auxiliary_loss_clip": 0.01072035,
      "auxiliary_loss_mlp": 0.0102765,
      "balance_loss_clip": 1.03475428,
      "balance_loss_mlp": 1.0155859,
      "epoch": 0.7329024500225462,
      "flos": 21689578638720.0,
      "grad_norm": 1.5422968370155854,
      "language_loss": 0.82072794,
      "learning_rate": 7.027112360815648e-07,
      "loss": 0.84172475,
      "num_input_tokens_seen": 263070835,
      "step": 12190,
      "time_per_iteration": 2.5835111141204834
    },
    {
      "auxiliary_loss_clip": 0.0106656,
      "auxiliary_loss_mlp": 0.01033957,
      "balance_loss_clip": 1.03545904,
      "balance_loss_mlp": 1.02045035,
      "epoch": 0.7329625732752142,
      "flos": 24164038661760.0,
      "grad_norm": 1.9226998516865073,
      "language_loss": 0.71865541,
      "learning_rate": 7.024148446550204e-07,
      "loss": 0.73966062,
      "num_input_tokens_seen": 263090070,
      "step": 12191,
      "time_per_iteration": 2.5783531665802
    },
    {
      "auxiliary_loss_clip": 0.01106655,
      "auxiliary_loss_mlp": 0.01033272,
      "balance_loss_clip": 1.0367496,
      "balance_loss_mlp": 1.02069545,
      "epoch": 0.7330226965278822,
      "flos": 30077915627520.0,
      "grad_norm": 1.5300601332581425,
      "language_loss": 0.69187987,
      "learning_rate": 7.021185024330361e-07,
      "loss": 0.71327913,
      "num_input_tokens_seen": 263110030,
      "step": 12192,
      "time_per_iteration": 2.5320897102355957
    },
    {
      "auxiliary_loss_clip": 0.01091668,
      "auxiliary_loss_mlp": 0.01030638,
      "balance_loss_clip": 1.03517592,
      "balance_loss_mlp": 1.01862121,
      "epoch": 0.7330828197805501,
      "flos": 23368294713600.0,
      "grad_norm": 1.5066739581154027,
      "language_loss": 0.73452491,
      "learning_rate": 7.01822209426848e-07,
      "loss": 0.75574797,
      "num_input_tokens_seen": 263129735,
      "step": 12193,
      "time_per_iteration": 2.5188703536987305
    },
    {
      "auxiliary_loss_clip": 0.01096999,
      "auxiliary_loss_mlp": 0.01029815,
      "balance_loss_clip": 1.03587246,
      "balance_loss_mlp": 1.01737523,
      "epoch": 0.7331429430332181,
      "flos": 21032808410880.0,
      "grad_norm": 1.7770110976381506,
      "language_loss": 0.77013612,
      "learning_rate": 7.015259656476911e-07,
      "loss": 0.79140425,
      "num_input_tokens_seen": 263149100,
      "step": 12194,
      "time_per_iteration": 2.4990944862365723
    },
    {
      "auxiliary_loss_clip": 0.01094517,
      "auxiliary_loss_mlp": 0.01030624,
      "balance_loss_clip": 1.03738928,
      "balance_loss_mlp": 1.01813054,
      "epoch": 0.733203066285886,
      "flos": 14647891466880.0,
      "grad_norm": 1.8465820776319488,
      "language_loss": 0.7045927,
      "learning_rate": 7.012297711067998e-07,
      "loss": 0.72584414,
      "num_input_tokens_seen": 263166620,
      "step": 12195,
      "time_per_iteration": 2.447614908218384
    },
    {
      "auxiliary_loss_clip": 0.01105374,
      "auxiliary_loss_mlp": 0.01039184,
      "balance_loss_clip": 1.03610265,
      "balance_loss_mlp": 1.02716208,
      "epoch": 0.7332631895385541,
      "flos": 17165301177600.0,
      "grad_norm": 1.8229861344175058,
      "language_loss": 0.72038531,
      "learning_rate": 7.009336258154057e-07,
      "loss": 0.74183095,
      "num_input_tokens_seen": 263184780,
      "step": 12196,
      "time_per_iteration": 2.4437386989593506
    },
    {
      "auxiliary_loss_clip": 0.01103608,
      "auxiliary_loss_mlp": 0.01028137,
      "balance_loss_clip": 1.03617895,
      "balance_loss_mlp": 1.01547742,
      "epoch": 0.733323312791222,
      "flos": 28658151676800.0,
      "grad_norm": 1.700197375118563,
      "language_loss": 0.71750987,
      "learning_rate": 7.006375297847394e-07,
      "loss": 0.73882729,
      "num_input_tokens_seen": 263204625,
      "step": 12197,
      "time_per_iteration": 2.5084803104400635
    },
    {
      "auxiliary_loss_clip": 0.01055234,
      "auxiliary_loss_mlp": 0.00784031,
      "balance_loss_clip": 1.03443265,
      "balance_loss_mlp": 1.008021,
      "epoch": 0.73338343604389,
      "flos": 16618417632000.0,
      "grad_norm": 2.042651736808952,
      "language_loss": 0.78240132,
      "learning_rate": 7.003414830260282e-07,
      "loss": 0.80079395,
      "num_input_tokens_seen": 263221565,
      "step": 12198,
      "time_per_iteration": 2.613983631134033
    },
    {
      "auxiliary_loss_clip": 0.01050999,
      "auxiliary_loss_mlp": 0.01033955,
      "balance_loss_clip": 1.03531742,
      "balance_loss_mlp": 1.02206969,
      "epoch": 0.7334435592965579,
      "flos": 21142084561920.0,
      "grad_norm": 2.028796718795781,
      "language_loss": 0.74023187,
      "learning_rate": 7.000454855504974e-07,
      "loss": 0.7610814,
      "num_input_tokens_seen": 263240620,
      "step": 12199,
      "time_per_iteration": 2.6060001850128174
    },
    {
      "auxiliary_loss_clip": 0.01087986,
      "auxiliary_loss_mlp": 0.01034948,
      "balance_loss_clip": 1.03723025,
      "balance_loss_mlp": 1.0224247,
      "epoch": 0.7335036825492259,
      "flos": 17125332318720.0,
      "grad_norm": 2.7626850160261682,
      "language_loss": 0.77309489,
      "learning_rate": 6.997495373693729e-07,
      "loss": 0.79432422,
      "num_input_tokens_seen": 263254365,
      "step": 12200,
      "time_per_iteration": 2.5130815505981445
    },
    {
      "auxiliary_loss_clip": 0.01067706,
      "auxiliary_loss_mlp": 0.01030035,
      "balance_loss_clip": 1.03616965,
      "balance_loss_mlp": 1.0187881,
      "epoch": 0.7335638058018938,
      "flos": 23731818307200.0,
      "grad_norm": 2.467902441565637,
      "language_loss": 0.61292195,
      "learning_rate": 6.994536384938754e-07,
      "loss": 0.63389933,
      "num_input_tokens_seen": 263275880,
      "step": 12201,
      "time_per_iteration": 2.5849809646606445
    },
    {
      "auxiliary_loss_clip": 0.01067545,
      "auxiliary_loss_mlp": 0.00783156,
      "balance_loss_clip": 1.03429484,
      "balance_loss_mlp": 1.00961471,
      "epoch": 0.7336239290545619,
      "flos": 34933289679360.0,
      "grad_norm": 2.364406813646872,
      "language_loss": 0.51949883,
      "learning_rate": 6.991577889352264e-07,
      "loss": 0.53800583,
      "num_input_tokens_seen": 263298315,
      "step": 12202,
      "time_per_iteration": 4.084860563278198
    },
    {
      "auxiliary_loss_clip": 0.01081135,
      "auxiliary_loss_mlp": 0.01030099,
      "balance_loss_clip": 1.03392434,
      "balance_loss_mlp": 1.01831484,
      "epoch": 0.7336840523072298,
      "flos": 21103049456640.0,
      "grad_norm": 1.6665454277277358,
      "language_loss": 0.69044393,
      "learning_rate": 6.98861988704645e-07,
      "loss": 0.71155632,
      "num_input_tokens_seen": 263318615,
      "step": 12203,
      "time_per_iteration": 2.5386900901794434
    },
    {
      "auxiliary_loss_clip": 0.01081233,
      "auxiliary_loss_mlp": 0.01034466,
      "balance_loss_clip": 1.03659964,
      "balance_loss_mlp": 1.02181184,
      "epoch": 0.7337441755598978,
      "flos": 24024418496640.0,
      "grad_norm": 3.7118217333262264,
      "language_loss": 0.66202009,
      "learning_rate": 6.985662378133474e-07,
      "loss": 0.68317711,
      "num_input_tokens_seen": 263336705,
      "step": 12204,
      "time_per_iteration": 2.5484304428100586
    },
    {
      "auxiliary_loss_clip": 0.01082386,
      "auxiliary_loss_mlp": 0.0103336,
      "balance_loss_clip": 1.03767836,
      "balance_loss_mlp": 1.02204669,
      "epoch": 0.7338042988125658,
      "flos": 22711309004160.0,
      "grad_norm": 1.8848198565473389,
      "language_loss": 0.77421916,
      "learning_rate": 6.982705362725479e-07,
      "loss": 0.7953766,
      "num_input_tokens_seen": 263355065,
      "step": 12205,
      "time_per_iteration": 2.5387749671936035
    },
    {
      "auxiliary_loss_clip": 0.01060258,
      "auxiliary_loss_mlp": 0.01026933,
      "balance_loss_clip": 1.03569722,
      "balance_loss_mlp": 1.01580477,
      "epoch": 0.7338644220652337,
      "flos": 21360996000000.0,
      "grad_norm": 1.627393583970014,
      "language_loss": 0.79693246,
      "learning_rate": 6.979748840934601e-07,
      "loss": 0.81780446,
      "num_input_tokens_seen": 263374460,
      "step": 12206,
      "time_per_iteration": 2.6074490547180176
    },
    {
      "auxiliary_loss_clip": 0.01065905,
      "auxiliary_loss_mlp": 0.0102845,
      "balance_loss_clip": 1.03248692,
      "balance_loss_mlp": 1.01597476,
      "epoch": 0.7339245453179017,
      "flos": 30920236536960.0,
      "grad_norm": 1.9012446703026653,
      "language_loss": 0.71481061,
      "learning_rate": 6.976792812872958e-07,
      "loss": 0.73575419,
      "num_input_tokens_seen": 263393610,
      "step": 12207,
      "time_per_iteration": 2.6919336318969727
    },
    {
      "auxiliary_loss_clip": 0.01015276,
      "auxiliary_loss_mlp": 0.01002401,
      "balance_loss_clip": 1.01121271,
      "balance_loss_mlp": 1.0011847,
      "epoch": 0.7339846685705697,
      "flos": 67899429072000.0,
      "grad_norm": 0.8817549127639255,
      "language_loss": 0.54792237,
      "learning_rate": 6.97383727865263e-07,
      "loss": 0.56809914,
      "num_input_tokens_seen": 263450340,
      "step": 12208,
      "time_per_iteration": 4.6046624183654785
    },
    {
      "auxiliary_loss_clip": 0.01105232,
      "auxiliary_loss_mlp": 0.0103176,
      "balance_loss_clip": 1.03628337,
      "balance_loss_mlp": 1.02085292,
      "epoch": 0.7340447918232377,
      "flos": 22236749493120.0,
      "grad_norm": 1.7937534731137583,
      "language_loss": 0.80357802,
      "learning_rate": 6.970882238385703e-07,
      "loss": 0.82494795,
      "num_input_tokens_seen": 263471735,
      "step": 12209,
      "time_per_iteration": 3.8980114459991455
    },
    {
      "auxiliary_loss_clip": 0.01101058,
      "auxiliary_loss_mlp": 0.01028161,
      "balance_loss_clip": 1.03381717,
      "balance_loss_mlp": 1.01648402,
      "epoch": 0.7341049150759056,
      "flos": 23764784014080.0,
      "grad_norm": 1.53708551415469,
      "language_loss": 0.78849733,
      "learning_rate": 6.96792769218423e-07,
      "loss": 0.80978948,
      "num_input_tokens_seen": 263493245,
      "step": 12210,
      "time_per_iteration": 2.51960825920105
    },
    {
      "auxiliary_loss_clip": 0.01103194,
      "auxiliary_loss_mlp": 0.01029659,
      "balance_loss_clip": 1.03597283,
      "balance_loss_mlp": 1.01757729,
      "epoch": 0.7341650383285736,
      "flos": 17236547804160.0,
      "grad_norm": 1.775541343299107,
      "language_loss": 0.76423192,
      "learning_rate": 6.964973640160236e-07,
      "loss": 0.78556049,
      "num_input_tokens_seen": 263511660,
      "step": 12211,
      "time_per_iteration": 2.4460318088531494
    },
    {
      "auxiliary_loss_clip": 0.01081861,
      "auxiliary_loss_mlp": 0.01024682,
      "balance_loss_clip": 1.03742695,
      "balance_loss_mlp": 1.01287997,
      "epoch": 0.7342251615812415,
      "flos": 23403953940480.0,
      "grad_norm": 1.8598422557350829,
      "language_loss": 0.72286868,
      "learning_rate": 6.962020082425748e-07,
      "loss": 0.74393415,
      "num_input_tokens_seen": 263530875,
      "step": 12212,
      "time_per_iteration": 2.5685925483703613
    },
    {
      "auxiliary_loss_clip": 0.01106175,
      "auxiliary_loss_mlp": 0.01034022,
      "balance_loss_clip": 1.0377115,
      "balance_loss_mlp": 1.02211261,
      "epoch": 0.7342852848339095,
      "flos": 22747183712640.0,
      "grad_norm": 1.712077440152238,
      "language_loss": 0.69073743,
      "learning_rate": 6.959067019092766e-07,
      "loss": 0.71213943,
      "num_input_tokens_seen": 263551585,
      "step": 12213,
      "time_per_iteration": 3.878291606903076
    },
    {
      "auxiliary_loss_clip": 0.01032499,
      "auxiliary_loss_mlp": 0.01002571,
      "balance_loss_clip": 1.00922632,
      "balance_loss_mlp": 1.00140846,
      "epoch": 0.7343454080865774,
      "flos": 53942353925760.0,
      "grad_norm": 0.7311735466843946,
      "language_loss": 0.54368991,
      "learning_rate": 6.956114450273276e-07,
      "loss": 0.5640406,
      "num_input_tokens_seen": 263609545,
      "step": 12214,
      "time_per_iteration": 2.9769935607910156
    },
    {
      "auxiliary_loss_clip": 0.01107015,
      "auxiliary_loss_mlp": 0.01025552,
      "balance_loss_clip": 1.03621316,
      "balance_loss_mlp": 1.01377451,
      "epoch": 0.7344055313392455,
      "flos": 12166859255040.0,
      "grad_norm": 2.0651299500865417,
      "language_loss": 0.70465046,
      "learning_rate": 6.953162376079233e-07,
      "loss": 0.72597611,
      "num_input_tokens_seen": 263627880,
      "step": 12215,
      "time_per_iteration": 2.4376790523529053
    },
    {
      "auxiliary_loss_clip": 0.01076394,
      "auxiliary_loss_mlp": 0.01028174,
      "balance_loss_clip": 1.03471756,
      "balance_loss_mlp": 1.0167774,
      "epoch": 0.7344656545919134,
      "flos": 18550052346240.0,
      "grad_norm": 1.5755018294224734,
      "language_loss": 0.72845769,
      "learning_rate": 6.950210796622573e-07,
      "loss": 0.74950337,
      "num_input_tokens_seen": 263645665,
      "step": 12216,
      "time_per_iteration": 2.488072156906128
    },
    {
      "auxiliary_loss_clip": 0.01113153,
      "auxiliary_loss_mlp": 0.01038336,
      "balance_loss_clip": 1.03816915,
      "balance_loss_mlp": 1.02357793,
      "epoch": 0.7345257778445814,
      "flos": 23661649088640.0,
      "grad_norm": 1.7350093420516886,
      "language_loss": 0.78005409,
      "learning_rate": 6.947259712015236e-07,
      "loss": 0.80156904,
      "num_input_tokens_seen": 263668170,
      "step": 12217,
      "time_per_iteration": 2.5204434394836426
    },
    {
      "auxiliary_loss_clip": 0.01065425,
      "auxiliary_loss_mlp": 0.01027452,
      "balance_loss_clip": 1.03477049,
      "balance_loss_mlp": 1.01670504,
      "epoch": 0.7345859010972494,
      "flos": 13808659127040.0,
      "grad_norm": 2.12153979932035,
      "language_loss": 0.78145021,
      "learning_rate": 6.94430912236911e-07,
      "loss": 0.80237895,
      "num_input_tokens_seen": 263684190,
      "step": 12218,
      "time_per_iteration": 2.5613982677459717
    },
    {
      "auxiliary_loss_clip": 0.01056501,
      "auxiliary_loss_mlp": 0.01037922,
      "balance_loss_clip": 1.03197646,
      "balance_loss_mlp": 1.02416527,
      "epoch": 0.7346460243499173,
      "flos": 22272731942400.0,
      "grad_norm": 2.0578586359050384,
      "language_loss": 0.72402221,
      "learning_rate": 6.941359027796092e-07,
      "loss": 0.74496645,
      "num_input_tokens_seen": 263702095,
      "step": 12219,
      "time_per_iteration": 2.5851259231567383
    },
    {
      "auxiliary_loss_clip": 0.01079254,
      "auxiliary_loss_mlp": 0.01030647,
      "balance_loss_clip": 1.03237462,
      "balance_loss_mlp": 1.0189054,
      "epoch": 0.7347061476025853,
      "flos": 23255247634560.0,
      "grad_norm": 1.7588658218794593,
      "language_loss": 0.74802715,
      "learning_rate": 6.938409428408061e-07,
      "loss": 0.76912618,
      "num_input_tokens_seen": 263721385,
      "step": 12220,
      "time_per_iteration": 2.5264031887054443
    },
    {
      "auxiliary_loss_clip": 0.01096564,
      "auxiliary_loss_mlp": 0.01030879,
      "balance_loss_clip": 1.0360992,
      "balance_loss_mlp": 1.01869631,
      "epoch": 0.7347662708552533,
      "flos": 15267565923840.0,
      "grad_norm": 1.6600898594294404,
      "language_loss": 0.66201377,
      "learning_rate": 6.93546032431684e-07,
      "loss": 0.68328822,
      "num_input_tokens_seen": 263737835,
      "step": 12221,
      "time_per_iteration": 2.4600467681884766
    },
    {
      "auxiliary_loss_clip": 0.01079308,
      "auxiliary_loss_mlp": 0.01034285,
      "balance_loss_clip": 1.03333628,
      "balance_loss_mlp": 1.02145219,
      "epoch": 0.7348263941079213,
      "flos": 24859987649280.0,
      "grad_norm": 2.115563016556717,
      "language_loss": 0.6899718,
      "learning_rate": 6.932511715634273e-07,
      "loss": 0.71110767,
      "num_input_tokens_seen": 263756480,
      "step": 12222,
      "time_per_iteration": 2.535013198852539
    },
    {
      "auxiliary_loss_clip": 0.01060002,
      "auxiliary_loss_mlp": 0.01031125,
      "balance_loss_clip": 1.03457642,
      "balance_loss_mlp": 1.02005041,
      "epoch": 0.7348865173605892,
      "flos": 24352103295360.0,
      "grad_norm": 1.6767165001044861,
      "language_loss": 0.65550077,
      "learning_rate": 6.92956360247217e-07,
      "loss": 0.67641211,
      "num_input_tokens_seen": 263776440,
      "step": 12223,
      "time_per_iteration": 2.625537157058716
    },
    {
      "auxiliary_loss_clip": 0.0108961,
      "auxiliary_loss_mlp": 0.01030071,
      "balance_loss_clip": 1.03562772,
      "balance_loss_mlp": 1.01768529,
      "epoch": 0.7349466406132572,
      "flos": 20004613597440.0,
      "grad_norm": 1.849969257871437,
      "language_loss": 0.72001213,
      "learning_rate": 6.926615984942332e-07,
      "loss": 0.74120891,
      "num_input_tokens_seen": 263793700,
      "step": 12224,
      "time_per_iteration": 2.5144588947296143
    },
    {
      "auxiliary_loss_clip": 0.01075618,
      "auxiliary_loss_mlp": 0.01028255,
      "balance_loss_clip": 1.036098,
      "balance_loss_mlp": 1.01657856,
      "epoch": 0.7350067638659251,
      "flos": 29825068815360.0,
      "grad_norm": 1.6646096660211382,
      "language_loss": 0.72225451,
      "learning_rate": 6.92366886315652e-07,
      "loss": 0.74329317,
      "num_input_tokens_seen": 263814620,
      "step": 12225,
      "time_per_iteration": 2.6347968578338623
    },
    {
      "auxiliary_loss_clip": 0.01108627,
      "auxiliary_loss_mlp": 0.01031511,
      "balance_loss_clip": 1.03639245,
      "balance_loss_mlp": 1.01850569,
      "epoch": 0.7350668871185931,
      "flos": 21866150920320.0,
      "grad_norm": 1.7135112745157375,
      "language_loss": 0.76180738,
      "learning_rate": 6.920722237226501e-07,
      "loss": 0.78320873,
      "num_input_tokens_seen": 263832725,
      "step": 12226,
      "time_per_iteration": 2.4667840003967285
    },
    {
      "auxiliary_loss_clip": 0.01078156,
      "auxiliary_loss_mlp": 0.01031413,
      "balance_loss_clip": 1.03343952,
      "balance_loss_mlp": 1.01797211,
      "epoch": 0.735127010371261,
      "flos": 22566122231040.0,
      "grad_norm": 3.452584062342471,
      "language_loss": 0.66756749,
      "learning_rate": 6.917776107264008e-07,
      "loss": 0.68866313,
      "num_input_tokens_seen": 263853850,
      "step": 12227,
      "time_per_iteration": 2.5390729904174805
    },
    {
      "auxiliary_loss_clip": 0.01096117,
      "auxiliary_loss_mlp": 0.01032699,
      "balance_loss_clip": 1.03560305,
      "balance_loss_mlp": 1.02129698,
      "epoch": 0.7351871336239291,
      "flos": 25884339707520.0,
      "grad_norm": 1.4870053125623355,
      "language_loss": 0.63405085,
      "learning_rate": 6.914830473380749e-07,
      "loss": 0.655339,
      "num_input_tokens_seen": 263874760,
      "step": 12228,
      "time_per_iteration": 2.54447340965271
    },
    {
      "auxiliary_loss_clip": 0.01084562,
      "auxiliary_loss_mlp": 0.01033984,
      "balance_loss_clip": 1.03640485,
      "balance_loss_mlp": 1.02297544,
      "epoch": 0.735247256876597,
      "flos": 17932173569280.0,
      "grad_norm": 1.6764424586960494,
      "language_loss": 0.63391674,
      "learning_rate": 6.911885335688427e-07,
      "loss": 0.65510225,
      "num_input_tokens_seen": 263893390,
      "step": 12229,
      "time_per_iteration": 2.5210559368133545
    },
    {
      "auxiliary_loss_clip": 0.01080849,
      "auxiliary_loss_mlp": 0.01035227,
      "balance_loss_clip": 1.03706431,
      "balance_loss_mlp": 1.02267456,
      "epoch": 0.735307380129265,
      "flos": 28875159694080.0,
      "grad_norm": 1.6462811326811897,
      "language_loss": 0.73378217,
      "learning_rate": 6.908940694298726e-07,
      "loss": 0.75494283,
      "num_input_tokens_seen": 263911180,
      "step": 12230,
      "time_per_iteration": 2.56764554977417
    },
    {
      "auxiliary_loss_clip": 0.01051782,
      "auxiliary_loss_mlp": 0.01028394,
      "balance_loss_clip": 1.03359604,
      "balance_loss_mlp": 1.01602566,
      "epoch": 0.7353675033819329,
      "flos": 13625658311040.0,
      "grad_norm": 2.282001459608554,
      "language_loss": 0.71901137,
      "learning_rate": 6.90599654932332e-07,
      "loss": 0.73981309,
      "num_input_tokens_seen": 263928975,
      "step": 12231,
      "time_per_iteration": 2.6062746047973633
    },
    {
      "auxiliary_loss_clip": 0.01097895,
      "auxiliary_loss_mlp": 0.01036533,
      "balance_loss_clip": 1.03812039,
      "balance_loss_mlp": 1.02279401,
      "epoch": 0.7354276266346009,
      "flos": 19463081178240.0,
      "grad_norm": 2.110570868040485,
      "language_loss": 0.63696784,
      "learning_rate": 6.903052900873823e-07,
      "loss": 0.65831208,
      "num_input_tokens_seen": 263944495,
      "step": 12232,
      "time_per_iteration": 2.486546754837036
    },
    {
      "auxiliary_loss_clip": 0.01085913,
      "auxiliary_loss_mlp": 0.01028669,
      "balance_loss_clip": 1.03530657,
      "balance_loss_mlp": 1.01647377,
      "epoch": 0.735487749887269,
      "flos": 15771858917760.0,
      "grad_norm": 1.962107713726942,
      "language_loss": 0.75356728,
      "learning_rate": 6.900109749061874e-07,
      "loss": 0.7747131,
      "num_input_tokens_seen": 263961325,
      "step": 12233,
      "time_per_iteration": 2.507244110107422
    },
    {
      "auxiliary_loss_clip": 0.01106017,
      "auxiliary_loss_mlp": 0.01026496,
      "balance_loss_clip": 1.03645921,
      "balance_loss_mlp": 1.01395512,
      "epoch": 0.7355478731399369,
      "flos": 18260648467200.0,
      "grad_norm": 1.571969347270664,
      "language_loss": 0.73178595,
      "learning_rate": 6.897167093999079e-07,
      "loss": 0.75311106,
      "num_input_tokens_seen": 263980445,
      "step": 12234,
      "time_per_iteration": 2.446500062942505
    },
    {
      "auxiliary_loss_clip": 0.01094377,
      "auxiliary_loss_mlp": 0.01029797,
      "balance_loss_clip": 1.03648901,
      "balance_loss_mlp": 1.01697004,
      "epoch": 0.7356079963926049,
      "flos": 26542043688960.0,
      "grad_norm": 2.6270870041554035,
      "language_loss": 0.60499299,
      "learning_rate": 6.894224935797017e-07,
      "loss": 0.62623477,
      "num_input_tokens_seen": 263999330,
      "step": 12235,
      "time_per_iteration": 2.547255516052246
    },
    {
      "auxiliary_loss_clip": 0.01082717,
      "auxiliary_loss_mlp": 0.01022857,
      "balance_loss_clip": 1.03621554,
      "balance_loss_mlp": 1.01113844,
      "epoch": 0.7356681196452728,
      "flos": 10778624467200.0,
      "grad_norm": 2.2443556436114975,
      "language_loss": 0.86113381,
      "learning_rate": 6.891283274567259e-07,
      "loss": 0.88218951,
      "num_input_tokens_seen": 264014150,
      "step": 12236,
      "time_per_iteration": 2.4753215312957764
    },
    {
      "auxiliary_loss_clip": 0.01096238,
      "auxiliary_loss_mlp": 0.00784168,
      "balance_loss_clip": 1.03739667,
      "balance_loss_mlp": 1.01160359,
      "epoch": 0.7357282428979408,
      "flos": 19718693337600.0,
      "grad_norm": 1.803286416183646,
      "language_loss": 0.69483215,
      "learning_rate": 6.888342110421364e-07,
      "loss": 0.71363616,
      "num_input_tokens_seen": 264033140,
      "step": 12237,
      "time_per_iteration": 2.505621910095215
    },
    {
      "auxiliary_loss_clip": 0.01022046,
      "auxiliary_loss_mlp": 0.0102911,
      "balance_loss_clip": 1.03254652,
      "balance_loss_mlp": 1.01680148,
      "epoch": 0.7357883661506087,
      "flos": 19464014931840.0,
      "grad_norm": 1.5371272079682532,
      "language_loss": 0.71948314,
      "learning_rate": 6.885401443470839e-07,
      "loss": 0.73999476,
      "num_input_tokens_seen": 264052105,
      "step": 12238,
      "time_per_iteration": 2.931239128112793
    },
    {
      "auxiliary_loss_clip": 0.01076327,
      "auxiliary_loss_mlp": 0.01028364,
      "balance_loss_clip": 1.03377461,
      "balance_loss_mlp": 1.01544809,
      "epoch": 0.7358484894032767,
      "flos": 27123006263040.0,
      "grad_norm": 1.8368415184220992,
      "language_loss": 0.72204721,
      "learning_rate": 6.882461273827205e-07,
      "loss": 0.74309409,
      "num_input_tokens_seen": 264070690,
      "step": 12239,
      "time_per_iteration": 2.8123621940612793
    },
    {
      "auxiliary_loss_clip": 0.01082607,
      "auxiliary_loss_mlp": 0.01029716,
      "balance_loss_clip": 1.03724265,
      "balance_loss_mlp": 1.01773548,
      "epoch": 0.7359086126559446,
      "flos": 24502282058880.0,
      "grad_norm": 1.4644767062299358,
      "language_loss": 0.78919506,
      "learning_rate": 6.879521601601954e-07,
      "loss": 0.81031829,
      "num_input_tokens_seen": 264094225,
      "step": 12240,
      "time_per_iteration": 2.5937659740448
    },
    {
      "auxiliary_loss_clip": 0.01095373,
      "auxiliary_loss_mlp": 0.01033866,
      "balance_loss_clip": 1.03681374,
      "balance_loss_mlp": 1.02128983,
      "epoch": 0.7359687359086127,
      "flos": 23331270769920.0,
      "grad_norm": 1.8557452940590293,
      "language_loss": 0.830567,
      "learning_rate": 6.876582426906565e-07,
      "loss": 0.85185939,
      "num_input_tokens_seen": 264113190,
      "step": 12241,
      "time_per_iteration": 4.27785062789917
    },
    {
      "auxiliary_loss_clip": 0.01092837,
      "auxiliary_loss_mlp": 0.01024156,
      "balance_loss_clip": 1.03455997,
      "balance_loss_mlp": 1.01252699,
      "epoch": 0.7360288591612806,
      "flos": 20193396503040.0,
      "grad_norm": 4.414466012716064,
      "language_loss": 0.78595948,
      "learning_rate": 6.873643749852484e-07,
      "loss": 0.80712938,
      "num_input_tokens_seen": 264132050,
      "step": 12242,
      "time_per_iteration": 2.534701108932495
    },
    {
      "auxiliary_loss_clip": 0.01057191,
      "auxiliary_loss_mlp": 0.01029853,
      "balance_loss_clip": 1.03385282,
      "balance_loss_mlp": 1.01757479,
      "epoch": 0.7360889824139486,
      "flos": 24972783333120.0,
      "grad_norm": 1.6923396884775845,
      "language_loss": 0.79179448,
      "learning_rate": 6.870705570551145e-07,
      "loss": 0.81266493,
      "num_input_tokens_seen": 264152800,
      "step": 12243,
      "time_per_iteration": 2.6337709426879883
    },
    {
      "auxiliary_loss_clip": 0.01095615,
      "auxiliary_loss_mlp": 0.01029306,
      "balance_loss_clip": 1.03552127,
      "balance_loss_mlp": 1.01640737,
      "epoch": 0.7361491056666165,
      "flos": 15012312900480.0,
      "grad_norm": 2.261697794486923,
      "language_loss": 0.74545825,
      "learning_rate": 6.867767889113969e-07,
      "loss": 0.76670742,
      "num_input_tokens_seen": 264169650,
      "step": 12244,
      "time_per_iteration": 2.4749157428741455
    },
    {
      "auxiliary_loss_clip": 0.01089297,
      "auxiliary_loss_mlp": 0.01030564,
      "balance_loss_clip": 1.03390479,
      "balance_loss_mlp": 1.01786196,
      "epoch": 0.7362092289192845,
      "flos": 22930400010240.0,
      "grad_norm": 1.7252549826813348,
      "language_loss": 0.69470078,
      "learning_rate": 6.864830705652347e-07,
      "loss": 0.71589941,
      "num_input_tokens_seen": 264190530,
      "step": 12245,
      "time_per_iteration": 2.5161588191986084
    },
    {
      "auxiliary_loss_clip": 0.01067915,
      "auxiliary_loss_mlp": 0.01032281,
      "balance_loss_clip": 1.03686202,
      "balance_loss_mlp": 1.02002048,
      "epoch": 0.7362693521719526,
      "flos": 20702681487360.0,
      "grad_norm": 1.4275159038208167,
      "language_loss": 0.73296982,
      "learning_rate": 6.861894020277658e-07,
      "loss": 0.75397182,
      "num_input_tokens_seen": 264210820,
      "step": 12246,
      "time_per_iteration": 4.397065162658691
    },
    {
      "auxiliary_loss_clip": 0.01080047,
      "auxiliary_loss_mlp": 0.01022964,
      "balance_loss_clip": 1.03430128,
      "balance_loss_mlp": 1.01182961,
      "epoch": 0.7363294754246205,
      "flos": 13111381336320.0,
      "grad_norm": 2.0499275787373454,
      "language_loss": 0.72952509,
      "learning_rate": 6.858957833101266e-07,
      "loss": 0.75055528,
      "num_input_tokens_seen": 264227430,
      "step": 12247,
      "time_per_iteration": 3.905771255493164
    },
    {
      "auxiliary_loss_clip": 0.01094424,
      "auxiliary_loss_mlp": 0.01028126,
      "balance_loss_clip": 1.03880322,
      "balance_loss_mlp": 1.01672959,
      "epoch": 0.7363895986772885,
      "flos": 14027426910720.0,
      "grad_norm": 2.848233783290102,
      "language_loss": 0.74280155,
      "learning_rate": 6.856022144234526e-07,
      "loss": 0.764027,
      "num_input_tokens_seen": 264245230,
      "step": 12248,
      "time_per_iteration": 2.473585605621338
    },
    {
      "auxiliary_loss_clip": 0.01084834,
      "auxiliary_loss_mlp": 0.01033503,
      "balance_loss_clip": 1.03551793,
      "balance_loss_mlp": 1.021052,
      "epoch": 0.7364497219299564,
      "flos": 19719986227200.0,
      "grad_norm": 2.0804214440191555,
      "language_loss": 0.72649658,
      "learning_rate": 6.853086953788727e-07,
      "loss": 0.74767995,
      "num_input_tokens_seen": 264263945,
      "step": 12249,
      "time_per_iteration": 2.5261425971984863
    },
    {
      "auxiliary_loss_clip": 0.01083441,
      "auxiliary_loss_mlp": 0.0103086,
      "balance_loss_clip": 1.0356549,
      "balance_loss_mlp": 1.01842618,
      "epoch": 0.7365098451826244,
      "flos": 21361391049600.0,
      "grad_norm": 1.8359377653879636,
      "language_loss": 0.76977873,
      "learning_rate": 6.850152261875189e-07,
      "loss": 0.79092169,
      "num_input_tokens_seen": 264281500,
      "step": 12250,
      "time_per_iteration": 2.5651333332061768
    },
    {
      "auxiliary_loss_clip": 0.01063446,
      "auxiliary_loss_mlp": 0.010277,
      "balance_loss_clip": 1.03628933,
      "balance_loss_mlp": 1.01519465,
      "epoch": 0.7365699684352923,
      "flos": 23368222886400.0,
      "grad_norm": 1.5159530573107787,
      "language_loss": 0.71277678,
      "learning_rate": 6.8472180686052e-07,
      "loss": 0.73368829,
      "num_input_tokens_seen": 264301625,
      "step": 12251,
      "time_per_iteration": 4.098764896392822
    },
    {
      "auxiliary_loss_clip": 0.01088823,
      "auxiliary_loss_mlp": 0.01031782,
      "balance_loss_clip": 1.0356282,
      "balance_loss_mlp": 1.0192951,
      "epoch": 0.7366300916879603,
      "flos": 59524879927680.0,
      "grad_norm": 1.54942445521682,
      "language_loss": 0.65834486,
      "learning_rate": 6.844284374090015e-07,
      "loss": 0.67955101,
      "num_input_tokens_seen": 264323975,
      "step": 12252,
      "time_per_iteration": 2.857226848602295
    },
    {
      "auxiliary_loss_clip": 0.0105745,
      "auxiliary_loss_mlp": 0.01033913,
      "balance_loss_clip": 1.03448367,
      "balance_loss_mlp": 1.02121139,
      "epoch": 0.7366902149406283,
      "flos": 20923137210240.0,
      "grad_norm": 1.9511667791874612,
      "language_loss": 0.79415089,
      "learning_rate": 6.841351178440884e-07,
      "loss": 0.81506449,
      "num_input_tokens_seen": 264343785,
      "step": 12253,
      "time_per_iteration": 2.606703281402588
    },
    {
      "auxiliary_loss_clip": 0.01101848,
      "auxiliary_loss_mlp": 0.00783242,
      "balance_loss_clip": 1.03584242,
      "balance_loss_mlp": 1.0096736,
      "epoch": 0.7367503381932963,
      "flos": 17348158339200.0,
      "grad_norm": 2.2140684375271307,
      "language_loss": 0.76428986,
      "learning_rate": 6.83841848176905e-07,
      "loss": 0.78314078,
      "num_input_tokens_seen": 264361130,
      "step": 12254,
      "time_per_iteration": 2.4646682739257812
    },
    {
      "auxiliary_loss_clip": 0.01079758,
      "auxiliary_loss_mlp": 0.01039489,
      "balance_loss_clip": 1.03461409,
      "balance_loss_mlp": 1.02540469,
      "epoch": 0.7368104614459642,
      "flos": 17821317219840.0,
      "grad_norm": 2.5344924309389207,
      "language_loss": 0.69225442,
      "learning_rate": 6.835486284185692e-07,
      "loss": 0.71344686,
      "num_input_tokens_seen": 264376965,
      "step": 12255,
      "time_per_iteration": 2.549440383911133
    },
    {
      "auxiliary_loss_clip": 0.01095776,
      "auxiliary_loss_mlp": 0.01033249,
      "balance_loss_clip": 1.03708506,
      "balance_loss_mlp": 1.02045751,
      "epoch": 0.7368705846986322,
      "flos": 24606099342720.0,
      "grad_norm": 1.6391626278783977,
      "language_loss": 0.75323492,
      "learning_rate": 6.832554585802012e-07,
      "loss": 0.77452517,
      "num_input_tokens_seen": 264396310,
      "step": 12256,
      "time_per_iteration": 2.5389461517333984
    },
    {
      "auxiliary_loss_clip": 0.01095182,
      "auxiliary_loss_mlp": 0.01026812,
      "balance_loss_clip": 1.03571129,
      "balance_loss_mlp": 1.01428962,
      "epoch": 0.7369307079513001,
      "flos": 34970169968640.0,
      "grad_norm": 1.5867398258313095,
      "language_loss": 0.73907888,
      "learning_rate": 6.829623386729182e-07,
      "loss": 0.76029879,
      "num_input_tokens_seen": 264418085,
      "step": 12257,
      "time_per_iteration": 2.6123385429382324
    },
    {
      "auxiliary_loss_clip": 0.01088575,
      "auxiliary_loss_mlp": 0.01035699,
      "balance_loss_clip": 1.03389502,
      "balance_loss_mlp": 1.02344441,
      "epoch": 0.7369908312039681,
      "flos": 21214588164480.0,
      "grad_norm": 1.5406634029883848,
      "language_loss": 0.78340244,
      "learning_rate": 6.826692687078362e-07,
      "loss": 0.80464518,
      "num_input_tokens_seen": 264437595,
      "step": 12258,
      "time_per_iteration": 2.5139074325561523
    },
    {
      "auxiliary_loss_clip": 0.01097596,
      "auxiliary_loss_mlp": 0.01032183,
      "balance_loss_clip": 1.03728247,
      "balance_loss_mlp": 1.02013099,
      "epoch": 0.7370509544566362,
      "flos": 23623655477760.0,
      "grad_norm": 1.3971096100920117,
      "language_loss": 0.66290808,
      "learning_rate": 6.823762486960674e-07,
      "loss": 0.68420589,
      "num_input_tokens_seen": 264457385,
      "step": 12259,
      "time_per_iteration": 2.4997944831848145
    },
    {
      "auxiliary_loss_clip": 0.01095797,
      "auxiliary_loss_mlp": 0.01034421,
      "balance_loss_clip": 1.0373981,
      "balance_loss_mlp": 1.02126622,
      "epoch": 0.7371110777093041,
      "flos": 24827704300800.0,
      "grad_norm": 1.649248196334524,
      "language_loss": 0.73059011,
      "learning_rate": 6.820832786487225e-07,
      "loss": 0.75189233,
      "num_input_tokens_seen": 264477205,
      "step": 12260,
      "time_per_iteration": 2.524329900741577
    },
    {
      "auxiliary_loss_clip": 0.01092731,
      "auxiliary_loss_mlp": 0.01031538,
      "balance_loss_clip": 1.0362463,
      "balance_loss_mlp": 1.01873493,
      "epoch": 0.7371712009619721,
      "flos": 23149491016320.0,
      "grad_norm": 1.6681404996888443,
      "language_loss": 0.73715812,
      "learning_rate": 6.817903585769125e-07,
      "loss": 0.7584008,
      "num_input_tokens_seen": 264497195,
      "step": 12261,
      "time_per_iteration": 2.501664638519287
    },
    {
      "auxiliary_loss_clip": 0.0108695,
      "auxiliary_loss_mlp": 0.01035216,
      "balance_loss_clip": 1.03537273,
      "balance_loss_mlp": 1.02181649,
      "epoch": 0.73723132421464,
      "flos": 23112898035840.0,
      "grad_norm": 1.9024526313002705,
      "language_loss": 0.66646618,
      "learning_rate": 6.814974884917438e-07,
      "loss": 0.68768781,
      "num_input_tokens_seen": 264516950,
      "step": 12262,
      "time_per_iteration": 2.56613826751709
    },
    {
      "auxiliary_loss_clip": 0.01109316,
      "auxiliary_loss_mlp": 0.01029665,
      "balance_loss_clip": 1.03809679,
      "balance_loss_mlp": 1.01686168,
      "epoch": 0.737291447467308,
      "flos": 19273328605440.0,
      "grad_norm": 1.6952044984541146,
      "language_loss": 0.88651508,
      "learning_rate": 6.81204668404322e-07,
      "loss": 0.90790492,
      "num_input_tokens_seen": 264532675,
      "step": 12263,
      "time_per_iteration": 2.4330577850341797
    },
    {
      "auxiliary_loss_clip": 0.01101626,
      "auxiliary_loss_mlp": 0.01029191,
      "balance_loss_clip": 1.03661752,
      "balance_loss_mlp": 1.01885581,
      "epoch": 0.7373515707199759,
      "flos": 25118257415040.0,
      "grad_norm": 1.4975919100971744,
      "language_loss": 0.67358983,
      "learning_rate": 6.809118983257522e-07,
      "loss": 0.69489795,
      "num_input_tokens_seen": 264555635,
      "step": 12264,
      "time_per_iteration": 2.5394296646118164
    },
    {
      "auxiliary_loss_clip": 0.01102409,
      "auxiliary_loss_mlp": 0.01029305,
      "balance_loss_clip": 1.03564763,
      "balance_loss_mlp": 1.01797998,
      "epoch": 0.737411693972644,
      "flos": 32408481767040.0,
      "grad_norm": 2.3049755868806123,
      "language_loss": 0.8009696,
      "learning_rate": 6.806191782671356e-07,
      "loss": 0.82228673,
      "num_input_tokens_seen": 264573140,
      "step": 12265,
      "time_per_iteration": 2.549098491668701
    },
    {
      "auxiliary_loss_clip": 0.01099647,
      "auxiliary_loss_mlp": 0.01031915,
      "balance_loss_clip": 1.03602409,
      "balance_loss_mlp": 1.01912344,
      "epoch": 0.7374718172253119,
      "flos": 24315797623680.0,
      "grad_norm": 1.755031647850366,
      "language_loss": 0.74719024,
      "learning_rate": 6.803265082395711e-07,
      "loss": 0.76850581,
      "num_input_tokens_seen": 264591610,
      "step": 12266,
      "time_per_iteration": 2.5248162746429443
    },
    {
      "auxiliary_loss_clip": 0.01096795,
      "auxiliary_loss_mlp": 0.01034999,
      "balance_loss_clip": 1.03730857,
      "balance_loss_mlp": 1.02242827,
      "epoch": 0.7375319404779799,
      "flos": 27156115624320.0,
      "grad_norm": 1.6436926747938332,
      "language_loss": 0.73596507,
      "learning_rate": 6.800338882541576e-07,
      "loss": 0.75728303,
      "num_input_tokens_seen": 264611170,
      "step": 12267,
      "time_per_iteration": 2.5502076148986816
    },
    {
      "auxiliary_loss_clip": 0.0107077,
      "auxiliary_loss_mlp": 0.01032983,
      "balance_loss_clip": 1.03500128,
      "balance_loss_mlp": 1.02118766,
      "epoch": 0.7375920637306478,
      "flos": 18879999701760.0,
      "grad_norm": 1.9677095060780894,
      "language_loss": 0.82768911,
      "learning_rate": 6.797413183219923e-07,
      "loss": 0.84872663,
      "num_input_tokens_seen": 264629365,
      "step": 12268,
      "time_per_iteration": 2.5520355701446533
    },
    {
      "auxiliary_loss_clip": 0.01104902,
      "auxiliary_loss_mlp": 0.01039711,
      "balance_loss_clip": 1.03674626,
      "balance_loss_mlp": 1.02727795,
      "epoch": 0.7376521869833158,
      "flos": 15669765486720.0,
      "grad_norm": 1.8374970787500802,
      "language_loss": 0.73165661,
      "learning_rate": 6.794487984541677e-07,
      "loss": 0.75310272,
      "num_input_tokens_seen": 264647915,
      "step": 12269,
      "time_per_iteration": 2.4777140617370605
    },
    {
      "auxiliary_loss_clip": 0.01082507,
      "auxiliary_loss_mlp": 0.01038705,
      "balance_loss_clip": 1.03474736,
      "balance_loss_mlp": 1.02518058,
      "epoch": 0.7377123102359837,
      "flos": 36971973901440.0,
      "grad_norm": 1.8391508250332866,
      "language_loss": 0.70199084,
      "learning_rate": 6.791563286617776e-07,
      "loss": 0.72320306,
      "num_input_tokens_seen": 264669620,
      "step": 12270,
      "time_per_iteration": 2.649378538131714
    },
    {
      "auxiliary_loss_clip": 0.01091606,
      "auxiliary_loss_mlp": 0.01028446,
      "balance_loss_clip": 1.03429186,
      "balance_loss_mlp": 1.01709723,
      "epoch": 0.7377724334886517,
      "flos": 24496284487680.0,
      "grad_norm": 1.889266105271503,
      "language_loss": 0.6950115,
      "learning_rate": 6.788639089559119e-07,
      "loss": 0.71621197,
      "num_input_tokens_seen": 264689345,
      "step": 12271,
      "time_per_iteration": 2.5344760417938232
    },
    {
      "auxiliary_loss_clip": 0.01082271,
      "auxiliary_loss_mlp": 0.01027022,
      "balance_loss_clip": 1.03763664,
      "balance_loss_mlp": 1.01420712,
      "epoch": 0.7378325567413198,
      "flos": 24390025079040.0,
      "grad_norm": 1.908794399890235,
      "language_loss": 0.67322421,
      "learning_rate": 6.785715393476586e-07,
      "loss": 0.69431716,
      "num_input_tokens_seen": 264707625,
      "step": 12272,
      "time_per_iteration": 2.5376393795013428
    },
    {
      "auxiliary_loss_clip": 0.01081993,
      "auxiliary_loss_mlp": 0.01029634,
      "balance_loss_clip": 1.03626585,
      "balance_loss_mlp": 1.01773095,
      "epoch": 0.7378926799939877,
      "flos": 17416388223360.0,
      "grad_norm": 1.8552444717244574,
      "language_loss": 0.78061867,
      "learning_rate": 6.782792198481049e-07,
      "loss": 0.80173498,
      "num_input_tokens_seen": 264725575,
      "step": 12273,
      "time_per_iteration": 2.5057506561279297
    },
    {
      "auxiliary_loss_clip": 0.0110413,
      "auxiliary_loss_mlp": 0.01031222,
      "balance_loss_clip": 1.03556728,
      "balance_loss_mlp": 1.01920021,
      "epoch": 0.7379528032466557,
      "flos": 18474208778880.0,
      "grad_norm": 1.8137083227894086,
      "language_loss": 0.83818686,
      "learning_rate": 6.779869504683355e-07,
      "loss": 0.85954034,
      "num_input_tokens_seen": 264742855,
      "step": 12274,
      "time_per_iteration": 2.4401180744171143
    },
    {
      "auxiliary_loss_clip": 0.01085144,
      "auxiliary_loss_mlp": 0.00784518,
      "balance_loss_clip": 1.03696907,
      "balance_loss_mlp": 1.00899816,
      "epoch": 0.7380129264993236,
      "flos": 17821999578240.0,
      "grad_norm": 2.118018273399142,
      "language_loss": 0.7386927,
      "learning_rate": 6.776947312194341e-07,
      "loss": 0.75738931,
      "num_input_tokens_seen": 264761155,
      "step": 12275,
      "time_per_iteration": 2.517084836959839
    },
    {
      "auxiliary_loss_clip": 0.01071827,
      "auxiliary_loss_mlp": 0.01048611,
      "balance_loss_clip": 1.03464818,
      "balance_loss_mlp": 1.03452075,
      "epoch": 0.7380730497519916,
      "flos": 22997372918400.0,
      "grad_norm": 3.7814404448269476,
      "language_loss": 0.73097539,
      "learning_rate": 6.774025621124813e-07,
      "loss": 0.7521798,
      "num_input_tokens_seen": 264780660,
      "step": 12276,
      "time_per_iteration": 2.5628018379211426
    },
    {
      "auxiliary_loss_clip": 0.01106768,
      "auxiliary_loss_mlp": 0.01030419,
      "balance_loss_clip": 1.03559589,
      "balance_loss_mlp": 1.01810443,
      "epoch": 0.7381331730046595,
      "flos": 20266259241600.0,
      "grad_norm": 1.8701447207174482,
      "language_loss": 0.77353621,
      "learning_rate": 6.771104431585551e-07,
      "loss": 0.79490805,
      "num_input_tokens_seen": 264798850,
      "step": 12277,
      "time_per_iteration": 2.464379072189331
    },
    {
      "auxiliary_loss_clip": 0.01104707,
      "auxiliary_loss_mlp": 0.01037948,
      "balance_loss_clip": 1.03740954,
      "balance_loss_mlp": 1.02572298,
      "epoch": 0.7381932962573275,
      "flos": 19754532132480.0,
      "grad_norm": 1.6640177124439044,
      "language_loss": 0.78713512,
      "learning_rate": 6.768183743687338e-07,
      "loss": 0.80856156,
      "num_input_tokens_seen": 264816795,
      "step": 12278,
      "time_per_iteration": 2.440934419631958
    },
    {
      "auxiliary_loss_clip": 0.01096238,
      "auxiliary_loss_mlp": 0.00784471,
      "balance_loss_clip": 1.03572643,
      "balance_loss_mlp": 1.00876486,
      "epoch": 0.7382534195099955,
      "flos": 17305316392320.0,
      "grad_norm": 2.1219333399385603,
      "language_loss": 0.72243154,
      "learning_rate": 6.765263557540921e-07,
      "loss": 0.74123865,
      "num_input_tokens_seen": 264834105,
      "step": 12279,
      "time_per_iteration": 2.4752302169799805
    },
    {
      "auxiliary_loss_clip": 0.01096934,
      "auxiliary_loss_mlp": 0.01034739,
      "balance_loss_clip": 1.03484464,
      "balance_loss_mlp": 1.0216198,
      "epoch": 0.7383135427626635,
      "flos": 18697358021760.0,
      "grad_norm": 2.117303878868669,
      "language_loss": 0.85729134,
      "learning_rate": 6.762343873257034e-07,
      "loss": 0.87860805,
      "num_input_tokens_seen": 264850895,
      "step": 12280,
      "time_per_iteration": 3.8616223335266113
    },
    {
      "auxiliary_loss_clip": 0.01068203,
      "auxiliary_loss_mlp": 0.01031194,
      "balance_loss_clip": 1.03401518,
      "balance_loss_mlp": 1.01855826,
      "epoch": 0.7383736660153314,
      "flos": 20881300844160.0,
      "grad_norm": 1.72940291397195,
      "language_loss": 0.72087502,
      "learning_rate": 6.759424690946408e-07,
      "loss": 0.74186897,
      "num_input_tokens_seen": 264869505,
      "step": 12281,
      "time_per_iteration": 2.5571165084838867
    },
    {
      "auxiliary_loss_clip": 0.01063207,
      "auxiliary_loss_mlp": 0.0102964,
      "balance_loss_clip": 1.03339911,
      "balance_loss_mlp": 1.01723623,
      "epoch": 0.7384337892679994,
      "flos": 20663215418880.0,
      "grad_norm": 1.5816832766456486,
      "language_loss": 0.6029408,
      "learning_rate": 6.756506010719711e-07,
      "loss": 0.62386918,
      "num_input_tokens_seen": 264886915,
      "step": 12282,
      "time_per_iteration": 2.5720293521881104
    },
    {
      "auxiliary_loss_clip": 0.01074013,
      "auxiliary_loss_mlp": 0.01031169,
      "balance_loss_clip": 1.03636336,
      "balance_loss_mlp": 1.01888466,
      "epoch": 0.7384939125206673,
      "flos": 29169627390720.0,
      "grad_norm": 1.8467512464283036,
      "language_loss": 0.67846388,
      "learning_rate": 6.753587832687632e-07,
      "loss": 0.6995157,
      "num_input_tokens_seen": 264910350,
      "step": 12283,
      "time_per_iteration": 2.624000072479248
    },
    {
      "auxiliary_loss_clip": 0.01106367,
      "auxiliary_loss_mlp": 0.00782278,
      "balance_loss_clip": 1.03740656,
      "balance_loss_mlp": 1.0073235,
      "epoch": 0.7385540357733353,
      "flos": 36312833376000.0,
      "grad_norm": 1.6976568284020885,
      "language_loss": 0.75800401,
      "learning_rate": 6.750670156960832e-07,
      "loss": 0.7768904,
      "num_input_tokens_seen": 264930705,
      "step": 12284,
      "time_per_iteration": 2.587664842605591
    },
    {
      "auxiliary_loss_clip": 0.01089458,
      "auxiliary_loss_mlp": 0.01033368,
      "balance_loss_clip": 1.03313136,
      "balance_loss_mlp": 1.02004623,
      "epoch": 0.7386141590260034,
      "flos": 20302600826880.0,
      "grad_norm": 1.834889170138869,
      "language_loss": 0.69750881,
      "learning_rate": 6.747752983649954e-07,
      "loss": 0.71873713,
      "num_input_tokens_seen": 264946975,
      "step": 12285,
      "time_per_iteration": 3.947385787963867
    },
    {
      "auxiliary_loss_clip": 0.01089034,
      "auxiliary_loss_mlp": 0.01033631,
      "balance_loss_clip": 1.03535342,
      "balance_loss_mlp": 1.02032685,
      "epoch": 0.7386742822786713,
      "flos": 25483792170240.0,
      "grad_norm": 1.801741176361625,
      "language_loss": 0.79848969,
      "learning_rate": 6.744836312865602e-07,
      "loss": 0.81971633,
      "num_input_tokens_seen": 264967665,
      "step": 12286,
      "time_per_iteration": 4.046239852905273
    },
    {
      "auxiliary_loss_clip": 0.01057757,
      "auxiliary_loss_mlp": 0.01030529,
      "balance_loss_clip": 1.03469312,
      "balance_loss_mlp": 1.01778603,
      "epoch": 0.7387344055313393,
      "flos": 13771958405760.0,
      "grad_norm": 4.586252862763562,
      "language_loss": 0.65795374,
      "learning_rate": 6.741920144718396e-07,
      "loss": 0.67883658,
      "num_input_tokens_seen": 264985480,
      "step": 12287,
      "time_per_iteration": 2.6108880043029785
    },
    {
      "auxiliary_loss_clip": 0.01079346,
      "auxiliary_loss_mlp": 0.01027462,
      "balance_loss_clip": 1.03249693,
      "balance_loss_mlp": 1.01565433,
      "epoch": 0.7387945287840072,
      "flos": 27855189095040.0,
      "grad_norm": 1.9892578809231651,
      "language_loss": 0.76945537,
      "learning_rate": 6.739004479318903e-07,
      "loss": 0.79052347,
      "num_input_tokens_seen": 265004790,
      "step": 12288,
      "time_per_iteration": 2.572993040084839
    },
    {
      "auxiliary_loss_clip": 0.01097648,
      "auxiliary_loss_mlp": 0.00786052,
      "balance_loss_clip": 1.03791738,
      "balance_loss_mlp": 1.00946975,
      "epoch": 0.7388546520366752,
      "flos": 44233039388160.0,
      "grad_norm": 1.5768673451927742,
      "language_loss": 0.58356863,
      "learning_rate": 6.736089316777684e-07,
      "loss": 0.60240567,
      "num_input_tokens_seen": 265028790,
      "step": 12289,
      "time_per_iteration": 2.7142226696014404
    },
    {
      "auxiliary_loss_clip": 0.01033508,
      "auxiliary_loss_mlp": 0.00761544,
      "balance_loss_clip": 1.01017082,
      "balance_loss_mlp": 0.9996047,
      "epoch": 0.7389147752893431,
      "flos": 70680890638080.0,
      "grad_norm": 0.6381458542438141,
      "language_loss": 0.49293876,
      "learning_rate": 6.733174657205287e-07,
      "loss": 0.51088929,
      "num_input_tokens_seen": 265096660,
      "step": 12290,
      "time_per_iteration": 4.5651984214782715
    },
    {
      "auxiliary_loss_clip": 0.01099319,
      "auxiliary_loss_mlp": 0.01028757,
      "balance_loss_clip": 1.03799224,
      "balance_loss_mlp": 1.01567376,
      "epoch": 0.7389748985420111,
      "flos": 25994980575360.0,
      "grad_norm": 1.8703093746417219,
      "language_loss": 0.67650044,
      "learning_rate": 6.730260500712237e-07,
      "loss": 0.69778121,
      "num_input_tokens_seen": 265116375,
      "step": 12291,
      "time_per_iteration": 2.532646894454956
    },
    {
      "auxiliary_loss_clip": 0.00994627,
      "auxiliary_loss_mlp": 0.01003747,
      "balance_loss_clip": 1.01327705,
      "balance_loss_mlp": 1.00258446,
      "epoch": 0.7390350217946791,
      "flos": 54403661318400.0,
      "grad_norm": 0.9939788614391721,
      "language_loss": 0.60947764,
      "learning_rate": 6.727346847409052e-07,
      "loss": 0.62946141,
      "num_input_tokens_seen": 265161230,
      "step": 12292,
      "time_per_iteration": 2.8518669605255127
    },
    {
      "auxiliary_loss_clip": 0.01058922,
      "auxiliary_loss_mlp": 0.01035387,
      "balance_loss_clip": 1.03556681,
      "balance_loss_mlp": 1.0235858,
      "epoch": 0.7390951450473471,
      "flos": 32196968530560.0,
      "grad_norm": 1.7828189491854698,
      "language_loss": 0.67029953,
      "learning_rate": 6.724433697406191e-07,
      "loss": 0.69124258,
      "num_input_tokens_seen": 265182515,
      "step": 12293,
      "time_per_iteration": 2.6730523109436035
    },
    {
      "auxiliary_loss_clip": 0.01094813,
      "auxiliary_loss_mlp": 0.01032163,
      "balance_loss_clip": 1.03514123,
      "balance_loss_mlp": 1.01978362,
      "epoch": 0.739155268300015,
      "flos": 16684241304960.0,
      "grad_norm": 1.8296831979601011,
      "language_loss": 0.83616722,
      "learning_rate": 6.721521050814134e-07,
      "loss": 0.85743701,
      "num_input_tokens_seen": 265198160,
      "step": 12294,
      "time_per_iteration": 2.498258352279663
    },
    {
      "auxiliary_loss_clip": 0.01069534,
      "auxiliary_loss_mlp": 0.0103202,
      "balance_loss_clip": 1.03427815,
      "balance_loss_mlp": 1.01982522,
      "epoch": 0.739215391552683,
      "flos": 31649761762560.0,
      "grad_norm": 1.7127353828117067,
      "language_loss": 0.73141944,
      "learning_rate": 6.718608907743337e-07,
      "loss": 0.75243503,
      "num_input_tokens_seen": 265218480,
      "step": 12295,
      "time_per_iteration": 2.6299564838409424
    },
    {
      "auxiliary_loss_clip": 0.01093521,
      "auxiliary_loss_mlp": 0.01035698,
      "balance_loss_clip": 1.03714108,
      "balance_loss_mlp": 1.02391422,
      "epoch": 0.7392755148053509,
      "flos": 29718522097920.0,
      "grad_norm": 1.8087195099266489,
      "language_loss": 0.78948843,
      "learning_rate": 6.715697268304215e-07,
      "loss": 0.81078064,
      "num_input_tokens_seen": 265240165,
      "step": 12296,
      "time_per_iteration": 2.5760204792022705
    },
    {
      "auxiliary_loss_clip": 0.01105426,
      "auxiliary_loss_mlp": 0.01033259,
      "balance_loss_clip": 1.03646839,
      "balance_loss_mlp": 1.02032471,
      "epoch": 0.7393356380580189,
      "flos": 37050475075200.0,
      "grad_norm": 1.9427972700579963,
      "language_loss": 0.66239613,
      "learning_rate": 6.712786132607182e-07,
      "loss": 0.68378294,
      "num_input_tokens_seen": 265263295,
      "step": 12297,
      "time_per_iteration": 2.597039222717285
    },
    {
      "auxiliary_loss_clip": 0.01080839,
      "auxiliary_loss_mlp": 0.01033425,
      "balance_loss_clip": 1.03562379,
      "balance_loss_mlp": 1.02058578,
      "epoch": 0.739395761310687,
      "flos": 19719627091200.0,
      "grad_norm": 1.6854586310371844,
      "language_loss": 0.68881655,
      "learning_rate": 6.709875500762645e-07,
      "loss": 0.70995915,
      "num_input_tokens_seen": 265282740,
      "step": 12298,
      "time_per_iteration": 2.512849807739258
    },
    {
      "auxiliary_loss_clip": 0.01085526,
      "auxiliary_loss_mlp": 0.01031034,
      "balance_loss_clip": 1.03578091,
      "balance_loss_mlp": 1.01883316,
      "epoch": 0.7394558845633549,
      "flos": 11801504067840.0,
      "grad_norm": 1.861003250412372,
      "language_loss": 0.74677885,
      "learning_rate": 6.706965372880946e-07,
      "loss": 0.76794446,
      "num_input_tokens_seen": 265300175,
      "step": 12299,
      "time_per_iteration": 2.497136116027832
    },
    {
      "auxiliary_loss_clip": 0.01018794,
      "auxiliary_loss_mlp": 0.01003528,
      "balance_loss_clip": 1.02172089,
      "balance_loss_mlp": 1.00219274,
      "epoch": 0.7395160078160229,
      "flos": 66195827850240.0,
      "grad_norm": 0.7275575326507032,
      "language_loss": 0.6084851,
      "learning_rate": 6.704055749072455e-07,
      "loss": 0.62870824,
      "num_input_tokens_seen": 265363275,
      "step": 12300,
      "time_per_iteration": 3.2091503143310547
    },
    {
      "auxiliary_loss_clip": 0.01085488,
      "auxiliary_loss_mlp": 0.01032779,
      "balance_loss_clip": 1.03776503,
      "balance_loss_mlp": 1.0198034,
      "epoch": 0.7395761310686908,
      "flos": 21249708687360.0,
      "grad_norm": 1.7993055451662745,
      "language_loss": 0.80176955,
      "learning_rate": 6.7011466294475e-07,
      "loss": 0.82295227,
      "num_input_tokens_seen": 265382935,
      "step": 12301,
      "time_per_iteration": 2.5261266231536865
    },
    {
      "auxiliary_loss_clip": 0.01103943,
      "auxiliary_loss_mlp": 0.01029259,
      "balance_loss_clip": 1.03638506,
      "balance_loss_mlp": 1.01818466,
      "epoch": 0.7396362543213588,
      "flos": 25955299025280.0,
      "grad_norm": 1.4947391976411573,
      "language_loss": 0.73079121,
      "learning_rate": 6.698238014116406e-07,
      "loss": 0.75212324,
      "num_input_tokens_seen": 265403245,
      "step": 12302,
      "time_per_iteration": 2.502429962158203
    },
    {
      "auxiliary_loss_clip": 0.01105628,
      "auxiliary_loss_mlp": 0.01037552,
      "balance_loss_clip": 1.03580451,
      "balance_loss_mlp": 1.02546453,
      "epoch": 0.7396963775740267,
      "flos": 27377936064000.0,
      "grad_norm": 1.8794527538342534,
      "language_loss": 0.73970133,
      "learning_rate": 6.695329903189451e-07,
      "loss": 0.76113307,
      "num_input_tokens_seen": 265423105,
      "step": 12303,
      "time_per_iteration": 2.502476453781128
    },
    {
      "auxiliary_loss_clip": 0.01102643,
      "auxiliary_loss_mlp": 0.01029876,
      "balance_loss_clip": 1.03536367,
      "balance_loss_mlp": 1.01846731,
      "epoch": 0.7397565008266948,
      "flos": 25520133755520.0,
      "grad_norm": 1.7548378874660553,
      "language_loss": 0.54318881,
      "learning_rate": 6.692422296776927e-07,
      "loss": 0.56451398,
      "num_input_tokens_seen": 265443445,
      "step": 12304,
      "time_per_iteration": 2.5142760276794434
    },
    {
      "auxiliary_loss_clip": 0.01083184,
      "auxiliary_loss_mlp": 0.01031921,
      "balance_loss_clip": 1.03501725,
      "balance_loss_mlp": 1.01979184,
      "epoch": 0.7398166240793627,
      "flos": 23727760070400.0,
      "grad_norm": 2.1053804559735085,
      "language_loss": 0.84618068,
      "learning_rate": 6.689515194989084e-07,
      "loss": 0.86733174,
      "num_input_tokens_seen": 265462085,
      "step": 12305,
      "time_per_iteration": 2.5424485206604004
    },
    {
      "auxiliary_loss_clip": 0.01010901,
      "auxiliary_loss_mlp": 0.01001353,
      "balance_loss_clip": 1.01084185,
      "balance_loss_mlp": 1.00019693,
      "epoch": 0.7398767473320307,
      "flos": 67267582882560.0,
      "grad_norm": 0.8663867832607923,
      "language_loss": 0.57716739,
      "learning_rate": 6.68660859793615e-07,
      "loss": 0.59728992,
      "num_input_tokens_seen": 265521190,
      "step": 12306,
      "time_per_iteration": 3.1225342750549316
    },
    {
      "auxiliary_loss_clip": 0.01086679,
      "auxiliary_loss_mlp": 0.01029753,
      "balance_loss_clip": 1.03750813,
      "balance_loss_mlp": 1.01712871,
      "epoch": 0.7399368705846986,
      "flos": 22018699981440.0,
      "grad_norm": 1.835490745343567,
      "language_loss": 0.81518805,
      "learning_rate": 6.683702505728355e-07,
      "loss": 0.83635235,
      "num_input_tokens_seen": 265539705,
      "step": 12307,
      "time_per_iteration": 2.5099899768829346
    },
    {
      "auxiliary_loss_clip": 0.01087925,
      "auxiliary_loss_mlp": 0.01028185,
      "balance_loss_clip": 1.03679669,
      "balance_loss_mlp": 1.01672864,
      "epoch": 0.7399969938373666,
      "flos": 14173870659840.0,
      "grad_norm": 1.673617090278493,
      "language_loss": 0.69768775,
      "learning_rate": 6.680796918475893e-07,
      "loss": 0.71884882,
      "num_input_tokens_seen": 265555855,
      "step": 12308,
      "time_per_iteration": 2.47898530960083
    },
    {
      "auxiliary_loss_clip": 0.01080927,
      "auxiliary_loss_mlp": 0.01027976,
      "balance_loss_clip": 1.03531528,
      "balance_loss_mlp": 1.0166986,
      "epoch": 0.7400571170900345,
      "flos": 25301473712640.0,
      "grad_norm": 1.7721586879990878,
      "language_loss": 0.81161582,
      "learning_rate": 6.67789183628896e-07,
      "loss": 0.8327049,
      "num_input_tokens_seen": 265575455,
      "step": 12309,
      "time_per_iteration": 2.567465305328369
    },
    {
      "auxiliary_loss_clip": 0.01093161,
      "auxiliary_loss_mlp": 0.01029939,
      "balance_loss_clip": 1.03703606,
      "balance_loss_mlp": 1.01773763,
      "epoch": 0.7401172403427025,
      "flos": 22711344917760.0,
      "grad_norm": 1.753373351604273,
      "language_loss": 0.72658122,
      "learning_rate": 6.674987259277692e-07,
      "loss": 0.74781227,
      "num_input_tokens_seen": 265595250,
      "step": 12310,
      "time_per_iteration": 2.517313241958618
    },
    {
      "auxiliary_loss_clip": 0.01072296,
      "auxiliary_loss_mlp": 0.01037185,
      "balance_loss_clip": 1.03922391,
      "balance_loss_mlp": 1.02424526,
      "epoch": 0.7401773635953706,
      "flos": 18067448188800.0,
      "grad_norm": 2.4601276452080842,
      "language_loss": 0.88534355,
      "learning_rate": 6.672083187552239e-07,
      "loss": 0.90643835,
      "num_input_tokens_seen": 265606945,
      "step": 12311,
      "time_per_iteration": 2.497124195098877
    },
    {
      "auxiliary_loss_clip": 0.01042454,
      "auxiliary_loss_mlp": 0.01028872,
      "balance_loss_clip": 1.03218579,
      "balance_loss_mlp": 1.01718962,
      "epoch": 0.7402374868480385,
      "flos": 22712135016960.0,
      "grad_norm": 1.4616140586283417,
      "language_loss": 0.80172503,
      "learning_rate": 6.669179621222738e-07,
      "loss": 0.82243836,
      "num_input_tokens_seen": 265626115,
      "step": 12312,
      "time_per_iteration": 2.6167969703674316
    },
    {
      "auxiliary_loss_clip": 0.01058117,
      "auxiliary_loss_mlp": 0.01034236,
      "balance_loss_clip": 1.03404331,
      "balance_loss_mlp": 1.02185607,
      "epoch": 0.7402976101007065,
      "flos": 22856675345280.0,
      "grad_norm": 1.7097445182311732,
      "language_loss": 0.78284639,
      "learning_rate": 6.666276560399273e-07,
      "loss": 0.80376995,
      "num_input_tokens_seen": 265646520,
      "step": 12313,
      "time_per_iteration": 2.6160778999328613
    },
    {
      "auxiliary_loss_clip": 0.01064311,
      "auxiliary_loss_mlp": 0.01032763,
      "balance_loss_clip": 1.03557706,
      "balance_loss_mlp": 1.0201093,
      "epoch": 0.7403577333533744,
      "flos": 12345801834240.0,
      "grad_norm": 2.4162185873192685,
      "language_loss": 0.78359282,
      "learning_rate": 6.663374005191937e-07,
      "loss": 0.80456352,
      "num_input_tokens_seen": 265661875,
      "step": 12314,
      "time_per_iteration": 2.560800790786743
    },
    {
      "auxiliary_loss_clip": 0.01022736,
      "auxiliary_loss_mlp": 0.0100622,
      "balance_loss_clip": 1.01002645,
      "balance_loss_mlp": 1.00514078,
      "epoch": 0.7404178566060424,
      "flos": 60327270869760.0,
      "grad_norm": 0.8441679286417137,
      "language_loss": 0.55193168,
      "learning_rate": 6.660471955710809e-07,
      "loss": 0.57222128,
      "num_input_tokens_seen": 265721255,
      "step": 12315,
      "time_per_iteration": 3.0791289806365967
    },
    {
      "auxiliary_loss_clip": 0.0108899,
      "auxiliary_loss_mlp": 0.01029597,
      "balance_loss_clip": 1.03799176,
      "balance_loss_mlp": 1.01783085,
      "epoch": 0.7404779798587103,
      "flos": 32014650072960.0,
      "grad_norm": 1.652530890145404,
      "language_loss": 0.80134356,
      "learning_rate": 6.65757041206591e-07,
      "loss": 0.82252944,
      "num_input_tokens_seen": 265743970,
      "step": 12316,
      "time_per_iteration": 2.594599723815918
    },
    {
      "auxiliary_loss_clip": 0.01092083,
      "auxiliary_loss_mlp": 0.01030317,
      "balance_loss_clip": 1.03412104,
      "balance_loss_mlp": 1.01852727,
      "epoch": 0.7405381031113784,
      "flos": 12889704551040.0,
      "grad_norm": 1.8418763347490485,
      "language_loss": 0.74720907,
      "learning_rate": 6.654669374367275e-07,
      "loss": 0.76843309,
      "num_input_tokens_seen": 265760890,
      "step": 12317,
      "time_per_iteration": 2.488734483718872
    },
    {
      "auxiliary_loss_clip": 0.01080639,
      "auxiliary_loss_mlp": 0.01030233,
      "balance_loss_clip": 1.03589463,
      "balance_loss_mlp": 1.01861584,
      "epoch": 0.7405982263640463,
      "flos": 20229127557120.0,
      "grad_norm": 1.5810904611640646,
      "language_loss": 0.81393635,
      "learning_rate": 6.651768842724917e-07,
      "loss": 0.8350451,
      "num_input_tokens_seen": 265779600,
      "step": 12318,
      "time_per_iteration": 3.8961055278778076
    },
    {
      "auxiliary_loss_clip": 0.01081175,
      "auxiliary_loss_mlp": 0.01029572,
      "balance_loss_clip": 1.03478515,
      "balance_loss_mlp": 1.0169301,
      "epoch": 0.7406583496167143,
      "flos": 17567213431680.0,
      "grad_norm": 2.3730958292156035,
      "language_loss": 0.76744026,
      "learning_rate": 6.648868817248827e-07,
      "loss": 0.78854775,
      "num_input_tokens_seen": 265797030,
      "step": 12319,
      "time_per_iteration": 2.5030930042266846
    },
    {
      "auxiliary_loss_clip": 0.01083132,
      "auxiliary_loss_mlp": 0.01033147,
      "balance_loss_clip": 1.03560364,
      "balance_loss_mlp": 1.02204227,
      "epoch": 0.7407184728693822,
      "flos": 18295733076480.0,
      "grad_norm": 1.8393706158431276,
      "language_loss": 0.63628805,
      "learning_rate": 6.64596929804897e-07,
      "loss": 0.6574508,
      "num_input_tokens_seen": 265815055,
      "step": 12320,
      "time_per_iteration": 2.483637809753418
    },
    {
      "auxiliary_loss_clip": 0.01097276,
      "auxiliary_loss_mlp": 0.01038758,
      "balance_loss_clip": 1.03716993,
      "balance_loss_mlp": 1.02654481,
      "epoch": 0.7407785961220502,
      "flos": 16690562098560.0,
      "grad_norm": 4.151621269767466,
      "language_loss": 0.82553804,
      "learning_rate": 6.643070285235288e-07,
      "loss": 0.84689832,
      "num_input_tokens_seen": 265828480,
      "step": 12321,
      "time_per_iteration": 2.4755165576934814
    },
    {
      "auxiliary_loss_clip": 0.0108683,
      "auxiliary_loss_mlp": 0.01048453,
      "balance_loss_clip": 1.03511024,
      "balance_loss_mlp": 1.03479135,
      "epoch": 0.7408387193747181,
      "flos": 22088330496000.0,
      "grad_norm": 2.4670302828978987,
      "language_loss": 0.71984625,
      "learning_rate": 6.640171778917727e-07,
      "loss": 0.74119914,
      "num_input_tokens_seen": 265845825,
      "step": 12322,
      "time_per_iteration": 2.5338587760925293
    },
    {
      "auxiliary_loss_clip": 0.01094989,
      "auxiliary_loss_mlp": 0.00784048,
      "balance_loss_clip": 1.03618085,
      "balance_loss_mlp": 1.00924671,
      "epoch": 0.7408988426273861,
      "flos": 24236721832320.0,
      "grad_norm": 1.7864011410045353,
      "language_loss": 0.64344293,
      "learning_rate": 6.637273779206183e-07,
      "loss": 0.66223329,
      "num_input_tokens_seen": 265866335,
      "step": 12323,
      "time_per_iteration": 2.5422515869140625
    },
    {
      "auxiliary_loss_clip": 0.01070597,
      "auxiliary_loss_mlp": 0.01027696,
      "balance_loss_clip": 1.03606427,
      "balance_loss_mlp": 1.01570964,
      "epoch": 0.7409589658800542,
      "flos": 29023004073600.0,
      "grad_norm": 1.3748630709286518,
      "language_loss": 0.75853634,
      "learning_rate": 6.634376286210559e-07,
      "loss": 0.77951926,
      "num_input_tokens_seen": 265888945,
      "step": 12324,
      "time_per_iteration": 5.414827108383179
    },
    {
      "auxiliary_loss_clip": 0.01074209,
      "auxiliary_loss_mlp": 0.01023961,
      "balance_loss_clip": 1.03523731,
      "balance_loss_mlp": 1.01224291,
      "epoch": 0.7410190891327221,
      "flos": 19351362902400.0,
      "grad_norm": 1.6829619577824186,
      "language_loss": 0.75183785,
      "learning_rate": 6.63147930004073e-07,
      "loss": 0.77281952,
      "num_input_tokens_seen": 265908030,
      "step": 12325,
      "time_per_iteration": 2.5139832496643066
    },
    {
      "auxiliary_loss_clip": 0.01065299,
      "auxiliary_loss_mlp": 0.01036906,
      "balance_loss_clip": 1.03511739,
      "balance_loss_mlp": 1.02403164,
      "epoch": 0.7410792123853901,
      "flos": 22747650589440.0,
      "grad_norm": 3.0018495681992885,
      "language_loss": 0.68063211,
      "learning_rate": 6.628582820806545e-07,
      "loss": 0.7016542,
      "num_input_tokens_seen": 265927030,
      "step": 12326,
      "time_per_iteration": 2.597381591796875
    },
    {
      "auxiliary_loss_clip": 0.0106756,
      "auxiliary_loss_mlp": 0.01027384,
      "balance_loss_clip": 1.03638649,
      "balance_loss_mlp": 1.01557088,
      "epoch": 0.741139335638058,
      "flos": 25372433030400.0,
      "grad_norm": 1.5910253416268154,
      "language_loss": 0.89418948,
      "learning_rate": 6.625686848617835e-07,
      "loss": 0.91513896,
      "num_input_tokens_seen": 265945490,
      "step": 12327,
      "time_per_iteration": 2.599994421005249
    },
    {
      "auxiliary_loss_clip": 0.01105752,
      "auxiliary_loss_mlp": 0.01033616,
      "balance_loss_clip": 1.03754878,
      "balance_loss_mlp": 1.02093208,
      "epoch": 0.741199458890726,
      "flos": 18585639745920.0,
      "grad_norm": 1.6548942386272454,
      "language_loss": 0.85585946,
      "learning_rate": 6.62279138358442e-07,
      "loss": 0.87725317,
      "num_input_tokens_seen": 265963265,
      "step": 12328,
      "time_per_iteration": 2.45192551612854
    },
    {
      "auxiliary_loss_clip": 0.0109154,
      "auxiliary_loss_mlp": 0.01029362,
      "balance_loss_clip": 1.03506887,
      "balance_loss_mlp": 1.01601684,
      "epoch": 0.7412595821433939,
      "flos": 22127078292480.0,
      "grad_norm": 1.7119229990889684,
      "language_loss": 0.66537726,
      "learning_rate": 6.619896425816103e-07,
      "loss": 0.68658626,
      "num_input_tokens_seen": 265982270,
      "step": 12329,
      "time_per_iteration": 3.982621669769287
    },
    {
      "auxiliary_loss_clip": 0.01073697,
      "auxiliary_loss_mlp": 0.01040504,
      "balance_loss_clip": 1.03544497,
      "balance_loss_mlp": 1.02766478,
      "epoch": 0.741319705396062,
      "flos": 29169699217920.0,
      "grad_norm": 1.6500569253780875,
      "language_loss": 0.66855556,
      "learning_rate": 6.617001975422647e-07,
      "loss": 0.6896975,
      "num_input_tokens_seen": 266003835,
      "step": 12330,
      "time_per_iteration": 2.648768186569214
    },
    {
      "auxiliary_loss_clip": 0.01072762,
      "auxiliary_loss_mlp": 0.01032062,
      "balance_loss_clip": 1.03922224,
      "balance_loss_mlp": 1.01750636,
      "epoch": 0.7413798286487299,
      "flos": 20667489137280.0,
      "grad_norm": 1.8182825956932047,
      "language_loss": 0.85612541,
      "learning_rate": 6.614108032513823e-07,
      "loss": 0.87717366,
      "num_input_tokens_seen": 266021595,
      "step": 12331,
      "time_per_iteration": 2.578341007232666
    },
    {
      "auxiliary_loss_clip": 0.01045755,
      "auxiliary_loss_mlp": 0.01028857,
      "balance_loss_clip": 1.03305197,
      "balance_loss_mlp": 1.01625109,
      "epoch": 0.7414399519013979,
      "flos": 16398895662720.0,
      "grad_norm": 2.2446176296108122,
      "language_loss": 0.69676352,
      "learning_rate": 6.611214597199364e-07,
      "loss": 0.71750963,
      "num_input_tokens_seen": 266039860,
      "step": 12332,
      "time_per_iteration": 2.6799416542053223
    },
    {
      "auxiliary_loss_clip": 0.01107126,
      "auxiliary_loss_mlp": 0.01040081,
      "balance_loss_clip": 1.03760672,
      "balance_loss_mlp": 1.02712309,
      "epoch": 0.7415000751540658,
      "flos": 25630235919360.0,
      "grad_norm": 10.903818210350364,
      "language_loss": 0.63917887,
      "learning_rate": 6.608321669588984e-07,
      "loss": 0.66065097,
      "num_input_tokens_seen": 266058050,
      "step": 12333,
      "time_per_iteration": 2.6797683238983154
    },
    {
      "auxiliary_loss_clip": 0.01081228,
      "auxiliary_loss_mlp": 0.01033907,
      "balance_loss_clip": 1.03698194,
      "balance_loss_mlp": 1.02255869,
      "epoch": 0.7415601984067338,
      "flos": 24499732193280.0,
      "grad_norm": 1.71502229609218,
      "language_loss": 0.71416521,
      "learning_rate": 6.605429249792387e-07,
      "loss": 0.73531657,
      "num_input_tokens_seen": 266078060,
      "step": 12334,
      "time_per_iteration": 2.564955234527588
    },
    {
      "auxiliary_loss_clip": 0.01067013,
      "auxiliary_loss_mlp": 0.01024246,
      "balance_loss_clip": 1.03531671,
      "balance_loss_mlp": 1.01239693,
      "epoch": 0.7416203216594017,
      "flos": 20887154760960.0,
      "grad_norm": 1.8416226582450987,
      "language_loss": 0.82206595,
      "learning_rate": 6.602537337919257e-07,
      "loss": 0.8429786,
      "num_input_tokens_seen": 266097110,
      "step": 12335,
      "time_per_iteration": 2.5671515464782715
    },
    {
      "auxiliary_loss_clip": 0.01105511,
      "auxiliary_loss_mlp": 0.01028842,
      "balance_loss_clip": 1.03610218,
      "balance_loss_mlp": 1.01612234,
      "epoch": 0.7416804449120697,
      "flos": 15624265933440.0,
      "grad_norm": 2.337866585890216,
      "language_loss": 0.74862885,
      "learning_rate": 6.599645934079259e-07,
      "loss": 0.76997232,
      "num_input_tokens_seen": 266110870,
      "step": 12336,
      "time_per_iteration": 2.404961109161377
    },
    {
      "auxiliary_loss_clip": 0.01061937,
      "auxiliary_loss_mlp": 0.01031038,
      "balance_loss_clip": 1.03492153,
      "balance_loss_mlp": 1.01883698,
      "epoch": 0.7417405681647377,
      "flos": 17120483982720.0,
      "grad_norm": 1.9498925137446024,
      "language_loss": 0.73258483,
      "learning_rate": 6.596755038382029e-07,
      "loss": 0.75351465,
      "num_input_tokens_seen": 266127845,
      "step": 12337,
      "time_per_iteration": 2.5827932357788086
    },
    {
      "auxiliary_loss_clip": 0.01078071,
      "auxiliary_loss_mlp": 0.01034637,
      "balance_loss_clip": 1.03589034,
      "balance_loss_mlp": 1.02296686,
      "epoch": 0.7418006914174057,
      "flos": 18880322924160.0,
      "grad_norm": 1.7010772018577223,
      "language_loss": 0.76565099,
      "learning_rate": 6.593864650937186e-07,
      "loss": 0.78677809,
      "num_input_tokens_seen": 266145400,
      "step": 12338,
      "time_per_iteration": 2.5041651725769043
    },
    {
      "auxiliary_loss_clip": 0.01093373,
      "auxiliary_loss_mlp": 0.01026362,
      "balance_loss_clip": 1.03492701,
      "balance_loss_mlp": 1.01569307,
      "epoch": 0.7418608146700737,
      "flos": 21580733450880.0,
      "grad_norm": 1.7085085487834184,
      "language_loss": 0.73356211,
      "learning_rate": 6.590974771854345e-07,
      "loss": 0.75475949,
      "num_input_tokens_seen": 266164430,
      "step": 12339,
      "time_per_iteration": 2.5216386318206787
    },
    {
      "auxiliary_loss_clip": 0.01085525,
      "auxiliary_loss_mlp": 0.0103083,
      "balance_loss_clip": 1.03519428,
      "balance_loss_mlp": 1.01814032,
      "epoch": 0.7419209379227416,
      "flos": 22340459036160.0,
      "grad_norm": 3.120066154988947,
      "language_loss": 0.79338944,
      "learning_rate": 6.588085401243077e-07,
      "loss": 0.81455296,
      "num_input_tokens_seen": 266183855,
      "step": 12340,
      "time_per_iteration": 2.547351598739624
    },
    {
      "auxiliary_loss_clip": 0.01072017,
      "auxiliary_loss_mlp": 0.01035189,
      "balance_loss_clip": 1.03487277,
      "balance_loss_mlp": 1.0226835,
      "epoch": 0.7419810611754096,
      "flos": 16762275601920.0,
      "grad_norm": 1.4415086769101915,
      "language_loss": 0.75432253,
      "learning_rate": 6.585196539212958e-07,
      "loss": 0.77539456,
      "num_input_tokens_seen": 266202085,
      "step": 12341,
      "time_per_iteration": 2.5167527198791504
    },
    {
      "auxiliary_loss_clip": 0.01073224,
      "auxiliary_loss_mlp": 0.01038377,
      "balance_loss_clip": 1.03403294,
      "balance_loss_mlp": 1.02513885,
      "epoch": 0.7420411844280775,
      "flos": 26212958259840.0,
      "grad_norm": 1.3923877826395394,
      "language_loss": 0.80291617,
      "learning_rate": 6.582308185873535e-07,
      "loss": 0.82403213,
      "num_input_tokens_seen": 266223445,
      "step": 12342,
      "time_per_iteration": 2.595031261444092
    },
    {
      "auxiliary_loss_clip": 0.01069508,
      "auxiliary_loss_mlp": 0.01030757,
      "balance_loss_clip": 1.03325105,
      "balance_loss_mlp": 1.01841235,
      "epoch": 0.7421013076807456,
      "flos": 68529371840640.0,
      "grad_norm": 1.724114013215581,
      "language_loss": 0.77445161,
      "learning_rate": 6.57942034133433e-07,
      "loss": 0.79545426,
      "num_input_tokens_seen": 266246575,
      "step": 12343,
      "time_per_iteration": 2.954693078994751
    },
    {
      "auxiliary_loss_clip": 0.01079004,
      "auxiliary_loss_mlp": 0.01032519,
      "balance_loss_clip": 1.03232431,
      "balance_loss_mlp": 1.02053857,
      "epoch": 0.7421614309334135,
      "flos": 24425325169920.0,
      "grad_norm": 8.138685087762191,
      "language_loss": 0.67979014,
      "learning_rate": 6.576533005704843e-07,
      "loss": 0.70090538,
      "num_input_tokens_seen": 266266055,
      "step": 12344,
      "time_per_iteration": 2.5676352977752686
    },
    {
      "auxiliary_loss_clip": 0.01060988,
      "auxiliary_loss_mlp": 0.01033631,
      "balance_loss_clip": 1.03635979,
      "balance_loss_mlp": 1.02028537,
      "epoch": 0.7422215541860815,
      "flos": 12311076360960.0,
      "grad_norm": 2.2511616893958717,
      "language_loss": 0.80972463,
      "learning_rate": 6.573646179094572e-07,
      "loss": 0.83067077,
      "num_input_tokens_seen": 266282240,
      "step": 12345,
      "time_per_iteration": 2.5678532123565674
    },
    {
      "auxiliary_loss_clip": 0.01074601,
      "auxiliary_loss_mlp": 0.01035692,
      "balance_loss_clip": 1.03472865,
      "balance_loss_mlp": 1.02302647,
      "epoch": 0.7422816774387494,
      "flos": 19645579203840.0,
      "grad_norm": 1.8900096443139274,
      "language_loss": 0.70636898,
      "learning_rate": 6.570759861612988e-07,
      "loss": 0.72747195,
      "num_input_tokens_seen": 266300980,
      "step": 12346,
      "time_per_iteration": 2.6218819618225098
    },
    {
      "auxiliary_loss_clip": 0.01095759,
      "auxiliary_loss_mlp": 0.01031843,
      "balance_loss_clip": 1.03721273,
      "balance_loss_mlp": 1.01966548,
      "epoch": 0.7423418006914174,
      "flos": 32015978876160.0,
      "grad_norm": 1.6397783502379655,
      "language_loss": 0.73199624,
      "learning_rate": 6.56787405336953e-07,
      "loss": 0.75327224,
      "num_input_tokens_seen": 266322215,
      "step": 12347,
      "time_per_iteration": 2.576030731201172
    },
    {
      "auxiliary_loss_clip": 0.01084082,
      "auxiliary_loss_mlp": 0.01033011,
      "balance_loss_clip": 1.03433132,
      "balance_loss_mlp": 1.02061951,
      "epoch": 0.7424019239440853,
      "flos": 18916951818240.0,
      "grad_norm": 1.7095756526717656,
      "language_loss": 0.80894411,
      "learning_rate": 6.564988754473642e-07,
      "loss": 0.83011502,
      "num_input_tokens_seen": 266341600,
      "step": 12348,
      "time_per_iteration": 2.5361597537994385
    },
    {
      "auxiliary_loss_clip": 0.01102773,
      "auxiliary_loss_mlp": 0.01031144,
      "balance_loss_clip": 1.03477228,
      "balance_loss_mlp": 1.01919937,
      "epoch": 0.7424620471967533,
      "flos": 35876518871040.0,
      "grad_norm": 1.7268640218387012,
      "language_loss": 0.7239306,
      "learning_rate": 6.562103965034724e-07,
      "loss": 0.74526978,
      "num_input_tokens_seen": 266362895,
      "step": 12349,
      "time_per_iteration": 2.5692129135131836
    },
    {
      "auxiliary_loss_clip": 0.01085507,
      "auxiliary_loss_mlp": 0.01032177,
      "balance_loss_clip": 1.03352368,
      "balance_loss_mlp": 1.01831913,
      "epoch": 0.7425221704494213,
      "flos": 27016603200000.0,
      "grad_norm": 2.013973407818162,
      "language_loss": 0.78762138,
      "learning_rate": 6.559219685162165e-07,
      "loss": 0.80879819,
      "num_input_tokens_seen": 266384015,
      "step": 12350,
      "time_per_iteration": 2.5644924640655518
    },
    {
      "auxiliary_loss_clip": 0.01060287,
      "auxiliary_loss_mlp": 0.01029829,
      "balance_loss_clip": 1.03541899,
      "balance_loss_mlp": 1.0181222,
      "epoch": 0.7425822937020893,
      "flos": 34167135559680.0,
      "grad_norm": 2.136406554035744,
      "language_loss": 0.75356221,
      "learning_rate": 6.556335914965343e-07,
      "loss": 0.77446342,
      "num_input_tokens_seen": 266405990,
      "step": 12351,
      "time_per_iteration": 2.6857399940490723
    },
    {
      "auxiliary_loss_clip": 0.01046752,
      "auxiliary_loss_mlp": 0.01023686,
      "balance_loss_clip": 1.03315115,
      "balance_loss_mlp": 1.01181281,
      "epoch": 0.7426424169547573,
      "flos": 21283572234240.0,
      "grad_norm": 2.1949301367231757,
      "language_loss": 0.8160581,
      "learning_rate": 6.553452654553611e-07,
      "loss": 0.83676243,
      "num_input_tokens_seen": 266424260,
      "step": 12352,
      "time_per_iteration": 2.6549971103668213
    },
    {
      "auxiliary_loss_clip": 0.0109629,
      "auxiliary_loss_mlp": 0.0103799,
      "balance_loss_clip": 1.03723967,
      "balance_loss_mlp": 1.02616394,
      "epoch": 0.7427025402074252,
      "flos": 22448442297600.0,
      "grad_norm": 1.7431938699053586,
      "language_loss": 0.71706545,
      "learning_rate": 6.550569904036307e-07,
      "loss": 0.73840821,
      "num_input_tokens_seen": 266444580,
      "step": 12353,
      "time_per_iteration": 2.509568929672241
    },
    {
      "auxiliary_loss_clip": 0.01097072,
      "auxiliary_loss_mlp": 0.01033797,
      "balance_loss_clip": 1.03870034,
      "balance_loss_mlp": 1.02247167,
      "epoch": 0.7427626634600932,
      "flos": 22524609087360.0,
      "grad_norm": 2.9185929903929386,
      "language_loss": 0.72465849,
      "learning_rate": 6.547687663522739e-07,
      "loss": 0.74596721,
      "num_input_tokens_seen": 266465640,
      "step": 12354,
      "time_per_iteration": 2.532309055328369
    },
    {
      "auxiliary_loss_clip": 0.01022298,
      "auxiliary_loss_mlp": 0.01003226,
      "balance_loss_clip": 1.00884223,
      "balance_loss_mlp": 1.00211143,
      "epoch": 0.7428227867127611,
      "flos": 67209477655680.0,
      "grad_norm": 0.6943867093696149,
      "language_loss": 0.59576714,
      "learning_rate": 6.544805933122199e-07,
      "loss": 0.61602235,
      "num_input_tokens_seen": 266531950,
      "step": 12355,
      "time_per_iteration": 3.2189323902130127
    },
    {
      "auxiliary_loss_clip": 0.01106296,
      "auxiliary_loss_mlp": 0.01026711,
      "balance_loss_clip": 1.03712058,
      "balance_loss_mlp": 1.01461148,
      "epoch": 0.7428829099654292,
      "flos": 14721221082240.0,
      "grad_norm": 1.70541606006716,
      "language_loss": 0.67694616,
      "learning_rate": 6.541924712943971e-07,
      "loss": 0.69827622,
      "num_input_tokens_seen": 266550665,
      "step": 12356,
      "time_per_iteration": 2.443842887878418
    },
    {
      "auxiliary_loss_clip": 0.01095071,
      "auxiliary_loss_mlp": 0.00783531,
      "balance_loss_clip": 1.03349912,
      "balance_loss_mlp": 1.00864649,
      "epoch": 0.7429430332180971,
      "flos": 48646496413440.0,
      "grad_norm": 1.9338474509350345,
      "language_loss": 0.72090971,
      "learning_rate": 6.539044003097301e-07,
      "loss": 0.73969573,
      "num_input_tokens_seen": 266572455,
      "step": 12357,
      "time_per_iteration": 4.112387657165527
    },
    {
      "auxiliary_loss_clip": 0.01084097,
      "auxiliary_loss_mlp": 0.0102925,
      "balance_loss_clip": 1.03950191,
      "balance_loss_mlp": 1.01800227,
      "epoch": 0.7430031564707651,
      "flos": 16764071281920.0,
      "grad_norm": 1.7276765347156942,
      "language_loss": 0.64835256,
      "learning_rate": 6.53616380369143e-07,
      "loss": 0.66948605,
      "num_input_tokens_seen": 266590895,
      "step": 12358,
      "time_per_iteration": 2.5044400691986084
    },
    {
      "auxiliary_loss_clip": 0.01066261,
      "auxiliary_loss_mlp": 0.01035521,
      "balance_loss_clip": 1.03662503,
      "balance_loss_mlp": 1.02224755,
      "epoch": 0.743063279723433,
      "flos": 23870576545920.0,
      "grad_norm": 1.7846772596610843,
      "language_loss": 0.80505908,
      "learning_rate": 6.533284114835591e-07,
      "loss": 0.82607692,
      "num_input_tokens_seen": 266607660,
      "step": 12359,
      "time_per_iteration": 2.603227376937866
    },
    {
      "auxiliary_loss_clip": 0.01093581,
      "auxiliary_loss_mlp": 0.0102616,
      "balance_loss_clip": 1.03496408,
      "balance_loss_mlp": 1.01480496,
      "epoch": 0.743123402976101,
      "flos": 14391704689920.0,
      "grad_norm": 1.939653931044939,
      "language_loss": 0.68305266,
      "learning_rate": 6.530404936638956e-07,
      "loss": 0.7042501,
      "num_input_tokens_seen": 266624260,
      "step": 12360,
      "time_per_iteration": 2.4554591178894043
    },
    {
      "auxiliary_loss_clip": 0.01091665,
      "auxiliary_loss_mlp": 0.00782153,
      "balance_loss_clip": 1.03448796,
      "balance_loss_mlp": 1.00752115,
      "epoch": 0.7431835262287689,
      "flos": 27454318335360.0,
      "grad_norm": 1.6272758711568318,
      "language_loss": 0.72789836,
      "learning_rate": 6.527526269210715e-07,
      "loss": 0.74663651,
      "num_input_tokens_seen": 266644210,
      "step": 12361,
      "time_per_iteration": 2.5588886737823486
    },
    {
      "auxiliary_loss_clip": 0.01056127,
      "auxiliary_loss_mlp": 0.01040713,
      "balance_loss_clip": 1.03247416,
      "balance_loss_mlp": 1.02715278,
      "epoch": 0.743243649481437,
      "flos": 20959514709120.0,
      "grad_norm": 1.8020576247952715,
      "language_loss": 0.56229144,
      "learning_rate": 6.524648112660027e-07,
      "loss": 0.58325982,
      "num_input_tokens_seen": 266664230,
      "step": 12362,
      "time_per_iteration": 2.584265947341919
    },
    {
      "auxiliary_loss_clip": 0.01068421,
      "auxiliary_loss_mlp": 0.01031904,
      "balance_loss_clip": 1.03451073,
      "balance_loss_mlp": 1.01933289,
      "epoch": 0.7433037727341049,
      "flos": 22783166161920.0,
      "grad_norm": 1.6320370110785174,
      "language_loss": 0.77455056,
      "learning_rate": 6.521770467096039e-07,
      "loss": 0.7955538,
      "num_input_tokens_seen": 266683270,
      "step": 12363,
      "time_per_iteration": 5.753237962722778
    },
    {
      "auxiliary_loss_clip": 0.01076774,
      "auxiliary_loss_mlp": 0.01032083,
      "balance_loss_clip": 1.03360009,
      "balance_loss_mlp": 1.02062082,
      "epoch": 0.7433638959867729,
      "flos": 22196708807040.0,
      "grad_norm": 1.7037097129540621,
      "language_loss": 0.77834237,
      "learning_rate": 6.518893332627862e-07,
      "loss": 0.79943097,
      "num_input_tokens_seen": 266701235,
      "step": 12364,
      "time_per_iteration": 2.541304349899292
    },
    {
      "auxiliary_loss_clip": 0.01093519,
      "auxiliary_loss_mlp": 0.01029548,
      "balance_loss_clip": 1.03546011,
      "balance_loss_mlp": 1.01789498,
      "epoch": 0.7434240192394409,
      "flos": 23296760778240.0,
      "grad_norm": 1.9560591689170765,
      "language_loss": 0.78389537,
      "learning_rate": 6.516016709364604e-07,
      "loss": 0.80512607,
      "num_input_tokens_seen": 266721495,
      "step": 12365,
      "time_per_iteration": 2.5341920852661133
    },
    {
      "auxiliary_loss_clip": 0.01081957,
      "auxiliary_loss_mlp": 0.01030996,
      "balance_loss_clip": 1.03560722,
      "balance_loss_mlp": 1.01874685,
      "epoch": 0.7434841424921088,
      "flos": 54009575251200.0,
      "grad_norm": 1.6890411966437404,
      "language_loss": 0.76614076,
      "learning_rate": 6.513140597415346e-07,
      "loss": 0.78727031,
      "num_input_tokens_seen": 266747400,
      "step": 12366,
      "time_per_iteration": 2.818577289581299
    },
    {
      "auxiliary_loss_clip": 0.01093982,
      "auxiliary_loss_mlp": 0.01028011,
      "balance_loss_clip": 1.03683269,
      "balance_loss_mlp": 1.01762748,
      "epoch": 0.7435442657447768,
      "flos": 21433966479360.0,
      "grad_norm": 1.6653859548796892,
      "language_loss": 0.71385932,
      "learning_rate": 6.510264996889141e-07,
      "loss": 0.73507923,
      "num_input_tokens_seen": 266767630,
      "step": 12367,
      "time_per_iteration": 2.4971561431884766
    },
    {
      "auxiliary_loss_clip": 0.01076806,
      "auxiliary_loss_mlp": 0.01031649,
      "balance_loss_clip": 1.03513336,
      "balance_loss_mlp": 1.01959705,
      "epoch": 0.7436043889974447,
      "flos": 24499408970880.0,
      "grad_norm": 1.5001740842134808,
      "language_loss": 0.7435993,
      "learning_rate": 6.507389907895038e-07,
      "loss": 0.76468384,
      "num_input_tokens_seen": 266788015,
      "step": 12368,
      "time_per_iteration": 3.97795033454895
    },
    {
      "auxiliary_loss_clip": 0.01089552,
      "auxiliary_loss_mlp": 0.01033103,
      "balance_loss_clip": 1.03580618,
      "balance_loss_mlp": 1.02266657,
      "epoch": 0.7436645122501128,
      "flos": 40698388512000.0,
      "grad_norm": 1.5744963604909499,
      "language_loss": 0.6956799,
      "learning_rate": 6.50451533054207e-07,
      "loss": 0.71690643,
      "num_input_tokens_seen": 266809010,
      "step": 12369,
      "time_per_iteration": 2.6392264366149902
    },
    {
      "auxiliary_loss_clip": 0.01083224,
      "auxiliary_loss_mlp": 0.00784156,
      "balance_loss_clip": 1.03649831,
      "balance_loss_mlp": 1.01199293,
      "epoch": 0.7437246355027807,
      "flos": 18908835344640.0,
      "grad_norm": 1.6692159850427977,
      "language_loss": 0.75667584,
      "learning_rate": 6.501641264939233e-07,
      "loss": 0.77534962,
      "num_input_tokens_seen": 266825390,
      "step": 12370,
      "time_per_iteration": 2.490330219268799
    },
    {
      "auxiliary_loss_clip": 0.01106166,
      "auxiliary_loss_mlp": 0.01034591,
      "balance_loss_clip": 1.03786302,
      "balance_loss_mlp": 1.02240181,
      "epoch": 0.7437847587554487,
      "flos": 21543817248000.0,
      "grad_norm": 1.5423763933580898,
      "language_loss": 0.78402841,
      "learning_rate": 6.498767711195503e-07,
      "loss": 0.80543602,
      "num_input_tokens_seen": 266844675,
      "step": 12371,
      "time_per_iteration": 2.460832118988037
    },
    {
      "auxiliary_loss_clip": 0.01084634,
      "auxiliary_loss_mlp": 0.01023797,
      "balance_loss_clip": 1.03567469,
      "balance_loss_mlp": 1.01183438,
      "epoch": 0.7438448820081166,
      "flos": 27782470010880.0,
      "grad_norm": 1.637399046615536,
      "language_loss": 0.6951288,
      "learning_rate": 6.495894669419857e-07,
      "loss": 0.71621311,
      "num_input_tokens_seen": 266865160,
      "step": 12372,
      "time_per_iteration": 2.568676471710205
    },
    {
      "auxiliary_loss_clip": 0.01073167,
      "auxiliary_loss_mlp": 0.01031017,
      "balance_loss_clip": 1.0343678,
      "balance_loss_mlp": 1.01882756,
      "epoch": 0.7439050052607846,
      "flos": 17967832796160.0,
      "grad_norm": 1.7961501093872643,
      "language_loss": 0.74712527,
      "learning_rate": 6.493022139721245e-07,
      "loss": 0.76816708,
      "num_input_tokens_seen": 266883285,
      "step": 12373,
      "time_per_iteration": 2.5030784606933594
    },
    {
      "auxiliary_loss_clip": 0.01051958,
      "auxiliary_loss_mlp": 0.0103257,
      "balance_loss_clip": 1.03185654,
      "balance_loss_mlp": 1.01883125,
      "epoch": 0.7439651285134525,
      "flos": 22958696949120.0,
      "grad_norm": 1.7475309172826103,
      "language_loss": 0.77259886,
      "learning_rate": 6.49015012220858e-07,
      "loss": 0.79344416,
      "num_input_tokens_seen": 266900960,
      "step": 12374,
      "time_per_iteration": 2.5844027996063232
    },
    {
      "auxiliary_loss_clip": 0.01051283,
      "auxiliary_loss_mlp": 0.0103251,
      "balance_loss_clip": 1.03596759,
      "balance_loss_mlp": 1.02014244,
      "epoch": 0.7440252517661206,
      "flos": 18806777827200.0,
      "grad_norm": 2.329495456202981,
      "language_loss": 0.7633515,
      "learning_rate": 6.487278616990774e-07,
      "loss": 0.78418946,
      "num_input_tokens_seen": 266917710,
      "step": 12375,
      "time_per_iteration": 2.6013636589050293
    },
    {
      "auxiliary_loss_clip": 0.01090345,
      "auxiliary_loss_mlp": 0.01025308,
      "balance_loss_clip": 1.03409648,
      "balance_loss_mlp": 1.0140965,
      "epoch": 0.7440853750187885,
      "flos": 20266295155200.0,
      "grad_norm": 1.9240422528425,
      "language_loss": 0.77238595,
      "learning_rate": 6.484407624176733e-07,
      "loss": 0.79354256,
      "num_input_tokens_seen": 266934220,
      "step": 12376,
      "time_per_iteration": 2.486158847808838
    },
    {
      "auxiliary_loss_clip": 0.01067018,
      "auxiliary_loss_mlp": 0.01037458,
      "balance_loss_clip": 1.03321755,
      "balance_loss_mlp": 1.02337909,
      "epoch": 0.7441454982714565,
      "flos": 25337276593920.0,
      "grad_norm": 1.5981571861867867,
      "language_loss": 0.79265428,
      "learning_rate": 6.481537143875296e-07,
      "loss": 0.81369901,
      "num_input_tokens_seen": 266955210,
      "step": 12377,
      "time_per_iteration": 2.5915939807891846
    },
    {
      "auxiliary_loss_clip": 0.01093134,
      "auxiliary_loss_mlp": 0.01028949,
      "balance_loss_clip": 1.03697896,
      "balance_loss_mlp": 1.0165391,
      "epoch": 0.7442056215241245,
      "flos": 64480910866560.0,
      "grad_norm": 1.8691545002945873,
      "language_loss": 0.66722727,
      "learning_rate": 6.478667176195322e-07,
      "loss": 0.68844807,
      "num_input_tokens_seen": 266976555,
      "step": 12378,
      "time_per_iteration": 2.861246109008789
    },
    {
      "auxiliary_loss_clip": 0.01071779,
      "auxiliary_loss_mlp": 0.01035166,
      "balance_loss_clip": 1.0362227,
      "balance_loss_mlp": 1.02218449,
      "epoch": 0.7442657447767924,
      "flos": 31285376242560.0,
      "grad_norm": 1.778539145163772,
      "language_loss": 0.7202723,
      "learning_rate": 6.475797721245648e-07,
      "loss": 0.74134183,
      "num_input_tokens_seen": 266997640,
      "step": 12379,
      "time_per_iteration": 2.6407508850097656
    },
    {
      "auxiliary_loss_clip": 0.01067882,
      "auxiliary_loss_mlp": 0.00786623,
      "balance_loss_clip": 1.03406501,
      "balance_loss_mlp": 1.01042354,
      "epoch": 0.7443258680294604,
      "flos": 20807899401600.0,
      "grad_norm": 1.725268945413912,
      "language_loss": 0.65343481,
      "learning_rate": 6.472928779135085e-07,
      "loss": 0.6719799,
      "num_input_tokens_seen": 267016165,
      "step": 12380,
      "time_per_iteration": 2.6069605350494385
    },
    {
      "auxiliary_loss_clip": 0.0109505,
      "auxiliary_loss_mlp": 0.01029185,
      "balance_loss_clip": 1.03596568,
      "balance_loss_mlp": 1.01728165,
      "epoch": 0.7443859912821283,
      "flos": 22199833290240.0,
      "grad_norm": 2.315025992082962,
      "language_loss": 0.78591037,
      "learning_rate": 6.470060349972411e-07,
      "loss": 0.80715275,
      "num_input_tokens_seen": 267034075,
      "step": 12381,
      "time_per_iteration": 2.4874672889709473
    },
    {
      "auxiliary_loss_clip": 0.01067122,
      "auxiliary_loss_mlp": 0.01039492,
      "balance_loss_clip": 1.03703356,
      "balance_loss_mlp": 1.02474594,
      "epoch": 0.7444461145347964,
      "flos": 22017838055040.0,
      "grad_norm": 2.0538053762110113,
      "language_loss": 0.7226764,
      "learning_rate": 6.467192433866411e-07,
      "loss": 0.74374253,
      "num_input_tokens_seen": 267053645,
      "step": 12382,
      "time_per_iteration": 2.5615084171295166
    },
    {
      "auxiliary_loss_clip": 0.01004166,
      "auxiliary_loss_mlp": 0.01000023,
      "balance_loss_clip": 1.01239932,
      "balance_loss_mlp": 0.99884856,
      "epoch": 0.7445062377874643,
      "flos": 70559047704960.0,
      "grad_norm": 0.6500777099565485,
      "language_loss": 0.54624617,
      "learning_rate": 6.464325030925831e-07,
      "loss": 0.56628805,
      "num_input_tokens_seen": 267121830,
      "step": 12383,
      "time_per_iteration": 3.3125407695770264
    },
    {
      "auxiliary_loss_clip": 0.01081417,
      "auxiliary_loss_mlp": 0.01028936,
      "balance_loss_clip": 1.03586936,
      "balance_loss_mlp": 1.01667547,
      "epoch": 0.7445663610401323,
      "flos": 22164425458560.0,
      "grad_norm": 2.1594034218612608,
      "language_loss": 0.76310909,
      "learning_rate": 6.461458141259395e-07,
      "loss": 0.78421259,
      "num_input_tokens_seen": 267141145,
      "step": 12384,
      "time_per_iteration": 2.5615503787994385
    },
    {
      "auxiliary_loss_clip": 0.01092791,
      "auxiliary_loss_mlp": 0.01029182,
      "balance_loss_clip": 1.03476858,
      "balance_loss_mlp": 1.01759446,
      "epoch": 0.7446264842928002,
      "flos": 24170251714560.0,
      "grad_norm": 1.9088672554956119,
      "language_loss": 0.79587984,
      "learning_rate": 6.458591764975823e-07,
      "loss": 0.81709957,
      "num_input_tokens_seen": 267159280,
      "step": 12385,
      "time_per_iteration": 2.532036542892456
    },
    {
      "auxiliary_loss_clip": 0.01073859,
      "auxiliary_loss_mlp": 0.01032025,
      "balance_loss_clip": 1.03846562,
      "balance_loss_mlp": 1.01887012,
      "epoch": 0.7446866075454682,
      "flos": 24134556574080.0,
      "grad_norm": 1.6521084208640353,
      "language_loss": 0.81345713,
      "learning_rate": 6.455725902183813e-07,
      "loss": 0.83451599,
      "num_input_tokens_seen": 267179390,
      "step": 12386,
      "time_per_iteration": 2.5866873264312744
    },
    {
      "auxiliary_loss_clip": 0.01089859,
      "auxiliary_loss_mlp": 0.01031004,
      "balance_loss_clip": 1.03620601,
      "balance_loss_mlp": 1.01942277,
      "epoch": 0.7447467307981361,
      "flos": 23548063305600.0,
      "grad_norm": 1.6122995564365425,
      "language_loss": 0.70803332,
      "learning_rate": 6.452860552992037e-07,
      "loss": 0.72924191,
      "num_input_tokens_seen": 267198165,
      "step": 12387,
      "time_per_iteration": 2.4957101345062256
    },
    {
      "auxiliary_loss_clip": 0.01073316,
      "auxiliary_loss_mlp": 0.01031503,
      "balance_loss_clip": 1.0358845,
      "balance_loss_mlp": 1.02004147,
      "epoch": 0.7448068540508042,
      "flos": 19567832215680.0,
      "grad_norm": 2.023178111803868,
      "language_loss": 0.70597714,
      "learning_rate": 6.449995717509138e-07,
      "loss": 0.72702539,
      "num_input_tokens_seen": 267214520,
      "step": 12388,
      "time_per_iteration": 2.551757574081421
    },
    {
      "auxiliary_loss_clip": 0.0109343,
      "auxiliary_loss_mlp": 0.01029424,
      "balance_loss_clip": 1.03460765,
      "balance_loss_mlp": 1.01769376,
      "epoch": 0.7448669773034721,
      "flos": 21839721488640.0,
      "grad_norm": 1.5977948183582167,
      "language_loss": 0.84676385,
      "learning_rate": 6.447131395843761e-07,
      "loss": 0.8679924,
      "num_input_tokens_seen": 267236555,
      "step": 12389,
      "time_per_iteration": 2.5446979999542236
    },
    {
      "auxiliary_loss_clip": 0.01065121,
      "auxiliary_loss_mlp": 0.01031811,
      "balance_loss_clip": 1.03659713,
      "balance_loss_mlp": 1.01970506,
      "epoch": 0.7449271005561401,
      "flos": 25155389099520.0,
      "grad_norm": 1.8767072819941821,
      "language_loss": 0.79346383,
      "learning_rate": 6.444267588104526e-07,
      "loss": 0.8144331,
      "num_input_tokens_seen": 267254800,
      "step": 12390,
      "time_per_iteration": 2.6248862743377686
    },
    {
      "auxiliary_loss_clip": 0.01085863,
      "auxiliary_loss_mlp": 0.01029854,
      "balance_loss_clip": 1.03658593,
      "balance_loss_mlp": 1.01717639,
      "epoch": 0.7449872238088081,
      "flos": 22273342473600.0,
      "grad_norm": 1.6217239665237047,
      "language_loss": 0.84840083,
      "learning_rate": 6.441404294400014e-07,
      "loss": 0.86955798,
      "num_input_tokens_seen": 267274610,
      "step": 12391,
      "time_per_iteration": 2.5180253982543945
    },
    {
      "auxiliary_loss_clip": 0.01104027,
      "auxiliary_loss_mlp": 0.01026877,
      "balance_loss_clip": 1.03569233,
      "balance_loss_mlp": 1.0157485,
      "epoch": 0.745047347061476,
      "flos": 20594805966720.0,
      "grad_norm": 2.1250168343804456,
      "language_loss": 0.73667687,
      "learning_rate": 6.438541514838811e-07,
      "loss": 0.75798595,
      "num_input_tokens_seen": 267292600,
      "step": 12392,
      "time_per_iteration": 2.4667131900787354
    },
    {
      "auxiliary_loss_clip": 0.01091324,
      "auxiliary_loss_mlp": 0.01033209,
      "balance_loss_clip": 1.03617072,
      "balance_loss_mlp": 1.02147293,
      "epoch": 0.745107470314144,
      "flos": 22127545169280.0,
      "grad_norm": 1.6255926631750188,
      "language_loss": 0.76883394,
      "learning_rate": 6.435679249529487e-07,
      "loss": 0.7900793,
      "num_input_tokens_seen": 267311295,
      "step": 12393,
      "time_per_iteration": 2.480015993118286
    },
    {
      "auxiliary_loss_clip": 0.01094961,
      "auxiliary_loss_mlp": 0.01032064,
      "balance_loss_clip": 1.03727579,
      "balance_loss_mlp": 1.01919568,
      "epoch": 0.745167593566812,
      "flos": 22236498097920.0,
      "grad_norm": 1.8572618435726926,
      "language_loss": 0.730129,
      "learning_rate": 6.432817498580552e-07,
      "loss": 0.75139928,
      "num_input_tokens_seen": 267328390,
      "step": 12394,
      "time_per_iteration": 2.502887010574341
    },
    {
      "auxiliary_loss_clip": 0.01048749,
      "auxiliary_loss_mlp": 0.00782923,
      "balance_loss_clip": 1.03658748,
      "balance_loss_mlp": 1.00961971,
      "epoch": 0.74522771681948,
      "flos": 20666232161280.0,
      "grad_norm": 1.5428169771739302,
      "language_loss": 0.81325608,
      "learning_rate": 6.429956262100535e-07,
      "loss": 0.83157283,
      "num_input_tokens_seen": 267348185,
      "step": 12395,
      "time_per_iteration": 4.051112413406372
    },
    {
      "auxiliary_loss_clip": 0.0109814,
      "auxiliary_loss_mlp": 0.01035027,
      "balance_loss_clip": 1.03646302,
      "balance_loss_mlp": 1.02252841,
      "epoch": 0.7452878400721479,
      "flos": 21106999952640.0,
      "grad_norm": 3.536633369747469,
      "language_loss": 0.71512079,
      "learning_rate": 6.427095540197937e-07,
      "loss": 0.7364524,
      "num_input_tokens_seen": 267367010,
      "step": 12396,
      "time_per_iteration": 2.4796435832977295
    },
    {
      "auxiliary_loss_clip": 0.01065593,
      "auxiliary_loss_mlp": 0.01029649,
      "balance_loss_clip": 1.03580785,
      "balance_loss_mlp": 1.01740015,
      "epoch": 0.7453479633248159,
      "flos": 26688056474880.0,
      "grad_norm": 2.0367224714382868,
      "language_loss": 0.68158466,
      "learning_rate": 6.424235332981245e-07,
      "loss": 0.70253706,
      "num_input_tokens_seen": 267386605,
      "step": 12397,
      "time_per_iteration": 2.6405365467071533
    },
    {
      "auxiliary_loss_clip": 0.01104122,
      "auxiliary_loss_mlp": 0.01040304,
      "balance_loss_clip": 1.03556001,
      "balance_loss_mlp": 1.02801943,
      "epoch": 0.7454080865774838,
      "flos": 17016056167680.0,
      "grad_norm": 1.7539178697213411,
      "language_loss": 0.76382834,
      "learning_rate": 6.421375640558908e-07,
      "loss": 0.7852726,
      "num_input_tokens_seen": 267404135,
      "step": 12398,
      "time_per_iteration": 2.4339704513549805
    },
    {
      "auxiliary_loss_clip": 0.01091969,
      "auxiliary_loss_mlp": 0.01025089,
      "balance_loss_clip": 1.03521514,
      "balance_loss_mlp": 1.01326358,
      "epoch": 0.7454682098301518,
      "flos": 21323900229120.0,
      "grad_norm": 2.120052450037395,
      "language_loss": 0.77946609,
      "learning_rate": 6.418516463039363e-07,
      "loss": 0.80063671,
      "num_input_tokens_seen": 267423120,
      "step": 12399,
      "time_per_iteration": 2.4947004318237305
    },
    {
      "auxiliary_loss_clip": 0.01079233,
      "auxiliary_loss_mlp": 0.01033017,
      "balance_loss_clip": 1.033355,
      "balance_loss_mlp": 1.02170444,
      "epoch": 0.7455283330828197,
      "flos": 17858341163520.0,
      "grad_norm": 1.860746586333181,
      "language_loss": 0.73501873,
      "learning_rate": 6.415657800531038e-07,
      "loss": 0.75614119,
      "num_input_tokens_seen": 267441250,
      "step": 12400,
      "time_per_iteration": 2.500904083251953
    },
    {
      "auxiliary_loss_clip": 0.01091335,
      "auxiliary_loss_mlp": 0.01032293,
      "balance_loss_clip": 1.03482556,
      "balance_loss_mlp": 1.02113485,
      "epoch": 0.7455884563354878,
      "flos": 30774259664640.0,
      "grad_norm": 1.766492035128556,
      "language_loss": 0.8214227,
      "learning_rate": 6.412799653142327e-07,
      "loss": 0.84265894,
      "num_input_tokens_seen": 267462820,
      "step": 12401,
      "time_per_iteration": 3.941330909729004
    },
    {
      "auxiliary_loss_clip": 0.01071564,
      "auxiliary_loss_mlp": 0.01031387,
      "balance_loss_clip": 1.0353992,
      "balance_loss_mlp": 1.01975846,
      "epoch": 0.7456485795881557,
      "flos": 23185545292800.0,
      "grad_norm": 1.9466936697628179,
      "language_loss": 0.652969,
      "learning_rate": 6.409942020981611e-07,
      "loss": 0.67399848,
      "num_input_tokens_seen": 267483065,
      "step": 12402,
      "time_per_iteration": 3.9990975856781006
    },
    {
      "auxiliary_loss_clip": 0.01072266,
      "auxiliary_loss_mlp": 0.01032889,
      "balance_loss_clip": 1.0338726,
      "balance_loss_mlp": 1.02146912,
      "epoch": 0.7457087028408237,
      "flos": 38727144074880.0,
      "grad_norm": 1.5580717337306913,
      "language_loss": 0.73367298,
      "learning_rate": 6.407084904157265e-07,
      "loss": 0.7547245,
      "num_input_tokens_seen": 267504825,
      "step": 12403,
      "time_per_iteration": 2.7135751247406006
    },
    {
      "auxiliary_loss_clip": 0.01001921,
      "auxiliary_loss_mlp": 0.01002808,
      "balance_loss_clip": 1.0113852,
      "balance_loss_mlp": 1.00159764,
      "epoch": 0.7457688260934917,
      "flos": 56043737337600.0,
      "grad_norm": 0.8256386603819094,
      "language_loss": 0.58750796,
      "learning_rate": 6.404228302777621e-07,
      "loss": 0.60755527,
      "num_input_tokens_seen": 267559260,
      "step": 12404,
      "time_per_iteration": 3.0052683353424072
    },
    {
      "auxiliary_loss_clip": 0.01103339,
      "auxiliary_loss_mlp": 0.01031334,
      "balance_loss_clip": 1.03452325,
      "balance_loss_mlp": 1.01993132,
      "epoch": 0.7458289493461596,
      "flos": 20116152305280.0,
      "grad_norm": 2.4351817039950974,
      "language_loss": 0.77622783,
      "learning_rate": 6.401372216950995e-07,
      "loss": 0.79757452,
      "num_input_tokens_seen": 267578720,
      "step": 12405,
      "time_per_iteration": 2.461237907409668
    },
    {
      "auxiliary_loss_clip": 0.01078406,
      "auxiliary_loss_mlp": 0.01038069,
      "balance_loss_clip": 1.03384829,
      "balance_loss_mlp": 1.02576089,
      "epoch": 0.7458890725988276,
      "flos": 20193073280640.0,
      "grad_norm": 1.5406022317438208,
      "language_loss": 0.69120997,
      "learning_rate": 6.398516646785698e-07,
      "loss": 0.71237475,
      "num_input_tokens_seen": 267598250,
      "step": 12406,
      "time_per_iteration": 2.5138683319091797
    },
    {
      "auxiliary_loss_clip": 0.01046941,
      "auxiliary_loss_mlp": 0.01035865,
      "balance_loss_clip": 1.03546214,
      "balance_loss_mlp": 1.02212024,
      "epoch": 0.7459491958514956,
      "flos": 17018749687680.0,
      "grad_norm": 1.7497696791019406,
      "language_loss": 0.65075558,
      "learning_rate": 6.39566159239002e-07,
      "loss": 0.67158365,
      "num_input_tokens_seen": 267615430,
      "step": 12407,
      "time_per_iteration": 4.109792470932007
    },
    {
      "auxiliary_loss_clip": 0.01064132,
      "auxiliary_loss_mlp": 0.01033954,
      "balance_loss_clip": 1.0344516,
      "balance_loss_mlp": 1.02079284,
      "epoch": 0.7460093191041636,
      "flos": 25078719519360.0,
      "grad_norm": 1.720993934943781,
      "language_loss": 0.7200129,
      "learning_rate": 6.392807053872212e-07,
      "loss": 0.74099374,
      "num_input_tokens_seen": 267635075,
      "step": 12408,
      "time_per_iteration": 2.8240721225738525
    },
    {
      "auxiliary_loss_clip": 0.01100194,
      "auxiliary_loss_mlp": 0.01033706,
      "balance_loss_clip": 1.03724384,
      "balance_loss_mlp": 1.01985967,
      "epoch": 0.7460694423568315,
      "flos": 21908525990400.0,
      "grad_norm": 1.8452217382321383,
      "language_loss": 0.72992671,
      "learning_rate": 6.38995303134053e-07,
      "loss": 0.75126576,
      "num_input_tokens_seen": 267654105,
      "step": 12409,
      "time_per_iteration": 2.550278425216675
    },
    {
      "auxiliary_loss_clip": 0.01090938,
      "auxiliary_loss_mlp": 0.0102981,
      "balance_loss_clip": 1.03478551,
      "balance_loss_mlp": 1.01861048,
      "epoch": 0.7461295656094995,
      "flos": 21215737399680.0,
      "grad_norm": 1.7073036449418648,
      "language_loss": 0.65929961,
      "learning_rate": 6.38709952490319e-07,
      "loss": 0.68050712,
      "num_input_tokens_seen": 267673090,
      "step": 12410,
      "time_per_iteration": 2.4936752319335938
    },
    {
      "auxiliary_loss_clip": 0.01087737,
      "auxiliary_loss_mlp": 0.00782981,
      "balance_loss_clip": 1.03485405,
      "balance_loss_mlp": 1.01002574,
      "epoch": 0.7461896888621674,
      "flos": 22346851656960.0,
      "grad_norm": 1.8982360030683214,
      "language_loss": 0.84597725,
      "learning_rate": 6.384246534668396e-07,
      "loss": 0.86468446,
      "num_input_tokens_seen": 267690605,
      "step": 12411,
      "time_per_iteration": 2.505321741104126
    },
    {
      "auxiliary_loss_clip": 0.01070281,
      "auxiliary_loss_mlp": 0.01028284,
      "balance_loss_clip": 1.03437328,
      "balance_loss_mlp": 1.01594567,
      "epoch": 0.7462498121148354,
      "flos": 25482930243840.0,
      "grad_norm": 1.572197790781628,
      "language_loss": 0.77791572,
      "learning_rate": 6.381394060744339e-07,
      "loss": 0.79890138,
      "num_input_tokens_seen": 267710540,
      "step": 12412,
      "time_per_iteration": 2.5694098472595215
    },
    {
      "auxiliary_loss_clip": 0.01065352,
      "auxiliary_loss_mlp": 0.01037015,
      "balance_loss_clip": 1.03409457,
      "balance_loss_mlp": 1.0245223,
      "epoch": 0.7463099353675033,
      "flos": 33947936812800.0,
      "grad_norm": 1.8897931548494664,
      "language_loss": 0.62467396,
      "learning_rate": 6.378542103239188e-07,
      "loss": 0.64569771,
      "num_input_tokens_seen": 267730780,
      "step": 12413,
      "time_per_iteration": 2.697270631790161
    },
    {
      "auxiliary_loss_clip": 0.0102444,
      "auxiliary_loss_mlp": 0.00762593,
      "balance_loss_clip": 1.00989246,
      "balance_loss_mlp": 1.00180292,
      "epoch": 0.7463700586201714,
      "flos": 62767723691520.0,
      "grad_norm": 0.7268810703358441,
      "language_loss": 0.54919755,
      "learning_rate": 6.375690662261082e-07,
      "loss": 0.56706786,
      "num_input_tokens_seen": 267794240,
      "step": 12414,
      "time_per_iteration": 3.1534206867218018
    },
    {
      "auxiliary_loss_clip": 0.01077667,
      "auxiliary_loss_mlp": 0.01029731,
      "balance_loss_clip": 1.03545654,
      "balance_loss_mlp": 1.01735735,
      "epoch": 0.7464301818728393,
      "flos": 33432654257280.0,
      "grad_norm": 1.5880463360297183,
      "language_loss": 0.54558623,
      "learning_rate": 6.372839737918154e-07,
      "loss": 0.56666023,
      "num_input_tokens_seen": 267817190,
      "step": 12415,
      "time_per_iteration": 2.6484315395355225
    },
    {
      "auxiliary_loss_clip": 0.01049329,
      "auxiliary_loss_mlp": 0.01035325,
      "balance_loss_clip": 1.03477812,
      "balance_loss_mlp": 1.02133012,
      "epoch": 0.7464903051255073,
      "flos": 26869872142080.0,
      "grad_norm": 1.608871418961689,
      "language_loss": 0.75192833,
      "learning_rate": 6.369989330318506e-07,
      "loss": 0.77277488,
      "num_input_tokens_seen": 267836245,
      "step": 12416,
      "time_per_iteration": 2.6346628665924072
    },
    {
      "auxiliary_loss_clip": 0.0105569,
      "auxiliary_loss_mlp": 0.01037665,
      "balance_loss_clip": 1.03170812,
      "balance_loss_mlp": 1.02483857,
      "epoch": 0.7465504283781753,
      "flos": 44086954775040.0,
      "grad_norm": 1.6389189809097362,
      "language_loss": 0.69190031,
      "learning_rate": 6.367139439570233e-07,
      "loss": 0.71283388,
      "num_input_tokens_seen": 267858310,
      "step": 12417,
      "time_per_iteration": 2.7954540252685547
    },
    {
      "auxiliary_loss_clip": 0.01077469,
      "auxiliary_loss_mlp": 0.0103233,
      "balance_loss_clip": 1.03752697,
      "balance_loss_mlp": 1.0195272,
      "epoch": 0.7466105516308432,
      "flos": 19676102785920.0,
      "grad_norm": 2.5094616068259548,
      "language_loss": 0.73859775,
      "learning_rate": 6.364290065781392e-07,
      "loss": 0.75969577,
      "num_input_tokens_seen": 267876345,
      "step": 12418,
      "time_per_iteration": 2.548858642578125
    },
    {
      "auxiliary_loss_clip": 0.01094606,
      "auxiliary_loss_mlp": 0.01034931,
      "balance_loss_clip": 1.03652477,
      "balance_loss_mlp": 1.02323675,
      "epoch": 0.7466706748835112,
      "flos": 20520722165760.0,
      "grad_norm": 1.8629605609829527,
      "language_loss": 0.69168425,
      "learning_rate": 6.361441209060039e-07,
      "loss": 0.71297956,
      "num_input_tokens_seen": 267896740,
      "step": 12419,
      "time_per_iteration": 2.5440244674682617
    },
    {
      "auxiliary_loss_clip": 0.01099196,
      "auxiliary_loss_mlp": 0.01031123,
      "balance_loss_clip": 1.03452468,
      "balance_loss_mlp": 1.01975608,
      "epoch": 0.7467307981361792,
      "flos": 21690260997120.0,
      "grad_norm": 1.8581267420314864,
      "language_loss": 0.75032157,
      "learning_rate": 6.358592869514216e-07,
      "loss": 0.77162468,
      "num_input_tokens_seen": 267914765,
      "step": 12420,
      "time_per_iteration": 2.4803249835968018
    },
    {
      "auxiliary_loss_clip": 0.01099485,
      "auxiliary_loss_mlp": 0.01030695,
      "balance_loss_clip": 1.03826952,
      "balance_loss_mlp": 1.01818371,
      "epoch": 0.7467909213888472,
      "flos": 19573686132480.0,
      "grad_norm": 1.8293994625569678,
      "language_loss": 0.67677462,
      "learning_rate": 6.355745047251904e-07,
      "loss": 0.69807643,
      "num_input_tokens_seen": 267934085,
      "step": 12421,
      "time_per_iteration": 2.502230167388916
    },
    {
      "auxiliary_loss_clip": 0.01081668,
      "auxiliary_loss_mlp": 0.01033915,
      "balance_loss_clip": 1.03578043,
      "balance_loss_mlp": 1.0208621,
      "epoch": 0.7468510446415151,
      "flos": 23695225326720.0,
      "grad_norm": 1.6103472251205657,
      "language_loss": 0.7243799,
      "learning_rate": 6.352897742381107e-07,
      "loss": 0.74553567,
      "num_input_tokens_seen": 267955170,
      "step": 12422,
      "time_per_iteration": 2.564944267272949
    },
    {
      "auxiliary_loss_clip": 0.01067413,
      "auxiliary_loss_mlp": 0.01033846,
      "balance_loss_clip": 1.0354979,
      "balance_loss_mlp": 1.02094126,
      "epoch": 0.7469111678941831,
      "flos": 29315783831040.0,
      "grad_norm": 1.7830951979353022,
      "language_loss": 0.74818158,
      "learning_rate": 6.350050955009796e-07,
      "loss": 0.76919425,
      "num_input_tokens_seen": 267974980,
      "step": 12423,
      "time_per_iteration": 2.6318697929382324
    },
    {
      "auxiliary_loss_clip": 0.01090645,
      "auxiliary_loss_mlp": 0.01024671,
      "balance_loss_clip": 1.03428459,
      "balance_loss_mlp": 1.01361394,
      "epoch": 0.746971291146851,
      "flos": 21798639308160.0,
      "grad_norm": 1.302575394146366,
      "language_loss": 0.67558765,
      "learning_rate": 6.347204685245929e-07,
      "loss": 0.69674081,
      "num_input_tokens_seen": 267994985,
      "step": 12424,
      "time_per_iteration": 2.5149121284484863
    },
    {
      "auxiliary_loss_clip": 0.0110027,
      "auxiliary_loss_mlp": 0.0103526,
      "balance_loss_clip": 1.03860736,
      "balance_loss_mlp": 1.02290988,
      "epoch": 0.747031414399519,
      "flos": 36245070368640.0,
      "grad_norm": 1.9662124985774336,
      "language_loss": 0.74729902,
      "learning_rate": 6.344358933197418e-07,
      "loss": 0.76865429,
      "num_input_tokens_seen": 268014985,
      "step": 12425,
      "time_per_iteration": 2.635683298110962
    },
    {
      "auxiliary_loss_clip": 0.01066407,
      "auxiliary_loss_mlp": 0.01034863,
      "balance_loss_clip": 1.03403187,
      "balance_loss_mlp": 1.02167857,
      "epoch": 0.7470915376521869,
      "flos": 19974916028160.0,
      "grad_norm": 2.8282388979012363,
      "language_loss": 0.69383752,
      "learning_rate": 6.341513698972194e-07,
      "loss": 0.71485019,
      "num_input_tokens_seen": 268034395,
      "step": 12426,
      "time_per_iteration": 2.5815205574035645
    },
    {
      "auxiliary_loss_clip": 0.01065373,
      "auxiliary_loss_mlp": 0.01034693,
      "balance_loss_clip": 1.03642392,
      "balance_loss_mlp": 1.02302885,
      "epoch": 0.747151660904855,
      "flos": 20084299920000.0,
      "grad_norm": 1.5758304755642805,
      "language_loss": 0.65460479,
      "learning_rate": 6.338668982678139e-07,
      "loss": 0.67560542,
      "num_input_tokens_seen": 268054485,
      "step": 12427,
      "time_per_iteration": 2.5713250637054443
    },
    {
      "auxiliary_loss_clip": 0.01106475,
      "auxiliary_loss_mlp": 0.01029511,
      "balance_loss_clip": 1.03739572,
      "balance_loss_mlp": 1.01683879,
      "epoch": 0.7472117841575229,
      "flos": 16290373697280.0,
      "grad_norm": 1.5791368985642704,
      "language_loss": 0.74561083,
      "learning_rate": 6.335824784423118e-07,
      "loss": 0.76697063,
      "num_input_tokens_seen": 268072250,
      "step": 12428,
      "time_per_iteration": 2.4341745376586914
    },
    {
      "auxiliary_loss_clip": 0.01100511,
      "auxiliary_loss_mlp": 0.01033425,
      "balance_loss_clip": 1.03712916,
      "balance_loss_mlp": 1.02031231,
      "epoch": 0.7472719074101909,
      "flos": 21389939383680.0,
      "grad_norm": 1.9086889270479022,
      "language_loss": 0.58178741,
      "learning_rate": 6.33298110431499e-07,
      "loss": 0.60312676,
      "num_input_tokens_seen": 268089840,
      "step": 12429,
      "time_per_iteration": 2.4761509895324707
    },
    {
      "auxiliary_loss_clip": 0.01094534,
      "auxiliary_loss_mlp": 0.01030321,
      "balance_loss_clip": 1.03719127,
      "balance_loss_mlp": 1.01782751,
      "epoch": 0.7473320306628589,
      "flos": 29643289061760.0,
      "grad_norm": 2.044743133589867,
      "language_loss": 0.605986,
      "learning_rate": 6.330137942461595e-07,
      "loss": 0.62723458,
      "num_input_tokens_seen": 268109360,
      "step": 12430,
      "time_per_iteration": 2.5812690258026123
    },
    {
      "auxiliary_loss_clip": 0.01084736,
      "auxiliary_loss_mlp": 0.01028562,
      "balance_loss_clip": 1.03627658,
      "balance_loss_mlp": 1.01667726,
      "epoch": 0.7473921539155268,
      "flos": 24136100858880.0,
      "grad_norm": 1.479504046573755,
      "language_loss": 0.75575566,
      "learning_rate": 6.327295298970734e-07,
      "loss": 0.77688867,
      "num_input_tokens_seen": 268131840,
      "step": 12431,
      "time_per_iteration": 2.5966789722442627
    },
    {
      "auxiliary_loss_clip": 0.01095479,
      "auxiliary_loss_mlp": 0.01028713,
      "balance_loss_clip": 1.03631783,
      "balance_loss_mlp": 1.01689363,
      "epoch": 0.7474522771681948,
      "flos": 17487958072320.0,
      "grad_norm": 1.9919547552345311,
      "language_loss": 0.75544977,
      "learning_rate": 6.32445317395021e-07,
      "loss": 0.77669168,
      "num_input_tokens_seen": 268148300,
      "step": 12432,
      "time_per_iteration": 2.478036403656006
    },
    {
      "auxiliary_loss_clip": 0.01084649,
      "auxiliary_loss_mlp": 0.01032802,
      "balance_loss_clip": 1.0370667,
      "balance_loss_mlp": 1.01924181,
      "epoch": 0.7475124004208628,
      "flos": 16727298733440.0,
      "grad_norm": 2.414570790059739,
      "language_loss": 0.69943875,
      "learning_rate": 6.321611567507787e-07,
      "loss": 0.7206133,
      "num_input_tokens_seen": 268166450,
      "step": 12433,
      "time_per_iteration": 2.5433192253112793
    },
    {
      "auxiliary_loss_clip": 0.01059251,
      "auxiliary_loss_mlp": 0.01032843,
      "balance_loss_clip": 1.0360868,
      "balance_loss_mlp": 1.02018857,
      "epoch": 0.7475725236735308,
      "flos": 19720237622400.0,
      "grad_norm": 1.6620386462184908,
      "language_loss": 0.66803324,
      "learning_rate": 6.318770479751232e-07,
      "loss": 0.68895423,
      "num_input_tokens_seen": 268186165,
      "step": 12434,
      "time_per_iteration": 4.218998432159424
    },
    {
      "auxiliary_loss_clip": 0.01099636,
      "auxiliary_loss_mlp": 0.01032653,
      "balance_loss_clip": 1.03534472,
      "balance_loss_mlp": 1.02177536,
      "epoch": 0.7476326469261987,
      "flos": 26286000566400.0,
      "grad_norm": 1.434380456712659,
      "language_loss": 0.79551721,
      "learning_rate": 6.315929910788263e-07,
      "loss": 0.81684005,
      "num_input_tokens_seen": 268208145,
      "step": 12435,
      "time_per_iteration": 2.518249034881592
    },
    {
      "auxiliary_loss_clip": 0.01075397,
      "auxiliary_loss_mlp": 0.01030022,
      "balance_loss_clip": 1.03368044,
      "balance_loss_mlp": 1.01780868,
      "epoch": 0.7476927701788667,
      "flos": 31831828824960.0,
      "grad_norm": 32.027791172425424,
      "language_loss": 0.67792928,
      "learning_rate": 6.313089860726604e-07,
      "loss": 0.69898343,
      "num_input_tokens_seen": 268228345,
      "step": 12436,
      "time_per_iteration": 2.635502338409424
    },
    {
      "auxiliary_loss_clip": 0.01079236,
      "auxiliary_loss_mlp": 0.01033935,
      "balance_loss_clip": 1.03418994,
      "balance_loss_mlp": 1.02155542,
      "epoch": 0.7477528934315346,
      "flos": 31795487239680.0,
      "grad_norm": 1.655573065822874,
      "language_loss": 0.70599282,
      "learning_rate": 6.31025032967396e-07,
      "loss": 0.72712457,
      "num_input_tokens_seen": 268250260,
      "step": 12437,
      "time_per_iteration": 2.630236864089966
    },
    {
      "auxiliary_loss_clip": 0.01066568,
      "auxiliary_loss_mlp": 0.0102678,
      "balance_loss_clip": 1.0330534,
      "balance_loss_mlp": 1.01550913,
      "epoch": 0.7478130166842026,
      "flos": 20371979946240.0,
      "grad_norm": 1.8328517645202864,
      "language_loss": 0.67336512,
      "learning_rate": 6.307411317737986e-07,
      "loss": 0.69429862,
      "num_input_tokens_seen": 268268440,
      "step": 12438,
      "time_per_iteration": 2.575533151626587
    },
    {
      "auxiliary_loss_clip": 0.01082883,
      "auxiliary_loss_mlp": 0.01034648,
      "balance_loss_clip": 1.03486514,
      "balance_loss_mlp": 1.02267957,
      "epoch": 0.7478731399368705,
      "flos": 18148930191360.0,
      "grad_norm": 1.6858175316956214,
      "language_loss": 0.80595887,
      "learning_rate": 6.304572825026344e-07,
      "loss": 0.82713413,
      "num_input_tokens_seen": 268285765,
      "step": 12439,
      "time_per_iteration": 2.5026497840881348
    },
    {
      "auxiliary_loss_clip": 0.01069517,
      "auxiliary_loss_mlp": 0.01038072,
      "balance_loss_clip": 1.03429341,
      "balance_loss_mlp": 1.02615678,
      "epoch": 0.7479332631895386,
      "flos": 15267889146240.0,
      "grad_norm": 1.9812843426398712,
      "language_loss": 0.71043396,
      "learning_rate": 6.301734851646674e-07,
      "loss": 0.73150986,
      "num_input_tokens_seen": 268304015,
      "step": 12440,
      "time_per_iteration": 3.9029359817504883
    },
    {
      "auxiliary_loss_clip": 0.0108216,
      "auxiliary_loss_mlp": 0.01026581,
      "balance_loss_clip": 1.03732789,
      "balance_loss_mlp": 1.01485646,
      "epoch": 0.7479933864422065,
      "flos": 21142515525120.0,
      "grad_norm": 1.726529693038754,
      "language_loss": 0.74414635,
      "learning_rate": 6.298897397706597e-07,
      "loss": 0.76523376,
      "num_input_tokens_seen": 268323290,
      "step": 12441,
      "time_per_iteration": 3.974076509475708
    },
    {
      "auxiliary_loss_clip": 0.01097548,
      "auxiliary_loss_mlp": 0.00783987,
      "balance_loss_clip": 1.03742123,
      "balance_loss_mlp": 1.01083052,
      "epoch": 0.7480535096948745,
      "flos": 14392027912320.0,
      "grad_norm": 2.166831080165801,
      "language_loss": 0.82586074,
      "learning_rate": 6.296060463313698e-07,
      "loss": 0.84467608,
      "num_input_tokens_seen": 268339490,
      "step": 12442,
      "time_per_iteration": 2.477626323699951
    },
    {
      "auxiliary_loss_clip": 0.01050288,
      "auxiliary_loss_mlp": 0.01033684,
      "balance_loss_clip": 1.03520107,
      "balance_loss_mlp": 1.02030301,
      "epoch": 0.7481136329475425,
      "flos": 27344683048320.0,
      "grad_norm": 1.8598648867797591,
      "language_loss": 0.62883502,
      "learning_rate": 6.293224048575565e-07,
      "loss": 0.64967471,
      "num_input_tokens_seen": 268359865,
      "step": 12443,
      "time_per_iteration": 2.6732797622680664
    },
    {
      "auxiliary_loss_clip": 0.01063522,
      "auxiliary_loss_mlp": 0.01028352,
      "balance_loss_clip": 1.03391886,
      "balance_loss_mlp": 1.01665151,
      "epoch": 0.7481737562002104,
      "flos": 19531454716800.0,
      "grad_norm": 1.67046864570416,
      "language_loss": 0.7110272,
      "learning_rate": 6.29038815359975e-07,
      "loss": 0.73194593,
      "num_input_tokens_seen": 268377065,
      "step": 12444,
      "time_per_iteration": 2.548701763153076
    },
    {
      "auxiliary_loss_clip": 0.01054697,
      "auxiliary_loss_mlp": 0.01033517,
      "balance_loss_clip": 1.03509355,
      "balance_loss_mlp": 1.02075005,
      "epoch": 0.7482338794528784,
      "flos": 21760035166080.0,
      "grad_norm": 1.4093311387286585,
      "language_loss": 0.68831599,
      "learning_rate": 6.287552778493786e-07,
      "loss": 0.70919818,
      "num_input_tokens_seen": 268396935,
      "step": 12445,
      "time_per_iteration": 4.071219444274902
    },
    {
      "auxiliary_loss_clip": 0.01091384,
      "auxiliary_loss_mlp": 0.01024757,
      "balance_loss_clip": 1.03504121,
      "balance_loss_mlp": 1.01325965,
      "epoch": 0.7482940027055464,
      "flos": 18697358021760.0,
      "grad_norm": 1.722012330933655,
      "language_loss": 0.74150407,
      "learning_rate": 6.28471792336519e-07,
      "loss": 0.76266551,
      "num_input_tokens_seen": 268414460,
      "step": 12446,
      "time_per_iteration": 2.4746510982513428
    },
    {
      "auxiliary_loss_clip": 0.01088461,
      "auxiliary_loss_mlp": 0.00785153,
      "balance_loss_clip": 1.03661585,
      "balance_loss_mlp": 1.01037312,
      "epoch": 0.7483541259582144,
      "flos": 15998024903040.0,
      "grad_norm": 2.2092870502886854,
      "language_loss": 0.73504388,
      "learning_rate": 6.281883588321475e-07,
      "loss": 0.75377995,
      "num_input_tokens_seen": 268432225,
      "step": 12447,
      "time_per_iteration": 2.5000741481781006
    },
    {
      "auxiliary_loss_clip": 0.01061349,
      "auxiliary_loss_mlp": 0.01031328,
      "balance_loss_clip": 1.03255415,
      "balance_loss_mlp": 1.01981282,
      "epoch": 0.7484142492108823,
      "flos": 25556295772800.0,
      "grad_norm": 4.867366876804958,
      "language_loss": 0.7201705,
      "learning_rate": 6.279049773470109e-07,
      "loss": 0.74109727,
      "num_input_tokens_seen": 268449270,
      "step": 12448,
      "time_per_iteration": 2.616562843322754
    },
    {
      "auxiliary_loss_clip": 0.01107663,
      "auxiliary_loss_mlp": 0.01035201,
      "balance_loss_clip": 1.03737342,
      "balance_loss_mlp": 1.02313721,
      "epoch": 0.7484743724635503,
      "flos": 22887737631360.0,
      "grad_norm": 1.9151443353946096,
      "language_loss": 0.73624802,
      "learning_rate": 6.276216478918543e-07,
      "loss": 0.75767672,
      "num_input_tokens_seen": 268467250,
      "step": 12449,
      "time_per_iteration": 2.479686975479126
    },
    {
      "auxiliary_loss_clip": 0.01080436,
      "auxiliary_loss_mlp": 0.01034332,
      "balance_loss_clip": 1.03797388,
      "balance_loss_mlp": 1.02139759,
      "epoch": 0.7485344957162182,
      "flos": 25300288563840.0,
      "grad_norm": 1.8881210961400068,
      "language_loss": 0.60763562,
      "learning_rate": 6.273383704774225e-07,
      "loss": 0.62878323,
      "num_input_tokens_seen": 268487270,
      "step": 12450,
      "time_per_iteration": 2.578500270843506
    },
    {
      "auxiliary_loss_clip": 0.01099914,
      "auxiliary_loss_mlp": 0.01025249,
      "balance_loss_clip": 1.03453016,
      "balance_loss_mlp": 1.01414514,
      "epoch": 0.7485946189688862,
      "flos": 27053016612480.0,
      "grad_norm": 3.0501001132599326,
      "language_loss": 0.70339847,
      "learning_rate": 6.270551451144577e-07,
      "loss": 0.72465014,
      "num_input_tokens_seen": 268508020,
      "step": 12451,
      "time_per_iteration": 2.5189316272735596
    },
    {
      "auxiliary_loss_clip": 0.01098567,
      "auxiliary_loss_mlp": 0.01027996,
      "balance_loss_clip": 1.03562903,
      "balance_loss_mlp": 1.01577067,
      "epoch": 0.7486547422215541,
      "flos": 26906752431360.0,
      "grad_norm": 3.161847754308371,
      "language_loss": 0.80074203,
      "learning_rate": 6.267719718136988e-07,
      "loss": 0.82200766,
      "num_input_tokens_seen": 268527375,
      "step": 12452,
      "time_per_iteration": 2.5053582191467285
    },
    {
      "auxiliary_loss_clip": 0.01113413,
      "auxiliary_loss_mlp": 0.01032466,
      "balance_loss_clip": 1.04023433,
      "balance_loss_mlp": 1.01972818,
      "epoch": 0.7487148654742222,
      "flos": 22346277039360.0,
      "grad_norm": 2.2011923882668527,
      "language_loss": 0.71658796,
      "learning_rate": 6.264888505858843e-07,
      "loss": 0.73804677,
      "num_input_tokens_seen": 268544870,
      "step": 12453,
      "time_per_iteration": 2.4709975719451904
    },
    {
      "auxiliary_loss_clip": 0.01085476,
      "auxiliary_loss_mlp": 0.01032094,
      "balance_loss_clip": 1.03777647,
      "balance_loss_mlp": 1.01995873,
      "epoch": 0.7487749887268901,
      "flos": 23038814234880.0,
      "grad_norm": 1.520752990944129,
      "language_loss": 0.74211639,
      "learning_rate": 6.262057814417517e-07,
      "loss": 0.76329207,
      "num_input_tokens_seen": 268564580,
      "step": 12454,
      "time_per_iteration": 2.5405733585357666
    },
    {
      "auxiliary_loss_clip": 0.01013916,
      "auxiliary_loss_mlp": 0.01017857,
      "balance_loss_clip": 1.01334596,
      "balance_loss_mlp": 1.01631916,
      "epoch": 0.7488351119795581,
      "flos": 71525294536320.0,
      "grad_norm": 0.7342577747056931,
      "language_loss": 0.59430796,
      "learning_rate": 6.259227643920322e-07,
      "loss": 0.61462569,
      "num_input_tokens_seen": 268629550,
      "step": 12455,
      "time_per_iteration": 3.320308208465576
    },
    {
      "auxiliary_loss_clip": 0.01069064,
      "auxiliary_loss_mlp": 0.01027563,
      "balance_loss_clip": 1.03719401,
      "balance_loss_mlp": 1.01523089,
      "epoch": 0.748895235232226,
      "flos": 17196255722880.0,
      "grad_norm": 2.032098726270048,
      "language_loss": 0.79506797,
      "learning_rate": 6.256397994474592e-07,
      "loss": 0.81603426,
      "num_input_tokens_seen": 268646645,
      "step": 12456,
      "time_per_iteration": 2.550816059112549
    },
    {
      "auxiliary_loss_clip": 0.01025797,
      "auxiliary_loss_mlp": 0.01002083,
      "balance_loss_clip": 1.01129258,
      "balance_loss_mlp": 1.00102806,
      "epoch": 0.748955358484894,
      "flos": 58979256336000.0,
      "grad_norm": 0.8613344432054116,
      "language_loss": 0.6140939,
      "learning_rate": 6.25356886618763e-07,
      "loss": 0.63437271,
      "num_input_tokens_seen": 268702275,
      "step": 12457,
      "time_per_iteration": 3.045292615890503
    },
    {
      "auxiliary_loss_clip": 0.01090652,
      "auxiliary_loss_mlp": 0.01031206,
      "balance_loss_clip": 1.0386076,
      "balance_loss_mlp": 1.01901734,
      "epoch": 0.749015481737562,
      "flos": 11360413054080.0,
      "grad_norm": 2.016159158164655,
      "language_loss": 0.67491674,
      "learning_rate": 6.250740259166711e-07,
      "loss": 0.6961354,
      "num_input_tokens_seen": 268716265,
      "step": 12458,
      "time_per_iteration": 2.501405954360962
    },
    {
      "auxiliary_loss_clip": 0.01052039,
      "auxiliary_loss_mlp": 0.01029212,
      "balance_loss_clip": 1.03448761,
      "balance_loss_mlp": 1.01765501,
      "epoch": 0.74907560499023,
      "flos": 21106497162240.0,
      "grad_norm": 1.6746090213527325,
      "language_loss": 0.79762006,
      "learning_rate": 6.247912173519106e-07,
      "loss": 0.81843257,
      "num_input_tokens_seen": 268734330,
      "step": 12459,
      "time_per_iteration": 2.6022870540618896
    },
    {
      "auxiliary_loss_clip": 0.01066764,
      "auxiliary_loss_mlp": 0.0103338,
      "balance_loss_clip": 1.0335784,
      "balance_loss_mlp": 1.02085125,
      "epoch": 0.749135728242898,
      "flos": 22268027260800.0,
      "grad_norm": 1.8603524967098515,
      "language_loss": 0.80517602,
      "learning_rate": 6.245084609352043e-07,
      "loss": 0.82617736,
      "num_input_tokens_seen": 268753500,
      "step": 12460,
      "time_per_iteration": 2.566892385482788
    },
    {
      "auxiliary_loss_clip": 0.01078156,
      "auxiliary_loss_mlp": 0.01033863,
      "balance_loss_clip": 1.03513861,
      "balance_loss_mlp": 1.02080345,
      "epoch": 0.7491958514955659,
      "flos": 24057527857920.0,
      "grad_norm": 1.7506406240741834,
      "language_loss": 0.86306703,
      "learning_rate": 6.242257566772755e-07,
      "loss": 0.88418722,
      "num_input_tokens_seen": 268772055,
      "step": 12461,
      "time_per_iteration": 2.563171863555908
    },
    {
      "auxiliary_loss_clip": 0.01087427,
      "auxiliary_loss_mlp": 0.01032833,
      "balance_loss_clip": 1.03709018,
      "balance_loss_mlp": 1.02112687,
      "epoch": 0.7492559747482339,
      "flos": 24492118510080.0,
      "grad_norm": 1.7608249579379034,
      "language_loss": 0.68932426,
      "learning_rate": 6.239431045888435e-07,
      "loss": 0.71052694,
      "num_input_tokens_seen": 268792265,
      "step": 12462,
      "time_per_iteration": 2.530583620071411
    },
    {
      "auxiliary_loss_clip": 0.01104257,
      "auxiliary_loss_mlp": 0.01030706,
      "balance_loss_clip": 1.03525817,
      "balance_loss_mlp": 1.01821339,
      "epoch": 0.7493160980009018,
      "flos": 27745338326400.0,
      "grad_norm": 2.468987515788426,
      "language_loss": 0.70595902,
      "learning_rate": 6.236605046806267e-07,
      "loss": 0.72730863,
      "num_input_tokens_seen": 268812735,
      "step": 12463,
      "time_per_iteration": 2.5265681743621826
    },
    {
      "auxiliary_loss_clip": 0.01069538,
      "auxiliary_loss_mlp": 0.01027795,
      "balance_loss_clip": 1.03509438,
      "balance_loss_mlp": 1.01623762,
      "epoch": 0.7493762212535698,
      "flos": 30226190970240.0,
      "grad_norm": 1.5948281881498696,
      "language_loss": 0.7773369,
      "learning_rate": 6.233779569633419e-07,
      "loss": 0.79831022,
      "num_input_tokens_seen": 268833090,
      "step": 12464,
      "time_per_iteration": 2.622408151626587
    },
    {
      "auxiliary_loss_clip": 0.01079622,
      "auxiliary_loss_mlp": 0.01025579,
      "balance_loss_clip": 1.03282642,
      "balance_loss_mlp": 1.01421201,
      "epoch": 0.7494363445062378,
      "flos": 21944472526080.0,
      "grad_norm": 1.6152166143014708,
      "language_loss": 0.78504294,
      "learning_rate": 6.230954614477034e-07,
      "loss": 0.80609494,
      "num_input_tokens_seen": 268851880,
      "step": 12465,
      "time_per_iteration": 2.538890838623047
    },
    {
      "auxiliary_loss_clip": 0.01077263,
      "auxiliary_loss_mlp": 0.01035251,
      "balance_loss_clip": 1.03739274,
      "balance_loss_mlp": 1.02167296,
      "epoch": 0.7494964677589058,
      "flos": 12490342162560.0,
      "grad_norm": 2.3752099787136225,
      "language_loss": 0.74618471,
      "learning_rate": 6.22813018144422e-07,
      "loss": 0.7673099,
      "num_input_tokens_seen": 268867910,
      "step": 12466,
      "time_per_iteration": 2.533190965652466
    },
    {
      "auxiliary_loss_clip": 0.01095695,
      "auxiliary_loss_mlp": 0.01031894,
      "balance_loss_clip": 1.0356679,
      "balance_loss_mlp": 1.01965129,
      "epoch": 0.7495565910115737,
      "flos": 21653057485440.0,
      "grad_norm": 2.0377570435141865,
      "language_loss": 0.66279203,
      "learning_rate": 6.22530627064209e-07,
      "loss": 0.6840679,
      "num_input_tokens_seen": 268887260,
      "step": 12467,
      "time_per_iteration": 2.4945473670959473
    },
    {
      "auxiliary_loss_clip": 0.01058624,
      "auxiliary_loss_mlp": 0.00786945,
      "balance_loss_clip": 1.03659654,
      "balance_loss_mlp": 1.01122284,
      "epoch": 0.7496167142642417,
      "flos": 15268535591040.0,
      "grad_norm": 2.0651796327331913,
      "language_loss": 0.76126587,
      "learning_rate": 6.222482882177735e-07,
      "loss": 0.77972156,
      "num_input_tokens_seen": 268902520,
      "step": 12468,
      "time_per_iteration": 2.5627007484436035
    },
    {
      "auxiliary_loss_clip": 0.01066476,
      "auxiliary_loss_mlp": 0.01030823,
      "balance_loss_clip": 1.03748989,
      "balance_loss_mlp": 1.0180918,
      "epoch": 0.7496768375169096,
      "flos": 22054933825920.0,
      "grad_norm": 1.8753619878756969,
      "language_loss": 0.69013447,
      "learning_rate": 6.219660016158201e-07,
      "loss": 0.71110749,
      "num_input_tokens_seen": 268920970,
      "step": 12469,
      "time_per_iteration": 2.570352077484131
    },
    {
      "auxiliary_loss_clip": 0.01086475,
      "auxiliary_loss_mlp": 0.01032539,
      "balance_loss_clip": 1.03749919,
      "balance_loss_mlp": 1.01999784,
      "epoch": 0.7497369607695776,
      "flos": 19057038860160.0,
      "grad_norm": 1.983508551044083,
      "language_loss": 0.69340867,
      "learning_rate": 6.216837672690543e-07,
      "loss": 0.71459877,
      "num_input_tokens_seen": 268936600,
      "step": 12470,
      "time_per_iteration": 2.512080430984497
    },
    {
      "auxiliary_loss_clip": 0.01088825,
      "auxiliary_loss_mlp": 0.01031489,
      "balance_loss_clip": 1.04165363,
      "balance_loss_mlp": 1.01796484,
      "epoch": 0.7497970840222457,
      "flos": 21617434172160.0,
      "grad_norm": 2.0300373116112937,
      "language_loss": 0.75586671,
      "learning_rate": 6.214015851881793e-07,
      "loss": 0.77706981,
      "num_input_tokens_seen": 268956560,
      "step": 12471,
      "time_per_iteration": 2.5457096099853516
    },
    {
      "auxiliary_loss_clip": 0.01080013,
      "auxiliary_loss_mlp": 0.01029752,
      "balance_loss_clip": 1.03437686,
      "balance_loss_mlp": 1.01577497,
      "epoch": 0.7498572072749136,
      "flos": 13735580906880.0,
      "grad_norm": 2.4843794320911643,
      "language_loss": 0.76614523,
      "learning_rate": 6.211194553838929e-07,
      "loss": 0.78724289,
      "num_input_tokens_seen": 268973945,
      "step": 12472,
      "time_per_iteration": 3.8707194328308105
    },
    {
      "auxiliary_loss_clip": 0.01093498,
      "auxiliary_loss_mlp": 0.00782876,
      "balance_loss_clip": 1.03602827,
      "balance_loss_mlp": 1.00894523,
      "epoch": 0.7499173305275816,
      "flos": 22966526113920.0,
      "grad_norm": 1.5025750254311407,
      "language_loss": 0.84220791,
      "learning_rate": 6.208373778668951e-07,
      "loss": 0.86097163,
      "num_input_tokens_seen": 268993245,
      "step": 12473,
      "time_per_iteration": 2.5149974822998047
    },
    {
      "auxiliary_loss_clip": 0.01076569,
      "auxiliary_loss_mlp": 0.01036294,
      "balance_loss_clip": 1.03507638,
      "balance_loss_mlp": 1.0226028,
      "epoch": 0.7499774537802495,
      "flos": 22740467869440.0,
      "grad_norm": 2.2100310436147983,
      "language_loss": 0.7370646,
      "learning_rate": 6.205553526478829e-07,
      "loss": 0.75819314,
      "num_input_tokens_seen": 269012125,
      "step": 12474,
      "time_per_iteration": 2.5586583614349365
    },
    {
      "auxiliary_loss_clip": 0.01083043,
      "auxiliary_loss_mlp": 0.01034206,
      "balance_loss_clip": 1.03558993,
      "balance_loss_mlp": 1.0214746,
      "epoch": 0.7500375770329175,
      "flos": 18296559089280.0,
      "grad_norm": 1.797763285121591,
      "language_loss": 0.74605834,
      "learning_rate": 6.202733797375492e-07,
      "loss": 0.76723075,
      "num_input_tokens_seen": 269030545,
      "step": 12475,
      "time_per_iteration": 2.497840642929077
    },
    {
      "auxiliary_loss_clip": 0.0110244,
      "auxiliary_loss_mlp": 0.01034104,
      "balance_loss_clip": 1.03876638,
      "balance_loss_mlp": 1.02069867,
      "epoch": 0.7500977002855854,
      "flos": 19169978198400.0,
      "grad_norm": 1.7779836276187349,
      "language_loss": 0.79786074,
      "learning_rate": 6.199914591465878e-07,
      "loss": 0.81922615,
      "num_input_tokens_seen": 269048180,
      "step": 12476,
      "time_per_iteration": 2.4936680793762207
    },
    {
      "auxiliary_loss_clip": 0.01070962,
      "auxiliary_loss_mlp": 0.01033361,
      "balance_loss_clip": 1.03539705,
      "balance_loss_mlp": 1.02170229,
      "epoch": 0.7501578235382534,
      "flos": 22163886754560.0,
      "grad_norm": 1.7544852382350766,
      "language_loss": 0.77934766,
      "learning_rate": 6.19709590885688e-07,
      "loss": 0.8003909,
      "num_input_tokens_seen": 269068600,
      "step": 12477,
      "time_per_iteration": 2.577725648880005
    },
    {
      "auxiliary_loss_clip": 0.01012965,
      "auxiliary_loss_mlp": 0.0099957,
      "balance_loss_clip": 1.01122522,
      "balance_loss_mlp": 0.9983902,
      "epoch": 0.7502179467909214,
      "flos": 64465040033280.0,
      "grad_norm": 0.805701740048689,
      "language_loss": 0.54472798,
      "learning_rate": 6.194277749655394e-07,
      "loss": 0.56485331,
      "num_input_tokens_seen": 269119045,
      "step": 12478,
      "time_per_iteration": 4.483763217926025
    },
    {
      "auxiliary_loss_clip": 0.01077599,
      "auxiliary_loss_mlp": 0.01033383,
      "balance_loss_clip": 1.03477907,
      "balance_loss_mlp": 1.02137256,
      "epoch": 0.7502780700435894,
      "flos": 20478275268480.0,
      "grad_norm": 1.6477925934653805,
      "language_loss": 0.79980731,
      "learning_rate": 6.191460113968272e-07,
      "loss": 0.82091713,
      "num_input_tokens_seen": 269136755,
      "step": 12479,
      "time_per_iteration": 2.594290018081665
    },
    {
      "auxiliary_loss_clip": 0.01098884,
      "auxiliary_loss_mlp": 0.0103784,
      "balance_loss_clip": 1.03707492,
      "balance_loss_mlp": 1.0246439,
      "epoch": 0.7503381932962573,
      "flos": 20445273648000.0,
      "grad_norm": 1.9395856420488617,
      "language_loss": 0.62359059,
      "learning_rate": 6.188643001902369e-07,
      "loss": 0.64495784,
      "num_input_tokens_seen": 269156120,
      "step": 12480,
      "time_per_iteration": 3.9500463008880615
    },
    {
      "auxiliary_loss_clip": 0.01078388,
      "auxiliary_loss_mlp": 0.01036415,
      "balance_loss_clip": 1.03468585,
      "balance_loss_mlp": 1.02489924,
      "epoch": 0.7503983165489253,
      "flos": 22381936266240.0,
      "grad_norm": 1.5634155071932831,
      "language_loss": 0.7767731,
      "learning_rate": 6.185826413564512e-07,
      "loss": 0.79792106,
      "num_input_tokens_seen": 269175650,
      "step": 12481,
      "time_per_iteration": 2.535130500793457
    },
    {
      "auxiliary_loss_clip": 0.01065351,
      "auxiliary_loss_mlp": 0.01032879,
      "balance_loss_clip": 1.03530777,
      "balance_loss_mlp": 1.01961124,
      "epoch": 0.7504584398015932,
      "flos": 24899453717760.0,
      "grad_norm": 1.6055308566556097,
      "language_loss": 0.71363711,
      "learning_rate": 6.183010349061501e-07,
      "loss": 0.7346195,
      "num_input_tokens_seen": 269197080,
      "step": 12482,
      "time_per_iteration": 2.599626302719116
    },
    {
      "auxiliary_loss_clip": 0.01107663,
      "auxiliary_loss_mlp": 0.01035861,
      "balance_loss_clip": 1.03783417,
      "balance_loss_mlp": 1.02360666,
      "epoch": 0.7505185630542612,
      "flos": 25885237547520.0,
      "grad_norm": 1.6924906764911605,
      "language_loss": 0.70293677,
      "learning_rate": 6.180194808500118e-07,
      "loss": 0.72437203,
      "num_input_tokens_seen": 269218600,
      "step": 12483,
      "time_per_iteration": 2.4972894191741943
    },
    {
      "auxiliary_loss_clip": 0.01105686,
      "auxiliary_loss_mlp": 0.01028894,
      "balance_loss_clip": 1.03642642,
      "balance_loss_mlp": 1.01761651,
      "epoch": 0.7505786863069293,
      "flos": 23143852581120.0,
      "grad_norm": 2.0273934120648063,
      "language_loss": 0.74658406,
      "learning_rate": 6.177379791987131e-07,
      "loss": 0.76792991,
      "num_input_tokens_seen": 269239245,
      "step": 12484,
      "time_per_iteration": 3.8632073402404785
    },
    {
      "auxiliary_loss_clip": 0.01083175,
      "auxiliary_loss_mlp": 0.01029149,
      "balance_loss_clip": 1.03525305,
      "balance_loss_mlp": 1.01643491,
      "epoch": 0.7506388095595972,
      "flos": 16983377769600.0,
      "grad_norm": 2.602467666306162,
      "language_loss": 0.84431762,
      "learning_rate": 6.174565299629295e-07,
      "loss": 0.86544085,
      "num_input_tokens_seen": 269258520,
      "step": 12485,
      "time_per_iteration": 2.5065524578094482
    },
    {
      "auxiliary_loss_clip": 0.01069283,
      "auxiliary_loss_mlp": 0.01029851,
      "balance_loss_clip": 1.03610015,
      "balance_loss_mlp": 1.01759017,
      "epoch": 0.7506989328122652,
      "flos": 22344984149760.0,
      "grad_norm": 1.5288288394289904,
      "language_loss": 0.78119469,
      "learning_rate": 6.171751331533323e-07,
      "loss": 0.80218601,
      "num_input_tokens_seen": 269278320,
      "step": 12486,
      "time_per_iteration": 2.5666418075561523
    },
    {
      "auxiliary_loss_clip": 0.01095877,
      "auxiliary_loss_mlp": 0.01030903,
      "balance_loss_clip": 1.03617835,
      "balance_loss_mlp": 1.01789725,
      "epoch": 0.7507590560649331,
      "flos": 25776069137280.0,
      "grad_norm": 2.077392676872717,
      "language_loss": 0.72770011,
      "learning_rate": 6.168937887805932e-07,
      "loss": 0.74896801,
      "num_input_tokens_seen": 269298025,
      "step": 12487,
      "time_per_iteration": 2.5141847133636475
    },
    {
      "auxiliary_loss_clip": 0.01080616,
      "auxiliary_loss_mlp": 0.01027643,
      "balance_loss_clip": 1.03378534,
      "balance_loss_mlp": 1.01570427,
      "epoch": 0.7508191793176011,
      "flos": 24279420124800.0,
      "grad_norm": 1.9813379377058418,
      "language_loss": 0.67140996,
      "learning_rate": 6.166124968553801e-07,
      "loss": 0.6924926,
      "num_input_tokens_seen": 269316770,
      "step": 12488,
      "time_per_iteration": 2.553664445877075
    },
    {
      "auxiliary_loss_clip": 0.01040443,
      "auxiliary_loss_mlp": 0.0103401,
      "balance_loss_clip": 1.03276873,
      "balance_loss_mlp": 1.02143979,
      "epoch": 0.750879302570269,
      "flos": 19899575251200.0,
      "grad_norm": 1.7317282443134243,
      "language_loss": 0.77263552,
      "learning_rate": 6.163312573883592e-07,
      "loss": 0.79338002,
      "num_input_tokens_seen": 269334755,
      "step": 12489,
      "time_per_iteration": 2.6048166751861572
    },
    {
      "auxiliary_loss_clip": 0.0109321,
      "auxiliary_loss_mlp": 0.01030344,
      "balance_loss_clip": 1.03730869,
      "balance_loss_mlp": 1.019472,
      "epoch": 0.750939425822937,
      "flos": 29205681667200.0,
      "grad_norm": 2.0422525380887424,
      "language_loss": 0.75238216,
      "learning_rate": 6.160500703901956e-07,
      "loss": 0.77361763,
      "num_input_tokens_seen": 269353810,
      "step": 12490,
      "time_per_iteration": 2.542771339416504
    },
    {
      "auxiliary_loss_clip": 0.0110468,
      "auxiliary_loss_mlp": 0.01028058,
      "balance_loss_clip": 1.0369184,
      "balance_loss_mlp": 1.01622677,
      "epoch": 0.750999549075605,
      "flos": 21142300043520.0,
      "grad_norm": 1.558997955361212,
      "language_loss": 0.7814759,
      "learning_rate": 6.157689358715527e-07,
      "loss": 0.80280328,
      "num_input_tokens_seen": 269372910,
      "step": 12491,
      "time_per_iteration": 2.449381113052368
    },
    {
      "auxiliary_loss_clip": 0.01092419,
      "auxiliary_loss_mlp": 0.01029256,
      "balance_loss_clip": 1.03472555,
      "balance_loss_mlp": 1.0186944,
      "epoch": 0.751059672328273,
      "flos": 23547740083200.0,
      "grad_norm": 1.9051801324386155,
      "language_loss": 0.76483464,
      "learning_rate": 6.154878538430899e-07,
      "loss": 0.78605139,
      "num_input_tokens_seen": 269391545,
      "step": 12492,
      "time_per_iteration": 2.518357753753662
    },
    {
      "auxiliary_loss_clip": 0.01073169,
      "auxiliary_loss_mlp": 0.01029318,
      "balance_loss_clip": 1.03490686,
      "balance_loss_mlp": 1.01789141,
      "epoch": 0.7511197955809409,
      "flos": 18989742729600.0,
      "grad_norm": 2.0052366052901744,
      "language_loss": 0.7098459,
      "learning_rate": 6.152068243154671e-07,
      "loss": 0.73087078,
      "num_input_tokens_seen": 269408530,
      "step": 12493,
      "time_per_iteration": 2.529813766479492
    },
    {
      "auxiliary_loss_clip": 0.01096058,
      "auxiliary_loss_mlp": 0.00783117,
      "balance_loss_clip": 1.03725648,
      "balance_loss_mlp": 1.01124501,
      "epoch": 0.7511799188336089,
      "flos": 22046961006720.0,
      "grad_norm": 1.6436004341278811,
      "language_loss": 0.80853724,
      "learning_rate": 6.149258472993395e-07,
      "loss": 0.82732892,
      "num_input_tokens_seen": 269425930,
      "step": 12494,
      "time_per_iteration": 2.5087790489196777
    },
    {
      "auxiliary_loss_clip": 0.01106553,
      "auxiliary_loss_mlp": 0.01028071,
      "balance_loss_clip": 1.0369935,
      "balance_loss_mlp": 1.01572108,
      "epoch": 0.7512400420862768,
      "flos": 16467125546880.0,
      "grad_norm": 1.9267796723662196,
      "language_loss": 0.78950649,
      "learning_rate": 6.146449228053634e-07,
      "loss": 0.81085277,
      "num_input_tokens_seen": 269443945,
      "step": 12495,
      "time_per_iteration": 2.472052812576294
    },
    {
      "auxiliary_loss_clip": 0.01104882,
      "auxiliary_loss_mlp": 0.00781977,
      "balance_loss_clip": 1.03623748,
      "balance_loss_mlp": 1.00880241,
      "epoch": 0.7513001653389448,
      "flos": 20448326304000.0,
      "grad_norm": 2.030564241967498,
      "language_loss": 0.71490884,
      "learning_rate": 6.143640508441898e-07,
      "loss": 0.7337774,
      "num_input_tokens_seen": 269463625,
      "step": 12496,
      "time_per_iteration": 2.482271432876587
    },
    {
      "auxiliary_loss_clip": 0.01059889,
      "auxiliary_loss_mlp": 0.01032734,
      "balance_loss_clip": 1.03244698,
      "balance_loss_mlp": 1.02077127,
      "epoch": 0.7513602885916129,
      "flos": 23476816679040.0,
      "grad_norm": 1.5967325430846957,
      "language_loss": 0.78375411,
      "learning_rate": 6.140832314264705e-07,
      "loss": 0.80468035,
      "num_input_tokens_seen": 269483415,
      "step": 12497,
      "time_per_iteration": 2.6006524562835693
    },
    {
      "auxiliary_loss_clip": 0.01097504,
      "auxiliary_loss_mlp": 0.01035633,
      "balance_loss_clip": 1.03721046,
      "balance_loss_mlp": 1.02287197,
      "epoch": 0.7514204118442808,
      "flos": 26797224885120.0,
      "grad_norm": 1.5108013816480954,
      "language_loss": 0.76755548,
      "learning_rate": 6.13802464562855e-07,
      "loss": 0.78888679,
      "num_input_tokens_seen": 269504635,
      "step": 12498,
      "time_per_iteration": 2.549534320831299
    },
    {
      "auxiliary_loss_clip": 0.01082991,
      "auxiliary_loss_mlp": 0.0103486,
      "balance_loss_clip": 1.03844893,
      "balance_loss_mlp": 1.02326155,
      "epoch": 0.7514805350969488,
      "flos": 19865639877120.0,
      "grad_norm": 1.7289059979201555,
      "language_loss": 0.74087769,
      "learning_rate": 6.135217502639878e-07,
      "loss": 0.76205623,
      "num_input_tokens_seen": 269523955,
      "step": 12499,
      "time_per_iteration": 2.5353164672851562
    },
    {
      "auxiliary_loss_clip": 0.01091207,
      "auxiliary_loss_mlp": 0.01025764,
      "balance_loss_clip": 1.03311515,
      "balance_loss_mlp": 1.01483274,
      "epoch": 0.7515406583496167,
      "flos": 24571553437440.0,
      "grad_norm": 1.851485553278697,
      "language_loss": 0.79190719,
      "learning_rate": 6.132410885405148e-07,
      "loss": 0.81307691,
      "num_input_tokens_seen": 269544410,
      "step": 12500,
      "time_per_iteration": 2.5767288208007812
    },
    {
      "auxiliary_loss_clip": 0.01103505,
      "auxiliary_loss_mlp": 0.01035638,
      "balance_loss_clip": 1.03817797,
      "balance_loss_mlp": 1.0216428,
      "epoch": 0.7516007816022847,
      "flos": 20120246455680.0,
      "grad_norm": 1.8585060915413656,
      "language_loss": 0.73630524,
      "learning_rate": 6.129604794030794e-07,
      "loss": 0.75769669,
      "num_input_tokens_seen": 269563315,
      "step": 12501,
      "time_per_iteration": 2.487550735473633
    },
    {
      "auxiliary_loss_clip": 0.01082531,
      "auxiliary_loss_mlp": 0.01024632,
      "balance_loss_clip": 1.03509104,
      "balance_loss_mlp": 1.01224041,
      "epoch": 0.7516609048549526,
      "flos": 22784638619520.0,
      "grad_norm": 1.5630119983134898,
      "language_loss": 0.78517497,
      "learning_rate": 6.126799228623207e-07,
      "loss": 0.80624664,
      "num_input_tokens_seen": 269583950,
      "step": 12502,
      "time_per_iteration": 2.587296962738037
    },
    {
      "auxiliary_loss_clip": 0.01083312,
      "auxiliary_loss_mlp": 0.01031257,
      "balance_loss_clip": 1.0350852,
      "balance_loss_mlp": 1.01960981,
      "epoch": 0.7517210281076206,
      "flos": 10634012311680.0,
      "grad_norm": 2.3861117684436914,
      "language_loss": 0.70581466,
      "learning_rate": 6.123994189288786e-07,
      "loss": 0.72696036,
      "num_input_tokens_seen": 269600120,
      "step": 12503,
      "time_per_iteration": 2.4832592010498047
    },
    {
      "auxiliary_loss_clip": 0.01032322,
      "auxiliary_loss_mlp": 0.01003837,
      "balance_loss_clip": 1.00914514,
      "balance_loss_mlp": 1.00271034,
      "epoch": 0.7517811513602886,
      "flos": 66052221275520.0,
      "grad_norm": 0.9838714999072278,
      "language_loss": 0.6398136,
      "learning_rate": 6.121189676133903e-07,
      "loss": 0.66017526,
      "num_input_tokens_seen": 269659815,
      "step": 12504,
      "time_per_iteration": 3.0037269592285156
    },
    {
      "auxiliary_loss_clip": 0.01062527,
      "auxiliary_loss_mlp": 0.0103125,
      "balance_loss_clip": 1.03114057,
      "balance_loss_mlp": 1.01940703,
      "epoch": 0.7518412746129566,
      "flos": 37268345018880.0,
      "grad_norm": 1.4955568079651682,
      "language_loss": 0.68525279,
      "learning_rate": 6.118385689264896e-07,
      "loss": 0.70619053,
      "num_input_tokens_seen": 269684565,
      "step": 12505,
      "time_per_iteration": 2.7339110374450684
    },
    {
      "auxiliary_loss_clip": 0.01022613,
      "auxiliary_loss_mlp": 0.00762797,
      "balance_loss_clip": 1.00993311,
      "balance_loss_mlp": 1.00259876,
      "epoch": 0.7519013978656245,
      "flos": 60518567727360.0,
      "grad_norm": 0.6436769394826811,
      "language_loss": 0.55091321,
      "learning_rate": 6.11558222878809e-07,
      "loss": 0.56876731,
      "num_input_tokens_seen": 269752325,
      "step": 12506,
      "time_per_iteration": 3.2125961780548096
    },
    {
      "auxiliary_loss_clip": 0.01094191,
      "auxiliary_loss_mlp": 0.01034972,
      "balance_loss_clip": 1.03673744,
      "balance_loss_mlp": 1.02273536,
      "epoch": 0.7519615211182925,
      "flos": 18806885568000.0,
      "grad_norm": 1.8599667222002583,
      "language_loss": 0.78526068,
      "learning_rate": 6.112779294809796e-07,
      "loss": 0.80655229,
      "num_input_tokens_seen": 269770630,
      "step": 12507,
      "time_per_iteration": 2.51926326751709
    },
    {
      "auxiliary_loss_clip": 0.01078328,
      "auxiliary_loss_mlp": 0.01031769,
      "balance_loss_clip": 1.03709793,
      "balance_loss_mlp": 1.02025294,
      "epoch": 0.7520216443709604,
      "flos": 14575244209920.0,
      "grad_norm": 1.5890886033139742,
      "language_loss": 0.71113914,
      "learning_rate": 6.10997688743631e-07,
      "loss": 0.73224008,
      "num_input_tokens_seen": 269787280,
      "step": 12508,
      "time_per_iteration": 2.497087001800537
    },
    {
      "auxiliary_loss_clip": 0.01088883,
      "auxiliary_loss_mlp": 0.01028601,
      "balance_loss_clip": 1.0339818,
      "balance_loss_mlp": 1.01647186,
      "epoch": 0.7520817676236284,
      "flos": 17056599644160.0,
      "grad_norm": 1.5343955525117614,
      "language_loss": 0.7161088,
      "learning_rate": 6.107175006773885e-07,
      "loss": 0.73728359,
      "num_input_tokens_seen": 269805205,
      "step": 12509,
      "time_per_iteration": 2.490513801574707
    },
    {
      "auxiliary_loss_clip": 0.01111069,
      "auxiliary_loss_mlp": 0.01038515,
      "balance_loss_clip": 1.03751802,
      "balance_loss_mlp": 1.02503824,
      "epoch": 0.7521418908762965,
      "flos": 25666397936640.0,
      "grad_norm": 1.7030303986704753,
      "language_loss": 0.61797899,
      "learning_rate": 6.104373652928785e-07,
      "loss": 0.63947481,
      "num_input_tokens_seen": 269824820,
      "step": 12510,
      "time_per_iteration": 2.4850845336914062
    },
    {
      "auxiliary_loss_clip": 0.01090995,
      "auxiliary_loss_mlp": 0.01030431,
      "balance_loss_clip": 1.03548336,
      "balance_loss_mlp": 1.01893318,
      "epoch": 0.7522020141289644,
      "flos": 20886759711360.0,
      "grad_norm": 1.7668134657850443,
      "language_loss": 0.81390828,
      "learning_rate": 6.10157282600722e-07,
      "loss": 0.83512253,
      "num_input_tokens_seen": 269842825,
      "step": 12511,
      "time_per_iteration": 3.950269937515259
    },
    {
      "auxiliary_loss_clip": 0.01081035,
      "auxiliary_loss_mlp": 0.01033566,
      "balance_loss_clip": 1.0358156,
      "balance_loss_mlp": 1.02067399,
      "epoch": 0.7522621373816324,
      "flos": 12640305444480.0,
      "grad_norm": 2.04263083550264,
      "language_loss": 0.75599831,
      "learning_rate": 6.098772526115412e-07,
      "loss": 0.77714431,
      "num_input_tokens_seen": 269859000,
      "step": 12512,
      "time_per_iteration": 2.5187625885009766
    },
    {
      "auxiliary_loss_clip": 0.01087366,
      "auxiliary_loss_mlp": 0.01024929,
      "balance_loss_clip": 1.0332253,
      "balance_loss_mlp": 1.014153,
      "epoch": 0.7523222606343003,
      "flos": 25626141768960.0,
      "grad_norm": 1.6884203900080537,
      "language_loss": 0.82305825,
      "learning_rate": 6.095972753359537e-07,
      "loss": 0.84418118,
      "num_input_tokens_seen": 269878895,
      "step": 12513,
      "time_per_iteration": 2.562002420425415
    },
    {
      "auxiliary_loss_clip": 0.01093579,
      "auxiliary_loss_mlp": 0.01035041,
      "balance_loss_clip": 1.03697681,
      "balance_loss_mlp": 1.02238655,
      "epoch": 0.7523823838869683,
      "flos": 20448900921600.0,
      "grad_norm": 3.515412038649654,
      "language_loss": 0.74769056,
      "learning_rate": 6.093173507845771e-07,
      "loss": 0.76897675,
      "num_input_tokens_seen": 269897280,
      "step": 12514,
      "time_per_iteration": 2.52724027633667
    },
    {
      "auxiliary_loss_clip": 0.01083775,
      "auxiliary_loss_mlp": 0.01028042,
      "balance_loss_clip": 1.03489125,
      "balance_loss_mlp": 1.01727104,
      "epoch": 0.7524425071396362,
      "flos": 14720610551040.0,
      "grad_norm": 1.9306911917256857,
      "language_loss": 0.69149435,
      "learning_rate": 6.090374789680271e-07,
      "loss": 0.71261251,
      "num_input_tokens_seen": 269914640,
      "step": 12515,
      "time_per_iteration": 2.526887893676758
    },
    {
      "auxiliary_loss_clip": 0.01094265,
      "auxiliary_loss_mlp": 0.01030656,
      "balance_loss_clip": 1.03618252,
      "balance_loss_mlp": 1.01940227,
      "epoch": 0.7525026303923043,
      "flos": 30592048947840.0,
      "grad_norm": 1.9783723164715672,
      "language_loss": 0.70058334,
      "learning_rate": 6.087576598969137e-07,
      "loss": 0.72183251,
      "num_input_tokens_seen": 269934960,
      "step": 12516,
      "time_per_iteration": 2.5820538997650146
    },
    {
      "auxiliary_loss_clip": 0.01057175,
      "auxiliary_loss_mlp": 0.01031096,
      "balance_loss_clip": 1.03764689,
      "balance_loss_mlp": 1.0199616,
      "epoch": 0.7525627536449722,
      "flos": 24791757765120.0,
      "grad_norm": 1.5046818591365116,
      "language_loss": 0.89729691,
      "learning_rate": 6.084778935818495e-07,
      "loss": 0.91817963,
      "num_input_tokens_seen": 269956655,
      "step": 12517,
      "time_per_iteration": 4.047922372817993
    },
    {
      "auxiliary_loss_clip": 0.01083932,
      "auxiliary_loss_mlp": 0.01033807,
      "balance_loss_clip": 1.03618479,
      "balance_loss_mlp": 1.02187979,
      "epoch": 0.7526228768976402,
      "flos": 20779782030720.0,
      "grad_norm": 1.6272729010569138,
      "language_loss": 0.74256539,
      "learning_rate": 6.081981800334437e-07,
      "loss": 0.7637428,
      "num_input_tokens_seen": 269976835,
      "step": 12518,
      "time_per_iteration": 3.9690399169921875
    },
    {
      "auxiliary_loss_clip": 0.01004286,
      "auxiliary_loss_mlp": 0.01004005,
      "balance_loss_clip": 1.02611184,
      "balance_loss_mlp": 1.00254488,
      "epoch": 0.7526830001503081,
      "flos": 66559243703040.0,
      "grad_norm": 0.7262731237667166,
      "language_loss": 0.55693161,
      "learning_rate": 6.079185192623017e-07,
      "loss": 0.57701451,
      "num_input_tokens_seen": 270040630,
      "step": 12519,
      "time_per_iteration": 3.279413938522339
    },
    {
      "auxiliary_loss_clip": 0.01087375,
      "auxiliary_loss_mlp": 0.01033509,
      "balance_loss_clip": 1.03565264,
      "balance_loss_mlp": 1.02261877,
      "epoch": 0.7527431234029761,
      "flos": 23477894087040.0,
      "grad_norm": 1.448688204103979,
      "language_loss": 0.77677774,
      "learning_rate": 6.07638911279029e-07,
      "loss": 0.79798663,
      "num_input_tokens_seen": 270059695,
      "step": 12520,
      "time_per_iteration": 2.5076589584350586
    },
    {
      "auxiliary_loss_clip": 0.01087091,
      "auxiliary_loss_mlp": 0.01036109,
      "balance_loss_clip": 1.03314614,
      "balance_loss_mlp": 1.02458107,
      "epoch": 0.752803246655644,
      "flos": 22049546785920.0,
      "grad_norm": 2.134228430347393,
      "language_loss": 0.74020475,
      "learning_rate": 6.07359356094229e-07,
      "loss": 0.76143676,
      "num_input_tokens_seen": 270078420,
      "step": 12521,
      "time_per_iteration": 2.5245373249053955
    },
    {
      "auxiliary_loss_clip": 0.01089561,
      "auxiliary_loss_mlp": 0.01034802,
      "balance_loss_clip": 1.03866649,
      "balance_loss_mlp": 1.02168345,
      "epoch": 0.752863369908312,
      "flos": 30153795108480.0,
      "grad_norm": 2.1155507287833553,
      "language_loss": 0.66979623,
      "learning_rate": 6.070798537185016e-07,
      "loss": 0.6910398,
      "num_input_tokens_seen": 270097040,
      "step": 12522,
      "time_per_iteration": 3.9793951511383057
    },
    {
      "auxiliary_loss_clip": 0.01097386,
      "auxiliary_loss_mlp": 0.01038021,
      "balance_loss_clip": 1.03719044,
      "balance_loss_mlp": 1.02598691,
      "epoch": 0.7529234931609801,
      "flos": 24567638855040.0,
      "grad_norm": 1.5635370742962917,
      "language_loss": 0.7860471,
      "learning_rate": 6.068004041624453e-07,
      "loss": 0.80740118,
      "num_input_tokens_seen": 270116365,
      "step": 12523,
      "time_per_iteration": 2.5221335887908936
    },
    {
      "auxiliary_loss_clip": 0.01103662,
      "auxiliary_loss_mlp": 0.01026722,
      "balance_loss_clip": 1.03685021,
      "balance_loss_mlp": 1.0150094,
      "epoch": 0.752983616413648,
      "flos": 23112395245440.0,
      "grad_norm": 1.8670551756808684,
      "language_loss": 0.80935454,
      "learning_rate": 6.065210074366571e-07,
      "loss": 0.83065838,
      "num_input_tokens_seen": 270135395,
      "step": 12524,
      "time_per_iteration": 2.483976125717163
    },
    {
      "auxiliary_loss_clip": 0.01094903,
      "auxiliary_loss_mlp": 0.00781325,
      "balance_loss_clip": 1.03669262,
      "balance_loss_mlp": 1.00840378,
      "epoch": 0.753043739666316,
      "flos": 24316946858880.0,
      "grad_norm": 1.5718628618359405,
      "language_loss": 0.74069113,
      "learning_rate": 6.062416635517326e-07,
      "loss": 0.75945342,
      "num_input_tokens_seen": 270156425,
      "step": 12525,
      "time_per_iteration": 2.519648790359497
    },
    {
      "auxiliary_loss_clip": 0.0107157,
      "auxiliary_loss_mlp": 0.01028598,
      "balance_loss_clip": 1.03531837,
      "balance_loss_mlp": 1.01717138,
      "epoch": 0.7531038629189839,
      "flos": 24243294021120.0,
      "grad_norm": 1.8726252235513057,
      "language_loss": 0.71784425,
      "learning_rate": 6.059623725182641e-07,
      "loss": 0.73884594,
      "num_input_tokens_seen": 270176905,
      "step": 12526,
      "time_per_iteration": 2.5989396572113037
    },
    {
      "auxiliary_loss_clip": 0.01080934,
      "auxiliary_loss_mlp": 0.01024202,
      "balance_loss_clip": 1.03348494,
      "balance_loss_mlp": 1.01316905,
      "epoch": 0.7531639861716519,
      "flos": 30188807890560.0,
      "grad_norm": 1.6636812045092084,
      "language_loss": 0.72094327,
      "learning_rate": 6.056831343468414e-07,
      "loss": 0.74199468,
      "num_input_tokens_seen": 270196640,
      "step": 12527,
      "time_per_iteration": 2.5888571739196777
    },
    {
      "auxiliary_loss_clip": 0.01069078,
      "auxiliary_loss_mlp": 0.01023696,
      "balance_loss_clip": 1.03412223,
      "balance_loss_mlp": 1.01316416,
      "epoch": 0.7532241094243198,
      "flos": 18223193560320.0,
      "grad_norm": 1.7982010468987335,
      "language_loss": 0.81197083,
      "learning_rate": 6.054039490480539e-07,
      "loss": 0.83289862,
      "num_input_tokens_seen": 270213905,
      "step": 12528,
      "time_per_iteration": 2.531092643737793
    },
    {
      "auxiliary_loss_clip": 0.01049364,
      "auxiliary_loss_mlp": 0.01032973,
      "balance_loss_clip": 1.03358698,
      "balance_loss_mlp": 1.02053392,
      "epoch": 0.7532842326769879,
      "flos": 20881049448960.0,
      "grad_norm": 1.7930087991184898,
      "language_loss": 0.8524074,
      "learning_rate": 6.051248166324892e-07,
      "loss": 0.87323081,
      "num_input_tokens_seen": 270231995,
      "step": 12529,
      "time_per_iteration": 2.596324920654297
    },
    {
      "auxiliary_loss_clip": 0.01077518,
      "auxiliary_loss_mlp": 0.01033372,
      "balance_loss_clip": 1.03750825,
      "balance_loss_mlp": 1.0209322,
      "epoch": 0.7533443559296558,
      "flos": 18078689145600.0,
      "grad_norm": 1.6657629802249088,
      "language_loss": 0.73556739,
      "learning_rate": 6.048457371107303e-07,
      "loss": 0.75667632,
      "num_input_tokens_seen": 270251480,
      "step": 12530,
      "time_per_iteration": 2.576198101043701
    },
    {
      "auxiliary_loss_clip": 0.00998683,
      "auxiliary_loss_mlp": 0.010039,
      "balance_loss_clip": 1.02384782,
      "balance_loss_mlp": 1.00237453,
      "epoch": 0.7534044791823238,
      "flos": 50254830766080.0,
      "grad_norm": 0.8250759001478801,
      "language_loss": 0.63679278,
      "learning_rate": 6.045667104933612e-07,
      "loss": 0.65681863,
      "num_input_tokens_seen": 270306480,
      "step": 12531,
      "time_per_iteration": 3.0545570850372314
    },
    {
      "auxiliary_loss_clip": 0.0108358,
      "auxiliary_loss_mlp": 0.01028914,
      "balance_loss_clip": 1.03597999,
      "balance_loss_mlp": 1.01665306,
      "epoch": 0.7534646024349917,
      "flos": 20850274471680.0,
      "grad_norm": 1.8824616198606245,
      "language_loss": 0.69817126,
      "learning_rate": 6.042877367909633e-07,
      "loss": 0.71929622,
      "num_input_tokens_seen": 270324595,
      "step": 12532,
      "time_per_iteration": 2.591109037399292
    },
    {
      "auxiliary_loss_clip": 0.01083798,
      "auxiliary_loss_mlp": 0.01027551,
      "balance_loss_clip": 1.03670406,
      "balance_loss_mlp": 1.01685739,
      "epoch": 0.7535247256876597,
      "flos": 23071779941760.0,
      "grad_norm": 1.599884934324779,
      "language_loss": 0.77381158,
      "learning_rate": 6.040088160141132e-07,
      "loss": 0.79492503,
      "num_input_tokens_seen": 270344375,
      "step": 12533,
      "time_per_iteration": 2.568091630935669
    },
    {
      "auxiliary_loss_clip": 0.01023034,
      "auxiliary_loss_mlp": 0.01007915,
      "balance_loss_clip": 1.00892282,
      "balance_loss_mlp": 1.00671053,
      "epoch": 0.7535848489403276,
      "flos": 58623418252800.0,
      "grad_norm": 0.7820394763904313,
      "language_loss": 0.57308382,
      "learning_rate": 6.037299481733886e-07,
      "loss": 0.59339333,
      "num_input_tokens_seen": 270405235,
      "step": 12534,
      "time_per_iteration": 3.141066551208496
    },
    {
      "auxiliary_loss_clip": 0.01082286,
      "auxiliary_loss_mlp": 0.01027243,
      "balance_loss_clip": 1.03443754,
      "balance_loss_mlp": 1.01547098,
      "epoch": 0.7536449721929956,
      "flos": 26577882483840.0,
      "grad_norm": 1.4859180462079278,
      "language_loss": 0.71509033,
      "learning_rate": 6.03451133279365e-07,
      "loss": 0.73618567,
      "num_input_tokens_seen": 270425820,
      "step": 12535,
      "time_per_iteration": 2.559091806411743
    },
    {
      "auxiliary_loss_clip": 0.01080593,
      "auxiliary_loss_mlp": 0.01028027,
      "balance_loss_clip": 1.03242683,
      "balance_loss_mlp": 1.01538539,
      "epoch": 0.7537050954456637,
      "flos": 25735992537600.0,
      "grad_norm": 1.5631944878000839,
      "language_loss": 0.80820239,
      "learning_rate": 6.031723713426135e-07,
      "loss": 0.8292886,
      "num_input_tokens_seen": 270447120,
      "step": 12536,
      "time_per_iteration": 2.571333885192871
    },
    {
      "auxiliary_loss_clip": 0.01078185,
      "auxiliary_loss_mlp": 0.01029911,
      "balance_loss_clip": 1.03323221,
      "balance_loss_mlp": 1.01847839,
      "epoch": 0.7537652186983316,
      "flos": 30224431203840.0,
      "grad_norm": 1.806411410896196,
      "language_loss": 0.74504244,
      "learning_rate": 6.028936623737067e-07,
      "loss": 0.76612341,
      "num_input_tokens_seen": 270468680,
      "step": 12537,
      "time_per_iteration": 2.6060614585876465
    },
    {
      "auxiliary_loss_clip": 0.0110462,
      "auxiliary_loss_mlp": 0.01032114,
      "balance_loss_clip": 1.03515291,
      "balance_loss_mlp": 1.01984143,
      "epoch": 0.7538253419509996,
      "flos": 12641239198080.0,
      "grad_norm": 1.6871873309217127,
      "language_loss": 0.74026674,
      "learning_rate": 6.026150063832111e-07,
      "loss": 0.76163411,
      "num_input_tokens_seen": 270486310,
      "step": 12538,
      "time_per_iteration": 2.466470241546631
    },
    {
      "auxiliary_loss_clip": 0.01068221,
      "auxiliary_loss_mlp": 0.01029948,
      "balance_loss_clip": 1.03620541,
      "balance_loss_mlp": 1.01786637,
      "epoch": 0.7538854652036675,
      "flos": 23185976256000.0,
      "grad_norm": 1.469429660159667,
      "language_loss": 0.67270184,
      "learning_rate": 6.023364033816956e-07,
      "loss": 0.69368351,
      "num_input_tokens_seen": 270507210,
      "step": 12539,
      "time_per_iteration": 2.5996882915496826
    },
    {
      "auxiliary_loss_clip": 0.01103253,
      "auxiliary_loss_mlp": 0.01027513,
      "balance_loss_clip": 1.03591561,
      "balance_loss_mlp": 1.01554477,
      "epoch": 0.7539455884563355,
      "flos": 23186227651200.0,
      "grad_norm": 1.5785818908289488,
      "language_loss": 0.74547398,
      "learning_rate": 6.020578533797229e-07,
      "loss": 0.76678169,
      "num_input_tokens_seen": 270525250,
      "step": 12540,
      "time_per_iteration": 2.4579107761383057
    },
    {
      "auxiliary_loss_clip": 0.01106308,
      "auxiliary_loss_mlp": 0.01029159,
      "balance_loss_clip": 1.03557646,
      "balance_loss_mlp": 1.01708865,
      "epoch": 0.7540057117090034,
      "flos": 13181155505280.0,
      "grad_norm": 2.895484024278151,
      "language_loss": 0.72901762,
      "learning_rate": 6.017793563878566e-07,
      "loss": 0.75037229,
      "num_input_tokens_seen": 270539295,
      "step": 12541,
      "time_per_iteration": 2.3987300395965576
    },
    {
      "auxiliary_loss_clip": 0.01103803,
      "auxiliary_loss_mlp": 0.01028437,
      "balance_loss_clip": 1.03627622,
      "balance_loss_mlp": 1.01649237,
      "epoch": 0.7540658349616715,
      "flos": 45478134478080.0,
      "grad_norm": 1.5354839012803267,
      "language_loss": 0.71453643,
      "learning_rate": 6.015009124166576e-07,
      "loss": 0.73585886,
      "num_input_tokens_seen": 270562815,
      "step": 12542,
      "time_per_iteration": 2.6646933555603027
    },
    {
      "auxiliary_loss_clip": 0.01078531,
      "auxiliary_loss_mlp": 0.01025069,
      "balance_loss_clip": 1.03324211,
      "balance_loss_mlp": 1.01319528,
      "epoch": 0.7541259582143394,
      "flos": 19930817105280.0,
      "grad_norm": 1.8421909229108038,
      "language_loss": 0.84473825,
      "learning_rate": 6.012225214766844e-07,
      "loss": 0.86577427,
      "num_input_tokens_seen": 270579055,
      "step": 12543,
      "time_per_iteration": 2.509392738342285
    },
    {
      "auxiliary_loss_clip": 0.01073306,
      "auxiliary_loss_mlp": 0.01031309,
      "balance_loss_clip": 1.03972697,
      "balance_loss_mlp": 1.01934648,
      "epoch": 0.7541860814670074,
      "flos": 27198239299200.0,
      "grad_norm": 2.800183314765061,
      "language_loss": 0.73245549,
      "learning_rate": 6.009441835784927e-07,
      "loss": 0.75350165,
      "num_input_tokens_seen": 270599080,
      "step": 12544,
      "time_per_iteration": 2.593473434448242
    },
    {
      "auxiliary_loss_clip": 0.0108918,
      "auxiliary_loss_mlp": 0.01032439,
      "balance_loss_clip": 1.03586507,
      "balance_loss_mlp": 1.02119708,
      "epoch": 0.7542462047196753,
      "flos": 21324151624320.0,
      "grad_norm": 2.0867028638219898,
      "language_loss": 0.68037403,
      "learning_rate": 6.006658987326383e-07,
      "loss": 0.70159018,
      "num_input_tokens_seen": 270618715,
      "step": 12545,
      "time_per_iteration": 2.5062661170959473
    },
    {
      "auxiliary_loss_clip": 0.01081374,
      "auxiliary_loss_mlp": 0.01031405,
      "balance_loss_clip": 1.03434205,
      "balance_loss_mlp": 1.01944232,
      "epoch": 0.7543063279723433,
      "flos": 11940944664960.0,
      "grad_norm": 1.6465688607053828,
      "language_loss": 0.68626565,
      "learning_rate": 6.003876669496728e-07,
      "loss": 0.70739347,
      "num_input_tokens_seen": 270635695,
      "step": 12546,
      "time_per_iteration": 2.4988033771514893
    },
    {
      "auxiliary_loss_clip": 0.01092461,
      "auxiliary_loss_mlp": 0.01033896,
      "balance_loss_clip": 1.0380646,
      "balance_loss_mlp": 1.02130127,
      "epoch": 0.7543664512250112,
      "flos": 22819974624000.0,
      "grad_norm": 6.308178284225634,
      "language_loss": 0.73575258,
      "learning_rate": 6.00109488240147e-07,
      "loss": 0.75701618,
      "num_input_tokens_seen": 270654325,
      "step": 12547,
      "time_per_iteration": 2.5134336948394775
    },
    {
      "auxiliary_loss_clip": 0.01104547,
      "auxiliary_loss_mlp": 0.01025581,
      "balance_loss_clip": 1.03601146,
      "balance_loss_mlp": 1.01328421,
      "epoch": 0.7544265744776792,
      "flos": 20923855482240.0,
      "grad_norm": 2.7950113755223205,
      "language_loss": 0.6750294,
      "learning_rate": 5.998313626146099e-07,
      "loss": 0.69633067,
      "num_input_tokens_seen": 270674260,
      "step": 12548,
      "time_per_iteration": 2.467850923538208
    },
    {
      "auxiliary_loss_clip": 0.01083624,
      "auxiliary_loss_mlp": 0.01030597,
      "balance_loss_clip": 1.03499579,
      "balance_loss_mlp": 1.01855087,
      "epoch": 0.7544866977303473,
      "flos": 15195493284480.0,
      "grad_norm": 1.8227839133353618,
      "language_loss": 0.87292594,
      "learning_rate": 5.995532900836088e-07,
      "loss": 0.89406812,
      "num_input_tokens_seen": 270692200,
      "step": 12549,
      "time_per_iteration": 2.5147600173950195
    },
    {
      "auxiliary_loss_clip": 0.01056661,
      "auxiliary_loss_mlp": 0.01034582,
      "balance_loss_clip": 1.03579426,
      "balance_loss_mlp": 1.02328074,
      "epoch": 0.7545468209830152,
      "flos": 27083683848960.0,
      "grad_norm": 1.7319896565436874,
      "language_loss": 0.77084649,
      "learning_rate": 5.992752706576865e-07,
      "loss": 0.79175889,
      "num_input_tokens_seen": 270709675,
      "step": 12550,
      "time_per_iteration": 4.002254486083984
    },
    {
      "auxiliary_loss_clip": 0.01104924,
      "auxiliary_loss_mlp": 0.01025555,
      "balance_loss_clip": 1.03534603,
      "balance_loss_mlp": 1.01381302,
      "epoch": 0.7546069442356832,
      "flos": 26871703735680.0,
      "grad_norm": 1.426147114766673,
      "language_loss": 0.69329214,
      "learning_rate": 5.98997304347386e-07,
      "loss": 0.71459699,
      "num_input_tokens_seen": 270733055,
      "step": 12551,
      "time_per_iteration": 2.522958278656006
    },
    {
      "auxiliary_loss_clip": 0.01082743,
      "auxiliary_loss_mlp": 0.01027337,
      "balance_loss_clip": 1.03661084,
      "balance_loss_mlp": 1.01498079,
      "epoch": 0.7546670674883511,
      "flos": 15743131015680.0,
      "grad_norm": 2.604516278348822,
      "language_loss": 0.86308199,
      "learning_rate": 5.987193911632487e-07,
      "loss": 0.88418275,
      "num_input_tokens_seen": 270749275,
      "step": 12552,
      "time_per_iteration": 2.49202036857605
    },
    {
      "auxiliary_loss_clip": 0.01095472,
      "auxiliary_loss_mlp": 0.01032312,
      "balance_loss_clip": 1.03558123,
      "balance_loss_mlp": 1.0205338,
      "epoch": 0.7547271907410191,
      "flos": 23477714519040.0,
      "grad_norm": 1.8137954048237637,
      "language_loss": 0.77973974,
      "learning_rate": 5.98441531115812e-07,
      "loss": 0.80101752,
      "num_input_tokens_seen": 270768230,
      "step": 12553,
      "time_per_iteration": 2.511488437652588
    },
    {
      "auxiliary_loss_clip": 0.01095194,
      "auxiliary_loss_mlp": 0.01034825,
      "balance_loss_clip": 1.03808558,
      "balance_loss_mlp": 1.02274966,
      "epoch": 0.754787313993687,
      "flos": 31722804069120.0,
      "grad_norm": 1.919775563073745,
      "language_loss": 0.62590015,
      "learning_rate": 5.981637242156135e-07,
      "loss": 0.64720035,
      "num_input_tokens_seen": 270786285,
      "step": 12554,
      "time_per_iteration": 2.5777547359466553
    },
    {
      "auxiliary_loss_clip": 0.01082712,
      "auxiliary_loss_mlp": 0.01036676,
      "balance_loss_clip": 1.03402054,
      "balance_loss_mlp": 1.02493989,
      "epoch": 0.7548474372463551,
      "flos": 27563055782400.0,
      "grad_norm": 1.5416249450200699,
      "language_loss": 0.7346406,
      "learning_rate": 5.978859704731864e-07,
      "loss": 0.75583446,
      "num_input_tokens_seen": 270805505,
      "step": 12555,
      "time_per_iteration": 3.9438347816467285
    },
    {
      "auxiliary_loss_clip": 0.0108072,
      "auxiliary_loss_mlp": 0.01026518,
      "balance_loss_clip": 1.03783202,
      "balance_loss_mlp": 1.01444221,
      "epoch": 0.754907560499023,
      "flos": 19318576763520.0,
      "grad_norm": 1.916613473330727,
      "language_loss": 0.78306162,
      "learning_rate": 5.976082698990645e-07,
      "loss": 0.80413401,
      "num_input_tokens_seen": 270824610,
      "step": 12556,
      "time_per_iteration": 2.562962055206299
    },
    {
      "auxiliary_loss_clip": 0.0102206,
      "auxiliary_loss_mlp": 0.01003406,
      "balance_loss_clip": 1.00925016,
      "balance_loss_mlp": 1.0023272,
      "epoch": 0.754967683751691,
      "flos": 69744628684800.0,
      "grad_norm": 0.7081316104057256,
      "language_loss": 0.50447935,
      "learning_rate": 5.973306225037769e-07,
      "loss": 0.52473402,
      "num_input_tokens_seen": 270886155,
      "step": 12557,
      "time_per_iteration": 4.5392725467681885
    },
    {
      "auxiliary_loss_clip": 0.0109477,
      "auxiliary_loss_mlp": 0.01029487,
      "balance_loss_clip": 1.03975236,
      "balance_loss_mlp": 1.01753044,
      "epoch": 0.7550278070043589,
      "flos": 24421913377920.0,
      "grad_norm": 1.717825602062205,
      "language_loss": 0.7165103,
      "learning_rate": 5.970530282978525e-07,
      "loss": 0.73775291,
      "num_input_tokens_seen": 270905325,
      "step": 12558,
      "time_per_iteration": 2.5575428009033203
    },
    {
      "auxiliary_loss_clip": 0.01080176,
      "auxiliary_loss_mlp": 0.01036081,
      "balance_loss_clip": 1.03380513,
      "balance_loss_mlp": 1.02319479,
      "epoch": 0.7550879302570269,
      "flos": 32634611838720.0,
      "grad_norm": 1.7613666083843844,
      "language_loss": 0.80310762,
      "learning_rate": 5.967754872918187e-07,
      "loss": 0.82427019,
      "num_input_tokens_seen": 270927535,
      "step": 12559,
      "time_per_iteration": 2.625265121459961
    },
    {
      "auxiliary_loss_clip": 0.01058017,
      "auxiliary_loss_mlp": 0.01028381,
      "balance_loss_clip": 1.03590155,
      "balance_loss_mlp": 1.01622212,
      "epoch": 0.7551480535096948,
      "flos": 21795550738560.0,
      "grad_norm": 1.608551282811912,
      "language_loss": 0.78935665,
      "learning_rate": 5.96497999496199e-07,
      "loss": 0.8102206,
      "num_input_tokens_seen": 270946920,
      "step": 12560,
      "time_per_iteration": 2.6259067058563232
    },
    {
      "auxiliary_loss_clip": 0.01054143,
      "auxiliary_loss_mlp": 0.01040505,
      "balance_loss_clip": 1.03449297,
      "balance_loss_mlp": 1.02758253,
      "epoch": 0.7552081767623628,
      "flos": 18515111391360.0,
      "grad_norm": 1.6787559628144801,
      "language_loss": 0.70545614,
      "learning_rate": 5.96220564921515e-07,
      "loss": 0.72640258,
      "num_input_tokens_seen": 270965705,
      "step": 12561,
      "time_per_iteration": 2.5699119567871094
    },
    {
      "auxiliary_loss_clip": 0.01077531,
      "auxiliary_loss_mlp": 0.00786037,
      "balance_loss_clip": 1.03228879,
      "balance_loss_mlp": 1.00926745,
      "epoch": 0.7552683000150308,
      "flos": 27634805199360.0,
      "grad_norm": 1.5519131804653503,
      "language_loss": 0.75704336,
      "learning_rate": 5.959431835782889e-07,
      "loss": 0.77567905,
      "num_input_tokens_seen": 270986550,
      "step": 12562,
      "time_per_iteration": 4.006548643112183
    },
    {
      "auxiliary_loss_clip": 0.01078907,
      "auxiliary_loss_mlp": 0.01028969,
      "balance_loss_clip": 1.03495061,
      "balance_loss_mlp": 1.01657152,
      "epoch": 0.7553284232676988,
      "flos": 20302924049280.0,
      "grad_norm": 2.0846556178086386,
      "language_loss": 0.75936139,
      "learning_rate": 5.956658554770371e-07,
      "loss": 0.78044015,
      "num_input_tokens_seen": 271006250,
      "step": 12563,
      "time_per_iteration": 2.531672239303589
    },
    {
      "auxiliary_loss_clip": 0.0107612,
      "auxiliary_loss_mlp": 0.01034072,
      "balance_loss_clip": 1.03459466,
      "balance_loss_mlp": 1.01937306,
      "epoch": 0.7553885465203668,
      "flos": 33255471444480.0,
      "grad_norm": 2.212510371592076,
      "language_loss": 0.67313796,
      "learning_rate": 5.953885806282768e-07,
      "loss": 0.69423985,
      "num_input_tokens_seen": 271025575,
      "step": 12564,
      "time_per_iteration": 2.6571788787841797
    },
    {
      "auxiliary_loss_clip": 0.01078216,
      "auxiliary_loss_mlp": 0.01035662,
      "balance_loss_clip": 1.03646719,
      "balance_loss_mlp": 1.02270389,
      "epoch": 0.7554486697730347,
      "flos": 21616249023360.0,
      "grad_norm": 1.9247829908605825,
      "language_loss": 0.68673909,
      "learning_rate": 5.951113590425228e-07,
      "loss": 0.70787787,
      "num_input_tokens_seen": 271045805,
      "step": 12565,
      "time_per_iteration": 2.5617005825042725
    },
    {
      "auxiliary_loss_clip": 0.01085141,
      "auxiliary_loss_mlp": 0.01029207,
      "balance_loss_clip": 1.03433943,
      "balance_loss_mlp": 1.01629639,
      "epoch": 0.7555087930257027,
      "flos": 27632973605760.0,
      "grad_norm": 2.5598997788851743,
      "language_loss": 0.74914777,
      "learning_rate": 5.94834190730287e-07,
      "loss": 0.77029127,
      "num_input_tokens_seen": 271066065,
      "step": 12566,
      "time_per_iteration": 2.585289716720581
    },
    {
      "auxiliary_loss_clip": 0.01099934,
      "auxiliary_loss_mlp": 0.01034406,
      "balance_loss_clip": 1.03795004,
      "balance_loss_mlp": 1.02060783,
      "epoch": 0.7555689162783706,
      "flos": 23621644316160.0,
      "grad_norm": 1.9553463244571208,
      "language_loss": 0.73576194,
      "learning_rate": 5.945570757020789e-07,
      "loss": 0.75710535,
      "num_input_tokens_seen": 271085870,
      "step": 12567,
      "time_per_iteration": 2.5140371322631836
    },
    {
      "auxiliary_loss_clip": 0.01103674,
      "auxiliary_loss_mlp": 0.01028033,
      "balance_loss_clip": 1.03526151,
      "balance_loss_mlp": 1.01642847,
      "epoch": 0.7556290395310387,
      "flos": 24863076218880.0,
      "grad_norm": 1.9650130063807358,
      "language_loss": 0.62676853,
      "learning_rate": 5.942800139684073e-07,
      "loss": 0.64808559,
      "num_input_tokens_seen": 271104260,
      "step": 12568,
      "time_per_iteration": 2.493835210800171
    },
    {
      "auxiliary_loss_clip": 0.01019588,
      "auxiliary_loss_mlp": 0.01037245,
      "balance_loss_clip": 1.03341794,
      "balance_loss_mlp": 1.02446628,
      "epoch": 0.7556891627837066,
      "flos": 43543770330240.0,
      "grad_norm": 1.8441000260839275,
      "language_loss": 0.66555429,
      "learning_rate": 5.940030055397789e-07,
      "loss": 0.68612266,
      "num_input_tokens_seen": 271125745,
      "step": 12569,
      "time_per_iteration": 2.93245792388916
    },
    {
      "auxiliary_loss_clip": 0.01099563,
      "auxiliary_loss_mlp": 0.01036004,
      "balance_loss_clip": 1.03866887,
      "balance_loss_mlp": 1.02230644,
      "epoch": 0.7557492860363746,
      "flos": 26650924790400.0,
      "grad_norm": 1.6759615480978152,
      "language_loss": 0.67455512,
      "learning_rate": 5.93726050426697e-07,
      "loss": 0.69591075,
      "num_input_tokens_seen": 271147145,
      "step": 12570,
      "time_per_iteration": 2.66404390335083
    },
    {
      "auxiliary_loss_clip": 0.01105573,
      "auxiliary_loss_mlp": 0.01030404,
      "balance_loss_clip": 1.03657269,
      "balance_loss_mlp": 1.01815557,
      "epoch": 0.7558094092890425,
      "flos": 55182885010560.0,
      "grad_norm": 2.059497901712733,
      "language_loss": 0.71991998,
      "learning_rate": 5.934491486396647e-07,
      "loss": 0.74127972,
      "num_input_tokens_seen": 271170865,
      "step": 12571,
      "time_per_iteration": 2.7414238452911377
    },
    {
      "auxiliary_loss_clip": 0.01061355,
      "auxiliary_loss_mlp": 0.01035207,
      "balance_loss_clip": 1.03534317,
      "balance_loss_mlp": 1.02158689,
      "epoch": 0.7558695325417105,
      "flos": 23988292392960.0,
      "grad_norm": 1.707328548729647,
      "language_loss": 0.73390609,
      "learning_rate": 5.931723001891811e-07,
      "loss": 0.75487173,
      "num_input_tokens_seen": 271191450,
      "step": 12572,
      "time_per_iteration": 2.656403064727783
    },
    {
      "auxiliary_loss_clip": 0.01086869,
      "auxiliary_loss_mlp": 0.01029915,
      "balance_loss_clip": 1.03828573,
      "balance_loss_mlp": 1.0181849,
      "epoch": 0.7559296557943784,
      "flos": 14611262572800.0,
      "grad_norm": 2.00441253244529,
      "language_loss": 0.76875353,
      "learning_rate": 5.928955050857456e-07,
      "loss": 0.78992134,
      "num_input_tokens_seen": 271207335,
      "step": 12573,
      "time_per_iteration": 2.507458209991455
    },
    {
      "auxiliary_loss_clip": 0.01080076,
      "auxiliary_loss_mlp": 0.01029996,
      "balance_loss_clip": 1.03691816,
      "balance_loss_mlp": 1.01781869,
      "epoch": 0.7559897790470465,
      "flos": 18550483309440.0,
      "grad_norm": 1.6240277338987166,
      "language_loss": 0.69233268,
      "learning_rate": 5.926187633398527e-07,
      "loss": 0.71343338,
      "num_input_tokens_seen": 271226895,
      "step": 12574,
      "time_per_iteration": 2.5767080783843994
    },
    {
      "auxiliary_loss_clip": 0.01066926,
      "auxiliary_loss_mlp": 0.01037441,
      "balance_loss_clip": 1.03067338,
      "balance_loss_mlp": 1.02452469,
      "epoch": 0.7560499022997144,
      "flos": 17967868709760.0,
      "grad_norm": 2.1120611812689547,
      "language_loss": 0.71469426,
      "learning_rate": 5.923420749619974e-07,
      "loss": 0.73573792,
      "num_input_tokens_seen": 271244375,
      "step": 12575,
      "time_per_iteration": 2.5312037467956543
    },
    {
      "auxiliary_loss_clip": 0.01102521,
      "auxiliary_loss_mlp": 0.00783355,
      "balance_loss_clip": 1.03454792,
      "balance_loss_mlp": 1.00970984,
      "epoch": 0.7561100255523824,
      "flos": 15737815802880.0,
      "grad_norm": 2.1288585560103024,
      "language_loss": 0.72136497,
      "learning_rate": 5.92065439962673e-07,
      "loss": 0.74022377,
      "num_input_tokens_seen": 271259530,
      "step": 12576,
      "time_per_iteration": 2.4883334636688232
    },
    {
      "auxiliary_loss_clip": 0.01069856,
      "auxiliary_loss_mlp": 0.0103067,
      "balance_loss_clip": 1.03532112,
      "balance_loss_mlp": 1.01868391,
      "epoch": 0.7561701488050504,
      "flos": 15888102307200.0,
      "grad_norm": 1.7928768681560392,
      "language_loss": 0.67228949,
      "learning_rate": 5.917888583523669e-07,
      "loss": 0.69329476,
      "num_input_tokens_seen": 271276835,
      "step": 12577,
      "time_per_iteration": 2.542865753173828
    },
    {
      "auxiliary_loss_clip": 0.01082363,
      "auxiliary_loss_mlp": 0.01034448,
      "balance_loss_clip": 1.03500628,
      "balance_loss_mlp": 1.02246106,
      "epoch": 0.7562302720577183,
      "flos": 20339157893760.0,
      "grad_norm": 1.7313226915970445,
      "language_loss": 0.78307045,
      "learning_rate": 5.915123301415685e-07,
      "loss": 0.80423856,
      "num_input_tokens_seen": 271296275,
      "step": 12578,
      "time_per_iteration": 2.518136739730835
    },
    {
      "auxiliary_loss_clip": 0.01094503,
      "auxiliary_loss_mlp": 0.01033351,
      "balance_loss_clip": 1.03544104,
      "balance_loss_mlp": 1.02135301,
      "epoch": 0.7562903953103863,
      "flos": 20812209033600.0,
      "grad_norm": 1.5107610521350894,
      "language_loss": 0.75935221,
      "learning_rate": 5.912358553407641e-07,
      "loss": 0.78063071,
      "num_input_tokens_seen": 271315685,
      "step": 12579,
      "time_per_iteration": 2.527510643005371
    },
    {
      "auxiliary_loss_clip": 0.01064318,
      "auxiliary_loss_mlp": 0.0103221,
      "balance_loss_clip": 1.03724074,
      "balance_loss_mlp": 1.01888227,
      "epoch": 0.7563505185630542,
      "flos": 37596999484800.0,
      "grad_norm": 2.0819873667994235,
      "language_loss": 0.62743121,
      "learning_rate": 5.90959433960437e-07,
      "loss": 0.64839649,
      "num_input_tokens_seen": 271336790,
      "step": 12580,
      "time_per_iteration": 2.731679677963257
    },
    {
      "auxiliary_loss_clip": 0.01060936,
      "auxiliary_loss_mlp": 0.01028894,
      "balance_loss_clip": 1.03377247,
      "balance_loss_mlp": 1.01682436,
      "epoch": 0.7564106418157223,
      "flos": 20230995064320.0,
      "grad_norm": 1.783254124912878,
      "language_loss": 0.74953419,
      "learning_rate": 5.906830660110691e-07,
      "loss": 0.77043247,
      "num_input_tokens_seen": 271355470,
      "step": 12581,
      "time_per_iteration": 2.6006104946136475
    },
    {
      "auxiliary_loss_clip": 0.01065173,
      "auxiliary_loss_mlp": 0.01030354,
      "balance_loss_clip": 1.03660464,
      "balance_loss_mlp": 1.01802754,
      "epoch": 0.7564707650683902,
      "flos": 24754877475840.0,
      "grad_norm": 1.8200394118983614,
      "language_loss": 0.62609512,
      "learning_rate": 5.904067515031412e-07,
      "loss": 0.64705038,
      "num_input_tokens_seen": 271375810,
      "step": 12582,
      "time_per_iteration": 2.6009130477905273
    },
    {
      "auxiliary_loss_clip": 0.0103191,
      "auxiliary_loss_mlp": 0.01005695,
      "balance_loss_clip": 1.00866318,
      "balance_loss_mlp": 1.00454509,
      "epoch": 0.7565308883210582,
      "flos": 48530076433920.0,
      "grad_norm": 0.9515085202372157,
      "language_loss": 0.60717702,
      "learning_rate": 5.901304904471307e-07,
      "loss": 0.62755311,
      "num_input_tokens_seen": 271424775,
      "step": 12583,
      "time_per_iteration": 2.8540520668029785
    },
    {
      "auxiliary_loss_clip": 0.01078866,
      "auxiliary_loss_mlp": 0.01031519,
      "balance_loss_clip": 1.03818893,
      "balance_loss_mlp": 1.01901364,
      "epoch": 0.7565910115737261,
      "flos": 12495082757760.0,
      "grad_norm": 1.8821615709627735,
      "language_loss": 0.79006904,
      "learning_rate": 5.898542828535125e-07,
      "loss": 0.81117284,
      "num_input_tokens_seen": 271440500,
      "step": 12584,
      "time_per_iteration": 2.5076582431793213
    },
    {
      "auxiliary_loss_clip": 0.01075,
      "auxiliary_loss_mlp": 0.01041526,
      "balance_loss_clip": 1.03339314,
      "balance_loss_mlp": 1.02708995,
      "epoch": 0.7566511348263941,
      "flos": 21173003193600.0,
      "grad_norm": 1.9247729836367704,
      "language_loss": 0.77520859,
      "learning_rate": 5.895781287327612e-07,
      "loss": 0.79637384,
      "num_input_tokens_seen": 271458180,
      "step": 12585,
      "time_per_iteration": 2.5434367656707764
    },
    {
      "auxiliary_loss_clip": 0.01110213,
      "auxiliary_loss_mlp": 0.01033819,
      "balance_loss_clip": 1.0390811,
      "balance_loss_mlp": 1.02105761,
      "epoch": 0.756711258079062,
      "flos": 21754827694080.0,
      "grad_norm": 1.921569916424512,
      "language_loss": 0.8295145,
      "learning_rate": 5.893020280953493e-07,
      "loss": 0.85095489,
      "num_input_tokens_seen": 271475730,
      "step": 12586,
      "time_per_iteration": 2.4690253734588623
    },
    {
      "auxiliary_loss_clip": 0.01108811,
      "auxiliary_loss_mlp": 0.01030989,
      "balance_loss_clip": 1.03712201,
      "balance_loss_mlp": 1.0193243,
      "epoch": 0.75677138133173,
      "flos": 22382905933440.0,
      "grad_norm": 2.235983633684025,
      "language_loss": 0.83533657,
      "learning_rate": 5.890259809517459e-07,
      "loss": 0.85673457,
      "num_input_tokens_seen": 271495030,
      "step": 12587,
      "time_per_iteration": 2.4896111488342285
    },
    {
      "auxiliary_loss_clip": 0.01071148,
      "auxiliary_loss_mlp": 0.01027418,
      "balance_loss_clip": 1.03472567,
      "balance_loss_mlp": 1.01528859,
      "epoch": 0.756831504584398,
      "flos": 22708974620160.0,
      "grad_norm": 1.5910895723602476,
      "language_loss": 0.71358538,
      "learning_rate": 5.88749987312418e-07,
      "loss": 0.7345711,
      "num_input_tokens_seen": 271515355,
      "step": 12588,
      "time_per_iteration": 2.571631908416748
    },
    {
      "auxiliary_loss_clip": 0.01108338,
      "auxiliary_loss_mlp": 0.00782705,
      "balance_loss_clip": 1.0375483,
      "balance_loss_mlp": 1.00722373,
      "epoch": 0.756891627837066,
      "flos": 24098358643200.0,
      "grad_norm": 1.9239347573246157,
      "language_loss": 0.69060731,
      "learning_rate": 5.884740471878327e-07,
      "loss": 0.70951772,
      "num_input_tokens_seen": 271535090,
      "step": 12589,
      "time_per_iteration": 3.940742254257202
    },
    {
      "auxiliary_loss_clip": 0.0109379,
      "auxiliary_loss_mlp": 0.01027462,
      "balance_loss_clip": 1.03518534,
      "balance_loss_mlp": 1.01538062,
      "epoch": 0.756951751089734,
      "flos": 19749001438080.0,
      "grad_norm": 1.6486989084072432,
      "language_loss": 0.92053962,
      "learning_rate": 5.881981605884522e-07,
      "loss": 0.9417522,
      "num_input_tokens_seen": 271551075,
      "step": 12590,
      "time_per_iteration": 2.4786465167999268
    },
    {
      "auxiliary_loss_clip": 0.0107731,
      "auxiliary_loss_mlp": 0.01029098,
      "balance_loss_clip": 1.03429914,
      "balance_loss_mlp": 1.01674247,
      "epoch": 0.7570118743424019,
      "flos": 35079266551680.0,
      "grad_norm": 1.77528228200888,
      "language_loss": 0.65549266,
      "learning_rate": 5.879223275247391e-07,
      "loss": 0.67655677,
      "num_input_tokens_seen": 271571035,
      "step": 12591,
      "time_per_iteration": 2.6804704666137695
    },
    {
      "auxiliary_loss_clip": 0.01094027,
      "auxiliary_loss_mlp": 0.01022957,
      "balance_loss_clip": 1.03689861,
      "balance_loss_mlp": 1.01192403,
      "epoch": 0.7570719975950699,
      "flos": 25594540778880.0,
      "grad_norm": 2.121121914421505,
      "language_loss": 0.73631394,
      "learning_rate": 5.876465480071528e-07,
      "loss": 0.75748378,
      "num_input_tokens_seen": 271592950,
      "step": 12592,
      "time_per_iteration": 2.5370850563049316
    },
    {
      "auxiliary_loss_clip": 0.01095899,
      "auxiliary_loss_mlp": 0.01034458,
      "balance_loss_clip": 1.03613615,
      "balance_loss_mlp": 1.0215652,
      "epoch": 0.7571321208477378,
      "flos": 10816223028480.0,
      "grad_norm": 2.3114374577424996,
      "language_loss": 0.71476865,
      "learning_rate": 5.873708220461522e-07,
      "loss": 0.73607218,
      "num_input_tokens_seen": 271608835,
      "step": 12593,
      "time_per_iteration": 2.4616951942443848
    },
    {
      "auxiliary_loss_clip": 0.01105912,
      "auxiliary_loss_mlp": 0.01029827,
      "balance_loss_clip": 1.03548431,
      "balance_loss_mlp": 1.01778698,
      "epoch": 0.7571922441004059,
      "flos": 18260109763200.0,
      "grad_norm": 2.584148050491652,
      "language_loss": 0.66250694,
      "learning_rate": 5.870951496521903e-07,
      "loss": 0.68386436,
      "num_input_tokens_seen": 271627730,
      "step": 12594,
      "time_per_iteration": 3.82405948638916
    },
    {
      "auxiliary_loss_clip": 0.01075886,
      "auxiliary_loss_mlp": 0.01030794,
      "balance_loss_clip": 1.03693271,
      "balance_loss_mlp": 1.01895595,
      "epoch": 0.7572523673530738,
      "flos": 22890502978560.0,
      "grad_norm": 1.7144933778907412,
      "language_loss": 0.80477899,
      "learning_rate": 5.86819530835722e-07,
      "loss": 0.82584572,
      "num_input_tokens_seen": 271646415,
      "step": 12595,
      "time_per_iteration": 3.924114227294922
    },
    {
      "auxiliary_loss_clip": 0.01072284,
      "auxiliary_loss_mlp": 0.01031603,
      "balance_loss_clip": 1.03576863,
      "balance_loss_mlp": 1.01985455,
      "epoch": 0.7573124906057418,
      "flos": 20996323171200.0,
      "grad_norm": 1.9461362085533571,
      "language_loss": 0.71524084,
      "learning_rate": 5.865439656071993e-07,
      "loss": 0.73627973,
      "num_input_tokens_seen": 271666240,
      "step": 12596,
      "time_per_iteration": 2.5494489669799805
    },
    {
      "auxiliary_loss_clip": 0.01016757,
      "auxiliary_loss_mlp": 0.01029505,
      "balance_loss_clip": 1.03622484,
      "balance_loss_mlp": 1.01790035,
      "epoch": 0.7573726138584097,
      "flos": 20886292834560.0,
      "grad_norm": 1.4871680137227337,
      "language_loss": 0.80514908,
      "learning_rate": 5.862684539770706e-07,
      "loss": 0.82561171,
      "num_input_tokens_seen": 271686370,
      "step": 12597,
      "time_per_iteration": 2.881378650665283
    },
    {
      "auxiliary_loss_clip": 0.01078068,
      "auxiliary_loss_mlp": 0.01030665,
      "balance_loss_clip": 1.03724849,
      "balance_loss_mlp": 1.01749206,
      "epoch": 0.7574327371110777,
      "flos": 24530507170560.0,
      "grad_norm": 2.2285457079882334,
      "language_loss": 0.83117652,
      "learning_rate": 5.859929959557835e-07,
      "loss": 0.85226387,
      "num_input_tokens_seen": 271705050,
      "step": 12598,
      "time_per_iteration": 2.8816792964935303
    },
    {
      "auxiliary_loss_clip": 0.01077399,
      "auxiliary_loss_mlp": 0.01024905,
      "balance_loss_clip": 1.035725,
      "balance_loss_mlp": 1.01395583,
      "epoch": 0.7574928603637456,
      "flos": 23364523785600.0,
      "grad_norm": 1.6962601443753038,
      "language_loss": 0.62943125,
      "learning_rate": 5.857175915537845e-07,
      "loss": 0.65045428,
      "num_input_tokens_seen": 271724915,
      "step": 12599,
      "time_per_iteration": 2.547057867050171
    },
    {
      "auxiliary_loss_clip": 0.01088986,
      "auxiliary_loss_mlp": 0.00786492,
      "balance_loss_clip": 1.03682041,
      "balance_loss_mlp": 1.01319206,
      "epoch": 0.7575529836164137,
      "flos": 13516274419200.0,
      "grad_norm": 2.7796289012513493,
      "language_loss": 0.63313484,
      "learning_rate": 5.854422407815161e-07,
      "loss": 0.65188968,
      "num_input_tokens_seen": 271742410,
      "step": 12600,
      "time_per_iteration": 4.071700572967529
    },
    {
      "auxiliary_loss_clip": 0.01077064,
      "auxiliary_loss_mlp": 0.01031884,
      "balance_loss_clip": 1.03379154,
      "balance_loss_mlp": 1.01876462,
      "epoch": 0.7576131068690816,
      "flos": 19646584784640.0,
      "grad_norm": 1.68781142331712,
      "language_loss": 0.66405272,
      "learning_rate": 5.851669436494191e-07,
      "loss": 0.68514222,
      "num_input_tokens_seen": 271761425,
      "step": 12601,
      "time_per_iteration": 2.6218433380126953
    },
    {
      "auxiliary_loss_clip": 0.01083296,
      "auxiliary_loss_mlp": 0.01027784,
      "balance_loss_clip": 1.03692317,
      "balance_loss_mlp": 1.01666141,
      "epoch": 0.7576732301217496,
      "flos": 20048245643520.0,
      "grad_norm": 1.6599385668428057,
      "language_loss": 0.68031013,
      "learning_rate": 5.848917001679335e-07,
      "loss": 0.70142096,
      "num_input_tokens_seen": 271780875,
      "step": 12602,
      "time_per_iteration": 2.5062079429626465
    },
    {
      "auxiliary_loss_clip": 0.01097119,
      "auxiliary_loss_mlp": 0.01031459,
      "balance_loss_clip": 1.03715038,
      "balance_loss_mlp": 1.01835775,
      "epoch": 0.7577333533744176,
      "flos": 15377093470080.0,
      "grad_norm": 1.8516900415586521,
      "language_loss": 0.67109895,
      "learning_rate": 5.846165103474967e-07,
      "loss": 0.69238478,
      "num_input_tokens_seen": 271799490,
      "step": 12603,
      "time_per_iteration": 2.4768238067626953
    },
    {
      "auxiliary_loss_clip": 0.01080362,
      "auxiliary_loss_mlp": 0.01028668,
      "balance_loss_clip": 1.03312159,
      "balance_loss_mlp": 1.01745057,
      "epoch": 0.7577934766270855,
      "flos": 17894862316800.0,
      "grad_norm": 1.8641923021586533,
      "language_loss": 0.61696744,
      "learning_rate": 5.843413741985439e-07,
      "loss": 0.63805771,
      "num_input_tokens_seen": 271817040,
      "step": 12604,
      "time_per_iteration": 2.4764058589935303
    },
    {
      "auxiliary_loss_clip": 0.01106686,
      "auxiliary_loss_mlp": 0.0103426,
      "balance_loss_clip": 1.03893471,
      "balance_loss_mlp": 1.02168298,
      "epoch": 0.7578535998797535,
      "flos": 21613770984960.0,
      "grad_norm": 1.7453052938972013,
      "language_loss": 0.79959059,
      "learning_rate": 5.840662917315076e-07,
      "loss": 0.82100004,
      "num_input_tokens_seen": 271835480,
      "step": 12605,
      "time_per_iteration": 2.4607107639312744
    },
    {
      "auxiliary_loss_clip": 0.01108513,
      "auxiliary_loss_mlp": 0.01029739,
      "balance_loss_clip": 1.03658283,
      "balance_loss_mlp": 1.01736474,
      "epoch": 0.7579137231324214,
      "flos": 18478374756480.0,
      "grad_norm": 2.6829601673299703,
      "language_loss": 0.79759097,
      "learning_rate": 5.837912629568198e-07,
      "loss": 0.81897342,
      "num_input_tokens_seen": 271849835,
      "step": 12606,
      "time_per_iteration": 2.4144935607910156
    },
    {
      "auxiliary_loss_clip": 0.01088517,
      "auxiliary_loss_mlp": 0.01026193,
      "balance_loss_clip": 1.03510714,
      "balance_loss_mlp": 1.01567864,
      "epoch": 0.7579738463850895,
      "flos": 23255032152960.0,
      "grad_norm": 1.530363428783069,
      "language_loss": 0.72812986,
      "learning_rate": 5.835162878849087e-07,
      "loss": 0.74927694,
      "num_input_tokens_seen": 271869560,
      "step": 12607,
      "time_per_iteration": 2.5196456909179688
    },
    {
      "auxiliary_loss_clip": 0.01078999,
      "auxiliary_loss_mlp": 0.01028471,
      "balance_loss_clip": 1.03577757,
      "balance_loss_mlp": 1.01623976,
      "epoch": 0.7580339696377574,
      "flos": 14027031861120.0,
      "grad_norm": 2.011897558219424,
      "language_loss": 0.74909025,
      "learning_rate": 5.83241366526202e-07,
      "loss": 0.77016497,
      "num_input_tokens_seen": 271887950,
      "step": 12608,
      "time_per_iteration": 2.490443706512451
    },
    {
      "auxiliary_loss_clip": 0.01070775,
      "auxiliary_loss_mlp": 0.00783847,
      "balance_loss_clip": 1.0359292,
      "balance_loss_mlp": 1.01072693,
      "epoch": 0.7580940928904254,
      "flos": 25082777756160.0,
      "grad_norm": 1.5202469383561665,
      "language_loss": 0.71251744,
      "learning_rate": 5.829664988911245e-07,
      "loss": 0.73106366,
      "num_input_tokens_seen": 271907700,
      "step": 12609,
      "time_per_iteration": 2.6089119911193848
    },
    {
      "auxiliary_loss_clip": 0.01106803,
      "auxiliary_loss_mlp": 0.01030581,
      "balance_loss_clip": 1.0359683,
      "balance_loss_mlp": 1.01733649,
      "epoch": 0.7581542161430933,
      "flos": 23836425690240.0,
      "grad_norm": 1.693848724515262,
      "language_loss": 0.81674874,
      "learning_rate": 5.826916849901007e-07,
      "loss": 0.83812261,
      "num_input_tokens_seen": 271926840,
      "step": 12610,
      "time_per_iteration": 2.457559823989868
    },
    {
      "auxiliary_loss_clip": 0.01089723,
      "auxiliary_loss_mlp": 0.01029443,
      "balance_loss_clip": 1.0369916,
      "balance_loss_mlp": 1.01683116,
      "epoch": 0.7582143393957613,
      "flos": 22237000888320.0,
      "grad_norm": 1.7556565398519286,
      "language_loss": 0.70132506,
      "learning_rate": 5.824169248335488e-07,
      "loss": 0.72251678,
      "num_input_tokens_seen": 271946465,
      "step": 12611,
      "time_per_iteration": 2.544071674346924
    },
    {
      "auxiliary_loss_clip": 0.0110591,
      "auxiliary_loss_mlp": 0.01028494,
      "balance_loss_clip": 1.03743434,
      "balance_loss_mlp": 1.01623964,
      "epoch": 0.7582744626484292,
      "flos": 21106389421440.0,
      "grad_norm": 1.4905088915954563,
      "language_loss": 0.71204299,
      "learning_rate": 5.821422184318893e-07,
      "loss": 0.73338699,
      "num_input_tokens_seen": 271967295,
      "step": 12612,
      "time_per_iteration": 2.4544012546539307
    },
    {
      "auxiliary_loss_clip": 0.0104695,
      "auxiliary_loss_mlp": 0.01038804,
      "balance_loss_clip": 1.0346837,
      "balance_loss_mlp": 1.0268358,
      "epoch": 0.7583345859010973,
      "flos": 24604770539520.0,
      "grad_norm": 2.7726650874562218,
      "language_loss": 0.59545356,
      "learning_rate": 5.818675657955397e-07,
      "loss": 0.61631107,
      "num_input_tokens_seen": 271987960,
      "step": 12613,
      "time_per_iteration": 2.660717487335205
    },
    {
      "auxiliary_loss_clip": 0.01080379,
      "auxiliary_loss_mlp": 0.01039381,
      "balance_loss_clip": 1.03421724,
      "balance_loss_mlp": 1.02601147,
      "epoch": 0.7583947091537652,
      "flos": 33546814657920.0,
      "grad_norm": 1.5328929722229196,
      "language_loss": 0.59859818,
      "learning_rate": 5.815929669349135e-07,
      "loss": 0.6197958,
      "num_input_tokens_seen": 272011780,
      "step": 12614,
      "time_per_iteration": 2.6304147243499756
    },
    {
      "auxiliary_loss_clip": 0.01070829,
      "auxiliary_loss_mlp": 0.01027296,
      "balance_loss_clip": 1.03317046,
      "balance_loss_mlp": 1.01450503,
      "epoch": 0.7584548324064332,
      "flos": 20121000641280.0,
      "grad_norm": 1.854123343796536,
      "language_loss": 0.73284352,
      "learning_rate": 5.813184218604246e-07,
      "loss": 0.75382483,
      "num_input_tokens_seen": 272030825,
      "step": 12615,
      "time_per_iteration": 2.555500030517578
    },
    {
      "auxiliary_loss_clip": 0.0101538,
      "auxiliary_loss_mlp": 0.01000042,
      "balance_loss_clip": 1.01524079,
      "balance_loss_mlp": 0.99887389,
      "epoch": 0.7585149556591012,
      "flos": 70402584061440.0,
      "grad_norm": 0.8063946325322698,
      "language_loss": 0.6776129,
      "learning_rate": 5.810439305824828e-07,
      "loss": 0.69776702,
      "num_input_tokens_seen": 272095825,
      "step": 12616,
      "time_per_iteration": 3.1808230876922607
    },
    {
      "auxiliary_loss_clip": 0.01074695,
      "auxiliary_loss_mlp": 0.01034551,
      "balance_loss_clip": 1.03608155,
      "balance_loss_mlp": 1.0217005,
      "epoch": 0.7585750789117691,
      "flos": 16143786293760.0,
      "grad_norm": 1.8243241536540327,
      "language_loss": 0.84895813,
      "learning_rate": 5.807694931114979e-07,
      "loss": 0.87005067,
      "num_input_tokens_seen": 272113950,
      "step": 12617,
      "time_per_iteration": 2.549408435821533
    },
    {
      "auxiliary_loss_clip": 0.01071265,
      "auxiliary_loss_mlp": 0.01031002,
      "balance_loss_clip": 1.03545165,
      "balance_loss_mlp": 1.0194689,
      "epoch": 0.7586352021644371,
      "flos": 17493165544320.0,
      "grad_norm": 4.08440386960874,
      "language_loss": 0.73995215,
      "learning_rate": 5.804951094578757e-07,
      "loss": 0.76097488,
      "num_input_tokens_seen": 272130315,
      "step": 12618,
      "time_per_iteration": 2.514948606491089
    },
    {
      "auxiliary_loss_clip": 0.01085187,
      "auxiliary_loss_mlp": 0.01031913,
      "balance_loss_clip": 1.03813004,
      "balance_loss_mlp": 1.01942623,
      "epoch": 0.758695325417105,
      "flos": 17275187859840.0,
      "grad_norm": 2.0077368300169973,
      "language_loss": 0.77479726,
      "learning_rate": 5.802207796320209e-07,
      "loss": 0.79596823,
      "num_input_tokens_seen": 272149080,
      "step": 12619,
      "time_per_iteration": 2.540688991546631
    },
    {
      "auxiliary_loss_clip": 0.01064775,
      "auxiliary_loss_mlp": 0.01032063,
      "balance_loss_clip": 1.03250968,
      "balance_loss_mlp": 1.01937318,
      "epoch": 0.7587554486697731,
      "flos": 29495660163840.0,
      "grad_norm": 1.8083885897955416,
      "language_loss": 0.82405144,
      "learning_rate": 5.79946503644337e-07,
      "loss": 0.84501982,
      "num_input_tokens_seen": 272168285,
      "step": 12620,
      "time_per_iteration": 2.6320033073425293
    },
    {
      "auxiliary_loss_clip": 0.01081645,
      "auxiliary_loss_mlp": 0.01035369,
      "balance_loss_clip": 1.03611541,
      "balance_loss_mlp": 1.02191067,
      "epoch": 0.758815571922441,
      "flos": 16100800692480.0,
      "grad_norm": 2.2649493735939425,
      "language_loss": 0.81884193,
      "learning_rate": 5.796722815052242e-07,
      "loss": 0.84001207,
      "num_input_tokens_seen": 272184585,
      "step": 12621,
      "time_per_iteration": 2.5399515628814697
    },
    {
      "auxiliary_loss_clip": 0.01084787,
      "auxiliary_loss_mlp": 0.01034603,
      "balance_loss_clip": 1.03654623,
      "balance_loss_mlp": 1.0217998,
      "epoch": 0.758875695175109,
      "flos": 16143714466560.0,
      "grad_norm": 1.903456257653464,
      "language_loss": 0.7345432,
      "learning_rate": 5.7939811322508e-07,
      "loss": 0.75573707,
      "num_input_tokens_seen": 272200205,
      "step": 12622,
      "time_per_iteration": 2.5053815841674805
    },
    {
      "auxiliary_loss_clip": 0.01020989,
      "auxiliary_loss_mlp": 0.01004529,
      "balance_loss_clip": 1.00829422,
      "balance_loss_mlp": 1.00332499,
      "epoch": 0.7589358184277769,
      "flos": 68462006860800.0,
      "grad_norm": 0.8126635712276263,
      "language_loss": 0.60850495,
      "learning_rate": 5.791239988143024e-07,
      "loss": 0.62876016,
      "num_input_tokens_seen": 272259670,
      "step": 12623,
      "time_per_iteration": 3.1831858158111572
    },
    {
      "auxiliary_loss_clip": 0.0110369,
      "auxiliary_loss_mlp": 0.0103602,
      "balance_loss_clip": 1.03729129,
      "balance_loss_mlp": 1.02448678,
      "epoch": 0.7589959416804449,
      "flos": 20047311889920.0,
      "grad_norm": 1.9586233528729622,
      "language_loss": 0.67420292,
      "learning_rate": 5.788499382832847e-07,
      "loss": 0.69560009,
      "num_input_tokens_seen": 272277925,
      "step": 12624,
      "time_per_iteration": 2.4545326232910156
    },
    {
      "auxiliary_loss_clip": 0.01102841,
      "auxiliary_loss_mlp": 0.01024177,
      "balance_loss_clip": 1.03599167,
      "balance_loss_mlp": 1.01179171,
      "epoch": 0.7590560649331128,
      "flos": 18771800958720.0,
      "grad_norm": 1.7488200894997303,
      "language_loss": 0.7617681,
      "learning_rate": 5.785759316424196e-07,
      "loss": 0.78303826,
      "num_input_tokens_seen": 272296010,
      "step": 12625,
      "time_per_iteration": 2.4783124923706055
    },
    {
      "auxiliary_loss_clip": 0.01078995,
      "auxiliary_loss_mlp": 0.01044152,
      "balance_loss_clip": 1.03497446,
      "balance_loss_mlp": 1.03069925,
      "epoch": 0.7591161881857809,
      "flos": 29825284296960.0,
      "grad_norm": 1.775847563571098,
      "language_loss": 0.63042182,
      "learning_rate": 5.783019789020977e-07,
      "loss": 0.65165329,
      "num_input_tokens_seen": 272318330,
      "step": 12626,
      "time_per_iteration": 2.589083433151245
    },
    {
      "auxiliary_loss_clip": 0.01069752,
      "auxiliary_loss_mlp": 0.00786767,
      "balance_loss_clip": 1.03814793,
      "balance_loss_mlp": 1.01197481,
      "epoch": 0.7591763114384488,
      "flos": 20302708567680.0,
      "grad_norm": 1.9359127257216533,
      "language_loss": 0.73818183,
      "learning_rate": 5.780280800727084e-07,
      "loss": 0.75674701,
      "num_input_tokens_seen": 272335265,
      "step": 12627,
      "time_per_iteration": 2.6098482608795166
    },
    {
      "auxiliary_loss_clip": 0.01096321,
      "auxiliary_loss_mlp": 0.01029,
      "balance_loss_clip": 1.03741527,
      "balance_loss_mlp": 1.01680481,
      "epoch": 0.7592364346911168,
      "flos": 20813609664000.0,
      "grad_norm": 2.6868463709306245,
      "language_loss": 0.68989974,
      "learning_rate": 5.777542351646356e-07,
      "loss": 0.71115297,
      "num_input_tokens_seen": 272354795,
      "step": 12628,
      "time_per_iteration": 4.16873836517334
    },
    {
      "auxiliary_loss_clip": 0.01098605,
      "auxiliary_loss_mlp": 0.01034412,
      "balance_loss_clip": 1.03926229,
      "balance_loss_mlp": 1.02113783,
      "epoch": 0.7592965579437848,
      "flos": 21251504367360.0,
      "grad_norm": 2.019192101201329,
      "language_loss": 0.62568164,
      "learning_rate": 5.774804441882648e-07,
      "loss": 0.64701188,
      "num_input_tokens_seen": 272372875,
      "step": 12629,
      "time_per_iteration": 2.544950008392334
    },
    {
      "auxiliary_loss_clip": 0.01080122,
      "auxiliary_loss_mlp": 0.01030015,
      "balance_loss_clip": 1.03461123,
      "balance_loss_mlp": 1.01789165,
      "epoch": 0.7593566811964527,
      "flos": 26213604704640.0,
      "grad_norm": 1.4350798949220849,
      "language_loss": 0.77448779,
      "learning_rate": 5.772067071539786e-07,
      "loss": 0.79558921,
      "num_input_tokens_seen": 272394715,
      "step": 12630,
      "time_per_iteration": 2.568291664123535
    },
    {
      "auxiliary_loss_clip": 0.01031524,
      "auxiliary_loss_mlp": 0.01002081,
      "balance_loss_clip": 1.0085001,
      "balance_loss_mlp": 1.00096655,
      "epoch": 0.7594168044491207,
      "flos": 71237255374080.0,
      "grad_norm": 0.8106193984301918,
      "language_loss": 0.61534786,
      "learning_rate": 5.769330240721562e-07,
      "loss": 0.63568389,
      "num_input_tokens_seen": 272458775,
      "step": 12631,
      "time_per_iteration": 3.167537212371826
    },
    {
      "auxiliary_loss_clip": 0.0107902,
      "auxiliary_loss_mlp": 0.00788442,
      "balance_loss_clip": 1.03745031,
      "balance_loss_mlp": 1.01498079,
      "epoch": 0.7594769277017887,
      "flos": 26613326229120.0,
      "grad_norm": 1.8997286267485045,
      "language_loss": 0.739474,
      "learning_rate": 5.766593949531767e-07,
      "loss": 0.75814867,
      "num_input_tokens_seen": 272479355,
      "step": 12632,
      "time_per_iteration": 3.9765303134918213
    },
    {
      "auxiliary_loss_clip": 0.01085124,
      "auxiliary_loss_mlp": 0.0102901,
      "balance_loss_clip": 1.03719687,
      "balance_loss_mlp": 1.0172503,
      "epoch": 0.7595370509544567,
      "flos": 17595941333760.0,
      "grad_norm": 1.9101910984548796,
      "language_loss": 0.7500608,
      "learning_rate": 5.763858198074154e-07,
      "loss": 0.77120215,
      "num_input_tokens_seen": 272493555,
      "step": 12633,
      "time_per_iteration": 2.52139949798584
    },
    {
      "auxiliary_loss_clip": 0.01081262,
      "auxiliary_loss_mlp": 0.01027669,
      "balance_loss_clip": 1.036165,
      "balance_loss_mlp": 1.01657104,
      "epoch": 0.7595971742071246,
      "flos": 18002953319040.0,
      "grad_norm": 1.9569991223371972,
      "language_loss": 0.73648143,
      "learning_rate": 5.76112298645246e-07,
      "loss": 0.75757074,
      "num_input_tokens_seen": 272508925,
      "step": 12634,
      "time_per_iteration": 3.8411355018615723
    },
    {
      "auxiliary_loss_clip": 0.01107481,
      "auxiliary_loss_mlp": 0.01034538,
      "balance_loss_clip": 1.03772259,
      "balance_loss_mlp": 1.02233672,
      "epoch": 0.7596572974597926,
      "flos": 28840326480000.0,
      "grad_norm": 1.6656391678588975,
      "language_loss": 0.64730805,
      "learning_rate": 5.758388314770408e-07,
      "loss": 0.66872823,
      "num_input_tokens_seen": 272528805,
      "step": 12635,
      "time_per_iteration": 2.5021257400512695
    },
    {
      "auxiliary_loss_clip": 0.01049032,
      "auxiliary_loss_mlp": 0.01039762,
      "balance_loss_clip": 1.03365219,
      "balance_loss_mlp": 1.02453923,
      "epoch": 0.7597174207124605,
      "flos": 14282823588480.0,
      "grad_norm": 1.68526363091464,
      "language_loss": 0.68572658,
      "learning_rate": 5.7556541831317e-07,
      "loss": 0.70661455,
      "num_input_tokens_seen": 272546655,
      "step": 12636,
      "time_per_iteration": 2.5675365924835205
    },
    {
      "auxiliary_loss_clip": 0.01089422,
      "auxiliary_loss_mlp": 0.01036823,
      "balance_loss_clip": 1.03919232,
      "balance_loss_mlp": 1.02490187,
      "epoch": 0.7597775439651285,
      "flos": 21688932193920.0,
      "grad_norm": 1.7701478691444847,
      "language_loss": 0.80990887,
      "learning_rate": 5.752920591640018e-07,
      "loss": 0.83117127,
      "num_input_tokens_seen": 272564010,
      "step": 12637,
      "time_per_iteration": 2.5088589191436768
    },
    {
      "auxiliary_loss_clip": 0.01093926,
      "auxiliary_loss_mlp": 0.01035723,
      "balance_loss_clip": 1.03572106,
      "balance_loss_mlp": 1.02388,
      "epoch": 0.7598376672177964,
      "flos": 36101248312320.0,
      "grad_norm": 2.0227002142260897,
      "language_loss": 0.66355217,
      "learning_rate": 5.750187540399017e-07,
      "loss": 0.68484867,
      "num_input_tokens_seen": 272585840,
      "step": 12638,
      "time_per_iteration": 2.6175339221954346
    },
    {
      "auxiliary_loss_clip": 0.01106802,
      "auxiliary_loss_mlp": 0.01037713,
      "balance_loss_clip": 1.03679276,
      "balance_loss_mlp": 1.02386117,
      "epoch": 0.7598977904704645,
      "flos": 18332326056960.0,
      "grad_norm": 2.3459312606424727,
      "language_loss": 0.65335482,
      "learning_rate": 5.747455029512323e-07,
      "loss": 0.67479998,
      "num_input_tokens_seen": 272602300,
      "step": 12639,
      "time_per_iteration": 3.839275598526001
    },
    {
      "auxiliary_loss_clip": 0.01093411,
      "auxiliary_loss_mlp": 0.01029015,
      "balance_loss_clip": 1.03469777,
      "balance_loss_mlp": 1.01667118,
      "epoch": 0.7599579137231324,
      "flos": 20192642317440.0,
      "grad_norm": 1.917374208932494,
      "language_loss": 0.69718921,
      "learning_rate": 5.744723059083572e-07,
      "loss": 0.71841347,
      "num_input_tokens_seen": 272619595,
      "step": 12640,
      "time_per_iteration": 2.465106725692749
    },
    {
      "auxiliary_loss_clip": 0.01085077,
      "auxiliary_loss_mlp": 0.01031132,
      "balance_loss_clip": 1.03728056,
      "balance_loss_mlp": 1.0179714,
      "epoch": 0.7600180369758004,
      "flos": 24024849459840.0,
      "grad_norm": 1.7460526424857297,
      "language_loss": 0.67235816,
      "learning_rate": 5.741991629216343e-07,
      "loss": 0.69352019,
      "num_input_tokens_seen": 272638825,
      "step": 12641,
      "time_per_iteration": 2.532058000564575
    },
    {
      "auxiliary_loss_clip": 0.01093043,
      "auxiliary_loss_mlp": 0.01031873,
      "balance_loss_clip": 1.03489494,
      "balance_loss_mlp": 1.0189451,
      "epoch": 0.7600781602284684,
      "flos": 18989527248000.0,
      "grad_norm": 2.2235172760308153,
      "language_loss": 0.66631711,
      "learning_rate": 5.73926074001422e-07,
      "loss": 0.68756628,
      "num_input_tokens_seen": 272657240,
      "step": 12642,
      "time_per_iteration": 2.4637317657470703
    },
    {
      "auxiliary_loss_clip": 0.01086106,
      "auxiliary_loss_mlp": 0.01030429,
      "balance_loss_clip": 1.03823185,
      "balance_loss_mlp": 1.01816189,
      "epoch": 0.7601382834811363,
      "flos": 26067520091520.0,
      "grad_norm": 1.793887868643889,
      "language_loss": 0.75215214,
      "learning_rate": 5.736530391580765e-07,
      "loss": 0.77331752,
      "num_input_tokens_seen": 272677520,
      "step": 12643,
      "time_per_iteration": 2.5543930530548096
    },
    {
      "auxiliary_loss_clip": 0.01070055,
      "auxiliary_loss_mlp": 0.01036117,
      "balance_loss_clip": 1.03625584,
      "balance_loss_mlp": 1.02273548,
      "epoch": 0.7601984067338043,
      "flos": 18844232734080.0,
      "grad_norm": 1.9296953518833748,
      "language_loss": 0.78826934,
      "learning_rate": 5.733800584019508e-07,
      "loss": 0.80933106,
      "num_input_tokens_seen": 272696770,
      "step": 12644,
      "time_per_iteration": 2.5304133892059326
    },
    {
      "auxiliary_loss_clip": 0.01079793,
      "auxiliary_loss_mlp": 0.01031193,
      "balance_loss_clip": 1.03393292,
      "balance_loss_mlp": 1.01877713,
      "epoch": 0.7602585299864723,
      "flos": 24646391424000.0,
      "grad_norm": 1.626303169780059,
      "language_loss": 0.80164802,
      "learning_rate": 5.731071317433957e-07,
      "loss": 0.82275784,
      "num_input_tokens_seen": 272718340,
      "step": 12645,
      "time_per_iteration": 2.5612692832946777
    },
    {
      "auxiliary_loss_clip": 0.01090466,
      "auxiliary_loss_mlp": 0.01031568,
      "balance_loss_clip": 1.03879917,
      "balance_loss_mlp": 1.0190208,
      "epoch": 0.7603186532391403,
      "flos": 23842100039040.0,
      "grad_norm": 1.7073777372649244,
      "language_loss": 0.73047274,
      "learning_rate": 5.728342591927611e-07,
      "loss": 0.75169307,
      "num_input_tokens_seen": 272739575,
      "step": 12646,
      "time_per_iteration": 2.5349104404449463
    },
    {
      "auxiliary_loss_clip": 0.01093884,
      "auxiliary_loss_mlp": 0.01035402,
      "balance_loss_clip": 1.03645074,
      "balance_loss_mlp": 1.02363062,
      "epoch": 0.7603787764918082,
      "flos": 22199905117440.0,
      "grad_norm": 1.9078607738330249,
      "language_loss": 0.67156005,
      "learning_rate": 5.725614407603949e-07,
      "loss": 0.69285285,
      "num_input_tokens_seen": 272758710,
      "step": 12647,
      "time_per_iteration": 2.5044877529144287
    },
    {
      "auxiliary_loss_clip": 0.01026456,
      "auxiliary_loss_mlp": 0.01019864,
      "balance_loss_clip": 1.01330328,
      "balance_loss_mlp": 1.01837397,
      "epoch": 0.7604388997444762,
      "flos": 54086894254080.0,
      "grad_norm": 0.6895456355460703,
      "language_loss": 0.49006724,
      "learning_rate": 5.722886764566415e-07,
      "loss": 0.51053047,
      "num_input_tokens_seen": 272814855,
      "step": 12648,
      "time_per_iteration": 3.07966947555542
    },
    {
      "auxiliary_loss_clip": 0.01091437,
      "auxiliary_loss_mlp": 0.01032861,
      "balance_loss_clip": 1.03512287,
      "balance_loss_mlp": 1.0211066,
      "epoch": 0.7604990229971441,
      "flos": 19681920789120.0,
      "grad_norm": 2.337665925509676,
      "language_loss": 0.76490283,
      "learning_rate": 5.720159662918451e-07,
      "loss": 0.78614581,
      "num_input_tokens_seen": 272834400,
      "step": 12649,
      "time_per_iteration": 2.583862781524658
    },
    {
      "auxiliary_loss_clip": 0.0106258,
      "auxiliary_loss_mlp": 0.01031246,
      "balance_loss_clip": 1.03412223,
      "balance_loss_mlp": 1.01951027,
      "epoch": 0.7605591462498121,
      "flos": 25228036356480.0,
      "grad_norm": 1.5043419811876233,
      "language_loss": 0.6879248,
      "learning_rate": 5.717433102763462e-07,
      "loss": 0.70886302,
      "num_input_tokens_seen": 272854760,
      "step": 12650,
      "time_per_iteration": 2.572755813598633
    },
    {
      "auxiliary_loss_clip": 0.01021187,
      "auxiliary_loss_mlp": 0.01003987,
      "balance_loss_clip": 1.00846362,
      "balance_loss_mlp": 1.00284886,
      "epoch": 0.76061926950248,
      "flos": 66783757662720.0,
      "grad_norm": 0.7544929821757882,
      "language_loss": 0.62698829,
      "learning_rate": 5.714707084204838e-07,
      "loss": 0.64724004,
      "num_input_tokens_seen": 272919030,
      "step": 12651,
      "time_per_iteration": 3.109058141708374
    },
    {
      "auxiliary_loss_clip": 0.01071034,
      "auxiliary_loss_mlp": 0.0103321,
      "balance_loss_clip": 1.03505373,
      "balance_loss_mlp": 1.02149725,
      "epoch": 0.7606793927551481,
      "flos": 25338354001920.0,
      "grad_norm": 1.412892542170891,
      "language_loss": 0.71393794,
      "learning_rate": 5.711981607345951e-07,
      "loss": 0.73498034,
      "num_input_tokens_seen": 272938925,
      "step": 12652,
      "time_per_iteration": 2.6114797592163086
    },
    {
      "auxiliary_loss_clip": 0.01054204,
      "auxiliary_loss_mlp": 0.01036416,
      "balance_loss_clip": 1.03515172,
      "balance_loss_mlp": 1.02395272,
      "epoch": 0.760739516007816,
      "flos": 18223624523520.0,
      "grad_norm": 1.790837276620901,
      "language_loss": 0.79994977,
      "learning_rate": 5.709256672290152e-07,
      "loss": 0.82085598,
      "num_input_tokens_seen": 272954945,
      "step": 12653,
      "time_per_iteration": 2.596724510192871
    },
    {
      "auxiliary_loss_clip": 0.01109941,
      "auxiliary_loss_mlp": 0.01030144,
      "balance_loss_clip": 1.03820384,
      "balance_loss_mlp": 1.01793051,
      "epoch": 0.760799639260484,
      "flos": 22559119079040.0,
      "grad_norm": 1.522474760179085,
      "language_loss": 0.80318362,
      "learning_rate": 5.706532279140785e-07,
      "loss": 0.82458442,
      "num_input_tokens_seen": 272972855,
      "step": 12654,
      "time_per_iteration": 2.4586777687072754
    },
    {
      "auxiliary_loss_clip": 0.01069662,
      "auxiliary_loss_mlp": 0.010364,
      "balance_loss_clip": 1.0340836,
      "balance_loss_mlp": 1.02294719,
      "epoch": 0.760859762513152,
      "flos": 22309324922880.0,
      "grad_norm": 1.9488485380738214,
      "language_loss": 0.79734337,
      "learning_rate": 5.703808428001136e-07,
      "loss": 0.81840396,
      "num_input_tokens_seen": 272989895,
      "step": 12655,
      "time_per_iteration": 2.5570497512817383
    },
    {
      "auxiliary_loss_clip": 0.01086731,
      "auxiliary_loss_mlp": 0.01024837,
      "balance_loss_clip": 1.03572011,
      "balance_loss_mlp": 1.01482368,
      "epoch": 0.7609198857658199,
      "flos": 24863902231680.0,
      "grad_norm": 1.829672254998866,
      "language_loss": 0.68527585,
      "learning_rate": 5.701085118974505e-07,
      "loss": 0.70639145,
      "num_input_tokens_seen": 273011695,
      "step": 12656,
      "time_per_iteration": 2.5388708114624023
    },
    {
      "auxiliary_loss_clip": 0.01095574,
      "auxiliary_loss_mlp": 0.01029683,
      "balance_loss_clip": 1.03331339,
      "balance_loss_mlp": 1.01667702,
      "epoch": 0.760980009018488,
      "flos": 16836790366080.0,
      "grad_norm": 3.7074335310627107,
      "language_loss": 0.73577452,
      "learning_rate": 5.698362352164164e-07,
      "loss": 0.75702703,
      "num_input_tokens_seen": 273028815,
      "step": 12657,
      "time_per_iteration": 2.4616217613220215
    },
    {
      "auxiliary_loss_clip": 0.0101472,
      "auxiliary_loss_mlp": 0.01002872,
      "balance_loss_clip": 1.00986052,
      "balance_loss_mlp": 1.00167942,
      "epoch": 0.7610401322711559,
      "flos": 61230603029760.0,
      "grad_norm": 0.8591108360913255,
      "language_loss": 0.64944541,
      "learning_rate": 5.695640127673347e-07,
      "loss": 0.66962135,
      "num_input_tokens_seen": 273084080,
      "step": 12658,
      "time_per_iteration": 3.0704715251922607
    },
    {
      "auxiliary_loss_clip": 0.01087869,
      "auxiliary_loss_mlp": 0.01033401,
      "balance_loss_clip": 1.0366621,
      "balance_loss_mlp": 1.02090764,
      "epoch": 0.7611002555238239,
      "flos": 19640730867840.0,
      "grad_norm": 1.6236491480527129,
      "language_loss": 0.7924844,
      "learning_rate": 5.692918445605293e-07,
      "loss": 0.81369716,
      "num_input_tokens_seen": 273102295,
      "step": 12659,
      "time_per_iteration": 2.4778554439544678
    },
    {
      "auxiliary_loss_clip": 0.01091775,
      "auxiliary_loss_mlp": 0.01027012,
      "balance_loss_clip": 1.0342679,
      "balance_loss_mlp": 1.01504922,
      "epoch": 0.7611603787764918,
      "flos": 26872206526080.0,
      "grad_norm": 1.526035034454464,
      "language_loss": 0.68629593,
      "learning_rate": 5.690197306063209e-07,
      "loss": 0.70748377,
      "num_input_tokens_seen": 273123400,
      "step": 12660,
      "time_per_iteration": 2.54925799369812
    },
    {
      "auxiliary_loss_clip": 0.0110517,
      "auxiliary_loss_mlp": 0.01029143,
      "balance_loss_clip": 1.03561664,
      "balance_loss_mlp": 1.01741862,
      "epoch": 0.7612205020291598,
      "flos": 27344252085120.0,
      "grad_norm": 1.5201468402166578,
      "language_loss": 0.70361006,
      "learning_rate": 5.687476709150281e-07,
      "loss": 0.72495317,
      "num_input_tokens_seen": 273145150,
      "step": 12661,
      "time_per_iteration": 2.4921281337738037
    },
    {
      "auxiliary_loss_clip": 0.01092043,
      "auxiliary_loss_mlp": 0.01029751,
      "balance_loss_clip": 1.03391039,
      "balance_loss_mlp": 1.01832438,
      "epoch": 0.7612806252818277,
      "flos": 29314598682240.0,
      "grad_norm": 1.44674272403428,
      "language_loss": 0.83258855,
      "learning_rate": 5.68475665496966e-07,
      "loss": 0.85380644,
      "num_input_tokens_seen": 273165180,
      "step": 12662,
      "time_per_iteration": 2.554934024810791
    },
    {
      "auxiliary_loss_clip": 0.01081741,
      "auxiliary_loss_mlp": 0.01042987,
      "balance_loss_clip": 1.03474498,
      "balance_loss_mlp": 1.03102422,
      "epoch": 0.7613407485344957,
      "flos": 19026048401280.0,
      "grad_norm": 1.7361906176983537,
      "language_loss": 0.68539822,
      "learning_rate": 5.682037143624505e-07,
      "loss": 0.70664549,
      "num_input_tokens_seen": 273184005,
      "step": 12663,
      "time_per_iteration": 2.508544683456421
    },
    {
      "auxiliary_loss_clip": 0.01090806,
      "auxiliary_loss_mlp": 0.01024805,
      "balance_loss_clip": 1.03492165,
      "balance_loss_mlp": 1.01323569,
      "epoch": 0.7614008717871636,
      "flos": 23256037733760.0,
      "grad_norm": 1.8770742666827729,
      "language_loss": 0.70142728,
      "learning_rate": 5.67931817521794e-07,
      "loss": 0.72258341,
      "num_input_tokens_seen": 273203565,
      "step": 12664,
      "time_per_iteration": 2.5068535804748535
    },
    {
      "auxiliary_loss_clip": 0.01098827,
      "auxiliary_loss_mlp": 0.01036585,
      "balance_loss_clip": 1.03818107,
      "balance_loss_mlp": 1.02375841,
      "epoch": 0.7614609950398317,
      "flos": 21579907438080.0,
      "grad_norm": 1.629522900379031,
      "language_loss": 0.79246885,
      "learning_rate": 5.676599749853066e-07,
      "loss": 0.81382298,
      "num_input_tokens_seen": 273221645,
      "step": 12665,
      "time_per_iteration": 2.4873647689819336
    },
    {
      "auxiliary_loss_clip": 0.01104301,
      "auxiliary_loss_mlp": 0.0078218,
      "balance_loss_clip": 1.03807521,
      "balance_loss_mlp": 1.00856769,
      "epoch": 0.7615211182924996,
      "flos": 29277897960960.0,
      "grad_norm": 1.6084873614930912,
      "language_loss": 0.87998295,
      "learning_rate": 5.673881867632959e-07,
      "loss": 0.89884776,
      "num_input_tokens_seen": 273242040,
      "step": 12666,
      "time_per_iteration": 2.546276569366455
    },
    {
      "auxiliary_loss_clip": 0.01049757,
      "auxiliary_loss_mlp": 0.01032328,
      "balance_loss_clip": 1.03375638,
      "balance_loss_mlp": 1.0193758,
      "epoch": 0.7615812415451676,
      "flos": 13261129136640.0,
      "grad_norm": 2.4053441903649766,
      "language_loss": 0.83154428,
      "learning_rate": 5.671164528660693e-07,
      "loss": 0.8523652,
      "num_input_tokens_seen": 273257365,
      "step": 12667,
      "time_per_iteration": 3.943321704864502
    },
    {
      "auxiliary_loss_clip": 0.01081843,
      "auxiliary_loss_mlp": 0.01034692,
      "balance_loss_clip": 1.03559148,
      "balance_loss_mlp": 1.02337861,
      "epoch": 0.7616413647978356,
      "flos": 18584741905920.0,
      "grad_norm": 1.5741337388476715,
      "language_loss": 0.78520072,
      "learning_rate": 5.668447733039296e-07,
      "loss": 0.80636609,
      "num_input_tokens_seen": 273274710,
      "step": 12668,
      "time_per_iteration": 2.5610439777374268
    },
    {
      "auxiliary_loss_clip": 0.01069229,
      "auxiliary_loss_mlp": 0.01027278,
      "balance_loss_clip": 1.03296232,
      "balance_loss_mlp": 1.01555419,
      "epoch": 0.7617014880505035,
      "flos": 18516188799360.0,
      "grad_norm": 1.9133164203322024,
      "language_loss": 0.64428639,
      "learning_rate": 5.6657314808718e-07,
      "loss": 0.66525149,
      "num_input_tokens_seen": 273292870,
      "step": 12669,
      "time_per_iteration": 2.5123379230499268
    },
    {
      "auxiliary_loss_clip": 0.01079073,
      "auxiliary_loss_mlp": 0.01038598,
      "balance_loss_clip": 1.03297484,
      "balance_loss_mlp": 1.02496028,
      "epoch": 0.7617616113031715,
      "flos": 24973178382720.0,
      "grad_norm": 1.8544854934285684,
      "language_loss": 0.66342235,
      "learning_rate": 5.663015772261202e-07,
      "loss": 0.68459904,
      "num_input_tokens_seen": 273312375,
      "step": 12670,
      "time_per_iteration": 2.587808847427368
    },
    {
      "auxiliary_loss_clip": 0.0109621,
      "auxiliary_loss_mlp": 0.01034164,
      "balance_loss_clip": 1.03640008,
      "balance_loss_mlp": 1.02230251,
      "epoch": 0.7618217345558395,
      "flos": 23295036925440.0,
      "grad_norm": 1.7300877599952822,
      "language_loss": 0.73121452,
      "learning_rate": 5.660300607310493e-07,
      "loss": 0.7525183,
      "num_input_tokens_seen": 273332590,
      "step": 12671,
      "time_per_iteration": 3.940094232559204
    },
    {
      "auxiliary_loss_clip": 0.01067127,
      "auxiliary_loss_mlp": 0.01035555,
      "balance_loss_clip": 1.0324285,
      "balance_loss_mlp": 1.02276397,
      "epoch": 0.7618818578085075,
      "flos": 25482894330240.0,
      "grad_norm": 1.674856908289275,
      "language_loss": 0.73448813,
      "learning_rate": 5.657585986122613e-07,
      "loss": 0.75551498,
      "num_input_tokens_seen": 273352885,
      "step": 12672,
      "time_per_iteration": 3.9829635620117188
    },
    {
      "auxiliary_loss_clip": 0.01009578,
      "auxiliary_loss_mlp": 0.01001001,
      "balance_loss_clip": 1.00864458,
      "balance_loss_mlp": 0.99988633,
      "epoch": 0.7619419810611754,
      "flos": 61151994115200.0,
      "grad_norm": 0.7549985006659468,
      "language_loss": 0.56632227,
      "learning_rate": 5.654871908800506e-07,
      "loss": 0.58642805,
      "num_input_tokens_seen": 273411730,
      "step": 12673,
      "time_per_iteration": 3.0915274620056152
    },
    {
      "auxiliary_loss_clip": 0.0109338,
      "auxiliary_loss_mlp": 0.01031168,
      "balance_loss_clip": 1.03588951,
      "balance_loss_mlp": 1.01846075,
      "epoch": 0.7620021043138434,
      "flos": 23258659426560.0,
      "grad_norm": 1.8674317137531766,
      "language_loss": 0.74939716,
      "learning_rate": 5.652158375447102e-07,
      "loss": 0.77064264,
      "num_input_tokens_seen": 273430020,
      "step": 12674,
      "time_per_iteration": 2.5041821002960205
    },
    {
      "auxiliary_loss_clip": 0.01074143,
      "auxiliary_loss_mlp": 0.01036311,
      "balance_loss_clip": 1.03267705,
      "balance_loss_mlp": 1.02294219,
      "epoch": 0.7620622275665113,
      "flos": 25082490447360.0,
      "grad_norm": 1.8697032906046793,
      "language_loss": 0.7222293,
      "learning_rate": 5.649445386165286e-07,
      "loss": 0.74333382,
      "num_input_tokens_seen": 273448690,
      "step": 12675,
      "time_per_iteration": 2.548712968826294
    },
    {
      "auxiliary_loss_clip": 0.01091885,
      "auxiliary_loss_mlp": 0.01028021,
      "balance_loss_clip": 1.03577948,
      "balance_loss_mlp": 1.01654136,
      "epoch": 0.7621223508191793,
      "flos": 20155007842560.0,
      "grad_norm": 2.263562855700515,
      "language_loss": 0.73041373,
      "learning_rate": 5.646732941057936e-07,
      "loss": 0.75161278,
      "num_input_tokens_seen": 273465190,
      "step": 12676,
      "time_per_iteration": 2.4798154830932617
    },
    {
      "auxiliary_loss_clip": 0.01075955,
      "auxiliary_loss_mlp": 0.00786274,
      "balance_loss_clip": 1.03884649,
      "balance_loss_mlp": 1.01154947,
      "epoch": 0.7621824740718472,
      "flos": 18000187971840.0,
      "grad_norm": 2.666538220917961,
      "language_loss": 0.53498459,
      "learning_rate": 5.644021040227927e-07,
      "loss": 0.55360693,
      "num_input_tokens_seen": 273478620,
      "step": 12677,
      "time_per_iteration": 2.508429765701294
    },
    {
      "auxiliary_loss_clip": 0.01056399,
      "auxiliary_loss_mlp": 0.01029639,
      "balance_loss_clip": 1.03478682,
      "balance_loss_mlp": 1.01670516,
      "epoch": 0.7622425973245153,
      "flos": 21725668828800.0,
      "grad_norm": 1.8915611992771084,
      "language_loss": 0.79006183,
      "learning_rate": 5.641309683778064e-07,
      "loss": 0.81092227,
      "num_input_tokens_seen": 273497635,
      "step": 12678,
      "time_per_iteration": 3.9600744247436523
    },
    {
      "auxiliary_loss_clip": 0.01069441,
      "auxiliary_loss_mlp": 0.0103636,
      "balance_loss_clip": 1.0335927,
      "balance_loss_mlp": 1.02260947,
      "epoch": 0.7623027205771832,
      "flos": 19718549683200.0,
      "grad_norm": 2.1229357382465626,
      "language_loss": 0.77471614,
      "learning_rate": 5.638598871811175e-07,
      "loss": 0.7957741,
      "num_input_tokens_seen": 273513955,
      "step": 12679,
      "time_per_iteration": 2.5113048553466797
    },
    {
      "auxiliary_loss_clip": 0.01092724,
      "auxiliary_loss_mlp": 0.01023954,
      "balance_loss_clip": 1.03499031,
      "balance_loss_mlp": 1.01168156,
      "epoch": 0.7623628438298512,
      "flos": 23988831096960.0,
      "grad_norm": 1.589777156068415,
      "language_loss": 0.80097955,
      "learning_rate": 5.635888604430059e-07,
      "loss": 0.8221463,
      "num_input_tokens_seen": 273533970,
      "step": 12680,
      "time_per_iteration": 2.546685218811035
    },
    {
      "auxiliary_loss_clip": 0.01084993,
      "auxiliary_loss_mlp": 0.01031342,
      "balance_loss_clip": 1.0373652,
      "balance_loss_mlp": 1.01765656,
      "epoch": 0.7624229670825191,
      "flos": 22345702421760.0,
      "grad_norm": 1.7026043582740156,
      "language_loss": 0.62799972,
      "learning_rate": 5.633178881737493e-07,
      "loss": 0.64916307,
      "num_input_tokens_seen": 273553090,
      "step": 12681,
      "time_per_iteration": 2.5129544734954834
    },
    {
      "auxiliary_loss_clip": 0.01068778,
      "auxiliary_loss_mlp": 0.01029052,
      "balance_loss_clip": 1.03604901,
      "balance_loss_mlp": 1.017465,
      "epoch": 0.7624830903351871,
      "flos": 22711775880960.0,
      "grad_norm": 1.9816637790571414,
      "language_loss": 0.75933564,
      "learning_rate": 5.63046970383622e-07,
      "loss": 0.78031385,
      "num_input_tokens_seen": 273572460,
      "step": 12682,
      "time_per_iteration": 2.578129768371582
    },
    {
      "auxiliary_loss_clip": 0.01080462,
      "auxiliary_loss_mlp": 0.01029623,
      "balance_loss_clip": 1.03408694,
      "balance_loss_mlp": 1.0186677,
      "epoch": 0.7625432135878552,
      "flos": 25593714766080.0,
      "grad_norm": 1.427002460060655,
      "language_loss": 0.67946064,
      "learning_rate": 5.627761070828974e-07,
      "loss": 0.70056152,
      "num_input_tokens_seen": 273592815,
      "step": 12683,
      "time_per_iteration": 2.5489578247070312
    },
    {
      "auxiliary_loss_clip": 0.01065731,
      "auxiliary_loss_mlp": 0.0078609,
      "balance_loss_clip": 1.03382349,
      "balance_loss_mlp": 1.01001489,
      "epoch": 0.7626033368405231,
      "flos": 23987645948160.0,
      "grad_norm": 2.4977853433011443,
      "language_loss": 0.83452785,
      "learning_rate": 5.625052982818472e-07,
      "loss": 0.85304606,
      "num_input_tokens_seen": 273611790,
      "step": 12684,
      "time_per_iteration": 2.586915969848633
    },
    {
      "auxiliary_loss_clip": 0.01084125,
      "auxiliary_loss_mlp": 0.01030753,
      "balance_loss_clip": 1.03573012,
      "balance_loss_mlp": 1.01718116,
      "epoch": 0.7626634600931911,
      "flos": 12599115523200.0,
      "grad_norm": 3.943461453176747,
      "language_loss": 0.82525241,
      "learning_rate": 5.622345439907396e-07,
      "loss": 0.84640121,
      "num_input_tokens_seen": 273628340,
      "step": 12685,
      "time_per_iteration": 2.4941577911376953
    },
    {
      "auxiliary_loss_clip": 0.01075552,
      "auxiliary_loss_mlp": 0.0078518,
      "balance_loss_clip": 1.03513885,
      "balance_loss_mlp": 1.01050735,
      "epoch": 0.762723583345859,
      "flos": 26322593546880.0,
      "grad_norm": 1.9297028170463524,
      "language_loss": 0.77376783,
      "learning_rate": 5.619638442198422e-07,
      "loss": 0.79237521,
      "num_input_tokens_seen": 273646585,
      "step": 12686,
      "time_per_iteration": 2.5836827754974365
    },
    {
      "auxiliary_loss_clip": 0.01048715,
      "auxiliary_loss_mlp": 0.01045521,
      "balance_loss_clip": 1.0328517,
      "balance_loss_mlp": 1.03003585,
      "epoch": 0.762783706598527,
      "flos": 21907053532800.0,
      "grad_norm": 1.7209627421678655,
      "language_loss": 0.7217195,
      "learning_rate": 5.616931989794198e-07,
      "loss": 0.74266189,
      "num_input_tokens_seen": 273665410,
      "step": 12687,
      "time_per_iteration": 2.61552095413208
    },
    {
      "auxiliary_loss_clip": 0.01081793,
      "auxiliary_loss_mlp": 0.01045851,
      "balance_loss_clip": 1.03582585,
      "balance_loss_mlp": 1.03122413,
      "epoch": 0.7628438298511949,
      "flos": 15339782217600.0,
      "grad_norm": 2.0478197250130004,
      "language_loss": 0.64738488,
      "learning_rate": 5.614226082797369e-07,
      "loss": 0.66866136,
      "num_input_tokens_seen": 273683035,
      "step": 12688,
      "time_per_iteration": 2.5080111026763916
    },
    {
      "auxiliary_loss_clip": 0.01094198,
      "auxiliary_loss_mlp": 0.01027838,
      "balance_loss_clip": 1.03629017,
      "balance_loss_mlp": 1.01645327,
      "epoch": 0.7629039531038629,
      "flos": 13006307076480.0,
      "grad_norm": 1.9184383376244654,
      "language_loss": 0.70836234,
      "learning_rate": 5.611520721310515e-07,
      "loss": 0.72958267,
      "num_input_tokens_seen": 273700130,
      "step": 12689,
      "time_per_iteration": 2.482851028442383
    },
    {
      "auxiliary_loss_clip": 0.0107458,
      "auxiliary_loss_mlp": 0.01037624,
      "balance_loss_clip": 1.03469253,
      "balance_loss_mlp": 1.02487445,
      "epoch": 0.7629640763565309,
      "flos": 26171660597760.0,
      "grad_norm": 1.9524429470624853,
      "language_loss": 0.6984669,
      "learning_rate": 5.608815905436238e-07,
      "loss": 0.71958888,
      "num_input_tokens_seen": 273720310,
      "step": 12690,
      "time_per_iteration": 2.610696315765381
    },
    {
      "auxiliary_loss_clip": 0.01079595,
      "auxiliary_loss_mlp": 0.01036313,
      "balance_loss_clip": 1.03440261,
      "balance_loss_mlp": 1.02344441,
      "epoch": 0.7630241996091989,
      "flos": 36793713680640.0,
      "grad_norm": 1.4563334060608215,
      "language_loss": 0.69403744,
      "learning_rate": 5.606111635277109e-07,
      "loss": 0.71519655,
      "num_input_tokens_seen": 273744475,
      "step": 12691,
      "time_per_iteration": 2.668941020965576
    },
    {
      "auxiliary_loss_clip": 0.01088885,
      "auxiliary_loss_mlp": 0.01035721,
      "balance_loss_clip": 1.03546226,
      "balance_loss_mlp": 1.02449703,
      "epoch": 0.7630843228618668,
      "flos": 21835160461440.0,
      "grad_norm": 1.6197845181352042,
      "language_loss": 0.81429905,
      "learning_rate": 5.603407910935662e-07,
      "loss": 0.83554512,
      "num_input_tokens_seen": 273764635,
      "step": 12692,
      "time_per_iteration": 2.495760440826416
    },
    {
      "auxiliary_loss_clip": 0.01072051,
      "auxiliary_loss_mlp": 0.01028367,
      "balance_loss_clip": 1.03776157,
      "balance_loss_mlp": 1.01657724,
      "epoch": 0.7631444461145348,
      "flos": 12640520926080.0,
      "grad_norm": 2.253722798539994,
      "language_loss": 0.76997781,
      "learning_rate": 5.600704732514438e-07,
      "loss": 0.79098195,
      "num_input_tokens_seen": 273780115,
      "step": 12693,
      "time_per_iteration": 2.5511112213134766
    },
    {
      "auxiliary_loss_clip": 0.01066929,
      "auxiliary_loss_mlp": 0.0102801,
      "balance_loss_clip": 1.03592229,
      "balance_loss_mlp": 1.01528478,
      "epoch": 0.7632045693672027,
      "flos": 16836610798080.0,
      "grad_norm": 2.693808251305855,
      "language_loss": 0.72598898,
      "learning_rate": 5.598002100115933e-07,
      "loss": 0.74693835,
      "num_input_tokens_seen": 273796605,
      "step": 12694,
      "time_per_iteration": 2.511033058166504
    },
    {
      "auxiliary_loss_clip": 0.01091232,
      "auxiliary_loss_mlp": 0.01025423,
      "balance_loss_clip": 1.03426623,
      "balance_loss_mlp": 1.0130074,
      "epoch": 0.7632646926198707,
      "flos": 22017335264640.0,
      "grad_norm": 1.7411689628442022,
      "language_loss": 0.70658147,
      "learning_rate": 5.595300013842625e-07,
      "loss": 0.72774804,
      "num_input_tokens_seen": 273816515,
      "step": 12695,
      "time_per_iteration": 2.524937629699707
    },
    {
      "auxiliary_loss_clip": 0.01104409,
      "auxiliary_loss_mlp": 0.0103082,
      "balance_loss_clip": 1.03597522,
      "balance_loss_mlp": 1.01913786,
      "epoch": 0.7633248158725388,
      "flos": 23114011357440.0,
      "grad_norm": 1.4340338975923714,
      "language_loss": 0.7227546,
      "learning_rate": 5.592598473796985e-07,
      "loss": 0.74410689,
      "num_input_tokens_seen": 273837060,
      "step": 12696,
      "time_per_iteration": 2.484015941619873
    },
    {
      "auxiliary_loss_clip": 0.01046201,
      "auxiliary_loss_mlp": 0.01039197,
      "balance_loss_clip": 1.03418827,
      "balance_loss_mlp": 1.02575588,
      "epoch": 0.7633849391252067,
      "flos": 10889839952640.0,
      "grad_norm": 2.259469705165791,
      "language_loss": 0.71179831,
      "learning_rate": 5.589897480081453e-07,
      "loss": 0.73265231,
      "num_input_tokens_seen": 273853365,
      "step": 12697,
      "time_per_iteration": 2.5924935340881348
    },
    {
      "auxiliary_loss_clip": 0.01070493,
      "auxiliary_loss_mlp": 0.01026895,
      "balance_loss_clip": 1.0392586,
      "balance_loss_mlp": 1.01512313,
      "epoch": 0.7634450623778747,
      "flos": 20994168355200.0,
      "grad_norm": 1.9953388312114715,
      "language_loss": 0.66791278,
      "learning_rate": 5.587197032798461e-07,
      "loss": 0.68888664,
      "num_input_tokens_seen": 273870750,
      "step": 12698,
      "time_per_iteration": 2.6198856830596924
    },
    {
      "auxiliary_loss_clip": 0.010923,
      "auxiliary_loss_mlp": 0.01026764,
      "balance_loss_clip": 1.03349221,
      "balance_loss_mlp": 1.01453912,
      "epoch": 0.7635051856305426,
      "flos": 18882046776960.0,
      "grad_norm": 1.7365582668133617,
      "language_loss": 0.71890408,
      "learning_rate": 5.5844971320504e-07,
      "loss": 0.74009472,
      "num_input_tokens_seen": 273890890,
      "step": 12699,
      "time_per_iteration": 2.5164833068847656
    },
    {
      "auxiliary_loss_clip": 0.01082944,
      "auxiliary_loss_mlp": 0.01032957,
      "balance_loss_clip": 1.03531766,
      "balance_loss_mlp": 1.02138162,
      "epoch": 0.7635653088832106,
      "flos": 34786989584640.0,
      "grad_norm": 1.7144279871177495,
      "language_loss": 0.73210061,
      "learning_rate": 5.581797777939648e-07,
      "loss": 0.7532596,
      "num_input_tokens_seen": 273914015,
      "step": 12700,
      "time_per_iteration": 2.6401526927948
    },
    {
      "auxiliary_loss_clip": 0.01104428,
      "auxiliary_loss_mlp": 0.01031516,
      "balance_loss_clip": 1.03475833,
      "balance_loss_mlp": 1.01954174,
      "epoch": 0.7636254321358785,
      "flos": 23178434400000.0,
      "grad_norm": 2.4403303197931048,
      "language_loss": 0.69365031,
      "learning_rate": 5.579098970568574e-07,
      "loss": 0.71500975,
      "num_input_tokens_seen": 273927415,
      "step": 12701,
      "time_per_iteration": 2.4446098804473877
    },
    {
      "auxiliary_loss_clip": 0.01077639,
      "auxiliary_loss_mlp": 0.01026977,
      "balance_loss_clip": 1.03890824,
      "balance_loss_mlp": 1.01543128,
      "epoch": 0.7636855553885465,
      "flos": 21325229032320.0,
      "grad_norm": 1.52687546361391,
      "language_loss": 0.64216506,
      "learning_rate": 5.576400710039508e-07,
      "loss": 0.66321123,
      "num_input_tokens_seen": 273946690,
      "step": 12702,
      "time_per_iteration": 2.5328667163848877
    },
    {
      "auxiliary_loss_clip": 0.01073337,
      "auxiliary_loss_mlp": 0.01033019,
      "balance_loss_clip": 1.03607917,
      "balance_loss_mlp": 1.02088976,
      "epoch": 0.7637456786412145,
      "flos": 28658079849600.0,
      "grad_norm": 1.9740754643474494,
      "language_loss": 0.6529851,
      "learning_rate": 5.57370299645477e-07,
      "loss": 0.6740486,
      "num_input_tokens_seen": 273966870,
      "step": 12703,
      "time_per_iteration": 2.617490768432617
    },
    {
      "auxiliary_loss_clip": 0.01081056,
      "auxiliary_loss_mlp": 0.01025523,
      "balance_loss_clip": 1.03624392,
      "balance_loss_mlp": 1.01369786,
      "epoch": 0.7638058018938825,
      "flos": 21907269014400.0,
      "grad_norm": 1.7978709100130605,
      "language_loss": 0.83977789,
      "learning_rate": 5.571005829916668e-07,
      "loss": 0.86084366,
      "num_input_tokens_seen": 273986360,
      "step": 12704,
      "time_per_iteration": 2.5362422466278076
    },
    {
      "auxiliary_loss_clip": 0.01083937,
      "auxiliary_loss_mlp": 0.01030711,
      "balance_loss_clip": 1.03655303,
      "balance_loss_mlp": 1.01842034,
      "epoch": 0.7638659251465504,
      "flos": 29643899592960.0,
      "grad_norm": 1.4234525255075872,
      "language_loss": 0.67991906,
      "learning_rate": 5.568309210527469e-07,
      "loss": 0.70106554,
      "num_input_tokens_seen": 274009745,
      "step": 12705,
      "time_per_iteration": 2.635594129562378
    },
    {
      "auxiliary_loss_clip": 0.01078872,
      "auxiliary_loss_mlp": 0.01029255,
      "balance_loss_clip": 1.03464711,
      "balance_loss_mlp": 1.0170902,
      "epoch": 0.7639260483992184,
      "flos": 26141172929280.0,
      "grad_norm": 1.6579830755573166,
      "language_loss": 0.74411571,
      "learning_rate": 5.565613138389427e-07,
      "loss": 0.76519704,
      "num_input_tokens_seen": 274028775,
      "step": 12706,
      "time_per_iteration": 4.063490867614746
    },
    {
      "auxiliary_loss_clip": 0.01089691,
      "auxiliary_loss_mlp": 0.01034201,
      "balance_loss_clip": 1.03552985,
      "balance_loss_mlp": 1.02163589,
      "epoch": 0.7639861716518863,
      "flos": 20156695781760.0,
      "grad_norm": 1.874713618698413,
      "language_loss": 0.78395748,
      "learning_rate": 5.562917613604781e-07,
      "loss": 0.8051964,
      "num_input_tokens_seen": 274047520,
      "step": 12707,
      "time_per_iteration": 2.503784656524658
    },
    {
      "auxiliary_loss_clip": 0.01077469,
      "auxiliary_loss_mlp": 0.01026579,
      "balance_loss_clip": 1.03400552,
      "balance_loss_mlp": 1.01431274,
      "epoch": 0.7640462949045543,
      "flos": 18583125793920.0,
      "grad_norm": 1.676659069403074,
      "language_loss": 0.79733527,
      "learning_rate": 5.560222636275751e-07,
      "loss": 0.81837571,
      "num_input_tokens_seen": 274065350,
      "step": 12708,
      "time_per_iteration": 2.499134063720703
    },
    {
      "auxiliary_loss_clip": 0.01019374,
      "auxiliary_loss_mlp": 0.01002838,
      "balance_loss_clip": 1.00931716,
      "balance_loss_mlp": 1.00164604,
      "epoch": 0.7641064181572224,
      "flos": 68321991646080.0,
      "grad_norm": 0.8180812717830693,
      "language_loss": 0.56488198,
      "learning_rate": 5.557528206504521e-07,
      "loss": 0.58510411,
      "num_input_tokens_seen": 274122315,
      "step": 12709,
      "time_per_iteration": 4.506904602050781
    },
    {
      "auxiliary_loss_clip": 0.0109506,
      "auxiliary_loss_mlp": 0.01037192,
      "balance_loss_clip": 1.03596592,
      "balance_loss_mlp": 1.02366757,
      "epoch": 0.7641665414098903,
      "flos": 17968982031360.0,
      "grad_norm": 17.013257312367852,
      "language_loss": 0.63530135,
      "learning_rate": 5.554834324393271e-07,
      "loss": 0.6566239,
      "num_input_tokens_seen": 274140555,
      "step": 12710,
      "time_per_iteration": 3.945692539215088
    },
    {
      "auxiliary_loss_clip": 0.01060167,
      "auxiliary_loss_mlp": 0.0078507,
      "balance_loss_clip": 1.03596961,
      "balance_loss_mlp": 1.00793052,
      "epoch": 0.7642266646625583,
      "flos": 21252078984960.0,
      "grad_norm": 1.910032240154901,
      "language_loss": 0.64281499,
      "learning_rate": 5.552140990044154e-07,
      "loss": 0.6612674,
      "num_input_tokens_seen": 274161125,
      "step": 12711,
      "time_per_iteration": 2.606048583984375
    },
    {
      "auxiliary_loss_clip": 0.01082423,
      "auxiliary_loss_mlp": 0.01030959,
      "balance_loss_clip": 1.03513443,
      "balance_loss_mlp": 1.01938951,
      "epoch": 0.7642867879152262,
      "flos": 22747794243840.0,
      "grad_norm": 1.7290597510589432,
      "language_loss": 0.73101598,
      "learning_rate": 5.549448203559293e-07,
      "loss": 0.75214982,
      "num_input_tokens_seen": 274180835,
      "step": 12712,
      "time_per_iteration": 2.5225303173065186
    },
    {
      "auxiliary_loss_clip": 0.01070185,
      "auxiliary_loss_mlp": 0.01028432,
      "balance_loss_clip": 1.0356431,
      "balance_loss_mlp": 1.01724386,
      "epoch": 0.7643469111678942,
      "flos": 23332132696320.0,
      "grad_norm": 1.4822014207304208,
      "language_loss": 0.80438852,
      "learning_rate": 5.546755965040804e-07,
      "loss": 0.82537472,
      "num_input_tokens_seen": 274201190,
      "step": 12713,
      "time_per_iteration": 2.599271535873413
    },
    {
      "auxiliary_loss_clip": 0.010964,
      "auxiliary_loss_mlp": 0.00784186,
      "balance_loss_clip": 1.03498697,
      "balance_loss_mlp": 1.01001644,
      "epoch": 0.7644070344205621,
      "flos": 19857092440320.0,
      "grad_norm": 2.208101730158308,
      "language_loss": 0.83271867,
      "learning_rate": 5.544064274590776e-07,
      "loss": 0.85152459,
      "num_input_tokens_seen": 274217595,
      "step": 12714,
      "time_per_iteration": 2.4917168617248535
    },
    {
      "auxiliary_loss_clip": 0.01099611,
      "auxiliary_loss_mlp": 0.01034801,
      "balance_loss_clip": 1.03850365,
      "balance_loss_mlp": 1.02212274,
      "epoch": 0.7644671576732301,
      "flos": 22090628966400.0,
      "grad_norm": 1.560705328145743,
      "language_loss": 0.72881591,
      "learning_rate": 5.541373132311287e-07,
      "loss": 0.75015998,
      "num_input_tokens_seen": 274237885,
      "step": 12715,
      "time_per_iteration": 2.5132856369018555
    },
    {
      "auxiliary_loss_clip": 0.01065237,
      "auxiliary_loss_mlp": 0.01028269,
      "balance_loss_clip": 1.03436804,
      "balance_loss_mlp": 1.01609755,
      "epoch": 0.7645272809258981,
      "flos": 25481421872640.0,
      "grad_norm": 1.5405521708423653,
      "language_loss": 0.6311084,
      "learning_rate": 5.538682538304376e-07,
      "loss": 0.65204346,
      "num_input_tokens_seen": 274258820,
      "step": 12716,
      "time_per_iteration": 3.9953503608703613
    },
    {
      "auxiliary_loss_clip": 0.01109554,
      "auxiliary_loss_mlp": 0.01032222,
      "balance_loss_clip": 1.03685713,
      "balance_loss_mlp": 1.01895404,
      "epoch": 0.7645874041785661,
      "flos": 21541877913600.0,
      "grad_norm": 1.4798090882089021,
      "language_loss": 0.79953456,
      "learning_rate": 5.535992492672068e-07,
      "loss": 0.8209523,
      "num_input_tokens_seen": 274278835,
      "step": 12717,
      "time_per_iteration": 2.4645755290985107
    },
    {
      "auxiliary_loss_clip": 0.01105539,
      "auxiliary_loss_mlp": 0.01033144,
      "balance_loss_clip": 1.0371151,
      "balance_loss_mlp": 1.02106762,
      "epoch": 0.764647527431234,
      "flos": 20630896156800.0,
      "grad_norm": 2.2484592947020223,
      "language_loss": 0.66562712,
      "learning_rate": 5.53330299551638e-07,
      "loss": 0.68701398,
      "num_input_tokens_seen": 274297110,
      "step": 12718,
      "time_per_iteration": 2.4482691287994385
    },
    {
      "auxiliary_loss_clip": 0.01066899,
      "auxiliary_loss_mlp": 0.01033502,
      "balance_loss_clip": 1.0361588,
      "balance_loss_mlp": 1.02221847,
      "epoch": 0.764707650683902,
      "flos": 21434074220160.0,
      "grad_norm": 1.9402739293904379,
      "language_loss": 0.7709583,
      "learning_rate": 5.530614046939286e-07,
      "loss": 0.79196233,
      "num_input_tokens_seen": 274315610,
      "step": 12719,
      "time_per_iteration": 2.542832851409912
    },
    {
      "auxiliary_loss_clip": 0.01105208,
      "auxiliary_loss_mlp": 0.01027923,
      "balance_loss_clip": 1.03555107,
      "balance_loss_mlp": 1.01519752,
      "epoch": 0.7647677739365699,
      "flos": 22711201263360.0,
      "grad_norm": 1.7032412327501656,
      "language_loss": 0.69882929,
      "learning_rate": 5.527925647042754e-07,
      "loss": 0.7201606,
      "num_input_tokens_seen": 274333975,
      "step": 12720,
      "time_per_iteration": 2.4664902687072754
    },
    {
      "auxiliary_loss_clip": 0.01072911,
      "auxiliary_loss_mlp": 0.01035403,
      "balance_loss_clip": 1.03581071,
      "balance_loss_mlp": 1.0236429,
      "epoch": 0.7648278971892379,
      "flos": 21324115710720.0,
      "grad_norm": 2.3858028766462955,
      "language_loss": 0.73863447,
      "learning_rate": 5.52523779592875e-07,
      "loss": 0.75971758,
      "num_input_tokens_seen": 274353695,
      "step": 12721,
      "time_per_iteration": 2.5325324535369873
    },
    {
      "auxiliary_loss_clip": 0.01066484,
      "auxiliary_loss_mlp": 0.01030789,
      "balance_loss_clip": 1.03545034,
      "balance_loss_mlp": 1.01856399,
      "epoch": 0.764888020441906,
      "flos": 20667345482880.0,
      "grad_norm": 1.9260520949013944,
      "language_loss": 0.73537505,
      "learning_rate": 5.522550493699163e-07,
      "loss": 0.75634778,
      "num_input_tokens_seen": 274371120,
      "step": 12722,
      "time_per_iteration": 2.547100782394409
    },
    {
      "auxiliary_loss_clip": 0.0109502,
      "auxiliary_loss_mlp": 0.01033407,
      "balance_loss_clip": 1.03529525,
      "balance_loss_mlp": 1.02128899,
      "epoch": 0.7649481436945739,
      "flos": 25082526360960.0,
      "grad_norm": 1.7951258048533811,
      "language_loss": 0.73822641,
      "learning_rate": 5.519863740455912e-07,
      "loss": 0.75951064,
      "num_input_tokens_seen": 274389665,
      "step": 12723,
      "time_per_iteration": 2.5223472118377686
    },
    {
      "auxiliary_loss_clip": 0.01105881,
      "auxiliary_loss_mlp": 0.01029962,
      "balance_loss_clip": 1.03490245,
      "balance_loss_mlp": 1.01742673,
      "epoch": 0.7650082669472419,
      "flos": 24900890261760.0,
      "grad_norm": 1.713732460691786,
      "language_loss": 0.73062921,
      "learning_rate": 5.517177536300881e-07,
      "loss": 0.75198764,
      "num_input_tokens_seen": 274408750,
      "step": 12724,
      "time_per_iteration": 2.4960367679595947
    },
    {
      "auxiliary_loss_clip": 0.01091411,
      "auxiliary_loss_mlp": 0.01026264,
      "balance_loss_clip": 1.03589463,
      "balance_loss_mlp": 1.01452219,
      "epoch": 0.7650683901999098,
      "flos": 14647388676480.0,
      "grad_norm": 1.8584106642280627,
      "language_loss": 0.84055841,
      "learning_rate": 5.514491881335935e-07,
      "loss": 0.86173517,
      "num_input_tokens_seen": 274424600,
      "step": 12725,
      "time_per_iteration": 2.4505133628845215
    },
    {
      "auxiliary_loss_clip": 0.0106452,
      "auxiliary_loss_mlp": 0.01033449,
      "balance_loss_clip": 1.03541112,
      "balance_loss_mlp": 1.0204668,
      "epoch": 0.7651285134525778,
      "flos": 26352434770560.0,
      "grad_norm": 1.7805843509886905,
      "language_loss": 0.77419645,
      "learning_rate": 5.511806775662901e-07,
      "loss": 0.79517615,
      "num_input_tokens_seen": 274443075,
      "step": 12726,
      "time_per_iteration": 2.6109695434570312
    },
    {
      "auxiliary_loss_clip": 0.01095479,
      "auxiliary_loss_mlp": 0.01032449,
      "balance_loss_clip": 1.03606594,
      "balance_loss_mlp": 1.02020597,
      "epoch": 0.7651886367052457,
      "flos": 26646866553600.0,
      "grad_norm": 2.5005385961415616,
      "language_loss": 0.70800877,
      "learning_rate": 5.509122219383615e-07,
      "loss": 0.72928798,
      "num_input_tokens_seen": 274463240,
      "step": 12727,
      "time_per_iteration": 2.5367915630340576
    },
    {
      "auxiliary_loss_clip": 0.01101292,
      "auxiliary_loss_mlp": 0.01026805,
      "balance_loss_clip": 1.03494048,
      "balance_loss_mlp": 1.01530194,
      "epoch": 0.7652487599579137,
      "flos": 25702847262720.0,
      "grad_norm": 1.6851033135092783,
      "language_loss": 0.79577768,
      "learning_rate": 5.506438212599864e-07,
      "loss": 0.81705862,
      "num_input_tokens_seen": 274482750,
      "step": 12728,
      "time_per_iteration": 2.50215220451355
    },
    {
      "auxiliary_loss_clip": 0.01107681,
      "auxiliary_loss_mlp": 0.010278,
      "balance_loss_clip": 1.03760886,
      "balance_loss_mlp": 1.01517534,
      "epoch": 0.7653088832105817,
      "flos": 28585576247040.0,
      "grad_norm": 1.8695902203833261,
      "language_loss": 0.55167186,
      "learning_rate": 5.503754755413424e-07,
      "loss": 0.57302666,
      "num_input_tokens_seen": 274503545,
      "step": 12729,
      "time_per_iteration": 2.5152390003204346
    },
    {
      "auxiliary_loss_clip": 0.01081257,
      "auxiliary_loss_mlp": 0.00784637,
      "balance_loss_clip": 1.03444195,
      "balance_loss_mlp": 1.01027954,
      "epoch": 0.7653690064632497,
      "flos": 23366750428800.0,
      "grad_norm": 1.568604240183008,
      "language_loss": 0.777421,
      "learning_rate": 5.501071847926055e-07,
      "loss": 0.79607999,
      "num_input_tokens_seen": 274523825,
      "step": 12730,
      "time_per_iteration": 2.570303201675415
    },
    {
      "auxiliary_loss_clip": 0.01100351,
      "auxiliary_loss_mlp": 0.01036867,
      "balance_loss_clip": 1.04006398,
      "balance_loss_mlp": 1.02396214,
      "epoch": 0.7654291297159176,
      "flos": 15773905992960.0,
      "grad_norm": 1.6128678807957468,
      "language_loss": 0.68976015,
      "learning_rate": 5.498389490239495e-07,
      "loss": 0.71113229,
      "num_input_tokens_seen": 274541625,
      "step": 12731,
      "time_per_iteration": 2.45709490776062
    },
    {
      "auxiliary_loss_clip": 0.01108371,
      "auxiliary_loss_mlp": 0.01030422,
      "balance_loss_clip": 1.03767419,
      "balance_loss_mlp": 1.01837552,
      "epoch": 0.7654892529685856,
      "flos": 18033800123520.0,
      "grad_norm": 2.400353312672418,
      "language_loss": 0.70332956,
      "learning_rate": 5.495707682455471e-07,
      "loss": 0.7247175,
      "num_input_tokens_seen": 274557580,
      "step": 12732,
      "time_per_iteration": 2.4380133152008057
    },
    {
      "auxiliary_loss_clip": 0.01086896,
      "auxiliary_loss_mlp": 0.01028199,
      "balance_loss_clip": 1.0361433,
      "balance_loss_mlp": 1.01570606,
      "epoch": 0.7655493762212535,
      "flos": 27236017428480.0,
      "grad_norm": 1.5764152119765142,
      "language_loss": 0.78390157,
      "learning_rate": 5.493026424675653e-07,
      "loss": 0.80505252,
      "num_input_tokens_seen": 274578135,
      "step": 12733,
      "time_per_iteration": 2.5690832138061523
    },
    {
      "auxiliary_loss_clip": 0.01094452,
      "auxiliary_loss_mlp": 0.01031725,
      "balance_loss_clip": 1.03615165,
      "balance_loss_mlp": 1.01942253,
      "epoch": 0.7656094994739215,
      "flos": 20773964027520.0,
      "grad_norm": 1.935481264600818,
      "language_loss": 0.7720449,
      "learning_rate": 5.490345717001726e-07,
      "loss": 0.79330671,
      "num_input_tokens_seen": 274595655,
      "step": 12734,
      "time_per_iteration": 2.5017309188842773
    },
    {
      "auxiliary_loss_clip": 0.01082584,
      "auxiliary_loss_mlp": 0.01029396,
      "balance_loss_clip": 1.03818333,
      "balance_loss_mlp": 1.01575291,
      "epoch": 0.7656696227265896,
      "flos": 23039245198080.0,
      "grad_norm": 1.5631254803635872,
      "language_loss": 0.73276001,
      "learning_rate": 5.48766555953535e-07,
      "loss": 0.75387985,
      "num_input_tokens_seen": 274616305,
      "step": 12735,
      "time_per_iteration": 2.5621604919433594
    },
    {
      "auxiliary_loss_clip": 0.01084146,
      "auxiliary_loss_mlp": 0.01029769,
      "balance_loss_clip": 1.03876913,
      "balance_loss_mlp": 1.01751447,
      "epoch": 0.7657297459792575,
      "flos": 27525636789120.0,
      "grad_norm": 1.986157288815839,
      "language_loss": 0.72643936,
      "learning_rate": 5.484985952378145e-07,
      "loss": 0.7475785,
      "num_input_tokens_seen": 274638110,
      "step": 12736,
      "time_per_iteration": 2.5855581760406494
    },
    {
      "auxiliary_loss_clip": 0.01097447,
      "auxiliary_loss_mlp": 0.00786265,
      "balance_loss_clip": 1.03972292,
      "balance_loss_mlp": 1.01157999,
      "epoch": 0.7657898692319255,
      "flos": 17128456801920.0,
      "grad_norm": 2.320909965491637,
      "language_loss": 0.77772641,
      "learning_rate": 5.482306895631728e-07,
      "loss": 0.79656363,
      "num_input_tokens_seen": 274656565,
      "step": 12737,
      "time_per_iteration": 2.4691450595855713
    },
    {
      "auxiliary_loss_clip": 0.01079324,
      "auxiliary_loss_mlp": 0.01032629,
      "balance_loss_clip": 1.03381824,
      "balance_loss_mlp": 1.01985013,
      "epoch": 0.7658499924845934,
      "flos": 21465747037440.0,
      "grad_norm": 1.6363883816890863,
      "language_loss": 0.76772445,
      "learning_rate": 5.479628389397699e-07,
      "loss": 0.78884399,
      "num_input_tokens_seen": 274674215,
      "step": 12738,
      "time_per_iteration": 2.4983088970184326
    },
    {
      "auxiliary_loss_clip": 0.01085627,
      "auxiliary_loss_mlp": 0.01029605,
      "balance_loss_clip": 1.03536463,
      "balance_loss_mlp": 1.01669431,
      "epoch": 0.7659101157372614,
      "flos": 29496665744640.0,
      "grad_norm": 1.860170500672601,
      "language_loss": 0.62653166,
      "learning_rate": 5.476950433777603e-07,
      "loss": 0.64768398,
      "num_input_tokens_seen": 274693445,
      "step": 12739,
      "time_per_iteration": 2.5684781074523926
    },
    {
      "auxiliary_loss_clip": 0.0110796,
      "auxiliary_loss_mlp": 0.01034303,
      "balance_loss_clip": 1.03784513,
      "balance_loss_mlp": 1.0214045,
      "epoch": 0.7659702389899293,
      "flos": 18551812112640.0,
      "grad_norm": 1.8685906667604195,
      "language_loss": 0.7918427,
      "learning_rate": 5.474273028873004e-07,
      "loss": 0.81326532,
      "num_input_tokens_seen": 274712815,
      "step": 12740,
      "time_per_iteration": 2.4931511878967285
    },
    {
      "auxiliary_loss_clip": 0.01095315,
      "auxiliary_loss_mlp": 0.01034415,
      "balance_loss_clip": 1.03446615,
      "balance_loss_mlp": 1.02142143,
      "epoch": 0.7660303622425974,
      "flos": 23549176627200.0,
      "grad_norm": 1.58802686115676,
      "language_loss": 0.65662527,
      "learning_rate": 5.471596174785429e-07,
      "loss": 0.67792255,
      "num_input_tokens_seen": 274732690,
      "step": 12741,
      "time_per_iteration": 2.5480456352233887
    },
    {
      "auxiliary_loss_clip": 0.01080165,
      "auxiliary_loss_mlp": 0.01029238,
      "balance_loss_clip": 1.03564978,
      "balance_loss_mlp": 1.01629806,
      "epoch": 0.7660904854952653,
      "flos": 18916736336640.0,
      "grad_norm": 1.4803089009429498,
      "language_loss": 0.76006705,
      "learning_rate": 5.468919871616386e-07,
      "loss": 0.78116113,
      "num_input_tokens_seen": 274752460,
      "step": 12742,
      "time_per_iteration": 2.5261166095733643
    },
    {
      "auxiliary_loss_clip": 0.01080521,
      "auxiliary_loss_mlp": 0.01030544,
      "balance_loss_clip": 1.03672504,
      "balance_loss_mlp": 1.01892686,
      "epoch": 0.7661506087479333,
      "flos": 23147515768320.0,
      "grad_norm": 1.3397860004290465,
      "language_loss": 0.76493883,
      "learning_rate": 5.46624411946736e-07,
      "loss": 0.78604949,
      "num_input_tokens_seen": 274773070,
      "step": 12743,
      "time_per_iteration": 2.525453567504883
    },
    {
      "auxiliary_loss_clip": 0.01080223,
      "auxiliary_loss_mlp": 0.0103319,
      "balance_loss_clip": 1.03354609,
      "balance_loss_mlp": 1.02110863,
      "epoch": 0.7662107320006012,
      "flos": 17565776887680.0,
      "grad_norm": 1.938449710555847,
      "language_loss": 0.74933881,
      "learning_rate": 5.463568918439805e-07,
      "loss": 0.77047288,
      "num_input_tokens_seen": 274790220,
      "step": 12744,
      "time_per_iteration": 3.8911166191101074
    },
    {
      "auxiliary_loss_clip": 0.01096402,
      "auxiliary_loss_mlp": 0.01030414,
      "balance_loss_clip": 1.03625226,
      "balance_loss_mlp": 1.01739073,
      "epoch": 0.7662708552532692,
      "flos": 22303075956480.0,
      "grad_norm": 2.2451181301662206,
      "language_loss": 0.71222448,
      "learning_rate": 5.460894268635181e-07,
      "loss": 0.73349261,
      "num_input_tokens_seen": 274805095,
      "step": 12745,
      "time_per_iteration": 2.4627761840820312
    },
    {
      "auxiliary_loss_clip": 0.01090412,
      "auxiliary_loss_mlp": 0.01036736,
      "balance_loss_clip": 1.03370416,
      "balance_loss_mlp": 1.0224185,
      "epoch": 0.7663309785059371,
      "flos": 15742053607680.0,
      "grad_norm": 2.348610615936546,
      "language_loss": 0.76789314,
      "learning_rate": 5.458220170154896e-07,
      "loss": 0.7891646,
      "num_input_tokens_seen": 274821800,
      "step": 12746,
      "time_per_iteration": 2.474148988723755
    },
    {
      "auxiliary_loss_clip": 0.01004799,
      "auxiliary_loss_mlp": 0.01000908,
      "balance_loss_clip": 1.01124573,
      "balance_loss_mlp": 0.99981159,
      "epoch": 0.7663911017586051,
      "flos": 62163312514560.0,
      "grad_norm": 0.6644517547792831,
      "language_loss": 0.56804311,
      "learning_rate": 5.455546623100362e-07,
      "loss": 0.58810019,
      "num_input_tokens_seen": 274886970,
      "step": 12747,
      "time_per_iteration": 3.2208662033081055
    },
    {
      "auxiliary_loss_clip": 0.01103575,
      "auxiliary_loss_mlp": 0.01033398,
      "balance_loss_clip": 1.0360949,
      "balance_loss_mlp": 1.02263284,
      "epoch": 0.7664512250112732,
      "flos": 26506025326080.0,
      "grad_norm": 1.4843315853474126,
      "language_loss": 0.72147238,
      "learning_rate": 5.452873627572956e-07,
      "loss": 0.7428422,
      "num_input_tokens_seen": 274907240,
      "step": 12748,
      "time_per_iteration": 3.8783538341522217
    },
    {
      "auxiliary_loss_clip": 0.01070414,
      "auxiliary_loss_mlp": 0.01030562,
      "balance_loss_clip": 1.03423977,
      "balance_loss_mlp": 1.01757371,
      "epoch": 0.7665113482639411,
      "flos": 16249542912000.0,
      "grad_norm": 2.102250658225341,
      "language_loss": 0.69326097,
      "learning_rate": 5.450201183674052e-07,
      "loss": 0.71427071,
      "num_input_tokens_seen": 274924650,
      "step": 12749,
      "time_per_iteration": 3.987619638442993
    },
    {
      "auxiliary_loss_clip": 0.01095417,
      "auxiliary_loss_mlp": 0.0103108,
      "balance_loss_clip": 1.03593063,
      "balance_loss_mlp": 1.01835406,
      "epoch": 0.7665714715166091,
      "flos": 27197880163200.0,
      "grad_norm": 1.6257675530864106,
      "language_loss": 0.73549503,
      "learning_rate": 5.447529291504967e-07,
      "loss": 0.75676,
      "num_input_tokens_seen": 274944550,
      "step": 12750,
      "time_per_iteration": 2.557943344116211
    },
    {
      "auxiliary_loss_clip": 0.01091594,
      "auxiliary_loss_mlp": 0.01031326,
      "balance_loss_clip": 1.03566647,
      "balance_loss_mlp": 1.01929784,
      "epoch": 0.766631594769277,
      "flos": 21067785279360.0,
      "grad_norm": 2.011661877971584,
      "language_loss": 0.75765753,
      "learning_rate": 5.444857951167026e-07,
      "loss": 0.77888674,
      "num_input_tokens_seen": 274961330,
      "step": 12751,
      "time_per_iteration": 2.4928882122039795
    },
    {
      "auxiliary_loss_clip": 0.01071129,
      "auxiliary_loss_mlp": 0.01036981,
      "balance_loss_clip": 1.0348177,
      "balance_loss_mlp": 1.02401733,
      "epoch": 0.766691718021945,
      "flos": 24097963593600.0,
      "grad_norm": 1.6824234020545232,
      "language_loss": 0.61249399,
      "learning_rate": 5.442187162761537e-07,
      "loss": 0.63357508,
      "num_input_tokens_seen": 274981655,
      "step": 12752,
      "time_per_iteration": 2.5979862213134766
    },
    {
      "auxiliary_loss_clip": 0.01096721,
      "auxiliary_loss_mlp": 0.01031839,
      "balance_loss_clip": 1.03667879,
      "balance_loss_mlp": 1.01891661,
      "epoch": 0.7667518412746129,
      "flos": 23440654661760.0,
      "grad_norm": 2.0845765916472017,
      "language_loss": 0.69597387,
      "learning_rate": 5.439516926389767e-07,
      "loss": 0.71725947,
      "num_input_tokens_seen": 274999970,
      "step": 12753,
      "time_per_iteration": 2.503129005432129
    },
    {
      "auxiliary_loss_clip": 0.01096391,
      "auxiliary_loss_mlp": 0.01036559,
      "balance_loss_clip": 1.03718638,
      "balance_loss_mlp": 1.02427495,
      "epoch": 0.766811964527281,
      "flos": 18148786536960.0,
      "grad_norm": 1.9694118502455005,
      "language_loss": 0.6233899,
      "learning_rate": 5.436847242152971e-07,
      "loss": 0.64471942,
      "num_input_tokens_seen": 275015805,
      "step": 12754,
      "time_per_iteration": 2.4710001945495605
    },
    {
      "auxiliary_loss_clip": 0.01106903,
      "auxiliary_loss_mlp": 0.01031367,
      "balance_loss_clip": 1.0387547,
      "balance_loss_mlp": 1.01960707,
      "epoch": 0.7668720877799489,
      "flos": 19536051657600.0,
      "grad_norm": 2.698192894147036,
      "language_loss": 0.79756403,
      "learning_rate": 5.434178110152401e-07,
      "loss": 0.81894672,
      "num_input_tokens_seen": 275031810,
      "step": 12755,
      "time_per_iteration": 3.812110424041748
    },
    {
      "auxiliary_loss_clip": 0.01104165,
      "auxiliary_loss_mlp": 0.01030413,
      "balance_loss_clip": 1.03604913,
      "balance_loss_mlp": 1.01864743,
      "epoch": 0.7669322110326169,
      "flos": 22674320974080.0,
      "grad_norm": 1.9633767341617367,
      "language_loss": 0.70682645,
      "learning_rate": 5.431509530489242e-07,
      "loss": 0.72817218,
      "num_input_tokens_seen": 275049325,
      "step": 12756,
      "time_per_iteration": 2.470271110534668
    },
    {
      "auxiliary_loss_clip": 0.01095235,
      "auxiliary_loss_mlp": 0.01037207,
      "balance_loss_clip": 1.0366925,
      "balance_loss_mlp": 1.02563167,
      "epoch": 0.7669923342852848,
      "flos": 26469396432000.0,
      "grad_norm": 1.5202718272173454,
      "language_loss": 0.69719505,
      "learning_rate": 5.428841503264706e-07,
      "loss": 0.71851957,
      "num_input_tokens_seen": 275070865,
      "step": 12757,
      "time_per_iteration": 2.5235588550567627
    },
    {
      "auxiliary_loss_clip": 0.01086972,
      "auxiliary_loss_mlp": 0.01037913,
      "balance_loss_clip": 1.03807592,
      "balance_loss_mlp": 1.02490711,
      "epoch": 0.7670524575379528,
      "flos": 22856136641280.0,
      "grad_norm": 1.9461270919814544,
      "language_loss": 0.76210648,
      "learning_rate": 5.426174028579955e-07,
      "loss": 0.7833553,
      "num_input_tokens_seen": 275088015,
      "step": 12758,
      "time_per_iteration": 2.544605016708374
    },
    {
      "auxiliary_loss_clip": 0.01091959,
      "auxiliary_loss_mlp": 0.01034529,
      "balance_loss_clip": 1.03460205,
      "balance_loss_mlp": 1.02249467,
      "epoch": 0.7671125807906207,
      "flos": 22452141398400.0,
      "grad_norm": 1.952729486379786,
      "language_loss": 0.76411438,
      "learning_rate": 5.423507106536156e-07,
      "loss": 0.78537929,
      "num_input_tokens_seen": 275106975,
      "step": 12759,
      "time_per_iteration": 2.484149932861328
    },
    {
      "auxiliary_loss_clip": 0.01083642,
      "auxiliary_loss_mlp": 0.01027516,
      "balance_loss_clip": 1.03435481,
      "balance_loss_mlp": 1.01574993,
      "epoch": 0.7671727040432887,
      "flos": 35371543518720.0,
      "grad_norm": 1.9752245876113859,
      "language_loss": 0.68165565,
      "learning_rate": 5.420840737234425e-07,
      "loss": 0.70276725,
      "num_input_tokens_seen": 275129560,
      "step": 12760,
      "time_per_iteration": 2.643575668334961
    },
    {
      "auxiliary_loss_clip": 0.01084445,
      "auxiliary_loss_mlp": 0.01031331,
      "balance_loss_clip": 1.03582025,
      "balance_loss_mlp": 1.01819992,
      "epoch": 0.7672328272959568,
      "flos": 22494947431680.0,
      "grad_norm": 1.4710674423242924,
      "language_loss": 0.7932936,
      "learning_rate": 5.418174920775871e-07,
      "loss": 0.8144514,
      "num_input_tokens_seen": 275151180,
      "step": 12761,
      "time_per_iteration": 2.5579490661621094
    },
    {
      "auxiliary_loss_clip": 0.01081349,
      "auxiliary_loss_mlp": 0.01030936,
      "balance_loss_clip": 1.03564322,
      "balance_loss_mlp": 1.01844871,
      "epoch": 0.7672929505486247,
      "flos": 22815557251200.0,
      "grad_norm": 1.8225944135678371,
      "language_loss": 0.6625793,
      "learning_rate": 5.415509657261589e-07,
      "loss": 0.68370211,
      "num_input_tokens_seen": 275170605,
      "step": 12762,
      "time_per_iteration": 2.5468802452087402
    },
    {
      "auxiliary_loss_clip": 0.0109645,
      "auxiliary_loss_mlp": 0.01030692,
      "balance_loss_clip": 1.03634048,
      "balance_loss_mlp": 1.01748323,
      "epoch": 0.7673530738012927,
      "flos": 20338834671360.0,
      "grad_norm": 1.7772141296572048,
      "language_loss": 0.74057066,
      "learning_rate": 5.412844946792639e-07,
      "loss": 0.76184213,
      "num_input_tokens_seen": 275188750,
      "step": 12763,
      "time_per_iteration": 2.515439748764038
    },
    {
      "auxiliary_loss_clip": 0.0108326,
      "auxiliary_loss_mlp": 0.01030305,
      "balance_loss_clip": 1.03791225,
      "balance_loss_mlp": 1.01815188,
      "epoch": 0.7674131970539606,
      "flos": 34933576988160.0,
      "grad_norm": 1.417648606839019,
      "language_loss": 0.70592403,
      "learning_rate": 5.410180789470067e-07,
      "loss": 0.72705966,
      "num_input_tokens_seen": 275211365,
      "step": 12764,
      "time_per_iteration": 2.6685361862182617
    },
    {
      "auxiliary_loss_clip": 0.01095754,
      "auxiliary_loss_mlp": 0.0103191,
      "balance_loss_clip": 1.0371325,
      "balance_loss_mlp": 1.01986945,
      "epoch": 0.7674733203066286,
      "flos": 28328850766080.0,
      "grad_norm": 1.6056250925169935,
      "language_loss": 0.69558728,
      "learning_rate": 5.40751718539491e-07,
      "loss": 0.71686399,
      "num_input_tokens_seen": 275231670,
      "step": 12765,
      "time_per_iteration": 2.561206102371216
    },
    {
      "auxiliary_loss_clip": 0.01082477,
      "auxiliary_loss_mlp": 0.01031815,
      "balance_loss_clip": 1.03390765,
      "balance_loss_mlp": 1.02079964,
      "epoch": 0.7675334435592965,
      "flos": 16289727252480.0,
      "grad_norm": 1.7213541813042743,
      "language_loss": 0.60957682,
      "learning_rate": 5.404854134668162e-07,
      "loss": 0.63071978,
      "num_input_tokens_seen": 275249425,
      "step": 12766,
      "time_per_iteration": 2.4870307445526123
    },
    {
      "auxiliary_loss_clip": 0.01008894,
      "auxiliary_loss_mlp": 0.0100161,
      "balance_loss_clip": 1.02275181,
      "balance_loss_mlp": 1.000144,
      "epoch": 0.7675935668119646,
      "flos": 64826232220800.0,
      "grad_norm": 0.7298070534310702,
      "language_loss": 0.60776448,
      "learning_rate": 5.402191637390803e-07,
      "loss": 0.62786949,
      "num_input_tokens_seen": 275312485,
      "step": 12767,
      "time_per_iteration": 3.2905070781707764
    },
    {
      "auxiliary_loss_clip": 0.01082129,
      "auxiliary_loss_mlp": 0.01024546,
      "balance_loss_clip": 1.03700757,
      "balance_loss_mlp": 1.01352537,
      "epoch": 0.7676536900646325,
      "flos": 22675398382080.0,
      "grad_norm": 1.737800582710968,
      "language_loss": 0.69661635,
      "learning_rate": 5.399529693663801e-07,
      "loss": 0.71768308,
      "num_input_tokens_seen": 275331680,
      "step": 12768,
      "time_per_iteration": 2.5282299518585205
    },
    {
      "auxiliary_loss_clip": 0.01101051,
      "auxiliary_loss_mlp": 0.01036548,
      "balance_loss_clip": 1.03893924,
      "balance_loss_mlp": 1.02336383,
      "epoch": 0.7677138133173005,
      "flos": 26939682224640.0,
      "grad_norm": 2.630385377830171,
      "language_loss": 0.70802271,
      "learning_rate": 5.3968683035881e-07,
      "loss": 0.72939867,
      "num_input_tokens_seen": 275351615,
      "step": 12769,
      "time_per_iteration": 2.5584893226623535
    },
    {
      "auxiliary_loss_clip": 0.01096115,
      "auxiliary_loss_mlp": 0.01028763,
      "balance_loss_clip": 1.03635037,
      "balance_loss_mlp": 1.01615679,
      "epoch": 0.7677739365699684,
      "flos": 23799545400960.0,
      "grad_norm": 2.493640318957474,
      "language_loss": 0.80463886,
      "learning_rate": 5.394207467264611e-07,
      "loss": 0.82588762,
      "num_input_tokens_seen": 275368815,
      "step": 12770,
      "time_per_iteration": 2.497979164123535
    },
    {
      "auxiliary_loss_clip": 0.01067019,
      "auxiliary_loss_mlp": 0.0104101,
      "balance_loss_clip": 1.0328331,
      "balance_loss_mlp": 1.0284096,
      "epoch": 0.7678340598226364,
      "flos": 34455497944320.0,
      "grad_norm": 1.6466426147155542,
      "language_loss": 0.78765714,
      "learning_rate": 5.391547184794245e-07,
      "loss": 0.8087374,
      "num_input_tokens_seen": 275389345,
      "step": 12771,
      "time_per_iteration": 2.6543772220611572
    },
    {
      "auxiliary_loss_clip": 0.01103864,
      "auxiliary_loss_mlp": 0.01030647,
      "balance_loss_clip": 1.03500128,
      "balance_loss_mlp": 1.0188216,
      "epoch": 0.7678941830753043,
      "flos": 23841740903040.0,
      "grad_norm": 1.3475323037838656,
      "language_loss": 0.68409783,
      "learning_rate": 5.388887456277876e-07,
      "loss": 0.70544291,
      "num_input_tokens_seen": 275411240,
      "step": 12772,
      "time_per_iteration": 2.477107524871826
    },
    {
      "auxiliary_loss_clip": 0.01090038,
      "auxiliary_loss_mlp": 0.01024566,
      "balance_loss_clip": 1.03542042,
      "balance_loss_mlp": 1.01293683,
      "epoch": 0.7679543063279723,
      "flos": 25410929431680.0,
      "grad_norm": 1.5253605028037867,
      "language_loss": 0.73342824,
      "learning_rate": 5.386228281816349e-07,
      "loss": 0.7545743,
      "num_input_tokens_seen": 275432010,
      "step": 12773,
      "time_per_iteration": 2.5394811630249023
    },
    {
      "auxiliary_loss_clip": 0.01065448,
      "auxiliary_loss_mlp": 0.01029136,
      "balance_loss_clip": 1.03290677,
      "balance_loss_mlp": 1.01801968,
      "epoch": 0.7680144295806404,
      "flos": 27962382257280.0,
      "grad_norm": 1.9915132351779838,
      "language_loss": 0.80785263,
      "learning_rate": 5.383569661510512e-07,
      "loss": 0.82879841,
      "num_input_tokens_seen": 275453710,
      "step": 12774,
      "time_per_iteration": 2.5962765216827393
    },
    {
      "auxiliary_loss_clip": 0.01097178,
      "auxiliary_loss_mlp": 0.0078436,
      "balance_loss_clip": 1.03836524,
      "balance_loss_mlp": 1.00978673,
      "epoch": 0.7680745528333083,
      "flos": 20412810731520.0,
      "grad_norm": 1.6172900178557381,
      "language_loss": 0.69950461,
      "learning_rate": 5.380911595461177e-07,
      "loss": 0.71832001,
      "num_input_tokens_seen": 275472915,
      "step": 12775,
      "time_per_iteration": 2.5239555835723877
    },
    {
      "auxiliary_loss_clip": 0.00995853,
      "auxiliary_loss_mlp": 0.01007484,
      "balance_loss_clip": 1.01503897,
      "balance_loss_mlp": 1.00637567,
      "epoch": 0.7681346760859763,
      "flos": 68401103351040.0,
      "grad_norm": 0.7051962527432869,
      "language_loss": 0.56812096,
      "learning_rate": 5.378254083769147e-07,
      "loss": 0.58815432,
      "num_input_tokens_seen": 275534785,
      "step": 12776,
      "time_per_iteration": 3.2369422912597656
    },
    {
      "auxiliary_loss_clip": 0.01091717,
      "auxiliary_loss_mlp": 0.01035192,
      "balance_loss_clip": 1.03489566,
      "balance_loss_mlp": 1.023283,
      "epoch": 0.7681947993386442,
      "flos": 21251468453760.0,
      "grad_norm": 1.8868067745225896,
      "language_loss": 0.73809528,
      "learning_rate": 5.375597126535188e-07,
      "loss": 0.75936431,
      "num_input_tokens_seen": 275553205,
      "step": 12777,
      "time_per_iteration": 2.5061800479888916
    },
    {
      "auxiliary_loss_clip": 0.01075004,
      "auxiliary_loss_mlp": 0.01031721,
      "balance_loss_clip": 1.03539109,
      "balance_loss_mlp": 1.02010453,
      "epoch": 0.7682549225913122,
      "flos": 21397696721280.0,
      "grad_norm": 2.1312070870735105,
      "language_loss": 0.70332146,
      "learning_rate": 5.372940723860043e-07,
      "loss": 0.72438872,
      "num_input_tokens_seen": 275571490,
      "step": 12778,
      "time_per_iteration": 2.5399656295776367
    },
    {
      "auxiliary_loss_clip": 0.01091827,
      "auxiliary_loss_mlp": 0.01030114,
      "balance_loss_clip": 1.03885674,
      "balance_loss_mlp": 1.01872945,
      "epoch": 0.7683150458439801,
      "flos": 23038921975680.0,
      "grad_norm": 1.7470664705128751,
      "language_loss": 0.70290995,
      "learning_rate": 5.37028487584446e-07,
      "loss": 0.72412932,
      "num_input_tokens_seen": 275589665,
      "step": 12779,
      "time_per_iteration": 2.5186898708343506
    },
    {
      "auxiliary_loss_clip": 0.01083231,
      "auxiliary_loss_mlp": 0.01028435,
      "balance_loss_clip": 1.03877604,
      "balance_loss_mlp": 1.01615,
      "epoch": 0.7683751690966482,
      "flos": 67332397996800.0,
      "grad_norm": 1.5940936879832504,
      "language_loss": 0.59133524,
      "learning_rate": 5.367629582589133e-07,
      "loss": 0.61245185,
      "num_input_tokens_seen": 275615605,
      "step": 12780,
      "time_per_iteration": 2.940070629119873
    },
    {
      "auxiliary_loss_clip": 0.01098765,
      "auxiliary_loss_mlp": 0.01041159,
      "balance_loss_clip": 1.0367558,
      "balance_loss_mlp": 1.02652001,
      "epoch": 0.7684352923493161,
      "flos": 21798890703360.0,
      "grad_norm": 1.7423012194162606,
      "language_loss": 0.68012071,
      "learning_rate": 5.364974844194759e-07,
      "loss": 0.70151997,
      "num_input_tokens_seen": 275634965,
      "step": 12781,
      "time_per_iteration": 2.50557017326355
    },
    {
      "auxiliary_loss_clip": 0.01062483,
      "auxiliary_loss_mlp": 0.01029539,
      "balance_loss_clip": 1.03486705,
      "balance_loss_mlp": 1.01795805,
      "epoch": 0.7684954156019841,
      "flos": 25847603072640.0,
      "grad_norm": 1.4723753684940493,
      "language_loss": 0.79452109,
      "learning_rate": 5.362320660762016e-07,
      "loss": 0.81544131,
      "num_input_tokens_seen": 275655785,
      "step": 12782,
      "time_per_iteration": 2.6484766006469727
    },
    {
      "auxiliary_loss_clip": 0.0108642,
      "auxiliary_loss_mlp": 0.01031303,
      "balance_loss_clip": 1.03751945,
      "balance_loss_mlp": 1.0188334,
      "epoch": 0.768555538854652,
      "flos": 25447378757760.0,
      "grad_norm": 1.6984795874312189,
      "language_loss": 0.66695386,
      "learning_rate": 5.35966703239153e-07,
      "loss": 0.68813109,
      "num_input_tokens_seen": 275676160,
      "step": 12783,
      "time_per_iteration": 3.9207050800323486
    },
    {
      "auxiliary_loss_clip": 0.01083458,
      "auxiliary_loss_mlp": 0.01034819,
      "balance_loss_clip": 1.03577709,
      "balance_loss_mlp": 1.02173567,
      "epoch": 0.76861566210732,
      "flos": 19646369303040.0,
      "grad_norm": 2.0920688623122965,
      "language_loss": 0.69007111,
      "learning_rate": 5.357013959183938e-07,
      "loss": 0.71125388,
      "num_input_tokens_seen": 275695660,
      "step": 12784,
      "time_per_iteration": 2.521996021270752
    },
    {
      "auxiliary_loss_clip": 0.01062042,
      "auxiliary_loss_mlp": 0.0102613,
      "balance_loss_clip": 1.03625476,
      "balance_loss_mlp": 1.01497841,
      "epoch": 0.7686757853599879,
      "flos": 22419032037120.0,
      "grad_norm": 1.6937361435637195,
      "language_loss": 0.80521554,
      "learning_rate": 5.354361441239843e-07,
      "loss": 0.82609725,
      "num_input_tokens_seen": 275714025,
      "step": 12785,
      "time_per_iteration": 2.582792043685913
    },
    {
      "auxiliary_loss_clip": 0.01091544,
      "auxiliary_loss_mlp": 0.01037134,
      "balance_loss_clip": 1.035501,
      "balance_loss_mlp": 1.02240014,
      "epoch": 0.768735908612656,
      "flos": 47774262453120.0,
      "grad_norm": 1.5644328374093424,
      "language_loss": 0.7744211,
      "learning_rate": 5.351709478659836e-07,
      "loss": 0.79570788,
      "num_input_tokens_seen": 275737300,
      "step": 12786,
      "time_per_iteration": 4.225131988525391
    },
    {
      "auxiliary_loss_clip": 0.01104102,
      "auxiliary_loss_mlp": 0.01030218,
      "balance_loss_clip": 1.03560758,
      "balance_loss_mlp": 1.01830935,
      "epoch": 0.7687960318653239,
      "flos": 30263179000320.0,
      "grad_norm": 2.0329388969722126,
      "language_loss": 0.58764386,
      "learning_rate": 5.349058071544468e-07,
      "loss": 0.60898703,
      "num_input_tokens_seen": 275757895,
      "step": 12787,
      "time_per_iteration": 2.5639567375183105
    },
    {
      "auxiliary_loss_clip": 0.01078998,
      "auxiliary_loss_mlp": 0.01029588,
      "balance_loss_clip": 1.03350496,
      "balance_loss_mlp": 1.01707125,
      "epoch": 0.7688561551179919,
      "flos": 19573434737280.0,
      "grad_norm": 1.6256673595017697,
      "language_loss": 0.76087809,
      "learning_rate": 5.346407219994292e-07,
      "loss": 0.78196388,
      "num_input_tokens_seen": 275776745,
      "step": 12788,
      "time_per_iteration": 3.90998911857605
    },
    {
      "auxiliary_loss_clip": 0.01064537,
      "auxiliary_loss_mlp": 0.00784033,
      "balance_loss_clip": 1.03595209,
      "balance_loss_mlp": 1.00815022,
      "epoch": 0.7689162783706599,
      "flos": 22783776693120.0,
      "grad_norm": 1.5657756541470398,
      "language_loss": 0.66660833,
      "learning_rate": 5.343756924109821e-07,
      "loss": 0.685094,
      "num_input_tokens_seen": 275797205,
      "step": 12789,
      "time_per_iteration": 2.6016578674316406
    },
    {
      "auxiliary_loss_clip": 0.01085839,
      "auxiliary_loss_mlp": 0.01033509,
      "balance_loss_clip": 1.03585863,
      "balance_loss_mlp": 1.01990747,
      "epoch": 0.7689764016233278,
      "flos": 34204195416960.0,
      "grad_norm": 2.020502403401883,
      "language_loss": 0.68399131,
      "learning_rate": 5.341107183991553e-07,
      "loss": 0.70518476,
      "num_input_tokens_seen": 275817935,
      "step": 12790,
      "time_per_iteration": 2.640934944152832
    },
    {
      "auxiliary_loss_clip": 0.01079173,
      "auxiliary_loss_mlp": 0.01030039,
      "balance_loss_clip": 1.03602147,
      "balance_loss_mlp": 1.0175879,
      "epoch": 0.7690365248759958,
      "flos": 17274469587840.0,
      "grad_norm": 1.5478718430920804,
      "language_loss": 0.68759018,
      "learning_rate": 5.338457999739969e-07,
      "loss": 0.70868236,
      "num_input_tokens_seen": 275837145,
      "step": 12791,
      "time_per_iteration": 2.5094332695007324
    },
    {
      "auxiliary_loss_clip": 0.01091947,
      "auxiliary_loss_mlp": 0.01031113,
      "balance_loss_clip": 1.03612328,
      "balance_loss_mlp": 1.01939452,
      "epoch": 0.7690966481286637,
      "flos": 18223157646720.0,
      "grad_norm": 1.7912425261392715,
      "language_loss": 0.79689276,
      "learning_rate": 5.335809371455526e-07,
      "loss": 0.81812346,
      "num_input_tokens_seen": 275855705,
      "step": 12792,
      "time_per_iteration": 2.4814069271087646
    },
    {
      "auxiliary_loss_clip": 0.01079207,
      "auxiliary_loss_mlp": 0.00786635,
      "balance_loss_clip": 1.03969622,
      "balance_loss_mlp": 1.01242733,
      "epoch": 0.7691567713813318,
      "flos": 21537568281600.0,
      "grad_norm": 1.8476481545668038,
      "language_loss": 0.72403139,
      "learning_rate": 5.333161299238673e-07,
      "loss": 0.74268985,
      "num_input_tokens_seen": 275873930,
      "step": 12793,
      "time_per_iteration": 2.5596067905426025
    },
    {
      "auxiliary_loss_clip": 0.01068164,
      "auxiliary_loss_mlp": 0.01032311,
      "balance_loss_clip": 1.03857279,
      "balance_loss_mlp": 1.02027082,
      "epoch": 0.7692168946339997,
      "flos": 39379999720320.0,
      "grad_norm": 1.8383572918807012,
      "language_loss": 0.63565028,
      "learning_rate": 5.330513783189803e-07,
      "loss": 0.65665501,
      "num_input_tokens_seen": 275895895,
      "step": 12794,
      "time_per_iteration": 4.091511487960815
    },
    {
      "auxiliary_loss_clip": 0.01082856,
      "auxiliary_loss_mlp": 0.0103706,
      "balance_loss_clip": 1.03563094,
      "balance_loss_mlp": 1.02448928,
      "epoch": 0.7692770178866677,
      "flos": 25009950931200.0,
      "grad_norm": 1.4599955268337872,
      "language_loss": 0.764292,
      "learning_rate": 5.327866823409319e-07,
      "loss": 0.78549111,
      "num_input_tokens_seen": 275917825,
      "step": 12795,
      "time_per_iteration": 2.5398802757263184
    },
    {
      "auxiliary_loss_clip": 0.0106983,
      "auxiliary_loss_mlp": 0.01028011,
      "balance_loss_clip": 1.03577769,
      "balance_loss_mlp": 1.01586318,
      "epoch": 0.7693371411393356,
      "flos": 24716273333760.0,
      "grad_norm": 1.4523009428308318,
      "language_loss": 0.71834326,
      "learning_rate": 5.325220419997601e-07,
      "loss": 0.73932165,
      "num_input_tokens_seen": 275937890,
      "step": 12796,
      "time_per_iteration": 2.592381477355957
    },
    {
      "auxiliary_loss_clip": 0.01105366,
      "auxiliary_loss_mlp": 0.01026803,
      "balance_loss_clip": 1.03589821,
      "balance_loss_mlp": 1.01441669,
      "epoch": 0.7693972643920036,
      "flos": 15924803028480.0,
      "grad_norm": 1.8974975529529157,
      "language_loss": 0.64765334,
      "learning_rate": 5.32257457305499e-07,
      "loss": 0.66897511,
      "num_input_tokens_seen": 275954495,
      "step": 12797,
      "time_per_iteration": 2.440858840942383
    },
    {
      "auxiliary_loss_clip": 0.0106958,
      "auxiliary_loss_mlp": 0.01033792,
      "balance_loss_clip": 1.03477359,
      "balance_loss_mlp": 1.02030301,
      "epoch": 0.7694573876446715,
      "flos": 25405901527680.0,
      "grad_norm": 1.8445992955216743,
      "language_loss": 0.91545862,
      "learning_rate": 5.319929282681823e-07,
      "loss": 0.93649232,
      "num_input_tokens_seen": 275972395,
      "step": 12798,
      "time_per_iteration": 2.5899174213409424
    },
    {
      "auxiliary_loss_clip": 0.01063103,
      "auxiliary_loss_mlp": 0.01024798,
      "balance_loss_clip": 1.03533494,
      "balance_loss_mlp": 1.01313889,
      "epoch": 0.7695175108973396,
      "flos": 16654220513280.0,
      "grad_norm": 2.20104586473166,
      "language_loss": 0.82254064,
      "learning_rate": 5.317284548978418e-07,
      "loss": 0.84341967,
      "num_input_tokens_seen": 275989020,
      "step": 12799,
      "time_per_iteration": 2.552102565765381
    },
    {
      "auxiliary_loss_clip": 0.01052663,
      "auxiliary_loss_mlp": 0.01026426,
      "balance_loss_clip": 1.03598535,
      "balance_loss_mlp": 1.01382542,
      "epoch": 0.7695776341500075,
      "flos": 13626520237440.0,
      "grad_norm": 1.9254303479503816,
      "language_loss": 0.78411353,
      "learning_rate": 5.314640372045045e-07,
      "loss": 0.80490446,
      "num_input_tokens_seen": 276006525,
      "step": 12800,
      "time_per_iteration": 2.607235908508301
    },
    {
      "auxiliary_loss_clip": 0.0108768,
      "auxiliary_loss_mlp": 0.01024928,
      "balance_loss_clip": 1.03537083,
      "balance_loss_mlp": 1.01174903,
      "epoch": 0.7696377574026755,
      "flos": 24276690691200.0,
      "grad_norm": 1.5456358546841908,
      "language_loss": 0.83807737,
      "learning_rate": 5.31199675198198e-07,
      "loss": 0.85920346,
      "num_input_tokens_seen": 276027130,
      "step": 12801,
      "time_per_iteration": 2.557515859603882
    },
    {
      "auxiliary_loss_clip": 0.01084126,
      "auxiliary_loss_mlp": 0.01027013,
      "balance_loss_clip": 1.03558373,
      "balance_loss_mlp": 1.01469278,
      "epoch": 0.7696978806553435,
      "flos": 20923137210240.0,
      "grad_norm": 1.9093711411420022,
      "language_loss": 0.72136998,
      "learning_rate": 5.30935368888947e-07,
      "loss": 0.74248135,
      "num_input_tokens_seen": 276045715,
      "step": 12802,
      "time_per_iteration": 2.572239637374878
    },
    {
      "auxiliary_loss_clip": 0.01079966,
      "auxiliary_loss_mlp": 0.01032613,
      "balance_loss_clip": 1.03480852,
      "balance_loss_mlp": 1.02014971,
      "epoch": 0.7697580039080114,
      "flos": 22929609911040.0,
      "grad_norm": 1.7594288205026527,
      "language_loss": 0.76317799,
      "learning_rate": 5.306711182867747e-07,
      "loss": 0.78430372,
      "num_input_tokens_seen": 276065375,
      "step": 12803,
      "time_per_iteration": 2.6012067794799805
    },
    {
      "auxiliary_loss_clip": 0.01017215,
      "auxiliary_loss_mlp": 0.01004345,
      "balance_loss_clip": 1.0137434,
      "balance_loss_mlp": 1.00306976,
      "epoch": 0.7698181271606794,
      "flos": 68717654933760.0,
      "grad_norm": 0.7341073536466035,
      "language_loss": 0.55844843,
      "learning_rate": 5.304069234017001e-07,
      "loss": 0.57866406,
      "num_input_tokens_seen": 276131405,
      "step": 12804,
      "time_per_iteration": 3.174926996231079
    },
    {
      "auxiliary_loss_clip": 0.01015219,
      "auxiliary_loss_mlp": 0.01004636,
      "balance_loss_clip": 1.01170886,
      "balance_loss_mlp": 1.00361109,
      "epoch": 0.7698782504133473,
      "flos": 67409716999680.0,
      "grad_norm": 0.7475831035439705,
      "language_loss": 0.54013586,
      "learning_rate": 5.301427842437429e-07,
      "loss": 0.56033444,
      "num_input_tokens_seen": 276200755,
      "step": 12805,
      "time_per_iteration": 3.303776502609253
    },
    {
      "auxiliary_loss_clip": 0.01075105,
      "auxiliary_loss_mlp": 0.01033129,
      "balance_loss_clip": 1.03778648,
      "balance_loss_mlp": 1.02026641,
      "epoch": 0.7699383736660154,
      "flos": 22488842119680.0,
      "grad_norm": 1.977605556067504,
      "language_loss": 0.72660744,
      "learning_rate": 5.298787008229187e-07,
      "loss": 0.74768978,
      "num_input_tokens_seen": 276217880,
      "step": 12806,
      "time_per_iteration": 2.5716021060943604
    },
    {
      "auxiliary_loss_clip": 0.01081905,
      "auxiliary_loss_mlp": 0.0103509,
      "balance_loss_clip": 1.03661585,
      "balance_loss_mlp": 1.02271605,
      "epoch": 0.7699984969186833,
      "flos": 21539723097600.0,
      "grad_norm": 1.8908332029052761,
      "language_loss": 0.74887365,
      "learning_rate": 5.296146731492408e-07,
      "loss": 0.77004361,
      "num_input_tokens_seen": 276234810,
      "step": 12807,
      "time_per_iteration": 2.5444610118865967
    },
    {
      "auxiliary_loss_clip": 0.0109978,
      "auxiliary_loss_mlp": 0.0103541,
      "balance_loss_clip": 1.0377326,
      "balance_loss_mlp": 1.02279758,
      "epoch": 0.7700586201713513,
      "flos": 21719096640000.0,
      "grad_norm": 2.262278742494346,
      "language_loss": 0.80169064,
      "learning_rate": 5.293507012327218e-07,
      "loss": 0.82304251,
      "num_input_tokens_seen": 276252850,
      "step": 12808,
      "time_per_iteration": 2.502634286880493
    },
    {
      "auxiliary_loss_clip": 0.01098539,
      "auxiliary_loss_mlp": 0.01033812,
      "balance_loss_clip": 1.0379343,
      "balance_loss_mlp": 1.0218792,
      "epoch": 0.7701187434240192,
      "flos": 27856015107840.0,
      "grad_norm": 1.931795885772982,
      "language_loss": 0.79230249,
      "learning_rate": 5.290867850833718e-07,
      "loss": 0.81362605,
      "num_input_tokens_seen": 276272525,
      "step": 12809,
      "time_per_iteration": 2.554884433746338
    },
    {
      "auxiliary_loss_clip": 0.01070432,
      "auxiliary_loss_mlp": 0.01026467,
      "balance_loss_clip": 1.03557098,
      "balance_loss_mlp": 1.01510668,
      "epoch": 0.7701788666766872,
      "flos": 28621307301120.0,
      "grad_norm": 1.4202725467125998,
      "language_loss": 0.70138264,
      "learning_rate": 5.288229247111993e-07,
      "loss": 0.72235167,
      "num_input_tokens_seen": 276294210,
      "step": 12810,
      "time_per_iteration": 2.599517822265625
    },
    {
      "auxiliary_loss_clip": 0.01082259,
      "auxiliary_loss_mlp": 0.0103901,
      "balance_loss_clip": 1.03460526,
      "balance_loss_mlp": 1.02326214,
      "epoch": 0.7702389899293551,
      "flos": 14246446089600.0,
      "grad_norm": 2.334487869428652,
      "language_loss": 0.78669435,
      "learning_rate": 5.285591201262079e-07,
      "loss": 0.80790699,
      "num_input_tokens_seen": 276310290,
      "step": 12811,
      "time_per_iteration": 2.5000243186950684
    },
    {
      "auxiliary_loss_clip": 0.01014159,
      "auxiliary_loss_mlp": 0.01019602,
      "balance_loss_clip": 1.01294971,
      "balance_loss_mlp": 1.01806402,
      "epoch": 0.7702991131820232,
      "flos": 70574128439040.0,
      "grad_norm": 0.810492115759281,
      "language_loss": 0.56637633,
      "learning_rate": 5.28295371338402e-07,
      "loss": 0.58671403,
      "num_input_tokens_seen": 276371715,
      "step": 12812,
      "time_per_iteration": 3.1848747730255127
    },
    {
      "auxiliary_loss_clip": 0.01075142,
      "auxiliary_loss_mlp": 0.01034388,
      "balance_loss_clip": 1.03562903,
      "balance_loss_mlp": 1.02244925,
      "epoch": 0.7703592364346911,
      "flos": 25480021242240.0,
      "grad_norm": 1.5828252839654038,
      "language_loss": 0.71707237,
      "learning_rate": 5.280316783577836e-07,
      "loss": 0.7381677,
      "num_input_tokens_seen": 276389895,
      "step": 12813,
      "time_per_iteration": 2.5900189876556396
    },
    {
      "auxiliary_loss_clip": 0.01096871,
      "auxiliary_loss_mlp": 0.01027977,
      "balance_loss_clip": 1.03604746,
      "balance_loss_mlp": 1.01477432,
      "epoch": 0.7704193596873591,
      "flos": 19280906375040.0,
      "grad_norm": 2.0001819861113272,
      "language_loss": 0.66381699,
      "learning_rate": 5.27768041194351e-07,
      "loss": 0.68506545,
      "num_input_tokens_seen": 276408990,
      "step": 12814,
      "time_per_iteration": 2.5317184925079346
    },
    {
      "auxiliary_loss_clip": 0.01082907,
      "auxiliary_loss_mlp": 0.01035732,
      "balance_loss_clip": 1.03482854,
      "balance_loss_mlp": 1.02348971,
      "epoch": 0.7704794829400271,
      "flos": 23658452778240.0,
      "grad_norm": 1.811959076872331,
      "language_loss": 0.65576565,
      "learning_rate": 5.275044598581018e-07,
      "loss": 0.676952,
      "num_input_tokens_seen": 276428190,
      "step": 12815,
      "time_per_iteration": 2.548947811126709
    },
    {
      "auxiliary_loss_clip": 0.01093691,
      "auxiliary_loss_mlp": 0.01032063,
      "balance_loss_clip": 1.03550398,
      "balance_loss_mlp": 1.01962972,
      "epoch": 0.770539606192695,
      "flos": 18989311766400.0,
      "grad_norm": 2.0657346600618967,
      "language_loss": 0.6466924,
      "learning_rate": 5.272409343590322e-07,
      "loss": 0.66794991,
      "num_input_tokens_seen": 276446855,
      "step": 12816,
      "time_per_iteration": 2.4844844341278076
    },
    {
      "auxiliary_loss_clip": 0.0109652,
      "auxiliary_loss_mlp": 0.01033189,
      "balance_loss_clip": 1.03742254,
      "balance_loss_mlp": 1.02119064,
      "epoch": 0.770599729445363,
      "flos": 11830160142720.0,
      "grad_norm": 2.039349826403797,
      "language_loss": 0.71837872,
      "learning_rate": 5.26977464707133e-07,
      "loss": 0.73967576,
      "num_input_tokens_seen": 276462000,
      "step": 12817,
      "time_per_iteration": 2.45175838470459
    },
    {
      "auxiliary_loss_clip": 0.0106389,
      "auxiliary_loss_mlp": 0.01029471,
      "balance_loss_clip": 1.03669989,
      "balance_loss_mlp": 1.01759768,
      "epoch": 0.770659852698031,
      "flos": 17822610109440.0,
      "grad_norm": 3.857868198964955,
      "language_loss": 0.61035085,
      "learning_rate": 5.267140509123957e-07,
      "loss": 0.63128448,
      "num_input_tokens_seen": 276481190,
      "step": 12818,
      "time_per_iteration": 2.576406240463257
    },
    {
      "auxiliary_loss_clip": 0.0109336,
      "auxiliary_loss_mlp": 0.0102795,
      "balance_loss_clip": 1.03676832,
      "balance_loss_mlp": 1.0172981,
      "epoch": 0.770719975950699,
      "flos": 21871968923520.0,
      "grad_norm": 1.6504535971655625,
      "language_loss": 0.67155415,
      "learning_rate": 5.264506929848093e-07,
      "loss": 0.69276726,
      "num_input_tokens_seen": 276499520,
      "step": 12819,
      "time_per_iteration": 2.495257616043091
    },
    {
      "auxiliary_loss_clip": 0.01107168,
      "auxiliary_loss_mlp": 0.01031521,
      "balance_loss_clip": 1.03665185,
      "balance_loss_mlp": 1.01935554,
      "epoch": 0.7707800992033669,
      "flos": 21325049464320.0,
      "grad_norm": 1.7333373051393226,
      "language_loss": 0.57426894,
      "learning_rate": 5.261873909343608e-07,
      "loss": 0.5956558,
      "num_input_tokens_seen": 276519110,
      "step": 12820,
      "time_per_iteration": 2.4853034019470215
    },
    {
      "auxiliary_loss_clip": 0.01077188,
      "auxiliary_loss_mlp": 0.01029643,
      "balance_loss_clip": 1.03572035,
      "balance_loss_mlp": 1.01718569,
      "epoch": 0.7708402224560349,
      "flos": 28179426188160.0,
      "grad_norm": 1.7856003471020523,
      "language_loss": 0.81137687,
      "learning_rate": 5.259241447710343e-07,
      "loss": 0.83244526,
      "num_input_tokens_seen": 276538805,
      "step": 12821,
      "time_per_iteration": 2.5758302211761475
    },
    {
      "auxiliary_loss_clip": 0.0110704,
      "auxiliary_loss_mlp": 0.0103108,
      "balance_loss_clip": 1.03720808,
      "balance_loss_mlp": 1.01859903,
      "epoch": 0.7709003457087028,
      "flos": 15377057556480.0,
      "grad_norm": 6.69579716423254,
      "language_loss": 0.68804806,
      "learning_rate": 5.256609545048114e-07,
      "loss": 0.70942926,
      "num_input_tokens_seen": 276554770,
      "step": 12822,
      "time_per_iteration": 3.827915668487549
    },
    {
      "auxiliary_loss_clip": 0.01082006,
      "auxiliary_loss_mlp": 0.0103536,
      "balance_loss_clip": 1.03473878,
      "balance_loss_mlp": 1.02257514,
      "epoch": 0.7709604689613708,
      "flos": 30621854257920.0,
      "grad_norm": 1.7375377098480962,
      "language_loss": 0.72474754,
      "learning_rate": 5.253978201456733e-07,
      "loss": 0.74592113,
      "num_input_tokens_seen": 276574535,
      "step": 12823,
      "time_per_iteration": 2.5875067710876465
    },
    {
      "auxiliary_loss_clip": 0.01100606,
      "auxiliary_loss_mlp": 0.01039622,
      "balance_loss_clip": 1.03746247,
      "balance_loss_mlp": 1.02537012,
      "epoch": 0.7710205922140387,
      "flos": 20301272023680.0,
      "grad_norm": 1.6187982955998403,
      "language_loss": 0.76372194,
      "learning_rate": 5.251347417035969e-07,
      "loss": 0.78512424,
      "num_input_tokens_seen": 276592925,
      "step": 12824,
      "time_per_iteration": 2.495988130569458
    },
    {
      "auxiliary_loss_clip": 0.01082933,
      "auxiliary_loss_mlp": 0.0102914,
      "balance_loss_clip": 1.03871047,
      "balance_loss_mlp": 1.01649153,
      "epoch": 0.7710807154667068,
      "flos": 19644214487040.0,
      "grad_norm": 1.8043239632398065,
      "language_loss": 0.7240535,
      "learning_rate": 5.248717191885592e-07,
      "loss": 0.74517417,
      "num_input_tokens_seen": 276610540,
      "step": 12825,
      "time_per_iteration": 3.973560333251953
    },
    {
      "auxiliary_loss_clip": 0.01102457,
      "auxiliary_loss_mlp": 0.01037269,
      "balance_loss_clip": 1.03645158,
      "balance_loss_mlp": 1.02626646,
      "epoch": 0.7711408387193747,
      "flos": 20006337450240.0,
      "grad_norm": 1.4105632343658252,
      "language_loss": 0.73708212,
      "learning_rate": 5.246087526105343e-07,
      "loss": 0.75847936,
      "num_input_tokens_seen": 276629200,
      "step": 12826,
      "time_per_iteration": 2.471680164337158
    },
    {
      "auxiliary_loss_clip": 0.01107573,
      "auxiliary_loss_mlp": 0.01035956,
      "balance_loss_clip": 1.03560567,
      "balance_loss_mlp": 1.02240849,
      "epoch": 0.7712009619720427,
      "flos": 24971131307520.0,
      "grad_norm": 1.4921595501791176,
      "language_loss": 0.81317747,
      "learning_rate": 5.243458419794933e-07,
      "loss": 0.83461285,
      "num_input_tokens_seen": 276648655,
      "step": 12827,
      "time_per_iteration": 3.969805955886841
    },
    {
      "auxiliary_loss_clip": 0.01032215,
      "auxiliary_loss_mlp": 0.01002316,
      "balance_loss_clip": 1.00913978,
      "balance_loss_mlp": 1.0011301,
      "epoch": 0.7712610852247107,
      "flos": 63249681404160.0,
      "grad_norm": 0.8629111488922488,
      "language_loss": 0.55171943,
      "learning_rate": 5.240829873054051e-07,
      "loss": 0.57206476,
      "num_input_tokens_seen": 276716500,
      "step": 12828,
      "time_per_iteration": 3.2220711708068848
    },
    {
      "auxiliary_loss_clip": 0.01064932,
      "auxiliary_loss_mlp": 0.01032844,
      "balance_loss_clip": 1.03336239,
      "balance_loss_mlp": 1.021227,
      "epoch": 0.7713212084773786,
      "flos": 18697860812160.0,
      "grad_norm": 2.236184037837105,
      "language_loss": 0.69423324,
      "learning_rate": 5.23820188598238e-07,
      "loss": 0.71521103,
      "num_input_tokens_seen": 276733535,
      "step": 12829,
      "time_per_iteration": 2.515791654586792
    },
    {
      "auxiliary_loss_clip": 0.01085029,
      "auxiliary_loss_mlp": 0.01030275,
      "balance_loss_clip": 1.03971767,
      "balance_loss_mlp": 1.01715636,
      "epoch": 0.7713813317300466,
      "flos": 14173367869440.0,
      "grad_norm": 2.8080193338262878,
      "language_loss": 0.8026126,
      "learning_rate": 5.235574458679579e-07,
      "loss": 0.82376564,
      "num_input_tokens_seen": 276749575,
      "step": 12830,
      "time_per_iteration": 2.4987292289733887
    },
    {
      "auxiliary_loss_clip": 0.01098103,
      "auxiliary_loss_mlp": 0.01038453,
      "balance_loss_clip": 1.03745747,
      "balance_loss_mlp": 1.02513778,
      "epoch": 0.7714414549827145,
      "flos": 25703960584320.0,
      "grad_norm": 1.7036685134966296,
      "language_loss": 0.77744859,
      "learning_rate": 5.232947591245269e-07,
      "loss": 0.79881412,
      "num_input_tokens_seen": 276769460,
      "step": 12831,
      "time_per_iteration": 2.5265276432037354
    },
    {
      "auxiliary_loss_clip": 0.01075452,
      "auxiliary_loss_mlp": 0.0103404,
      "balance_loss_clip": 1.03393602,
      "balance_loss_mlp": 1.02057528,
      "epoch": 0.7715015782353826,
      "flos": 30555312312960.0,
      "grad_norm": 1.446187031438796,
      "language_loss": 0.61033809,
      "learning_rate": 5.230321283779071e-07,
      "loss": 0.63143301,
      "num_input_tokens_seen": 276790820,
      "step": 12832,
      "time_per_iteration": 4.0561299324035645
    },
    {
      "auxiliary_loss_clip": 0.01079263,
      "auxiliary_loss_mlp": 0.01036316,
      "balance_loss_clip": 1.03494036,
      "balance_loss_mlp": 1.02373326,
      "epoch": 0.7715617014880505,
      "flos": 20229343038720.0,
      "grad_norm": 1.5677496436205776,
      "language_loss": 0.79475713,
      "learning_rate": 5.227695536380572e-07,
      "loss": 0.81591296,
      "num_input_tokens_seen": 276811345,
      "step": 12833,
      "time_per_iteration": 2.5706682205200195
    },
    {
      "auxiliary_loss_clip": 0.00993473,
      "auxiliary_loss_mlp": 0.00999943,
      "balance_loss_clip": 1.01433945,
      "balance_loss_mlp": 0.9986912,
      "epoch": 0.7716218247407185,
      "flos": 63664770971520.0,
      "grad_norm": 0.8413287250587307,
      "language_loss": 0.55366367,
      "learning_rate": 5.22507034914933e-07,
      "loss": 0.57359779,
      "num_input_tokens_seen": 276870950,
      "step": 12834,
      "time_per_iteration": 3.2120351791381836
    },
    {
      "auxiliary_loss_clip": 0.01058304,
      "auxiliary_loss_mlp": 0.01032288,
      "balance_loss_clip": 1.03175056,
      "balance_loss_mlp": 1.01904392,
      "epoch": 0.7716819479933864,
      "flos": 19791807471360.0,
      "grad_norm": 2.8875431454830265,
      "language_loss": 0.7275089,
      "learning_rate": 5.222445722184903e-07,
      "loss": 0.74841487,
      "num_input_tokens_seen": 276890760,
      "step": 12835,
      "time_per_iteration": 2.64113450050354
    },
    {
      "auxiliary_loss_clip": 0.0107206,
      "auxiliary_loss_mlp": 0.00784714,
      "balance_loss_clip": 1.03407133,
      "balance_loss_mlp": 1.01056337,
      "epoch": 0.7717420712460544,
      "flos": 18442176825600.0,
      "grad_norm": 1.6541349596803399,
      "language_loss": 0.70040482,
      "learning_rate": 5.219821655586814e-07,
      "loss": 0.71897256,
      "num_input_tokens_seen": 276909625,
      "step": 12836,
      "time_per_iteration": 2.5338456630706787
    },
    {
      "auxiliary_loss_clip": 0.01083471,
      "auxiliary_loss_mlp": 0.01030685,
      "balance_loss_clip": 1.03568006,
      "balance_loss_mlp": 1.01816249,
      "epoch": 0.7718021944987223,
      "flos": 35189476456320.0,
      "grad_norm": 2.0268943455812347,
      "language_loss": 0.59288329,
      "learning_rate": 5.217198149454575e-07,
      "loss": 0.61402482,
      "num_input_tokens_seen": 276930760,
      "step": 12837,
      "time_per_iteration": 2.658083915710449
    },
    {
      "auxiliary_loss_clip": 0.01025113,
      "auxiliary_loss_mlp": 0.01002826,
      "balance_loss_clip": 1.01911259,
      "balance_loss_mlp": 1.00147343,
      "epoch": 0.7718623177513904,
      "flos": 67923167961600.0,
      "grad_norm": 0.8625793663627638,
      "language_loss": 0.55769199,
      "learning_rate": 5.214575203887666e-07,
      "loss": 0.57797134,
      "num_input_tokens_seen": 276989580,
      "step": 12838,
      "time_per_iteration": 3.0817508697509766
    },
    {
      "auxiliary_loss_clip": 0.01095124,
      "auxiliary_loss_mlp": 0.0102855,
      "balance_loss_clip": 1.03618503,
      "balance_loss_mlp": 1.01718342,
      "epoch": 0.7719224410040583,
      "flos": 18581401941120.0,
      "grad_norm": 2.525487118789534,
      "language_loss": 0.69430709,
      "learning_rate": 5.211952818985538e-07,
      "loss": 0.71554387,
      "num_input_tokens_seen": 277005450,
      "step": 12839,
      "time_per_iteration": 2.4914534091949463
    },
    {
      "auxiliary_loss_clip": 0.01093023,
      "auxiliary_loss_mlp": 0.01026155,
      "balance_loss_clip": 1.0364368,
      "balance_loss_mlp": 1.01431131,
      "epoch": 0.7719825642567263,
      "flos": 23075802264960.0,
      "grad_norm": 2.552612767149446,
      "language_loss": 0.80033261,
      "learning_rate": 5.209330994847647e-07,
      "loss": 0.82152438,
      "num_input_tokens_seen": 277023055,
      "step": 12840,
      "time_per_iteration": 2.501300811767578
    },
    {
      "auxiliary_loss_clip": 0.01094042,
      "auxiliary_loss_mlp": 0.00784336,
      "balance_loss_clip": 1.03548217,
      "balance_loss_mlp": 1.01049852,
      "epoch": 0.7720426875093943,
      "flos": 20339086066560.0,
      "grad_norm": 1.7545138080351081,
      "language_loss": 0.80459869,
      "learning_rate": 5.206709731573402e-07,
      "loss": 0.82338244,
      "num_input_tokens_seen": 277041150,
      "step": 12841,
      "time_per_iteration": 2.5766897201538086
    },
    {
      "auxiliary_loss_clip": 0.01068646,
      "auxiliary_loss_mlp": 0.01029794,
      "balance_loss_clip": 1.03618455,
      "balance_loss_mlp": 1.01749778,
      "epoch": 0.7721028107620622,
      "flos": 23880704181120.0,
      "grad_norm": 1.429064390350467,
      "language_loss": 0.76078552,
      "learning_rate": 5.204089029262208e-07,
      "loss": 0.78176993,
      "num_input_tokens_seen": 277063895,
      "step": 12842,
      "time_per_iteration": 2.5987627506256104
    },
    {
      "auxiliary_loss_clip": 0.0106023,
      "auxiliary_loss_mlp": 0.00786152,
      "balance_loss_clip": 1.03664851,
      "balance_loss_mlp": 1.01227021,
      "epoch": 0.7721629340147302,
      "flos": 26651571235200.0,
      "grad_norm": 1.8124624665391602,
      "language_loss": 0.6872108,
      "learning_rate": 5.201468888013445e-07,
      "loss": 0.70567465,
      "num_input_tokens_seen": 277084045,
      "step": 12843,
      "time_per_iteration": 2.66969633102417
    },
    {
      "auxiliary_loss_clip": 0.01081744,
      "auxiliary_loss_mlp": 0.01028079,
      "balance_loss_clip": 1.0319314,
      "balance_loss_mlp": 1.01619935,
      "epoch": 0.7722230572673981,
      "flos": 21178857110400.0,
      "grad_norm": 2.0607072607211427,
      "language_loss": 0.7357223,
      "learning_rate": 5.198849307926465e-07,
      "loss": 0.75682056,
      "num_input_tokens_seen": 277102625,
      "step": 12844,
      "time_per_iteration": 2.548755645751953
    },
    {
      "auxiliary_loss_clip": 0.01090724,
      "auxiliary_loss_mlp": 0.01041073,
      "balance_loss_clip": 1.03527331,
      "balance_loss_mlp": 1.02731609,
      "epoch": 0.7722831805200662,
      "flos": 27964644814080.0,
      "grad_norm": 1.3927089722591668,
      "language_loss": 0.71669269,
      "learning_rate": 5.196230289100596e-07,
      "loss": 0.73801064,
      "num_input_tokens_seen": 277123210,
      "step": 12845,
      "time_per_iteration": 2.542170286178589
    },
    {
      "auxiliary_loss_clip": 0.01102499,
      "auxiliary_loss_mlp": 0.01031733,
      "balance_loss_clip": 1.03568745,
      "balance_loss_mlp": 1.02026498,
      "epoch": 0.7723433037727341,
      "flos": 33875576864640.0,
      "grad_norm": 1.9749642509709284,
      "language_loss": 0.64235836,
      "learning_rate": 5.193611831635159e-07,
      "loss": 0.6637007,
      "num_input_tokens_seen": 277144895,
      "step": 12846,
      "time_per_iteration": 2.555647134780884
    },
    {
      "auxiliary_loss_clip": 0.01023505,
      "auxiliary_loss_mlp": 0.00763444,
      "balance_loss_clip": 1.00914526,
      "balance_loss_mlp": 1.00374627,
      "epoch": 0.7724034270254021,
      "flos": 62848271940480.0,
      "grad_norm": 0.7925767752665814,
      "language_loss": 0.61749732,
      "learning_rate": 5.19099393562945e-07,
      "loss": 0.6353668,
      "num_input_tokens_seen": 277205160,
      "step": 12847,
      "time_per_iteration": 3.0515732765197754
    },
    {
      "auxiliary_loss_clip": 0.01103869,
      "auxiliary_loss_mlp": 0.01026873,
      "balance_loss_clip": 1.03438878,
      "balance_loss_mlp": 1.01480889,
      "epoch": 0.77246355027807,
      "flos": 23295467888640.0,
      "grad_norm": 1.7693170715561644,
      "language_loss": 0.79242843,
      "learning_rate": 5.188376601182732e-07,
      "loss": 0.81373584,
      "num_input_tokens_seen": 277223005,
      "step": 12848,
      "time_per_iteration": 2.467751979827881
    },
    {
      "auxiliary_loss_clip": 0.01068881,
      "auxiliary_loss_mlp": 0.01041512,
      "balance_loss_clip": 1.03671217,
      "balance_loss_mlp": 1.0279218,
      "epoch": 0.772523673530738,
      "flos": 20121287950080.0,
      "grad_norm": 1.5612725939448235,
      "language_loss": 0.72835159,
      "learning_rate": 5.185759828394261e-07,
      "loss": 0.74945551,
      "num_input_tokens_seen": 277241785,
      "step": 12849,
      "time_per_iteration": 2.5606796741485596
    },
    {
      "auxiliary_loss_clip": 0.01103645,
      "auxiliary_loss_mlp": 0.01032594,
      "balance_loss_clip": 1.03499579,
      "balance_loss_mlp": 1.02017212,
      "epoch": 0.7725837967834059,
      "flos": 17820096157440.0,
      "grad_norm": 1.9126752188721703,
      "language_loss": 0.78369349,
      "learning_rate": 5.183143617363261e-07,
      "loss": 0.80505592,
      "num_input_tokens_seen": 277259050,
      "step": 12850,
      "time_per_iteration": 2.4162545204162598
    },
    {
      "auxiliary_loss_clip": 0.01049013,
      "auxiliary_loss_mlp": 0.00784858,
      "balance_loss_clip": 1.03347754,
      "balance_loss_mlp": 1.00920463,
      "epoch": 0.772643920036074,
      "flos": 27198921657600.0,
      "grad_norm": 1.5062023186817235,
      "language_loss": 0.79853922,
      "learning_rate": 5.180527968188935e-07,
      "loss": 0.81687796,
      "num_input_tokens_seen": 277278235,
      "step": 12851,
      "time_per_iteration": 2.727858304977417
    },
    {
      "auxiliary_loss_clip": 0.01094172,
      "auxiliary_loss_mlp": 0.01029736,
      "balance_loss_clip": 1.03687274,
      "balance_loss_mlp": 1.01605678,
      "epoch": 0.7727040432887419,
      "flos": 21579512388480.0,
      "grad_norm": 1.6296018580622826,
      "language_loss": 0.73873973,
      "learning_rate": 5.177912880970474e-07,
      "loss": 0.75997883,
      "num_input_tokens_seen": 277298355,
      "step": 12852,
      "time_per_iteration": 2.6451754570007324
    },
    {
      "auxiliary_loss_clip": 0.01102929,
      "auxiliary_loss_mlp": 0.01034034,
      "balance_loss_clip": 1.0345161,
      "balance_loss_mlp": 1.02166569,
      "epoch": 0.7727641665414099,
      "flos": 22236641752320.0,
      "grad_norm": 1.674738145271165,
      "language_loss": 0.82096112,
      "learning_rate": 5.17529835580704e-07,
      "loss": 0.84233081,
      "num_input_tokens_seen": 277316095,
      "step": 12853,
      "time_per_iteration": 2.469040870666504
    },
    {
      "auxiliary_loss_clip": 0.01032215,
      "auxiliary_loss_mlp": 0.01002478,
      "balance_loss_clip": 1.00922358,
      "balance_loss_mlp": 1.00130939,
      "epoch": 0.7728242897940779,
      "flos": 54832221463680.0,
      "grad_norm": 0.8722879249591857,
      "language_loss": 0.54522181,
      "learning_rate": 5.172684392797786e-07,
      "loss": 0.56556875,
      "num_input_tokens_seen": 277380130,
      "step": 12854,
      "time_per_iteration": 3.1401758193969727
    },
    {
      "auxiliary_loss_clip": 0.01097204,
      "auxiliary_loss_mlp": 0.0103051,
      "balance_loss_clip": 1.03656387,
      "balance_loss_mlp": 1.01679492,
      "epoch": 0.7728844130467458,
      "flos": 34461962392320.0,
      "grad_norm": 1.6658056097746574,
      "language_loss": 0.71803451,
      "learning_rate": 5.170070992041826e-07,
      "loss": 0.73931158,
      "num_input_tokens_seen": 277404015,
      "step": 12855,
      "time_per_iteration": 2.6139421463012695
    },
    {
      "auxiliary_loss_clip": 0.01105138,
      "auxiliary_loss_mlp": 0.01029948,
      "balance_loss_clip": 1.03576529,
      "balance_loss_mlp": 1.0168407,
      "epoch": 0.7729445362994138,
      "flos": 18916341287040.0,
      "grad_norm": 1.6396582811302933,
      "language_loss": 0.6771338,
      "learning_rate": 5.167458153638254e-07,
      "loss": 0.69848466,
      "num_input_tokens_seen": 277421375,
      "step": 12856,
      "time_per_iteration": 2.4586925506591797
    },
    {
      "auxiliary_loss_clip": 0.01075173,
      "auxiliary_loss_mlp": 0.0102987,
      "balance_loss_clip": 1.03489864,
      "balance_loss_mlp": 1.01795459,
      "epoch": 0.7730046595520818,
      "flos": 22200048771840.0,
      "grad_norm": 1.5090834368929613,
      "language_loss": 0.78801906,
      "learning_rate": 5.164845877686162e-07,
      "loss": 0.80906945,
      "num_input_tokens_seen": 277440170,
      "step": 12857,
      "time_per_iteration": 2.5477378368377686
    },
    {
      "auxiliary_loss_clip": 0.01055116,
      "auxiliary_loss_mlp": 0.00783478,
      "balance_loss_clip": 1.03626943,
      "balance_loss_mlp": 1.00894868,
      "epoch": 0.7730647828047498,
      "flos": 13552328695680.0,
      "grad_norm": 1.6656244384776142,
      "language_loss": 0.7858988,
      "learning_rate": 5.162234164284591e-07,
      "loss": 0.80428469,
      "num_input_tokens_seen": 277456880,
      "step": 12858,
      "time_per_iteration": 2.628862142562866
    },
    {
      "auxiliary_loss_clip": 0.01104806,
      "auxiliary_loss_mlp": 0.01031499,
      "balance_loss_clip": 1.03503466,
      "balance_loss_mlp": 1.0192976,
      "epoch": 0.7731249060574177,
      "flos": 21976037602560.0,
      "grad_norm": 1.9714962257832367,
      "language_loss": 0.76945198,
      "learning_rate": 5.159623013532591e-07,
      "loss": 0.790815,
      "num_input_tokens_seen": 277475365,
      "step": 12859,
      "time_per_iteration": 2.4592201709747314
    },
    {
      "auxiliary_loss_clip": 0.01093267,
      "auxiliary_loss_mlp": 0.0102761,
      "balance_loss_clip": 1.03840411,
      "balance_loss_mlp": 1.01702464,
      "epoch": 0.7731850293100857,
      "flos": 22601817371520.0,
      "grad_norm": 1.5259192437773528,
      "language_loss": 0.68043607,
      "learning_rate": 5.157012425529186e-07,
      "loss": 0.70164484,
      "num_input_tokens_seen": 277494975,
      "step": 12860,
      "time_per_iteration": 2.5339112281799316
    },
    {
      "auxiliary_loss_clip": 0.01107958,
      "auxiliary_loss_mlp": 0.01034981,
      "balance_loss_clip": 1.0355531,
      "balance_loss_mlp": 1.02216041,
      "epoch": 0.7732451525627536,
      "flos": 14098422142080.0,
      "grad_norm": 9.239583151699152,
      "language_loss": 0.74312508,
      "learning_rate": 5.154402400373343e-07,
      "loss": 0.7645545,
      "num_input_tokens_seen": 277510520,
      "step": 12861,
      "time_per_iteration": 3.884416341781616
    },
    {
      "auxiliary_loss_clip": 0.01098985,
      "auxiliary_loss_mlp": 0.01031664,
      "balance_loss_clip": 1.03782606,
      "balance_loss_mlp": 1.01859236,
      "epoch": 0.7733052758154216,
      "flos": 21470020755840.0,
      "grad_norm": 1.565709952876274,
      "language_loss": 0.7450549,
      "learning_rate": 5.15179293816405e-07,
      "loss": 0.76636142,
      "num_input_tokens_seen": 277530505,
      "step": 12862,
      "time_per_iteration": 2.5455410480499268
    },
    {
      "auxiliary_loss_clip": 0.01059312,
      "auxiliary_loss_mlp": 0.01031572,
      "balance_loss_clip": 1.03323889,
      "balance_loss_mlp": 1.02003241,
      "epoch": 0.7733653990680895,
      "flos": 21394284929280.0,
      "grad_norm": 1.5762928363583804,
      "language_loss": 0.83227026,
      "learning_rate": 5.149184039000256e-07,
      "loss": 0.8531791,
      "num_input_tokens_seen": 277550810,
      "step": 12863,
      "time_per_iteration": 2.6097006797790527
    },
    {
      "auxiliary_loss_clip": 0.01104908,
      "auxiliary_loss_mlp": 0.01029399,
      "balance_loss_clip": 1.0362376,
      "balance_loss_mlp": 1.01764488,
      "epoch": 0.7734255223207576,
      "flos": 17676058619520.0,
      "grad_norm": 1.6347509127155062,
      "language_loss": 0.73389584,
      "learning_rate": 5.146575702980898e-07,
      "loss": 0.75523889,
      "num_input_tokens_seen": 277567680,
      "step": 12864,
      "time_per_iteration": 3.8390305042266846
    },
    {
      "auxiliary_loss_clip": 0.01082809,
      "auxiliary_loss_mlp": 0.01031417,
      "balance_loss_clip": 1.03375387,
      "balance_loss_mlp": 1.01985335,
      "epoch": 0.7734856455734255,
      "flos": 25230837617280.0,
      "grad_norm": 1.7381131139852526,
      "language_loss": 0.82632494,
      "learning_rate": 5.143967930204871e-07,
      "loss": 0.84746718,
      "num_input_tokens_seen": 277588970,
      "step": 12865,
      "time_per_iteration": 3.9596285820007324
    },
    {
      "auxiliary_loss_clip": 0.01110946,
      "auxiliary_loss_mlp": 0.01032631,
      "balance_loss_clip": 1.03860354,
      "balance_loss_mlp": 1.01857591,
      "epoch": 0.7735457688260935,
      "flos": 23433112805760.0,
      "grad_norm": 2.0023546729560238,
      "language_loss": 0.71979547,
      "learning_rate": 5.141360720771077e-07,
      "loss": 0.7412312,
      "num_input_tokens_seen": 277605450,
      "step": 12866,
      "time_per_iteration": 2.4728803634643555
    },
    {
      "auxiliary_loss_clip": 0.01061766,
      "auxiliary_loss_mlp": 0.00782969,
      "balance_loss_clip": 1.03762388,
      "balance_loss_mlp": 1.0085429,
      "epoch": 0.7736058920787615,
      "flos": 18729246320640.0,
      "grad_norm": 2.874493838284572,
      "language_loss": 0.64914107,
      "learning_rate": 5.138754074778371e-07,
      "loss": 0.66758847,
      "num_input_tokens_seen": 277622530,
      "step": 12867,
      "time_per_iteration": 2.552341938018799
    },
    {
      "auxiliary_loss_clip": 0.01092206,
      "auxiliary_loss_mlp": 0.01033408,
      "balance_loss_clip": 1.03549671,
      "balance_loss_mlp": 1.02179718,
      "epoch": 0.7736660153314294,
      "flos": 22893304239360.0,
      "grad_norm": 1.6016792432355687,
      "language_loss": 0.71021533,
      "learning_rate": 5.136147992325595e-07,
      "loss": 0.73147148,
      "num_input_tokens_seen": 277642700,
      "step": 12868,
      "time_per_iteration": 2.5141184329986572
    },
    {
      "auxiliary_loss_clip": 0.0109834,
      "auxiliary_loss_mlp": 0.01029318,
      "balance_loss_clip": 1.03739738,
      "balance_loss_mlp": 1.01713467,
      "epoch": 0.7737261385840974,
      "flos": 13800901789440.0,
      "grad_norm": 2.0821329852335464,
      "language_loss": 0.78156292,
      "learning_rate": 5.133542473511578e-07,
      "loss": 0.80283952,
      "num_input_tokens_seen": 277660005,
      "step": 12869,
      "time_per_iteration": 2.4598798751831055
    },
    {
      "auxiliary_loss_clip": 0.0108998,
      "auxiliary_loss_mlp": 0.01026195,
      "balance_loss_clip": 1.03434563,
      "balance_loss_mlp": 1.01431012,
      "epoch": 0.7737862618367654,
      "flos": 28730727106560.0,
      "grad_norm": 1.7952855395044343,
      "language_loss": 0.73752189,
      "learning_rate": 5.130937518435124e-07,
      "loss": 0.75868362,
      "num_input_tokens_seen": 277682890,
      "step": 12870,
      "time_per_iteration": 2.5559346675872803
    },
    {
      "auxiliary_loss_clip": 0.01096023,
      "auxiliary_loss_mlp": 0.01032437,
      "balance_loss_clip": 1.03571701,
      "balance_loss_mlp": 1.01983631,
      "epoch": 0.7738463850894334,
      "flos": 17018570119680.0,
      "grad_norm": 1.9502380444041627,
      "language_loss": 0.75931907,
      "learning_rate": 5.12833312719501e-07,
      "loss": 0.78060365,
      "num_input_tokens_seen": 277699330,
      "step": 12871,
      "time_per_iteration": 3.8365724086761475
    },
    {
      "auxiliary_loss_clip": 0.0108026,
      "auxiliary_loss_mlp": 0.01030703,
      "balance_loss_clip": 1.0335499,
      "balance_loss_mlp": 1.01955736,
      "epoch": 0.7739065083421013,
      "flos": 20704010290560.0,
      "grad_norm": 1.6470968031695052,
      "language_loss": 0.69112808,
      "learning_rate": 5.12572929988999e-07,
      "loss": 0.71223772,
      "num_input_tokens_seen": 277718750,
      "step": 12872,
      "time_per_iteration": 2.6436235904693604
    },
    {
      "auxiliary_loss_clip": 0.01105758,
      "auxiliary_loss_mlp": 0.0103282,
      "balance_loss_clip": 1.03599501,
      "balance_loss_mlp": 1.01971924,
      "epoch": 0.7739666315947693,
      "flos": 20697222620160.0,
      "grad_norm": 2.0074334248013432,
      "language_loss": 0.85347533,
      "learning_rate": 5.123126036618804e-07,
      "loss": 0.87486112,
      "num_input_tokens_seen": 277734645,
      "step": 12873,
      "time_per_iteration": 2.4310896396636963
    },
    {
      "auxiliary_loss_clip": 0.0110662,
      "auxiliary_loss_mlp": 0.01033337,
      "balance_loss_clip": 1.03674269,
      "balance_loss_mlp": 1.02122593,
      "epoch": 0.7740267548474372,
      "flos": 29570677718400.0,
      "grad_norm": 2.3220493277297845,
      "language_loss": 0.65370262,
      "learning_rate": 5.120523337480174e-07,
      "loss": 0.67510217,
      "num_input_tokens_seen": 277755535,
      "step": 12874,
      "time_per_iteration": 2.531129837036133
    },
    {
      "auxiliary_loss_clip": 0.01063374,
      "auxiliary_loss_mlp": 0.01030075,
      "balance_loss_clip": 1.03622913,
      "balance_loss_mlp": 1.01757014,
      "epoch": 0.7740868781001052,
      "flos": 23659099223040.0,
      "grad_norm": 1.7701985424627638,
      "language_loss": 0.62335628,
      "learning_rate": 5.117921202572785e-07,
      "loss": 0.6442908,
      "num_input_tokens_seen": 277775585,
      "step": 12875,
      "time_per_iteration": 2.6034257411956787
    },
    {
      "auxiliary_loss_clip": 0.01095873,
      "auxiliary_loss_mlp": 0.01030449,
      "balance_loss_clip": 1.03510857,
      "balance_loss_mlp": 1.01811051,
      "epoch": 0.7741470013527731,
      "flos": 24717314828160.0,
      "grad_norm": 2.8315283171337255,
      "language_loss": 0.65323097,
      "learning_rate": 5.115319631995318e-07,
      "loss": 0.67449415,
      "num_input_tokens_seen": 277794795,
      "step": 12876,
      "time_per_iteration": 2.521559000015259
    },
    {
      "auxiliary_loss_clip": 0.01079685,
      "auxiliary_loss_mlp": 0.01032453,
      "balance_loss_clip": 1.0367167,
      "balance_loss_mlp": 1.02072287,
      "epoch": 0.7742071246054412,
      "flos": 21871645701120.0,
      "grad_norm": 2.329977414569712,
      "language_loss": 0.71151114,
      "learning_rate": 5.112718625846433e-07,
      "loss": 0.73263252,
      "num_input_tokens_seen": 277813235,
      "step": 12877,
      "time_per_iteration": 2.5230207443237305
    },
    {
      "auxiliary_loss_clip": 0.01066141,
      "auxiliary_loss_mlp": 0.01037904,
      "balance_loss_clip": 1.03405213,
      "balance_loss_mlp": 1.02376533,
      "epoch": 0.7742672478581091,
      "flos": 22674249146880.0,
      "grad_norm": 1.6489017128494265,
      "language_loss": 0.82982671,
      "learning_rate": 5.110118184224736e-07,
      "loss": 0.85086721,
      "num_input_tokens_seen": 277832560,
      "step": 12878,
      "time_per_iteration": 2.5750985145568848
    },
    {
      "auxiliary_loss_clip": 0.01084042,
      "auxiliary_loss_mlp": 0.01031945,
      "balance_loss_clip": 1.03593338,
      "balance_loss_mlp": 1.01877785,
      "epoch": 0.7743273711107771,
      "flos": 18840892769280.0,
      "grad_norm": 1.753877777350819,
      "language_loss": 0.73825979,
      "learning_rate": 5.10751830722885e-07,
      "loss": 0.75941968,
      "num_input_tokens_seen": 277850120,
      "step": 12879,
      "time_per_iteration": 2.4994149208068848
    },
    {
      "auxiliary_loss_clip": 0.01079107,
      "auxiliary_loss_mlp": 0.01031164,
      "balance_loss_clip": 1.033687,
      "balance_loss_mlp": 1.01877236,
      "epoch": 0.7743874943634451,
      "flos": 28729326476160.0,
      "grad_norm": 1.826012095180428,
      "language_loss": 0.79511273,
      "learning_rate": 5.104918994957364e-07,
      "loss": 0.8162154,
      "num_input_tokens_seen": 277871020,
      "step": 12880,
      "time_per_iteration": 2.580899238586426
    },
    {
      "auxiliary_loss_clip": 0.01082381,
      "auxiliary_loss_mlp": 0.01036028,
      "balance_loss_clip": 1.03613126,
      "balance_loss_mlp": 1.02333856,
      "epoch": 0.774447617616113,
      "flos": 21909639312000.0,
      "grad_norm": 1.975896544591322,
      "language_loss": 0.70363802,
      "learning_rate": 5.102320247508847e-07,
      "loss": 0.7248221,
      "num_input_tokens_seen": 277891525,
      "step": 12881,
      "time_per_iteration": 2.524547815322876
    },
    {
      "auxiliary_loss_clip": 0.01086692,
      "auxiliary_loss_mlp": 0.01042475,
      "balance_loss_clip": 1.03522718,
      "balance_loss_mlp": 1.02852178,
      "epoch": 0.774507740868781,
      "flos": 19500643825920.0,
      "grad_norm": 2.592829029585313,
      "language_loss": 0.84567511,
      "learning_rate": 5.099722064981832e-07,
      "loss": 0.86696678,
      "num_input_tokens_seen": 277910425,
      "step": 12882,
      "time_per_iteration": 2.5531561374664307
    },
    {
      "auxiliary_loss_clip": 0.01004385,
      "auxiliary_loss_mlp": 0.01004458,
      "balance_loss_clip": 1.01874757,
      "balance_loss_mlp": 1.00301576,
      "epoch": 0.774567864121449,
      "flos": 59426560402560.0,
      "grad_norm": 0.7703362137093294,
      "language_loss": 0.60428351,
      "learning_rate": 5.097124447474858e-07,
      "loss": 0.62437195,
      "num_input_tokens_seen": 277972795,
      "step": 12883,
      "time_per_iteration": 3.12441349029541
    },
    {
      "auxiliary_loss_clip": 0.01058643,
      "auxiliary_loss_mlp": 0.0103679,
      "balance_loss_clip": 1.03379369,
      "balance_loss_mlp": 1.02231228,
      "epoch": 0.774627987374117,
      "flos": 13225326255360.0,
      "grad_norm": 1.8773506624971381,
      "language_loss": 0.72758168,
      "learning_rate": 5.094527395086416e-07,
      "loss": 0.74853599,
      "num_input_tokens_seen": 277990675,
      "step": 12884,
      "time_per_iteration": 2.53611159324646
    },
    {
      "auxiliary_loss_clip": 0.01094114,
      "auxiliary_loss_mlp": 0.01033708,
      "balance_loss_clip": 1.03651059,
      "balance_loss_mlp": 1.02263331,
      "epoch": 0.7746881106267849,
      "flos": 21394033534080.0,
      "grad_norm": 1.6085790226983008,
      "language_loss": 0.80952728,
      "learning_rate": 5.091930907914986e-07,
      "loss": 0.83080548,
      "num_input_tokens_seen": 278010050,
      "step": 12885,
      "time_per_iteration": 2.493065595626831
    },
    {
      "auxiliary_loss_clip": 0.01102695,
      "auxiliary_loss_mlp": 0.01030324,
      "balance_loss_clip": 1.03513694,
      "balance_loss_mlp": 1.01949382,
      "epoch": 0.7747482338794529,
      "flos": 25629338079360.0,
      "grad_norm": 1.6843929265629287,
      "language_loss": 0.63587934,
      "learning_rate": 5.089334986059029e-07,
      "loss": 0.65720952,
      "num_input_tokens_seen": 278030660,
      "step": 12886,
      "time_per_iteration": 2.4858391284942627
    },
    {
      "auxiliary_loss_clip": 0.01076374,
      "auxiliary_loss_mlp": 0.01032707,
      "balance_loss_clip": 1.03596902,
      "balance_loss_mlp": 1.02129269,
      "epoch": 0.7748083571321208,
      "flos": 11546933402880.0,
      "grad_norm": 1.8596779462921018,
      "language_loss": 0.69464242,
      "learning_rate": 5.086739629616987e-07,
      "loss": 0.71573323,
      "num_input_tokens_seen": 278047645,
      "step": 12887,
      "time_per_iteration": 2.535980224609375
    },
    {
      "auxiliary_loss_clip": 0.01094241,
      "auxiliary_loss_mlp": 0.01027711,
      "balance_loss_clip": 1.0353514,
      "balance_loss_mlp": 1.01640451,
      "epoch": 0.7748684803847888,
      "flos": 19062425900160.0,
      "grad_norm": 1.6941714927295886,
      "language_loss": 0.70730853,
      "learning_rate": 5.084144838687275e-07,
      "loss": 0.72852802,
      "num_input_tokens_seen": 278066170,
      "step": 12888,
      "time_per_iteration": 2.4779396057128906
    },
    {
      "auxiliary_loss_clip": 0.01095742,
      "auxiliary_loss_mlp": 0.01032168,
      "balance_loss_clip": 1.0352267,
      "balance_loss_mlp": 1.01969242,
      "epoch": 0.7749286036374567,
      "flos": 22273162905600.0,
      "grad_norm": 1.6709601342722726,
      "language_loss": 0.8185004,
      "learning_rate": 5.081550613368279e-07,
      "loss": 0.8397795,
      "num_input_tokens_seen": 278085545,
      "step": 12889,
      "time_per_iteration": 2.513124704360962
    },
    {
      "auxiliary_loss_clip": 0.01072397,
      "auxiliary_loss_mlp": 0.0102948,
      "balance_loss_clip": 1.03545356,
      "balance_loss_mlp": 1.01720715,
      "epoch": 0.7749887268901248,
      "flos": 20192462749440.0,
      "grad_norm": 1.9040590529086199,
      "language_loss": 0.79553878,
      "learning_rate": 5.07895695375838e-07,
      "loss": 0.81655753,
      "num_input_tokens_seen": 278102995,
      "step": 12890,
      "time_per_iteration": 2.5174496173858643
    },
    {
      "auxiliary_loss_clip": 0.01079881,
      "auxiliary_loss_mlp": 0.01027596,
      "balance_loss_clip": 1.03914952,
      "balance_loss_mlp": 1.01492977,
      "epoch": 0.7750488501427927,
      "flos": 20337541781760.0,
      "grad_norm": 1.778522388927819,
      "language_loss": 0.66399539,
      "learning_rate": 5.076363859955932e-07,
      "loss": 0.68507022,
      "num_input_tokens_seen": 278121460,
      "step": 12891,
      "time_per_iteration": 2.550403118133545
    },
    {
      "auxiliary_loss_clip": 0.01095306,
      "auxiliary_loss_mlp": 0.01031443,
      "balance_loss_clip": 1.03537035,
      "balance_loss_mlp": 1.01900363,
      "epoch": 0.7751089733954607,
      "flos": 28364043116160.0,
      "grad_norm": 1.4128962772659446,
      "language_loss": 0.7863887,
      "learning_rate": 5.073771332059257e-07,
      "loss": 0.80765617,
      "num_input_tokens_seen": 278143905,
      "step": 12892,
      "time_per_iteration": 2.5386195182800293
    },
    {
      "auxiliary_loss_clip": 0.01097278,
      "auxiliary_loss_mlp": 0.01030534,
      "balance_loss_clip": 1.0372479,
      "balance_loss_mlp": 1.01768947,
      "epoch": 0.7751690966481286,
      "flos": 16943803960320.0,
      "grad_norm": 2.4586518459598303,
      "language_loss": 0.67211932,
      "learning_rate": 5.071179370166669e-07,
      "loss": 0.6933974,
      "num_input_tokens_seen": 278160850,
      "step": 12893,
      "time_per_iteration": 2.4694736003875732
    },
    {
      "auxiliary_loss_clip": 0.0102363,
      "auxiliary_loss_mlp": 0.01003171,
      "balance_loss_clip": 1.0105325,
      "balance_loss_mlp": 1.00200868,
      "epoch": 0.7752292199007966,
      "flos": 65668050339840.0,
      "grad_norm": 0.8123943206604832,
      "language_loss": 0.58554661,
      "learning_rate": 5.068587974376468e-07,
      "loss": 0.60581458,
      "num_input_tokens_seen": 278219950,
      "step": 12894,
      "time_per_iteration": 3.1675500869750977
    },
    {
      "auxiliary_loss_clip": 0.01087477,
      "auxiliary_loss_mlp": 0.01031255,
      "balance_loss_clip": 1.03722823,
      "balance_loss_mlp": 1.01804066,
      "epoch": 0.7752893431534646,
      "flos": 20594662312320.0,
      "grad_norm": 2.000014549500338,
      "language_loss": 0.78188103,
      "learning_rate": 5.065997144786895e-07,
      "loss": 0.8030684,
      "num_input_tokens_seen": 278237805,
      "step": 12895,
      "time_per_iteration": 2.5408387184143066
    },
    {
      "auxiliary_loss_clip": 0.01066107,
      "auxiliary_loss_mlp": 0.0103242,
      "balance_loss_clip": 1.03516161,
      "balance_loss_mlp": 1.01815677,
      "epoch": 0.7753494664061326,
      "flos": 20485350247680.0,
      "grad_norm": 1.7359405239578654,
      "language_loss": 0.6766358,
      "learning_rate": 5.063406881496209e-07,
      "loss": 0.69762111,
      "num_input_tokens_seen": 278257660,
      "step": 12896,
      "time_per_iteration": 2.5418779850006104
    },
    {
      "auxiliary_loss_clip": 0.01083477,
      "auxiliary_loss_mlp": 0.01034399,
      "balance_loss_clip": 1.03624821,
      "balance_loss_mlp": 1.02282357,
      "epoch": 0.7754095896588006,
      "flos": 20265900105600.0,
      "grad_norm": 1.6755887197366992,
      "language_loss": 0.68937492,
      "learning_rate": 5.060817184602629e-07,
      "loss": 0.71055365,
      "num_input_tokens_seen": 278275110,
      "step": 12897,
      "time_per_iteration": 2.5474109649658203
    },
    {
      "auxiliary_loss_clip": 0.01108961,
      "auxiliary_loss_mlp": 0.01033658,
      "balance_loss_clip": 1.03895187,
      "balance_loss_mlp": 1.0205394,
      "epoch": 0.7754697129114685,
      "flos": 23331091201920.0,
      "grad_norm": 1.6138196934878455,
      "language_loss": 0.74925184,
      "learning_rate": 5.058228054204364e-07,
      "loss": 0.77067804,
      "num_input_tokens_seen": 278293035,
      "step": 12898,
      "time_per_iteration": 2.4736642837524414
    },
    {
      "auxiliary_loss_clip": 0.01094929,
      "auxiliary_loss_mlp": 0.00787603,
      "balance_loss_clip": 1.03559113,
      "balance_loss_mlp": 1.01348627,
      "epoch": 0.7755298361641365,
      "flos": 17347619635200.0,
      "grad_norm": 2.060665734551727,
      "language_loss": 0.70018786,
      "learning_rate": 5.055639490399588e-07,
      "loss": 0.71901315,
      "num_input_tokens_seen": 278311010,
      "step": 12899,
      "time_per_iteration": 3.974762201309204
    },
    {
      "auxiliary_loss_clip": 0.01072895,
      "auxiliary_loss_mlp": 0.01036156,
      "balance_loss_clip": 1.0356282,
      "balance_loss_mlp": 1.02291751,
      "epoch": 0.7755899594168044,
      "flos": 19645866512640.0,
      "grad_norm": 1.8978528963525316,
      "language_loss": 0.75144708,
      "learning_rate": 5.053051493286453e-07,
      "loss": 0.77253759,
      "num_input_tokens_seen": 278329900,
      "step": 12900,
      "time_per_iteration": 2.5825138092041016
    },
    {
      "auxiliary_loss_clip": 0.01086255,
      "auxiliary_loss_mlp": 0.01032898,
      "balance_loss_clip": 1.03531981,
      "balance_loss_mlp": 1.02147818,
      "epoch": 0.7756500826694724,
      "flos": 27414457217280.0,
      "grad_norm": 2.133929518407901,
      "language_loss": 0.77101606,
      "learning_rate": 5.050464062963113e-07,
      "loss": 0.7922076,
      "num_input_tokens_seen": 278349980,
      "step": 12901,
      "time_per_iteration": 2.5485048294067383
    },
    {
      "auxiliary_loss_clip": 0.01095639,
      "auxiliary_loss_mlp": 0.01029926,
      "balance_loss_clip": 1.0384109,
      "balance_loss_mlp": 1.01743293,
      "epoch": 0.7757102059221404,
      "flos": 28730511624960.0,
      "grad_norm": 1.4391343791859106,
      "language_loss": 0.77400208,
      "learning_rate": 5.047877199527666e-07,
      "loss": 0.79525763,
      "num_input_tokens_seen": 278372485,
      "step": 12902,
      "time_per_iteration": 2.560375452041626
    },
    {
      "auxiliary_loss_clip": 0.01095245,
      "auxiliary_loss_mlp": 0.01030364,
      "balance_loss_clip": 1.0360148,
      "balance_loss_mlp": 1.01846075,
      "epoch": 0.7757703291748084,
      "flos": 22486795044480.0,
      "grad_norm": 3.003738254486541,
      "language_loss": 0.73018539,
      "learning_rate": 5.045290903078215e-07,
      "loss": 0.75144148,
      "num_input_tokens_seen": 278391660,
      "step": 12903,
      "time_per_iteration": 5.283673048019409
    },
    {
      "auxiliary_loss_clip": 0.01084367,
      "auxiliary_loss_mlp": 0.01025487,
      "balance_loss_clip": 1.03799486,
      "balance_loss_mlp": 1.01376295,
      "epoch": 0.7758304524274763,
      "flos": 21430159637760.0,
      "grad_norm": 2.254434906746113,
      "language_loss": 0.76045573,
      "learning_rate": 5.042705173712835e-07,
      "loss": 0.78155422,
      "num_input_tokens_seen": 278409125,
      "step": 12904,
      "time_per_iteration": 2.514991044998169
    },
    {
      "auxiliary_loss_clip": 0.01102528,
      "auxiliary_loss_mlp": 0.01027726,
      "balance_loss_clip": 1.03629446,
      "balance_loss_mlp": 1.01615047,
      "epoch": 0.7758905756801443,
      "flos": 23659242877440.0,
      "grad_norm": 2.102372589575635,
      "language_loss": 0.68248528,
      "learning_rate": 5.040120011529576e-07,
      "loss": 0.7037878,
      "num_input_tokens_seen": 278429450,
      "step": 12905,
      "time_per_iteration": 2.475625514984131
    },
    {
      "auxiliary_loss_clip": 0.01090951,
      "auxiliary_loss_mlp": 0.00785341,
      "balance_loss_clip": 1.03968632,
      "balance_loss_mlp": 1.01316357,
      "epoch": 0.7759506989328122,
      "flos": 28365479660160.0,
      "grad_norm": 1.6393277936426756,
      "language_loss": 0.675403,
      "learning_rate": 5.037535416626459e-07,
      "loss": 0.69416595,
      "num_input_tokens_seen": 278449925,
      "step": 12906,
      "time_per_iteration": 2.5448105335235596
    },
    {
      "auxiliary_loss_clip": 0.01068475,
      "auxiliary_loss_mlp": 0.01031465,
      "balance_loss_clip": 1.03376281,
      "balance_loss_mlp": 1.01879883,
      "epoch": 0.7760108221854802,
      "flos": 14902785354240.0,
      "grad_norm": 1.8986056788164603,
      "language_loss": 0.80986983,
      "learning_rate": 5.034951389101498e-07,
      "loss": 0.8308692,
      "num_input_tokens_seen": 278467255,
      "step": 12907,
      "time_per_iteration": 2.529893159866333
    },
    {
      "auxiliary_loss_clip": 0.01091782,
      "auxiliary_loss_mlp": 0.01034434,
      "balance_loss_clip": 1.03646803,
      "balance_loss_mlp": 1.02251351,
      "epoch": 0.7760709454381483,
      "flos": 14792503622400.0,
      "grad_norm": 2.055720138657292,
      "language_loss": 0.67554367,
      "learning_rate": 5.032367929052685e-07,
      "loss": 0.69680583,
      "num_input_tokens_seen": 278484250,
      "step": 12908,
      "time_per_iteration": 2.463670015335083
    },
    {
      "auxiliary_loss_clip": 0.01074398,
      "auxiliary_loss_mlp": 0.01038094,
      "balance_loss_clip": 1.03530741,
      "balance_loss_mlp": 1.02572584,
      "epoch": 0.7761310686908162,
      "flos": 17379831156480.0,
      "grad_norm": 1.6513786656273939,
      "language_loss": 0.70369214,
      "learning_rate": 5.029785036577976e-07,
      "loss": 0.72481704,
      "num_input_tokens_seen": 278502740,
      "step": 12909,
      "time_per_iteration": 3.933603525161743
    },
    {
      "auxiliary_loss_clip": 0.01091856,
      "auxiliary_loss_mlp": 0.01034501,
      "balance_loss_clip": 1.03556156,
      "balance_loss_mlp": 1.02287793,
      "epoch": 0.7761911919434842,
      "flos": 25556547168000.0,
      "grad_norm": 1.66737617909252,
      "language_loss": 0.67869174,
      "learning_rate": 5.027202711775324e-07,
      "loss": 0.69995534,
      "num_input_tokens_seen": 278523890,
      "step": 12910,
      "time_per_iteration": 2.5230565071105957
    },
    {
      "auxiliary_loss_clip": 0.01060684,
      "auxiliary_loss_mlp": 0.01034601,
      "balance_loss_clip": 1.0361762,
      "balance_loss_mlp": 1.02315676,
      "epoch": 0.7762513151961521,
      "flos": 23179763203200.0,
      "grad_norm": 1.5998886340337886,
      "language_loss": 0.71941864,
      "learning_rate": 5.024620954742646e-07,
      "loss": 0.74037147,
      "num_input_tokens_seen": 278543185,
      "step": 12911,
      "time_per_iteration": 2.6369431018829346
    },
    {
      "auxiliary_loss_clip": 0.01108764,
      "auxiliary_loss_mlp": 0.00785307,
      "balance_loss_clip": 1.03829813,
      "balance_loss_mlp": 1.00792587,
      "epoch": 0.7763114384488201,
      "flos": 21689614552320.0,
      "grad_norm": 2.8794336503562725,
      "language_loss": 0.63379043,
      "learning_rate": 5.022039765577836e-07,
      "loss": 0.65273118,
      "num_input_tokens_seen": 278559220,
      "step": 12912,
      "time_per_iteration": 2.4639651775360107
    },
    {
      "auxiliary_loss_clip": 0.01011859,
      "auxiliary_loss_mlp": 0.01003198,
      "balance_loss_clip": 1.01132727,
      "balance_loss_mlp": 1.00188696,
      "epoch": 0.776371561701488,
      "flos": 69025554316800.0,
      "grad_norm": 0.7783884063267448,
      "language_loss": 0.53276455,
      "learning_rate": 5.019459144378779e-07,
      "loss": 0.55291522,
      "num_input_tokens_seen": 278618185,
      "step": 12913,
      "time_per_iteration": 3.2307779788970947
    },
    {
      "auxiliary_loss_clip": 0.01085818,
      "auxiliary_loss_mlp": 0.01031065,
      "balance_loss_clip": 1.03763425,
      "balance_loss_mlp": 1.01878595,
      "epoch": 0.776431684954156,
      "flos": 22893914770560.0,
      "grad_norm": 2.0391695118996567,
      "language_loss": 0.61929512,
      "learning_rate": 5.016879091243338e-07,
      "loss": 0.64046395,
      "num_input_tokens_seen": 278636210,
      "step": 12914,
      "time_per_iteration": 2.580322027206421
    },
    {
      "auxiliary_loss_clip": 0.0108508,
      "auxiliary_loss_mlp": 0.01032367,
      "balance_loss_clip": 1.03780627,
      "balance_loss_mlp": 1.0201602,
      "epoch": 0.776491808206824,
      "flos": 20261554560000.0,
      "grad_norm": 2.3077914384482345,
      "language_loss": 0.82444942,
      "learning_rate": 5.014299606269339e-07,
      "loss": 0.84562391,
      "num_input_tokens_seen": 278653305,
      "step": 12915,
      "time_per_iteration": 2.5242555141448975
    },
    {
      "auxiliary_loss_clip": 0.01094705,
      "auxiliary_loss_mlp": 0.01033688,
      "balance_loss_clip": 1.03775442,
      "balance_loss_mlp": 1.02068806,
      "epoch": 0.776551931459492,
      "flos": 26759051706240.0,
      "grad_norm": 1.6455940207899116,
      "language_loss": 0.74796617,
      "learning_rate": 5.011720689554603e-07,
      "loss": 0.76925009,
      "num_input_tokens_seen": 278671850,
      "step": 12916,
      "time_per_iteration": 2.5551278591156006
    },
    {
      "auxiliary_loss_clip": 0.01052424,
      "auxiliary_loss_mlp": 0.01035699,
      "balance_loss_clip": 1.03647602,
      "balance_loss_mlp": 1.02183461,
      "epoch": 0.7766120547121599,
      "flos": 52665080250240.0,
      "grad_norm": 1.4817817907499973,
      "language_loss": 0.65755397,
      "learning_rate": 5.009142341196919e-07,
      "loss": 0.67843521,
      "num_input_tokens_seen": 278697860,
      "step": 12917,
      "time_per_iteration": 2.9118571281433105
    },
    {
      "auxiliary_loss_clip": 0.01094318,
      "auxiliary_loss_mlp": 0.0103097,
      "balance_loss_clip": 1.03384984,
      "balance_loss_mlp": 1.018978,
      "epoch": 0.7766721779648279,
      "flos": 25156215112320.0,
      "grad_norm": 1.4138640183735087,
      "language_loss": 0.64509696,
      "learning_rate": 5.006564561294065e-07,
      "loss": 0.66634983,
      "num_input_tokens_seen": 278720655,
      "step": 12918,
      "time_per_iteration": 2.5666329860687256
    },
    {
      "auxiliary_loss_clip": 0.01104898,
      "auxiliary_loss_mlp": 0.01030696,
      "balance_loss_clip": 1.03653657,
      "balance_loss_mlp": 1.01881671,
      "epoch": 0.7767323012174958,
      "flos": 23760761690880.0,
      "grad_norm": 2.0305901212625934,
      "language_loss": 0.73539841,
      "learning_rate": 5.003987349943777e-07,
      "loss": 0.75675428,
      "num_input_tokens_seen": 278737375,
      "step": 12919,
      "time_per_iteration": 2.493507146835327
    },
    {
      "auxiliary_loss_clip": 0.01064181,
      "auxiliary_loss_mlp": 0.01031484,
      "balance_loss_clip": 1.03635013,
      "balance_loss_mlp": 1.01829338,
      "epoch": 0.7767924244701638,
      "flos": 22086642556800.0,
      "grad_norm": 1.7212368572927992,
      "language_loss": 0.79109669,
      "learning_rate": 5.001410707243792e-07,
      "loss": 0.81205332,
      "num_input_tokens_seen": 278756510,
      "step": 12920,
      "time_per_iteration": 2.623487949371338
    },
    {
      "auxiliary_loss_clip": 0.01096629,
      "auxiliary_loss_mlp": 0.01029992,
      "balance_loss_clip": 1.03744912,
      "balance_loss_mlp": 1.01791596,
      "epoch": 0.7768525477228319,
      "flos": 21981640124160.0,
      "grad_norm": 1.5820370882808008,
      "language_loss": 0.70773095,
      "learning_rate": 4.998834633291829e-07,
      "loss": 0.72899717,
      "num_input_tokens_seen": 278775410,
      "step": 12921,
      "time_per_iteration": 2.49204158782959
    },
    {
      "auxiliary_loss_clip": 0.01099695,
      "auxiliary_loss_mlp": 0.01031532,
      "balance_loss_clip": 1.03784323,
      "balance_loss_mlp": 1.01837134,
      "epoch": 0.7769126709754998,
      "flos": 21794581071360.0,
      "grad_norm": 1.71344842786542,
      "language_loss": 0.76095855,
      "learning_rate": 4.996259128185547e-07,
      "loss": 0.78227085,
      "num_input_tokens_seen": 278794260,
      "step": 12922,
      "time_per_iteration": 2.545332431793213
    },
    {
      "auxiliary_loss_clip": 0.01058167,
      "auxiliary_loss_mlp": 0.01036864,
      "balance_loss_clip": 1.03500843,
      "balance_loss_mlp": 1.02432954,
      "epoch": 0.7769727942281678,
      "flos": 20047994248320.0,
      "grad_norm": 1.6932111265116463,
      "language_loss": 0.80342996,
      "learning_rate": 4.993684192022625e-07,
      "loss": 0.82438022,
      "num_input_tokens_seen": 278813290,
      "step": 12923,
      "time_per_iteration": 2.5614752769470215
    },
    {
      "auxiliary_loss_clip": 0.01068037,
      "auxiliary_loss_mlp": 0.01031441,
      "balance_loss_clip": 1.03732896,
      "balance_loss_mlp": 1.01944304,
      "epoch": 0.7770329174808357,
      "flos": 21686777377920.0,
      "grad_norm": 1.9328724881114492,
      "language_loss": 0.92231548,
      "learning_rate": 4.991109824900699e-07,
      "loss": 0.94331026,
      "num_input_tokens_seen": 278830610,
      "step": 12924,
      "time_per_iteration": 2.565948486328125
    },
    {
      "auxiliary_loss_clip": 0.01093495,
      "auxiliary_loss_mlp": 0.0102949,
      "balance_loss_clip": 1.03521216,
      "balance_loss_mlp": 1.01697922,
      "epoch": 0.7770930407335037,
      "flos": 25849255098240.0,
      "grad_norm": 2.048887916207168,
      "language_loss": 0.66275996,
      "learning_rate": 4.988536026917401e-07,
      "loss": 0.68398976,
      "num_input_tokens_seen": 278849530,
      "step": 12925,
      "time_per_iteration": 2.5196340084075928
    },
    {
      "auxiliary_loss_clip": 0.01074588,
      "auxiliary_loss_mlp": 0.0103152,
      "balance_loss_clip": 1.03689957,
      "balance_loss_mlp": 1.01944995,
      "epoch": 0.7771531639861716,
      "flos": 24347865490560.0,
      "grad_norm": 1.6469959909624128,
      "language_loss": 0.71788496,
      "learning_rate": 4.985962798170314e-07,
      "loss": 0.73894602,
      "num_input_tokens_seen": 278869005,
      "step": 12926,
      "time_per_iteration": 2.5999913215637207
    },
    {
      "auxiliary_loss_clip": 0.01096656,
      "auxiliary_loss_mlp": 0.01027679,
      "balance_loss_clip": 1.03615761,
      "balance_loss_mlp": 1.01508999,
      "epoch": 0.7772132872388396,
      "flos": 25629948610560.0,
      "grad_norm": 1.7004420593574334,
      "language_loss": 0.65462631,
      "learning_rate": 4.983390138757027e-07,
      "loss": 0.67586958,
      "num_input_tokens_seen": 278888790,
      "step": 12927,
      "time_per_iteration": 2.5170023441314697
    },
    {
      "auxiliary_loss_clip": 0.01084328,
      "auxiliary_loss_mlp": 0.01035967,
      "balance_loss_clip": 1.03704381,
      "balance_loss_mlp": 1.02313423,
      "epoch": 0.7772734104915076,
      "flos": 26067412350720.0,
      "grad_norm": 1.778573033140268,
      "language_loss": 0.72477388,
      "learning_rate": 4.980818048775093e-07,
      "loss": 0.74597692,
      "num_input_tokens_seen": 278908150,
      "step": 12928,
      "time_per_iteration": 2.5482072830200195
    },
    {
      "auxiliary_loss_clip": 0.01062697,
      "auxiliary_loss_mlp": 0.01031426,
      "balance_loss_clip": 1.03371322,
      "balance_loss_mlp": 1.01871252,
      "epoch": 0.7773335337441756,
      "flos": 22925048883840.0,
      "grad_norm": 1.704993161563091,
      "language_loss": 0.74435419,
      "learning_rate": 4.978246528322036e-07,
      "loss": 0.76529545,
      "num_input_tokens_seen": 278927425,
      "step": 12929,
      "time_per_iteration": 2.587257146835327
    },
    {
      "auxiliary_loss_clip": 0.01069701,
      "auxiliary_loss_mlp": 0.01030489,
      "balance_loss_clip": 1.03721619,
      "balance_loss_mlp": 1.01828194,
      "epoch": 0.7773936569968435,
      "flos": 20776765288320.0,
      "grad_norm": 1.7848517174143814,
      "language_loss": 0.7757982,
      "learning_rate": 4.975675577495377e-07,
      "loss": 0.79680002,
      "num_input_tokens_seen": 278946475,
      "step": 12930,
      "time_per_iteration": 2.559598207473755
    },
    {
      "auxiliary_loss_clip": 0.01107516,
      "auxiliary_loss_mlp": 0.01032414,
      "balance_loss_clip": 1.0388751,
      "balance_loss_mlp": 1.01999879,
      "epoch": 0.7774537802495115,
      "flos": 20372267255040.0,
      "grad_norm": 1.924953217864129,
      "language_loss": 0.80043161,
      "learning_rate": 4.973105196392613e-07,
      "loss": 0.82183093,
      "num_input_tokens_seen": 278964345,
      "step": 12931,
      "time_per_iteration": 2.4600746631622314
    },
    {
      "auxiliary_loss_clip": 0.01013554,
      "auxiliary_loss_mlp": 0.01002453,
      "balance_loss_clip": 1.03431642,
      "balance_loss_mlp": 1.00084364,
      "epoch": 0.7775139035021794,
      "flos": 53912081738880.0,
      "grad_norm": 0.8164212293045781,
      "language_loss": 0.59782535,
      "learning_rate": 4.970535385111199e-07,
      "loss": 0.61798543,
      "num_input_tokens_seen": 279022380,
      "step": 12932,
      "time_per_iteration": 3.1208577156066895
    },
    {
      "auxiliary_loss_clip": 0.01096798,
      "auxiliary_loss_mlp": 0.01031218,
      "balance_loss_clip": 1.03758717,
      "balance_loss_mlp": 1.01965475,
      "epoch": 0.7775740267548474,
      "flos": 28842481296000.0,
      "grad_norm": 1.4289576189573685,
      "language_loss": 0.76217765,
      "learning_rate": 4.967966143748595e-07,
      "loss": 0.78345782,
      "num_input_tokens_seen": 279044275,
      "step": 12933,
      "time_per_iteration": 2.5824007987976074
    },
    {
      "auxiliary_loss_clip": 0.01087262,
      "auxiliary_loss_mlp": 0.01033477,
      "balance_loss_clip": 1.03740454,
      "balance_loss_mlp": 1.02067435,
      "epoch": 0.7776341500075155,
      "flos": 21872471713920.0,
      "grad_norm": 2.3520390802465974,
      "language_loss": 0.72849488,
      "learning_rate": 4.965397472402215e-07,
      "loss": 0.74970233,
      "num_input_tokens_seen": 279063375,
      "step": 12934,
      "time_per_iteration": 2.5667359828948975
    },
    {
      "auxiliary_loss_clip": 0.01062665,
      "auxiliary_loss_mlp": 0.01026732,
      "balance_loss_clip": 1.03363681,
      "balance_loss_mlp": 1.01379132,
      "epoch": 0.7776942732601834,
      "flos": 20229845829120.0,
      "grad_norm": 1.853996513206363,
      "language_loss": 0.70207101,
      "learning_rate": 4.962829371169475e-07,
      "loss": 0.722965,
      "num_input_tokens_seen": 279082680,
      "step": 12935,
      "time_per_iteration": 2.5957260131835938
    },
    {
      "auxiliary_loss_clip": 0.01081086,
      "auxiliary_loss_mlp": 0.0078621,
      "balance_loss_clip": 1.03693163,
      "balance_loss_mlp": 1.01139534,
      "epoch": 0.7777543965128514,
      "flos": 22231829329920.0,
      "grad_norm": 1.7640495639235767,
      "language_loss": 0.83631837,
      "learning_rate": 4.960261840147746e-07,
      "loss": 0.85499132,
      "num_input_tokens_seen": 279099805,
      "step": 12936,
      "time_per_iteration": 2.5335693359375
    },
    {
      "auxiliary_loss_clip": 0.01099009,
      "auxiliary_loss_mlp": 0.01027458,
      "balance_loss_clip": 1.03612542,
      "balance_loss_mlp": 1.01579905,
      "epoch": 0.7778145197655193,
      "flos": 14501950508160.0,
      "grad_norm": 1.9114266800232635,
      "language_loss": 0.67606688,
      "learning_rate": 4.957694879434397e-07,
      "loss": 0.69733155,
      "num_input_tokens_seen": 279117975,
      "step": 12937,
      "time_per_iteration": 2.483898878097534
    },
    {
      "auxiliary_loss_clip": 0.01106144,
      "auxiliary_loss_mlp": 0.01028766,
      "balance_loss_clip": 1.03572464,
      "balance_loss_mlp": 1.01664877,
      "epoch": 0.7778746430181873,
      "flos": 21140288881920.0,
      "grad_norm": 1.4442072333579719,
      "language_loss": 0.87286329,
      "learning_rate": 4.955128489126777e-07,
      "loss": 0.89421242,
      "num_input_tokens_seen": 279137255,
      "step": 12938,
      "time_per_iteration": 3.897878646850586
    },
    {
      "auxiliary_loss_clip": 0.01094748,
      "auxiliary_loss_mlp": 0.01032333,
      "balance_loss_clip": 1.03606975,
      "balance_loss_mlp": 1.01972044,
      "epoch": 0.7779347662708552,
      "flos": 20266366982400.0,
      "grad_norm": 1.9771756057164271,
      "language_loss": 0.85028434,
      "learning_rate": 4.95256266932218e-07,
      "loss": 0.87155515,
      "num_input_tokens_seen": 279154500,
      "step": 12939,
      "time_per_iteration": 2.5126852989196777
    },
    {
      "auxiliary_loss_clip": 0.01103551,
      "auxiliary_loss_mlp": 0.00785608,
      "balance_loss_clip": 1.03654099,
      "balance_loss_mlp": 1.01334047,
      "epoch": 0.7779948895235232,
      "flos": 19209013303680.0,
      "grad_norm": 1.68106654703397,
      "language_loss": 0.6903913,
      "learning_rate": 4.949997420117915e-07,
      "loss": 0.70928288,
      "num_input_tokens_seen": 279173635,
      "step": 12940,
      "time_per_iteration": 2.45277738571167
    },
    {
      "auxiliary_loss_clip": 0.0107101,
      "auxiliary_loss_mlp": 0.01026806,
      "balance_loss_clip": 1.03447509,
      "balance_loss_mlp": 1.01541531,
      "epoch": 0.7780550127761912,
      "flos": 23914711382400.0,
      "grad_norm": 1.5054133372951808,
      "language_loss": 0.77872694,
      "learning_rate": 4.947432741611255e-07,
      "loss": 0.79970515,
      "num_input_tokens_seen": 279194430,
      "step": 12941,
      "time_per_iteration": 3.989126205444336
    },
    {
      "auxiliary_loss_clip": 0.01100105,
      "auxiliary_loss_mlp": 0.01033884,
      "balance_loss_clip": 1.03739524,
      "balance_loss_mlp": 1.02048492,
      "epoch": 0.7781151360288592,
      "flos": 32415951795840.0,
      "grad_norm": 2.2106376287375618,
      "language_loss": 0.73262519,
      "learning_rate": 4.944868633899462e-07,
      "loss": 0.75396514,
      "num_input_tokens_seen": 279212920,
      "step": 12942,
      "time_per_iteration": 4.04442286491394
    },
    {
      "auxiliary_loss_clip": 0.01051446,
      "auxiliary_loss_mlp": 0.01039755,
      "balance_loss_clip": 1.03435421,
      "balance_loss_mlp": 1.02670765,
      "epoch": 0.7781752592815271,
      "flos": 22346384780160.0,
      "grad_norm": 2.841527811606394,
      "language_loss": 0.67912912,
      "learning_rate": 4.942305097079751e-07,
      "loss": 0.70004112,
      "num_input_tokens_seen": 279232310,
      "step": 12943,
      "time_per_iteration": 2.588712692260742
    },
    {
      "auxiliary_loss_clip": 0.01014298,
      "auxiliary_loss_mlp": 0.01015442,
      "balance_loss_clip": 1.01334476,
      "balance_loss_mlp": 1.01390445,
      "epoch": 0.7782353825341951,
      "flos": 70460183520000.0,
      "grad_norm": 0.7808201226025938,
      "language_loss": 0.58488983,
      "learning_rate": 4.939742131249347e-07,
      "loss": 0.60518724,
      "num_input_tokens_seen": 279295375,
      "step": 12944,
      "time_per_iteration": 3.306283712387085
    },
    {
      "auxiliary_loss_clip": 0.01107599,
      "auxiliary_loss_mlp": 0.01036376,
      "balance_loss_clip": 1.03653681,
      "balance_loss_mlp": 1.02300692,
      "epoch": 0.778295505786863,
      "flos": 19062569554560.0,
      "grad_norm": 2.444858844469556,
      "language_loss": 0.67997372,
      "learning_rate": 4.937179736505428e-07,
      "loss": 0.70141351,
      "num_input_tokens_seen": 279313660,
      "step": 12945,
      "time_per_iteration": 2.4550700187683105
    },
    {
      "auxiliary_loss_clip": 0.01093082,
      "auxiliary_loss_mlp": 0.01033102,
      "balance_loss_clip": 1.03604198,
      "balance_loss_mlp": 1.02049005,
      "epoch": 0.778355629039531,
      "flos": 20999734963200.0,
      "grad_norm": 1.8182410455411544,
      "language_loss": 0.69411945,
      "learning_rate": 4.93461791294516e-07,
      "loss": 0.71538132,
      "num_input_tokens_seen": 279334495,
      "step": 12946,
      "time_per_iteration": 2.5272586345672607
    },
    {
      "auxiliary_loss_clip": 0.01106675,
      "auxiliary_loss_mlp": 0.01029618,
      "balance_loss_clip": 1.0377512,
      "balance_loss_mlp": 1.0169524,
      "epoch": 0.7784157522921991,
      "flos": 21398091770880.0,
      "grad_norm": 1.7815527393535227,
      "language_loss": 0.65527618,
      "learning_rate": 4.932056660665689e-07,
      "loss": 0.6766392,
      "num_input_tokens_seen": 279352985,
      "step": 12947,
      "time_per_iteration": 2.4667985439300537
    },
    {
      "auxiliary_loss_clip": 0.01044243,
      "auxiliary_loss_mlp": 0.01038832,
      "balance_loss_clip": 1.03271627,
      "balance_loss_mlp": 1.02490819,
      "epoch": 0.778475875544867,
      "flos": 20813861059200.0,
      "grad_norm": 2.5029606321594815,
      "language_loss": 0.65083426,
      "learning_rate": 4.929495979764147e-07,
      "loss": 0.67166501,
      "num_input_tokens_seen": 279371360,
      "step": 12948,
      "time_per_iteration": 4.001905679702759
    },
    {
      "auxiliary_loss_clip": 0.01106456,
      "auxiliary_loss_mlp": 0.01032234,
      "balance_loss_clip": 1.03699863,
      "balance_loss_mlp": 1.01980615,
      "epoch": 0.778535998797535,
      "flos": 14355363104640.0,
      "grad_norm": 1.704056838096069,
      "language_loss": 0.75091684,
      "learning_rate": 4.926935870337625e-07,
      "loss": 0.77230376,
      "num_input_tokens_seen": 279389400,
      "step": 12949,
      "time_per_iteration": 2.46199107170105
    },
    {
      "auxiliary_loss_clip": 0.01110088,
      "auxiliary_loss_mlp": 0.01032456,
      "balance_loss_clip": 1.03796172,
      "balance_loss_mlp": 1.01987922,
      "epoch": 0.7785961220502029,
      "flos": 19209552007680.0,
      "grad_norm": 1.6082704531930117,
      "language_loss": 0.69083387,
      "learning_rate": 4.924376332483202e-07,
      "loss": 0.71225929,
      "num_input_tokens_seen": 279409715,
      "step": 12950,
      "time_per_iteration": 2.475525379180908
    },
    {
      "auxiliary_loss_clip": 0.01089174,
      "auxiliary_loss_mlp": 0.01032574,
      "balance_loss_clip": 1.03557587,
      "balance_loss_mlp": 1.02020586,
      "epoch": 0.7786562453028709,
      "flos": 25738757884800.0,
      "grad_norm": 1.706537559210045,
      "language_loss": 0.71940571,
      "learning_rate": 4.921817366297938e-07,
      "loss": 0.74062324,
      "num_input_tokens_seen": 279427705,
      "step": 12951,
      "time_per_iteration": 2.5303313732147217
    },
    {
      "auxiliary_loss_clip": 0.01082316,
      "auxiliary_loss_mlp": 0.01034274,
      "balance_loss_clip": 1.03553927,
      "balance_loss_mlp": 1.02209067,
      "epoch": 0.7787163685555388,
      "flos": 25739440243200.0,
      "grad_norm": 1.859370861220475,
      "language_loss": 0.65541434,
      "learning_rate": 4.919258971878877e-07,
      "loss": 0.67658025,
      "num_input_tokens_seen": 279448215,
      "step": 12952,
      "time_per_iteration": 2.5791444778442383
    },
    {
      "auxiliary_loss_clip": 0.01075958,
      "auxiliary_loss_mlp": 0.01027231,
      "balance_loss_clip": 1.03353834,
      "balance_loss_mlp": 1.01609087,
      "epoch": 0.7787764918082068,
      "flos": 22747722416640.0,
      "grad_norm": 1.712883752670185,
      "language_loss": 0.81388128,
      "learning_rate": 4.916701149323022e-07,
      "loss": 0.83491313,
      "num_input_tokens_seen": 279466260,
      "step": 12953,
      "time_per_iteration": 2.5463006496429443
    },
    {
      "auxiliary_loss_clip": 0.01111478,
      "auxiliary_loss_mlp": 0.01030696,
      "balance_loss_clip": 1.04009604,
      "balance_loss_mlp": 1.018489,
      "epoch": 0.7788366150608748,
      "flos": 15190860430080.0,
      "grad_norm": 2.1618789658045756,
      "language_loss": 0.7686497,
      "learning_rate": 4.91414389872737e-07,
      "loss": 0.79007137,
      "num_input_tokens_seen": 279484520,
      "step": 12954,
      "time_per_iteration": 2.450113296508789
    },
    {
      "auxiliary_loss_clip": 0.01095679,
      "auxiliary_loss_mlp": 0.01031767,
      "balance_loss_clip": 1.03582919,
      "balance_loss_mlp": 1.01986408,
      "epoch": 0.7788967383135428,
      "flos": 21210242618880.0,
      "grad_norm": 1.6873628181255842,
      "language_loss": 0.72919893,
      "learning_rate": 4.911587220188905e-07,
      "loss": 0.75047338,
      "num_input_tokens_seen": 279503130,
      "step": 12955,
      "time_per_iteration": 2.502089500427246
    },
    {
      "auxiliary_loss_clip": 0.01072583,
      "auxiliary_loss_mlp": 0.01034651,
      "balance_loss_clip": 1.03347862,
      "balance_loss_mlp": 1.02202129,
      "epoch": 0.7789568615662107,
      "flos": 21682970536320.0,
      "grad_norm": 1.9662635334752019,
      "language_loss": 0.68497795,
      "learning_rate": 4.909031113804551e-07,
      "loss": 0.70605028,
      "num_input_tokens_seen": 279521930,
      "step": 12956,
      "time_per_iteration": 2.551457405090332
    },
    {
      "auxiliary_loss_clip": 0.01071113,
      "auxiliary_loss_mlp": 0.01032668,
      "balance_loss_clip": 1.03547525,
      "balance_loss_mlp": 1.02085447,
      "epoch": 0.7790169848188787,
      "flos": 26360371676160.0,
      "grad_norm": 1.5888472766298276,
      "language_loss": 0.75827169,
      "learning_rate": 4.906475579671252e-07,
      "loss": 0.77930951,
      "num_input_tokens_seen": 279542375,
      "step": 12957,
      "time_per_iteration": 2.6357386112213135
    },
    {
      "auxiliary_loss_clip": 0.01035153,
      "auxiliary_loss_mlp": 0.01028722,
      "balance_loss_clip": 1.03603792,
      "balance_loss_mlp": 1.0165925,
      "epoch": 0.7790771080715466,
      "flos": 25516183259520.0,
      "grad_norm": 2.050122423796401,
      "language_loss": 0.77743489,
      "learning_rate": 4.903920617885917e-07,
      "loss": 0.79807359,
      "num_input_tokens_seen": 279561885,
      "step": 12958,
      "time_per_iteration": 2.762251615524292
    },
    {
      "auxiliary_loss_clip": 0.01092258,
      "auxiliary_loss_mlp": 0.01042206,
      "balance_loss_clip": 1.03527546,
      "balance_loss_mlp": 1.02774,
      "epoch": 0.7791372313242146,
      "flos": 16034186920320.0,
      "grad_norm": 2.278057625011404,
      "language_loss": 0.71247244,
      "learning_rate": 4.901366228545418e-07,
      "loss": 0.7338171,
      "num_input_tokens_seen": 279579965,
      "step": 12959,
      "time_per_iteration": 2.6730854511260986
    },
    {
      "auxiliary_loss_clip": 0.01096092,
      "auxiliary_loss_mlp": 0.00784726,
      "balance_loss_clip": 1.0374707,
      "balance_loss_mlp": 1.01091146,
      "epoch": 0.7791973545768827,
      "flos": 23842207779840.0,
      "grad_norm": 1.5703243308243429,
      "language_loss": 0.77918518,
      "learning_rate": 4.898812411746632e-07,
      "loss": 0.79799342,
      "num_input_tokens_seen": 279599030,
      "step": 12960,
      "time_per_iteration": 2.5081260204315186
    },
    {
      "auxiliary_loss_clip": 0.01098066,
      "auxiliary_loss_mlp": 0.01034913,
      "balance_loss_clip": 1.03739309,
      "balance_loss_mlp": 1.02242017,
      "epoch": 0.7792574778295506,
      "flos": 24168384207360.0,
      "grad_norm": 1.782267142108547,
      "language_loss": 0.75358009,
      "learning_rate": 4.896259167586385e-07,
      "loss": 0.77490985,
      "num_input_tokens_seen": 279614400,
      "step": 12961,
      "time_per_iteration": 2.506835460662842
    },
    {
      "auxiliary_loss_clip": 0.01082938,
      "auxiliary_loss_mlp": 0.01036352,
      "balance_loss_clip": 1.03910542,
      "balance_loss_mlp": 1.02468777,
      "epoch": 0.7793176010822186,
      "flos": 21464921024640.0,
      "grad_norm": 1.583035338536796,
      "language_loss": 0.73798847,
      "learning_rate": 4.893706496161511e-07,
      "loss": 0.75918138,
      "num_input_tokens_seen": 279633745,
      "step": 12962,
      "time_per_iteration": 2.5578320026397705
    },
    {
      "auxiliary_loss_clip": 0.01094711,
      "auxiliary_loss_mlp": 0.01027507,
      "balance_loss_clip": 1.03663731,
      "balance_loss_mlp": 1.0159024,
      "epoch": 0.7793777243348865,
      "flos": 20666699038080.0,
      "grad_norm": 1.9736474787128522,
      "language_loss": 0.69942188,
      "learning_rate": 4.891154397568795e-07,
      "loss": 0.72064412,
      "num_input_tokens_seen": 279651165,
      "step": 12963,
      "time_per_iteration": 2.4803755283355713
    },
    {
      "auxiliary_loss_clip": 0.0109519,
      "auxiliary_loss_mlp": 0.00784175,
      "balance_loss_clip": 1.0374186,
      "balance_loss_mlp": 1.01131034,
      "epoch": 0.7794378475875545,
      "flos": 27125771610240.0,
      "grad_norm": 1.7729299483010115,
      "language_loss": 0.63710546,
      "learning_rate": 4.888602871905019e-07,
      "loss": 0.65589911,
      "num_input_tokens_seen": 279671175,
      "step": 12964,
      "time_per_iteration": 2.5842459201812744
    },
    {
      "auxiliary_loss_clip": 0.01087018,
      "auxiliary_loss_mlp": 0.01031903,
      "balance_loss_clip": 1.03650248,
      "balance_loss_mlp": 1.01992297,
      "epoch": 0.7794979708402224,
      "flos": 28074136446720.0,
      "grad_norm": 1.623509227811248,
      "language_loss": 0.76634622,
      "learning_rate": 4.88605191926694e-07,
      "loss": 0.78753543,
      "num_input_tokens_seen": 279688675,
      "step": 12965,
      "time_per_iteration": 2.573005199432373
    },
    {
      "auxiliary_loss_clip": 0.01085036,
      "auxiliary_loss_mlp": 0.01036238,
      "balance_loss_clip": 1.03390372,
      "balance_loss_mlp": 1.02366734,
      "epoch": 0.7795580940928905,
      "flos": 26869548919680.0,
      "grad_norm": 1.4252090849960406,
      "language_loss": 0.72780955,
      "learning_rate": 4.883501539751289e-07,
      "loss": 0.74902236,
      "num_input_tokens_seen": 279710245,
      "step": 12966,
      "time_per_iteration": 2.541987180709839
    },
    {
      "auxiliary_loss_clip": 0.01083125,
      "auxiliary_loss_mlp": 0.00781556,
      "balance_loss_clip": 1.03764391,
      "balance_loss_mlp": 1.00690246,
      "epoch": 0.7796182173455584,
      "flos": 23835384195840.0,
      "grad_norm": 1.4780310990132308,
      "language_loss": 0.74451321,
      "learning_rate": 4.880951733454768e-07,
      "loss": 0.76315999,
      "num_input_tokens_seen": 279729045,
      "step": 12967,
      "time_per_iteration": 2.5361344814300537
    },
    {
      "auxiliary_loss_clip": 0.01107239,
      "auxiliary_loss_mlp": 0.01031308,
      "balance_loss_clip": 1.03791785,
      "balance_loss_mlp": 1.01852298,
      "epoch": 0.7796783405982264,
      "flos": 19792238434560.0,
      "grad_norm": 2.275931410346895,
      "language_loss": 0.71849239,
      "learning_rate": 4.878402500474073e-07,
      "loss": 0.73987782,
      "num_input_tokens_seen": 279748350,
      "step": 12968,
      "time_per_iteration": 2.47818660736084
    },
    {
      "auxiliary_loss_clip": 0.01081279,
      "auxiliary_loss_mlp": 0.01035789,
      "balance_loss_clip": 1.03751874,
      "balance_loss_mlp": 1.02350473,
      "epoch": 0.7797384638508943,
      "flos": 15450207603840.0,
      "grad_norm": 1.8155442411700582,
      "language_loss": 0.607952,
      "learning_rate": 4.875853840905874e-07,
      "loss": 0.62912267,
      "num_input_tokens_seen": 279765620,
      "step": 12969,
      "time_per_iteration": 2.4996719360351562
    },
    {
      "auxiliary_loss_clip": 0.01085669,
      "auxiliary_loss_mlp": 0.0103345,
      "balance_loss_clip": 1.03537416,
      "balance_loss_mlp": 1.02180302,
      "epoch": 0.7797985871035623,
      "flos": 20922742160640.0,
      "grad_norm": 1.991097124185547,
      "language_loss": 0.70322776,
      "learning_rate": 4.873305754846811e-07,
      "loss": 0.724419,
      "num_input_tokens_seen": 279782485,
      "step": 12970,
      "time_per_iteration": 2.5082459449768066
    },
    {
      "auxiliary_loss_clip": 0.01075948,
      "auxiliary_loss_mlp": 0.00784744,
      "balance_loss_clip": 1.03868032,
      "balance_loss_mlp": 1.01065886,
      "epoch": 0.7798587103562302,
      "flos": 36937212514560.0,
      "grad_norm": 1.6565343189392445,
      "language_loss": 0.71988404,
      "learning_rate": 4.870758242393507e-07,
      "loss": 0.73849094,
      "num_input_tokens_seen": 279804170,
      "step": 12971,
      "time_per_iteration": 2.674285888671875
    },
    {
      "auxiliary_loss_clip": 0.01072076,
      "auxiliary_loss_mlp": 0.0103361,
      "balance_loss_clip": 1.03513098,
      "balance_loss_mlp": 1.02062833,
      "epoch": 0.7799188336088982,
      "flos": 22419283432320.0,
      "grad_norm": 1.6269681296274814,
      "language_loss": 0.74558151,
      "learning_rate": 4.868211303642578e-07,
      "loss": 0.7666384,
      "num_input_tokens_seen": 279823730,
      "step": 12972,
      "time_per_iteration": 2.5682570934295654
    },
    {
      "auxiliary_loss_clip": 0.01104148,
      "auxiliary_loss_mlp": 0.01025523,
      "balance_loss_clip": 1.03504503,
      "balance_loss_mlp": 1.01314282,
      "epoch": 0.7799789568615663,
      "flos": 18880466578560.0,
      "grad_norm": 1.771861243927943,
      "language_loss": 0.71290284,
      "learning_rate": 4.865664938690584e-07,
      "loss": 0.73419952,
      "num_input_tokens_seen": 279843035,
      "step": 12973,
      "time_per_iteration": 2.44003963470459
    },
    {
      "auxiliary_loss_clip": 0.01091771,
      "auxiliary_loss_mlp": 0.01032855,
      "balance_loss_clip": 1.03592277,
      "balance_loss_mlp": 1.02166748,
      "epoch": 0.7800390801142342,
      "flos": 20262272832000.0,
      "grad_norm": 2.0215921607350023,
      "language_loss": 0.7764262,
      "learning_rate": 4.863119147634089e-07,
      "loss": 0.79767251,
      "num_input_tokens_seen": 279861450,
      "step": 12974,
      "time_per_iteration": 2.4945425987243652
    },
    {
      "auxiliary_loss_clip": 0.01069761,
      "auxiliary_loss_mlp": 0.01031,
      "balance_loss_clip": 1.03347921,
      "balance_loss_mlp": 1.01840568,
      "epoch": 0.7800992033669022,
      "flos": 16690310703360.0,
      "grad_norm": 1.4783044117133326,
      "language_loss": 0.6919719,
      "learning_rate": 4.86057393056964e-07,
      "loss": 0.71297944,
      "num_input_tokens_seen": 279878660,
      "step": 12975,
      "time_per_iteration": 2.521122932434082
    },
    {
      "auxiliary_loss_clip": 0.01068899,
      "auxiliary_loss_mlp": 0.01029482,
      "balance_loss_clip": 1.03532207,
      "balance_loss_mlp": 1.01769781,
      "epoch": 0.7801593266195701,
      "flos": 18585208782720.0,
      "grad_norm": 1.8185517625165502,
      "language_loss": 0.81801057,
      "learning_rate": 4.858029287593739e-07,
      "loss": 0.83899438,
      "num_input_tokens_seen": 279895685,
      "step": 12976,
      "time_per_iteration": 2.561774730682373
    },
    {
      "auxiliary_loss_clip": 0.01084957,
      "auxiliary_loss_mlp": 0.00784574,
      "balance_loss_clip": 1.03467453,
      "balance_loss_mlp": 1.01215589,
      "epoch": 0.7802194498722381,
      "flos": 25484941405440.0,
      "grad_norm": 1.4444527302242933,
      "language_loss": 0.6576066,
      "learning_rate": 4.85548521880289e-07,
      "loss": 0.6763019,
      "num_input_tokens_seen": 279917240,
      "step": 12977,
      "time_per_iteration": 3.9706859588623047
    },
    {
      "auxiliary_loss_clip": 0.01083148,
      "auxiliary_loss_mlp": 0.01028803,
      "balance_loss_clip": 1.03658593,
      "balance_loss_mlp": 1.01731682,
      "epoch": 0.780279573124906,
      "flos": 31176315573120.0,
      "grad_norm": 2.100351325099898,
      "language_loss": 0.74873257,
      "learning_rate": 4.852941724293554e-07,
      "loss": 0.76985204,
      "num_input_tokens_seen": 279938665,
      "step": 12978,
      "time_per_iteration": 2.6065220832824707
    },
    {
      "auxiliary_loss_clip": 0.01080681,
      "auxiliary_loss_mlp": 0.01038509,
      "balance_loss_clip": 1.03404391,
      "balance_loss_mlp": 1.02378631,
      "epoch": 0.780339696377574,
      "flos": 26944027770240.0,
      "grad_norm": 1.8108874370288688,
      "language_loss": 0.6227107,
      "learning_rate": 4.85039880416219e-07,
      "loss": 0.64390266,
      "num_input_tokens_seen": 279957965,
      "step": 12979,
      "time_per_iteration": 2.5532548427581787
    },
    {
      "auxiliary_loss_clip": 0.01105608,
      "auxiliary_loss_mlp": 0.01031689,
      "balance_loss_clip": 1.03745246,
      "balance_loss_mlp": 1.01914787,
      "epoch": 0.780399819630242,
      "flos": 27957426180480.0,
      "grad_norm": 1.9458233394351334,
      "language_loss": 0.76713705,
      "learning_rate": 4.847856458505217e-07,
      "loss": 0.78851008,
      "num_input_tokens_seen": 279977490,
      "step": 12980,
      "time_per_iteration": 3.9115207195281982
    },
    {
      "auxiliary_loss_clip": 0.01107574,
      "auxiliary_loss_mlp": 0.01034625,
      "balance_loss_clip": 1.03743875,
      "balance_loss_mlp": 1.02305007,
      "epoch": 0.78045994288291,
      "flos": 22486795044480.0,
      "grad_norm": 2.024296450085091,
      "language_loss": 0.7743414,
      "learning_rate": 4.845314687419046e-07,
      "loss": 0.79576337,
      "num_input_tokens_seen": 279994220,
      "step": 12981,
      "time_per_iteration": 3.8836231231689453
    },
    {
      "auxiliary_loss_clip": 0.01071359,
      "auxiliary_loss_mlp": 0.01035038,
      "balance_loss_clip": 1.03703094,
      "balance_loss_mlp": 1.02284837,
      "epoch": 0.7805200661355779,
      "flos": 20850849089280.0,
      "grad_norm": 1.6773429407857483,
      "language_loss": 0.72400403,
      "learning_rate": 4.842773491000067e-07,
      "loss": 0.74506795,
      "num_input_tokens_seen": 280012590,
      "step": 12982,
      "time_per_iteration": 2.59193754196167
    },
    {
      "auxiliary_loss_clip": 0.01079035,
      "auxiliary_loss_mlp": 0.01030606,
      "balance_loss_clip": 1.0359745,
      "balance_loss_mlp": 1.01914978,
      "epoch": 0.7805801893882459,
      "flos": 25665966973440.0,
      "grad_norm": 1.328367214279769,
      "language_loss": 0.73397791,
      "learning_rate": 4.840232869344636e-07,
      "loss": 0.75507432,
      "num_input_tokens_seen": 280033700,
      "step": 12983,
      "time_per_iteration": 2.564222812652588
    },
    {
      "auxiliary_loss_clip": 0.0107844,
      "auxiliary_loss_mlp": 0.01028686,
      "balance_loss_clip": 1.03594995,
      "balance_loss_mlp": 1.01716483,
      "epoch": 0.7806403126409138,
      "flos": 11327806483200.0,
      "grad_norm": 1.791045093096207,
      "language_loss": 0.74619615,
      "learning_rate": 4.837692822549086e-07,
      "loss": 0.76726735,
      "num_input_tokens_seen": 280052215,
      "step": 12984,
      "time_per_iteration": 2.514319658279419
    },
    {
      "auxiliary_loss_clip": 0.01077598,
      "auxiliary_loss_mlp": 0.01037195,
      "balance_loss_clip": 1.03332996,
      "balance_loss_mlp": 1.02554822,
      "epoch": 0.7807004358935818,
      "flos": 19573362910080.0,
      "grad_norm": 1.8340869227038836,
      "language_loss": 0.81187469,
      "learning_rate": 4.835153350709746e-07,
      "loss": 0.83302265,
      "num_input_tokens_seen": 280070525,
      "step": 12985,
      "time_per_iteration": 2.507357358932495
    },
    {
      "auxiliary_loss_clip": 0.01083814,
      "auxiliary_loss_mlp": 0.01031308,
      "balance_loss_clip": 1.03588116,
      "balance_loss_mlp": 1.01913059,
      "epoch": 0.7807605591462499,
      "flos": 19135827342720.0,
      "grad_norm": 1.5624243066085874,
      "language_loss": 0.77335191,
      "learning_rate": 4.832614453922915e-07,
      "loss": 0.79450309,
      "num_input_tokens_seen": 280089855,
      "step": 12986,
      "time_per_iteration": 2.529770612716675
    },
    {
      "auxiliary_loss_clip": 0.01094501,
      "auxiliary_loss_mlp": 0.01033452,
      "balance_loss_clip": 1.03462696,
      "balance_loss_mlp": 1.02159691,
      "epoch": 0.7808206823989178,
      "flos": 32374654133760.0,
      "grad_norm": 1.701468706056767,
      "language_loss": 0.74140358,
      "learning_rate": 4.830076132284859e-07,
      "loss": 0.76268303,
      "num_input_tokens_seen": 280109960,
      "step": 12987,
      "time_per_iteration": 3.9619791507720947
    },
    {
      "auxiliary_loss_clip": 0.01021872,
      "auxiliary_loss_mlp": 0.0099874,
      "balance_loss_clip": 1.0091846,
      "balance_loss_mlp": 0.9976905,
      "epoch": 0.7808808056515858,
      "flos": 55050235061760.0,
      "grad_norm": 0.745797502437112,
      "language_loss": 0.55118859,
      "learning_rate": 4.82753838589184e-07,
      "loss": 0.57139468,
      "num_input_tokens_seen": 280169805,
      "step": 12988,
      "time_per_iteration": 3.128983736038208
    },
    {
      "auxiliary_loss_clip": 0.01079788,
      "auxiliary_loss_mlp": 0.01034476,
      "balance_loss_clip": 1.03476262,
      "balance_loss_mlp": 1.02284157,
      "epoch": 0.7809409289042537,
      "flos": 12859468277760.0,
      "grad_norm": 3.9449606509766446,
      "language_loss": 0.80530286,
      "learning_rate": 4.82500121484009e-07,
      "loss": 0.82644558,
      "num_input_tokens_seen": 280184630,
      "step": 12989,
      "time_per_iteration": 2.504418134689331
    },
    {
      "auxiliary_loss_clip": 0.0107102,
      "auxiliary_loss_mlp": 0.0102809,
      "balance_loss_clip": 1.03460813,
      "balance_loss_mlp": 1.01622236,
      "epoch": 0.7810010521569217,
      "flos": 21687244254720.0,
      "grad_norm": 1.5126927566730415,
      "language_loss": 0.70363891,
      "learning_rate": 4.822464619225806e-07,
      "loss": 0.72463,
      "num_input_tokens_seen": 280203880,
      "step": 12990,
      "time_per_iteration": 2.5748088359832764
    },
    {
      "auxiliary_loss_clip": 0.01080084,
      "auxiliary_loss_mlp": 0.01030561,
      "balance_loss_clip": 1.03425789,
      "balance_loss_mlp": 1.01707268,
      "epoch": 0.7810611754095896,
      "flos": 16757068129920.0,
      "grad_norm": 1.9854706215796256,
      "language_loss": 0.7772373,
      "learning_rate": 4.819928599145184e-07,
      "loss": 0.79834378,
      "num_input_tokens_seen": 280220460,
      "step": 12991,
      "time_per_iteration": 2.5005359649658203
    },
    {
      "auxiliary_loss_clip": 0.01066697,
      "auxiliary_loss_mlp": 0.0103227,
      "balance_loss_clip": 1.0336895,
      "balance_loss_mlp": 1.01991987,
      "epoch": 0.7811212986622577,
      "flos": 43507464658560.0,
      "grad_norm": 1.4971879407099793,
      "language_loss": 0.66160643,
      "learning_rate": 4.817393154694398e-07,
      "loss": 0.68259609,
      "num_input_tokens_seen": 280242680,
      "step": 12992,
      "time_per_iteration": 2.7561628818511963
    },
    {
      "auxiliary_loss_clip": 0.01106391,
      "auxiliary_loss_mlp": 0.01029384,
      "balance_loss_clip": 1.03698492,
      "balance_loss_mlp": 1.0173018,
      "epoch": 0.7811814219149256,
      "flos": 21757700782080.0,
      "grad_norm": 1.788657001942646,
      "language_loss": 0.61757773,
      "learning_rate": 4.814858285969578e-07,
      "loss": 0.63893545,
      "num_input_tokens_seen": 280260655,
      "step": 12993,
      "time_per_iteration": 2.4816274642944336
    },
    {
      "auxiliary_loss_clip": 0.0108075,
      "auxiliary_loss_mlp": 0.01028463,
      "balance_loss_clip": 1.03463018,
      "balance_loss_mlp": 1.0161128,
      "epoch": 0.7812415451675936,
      "flos": 24061514267520.0,
      "grad_norm": 1.67214780946007,
      "language_loss": 0.6848135,
      "learning_rate": 4.812323993066862e-07,
      "loss": 0.70590562,
      "num_input_tokens_seen": 280281185,
      "step": 12994,
      "time_per_iteration": 2.556504487991333
    },
    {
      "auxiliary_loss_clip": 0.01102348,
      "auxiliary_loss_mlp": 0.01026524,
      "balance_loss_clip": 1.03480291,
      "balance_loss_mlp": 1.01518703,
      "epoch": 0.7813016684202615,
      "flos": 18989706816000.0,
      "grad_norm": 2.019807541894518,
      "language_loss": 0.68796498,
      "learning_rate": 4.809790276082335e-07,
      "loss": 0.70925367,
      "num_input_tokens_seen": 280298255,
      "step": 12995,
      "time_per_iteration": 2.4788713455200195
    },
    {
      "auxiliary_loss_clip": 0.01069015,
      "auxiliary_loss_mlp": 0.01028226,
      "balance_loss_clip": 1.03596354,
      "balance_loss_mlp": 1.01721692,
      "epoch": 0.7813617916729295,
      "flos": 25260786581760.0,
      "grad_norm": 1.736260922343832,
      "language_loss": 0.74691796,
      "learning_rate": 4.807257135112088e-07,
      "loss": 0.76789039,
      "num_input_tokens_seen": 280319000,
      "step": 12996,
      "time_per_iteration": 2.5799834728240967
    },
    {
      "auxiliary_loss_clip": 0.01110034,
      "auxiliary_loss_mlp": 0.01033412,
      "balance_loss_clip": 1.03784966,
      "balance_loss_mlp": 1.02097225,
      "epoch": 0.7814219149255974,
      "flos": 17966037116160.0,
      "grad_norm": 5.2576105419936106,
      "language_loss": 0.68117839,
      "learning_rate": 4.804724570252167e-07,
      "loss": 0.70261282,
      "num_input_tokens_seen": 280336375,
      "step": 12997,
      "time_per_iteration": 2.4372987747192383
    },
    {
      "auxiliary_loss_clip": 0.01108863,
      "auxiliary_loss_mlp": 0.01033344,
      "balance_loss_clip": 1.03705883,
      "balance_loss_mlp": 1.0205822,
      "epoch": 0.7814820381782654,
      "flos": 25776176878080.0,
      "grad_norm": 1.6294597606476013,
      "language_loss": 0.81947386,
      "learning_rate": 4.802192581598614e-07,
      "loss": 0.84089589,
      "num_input_tokens_seen": 280358760,
      "step": 12998,
      "time_per_iteration": 2.4848928451538086
    },
    {
      "auxiliary_loss_clip": 0.01077148,
      "auxiliary_loss_mlp": 0.01036053,
      "balance_loss_clip": 1.03133488,
      "balance_loss_mlp": 1.02274287,
      "epoch": 0.7815421614309335,
      "flos": 20519572930560.0,
      "grad_norm": 2.8903696000746297,
      "language_loss": 0.74348688,
      "learning_rate": 4.799661169247453e-07,
      "loss": 0.76461893,
      "num_input_tokens_seen": 280377085,
      "step": 12999,
      "time_per_iteration": 2.520977020263672
    },
    {
      "auxiliary_loss_clip": 0.01092689,
      "auxiliary_loss_mlp": 0.01035565,
      "balance_loss_clip": 1.03556609,
      "balance_loss_mlp": 1.02176094,
      "epoch": 0.7816022846836014,
      "flos": 21287666384640.0,
      "grad_norm": 1.4623253388498227,
      "language_loss": 0.84600073,
      "learning_rate": 4.797130333294652e-07,
      "loss": 0.86728334,
      "num_input_tokens_seen": 280395465,
      "step": 13000,
      "time_per_iteration": 2.4857125282287598
    },
    {
      "auxiliary_loss_clip": 0.01096175,
      "auxiliary_loss_mlp": 0.01031254,
      "balance_loss_clip": 1.03685081,
      "balance_loss_mlp": 1.01916647,
      "epoch": 0.7816624079362694,
      "flos": 19208402772480.0,
      "grad_norm": 1.7203575673763005,
      "language_loss": 0.66170025,
      "learning_rate": 4.794600073836192e-07,
      "loss": 0.68297446,
      "num_input_tokens_seen": 280412775,
      "step": 13001,
      "time_per_iteration": 2.50039005279541
    },
    {
      "auxiliary_loss_clip": 0.01067753,
      "auxiliary_loss_mlp": 0.01034241,
      "balance_loss_clip": 1.03512573,
      "balance_loss_mlp": 1.02234387,
      "epoch": 0.7817225311889373,
      "flos": 26104687689600.0,
      "grad_norm": 1.4573836362276584,
      "language_loss": 0.66849542,
      "learning_rate": 4.792070390968027e-07,
      "loss": 0.68951535,
      "num_input_tokens_seen": 280432905,
      "step": 13002,
      "time_per_iteration": 2.5946924686431885
    },
    {
      "auxiliary_loss_clip": 0.01098044,
      "auxiliary_loss_mlp": 0.01033211,
      "balance_loss_clip": 1.03766751,
      "balance_loss_mlp": 1.02054453,
      "epoch": 0.7817826544416053,
      "flos": 21250929749760.0,
      "grad_norm": 2.0913026452602153,
      "language_loss": 0.73451805,
      "learning_rate": 4.78954128478607e-07,
      "loss": 0.75583059,
      "num_input_tokens_seen": 280450785,
      "step": 13003,
      "time_per_iteration": 2.502394199371338
    },
    {
      "auxiliary_loss_clip": 0.01095045,
      "auxiliary_loss_mlp": 0.01033343,
      "balance_loss_clip": 1.03659058,
      "balance_loss_mlp": 1.02110648,
      "epoch": 0.7818427776942732,
      "flos": 19932181822080.0,
      "grad_norm": 1.7069110243573784,
      "language_loss": 0.62243938,
      "learning_rate": 4.787012755386233e-07,
      "loss": 0.64372325,
      "num_input_tokens_seen": 280468400,
      "step": 13004,
      "time_per_iteration": 2.474656581878662
    },
    {
      "auxiliary_loss_clip": 0.01099248,
      "auxiliary_loss_mlp": 0.01030366,
      "balance_loss_clip": 1.03434944,
      "balance_loss_mlp": 1.01954174,
      "epoch": 0.7819029009469413,
      "flos": 11363753018880.0,
      "grad_norm": 1.8318114520549877,
      "language_loss": 0.82880878,
      "learning_rate": 4.784484802864403e-07,
      "loss": 0.85010493,
      "num_input_tokens_seen": 280483930,
      "step": 13005,
      "time_per_iteration": 2.4471733570098877
    },
    {
      "auxiliary_loss_clip": 0.01064237,
      "auxiliary_loss_mlp": 0.00784212,
      "balance_loss_clip": 1.03194928,
      "balance_loss_mlp": 1.00868332,
      "epoch": 0.7819630241996092,
      "flos": 24279276470400.0,
      "grad_norm": 1.714272265774244,
      "language_loss": 0.72664309,
      "learning_rate": 4.781957427316432e-07,
      "loss": 0.74512762,
      "num_input_tokens_seen": 280503465,
      "step": 13006,
      "time_per_iteration": 2.5689024925231934
    },
    {
      "auxiliary_loss_clip": 0.01097436,
      "auxiliary_loss_mlp": 0.00785463,
      "balance_loss_clip": 1.03759658,
      "balance_loss_mlp": 1.01270545,
      "epoch": 0.7820231474522772,
      "flos": 22708902792960.0,
      "grad_norm": 1.5638143026775078,
      "language_loss": 0.72039449,
      "learning_rate": 4.779430628838157e-07,
      "loss": 0.73922348,
      "num_input_tokens_seen": 280523375,
      "step": 13007,
      "time_per_iteration": 2.516085386276245
    },
    {
      "auxiliary_loss_clip": 0.0110565,
      "auxiliary_loss_mlp": 0.01030361,
      "balance_loss_clip": 1.03481567,
      "balance_loss_mlp": 1.01782572,
      "epoch": 0.7820832707049451,
      "flos": 20047419630720.0,
      "grad_norm": 1.9622807444910655,
      "language_loss": 0.689448,
      "learning_rate": 4.776904407525397e-07,
      "loss": 0.71080804,
      "num_input_tokens_seen": 280542920,
      "step": 13008,
      "time_per_iteration": 2.4463868141174316
    },
    {
      "auxiliary_loss_clip": 0.01077654,
      "auxiliary_loss_mlp": 0.01026732,
      "balance_loss_clip": 1.03578496,
      "balance_loss_mlp": 1.01400638,
      "epoch": 0.7821433939576131,
      "flos": 27162795553920.0,
      "grad_norm": 1.89510750336667,
      "language_loss": 0.69680274,
      "learning_rate": 4.774378763473954e-07,
      "loss": 0.71784663,
      "num_input_tokens_seen": 280561700,
      "step": 13009,
      "time_per_iteration": 2.610992193222046
    },
    {
      "auxiliary_loss_clip": 0.01064464,
      "auxiliary_loss_mlp": 0.01027414,
      "balance_loss_clip": 1.03207827,
      "balance_loss_mlp": 1.01505172,
      "epoch": 0.782203517210281,
      "flos": 22602068766720.0,
      "grad_norm": 1.6276754164285818,
      "language_loss": 0.81676781,
      "learning_rate": 4.771853696779586e-07,
      "loss": 0.8376866,
      "num_input_tokens_seen": 280580605,
      "step": 13010,
      "time_per_iteration": 2.551722526550293
    },
    {
      "auxiliary_loss_clip": 0.01091634,
      "auxiliary_loss_mlp": 0.01032538,
      "balance_loss_clip": 1.03434801,
      "balance_loss_mlp": 1.02124262,
      "epoch": 0.782263640462949,
      "flos": 29059812535680.0,
      "grad_norm": 1.5557890986180725,
      "language_loss": 0.62443864,
      "learning_rate": 4.76932920753806e-07,
      "loss": 0.64568043,
      "num_input_tokens_seen": 280601495,
      "step": 13011,
      "time_per_iteration": 2.5515785217285156
    },
    {
      "auxiliary_loss_clip": 0.01094121,
      "auxiliary_loss_mlp": 0.01028701,
      "balance_loss_clip": 1.03753996,
      "balance_loss_mlp": 1.01829982,
      "epoch": 0.782323763715617,
      "flos": 25299498464640.0,
      "grad_norm": 1.8733405806923116,
      "language_loss": 0.69891208,
      "learning_rate": 4.7668052958450913e-07,
      "loss": 0.72014034,
      "num_input_tokens_seen": 280622760,
      "step": 13012,
      "time_per_iteration": 2.535952091217041
    },
    {
      "auxiliary_loss_clip": 0.01031667,
      "auxiliary_loss_mlp": 0.0100608,
      "balance_loss_clip": 1.00864577,
      "balance_loss_mlp": 1.00479221,
      "epoch": 0.782383886968285,
      "flos": 65194388668800.0,
      "grad_norm": 0.7039131077630228,
      "language_loss": 0.55023277,
      "learning_rate": 4.764281961796395e-07,
      "loss": 0.57061028,
      "num_input_tokens_seen": 280687115,
      "step": 13013,
      "time_per_iteration": 3.1727874279022217
    },
    {
      "auxiliary_loss_clip": 0.01081626,
      "auxiliary_loss_mlp": 0.01034678,
      "balance_loss_clip": 1.03647566,
      "balance_loss_mlp": 1.02260244,
      "epoch": 0.782444010220953,
      "flos": 18405440190720.0,
      "grad_norm": 1.9409096026699597,
      "language_loss": 0.65432155,
      "learning_rate": 4.76175920548765e-07,
      "loss": 0.67548466,
      "num_input_tokens_seen": 280705000,
      "step": 13014,
      "time_per_iteration": 2.500490665435791
    },
    {
      "auxiliary_loss_clip": 0.01008406,
      "auxiliary_loss_mlp": 0.01000189,
      "balance_loss_clip": 1.01106358,
      "balance_loss_mlp": 0.99872255,
      "epoch": 0.7825041334736209,
      "flos": 63955003841280.0,
      "grad_norm": 0.7263432952175565,
      "language_loss": 0.5843733,
      "learning_rate": 4.759237027014524e-07,
      "loss": 0.60445923,
      "num_input_tokens_seen": 280773525,
      "step": 13015,
      "time_per_iteration": 3.2573177814483643
    },
    {
      "auxiliary_loss_clip": 0.01084423,
      "auxiliary_loss_mlp": 0.01031502,
      "balance_loss_clip": 1.03682351,
      "balance_loss_mlp": 1.02027893,
      "epoch": 0.7825642567262889,
      "flos": 20339373375360.0,
      "grad_norm": 1.6389407790932553,
      "language_loss": 0.74602288,
      "learning_rate": 4.756715426472666e-07,
      "loss": 0.76718211,
      "num_input_tokens_seen": 280791915,
      "step": 13016,
      "time_per_iteration": 3.962952136993408
    },
    {
      "auxiliary_loss_clip": 0.01105486,
      "auxiliary_loss_mlp": 0.0102872,
      "balance_loss_clip": 1.03615713,
      "balance_loss_mlp": 1.01526117,
      "epoch": 0.7826243799789568,
      "flos": 20262955190400.0,
      "grad_norm": 1.6301409706473726,
      "language_loss": 0.75054812,
      "learning_rate": 4.7541944039576766e-07,
      "loss": 0.77189016,
      "num_input_tokens_seen": 280811460,
      "step": 13017,
      "time_per_iteration": 2.4528629779815674
    },
    {
      "auxiliary_loss_clip": 0.01079864,
      "auxiliary_loss_mlp": 0.01031642,
      "balance_loss_clip": 1.03523207,
      "balance_loss_mlp": 1.01925611,
      "epoch": 0.7826845032316249,
      "flos": 21132926593920.0,
      "grad_norm": 2.0321698843195586,
      "language_loss": 0.75407892,
      "learning_rate": 4.7516739595651636e-07,
      "loss": 0.77519399,
      "num_input_tokens_seen": 280825415,
      "step": 13018,
      "time_per_iteration": 3.8766555786132812
    },
    {
      "auxiliary_loss_clip": 0.01103305,
      "auxiliary_loss_mlp": 0.01028295,
      "balance_loss_clip": 1.03526759,
      "balance_loss_mlp": 1.01592731,
      "epoch": 0.7827446264842928,
      "flos": 22492253911680.0,
      "grad_norm": 1.4610165715352714,
      "language_loss": 0.77224791,
      "learning_rate": 4.749154093390708e-07,
      "loss": 0.7935639,
      "num_input_tokens_seen": 280845335,
      "step": 13019,
      "time_per_iteration": 2.4723141193389893
    },
    {
      "auxiliary_loss_clip": 0.01057315,
      "auxiliary_loss_mlp": 0.01021197,
      "balance_loss_clip": 1.03474164,
      "balance_loss_mlp": 1.0103786,
      "epoch": 0.7828047497369608,
      "flos": 28840649702400.0,
      "grad_norm": 1.4058647528959523,
      "language_loss": 0.6745857,
      "learning_rate": 4.746634805529852e-07,
      "loss": 0.69537085,
      "num_input_tokens_seen": 280867145,
      "step": 13020,
      "time_per_iteration": 4.065423488616943
    },
    {
      "auxiliary_loss_clip": 0.01092241,
      "auxiliary_loss_mlp": 0.0102802,
      "balance_loss_clip": 1.03859496,
      "balance_loss_mlp": 1.01623619,
      "epoch": 0.7828648729896287,
      "flos": 23257689759360.0,
      "grad_norm": 1.7361485863291195,
      "language_loss": 0.62358892,
      "learning_rate": 4.7441160960781325e-07,
      "loss": 0.64479148,
      "num_input_tokens_seen": 280886185,
      "step": 13021,
      "time_per_iteration": 2.5113656520843506
    },
    {
      "auxiliary_loss_clip": 0.01101525,
      "auxiliary_loss_mlp": 0.01031373,
      "balance_loss_clip": 1.03526938,
      "balance_loss_mlp": 1.02017307,
      "epoch": 0.7829249962422967,
      "flos": 25265670831360.0,
      "grad_norm": 1.6762412424569924,
      "language_loss": 0.69607687,
      "learning_rate": 4.7415979651310636e-07,
      "loss": 0.71740586,
      "num_input_tokens_seen": 280907665,
      "step": 13022,
      "time_per_iteration": 2.5003914833068848
    },
    {
      "auxiliary_loss_clip": 0.00992377,
      "auxiliary_loss_mlp": 0.00999564,
      "balance_loss_clip": 1.01559985,
      "balance_loss_mlp": 0.9984318,
      "epoch": 0.7829851194949646,
      "flos": 70722044645760.0,
      "grad_norm": 0.6432050880553647,
      "language_loss": 0.56200415,
      "learning_rate": 4.739080412784131e-07,
      "loss": 0.58192354,
      "num_input_tokens_seen": 280971405,
      "step": 13023,
      "time_per_iteration": 3.3136653900146484
    },
    {
      "auxiliary_loss_clip": 0.0107469,
      "auxiliary_loss_mlp": 0.01032981,
      "balance_loss_clip": 1.03266072,
      "balance_loss_mlp": 1.02040994,
      "epoch": 0.7830452427476327,
      "flos": 25660795415040.0,
      "grad_norm": 1.603806008777355,
      "language_loss": 0.66886234,
      "learning_rate": 4.736563439132792e-07,
      "loss": 0.68993902,
      "num_input_tokens_seen": 280989615,
      "step": 13024,
      "time_per_iteration": 2.529069662094116
    },
    {
      "auxiliary_loss_clip": 0.01106649,
      "auxiliary_loss_mlp": 0.01027539,
      "balance_loss_clip": 1.03682446,
      "balance_loss_mlp": 1.01479554,
      "epoch": 0.7831053660003006,
      "flos": 22784315397120.0,
      "grad_norm": 2.8350822299388843,
      "language_loss": 0.77817899,
      "learning_rate": 4.734047044272498e-07,
      "loss": 0.79952085,
      "num_input_tokens_seen": 281009450,
      "step": 13025,
      "time_per_iteration": 2.508657217025757
    },
    {
      "auxiliary_loss_clip": 0.01079287,
      "auxiliary_loss_mlp": 0.01032264,
      "balance_loss_clip": 1.03499103,
      "balance_loss_mlp": 1.02055812,
      "epoch": 0.7831654892529686,
      "flos": 25812267068160.0,
      "grad_norm": 1.6226034178925606,
      "language_loss": 0.78553587,
      "learning_rate": 4.731531228298673e-07,
      "loss": 0.80665135,
      "num_input_tokens_seen": 281028120,
      "step": 13026,
      "time_per_iteration": 4.046311855316162
    },
    {
      "auxiliary_loss_clip": 0.01092607,
      "auxiliary_loss_mlp": 0.01024486,
      "balance_loss_clip": 1.03731847,
      "balance_loss_mlp": 1.01300025,
      "epoch": 0.7832256125056366,
      "flos": 20771557816320.0,
      "grad_norm": 2.261926749247982,
      "language_loss": 0.75479025,
      "learning_rate": 4.729015991306715e-07,
      "loss": 0.77596116,
      "num_input_tokens_seen": 281042130,
      "step": 13027,
      "time_per_iteration": 2.4544484615325928
    },
    {
      "auxiliary_loss_clip": 0.01094957,
      "auxiliary_loss_mlp": 0.01027957,
      "balance_loss_clip": 1.03699625,
      "balance_loss_mlp": 1.01656055,
      "epoch": 0.7832857357583045,
      "flos": 21506541909120.0,
      "grad_norm": 1.7050326646645175,
      "language_loss": 0.70366257,
      "learning_rate": 4.726501333391997e-07,
      "loss": 0.72489178,
      "num_input_tokens_seen": 281060945,
      "step": 13028,
      "time_per_iteration": 2.500354051589966
    },
    {
      "auxiliary_loss_clip": 0.01058275,
      "auxiliary_loss_mlp": 0.01040822,
      "balance_loss_clip": 1.03644896,
      "balance_loss_mlp": 1.02795959,
      "epoch": 0.7833458590109725,
      "flos": 18077791305600.0,
      "grad_norm": 2.006215113839031,
      "language_loss": 0.68972403,
      "learning_rate": 4.7239872546498774e-07,
      "loss": 0.710715,
      "num_input_tokens_seen": 281079270,
      "step": 13029,
      "time_per_iteration": 2.554739475250244
    },
    {
      "auxiliary_loss_clip": 0.01074946,
      "auxiliary_loss_mlp": 0.01027366,
      "balance_loss_clip": 1.03339505,
      "balance_loss_mlp": 1.01523054,
      "epoch": 0.7834059822636404,
      "flos": 28288738252800.0,
      "grad_norm": 1.6863230610323796,
      "language_loss": 0.80827534,
      "learning_rate": 4.721473755175698e-07,
      "loss": 0.8292985,
      "num_input_tokens_seen": 281099500,
      "step": 13030,
      "time_per_iteration": 2.618567705154419
    },
    {
      "auxiliary_loss_clip": 0.01096877,
      "auxiliary_loss_mlp": 0.01032566,
      "balance_loss_clip": 1.03500688,
      "balance_loss_mlp": 1.0205493,
      "epoch": 0.7834661055163085,
      "flos": 31686211088640.0,
      "grad_norm": 1.6126103299959011,
      "language_loss": 0.70747888,
      "learning_rate": 4.71896083506476e-07,
      "loss": 0.72877336,
      "num_input_tokens_seen": 281121250,
      "step": 13031,
      "time_per_iteration": 2.5619826316833496
    },
    {
      "auxiliary_loss_clip": 0.01066871,
      "auxiliary_loss_mlp": 0.01029224,
      "balance_loss_clip": 1.03436875,
      "balance_loss_mlp": 1.01705885,
      "epoch": 0.7835262287689764,
      "flos": 12933192942720.0,
      "grad_norm": 1.8779739007097658,
      "language_loss": 0.78391206,
      "learning_rate": 4.7164484944123574e-07,
      "loss": 0.80487299,
      "num_input_tokens_seen": 281138760,
      "step": 13032,
      "time_per_iteration": 2.5481278896331787
    },
    {
      "auxiliary_loss_clip": 0.01098694,
      "auxiliary_loss_mlp": 0.01037867,
      "balance_loss_clip": 1.03698277,
      "balance_loss_mlp": 1.02532053,
      "epoch": 0.7835863520216444,
      "flos": 16143211676160.0,
      "grad_norm": 2.287083433098722,
      "language_loss": 0.63298941,
      "learning_rate": 4.7139367333137726e-07,
      "loss": 0.65435505,
      "num_input_tokens_seen": 281157420,
      "step": 13033,
      "time_per_iteration": 2.456186294555664
    },
    {
      "auxiliary_loss_clip": 0.0109225,
      "auxiliary_loss_mlp": 0.01033308,
      "balance_loss_clip": 1.03514242,
      "balance_loss_mlp": 1.02130342,
      "epoch": 0.7836464752743123,
      "flos": 11509909459200.0,
      "grad_norm": 1.6689414425545086,
      "language_loss": 0.71643627,
      "learning_rate": 4.7114255518642255e-07,
      "loss": 0.73769188,
      "num_input_tokens_seen": 281174620,
      "step": 13034,
      "time_per_iteration": 2.5267012119293213
    },
    {
      "auxiliary_loss_clip": 0.01105471,
      "auxiliary_loss_mlp": 0.00785165,
      "balance_loss_clip": 1.03590417,
      "balance_loss_mlp": 1.01183224,
      "epoch": 0.7837065985269803,
      "flos": 18223696350720.0,
      "grad_norm": 1.5907583299067307,
      "language_loss": 0.71971524,
      "learning_rate": 4.7089149501589555e-07,
      "loss": 0.73862159,
      "num_input_tokens_seen": 281193865,
      "step": 13035,
      "time_per_iteration": 2.4351465702056885
    },
    {
      "auxiliary_loss_clip": 0.01106774,
      "auxiliary_loss_mlp": 0.01036452,
      "balance_loss_clip": 1.03759432,
      "balance_loss_mlp": 1.02400041,
      "epoch": 0.7837667217796482,
      "flos": 24754410599040.0,
      "grad_norm": 1.9274046117459727,
      "language_loss": 0.66145921,
      "learning_rate": 4.7064049282931664e-07,
      "loss": 0.68289149,
      "num_input_tokens_seen": 281212250,
      "step": 13036,
      "time_per_iteration": 2.478936195373535
    },
    {
      "auxiliary_loss_clip": 0.0109815,
      "auxiliary_loss_mlp": 0.01037405,
      "balance_loss_clip": 1.03683949,
      "balance_loss_mlp": 1.02480483,
      "epoch": 0.7838268450323163,
      "flos": 22383121415040.0,
      "grad_norm": 2.8846305398118792,
      "language_loss": 0.72620767,
      "learning_rate": 4.703895486362031e-07,
      "loss": 0.74756324,
      "num_input_tokens_seen": 281230850,
      "step": 13037,
      "time_per_iteration": 2.487478017807007
    },
    {
      "auxiliary_loss_clip": 0.01064565,
      "auxiliary_loss_mlp": 0.01036234,
      "balance_loss_clip": 1.03090465,
      "balance_loss_mlp": 1.02334118,
      "epoch": 0.7838869682849842,
      "flos": 19500284689920.0,
      "grad_norm": 2.8311827451458607,
      "language_loss": 0.59689486,
      "learning_rate": 4.701386624460717e-07,
      "loss": 0.61790287,
      "num_input_tokens_seen": 281249810,
      "step": 13038,
      "time_per_iteration": 2.5455222129821777
    },
    {
      "auxiliary_loss_clip": 0.01082551,
      "auxiliary_loss_mlp": 0.01027816,
      "balance_loss_clip": 1.03564358,
      "balance_loss_mlp": 1.01677144,
      "epoch": 0.7839470915376522,
      "flos": 32892845690880.0,
      "grad_norm": 1.5607932424089606,
      "language_loss": 0.67832506,
      "learning_rate": 4.698878342684349e-07,
      "loss": 0.6994288,
      "num_input_tokens_seen": 281273730,
      "step": 13039,
      "time_per_iteration": 2.625549793243408
    },
    {
      "auxiliary_loss_clip": 0.010662,
      "auxiliary_loss_mlp": 0.01023778,
      "balance_loss_clip": 1.03259349,
      "balance_loss_mlp": 1.01331174,
      "epoch": 0.7840072147903202,
      "flos": 29676003373440.0,
      "grad_norm": 2.679270654770113,
      "language_loss": 0.69068098,
      "learning_rate": 4.6963706411280537e-07,
      "loss": 0.71158075,
      "num_input_tokens_seen": 281293670,
      "step": 13040,
      "time_per_iteration": 2.622471809387207
    },
    {
      "auxiliary_loss_clip": 0.01061949,
      "auxiliary_loss_mlp": 0.01031805,
      "balance_loss_clip": 1.03586173,
      "balance_loss_mlp": 1.01931179,
      "epoch": 0.7840673380429881,
      "flos": 18186744234240.0,
      "grad_norm": 1.6540891068651573,
      "language_loss": 0.67120779,
      "learning_rate": 4.6938635198869116e-07,
      "loss": 0.69214535,
      "num_input_tokens_seen": 281313070,
      "step": 13041,
      "time_per_iteration": 2.569120168685913
    },
    {
      "auxiliary_loss_clip": 0.01022386,
      "auxiliary_loss_mlp": 0.00762788,
      "balance_loss_clip": 1.00931418,
      "balance_loss_mlp": 1.00306213,
      "epoch": 0.7841274612956561,
      "flos": 66346006613760.0,
      "grad_norm": 0.6605618267714153,
      "language_loss": 0.57473719,
      "learning_rate": 4.691356979055998e-07,
      "loss": 0.59258902,
      "num_input_tokens_seen": 281374880,
      "step": 13042,
      "time_per_iteration": 3.074313163757324
    },
    {
      "auxiliary_loss_clip": 0.01081227,
      "auxiliary_loss_mlp": 0.01028077,
      "balance_loss_clip": 1.03637004,
      "balance_loss_mlp": 1.0155834,
      "epoch": 0.784187584548324,
      "flos": 26648482665600.0,
      "grad_norm": 1.867128931269078,
      "language_loss": 0.83674157,
      "learning_rate": 4.688851018730369e-07,
      "loss": 0.85783458,
      "num_input_tokens_seen": 281392620,
      "step": 13043,
      "time_per_iteration": 2.5671727657318115
    },
    {
      "auxiliary_loss_clip": 0.01089805,
      "auxiliary_loss_mlp": 0.0102975,
      "balance_loss_clip": 1.03536165,
      "balance_loss_mlp": 1.0184077,
      "epoch": 0.7842477078009921,
      "flos": 25740158515200.0,
      "grad_norm": 1.3778999240365586,
      "language_loss": 0.8832081,
      "learning_rate": 4.6863456390050425e-07,
      "loss": 0.90440363,
      "num_input_tokens_seen": 281413140,
      "step": 13044,
      "time_per_iteration": 2.5377919673919678
    },
    {
      "auxiliary_loss_clip": 0.01086918,
      "auxiliary_loss_mlp": 0.01029808,
      "balance_loss_clip": 1.03550768,
      "balance_loss_mlp": 1.01792836,
      "epoch": 0.78430783105366,
      "flos": 21980957765760.0,
      "grad_norm": 1.6172235906076722,
      "language_loss": 0.79086924,
      "learning_rate": 4.6838408399750195e-07,
      "loss": 0.81203645,
      "num_input_tokens_seen": 281430860,
      "step": 13045,
      "time_per_iteration": 2.538931369781494
    },
    {
      "auxiliary_loss_clip": 0.01076271,
      "auxiliary_loss_mlp": 0.01027621,
      "balance_loss_clip": 1.03443098,
      "balance_loss_mlp": 1.01631403,
      "epoch": 0.784367954306328,
      "flos": 23842279607040.0,
      "grad_norm": 1.4929639509408479,
      "language_loss": 0.72375071,
      "learning_rate": 4.6813366217352925e-07,
      "loss": 0.74478966,
      "num_input_tokens_seen": 281451385,
      "step": 13046,
      "time_per_iteration": 2.55049467086792
    },
    {
      "auxiliary_loss_clip": 0.01060347,
      "auxiliary_loss_mlp": 0.0103469,
      "balance_loss_clip": 1.03447485,
      "balance_loss_mlp": 1.02225029,
      "epoch": 0.7844280775589959,
      "flos": 24826662806400.0,
      "grad_norm": 1.5165785232665245,
      "language_loss": 0.63055003,
      "learning_rate": 4.678832984380809e-07,
      "loss": 0.6515004,
      "num_input_tokens_seen": 281472255,
      "step": 13047,
      "time_per_iteration": 2.6557538509368896
    },
    {
      "auxiliary_loss_clip": 0.0109185,
      "auxiliary_loss_mlp": 0.01028559,
      "balance_loss_clip": 1.03458762,
      "balance_loss_mlp": 1.01693642,
      "epoch": 0.7844882008116639,
      "flos": 22455660931200.0,
      "grad_norm": 1.4668914840696121,
      "language_loss": 0.73015583,
      "learning_rate": 4.676329928006515e-07,
      "loss": 0.75135994,
      "num_input_tokens_seen": 281492860,
      "step": 13048,
      "time_per_iteration": 2.4971556663513184
    },
    {
      "auxiliary_loss_clip": 0.01081893,
      "auxiliary_loss_mlp": 0.01028423,
      "balance_loss_clip": 1.03743148,
      "balance_loss_mlp": 1.01675892,
      "epoch": 0.7845483240643318,
      "flos": 26104041244800.0,
      "grad_norm": 1.9293034234921174,
      "language_loss": 0.74602568,
      "learning_rate": 4.6738274527073243e-07,
      "loss": 0.76712888,
      "num_input_tokens_seen": 281511815,
      "step": 13049,
      "time_per_iteration": 2.57753849029541
    },
    {
      "auxiliary_loss_clip": 0.01107879,
      "auxiliary_loss_mlp": 0.01030696,
      "balance_loss_clip": 1.0360074,
      "balance_loss_mlp": 1.01778531,
      "epoch": 0.7846084473169999,
      "flos": 19354307817600.0,
      "grad_norm": 1.8862345783298364,
      "language_loss": 0.72608942,
      "learning_rate": 4.6713255585781454e-07,
      "loss": 0.74747515,
      "num_input_tokens_seen": 281530090,
      "step": 13050,
      "time_per_iteration": 2.4583442211151123
    },
    {
      "auxiliary_loss_clip": 0.01091955,
      "auxiliary_loss_mlp": 0.01031727,
      "balance_loss_clip": 1.03504801,
      "balance_loss_mlp": 1.01947188,
      "epoch": 0.7846685705696678,
      "flos": 23325811902720.0,
      "grad_norm": 1.9531420701643483,
      "language_loss": 0.73514378,
      "learning_rate": 4.668824245713825e-07,
      "loss": 0.75638056,
      "num_input_tokens_seen": 281547075,
      "step": 13051,
      "time_per_iteration": 2.5108530521392822
    },
    {
      "auxiliary_loss_clip": 0.01107652,
      "auxiliary_loss_mlp": 0.0103368,
      "balance_loss_clip": 1.03739297,
      "balance_loss_mlp": 1.0210079,
      "epoch": 0.7847286938223358,
      "flos": 35809545962880.0,
      "grad_norm": 1.8890484104551903,
      "language_loss": 0.72775108,
      "learning_rate": 4.666323514209227e-07,
      "loss": 0.74916434,
      "num_input_tokens_seen": 281568080,
      "step": 13052,
      "time_per_iteration": 2.5786654949188232
    },
    {
      "auxiliary_loss_clip": 0.010784,
      "auxiliary_loss_mlp": 0.0103276,
      "balance_loss_clip": 1.03587198,
      "balance_loss_mlp": 1.02132154,
      "epoch": 0.7847888170750038,
      "flos": 18478159274880.0,
      "grad_norm": 2.9168964052599478,
      "language_loss": 0.68898618,
      "learning_rate": 4.663823364159183e-07,
      "loss": 0.71009785,
      "num_input_tokens_seen": 281586925,
      "step": 13053,
      "time_per_iteration": 2.54837703704834
    },
    {
      "auxiliary_loss_clip": 0.01094891,
      "auxiliary_loss_mlp": 0.01031663,
      "balance_loss_clip": 1.03759289,
      "balance_loss_mlp": 1.02056456,
      "epoch": 0.7848489403276717,
      "flos": 25119155255040.0,
      "grad_norm": 2.03516021197348,
      "language_loss": 0.70019245,
      "learning_rate": 4.6613237956584893e-07,
      "loss": 0.72145802,
      "num_input_tokens_seen": 281603915,
      "step": 13054,
      "time_per_iteration": 3.9188766479492188
    },
    {
      "auxiliary_loss_clip": 0.01096444,
      "auxiliary_loss_mlp": 0.01028522,
      "balance_loss_clip": 1.03641319,
      "balance_loss_mlp": 1.01628566,
      "epoch": 0.7849090635803397,
      "flos": 26502433966080.0,
      "grad_norm": 1.6254964808833352,
      "language_loss": 0.75557363,
      "learning_rate": 4.658824808801938e-07,
      "loss": 0.77682328,
      "num_input_tokens_seen": 281624220,
      "step": 13055,
      "time_per_iteration": 2.5192887783050537
    },
    {
      "auxiliary_loss_clip": 0.01109669,
      "auxiliary_loss_mlp": 0.01032375,
      "balance_loss_clip": 1.03737414,
      "balance_loss_mlp": 1.01971531,
      "epoch": 0.7849691868330076,
      "flos": 20959658363520.0,
      "grad_norm": 2.0034261693037414,
      "language_loss": 0.74428737,
      "learning_rate": 4.656326403684283e-07,
      "loss": 0.76570779,
      "num_input_tokens_seen": 281642325,
      "step": 13056,
      "time_per_iteration": 2.4535651206970215
    },
    {
      "auxiliary_loss_clip": 0.01049099,
      "auxiliary_loss_mlp": 0.01028511,
      "balance_loss_clip": 1.03606915,
      "balance_loss_mlp": 1.01651287,
      "epoch": 0.7850293100856757,
      "flos": 26067484177920.0,
      "grad_norm": 1.6997047032754706,
      "language_loss": 0.70140338,
      "learning_rate": 4.6538285804002744e-07,
      "loss": 0.72217953,
      "num_input_tokens_seen": 281663065,
      "step": 13057,
      "time_per_iteration": 4.3238747119903564
    },
    {
      "auxiliary_loss_clip": 0.01066938,
      "auxiliary_loss_mlp": 0.01030874,
      "balance_loss_clip": 1.03786623,
      "balance_loss_mlp": 1.01917362,
      "epoch": 0.7850894333383436,
      "flos": 22491894775680.0,
      "grad_norm": 1.849544944620275,
      "language_loss": 0.76612771,
      "learning_rate": 4.6513313390446175e-07,
      "loss": 0.7871058,
      "num_input_tokens_seen": 281681005,
      "step": 13058,
      "time_per_iteration": 2.5695457458496094
    },
    {
      "auxiliary_loss_clip": 0.01096172,
      "auxiliary_loss_mlp": 0.010343,
      "balance_loss_clip": 1.03762984,
      "balance_loss_mlp": 1.02188468,
      "epoch": 0.7851495565910116,
      "flos": 20558643949440.0,
      "grad_norm": 1.768928003588431,
      "language_loss": 0.70775157,
      "learning_rate": 4.6488346797120146e-07,
      "loss": 0.72905636,
      "num_input_tokens_seen": 281697965,
      "step": 13059,
      "time_per_iteration": 3.8780956268310547
    },
    {
      "auxiliary_loss_clip": 0.01076102,
      "auxiliary_loss_mlp": 0.01039252,
      "balance_loss_clip": 1.03482485,
      "balance_loss_mlp": 1.02635944,
      "epoch": 0.7852096798436795,
      "flos": 15924838942080.0,
      "grad_norm": 2.0005750833308413,
      "language_loss": 0.7698831,
      "learning_rate": 4.646338602497144e-07,
      "loss": 0.79103661,
      "num_input_tokens_seen": 281716035,
      "step": 13060,
      "time_per_iteration": 2.5143232345581055
    },
    {
      "auxiliary_loss_clip": 0.01085531,
      "auxiliary_loss_mlp": 0.01028721,
      "balance_loss_clip": 1.03709698,
      "balance_loss_mlp": 1.01630569,
      "epoch": 0.7852698030963475,
      "flos": 19062282245760.0,
      "grad_norm": 2.148073697765887,
      "language_loss": 0.76926279,
      "learning_rate": 4.643843107494654e-07,
      "loss": 0.79040527,
      "num_input_tokens_seen": 281732815,
      "step": 13061,
      "time_per_iteration": 2.5092647075653076
    },
    {
      "auxiliary_loss_clip": 0.01068838,
      "auxiliary_loss_mlp": 0.01032616,
      "balance_loss_clip": 1.03407431,
      "balance_loss_mlp": 1.01986086,
      "epoch": 0.7853299263490154,
      "flos": 24644380262400.0,
      "grad_norm": 1.9600387403927144,
      "language_loss": 0.73906589,
      "learning_rate": 4.641348194799164e-07,
      "loss": 0.76008046,
      "num_input_tokens_seen": 281751980,
      "step": 13062,
      "time_per_iteration": 2.5772688388824463
    },
    {
      "auxiliary_loss_clip": 0.01092107,
      "auxiliary_loss_mlp": 0.01031381,
      "balance_loss_clip": 1.03458726,
      "balance_loss_mlp": 1.01969218,
      "epoch": 0.7853900496016835,
      "flos": 22017981709440.0,
      "grad_norm": 1.403548412349145,
      "language_loss": 0.68427092,
      "learning_rate": 4.638853864505297e-07,
      "loss": 0.70550585,
      "num_input_tokens_seen": 281772670,
      "step": 13063,
      "time_per_iteration": 3.9192802906036377
    },
    {
      "auxiliary_loss_clip": 0.01093886,
      "auxiliary_loss_mlp": 0.01032308,
      "balance_loss_clip": 1.03846264,
      "balance_loss_mlp": 1.02016675,
      "epoch": 0.7854501728543514,
      "flos": 30227412032640.0,
      "grad_norm": 1.8624799989300924,
      "language_loss": 0.72707045,
      "learning_rate": 4.636360116707625e-07,
      "loss": 0.74833238,
      "num_input_tokens_seen": 281792930,
      "step": 13064,
      "time_per_iteration": 2.55252742767334
    },
    {
      "auxiliary_loss_clip": 0.01075304,
      "auxiliary_loss_mlp": 0.01030474,
      "balance_loss_clip": 1.03412318,
      "balance_loss_mlp": 1.01866031,
      "epoch": 0.7855102961070194,
      "flos": 18843694030080.0,
      "grad_norm": 1.7300841396453583,
      "language_loss": 0.6782558,
      "learning_rate": 4.633866951500718e-07,
      "loss": 0.69931358,
      "num_input_tokens_seen": 281811805,
      "step": 13065,
      "time_per_iteration": 2.5524344444274902
    },
    {
      "auxiliary_loss_clip": 0.0109332,
      "auxiliary_loss_mlp": 0.01033893,
      "balance_loss_clip": 1.03964853,
      "balance_loss_mlp": 1.02211511,
      "epoch": 0.7855704193596874,
      "flos": 22309971367680.0,
      "grad_norm": 1.9047258839500067,
      "language_loss": 0.76357853,
      "learning_rate": 4.6313743689791196e-07,
      "loss": 0.78485072,
      "num_input_tokens_seen": 281831885,
      "step": 13066,
      "time_per_iteration": 2.4977493286132812
    },
    {
      "auxiliary_loss_clip": 0.0103131,
      "auxiliary_loss_mlp": 0.01005249,
      "balance_loss_clip": 1.00839877,
      "balance_loss_mlp": 1.00403905,
      "epoch": 0.7856305426123553,
      "flos": 60004434407040.0,
      "grad_norm": 0.7168198979877674,
      "language_loss": 0.53436494,
      "learning_rate": 4.628882369237346e-07,
      "loss": 0.55473053,
      "num_input_tokens_seen": 281900310,
      "step": 13067,
      "time_per_iteration": 3.1492693424224854
    },
    {
      "auxiliary_loss_clip": 0.01060122,
      "auxiliary_loss_mlp": 0.01031839,
      "balance_loss_clip": 1.03481066,
      "balance_loss_mlp": 1.0191617,
      "epoch": 0.7856906658650233,
      "flos": 21868593045120.0,
      "grad_norm": 1.6338137100815981,
      "language_loss": 0.6781888,
      "learning_rate": 4.62639095236989e-07,
      "loss": 0.69910836,
      "num_input_tokens_seen": 281918870,
      "step": 13068,
      "time_per_iteration": 2.5973987579345703
    },
    {
      "auxiliary_loss_clip": 0.01071862,
      "auxiliary_loss_mlp": 0.01030228,
      "balance_loss_clip": 1.03492546,
      "balance_loss_mlp": 1.01836061,
      "epoch": 0.7857507891176913,
      "flos": 23622937205760.0,
      "grad_norm": 1.8355498958734748,
      "language_loss": 0.68275356,
      "learning_rate": 4.6239001184712267e-07,
      "loss": 0.70377445,
      "num_input_tokens_seen": 281936905,
      "step": 13069,
      "time_per_iteration": 2.5731067657470703
    },
    {
      "auxiliary_loss_clip": 0.01096966,
      "auxiliary_loss_mlp": 0.01033684,
      "balance_loss_clip": 1.03761208,
      "balance_loss_mlp": 1.02147102,
      "epoch": 0.7858109123703593,
      "flos": 25520061928320.0,
      "grad_norm": 1.5234092608260015,
      "language_loss": 0.76775885,
      "learning_rate": 4.6214098676358195e-07,
      "loss": 0.78906536,
      "num_input_tokens_seen": 281955625,
      "step": 13070,
      "time_per_iteration": 2.5345969200134277
    },
    {
      "auxiliary_loss_clip": 0.01047378,
      "auxiliary_loss_mlp": 0.01035053,
      "balance_loss_clip": 1.03285515,
      "balance_loss_mlp": 1.02238703,
      "epoch": 0.7858710356230272,
      "flos": 17457398576640.0,
      "grad_norm": 1.5871325164000656,
      "language_loss": 0.65802586,
      "learning_rate": 4.618920199958083e-07,
      "loss": 0.67885017,
      "num_input_tokens_seen": 281973285,
      "step": 13071,
      "time_per_iteration": 2.5797383785247803
    },
    {
      "auxiliary_loss_clip": 0.01059805,
      "auxiliary_loss_mlp": 0.01031308,
      "balance_loss_clip": 1.0340519,
      "balance_loss_mlp": 1.01945806,
      "epoch": 0.7859311588756952,
      "flos": 24679680353280.0,
      "grad_norm": 1.8708565495733256,
      "language_loss": 0.7397691,
      "learning_rate": 4.616431115532442e-07,
      "loss": 0.76068026,
      "num_input_tokens_seen": 281991410,
      "step": 13072,
      "time_per_iteration": 2.6234352588653564
    },
    {
      "auxiliary_loss_clip": 0.0109132,
      "auxiliary_loss_mlp": 0.01029694,
      "balance_loss_clip": 1.03782225,
      "balance_loss_mlp": 1.01698625,
      "epoch": 0.7859912821283631,
      "flos": 21799142098560.0,
      "grad_norm": 2.469743787467539,
      "language_loss": 0.71712661,
      "learning_rate": 4.613942614453268e-07,
      "loss": 0.73833674,
      "num_input_tokens_seen": 282010845,
      "step": 13073,
      "time_per_iteration": 2.526869058609009
    },
    {
      "auxiliary_loss_clip": 0.01079831,
      "auxiliary_loss_mlp": 0.01031907,
      "balance_loss_clip": 1.03774846,
      "balance_loss_mlp": 1.01964676,
      "epoch": 0.7860514053810311,
      "flos": 20847293642880.0,
      "grad_norm": 1.5774651732552534,
      "language_loss": 0.76554757,
      "learning_rate": 4.611454696814938e-07,
      "loss": 0.78666496,
      "num_input_tokens_seen": 282029635,
      "step": 13074,
      "time_per_iteration": 2.542867422103882
    },
    {
      "auxiliary_loss_clip": 0.01069869,
      "auxiliary_loss_mlp": 0.01029825,
      "balance_loss_clip": 1.03496373,
      "balance_loss_mlp": 1.01813626,
      "epoch": 0.786111528633699,
      "flos": 24315689882880.0,
      "grad_norm": 1.606608931650668,
      "language_loss": 0.75167197,
      "learning_rate": 4.608967362711782e-07,
      "loss": 0.77266896,
      "num_input_tokens_seen": 282050285,
      "step": 13075,
      "time_per_iteration": 2.5756590366363525
    },
    {
      "auxiliary_loss_clip": 0.01074737,
      "auxiliary_loss_mlp": 0.01027255,
      "balance_loss_clip": 1.03638673,
      "balance_loss_mlp": 1.01649642,
      "epoch": 0.7861716518863671,
      "flos": 24353180703360.0,
      "grad_norm": 1.840073586287196,
      "language_loss": 0.68871188,
      "learning_rate": 4.6064806122381283e-07,
      "loss": 0.70973182,
      "num_input_tokens_seen": 282071040,
      "step": 13076,
      "time_per_iteration": 2.591090202331543
    },
    {
      "auxiliary_loss_clip": 0.01090509,
      "auxiliary_loss_mlp": 0.01030573,
      "balance_loss_clip": 1.0349431,
      "balance_loss_mlp": 1.01825273,
      "epoch": 0.786231775139035,
      "flos": 14022399006720.0,
      "grad_norm": 1.9081288645049805,
      "language_loss": 0.80165279,
      "learning_rate": 4.603994445488282e-07,
      "loss": 0.82286364,
      "num_input_tokens_seen": 282086610,
      "step": 13077,
      "time_per_iteration": 2.4528255462646484
    },
    {
      "auxiliary_loss_clip": 0.01093946,
      "auxiliary_loss_mlp": 0.01031499,
      "balance_loss_clip": 1.03684235,
      "balance_loss_mlp": 1.01916122,
      "epoch": 0.786291898391703,
      "flos": 33724248865920.0,
      "grad_norm": 1.5207188020111702,
      "language_loss": 0.70889884,
      "learning_rate": 4.6015088625564956e-07,
      "loss": 0.73015332,
      "num_input_tokens_seen": 282107440,
      "step": 13078,
      "time_per_iteration": 2.609437942504883
    },
    {
      "auxiliary_loss_clip": 0.01091825,
      "auxiliary_loss_mlp": 0.01033047,
      "balance_loss_clip": 1.03689289,
      "balance_loss_mlp": 1.02148366,
      "epoch": 0.786352021644371,
      "flos": 25811476968960.0,
      "grad_norm": 1.427760843751378,
      "language_loss": 0.81339097,
      "learning_rate": 4.599023863537039e-07,
      "loss": 0.83463967,
      "num_input_tokens_seen": 282127290,
      "step": 13079,
      "time_per_iteration": 2.532857894897461
    },
    {
      "auxiliary_loss_clip": 0.01070329,
      "auxiliary_loss_mlp": 0.01030742,
      "balance_loss_clip": 1.035586,
      "balance_loss_mlp": 1.01869035,
      "epoch": 0.7864121448970389,
      "flos": 28910818920960.0,
      "grad_norm": 1.4601848117945384,
      "language_loss": 0.68267453,
      "learning_rate": 4.596539448524146e-07,
      "loss": 0.70368528,
      "num_input_tokens_seen": 282147505,
      "step": 13080,
      "time_per_iteration": 2.624238967895508
    },
    {
      "auxiliary_loss_clip": 0.01093574,
      "auxiliary_loss_mlp": 0.01033164,
      "balance_loss_clip": 1.0361402,
      "balance_loss_mlp": 1.02115333,
      "epoch": 0.7864722681497069,
      "flos": 19208833735680.0,
      "grad_norm": 1.5478637255103993,
      "language_loss": 0.69401526,
      "learning_rate": 4.594055617612016e-07,
      "loss": 0.71528268,
      "num_input_tokens_seen": 282166450,
      "step": 13081,
      "time_per_iteration": 2.4859063625335693
    },
    {
      "auxiliary_loss_clip": 0.01080701,
      "auxiliary_loss_mlp": 0.01034731,
      "balance_loss_clip": 1.03482664,
      "balance_loss_mlp": 1.0232451,
      "epoch": 0.7865323914023749,
      "flos": 21871573873920.0,
      "grad_norm": 1.50379712103308,
      "language_loss": 0.68448901,
      "learning_rate": 4.591572370894838e-07,
      "loss": 0.7056433,
      "num_input_tokens_seen": 282186465,
      "step": 13082,
      "time_per_iteration": 2.543154239654541
    },
    {
      "auxiliary_loss_clip": 0.01074524,
      "auxiliary_loss_mlp": 0.01031866,
      "balance_loss_clip": 1.03406572,
      "balance_loss_mlp": 1.01990914,
      "epoch": 0.7865925146550429,
      "flos": 25520313323520.0,
      "grad_norm": 1.5204480489992114,
      "language_loss": 0.66300863,
      "learning_rate": 4.589089708466789e-07,
      "loss": 0.68407261,
      "num_input_tokens_seen": 282207180,
      "step": 13083,
      "time_per_iteration": 2.53957462310791
    },
    {
      "auxiliary_loss_clip": 0.01085445,
      "auxiliary_loss_mlp": 0.01032233,
      "balance_loss_clip": 1.03700066,
      "balance_loss_mlp": 1.01919162,
      "epoch": 0.7866526379077108,
      "flos": 19097366855040.0,
      "grad_norm": 2.0402210108742804,
      "language_loss": 0.74655467,
      "learning_rate": 4.5866076304220015e-07,
      "loss": 0.76773143,
      "num_input_tokens_seen": 282225865,
      "step": 13084,
      "time_per_iteration": 2.503448963165283
    },
    {
      "auxiliary_loss_clip": 0.01079541,
      "auxiliary_loss_mlp": 0.01037954,
      "balance_loss_clip": 1.0366056,
      "balance_loss_mlp": 1.0251807,
      "epoch": 0.7867127611603788,
      "flos": 16173771171840.0,
      "grad_norm": 3.0594407408908477,
      "language_loss": 0.70680964,
      "learning_rate": 4.584126136854591e-07,
      "loss": 0.72798455,
      "num_input_tokens_seen": 282242895,
      "step": 13085,
      "time_per_iteration": 2.4677326679229736
    },
    {
      "auxiliary_loss_clip": 0.01079962,
      "auxiliary_loss_mlp": 0.01029856,
      "balance_loss_clip": 1.03393841,
      "balance_loss_mlp": 1.01734531,
      "epoch": 0.7867728844130467,
      "flos": 20773640805120.0,
      "grad_norm": 1.703855984410092,
      "language_loss": 0.72090673,
      "learning_rate": 4.5816452278586617e-07,
      "loss": 0.74200499,
      "num_input_tokens_seen": 282260425,
      "step": 13086,
      "time_per_iteration": 2.50201416015625
    },
    {
      "auxiliary_loss_clip": 0.01104704,
      "auxiliary_loss_mlp": 0.01028291,
      "balance_loss_clip": 1.03597617,
      "balance_loss_mlp": 1.01669168,
      "epoch": 0.7868330076657147,
      "flos": 21760106993280.0,
      "grad_norm": 2.154269312621411,
      "language_loss": 0.74566287,
      "learning_rate": 4.5791649035282965e-07,
      "loss": 0.76699281,
      "num_input_tokens_seen": 282279335,
      "step": 13087,
      "time_per_iteration": 2.452317953109741
    },
    {
      "auxiliary_loss_clip": 0.01082306,
      "auxiliary_loss_mlp": 0.0103144,
      "balance_loss_clip": 1.03550005,
      "balance_loss_mlp": 1.02043092,
      "epoch": 0.7868931309183826,
      "flos": 25700692446720.0,
      "grad_norm": 1.5588390951709792,
      "language_loss": 0.71306276,
      "learning_rate": 4.5766851639575456e-07,
      "loss": 0.73420024,
      "num_input_tokens_seen": 282299905,
      "step": 13088,
      "time_per_iteration": 2.5520641803741455
    },
    {
      "auxiliary_loss_clip": 0.01032658,
      "auxiliary_loss_mlp": 0.01004346,
      "balance_loss_clip": 1.00976396,
      "balance_loss_mlp": 1.00321341,
      "epoch": 0.7869532541710507,
      "flos": 64644883430400.0,
      "grad_norm": 0.6732884845195526,
      "language_loss": 0.55472881,
      "learning_rate": 4.574206009240431e-07,
      "loss": 0.57509881,
      "num_input_tokens_seen": 282367620,
      "step": 13089,
      "time_per_iteration": 3.146073818206787
    },
    {
      "auxiliary_loss_clip": 0.01021905,
      "auxiliary_loss_mlp": 0.01004526,
      "balance_loss_clip": 1.01085818,
      "balance_loss_mlp": 1.00342321,
      "epoch": 0.7870133774237186,
      "flos": 67453600440960.0,
      "grad_norm": 0.7252255157912649,
      "language_loss": 0.49934068,
      "learning_rate": 4.571727439470976e-07,
      "loss": 0.51960492,
      "num_input_tokens_seen": 282435695,
      "step": 13090,
      "time_per_iteration": 3.2068629264831543
    },
    {
      "auxiliary_loss_clip": 0.01092593,
      "auxiliary_loss_mlp": 0.0102926,
      "balance_loss_clip": 1.036098,
      "balance_loss_mlp": 1.01809001,
      "epoch": 0.7870735006763866,
      "flos": 26068310190720.0,
      "grad_norm": 1.433581849038391,
      "language_loss": 0.83761144,
      "learning_rate": 4.5692494547431583e-07,
      "loss": 0.85882998,
      "num_input_tokens_seen": 282456025,
      "step": 13091,
      "time_per_iteration": 2.5248043537139893
    },
    {
      "auxiliary_loss_clip": 0.01022318,
      "auxiliary_loss_mlp": 0.01003057,
      "balance_loss_clip": 1.00952053,
      "balance_loss_mlp": 1.00196028,
      "epoch": 0.7871336239290546,
      "flos": 70289572896000.0,
      "grad_norm": 0.7111967543549162,
      "language_loss": 0.64026529,
      "learning_rate": 4.566772055150947e-07,
      "loss": 0.66051906,
      "num_input_tokens_seen": 282520995,
      "step": 13092,
      "time_per_iteration": 3.1398584842681885
    },
    {
      "auxiliary_loss_clip": 0.01084056,
      "auxiliary_loss_mlp": 0.01034103,
      "balance_loss_clip": 1.03828132,
      "balance_loss_mlp": 1.02163339,
      "epoch": 0.7871937471817225,
      "flos": 15778574760960.0,
      "grad_norm": 2.6261363255698202,
      "language_loss": 0.79096669,
      "learning_rate": 4.564295240788285e-07,
      "loss": 0.81214827,
      "num_input_tokens_seen": 282539355,
      "step": 13093,
      "time_per_iteration": 2.511087417602539
    },
    {
      "auxiliary_loss_clip": 0.01076244,
      "auxiliary_loss_mlp": 0.010245,
      "balance_loss_clip": 1.03712618,
      "balance_loss_mlp": 1.01297295,
      "epoch": 0.7872538704343905,
      "flos": 20485242506880.0,
      "grad_norm": 1.7261222886463472,
      "language_loss": 0.7532028,
      "learning_rate": 4.561819011749106e-07,
      "loss": 0.77421027,
      "num_input_tokens_seen": 282555735,
      "step": 13094,
      "time_per_iteration": 5.309385061264038
    },
    {
      "auxiliary_loss_clip": 0.01059652,
      "auxiliary_loss_mlp": 0.01044535,
      "balance_loss_clip": 1.03510666,
      "balance_loss_mlp": 1.0307188,
      "epoch": 0.7873139936870585,
      "flos": 25082670015360.0,
      "grad_norm": 1.6172613991369318,
      "language_loss": 0.80051059,
      "learning_rate": 4.5593433681272884e-07,
      "loss": 0.82155246,
      "num_input_tokens_seen": 282574550,
      "step": 13095,
      "time_per_iteration": 2.60383677482605
    },
    {
      "auxiliary_loss_clip": 0.01095353,
      "auxiliary_loss_mlp": 0.01031162,
      "balance_loss_clip": 1.03643203,
      "balance_loss_mlp": 1.01908648,
      "epoch": 0.7873741169397265,
      "flos": 30883176679680.0,
      "grad_norm": 1.7025478642646834,
      "language_loss": 0.6794678,
      "learning_rate": 4.556868310016715e-07,
      "loss": 0.70073295,
      "num_input_tokens_seen": 282596520,
      "step": 13096,
      "time_per_iteration": 2.5673410892486572
    },
    {
      "auxiliary_loss_clip": 0.01076248,
      "auxiliary_loss_mlp": 0.01027635,
      "balance_loss_clip": 1.0332557,
      "balance_loss_mlp": 1.01740062,
      "epoch": 0.7874342401923944,
      "flos": 46791962242560.0,
      "grad_norm": 1.4487673323533563,
      "language_loss": 0.70518196,
      "learning_rate": 4.55439383751125e-07,
      "loss": 0.72622085,
      "num_input_tokens_seen": 282620560,
      "step": 13097,
      "time_per_iteration": 4.162252426147461
    },
    {
      "auxiliary_loss_clip": 0.01086456,
      "auxiliary_loss_mlp": 0.01033118,
      "balance_loss_clip": 1.03799474,
      "balance_loss_mlp": 1.02109528,
      "epoch": 0.7874943634450624,
      "flos": 23584548545280.0,
      "grad_norm": 2.065542326342619,
      "language_loss": 0.80700362,
      "learning_rate": 4.5519199507047126e-07,
      "loss": 0.82819939,
      "num_input_tokens_seen": 282639830,
      "step": 13098,
      "time_per_iteration": 2.5486583709716797
    },
    {
      "auxiliary_loss_clip": 0.01066292,
      "auxiliary_loss_mlp": 0.01030981,
      "balance_loss_clip": 1.03537273,
      "balance_loss_mlp": 1.01944125,
      "epoch": 0.7875544866977303,
      "flos": 20191169859840.0,
      "grad_norm": 1.6617089773634088,
      "language_loss": 0.74348581,
      "learning_rate": 4.5494466496909177e-07,
      "loss": 0.76445854,
      "num_input_tokens_seen": 282660130,
      "step": 13099,
      "time_per_iteration": 2.5750772953033447
    },
    {
      "auxiliary_loss_clip": 0.01082884,
      "auxiliary_loss_mlp": 0.01023505,
      "balance_loss_clip": 1.03533471,
      "balance_loss_mlp": 1.01165605,
      "epoch": 0.7876146099503983,
      "flos": 22602571557120.0,
      "grad_norm": 1.7609556189803828,
      "language_loss": 0.78126663,
      "learning_rate": 4.5469739345636603e-07,
      "loss": 0.80233049,
      "num_input_tokens_seen": 282681125,
      "step": 13100,
      "time_per_iteration": 2.5394973754882812
    },
    {
      "auxiliary_loss_clip": 0.01097357,
      "auxiliary_loss_mlp": 0.00786932,
      "balance_loss_clip": 1.03718638,
      "balance_loss_mlp": 1.01265693,
      "epoch": 0.7876747332030662,
      "flos": 10705833555840.0,
      "grad_norm": 2.0139717009989613,
      "language_loss": 0.65972173,
      "learning_rate": 4.5445018054167007e-07,
      "loss": 0.67856461,
      "num_input_tokens_seen": 282696690,
      "step": 13101,
      "time_per_iteration": 2.4449574947357178
    },
    {
      "auxiliary_loss_clip": 0.01081696,
      "auxiliary_loss_mlp": 0.01030825,
      "balance_loss_clip": 1.03555715,
      "balance_loss_mlp": 1.01893973,
      "epoch": 0.7877348564557343,
      "flos": 38399315621760.0,
      "grad_norm": 1.418390588387776,
      "language_loss": 0.77796876,
      "learning_rate": 4.5420302623437745e-07,
      "loss": 0.79909396,
      "num_input_tokens_seen": 282721210,
      "step": 13102,
      "time_per_iteration": 4.0586512088775635
    },
    {
      "auxiliary_loss_clip": 0.01092671,
      "auxiliary_loss_mlp": 0.01035095,
      "balance_loss_clip": 1.03524148,
      "balance_loss_mlp": 1.0233891,
      "epoch": 0.7877949797084022,
      "flos": 18329524796160.0,
      "grad_norm": 1.8399714163393464,
      "language_loss": 0.82297242,
      "learning_rate": 4.5395593054386093e-07,
      "loss": 0.84425008,
      "num_input_tokens_seen": 282738505,
      "step": 13103,
      "time_per_iteration": 2.483966112136841
    },
    {
      "auxiliary_loss_clip": 0.01096803,
      "auxiliary_loss_mlp": 0.01031949,
      "balance_loss_clip": 1.03702545,
      "balance_loss_mlp": 1.01949167,
      "epoch": 0.7878551029610702,
      "flos": 25806736373760.0,
      "grad_norm": 2.0466430645179208,
      "language_loss": 0.80789924,
      "learning_rate": 4.537088934794913e-07,
      "loss": 0.8291868,
      "num_input_tokens_seen": 282756895,
      "step": 13104,
      "time_per_iteration": 2.519477128982544
    },
    {
      "auxiliary_loss_clip": 0.01106625,
      "auxiliary_loss_mlp": 0.01032787,
      "balance_loss_clip": 1.03680897,
      "balance_loss_mlp": 1.020854,
      "epoch": 0.7879152262137382,
      "flos": 22342685679360.0,
      "grad_norm": 1.648149375016571,
      "language_loss": 0.7376399,
      "learning_rate": 4.5346191505063515e-07,
      "loss": 0.75903404,
      "num_input_tokens_seen": 282774955,
      "step": 13105,
      "time_per_iteration": 2.4851861000061035
    },
    {
      "auxiliary_loss_clip": 0.01047081,
      "auxiliary_loss_mlp": 0.01037467,
      "balance_loss_clip": 1.03395295,
      "balance_loss_mlp": 1.02527833,
      "epoch": 0.7879753494664061,
      "flos": 24785329230720.0,
      "grad_norm": 1.8479717583421433,
      "language_loss": 0.75870419,
      "learning_rate": 4.5321499526665776e-07,
      "loss": 0.77954966,
      "num_input_tokens_seen": 282793165,
      "step": 13106,
      "time_per_iteration": 2.642749309539795
    },
    {
      "auxiliary_loss_clip": 0.01056563,
      "auxiliary_loss_mlp": 0.01031538,
      "balance_loss_clip": 1.03815031,
      "balance_loss_mlp": 1.01988482,
      "epoch": 0.7880354727190741,
      "flos": 16909078487040.0,
      "grad_norm": 2.1418264226565626,
      "language_loss": 0.73345172,
      "learning_rate": 4.5296813413692337e-07,
      "loss": 0.75433278,
      "num_input_tokens_seen": 282809820,
      "step": 13107,
      "time_per_iteration": 2.5739641189575195
    },
    {
      "auxiliary_loss_clip": 0.01104477,
      "auxiliary_loss_mlp": 0.01032354,
      "balance_loss_clip": 1.03667927,
      "balance_loss_mlp": 1.02018237,
      "epoch": 0.7880955959717421,
      "flos": 22230500526720.0,
      "grad_norm": 1.6596467975797882,
      "language_loss": 0.73405492,
      "learning_rate": 4.5272133167079165e-07,
      "loss": 0.75542319,
      "num_input_tokens_seen": 282828600,
      "step": 13108,
      "time_per_iteration": 2.469514846801758
    },
    {
      "auxiliary_loss_clip": 0.0103179,
      "auxiliary_loss_mlp": 0.01001777,
      "balance_loss_clip": 1.00887656,
      "balance_loss_mlp": 1.00062692,
      "epoch": 0.7881557192244101,
      "flos": 69183200131200.0,
      "grad_norm": 0.8825754435429964,
      "language_loss": 0.60356319,
      "learning_rate": 4.5247458787762216e-07,
      "loss": 0.62389886,
      "num_input_tokens_seen": 282882775,
      "step": 13109,
      "time_per_iteration": 3.029965400695801
    },
    {
      "auxiliary_loss_clip": 0.01070722,
      "auxiliary_loss_mlp": 0.01030134,
      "balance_loss_clip": 1.03677189,
      "balance_loss_mlp": 1.0182668,
      "epoch": 0.788215842477078,
      "flos": 24935436167040.0,
      "grad_norm": 1.5725876567387969,
      "language_loss": 0.7217921,
      "learning_rate": 4.5222790276677126e-07,
      "loss": 0.74280065,
      "num_input_tokens_seen": 282902680,
      "step": 13110,
      "time_per_iteration": 2.619856595993042
    },
    {
      "auxiliary_loss_clip": 0.01053989,
      "auxiliary_loss_mlp": 0.01027188,
      "balance_loss_clip": 1.03429985,
      "balance_loss_mlp": 1.01577961,
      "epoch": 0.788275965729746,
      "flos": 26106483369600.0,
      "grad_norm": 1.3576226973574288,
      "language_loss": 0.75164557,
      "learning_rate": 4.5198127634759455e-07,
      "loss": 0.77245736,
      "num_input_tokens_seen": 282923625,
      "step": 13111,
      "time_per_iteration": 2.625410795211792
    },
    {
      "auxiliary_loss_clip": 0.01088734,
      "auxiliary_loss_mlp": 0.010347,
      "balance_loss_clip": 1.03456223,
      "balance_loss_mlp": 1.02201664,
      "epoch": 0.7883360889824139,
      "flos": 21214803646080.0,
      "grad_norm": 2.0199596380797065,
      "language_loss": 0.61126029,
      "learning_rate": 4.5173470862944206e-07,
      "loss": 0.63249469,
      "num_input_tokens_seen": 282941955,
      "step": 13112,
      "time_per_iteration": 2.512524366378784
    },
    {
      "auxiliary_loss_clip": 0.01081873,
      "auxiliary_loss_mlp": 0.01028885,
      "balance_loss_clip": 1.03400826,
      "balance_loss_mlp": 1.01644576,
      "epoch": 0.7883962122350819,
      "flos": 21142551438720.0,
      "grad_norm": 1.786407504260886,
      "language_loss": 0.67571372,
      "learning_rate": 4.514881996216644e-07,
      "loss": 0.69682133,
      "num_input_tokens_seen": 282961280,
      "step": 13113,
      "time_per_iteration": 2.52264404296875
    },
    {
      "auxiliary_loss_clip": 0.01067421,
      "auxiliary_loss_mlp": 0.01031263,
      "balance_loss_clip": 1.03552139,
      "balance_loss_mlp": 1.0185312,
      "epoch": 0.7884563354877498,
      "flos": 15302901928320.0,
      "grad_norm": 2.2866354632409087,
      "language_loss": 0.58427382,
      "learning_rate": 4.5124174933361e-07,
      "loss": 0.60526061,
      "num_input_tokens_seen": 282978210,
      "step": 13114,
      "time_per_iteration": 2.5269784927368164
    },
    {
      "auxiliary_loss_clip": 0.01058269,
      "auxiliary_loss_mlp": 0.01028801,
      "balance_loss_clip": 1.03744376,
      "balance_loss_mlp": 1.01613533,
      "epoch": 0.7885164587404179,
      "flos": 24388301226240.0,
      "grad_norm": 1.7302605455778182,
      "language_loss": 0.66649902,
      "learning_rate": 4.5099535777462306e-07,
      "loss": 0.6873697,
      "num_input_tokens_seen": 282998845,
      "step": 13115,
      "time_per_iteration": 2.6236155033111572
    },
    {
      "auxiliary_loss_clip": 0.01081372,
      "auxiliary_loss_mlp": 0.01031643,
      "balance_loss_clip": 1.03555477,
      "balance_loss_mlp": 1.01838648,
      "epoch": 0.7885765819930858,
      "flos": 14385886686720.0,
      "grad_norm": 1.8340264697151658,
      "language_loss": 0.88450432,
      "learning_rate": 4.50749024954048e-07,
      "loss": 0.90563446,
      "num_input_tokens_seen": 283015200,
      "step": 13116,
      "time_per_iteration": 2.517118215560913
    },
    {
      "auxiliary_loss_clip": 0.01089341,
      "auxiliary_loss_mlp": 0.01032422,
      "balance_loss_clip": 1.03612733,
      "balance_loss_mlp": 1.01934445,
      "epoch": 0.7886367052457538,
      "flos": 18259930195200.0,
      "grad_norm": 1.6327572938292425,
      "language_loss": 0.72720528,
      "learning_rate": 4.505027508812245e-07,
      "loss": 0.74842286,
      "num_input_tokens_seen": 283033680,
      "step": 13117,
      "time_per_iteration": 2.505702257156372
    },
    {
      "auxiliary_loss_clip": 0.01092393,
      "auxiliary_loss_mlp": 0.01027021,
      "balance_loss_clip": 1.03639269,
      "balance_loss_mlp": 1.01582682,
      "epoch": 0.7886968284984217,
      "flos": 15305092657920.0,
      "grad_norm": 1.4172235578552708,
      "language_loss": 0.80048156,
      "learning_rate": 4.502565355654926e-07,
      "loss": 0.82167578,
      "num_input_tokens_seen": 283050620,
      "step": 13118,
      "time_per_iteration": 2.4778311252593994
    },
    {
      "auxiliary_loss_clip": 0.01093148,
      "auxiliary_loss_mlp": 0.01025489,
      "balance_loss_clip": 1.03625679,
      "balance_loss_mlp": 1.01387787,
      "epoch": 0.7887569517510897,
      "flos": 21215450090880.0,
      "grad_norm": 3.314945247564775,
      "language_loss": 0.73212045,
      "learning_rate": 4.500103790161878e-07,
      "loss": 0.75330681,
      "num_input_tokens_seen": 283070215,
      "step": 13119,
      "time_per_iteration": 2.482278347015381
    },
    {
      "auxiliary_loss_clip": 0.01092249,
      "auxiliary_loss_mlp": 0.01027303,
      "balance_loss_clip": 1.03536367,
      "balance_loss_mlp": 1.01490545,
      "epoch": 0.7888170750037578,
      "flos": 22711237176960.0,
      "grad_norm": 1.3270408164428418,
      "language_loss": 0.71850485,
      "learning_rate": 4.4976428124264454e-07,
      "loss": 0.73970032,
      "num_input_tokens_seen": 283091485,
      "step": 13120,
      "time_per_iteration": 2.515949249267578
    },
    {
      "auxiliary_loss_clip": 0.01079254,
      "auxiliary_loss_mlp": 0.00786952,
      "balance_loss_clip": 1.03473234,
      "balance_loss_mlp": 1.01012182,
      "epoch": 0.7888771982564257,
      "flos": 36429148592640.0,
      "grad_norm": 1.4010531520633145,
      "language_loss": 0.78735566,
      "learning_rate": 4.4951824225419564e-07,
      "loss": 0.80601776,
      "num_input_tokens_seen": 283115040,
      "step": 13121,
      "time_per_iteration": 2.633202075958252
    },
    {
      "auxiliary_loss_clip": 0.01091001,
      "auxiliary_loss_mlp": 0.01031177,
      "balance_loss_clip": 1.03459418,
      "balance_loss_mlp": 1.01877928,
      "epoch": 0.7889373215090937,
      "flos": 27309993488640.0,
      "grad_norm": 1.4813101558450685,
      "language_loss": 0.80203754,
      "learning_rate": 4.4927226206017057e-07,
      "loss": 0.82325935,
      "num_input_tokens_seen": 283136925,
      "step": 13122,
      "time_per_iteration": 2.5559046268463135
    },
    {
      "auxiliary_loss_clip": 0.01083592,
      "auxiliary_loss_mlp": 0.01025626,
      "balance_loss_clip": 1.03569484,
      "balance_loss_mlp": 1.01416361,
      "epoch": 0.7889974447617616,
      "flos": 19829010983040.0,
      "grad_norm": 5.139496870167601,
      "language_loss": 0.78098136,
      "learning_rate": 4.4902634066989597e-07,
      "loss": 0.80207348,
      "num_input_tokens_seen": 283155725,
      "step": 13123,
      "time_per_iteration": 2.550250768661499
    },
    {
      "auxiliary_loss_clip": 0.0108235,
      "auxiliary_loss_mlp": 0.0103331,
      "balance_loss_clip": 1.03766835,
      "balance_loss_mlp": 1.0208168,
      "epoch": 0.7890575680144296,
      "flos": 17271201450240.0,
      "grad_norm": 1.8562750753026471,
      "language_loss": 0.67136782,
      "learning_rate": 4.487804780926985e-07,
      "loss": 0.69252443,
      "num_input_tokens_seen": 283173845,
      "step": 13124,
      "time_per_iteration": 2.552367687225342
    },
    {
      "auxiliary_loss_clip": 0.01083058,
      "auxiliary_loss_mlp": 0.01027389,
      "balance_loss_clip": 1.03540063,
      "balance_loss_mlp": 1.01498568,
      "epoch": 0.7891176912670975,
      "flos": 27600151553280.0,
      "grad_norm": 2.3676472898377026,
      "language_loss": 0.72788393,
      "learning_rate": 4.4853467433790036e-07,
      "loss": 0.74898839,
      "num_input_tokens_seen": 283191985,
      "step": 13125,
      "time_per_iteration": 2.5396409034729004
    },
    {
      "auxiliary_loss_clip": 0.01086847,
      "auxiliary_loss_mlp": 0.01027636,
      "balance_loss_clip": 1.03387868,
      "balance_loss_mlp": 1.01491618,
      "epoch": 0.7891778145197655,
      "flos": 22711668140160.0,
      "grad_norm": 1.9353253072029226,
      "language_loss": 0.72315526,
      "learning_rate": 4.4828892941482267e-07,
      "loss": 0.74430007,
      "num_input_tokens_seen": 283210855,
      "step": 13126,
      "time_per_iteration": 2.549556016921997
    },
    {
      "auxiliary_loss_clip": 0.01086424,
      "auxiliary_loss_mlp": 0.01030437,
      "balance_loss_clip": 1.0345099,
      "balance_loss_mlp": 1.01744938,
      "epoch": 0.7892379377724335,
      "flos": 17310775259520.0,
      "grad_norm": 1.7157654873688692,
      "language_loss": 0.76728386,
      "learning_rate": 4.480432433327845e-07,
      "loss": 0.78845251,
      "num_input_tokens_seen": 283229665,
      "step": 13127,
      "time_per_iteration": 2.4817049503326416
    },
    {
      "auxiliary_loss_clip": 0.01088985,
      "auxiliary_loss_mlp": 0.01034169,
      "balance_loss_clip": 1.0358814,
      "balance_loss_mlp": 1.02105594,
      "epoch": 0.7892980610251015,
      "flos": 25775674087680.0,
      "grad_norm": 1.6195762634593,
      "language_loss": 0.85785109,
      "learning_rate": 4.47797616101103e-07,
      "loss": 0.87908268,
      "num_input_tokens_seen": 283248615,
      "step": 13128,
      "time_per_iteration": 2.534285068511963
    },
    {
      "auxiliary_loss_clip": 0.01093759,
      "auxiliary_loss_mlp": 0.01032251,
      "balance_loss_clip": 1.03633285,
      "balance_loss_mlp": 1.02127814,
      "epoch": 0.7893581842777694,
      "flos": 21579943351680.0,
      "grad_norm": 2.229045492202683,
      "language_loss": 0.69179684,
      "learning_rate": 4.475520477290904e-07,
      "loss": 0.71305692,
      "num_input_tokens_seen": 283267135,
      "step": 13129,
      "time_per_iteration": 2.484806537628174
    },
    {
      "auxiliary_loss_clip": 0.01025263,
      "auxiliary_loss_mlp": 0.01015608,
      "balance_loss_clip": 1.0127455,
      "balance_loss_mlp": 1.01408184,
      "epoch": 0.7894183075304374,
      "flos": 69016468176000.0,
      "grad_norm": 0.7157991328470195,
      "language_loss": 0.6161499,
      "learning_rate": 4.473065382260597e-07,
      "loss": 0.63655865,
      "num_input_tokens_seen": 283328940,
      "step": 13130,
      "time_per_iteration": 3.1297969818115234
    },
    {
      "auxiliary_loss_clip": 0.0109669,
      "auxiliary_loss_mlp": 0.01030883,
      "balance_loss_clip": 1.03827095,
      "balance_loss_mlp": 1.01931381,
      "epoch": 0.7894784307831053,
      "flos": 24243258107520.0,
      "grad_norm": 1.559947820501089,
      "language_loss": 0.73948091,
      "learning_rate": 4.4706108760132124e-07,
      "loss": 0.76075661,
      "num_input_tokens_seen": 283350000,
      "step": 13131,
      "time_per_iteration": 2.5693306922912598
    },
    {
      "auxiliary_loss_clip": 0.01093426,
      "auxiliary_loss_mlp": 0.01026946,
      "balance_loss_clip": 1.03802848,
      "balance_loss_mlp": 1.01333833,
      "epoch": 0.7895385540357733,
      "flos": 20266546550400.0,
      "grad_norm": 5.0011340212951865,
      "language_loss": 0.68973231,
      "learning_rate": 4.4681569586418153e-07,
      "loss": 0.71093601,
      "num_input_tokens_seen": 283368020,
      "step": 13132,
      "time_per_iteration": 3.8918983936309814
    },
    {
      "auxiliary_loss_clip": 0.01098247,
      "auxiliary_loss_mlp": 0.01039925,
      "balance_loss_clip": 1.03797698,
      "balance_loss_mlp": 1.02728915,
      "epoch": 0.7895986772884414,
      "flos": 20996574566400.0,
      "grad_norm": 1.9229138751777557,
      "language_loss": 0.62084764,
      "learning_rate": 4.465703630239468e-07,
      "loss": 0.64222944,
      "num_input_tokens_seen": 283387030,
      "step": 13133,
      "time_per_iteration": 3.8475501537323
    },
    {
      "auxiliary_loss_clip": 0.01076824,
      "auxiliary_loss_mlp": 0.01034981,
      "balance_loss_clip": 1.03668118,
      "balance_loss_mlp": 1.0217663,
      "epoch": 0.7896588005411093,
      "flos": 18657999694080.0,
      "grad_norm": 2.057836366561185,
      "language_loss": 0.79681021,
      "learning_rate": 4.463250890899195e-07,
      "loss": 0.81792825,
      "num_input_tokens_seen": 283402090,
      "step": 13134,
      "time_per_iteration": 2.5017220973968506
    },
    {
      "auxiliary_loss_clip": 0.01093158,
      "auxiliary_loss_mlp": 0.01032016,
      "balance_loss_clip": 1.03425729,
      "balance_loss_mlp": 1.01987493,
      "epoch": 0.7897189237937773,
      "flos": 18405907067520.0,
      "grad_norm": 4.904785344937257,
      "language_loss": 0.80519736,
      "learning_rate": 4.460798740713998e-07,
      "loss": 0.82644904,
      "num_input_tokens_seen": 283421035,
      "step": 13135,
      "time_per_iteration": 3.8616526126861572
    },
    {
      "auxiliary_loss_clip": 0.0109389,
      "auxiliary_loss_mlp": 0.01029043,
      "balance_loss_clip": 1.03639662,
      "balance_loss_mlp": 1.01654434,
      "epoch": 0.7897790470464452,
      "flos": 23731602825600.0,
      "grad_norm": 1.5038441853985973,
      "language_loss": 0.72485703,
      "learning_rate": 4.4583471797768733e-07,
      "loss": 0.74608636,
      "num_input_tokens_seen": 283441830,
      "step": 13136,
      "time_per_iteration": 2.5261356830596924
    },
    {
      "auxiliary_loss_clip": 0.01112291,
      "auxiliary_loss_mlp": 0.01036187,
      "balance_loss_clip": 1.03754139,
      "balance_loss_mlp": 1.0234791,
      "epoch": 0.7898391702991132,
      "flos": 15918949111680.0,
      "grad_norm": 2.2669578990135415,
      "language_loss": 0.71208918,
      "learning_rate": 4.455896208180778e-07,
      "loss": 0.73357397,
      "num_input_tokens_seen": 283459540,
      "step": 13137,
      "time_per_iteration": 2.4155349731445312
    },
    {
      "auxiliary_loss_clip": 0.01104126,
      "auxiliary_loss_mlp": 0.01031417,
      "balance_loss_clip": 1.03664231,
      "balance_loss_mlp": 1.01868606,
      "epoch": 0.7898992935517811,
      "flos": 19829046896640.0,
      "grad_norm": 1.6711437437807006,
      "language_loss": 0.73751533,
      "learning_rate": 4.4534458260186645e-07,
      "loss": 0.75887084,
      "num_input_tokens_seen": 283478790,
      "step": 13138,
      "time_per_iteration": 2.457265853881836
    },
    {
      "auxiliary_loss_clip": 0.01066468,
      "auxiliary_loss_mlp": 0.01030326,
      "balance_loss_clip": 1.03719032,
      "balance_loss_mlp": 1.01844716,
      "epoch": 0.7899594168044491,
      "flos": 16216253982720.0,
      "grad_norm": 2.149827731168992,
      "language_loss": 0.68059599,
      "learning_rate": 4.4509960333834426e-07,
      "loss": 0.70156395,
      "num_input_tokens_seen": 283495720,
      "step": 13139,
      "time_per_iteration": 2.5211362838745117
    },
    {
      "auxiliary_loss_clip": 0.01023241,
      "auxiliary_loss_mlp": 0.01003689,
      "balance_loss_clip": 1.00976491,
      "balance_loss_mlp": 1.00252104,
      "epoch": 0.790019540057117,
      "flos": 68331005959680.0,
      "grad_norm": 0.8498170750344725,
      "language_loss": 0.60205382,
      "learning_rate": 4.448546830368003e-07,
      "loss": 0.6223231,
      "num_input_tokens_seen": 283558795,
      "step": 13140,
      "time_per_iteration": 3.182072162628174
    },
    {
      "auxiliary_loss_clip": 0.01106986,
      "auxiliary_loss_mlp": 0.0103348,
      "balance_loss_clip": 1.03769672,
      "balance_loss_mlp": 1.02109361,
      "epoch": 0.7900796633097851,
      "flos": 30332773601280.0,
      "grad_norm": 1.7000944921320733,
      "language_loss": 0.76008761,
      "learning_rate": 4.4460982170652304e-07,
      "loss": 0.78149223,
      "num_input_tokens_seen": 283579305,
      "step": 13141,
      "time_per_iteration": 3.991665840148926
    },
    {
      "auxiliary_loss_clip": 0.01097005,
      "auxiliary_loss_mlp": 0.01031862,
      "balance_loss_clip": 1.03646195,
      "balance_loss_mlp": 1.01949394,
      "epoch": 0.790139786562453,
      "flos": 22126790983680.0,
      "grad_norm": 1.8610218559269482,
      "language_loss": 0.68312669,
      "learning_rate": 4.4436501935679694e-07,
      "loss": 0.70441538,
      "num_input_tokens_seen": 283597840,
      "step": 13142,
      "time_per_iteration": 2.4930179119110107
    },
    {
      "auxiliary_loss_clip": 0.00988601,
      "auxiliary_loss_mlp": 0.01010513,
      "balance_loss_clip": 1.0172286,
      "balance_loss_mlp": 1.00913072,
      "epoch": 0.790199909815121,
      "flos": 58207284213120.0,
      "grad_norm": 0.8203421553660541,
      "language_loss": 0.59952366,
      "learning_rate": 4.441202759969049e-07,
      "loss": 0.61951482,
      "num_input_tokens_seen": 283647950,
      "step": 13143,
      "time_per_iteration": 3.0622165203094482
    },
    {
      "auxiliary_loss_clip": 0.01077566,
      "auxiliary_loss_mlp": 0.01032187,
      "balance_loss_clip": 1.03758168,
      "balance_loss_mlp": 1.02001595,
      "epoch": 0.7902600330677889,
      "flos": 34533316759680.0,
      "grad_norm": 1.4622068273483844,
      "language_loss": 0.74529731,
      "learning_rate": 4.4387559163612875e-07,
      "loss": 0.76639485,
      "num_input_tokens_seen": 283670645,
      "step": 13144,
      "time_per_iteration": 3.2034802436828613
    },
    {
      "auxiliary_loss_clip": 0.01098556,
      "auxiliary_loss_mlp": 0.01031771,
      "balance_loss_clip": 1.03607714,
      "balance_loss_mlp": 1.01874161,
      "epoch": 0.7903201563204569,
      "flos": 22346384780160.0,
      "grad_norm": 2.148375051385357,
      "language_loss": 0.83436692,
      "learning_rate": 4.4363096628374605e-07,
      "loss": 0.85567015,
      "num_input_tokens_seen": 283688830,
      "step": 13145,
      "time_per_iteration": 2.5351593494415283
    },
    {
      "auxiliary_loss_clip": 0.01088333,
      "auxiliary_loss_mlp": 0.01030275,
      "balance_loss_clip": 1.0336169,
      "balance_loss_mlp": 1.01965964,
      "epoch": 0.790380279573125,
      "flos": 22053533195520.0,
      "grad_norm": 1.9545732641052236,
      "language_loss": 0.7290889,
      "learning_rate": 4.4338639994903235e-07,
      "loss": 0.75027502,
      "num_input_tokens_seen": 283708625,
      "step": 13146,
      "time_per_iteration": 2.501826286315918
    },
    {
      "auxiliary_loss_clip": 0.01107461,
      "auxiliary_loss_mlp": 0.01031846,
      "balance_loss_clip": 1.03591585,
      "balance_loss_mlp": 1.01985908,
      "epoch": 0.7904404028257929,
      "flos": 20302600826880.0,
      "grad_norm": 1.7606279585998483,
      "language_loss": 0.7557137,
      "learning_rate": 4.4314189264126246e-07,
      "loss": 0.77710676,
      "num_input_tokens_seen": 283725710,
      "step": 13147,
      "time_per_iteration": 2.471686601638794
    },
    {
      "auxiliary_loss_clip": 0.01089036,
      "auxiliary_loss_mlp": 0.01037365,
      "balance_loss_clip": 1.03563428,
      "balance_loss_mlp": 1.02424002,
      "epoch": 0.7905005260784609,
      "flos": 20008923229440.0,
      "grad_norm": 1.6746092497362883,
      "language_loss": 0.71910417,
      "learning_rate": 4.428974443697087e-07,
      "loss": 0.74036825,
      "num_input_tokens_seen": 283744150,
      "step": 13148,
      "time_per_iteration": 2.486349105834961
    },
    {
      "auxiliary_loss_clip": 0.01090842,
      "auxiliary_loss_mlp": 0.01030849,
      "balance_loss_clip": 1.03349447,
      "balance_loss_mlp": 1.0183444,
      "epoch": 0.7905606493311288,
      "flos": 26905926418560.0,
      "grad_norm": 1.8919445658321192,
      "language_loss": 0.71586269,
      "learning_rate": 4.4265305514363913e-07,
      "loss": 0.73707962,
      "num_input_tokens_seen": 283764170,
      "step": 13149,
      "time_per_iteration": 2.5693464279174805
    },
    {
      "auxiliary_loss_clip": 0.01072038,
      "auxiliary_loss_mlp": 0.01035375,
      "balance_loss_clip": 1.03454804,
      "balance_loss_mlp": 1.02061105,
      "epoch": 0.7906207725837968,
      "flos": 23696230907520.0,
      "grad_norm": 2.459241666707848,
      "language_loss": 0.6506266,
      "learning_rate": 4.424087249723225e-07,
      "loss": 0.67170072,
      "num_input_tokens_seen": 283784305,
      "step": 13150,
      "time_per_iteration": 2.589700937271118
    },
    {
      "auxiliary_loss_clip": 0.01103664,
      "auxiliary_loss_mlp": 0.01031073,
      "balance_loss_clip": 1.03521204,
      "balance_loss_mlp": 1.01946211,
      "epoch": 0.7906808958364647,
      "flos": 20848837927680.0,
      "grad_norm": 1.532853171368158,
      "language_loss": 0.69829631,
      "learning_rate": 4.421644538650231e-07,
      "loss": 0.71964371,
      "num_input_tokens_seen": 283804040,
      "step": 13151,
      "time_per_iteration": 2.481691837310791
    },
    {
      "auxiliary_loss_clip": 0.01087547,
      "auxiliary_loss_mlp": 0.01035107,
      "balance_loss_clip": 1.03673768,
      "balance_loss_mlp": 1.02185702,
      "epoch": 0.7907410190891327,
      "flos": 40735196974080.0,
      "grad_norm": 1.6565763032149325,
      "language_loss": 0.70518172,
      "learning_rate": 4.4192024183100306e-07,
      "loss": 0.72640824,
      "num_input_tokens_seen": 283827120,
      "step": 13152,
      "time_per_iteration": 2.694915294647217
    },
    {
      "auxiliary_loss_clip": 0.01072389,
      "auxiliary_loss_mlp": 0.00783579,
      "balance_loss_clip": 1.03537202,
      "balance_loss_mlp": 1.00900316,
      "epoch": 0.7908011423418007,
      "flos": 13261165050240.0,
      "grad_norm": 1.771473331837769,
      "language_loss": 0.7271086,
      "learning_rate": 4.4167608887952367e-07,
      "loss": 0.74566829,
      "num_input_tokens_seen": 283844820,
      "step": 13153,
      "time_per_iteration": 2.5986404418945312
    },
    {
      "auxiliary_loss_clip": 0.01105825,
      "auxiliary_loss_mlp": 0.01027764,
      "balance_loss_clip": 1.03641272,
      "balance_loss_mlp": 1.01568222,
      "epoch": 0.7908612655944687,
      "flos": 19754747614080.0,
      "grad_norm": 1.4732141699951657,
      "language_loss": 0.78885978,
      "learning_rate": 4.4143199501984306e-07,
      "loss": 0.81019568,
      "num_input_tokens_seen": 283862870,
      "step": 13154,
      "time_per_iteration": 2.430135726928711
    },
    {
      "auxiliary_loss_clip": 0.01099802,
      "auxiliary_loss_mlp": 0.01029504,
      "balance_loss_clip": 1.03656936,
      "balance_loss_mlp": 1.01643252,
      "epoch": 0.7909213888471366,
      "flos": 21287738211840.0,
      "grad_norm": 2.3202896864526443,
      "language_loss": 0.7003839,
      "learning_rate": 4.411879602612185e-07,
      "loss": 0.72167701,
      "num_input_tokens_seen": 283882405,
      "step": 13155,
      "time_per_iteration": 2.492530345916748
    },
    {
      "auxiliary_loss_clip": 0.01106642,
      "auxiliary_loss_mlp": 0.01029164,
      "balance_loss_clip": 1.03670502,
      "balance_loss_mlp": 1.01669455,
      "epoch": 0.7909815120998046,
      "flos": 22528882805760.0,
      "grad_norm": 1.6288749333965507,
      "language_loss": 0.76997286,
      "learning_rate": 4.4094398461290174e-07,
      "loss": 0.79133093,
      "num_input_tokens_seen": 283902070,
      "step": 13156,
      "time_per_iteration": 2.446924924850464
    },
    {
      "auxiliary_loss_clip": 0.01071904,
      "auxiliary_loss_mlp": 0.01028777,
      "balance_loss_clip": 1.03189445,
      "balance_loss_mlp": 1.01630199,
      "epoch": 0.7910416353524725,
      "flos": 26727702111360.0,
      "grad_norm": 1.6176294498247645,
      "language_loss": 0.65791535,
      "learning_rate": 4.4070006808414526e-07,
      "loss": 0.67892212,
      "num_input_tokens_seen": 283924100,
      "step": 13157,
      "time_per_iteration": 2.6245779991149902
    },
    {
      "auxiliary_loss_clip": 0.0109523,
      "auxiliary_loss_mlp": 0.0103637,
      "balance_loss_clip": 1.03461266,
      "balance_loss_mlp": 1.02297664,
      "epoch": 0.7911017586051405,
      "flos": 24644847139200.0,
      "grad_norm": 1.7769083835316202,
      "language_loss": 0.73949814,
      "learning_rate": 4.4045621068419894e-07,
      "loss": 0.76081419,
      "num_input_tokens_seen": 283944955,
      "step": 13158,
      "time_per_iteration": 2.5131125450134277
    },
    {
      "auxiliary_loss_clip": 0.01090593,
      "auxiliary_loss_mlp": 0.01027114,
      "balance_loss_clip": 1.03470421,
      "balance_loss_mlp": 1.01611137,
      "epoch": 0.7911618818578086,
      "flos": 17565489578880.0,
      "grad_norm": 2.1646325732530265,
      "language_loss": 0.67133236,
      "learning_rate": 4.40212412422309e-07,
      "loss": 0.69250941,
      "num_input_tokens_seen": 283963125,
      "step": 13159,
      "time_per_iteration": 2.4795029163360596
    },
    {
      "auxiliary_loss_clip": 0.01093162,
      "auxiliary_loss_mlp": 0.01030771,
      "balance_loss_clip": 1.03616095,
      "balance_loss_mlp": 1.01932096,
      "epoch": 0.7912220051104765,
      "flos": 16721660298240.0,
      "grad_norm": 1.948005949684928,
      "language_loss": 0.67448503,
      "learning_rate": 4.399686733077206e-07,
      "loss": 0.69572431,
      "num_input_tokens_seen": 283982850,
      "step": 13160,
      "time_per_iteration": 2.478736639022827
    },
    {
      "auxiliary_loss_clip": 0.01076698,
      "auxiliary_loss_mlp": 0.01027706,
      "balance_loss_clip": 1.03177845,
      "balance_loss_mlp": 1.01739454,
      "epoch": 0.7912821283631445,
      "flos": 13698736531200.0,
      "grad_norm": 2.432144149022387,
      "language_loss": 0.7306186,
      "learning_rate": 4.3972499334967694e-07,
      "loss": 0.75166267,
      "num_input_tokens_seen": 283998275,
      "step": 13161,
      "time_per_iteration": 2.4970364570617676
    },
    {
      "auxiliary_loss_clip": 0.01076408,
      "auxiliary_loss_mlp": 0.01029702,
      "balance_loss_clip": 1.03485048,
      "balance_loss_mlp": 1.01738214,
      "epoch": 0.7913422516158124,
      "flos": 23769021818880.0,
      "grad_norm": 2.723981496451493,
      "language_loss": 0.7346347,
      "learning_rate": 4.39481372557418e-07,
      "loss": 0.75569582,
      "num_input_tokens_seen": 284018750,
      "step": 13162,
      "time_per_iteration": 2.529572010040283
    },
    {
      "auxiliary_loss_clip": 0.01083709,
      "auxiliary_loss_mlp": 0.01028823,
      "balance_loss_clip": 1.03615499,
      "balance_loss_mlp": 1.01667571,
      "epoch": 0.7914023748684804,
      "flos": 19938251220480.0,
      "grad_norm": 1.7924277220476235,
      "language_loss": 0.71859634,
      "learning_rate": 4.392378109401811e-07,
      "loss": 0.73972166,
      "num_input_tokens_seen": 284037850,
      "step": 13163,
      "time_per_iteration": 2.5426783561706543
    },
    {
      "auxiliary_loss_clip": 0.01070134,
      "auxiliary_loss_mlp": 0.01032344,
      "balance_loss_clip": 1.03325009,
      "balance_loss_mlp": 1.01911795,
      "epoch": 0.7914624981211483,
      "flos": 20594805966720.0,
      "grad_norm": 1.8718273928293514,
      "language_loss": 0.70016432,
      "learning_rate": 4.3899430850720296e-07,
      "loss": 0.72118914,
      "num_input_tokens_seen": 284056380,
      "step": 13164,
      "time_per_iteration": 2.520493745803833
    },
    {
      "auxiliary_loss_clip": 0.01065211,
      "auxiliary_loss_mlp": 0.01029574,
      "balance_loss_clip": 1.0348196,
      "balance_loss_mlp": 1.01807642,
      "epoch": 0.7915226213738163,
      "flos": 21799465320960.0,
      "grad_norm": 1.9144319277662667,
      "language_loss": 0.6636892,
      "learning_rate": 4.387508652677177e-07,
      "loss": 0.68463707,
      "num_input_tokens_seen": 284074945,
      "step": 13165,
      "time_per_iteration": 2.565697193145752
    },
    {
      "auxiliary_loss_clip": 0.01055918,
      "auxiliary_loss_mlp": 0.01025302,
      "balance_loss_clip": 1.03543246,
      "balance_loss_mlp": 1.01461518,
      "epoch": 0.7915827446264843,
      "flos": 16288362535680.0,
      "grad_norm": 1.8207589694483788,
      "language_loss": 0.72234142,
      "learning_rate": 4.385074812309557e-07,
      "loss": 0.74315363,
      "num_input_tokens_seen": 284092070,
      "step": 13166,
      "time_per_iteration": 2.533301830291748
    },
    {
      "auxiliary_loss_clip": 0.01103766,
      "auxiliary_loss_mlp": 0.01033108,
      "balance_loss_clip": 1.03521764,
      "balance_loss_mlp": 1.01955962,
      "epoch": 0.7916428678791523,
      "flos": 25702595867520.0,
      "grad_norm": 1.691965327005288,
      "language_loss": 0.77659774,
      "learning_rate": 4.382641564061462e-07,
      "loss": 0.79796648,
      "num_input_tokens_seen": 284112255,
      "step": 13167,
      "time_per_iteration": 2.5208961963653564
    },
    {
      "auxiliary_loss_clip": 0.01071453,
      "auxiliary_loss_mlp": 0.01028607,
      "balance_loss_clip": 1.03584671,
      "balance_loss_mlp": 1.01721644,
      "epoch": 0.7917029911318202,
      "flos": 23878513451520.0,
      "grad_norm": 1.6004389293603498,
      "language_loss": 0.84378529,
      "learning_rate": 4.3802089080251713e-07,
      "loss": 0.86478591,
      "num_input_tokens_seen": 284132330,
      "step": 13168,
      "time_per_iteration": 2.5565226078033447
    },
    {
      "auxiliary_loss_clip": 0.01106639,
      "auxiliary_loss_mlp": 0.01030085,
      "balance_loss_clip": 1.03712451,
      "balance_loss_mlp": 1.01834917,
      "epoch": 0.7917631143844882,
      "flos": 21646593037440.0,
      "grad_norm": 1.971303424471412,
      "language_loss": 0.72780633,
      "learning_rate": 4.3777768442929155e-07,
      "loss": 0.74917364,
      "num_input_tokens_seen": 284150640,
      "step": 13169,
      "time_per_iteration": 2.5167088508605957
    },
    {
      "auxiliary_loss_clip": 0.01106629,
      "auxiliary_loss_mlp": 0.01034456,
      "balance_loss_clip": 1.03564358,
      "balance_loss_mlp": 1.02152824,
      "epoch": 0.7918232376371561,
      "flos": 38874198355200.0,
      "grad_norm": 1.8151550616878152,
      "language_loss": 0.67684901,
      "learning_rate": 4.3753453729569287e-07,
      "loss": 0.69825989,
      "num_input_tokens_seen": 284171910,
      "step": 13170,
      "time_per_iteration": 3.9719409942626953
    },
    {
      "auxiliary_loss_clip": 0.01093914,
      "auxiliary_loss_mlp": 0.01023028,
      "balance_loss_clip": 1.03469682,
      "balance_loss_mlp": 1.01175094,
      "epoch": 0.7918833608898241,
      "flos": 20775544225920.0,
      "grad_norm": 1.6365779857229437,
      "language_loss": 0.70617974,
      "learning_rate": 4.372914494109412e-07,
      "loss": 0.72734916,
      "num_input_tokens_seen": 284191340,
      "step": 13171,
      "time_per_iteration": 2.4768433570861816
    },
    {
      "auxiliary_loss_clip": 0.01092736,
      "auxiliary_loss_mlp": 0.01027201,
      "balance_loss_clip": 1.03536153,
      "balance_loss_mlp": 1.01502371,
      "epoch": 0.7919434841424922,
      "flos": 33910122769920.0,
      "grad_norm": 1.904992849524794,
      "language_loss": 0.66878241,
      "learning_rate": 4.370484207842553e-07,
      "loss": 0.68998182,
      "num_input_tokens_seen": 284212495,
      "step": 13172,
      "time_per_iteration": 3.9697887897491455
    },
    {
      "auxiliary_loss_clip": 0.01080224,
      "auxiliary_loss_mlp": 0.01033855,
      "balance_loss_clip": 1.03549528,
      "balance_loss_mlp": 1.02185094,
      "epoch": 0.7920036073951601,
      "flos": 21064660796160.0,
      "grad_norm": 2.0370204503780154,
      "language_loss": 0.79512984,
      "learning_rate": 4.3680545142484893e-07,
      "loss": 0.81627065,
      "num_input_tokens_seen": 284230825,
      "step": 13173,
      "time_per_iteration": 3.9237518310546875
    },
    {
      "auxiliary_loss_clip": 0.01067539,
      "auxiliary_loss_mlp": 0.01034186,
      "balance_loss_clip": 1.03231907,
      "balance_loss_mlp": 1.02147222,
      "epoch": 0.7920637306478281,
      "flos": 23655974739840.0,
      "grad_norm": 1.8273587430240612,
      "language_loss": 0.76658159,
      "learning_rate": 4.365625413419365e-07,
      "loss": 0.78759879,
      "num_input_tokens_seen": 284250365,
      "step": 13174,
      "time_per_iteration": 2.6060094833374023
    },
    {
      "auxiliary_loss_clip": 0.01078265,
      "auxiliary_loss_mlp": 0.01031215,
      "balance_loss_clip": 1.03224182,
      "balance_loss_mlp": 1.01978922,
      "epoch": 0.792123853900496,
      "flos": 27195438038400.0,
      "grad_norm": 1.6850920827691889,
      "language_loss": 0.71889567,
      "learning_rate": 4.363196905447297e-07,
      "loss": 0.73999047,
      "num_input_tokens_seen": 284269635,
      "step": 13175,
      "time_per_iteration": 2.574516534805298
    },
    {
      "auxiliary_loss_clip": 0.01092909,
      "auxiliary_loss_mlp": 0.0102987,
      "balance_loss_clip": 1.03448963,
      "balance_loss_mlp": 1.01759791,
      "epoch": 0.792183977153164,
      "flos": 19098659744640.0,
      "grad_norm": 1.8611115816172499,
      "language_loss": 0.59508944,
      "learning_rate": 4.360768990424364e-07,
      "loss": 0.61631727,
      "num_input_tokens_seen": 284288380,
      "step": 13176,
      "time_per_iteration": 2.451698064804077
    },
    {
      "auxiliary_loss_clip": 0.01106869,
      "auxiliary_loss_mlp": 0.01030684,
      "balance_loss_clip": 1.03906846,
      "balance_loss_mlp": 1.01872683,
      "epoch": 0.7922441004058319,
      "flos": 17128851851520.0,
      "grad_norm": 1.7986649271149069,
      "language_loss": 0.73413515,
      "learning_rate": 4.3583416684426376e-07,
      "loss": 0.75551069,
      "num_input_tokens_seen": 284306920,
      "step": 13177,
      "time_per_iteration": 2.4519453048706055
    },
    {
      "auxiliary_loss_clip": 0.01087542,
      "auxiliary_loss_mlp": 0.01031722,
      "balance_loss_clip": 1.03494608,
      "balance_loss_mlp": 1.01962256,
      "epoch": 0.7923042236585,
      "flos": 17821640442240.0,
      "grad_norm": 1.9093988702350015,
      "language_loss": 0.64183331,
      "learning_rate": 4.355914939594174e-07,
      "loss": 0.66302598,
      "num_input_tokens_seen": 284324700,
      "step": 13178,
      "time_per_iteration": 2.441577434539795
    },
    {
      "auxiliary_loss_clip": 0.01079865,
      "auxiliary_loss_mlp": 0.01031732,
      "balance_loss_clip": 1.03361952,
      "balance_loss_mlp": 1.02102745,
      "epoch": 0.7923643469111679,
      "flos": 29935206892800.0,
      "grad_norm": 1.5421284892907268,
      "language_loss": 0.68544,
      "learning_rate": 4.3534888039709726e-07,
      "loss": 0.70655596,
      "num_input_tokens_seen": 284345985,
      "step": 13179,
      "time_per_iteration": 2.60136342048645
    },
    {
      "auxiliary_loss_clip": 0.01102273,
      "auxiliary_loss_mlp": 0.01027685,
      "balance_loss_clip": 1.03480196,
      "balance_loss_mlp": 1.01571596,
      "epoch": 0.7924244701638359,
      "flos": 22674716023680.0,
      "grad_norm": 1.9935971334300522,
      "language_loss": 0.74359739,
      "learning_rate": 4.3510632616650444e-07,
      "loss": 0.76489699,
      "num_input_tokens_seen": 284364475,
      "step": 13180,
      "time_per_iteration": 3.8719544410705566
    },
    {
      "auxiliary_loss_clip": 0.01093669,
      "auxiliary_loss_mlp": 0.01035238,
      "balance_loss_clip": 1.03767753,
      "balance_loss_mlp": 1.02230418,
      "epoch": 0.7924845934165038,
      "flos": 17968156018560.0,
      "grad_norm": 2.009058378286736,
      "language_loss": 0.81420845,
      "learning_rate": 4.3486383127683646e-07,
      "loss": 0.8354975,
      "num_input_tokens_seen": 284382125,
      "step": 13181,
      "time_per_iteration": 2.4489307403564453
    },
    {
      "auxiliary_loss_clip": 0.01075774,
      "auxiliary_loss_mlp": 0.01035403,
      "balance_loss_clip": 1.03476405,
      "balance_loss_mlp": 1.02276635,
      "epoch": 0.7925447166691718,
      "flos": 23476960333440.0,
      "grad_norm": 1.779442516650618,
      "language_loss": 0.77660525,
      "learning_rate": 4.346213957372895e-07,
      "loss": 0.79771698,
      "num_input_tokens_seen": 284401585,
      "step": 13182,
      "time_per_iteration": 2.5456299781799316
    },
    {
      "auxiliary_loss_clip": 0.01099877,
      "auxiliary_loss_mlp": 0.01036827,
      "balance_loss_clip": 1.03801358,
      "balance_loss_mlp": 1.02331448,
      "epoch": 0.7926048399218397,
      "flos": 20447572118400.0,
      "grad_norm": 1.850407664163874,
      "language_loss": 0.73844552,
      "learning_rate": 4.34379019557056e-07,
      "loss": 0.75981259,
      "num_input_tokens_seen": 284419125,
      "step": 13183,
      "time_per_iteration": 2.4586498737335205
    },
    {
      "auxiliary_loss_clip": 0.01073653,
      "auxiliary_loss_mlp": 0.01026646,
      "balance_loss_clip": 1.03509855,
      "balance_loss_mlp": 1.01337183,
      "epoch": 0.7926649631745077,
      "flos": 37160038535040.0,
      "grad_norm": 1.8123674292201173,
      "language_loss": 0.68082094,
      "learning_rate": 4.341367027453264e-07,
      "loss": 0.70182383,
      "num_input_tokens_seen": 284440445,
      "step": 13184,
      "time_per_iteration": 2.6618990898132324
    },
    {
      "auxiliary_loss_clip": 0.01066278,
      "auxiliary_loss_mlp": 0.01033232,
      "balance_loss_clip": 1.03717566,
      "balance_loss_mlp": 1.02097154,
      "epoch": 0.7927250864271758,
      "flos": 17018606033280.0,
      "grad_norm": 1.7362120078361776,
      "language_loss": 0.71063006,
      "learning_rate": 4.338944453112907e-07,
      "loss": 0.7316252,
      "num_input_tokens_seen": 284459370,
      "step": 13185,
      "time_per_iteration": 2.537388563156128
    },
    {
      "auxiliary_loss_clip": 0.01090846,
      "auxiliary_loss_mlp": 0.01028257,
      "balance_loss_clip": 1.03662205,
      "balance_loss_mlp": 1.01569211,
      "epoch": 0.7927852096798437,
      "flos": 17749208666880.0,
      "grad_norm": 3.9185797132809785,
      "language_loss": 0.65336668,
      "learning_rate": 4.3365224726413375e-07,
      "loss": 0.67455769,
      "num_input_tokens_seen": 284477525,
      "step": 13186,
      "time_per_iteration": 2.489509344100952
    },
    {
      "auxiliary_loss_clip": 0.01087162,
      "auxiliary_loss_mlp": 0.0103267,
      "balance_loss_clip": 1.0338335,
      "balance_loss_mlp": 1.02105927,
      "epoch": 0.7928453329325117,
      "flos": 23838436851840.0,
      "grad_norm": 1.75431095837258,
      "language_loss": 0.76763535,
      "learning_rate": 4.334101086130408e-07,
      "loss": 0.78883362,
      "num_input_tokens_seen": 284496590,
      "step": 13187,
      "time_per_iteration": 2.5379977226257324
    },
    {
      "auxiliary_loss_clip": 0.01084322,
      "auxiliary_loss_mlp": 0.01026384,
      "balance_loss_clip": 1.03621733,
      "balance_loss_mlp": 1.01451683,
      "epoch": 0.7929054561851796,
      "flos": 17454920538240.0,
      "grad_norm": 1.921777012413162,
      "language_loss": 0.72297192,
      "learning_rate": 4.3316802936719334e-07,
      "loss": 0.74407899,
      "num_input_tokens_seen": 284511470,
      "step": 13188,
      "time_per_iteration": 2.497368812561035
    },
    {
      "auxiliary_loss_clip": 0.01105984,
      "auxiliary_loss_mlp": 0.00784344,
      "balance_loss_clip": 1.03521633,
      "balance_loss_mlp": 1.00993431,
      "epoch": 0.7929655794378476,
      "flos": 21981280988160.0,
      "grad_norm": 2.0516926783239966,
      "language_loss": 0.63006079,
      "learning_rate": 4.329260095357725e-07,
      "loss": 0.64896405,
      "num_input_tokens_seen": 284531125,
      "step": 13189,
      "time_per_iteration": 2.448115825653076
    },
    {
      "auxiliary_loss_clip": 0.0105789,
      "auxiliary_loss_mlp": 0.01029456,
      "balance_loss_clip": 1.03458059,
      "balance_loss_mlp": 1.01847649,
      "epoch": 0.7930257026905155,
      "flos": 17273930883840.0,
      "grad_norm": 1.8738646153742748,
      "language_loss": 0.72152179,
      "learning_rate": 4.3268404912795307e-07,
      "loss": 0.74239528,
      "num_input_tokens_seen": 284549340,
      "step": 13190,
      "time_per_iteration": 2.5836496353149414
    },
    {
      "auxiliary_loss_clip": 0.01091189,
      "auxiliary_loss_mlp": 0.01027661,
      "balance_loss_clip": 1.03757,
      "balance_loss_mlp": 1.01726007,
      "epoch": 0.7930858259431836,
      "flos": 27300584125440.0,
      "grad_norm": 1.7730795406922149,
      "language_loss": 0.73115295,
      "learning_rate": 4.3244214815291166e-07,
      "loss": 0.75234145,
      "num_input_tokens_seen": 284567060,
      "step": 13191,
      "time_per_iteration": 2.5335373878479004
    },
    {
      "auxiliary_loss_clip": 0.0109211,
      "auxiliary_loss_mlp": 0.01035421,
      "balance_loss_clip": 1.0351212,
      "balance_loss_mlp": 1.0226897,
      "epoch": 0.7931459491958515,
      "flos": 19863736456320.0,
      "grad_norm": 1.8408789420606098,
      "language_loss": 0.68683851,
      "learning_rate": 4.322003066198219e-07,
      "loss": 0.70811379,
      "num_input_tokens_seen": 284586600,
      "step": 13192,
      "time_per_iteration": 2.510612964630127
    },
    {
      "auxiliary_loss_clip": 0.01070936,
      "auxiliary_loss_mlp": 0.01037031,
      "balance_loss_clip": 1.03412068,
      "balance_loss_mlp": 1.02485394,
      "epoch": 0.7932060724485195,
      "flos": 23147120718720.0,
      "grad_norm": 1.5437493258656407,
      "language_loss": 0.74796194,
      "learning_rate": 4.3195852453785274e-07,
      "loss": 0.7690416,
      "num_input_tokens_seen": 284605715,
      "step": 13193,
      "time_per_iteration": 2.53916072845459
    },
    {
      "auxiliary_loss_clip": 0.01092643,
      "auxiliary_loss_mlp": 0.01030047,
      "balance_loss_clip": 1.03728032,
      "balance_loss_mlp": 1.01689816,
      "epoch": 0.7932661957011874,
      "flos": 29934847756800.0,
      "grad_norm": 1.4716361325591532,
      "language_loss": 0.71890461,
      "learning_rate": 4.317168019161741e-07,
      "loss": 0.7401315,
      "num_input_tokens_seen": 284628540,
      "step": 13194,
      "time_per_iteration": 2.5463688373565674
    },
    {
      "auxiliary_loss_clip": 0.01107877,
      "auxiliary_loss_mlp": 0.01033266,
      "balance_loss_clip": 1.03650522,
      "balance_loss_mlp": 1.02079034,
      "epoch": 0.7933263189538554,
      "flos": 22559119079040.0,
      "grad_norm": 1.9906596521576876,
      "language_loss": 0.7010386,
      "learning_rate": 4.314751387639517e-07,
      "loss": 0.72245002,
      "num_input_tokens_seen": 284646040,
      "step": 13195,
      "time_per_iteration": 2.433253288269043
    },
    {
      "auxiliary_loss_clip": 0.01053823,
      "auxiliary_loss_mlp": 0.01026054,
      "balance_loss_clip": 1.03632808,
      "balance_loss_mlp": 1.01435328,
      "epoch": 0.7933864422065233,
      "flos": 25479051575040.0,
      "grad_norm": 1.6788474722170903,
      "language_loss": 0.77445751,
      "learning_rate": 4.3123353509034844e-07,
      "loss": 0.79525626,
      "num_input_tokens_seen": 284665110,
      "step": 13196,
      "time_per_iteration": 2.6459922790527344
    },
    {
      "auxiliary_loss_clip": 0.01073242,
      "auxiliary_loss_mlp": 0.01037594,
      "balance_loss_clip": 1.03576493,
      "balance_loss_mlp": 1.02542233,
      "epoch": 0.7934465654591913,
      "flos": 33583156243200.0,
      "grad_norm": 1.5861764964000218,
      "language_loss": 0.68631828,
      "learning_rate": 4.309919909045268e-07,
      "loss": 0.70742667,
      "num_input_tokens_seen": 284686515,
      "step": 13197,
      "time_per_iteration": 2.6409270763397217
    },
    {
      "auxiliary_loss_clip": 0.01091843,
      "auxiliary_loss_mlp": 0.01028593,
      "balance_loss_clip": 1.03540528,
      "balance_loss_mlp": 1.01679754,
      "epoch": 0.7935066887118594,
      "flos": 31432538263680.0,
      "grad_norm": 1.6479064261612526,
      "language_loss": 0.64929438,
      "learning_rate": 4.30750506215646e-07,
      "loss": 0.67049873,
      "num_input_tokens_seen": 284707300,
      "step": 13198,
      "time_per_iteration": 2.5902647972106934
    },
    {
      "auxiliary_loss_clip": 0.01056793,
      "auxiliary_loss_mlp": 0.01040996,
      "balance_loss_clip": 1.03469157,
      "balance_loss_mlp": 1.02625012,
      "epoch": 0.7935668119645273,
      "flos": 14682616940160.0,
      "grad_norm": 1.9895710891339322,
      "language_loss": 0.72520709,
      "learning_rate": 4.30509081032864e-07,
      "loss": 0.74618506,
      "num_input_tokens_seen": 284723545,
      "step": 13199,
      "time_per_iteration": 2.532724380493164
    },
    {
      "auxiliary_loss_clip": 0.01077953,
      "auxiliary_loss_mlp": 0.01030496,
      "balance_loss_clip": 1.03364444,
      "balance_loss_mlp": 1.01874197,
      "epoch": 0.7936269352171953,
      "flos": 18004246208640.0,
      "grad_norm": 1.8672598215577085,
      "language_loss": 0.8063162,
      "learning_rate": 4.302677153653349e-07,
      "loss": 0.82740068,
      "num_input_tokens_seen": 284742650,
      "step": 13200,
      "time_per_iteration": 2.5364186763763428
    },
    {
      "auxiliary_loss_clip": 0.01091815,
      "auxiliary_loss_mlp": 0.01029356,
      "balance_loss_clip": 1.03686976,
      "balance_loss_mlp": 1.01788771,
      "epoch": 0.7936870584698632,
      "flos": 18880215183360.0,
      "grad_norm": 1.640679408302962,
      "language_loss": 0.77306491,
      "learning_rate": 4.3002640922221077e-07,
      "loss": 0.7942766,
      "num_input_tokens_seen": 284760955,
      "step": 13201,
      "time_per_iteration": 2.4687345027923584
    },
    {
      "auxiliary_loss_clip": 0.01103096,
      "auxiliary_loss_mlp": 0.01031569,
      "balance_loss_clip": 1.03541076,
      "balance_loss_mlp": 1.01979685,
      "epoch": 0.7937471817225312,
      "flos": 23367001824000.0,
      "grad_norm": 1.4717353306747516,
      "language_loss": 0.67109442,
      "learning_rate": 4.2978516261264296e-07,
      "loss": 0.69244105,
      "num_input_tokens_seen": 284780745,
      "step": 13202,
      "time_per_iteration": 2.548905372619629
    },
    {
      "auxiliary_loss_clip": 0.01094116,
      "auxiliary_loss_mlp": 0.01031666,
      "balance_loss_clip": 1.03601706,
      "balance_loss_mlp": 1.01892209,
      "epoch": 0.7938073049751991,
      "flos": 22674428714880.0,
      "grad_norm": 1.8146861707796742,
      "language_loss": 0.74739659,
      "learning_rate": 4.2954397554577884e-07,
      "loss": 0.76865447,
      "num_input_tokens_seen": 284799000,
      "step": 13203,
      "time_per_iteration": 2.4939615726470947
    },
    {
      "auxiliary_loss_clip": 0.0104977,
      "auxiliary_loss_mlp": 0.01030695,
      "balance_loss_clip": 1.03276849,
      "balance_loss_mlp": 1.01925111,
      "epoch": 0.7938674282278672,
      "flos": 22851431959680.0,
      "grad_norm": 1.806018109897258,
      "language_loss": 0.66436803,
      "learning_rate": 4.293028480307643e-07,
      "loss": 0.68517268,
      "num_input_tokens_seen": 284817450,
      "step": 13204,
      "time_per_iteration": 2.6724603176116943
    },
    {
      "auxiliary_loss_clip": 0.01046284,
      "auxiliary_loss_mlp": 0.0102724,
      "balance_loss_clip": 1.0355438,
      "balance_loss_mlp": 1.01584363,
      "epoch": 0.7939275514805351,
      "flos": 27012509049600.0,
      "grad_norm": 1.3111537033805436,
      "language_loss": 0.79516029,
      "learning_rate": 4.290617800767438e-07,
      "loss": 0.8158955,
      "num_input_tokens_seen": 284838865,
      "step": 13205,
      "time_per_iteration": 2.648261308670044
    },
    {
      "auxiliary_loss_clip": 0.01070285,
      "auxiliary_loss_mlp": 0.01030051,
      "balance_loss_clip": 1.03424215,
      "balance_loss_mlp": 1.01810575,
      "epoch": 0.7939876747332031,
      "flos": 21142838747520.0,
      "grad_norm": 1.9812185113480505,
      "language_loss": 0.7787407,
      "learning_rate": 4.28820771692858e-07,
      "loss": 0.79974407,
      "num_input_tokens_seen": 284857975,
      "step": 13206,
      "time_per_iteration": 2.5571513175964355
    },
    {
      "auxiliary_loss_clip": 0.01081585,
      "auxiliary_loss_mlp": 0.0103942,
      "balance_loss_clip": 1.03598583,
      "balance_loss_mlp": 1.02530515,
      "epoch": 0.794047797985871,
      "flos": 23289075267840.0,
      "grad_norm": 1.9458918681852955,
      "language_loss": 0.79419506,
      "learning_rate": 4.285798228882456e-07,
      "loss": 0.81540507,
      "num_input_tokens_seen": 284877145,
      "step": 13207,
      "time_per_iteration": 2.553318738937378
    },
    {
      "auxiliary_loss_clip": 0.01068627,
      "auxiliary_loss_mlp": 0.01031127,
      "balance_loss_clip": 1.03629017,
      "balance_loss_mlp": 1.01931345,
      "epoch": 0.794107921238539,
      "flos": 24608074590720.0,
      "grad_norm": 1.771933747673076,
      "language_loss": 0.83872581,
      "learning_rate": 4.2833893367204375e-07,
      "loss": 0.85972333,
      "num_input_tokens_seen": 284895560,
      "step": 13208,
      "time_per_iteration": 2.597020387649536
    },
    {
      "auxiliary_loss_clip": 0.00994371,
      "auxiliary_loss_mlp": 0.01004227,
      "balance_loss_clip": 1.00969017,
      "balance_loss_mlp": 1.00292766,
      "epoch": 0.7941680444912069,
      "flos": 64093690252800.0,
      "grad_norm": 0.7163224593324753,
      "language_loss": 0.58325076,
      "learning_rate": 4.280981040533875e-07,
      "loss": 0.60323679,
      "num_input_tokens_seen": 284963135,
      "step": 13209,
      "time_per_iteration": 4.748367547988892
    },
    {
      "auxiliary_loss_clip": 0.01070072,
      "auxiliary_loss_mlp": 0.0103147,
      "balance_loss_clip": 1.03370738,
      "balance_loss_mlp": 1.01837456,
      "epoch": 0.794228167743875,
      "flos": 24388839930240.0,
      "grad_norm": 2.0830480249151884,
      "language_loss": 0.62873131,
      "learning_rate": 4.2785733404140825e-07,
      "loss": 0.64974672,
      "num_input_tokens_seen": 284981755,
      "step": 13210,
      "time_per_iteration": 2.571667194366455
    },
    {
      "auxiliary_loss_clip": 0.01088616,
      "auxiliary_loss_mlp": 0.01032132,
      "balance_loss_clip": 1.03571844,
      "balance_loss_mlp": 1.02065253,
      "epoch": 0.794288290996543,
      "flos": 28512498026880.0,
      "grad_norm": 1.5655829167959905,
      "language_loss": 0.69090295,
      "learning_rate": 4.2761662364523676e-07,
      "loss": 0.7121104,
      "num_input_tokens_seen": 285003060,
      "step": 13211,
      "time_per_iteration": 3.9563143253326416
    },
    {
      "auxiliary_loss_clip": 0.01096623,
      "auxiliary_loss_mlp": 0.01034004,
      "balance_loss_clip": 1.03629756,
      "balance_loss_mlp": 1.02164829,
      "epoch": 0.7943484142492109,
      "flos": 25922117836800.0,
      "grad_norm": 1.7288808238498627,
      "language_loss": 0.72254169,
      "learning_rate": 4.2737597287400074e-07,
      "loss": 0.74384803,
      "num_input_tokens_seen": 285021640,
      "step": 13212,
      "time_per_iteration": 3.8920576572418213
    },
    {
      "auxiliary_loss_clip": 0.01091172,
      "auxiliary_loss_mlp": 0.01027743,
      "balance_loss_clip": 1.03522134,
      "balance_loss_mlp": 1.01619196,
      "epoch": 0.7944085375018789,
      "flos": 23915286000000.0,
      "grad_norm": 1.5978588177964779,
      "language_loss": 0.81014895,
      "learning_rate": 4.271353817368246e-07,
      "loss": 0.83133805,
      "num_input_tokens_seen": 285040490,
      "step": 13213,
      "time_per_iteration": 2.480397939682007
    },
    {
      "auxiliary_loss_clip": 0.01098057,
      "auxiliary_loss_mlp": 0.0103006,
      "balance_loss_clip": 1.03718746,
      "balance_loss_mlp": 1.01766229,
      "epoch": 0.7944686607545468,
      "flos": 20229953569920.0,
      "grad_norm": 2.090563871805547,
      "language_loss": 0.68053931,
      "learning_rate": 4.268948502428327e-07,
      "loss": 0.70182049,
      "num_input_tokens_seen": 285059270,
      "step": 13214,
      "time_per_iteration": 2.490478754043579
    },
    {
      "auxiliary_loss_clip": 0.01102318,
      "auxiliary_loss_mlp": 0.0102845,
      "balance_loss_clip": 1.03578985,
      "balance_loss_mlp": 1.01715505,
      "epoch": 0.7945287840072148,
      "flos": 21980993679360.0,
      "grad_norm": 2.14047709542354,
      "language_loss": 0.72620785,
      "learning_rate": 4.2665437840114535e-07,
      "loss": 0.74751556,
      "num_input_tokens_seen": 285075390,
      "step": 13215,
      "time_per_iteration": 2.431408643722534
    },
    {
      "auxiliary_loss_clip": 0.01053908,
      "auxiliary_loss_mlp": 0.01034056,
      "balance_loss_clip": 1.03493834,
      "balance_loss_mlp": 1.02115154,
      "epoch": 0.7945889072598827,
      "flos": 26397718842240.0,
      "grad_norm": 1.5629648225577593,
      "language_loss": 0.78813583,
      "learning_rate": 4.2641396622088253e-07,
      "loss": 0.80901545,
      "num_input_tokens_seen": 285096290,
      "step": 13216,
      "time_per_iteration": 2.62217378616333
    },
    {
      "auxiliary_loss_clip": 0.01091641,
      "auxiliary_loss_mlp": 0.01032388,
      "balance_loss_clip": 1.03552306,
      "balance_loss_mlp": 1.0201211,
      "epoch": 0.7946490305125508,
      "flos": 25810255906560.0,
      "grad_norm": 1.584101574006816,
      "language_loss": 0.73852062,
      "learning_rate": 4.261736137111598e-07,
      "loss": 0.75976086,
      "num_input_tokens_seen": 285116020,
      "step": 13217,
      "time_per_iteration": 2.5042057037353516
    },
    {
      "auxiliary_loss_clip": 0.01075964,
      "auxiliary_loss_mlp": 0.01032857,
      "balance_loss_clip": 1.03357637,
      "balance_loss_mlp": 1.02032804,
      "epoch": 0.7947091537652187,
      "flos": 15960965045760.0,
      "grad_norm": 1.8088663320061202,
      "language_loss": 0.74436778,
      "learning_rate": 4.259333208810907e-07,
      "loss": 0.76545602,
      "num_input_tokens_seen": 285133510,
      "step": 13218,
      "time_per_iteration": 2.5118370056152344
    },
    {
      "auxiliary_loss_clip": 0.01095708,
      "auxiliary_loss_mlp": 0.01035586,
      "balance_loss_clip": 1.03554535,
      "balance_loss_mlp": 1.02240777,
      "epoch": 0.7947692770178867,
      "flos": 18587866389120.0,
      "grad_norm": 1.9210050251752622,
      "language_loss": 0.83219767,
      "learning_rate": 4.2569308773978817e-07,
      "loss": 0.85351062,
      "num_input_tokens_seen": 285151690,
      "step": 13219,
      "time_per_iteration": 3.8837080001831055
    },
    {
      "auxiliary_loss_clip": 0.01094231,
      "auxiliary_loss_mlp": 0.01040385,
      "balance_loss_clip": 1.03591704,
      "balance_loss_mlp": 1.02568626,
      "epoch": 0.7948294002705546,
      "flos": 20442220992000.0,
      "grad_norm": 3.27541271630113,
      "language_loss": 0.75475824,
      "learning_rate": 4.2545291429636123e-07,
      "loss": 0.77610445,
      "num_input_tokens_seen": 285170485,
      "step": 13220,
      "time_per_iteration": 2.5024161338806152
    },
    {
      "auxiliary_loss_clip": 0.01088128,
      "auxiliary_loss_mlp": 0.01032374,
      "balance_loss_clip": 1.03603351,
      "balance_loss_mlp": 1.01991653,
      "epoch": 0.7948895235232226,
      "flos": 38181194282880.0,
      "grad_norm": 1.710344383739916,
      "language_loss": 0.72549617,
      "learning_rate": 4.252128005599176e-07,
      "loss": 0.74670118,
      "num_input_tokens_seen": 285191050,
      "step": 13221,
      "time_per_iteration": 2.669774055480957
    },
    {
      "auxiliary_loss_clip": 0.01093125,
      "auxiliary_loss_mlp": 0.01025907,
      "balance_loss_clip": 1.03645897,
      "balance_loss_mlp": 1.01458788,
      "epoch": 0.7949496467758905,
      "flos": 15559806977280.0,
      "grad_norm": 2.1230476212234985,
      "language_loss": 0.74712038,
      "learning_rate": 4.249727465395634e-07,
      "loss": 0.76831067,
      "num_input_tokens_seen": 285208750,
      "step": 13222,
      "time_per_iteration": 2.444411516189575
    },
    {
      "auxiliary_loss_clip": 0.01011684,
      "auxiliary_loss_mlp": 0.01001703,
      "balance_loss_clip": 1.00916624,
      "balance_loss_mlp": 1.00058198,
      "epoch": 0.7950097700285585,
      "flos": 70897036728960.0,
      "grad_norm": 0.7666821029573112,
      "language_loss": 0.67073858,
      "learning_rate": 4.247327522443993e-07,
      "loss": 0.69087243,
      "num_input_tokens_seen": 285264605,
      "step": 13223,
      "time_per_iteration": 3.0004827976226807
    },
    {
      "auxiliary_loss_clip": 0.01092703,
      "auxiliary_loss_mlp": 0.0102909,
      "balance_loss_clip": 1.03508735,
      "balance_loss_mlp": 1.01639402,
      "epoch": 0.7950698932812266,
      "flos": 23951627585280.0,
      "grad_norm": 1.6378725692833767,
      "language_loss": 0.71279943,
      "learning_rate": 4.2449281768352717e-07,
      "loss": 0.73401731,
      "num_input_tokens_seen": 285283940,
      "step": 13224,
      "time_per_iteration": 2.4958302974700928
    },
    {
      "auxiliary_loss_clip": 0.01031663,
      "auxiliary_loss_mlp": 0.01003705,
      "balance_loss_clip": 1.00885916,
      "balance_loss_mlp": 1.00239396,
      "epoch": 0.7951300165338945,
      "flos": 60282561415680.0,
      "grad_norm": 0.6676627337500398,
      "language_loss": 0.55005997,
      "learning_rate": 4.2425294286604527e-07,
      "loss": 0.57041365,
      "num_input_tokens_seen": 285349525,
      "step": 13225,
      "time_per_iteration": 3.135474443435669
    },
    {
      "auxiliary_loss_clip": 0.01078766,
      "auxiliary_loss_mlp": 0.01021389,
      "balance_loss_clip": 1.03261662,
      "balance_loss_mlp": 1.01011777,
      "epoch": 0.7951901397865625,
      "flos": 22819004956800.0,
      "grad_norm": 1.990812863730598,
      "language_loss": 0.65203363,
      "learning_rate": 4.2401312780105034e-07,
      "loss": 0.67303514,
      "num_input_tokens_seen": 285367355,
      "step": 13226,
      "time_per_iteration": 2.505829334259033
    },
    {
      "auxiliary_loss_clip": 0.01060602,
      "auxiliary_loss_mlp": 0.01035904,
      "balance_loss_clip": 1.03619623,
      "balance_loss_mlp": 1.02419734,
      "epoch": 0.7952502630392304,
      "flos": 35695672871040.0,
      "grad_norm": 2.10313229420545,
      "language_loss": 0.70109773,
      "learning_rate": 4.237733724976349e-07,
      "loss": 0.72206283,
      "num_input_tokens_seen": 285386190,
      "step": 13227,
      "time_per_iteration": 2.7088866233825684
    },
    {
      "auxiliary_loss_clip": 0.01068614,
      "auxiliary_loss_mlp": 0.01028778,
      "balance_loss_clip": 1.03524375,
      "balance_loss_mlp": 1.01766777,
      "epoch": 0.7953103862918984,
      "flos": 25629840869760.0,
      "grad_norm": 1.6615762884527814,
      "language_loss": 0.6892125,
      "learning_rate": 4.2353367696489184e-07,
      "loss": 0.71018636,
      "num_input_tokens_seen": 285406150,
      "step": 13228,
      "time_per_iteration": 2.6202423572540283
    },
    {
      "auxiliary_loss_clip": 0.01057116,
      "auxiliary_loss_mlp": 0.01038555,
      "balance_loss_clip": 1.03179586,
      "balance_loss_mlp": 1.02570426,
      "epoch": 0.7953705095445663,
      "flos": 40551980676480.0,
      "grad_norm": 1.4161574405640922,
      "language_loss": 0.7077204,
      "learning_rate": 4.232940412119095e-07,
      "loss": 0.72867709,
      "num_input_tokens_seen": 285429900,
      "step": 13229,
      "time_per_iteration": 2.742006778717041
    },
    {
      "auxiliary_loss_clip": 0.01099299,
      "auxiliary_loss_mlp": 0.01032909,
      "balance_loss_clip": 1.03892946,
      "balance_loss_mlp": 1.02045798,
      "epoch": 0.7954306327972344,
      "flos": 27636672706560.0,
      "grad_norm": 1.6952991344465926,
      "language_loss": 0.71915197,
      "learning_rate": 4.2305446524777457e-07,
      "loss": 0.7404741,
      "num_input_tokens_seen": 285452555,
      "step": 13230,
      "time_per_iteration": 2.57804799079895
    },
    {
      "auxiliary_loss_clip": 0.01013509,
      "auxiliary_loss_mlp": 0.01000038,
      "balance_loss_clip": 1.01167798,
      "balance_loss_mlp": 0.99892348,
      "epoch": 0.7954907560499023,
      "flos": 59504055995520.0,
      "grad_norm": 0.8919757519885787,
      "language_loss": 0.63613057,
      "learning_rate": 4.2281494908157247e-07,
      "loss": 0.65626609,
      "num_input_tokens_seen": 285515700,
      "step": 13231,
      "time_per_iteration": 3.1293840408325195
    },
    {
      "auxiliary_loss_clip": 0.01081326,
      "auxiliary_loss_mlp": 0.01026459,
      "balance_loss_clip": 1.03424132,
      "balance_loss_mlp": 1.01477087,
      "epoch": 0.7955508793025703,
      "flos": 20120533764480.0,
      "grad_norm": 1.7060056902447116,
      "language_loss": 0.69970542,
      "learning_rate": 4.2257549272238566e-07,
      "loss": 0.72078335,
      "num_input_tokens_seen": 285533910,
      "step": 13232,
      "time_per_iteration": 2.521815299987793
    },
    {
      "auxiliary_loss_clip": 0.01093235,
      "auxiliary_loss_mlp": 0.01027017,
      "balance_loss_clip": 1.03515947,
      "balance_loss_mlp": 1.01507258,
      "epoch": 0.7956110025552382,
      "flos": 26505378881280.0,
      "grad_norm": 2.9374979561657177,
      "language_loss": 0.77852917,
      "learning_rate": 4.223360961792952e-07,
      "loss": 0.79973167,
      "num_input_tokens_seen": 285554080,
      "step": 13233,
      "time_per_iteration": 2.52329421043396
    },
    {
      "auxiliary_loss_clip": 0.01094564,
      "auxiliary_loss_mlp": 0.01032111,
      "balance_loss_clip": 1.03553724,
      "balance_loss_mlp": 1.02011251,
      "epoch": 0.7956711258079062,
      "flos": 22565475786240.0,
      "grad_norm": 1.92927225006163,
      "language_loss": 0.79039514,
      "learning_rate": 4.220967594613769e-07,
      "loss": 0.81166184,
      "num_input_tokens_seen": 285572325,
      "step": 13234,
      "time_per_iteration": 2.501094102859497
    },
    {
      "auxiliary_loss_clip": 0.01082979,
      "auxiliary_loss_mlp": 0.00780932,
      "balance_loss_clip": 1.03599429,
      "balance_loss_mlp": 1.00659692,
      "epoch": 0.7957312490605741,
      "flos": 17379005143680.0,
      "grad_norm": 1.5899201878544813,
      "language_loss": 0.70076466,
      "learning_rate": 4.218574825777077e-07,
      "loss": 0.7194038,
      "num_input_tokens_seen": 285589770,
      "step": 13235,
      "time_per_iteration": 2.4995243549346924
    },
    {
      "auxiliary_loss_clip": 0.01066829,
      "auxiliary_loss_mlp": 0.01027448,
      "balance_loss_clip": 1.03611755,
      "balance_loss_mlp": 1.01537848,
      "epoch": 0.7957913723132422,
      "flos": 22491427898880.0,
      "grad_norm": 1.578202652555452,
      "language_loss": 0.6809094,
      "learning_rate": 4.2161826553736145e-07,
      "loss": 0.7018522,
      "num_input_tokens_seen": 285610065,
      "step": 13236,
      "time_per_iteration": 2.5930373668670654
    },
    {
      "auxiliary_loss_clip": 0.01055245,
      "auxiliary_loss_mlp": 0.01025982,
      "balance_loss_clip": 1.0334146,
      "balance_loss_mlp": 1.01381111,
      "epoch": 0.7958514955659101,
      "flos": 22638087129600.0,
      "grad_norm": 1.7392182004996524,
      "language_loss": 0.75212389,
      "learning_rate": 4.2137910834940826e-07,
      "loss": 0.77293617,
      "num_input_tokens_seen": 285628480,
      "step": 13237,
      "time_per_iteration": 2.576793670654297
    },
    {
      "auxiliary_loss_clip": 0.01094473,
      "auxiliary_loss_mlp": 0.01034664,
      "balance_loss_clip": 1.03648806,
      "balance_loss_mlp": 1.02224243,
      "epoch": 0.7959116188185781,
      "flos": 20704225772160.0,
      "grad_norm": 3.2927590053638895,
      "language_loss": 0.71859026,
      "learning_rate": 4.211400110229175e-07,
      "loss": 0.73988163,
      "num_input_tokens_seen": 285647805,
      "step": 13238,
      "time_per_iteration": 2.5159196853637695
    },
    {
      "auxiliary_loss_clip": 0.01089385,
      "auxiliary_loss_mlp": 0.01026902,
      "balance_loss_clip": 1.03478003,
      "balance_loss_mlp": 1.01488566,
      "epoch": 0.7959717420712461,
      "flos": 19024683684480.0,
      "grad_norm": 1.7024174191986974,
      "language_loss": 0.7349658,
      "learning_rate": 4.2090097356695684e-07,
      "loss": 0.75612867,
      "num_input_tokens_seen": 285665505,
      "step": 13239,
      "time_per_iteration": 2.463737726211548
    },
    {
      "auxiliary_loss_clip": 0.01107563,
      "auxiliary_loss_mlp": 0.01032929,
      "balance_loss_clip": 1.03641343,
      "balance_loss_mlp": 1.02084136,
      "epoch": 0.796031865323914,
      "flos": 26356636661760.0,
      "grad_norm": 1.8389488216304122,
      "language_loss": 0.69676816,
      "learning_rate": 4.2066199599058814e-07,
      "loss": 0.71817309,
      "num_input_tokens_seen": 285685855,
      "step": 13240,
      "time_per_iteration": 2.5436160564422607
    },
    {
      "auxiliary_loss_clip": 0.01024932,
      "auxiliary_loss_mlp": 0.01002265,
      "balance_loss_clip": 1.01187158,
      "balance_loss_mlp": 1.00110841,
      "epoch": 0.796091988576582,
      "flos": 62069440320000.0,
      "grad_norm": 0.8894929062282761,
      "language_loss": 0.58694494,
      "learning_rate": 4.2042307830287526e-07,
      "loss": 0.60721689,
      "num_input_tokens_seen": 285735710,
      "step": 13241,
      "time_per_iteration": 2.882469654083252
    },
    {
      "auxiliary_loss_clip": 0.01074145,
      "auxiliary_loss_mlp": 0.01030897,
      "balance_loss_clip": 1.03696966,
      "balance_loss_mlp": 1.01965594,
      "epoch": 0.7961521118292499,
      "flos": 39020103400320.0,
      "grad_norm": 2.4998962379746863,
      "language_loss": 0.64260161,
      "learning_rate": 4.201842205128772e-07,
      "loss": 0.663652,
      "num_input_tokens_seen": 285757045,
      "step": 13242,
      "time_per_iteration": 2.6980950832366943
    },
    {
      "auxiliary_loss_clip": 0.01104646,
      "auxiliary_loss_mlp": 0.01034399,
      "balance_loss_clip": 1.03546596,
      "balance_loss_mlp": 1.02191782,
      "epoch": 0.796212235081918,
      "flos": 21762836426880.0,
      "grad_norm": 1.9391778549737408,
      "language_loss": 0.75975037,
      "learning_rate": 4.199454226296526e-07,
      "loss": 0.78114086,
      "num_input_tokens_seen": 285776050,
      "step": 13243,
      "time_per_iteration": 2.4617419242858887
    },
    {
      "auxiliary_loss_clip": 0.01077169,
      "auxiliary_loss_mlp": 0.01027813,
      "balance_loss_clip": 1.03622675,
      "balance_loss_mlp": 1.01537967,
      "epoch": 0.7962723583345859,
      "flos": 21178857110400.0,
      "grad_norm": 1.6054770026277123,
      "language_loss": 0.7939235,
      "learning_rate": 4.1970668466225565e-07,
      "loss": 0.81497335,
      "num_input_tokens_seen": 285796830,
      "step": 13244,
      "time_per_iteration": 2.597487211227417
    },
    {
      "auxiliary_loss_clip": 0.0109673,
      "auxiliary_loss_mlp": 0.01028699,
      "balance_loss_clip": 1.03562558,
      "balance_loss_mlp": 1.01596761,
      "epoch": 0.7963324815872539,
      "flos": 17128636369920.0,
      "grad_norm": 2.0544943145311882,
      "language_loss": 0.68209231,
      "learning_rate": 4.1946800661973934e-07,
      "loss": 0.70334661,
      "num_input_tokens_seen": 285814755,
      "step": 13245,
      "time_per_iteration": 2.4767088890075684
    },
    {
      "auxiliary_loss_clip": 0.01084122,
      "auxiliary_loss_mlp": 0.01032389,
      "balance_loss_clip": 1.03559971,
      "balance_loss_mlp": 1.02006912,
      "epoch": 0.7963926048399218,
      "flos": 21397481239680.0,
      "grad_norm": 1.4688082024812112,
      "language_loss": 0.79095185,
      "learning_rate": 4.192293885111549e-07,
      "loss": 0.81211698,
      "num_input_tokens_seen": 285834255,
      "step": 13246,
      "time_per_iteration": 2.536513328552246
    },
    {
      "auxiliary_loss_clip": 0.01084447,
      "auxiliary_loss_mlp": 0.01027008,
      "balance_loss_clip": 1.03446627,
      "balance_loss_mlp": 1.01505744,
      "epoch": 0.7964527280925898,
      "flos": 25184188828800.0,
      "grad_norm": 1.6704279361428402,
      "language_loss": 0.65969992,
      "learning_rate": 4.1899083034555007e-07,
      "loss": 0.6808145,
      "num_input_tokens_seen": 285853540,
      "step": 13247,
      "time_per_iteration": 2.5416057109832764
    },
    {
      "auxiliary_loss_clip": 0.01077923,
      "auxiliary_loss_mlp": 0.01027265,
      "balance_loss_clip": 1.03308988,
      "balance_loss_mlp": 1.01664352,
      "epoch": 0.7965128513452577,
      "flos": 27015884928000.0,
      "grad_norm": 1.9137652874781979,
      "language_loss": 0.71459162,
      "learning_rate": 4.1875233213197123e-07,
      "loss": 0.73564357,
      "num_input_tokens_seen": 285872705,
      "step": 13248,
      "time_per_iteration": 5.3422019481658936
    },
    {
      "auxiliary_loss_clip": 0.01086275,
      "auxiliary_loss_mlp": 0.01028068,
      "balance_loss_clip": 1.03481734,
      "balance_loss_mlp": 1.01577759,
      "epoch": 0.7965729745979258,
      "flos": 24419578993920.0,
      "grad_norm": 1.9362713061569736,
      "language_loss": 0.75956547,
      "learning_rate": 4.1851389387946255e-07,
      "loss": 0.78070891,
      "num_input_tokens_seen": 285890290,
      "step": 13249,
      "time_per_iteration": 2.548143148422241
    },
    {
      "auxiliary_loss_clip": 0.01076376,
      "auxiliary_loss_mlp": 0.01032786,
      "balance_loss_clip": 1.03576922,
      "balance_loss_mlp": 1.02138352,
      "epoch": 0.7966330978505937,
      "flos": 18840389978880.0,
      "grad_norm": 2.0231233864144733,
      "language_loss": 0.62236667,
      "learning_rate": 4.1827551559706674e-07,
      "loss": 0.64345825,
      "num_input_tokens_seen": 285909190,
      "step": 13250,
      "time_per_iteration": 2.4835221767425537
    },
    {
      "auxiliary_loss_clip": 0.01083085,
      "auxiliary_loss_mlp": 0.01024134,
      "balance_loss_clip": 1.0347507,
      "balance_loss_mlp": 1.01177835,
      "epoch": 0.7966932211032617,
      "flos": 13152319862400.0,
      "grad_norm": 2.44142669807278,
      "language_loss": 0.71697384,
      "learning_rate": 4.180371972938206e-07,
      "loss": 0.73804605,
      "num_input_tokens_seen": 285927570,
      "step": 13251,
      "time_per_iteration": 3.9977009296417236
    },
    {
      "auxiliary_loss_clip": 0.01108781,
      "auxiliary_loss_mlp": 0.0103249,
      "balance_loss_clip": 1.03724337,
      "balance_loss_mlp": 1.01948452,
      "epoch": 0.7967533443559297,
      "flos": 23949760078080.0,
      "grad_norm": 2.9866308401023502,
      "language_loss": 0.7325269,
      "learning_rate": 4.177989389787624e-07,
      "loss": 0.75393957,
      "num_input_tokens_seen": 285945810,
      "step": 13252,
      "time_per_iteration": 2.4762418270111084
    },
    {
      "auxiliary_loss_clip": 0.01101281,
      "auxiliary_loss_mlp": 0.01030484,
      "balance_loss_clip": 1.03517151,
      "balance_loss_mlp": 1.01860452,
      "epoch": 0.7968134676085976,
      "flos": 30368791964160.0,
      "grad_norm": 2.0005914239079665,
      "language_loss": 0.66025251,
      "learning_rate": 4.175607406609278e-07,
      "loss": 0.68157017,
      "num_input_tokens_seen": 285964235,
      "step": 13253,
      "time_per_iteration": 2.5017457008361816
    },
    {
      "auxiliary_loss_clip": 0.01079632,
      "auxiliary_loss_mlp": 0.01035236,
      "balance_loss_clip": 1.03945887,
      "balance_loss_mlp": 1.02257013,
      "epoch": 0.7968735908612656,
      "flos": 23075048079360.0,
      "grad_norm": 1.4361955491350193,
      "language_loss": 0.68002558,
      "learning_rate": 4.1732260234934767e-07,
      "loss": 0.7011742,
      "num_input_tokens_seen": 285983710,
      "step": 13254,
      "time_per_iteration": 2.5740509033203125
    },
    {
      "auxiliary_loss_clip": 0.01092712,
      "auxiliary_loss_mlp": 0.01033665,
      "balance_loss_clip": 1.03440809,
      "balance_loss_mlp": 1.0220598,
      "epoch": 0.7969337141139335,
      "flos": 23582250074880.0,
      "grad_norm": 1.7871829556532768,
      "language_loss": 0.69200301,
      "learning_rate": 4.1708452405305314e-07,
      "loss": 0.71326685,
      "num_input_tokens_seen": 286003425,
      "step": 13255,
      "time_per_iteration": 2.4927141666412354
    },
    {
      "auxiliary_loss_clip": 0.01101539,
      "auxiliary_loss_mlp": 0.01028429,
      "balance_loss_clip": 1.03407741,
      "balance_loss_mlp": 1.01690722,
      "epoch": 0.7969938373666016,
      "flos": 19755860935680.0,
      "grad_norm": 1.8589655606821995,
      "language_loss": 0.78862518,
      "learning_rate": 4.168465057810733e-07,
      "loss": 0.80992496,
      "num_input_tokens_seen": 286020130,
      "step": 13256,
      "time_per_iteration": 2.4588615894317627
    },
    {
      "auxiliary_loss_clip": 0.01093001,
      "auxiliary_loss_mlp": 0.01027585,
      "balance_loss_clip": 1.03631926,
      "balance_loss_mlp": 1.01544392,
      "epoch": 0.7970539606192695,
      "flos": 24134089697280.0,
      "grad_norm": 1.825487247715329,
      "language_loss": 0.65966737,
      "learning_rate": 4.166085475424315e-07,
      "loss": 0.68087316,
      "num_input_tokens_seen": 286040230,
      "step": 13257,
      "time_per_iteration": 2.486119031906128
    },
    {
      "auxiliary_loss_clip": 0.01085986,
      "auxiliary_loss_mlp": 0.01032099,
      "balance_loss_clip": 1.03579354,
      "balance_loss_mlp": 1.02010036,
      "epoch": 0.7971140838719375,
      "flos": 17968622895360.0,
      "grad_norm": 1.6811890693598976,
      "language_loss": 0.72089088,
      "learning_rate": 4.163706493461523e-07,
      "loss": 0.74207175,
      "num_input_tokens_seen": 286059475,
      "step": 13258,
      "time_per_iteration": 3.8730008602142334
    },
    {
      "auxiliary_loss_clip": 0.01096755,
      "auxiliary_loss_mlp": 0.01031452,
      "balance_loss_clip": 1.03625035,
      "balance_loss_mlp": 1.0189414,
      "epoch": 0.7971742071246054,
      "flos": 19169547235200.0,
      "grad_norm": 1.6644352046972792,
      "language_loss": 0.69202065,
      "learning_rate": 4.1613281120125655e-07,
      "loss": 0.71330273,
      "num_input_tokens_seen": 286077820,
      "step": 13259,
      "time_per_iteration": 2.4964866638183594
    },
    {
      "auxiliary_loss_clip": 0.01090813,
      "auxiliary_loss_mlp": 0.0102897,
      "balance_loss_clip": 1.03520584,
      "balance_loss_mlp": 1.01792526,
      "epoch": 0.7972343303772734,
      "flos": 27125951178240.0,
      "grad_norm": 2.6200728821374897,
      "language_loss": 0.73817325,
      "learning_rate": 4.158950331167641e-07,
      "loss": 0.7593711,
      "num_input_tokens_seen": 286097285,
      "step": 13260,
      "time_per_iteration": 2.5178604125976562
    },
    {
      "auxiliary_loss_clip": 0.01078427,
      "auxiliary_loss_mlp": 0.01028042,
      "balance_loss_clip": 1.03261471,
      "balance_loss_mlp": 1.0165565,
      "epoch": 0.7972944536299413,
      "flos": 20996646393600.0,
      "grad_norm": 1.702181950932197,
      "language_loss": 0.78287929,
      "learning_rate": 4.1565731510169065e-07,
      "loss": 0.80394393,
      "num_input_tokens_seen": 286116000,
      "step": 13261,
      "time_per_iteration": 2.5473992824554443
    },
    {
      "auxiliary_loss_clip": 0.01089867,
      "auxiliary_loss_mlp": 0.01027049,
      "balance_loss_clip": 1.03540909,
      "balance_loss_mlp": 1.01646376,
      "epoch": 0.7973545768826094,
      "flos": 21580015178880.0,
      "grad_norm": 1.6181045657758593,
      "language_loss": 0.76223749,
      "learning_rate": 4.154196571650501e-07,
      "loss": 0.78340662,
      "num_input_tokens_seen": 286135110,
      "step": 13262,
      "time_per_iteration": 2.48492693901062
    },
    {
      "auxiliary_loss_clip": 0.01075886,
      "auxiliary_loss_mlp": 0.01030586,
      "balance_loss_clip": 1.03774381,
      "balance_loss_mlp": 1.01699042,
      "epoch": 0.7974147001352773,
      "flos": 20558536208640.0,
      "grad_norm": 2.3572436446602576,
      "language_loss": 0.70475858,
      "learning_rate": 4.1518205931585524e-07,
      "loss": 0.72582334,
      "num_input_tokens_seen": 286152835,
      "step": 13263,
      "time_per_iteration": 2.5712146759033203
    },
    {
      "auxiliary_loss_clip": 0.01098736,
      "auxiliary_loss_mlp": 0.0103172,
      "balance_loss_clip": 1.03561234,
      "balance_loss_mlp": 1.01851797,
      "epoch": 0.7974748233879453,
      "flos": 20996790048000.0,
      "grad_norm": 1.621283840775451,
      "language_loss": 0.70957935,
      "learning_rate": 4.149445215631153e-07,
      "loss": 0.73088384,
      "num_input_tokens_seen": 286171785,
      "step": 13264,
      "time_per_iteration": 2.4827563762664795
    },
    {
      "auxiliary_loss_clip": 0.01101903,
      "auxiliary_loss_mlp": 0.01031176,
      "balance_loss_clip": 1.03533113,
      "balance_loss_mlp": 1.0194217,
      "epoch": 0.7975349466406133,
      "flos": 22565188477440.0,
      "grad_norm": 1.9036258143621274,
      "language_loss": 0.76808614,
      "learning_rate": 4.1470704391583776e-07,
      "loss": 0.78941697,
      "num_input_tokens_seen": 286190420,
      "step": 13265,
      "time_per_iteration": 2.4708216190338135
    },
    {
      "auxiliary_loss_clip": 0.01072534,
      "auxiliary_loss_mlp": 0.01028343,
      "balance_loss_clip": 1.0350951,
      "balance_loss_mlp": 1.01668477,
      "epoch": 0.7975950698932812,
      "flos": 21689542725120.0,
      "grad_norm": 1.7953271891447327,
      "language_loss": 0.75471693,
      "learning_rate": 4.144696263830285e-07,
      "loss": 0.77572572,
      "num_input_tokens_seen": 286210105,
      "step": 13266,
      "time_per_iteration": 2.5436112880706787
    },
    {
      "auxiliary_loss_clip": 0.0107806,
      "auxiliary_loss_mlp": 0.01025556,
      "balance_loss_clip": 1.03577304,
      "balance_loss_mlp": 1.01386189,
      "epoch": 0.7976551931459492,
      "flos": 19604568850560.0,
      "grad_norm": 1.5762933137923194,
      "language_loss": 0.84064019,
      "learning_rate": 4.1423226897369015e-07,
      "loss": 0.86167628,
      "num_input_tokens_seen": 286228180,
      "step": 13267,
      "time_per_iteration": 2.526582956314087
    },
    {
      "auxiliary_loss_clip": 0.01092371,
      "auxiliary_loss_mlp": 0.01031043,
      "balance_loss_clip": 1.03542495,
      "balance_loss_mlp": 1.01865745,
      "epoch": 0.7977153163986171,
      "flos": 21687603390720.0,
      "grad_norm": 1.6917749328555682,
      "language_loss": 0.7601707,
      "learning_rate": 4.139949716968223e-07,
      "loss": 0.78140485,
      "num_input_tokens_seen": 286247305,
      "step": 13268,
      "time_per_iteration": 2.4846396446228027
    },
    {
      "auxiliary_loss_clip": 0.0110359,
      "auxiliary_loss_mlp": 0.01028898,
      "balance_loss_clip": 1.03541636,
      "balance_loss_mlp": 1.01642859,
      "epoch": 0.7977754396512852,
      "flos": 23476780765440.0,
      "grad_norm": 1.5600846453866883,
      "language_loss": 0.77902615,
      "learning_rate": 4.1375773456142403e-07,
      "loss": 0.80035108,
      "num_input_tokens_seen": 286268145,
      "step": 13269,
      "time_per_iteration": 2.4935390949249268
    },
    {
      "auxiliary_loss_clip": 0.01085373,
      "auxiliary_loss_mlp": 0.01036293,
      "balance_loss_clip": 1.03227603,
      "balance_loss_mlp": 1.0240922,
      "epoch": 0.7978355629039531,
      "flos": 22382223575040.0,
      "grad_norm": 1.7544193423281385,
      "language_loss": 0.82239574,
      "learning_rate": 4.135205575764922e-07,
      "loss": 0.84361237,
      "num_input_tokens_seen": 286286775,
      "step": 13270,
      "time_per_iteration": 2.483074188232422
    },
    {
      "auxiliary_loss_clip": 0.01064749,
      "auxiliary_loss_mlp": 0.01034752,
      "balance_loss_clip": 1.03543353,
      "balance_loss_mlp": 1.02169323,
      "epoch": 0.7978956861566211,
      "flos": 20266331068800.0,
      "grad_norm": 2.3893799784034333,
      "language_loss": 0.59755838,
      "learning_rate": 4.1328344075101905e-07,
      "loss": 0.6185534,
      "num_input_tokens_seen": 286305590,
      "step": 13271,
      "time_per_iteration": 2.5559189319610596
    },
    {
      "auxiliary_loss_clip": 0.01086475,
      "auxiliary_loss_mlp": 0.01028522,
      "balance_loss_clip": 1.03586292,
      "balance_loss_mlp": 1.01646376,
      "epoch": 0.797955809409289,
      "flos": 28112417366400.0,
      "grad_norm": 1.603264273973951,
      "language_loss": 0.73160207,
      "learning_rate": 4.130463840939975e-07,
      "loss": 0.75275207,
      "num_input_tokens_seen": 286328050,
      "step": 13272,
      "time_per_iteration": 2.574772596359253
    },
    {
      "auxiliary_loss_clip": 0.01042768,
      "auxiliary_loss_mlp": 0.01033047,
      "balance_loss_clip": 1.03184223,
      "balance_loss_mlp": 1.0197854,
      "epoch": 0.798015932661957,
      "flos": 15559591495680.0,
      "grad_norm": 1.749781061764243,
      "language_loss": 0.71465576,
      "learning_rate": 4.128093876144161e-07,
      "loss": 0.73541391,
      "num_input_tokens_seen": 286345265,
      "step": 13273,
      "time_per_iteration": 2.5914793014526367
    },
    {
      "auxiliary_loss_clip": 0.01083184,
      "auxiliary_loss_mlp": 0.01033997,
      "balance_loss_clip": 1.0364995,
      "balance_loss_mlp": 1.02133656,
      "epoch": 0.7980760559146249,
      "flos": 23951196622080.0,
      "grad_norm": 1.7345356106681178,
      "language_loss": 0.75649399,
      "learning_rate": 4.1257245132126117e-07,
      "loss": 0.77766579,
      "num_input_tokens_seen": 286364465,
      "step": 13274,
      "time_per_iteration": 2.5294363498687744
    },
    {
      "auxiliary_loss_clip": 0.01051479,
      "auxiliary_loss_mlp": 0.01028277,
      "balance_loss_clip": 1.03218484,
      "balance_loss_mlp": 1.01716113,
      "epoch": 0.798136179167293,
      "flos": 28038082170240.0,
      "grad_norm": 1.3081115644855235,
      "language_loss": 0.7779181,
      "learning_rate": 4.12335575223518e-07,
      "loss": 0.79871571,
      "num_input_tokens_seen": 286385565,
      "step": 13275,
      "time_per_iteration": 2.6309478282928467
    },
    {
      "auxiliary_loss_clip": 0.0109561,
      "auxiliary_loss_mlp": 0.0103434,
      "balance_loss_clip": 1.03588104,
      "balance_loss_mlp": 1.02154255,
      "epoch": 0.7981963024199609,
      "flos": 35984538046080.0,
      "grad_norm": 1.9308364009838062,
      "language_loss": 0.64156985,
      "learning_rate": 4.1209875933016877e-07,
      "loss": 0.66286939,
      "num_input_tokens_seen": 286403950,
      "step": 13276,
      "time_per_iteration": 2.5662143230438232
    },
    {
      "auxiliary_loss_clip": 0.0106286,
      "auxiliary_loss_mlp": 0.01032025,
      "balance_loss_clip": 1.0339601,
      "balance_loss_mlp": 1.02040827,
      "epoch": 0.7982564256726289,
      "flos": 25884914325120.0,
      "grad_norm": 1.6493748511605029,
      "language_loss": 0.60804826,
      "learning_rate": 4.118620036501945e-07,
      "loss": 0.62899709,
      "num_input_tokens_seen": 286426160,
      "step": 13277,
      "time_per_iteration": 2.5986385345458984
    },
    {
      "auxiliary_loss_clip": 0.01080978,
      "auxiliary_loss_mlp": 0.01035291,
      "balance_loss_clip": 1.03588653,
      "balance_loss_mlp": 1.02277994,
      "epoch": 0.7983165489252969,
      "flos": 25739152934400.0,
      "grad_norm": 1.8630638729775306,
      "language_loss": 0.79970646,
      "learning_rate": 4.1162530819257227e-07,
      "loss": 0.82086915,
      "num_input_tokens_seen": 286446610,
      "step": 13278,
      "time_per_iteration": 2.534487247467041
    },
    {
      "auxiliary_loss_clip": 0.01084434,
      "auxiliary_loss_mlp": 0.01034561,
      "balance_loss_clip": 1.03529048,
      "balance_loss_mlp": 1.02203798,
      "epoch": 0.7983766721779648,
      "flos": 21908202768000.0,
      "grad_norm": 1.9536461288656146,
      "language_loss": 0.63583457,
      "learning_rate": 4.113886729662768e-07,
      "loss": 0.6570245,
      "num_input_tokens_seen": 286465460,
      "step": 13279,
      "time_per_iteration": 2.535396099090576
    },
    {
      "auxiliary_loss_clip": 0.01089529,
      "auxiliary_loss_mlp": 0.0102658,
      "balance_loss_clip": 1.03578258,
      "balance_loss_mlp": 1.01566005,
      "epoch": 0.7984367954306328,
      "flos": 29347420734720.0,
      "grad_norm": 1.8358511408257965,
      "language_loss": 0.71084499,
      "learning_rate": 4.111520979802825e-07,
      "loss": 0.73200607,
      "num_input_tokens_seen": 286485720,
      "step": 13280,
      "time_per_iteration": 2.5524742603302
    },
    {
      "auxiliary_loss_clip": 0.01065445,
      "auxiliary_loss_mlp": 0.01035009,
      "balance_loss_clip": 1.03397536,
      "balance_loss_mlp": 1.02043009,
      "epoch": 0.7984969186833007,
      "flos": 31357772104320.0,
      "grad_norm": 1.8005110732560285,
      "language_loss": 0.62975854,
      "learning_rate": 4.1091558324355955e-07,
      "loss": 0.65076309,
      "num_input_tokens_seen": 286507465,
      "step": 13281,
      "time_per_iteration": 2.6515886783599854
    },
    {
      "auxiliary_loss_clip": 0.01093673,
      "auxiliary_loss_mlp": 0.01033915,
      "balance_loss_clip": 1.03361416,
      "balance_loss_mlp": 1.02135003,
      "epoch": 0.7985570419359688,
      "flos": 24312924535680.0,
      "grad_norm": 2.0003697898732895,
      "language_loss": 0.80217367,
      "learning_rate": 4.1067912876507683e-07,
      "loss": 0.82344949,
      "num_input_tokens_seen": 286526345,
      "step": 13282,
      "time_per_iteration": 2.496497869491577
    },
    {
      "auxiliary_loss_clip": 0.01071686,
      "auxiliary_loss_mlp": 0.00783428,
      "balance_loss_clip": 1.03334069,
      "balance_loss_mlp": 1.01083875,
      "epoch": 0.7986171651886367,
      "flos": 15742233175680.0,
      "grad_norm": 1.8014220771384368,
      "language_loss": 0.71762258,
      "learning_rate": 4.10442734553802e-07,
      "loss": 0.73617375,
      "num_input_tokens_seen": 286544095,
      "step": 13283,
      "time_per_iteration": 2.5499420166015625
    },
    {
      "auxiliary_loss_clip": 0.01089089,
      "auxiliary_loss_mlp": 0.01026601,
      "balance_loss_clip": 1.03343511,
      "balance_loss_mlp": 1.0153234,
      "epoch": 0.7986772884413047,
      "flos": 11619401091840.0,
      "grad_norm": 1.8885048269022147,
      "language_loss": 0.73305601,
      "learning_rate": 4.102064006186967e-07,
      "loss": 0.75421292,
      "num_input_tokens_seen": 286560960,
      "step": 13284,
      "time_per_iteration": 2.457382917404175
    },
    {
      "auxiliary_loss_clip": 0.01076905,
      "auxiliary_loss_mlp": 0.01036417,
      "balance_loss_clip": 1.03342867,
      "balance_loss_mlp": 1.02521741,
      "epoch": 0.7987374116939726,
      "flos": 22091059929600.0,
      "grad_norm": 1.4303897242407437,
      "language_loss": 0.70467502,
      "learning_rate": 4.0997012696872415e-07,
      "loss": 0.72580826,
      "num_input_tokens_seen": 286579865,
      "step": 13285,
      "time_per_iteration": 2.566587448120117
    },
    {
      "auxiliary_loss_clip": 0.01076045,
      "auxiliary_loss_mlp": 0.01029261,
      "balance_loss_clip": 1.03250575,
      "balance_loss_mlp": 1.01773357,
      "epoch": 0.7987975349466406,
      "flos": 17890696339200.0,
      "grad_norm": 1.6930351777122357,
      "language_loss": 0.73077178,
      "learning_rate": 4.097339136128437e-07,
      "loss": 0.7518248,
      "num_input_tokens_seen": 286597295,
      "step": 13286,
      "time_per_iteration": 2.485409736633301
    },
    {
      "auxiliary_loss_clip": 0.01082288,
      "auxiliary_loss_mlp": 0.0103162,
      "balance_loss_clip": 1.03469586,
      "balance_loss_mlp": 1.01978886,
      "epoch": 0.7988576581993085,
      "flos": 19719232041600.0,
      "grad_norm": 1.7727940748981907,
      "language_loss": 0.75125241,
      "learning_rate": 4.0949776056001296e-07,
      "loss": 0.7723915,
      "num_input_tokens_seen": 286616270,
      "step": 13287,
      "time_per_iteration": 5.264992952346802
    },
    {
      "auxiliary_loss_clip": 0.01080884,
      "auxiliary_loss_mlp": 0.01026153,
      "balance_loss_clip": 1.03751624,
      "balance_loss_mlp": 1.01467311,
      "epoch": 0.7989177814519766,
      "flos": 28036358317440.0,
      "grad_norm": 1.474529186062847,
      "language_loss": 0.61799055,
      "learning_rate": 4.092616678191863e-07,
      "loss": 0.63906091,
      "num_input_tokens_seen": 286638315,
      "step": 13288,
      "time_per_iteration": 2.6056551933288574
    },
    {
      "auxiliary_loss_clip": 0.01093123,
      "auxiliary_loss_mlp": 0.01028934,
      "balance_loss_clip": 1.0367918,
      "balance_loss_mlp": 1.01809835,
      "epoch": 0.7989779047046445,
      "flos": 28871029630080.0,
      "grad_norm": 1.8993816038295326,
      "language_loss": 0.69960684,
      "learning_rate": 4.090256353993169e-07,
      "loss": 0.72082734,
      "num_input_tokens_seen": 286658630,
      "step": 13289,
      "time_per_iteration": 2.52897572517395
    },
    {
      "auxiliary_loss_clip": 0.01069323,
      "auxiliary_loss_mlp": 0.01034115,
      "balance_loss_clip": 1.03562832,
      "balance_loss_mlp": 1.02174103,
      "epoch": 0.7990380279573125,
      "flos": 18186887888640.0,
      "grad_norm": 1.998427638628982,
      "language_loss": 0.62400842,
      "learning_rate": 4.0878966330935506e-07,
      "loss": 0.64504278,
      "num_input_tokens_seen": 286676870,
      "step": 13290,
      "time_per_iteration": 4.05820369720459
    },
    {
      "auxiliary_loss_clip": 0.01096334,
      "auxiliary_loss_mlp": 0.0103051,
      "balance_loss_clip": 1.03586972,
      "balance_loss_mlp": 1.0175221,
      "epoch": 0.7990981512099805,
      "flos": 20879936127360.0,
      "grad_norm": 2.4688692287134515,
      "language_loss": 0.71719831,
      "learning_rate": 4.08553751558248e-07,
      "loss": 0.73846674,
      "num_input_tokens_seen": 286694300,
      "step": 13291,
      "time_per_iteration": 2.529148817062378
    },
    {
      "auxiliary_loss_clip": 0.01064974,
      "auxiliary_loss_mlp": 0.01025949,
      "balance_loss_clip": 1.0340817,
      "balance_loss_mlp": 1.01501119,
      "epoch": 0.7991582744626484,
      "flos": 26099911180800.0,
      "grad_norm": 1.5208533606290249,
      "language_loss": 0.63467246,
      "learning_rate": 4.083179001549422e-07,
      "loss": 0.65558165,
      "num_input_tokens_seen": 286714545,
      "step": 13292,
      "time_per_iteration": 2.567009687423706
    },
    {
      "auxiliary_loss_clip": 0.01091776,
      "auxiliary_loss_mlp": 0.0103245,
      "balance_loss_clip": 1.03426325,
      "balance_loss_mlp": 1.02125025,
      "epoch": 0.7992183977153164,
      "flos": 35295843605760.0,
      "grad_norm": 1.9084777367447023,
      "language_loss": 0.56018895,
      "learning_rate": 4.0808210910838105e-07,
      "loss": 0.58143121,
      "num_input_tokens_seen": 286734525,
      "step": 13293,
      "time_per_iteration": 2.648062229156494
    },
    {
      "auxiliary_loss_clip": 0.0107842,
      "auxiliary_loss_mlp": 0.01031636,
      "balance_loss_clip": 1.03742754,
      "balance_loss_mlp": 1.01972067,
      "epoch": 0.7992785209679844,
      "flos": 51853426577280.0,
      "grad_norm": 2.7893960989675644,
      "language_loss": 0.71615899,
      "learning_rate": 4.0784637842750704e-07,
      "loss": 0.73725963,
      "num_input_tokens_seen": 286753430,
      "step": 13294,
      "time_per_iteration": 2.7580318450927734
    },
    {
      "auxiliary_loss_clip": 0.01069245,
      "auxiliary_loss_mlp": 0.01035752,
      "balance_loss_clip": 1.03507876,
      "balance_loss_mlp": 1.02312791,
      "epoch": 0.7993386442206524,
      "flos": 22565116650240.0,
      "grad_norm": 1.7888106865134243,
      "language_loss": 0.72384816,
      "learning_rate": 4.0761070812125675e-07,
      "loss": 0.74489814,
      "num_input_tokens_seen": 286771915,
      "step": 13295,
      "time_per_iteration": 2.558220386505127
    },
    {
      "auxiliary_loss_clip": 0.01070751,
      "auxiliary_loss_mlp": 0.01035757,
      "balance_loss_clip": 1.03565621,
      "balance_loss_mlp": 1.0244441,
      "epoch": 0.7993987674733203,
      "flos": 18800277465600.0,
      "grad_norm": 1.6881909113370248,
      "language_loss": 0.76558661,
      "learning_rate": 4.0737509819856797e-07,
      "loss": 0.78665173,
      "num_input_tokens_seen": 286789835,
      "step": 13296,
      "time_per_iteration": 2.525397777557373
    },
    {
      "auxiliary_loss_clip": 0.01004385,
      "auxiliary_loss_mlp": 0.00997161,
      "balance_loss_clip": 1.01361716,
      "balance_loss_mlp": 0.99598122,
      "epoch": 0.7994588907259883,
      "flos": 69421720394880.0,
      "grad_norm": 0.6948636867311111,
      "language_loss": 0.60819817,
      "learning_rate": 4.0713954866837573e-07,
      "loss": 0.62821364,
      "num_input_tokens_seen": 286855580,
      "step": 13297,
      "time_per_iteration": 4.623366117477417
    },
    {
      "auxiliary_loss_clip": 0.01082448,
      "auxiliary_loss_mlp": 0.01030197,
      "balance_loss_clip": 1.03628063,
      "balance_loss_mlp": 1.01900351,
      "epoch": 0.7995190139786562,
      "flos": 13480327883520.0,
      "grad_norm": 2.123015536565371,
      "language_loss": 0.69865847,
      "learning_rate": 4.0690405953961073e-07,
      "loss": 0.71978498,
      "num_input_tokens_seen": 286874360,
      "step": 13298,
      "time_per_iteration": 2.5298879146575928
    },
    {
      "auxiliary_loss_clip": 0.01067275,
      "auxiliary_loss_mlp": 0.01035168,
      "balance_loss_clip": 1.03426349,
      "balance_loss_mlp": 1.02180445,
      "epoch": 0.7995791372313242,
      "flos": 21652842003840.0,
      "grad_norm": 1.9641277104959538,
      "language_loss": 0.75562048,
      "learning_rate": 4.066686308212037e-07,
      "loss": 0.77664495,
      "num_input_tokens_seen": 286891950,
      "step": 13299,
      "time_per_iteration": 2.53023362159729
    },
    {
      "auxiliary_loss_clip": 0.01078013,
      "auxiliary_loss_mlp": 0.01030461,
      "balance_loss_clip": 1.03389287,
      "balance_loss_mlp": 1.01941013,
      "epoch": 0.7996392604839921,
      "flos": 26068130622720.0,
      "grad_norm": 1.6332192841080453,
      "language_loss": 0.77607715,
      "learning_rate": 4.064332625220828e-07,
      "loss": 0.79716188,
      "num_input_tokens_seen": 286911725,
      "step": 13300,
      "time_per_iteration": 2.6062376499176025
    },
    {
      "auxiliary_loss_clip": 0.01064062,
      "auxiliary_loss_mlp": 0.01039573,
      "balance_loss_clip": 1.03187001,
      "balance_loss_mlp": 1.02516055,
      "epoch": 0.7996993837366602,
      "flos": 24606889441920.0,
      "grad_norm": 2.209297622575398,
      "language_loss": 0.63990778,
      "learning_rate": 4.0619795465117115e-07,
      "loss": 0.6609441,
      "num_input_tokens_seen": 286931400,
      "step": 13301,
      "time_per_iteration": 2.5811424255371094
    },
    {
      "auxiliary_loss_clip": 0.01092073,
      "auxiliary_loss_mlp": 0.0103773,
      "balance_loss_clip": 1.03642178,
      "balance_loss_mlp": 1.02567172,
      "epoch": 0.7997595069893281,
      "flos": 20992049452800.0,
      "grad_norm": 1.606530216888061,
      "language_loss": 0.72011799,
      "learning_rate": 4.059627072173928e-07,
      "loss": 0.74141598,
      "num_input_tokens_seen": 286949795,
      "step": 13302,
      "time_per_iteration": 2.5813241004943848
    },
    {
      "auxiliary_loss_clip": 0.01106842,
      "auxiliary_loss_mlp": 0.0078726,
      "balance_loss_clip": 1.03603709,
      "balance_loss_mlp": 1.01529384,
      "epoch": 0.7998196302419961,
      "flos": 24426510318720.0,
      "grad_norm": 1.7861110881445073,
      "language_loss": 0.8350457,
      "learning_rate": 4.057275202296684e-07,
      "loss": 0.85398674,
      "num_input_tokens_seen": 286968805,
      "step": 13303,
      "time_per_iteration": 2.4809443950653076
    },
    {
      "auxiliary_loss_clip": 0.01101855,
      "auxiliary_loss_mlp": 0.01030243,
      "balance_loss_clip": 1.03551757,
      "balance_loss_mlp": 1.01892436,
      "epoch": 0.7998797534946641,
      "flos": 30264651457920.0,
      "grad_norm": 1.728649453475977,
      "language_loss": 0.58999354,
      "learning_rate": 4.054923936969166e-07,
      "loss": 0.61131454,
      "num_input_tokens_seen": 286990235,
      "step": 13304,
      "time_per_iteration": 2.5681190490722656
    },
    {
      "auxiliary_loss_clip": 0.01105463,
      "auxiliary_loss_mlp": 0.0102892,
      "balance_loss_clip": 1.03448009,
      "balance_loss_mlp": 1.01599777,
      "epoch": 0.799939876747332,
      "flos": 23513984277120.0,
      "grad_norm": 1.5442205248312337,
      "language_loss": 0.6893127,
      "learning_rate": 4.0525732762805265e-07,
      "loss": 0.71065658,
      "num_input_tokens_seen": 287011060,
      "step": 13305,
      "time_per_iteration": 2.4815049171447754
    },
    {
      "auxiliary_loss_clip": 0.01067717,
      "auxiliary_loss_mlp": 0.01028779,
      "balance_loss_clip": 1.03604078,
      "balance_loss_mlp": 1.01733506,
      "epoch": 0.8,
      "flos": 19318109886720.0,
      "grad_norm": 1.5769833032011886,
      "language_loss": 0.6946187,
      "learning_rate": 4.0502232203199107e-07,
      "loss": 0.71558368,
      "num_input_tokens_seen": 287029215,
      "step": 13306,
      "time_per_iteration": 2.586491584777832
    },
    {
      "auxiliary_loss_clip": 0.01093162,
      "auxiliary_loss_mlp": 0.0103286,
      "balance_loss_clip": 1.035622,
      "balance_loss_mlp": 1.02109456,
      "epoch": 0.800060123252668,
      "flos": 32412432263040.0,
      "grad_norm": 1.6390274856589602,
      "language_loss": 0.69741178,
      "learning_rate": 4.0478737691764286e-07,
      "loss": 0.71867198,
      "num_input_tokens_seen": 287050855,
      "step": 13307,
      "time_per_iteration": 2.5763046741485596
    },
    {
      "auxiliary_loss_clip": 0.01082871,
      "auxiliary_loss_mlp": 0.01034815,
      "balance_loss_clip": 1.03584003,
      "balance_loss_mlp": 1.02325821,
      "epoch": 0.800120246505336,
      "flos": 20010611168640.0,
      "grad_norm": 2.1679408904688895,
      "language_loss": 0.7694096,
      "learning_rate": 4.0455249229391677e-07,
      "loss": 0.79058647,
      "num_input_tokens_seen": 287069915,
      "step": 13308,
      "time_per_iteration": 2.5316720008850098
    },
    {
      "auxiliary_loss_clip": 0.01058498,
      "auxiliary_loss_mlp": 0.01029419,
      "balance_loss_clip": 1.03469729,
      "balance_loss_mlp": 1.01638341,
      "epoch": 0.8001803697580039,
      "flos": 31868278151040.0,
      "grad_norm": 1.3860982236034685,
      "language_loss": 0.78555453,
      "learning_rate": 4.0431766816972e-07,
      "loss": 0.80643368,
      "num_input_tokens_seen": 287091450,
      "step": 13309,
      "time_per_iteration": 2.6407670974731445
    },
    {
      "auxiliary_loss_clip": 0.01030828,
      "auxiliary_loss_mlp": 0.01002371,
      "balance_loss_clip": 1.00798631,
      "balance_loss_mlp": 1.00122106,
      "epoch": 0.8002404930106719,
      "flos": 63392066916480.0,
      "grad_norm": 0.9408338294991487,
      "language_loss": 0.64609748,
      "learning_rate": 4.040829045539571e-07,
      "loss": 0.66642946,
      "num_input_tokens_seen": 287148365,
      "step": 13310,
      "time_per_iteration": 3.0011370182037354
    },
    {
      "auxiliary_loss_clip": 0.0109379,
      "auxiliary_loss_mlp": 0.0103379,
      "balance_loss_clip": 1.03590107,
      "balance_loss_mlp": 1.0222801,
      "epoch": 0.8003006162633398,
      "flos": 27855476403840.0,
      "grad_norm": 1.833716676248296,
      "language_loss": 0.82588959,
      "learning_rate": 4.0384820145553156e-07,
      "loss": 0.84716541,
      "num_input_tokens_seen": 287168280,
      "step": 13311,
      "time_per_iteration": 2.5320956707000732
    },
    {
      "auxiliary_loss_clip": 0.0109512,
      "auxiliary_loss_mlp": 0.01033419,
      "balance_loss_clip": 1.03629231,
      "balance_loss_mlp": 1.0216893,
      "epoch": 0.8003607395160078,
      "flos": 18223337214720.0,
      "grad_norm": 2.6590635902118795,
      "language_loss": 0.66220164,
      "learning_rate": 4.0361355888334116e-07,
      "loss": 0.68348706,
      "num_input_tokens_seen": 287185980,
      "step": 13312,
      "time_per_iteration": 2.4875237941741943
    },
    {
      "auxiliary_loss_clip": 0.01109762,
      "auxiliary_loss_mlp": 0.01033877,
      "balance_loss_clip": 1.03905129,
      "balance_loss_mlp": 1.02118707,
      "epoch": 0.8004208627686757,
      "flos": 20886975192960.0,
      "grad_norm": 1.703268378942902,
      "language_loss": 0.75325143,
      "learning_rate": 4.033789768462843e-07,
      "loss": 0.77468783,
      "num_input_tokens_seen": 287203875,
      "step": 13313,
      "time_per_iteration": 2.500614643096924
    },
    {
      "auxiliary_loss_clip": 0.01091577,
      "auxiliary_loss_mlp": 0.0103246,
      "balance_loss_clip": 1.03376472,
      "balance_loss_mlp": 1.02019298,
      "epoch": 0.8004809860213438,
      "flos": 26436143416320.0,
      "grad_norm": 1.3199652453216375,
      "language_loss": 0.75586075,
      "learning_rate": 4.031444553532575e-07,
      "loss": 0.7771011,
      "num_input_tokens_seen": 287226445,
      "step": 13314,
      "time_per_iteration": 2.560427188873291
    },
    {
      "auxiliary_loss_clip": 0.00992168,
      "auxiliary_loss_mlp": 0.0100162,
      "balance_loss_clip": 1.01326847,
      "balance_loss_mlp": 1.00057149,
      "epoch": 0.8005411092740117,
      "flos": 63648612829440.0,
      "grad_norm": 0.8145517156108162,
      "language_loss": 0.53761435,
      "learning_rate": 4.029099944131522e-07,
      "loss": 0.55755222,
      "num_input_tokens_seen": 287286240,
      "step": 13315,
      "time_per_iteration": 3.121859550476074
    },
    {
      "auxiliary_loss_clip": 0.01080576,
      "auxiliary_loss_mlp": 0.01032077,
      "balance_loss_clip": 1.03448415,
      "balance_loss_mlp": 1.01997709,
      "epoch": 0.8006012325266797,
      "flos": 36138056774400.0,
      "grad_norm": 1.5119309087865334,
      "language_loss": 0.71077794,
      "learning_rate": 4.026755940348603e-07,
      "loss": 0.73190445,
      "num_input_tokens_seen": 287310265,
      "step": 13316,
      "time_per_iteration": 2.670164108276367
    },
    {
      "auxiliary_loss_clip": 0.01084186,
      "auxiliary_loss_mlp": 0.01029162,
      "balance_loss_clip": 1.03592563,
      "balance_loss_mlp": 1.01689506,
      "epoch": 0.8006613557793477,
      "flos": 33838947970560.0,
      "grad_norm": 1.810500196853428,
      "language_loss": 0.64524603,
      "learning_rate": 4.024412542272706e-07,
      "loss": 0.66637951,
      "num_input_tokens_seen": 287331610,
      "step": 13317,
      "time_per_iteration": 2.6386115550994873
    },
    {
      "auxiliary_loss_clip": 0.01030593,
      "auxiliary_loss_mlp": 0.01005247,
      "balance_loss_clip": 1.00785232,
      "balance_loss_mlp": 1.00411439,
      "epoch": 0.8007214790320156,
      "flos": 67348310699520.0,
      "grad_norm": 0.7873625651663488,
      "language_loss": 0.58995062,
      "learning_rate": 4.0220697499926783e-07,
      "loss": 0.610309,
      "num_input_tokens_seen": 287394795,
      "step": 13318,
      "time_per_iteration": 3.158604621887207
    },
    {
      "auxiliary_loss_clip": 0.01070179,
      "auxiliary_loss_mlp": 0.01025886,
      "balance_loss_clip": 1.03537011,
      "balance_loss_mlp": 1.01445413,
      "epoch": 0.8007816022846836,
      "flos": 23185653033600.0,
      "grad_norm": 1.640218059856688,
      "language_loss": 0.66601396,
      "learning_rate": 4.019727563597366e-07,
      "loss": 0.68697459,
      "num_input_tokens_seen": 287414595,
      "step": 13319,
      "time_per_iteration": 2.5701730251312256
    },
    {
      "auxiliary_loss_clip": 0.01105446,
      "auxiliary_loss_mlp": 0.00786595,
      "balance_loss_clip": 1.03482199,
      "balance_loss_mlp": 1.01220965,
      "epoch": 0.8008417255373516,
      "flos": 21981388728960.0,
      "grad_norm": 2.322316088063077,
      "language_loss": 0.73790193,
      "learning_rate": 4.0173859831755873e-07,
      "loss": 0.75682235,
      "num_input_tokens_seen": 287434395,
      "step": 13320,
      "time_per_iteration": 2.5361669063568115
    },
    {
      "auxiliary_loss_clip": 0.01091484,
      "auxiliary_loss_mlp": 0.01026025,
      "balance_loss_clip": 1.0356741,
      "balance_loss_mlp": 1.01422358,
      "epoch": 0.8009018487900196,
      "flos": 16727334647040.0,
      "grad_norm": 2.4874330352689267,
      "language_loss": 0.80477959,
      "learning_rate": 4.015045008816138e-07,
      "loss": 0.82595462,
      "num_input_tokens_seen": 287450590,
      "step": 13321,
      "time_per_iteration": 2.4656593799591064
    },
    {
      "auxiliary_loss_clip": 0.01038328,
      "auxiliary_loss_mlp": 0.01033839,
      "balance_loss_clip": 1.02967834,
      "balance_loss_mlp": 1.02192414,
      "epoch": 0.8009619720426875,
      "flos": 20813609664000.0,
      "grad_norm": 1.6840438039764396,
      "language_loss": 0.66068524,
      "learning_rate": 4.0127046406077825e-07,
      "loss": 0.68140692,
      "num_input_tokens_seen": 287468455,
      "step": 13322,
      "time_per_iteration": 2.6366305351257324
    },
    {
      "auxiliary_loss_clip": 0.01093089,
      "auxiliary_loss_mlp": 0.01026597,
      "balance_loss_clip": 1.03435409,
      "balance_loss_mlp": 1.01495051,
      "epoch": 0.8010220952953555,
      "flos": 17931096161280.0,
      "grad_norm": 1.7304915018033709,
      "language_loss": 0.77987981,
      "learning_rate": 4.010364878639265e-07,
      "loss": 0.80107671,
      "num_input_tokens_seen": 287486485,
      "step": 13323,
      "time_per_iteration": 2.4652516841888428
    },
    {
      "auxiliary_loss_clip": 0.01105903,
      "auxiliary_loss_mlp": 0.01034504,
      "balance_loss_clip": 1.03536582,
      "balance_loss_mlp": 1.02235067,
      "epoch": 0.8010822185480234,
      "flos": 24572235795840.0,
      "grad_norm": 3.278598536802285,
      "language_loss": 0.71834219,
      "learning_rate": 4.00802572299932e-07,
      "loss": 0.73974621,
      "num_input_tokens_seen": 287503940,
      "step": 13324,
      "time_per_iteration": 2.5188510417938232
    },
    {
      "auxiliary_loss_clip": 0.01070718,
      "auxiliary_loss_mlp": 0.01031457,
      "balance_loss_clip": 1.03224802,
      "balance_loss_mlp": 1.01891005,
      "epoch": 0.8011423418006914,
      "flos": 21829988903040.0,
      "grad_norm": 2.1078516340882514,
      "language_loss": 0.76646745,
      "learning_rate": 4.005687173776635e-07,
      "loss": 0.78748918,
      "num_input_tokens_seen": 287521660,
      "step": 13325,
      "time_per_iteration": 5.3655829429626465
    },
    {
      "auxiliary_loss_clip": 0.0108614,
      "auxiliary_loss_mlp": 0.01026314,
      "balance_loss_clip": 1.03312159,
      "balance_loss_mlp": 1.0157758,
      "epoch": 0.8012024650533593,
      "flos": 23915178259200.0,
      "grad_norm": 1.6978798092074072,
      "language_loss": 0.79394805,
      "learning_rate": 4.003349231059898e-07,
      "loss": 0.8150726,
      "num_input_tokens_seen": 287541505,
      "step": 13326,
      "time_per_iteration": 2.5177502632141113
    },
    {
      "auxiliary_loss_clip": 0.01090485,
      "auxiliary_loss_mlp": 0.01031668,
      "balance_loss_clip": 1.03396702,
      "balance_loss_mlp": 1.02050436,
      "epoch": 0.8012625883060274,
      "flos": 23587062497280.0,
      "grad_norm": 2.912477557845531,
      "language_loss": 0.66161281,
      "learning_rate": 4.001011894937765e-07,
      "loss": 0.68283439,
      "num_input_tokens_seen": 287560015,
      "step": 13327,
      "time_per_iteration": 2.5111331939697266
    },
    {
      "auxiliary_loss_clip": 0.01089458,
      "auxiliary_loss_mlp": 0.01027476,
      "balance_loss_clip": 1.0354948,
      "balance_loss_mlp": 1.01665151,
      "epoch": 0.8013227115586953,
      "flos": 20813932886400.0,
      "grad_norm": 1.8205493290709498,
      "language_loss": 0.73709273,
      "learning_rate": 3.9986751654988636e-07,
      "loss": 0.75826204,
      "num_input_tokens_seen": 287579150,
      "step": 13328,
      "time_per_iteration": 4.1316986083984375
    },
    {
      "auxiliary_loss_clip": 0.01054057,
      "auxiliary_loss_mlp": 0.01032132,
      "balance_loss_clip": 1.03644693,
      "balance_loss_mlp": 1.01923966,
      "epoch": 0.8013828348113633,
      "flos": 15888317788800.0,
      "grad_norm": 1.9041794062794475,
      "language_loss": 0.7410742,
      "learning_rate": 3.996339042831798e-07,
      "loss": 0.76193607,
      "num_input_tokens_seen": 287597420,
      "step": 13329,
      "time_per_iteration": 2.549942970275879
    },
    {
      "auxiliary_loss_clip": 0.01023602,
      "auxiliary_loss_mlp": 0.01001551,
      "balance_loss_clip": 1.00968885,
      "balance_loss_mlp": 1.00050759,
      "epoch": 0.8014429580640313,
      "flos": 71062981562880.0,
      "grad_norm": 0.6905274449975086,
      "language_loss": 0.52924335,
      "learning_rate": 3.9940035270251605e-07,
      "loss": 0.54949486,
      "num_input_tokens_seen": 287667280,
      "step": 13330,
      "time_per_iteration": 3.201615810394287
    },
    {
      "auxiliary_loss_clip": 0.01080285,
      "auxiliary_loss_mlp": 0.01036242,
      "balance_loss_clip": 1.03457117,
      "balance_loss_mlp": 1.02262878,
      "epoch": 0.8015030813166992,
      "flos": 23076340968960.0,
      "grad_norm": 1.6884841419359697,
      "language_loss": 0.72646618,
      "learning_rate": 3.991668618167519e-07,
      "loss": 0.74763143,
      "num_input_tokens_seen": 287687375,
      "step": 13331,
      "time_per_iteration": 2.52752685546875
    },
    {
      "auxiliary_loss_clip": 0.01087159,
      "auxiliary_loss_mlp": 0.01030235,
      "balance_loss_clip": 1.03454638,
      "balance_loss_mlp": 1.01954794,
      "epoch": 0.8015632045693672,
      "flos": 21872328059520.0,
      "grad_norm": 1.951989496180462,
      "language_loss": 0.77215278,
      "learning_rate": 3.989334316347401e-07,
      "loss": 0.79332674,
      "num_input_tokens_seen": 287707895,
      "step": 13332,
      "time_per_iteration": 2.516373872756958
    },
    {
      "auxiliary_loss_clip": 0.0110572,
      "auxiliary_loss_mlp": 0.01027592,
      "balance_loss_clip": 1.03707957,
      "balance_loss_mlp": 1.01533175,
      "epoch": 0.8016233278220352,
      "flos": 23656728925440.0,
      "grad_norm": 1.7369412907283384,
      "language_loss": 0.83228683,
      "learning_rate": 3.987000621653338e-07,
      "loss": 0.85361993,
      "num_input_tokens_seen": 287723990,
      "step": 13333,
      "time_per_iteration": 2.4531333446502686
    },
    {
      "auxiliary_loss_clip": 0.01082802,
      "auxiliary_loss_mlp": 0.01026443,
      "balance_loss_clip": 1.03421903,
      "balance_loss_mlp": 1.01433158,
      "epoch": 0.8016834510747032,
      "flos": 16253170185600.0,
      "grad_norm": 1.996126962887249,
      "language_loss": 0.73358667,
      "learning_rate": 3.9846675341738133e-07,
      "loss": 0.75467908,
      "num_input_tokens_seen": 287742380,
      "step": 13334,
      "time_per_iteration": 3.900388717651367
    },
    {
      "auxiliary_loss_clip": 0.01065096,
      "auxiliary_loss_mlp": 0.01033372,
      "balance_loss_clip": 1.03323591,
      "balance_loss_mlp": 1.02038431,
      "epoch": 0.8017435743273711,
      "flos": 12276027665280.0,
      "grad_norm": 1.8754934942047101,
      "language_loss": 0.74536192,
      "learning_rate": 3.9823350539972967e-07,
      "loss": 0.76634657,
      "num_input_tokens_seen": 287760130,
      "step": 13335,
      "time_per_iteration": 2.55130672454834
    },
    {
      "auxiliary_loss_clip": 0.01062914,
      "auxiliary_loss_mlp": 0.01031054,
      "balance_loss_clip": 1.03246295,
      "balance_loss_mlp": 1.01859641,
      "epoch": 0.8018036975800391,
      "flos": 17196112068480.0,
      "grad_norm": 1.7703856834538236,
      "language_loss": 0.75363553,
      "learning_rate": 3.9800031812122416e-07,
      "loss": 0.77457523,
      "num_input_tokens_seen": 287777565,
      "step": 13336,
      "time_per_iteration": 2.567640781402588
    },
    {
      "auxiliary_loss_clip": 0.01072967,
      "auxiliary_loss_mlp": 0.01035988,
      "balance_loss_clip": 1.03711772,
      "balance_loss_mlp": 1.02312565,
      "epoch": 0.801863820832707,
      "flos": 20631865824000.0,
      "grad_norm": 2.0758767605761075,
      "language_loss": 0.75050652,
      "learning_rate": 3.977671915907068e-07,
      "loss": 0.77159613,
      "num_input_tokens_seen": 287796310,
      "step": 13337,
      "time_per_iteration": 2.5829451084136963
    },
    {
      "auxiliary_loss_clip": 0.01046574,
      "auxiliary_loss_mlp": 0.0078395,
      "balance_loss_clip": 1.03871942,
      "balance_loss_mlp": 1.00954533,
      "epoch": 0.801923944085375,
      "flos": 30445569285120.0,
      "grad_norm": 1.705399119771899,
      "language_loss": 0.80342925,
      "learning_rate": 3.9753412581701883e-07,
      "loss": 0.82173449,
      "num_input_tokens_seen": 287817330,
      "step": 13338,
      "time_per_iteration": 2.6872522830963135
    },
    {
      "auxiliary_loss_clip": 0.01068863,
      "auxiliary_loss_mlp": 0.01030846,
      "balance_loss_clip": 1.03286004,
      "balance_loss_mlp": 1.01784015,
      "epoch": 0.801984067338043,
      "flos": 20010575255040.0,
      "grad_norm": 1.8136746244296729,
      "language_loss": 0.74446017,
      "learning_rate": 3.9730112080899733e-07,
      "loss": 0.76545727,
      "num_input_tokens_seen": 287835095,
      "step": 13339,
      "time_per_iteration": 2.5608227252960205
    },
    {
      "auxiliary_loss_clip": 0.01089424,
      "auxiliary_loss_mlp": 0.01031364,
      "balance_loss_clip": 1.03348064,
      "balance_loss_mlp": 1.02018821,
      "epoch": 0.802044190590711,
      "flos": 22784028088320.0,
      "grad_norm": 1.6774071267319046,
      "language_loss": 0.7927376,
      "learning_rate": 3.970681765754775e-07,
      "loss": 0.81394547,
      "num_input_tokens_seen": 287854595,
      "step": 13340,
      "time_per_iteration": 2.511946678161621
    },
    {
      "auxiliary_loss_clip": 0.01074949,
      "auxiliary_loss_mlp": 0.01028007,
      "balance_loss_clip": 1.03514183,
      "balance_loss_mlp": 1.01667023,
      "epoch": 0.8021043138433789,
      "flos": 27600115639680.0,
      "grad_norm": 1.7574466578034205,
      "language_loss": 0.67706728,
      "learning_rate": 3.968352931252936e-07,
      "loss": 0.69809687,
      "num_input_tokens_seen": 287876960,
      "step": 13341,
      "time_per_iteration": 2.6934967041015625
    },
    {
      "auxiliary_loss_clip": 0.01012686,
      "auxiliary_loss_mlp": 0.01006435,
      "balance_loss_clip": 1.00865841,
      "balance_loss_mlp": 1.00518882,
      "epoch": 0.8021644370960469,
      "flos": 62063730057600.0,
      "grad_norm": 0.8109848907676462,
      "language_loss": 0.61618185,
      "learning_rate": 3.9660247046727547e-07,
      "loss": 0.6363731,
      "num_input_tokens_seen": 287936530,
      "step": 13342,
      "time_per_iteration": 3.06744122505188
    },
    {
      "auxiliary_loss_clip": 0.01086778,
      "auxiliary_loss_mlp": 0.0103317,
      "balance_loss_clip": 1.03697562,
      "balance_loss_mlp": 1.02065945,
      "epoch": 0.8022245603487148,
      "flos": 23361794352000.0,
      "grad_norm": 1.8333080454217499,
      "language_loss": 0.63677675,
      "learning_rate": 3.963697086102522e-07,
      "loss": 0.65797627,
      "num_input_tokens_seen": 287954285,
      "step": 13343,
      "time_per_iteration": 2.544748067855835
    },
    {
      "auxiliary_loss_clip": 0.01077782,
      "auxiliary_loss_mlp": 0.01025551,
      "balance_loss_clip": 1.03361297,
      "balance_loss_mlp": 1.0145241,
      "epoch": 0.8022846836013828,
      "flos": 10853354712960.0,
      "grad_norm": 1.8176144429317986,
      "language_loss": 0.68549788,
      "learning_rate": 3.96137007563051e-07,
      "loss": 0.70653117,
      "num_input_tokens_seen": 287971595,
      "step": 13344,
      "time_per_iteration": 2.497309923171997
    },
    {
      "auxiliary_loss_clip": 0.01094821,
      "auxiliary_loss_mlp": 0.01027528,
      "balance_loss_clip": 1.03718472,
      "balance_loss_mlp": 1.01524353,
      "epoch": 0.8023448068540509,
      "flos": 29240443054080.0,
      "grad_norm": 1.6303371131209852,
      "language_loss": 0.70206916,
      "learning_rate": 3.9590436733449506e-07,
      "loss": 0.72329271,
      "num_input_tokens_seen": 287992540,
      "step": 13345,
      "time_per_iteration": 2.561580181121826
    },
    {
      "auxiliary_loss_clip": 0.01012069,
      "auxiliary_loss_mlp": 0.01004421,
      "balance_loss_clip": 1.00843906,
      "balance_loss_mlp": 1.00326467,
      "epoch": 0.8024049301067188,
      "flos": 64153588181760.0,
      "grad_norm": 0.883928676296958,
      "language_loss": 0.62981355,
      "learning_rate": 3.956717879334059e-07,
      "loss": 0.64997846,
      "num_input_tokens_seen": 288052810,
      "step": 13346,
      "time_per_iteration": 3.2149910926818848
    },
    {
      "auxiliary_loss_clip": 0.0107944,
      "auxiliary_loss_mlp": 0.0103221,
      "balance_loss_clip": 1.03697801,
      "balance_loss_mlp": 1.02046168,
      "epoch": 0.8024650533593868,
      "flos": 28585360765440.0,
      "grad_norm": 1.526223294966733,
      "language_loss": 0.72732967,
      "learning_rate": 3.9543926936860327e-07,
      "loss": 0.74844623,
      "num_input_tokens_seen": 288073045,
      "step": 13347,
      "time_per_iteration": 2.6037845611572266
    },
    {
      "auxiliary_loss_clip": 0.01096008,
      "auxiliary_loss_mlp": 0.010326,
      "balance_loss_clip": 1.03683412,
      "balance_loss_mlp": 1.02041054,
      "epoch": 0.8025251766120547,
      "flos": 16982264448000.0,
      "grad_norm": 3.005347487713174,
      "language_loss": 0.72915757,
      "learning_rate": 3.9520681164890493e-07,
      "loss": 0.7504437,
      "num_input_tokens_seen": 288091165,
      "step": 13348,
      "time_per_iteration": 2.5179545879364014
    },
    {
      "auxiliary_loss_clip": 0.01083995,
      "auxiliary_loss_mlp": 0.01025399,
      "balance_loss_clip": 1.03721523,
      "balance_loss_mlp": 1.01360273,
      "epoch": 0.8025852998647227,
      "flos": 22163671272960.0,
      "grad_norm": 2.098107771364205,
      "language_loss": 0.75502479,
      "learning_rate": 3.9497441478312444e-07,
      "loss": 0.77611876,
      "num_input_tokens_seen": 288110595,
      "step": 13349,
      "time_per_iteration": 2.5509252548217773
    },
    {
      "auxiliary_loss_clip": 0.01105502,
      "auxiliary_loss_mlp": 0.01035276,
      "balance_loss_clip": 1.03753901,
      "balance_loss_mlp": 1.0238142,
      "epoch": 0.8026454231173906,
      "flos": 22017012042240.0,
      "grad_norm": 2.01755459234755,
      "language_loss": 0.83377945,
      "learning_rate": 3.947420787800755e-07,
      "loss": 0.85518724,
      "num_input_tokens_seen": 288128995,
      "step": 13350,
      "time_per_iteration": 2.4717376232147217
    },
    {
      "auxiliary_loss_clip": 0.01094964,
      "auxiliary_loss_mlp": 0.0103375,
      "balance_loss_clip": 1.03679359,
      "balance_loss_mlp": 1.02193677,
      "epoch": 0.8027055463700586,
      "flos": 22491320158080.0,
      "grad_norm": 2.497726803890766,
      "language_loss": 0.71650499,
      "learning_rate": 3.945098036485679e-07,
      "loss": 0.73779213,
      "num_input_tokens_seen": 288149265,
      "step": 13351,
      "time_per_iteration": 2.4903812408447266
    },
    {
      "auxiliary_loss_clip": 0.0106559,
      "auxiliary_loss_mlp": 0.01029994,
      "balance_loss_clip": 1.03362751,
      "balance_loss_mlp": 1.01760173,
      "epoch": 0.8027656696227266,
      "flos": 28912901909760.0,
      "grad_norm": 1.6372182576046679,
      "language_loss": 0.61719322,
      "learning_rate": 3.9427758939740885e-07,
      "loss": 0.63814902,
      "num_input_tokens_seen": 288170745,
      "step": 13352,
      "time_per_iteration": 2.616285562515259
    },
    {
      "auxiliary_loss_clip": 0.01093321,
      "auxiliary_loss_mlp": 0.01035265,
      "balance_loss_clip": 1.03710389,
      "balance_loss_mlp": 1.02341533,
      "epoch": 0.8028257928753946,
      "flos": 18589374760320.0,
      "grad_norm": 1.6198038648675461,
      "language_loss": 0.76510072,
      "learning_rate": 3.940454360354046e-07,
      "loss": 0.78638661,
      "num_input_tokens_seen": 288189415,
      "step": 13353,
      "time_per_iteration": 2.466602325439453
    },
    {
      "auxiliary_loss_clip": 0.01047623,
      "auxiliary_loss_mlp": 0.01027588,
      "balance_loss_clip": 1.03648007,
      "balance_loss_mlp": 1.01393247,
      "epoch": 0.8028859161280625,
      "flos": 19130009339520.0,
      "grad_norm": 2.479525279796178,
      "language_loss": 0.73323971,
      "learning_rate": 3.938133435713582e-07,
      "loss": 0.75399178,
      "num_input_tokens_seen": 288206900,
      "step": 13354,
      "time_per_iteration": 2.6257808208465576
    },
    {
      "auxiliary_loss_clip": 0.01066113,
      "auxiliary_loss_mlp": 0.01037891,
      "balance_loss_clip": 1.0348922,
      "balance_loss_mlp": 1.02481413,
      "epoch": 0.8029460393807305,
      "flos": 20229881742720.0,
      "grad_norm": 1.9382328274244234,
      "language_loss": 0.65956122,
      "learning_rate": 3.935813120140714e-07,
      "loss": 0.68060124,
      "num_input_tokens_seen": 288224800,
      "step": 13355,
      "time_per_iteration": 2.550434112548828
    },
    {
      "auxiliary_loss_clip": 0.01070555,
      "auxiliary_loss_mlp": 0.01033711,
      "balance_loss_clip": 1.03337026,
      "balance_loss_mlp": 1.02015126,
      "epoch": 0.8030061626333984,
      "flos": 49783320933120.0,
      "grad_norm": 2.075622788260489,
      "language_loss": 0.6896823,
      "learning_rate": 3.9334934137234235e-07,
      "loss": 0.71072495,
      "num_input_tokens_seen": 288249400,
      "step": 13356,
      "time_per_iteration": 2.8160178661346436
    },
    {
      "auxiliary_loss_clip": 0.01057821,
      "auxiliary_loss_mlp": 0.01027584,
      "balance_loss_clip": 1.03479362,
      "balance_loss_mlp": 1.01591933,
      "epoch": 0.8030662858860664,
      "flos": 21615243442560.0,
      "grad_norm": 1.5973810549953866,
      "language_loss": 0.77349156,
      "learning_rate": 3.931174316549666e-07,
      "loss": 0.79434562,
      "num_input_tokens_seen": 288268780,
      "step": 13357,
      "time_per_iteration": 2.597038984298706
    },
    {
      "auxiliary_loss_clip": 0.01068725,
      "auxiliary_loss_mlp": 0.01030049,
      "balance_loss_clip": 1.03242028,
      "balance_loss_mlp": 1.01715004,
      "epoch": 0.8031264091387345,
      "flos": 25630056351360.0,
      "grad_norm": 1.4646503966642257,
      "language_loss": 0.76954699,
      "learning_rate": 3.9288558287073937e-07,
      "loss": 0.79053473,
      "num_input_tokens_seen": 288290830,
      "step": 13358,
      "time_per_iteration": 2.6498491764068604
    },
    {
      "auxiliary_loss_clip": 0.01093022,
      "auxiliary_loss_mlp": 0.01029509,
      "balance_loss_clip": 1.03516746,
      "balance_loss_mlp": 1.01802373,
      "epoch": 0.8031865323914024,
      "flos": 19646225648640.0,
      "grad_norm": 1.7607498497660674,
      "language_loss": 0.84897596,
      "learning_rate": 3.9265379502845143e-07,
      "loss": 0.87020129,
      "num_input_tokens_seen": 288308865,
      "step": 13359,
      "time_per_iteration": 2.494602918624878
    },
    {
      "auxiliary_loss_clip": 0.01079035,
      "auxiliary_loss_mlp": 0.01026224,
      "balance_loss_clip": 1.03665912,
      "balance_loss_mlp": 1.01511943,
      "epoch": 0.8032466556440704,
      "flos": 26169110732160.0,
      "grad_norm": 1.7605318112990773,
      "language_loss": 0.73083961,
      "learning_rate": 3.924220681368928e-07,
      "loss": 0.75189215,
      "num_input_tokens_seen": 288327325,
      "step": 13360,
      "time_per_iteration": 2.584104061126709
    },
    {
      "auxiliary_loss_clip": 0.01105864,
      "auxiliary_loss_mlp": 0.01029618,
      "balance_loss_clip": 1.03630781,
      "balance_loss_mlp": 1.01803064,
      "epoch": 0.8033067788967383,
      "flos": 25520026014720.0,
      "grad_norm": 1.8986745463187606,
      "language_loss": 0.69647849,
      "learning_rate": 3.921904022048512e-07,
      "loss": 0.71783334,
      "num_input_tokens_seen": 288347285,
      "step": 13361,
      "time_per_iteration": 2.484532594680786
    },
    {
      "auxiliary_loss_clip": 0.01108124,
      "auxiliary_loss_mlp": 0.01035112,
      "balance_loss_clip": 1.03573632,
      "balance_loss_mlp": 1.02229059,
      "epoch": 0.8033669021494063,
      "flos": 24024274842240.0,
      "grad_norm": 2.6031135984086875,
      "language_loss": 0.70269454,
      "learning_rate": 3.919587972411098e-07,
      "loss": 0.72412688,
      "num_input_tokens_seen": 288367785,
      "step": 13362,
      "time_per_iteration": 2.5250771045684814
    },
    {
      "auxiliary_loss_clip": 0.01112728,
      "auxiliary_loss_mlp": 0.01037203,
      "balance_loss_clip": 1.03802633,
      "balance_loss_mlp": 1.02315986,
      "epoch": 0.8034270254020742,
      "flos": 13588059749760.0,
      "grad_norm": 2.284922339814391,
      "language_loss": 0.78394192,
      "learning_rate": 3.91727253254452e-07,
      "loss": 0.8054412,
      "num_input_tokens_seen": 288384135,
      "step": 13363,
      "time_per_iteration": 3.863267660140991
    },
    {
      "auxiliary_loss_clip": 0.01093295,
      "auxiliary_loss_mlp": 0.01028384,
      "balance_loss_clip": 1.03439474,
      "balance_loss_mlp": 1.01577783,
      "epoch": 0.8034871486547422,
      "flos": 27412661537280.0,
      "grad_norm": 1.9069995209118715,
      "language_loss": 0.74891394,
      "learning_rate": 3.9149577025365787e-07,
      "loss": 0.77013075,
      "num_input_tokens_seen": 288403805,
      "step": 13364,
      "time_per_iteration": 3.926029682159424
    },
    {
      "auxiliary_loss_clip": 0.01092619,
      "auxiliary_loss_mlp": 0.01028981,
      "balance_loss_clip": 1.03906858,
      "balance_loss_mlp": 1.01751339,
      "epoch": 0.8035472719074102,
      "flos": 32598593475840.0,
      "grad_norm": 2.0645978139304915,
      "language_loss": 0.60653353,
      "learning_rate": 3.9126434824750596e-07,
      "loss": 0.62774956,
      "num_input_tokens_seen": 288424895,
      "step": 13365,
      "time_per_iteration": 2.568396806716919
    },
    {
      "auxiliary_loss_clip": 0.01083699,
      "auxiliary_loss_mlp": 0.01032522,
      "balance_loss_clip": 1.03456807,
      "balance_loss_mlp": 1.01996303,
      "epoch": 0.8036073951600782,
      "flos": 21287989607040.0,
      "grad_norm": 27.646359262791382,
      "language_loss": 0.66233498,
      "learning_rate": 3.910329872447706e-07,
      "loss": 0.68349719,
      "num_input_tokens_seen": 288443865,
      "step": 13366,
      "time_per_iteration": 2.5014655590057373
    },
    {
      "auxiliary_loss_clip": 0.01102092,
      "auxiliary_loss_mlp": 0.01029092,
      "balance_loss_clip": 1.03492355,
      "balance_loss_mlp": 1.01702809,
      "epoch": 0.8036675184127461,
      "flos": 18113845582080.0,
      "grad_norm": 2.8297758421375376,
      "language_loss": 0.74963737,
      "learning_rate": 3.908016872542259e-07,
      "loss": 0.77094918,
      "num_input_tokens_seen": 288461065,
      "step": 13367,
      "time_per_iteration": 3.846749782562256
    },
    {
      "auxiliary_loss_clip": 0.0110262,
      "auxiliary_loss_mlp": 0.01027513,
      "balance_loss_clip": 1.03518891,
      "balance_loss_mlp": 1.01629496,
      "epoch": 0.8037276416654141,
      "flos": 26030280666240.0,
      "grad_norm": 1.4466072852806793,
      "language_loss": 0.73801064,
      "learning_rate": 3.905704482846428e-07,
      "loss": 0.75931191,
      "num_input_tokens_seen": 288481865,
      "step": 13368,
      "time_per_iteration": 2.5061404705047607
    },
    {
      "auxiliary_loss_clip": 0.0110566,
      "auxiliary_loss_mlp": 0.01031791,
      "balance_loss_clip": 1.03477788,
      "balance_loss_mlp": 1.01993585,
      "epoch": 0.803787764918082,
      "flos": 18802180886400.0,
      "grad_norm": 1.895948819279036,
      "language_loss": 0.69739908,
      "learning_rate": 3.90339270344789e-07,
      "loss": 0.7187736,
      "num_input_tokens_seen": 288499345,
      "step": 13369,
      "time_per_iteration": 2.429579973220825
    },
    {
      "auxiliary_loss_clip": 0.01082592,
      "auxiliary_loss_mlp": 0.01030337,
      "balance_loss_clip": 1.03410947,
      "balance_loss_mlp": 1.01908994,
      "epoch": 0.80384788817075,
      "flos": 20225787592320.0,
      "grad_norm": 1.648630609448983,
      "language_loss": 0.73570299,
      "learning_rate": 3.901081534434312e-07,
      "loss": 0.7568323,
      "num_input_tokens_seen": 288517660,
      "step": 13370,
      "time_per_iteration": 2.5294971466064453
    },
    {
      "auxiliary_loss_clip": 0.01082692,
      "auxiliary_loss_mlp": 0.0103218,
      "balance_loss_clip": 1.0339241,
      "balance_loss_mlp": 1.01873326,
      "epoch": 0.8039080114234181,
      "flos": 18515290959360.0,
      "grad_norm": 2.874213982351197,
      "language_loss": 0.87185812,
      "learning_rate": 3.898770975893342e-07,
      "loss": 0.89300686,
      "num_input_tokens_seen": 288534180,
      "step": 13371,
      "time_per_iteration": 2.496948003768921
    },
    {
      "auxiliary_loss_clip": 0.01097101,
      "auxiliary_loss_mlp": 0.010292,
      "balance_loss_clip": 1.03483593,
      "balance_loss_mlp": 1.01584244,
      "epoch": 0.803968134676086,
      "flos": 22382510883840.0,
      "grad_norm": 2.5891342049582806,
      "language_loss": 0.74791694,
      "learning_rate": 3.89646102791259e-07,
      "loss": 0.76917994,
      "num_input_tokens_seen": 288553350,
      "step": 13372,
      "time_per_iteration": 2.491957664489746
    },
    {
      "auxiliary_loss_clip": 0.01063845,
      "auxiliary_loss_mlp": 0.01029084,
      "balance_loss_clip": 1.03591871,
      "balance_loss_mlp": 1.01587534,
      "epoch": 0.804028257928754,
      "flos": 23842566915840.0,
      "grad_norm": 2.134721906471902,
      "language_loss": 0.79230094,
      "learning_rate": 3.894151690579646e-07,
      "loss": 0.81323022,
      "num_input_tokens_seen": 288571325,
      "step": 13373,
      "time_per_iteration": 3.9948012828826904
    },
    {
      "auxiliary_loss_clip": 0.01080058,
      "auxiliary_loss_mlp": 0.01034594,
      "balance_loss_clip": 1.03522956,
      "balance_loss_mlp": 1.0230248,
      "epoch": 0.8040883811814219,
      "flos": 23550720912000.0,
      "grad_norm": 1.7751817690024492,
      "language_loss": 0.74281162,
      "learning_rate": 3.8918429639820815e-07,
      "loss": 0.76395816,
      "num_input_tokens_seen": 288592100,
      "step": 13374,
      "time_per_iteration": 2.554072618484497
    },
    {
      "auxiliary_loss_clip": 0.01057047,
      "auxiliary_loss_mlp": 0.01039245,
      "balance_loss_clip": 1.03215492,
      "balance_loss_mlp": 1.02492189,
      "epoch": 0.8041485044340899,
      "flos": 19026263882880.0,
      "grad_norm": 1.9378163270766697,
      "language_loss": 0.68572056,
      "learning_rate": 3.889534848207452e-07,
      "loss": 0.70668352,
      "num_input_tokens_seen": 288612305,
      "step": 13375,
      "time_per_iteration": 2.5922799110412598
    },
    {
      "auxiliary_loss_clip": 0.01010982,
      "auxiliary_loss_mlp": 0.01003688,
      "balance_loss_clip": 1.01783705,
      "balance_loss_mlp": 1.00241232,
      "epoch": 0.8042086276867578,
      "flos": 70005663797760.0,
      "grad_norm": 0.7253296602307644,
      "language_loss": 0.55664819,
      "learning_rate": 3.887227343343271e-07,
      "loss": 0.57679492,
      "num_input_tokens_seen": 288676015,
      "step": 13376,
      "time_per_iteration": 3.249617576599121
    },
    {
      "auxiliary_loss_clip": 0.0106207,
      "auxiliary_loss_mlp": 0.01033614,
      "balance_loss_clip": 1.03444219,
      "balance_loss_mlp": 1.02008986,
      "epoch": 0.8042687509394258,
      "flos": 21872435800320.0,
      "grad_norm": 1.638577443839273,
      "language_loss": 0.72976846,
      "learning_rate": 3.8849204494770425e-07,
      "loss": 0.75072527,
      "num_input_tokens_seen": 288696455,
      "step": 13377,
      "time_per_iteration": 2.6367011070251465
    },
    {
      "auxiliary_loss_clip": 0.01091014,
      "auxiliary_loss_mlp": 0.0102918,
      "balance_loss_clip": 1.03343606,
      "balance_loss_mlp": 1.01675868,
      "epoch": 0.8043288741920938,
      "flos": 26614870513920.0,
      "grad_norm": 1.9333208146645127,
      "language_loss": 0.70256734,
      "learning_rate": 3.8826141666962567e-07,
      "loss": 0.72376931,
      "num_input_tokens_seen": 288715560,
      "step": 13378,
      "time_per_iteration": 2.528536319732666
    },
    {
      "auxiliary_loss_clip": 0.01097237,
      "auxiliary_loss_mlp": 0.01028914,
      "balance_loss_clip": 1.0367986,
      "balance_loss_mlp": 1.01628399,
      "epoch": 0.8043889974447618,
      "flos": 33403387651200.0,
      "grad_norm": 1.5888929887623413,
      "language_loss": 0.69366574,
      "learning_rate": 3.880308495088347e-07,
      "loss": 0.7149272,
      "num_input_tokens_seen": 288739485,
      "step": 13379,
      "time_per_iteration": 2.6258974075317383
    },
    {
      "auxiliary_loss_clip": 0.01110051,
      "auxiliary_loss_mlp": 0.01031791,
      "balance_loss_clip": 1.03821445,
      "balance_loss_mlp": 1.01854134,
      "epoch": 0.8044491206974297,
      "flos": 20375966355840.0,
      "grad_norm": 2.54464821124022,
      "language_loss": 0.76279229,
      "learning_rate": 3.8780034347407533e-07,
      "loss": 0.78421068,
      "num_input_tokens_seen": 288757420,
      "step": 13380,
      "time_per_iteration": 2.4461846351623535
    },
    {
      "auxiliary_loss_clip": 0.01058091,
      "auxiliary_loss_mlp": 0.01026033,
      "balance_loss_clip": 1.0330379,
      "balance_loss_mlp": 1.014768,
      "epoch": 0.8045092439500977,
      "flos": 23403810286080.0,
      "grad_norm": 2.5703430379826204,
      "language_loss": 0.69191515,
      "learning_rate": 3.875698985740887e-07,
      "loss": 0.71275634,
      "num_input_tokens_seen": 288775535,
      "step": 13381,
      "time_per_iteration": 2.6355161666870117
    },
    {
      "auxiliary_loss_clip": 0.01096964,
      "auxiliary_loss_mlp": 0.01035413,
      "balance_loss_clip": 1.03715158,
      "balance_loss_mlp": 1.02312279,
      "epoch": 0.8045693672027656,
      "flos": 24097245321600.0,
      "grad_norm": 1.7927408818103339,
      "language_loss": 0.6381799,
      "learning_rate": 3.873395148176135e-07,
      "loss": 0.65950364,
      "num_input_tokens_seen": 288795035,
      "step": 13382,
      "time_per_iteration": 2.5027074813842773
    },
    {
      "auxiliary_loss_clip": 0.01083104,
      "auxiliary_loss_mlp": 0.01032505,
      "balance_loss_clip": 1.03678763,
      "balance_loss_mlp": 1.02120996,
      "epoch": 0.8046294904554336,
      "flos": 27707165147520.0,
      "grad_norm": 7.044418196142844,
      "language_loss": 0.7652936,
      "learning_rate": 3.8710919221338487e-07,
      "loss": 0.78644967,
      "num_input_tokens_seen": 288816270,
      "step": 13383,
      "time_per_iteration": 2.590162515640259
    },
    {
      "auxiliary_loss_clip": 0.01090732,
      "auxiliary_loss_mlp": 0.01041219,
      "balance_loss_clip": 1.03546405,
      "balance_loss_mlp": 1.02777815,
      "epoch": 0.8046896137081017,
      "flos": 24972998814720.0,
      "grad_norm": 2.2426722541602966,
      "language_loss": 0.69919097,
      "learning_rate": 3.868789307701381e-07,
      "loss": 0.72051048,
      "num_input_tokens_seen": 288836050,
      "step": 13384,
      "time_per_iteration": 2.4977378845214844
    },
    {
      "auxiliary_loss_clip": 0.01094484,
      "auxiliary_loss_mlp": 0.0103708,
      "balance_loss_clip": 1.03389359,
      "balance_loss_mlp": 1.02413428,
      "epoch": 0.8047497369607696,
      "flos": 17675484001920.0,
      "grad_norm": 2.6131138455864726,
      "language_loss": 0.79030651,
      "learning_rate": 3.8664873049660375e-07,
      "loss": 0.81162214,
      "num_input_tokens_seen": 288852900,
      "step": 13385,
      "time_per_iteration": 2.4689950942993164
    },
    {
      "auxiliary_loss_clip": 0.01105687,
      "auxiliary_loss_mlp": 0.01034181,
      "balance_loss_clip": 1.03552186,
      "balance_loss_mlp": 1.02172971,
      "epoch": 0.8048098602134376,
      "flos": 22382079920640.0,
      "grad_norm": 1.7381334569539308,
      "language_loss": 0.72058332,
      "learning_rate": 3.864185914015108e-07,
      "loss": 0.74198198,
      "num_input_tokens_seen": 288872625,
      "step": 13386,
      "time_per_iteration": 2.4455037117004395
    },
    {
      "auxiliary_loss_clip": 0.01009391,
      "auxiliary_loss_mlp": 0.01003217,
      "balance_loss_clip": 1.00914633,
      "balance_loss_mlp": 1.00205505,
      "epoch": 0.8048699834661055,
      "flos": 71200949702400.0,
      "grad_norm": 0.6736834872445414,
      "language_loss": 0.51287585,
      "learning_rate": 3.861885134935865e-07,
      "loss": 0.53300196,
      "num_input_tokens_seen": 288939180,
      "step": 13387,
      "time_per_iteration": 3.19726300239563
    },
    {
      "auxiliary_loss_clip": 0.0110542,
      "auxiliary_loss_mlp": 0.01032369,
      "balance_loss_clip": 1.0349648,
      "balance_loss_mlp": 1.01874375,
      "epoch": 0.8049301067187735,
      "flos": 23660320285440.0,
      "grad_norm": 1.7523593498265764,
      "language_loss": 0.73828721,
      "learning_rate": 3.859584967815559e-07,
      "loss": 0.75966513,
      "num_input_tokens_seen": 288958925,
      "step": 13388,
      "time_per_iteration": 2.4600484371185303
    },
    {
      "auxiliary_loss_clip": 0.01073292,
      "auxiliary_loss_mlp": 0.01027567,
      "balance_loss_clip": 1.03588259,
      "balance_loss_mlp": 1.01605117,
      "epoch": 0.8049902299714414,
      "flos": 24426330750720.0,
      "grad_norm": 2.0910349843915954,
      "language_loss": 0.71611261,
      "learning_rate": 3.857285412741411e-07,
      "loss": 0.73712122,
      "num_input_tokens_seen": 288980935,
      "step": 13389,
      "time_per_iteration": 2.6151931285858154
    },
    {
      "auxiliary_loss_clip": 0.01079584,
      "auxiliary_loss_mlp": 0.0103508,
      "balance_loss_clip": 1.03768218,
      "balance_loss_mlp": 1.02238369,
      "epoch": 0.8050503532241094,
      "flos": 17492626840320.0,
      "grad_norm": 2.262386953923954,
      "language_loss": 0.82705092,
      "learning_rate": 3.8549864698006097e-07,
      "loss": 0.84819758,
      "num_input_tokens_seen": 288996780,
      "step": 13390,
      "time_per_iteration": 2.4888815879821777
    },
    {
      "auxiliary_loss_clip": 0.01022125,
      "auxiliary_loss_mlp": 0.0100259,
      "balance_loss_clip": 1.00812674,
      "balance_loss_mlp": 1.00140381,
      "epoch": 0.8051104764767774,
      "flos": 57658030369920.0,
      "grad_norm": 0.7756338311084859,
      "language_loss": 0.55528748,
      "learning_rate": 3.8526881390803424e-07,
      "loss": 0.57553464,
      "num_input_tokens_seen": 289057590,
      "step": 13391,
      "time_per_iteration": 3.0758655071258545
    },
    {
      "auxiliary_loss_clip": 0.01091345,
      "auxiliary_loss_mlp": 0.01032273,
      "balance_loss_clip": 1.03579521,
      "balance_loss_mlp": 1.02056086,
      "epoch": 0.8051705997294454,
      "flos": 18003456109440.0,
      "grad_norm": 1.5264551630611596,
      "language_loss": 0.84814405,
      "learning_rate": 3.850390420667762e-07,
      "loss": 0.86938024,
      "num_input_tokens_seen": 289076285,
      "step": 13392,
      "time_per_iteration": 2.464813470840454
    },
    {
      "auxiliary_loss_clip": 0.01071361,
      "auxiliary_loss_mlp": 0.01026384,
      "balance_loss_clip": 1.03320336,
      "balance_loss_mlp": 1.01491046,
      "epoch": 0.8052307229821133,
      "flos": 26397754755840.0,
      "grad_norm": 1.4276030921994185,
      "language_loss": 0.70267677,
      "learning_rate": 3.8480933146499914e-07,
      "loss": 0.72365421,
      "num_input_tokens_seen": 289097585,
      "step": 13393,
      "time_per_iteration": 2.6039328575134277
    },
    {
      "auxiliary_loss_clip": 0.01095367,
      "auxiliary_loss_mlp": 0.01030987,
      "balance_loss_clip": 1.03550065,
      "balance_loss_mlp": 1.01820743,
      "epoch": 0.8052908462347813,
      "flos": 21757018423680.0,
      "grad_norm": 1.9576355407722206,
      "language_loss": 0.76198173,
      "learning_rate": 3.84579682111414e-07,
      "loss": 0.78324521,
      "num_input_tokens_seen": 289116890,
      "step": 13394,
      "time_per_iteration": 2.480156421661377
    },
    {
      "auxiliary_loss_clip": 0.01107688,
      "auxiliary_loss_mlp": 0.01030641,
      "balance_loss_clip": 1.03826773,
      "balance_loss_mlp": 1.01890469,
      "epoch": 0.8053509694874492,
      "flos": 25442279026560.0,
      "grad_norm": 1.6707155604193806,
      "language_loss": 0.648853,
      "learning_rate": 3.843500940147304e-07,
      "loss": 0.67023635,
      "num_input_tokens_seen": 289136670,
      "step": 13395,
      "time_per_iteration": 2.503253936767578
    },
    {
      "auxiliary_loss_clip": 0.01021941,
      "auxiliary_loss_mlp": 0.01003154,
      "balance_loss_clip": 1.00826669,
      "balance_loss_mlp": 1.00198007,
      "epoch": 0.8054110927401172,
      "flos": 57668122091520.0,
      "grad_norm": 0.7620429408536284,
      "language_loss": 0.57322359,
      "learning_rate": 3.8412056718365206e-07,
      "loss": 0.59347451,
      "num_input_tokens_seen": 289200150,
      "step": 13396,
      "time_per_iteration": 3.2477612495422363
    },
    {
      "auxiliary_loss_clip": 0.01095375,
      "auxiliary_loss_mlp": 0.01037161,
      "balance_loss_clip": 1.0359447,
      "balance_loss_mlp": 1.02401209,
      "epoch": 0.8054712159927853,
      "flos": 19276201693440.0,
      "grad_norm": 1.6559970727412714,
      "language_loss": 0.77517152,
      "learning_rate": 3.8389110162688353e-07,
      "loss": 0.79649687,
      "num_input_tokens_seen": 289218125,
      "step": 13397,
      "time_per_iteration": 2.4827191829681396
    },
    {
      "auxiliary_loss_clip": 0.01090485,
      "auxiliary_loss_mlp": 0.01027295,
      "balance_loss_clip": 1.03821337,
      "balance_loss_mlp": 1.01590443,
      "epoch": 0.8055313392454532,
      "flos": 17967617314560.0,
      "grad_norm": 1.74114740982197,
      "language_loss": 0.70399547,
      "learning_rate": 3.836616973531266e-07,
      "loss": 0.72517329,
      "num_input_tokens_seen": 289237115,
      "step": 13398,
      "time_per_iteration": 2.4957783222198486
    },
    {
      "auxiliary_loss_clip": 0.01082653,
      "auxiliary_loss_mlp": 0.01028805,
      "balance_loss_clip": 1.03432429,
      "balance_loss_mlp": 1.01742029,
      "epoch": 0.8055914624981212,
      "flos": 13478352635520.0,
      "grad_norm": 3.108877753155029,
      "language_loss": 0.69269228,
      "learning_rate": 3.834323543710805e-07,
      "loss": 0.71380687,
      "num_input_tokens_seen": 289253635,
      "step": 13399,
      "time_per_iteration": 2.4747018814086914
    },
    {
      "auxiliary_loss_clip": 0.01105769,
      "auxiliary_loss_mlp": 0.01036363,
      "balance_loss_clip": 1.03641081,
      "balance_loss_mlp": 1.0247519,
      "epoch": 0.8056515857507891,
      "flos": 13224787551360.0,
      "grad_norm": 2.086437665809118,
      "language_loss": 0.72065437,
      "learning_rate": 3.8320307268944153e-07,
      "loss": 0.74207562,
      "num_input_tokens_seen": 289270085,
      "step": 13400,
      "time_per_iteration": 2.4518284797668457
    },
    {
      "auxiliary_loss_clip": 0.0109153,
      "auxiliary_loss_mlp": 0.01030844,
      "balance_loss_clip": 1.03255856,
      "balance_loss_mlp": 1.01833332,
      "epoch": 0.8057117090034571,
      "flos": 23878190229120.0,
      "grad_norm": 3.1195764920304203,
      "language_loss": 0.63832927,
      "learning_rate": 3.829738523169037e-07,
      "loss": 0.65955305,
      "num_input_tokens_seen": 289289645,
      "step": 13401,
      "time_per_iteration": 2.506031036376953
    },
    {
      "auxiliary_loss_clip": 0.01093513,
      "auxiliary_loss_mlp": 0.01032565,
      "balance_loss_clip": 1.03403246,
      "balance_loss_mlp": 1.02068567,
      "epoch": 0.805771832256125,
      "flos": 21214300855680.0,
      "grad_norm": 2.0413766331904926,
      "language_loss": 0.83736944,
      "learning_rate": 3.8274469326215985e-07,
      "loss": 0.85863018,
      "num_input_tokens_seen": 289306630,
      "step": 13402,
      "time_per_iteration": 3.88748836517334
    },
    {
      "auxiliary_loss_clip": 0.01054202,
      "auxiliary_loss_mlp": 0.01031427,
      "balance_loss_clip": 1.03641808,
      "balance_loss_mlp": 1.01970243,
      "epoch": 0.805831955508793,
      "flos": 17566818382080.0,
      "grad_norm": 2.7062480387105357,
      "language_loss": 0.68047547,
      "learning_rate": 3.8251559553389876e-07,
      "loss": 0.70133173,
      "num_input_tokens_seen": 289324960,
      "step": 13403,
      "time_per_iteration": 4.051624536514282
    },
    {
      "auxiliary_loss_clip": 0.01067137,
      "auxiliary_loss_mlp": 0.00783587,
      "balance_loss_clip": 1.03438258,
      "balance_loss_mlp": 1.00909245,
      "epoch": 0.805892078761461,
      "flos": 26907542530560.0,
      "grad_norm": 1.7365389480847733,
      "language_loss": 0.84814215,
      "learning_rate": 3.822865591408084e-07,
      "loss": 0.86664939,
      "num_input_tokens_seen": 289344980,
      "step": 13404,
      "time_per_iteration": 2.5914323329925537
    },
    {
      "auxiliary_loss_clip": 0.01069097,
      "auxiliary_loss_mlp": 0.01030865,
      "balance_loss_clip": 1.03432369,
      "balance_loss_mlp": 1.01997542,
      "epoch": 0.805952202014129,
      "flos": 31506442496640.0,
      "grad_norm": 1.464944311347173,
      "language_loss": 0.70487094,
      "learning_rate": 3.820575840915743e-07,
      "loss": 0.72587055,
      "num_input_tokens_seen": 289367500,
      "step": 13405,
      "time_per_iteration": 2.6530344486236572
    },
    {
      "auxiliary_loss_clip": 0.01093679,
      "auxiliary_loss_mlp": 0.01025353,
      "balance_loss_clip": 1.03617704,
      "balance_loss_mlp": 1.01398087,
      "epoch": 0.8060123252667969,
      "flos": 24389953251840.0,
      "grad_norm": 2.8450230937305396,
      "language_loss": 0.75031948,
      "learning_rate": 3.818286703948788e-07,
      "loss": 0.77150983,
      "num_input_tokens_seen": 289385930,
      "step": 13406,
      "time_per_iteration": 3.9375033378601074
    },
    {
      "auxiliary_loss_clip": 0.01096299,
      "auxiliary_loss_mlp": 0.01034288,
      "balance_loss_clip": 1.03672886,
      "balance_loss_mlp": 1.02162194,
      "epoch": 0.8060724485194649,
      "flos": 23479941162240.0,
      "grad_norm": 1.4488799515543587,
      "language_loss": 0.76265389,
      "learning_rate": 3.815998180594018e-07,
      "loss": 0.78395975,
      "num_input_tokens_seen": 289408025,
      "step": 13407,
      "time_per_iteration": 2.549997091293335
    },
    {
      "auxiliary_loss_clip": 0.01076352,
      "auxiliary_loss_mlp": 0.00785515,
      "balance_loss_clip": 1.03316903,
      "balance_loss_mlp": 1.01067543,
      "epoch": 0.8061325717721328,
      "flos": 18624495283200.0,
      "grad_norm": 1.8418156486840802,
      "language_loss": 0.73281717,
      "learning_rate": 3.81371027093822e-07,
      "loss": 0.75143588,
      "num_input_tokens_seen": 289426575,
      "step": 13408,
      "time_per_iteration": 2.50502347946167
    },
    {
      "auxiliary_loss_clip": 0.01076618,
      "auxiliary_loss_mlp": 0.010338,
      "balance_loss_clip": 1.03589356,
      "balance_loss_mlp": 1.02006698,
      "epoch": 0.8061926950248008,
      "flos": 23582752865280.0,
      "grad_norm": 1.918956520883007,
      "language_loss": 0.70756918,
      "learning_rate": 3.8114229750681523e-07,
      "loss": 0.72867334,
      "num_input_tokens_seen": 289447760,
      "step": 13409,
      "time_per_iteration": 2.546689748764038
    },
    {
      "auxiliary_loss_clip": 0.01104477,
      "auxiliary_loss_mlp": 0.01025187,
      "balance_loss_clip": 1.03469849,
      "balance_loss_mlp": 1.01278317,
      "epoch": 0.8062528182774689,
      "flos": 11143333209600.0,
      "grad_norm": 2.0672452244505677,
      "language_loss": 0.77130526,
      "learning_rate": 3.809136293070545e-07,
      "loss": 0.79260194,
      "num_input_tokens_seen": 289463920,
      "step": 13410,
      "time_per_iteration": 2.4543099403381348
    },
    {
      "auxiliary_loss_clip": 0.01093148,
      "auxiliary_loss_mlp": 0.0103217,
      "balance_loss_clip": 1.03596997,
      "balance_loss_mlp": 1.01992154,
      "epoch": 0.8063129415301368,
      "flos": 22346815743360.0,
      "grad_norm": 2.2403458045294555,
      "language_loss": 0.6840862,
      "learning_rate": 3.806850225032117e-07,
      "loss": 0.70533943,
      "num_input_tokens_seen": 289482635,
      "step": 13411,
      "time_per_iteration": 2.494877815246582
    },
    {
      "auxiliary_loss_clip": 0.01071203,
      "auxiliary_loss_mlp": 0.01028131,
      "balance_loss_clip": 1.03417981,
      "balance_loss_mlp": 1.01596546,
      "epoch": 0.8063730647828048,
      "flos": 23988400133760.0,
      "grad_norm": 2.1307343396564584,
      "language_loss": 0.68136013,
      "learning_rate": 3.804564771039551e-07,
      "loss": 0.70235348,
      "num_input_tokens_seen": 289502040,
      "step": 13412,
      "time_per_iteration": 3.9720804691314697
    },
    {
      "auxiliary_loss_clip": 0.010962,
      "auxiliary_loss_mlp": 0.01033455,
      "balance_loss_clip": 1.03923547,
      "balance_loss_mlp": 1.01978803,
      "epoch": 0.8064331880354727,
      "flos": 21321494017920.0,
      "grad_norm": 1.5952779005573559,
      "language_loss": 0.81734228,
      "learning_rate": 3.8022799311795064e-07,
      "loss": 0.83863884,
      "num_input_tokens_seen": 289520740,
      "step": 13413,
      "time_per_iteration": 2.4875619411468506
    },
    {
      "auxiliary_loss_clip": 0.01089353,
      "auxiliary_loss_mlp": 0.01041897,
      "balance_loss_clip": 1.03431118,
      "balance_loss_mlp": 1.02831316,
      "epoch": 0.8064933112881407,
      "flos": 19682890456320.0,
      "grad_norm": 1.9267441179905571,
      "language_loss": 0.8497321,
      "learning_rate": 3.7999957055386303e-07,
      "loss": 0.87104464,
      "num_input_tokens_seen": 289535840,
      "step": 13414,
      "time_per_iteration": 2.493429183959961
    },
    {
      "auxiliary_loss_clip": 0.01079702,
      "auxiliary_loss_mlp": 0.01031093,
      "balance_loss_clip": 1.03368402,
      "balance_loss_mlp": 1.01935101,
      "epoch": 0.8065534345408086,
      "flos": 19279721226240.0,
      "grad_norm": 1.8762615920537162,
      "language_loss": 0.67145783,
      "learning_rate": 3.7977120942035467e-07,
      "loss": 0.69256574,
      "num_input_tokens_seen": 289555205,
      "step": 13415,
      "time_per_iteration": 2.5439505577087402
    },
    {
      "auxiliary_loss_clip": 0.01068708,
      "auxiliary_loss_mlp": 0.01024117,
      "balance_loss_clip": 1.03347218,
      "balance_loss_mlp": 1.01267314,
      "epoch": 0.8066135577934767,
      "flos": 19677718897920.0,
      "grad_norm": 1.7666236406041032,
      "language_loss": 0.76289964,
      "learning_rate": 3.7954290972608383e-07,
      "loss": 0.7838279,
      "num_input_tokens_seen": 289573000,
      "step": 13416,
      "time_per_iteration": 2.537405252456665
    },
    {
      "auxiliary_loss_clip": 0.01091277,
      "auxiliary_loss_mlp": 0.01033355,
      "balance_loss_clip": 1.03379202,
      "balance_loss_mlp": 1.02090335,
      "epoch": 0.8066736810461446,
      "flos": 21143592933120.0,
      "grad_norm": 1.4748844327515631,
      "language_loss": 0.65509415,
      "learning_rate": 3.793146714797086e-07,
      "loss": 0.67634046,
      "num_input_tokens_seen": 289592625,
      "step": 13417,
      "time_per_iteration": 2.491903305053711
    },
    {
      "auxiliary_loss_clip": 0.01072642,
      "auxiliary_loss_mlp": 0.01048617,
      "balance_loss_clip": 1.03477871,
      "balance_loss_mlp": 1.03518796,
      "epoch": 0.8067338042988126,
      "flos": 22598261925120.0,
      "grad_norm": 1.575936612542222,
      "language_loss": 0.80540264,
      "learning_rate": 3.7908649468988306e-07,
      "loss": 0.82661521,
      "num_input_tokens_seen": 289610780,
      "step": 13418,
      "time_per_iteration": 2.575634002685547
    },
    {
      "auxiliary_loss_clip": 0.01085257,
      "auxiliary_loss_mlp": 0.0102928,
      "balance_loss_clip": 1.03670061,
      "balance_loss_mlp": 1.01654887,
      "epoch": 0.8067939275514805,
      "flos": 16508423208960.0,
      "grad_norm": 1.6115370315105824,
      "language_loss": 0.85017633,
      "learning_rate": 3.7885837936526066e-07,
      "loss": 0.87132174,
      "num_input_tokens_seen": 289628890,
      "step": 13419,
      "time_per_iteration": 2.4832308292388916
    },
    {
      "auxiliary_loss_clip": 0.01074181,
      "auxiliary_loss_mlp": 0.0078695,
      "balance_loss_clip": 1.03498042,
      "balance_loss_mlp": 1.01431358,
      "epoch": 0.8068540508041485,
      "flos": 28541836460160.0,
      "grad_norm": 1.5995880353382983,
      "language_loss": 0.75751603,
      "learning_rate": 3.7863032551449047e-07,
      "loss": 0.77612734,
      "num_input_tokens_seen": 289647220,
      "step": 13420,
      "time_per_iteration": 2.6008808612823486
    },
    {
      "auxiliary_loss_clip": 0.01085227,
      "auxiliary_loss_mlp": 0.00783162,
      "balance_loss_clip": 1.03190923,
      "balance_loss_mlp": 1.00888848,
      "epoch": 0.8069141740568164,
      "flos": 21652482867840.0,
      "grad_norm": 1.6855881538406192,
      "language_loss": 0.78622478,
      "learning_rate": 3.784023331462207e-07,
      "loss": 0.80490869,
      "num_input_tokens_seen": 289665800,
      "step": 13421,
      "time_per_iteration": 2.524552583694458
    },
    {
      "auxiliary_loss_clip": 0.01074514,
      "auxiliary_loss_mlp": 0.01025281,
      "balance_loss_clip": 1.03675735,
      "balance_loss_mlp": 1.01319885,
      "epoch": 0.8069742973094844,
      "flos": 17529327561600.0,
      "grad_norm": 1.683589976901725,
      "language_loss": 0.80048919,
      "learning_rate": 3.78174402269098e-07,
      "loss": 0.82148719,
      "num_input_tokens_seen": 289682705,
      "step": 13422,
      "time_per_iteration": 2.5435802936553955
    },
    {
      "auxiliary_loss_clip": 0.01103393,
      "auxiliary_loss_mlp": 0.01031193,
      "balance_loss_clip": 1.03422213,
      "balance_loss_mlp": 1.0193913,
      "epoch": 0.8070344205621525,
      "flos": 23367037737600.0,
      "grad_norm": 1.6750883488668717,
      "language_loss": 0.68208474,
      "learning_rate": 3.7794653289176347e-07,
      "loss": 0.70343059,
      "num_input_tokens_seen": 289702920,
      "step": 13423,
      "time_per_iteration": 2.459404468536377
    },
    {
      "auxiliary_loss_clip": 0.01080116,
      "auxiliary_loss_mlp": 0.01035992,
      "balance_loss_clip": 1.03597116,
      "balance_loss_mlp": 1.0233674,
      "epoch": 0.8070945438148204,
      "flos": 22930184528640.0,
      "grad_norm": 1.7268804224316383,
      "language_loss": 0.80255085,
      "learning_rate": 3.7771872502285904e-07,
      "loss": 0.82371187,
      "num_input_tokens_seen": 289723280,
      "step": 13424,
      "time_per_iteration": 2.550837516784668
    },
    {
      "auxiliary_loss_clip": 0.01094859,
      "auxiliary_loss_mlp": 0.0102862,
      "balance_loss_clip": 1.03429639,
      "balance_loss_mlp": 1.01643729,
      "epoch": 0.8071546670674884,
      "flos": 25300683613440.0,
      "grad_norm": 1.4662367875856352,
      "language_loss": 0.78841656,
      "learning_rate": 3.774909786710232e-07,
      "loss": 0.80965137,
      "num_input_tokens_seen": 289743475,
      "step": 13425,
      "time_per_iteration": 2.5116984844207764
    },
    {
      "auxiliary_loss_clip": 0.01073018,
      "auxiliary_loss_mlp": 0.01031133,
      "balance_loss_clip": 1.03331006,
      "balance_loss_mlp": 1.01942658,
      "epoch": 0.8072147903201563,
      "flos": 18113701927680.0,
      "grad_norm": 2.411618764588188,
      "language_loss": 0.75675893,
      "learning_rate": 3.772632938448923e-07,
      "loss": 0.77780038,
      "num_input_tokens_seen": 289761400,
      "step": 13426,
      "time_per_iteration": 2.4966671466827393
    },
    {
      "auxiliary_loss_clip": 0.01093396,
      "auxiliary_loss_mlp": 0.01023491,
      "balance_loss_clip": 1.03468275,
      "balance_loss_mlp": 1.01238632,
      "epoch": 0.8072749135728243,
      "flos": 26688164215680.0,
      "grad_norm": 1.693231110991996,
      "language_loss": 0.72957766,
      "learning_rate": 3.770356705530997e-07,
      "loss": 0.75074655,
      "num_input_tokens_seen": 289781025,
      "step": 13427,
      "time_per_iteration": 2.5264010429382324
    },
    {
      "auxiliary_loss_clip": 0.01055804,
      "auxiliary_loss_mlp": 0.01033974,
      "balance_loss_clip": 1.03648925,
      "balance_loss_mlp": 1.02065885,
      "epoch": 0.8073350368254922,
      "flos": 19240291071360.0,
      "grad_norm": 1.7064019518094282,
      "language_loss": 0.70047659,
      "learning_rate": 3.768081088042774e-07,
      "loss": 0.72137439,
      "num_input_tokens_seen": 289798380,
      "step": 13428,
      "time_per_iteration": 2.6026830673217773
    },
    {
      "auxiliary_loss_clip": 0.01082746,
      "auxiliary_loss_mlp": 0.0102687,
      "balance_loss_clip": 1.0331459,
      "balance_loss_mlp": 1.01543772,
      "epoch": 0.8073951600781603,
      "flos": 13334530579200.0,
      "grad_norm": 1.8045441457882236,
      "language_loss": 0.75076199,
      "learning_rate": 3.765806086070544e-07,
      "loss": 0.77185816,
      "num_input_tokens_seen": 289814515,
      "step": 13429,
      "time_per_iteration": 2.482254981994629
    },
    {
      "auxiliary_loss_clip": 0.01091627,
      "auxiliary_loss_mlp": 0.01029735,
      "balance_loss_clip": 1.03545237,
      "balance_loss_mlp": 1.0181601,
      "epoch": 0.8074552833308282,
      "flos": 22853191726080.0,
      "grad_norm": 2.768916179194125,
      "language_loss": 0.67096603,
      "learning_rate": 3.763531699700568e-07,
      "loss": 0.69217968,
      "num_input_tokens_seen": 289834315,
      "step": 13430,
      "time_per_iteration": 2.5008552074432373
    },
    {
      "auxiliary_loss_clip": 0.01066664,
      "auxiliary_loss_mlp": 0.01027877,
      "balance_loss_clip": 1.03568578,
      "balance_loss_mlp": 1.01611114,
      "epoch": 0.8075154065834962,
      "flos": 20339409288960.0,
      "grad_norm": 1.698210724412682,
      "language_loss": 0.80346692,
      "learning_rate": 3.7612579290190994e-07,
      "loss": 0.82441235,
      "num_input_tokens_seen": 289853770,
      "step": 13431,
      "time_per_iteration": 2.542090892791748
    },
    {
      "auxiliary_loss_clip": 0.01080042,
      "auxiliary_loss_mlp": 0.01029153,
      "balance_loss_clip": 1.03427076,
      "balance_loss_mlp": 1.01626039,
      "epoch": 0.8075755298361641,
      "flos": 21908059113600.0,
      "grad_norm": 2.1391201367333386,
      "language_loss": 0.80411553,
      "learning_rate": 3.7589847741123593e-07,
      "loss": 0.82520747,
      "num_input_tokens_seen": 289870480,
      "step": 13432,
      "time_per_iteration": 2.5037527084350586
    },
    {
      "auxiliary_loss_clip": 0.01080012,
      "auxiliary_loss_mlp": 0.0103309,
      "balance_loss_clip": 1.03870082,
      "balance_loss_mlp": 1.01999521,
      "epoch": 0.8076356530888321,
      "flos": 15669298609920.0,
      "grad_norm": 1.8506030304519465,
      "language_loss": 0.70320427,
      "learning_rate": 3.7567122350665415e-07,
      "loss": 0.72433531,
      "num_input_tokens_seen": 289888275,
      "step": 13433,
      "time_per_iteration": 2.522468090057373
    },
    {
      "auxiliary_loss_clip": 0.01079302,
      "auxiliary_loss_mlp": 0.01028595,
      "balance_loss_clip": 1.03525591,
      "balance_loss_mlp": 1.01657331,
      "epoch": 0.8076957763415,
      "flos": 37777414521600.0,
      "grad_norm": 1.6233841021150393,
      "language_loss": 0.72751486,
      "learning_rate": 3.754440311967828e-07,
      "loss": 0.74859381,
      "num_input_tokens_seen": 289911495,
      "step": 13434,
      "time_per_iteration": 2.661424398422241
    },
    {
      "auxiliary_loss_clip": 0.01067776,
      "auxiliary_loss_mlp": 0.01028668,
      "balance_loss_clip": 1.03768873,
      "balance_loss_mlp": 1.01646113,
      "epoch": 0.807755899594168,
      "flos": 19610781903360.0,
      "grad_norm": 4.08174125061118,
      "language_loss": 0.68102163,
      "learning_rate": 3.752169004902361e-07,
      "loss": 0.70198607,
      "num_input_tokens_seen": 289930045,
      "step": 13435,
      "time_per_iteration": 2.5333874225616455
    },
    {
      "auxiliary_loss_clip": 0.01064306,
      "auxiliary_loss_mlp": 0.0103321,
      "balance_loss_clip": 1.03654099,
      "balance_loss_mlp": 1.01864266,
      "epoch": 0.8078160228468361,
      "flos": 23294893271040.0,
      "grad_norm": 1.5764106753399945,
      "language_loss": 0.75116932,
      "learning_rate": 3.749898313956279e-07,
      "loss": 0.7721445,
      "num_input_tokens_seen": 289950815,
      "step": 13436,
      "time_per_iteration": 2.6830732822418213
    },
    {
      "auxiliary_loss_clip": 0.01101594,
      "auxiliary_loss_mlp": 0.0102967,
      "balance_loss_clip": 1.03435826,
      "balance_loss_mlp": 1.01730788,
      "epoch": 0.807876146099504,
      "flos": 27162651899520.0,
      "grad_norm": 1.7519629244472612,
      "language_loss": 0.70325857,
      "learning_rate": 3.747628239215674e-07,
      "loss": 0.72457123,
      "num_input_tokens_seen": 289971730,
      "step": 13437,
      "time_per_iteration": 2.5000627040863037
    },
    {
      "auxiliary_loss_clip": 0.01085395,
      "auxiliary_loss_mlp": 0.01031408,
      "balance_loss_clip": 1.03809178,
      "balance_loss_mlp": 1.01968384,
      "epoch": 0.807936269352172,
      "flos": 27160030206720.0,
      "grad_norm": 1.6085573392273413,
      "language_loss": 0.72808737,
      "learning_rate": 3.745358780766636e-07,
      "loss": 0.74925536,
      "num_input_tokens_seen": 289992995,
      "step": 13438,
      "time_per_iteration": 2.5887696743011475
    },
    {
      "auxiliary_loss_clip": 0.010814,
      "auxiliary_loss_mlp": 0.01029826,
      "balance_loss_clip": 1.03543901,
      "balance_loss_mlp": 1.01803637,
      "epoch": 0.8079963926048399,
      "flos": 20740423703040.0,
      "grad_norm": 1.9089901041630044,
      "language_loss": 0.77073938,
      "learning_rate": 3.7430899386952344e-07,
      "loss": 0.79185164,
      "num_input_tokens_seen": 290009405,
      "step": 13439,
      "time_per_iteration": 2.5075926780700684
    },
    {
      "auxiliary_loss_clip": 0.01104023,
      "auxiliary_loss_mlp": 0.01034351,
      "balance_loss_clip": 1.03549266,
      "balance_loss_mlp": 1.02220356,
      "epoch": 0.8080565158575079,
      "flos": 25009663622400.0,
      "grad_norm": 1.558408542716528,
      "language_loss": 0.7884897,
      "learning_rate": 3.7408217130874786e-07,
      "loss": 0.8098734,
      "num_input_tokens_seen": 290031085,
      "step": 13440,
      "time_per_iteration": 2.510411500930786
    },
    {
      "auxiliary_loss_clip": 0.01084131,
      "auxiliary_loss_mlp": 0.00785353,
      "balance_loss_clip": 1.03496099,
      "balance_loss_mlp": 1.01234186,
      "epoch": 0.8081166391101758,
      "flos": 18698076293760.0,
      "grad_norm": 7.468712660480382,
      "language_loss": 0.58883297,
      "learning_rate": 3.7385541040293946e-07,
      "loss": 0.60752779,
      "num_input_tokens_seen": 290048670,
      "step": 13441,
      "time_per_iteration": 5.350764751434326
    },
    {
      "auxiliary_loss_clip": 0.01092841,
      "auxiliary_loss_mlp": 0.01032253,
      "balance_loss_clip": 1.03412437,
      "balance_loss_mlp": 1.01970029,
      "epoch": 0.8081767623628439,
      "flos": 19828651847040.0,
      "grad_norm": 1.9449709556750394,
      "language_loss": 0.76070184,
      "learning_rate": 3.7362871116069684e-07,
      "loss": 0.78195274,
      "num_input_tokens_seen": 290064085,
      "step": 13442,
      "time_per_iteration": 2.4739067554473877
    },
    {
      "auxiliary_loss_clip": 0.01082439,
      "auxiliary_loss_mlp": 0.01030083,
      "balance_loss_clip": 1.03511381,
      "balance_loss_mlp": 1.01801896,
      "epoch": 0.8082368856155118,
      "flos": 35772952982400.0,
      "grad_norm": 1.4150970132857144,
      "language_loss": 0.70640761,
      "learning_rate": 3.734020735906169e-07,
      "loss": 0.72753286,
      "num_input_tokens_seen": 290086255,
      "step": 13443,
      "time_per_iteration": 2.6232006549835205
    },
    {
      "auxiliary_loss_clip": 0.01062298,
      "auxiliary_loss_mlp": 0.01030582,
      "balance_loss_clip": 1.03704715,
      "balance_loss_mlp": 1.01870322,
      "epoch": 0.8082970088681798,
      "flos": 17198015489280.0,
      "grad_norm": 1.7571397466860874,
      "language_loss": 0.82402211,
      "learning_rate": 3.7317549770129286e-07,
      "loss": 0.84495091,
      "num_input_tokens_seen": 290103995,
      "step": 13444,
      "time_per_iteration": 3.9211857318878174
    },
    {
      "auxiliary_loss_clip": 0.01007476,
      "auxiliary_loss_mlp": 0.00761876,
      "balance_loss_clip": 1.01642799,
      "balance_loss_mlp": 0.99956733,
      "epoch": 0.8083571321208477,
      "flos": 63555207511680.0,
      "grad_norm": 1.12124287238449,
      "language_loss": 0.53639859,
      "learning_rate": 3.7294898350131754e-07,
      "loss": 0.55409211,
      "num_input_tokens_seen": 290157245,
      "step": 13445,
      "time_per_iteration": 3.014122247695923
    },
    {
      "auxiliary_loss_clip": 0.01068396,
      "auxiliary_loss_mlp": 0.01032494,
      "balance_loss_clip": 1.0362072,
      "balance_loss_mlp": 1.0195775,
      "epoch": 0.8084172553735157,
      "flos": 17930701111680.0,
      "grad_norm": 3.051317029308302,
      "language_loss": 0.72407377,
      "learning_rate": 3.7272253099927964e-07,
      "loss": 0.74508262,
      "num_input_tokens_seen": 290174970,
      "step": 13446,
      "time_per_iteration": 2.5582425594329834
    },
    {
      "auxiliary_loss_clip": 0.0108102,
      "auxiliary_loss_mlp": 0.01034582,
      "balance_loss_clip": 1.03554094,
      "balance_loss_mlp": 1.02103996,
      "epoch": 0.8084773786261836,
      "flos": 24097999507200.0,
      "grad_norm": 1.6813885176281118,
      "language_loss": 0.71171558,
      "learning_rate": 3.7249614020376606e-07,
      "loss": 0.73287159,
      "num_input_tokens_seen": 290194395,
      "step": 13447,
      "time_per_iteration": 2.5343635082244873
    },
    {
      "auxiliary_loss_clip": 0.01041226,
      "auxiliary_loss_mlp": 0.01033975,
      "balance_loss_clip": 1.0328629,
      "balance_loss_mlp": 1.01984847,
      "epoch": 0.8085375018788516,
      "flos": 15588211656960.0,
      "grad_norm": 2.106140692522029,
      "language_loss": 0.74815637,
      "learning_rate": 3.7226981112336197e-07,
      "loss": 0.76890838,
      "num_input_tokens_seen": 290209200,
      "step": 13448,
      "time_per_iteration": 2.5801353454589844
    },
    {
      "auxiliary_loss_clip": 0.01030218,
      "auxiliary_loss_mlp": 0.01001979,
      "balance_loss_clip": 1.00744534,
      "balance_loss_mlp": 1.00067353,
      "epoch": 0.8085976251315197,
      "flos": 67561296393600.0,
      "grad_norm": 0.858274181069563,
      "language_loss": 0.63867569,
      "learning_rate": 3.7204354376665024e-07,
      "loss": 0.65899765,
      "num_input_tokens_seen": 290274565,
      "step": 13449,
      "time_per_iteration": 3.098720073699951
    },
    {
      "auxiliary_loss_clip": 0.01098404,
      "auxiliary_loss_mlp": 0.01028155,
      "balance_loss_clip": 1.03885007,
      "balance_loss_mlp": 1.01491094,
      "epoch": 0.8086577483841876,
      "flos": 22561453463040.0,
      "grad_norm": 1.6523478310728965,
      "language_loss": 0.73980927,
      "learning_rate": 3.718173381422105e-07,
      "loss": 0.76107484,
      "num_input_tokens_seen": 290293630,
      "step": 13450,
      "time_per_iteration": 3.8686087131500244
    },
    {
      "auxiliary_loss_clip": 0.01080681,
      "auxiliary_loss_mlp": 0.00786114,
      "balance_loss_clip": 1.03282404,
      "balance_loss_mlp": 1.01202238,
      "epoch": 0.8087178716368556,
      "flos": 17968084191360.0,
      "grad_norm": 1.7816325469559289,
      "language_loss": 0.74050063,
      "learning_rate": 3.7159119425861986e-07,
      "loss": 0.75916862,
      "num_input_tokens_seen": 290311450,
      "step": 13451,
      "time_per_iteration": 2.5282537937164307
    },
    {
      "auxiliary_loss_clip": 0.01082283,
      "auxiliary_loss_mlp": 0.01030696,
      "balance_loss_clip": 1.0331763,
      "balance_loss_mlp": 1.01661754,
      "epoch": 0.8087779948895235,
      "flos": 21719527603200.0,
      "grad_norm": 1.7642485045176768,
      "language_loss": 0.80379581,
      "learning_rate": 3.713651121244543e-07,
      "loss": 0.82492566,
      "num_input_tokens_seen": 290330165,
      "step": 13452,
      "time_per_iteration": 2.508531332015991
    },
    {
      "auxiliary_loss_clip": 0.01096691,
      "auxiliary_loss_mlp": 0.0103464,
      "balance_loss_clip": 1.03636551,
      "balance_loss_mlp": 1.02264214,
      "epoch": 0.8088381181421915,
      "flos": 29092885983360.0,
      "grad_norm": 1.5704841381602948,
      "language_loss": 0.78501844,
      "learning_rate": 3.711390917482875e-07,
      "loss": 0.80633175,
      "num_input_tokens_seen": 290350815,
      "step": 13453,
      "time_per_iteration": 2.5479650497436523
    },
    {
      "auxiliary_loss_clip": 0.01053357,
      "auxiliary_loss_mlp": 0.01031621,
      "balance_loss_clip": 1.03145123,
      "balance_loss_mlp": 1.01869321,
      "epoch": 0.8088982413948594,
      "flos": 22198432659840.0,
      "grad_norm": 2.873101353610194,
      "language_loss": 0.77234232,
      "learning_rate": 3.709131331386892e-07,
      "loss": 0.79319209,
      "num_input_tokens_seen": 290367380,
      "step": 13454,
      "time_per_iteration": 2.5402090549468994
    },
    {
      "auxiliary_loss_clip": 0.01073785,
      "auxiliary_loss_mlp": 0.01031142,
      "balance_loss_clip": 1.03669643,
      "balance_loss_mlp": 1.01814246,
      "epoch": 0.8089583646475275,
      "flos": 28036717453440.0,
      "grad_norm": 2.669215904540028,
      "language_loss": 0.76466572,
      "learning_rate": 3.7068723630422795e-07,
      "loss": 0.78571504,
      "num_input_tokens_seen": 290387965,
      "step": 13455,
      "time_per_iteration": 2.6208949089050293
    },
    {
      "auxiliary_loss_clip": 0.01078595,
      "auxiliary_loss_mlp": 0.0103231,
      "balance_loss_clip": 1.0331955,
      "balance_loss_mlp": 1.01932812,
      "epoch": 0.8090184879001954,
      "flos": 16617735273600.0,
      "grad_norm": 1.8141928330035448,
      "language_loss": 0.78921223,
      "learning_rate": 3.70461401253471e-07,
      "loss": 0.81032121,
      "num_input_tokens_seen": 290404150,
      "step": 13456,
      "time_per_iteration": 2.4786179065704346
    },
    {
      "auxiliary_loss_clip": 0.01105476,
      "auxiliary_loss_mlp": 0.01036385,
      "balance_loss_clip": 1.0380429,
      "balance_loss_mlp": 1.02474427,
      "epoch": 0.8090786111528634,
      "flos": 27340804379520.0,
      "grad_norm": 1.8386846667132055,
      "language_loss": 0.71925366,
      "learning_rate": 3.702356279949801e-07,
      "loss": 0.74067235,
      "num_input_tokens_seen": 290422370,
      "step": 13457,
      "time_per_iteration": 2.5039196014404297
    },
    {
      "auxiliary_loss_clip": 0.01082018,
      "auxiliary_loss_mlp": 0.01028426,
      "balance_loss_clip": 1.03408682,
      "balance_loss_mlp": 1.01717234,
      "epoch": 0.8091387344055313,
      "flos": 21105742976640.0,
      "grad_norm": 1.7572415395313743,
      "language_loss": 0.72606707,
      "learning_rate": 3.700099165373176e-07,
      "loss": 0.74717152,
      "num_input_tokens_seen": 290442645,
      "step": 13458,
      "time_per_iteration": 2.509395122528076
    },
    {
      "auxiliary_loss_clip": 0.01094003,
      "auxiliary_loss_mlp": 0.01034391,
      "balance_loss_clip": 1.03518605,
      "balance_loss_mlp": 1.02222633,
      "epoch": 0.8091988576581993,
      "flos": 11655060318720.0,
      "grad_norm": 6.613367883645332,
      "language_loss": 0.79333818,
      "learning_rate": 3.6978426688904275e-07,
      "loss": 0.81462216,
      "num_input_tokens_seen": 290458520,
      "step": 13459,
      "time_per_iteration": 2.4819753170013428
    },
    {
      "auxiliary_loss_clip": 0.01079732,
      "auxiliary_loss_mlp": 0.01025171,
      "balance_loss_clip": 1.03627229,
      "balance_loss_mlp": 1.01233184,
      "epoch": 0.8092589809108672,
      "flos": 22963329803520.0,
      "grad_norm": 1.996948170732086,
      "language_loss": 0.7995311,
      "learning_rate": 3.695586790587113e-07,
      "loss": 0.82058024,
      "num_input_tokens_seen": 290474465,
      "step": 13460,
      "time_per_iteration": 2.5079989433288574
    },
    {
      "auxiliary_loss_clip": 0.0108349,
      "auxiliary_loss_mlp": 0.01033416,
      "balance_loss_clip": 1.03395259,
      "balance_loss_mlp": 1.02035093,
      "epoch": 0.8093191041635353,
      "flos": 13260985482240.0,
      "grad_norm": 1.8966764052107896,
      "language_loss": 0.8457011,
      "learning_rate": 3.693331530548789e-07,
      "loss": 0.86687016,
      "num_input_tokens_seen": 290492060,
      "step": 13461,
      "time_per_iteration": 2.530317544937134
    },
    {
      "auxiliary_loss_clip": 0.01096876,
      "auxiliary_loss_mlp": 0.01037091,
      "balance_loss_clip": 1.03728783,
      "balance_loss_mlp": 1.02394819,
      "epoch": 0.8093792274162032,
      "flos": 25516003691520.0,
      "grad_norm": 1.7300795018869375,
      "language_loss": 0.76377636,
      "learning_rate": 3.69107688886096e-07,
      "loss": 0.78511602,
      "num_input_tokens_seen": 290511510,
      "step": 13462,
      "time_per_iteration": 2.5187902450561523
    },
    {
      "auxiliary_loss_clip": 0.01085222,
      "auxiliary_loss_mlp": 0.01036295,
      "balance_loss_clip": 1.03907633,
      "balance_loss_mlp": 1.02297354,
      "epoch": 0.8094393506688712,
      "flos": 23546483107200.0,
      "grad_norm": 1.6757498685713699,
      "language_loss": 0.83107698,
      "learning_rate": 3.6888228656091357e-07,
      "loss": 0.85229218,
      "num_input_tokens_seen": 290530035,
      "step": 13463,
      "time_per_iteration": 2.570019245147705
    },
    {
      "auxiliary_loss_clip": 0.01103852,
      "auxiliary_loss_mlp": 0.01035285,
      "balance_loss_clip": 1.03602386,
      "balance_loss_mlp": 1.02386463,
      "epoch": 0.8094994739215392,
      "flos": 17055917285760.0,
      "grad_norm": 1.815064688933616,
      "language_loss": 0.62185377,
      "learning_rate": 3.686569460878779e-07,
      "loss": 0.64324516,
      "num_input_tokens_seen": 290548245,
      "step": 13464,
      "time_per_iteration": 2.419294834136963
    },
    {
      "auxiliary_loss_clip": 0.01103214,
      "auxiliary_loss_mlp": 0.01029491,
      "balance_loss_clip": 1.03595531,
      "balance_loss_mlp": 1.01804662,
      "epoch": 0.8095595971742071,
      "flos": 23551223702400.0,
      "grad_norm": 1.5152508608648634,
      "language_loss": 0.61854351,
      "learning_rate": 3.684316674755341e-07,
      "loss": 0.63987052,
      "num_input_tokens_seen": 290568625,
      "step": 13465,
      "time_per_iteration": 2.4879698753356934
    },
    {
      "auxiliary_loss_clip": 0.01095709,
      "auxiliary_loss_mlp": 0.01035672,
      "balance_loss_clip": 1.03835309,
      "balance_loss_mlp": 1.02337599,
      "epoch": 0.8096197204268751,
      "flos": 20373201008640.0,
      "grad_norm": 2.4631838056043285,
      "language_loss": 0.81999326,
      "learning_rate": 3.682064507324256e-07,
      "loss": 0.8413071,
      "num_input_tokens_seen": 290586575,
      "step": 13466,
      "time_per_iteration": 2.459765672683716
    },
    {
      "auxiliary_loss_clip": 0.01089907,
      "auxiliary_loss_mlp": 0.00788407,
      "balance_loss_clip": 1.0361135,
      "balance_loss_mlp": 1.01686001,
      "epoch": 0.809679843679543,
      "flos": 27818775682560.0,
      "grad_norm": 1.9302829005677207,
      "language_loss": 0.76190174,
      "learning_rate": 3.6798129586709204e-07,
      "loss": 0.78068489,
      "num_input_tokens_seen": 290606790,
      "step": 13467,
      "time_per_iteration": 2.5741984844207764
    },
    {
      "auxiliary_loss_clip": 0.01070125,
      "auxiliary_loss_mlp": 0.01026328,
      "balance_loss_clip": 1.03079128,
      "balance_loss_mlp": 1.01400757,
      "epoch": 0.8097399669322111,
      "flos": 22014103040640.0,
      "grad_norm": 1.6515073553760422,
      "language_loss": 0.79379702,
      "learning_rate": 3.6775620288807073e-07,
      "loss": 0.81476152,
      "num_input_tokens_seen": 290625525,
      "step": 13468,
      "time_per_iteration": 2.5347628593444824
    },
    {
      "auxiliary_loss_clip": 0.01092132,
      "auxiliary_loss_mlp": 0.01028504,
      "balance_loss_clip": 1.03517032,
      "balance_loss_mlp": 1.0168395,
      "epoch": 0.809800090184879,
      "flos": 18988988544000.0,
      "grad_norm": 1.8602259438843383,
      "language_loss": 0.67125869,
      "learning_rate": 3.675311718038978e-07,
      "loss": 0.69246507,
      "num_input_tokens_seen": 290644935,
      "step": 13469,
      "time_per_iteration": 2.4925785064697266
    },
    {
      "auxiliary_loss_clip": 0.01005176,
      "auxiliary_loss_mlp": 0.01016335,
      "balance_loss_clip": 1.01454377,
      "balance_loss_mlp": 1.01491094,
      "epoch": 0.809860213437547,
      "flos": 66099516508800.0,
      "grad_norm": 0.7012125986289238,
      "language_loss": 0.54697812,
      "learning_rate": 3.6730620262310683e-07,
      "loss": 0.56719327,
      "num_input_tokens_seen": 290710735,
      "step": 13470,
      "time_per_iteration": 3.2326226234436035
    },
    {
      "auxiliary_loss_clip": 0.0110376,
      "auxiliary_loss_mlp": 0.01032374,
      "balance_loss_clip": 1.03468418,
      "balance_loss_mlp": 1.02063775,
      "epoch": 0.8099203366902149,
      "flos": 20882485992960.0,
      "grad_norm": 1.7133434799370288,
      "language_loss": 0.69548899,
      "learning_rate": 3.670812953542279e-07,
      "loss": 0.7168504,
      "num_input_tokens_seen": 290729565,
      "step": 13471,
      "time_per_iteration": 2.454738140106201
    },
    {
      "auxiliary_loss_clip": 0.01096443,
      "auxiliary_loss_mlp": 0.01027477,
      "balance_loss_clip": 1.03712296,
      "balance_loss_mlp": 1.01586008,
      "epoch": 0.8099804599428829,
      "flos": 26030927111040.0,
      "grad_norm": 1.8888712990381693,
      "language_loss": 0.79671299,
      "learning_rate": 3.6685645000579003e-07,
      "loss": 0.81795216,
      "num_input_tokens_seen": 290749360,
      "step": 13472,
      "time_per_iteration": 2.5263824462890625
    },
    {
      "auxiliary_loss_clip": 0.01022152,
      "auxiliary_loss_mlp": 0.01005157,
      "balance_loss_clip": 1.00793135,
      "balance_loss_mlp": 1.00386381,
      "epoch": 0.8100405831955508,
      "flos": 69303573584640.0,
      "grad_norm": 0.748138137928801,
      "language_loss": 0.57797754,
      "learning_rate": 3.666316665863201e-07,
      "loss": 0.59825063,
      "num_input_tokens_seen": 290812145,
      "step": 13473,
      "time_per_iteration": 3.054250717163086
    },
    {
      "auxiliary_loss_clip": 0.01054752,
      "auxiliary_loss_mlp": 0.01028679,
      "balance_loss_clip": 1.03534043,
      "balance_loss_mlp": 1.01650167,
      "epoch": 0.8101007064482189,
      "flos": 15012492468480.0,
      "grad_norm": 1.7327519745334876,
      "language_loss": 0.73618472,
      "learning_rate": 3.664069451043399e-07,
      "loss": 0.75701904,
      "num_input_tokens_seen": 290829845,
      "step": 13474,
      "time_per_iteration": 2.5743112564086914
    },
    {
      "auxiliary_loss_clip": 0.01094536,
      "auxiliary_loss_mlp": 0.01036788,
      "balance_loss_clip": 1.03770804,
      "balance_loss_mlp": 1.02439654,
      "epoch": 0.8101608297008868,
      "flos": 21067210661760.0,
      "grad_norm": 1.707475092931714,
      "language_loss": 0.78764009,
      "learning_rate": 3.661822855683723e-07,
      "loss": 0.80895334,
      "num_input_tokens_seen": 290848815,
      "step": 13475,
      "time_per_iteration": 2.4810335636138916
    },
    {
      "auxiliary_loss_clip": 0.01092139,
      "auxiliary_loss_mlp": 0.01033739,
      "balance_loss_clip": 1.03471994,
      "balance_loss_mlp": 1.0220623,
      "epoch": 0.8102209529535548,
      "flos": 23731279603200.0,
      "grad_norm": 1.550863660263918,
      "language_loss": 0.75135654,
      "learning_rate": 3.659576879869364e-07,
      "loss": 0.77261531,
      "num_input_tokens_seen": 290868580,
      "step": 13476,
      "time_per_iteration": 2.530224084854126
    },
    {
      "auxiliary_loss_clip": 0.01087086,
      "auxiliary_loss_mlp": 0.0103661,
      "balance_loss_clip": 1.03534293,
      "balance_loss_mlp": 1.02340794,
      "epoch": 0.8102810762062228,
      "flos": 10955879107200.0,
      "grad_norm": 2.1682605014044207,
      "language_loss": 0.74054408,
      "learning_rate": 3.657331523685485e-07,
      "loss": 0.76178104,
      "num_input_tokens_seen": 290883540,
      "step": 13477,
      "time_per_iteration": 2.4677090644836426
    },
    {
      "auxiliary_loss_clip": 0.01086216,
      "auxiliary_loss_mlp": 0.01034979,
      "balance_loss_clip": 1.03608203,
      "balance_loss_mlp": 1.02320707,
      "epoch": 0.8103411994588907,
      "flos": 14648825220480.0,
      "grad_norm": 2.06521305020464,
      "language_loss": 0.69511408,
      "learning_rate": 3.6550867872172365e-07,
      "loss": 0.716326,
      "num_input_tokens_seen": 290901560,
      "step": 13478,
      "time_per_iteration": 2.5183346271514893
    },
    {
      "auxiliary_loss_clip": 0.01029985,
      "auxiliary_loss_mlp": 0.01005614,
      "balance_loss_clip": 1.00714529,
      "balance_loss_mlp": 1.00433862,
      "epoch": 0.8104013227115587,
      "flos": 59153314665600.0,
      "grad_norm": 0.6879278010860218,
      "language_loss": 0.52166045,
      "learning_rate": 3.6528426705497293e-07,
      "loss": 0.54201639,
      "num_input_tokens_seen": 290959185,
      "step": 13479,
      "time_per_iteration": 4.409225702285767
    },
    {
      "auxiliary_loss_clip": 0.01061499,
      "auxiliary_loss_mlp": 0.01030951,
      "balance_loss_clip": 1.03574347,
      "balance_loss_mlp": 1.01821303,
      "epoch": 0.8104614459642266,
      "flos": 19828687760640.0,
      "grad_norm": 2.3654490178000445,
      "language_loss": 0.71154708,
      "learning_rate": 3.650599173768072e-07,
      "loss": 0.73247159,
      "num_input_tokens_seen": 290979585,
      "step": 13480,
      "time_per_iteration": 3.939840316772461
    },
    {
      "auxiliary_loss_clip": 0.0110535,
      "auxiliary_loss_mlp": 0.01033585,
      "balance_loss_clip": 1.03607583,
      "balance_loss_mlp": 1.02175355,
      "epoch": 0.8105215692168947,
      "flos": 25374264624000.0,
      "grad_norm": 1.724476872368226,
      "language_loss": 0.7970953,
      "learning_rate": 3.648356296957327e-07,
      "loss": 0.81848466,
      "num_input_tokens_seen": 291000865,
      "step": 13481,
      "time_per_iteration": 2.495518445968628
    },
    {
      "auxiliary_loss_clip": 0.0108088,
      "auxiliary_loss_mlp": 0.0103336,
      "balance_loss_clip": 1.03411293,
      "balance_loss_mlp": 1.02155876,
      "epoch": 0.8105816924695626,
      "flos": 20481722974080.0,
      "grad_norm": 1.6465191533476875,
      "language_loss": 0.72511113,
      "learning_rate": 3.646114040202548e-07,
      "loss": 0.74625355,
      "num_input_tokens_seen": 291018285,
      "step": 13482,
      "time_per_iteration": 2.477145195007324
    },
    {
      "auxiliary_loss_clip": 0.01047882,
      "auxiliary_loss_mlp": 0.01025158,
      "balance_loss_clip": 1.03425491,
      "balance_loss_mlp": 1.01296318,
      "epoch": 0.8106418157222306,
      "flos": 14538687143040.0,
      "grad_norm": 2.3981835504458635,
      "language_loss": 0.65720528,
      "learning_rate": 3.6438724035887705e-07,
      "loss": 0.67793566,
      "num_input_tokens_seen": 291035745,
      "step": 13483,
      "time_per_iteration": 3.993739604949951
    },
    {
      "auxiliary_loss_clip": 0.01077558,
      "auxiliary_loss_mlp": 0.0103231,
      "balance_loss_clip": 1.03226781,
      "balance_loss_mlp": 1.01907146,
      "epoch": 0.8107019389748985,
      "flos": 22564470205440.0,
      "grad_norm": 1.562378538015318,
      "language_loss": 0.7635231,
      "learning_rate": 3.641631387200992e-07,
      "loss": 0.78462172,
      "num_input_tokens_seen": 291053280,
      "step": 13484,
      "time_per_iteration": 2.538999557495117
    },
    {
      "auxiliary_loss_clip": 0.0108704,
      "auxiliary_loss_mlp": 0.0103509,
      "balance_loss_clip": 1.03529298,
      "balance_loss_mlp": 1.02181602,
      "epoch": 0.8107620622275665,
      "flos": 19609560840960.0,
      "grad_norm": 1.6606371645948337,
      "language_loss": 0.7209937,
      "learning_rate": 3.639390991124183e-07,
      "loss": 0.74221498,
      "num_input_tokens_seen": 291072855,
      "step": 13485,
      "time_per_iteration": 2.6343650817871094
    },
    {
      "auxiliary_loss_clip": 0.01058938,
      "auxiliary_loss_mlp": 0.01033139,
      "balance_loss_clip": 1.0313189,
      "balance_loss_mlp": 1.0196923,
      "epoch": 0.8108221854802344,
      "flos": 16143498984960.0,
      "grad_norm": 1.8824491943826365,
      "language_loss": 0.75819612,
      "learning_rate": 3.637151215443308e-07,
      "loss": 0.77911687,
      "num_input_tokens_seen": 291090285,
      "step": 13486,
      "time_per_iteration": 2.5341413021087646
    },
    {
      "auxiliary_loss_clip": 0.01086423,
      "auxiliary_loss_mlp": 0.01030453,
      "balance_loss_clip": 1.03612077,
      "balance_loss_mlp": 1.01828218,
      "epoch": 0.8108823087329025,
      "flos": 21106209853440.0,
      "grad_norm": 2.0111423069326153,
      "language_loss": 0.71953094,
      "learning_rate": 3.6349120602433045e-07,
      "loss": 0.74069971,
      "num_input_tokens_seen": 291107675,
      "step": 13487,
      "time_per_iteration": 2.5038161277770996
    },
    {
      "auxiliary_loss_clip": 0.01042588,
      "auxiliary_loss_mlp": 0.01033496,
      "balance_loss_clip": 1.03599226,
      "balance_loss_mlp": 1.02201664,
      "epoch": 0.8109424319855704,
      "flos": 29199648182400.0,
      "grad_norm": 1.7958456300690162,
      "language_loss": 0.84113014,
      "learning_rate": 3.6326735256090715e-07,
      "loss": 0.86189097,
      "num_input_tokens_seen": 291126900,
      "step": 13488,
      "time_per_iteration": 2.701881170272827
    },
    {
      "auxiliary_loss_clip": 0.01106916,
      "auxiliary_loss_mlp": 0.01031909,
      "balance_loss_clip": 1.03649688,
      "balance_loss_mlp": 1.01937437,
      "epoch": 0.8110025552382384,
      "flos": 23111856541440.0,
      "grad_norm": 1.7998419748350067,
      "language_loss": 0.73870081,
      "learning_rate": 3.630435611625502e-07,
      "loss": 0.76008904,
      "num_input_tokens_seen": 291145285,
      "step": 13489,
      "time_per_iteration": 3.87990665435791
    },
    {
      "auxiliary_loss_clip": 0.01060544,
      "auxiliary_loss_mlp": 0.00782417,
      "balance_loss_clip": 1.03533447,
      "balance_loss_mlp": 1.00914299,
      "epoch": 0.8110626784909064,
      "flos": 22379961018240.0,
      "grad_norm": 1.5931549164924441,
      "language_loss": 0.71697903,
      "learning_rate": 3.628198318377453e-07,
      "loss": 0.7354086,
      "num_input_tokens_seen": 291163485,
      "step": 13490,
      "time_per_iteration": 2.5707807540893555
    },
    {
      "auxiliary_loss_clip": 0.01073224,
      "auxiliary_loss_mlp": 0.01048269,
      "balance_loss_clip": 1.03517735,
      "balance_loss_mlp": 1.03305173,
      "epoch": 0.8111228017435743,
      "flos": 23368043318400.0,
      "grad_norm": 1.9824391329981224,
      "language_loss": 0.71231234,
      "learning_rate": 3.625961645949762e-07,
      "loss": 0.7335273,
      "num_input_tokens_seen": 291182215,
      "step": 13491,
      "time_per_iteration": 2.5717296600341797
    },
    {
      "auxiliary_loss_clip": 0.01104483,
      "auxiliary_loss_mlp": 0.01030032,
      "balance_loss_clip": 1.03486896,
      "balance_loss_mlp": 1.01777148,
      "epoch": 0.8111829249962423,
      "flos": 21286553063040.0,
      "grad_norm": 1.3102271601740754,
      "language_loss": 0.67793602,
      "learning_rate": 3.623725594427245e-07,
      "loss": 0.6992811,
      "num_input_tokens_seen": 291203145,
      "step": 13492,
      "time_per_iteration": 2.4872841835021973
    },
    {
      "auxiliary_loss_clip": 0.01062552,
      "auxiliary_loss_mlp": 0.01030711,
      "balance_loss_clip": 1.03749609,
      "balance_loss_mlp": 1.0184505,
      "epoch": 0.8112430482489102,
      "flos": 22345558767360.0,
      "grad_norm": 1.8016541171844693,
      "language_loss": 0.72058529,
      "learning_rate": 3.6214901638947006e-07,
      "loss": 0.7415179,
      "num_input_tokens_seen": 291220600,
      "step": 13493,
      "time_per_iteration": 2.631155252456665
    },
    {
      "auxiliary_loss_clip": 0.01088338,
      "auxiliary_loss_mlp": 0.01042449,
      "balance_loss_clip": 1.03285217,
      "balance_loss_mlp": 1.02892447,
      "epoch": 0.8113031715015783,
      "flos": 31138321962240.0,
      "grad_norm": 1.5812226710503685,
      "language_loss": 0.70608079,
      "learning_rate": 3.619255354436885e-07,
      "loss": 0.72738874,
      "num_input_tokens_seen": 291241195,
      "step": 13494,
      "time_per_iteration": 2.5480966567993164
    },
    {
      "auxiliary_loss_clip": 0.0109842,
      "auxiliary_loss_mlp": 0.01033908,
      "balance_loss_clip": 1.03720641,
      "balance_loss_mlp": 1.02025878,
      "epoch": 0.8113632947542462,
      "flos": 25335445000320.0,
      "grad_norm": 2.298252046677942,
      "language_loss": 0.76666605,
      "learning_rate": 3.6170211661385543e-07,
      "loss": 0.78798938,
      "num_input_tokens_seen": 291258715,
      "step": 13495,
      "time_per_iteration": 2.5338222980499268
    },
    {
      "auxiliary_loss_clip": 0.01085563,
      "auxiliary_loss_mlp": 0.01033209,
      "balance_loss_clip": 1.03513908,
      "balance_loss_mlp": 1.02063286,
      "epoch": 0.8114234180069142,
      "flos": 28439168411520.0,
      "grad_norm": 2.1458204717416125,
      "language_loss": 0.79555148,
      "learning_rate": 3.614787599084417e-07,
      "loss": 0.8167392,
      "num_input_tokens_seen": 291278030,
      "step": 13496,
      "time_per_iteration": 2.5588297843933105
    },
    {
      "auxiliary_loss_clip": 0.01095101,
      "auxiliary_loss_mlp": 0.01035964,
      "balance_loss_clip": 1.03487217,
      "balance_loss_mlp": 1.02254128,
      "epoch": 0.8114835412595821,
      "flos": 20338870584960.0,
      "grad_norm": 1.6391168878361722,
      "language_loss": 0.71166933,
      "learning_rate": 3.6125546533591787e-07,
      "loss": 0.73297989,
      "num_input_tokens_seen": 291296740,
      "step": 13497,
      "time_per_iteration": 2.504910707473755
    },
    {
      "auxiliary_loss_clip": 0.01070887,
      "auxiliary_loss_mlp": 0.01032273,
      "balance_loss_clip": 1.03443825,
      "balance_loss_mlp": 1.02035248,
      "epoch": 0.8115436645122501,
      "flos": 22490889194880.0,
      "grad_norm": 1.7138158703376067,
      "language_loss": 0.76797378,
      "learning_rate": 3.610322329047508e-07,
      "loss": 0.7890054,
      "num_input_tokens_seen": 291318730,
      "step": 13498,
      "time_per_iteration": 2.5896711349487305
    },
    {
      "auxiliary_loss_clip": 0.01104917,
      "auxiliary_loss_mlp": 0.01033714,
      "balance_loss_clip": 1.0350976,
      "balance_loss_mlp": 1.02147174,
      "epoch": 0.811603787764918,
      "flos": 13845288021120.0,
      "grad_norm": 1.8346128630326792,
      "language_loss": 0.83932143,
      "learning_rate": 3.608090626234055e-07,
      "loss": 0.86070776,
      "num_input_tokens_seen": 291336755,
      "step": 13499,
      "time_per_iteration": 2.4593067169189453
    },
    {
      "auxiliary_loss_clip": 0.01071585,
      "auxiliary_loss_mlp": 0.01029635,
      "balance_loss_clip": 1.0340178,
      "balance_loss_mlp": 1.01600969,
      "epoch": 0.8116639110175861,
      "flos": 21614632911360.0,
      "grad_norm": 1.478332255748856,
      "language_loss": 0.7634784,
      "learning_rate": 3.6058595450034603e-07,
      "loss": 0.7844907,
      "num_input_tokens_seen": 291356795,
      "step": 13500,
      "time_per_iteration": 2.5429863929748535
    },
    {
      "auxiliary_loss_clip": 0.01009823,
      "auxiliary_loss_mlp": 0.01001923,
      "balance_loss_clip": 1.00815535,
      "balance_loss_mlp": 1.00069511,
      "epoch": 0.811724034270254,
      "flos": 64459799625600.0,
      "grad_norm": 0.8041681984532766,
      "language_loss": 0.59894121,
      "learning_rate": 3.603629085440303e-07,
      "loss": 0.61905861,
      "num_input_tokens_seen": 291416005,
      "step": 13501,
      "time_per_iteration": 3.1992151737213135
    },
    {
      "auxiliary_loss_clip": 0.01090532,
      "auxiliary_loss_mlp": 0.01027251,
      "balance_loss_clip": 1.03479815,
      "balance_loss_mlp": 1.01560998,
      "epoch": 0.811784157522922,
      "flos": 24754123290240.0,
      "grad_norm": 1.812669804547587,
      "language_loss": 0.79299885,
      "learning_rate": 3.6013992476291753e-07,
      "loss": 0.81417668,
      "num_input_tokens_seen": 291434870,
      "step": 13502,
      "time_per_iteration": 2.5064752101898193
    },
    {
      "auxiliary_loss_clip": 0.0107928,
      "auxiliary_loss_mlp": 0.01039213,
      "balance_loss_clip": 1.0335567,
      "balance_loss_mlp": 1.02579582,
      "epoch": 0.81184428077559,
      "flos": 12167146563840.0,
      "grad_norm": 1.8222646766497628,
      "language_loss": 0.71218187,
      "learning_rate": 3.599170031654635e-07,
      "loss": 0.73336679,
      "num_input_tokens_seen": 291452230,
      "step": 13503,
      "time_per_iteration": 2.5078284740448
    },
    {
      "auxiliary_loss_clip": 0.01075117,
      "auxiliary_loss_mlp": 0.01031684,
      "balance_loss_clip": 1.03420186,
      "balance_loss_mlp": 1.01793957,
      "epoch": 0.8119044040282579,
      "flos": 44422037775360.0,
      "grad_norm": 1.7149152056760173,
      "language_loss": 0.67949259,
      "learning_rate": 3.5969414376012065e-07,
      "loss": 0.70056063,
      "num_input_tokens_seen": 291477425,
      "step": 13504,
      "time_per_iteration": 2.7182419300079346
    },
    {
      "auxiliary_loss_clip": 0.01079532,
      "auxiliary_loss_mlp": 0.01031261,
      "balance_loss_clip": 1.03539801,
      "balance_loss_mlp": 1.01796365,
      "epoch": 0.8119645272809259,
      "flos": 52155507957120.0,
      "grad_norm": 1.9108291380399665,
      "language_loss": 0.74560642,
      "learning_rate": 3.594713465553403e-07,
      "loss": 0.76671439,
      "num_input_tokens_seen": 291501070,
      "step": 13505,
      "time_per_iteration": 2.8055641651153564
    },
    {
      "auxiliary_loss_clip": 0.01083962,
      "auxiliary_loss_mlp": 0.01028016,
      "balance_loss_clip": 1.0357759,
      "balance_loss_mlp": 1.01444459,
      "epoch": 0.8120246505335939,
      "flos": 30232978640640.0,
      "grad_norm": 1.8377068530784149,
      "language_loss": 0.72351646,
      "learning_rate": 3.5924861155957123e-07,
      "loss": 0.7446363,
      "num_input_tokens_seen": 291524945,
      "step": 13506,
      "time_per_iteration": 2.6256158351898193
    },
    {
      "auxiliary_loss_clip": 0.01110258,
      "auxiliary_loss_mlp": 0.01029672,
      "balance_loss_clip": 1.03618622,
      "balance_loss_mlp": 1.01736999,
      "epoch": 0.8120847737862619,
      "flos": 22127652910080.0,
      "grad_norm": 2.286956278737283,
      "language_loss": 0.76284623,
      "learning_rate": 3.590259387812593e-07,
      "loss": 0.78424561,
      "num_input_tokens_seen": 291544605,
      "step": 13507,
      "time_per_iteration": 2.5013647079467773
    },
    {
      "auxiliary_loss_clip": 0.01105628,
      "auxiliary_loss_mlp": 0.01026437,
      "balance_loss_clip": 1.03398085,
      "balance_loss_mlp": 1.01435494,
      "epoch": 0.8121448970389298,
      "flos": 23295180579840.0,
      "grad_norm": 1.5975871882106458,
      "language_loss": 0.69977462,
      "learning_rate": 3.5880332822884783e-07,
      "loss": 0.7210952,
      "num_input_tokens_seen": 291563850,
      "step": 13508,
      "time_per_iteration": 2.456258773803711
    },
    {
      "auxiliary_loss_clip": 0.01093072,
      "auxiliary_loss_mlp": 0.01030895,
      "balance_loss_clip": 1.03516197,
      "balance_loss_mlp": 1.01886106,
      "epoch": 0.8122050202915978,
      "flos": 22164138149760.0,
      "grad_norm": 3.3546064926186117,
      "language_loss": 0.75630605,
      "learning_rate": 3.585807799107785e-07,
      "loss": 0.77754569,
      "num_input_tokens_seen": 291581730,
      "step": 13509,
      "time_per_iteration": 2.5372109413146973
    },
    {
      "auxiliary_loss_clip": 0.0110764,
      "auxiliary_loss_mlp": 0.01031588,
      "balance_loss_clip": 1.03623891,
      "balance_loss_mlp": 1.01870751,
      "epoch": 0.8122651435442657,
      "flos": 23258946735360.0,
      "grad_norm": 1.6912726136113183,
      "language_loss": 0.76919103,
      "learning_rate": 3.58358293835491e-07,
      "loss": 0.79058325,
      "num_input_tokens_seen": 291601225,
      "step": 13510,
      "time_per_iteration": 2.4911715984344482
    },
    {
      "auxiliary_loss_clip": 0.01096388,
      "auxiliary_loss_mlp": 0.01034948,
      "balance_loss_clip": 1.03491104,
      "balance_loss_mlp": 1.0214715,
      "epoch": 0.8123252667969337,
      "flos": 16140015365760.0,
      "grad_norm": 1.7936784902429423,
      "language_loss": 0.70019907,
      "learning_rate": 3.581358700114212e-07,
      "loss": 0.72151244,
      "num_input_tokens_seen": 291616995,
      "step": 13511,
      "time_per_iteration": 2.478416919708252
    },
    {
      "auxiliary_loss_clip": 0.0108759,
      "auxiliary_loss_mlp": 0.01034535,
      "balance_loss_clip": 1.03726733,
      "balance_loss_mlp": 1.02233362,
      "epoch": 0.8123853900496016,
      "flos": 21245399055360.0,
      "grad_norm": 1.55542383695618,
      "language_loss": 0.79775178,
      "learning_rate": 3.57913508447004e-07,
      "loss": 0.818973,
      "num_input_tokens_seen": 291636145,
      "step": 13512,
      "time_per_iteration": 2.5116941928863525
    },
    {
      "auxiliary_loss_clip": 0.01092251,
      "auxiliary_loss_mlp": 0.01031714,
      "balance_loss_clip": 1.0342021,
      "balance_loss_mlp": 1.02025247,
      "epoch": 0.8124455133022697,
      "flos": 64377596373120.0,
      "grad_norm": 1.8320881960268358,
      "language_loss": 0.63341737,
      "learning_rate": 3.5769120915067076e-07,
      "loss": 0.65465701,
      "num_input_tokens_seen": 291662440,
      "step": 13513,
      "time_per_iteration": 2.8999693393707275
    },
    {
      "auxiliary_loss_clip": 0.01058155,
      "auxiliary_loss_mlp": 0.0103419,
      "balance_loss_clip": 1.03473771,
      "balance_loss_mlp": 1.02126169,
      "epoch": 0.8125056365549376,
      "flos": 23842207779840.0,
      "grad_norm": 1.6412105492341407,
      "language_loss": 0.71628612,
      "learning_rate": 3.5746897213085194e-07,
      "loss": 0.73720956,
      "num_input_tokens_seen": 291680950,
      "step": 13514,
      "time_per_iteration": 2.615200996398926
    },
    {
      "auxiliary_loss_clip": 0.01067839,
      "auxiliary_loss_mlp": 0.0102964,
      "balance_loss_clip": 1.03342426,
      "balance_loss_mlp": 1.01760554,
      "epoch": 0.8125657598076056,
      "flos": 23550325862400.0,
      "grad_norm": 1.593813484702048,
      "language_loss": 0.62784863,
      "learning_rate": 3.5724679739597364e-07,
      "loss": 0.64882338,
      "num_input_tokens_seen": 291702395,
      "step": 13515,
      "time_per_iteration": 2.5983948707580566
    },
    {
      "auxiliary_loss_clip": 0.01099002,
      "auxiliary_loss_mlp": 0.0078276,
      "balance_loss_clip": 1.03380692,
      "balance_loss_mlp": 1.01055503,
      "epoch": 0.8126258830602736,
      "flos": 20704225772160.0,
      "grad_norm": 1.5750285597602072,
      "language_loss": 0.7534306,
      "learning_rate": 3.570246849544616e-07,
      "loss": 0.77224827,
      "num_input_tokens_seen": 291721135,
      "step": 13516,
      "time_per_iteration": 2.4811291694641113
    },
    {
      "auxiliary_loss_clip": 0.01064808,
      "auxiliary_loss_mlp": 0.01030118,
      "balance_loss_clip": 1.03614044,
      "balance_loss_mlp": 1.01800644,
      "epoch": 0.8126860063129415,
      "flos": 23618160696960.0,
      "grad_norm": 1.4882806905598813,
      "language_loss": 0.91422576,
      "learning_rate": 3.5680263481473907e-07,
      "loss": 0.93517494,
      "num_input_tokens_seen": 291741235,
      "step": 13517,
      "time_per_iteration": 2.62703537940979
    },
    {
      "auxiliary_loss_clip": 0.01097365,
      "auxiliary_loss_mlp": 0.00785466,
      "balance_loss_clip": 1.0381844,
      "balance_loss_mlp": 1.01018977,
      "epoch": 0.8127461295656095,
      "flos": 25007149670400.0,
      "grad_norm": 1.4841449349530553,
      "language_loss": 0.78627187,
      "learning_rate": 3.565806469852244e-07,
      "loss": 0.8051002,
      "num_input_tokens_seen": 291761430,
      "step": 13518,
      "time_per_iteration": 3.889030933380127
    },
    {
      "auxiliary_loss_clip": 0.0109342,
      "auxiliary_loss_mlp": 0.01029661,
      "balance_loss_clip": 1.03610229,
      "balance_loss_mlp": 1.01869369,
      "epoch": 0.8128062528182775,
      "flos": 27342169096320.0,
      "grad_norm": 1.611181259856512,
      "language_loss": 0.78869343,
      "learning_rate": 3.56358721474336e-07,
      "loss": 0.80992424,
      "num_input_tokens_seen": 291781755,
      "step": 13519,
      "time_per_iteration": 2.5348429679870605
    },
    {
      "auxiliary_loss_clip": 0.01104895,
      "auxiliary_loss_mlp": 0.01037149,
      "balance_loss_clip": 1.03388977,
      "balance_loss_mlp": 1.02483499,
      "epoch": 0.8128663760709455,
      "flos": 26506312634880.0,
      "grad_norm": 1.5842883800319059,
      "language_loss": 0.7020728,
      "learning_rate": 3.561368582904905e-07,
      "loss": 0.72349328,
      "num_input_tokens_seen": 291804410,
      "step": 13520,
      "time_per_iteration": 3.924917221069336
    },
    {
      "auxiliary_loss_clip": 0.0108721,
      "auxiliary_loss_mlp": 0.01029831,
      "balance_loss_clip": 1.03474569,
      "balance_loss_mlp": 1.01796412,
      "epoch": 0.8129264993236134,
      "flos": 17931239815680.0,
      "grad_norm": 1.4739678049306009,
      "language_loss": 0.72440779,
      "learning_rate": 3.5591505744209925e-07,
      "loss": 0.74557817,
      "num_input_tokens_seen": 291823285,
      "step": 13521,
      "time_per_iteration": 2.5779337882995605
    },
    {
      "auxiliary_loss_clip": 0.01096626,
      "auxiliary_loss_mlp": 0.01030671,
      "balance_loss_clip": 1.0351212,
      "balance_loss_mlp": 1.01834524,
      "epoch": 0.8129866225762814,
      "flos": 26177694082560.0,
      "grad_norm": 1.8198246600742918,
      "language_loss": 0.69780672,
      "learning_rate": 3.5569331893757394e-07,
      "loss": 0.71907967,
      "num_input_tokens_seen": 291845305,
      "step": 13522,
      "time_per_iteration": 3.965945243835449
    },
    {
      "auxiliary_loss_clip": 0.01088897,
      "auxiliary_loss_mlp": 0.01034398,
      "balance_loss_clip": 1.03521621,
      "balance_loss_mlp": 1.023157,
      "epoch": 0.8130467458289493,
      "flos": 21032197879680.0,
      "grad_norm": 1.519661192449538,
      "language_loss": 0.70685053,
      "learning_rate": 3.554716427853233e-07,
      "loss": 0.72808349,
      "num_input_tokens_seen": 291863715,
      "step": 13523,
      "time_per_iteration": 2.5087697505950928
    },
    {
      "auxiliary_loss_clip": 0.01092145,
      "auxiliary_loss_mlp": 0.01029561,
      "balance_loss_clip": 1.03326011,
      "balance_loss_mlp": 1.01700783,
      "epoch": 0.8131068690816173,
      "flos": 15487051979520.0,
      "grad_norm": 2.3394092640195483,
      "language_loss": 0.71266526,
      "learning_rate": 3.5525002899375256e-07,
      "loss": 0.73388231,
      "num_input_tokens_seen": 291880735,
      "step": 13524,
      "time_per_iteration": 2.434755563735962
    },
    {
      "auxiliary_loss_clip": 0.0109314,
      "auxiliary_loss_mlp": 0.01026305,
      "balance_loss_clip": 1.03505576,
      "balance_loss_mlp": 1.0149622,
      "epoch": 0.8131669923342852,
      "flos": 29351227576320.0,
      "grad_norm": 1.845289447033173,
      "language_loss": 0.62384439,
      "learning_rate": 3.550284775712653e-07,
      "loss": 0.64503884,
      "num_input_tokens_seen": 291900535,
      "step": 13525,
      "time_per_iteration": 2.5842974185943604
    },
    {
      "auxiliary_loss_clip": 0.01085084,
      "auxiliary_loss_mlp": 0.01036999,
      "balance_loss_clip": 1.03664339,
      "balance_loss_mlp": 1.02538776,
      "epoch": 0.8132271155869533,
      "flos": 35256162055680.0,
      "grad_norm": 1.4695521410933603,
      "language_loss": 0.65372926,
      "learning_rate": 3.548069885262628e-07,
      "loss": 0.67495012,
      "num_input_tokens_seen": 291919760,
      "step": 13526,
      "time_per_iteration": 2.6480839252471924
    },
    {
      "auxiliary_loss_clip": 0.01079962,
      "auxiliary_loss_mlp": 0.01029138,
      "balance_loss_clip": 1.03423381,
      "balance_loss_mlp": 1.01761639,
      "epoch": 0.8132872388396212,
      "flos": 27781895393280.0,
      "grad_norm": 2.026405696812929,
      "language_loss": 0.74924183,
      "learning_rate": 3.5458556186714473e-07,
      "loss": 0.77033281,
      "num_input_tokens_seen": 291938915,
      "step": 13527,
      "time_per_iteration": 2.5842063426971436
    },
    {
      "auxiliary_loss_clip": 0.01101781,
      "auxiliary_loss_mlp": 0.01028369,
      "balance_loss_clip": 1.03387499,
      "balance_loss_mlp": 1.01675773,
      "epoch": 0.8133473620922892,
      "flos": 27819601695360.0,
      "grad_norm": 1.7179378036664423,
      "language_loss": 0.70519549,
      "learning_rate": 3.5436419760230706e-07,
      "loss": 0.72649699,
      "num_input_tokens_seen": 291958145,
      "step": 13528,
      "time_per_iteration": 3.8957135677337646
    },
    {
      "auxiliary_loss_clip": 0.01104764,
      "auxiliary_loss_mlp": 0.01029652,
      "balance_loss_clip": 1.03471267,
      "balance_loss_mlp": 1.01787364,
      "epoch": 0.8134074853449572,
      "flos": 18989527248000.0,
      "grad_norm": 1.9692142502019614,
      "language_loss": 0.68746084,
      "learning_rate": 3.5414289574014357e-07,
      "loss": 0.70880497,
      "num_input_tokens_seen": 291976860,
      "step": 13529,
      "time_per_iteration": 2.4437363147735596
    },
    {
      "auxiliary_loss_clip": 0.01093295,
      "auxiliary_loss_mlp": 0.0103075,
      "balance_loss_clip": 1.03613162,
      "balance_loss_mlp": 1.0194844,
      "epoch": 0.8134676085976251,
      "flos": 24242863057920.0,
      "grad_norm": 1.3146574939297855,
      "language_loss": 0.77465117,
      "learning_rate": 3.5392165628904635e-07,
      "loss": 0.79589152,
      "num_input_tokens_seen": 291998085,
      "step": 13530,
      "time_per_iteration": 2.5460402965545654
    },
    {
      "auxiliary_loss_clip": 0.01092972,
      "auxiliary_loss_mlp": 0.01031412,
      "balance_loss_clip": 1.03550076,
      "balance_loss_mlp": 1.01879334,
      "epoch": 0.8135277318502931,
      "flos": 19062389986560.0,
      "grad_norm": 1.8515373636013563,
      "language_loss": 0.81803614,
      "learning_rate": 3.537004792574052e-07,
      "loss": 0.83928001,
      "num_input_tokens_seen": 292016585,
      "step": 13531,
      "time_per_iteration": 2.4839963912963867
    },
    {
      "auxiliary_loss_clip": 0.01081878,
      "auxiliary_loss_mlp": 0.01033524,
      "balance_loss_clip": 1.03491652,
      "balance_loss_mlp": 1.01951718,
      "epoch": 0.813587855102961,
      "flos": 17269728992640.0,
      "grad_norm": 8.520391249064106,
      "language_loss": 0.71506333,
      "learning_rate": 3.534793646536065e-07,
      "loss": 0.73621738,
      "num_input_tokens_seen": 292033255,
      "step": 13532,
      "time_per_iteration": 2.4987785816192627
    },
    {
      "auxiliary_loss_clip": 0.01071094,
      "auxiliary_loss_mlp": 0.01028684,
      "balance_loss_clip": 1.03400362,
      "balance_loss_mlp": 1.01697183,
      "epoch": 0.8136479783556291,
      "flos": 20157593621760.0,
      "grad_norm": 1.8225666960864597,
      "language_loss": 0.76695126,
      "learning_rate": 3.5325831248603533e-07,
      "loss": 0.78794903,
      "num_input_tokens_seen": 292051800,
      "step": 13533,
      "time_per_iteration": 2.5464775562286377
    },
    {
      "auxiliary_loss_clip": 0.01108855,
      "auxiliary_loss_mlp": 0.00783941,
      "balance_loss_clip": 1.03598082,
      "balance_loss_mlp": 1.00973785,
      "epoch": 0.813708101608297,
      "flos": 22052348046720.0,
      "grad_norm": 2.055988204989701,
      "language_loss": 0.76259983,
      "learning_rate": 3.5303732276307495e-07,
      "loss": 0.78152776,
      "num_input_tokens_seen": 292072215,
      "step": 13534,
      "time_per_iteration": 2.475858449935913
    },
    {
      "auxiliary_loss_clip": 0.01094646,
      "auxiliary_loss_mlp": 0.01025133,
      "balance_loss_clip": 1.03644407,
      "balance_loss_mlp": 1.01457083,
      "epoch": 0.813768224860965,
      "flos": 16173412035840.0,
      "grad_norm": 2.0431229740037296,
      "language_loss": 0.9310838,
      "learning_rate": 3.5281639549310336e-07,
      "loss": 0.95228159,
      "num_input_tokens_seen": 292088830,
      "step": 13535,
      "time_per_iteration": 2.4759867191314697
    },
    {
      "auxiliary_loss_clip": 0.01065046,
      "auxiliary_loss_mlp": 0.01025381,
      "balance_loss_clip": 1.03605533,
      "balance_loss_mlp": 1.0139432,
      "epoch": 0.8138283481136329,
      "flos": 24352318776960.0,
      "grad_norm": 1.755710562588575,
      "language_loss": 0.70697522,
      "learning_rate": 3.52595530684499e-07,
      "loss": 0.72787952,
      "num_input_tokens_seen": 292109225,
      "step": 13536,
      "time_per_iteration": 2.6056900024414062
    },
    {
      "auxiliary_loss_clip": 0.01069448,
      "auxiliary_loss_mlp": 0.01033083,
      "balance_loss_clip": 1.0354321,
      "balance_loss_mlp": 1.02050614,
      "epoch": 0.8138884713663009,
      "flos": 25516362827520.0,
      "grad_norm": 1.782050419702371,
      "language_loss": 0.75821096,
      "learning_rate": 3.5237472834563775e-07,
      "loss": 0.77923632,
      "num_input_tokens_seen": 292129660,
      "step": 13537,
      "time_per_iteration": 2.5904340744018555
    },
    {
      "auxiliary_loss_clip": 0.01079081,
      "auxiliary_loss_mlp": 0.01034866,
      "balance_loss_clip": 1.03461862,
      "balance_loss_mlp": 1.02218175,
      "epoch": 0.8139485946189688,
      "flos": 22454368041600.0,
      "grad_norm": 1.4133561619925668,
      "language_loss": 0.76404613,
      "learning_rate": 3.5215398848489163e-07,
      "loss": 0.78518564,
      "num_input_tokens_seen": 292149090,
      "step": 13538,
      "time_per_iteration": 2.539875030517578
    },
    {
      "auxiliary_loss_clip": 0.01087134,
      "auxiliary_loss_mlp": 0.01027777,
      "balance_loss_clip": 1.0333184,
      "balance_loss_mlp": 1.01637423,
      "epoch": 0.8140087178716369,
      "flos": 21250391045760.0,
      "grad_norm": 1.5985171328388066,
      "language_loss": 0.78040045,
      "learning_rate": 3.5193331111063176e-07,
      "loss": 0.80154955,
      "num_input_tokens_seen": 292169260,
      "step": 13539,
      "time_per_iteration": 2.4967734813690186
    },
    {
      "auxiliary_loss_clip": 0.0106309,
      "auxiliary_loss_mlp": 0.01031294,
      "balance_loss_clip": 1.03829801,
      "balance_loss_mlp": 1.02015424,
      "epoch": 0.8140688411243048,
      "flos": 39415730774400.0,
      "grad_norm": 2.5391137604485783,
      "language_loss": 0.66308838,
      "learning_rate": 3.5171269623122533e-07,
      "loss": 0.6840322,
      "num_input_tokens_seen": 292188145,
      "step": 13540,
      "time_per_iteration": 2.7033023834228516
    },
    {
      "auxiliary_loss_clip": 0.01096167,
      "auxiliary_loss_mlp": 0.01030975,
      "balance_loss_clip": 1.03661299,
      "balance_loss_mlp": 1.01987624,
      "epoch": 0.8141289643769728,
      "flos": 25415885508480.0,
      "grad_norm": 1.5519137507032548,
      "language_loss": 0.67653126,
      "learning_rate": 3.5149214385503913e-07,
      "loss": 0.69780266,
      "num_input_tokens_seen": 292212135,
      "step": 13541,
      "time_per_iteration": 2.564786434173584
    },
    {
      "auxiliary_loss_clip": 0.01101681,
      "auxiliary_loss_mlp": 0.01032876,
      "balance_loss_clip": 1.0338068,
      "balance_loss_mlp": 1.0203886,
      "epoch": 0.8141890876296408,
      "flos": 12568053237120.0,
      "grad_norm": 1.7719689065839233,
      "language_loss": 0.68919861,
      "learning_rate": 3.512716539904355e-07,
      "loss": 0.71054423,
      "num_input_tokens_seen": 292230645,
      "step": 13542,
      "time_per_iteration": 2.4474687576293945
    },
    {
      "auxiliary_loss_clip": 0.01107201,
      "auxiliary_loss_mlp": 0.0103377,
      "balance_loss_clip": 1.03469825,
      "balance_loss_mlp": 1.0209012,
      "epoch": 0.8142492108823087,
      "flos": 14967172483200.0,
      "grad_norm": 2.912462067731349,
      "language_loss": 0.79462683,
      "learning_rate": 3.5105122664577613e-07,
      "loss": 0.81603658,
      "num_input_tokens_seen": 292243540,
      "step": 13543,
      "time_per_iteration": 2.3833887577056885
    },
    {
      "auxiliary_loss_clip": 0.01078032,
      "auxiliary_loss_mlp": 0.01037846,
      "balance_loss_clip": 1.03728092,
      "balance_loss_mlp": 1.02463722,
      "epoch": 0.8143093341349767,
      "flos": 12422004537600.0,
      "grad_norm": 2.090323159783881,
      "language_loss": 0.77462351,
      "learning_rate": 3.5083086182942003e-07,
      "loss": 0.79578233,
      "num_input_tokens_seen": 292261715,
      "step": 13544,
      "time_per_iteration": 2.5320842266082764
    },
    {
      "auxiliary_loss_clip": 0.01112318,
      "auxiliary_loss_mlp": 0.01032004,
      "balance_loss_clip": 1.03751612,
      "balance_loss_mlp": 1.01809263,
      "epoch": 0.8143694573876447,
      "flos": 11910564737280.0,
      "grad_norm": 3.2820749636441406,
      "language_loss": 0.73307717,
      "learning_rate": 3.5061055954972264e-07,
      "loss": 0.75452036,
      "num_input_tokens_seen": 292275080,
      "step": 13545,
      "time_per_iteration": 2.401869773864746
    },
    {
      "auxiliary_loss_clip": 0.0109159,
      "auxiliary_loss_mlp": 0.01029866,
      "balance_loss_clip": 1.0350585,
      "balance_loss_mlp": 1.01804614,
      "epoch": 0.8144295806403127,
      "flos": 21212900225280.0,
      "grad_norm": 1.6890185228899988,
      "language_loss": 0.77024072,
      "learning_rate": 3.5039031981503776e-07,
      "loss": 0.79145527,
      "num_input_tokens_seen": 292294635,
      "step": 13546,
      "time_per_iteration": 2.4939804077148438
    },
    {
      "auxiliary_loss_clip": 0.01096359,
      "auxiliary_loss_mlp": 0.01031826,
      "balance_loss_clip": 1.03724396,
      "balance_loss_mlp": 1.02047157,
      "epoch": 0.8144897038929806,
      "flos": 19865280741120.0,
      "grad_norm": 3.5037280246615166,
      "language_loss": 0.70575327,
      "learning_rate": 3.501701426337178e-07,
      "loss": 0.72703511,
      "num_input_tokens_seen": 292312695,
      "step": 13547,
      "time_per_iteration": 2.470778226852417
    },
    {
      "auxiliary_loss_clip": 0.01107384,
      "auxiliary_loss_mlp": 0.01032735,
      "balance_loss_clip": 1.03612947,
      "balance_loss_mlp": 1.01920462,
      "epoch": 0.8145498271456486,
      "flos": 24571733005440.0,
      "grad_norm": 2.346199082109253,
      "language_loss": 0.70938724,
      "learning_rate": 3.49950028014111e-07,
      "loss": 0.73078847,
      "num_input_tokens_seen": 292332005,
      "step": 13548,
      "time_per_iteration": 2.4791665077209473
    },
    {
      "auxiliary_loss_clip": 0.01095544,
      "auxiliary_loss_mlp": 0.01034878,
      "balance_loss_clip": 1.03949201,
      "balance_loss_mlp": 1.02166986,
      "epoch": 0.8146099503983165,
      "flos": 20193037367040.0,
      "grad_norm": 2.184491874714152,
      "language_loss": 0.7702992,
      "learning_rate": 3.4972997596456444e-07,
      "loss": 0.79160351,
      "num_input_tokens_seen": 292348365,
      "step": 13549,
      "time_per_iteration": 2.4777755737304688
    },
    {
      "auxiliary_loss_clip": 0.01106883,
      "auxiliary_loss_mlp": 0.01031068,
      "balance_loss_clip": 1.03761077,
      "balance_loss_mlp": 1.01922441,
      "epoch": 0.8146700736509845,
      "flos": 19536949497600.0,
      "grad_norm": 1.981281576171727,
      "language_loss": 0.70936334,
      "learning_rate": 3.4950998649342233e-07,
      "loss": 0.73074281,
      "num_input_tokens_seen": 292368050,
      "step": 13550,
      "time_per_iteration": 2.4631240367889404
    },
    {
      "auxiliary_loss_clip": 0.01089189,
      "auxiliary_loss_mlp": 0.01026405,
      "balance_loss_clip": 1.03480601,
      "balance_loss_mlp": 1.01476479,
      "epoch": 0.8147301969036524,
      "flos": 18041341979520.0,
      "grad_norm": 1.7136808585282315,
      "language_loss": 0.7237668,
      "learning_rate": 3.4929005960902826e-07,
      "loss": 0.74492276,
      "num_input_tokens_seen": 292385315,
      "step": 13551,
      "time_per_iteration": 2.4771616458892822
    },
    {
      "auxiliary_loss_clip": 0.01067841,
      "auxiliary_loss_mlp": 0.01036273,
      "balance_loss_clip": 1.03745413,
      "balance_loss_mlp": 1.02292776,
      "epoch": 0.8147903201563205,
      "flos": 18004713085440.0,
      "grad_norm": 1.9190173988509487,
      "language_loss": 0.68067896,
      "learning_rate": 3.4907019531971926e-07,
      "loss": 0.70172012,
      "num_input_tokens_seen": 292403375,
      "step": 13552,
      "time_per_iteration": 2.5980613231658936
    },
    {
      "auxiliary_loss_clip": 0.01103376,
      "auxiliary_loss_mlp": 0.01039227,
      "balance_loss_clip": 1.03444028,
      "balance_loss_mlp": 1.02730072,
      "epoch": 0.8148504434089884,
      "flos": 20259327916800.0,
      "grad_norm": 1.9177677032248008,
      "language_loss": 0.82577753,
      "learning_rate": 3.4885039363383407e-07,
      "loss": 0.84720349,
      "num_input_tokens_seen": 292419260,
      "step": 13553,
      "time_per_iteration": 2.4379117488861084
    },
    {
      "auxiliary_loss_clip": 0.01092757,
      "auxiliary_loss_mlp": 0.01027849,
      "balance_loss_clip": 1.03333664,
      "balance_loss_mlp": 1.01551712,
      "epoch": 0.8149105666616564,
      "flos": 12494723621760.0,
      "grad_norm": 2.0469017364000694,
      "language_loss": 0.67756009,
      "learning_rate": 3.4863065455970795e-07,
      "loss": 0.69876611,
      "num_input_tokens_seen": 292436095,
      "step": 13554,
      "time_per_iteration": 2.457012414932251
    },
    {
      "auxiliary_loss_clip": 0.01079537,
      "auxiliary_loss_mlp": 0.01036998,
      "balance_loss_clip": 1.03630757,
      "balance_loss_mlp": 1.02286577,
      "epoch": 0.8149706899143244,
      "flos": 32523683662080.0,
      "grad_norm": 1.9934282604056042,
      "language_loss": 0.66547179,
      "learning_rate": 3.484109781056723e-07,
      "loss": 0.68663716,
      "num_input_tokens_seen": 292457190,
      "step": 13555,
      "time_per_iteration": 2.5819034576416016
    },
    {
      "auxiliary_loss_clip": 0.01097736,
      "auxiliary_loss_mlp": 0.01033368,
      "balance_loss_clip": 1.03538704,
      "balance_loss_mlp": 1.02070785,
      "epoch": 0.8150308131669923,
      "flos": 19386088375680.0,
      "grad_norm": 2.1312635431446947,
      "language_loss": 0.73022044,
      "learning_rate": 3.4819136428005844e-07,
      "loss": 0.75153154,
      "num_input_tokens_seen": 292474300,
      "step": 13556,
      "time_per_iteration": 2.489006519317627
    },
    {
      "auxiliary_loss_clip": 0.01094839,
      "auxiliary_loss_mlp": 0.01029001,
      "balance_loss_clip": 1.03802407,
      "balance_loss_mlp": 1.01787317,
      "epoch": 0.8150909364196604,
      "flos": 17421380213760.0,
      "grad_norm": 1.5875672247731132,
      "language_loss": 0.80572408,
      "learning_rate": 3.4797181309119307e-07,
      "loss": 0.82696253,
      "num_input_tokens_seen": 292492420,
      "step": 13557,
      "time_per_iteration": 3.8833038806915283
    },
    {
      "auxiliary_loss_clip": 0.01087686,
      "auxiliary_loss_mlp": 0.01033633,
      "balance_loss_clip": 1.03637493,
      "balance_loss_mlp": 1.02120507,
      "epoch": 0.8151510596723283,
      "flos": 27162795553920.0,
      "grad_norm": 1.691390466246969,
      "language_loss": 0.65936208,
      "learning_rate": 3.4775232454740255e-07,
      "loss": 0.68057525,
      "num_input_tokens_seen": 292512895,
      "step": 13558,
      "time_per_iteration": 3.944021224975586
    },
    {
      "auxiliary_loss_clip": 0.01029646,
      "auxiliary_loss_mlp": 0.01002517,
      "balance_loss_clip": 1.00677872,
      "balance_loss_mlp": 1.00128353,
      "epoch": 0.8152111829249963,
      "flos": 64219052718720.0,
      "grad_norm": 0.7897487522819733,
      "language_loss": 0.56933069,
      "learning_rate": 3.4753289865700896e-07,
      "loss": 0.5896523,
      "num_input_tokens_seen": 292566580,
      "step": 13559,
      "time_per_iteration": 2.9919676780700684
    },
    {
      "auxiliary_loss_clip": 0.01018139,
      "auxiliary_loss_mlp": 0.01020772,
      "balance_loss_clip": 1.01424897,
      "balance_loss_mlp": 1.01937759,
      "epoch": 0.8152713061776642,
      "flos": 67072012306560.0,
      "grad_norm": 0.6816132319102937,
      "language_loss": 0.55299163,
      "learning_rate": 3.473135354283334e-07,
      "loss": 0.57338071,
      "num_input_tokens_seen": 292621490,
      "step": 13560,
      "time_per_iteration": 4.406762361526489
    },
    {
      "auxiliary_loss_clip": 0.01080119,
      "auxiliary_loss_mlp": 0.01028336,
      "balance_loss_clip": 1.03228021,
      "balance_loss_mlp": 1.01683259,
      "epoch": 0.8153314294303322,
      "flos": 14391130072320.0,
      "grad_norm": 1.650364870230622,
      "language_loss": 0.6755268,
      "learning_rate": 3.470942348696948e-07,
      "loss": 0.69661129,
      "num_input_tokens_seen": 292638660,
      "step": 13561,
      "time_per_iteration": 2.4823617935180664
    },
    {
      "auxiliary_loss_clip": 0.0109665,
      "auxiliary_loss_mlp": 0.01032573,
      "balance_loss_clip": 1.03571463,
      "balance_loss_mlp": 1.02083707,
      "epoch": 0.8153915526830001,
      "flos": 25623520076160.0,
      "grad_norm": 1.5872994748801734,
      "language_loss": 0.81438971,
      "learning_rate": 3.468749969894085e-07,
      "loss": 0.83568192,
      "num_input_tokens_seen": 292658545,
      "step": 13562,
      "time_per_iteration": 2.5219523906707764
    },
    {
      "auxiliary_loss_clip": 0.01078577,
      "auxiliary_loss_mlp": 0.01029168,
      "balance_loss_clip": 1.03492498,
      "balance_loss_mlp": 1.01721156,
      "epoch": 0.8154516759356681,
      "flos": 23369156640000.0,
      "grad_norm": 1.530478104990854,
      "language_loss": 0.71842325,
      "learning_rate": 3.4665582179578734e-07,
      "loss": 0.73950064,
      "num_input_tokens_seen": 292678460,
      "step": 13563,
      "time_per_iteration": 2.548308849334717
    },
    {
      "auxiliary_loss_clip": 0.01026628,
      "auxiliary_loss_mlp": 0.01027793,
      "balance_loss_clip": 1.03701079,
      "balance_loss_mlp": 1.01469779,
      "epoch": 0.815511799188336,
      "flos": 28149189914880.0,
      "grad_norm": 1.5818168745473526,
      "language_loss": 0.70124841,
      "learning_rate": 3.4643670929714387e-07,
      "loss": 0.72179264,
      "num_input_tokens_seen": 292699815,
      "step": 13564,
      "time_per_iteration": 2.859358549118042
    },
    {
      "auxiliary_loss_clip": 0.01078795,
      "auxiliary_loss_mlp": 0.01027546,
      "balance_loss_clip": 1.03570282,
      "balance_loss_mlp": 1.01577377,
      "epoch": 0.8155719224410041,
      "flos": 16983413683200.0,
      "grad_norm": 2.277187121338606,
      "language_loss": 0.70516682,
      "learning_rate": 3.462176595017854e-07,
      "loss": 0.72623026,
      "num_input_tokens_seen": 292717370,
      "step": 13565,
      "time_per_iteration": 2.8851611614227295
    },
    {
      "auxiliary_loss_clip": 0.01094003,
      "auxiliary_loss_mlp": 0.01033756,
      "balance_loss_clip": 1.03637087,
      "balance_loss_mlp": 1.02175736,
      "epoch": 0.815632045693672,
      "flos": 24681727428480.0,
      "grad_norm": 1.6217543820642966,
      "language_loss": 0.79120386,
      "learning_rate": 3.459986724180188e-07,
      "loss": 0.81248146,
      "num_input_tokens_seen": 292737110,
      "step": 13566,
      "time_per_iteration": 3.900590658187866
    },
    {
      "auxiliary_loss_clip": 0.01083001,
      "auxiliary_loss_mlp": 0.01032428,
      "balance_loss_clip": 1.03693962,
      "balance_loss_mlp": 1.02127564,
      "epoch": 0.81569216894634,
      "flos": 19938323047680.0,
      "grad_norm": 1.6011830008830816,
      "language_loss": 0.82385451,
      "learning_rate": 3.457797480541491e-07,
      "loss": 0.84500873,
      "num_input_tokens_seen": 292756510,
      "step": 13567,
      "time_per_iteration": 2.5866122245788574
    },
    {
      "auxiliary_loss_clip": 0.0110162,
      "auxiliary_loss_mlp": 0.01026908,
      "balance_loss_clip": 1.0354867,
      "balance_loss_mlp": 1.01633406,
      "epoch": 0.8157522921990079,
      "flos": 21799393493760.0,
      "grad_norm": 2.0364336640505796,
      "language_loss": 0.79666567,
      "learning_rate": 3.455608864184771e-07,
      "loss": 0.81795096,
      "num_input_tokens_seen": 292776710,
      "step": 13568,
      "time_per_iteration": 2.451289415359497
    },
    {
      "auxiliary_loss_clip": 0.0108058,
      "auxiliary_loss_mlp": 0.01029626,
      "balance_loss_clip": 1.03389859,
      "balance_loss_mlp": 1.0182054,
      "epoch": 0.8158124154516759,
      "flos": 18508323720960.0,
      "grad_norm": 1.7727788120967456,
      "language_loss": 0.77277756,
      "learning_rate": 3.453420875193016e-07,
      "loss": 0.79387963,
      "num_input_tokens_seen": 292794350,
      "step": 13569,
      "time_per_iteration": 2.5868709087371826
    },
    {
      "auxiliary_loss_clip": 0.01101606,
      "auxiliary_loss_mlp": 0.01035502,
      "balance_loss_clip": 1.0342871,
      "balance_loss_mlp": 1.02450454,
      "epoch": 0.815872538704344,
      "flos": 26830801123200.0,
      "grad_norm": 1.9361530607548736,
      "language_loss": 0.58307368,
      "learning_rate": 3.451233513649199e-07,
      "loss": 0.6044448,
      "num_input_tokens_seen": 292814005,
      "step": 13570,
      "time_per_iteration": 2.4858391284942627
    },
    {
      "auxiliary_loss_clip": 0.01096255,
      "auxiliary_loss_mlp": 0.01040101,
      "balance_loss_clip": 1.03535354,
      "balance_loss_mlp": 1.02698827,
      "epoch": 0.8159326619570119,
      "flos": 21725704742400.0,
      "grad_norm": 1.7553400442274638,
      "language_loss": 0.82363737,
      "learning_rate": 3.4490467796362687e-07,
      "loss": 0.84500086,
      "num_input_tokens_seen": 292833485,
      "step": 13571,
      "time_per_iteration": 2.4974112510681152
    },
    {
      "auxiliary_loss_clip": 0.01077592,
      "auxiliary_loss_mlp": 0.01041896,
      "balance_loss_clip": 1.03449714,
      "balance_loss_mlp": 1.02922988,
      "epoch": 0.8159927852096799,
      "flos": 13840726993920.0,
      "grad_norm": 2.039274900691867,
      "language_loss": 0.78633189,
      "learning_rate": 3.446860673237142e-07,
      "loss": 0.80752671,
      "num_input_tokens_seen": 292848045,
      "step": 13572,
      "time_per_iteration": 2.473050117492676
    },
    {
      "auxiliary_loss_clip": 0.01104715,
      "auxiliary_loss_mlp": 0.01031012,
      "balance_loss_clip": 1.03492844,
      "balance_loss_mlp": 1.01947284,
      "epoch": 0.8160529084623478,
      "flos": 24499516711680.0,
      "grad_norm": 1.5746125153205963,
      "language_loss": 0.65153193,
      "learning_rate": 3.4446751945347186e-07,
      "loss": 0.67288917,
      "num_input_tokens_seen": 292869965,
      "step": 13573,
      "time_per_iteration": 2.4947822093963623
    },
    {
      "auxiliary_loss_clip": 0.01070467,
      "auxiliary_loss_mlp": 0.01029537,
      "balance_loss_clip": 1.03731322,
      "balance_loss_mlp": 1.01864767,
      "epoch": 0.8161130317150158,
      "flos": 24826339584000.0,
      "grad_norm": 1.4473009587498493,
      "language_loss": 0.75349635,
      "learning_rate": 3.442490343611868e-07,
      "loss": 0.77449644,
      "num_input_tokens_seen": 292889680,
      "step": 13574,
      "time_per_iteration": 2.655406951904297
    },
    {
      "auxiliary_loss_clip": 0.01096649,
      "auxiliary_loss_mlp": 0.01033733,
      "balance_loss_clip": 1.03601217,
      "balance_loss_mlp": 1.02149057,
      "epoch": 0.8161731549676837,
      "flos": 30956542208640.0,
      "grad_norm": 1.6481025268391045,
      "language_loss": 0.59924006,
      "learning_rate": 3.4403061205514485e-07,
      "loss": 0.6205439,
      "num_input_tokens_seen": 292912360,
      "step": 13575,
      "time_per_iteration": 2.5666818618774414
    },
    {
      "auxiliary_loss_clip": 0.01027042,
      "auxiliary_loss_mlp": 0.01035343,
      "balance_loss_clip": 1.03236318,
      "balance_loss_mlp": 1.02124012,
      "epoch": 0.8162332782203517,
      "flos": 18551991680640.0,
      "grad_norm": 1.8122412548274236,
      "language_loss": 0.7423265,
      "learning_rate": 3.4381225254362736e-07,
      "loss": 0.7629503,
      "num_input_tokens_seen": 292928325,
      "step": 13576,
      "time_per_iteration": 2.6165342330932617
    },
    {
      "auxiliary_loss_clip": 0.01010756,
      "auxiliary_loss_mlp": 0.01002977,
      "balance_loss_clip": 1.00870681,
      "balance_loss_mlp": 1.00184441,
      "epoch": 0.8162934014730197,
      "flos": 70386853904640.0,
      "grad_norm": 0.8273644605478022,
      "language_loss": 0.58575475,
      "learning_rate": 3.435939558349155e-07,
      "loss": 0.60589218,
      "num_input_tokens_seen": 292992795,
      "step": 13577,
      "time_per_iteration": 3.1288561820983887
    },
    {
      "auxiliary_loss_clip": 0.01052373,
      "auxiliary_loss_mlp": 0.01028773,
      "balance_loss_clip": 1.03184271,
      "balance_loss_mlp": 1.01651216,
      "epoch": 0.8163535247256877,
      "flos": 21214839559680.0,
      "grad_norm": 1.66057969196801,
      "language_loss": 0.70970964,
      "learning_rate": 3.4337572193728747e-07,
      "loss": 0.73052114,
      "num_input_tokens_seen": 293011950,
      "step": 13578,
      "time_per_iteration": 2.6052675247192383
    },
    {
      "auxiliary_loss_clip": 0.01062408,
      "auxiliary_loss_mlp": 0.01028978,
      "balance_loss_clip": 1.03313494,
      "balance_loss_mlp": 1.01755786,
      "epoch": 0.8164136479783556,
      "flos": 21098847565440.0,
      "grad_norm": 1.7132886166005252,
      "language_loss": 0.73684269,
      "learning_rate": 3.431575508590172e-07,
      "loss": 0.75775659,
      "num_input_tokens_seen": 293030175,
      "step": 13579,
      "time_per_iteration": 2.537247657775879
    },
    {
      "auxiliary_loss_clip": 0.01105385,
      "auxiliary_loss_mlp": 0.01029003,
      "balance_loss_clip": 1.03530133,
      "balance_loss_mlp": 1.01761258,
      "epoch": 0.8164737712310236,
      "flos": 21720640924800.0,
      "grad_norm": 1.943652330688153,
      "language_loss": 0.78824568,
      "learning_rate": 3.4293944260837873e-07,
      "loss": 0.80958951,
      "num_input_tokens_seen": 293047980,
      "step": 13580,
      "time_per_iteration": 2.467878818511963
    },
    {
      "auxiliary_loss_clip": 0.01067024,
      "auxiliary_loss_mlp": 0.01037344,
      "balance_loss_clip": 1.03240097,
      "balance_loss_mlp": 1.02484465,
      "epoch": 0.8165338944836915,
      "flos": 19536805843200.0,
      "grad_norm": 1.677196354206817,
      "language_loss": 0.68868029,
      "learning_rate": 3.4272139719364314e-07,
      "loss": 0.70972407,
      "num_input_tokens_seen": 293067030,
      "step": 13581,
      "time_per_iteration": 2.535973072052002
    },
    {
      "auxiliary_loss_clip": 0.01104102,
      "auxiliary_loss_mlp": 0.0102708,
      "balance_loss_clip": 1.0353446,
      "balance_loss_mlp": 1.0152719,
      "epoch": 0.8165940177363595,
      "flos": 22928568416640.0,
      "grad_norm": 1.714409022979012,
      "language_loss": 0.59906179,
      "learning_rate": 3.4250341462307786e-07,
      "loss": 0.62037361,
      "num_input_tokens_seen": 293085575,
      "step": 13582,
      "time_per_iteration": 2.4696834087371826
    },
    {
      "auxiliary_loss_clip": 0.01068726,
      "auxiliary_loss_mlp": 0.0078309,
      "balance_loss_clip": 1.03412199,
      "balance_loss_mlp": 1.00901771,
      "epoch": 0.8166541409890276,
      "flos": 23370377702400.0,
      "grad_norm": 1.340799806078612,
      "language_loss": 0.82214427,
      "learning_rate": 3.4228549490494897e-07,
      "loss": 0.84066248,
      "num_input_tokens_seen": 293108200,
      "step": 13583,
      "time_per_iteration": 2.6061742305755615
    },
    {
      "auxiliary_loss_clip": 0.01082284,
      "auxiliary_loss_mlp": 0.01030523,
      "balance_loss_clip": 1.0332098,
      "balance_loss_mlp": 1.01864338,
      "epoch": 0.8167142642416955,
      "flos": 18441997257600.0,
      "grad_norm": 2.9637558515009395,
      "language_loss": 0.74376541,
      "learning_rate": 3.4206763804752093e-07,
      "loss": 0.76489353,
      "num_input_tokens_seen": 293126020,
      "step": 13584,
      "time_per_iteration": 2.4934563636779785
    },
    {
      "auxiliary_loss_clip": 0.01096263,
      "auxiliary_loss_mlp": 0.01027596,
      "balance_loss_clip": 1.03844726,
      "balance_loss_mlp": 1.01558614,
      "epoch": 0.8167743874943635,
      "flos": 21214983214080.0,
      "grad_norm": 1.5663492417279683,
      "language_loss": 0.74493474,
      "learning_rate": 3.4184984405905405e-07,
      "loss": 0.7661733,
      "num_input_tokens_seen": 293144620,
      "step": 13585,
      "time_per_iteration": 2.4752769470214844
    },
    {
      "auxiliary_loss_clip": 0.010778,
      "auxiliary_loss_mlp": 0.01032337,
      "balance_loss_clip": 1.03607142,
      "balance_loss_mlp": 1.0201714,
      "epoch": 0.8168345107470314,
      "flos": 18697681244160.0,
      "grad_norm": 1.7796479867452657,
      "language_loss": 0.69661337,
      "learning_rate": 3.416321129478068e-07,
      "loss": 0.71771473,
      "num_input_tokens_seen": 293162850,
      "step": 13586,
      "time_per_iteration": 2.526461362838745
    },
    {
      "auxiliary_loss_clip": 0.01045522,
      "auxiliary_loss_mlp": 0.01038424,
      "balance_loss_clip": 1.03319073,
      "balance_loss_mlp": 1.02633035,
      "epoch": 0.8168946339996994,
      "flos": 16253098358400.0,
      "grad_norm": 7.028573781724938,
      "language_loss": 0.60927391,
      "learning_rate": 3.4141444472203594e-07,
      "loss": 0.63011336,
      "num_input_tokens_seen": 293181620,
      "step": 13587,
      "time_per_iteration": 2.5828168392181396
    },
    {
      "auxiliary_loss_clip": 0.01095264,
      "auxiliary_loss_mlp": 0.01031294,
      "balance_loss_clip": 1.03410578,
      "balance_loss_mlp": 1.01874101,
      "epoch": 0.8169547572523673,
      "flos": 26941585645440.0,
      "grad_norm": 2.7308603473180444,
      "language_loss": 0.68997347,
      "learning_rate": 3.4119683938999624e-07,
      "loss": 0.71123904,
      "num_input_tokens_seen": 293200270,
      "step": 13588,
      "time_per_iteration": 2.559894561767578
    },
    {
      "auxiliary_loss_clip": 0.01080969,
      "auxiliary_loss_mlp": 0.01035938,
      "balance_loss_clip": 1.03496826,
      "balance_loss_mlp": 1.02207422,
      "epoch": 0.8170148805050353,
      "flos": 18952323736320.0,
      "grad_norm": 1.4902606645104202,
      "language_loss": 0.73181069,
      "learning_rate": 3.4097929695993854e-07,
      "loss": 0.75297976,
      "num_input_tokens_seen": 293218960,
      "step": 13589,
      "time_per_iteration": 2.5112459659576416
    },
    {
      "auxiliary_loss_clip": 0.01088059,
      "auxiliary_loss_mlp": 0.01030566,
      "balance_loss_clip": 1.03600883,
      "balance_loss_mlp": 1.01877594,
      "epoch": 0.8170750037577033,
      "flos": 21834909066240.0,
      "grad_norm": 1.8514627965744839,
      "language_loss": 0.73420519,
      "learning_rate": 3.4076181744011166e-07,
      "loss": 0.75539142,
      "num_input_tokens_seen": 293236450,
      "step": 13590,
      "time_per_iteration": 2.5109808444976807
    },
    {
      "auxiliary_loss_clip": 0.01108094,
      "auxiliary_loss_mlp": 0.0103311,
      "balance_loss_clip": 1.03582394,
      "balance_loss_mlp": 1.01953816,
      "epoch": 0.8171351270103713,
      "flos": 33507169021440.0,
      "grad_norm": 2.1641791270026256,
      "language_loss": 0.65585053,
      "learning_rate": 3.4054440083876345e-07,
      "loss": 0.67726254,
      "num_input_tokens_seen": 293256480,
      "step": 13591,
      "time_per_iteration": 2.5616204738616943
    },
    {
      "auxiliary_loss_clip": 0.01106105,
      "auxiliary_loss_mlp": 0.01034616,
      "balance_loss_clip": 1.03462934,
      "balance_loss_mlp": 1.02267134,
      "epoch": 0.8171952502630392,
      "flos": 22708184520960.0,
      "grad_norm": 2.850677771664054,
      "language_loss": 0.6832099,
      "learning_rate": 3.403270471641373e-07,
      "loss": 0.70461708,
      "num_input_tokens_seen": 293274960,
      "step": 13592,
      "time_per_iteration": 2.4719650745391846
    },
    {
      "auxiliary_loss_clip": 0.0108137,
      "auxiliary_loss_mlp": 0.01028903,
      "balance_loss_clip": 1.0340271,
      "balance_loss_mlp": 1.0159626,
      "epoch": 0.8172553735157072,
      "flos": 26723715701760.0,
      "grad_norm": 1.9055449407063152,
      "language_loss": 0.66370976,
      "learning_rate": 3.401097564244759e-07,
      "loss": 0.68481249,
      "num_input_tokens_seen": 293295945,
      "step": 13593,
      "time_per_iteration": 2.56549334526062
    },
    {
      "auxiliary_loss_clip": 0.01092839,
      "auxiliary_loss_mlp": 0.0103202,
      "balance_loss_clip": 1.03388548,
      "balance_loss_mlp": 1.02058208,
      "epoch": 0.8173154967683751,
      "flos": 15961072786560.0,
      "grad_norm": 2.0465152479210214,
      "language_loss": 0.69500703,
      "learning_rate": 3.398925286280188e-07,
      "loss": 0.71625566,
      "num_input_tokens_seen": 293313300,
      "step": 13594,
      "time_per_iteration": 2.4884448051452637
    },
    {
      "auxiliary_loss_clip": 0.01105543,
      "auxiliary_loss_mlp": 0.01028923,
      "balance_loss_clip": 1.03545403,
      "balance_loss_mlp": 1.01696074,
      "epoch": 0.8173756200210431,
      "flos": 25986720447360.0,
      "grad_norm": 2.5974578288372605,
      "language_loss": 0.65663415,
      "learning_rate": 3.3967536378300456e-07,
      "loss": 0.67797887,
      "num_input_tokens_seen": 293333085,
      "step": 13595,
      "time_per_iteration": 3.850938081741333
    },
    {
      "auxiliary_loss_clip": 0.0107287,
      "auxiliary_loss_mlp": 0.01028601,
      "balance_loss_clip": 1.03698587,
      "balance_loss_mlp": 1.01627469,
      "epoch": 0.8174357432737112,
      "flos": 25664422688640.0,
      "grad_norm": 1.5409967877897346,
      "language_loss": 0.78580201,
      "learning_rate": 3.394582618976658e-07,
      "loss": 0.8068167,
      "num_input_tokens_seen": 293351895,
      "step": 13596,
      "time_per_iteration": 4.027684211730957
    },
    {
      "auxiliary_loss_clip": 0.01077059,
      "auxiliary_loss_mlp": 0.01027169,
      "balance_loss_clip": 1.03183174,
      "balance_loss_mlp": 1.01476562,
      "epoch": 0.8174958665263791,
      "flos": 21835088634240.0,
      "grad_norm": 3.727220442715241,
      "language_loss": 0.58424461,
      "learning_rate": 3.392412229802362e-07,
      "loss": 0.60528696,
      "num_input_tokens_seen": 293371165,
      "step": 13597,
      "time_per_iteration": 2.6559948921203613
    },
    {
      "auxiliary_loss_clip": 0.01061061,
      "auxiliary_loss_mlp": 0.01031446,
      "balance_loss_clip": 1.03440976,
      "balance_loss_mlp": 1.01978731,
      "epoch": 0.8175559897790471,
      "flos": 22455517276800.0,
      "grad_norm": 1.9424358582043217,
      "language_loss": 0.82464492,
      "learning_rate": 3.390242470389462e-07,
      "loss": 0.84556997,
      "num_input_tokens_seen": 293391150,
      "step": 13598,
      "time_per_iteration": 2.591569185256958
    },
    {
      "auxiliary_loss_clip": 0.01043308,
      "auxiliary_loss_mlp": 0.01027802,
      "balance_loss_clip": 1.03484011,
      "balance_loss_mlp": 1.0163343,
      "epoch": 0.817616113031715,
      "flos": 23615790399360.0,
      "grad_norm": 1.9952123747554074,
      "language_loss": 0.82665056,
      "learning_rate": 3.3880733408202277e-07,
      "loss": 0.84736168,
      "num_input_tokens_seen": 293409440,
      "step": 13599,
      "time_per_iteration": 4.190845251083374
    },
    {
      "auxiliary_loss_clip": 0.01055064,
      "auxiliary_loss_mlp": 0.01040916,
      "balance_loss_clip": 1.03189421,
      "balance_loss_mlp": 1.02794051,
      "epoch": 0.817676236284383,
      "flos": 27672260106240.0,
      "grad_norm": 1.7443460623456057,
      "language_loss": 0.83873987,
      "learning_rate": 3.3859048411769186e-07,
      "loss": 0.85969961,
      "num_input_tokens_seen": 293428995,
      "step": 13600,
      "time_per_iteration": 2.776707410812378
    },
    {
      "auxiliary_loss_clip": 0.0107071,
      "auxiliary_loss_mlp": 0.01033565,
      "balance_loss_clip": 1.03402996,
      "balance_loss_mlp": 1.02128649,
      "epoch": 0.8177363595370509,
      "flos": 24681009156480.0,
      "grad_norm": 1.7714872267482222,
      "language_loss": 0.7403214,
      "learning_rate": 3.383736971541766e-07,
      "loss": 0.7613641,
      "num_input_tokens_seen": 293449155,
      "step": 13601,
      "time_per_iteration": 2.5699915885925293
    },
    {
      "auxiliary_loss_clip": 0.01073289,
      "auxiliary_loss_mlp": 0.01029176,
      "balance_loss_clip": 1.03535736,
      "balance_loss_mlp": 1.01702225,
      "epoch": 0.817796482789719,
      "flos": 17346326745600.0,
      "grad_norm": 2.168196644548913,
      "language_loss": 0.67939508,
      "learning_rate": 3.3815697319969737e-07,
      "loss": 0.70041972,
      "num_input_tokens_seen": 293466125,
      "step": 13602,
      "time_per_iteration": 2.537374496459961
    },
    {
      "auxiliary_loss_clip": 0.010636,
      "auxiliary_loss_mlp": 0.01032231,
      "balance_loss_clip": 1.03252363,
      "balance_loss_mlp": 1.01946378,
      "epoch": 0.8178566060423869,
      "flos": 17778475272960.0,
      "grad_norm": 2.200777772428991,
      "language_loss": 0.83731377,
      "learning_rate": 3.379403122624718e-07,
      "loss": 0.85827208,
      "num_input_tokens_seen": 293481345,
      "step": 13603,
      "time_per_iteration": 2.500786781311035
    },
    {
      "auxiliary_loss_clip": 0.01056635,
      "auxiliary_loss_mlp": 0.01024563,
      "balance_loss_clip": 1.03612053,
      "balance_loss_mlp": 1.01325011,
      "epoch": 0.8179167292950549,
      "flos": 24973250209920.0,
      "grad_norm": 1.680642849720776,
      "language_loss": 0.68960428,
      "learning_rate": 3.377237143507159e-07,
      "loss": 0.7104162,
      "num_input_tokens_seen": 293502330,
      "step": 13604,
      "time_per_iteration": 2.635911226272583
    },
    {
      "auxiliary_loss_clip": 0.01085188,
      "auxiliary_loss_mlp": 0.01035662,
      "balance_loss_clip": 1.03779221,
      "balance_loss_mlp": 1.02321625,
      "epoch": 0.8179768525477228,
      "flos": 22856783086080.0,
      "grad_norm": 2.4624179208286563,
      "language_loss": 0.74442029,
      "learning_rate": 3.3750717947264406e-07,
      "loss": 0.76562881,
      "num_input_tokens_seen": 293521415,
      "step": 13605,
      "time_per_iteration": 3.895524501800537
    },
    {
      "auxiliary_loss_clip": 0.01065079,
      "auxiliary_loss_mlp": 0.01035472,
      "balance_loss_clip": 1.03815806,
      "balance_loss_mlp": 1.02389622,
      "epoch": 0.8180369758003908,
      "flos": 18515147304960.0,
      "grad_norm": 1.8686400263906888,
      "language_loss": 0.7424525,
      "learning_rate": 3.372907076364666e-07,
      "loss": 0.76345801,
      "num_input_tokens_seen": 293539245,
      "step": 13606,
      "time_per_iteration": 2.540588140487671
    },
    {
      "auxiliary_loss_clip": 0.01103171,
      "auxiliary_loss_mlp": 0.01030056,
      "balance_loss_clip": 1.035748,
      "balance_loss_mlp": 1.01847482,
      "epoch": 0.8180970990530587,
      "flos": 33182105915520.0,
      "grad_norm": 1.8433118813121732,
      "language_loss": 0.65450025,
      "learning_rate": 3.370742988503916e-07,
      "loss": 0.67583251,
      "num_input_tokens_seen": 293560640,
      "step": 13607,
      "time_per_iteration": 2.5554890632629395
    },
    {
      "auxiliary_loss_clip": 0.01081825,
      "auxiliary_loss_mlp": 0.01029636,
      "balance_loss_clip": 1.0347898,
      "balance_loss_mlp": 1.01719666,
      "epoch": 0.8181572223057267,
      "flos": 25010022758400.0,
      "grad_norm": 1.7730750084785125,
      "language_loss": 0.70206821,
      "learning_rate": 3.3685795312262634e-07,
      "loss": 0.72318286,
      "num_input_tokens_seen": 293579465,
      "step": 13608,
      "time_per_iteration": 2.554417371749878
    },
    {
      "auxiliary_loss_clip": 0.01089459,
      "auxiliary_loss_mlp": 0.01033008,
      "balance_loss_clip": 1.03296065,
      "balance_loss_mlp": 1.0212245,
      "epoch": 0.8182173455583948,
      "flos": 28548731871360.0,
      "grad_norm": 2.045575366651012,
      "language_loss": 0.79400682,
      "learning_rate": 3.366416704613735e-07,
      "loss": 0.81523144,
      "num_input_tokens_seen": 293600540,
      "step": 13609,
      "time_per_iteration": 2.5795366764068604
    },
    {
      "auxiliary_loss_clip": 0.01012259,
      "auxiliary_loss_mlp": 0.01002389,
      "balance_loss_clip": 1.01799703,
      "balance_loss_mlp": 1.00126886,
      "epoch": 0.8182774688110627,
      "flos": 72028043245440.0,
      "grad_norm": 0.753442347892229,
      "language_loss": 0.55942738,
      "learning_rate": 3.3642545087483544e-07,
      "loss": 0.57957387,
      "num_input_tokens_seen": 293665160,
      "step": 13610,
      "time_per_iteration": 3.23667311668396
    },
    {
      "auxiliary_loss_clip": 0.01041775,
      "auxiliary_loss_mlp": 0.00784647,
      "balance_loss_clip": 1.03211439,
      "balance_loss_mlp": 1.01157618,
      "epoch": 0.8183375920637307,
      "flos": 19755358145280.0,
      "grad_norm": 1.833400791965003,
      "language_loss": 0.77604067,
      "learning_rate": 3.362092943712107e-07,
      "loss": 0.79430485,
      "num_input_tokens_seen": 293683995,
      "step": 13611,
      "time_per_iteration": 2.7209906578063965
    },
    {
      "auxiliary_loss_clip": 0.01071679,
      "auxiliary_loss_mlp": 0.01036092,
      "balance_loss_clip": 1.03465891,
      "balance_loss_mlp": 1.0224781,
      "epoch": 0.8183977153163986,
      "flos": 22341895580160.0,
      "grad_norm": 2.1169755605568987,
      "language_loss": 0.77135003,
      "learning_rate": 3.3599320095869745e-07,
      "loss": 0.79242772,
      "num_input_tokens_seen": 293704115,
      "step": 13612,
      "time_per_iteration": 2.6320464611053467
    },
    {
      "auxiliary_loss_clip": 0.01068066,
      "auxiliary_loss_mlp": 0.01024943,
      "balance_loss_clip": 1.03301203,
      "balance_loss_mlp": 1.01334381,
      "epoch": 0.8184578385690666,
      "flos": 17712472032000.0,
      "grad_norm": 1.792583756654176,
      "language_loss": 0.86167949,
      "learning_rate": 3.3577717064548793e-07,
      "loss": 0.88260961,
      "num_input_tokens_seen": 293722225,
      "step": 13613,
      "time_per_iteration": 2.619736671447754
    },
    {
      "auxiliary_loss_clip": 0.01093632,
      "auxiliary_loss_mlp": 0.01041825,
      "balance_loss_clip": 1.03639579,
      "balance_loss_mlp": 1.03053021,
      "epoch": 0.8185179618217345,
      "flos": 25701159323520.0,
      "grad_norm": 2.318425706684189,
      "language_loss": 0.72771609,
      "learning_rate": 3.355612034397746e-07,
      "loss": 0.7490707,
      "num_input_tokens_seen": 293743995,
      "step": 13614,
      "time_per_iteration": 2.5543928146362305
    },
    {
      "auxiliary_loss_clip": 0.01082808,
      "auxiliary_loss_mlp": 0.01037069,
      "balance_loss_clip": 1.03510046,
      "balance_loss_mlp": 1.02466547,
      "epoch": 0.8185780850744026,
      "flos": 25960326929280.0,
      "grad_norm": 1.482047088190772,
      "language_loss": 0.80885154,
      "learning_rate": 3.353452993497479e-07,
      "loss": 0.83005035,
      "num_input_tokens_seen": 293764935,
      "step": 13615,
      "time_per_iteration": 2.6628401279449463
    },
    {
      "auxiliary_loss_clip": 0.01090227,
      "auxiliary_loss_mlp": 0.01031873,
      "balance_loss_clip": 1.03264427,
      "balance_loss_mlp": 1.01922512,
      "epoch": 0.8186382083270705,
      "flos": 25228431406080.0,
      "grad_norm": 1.8363085873263219,
      "language_loss": 0.75634378,
      "learning_rate": 3.3512945838359375e-07,
      "loss": 0.77756476,
      "num_input_tokens_seen": 293784035,
      "step": 13616,
      "time_per_iteration": 2.5567097663879395
    },
    {
      "auxiliary_loss_clip": 0.01063369,
      "auxiliary_loss_mlp": 0.01039674,
      "balance_loss_clip": 1.03028727,
      "balance_loss_mlp": 1.02616143,
      "epoch": 0.8186983315797385,
      "flos": 22415009713920.0,
      "grad_norm": 1.679621033385655,
      "language_loss": 0.75378454,
      "learning_rate": 3.349136805494979e-07,
      "loss": 0.7748149,
      "num_input_tokens_seen": 293803360,
      "step": 13617,
      "time_per_iteration": 2.6009321212768555
    },
    {
      "auxiliary_loss_clip": 0.01075107,
      "auxiliary_loss_mlp": 0.01029677,
      "balance_loss_clip": 1.03401208,
      "balance_loss_mlp": 1.01868558,
      "epoch": 0.8187584548324064,
      "flos": 22018017623040.0,
      "grad_norm": 2.040995170042844,
      "language_loss": 0.68264806,
      "learning_rate": 3.346979658556415e-07,
      "loss": 0.70369589,
      "num_input_tokens_seen": 293821325,
      "step": 13618,
      "time_per_iteration": 2.5578482151031494
    },
    {
      "auxiliary_loss_clip": 0.01083308,
      "auxiliary_loss_mlp": 0.01030815,
      "balance_loss_clip": 1.03670156,
      "balance_loss_mlp": 1.01835728,
      "epoch": 0.8188185780850744,
      "flos": 29241664116480.0,
      "grad_norm": 1.9561942113004454,
      "language_loss": 0.69766688,
      "learning_rate": 3.344823143102058e-07,
      "loss": 0.71880811,
      "num_input_tokens_seen": 293840315,
      "step": 13619,
      "time_per_iteration": 2.633636474609375
    },
    {
      "auxiliary_loss_clip": 0.01043623,
      "auxiliary_loss_mlp": 0.01025042,
      "balance_loss_clip": 1.03748441,
      "balance_loss_mlp": 1.01344264,
      "epoch": 0.8188787013377423,
      "flos": 20696504348160.0,
      "grad_norm": 1.837527021084957,
      "language_loss": 0.73614979,
      "learning_rate": 3.3426672592136694e-07,
      "loss": 0.75683641,
      "num_input_tokens_seen": 293855685,
      "step": 13620,
      "time_per_iteration": 2.6340267658233643
    },
    {
      "auxiliary_loss_clip": 0.01075958,
      "auxiliary_loss_mlp": 0.00786201,
      "balance_loss_clip": 1.03217959,
      "balance_loss_mlp": 1.0104897,
      "epoch": 0.8189388245904103,
      "flos": 23732967542400.0,
      "grad_norm": 1.6486186623549157,
      "language_loss": 0.76210439,
      "learning_rate": 3.340512006973011e-07,
      "loss": 0.78072596,
      "num_input_tokens_seen": 293875540,
      "step": 13621,
      "time_per_iteration": 2.640263080596924
    },
    {
      "auxiliary_loss_clip": 0.01078672,
      "auxiliary_loss_mlp": 0.01027392,
      "balance_loss_clip": 1.03168559,
      "balance_loss_mlp": 1.01482105,
      "epoch": 0.8189989478430784,
      "flos": 28255090187520.0,
      "grad_norm": 2.1561168377058753,
      "language_loss": 0.65768003,
      "learning_rate": 3.3383573864618076e-07,
      "loss": 0.67874062,
      "num_input_tokens_seen": 293896570,
      "step": 13622,
      "time_per_iteration": 2.627437114715576
    },
    {
      "auxiliary_loss_clip": 0.01107787,
      "auxiliary_loss_mlp": 0.01029177,
      "balance_loss_clip": 1.03781343,
      "balance_loss_mlp": 1.01586723,
      "epoch": 0.8190590710957463,
      "flos": 21397696721280.0,
      "grad_norm": 1.7817095338263493,
      "language_loss": 0.74814135,
      "learning_rate": 3.3362033977617653e-07,
      "loss": 0.76951098,
      "num_input_tokens_seen": 293914680,
      "step": 13623,
      "time_per_iteration": 2.476555109024048
    },
    {
      "auxiliary_loss_clip": 0.01081814,
      "auxiliary_loss_mlp": 0.01038298,
      "balance_loss_clip": 1.03331041,
      "balance_loss_mlp": 1.02517951,
      "epoch": 0.8191191943484143,
      "flos": 38796451367040.0,
      "grad_norm": 2.0050310649594176,
      "language_loss": 0.63451326,
      "learning_rate": 3.3340500409545527e-07,
      "loss": 0.65571439,
      "num_input_tokens_seen": 293936480,
      "step": 13624,
      "time_per_iteration": 2.736358165740967
    },
    {
      "auxiliary_loss_clip": 0.01101912,
      "auxiliary_loss_mlp": 0.01033466,
      "balance_loss_clip": 1.03441083,
      "balance_loss_mlp": 1.0222007,
      "epoch": 0.8191793176010822,
      "flos": 25446516831360.0,
      "grad_norm": 1.521089496069259,
      "language_loss": 0.78236181,
      "learning_rate": 3.3318973161218386e-07,
      "loss": 0.80371559,
      "num_input_tokens_seen": 293957815,
      "step": 13625,
      "time_per_iteration": 2.4797842502593994
    },
    {
      "auxiliary_loss_clip": 0.01099048,
      "auxiliary_loss_mlp": 0.00784895,
      "balance_loss_clip": 1.03452766,
      "balance_loss_mlp": 1.00940824,
      "epoch": 0.8192394408537502,
      "flos": 25083029151360.0,
      "grad_norm": 1.9766321856734506,
      "language_loss": 0.76074558,
      "learning_rate": 3.329745223345244e-07,
      "loss": 0.779585,
      "num_input_tokens_seen": 293975440,
      "step": 13626,
      "time_per_iteration": 2.536447048187256
    },
    {
      "auxiliary_loss_clip": 0.01093194,
      "auxiliary_loss_mlp": 0.01035542,
      "balance_loss_clip": 1.03544378,
      "balance_loss_mlp": 1.02440178,
      "epoch": 0.8192995641064181,
      "flos": 27673732563840.0,
      "grad_norm": 1.466530954930497,
      "language_loss": 0.73513758,
      "learning_rate": 3.3275937627063823e-07,
      "loss": 0.7564249,
      "num_input_tokens_seen": 293997540,
      "step": 13627,
      "time_per_iteration": 2.522843599319458
    },
    {
      "auxiliary_loss_clip": 0.01105925,
      "auxiliary_loss_mlp": 0.01033223,
      "balance_loss_clip": 1.03607559,
      "balance_loss_mlp": 1.02100408,
      "epoch": 0.8193596873590862,
      "flos": 21288492397440.0,
      "grad_norm": 1.6861431815739538,
      "language_loss": 0.68878531,
      "learning_rate": 3.3254429342868353e-07,
      "loss": 0.71017683,
      "num_input_tokens_seen": 294017030,
      "step": 13628,
      "time_per_iteration": 2.4668054580688477
    },
    {
      "auxiliary_loss_clip": 0.01083921,
      "auxiliary_loss_mlp": 0.01036271,
      "balance_loss_clip": 1.03582215,
      "balance_loss_mlp": 1.02216899,
      "epoch": 0.8194198106117541,
      "flos": 17492626840320.0,
      "grad_norm": 1.5530157048371536,
      "language_loss": 0.85365415,
      "learning_rate": 3.323292738168171e-07,
      "loss": 0.874856,
      "num_input_tokens_seen": 294035700,
      "step": 13629,
      "time_per_iteration": 2.4703521728515625
    },
    {
      "auxiliary_loss_clip": 0.01103533,
      "auxiliary_loss_mlp": 0.01026691,
      "balance_loss_clip": 1.03544593,
      "balance_loss_mlp": 1.01521683,
      "epoch": 0.8194799338644221,
      "flos": 15267925059840.0,
      "grad_norm": 2.1003452826714963,
      "language_loss": 0.73666281,
      "learning_rate": 3.3211431744319084e-07,
      "loss": 0.75796503,
      "num_input_tokens_seen": 294049730,
      "step": 13630,
      "time_per_iteration": 2.4221086502075195
    },
    {
      "auxiliary_loss_clip": 0.01091703,
      "auxiliary_loss_mlp": 0.01034311,
      "balance_loss_clip": 1.03456259,
      "balance_loss_mlp": 1.02123427,
      "epoch": 0.81954005711709,
      "flos": 14718814871040.0,
      "grad_norm": 1.7995922731098632,
      "language_loss": 0.71850216,
      "learning_rate": 3.31899424315957e-07,
      "loss": 0.73976231,
      "num_input_tokens_seen": 294066545,
      "step": 13631,
      "time_per_iteration": 2.459794521331787
    },
    {
      "auxiliary_loss_clip": 0.01103851,
      "auxiliary_loss_mlp": 0.01031744,
      "balance_loss_clip": 1.03425431,
      "balance_loss_mlp": 1.02022195,
      "epoch": 0.819600180369758,
      "flos": 23074042498560.0,
      "grad_norm": 1.5320132319241526,
      "language_loss": 0.76738596,
      "learning_rate": 3.3168459444326447e-07,
      "loss": 0.78874195,
      "num_input_tokens_seen": 294087455,
      "step": 13632,
      "time_per_iteration": 2.515244960784912
    },
    {
      "auxiliary_loss_clip": 0.01078361,
      "auxiliary_loss_mlp": 0.01034789,
      "balance_loss_clip": 1.03277707,
      "balance_loss_mlp": 1.0226593,
      "epoch": 0.8196603036224259,
      "flos": 27599792417280.0,
      "grad_norm": 1.971867432745887,
      "language_loss": 0.65728033,
      "learning_rate": 3.314698278332588e-07,
      "loss": 0.67841178,
      "num_input_tokens_seen": 294107480,
      "step": 13633,
      "time_per_iteration": 2.5458459854125977
    },
    {
      "auxiliary_loss_clip": 0.01088514,
      "auxiliary_loss_mlp": 0.01032835,
      "balance_loss_clip": 1.03541207,
      "balance_loss_mlp": 1.02165961,
      "epoch": 0.8197204268750939,
      "flos": 28582020800640.0,
      "grad_norm": 1.5127356923893243,
      "language_loss": 0.75862938,
      "learning_rate": 3.3125512449408513e-07,
      "loss": 0.77984285,
      "num_input_tokens_seen": 294130115,
      "step": 13634,
      "time_per_iteration": 4.373814105987549
    },
    {
      "auxiliary_loss_clip": 0.01043686,
      "auxiliary_loss_mlp": 0.00782276,
      "balance_loss_clip": 1.03310311,
      "balance_loss_mlp": 1.00758839,
      "epoch": 0.819780550127762,
      "flos": 23258300290560.0,
      "grad_norm": 1.9151579934068332,
      "language_loss": 0.82046998,
      "learning_rate": 3.310404844338841e-07,
      "loss": 0.83872962,
      "num_input_tokens_seen": 294148495,
      "step": 13635,
      "time_per_iteration": 4.030940771102905
    },
    {
      "auxiliary_loss_clip": 0.01091461,
      "auxiliary_loss_mlp": 0.01032061,
      "balance_loss_clip": 1.03377008,
      "balance_loss_mlp": 1.01882255,
      "epoch": 0.8198406733804299,
      "flos": 26685255214080.0,
      "grad_norm": 1.6111697814940498,
      "language_loss": 0.75836396,
      "learning_rate": 3.308259076607949e-07,
      "loss": 0.77959919,
      "num_input_tokens_seen": 294169595,
      "step": 13636,
      "time_per_iteration": 2.515103816986084
    },
    {
      "auxiliary_loss_clip": 0.01072291,
      "auxiliary_loss_mlp": 0.01030688,
      "balance_loss_clip": 1.03309703,
      "balance_loss_mlp": 1.01873708,
      "epoch": 0.8199007966330979,
      "flos": 20084084438400.0,
      "grad_norm": 2.1047594237921263,
      "language_loss": 0.81038451,
      "learning_rate": 3.3061139418295445e-07,
      "loss": 0.83141422,
      "num_input_tokens_seen": 294183885,
      "step": 13637,
      "time_per_iteration": 2.5411343574523926
    },
    {
      "auxiliary_loss_clip": 0.01093053,
      "auxiliary_loss_mlp": 0.01030581,
      "balance_loss_clip": 1.03653443,
      "balance_loss_mlp": 1.01873136,
      "epoch": 0.8199609198857658,
      "flos": 31902788142720.0,
      "grad_norm": 2.3438146275047993,
      "language_loss": 0.71505821,
      "learning_rate": 3.3039694400849725e-07,
      "loss": 0.73629451,
      "num_input_tokens_seen": 294200150,
      "step": 13638,
      "time_per_iteration": 3.957608699798584
    },
    {
      "auxiliary_loss_clip": 0.01058661,
      "auxiliary_loss_mlp": 0.01034784,
      "balance_loss_clip": 1.03301513,
      "balance_loss_mlp": 1.02053261,
      "epoch": 0.8200210431384338,
      "flos": 26470150617600.0,
      "grad_norm": 2.072550967816083,
      "language_loss": 0.79594857,
      "learning_rate": 3.3018255714555564e-07,
      "loss": 0.81688303,
      "num_input_tokens_seen": 294220385,
      "step": 13639,
      "time_per_iteration": 2.6242294311523438
    },
    {
      "auxiliary_loss_clip": 0.01058942,
      "auxiliary_loss_mlp": 0.01030125,
      "balance_loss_clip": 1.03252077,
      "balance_loss_mlp": 1.01811492,
      "epoch": 0.8200811663911017,
      "flos": 22091454979200.0,
      "grad_norm": 1.7876894018643867,
      "language_loss": 0.78850585,
      "learning_rate": 3.299682336022589e-07,
      "loss": 0.80939651,
      "num_input_tokens_seen": 294239355,
      "step": 13640,
      "time_per_iteration": 2.5684409141540527
    },
    {
      "auxiliary_loss_clip": 0.0107034,
      "auxiliary_loss_mlp": 0.01032937,
      "balance_loss_clip": 1.03231859,
      "balance_loss_mlp": 1.02021098,
      "epoch": 0.8201412896437698,
      "flos": 37593659520000.0,
      "grad_norm": 1.9513420401955777,
      "language_loss": 0.63180339,
      "learning_rate": 3.297539733867336e-07,
      "loss": 0.65283614,
      "num_input_tokens_seen": 294259395,
      "step": 13641,
      "time_per_iteration": 2.6585452556610107
    },
    {
      "auxiliary_loss_clip": 0.01050604,
      "auxiliary_loss_mlp": 0.01026391,
      "balance_loss_clip": 1.0349462,
      "balance_loss_mlp": 1.01391602,
      "epoch": 0.8202014128964377,
      "flos": 19646333389440.0,
      "grad_norm": 1.8706528312155397,
      "language_loss": 0.73666883,
      "learning_rate": 3.295397765071055e-07,
      "loss": 0.75743878,
      "num_input_tokens_seen": 294277365,
      "step": 13642,
      "time_per_iteration": 2.592912197113037
    },
    {
      "auxiliary_loss_clip": 0.01079795,
      "auxiliary_loss_mlp": 0.01029752,
      "balance_loss_clip": 1.03697014,
      "balance_loss_mlp": 1.01834989,
      "epoch": 0.8202615361491057,
      "flos": 31467335564160.0,
      "grad_norm": 1.53471131852201,
      "language_loss": 0.7009362,
      "learning_rate": 3.2932564297149615e-07,
      "loss": 0.72203165,
      "num_input_tokens_seen": 294297555,
      "step": 13643,
      "time_per_iteration": 2.5977113246917725
    },
    {
      "auxiliary_loss_clip": 0.01092534,
      "auxiliary_loss_mlp": 0.01032413,
      "balance_loss_clip": 1.03551948,
      "balance_loss_mlp": 1.02079034,
      "epoch": 0.8203216594017736,
      "flos": 24715555061760.0,
      "grad_norm": 1.8633377340604833,
      "language_loss": 0.65547168,
      "learning_rate": 3.291115727880256e-07,
      "loss": 0.67672122,
      "num_input_tokens_seen": 294317600,
      "step": 13644,
      "time_per_iteration": 3.9514551162719727
    },
    {
      "auxiliary_loss_clip": 0.01073664,
      "auxiliary_loss_mlp": 0.01034362,
      "balance_loss_clip": 1.03498697,
      "balance_loss_mlp": 1.02239954,
      "epoch": 0.8203817826544416,
      "flos": 26031824951040.0,
      "grad_norm": 1.3976702383308246,
      "language_loss": 0.70830333,
      "learning_rate": 3.2889756596481234e-07,
      "loss": 0.72938359,
      "num_input_tokens_seen": 294340215,
      "step": 13645,
      "time_per_iteration": 2.575320243835449
    },
    {
      "auxiliary_loss_clip": 0.01078416,
      "auxiliary_loss_mlp": 0.01029354,
      "balance_loss_clip": 1.03438485,
      "balance_loss_mlp": 1.01761174,
      "epoch": 0.8204419059071095,
      "flos": 25954544839680.0,
      "grad_norm": 1.8967401147659033,
      "language_loss": 0.71471405,
      "learning_rate": 3.286836225099707e-07,
      "loss": 0.73579168,
      "num_input_tokens_seen": 294358590,
      "step": 13646,
      "time_per_iteration": 2.5499613285064697
    },
    {
      "auxiliary_loss_clip": 0.01086156,
      "auxiliary_loss_mlp": 0.01033701,
      "balance_loss_clip": 1.03536224,
      "balance_loss_mlp": 1.02135706,
      "epoch": 0.8205020291597775,
      "flos": 23580059345280.0,
      "grad_norm": 2.2342565405831243,
      "language_loss": 0.78389859,
      "learning_rate": 3.284697424316132e-07,
      "loss": 0.8050971,
      "num_input_tokens_seen": 294375825,
      "step": 13647,
      "time_per_iteration": 2.503262519836426
    },
    {
      "auxiliary_loss_clip": 0.01101734,
      "auxiliary_loss_mlp": 0.01031698,
      "balance_loss_clip": 1.03617597,
      "balance_loss_mlp": 1.01999116,
      "epoch": 0.8205621524124456,
      "flos": 26799164219520.0,
      "grad_norm": 1.2986811431378495,
      "language_loss": 0.67869347,
      "learning_rate": 3.2825592573785034e-07,
      "loss": 0.70002782,
      "num_input_tokens_seen": 294398500,
      "step": 13648,
      "time_per_iteration": 2.5273921489715576
    },
    {
      "auxiliary_loss_clip": 0.01078492,
      "auxiliary_loss_mlp": 0.01028254,
      "balance_loss_clip": 1.03319144,
      "balance_loss_mlp": 1.01599288,
      "epoch": 0.8206222756651135,
      "flos": 27527863432320.0,
      "grad_norm": 1.7185618064278274,
      "language_loss": 0.80244482,
      "learning_rate": 3.28042172436791e-07,
      "loss": 0.82351232,
      "num_input_tokens_seen": 294418840,
      "step": 13649,
      "time_per_iteration": 2.5790982246398926
    },
    {
      "auxiliary_loss_clip": 0.01096329,
      "auxiliary_loss_mlp": 0.01032914,
      "balance_loss_clip": 1.0376687,
      "balance_loss_mlp": 1.01972342,
      "epoch": 0.8206823989177815,
      "flos": 21178605715200.0,
      "grad_norm": 1.6657161046104314,
      "language_loss": 0.68771458,
      "learning_rate": 3.278284825365396e-07,
      "loss": 0.70900702,
      "num_input_tokens_seen": 294438215,
      "step": 13650,
      "time_per_iteration": 2.512507200241089
    },
    {
      "auxiliary_loss_clip": 0.01087127,
      "auxiliary_loss_mlp": 0.01034603,
      "balance_loss_clip": 1.03741634,
      "balance_loss_mlp": 1.02141809,
      "epoch": 0.8207425221704494,
      "flos": 11509622150400.0,
      "grad_norm": 2.269813050640077,
      "language_loss": 0.61126071,
      "learning_rate": 3.276148560452001e-07,
      "loss": 0.632478,
      "num_input_tokens_seen": 294455260,
      "step": 13651,
      "time_per_iteration": 2.4748425483703613
    },
    {
      "auxiliary_loss_clip": 0.01068374,
      "auxiliary_loss_mlp": 0.00788703,
      "balance_loss_clip": 1.03637385,
      "balance_loss_mlp": 1.01188755,
      "epoch": 0.8208026454231174,
      "flos": 19791987039360.0,
      "grad_norm": 2.5836343012278626,
      "language_loss": 0.71848238,
      "learning_rate": 3.2740129297087293e-07,
      "loss": 0.73705316,
      "num_input_tokens_seen": 294473205,
      "step": 13652,
      "time_per_iteration": 2.5552477836608887
    },
    {
      "auxiliary_loss_clip": 0.01078249,
      "auxiliary_loss_mlp": 0.0102843,
      "balance_loss_clip": 1.034899,
      "balance_loss_mlp": 1.01794577,
      "epoch": 0.8208627686757853,
      "flos": 15667538843520.0,
      "grad_norm": 1.8181644722760613,
      "language_loss": 0.72845685,
      "learning_rate": 3.271877933216558e-07,
      "loss": 0.74952364,
      "num_input_tokens_seen": 294490645,
      "step": 13653,
      "time_per_iteration": 2.4920341968536377
    },
    {
      "auxiliary_loss_clip": 0.01068152,
      "auxiliary_loss_mlp": 0.01033728,
      "balance_loss_clip": 1.03554118,
      "balance_loss_mlp": 1.0202868,
      "epoch": 0.8209228919284534,
      "flos": 37482659516160.0,
      "grad_norm": 2.2113260100534338,
      "language_loss": 0.62941128,
      "learning_rate": 3.269743571056451e-07,
      "loss": 0.65043008,
      "num_input_tokens_seen": 294513500,
      "step": 13654,
      "time_per_iteration": 2.7248647212982178
    },
    {
      "auxiliary_loss_clip": 0.01076452,
      "auxiliary_loss_mlp": 0.01029368,
      "balance_loss_clip": 1.03236246,
      "balance_loss_mlp": 1.01723218,
      "epoch": 0.8209830151811213,
      "flos": 23112969863040.0,
      "grad_norm": 1.4513552023399998,
      "language_loss": 0.69906998,
      "learning_rate": 3.2676098433093447e-07,
      "loss": 0.72012818,
      "num_input_tokens_seen": 294535710,
      "step": 13655,
      "time_per_iteration": 2.59395170211792
    },
    {
      "auxiliary_loss_clip": 0.01080275,
      "auxiliary_loss_mlp": 0.01034668,
      "balance_loss_clip": 1.03584003,
      "balance_loss_mlp": 1.02259779,
      "epoch": 0.8210431384337893,
      "flos": 21288169175040.0,
      "grad_norm": 4.588960162398,
      "language_loss": 0.81951815,
      "learning_rate": 3.265476750056162e-07,
      "loss": 0.84066761,
      "num_input_tokens_seen": 294554055,
      "step": 13656,
      "time_per_iteration": 2.521470308303833
    },
    {
      "auxiliary_loss_clip": 0.01076554,
      "auxiliary_loss_mlp": 0.01031652,
      "balance_loss_clip": 1.03485394,
      "balance_loss_mlp": 1.01975489,
      "epoch": 0.8211032616864572,
      "flos": 11502403516800.0,
      "grad_norm": 2.1739220150291065,
      "language_loss": 0.74113584,
      "learning_rate": 3.2633442913777654e-07,
      "loss": 0.76221794,
      "num_input_tokens_seen": 294570390,
      "step": 13657,
      "time_per_iteration": 2.501217842102051
    },
    {
      "auxiliary_loss_clip": 0.01072262,
      "auxiliary_loss_mlp": 0.01034336,
      "balance_loss_clip": 1.03376555,
      "balance_loss_mlp": 1.0221591,
      "epoch": 0.8211633849391252,
      "flos": 29821477455360.0,
      "grad_norm": 1.7048304578190787,
      "language_loss": 0.55807281,
      "learning_rate": 3.2612124673550325e-07,
      "loss": 0.57913876,
      "num_input_tokens_seen": 294593050,
      "step": 13658,
      "time_per_iteration": 2.608198404312134
    },
    {
      "auxiliary_loss_clip": 0.01050198,
      "auxiliary_loss_mlp": 0.01030105,
      "balance_loss_clip": 1.03344584,
      "balance_loss_mlp": 1.01796949,
      "epoch": 0.8212235081917931,
      "flos": 13115439573120.0,
      "grad_norm": 2.0928158106623447,
      "language_loss": 0.78947854,
      "learning_rate": 3.259081278068805e-07,
      "loss": 0.81028163,
      "num_input_tokens_seen": 294608550,
      "step": 13659,
      "time_per_iteration": 2.590679883956909
    },
    {
      "auxiliary_loss_clip": 0.01087505,
      "auxiliary_loss_mlp": 0.01024892,
      "balance_loss_clip": 1.03294051,
      "balance_loss_mlp": 1.01449049,
      "epoch": 0.8212836314444611,
      "flos": 40515351782400.0,
      "grad_norm": 1.5781714304596575,
      "language_loss": 0.60070872,
      "learning_rate": 3.256950723599887e-07,
      "loss": 0.62183273,
      "num_input_tokens_seen": 294630380,
      "step": 13660,
      "time_per_iteration": 2.6639597415924072
    },
    {
      "auxiliary_loss_clip": 0.01092637,
      "auxiliary_loss_mlp": 0.01033663,
      "balance_loss_clip": 1.0353483,
      "balance_loss_mlp": 1.02031183,
      "epoch": 0.8213437546971292,
      "flos": 18770543982720.0,
      "grad_norm": 2.554792591164921,
      "language_loss": 0.73019308,
      "learning_rate": 3.254820804029075e-07,
      "loss": 0.75145614,
      "num_input_tokens_seen": 294648655,
      "step": 13661,
      "time_per_iteration": 2.4931888580322266
    },
    {
      "auxiliary_loss_clip": 0.01094715,
      "auxiliary_loss_mlp": 0.0103066,
      "balance_loss_clip": 1.03349352,
      "balance_loss_mlp": 1.01852453,
      "epoch": 0.8214038779497971,
      "flos": 19682279925120.0,
      "grad_norm": 2.9717399954792474,
      "language_loss": 0.7487275,
      "learning_rate": 3.252691519437143e-07,
      "loss": 0.76998127,
      "num_input_tokens_seen": 294666915,
      "step": 13662,
      "time_per_iteration": 2.4629268646240234
    },
    {
      "auxiliary_loss_clip": 0.01030024,
      "auxiliary_loss_mlp": 0.01003392,
      "balance_loss_clip": 1.00718331,
      "balance_loss_mlp": 1.00215781,
      "epoch": 0.8214640012024651,
      "flos": 71602969697280.0,
      "grad_norm": 0.7649024242008559,
      "language_loss": 0.5404439,
      "learning_rate": 3.250562869904825e-07,
      "loss": 0.56077802,
      "num_input_tokens_seen": 294731545,
      "step": 13663,
      "time_per_iteration": 3.1974599361419678
    },
    {
      "auxiliary_loss_clip": 0.01059223,
      "auxiliary_loss_mlp": 0.01035769,
      "balance_loss_clip": 1.03294849,
      "balance_loss_mlp": 1.02291214,
      "epoch": 0.821524124455133,
      "flos": 14757203531520.0,
      "grad_norm": 2.0237780159437397,
      "language_loss": 0.65492028,
      "learning_rate": 3.248434855512838e-07,
      "loss": 0.67587018,
      "num_input_tokens_seen": 294748745,
      "step": 13664,
      "time_per_iteration": 2.5572309494018555
    },
    {
      "auxiliary_loss_clip": 0.01080509,
      "auxiliary_loss_mlp": 0.01029905,
      "balance_loss_clip": 1.03517604,
      "balance_loss_mlp": 1.01896787,
      "epoch": 0.821584247707801,
      "flos": 25082274965760.0,
      "grad_norm": 1.4339926563262932,
      "language_loss": 0.74935794,
      "learning_rate": 3.246307476341881e-07,
      "loss": 0.77046204,
      "num_input_tokens_seen": 294768955,
      "step": 13665,
      "time_per_iteration": 2.582526445388794
    },
    {
      "auxiliary_loss_clip": 0.01089758,
      "auxiliary_loss_mlp": 0.00785812,
      "balance_loss_clip": 1.03671193,
      "balance_loss_mlp": 1.01436138,
      "epoch": 0.8216443709604689,
      "flos": 36830701710720.0,
      "grad_norm": 2.207159008959412,
      "language_loss": 0.65348387,
      "learning_rate": 3.2441807324726256e-07,
      "loss": 0.67223954,
      "num_input_tokens_seen": 294789250,
      "step": 13666,
      "time_per_iteration": 2.6086766719818115
    },
    {
      "auxiliary_loss_clip": 0.01055472,
      "auxiliary_loss_mlp": 0.01028959,
      "balance_loss_clip": 1.03491378,
      "balance_loss_mlp": 1.01772332,
      "epoch": 0.821704494213137,
      "flos": 25081808088960.0,
      "grad_norm": 1.680661279867439,
      "language_loss": 0.77221292,
      "learning_rate": 3.2420546239857174e-07,
      "loss": 0.7930572,
      "num_input_tokens_seen": 294809760,
      "step": 13667,
      "time_per_iteration": 2.637730836868286
    },
    {
      "auxiliary_loss_clip": 0.01072407,
      "auxiliary_loss_mlp": 0.01034544,
      "balance_loss_clip": 1.03586555,
      "balance_loss_mlp": 1.02197945,
      "epoch": 0.8217646174658049,
      "flos": 14356117290240.0,
      "grad_norm": 1.8985613780444102,
      "language_loss": 0.77127242,
      "learning_rate": 3.239929150961773e-07,
      "loss": 0.79234195,
      "num_input_tokens_seen": 294826495,
      "step": 13668,
      "time_per_iteration": 2.5103647708892822
    },
    {
      "auxiliary_loss_clip": 0.01057666,
      "auxiliary_loss_mlp": 0.01029586,
      "balance_loss_clip": 1.03337479,
      "balance_loss_mlp": 1.01808226,
      "epoch": 0.8218247407184729,
      "flos": 22090557139200.0,
      "grad_norm": 2.0650135392651205,
      "language_loss": 0.73470962,
      "learning_rate": 3.2378043134813984e-07,
      "loss": 0.75558209,
      "num_input_tokens_seen": 294845370,
      "step": 13669,
      "time_per_iteration": 2.6220126152038574
    },
    {
      "auxiliary_loss_clip": 0.01092746,
      "auxiliary_loss_mlp": 0.01024615,
      "balance_loss_clip": 1.03550231,
      "balance_loss_mlp": 1.01335621,
      "epoch": 0.8218848639711408,
      "flos": 16764035368320.0,
      "grad_norm": 1.6693339922808688,
      "language_loss": 0.78677434,
      "learning_rate": 3.235680111625161e-07,
      "loss": 0.80794787,
      "num_input_tokens_seen": 294863740,
      "step": 13670,
      "time_per_iteration": 2.467154026031494
    },
    {
      "auxiliary_loss_clip": 0.01097959,
      "auxiliary_loss_mlp": 0.01032516,
      "balance_loss_clip": 1.03723431,
      "balance_loss_mlp": 1.01961184,
      "epoch": 0.8219449872238088,
      "flos": 25994801007360.0,
      "grad_norm": 1.802106749376981,
      "language_loss": 0.7481057,
      "learning_rate": 3.2335565454736123e-07,
      "loss": 0.76941049,
      "num_input_tokens_seen": 294882815,
      "step": 13671,
      "time_per_iteration": 2.5314815044403076
    },
    {
      "auxiliary_loss_clip": 0.0109922,
      "auxiliary_loss_mlp": 0.0103018,
      "balance_loss_clip": 1.03662634,
      "balance_loss_mlp": 1.01744294,
      "epoch": 0.8220051104764767,
      "flos": 20778094091520.0,
      "grad_norm": 1.7123744114769295,
      "language_loss": 0.7639032,
      "learning_rate": 3.23143361510728e-07,
      "loss": 0.7851972,
      "num_input_tokens_seen": 294901985,
      "step": 13672,
      "time_per_iteration": 3.9099624156951904
    },
    {
      "auxiliary_loss_clip": 0.01054271,
      "auxiliary_loss_mlp": 0.01032737,
      "balance_loss_clip": 1.03330731,
      "balance_loss_mlp": 1.01970768,
      "epoch": 0.8220652337291448,
      "flos": 14574849160320.0,
      "grad_norm": 2.810477976613284,
      "language_loss": 0.74700928,
      "learning_rate": 3.2293113206066733e-07,
      "loss": 0.76787937,
      "num_input_tokens_seen": 294919705,
      "step": 13673,
      "time_per_iteration": 2.548940658569336
    },
    {
      "auxiliary_loss_clip": 0.01084029,
      "auxiliary_loss_mlp": 0.01033828,
      "balance_loss_clip": 1.03578603,
      "balance_loss_mlp": 1.02157962,
      "epoch": 0.8221253569818128,
      "flos": 23805866194560.0,
      "grad_norm": 1.6044431154163046,
      "language_loss": 0.79409277,
      "learning_rate": 3.227189662052254e-07,
      "loss": 0.81527132,
      "num_input_tokens_seen": 294939900,
      "step": 13674,
      "time_per_iteration": 3.907717227935791
    },
    {
      "auxiliary_loss_clip": 0.01079418,
      "auxiliary_loss_mlp": 0.01034711,
      "balance_loss_clip": 1.03271472,
      "balance_loss_mlp": 1.0226891,
      "epoch": 0.8221854802344807,
      "flos": 21288241002240.0,
      "grad_norm": 1.8118395529851583,
      "language_loss": 0.70332783,
      "learning_rate": 3.225068639524484e-07,
      "loss": 0.72446913,
      "num_input_tokens_seen": 294959110,
      "step": 13675,
      "time_per_iteration": 2.4971230030059814
    },
    {
      "auxiliary_loss_clip": 0.0108582,
      "auxiliary_loss_mlp": 0.01036773,
      "balance_loss_clip": 1.03312349,
      "balance_loss_mlp": 1.02460146,
      "epoch": 0.8222456034871487,
      "flos": 20956785275520.0,
      "grad_norm": 1.5667996081462106,
      "language_loss": 0.74571884,
      "learning_rate": 3.2229482531037965e-07,
      "loss": 0.76694477,
      "num_input_tokens_seen": 294978660,
      "step": 13676,
      "time_per_iteration": 2.4892139434814453
    },
    {
      "auxiliary_loss_clip": 0.01076757,
      "auxiliary_loss_mlp": 0.0103037,
      "balance_loss_clip": 1.03434634,
      "balance_loss_mlp": 1.01877689,
      "epoch": 0.8223057267398166,
      "flos": 21397517153280.0,
      "grad_norm": 1.7154149904632143,
      "language_loss": 0.80215526,
      "learning_rate": 3.2208285028705893e-07,
      "loss": 0.82322657,
      "num_input_tokens_seen": 294998075,
      "step": 13677,
      "time_per_iteration": 4.0228049755096436
    },
    {
      "auxiliary_loss_clip": 0.01089259,
      "auxiliary_loss_mlp": 0.01034711,
      "balance_loss_clip": 1.03358412,
      "balance_loss_mlp": 1.02207518,
      "epoch": 0.8223658499924846,
      "flos": 15268212368640.0,
      "grad_norm": 1.7338524056658107,
      "language_loss": 0.69760221,
      "learning_rate": 3.218709388905245e-07,
      "loss": 0.71884191,
      "num_input_tokens_seen": 295015950,
      "step": 13678,
      "time_per_iteration": 2.4472744464874268
    },
    {
      "auxiliary_loss_clip": 0.01102003,
      "auxiliary_loss_mlp": 0.01033352,
      "balance_loss_clip": 1.0340749,
      "balance_loss_mlp": 1.02105546,
      "epoch": 0.8224259732451525,
      "flos": 31249537447680.0,
      "grad_norm": 1.5677654115587853,
      "language_loss": 0.71698064,
      "learning_rate": 3.216590911288133e-07,
      "loss": 0.73833418,
      "num_input_tokens_seen": 295036800,
      "step": 13679,
      "time_per_iteration": 2.5593512058258057
    },
    {
      "auxiliary_loss_clip": 0.01074924,
      "auxiliary_loss_mlp": 0.01031752,
      "balance_loss_clip": 1.03245723,
      "balance_loss_mlp": 1.01858592,
      "epoch": 0.8224860964978206,
      "flos": 21574628138880.0,
      "grad_norm": 1.9068074787432832,
      "language_loss": 0.70029783,
      "learning_rate": 3.214473070099564e-07,
      "loss": 0.72136462,
      "num_input_tokens_seen": 295055300,
      "step": 13680,
      "time_per_iteration": 2.492856502532959
    },
    {
      "auxiliary_loss_clip": 0.01072729,
      "auxiliary_loss_mlp": 0.01033304,
      "balance_loss_clip": 1.03610277,
      "balance_loss_mlp": 1.02216935,
      "epoch": 0.8225462197504885,
      "flos": 25483217552640.0,
      "grad_norm": 2.187967599563728,
      "language_loss": 0.59917259,
      "learning_rate": 3.21235586541986e-07,
      "loss": 0.62023288,
      "num_input_tokens_seen": 295076420,
      "step": 13681,
      "time_per_iteration": 2.6170737743377686
    },
    {
      "auxiliary_loss_clip": 0.01080055,
      "auxiliary_loss_mlp": 0.01037126,
      "balance_loss_clip": 1.0335288,
      "balance_loss_mlp": 1.02433515,
      "epoch": 0.8226063430031565,
      "flos": 39385458587520.0,
      "grad_norm": 1.58199692836229,
      "language_loss": 0.697366,
      "learning_rate": 3.2102392973293047e-07,
      "loss": 0.71853781,
      "num_input_tokens_seen": 295100540,
      "step": 13682,
      "time_per_iteration": 4.143878936767578
    },
    {
      "auxiliary_loss_clip": 0.01104473,
      "auxiliary_loss_mlp": 0.01031683,
      "balance_loss_clip": 1.03428376,
      "balance_loss_mlp": 1.01860011,
      "epoch": 0.8226664662558244,
      "flos": 22815269942400.0,
      "grad_norm": 1.9859477843324669,
      "language_loss": 0.78903198,
      "learning_rate": 3.20812336590816e-07,
      "loss": 0.81039357,
      "num_input_tokens_seen": 295120180,
      "step": 13683,
      "time_per_iteration": 2.4602532386779785
    },
    {
      "auxiliary_loss_clip": 0.01099442,
      "auxiliary_loss_mlp": 0.01030616,
      "balance_loss_clip": 1.03440166,
      "balance_loss_mlp": 1.01994634,
      "epoch": 0.8227265895084924,
      "flos": 25665607837440.0,
      "grad_norm": 1.9244657083526502,
      "language_loss": 0.86636102,
      "learning_rate": 3.206008071236661e-07,
      "loss": 0.88766158,
      "num_input_tokens_seen": 295138530,
      "step": 13684,
      "time_per_iteration": 2.485523223876953
    },
    {
      "auxiliary_loss_clip": 0.01099585,
      "auxiliary_loss_mlp": 0.0102894,
      "balance_loss_clip": 1.0339402,
      "balance_loss_mlp": 1.0176146,
      "epoch": 0.8227867127611603,
      "flos": 26179274280960.0,
      "grad_norm": 1.4509080662168439,
      "language_loss": 0.79602337,
      "learning_rate": 3.2038934133950157e-07,
      "loss": 0.8173086,
      "num_input_tokens_seen": 295160260,
      "step": 13685,
      "time_per_iteration": 2.4974024295806885
    },
    {
      "auxiliary_loss_clip": 0.01066236,
      "auxiliary_loss_mlp": 0.01029583,
      "balance_loss_clip": 1.03304815,
      "balance_loss_mlp": 1.01768613,
      "epoch": 0.8228468360138284,
      "flos": 22018053536640.0,
      "grad_norm": 2.3427416594496715,
      "language_loss": 0.69118285,
      "learning_rate": 3.2017793924634194e-07,
      "loss": 0.71214104,
      "num_input_tokens_seen": 295177055,
      "step": 13686,
      "time_per_iteration": 2.5387625694274902
    },
    {
      "auxiliary_loss_clip": 0.01070681,
      "auxiliary_loss_mlp": 0.01029449,
      "balance_loss_clip": 1.03311229,
      "balance_loss_mlp": 1.0176115,
      "epoch": 0.8229069592664963,
      "flos": 14903359971840.0,
      "grad_norm": 1.868689744868676,
      "language_loss": 0.7795707,
      "learning_rate": 3.1996660085220263e-07,
      "loss": 0.80057198,
      "num_input_tokens_seen": 295193870,
      "step": 13687,
      "time_per_iteration": 2.514427900314331
    },
    {
      "auxiliary_loss_clip": 0.01092386,
      "auxiliary_loss_mlp": 0.01029468,
      "balance_loss_clip": 1.03418863,
      "balance_loss_mlp": 1.01702273,
      "epoch": 0.8229670825191643,
      "flos": 15669478177920.0,
      "grad_norm": 1.9423968478733573,
      "language_loss": 0.72330236,
      "learning_rate": 3.1975532616509825e-07,
      "loss": 0.7445209,
      "num_input_tokens_seen": 295211040,
      "step": 13688,
      "time_per_iteration": 2.482572317123413
    },
    {
      "auxiliary_loss_clip": 0.01103779,
      "auxiliary_loss_mlp": 0.00782343,
      "balance_loss_clip": 1.03535485,
      "balance_loss_mlp": 1.00832474,
      "epoch": 0.8230272057718323,
      "flos": 23183498217600.0,
      "grad_norm": 1.609548191771626,
      "language_loss": 0.73445487,
      "learning_rate": 3.1954411519304025e-07,
      "loss": 0.75331616,
      "num_input_tokens_seen": 295231300,
      "step": 13689,
      "time_per_iteration": 2.4818973541259766
    },
    {
      "auxiliary_loss_clip": 0.01094857,
      "auxiliary_loss_mlp": 0.01031768,
      "balance_loss_clip": 1.03474307,
      "balance_loss_mlp": 1.01971042,
      "epoch": 0.8230873290245002,
      "flos": 21032413361280.0,
      "grad_norm": 3.100656229255575,
      "language_loss": 0.69278246,
      "learning_rate": 3.1933296794403887e-07,
      "loss": 0.71404874,
      "num_input_tokens_seen": 295251045,
      "step": 13690,
      "time_per_iteration": 2.4873337745666504
    },
    {
      "auxiliary_loss_clip": 0.01058357,
      "auxiliary_loss_mlp": 0.01037232,
      "balance_loss_clip": 1.03189528,
      "balance_loss_mlp": 1.02351689,
      "epoch": 0.8231474522771682,
      "flos": 21250139650560.0,
      "grad_norm": 8.077849253777021,
      "language_loss": 0.8530916,
      "learning_rate": 3.191218844260988e-07,
      "loss": 0.87404752,
      "num_input_tokens_seen": 295270225,
      "step": 13691,
      "time_per_iteration": 2.5592188835144043
    },
    {
      "auxiliary_loss_clip": 0.01096161,
      "auxiliary_loss_mlp": 0.01032788,
      "balance_loss_clip": 1.03564537,
      "balance_loss_mlp": 1.02096891,
      "epoch": 0.8232075755298361,
      "flos": 23842028211840.0,
      "grad_norm": 1.8553114097976604,
      "language_loss": 0.7695936,
      "learning_rate": 3.189108646472252e-07,
      "loss": 0.79088306,
      "num_input_tokens_seen": 295288950,
      "step": 13692,
      "time_per_iteration": 2.50541090965271
    },
    {
      "auxiliary_loss_clip": 0.01091517,
      "auxiliary_loss_mlp": 0.01025823,
      "balance_loss_clip": 1.03505886,
      "balance_loss_mlp": 1.01414609,
      "epoch": 0.8232676987825042,
      "flos": 21653955325440.0,
      "grad_norm": 1.5158864203035978,
      "language_loss": 0.7168805,
      "learning_rate": 3.186999086154205e-07,
      "loss": 0.73805386,
      "num_input_tokens_seen": 295309405,
      "step": 13693,
      "time_per_iteration": 2.4617600440979004
    },
    {
      "auxiliary_loss_clip": 0.01066569,
      "auxiliary_loss_mlp": 0.01033465,
      "balance_loss_clip": 1.03268075,
      "balance_loss_mlp": 1.02244401,
      "epoch": 0.8233278220351721,
      "flos": 26322701287680.0,
      "grad_norm": 1.3225183433813361,
      "language_loss": 0.83800626,
      "learning_rate": 3.1848901633868355e-07,
      "loss": 0.85900664,
      "num_input_tokens_seen": 295331115,
      "step": 13694,
      "time_per_iteration": 2.5937135219573975
    },
    {
      "auxiliary_loss_clip": 0.01041593,
      "auxiliary_loss_mlp": 0.01027318,
      "balance_loss_clip": 1.03370857,
      "balance_loss_mlp": 1.01516485,
      "epoch": 0.8233879452878401,
      "flos": 21725812483200.0,
      "grad_norm": 1.7202923222517448,
      "language_loss": 0.77389884,
      "learning_rate": 3.182781878250118e-07,
      "loss": 0.79458797,
      "num_input_tokens_seen": 295350495,
      "step": 13695,
      "time_per_iteration": 2.612786293029785
    },
    {
      "auxiliary_loss_clip": 0.01082968,
      "auxiliary_loss_mlp": 0.01032596,
      "balance_loss_clip": 1.03539205,
      "balance_loss_mlp": 1.02094924,
      "epoch": 0.823448068540508,
      "flos": 20557746109440.0,
      "grad_norm": 2.3466821973900442,
      "language_loss": 0.8065179,
      "learning_rate": 3.1806742308239985e-07,
      "loss": 0.82767355,
      "num_input_tokens_seen": 295368225,
      "step": 13696,
      "time_per_iteration": 2.5230674743652344
    },
    {
      "auxiliary_loss_clip": 0.01021469,
      "auxiliary_loss_mlp": 0.00999699,
      "balance_loss_clip": 1.00818098,
      "balance_loss_mlp": 0.99845356,
      "epoch": 0.823508191793176,
      "flos": 67273688194560.0,
      "grad_norm": 0.7374254051037429,
      "language_loss": 0.63926387,
      "learning_rate": 3.178567221188393e-07,
      "loss": 0.65947556,
      "num_input_tokens_seen": 295430035,
      "step": 13697,
      "time_per_iteration": 3.153226375579834
    },
    {
      "auxiliary_loss_clip": 0.01066504,
      "auxiliary_loss_mlp": 0.01026182,
      "balance_loss_clip": 1.03372622,
      "balance_loss_mlp": 1.01573324,
      "epoch": 0.8235683150458439,
      "flos": 17928402641280.0,
      "grad_norm": 1.5927256151393094,
      "language_loss": 0.72704273,
      "learning_rate": 3.1764608494232037e-07,
      "loss": 0.74796963,
      "num_input_tokens_seen": 295447765,
      "step": 13698,
      "time_per_iteration": 2.523817777633667
    },
    {
      "auxiliary_loss_clip": 0.01059584,
      "auxiliary_loss_mlp": 0.01040979,
      "balance_loss_clip": 1.03027296,
      "balance_loss_mlp": 1.0258038,
      "epoch": 0.823628438298512,
      "flos": 18916089891840.0,
      "grad_norm": 2.093301578173963,
      "language_loss": 0.71874332,
      "learning_rate": 3.174355115608305e-07,
      "loss": 0.7397489,
      "num_input_tokens_seen": 295464810,
      "step": 13699,
      "time_per_iteration": 2.523703098297119
    },
    {
      "auxiliary_loss_clip": 0.01080444,
      "auxiliary_loss_mlp": 0.01028774,
      "balance_loss_clip": 1.03302503,
      "balance_loss_mlp": 1.01710892,
      "epoch": 0.8236885615511799,
      "flos": 18696460181760.0,
      "grad_norm": 1.978993727333208,
      "language_loss": 0.81838828,
      "learning_rate": 3.1722500198235526e-07,
      "loss": 0.8394804,
      "num_input_tokens_seen": 295482605,
      "step": 13700,
      "time_per_iteration": 2.499953031539917
    },
    {
      "auxiliary_loss_clip": 0.01082097,
      "auxiliary_loss_mlp": 0.01033589,
      "balance_loss_clip": 1.03448617,
      "balance_loss_mlp": 1.02209175,
      "epoch": 0.8237486848038479,
      "flos": 23695009845120.0,
      "grad_norm": 1.6378758960861395,
      "language_loss": 0.72911727,
      "learning_rate": 3.170145562148763e-07,
      "loss": 0.75027412,
      "num_input_tokens_seen": 295503780,
      "step": 13701,
      "time_per_iteration": 2.5373096466064453
    },
    {
      "auxiliary_loss_clip": 0.01092889,
      "auxiliary_loss_mlp": 0.01033914,
      "balance_loss_clip": 1.0329392,
      "balance_loss_mlp": 1.02123058,
      "epoch": 0.8238088080565159,
      "flos": 23441301106560.0,
      "grad_norm": 1.8900752714771187,
      "language_loss": 0.68994635,
      "learning_rate": 3.1680417426637384e-07,
      "loss": 0.71121436,
      "num_input_tokens_seen": 295522035,
      "step": 13702,
      "time_per_iteration": 2.5409021377563477
    },
    {
      "auxiliary_loss_clip": 0.01056981,
      "auxiliary_loss_mlp": 0.01028465,
      "balance_loss_clip": 1.03499293,
      "balance_loss_mlp": 1.01635909,
      "epoch": 0.8238689313091838,
      "flos": 22746537267840.0,
      "grad_norm": 1.615780496698445,
      "language_loss": 0.7487765,
      "learning_rate": 3.1659385614482603e-07,
      "loss": 0.76963091,
      "num_input_tokens_seen": 295541190,
      "step": 13703,
      "time_per_iteration": 2.5793471336364746
    },
    {
      "auxiliary_loss_clip": 0.01108196,
      "auxiliary_loss_mlp": 0.01038802,
      "balance_loss_clip": 1.03539348,
      "balance_loss_mlp": 1.02518249,
      "epoch": 0.8239290545618518,
      "flos": 25630092264960.0,
      "grad_norm": 1.7831004340503915,
      "language_loss": 0.69609576,
      "learning_rate": 3.1638360185820755e-07,
      "loss": 0.71756566,
      "num_input_tokens_seen": 295558860,
      "step": 13704,
      "time_per_iteration": 2.5044496059417725
    },
    {
      "auxiliary_loss_clip": 0.01100895,
      "auxiliary_loss_mlp": 0.01029771,
      "balance_loss_clip": 1.03351855,
      "balance_loss_mlp": 1.01824343,
      "epoch": 0.8239891778145197,
      "flos": 26026473824640.0,
      "grad_norm": 1.9697894718771298,
      "language_loss": 0.63909256,
      "learning_rate": 3.161734114144916e-07,
      "loss": 0.6603992,
      "num_input_tokens_seen": 295578155,
      "step": 13705,
      "time_per_iteration": 2.4783847332000732
    },
    {
      "auxiliary_loss_clip": 0.01106058,
      "auxiliary_loss_mlp": 0.01028255,
      "balance_loss_clip": 1.0354917,
      "balance_loss_mlp": 1.01595867,
      "epoch": 0.8240493010671878,
      "flos": 21833257040640.0,
      "grad_norm": 1.6315001131099398,
      "language_loss": 0.69382554,
      "learning_rate": 3.1596328482164915e-07,
      "loss": 0.71516871,
      "num_input_tokens_seen": 295599170,
      "step": 13706,
      "time_per_iteration": 2.4862122535705566
    },
    {
      "auxiliary_loss_clip": 0.0107947,
      "auxiliary_loss_mlp": 0.01036227,
      "balance_loss_clip": 1.03523088,
      "balance_loss_mlp": 1.0229888,
      "epoch": 0.8241094243198557,
      "flos": 18551919853440.0,
      "grad_norm": 1.8288316897918095,
      "language_loss": 0.69483256,
      "learning_rate": 3.157532220876475e-07,
      "loss": 0.71598953,
      "num_input_tokens_seen": 295617465,
      "step": 13707,
      "time_per_iteration": 2.483332872390747
    },
    {
      "auxiliary_loss_clip": 0.01068883,
      "auxiliary_loss_mlp": 0.01039615,
      "balance_loss_clip": 1.0328238,
      "balance_loss_mlp": 1.02593517,
      "epoch": 0.8241695475725237,
      "flos": 25447163276160.0,
      "grad_norm": 1.7328909880435355,
      "language_loss": 0.78894472,
      "learning_rate": 3.1554322322045226e-07,
      "loss": 0.81002969,
      "num_input_tokens_seen": 295634960,
      "step": 13708,
      "time_per_iteration": 2.5995748043060303
    },
    {
      "auxiliary_loss_clip": 0.01088136,
      "auxiliary_loss_mlp": 0.01031283,
      "balance_loss_clip": 1.0324707,
      "balance_loss_mlp": 1.01811671,
      "epoch": 0.8242296708251916,
      "flos": 18989670902400.0,
      "grad_norm": 2.349072748548048,
      "language_loss": 0.68543988,
      "learning_rate": 3.1533328822802664e-07,
      "loss": 0.70663404,
      "num_input_tokens_seen": 295652725,
      "step": 13709,
      "time_per_iteration": 2.4535481929779053
    },
    {
      "auxiliary_loss_clip": 0.0106199,
      "auxiliary_loss_mlp": 0.01030631,
      "balance_loss_clip": 1.03290117,
      "balance_loss_mlp": 1.019032,
      "epoch": 0.8242897940778596,
      "flos": 22600883617920.0,
      "grad_norm": 1.7986063604465334,
      "language_loss": 0.82310599,
      "learning_rate": 3.151234171183319e-07,
      "loss": 0.84403223,
      "num_input_tokens_seen": 295671195,
      "step": 13710,
      "time_per_iteration": 2.5865705013275146
    },
    {
      "auxiliary_loss_clip": 0.01092754,
      "auxiliary_loss_mlp": 0.01031999,
      "balance_loss_clip": 1.03430986,
      "balance_loss_mlp": 1.01969016,
      "epoch": 0.8243499173305275,
      "flos": 21468153248640.0,
      "grad_norm": 1.8642906690199172,
      "language_loss": 0.78436685,
      "learning_rate": 3.149136098993257e-07,
      "loss": 0.80561441,
      "num_input_tokens_seen": 295689130,
      "step": 13711,
      "time_per_iteration": 3.9194467067718506
    },
    {
      "auxiliary_loss_clip": 0.0107078,
      "auxiliary_loss_mlp": 0.01030353,
      "balance_loss_clip": 1.0339545,
      "balance_loss_mlp": 1.01800323,
      "epoch": 0.8244100405831956,
      "flos": 20010359773440.0,
      "grad_norm": 1.7782303627751168,
      "language_loss": 0.65944815,
      "learning_rate": 3.1470386657896473e-07,
      "loss": 0.6804595,
      "num_input_tokens_seen": 295706385,
      "step": 13712,
      "time_per_iteration": 2.5072569847106934
    },
    {
      "auxiliary_loss_clip": 0.01094069,
      "auxiliary_loss_mlp": 0.01029002,
      "balance_loss_clip": 1.03536391,
      "balance_loss_mlp": 1.0175277,
      "epoch": 0.8244701638358635,
      "flos": 26430684549120.0,
      "grad_norm": 1.8237788780923978,
      "language_loss": 0.74183524,
      "learning_rate": 3.14494187165202e-07,
      "loss": 0.76306593,
      "num_input_tokens_seen": 295727925,
      "step": 13713,
      "time_per_iteration": 3.979630708694458
    },
    {
      "auxiliary_loss_clip": 0.01091726,
      "auxiliary_loss_mlp": 0.01029317,
      "balance_loss_clip": 1.03298283,
      "balance_loss_mlp": 1.01747346,
      "epoch": 0.8245302870885315,
      "flos": 17640004343040.0,
      "grad_norm": 2.1918878791904195,
      "language_loss": 0.81352758,
      "learning_rate": 3.1428457166598833e-07,
      "loss": 0.83473796,
      "num_input_tokens_seen": 295744420,
      "step": 13714,
      "time_per_iteration": 2.434802770614624
    },
    {
      "auxiliary_loss_clip": 0.01095187,
      "auxiliary_loss_mlp": 0.01034567,
      "balance_loss_clip": 1.03786731,
      "balance_loss_mlp": 1.02196693,
      "epoch": 0.8245904103411995,
      "flos": 26209510554240.0,
      "grad_norm": 1.7143449370553712,
      "language_loss": 0.66366428,
      "learning_rate": 3.1407502008927235e-07,
      "loss": 0.6849618,
      "num_input_tokens_seen": 295765105,
      "step": 13715,
      "time_per_iteration": 3.9443399906158447
    },
    {
      "auxiliary_loss_clip": 0.01074572,
      "auxiliary_loss_mlp": 0.01030018,
      "balance_loss_clip": 1.03460586,
      "balance_loss_mlp": 1.01728106,
      "epoch": 0.8246505335938674,
      "flos": 24205084928640.0,
      "grad_norm": 2.0134213377978587,
      "language_loss": 0.74945915,
      "learning_rate": 3.1386553244300086e-07,
      "loss": 0.77050507,
      "num_input_tokens_seen": 295784200,
      "step": 13716,
      "time_per_iteration": 2.568373203277588
    },
    {
      "auxiliary_loss_clip": 0.01001942,
      "auxiliary_loss_mlp": 0.01002397,
      "balance_loss_clip": 1.01367021,
      "balance_loss_mlp": 1.00121677,
      "epoch": 0.8247106568465354,
      "flos": 67092195749760.0,
      "grad_norm": 0.7202419950721755,
      "language_loss": 0.58977371,
      "learning_rate": 3.136561087351175e-07,
      "loss": 0.60981709,
      "num_input_tokens_seen": 295846555,
      "step": 13717,
      "time_per_iteration": 3.259536027908325
    },
    {
      "auxiliary_loss_clip": 0.01090692,
      "auxiliary_loss_mlp": 0.00783384,
      "balance_loss_clip": 1.03577244,
      "balance_loss_mlp": 1.00851965,
      "epoch": 0.8247707800992033,
      "flos": 12568232805120.0,
      "grad_norm": 2.013133513603316,
      "language_loss": 0.79507363,
      "learning_rate": 3.1344674897356373e-07,
      "loss": 0.8138144,
      "num_input_tokens_seen": 295863425,
      "step": 13718,
      "time_per_iteration": 2.473606586456299
    },
    {
      "auxiliary_loss_clip": 0.01089487,
      "auxiliary_loss_mlp": 0.0102947,
      "balance_loss_clip": 1.03392982,
      "balance_loss_mlp": 1.01825237,
      "epoch": 0.8248309033518714,
      "flos": 15923617879680.0,
      "grad_norm": 1.648434828225315,
      "language_loss": 0.68957287,
      "learning_rate": 3.132374531662778e-07,
      "loss": 0.71076238,
      "num_input_tokens_seen": 295880925,
      "step": 13719,
      "time_per_iteration": 2.4307472705841064
    },
    {
      "auxiliary_loss_clip": 0.01075414,
      "auxiliary_loss_mlp": 0.01034017,
      "balance_loss_clip": 1.03289819,
      "balance_loss_mlp": 1.01958036,
      "epoch": 0.8248910266045393,
      "flos": 17564735393280.0,
      "grad_norm": 2.3958643974517235,
      "language_loss": 0.69568467,
      "learning_rate": 3.13028221321197e-07,
      "loss": 0.71677899,
      "num_input_tokens_seen": 295898205,
      "step": 13720,
      "time_per_iteration": 2.468945026397705
    },
    {
      "auxiliary_loss_clip": 0.01037839,
      "auxiliary_loss_mlp": 0.01028437,
      "balance_loss_clip": 1.03569627,
      "balance_loss_mlp": 1.01640284,
      "epoch": 0.8249511498572073,
      "flos": 28619655275520.0,
      "grad_norm": 1.6743324292830122,
      "language_loss": 0.75658166,
      "learning_rate": 3.1281905344625467e-07,
      "loss": 0.77724445,
      "num_input_tokens_seen": 295918130,
      "step": 13721,
      "time_per_iteration": 4.2036614418029785
    },
    {
      "auxiliary_loss_clip": 0.01053507,
      "auxiliary_loss_mlp": 0.01027709,
      "balance_loss_clip": 1.03540945,
      "balance_loss_mlp": 1.01618183,
      "epoch": 0.8250112731098752,
      "flos": 25556583081600.0,
      "grad_norm": 1.8088309522601314,
      "language_loss": 0.77921855,
      "learning_rate": 3.1260994954938305e-07,
      "loss": 0.80003071,
      "num_input_tokens_seen": 295937760,
      "step": 13722,
      "time_per_iteration": 2.8400325775146484
    },
    {
      "auxiliary_loss_clip": 0.01101682,
      "auxiliary_loss_mlp": 0.01034255,
      "balance_loss_clip": 1.03517365,
      "balance_loss_mlp": 1.02239323,
      "epoch": 0.8250713963625432,
      "flos": 27746164339200.0,
      "grad_norm": 1.8696950077764947,
      "language_loss": 0.62869155,
      "learning_rate": 3.1240090963851205e-07,
      "loss": 0.650051,
      "num_input_tokens_seen": 295957585,
      "step": 13723,
      "time_per_iteration": 2.5200655460357666
    },
    {
      "auxiliary_loss_clip": 0.01105754,
      "auxiliary_loss_mlp": 0.01033081,
      "balance_loss_clip": 1.03584373,
      "balance_loss_mlp": 1.02116573,
      "epoch": 0.8251315196152111,
      "flos": 21610610588160.0,
      "grad_norm": 1.4692067555428319,
      "language_loss": 0.74447739,
      "learning_rate": 3.121919337215666e-07,
      "loss": 0.76586574,
      "num_input_tokens_seen": 295977135,
      "step": 13724,
      "time_per_iteration": 2.4444024562835693
    },
    {
      "auxiliary_loss_clip": 0.01071195,
      "auxiliary_loss_mlp": 0.0103419,
      "balance_loss_clip": 1.03405833,
      "balance_loss_mlp": 1.02178621,
      "epoch": 0.8251916428678792,
      "flos": 28579363194240.0,
      "grad_norm": 1.7510220113255721,
      "language_loss": 0.64093155,
      "learning_rate": 3.1198302180647253e-07,
      "loss": 0.6619854,
      "num_input_tokens_seen": 295996265,
      "step": 13725,
      "time_per_iteration": 2.618880033493042
    },
    {
      "auxiliary_loss_clip": 0.01077037,
      "auxiliary_loss_mlp": 0.01031377,
      "balance_loss_clip": 1.03335309,
      "balance_loss_mlp": 1.01893699,
      "epoch": 0.8252517661205471,
      "flos": 23075191733760.0,
      "grad_norm": 1.6585041995631224,
      "language_loss": 0.81898594,
      "learning_rate": 3.1177417390115125e-07,
      "loss": 0.84007013,
      "num_input_tokens_seen": 296014745,
      "step": 13726,
      "time_per_iteration": 2.5210978984832764
    },
    {
      "auxiliary_loss_clip": 0.01085416,
      "auxiliary_loss_mlp": 0.01035761,
      "balance_loss_clip": 1.03054428,
      "balance_loss_mlp": 1.02407885,
      "epoch": 0.8253118893732151,
      "flos": 31759576617600.0,
      "grad_norm": 3.4632511965942445,
      "language_loss": 0.70374268,
      "learning_rate": 3.1156539001352286e-07,
      "loss": 0.72495449,
      "num_input_tokens_seen": 296036960,
      "step": 13727,
      "time_per_iteration": 2.5621700286865234
    },
    {
      "auxiliary_loss_clip": 0.0109617,
      "auxiliary_loss_mlp": 0.01031064,
      "balance_loss_clip": 1.03564608,
      "balance_loss_mlp": 1.01821899,
      "epoch": 0.8253720126258831,
      "flos": 18296415434880.0,
      "grad_norm": 1.9043862282464106,
      "language_loss": 0.62805068,
      "learning_rate": 3.113566701515036e-07,
      "loss": 0.64932305,
      "num_input_tokens_seen": 296056540,
      "step": 13728,
      "time_per_iteration": 2.477975606918335
    },
    {
      "auxiliary_loss_clip": 0.01088485,
      "auxiliary_loss_mlp": 0.01031677,
      "balance_loss_clip": 1.03745556,
      "balance_loss_mlp": 1.01922524,
      "epoch": 0.825432135878551,
      "flos": 26797332625920.0,
      "grad_norm": 1.722085643201409,
      "language_loss": 0.71140647,
      "learning_rate": 3.111480143230092e-07,
      "loss": 0.73260808,
      "num_input_tokens_seen": 296077950,
      "step": 13729,
      "time_per_iteration": 2.563001871109009
    },
    {
      "auxiliary_loss_clip": 0.01013483,
      "auxiliary_loss_mlp": 0.01001801,
      "balance_loss_clip": 1.00940061,
      "balance_loss_mlp": 1.00049531,
      "epoch": 0.825492259131219,
      "flos": 54219116217600.0,
      "grad_norm": 0.8637344057388846,
      "language_loss": 0.62780476,
      "learning_rate": 3.109394225359514e-07,
      "loss": 0.64795756,
      "num_input_tokens_seen": 296127060,
      "step": 13730,
      "time_per_iteration": 2.9335312843322754
    },
    {
      "auxiliary_loss_clip": 0.01054594,
      "auxiliary_loss_mlp": 0.01035539,
      "balance_loss_clip": 1.03356457,
      "balance_loss_mlp": 1.02323031,
      "epoch": 0.825552382383887,
      "flos": 43756145493120.0,
      "grad_norm": 2.3652174757975866,
      "language_loss": 0.63461334,
      "learning_rate": 3.1073089479823945e-07,
      "loss": 0.65551466,
      "num_input_tokens_seen": 296147775,
      "step": 13731,
      "time_per_iteration": 2.761463165283203
    },
    {
      "auxiliary_loss_clip": 0.0106972,
      "auxiliary_loss_mlp": 0.00787729,
      "balance_loss_clip": 1.03322649,
      "balance_loss_mlp": 1.01110899,
      "epoch": 0.825612505636555,
      "flos": 12602814624000.0,
      "grad_norm": 2.0806822124928184,
      "language_loss": 0.69765472,
      "learning_rate": 3.105224311177812e-07,
      "loss": 0.7162292,
      "num_input_tokens_seen": 296163560,
      "step": 13732,
      "time_per_iteration": 2.4991185665130615
    },
    {
      "auxiliary_loss_clip": 0.01094547,
      "auxiliary_loss_mlp": 0.01039712,
      "balance_loss_clip": 1.03424692,
      "balance_loss_mlp": 1.02729595,
      "epoch": 0.8256726288892229,
      "flos": 17595618111360.0,
      "grad_norm": 2.493802120757609,
      "language_loss": 0.70620871,
      "learning_rate": 3.103140315024817e-07,
      "loss": 0.72755134,
      "num_input_tokens_seen": 296178730,
      "step": 13733,
      "time_per_iteration": 2.4632673263549805
    },
    {
      "auxiliary_loss_clip": 0.01101348,
      "auxiliary_loss_mlp": 0.01029675,
      "balance_loss_clip": 1.03382945,
      "balance_loss_mlp": 1.01761103,
      "epoch": 0.8257327521418909,
      "flos": 23805794367360.0,
      "grad_norm": 1.5210412149745005,
      "language_loss": 0.82556534,
      "learning_rate": 3.1010569596024437e-07,
      "loss": 0.84687561,
      "num_input_tokens_seen": 296200175,
      "step": 13734,
      "time_per_iteration": 2.4904592037200928
    },
    {
      "auxiliary_loss_clip": 0.01072549,
      "auxiliary_loss_mlp": 0.01032731,
      "balance_loss_clip": 1.03144789,
      "balance_loss_mlp": 1.02042294,
      "epoch": 0.8257928753945588,
      "flos": 19281121856640.0,
      "grad_norm": 1.8850320151325466,
      "language_loss": 0.82668889,
      "learning_rate": 3.098974244989676e-07,
      "loss": 0.84774166,
      "num_input_tokens_seen": 296219305,
      "step": 13735,
      "time_per_iteration": 2.5090553760528564
    },
    {
      "auxiliary_loss_clip": 0.0109691,
      "auxiliary_loss_mlp": 0.01029121,
      "balance_loss_clip": 1.03692651,
      "balance_loss_mlp": 1.01808822,
      "epoch": 0.8258529986472268,
      "flos": 18478841633280.0,
      "grad_norm": 1.8286896147125027,
      "language_loss": 0.70855117,
      "learning_rate": 3.096892171265497e-07,
      "loss": 0.72981143,
      "num_input_tokens_seen": 296236945,
      "step": 13736,
      "time_per_iteration": 2.459522247314453
    },
    {
      "auxiliary_loss_clip": 0.01022068,
      "auxiliary_loss_mlp": 0.0100264,
      "balance_loss_clip": 1.00921702,
      "balance_loss_mlp": 1.00138271,
      "epoch": 0.8259131218998947,
      "flos": 62137957512960.0,
      "grad_norm": 0.8498606825105989,
      "language_loss": 0.67961681,
      "learning_rate": 3.0948107385088665e-07,
      "loss": 0.69986391,
      "num_input_tokens_seen": 296294685,
      "step": 13737,
      "time_per_iteration": 3.0983591079711914
    },
    {
      "auxiliary_loss_clip": 0.0108054,
      "auxiliary_loss_mlp": 0.01033971,
      "balance_loss_clip": 1.03396869,
      "balance_loss_mlp": 1.02228808,
      "epoch": 0.8259732451525628,
      "flos": 22159038418560.0,
      "grad_norm": 1.782762228498345,
      "language_loss": 0.69691104,
      "learning_rate": 3.0927299467987e-07,
      "loss": 0.7180562,
      "num_input_tokens_seen": 296314790,
      "step": 13738,
      "time_per_iteration": 2.529181957244873
    },
    {
      "auxiliary_loss_clip": 0.01087054,
      "auxiliary_loss_mlp": 0.01031581,
      "balance_loss_clip": 1.03654051,
      "balance_loss_mlp": 1.01796687,
      "epoch": 0.8260333684052307,
      "flos": 38361645233280.0,
      "grad_norm": 2.4324392874212193,
      "language_loss": 0.62767243,
      "learning_rate": 3.090649796213911e-07,
      "loss": 0.64885885,
      "num_input_tokens_seen": 296335355,
      "step": 13739,
      "time_per_iteration": 2.6497867107391357
    },
    {
      "auxiliary_loss_clip": 0.01012283,
      "auxiliary_loss_mlp": 0.01001245,
      "balance_loss_clip": 1.00888562,
      "balance_loss_mlp": 1.00006521,
      "epoch": 0.8260934916578987,
      "flos": 62185611882240.0,
      "grad_norm": 0.8241126698395173,
      "language_loss": 0.59363127,
      "learning_rate": 3.0885702868333853e-07,
      "loss": 0.61376655,
      "num_input_tokens_seen": 296399885,
      "step": 13740,
      "time_per_iteration": 3.1825897693634033
    },
    {
      "auxiliary_loss_clip": 0.01108456,
      "auxiliary_loss_mlp": 0.01033387,
      "balance_loss_clip": 1.03647816,
      "balance_loss_mlp": 1.02022004,
      "epoch": 0.8261536149105667,
      "flos": 22565475786240.0,
      "grad_norm": 1.8603904323101035,
      "language_loss": 0.75504613,
      "learning_rate": 3.086491418735959e-07,
      "loss": 0.77646458,
      "num_input_tokens_seen": 296417660,
      "step": 13741,
      "time_per_iteration": 2.4529848098754883
    },
    {
      "auxiliary_loss_clip": 0.01091534,
      "auxiliary_loss_mlp": 0.01032995,
      "balance_loss_clip": 1.0343591,
      "balance_loss_mlp": 1.02131224,
      "epoch": 0.8262137381632346,
      "flos": 32525479342080.0,
      "grad_norm": 1.8763066507107642,
      "language_loss": 0.62486029,
      "learning_rate": 3.0844131920004726e-07,
      "loss": 0.64610553,
      "num_input_tokens_seen": 296438255,
      "step": 13742,
      "time_per_iteration": 2.620152711868286
    },
    {
      "auxiliary_loss_clip": 0.01061697,
      "auxiliary_loss_mlp": 0.01036615,
      "balance_loss_clip": 1.03373218,
      "balance_loss_mlp": 1.02297115,
      "epoch": 0.8262738614159026,
      "flos": 14136451666560.0,
      "grad_norm": 2.772873236102213,
      "language_loss": 0.66005862,
      "learning_rate": 3.0823356067057327e-07,
      "loss": 0.68104172,
      "num_input_tokens_seen": 296454485,
      "step": 13743,
      "time_per_iteration": 2.5487794876098633
    },
    {
      "auxiliary_loss_clip": 0.01083939,
      "auxiliary_loss_mlp": 0.01034054,
      "balance_loss_clip": 1.0339421,
      "balance_loss_mlp": 1.02177501,
      "epoch": 0.8263339846685706,
      "flos": 19825347795840.0,
      "grad_norm": 1.9548101938603908,
      "language_loss": 0.67138809,
      "learning_rate": 3.0802586629305283e-07,
      "loss": 0.69256794,
      "num_input_tokens_seen": 296473740,
      "step": 13744,
      "time_per_iteration": 2.598337411880493
    },
    {
      "auxiliary_loss_clip": 0.01072062,
      "auxiliary_loss_mlp": 0.01028689,
      "balance_loss_clip": 1.03692698,
      "balance_loss_mlp": 1.0175612,
      "epoch": 0.8263941079212386,
      "flos": 22745962650240.0,
      "grad_norm": 1.7775720241861561,
      "language_loss": 0.75306547,
      "learning_rate": 3.078182360753612e-07,
      "loss": 0.77407295,
      "num_input_tokens_seen": 296493355,
      "step": 13745,
      "time_per_iteration": 2.5919029712677
    },
    {
      "auxiliary_loss_clip": 0.01073594,
      "auxiliary_loss_mlp": 0.00784532,
      "balance_loss_clip": 1.03445387,
      "balance_loss_mlp": 1.00914145,
      "epoch": 0.8264542311739065,
      "flos": 20120641505280.0,
      "grad_norm": 1.8691808651503856,
      "language_loss": 0.78841329,
      "learning_rate": 3.076106700253709e-07,
      "loss": 0.80699456,
      "num_input_tokens_seen": 296510520,
      "step": 13746,
      "time_per_iteration": 2.535961627960205
    },
    {
      "auxiliary_loss_clip": 0.01099254,
      "auxiliary_loss_mlp": 0.01036161,
      "balance_loss_clip": 1.03758848,
      "balance_loss_mlp": 1.02335191,
      "epoch": 0.8265143544265745,
      "flos": 16837149502080.0,
      "grad_norm": 1.9835930709111191,
      "language_loss": 0.67544699,
      "learning_rate": 3.0740316815095415e-07,
      "loss": 0.69680113,
      "num_input_tokens_seen": 296528265,
      "step": 13747,
      "time_per_iteration": 2.4539108276367188
    },
    {
      "auxiliary_loss_clip": 0.01090791,
      "auxiliary_loss_mlp": 0.01041672,
      "balance_loss_clip": 1.0341264,
      "balance_loss_mlp": 1.02752149,
      "epoch": 0.8265744776792424,
      "flos": 22018592240640.0,
      "grad_norm": 1.962960738608889,
      "language_loss": 0.75489652,
      "learning_rate": 3.0719573045997835e-07,
      "loss": 0.77622104,
      "num_input_tokens_seen": 296547810,
      "step": 13748,
      "time_per_iteration": 2.5020344257354736
    },
    {
      "auxiliary_loss_clip": 0.01065613,
      "auxiliary_loss_mlp": 0.0103158,
      "balance_loss_clip": 1.03699827,
      "balance_loss_mlp": 1.02113724,
      "epoch": 0.8266346009319104,
      "flos": 19244852098560.0,
      "grad_norm": 1.7292026949672679,
      "language_loss": 0.63622987,
      "learning_rate": 3.069883569603102e-07,
      "loss": 0.65720177,
      "num_input_tokens_seen": 296565940,
      "step": 13749,
      "time_per_iteration": 2.530625820159912
    },
    {
      "auxiliary_loss_clip": 0.01078913,
      "auxiliary_loss_mlp": 0.01026547,
      "balance_loss_clip": 1.03227162,
      "balance_loss_mlp": 1.01492405,
      "epoch": 0.8266947241845783,
      "flos": 24166768095360.0,
      "grad_norm": 1.8517908756232848,
      "language_loss": 0.73478484,
      "learning_rate": 3.067810476598132e-07,
      "loss": 0.75583947,
      "num_input_tokens_seen": 296585090,
      "step": 13750,
      "time_per_iteration": 4.19384503364563
    },
    {
      "auxiliary_loss_clip": 0.01094817,
      "auxiliary_loss_mlp": 0.01036772,
      "balance_loss_clip": 1.03591979,
      "balance_loss_mlp": 1.02435076,
      "epoch": 0.8267548474372464,
      "flos": 21105814803840.0,
      "grad_norm": 1.8908309914416515,
      "language_loss": 0.66124713,
      "learning_rate": 3.065738025663496e-07,
      "loss": 0.68256301,
      "num_input_tokens_seen": 296604950,
      "step": 13751,
      "time_per_iteration": 3.8950278759002686
    },
    {
      "auxiliary_loss_clip": 0.01076547,
      "auxiliary_loss_mlp": 0.01028408,
      "balance_loss_clip": 1.03204966,
      "balance_loss_mlp": 1.01760149,
      "epoch": 0.8268149706899143,
      "flos": 39968288668800.0,
      "grad_norm": 1.434714228303696,
      "language_loss": 0.60291088,
      "learning_rate": 3.0636662168777607e-07,
      "loss": 0.62396038,
      "num_input_tokens_seen": 296627780,
      "step": 13752,
      "time_per_iteration": 2.67452335357666
    },
    {
      "auxiliary_loss_clip": 0.01021134,
      "auxiliary_loss_mlp": 0.01002853,
      "balance_loss_clip": 1.00888729,
      "balance_loss_mlp": 1.00165546,
      "epoch": 0.8268750939425823,
      "flos": 65782423244160.0,
      "grad_norm": 0.7697279688427325,
      "language_loss": 0.57408082,
      "learning_rate": 3.0615950503194986e-07,
      "loss": 0.59432065,
      "num_input_tokens_seen": 296683850,
      "step": 13753,
      "time_per_iteration": 3.139103889465332
    },
    {
      "auxiliary_loss_clip": 0.00993979,
      "auxiliary_loss_mlp": 0.00763476,
      "balance_loss_clip": 1.01328897,
      "balance_loss_mlp": 1.00198162,
      "epoch": 0.8269352171952503,
      "flos": 52981455242880.0,
      "grad_norm": 0.6991419744479438,
      "language_loss": 0.55042869,
      "learning_rate": 3.0595245260672563e-07,
      "loss": 0.56800324,
      "num_input_tokens_seen": 296741420,
      "step": 13754,
      "time_per_iteration": 4.670770645141602
    },
    {
      "auxiliary_loss_clip": 0.01060498,
      "auxiliary_loss_mlp": 0.01036217,
      "balance_loss_clip": 1.03172314,
      "balance_loss_mlp": 1.02561903,
      "epoch": 0.8269953404479182,
      "flos": 23076125487360.0,
      "grad_norm": 1.902709589106428,
      "language_loss": 0.6907562,
      "learning_rate": 3.0574546441995354e-07,
      "loss": 0.71172333,
      "num_input_tokens_seen": 296759620,
      "step": 13755,
      "time_per_iteration": 2.783496141433716
    },
    {
      "auxiliary_loss_clip": 0.01058689,
      "auxiliary_loss_mlp": 0.01028384,
      "balance_loss_clip": 1.03397417,
      "balance_loss_mlp": 1.01745272,
      "epoch": 0.8270554637005862,
      "flos": 14209996763520.0,
      "grad_norm": 1.930453879655492,
      "language_loss": 0.69983971,
      "learning_rate": 3.0553854047948324e-07,
      "loss": 0.7207104,
      "num_input_tokens_seen": 296777275,
      "step": 13756,
      "time_per_iteration": 2.563490867614746
    },
    {
      "auxiliary_loss_clip": 0.01094853,
      "auxiliary_loss_mlp": 0.0103014,
      "balance_loss_clip": 1.03713751,
      "balance_loss_mlp": 1.01839161,
      "epoch": 0.8271155869532542,
      "flos": 21762046327680.0,
      "grad_norm": 1.8496955098919012,
      "language_loss": 0.72195375,
      "learning_rate": 3.053316807931623e-07,
      "loss": 0.7432037,
      "num_input_tokens_seen": 296796655,
      "step": 13757,
      "time_per_iteration": 2.5056746006011963
    },
    {
      "auxiliary_loss_clip": 0.01097083,
      "auxiliary_loss_mlp": 0.01031585,
      "balance_loss_clip": 1.03586292,
      "balance_loss_mlp": 1.01759017,
      "epoch": 0.8271757102059222,
      "flos": 15120475729920.0,
      "grad_norm": 2.408019644756395,
      "language_loss": 0.69047558,
      "learning_rate": 3.0512488536883283e-07,
      "loss": 0.71176219,
      "num_input_tokens_seen": 296813705,
      "step": 13758,
      "time_per_iteration": 2.453334093093872
    },
    {
      "auxiliary_loss_clip": 0.01079008,
      "auxiliary_loss_mlp": 0.01029853,
      "balance_loss_clip": 1.03284168,
      "balance_loss_mlp": 1.01833153,
      "epoch": 0.8272358334585901,
      "flos": 24133730561280.0,
      "grad_norm": 1.5399483567234775,
      "language_loss": 0.69934583,
      "learning_rate": 3.0491815421433775e-07,
      "loss": 0.72043443,
      "num_input_tokens_seen": 296833985,
      "step": 13759,
      "time_per_iteration": 3.9600484371185303
    },
    {
      "auxiliary_loss_clip": 0.01081206,
      "auxiliary_loss_mlp": 0.01030289,
      "balance_loss_clip": 1.03398252,
      "balance_loss_mlp": 1.01810598,
      "epoch": 0.8272959567112581,
      "flos": 18990712396800.0,
      "grad_norm": 1.698754179967429,
      "language_loss": 0.70726383,
      "learning_rate": 3.047114873375161e-07,
      "loss": 0.72837877,
      "num_input_tokens_seen": 296850150,
      "step": 13760,
      "time_per_iteration": 2.4885857105255127
    },
    {
      "auxiliary_loss_clip": 0.01062052,
      "auxiliary_loss_mlp": 0.01033523,
      "balance_loss_clip": 1.03340101,
      "balance_loss_mlp": 1.02166164,
      "epoch": 0.827356079963926,
      "flos": 20631614428800.0,
      "grad_norm": 1.6967136424717588,
      "language_loss": 0.77847427,
      "learning_rate": 3.0450488474620505e-07,
      "loss": 0.79943001,
      "num_input_tokens_seen": 296869585,
      "step": 13761,
      "time_per_iteration": 2.567659854888916
    },
    {
      "auxiliary_loss_clip": 0.01068397,
      "auxiliary_loss_mlp": 0.01030687,
      "balance_loss_clip": 1.03444731,
      "balance_loss_mlp": 1.02010703,
      "epoch": 0.827416203216594,
      "flos": 22416625825920.0,
      "grad_norm": 1.737762983649448,
      "language_loss": 0.7002281,
      "learning_rate": 3.042983464482387e-07,
      "loss": 0.72121894,
      "num_input_tokens_seen": 296887710,
      "step": 13762,
      "time_per_iteration": 2.5412840843200684
    },
    {
      "auxiliary_loss_clip": 0.01049564,
      "auxiliary_loss_mlp": 0.01030379,
      "balance_loss_clip": 1.03199434,
      "balance_loss_mlp": 1.01861298,
      "epoch": 0.827476326469262,
      "flos": 19026192055680.0,
      "grad_norm": 1.7867537035933192,
      "language_loss": 0.69872546,
      "learning_rate": 3.0409187245144853e-07,
      "loss": 0.71952486,
      "num_input_tokens_seen": 296906265,
      "step": 13763,
      "time_per_iteration": 2.599407434463501
    },
    {
      "auxiliary_loss_clip": 0.01013088,
      "auxiliary_loss_mlp": 0.01000778,
      "balance_loss_clip": 1.01533723,
      "balance_loss_mlp": 0.9994961,
      "epoch": 0.82753644972193,
      "flos": 68500575089280.0,
      "grad_norm": 0.8395302397199025,
      "language_loss": 0.65137666,
      "learning_rate": 3.038854627636651e-07,
      "loss": 0.67151541,
      "num_input_tokens_seen": 296971290,
      "step": 13764,
      "time_per_iteration": 3.254702091217041
    },
    {
      "auxiliary_loss_clip": 0.01094302,
      "auxiliary_loss_mlp": 0.01032949,
      "balance_loss_clip": 1.03555942,
      "balance_loss_mlp": 1.02062845,
      "epoch": 0.8275965729745979,
      "flos": 18405404277120.0,
      "grad_norm": 2.1923458216558034,
      "language_loss": 0.7760756,
      "learning_rate": 3.0367911739271423e-07,
      "loss": 0.79734808,
      "num_input_tokens_seen": 296989060,
      "step": 13765,
      "time_per_iteration": 2.4713244438171387
    },
    {
      "auxiliary_loss_clip": 0.01050868,
      "auxiliary_loss_mlp": 0.01029878,
      "balance_loss_clip": 1.03419542,
      "balance_loss_mlp": 1.01722991,
      "epoch": 0.8276566962272659,
      "flos": 28512067063680.0,
      "grad_norm": 1.5064561121929607,
      "language_loss": 0.62473631,
      "learning_rate": 3.034728363464214e-07,
      "loss": 0.64554381,
      "num_input_tokens_seen": 297011300,
      "step": 13766,
      "time_per_iteration": 2.672788143157959
    },
    {
      "auxiliary_loss_clip": 0.01069769,
      "auxiliary_loss_mlp": 0.01030821,
      "balance_loss_clip": 1.0358423,
      "balance_loss_mlp": 1.018543,
      "epoch": 0.8277168194799339,
      "flos": 20230240878720.0,
      "grad_norm": 1.6333018619521407,
      "language_loss": 0.82568395,
      "learning_rate": 3.03266619632609e-07,
      "loss": 0.84668982,
      "num_input_tokens_seen": 297030350,
      "step": 13767,
      "time_per_iteration": 2.557683229446411
    },
    {
      "auxiliary_loss_clip": 0.01078804,
      "auxiliary_loss_mlp": 0.01028231,
      "balance_loss_clip": 1.03945065,
      "balance_loss_mlp": 1.0163641,
      "epoch": 0.8277769427326018,
      "flos": 28476623318400.0,
      "grad_norm": 1.5494839278570682,
      "language_loss": 0.69281024,
      "learning_rate": 3.030604672590964e-07,
      "loss": 0.7138806,
      "num_input_tokens_seen": 297049710,
      "step": 13768,
      "time_per_iteration": 2.5780515670776367
    },
    {
      "auxiliary_loss_clip": 0.01045041,
      "auxiliary_loss_mlp": 0.01034252,
      "balance_loss_clip": 1.03281331,
      "balance_loss_mlp": 1.02141929,
      "epoch": 0.8278370659852698,
      "flos": 27197628768000.0,
      "grad_norm": 1.8816555559976995,
      "language_loss": 0.74724287,
      "learning_rate": 3.028543792337006e-07,
      "loss": 0.76803583,
      "num_input_tokens_seen": 297070510,
      "step": 13769,
      "time_per_iteration": 2.6421194076538086
    },
    {
      "auxiliary_loss_clip": 0.01080093,
      "auxiliary_loss_mlp": 0.01029957,
      "balance_loss_clip": 1.03425241,
      "balance_loss_mlp": 1.01829815,
      "epoch": 0.8278971892379378,
      "flos": 37816126404480.0,
      "grad_norm": 1.7921071783067353,
      "language_loss": 0.74576354,
      "learning_rate": 3.0264835556423675e-07,
      "loss": 0.76686406,
      "num_input_tokens_seen": 297092585,
      "step": 13770,
      "time_per_iteration": 2.6709301471710205
    },
    {
      "auxiliary_loss_clip": 0.01074229,
      "auxiliary_loss_mlp": 0.01032244,
      "balance_loss_clip": 1.03519368,
      "balance_loss_mlp": 1.01975131,
      "epoch": 0.8279573124906058,
      "flos": 22560160573440.0,
      "grad_norm": 1.6819123251030312,
      "language_loss": 0.75783736,
      "learning_rate": 3.0244239625851785e-07,
      "loss": 0.77890205,
      "num_input_tokens_seen": 297110055,
      "step": 13771,
      "time_per_iteration": 2.5439581871032715
    },
    {
      "auxiliary_loss_clip": 0.01103143,
      "auxiliary_loss_mlp": 0.01031609,
      "balance_loss_clip": 1.03453314,
      "balance_loss_mlp": 1.01994419,
      "epoch": 0.8280174357432737,
      "flos": 36064619418240.0,
      "grad_norm": 1.590593537573226,
      "language_loss": 0.72566205,
      "learning_rate": 3.0223650132435284e-07,
      "loss": 0.74700952,
      "num_input_tokens_seen": 297132170,
      "step": 13772,
      "time_per_iteration": 2.582756996154785
    },
    {
      "auxiliary_loss_clip": 0.01081964,
      "auxiliary_loss_mlp": 0.01030873,
      "balance_loss_clip": 1.03480983,
      "balance_loss_mlp": 1.0183382,
      "epoch": 0.8280775589959417,
      "flos": 22961067246720.0,
      "grad_norm": 2.501083071461501,
      "language_loss": 0.74649966,
      "learning_rate": 3.0203067076955035e-07,
      "loss": 0.76762795,
      "num_input_tokens_seen": 297149515,
      "step": 13773,
      "time_per_iteration": 2.504692316055298
    },
    {
      "auxiliary_loss_clip": 0.01058183,
      "auxiliary_loss_mlp": 0.01031804,
      "balance_loss_clip": 1.03691006,
      "balance_loss_mlp": 1.02077711,
      "epoch": 0.8281376822486096,
      "flos": 26063282286720.0,
      "grad_norm": 1.7383643570029466,
      "language_loss": 0.75776052,
      "learning_rate": 3.01824904601915e-07,
      "loss": 0.77866042,
      "num_input_tokens_seen": 297170320,
      "step": 13774,
      "time_per_iteration": 2.6227054595947266
    },
    {
      "auxiliary_loss_clip": 0.0107624,
      "auxiliary_loss_mlp": 0.0078386,
      "balance_loss_clip": 1.03709388,
      "balance_loss_mlp": 1.01075077,
      "epoch": 0.8281978055012776,
      "flos": 20667776446080.0,
      "grad_norm": 1.702503160144713,
      "language_loss": 0.75307691,
      "learning_rate": 3.01619202829249e-07,
      "loss": 0.77167785,
      "num_input_tokens_seen": 297189935,
      "step": 13775,
      "time_per_iteration": 2.5441153049468994
    },
    {
      "auxiliary_loss_clip": 0.01107543,
      "auxiliary_loss_mlp": 0.01032455,
      "balance_loss_clip": 1.03484106,
      "balance_loss_mlp": 1.01955688,
      "epoch": 0.8282579287539455,
      "flos": 29315281040640.0,
      "grad_norm": 4.621910496719665,
      "language_loss": 0.73738182,
      "learning_rate": 3.01413565459353e-07,
      "loss": 0.75878179,
      "num_input_tokens_seen": 297210885,
      "step": 13776,
      "time_per_iteration": 2.5880653858184814
    },
    {
      "auxiliary_loss_clip": 0.01050752,
      "auxiliary_loss_mlp": 0.01032762,
      "balance_loss_clip": 1.02933276,
      "balance_loss_mlp": 1.01912427,
      "epoch": 0.8283180520066136,
      "flos": 15706178899200.0,
      "grad_norm": 1.900685674268679,
      "language_loss": 0.77318406,
      "learning_rate": 3.0120799250002483e-07,
      "loss": 0.79401922,
      "num_input_tokens_seen": 297228500,
      "step": 13777,
      "time_per_iteration": 2.562140703201294
    },
    {
      "auxiliary_loss_clip": 0.01091418,
      "auxiliary_loss_mlp": 0.01028446,
      "balance_loss_clip": 1.03588605,
      "balance_loss_mlp": 1.01757383,
      "epoch": 0.8283781752592815,
      "flos": 24791470456320.0,
      "grad_norm": 1.5145547550286076,
      "language_loss": 0.82535315,
      "learning_rate": 3.010024839590604e-07,
      "loss": 0.84655178,
      "num_input_tokens_seen": 297249470,
      "step": 13778,
      "time_per_iteration": 2.530228853225708
    },
    {
      "auxiliary_loss_clip": 0.01087691,
      "auxiliary_loss_mlp": 0.01024147,
      "balance_loss_clip": 1.03319943,
      "balance_loss_mlp": 1.01250076,
      "epoch": 0.8284382985119495,
      "flos": 18982811404800.0,
      "grad_norm": 1.8060068025394622,
      "language_loss": 0.74942815,
      "learning_rate": 3.0079703984425187e-07,
      "loss": 0.77054656,
      "num_input_tokens_seen": 297265970,
      "step": 13779,
      "time_per_iteration": 2.4607298374176025
    },
    {
      "auxiliary_loss_clip": 0.0100385,
      "auxiliary_loss_mlp": 0.01002813,
      "balance_loss_clip": 1.00879514,
      "balance_loss_mlp": 1.00147188,
      "epoch": 0.8284984217646175,
      "flos": 61034460814080.0,
      "grad_norm": 0.7660120474984011,
      "language_loss": 0.56719732,
      "learning_rate": 3.0059166016338954e-07,
      "loss": 0.58726394,
      "num_input_tokens_seen": 297325525,
      "step": 13780,
      "time_per_iteration": 3.2000057697296143
    },
    {
      "auxiliary_loss_clip": 0.0106311,
      "auxiliary_loss_mlp": 0.01026504,
      "balance_loss_clip": 1.03430843,
      "balance_loss_mlp": 1.01378989,
      "epoch": 0.8285585450172854,
      "flos": 19714635100800.0,
      "grad_norm": 1.5863784598446904,
      "language_loss": 0.79727173,
      "learning_rate": 3.0038634492426205e-07,
      "loss": 0.81816792,
      "num_input_tokens_seen": 297345025,
      "step": 13781,
      "time_per_iteration": 2.595160722732544
    },
    {
      "auxiliary_loss_clip": 0.01062029,
      "auxiliary_loss_mlp": 0.01028864,
      "balance_loss_clip": 1.03728259,
      "balance_loss_mlp": 1.0159297,
      "epoch": 0.8286186682699535,
      "flos": 21688896280320.0,
      "grad_norm": 2.8468546744039984,
      "language_loss": 0.75193179,
      "learning_rate": 3.001810941346543e-07,
      "loss": 0.77284068,
      "num_input_tokens_seen": 297363570,
      "step": 13782,
      "time_per_iteration": 2.610928773880005
    },
    {
      "auxiliary_loss_clip": 0.0109132,
      "auxiliary_loss_mlp": 0.01026811,
      "balance_loss_clip": 1.03297091,
      "balance_loss_mlp": 1.01511705,
      "epoch": 0.8286787915226214,
      "flos": 25775566346880.0,
      "grad_norm": 1.7688811977450976,
      "language_loss": 0.7619791,
      "learning_rate": 2.9997590780234983e-07,
      "loss": 0.78316045,
      "num_input_tokens_seen": 297385385,
      "step": 13783,
      "time_per_iteration": 2.543686866760254
    },
    {
      "auxiliary_loss_clip": 0.01103373,
      "auxiliary_loss_mlp": 0.01026406,
      "balance_loss_clip": 1.03487945,
      "balance_loss_mlp": 1.01486063,
      "epoch": 0.8287389147752894,
      "flos": 21288348743040.0,
      "grad_norm": 1.5471168193635902,
      "language_loss": 0.73771727,
      "learning_rate": 2.997707859351304e-07,
      "loss": 0.75901508,
      "num_input_tokens_seen": 297403950,
      "step": 13784,
      "time_per_iteration": 2.5256195068359375
    },
    {
      "auxiliary_loss_clip": 0.01095283,
      "auxiliary_loss_mlp": 0.01031746,
      "balance_loss_clip": 1.03363323,
      "balance_loss_mlp": 1.01906776,
      "epoch": 0.8287990380279573,
      "flos": 33544875323520.0,
      "grad_norm": 1.4846059376341723,
      "language_loss": 0.69877672,
      "learning_rate": 2.99565728540772e-07,
      "loss": 0.720047,
      "num_input_tokens_seen": 297424565,
      "step": 13785,
      "time_per_iteration": 2.585559368133545
    },
    {
      "auxiliary_loss_clip": 0.01080353,
      "auxiliary_loss_mlp": 0.01030006,
      "balance_loss_clip": 1.03636718,
      "balance_loss_mlp": 1.01797187,
      "epoch": 0.8288591612806253,
      "flos": 22966346545920.0,
      "grad_norm": 1.4327602047461956,
      "language_loss": 0.68304807,
      "learning_rate": 2.993607356270516e-07,
      "loss": 0.70415169,
      "num_input_tokens_seen": 297445180,
      "step": 13786,
      "time_per_iteration": 2.564075231552124
    },
    {
      "auxiliary_loss_clip": 0.01064921,
      "auxiliary_loss_mlp": 0.01036734,
      "balance_loss_clip": 1.03533983,
      "balance_loss_mlp": 1.02462244,
      "epoch": 0.8289192845332932,
      "flos": 18588979710720.0,
      "grad_norm": 1.7445751539365408,
      "language_loss": 0.76785195,
      "learning_rate": 2.991558072017426e-07,
      "loss": 0.78886855,
      "num_input_tokens_seen": 297463790,
      "step": 13787,
      "time_per_iteration": 2.6029598712921143
    },
    {
      "auxiliary_loss_clip": 0.01092682,
      "auxiliary_loss_mlp": 0.01035574,
      "balance_loss_clip": 1.03568709,
      "balance_loss_mlp": 1.02399325,
      "epoch": 0.8289794077859612,
      "flos": 15450423085440.0,
      "grad_norm": 1.6825528578457583,
      "language_loss": 0.80367959,
      "learning_rate": 2.989509432726163e-07,
      "loss": 0.82496208,
      "num_input_tokens_seen": 297480100,
      "step": 13788,
      "time_per_iteration": 3.91154146194458
    },
    {
      "auxiliary_loss_clip": 0.01079389,
      "auxiliary_loss_mlp": 0.01031316,
      "balance_loss_clip": 1.03679669,
      "balance_loss_mlp": 1.02001524,
      "epoch": 0.8290395310386292,
      "flos": 28877853214080.0,
      "grad_norm": 1.6036075158245275,
      "language_loss": 0.70807004,
      "learning_rate": 2.9874614384744014e-07,
      "loss": 0.72917706,
      "num_input_tokens_seen": 297499890,
      "step": 13789,
      "time_per_iteration": 4.1907854080200195
    },
    {
      "auxiliary_loss_clip": 0.0107124,
      "auxiliary_loss_mlp": 0.01028522,
      "balance_loss_clip": 1.03079355,
      "balance_loss_mlp": 1.01590967,
      "epoch": 0.8290996542912972,
      "flos": 36576274700160.0,
      "grad_norm": 1.717551870176779,
      "language_loss": 0.68069875,
      "learning_rate": 2.985414089339813e-07,
      "loss": 0.7016964,
      "num_input_tokens_seen": 297521440,
      "step": 13790,
      "time_per_iteration": 2.718311071395874
    },
    {
      "auxiliary_loss_clip": 0.01094901,
      "auxiliary_loss_mlp": 0.01028559,
      "balance_loss_clip": 1.03483725,
      "balance_loss_mlp": 1.01526749,
      "epoch": 0.8291597775439651,
      "flos": 23623009032960.0,
      "grad_norm": 1.8102400257496356,
      "language_loss": 0.77587438,
      "learning_rate": 2.9833673854000265e-07,
      "loss": 0.79710895,
      "num_input_tokens_seen": 297539920,
      "step": 13791,
      "time_per_iteration": 2.5219216346740723
    },
    {
      "auxiliary_loss_clip": 0.01078139,
      "auxiliary_loss_mlp": 0.01028731,
      "balance_loss_clip": 1.03471017,
      "balance_loss_mlp": 1.01570153,
      "epoch": 0.8292199007966331,
      "flos": 21397481239680.0,
      "grad_norm": 1.4760769734780976,
      "language_loss": 0.6996274,
      "learning_rate": 2.981321326732651e-07,
      "loss": 0.72069615,
      "num_input_tokens_seen": 297560000,
      "step": 13792,
      "time_per_iteration": 3.9157023429870605
    },
    {
      "auxiliary_loss_clip": 0.01082569,
      "auxiliary_loss_mlp": 0.01034323,
      "balance_loss_clip": 1.03436422,
      "balance_loss_mlp": 1.02202034,
      "epoch": 0.829280024049301,
      "flos": 28767607395840.0,
      "grad_norm": 2.2196112551783695,
      "language_loss": 0.6511718,
      "learning_rate": 2.9792759134152736e-07,
      "loss": 0.67234069,
      "num_input_tokens_seen": 297579300,
      "step": 13793,
      "time_per_iteration": 2.5637457370758057
    },
    {
      "auxiliary_loss_clip": 0.01049124,
      "auxiliary_loss_mlp": 0.01035507,
      "balance_loss_clip": 1.03253913,
      "balance_loss_mlp": 1.02180433,
      "epoch": 0.829340147301969,
      "flos": 19938071652480.0,
      "grad_norm": 1.7392558847816544,
      "language_loss": 0.66164374,
      "learning_rate": 2.977231145525461e-07,
      "loss": 0.68248999,
      "num_input_tokens_seen": 297598095,
      "step": 13794,
      "time_per_iteration": 2.613039016723633
    },
    {
      "auxiliary_loss_clip": 0.01102841,
      "auxiliary_loss_mlp": 0.01033384,
      "balance_loss_clip": 1.03349638,
      "balance_loss_mlp": 1.02089703,
      "epoch": 0.829400270554637,
      "flos": 25228575060480.0,
      "grad_norm": 1.9679588470097593,
      "language_loss": 0.66119969,
      "learning_rate": 2.975187023140757e-07,
      "loss": 0.68256193,
      "num_input_tokens_seen": 297615955,
      "step": 13795,
      "time_per_iteration": 2.4749884605407715
    },
    {
      "auxiliary_loss_clip": 0.01012265,
      "auxiliary_loss_mlp": 0.01042572,
      "balance_loss_clip": 1.03154027,
      "balance_loss_mlp": 1.02895796,
      "epoch": 0.829460393807305,
      "flos": 24463570176000.0,
      "grad_norm": 1.7900305576602134,
      "language_loss": 0.66302365,
      "learning_rate": 2.973143546338661e-07,
      "loss": 0.68357205,
      "num_input_tokens_seen": 297636285,
      "step": 13796,
      "time_per_iteration": 2.789785623550415
    },
    {
      "auxiliary_loss_clip": 0.01055113,
      "auxiliary_loss_mlp": 0.01035967,
      "balance_loss_clip": 1.03075981,
      "balance_loss_mlp": 1.02253842,
      "epoch": 0.829520517059973,
      "flos": 15122486891520.0,
      "grad_norm": 1.7011874035943528,
      "language_loss": 0.71888673,
      "learning_rate": 2.971100715196666e-07,
      "loss": 0.73979759,
      "num_input_tokens_seen": 297653315,
      "step": 13797,
      "time_per_iteration": 2.688265085220337
    },
    {
      "auxiliary_loss_clip": 0.01035813,
      "auxiliary_loss_mlp": 0.01030838,
      "balance_loss_clip": 1.03291166,
      "balance_loss_mlp": 1.01884604,
      "epoch": 0.8295806403126409,
      "flos": 21579979265280.0,
      "grad_norm": 1.7243802918064146,
      "language_loss": 0.72436374,
      "learning_rate": 2.969058529792243e-07,
      "loss": 0.74503028,
      "num_input_tokens_seen": 297673480,
      "step": 13798,
      "time_per_iteration": 4.167632341384888
    },
    {
      "auxiliary_loss_clip": 0.01065658,
      "auxiliary_loss_mlp": 0.01032061,
      "balance_loss_clip": 1.0317421,
      "balance_loss_mlp": 1.01960325,
      "epoch": 0.8296407635653089,
      "flos": 21726566668800.0,
      "grad_norm": 1.5710904312101868,
      "language_loss": 0.75959647,
      "learning_rate": 2.967016990202822e-07,
      "loss": 0.78057361,
      "num_input_tokens_seen": 297693250,
      "step": 13799,
      "time_per_iteration": 2.795722246170044
    },
    {
      "auxiliary_loss_clip": 0.01103542,
      "auxiliary_loss_mlp": 0.01032723,
      "balance_loss_clip": 1.03575134,
      "balance_loss_mlp": 1.0208317,
      "epoch": 0.8297008868179768,
      "flos": 11181147252480.0,
      "grad_norm": 1.7785422132184336,
      "language_loss": 0.67173135,
      "learning_rate": 2.9649760965058245e-07,
      "loss": 0.69309396,
      "num_input_tokens_seen": 297710975,
      "step": 13800,
      "time_per_iteration": 2.444925308227539
    },
    {
      "auxiliary_loss_clip": 0.01068756,
      "auxiliary_loss_mlp": 0.01035508,
      "balance_loss_clip": 1.036057,
      "balance_loss_mlp": 1.02204311,
      "epoch": 0.8297610100706448,
      "flos": 20664041431680.0,
      "grad_norm": 1.6222500305311753,
      "language_loss": 0.74729437,
      "learning_rate": 2.9629358487786515e-07,
      "loss": 0.76833701,
      "num_input_tokens_seen": 297730860,
      "step": 13801,
      "time_per_iteration": 2.6141347885131836
    },
    {
      "auxiliary_loss_clip": 0.01061981,
      "auxiliary_loss_mlp": 0.01028833,
      "balance_loss_clip": 1.03683972,
      "balance_loss_mlp": 1.01760316,
      "epoch": 0.8298211333233128,
      "flos": 20376325491840.0,
      "grad_norm": 1.513138674732876,
      "language_loss": 0.73562396,
      "learning_rate": 2.9608962470986476e-07,
      "loss": 0.75653219,
      "num_input_tokens_seen": 297749765,
      "step": 13802,
      "time_per_iteration": 2.6218268871307373
    },
    {
      "auxiliary_loss_clip": 0.01084699,
      "auxiliary_loss_mlp": 0.0103201,
      "balance_loss_clip": 1.03463674,
      "balance_loss_mlp": 1.02008963,
      "epoch": 0.8298812565759808,
      "flos": 21508696725120.0,
      "grad_norm": 1.579773479724805,
      "language_loss": 0.74645382,
      "learning_rate": 2.9588572915431644e-07,
      "loss": 0.76762092,
      "num_input_tokens_seen": 297770380,
      "step": 13803,
      "time_per_iteration": 2.554824113845825
    },
    {
      "auxiliary_loss_clip": 0.01093958,
      "auxiliary_loss_mlp": 0.01030482,
      "balance_loss_clip": 1.03604555,
      "balance_loss_mlp": 1.01892471,
      "epoch": 0.8299413798286487,
      "flos": 22818681734400.0,
      "grad_norm": 1.6946292556468103,
      "language_loss": 0.79357982,
      "learning_rate": 2.9568189821895215e-07,
      "loss": 0.81482422,
      "num_input_tokens_seen": 297789440,
      "step": 13804,
      "time_per_iteration": 2.4915390014648438
    },
    {
      "auxiliary_loss_clip": 0.01102306,
      "auxiliary_loss_mlp": 0.01029471,
      "balance_loss_clip": 1.03466928,
      "balance_loss_mlp": 1.01827717,
      "epoch": 0.8300015030813167,
      "flos": 29679199683840.0,
      "grad_norm": 1.672390215027454,
      "language_loss": 0.7325207,
      "learning_rate": 2.954781319115016e-07,
      "loss": 0.75383854,
      "num_input_tokens_seen": 297810425,
      "step": 13805,
      "time_per_iteration": 2.5287089347839355
    },
    {
      "auxiliary_loss_clip": 0.01095143,
      "auxiliary_loss_mlp": 0.00781485,
      "balance_loss_clip": 1.03554392,
      "balance_loss_mlp": 1.00623059,
      "epoch": 0.8300616263339846,
      "flos": 19719483436800.0,
      "grad_norm": 2.097453678602502,
      "language_loss": 0.77370495,
      "learning_rate": 2.952744302396906e-07,
      "loss": 0.79247117,
      "num_input_tokens_seen": 297827680,
      "step": 13806,
      "time_per_iteration": 2.4910733699798584
    },
    {
      "auxiliary_loss_clip": 0.01097062,
      "auxiliary_loss_mlp": 0.01034323,
      "balance_loss_clip": 1.03546321,
      "balance_loss_mlp": 1.02112675,
      "epoch": 0.8301217495866526,
      "flos": 19901945548800.0,
      "grad_norm": 1.6402176903222554,
      "language_loss": 0.63401306,
      "learning_rate": 2.950707932112444e-07,
      "loss": 0.6553269,
      "num_input_tokens_seen": 297848005,
      "step": 13807,
      "time_per_iteration": 2.4969959259033203
    },
    {
      "auxiliary_loss_clip": 0.01092079,
      "auxiliary_loss_mlp": 0.01029069,
      "balance_loss_clip": 1.03859377,
      "balance_loss_mlp": 1.01729679,
      "epoch": 0.8301818728393207,
      "flos": 19715784336000.0,
      "grad_norm": 1.7440578161079283,
      "language_loss": 0.7306065,
      "learning_rate": 2.948672208338847e-07,
      "loss": 0.751818,
      "num_input_tokens_seen": 297866730,
      "step": 13808,
      "time_per_iteration": 2.510754346847534
    },
    {
      "auxiliary_loss_clip": 0.01090179,
      "auxiliary_loss_mlp": 0.01045223,
      "balance_loss_clip": 1.03915524,
      "balance_loss_mlp": 1.03149557,
      "epoch": 0.8302419960919886,
      "flos": 28293658416000.0,
      "grad_norm": 1.7964508937190282,
      "language_loss": 0.66378236,
      "learning_rate": 2.9466371311533046e-07,
      "loss": 0.68513632,
      "num_input_tokens_seen": 297886390,
      "step": 13809,
      "time_per_iteration": 2.570053815841675
    },
    {
      "auxiliary_loss_clip": 0.01105757,
      "auxiliary_loss_mlp": 0.01025009,
      "balance_loss_clip": 1.03547847,
      "balance_loss_mlp": 1.01345801,
      "epoch": 0.8303021193446566,
      "flos": 18223444955520.0,
      "grad_norm": 1.9334740234690653,
      "language_loss": 0.74251157,
      "learning_rate": 2.9446027006329896e-07,
      "loss": 0.76381922,
      "num_input_tokens_seen": 297905110,
      "step": 13810,
      "time_per_iteration": 2.44181227684021
    },
    {
      "auxiliary_loss_clip": 0.01070984,
      "auxiliary_loss_mlp": 0.0103511,
      "balance_loss_clip": 1.03474879,
      "balance_loss_mlp": 1.02404118,
      "epoch": 0.8303622425973245,
      "flos": 23111425578240.0,
      "grad_norm": 1.8897016823174166,
      "language_loss": 0.81176174,
      "learning_rate": 2.94256891685505e-07,
      "loss": 0.83282268,
      "num_input_tokens_seen": 297925460,
      "step": 13811,
      "time_per_iteration": 2.5757362842559814
    },
    {
      "auxiliary_loss_clip": 0.01072704,
      "auxiliary_loss_mlp": 0.01044877,
      "balance_loss_clip": 1.03539741,
      "balance_loss_mlp": 1.03185368,
      "epoch": 0.8304223658499925,
      "flos": 19572860119680.0,
      "grad_norm": 1.8374575824274413,
      "language_loss": 0.73499203,
      "learning_rate": 2.9405357798966156e-07,
      "loss": 0.75616789,
      "num_input_tokens_seen": 297941760,
      "step": 13812,
      "time_per_iteration": 2.5577752590179443
    },
    {
      "auxiliary_loss_clip": 0.01077602,
      "auxiliary_loss_mlp": 0.01027388,
      "balance_loss_clip": 1.03615165,
      "balance_loss_mlp": 1.01551461,
      "epoch": 0.8304824891026604,
      "flos": 24426115269120.0,
      "grad_norm": 2.0168874191529795,
      "language_loss": 0.78330684,
      "learning_rate": 2.9385032898347664e-07,
      "loss": 0.80435669,
      "num_input_tokens_seen": 297959745,
      "step": 13813,
      "time_per_iteration": 2.5610148906707764
    },
    {
      "auxiliary_loss_clip": 0.01049527,
      "auxiliary_loss_mlp": 0.0078333,
      "balance_loss_clip": 1.03454852,
      "balance_loss_mlp": 1.01110053,
      "epoch": 0.8305426123553284,
      "flos": 22381792611840.0,
      "grad_norm": 2.0718368028024976,
      "language_loss": 0.70974678,
      "learning_rate": 2.93647144674658e-07,
      "loss": 0.72807539,
      "num_input_tokens_seen": 297977665,
      "step": 13814,
      "time_per_iteration": 2.6469643115997314
    },
    {
      "auxiliary_loss_clip": 0.01112308,
      "auxiliary_loss_mlp": 0.0104137,
      "balance_loss_clip": 1.03606236,
      "balance_loss_mlp": 1.0270766,
      "epoch": 0.8306027356079964,
      "flos": 14903575453440.0,
      "grad_norm": 2.579497449831375,
      "language_loss": 0.67571127,
      "learning_rate": 2.9344402507091116e-07,
      "loss": 0.69724804,
      "num_input_tokens_seen": 297993525,
      "step": 13815,
      "time_per_iteration": 2.430887460708618
    },
    {
      "auxiliary_loss_clip": 0.01091878,
      "auxiliary_loss_mlp": 0.0102957,
      "balance_loss_clip": 1.03755379,
      "balance_loss_mlp": 1.01752949,
      "epoch": 0.8306628588606644,
      "flos": 19644573623040.0,
      "grad_norm": 1.7966631091219092,
      "language_loss": 0.75682199,
      "learning_rate": 2.9324097017993745e-07,
      "loss": 0.77803648,
      "num_input_tokens_seen": 298012920,
      "step": 13816,
      "time_per_iteration": 2.5072364807128906
    },
    {
      "auxiliary_loss_clip": 0.01070114,
      "auxiliary_loss_mlp": 0.01033513,
      "balance_loss_clip": 1.03294337,
      "balance_loss_mlp": 1.0224328,
      "epoch": 0.8307229821133323,
      "flos": 24389737770240.0,
      "grad_norm": 1.920358366740514,
      "language_loss": 0.81662488,
      "learning_rate": 2.930379800094371e-07,
      "loss": 0.83766115,
      "num_input_tokens_seen": 298033310,
      "step": 13817,
      "time_per_iteration": 2.571570873260498
    },
    {
      "auxiliary_loss_clip": 0.0109636,
      "auxiliary_loss_mlp": 0.01037353,
      "balance_loss_clip": 1.0369544,
      "balance_loss_mlp": 1.0242219,
      "epoch": 0.8307831053660003,
      "flos": 20996933702400.0,
      "grad_norm": 1.711400714973582,
      "language_loss": 0.7787714,
      "learning_rate": 2.9283505456710875e-07,
      "loss": 0.80010855,
      "num_input_tokens_seen": 298053530,
      "step": 13818,
      "time_per_iteration": 2.5033414363861084
    },
    {
      "auxiliary_loss_clip": 0.01088483,
      "auxiliary_loss_mlp": 0.01034952,
      "balance_loss_clip": 1.03772843,
      "balance_loss_mlp": 1.02243519,
      "epoch": 0.8308432286186682,
      "flos": 21397301671680.0,
      "grad_norm": 1.7619217796044915,
      "language_loss": 0.82070011,
      "learning_rate": 2.926321938606453e-07,
      "loss": 0.8419345,
      "num_input_tokens_seen": 298069305,
      "step": 13819,
      "time_per_iteration": 2.518442153930664
    },
    {
      "auxiliary_loss_clip": 0.01020673,
      "auxiliary_loss_mlp": 0.01002921,
      "balance_loss_clip": 1.00827646,
      "balance_loss_mlp": 1.00174642,
      "epoch": 0.8309033518713362,
      "flos": 62533656714240.0,
      "grad_norm": 0.7740868584755801,
      "language_loss": 0.56237477,
      "learning_rate": 2.924293978977399e-07,
      "loss": 0.58261067,
      "num_input_tokens_seen": 298125830,
      "step": 13820,
      "time_per_iteration": 3.118082046508789
    },
    {
      "auxiliary_loss_clip": 0.01091261,
      "auxiliary_loss_mlp": 0.01026623,
      "balance_loss_clip": 1.03431511,
      "balance_loss_mlp": 1.01473784,
      "epoch": 0.8309634751240043,
      "flos": 16979104051200.0,
      "grad_norm": 1.8591452375605304,
      "language_loss": 0.68536413,
      "learning_rate": 2.922266666860831e-07,
      "loss": 0.70654309,
      "num_input_tokens_seen": 298142320,
      "step": 13821,
      "time_per_iteration": 2.459219217300415
    },
    {
      "auxiliary_loss_clip": 0.01041116,
      "auxiliary_loss_mlp": 0.01037156,
      "balance_loss_clip": 1.03171325,
      "balance_loss_mlp": 1.02345252,
      "epoch": 0.8310235983766722,
      "flos": 22674464628480.0,
      "grad_norm": 1.7266902105749462,
      "language_loss": 0.69199753,
      "learning_rate": 2.920240002333625e-07,
      "loss": 0.71278024,
      "num_input_tokens_seen": 298161845,
      "step": 13822,
      "time_per_iteration": 2.659160852432251
    },
    {
      "auxiliary_loss_clip": 0.01061073,
      "auxiliary_loss_mlp": 0.0103531,
      "balance_loss_clip": 1.03485012,
      "balance_loss_mlp": 1.0238719,
      "epoch": 0.8310837216293402,
      "flos": 30811463176320.0,
      "grad_norm": 1.8004188325569508,
      "language_loss": 0.62330663,
      "learning_rate": 2.918213985472631e-07,
      "loss": 0.64427042,
      "num_input_tokens_seen": 298184165,
      "step": 13823,
      "time_per_iteration": 2.6619222164154053
    },
    {
      "auxiliary_loss_clip": 0.01012857,
      "auxiliary_loss_mlp": 0.01003457,
      "balance_loss_clip": 1.00944126,
      "balance_loss_mlp": 1.00211608,
      "epoch": 0.8311438448820081,
      "flos": 71276074997760.0,
      "grad_norm": 0.8640424218083849,
      "language_loss": 0.6194942,
      "learning_rate": 2.916188616354669e-07,
      "loss": 0.63965732,
      "num_input_tokens_seen": 298251720,
      "step": 13824,
      "time_per_iteration": 3.2104811668395996
    },
    {
      "auxiliary_loss_clip": 0.01104163,
      "auxiliary_loss_mlp": 0.01032528,
      "balance_loss_clip": 1.03566754,
      "balance_loss_mlp": 1.02095318,
      "epoch": 0.8312039681346761,
      "flos": 20887082933760.0,
      "grad_norm": 1.6416754556244744,
      "language_loss": 0.7407347,
      "learning_rate": 2.914163895056552e-07,
      "loss": 0.76210153,
      "num_input_tokens_seen": 298271910,
      "step": 13825,
      "time_per_iteration": 2.470891237258911
    },
    {
      "auxiliary_loss_clip": 0.0106237,
      "auxiliary_loss_mlp": 0.00784129,
      "balance_loss_clip": 1.03376889,
      "balance_loss_mlp": 1.0080713,
      "epoch": 0.831264091387344,
      "flos": 17017528625280.0,
      "grad_norm": 2.964410229079954,
      "language_loss": 0.80369389,
      "learning_rate": 2.9121398216550486e-07,
      "loss": 0.82215887,
      "num_input_tokens_seen": 298288105,
      "step": 13826,
      "time_per_iteration": 3.95770001411438
    },
    {
      "auxiliary_loss_clip": 0.01104597,
      "auxiliary_loss_mlp": 0.01034557,
      "balance_loss_clip": 1.03528357,
      "balance_loss_mlp": 1.02198648,
      "epoch": 0.831324214640012,
      "flos": 24419578993920.0,
      "grad_norm": 1.6786912884258096,
      "language_loss": 0.67957854,
      "learning_rate": 2.910116396226914e-07,
      "loss": 0.70097017,
      "num_input_tokens_seen": 298307600,
      "step": 13827,
      "time_per_iteration": 2.506308078765869
    },
    {
      "auxiliary_loss_clip": 0.01091566,
      "auxiliary_loss_mlp": 0.01030623,
      "balance_loss_clip": 1.03189003,
      "balance_loss_mlp": 1.01947069,
      "epoch": 0.83138433789268,
      "flos": 13545576938880.0,
      "grad_norm": 1.7289220191035004,
      "language_loss": 0.74355602,
      "learning_rate": 2.9080936188488834e-07,
      "loss": 0.7647779,
      "num_input_tokens_seen": 298323055,
      "step": 13828,
      "time_per_iteration": 4.398099899291992
    },
    {
      "auxiliary_loss_clip": 0.01068505,
      "auxiliary_loss_mlp": 0.01037408,
      "balance_loss_clip": 1.03110981,
      "balance_loss_mlp": 1.02451599,
      "epoch": 0.831444461145348,
      "flos": 44492386561920.0,
      "grad_norm": 1.587930216290016,
      "language_loss": 0.67384195,
      "learning_rate": 2.906071489597657e-07,
      "loss": 0.69490117,
      "num_input_tokens_seen": 298346950,
      "step": 13829,
      "time_per_iteration": 2.765392303466797
    },
    {
      "auxiliary_loss_clip": 0.01079545,
      "auxiliary_loss_mlp": 0.01029366,
      "balance_loss_clip": 1.03434873,
      "balance_loss_mlp": 1.01675344,
      "epoch": 0.8315045843980159,
      "flos": 22705024124160.0,
      "grad_norm": 1.6465641042700396,
      "language_loss": 0.82832253,
      "learning_rate": 2.9040500085499054e-07,
      "loss": 0.84941155,
      "num_input_tokens_seen": 298366315,
      "step": 13830,
      "time_per_iteration": 2.541916608810425
    },
    {
      "auxiliary_loss_clip": 0.01092817,
      "auxiliary_loss_mlp": 0.01033307,
      "balance_loss_clip": 1.03513122,
      "balance_loss_mlp": 1.02185738,
      "epoch": 0.8315647076506839,
      "flos": 16873491087360.0,
      "grad_norm": 2.265703257620473,
      "language_loss": 0.74169832,
      "learning_rate": 2.9020291757822925e-07,
      "loss": 0.7629596,
      "num_input_tokens_seen": 298385185,
      "step": 13831,
      "time_per_iteration": 3.9343345165252686
    },
    {
      "auxiliary_loss_clip": 0.01106135,
      "auxiliary_loss_mlp": 0.01037668,
      "balance_loss_clip": 1.03678167,
      "balance_loss_mlp": 1.02428615,
      "epoch": 0.8316248309033518,
      "flos": 13808730954240.0,
      "grad_norm": 1.7406278027767697,
      "language_loss": 0.71125388,
      "learning_rate": 2.9000089913714523e-07,
      "loss": 0.73269188,
      "num_input_tokens_seen": 298402335,
      "step": 13832,
      "time_per_iteration": 2.4564247131347656
    },
    {
      "auxiliary_loss_clip": 0.01076447,
      "auxiliary_loss_mlp": 0.0103107,
      "balance_loss_clip": 1.0330745,
      "balance_loss_mlp": 1.01934624,
      "epoch": 0.8316849541560198,
      "flos": 23512511819520.0,
      "grad_norm": 1.6165300137387997,
      "language_loss": 0.84605062,
      "learning_rate": 2.897989455393979e-07,
      "loss": 0.86712581,
      "num_input_tokens_seen": 298423370,
      "step": 13833,
      "time_per_iteration": 2.5408313274383545
    },
    {
      "auxiliary_loss_clip": 0.01087161,
      "auxiliary_loss_mlp": 0.01035786,
      "balance_loss_clip": 1.03641653,
      "balance_loss_mlp": 1.02314949,
      "epoch": 0.8317450774086879,
      "flos": 23771356202880.0,
      "grad_norm": 1.616143626168053,
      "language_loss": 0.76063037,
      "learning_rate": 2.8959705679264625e-07,
      "loss": 0.78185987,
      "num_input_tokens_seen": 298444835,
      "step": 13834,
      "time_per_iteration": 2.5705676078796387
    },
    {
      "auxiliary_loss_clip": 0.01100134,
      "auxiliary_loss_mlp": 0.0078275,
      "balance_loss_clip": 1.03345323,
      "balance_loss_mlp": 1.00883603,
      "epoch": 0.8318052006613558,
      "flos": 16215535710720.0,
      "grad_norm": 2.0382646882263553,
      "language_loss": 0.79657006,
      "learning_rate": 2.893952329045459e-07,
      "loss": 0.81539893,
      "num_input_tokens_seen": 298461845,
      "step": 13835,
      "time_per_iteration": 2.42704176902771
    },
    {
      "auxiliary_loss_clip": 0.01098765,
      "auxiliary_loss_mlp": 0.01035844,
      "balance_loss_clip": 1.03730965,
      "balance_loss_mlp": 1.02221799,
      "epoch": 0.8318653239140238,
      "flos": 19974556892160.0,
      "grad_norm": 1.9011712807554872,
      "language_loss": 0.80940855,
      "learning_rate": 2.8919347388274905e-07,
      "loss": 0.83075464,
      "num_input_tokens_seen": 298479095,
      "step": 13836,
      "time_per_iteration": 3.959326982498169
    },
    {
      "auxiliary_loss_clip": 0.01081484,
      "auxiliary_loss_mlp": 0.01028271,
      "balance_loss_clip": 1.03530109,
      "balance_loss_mlp": 1.01702929,
      "epoch": 0.8319254471666917,
      "flos": 17704714694400.0,
      "grad_norm": 2.1507823156936947,
      "language_loss": 0.77705002,
      "learning_rate": 2.8899177973490727e-07,
      "loss": 0.79814756,
      "num_input_tokens_seen": 298494475,
      "step": 13837,
      "time_per_iteration": 2.58900785446167
    },
    {
      "auxiliary_loss_clip": 0.01109476,
      "auxiliary_loss_mlp": 0.01026331,
      "balance_loss_clip": 1.03574419,
      "balance_loss_mlp": 1.01305699,
      "epoch": 0.8319855704193597,
      "flos": 19536554448000.0,
      "grad_norm": 2.0208029252008335,
      "language_loss": 0.83265197,
      "learning_rate": 2.887901504686685e-07,
      "loss": 0.85401005,
      "num_input_tokens_seen": 298513185,
      "step": 13838,
      "time_per_iteration": 2.4313042163848877
    },
    {
      "auxiliary_loss_clip": 0.01077408,
      "auxiliary_loss_mlp": 0.0103128,
      "balance_loss_clip": 1.03346229,
      "balance_loss_mlp": 1.01806569,
      "epoch": 0.8320456936720276,
      "flos": 21178067011200.0,
      "grad_norm": 1.819072298477902,
      "language_loss": 0.74392223,
      "learning_rate": 2.885885860916795e-07,
      "loss": 0.76500916,
      "num_input_tokens_seen": 298531885,
      "step": 13839,
      "time_per_iteration": 2.5341439247131348
    },
    {
      "auxiliary_loss_clip": 0.0109244,
      "auxiliary_loss_mlp": 0.01033279,
      "balance_loss_clip": 1.03589034,
      "balance_loss_mlp": 1.02057719,
      "epoch": 0.8321058169246957,
      "flos": 33250874503680.0,
      "grad_norm": 1.5668461158895621,
      "language_loss": 0.67932194,
      "learning_rate": 2.8838708661158253e-07,
      "loss": 0.70057917,
      "num_input_tokens_seen": 298554905,
      "step": 13840,
      "time_per_iteration": 2.566868543624878
    },
    {
      "auxiliary_loss_clip": 0.01055573,
      "auxiliary_loss_mlp": 0.01034991,
      "balance_loss_clip": 1.03243506,
      "balance_loss_mlp": 1.02180028,
      "epoch": 0.8321659401773636,
      "flos": 14208129256320.0,
      "grad_norm": 1.980460929420664,
      "language_loss": 0.79520911,
      "learning_rate": 2.8818565203601843e-07,
      "loss": 0.81611472,
      "num_input_tokens_seen": 298571185,
      "step": 13841,
      "time_per_iteration": 2.544855833053589
    },
    {
      "auxiliary_loss_clip": 0.0106488,
      "auxiliary_loss_mlp": 0.01027616,
      "balance_loss_clip": 1.03826475,
      "balance_loss_mlp": 1.01561713,
      "epoch": 0.8322260634300316,
      "flos": 15158253859200.0,
      "grad_norm": 2.061346407831117,
      "language_loss": 0.68137437,
      "learning_rate": 2.879842823726262e-07,
      "loss": 0.70229936,
      "num_input_tokens_seen": 298588505,
      "step": 13842,
      "time_per_iteration": 2.51357364654541
    },
    {
      "auxiliary_loss_clip": 0.01085373,
      "auxiliary_loss_mlp": 0.0103034,
      "balance_loss_clip": 1.03674889,
      "balance_loss_mlp": 1.01733971,
      "epoch": 0.8322861866826995,
      "flos": 25300827267840.0,
      "grad_norm": 1.617530683105261,
      "language_loss": 0.73072314,
      "learning_rate": 2.8778297762904124e-07,
      "loss": 0.75188029,
      "num_input_tokens_seen": 298609295,
      "step": 13843,
      "time_per_iteration": 2.576111316680908
    },
    {
      "auxiliary_loss_clip": 0.01073764,
      "auxiliary_loss_mlp": 0.0103121,
      "balance_loss_clip": 1.03537178,
      "balance_loss_mlp": 1.01881194,
      "epoch": 0.8323463099353675,
      "flos": 17019360218880.0,
      "grad_norm": 1.9310246798231765,
      "language_loss": 0.77574241,
      "learning_rate": 2.875817378128975e-07,
      "loss": 0.79679215,
      "num_input_tokens_seen": 298625765,
      "step": 13844,
      "time_per_iteration": 2.511235237121582
    },
    {
      "auxiliary_loss_clip": 0.01009275,
      "auxiliary_loss_mlp": 0.01004387,
      "balance_loss_clip": 1.0090301,
      "balance_loss_mlp": 1.0033263,
      "epoch": 0.8324064331880354,
      "flos": 55607889709440.0,
      "grad_norm": 0.7746739475899296,
      "language_loss": 0.55243498,
      "learning_rate": 2.8738056293182624e-07,
      "loss": 0.57257158,
      "num_input_tokens_seen": 298683005,
      "step": 13845,
      "time_per_iteration": 3.039168119430542
    },
    {
      "auxiliary_loss_clip": 0.01096529,
      "auxiliary_loss_mlp": 0.01043849,
      "balance_loss_clip": 1.03501606,
      "balance_loss_mlp": 1.03061688,
      "epoch": 0.8324665564407034,
      "flos": 26138623063680.0,
      "grad_norm": 1.5738952989929795,
      "language_loss": 0.75534993,
      "learning_rate": 2.871794529934555e-07,
      "loss": 0.77675366,
      "num_input_tokens_seen": 298703060,
      "step": 13846,
      "time_per_iteration": 2.530256509780884
    },
    {
      "auxiliary_loss_clip": 0.0105329,
      "auxiliary_loss_mlp": 0.01029845,
      "balance_loss_clip": 1.03352284,
      "balance_loss_mlp": 1.01528943,
      "epoch": 0.8325266796933715,
      "flos": 22049187649920.0,
      "grad_norm": 1.5826268172048579,
      "language_loss": 0.78763586,
      "learning_rate": 2.8697840800541115e-07,
      "loss": 0.80846721,
      "num_input_tokens_seen": 298721765,
      "step": 13847,
      "time_per_iteration": 2.6231958866119385
    },
    {
      "auxiliary_loss_clip": 0.01047177,
      "auxiliary_loss_mlp": 0.01028347,
      "balance_loss_clip": 1.03402328,
      "balance_loss_mlp": 1.01719463,
      "epoch": 0.8325868029460394,
      "flos": 22816634659200.0,
      "grad_norm": 1.7718381518640798,
      "language_loss": 0.74497104,
      "learning_rate": 2.867774279753175e-07,
      "loss": 0.76572627,
      "num_input_tokens_seen": 298740825,
      "step": 13848,
      "time_per_iteration": 2.6224231719970703
    },
    {
      "auxiliary_loss_clip": 0.0109331,
      "auxiliary_loss_mlp": 0.01028276,
      "balance_loss_clip": 1.03567147,
      "balance_loss_mlp": 1.01674891,
      "epoch": 0.8326469261987074,
      "flos": 14757454926720.0,
      "grad_norm": 1.9043477972370058,
      "language_loss": 0.63249213,
      "learning_rate": 2.8657651291079554e-07,
      "loss": 0.65370804,
      "num_input_tokens_seen": 298758515,
      "step": 13849,
      "time_per_iteration": 2.4877092838287354
    },
    {
      "auxiliary_loss_clip": 0.0108169,
      "auxiliary_loss_mlp": 0.01030732,
      "balance_loss_clip": 1.03374314,
      "balance_loss_mlp": 1.01834035,
      "epoch": 0.8327070494513753,
      "flos": 22926126291840.0,
      "grad_norm": 2.083819047575844,
      "language_loss": 0.79501992,
      "learning_rate": 2.863756628194638e-07,
      "loss": 0.81614411,
      "num_input_tokens_seen": 298776375,
      "step": 13850,
      "time_per_iteration": 2.540769338607788
    },
    {
      "auxiliary_loss_clip": 0.0106377,
      "auxiliary_loss_mlp": 0.0103013,
      "balance_loss_clip": 1.03361177,
      "balance_loss_mlp": 1.01963317,
      "epoch": 0.8327671727040433,
      "flos": 20665334321280.0,
      "grad_norm": 1.7643637275965747,
      "language_loss": 0.78142476,
      "learning_rate": 2.8617487770893877e-07,
      "loss": 0.80236375,
      "num_input_tokens_seen": 298795135,
      "step": 13851,
      "time_per_iteration": 2.6344423294067383
    },
    {
      "auxiliary_loss_clip": 0.01024286,
      "auxiliary_loss_mlp": 0.01002848,
      "balance_loss_clip": 1.01124871,
      "balance_loss_mlp": 1.00162017,
      "epoch": 0.8328272959567112,
      "flos": 56060760384000.0,
      "grad_norm": 0.7605292237615213,
      "language_loss": 0.55796814,
      "learning_rate": 2.859741575868344e-07,
      "loss": 0.57823956,
      "num_input_tokens_seen": 298855475,
      "step": 13852,
      "time_per_iteration": 3.1143434047698975
    },
    {
      "auxiliary_loss_clip": 0.01093278,
      "auxiliary_loss_mlp": 0.01031995,
      "balance_loss_clip": 1.0362761,
      "balance_loss_mlp": 1.02005041,
      "epoch": 0.8328874192093793,
      "flos": 32303084284800.0,
      "grad_norm": 2.4400994335026964,
      "language_loss": 0.67189711,
      "learning_rate": 2.8577350246076125e-07,
      "loss": 0.69314981,
      "num_input_tokens_seen": 298875875,
      "step": 13853,
      "time_per_iteration": 2.595062255859375
    },
    {
      "auxiliary_loss_clip": 0.01078878,
      "auxiliary_loss_mlp": 0.01031398,
      "balance_loss_clip": 1.03696406,
      "balance_loss_mlp": 1.01920259,
      "epoch": 0.8329475424620472,
      "flos": 23512691387520.0,
      "grad_norm": 3.477963594565376,
      "language_loss": 0.7839483,
      "learning_rate": 2.855729123383286e-07,
      "loss": 0.80505109,
      "num_input_tokens_seen": 298895950,
      "step": 13854,
      "time_per_iteration": 2.5508546829223633
    },
    {
      "auxiliary_loss_clip": 0.01029396,
      "auxiliary_loss_mlp": 0.01002099,
      "balance_loss_clip": 1.00651121,
      "balance_loss_mlp": 1.00091922,
      "epoch": 0.8330076657147152,
      "flos": 67840680378240.0,
      "grad_norm": 0.7607812289757289,
      "language_loss": 0.5863871,
      "learning_rate": 2.8537238722714295e-07,
      "loss": 0.60670203,
      "num_input_tokens_seen": 298955770,
      "step": 13855,
      "time_per_iteration": 2.952234983444214
    },
    {
      "auxiliary_loss_clip": 0.0109144,
      "auxiliary_loss_mlp": 0.01027962,
      "balance_loss_clip": 1.03478849,
      "balance_loss_mlp": 1.01604092,
      "epoch": 0.8330677889673831,
      "flos": 22892801448960.0,
      "grad_norm": 1.701846098515264,
      "language_loss": 0.71314585,
      "learning_rate": 2.8517192713480853e-07,
      "loss": 0.73433989,
      "num_input_tokens_seen": 298976545,
      "step": 13856,
      "time_per_iteration": 2.5239620208740234
    },
    {
      "auxiliary_loss_clip": 0.01095465,
      "auxiliary_loss_mlp": 0.01030126,
      "balance_loss_clip": 1.03698039,
      "balance_loss_mlp": 1.01823473,
      "epoch": 0.8331279122200511,
      "flos": 27345042184320.0,
      "grad_norm": 1.5725319289184696,
      "language_loss": 0.75440431,
      "learning_rate": 2.8497153206892677e-07,
      "loss": 0.77566016,
      "num_input_tokens_seen": 298996750,
      "step": 13857,
      "time_per_iteration": 2.5327212810516357
    },
    {
      "auxiliary_loss_clip": 0.01054019,
      "auxiliary_loss_mlp": 0.01025877,
      "balance_loss_clip": 1.03555489,
      "balance_loss_mlp": 1.01554775,
      "epoch": 0.833188035472719,
      "flos": 19938179393280.0,
      "grad_norm": 1.5189819957092017,
      "language_loss": 0.73512679,
      "learning_rate": 2.847712020370958e-07,
      "loss": 0.75592577,
      "num_input_tokens_seen": 299014895,
      "step": 13858,
      "time_per_iteration": 2.618290424346924
    },
    {
      "auxiliary_loss_clip": 0.01106863,
      "auxiliary_loss_mlp": 0.01034178,
      "balance_loss_clip": 1.03421783,
      "balance_loss_mlp": 1.02169085,
      "epoch": 0.833248158725387,
      "flos": 15232624968960.0,
      "grad_norm": 1.7130368332875345,
      "language_loss": 0.73048264,
      "learning_rate": 2.8457093704691316e-07,
      "loss": 0.75189304,
      "num_input_tokens_seen": 299032855,
      "step": 13859,
      "time_per_iteration": 2.4309744834899902
    },
    {
      "auxiliary_loss_clip": 0.01088037,
      "auxiliary_loss_mlp": 0.01025161,
      "balance_loss_clip": 1.03379607,
      "balance_loss_mlp": 1.01396132,
      "epoch": 0.8333082819780551,
      "flos": 24535535074560.0,
      "grad_norm": 1.5834021951615072,
      "language_loss": 0.78791153,
      "learning_rate": 2.8437073710597205e-07,
      "loss": 0.80904353,
      "num_input_tokens_seen": 299052055,
      "step": 13860,
      "time_per_iteration": 2.55118465423584
    },
    {
      "auxiliary_loss_clip": 0.01031671,
      "auxiliary_loss_mlp": 0.01030321,
      "balance_loss_clip": 1.03313291,
      "balance_loss_mlp": 1.01856136,
      "epoch": 0.833368405230723,
      "flos": 31467407391360.0,
      "grad_norm": 1.4079100894360363,
      "language_loss": 0.82080173,
      "learning_rate": 2.841706022218644e-07,
      "loss": 0.84142166,
      "num_input_tokens_seen": 299075285,
      "step": 13861,
      "time_per_iteration": 2.7116944789886475
    },
    {
      "auxiliary_loss_clip": 0.01106311,
      "auxiliary_loss_mlp": 0.01032977,
      "balance_loss_clip": 1.0372932,
      "balance_loss_mlp": 1.02054954,
      "epoch": 0.833428528483391,
      "flos": 14902713527040.0,
      "grad_norm": 3.6154981789090668,
      "language_loss": 0.78941476,
      "learning_rate": 2.839705324021806e-07,
      "loss": 0.81080765,
      "num_input_tokens_seen": 299092520,
      "step": 13862,
      "time_per_iteration": 2.438392162322998
    },
    {
      "auxiliary_loss_clip": 0.01094642,
      "auxiliary_loss_mlp": 0.01032381,
      "balance_loss_clip": 1.03429842,
      "balance_loss_mlp": 1.01993585,
      "epoch": 0.8334886517360589,
      "flos": 22199833290240.0,
      "grad_norm": 2.1038970219088577,
      "language_loss": 0.75296247,
      "learning_rate": 2.83770527654505e-07,
      "loss": 0.77423275,
      "num_input_tokens_seen": 299109450,
      "step": 13863,
      "time_per_iteration": 2.4769136905670166
    },
    {
      "auxiliary_loss_clip": 0.01052962,
      "auxiliary_loss_mlp": 0.007851,
      "balance_loss_clip": 1.03492415,
      "balance_loss_mlp": 1.00957131,
      "epoch": 0.8335487749887269,
      "flos": 30372562892160.0,
      "grad_norm": 2.130234188607242,
      "language_loss": 0.75570917,
      "learning_rate": 2.835705879864232e-07,
      "loss": 0.77408981,
      "num_input_tokens_seen": 299129540,
      "step": 13864,
      "time_per_iteration": 2.667254686355591
    },
    {
      "auxiliary_loss_clip": 0.01082085,
      "auxiliary_loss_mlp": 0.01034364,
      "balance_loss_clip": 1.03492522,
      "balance_loss_mlp": 1.02142942,
      "epoch": 0.8336088982413948,
      "flos": 24681152810880.0,
      "grad_norm": 1.9931325777105322,
      "language_loss": 0.69256824,
      "learning_rate": 2.833707134055168e-07,
      "loss": 0.71373272,
      "num_input_tokens_seen": 299148670,
      "step": 13865,
      "time_per_iteration": 4.108296632766724
    },
    {
      "auxiliary_loss_clip": 0.01094064,
      "auxiliary_loss_mlp": 0.01033506,
      "balance_loss_clip": 1.03567505,
      "balance_loss_mlp": 1.02170408,
      "epoch": 0.8336690214940629,
      "flos": 38177207873280.0,
      "grad_norm": 1.9767910396076345,
      "language_loss": 0.75635284,
      "learning_rate": 2.831709039193653e-07,
      "loss": 0.77762848,
      "num_input_tokens_seen": 299169330,
      "step": 13866,
      "time_per_iteration": 4.078115224838257
    },
    {
      "auxiliary_loss_clip": 0.01012891,
      "auxiliary_loss_mlp": 0.01013991,
      "balance_loss_clip": 1.01313329,
      "balance_loss_mlp": 1.01253116,
      "epoch": 0.8337291447467308,
      "flos": 55565119589760.0,
      "grad_norm": 0.8901576806567469,
      "language_loss": 0.63053864,
      "learning_rate": 2.8297115953554465e-07,
      "loss": 0.65080756,
      "num_input_tokens_seen": 299220980,
      "step": 13867,
      "time_per_iteration": 3.0569379329681396
    },
    {
      "auxiliary_loss_clip": 0.01079657,
      "auxiliary_loss_mlp": 0.01032234,
      "balance_loss_clip": 1.03479934,
      "balance_loss_mlp": 1.02143991,
      "epoch": 0.8337892679993988,
      "flos": 24133550993280.0,
      "grad_norm": 2.423548606391088,
      "language_loss": 0.72072333,
      "learning_rate": 2.827714802616301e-07,
      "loss": 0.74184227,
      "num_input_tokens_seen": 299240130,
      "step": 13868,
      "time_per_iteration": 2.5332343578338623
    },
    {
      "auxiliary_loss_clip": 0.01081056,
      "auxiliary_loss_mlp": 0.01030113,
      "balance_loss_clip": 1.03814507,
      "balance_loss_mlp": 1.01782227,
      "epoch": 0.8338493912520667,
      "flos": 28183915388160.0,
      "grad_norm": 1.475703808554861,
      "language_loss": 0.80486089,
      "learning_rate": 2.8257186610519325e-07,
      "loss": 0.82597256,
      "num_input_tokens_seen": 299260705,
      "step": 13869,
      "time_per_iteration": 2.554749011993408
    },
    {
      "auxiliary_loss_clip": 0.01096348,
      "auxiliary_loss_mlp": 0.01033725,
      "balance_loss_clip": 1.03676331,
      "balance_loss_mlp": 1.02152383,
      "epoch": 0.8339095145047347,
      "flos": 22158356060160.0,
      "grad_norm": 1.6507595584300727,
      "language_loss": 0.82589674,
      "learning_rate": 2.823723170738028e-07,
      "loss": 0.84719741,
      "num_input_tokens_seen": 299278925,
      "step": 13870,
      "time_per_iteration": 3.8960795402526855
    },
    {
      "auxiliary_loss_clip": 0.01080182,
      "auxiliary_loss_mlp": 0.01028251,
      "balance_loss_clip": 1.03505039,
      "balance_loss_mlp": 1.01568639,
      "epoch": 0.8339696377574026,
      "flos": 17307112072320.0,
      "grad_norm": 2.541368547540725,
      "language_loss": 0.70527685,
      "learning_rate": 2.821728331750264e-07,
      "loss": 0.72636122,
      "num_input_tokens_seen": 299291580,
      "step": 13871,
      "time_per_iteration": 2.4710822105407715
    },
    {
      "auxiliary_loss_clip": 0.01092252,
      "auxiliary_loss_mlp": 0.01030798,
      "balance_loss_clip": 1.03607607,
      "balance_loss_mlp": 1.01919293,
      "epoch": 0.8340297610100706,
      "flos": 20668351063680.0,
      "grad_norm": 1.8044358301572225,
      "language_loss": 0.68934268,
      "learning_rate": 2.8197341441642853e-07,
      "loss": 0.7105732,
      "num_input_tokens_seen": 299310385,
      "step": 13872,
      "time_per_iteration": 2.502723455429077
    },
    {
      "auxiliary_loss_clip": 0.01081147,
      "auxiliary_loss_mlp": 0.01025923,
      "balance_loss_clip": 1.03466702,
      "balance_loss_mlp": 1.01437747,
      "epoch": 0.8340898842627387,
      "flos": 20515442866560.0,
      "grad_norm": 1.9107878354892271,
      "language_loss": 0.73293805,
      "learning_rate": 2.817740608055712e-07,
      "loss": 0.75400871,
      "num_input_tokens_seen": 299327660,
      "step": 13873,
      "time_per_iteration": 2.505671977996826
    },
    {
      "auxiliary_loss_clip": 0.01081725,
      "auxiliary_loss_mlp": 0.01033788,
      "balance_loss_clip": 1.03444195,
      "balance_loss_mlp": 1.01992953,
      "epoch": 0.8341500075154066,
      "flos": 21425850005760.0,
      "grad_norm": 1.9721039835565375,
      "language_loss": 0.75145042,
      "learning_rate": 2.81574772350013e-07,
      "loss": 0.77260554,
      "num_input_tokens_seen": 299343685,
      "step": 13874,
      "time_per_iteration": 3.913151979446411
    },
    {
      "auxiliary_loss_clip": 0.01074839,
      "auxiliary_loss_mlp": 0.01026649,
      "balance_loss_clip": 1.03305435,
      "balance_loss_mlp": 1.01549101,
      "epoch": 0.8342101307680746,
      "flos": 22090988102400.0,
      "grad_norm": 1.7795245170499925,
      "language_loss": 0.66055781,
      "learning_rate": 2.813755490573118e-07,
      "loss": 0.68157268,
      "num_input_tokens_seen": 299363305,
      "step": 13875,
      "time_per_iteration": 2.564612865447998
    },
    {
      "auxiliary_loss_clip": 0.01061669,
      "auxiliary_loss_mlp": 0.01037671,
      "balance_loss_clip": 1.03508067,
      "balance_loss_mlp": 1.02536869,
      "epoch": 0.8342702540207425,
      "flos": 21871466133120.0,
      "grad_norm": 1.9364064207484857,
      "language_loss": 0.79824018,
      "learning_rate": 2.8117639093502243e-07,
      "loss": 0.81923354,
      "num_input_tokens_seen": 299382630,
      "step": 13876,
      "time_per_iteration": 2.595637083053589
    },
    {
      "auxiliary_loss_clip": 0.01088402,
      "auxiliary_loss_mlp": 0.01038255,
      "balance_loss_clip": 1.03363967,
      "balance_loss_mlp": 1.02479601,
      "epoch": 0.8343303772734105,
      "flos": 22528487756160.0,
      "grad_norm": 1.8512335187722102,
      "language_loss": 0.87282461,
      "learning_rate": 2.8097729799069615e-07,
      "loss": 0.89409113,
      "num_input_tokens_seen": 299402385,
      "step": 13877,
      "time_per_iteration": 2.5206453800201416
    },
    {
      "auxiliary_loss_clip": 0.01068984,
      "auxiliary_loss_mlp": 0.01029031,
      "balance_loss_clip": 1.03490162,
      "balance_loss_mlp": 1.01751506,
      "epoch": 0.8343905005260784,
      "flos": 14939773384320.0,
      "grad_norm": 1.7847799874166084,
      "language_loss": 0.69390225,
      "learning_rate": 2.807782702318828e-07,
      "loss": 0.71488237,
      "num_input_tokens_seen": 299419820,
      "step": 13878,
      "time_per_iteration": 2.5035104751586914
    },
    {
      "auxiliary_loss_clip": 0.01081611,
      "auxiliary_loss_mlp": 0.01028885,
      "balance_loss_clip": 1.03513813,
      "balance_loss_mlp": 1.01743484,
      "epoch": 0.8344506237787465,
      "flos": 15012456554880.0,
      "grad_norm": 1.93915079639133,
      "language_loss": 0.79414845,
      "learning_rate": 2.805793076661309e-07,
      "loss": 0.81525344,
      "num_input_tokens_seen": 299436265,
      "step": 13879,
      "time_per_iteration": 2.5225470066070557
    },
    {
      "auxiliary_loss_clip": 0.01053377,
      "auxiliary_loss_mlp": 0.01030574,
      "balance_loss_clip": 1.03475571,
      "balance_loss_mlp": 1.02007759,
      "epoch": 0.8345107470314144,
      "flos": 17560389847680.0,
      "grad_norm": 2.2829282878134793,
      "language_loss": 0.83192712,
      "learning_rate": 2.803804103009828e-07,
      "loss": 0.85276663,
      "num_input_tokens_seen": 299451660,
      "step": 13880,
      "time_per_iteration": 2.5448951721191406
    },
    {
      "auxiliary_loss_clip": 0.01085892,
      "auxiliary_loss_mlp": 0.01029446,
      "balance_loss_clip": 1.03489232,
      "balance_loss_mlp": 1.01794839,
      "epoch": 0.8345708702840824,
      "flos": 25187277398400.0,
      "grad_norm": 1.472788727312601,
      "language_loss": 0.78227401,
      "learning_rate": 2.80181578143982e-07,
      "loss": 0.80342734,
      "num_input_tokens_seen": 299472070,
      "step": 13881,
      "time_per_iteration": 2.5989532470703125
    },
    {
      "auxiliary_loss_clip": 0.01060914,
      "auxiliary_loss_mlp": 0.01026204,
      "balance_loss_clip": 1.03454423,
      "balance_loss_mlp": 1.01561821,
      "epoch": 0.8346309935367503,
      "flos": 15083559527040.0,
      "grad_norm": 2.3211582985882,
      "language_loss": 0.77961683,
      "learning_rate": 2.7998281120266807e-07,
      "loss": 0.80048805,
      "num_input_tokens_seen": 299486725,
      "step": 13882,
      "time_per_iteration": 2.505530595779419
    },
    {
      "auxiliary_loss_clip": 0.01066294,
      "auxiliary_loss_mlp": 0.01047789,
      "balance_loss_clip": 1.03276217,
      "balance_loss_mlp": 1.03438401,
      "epoch": 0.8346911167894183,
      "flos": 22930615491840.0,
      "grad_norm": 1.6966508090055339,
      "language_loss": 0.80450737,
      "learning_rate": 2.79784109484579e-07,
      "loss": 0.82564825,
      "num_input_tokens_seen": 299505435,
      "step": 13883,
      "time_per_iteration": 2.5603532791137695
    },
    {
      "auxiliary_loss_clip": 0.0109173,
      "auxiliary_loss_mlp": 0.010298,
      "balance_loss_clip": 1.0357182,
      "balance_loss_mlp": 1.01749778,
      "epoch": 0.8347512400420862,
      "flos": 20193037367040.0,
      "grad_norm": 1.8003296327033074,
      "language_loss": 0.74219984,
      "learning_rate": 2.795854729972482e-07,
      "loss": 0.76341522,
      "num_input_tokens_seen": 299523555,
      "step": 13884,
      "time_per_iteration": 2.4747161865234375
    },
    {
      "auxiliary_loss_clip": 0.01089756,
      "auxiliary_loss_mlp": 0.01034445,
      "balance_loss_clip": 1.03673244,
      "balance_loss_mlp": 1.0210278,
      "epoch": 0.8348113632947542,
      "flos": 25954832148480.0,
      "grad_norm": 1.8330064031024795,
      "language_loss": 0.70304674,
      "learning_rate": 2.7938690174820913e-07,
      "loss": 0.72428876,
      "num_input_tokens_seen": 299541660,
      "step": 13885,
      "time_per_iteration": 2.5866193771362305
    },
    {
      "auxiliary_loss_clip": 0.01071466,
      "auxiliary_loss_mlp": 0.01029314,
      "balance_loss_clip": 1.03420997,
      "balance_loss_mlp": 1.01707125,
      "epoch": 0.8348714865474223,
      "flos": 34204554552960.0,
      "grad_norm": 2.181688033740228,
      "language_loss": 0.70081067,
      "learning_rate": 2.791883957449912e-07,
      "loss": 0.72181851,
      "num_input_tokens_seen": 299562465,
      "step": 13886,
      "time_per_iteration": 2.6421046257019043
    },
    {
      "auxiliary_loss_clip": 0.01067265,
      "auxiliary_loss_mlp": 0.01033304,
      "balance_loss_clip": 1.03324401,
      "balance_loss_mlp": 1.02054238,
      "epoch": 0.8349316098000902,
      "flos": 24390132819840.0,
      "grad_norm": 1.8577313011952379,
      "language_loss": 0.78977764,
      "learning_rate": 2.7898995499512134e-07,
      "loss": 0.81078339,
      "num_input_tokens_seen": 299582700,
      "step": 13887,
      "time_per_iteration": 2.6188101768493652
    },
    {
      "auxiliary_loss_clip": 0.01086795,
      "auxiliary_loss_mlp": 0.00785695,
      "balance_loss_clip": 1.0364598,
      "balance_loss_mlp": 1.0102936,
      "epoch": 0.8349917330527582,
      "flos": 23032744836480.0,
      "grad_norm": 2.2957056243692335,
      "language_loss": 0.64211953,
      "learning_rate": 2.7879157950612467e-07,
      "loss": 0.66084439,
      "num_input_tokens_seen": 299600310,
      "step": 13888,
      "time_per_iteration": 2.5228805541992188
    },
    {
      "auxiliary_loss_clip": 0.0108195,
      "auxiliary_loss_mlp": 0.01028478,
      "balance_loss_clip": 1.03353035,
      "balance_loss_mlp": 1.01628852,
      "epoch": 0.8350518563054261,
      "flos": 13625873792640.0,
      "grad_norm": 2.186513991467204,
      "language_loss": 0.66695374,
      "learning_rate": 2.785932692855244e-07,
      "loss": 0.68805802,
      "num_input_tokens_seen": 299617025,
      "step": 13889,
      "time_per_iteration": 2.5198302268981934
    },
    {
      "auxiliary_loss_clip": 0.01085435,
      "auxiliary_loss_mlp": 0.01028587,
      "balance_loss_clip": 1.03172445,
      "balance_loss_mlp": 1.01662469,
      "epoch": 0.8351119795580941,
      "flos": 21579799697280.0,
      "grad_norm": 1.8138056790320118,
      "language_loss": 0.68607134,
      "learning_rate": 2.783950243408399e-07,
      "loss": 0.70721161,
      "num_input_tokens_seen": 299633050,
      "step": 13890,
      "time_per_iteration": 2.476487398147583
    },
    {
      "auxiliary_loss_clip": 0.01084932,
      "auxiliary_loss_mlp": 0.0103621,
      "balance_loss_clip": 1.03664231,
      "balance_loss_mlp": 1.02437282,
      "epoch": 0.835172102810762,
      "flos": 20038297576320.0,
      "grad_norm": 2.283780037424424,
      "language_loss": 0.59399426,
      "learning_rate": 2.7819684467958817e-07,
      "loss": 0.61520571,
      "num_input_tokens_seen": 299646445,
      "step": 13891,
      "time_per_iteration": 2.503169059753418
    },
    {
      "auxiliary_loss_clip": 0.01094638,
      "auxiliary_loss_mlp": 0.0102716,
      "balance_loss_clip": 1.03630149,
      "balance_loss_mlp": 1.01553726,
      "epoch": 0.8352322260634301,
      "flos": 25111577485440.0,
      "grad_norm": 1.7903443509030557,
      "language_loss": 0.71742284,
      "learning_rate": 2.779987303092846e-07,
      "loss": 0.73864079,
      "num_input_tokens_seen": 299662665,
      "step": 13892,
      "time_per_iteration": 2.5205421447753906
    },
    {
      "auxiliary_loss_clip": 0.01100877,
      "auxiliary_loss_mlp": 0.01029976,
      "balance_loss_clip": 1.03415322,
      "balance_loss_mlp": 1.01791167,
      "epoch": 0.835292349316098,
      "flos": 24863758577280.0,
      "grad_norm": 1.5547044601312565,
      "language_loss": 0.65830636,
      "learning_rate": 2.7780068123744207e-07,
      "loss": 0.6796149,
      "num_input_tokens_seen": 299683585,
      "step": 13893,
      "time_per_iteration": 2.5199155807495117
    },
    {
      "auxiliary_loss_clip": 0.01081033,
      "auxiliary_loss_mlp": 0.01026748,
      "balance_loss_clip": 1.03322959,
      "balance_loss_mlp": 1.01526856,
      "epoch": 0.835352472568766,
      "flos": 19865568049920.0,
      "grad_norm": 2.059871882482799,
      "language_loss": 0.78463113,
      "learning_rate": 2.7760269747156996e-07,
      "loss": 0.80570894,
      "num_input_tokens_seen": 299702680,
      "step": 13894,
      "time_per_iteration": 2.5039217472076416
    },
    {
      "auxiliary_loss_clip": 0.01085461,
      "auxiliary_loss_mlp": 0.01030684,
      "balance_loss_clip": 1.03394771,
      "balance_loss_mlp": 1.01836336,
      "epoch": 0.8354125958214339,
      "flos": 22054754257920.0,
      "grad_norm": 1.6405816990444655,
      "language_loss": 0.72453892,
      "learning_rate": 2.7740477901917625e-07,
      "loss": 0.74570036,
      "num_input_tokens_seen": 299721050,
      "step": 13895,
      "time_per_iteration": 2.513826370239258
    },
    {
      "auxiliary_loss_clip": 0.01091558,
      "auxiliary_loss_mlp": 0.01039436,
      "balance_loss_clip": 1.03588152,
      "balance_loss_mlp": 1.0260905,
      "epoch": 0.8354727190741019,
      "flos": 21397804462080.0,
      "grad_norm": 2.030057774241665,
      "language_loss": 0.71964276,
      "learning_rate": 2.772069258877667e-07,
      "loss": 0.74095273,
      "num_input_tokens_seen": 299738255,
      "step": 13896,
      "time_per_iteration": 2.5036683082580566
    },
    {
      "auxiliary_loss_clip": 0.01092716,
      "auxiliary_loss_mlp": 0.01031574,
      "balance_loss_clip": 1.03503382,
      "balance_loss_mlp": 1.01978397,
      "epoch": 0.8355328423267698,
      "flos": 50840997834240.0,
      "grad_norm": 2.1632513583616846,
      "language_loss": 0.58833677,
      "learning_rate": 2.770091380848423e-07,
      "loss": 0.60957968,
      "num_input_tokens_seen": 299761315,
      "step": 13897,
      "time_per_iteration": 2.733217477798462
    },
    {
      "auxiliary_loss_clip": 0.01029532,
      "auxiliary_loss_mlp": 0.00764137,
      "balance_loss_clip": 1.00654733,
      "balance_loss_mlp": 1.00675249,
      "epoch": 0.8355929655794379,
      "flos": 65551052764800.0,
      "grad_norm": 0.6938136141987127,
      "language_loss": 0.57626593,
      "learning_rate": 2.7681141561790423e-07,
      "loss": 0.59420264,
      "num_input_tokens_seen": 299828735,
      "step": 13898,
      "time_per_iteration": 3.13100528717041
    },
    {
      "auxiliary_loss_clip": 0.01097078,
      "auxiliary_loss_mlp": 0.01036137,
      "balance_loss_clip": 1.0362078,
      "balance_loss_mlp": 1.02306545,
      "epoch": 0.8356530888321058,
      "flos": 19170516902400.0,
      "grad_norm": 2.466985144977586,
      "language_loss": 0.79815131,
      "learning_rate": 2.7661375849444967e-07,
      "loss": 0.81948346,
      "num_input_tokens_seen": 299848395,
      "step": 13899,
      "time_per_iteration": 2.501960277557373
    },
    {
      "auxiliary_loss_clip": 0.0110481,
      "auxiliary_loss_mlp": 0.01032395,
      "balance_loss_clip": 1.03544998,
      "balance_loss_mlp": 1.02098072,
      "epoch": 0.8357132120847738,
      "flos": 44126672238720.0,
      "grad_norm": 1.5597706485444327,
      "language_loss": 0.68635029,
      "learning_rate": 2.764161667219749e-07,
      "loss": 0.70772231,
      "num_input_tokens_seen": 299871665,
      "step": 13900,
      "time_per_iteration": 2.669365167617798
    },
    {
      "auxiliary_loss_clip": 0.01082798,
      "auxiliary_loss_mlp": 0.01029843,
      "balance_loss_clip": 1.0392127,
      "balance_loss_mlp": 1.01846409,
      "epoch": 0.8357733353374418,
      "flos": 24389701856640.0,
      "grad_norm": 1.370612517600156,
      "language_loss": 0.70928055,
      "learning_rate": 2.762186403079716e-07,
      "loss": 0.730407,
      "num_input_tokens_seen": 299891960,
      "step": 13901,
      "time_per_iteration": 2.554680585861206
    },
    {
      "auxiliary_loss_clip": 0.0106153,
      "auxiliary_loss_mlp": 0.01037029,
      "balance_loss_clip": 1.03442097,
      "balance_loss_mlp": 1.02460766,
      "epoch": 0.8358334585901097,
      "flos": 20916313626240.0,
      "grad_norm": 2.0228589356738453,
      "language_loss": 0.799573,
      "learning_rate": 2.7602117925992963e-07,
      "loss": 0.82055861,
      "num_input_tokens_seen": 299905070,
      "step": 13902,
      "time_per_iteration": 2.5529561042785645
    },
    {
      "auxiliary_loss_clip": 0.01092139,
      "auxiliary_loss_mlp": 0.01031424,
      "balance_loss_clip": 1.0354079,
      "balance_loss_mlp": 1.01937795,
      "epoch": 0.8358935818427777,
      "flos": 19244169740160.0,
      "grad_norm": 1.463148152753567,
      "language_loss": 0.62521505,
      "learning_rate": 2.758237835853379e-07,
      "loss": 0.64645064,
      "num_input_tokens_seen": 299925130,
      "step": 13903,
      "time_per_iteration": 3.8545761108398438
    },
    {
      "auxiliary_loss_clip": 0.01080627,
      "auxiliary_loss_mlp": 0.01037547,
      "balance_loss_clip": 1.03430605,
      "balance_loss_mlp": 1.02481532,
      "epoch": 0.8359537050954456,
      "flos": 24134053783680.0,
      "grad_norm": 1.847160944925841,
      "language_loss": 0.74571294,
      "learning_rate": 2.7562645329168054e-07,
      "loss": 0.7668947,
      "num_input_tokens_seen": 299943845,
      "step": 13904,
      "time_per_iteration": 2.548473358154297
    },
    {
      "auxiliary_loss_clip": 0.01076117,
      "auxiliary_loss_mlp": 0.01028854,
      "balance_loss_clip": 1.03214836,
      "balance_loss_mlp": 1.01680231,
      "epoch": 0.8360138283481137,
      "flos": 16180415187840.0,
      "grad_norm": 1.7533997525022829,
      "language_loss": 0.72720921,
      "learning_rate": 2.7542918838644104e-07,
      "loss": 0.74825895,
      "num_input_tokens_seen": 299961620,
      "step": 13905,
      "time_per_iteration": 3.906623601913452
    },
    {
      "auxiliary_loss_clip": 0.01092753,
      "auxiliary_loss_mlp": 0.01034268,
      "balance_loss_clip": 1.0366708,
      "balance_loss_mlp": 1.02333105,
      "epoch": 0.8360739516007816,
      "flos": 22198899536640.0,
      "grad_norm": 1.608054588213091,
      "language_loss": 0.66458142,
      "learning_rate": 2.752319888771e-07,
      "loss": 0.68585157,
      "num_input_tokens_seen": 299982170,
      "step": 13906,
      "time_per_iteration": 2.499828815460205
    },
    {
      "auxiliary_loss_clip": 0.01091793,
      "auxiliary_loss_mlp": 0.01028962,
      "balance_loss_clip": 1.03503489,
      "balance_loss_mlp": 1.01725519,
      "epoch": 0.8361340748534496,
      "flos": 20923137210240.0,
      "grad_norm": 1.468936455740421,
      "language_loss": 0.74027067,
      "learning_rate": 2.7503485477113475e-07,
      "loss": 0.76147825,
      "num_input_tokens_seen": 300001330,
      "step": 13907,
      "time_per_iteration": 2.507209539413452
    },
    {
      "auxiliary_loss_clip": 0.01072263,
      "auxiliary_loss_mlp": 0.01034414,
      "balance_loss_clip": 1.03366566,
      "balance_loss_mlp": 1.02233791,
      "epoch": 0.8361941981061175,
      "flos": 26173599932160.0,
      "grad_norm": 1.606919285973566,
      "language_loss": 0.75386679,
      "learning_rate": 2.7483778607602005e-07,
      "loss": 0.77493346,
      "num_input_tokens_seen": 300020645,
      "step": 13908,
      "time_per_iteration": 4.042942523956299
    },
    {
      "auxiliary_loss_clip": 0.01096012,
      "auxiliary_loss_mlp": 0.01032852,
      "balance_loss_clip": 1.03608489,
      "balance_loss_mlp": 1.01945901,
      "epoch": 0.8362543213587855,
      "flos": 24419363512320.0,
      "grad_norm": 2.218825363529686,
      "language_loss": 0.71497732,
      "learning_rate": 2.7464078279922964e-07,
      "loss": 0.73626602,
      "num_input_tokens_seen": 300039945,
      "step": 13909,
      "time_per_iteration": 2.5318310260772705
    },
    {
      "auxiliary_loss_clip": 0.01106626,
      "auxiliary_loss_mlp": 0.00784047,
      "balance_loss_clip": 1.03529763,
      "balance_loss_mlp": 1.00910175,
      "epoch": 0.8363144446114534,
      "flos": 17202396948480.0,
      "grad_norm": 1.8327195021422595,
      "language_loss": 0.73062122,
      "learning_rate": 2.744438449482338e-07,
      "loss": 0.74952799,
      "num_input_tokens_seen": 300058260,
      "step": 13910,
      "time_per_iteration": 2.4225032329559326
    },
    {
      "auxiliary_loss_clip": 0.01089592,
      "auxiliary_loss_mlp": 0.00784057,
      "balance_loss_clip": 1.03404248,
      "balance_loss_mlp": 1.00765872,
      "epoch": 0.8363745678641215,
      "flos": 19279398003840.0,
      "grad_norm": 1.6823936594978595,
      "language_loss": 0.73290712,
      "learning_rate": 2.742469725305001e-07,
      "loss": 0.7516436,
      "num_input_tokens_seen": 300076720,
      "step": 13911,
      "time_per_iteration": 2.489025115966797
    },
    {
      "auxiliary_loss_clip": 0.01084984,
      "auxiliary_loss_mlp": 0.01039794,
      "balance_loss_clip": 1.03482008,
      "balance_loss_mlp": 1.02791464,
      "epoch": 0.8364346911167894,
      "flos": 11874869596800.0,
      "grad_norm": 2.093106860618108,
      "language_loss": 0.78895736,
      "learning_rate": 2.740501655534946e-07,
      "loss": 0.8102051,
      "num_input_tokens_seen": 300092950,
      "step": 13912,
      "time_per_iteration": 2.585545539855957
    },
    {
      "auxiliary_loss_clip": 0.0109712,
      "auxiliary_loss_mlp": 0.01033326,
      "balance_loss_clip": 1.0381577,
      "balance_loss_mlp": 1.02193522,
      "epoch": 0.8364948143694574,
      "flos": 20225212974720.0,
      "grad_norm": 1.748465039759902,
      "language_loss": 0.79009497,
      "learning_rate": 2.738534240246797e-07,
      "loss": 0.8113994,
      "num_input_tokens_seen": 300110950,
      "step": 13913,
      "time_per_iteration": 3.9602251052856445
    },
    {
      "auxiliary_loss_clip": 0.01091791,
      "auxiliary_loss_mlp": 0.01034412,
      "balance_loss_clip": 1.03370547,
      "balance_loss_mlp": 1.02223492,
      "epoch": 0.8365549376221254,
      "flos": 21612909058560.0,
      "grad_norm": 1.9147541466787323,
      "language_loss": 0.72987872,
      "learning_rate": 2.736567479515153e-07,
      "loss": 0.75114071,
      "num_input_tokens_seen": 300128705,
      "step": 13914,
      "time_per_iteration": 2.560734987258911
    },
    {
      "auxiliary_loss_clip": 0.01057013,
      "auxiliary_loss_mlp": 0.01028994,
      "balance_loss_clip": 1.0362581,
      "balance_loss_mlp": 1.0168705,
      "epoch": 0.8366150608747933,
      "flos": 23294210912640.0,
      "grad_norm": 1.591561488744853,
      "language_loss": 0.71130073,
      "learning_rate": 2.7346013734146025e-07,
      "loss": 0.73216081,
      "num_input_tokens_seen": 300148635,
      "step": 13915,
      "time_per_iteration": 2.6519644260406494
    },
    {
      "auxiliary_loss_clip": 0.0107284,
      "auxiliary_loss_mlp": 0.01031662,
      "balance_loss_clip": 1.0339992,
      "balance_loss_mlp": 1.02046824,
      "epoch": 0.8366751841274613,
      "flos": 15267673664640.0,
      "grad_norm": 1.9263281553753617,
      "language_loss": 0.7234177,
      "learning_rate": 2.7326359220197035e-07,
      "loss": 0.74446273,
      "num_input_tokens_seen": 300165490,
      "step": 13916,
      "time_per_iteration": 2.614776372909546
    },
    {
      "auxiliary_loss_clip": 0.0107219,
      "auxiliary_loss_mlp": 0.00782171,
      "balance_loss_clip": 1.03526783,
      "balance_loss_mlp": 1.00768495,
      "epoch": 0.8367353073801292,
      "flos": 13224931205760.0,
      "grad_norm": 1.9438688982750534,
      "language_loss": 0.74707592,
      "learning_rate": 2.7306711254049755e-07,
      "loss": 0.76561958,
      "num_input_tokens_seen": 300182130,
      "step": 13917,
      "time_per_iteration": 2.575924873352051
    },
    {
      "auxiliary_loss_clip": 0.01101537,
      "auxiliary_loss_mlp": 0.01029969,
      "balance_loss_clip": 1.03658068,
      "balance_loss_mlp": 1.0187459,
      "epoch": 0.8367954306327973,
      "flos": 24205084928640.0,
      "grad_norm": 1.7982244206097922,
      "language_loss": 0.78647578,
      "learning_rate": 2.728706983644933e-07,
      "loss": 0.80779088,
      "num_input_tokens_seen": 300203050,
      "step": 13918,
      "time_per_iteration": 2.561345338821411
    },
    {
      "auxiliary_loss_clip": 0.01059896,
      "auxiliary_loss_mlp": 0.01032992,
      "balance_loss_clip": 1.03438079,
      "balance_loss_mlp": 1.02118444,
      "epoch": 0.8368555538854652,
      "flos": 24534744975360.0,
      "grad_norm": 1.5559761182797092,
      "language_loss": 0.68017256,
      "learning_rate": 2.7267434968140457e-07,
      "loss": 0.70110142,
      "num_input_tokens_seen": 300224380,
      "step": 13919,
      "time_per_iteration": 2.6836633682250977
    },
    {
      "auxiliary_loss_clip": 0.01086504,
      "auxiliary_loss_mlp": 0.0103474,
      "balance_loss_clip": 1.03212237,
      "balance_loss_mlp": 1.02231812,
      "epoch": 0.8369156771381332,
      "flos": 20259363830400.0,
      "grad_norm": 1.7097027352243663,
      "language_loss": 0.73852497,
      "learning_rate": 2.7247806649867835e-07,
      "loss": 0.75973737,
      "num_input_tokens_seen": 300242915,
      "step": 13920,
      "time_per_iteration": 2.5117666721343994
    },
    {
      "auxiliary_loss_clip": 0.01083805,
      "auxiliary_loss_mlp": 0.01030657,
      "balance_loss_clip": 1.03408778,
      "balance_loss_mlp": 1.01815772,
      "epoch": 0.8369758003908011,
      "flos": 21835555511040.0,
      "grad_norm": 1.960167735947292,
      "language_loss": 0.6860438,
      "learning_rate": 2.722818488237566e-07,
      "loss": 0.70718837,
      "num_input_tokens_seen": 300261905,
      "step": 13921,
      "time_per_iteration": 2.547477960586548
    },
    {
      "auxiliary_loss_clip": 0.01097502,
      "auxiliary_loss_mlp": 0.0103541,
      "balance_loss_clip": 1.03624272,
      "balance_loss_mlp": 1.0238471,
      "epoch": 0.8370359236434691,
      "flos": 21719312121600.0,
      "grad_norm": 1.8894399067133132,
      "language_loss": 0.85623074,
      "learning_rate": 2.720856966640801e-07,
      "loss": 0.87755984,
      "num_input_tokens_seen": 300281145,
      "step": 13922,
      "time_per_iteration": 2.4766440391540527
    },
    {
      "auxiliary_loss_clip": 0.01061632,
      "auxiliary_loss_mlp": 0.00782387,
      "balance_loss_clip": 1.03312671,
      "balance_loss_mlp": 1.00959134,
      "epoch": 0.837096046896137,
      "flos": 23148880485120.0,
      "grad_norm": 1.7009172480493002,
      "language_loss": 0.71826839,
      "learning_rate": 2.71889610027088e-07,
      "loss": 0.73670852,
      "num_input_tokens_seen": 300301610,
      "step": 13923,
      "time_per_iteration": 2.585829973220825
    },
    {
      "auxiliary_loss_clip": 0.01078167,
      "auxiliary_loss_mlp": 0.01029791,
      "balance_loss_clip": 1.0351634,
      "balance_loss_mlp": 1.01702404,
      "epoch": 0.8371561701488051,
      "flos": 24492872695680.0,
      "grad_norm": 1.8693050149404153,
      "language_loss": 0.76247787,
      "learning_rate": 2.7169358892021433e-07,
      "loss": 0.78355747,
      "num_input_tokens_seen": 300319420,
      "step": 13924,
      "time_per_iteration": 2.540253162384033
    },
    {
      "auxiliary_loss_clip": 0.01081174,
      "auxiliary_loss_mlp": 0.01024728,
      "balance_loss_clip": 1.03393817,
      "balance_loss_mlp": 1.01284266,
      "epoch": 0.837216293401473,
      "flos": 29206723161600.0,
      "grad_norm": 1.4855306657387035,
      "language_loss": 0.64792436,
      "learning_rate": 2.7149763335089293e-07,
      "loss": 0.6689834,
      "num_input_tokens_seen": 300341325,
      "step": 13925,
      "time_per_iteration": 2.586538314819336
    },
    {
      "auxiliary_loss_clip": 0.01083067,
      "auxiliary_loss_mlp": 0.01031806,
      "balance_loss_clip": 1.03688991,
      "balance_loss_mlp": 1.01943815,
      "epoch": 0.837276416654141,
      "flos": 25265275781760.0,
      "grad_norm": 1.5332352944294216,
      "language_loss": 0.74442226,
      "learning_rate": 2.713017433265543e-07,
      "loss": 0.765571,
      "num_input_tokens_seen": 300361620,
      "step": 13926,
      "time_per_iteration": 2.5796051025390625
    },
    {
      "auxiliary_loss_clip": 0.01094462,
      "auxiliary_loss_mlp": 0.0103297,
      "balance_loss_clip": 1.03693533,
      "balance_loss_mlp": 1.02055407,
      "epoch": 0.837336539906809,
      "flos": 13882024656000.0,
      "grad_norm": 1.7498484354281776,
      "language_loss": 0.71085656,
      "learning_rate": 2.711059188546274e-07,
      "loss": 0.73213089,
      "num_input_tokens_seen": 300378675,
      "step": 13927,
      "time_per_iteration": 2.4699366092681885
    },
    {
      "auxiliary_loss_clip": 0.01008536,
      "auxiliary_loss_mlp": 0.01002315,
      "balance_loss_clip": 1.00784945,
      "balance_loss_mlp": 1.00118244,
      "epoch": 0.8373966631594769,
      "flos": 68870599044480.0,
      "grad_norm": 0.709560756023474,
      "language_loss": 0.58822608,
      "learning_rate": 2.7091015994253695e-07,
      "loss": 0.6083346,
      "num_input_tokens_seen": 300449740,
      "step": 13928,
      "time_per_iteration": 3.261688232421875
    },
    {
      "auxiliary_loss_clip": 0.01068119,
      "auxiliary_loss_mlp": 0.0103781,
      "balance_loss_clip": 1.03757548,
      "balance_loss_mlp": 1.02457142,
      "epoch": 0.8374567864121449,
      "flos": 20448972748800.0,
      "grad_norm": 1.7361076025926911,
      "language_loss": 0.6992979,
      "learning_rate": 2.707144665977068e-07,
      "loss": 0.72035718,
      "num_input_tokens_seen": 300470000,
      "step": 13929,
      "time_per_iteration": 2.5716052055358887
    },
    {
      "auxiliary_loss_clip": 0.01094615,
      "auxiliary_loss_mlp": 0.01027518,
      "balance_loss_clip": 1.03534102,
      "balance_loss_mlp": 1.01505494,
      "epoch": 0.8375169096648128,
      "flos": 41904197101440.0,
      "grad_norm": 1.4357393147939819,
      "language_loss": 0.66971534,
      "learning_rate": 2.705188388275574e-07,
      "loss": 0.69093668,
      "num_input_tokens_seen": 300494975,
      "step": 13930,
      "time_per_iteration": 2.6938843727111816
    },
    {
      "auxiliary_loss_clip": 0.01063236,
      "auxiliary_loss_mlp": 0.01027354,
      "balance_loss_clip": 1.03734064,
      "balance_loss_mlp": 1.01552224,
      "epoch": 0.8375770329174809,
      "flos": 20009354192640.0,
      "grad_norm": 1.6162014239351166,
      "language_loss": 0.71177053,
      "learning_rate": 2.703232766395067e-07,
      "loss": 0.73267639,
      "num_input_tokens_seen": 300513175,
      "step": 13931,
      "time_per_iteration": 2.5953662395477295
    },
    {
      "auxiliary_loss_clip": 0.01068282,
      "auxiliary_loss_mlp": 0.01032298,
      "balance_loss_clip": 1.03158212,
      "balance_loss_mlp": 1.0201385,
      "epoch": 0.8376371561701488,
      "flos": 22783597125120.0,
      "grad_norm": 1.58174218051321,
      "language_loss": 0.71856117,
      "learning_rate": 2.701277800409705e-07,
      "loss": 0.73956692,
      "num_input_tokens_seen": 300533770,
      "step": 13932,
      "time_per_iteration": 2.554008960723877
    },
    {
      "auxiliary_loss_clip": 0.0103908,
      "auxiliary_loss_mlp": 0.0103113,
      "balance_loss_clip": 1.03232956,
      "balance_loss_mlp": 1.02050209,
      "epoch": 0.8376972794228168,
      "flos": 23914459987200.0,
      "grad_norm": 2.0348648832372316,
      "language_loss": 0.66496444,
      "learning_rate": 2.699323490393628e-07,
      "loss": 0.6856665,
      "num_input_tokens_seen": 300552995,
      "step": 13933,
      "time_per_iteration": 2.6921565532684326
    },
    {
      "auxiliary_loss_clip": 0.01074007,
      "auxiliary_loss_mlp": 0.01039179,
      "balance_loss_clip": 1.03422058,
      "balance_loss_mlp": 1.02747893,
      "epoch": 0.8377574026754847,
      "flos": 13734718980480.0,
      "grad_norm": 2.0740067893872185,
      "language_loss": 0.76653624,
      "learning_rate": 2.697369836420933e-07,
      "loss": 0.78766817,
      "num_input_tokens_seen": 300570275,
      "step": 13934,
      "time_per_iteration": 2.492398738861084
    },
    {
      "auxiliary_loss_clip": 0.01093979,
      "auxiliary_loss_mlp": 0.01027898,
      "balance_loss_clip": 1.03797972,
      "balance_loss_mlp": 1.01608396,
      "epoch": 0.8378175259281527,
      "flos": 21651333632640.0,
      "grad_norm": 1.5593046782831266,
      "language_loss": 0.77276808,
      "learning_rate": 2.6954168385657115e-07,
      "loss": 0.79398692,
      "num_input_tokens_seen": 300590875,
      "step": 13935,
      "time_per_iteration": 2.5162315368652344
    },
    {
      "auxiliary_loss_clip": 0.01063486,
      "auxiliary_loss_mlp": 0.01029147,
      "balance_loss_clip": 1.03644395,
      "balance_loss_mlp": 1.01746464,
      "epoch": 0.8378776491808206,
      "flos": 15448806973440.0,
      "grad_norm": 3.002655436567441,
      "language_loss": 0.55746806,
      "learning_rate": 2.6934644969020135e-07,
      "loss": 0.57839435,
      "num_input_tokens_seen": 300607490,
      "step": 13936,
      "time_per_iteration": 2.5022873878479004
    },
    {
      "auxiliary_loss_clip": 0.01087211,
      "auxiliary_loss_mlp": 0.01032981,
      "balance_loss_clip": 1.03261173,
      "balance_loss_mlp": 1.02069592,
      "epoch": 0.8379377724334887,
      "flos": 14720395069440.0,
      "grad_norm": 1.7398262323607545,
      "language_loss": 0.89500827,
      "learning_rate": 2.691512811503882e-07,
      "loss": 0.91621017,
      "num_input_tokens_seen": 300623635,
      "step": 13937,
      "time_per_iteration": 2.4685986042022705
    },
    {
      "auxiliary_loss_clip": 0.01095015,
      "auxiliary_loss_mlp": 0.01029308,
      "balance_loss_clip": 1.03565812,
      "balance_loss_mlp": 1.01747704,
      "epoch": 0.8379978956861566,
      "flos": 24535247765760.0,
      "grad_norm": 1.6732919989714432,
      "language_loss": 0.81909549,
      "learning_rate": 2.689561782445313e-07,
      "loss": 0.84033871,
      "num_input_tokens_seen": 300643835,
      "step": 13938,
      "time_per_iteration": 2.496149778366089
    },
    {
      "auxiliary_loss_clip": 0.01096508,
      "auxiliary_loss_mlp": 0.01030321,
      "balance_loss_clip": 1.03545737,
      "balance_loss_mlp": 1.01774406,
      "epoch": 0.8380580189388246,
      "flos": 18952611045120.0,
      "grad_norm": 1.7334849160631554,
      "language_loss": 0.70441115,
      "learning_rate": 2.6876114098002965e-07,
      "loss": 0.7256794,
      "num_input_tokens_seen": 300662500,
      "step": 13939,
      "time_per_iteration": 2.5023746490478516
    },
    {
      "auxiliary_loss_clip": 0.01074238,
      "auxiliary_loss_mlp": 0.01038421,
      "balance_loss_clip": 1.03484201,
      "balance_loss_mlp": 1.02502775,
      "epoch": 0.8381181421914926,
      "flos": 26540283922560.0,
      "grad_norm": 1.6437890385274052,
      "language_loss": 0.76120734,
      "learning_rate": 2.6856616936428e-07,
      "loss": 0.78233391,
      "num_input_tokens_seen": 300681480,
      "step": 13940,
      "time_per_iteration": 2.580244541168213
    },
    {
      "auxiliary_loss_clip": 0.01090319,
      "auxiliary_loss_mlp": 0.01033935,
      "balance_loss_clip": 1.03547549,
      "balance_loss_mlp": 1.02160907,
      "epoch": 0.8381782654441605,
      "flos": 23291481479040.0,
      "grad_norm": 1.9761393224990804,
      "language_loss": 0.76670063,
      "learning_rate": 2.6837126340467374e-07,
      "loss": 0.78794312,
      "num_input_tokens_seen": 300699165,
      "step": 13941,
      "time_per_iteration": 2.5179035663604736
    },
    {
      "auxiliary_loss_clip": 0.01062542,
      "auxiliary_loss_mlp": 0.01029597,
      "balance_loss_clip": 1.03330958,
      "balance_loss_mlp": 1.01706195,
      "epoch": 0.8382383886968285,
      "flos": 26758800311040.0,
      "grad_norm": 2.098510605216128,
      "language_loss": 0.73601764,
      "learning_rate": 2.6817642310860276e-07,
      "loss": 0.75693905,
      "num_input_tokens_seen": 300714615,
      "step": 13942,
      "time_per_iteration": 3.983585834503174
    },
    {
      "auxiliary_loss_clip": 0.01062537,
      "auxiliary_loss_mlp": 0.0103587,
      "balance_loss_clip": 1.03512776,
      "balance_loss_mlp": 1.02234006,
      "epoch": 0.8382985119494964,
      "flos": 26104544035200.0,
      "grad_norm": 1.4196172572646153,
      "language_loss": 0.79349184,
      "learning_rate": 2.679816484834554e-07,
      "loss": 0.81447589,
      "num_input_tokens_seen": 300734860,
      "step": 13943,
      "time_per_iteration": 3.9789185523986816
    },
    {
      "auxiliary_loss_clip": 0.01053736,
      "auxiliary_loss_mlp": 0.01030665,
      "balance_loss_clip": 1.03159308,
      "balance_loss_mlp": 1.01898861,
      "epoch": 0.8383586352021645,
      "flos": 16435129507200.0,
      "grad_norm": 1.9165464582437033,
      "language_loss": 0.84908515,
      "learning_rate": 2.6778693953661766e-07,
      "loss": 0.86992919,
      "num_input_tokens_seen": 300752735,
      "step": 13944,
      "time_per_iteration": 2.5898139476776123
    },
    {
      "auxiliary_loss_clip": 0.01017139,
      "auxiliary_loss_mlp": 0.00762318,
      "balance_loss_clip": 1.0065372,
      "balance_loss_mlp": 1.0033344,
      "epoch": 0.8384187584548324,
      "flos": 64195532288640.0,
      "grad_norm": 0.7029654488139638,
      "language_loss": 0.5023905,
      "learning_rate": 2.6759229627547263e-07,
      "loss": 0.52018511,
      "num_input_tokens_seen": 300820760,
      "step": 13945,
      "time_per_iteration": 3.2068533897399902
    },
    {
      "auxiliary_loss_clip": 0.01058623,
      "auxiliary_loss_mlp": 0.01030903,
      "balance_loss_clip": 1.0344497,
      "balance_loss_mlp": 1.01924479,
      "epoch": 0.8384788817075004,
      "flos": 22382905933440.0,
      "grad_norm": 1.5479795236038068,
      "language_loss": 0.65014857,
      "learning_rate": 2.673977187074017e-07,
      "loss": 0.67104387,
      "num_input_tokens_seen": 300840025,
      "step": 13946,
      "time_per_iteration": 2.631258010864258
    },
    {
      "auxiliary_loss_clip": 0.01054132,
      "auxiliary_loss_mlp": 0.0103326,
      "balance_loss_clip": 1.03125811,
      "balance_loss_mlp": 1.02062368,
      "epoch": 0.8385390049601683,
      "flos": 29496845312640.0,
      "grad_norm": 1.6923282852596198,
      "language_loss": 0.67612338,
      "learning_rate": 2.672032068397829e-07,
      "loss": 0.69699728,
      "num_input_tokens_seen": 300860380,
      "step": 13947,
      "time_per_iteration": 4.000232934951782
    },
    {
      "auxiliary_loss_clip": 0.01079025,
      "auxiliary_loss_mlp": 0.01033282,
      "balance_loss_clip": 1.03455901,
      "balance_loss_mlp": 1.02030659,
      "epoch": 0.8385991282128363,
      "flos": 32707797799680.0,
      "grad_norm": 1.5194661912900052,
      "language_loss": 0.69700718,
      "learning_rate": 2.6700876067999176e-07,
      "loss": 0.71813023,
      "num_input_tokens_seen": 300881895,
      "step": 13948,
      "time_per_iteration": 2.6137001514434814
    },
    {
      "auxiliary_loss_clip": 0.01078556,
      "auxiliary_loss_mlp": 0.01032407,
      "balance_loss_clip": 1.03425431,
      "balance_loss_mlp": 1.02185082,
      "epoch": 0.8386592514655042,
      "flos": 25441022050560.0,
      "grad_norm": 1.7931961525172722,
      "language_loss": 0.84667385,
      "learning_rate": 2.6681438023540194e-07,
      "loss": 0.86778349,
      "num_input_tokens_seen": 300901575,
      "step": 13949,
      "time_per_iteration": 2.560199737548828
    },
    {
      "auxiliary_loss_clip": 0.01076927,
      "auxiliary_loss_mlp": 0.01025536,
      "balance_loss_clip": 1.03681612,
      "balance_loss_mlp": 1.01393104,
      "epoch": 0.8387193747181723,
      "flos": 22015898720640.0,
      "grad_norm": 3.0973921407735823,
      "language_loss": 0.70326722,
      "learning_rate": 2.66620065513385e-07,
      "loss": 0.7242918,
      "num_input_tokens_seen": 300919735,
      "step": 13950,
      "time_per_iteration": 2.521230459213257
    },
    {
      "auxiliary_loss_clip": 0.01093724,
      "auxiliary_loss_mlp": 0.01027066,
      "balance_loss_clip": 1.03604341,
      "balance_loss_mlp": 1.01494241,
      "epoch": 0.8387794979708402,
      "flos": 18150223080960.0,
      "grad_norm": 2.270612636817188,
      "language_loss": 0.64484739,
      "learning_rate": 2.6642581652130913e-07,
      "loss": 0.66605532,
      "num_input_tokens_seen": 300939150,
      "step": 13951,
      "time_per_iteration": 2.491940975189209
    },
    {
      "auxiliary_loss_clip": 0.01094124,
      "auxiliary_loss_mlp": 0.010292,
      "balance_loss_clip": 1.03597164,
      "balance_loss_mlp": 1.0177083,
      "epoch": 0.8388396212235082,
      "flos": 25411216740480.0,
      "grad_norm": 3.0508049334135254,
      "language_loss": 0.70184088,
      "learning_rate": 2.662316332665393e-07,
      "loss": 0.72307414,
      "num_input_tokens_seen": 300959730,
      "step": 13952,
      "time_per_iteration": 3.932929277420044
    },
    {
      "auxiliary_loss_clip": 0.01090912,
      "auxiliary_loss_mlp": 0.0102734,
      "balance_loss_clip": 1.035182,
      "balance_loss_mlp": 1.01589584,
      "epoch": 0.8388997444761762,
      "flos": 22273055164800.0,
      "grad_norm": 1.9667533939445316,
      "language_loss": 0.72458172,
      "learning_rate": 2.6603751575643987e-07,
      "loss": 0.7457642,
      "num_input_tokens_seen": 300976120,
      "step": 13953,
      "time_per_iteration": 2.4690165519714355
    },
    {
      "auxiliary_loss_clip": 0.01039961,
      "auxiliary_loss_mlp": 0.01030798,
      "balance_loss_clip": 1.03356886,
      "balance_loss_mlp": 1.01727414,
      "epoch": 0.8389598677288441,
      "flos": 19573219255680.0,
      "grad_norm": 2.0272414186837606,
      "language_loss": 0.68074054,
      "learning_rate": 2.6584346399837176e-07,
      "loss": 0.70144808,
      "num_input_tokens_seen": 300995080,
      "step": 13954,
      "time_per_iteration": 2.614821195602417
    },
    {
      "auxiliary_loss_clip": 0.01083701,
      "auxiliary_loss_mlp": 0.01031049,
      "balance_loss_clip": 1.03689027,
      "balance_loss_mlp": 1.01990318,
      "epoch": 0.8390199909815121,
      "flos": 17384715406080.0,
      "grad_norm": 1.7864248424906386,
      "language_loss": 0.73158157,
      "learning_rate": 2.656494779996932e-07,
      "loss": 0.75272906,
      "num_input_tokens_seen": 301012920,
      "step": 13955,
      "time_per_iteration": 2.5112876892089844
    },
    {
      "auxiliary_loss_clip": 0.01039473,
      "auxiliary_loss_mlp": 0.01030339,
      "balance_loss_clip": 1.03133523,
      "balance_loss_mlp": 1.01797748,
      "epoch": 0.83908011423418,
      "flos": 24639639667200.0,
      "grad_norm": 2.345639350732659,
      "language_loss": 0.66278088,
      "learning_rate": 2.6545555776775995e-07,
      "loss": 0.68347907,
      "num_input_tokens_seen": 301028875,
      "step": 13956,
      "time_per_iteration": 2.6299521923065186
    },
    {
      "auxiliary_loss_clip": 0.01094636,
      "auxiliary_loss_mlp": 0.01033009,
      "balance_loss_clip": 1.03535938,
      "balance_loss_mlp": 1.02067065,
      "epoch": 0.8391402374868481,
      "flos": 24718356322560.0,
      "grad_norm": 1.6916541217753913,
      "language_loss": 0.79714274,
      "learning_rate": 2.6526170330992667e-07,
      "loss": 0.81841922,
      "num_input_tokens_seen": 301050115,
      "step": 13957,
      "time_per_iteration": 2.516850709915161
    },
    {
      "auxiliary_loss_clip": 0.00993873,
      "auxiliary_loss_mlp": 0.01008596,
      "balance_loss_clip": 1.02593207,
      "balance_loss_mlp": 1.00723076,
      "epoch": 0.839200360739516,
      "flos": 56871695784960.0,
      "grad_norm": 0.7634375852212392,
      "language_loss": 0.53354728,
      "learning_rate": 2.6506791463354283e-07,
      "loss": 0.55357194,
      "num_input_tokens_seen": 301114155,
      "step": 13958,
      "time_per_iteration": 3.3439841270446777
    },
    {
      "auxiliary_loss_clip": 0.01093186,
      "auxiliary_loss_mlp": 0.01030427,
      "balance_loss_clip": 1.03598833,
      "balance_loss_mlp": 1.01810694,
      "epoch": 0.839260483992184,
      "flos": 18332792933760.0,
      "grad_norm": 1.810273997402119,
      "language_loss": 0.73224294,
      "learning_rate": 2.648741917459574e-07,
      "loss": 0.75347912,
      "num_input_tokens_seen": 301133150,
      "step": 13959,
      "time_per_iteration": 2.50077486038208
    },
    {
      "auxiliary_loss_clip": 0.01068807,
      "auxiliary_loss_mlp": 0.01024245,
      "balance_loss_clip": 1.03745222,
      "balance_loss_mlp": 1.01326549,
      "epoch": 0.8393206072448519,
      "flos": 27087921653760.0,
      "grad_norm": 1.778078653647245,
      "language_loss": 0.55045342,
      "learning_rate": 2.646805346545169e-07,
      "loss": 0.57138395,
      "num_input_tokens_seen": 301153600,
      "step": 13960,
      "time_per_iteration": 2.621269941329956
    },
    {
      "auxiliary_loss_clip": 0.01003728,
      "auxiliary_loss_mlp": 0.01002952,
      "balance_loss_clip": 1.00954449,
      "balance_loss_mlp": 1.00183129,
      "epoch": 0.8393807304975199,
      "flos": 61521192057600.0,
      "grad_norm": 0.773854739093631,
      "language_loss": 0.60693908,
      "learning_rate": 2.6448694336656397e-07,
      "loss": 0.62700588,
      "num_input_tokens_seen": 301214335,
      "step": 13961,
      "time_per_iteration": 3.245011806488037
    },
    {
      "auxiliary_loss_clip": 0.01049391,
      "auxiliary_loss_mlp": 0.01036439,
      "balance_loss_clip": 1.03092337,
      "balance_loss_mlp": 1.02413058,
      "epoch": 0.8394408537501878,
      "flos": 14894848448640.0,
      "grad_norm": 2.013106193141199,
      "language_loss": 0.68211877,
      "learning_rate": 2.642934178894405e-07,
      "loss": 0.70297706,
      "num_input_tokens_seen": 301228960,
      "step": 13962,
      "time_per_iteration": 2.5503060817718506
    },
    {
      "auxiliary_loss_clip": 0.01072379,
      "auxiliary_loss_mlp": 0.01030834,
      "balance_loss_clip": 1.03278184,
      "balance_loss_mlp": 1.01891851,
      "epoch": 0.8395009770028559,
      "flos": 17412186332160.0,
      "grad_norm": 1.8004332406285057,
      "language_loss": 0.73236167,
      "learning_rate": 2.640999582304841e-07,
      "loss": 0.75339389,
      "num_input_tokens_seen": 301245875,
      "step": 13963,
      "time_per_iteration": 2.5347201824188232
    },
    {
      "auxiliary_loss_clip": 0.0108082,
      "auxiliary_loss_mlp": 0.01034277,
      "balance_loss_clip": 1.03283024,
      "balance_loss_mlp": 1.02285051,
      "epoch": 0.8395611002555238,
      "flos": 27924747782400.0,
      "grad_norm": 1.5284985692084379,
      "language_loss": 0.76600415,
      "learning_rate": 2.6390656439703173e-07,
      "loss": 0.78715515,
      "num_input_tokens_seen": 301265550,
      "step": 13964,
      "time_per_iteration": 2.5668647289276123
    },
    {
      "auxiliary_loss_clip": 0.01085647,
      "auxiliary_loss_mlp": 0.01033763,
      "balance_loss_clip": 1.03607941,
      "balance_loss_mlp": 1.02073944,
      "epoch": 0.8396212235081918,
      "flos": 11100922225920.0,
      "grad_norm": 2.418753717242034,
      "language_loss": 0.78081113,
      "learning_rate": 2.637132363964161e-07,
      "loss": 0.80200529,
      "num_input_tokens_seen": 301282035,
      "step": 13965,
      "time_per_iteration": 2.5139474868774414
    },
    {
      "auxiliary_loss_clip": 0.01085956,
      "auxiliary_loss_mlp": 0.01031783,
      "balance_loss_clip": 1.03328383,
      "balance_loss_mlp": 1.0206728,
      "epoch": 0.8396813467608598,
      "flos": 35735641729920.0,
      "grad_norm": 1.6206627418033916,
      "language_loss": 0.65721214,
      "learning_rate": 2.635199742359684e-07,
      "loss": 0.67838955,
      "num_input_tokens_seen": 301305210,
      "step": 13966,
      "time_per_iteration": 2.618116617202759
    },
    {
      "auxiliary_loss_clip": 0.01081813,
      "auxiliary_loss_mlp": 0.01029632,
      "balance_loss_clip": 1.03488588,
      "balance_loss_mlp": 1.01802099,
      "epoch": 0.8397414700135277,
      "flos": 26176724415360.0,
      "grad_norm": 1.6745473593323255,
      "language_loss": 0.74263608,
      "learning_rate": 2.633267779230177e-07,
      "loss": 0.76375049,
      "num_input_tokens_seen": 301324885,
      "step": 13967,
      "time_per_iteration": 2.589142322540283
    },
    {
      "auxiliary_loss_clip": 0.01077357,
      "auxiliary_loss_mlp": 0.01029236,
      "balance_loss_clip": 1.03583789,
      "balance_loss_mlp": 1.01732087,
      "epoch": 0.8398015932661957,
      "flos": 18333116156160.0,
      "grad_norm": 2.119495905415382,
      "language_loss": 0.82924747,
      "learning_rate": 2.6313364746488974e-07,
      "loss": 0.85031343,
      "num_input_tokens_seen": 301343070,
      "step": 13968,
      "time_per_iteration": 2.5202994346618652
    },
    {
      "auxiliary_loss_clip": 0.01085164,
      "auxiliary_loss_mlp": 0.01030255,
      "balance_loss_clip": 1.03675699,
      "balance_loss_mlp": 1.01857305,
      "epoch": 0.8398617165188637,
      "flos": 17379507934080.0,
      "grad_norm": 1.9470270704890418,
      "language_loss": 0.77535272,
      "learning_rate": 2.629405828689075e-07,
      "loss": 0.79650688,
      "num_input_tokens_seen": 301359280,
      "step": 13969,
      "time_per_iteration": 2.5028836727142334
    },
    {
      "auxiliary_loss_clip": 0.01085424,
      "auxiliary_loss_mlp": 0.01026436,
      "balance_loss_clip": 1.03443527,
      "balance_loss_mlp": 1.01371026,
      "epoch": 0.8399218397715317,
      "flos": 22929681738240.0,
      "grad_norm": 2.017098627491471,
      "language_loss": 0.77053446,
      "learning_rate": 2.627475841423923e-07,
      "loss": 0.7916531,
      "num_input_tokens_seen": 301376465,
      "step": 13970,
      "time_per_iteration": 2.571639060974121
    },
    {
      "auxiliary_loss_clip": 0.01081698,
      "auxiliary_loss_mlp": 0.01036773,
      "balance_loss_clip": 1.03393054,
      "balance_loss_mlp": 1.02491724,
      "epoch": 0.8399819630241996,
      "flos": 23149562843520.0,
      "grad_norm": 1.8289167374169204,
      "language_loss": 0.72192836,
      "learning_rate": 2.625546512926633e-07,
      "loss": 0.74311304,
      "num_input_tokens_seen": 301396000,
      "step": 13971,
      "time_per_iteration": 2.5542821884155273
    },
    {
      "auxiliary_loss_clip": 0.01082372,
      "auxiliary_loss_mlp": 0.0103033,
      "balance_loss_clip": 1.03391576,
      "balance_loss_mlp": 1.01792026,
      "epoch": 0.8400420862768676,
      "flos": 16397423205120.0,
      "grad_norm": 1.9133947573434664,
      "language_loss": 0.77420157,
      "learning_rate": 2.623617843270358e-07,
      "loss": 0.79532856,
      "num_input_tokens_seen": 301413160,
      "step": 13972,
      "time_per_iteration": 2.4767229557037354
    },
    {
      "auxiliary_loss_clip": 0.01038683,
      "auxiliary_loss_mlp": 0.0103339,
      "balance_loss_clip": 1.03441715,
      "balance_loss_mlp": 1.02083683,
      "epoch": 0.8401022095295355,
      "flos": 21287486816640.0,
      "grad_norm": 1.3656360380535346,
      "language_loss": 0.68362391,
      "learning_rate": 2.6216898325282333e-07,
      "loss": 0.70434463,
      "num_input_tokens_seen": 301433325,
      "step": 13973,
      "time_per_iteration": 2.641472578048706
    },
    {
      "auxiliary_loss_clip": 0.01082863,
      "auxiliary_loss_mlp": 0.01027512,
      "balance_loss_clip": 1.03468347,
      "balance_loss_mlp": 1.01550186,
      "epoch": 0.8401623327822035,
      "flos": 17311313963520.0,
      "grad_norm": 1.8998949556439089,
      "language_loss": 0.78131318,
      "learning_rate": 2.619762480773382e-07,
      "loss": 0.80241698,
      "num_input_tokens_seen": 301450265,
      "step": 13974,
      "time_per_iteration": 2.5020694732666016
    },
    {
      "auxiliary_loss_clip": 0.01087934,
      "auxiliary_loss_mlp": 0.01027188,
      "balance_loss_clip": 1.03528869,
      "balance_loss_mlp": 1.01534462,
      "epoch": 0.8402224560348714,
      "flos": 22236677665920.0,
      "grad_norm": 1.5417594706075974,
      "language_loss": 0.72944021,
      "learning_rate": 2.617835788078868e-07,
      "loss": 0.7505914,
      "num_input_tokens_seen": 301470760,
      "step": 13975,
      "time_per_iteration": 2.521355152130127
    },
    {
      "auxiliary_loss_clip": 0.01082701,
      "auxiliary_loss_mlp": 0.01031728,
      "balance_loss_clip": 1.03402424,
      "balance_loss_mlp": 1.01937795,
      "epoch": 0.8402825792875395,
      "flos": 20229953569920.0,
      "grad_norm": 1.6902734931603942,
      "language_loss": 0.72679693,
      "learning_rate": 2.6159097545177645e-07,
      "loss": 0.74794126,
      "num_input_tokens_seen": 301489425,
      "step": 13976,
      "time_per_iteration": 2.4927144050598145
    },
    {
      "auxiliary_loss_clip": 0.01100953,
      "auxiliary_loss_mlp": 0.00785152,
      "balance_loss_clip": 1.03382206,
      "balance_loss_mlp": 1.01363933,
      "epoch": 0.8403427025402074,
      "flos": 23289973107840.0,
      "grad_norm": 1.7799803976507613,
      "language_loss": 0.71813053,
      "learning_rate": 2.61398438016311e-07,
      "loss": 0.73699158,
      "num_input_tokens_seen": 301508885,
      "step": 13977,
      "time_per_iteration": 2.475717306137085
    },
    {
      "auxiliary_loss_clip": 0.01090377,
      "auxiliary_loss_mlp": 0.01028727,
      "balance_loss_clip": 1.03180933,
      "balance_loss_mlp": 1.01714611,
      "epoch": 0.8404028257928754,
      "flos": 32675586278400.0,
      "grad_norm": 1.606603836356689,
      "language_loss": 0.68375093,
      "learning_rate": 2.6120596650879043e-07,
      "loss": 0.70494199,
      "num_input_tokens_seen": 301533780,
      "step": 13978,
      "time_per_iteration": 2.6183300018310547
    },
    {
      "auxiliary_loss_clip": 0.01068185,
      "auxiliary_loss_mlp": 0.01030248,
      "balance_loss_clip": 1.03433323,
      "balance_loss_mlp": 1.01850593,
      "epoch": 0.8404629490455434,
      "flos": 16180522928640.0,
      "grad_norm": 1.8049160861709796,
      "language_loss": 0.7800808,
      "learning_rate": 2.610135609365145e-07,
      "loss": 0.80106515,
      "num_input_tokens_seen": 301551775,
      "step": 13979,
      "time_per_iteration": 2.540816307067871
    },
    {
      "auxiliary_loss_clip": 0.01088555,
      "auxiliary_loss_mlp": 0.01028863,
      "balance_loss_clip": 1.03693974,
      "balance_loss_mlp": 1.01704335,
      "epoch": 0.8405230722982113,
      "flos": 15194451790080.0,
      "grad_norm": 2.2874958593142316,
      "language_loss": 0.78039193,
      "learning_rate": 2.60821221306778e-07,
      "loss": 0.80156612,
      "num_input_tokens_seen": 301570495,
      "step": 13980,
      "time_per_iteration": 2.458989381790161
    },
    {
      "auxiliary_loss_clip": 0.01068566,
      "auxiliary_loss_mlp": 0.01027334,
      "balance_loss_clip": 1.03439975,
      "balance_loss_mlp": 1.01623607,
      "epoch": 0.8405831955508793,
      "flos": 27812418975360.0,
      "grad_norm": 1.728087695129275,
      "language_loss": 0.86739993,
      "learning_rate": 2.606289476268757e-07,
      "loss": 0.88835895,
      "num_input_tokens_seen": 301591705,
      "step": 13981,
      "time_per_iteration": 3.987117290496826
    },
    {
      "auxiliary_loss_clip": 0.01093165,
      "auxiliary_loss_mlp": 0.01031072,
      "balance_loss_clip": 1.03544235,
      "balance_loss_mlp": 1.01940739,
      "epoch": 0.8406433188035473,
      "flos": 23769452782080.0,
      "grad_norm": 1.771615046201088,
      "language_loss": 0.67289615,
      "learning_rate": 2.6043673990409745e-07,
      "loss": 0.69413853,
      "num_input_tokens_seen": 301611670,
      "step": 13982,
      "time_per_iteration": 3.9059979915618896
    },
    {
      "auxiliary_loss_clip": 0.01056478,
      "auxiliary_loss_mlp": 0.01034489,
      "balance_loss_clip": 1.03617597,
      "balance_loss_mlp": 1.02078009,
      "epoch": 0.8407034420562153,
      "flos": 29205681667200.0,
      "grad_norm": 2.537158398339885,
      "language_loss": 0.68061149,
      "learning_rate": 2.602445981457324e-07,
      "loss": 0.70152116,
      "num_input_tokens_seen": 301632540,
      "step": 13983,
      "time_per_iteration": 2.6372921466827393
    },
    {
      "auxiliary_loss_clip": 0.01064651,
      "auxiliary_loss_mlp": 0.01032084,
      "balance_loss_clip": 1.02943587,
      "balance_loss_mlp": 1.01949549,
      "epoch": 0.8407635653088832,
      "flos": 26360084367360.0,
      "grad_norm": 1.8411722007302604,
      "language_loss": 0.79041213,
      "learning_rate": 2.6005252235906684e-07,
      "loss": 0.81137943,
      "num_input_tokens_seen": 301651480,
      "step": 13984,
      "time_per_iteration": 2.604414701461792
    },
    {
      "auxiliary_loss_clip": 0.01089281,
      "auxiliary_loss_mlp": 0.01029984,
      "balance_loss_clip": 1.03198361,
      "balance_loss_mlp": 1.01797962,
      "epoch": 0.8408236885615512,
      "flos": 21468799693440.0,
      "grad_norm": 2.7126331163489503,
      "language_loss": 0.60068929,
      "learning_rate": 2.598605125513842e-07,
      "loss": 0.62188196,
      "num_input_tokens_seen": 301670010,
      "step": 13985,
      "time_per_iteration": 3.874861001968384
    },
    {
      "auxiliary_loss_clip": 0.01062576,
      "auxiliary_loss_mlp": 0.01028111,
      "balance_loss_clip": 1.03302062,
      "balance_loss_mlp": 1.01551032,
      "epoch": 0.8408838118142191,
      "flos": 22963724853120.0,
      "grad_norm": 2.383817988528982,
      "language_loss": 0.8203634,
      "learning_rate": 2.5966856872996467e-07,
      "loss": 0.84127033,
      "num_input_tokens_seen": 301689785,
      "step": 13986,
      "time_per_iteration": 2.6269054412841797
    },
    {
      "auxiliary_loss_clip": 0.01082853,
      "auxiliary_loss_mlp": 0.00782968,
      "balance_loss_clip": 1.0376066,
      "balance_loss_mlp": 1.01029432,
      "epoch": 0.8409439350668871,
      "flos": 26800026145920.0,
      "grad_norm": 1.4359836520669882,
      "language_loss": 0.65803772,
      "learning_rate": 2.5947669090208755e-07,
      "loss": 0.676696,
      "num_input_tokens_seen": 301712225,
      "step": 13987,
      "time_per_iteration": 2.671391725540161
    },
    {
      "auxiliary_loss_clip": 0.01103273,
      "auxiliary_loss_mlp": 0.00784682,
      "balance_loss_clip": 1.03514886,
      "balance_loss_mlp": 1.01276839,
      "epoch": 0.841004058319555,
      "flos": 26578672583040.0,
      "grad_norm": 1.9435432036470026,
      "language_loss": 0.67448819,
      "learning_rate": 2.5928487907502906e-07,
      "loss": 0.69336772,
      "num_input_tokens_seen": 301730955,
      "step": 13988,
      "time_per_iteration": 2.538529634475708
    },
    {
      "auxiliary_loss_clip": 0.01093438,
      "auxiliary_loss_mlp": 0.010345,
      "balance_loss_clip": 1.03819144,
      "balance_loss_mlp": 1.02176225,
      "epoch": 0.8410641815722231,
      "flos": 14501878680960.0,
      "grad_norm": 2.149081050016251,
      "language_loss": 0.80806673,
      "learning_rate": 2.590931332560622e-07,
      "loss": 0.82934606,
      "num_input_tokens_seen": 301746930,
      "step": 13989,
      "time_per_iteration": 2.53222393989563
    },
    {
      "auxiliary_loss_clip": 0.01091207,
      "auxiliary_loss_mlp": 0.01027896,
      "balance_loss_clip": 1.03337288,
      "balance_loss_mlp": 1.01594567,
      "epoch": 0.841124304824891,
      "flos": 29166682475520.0,
      "grad_norm": 1.6713849724804943,
      "language_loss": 0.75513458,
      "learning_rate": 2.5890145345245826e-07,
      "loss": 0.77632564,
      "num_input_tokens_seen": 301766945,
      "step": 13990,
      "time_per_iteration": 2.5384159088134766
    },
    {
      "auxiliary_loss_clip": 0.01086121,
      "auxiliary_loss_mlp": 0.01030257,
      "balance_loss_clip": 1.03212571,
      "balance_loss_mlp": 1.01887822,
      "epoch": 0.841184428077559,
      "flos": 22412028885120.0,
      "grad_norm": 1.5709025546696422,
      "language_loss": 0.8107233,
      "learning_rate": 2.5870983967148597e-07,
      "loss": 0.83188713,
      "num_input_tokens_seen": 301785460,
      "step": 13991,
      "time_per_iteration": 3.9016592502593994
    },
    {
      "auxiliary_loss_clip": 0.01064524,
      "auxiliary_loss_mlp": 0.01027877,
      "balance_loss_clip": 1.03518748,
      "balance_loss_mlp": 1.01655817,
      "epoch": 0.841244551330227,
      "flos": 22962791099520.0,
      "grad_norm": 2.1036693991442967,
      "language_loss": 0.70923877,
      "learning_rate": 2.585182919204105e-07,
      "loss": 0.73016274,
      "num_input_tokens_seen": 301804180,
      "step": 13992,
      "time_per_iteration": 2.551100730895996
    },
    {
      "auxiliary_loss_clip": 0.01069445,
      "auxiliary_loss_mlp": 0.01025118,
      "balance_loss_clip": 1.03370273,
      "balance_loss_mlp": 1.01325631,
      "epoch": 0.8413046745828949,
      "flos": 21032736583680.0,
      "grad_norm": 1.5564021866570537,
      "language_loss": 0.7678535,
      "learning_rate": 2.583268102064959e-07,
      "loss": 0.78879911,
      "num_input_tokens_seen": 301823670,
      "step": 13993,
      "time_per_iteration": 2.5669591426849365
    },
    {
      "auxiliary_loss_clip": 0.01098042,
      "auxiliary_loss_mlp": 0.01034219,
      "balance_loss_clip": 1.03394711,
      "balance_loss_mlp": 1.02056968,
      "epoch": 0.841364797835563,
      "flos": 27052082858880.0,
      "grad_norm": 2.0745629802378973,
      "language_loss": 0.74213797,
      "learning_rate": 2.5813539453700393e-07,
      "loss": 0.76346058,
      "num_input_tokens_seen": 301845890,
      "step": 13994,
      "time_per_iteration": 2.521392583847046
    },
    {
      "auxiliary_loss_clip": 0.01091004,
      "auxiliary_loss_mlp": 0.01029074,
      "balance_loss_clip": 1.03465939,
      "balance_loss_mlp": 1.01829708,
      "epoch": 0.8414249210882309,
      "flos": 17895688329600.0,
      "grad_norm": 1.583370449752617,
      "language_loss": 0.59623295,
      "learning_rate": 2.5794404491919163e-07,
      "loss": 0.61743367,
      "num_input_tokens_seen": 301863985,
      "step": 13995,
      "time_per_iteration": 2.4629693031311035
    },
    {
      "auxiliary_loss_clip": 0.01092112,
      "auxiliary_loss_mlp": 0.01032449,
      "balance_loss_clip": 1.03461838,
      "balance_loss_mlp": 1.01974773,
      "epoch": 0.8414850443408989,
      "flos": 25441201618560.0,
      "grad_norm": 1.6909463511267706,
      "language_loss": 0.71822989,
      "learning_rate": 2.577527613603163e-07,
      "loss": 0.73947549,
      "num_input_tokens_seen": 301882765,
      "step": 13996,
      "time_per_iteration": 2.5143539905548096
    },
    {
      "auxiliary_loss_clip": 0.01078346,
      "auxiliary_loss_mlp": 0.01027263,
      "balance_loss_clip": 1.03374696,
      "balance_loss_mlp": 1.01547337,
      "epoch": 0.8415451675935668,
      "flos": 23220055284480.0,
      "grad_norm": 1.9078117280203712,
      "language_loss": 0.63923883,
      "learning_rate": 2.5756154386763017e-07,
      "loss": 0.66029489,
      "num_input_tokens_seen": 301902720,
      "step": 13997,
      "time_per_iteration": 2.5474588871002197
    },
    {
      "auxiliary_loss_clip": 0.01079045,
      "auxiliary_loss_mlp": 0.01031398,
      "balance_loss_clip": 1.03478193,
      "balance_loss_mlp": 1.01873231,
      "epoch": 0.8416052908462348,
      "flos": 18546496899840.0,
      "grad_norm": 2.962944862750636,
      "language_loss": 0.81865853,
      "learning_rate": 2.5737039244838565e-07,
      "loss": 0.83976293,
      "num_input_tokens_seen": 301921245,
      "step": 13998,
      "time_per_iteration": 2.513605833053589
    },
    {
      "auxiliary_loss_clip": 0.01094261,
      "auxiliary_loss_mlp": 0.00782892,
      "balance_loss_clip": 1.03592634,
      "balance_loss_mlp": 1.00878632,
      "epoch": 0.8416654140989027,
      "flos": 26105190480000.0,
      "grad_norm": 1.9519178945856186,
      "language_loss": 0.80074787,
      "learning_rate": 2.5717930710982984e-07,
      "loss": 0.81951934,
      "num_input_tokens_seen": 301942320,
      "step": 13999,
      "time_per_iteration": 2.545562982559204
    },
    {
      "auxiliary_loss_clip": 0.01093369,
      "auxiliary_loss_mlp": 0.01032148,
      "balance_loss_clip": 1.03459144,
      "balance_loss_mlp": 1.0189991,
      "epoch": 0.8417255373515707,
      "flos": 26433270328320.0,
      "grad_norm": 2.2467735973041174,
      "language_loss": 0.66802597,
      "learning_rate": 2.569882878592096e-07,
      "loss": 0.68928123,
      "num_input_tokens_seen": 301963110,
      "step": 14000,
      "time_per_iteration": 2.531163215637207
    },
    {
      "auxiliary_loss_clip": 0.01097972,
      "auxiliary_loss_mlp": 0.01030185,
      "balance_loss_clip": 1.03605437,
      "balance_loss_mlp": 1.0178225,
      "epoch": 0.8417856606042387,
      "flos": 24717745791360.0,
      "grad_norm": 1.4579318033897304,
      "language_loss": 0.79422134,
      "learning_rate": 2.5679733470376885e-07,
      "loss": 0.81550288,
      "num_input_tokens_seen": 301984915,
      "step": 14001,
      "time_per_iteration": 2.5269150733947754
    },
    {
      "auxiliary_loss_clip": 0.01040679,
      "auxiliary_loss_mlp": 0.01028356,
      "balance_loss_clip": 1.03467679,
      "balance_loss_mlp": 1.01754951,
      "epoch": 0.8418457838569067,
      "flos": 20850849089280.0,
      "grad_norm": 1.5895008759139129,
      "language_loss": 0.7858814,
      "learning_rate": 2.5660644765074703e-07,
      "loss": 0.80657172,
      "num_input_tokens_seen": 302004095,
      "step": 14002,
      "time_per_iteration": 2.6270828247070312
    },
    {
      "auxiliary_loss_clip": 0.01056785,
      "auxiliary_loss_mlp": 0.0078021,
      "balance_loss_clip": 1.03304863,
      "balance_loss_mlp": 1.00545847,
      "epoch": 0.8419059071095746,
      "flos": 28660629715200.0,
      "grad_norm": 1.4247947456424408,
      "language_loss": 0.78281999,
      "learning_rate": 2.5641562670738334e-07,
      "loss": 0.8011899,
      "num_input_tokens_seen": 302027250,
      "step": 14003,
      "time_per_iteration": 2.653022527694702
    },
    {
      "auxiliary_loss_clip": 0.01077054,
      "auxiliary_loss_mlp": 0.01028494,
      "balance_loss_clip": 1.03611934,
      "balance_loss_mlp": 1.01657319,
      "epoch": 0.8419660303622426,
      "flos": 21653596189440.0,
      "grad_norm": 1.698481793283611,
      "language_loss": 0.65640002,
      "learning_rate": 2.5622487188091436e-07,
      "loss": 0.67745548,
      "num_input_tokens_seen": 302046950,
      "step": 14004,
      "time_per_iteration": 2.540910005569458
    },
    {
      "auxiliary_loss_clip": 0.01093576,
      "auxiliary_loss_mlp": 0.01029151,
      "balance_loss_clip": 1.03463483,
      "balance_loss_mlp": 1.01649106,
      "epoch": 0.8420261536149106,
      "flos": 25301114576640.0,
      "grad_norm": 1.9232798663906334,
      "language_loss": 0.76172906,
      "learning_rate": 2.560341831785724e-07,
      "loss": 0.78295636,
      "num_input_tokens_seen": 302065470,
      "step": 14005,
      "time_per_iteration": 2.5526506900787354
    },
    {
      "auxiliary_loss_clip": 0.01066479,
      "auxiliary_loss_mlp": 0.00784196,
      "balance_loss_clip": 1.03173184,
      "balance_loss_mlp": 1.00902939,
      "epoch": 0.8420862768675785,
      "flos": 18763397176320.0,
      "grad_norm": 1.6247743149622682,
      "language_loss": 0.77698839,
      "learning_rate": 2.5584356060758906e-07,
      "loss": 0.79549515,
      "num_input_tokens_seen": 302083190,
      "step": 14006,
      "time_per_iteration": 2.565410852432251
    },
    {
      "auxiliary_loss_clip": 0.01092755,
      "auxiliary_loss_mlp": 0.01032139,
      "balance_loss_clip": 1.03482866,
      "balance_loss_mlp": 1.02033162,
      "epoch": 0.8421464001202466,
      "flos": 18328052338560.0,
      "grad_norm": 1.9569419504515364,
      "language_loss": 0.77069658,
      "learning_rate": 2.556530041751932e-07,
      "loss": 0.79194552,
      "num_input_tokens_seen": 302098820,
      "step": 14007,
      "time_per_iteration": 2.4760329723358154
    },
    {
      "auxiliary_loss_clip": 0.0108419,
      "auxiliary_loss_mlp": 0.01029829,
      "balance_loss_clip": 1.03531241,
      "balance_loss_mlp": 1.01743114,
      "epoch": 0.8422065233729145,
      "flos": 31537181560320.0,
      "grad_norm": 1.9739930871568838,
      "language_loss": 0.6567415,
      "learning_rate": 2.554625138886102e-07,
      "loss": 0.67788172,
      "num_input_tokens_seen": 302117075,
      "step": 14008,
      "time_per_iteration": 2.6162703037261963
    },
    {
      "auxiliary_loss_clip": 0.01019803,
      "auxiliary_loss_mlp": 0.01004488,
      "balance_loss_clip": 1.0065589,
      "balance_loss_mlp": 1.00331962,
      "epoch": 0.8422666466255825,
      "flos": 64298128510080.0,
      "grad_norm": 0.7118714307314838,
      "language_loss": 0.56933463,
      "learning_rate": 2.552720897550631e-07,
      "loss": 0.5895775,
      "num_input_tokens_seen": 302179735,
      "step": 14009,
      "time_per_iteration": 3.172305107116699
    },
    {
      "auxiliary_loss_clip": 0.01040635,
      "auxiliary_loss_mlp": 0.01031535,
      "balance_loss_clip": 1.03151071,
      "balance_loss_mlp": 1.02041328,
      "epoch": 0.8423267698782504,
      "flos": 24316731377280.0,
      "grad_norm": 1.395751042761149,
      "language_loss": 0.77996492,
      "learning_rate": 2.5508173178177304e-07,
      "loss": 0.80068666,
      "num_input_tokens_seen": 302202055,
      "step": 14010,
      "time_per_iteration": 2.6547160148620605
    },
    {
      "auxiliary_loss_clip": 0.01107418,
      "auxiliary_loss_mlp": 0.01034284,
      "balance_loss_clip": 1.03675103,
      "balance_loss_mlp": 1.02132034,
      "epoch": 0.8423868931309184,
      "flos": 18296092212480.0,
      "grad_norm": 1.954332654757083,
      "language_loss": 0.72234249,
      "learning_rate": 2.548914399759592e-07,
      "loss": 0.74375945,
      "num_input_tokens_seen": 302221360,
      "step": 14011,
      "time_per_iteration": 2.4374656677246094
    },
    {
      "auxiliary_loss_clip": 0.01087721,
      "auxiliary_loss_mlp": 0.01033673,
      "balance_loss_clip": 1.03464723,
      "balance_loss_mlp": 1.02218711,
      "epoch": 0.8424470163835863,
      "flos": 23550218121600.0,
      "grad_norm": 1.7702280373626171,
      "language_loss": 0.83927464,
      "learning_rate": 2.5470121434483636e-07,
      "loss": 0.86048859,
      "num_input_tokens_seen": 302240715,
      "step": 14012,
      "time_per_iteration": 2.522848606109619
    },
    {
      "auxiliary_loss_clip": 0.01096023,
      "auxiliary_loss_mlp": 0.01029485,
      "balance_loss_clip": 1.03311956,
      "balance_loss_mlp": 1.01934028,
      "epoch": 0.8425071396362543,
      "flos": 23769488695680.0,
      "grad_norm": 1.8967647532420013,
      "language_loss": 0.6783185,
      "learning_rate": 2.5451105489561884e-07,
      "loss": 0.69957352,
      "num_input_tokens_seen": 302260950,
      "step": 14013,
      "time_per_iteration": 2.471808433532715
    },
    {
      "auxiliary_loss_clip": 0.01108591,
      "auxiliary_loss_mlp": 0.01029497,
      "balance_loss_clip": 1.03620195,
      "balance_loss_mlp": 1.01728415,
      "epoch": 0.8425672628889223,
      "flos": 16178906816640.0,
      "grad_norm": 5.062969196461233,
      "language_loss": 0.78640604,
      "learning_rate": 2.5432096163551644e-07,
      "loss": 0.80778694,
      "num_input_tokens_seen": 302277500,
      "step": 14014,
      "time_per_iteration": 2.4459071159362793
    },
    {
      "auxiliary_loss_clip": 0.01073563,
      "auxiliary_loss_mlp": 0.00784769,
      "balance_loss_clip": 1.03401959,
      "balance_loss_mlp": 1.01200593,
      "epoch": 0.8426273861415903,
      "flos": 23149131880320.0,
      "grad_norm": 1.748955673945598,
      "language_loss": 0.67746276,
      "learning_rate": 2.5413093457173884e-07,
      "loss": 0.69604611,
      "num_input_tokens_seen": 302297930,
      "step": 14015,
      "time_per_iteration": 2.5660958290100098
    },
    {
      "auxiliary_loss_clip": 0.01103092,
      "auxiliary_loss_mlp": 0.01030047,
      "balance_loss_clip": 1.03508639,
      "balance_loss_mlp": 1.01726747,
      "epoch": 0.8426875093942582,
      "flos": 17457757712640.0,
      "grad_norm": 2.216353873602798,
      "language_loss": 0.76214898,
      "learning_rate": 2.5394097371149036e-07,
      "loss": 0.78348029,
      "num_input_tokens_seen": 302315735,
      "step": 14016,
      "time_per_iteration": 2.4463095664978027
    },
    {
      "auxiliary_loss_clip": 0.01080701,
      "auxiliary_loss_mlp": 0.01033382,
      "balance_loss_clip": 1.03428483,
      "balance_loss_mlp": 1.02158642,
      "epoch": 0.8427476326469262,
      "flos": 19640551299840.0,
      "grad_norm": 1.751734608495614,
      "language_loss": 0.79380459,
      "learning_rate": 2.5375107906197544e-07,
      "loss": 0.81494546,
      "num_input_tokens_seen": 302332790,
      "step": 14017,
      "time_per_iteration": 2.534085750579834
    },
    {
      "auxiliary_loss_clip": 0.01081577,
      "auxiliary_loss_mlp": 0.01030787,
      "balance_loss_clip": 1.03504086,
      "balance_loss_mlp": 1.01871085,
      "epoch": 0.8428077558995941,
      "flos": 11941160146560.0,
      "grad_norm": 1.9772314494697347,
      "language_loss": 0.62621224,
      "learning_rate": 2.5356125063039525e-07,
      "loss": 0.64733589,
      "num_input_tokens_seen": 302346490,
      "step": 14018,
      "time_per_iteration": 2.4998939037323
    },
    {
      "auxiliary_loss_clip": 0.01090614,
      "auxiliary_loss_mlp": 0.01029885,
      "balance_loss_clip": 1.03597367,
      "balance_loss_mlp": 1.01869774,
      "epoch": 0.8428678791522621,
      "flos": 10451729767680.0,
      "grad_norm": 1.8371737500607932,
      "language_loss": 0.79249346,
      "learning_rate": 2.5337148842394687e-07,
      "loss": 0.81369841,
      "num_input_tokens_seen": 302363235,
      "step": 14019,
      "time_per_iteration": 2.4686081409454346
    },
    {
      "auxiliary_loss_clip": 0.01062825,
      "auxiliary_loss_mlp": 0.01039431,
      "balance_loss_clip": 1.03201985,
      "balance_loss_mlp": 1.02476883,
      "epoch": 0.8429280024049302,
      "flos": 28767248259840.0,
      "grad_norm": 1.7723844039959444,
      "language_loss": 0.78175735,
      "learning_rate": 2.531817924498265e-07,
      "loss": 0.80277997,
      "num_input_tokens_seen": 302383270,
      "step": 14020,
      "time_per_iteration": 3.9817473888397217
    },
    {
      "auxiliary_loss_clip": 0.01089369,
      "auxiliary_loss_mlp": 0.01026378,
      "balance_loss_clip": 1.03532529,
      "balance_loss_mlp": 1.01473689,
      "epoch": 0.8429881256575981,
      "flos": 19537093152000.0,
      "grad_norm": 1.5355613316661234,
      "language_loss": 0.71187341,
      "learning_rate": 2.5299216271522805e-07,
      "loss": 0.73303092,
      "num_input_tokens_seen": 302401355,
      "step": 14021,
      "time_per_iteration": 3.8699326515197754
    },
    {
      "auxiliary_loss_clip": 0.0108313,
      "auxiliary_loss_mlp": 0.01035574,
      "balance_loss_clip": 1.03807497,
      "balance_loss_mlp": 1.0241245,
      "epoch": 0.8430482489102661,
      "flos": 24790931752320.0,
      "grad_norm": 1.8048883217073266,
      "language_loss": 0.69811922,
      "learning_rate": 2.5280259922734125e-07,
      "loss": 0.71930623,
      "num_input_tokens_seen": 302419515,
      "step": 14022,
      "time_per_iteration": 2.5486457347869873
    },
    {
      "auxiliary_loss_clip": 0.01048686,
      "auxiliary_loss_mlp": 0.01036429,
      "balance_loss_clip": 1.03414106,
      "balance_loss_mlp": 1.02338731,
      "epoch": 0.843108372162934,
      "flos": 21544248211200.0,
      "grad_norm": 1.9902692796211197,
      "language_loss": 0.72391009,
      "learning_rate": 2.526131019933553e-07,
      "loss": 0.74476129,
      "num_input_tokens_seen": 302438280,
      "step": 14023,
      "time_per_iteration": 2.629359483718872
    },
    {
      "auxiliary_loss_clip": 0.01094393,
      "auxiliary_loss_mlp": 0.01034424,
      "balance_loss_clip": 1.03677046,
      "balance_loss_mlp": 1.02166855,
      "epoch": 0.843168495415602,
      "flos": 24608792862720.0,
      "grad_norm": 1.537324373175969,
      "language_loss": 0.66834784,
      "learning_rate": 2.524236710204559e-07,
      "loss": 0.68963599,
      "num_input_tokens_seen": 302460860,
      "step": 14024,
      "time_per_iteration": 4.082241058349609
    },
    {
      "auxiliary_loss_clip": 0.01089786,
      "auxiliary_loss_mlp": 0.01030379,
      "balance_loss_clip": 1.0365355,
      "balance_loss_mlp": 1.01845229,
      "epoch": 0.8432286186682699,
      "flos": 15122738286720.0,
      "grad_norm": 1.7845263323708536,
      "language_loss": 0.81111634,
      "learning_rate": 2.522343063158261e-07,
      "loss": 0.83231801,
      "num_input_tokens_seen": 302476980,
      "step": 14025,
      "time_per_iteration": 2.476935863494873
    },
    {
      "auxiliary_loss_clip": 0.01087684,
      "auxiliary_loss_mlp": 0.01031265,
      "balance_loss_clip": 1.03390479,
      "balance_loss_mlp": 1.02139461,
      "epoch": 0.843288741920938,
      "flos": 20301882554880.0,
      "grad_norm": 1.4883081501714226,
      "language_loss": 0.77732062,
      "learning_rate": 2.5204500788664606e-07,
      "loss": 0.79851007,
      "num_input_tokens_seen": 302496380,
      "step": 14026,
      "time_per_iteration": 2.521749496459961
    },
    {
      "auxiliary_loss_clip": 0.01078573,
      "auxiliary_loss_mlp": 0.01034784,
      "balance_loss_clip": 1.03299022,
      "balance_loss_mlp": 1.02280331,
      "epoch": 0.8433488651736059,
      "flos": 23332096782720.0,
      "grad_norm": 1.3895842195320856,
      "language_loss": 0.82448435,
      "learning_rate": 2.518557757400945e-07,
      "loss": 0.84561783,
      "num_input_tokens_seen": 302516845,
      "step": 14027,
      "time_per_iteration": 2.5673789978027344
    },
    {
      "auxiliary_loss_clip": 0.01078259,
      "auxiliary_loss_mlp": 0.01030415,
      "balance_loss_clip": 1.03378427,
      "balance_loss_mlp": 1.01898289,
      "epoch": 0.8434089884262739,
      "flos": 39458105844480.0,
      "grad_norm": 1.8099156712001558,
      "language_loss": 0.56103534,
      "learning_rate": 2.5166660988334754e-07,
      "loss": 0.58212209,
      "num_input_tokens_seen": 302538865,
      "step": 14028,
      "time_per_iteration": 2.677839517593384
    },
    {
      "auxiliary_loss_clip": 0.01078483,
      "auxiliary_loss_mlp": 0.01027121,
      "balance_loss_clip": 1.03435373,
      "balance_loss_mlp": 1.015957,
      "epoch": 0.8434691116789418,
      "flos": 23768842250880.0,
      "grad_norm": 1.7237342874627883,
      "language_loss": 0.6350342,
      "learning_rate": 2.51477510323578e-07,
      "loss": 0.65609026,
      "num_input_tokens_seen": 302557970,
      "step": 14029,
      "time_per_iteration": 2.5209009647369385
    },
    {
      "auxiliary_loss_clip": 0.01099711,
      "auxiliary_loss_mlp": 0.0103116,
      "balance_loss_clip": 1.03546858,
      "balance_loss_mlp": 1.02019882,
      "epoch": 0.8435292349316098,
      "flos": 22671411972480.0,
      "grad_norm": 1.5717717025227453,
      "language_loss": 0.75454021,
      "learning_rate": 2.51288477067956e-07,
      "loss": 0.77584887,
      "num_input_tokens_seen": 302578915,
      "step": 14030,
      "time_per_iteration": 3.918175458908081
    },
    {
      "auxiliary_loss_clip": 0.01083598,
      "auxiliary_loss_mlp": 0.01032796,
      "balance_loss_clip": 1.03629732,
      "balance_loss_mlp": 1.02082109,
      "epoch": 0.8435893581842777,
      "flos": 18843622202880.0,
      "grad_norm": 1.6951699207400748,
      "language_loss": 0.82762295,
      "learning_rate": 2.510995101236502e-07,
      "loss": 0.84878689,
      "num_input_tokens_seen": 302596300,
      "step": 14031,
      "time_per_iteration": 2.512314558029175
    },
    {
      "auxiliary_loss_clip": 0.01078878,
      "auxiliary_loss_mlp": 0.01028744,
      "balance_loss_clip": 1.03276193,
      "balance_loss_mlp": 1.01758575,
      "epoch": 0.8436494814369457,
      "flos": 20704225772160.0,
      "grad_norm": 1.8272565967415897,
      "language_loss": 0.80015987,
      "learning_rate": 2.509106094978266e-07,
      "loss": 0.82123613,
      "num_input_tokens_seen": 302614975,
      "step": 14032,
      "time_per_iteration": 2.5044937133789062
    },
    {
      "auxiliary_loss_clip": 0.01068522,
      "auxiliary_loss_mlp": 0.01034235,
      "balance_loss_clip": 1.0336535,
      "balance_loss_mlp": 1.02047241,
      "epoch": 0.8437096046896138,
      "flos": 22674177319680.0,
      "grad_norm": 1.5749652101396596,
      "language_loss": 0.756598,
      "learning_rate": 2.507217751976478e-07,
      "loss": 0.77762556,
      "num_input_tokens_seen": 302636415,
      "step": 14033,
      "time_per_iteration": 2.5657293796539307
    },
    {
      "auxiliary_loss_clip": 0.01065545,
      "auxiliary_loss_mlp": 0.01035736,
      "balance_loss_clip": 1.03338599,
      "balance_loss_mlp": 1.02447701,
      "epoch": 0.8437697279422817,
      "flos": 16180127879040.0,
      "grad_norm": 1.7226434210672155,
      "language_loss": 0.838121,
      "learning_rate": 2.505330072302743e-07,
      "loss": 0.85913384,
      "num_input_tokens_seen": 302653605,
      "step": 14034,
      "time_per_iteration": 2.501718282699585
    },
    {
      "auxiliary_loss_clip": 0.01068517,
      "auxiliary_loss_mlp": 0.01030907,
      "balance_loss_clip": 1.03270578,
      "balance_loss_mlp": 1.01714396,
      "epoch": 0.8438298511949497,
      "flos": 28765847629440.0,
      "grad_norm": 1.4612808226077214,
      "language_loss": 0.78432345,
      "learning_rate": 2.503443056028656e-07,
      "loss": 0.80531764,
      "num_input_tokens_seen": 302673965,
      "step": 14035,
      "time_per_iteration": 2.632803440093994
    },
    {
      "auxiliary_loss_clip": 0.01089127,
      "auxiliary_loss_mlp": 0.0103176,
      "balance_loss_clip": 1.03444612,
      "balance_loss_mlp": 1.01995182,
      "epoch": 0.8438899744476176,
      "flos": 33724284779520.0,
      "grad_norm": 1.3332181286345353,
      "language_loss": 0.72261345,
      "learning_rate": 2.501556703225751e-07,
      "loss": 0.74382228,
      "num_input_tokens_seen": 302695560,
      "step": 14036,
      "time_per_iteration": 2.6750521659851074
    },
    {
      "auxiliary_loss_clip": 0.01098724,
      "auxiliary_loss_mlp": 0.01025102,
      "balance_loss_clip": 1.03474045,
      "balance_loss_mlp": 1.01498175,
      "epoch": 0.8439500977002856,
      "flos": 25110787386240.0,
      "grad_norm": 1.6817880078587948,
      "language_loss": 0.69759822,
      "learning_rate": 2.49967101396557e-07,
      "loss": 0.71883655,
      "num_input_tokens_seen": 302713480,
      "step": 14037,
      "time_per_iteration": 2.5505809783935547
    },
    {
      "auxiliary_loss_clip": 0.0110179,
      "auxiliary_loss_mlp": 0.01027988,
      "balance_loss_clip": 1.03467941,
      "balance_loss_mlp": 1.01638937,
      "epoch": 0.8440102209529535,
      "flos": 32850362880000.0,
      "grad_norm": 1.5956128940459593,
      "language_loss": 0.68912709,
      "learning_rate": 2.4977859883196227e-07,
      "loss": 0.7104249,
      "num_input_tokens_seen": 302736860,
      "step": 14038,
      "time_per_iteration": 2.558093786239624
    },
    {
      "auxiliary_loss_clip": 0.01055691,
      "auxiliary_loss_mlp": 0.01038713,
      "balance_loss_clip": 1.03224111,
      "balance_loss_mlp": 1.02611232,
      "epoch": 0.8440703442056215,
      "flos": 23730202195200.0,
      "grad_norm": 1.5304174722996131,
      "language_loss": 0.7633636,
      "learning_rate": 2.49590162635938e-07,
      "loss": 0.7843076,
      "num_input_tokens_seen": 302757745,
      "step": 14039,
      "time_per_iteration": 2.6460976600646973
    },
    {
      "auxiliary_loss_clip": 0.01110009,
      "auxiliary_loss_mlp": 0.01027604,
      "balance_loss_clip": 1.0379467,
      "balance_loss_mlp": 1.01541507,
      "epoch": 0.8441304674582895,
      "flos": 20193719725440.0,
      "grad_norm": 4.108290851526094,
      "language_loss": 0.79172647,
      "learning_rate": 2.4940179281563046e-07,
      "loss": 0.8131026,
      "num_input_tokens_seen": 302774885,
      "step": 14040,
      "time_per_iteration": 2.4590845108032227
    },
    {
      "auxiliary_loss_clip": 0.01071719,
      "auxiliary_loss_mlp": 0.01031416,
      "balance_loss_clip": 1.035303,
      "balance_loss_mlp": 1.01897073,
      "epoch": 0.8441905907109575,
      "flos": 20219897761920.0,
      "grad_norm": 2.2654785945407006,
      "language_loss": 0.69492865,
      "learning_rate": 2.492134893781821e-07,
      "loss": 0.71596003,
      "num_input_tokens_seen": 302791035,
      "step": 14041,
      "time_per_iteration": 2.5832502841949463
    },
    {
      "auxiliary_loss_clip": 0.01084328,
      "auxiliary_loss_mlp": 0.01037117,
      "balance_loss_clip": 1.03454852,
      "balance_loss_mlp": 1.02505922,
      "epoch": 0.8442507139636254,
      "flos": 13516453987200.0,
      "grad_norm": 1.7918305014415317,
      "language_loss": 0.68602771,
      "learning_rate": 2.490252523307341e-07,
      "loss": 0.70724219,
      "num_input_tokens_seen": 302808650,
      "step": 14042,
      "time_per_iteration": 2.474485397338867
    },
    {
      "auxiliary_loss_clip": 0.01078959,
      "auxiliary_loss_mlp": 0.01030605,
      "balance_loss_clip": 1.03403521,
      "balance_loss_mlp": 1.01946449,
      "epoch": 0.8443108372162934,
      "flos": 18220212731520.0,
      "grad_norm": 1.7800166503663042,
      "language_loss": 0.74578238,
      "learning_rate": 2.4883708168042373e-07,
      "loss": 0.76687801,
      "num_input_tokens_seen": 302824605,
      "step": 14043,
      "time_per_iteration": 2.5255541801452637
    },
    {
      "auxiliary_loss_clip": 0.01102315,
      "auxiliary_loss_mlp": 0.00781344,
      "balance_loss_clip": 1.03519988,
      "balance_loss_mlp": 1.00631893,
      "epoch": 0.8443709604689613,
      "flos": 16105110324480.0,
      "grad_norm": 2.0457704920313566,
      "language_loss": 0.72132307,
      "learning_rate": 2.486489774343865e-07,
      "loss": 0.74015963,
      "num_input_tokens_seen": 302840170,
      "step": 14044,
      "time_per_iteration": 2.4234249591827393
    },
    {
      "auxiliary_loss_clip": 0.01079433,
      "auxiliary_loss_mlp": 0.01028772,
      "balance_loss_clip": 1.03339744,
      "balance_loss_mlp": 1.01685143,
      "epoch": 0.8444310837216293,
      "flos": 18512130562560.0,
      "grad_norm": 1.60500141705661,
      "language_loss": 0.74767393,
      "learning_rate": 2.484609395997559e-07,
      "loss": 0.76875591,
      "num_input_tokens_seen": 302858320,
      "step": 14045,
      "time_per_iteration": 2.589599609375
    },
    {
      "auxiliary_loss_clip": 0.01080834,
      "auxiliary_loss_mlp": 0.0078306,
      "balance_loss_clip": 1.03113544,
      "balance_loss_mlp": 1.01022029,
      "epoch": 0.8444912069742974,
      "flos": 14939845211520.0,
      "grad_norm": 1.7360266960947157,
      "language_loss": 0.78676397,
      "learning_rate": 2.4827296818366216e-07,
      "loss": 0.80540299,
      "num_input_tokens_seen": 302875255,
      "step": 14046,
      "time_per_iteration": 2.471883773803711
    },
    {
      "auxiliary_loss_clip": 0.0107737,
      "auxiliary_loss_mlp": 0.01037527,
      "balance_loss_clip": 1.03496397,
      "balance_loss_mlp": 1.02389526,
      "epoch": 0.8445513302269653,
      "flos": 20120318282880.0,
      "grad_norm": 1.897780690441939,
      "language_loss": 0.78452039,
      "learning_rate": 2.4808506319323255e-07,
      "loss": 0.80566931,
      "num_input_tokens_seen": 302894690,
      "step": 14047,
      "time_per_iteration": 2.544961929321289
    },
    {
      "auxiliary_loss_clip": 0.01080596,
      "auxiliary_loss_mlp": 0.01029735,
      "balance_loss_clip": 1.03685689,
      "balance_loss_mlp": 1.01825547,
      "epoch": 0.8446114534796333,
      "flos": 31170928533120.0,
      "grad_norm": 1.813930869940709,
      "language_loss": 0.72309071,
      "learning_rate": 2.478972246355935e-07,
      "loss": 0.74419403,
      "num_input_tokens_seen": 302912405,
      "step": 14048,
      "time_per_iteration": 2.5702719688415527
    },
    {
      "auxiliary_loss_clip": 0.01023983,
      "auxiliary_loss_mlp": 0.01034279,
      "balance_loss_clip": 1.03570652,
      "balance_loss_mlp": 1.02266169,
      "epoch": 0.8446715767323012,
      "flos": 23948323534080.0,
      "grad_norm": 1.5439193109987108,
      "language_loss": 0.73356086,
      "learning_rate": 2.477094525178667e-07,
      "loss": 0.75414348,
      "num_input_tokens_seen": 302932525,
      "step": 14049,
      "time_per_iteration": 2.8575172424316406
    },
    {
      "auxiliary_loss_clip": 0.0101932,
      "auxiliary_loss_mlp": 0.00761628,
      "balance_loss_clip": 1.00609303,
      "balance_loss_mlp": 1.0011059,
      "epoch": 0.8447316999849692,
      "flos": 67984897484160.0,
      "grad_norm": 0.8045335048243902,
      "language_loss": 0.6062845,
      "learning_rate": 2.475217468471729e-07,
      "loss": 0.62409389,
      "num_input_tokens_seen": 302991285,
      "step": 14050,
      "time_per_iteration": 3.266545534133911
    },
    {
      "auxiliary_loss_clip": 0.01078514,
      "auxiliary_loss_mlp": 0.00786401,
      "balance_loss_clip": 1.03199875,
      "balance_loss_mlp": 1.01073647,
      "epoch": 0.8447918232376371,
      "flos": 22418924296320.0,
      "grad_norm": 2.3587522518857824,
      "language_loss": 0.72230273,
      "learning_rate": 2.473341076306303e-07,
      "loss": 0.7409519,
      "num_input_tokens_seen": 303009515,
      "step": 14051,
      "time_per_iteration": 2.582123041152954
    },
    {
      "auxiliary_loss_clip": 0.01090675,
      "auxiliary_loss_mlp": 0.01026639,
      "balance_loss_clip": 1.03469658,
      "balance_loss_mlp": 1.01496887,
      "epoch": 0.8448519464903052,
      "flos": 23694147918720.0,
      "grad_norm": 1.8581439430208904,
      "language_loss": 0.74280465,
      "learning_rate": 2.471465348753547e-07,
      "loss": 0.76397777,
      "num_input_tokens_seen": 303026905,
      "step": 14052,
      "time_per_iteration": 2.4827253818511963
    },
    {
      "auxiliary_loss_clip": 0.01072305,
      "auxiliary_loss_mlp": 0.01024554,
      "balance_loss_clip": 1.035115,
      "balance_loss_mlp": 1.01431346,
      "epoch": 0.8449120697429731,
      "flos": 13735904129280.0,
      "grad_norm": 1.7915473137272608,
      "language_loss": 0.73883557,
      "learning_rate": 2.469590285884575e-07,
      "loss": 0.75980413,
      "num_input_tokens_seen": 303045245,
      "step": 14053,
      "time_per_iteration": 2.5451111793518066
    },
    {
      "auxiliary_loss_clip": 0.01086003,
      "auxiliary_loss_mlp": 0.0102769,
      "balance_loss_clip": 1.03313637,
      "balance_loss_mlp": 1.01578736,
      "epoch": 0.8449721929956411,
      "flos": 20886795624960.0,
      "grad_norm": 1.9209065089637507,
      "language_loss": 0.73994362,
      "learning_rate": 2.467715887770494e-07,
      "loss": 0.76108062,
      "num_input_tokens_seen": 303065205,
      "step": 14054,
      "time_per_iteration": 2.4609224796295166
    },
    {
      "auxiliary_loss_clip": 0.01096468,
      "auxiliary_loss_mlp": 0.01030247,
      "balance_loss_clip": 1.03490734,
      "balance_loss_mlp": 1.01819539,
      "epoch": 0.845032316248309,
      "flos": 33216939129600.0,
      "grad_norm": 1.4921050930352815,
      "language_loss": 0.78246403,
      "learning_rate": 2.4658421544823895e-07,
      "loss": 0.8037312,
      "num_input_tokens_seen": 303088250,
      "step": 14055,
      "time_per_iteration": 2.6078484058380127
    },
    {
      "auxiliary_loss_clip": 0.01089834,
      "auxiliary_loss_mlp": 0.01030806,
      "balance_loss_clip": 1.03365564,
      "balance_loss_mlp": 1.01949859,
      "epoch": 0.845092439500977,
      "flos": 23585230903680.0,
      "grad_norm": 1.7872732960435973,
      "language_loss": 0.73167539,
      "learning_rate": 2.463969086091302e-07,
      "loss": 0.75288177,
      "num_input_tokens_seen": 303109280,
      "step": 14056,
      "time_per_iteration": 2.489408254623413
    },
    {
      "auxiliary_loss_clip": 0.01097155,
      "auxiliary_loss_mlp": 0.01036085,
      "balance_loss_clip": 1.03612912,
      "balance_loss_mlp": 1.0234791,
      "epoch": 0.8451525627536449,
      "flos": 13333920048000.0,
      "grad_norm": 1.9907839074685436,
      "language_loss": 0.67089581,
      "learning_rate": 2.4620966826682686e-07,
      "loss": 0.6922282,
      "num_input_tokens_seen": 303126075,
      "step": 14057,
      "time_per_iteration": 2.5094311237335205
    },
    {
      "auxiliary_loss_clip": 0.01056951,
      "auxiliary_loss_mlp": 0.01028736,
      "balance_loss_clip": 1.03584182,
      "balance_loss_mlp": 1.0165652,
      "epoch": 0.8452126860063129,
      "flos": 27817985583360.0,
      "grad_norm": 1.6147150786226652,
      "language_loss": 0.77387506,
      "learning_rate": 2.460224944284284e-07,
      "loss": 0.79473186,
      "num_input_tokens_seen": 303146920,
      "step": 14058,
      "time_per_iteration": 2.5946662425994873
    },
    {
      "auxiliary_loss_clip": 0.01103458,
      "auxiliary_loss_mlp": 0.01032766,
      "balance_loss_clip": 1.03438926,
      "balance_loss_mlp": 1.02108383,
      "epoch": 0.845272809258981,
      "flos": 27124694202240.0,
      "grad_norm": 1.5967886267850524,
      "language_loss": 0.69922948,
      "learning_rate": 2.45835387101033e-07,
      "loss": 0.72059178,
      "num_input_tokens_seen": 303167885,
      "step": 14059,
      "time_per_iteration": 4.245558500289917
    },
    {
      "auxiliary_loss_clip": 0.01108122,
      "auxiliary_loss_mlp": 0.01036133,
      "balance_loss_clip": 1.03610373,
      "balance_loss_mlp": 1.0234071,
      "epoch": 0.8453329325116489,
      "flos": 18332577452160.0,
      "grad_norm": 2.2691220217651638,
      "language_loss": 0.57473195,
      "learning_rate": 2.4564834629173516e-07,
      "loss": 0.59617448,
      "num_input_tokens_seen": 303185000,
      "step": 14060,
      "time_per_iteration": 3.8222591876983643
    },
    {
      "auxiliary_loss_clip": 0.01085145,
      "auxiliary_loss_mlp": 0.01032963,
      "balance_loss_clip": 1.03363276,
      "balance_loss_mlp": 1.02002835,
      "epoch": 0.8453930557643169,
      "flos": 22675254727680.0,
      "grad_norm": 1.6864479877151286,
      "language_loss": 0.75851357,
      "learning_rate": 2.454613720076277e-07,
      "loss": 0.77969462,
      "num_input_tokens_seen": 303205210,
      "step": 14061,
      "time_per_iteration": 2.565387725830078
    },
    {
      "auxiliary_loss_clip": 0.01085786,
      "auxiliary_loss_mlp": 0.01025654,
      "balance_loss_clip": 1.03513992,
      "balance_loss_mlp": 1.01338172,
      "epoch": 0.8454531790169848,
      "flos": 22487261921280.0,
      "grad_norm": 2.0180007894499505,
      "language_loss": 0.70896006,
      "learning_rate": 2.452744642558013e-07,
      "loss": 0.73007452,
      "num_input_tokens_seen": 303224655,
      "step": 14062,
      "time_per_iteration": 2.5023956298828125
    },
    {
      "auxiliary_loss_clip": 0.01001815,
      "auxiliary_loss_mlp": 0.01002755,
      "balance_loss_clip": 1.0267868,
      "balance_loss_mlp": 1.00160432,
      "epoch": 0.8455133022696528,
      "flos": 58277848481280.0,
      "grad_norm": 0.6338222040276487,
      "language_loss": 0.52665424,
      "learning_rate": 2.450876230433432e-07,
      "loss": 0.54669994,
      "num_input_tokens_seen": 303289645,
      "step": 14063,
      "time_per_iteration": 4.707958698272705
    },
    {
      "auxiliary_loss_clip": 0.01061753,
      "auxiliary_loss_mlp": 0.0102503,
      "balance_loss_clip": 1.03666127,
      "balance_loss_mlp": 1.01452804,
      "epoch": 0.8455734255223207,
      "flos": 21361283308800.0,
      "grad_norm": 1.8195613933235573,
      "language_loss": 0.82294381,
      "learning_rate": 2.449008483773378e-07,
      "loss": 0.84381163,
      "num_input_tokens_seen": 303308350,
      "step": 14064,
      "time_per_iteration": 3.1661176681518555
    },
    {
      "auxiliary_loss_clip": 0.01097077,
      "auxiliary_loss_mlp": 0.01032169,
      "balance_loss_clip": 1.03815269,
      "balance_loss_mlp": 1.0195986,
      "epoch": 0.8456335487749888,
      "flos": 20449260057600.0,
      "grad_norm": 1.8367215900267515,
      "language_loss": 0.72588086,
      "learning_rate": 2.447141402648685e-07,
      "loss": 0.74717331,
      "num_input_tokens_seen": 303325230,
      "step": 14065,
      "time_per_iteration": 2.4744772911071777
    },
    {
      "auxiliary_loss_clip": 0.01069658,
      "auxiliary_loss_mlp": 0.01030534,
      "balance_loss_clip": 1.03472042,
      "balance_loss_mlp": 1.0191077,
      "epoch": 0.8456936720276567,
      "flos": 28840901097600.0,
      "grad_norm": 1.5100051932201923,
      "language_loss": 0.77498901,
      "learning_rate": 2.445274987130146e-07,
      "loss": 0.79599094,
      "num_input_tokens_seen": 303345810,
      "step": 14066,
      "time_per_iteration": 2.6412084102630615
    },
    {
      "auxiliary_loss_clip": 0.01075142,
      "auxiliary_loss_mlp": 0.01029962,
      "balance_loss_clip": 1.03725457,
      "balance_loss_mlp": 1.01823139,
      "epoch": 0.8457537952803247,
      "flos": 22672884430080.0,
      "grad_norm": 1.4751618865835265,
      "language_loss": 0.69787908,
      "learning_rate": 2.4434092372885363e-07,
      "loss": 0.71893013,
      "num_input_tokens_seen": 303365140,
      "step": 14067,
      "time_per_iteration": 2.5382070541381836
    },
    {
      "auxiliary_loss_clip": 0.01065746,
      "auxiliary_loss_mlp": 0.01027482,
      "balance_loss_clip": 1.03177154,
      "balance_loss_mlp": 1.01575208,
      "epoch": 0.8458139185329926,
      "flos": 33802929607680.0,
      "grad_norm": 3.384315473705079,
      "language_loss": 0.71017116,
      "learning_rate": 2.4415441531946144e-07,
      "loss": 0.73110342,
      "num_input_tokens_seen": 303386150,
      "step": 14068,
      "time_per_iteration": 2.741776466369629
    },
    {
      "auxiliary_loss_clip": 0.00999167,
      "auxiliary_loss_mlp": 0.01002843,
      "balance_loss_clip": 1.00795293,
      "balance_loss_mlp": 1.00168061,
      "epoch": 0.8458740417856606,
      "flos": 70295929603200.0,
      "grad_norm": 0.6922893585955606,
      "language_loss": 0.60522616,
      "learning_rate": 2.4396797349190976e-07,
      "loss": 0.62524617,
      "num_input_tokens_seen": 303453770,
      "step": 14069,
      "time_per_iteration": 4.645868539810181
    },
    {
      "auxiliary_loss_clip": 0.01078333,
      "auxiliary_loss_mlp": 0.01026773,
      "balance_loss_clip": 1.0350312,
      "balance_loss_mlp": 1.01559138,
      "epoch": 0.8459341650383285,
      "flos": 24170862245760.0,
      "grad_norm": 1.5825425501155832,
      "language_loss": 0.74604785,
      "learning_rate": 2.4378159825326804e-07,
      "loss": 0.7670989,
      "num_input_tokens_seen": 303474520,
      "step": 14070,
      "time_per_iteration": 2.530360221862793
    },
    {
      "auxiliary_loss_clip": 0.01056579,
      "auxiliary_loss_mlp": 0.01034527,
      "balance_loss_clip": 1.03294134,
      "balance_loss_mlp": 1.02181911,
      "epoch": 0.8459942882909965,
      "flos": 38181158369280.0,
      "grad_norm": 1.6307845257461309,
      "language_loss": 0.67315912,
      "learning_rate": 2.435952896106039e-07,
      "loss": 0.69407022,
      "num_input_tokens_seen": 303497345,
      "step": 14071,
      "time_per_iteration": 2.720252513885498
    },
    {
      "auxiliary_loss_clip": 0.01019171,
      "auxiliary_loss_mlp": 0.00760002,
      "balance_loss_clip": 1.00662971,
      "balance_loss_mlp": 0.99795038,
      "epoch": 0.8460544115436646,
      "flos": 64118252177280.0,
      "grad_norm": 0.7336920617435406,
      "language_loss": 0.6105181,
      "learning_rate": 2.4340904757098313e-07,
      "loss": 0.62830979,
      "num_input_tokens_seen": 303554890,
      "step": 14072,
      "time_per_iteration": 2.968027353286743
    },
    {
      "auxiliary_loss_clip": 0.01064339,
      "auxiliary_loss_mlp": 0.01030739,
      "balance_loss_clip": 1.03523159,
      "balance_loss_mlp": 1.01768565,
      "epoch": 0.8461145347963325,
      "flos": 24170826332160.0,
      "grad_norm": 2.0818188172639926,
      "language_loss": 0.72158712,
      "learning_rate": 2.4322287214146664e-07,
      "loss": 0.74253792,
      "num_input_tokens_seen": 303574380,
      "step": 14073,
      "time_per_iteration": 2.594224214553833
    },
    {
      "auxiliary_loss_clip": 0.01088675,
      "auxiliary_loss_mlp": 0.0103648,
      "balance_loss_clip": 1.03770626,
      "balance_loss_mlp": 1.02368295,
      "epoch": 0.8461746580490005,
      "flos": 34893787697280.0,
      "grad_norm": 2.017372376353228,
      "language_loss": 0.77925032,
      "learning_rate": 2.430367633291155e-07,
      "loss": 0.80050182,
      "num_input_tokens_seen": 303594910,
      "step": 14074,
      "time_per_iteration": 2.6565566062927246
    },
    {
      "auxiliary_loss_clip": 0.01094427,
      "auxiliary_loss_mlp": 0.01034084,
      "balance_loss_clip": 1.03641844,
      "balance_loss_mlp": 1.02201462,
      "epoch": 0.8462347813016684,
      "flos": 25557014044800.0,
      "grad_norm": 2.4252636067830537,
      "language_loss": 0.7522229,
      "learning_rate": 2.4285072114098583e-07,
      "loss": 0.77350795,
      "num_input_tokens_seen": 303613520,
      "step": 14075,
      "time_per_iteration": 2.589517593383789
    },
    {
      "auxiliary_loss_clip": 0.01079856,
      "auxiliary_loss_mlp": 0.01028867,
      "balance_loss_clip": 1.03401709,
      "balance_loss_mlp": 1.01647484,
      "epoch": 0.8462949045543364,
      "flos": 21325336773120.0,
      "grad_norm": 1.9468372890477397,
      "language_loss": 0.72926062,
      "learning_rate": 2.4266474558413355e-07,
      "loss": 0.75034785,
      "num_input_tokens_seen": 303631225,
      "step": 14076,
      "time_per_iteration": 2.5452277660369873
    },
    {
      "auxiliary_loss_clip": 0.01086269,
      "auxiliary_loss_mlp": 0.01031721,
      "balance_loss_clip": 1.0349685,
      "balance_loss_mlp": 1.02010369,
      "epoch": 0.8463550278070043,
      "flos": 22637440684800.0,
      "grad_norm": 1.9621082074976173,
      "language_loss": 0.77829742,
      "learning_rate": 2.4247883666560945e-07,
      "loss": 0.79947734,
      "num_input_tokens_seen": 303649175,
      "step": 14077,
      "time_per_iteration": 2.5348451137542725
    },
    {
      "auxiliary_loss_clip": 0.01072122,
      "auxiliary_loss_mlp": 0.01033461,
      "balance_loss_clip": 1.03480339,
      "balance_loss_mlp": 1.02135575,
      "epoch": 0.8464151510596724,
      "flos": 13005588804480.0,
      "grad_norm": 1.902937237994731,
      "language_loss": 0.75022137,
      "learning_rate": 2.422929943924643e-07,
      "loss": 0.77127719,
      "num_input_tokens_seen": 303665915,
      "step": 14078,
      "time_per_iteration": 2.5139570236206055
    },
    {
      "auxiliary_loss_clip": 0.01062122,
      "auxiliary_loss_mlp": 0.01024651,
      "balance_loss_clip": 1.03420722,
      "balance_loss_mlp": 1.01269984,
      "epoch": 0.8464752743123403,
      "flos": 15704921923200.0,
      "grad_norm": 2.314330182511338,
      "language_loss": 0.85242188,
      "learning_rate": 2.4210721877174565e-07,
      "loss": 0.87328959,
      "num_input_tokens_seen": 303679985,
      "step": 14079,
      "time_per_iteration": 2.581977128982544
    },
    {
      "auxiliary_loss_clip": 0.01079252,
      "auxiliary_loss_mlp": 0.0103575,
      "balance_loss_clip": 1.03505743,
      "balance_loss_mlp": 1.02228546,
      "epoch": 0.8465353975650083,
      "flos": 21653955325440.0,
      "grad_norm": 1.9099246251347561,
      "language_loss": 0.59306639,
      "learning_rate": 2.419215098104965e-07,
      "loss": 0.61421645,
      "num_input_tokens_seen": 303698470,
      "step": 14080,
      "time_per_iteration": 2.600745439529419
    },
    {
      "auxiliary_loss_clip": 0.01079317,
      "auxiliary_loss_mlp": 0.01029621,
      "balance_loss_clip": 1.03448796,
      "balance_loss_mlp": 1.01712227,
      "epoch": 0.8465955208176762,
      "flos": 18515650095360.0,
      "grad_norm": 2.6851014224414365,
      "language_loss": 0.65955001,
      "learning_rate": 2.4173586751576014e-07,
      "loss": 0.68063939,
      "num_input_tokens_seen": 303716415,
      "step": 14081,
      "time_per_iteration": 2.5805532932281494
    },
    {
      "auxiliary_loss_clip": 0.01090857,
      "auxiliary_loss_mlp": 0.0103716,
      "balance_loss_clip": 1.03397918,
      "balance_loss_mlp": 1.02480984,
      "epoch": 0.8466556440703442,
      "flos": 24200559815040.0,
      "grad_norm": 1.850476146642062,
      "language_loss": 0.72766614,
      "learning_rate": 2.41550291894576e-07,
      "loss": 0.74894631,
      "num_input_tokens_seen": 303734490,
      "step": 14082,
      "time_per_iteration": 2.509004592895508
    },
    {
      "auxiliary_loss_clip": 0.01051641,
      "auxiliary_loss_mlp": 0.01031349,
      "balance_loss_clip": 1.0312283,
      "balance_loss_mlp": 1.01800978,
      "epoch": 0.8467157673230121,
      "flos": 20375894528640.0,
      "grad_norm": 1.8547312696622418,
      "language_loss": 0.76433325,
      "learning_rate": 2.413647829539809e-07,
      "loss": 0.78516316,
      "num_input_tokens_seen": 303752310,
      "step": 14083,
      "time_per_iteration": 2.5720412731170654
    },
    {
      "auxiliary_loss_clip": 0.01057657,
      "auxiliary_loss_mlp": 0.01033447,
      "balance_loss_clip": 1.03230369,
      "balance_loss_mlp": 1.01969028,
      "epoch": 0.8467758905756801,
      "flos": 28473642489600.0,
      "grad_norm": 1.7358334638441375,
      "language_loss": 0.66073859,
      "learning_rate": 2.411793407010092e-07,
      "loss": 0.68164968,
      "num_input_tokens_seen": 303776065,
      "step": 14084,
      "time_per_iteration": 2.6330409049987793
    },
    {
      "auxiliary_loss_clip": 0.01060923,
      "auxiliary_loss_mlp": 0.01029156,
      "balance_loss_clip": 1.03507411,
      "balance_loss_mlp": 1.01795053,
      "epoch": 0.8468360138283482,
      "flos": 11692551139200.0,
      "grad_norm": 2.0470669525957907,
      "language_loss": 0.69878274,
      "learning_rate": 2.409939651426938e-07,
      "loss": 0.71968353,
      "num_input_tokens_seen": 303793500,
      "step": 14085,
      "time_per_iteration": 2.56280255317688
    },
    {
      "auxiliary_loss_clip": 0.01062133,
      "auxiliary_loss_mlp": 0.01030145,
      "balance_loss_clip": 1.03360188,
      "balance_loss_mlp": 1.01862931,
      "epoch": 0.8468961370810161,
      "flos": 24607859109120.0,
      "grad_norm": 1.5405991236685308,
      "language_loss": 0.70902568,
      "learning_rate": 2.408086562860634e-07,
      "loss": 0.7299484,
      "num_input_tokens_seen": 303814835,
      "step": 14086,
      "time_per_iteration": 2.652015209197998
    },
    {
      "auxiliary_loss_clip": 0.01086322,
      "auxiliary_loss_mlp": 0.01032236,
      "balance_loss_clip": 1.03268695,
      "balance_loss_mlp": 1.02003527,
      "epoch": 0.8469562603336841,
      "flos": 19609812236160.0,
      "grad_norm": 1.7096072556629764,
      "language_loss": 0.74442172,
      "learning_rate": 2.4062341413814445e-07,
      "loss": 0.7656073,
      "num_input_tokens_seen": 303834505,
      "step": 14087,
      "time_per_iteration": 2.475109100341797
    },
    {
      "auxiliary_loss_clip": 0.01081854,
      "auxiliary_loss_mlp": 0.01022362,
      "balance_loss_clip": 1.03669381,
      "balance_loss_mlp": 1.01069188,
      "epoch": 0.847016383586352,
      "flos": 22638949056000.0,
      "grad_norm": 1.4968730906851149,
      "language_loss": 0.74009812,
      "learning_rate": 2.4043823870596227e-07,
      "loss": 0.76114023,
      "num_input_tokens_seen": 303855050,
      "step": 14088,
      "time_per_iteration": 2.5299978256225586
    },
    {
      "auxiliary_loss_clip": 0.01094735,
      "auxiliary_loss_mlp": 0.01033037,
      "balance_loss_clip": 1.0351814,
      "balance_loss_mlp": 1.02082419,
      "epoch": 0.84707650683902,
      "flos": 20960161153920.0,
      "grad_norm": 1.9672384279518031,
      "language_loss": 0.71826535,
      "learning_rate": 2.402531299965387e-07,
      "loss": 0.73954308,
      "num_input_tokens_seen": 303875635,
      "step": 14089,
      "time_per_iteration": 2.478297710418701
    },
    {
      "auxiliary_loss_clip": 0.01102755,
      "auxiliary_loss_mlp": 0.01027323,
      "balance_loss_clip": 1.03644717,
      "balance_loss_mlp": 1.01602244,
      "epoch": 0.8471366300916879,
      "flos": 24093007516800.0,
      "grad_norm": 1.4374383668614616,
      "language_loss": 0.79151487,
      "learning_rate": 2.400680880168928e-07,
      "loss": 0.81281567,
      "num_input_tokens_seen": 303896750,
      "step": 14090,
      "time_per_iteration": 2.483769416809082
    },
    {
      "auxiliary_loss_clip": 0.0104949,
      "auxiliary_loss_mlp": 0.01040797,
      "balance_loss_clip": 1.03132999,
      "balance_loss_mlp": 1.02661681,
      "epoch": 0.847196753344356,
      "flos": 18332900674560.0,
      "grad_norm": 1.9623267959299169,
      "language_loss": 0.77041411,
      "learning_rate": 2.3988311277404085e-07,
      "loss": 0.79131699,
      "num_input_tokens_seen": 303915435,
      "step": 14091,
      "time_per_iteration": 2.590689182281494
    },
    {
      "auxiliary_loss_clip": 0.01028129,
      "auxiliary_loss_mlp": 0.01006477,
      "balance_loss_clip": 1.00521851,
      "balance_loss_mlp": 1.00532103,
      "epoch": 0.8472568765970239,
      "flos": 49567536956160.0,
      "grad_norm": 0.819561073868323,
      "language_loss": 0.59415406,
      "learning_rate": 2.396982042749982e-07,
      "loss": 0.61450011,
      "num_input_tokens_seen": 303977245,
      "step": 14092,
      "time_per_iteration": 3.131436347961426
    },
    {
      "auxiliary_loss_clip": 0.01077946,
      "auxiliary_loss_mlp": 0.0103417,
      "balance_loss_clip": 1.03286695,
      "balance_loss_mlp": 1.021492,
      "epoch": 0.8473169998496919,
      "flos": 19279074781440.0,
      "grad_norm": 1.8844147639178976,
      "language_loss": 0.70294654,
      "learning_rate": 2.395133625267756e-07,
      "loss": 0.72406769,
      "num_input_tokens_seen": 303996055,
      "step": 14093,
      "time_per_iteration": 2.4793708324432373
    },
    {
      "auxiliary_loss_clip": 0.01100473,
      "auxiliary_loss_mlp": 0.01026116,
      "balance_loss_clip": 1.03390956,
      "balance_loss_mlp": 1.01492202,
      "epoch": 0.8473771231023598,
      "flos": 17675555829120.0,
      "grad_norm": 1.8885159574751447,
      "language_loss": 0.83456242,
      "learning_rate": 2.3932858753638263e-07,
      "loss": 0.85582829,
      "num_input_tokens_seen": 304012205,
      "step": 14094,
      "time_per_iteration": 2.4148824214935303
    },
    {
      "auxiliary_loss_clip": 0.01091184,
      "auxiliary_loss_mlp": 0.01032408,
      "balance_loss_clip": 1.03609872,
      "balance_loss_mlp": 1.02073133,
      "epoch": 0.8474372463550278,
      "flos": 26359761144960.0,
      "grad_norm": 1.6116688491759743,
      "language_loss": 0.71133077,
      "learning_rate": 2.3914387931082626e-07,
      "loss": 0.73256671,
      "num_input_tokens_seen": 304033475,
      "step": 14095,
      "time_per_iteration": 2.503856897354126
    },
    {
      "auxiliary_loss_clip": 0.01090766,
      "auxiliary_loss_mlp": 0.00783318,
      "balance_loss_clip": 1.03490758,
      "balance_loss_mlp": 1.01004636,
      "epoch": 0.8474973696076957,
      "flos": 23402050519680.0,
      "grad_norm": 1.5784662606792474,
      "language_loss": 0.80846167,
      "learning_rate": 2.3895923785711105e-07,
      "loss": 0.82720256,
      "num_input_tokens_seen": 304051845,
      "step": 14096,
      "time_per_iteration": 2.5179052352905273
    },
    {
      "auxiliary_loss_clip": 0.0109456,
      "auxiliary_loss_mlp": 0.01030104,
      "balance_loss_clip": 1.03461075,
      "balance_loss_mlp": 1.01728272,
      "epoch": 0.8475574928603637,
      "flos": 25075666863360.0,
      "grad_norm": 1.6929097450202593,
      "language_loss": 0.77757597,
      "learning_rate": 2.387746631822374e-07,
      "loss": 0.79882252,
      "num_input_tokens_seen": 304069965,
      "step": 14097,
      "time_per_iteration": 3.943601369857788
    },
    {
      "auxiliary_loss_clip": 0.01072052,
      "auxiliary_loss_mlp": 0.01023717,
      "balance_loss_clip": 1.03351152,
      "balance_loss_mlp": 1.01252949,
      "epoch": 0.8476176161130318,
      "flos": 19966691813760.0,
      "grad_norm": 1.7864901917689309,
      "language_loss": 0.80571234,
      "learning_rate": 2.385901552932048e-07,
      "loss": 0.82667011,
      "num_input_tokens_seen": 304086805,
      "step": 14098,
      "time_per_iteration": 2.5659918785095215
    },
    {
      "auxiliary_loss_clip": 0.01087599,
      "auxiliary_loss_mlp": 0.00783696,
      "balance_loss_clip": 1.03403211,
      "balance_loss_mlp": 1.00955796,
      "epoch": 0.8476777393656997,
      "flos": 21285834791040.0,
      "grad_norm": 4.575822037304887,
      "language_loss": 0.71859336,
      "learning_rate": 2.3840571419701062e-07,
      "loss": 0.73730624,
      "num_input_tokens_seen": 304105865,
      "step": 14099,
      "time_per_iteration": 3.871368646621704
    },
    {
      "auxiliary_loss_clip": 0.01091198,
      "auxiliary_loss_mlp": 0.01032058,
      "balance_loss_clip": 1.03305697,
      "balance_loss_mlp": 1.01914155,
      "epoch": 0.8477378626183677,
      "flos": 29971476650880.0,
      "grad_norm": 1.974146275469105,
      "language_loss": 0.63342202,
      "learning_rate": 2.3822133990064787e-07,
      "loss": 0.65465462,
      "num_input_tokens_seen": 304128300,
      "step": 14100,
      "time_per_iteration": 2.5426323413848877
    },
    {
      "auxiliary_loss_clip": 0.01094262,
      "auxiliary_loss_mlp": 0.01031271,
      "balance_loss_clip": 1.03503227,
      "balance_loss_mlp": 1.01858664,
      "epoch": 0.8477979858710356,
      "flos": 24237727413120.0,
      "grad_norm": 2.671279564174652,
      "language_loss": 0.73419583,
      "learning_rate": 2.380370324111085e-07,
      "loss": 0.75545108,
      "num_input_tokens_seen": 304143695,
      "step": 14101,
      "time_per_iteration": 2.502648115158081
    },
    {
      "auxiliary_loss_clip": 0.01092979,
      "auxiliary_loss_mlp": 0.01030605,
      "balance_loss_clip": 1.03321958,
      "balance_loss_mlp": 1.01896381,
      "epoch": 0.8478581091237036,
      "flos": 25593678852480.0,
      "grad_norm": 1.5297736407856617,
      "language_loss": 0.71154082,
      "learning_rate": 2.3785279173538163e-07,
      "loss": 0.73277664,
      "num_input_tokens_seen": 304165800,
      "step": 14102,
      "time_per_iteration": 3.9205312728881836
    },
    {
      "auxiliary_loss_clip": 0.01079097,
      "auxiliary_loss_mlp": 0.01033153,
      "balance_loss_clip": 1.03364706,
      "balance_loss_mlp": 1.01986098,
      "epoch": 0.8479182323763715,
      "flos": 12057116227200.0,
      "grad_norm": 2.3070508981337126,
      "language_loss": 0.81704056,
      "learning_rate": 2.3766861788045366e-07,
      "loss": 0.83816302,
      "num_input_tokens_seen": 304182910,
      "step": 14103,
      "time_per_iteration": 2.5113487243652344
    },
    {
      "auxiliary_loss_clip": 0.01103902,
      "auxiliary_loss_mlp": 0.01030793,
      "balance_loss_clip": 1.03582418,
      "balance_loss_mlp": 1.01943874,
      "epoch": 0.8479783556290396,
      "flos": 21433391861760.0,
      "grad_norm": 2.162711540947283,
      "language_loss": 0.78385139,
      "learning_rate": 2.374845108533079e-07,
      "loss": 0.80519831,
      "num_input_tokens_seen": 304200175,
      "step": 14104,
      "time_per_iteration": 2.475825786590576
    },
    {
      "auxiliary_loss_clip": 0.01096695,
      "auxiliary_loss_mlp": 0.010358,
      "balance_loss_clip": 1.0373385,
      "balance_loss_mlp": 1.02362323,
      "epoch": 0.8480384788817075,
      "flos": 19642634288640.0,
      "grad_norm": 1.8437295592627636,
      "language_loss": 0.78845543,
      "learning_rate": 2.3730047066092607e-07,
      "loss": 0.80978036,
      "num_input_tokens_seen": 304217775,
      "step": 14105,
      "time_per_iteration": 2.468998432159424
    },
    {
      "auxiliary_loss_clip": 0.01079045,
      "auxiliary_loss_mlp": 0.01037126,
      "balance_loss_clip": 1.03435636,
      "balance_loss_mlp": 1.02273726,
      "epoch": 0.8480986021343755,
      "flos": 22489201255680.0,
      "grad_norm": 1.8096154208018684,
      "language_loss": 0.50279027,
      "learning_rate": 2.3711649731028749e-07,
      "loss": 0.52395195,
      "num_input_tokens_seen": 304235760,
      "step": 14106,
      "time_per_iteration": 2.522883892059326
    },
    {
      "auxiliary_loss_clip": 0.01069365,
      "auxiliary_loss_mlp": 0.01029256,
      "balance_loss_clip": 1.03344262,
      "balance_loss_mlp": 1.01793694,
      "epoch": 0.8481587253870434,
      "flos": 22090557139200.0,
      "grad_norm": 1.8970926906185903,
      "language_loss": 0.76042354,
      "learning_rate": 2.3693259080836792e-07,
      "loss": 0.78140968,
      "num_input_tokens_seen": 304253985,
      "step": 14107,
      "time_per_iteration": 3.924323558807373
    },
    {
      "auxiliary_loss_clip": 0.0107135,
      "auxiliary_loss_mlp": 0.01027528,
      "balance_loss_clip": 1.03385162,
      "balance_loss_mlp": 1.01578546,
      "epoch": 0.8482188486397114,
      "flos": 33582689366400.0,
      "grad_norm": 1.5984378804915558,
      "language_loss": 0.73499352,
      "learning_rate": 2.3674875116214087e-07,
      "loss": 0.75598228,
      "num_input_tokens_seen": 304276785,
      "step": 14108,
      "time_per_iteration": 2.642639398574829
    },
    {
      "auxiliary_loss_clip": 0.01101422,
      "auxiliary_loss_mlp": 0.01027726,
      "balance_loss_clip": 1.03567994,
      "balance_loss_mlp": 1.01363564,
      "epoch": 0.8482789718923793,
      "flos": 20919402195840.0,
      "grad_norm": 1.870347301287449,
      "language_loss": 0.72766477,
      "learning_rate": 2.3656497837857836e-07,
      "loss": 0.74895626,
      "num_input_tokens_seen": 304296310,
      "step": 14109,
      "time_per_iteration": 2.4398763179779053
    },
    {
      "auxiliary_loss_clip": 0.01042856,
      "auxiliary_loss_mlp": 0.01032999,
      "balance_loss_clip": 1.03353143,
      "balance_loss_mlp": 1.02061296,
      "epoch": 0.8483390951450474,
      "flos": 12896204912640.0,
      "grad_norm": 2.1400919358585164,
      "language_loss": 0.74439973,
      "learning_rate": 2.3638127246464811e-07,
      "loss": 0.7651583,
      "num_input_tokens_seen": 304311715,
      "step": 14110,
      "time_per_iteration": 2.5858349800109863
    },
    {
      "auxiliary_loss_clip": 0.01046008,
      "auxiliary_loss_mlp": 0.01033618,
      "balance_loss_clip": 1.03512549,
      "balance_loss_mlp": 1.02256703,
      "epoch": 0.8483992183977154,
      "flos": 25081628520960.0,
      "grad_norm": 1.7337658214913485,
      "language_loss": 0.76130813,
      "learning_rate": 2.3619763342731658e-07,
      "loss": 0.78210437,
      "num_input_tokens_seen": 304331910,
      "step": 14111,
      "time_per_iteration": 2.6229629516601562
    },
    {
      "auxiliary_loss_clip": 0.01102598,
      "auxiliary_loss_mlp": 0.01027497,
      "balance_loss_clip": 1.0354147,
      "balance_loss_mlp": 1.01644635,
      "epoch": 0.8484593416503833,
      "flos": 25557445008000.0,
      "grad_norm": 1.5497110591815617,
      "language_loss": 0.67754102,
      "learning_rate": 2.3601406127354772e-07,
      "loss": 0.69884193,
      "num_input_tokens_seen": 304351405,
      "step": 14112,
      "time_per_iteration": 2.4923081398010254
    },
    {
      "auxiliary_loss_clip": 0.01090615,
      "auxiliary_loss_mlp": 0.01036272,
      "balance_loss_clip": 1.03407049,
      "balance_loss_mlp": 1.02447009,
      "epoch": 0.8485194649030513,
      "flos": 27198454780800.0,
      "grad_norm": 1.384651642319795,
      "language_loss": 0.73778749,
      "learning_rate": 2.3583055601030312e-07,
      "loss": 0.75905633,
      "num_input_tokens_seen": 304372935,
      "step": 14113,
      "time_per_iteration": 2.529430627822876
    },
    {
      "auxiliary_loss_clip": 0.01064873,
      "auxiliary_loss_mlp": 0.01030777,
      "balance_loss_clip": 1.03446817,
      "balance_loss_mlp": 1.01931524,
      "epoch": 0.8485795881557192,
      "flos": 24205910941440.0,
      "grad_norm": 1.92221842147391,
      "language_loss": 0.66772324,
      "learning_rate": 2.3564711764454003e-07,
      "loss": 0.6886797,
      "num_input_tokens_seen": 304393070,
      "step": 14114,
      "time_per_iteration": 2.565100908279419
    },
    {
      "auxiliary_loss_clip": 0.01106674,
      "auxiliary_loss_mlp": 0.01031554,
      "balance_loss_clip": 1.03681612,
      "balance_loss_mlp": 1.0191685,
      "epoch": 0.8486397114083872,
      "flos": 21141653598720.0,
      "grad_norm": 1.8981617987180666,
      "language_loss": 0.78846258,
      "learning_rate": 2.3546374618321495e-07,
      "loss": 0.80984485,
      "num_input_tokens_seen": 304411195,
      "step": 14115,
      "time_per_iteration": 2.4441282749176025
    },
    {
      "auxiliary_loss_clip": 0.01102948,
      "auxiliary_loss_mlp": 0.01028425,
      "balance_loss_clip": 1.03415108,
      "balance_loss_mlp": 1.01697552,
      "epoch": 0.8486998346610551,
      "flos": 19974772373760.0,
      "grad_norm": 1.7782009621532118,
      "language_loss": 0.78745461,
      "learning_rate": 2.3528044163328187e-07,
      "loss": 0.80876839,
      "num_input_tokens_seen": 304429425,
      "step": 14116,
      "time_per_iteration": 2.442976236343384
    },
    {
      "auxiliary_loss_clip": 0.01094916,
      "auxiliary_loss_mlp": 0.01032614,
      "balance_loss_clip": 1.03463531,
      "balance_loss_mlp": 1.0201869,
      "epoch": 0.8487599579137232,
      "flos": 19792310261760.0,
      "grad_norm": 2.6104210410277124,
      "language_loss": 0.68558306,
      "learning_rate": 2.3509720400169076e-07,
      "loss": 0.70685834,
      "num_input_tokens_seen": 304447460,
      "step": 14117,
      "time_per_iteration": 2.462104082107544
    },
    {
      "auxiliary_loss_clip": 0.01084778,
      "auxiliary_loss_mlp": 0.0102776,
      "balance_loss_clip": 1.03276443,
      "balance_loss_mlp": 1.0155232,
      "epoch": 0.8488200811663911,
      "flos": 26396030903040.0,
      "grad_norm": 1.862336326254606,
      "language_loss": 0.64688456,
      "learning_rate": 2.3491403329539096e-07,
      "loss": 0.66801,
      "num_input_tokens_seen": 304468230,
      "step": 14118,
      "time_per_iteration": 2.5493640899658203
    },
    {
      "auxiliary_loss_clip": 0.01062264,
      "auxiliary_loss_mlp": 0.01028934,
      "balance_loss_clip": 1.03357267,
      "balance_loss_mlp": 1.01753736,
      "epoch": 0.8488802044190591,
      "flos": 16359285939840.0,
      "grad_norm": 1.5751926904734792,
      "language_loss": 0.73267996,
      "learning_rate": 2.3473092952132757e-07,
      "loss": 0.75359195,
      "num_input_tokens_seen": 304484860,
      "step": 14119,
      "time_per_iteration": 2.5180680751800537
    },
    {
      "auxiliary_loss_clip": 0.01071494,
      "auxiliary_loss_mlp": 0.01030911,
      "balance_loss_clip": 1.033023,
      "balance_loss_mlp": 1.01802492,
      "epoch": 0.848940327671727,
      "flos": 19208869649280.0,
      "grad_norm": 1.7581163751766475,
      "language_loss": 0.7798084,
      "learning_rate": 2.345478926864446e-07,
      "loss": 0.80083239,
      "num_input_tokens_seen": 304503575,
      "step": 14120,
      "time_per_iteration": 2.5387110710144043
    },
    {
      "auxiliary_loss_clip": 0.01089149,
      "auxiliary_loss_mlp": 0.01029675,
      "balance_loss_clip": 1.03632271,
      "balance_loss_mlp": 1.01776624,
      "epoch": 0.849000450924395,
      "flos": 21871178824320.0,
      "grad_norm": 1.8378986923723706,
      "language_loss": 0.7545563,
      "learning_rate": 2.3436492279768227e-07,
      "loss": 0.7757445,
      "num_input_tokens_seen": 304525005,
      "step": 14121,
      "time_per_iteration": 2.487318992614746
    },
    {
      "auxiliary_loss_clip": 0.00996109,
      "auxiliary_loss_mlp": 0.01004582,
      "balance_loss_clip": 1.00595069,
      "balance_loss_mlp": 1.00344372,
      "epoch": 0.8490605741770629,
      "flos": 71166475624320.0,
      "grad_norm": 0.8067539868655578,
      "language_loss": 0.60167289,
      "learning_rate": 2.3418201986197883e-07,
      "loss": 0.62167978,
      "num_input_tokens_seen": 304585220,
      "step": 14122,
      "time_per_iteration": 3.1411499977111816
    },
    {
      "auxiliary_loss_clip": 0.01093562,
      "auxiliary_loss_mlp": 0.01030558,
      "balance_loss_clip": 1.03564131,
      "balance_loss_mlp": 1.01931667,
      "epoch": 0.849120697429731,
      "flos": 24973357950720.0,
      "grad_norm": 1.7270855822877715,
      "language_loss": 0.79793346,
      "learning_rate": 2.3399918388627048e-07,
      "loss": 0.81917471,
      "num_input_tokens_seen": 304604665,
      "step": 14123,
      "time_per_iteration": 2.545786142349243
    },
    {
      "auxiliary_loss_clip": 0.01089983,
      "auxiliary_loss_mlp": 0.01028208,
      "balance_loss_clip": 1.03614223,
      "balance_loss_mlp": 1.01680517,
      "epoch": 0.8491808206823989,
      "flos": 23032277959680.0,
      "grad_norm": 1.9199175568145632,
      "language_loss": 0.83032608,
      "learning_rate": 2.3381641487749016e-07,
      "loss": 0.85150802,
      "num_input_tokens_seen": 304620600,
      "step": 14124,
      "time_per_iteration": 2.475848436355591
    },
    {
      "auxiliary_loss_clip": 0.01058745,
      "auxiliary_loss_mlp": 0.01029591,
      "balance_loss_clip": 1.03689098,
      "balance_loss_mlp": 1.01782489,
      "epoch": 0.8492409439350669,
      "flos": 23878549365120.0,
      "grad_norm": 3.8320789001463247,
      "language_loss": 0.7185483,
      "learning_rate": 2.3363371284256805e-07,
      "loss": 0.73943162,
      "num_input_tokens_seen": 304639540,
      "step": 14125,
      "time_per_iteration": 2.6198043823242188
    },
    {
      "auxiliary_loss_clip": 0.01107971,
      "auxiliary_loss_mlp": 0.01034083,
      "balance_loss_clip": 1.03599644,
      "balance_loss_mlp": 1.02102971,
      "epoch": 0.8493010671877349,
      "flos": 22419893963520.0,
      "grad_norm": 1.704454913903163,
      "language_loss": 0.73566735,
      "learning_rate": 2.3345107778843288e-07,
      "loss": 0.75708789,
      "num_input_tokens_seen": 304660595,
      "step": 14126,
      "time_per_iteration": 2.4594645500183105
    },
    {
      "auxiliary_loss_clip": 0.01055853,
      "auxiliary_loss_mlp": 0.01037893,
      "balance_loss_clip": 1.03201306,
      "balance_loss_mlp": 1.02572167,
      "epoch": 0.8493611904404028,
      "flos": 17529435302400.0,
      "grad_norm": 1.4167191825258882,
      "language_loss": 0.67398006,
      "learning_rate": 2.3326850972200928e-07,
      "loss": 0.69491744,
      "num_input_tokens_seen": 304679580,
      "step": 14127,
      "time_per_iteration": 2.5765721797943115
    },
    {
      "auxiliary_loss_clip": 0.01071268,
      "auxiliary_loss_mlp": 0.00785168,
      "balance_loss_clip": 1.03353012,
      "balance_loss_mlp": 1.01074445,
      "epoch": 0.8494213136930708,
      "flos": 19462937523840.0,
      "grad_norm": 1.6488876732713063,
      "language_loss": 0.69278514,
      "learning_rate": 2.330860086502211e-07,
      "loss": 0.71134955,
      "num_input_tokens_seen": 304698385,
      "step": 14128,
      "time_per_iteration": 2.535161018371582
    },
    {
      "auxiliary_loss_clip": 0.01077983,
      "auxiliary_loss_mlp": 0.01032799,
      "balance_loss_clip": 1.0334444,
      "balance_loss_mlp": 1.02043128,
      "epoch": 0.8494814369457387,
      "flos": 18770292587520.0,
      "grad_norm": 1.7334074051398476,
      "language_loss": 0.77590233,
      "learning_rate": 2.3290357457998855e-07,
      "loss": 0.79701012,
      "num_input_tokens_seen": 304715430,
      "step": 14129,
      "time_per_iteration": 2.5002689361572266
    },
    {
      "auxiliary_loss_clip": 0.0103863,
      "auxiliary_loss_mlp": 0.01030089,
      "balance_loss_clip": 1.03381157,
      "balance_loss_mlp": 1.01816177,
      "epoch": 0.8495415601984068,
      "flos": 23331486251520.0,
      "grad_norm": 1.9278608632691003,
      "language_loss": 0.68187523,
      "learning_rate": 2.3272120751823031e-07,
      "loss": 0.70256245,
      "num_input_tokens_seen": 304734345,
      "step": 14130,
      "time_per_iteration": 2.6199803352355957
    },
    {
      "auxiliary_loss_clip": 0.01094014,
      "auxiliary_loss_mlp": 0.01031642,
      "balance_loss_clip": 1.03515518,
      "balance_loss_mlp": 1.01947713,
      "epoch": 0.8496016834510747,
      "flos": 26612859352320.0,
      "grad_norm": 1.6706637887512492,
      "language_loss": 0.71418852,
      "learning_rate": 2.3253890747186e-07,
      "loss": 0.73544508,
      "num_input_tokens_seen": 304755030,
      "step": 14131,
      "time_per_iteration": 2.5323610305786133
    },
    {
      "auxiliary_loss_clip": 0.01071694,
      "auxiliary_loss_mlp": 0.01029549,
      "balance_loss_clip": 1.03387475,
      "balance_loss_mlp": 1.01832557,
      "epoch": 0.8496618067037427,
      "flos": 25480380378240.0,
      "grad_norm": 1.8390961951195566,
      "language_loss": 0.68433177,
      "learning_rate": 2.3235667444779162e-07,
      "loss": 0.7053442,
      "num_input_tokens_seen": 304774320,
      "step": 14132,
      "time_per_iteration": 2.5700089931488037
    },
    {
      "auxiliary_loss_clip": 0.0110059,
      "auxiliary_loss_mlp": 0.01034814,
      "balance_loss_clip": 1.03353918,
      "balance_loss_mlp": 1.02375758,
      "epoch": 0.8497219299564106,
      "flos": 25374587846400.0,
      "grad_norm": 1.563905314171235,
      "language_loss": 0.7038548,
      "learning_rate": 2.3217450845293564e-07,
      "loss": 0.72520888,
      "num_input_tokens_seen": 304795355,
      "step": 14133,
      "time_per_iteration": 2.4921376705169678
    },
    {
      "auxiliary_loss_clip": 0.00993693,
      "auxiliary_loss_mlp": 0.00762417,
      "balance_loss_clip": 1.00844145,
      "balance_loss_mlp": 0.99891168,
      "epoch": 0.8497820532090786,
      "flos": 67780279658880.0,
      "grad_norm": 0.727809195650896,
      "language_loss": 0.5767802,
      "learning_rate": 2.3199240949419918e-07,
      "loss": 0.59434128,
      "num_input_tokens_seen": 304863915,
      "step": 14134,
      "time_per_iteration": 3.286475896835327
    },
    {
      "auxiliary_loss_clip": 0.01069784,
      "auxiliary_loss_mlp": 0.0103025,
      "balance_loss_clip": 1.03596151,
      "balance_loss_mlp": 1.01795363,
      "epoch": 0.8498421764617465,
      "flos": 23440546920960.0,
      "grad_norm": 2.229311010484275,
      "language_loss": 0.79247069,
      "learning_rate": 2.3181037757848787e-07,
      "loss": 0.81347102,
      "num_input_tokens_seen": 304881555,
      "step": 14135,
      "time_per_iteration": 2.5837128162384033
    },
    {
      "auxiliary_loss_clip": 0.01092354,
      "auxiliary_loss_mlp": 0.01029001,
      "balance_loss_clip": 1.03611732,
      "balance_loss_mlp": 1.01659679,
      "epoch": 0.8499022997144146,
      "flos": 17712615686400.0,
      "grad_norm": 2.0056862609359296,
      "language_loss": 0.63309658,
      "learning_rate": 2.316284127127044e-07,
      "loss": 0.65431011,
      "num_input_tokens_seen": 304898760,
      "step": 14136,
      "time_per_iteration": 3.8931891918182373
    },
    {
      "auxiliary_loss_clip": 0.01096501,
      "auxiliary_loss_mlp": 0.01029849,
      "balance_loss_clip": 1.03618443,
      "balance_loss_mlp": 1.0169028,
      "epoch": 0.8499624229670825,
      "flos": 18588512833920.0,
      "grad_norm": 2.0511069003768543,
      "language_loss": 0.83951402,
      "learning_rate": 2.3144651490374835e-07,
      "loss": 0.86077756,
      "num_input_tokens_seen": 304915465,
      "step": 14137,
      "time_per_iteration": 3.866767406463623
    },
    {
      "auxiliary_loss_clip": 0.01071969,
      "auxiliary_loss_mlp": 0.01026371,
      "balance_loss_clip": 1.03540802,
      "balance_loss_mlp": 1.01562428,
      "epoch": 0.8500225462197505,
      "flos": 24345854328960.0,
      "grad_norm": 1.9649068662486422,
      "language_loss": 0.78583372,
      "learning_rate": 2.3126468415851773e-07,
      "loss": 0.80681711,
      "num_input_tokens_seen": 304933190,
      "step": 14138,
      "time_per_iteration": 2.5760514736175537
    },
    {
      "auxiliary_loss_clip": 0.01093,
      "auxiliary_loss_mlp": 0.0102843,
      "balance_loss_clip": 1.03564286,
      "balance_loss_mlp": 1.01710558,
      "epoch": 0.8500826694724185,
      "flos": 16545518979840.0,
      "grad_norm": 1.6660429519414741,
      "language_loss": 0.64471149,
      "learning_rate": 2.310829204839073e-07,
      "loss": 0.6659258,
      "num_input_tokens_seen": 304951110,
      "step": 14139,
      "time_per_iteration": 2.4487147331237793
    },
    {
      "auxiliary_loss_clip": 0.01063618,
      "auxiliary_loss_mlp": 0.01026415,
      "balance_loss_clip": 1.03653753,
      "balance_loss_mlp": 1.01513791,
      "epoch": 0.8501427927250864,
      "flos": 16289404030080.0,
      "grad_norm": 1.6033089485906502,
      "language_loss": 0.70879722,
      "learning_rate": 2.3090122388681043e-07,
      "loss": 0.72969759,
      "num_input_tokens_seen": 304969095,
      "step": 14140,
      "time_per_iteration": 2.5398292541503906
    },
    {
      "auxiliary_loss_clip": 0.01065768,
      "auxiliary_loss_mlp": 0.01030645,
      "balance_loss_clip": 1.0346024,
      "balance_loss_mlp": 1.01883113,
      "epoch": 0.8502029159777544,
      "flos": 26687912820480.0,
      "grad_norm": 1.8218745827263967,
      "language_loss": 0.63942623,
      "learning_rate": 2.3071959437411648e-07,
      "loss": 0.66039038,
      "num_input_tokens_seen": 304989315,
      "step": 14141,
      "time_per_iteration": 4.006561994552612
    },
    {
      "auxiliary_loss_clip": 0.01072577,
      "auxiliary_loss_mlp": 0.01029651,
      "balance_loss_clip": 1.03523278,
      "balance_loss_mlp": 1.01802242,
      "epoch": 0.8502630392304223,
      "flos": 35590778179200.0,
      "grad_norm": 1.4740217319572082,
      "language_loss": 0.70721501,
      "learning_rate": 2.3053803195271214e-07,
      "loss": 0.72823727,
      "num_input_tokens_seen": 305011020,
      "step": 14142,
      "time_per_iteration": 2.6839096546173096
    },
    {
      "auxiliary_loss_clip": 0.01055598,
      "auxiliary_loss_mlp": 0.01027796,
      "balance_loss_clip": 1.03139699,
      "balance_loss_mlp": 1.0165602,
      "epoch": 0.8503231624830904,
      "flos": 21649466125440.0,
      "grad_norm": 1.4192175292575733,
      "language_loss": 0.6556952,
      "learning_rate": 2.3035653662948375e-07,
      "loss": 0.67652917,
      "num_input_tokens_seen": 305033550,
      "step": 14143,
      "time_per_iteration": 2.639082431793213
    },
    {
      "auxiliary_loss_clip": 0.01074749,
      "auxiliary_loss_mlp": 0.00781614,
      "balance_loss_clip": 1.0339905,
      "balance_loss_mlp": 1.00609374,
      "epoch": 0.8503832857357583,
      "flos": 22417451838720.0,
      "grad_norm": 2.072766840753015,
      "language_loss": 0.6798799,
      "learning_rate": 2.3017510841131216e-07,
      "loss": 0.69844353,
      "num_input_tokens_seen": 305052885,
      "step": 14144,
      "time_per_iteration": 2.5525529384613037
    },
    {
      "auxiliary_loss_clip": 0.01044867,
      "auxiliary_loss_mlp": 0.01034892,
      "balance_loss_clip": 1.03326249,
      "balance_loss_mlp": 1.02241123,
      "epoch": 0.8504434089884263,
      "flos": 18697968552960.0,
      "grad_norm": 1.9086751241251951,
      "language_loss": 0.6453703,
      "learning_rate": 2.299937473050777e-07,
      "loss": 0.66616791,
      "num_input_tokens_seen": 305071995,
      "step": 14145,
      "time_per_iteration": 2.6046791076660156
    },
    {
      "auxiliary_loss_clip": 0.01082816,
      "auxiliary_loss_mlp": 0.010334,
      "balance_loss_clip": 1.03450787,
      "balance_loss_mlp": 1.02082324,
      "epoch": 0.8505035322410942,
      "flos": 20007989475840.0,
      "grad_norm": 1.7277721483402266,
      "language_loss": 0.85777175,
      "learning_rate": 2.2981245331765842e-07,
      "loss": 0.87893391,
      "num_input_tokens_seen": 305090190,
      "step": 14146,
      "time_per_iteration": 3.9808478355407715
    },
    {
      "auxiliary_loss_clip": 0.01101287,
      "auxiliary_loss_mlp": 0.01028423,
      "balance_loss_clip": 1.03360415,
      "balance_loss_mlp": 1.01690149,
      "epoch": 0.8505636554937622,
      "flos": 20812173120000.0,
      "grad_norm": 1.6439957464903876,
      "language_loss": 0.83649123,
      "learning_rate": 2.2963122645592814e-07,
      "loss": 0.85778832,
      "num_input_tokens_seen": 305109355,
      "step": 14147,
      "time_per_iteration": 2.442958354949951
    },
    {
      "auxiliary_loss_clip": 0.01094397,
      "auxiliary_loss_mlp": 0.01031355,
      "balance_loss_clip": 1.03535521,
      "balance_loss_mlp": 1.01884389,
      "epoch": 0.8506237787464301,
      "flos": 14174445277440.0,
      "grad_norm": 2.557360920053027,
      "language_loss": 0.85710573,
      "learning_rate": 2.2945006672675894e-07,
      "loss": 0.87836331,
      "num_input_tokens_seen": 305124165,
      "step": 14148,
      "time_per_iteration": 2.4650378227233887
    },
    {
      "auxiliary_loss_clip": 0.01081239,
      "auxiliary_loss_mlp": 0.01028625,
      "balance_loss_clip": 1.03401756,
      "balance_loss_mlp": 1.01605463,
      "epoch": 0.8506839019990982,
      "flos": 23258372117760.0,
      "grad_norm": 1.6225686915835031,
      "language_loss": 0.71912521,
      "learning_rate": 2.292689741370204e-07,
      "loss": 0.74022382,
      "num_input_tokens_seen": 305143940,
      "step": 14149,
      "time_per_iteration": 2.521343231201172
    },
    {
      "auxiliary_loss_clip": 0.01081693,
      "auxiliary_loss_mlp": 0.01031253,
      "balance_loss_clip": 1.03470421,
      "balance_loss_mlp": 1.01926017,
      "epoch": 0.8507440252517661,
      "flos": 23659206963840.0,
      "grad_norm": 1.58267397094711,
      "language_loss": 0.76001811,
      "learning_rate": 2.290879486935804e-07,
      "loss": 0.7811476,
      "num_input_tokens_seen": 305163505,
      "step": 14150,
      "time_per_iteration": 2.55377197265625
    },
    {
      "auxiliary_loss_clip": 0.01064761,
      "auxiliary_loss_mlp": 0.01036739,
      "balance_loss_clip": 1.03445184,
      "balance_loss_mlp": 1.02496099,
      "epoch": 0.8508041485044341,
      "flos": 18661339658880.0,
      "grad_norm": 1.6288550917396096,
      "language_loss": 0.7220161,
      "learning_rate": 2.2890699040330231e-07,
      "loss": 0.74303114,
      "num_input_tokens_seen": 305182325,
      "step": 14151,
      "time_per_iteration": 2.5338752269744873
    },
    {
      "auxiliary_loss_clip": 0.00985951,
      "auxiliary_loss_mlp": 0.01006058,
      "balance_loss_clip": 1.01238275,
      "balance_loss_mlp": 1.00481272,
      "epoch": 0.8508642717571021,
      "flos": 52510918055040.0,
      "grad_norm": 0.883326108758271,
      "language_loss": 0.59638959,
      "learning_rate": 2.2872609927304909e-07,
      "loss": 0.6163097,
      "num_input_tokens_seen": 305230775,
      "step": 14152,
      "time_per_iteration": 3.02317214012146
    },
    {
      "auxiliary_loss_clip": 0.01009283,
      "auxiliary_loss_mlp": 0.01004267,
      "balance_loss_clip": 1.00553918,
      "balance_loss_mlp": 1.00297928,
      "epoch": 0.85092439500977,
      "flos": 69297145050240.0,
      "grad_norm": 0.6911959934000129,
      "language_loss": 0.61171091,
      "learning_rate": 2.285452753096797e-07,
      "loss": 0.63184643,
      "num_input_tokens_seen": 305296000,
      "step": 14153,
      "time_per_iteration": 3.244988203048706
    },
    {
      "auxiliary_loss_clip": 0.01093757,
      "auxiliary_loss_mlp": 0.01031237,
      "balance_loss_clip": 1.03437066,
      "balance_loss_mlp": 1.0185771,
      "epoch": 0.850984518262438,
      "flos": 24389737770240.0,
      "grad_norm": 2.916465364309451,
      "language_loss": 0.80777705,
      "learning_rate": 2.2836451852005067e-07,
      "loss": 0.82902694,
      "num_input_tokens_seen": 305314705,
      "step": 14154,
      "time_per_iteration": 2.517843723297119
    },
    {
      "auxiliary_loss_clip": 0.01065902,
      "auxiliary_loss_mlp": 0.01033533,
      "balance_loss_clip": 1.03301489,
      "balance_loss_mlp": 1.02285171,
      "epoch": 0.851044641515106,
      "flos": 23294821443840.0,
      "grad_norm": 1.603262251837608,
      "language_loss": 0.79350364,
      "learning_rate": 2.281838289110165e-07,
      "loss": 0.81449801,
      "num_input_tokens_seen": 305333870,
      "step": 14155,
      "time_per_iteration": 2.5622315406799316
    },
    {
      "auxiliary_loss_clip": 0.0106923,
      "auxiliary_loss_mlp": 0.01026522,
      "balance_loss_clip": 1.03393579,
      "balance_loss_mlp": 1.01457739,
      "epoch": 0.851104764767774,
      "flos": 22050085489920.0,
      "grad_norm": 1.6473491984066035,
      "language_loss": 0.70981133,
      "learning_rate": 2.2800320648942904e-07,
      "loss": 0.73076886,
      "num_input_tokens_seen": 305352780,
      "step": 14156,
      "time_per_iteration": 2.5768892765045166
    },
    {
      "auxiliary_loss_clip": 0.01061885,
      "auxiliary_loss_mlp": 0.01029333,
      "balance_loss_clip": 1.03518224,
      "balance_loss_mlp": 1.01813912,
      "epoch": 0.8511648880204419,
      "flos": 20704728562560.0,
      "grad_norm": 1.7672695944681798,
      "language_loss": 0.73470837,
      "learning_rate": 2.278226512621386e-07,
      "loss": 0.75562054,
      "num_input_tokens_seen": 305371370,
      "step": 14157,
      "time_per_iteration": 2.558565378189087
    },
    {
      "auxiliary_loss_clip": 0.01033577,
      "auxiliary_loss_mlp": 0.01025899,
      "balance_loss_clip": 1.03344107,
      "balance_loss_mlp": 1.01559913,
      "epoch": 0.8512250112731099,
      "flos": 24024669891840.0,
      "grad_norm": 3.3276033001405962,
      "language_loss": 0.79525733,
      "learning_rate": 2.2764216323598995e-07,
      "loss": 0.81585211,
      "num_input_tokens_seen": 305387955,
      "step": 14158,
      "time_per_iteration": 2.6828107833862305
    },
    {
      "auxiliary_loss_clip": 0.01089831,
      "auxiliary_loss_mlp": 0.01035738,
      "balance_loss_clip": 1.03405762,
      "balance_loss_mlp": 1.02249408,
      "epoch": 0.8512851345257778,
      "flos": 22015467757440.0,
      "grad_norm": 2.1500544650160993,
      "language_loss": 0.79548883,
      "learning_rate": 2.27461742417828e-07,
      "loss": 0.81674457,
      "num_input_tokens_seen": 305406285,
      "step": 14159,
      "time_per_iteration": 2.476229190826416
    },
    {
      "auxiliary_loss_clip": 0.01084187,
      "auxiliary_loss_mlp": 0.01033503,
      "balance_loss_clip": 1.03510988,
      "balance_loss_mlp": 1.02166533,
      "epoch": 0.8513452577784458,
      "flos": 14830209924480.0,
      "grad_norm": 1.6967227287271132,
      "language_loss": 0.71410012,
      "learning_rate": 2.2728138881449488e-07,
      "loss": 0.73527706,
      "num_input_tokens_seen": 305424500,
      "step": 14160,
      "time_per_iteration": 2.533555030822754
    },
    {
      "auxiliary_loss_clip": 0.01098521,
      "auxiliary_loss_mlp": 0.01034041,
      "balance_loss_clip": 1.03594351,
      "balance_loss_mlp": 1.02097583,
      "epoch": 0.8514053810311137,
      "flos": 33035662166400.0,
      "grad_norm": 1.9799193221414308,
      "language_loss": 0.70296329,
      "learning_rate": 2.2710110243282866e-07,
      "loss": 0.72428888,
      "num_input_tokens_seen": 305442990,
      "step": 14161,
      "time_per_iteration": 2.5875399112701416
    },
    {
      "auxiliary_loss_clip": 0.01093621,
      "auxiliary_loss_mlp": 0.01028803,
      "balance_loss_clip": 1.03238595,
      "balance_loss_mlp": 1.01719809,
      "epoch": 0.8514655042837818,
      "flos": 27564456412800.0,
      "grad_norm": 2.119349827986808,
      "language_loss": 0.77529407,
      "learning_rate": 2.2692088327966653e-07,
      "loss": 0.79651833,
      "num_input_tokens_seen": 305463065,
      "step": 14162,
      "time_per_iteration": 2.540135622024536
    },
    {
      "auxiliary_loss_clip": 0.01095944,
      "auxiliary_loss_mlp": 0.01033887,
      "balance_loss_clip": 1.03711891,
      "balance_loss_mlp": 1.02172804,
      "epoch": 0.8515256275364497,
      "flos": 35556052705920.0,
      "grad_norm": 1.8020287799985892,
      "language_loss": 0.76500493,
      "learning_rate": 2.2674073136184235e-07,
      "loss": 0.78630328,
      "num_input_tokens_seen": 305489070,
      "step": 14163,
      "time_per_iteration": 2.6160409450531006
    },
    {
      "auxiliary_loss_clip": 0.01014062,
      "auxiliary_loss_mlp": 0.01000156,
      "balance_loss_clip": 1.0113554,
      "balance_loss_mlp": 0.99902326,
      "epoch": 0.8515857507891177,
      "flos": 70207372621440.0,
      "grad_norm": 0.6991597176287263,
      "language_loss": 0.55055022,
      "learning_rate": 2.2656064668618735e-07,
      "loss": 0.57069242,
      "num_input_tokens_seen": 305551490,
      "step": 14164,
      "time_per_iteration": 3.183668613433838
    },
    {
      "auxiliary_loss_clip": 0.01093667,
      "auxiliary_loss_mlp": 0.01035691,
      "balance_loss_clip": 1.03393924,
      "balance_loss_mlp": 1.02346039,
      "epoch": 0.8516458740417857,
      "flos": 22675290641280.0,
      "grad_norm": 9.771071496145192,
      "language_loss": 0.72608566,
      "learning_rate": 2.2638062925953005e-07,
      "loss": 0.74737918,
      "num_input_tokens_seen": 305570535,
      "step": 14165,
      "time_per_iteration": 2.482961416244507
    },
    {
      "auxiliary_loss_clip": 0.0106528,
      "auxiliary_loss_mlp": 0.01026883,
      "balance_loss_clip": 1.03492188,
      "balance_loss_mlp": 1.01509309,
      "epoch": 0.8517059972944536,
      "flos": 22747435107840.0,
      "grad_norm": 1.6397680816421436,
      "language_loss": 0.67177027,
      "learning_rate": 2.26200679088697e-07,
      "loss": 0.69269192,
      "num_input_tokens_seen": 305590800,
      "step": 14166,
      "time_per_iteration": 2.5696189403533936
    },
    {
      "auxiliary_loss_clip": 0.01079157,
      "auxiliary_loss_mlp": 0.01031445,
      "balance_loss_clip": 1.0313096,
      "balance_loss_mlp": 1.0198462,
      "epoch": 0.8517661205471216,
      "flos": 21689147675520.0,
      "grad_norm": 1.900345302549379,
      "language_loss": 0.73409176,
      "learning_rate": 2.260207961805125e-07,
      "loss": 0.75519788,
      "num_input_tokens_seen": 305609495,
      "step": 14167,
      "time_per_iteration": 2.521881341934204
    },
    {
      "auxiliary_loss_clip": 0.01102356,
      "auxiliary_loss_mlp": 0.0102876,
      "balance_loss_clip": 1.0347811,
      "balance_loss_mlp": 1.01754189,
      "epoch": 0.8518262437997896,
      "flos": 25374839241600.0,
      "grad_norm": 1.6668330261377853,
      "language_loss": 0.80008006,
      "learning_rate": 2.258409805417969e-07,
      "loss": 0.82139117,
      "num_input_tokens_seen": 305629420,
      "step": 14168,
      "time_per_iteration": 2.483902931213379
    },
    {
      "auxiliary_loss_clip": 0.0110223,
      "auxiliary_loss_mlp": 0.0102482,
      "balance_loss_clip": 1.0341897,
      "balance_loss_mlp": 1.01381731,
      "epoch": 0.8518863670524576,
      "flos": 27235406897280.0,
      "grad_norm": 2.0470967222298695,
      "language_loss": 0.76563877,
      "learning_rate": 2.2566123217936893e-07,
      "loss": 0.78690922,
      "num_input_tokens_seen": 305649835,
      "step": 14169,
      "time_per_iteration": 2.5062568187713623
    },
    {
      "auxiliary_loss_clip": 0.01106872,
      "auxiliary_loss_mlp": 0.01029501,
      "balance_loss_clip": 1.03632689,
      "balance_loss_mlp": 1.01736569,
      "epoch": 0.8519464903051255,
      "flos": 20959514709120.0,
      "grad_norm": 1.6683032167453562,
      "language_loss": 0.63710928,
      "learning_rate": 2.254815511000452e-07,
      "loss": 0.65847301,
      "num_input_tokens_seen": 305668840,
      "step": 14170,
      "time_per_iteration": 2.4369845390319824
    },
    {
      "auxiliary_loss_clip": 0.01091008,
      "auxiliary_loss_mlp": 0.01025454,
      "balance_loss_clip": 1.03321993,
      "balance_loss_mlp": 1.0138607,
      "epoch": 0.8520066135577935,
      "flos": 18441745862400.0,
      "grad_norm": 2.10093560909819,
      "language_loss": 0.8623715,
      "learning_rate": 2.253019373106384e-07,
      "loss": 0.8835361,
      "num_input_tokens_seen": 305686955,
      "step": 14171,
      "time_per_iteration": 2.4422199726104736
    },
    {
      "auxiliary_loss_clip": 0.01089177,
      "auxiliary_loss_mlp": 0.01034556,
      "balance_loss_clip": 1.03435445,
      "balance_loss_mlp": 1.0227896,
      "epoch": 0.8520667368104614,
      "flos": 29130233149440.0,
      "grad_norm": 1.7458461064980582,
      "language_loss": 0.54680502,
      "learning_rate": 2.2512239081796003e-07,
      "loss": 0.5680424,
      "num_input_tokens_seen": 305706290,
      "step": 14172,
      "time_per_iteration": 2.5270235538482666
    },
    {
      "auxiliary_loss_clip": 0.01078712,
      "auxiliary_loss_mlp": 0.01028585,
      "balance_loss_clip": 1.03345823,
      "balance_loss_mlp": 1.01848149,
      "epoch": 0.8521268600631294,
      "flos": 16034366488320.0,
      "grad_norm": 2.471421680359514,
      "language_loss": 0.69415915,
      "learning_rate": 2.2494291162881862e-07,
      "loss": 0.71523213,
      "num_input_tokens_seen": 305723835,
      "step": 14173,
      "time_per_iteration": 2.5105814933776855
    },
    {
      "auxiliary_loss_clip": 0.01082208,
      "auxiliary_loss_mlp": 0.00783462,
      "balance_loss_clip": 1.03396213,
      "balance_loss_mlp": 1.00942802,
      "epoch": 0.8521869833157973,
      "flos": 22454870832000.0,
      "grad_norm": 3.282163716807284,
      "language_loss": 0.76738739,
      "learning_rate": 2.247634997500205e-07,
      "loss": 0.78604412,
      "num_input_tokens_seen": 305741655,
      "step": 14174,
      "time_per_iteration": 2.5092735290527344
    },
    {
      "auxiliary_loss_clip": 0.01070018,
      "auxiliary_loss_mlp": 0.00783539,
      "balance_loss_clip": 1.03308535,
      "balance_loss_mlp": 1.00826263,
      "epoch": 0.8522471065684654,
      "flos": 24972029147520.0,
      "grad_norm": 1.6499097440724706,
      "language_loss": 0.81897455,
      "learning_rate": 2.245841551883676e-07,
      "loss": 0.83751017,
      "num_input_tokens_seen": 305761890,
      "step": 14175,
      "time_per_iteration": 3.9617342948913574
    },
    {
      "auxiliary_loss_clip": 0.01108328,
      "auxiliary_loss_mlp": 0.01029078,
      "balance_loss_clip": 1.03704512,
      "balance_loss_mlp": 1.01651335,
      "epoch": 0.8523072298211333,
      "flos": 17710604524800.0,
      "grad_norm": 2.2944007779187254,
      "language_loss": 0.6548093,
      "learning_rate": 2.2440487795066153e-07,
      "loss": 0.67618328,
      "num_input_tokens_seen": 305779190,
      "step": 14176,
      "time_per_iteration": 2.420305013656616
    },
    {
      "auxiliary_loss_clip": 0.01080118,
      "auxiliary_loss_mlp": 0.00784199,
      "balance_loss_clip": 1.03374219,
      "balance_loss_mlp": 1.01085472,
      "epoch": 0.8523673530738013,
      "flos": 25446193608960.0,
      "grad_norm": 2.017664337243453,
      "language_loss": 0.78486341,
      "learning_rate": 2.2422566804370068e-07,
      "loss": 0.80350661,
      "num_input_tokens_seen": 305799870,
      "step": 14177,
      "time_per_iteration": 3.954669237136841
    },
    {
      "auxiliary_loss_clip": 0.01082854,
      "auxiliary_loss_mlp": 0.01030355,
      "balance_loss_clip": 1.03528214,
      "balance_loss_mlp": 1.01830328,
      "epoch": 0.8524274763264693,
      "flos": 31429593348480.0,
      "grad_norm": 1.6315630990400944,
      "language_loss": 0.73356998,
      "learning_rate": 2.2404652547428026e-07,
      "loss": 0.75470209,
      "num_input_tokens_seen": 305819695,
      "step": 14178,
      "time_per_iteration": 2.6070971488952637
    },
    {
      "auxiliary_loss_clip": 0.01069076,
      "auxiliary_loss_mlp": 0.01032606,
      "balance_loss_clip": 1.03593159,
      "balance_loss_mlp": 1.02086949,
      "epoch": 0.8524875995791372,
      "flos": 17712651600000.0,
      "grad_norm": 1.684953669805899,
      "language_loss": 0.74868852,
      "learning_rate": 2.238674502491935e-07,
      "loss": 0.76970536,
      "num_input_tokens_seen": 305837270,
      "step": 14179,
      "time_per_iteration": 4.092430591583252
    },
    {
      "auxiliary_loss_clip": 0.01102136,
      "auxiliary_loss_mlp": 0.01028294,
      "balance_loss_clip": 1.03537166,
      "balance_loss_mlp": 1.01700521,
      "epoch": 0.8525477228318052,
      "flos": 21687316081920.0,
      "grad_norm": 2.1660338673778354,
      "language_loss": 0.81879807,
      "learning_rate": 2.2368844237523165e-07,
      "loss": 0.84010231,
      "num_input_tokens_seen": 305855250,
      "step": 14180,
      "time_per_iteration": 2.482255458831787
    },
    {
      "auxiliary_loss_clip": 0.01046749,
      "auxiliary_loss_mlp": 0.0103557,
      "balance_loss_clip": 1.03277707,
      "balance_loss_mlp": 1.02357149,
      "epoch": 0.8526078460844732,
      "flos": 24827057856000.0,
      "grad_norm": 3.172850014138972,
      "language_loss": 0.60997713,
      "learning_rate": 2.235095018591815e-07,
      "loss": 0.63080037,
      "num_input_tokens_seen": 305875660,
      "step": 14181,
      "time_per_iteration": 2.611671209335327
    },
    {
      "auxiliary_loss_clip": 0.0110334,
      "auxiliary_loss_mlp": 0.01035559,
      "balance_loss_clip": 1.03664684,
      "balance_loss_mlp": 1.02436495,
      "epoch": 0.8526679693371412,
      "flos": 13516418073600.0,
      "grad_norm": 2.087761396730017,
      "language_loss": 0.72542822,
      "learning_rate": 2.2333062870782894e-07,
      "loss": 0.74681723,
      "num_input_tokens_seen": 305892415,
      "step": 14182,
      "time_per_iteration": 2.433070659637451
    },
    {
      "auxiliary_loss_clip": 0.01053108,
      "auxiliary_loss_mlp": 0.01031751,
      "balance_loss_clip": 1.03531647,
      "balance_loss_mlp": 1.01987755,
      "epoch": 0.8527280925898091,
      "flos": 23514092017920.0,
      "grad_norm": 1.4770399771783616,
      "language_loss": 0.70880985,
      "learning_rate": 2.2315182292795697e-07,
      "loss": 0.72965842,
      "num_input_tokens_seen": 305912665,
      "step": 14183,
      "time_per_iteration": 2.64776349067688
    },
    {
      "auxiliary_loss_clip": 0.01077573,
      "auxiliary_loss_mlp": 0.0102905,
      "balance_loss_clip": 1.03716302,
      "balance_loss_mlp": 1.01803553,
      "epoch": 0.8527882158424771,
      "flos": 20303031790080.0,
      "grad_norm": 2.0572949274127708,
      "language_loss": 0.73037577,
      "learning_rate": 2.2297308452634644e-07,
      "loss": 0.75144196,
      "num_input_tokens_seen": 305931515,
      "step": 14184,
      "time_per_iteration": 2.594871997833252
    },
    {
      "auxiliary_loss_clip": 0.01105089,
      "auxiliary_loss_mlp": 0.01032735,
      "balance_loss_clip": 1.03631616,
      "balance_loss_mlp": 1.02095699,
      "epoch": 0.852848339095145,
      "flos": 17202504689280.0,
      "grad_norm": 1.7823785791244964,
      "language_loss": 0.76605058,
      "learning_rate": 2.2279441350977457e-07,
      "loss": 0.7874288,
      "num_input_tokens_seen": 305949965,
      "step": 14185,
      "time_per_iteration": 3.901979446411133
    },
    {
      "auxiliary_loss_clip": 0.01062103,
      "auxiliary_loss_mlp": 0.01031258,
      "balance_loss_clip": 1.03219318,
      "balance_loss_mlp": 1.01797807,
      "epoch": 0.852908462347813,
      "flos": 18368990864640.0,
      "grad_norm": 2.1101798272930683,
      "language_loss": 0.79664183,
      "learning_rate": 2.2261580988501637e-07,
      "loss": 0.81757545,
      "num_input_tokens_seen": 305967820,
      "step": 14186,
      "time_per_iteration": 2.5126733779907227
    },
    {
      "auxiliary_loss_clip": 0.01079413,
      "auxiliary_loss_mlp": 0.01028017,
      "balance_loss_clip": 1.03166437,
      "balance_loss_mlp": 1.01526797,
      "epoch": 0.8529685856004809,
      "flos": 18624890332800.0,
      "grad_norm": 1.6380722279918796,
      "language_loss": 0.62965655,
      "learning_rate": 2.224372736588449e-07,
      "loss": 0.65073085,
      "num_input_tokens_seen": 305985505,
      "step": 14187,
      "time_per_iteration": 2.499657154083252
    },
    {
      "auxiliary_loss_clip": 0.01059711,
      "auxiliary_loss_mlp": 0.01029783,
      "balance_loss_clip": 1.0336647,
      "balance_loss_mlp": 1.01691997,
      "epoch": 0.853028708853149,
      "flos": 29607665748480.0,
      "grad_norm": 1.9459516537144488,
      "language_loss": 0.76840281,
      "learning_rate": 2.2225880483803005e-07,
      "loss": 0.7892977,
      "num_input_tokens_seen": 306005220,
      "step": 14188,
      "time_per_iteration": 2.625091075897217
    },
    {
      "auxiliary_loss_clip": 0.01094711,
      "auxiliary_loss_mlp": 0.01029306,
      "balance_loss_clip": 1.03553605,
      "balance_loss_mlp": 1.01643729,
      "epoch": 0.8530888321058169,
      "flos": 26353153042560.0,
      "grad_norm": 1.7427813822156295,
      "language_loss": 0.78513217,
      "learning_rate": 2.2208040342933932e-07,
      "loss": 0.8063724,
      "num_input_tokens_seen": 306023785,
      "step": 14189,
      "time_per_iteration": 2.522097110748291
    },
    {
      "auxiliary_loss_clip": 0.0108069,
      "auxiliary_loss_mlp": 0.01031726,
      "balance_loss_clip": 1.03322196,
      "balance_loss_mlp": 1.01977539,
      "epoch": 0.8531489553584849,
      "flos": 20521979141760.0,
      "grad_norm": 1.966226015971542,
      "language_loss": 0.79718214,
      "learning_rate": 2.2190206943953793e-07,
      "loss": 0.81830633,
      "num_input_tokens_seen": 306041600,
      "step": 14190,
      "time_per_iteration": 2.4943337440490723
    },
    {
      "auxiliary_loss_clip": 0.01054129,
      "auxiliary_loss_mlp": 0.01028791,
      "balance_loss_clip": 1.03474152,
      "balance_loss_mlp": 1.01674485,
      "epoch": 0.8532090786111529,
      "flos": 20704297599360.0,
      "grad_norm": 1.8349390249070154,
      "language_loss": 0.75963438,
      "learning_rate": 2.2172380287538894e-07,
      "loss": 0.78046364,
      "num_input_tokens_seen": 306060345,
      "step": 14191,
      "time_per_iteration": 2.5829691886901855
    },
    {
      "auxiliary_loss_clip": 0.01093058,
      "auxiliary_loss_mlp": 0.01028857,
      "balance_loss_clip": 1.03618312,
      "balance_loss_mlp": 1.01716876,
      "epoch": 0.8532692018638208,
      "flos": 19828903242240.0,
      "grad_norm": 1.730621034631871,
      "language_loss": 0.68826729,
      "learning_rate": 2.2154560374365073e-07,
      "loss": 0.70948648,
      "num_input_tokens_seen": 306078285,
      "step": 14192,
      "time_per_iteration": 2.4670867919921875
    },
    {
      "auxiliary_loss_clip": 0.01096736,
      "auxiliary_loss_mlp": 0.01034719,
      "balance_loss_clip": 1.0368526,
      "balance_loss_mlp": 1.02124858,
      "epoch": 0.8533293251164888,
      "flos": 20996790048000.0,
      "grad_norm": 2.2710422246857633,
      "language_loss": 0.6274085,
      "learning_rate": 2.2136747205108164e-07,
      "loss": 0.64872307,
      "num_input_tokens_seen": 306093760,
      "step": 14193,
      "time_per_iteration": 2.5021474361419678
    },
    {
      "auxiliary_loss_clip": 0.01083246,
      "auxiliary_loss_mlp": 0.01030261,
      "balance_loss_clip": 1.03396463,
      "balance_loss_mlp": 1.01824522,
      "epoch": 0.8533894483691568,
      "flos": 22419606654720.0,
      "grad_norm": 1.7280750495027208,
      "language_loss": 0.76807207,
      "learning_rate": 2.211894078044365e-07,
      "loss": 0.78920716,
      "num_input_tokens_seen": 306112595,
      "step": 14194,
      "time_per_iteration": 2.5220136642456055
    },
    {
      "auxiliary_loss_clip": 0.01103912,
      "auxiliary_loss_mlp": 0.01026245,
      "balance_loss_clip": 1.03500044,
      "balance_loss_mlp": 1.01441956,
      "epoch": 0.8534495716218248,
      "flos": 21616536332160.0,
      "grad_norm": 1.6290172482773155,
      "language_loss": 0.69489813,
      "learning_rate": 2.2101141101046705e-07,
      "loss": 0.71619976,
      "num_input_tokens_seen": 306131800,
      "step": 14195,
      "time_per_iteration": 2.4679746627807617
    },
    {
      "auxiliary_loss_clip": 0.01077788,
      "auxiliary_loss_mlp": 0.01032555,
      "balance_loss_clip": 1.03439927,
      "balance_loss_mlp": 1.01980543,
      "epoch": 0.8535096948744927,
      "flos": 22346277039360.0,
      "grad_norm": 2.115961317791182,
      "language_loss": 0.85969949,
      "learning_rate": 2.2083348167592343e-07,
      "loss": 0.88080293,
      "num_input_tokens_seen": 306150590,
      "step": 14196,
      "time_per_iteration": 2.5139529705047607
    },
    {
      "auxiliary_loss_clip": 0.01013619,
      "auxiliary_loss_mlp": 0.00999193,
      "balance_loss_clip": 1.0109235,
      "balance_loss_mlp": 0.99797118,
      "epoch": 0.8535698181271607,
      "flos": 52762507891200.0,
      "grad_norm": 0.7655985569086489,
      "language_loss": 0.55062568,
      "learning_rate": 2.2065561980755243e-07,
      "loss": 0.57075387,
      "num_input_tokens_seen": 306205850,
      "step": 14197,
      "time_per_iteration": 3.0742886066436768
    },
    {
      "auxiliary_loss_clip": 0.01068231,
      "auxiliary_loss_mlp": 0.00782198,
      "balance_loss_clip": 1.03330529,
      "balance_loss_mlp": 1.00859094,
      "epoch": 0.8536299413798286,
      "flos": 19062892776960.0,
      "grad_norm": 1.479172641609595,
      "language_loss": 0.81453049,
      "learning_rate": 2.2047782541209826e-07,
      "loss": 0.83303475,
      "num_input_tokens_seen": 306225220,
      "step": 14198,
      "time_per_iteration": 2.56241512298584
    },
    {
      "auxiliary_loss_clip": 0.01103288,
      "auxiliary_loss_mlp": 0.01028865,
      "balance_loss_clip": 1.03519845,
      "balance_loss_mlp": 1.01819599,
      "epoch": 0.8536900646324966,
      "flos": 49344743871360.0,
      "grad_norm": 1.430285643911617,
      "language_loss": 0.68343496,
      "learning_rate": 2.203000984963035e-07,
      "loss": 0.7047565,
      "num_input_tokens_seen": 306249865,
      "step": 14199,
      "time_per_iteration": 2.7108147144317627
    },
    {
      "auxiliary_loss_clip": 0.0106516,
      "auxiliary_loss_mlp": 0.01026431,
      "balance_loss_clip": 1.03130221,
      "balance_loss_mlp": 1.01494575,
      "epoch": 0.8537501878851645,
      "flos": 21762333636480.0,
      "grad_norm": 1.5183751627010416,
      "language_loss": 0.8619169,
      "learning_rate": 2.201224390669072e-07,
      "loss": 0.88283277,
      "num_input_tokens_seen": 306270215,
      "step": 14200,
      "time_per_iteration": 2.570683479309082
    },
    {
      "auxiliary_loss_clip": 0.01072021,
      "auxiliary_loss_mlp": 0.01027624,
      "balance_loss_clip": 1.03360462,
      "balance_loss_mlp": 1.01704431,
      "epoch": 0.8538103111378326,
      "flos": 22269176496000.0,
      "grad_norm": 1.8694719930850656,
      "language_loss": 0.78267598,
      "learning_rate": 2.1994484713064666e-07,
      "loss": 0.80367243,
      "num_input_tokens_seen": 306288960,
      "step": 14201,
      "time_per_iteration": 2.551252841949463
    },
    {
      "auxiliary_loss_clip": 0.01075997,
      "auxiliary_loss_mlp": 0.01028007,
      "balance_loss_clip": 1.0346024,
      "balance_loss_mlp": 1.01675391,
      "epoch": 0.8538704343905005,
      "flos": 20303929630080.0,
      "grad_norm": 1.748905396646379,
      "language_loss": 0.6884315,
      "learning_rate": 2.19767322694256e-07,
      "loss": 0.70947152,
      "num_input_tokens_seen": 306308735,
      "step": 14202,
      "time_per_iteration": 2.5191595554351807
    },
    {
      "auxiliary_loss_clip": 0.01093129,
      "auxiliary_loss_mlp": 0.01029801,
      "balance_loss_clip": 1.03502226,
      "balance_loss_mlp": 1.01805925,
      "epoch": 0.8539305576431685,
      "flos": 24755164784640.0,
      "grad_norm": 1.5069387146338764,
      "language_loss": 0.79944849,
      "learning_rate": 2.195898657644666e-07,
      "loss": 0.82067776,
      "num_input_tokens_seen": 306329015,
      "step": 14203,
      "time_per_iteration": 2.511061668395996
    },
    {
      "auxiliary_loss_clip": 0.0108584,
      "auxiliary_loss_mlp": 0.01031145,
      "balance_loss_clip": 1.03394365,
      "balance_loss_mlp": 1.01838946,
      "epoch": 0.8539906808958365,
      "flos": 26687625511680.0,
      "grad_norm": 1.9318096386198393,
      "language_loss": 0.65755635,
      "learning_rate": 2.1941247634800808e-07,
      "loss": 0.6787262,
      "num_input_tokens_seen": 306349085,
      "step": 14204,
      "time_per_iteration": 2.564866542816162
    },
    {
      "auxiliary_loss_clip": 0.01105588,
      "auxiliary_loss_mlp": 0.01032723,
      "balance_loss_clip": 1.03594732,
      "balance_loss_mlp": 1.02040303,
      "epoch": 0.8540508041485044,
      "flos": 13365521038080.0,
      "grad_norm": 2.0118845295996337,
      "language_loss": 0.59230334,
      "learning_rate": 2.1923515445160667e-07,
      "loss": 0.61368644,
      "num_input_tokens_seen": 306365385,
      "step": 14205,
      "time_per_iteration": 2.4015893936157227
    },
    {
      "auxiliary_loss_clip": 0.01081129,
      "auxiliary_loss_mlp": 0.01027511,
      "balance_loss_clip": 1.03523052,
      "balance_loss_mlp": 1.01574504,
      "epoch": 0.8541109274011724,
      "flos": 32780876019840.0,
      "grad_norm": 2.0118696105433345,
      "language_loss": 0.72009254,
      "learning_rate": 2.1905790008198655e-07,
      "loss": 0.74117887,
      "num_input_tokens_seen": 306384585,
      "step": 14206,
      "time_per_iteration": 2.607109308242798
    },
    {
      "auxiliary_loss_clip": 0.0109511,
      "auxiliary_loss_mlp": 0.01029399,
      "balance_loss_clip": 1.03618598,
      "balance_loss_mlp": 1.01748967,
      "epoch": 0.8541710506538404,
      "flos": 17639286071040.0,
      "grad_norm": 3.019384669411787,
      "language_loss": 0.76145703,
      "learning_rate": 2.1888071324586987e-07,
      "loss": 0.78270209,
      "num_input_tokens_seen": 306401565,
      "step": 14207,
      "time_per_iteration": 2.4479048252105713
    },
    {
      "auxiliary_loss_clip": 0.01105243,
      "auxiliary_loss_mlp": 0.01031546,
      "balance_loss_clip": 1.03614855,
      "balance_loss_mlp": 1.01888609,
      "epoch": 0.8542311739065084,
      "flos": 20263062931200.0,
      "grad_norm": 1.7481748553845422,
      "language_loss": 0.85042918,
      "learning_rate": 2.1870359394997485e-07,
      "loss": 0.87179708,
      "num_input_tokens_seen": 306419995,
      "step": 14208,
      "time_per_iteration": 2.4447214603424072
    },
    {
      "auxiliary_loss_clip": 0.01084102,
      "auxiliary_loss_mlp": 0.01034331,
      "balance_loss_clip": 1.03413177,
      "balance_loss_mlp": 1.02287495,
      "epoch": 0.8542912971591763,
      "flos": 17785657992960.0,
      "grad_norm": 1.8643412865847393,
      "language_loss": 0.6645546,
      "learning_rate": 2.1852654220101785e-07,
      "loss": 0.68573892,
      "num_input_tokens_seen": 306439240,
      "step": 14209,
      "time_per_iteration": 2.477638006210327
    },
    {
      "auxiliary_loss_clip": 0.01059118,
      "auxiliary_loss_mlp": 0.01027599,
      "balance_loss_clip": 1.03400397,
      "balance_loss_mlp": 1.01600027,
      "epoch": 0.8543514204118443,
      "flos": 26979507429120.0,
      "grad_norm": 1.8274511630734058,
      "language_loss": 0.70592278,
      "learning_rate": 2.1834955800571287e-07,
      "loss": 0.72678995,
      "num_input_tokens_seen": 306458425,
      "step": 14210,
      "time_per_iteration": 2.6127350330352783
    },
    {
      "auxiliary_loss_clip": 0.01079038,
      "auxiliary_loss_mlp": 0.01028882,
      "balance_loss_clip": 1.03498983,
      "balance_loss_mlp": 1.01733088,
      "epoch": 0.8544115436645122,
      "flos": 24024598064640.0,
      "grad_norm": 1.3182432568700784,
      "language_loss": 0.70362633,
      "learning_rate": 2.1817264137077141e-07,
      "loss": 0.72470552,
      "num_input_tokens_seen": 306477210,
      "step": 14211,
      "time_per_iteration": 2.5296895503997803
    },
    {
      "auxiliary_loss_clip": 0.01083924,
      "auxiliary_loss_mlp": 0.01030368,
      "balance_loss_clip": 1.03495336,
      "balance_loss_mlp": 1.01838183,
      "epoch": 0.8544716669171802,
      "flos": 16617986668800.0,
      "grad_norm": 2.182151183848599,
      "language_loss": 0.81616569,
      "learning_rate": 2.1799579230290166e-07,
      "loss": 0.83730865,
      "num_input_tokens_seen": 306495820,
      "step": 14212,
      "time_per_iteration": 2.5101587772369385
    },
    {
      "auxiliary_loss_clip": 0.0107117,
      "auxiliary_loss_mlp": 0.01033967,
      "balance_loss_clip": 1.03285813,
      "balance_loss_mlp": 1.02067554,
      "epoch": 0.8545317901698481,
      "flos": 40005779489280.0,
      "grad_norm": 1.8215128774697689,
      "language_loss": 0.66758728,
      "learning_rate": 2.178190108088105e-07,
      "loss": 0.68863863,
      "num_input_tokens_seen": 306516420,
      "step": 14213,
      "time_per_iteration": 2.7067058086395264
    },
    {
      "auxiliary_loss_clip": 0.01102012,
      "auxiliary_loss_mlp": 0.01028387,
      "balance_loss_clip": 1.03441882,
      "balance_loss_mlp": 1.01685357,
      "epoch": 0.8545919134225162,
      "flos": 19902520166400.0,
      "grad_norm": 2.5710208959978527,
      "language_loss": 0.78354704,
      "learning_rate": 2.1764229689520098e-07,
      "loss": 0.804851,
      "num_input_tokens_seen": 306534785,
      "step": 14214,
      "time_per_iteration": 3.8267557621002197
    },
    {
      "auxiliary_loss_clip": 0.01084616,
      "auxiliary_loss_mlp": 0.0102879,
      "balance_loss_clip": 1.03325379,
      "balance_loss_mlp": 1.01580238,
      "epoch": 0.8546520366751841,
      "flos": 18952970181120.0,
      "grad_norm": 2.341442861165409,
      "language_loss": 0.66497749,
      "learning_rate": 2.1746565056877397e-07,
      "loss": 0.68611157,
      "num_input_tokens_seen": 306552440,
      "step": 14215,
      "time_per_iteration": 3.9914908409118652
    },
    {
      "auxiliary_loss_clip": 0.01105187,
      "auxiliary_loss_mlp": 0.01026956,
      "balance_loss_clip": 1.03667164,
      "balance_loss_mlp": 1.01541078,
      "epoch": 0.8547121599278521,
      "flos": 35621445415680.0,
      "grad_norm": 2.1908378435195095,
      "language_loss": 0.62670398,
      "learning_rate": 2.172890718362279e-07,
      "loss": 0.64802539,
      "num_input_tokens_seen": 306573600,
      "step": 14216,
      "time_per_iteration": 2.5691967010498047
    },
    {
      "auxiliary_loss_clip": 0.01068466,
      "auxiliary_loss_mlp": 0.01031958,
      "balance_loss_clip": 1.03284872,
      "balance_loss_mlp": 1.0201323,
      "epoch": 0.8547722831805201,
      "flos": 16910048154240.0,
      "grad_norm": 1.6181862235819566,
      "language_loss": 0.65354621,
      "learning_rate": 2.17112560704259e-07,
      "loss": 0.67455047,
      "num_input_tokens_seen": 306592840,
      "step": 14217,
      "time_per_iteration": 2.5615954399108887
    },
    {
      "auxiliary_loss_clip": 0.01089684,
      "auxiliary_loss_mlp": 0.01029546,
      "balance_loss_clip": 1.03622019,
      "balance_loss_mlp": 1.01839352,
      "epoch": 0.854832406433188,
      "flos": 23002616304000.0,
      "grad_norm": 1.4659749005987077,
      "language_loss": 0.64876676,
      "learning_rate": 2.1693611717956072e-07,
      "loss": 0.66995907,
      "num_input_tokens_seen": 306613210,
      "step": 14218,
      "time_per_iteration": 3.9467499256134033
    },
    {
      "auxiliary_loss_clip": 0.01089866,
      "auxiliary_loss_mlp": 0.01033727,
      "balance_loss_clip": 1.03288603,
      "balance_loss_mlp": 1.02121592,
      "epoch": 0.854892529685856,
      "flos": 20412595249920.0,
      "grad_norm": 1.8758218508031705,
      "language_loss": 0.70120585,
      "learning_rate": 2.167597412688238e-07,
      "loss": 0.72244179,
      "num_input_tokens_seen": 306631620,
      "step": 14219,
      "time_per_iteration": 2.4735360145568848
    },
    {
      "auxiliary_loss_clip": 0.01081363,
      "auxiliary_loss_mlp": 0.01033288,
      "balance_loss_clip": 1.0328095,
      "balance_loss_mlp": 1.02136731,
      "epoch": 0.854952652938524,
      "flos": 16398716094720.0,
      "grad_norm": 2.4833704075125205,
      "language_loss": 0.67178315,
      "learning_rate": 2.1658343297873549e-07,
      "loss": 0.69292963,
      "num_input_tokens_seen": 306646695,
      "step": 14220,
      "time_per_iteration": 2.47493314743042
    },
    {
      "auxiliary_loss_clip": 0.01100037,
      "auxiliary_loss_mlp": 0.01026801,
      "balance_loss_clip": 1.03446865,
      "balance_loss_mlp": 1.01569033,
      "epoch": 0.855012776191192,
      "flos": 21178677542400.0,
      "grad_norm": 1.98314001041562,
      "language_loss": 0.71685874,
      "learning_rate": 2.164071923159827e-07,
      "loss": 0.73812717,
      "num_input_tokens_seen": 306665465,
      "step": 14221,
      "time_per_iteration": 2.432995557785034
    },
    {
      "auxiliary_loss_clip": 0.0107148,
      "auxiliary_loss_mlp": 0.01036388,
      "balance_loss_clip": 1.03306842,
      "balance_loss_mlp": 1.02410364,
      "epoch": 0.8550728994438599,
      "flos": 26140993361280.0,
      "grad_norm": 2.0025949116107116,
      "language_loss": 0.60234261,
      "learning_rate": 2.1623101928724763e-07,
      "loss": 0.62342131,
      "num_input_tokens_seen": 306685950,
      "step": 14222,
      "time_per_iteration": 2.584054946899414
    },
    {
      "auxiliary_loss_clip": 0.01075521,
      "auxiliary_loss_mlp": 0.01032011,
      "balance_loss_clip": 1.03198457,
      "balance_loss_mlp": 1.01998866,
      "epoch": 0.8551330226965279,
      "flos": 22786793435520.0,
      "grad_norm": 1.7762996236042858,
      "language_loss": 0.83924383,
      "learning_rate": 2.1605491389921093e-07,
      "loss": 0.86031914,
      "num_input_tokens_seen": 306705740,
      "step": 14223,
      "time_per_iteration": 2.536123037338257
    },
    {
      "auxiliary_loss_clip": 0.01090949,
      "auxiliary_loss_mlp": 0.01031447,
      "balance_loss_clip": 1.03513265,
      "balance_loss_mlp": 1.02022338,
      "epoch": 0.8551931459491958,
      "flos": 22419032037120.0,
      "grad_norm": 1.6147154175218947,
      "language_loss": 0.7390101,
      "learning_rate": 2.158788761585515e-07,
      "loss": 0.760234,
      "num_input_tokens_seen": 306725065,
      "step": 14224,
      "time_per_iteration": 3.9044668674468994
    },
    {
      "auxiliary_loss_clip": 0.01076623,
      "auxiliary_loss_mlp": 0.0078513,
      "balance_loss_clip": 1.03293705,
      "balance_loss_mlp": 1.00929618,
      "epoch": 0.8552532692018638,
      "flos": 19573183342080.0,
      "grad_norm": 2.26535762279399,
      "language_loss": 0.75602531,
      "learning_rate": 2.1570290607194307e-07,
      "loss": 0.77464283,
      "num_input_tokens_seen": 306743630,
      "step": 14225,
      "time_per_iteration": 2.4855306148529053
    },
    {
      "auxiliary_loss_clip": 0.01042415,
      "auxiliary_loss_mlp": 0.01040181,
      "balance_loss_clip": 1.03218937,
      "balance_loss_mlp": 1.02865338,
      "epoch": 0.8553133924545318,
      "flos": 26432767537920.0,
      "grad_norm": 1.6038249932643707,
      "language_loss": 0.7730273,
      "learning_rate": 2.1552700364605925e-07,
      "loss": 0.79385328,
      "num_input_tokens_seen": 306763105,
      "step": 14226,
      "time_per_iteration": 2.6719436645507812
    },
    {
      "auxiliary_loss_clip": 0.01106971,
      "auxiliary_loss_mlp": 0.01035252,
      "balance_loss_clip": 1.03560424,
      "balance_loss_mlp": 1.0225265,
      "epoch": 0.8553735157071998,
      "flos": 16362446336640.0,
      "grad_norm": 2.0052042539839694,
      "language_loss": 0.54224384,
      "learning_rate": 2.153511688875702e-07,
      "loss": 0.56366611,
      "num_input_tokens_seen": 306779875,
      "step": 14227,
      "time_per_iteration": 2.4174771308898926
    },
    {
      "auxiliary_loss_clip": 0.01072389,
      "auxiliary_loss_mlp": 0.00781978,
      "balance_loss_clip": 1.03412163,
      "balance_loss_mlp": 1.00854897,
      "epoch": 0.8554336389598677,
      "flos": 20887334328960.0,
      "grad_norm": 2.944814921505782,
      "language_loss": 0.65535402,
      "learning_rate": 2.151754018031442e-07,
      "loss": 0.67389768,
      "num_input_tokens_seen": 306800015,
      "step": 14228,
      "time_per_iteration": 2.5504443645477295
    },
    {
      "auxiliary_loss_clip": 0.01072604,
      "auxiliary_loss_mlp": 0.01033922,
      "balance_loss_clip": 1.03615665,
      "balance_loss_mlp": 1.02151275,
      "epoch": 0.8554937622125357,
      "flos": 21284721469440.0,
      "grad_norm": 2.0956645385979096,
      "language_loss": 0.73854029,
      "learning_rate": 2.1499970239944542e-07,
      "loss": 0.75960553,
      "num_input_tokens_seen": 306814160,
      "step": 14229,
      "time_per_iteration": 2.518414258956909
    },
    {
      "auxiliary_loss_clip": 0.01090552,
      "auxiliary_loss_mlp": 0.0102684,
      "balance_loss_clip": 1.03385735,
      "balance_loss_mlp": 1.01593208,
      "epoch": 0.8555538854652037,
      "flos": 22413178120320.0,
      "grad_norm": 1.6693394487657833,
      "language_loss": 0.72847921,
      "learning_rate": 2.1482407068313724e-07,
      "loss": 0.74965316,
      "num_input_tokens_seen": 306833310,
      "step": 14230,
      "time_per_iteration": 2.4915170669555664
    },
    {
      "auxiliary_loss_clip": 0.01089916,
      "auxiliary_loss_mlp": 0.01028043,
      "balance_loss_clip": 1.03360891,
      "balance_loss_mlp": 1.01619983,
      "epoch": 0.8556140087178716,
      "flos": 20193719725440.0,
      "grad_norm": 1.7212542203417283,
      "language_loss": 0.82500911,
      "learning_rate": 2.1464850666087897e-07,
      "loss": 0.84618866,
      "num_input_tokens_seen": 306851345,
      "step": 14231,
      "time_per_iteration": 2.453489303588867
    },
    {
      "auxiliary_loss_clip": 0.0109564,
      "auxiliary_loss_mlp": 0.01033287,
      "balance_loss_clip": 1.03570879,
      "balance_loss_mlp": 1.02052522,
      "epoch": 0.8556741319705397,
      "flos": 22638123043200.0,
      "grad_norm": 1.804602863757462,
      "language_loss": 0.6762082,
      "learning_rate": 2.1447301033932796e-07,
      "loss": 0.69749749,
      "num_input_tokens_seen": 306871040,
      "step": 14232,
      "time_per_iteration": 2.5174789428710938
    },
    {
      "auxiliary_loss_clip": 0.01083545,
      "auxiliary_loss_mlp": 0.01031022,
      "balance_loss_clip": 1.03523064,
      "balance_loss_mlp": 1.01868951,
      "epoch": 0.8557342552232076,
      "flos": 23549320281600.0,
      "grad_norm": 1.4249017830594095,
      "language_loss": 0.67252183,
      "learning_rate": 2.1429758172513955e-07,
      "loss": 0.69366759,
      "num_input_tokens_seen": 306891625,
      "step": 14233,
      "time_per_iteration": 2.533970355987549
    },
    {
      "auxiliary_loss_clip": 0.01090073,
      "auxiliary_loss_mlp": 0.01029811,
      "balance_loss_clip": 1.03330576,
      "balance_loss_mlp": 1.01844454,
      "epoch": 0.8557943784758756,
      "flos": 19609884063360.0,
      "grad_norm": 1.6701839511822696,
      "language_loss": 0.76428008,
      "learning_rate": 2.1412222082496556e-07,
      "loss": 0.78547895,
      "num_input_tokens_seen": 306910020,
      "step": 14234,
      "time_per_iteration": 2.4891719818115234
    },
    {
      "auxiliary_loss_clip": 0.01004668,
      "auxiliary_loss_mlp": 0.01016587,
      "balance_loss_clip": 1.01053071,
      "balance_loss_mlp": 1.01521015,
      "epoch": 0.8558545017285435,
      "flos": 70641891446400.0,
      "grad_norm": 0.7541465257032627,
      "language_loss": 0.58010322,
      "learning_rate": 2.1394692764545684e-07,
      "loss": 0.60031575,
      "num_input_tokens_seen": 306969505,
      "step": 14235,
      "time_per_iteration": 3.1210572719573975
    },
    {
      "auxiliary_loss_clip": 0.01012124,
      "auxiliary_loss_mlp": 0.00999553,
      "balance_loss_clip": 1.00877607,
      "balance_loss_mlp": 0.99835455,
      "epoch": 0.8559146249812115,
      "flos": 56649983086080.0,
      "grad_norm": 0.7874319510846941,
      "language_loss": 0.56692219,
      "learning_rate": 2.1377170219325858e-07,
      "loss": 0.58703887,
      "num_input_tokens_seen": 307027710,
      "step": 14236,
      "time_per_iteration": 3.034252166748047
    },
    {
      "auxiliary_loss_clip": 0.01081086,
      "auxiliary_loss_mlp": 0.01032947,
      "balance_loss_clip": 1.03403711,
      "balance_loss_mlp": 1.02117562,
      "epoch": 0.8559747482338794,
      "flos": 22888240421760.0,
      "grad_norm": 1.767999131162567,
      "language_loss": 0.7044909,
      "learning_rate": 2.1359654447501673e-07,
      "loss": 0.72563124,
      "num_input_tokens_seen": 307045515,
      "step": 14237,
      "time_per_iteration": 2.5484769344329834
    },
    {
      "auxiliary_loss_clip": 0.01079988,
      "auxiliary_loss_mlp": 0.01029127,
      "balance_loss_clip": 1.03242695,
      "balance_loss_mlp": 1.01788568,
      "epoch": 0.8560348714865474,
      "flos": 22601925112320.0,
      "grad_norm": 2.072055720007472,
      "language_loss": 0.63201374,
      "learning_rate": 2.1342145449737314e-07,
      "loss": 0.6531049,
      "num_input_tokens_seen": 307064470,
      "step": 14238,
      "time_per_iteration": 2.531428337097168
    },
    {
      "auxiliary_loss_clip": 0.01097405,
      "auxiliary_loss_mlp": 0.01032948,
      "balance_loss_clip": 1.03291106,
      "balance_loss_mlp": 1.02288055,
      "epoch": 0.8560949947392154,
      "flos": 17931455297280.0,
      "grad_norm": 1.4404118788031197,
      "language_loss": 0.69365078,
      "learning_rate": 2.1324643226696648e-07,
      "loss": 0.71495426,
      "num_input_tokens_seen": 307083900,
      "step": 14239,
      "time_per_iteration": 2.471531629562378
    },
    {
      "auxiliary_loss_clip": 0.01106801,
      "auxiliary_loss_mlp": 0.01035742,
      "balance_loss_clip": 1.03567004,
      "balance_loss_mlp": 1.02383351,
      "epoch": 0.8561551179918834,
      "flos": 31026208636800.0,
      "grad_norm": 1.9442797939117318,
      "language_loss": 0.66547358,
      "learning_rate": 2.1307147779043455e-07,
      "loss": 0.68689907,
      "num_input_tokens_seen": 307104590,
      "step": 14240,
      "time_per_iteration": 2.51489520072937
    },
    {
      "auxiliary_loss_clip": 0.01063751,
      "auxiliary_loss_mlp": 0.01037402,
      "balance_loss_clip": 1.03252459,
      "balance_loss_mlp": 1.02359152,
      "epoch": 0.8562152412445513,
      "flos": 30665198995200.0,
      "grad_norm": 1.5014886883222083,
      "language_loss": 0.62215006,
      "learning_rate": 2.1289659107441182e-07,
      "loss": 0.64316154,
      "num_input_tokens_seen": 307125580,
      "step": 14241,
      "time_per_iteration": 2.616015911102295
    },
    {
      "auxiliary_loss_clip": 0.01107094,
      "auxiliary_loss_mlp": 0.01035697,
      "balance_loss_clip": 1.03466296,
      "balance_loss_mlp": 1.02318025,
      "epoch": 0.8562753644972193,
      "flos": 31576144838400.0,
      "grad_norm": 1.6925696368508898,
      "language_loss": 0.74634409,
      "learning_rate": 2.1272177212552855e-07,
      "loss": 0.76777196,
      "num_input_tokens_seen": 307147625,
      "step": 14242,
      "time_per_iteration": 2.5192947387695312
    },
    {
      "auxiliary_loss_clip": 0.01035143,
      "auxiliary_loss_mlp": 0.01046928,
      "balance_loss_clip": 1.03483367,
      "balance_loss_mlp": 1.03340364,
      "epoch": 0.8563354877498872,
      "flos": 26213640618240.0,
      "grad_norm": 1.9780897043750816,
      "language_loss": 0.76450747,
      "learning_rate": 2.1254702095041498e-07,
      "loss": 0.78532815,
      "num_input_tokens_seen": 307164665,
      "step": 14243,
      "time_per_iteration": 2.747976541519165
    },
    {
      "auxiliary_loss_clip": 0.01081127,
      "auxiliary_loss_mlp": 0.00781612,
      "balance_loss_clip": 1.03476501,
      "balance_loss_mlp": 1.00728667,
      "epoch": 0.8563956110025552,
      "flos": 24134341092480.0,
      "grad_norm": 2.06046157347618,
      "language_loss": 0.68060833,
      "learning_rate": 2.123723375556974e-07,
      "loss": 0.69923574,
      "num_input_tokens_seen": 307182530,
      "step": 14244,
      "time_per_iteration": 2.6322810649871826
    },
    {
      "auxiliary_loss_clip": 0.01020392,
      "auxiliary_loss_mlp": 0.01001587,
      "balance_loss_clip": 1.00670207,
      "balance_loss_mlp": 1.00050199,
      "epoch": 0.8564557342552233,
      "flos": 56271986311680.0,
      "grad_norm": 0.7548785631762144,
      "language_loss": 0.58441043,
      "learning_rate": 2.1219772194800046e-07,
      "loss": 0.60463023,
      "num_input_tokens_seen": 307241240,
      "step": 14245,
      "time_per_iteration": 3.0466201305389404
    },
    {
      "auxiliary_loss_clip": 0.01095695,
      "auxiliary_loss_mlp": 0.01032081,
      "balance_loss_clip": 1.03511477,
      "balance_loss_mlp": 1.01981497,
      "epoch": 0.8565158575078912,
      "flos": 23440618748160.0,
      "grad_norm": 1.7485144631659006,
      "language_loss": 0.77790177,
      "learning_rate": 2.1202317413394488e-07,
      "loss": 0.79917955,
      "num_input_tokens_seen": 307261485,
      "step": 14246,
      "time_per_iteration": 2.49139142036438
    },
    {
      "auxiliary_loss_clip": 0.01077203,
      "auxiliary_loss_mlp": 0.01027851,
      "balance_loss_clip": 1.02971983,
      "balance_loss_mlp": 1.01585269,
      "epoch": 0.8565759807605592,
      "flos": 20375930442240.0,
      "grad_norm": 2.9165733779687795,
      "language_loss": 0.81340325,
      "learning_rate": 2.1184869412014938e-07,
      "loss": 0.83445382,
      "num_input_tokens_seen": 307279160,
      "step": 14247,
      "time_per_iteration": 2.521484375
    },
    {
      "auxiliary_loss_clip": 0.01081041,
      "auxiliary_loss_mlp": 0.01030822,
      "balance_loss_clip": 1.03401113,
      "balance_loss_mlp": 1.01885378,
      "epoch": 0.8566361040132271,
      "flos": 18807101049600.0,
      "grad_norm": 1.7115899255142006,
      "language_loss": 0.77373379,
      "learning_rate": 2.1167428191323112e-07,
      "loss": 0.79485238,
      "num_input_tokens_seen": 307297920,
      "step": 14248,
      "time_per_iteration": 2.4825663566589355
    },
    {
      "auxiliary_loss_clip": 0.01054106,
      "auxiliary_loss_mlp": 0.01043294,
      "balance_loss_clip": 1.03030658,
      "balance_loss_mlp": 1.02904248,
      "epoch": 0.8566962272658951,
      "flos": 24535355506560.0,
      "grad_norm": 1.65538887039535,
      "language_loss": 0.77483851,
      "learning_rate": 2.1149993751980278e-07,
      "loss": 0.79581255,
      "num_input_tokens_seen": 307318320,
      "step": 14249,
      "time_per_iteration": 2.5996360778808594
    },
    {
      "auxiliary_loss_clip": 0.01077139,
      "auxiliary_loss_mlp": 0.01035319,
      "balance_loss_clip": 1.03209794,
      "balance_loss_mlp": 1.02339244,
      "epoch": 0.856756350518563,
      "flos": 23178506227200.0,
      "grad_norm": 2.0090874099187572,
      "language_loss": 0.78519136,
      "learning_rate": 2.1132566094647597e-07,
      "loss": 0.80631602,
      "num_input_tokens_seen": 307336720,
      "step": 14250,
      "time_per_iteration": 2.5110654830932617
    },
    {
      "auxiliary_loss_clip": 0.01080047,
      "auxiliary_loss_mlp": 0.01028554,
      "balance_loss_clip": 1.03447545,
      "balance_loss_mlp": 1.01764619,
      "epoch": 0.856816473771231,
      "flos": 20808581760000.0,
      "grad_norm": 1.7005950997008479,
      "language_loss": 0.79692328,
      "learning_rate": 2.1115145219985942e-07,
      "loss": 0.81800926,
      "num_input_tokens_seen": 307354120,
      "step": 14251,
      "time_per_iteration": 2.5197596549987793
    },
    {
      "auxiliary_loss_clip": 0.01063292,
      "auxiliary_loss_mlp": 0.01028516,
      "balance_loss_clip": 1.03299737,
      "balance_loss_mlp": 1.01711941,
      "epoch": 0.856876597023899,
      "flos": 20228157889920.0,
      "grad_norm": 1.8337806562726102,
      "language_loss": 0.61841136,
      "learning_rate": 2.1097731128656005e-07,
      "loss": 0.63932943,
      "num_input_tokens_seen": 307373165,
      "step": 14252,
      "time_per_iteration": 3.9161062240600586
    },
    {
      "auxiliary_loss_clip": 0.01081735,
      "auxiliary_loss_mlp": 0.01030451,
      "balance_loss_clip": 1.03789473,
      "balance_loss_mlp": 1.01799941,
      "epoch": 0.856936720276567,
      "flos": 18296128126080.0,
      "grad_norm": 1.795150721181122,
      "language_loss": 0.69805861,
      "learning_rate": 2.1080323821317924e-07,
      "loss": 0.71918046,
      "num_input_tokens_seen": 307391000,
      "step": 14253,
      "time_per_iteration": 3.887320041656494
    },
    {
      "auxiliary_loss_clip": 0.01014419,
      "auxiliary_loss_mlp": 0.0100169,
      "balance_loss_clip": 1.01181149,
      "balance_loss_mlp": 1.00063539,
      "epoch": 0.8569968435292349,
      "flos": 69878394933120.0,
      "grad_norm": 0.7877343905341255,
      "language_loss": 0.59210628,
      "learning_rate": 2.1062923298631907e-07,
      "loss": 0.61226737,
      "num_input_tokens_seen": 307452865,
      "step": 14254,
      "time_per_iteration": 3.1756110191345215
    },
    {
      "auxiliary_loss_clip": 0.01077934,
      "auxiliary_loss_mlp": 0.01033252,
      "balance_loss_clip": 1.03265357,
      "balance_loss_mlp": 1.0206871,
      "epoch": 0.8570569667819029,
      "flos": 25848572739840.0,
      "grad_norm": 1.7684076771483068,
      "language_loss": 0.81241775,
      "learning_rate": 2.1045529561257825e-07,
      "loss": 0.83352959,
      "num_input_tokens_seen": 307471940,
      "step": 14255,
      "time_per_iteration": 2.5554826259613037
    },
    {
      "auxiliary_loss_clip": 0.01102177,
      "auxiliary_loss_mlp": 0.01024557,
      "balance_loss_clip": 1.03540564,
      "balance_loss_mlp": 1.01315415,
      "epoch": 0.8571170900345708,
      "flos": 23257115141760.0,
      "grad_norm": 2.358676612892262,
      "language_loss": 0.67082119,
      "learning_rate": 2.1028142609855126e-07,
      "loss": 0.69208848,
      "num_input_tokens_seen": 307488745,
      "step": 14256,
      "time_per_iteration": 3.879448652267456
    },
    {
      "auxiliary_loss_clip": 0.01092688,
      "auxiliary_loss_mlp": 0.01028799,
      "balance_loss_clip": 1.03487706,
      "balance_loss_mlp": 1.0175041,
      "epoch": 0.8571772132872388,
      "flos": 18917670090240.0,
      "grad_norm": 1.7469183003816589,
      "language_loss": 0.69946647,
      "learning_rate": 2.1010762445083218e-07,
      "loss": 0.72068137,
      "num_input_tokens_seen": 307506855,
      "step": 14257,
      "time_per_iteration": 2.499655246734619
    },
    {
      "auxiliary_loss_clip": 0.01068502,
      "auxiliary_loss_mlp": 0.01031453,
      "balance_loss_clip": 1.0323472,
      "balance_loss_mlp": 1.0192759,
      "epoch": 0.8572373365399069,
      "flos": 33250120318080.0,
      "grad_norm": 1.7850743949077634,
      "language_loss": 0.76754773,
      "learning_rate": 2.0993389067601197e-07,
      "loss": 0.78854728,
      "num_input_tokens_seen": 307526115,
      "step": 14258,
      "time_per_iteration": 2.6324350833892822
    },
    {
      "auxiliary_loss_clip": 0.0109283,
      "auxiliary_loss_mlp": 0.00785073,
      "balance_loss_clip": 1.0352273,
      "balance_loss_mlp": 1.01386023,
      "epoch": 0.8572974597925748,
      "flos": 23327535755520.0,
      "grad_norm": 1.6588867108431231,
      "language_loss": 0.678249,
      "learning_rate": 2.0976022478067735e-07,
      "loss": 0.69702804,
      "num_input_tokens_seen": 307545230,
      "step": 14259,
      "time_per_iteration": 2.5161831378936768
    },
    {
      "auxiliary_loss_clip": 0.01090422,
      "auxiliary_loss_mlp": 0.010294,
      "balance_loss_clip": 1.03235483,
      "balance_loss_mlp": 1.01697206,
      "epoch": 0.8573575830452428,
      "flos": 24535858296960.0,
      "grad_norm": 1.6318643798702532,
      "language_loss": 0.76877153,
      "learning_rate": 2.0958662677141437e-07,
      "loss": 0.78996974,
      "num_input_tokens_seen": 307564900,
      "step": 14260,
      "time_per_iteration": 2.498394012451172
    },
    {
      "auxiliary_loss_clip": 0.01079648,
      "auxiliary_loss_mlp": 0.01027237,
      "balance_loss_clip": 1.03406107,
      "balance_loss_mlp": 1.01486325,
      "epoch": 0.8574177062979107,
      "flos": 24165403378560.0,
      "grad_norm": 1.8538371859827818,
      "language_loss": 0.74226093,
      "learning_rate": 2.09413096654806e-07,
      "loss": 0.7633298,
      "num_input_tokens_seen": 307583500,
      "step": 14261,
      "time_per_iteration": 2.5428829193115234
    },
    {
      "auxiliary_loss_clip": 0.01097815,
      "auxiliary_loss_mlp": 0.01033236,
      "balance_loss_clip": 1.03669095,
      "balance_loss_mlp": 1.02048671,
      "epoch": 0.8574778295505787,
      "flos": 17930737025280.0,
      "grad_norm": 1.8828307950556056,
      "language_loss": 0.78455746,
      "learning_rate": 2.0923963443743276e-07,
      "loss": 0.80586803,
      "num_input_tokens_seen": 307601430,
      "step": 14262,
      "time_per_iteration": 2.4664199352264404
    },
    {
      "auxiliary_loss_clip": 0.01067702,
      "auxiliary_loss_mlp": 0.01031751,
      "balance_loss_clip": 1.03408027,
      "balance_loss_mlp": 1.02073622,
      "epoch": 0.8575379528032466,
      "flos": 21580697537280.0,
      "grad_norm": 1.5852063800594351,
      "language_loss": 0.67995787,
      "learning_rate": 2.0906624012587203e-07,
      "loss": 0.70095241,
      "num_input_tokens_seen": 307621495,
      "step": 14263,
      "time_per_iteration": 3.9632153511047363
    },
    {
      "auxiliary_loss_clip": 0.01067727,
      "auxiliary_loss_mlp": 0.00783114,
      "balance_loss_clip": 1.03517449,
      "balance_loss_mlp": 1.00950181,
      "epoch": 0.8575980760559146,
      "flos": 21761579450880.0,
      "grad_norm": 1.429847520878928,
      "language_loss": 0.79692626,
      "learning_rate": 2.088929137266986e-07,
      "loss": 0.81543463,
      "num_input_tokens_seen": 307640840,
      "step": 14264,
      "time_per_iteration": 2.5703418254852295
    },
    {
      "auxiliary_loss_clip": 0.01068387,
      "auxiliary_loss_mlp": 0.01037375,
      "balance_loss_clip": 1.03448045,
      "balance_loss_mlp": 1.02447665,
      "epoch": 0.8576581993085826,
      "flos": 34386442047360.0,
      "grad_norm": 1.3430773218525816,
      "language_loss": 0.69331408,
      "learning_rate": 2.0871965524648582e-07,
      "loss": 0.71437168,
      "num_input_tokens_seen": 307663820,
      "step": 14265,
      "time_per_iteration": 2.6521377563476562
    },
    {
      "auxiliary_loss_clip": 0.01098823,
      "auxiliary_loss_mlp": 0.01025112,
      "balance_loss_clip": 1.03377771,
      "balance_loss_mlp": 1.01409125,
      "epoch": 0.8577183225612506,
      "flos": 23222497409280.0,
      "grad_norm": 1.686631657684879,
      "language_loss": 0.66219926,
      "learning_rate": 2.085464646918027e-07,
      "loss": 0.68343866,
      "num_input_tokens_seen": 307682385,
      "step": 14266,
      "time_per_iteration": 2.469451427459717
    },
    {
      "auxiliary_loss_clip": 0.01082936,
      "auxiliary_loss_mlp": 0.01033043,
      "balance_loss_clip": 1.03606606,
      "balance_loss_mlp": 1.02137268,
      "epoch": 0.8577784458139185,
      "flos": 28804164462720.0,
      "grad_norm": 1.5763544922898534,
      "language_loss": 0.75300914,
      "learning_rate": 2.0837334206921731e-07,
      "loss": 0.77416897,
      "num_input_tokens_seen": 307704680,
      "step": 14267,
      "time_per_iteration": 2.5644521713256836
    },
    {
      "auxiliary_loss_clip": 0.01090082,
      "auxiliary_loss_mlp": 0.01030303,
      "balance_loss_clip": 1.03467512,
      "balance_loss_mlp": 1.01922274,
      "epoch": 0.8578385690665865,
      "flos": 19755573626880.0,
      "grad_norm": 1.885405341788224,
      "language_loss": 0.87863433,
      "learning_rate": 2.082002873852946e-07,
      "loss": 0.89983821,
      "num_input_tokens_seen": 307723245,
      "step": 14268,
      "time_per_iteration": 2.5018723011016846
    },
    {
      "auxiliary_loss_clip": 0.01094005,
      "auxiliary_loss_mlp": 0.01036373,
      "balance_loss_clip": 1.03609145,
      "balance_loss_mlp": 1.02466106,
      "epoch": 0.8578986923192544,
      "flos": 20704082117760.0,
      "grad_norm": 1.871110117676521,
      "language_loss": 0.72763866,
      "learning_rate": 2.0802730064659667e-07,
      "loss": 0.74894243,
      "num_input_tokens_seen": 307742510,
      "step": 14269,
      "time_per_iteration": 2.475130081176758
    },
    {
      "auxiliary_loss_clip": 0.01094268,
      "auxiliary_loss_mlp": 0.01031458,
      "balance_loss_clip": 1.03496873,
      "balance_loss_mlp": 1.01947165,
      "epoch": 0.8579588155719224,
      "flos": 36101715189120.0,
      "grad_norm": 1.5093400776272106,
      "language_loss": 0.66560447,
      "learning_rate": 2.0785438185968252e-07,
      "loss": 0.68686175,
      "num_input_tokens_seen": 307766030,
      "step": 14270,
      "time_per_iteration": 2.6249141693115234
    },
    {
      "auxiliary_loss_clip": 0.01076669,
      "auxiliary_loss_mlp": 0.01025517,
      "balance_loss_clip": 1.03083229,
      "balance_loss_mlp": 1.01343536,
      "epoch": 0.8580189388245905,
      "flos": 22853479034880.0,
      "grad_norm": 1.7457115568552428,
      "language_loss": 0.73991978,
      "learning_rate": 2.0768153103110997e-07,
      "loss": 0.76094162,
      "num_input_tokens_seen": 307785800,
      "step": 14271,
      "time_per_iteration": 2.507211685180664
    },
    {
      "auxiliary_loss_clip": 0.00991631,
      "auxiliary_loss_mlp": 0.00760857,
      "balance_loss_clip": 1.00921059,
      "balance_loss_mlp": 0.99931669,
      "epoch": 0.8580790620772584,
      "flos": 69642104290560.0,
      "grad_norm": 0.7996552781615845,
      "language_loss": 0.59445429,
      "learning_rate": 2.0750874816743358e-07,
      "loss": 0.61197913,
      "num_input_tokens_seen": 307850995,
      "step": 14272,
      "time_per_iteration": 3.257549285888672
    },
    {
      "auxiliary_loss_clip": 0.01082613,
      "auxiliary_loss_mlp": 0.01036437,
      "balance_loss_clip": 1.03523469,
      "balance_loss_mlp": 1.02300835,
      "epoch": 0.8581391853299264,
      "flos": 13334243270400.0,
      "grad_norm": 1.7569647593561282,
      "language_loss": 0.75171065,
      "learning_rate": 2.0733603327520499e-07,
      "loss": 0.77290118,
      "num_input_tokens_seen": 307868585,
      "step": 14273,
      "time_per_iteration": 2.476038694381714
    },
    {
      "auxiliary_loss_clip": 0.01091017,
      "auxiliary_loss_mlp": 0.01030891,
      "balance_loss_clip": 1.03343344,
      "balance_loss_mlp": 1.01895177,
      "epoch": 0.8581993085825943,
      "flos": 19645651031040.0,
      "grad_norm": 1.9948670066156267,
      "language_loss": 0.82071626,
      "learning_rate": 2.0716338636097385e-07,
      "loss": 0.8419354,
      "num_input_tokens_seen": 307886820,
      "step": 14274,
      "time_per_iteration": 2.4740488529205322
    },
    {
      "auxiliary_loss_clip": 0.01019513,
      "auxiliary_loss_mlp": 0.01000916,
      "balance_loss_clip": 1.00645566,
      "balance_loss_mlp": 0.99980116,
      "epoch": 0.8582594318352623,
      "flos": 55825077294720.0,
      "grad_norm": 0.7948898245030853,
      "language_loss": 0.60849142,
      "learning_rate": 2.0699080743128672e-07,
      "loss": 0.62869573,
      "num_input_tokens_seen": 307944020,
      "step": 14275,
      "time_per_iteration": 3.1388211250305176
    },
    {
      "auxiliary_loss_clip": 0.01088746,
      "auxiliary_loss_mlp": 0.01024873,
      "balance_loss_clip": 1.03509712,
      "balance_loss_mlp": 1.01221895,
      "epoch": 0.8583195550879302,
      "flos": 24279563779200.0,
      "grad_norm": 1.954722737147305,
      "language_loss": 0.59397674,
      "learning_rate": 2.0681829649268768e-07,
      "loss": 0.6151129,
      "num_input_tokens_seen": 307961055,
      "step": 14276,
      "time_per_iteration": 2.5775272846221924
    },
    {
      "auxiliary_loss_clip": 0.0108091,
      "auxiliary_loss_mlp": 0.01030453,
      "balance_loss_clip": 1.03376591,
      "balance_loss_mlp": 1.01884842,
      "epoch": 0.8583796783405983,
      "flos": 13444129952640.0,
      "grad_norm": 1.960624972890865,
      "language_loss": 0.76271629,
      "learning_rate": 2.0664585355171838e-07,
      "loss": 0.78382993,
      "num_input_tokens_seen": 307978690,
      "step": 14277,
      "time_per_iteration": 2.456404447555542
    },
    {
      "auxiliary_loss_clip": 0.01080615,
      "auxiliary_loss_mlp": 0.01028242,
      "balance_loss_clip": 1.03358054,
      "balance_loss_mlp": 1.01631546,
      "epoch": 0.8584398015932662,
      "flos": 16180271533440.0,
      "grad_norm": 1.5219061636297668,
      "language_loss": 0.83456302,
      "learning_rate": 2.0647347861491803e-07,
      "loss": 0.85565162,
      "num_input_tokens_seen": 307995870,
      "step": 14278,
      "time_per_iteration": 2.4925217628479004
    },
    {
      "auxiliary_loss_clip": 0.01081358,
      "auxiliary_loss_mlp": 0.01031382,
      "balance_loss_clip": 1.03625965,
      "balance_loss_mlp": 1.0192945,
      "epoch": 0.8584999248459342,
      "flos": 17450431338240.0,
      "grad_norm": 2.766120064029242,
      "language_loss": 0.74424613,
      "learning_rate": 2.0630117168882366e-07,
      "loss": 0.76537347,
      "num_input_tokens_seen": 308013645,
      "step": 14279,
      "time_per_iteration": 2.4973630905151367
    },
    {
      "auxiliary_loss_clip": 0.01102133,
      "auxiliary_loss_mlp": 0.01029565,
      "balance_loss_clip": 1.03469563,
      "balance_loss_mlp": 1.01792431,
      "epoch": 0.8585600480986021,
      "flos": 23441013797760.0,
      "grad_norm": 2.39658558557705,
      "language_loss": 0.66223925,
      "learning_rate": 2.0612893277996845e-07,
      "loss": 0.68355626,
      "num_input_tokens_seen": 308032490,
      "step": 14280,
      "time_per_iteration": 2.4938676357269287
    },
    {
      "auxiliary_loss_clip": 0.01090233,
      "auxiliary_loss_mlp": 0.01029612,
      "balance_loss_clip": 1.03424621,
      "balance_loss_mlp": 1.01838875,
      "epoch": 0.8586201713512701,
      "flos": 19937927998080.0,
      "grad_norm": 1.7848614455723533,
      "language_loss": 0.62674105,
      "learning_rate": 2.0595676189488343e-07,
      "loss": 0.64793956,
      "num_input_tokens_seen": 308052110,
      "step": 14281,
      "time_per_iteration": 2.4762024879455566
    },
    {
      "auxiliary_loss_clip": 0.01079484,
      "auxiliary_loss_mlp": 0.0078411,
      "balance_loss_clip": 1.03490758,
      "balance_loss_mlp": 1.01191056,
      "epoch": 0.858680294603938,
      "flos": 15304769435520.0,
      "grad_norm": 1.8808242978369565,
      "language_loss": 0.73132908,
      "learning_rate": 2.0578465904009845e-07,
      "loss": 0.74996501,
      "num_input_tokens_seen": 308070660,
      "step": 14282,
      "time_per_iteration": 2.5190725326538086
    },
    {
      "auxiliary_loss_clip": 0.01074707,
      "auxiliary_loss_mlp": 0.01027549,
      "balance_loss_clip": 1.03109586,
      "balance_loss_mlp": 1.01635504,
      "epoch": 0.858740417856606,
      "flos": 22711237176960.0,
      "grad_norm": 1.707948690752795,
      "language_loss": 0.75689209,
      "learning_rate": 2.0561262422213832e-07,
      "loss": 0.77791464,
      "num_input_tokens_seen": 308089520,
      "step": 14283,
      "time_per_iteration": 2.504492998123169
    },
    {
      "auxiliary_loss_clip": 0.01089428,
      "auxiliary_loss_mlp": 0.01031701,
      "balance_loss_clip": 1.03309357,
      "balance_loss_mlp": 1.0195719,
      "epoch": 0.8588005411092741,
      "flos": 34054303962240.0,
      "grad_norm": 1.8555362639405433,
      "language_loss": 0.59566772,
      "learning_rate": 2.0544065744752736e-07,
      "loss": 0.61687899,
      "num_input_tokens_seen": 308111545,
      "step": 14284,
      "time_per_iteration": 2.5965402126312256
    },
    {
      "auxiliary_loss_clip": 0.01076492,
      "auxiliary_loss_mlp": 0.01028871,
      "balance_loss_clip": 1.03537869,
      "balance_loss_mlp": 1.01737356,
      "epoch": 0.858860664361942,
      "flos": 28913584268160.0,
      "grad_norm": 1.829371845947838,
      "language_loss": 0.7567181,
      "learning_rate": 2.0526875872278749e-07,
      "loss": 0.77777171,
      "num_input_tokens_seen": 308129690,
      "step": 14285,
      "time_per_iteration": 2.58284068107605
    },
    {
      "auxiliary_loss_clip": 0.01090381,
      "auxiliary_loss_mlp": 0.01030715,
      "balance_loss_clip": 1.03766406,
      "balance_loss_mlp": 1.0185616,
      "epoch": 0.85892078761461,
      "flos": 19792525743360.0,
      "grad_norm": 1.8346033553391012,
      "language_loss": 0.74366474,
      "learning_rate": 2.0509692805443524e-07,
      "loss": 0.76487565,
      "num_input_tokens_seen": 308147410,
      "step": 14286,
      "time_per_iteration": 2.489820957183838
    },
    {
      "auxiliary_loss_clip": 0.01009068,
      "auxiliary_loss_mlp": 0.00761475,
      "balance_loss_clip": 1.0074048,
      "balance_loss_mlp": 1.00033748,
      "epoch": 0.8589809108672779,
      "flos": 67106630039040.0,
      "grad_norm": 0.7862165861254445,
      "language_loss": 0.49466842,
      "learning_rate": 2.0492516544898718e-07,
      "loss": 0.51237386,
      "num_input_tokens_seen": 308204875,
      "step": 14287,
      "time_per_iteration": 3.0912904739379883
    },
    {
      "auxiliary_loss_clip": 0.01094695,
      "auxiliary_loss_mlp": 0.01032416,
      "balance_loss_clip": 1.03683281,
      "balance_loss_mlp": 1.02065051,
      "epoch": 0.8590410341199459,
      "flos": 29716259541120.0,
      "grad_norm": 1.83524751104603,
      "language_loss": 0.79105794,
      "learning_rate": 2.0475347091295704e-07,
      "loss": 0.81232905,
      "num_input_tokens_seen": 308225690,
      "step": 14288,
      "time_per_iteration": 2.551358461380005
    },
    {
      "auxiliary_loss_clip": 0.0106313,
      "auxiliary_loss_mlp": 0.01027999,
      "balance_loss_clip": 1.03534043,
      "balance_loss_mlp": 1.01581001,
      "epoch": 0.8591011573726138,
      "flos": 23987430466560.0,
      "grad_norm": 2.057116476758319,
      "language_loss": 0.81246138,
      "learning_rate": 2.045818444528553e-07,
      "loss": 0.83337259,
      "num_input_tokens_seen": 308245255,
      "step": 14289,
      "time_per_iteration": 2.585846185684204
    },
    {
      "auxiliary_loss_clip": 0.01096718,
      "auxiliary_loss_mlp": 0.01026494,
      "balance_loss_clip": 1.03780103,
      "balance_loss_mlp": 1.01476347,
      "epoch": 0.8591612806252819,
      "flos": 14428656806400.0,
      "grad_norm": 2.1805359105371904,
      "language_loss": 0.65161681,
      "learning_rate": 2.0441028607518973e-07,
      "loss": 0.67284894,
      "num_input_tokens_seen": 308261755,
      "step": 14290,
      "time_per_iteration": 2.458813428878784
    },
    {
      "auxiliary_loss_clip": 0.01083658,
      "auxiliary_loss_mlp": 0.01028905,
      "balance_loss_clip": 1.03462124,
      "balance_loss_mlp": 1.01672745,
      "epoch": 0.8592214038779498,
      "flos": 31577150419200.0,
      "grad_norm": 1.885358043451525,
      "language_loss": 0.55216897,
      "learning_rate": 2.0423879578646642e-07,
      "loss": 0.57329458,
      "num_input_tokens_seen": 308285145,
      "step": 14291,
      "time_per_iteration": 4.023146629333496
    },
    {
      "auxiliary_loss_clip": 0.01093583,
      "auxiliary_loss_mlp": 0.01029994,
      "balance_loss_clip": 1.03433406,
      "balance_loss_mlp": 1.0180676,
      "epoch": 0.8592815271306178,
      "flos": 17457290835840.0,
      "grad_norm": 1.9853070736969745,
      "language_loss": 0.71534693,
      "learning_rate": 2.0406737359318792e-07,
      "loss": 0.73658276,
      "num_input_tokens_seen": 308304130,
      "step": 14292,
      "time_per_iteration": 3.877232551574707
    },
    {
      "auxiliary_loss_clip": 0.01092822,
      "auxiliary_loss_mlp": 0.01030984,
      "balance_loss_clip": 1.03435445,
      "balance_loss_mlp": 1.01945615,
      "epoch": 0.8593416503832857,
      "flos": 25411360394880.0,
      "grad_norm": 1.483361975742606,
      "language_loss": 0.71222097,
      "learning_rate": 2.038960195018542e-07,
      "loss": 0.73345912,
      "num_input_tokens_seen": 308324670,
      "step": 14293,
      "time_per_iteration": 2.509713888168335
    },
    {
      "auxiliary_loss_clip": 0.01074019,
      "auxiliary_loss_mlp": 0.01033609,
      "balance_loss_clip": 1.03437448,
      "balance_loss_mlp": 1.0218668,
      "epoch": 0.8594017736359537,
      "flos": 20996646393600.0,
      "grad_norm": 1.5647116355965973,
      "language_loss": 0.68955636,
      "learning_rate": 2.0372473351896358e-07,
      "loss": 0.71063262,
      "num_input_tokens_seen": 308344215,
      "step": 14294,
      "time_per_iteration": 3.9416236877441406
    },
    {
      "auxiliary_loss_clip": 0.01099332,
      "auxiliary_loss_mlp": 0.01030115,
      "balance_loss_clip": 1.03293359,
      "balance_loss_mlp": 1.01877236,
      "epoch": 0.8594618968886216,
      "flos": 22091059929600.0,
      "grad_norm": 1.860477358135291,
      "language_loss": 0.77960026,
      "learning_rate": 2.0355351565101087e-07,
      "loss": 0.8008948,
      "num_input_tokens_seen": 308360520,
      "step": 14295,
      "time_per_iteration": 2.4700238704681396
    },
    {
      "auxiliary_loss_clip": 0.01079904,
      "auxiliary_loss_mlp": 0.01039086,
      "balance_loss_clip": 1.03316832,
      "balance_loss_mlp": 1.02543068,
      "epoch": 0.8595220201412896,
      "flos": 11656245467520.0,
      "grad_norm": 3.0833437482347708,
      "language_loss": 0.68570113,
      "learning_rate": 2.0338236590448975e-07,
      "loss": 0.70689106,
      "num_input_tokens_seen": 308376865,
      "step": 14296,
      "time_per_iteration": 2.4878060817718506
    },
    {
      "auxiliary_loss_clip": 0.01075798,
      "auxiliary_loss_mlp": 0.01028847,
      "balance_loss_clip": 1.03224373,
      "balance_loss_mlp": 1.01662803,
      "epoch": 0.8595821433939577,
      "flos": 25040366772480.0,
      "grad_norm": 1.9864974884780775,
      "language_loss": 0.79602861,
      "learning_rate": 2.0321128428588842e-07,
      "loss": 0.81707507,
      "num_input_tokens_seen": 308395870,
      "step": 14297,
      "time_per_iteration": 2.5417251586914062
    },
    {
      "auxiliary_loss_clip": 0.01088604,
      "auxiliary_loss_mlp": 0.01027952,
      "balance_loss_clip": 1.03294694,
      "balance_loss_mlp": 1.01706254,
      "epoch": 0.8596422666466256,
      "flos": 28511528359680.0,
      "grad_norm": 1.7530587740965007,
      "language_loss": 0.6832937,
      "learning_rate": 2.030402708016954e-07,
      "loss": 0.70445925,
      "num_input_tokens_seen": 308417250,
      "step": 14298,
      "time_per_iteration": 2.534247636795044
    },
    {
      "auxiliary_loss_clip": 0.01078518,
      "auxiliary_loss_mlp": 0.01032632,
      "balance_loss_clip": 1.03400171,
      "balance_loss_mlp": 1.02102065,
      "epoch": 0.8597023898992936,
      "flos": 13589137157760.0,
      "grad_norm": 1.9480448322538344,
      "language_loss": 0.68880832,
      "learning_rate": 2.0286932545839576e-07,
      "loss": 0.70991969,
      "num_input_tokens_seen": 308434565,
      "step": 14299,
      "time_per_iteration": 2.4974045753479004
    },
    {
      "auxiliary_loss_clip": 0.0107311,
      "auxiliary_loss_mlp": 0.01034367,
      "balance_loss_clip": 1.03514194,
      "balance_loss_mlp": 1.02263141,
      "epoch": 0.8597625131519615,
      "flos": 32300821728000.0,
      "grad_norm": 2.3068929476639917,
      "language_loss": 0.71551311,
      "learning_rate": 2.0269844826247096e-07,
      "loss": 0.73658794,
      "num_input_tokens_seen": 308450040,
      "step": 14300,
      "time_per_iteration": 2.5992512702941895
    },
    {
      "auxiliary_loss_clip": 0.01077215,
      "auxiliary_loss_mlp": 0.01033142,
      "balance_loss_clip": 1.03141785,
      "balance_loss_mlp": 1.02026761,
      "epoch": 0.8598226364046295,
      "flos": 28730367970560.0,
      "grad_norm": 1.4352327900068327,
      "language_loss": 0.6935631,
      "learning_rate": 2.0252763922040116e-07,
      "loss": 0.71466672,
      "num_input_tokens_seen": 308470545,
      "step": 14301,
      "time_per_iteration": 3.9283857345581055
    },
    {
      "auxiliary_loss_clip": 0.01052666,
      "auxiliary_loss_mlp": 0.01030382,
      "balance_loss_clip": 1.03405213,
      "balance_loss_mlp": 1.0185442,
      "epoch": 0.8598827596572974,
      "flos": 21871825269120.0,
      "grad_norm": 2.1894381784055104,
      "language_loss": 0.73887759,
      "learning_rate": 2.023568983386641e-07,
      "loss": 0.75970811,
      "num_input_tokens_seen": 308490020,
      "step": 14302,
      "time_per_iteration": 2.574747085571289
    },
    {
      "auxiliary_loss_clip": 0.01087849,
      "auxiliary_loss_mlp": 0.01025862,
      "balance_loss_clip": 1.03291464,
      "balance_loss_mlp": 1.01496005,
      "epoch": 0.8599428829099655,
      "flos": 23767297966080.0,
      "grad_norm": 1.8153183580818106,
      "language_loss": 0.83705652,
      "learning_rate": 2.02186225623733e-07,
      "loss": 0.85819364,
      "num_input_tokens_seen": 308509065,
      "step": 14303,
      "time_per_iteration": 2.4991750717163086
    },
    {
      "auxiliary_loss_clip": 0.01091105,
      "auxiliary_loss_mlp": 0.01034373,
      "balance_loss_clip": 1.03346276,
      "balance_loss_mlp": 1.02218378,
      "epoch": 0.8600030061626334,
      "flos": 16212770363520.0,
      "grad_norm": 5.968230029996946,
      "language_loss": 0.77086902,
      "learning_rate": 2.0201562108208025e-07,
      "loss": 0.79212379,
      "num_input_tokens_seen": 308524725,
      "step": 14304,
      "time_per_iteration": 2.4447391033172607
    },
    {
      "auxiliary_loss_clip": 0.01105042,
      "auxiliary_loss_mlp": 0.01036463,
      "balance_loss_clip": 1.03595459,
      "balance_loss_mlp": 1.02365434,
      "epoch": 0.8600631294153014,
      "flos": 15669370437120.0,
      "grad_norm": 2.460197556412072,
      "language_loss": 0.5337168,
      "learning_rate": 2.0184508472017537e-07,
      "loss": 0.55513185,
      "num_input_tokens_seen": 308543525,
      "step": 14305,
      "time_per_iteration": 2.437100648880005
    },
    {
      "auxiliary_loss_clip": 0.01102216,
      "auxiliary_loss_mlp": 0.01025396,
      "balance_loss_clip": 1.03509426,
      "balance_loss_mlp": 1.01351714,
      "epoch": 0.8601232526679693,
      "flos": 17493093717120.0,
      "grad_norm": 2.111510066055082,
      "language_loss": 0.84273958,
      "learning_rate": 2.0167461654448558e-07,
      "loss": 0.8640157,
      "num_input_tokens_seen": 308557995,
      "step": 14306,
      "time_per_iteration": 2.4166693687438965
    },
    {
      "auxiliary_loss_clip": 0.01090651,
      "auxiliary_loss_mlp": 0.00780487,
      "balance_loss_clip": 1.03398955,
      "balance_loss_mlp": 1.00538898,
      "epoch": 0.8601833759206373,
      "flos": 26985935963520.0,
      "grad_norm": 1.3955048729242354,
      "language_loss": 0.71431077,
      "learning_rate": 2.01504216561474e-07,
      "loss": 0.73302221,
      "num_input_tokens_seen": 308582750,
      "step": 14307,
      "time_per_iteration": 2.5940725803375244
    },
    {
      "auxiliary_loss_clip": 0.01092166,
      "auxiliary_loss_mlp": 0.00784889,
      "balance_loss_clip": 1.03242755,
      "balance_loss_mlp": 1.00941682,
      "epoch": 0.8602434991733052,
      "flos": 25229760209280.0,
      "grad_norm": 1.6164513888427787,
      "language_loss": 0.6353693,
      "learning_rate": 2.0133388477760316e-07,
      "loss": 0.65413988,
      "num_input_tokens_seen": 308603770,
      "step": 14308,
      "time_per_iteration": 2.520097017288208
    },
    {
      "auxiliary_loss_clip": 0.01010494,
      "auxiliary_loss_mlp": 0.01006206,
      "balance_loss_clip": 1.00744319,
      "balance_loss_mlp": 1.00506747,
      "epoch": 0.8603036224259732,
      "flos": 71015363107200.0,
      "grad_norm": 0.6248265596218042,
      "language_loss": 0.48486665,
      "learning_rate": 2.0116362119933172e-07,
      "loss": 0.50503367,
      "num_input_tokens_seen": 308667735,
      "step": 14309,
      "time_per_iteration": 3.206472873687744
    },
    {
      "auxiliary_loss_clip": 0.01048885,
      "auxiliary_loss_mlp": 0.01035911,
      "balance_loss_clip": 1.03398836,
      "balance_loss_mlp": 1.02276826,
      "epoch": 0.8603637456786413,
      "flos": 20300625578880.0,
      "grad_norm": 1.7157077335028457,
      "language_loss": 0.67163658,
      "learning_rate": 2.0099342583311563e-07,
      "loss": 0.6924845,
      "num_input_tokens_seen": 308686300,
      "step": 14310,
      "time_per_iteration": 2.6153416633605957
    },
    {
      "auxiliary_loss_clip": 0.01037394,
      "auxiliary_loss_mlp": 0.01033284,
      "balance_loss_clip": 1.03099966,
      "balance_loss_mlp": 1.02116013,
      "epoch": 0.8604238689313092,
      "flos": 21835842819840.0,
      "grad_norm": 1.83250936546361,
      "language_loss": 0.78786892,
      "learning_rate": 2.0082329868540905e-07,
      "loss": 0.80857563,
      "num_input_tokens_seen": 308705825,
      "step": 14311,
      "time_per_iteration": 2.627955913543701
    },
    {
      "auxiliary_loss_clip": 0.01090366,
      "auxiliary_loss_mlp": 0.01026353,
      "balance_loss_clip": 1.03393245,
      "balance_loss_mlp": 1.0148139,
      "epoch": 0.8604839921839772,
      "flos": 18004210295040.0,
      "grad_norm": 2.0810465927289172,
      "language_loss": 0.71358752,
      "learning_rate": 2.006532397626639e-07,
      "loss": 0.73475474,
      "num_input_tokens_seen": 308723340,
      "step": 14312,
      "time_per_iteration": 2.487879514694214
    },
    {
      "auxiliary_loss_clip": 0.01075194,
      "auxiliary_loss_mlp": 0.01032276,
      "balance_loss_clip": 1.03186238,
      "balance_loss_mlp": 1.01998591,
      "epoch": 0.8605441154366451,
      "flos": 16252164604800.0,
      "grad_norm": 1.965276591745554,
      "language_loss": 0.78112292,
      "learning_rate": 2.0048324907132797e-07,
      "loss": 0.80219758,
      "num_input_tokens_seen": 308741280,
      "step": 14313,
      "time_per_iteration": 2.4891586303710938
    },
    {
      "auxiliary_loss_clip": 0.01079302,
      "auxiliary_loss_mlp": 0.01032344,
      "balance_loss_clip": 1.03360784,
      "balance_loss_mlp": 1.01952255,
      "epoch": 0.8606042386893131,
      "flos": 32267065921920.0,
      "grad_norm": 1.6508441725652045,
      "language_loss": 0.73121333,
      "learning_rate": 2.003133266178474e-07,
      "loss": 0.75232983,
      "num_input_tokens_seen": 308762875,
      "step": 14314,
      "time_per_iteration": 2.602266550064087
    },
    {
      "auxiliary_loss_clip": 0.01078745,
      "auxiliary_loss_mlp": 0.01025625,
      "balance_loss_clip": 1.03281391,
      "balance_loss_mlp": 1.01400161,
      "epoch": 0.860664361941981,
      "flos": 20229774001920.0,
      "grad_norm": 1.6673238488592372,
      "language_loss": 0.69107068,
      "learning_rate": 2.001434724086657e-07,
      "loss": 0.71211433,
      "num_input_tokens_seen": 308780315,
      "step": 14315,
      "time_per_iteration": 2.4937996864318848
    },
    {
      "auxiliary_loss_clip": 0.01091351,
      "auxiliary_loss_mlp": 0.01031948,
      "balance_loss_clip": 1.03476632,
      "balance_loss_mlp": 1.02037859,
      "epoch": 0.8607244851946491,
      "flos": 25191622944000.0,
      "grad_norm": 1.810438416747908,
      "language_loss": 0.72013074,
      "learning_rate": 1.9997368645022418e-07,
      "loss": 0.74136364,
      "num_input_tokens_seen": 308799435,
      "step": 14316,
      "time_per_iteration": 2.538911819458008
    },
    {
      "auxiliary_loss_clip": 0.0108641,
      "auxiliary_loss_mlp": 0.01029243,
      "balance_loss_clip": 1.03850555,
      "balance_loss_mlp": 1.01763809,
      "epoch": 0.860784608447317,
      "flos": 20482082110080.0,
      "grad_norm": 2.4384086151207174,
      "language_loss": 0.82578897,
      "learning_rate": 1.9980396874896056e-07,
      "loss": 0.84694552,
      "num_input_tokens_seen": 308817730,
      "step": 14317,
      "time_per_iteration": 2.5049307346343994
    },
    {
      "auxiliary_loss_clip": 0.01082975,
      "auxiliary_loss_mlp": 0.01031535,
      "balance_loss_clip": 1.03458309,
      "balance_loss_mlp": 1.01948905,
      "epoch": 0.860844731699985,
      "flos": 50476037696640.0,
      "grad_norm": 1.6615638224618794,
      "language_loss": 0.67130899,
      "learning_rate": 1.996343193113108e-07,
      "loss": 0.6924541,
      "num_input_tokens_seen": 308841735,
      "step": 14318,
      "time_per_iteration": 2.770570993423462
    },
    {
      "auxiliary_loss_clip": 0.01088635,
      "auxiliary_loss_mlp": 0.0102473,
      "balance_loss_clip": 1.03418708,
      "balance_loss_mlp": 1.0135895,
      "epoch": 0.8609048549526529,
      "flos": 41172768455040.0,
      "grad_norm": 1.549422841203104,
      "language_loss": 0.71270728,
      "learning_rate": 1.9946473814370911e-07,
      "loss": 0.73384094,
      "num_input_tokens_seen": 308865050,
      "step": 14319,
      "time_per_iteration": 2.654040575027466
    },
    {
      "auxiliary_loss_clip": 0.01085178,
      "auxiliary_loss_mlp": 0.00783745,
      "balance_loss_clip": 1.03554499,
      "balance_loss_mlp": 1.01054478,
      "epoch": 0.8609649782053209,
      "flos": 23951196622080.0,
      "grad_norm": 1.821585770455099,
      "language_loss": 0.67141396,
      "learning_rate": 1.992952252525839e-07,
      "loss": 0.69010323,
      "num_input_tokens_seen": 308885375,
      "step": 14320,
      "time_per_iteration": 2.5576584339141846
    },
    {
      "auxiliary_loss_clip": 0.01075796,
      "auxiliary_loss_mlp": 0.01035305,
      "balance_loss_clip": 1.03235412,
      "balance_loss_mlp": 1.0214169,
      "epoch": 0.8610251014579888,
      "flos": 23112574813440.0,
      "grad_norm": 2.026088616953404,
      "language_loss": 0.80412966,
      "learning_rate": 1.9912578064436446e-07,
      "loss": 0.82524067,
      "num_input_tokens_seen": 308904700,
      "step": 14321,
      "time_per_iteration": 2.5101845264434814
    },
    {
      "auxiliary_loss_clip": 0.01087167,
      "auxiliary_loss_mlp": 0.00783265,
      "balance_loss_clip": 1.03275192,
      "balance_loss_mlp": 1.00874937,
      "epoch": 0.8610852247106568,
      "flos": 19426811420160.0,
      "grad_norm": 1.87582374077272,
      "language_loss": 0.71013534,
      "learning_rate": 1.9895640432547567e-07,
      "loss": 0.72883964,
      "num_input_tokens_seen": 308922985,
      "step": 14322,
      "time_per_iteration": 2.478645086288452
    },
    {
      "auxiliary_loss_clip": 0.01084056,
      "auxiliary_loss_mlp": 0.01036624,
      "balance_loss_clip": 1.03452206,
      "balance_loss_mlp": 1.02323127,
      "epoch": 0.8611453479633249,
      "flos": 19312076401920.0,
      "grad_norm": 1.8366979998199853,
      "language_loss": 0.56087363,
      "learning_rate": 1.9878709630234102e-07,
      "loss": 0.58208042,
      "num_input_tokens_seen": 308940765,
      "step": 14323,
      "time_per_iteration": 2.5026872158050537
    },
    {
      "auxiliary_loss_clip": 0.01068569,
      "auxiliary_loss_mlp": 0.01027063,
      "balance_loss_clip": 1.03322744,
      "balance_loss_mlp": 1.01533866,
      "epoch": 0.8612054712159928,
      "flos": 23253667436160.0,
      "grad_norm": 1.9072317652752087,
      "language_loss": 0.75941247,
      "learning_rate": 1.986178565813801e-07,
      "loss": 0.7803688,
      "num_input_tokens_seen": 308960110,
      "step": 14324,
      "time_per_iteration": 2.5698184967041016
    },
    {
      "auxiliary_loss_clip": 0.01058843,
      "auxiliary_loss_mlp": 0.01037779,
      "balance_loss_clip": 1.03448677,
      "balance_loss_mlp": 1.0232172,
      "epoch": 0.8612655944686608,
      "flos": 16028440744320.0,
      "grad_norm": 1.9922937098769982,
      "language_loss": 0.66591293,
      "learning_rate": 1.9844868516901036e-07,
      "loss": 0.6868791,
      "num_input_tokens_seen": 308976665,
      "step": 14325,
      "time_per_iteration": 2.5475683212280273
    },
    {
      "auxiliary_loss_clip": 0.01094759,
      "auxiliary_loss_mlp": 0.01029057,
      "balance_loss_clip": 1.03556681,
      "balance_loss_mlp": 1.0166707,
      "epoch": 0.8613257177213287,
      "flos": 22492720788480.0,
      "grad_norm": 1.570178396216154,
      "language_loss": 0.64752221,
      "learning_rate": 1.982795820716472e-07,
      "loss": 0.6687603,
      "num_input_tokens_seen": 308997015,
      "step": 14326,
      "time_per_iteration": 2.517493486404419
    },
    {
      "auxiliary_loss_clip": 0.01081335,
      "auxiliary_loss_mlp": 0.01033407,
      "balance_loss_clip": 1.03310144,
      "balance_loss_mlp": 1.02114034,
      "epoch": 0.8613858409739967,
      "flos": 17238056175360.0,
      "grad_norm": 2.112833627668595,
      "language_loss": 0.84372538,
      "learning_rate": 1.9811054729570253e-07,
      "loss": 0.86487281,
      "num_input_tokens_seen": 309015250,
      "step": 14327,
      "time_per_iteration": 2.482632875442505
    },
    {
      "auxiliary_loss_clip": 0.01092503,
      "auxiliary_loss_mlp": 0.01032505,
      "balance_loss_clip": 1.03413916,
      "balance_loss_mlp": 1.02034569,
      "epoch": 0.8614459642266646,
      "flos": 22821123859200.0,
      "grad_norm": 2.312221904073291,
      "language_loss": 0.74732393,
      "learning_rate": 1.9794158084758661e-07,
      "loss": 0.768574,
      "num_input_tokens_seen": 309034140,
      "step": 14328,
      "time_per_iteration": 2.5242748260498047
    },
    {
      "auxiliary_loss_clip": 0.01092885,
      "auxiliary_loss_mlp": 0.01025423,
      "balance_loss_clip": 1.03479242,
      "balance_loss_mlp": 1.01364517,
      "epoch": 0.8615060874793327,
      "flos": 26504301473280.0,
      "grad_norm": 1.8464338534804243,
      "language_loss": 0.79932523,
      "learning_rate": 1.9777268273370673e-07,
      "loss": 0.8205083,
      "num_input_tokens_seen": 309055075,
      "step": 14329,
      "time_per_iteration": 3.909661054611206
    },
    {
      "auxiliary_loss_clip": 0.01076774,
      "auxiliary_loss_mlp": 0.01027184,
      "balance_loss_clip": 1.03517699,
      "balance_loss_mlp": 1.01517963,
      "epoch": 0.8615662107320006,
      "flos": 24061011477120.0,
      "grad_norm": 1.863513737111843,
      "language_loss": 0.77016842,
      "learning_rate": 1.9760385296046757e-07,
      "loss": 0.79120803,
      "num_input_tokens_seen": 309074650,
      "step": 14330,
      "time_per_iteration": 3.9895944595336914
    },
    {
      "auxiliary_loss_clip": 0.01091264,
      "auxiliary_loss_mlp": 0.01027821,
      "balance_loss_clip": 1.03358364,
      "balance_loss_mlp": 1.01604283,
      "epoch": 0.8616263339846686,
      "flos": 24165044242560.0,
      "grad_norm": 1.85382558855581,
      "language_loss": 0.64455986,
      "learning_rate": 1.974350915342702e-07,
      "loss": 0.66575068,
      "num_input_tokens_seen": 309094385,
      "step": 14331,
      "time_per_iteration": 2.508577823638916
    },
    {
      "auxiliary_loss_clip": 0.01078181,
      "auxiliary_loss_mlp": 0.010288,
      "balance_loss_clip": 1.03649521,
      "balance_loss_mlp": 1.018327,
      "epoch": 0.8616864572373365,
      "flos": 21724340025600.0,
      "grad_norm": 1.683165370703536,
      "language_loss": 0.76183689,
      "learning_rate": 1.9726639846151506e-07,
      "loss": 0.78290665,
      "num_input_tokens_seen": 309111815,
      "step": 14332,
      "time_per_iteration": 3.9387359619140625
    },
    {
      "auxiliary_loss_clip": 0.01091403,
      "auxiliary_loss_mlp": 0.01030477,
      "balance_loss_clip": 1.03475785,
      "balance_loss_mlp": 1.01739931,
      "epoch": 0.8617465804900045,
      "flos": 23766651521280.0,
      "grad_norm": 1.7162102180074712,
      "language_loss": 0.66720891,
      "learning_rate": 1.9709777374859904e-07,
      "loss": 0.68842769,
      "num_input_tokens_seen": 309131385,
      "step": 14333,
      "time_per_iteration": 2.4972712993621826
    },
    {
      "auxiliary_loss_clip": 0.01079603,
      "auxiliary_loss_mlp": 0.01035259,
      "balance_loss_clip": 1.03443372,
      "balance_loss_mlp": 1.02157354,
      "epoch": 0.8618067037426724,
      "flos": 37703941251840.0,
      "grad_norm": 1.7573448555395703,
      "language_loss": 0.62351131,
      "learning_rate": 1.969292174019157e-07,
      "loss": 0.64465988,
      "num_input_tokens_seen": 309155020,
      "step": 14334,
      "time_per_iteration": 2.667008638381958
    },
    {
      "auxiliary_loss_clip": 0.01074176,
      "auxiliary_loss_mlp": 0.01049724,
      "balance_loss_clip": 1.03511453,
      "balance_loss_mlp": 1.036057,
      "epoch": 0.8618668269953405,
      "flos": 21471026336640.0,
      "grad_norm": 1.894903500501414,
      "language_loss": 0.69301027,
      "learning_rate": 1.967607294278577e-07,
      "loss": 0.71424931,
      "num_input_tokens_seen": 309172865,
      "step": 14335,
      "time_per_iteration": 2.519529104232788
    },
    {
      "auxiliary_loss_clip": 0.01094576,
      "auxiliary_loss_mlp": 0.01031409,
      "balance_loss_clip": 1.03611839,
      "balance_loss_mlp": 1.01975012,
      "epoch": 0.8619269502480085,
      "flos": 22232691256320.0,
      "grad_norm": 1.448608804390458,
      "language_loss": 0.8315804,
      "learning_rate": 1.965923098328135e-07,
      "loss": 0.85284019,
      "num_input_tokens_seen": 309193575,
      "step": 14336,
      "time_per_iteration": 2.5016698837280273
    },
    {
      "auxiliary_loss_clip": 0.01106742,
      "auxiliary_loss_mlp": 0.01030358,
      "balance_loss_clip": 1.03556263,
      "balance_loss_mlp": 1.01821637,
      "epoch": 0.8619870735006764,
      "flos": 22710626645760.0,
      "grad_norm": 1.6970126728312245,
      "language_loss": 0.6748361,
      "learning_rate": 1.9642395862316907e-07,
      "loss": 0.69620711,
      "num_input_tokens_seen": 309212680,
      "step": 14337,
      "time_per_iteration": 2.444659948348999
    },
    {
      "auxiliary_loss_clip": 0.01065882,
      "auxiliary_loss_mlp": 0.01029599,
      "balance_loss_clip": 1.03069496,
      "balance_loss_mlp": 1.01736772,
      "epoch": 0.8620471967533444,
      "flos": 37520293991040.0,
      "grad_norm": 1.5307307166339434,
      "language_loss": 0.67021108,
      "learning_rate": 1.962556758053089e-07,
      "loss": 0.69116592,
      "num_input_tokens_seen": 309234485,
      "step": 14338,
      "time_per_iteration": 2.699268102645874
    },
    {
      "auxiliary_loss_clip": 0.01082883,
      "auxiliary_loss_mlp": 0.01031943,
      "balance_loss_clip": 1.0358398,
      "balance_loss_mlp": 1.02022457,
      "epoch": 0.8621073200060123,
      "flos": 19682459493120.0,
      "grad_norm": 2.0141022283664767,
      "language_loss": 0.61777782,
      "learning_rate": 1.9608746138561448e-07,
      "loss": 0.63892603,
      "num_input_tokens_seen": 309253630,
      "step": 14339,
      "time_per_iteration": 3.893486499786377
    },
    {
      "auxiliary_loss_clip": 0.01080691,
      "auxiliary_loss_mlp": 0.00782501,
      "balance_loss_clip": 1.03234971,
      "balance_loss_mlp": 1.00840592,
      "epoch": 0.8621674432586803,
      "flos": 14536855549440.0,
      "grad_norm": 1.8359739096786403,
      "language_loss": 0.62706578,
      "learning_rate": 1.9591931537046458e-07,
      "loss": 0.64569771,
      "num_input_tokens_seen": 309270950,
      "step": 14340,
      "time_per_iteration": 2.5031309127807617
    },
    {
      "auxiliary_loss_clip": 0.01056317,
      "auxiliary_loss_mlp": 0.01024086,
      "balance_loss_clip": 1.03533232,
      "balance_loss_mlp": 1.0131005,
      "epoch": 0.8622275665113482,
      "flos": 20740100480640.0,
      "grad_norm": 1.651375475473071,
      "language_loss": 0.80155182,
      "learning_rate": 1.9575123776623493e-07,
      "loss": 0.82235587,
      "num_input_tokens_seen": 309288780,
      "step": 14341,
      "time_per_iteration": 2.5625181198120117
    },
    {
      "auxiliary_loss_clip": 0.01087698,
      "auxiliary_loss_mlp": 0.01030608,
      "balance_loss_clip": 1.03503942,
      "balance_loss_mlp": 1.01934266,
      "epoch": 0.8622876897640163,
      "flos": 24715914197760.0,
      "grad_norm": 1.8162647383247241,
      "language_loss": 0.74722457,
      "learning_rate": 1.9558322857929887e-07,
      "loss": 0.76840758,
      "num_input_tokens_seen": 309310875,
      "step": 14342,
      "time_per_iteration": 2.5396804809570312
    },
    {
      "auxiliary_loss_clip": 0.01072705,
      "auxiliary_loss_mlp": 0.01028097,
      "balance_loss_clip": 1.03487611,
      "balance_loss_mlp": 1.01559782,
      "epoch": 0.8623478130166842,
      "flos": 17457362663040.0,
      "grad_norm": 2.7582309081837932,
      "language_loss": 0.68724787,
      "learning_rate": 1.95415287816028e-07,
      "loss": 0.70825589,
      "num_input_tokens_seen": 309329900,
      "step": 14343,
      "time_per_iteration": 2.5160303115844727
    },
    {
      "auxiliary_loss_clip": 0.01092105,
      "auxiliary_loss_mlp": 0.01041669,
      "balance_loss_clip": 1.03420448,
      "balance_loss_mlp": 1.02881789,
      "epoch": 0.8624079362693522,
      "flos": 18109176814080.0,
      "grad_norm": 1.6608740399308295,
      "language_loss": 0.68347597,
      "learning_rate": 1.9524741548278967e-07,
      "loss": 0.70481372,
      "num_input_tokens_seen": 309347870,
      "step": 14344,
      "time_per_iteration": 2.4695301055908203
    },
    {
      "auxiliary_loss_clip": 0.01070416,
      "auxiliary_loss_mlp": 0.01037512,
      "balance_loss_clip": 1.03410816,
      "balance_loss_mlp": 1.02430975,
      "epoch": 0.8624680595220201,
      "flos": 30666455971200.0,
      "grad_norm": 1.4683342320131074,
      "language_loss": 0.81459439,
      "learning_rate": 1.9507961158595054e-07,
      "loss": 0.83567369,
      "num_input_tokens_seen": 309371695,
      "step": 14345,
      "time_per_iteration": 2.606796979904175
    },
    {
      "auxiliary_loss_clip": 0.01096209,
      "auxiliary_loss_mlp": 0.01033669,
      "balance_loss_clip": 1.0355382,
      "balance_loss_mlp": 1.02121711,
      "epoch": 0.8625281827746881,
      "flos": 37998588516480.0,
      "grad_norm": 2.000521816766669,
      "language_loss": 0.50633645,
      "learning_rate": 1.9491187613187355e-07,
      "loss": 0.52763522,
      "num_input_tokens_seen": 309394645,
      "step": 14346,
      "time_per_iteration": 2.632624387741089
    },
    {
      "auxiliary_loss_clip": 0.0102562,
      "auxiliary_loss_mlp": 0.0103145,
      "balance_loss_clip": 1.03313208,
      "balance_loss_mlp": 1.01908791,
      "epoch": 0.862588306027356,
      "flos": 26249730808320.0,
      "grad_norm": 1.4318276822531815,
      "language_loss": 0.7495929,
      "learning_rate": 1.9474420912691913e-07,
      "loss": 0.7701636,
      "num_input_tokens_seen": 309413170,
      "step": 14347,
      "time_per_iteration": 2.678907632827759
    },
    {
      "auxiliary_loss_clip": 0.01078802,
      "auxiliary_loss_mlp": 0.01033233,
      "balance_loss_clip": 1.03469706,
      "balance_loss_mlp": 1.02023304,
      "epoch": 0.862648429280024,
      "flos": 25878809013120.0,
      "grad_norm": 1.956767589993399,
      "language_loss": 0.80738819,
      "learning_rate": 1.945766105774449e-07,
      "loss": 0.82850856,
      "num_input_tokens_seen": 309431315,
      "step": 14348,
      "time_per_iteration": 2.5597450733184814
    },
    {
      "auxiliary_loss_clip": 0.01087013,
      "auxiliary_loss_mlp": 0.0102843,
      "balance_loss_clip": 1.03352857,
      "balance_loss_mlp": 1.01718259,
      "epoch": 0.862708552532692,
      "flos": 37816413713280.0,
      "grad_norm": 1.8361032489845635,
      "language_loss": 0.66073453,
      "learning_rate": 1.9440908048980665e-07,
      "loss": 0.681889,
      "num_input_tokens_seen": 309453020,
      "step": 14349,
      "time_per_iteration": 2.626281976699829
    },
    {
      "auxiliary_loss_clip": 0.01092846,
      "auxiliary_loss_mlp": 0.01037783,
      "balance_loss_clip": 1.03457177,
      "balance_loss_mlp": 1.02550411,
      "epoch": 0.86276867578536,
      "flos": 19091800247040.0,
      "grad_norm": 2.1198271869230716,
      "language_loss": 0.69822478,
      "learning_rate": 1.942416188703573e-07,
      "loss": 0.71953106,
      "num_input_tokens_seen": 309469780,
      "step": 14350,
      "time_per_iteration": 2.453994035720825
    },
    {
      "auxiliary_loss_clip": 0.01075058,
      "auxiliary_loss_mlp": 0.01035441,
      "balance_loss_clip": 1.03272963,
      "balance_loss_mlp": 1.02345443,
      "epoch": 0.862828799038028,
      "flos": 22164281804160.0,
      "grad_norm": 1.7639516240439652,
      "language_loss": 0.77168167,
      "learning_rate": 1.9407422572544618e-07,
      "loss": 0.79278666,
      "num_input_tokens_seen": 309489610,
      "step": 14351,
      "time_per_iteration": 2.550434112548828
    },
    {
      "auxiliary_loss_clip": 0.01091928,
      "auxiliary_loss_mlp": 0.01028296,
      "balance_loss_clip": 1.03469586,
      "balance_loss_mlp": 1.01689959,
      "epoch": 0.8628889222906959,
      "flos": 23145576433920.0,
      "grad_norm": 1.7300266489273186,
      "language_loss": 0.84721148,
      "learning_rate": 1.9390690106142204e-07,
      "loss": 0.86841369,
      "num_input_tokens_seen": 309508295,
      "step": 14352,
      "time_per_iteration": 2.5152041912078857
    },
    {
      "auxiliary_loss_clip": 0.01020811,
      "auxiliary_loss_mlp": 0.01001823,
      "balance_loss_clip": 1.00740695,
      "balance_loss_mlp": 1.00064325,
      "epoch": 0.8629490455433639,
      "flos": 57817762151040.0,
      "grad_norm": 0.7941488329985426,
      "language_loss": 0.61982977,
      "learning_rate": 1.9373964488462913e-07,
      "loss": 0.64005613,
      "num_input_tokens_seen": 309567960,
      "step": 14353,
      "time_per_iteration": 3.111920118331909
    },
    {
      "auxiliary_loss_clip": 0.01103261,
      "auxiliary_loss_mlp": 0.01025705,
      "balance_loss_clip": 1.03635216,
      "balance_loss_mlp": 1.01483262,
      "epoch": 0.8630091687960318,
      "flos": 15919667383680.0,
      "grad_norm": 1.6955655197223638,
      "language_loss": 0.81897503,
      "learning_rate": 1.9357245720140948e-07,
      "loss": 0.84026468,
      "num_input_tokens_seen": 309586050,
      "step": 14354,
      "time_per_iteration": 2.474431276321411
    },
    {
      "auxiliary_loss_clip": 0.01083193,
      "auxiliary_loss_mlp": 0.01029569,
      "balance_loss_clip": 1.0332973,
      "balance_loss_mlp": 1.01702261,
      "epoch": 0.8630692920486999,
      "flos": 17961691570560.0,
      "grad_norm": 1.989736609800371,
      "language_loss": 0.85536575,
      "learning_rate": 1.934053380181031e-07,
      "loss": 0.87649333,
      "num_input_tokens_seen": 309602910,
      "step": 14355,
      "time_per_iteration": 2.48223614692688
    },
    {
      "auxiliary_loss_clip": 0.01072057,
      "auxiliary_loss_mlp": 0.01032259,
      "balance_loss_clip": 1.03292203,
      "balance_loss_mlp": 1.01925969,
      "epoch": 0.8631294153013678,
      "flos": 22455158140800.0,
      "grad_norm": 2.0269166586533167,
      "language_loss": 0.58799648,
      "learning_rate": 1.9323828734104763e-07,
      "loss": 0.60903966,
      "num_input_tokens_seen": 309621175,
      "step": 14356,
      "time_per_iteration": 2.5295395851135254
    },
    {
      "auxiliary_loss_clip": 0.01059958,
      "auxiliary_loss_mlp": 0.01032149,
      "balance_loss_clip": 1.03278542,
      "balance_loss_mlp": 1.01921439,
      "epoch": 0.8631895385540358,
      "flos": 16837005847680.0,
      "grad_norm": 1.7952721581564965,
      "language_loss": 0.77021933,
      "learning_rate": 1.9307130517657756e-07,
      "loss": 0.79114038,
      "num_input_tokens_seen": 309639395,
      "step": 14357,
      "time_per_iteration": 2.558746337890625
    },
    {
      "auxiliary_loss_clip": 0.01094416,
      "auxiliary_loss_mlp": 0.01028899,
      "balance_loss_clip": 1.03643799,
      "balance_loss_mlp": 1.01703155,
      "epoch": 0.8632496618067037,
      "flos": 18697214367360.0,
      "grad_norm": 2.117517734745847,
      "language_loss": 0.77831018,
      "learning_rate": 1.9290439153102468e-07,
      "loss": 0.79954326,
      "num_input_tokens_seen": 309657265,
      "step": 14358,
      "time_per_iteration": 2.5504045486450195
    },
    {
      "auxiliary_loss_clip": 0.01067716,
      "auxiliary_loss_mlp": 0.01033026,
      "balance_loss_clip": 1.03122473,
      "balance_loss_mlp": 1.01926899,
      "epoch": 0.8633097850593717,
      "flos": 24279922915200.0,
      "grad_norm": 1.3536098979566764,
      "language_loss": 0.75162554,
      "learning_rate": 1.9273754641071816e-07,
      "loss": 0.77263296,
      "num_input_tokens_seen": 309678610,
      "step": 14359,
      "time_per_iteration": 2.581483840942383
    },
    {
      "auxiliary_loss_clip": 0.01042262,
      "auxiliary_loss_mlp": 0.01028256,
      "balance_loss_clip": 1.0327512,
      "balance_loss_mlp": 1.01625752,
      "epoch": 0.8633699083120396,
      "flos": 21178569801600.0,
      "grad_norm": 1.9099360479195766,
      "language_loss": 0.70518911,
      "learning_rate": 1.9257076982198517e-07,
      "loss": 0.72589433,
      "num_input_tokens_seen": 309697710,
      "step": 14360,
      "time_per_iteration": 2.5933117866516113
    },
    {
      "auxiliary_loss_clip": 0.01072772,
      "auxiliary_loss_mlp": 0.01033005,
      "balance_loss_clip": 1.03513563,
      "balance_loss_mlp": 1.01996303,
      "epoch": 0.8634300315647077,
      "flos": 19244888012160.0,
      "grad_norm": 1.8664431560653567,
      "language_loss": 0.76490766,
      "learning_rate": 1.9240406177114953e-07,
      "loss": 0.78596544,
      "num_input_tokens_seen": 309715985,
      "step": 14361,
      "time_per_iteration": 2.5299808979034424
    },
    {
      "auxiliary_loss_clip": 0.01028156,
      "auxiliary_loss_mlp": 0.01003259,
      "balance_loss_clip": 1.00536895,
      "balance_loss_mlp": 1.00217402,
      "epoch": 0.8634901548173756,
      "flos": 66195648282240.0,
      "grad_norm": 0.9649512638840279,
      "language_loss": 0.58846861,
      "learning_rate": 1.922374222645329e-07,
      "loss": 0.60878271,
      "num_input_tokens_seen": 309779930,
      "step": 14362,
      "time_per_iteration": 3.070331335067749
    },
    {
      "auxiliary_loss_clip": 0.01038368,
      "auxiliary_loss_mlp": 0.01030611,
      "balance_loss_clip": 1.03531682,
      "balance_loss_mlp": 1.01814187,
      "epoch": 0.8635502780700436,
      "flos": 24789531121920.0,
      "grad_norm": 1.7186453780768944,
      "language_loss": 0.80659425,
      "learning_rate": 1.9207085130845524e-07,
      "loss": 0.8272841,
      "num_input_tokens_seen": 309800580,
      "step": 14363,
      "time_per_iteration": 2.694070816040039
    },
    {
      "auxiliary_loss_clip": 0.01077513,
      "auxiliary_loss_mlp": 0.01039517,
      "balance_loss_clip": 1.03163934,
      "balance_loss_mlp": 1.02519369,
      "epoch": 0.8636104013227116,
      "flos": 25189970918400.0,
      "grad_norm": 3.23297114591796,
      "language_loss": 0.72947192,
      "learning_rate": 1.9190434890923112e-07,
      "loss": 0.75064224,
      "num_input_tokens_seen": 309821725,
      "step": 14364,
      "time_per_iteration": 2.5426387786865234
    },
    {
      "auxiliary_loss_clip": 0.01080671,
      "auxiliary_loss_mlp": 0.01031693,
      "balance_loss_clip": 1.03264809,
      "balance_loss_mlp": 1.01888967,
      "epoch": 0.8636705245753795,
      "flos": 23878441624320.0,
      "grad_norm": 2.144946792356575,
      "language_loss": 0.71518731,
      "learning_rate": 1.917379150731755e-07,
      "loss": 0.7363109,
      "num_input_tokens_seen": 309841565,
      "step": 14365,
      "time_per_iteration": 2.553417205810547
    },
    {
      "auxiliary_loss_clip": 0.01081324,
      "auxiliary_loss_mlp": 0.01042303,
      "balance_loss_clip": 1.03474712,
      "balance_loss_mlp": 1.0273962,
      "epoch": 0.8637306478280475,
      "flos": 23110455911040.0,
      "grad_norm": 2.107526216804772,
      "language_loss": 0.7121228,
      "learning_rate": 1.915715498065993e-07,
      "loss": 0.7333591,
      "num_input_tokens_seen": 309858635,
      "step": 14366,
      "time_per_iteration": 2.507657051086426
    },
    {
      "auxiliary_loss_clip": 0.01075387,
      "auxiliary_loss_mlp": 0.0102728,
      "balance_loss_clip": 1.03480375,
      "balance_loss_mlp": 1.01637888,
      "epoch": 0.8637907710807154,
      "flos": 21906802137600.0,
      "grad_norm": 1.5622021440368534,
      "language_loss": 0.81649971,
      "learning_rate": 1.9140525311581146e-07,
      "loss": 0.83752638,
      "num_input_tokens_seen": 309877885,
      "step": 14367,
      "time_per_iteration": 2.5285778045654297
    },
    {
      "auxiliary_loss_clip": 0.01078902,
      "auxiliary_loss_mlp": 0.01027228,
      "balance_loss_clip": 1.03485274,
      "balance_loss_mlp": 1.01437092,
      "epoch": 0.8638508943333835,
      "flos": 23580526222080.0,
      "grad_norm": 2.3252829249218325,
      "language_loss": 0.61954284,
      "learning_rate": 1.9123902500711743e-07,
      "loss": 0.6406042,
      "num_input_tokens_seen": 309893140,
      "step": 14368,
      "time_per_iteration": 3.9515724182128906
    },
    {
      "auxiliary_loss_clip": 0.01093838,
      "auxiliary_loss_mlp": 0.01029148,
      "balance_loss_clip": 1.03651333,
      "balance_loss_mlp": 1.01722121,
      "epoch": 0.8639110175860514,
      "flos": 25775853655680.0,
      "grad_norm": 1.9229075993352684,
      "language_loss": 0.76262629,
      "learning_rate": 1.91072865486821e-07,
      "loss": 0.78385615,
      "num_input_tokens_seen": 309914175,
      "step": 14369,
      "time_per_iteration": 4.005048036575317
    },
    {
      "auxiliary_loss_clip": 0.01080002,
      "auxiliary_loss_mlp": 0.01036432,
      "balance_loss_clip": 1.03517222,
      "balance_loss_mlp": 1.02372432,
      "epoch": 0.8639711408387194,
      "flos": 23369443948800.0,
      "grad_norm": 1.6307834614385197,
      "language_loss": 0.64065617,
      "learning_rate": 1.9090677456122294e-07,
      "loss": 0.66182053,
      "num_input_tokens_seen": 309932395,
      "step": 14370,
      "time_per_iteration": 2.5659241676330566
    },
    {
      "auxiliary_loss_clip": 0.0104499,
      "auxiliary_loss_mlp": 0.01031135,
      "balance_loss_clip": 1.0356369,
      "balance_loss_mlp": 1.01872516,
      "epoch": 0.8640312640913873,
      "flos": 22127221946880.0,
      "grad_norm": 1.5236329362948935,
      "language_loss": 0.66428733,
      "learning_rate": 1.907407522366209e-07,
      "loss": 0.68504858,
      "num_input_tokens_seen": 309951720,
      "step": 14371,
      "time_per_iteration": 4.12037992477417
    },
    {
      "auxiliary_loss_clip": 0.01011906,
      "auxiliary_loss_mlp": 0.00999251,
      "balance_loss_clip": 1.00835991,
      "balance_loss_mlp": 0.99807054,
      "epoch": 0.8640913873440553,
      "flos": 57571735944960.0,
      "grad_norm": 0.8648720129938896,
      "language_loss": 0.5696491,
      "learning_rate": 1.905747985193107e-07,
      "loss": 0.58976066,
      "num_input_tokens_seen": 310006120,
      "step": 14372,
      "time_per_iteration": 3.0109968185424805
    },
    {
      "auxiliary_loss_clip": 0.01101632,
      "auxiliary_loss_mlp": 0.01032255,
      "balance_loss_clip": 1.03572536,
      "balance_loss_mlp": 1.01992261,
      "epoch": 0.8641515105967232,
      "flos": 23987430466560.0,
      "grad_norm": 1.6809363842205356,
      "language_loss": 0.79703593,
      "learning_rate": 1.9040891341558597e-07,
      "loss": 0.81837475,
      "num_input_tokens_seen": 310026740,
      "step": 14373,
      "time_per_iteration": 2.511866807937622
    },
    {
      "auxiliary_loss_clip": 0.01103569,
      "auxiliary_loss_mlp": 0.01026325,
      "balance_loss_clip": 1.03484464,
      "balance_loss_mlp": 1.0143261,
      "epoch": 0.8642116338493913,
      "flos": 19062749122560.0,
      "grad_norm": 2.140796543609451,
      "language_loss": 0.64072603,
      "learning_rate": 1.9024309693173656e-07,
      "loss": 0.66202497,
      "num_input_tokens_seen": 310044135,
      "step": 14374,
      "time_per_iteration": 2.442509412765503
    },
    {
      "auxiliary_loss_clip": 0.01071509,
      "auxiliary_loss_mlp": 0.0103304,
      "balance_loss_clip": 1.03320074,
      "balance_loss_mlp": 1.02101195,
      "epoch": 0.8642717571020592,
      "flos": 18254148105600.0,
      "grad_norm": 2.7212156678102994,
      "language_loss": 0.77190328,
      "learning_rate": 1.9007734907404993e-07,
      "loss": 0.79294878,
      "num_input_tokens_seen": 310061560,
      "step": 14375,
      "time_per_iteration": 2.5920231342315674
    },
    {
      "auxiliary_loss_clip": 0.01060791,
      "auxiliary_loss_mlp": 0.00782888,
      "balance_loss_clip": 1.0340364,
      "balance_loss_mlp": 1.01010776,
      "epoch": 0.8643318803547272,
      "flos": 57663270777600.0,
      "grad_norm": 1.660335461768411,
      "language_loss": 0.606264,
      "learning_rate": 1.899116698488117e-07,
      "loss": 0.62470078,
      "num_input_tokens_seen": 310087310,
      "step": 14376,
      "time_per_iteration": 2.9682211875915527
    },
    {
      "auxiliary_loss_clip": 0.01068613,
      "auxiliary_loss_mlp": 0.01035262,
      "balance_loss_clip": 1.03394544,
      "balance_loss_mlp": 1.0238831,
      "epoch": 0.8643920036073952,
      "flos": 19609524927360.0,
      "grad_norm": 1.4634806683649446,
      "language_loss": 0.66316885,
      "learning_rate": 1.8974605926230457e-07,
      "loss": 0.68420762,
      "num_input_tokens_seen": 310106260,
      "step": 14377,
      "time_per_iteration": 2.552854537963867
    },
    {
      "auxiliary_loss_clip": 0.01079554,
      "auxiliary_loss_mlp": 0.01037167,
      "balance_loss_clip": 1.03336227,
      "balance_loss_mlp": 1.02353525,
      "epoch": 0.8644521268600631,
      "flos": 20850346298880.0,
      "grad_norm": 1.5185510818639527,
      "language_loss": 0.70227075,
      "learning_rate": 1.8958051732080804e-07,
      "loss": 0.72343791,
      "num_input_tokens_seen": 310125305,
      "step": 14378,
      "time_per_iteration": 3.91347336769104
    },
    {
      "auxiliary_loss_clip": 0.01020441,
      "auxiliary_loss_mlp": 0.01002364,
      "balance_loss_clip": 1.00831139,
      "balance_loss_mlp": 1.00134468,
      "epoch": 0.8645122501127311,
      "flos": 66719550101760.0,
      "grad_norm": 0.805008810325254,
      "language_loss": 0.6030696,
      "learning_rate": 1.894150440305995e-07,
      "loss": 0.62329763,
      "num_input_tokens_seen": 310189270,
      "step": 14379,
      "time_per_iteration": 3.107417345046997
    },
    {
      "auxiliary_loss_clip": 0.01080513,
      "auxiliary_loss_mlp": 0.01030926,
      "balance_loss_clip": 1.03432047,
      "balance_loss_mlp": 1.01886165,
      "epoch": 0.864572373365399,
      "flos": 21690009601920.0,
      "grad_norm": 1.9557671766676492,
      "language_loss": 0.74418366,
      "learning_rate": 1.8924963939795478e-07,
      "loss": 0.76529813,
      "num_input_tokens_seen": 310208395,
      "step": 14380,
      "time_per_iteration": 2.5550310611724854
    },
    {
      "auxiliary_loss_clip": 0.01076811,
      "auxiliary_loss_mlp": 0.01032901,
      "balance_loss_clip": 1.03350282,
      "balance_loss_mlp": 1.02015209,
      "epoch": 0.8646324966180671,
      "flos": 20266402896000.0,
      "grad_norm": 2.3066181854298464,
      "language_loss": 0.75574088,
      "learning_rate": 1.8908430342914473e-07,
      "loss": 0.776838,
      "num_input_tokens_seen": 310227415,
      "step": 14381,
      "time_per_iteration": 2.50007963180542
    },
    {
      "auxiliary_loss_clip": 0.01078917,
      "auxiliary_loss_mlp": 0.01031227,
      "balance_loss_clip": 1.03442943,
      "balance_loss_mlp": 1.02019453,
      "epoch": 0.864692619870735,
      "flos": 11946188050560.0,
      "grad_norm": 2.542596483785563,
      "language_loss": 0.84089983,
      "learning_rate": 1.8891903613043892e-07,
      "loss": 0.8620013,
      "num_input_tokens_seen": 310242625,
      "step": 14382,
      "time_per_iteration": 2.487907648086548
    },
    {
      "auxiliary_loss_clip": 0.01095768,
      "auxiliary_loss_mlp": 0.01034164,
      "balance_loss_clip": 1.03592396,
      "balance_loss_mlp": 1.02171862,
      "epoch": 0.864752743123403,
      "flos": 21470703114240.0,
      "grad_norm": 2.422291245023437,
      "language_loss": 0.75641978,
      "learning_rate": 1.8875383750810504e-07,
      "loss": 0.77771914,
      "num_input_tokens_seen": 310260585,
      "step": 14383,
      "time_per_iteration": 2.4710464477539062
    },
    {
      "auxiliary_loss_clip": 0.01083059,
      "auxiliary_loss_mlp": 0.01027254,
      "balance_loss_clip": 1.03694248,
      "balance_loss_mlp": 1.01536274,
      "epoch": 0.8648128663760709,
      "flos": 19530018172800.0,
      "grad_norm": 1.7605897013623193,
      "language_loss": 0.85087413,
      "learning_rate": 1.8858870756840738e-07,
      "loss": 0.87197721,
      "num_input_tokens_seen": 310277210,
      "step": 14384,
      "time_per_iteration": 2.516620635986328
    },
    {
      "auxiliary_loss_clip": 0.01089243,
      "auxiliary_loss_mlp": 0.01031646,
      "balance_loss_clip": 1.0334456,
      "balance_loss_mlp": 1.01996899,
      "epoch": 0.8648729896287389,
      "flos": 21287953693440.0,
      "grad_norm": 1.7114596218779896,
      "language_loss": 0.8087551,
      "learning_rate": 1.884236463176072e-07,
      "loss": 0.82996398,
      "num_input_tokens_seen": 310296610,
      "step": 14385,
      "time_per_iteration": 2.480952739715576
    },
    {
      "auxiliary_loss_clip": 0.01088159,
      "auxiliary_loss_mlp": 0.01029558,
      "balance_loss_clip": 1.03755355,
      "balance_loss_mlp": 1.01721978,
      "epoch": 0.8649331128814068,
      "flos": 24604483230720.0,
      "grad_norm": 1.9821178530434644,
      "language_loss": 0.7264657,
      "learning_rate": 1.8825865376196437e-07,
      "loss": 0.74764287,
      "num_input_tokens_seen": 310316830,
      "step": 14386,
      "time_per_iteration": 2.5641837120056152
    },
    {
      "auxiliary_loss_clip": 0.01089755,
      "auxiliary_loss_mlp": 0.01034655,
      "balance_loss_clip": 1.03458083,
      "balance_loss_mlp": 1.0221796,
      "epoch": 0.8649932361340749,
      "flos": 15377811742080.0,
      "grad_norm": 1.8100123878409637,
      "language_loss": 0.8214758,
      "learning_rate": 1.8809372990773476e-07,
      "loss": 0.84271991,
      "num_input_tokens_seen": 310334355,
      "step": 14387,
      "time_per_iteration": 2.4554600715637207
    },
    {
      "auxiliary_loss_clip": 0.01101359,
      "auxiliary_loss_mlp": 0.01026838,
      "balance_loss_clip": 1.03522468,
      "balance_loss_mlp": 1.01523256,
      "epoch": 0.8650533593867428,
      "flos": 19901227276800.0,
      "grad_norm": 1.8428830786662658,
      "language_loss": 0.68590343,
      "learning_rate": 1.8792887476117224e-07,
      "loss": 0.70718551,
      "num_input_tokens_seen": 310352900,
      "step": 14388,
      "time_per_iteration": 2.4846878051757812
    },
    {
      "auxiliary_loss_clip": 0.01069725,
      "auxiliary_loss_mlp": 0.01032622,
      "balance_loss_clip": 1.03550351,
      "balance_loss_mlp": 1.02145803,
      "epoch": 0.8651134826394108,
      "flos": 25626931868160.0,
      "grad_norm": 1.5004591300114027,
      "language_loss": 0.90366977,
      "learning_rate": 1.877640883285283e-07,
      "loss": 0.92469323,
      "num_input_tokens_seen": 310372855,
      "step": 14389,
      "time_per_iteration": 2.5894806385040283
    },
    {
      "auxiliary_loss_clip": 0.01060672,
      "auxiliary_loss_mlp": 0.00781095,
      "balance_loss_clip": 1.03655303,
      "balance_loss_mlp": 1.00806236,
      "epoch": 0.8651736058920788,
      "flos": 18734525619840.0,
      "grad_norm": 1.602465787174566,
      "language_loss": 0.70683658,
      "learning_rate": 1.8759937061605212e-07,
      "loss": 0.72525418,
      "num_input_tokens_seen": 310391595,
      "step": 14390,
      "time_per_iteration": 2.6144251823425293
    },
    {
      "auxiliary_loss_clip": 0.01104901,
      "auxiliary_loss_mlp": 0.01032992,
      "balance_loss_clip": 1.03546166,
      "balance_loss_mlp": 1.02063549,
      "epoch": 0.8652337291447467,
      "flos": 20776765288320.0,
      "grad_norm": 1.573645123752967,
      "language_loss": 0.82385993,
      "learning_rate": 1.8743472162998941e-07,
      "loss": 0.8452388,
      "num_input_tokens_seen": 310410090,
      "step": 14391,
      "time_per_iteration": 2.4699625968933105
    },
    {
      "auxiliary_loss_clip": 0.01002717,
      "auxiliary_loss_mlp": 0.01000466,
      "balance_loss_clip": 1.01336384,
      "balance_loss_mlp": 0.99918431,
      "epoch": 0.8652938523974147,
      "flos": 64227887464320.0,
      "grad_norm": 0.8012621255877996,
      "language_loss": 0.6805166,
      "learning_rate": 1.8727014137658337e-07,
      "loss": 0.70054841,
      "num_input_tokens_seen": 310470055,
      "step": 14392,
      "time_per_iteration": 3.060218334197998
    },
    {
      "auxiliary_loss_clip": 0.01097197,
      "auxiliary_loss_mlp": 0.01031718,
      "balance_loss_clip": 1.035092,
      "balance_loss_mlp": 1.01893318,
      "epoch": 0.8653539756500827,
      "flos": 18040587793920.0,
      "grad_norm": 2.0360508004548556,
      "language_loss": 0.75912416,
      "learning_rate": 1.8710562986207523e-07,
      "loss": 0.78041327,
      "num_input_tokens_seen": 310487665,
      "step": 14393,
      "time_per_iteration": 2.4742772579193115
    },
    {
      "auxiliary_loss_clip": 0.01081491,
      "auxiliary_loss_mlp": 0.01033347,
      "balance_loss_clip": 1.03193223,
      "balance_loss_mlp": 1.02119327,
      "epoch": 0.8654140989027507,
      "flos": 17382416935680.0,
      "grad_norm": 2.404715480824046,
      "language_loss": 0.73675168,
      "learning_rate": 1.8694118709270357e-07,
      "loss": 0.75790012,
      "num_input_tokens_seen": 310506130,
      "step": 14394,
      "time_per_iteration": 2.4694535732269287
    },
    {
      "auxiliary_loss_clip": 0.01093753,
      "auxiliary_loss_mlp": 0.01033009,
      "balance_loss_clip": 1.03391612,
      "balance_loss_mlp": 1.0199796,
      "epoch": 0.8654742221554186,
      "flos": 53284862448000.0,
      "grad_norm": 2.483713687961178,
      "language_loss": 0.64778,
      "learning_rate": 1.867768130747036e-07,
      "loss": 0.66904759,
      "num_input_tokens_seen": 310532445,
      "step": 14395,
      "time_per_iteration": 2.7875232696533203
    },
    {
      "auxiliary_loss_clip": 0.01086695,
      "auxiliary_loss_mlp": 0.01035431,
      "balance_loss_clip": 1.03361583,
      "balance_loss_mlp": 1.02336121,
      "epoch": 0.8655343454080866,
      "flos": 23914711382400.0,
      "grad_norm": 1.5740985234907725,
      "language_loss": 0.67918807,
      "learning_rate": 1.8661250781430838e-07,
      "loss": 0.70040929,
      "num_input_tokens_seen": 310552300,
      "step": 14396,
      "time_per_iteration": 2.4845588207244873
    },
    {
      "auxiliary_loss_clip": 0.01095702,
      "auxiliary_loss_mlp": 0.0103605,
      "balance_loss_clip": 1.03603137,
      "balance_loss_mlp": 1.02378893,
      "epoch": 0.8655944686607545,
      "flos": 24097209408000.0,
      "grad_norm": 4.906784461845033,
      "language_loss": 0.69095188,
      "learning_rate": 1.8644827131774954e-07,
      "loss": 0.71226937,
      "num_input_tokens_seen": 310572710,
      "step": 14397,
      "time_per_iteration": 2.5117950439453125
    },
    {
      "auxiliary_loss_clip": 0.01084867,
      "auxiliary_loss_mlp": 0.01028302,
      "balance_loss_clip": 1.03384447,
      "balance_loss_mlp": 1.01691771,
      "epoch": 0.8656545919134225,
      "flos": 23112718467840.0,
      "grad_norm": 1.7726908334626978,
      "language_loss": 0.63292503,
      "learning_rate": 1.86284103591253e-07,
      "loss": 0.65405679,
      "num_input_tokens_seen": 310592460,
      "step": 14398,
      "time_per_iteration": 2.52276349067688
    },
    {
      "auxiliary_loss_clip": 0.01067604,
      "auxiliary_loss_mlp": 0.0103137,
      "balance_loss_clip": 1.03581762,
      "balance_loss_mlp": 1.01826334,
      "epoch": 0.8657147151660904,
      "flos": 21141761339520.0,
      "grad_norm": 2.2199238416803326,
      "language_loss": 0.76109624,
      "learning_rate": 1.8612000464104517e-07,
      "loss": 0.78208596,
      "num_input_tokens_seen": 310609375,
      "step": 14399,
      "time_per_iteration": 2.5639255046844482
    },
    {
      "auxiliary_loss_clip": 0.01084803,
      "auxiliary_loss_mlp": 0.01026827,
      "balance_loss_clip": 1.03411913,
      "balance_loss_mlp": 1.01543081,
      "epoch": 0.8657748384187585,
      "flos": 16289439943680.0,
      "grad_norm": 1.936478877469833,
      "language_loss": 0.93071848,
      "learning_rate": 1.8595597447334855e-07,
      "loss": 0.9518348,
      "num_input_tokens_seen": 310627405,
      "step": 14400,
      "time_per_iteration": 2.4516475200653076
    },
    {
      "auxiliary_loss_clip": 0.01047964,
      "auxiliary_loss_mlp": 0.01036984,
      "balance_loss_clip": 1.03507876,
      "balance_loss_mlp": 1.02525997,
      "epoch": 0.8658349616714264,
      "flos": 30843890179200.0,
      "grad_norm": 2.031899192966752,
      "language_loss": 0.67453265,
      "learning_rate": 1.8579201309438353e-07,
      "loss": 0.69538212,
      "num_input_tokens_seen": 310649945,
      "step": 14401,
      "time_per_iteration": 2.714660406112671
    },
    {
      "auxiliary_loss_clip": 0.01093621,
      "auxiliary_loss_mlp": 0.01029314,
      "balance_loss_clip": 1.03379726,
      "balance_loss_mlp": 1.01738095,
      "epoch": 0.8658950849240944,
      "flos": 18952862440320.0,
      "grad_norm": 1.9808252493382519,
      "language_loss": 0.73835278,
      "learning_rate": 1.8562812051036714e-07,
      "loss": 0.75958216,
      "num_input_tokens_seen": 310668285,
      "step": 14402,
      "time_per_iteration": 2.472064256668091
    },
    {
      "auxiliary_loss_clip": 0.01037346,
      "auxiliary_loss_mlp": 0.0103086,
      "balance_loss_clip": 1.03284478,
      "balance_loss_mlp": 1.01915371,
      "epoch": 0.8659552081767624,
      "flos": 23364344217600.0,
      "grad_norm": 2.222115425870408,
      "language_loss": 0.75052655,
      "learning_rate": 1.8546429672751397e-07,
      "loss": 0.77120864,
      "num_input_tokens_seen": 310687015,
      "step": 14403,
      "time_per_iteration": 2.668152093887329
    },
    {
      "auxiliary_loss_clip": 0.01078057,
      "auxiliary_loss_mlp": 0.01034306,
      "balance_loss_clip": 1.0337522,
      "balance_loss_mlp": 1.02131212,
      "epoch": 0.8660153314294303,
      "flos": 23841992298240.0,
      "grad_norm": 1.822312171553705,
      "language_loss": 0.73384976,
      "learning_rate": 1.853005417520368e-07,
      "loss": 0.75497341,
      "num_input_tokens_seen": 310707580,
      "step": 14404,
      "time_per_iteration": 2.5422472953796387
    },
    {
      "auxiliary_loss_clip": 0.0106603,
      "auxiliary_loss_mlp": 0.01030591,
      "balance_loss_clip": 1.03504014,
      "balance_loss_mlp": 1.01828837,
      "epoch": 0.8660754546820983,
      "flos": 23112467072640.0,
      "grad_norm": 1.7998829825882579,
      "language_loss": 0.70683181,
      "learning_rate": 1.851368555901447e-07,
      "loss": 0.7277981,
      "num_input_tokens_seen": 310727300,
      "step": 14405,
      "time_per_iteration": 2.5714354515075684
    },
    {
      "auxiliary_loss_clip": 0.01094097,
      "auxiliary_loss_mlp": 0.00784018,
      "balance_loss_clip": 1.03429174,
      "balance_loss_mlp": 1.01051497,
      "epoch": 0.8661355779347663,
      "flos": 14391991998720.0,
      "grad_norm": 1.717676860369089,
      "language_loss": 0.66417897,
      "learning_rate": 1.8497323824804467e-07,
      "loss": 0.68296015,
      "num_input_tokens_seen": 310744935,
      "step": 14406,
      "time_per_iteration": 2.466764450073242
    },
    {
      "auxiliary_loss_clip": 0.01082548,
      "auxiliary_loss_mlp": 0.01024174,
      "balance_loss_clip": 1.03513241,
      "balance_loss_mlp": 1.01311779,
      "epoch": 0.8661957011874343,
      "flos": 21870137329920.0,
      "grad_norm": 2.044169927104571,
      "language_loss": 0.82976329,
      "learning_rate": 1.8480968973194177e-07,
      "loss": 0.8508305,
      "num_input_tokens_seen": 310765085,
      "step": 14407,
      "time_per_iteration": 5.404285907745361
    },
    {
      "auxiliary_loss_clip": 0.01093215,
      "auxiliary_loss_mlp": 0.0103508,
      "balance_loss_clip": 1.03629041,
      "balance_loss_mlp": 1.02371335,
      "epoch": 0.8662558244401022,
      "flos": 21835160461440.0,
      "grad_norm": 1.7812882157547132,
      "language_loss": 0.69898272,
      "learning_rate": 1.8464621004803748e-07,
      "loss": 0.72026575,
      "num_input_tokens_seen": 310783260,
      "step": 14408,
      "time_per_iteration": 2.507185935974121
    },
    {
      "auxiliary_loss_clip": 0.01082447,
      "auxiliary_loss_mlp": 0.01031483,
      "balance_loss_clip": 1.03394604,
      "balance_loss_mlp": 1.02035511,
      "epoch": 0.8663159476927702,
      "flos": 17384104874880.0,
      "grad_norm": 1.8654923238140926,
      "language_loss": 0.77419215,
      "learning_rate": 1.844827992025304e-07,
      "loss": 0.79533148,
      "num_input_tokens_seen": 310801970,
      "step": 14409,
      "time_per_iteration": 2.485217809677124
    },
    {
      "auxiliary_loss_clip": 0.01095508,
      "auxiliary_loss_mlp": 0.01030499,
      "balance_loss_clip": 1.0363729,
      "balance_loss_mlp": 1.01785684,
      "epoch": 0.8663760709454381,
      "flos": 22747722416640.0,
      "grad_norm": 1.9289014296633187,
      "language_loss": 0.77215588,
      "learning_rate": 1.8431945720161757e-07,
      "loss": 0.79341596,
      "num_input_tokens_seen": 310822070,
      "step": 14410,
      "time_per_iteration": 3.892517328262329
    },
    {
      "auxiliary_loss_clip": 0.01062061,
      "auxiliary_loss_mlp": 0.0103139,
      "balance_loss_clip": 1.03306997,
      "balance_loss_mlp": 1.01929069,
      "epoch": 0.8664361941981061,
      "flos": 17376850327680.0,
      "grad_norm": 1.965404520009851,
      "language_loss": 0.77610034,
      "learning_rate": 1.8415618405149315e-07,
      "loss": 0.79703486,
      "num_input_tokens_seen": 310838355,
      "step": 14411,
      "time_per_iteration": 2.5273244380950928
    },
    {
      "auxiliary_loss_clip": 0.01074831,
      "auxiliary_loss_mlp": 0.01031965,
      "balance_loss_clip": 1.03307319,
      "balance_loss_mlp": 1.02078366,
      "epoch": 0.866496317450774,
      "flos": 16034438315520.0,
      "grad_norm": 2.141652419848165,
      "language_loss": 0.73671895,
      "learning_rate": 1.8399297975834794e-07,
      "loss": 0.75778693,
      "num_input_tokens_seen": 310856055,
      "step": 14412,
      "time_per_iteration": 2.504023551940918
    },
    {
      "auxiliary_loss_clip": 0.01086971,
      "auxiliary_loss_mlp": 0.0078536,
      "balance_loss_clip": 1.03345585,
      "balance_loss_mlp": 1.01084685,
      "epoch": 0.8665564407034421,
      "flos": 20814830726400.0,
      "grad_norm": 1.8730450840445714,
      "language_loss": 0.69458073,
      "learning_rate": 1.83829844328371e-07,
      "loss": 0.71330404,
      "num_input_tokens_seen": 310876695,
      "step": 14413,
      "time_per_iteration": 2.509875774383545
    },
    {
      "auxiliary_loss_clip": 0.01093308,
      "auxiliary_loss_mlp": 0.01031886,
      "balance_loss_clip": 1.03592503,
      "balance_loss_mlp": 1.02008462,
      "epoch": 0.86661656395611,
      "flos": 15815167741440.0,
      "grad_norm": 2.066163244266518,
      "language_loss": 0.6326617,
      "learning_rate": 1.8366677776774874e-07,
      "loss": 0.65391362,
      "num_input_tokens_seen": 310893880,
      "step": 14414,
      "time_per_iteration": 2.4538064002990723
    },
    {
      "auxiliary_loss_clip": 0.0107319,
      "auxiliary_loss_mlp": 0.0078202,
      "balance_loss_clip": 1.0353775,
      "balance_loss_mlp": 1.00707245,
      "epoch": 0.866676687208778,
      "flos": 23036910814080.0,
      "grad_norm": 1.5689687316819567,
      "language_loss": 0.63706315,
      "learning_rate": 1.8350378008266377e-07,
      "loss": 0.65561521,
      "num_input_tokens_seen": 310914145,
      "step": 14415,
      "time_per_iteration": 2.575549364089966
    },
    {
      "auxiliary_loss_clip": 0.01003033,
      "auxiliary_loss_mlp": 0.00999894,
      "balance_loss_clip": 1.01099765,
      "balance_loss_mlp": 0.99879724,
      "epoch": 0.866736810461446,
      "flos": 63802275212160.0,
      "grad_norm": 0.7960438584738667,
      "language_loss": 0.60397005,
      "learning_rate": 1.8334085127929754e-07,
      "loss": 0.6239993,
      "num_input_tokens_seen": 310972825,
      "step": 14416,
      "time_per_iteration": 4.601866006851196
    },
    {
      "auxiliary_loss_clip": 0.01093988,
      "auxiliary_loss_mlp": 0.00784744,
      "balance_loss_clip": 1.0338372,
      "balance_loss_mlp": 1.01063156,
      "epoch": 0.8667969337141139,
      "flos": 20449367798400.0,
      "grad_norm": 1.7660233120590305,
      "language_loss": 0.74721551,
      "learning_rate": 1.831779913638285e-07,
      "loss": 0.76600289,
      "num_input_tokens_seen": 310992050,
      "step": 14417,
      "time_per_iteration": 2.479079008102417
    },
    {
      "auxiliary_loss_clip": 0.01079034,
      "auxiliary_loss_mlp": 0.01034942,
      "balance_loss_clip": 1.03375292,
      "balance_loss_mlp": 1.02368259,
      "epoch": 0.866857056966782,
      "flos": 21653703930240.0,
      "grad_norm": 2.009781431927632,
      "language_loss": 0.75131309,
      "learning_rate": 1.830152003424319e-07,
      "loss": 0.77245277,
      "num_input_tokens_seen": 311011105,
      "step": 14418,
      "time_per_iteration": 2.523104429244995
    },
    {
      "auxiliary_loss_clip": 0.0109068,
      "auxiliary_loss_mlp": 0.01033241,
      "balance_loss_clip": 1.03390741,
      "balance_loss_mlp": 1.02147484,
      "epoch": 0.8669171802194499,
      "flos": 22852832590080.0,
      "grad_norm": 1.5096522633567193,
      "language_loss": 0.68330371,
      "learning_rate": 1.8285247822128126e-07,
      "loss": 0.70454288,
      "num_input_tokens_seen": 311032080,
      "step": 14419,
      "time_per_iteration": 2.4989635944366455
    },
    {
      "auxiliary_loss_clip": 0.01094745,
      "auxiliary_loss_mlp": 0.01030226,
      "balance_loss_clip": 1.03558409,
      "balance_loss_mlp": 1.01924729,
      "epoch": 0.8669773034721179,
      "flos": 18734166483840.0,
      "grad_norm": 1.678164629918293,
      "language_loss": 0.78284943,
      "learning_rate": 1.826898250065465e-07,
      "loss": 0.8040992,
      "num_input_tokens_seen": 311049735,
      "step": 14420,
      "time_per_iteration": 2.4932966232299805
    },
    {
      "auxiliary_loss_clip": 0.01088878,
      "auxiliary_loss_mlp": 0.0103092,
      "balance_loss_clip": 1.03422034,
      "balance_loss_mlp": 1.01876068,
      "epoch": 0.8670374267247858,
      "flos": 18916018064640.0,
      "grad_norm": 1.5465729079255368,
      "language_loss": 0.8370347,
      "learning_rate": 1.8252724070439586e-07,
      "loss": 0.85823268,
      "num_input_tokens_seen": 311067675,
      "step": 14421,
      "time_per_iteration": 2.460062265396118
    },
    {
      "auxiliary_loss_clip": 0.01009585,
      "auxiliary_loss_mlp": 0.00999574,
      "balance_loss_clip": 1.01100755,
      "balance_loss_mlp": 0.99835247,
      "epoch": 0.8670975499774538,
      "flos": 48814527214080.0,
      "grad_norm": 0.7032300346641079,
      "language_loss": 0.49142066,
      "learning_rate": 1.823647253209941e-07,
      "loss": 0.51151228,
      "num_input_tokens_seen": 311126605,
      "step": 14422,
      "time_per_iteration": 3.1209442615509033
    },
    {
      "auxiliary_loss_clip": 0.01080869,
      "auxiliary_loss_mlp": 0.0078321,
      "balance_loss_clip": 1.03461194,
      "balance_loss_mlp": 1.01089549,
      "epoch": 0.8671576732301217,
      "flos": 26136145025280.0,
      "grad_norm": 1.5199134324055197,
      "language_loss": 0.73690403,
      "learning_rate": 1.8220227886250417e-07,
      "loss": 0.75554484,
      "num_input_tokens_seen": 311147325,
      "step": 14423,
      "time_per_iteration": 2.553122043609619
    },
    {
      "auxiliary_loss_clip": 0.01057291,
      "auxiliary_loss_mlp": 0.01028422,
      "balance_loss_clip": 1.03032231,
      "balance_loss_mlp": 1.01584005,
      "epoch": 0.8672177964827897,
      "flos": 18367446579840.0,
      "grad_norm": 1.5072241248761646,
      "language_loss": 0.76713389,
      "learning_rate": 1.8203990133508684e-07,
      "loss": 0.78799105,
      "num_input_tokens_seen": 311165385,
      "step": 14424,
      "time_per_iteration": 2.584876298904419
    },
    {
      "auxiliary_loss_clip": 0.01062704,
      "auxiliary_loss_mlp": 0.01035317,
      "balance_loss_clip": 1.03180504,
      "balance_loss_mlp": 1.02323532,
      "epoch": 0.8672779197354576,
      "flos": 28545355992960.0,
      "grad_norm": 2.844823730131768,
      "language_loss": 0.712111,
      "learning_rate": 1.8187759274489767e-07,
      "loss": 0.73309124,
      "num_input_tokens_seen": 311185860,
      "step": 14425,
      "time_per_iteration": 2.6039299964904785
    },
    {
      "auxiliary_loss_clip": 0.01095273,
      "auxiliary_loss_mlp": 0.01030853,
      "balance_loss_clip": 1.03590155,
      "balance_loss_mlp": 1.01853871,
      "epoch": 0.8673380429881257,
      "flos": 22382474970240.0,
      "grad_norm": 1.785028220009416,
      "language_loss": 0.68044257,
      "learning_rate": 1.817153530980926e-07,
      "loss": 0.70170385,
      "num_input_tokens_seen": 311205810,
      "step": 14426,
      "time_per_iteration": 2.55110239982605
    },
    {
      "auxiliary_loss_clip": 0.0106723,
      "auxiliary_loss_mlp": 0.01025273,
      "balance_loss_clip": 1.03578138,
      "balance_loss_mlp": 1.01274967,
      "epoch": 0.8673981662407936,
      "flos": 20996430912000.0,
      "grad_norm": 1.7291260375608357,
      "language_loss": 0.70591545,
      "learning_rate": 1.815531824008234e-07,
      "loss": 0.72684044,
      "num_input_tokens_seen": 311226080,
      "step": 14427,
      "time_per_iteration": 2.547909736633301
    },
    {
      "auxiliary_loss_clip": 0.01075908,
      "auxiliary_loss_mlp": 0.01028645,
      "balance_loss_clip": 1.03526962,
      "balance_loss_mlp": 1.01671243,
      "epoch": 0.8674582894934616,
      "flos": 24426797627520.0,
      "grad_norm": 1.6232513705115956,
      "language_loss": 0.68088883,
      "learning_rate": 1.8139108065924004e-07,
      "loss": 0.70193434,
      "num_input_tokens_seen": 311246380,
      "step": 14428,
      "time_per_iteration": 2.6174912452697754
    },
    {
      "auxiliary_loss_clip": 0.01074579,
      "auxiliary_loss_mlp": 0.0102734,
      "balance_loss_clip": 1.03597963,
      "balance_loss_mlp": 1.01645005,
      "epoch": 0.8675184127461296,
      "flos": 20737514701440.0,
      "grad_norm": 2.6067475841888776,
      "language_loss": 0.70718443,
      "learning_rate": 1.812290478794889e-07,
      "loss": 0.72820359,
      "num_input_tokens_seen": 311266465,
      "step": 14429,
      "time_per_iteration": 2.5142343044281006
    },
    {
      "auxiliary_loss_clip": 0.0108277,
      "auxiliary_loss_mlp": 0.0102777,
      "balance_loss_clip": 1.03497553,
      "balance_loss_mlp": 1.01598644,
      "epoch": 0.8675785359987975,
      "flos": 19135647774720.0,
      "grad_norm": 2.046750130156718,
      "language_loss": 0.66992331,
      "learning_rate": 1.810670840677151e-07,
      "loss": 0.69102871,
      "num_input_tokens_seen": 311285075,
      "step": 14430,
      "time_per_iteration": 2.5104892253875732
    },
    {
      "auxiliary_loss_clip": 0.01059568,
      "auxiliary_loss_mlp": 0.01033484,
      "balance_loss_clip": 1.03500676,
      "balance_loss_mlp": 1.02075243,
      "epoch": 0.8676386592514655,
      "flos": 22710662559360.0,
      "grad_norm": 2.268956116768792,
      "language_loss": 0.68991667,
      "learning_rate": 1.8090518923005948e-07,
      "loss": 0.71084726,
      "num_input_tokens_seen": 311303230,
      "step": 14431,
      "time_per_iteration": 2.5608632564544678
    },
    {
      "auxiliary_loss_clip": 0.01091441,
      "auxiliary_loss_mlp": 0.01032983,
      "balance_loss_clip": 1.03343737,
      "balance_loss_mlp": 1.0208118,
      "epoch": 0.8676987825041335,
      "flos": 14209853109120.0,
      "grad_norm": 2.7239039204909976,
      "language_loss": 0.63489181,
      "learning_rate": 1.8074336337266116e-07,
      "loss": 0.65613604,
      "num_input_tokens_seen": 311318070,
      "step": 14432,
      "time_per_iteration": 2.501511335372925
    },
    {
      "auxiliary_loss_clip": 0.01094395,
      "auxiliary_loss_mlp": 0.01035998,
      "balance_loss_clip": 1.03568792,
      "balance_loss_mlp": 1.02501893,
      "epoch": 0.8677589057568015,
      "flos": 13589927256960.0,
      "grad_norm": 1.916569742549719,
      "language_loss": 0.78306913,
      "learning_rate": 1.8058160650165656e-07,
      "loss": 0.80437309,
      "num_input_tokens_seen": 311334885,
      "step": 14433,
      "time_per_iteration": 2.4491987228393555
    },
    {
      "auxiliary_loss_clip": 0.0101282,
      "auxiliary_loss_mlp": 0.01001889,
      "balance_loss_clip": 1.01037836,
      "balance_loss_mlp": 1.00071478,
      "epoch": 0.8678190290094694,
      "flos": 68933657370240.0,
      "grad_norm": 1.196361005493811,
      "language_loss": 0.58438575,
      "learning_rate": 1.804199186231805e-07,
      "loss": 0.60453284,
      "num_input_tokens_seen": 311399780,
      "step": 14434,
      "time_per_iteration": 3.204069137573242
    },
    {
      "auxiliary_loss_clip": 0.01078974,
      "auxiliary_loss_mlp": 0.01032466,
      "balance_loss_clip": 1.03241467,
      "balance_loss_mlp": 1.02138579,
      "epoch": 0.8678791522621374,
      "flos": 32557726776960.0,
      "grad_norm": 1.772296103460185,
      "language_loss": 0.80097556,
      "learning_rate": 1.802582997433628e-07,
      "loss": 0.82209003,
      "num_input_tokens_seen": 311419610,
      "step": 14435,
      "time_per_iteration": 2.595299243927002
    },
    {
      "auxiliary_loss_clip": 0.01081378,
      "auxiliary_loss_mlp": 0.00782678,
      "balance_loss_clip": 1.03216851,
      "balance_loss_mlp": 1.00764465,
      "epoch": 0.8679392755148053,
      "flos": 35042637657600.0,
      "grad_norm": 1.8524409255509766,
      "language_loss": 0.617226,
      "learning_rate": 1.8009674986833322e-07,
      "loss": 0.63586652,
      "num_input_tokens_seen": 311440045,
      "step": 14436,
      "time_per_iteration": 2.6380438804626465
    },
    {
      "auxiliary_loss_clip": 0.01081272,
      "auxiliary_loss_mlp": 0.01028358,
      "balance_loss_clip": 1.03798223,
      "balance_loss_mlp": 1.01539993,
      "epoch": 0.8679993987674733,
      "flos": 18552494471040.0,
      "grad_norm": 2.1895988669964876,
      "language_loss": 0.70453453,
      "learning_rate": 1.7993526900421706e-07,
      "loss": 0.72563076,
      "num_input_tokens_seen": 311456660,
      "step": 14437,
      "time_per_iteration": 2.4976179599761963
    },
    {
      "auxiliary_loss_clip": 0.01070212,
      "auxiliary_loss_mlp": 0.0102767,
      "balance_loss_clip": 1.03434575,
      "balance_loss_mlp": 1.01578474,
      "epoch": 0.8680595220201412,
      "flos": 27454390162560.0,
      "grad_norm": 1.9853595822886019,
      "language_loss": 0.80282462,
      "learning_rate": 1.797738571571381e-07,
      "loss": 0.82380342,
      "num_input_tokens_seen": 311475460,
      "step": 14438,
      "time_per_iteration": 2.6041011810302734
    },
    {
      "auxiliary_loss_clip": 0.01084126,
      "auxiliary_loss_mlp": 0.01022258,
      "balance_loss_clip": 1.03354406,
      "balance_loss_mlp": 1.0106051,
      "epoch": 0.8681196452728093,
      "flos": 19208797822080.0,
      "grad_norm": 2.53776374697169,
      "language_loss": 0.67522347,
      "learning_rate": 1.7961251433321656e-07,
      "loss": 0.69628733,
      "num_input_tokens_seen": 311494575,
      "step": 14439,
      "time_per_iteration": 2.4864957332611084
    },
    {
      "auxiliary_loss_clip": 0.0109085,
      "auxiliary_loss_mlp": 0.01032426,
      "balance_loss_clip": 1.03418827,
      "balance_loss_mlp": 1.02135706,
      "epoch": 0.8681797685254772,
      "flos": 37560442417920.0,
      "grad_norm": 1.5248988350073533,
      "language_loss": 0.63853621,
      "learning_rate": 1.7945124053857085e-07,
      "loss": 0.65976894,
      "num_input_tokens_seen": 311515805,
      "step": 14440,
      "time_per_iteration": 2.6327784061431885
    },
    {
      "auxiliary_loss_clip": 0.01091593,
      "auxiliary_loss_mlp": 0.01028846,
      "balance_loss_clip": 1.03546727,
      "balance_loss_mlp": 1.01683593,
      "epoch": 0.8682398917781452,
      "flos": 23289937194240.0,
      "grad_norm": 1.6098979563491111,
      "language_loss": 0.65765601,
      "learning_rate": 1.7929003577931722e-07,
      "loss": 0.67886043,
      "num_input_tokens_seen": 311536000,
      "step": 14441,
      "time_per_iteration": 2.5107715129852295
    },
    {
      "auxiliary_loss_clip": 0.01085846,
      "auxiliary_loss_mlp": 0.01024265,
      "balance_loss_clip": 1.03647733,
      "balance_loss_mlp": 1.01352477,
      "epoch": 0.8683000150308132,
      "flos": 21872794936320.0,
      "grad_norm": 1.56127051517466,
      "language_loss": 0.66406894,
      "learning_rate": 1.7912890006156722e-07,
      "loss": 0.68517005,
      "num_input_tokens_seen": 311556220,
      "step": 14442,
      "time_per_iteration": 2.4997498989105225
    },
    {
      "auxiliary_loss_clip": 0.01078658,
      "auxiliary_loss_mlp": 0.01032682,
      "balance_loss_clip": 1.03345203,
      "balance_loss_mlp": 1.0188899,
      "epoch": 0.8683601382834811,
      "flos": 14647209108480.0,
      "grad_norm": 1.7886701385723607,
      "language_loss": 0.72135437,
      "learning_rate": 1.7896783339143195e-07,
      "loss": 0.74246782,
      "num_input_tokens_seen": 311572530,
      "step": 14443,
      "time_per_iteration": 2.4866933822631836
    },
    {
      "auxiliary_loss_clip": 0.01103796,
      "auxiliary_loss_mlp": 0.01029027,
      "balance_loss_clip": 1.0349853,
      "balance_loss_mlp": 1.01674294,
      "epoch": 0.8684202615361492,
      "flos": 26359904799360.0,
      "grad_norm": 1.901331424415087,
      "language_loss": 0.83502603,
      "learning_rate": 1.7880683577501877e-07,
      "loss": 0.85635424,
      "num_input_tokens_seen": 311591105,
      "step": 14444,
      "time_per_iteration": 2.5147132873535156
    },
    {
      "auxiliary_loss_clip": 0.01064798,
      "auxiliary_loss_mlp": 0.01027388,
      "balance_loss_clip": 1.03559351,
      "balance_loss_mlp": 1.01502037,
      "epoch": 0.8684803847888171,
      "flos": 20704010290560.0,
      "grad_norm": 1.8192943280136162,
      "language_loss": 0.77091241,
      "learning_rate": 1.7864590721843342e-07,
      "loss": 0.79183424,
      "num_input_tokens_seen": 311608350,
      "step": 14445,
      "time_per_iteration": 3.909639596939087
    },
    {
      "auxiliary_loss_clip": 0.01094656,
      "auxiliary_loss_mlp": 0.0102989,
      "balance_loss_clip": 1.03608871,
      "balance_loss_mlp": 1.0175935,
      "epoch": 0.8685405080414851,
      "flos": 22638123043200.0,
      "grad_norm": 1.9177722077796089,
      "language_loss": 0.67980933,
      "learning_rate": 1.7848504772777728e-07,
      "loss": 0.70105481,
      "num_input_tokens_seen": 311626380,
      "step": 14446,
      "time_per_iteration": 3.9154200553894043
    },
    {
      "auxiliary_loss_clip": 0.01090778,
      "auxiliary_loss_mlp": 0.01031668,
      "balance_loss_clip": 1.03613257,
      "balance_loss_mlp": 1.01945543,
      "epoch": 0.868600631294153,
      "flos": 24822065865600.0,
      "grad_norm": 1.588005544073923,
      "language_loss": 0.8299374,
      "learning_rate": 1.7832425730915102e-07,
      "loss": 0.85116184,
      "num_input_tokens_seen": 311644345,
      "step": 14447,
      "time_per_iteration": 2.515702962875366
    },
    {
      "auxiliary_loss_clip": 0.01025224,
      "auxiliary_loss_mlp": 0.01025456,
      "balance_loss_clip": 1.03283608,
      "balance_loss_mlp": 1.01363635,
      "epoch": 0.868660754546821,
      "flos": 25113983696640.0,
      "grad_norm": 1.8544304848122817,
      "language_loss": 0.74066603,
      "learning_rate": 1.781635359686515e-07,
      "loss": 0.76117283,
      "num_input_tokens_seen": 311663340,
      "step": 14448,
      "time_per_iteration": 2.7022008895874023
    },
    {
      "auxiliary_loss_clip": 0.01078841,
      "auxiliary_loss_mlp": 0.01032774,
      "balance_loss_clip": 1.0335288,
      "balance_loss_mlp": 1.01959515,
      "epoch": 0.8687208777994889,
      "flos": 12677832178560.0,
      "grad_norm": 2.0466642720063155,
      "language_loss": 0.80271941,
      "learning_rate": 1.7800288371237303e-07,
      "loss": 0.82383555,
      "num_input_tokens_seen": 311679860,
      "step": 14449,
      "time_per_iteration": 3.853191614151001
    },
    {
      "auxiliary_loss_clip": 0.01000724,
      "auxiliary_loss_mlp": 0.01003839,
      "balance_loss_clip": 1.00799966,
      "balance_loss_mlp": 1.00266504,
      "epoch": 0.8687810010521569,
      "flos": 65617235573760.0,
      "grad_norm": 0.7999466041093845,
      "language_loss": 0.6060611,
      "learning_rate": 1.7784230054640758e-07,
      "loss": 0.62610674,
      "num_input_tokens_seen": 311738135,
      "step": 14450,
      "time_per_iteration": 3.082252025604248
    },
    {
      "auxiliary_loss_clip": 0.01076951,
      "auxiliary_loss_mlp": 0.0102887,
      "balance_loss_clip": 1.0331614,
      "balance_loss_mlp": 1.0166328,
      "epoch": 0.8688411243048249,
      "flos": 24244012293120.0,
      "grad_norm": 2.17160081883365,
      "language_loss": 0.76062059,
      "learning_rate": 1.7768178647684517e-07,
      "loss": 0.7816788,
      "num_input_tokens_seen": 311756975,
      "step": 14451,
      "time_per_iteration": 2.5809717178344727
    },
    {
      "auxiliary_loss_clip": 0.01090462,
      "auxiliary_loss_mlp": 0.01025263,
      "balance_loss_clip": 1.03428805,
      "balance_loss_mlp": 1.0134728,
      "epoch": 0.8689012475574929,
      "flos": 18221828843520.0,
      "grad_norm": 2.616801230320457,
      "language_loss": 0.72009099,
      "learning_rate": 1.7752134150977205e-07,
      "loss": 0.74124825,
      "num_input_tokens_seen": 311771830,
      "step": 14452,
      "time_per_iteration": 2.4371657371520996
    },
    {
      "auxiliary_loss_clip": 0.01080744,
      "auxiliary_loss_mlp": 0.00786409,
      "balance_loss_clip": 1.03462815,
      "balance_loss_mlp": 1.01444221,
      "epoch": 0.8689613708101608,
      "flos": 19646728439040.0,
      "grad_norm": 1.4916465186526864,
      "language_loss": 0.72074056,
      "learning_rate": 1.7736096565127201e-07,
      "loss": 0.73941207,
      "num_input_tokens_seen": 311790130,
      "step": 14453,
      "time_per_iteration": 2.512530565261841
    },
    {
      "auxiliary_loss_clip": 0.01091214,
      "auxiliary_loss_mlp": 0.0103269,
      "balance_loss_clip": 1.03448749,
      "balance_loss_mlp": 1.02064419,
      "epoch": 0.8690214940628288,
      "flos": 11728749070080.0,
      "grad_norm": 2.1907338717695257,
      "language_loss": 0.73390532,
      "learning_rate": 1.7720065890742664e-07,
      "loss": 0.75514442,
      "num_input_tokens_seen": 311808360,
      "step": 14454,
      "time_per_iteration": 2.483767032623291
    },
    {
      "auxiliary_loss_clip": 0.01103057,
      "auxiliary_loss_mlp": 0.01031957,
      "balance_loss_clip": 1.03573525,
      "balance_loss_mlp": 1.02036417,
      "epoch": 0.8690816173154968,
      "flos": 34936450076160.0,
      "grad_norm": 1.6546090859021734,
      "language_loss": 0.59430444,
      "learning_rate": 1.7704042128431552e-07,
      "loss": 0.61565453,
      "num_input_tokens_seen": 311831325,
      "step": 14455,
      "time_per_iteration": 3.937310218811035
    },
    {
      "auxiliary_loss_clip": 0.01084577,
      "auxiliary_loss_mlp": 0.01028731,
      "balance_loss_clip": 1.0349021,
      "balance_loss_mlp": 1.01670325,
      "epoch": 0.8691417405681647,
      "flos": 11614804151040.0,
      "grad_norm": 2.0000630580073895,
      "language_loss": 0.79940468,
      "learning_rate": 1.7688025278801378e-07,
      "loss": 0.82053781,
      "num_input_tokens_seen": 311848090,
      "step": 14456,
      "time_per_iteration": 2.484936237335205
    },
    {
      "auxiliary_loss_clip": 0.01049468,
      "auxiliary_loss_mlp": 0.01039051,
      "balance_loss_clip": 1.03470969,
      "balance_loss_mlp": 1.02407849,
      "epoch": 0.8692018638208328,
      "flos": 24608038677120.0,
      "grad_norm": 2.8988203474634804,
      "language_loss": 0.74809521,
      "learning_rate": 1.7672015342459568e-07,
      "loss": 0.76898044,
      "num_input_tokens_seen": 311867855,
      "step": 14457,
      "time_per_iteration": 2.6505770683288574
    },
    {
      "auxiliary_loss_clip": 0.0104385,
      "auxiliary_loss_mlp": 0.01026705,
      "balance_loss_clip": 1.03481281,
      "balance_loss_mlp": 1.01511228,
      "epoch": 0.8692619870735007,
      "flos": 25995124229760.0,
      "grad_norm": 1.5840299630611727,
      "language_loss": 0.78437126,
      "learning_rate": 1.765601232001328e-07,
      "loss": 0.80507684,
      "num_input_tokens_seen": 311888675,
      "step": 14458,
      "time_per_iteration": 2.6167569160461426
    },
    {
      "auxiliary_loss_clip": 0.01093402,
      "auxiliary_loss_mlp": 0.01034885,
      "balance_loss_clip": 1.03428638,
      "balance_loss_mlp": 1.02195096,
      "epoch": 0.8693221103261687,
      "flos": 18041808856320.0,
      "grad_norm": 1.5852940255091947,
      "language_loss": 0.70960331,
      "learning_rate": 1.7640016212069187e-07,
      "loss": 0.73088622,
      "num_input_tokens_seen": 311907310,
      "step": 14459,
      "time_per_iteration": 2.492278575897217
    },
    {
      "auxiliary_loss_clip": 0.01078583,
      "auxiliary_loss_mlp": 0.01028031,
      "balance_loss_clip": 1.0345993,
      "balance_loss_mlp": 1.01737928,
      "epoch": 0.8693822335788366,
      "flos": 27492347859840.0,
      "grad_norm": 1.4954399400340253,
      "language_loss": 0.73957169,
      "learning_rate": 1.762402701923398e-07,
      "loss": 0.76063782,
      "num_input_tokens_seen": 311929635,
      "step": 14460,
      "time_per_iteration": 2.582257032394409
    },
    {
      "auxiliary_loss_clip": 0.01085797,
      "auxiliary_loss_mlp": 0.01032534,
      "balance_loss_clip": 1.03457797,
      "balance_loss_mlp": 1.02036858,
      "epoch": 0.8694423568315046,
      "flos": 24097712198400.0,
      "grad_norm": 1.884233682778875,
      "language_loss": 0.64673662,
      "learning_rate": 1.7608044742113947e-07,
      "loss": 0.66791999,
      "num_input_tokens_seen": 311948800,
      "step": 14461,
      "time_per_iteration": 2.5485026836395264
    },
    {
      "auxiliary_loss_clip": 0.01088723,
      "auxiliary_loss_mlp": 0.01034873,
      "balance_loss_clip": 1.03294492,
      "balance_loss_mlp": 1.02224255,
      "epoch": 0.8695024800841725,
      "flos": 18362131367040.0,
      "grad_norm": 2.0185285472477954,
      "language_loss": 0.82842213,
      "learning_rate": 1.7592069381315123e-07,
      "loss": 0.84965807,
      "num_input_tokens_seen": 311964090,
      "step": 14462,
      "time_per_iteration": 2.4405837059020996
    },
    {
      "auxiliary_loss_clip": 0.01092408,
      "auxiliary_loss_mlp": 0.01031945,
      "balance_loss_clip": 1.03345609,
      "balance_loss_mlp": 1.01976204,
      "epoch": 0.8695626033368405,
      "flos": 14027750133120.0,
      "grad_norm": 1.8486719102148033,
      "language_loss": 0.65381539,
      "learning_rate": 1.757610093744335e-07,
      "loss": 0.6750589,
      "num_input_tokens_seen": 311981460,
      "step": 14463,
      "time_per_iteration": 2.5104520320892334
    },
    {
      "auxiliary_loss_clip": 0.01085306,
      "auxiliary_loss_mlp": 0.0103733,
      "balance_loss_clip": 1.03665257,
      "balance_loss_mlp": 1.02454531,
      "epoch": 0.8696227265895085,
      "flos": 16836862193280.0,
      "grad_norm": 2.0688561092061626,
      "language_loss": 0.66466784,
      "learning_rate": 1.7560139411104058e-07,
      "loss": 0.68589419,
      "num_input_tokens_seen": 312000115,
      "step": 14464,
      "time_per_iteration": 2.518507957458496
    },
    {
      "auxiliary_loss_clip": 0.01083401,
      "auxiliary_loss_mlp": 0.01032559,
      "balance_loss_clip": 1.03325176,
      "balance_loss_mlp": 1.02051866,
      "epoch": 0.8696828498421765,
      "flos": 21799070271360.0,
      "grad_norm": 1.9476867493837906,
      "language_loss": 0.6243614,
      "learning_rate": 1.7544184802902607e-07,
      "loss": 0.64552099,
      "num_input_tokens_seen": 312020770,
      "step": 14465,
      "time_per_iteration": 2.5843963623046875
    },
    {
      "auxiliary_loss_clip": 0.01087497,
      "auxiliary_loss_mlp": 0.01036074,
      "balance_loss_clip": 1.0340544,
      "balance_loss_mlp": 1.02529752,
      "epoch": 0.8697429730948444,
      "flos": 22894812610560.0,
      "grad_norm": 1.464881963209008,
      "language_loss": 0.84582412,
      "learning_rate": 1.7528237113443934e-07,
      "loss": 0.86705983,
      "num_input_tokens_seen": 312041870,
      "step": 14466,
      "time_per_iteration": 2.5624406337738037
    },
    {
      "auxiliary_loss_clip": 0.01075796,
      "auxiliary_loss_mlp": 0.01039592,
      "balance_loss_clip": 1.03538465,
      "balance_loss_mlp": 1.02548969,
      "epoch": 0.8698030963475124,
      "flos": 24717458482560.0,
      "grad_norm": 2.324717994135291,
      "language_loss": 0.61475599,
      "learning_rate": 1.7512296343332779e-07,
      "loss": 0.63590986,
      "num_input_tokens_seen": 312058210,
      "step": 14467,
      "time_per_iteration": 2.56168270111084
    },
    {
      "auxiliary_loss_clip": 0.01097371,
      "auxiliary_loss_mlp": 0.01027467,
      "balance_loss_clip": 1.03236079,
      "balance_loss_mlp": 1.01627898,
      "epoch": 0.8698632196001803,
      "flos": 28442221067520.0,
      "grad_norm": 1.3430913866157304,
      "language_loss": 0.68619466,
      "learning_rate": 1.7496362493173655e-07,
      "loss": 0.707443,
      "num_input_tokens_seen": 312082665,
      "step": 14468,
      "time_per_iteration": 2.533604621887207
    },
    {
      "auxiliary_loss_clip": 0.01078447,
      "auxiliary_loss_mlp": 0.01025813,
      "balance_loss_clip": 1.03276682,
      "balance_loss_mlp": 1.01504827,
      "epoch": 0.8699233428528483,
      "flos": 27636457224960.0,
      "grad_norm": 1.6113196312974467,
      "language_loss": 0.71139991,
      "learning_rate": 1.7480435563570773e-07,
      "loss": 0.73244256,
      "num_input_tokens_seen": 312101960,
      "step": 14469,
      "time_per_iteration": 2.562858819961548
    },
    {
      "auxiliary_loss_clip": 0.01087561,
      "auxiliary_loss_mlp": 0.01027018,
      "balance_loss_clip": 1.03471744,
      "balance_loss_mlp": 1.01600301,
      "epoch": 0.8699834661055164,
      "flos": 20045659864320.0,
      "grad_norm": 1.9449420600766056,
      "language_loss": 0.8418014,
      "learning_rate": 1.7464515555128024e-07,
      "loss": 0.86294723,
      "num_input_tokens_seen": 312117125,
      "step": 14470,
      "time_per_iteration": 2.4563589096069336
    },
    {
      "auxiliary_loss_clip": 0.01079028,
      "auxiliary_loss_mlp": 0.01032576,
      "balance_loss_clip": 1.03510642,
      "balance_loss_mlp": 1.01956391,
      "epoch": 0.8700435893581843,
      "flos": 23732787974400.0,
      "grad_norm": 2.265664215985485,
      "language_loss": 0.73113024,
      "learning_rate": 1.7448602468449148e-07,
      "loss": 0.75224626,
      "num_input_tokens_seen": 312135775,
      "step": 14471,
      "time_per_iteration": 2.549468994140625
    },
    {
      "auxiliary_loss_clip": 0.01102353,
      "auxiliary_loss_mlp": 0.0102652,
      "balance_loss_clip": 1.03502154,
      "balance_loss_mlp": 1.01549339,
      "epoch": 0.8701037126108523,
      "flos": 23548422441600.0,
      "grad_norm": 1.4437363697577246,
      "language_loss": 0.79050136,
      "learning_rate": 1.7432696304137573e-07,
      "loss": 0.81179011,
      "num_input_tokens_seen": 312156070,
      "step": 14472,
      "time_per_iteration": 2.4599857330322266
    },
    {
      "auxiliary_loss_clip": 0.01086606,
      "auxiliary_loss_mlp": 0.00783539,
      "balance_loss_clip": 1.03472519,
      "balance_loss_mlp": 1.0090065,
      "epoch": 0.8701638358635202,
      "flos": 18843442634880.0,
      "grad_norm": 1.7878599959121197,
      "language_loss": 0.72889066,
      "learning_rate": 1.741679706279644e-07,
      "loss": 0.74759209,
      "num_input_tokens_seen": 312174380,
      "step": 14473,
      "time_per_iteration": 2.4792652130126953
    },
    {
      "auxiliary_loss_clip": 0.01106412,
      "auxiliary_loss_mlp": 0.01027447,
      "balance_loss_clip": 1.03621221,
      "balance_loss_mlp": 1.01560974,
      "epoch": 0.8702239591161882,
      "flos": 27928339142400.0,
      "grad_norm": 1.4439053796727133,
      "language_loss": 0.72405684,
      "learning_rate": 1.7400904745028644e-07,
      "loss": 0.74539542,
      "num_input_tokens_seen": 312195130,
      "step": 14474,
      "time_per_iteration": 2.4827075004577637
    },
    {
      "auxiliary_loss_clip": 0.01080538,
      "auxiliary_loss_mlp": 0.01034079,
      "balance_loss_clip": 1.03345728,
      "balance_loss_mlp": 1.0215081,
      "epoch": 0.8702840823688561,
      "flos": 17233997938560.0,
      "grad_norm": 1.8559342912896644,
      "language_loss": 0.66813564,
      "learning_rate": 1.7385019351436925e-07,
      "loss": 0.68928182,
      "num_input_tokens_seen": 312212300,
      "step": 14475,
      "time_per_iteration": 2.497982978820801
    },
    {
      "auxiliary_loss_clip": 0.01101568,
      "auxiliary_loss_mlp": 0.0102344,
      "balance_loss_clip": 1.03238702,
      "balance_loss_mlp": 1.01142991,
      "epoch": 0.8703442056215241,
      "flos": 19427565605760.0,
      "grad_norm": 1.5829894581459116,
      "language_loss": 0.77785337,
      "learning_rate": 1.736914088262349e-07,
      "loss": 0.79910338,
      "num_input_tokens_seen": 312231735,
      "step": 14476,
      "time_per_iteration": 2.468050956726074
    },
    {
      "auxiliary_loss_clip": 0.01087118,
      "auxiliary_loss_mlp": 0.01035109,
      "balance_loss_clip": 1.03388596,
      "balance_loss_mlp": 1.0223949,
      "epoch": 0.8704043288741921,
      "flos": 22273845264000.0,
      "grad_norm": 2.1647822176639706,
      "language_loss": 0.72286153,
      "learning_rate": 1.7353269339190525e-07,
      "loss": 0.74408376,
      "num_input_tokens_seen": 312253060,
      "step": 14477,
      "time_per_iteration": 2.5451111793518066
    },
    {
      "auxiliary_loss_clip": 0.0109457,
      "auxiliary_loss_mlp": 0.01027466,
      "balance_loss_clip": 1.0352571,
      "balance_loss_mlp": 1.01553953,
      "epoch": 0.8704644521268601,
      "flos": 16648725732480.0,
      "grad_norm": 1.803896076857084,
      "language_loss": 0.5972842,
      "learning_rate": 1.7337404721739946e-07,
      "loss": 0.61850452,
      "num_input_tokens_seen": 312269460,
      "step": 14478,
      "time_per_iteration": 2.443573474884033
    },
    {
      "auxiliary_loss_clip": 0.01093613,
      "auxiliary_loss_mlp": 0.01030358,
      "balance_loss_clip": 1.03817952,
      "balance_loss_mlp": 1.0198679,
      "epoch": 0.870524575379528,
      "flos": 24280210224000.0,
      "grad_norm": 1.6267302262456167,
      "language_loss": 0.71425426,
      "learning_rate": 1.732154703087323e-07,
      "loss": 0.73549402,
      "num_input_tokens_seen": 312289830,
      "step": 14479,
      "time_per_iteration": 2.5244760513305664
    },
    {
      "auxiliary_loss_clip": 0.01080384,
      "auxiliary_loss_mlp": 0.01031566,
      "balance_loss_clip": 1.0346961,
      "balance_loss_mlp": 1.01940703,
      "epoch": 0.870584698632196,
      "flos": 28768684803840.0,
      "grad_norm": 1.4996150642173522,
      "language_loss": 0.71125418,
      "learning_rate": 1.7305696267191805e-07,
      "loss": 0.73237371,
      "num_input_tokens_seen": 312311320,
      "step": 14480,
      "time_per_iteration": 2.579730272293091
    },
    {
      "auxiliary_loss_clip": 0.01054677,
      "auxiliary_loss_mlp": 0.01033407,
      "balance_loss_clip": 1.03340805,
      "balance_loss_mlp": 1.02139664,
      "epoch": 0.8706448218848639,
      "flos": 32449635774720.0,
      "grad_norm": 1.8430625338238198,
      "language_loss": 0.70161456,
      "learning_rate": 1.728985243129666e-07,
      "loss": 0.72249544,
      "num_input_tokens_seen": 312332095,
      "step": 14481,
      "time_per_iteration": 2.671815872192383
    },
    {
      "auxiliary_loss_clip": 0.01090367,
      "auxiliary_loss_mlp": 0.01027345,
      "balance_loss_clip": 1.03411496,
      "balance_loss_mlp": 1.01594281,
      "epoch": 0.8707049451375319,
      "flos": 22748009725440.0,
      "grad_norm": 1.745740860809939,
      "language_loss": 0.76962852,
      "learning_rate": 1.7274015523788643e-07,
      "loss": 0.79080564,
      "num_input_tokens_seen": 312351225,
      "step": 14482,
      "time_per_iteration": 2.49041748046875
    },
    {
      "auxiliary_loss_clip": 0.01080802,
      "auxiliary_loss_mlp": 0.01030181,
      "balance_loss_clip": 1.03429139,
      "balance_loss_mlp": 1.01778948,
      "epoch": 0.8707650683902,
      "flos": 15851976203520.0,
      "grad_norm": 1.8584796014162464,
      "language_loss": 0.76497602,
      "learning_rate": 1.7258185545268234e-07,
      "loss": 0.78608578,
      "num_input_tokens_seen": 312369730,
      "step": 14483,
      "time_per_iteration": 3.9839916229248047
    },
    {
      "auxiliary_loss_clip": 0.01098055,
      "auxiliary_loss_mlp": 0.01034684,
      "balance_loss_clip": 1.03543746,
      "balance_loss_mlp": 1.02151752,
      "epoch": 0.8708251916428679,
      "flos": 16468131127680.0,
      "grad_norm": 2.0584274026196554,
      "language_loss": 0.623402,
      "learning_rate": 1.7242362496335749e-07,
      "loss": 0.64472944,
      "num_input_tokens_seen": 312386780,
      "step": 14484,
      "time_per_iteration": 3.851515054702759
    },
    {
      "auxiliary_loss_clip": 0.01103252,
      "auxiliary_loss_mlp": 0.0103235,
      "balance_loss_clip": 1.03564858,
      "balance_loss_mlp": 1.02085197,
      "epoch": 0.8708853148955359,
      "flos": 15377847655680.0,
      "grad_norm": 1.7480041648271312,
      "language_loss": 0.67929196,
      "learning_rate": 1.7226546377591222e-07,
      "loss": 0.70064795,
      "num_input_tokens_seen": 312404875,
      "step": 14485,
      "time_per_iteration": 2.446337938308716
    },
    {
      "auxiliary_loss_clip": 0.01046667,
      "auxiliary_loss_mlp": 0.00783896,
      "balance_loss_clip": 1.03148186,
      "balance_loss_mlp": 1.00909328,
      "epoch": 0.8709454381482038,
      "flos": 30551325903360.0,
      "grad_norm": 1.7893398432812508,
      "language_loss": 0.63026518,
      "learning_rate": 1.7210737189634373e-07,
      "loss": 0.6485709,
      "num_input_tokens_seen": 312425280,
      "step": 14486,
      "time_per_iteration": 2.682859182357788
    },
    {
      "auxiliary_loss_clip": 0.01105899,
      "auxiliary_loss_mlp": 0.01032828,
      "balance_loss_clip": 1.03433907,
      "balance_loss_mlp": 1.01970959,
      "epoch": 0.8710055614008718,
      "flos": 22601422321920.0,
      "grad_norm": 2.122709873989983,
      "language_loss": 0.62171674,
      "learning_rate": 1.7194934933064653e-07,
      "loss": 0.64310408,
      "num_input_tokens_seen": 312443835,
      "step": 14487,
      "time_per_iteration": 2.467771053314209
    },
    {
      "auxiliary_loss_clip": 0.01076231,
      "auxiliary_loss_mlp": 0.0078048,
      "balance_loss_clip": 1.03490591,
      "balance_loss_mlp": 1.00608921,
      "epoch": 0.8710656846535397,
      "flos": 18443146492800.0,
      "grad_norm": 2.3342853863908046,
      "language_loss": 0.6784597,
      "learning_rate": 1.7179139608481318e-07,
      "loss": 0.69702679,
      "num_input_tokens_seen": 312460830,
      "step": 14488,
      "time_per_iteration": 3.8791208267211914
    },
    {
      "auxiliary_loss_clip": 0.0108519,
      "auxiliary_loss_mlp": 0.00782809,
      "balance_loss_clip": 1.03512883,
      "balance_loss_mlp": 1.00827861,
      "epoch": 0.8711258079062077,
      "flos": 16503862181760.0,
      "grad_norm": 1.8263789023041945,
      "language_loss": 0.85678846,
      "learning_rate": 1.716335121648338e-07,
      "loss": 0.87546849,
      "num_input_tokens_seen": 312477575,
      "step": 14489,
      "time_per_iteration": 2.5048816204071045
    },
    {
      "auxiliary_loss_clip": 0.01098328,
      "auxiliary_loss_mlp": 0.01029371,
      "balance_loss_clip": 1.03648019,
      "balance_loss_mlp": 1.01661563,
      "epoch": 0.8711859311588757,
      "flos": 15663336952320.0,
      "grad_norm": 2.1089495041904844,
      "language_loss": 0.76048535,
      "learning_rate": 1.7147569757669445e-07,
      "loss": 0.78176236,
      "num_input_tokens_seen": 312492140,
      "step": 14490,
      "time_per_iteration": 2.4342377185821533
    },
    {
      "auxiliary_loss_clip": 0.01098459,
      "auxiliary_loss_mlp": 0.01029329,
      "balance_loss_clip": 1.0363549,
      "balance_loss_mlp": 1.01703274,
      "epoch": 0.8712460544115437,
      "flos": 15557544420480.0,
      "grad_norm": 2.1175640420006605,
      "language_loss": 0.76206827,
      "learning_rate": 1.7131795232638012e-07,
      "loss": 0.78334612,
      "num_input_tokens_seen": 312508400,
      "step": 14491,
      "time_per_iteration": 2.4756104946136475
    },
    {
      "auxiliary_loss_clip": 0.01074867,
      "auxiliary_loss_mlp": 0.01025694,
      "balance_loss_clip": 1.03833306,
      "balance_loss_mlp": 1.0139761,
      "epoch": 0.8713061776642116,
      "flos": 16763568491520.0,
      "grad_norm": 1.89405419696423,
      "language_loss": 0.66872376,
      "learning_rate": 1.711602764198723e-07,
      "loss": 0.68972939,
      "num_input_tokens_seen": 312525915,
      "step": 14492,
      "time_per_iteration": 2.543067455291748
    },
    {
      "auxiliary_loss_clip": 0.01089663,
      "auxiliary_loss_mlp": 0.01030317,
      "balance_loss_clip": 1.03426564,
      "balance_loss_mlp": 1.01945114,
      "epoch": 0.8713663009168796,
      "flos": 24279887001600.0,
      "grad_norm": 2.279462319158682,
      "language_loss": 0.69574946,
      "learning_rate": 1.7100266986314992e-07,
      "loss": 0.71694922,
      "num_input_tokens_seen": 312544735,
      "step": 14493,
      "time_per_iteration": 2.5088891983032227
    },
    {
      "auxiliary_loss_clip": 0.01107041,
      "auxiliary_loss_mlp": 0.01032471,
      "balance_loss_clip": 1.03784251,
      "balance_loss_mlp": 1.01971006,
      "epoch": 0.8714264241695475,
      "flos": 23795594904960.0,
      "grad_norm": 2.588389629360129,
      "language_loss": 0.89128768,
      "learning_rate": 1.7084513266218936e-07,
      "loss": 0.91268283,
      "num_input_tokens_seen": 312557910,
      "step": 14494,
      "time_per_iteration": 3.9085967540740967
    },
    {
      "auxiliary_loss_clip": 0.01071681,
      "auxiliary_loss_mlp": 0.01029645,
      "balance_loss_clip": 1.03894353,
      "balance_loss_mlp": 1.01834369,
      "epoch": 0.8714865474222155,
      "flos": 37997942071680.0,
      "grad_norm": 1.5976229463612197,
      "language_loss": 0.59336495,
      "learning_rate": 1.7068766482296514e-07,
      "loss": 0.61437821,
      "num_input_tokens_seen": 312580360,
      "step": 14495,
      "time_per_iteration": 2.6967506408691406
    },
    {
      "auxiliary_loss_clip": 0.01073221,
      "auxiliary_loss_mlp": 0.01035181,
      "balance_loss_clip": 1.03456044,
      "balance_loss_mlp": 1.02278876,
      "epoch": 0.8715466706748836,
      "flos": 22455696844800.0,
      "grad_norm": 1.8726672344198838,
      "language_loss": 0.80161214,
      "learning_rate": 1.7053026635144762e-07,
      "loss": 0.82269615,
      "num_input_tokens_seen": 312597550,
      "step": 14496,
      "time_per_iteration": 2.547055959701538
    },
    {
      "auxiliary_loss_clip": 0.01083483,
      "auxiliary_loss_mlp": 0.01033841,
      "balance_loss_clip": 1.03532505,
      "balance_loss_mlp": 1.02096581,
      "epoch": 0.8716067939275515,
      "flos": 21215126868480.0,
      "grad_norm": 1.9728769901333587,
      "language_loss": 0.78479195,
      "learning_rate": 1.7037293725360624e-07,
      "loss": 0.80596519,
      "num_input_tokens_seen": 312616435,
      "step": 14497,
      "time_per_iteration": 2.4969942569732666
    },
    {
      "auxiliary_loss_clip": 0.01106333,
      "auxiliary_loss_mlp": 0.01028753,
      "balance_loss_clip": 1.03578782,
      "balance_loss_mlp": 1.01603913,
      "epoch": 0.8716669171802195,
      "flos": 22997732054400.0,
      "grad_norm": 1.9418777230057958,
      "language_loss": 0.67094177,
      "learning_rate": 1.70215677535406e-07,
      "loss": 0.69229263,
      "num_input_tokens_seen": 312632770,
      "step": 14498,
      "time_per_iteration": 2.470106363296509
    },
    {
      "auxiliary_loss_clip": 0.01065321,
      "auxiliary_loss_mlp": 0.01031686,
      "balance_loss_clip": 1.03331375,
      "balance_loss_mlp": 1.02036095,
      "epoch": 0.8717270404328874,
      "flos": 29784058462080.0,
      "grad_norm": 1.6224232864914374,
      "language_loss": 0.56731594,
      "learning_rate": 1.700584872028108e-07,
      "loss": 0.58828598,
      "num_input_tokens_seen": 312651900,
      "step": 14499,
      "time_per_iteration": 2.6328461170196533
    },
    {
      "auxiliary_loss_clip": 0.01062815,
      "auxiliary_loss_mlp": 0.0103115,
      "balance_loss_clip": 1.03364325,
      "balance_loss_mlp": 1.0187819,
      "epoch": 0.8717871636855554,
      "flos": 22018125363840.0,
      "grad_norm": 1.896133786340536,
      "language_loss": 0.79983801,
      "learning_rate": 1.6990136626178097e-07,
      "loss": 0.82077765,
      "num_input_tokens_seen": 312671380,
      "step": 14500,
      "time_per_iteration": 2.5965576171875
    },
    {
      "auxiliary_loss_clip": 0.01087773,
      "auxiliary_loss_mlp": 0.01028602,
      "balance_loss_clip": 1.03624916,
      "balance_loss_mlp": 1.01674628,
      "epoch": 0.8718472869382233,
      "flos": 16654256426880.0,
      "grad_norm": 1.9416882176797718,
      "language_loss": 0.73199797,
      "learning_rate": 1.6974431471827466e-07,
      "loss": 0.75316173,
      "num_input_tokens_seen": 312689215,
      "step": 14501,
      "time_per_iteration": 2.468776226043701
    },
    {
      "auxiliary_loss_clip": 0.01069316,
      "auxiliary_loss_mlp": 0.01029343,
      "balance_loss_clip": 1.03317809,
      "balance_loss_mlp": 1.01636696,
      "epoch": 0.8719074101908914,
      "flos": 19495328613120.0,
      "grad_norm": 1.6464814885907133,
      "language_loss": 0.64399064,
      "learning_rate": 1.695873325782482e-07,
      "loss": 0.66497719,
      "num_input_tokens_seen": 312706400,
      "step": 14502,
      "time_per_iteration": 2.553511619567871
    },
    {
      "auxiliary_loss_clip": 0.01079648,
      "auxiliary_loss_mlp": 0.01036517,
      "balance_loss_clip": 1.03360605,
      "balance_loss_mlp": 1.02256942,
      "epoch": 0.8719675334435593,
      "flos": 33070890430080.0,
      "grad_norm": 1.6306886567538454,
      "language_loss": 0.68887353,
      "learning_rate": 1.6943041984765262e-07,
      "loss": 0.71003515,
      "num_input_tokens_seen": 312727985,
      "step": 14503,
      "time_per_iteration": 2.6167399883270264
    },
    {
      "auxiliary_loss_clip": 0.01079611,
      "auxiliary_loss_mlp": 0.01027499,
      "balance_loss_clip": 1.0359329,
      "balance_loss_mlp": 1.01571476,
      "epoch": 0.8720276566962273,
      "flos": 13626268842240.0,
      "grad_norm": 3.204719711539022,
      "language_loss": 0.69571936,
      "learning_rate": 1.6927357653243912e-07,
      "loss": 0.71679044,
      "num_input_tokens_seen": 312745025,
      "step": 14504,
      "time_per_iteration": 2.5312280654907227
    },
    {
      "auxiliary_loss_clip": 0.01091396,
      "auxiliary_loss_mlp": 0.00783609,
      "balance_loss_clip": 1.03666914,
      "balance_loss_mlp": 1.00974178,
      "epoch": 0.8720877799488952,
      "flos": 23514163845120.0,
      "grad_norm": 1.7338585429550317,
      "language_loss": 0.70429862,
      "learning_rate": 1.691168026385552e-07,
      "loss": 0.72304869,
      "num_input_tokens_seen": 312764170,
      "step": 14505,
      "time_per_iteration": 2.535022020339966
    },
    {
      "auxiliary_loss_clip": 0.01081895,
      "auxiliary_loss_mlp": 0.01025396,
      "balance_loss_clip": 1.03513813,
      "balance_loss_mlp": 1.01449454,
      "epoch": 0.8721479032015632,
      "flos": 20814148368000.0,
      "grad_norm": 1.5012725390387984,
      "language_loss": 0.78092557,
      "learning_rate": 1.6896009817194545e-07,
      "loss": 0.8019985,
      "num_input_tokens_seen": 312783830,
      "step": 14506,
      "time_per_iteration": 2.5358402729034424
    },
    {
      "auxiliary_loss_clip": 0.01083919,
      "auxiliary_loss_mlp": 0.01030521,
      "balance_loss_clip": 1.0345068,
      "balance_loss_mlp": 1.01861238,
      "epoch": 0.8722080264542311,
      "flos": 19463655795840.0,
      "grad_norm": 2.24164535751118,
      "language_loss": 0.73660278,
      "learning_rate": 1.6880346313855221e-07,
      "loss": 0.75774717,
      "num_input_tokens_seen": 312802015,
      "step": 14507,
      "time_per_iteration": 2.5163071155548096
    },
    {
      "auxiliary_loss_clip": 0.01047869,
      "auxiliary_loss_mlp": 0.01038987,
      "balance_loss_clip": 1.03292751,
      "balance_loss_mlp": 1.02429438,
      "epoch": 0.8722681497068991,
      "flos": 21761866759680.0,
      "grad_norm": 1.9880686947609014,
      "language_loss": 0.71956885,
      "learning_rate": 1.686468975443156e-07,
      "loss": 0.74043739,
      "num_input_tokens_seen": 312820650,
      "step": 14508,
      "time_per_iteration": 2.625107526779175
    },
    {
      "auxiliary_loss_clip": 0.01086963,
      "auxiliary_loss_mlp": 0.01031098,
      "balance_loss_clip": 1.03709435,
      "balance_loss_mlp": 1.01840186,
      "epoch": 0.8723282729595672,
      "flos": 28877134942080.0,
      "grad_norm": 1.539780432936753,
      "language_loss": 0.68794513,
      "learning_rate": 1.6849040139517202e-07,
      "loss": 0.7091257,
      "num_input_tokens_seen": 312841310,
      "step": 14509,
      "time_per_iteration": 2.570800542831421
    },
    {
      "auxiliary_loss_clip": 0.01078098,
      "auxiliary_loss_mlp": 0.01028473,
      "balance_loss_clip": 1.03546154,
      "balance_loss_mlp": 1.01696301,
      "epoch": 0.8723883962122351,
      "flos": 26469145036800.0,
      "grad_norm": 1.5793748124876146,
      "language_loss": 0.58393681,
      "learning_rate": 1.683339746970558e-07,
      "loss": 0.60500252,
      "num_input_tokens_seen": 312862100,
      "step": 14510,
      "time_per_iteration": 2.5701496601104736
    },
    {
      "auxiliary_loss_clip": 0.01109871,
      "auxiliary_loss_mlp": 0.01033338,
      "balance_loss_clip": 1.03588676,
      "balance_loss_mlp": 1.0203259,
      "epoch": 0.8724485194649031,
      "flos": 20521476351360.0,
      "grad_norm": 2.3061709623541793,
      "language_loss": 0.67629647,
      "learning_rate": 1.6817761745589865e-07,
      "loss": 0.69772863,
      "num_input_tokens_seen": 312880220,
      "step": 14511,
      "time_per_iteration": 2.4637279510498047
    },
    {
      "auxiliary_loss_clip": 0.01058602,
      "auxiliary_loss_mlp": 0.01032975,
      "balance_loss_clip": 1.03424048,
      "balance_loss_mlp": 1.01962948,
      "epoch": 0.872508642717571,
      "flos": 24353360271360.0,
      "grad_norm": 1.614337350136534,
      "language_loss": 0.81804478,
      "learning_rate": 1.6802132967763027e-07,
      "loss": 0.83896059,
      "num_input_tokens_seen": 312900765,
      "step": 14512,
      "time_per_iteration": 2.6793863773345947
    },
    {
      "auxiliary_loss_clip": 0.01011519,
      "auxiliary_loss_mlp": 0.01001223,
      "balance_loss_clip": 1.00792527,
      "balance_loss_mlp": 0.99994797,
      "epoch": 0.872568765970239,
      "flos": 61410012485760.0,
      "grad_norm": 0.7920912914676207,
      "language_loss": 0.58669311,
      "learning_rate": 1.6786511136817617e-07,
      "loss": 0.60682058,
      "num_input_tokens_seen": 312955840,
      "step": 14513,
      "time_per_iteration": 3.003101348876953
    },
    {
      "auxiliary_loss_clip": 0.01093814,
      "auxiliary_loss_mlp": 0.01030606,
      "balance_loss_clip": 1.03536367,
      "balance_loss_mlp": 1.01814914,
      "epoch": 0.8726288892229069,
      "flos": 22598046443520.0,
      "grad_norm": 1.7175454395014758,
      "language_loss": 0.76314211,
      "learning_rate": 1.6770896253346112e-07,
      "loss": 0.78438634,
      "num_input_tokens_seen": 312973565,
      "step": 14514,
      "time_per_iteration": 2.5098469257354736
    },
    {
      "auxiliary_loss_clip": 0.0109874,
      "auxiliary_loss_mlp": 0.01024035,
      "balance_loss_clip": 1.03673279,
      "balance_loss_mlp": 1.01293635,
      "epoch": 0.872689012475575,
      "flos": 25885201633920.0,
      "grad_norm": 2.06261463205412,
      "language_loss": 0.65305054,
      "learning_rate": 1.675528831794055e-07,
      "loss": 0.67427838,
      "num_input_tokens_seen": 312994660,
      "step": 14515,
      "time_per_iteration": 2.5189123153686523
    },
    {
      "auxiliary_loss_clip": 0.01093566,
      "auxiliary_loss_mlp": 0.01034379,
      "balance_loss_clip": 1.03455758,
      "balance_loss_mlp": 1.02112317,
      "epoch": 0.8727491357282429,
      "flos": 21506721477120.0,
      "grad_norm": 1.9455865745385206,
      "language_loss": 0.79243386,
      "learning_rate": 1.6739687331192842e-07,
      "loss": 0.81371337,
      "num_input_tokens_seen": 313009860,
      "step": 14516,
      "time_per_iteration": 2.499708890914917
    },
    {
      "auxiliary_loss_clip": 0.01106252,
      "auxiliary_loss_mlp": 0.01029992,
      "balance_loss_clip": 1.0356636,
      "balance_loss_mlp": 1.01776767,
      "epoch": 0.8728092589809109,
      "flos": 19207504932480.0,
      "grad_norm": 1.9860371991610217,
      "language_loss": 0.72492391,
      "learning_rate": 1.672409329369453e-07,
      "loss": 0.74628633,
      "num_input_tokens_seen": 313027025,
      "step": 14517,
      "time_per_iteration": 2.428126096725464
    },
    {
      "auxiliary_loss_clip": 0.01065566,
      "auxiliary_loss_mlp": 0.01025043,
      "balance_loss_clip": 1.03279614,
      "balance_loss_mlp": 1.01403403,
      "epoch": 0.8728693822335788,
      "flos": 20595308757120.0,
      "grad_norm": 2.011992841605645,
      "language_loss": 0.73029107,
      "learning_rate": 1.6708506206036966e-07,
      "loss": 0.75119716,
      "num_input_tokens_seen": 313046830,
      "step": 14518,
      "time_per_iteration": 2.5543506145477295
    },
    {
      "auxiliary_loss_clip": 0.01077562,
      "auxiliary_loss_mlp": 0.01038645,
      "balance_loss_clip": 1.03241658,
      "balance_loss_mlp": 1.02603889,
      "epoch": 0.8729295054862468,
      "flos": 21728613744000.0,
      "grad_norm": 1.3742065841182358,
      "language_loss": 0.74090624,
      "learning_rate": 1.6692926068811275e-07,
      "loss": 0.76206839,
      "num_input_tokens_seen": 313067715,
      "step": 14519,
      "time_per_iteration": 2.537863254547119
    },
    {
      "auxiliary_loss_clip": 0.01094647,
      "auxiliary_loss_mlp": 0.01029191,
      "balance_loss_clip": 1.03531075,
      "balance_loss_mlp": 1.01622725,
      "epoch": 0.8729896287389147,
      "flos": 17673436926720.0,
      "grad_norm": 2.6086829848184605,
      "language_loss": 0.76449239,
      "learning_rate": 1.6677352882608142e-07,
      "loss": 0.78573084,
      "num_input_tokens_seen": 313082305,
      "step": 14520,
      "time_per_iteration": 2.4626524448394775
    },
    {
      "auxiliary_loss_clip": 0.01084002,
      "auxiliary_loss_mlp": 0.01033399,
      "balance_loss_clip": 1.03582907,
      "balance_loss_mlp": 1.02034569,
      "epoch": 0.8730497519915827,
      "flos": 24571804832640.0,
      "grad_norm": 1.8422267914002988,
      "language_loss": 0.81892812,
      "learning_rate": 1.666178664801816e-07,
      "loss": 0.8401022,
      "num_input_tokens_seen": 313101190,
      "step": 14521,
      "time_per_iteration": 2.522062063217163
    },
    {
      "auxiliary_loss_clip": 0.01092574,
      "auxiliary_loss_mlp": 0.01032644,
      "balance_loss_clip": 1.03660512,
      "balance_loss_mlp": 1.01979899,
      "epoch": 0.8731098752442508,
      "flos": 13443734903040.0,
      "grad_norm": 2.2988661851321517,
      "language_loss": 0.7651605,
      "learning_rate": 1.6646227365631616e-07,
      "loss": 0.78641272,
      "num_input_tokens_seen": 313118965,
      "step": 14522,
      "time_per_iteration": 3.8551037311553955
    },
    {
      "auxiliary_loss_clip": 0.01090162,
      "auxiliary_loss_mlp": 0.00781302,
      "balance_loss_clip": 1.033041,
      "balance_loss_mlp": 1.00814891,
      "epoch": 0.8731699984969187,
      "flos": 23474446381440.0,
      "grad_norm": 1.6517816119740467,
      "language_loss": 0.75431484,
      "learning_rate": 1.66306750360385e-07,
      "loss": 0.77302945,
      "num_input_tokens_seen": 313139280,
      "step": 14523,
      "time_per_iteration": 3.888089418411255
    },
    {
      "auxiliary_loss_clip": 0.01091632,
      "auxiliary_loss_mlp": 0.01027447,
      "balance_loss_clip": 1.03374219,
      "balance_loss_mlp": 1.01591301,
      "epoch": 0.8732301217495867,
      "flos": 17712651600000.0,
      "grad_norm": 2.279223315746858,
      "language_loss": 0.79009879,
      "learning_rate": 1.6615129659828542e-07,
      "loss": 0.81128955,
      "num_input_tokens_seen": 313156655,
      "step": 14524,
      "time_per_iteration": 2.480048179626465
    },
    {
      "auxiliary_loss_clip": 0.01082378,
      "auxiliary_loss_mlp": 0.01028696,
      "balance_loss_clip": 1.03425133,
      "balance_loss_mlp": 1.0173595,
      "epoch": 0.8732902450022546,
      "flos": 22054359208320.0,
      "grad_norm": 2.0538145373900707,
      "language_loss": 0.77783436,
      "learning_rate": 1.6599591237591272e-07,
      "loss": 0.79894507,
      "num_input_tokens_seen": 313174050,
      "step": 14525,
      "time_per_iteration": 2.533050537109375
    },
    {
      "auxiliary_loss_clip": 0.01015446,
      "auxiliary_loss_mlp": 0.01030957,
      "balance_loss_clip": 1.03441167,
      "balance_loss_mlp": 1.01934552,
      "epoch": 0.8733503682549226,
      "flos": 22272983337600.0,
      "grad_norm": 1.6903131111111729,
      "language_loss": 0.69249833,
      "learning_rate": 1.6584059769915902e-07,
      "loss": 0.71296239,
      "num_input_tokens_seen": 313192765,
      "step": 14526,
      "time_per_iteration": 4.416518211364746
    },
    {
      "auxiliary_loss_clip": 0.0105566,
      "auxiliary_loss_mlp": 0.01039509,
      "balance_loss_clip": 1.03581607,
      "balance_loss_mlp": 1.02621686,
      "epoch": 0.8734104915075905,
      "flos": 23364344217600.0,
      "grad_norm": 1.8013766299143161,
      "language_loss": 0.60963643,
      "learning_rate": 1.6568535257391326e-07,
      "loss": 0.63058817,
      "num_input_tokens_seen": 313210925,
      "step": 14527,
      "time_per_iteration": 2.891195058822632
    },
    {
      "auxiliary_loss_clip": 0.01095572,
      "auxiliary_loss_mlp": 0.01031598,
      "balance_loss_clip": 1.03780401,
      "balance_loss_mlp": 1.01768672,
      "epoch": 0.8734706147602586,
      "flos": 17712292464000.0,
      "grad_norm": 2.089471573449631,
      "language_loss": 0.65547341,
      "learning_rate": 1.6553017700606265e-07,
      "loss": 0.67674518,
      "num_input_tokens_seen": 313228250,
      "step": 14528,
      "time_per_iteration": 2.483466625213623
    },
    {
      "auxiliary_loss_clip": 0.01074749,
      "auxiliary_loss_mlp": 0.01026412,
      "balance_loss_clip": 1.03688228,
      "balance_loss_mlp": 1.0149684,
      "epoch": 0.8735307380129265,
      "flos": 22049367217920.0,
      "grad_norm": 2.2016515872433575,
      "language_loss": 0.89648777,
      "learning_rate": 1.6537507100149205e-07,
      "loss": 0.9174993,
      "num_input_tokens_seen": 313247880,
      "step": 14529,
      "time_per_iteration": 2.566633701324463
    },
    {
      "auxiliary_loss_clip": 0.01080284,
      "auxiliary_loss_mlp": 0.01027964,
      "balance_loss_clip": 1.03341198,
      "balance_loss_mlp": 1.01622748,
      "epoch": 0.8735908612655945,
      "flos": 25338425829120.0,
      "grad_norm": 1.7776334760650847,
      "language_loss": 0.84897524,
      "learning_rate": 1.6522003456608258e-07,
      "loss": 0.87005776,
      "num_input_tokens_seen": 313266790,
      "step": 14530,
      "time_per_iteration": 3.067650556564331
    },
    {
      "auxiliary_loss_clip": 0.01080531,
      "auxiliary_loss_mlp": 0.01032828,
      "balance_loss_clip": 1.0332303,
      "balance_loss_mlp": 1.02183127,
      "epoch": 0.8736509845182624,
      "flos": 21540908246400.0,
      "grad_norm": 1.444286149496322,
      "language_loss": 0.74109018,
      "learning_rate": 1.650650677057128e-07,
      "loss": 0.76222372,
      "num_input_tokens_seen": 313286805,
      "step": 14531,
      "time_per_iteration": 2.5553054809570312
    },
    {
      "auxiliary_loss_clip": 0.01088911,
      "auxiliary_loss_mlp": 0.01027401,
      "balance_loss_clip": 1.03199172,
      "balance_loss_mlp": 1.01574874,
      "epoch": 0.8737111077709304,
      "flos": 22017227523840.0,
      "grad_norm": 2.1380601389213365,
      "language_loss": 0.61350036,
      "learning_rate": 1.6491017042625966e-07,
      "loss": 0.63466346,
      "num_input_tokens_seen": 313305415,
      "step": 14532,
      "time_per_iteration": 2.5060722827911377
    },
    {
      "auxiliary_loss_clip": 0.010197,
      "auxiliary_loss_mlp": 0.0100524,
      "balance_loss_clip": 1.00739288,
      "balance_loss_mlp": 1.00410163,
      "epoch": 0.8737712310235983,
      "flos": 70066315912320.0,
      "grad_norm": 0.8403163675784777,
      "language_loss": 0.58795118,
      "learning_rate": 1.6475534273359704e-07,
      "loss": 0.60820055,
      "num_input_tokens_seen": 313369940,
      "step": 14533,
      "time_per_iteration": 4.579978942871094
    },
    {
      "auxiliary_loss_clip": 0.01075978,
      "auxiliary_loss_mlp": 0.01032417,
      "balance_loss_clip": 1.03337288,
      "balance_loss_mlp": 1.02115142,
      "epoch": 0.8738313542762663,
      "flos": 28658331244800.0,
      "grad_norm": 1.533807282255495,
      "language_loss": 0.76642919,
      "learning_rate": 1.646005846335954e-07,
      "loss": 0.78751314,
      "num_input_tokens_seen": 313390965,
      "step": 14534,
      "time_per_iteration": 2.599895477294922
    },
    {
      "auxiliary_loss_clip": 0.01076143,
      "auxiliary_loss_mlp": 0.01031801,
      "balance_loss_clip": 1.03228128,
      "balance_loss_mlp": 1.01992798,
      "epoch": 0.8738914775289344,
      "flos": 22346384780160.0,
      "grad_norm": 1.872179470538046,
      "language_loss": 0.74987489,
      "learning_rate": 1.6444589613212357e-07,
      "loss": 0.77095437,
      "num_input_tokens_seen": 313409680,
      "step": 14535,
      "time_per_iteration": 2.5299248695373535
    },
    {
      "auxiliary_loss_clip": 0.01103086,
      "auxiliary_loss_mlp": 0.01031995,
      "balance_loss_clip": 1.03373444,
      "balance_loss_mlp": 1.01969874,
      "epoch": 0.8739516007816023,
      "flos": 31759648444800.0,
      "grad_norm": 1.8672328822855577,
      "language_loss": 0.7433362,
      "learning_rate": 1.64291277235048e-07,
      "loss": 0.76468694,
      "num_input_tokens_seen": 313431335,
      "step": 14536,
      "time_per_iteration": 2.5574238300323486
    },
    {
      "auxiliary_loss_clip": 0.01083049,
      "auxiliary_loss_mlp": 0.01032319,
      "balance_loss_clip": 1.03414083,
      "balance_loss_mlp": 1.02110124,
      "epoch": 0.8740117240342703,
      "flos": 21211715076480.0,
      "grad_norm": 1.858050998907757,
      "language_loss": 0.64153647,
      "learning_rate": 1.641367279482304e-07,
      "loss": 0.66269022,
      "num_input_tokens_seen": 313449225,
      "step": 14537,
      "time_per_iteration": 2.508390188217163
    },
    {
      "auxiliary_loss_clip": 0.01090373,
      "auxiliary_loss_mlp": 0.01026326,
      "balance_loss_clip": 1.03301692,
      "balance_loss_mlp": 1.01362383,
      "epoch": 0.8740718472869382,
      "flos": 25186666867200.0,
      "grad_norm": 1.828266733433781,
      "language_loss": 0.5797267,
      "learning_rate": 1.6398224827753216e-07,
      "loss": 0.60089368,
      "num_input_tokens_seen": 313467715,
      "step": 14538,
      "time_per_iteration": 2.5161564350128174
    },
    {
      "auxiliary_loss_clip": 0.01092891,
      "auxiliary_loss_mlp": 0.01026063,
      "balance_loss_clip": 1.03654075,
      "balance_loss_mlp": 1.01404083,
      "epoch": 0.8741319705396062,
      "flos": 19500931134720.0,
      "grad_norm": 2.1153223440588365,
      "language_loss": 0.68644047,
      "learning_rate": 1.6382783822881142e-07,
      "loss": 0.70762992,
      "num_input_tokens_seen": 313486805,
      "step": 14539,
      "time_per_iteration": 2.490931510925293
    },
    {
      "auxiliary_loss_clip": 0.01095147,
      "auxiliary_loss_mlp": 0.01028854,
      "balance_loss_clip": 1.03273487,
      "balance_loss_mlp": 1.01619995,
      "epoch": 0.8741920937922741,
      "flos": 14100900180480.0,
      "grad_norm": 1.9953083080500276,
      "language_loss": 0.74322772,
      "learning_rate": 1.6367349780792262e-07,
      "loss": 0.76446772,
      "num_input_tokens_seen": 313504880,
      "step": 14540,
      "time_per_iteration": 2.469717502593994
    },
    {
      "auxiliary_loss_clip": 0.01078347,
      "auxiliary_loss_mlp": 0.01039926,
      "balance_loss_clip": 1.0332731,
      "balance_loss_mlp": 1.02650928,
      "epoch": 0.8742522170449422,
      "flos": 27709858667520.0,
      "grad_norm": 2.427944396212384,
      "language_loss": 0.79068899,
      "learning_rate": 1.635192270207193e-07,
      "loss": 0.81187171,
      "num_input_tokens_seen": 313524995,
      "step": 14541,
      "time_per_iteration": 2.558685064315796
    },
    {
      "auxiliary_loss_clip": 0.01063901,
      "auxiliary_loss_mlp": 0.01031854,
      "balance_loss_clip": 1.03307247,
      "balance_loss_mlp": 1.01760828,
      "epoch": 0.8743123402976101,
      "flos": 21142587352320.0,
      "grad_norm": 1.9109300236808564,
      "language_loss": 0.66748273,
      "learning_rate": 1.6336502587305035e-07,
      "loss": 0.68844032,
      "num_input_tokens_seen": 313541740,
      "step": 14542,
      "time_per_iteration": 2.603910446166992
    },
    {
      "auxiliary_loss_clip": 0.01027918,
      "auxiliary_loss_mlp": 0.01004337,
      "balance_loss_clip": 1.00522876,
      "balance_loss_mlp": 1.00324631,
      "epoch": 0.8743724635502781,
      "flos": 60870024351360.0,
      "grad_norm": 0.7838761554241469,
      "language_loss": 0.54473031,
      "learning_rate": 1.632108943707642e-07,
      "loss": 0.56505287,
      "num_input_tokens_seen": 313593445,
      "step": 14543,
      "time_per_iteration": 2.89294171333313
    },
    {
      "auxiliary_loss_clip": 0.01088586,
      "auxiliary_loss_mlp": 0.01033665,
      "balance_loss_clip": 1.03711152,
      "balance_loss_mlp": 1.02116013,
      "epoch": 0.874432586802946,
      "flos": 28109292883200.0,
      "grad_norm": 1.9415013704489137,
      "language_loss": 0.69635707,
      "learning_rate": 1.6305683251970458e-07,
      "loss": 0.71757954,
      "num_input_tokens_seen": 313615640,
      "step": 14544,
      "time_per_iteration": 2.5610697269439697
    },
    {
      "auxiliary_loss_clip": 0.01056504,
      "auxiliary_loss_mlp": 0.01025469,
      "balance_loss_clip": 1.03308678,
      "balance_loss_mlp": 1.01497865,
      "epoch": 0.874492710055614,
      "flos": 23550289948800.0,
      "grad_norm": 1.5613934471429989,
      "language_loss": 0.75653601,
      "learning_rate": 1.62902840325714e-07,
      "loss": 0.77735567,
      "num_input_tokens_seen": 313635550,
      "step": 14545,
      "time_per_iteration": 2.6149206161499023
    },
    {
      "auxiliary_loss_clip": 0.01092702,
      "auxiliary_loss_mlp": 0.00784714,
      "balance_loss_clip": 1.03359628,
      "balance_loss_mlp": 1.00848711,
      "epoch": 0.8745528333082819,
      "flos": 40915647924480.0,
      "grad_norm": 1.5938681983991707,
      "language_loss": 0.66152233,
      "learning_rate": 1.6274891779463217e-07,
      "loss": 0.68029648,
      "num_input_tokens_seen": 313659275,
      "step": 14546,
      "time_per_iteration": 2.697160482406616
    },
    {
      "auxiliary_loss_clip": 0.01104465,
      "auxiliary_loss_mlp": 0.01031606,
      "balance_loss_clip": 1.03579926,
      "balance_loss_mlp": 1.01975632,
      "epoch": 0.87461295656095,
      "flos": 23622901292160.0,
      "grad_norm": 1.5750694499031774,
      "language_loss": 0.7285018,
      "learning_rate": 1.6259506493229536e-07,
      "loss": 0.74986255,
      "num_input_tokens_seen": 313680595,
      "step": 14547,
      "time_per_iteration": 2.4573020935058594
    },
    {
      "auxiliary_loss_clip": 0.01110175,
      "auxiliary_loss_mlp": 0.01037876,
      "balance_loss_clip": 1.03577089,
      "balance_loss_mlp": 1.02476263,
      "epoch": 0.874673079813618,
      "flos": 38794116983040.0,
      "grad_norm": 2.179392472517967,
      "language_loss": 0.69048089,
      "learning_rate": 1.6244128174453752e-07,
      "loss": 0.71196139,
      "num_input_tokens_seen": 313699730,
      "step": 14548,
      "time_per_iteration": 2.6329948902130127
    },
    {
      "auxiliary_loss_clip": 0.01086029,
      "auxiliary_loss_mlp": 0.01030769,
      "balance_loss_clip": 1.03565574,
      "balance_loss_mlp": 1.01859772,
      "epoch": 0.8747332030662859,
      "flos": 23696159080320.0,
      "grad_norm": 2.04689680578817,
      "language_loss": 0.70827186,
      "learning_rate": 1.6228756823719093e-07,
      "loss": 0.72943985,
      "num_input_tokens_seen": 313720090,
      "step": 14549,
      "time_per_iteration": 2.538029909133911
    },
    {
      "auxiliary_loss_clip": 0.01095451,
      "auxiliary_loss_mlp": 0.00785861,
      "balance_loss_clip": 1.03340721,
      "balance_loss_mlp": 1.01052642,
      "epoch": 0.8747933263189539,
      "flos": 24462456854400.0,
      "grad_norm": 2.365046695638516,
      "language_loss": 0.8380965,
      "learning_rate": 1.6213392441608352e-07,
      "loss": 0.85690963,
      "num_input_tokens_seen": 313736795,
      "step": 14550,
      "time_per_iteration": 2.514082431793213
    },
    {
      "auxiliary_loss_clip": 0.01097787,
      "auxiliary_loss_mlp": 0.01038482,
      "balance_loss_clip": 1.03674471,
      "balance_loss_mlp": 1.02677536,
      "epoch": 0.8748534495716218,
      "flos": 13809161917440.0,
      "grad_norm": 1.6144354798968827,
      "language_loss": 0.71824229,
      "learning_rate": 1.6198035028704183e-07,
      "loss": 0.73960495,
      "num_input_tokens_seen": 313754820,
      "step": 14551,
      "time_per_iteration": 2.4599106311798096
    },
    {
      "auxiliary_loss_clip": 0.01093604,
      "auxiliary_loss_mlp": 0.00786825,
      "balance_loss_clip": 1.03553975,
      "balance_loss_mlp": 1.01300359,
      "epoch": 0.8749135728242898,
      "flos": 29862092759040.0,
      "grad_norm": 2.3933249683528546,
      "language_loss": 0.64044458,
      "learning_rate": 1.6182684585588934e-07,
      "loss": 0.65924883,
      "num_input_tokens_seen": 313775830,
      "step": 14552,
      "time_per_iteration": 2.583329439163208
    },
    {
      "auxiliary_loss_clip": 0.01074925,
      "auxiliary_loss_mlp": 0.01027371,
      "balance_loss_clip": 1.03466797,
      "balance_loss_mlp": 1.01347113,
      "epoch": 0.8749736960769577,
      "flos": 24133479166080.0,
      "grad_norm": 1.667179063127772,
      "language_loss": 0.79350197,
      "learning_rate": 1.616734111284479e-07,
      "loss": 0.81452489,
      "num_input_tokens_seen": 313795745,
      "step": 14553,
      "time_per_iteration": 2.5570313930511475
    },
    {
      "auxiliary_loss_clip": 0.01087925,
      "auxiliary_loss_mlp": 0.01030085,
      "balance_loss_clip": 1.0314734,
      "balance_loss_mlp": 1.01825333,
      "epoch": 0.8750338193296258,
      "flos": 17202540602880.0,
      "grad_norm": 1.8165751547381694,
      "language_loss": 0.70272148,
      "learning_rate": 1.6152004611053416e-07,
      "loss": 0.72390157,
      "num_input_tokens_seen": 313813895,
      "step": 14554,
      "time_per_iteration": 2.478280782699585
    },
    {
      "auxiliary_loss_clip": 0.01083191,
      "auxiliary_loss_mlp": 0.00782024,
      "balance_loss_clip": 1.03508317,
      "balance_loss_mlp": 1.00645852,
      "epoch": 0.8750939425822937,
      "flos": 23733218937600.0,
      "grad_norm": 1.425732991644998,
      "language_loss": 0.83705807,
      "learning_rate": 1.6136675080796457e-07,
      "loss": 0.85571021,
      "num_input_tokens_seen": 313834225,
      "step": 14555,
      "time_per_iteration": 2.5370240211486816
    },
    {
      "auxiliary_loss_clip": 0.01092468,
      "auxiliary_loss_mlp": 0.01032242,
      "balance_loss_clip": 1.03440738,
      "balance_loss_mlp": 1.01966584,
      "epoch": 0.8751540658349617,
      "flos": 26541684552960.0,
      "grad_norm": 1.4908302467091437,
      "language_loss": 0.70787084,
      "learning_rate": 1.6121352522655252e-07,
      "loss": 0.72911793,
      "num_input_tokens_seen": 313854430,
      "step": 14556,
      "time_per_iteration": 2.531404972076416
    },
    {
      "auxiliary_loss_clip": 0.01086212,
      "auxiliary_loss_mlp": 0.01035577,
      "balance_loss_clip": 1.03260314,
      "balance_loss_mlp": 1.02210617,
      "epoch": 0.8752141890876296,
      "flos": 19386806647680.0,
      "grad_norm": 1.8407825862599383,
      "language_loss": 0.76676023,
      "learning_rate": 1.6106036937210732e-07,
      "loss": 0.78797817,
      "num_input_tokens_seen": 313871600,
      "step": 14557,
      "time_per_iteration": 2.4906272888183594
    },
    {
      "auxiliary_loss_clip": 0.01067937,
      "auxiliary_loss_mlp": 0.01039891,
      "balance_loss_clip": 1.03483462,
      "balance_loss_mlp": 1.02662885,
      "epoch": 0.8752743123402976,
      "flos": 25374408278400.0,
      "grad_norm": 3.364944176845108,
      "language_loss": 0.83096337,
      "learning_rate": 1.6090728325043767e-07,
      "loss": 0.8520416,
      "num_input_tokens_seen": 313891570,
      "step": 14558,
      "time_per_iteration": 2.596416711807251
    },
    {
      "auxiliary_loss_clip": 0.01028506,
      "auxiliary_loss_mlp": 0.01003476,
      "balance_loss_clip": 1.00585961,
      "balance_loss_mlp": 1.00242734,
      "epoch": 0.8753344355929655,
      "flos": 59952398578560.0,
      "grad_norm": 0.8042667483634577,
      "language_loss": 0.5615865,
      "learning_rate": 1.6075426686734784e-07,
      "loss": 0.58190632,
      "num_input_tokens_seen": 313951290,
      "step": 14559,
      "time_per_iteration": 3.0683300495147705
    },
    {
      "auxiliary_loss_clip": 0.01091594,
      "auxiliary_loss_mlp": 0.01034419,
      "balance_loss_clip": 1.03444648,
      "balance_loss_mlp": 1.02283812,
      "epoch": 0.8753945588456336,
      "flos": 17894646835200.0,
      "grad_norm": 2.244946716122349,
      "language_loss": 0.65819365,
      "learning_rate": 1.606013202286407e-07,
      "loss": 0.67945379,
      "num_input_tokens_seen": 313968645,
      "step": 14560,
      "time_per_iteration": 4.096326589584351
    },
    {
      "auxiliary_loss_clip": 0.01102807,
      "auxiliary_loss_mlp": 0.01029966,
      "balance_loss_clip": 1.03450477,
      "balance_loss_mlp": 1.01811683,
      "epoch": 0.8754546820983016,
      "flos": 30914885410560.0,
      "grad_norm": 1.9416890284537542,
      "language_loss": 0.78734833,
      "learning_rate": 1.6044844334011541e-07,
      "loss": 0.80867606,
      "num_input_tokens_seen": 313987580,
      "step": 14561,
      "time_per_iteration": 2.5741875171661377
    },
    {
      "auxiliary_loss_clip": 0.0110485,
      "auxiliary_loss_mlp": 0.01033718,
      "balance_loss_clip": 1.03378296,
      "balance_loss_mlp": 1.0207119,
      "epoch": 0.8755148053509695,
      "flos": 20631075724800.0,
      "grad_norm": 1.9513123189149806,
      "language_loss": 0.7754668,
      "learning_rate": 1.6029563620756982e-07,
      "loss": 0.79685247,
      "num_input_tokens_seen": 314004460,
      "step": 14562,
      "time_per_iteration": 4.323149681091309
    },
    {
      "auxiliary_loss_clip": 0.01098753,
      "auxiliary_loss_mlp": 0.01026768,
      "balance_loss_clip": 1.03347135,
      "balance_loss_mlp": 1.01552057,
      "epoch": 0.8755749286036375,
      "flos": 34969739005440.0,
      "grad_norm": 1.6983536249512006,
      "language_loss": 0.7164464,
      "learning_rate": 1.601428988367981e-07,
      "loss": 0.73770165,
      "num_input_tokens_seen": 314026855,
      "step": 14563,
      "time_per_iteration": 2.5797016620635986
    },
    {
      "auxiliary_loss_clip": 0.01109095,
      "auxiliary_loss_mlp": 0.01033758,
      "balance_loss_clip": 1.03735733,
      "balance_loss_mlp": 1.02156901,
      "epoch": 0.8756350518563054,
      "flos": 18186456925440.0,
      "grad_norm": 2.096513956613813,
      "language_loss": 0.66062307,
      "learning_rate": 1.5999023123359235e-07,
      "loss": 0.68205154,
      "num_input_tokens_seen": 314042830,
      "step": 14564,
      "time_per_iteration": 2.438516855239868
    },
    {
      "auxiliary_loss_clip": 0.0108971,
      "auxiliary_loss_mlp": 0.01035016,
      "balance_loss_clip": 1.0326314,
      "balance_loss_mlp": 1.02361977,
      "epoch": 0.8756951751089734,
      "flos": 20084012611200.0,
      "grad_norm": 1.5781452656232458,
      "language_loss": 0.70419621,
      "learning_rate": 1.598376334037408e-07,
      "loss": 0.72544348,
      "num_input_tokens_seen": 314062225,
      "step": 14565,
      "time_per_iteration": 3.921264410018921
    },
    {
      "auxiliary_loss_clip": 0.01087632,
      "auxiliary_loss_mlp": 0.01032123,
      "balance_loss_clip": 1.03539109,
      "balance_loss_mlp": 1.01882529,
      "epoch": 0.8757552983616413,
      "flos": 27525241739520.0,
      "grad_norm": 1.592546932430247,
      "language_loss": 0.77746964,
      "learning_rate": 1.5968510535303102e-07,
      "loss": 0.79866719,
      "num_input_tokens_seen": 314082325,
      "step": 14566,
      "time_per_iteration": 2.564500093460083
    },
    {
      "auxiliary_loss_clip": 0.01083693,
      "auxiliary_loss_mlp": 0.01033525,
      "balance_loss_clip": 1.03803194,
      "balance_loss_mlp": 1.02161002,
      "epoch": 0.8758154216143094,
      "flos": 18073014796800.0,
      "grad_norm": 1.9800775864986948,
      "language_loss": 0.71051723,
      "learning_rate": 1.5953264708724624e-07,
      "loss": 0.73168933,
      "num_input_tokens_seen": 314100310,
      "step": 14567,
      "time_per_iteration": 2.4667885303497314
    },
    {
      "auxiliary_loss_clip": 0.01082827,
      "auxiliary_loss_mlp": 0.00783174,
      "balance_loss_clip": 1.03322315,
      "balance_loss_mlp": 1.0083493,
      "epoch": 0.8758755448669773,
      "flos": 25045681985280.0,
      "grad_norm": 1.7891105693143545,
      "language_loss": 0.74608535,
      "learning_rate": 1.5938025861216776e-07,
      "loss": 0.76474535,
      "num_input_tokens_seen": 314121330,
      "step": 14568,
      "time_per_iteration": 2.5596694946289062
    },
    {
      "auxiliary_loss_clip": 0.01066874,
      "auxiliary_loss_mlp": 0.01028795,
      "balance_loss_clip": 1.03398633,
      "balance_loss_mlp": 1.01700544,
      "epoch": 0.8759356681196453,
      "flos": 22856818999680.0,
      "grad_norm": 2.06179169607315,
      "language_loss": 0.86734456,
      "learning_rate": 1.5922793993357475e-07,
      "loss": 0.88830125,
      "num_input_tokens_seen": 314139875,
      "step": 14569,
      "time_per_iteration": 2.5417699813842773
    },
    {
      "auxiliary_loss_clip": 0.01066735,
      "auxiliary_loss_mlp": 0.01030843,
      "balance_loss_clip": 1.03313971,
      "balance_loss_mlp": 1.0193336,
      "epoch": 0.8759957913723132,
      "flos": 21032521102080.0,
      "grad_norm": 1.5969689646711007,
      "language_loss": 0.73974818,
      "learning_rate": 1.5907569105724284e-07,
      "loss": 0.76072395,
      "num_input_tokens_seen": 314157850,
      "step": 14570,
      "time_per_iteration": 2.5479788780212402
    },
    {
      "auxiliary_loss_clip": 0.01095606,
      "auxiliary_loss_mlp": 0.00784426,
      "balance_loss_clip": 1.035043,
      "balance_loss_mlp": 1.01000667,
      "epoch": 0.8760559146249812,
      "flos": 20010467514240.0,
      "grad_norm": 1.6852835734327531,
      "language_loss": 0.67500937,
      "learning_rate": 1.5892351198894472e-07,
      "loss": 0.69380975,
      "num_input_tokens_seen": 314176720,
      "step": 14571,
      "time_per_iteration": 3.887359380722046
    },
    {
      "auxiliary_loss_clip": 0.01065655,
      "auxiliary_loss_mlp": 0.01027045,
      "balance_loss_clip": 1.03484082,
      "balance_loss_mlp": 1.01566696,
      "epoch": 0.8761160378776491,
      "flos": 19974161842560.0,
      "grad_norm": 1.9465823616354196,
      "language_loss": 0.62421691,
      "learning_rate": 1.5877140273445156e-07,
      "loss": 0.64514387,
      "num_input_tokens_seen": 314196645,
      "step": 14572,
      "time_per_iteration": 2.5310797691345215
    },
    {
      "auxiliary_loss_clip": 0.01090053,
      "auxiliary_loss_mlp": 0.01030011,
      "balance_loss_clip": 1.03448427,
      "balance_loss_mlp": 1.01950848,
      "epoch": 0.8761761611303172,
      "flos": 28804415857920.0,
      "grad_norm": 1.6354157221063528,
      "language_loss": 0.73516154,
      "learning_rate": 1.5861936329953162e-07,
      "loss": 0.75636214,
      "num_input_tokens_seen": 314217430,
      "step": 14573,
      "time_per_iteration": 2.5523860454559326
    },
    {
      "auxiliary_loss_clip": 0.0105487,
      "auxiliary_loss_mlp": 0.00780777,
      "balance_loss_clip": 1.03399897,
      "balance_loss_mlp": 1.00644588,
      "epoch": 0.8762362843829851,
      "flos": 18332505624960.0,
      "grad_norm": 1.9072896244341297,
      "language_loss": 0.729895,
      "learning_rate": 1.5846739368994966e-07,
      "loss": 0.7482515,
      "num_input_tokens_seen": 314235310,
      "step": 14574,
      "time_per_iteration": 2.564819574356079
    },
    {
      "auxiliary_loss_clip": 0.01091409,
      "auxiliary_loss_mlp": 0.01034666,
      "balance_loss_clip": 1.03436053,
      "balance_loss_mlp": 1.02261376,
      "epoch": 0.8762964076356531,
      "flos": 15779149378560.0,
      "grad_norm": 1.6874944080069947,
      "language_loss": 0.75699472,
      "learning_rate": 1.5831549391146903e-07,
      "loss": 0.77825546,
      "num_input_tokens_seen": 314252355,
      "step": 14575,
      "time_per_iteration": 2.487624406814575
    },
    {
      "auxiliary_loss_clip": 0.01077685,
      "auxiliary_loss_mlp": 0.01040615,
      "balance_loss_clip": 1.03544807,
      "balance_loss_mlp": 1.02886117,
      "epoch": 0.8763565308883211,
      "flos": 33176754789120.0,
      "grad_norm": 1.8187111074492424,
      "language_loss": 0.66896737,
      "learning_rate": 1.5816366396984916e-07,
      "loss": 0.69015038,
      "num_input_tokens_seen": 314272755,
      "step": 14576,
      "time_per_iteration": 2.630730152130127
    },
    {
      "auxiliary_loss_clip": 0.01073483,
      "auxiliary_loss_mlp": 0.01029415,
      "balance_loss_clip": 1.03060853,
      "balance_loss_mlp": 1.01809025,
      "epoch": 0.876416654140989,
      "flos": 15888102307200.0,
      "grad_norm": 1.715414850476694,
      "language_loss": 0.66708738,
      "learning_rate": 1.5801190387084806e-07,
      "loss": 0.68811631,
      "num_input_tokens_seen": 314291365,
      "step": 14577,
      "time_per_iteration": 2.5200281143188477
    },
    {
      "auxiliary_loss_clip": 0.01095008,
      "auxiliary_loss_mlp": 0.01033422,
      "balance_loss_clip": 1.03668463,
      "balance_loss_mlp": 1.02102399,
      "epoch": 0.876476777393657,
      "flos": 25885237547520.0,
      "grad_norm": 1.9745455725760268,
      "language_loss": 0.71480274,
      "learning_rate": 1.5786021362021962e-07,
      "loss": 0.73608708,
      "num_input_tokens_seen": 314310075,
      "step": 14578,
      "time_per_iteration": 2.505345344543457
    },
    {
      "auxiliary_loss_clip": 0.01105471,
      "auxiliary_loss_mlp": 0.01031462,
      "balance_loss_clip": 1.03463054,
      "balance_loss_mlp": 1.01912951,
      "epoch": 0.876536900646325,
      "flos": 13589675861760.0,
      "grad_norm": 2.2286462006731833,
      "language_loss": 0.71503162,
      "learning_rate": 1.5770859322371676e-07,
      "loss": 0.73640096,
      "num_input_tokens_seen": 314325695,
      "step": 14579,
      "time_per_iteration": 2.4751572608947754
    },
    {
      "auxiliary_loss_clip": 0.01073374,
      "auxiliary_loss_mlp": 0.01034647,
      "balance_loss_clip": 1.03231525,
      "balance_loss_mlp": 1.02158761,
      "epoch": 0.876597023898993,
      "flos": 12203344494720.0,
      "grad_norm": 1.9214771754445348,
      "language_loss": 0.70574963,
      "learning_rate": 1.5755704268708912e-07,
      "loss": 0.72682989,
      "num_input_tokens_seen": 314343605,
      "step": 14580,
      "time_per_iteration": 2.4873805046081543
    },
    {
      "auxiliary_loss_clip": 0.01101842,
      "auxiliary_loss_mlp": 0.00782608,
      "balance_loss_clip": 1.03538156,
      "balance_loss_mlp": 1.00991607,
      "epoch": 0.8766571471516609,
      "flos": 25336773803520.0,
      "grad_norm": 4.222731067858873,
      "language_loss": 0.65358198,
      "learning_rate": 1.5740556201608256e-07,
      "loss": 0.67242646,
      "num_input_tokens_seen": 314364275,
      "step": 14581,
      "time_per_iteration": 2.5160391330718994
    },
    {
      "auxiliary_loss_clip": 0.01076354,
      "auxiliary_loss_mlp": 0.01029096,
      "balance_loss_clip": 1.03531897,
      "balance_loss_mlp": 1.01787829,
      "epoch": 0.8767172704043289,
      "flos": 30113287545600.0,
      "grad_norm": 1.5175793336771028,
      "language_loss": 0.73717582,
      "learning_rate": 1.572541512164416e-07,
      "loss": 0.75823033,
      "num_input_tokens_seen": 314385140,
      "step": 14582,
      "time_per_iteration": 2.581895112991333
    },
    {
      "auxiliary_loss_clip": 0.01101807,
      "auxiliary_loss_mlp": 0.00781887,
      "balance_loss_clip": 1.03340447,
      "balance_loss_mlp": 1.00821602,
      "epoch": 0.8767773936569968,
      "flos": 19281157770240.0,
      "grad_norm": 2.026901181189871,
      "language_loss": 0.66488069,
      "learning_rate": 1.5710281029390826e-07,
      "loss": 0.68371761,
      "num_input_tokens_seen": 314403715,
      "step": 14583,
      "time_per_iteration": 2.449711322784424
    },
    {
      "auxiliary_loss_clip": 0.01095608,
      "auxiliary_loss_mlp": 0.00783612,
      "balance_loss_clip": 1.03504753,
      "balance_loss_mlp": 1.00975358,
      "epoch": 0.8768375169096648,
      "flos": 21247230648960.0,
      "grad_norm": 1.7368585197408792,
      "language_loss": 0.79376423,
      "learning_rate": 1.5695153925422067e-07,
      "loss": 0.81255651,
      "num_input_tokens_seen": 314421880,
      "step": 14584,
      "time_per_iteration": 2.470453977584839
    },
    {
      "auxiliary_loss_clip": 0.0107054,
      "auxiliary_loss_mlp": 0.01026652,
      "balance_loss_clip": 1.03278589,
      "balance_loss_mlp": 1.01493347,
      "epoch": 0.8768976401623327,
      "flos": 23295539715840.0,
      "grad_norm": 1.5368498975494587,
      "language_loss": 0.72249043,
      "learning_rate": 1.5680033810311555e-07,
      "loss": 0.74346232,
      "num_input_tokens_seen": 314441585,
      "step": 14585,
      "time_per_iteration": 2.5647361278533936
    },
    {
      "auxiliary_loss_clip": 0.01082983,
      "auxiliary_loss_mlp": 0.01030792,
      "balance_loss_clip": 1.03425741,
      "balance_loss_mlp": 1.01807868,
      "epoch": 0.8769577634150008,
      "flos": 21361247395200.0,
      "grad_norm": 1.8864474063646308,
      "language_loss": 0.74550539,
      "learning_rate": 1.5664920684632654e-07,
      "loss": 0.76664317,
      "num_input_tokens_seen": 314459020,
      "step": 14586,
      "time_per_iteration": 2.5111515522003174
    },
    {
      "auxiliary_loss_clip": 0.01101846,
      "auxiliary_loss_mlp": 0.01031065,
      "balance_loss_clip": 1.03293884,
      "balance_loss_mlp": 1.01848269,
      "epoch": 0.8770178866676687,
      "flos": 23514056104320.0,
      "grad_norm": 1.795107204544321,
      "language_loss": 0.78684056,
      "learning_rate": 1.564981454895844e-07,
      "loss": 0.8081696,
      "num_input_tokens_seen": 314478935,
      "step": 14587,
      "time_per_iteration": 2.472163200378418
    },
    {
      "auxiliary_loss_clip": 0.01093825,
      "auxiliary_loss_mlp": 0.01029794,
      "balance_loss_clip": 1.03567529,
      "balance_loss_mlp": 1.01601303,
      "epoch": 0.8770780099203367,
      "flos": 19719052473600.0,
      "grad_norm": 1.552354094586315,
      "language_loss": 0.73803759,
      "learning_rate": 1.5634715403861697e-07,
      "loss": 0.75927377,
      "num_input_tokens_seen": 314497635,
      "step": 14588,
      "time_per_iteration": 2.481138229370117
    },
    {
      "auxiliary_loss_clip": 0.01038633,
      "auxiliary_loss_mlp": 0.0078289,
      "balance_loss_clip": 1.0342468,
      "balance_loss_mlp": 1.00923073,
      "epoch": 0.8771381331730047,
      "flos": 21395901041280.0,
      "grad_norm": 1.8019944547569382,
      "language_loss": 0.66809684,
      "learning_rate": 1.5619623249915016e-07,
      "loss": 0.68631208,
      "num_input_tokens_seen": 314515445,
      "step": 14589,
      "time_per_iteration": 2.6260859966278076
    },
    {
      "auxiliary_loss_clip": 0.01093897,
      "auxiliary_loss_mlp": 0.010338,
      "balance_loss_clip": 1.03593969,
      "balance_loss_mlp": 1.02246296,
      "epoch": 0.8771982564256726,
      "flos": 20261770041600.0,
      "grad_norm": 2.683788195986943,
      "language_loss": 0.71015286,
      "learning_rate": 1.5604538087690732e-07,
      "loss": 0.73142982,
      "num_input_tokens_seen": 314533040,
      "step": 14590,
      "time_per_iteration": 2.471261501312256
    },
    {
      "auxiliary_loss_clip": 0.01079882,
      "auxiliary_loss_mlp": 0.01041977,
      "balance_loss_clip": 1.03390408,
      "balance_loss_mlp": 1.02844644,
      "epoch": 0.8772583796783406,
      "flos": 12489372495360.0,
      "grad_norm": 1.9787066279147494,
      "language_loss": 0.74756134,
      "learning_rate": 1.558945991776086e-07,
      "loss": 0.76877993,
      "num_input_tokens_seen": 314548280,
      "step": 14591,
      "time_per_iteration": 2.4989218711853027
    },
    {
      "auxiliary_loss_clip": 0.01099029,
      "auxiliary_loss_mlp": 0.01026882,
      "balance_loss_clip": 1.03485596,
      "balance_loss_mlp": 1.01545024,
      "epoch": 0.8773185029310085,
      "flos": 15921103927680.0,
      "grad_norm": 1.6266467508441815,
      "language_loss": 0.7981028,
      "learning_rate": 1.5574388740697096e-07,
      "loss": 0.81936193,
      "num_input_tokens_seen": 314565345,
      "step": 14592,
      "time_per_iteration": 2.438707113265991
    },
    {
      "auxiliary_loss_clip": 0.01099086,
      "auxiliary_loss_mlp": 0.01029673,
      "balance_loss_clip": 1.03406012,
      "balance_loss_mlp": 1.01871157,
      "epoch": 0.8773786261836766,
      "flos": 21504530747520.0,
      "grad_norm": 1.711805539664191,
      "language_loss": 0.82486278,
      "learning_rate": 1.5559324557071052e-07,
      "loss": 0.8461504,
      "num_input_tokens_seen": 314584190,
      "step": 14593,
      "time_per_iteration": 2.465970277786255
    },
    {
      "auxiliary_loss_clip": 0.01086301,
      "auxiliary_loss_mlp": 0.01026777,
      "balance_loss_clip": 1.03342962,
      "balance_loss_mlp": 1.01505268,
      "epoch": 0.8774387494363445,
      "flos": 26761493831040.0,
      "grad_norm": 1.820524202868089,
      "language_loss": 0.76015413,
      "learning_rate": 1.5544267367453845e-07,
      "loss": 0.78128493,
      "num_input_tokens_seen": 314605625,
      "step": 14594,
      "time_per_iteration": 2.5352697372436523
    },
    {
      "auxiliary_loss_clip": 0.01055159,
      "auxiliary_loss_mlp": 0.0104221,
      "balance_loss_clip": 1.03204441,
      "balance_loss_mlp": 1.02795291,
      "epoch": 0.8774988726890125,
      "flos": 18478841633280.0,
      "grad_norm": 1.9928945141529861,
      "language_loss": 0.77985477,
      "learning_rate": 1.552921717241651e-07,
      "loss": 0.80082846,
      "num_input_tokens_seen": 314622630,
      "step": 14595,
      "time_per_iteration": 2.5600056648254395
    },
    {
      "auxiliary_loss_clip": 0.01073209,
      "auxiliary_loss_mlp": 0.0103432,
      "balance_loss_clip": 1.036448,
      "balance_loss_mlp": 1.02183294,
      "epoch": 0.8775589959416804,
      "flos": 24426366664320.0,
      "grad_norm": 1.7789919688859896,
      "language_loss": 0.70664656,
      "learning_rate": 1.5514173972529743e-07,
      "loss": 0.72772193,
      "num_input_tokens_seen": 314642460,
      "step": 14596,
      "time_per_iteration": 2.5785014629364014
    },
    {
      "auxiliary_loss_clip": 0.01071226,
      "auxiliary_loss_mlp": 0.01025383,
      "balance_loss_clip": 1.03579295,
      "balance_loss_mlp": 1.01420724,
      "epoch": 0.8776191191943484,
      "flos": 23440151871360.0,
      "grad_norm": 2.0064663399457405,
      "language_loss": 0.86263156,
      "learning_rate": 1.5499137768364067e-07,
      "loss": 0.88359761,
      "num_input_tokens_seen": 314659875,
      "step": 14597,
      "time_per_iteration": 2.5707006454467773
    },
    {
      "auxiliary_loss_clip": 0.01092296,
      "auxiliary_loss_mlp": 0.01027897,
      "balance_loss_clip": 1.03555465,
      "balance_loss_mlp": 1.01653683,
      "epoch": 0.8776792424470163,
      "flos": 26830872950400.0,
      "grad_norm": 1.7838088985133032,
      "language_loss": 0.72879219,
      "learning_rate": 1.5484108560489494e-07,
      "loss": 0.7499941,
      "num_input_tokens_seen": 314680260,
      "step": 14598,
      "time_per_iteration": 2.507356882095337
    },
    {
      "auxiliary_loss_clip": 0.01088347,
      "auxiliary_loss_mlp": 0.00785162,
      "balance_loss_clip": 1.0338366,
      "balance_loss_mlp": 1.0109961,
      "epoch": 0.8777393656996844,
      "flos": 15626169354240.0,
      "grad_norm": 2.1692856652516745,
      "language_loss": 0.77458453,
      "learning_rate": 1.5469086349476036e-07,
      "loss": 0.79331964,
      "num_input_tokens_seen": 314696260,
      "step": 14599,
      "time_per_iteration": 3.909712314605713
    },
    {
      "auxiliary_loss_clip": 0.01073125,
      "auxiliary_loss_mlp": 0.01030035,
      "balance_loss_clip": 1.03344822,
      "balance_loss_mlp": 1.01804256,
      "epoch": 0.8777994889523523,
      "flos": 18879999701760.0,
      "grad_norm": 2.226580816326483,
      "language_loss": 0.68367678,
      "learning_rate": 1.545407113589332e-07,
      "loss": 0.7047084,
      "num_input_tokens_seen": 314714215,
      "step": 14600,
      "time_per_iteration": 2.561004400253296
    },
    {
      "auxiliary_loss_clip": 0.0109302,
      "auxiliary_loss_mlp": 0.01034532,
      "balance_loss_clip": 1.03397655,
      "balance_loss_mlp": 1.02271295,
      "epoch": 0.8778596122050203,
      "flos": 48826516400640.0,
      "grad_norm": 1.8325860328536763,
      "language_loss": 0.69647253,
      "learning_rate": 1.543906292031072e-07,
      "loss": 0.71774805,
      "num_input_tokens_seen": 314735700,
      "step": 14601,
      "time_per_iteration": 4.051278829574585
    },
    {
      "auxiliary_loss_clip": 0.01098554,
      "auxiliary_loss_mlp": 0.01030485,
      "balance_loss_clip": 1.03658617,
      "balance_loss_mlp": 1.0182004,
      "epoch": 0.8779197354576883,
      "flos": 25660184883840.0,
      "grad_norm": 1.8153754974772762,
      "language_loss": 0.7329421,
      "learning_rate": 1.542406170329733e-07,
      "loss": 0.75423253,
      "num_input_tokens_seen": 314753335,
      "step": 14602,
      "time_per_iteration": 2.5323855876922607
    },
    {
      "auxiliary_loss_clip": 0.01102541,
      "auxiliary_loss_mlp": 0.01033095,
      "balance_loss_clip": 1.0349673,
      "balance_loss_mlp": 1.02159119,
      "epoch": 0.8779798587103562,
      "flos": 18843227153280.0,
      "grad_norm": 1.7851284889505383,
      "language_loss": 0.71265376,
      "learning_rate": 1.5409067485422056e-07,
      "loss": 0.7340101,
      "num_input_tokens_seen": 314770800,
      "step": 14603,
      "time_per_iteration": 2.4247536659240723
    },
    {
      "auxiliary_loss_clip": 0.01010211,
      "auxiliary_loss_mlp": 0.01000679,
      "balance_loss_clip": 1.0078826,
      "balance_loss_mlp": 0.99949312,
      "epoch": 0.8780399819630242,
      "flos": 68613119377920.0,
      "grad_norm": 0.7427893166199683,
      "language_loss": 0.5415749,
      "learning_rate": 1.539408026725344e-07,
      "loss": 0.56168377,
      "num_input_tokens_seen": 314837275,
      "step": 14604,
      "time_per_iteration": 4.518057823181152
    },
    {
      "auxiliary_loss_clip": 0.01004615,
      "auxiliary_loss_mlp": 0.01019551,
      "balance_loss_clip": 1.01213431,
      "balance_loss_mlp": 1.01807296,
      "epoch": 0.8781001052156922,
      "flos": 65734807766400.0,
      "grad_norm": 0.7080399574980787,
      "language_loss": 0.59196031,
      "learning_rate": 1.537910004935976e-07,
      "loss": 0.61220193,
      "num_input_tokens_seen": 314902220,
      "step": 14605,
      "time_per_iteration": 3.1412038803100586
    },
    {
      "auxiliary_loss_clip": 0.01062915,
      "auxiliary_loss_mlp": 0.0103169,
      "balance_loss_clip": 1.03525138,
      "balance_loss_mlp": 1.01942921,
      "epoch": 0.8781602284683602,
      "flos": 22049654526720.0,
      "grad_norm": 1.5167309964175346,
      "language_loss": 0.84886128,
      "learning_rate": 1.536412683230912e-07,
      "loss": 0.86980736,
      "num_input_tokens_seen": 314921645,
      "step": 14606,
      "time_per_iteration": 2.5748403072357178
    },
    {
      "auxiliary_loss_clip": 0.01106073,
      "auxiliary_loss_mlp": 0.01028092,
      "balance_loss_clip": 1.03638864,
      "balance_loss_mlp": 1.015378,
      "epoch": 0.8782203517210281,
      "flos": 17562939713280.0,
      "grad_norm": 1.9873772268173087,
      "language_loss": 0.70566434,
      "learning_rate": 1.534916061666931e-07,
      "loss": 0.72700596,
      "num_input_tokens_seen": 314939390,
      "step": 14607,
      "time_per_iteration": 2.4378836154937744
    },
    {
      "auxiliary_loss_clip": 0.01082294,
      "auxiliary_loss_mlp": 0.01038853,
      "balance_loss_clip": 1.03491783,
      "balance_loss_mlp": 1.02785611,
      "epoch": 0.8782804749736961,
      "flos": 25520421064320.0,
      "grad_norm": 1.7134894075420333,
      "language_loss": 0.72059816,
      "learning_rate": 1.533420140300785e-07,
      "loss": 0.74180961,
      "num_input_tokens_seen": 314959205,
      "step": 14608,
      "time_per_iteration": 2.550614356994629
    },
    {
      "auxiliary_loss_clip": 0.01095433,
      "auxiliary_loss_mlp": 0.01037858,
      "balance_loss_clip": 1.03460586,
      "balance_loss_mlp": 1.02551997,
      "epoch": 0.878340598226364,
      "flos": 21798747048960.0,
      "grad_norm": 1.9279408882683684,
      "language_loss": 0.87873948,
      "learning_rate": 1.5319249191891936e-07,
      "loss": 0.9000724,
      "num_input_tokens_seen": 314977485,
      "step": 14609,
      "time_per_iteration": 2.4757397174835205
    },
    {
      "auxiliary_loss_clip": 0.01066868,
      "auxiliary_loss_mlp": 0.01029947,
      "balance_loss_clip": 1.03763747,
      "balance_loss_mlp": 1.0184077,
      "epoch": 0.878400721479032,
      "flos": 21102403011840.0,
      "grad_norm": 1.5044353162674386,
      "language_loss": 0.70069259,
      "learning_rate": 1.5304303983888643e-07,
      "loss": 0.72166073,
      "num_input_tokens_seen": 314997830,
      "step": 14610,
      "time_per_iteration": 3.966487169265747
    },
    {
      "auxiliary_loss_clip": 0.01086069,
      "auxiliary_loss_mlp": 0.00784036,
      "balance_loss_clip": 1.03596401,
      "balance_loss_mlp": 1.01020122,
      "epoch": 0.8784608447316999,
      "flos": 20923532259840.0,
      "grad_norm": 1.9118802260705674,
      "language_loss": 0.80287349,
      "learning_rate": 1.5289365779564612e-07,
      "loss": 0.82157457,
      "num_input_tokens_seen": 315016480,
      "step": 14611,
      "time_per_iteration": 2.4641449451446533
    },
    {
      "auxiliary_loss_clip": 0.01104367,
      "auxiliary_loss_mlp": 0.0103128,
      "balance_loss_clip": 1.03538036,
      "balance_loss_mlp": 1.01891816,
      "epoch": 0.878520967984368,
      "flos": 23330660238720.0,
      "grad_norm": 1.4809412094184047,
      "language_loss": 0.76516145,
      "learning_rate": 1.5274434579486338e-07,
      "loss": 0.78651792,
      "num_input_tokens_seen": 315036135,
      "step": 14612,
      "time_per_iteration": 2.4718329906463623
    },
    {
      "auxiliary_loss_clip": 0.01061547,
      "auxiliary_loss_mlp": 0.01031294,
      "balance_loss_clip": 1.03557396,
      "balance_loss_mlp": 1.0196774,
      "epoch": 0.8785810912370359,
      "flos": 25518984520320.0,
      "grad_norm": 1.3809331738800725,
      "language_loss": 0.72162074,
      "learning_rate": 1.525951038422002e-07,
      "loss": 0.74254918,
      "num_input_tokens_seen": 315057995,
      "step": 14613,
      "time_per_iteration": 2.600696086883545
    },
    {
      "auxiliary_loss_clip": 0.01006752,
      "auxiliary_loss_mlp": 0.01004465,
      "balance_loss_clip": 1.01227903,
      "balance_loss_mlp": 1.00344014,
      "epoch": 0.8786412144897039,
      "flos": 61841047691520.0,
      "grad_norm": 1.0336525191694408,
      "language_loss": 0.64586496,
      "learning_rate": 1.5244593194331667e-07,
      "loss": 0.66597712,
      "num_input_tokens_seen": 315104010,
      "step": 14614,
      "time_per_iteration": 2.914268732070923
    },
    {
      "auxiliary_loss_clip": 0.01028271,
      "auxiliary_loss_mlp": 0.01001734,
      "balance_loss_clip": 1.00559855,
      "balance_loss_mlp": 1.00056541,
      "epoch": 0.8787013377423719,
      "flos": 70989364638720.0,
      "grad_norm": 0.6574508025729802,
      "language_loss": 0.58588648,
      "learning_rate": 1.5229683010386762e-07,
      "loss": 0.60618651,
      "num_input_tokens_seen": 315174550,
      "step": 14615,
      "time_per_iteration": 3.1473212242126465
    },
    {
      "auxiliary_loss_clip": 0.01054919,
      "auxiliary_loss_mlp": 0.01027215,
      "balance_loss_clip": 1.03198671,
      "balance_loss_mlp": 1.01526451,
      "epoch": 0.8787614609950398,
      "flos": 17347404153600.0,
      "grad_norm": 2.2667091732037647,
      "language_loss": 0.72930664,
      "learning_rate": 1.5214779832950807e-07,
      "loss": 0.75012797,
      "num_input_tokens_seen": 315191825,
      "step": 14616,
      "time_per_iteration": 2.538909673690796
    },
    {
      "auxiliary_loss_clip": 0.01027894,
      "auxiliary_loss_mlp": 0.0100131,
      "balance_loss_clip": 1.00522876,
      "balance_loss_mlp": 1.0000701,
      "epoch": 0.8788215842477078,
      "flos": 72511401588480.0,
      "grad_norm": 0.834683340959482,
      "language_loss": 0.57955176,
      "learning_rate": 1.5199883662588953e-07,
      "loss": 0.5998438,
      "num_input_tokens_seen": 315255075,
      "step": 14617,
      "time_per_iteration": 3.1584932804107666
    },
    {
      "auxiliary_loss_clip": 0.01077042,
      "auxiliary_loss_mlp": 0.01036508,
      "balance_loss_clip": 1.03305912,
      "balance_loss_mlp": 1.02390718,
      "epoch": 0.8788817075003758,
      "flos": 24827452905600.0,
      "grad_norm": 2.082241990241764,
      "language_loss": 0.83742487,
      "learning_rate": 1.5184994499865987e-07,
      "loss": 0.85856038,
      "num_input_tokens_seen": 315273995,
      "step": 14618,
      "time_per_iteration": 2.5396628379821777
    },
    {
      "auxiliary_loss_clip": 0.0107741,
      "auxiliary_loss_mlp": 0.01027728,
      "balance_loss_clip": 1.03488469,
      "balance_loss_mlp": 1.016397,
      "epoch": 0.8789418307530438,
      "flos": 22638769488000.0,
      "grad_norm": 2.214115332468239,
      "language_loss": 0.68980002,
      "learning_rate": 1.5170112345346598e-07,
      "loss": 0.71085143,
      "num_input_tokens_seen": 315294485,
      "step": 14619,
      "time_per_iteration": 2.5766639709472656
    },
    {
      "auxiliary_loss_clip": 0.0106109,
      "auxiliary_loss_mlp": 0.01038206,
      "balance_loss_clip": 1.03250003,
      "balance_loss_mlp": 1.02638614,
      "epoch": 0.8790019540057117,
      "flos": 19785738072960.0,
      "grad_norm": 1.8797418118623481,
      "language_loss": 0.77363855,
      "learning_rate": 1.5155237199595016e-07,
      "loss": 0.79463148,
      "num_input_tokens_seen": 315310420,
      "step": 14620,
      "time_per_iteration": 2.5460996627807617
    },
    {
      "auxiliary_loss_clip": 0.01080473,
      "auxiliary_loss_mlp": 0.01029387,
      "balance_loss_clip": 1.03625977,
      "balance_loss_mlp": 1.01608872,
      "epoch": 0.8790620772583797,
      "flos": 20229774001920.0,
      "grad_norm": 1.8486336807159003,
      "language_loss": 0.79174066,
      "learning_rate": 1.514036906317542e-07,
      "loss": 0.81283921,
      "num_input_tokens_seen": 315330110,
      "step": 14621,
      "time_per_iteration": 2.570830821990967
    },
    {
      "auxiliary_loss_clip": 0.01083853,
      "auxiliary_loss_mlp": 0.01035455,
      "balance_loss_clip": 1.03409767,
      "balance_loss_mlp": 1.02384377,
      "epoch": 0.8791222005110476,
      "flos": 24130785646080.0,
      "grad_norm": 1.669181962793135,
      "language_loss": 0.67025912,
      "learning_rate": 1.5125507936651506e-07,
      "loss": 0.69145226,
      "num_input_tokens_seen": 315350080,
      "step": 14622,
      "time_per_iteration": 2.538210391998291
    },
    {
      "auxiliary_loss_clip": 0.01075707,
      "auxiliary_loss_mlp": 0.01035913,
      "balance_loss_clip": 1.03514075,
      "balance_loss_mlp": 1.02430773,
      "epoch": 0.8791823237637156,
      "flos": 21614201948160.0,
      "grad_norm": 1.936371781891554,
      "language_loss": 0.73272115,
      "learning_rate": 1.511065382058687e-07,
      "loss": 0.75383735,
      "num_input_tokens_seen": 315366360,
      "step": 14623,
      "time_per_iteration": 2.5156872272491455
    },
    {
      "auxiliary_loss_clip": 0.01052172,
      "auxiliary_loss_mlp": 0.01032132,
      "balance_loss_clip": 1.03256428,
      "balance_loss_mlp": 1.02033067,
      "epoch": 0.8792424470163835,
      "flos": 24243401761920.0,
      "grad_norm": 1.5709613469484087,
      "language_loss": 0.78553683,
      "learning_rate": 1.5095806715544801e-07,
      "loss": 0.80637985,
      "num_input_tokens_seen": 315385890,
      "step": 14624,
      "time_per_iteration": 2.586944580078125
    },
    {
      "auxiliary_loss_clip": 0.01093745,
      "auxiliary_loss_mlp": 0.01034132,
      "balance_loss_clip": 1.03369141,
      "balance_loss_mlp": 1.02078032,
      "epoch": 0.8793025702690516,
      "flos": 24893204751360.0,
      "grad_norm": 2.3012932189067903,
      "language_loss": 0.79610598,
      "learning_rate": 1.5080966622088265e-07,
      "loss": 0.81738472,
      "num_input_tokens_seen": 315403400,
      "step": 14625,
      "time_per_iteration": 2.5191662311553955
    },
    {
      "auxiliary_loss_clip": 0.01081683,
      "auxiliary_loss_mlp": 0.01038191,
      "balance_loss_clip": 1.03566432,
      "balance_loss_mlp": 1.02640724,
      "epoch": 0.8793626935217195,
      "flos": 25373115388800.0,
      "grad_norm": 1.5013243784376216,
      "language_loss": 0.74200475,
      "learning_rate": 1.5066133540779967e-07,
      "loss": 0.7632035,
      "num_input_tokens_seen": 315423670,
      "step": 14626,
      "time_per_iteration": 2.5541443824768066
    },
    {
      "auxiliary_loss_clip": 0.01091911,
      "auxiliary_loss_mlp": 0.01032248,
      "balance_loss_clip": 1.03332472,
      "balance_loss_mlp": 1.02013028,
      "epoch": 0.8794228167743875,
      "flos": 34678000742400.0,
      "grad_norm": 1.4973641746632622,
      "language_loss": 0.70965087,
      "learning_rate": 1.505130747218246e-07,
      "loss": 0.73089248,
      "num_input_tokens_seen": 315446265,
      "step": 14627,
      "time_per_iteration": 2.618730068206787
    },
    {
      "auxiliary_loss_clip": 0.01071453,
      "auxiliary_loss_mlp": 0.01027381,
      "balance_loss_clip": 1.03635728,
      "balance_loss_mlp": 1.0154779,
      "epoch": 0.8794829400270555,
      "flos": 19464014931840.0,
      "grad_norm": 1.771703489346919,
      "language_loss": 0.72234201,
      "learning_rate": 1.5036488416857873e-07,
      "loss": 0.74333036,
      "num_input_tokens_seen": 315464655,
      "step": 14628,
      "time_per_iteration": 2.52742075920105
    },
    {
      "auxiliary_loss_clip": 0.01072688,
      "auxiliary_loss_mlp": 0.01031903,
      "balance_loss_clip": 1.03300428,
      "balance_loss_mlp": 1.01903391,
      "epoch": 0.8795430632797234,
      "flos": 15231403906560.0,
      "grad_norm": 2.5972340846474524,
      "language_loss": 0.69117981,
      "learning_rate": 1.5021676375368175e-07,
      "loss": 0.71222568,
      "num_input_tokens_seen": 315481090,
      "step": 14629,
      "time_per_iteration": 2.5217175483703613
    },
    {
      "auxiliary_loss_clip": 0.01077037,
      "auxiliary_loss_mlp": 0.01031149,
      "balance_loss_clip": 1.03108597,
      "balance_loss_mlp": 1.02027082,
      "epoch": 0.8796031865323914,
      "flos": 27744727795200.0,
      "grad_norm": 1.8869240019281852,
      "language_loss": 0.68406713,
      "learning_rate": 1.5006871348275053e-07,
      "loss": 0.70514899,
      "num_input_tokens_seen": 315502010,
      "step": 14630,
      "time_per_iteration": 2.5296783447265625
    },
    {
      "auxiliary_loss_clip": 0.01077806,
      "auxiliary_loss_mlp": 0.0103239,
      "balance_loss_clip": 1.03352189,
      "balance_loss_mlp": 1.01968825,
      "epoch": 0.8796633097850594,
      "flos": 31285412156160.0,
      "grad_norm": 1.547710883112421,
      "language_loss": 0.74037564,
      "learning_rate": 1.499207333613999e-07,
      "loss": 0.76147759,
      "num_input_tokens_seen": 315523040,
      "step": 14631,
      "time_per_iteration": 2.600100040435791
    },
    {
      "auxiliary_loss_clip": 0.01079386,
      "auxiliary_loss_mlp": 0.00783065,
      "balance_loss_clip": 1.0341382,
      "balance_loss_mlp": 1.00977015,
      "epoch": 0.8797234330377274,
      "flos": 24243150366720.0,
      "grad_norm": 3.4256411302596463,
      "language_loss": 0.69272506,
      "learning_rate": 1.4977282339523954e-07,
      "loss": 0.71134961,
      "num_input_tokens_seen": 315541865,
      "step": 14632,
      "time_per_iteration": 2.5291330814361572
    },
    {
      "auxiliary_loss_clip": 0.01081826,
      "auxiliary_loss_mlp": 0.01028806,
      "balance_loss_clip": 1.03478491,
      "balance_loss_mlp": 1.01789832,
      "epoch": 0.8797835562903953,
      "flos": 24167414540160.0,
      "grad_norm": 1.8881425430915078,
      "language_loss": 0.65150779,
      "learning_rate": 1.4962498358987929e-07,
      "loss": 0.6726141,
      "num_input_tokens_seen": 315561470,
      "step": 14633,
      "time_per_iteration": 2.5611515045166016
    },
    {
      "auxiliary_loss_clip": 0.01072131,
      "auxiliary_loss_mlp": 0.01033045,
      "balance_loss_clip": 1.03304267,
      "balance_loss_mlp": 1.02104616,
      "epoch": 0.8798436795430633,
      "flos": 19284677303040.0,
      "grad_norm": 1.4599655657037083,
      "language_loss": 0.8416748,
      "learning_rate": 1.4947721395092528e-07,
      "loss": 0.86272657,
      "num_input_tokens_seen": 315583140,
      "step": 14634,
      "time_per_iteration": 2.5857927799224854
    },
    {
      "auxiliary_loss_clip": 0.0108044,
      "auxiliary_loss_mlp": 0.00783492,
      "balance_loss_clip": 1.03429651,
      "balance_loss_mlp": 1.00959742,
      "epoch": 0.8799038027957312,
      "flos": 28179390274560.0,
      "grad_norm": 1.6759555894037406,
      "language_loss": 0.79985827,
      "learning_rate": 1.4932951448398056e-07,
      "loss": 0.8184976,
      "num_input_tokens_seen": 315601935,
      "step": 14635,
      "time_per_iteration": 2.563082695007324
    },
    {
      "auxiliary_loss_clip": 0.01083706,
      "auxiliary_loss_mlp": 0.01025511,
      "balance_loss_clip": 1.03590381,
      "balance_loss_mlp": 1.01379299,
      "epoch": 0.8799639260483992,
      "flos": 24644703484800.0,
      "grad_norm": 4.079210051306064,
      "language_loss": 0.65131652,
      "learning_rate": 1.4918188519464648e-07,
      "loss": 0.6724087,
      "num_input_tokens_seen": 315619995,
      "step": 14636,
      "time_per_iteration": 2.5261261463165283
    },
    {
      "auxiliary_loss_clip": 0.0107866,
      "auxiliary_loss_mlp": 0.01033785,
      "balance_loss_clip": 1.03358221,
      "balance_loss_mlp": 1.0212146,
      "epoch": 0.8800240493010671,
      "flos": 22200479735040.0,
      "grad_norm": 1.5199540325047247,
      "language_loss": 0.70205718,
      "learning_rate": 1.4903432608852074e-07,
      "loss": 0.72318166,
      "num_input_tokens_seen": 315637895,
      "step": 14637,
      "time_per_iteration": 2.5262961387634277
    },
    {
      "auxiliary_loss_clip": 0.01082038,
      "auxiliary_loss_mlp": 0.01029176,
      "balance_loss_clip": 1.03789985,
      "balance_loss_mlp": 1.01779723,
      "epoch": 0.8800841725537352,
      "flos": 14246086953600.0,
      "grad_norm": 1.853161733216791,
      "language_loss": 0.65878826,
      "learning_rate": 1.4888683717119843e-07,
      "loss": 0.67990041,
      "num_input_tokens_seen": 315655520,
      "step": 14638,
      "time_per_iteration": 3.905273675918579
    },
    {
      "auxiliary_loss_clip": 0.01093555,
      "auxiliary_loss_mlp": 0.01027582,
      "balance_loss_clip": 1.03531361,
      "balance_loss_mlp": 1.01530337,
      "epoch": 0.8801442958064031,
      "flos": 37415794348800.0,
      "grad_norm": 1.786855018147461,
      "language_loss": 0.58091658,
      "learning_rate": 1.4873941844827286e-07,
      "loss": 0.60212791,
      "num_input_tokens_seen": 315678955,
      "step": 14639,
      "time_per_iteration": 4.027686357498169
    },
    {
      "auxiliary_loss_clip": 0.01084353,
      "auxiliary_loss_mlp": 0.0103635,
      "balance_loss_clip": 1.03490818,
      "balance_loss_mlp": 1.0239346,
      "epoch": 0.8802044190590711,
      "flos": 25047334010880.0,
      "grad_norm": 1.4941852273312795,
      "language_loss": 0.74387896,
      "learning_rate": 1.4859206992533402e-07,
      "loss": 0.76508605,
      "num_input_tokens_seen": 315700360,
      "step": 14640,
      "time_per_iteration": 2.56791353225708
    },
    {
      "auxiliary_loss_clip": 0.01081021,
      "auxiliary_loss_mlp": 0.01041326,
      "balance_loss_clip": 1.03451824,
      "balance_loss_mlp": 1.02859473,
      "epoch": 0.8802645423117391,
      "flos": 24133874215680.0,
      "grad_norm": 1.946175830057362,
      "language_loss": 0.69632304,
      "learning_rate": 1.4844479160796985e-07,
      "loss": 0.71754646,
      "num_input_tokens_seen": 315719270,
      "step": 14641,
      "time_per_iteration": 2.5268938541412354
    },
    {
      "auxiliary_loss_clip": 0.01091141,
      "auxiliary_loss_mlp": 0.01031897,
      "balance_loss_clip": 1.03554153,
      "balance_loss_mlp": 1.01895702,
      "epoch": 0.880324665564407,
      "flos": 17931203902080.0,
      "grad_norm": 2.0741513309650483,
      "language_loss": 0.84828615,
      "learning_rate": 1.4829758350176457e-07,
      "loss": 0.86951649,
      "num_input_tokens_seen": 315737425,
      "step": 14642,
      "time_per_iteration": 3.839522123336792
    },
    {
      "auxiliary_loss_clip": 0.01064313,
      "auxiliary_loss_mlp": 0.01032187,
      "balance_loss_clip": 1.03909349,
      "balance_loss_mlp": 1.01954496,
      "epoch": 0.880384788817075,
      "flos": 21287630471040.0,
      "grad_norm": 1.7098187124885011,
      "language_loss": 0.78632009,
      "learning_rate": 1.4815044561230038e-07,
      "loss": 0.80728513,
      "num_input_tokens_seen": 315755725,
      "step": 14643,
      "time_per_iteration": 2.610787868499756
    },
    {
      "auxiliary_loss_clip": 0.01086934,
      "auxiliary_loss_mlp": 0.01025768,
      "balance_loss_clip": 1.03188634,
      "balance_loss_mlp": 1.01492572,
      "epoch": 0.880444912069743,
      "flos": 12458489777280.0,
      "grad_norm": 1.617858352543953,
      "language_loss": 0.73052353,
      "learning_rate": 1.4800337794515705e-07,
      "loss": 0.75165057,
      "num_input_tokens_seen": 315773835,
      "step": 14644,
      "time_per_iteration": 2.4823882579803467
    },
    {
      "auxiliary_loss_clip": 0.01108051,
      "auxiliary_loss_mlp": 0.00784017,
      "balance_loss_clip": 1.03694463,
      "balance_loss_mlp": 1.01046312,
      "epoch": 0.880505035322411,
      "flos": 13625945619840.0,
      "grad_norm": 2.5912239350573403,
      "language_loss": 0.79616964,
      "learning_rate": 1.47856380505911e-07,
      "loss": 0.8150903,
      "num_input_tokens_seen": 315790615,
      "step": 14645,
      "time_per_iteration": 2.442098379135132
    },
    {
      "auxiliary_loss_clip": 0.0108865,
      "auxiliary_loss_mlp": 0.01033835,
      "balance_loss_clip": 1.03327823,
      "balance_loss_mlp": 1.02197969,
      "epoch": 0.8805651585750789,
      "flos": 23183067254400.0,
      "grad_norm": 2.5832450540930965,
      "language_loss": 0.64299381,
      "learning_rate": 1.477094533001364e-07,
      "loss": 0.6642186,
      "num_input_tokens_seen": 315811010,
      "step": 14646,
      "time_per_iteration": 2.483525514602661
    },
    {
      "auxiliary_loss_clip": 0.0108181,
      "auxiliary_loss_mlp": 0.01029655,
      "balance_loss_clip": 1.03689814,
      "balance_loss_mlp": 1.01623189,
      "epoch": 0.8806252818277469,
      "flos": 14903000835840.0,
      "grad_norm": 10.390273405312833,
      "language_loss": 0.77481598,
      "learning_rate": 1.475625963334055e-07,
      "loss": 0.79593062,
      "num_input_tokens_seen": 315828130,
      "step": 14647,
      "time_per_iteration": 2.519935369491577
    },
    {
      "auxiliary_loss_clip": 0.01101421,
      "auxiliary_loss_mlp": 0.01030386,
      "balance_loss_clip": 1.03483868,
      "balance_loss_mlp": 1.01928806,
      "epoch": 0.8806854050804148,
      "flos": 17639178330240.0,
      "grad_norm": 1.9201412971036431,
      "language_loss": 0.7492525,
      "learning_rate": 1.4741580961128652e-07,
      "loss": 0.77057052,
      "num_input_tokens_seen": 315844900,
      "step": 14648,
      "time_per_iteration": 3.785038471221924
    },
    {
      "auxiliary_loss_clip": 0.01082186,
      "auxiliary_loss_mlp": 0.01031442,
      "balance_loss_clip": 1.03259921,
      "balance_loss_mlp": 1.01973009,
      "epoch": 0.8807455283330828,
      "flos": 25332392344320.0,
      "grad_norm": 1.6155078525464013,
      "language_loss": 0.65388262,
      "learning_rate": 1.4726909313934522e-07,
      "loss": 0.67501891,
      "num_input_tokens_seen": 315863745,
      "step": 14649,
      "time_per_iteration": 2.5493342876434326
    },
    {
      "auxiliary_loss_clip": 0.01067826,
      "auxiliary_loss_mlp": 0.01028351,
      "balance_loss_clip": 1.03565502,
      "balance_loss_mlp": 1.01609683,
      "epoch": 0.8808056515857507,
      "flos": 25265168040960.0,
      "grad_norm": 1.3403565553693535,
      "language_loss": 0.62467444,
      "learning_rate": 1.4712244692314578e-07,
      "loss": 0.64563626,
      "num_input_tokens_seen": 315885765,
      "step": 14650,
      "time_per_iteration": 2.6117103099823
    },
    {
      "auxiliary_loss_clip": 0.01076632,
      "auxiliary_loss_mlp": 0.01031937,
      "balance_loss_clip": 1.03328896,
      "balance_loss_mlp": 1.02026033,
      "epoch": 0.8808657748384188,
      "flos": 26578852151040.0,
      "grad_norm": 1.4887103720652275,
      "language_loss": 0.72737384,
      "learning_rate": 1.4697587096824914e-07,
      "loss": 0.74845952,
      "num_input_tokens_seen": 315907340,
      "step": 14651,
      "time_per_iteration": 2.5557219982147217
    },
    {
      "auxiliary_loss_clip": 0.010954,
      "auxiliary_loss_mlp": 0.01031503,
      "balance_loss_clip": 1.0344609,
      "balance_loss_mlp": 1.01862276,
      "epoch": 0.8809258980910867,
      "flos": 18661231918080.0,
      "grad_norm": 1.7706394469700244,
      "language_loss": 0.71590579,
      "learning_rate": 1.4682936528021284e-07,
      "loss": 0.73717475,
      "num_input_tokens_seen": 315924935,
      "step": 14652,
      "time_per_iteration": 2.4707043170928955
    },
    {
      "auxiliary_loss_clip": 0.01091074,
      "auxiliary_loss_mlp": 0.01029051,
      "balance_loss_clip": 1.03380108,
      "balance_loss_mlp": 1.01753545,
      "epoch": 0.8809860213437547,
      "flos": 19792274348160.0,
      "grad_norm": 1.9643613176033254,
      "language_loss": 0.75027466,
      "learning_rate": 1.4668292986459286e-07,
      "loss": 0.77147591,
      "num_input_tokens_seen": 315943165,
      "step": 14653,
      "time_per_iteration": 2.4665112495422363
    },
    {
      "auxiliary_loss_clip": 0.01107304,
      "auxiliary_loss_mlp": 0.01028348,
      "balance_loss_clip": 1.03541207,
      "balance_loss_mlp": 1.01580727,
      "epoch": 0.8810461445964227,
      "flos": 17894467267200.0,
      "grad_norm": 1.7483420752303584,
      "language_loss": 0.70978028,
      "learning_rate": 1.465365647269421e-07,
      "loss": 0.7311368,
      "num_input_tokens_seen": 315961340,
      "step": 14654,
      "time_per_iteration": 2.4366378784179688
    },
    {
      "auxiliary_loss_clip": 0.01057033,
      "auxiliary_loss_mlp": 0.01034639,
      "balance_loss_clip": 1.03376842,
      "balance_loss_mlp": 1.02163959,
      "epoch": 0.8811062678490906,
      "flos": 29163917128320.0,
      "grad_norm": 1.5743215652977551,
      "language_loss": 0.71492249,
      "learning_rate": 1.4639026987281012e-07,
      "loss": 0.73583925,
      "num_input_tokens_seen": 315981335,
      "step": 14655,
      "time_per_iteration": 2.6360890865325928
    },
    {
      "auxiliary_loss_clip": 0.01055464,
      "auxiliary_loss_mlp": 0.01034065,
      "balance_loss_clip": 1.03229356,
      "balance_loss_mlp": 1.02126169,
      "epoch": 0.8811663911017587,
      "flos": 20338834671360.0,
      "grad_norm": 1.5748557833498082,
      "language_loss": 0.81368953,
      "learning_rate": 1.462440453077449e-07,
      "loss": 0.83458483,
      "num_input_tokens_seen": 316001325,
      "step": 14656,
      "time_per_iteration": 2.592865228652954
    },
    {
      "auxiliary_loss_clip": 0.01078218,
      "auxiliary_loss_mlp": 0.01033717,
      "balance_loss_clip": 1.03558683,
      "balance_loss_mlp": 1.02235603,
      "epoch": 0.8812265143544266,
      "flos": 25885704424320.0,
      "grad_norm": 1.6543748297349437,
      "language_loss": 0.6851685,
      "learning_rate": 1.460978910372914e-07,
      "loss": 0.70628786,
      "num_input_tokens_seen": 316022540,
      "step": 14657,
      "time_per_iteration": 2.54152512550354
    },
    {
      "auxiliary_loss_clip": 0.01079644,
      "auxiliary_loss_mlp": 0.01033629,
      "balance_loss_clip": 1.03610027,
      "balance_loss_mlp": 1.02218461,
      "epoch": 0.8812866376070946,
      "flos": 27195509865600.0,
      "grad_norm": 2.0700842546487483,
      "language_loss": 0.83929151,
      "learning_rate": 1.4595180706699207e-07,
      "loss": 0.86042428,
      "num_input_tokens_seen": 316037735,
      "step": 14658,
      "time_per_iteration": 2.6052651405334473
    },
    {
      "auxiliary_loss_clip": 0.01089295,
      "auxiliary_loss_mlp": 0.01034211,
      "balance_loss_clip": 1.03685451,
      "balance_loss_mlp": 1.02133012,
      "epoch": 0.8813467608597625,
      "flos": 23807194997760.0,
      "grad_norm": 1.8680963205169296,
      "language_loss": 0.77178687,
      "learning_rate": 1.4580579340238554e-07,
      "loss": 0.79302192,
      "num_input_tokens_seen": 316058105,
      "step": 14659,
      "time_per_iteration": 2.5348258018493652
    },
    {
      "auxiliary_loss_clip": 0.01080327,
      "auxiliary_loss_mlp": 0.01031207,
      "balance_loss_clip": 1.03407645,
      "balance_loss_mlp": 1.01942277,
      "epoch": 0.8814068841124305,
      "flos": 21105455667840.0,
      "grad_norm": 2.1239191118891414,
      "language_loss": 0.60359359,
      "learning_rate": 1.4565985004900894e-07,
      "loss": 0.62470895,
      "num_input_tokens_seen": 316074415,
      "step": 14660,
      "time_per_iteration": 2.513101577758789
    },
    {
      "auxiliary_loss_clip": 0.01086786,
      "auxiliary_loss_mlp": 0.01035785,
      "balance_loss_clip": 1.03552318,
      "balance_loss_mlp": 1.02235627,
      "epoch": 0.8814670073650984,
      "flos": 24716991605760.0,
      "grad_norm": 2.3045393297484273,
      "language_loss": 0.77875531,
      "learning_rate": 1.455139770123972e-07,
      "loss": 0.799981,
      "num_input_tokens_seen": 316094405,
      "step": 14661,
      "time_per_iteration": 2.535114049911499
    },
    {
      "auxiliary_loss_clip": 0.01069581,
      "auxiliary_loss_mlp": 0.01040205,
      "balance_loss_clip": 1.03667402,
      "balance_loss_mlp": 1.02768815,
      "epoch": 0.8815271306177664,
      "flos": 22966274718720.0,
      "grad_norm": 1.5827849139445775,
      "language_loss": 0.76583159,
      "learning_rate": 1.45368174298081e-07,
      "loss": 0.78692943,
      "num_input_tokens_seen": 316113390,
      "step": 14662,
      "time_per_iteration": 2.562130928039551
    },
    {
      "auxiliary_loss_clip": 0.01053869,
      "auxiliary_loss_mlp": 0.01028311,
      "balance_loss_clip": 1.03152478,
      "balance_loss_mlp": 1.0171175,
      "epoch": 0.8815872538704344,
      "flos": 19460064435840.0,
      "grad_norm": 2.1148474378225672,
      "language_loss": 0.73833179,
      "learning_rate": 1.4522244191158929e-07,
      "loss": 0.7591536,
      "num_input_tokens_seen": 316131085,
      "step": 14663,
      "time_per_iteration": 2.5558643341064453
    },
    {
      "auxiliary_loss_clip": 0.01093655,
      "auxiliary_loss_mlp": 0.00782393,
      "balance_loss_clip": 1.03564835,
      "balance_loss_mlp": 1.0080992,
      "epoch": 0.8816473771231024,
      "flos": 32156604622080.0,
      "grad_norm": 1.6146538127166332,
      "language_loss": 0.70134318,
      "learning_rate": 1.450767798584489e-07,
      "loss": 0.72010368,
      "num_input_tokens_seen": 316151440,
      "step": 14664,
      "time_per_iteration": 2.591186285018921
    },
    {
      "auxiliary_loss_clip": 0.01024166,
      "auxiliary_loss_mlp": 0.01034836,
      "balance_loss_clip": 1.03189337,
      "balance_loss_mlp": 1.02295065,
      "epoch": 0.8817075003757703,
      "flos": 19682279925120.0,
      "grad_norm": 1.5074119089308244,
      "language_loss": 0.81304532,
      "learning_rate": 1.449311881441828e-07,
      "loss": 0.83363533,
      "num_input_tokens_seen": 316170750,
      "step": 14665,
      "time_per_iteration": 2.612718105316162
    },
    {
      "auxiliary_loss_clip": 0.01080937,
      "auxiliary_loss_mlp": 0.01032526,
      "balance_loss_clip": 1.03623128,
      "balance_loss_mlp": 1.02107,
      "epoch": 0.8817676236284383,
      "flos": 15668616251520.0,
      "grad_norm": 2.8075249969635387,
      "language_loss": 0.58397007,
      "learning_rate": 1.447856667743117e-07,
      "loss": 0.60510468,
      "num_input_tokens_seen": 316187265,
      "step": 14666,
      "time_per_iteration": 2.515561103820801
    },
    {
      "auxiliary_loss_clip": 0.01096195,
      "auxiliary_loss_mlp": 0.01030454,
      "balance_loss_clip": 1.03710592,
      "balance_loss_mlp": 1.01750779,
      "epoch": 0.8818277468811063,
      "flos": 17895185539200.0,
      "grad_norm": 1.9440911163955525,
      "language_loss": 0.83874917,
      "learning_rate": 1.4464021575435403e-07,
      "loss": 0.86001563,
      "num_input_tokens_seen": 316206555,
      "step": 14667,
      "time_per_iteration": 2.455974578857422
    },
    {
      "auxiliary_loss_clip": 0.01103368,
      "auxiliary_loss_mlp": 0.01034996,
      "balance_loss_clip": 1.03497219,
      "balance_loss_mlp": 1.02215111,
      "epoch": 0.8818878701337742,
      "flos": 18770508069120.0,
      "grad_norm": 1.7799638198058874,
      "language_loss": 0.62551093,
      "learning_rate": 1.4449483508982563e-07,
      "loss": 0.64689457,
      "num_input_tokens_seen": 316225210,
      "step": 14668,
      "time_per_iteration": 2.454132318496704
    },
    {
      "auxiliary_loss_clip": 0.01089851,
      "auxiliary_loss_mlp": 0.01025677,
      "balance_loss_clip": 1.03451073,
      "balance_loss_mlp": 1.01538396,
      "epoch": 0.8819479933864423,
      "flos": 17712292464000.0,
      "grad_norm": 2.587715456975285,
      "language_loss": 0.57267392,
      "learning_rate": 1.4434952478623918e-07,
      "loss": 0.59382915,
      "num_input_tokens_seen": 316242685,
      "step": 14669,
      "time_per_iteration": 2.435689687728882
    },
    {
      "auxiliary_loss_clip": 0.01101709,
      "auxiliary_loss_mlp": 0.01030722,
      "balance_loss_clip": 1.03327,
      "balance_loss_mlp": 1.01896191,
      "epoch": 0.8820081166391102,
      "flos": 11728749070080.0,
      "grad_norm": 1.897807524866727,
      "language_loss": 0.71491289,
      "learning_rate": 1.442042848491043e-07,
      "loss": 0.73623723,
      "num_input_tokens_seen": 316260935,
      "step": 14670,
      "time_per_iteration": 2.4206132888793945
    },
    {
      "auxiliary_loss_clip": 0.01085181,
      "auxiliary_loss_mlp": 0.01032118,
      "balance_loss_clip": 1.03118598,
      "balance_loss_mlp": 1.0199883,
      "epoch": 0.8820682398917782,
      "flos": 27490372611840.0,
      "grad_norm": 2.369410977798449,
      "language_loss": 0.73812664,
      "learning_rate": 1.44059115283929e-07,
      "loss": 0.75929964,
      "num_input_tokens_seen": 316281190,
      "step": 14671,
      "time_per_iteration": 2.512629508972168
    },
    {
      "auxiliary_loss_clip": 0.01082141,
      "auxiliary_loss_mlp": 0.01028398,
      "balance_loss_clip": 1.03245974,
      "balance_loss_mlp": 1.01588655,
      "epoch": 0.8821283631444461,
      "flos": 16873850223360.0,
      "grad_norm": 1.9828685639254646,
      "language_loss": 0.84677428,
      "learning_rate": 1.43914016096218e-07,
      "loss": 0.86787969,
      "num_input_tokens_seen": 316297115,
      "step": 14672,
      "time_per_iteration": 2.4755032062530518
    },
    {
      "auxiliary_loss_clip": 0.01066182,
      "auxiliary_loss_mlp": 0.01028314,
      "balance_loss_clip": 1.03240764,
      "balance_loss_mlp": 1.01677465,
      "epoch": 0.8821884863971141,
      "flos": 24280964409600.0,
      "grad_norm": 1.5758692854779395,
      "language_loss": 0.7291432,
      "learning_rate": 1.4376898729147336e-07,
      "loss": 0.7500881,
      "num_input_tokens_seen": 316318235,
      "step": 14673,
      "time_per_iteration": 2.5563881397247314
    },
    {
      "auxiliary_loss_clip": 0.01010085,
      "auxiliary_loss_mlp": 0.0100078,
      "balance_loss_clip": 1.00747216,
      "balance_loss_mlp": 0.9996832,
      "epoch": 0.882248609649782,
      "flos": 59432342492160.0,
      "grad_norm": 0.9031262024706469,
      "language_loss": 0.49440524,
      "learning_rate": 1.4362402887519487e-07,
      "loss": 0.51451385,
      "num_input_tokens_seen": 316384705,
      "step": 14674,
      "time_per_iteration": 3.2296383380889893
    },
    {
      "auxiliary_loss_clip": 0.01079033,
      "auxiliary_loss_mlp": 0.00783945,
      "balance_loss_clip": 1.03297079,
      "balance_loss_mlp": 1.00936711,
      "epoch": 0.88230873290245,
      "flos": 19937784343680.0,
      "grad_norm": 2.0786343326397048,
      "language_loss": 0.76439524,
      "learning_rate": 1.4347914085287971e-07,
      "loss": 0.78302503,
      "num_input_tokens_seen": 316401165,
      "step": 14675,
      "time_per_iteration": 2.4851183891296387
    },
    {
      "auxiliary_loss_clip": 0.01077382,
      "auxiliary_loss_mlp": 0.0103157,
      "balance_loss_clip": 1.03577757,
      "balance_loss_mlp": 1.01994133,
      "epoch": 0.882368856155118,
      "flos": 16362769559040.0,
      "grad_norm": 1.7249630414055,
      "language_loss": 0.79319918,
      "learning_rate": 1.4333432323002105e-07,
      "loss": 0.81428874,
      "num_input_tokens_seen": 316418780,
      "step": 14676,
      "time_per_iteration": 2.5118613243103027
    },
    {
      "auxiliary_loss_clip": 0.01008989,
      "auxiliary_loss_mlp": 0.01008618,
      "balance_loss_clip": 1.0227859,
      "balance_loss_mlp": 1.00713396,
      "epoch": 0.882428979407786,
      "flos": 70594563277440.0,
      "grad_norm": 0.6904998444433798,
      "language_loss": 0.54740286,
      "learning_rate": 1.431895760121109e-07,
      "loss": 0.56757891,
      "num_input_tokens_seen": 316482030,
      "step": 14677,
      "time_per_iteration": 4.6408281326293945
    },
    {
      "auxiliary_loss_clip": 0.01100209,
      "auxiliary_loss_mlp": 0.01025366,
      "balance_loss_clip": 1.0331192,
      "balance_loss_mlp": 1.01342738,
      "epoch": 0.8824891026604539,
      "flos": 18150294908160.0,
      "grad_norm": 2.202151547236294,
      "language_loss": 0.64733076,
      "learning_rate": 1.4304489920463847e-07,
      "loss": 0.66858649,
      "num_input_tokens_seen": 316499175,
      "step": 14678,
      "time_per_iteration": 3.867016315460205
    },
    {
      "auxiliary_loss_clip": 0.01081652,
      "auxiliary_loss_mlp": 0.01030126,
      "balance_loss_clip": 1.03357279,
      "balance_loss_mlp": 1.01797879,
      "epoch": 0.8825492259131219,
      "flos": 27232713377280.0,
      "grad_norm": 2.948543293625489,
      "language_loss": 0.71026963,
      "learning_rate": 1.4290029281308936e-07,
      "loss": 0.73138738,
      "num_input_tokens_seen": 316519495,
      "step": 14679,
      "time_per_iteration": 2.6133668422698975
    },
    {
      "auxiliary_loss_clip": 0.0107917,
      "auxiliary_loss_mlp": 0.01029838,
      "balance_loss_clip": 1.03453362,
      "balance_loss_mlp": 1.01919246,
      "epoch": 0.8826093491657898,
      "flos": 22274419881600.0,
      "grad_norm": 1.6092427029812757,
      "language_loss": 0.64054632,
      "learning_rate": 1.4275575684294694e-07,
      "loss": 0.66163635,
      "num_input_tokens_seen": 316538180,
      "step": 14680,
      "time_per_iteration": 2.5834507942199707
    },
    {
      "auxiliary_loss_clip": 0.01101543,
      "auxiliary_loss_mlp": 0.01035048,
      "balance_loss_clip": 1.03419065,
      "balance_loss_mlp": 1.02286446,
      "epoch": 0.8826694724184578,
      "flos": 14204753377920.0,
      "grad_norm": 2.2507948300580862,
      "language_loss": 0.77246958,
      "learning_rate": 1.4261129129969328e-07,
      "loss": 0.79383552,
      "num_input_tokens_seen": 316551750,
      "step": 14681,
      "time_per_iteration": 3.8344619274139404
    },
    {
      "auxiliary_loss_clip": 0.01082377,
      "auxiliary_loss_mlp": 0.01032077,
      "balance_loss_clip": 1.03443694,
      "balance_loss_mlp": 1.01912498,
      "epoch": 0.8827295956711259,
      "flos": 20631686256000.0,
      "grad_norm": 1.6879881391423697,
      "language_loss": 0.7277016,
      "learning_rate": 1.424668961888047e-07,
      "loss": 0.74884617,
      "num_input_tokens_seen": 316570680,
      "step": 14682,
      "time_per_iteration": 2.5244860649108887
    },
    {
      "auxiliary_loss_clip": 0.01061083,
      "auxiliary_loss_mlp": 0.01029669,
      "balance_loss_clip": 1.03673244,
      "balance_loss_mlp": 1.01651478,
      "epoch": 0.8827897189237938,
      "flos": 18513064316160.0,
      "grad_norm": 1.72751017398126,
      "language_loss": 0.74535066,
      "learning_rate": 1.4232257151575765e-07,
      "loss": 0.76625812,
      "num_input_tokens_seen": 316588635,
      "step": 14683,
      "time_per_iteration": 2.5575673580169678
    },
    {
      "auxiliary_loss_clip": 0.01069967,
      "auxiliary_loss_mlp": 0.01032857,
      "balance_loss_clip": 1.03317022,
      "balance_loss_mlp": 1.02020335,
      "epoch": 0.8828498421764618,
      "flos": 22747399194240.0,
      "grad_norm": 2.2131483989048766,
      "language_loss": 0.66104287,
      "learning_rate": 1.4217831728602492e-07,
      "loss": 0.68207109,
      "num_input_tokens_seen": 316607550,
      "step": 14684,
      "time_per_iteration": 2.548283100128174
    },
    {
      "auxiliary_loss_clip": 0.01087725,
      "auxiliary_loss_mlp": 0.01026464,
      "balance_loss_clip": 1.03534544,
      "balance_loss_mlp": 1.01544905,
      "epoch": 0.8829099654291297,
      "flos": 15012384727680.0,
      "grad_norm": 2.3215052655237813,
      "language_loss": 0.69278669,
      "learning_rate": 1.4203413350507677e-07,
      "loss": 0.71392858,
      "num_input_tokens_seen": 316624460,
      "step": 14685,
      "time_per_iteration": 2.434131383895874
    },
    {
      "auxiliary_loss_clip": 0.01051374,
      "auxiliary_loss_mlp": 0.01034511,
      "balance_loss_clip": 1.03363085,
      "balance_loss_mlp": 1.02075982,
      "epoch": 0.8829700886817977,
      "flos": 16720546976640.0,
      "grad_norm": 1.891880050579418,
      "language_loss": 0.74578744,
      "learning_rate": 1.418900201783806e-07,
      "loss": 0.76664627,
      "num_input_tokens_seen": 316640765,
      "step": 14686,
      "time_per_iteration": 2.5848681926727295
    },
    {
      "auxiliary_loss_clip": 0.01052191,
      "auxiliary_loss_mlp": 0.01026832,
      "balance_loss_clip": 1.031479,
      "balance_loss_mlp": 1.01469111,
      "epoch": 0.8830302119344656,
      "flos": 15263256291840.0,
      "grad_norm": 1.7185441602709122,
      "language_loss": 0.63053674,
      "learning_rate": 1.417459773114007e-07,
      "loss": 0.65132701,
      "num_input_tokens_seen": 316656120,
      "step": 14687,
      "time_per_iteration": 3.9287831783294678
    },
    {
      "auxiliary_loss_clip": 0.01094763,
      "auxiliary_loss_mlp": 0.01035669,
      "balance_loss_clip": 1.03521419,
      "balance_loss_mlp": 1.02355123,
      "epoch": 0.8830903351871336,
      "flos": 28617751854720.0,
      "grad_norm": 1.7404872463509278,
      "language_loss": 0.68960643,
      "learning_rate": 1.4160200490959984e-07,
      "loss": 0.71091068,
      "num_input_tokens_seen": 316676095,
      "step": 14688,
      "time_per_iteration": 2.524101495742798
    },
    {
      "auxiliary_loss_clip": 0.01089368,
      "auxiliary_loss_mlp": 0.01024528,
      "balance_loss_clip": 1.03433037,
      "balance_loss_mlp": 1.01255989,
      "epoch": 0.8831504584398016,
      "flos": 28001632844160.0,
      "grad_norm": 2.1786073414679814,
      "language_loss": 0.6695354,
      "learning_rate": 1.4145810297843697e-07,
      "loss": 0.69067442,
      "num_input_tokens_seen": 316696235,
      "step": 14689,
      "time_per_iteration": 2.556540012359619
    },
    {
      "auxiliary_loss_clip": 0.01083804,
      "auxiliary_loss_mlp": 0.01029366,
      "balance_loss_clip": 1.03779721,
      "balance_loss_mlp": 1.01789832,
      "epoch": 0.8832105816924696,
      "flos": 26579642250240.0,
      "grad_norm": 1.3857737616894588,
      "language_loss": 0.74552488,
      "learning_rate": 1.4131427152336905e-07,
      "loss": 0.76665658,
      "num_input_tokens_seen": 316719680,
      "step": 14690,
      "time_per_iteration": 2.5918853282928467
    },
    {
      "auxiliary_loss_clip": 0.01080115,
      "auxiliary_loss_mlp": 0.01036753,
      "balance_loss_clip": 1.03350711,
      "balance_loss_mlp": 1.02381849,
      "epoch": 0.8832707049451375,
      "flos": 24898771359360.0,
      "grad_norm": 1.4312646398270445,
      "language_loss": 0.72932351,
      "learning_rate": 1.4117051054985018e-07,
      "loss": 0.75049222,
      "num_input_tokens_seen": 316739830,
      "step": 14691,
      "time_per_iteration": 2.5568630695343018
    },
    {
      "auxiliary_loss_clip": 0.01069511,
      "auxiliary_loss_mlp": 0.01031975,
      "balance_loss_clip": 1.03466439,
      "balance_loss_mlp": 1.01958942,
      "epoch": 0.8833308281978055,
      "flos": 15451141357440.0,
      "grad_norm": 2.0239292550251187,
      "language_loss": 0.52185607,
      "learning_rate": 1.4102682006333243e-07,
      "loss": 0.54287088,
      "num_input_tokens_seen": 316758105,
      "step": 14692,
      "time_per_iteration": 2.5213332176208496
    },
    {
      "auxiliary_loss_clip": 0.01067591,
      "auxiliary_loss_mlp": 0.01032058,
      "balance_loss_clip": 1.03589511,
      "balance_loss_mlp": 1.02032208,
      "epoch": 0.8833909514504734,
      "flos": 20301523418880.0,
      "grad_norm": 1.9377649427944639,
      "language_loss": 0.60379422,
      "learning_rate": 1.4088320006926346e-07,
      "loss": 0.62479067,
      "num_input_tokens_seen": 316777455,
      "step": 14693,
      "time_per_iteration": 2.5571722984313965
    },
    {
      "auxiliary_loss_clip": 0.01099251,
      "auxiliary_loss_mlp": 0.01027234,
      "balance_loss_clip": 1.03462863,
      "balance_loss_mlp": 1.01637459,
      "epoch": 0.8834510747031414,
      "flos": 20374027021440.0,
      "grad_norm": 1.5359638723286422,
      "language_loss": 0.75462133,
      "learning_rate": 1.407396505730898e-07,
      "loss": 0.77588618,
      "num_input_tokens_seen": 316796300,
      "step": 14694,
      "time_per_iteration": 2.436044931411743
    },
    {
      "auxiliary_loss_clip": 0.01094032,
      "auxiliary_loss_mlp": 0.01028754,
      "balance_loss_clip": 1.03343821,
      "balance_loss_mlp": 1.0175662,
      "epoch": 0.8835111979558095,
      "flos": 29752026508800.0,
      "grad_norm": 1.6551153299245214,
      "language_loss": 0.72606051,
      "learning_rate": 1.4059617158025527e-07,
      "loss": 0.74728835,
      "num_input_tokens_seen": 316819090,
      "step": 14695,
      "time_per_iteration": 2.568169355392456
    },
    {
      "auxiliary_loss_clip": 0.01087542,
      "auxiliary_loss_mlp": 0.0102505,
      "balance_loss_clip": 1.03412998,
      "balance_loss_mlp": 1.01413083,
      "epoch": 0.8835713212084774,
      "flos": 24134556574080.0,
      "grad_norm": 1.5764595092189218,
      "language_loss": 0.79756117,
      "learning_rate": 1.404527630961998e-07,
      "loss": 0.81868708,
      "num_input_tokens_seen": 316839250,
      "step": 14696,
      "time_per_iteration": 2.5145013332366943
    },
    {
      "auxiliary_loss_clip": 0.01062526,
      "auxiliary_loss_mlp": 0.01030234,
      "balance_loss_clip": 1.03328013,
      "balance_loss_mlp": 1.01860499,
      "epoch": 0.8836314444611454,
      "flos": 27672331933440.0,
      "grad_norm": 1.4560514621954532,
      "language_loss": 0.74691707,
      "learning_rate": 1.4030942512636236e-07,
      "loss": 0.76784462,
      "num_input_tokens_seen": 316861315,
      "step": 14697,
      "time_per_iteration": 2.633427381515503
    },
    {
      "auxiliary_loss_clip": 0.01081174,
      "auxiliary_loss_mlp": 0.01032452,
      "balance_loss_clip": 1.03431666,
      "balance_loss_mlp": 1.02053761,
      "epoch": 0.8836915677138133,
      "flos": 16836969934080.0,
      "grad_norm": 2.0537120406881497,
      "language_loss": 0.72145909,
      "learning_rate": 1.401661576761779e-07,
      "loss": 0.74259531,
      "num_input_tokens_seen": 316879325,
      "step": 14698,
      "time_per_iteration": 2.487354278564453
    },
    {
      "auxiliary_loss_clip": 0.01019065,
      "auxiliary_loss_mlp": 0.01001155,
      "balance_loss_clip": 1.00711155,
      "balance_loss_mlp": 1.00019538,
      "epoch": 0.8837516909664813,
      "flos": 69310540823040.0,
      "grad_norm": 0.8117746981532594,
      "language_loss": 0.53697282,
      "learning_rate": 1.4002296075107856e-07,
      "loss": 0.55717498,
      "num_input_tokens_seen": 316936425,
      "step": 14699,
      "time_per_iteration": 3.10251522064209
    },
    {
      "auxiliary_loss_clip": 0.01083333,
      "auxiliary_loss_mlp": 0.01030563,
      "balance_loss_clip": 1.03436518,
      "balance_loss_mlp": 1.01805198,
      "epoch": 0.8838118142191492,
      "flos": 21324726241920.0,
      "grad_norm": 1.7683874936073016,
      "language_loss": 0.76809037,
      "learning_rate": 1.3987983435649508e-07,
      "loss": 0.78922927,
      "num_input_tokens_seen": 316956360,
      "step": 14700,
      "time_per_iteration": 2.510392189025879
    },
    {
      "auxiliary_loss_clip": 0.01070713,
      "auxiliary_loss_mlp": 0.01031756,
      "balance_loss_clip": 1.03537393,
      "balance_loss_mlp": 1.01981187,
      "epoch": 0.8838719374718172,
      "flos": 21470559459840.0,
      "grad_norm": 1.8713608697115733,
      "language_loss": 0.73475015,
      "learning_rate": 1.3973677849785494e-07,
      "loss": 0.75577486,
      "num_input_tokens_seen": 316975295,
      "step": 14701,
      "time_per_iteration": 2.560547113418579
    },
    {
      "auxiliary_loss_clip": 0.01082183,
      "auxiliary_loss_mlp": 0.01036166,
      "balance_loss_clip": 1.03360057,
      "balance_loss_mlp": 1.02280855,
      "epoch": 0.8839320607244852,
      "flos": 26468929555200.0,
      "grad_norm": 1.6880839193340886,
      "language_loss": 0.7074756,
      "learning_rate": 1.3959379318058262e-07,
      "loss": 0.72865915,
      "num_input_tokens_seen": 316994520,
      "step": 14702,
      "time_per_iteration": 2.5355043411254883
    },
    {
      "auxiliary_loss_clip": 0.01069417,
      "auxiliary_loss_mlp": 0.01038347,
      "balance_loss_clip": 1.03474832,
      "balance_loss_mlp": 1.02406013,
      "epoch": 0.8839921839771532,
      "flos": 45222270923520.0,
      "grad_norm": 2.4070004990481486,
      "language_loss": 0.71486962,
      "learning_rate": 1.3945087841010006e-07,
      "loss": 0.73594725,
      "num_input_tokens_seen": 317018095,
      "step": 14703,
      "time_per_iteration": 2.73290753364563
    },
    {
      "auxiliary_loss_clip": 0.0105952,
      "auxiliary_loss_mlp": 0.01028749,
      "balance_loss_clip": 1.03427386,
      "balance_loss_mlp": 1.01755488,
      "epoch": 0.8840523072298211,
      "flos": 20006876154240.0,
      "grad_norm": 1.9813796405231894,
      "language_loss": 0.66497993,
      "learning_rate": 1.3930803419182645e-07,
      "loss": 0.68586266,
      "num_input_tokens_seen": 317035755,
      "step": 14704,
      "time_per_iteration": 2.5469162464141846
    },
    {
      "auxiliary_loss_clip": 0.01088756,
      "auxiliary_loss_mlp": 0.01028088,
      "balance_loss_clip": 1.03319812,
      "balance_loss_mlp": 1.01700783,
      "epoch": 0.8841124304824891,
      "flos": 24426007528320.0,
      "grad_norm": 1.6530049053100373,
      "language_loss": 0.70511425,
      "learning_rate": 1.3916526053117905e-07,
      "loss": 0.72628272,
      "num_input_tokens_seen": 317055765,
      "step": 14705,
      "time_per_iteration": 2.5158543586730957
    },
    {
      "auxiliary_loss_clip": 0.01081677,
      "auxiliary_loss_mlp": 0.01032608,
      "balance_loss_clip": 1.03519666,
      "balance_loss_mlp": 1.02203393,
      "epoch": 0.884172553735157,
      "flos": 31284622056960.0,
      "grad_norm": 1.5245456515797116,
      "language_loss": 0.70800078,
      "learning_rate": 1.3902255743357104e-07,
      "loss": 0.72914362,
      "num_input_tokens_seen": 317077955,
      "step": 14706,
      "time_per_iteration": 2.580504894256592
    },
    {
      "auxiliary_loss_clip": 0.01090714,
      "auxiliary_loss_mlp": 0.01029254,
      "balance_loss_clip": 1.03302944,
      "balance_loss_mlp": 1.01761329,
      "epoch": 0.884232676987825,
      "flos": 21391160446080.0,
      "grad_norm": 1.5751884574546837,
      "language_loss": 0.74496198,
      "learning_rate": 1.3887992490441413e-07,
      "loss": 0.76616168,
      "num_input_tokens_seen": 317095825,
      "step": 14707,
      "time_per_iteration": 2.4810638427734375
    },
    {
      "auxiliary_loss_clip": 0.01003581,
      "auxiliary_loss_mlp": 0.01002276,
      "balance_loss_clip": 1.01050758,
      "balance_loss_mlp": 1.00123298,
      "epoch": 0.8842928002404931,
      "flos": 57911451799680.0,
      "grad_norm": 0.7929664546380933,
      "language_loss": 0.60385525,
      "learning_rate": 1.387373629491173e-07,
      "loss": 0.62391376,
      "num_input_tokens_seen": 317152875,
      "step": 14708,
      "time_per_iteration": 3.0063724517822266
    },
    {
      "auxiliary_loss_clip": 0.01074449,
      "auxiliary_loss_mlp": 0.01030003,
      "balance_loss_clip": 1.03361762,
      "balance_loss_mlp": 1.01904774,
      "epoch": 0.884352923493161,
      "flos": 41463896186880.0,
      "grad_norm": 1.5609034431832263,
      "language_loss": 0.67199349,
      "learning_rate": 1.3859487157308625e-07,
      "loss": 0.69303799,
      "num_input_tokens_seen": 317176725,
      "step": 14709,
      "time_per_iteration": 2.686643123626709
    },
    {
      "auxiliary_loss_clip": 0.0108399,
      "auxiliary_loss_mlp": 0.0103337,
      "balance_loss_clip": 1.03337991,
      "balance_loss_mlp": 1.02003038,
      "epoch": 0.884413046745829,
      "flos": 46541234332800.0,
      "grad_norm": 1.6047509032564182,
      "language_loss": 0.62668788,
      "learning_rate": 1.3845245078172373e-07,
      "loss": 0.64786148,
      "num_input_tokens_seen": 317206880,
      "step": 14710,
      "time_per_iteration": 2.7785916328430176
    },
    {
      "auxiliary_loss_clip": 0.01070988,
      "auxiliary_loss_mlp": 0.0102527,
      "balance_loss_clip": 1.03519857,
      "balance_loss_mlp": 1.01426673,
      "epoch": 0.8844731699984969,
      "flos": 19135324552320.0,
      "grad_norm": 2.805871844069436,
      "language_loss": 0.63856041,
      "learning_rate": 1.38310100580431e-07,
      "loss": 0.65952301,
      "num_input_tokens_seen": 317224135,
      "step": 14711,
      "time_per_iteration": 2.5117268562316895
    },
    {
      "auxiliary_loss_clip": 0.01068056,
      "auxiliary_loss_mlp": 0.01029887,
      "balance_loss_clip": 1.0343039,
      "balance_loss_mlp": 1.01763868,
      "epoch": 0.8845332932511649,
      "flos": 23260634674560.0,
      "grad_norm": 3.500120460647278,
      "language_loss": 0.75931317,
      "learning_rate": 1.38167820974606e-07,
      "loss": 0.78029263,
      "num_input_tokens_seen": 317244505,
      "step": 14712,
      "time_per_iteration": 2.571178674697876
    },
    {
      "auxiliary_loss_clip": 0.01043568,
      "auxiliary_loss_mlp": 0.01025085,
      "balance_loss_clip": 1.03068244,
      "balance_loss_mlp": 1.01309228,
      "epoch": 0.8845934165038328,
      "flos": 17564591738880.0,
      "grad_norm": 4.258699843232303,
      "language_loss": 0.8122263,
      "learning_rate": 1.3802561196964368e-07,
      "loss": 0.83291286,
      "num_input_tokens_seen": 317257830,
      "step": 14713,
      "time_per_iteration": 2.5560410022735596
    },
    {
      "auxiliary_loss_clip": 0.01080459,
      "auxiliary_loss_mlp": 0.01026986,
      "balance_loss_clip": 1.03424251,
      "balance_loss_mlp": 1.01516652,
      "epoch": 0.8846535397565009,
      "flos": 27485739757440.0,
      "grad_norm": 1.7661143094873863,
      "language_loss": 0.55565417,
      "learning_rate": 1.3788347357093688e-07,
      "loss": 0.57672864,
      "num_input_tokens_seen": 317278430,
      "step": 14714,
      "time_per_iteration": 2.6099441051483154
    },
    {
      "auxiliary_loss_clip": 0.01045826,
      "auxiliary_loss_mlp": 0.01033251,
      "balance_loss_clip": 1.03342009,
      "balance_loss_mlp": 1.02083516,
      "epoch": 0.8847136630091688,
      "flos": 28761430256640.0,
      "grad_norm": 1.7715062804136124,
      "language_loss": 0.73766029,
      "learning_rate": 1.377414057838755e-07,
      "loss": 0.7584511,
      "num_input_tokens_seen": 317295970,
      "step": 14715,
      "time_per_iteration": 2.6590845584869385
    },
    {
      "auxiliary_loss_clip": 0.01094881,
      "auxiliary_loss_mlp": 0.01028432,
      "balance_loss_clip": 1.03508914,
      "balance_loss_mlp": 1.01708961,
      "epoch": 0.8847737862618368,
      "flos": 23476924419840.0,
      "grad_norm": 1.6074226115223864,
      "language_loss": 0.75384289,
      "learning_rate": 1.375994086138461e-07,
      "loss": 0.77507603,
      "num_input_tokens_seen": 317316185,
      "step": 14716,
      "time_per_iteration": 3.9274466037750244
    },
    {
      "auxiliary_loss_clip": 0.01069841,
      "auxiliary_loss_mlp": 0.01037711,
      "balance_loss_clip": 1.03366399,
      "balance_loss_mlp": 1.02596283,
      "epoch": 0.8848339095145047,
      "flos": 18660872782080.0,
      "grad_norm": 2.430115384069665,
      "language_loss": 0.71286523,
      "learning_rate": 1.3745748206623397e-07,
      "loss": 0.73394072,
      "num_input_tokens_seen": 317333275,
      "step": 14717,
      "time_per_iteration": 3.963810920715332
    },
    {
      "auxiliary_loss_clip": 0.01089152,
      "auxiliary_loss_mlp": 0.01030438,
      "balance_loss_clip": 1.0345366,
      "balance_loss_mlp": 1.01931596,
      "epoch": 0.8848940327671727,
      "flos": 32270298145920.0,
      "grad_norm": 2.5416591630696184,
      "language_loss": 0.7456373,
      "learning_rate": 1.373156261464208e-07,
      "loss": 0.76683319,
      "num_input_tokens_seen": 317351245,
      "step": 14718,
      "time_per_iteration": 2.5589540004730225
    },
    {
      "auxiliary_loss_clip": 0.0105134,
      "auxiliary_loss_mlp": 0.01028743,
      "balance_loss_clip": 1.03538573,
      "balance_loss_mlp": 1.01702452,
      "epoch": 0.8849541560198406,
      "flos": 24021832717440.0,
      "grad_norm": 1.6203747585430244,
      "language_loss": 0.78749609,
      "learning_rate": 1.3717384085978602e-07,
      "loss": 0.80829692,
      "num_input_tokens_seen": 317370740,
      "step": 14719,
      "time_per_iteration": 4.063754558563232
    },
    {
      "auxiliary_loss_clip": 0.01104162,
      "auxiliary_loss_mlp": 0.01028129,
      "balance_loss_clip": 1.03464103,
      "balance_loss_mlp": 1.01595747,
      "epoch": 0.8850142792725086,
      "flos": 16873060124160.0,
      "grad_norm": 1.850391752118759,
      "language_loss": 0.72303748,
      "learning_rate": 1.3703212621170579e-07,
      "loss": 0.74436039,
      "num_input_tokens_seen": 317388370,
      "step": 14720,
      "time_per_iteration": 2.4463114738464355
    },
    {
      "auxiliary_loss_clip": 0.01084453,
      "auxiliary_loss_mlp": 0.01028696,
      "balance_loss_clip": 1.0348649,
      "balance_loss_mlp": 1.01597071,
      "epoch": 0.8850744025251767,
      "flos": 24024059360640.0,
      "grad_norm": 2.3471835823788223,
      "language_loss": 0.82592547,
      "learning_rate": 1.3689048220755383e-07,
      "loss": 0.84705698,
      "num_input_tokens_seen": 317407390,
      "step": 14721,
      "time_per_iteration": 2.5212619304656982
    },
    {
      "auxiliary_loss_clip": 0.01081053,
      "auxiliary_loss_mlp": 0.01030067,
      "balance_loss_clip": 1.03179312,
      "balance_loss_mlp": 1.01726413,
      "epoch": 0.8851345257778446,
      "flos": 47955575329920.0,
      "grad_norm": 1.6427256260382548,
      "language_loss": 0.61951602,
      "learning_rate": 1.3674890885270186e-07,
      "loss": 0.64062721,
      "num_input_tokens_seen": 317430825,
      "step": 14722,
      "time_per_iteration": 2.7698121070861816
    },
    {
      "auxiliary_loss_clip": 0.01093937,
      "auxiliary_loss_mlp": 0.01029352,
      "balance_loss_clip": 1.03440809,
      "balance_loss_mlp": 1.01677537,
      "epoch": 0.8851946490305126,
      "flos": 36611000173440.0,
      "grad_norm": 2.2704666802211113,
      "language_loss": 0.69002271,
      "learning_rate": 1.3660740615251754e-07,
      "loss": 0.71125567,
      "num_input_tokens_seen": 317451905,
      "step": 14723,
      "time_per_iteration": 2.5940165519714355
    },
    {
      "auxiliary_loss_clip": 0.01069902,
      "auxiliary_loss_mlp": 0.01034337,
      "balance_loss_clip": 1.03331959,
      "balance_loss_mlp": 1.02195752,
      "epoch": 0.8852547722831805,
      "flos": 21544248211200.0,
      "grad_norm": 1.8800517641738776,
      "language_loss": 0.7790029,
      "learning_rate": 1.3646597411236703e-07,
      "loss": 0.80004525,
      "num_input_tokens_seen": 317470030,
      "step": 14724,
      "time_per_iteration": 2.56684947013855
    },
    {
      "auxiliary_loss_clip": 0.01018027,
      "auxiliary_loss_mlp": 0.01000593,
      "balance_loss_clip": 1.00530434,
      "balance_loss_mlp": 0.99947226,
      "epoch": 0.8853148955358485,
      "flos": 63059246472960.0,
      "grad_norm": 0.8003296917487726,
      "language_loss": 0.58938527,
      "learning_rate": 1.363246127376143e-07,
      "loss": 0.60957152,
      "num_input_tokens_seen": 317527460,
      "step": 14725,
      "time_per_iteration": 4.322714328765869
    },
    {
      "auxiliary_loss_clip": 0.01082871,
      "auxiliary_loss_mlp": 0.00783832,
      "balance_loss_clip": 1.03358173,
      "balance_loss_mlp": 1.00856876,
      "epoch": 0.8853750187885164,
      "flos": 18149828031360.0,
      "grad_norm": 2.1349725419992946,
      "language_loss": 0.68712574,
      "learning_rate": 1.3618332203361837e-07,
      "loss": 0.70579278,
      "num_input_tokens_seen": 317544070,
      "step": 14726,
      "time_per_iteration": 2.5208945274353027
    },
    {
      "auxiliary_loss_clip": 0.01090477,
      "auxiliary_loss_mlp": 0.00783266,
      "balance_loss_clip": 1.03449631,
      "balance_loss_mlp": 1.00910902,
      "epoch": 0.8854351420411845,
      "flos": 39570542392320.0,
      "grad_norm": 1.2005619166472559,
      "language_loss": 0.69271207,
      "learning_rate": 1.3604210200573785e-07,
      "loss": 0.7114495,
      "num_input_tokens_seen": 317570275,
      "step": 14727,
      "time_per_iteration": 2.692862033843994
    },
    {
      "auxiliary_loss_clip": 0.01084471,
      "auxiliary_loss_mlp": 0.0103045,
      "balance_loss_clip": 1.03864884,
      "balance_loss_mlp": 1.01828444,
      "epoch": 0.8854952652938524,
      "flos": 23769309127680.0,
      "grad_norm": 1.6319564950930607,
      "language_loss": 0.70072114,
      "learning_rate": 1.3590095265932733e-07,
      "loss": 0.72187042,
      "num_input_tokens_seen": 317590160,
      "step": 14728,
      "time_per_iteration": 2.5229618549346924
    },
    {
      "auxiliary_loss_clip": 0.0107137,
      "auxiliary_loss_mlp": 0.0102853,
      "balance_loss_clip": 1.03529286,
      "balance_loss_mlp": 1.01752687,
      "epoch": 0.8855553885465204,
      "flos": 18290310122880.0,
      "grad_norm": 2.0192388038660916,
      "language_loss": 0.6661098,
      "learning_rate": 1.3575987399973987e-07,
      "loss": 0.68710887,
      "num_input_tokens_seen": 317608340,
      "step": 14729,
      "time_per_iteration": 2.5396616458892822
    },
    {
      "auxiliary_loss_clip": 0.01084034,
      "auxiliary_loss_mlp": 0.01030442,
      "balance_loss_clip": 1.03839648,
      "balance_loss_mlp": 1.01963019,
      "epoch": 0.8856155117991883,
      "flos": 36867402432000.0,
      "grad_norm": 1.7309492565671563,
      "language_loss": 0.62881577,
      "learning_rate": 1.3561886603232453e-07,
      "loss": 0.64996058,
      "num_input_tokens_seen": 317629910,
      "step": 14730,
      "time_per_iteration": 2.6425654888153076
    },
    {
      "auxiliary_loss_clip": 0.0106568,
      "auxiliary_loss_mlp": 0.01028926,
      "balance_loss_clip": 1.03361201,
      "balance_loss_mlp": 1.01724911,
      "epoch": 0.8856756350518563,
      "flos": 22163886754560.0,
      "grad_norm": 1.515073822313099,
      "language_loss": 0.7949096,
      "learning_rate": 1.3547792876242904e-07,
      "loss": 0.81585562,
      "num_input_tokens_seen": 317650265,
      "step": 14731,
      "time_per_iteration": 2.5953400135040283
    },
    {
      "auxiliary_loss_clip": 0.01068989,
      "auxiliary_loss_mlp": 0.01032295,
      "balance_loss_clip": 1.03192234,
      "balance_loss_mlp": 1.02032614,
      "epoch": 0.8857357583045242,
      "flos": 20740962407040.0,
      "grad_norm": 1.9508176320846629,
      "language_loss": 0.83139884,
      "learning_rate": 1.3533706219539708e-07,
      "loss": 0.85241175,
      "num_input_tokens_seen": 317669045,
      "step": 14732,
      "time_per_iteration": 2.5527169704437256
    },
    {
      "auxiliary_loss_clip": 0.01009973,
      "auxiliary_loss_mlp": 0.01001079,
      "balance_loss_clip": 1.00782883,
      "balance_loss_mlp": 0.9999944,
      "epoch": 0.8857958815571922,
      "flos": 69892329409920.0,
      "grad_norm": 0.8989018659668119,
      "language_loss": 0.59919125,
      "learning_rate": 1.3519626633657045e-07,
      "loss": 0.61930174,
      "num_input_tokens_seen": 317728065,
      "step": 14733,
      "time_per_iteration": 3.158027172088623
    },
    {
      "auxiliary_loss_clip": 0.01105004,
      "auxiliary_loss_mlp": 0.00784363,
      "balance_loss_clip": 1.03618312,
      "balance_loss_mlp": 1.0097568,
      "epoch": 0.8858560048098603,
      "flos": 15121948187520.0,
      "grad_norm": 1.721681024884068,
      "language_loss": 0.66242415,
      "learning_rate": 1.3505554119128838e-07,
      "loss": 0.68131781,
      "num_input_tokens_seen": 317746120,
      "step": 14734,
      "time_per_iteration": 2.452162981033325
    },
    {
      "auxiliary_loss_clip": 0.01077546,
      "auxiliary_loss_mlp": 0.01032722,
      "balance_loss_clip": 1.03790116,
      "balance_loss_mlp": 1.02183807,
      "epoch": 0.8859161280625282,
      "flos": 16611019430400.0,
      "grad_norm": 1.8774719955246029,
      "language_loss": 0.75466239,
      "learning_rate": 1.3491488676488682e-07,
      "loss": 0.77576506,
      "num_input_tokens_seen": 317762280,
      "step": 14735,
      "time_per_iteration": 2.514533758163452
    },
    {
      "auxiliary_loss_clip": 0.01066789,
      "auxiliary_loss_mlp": 0.01032495,
      "balance_loss_clip": 1.03359962,
      "balance_loss_mlp": 1.02015686,
      "epoch": 0.8859762513151962,
      "flos": 18694484933760.0,
      "grad_norm": 1.7445060601385467,
      "language_loss": 0.70348167,
      "learning_rate": 1.3477430306270066e-07,
      "loss": 0.72447455,
      "num_input_tokens_seen": 317780615,
      "step": 14736,
      "time_per_iteration": 2.519740104675293
    },
    {
      "auxiliary_loss_clip": 0.01079927,
      "auxiliary_loss_mlp": 0.01027771,
      "balance_loss_clip": 1.03726709,
      "balance_loss_mlp": 1.0157665,
      "epoch": 0.8860363745678641,
      "flos": 19536877670400.0,
      "grad_norm": 1.6762464206381906,
      "language_loss": 0.84377038,
      "learning_rate": 1.3463379009005892e-07,
      "loss": 0.86484736,
      "num_input_tokens_seen": 317798830,
      "step": 14737,
      "time_per_iteration": 2.5320472717285156
    },
    {
      "auxiliary_loss_clip": 0.01079711,
      "auxiliary_loss_mlp": 0.01033277,
      "balance_loss_clip": 1.03519726,
      "balance_loss_mlp": 1.01992536,
      "epoch": 0.8860964978205321,
      "flos": 35954912304000.0,
      "grad_norm": 2.054614254820155,
      "language_loss": 0.68360722,
      "learning_rate": 1.3449334785229093e-07,
      "loss": 0.70473707,
      "num_input_tokens_seen": 317819235,
      "step": 14738,
      "time_per_iteration": 2.644205331802368
    },
    {
      "auxiliary_loss_clip": 0.0109574,
      "auxiliary_loss_mlp": 0.01030021,
      "balance_loss_clip": 1.03489685,
      "balance_loss_mlp": 1.01736069,
      "epoch": 0.8861566210732,
      "flos": 21212577002880.0,
      "grad_norm": 1.582178962390456,
      "language_loss": 0.74981225,
      "learning_rate": 1.343529763547222e-07,
      "loss": 0.77106988,
      "num_input_tokens_seen": 317836785,
      "step": 14739,
      "time_per_iteration": 2.4910991191864014
    },
    {
      "auxiliary_loss_clip": 0.0109109,
      "auxiliary_loss_mlp": 0.01034205,
      "balance_loss_clip": 1.03530526,
      "balance_loss_mlp": 1.02311277,
      "epoch": 0.886216744325868,
      "flos": 14609071843200.0,
      "grad_norm": 1.8054401356432992,
      "language_loss": 0.87166172,
      "learning_rate": 1.3421267560267559e-07,
      "loss": 0.89291465,
      "num_input_tokens_seen": 317854225,
      "step": 14740,
      "time_per_iteration": 2.4569904804229736
    },
    {
      "auxiliary_loss_clip": 0.01057717,
      "auxiliary_loss_mlp": 0.0103241,
      "balance_loss_clip": 1.03548276,
      "balance_loss_mlp": 1.02041721,
      "epoch": 0.886276867578536,
      "flos": 26651643062400.0,
      "grad_norm": 1.7017012765090198,
      "language_loss": 0.63412178,
      "learning_rate": 1.34072445601471e-07,
      "loss": 0.65502298,
      "num_input_tokens_seen": 317874865,
      "step": 14741,
      "time_per_iteration": 2.6319611072540283
    },
    {
      "auxiliary_loss_clip": 0.01103509,
      "auxiliary_loss_mlp": 0.01029948,
      "balance_loss_clip": 1.03541684,
      "balance_loss_mlp": 1.01817024,
      "epoch": 0.886336990831204,
      "flos": 16764071281920.0,
      "grad_norm": 2.1822745287557805,
      "language_loss": 0.72925043,
      "learning_rate": 1.3393228635642717e-07,
      "loss": 0.75058496,
      "num_input_tokens_seen": 317892830,
      "step": 14742,
      "time_per_iteration": 2.422395944595337
    },
    {
      "auxiliary_loss_clip": 0.01090956,
      "auxiliary_loss_mlp": 0.00782266,
      "balance_loss_clip": 1.03425586,
      "balance_loss_mlp": 1.00872898,
      "epoch": 0.8863971140838719,
      "flos": 25265275781760.0,
      "grad_norm": 1.7376554270139353,
      "language_loss": 0.59227335,
      "learning_rate": 1.3379219787285733e-07,
      "loss": 0.61100554,
      "num_input_tokens_seen": 317911780,
      "step": 14743,
      "time_per_iteration": 2.5296425819396973
    },
    {
      "auxiliary_loss_clip": 0.01074178,
      "auxiliary_loss_mlp": 0.01032906,
      "balance_loss_clip": 1.03521729,
      "balance_loss_mlp": 1.01938164,
      "epoch": 0.8864572373365399,
      "flos": 23404313076480.0,
      "grad_norm": 1.709770331678158,
      "language_loss": 0.60264146,
      "learning_rate": 1.3365218015607437e-07,
      "loss": 0.6237123,
      "num_input_tokens_seen": 317932855,
      "step": 14744,
      "time_per_iteration": 2.5690457820892334
    },
    {
      "auxiliary_loss_clip": 0.01095104,
      "auxiliary_loss_mlp": 0.00781752,
      "balance_loss_clip": 1.0362339,
      "balance_loss_mlp": 1.00569272,
      "epoch": 0.8865173605892078,
      "flos": 18548759456640.0,
      "grad_norm": 1.6060460276248727,
      "language_loss": 0.76577759,
      "learning_rate": 1.3351223321138762e-07,
      "loss": 0.78454614,
      "num_input_tokens_seen": 317952090,
      "step": 14745,
      "time_per_iteration": 2.5019371509552
    },
    {
      "auxiliary_loss_clip": 0.01102908,
      "auxiliary_loss_mlp": 0.00782088,
      "balance_loss_clip": 1.03538525,
      "balance_loss_mlp": 1.00735438,
      "epoch": 0.8865774838418758,
      "flos": 19025868833280.0,
      "grad_norm": 1.6545822118548315,
      "language_loss": 0.77204478,
      "learning_rate": 1.3337235704410454e-07,
      "loss": 0.79089475,
      "num_input_tokens_seen": 317970370,
      "step": 14746,
      "time_per_iteration": 2.444553852081299
    },
    {
      "auxiliary_loss_clip": 0.01080786,
      "auxiliary_loss_mlp": 0.01032274,
      "balance_loss_clip": 1.03704453,
      "balance_loss_mlp": 1.01943469,
      "epoch": 0.8866376070945439,
      "flos": 22163168482560.0,
      "grad_norm": 1.8208924385232765,
      "language_loss": 0.76624328,
      "learning_rate": 1.3323255165952873e-07,
      "loss": 0.7873739,
      "num_input_tokens_seen": 317989125,
      "step": 14747,
      "time_per_iteration": 2.5524051189422607
    },
    {
      "auxiliary_loss_clip": 0.01076886,
      "auxiliary_loss_mlp": 0.00781235,
      "balance_loss_clip": 1.03129816,
      "balance_loss_mlp": 1.00676525,
      "epoch": 0.8866977303472118,
      "flos": 20704261685760.0,
      "grad_norm": 1.7591568711909982,
      "language_loss": 0.82579404,
      "learning_rate": 1.3309281706296127e-07,
      "loss": 0.84437525,
      "num_input_tokens_seen": 318007820,
      "step": 14748,
      "time_per_iteration": 2.5114691257476807
    },
    {
      "auxiliary_loss_clip": 0.01092954,
      "auxiliary_loss_mlp": 0.01032586,
      "balance_loss_clip": 1.03504515,
      "balance_loss_mlp": 1.02012897,
      "epoch": 0.8867578535998798,
      "flos": 48794448533760.0,
      "grad_norm": 1.8081299110442413,
      "language_loss": 0.77563143,
      "learning_rate": 1.3295315325970148e-07,
      "loss": 0.79688674,
      "num_input_tokens_seen": 318030435,
      "step": 14749,
      "time_per_iteration": 2.7221243381500244
    },
    {
      "auxiliary_loss_clip": 0.01044386,
      "auxiliary_loss_mlp": 0.0078398,
      "balance_loss_clip": 1.03371394,
      "balance_loss_mlp": 1.00994527,
      "epoch": 0.8868179768525477,
      "flos": 21105312013440.0,
      "grad_norm": 2.3841849365546515,
      "language_loss": 0.69656599,
      "learning_rate": 1.328135602550451e-07,
      "loss": 0.71484965,
      "num_input_tokens_seen": 318049465,
      "step": 14750,
      "time_per_iteration": 2.611816167831421
    },
    {
      "auxiliary_loss_clip": 0.01092783,
      "auxiliary_loss_mlp": 0.01030647,
      "balance_loss_clip": 1.03492546,
      "balance_loss_mlp": 1.01880372,
      "epoch": 0.8868781001052157,
      "flos": 21830922656640.0,
      "grad_norm": 1.7794370063328793,
      "language_loss": 0.59201682,
      "learning_rate": 1.3267403805428546e-07,
      "loss": 0.61325115,
      "num_input_tokens_seen": 318067760,
      "step": 14751,
      "time_per_iteration": 2.524425983428955
    },
    {
      "auxiliary_loss_clip": 0.01104285,
      "auxiliary_loss_mlp": 0.01031014,
      "balance_loss_clip": 1.03621113,
      "balance_loss_mlp": 1.01889586,
      "epoch": 0.8869382233578836,
      "flos": 13516418073600.0,
      "grad_norm": 2.47668543849538,
      "language_loss": 0.81331301,
      "learning_rate": 1.3253458666271344e-07,
      "loss": 0.83466601,
      "num_input_tokens_seen": 318082785,
      "step": 14752,
      "time_per_iteration": 2.411102533340454
    },
    {
      "auxiliary_loss_clip": 0.0108565,
      "auxiliary_loss_mlp": 0.01031223,
      "balance_loss_clip": 1.03646433,
      "balance_loss_mlp": 1.01812816,
      "epoch": 0.8869983466105517,
      "flos": 22704988210560.0,
      "grad_norm": 2.3816892259347147,
      "language_loss": 0.80301303,
      "learning_rate": 1.3239520608561793e-07,
      "loss": 0.8241818,
      "num_input_tokens_seen": 318101925,
      "step": 14753,
      "time_per_iteration": 2.533090353012085
    },
    {
      "auxiliary_loss_clip": 0.01102647,
      "auxiliary_loss_mlp": 0.01029452,
      "balance_loss_clip": 1.03500199,
      "balance_loss_mlp": 1.01782322,
      "epoch": 0.8870584698632196,
      "flos": 15340751884800.0,
      "grad_norm": 1.6793872582519507,
      "language_loss": 0.65482378,
      "learning_rate": 1.3225589632828248e-07,
      "loss": 0.67614472,
      "num_input_tokens_seen": 318119945,
      "step": 14754,
      "time_per_iteration": 3.813854932785034
    },
    {
      "auxiliary_loss_clip": 0.01106389,
      "auxiliary_loss_mlp": 0.01031451,
      "balance_loss_clip": 1.03682303,
      "balance_loss_mlp": 1.01916599,
      "epoch": 0.8871185931158876,
      "flos": 26615624699520.0,
      "grad_norm": 1.9030618858297426,
      "language_loss": 0.74576867,
      "learning_rate": 1.3211665739599065e-07,
      "loss": 0.76714706,
      "num_input_tokens_seen": 318139685,
      "step": 14755,
      "time_per_iteration": 3.875464677810669
    },
    {
      "auxiliary_loss_clip": 0.0107939,
      "auxiliary_loss_mlp": 0.01032602,
      "balance_loss_clip": 1.03088272,
      "balance_loss_mlp": 1.01941156,
      "epoch": 0.8871787163685555,
      "flos": 21799034357760.0,
      "grad_norm": 1.474849311144646,
      "language_loss": 0.77886963,
      "learning_rate": 1.3197748929402262e-07,
      "loss": 0.79998958,
      "num_input_tokens_seen": 318160375,
      "step": 14756,
      "time_per_iteration": 2.5308098793029785
    },
    {
      "auxiliary_loss_clip": 0.01084752,
      "auxiliary_loss_mlp": 0.01032605,
      "balance_loss_clip": 1.03604031,
      "balance_loss_mlp": 1.02067757,
      "epoch": 0.8872388396212235,
      "flos": 14902964922240.0,
      "grad_norm": 1.8798215198700294,
      "language_loss": 0.76611865,
      "learning_rate": 1.3183839202765535e-07,
      "loss": 0.78729224,
      "num_input_tokens_seen": 318177995,
      "step": 14757,
      "time_per_iteration": 2.4772582054138184
    },
    {
      "auxiliary_loss_clip": 0.01044629,
      "auxiliary_loss_mlp": 0.01031253,
      "balance_loss_clip": 1.03207016,
      "balance_loss_mlp": 1.01939201,
      "epoch": 0.8872989628738914,
      "flos": 26432157006720.0,
      "grad_norm": 1.728699522006643,
      "language_loss": 0.67757869,
      "learning_rate": 1.316993656021632e-07,
      "loss": 0.6983375,
      "num_input_tokens_seen": 318197030,
      "step": 14758,
      "time_per_iteration": 4.030256032943726
    },
    {
      "auxiliary_loss_clip": 0.0110399,
      "auxiliary_loss_mlp": 0.01033423,
      "balance_loss_clip": 1.03540373,
      "balance_loss_mlp": 1.02042913,
      "epoch": 0.8873590861265594,
      "flos": 48142562555520.0,
      "grad_norm": 1.5811818165131137,
      "language_loss": 0.68864012,
      "learning_rate": 1.3156041002281915e-07,
      "loss": 0.71001422,
      "num_input_tokens_seen": 318221780,
      "step": 14759,
      "time_per_iteration": 2.690025568008423
    },
    {
      "auxiliary_loss_clip": 0.01101422,
      "auxiliary_loss_mlp": 0.01029218,
      "balance_loss_clip": 1.03390205,
      "balance_loss_mlp": 1.0176723,
      "epoch": 0.8874192093792275,
      "flos": 18332972501760.0,
      "grad_norm": 2.7193815919078435,
      "language_loss": 0.74549448,
      "learning_rate": 1.3142152529489092e-07,
      "loss": 0.76680088,
      "num_input_tokens_seen": 318239710,
      "step": 14760,
      "time_per_iteration": 2.425956964492798
    },
    {
      "auxiliary_loss_clip": 0.01083878,
      "auxiliary_loss_mlp": 0.01030601,
      "balance_loss_clip": 1.03521121,
      "balance_loss_mlp": 1.01878738,
      "epoch": 0.8874793326318954,
      "flos": 17894215872000.0,
      "grad_norm": 2.1119417143918184,
      "language_loss": 0.76600879,
      "learning_rate": 1.3128271142364565e-07,
      "loss": 0.7871536,
      "num_input_tokens_seen": 318257425,
      "step": 14761,
      "time_per_iteration": 2.4882891178131104
    },
    {
      "auxiliary_loss_clip": 0.01104126,
      "auxiliary_loss_mlp": 0.01034316,
      "balance_loss_clip": 1.0340451,
      "balance_loss_mlp": 1.02228212,
      "epoch": 0.8875394558845634,
      "flos": 31102231772160.0,
      "grad_norm": 1.5545149534964293,
      "language_loss": 0.61009598,
      "learning_rate": 1.3114396841434717e-07,
      "loss": 0.6314804,
      "num_input_tokens_seen": 318278485,
      "step": 14762,
      "time_per_iteration": 2.5407497882843018
    },
    {
      "auxiliary_loss_clip": 0.01089279,
      "auxiliary_loss_mlp": 0.01031926,
      "balance_loss_clip": 1.03346372,
      "balance_loss_mlp": 1.01941442,
      "epoch": 0.8875995791372313,
      "flos": 21142048648320.0,
      "grad_norm": 1.8292620808658695,
      "language_loss": 0.64460361,
      "learning_rate": 1.3100529627225697e-07,
      "loss": 0.66581559,
      "num_input_tokens_seen": 318297560,
      "step": 14763,
      "time_per_iteration": 2.5466930866241455
    },
    {
      "auxiliary_loss_clip": 0.01076362,
      "auxiliary_loss_mlp": 0.00783565,
      "balance_loss_clip": 1.03450727,
      "balance_loss_mlp": 1.00765848,
      "epoch": 0.8876597023898993,
      "flos": 17455136019840.0,
      "grad_norm": 2.10215323663168,
      "language_loss": 0.71066827,
      "learning_rate": 1.3086669500263335e-07,
      "loss": 0.72926748,
      "num_input_tokens_seen": 318313060,
      "step": 14764,
      "time_per_iteration": 3.8846302032470703
    },
    {
      "auxiliary_loss_clip": 0.01107413,
      "auxiliary_loss_mlp": 0.01034863,
      "balance_loss_clip": 1.03543496,
      "balance_loss_mlp": 1.02285266,
      "epoch": 0.8877198256425672,
      "flos": 22707933125760.0,
      "grad_norm": 2.121813034613621,
      "language_loss": 0.65924615,
      "learning_rate": 1.3072816461073166e-07,
      "loss": 0.68066895,
      "num_input_tokens_seen": 318332030,
      "step": 14765,
      "time_per_iteration": 2.45858097076416
    },
    {
      "auxiliary_loss_clip": 0.01063526,
      "auxiliary_loss_mlp": 0.01027082,
      "balance_loss_clip": 1.03469133,
      "balance_loss_mlp": 1.01621604,
      "epoch": 0.8877799488952353,
      "flos": 24535104111360.0,
      "grad_norm": 1.5766073722033274,
      "language_loss": 0.76549357,
      "learning_rate": 1.3058970510180568e-07,
      "loss": 0.7863996,
      "num_input_tokens_seen": 318351090,
      "step": 14766,
      "time_per_iteration": 2.605849504470825
    },
    {
      "auxiliary_loss_clip": 0.01072024,
      "auxiliary_loss_mlp": 0.01029606,
      "balance_loss_clip": 1.03188753,
      "balance_loss_mlp": 1.01804912,
      "epoch": 0.8878400721479032,
      "flos": 20959191486720.0,
      "grad_norm": 2.8834814202177954,
      "language_loss": 0.72847444,
      "learning_rate": 1.3045131648110496e-07,
      "loss": 0.74949074,
      "num_input_tokens_seen": 318372000,
      "step": 14767,
      "time_per_iteration": 2.5258867740631104
    },
    {
      "auxiliary_loss_clip": 0.01099604,
      "auxiliary_loss_mlp": 0.01027655,
      "balance_loss_clip": 1.03418326,
      "balance_loss_mlp": 1.01657486,
      "epoch": 0.8879001954005712,
      "flos": 25295260659840.0,
      "grad_norm": 1.7783967157743072,
      "language_loss": 0.70945758,
      "learning_rate": 1.303129987538778e-07,
      "loss": 0.73073018,
      "num_input_tokens_seen": 318391530,
      "step": 14768,
      "time_per_iteration": 2.500744581222534
    },
    {
      "auxiliary_loss_clip": 0.01090018,
      "auxiliary_loss_mlp": 0.0103171,
      "balance_loss_clip": 1.03513062,
      "balance_loss_mlp": 1.02019429,
      "epoch": 0.8879603186532391,
      "flos": 23185329811200.0,
      "grad_norm": 1.9982355560255454,
      "language_loss": 0.70016217,
      "learning_rate": 1.3017475192536932e-07,
      "loss": 0.7213794,
      "num_input_tokens_seen": 318410690,
      "step": 14769,
      "time_per_iteration": 2.503359317779541
    },
    {
      "auxiliary_loss_clip": 0.01077587,
      "auxiliary_loss_mlp": 0.0102958,
      "balance_loss_clip": 1.03518736,
      "balance_loss_mlp": 1.0182972,
      "epoch": 0.8880204419059071,
      "flos": 13655427707520.0,
      "grad_norm": 2.247767717259084,
      "language_loss": 0.66987288,
      "learning_rate": 1.3003657600082174e-07,
      "loss": 0.69094455,
      "num_input_tokens_seen": 318427380,
      "step": 14770,
      "time_per_iteration": 2.514626979827881
    },
    {
      "auxiliary_loss_clip": 0.01088453,
      "auxiliary_loss_mlp": 0.01033329,
      "balance_loss_clip": 1.03415215,
      "balance_loss_mlp": 1.02109182,
      "epoch": 0.888080565158575,
      "flos": 20631865824000.0,
      "grad_norm": 1.7413124356404244,
      "language_loss": 0.65281296,
      "learning_rate": 1.2989847098547424e-07,
      "loss": 0.67403078,
      "num_input_tokens_seen": 318448530,
      "step": 14771,
      "time_per_iteration": 2.5058789253234863
    },
    {
      "auxiliary_loss_clip": 0.0108045,
      "auxiliary_loss_mlp": 0.01025,
      "balance_loss_clip": 1.03417587,
      "balance_loss_mlp": 1.01354432,
      "epoch": 0.888140688411243,
      "flos": 28620014411520.0,
      "grad_norm": 1.4387291899530263,
      "language_loss": 0.82478821,
      "learning_rate": 1.2976043688456396e-07,
      "loss": 0.84584278,
      "num_input_tokens_seen": 318468655,
      "step": 14772,
      "time_per_iteration": 2.5801050662994385
    },
    {
      "auxiliary_loss_clip": 0.01076356,
      "auxiliary_loss_mlp": 0.01024441,
      "balance_loss_clip": 1.03207076,
      "balance_loss_mlp": 1.01362324,
      "epoch": 0.8882008116639111,
      "flos": 25520241496320.0,
      "grad_norm": 1.9418214720670446,
      "language_loss": 0.76508504,
      "learning_rate": 1.296224737033258e-07,
      "loss": 0.78609306,
      "num_input_tokens_seen": 318488740,
      "step": 14773,
      "time_per_iteration": 2.5675265789031982
    },
    {
      "auxiliary_loss_clip": 0.01079979,
      "auxiliary_loss_mlp": 0.01025352,
      "balance_loss_clip": 1.03463447,
      "balance_loss_mlp": 1.01413405,
      "epoch": 0.888260934916579,
      "flos": 27673696650240.0,
      "grad_norm": 1.6787941191927467,
      "language_loss": 0.74947166,
      "learning_rate": 1.294845814469907e-07,
      "loss": 0.77052498,
      "num_input_tokens_seen": 318508810,
      "step": 14774,
      "time_per_iteration": 2.569411039352417
    },
    {
      "auxiliary_loss_clip": 0.01057911,
      "auxiliary_loss_mlp": 0.00782558,
      "balance_loss_clip": 1.03535175,
      "balance_loss_mlp": 1.00685859,
      "epoch": 0.888321058169247,
      "flos": 21611077464960.0,
      "grad_norm": 2.2262384520349676,
      "language_loss": 0.71985286,
      "learning_rate": 1.2934676012078783e-07,
      "loss": 0.73825753,
      "num_input_tokens_seen": 318526860,
      "step": 14775,
      "time_per_iteration": 2.556896209716797
    },
    {
      "auxiliary_loss_clip": 0.01101961,
      "auxiliary_loss_mlp": 0.01028833,
      "balance_loss_clip": 1.03405714,
      "balance_loss_mlp": 1.01747251,
      "epoch": 0.8883811814219149,
      "flos": 18149109759360.0,
      "grad_norm": 1.536238641645253,
      "language_loss": 0.80110335,
      "learning_rate": 1.292090097299432e-07,
      "loss": 0.8224113,
      "num_input_tokens_seen": 318545180,
      "step": 14776,
      "time_per_iteration": 2.433401584625244
    },
    {
      "auxiliary_loss_clip": 0.01098021,
      "auxiliary_loss_mlp": 0.01031625,
      "balance_loss_clip": 1.03468013,
      "balance_loss_mlp": 1.01923943,
      "epoch": 0.8884413046745829,
      "flos": 28324648874880.0,
      "grad_norm": 2.306179104985734,
      "language_loss": 0.69289291,
      "learning_rate": 1.290713302796802e-07,
      "loss": 0.71418941,
      "num_input_tokens_seen": 318564350,
      "step": 14777,
      "time_per_iteration": 2.5101089477539062
    },
    {
      "auxiliary_loss_clip": 0.01087489,
      "auxiliary_loss_mlp": 0.01035777,
      "balance_loss_clip": 1.03151846,
      "balance_loss_mlp": 1.02376699,
      "epoch": 0.8885014279272508,
      "flos": 15158756649600.0,
      "grad_norm": 2.150791245760462,
      "language_loss": 0.70917165,
      "learning_rate": 1.2893372177522e-07,
      "loss": 0.73040426,
      "num_input_tokens_seen": 318582275,
      "step": 14778,
      "time_per_iteration": 2.4723336696624756
    },
    {
      "auxiliary_loss_clip": 0.0110359,
      "auxiliary_loss_mlp": 0.01029513,
      "balance_loss_clip": 1.03485346,
      "balance_loss_mlp": 1.01846266,
      "epoch": 0.8885615511799189,
      "flos": 19099593498240.0,
      "grad_norm": 1.6362460193717518,
      "language_loss": 0.77413034,
      "learning_rate": 1.287961842217804e-07,
      "loss": 0.79546136,
      "num_input_tokens_seen": 318601230,
      "step": 14779,
      "time_per_iteration": 2.428987741470337
    },
    {
      "auxiliary_loss_clip": 0.01008808,
      "auxiliary_loss_mlp": 0.01003652,
      "balance_loss_clip": 1.01062799,
      "balance_loss_mlp": 1.00254309,
      "epoch": 0.8886216744325868,
      "flos": 51186567605760.0,
      "grad_norm": 0.8719873376003487,
      "language_loss": 0.56821805,
      "learning_rate": 1.2865871762457747e-07,
      "loss": 0.58834267,
      "num_input_tokens_seen": 318645595,
      "step": 14780,
      "time_per_iteration": 2.9266011714935303
    },
    {
      "auxiliary_loss_clip": 0.01027864,
      "auxiliary_loss_mlp": 0.01001232,
      "balance_loss_clip": 1.00507498,
      "balance_loss_mlp": 1.00023103,
      "epoch": 0.8886817976852548,
      "flos": 61612981263360.0,
      "grad_norm": 0.7916369161470943,
      "language_loss": 0.62447226,
      "learning_rate": 1.2852132198882326e-07,
      "loss": 0.64476323,
      "num_input_tokens_seen": 318707850,
      "step": 14781,
      "time_per_iteration": 3.1335248947143555
    },
    {
      "auxiliary_loss_clip": 0.00963276,
      "auxiliary_loss_mlp": 0.01001523,
      "balance_loss_clip": 1.01312089,
      "balance_loss_mlp": 1.00039089,
      "epoch": 0.8887419209379227,
      "flos": 60646946935680.0,
      "grad_norm": 0.7966364611463298,
      "language_loss": 0.58177191,
      "learning_rate": 1.2838399731972805e-07,
      "loss": 0.60141987,
      "num_input_tokens_seen": 318764915,
      "step": 14782,
      "time_per_iteration": 3.4004065990448
    },
    {
      "auxiliary_loss_clip": 0.0110225,
      "auxiliary_loss_mlp": 0.01028507,
      "balance_loss_clip": 1.03579354,
      "balance_loss_mlp": 1.01746237,
      "epoch": 0.8888020441905907,
      "flos": 29205861235200.0,
      "grad_norm": 1.6360913313098524,
      "language_loss": 0.65877289,
      "learning_rate": 1.2824674362249922e-07,
      "loss": 0.68008041,
      "num_input_tokens_seen": 318785660,
      "step": 14783,
      "time_per_iteration": 2.7113358974456787
    },
    {
      "auxiliary_loss_clip": 0.01104368,
      "auxiliary_loss_mlp": 0.01030793,
      "balance_loss_clip": 1.03363192,
      "balance_loss_mlp": 1.01822233,
      "epoch": 0.8888621674432586,
      "flos": 22162701605760.0,
      "grad_norm": 1.5746138761182862,
      "language_loss": 0.77722031,
      "learning_rate": 1.281095609023415e-07,
      "loss": 0.79857194,
      "num_input_tokens_seen": 318806080,
      "step": 14784,
      "time_per_iteration": 2.4481658935546875
    },
    {
      "auxiliary_loss_clip": 0.0107759,
      "auxiliary_loss_mlp": 0.01033035,
      "balance_loss_clip": 1.03361952,
      "balance_loss_mlp": 1.02073908,
      "epoch": 0.8889222906959267,
      "flos": 27672834723840.0,
      "grad_norm": 2.126451106868472,
      "language_loss": 0.60150492,
      "learning_rate": 1.279724491644565e-07,
      "loss": 0.62261117,
      "num_input_tokens_seen": 318826445,
      "step": 14785,
      "time_per_iteration": 2.5773379802703857
    },
    {
      "auxiliary_loss_clip": 0.01070565,
      "auxiliary_loss_mlp": 0.01029843,
      "balance_loss_clip": 1.03501058,
      "balance_loss_mlp": 1.01762974,
      "epoch": 0.8889824139485947,
      "flos": 14168627274240.0,
      "grad_norm": 1.7443448220577387,
      "language_loss": 0.65019405,
      "learning_rate": 1.278354084140445e-07,
      "loss": 0.67119813,
      "num_input_tokens_seen": 318843915,
      "step": 14786,
      "time_per_iteration": 2.5079758167266846
    },
    {
      "auxiliary_loss_clip": 0.01072641,
      "auxiliary_loss_mlp": 0.00784171,
      "balance_loss_clip": 1.03492987,
      "balance_loss_mlp": 1.00780559,
      "epoch": 0.8890425372012626,
      "flos": 12853003829760.0,
      "grad_norm": 2.165595730300719,
      "language_loss": 0.85514522,
      "learning_rate": 1.276984386563009e-07,
      "loss": 0.87371331,
      "num_input_tokens_seen": 318859670,
      "step": 14787,
      "time_per_iteration": 2.5575172901153564
    },
    {
      "auxiliary_loss_clip": 0.01078191,
      "auxiliary_loss_mlp": 0.01027707,
      "balance_loss_clip": 1.03557718,
      "balance_loss_mlp": 1.01578569,
      "epoch": 0.8891026604539306,
      "flos": 21689291329920.0,
      "grad_norm": 2.501196251835476,
      "language_loss": 0.70919478,
      "learning_rate": 1.2756153989642027e-07,
      "loss": 0.73025376,
      "num_input_tokens_seen": 318877855,
      "step": 14788,
      "time_per_iteration": 2.5095324516296387
    },
    {
      "auxiliary_loss_clip": 0.01101192,
      "auxiliary_loss_mlp": 0.01029179,
      "balance_loss_clip": 1.03529084,
      "balance_loss_mlp": 1.01784229,
      "epoch": 0.8891627837065985,
      "flos": 21871430219520.0,
      "grad_norm": 1.6118745288161302,
      "language_loss": 0.70277691,
      "learning_rate": 1.274247121395935e-07,
      "loss": 0.72408056,
      "num_input_tokens_seen": 318896045,
      "step": 14789,
      "time_per_iteration": 2.4720683097839355
    },
    {
      "auxiliary_loss_clip": 0.01094914,
      "auxiliary_loss_mlp": 0.01025508,
      "balance_loss_clip": 1.03599107,
      "balance_loss_mlp": 1.01346159,
      "epoch": 0.8892229069592665,
      "flos": 21580230660480.0,
      "grad_norm": 1.6351552431997953,
      "language_loss": 0.70204329,
      "learning_rate": 1.2728795539100956e-07,
      "loss": 0.72324753,
      "num_input_tokens_seen": 318915515,
      "step": 14790,
      "time_per_iteration": 2.4887616634368896
    },
    {
      "auxiliary_loss_clip": 0.01082054,
      "auxiliary_loss_mlp": 0.01027027,
      "balance_loss_clip": 1.0345062,
      "balance_loss_mlp": 1.01565421,
      "epoch": 0.8892830302119344,
      "flos": 23075981832960.0,
      "grad_norm": 1.7059943107379352,
      "language_loss": 0.72621655,
      "learning_rate": 1.2715126965585387e-07,
      "loss": 0.74730736,
      "num_input_tokens_seen": 318934305,
      "step": 14791,
      "time_per_iteration": 2.5478217601776123
    },
    {
      "auxiliary_loss_clip": 0.01070836,
      "auxiliary_loss_mlp": 0.01032576,
      "balance_loss_clip": 1.03348541,
      "balance_loss_mlp": 1.02060747,
      "epoch": 0.8893431534646025,
      "flos": 23072139077760.0,
      "grad_norm": 1.4587296950427202,
      "language_loss": 0.7415024,
      "learning_rate": 1.2701465493931008e-07,
      "loss": 0.76253653,
      "num_input_tokens_seen": 318953880,
      "step": 14792,
      "time_per_iteration": 2.5476179122924805
    },
    {
      "auxiliary_loss_clip": 0.01035638,
      "auxiliary_loss_mlp": 0.01033517,
      "balance_loss_clip": 1.03098845,
      "balance_loss_mlp": 1.02042818,
      "epoch": 0.8894032767172704,
      "flos": 22454978572800.0,
      "grad_norm": 1.8456321843614887,
      "language_loss": 0.66111988,
      "learning_rate": 1.2687811124655801e-07,
      "loss": 0.68181145,
      "num_input_tokens_seen": 318971395,
      "step": 14793,
      "time_per_iteration": 4.343828916549683
    },
    {
      "auxiliary_loss_clip": 0.01075472,
      "auxiliary_loss_mlp": 0.01029674,
      "balance_loss_clip": 1.03482568,
      "balance_loss_mlp": 1.01687682,
      "epoch": 0.8894633999699384,
      "flos": 25338246261120.0,
      "grad_norm": 1.642017553759774,
      "language_loss": 0.71513343,
      "learning_rate": 1.2674163858277552e-07,
      "loss": 0.73618484,
      "num_input_tokens_seen": 318990580,
      "step": 14794,
      "time_per_iteration": 4.152606964111328
    },
    {
      "auxiliary_loss_clip": 0.01093972,
      "auxiliary_loss_mlp": 0.01032122,
      "balance_loss_clip": 1.03665054,
      "balance_loss_mlp": 1.01910436,
      "epoch": 0.8895235232226063,
      "flos": 20994096528000.0,
      "grad_norm": 1.5902845514723203,
      "language_loss": 0.75336468,
      "learning_rate": 1.2660523695313785e-07,
      "loss": 0.77462566,
      "num_input_tokens_seen": 319010040,
      "step": 14795,
      "time_per_iteration": 2.479736328125
    },
    {
      "auxiliary_loss_clip": 0.01002272,
      "auxiliary_loss_mlp": 0.00999726,
      "balance_loss_clip": 1.00910187,
      "balance_loss_mlp": 0.99849814,
      "epoch": 0.8895836464752743,
      "flos": 69732956764800.0,
      "grad_norm": 0.7717996439197335,
      "language_loss": 0.56109905,
      "learning_rate": 1.2646890636281727e-07,
      "loss": 0.58111894,
      "num_input_tokens_seen": 319063860,
      "step": 14796,
      "time_per_iteration": 3.0396525859832764
    },
    {
      "auxiliary_loss_clip": 0.01106425,
      "auxiliary_loss_mlp": 0.01032866,
      "balance_loss_clip": 1.03614402,
      "balance_loss_mlp": 1.01885295,
      "epoch": 0.8896437697279422,
      "flos": 23221815050880.0,
      "grad_norm": 1.8855348564433667,
      "language_loss": 0.70507228,
      "learning_rate": 1.263326468169843e-07,
      "loss": 0.72646523,
      "num_input_tokens_seen": 319082335,
      "step": 14797,
      "time_per_iteration": 3.870093584060669
    },
    {
      "auxiliary_loss_clip": 0.01016569,
      "auxiliary_loss_mlp": 0.01005536,
      "balance_loss_clip": 1.01219368,
      "balance_loss_mlp": 1.00450468,
      "epoch": 0.8897038929806103,
      "flos": 70752711882240.0,
      "grad_norm": 2.0555559270329384,
      "language_loss": 0.58038336,
      "learning_rate": 1.2619645832080417e-07,
      "loss": 0.60060441,
      "num_input_tokens_seen": 319147075,
      "step": 14798,
      "time_per_iteration": 3.1666107177734375
    },
    {
      "auxiliary_loss_clip": 0.01092677,
      "auxiliary_loss_mlp": 0.01026627,
      "balance_loss_clip": 1.0339886,
      "balance_loss_mlp": 1.01448512,
      "epoch": 0.8897640162332782,
      "flos": 19245103493760.0,
      "grad_norm": 1.50829272113656,
      "language_loss": 0.7957083,
      "learning_rate": 1.2606034087944251e-07,
      "loss": 0.81690139,
      "num_input_tokens_seen": 319166630,
      "step": 14799,
      "time_per_iteration": 2.469029426574707
    },
    {
      "auxiliary_loss_clip": 0.01018846,
      "auxiliary_loss_mlp": 0.00999343,
      "balance_loss_clip": 1.00569963,
      "balance_loss_mlp": 0.99832362,
      "epoch": 0.8898241394859462,
      "flos": 41356275039360.0,
      "grad_norm": 0.8805503724292374,
      "language_loss": 0.58097434,
      "learning_rate": 1.2592429449806053e-07,
      "loss": 0.60115623,
      "num_input_tokens_seen": 319221865,
      "step": 14800,
      "time_per_iteration": 3.023193120956421
    },
    {
      "auxiliary_loss_clip": 0.01093032,
      "auxiliary_loss_mlp": 0.01028791,
      "balance_loss_clip": 1.03571105,
      "balance_loss_mlp": 1.01762688,
      "epoch": 0.8898842627386142,
      "flos": 18986295024000.0,
      "grad_norm": 1.4722163058480664,
      "language_loss": 0.66260612,
      "learning_rate": 1.2578831918181698e-07,
      "loss": 0.6838243,
      "num_input_tokens_seen": 319240710,
      "step": 14801,
      "time_per_iteration": 2.4837498664855957
    },
    {
      "auxiliary_loss_clip": 0.01077825,
      "auxiliary_loss_mlp": 0.01035108,
      "balance_loss_clip": 1.03658533,
      "balance_loss_mlp": 1.02154231,
      "epoch": 0.8899443859912821,
      "flos": 13217173868160.0,
      "grad_norm": 13.143894500658542,
      "language_loss": 0.75125837,
      "learning_rate": 1.256524149358682e-07,
      "loss": 0.77238768,
      "num_input_tokens_seen": 319256495,
      "step": 14802,
      "time_per_iteration": 2.514037847518921
    },
    {
      "auxiliary_loss_clip": 0.01086707,
      "auxiliary_loss_mlp": 0.01031317,
      "balance_loss_clip": 1.03476036,
      "balance_loss_mlp": 1.02001595,
      "epoch": 0.8900045092439501,
      "flos": 22674680110080.0,
      "grad_norm": 1.8427096132597989,
      "language_loss": 0.73094642,
      "learning_rate": 1.2551658176536805e-07,
      "loss": 0.75212669,
      "num_input_tokens_seen": 319273620,
      "step": 14803,
      "time_per_iteration": 3.8748252391815186
    },
    {
      "auxiliary_loss_clip": 0.01082782,
      "auxiliary_loss_mlp": 0.0103305,
      "balance_loss_clip": 1.03505158,
      "balance_loss_mlp": 1.02118921,
      "epoch": 0.890064632496618,
      "flos": 21141617685120.0,
      "grad_norm": 1.821593265169082,
      "language_loss": 0.72203696,
      "learning_rate": 1.2538081967546664e-07,
      "loss": 0.7431953,
      "num_input_tokens_seen": 319291720,
      "step": 14804,
      "time_per_iteration": 2.5248665809631348
    },
    {
      "auxiliary_loss_clip": 0.01089002,
      "auxiliary_loss_mlp": 0.01032815,
      "balance_loss_clip": 1.03425944,
      "balance_loss_mlp": 1.02065003,
      "epoch": 0.8901247557492861,
      "flos": 23397058529280.0,
      "grad_norm": 1.762159693306876,
      "language_loss": 0.81022114,
      "learning_rate": 1.252451286713123e-07,
      "loss": 0.83143938,
      "num_input_tokens_seen": 319310380,
      "step": 14805,
      "time_per_iteration": 2.4990882873535156
    },
    {
      "auxiliary_loss_clip": 0.01094921,
      "auxiliary_loss_mlp": 0.0102949,
      "balance_loss_clip": 1.03525376,
      "balance_loss_mlp": 1.01700234,
      "epoch": 0.890184879001954,
      "flos": 29169591477120.0,
      "grad_norm": 2.0060838836161037,
      "language_loss": 0.6752671,
      "learning_rate": 1.251095087580505e-07,
      "loss": 0.69651121,
      "num_input_tokens_seen": 319331765,
      "step": 14806,
      "time_per_iteration": 2.556997776031494
    },
    {
      "auxiliary_loss_clip": 0.01080562,
      "auxiliary_loss_mlp": 0.0103268,
      "balance_loss_clip": 1.03347707,
      "balance_loss_mlp": 1.02044868,
      "epoch": 0.890245002254622,
      "flos": 14427830793600.0,
      "grad_norm": 1.9131407424449522,
      "language_loss": 0.67100191,
      "learning_rate": 1.2497395994082438e-07,
      "loss": 0.69213432,
      "num_input_tokens_seen": 319349135,
      "step": 14807,
      "time_per_iteration": 2.490407705307007
    },
    {
      "auxiliary_loss_clip": 0.0107875,
      "auxiliary_loss_mlp": 0.01028949,
      "balance_loss_clip": 1.03333783,
      "balance_loss_mlp": 1.01767778,
      "epoch": 0.8903051255072899,
      "flos": 22382187661440.0,
      "grad_norm": 1.7911575303453495,
      "language_loss": 0.75134987,
      "learning_rate": 1.248384822247732e-07,
      "loss": 0.77242684,
      "num_input_tokens_seen": 319368410,
      "step": 14808,
      "time_per_iteration": 2.531984329223633
    },
    {
      "auxiliary_loss_clip": 0.01071015,
      "auxiliary_loss_mlp": 0.0102974,
      "balance_loss_clip": 1.03402948,
      "balance_loss_mlp": 1.01822484,
      "epoch": 0.8903652487599579,
      "flos": 20777375819520.0,
      "grad_norm": 1.803196553119202,
      "language_loss": 0.81342244,
      "learning_rate": 1.2470307561503513e-07,
      "loss": 0.83442998,
      "num_input_tokens_seen": 319387535,
      "step": 14809,
      "time_per_iteration": 2.535243511199951
    },
    {
      "auxiliary_loss_clip": 0.01093863,
      "auxiliary_loss_mlp": 0.01027089,
      "balance_loss_clip": 1.03502798,
      "balance_loss_mlp": 1.01547766,
      "epoch": 0.8904253720126258,
      "flos": 24424499157120.0,
      "grad_norm": 1.8322106166513674,
      "language_loss": 0.68327099,
      "learning_rate": 1.2456774011674442e-07,
      "loss": 0.70448047,
      "num_input_tokens_seen": 319407210,
      "step": 14810,
      "time_per_iteration": 2.5826032161712646
    },
    {
      "auxiliary_loss_clip": 0.01070197,
      "auxiliary_loss_mlp": 0.01028299,
      "balance_loss_clip": 1.0324614,
      "balance_loss_mlp": 1.01540041,
      "epoch": 0.8904854952652939,
      "flos": 19463871277440.0,
      "grad_norm": 2.0638228425460916,
      "language_loss": 0.7016722,
      "learning_rate": 1.2443247573503257e-07,
      "loss": 0.72265714,
      "num_input_tokens_seen": 319425340,
      "step": 14811,
      "time_per_iteration": 2.5280051231384277
    },
    {
      "auxiliary_loss_clip": 0.01071359,
      "auxiliary_loss_mlp": 0.00783259,
      "balance_loss_clip": 1.03418541,
      "balance_loss_mlp": 1.00868833,
      "epoch": 0.8905456185179618,
      "flos": 50800741666560.0,
      "grad_norm": 1.983922041570675,
      "language_loss": 0.65255433,
      "learning_rate": 1.2429728247502924e-07,
      "loss": 0.67110056,
      "num_input_tokens_seen": 319448150,
      "step": 14812,
      "time_per_iteration": 2.801257610321045
    },
    {
      "auxiliary_loss_clip": 0.01056287,
      "auxiliary_loss_mlp": 0.01029348,
      "balance_loss_clip": 1.0345844,
      "balance_loss_mlp": 1.01802874,
      "epoch": 0.8906057417706298,
      "flos": 17784867893760.0,
      "grad_norm": 1.7444793365529172,
      "language_loss": 0.68781304,
      "learning_rate": 1.24162160341861e-07,
      "loss": 0.70866942,
      "num_input_tokens_seen": 319466115,
      "step": 14813,
      "time_per_iteration": 2.5375816822052
    },
    {
      "auxiliary_loss_clip": 0.01080457,
      "auxiliary_loss_mlp": 0.01038862,
      "balance_loss_clip": 1.03207254,
      "balance_loss_mlp": 1.02266169,
      "epoch": 0.8906658650232978,
      "flos": 21944867575680.0,
      "grad_norm": 1.726444062831347,
      "language_loss": 0.75219786,
      "learning_rate": 1.2402710934065198e-07,
      "loss": 0.77339107,
      "num_input_tokens_seen": 319485255,
      "step": 14814,
      "time_per_iteration": 2.50970196723938
    },
    {
      "auxiliary_loss_clip": 0.01094823,
      "auxiliary_loss_mlp": 0.01025946,
      "balance_loss_clip": 1.03426194,
      "balance_loss_mlp": 1.01357794,
      "epoch": 0.8907259882759657,
      "flos": 21287810039040.0,
      "grad_norm": 1.9579099469166052,
      "language_loss": 0.74138284,
      "learning_rate": 1.2389212947652229e-07,
      "loss": 0.76259053,
      "num_input_tokens_seen": 319501800,
      "step": 14815,
      "time_per_iteration": 2.456357002258301
    },
    {
      "auxiliary_loss_clip": 0.01065298,
      "auxiliary_loss_mlp": 0.01030802,
      "balance_loss_clip": 1.03272557,
      "balance_loss_mlp": 1.01843369,
      "epoch": 0.8907861115286337,
      "flos": 20120426023680.0,
      "grad_norm": 1.715445428657627,
      "language_loss": 0.74947739,
      "learning_rate": 1.237572207545914e-07,
      "loss": 0.77043843,
      "num_input_tokens_seen": 319520415,
      "step": 14816,
      "time_per_iteration": 2.547678232192993
    },
    {
      "auxiliary_loss_clip": 0.0108274,
      "auxiliary_loss_mlp": 0.01028312,
      "balance_loss_clip": 1.03338337,
      "balance_loss_mlp": 1.01657009,
      "epoch": 0.8908462347813016,
      "flos": 20084156265600.0,
      "grad_norm": 1.7277091325077811,
      "language_loss": 0.77749598,
      "learning_rate": 1.2362238317997476e-07,
      "loss": 0.79860651,
      "num_input_tokens_seen": 319538410,
      "step": 14817,
      "time_per_iteration": 2.534184694290161
    },
    {
      "auxiliary_loss_clip": 0.01002516,
      "auxiliary_loss_mlp": 0.01003476,
      "balance_loss_clip": 1.00792098,
      "balance_loss_mlp": 1.00229573,
      "epoch": 0.8909063580339697,
      "flos": 65503649790720.0,
      "grad_norm": 0.7506755006812162,
      "language_loss": 0.56473804,
      "learning_rate": 1.2348761675778517e-07,
      "loss": 0.58479798,
      "num_input_tokens_seen": 319602565,
      "step": 14818,
      "time_per_iteration": 3.1919074058532715
    },
    {
      "auxiliary_loss_clip": 0.01055291,
      "auxiliary_loss_mlp": 0.01032213,
      "balance_loss_clip": 1.0367713,
      "balance_loss_mlp": 1.02029812,
      "epoch": 0.8909664812866376,
      "flos": 29863062426240.0,
      "grad_norm": 1.7680304357569707,
      "language_loss": 0.64618576,
      "learning_rate": 1.2335292149313325e-07,
      "loss": 0.66706079,
      "num_input_tokens_seen": 319624645,
      "step": 14819,
      "time_per_iteration": 2.6521623134613037
    },
    {
      "auxiliary_loss_clip": 0.01093104,
      "auxiliary_loss_mlp": 0.01032363,
      "balance_loss_clip": 1.03435123,
      "balance_loss_mlp": 1.01943517,
      "epoch": 0.8910266045393056,
      "flos": 25447127362560.0,
      "grad_norm": 2.983978733856615,
      "language_loss": 0.7871325,
      "learning_rate": 1.2321829739112731e-07,
      "loss": 0.80838716,
      "num_input_tokens_seen": 319644040,
      "step": 14820,
      "time_per_iteration": 2.5172977447509766
    },
    {
      "auxiliary_loss_clip": 0.01071152,
      "auxiliary_loss_mlp": 0.00783284,
      "balance_loss_clip": 1.03392363,
      "balance_loss_mlp": 1.00963938,
      "epoch": 0.8910867277919735,
      "flos": 24499121662080.0,
      "grad_norm": 1.796542552394135,
      "language_loss": 0.76725578,
      "learning_rate": 1.2308374445687087e-07,
      "loss": 0.78580022,
      "num_input_tokens_seen": 319663930,
      "step": 14821,
      "time_per_iteration": 2.5661685466766357
    },
    {
      "auxiliary_loss_clip": 0.01018598,
      "auxiliary_loss_mlp": 0.00762632,
      "balance_loss_clip": 1.00964868,
      "balance_loss_mlp": 1.00024307,
      "epoch": 0.8911468510446415,
      "flos": 60688136856960.0,
      "grad_norm": 0.7951907419902979,
      "language_loss": 0.59265792,
      "learning_rate": 1.2294926269546712e-07,
      "loss": 0.61047018,
      "num_input_tokens_seen": 319721245,
      "step": 14822,
      "time_per_iteration": 2.9911487102508545
    },
    {
      "auxiliary_loss_clip": 0.01086627,
      "auxiliary_loss_mlp": 0.01030192,
      "balance_loss_clip": 1.03359771,
      "balance_loss_mlp": 1.018116,
      "epoch": 0.8912069742973094,
      "flos": 25337492075520.0,
      "grad_norm": 1.7370656772097506,
      "language_loss": 0.69278973,
      "learning_rate": 1.2281485211201515e-07,
      "loss": 0.71395791,
      "num_input_tokens_seen": 319741200,
      "step": 14823,
      "time_per_iteration": 2.5412843227386475
    },
    {
      "auxiliary_loss_clip": 0.01084445,
      "auxiliary_loss_mlp": 0.01030732,
      "balance_loss_clip": 1.03174222,
      "balance_loss_mlp": 1.01826835,
      "epoch": 0.8912670975499775,
      "flos": 18223516782720.0,
      "grad_norm": 1.667301244128644,
      "language_loss": 0.69171327,
      "learning_rate": 1.2268051271161262e-07,
      "loss": 0.712865,
      "num_input_tokens_seen": 319759265,
      "step": 14824,
      "time_per_iteration": 2.4994192123413086
    },
    {
      "auxiliary_loss_clip": 0.0105467,
      "auxiliary_loss_mlp": 0.01040823,
      "balance_loss_clip": 1.03438485,
      "balance_loss_mlp": 1.02657104,
      "epoch": 0.8913272208026454,
      "flos": 26504481041280.0,
      "grad_norm": 1.7268978086662,
      "language_loss": 0.70666462,
      "learning_rate": 1.2254624449935303e-07,
      "loss": 0.72761953,
      "num_input_tokens_seen": 319777560,
      "step": 14825,
      "time_per_iteration": 2.6672799587249756
    },
    {
      "auxiliary_loss_clip": 0.01080936,
      "auxiliary_loss_mlp": 0.01031129,
      "balance_loss_clip": 1.03293896,
      "balance_loss_mlp": 1.01839721,
      "epoch": 0.8913873440553134,
      "flos": 18802324540800.0,
      "grad_norm": 2.1196632409506795,
      "language_loss": 0.70680189,
      "learning_rate": 1.2241204748032786e-07,
      "loss": 0.72792256,
      "num_input_tokens_seen": 319794125,
      "step": 14826,
      "time_per_iteration": 2.5015711784362793
    },
    {
      "auxiliary_loss_clip": 0.0109254,
      "auxiliary_loss_mlp": 0.01023551,
      "balance_loss_clip": 1.03604841,
      "balance_loss_mlp": 1.01235127,
      "epoch": 0.8914474673079814,
      "flos": 20884892204160.0,
      "grad_norm": 2.1277681917413473,
      "language_loss": 0.74603981,
      "learning_rate": 1.2227792165962615e-07,
      "loss": 0.76720077,
      "num_input_tokens_seen": 319810310,
      "step": 14827,
      "time_per_iteration": 2.525562047958374
    },
    {
      "auxiliary_loss_clip": 0.01095199,
      "auxiliary_loss_mlp": 0.0103,
      "balance_loss_clip": 1.03514242,
      "balance_loss_mlp": 1.01782823,
      "epoch": 0.8915075905606493,
      "flos": 20952439729920.0,
      "grad_norm": 1.6888287470090129,
      "language_loss": 0.78134853,
      "learning_rate": 1.221438670423336e-07,
      "loss": 0.8026005,
      "num_input_tokens_seen": 319828505,
      "step": 14828,
      "time_per_iteration": 2.5199453830718994
    },
    {
      "auxiliary_loss_clip": 0.01070947,
      "auxiliary_loss_mlp": 0.01030759,
      "balance_loss_clip": 1.03514731,
      "balance_loss_mlp": 1.01863551,
      "epoch": 0.8915677138133173,
      "flos": 23076305055360.0,
      "grad_norm": 1.63639096198059,
      "language_loss": 0.75383073,
      "learning_rate": 1.2200988363353392e-07,
      "loss": 0.77484781,
      "num_input_tokens_seen": 319848680,
      "step": 14829,
      "time_per_iteration": 2.593977928161621
    },
    {
      "auxiliary_loss_clip": 0.01103455,
      "auxiliary_loss_mlp": 0.01034457,
      "balance_loss_clip": 1.03417838,
      "balance_loss_mlp": 1.02326345,
      "epoch": 0.8916278370659853,
      "flos": 23440259612160.0,
      "grad_norm": 1.5534190104338388,
      "language_loss": 0.84244943,
      "learning_rate": 1.2187597143830773e-07,
      "loss": 0.86382854,
      "num_input_tokens_seen": 319868835,
      "step": 14830,
      "time_per_iteration": 2.5071849822998047
    },
    {
      "auxiliary_loss_clip": 0.0108883,
      "auxiliary_loss_mlp": 0.01028294,
      "balance_loss_clip": 1.03409457,
      "balance_loss_mlp": 1.01748729,
      "epoch": 0.8916879603186533,
      "flos": 25160488830720.0,
      "grad_norm": 1.3563058533125143,
      "language_loss": 0.75068378,
      "learning_rate": 1.2174213046173299e-07,
      "loss": 0.77185506,
      "num_input_tokens_seen": 319891585,
      "step": 14831,
      "time_per_iteration": 3.9934234619140625
    },
    {
      "auxiliary_loss_clip": 0.01095038,
      "auxiliary_loss_mlp": 0.0102875,
      "balance_loss_clip": 1.03379619,
      "balance_loss_mlp": 1.0163765,
      "epoch": 0.8917480835713212,
      "flos": 20229845829120.0,
      "grad_norm": 1.9951289776671632,
      "language_loss": 0.72890419,
      "learning_rate": 1.216083607088847e-07,
      "loss": 0.7501421,
      "num_input_tokens_seen": 319910315,
      "step": 14832,
      "time_per_iteration": 3.893512487411499
    },
    {
      "auxiliary_loss_clip": 0.01049075,
      "auxiliary_loss_mlp": 0.00781434,
      "balance_loss_clip": 1.03263867,
      "balance_loss_mlp": 1.00638616,
      "epoch": 0.8918082068239892,
      "flos": 26101922342400.0,
      "grad_norm": 1.7330483966043138,
      "language_loss": 0.67078936,
      "learning_rate": 1.214746621848355e-07,
      "loss": 0.68909442,
      "num_input_tokens_seen": 319932275,
      "step": 14833,
      "time_per_iteration": 2.6621882915496826
    },
    {
      "auxiliary_loss_clip": 0.01099627,
      "auxiliary_loss_mlp": 0.01032899,
      "balance_loss_clip": 1.03659379,
      "balance_loss_mlp": 1.0197202,
      "epoch": 0.8918683300766571,
      "flos": 24831439315200.0,
      "grad_norm": 2.901661324015054,
      "language_loss": 0.73560345,
      "learning_rate": 1.2134103489465575e-07,
      "loss": 0.75692868,
      "num_input_tokens_seen": 319955335,
      "step": 14834,
      "time_per_iteration": 2.605165958404541
    },
    {
      "auxiliary_loss_clip": 0.01067344,
      "auxiliary_loss_mlp": 0.01032867,
      "balance_loss_clip": 1.03352952,
      "balance_loss_mlp": 1.020684,
      "epoch": 0.8919284533293251,
      "flos": 22305158945280.0,
      "grad_norm": 2.044618747781443,
      "language_loss": 0.78971827,
      "learning_rate": 1.2120747884341188e-07,
      "loss": 0.81072032,
      "num_input_tokens_seen": 319973990,
      "step": 14835,
      "time_per_iteration": 3.984410524368286
    },
    {
      "auxiliary_loss_clip": 0.01098856,
      "auxiliary_loss_mlp": 0.01028436,
      "balance_loss_clip": 1.03276169,
      "balance_loss_mlp": 1.01736784,
      "epoch": 0.891988576581993,
      "flos": 30373532559360.0,
      "grad_norm": 1.4952576564187399,
      "language_loss": 0.73750126,
      "learning_rate": 1.210739940361689e-07,
      "loss": 0.75877422,
      "num_input_tokens_seen": 319995555,
      "step": 14836,
      "time_per_iteration": 2.5052273273468018
    },
    {
      "auxiliary_loss_clip": 0.01082194,
      "auxiliary_loss_mlp": 0.01030876,
      "balance_loss_clip": 1.03291345,
      "balance_loss_mlp": 1.01902103,
      "epoch": 0.8920486998346611,
      "flos": 15552947479680.0,
      "grad_norm": 3.3135604027457166,
      "language_loss": 0.68894231,
      "learning_rate": 1.2094058047798838e-07,
      "loss": 0.71007311,
      "num_input_tokens_seen": 320012385,
      "step": 14837,
      "time_per_iteration": 2.487098217010498
    },
    {
      "auxiliary_loss_clip": 0.0105612,
      "auxiliary_loss_mlp": 0.01031705,
      "balance_loss_clip": 1.03514636,
      "balance_loss_mlp": 1.01869941,
      "epoch": 0.892108823087329,
      "flos": 21214983214080.0,
      "grad_norm": 1.9180944781174813,
      "language_loss": 0.67527866,
      "learning_rate": 1.2080723817392913e-07,
      "loss": 0.69615692,
      "num_input_tokens_seen": 320032390,
      "step": 14838,
      "time_per_iteration": 2.5784826278686523
    },
    {
      "auxiliary_loss_clip": 0.01092098,
      "auxiliary_loss_mlp": 0.01031229,
      "balance_loss_clip": 1.03270912,
      "balance_loss_mlp": 1.01799059,
      "epoch": 0.892168946339997,
      "flos": 21978982517760.0,
      "grad_norm": 2.09505321382696,
      "language_loss": 0.76307875,
      "learning_rate": 1.2067396712904777e-07,
      "loss": 0.78431201,
      "num_input_tokens_seen": 320052885,
      "step": 14839,
      "time_per_iteration": 2.485930919647217
    },
    {
      "auxiliary_loss_clip": 0.01001056,
      "auxiliary_loss_mlp": 0.00761769,
      "balance_loss_clip": 1.00795984,
      "balance_loss_mlp": 1.00041723,
      "epoch": 0.892229069592665,
      "flos": 67475289277440.0,
      "grad_norm": 0.6989223465462696,
      "language_loss": 0.49376684,
      "learning_rate": 1.205407673483978e-07,
      "loss": 0.51139516,
      "num_input_tokens_seen": 320113685,
      "step": 14840,
      "time_per_iteration": 3.1210856437683105
    },
    {
      "auxiliary_loss_clip": 0.01108701,
      "auxiliary_loss_mlp": 0.010326,
      "balance_loss_clip": 1.03476715,
      "balance_loss_mlp": 1.0192126,
      "epoch": 0.8922891928453329,
      "flos": 19459561645440.0,
      "grad_norm": 2.558231128969223,
      "language_loss": 0.6379419,
      "learning_rate": 1.2040763883703074e-07,
      "loss": 0.65935487,
      "num_input_tokens_seen": 320130810,
      "step": 14841,
      "time_per_iteration": 2.4866116046905518
    },
    {
      "auxiliary_loss_clip": 0.01069059,
      "auxiliary_loss_mlp": 0.00782464,
      "balance_loss_clip": 1.03446352,
      "balance_loss_mlp": 1.01076293,
      "epoch": 0.8923493160980009,
      "flos": 23367396873600.0,
      "grad_norm": 1.3989295485906519,
      "language_loss": 0.6815697,
      "learning_rate": 1.2027458159999438e-07,
      "loss": 0.70008492,
      "num_input_tokens_seen": 320152170,
      "step": 14842,
      "time_per_iteration": 3.9559473991394043
    },
    {
      "auxiliary_loss_clip": 0.0110259,
      "auxiliary_loss_mlp": 0.01031991,
      "balance_loss_clip": 1.03556252,
      "balance_loss_mlp": 1.02068448,
      "epoch": 0.8924094393506689,
      "flos": 26177047637760.0,
      "grad_norm": 2.01498606779671,
      "language_loss": 0.79842663,
      "learning_rate": 1.2014159564233373e-07,
      "loss": 0.81977248,
      "num_input_tokens_seen": 320172360,
      "step": 14843,
      "time_per_iteration": 2.481513023376465
    },
    {
      "auxiliary_loss_clip": 0.01080374,
      "auxiliary_loss_mlp": 0.01035067,
      "balance_loss_clip": 1.03354299,
      "balance_loss_mlp": 1.02257967,
      "epoch": 0.8924695626033369,
      "flos": 22018520413440.0,
      "grad_norm": 1.7121616816048129,
      "language_loss": 0.6806978,
      "learning_rate": 1.2000868096909257e-07,
      "loss": 0.70185226,
      "num_input_tokens_seen": 320192130,
      "step": 14844,
      "time_per_iteration": 2.544724464416504
    },
    {
      "auxiliary_loss_clip": 0.01059353,
      "auxiliary_loss_mlp": 0.01033595,
      "balance_loss_clip": 1.03332937,
      "balance_loss_mlp": 1.02171564,
      "epoch": 0.8925296858560048,
      "flos": 14793940166400.0,
      "grad_norm": 1.984629803593817,
      "language_loss": 0.91460389,
      "learning_rate": 1.1987583758531038e-07,
      "loss": 0.9355334,
      "num_input_tokens_seen": 320207760,
      "step": 14845,
      "time_per_iteration": 2.536334991455078
    },
    {
      "auxiliary_loss_clip": 0.0108508,
      "auxiliary_loss_mlp": 0.0102826,
      "balance_loss_clip": 1.03306246,
      "balance_loss_mlp": 1.01652968,
      "epoch": 0.8925898091086728,
      "flos": 22346636175360.0,
      "grad_norm": 1.9060547500087672,
      "language_loss": 0.72370416,
      "learning_rate": 1.1974306549602476e-07,
      "loss": 0.74483758,
      "num_input_tokens_seen": 320225325,
      "step": 14846,
      "time_per_iteration": 2.5018866062164307
    },
    {
      "auxiliary_loss_clip": 0.01075678,
      "auxiliary_loss_mlp": 0.01034353,
      "balance_loss_clip": 1.03419757,
      "balance_loss_mlp": 1.02231872,
      "epoch": 0.8926499323613407,
      "flos": 45806322067200.0,
      "grad_norm": 1.6872753150711388,
      "language_loss": 0.57176518,
      "learning_rate": 1.1961036470627094e-07,
      "loss": 0.59286547,
      "num_input_tokens_seen": 320247645,
      "step": 14847,
      "time_per_iteration": 2.7422313690185547
    },
    {
      "auxiliary_loss_clip": 0.01073082,
      "auxiliary_loss_mlp": 0.0103045,
      "balance_loss_clip": 1.03418136,
      "balance_loss_mlp": 1.0193572,
      "epoch": 0.8927100556140087,
      "flos": 22127042378880.0,
      "grad_norm": 1.889045887475548,
      "language_loss": 0.76863843,
      "learning_rate": 1.1947773522108052e-07,
      "loss": 0.78967369,
      "num_input_tokens_seen": 320266005,
      "step": 14848,
      "time_per_iteration": 2.5706076622009277
    },
    {
      "auxiliary_loss_clip": 0.01040097,
      "auxiliary_loss_mlp": 0.01042249,
      "balance_loss_clip": 1.03178644,
      "balance_loss_mlp": 1.02851057,
      "epoch": 0.8927701788666766,
      "flos": 28330143655680.0,
      "grad_norm": 1.7164142989798647,
      "language_loss": 0.68733108,
      "learning_rate": 1.1934517704548251e-07,
      "loss": 0.70815456,
      "num_input_tokens_seen": 320285555,
      "step": 14849,
      "time_per_iteration": 2.688546895980835
    },
    {
      "auxiliary_loss_clip": 0.01096332,
      "auxiliary_loss_mlp": 0.01033138,
      "balance_loss_clip": 1.03783417,
      "balance_loss_mlp": 1.02172351,
      "epoch": 0.8928303021193447,
      "flos": 25294973351040.0,
      "grad_norm": 1.7257529116425367,
      "language_loss": 0.8053782,
      "learning_rate": 1.1921269018450364e-07,
      "loss": 0.82667291,
      "num_input_tokens_seen": 320305395,
      "step": 14850,
      "time_per_iteration": 2.5390236377716064
    },
    {
      "auxiliary_loss_clip": 0.01082991,
      "auxiliary_loss_mlp": 0.01036406,
      "balance_loss_clip": 1.03431749,
      "balance_loss_mlp": 1.02451491,
      "epoch": 0.8928904253720126,
      "flos": 22236713579520.0,
      "grad_norm": 1.5519111866586504,
      "language_loss": 0.74881804,
      "learning_rate": 1.1908027464316872e-07,
      "loss": 0.77001202,
      "num_input_tokens_seen": 320324220,
      "step": 14851,
      "time_per_iteration": 2.5222365856170654
    },
    {
      "auxiliary_loss_clip": 0.01079661,
      "auxiliary_loss_mlp": 0.01030283,
      "balance_loss_clip": 1.03378797,
      "balance_loss_mlp": 1.01798081,
      "epoch": 0.8929505486246806,
      "flos": 27092374940160.0,
      "grad_norm": 1.6703688292997558,
      "language_loss": 0.78455478,
      "learning_rate": 1.1894793042649775e-07,
      "loss": 0.80565423,
      "num_input_tokens_seen": 320347195,
      "step": 14852,
      "time_per_iteration": 2.5879154205322266
    },
    {
      "auxiliary_loss_clip": 0.01085743,
      "auxiliary_loss_mlp": 0.01029479,
      "balance_loss_clip": 1.03635454,
      "balance_loss_mlp": 1.01793957,
      "epoch": 0.8930106718773486,
      "flos": 23039352938880.0,
      "grad_norm": 1.3746544291533502,
      "language_loss": 0.69433975,
      "learning_rate": 1.1881565753951006e-07,
      "loss": 0.71549201,
      "num_input_tokens_seen": 320366850,
      "step": 14853,
      "time_per_iteration": 2.4991390705108643
    },
    {
      "auxiliary_loss_clip": 0.01057754,
      "auxiliary_loss_mlp": 0.01031749,
      "balance_loss_clip": 1.03648472,
      "balance_loss_mlp": 1.01964307,
      "epoch": 0.8930707951300165,
      "flos": 35626652887680.0,
      "grad_norm": 1.8124723245572947,
      "language_loss": 0.67398167,
      "learning_rate": 1.1868345598722118e-07,
      "loss": 0.69487667,
      "num_input_tokens_seen": 320388895,
      "step": 14854,
      "time_per_iteration": 2.6998136043548584
    },
    {
      "auxiliary_loss_clip": 0.01074376,
      "auxiliary_loss_mlp": 0.01037634,
      "balance_loss_clip": 1.03111184,
      "balance_loss_mlp": 1.02526045,
      "epoch": 0.8931309183826845,
      "flos": 23039891642880.0,
      "grad_norm": 1.458941569900575,
      "language_loss": 0.7493661,
      "learning_rate": 1.1855132577464399e-07,
      "loss": 0.77048624,
      "num_input_tokens_seen": 320408520,
      "step": 14855,
      "time_per_iteration": 2.5093588829040527
    },
    {
      "auxiliary_loss_clip": 0.01080396,
      "auxiliary_loss_mlp": 0.01033189,
      "balance_loss_clip": 1.03362727,
      "balance_loss_mlp": 1.02145255,
      "epoch": 0.8931910416353525,
      "flos": 26504624695680.0,
      "grad_norm": 1.936283504185197,
      "language_loss": 0.64547491,
      "learning_rate": 1.1841926690678893e-07,
      "loss": 0.66661072,
      "num_input_tokens_seen": 320427400,
      "step": 14856,
      "time_per_iteration": 2.558302402496338
    },
    {
      "auxiliary_loss_clip": 0.01102405,
      "auxiliary_loss_mlp": 0.01028465,
      "balance_loss_clip": 1.03430128,
      "balance_loss_mlp": 1.01703298,
      "epoch": 0.8932511648880205,
      "flos": 24973609345920.0,
      "grad_norm": 1.6761835533049678,
      "language_loss": 0.66657883,
      "learning_rate": 1.1828727938866378e-07,
      "loss": 0.68788755,
      "num_input_tokens_seen": 320447570,
      "step": 14857,
      "time_per_iteration": 2.4735491275787354
    },
    {
      "auxiliary_loss_clip": 0.0106041,
      "auxiliary_loss_mlp": 0.01034734,
      "balance_loss_clip": 1.03466344,
      "balance_loss_mlp": 1.02222896,
      "epoch": 0.8933112881406884,
      "flos": 24460733001600.0,
      "grad_norm": 3.4618661539101416,
      "language_loss": 0.75820887,
      "learning_rate": 1.1815536322527408e-07,
      "loss": 0.77916026,
      "num_input_tokens_seen": 320464405,
      "step": 14858,
      "time_per_iteration": 2.6381285190582275
    },
    {
      "auxiliary_loss_clip": 0.01090864,
      "auxiliary_loss_mlp": 0.01030287,
      "balance_loss_clip": 1.03339136,
      "balance_loss_mlp": 1.01793647,
      "epoch": 0.8933714113933564,
      "flos": 28293083798400.0,
      "grad_norm": 1.4752533386606812,
      "language_loss": 0.69017041,
      "learning_rate": 1.1802351842162139e-07,
      "loss": 0.71138191,
      "num_input_tokens_seen": 320485525,
      "step": 14859,
      "time_per_iteration": 2.526698589324951
    },
    {
      "auxiliary_loss_clip": 0.01051402,
      "auxiliary_loss_mlp": 0.01027025,
      "balance_loss_clip": 1.03227973,
      "balance_loss_mlp": 1.01581335,
      "epoch": 0.8934315346460243,
      "flos": 21434864319360.0,
      "grad_norm": 1.5023634035192732,
      "language_loss": 0.7562325,
      "learning_rate": 1.1789174498270526e-07,
      "loss": 0.77701676,
      "num_input_tokens_seen": 320506725,
      "step": 14860,
      "time_per_iteration": 2.5832436084747314
    },
    {
      "auxiliary_loss_clip": 0.01082608,
      "auxiliary_loss_mlp": 0.0103235,
      "balance_loss_clip": 1.03555751,
      "balance_loss_mlp": 1.01893902,
      "epoch": 0.8934916578986923,
      "flos": 23769596436480.0,
      "grad_norm": 1.675799786382552,
      "language_loss": 0.5753504,
      "learning_rate": 1.1776004291352303e-07,
      "loss": 0.59649998,
      "num_input_tokens_seen": 320525425,
      "step": 14861,
      "time_per_iteration": 2.5200815200805664
    },
    {
      "auxiliary_loss_clip": 0.01072615,
      "auxiliary_loss_mlp": 0.01028938,
      "balance_loss_clip": 1.03154159,
      "balance_loss_mlp": 1.01730347,
      "epoch": 0.8935517811513602,
      "flos": 18916161719040.0,
      "grad_norm": 1.7750318806667615,
      "language_loss": 0.6301989,
      "learning_rate": 1.176284122190685e-07,
      "loss": 0.65121442,
      "num_input_tokens_seen": 320543010,
      "step": 14862,
      "time_per_iteration": 2.5323781967163086
    },
    {
      "auxiliary_loss_clip": 0.01089225,
      "auxiliary_loss_mlp": 0.01029881,
      "balance_loss_clip": 1.03266788,
      "balance_loss_mlp": 1.01821625,
      "epoch": 0.8936119044040283,
      "flos": 24061370613120.0,
      "grad_norm": 1.728849940634581,
      "language_loss": 0.78066665,
      "learning_rate": 1.1749685290433298e-07,
      "loss": 0.80185771,
      "num_input_tokens_seen": 320562180,
      "step": 14863,
      "time_per_iteration": 2.498753309249878
    },
    {
      "auxiliary_loss_clip": 0.0107515,
      "auxiliary_loss_mlp": 0.01033923,
      "balance_loss_clip": 1.03164506,
      "balance_loss_mlp": 1.02190685,
      "epoch": 0.8936720276566962,
      "flos": 21324079797120.0,
      "grad_norm": 2.746426770015608,
      "language_loss": 0.70994467,
      "learning_rate": 1.1736536497430627e-07,
      "loss": 0.73103541,
      "num_input_tokens_seen": 320580395,
      "step": 14864,
      "time_per_iteration": 2.5100224018096924
    },
    {
      "auxiliary_loss_clip": 0.01097325,
      "auxiliary_loss_mlp": 0.01035927,
      "balance_loss_clip": 1.03590667,
      "balance_loss_mlp": 1.02341628,
      "epoch": 0.8937321509093642,
      "flos": 18406122549120.0,
      "grad_norm": 1.9027977616912422,
      "language_loss": 0.75569171,
      "learning_rate": 1.1723394843397283e-07,
      "loss": 0.77702427,
      "num_input_tokens_seen": 320599505,
      "step": 14865,
      "time_per_iteration": 2.452277898788452
    },
    {
      "auxiliary_loss_clip": 0.01063322,
      "auxiliary_loss_mlp": 0.0103125,
      "balance_loss_clip": 1.03276753,
      "balance_loss_mlp": 1.02010965,
      "epoch": 0.8937922741620322,
      "flos": 22054754257920.0,
      "grad_norm": 1.6240393723412574,
      "language_loss": 0.7187292,
      "learning_rate": 1.1710260328831668e-07,
      "loss": 0.73967493,
      "num_input_tokens_seen": 320619825,
      "step": 14866,
      "time_per_iteration": 2.6329727172851562
    },
    {
      "auxiliary_loss_clip": 0.01094559,
      "auxiliary_loss_mlp": 0.01030552,
      "balance_loss_clip": 1.03526616,
      "balance_loss_mlp": 1.01760602,
      "epoch": 0.8938523974147001,
      "flos": 25664386775040.0,
      "grad_norm": 1.6146363515488438,
      "language_loss": 0.84189999,
      "learning_rate": 1.1697132954231869e-07,
      "loss": 0.86315107,
      "num_input_tokens_seen": 320638515,
      "step": 14867,
      "time_per_iteration": 2.5000522136688232
    },
    {
      "auxiliary_loss_clip": 0.01091298,
      "auxiliary_loss_mlp": 0.01026136,
      "balance_loss_clip": 1.03365493,
      "balance_loss_mlp": 1.0155859,
      "epoch": 0.8939125206673681,
      "flos": 25742852035200.0,
      "grad_norm": 1.7575772364006854,
      "language_loss": 0.80623162,
      "learning_rate": 1.168401272009567e-07,
      "loss": 0.82740599,
      "num_input_tokens_seen": 320659430,
      "step": 14868,
      "time_per_iteration": 2.542694091796875
    },
    {
      "auxiliary_loss_clip": 0.01085431,
      "auxiliary_loss_mlp": 0.01032583,
      "balance_loss_clip": 1.0357635,
      "balance_loss_mlp": 1.02065003,
      "epoch": 0.8939726439200361,
      "flos": 27344503480320.0,
      "grad_norm": 2.0494903338438837,
      "language_loss": 0.77438748,
      "learning_rate": 1.167089962692056e-07,
      "loss": 0.79556757,
      "num_input_tokens_seen": 320679295,
      "step": 14869,
      "time_per_iteration": 2.564875602722168
    },
    {
      "auxiliary_loss_clip": 0.01091952,
      "auxiliary_loss_mlp": 0.00783204,
      "balance_loss_clip": 1.03558803,
      "balance_loss_mlp": 1.01139808,
      "epoch": 0.8940327671727041,
      "flos": 20338834671360.0,
      "grad_norm": 2.3760820938832006,
      "language_loss": 0.65318108,
      "learning_rate": 1.1657793675203853e-07,
      "loss": 0.67193264,
      "num_input_tokens_seen": 320697535,
      "step": 14870,
      "time_per_iteration": 3.8619303703308105
    },
    {
      "auxiliary_loss_clip": 0.00994632,
      "auxiliary_loss_mlp": 0.01017681,
      "balance_loss_clip": 1.00991201,
      "balance_loss_mlp": 1.01629257,
      "epoch": 0.894092890425372,
      "flos": 58410573235200.0,
      "grad_norm": 0.7921533576190782,
      "language_loss": 0.55951607,
      "learning_rate": 1.1644694865442461e-07,
      "loss": 0.57963932,
      "num_input_tokens_seen": 320758635,
      "step": 14871,
      "time_per_iteration": 4.599390506744385
    },
    {
      "auxiliary_loss_clip": 0.01087089,
      "auxiliary_loss_mlp": 0.01031941,
      "balance_loss_clip": 1.03538799,
      "balance_loss_mlp": 1.02077687,
      "epoch": 0.89415301367804,
      "flos": 19829657427840.0,
      "grad_norm": 1.9545089171385395,
      "language_loss": 0.76684487,
      "learning_rate": 1.16316031981331e-07,
      "loss": 0.78803515,
      "num_input_tokens_seen": 320777175,
      "step": 14872,
      "time_per_iteration": 2.4769339561462402
    },
    {
      "auxiliary_loss_clip": 0.01088594,
      "auxiliary_loss_mlp": 0.01026053,
      "balance_loss_clip": 1.03450215,
      "balance_loss_mlp": 1.01517522,
      "epoch": 0.8942131369307079,
      "flos": 25775781828480.0,
      "grad_norm": 1.5420986092420024,
      "language_loss": 0.6726414,
      "learning_rate": 1.1618518673772215e-07,
      "loss": 0.69378787,
      "num_input_tokens_seen": 320797670,
      "step": 14873,
      "time_per_iteration": 3.932002544403076
    },
    {
      "auxiliary_loss_clip": 0.01101003,
      "auxiliary_loss_mlp": 0.01033557,
      "balance_loss_clip": 1.03478074,
      "balance_loss_mlp": 1.02188027,
      "epoch": 0.8942732601833759,
      "flos": 23149024139520.0,
      "grad_norm": 1.5747460232468395,
      "language_loss": 0.59560752,
      "learning_rate": 1.1605441292856033e-07,
      "loss": 0.61695313,
      "num_input_tokens_seen": 320817410,
      "step": 14874,
      "time_per_iteration": 2.469344139099121
    },
    {
      "auxiliary_loss_clip": 0.01074261,
      "auxiliary_loss_mlp": 0.01028796,
      "balance_loss_clip": 1.03692389,
      "balance_loss_mlp": 1.01636839,
      "epoch": 0.8943333834360438,
      "flos": 27855548231040.0,
      "grad_norm": 1.9742061437336478,
      "language_loss": 0.75722289,
      "learning_rate": 1.1592371055880356e-07,
      "loss": 0.77825344,
      "num_input_tokens_seen": 320836745,
      "step": 14875,
      "time_per_iteration": 2.569679021835327
    },
    {
      "auxiliary_loss_clip": 0.01073631,
      "auxiliary_loss_mlp": 0.01031098,
      "balance_loss_clip": 1.03467774,
      "balance_loss_mlp": 1.0173769,
      "epoch": 0.8943935066887119,
      "flos": 22163958581760.0,
      "grad_norm": 3.272810959356292,
      "language_loss": 0.77168202,
      "learning_rate": 1.1579307963340857e-07,
      "loss": 0.79272932,
      "num_input_tokens_seen": 320853305,
      "step": 14876,
      "time_per_iteration": 2.564401149749756
    },
    {
      "auxiliary_loss_clip": 0.01091658,
      "auxiliary_loss_mlp": 0.01025427,
      "balance_loss_clip": 1.03508711,
      "balance_loss_mlp": 1.01473379,
      "epoch": 0.8944536299413798,
      "flos": 21470056669440.0,
      "grad_norm": 2.466172652412905,
      "language_loss": 0.78542805,
      "learning_rate": 1.156625201573287e-07,
      "loss": 0.80659896,
      "num_input_tokens_seen": 320872885,
      "step": 14877,
      "time_per_iteration": 2.4794716835021973
    },
    {
      "auxiliary_loss_clip": 0.01056599,
      "auxiliary_loss_mlp": 0.01036752,
      "balance_loss_clip": 1.03301215,
      "balance_loss_mlp": 1.02276278,
      "epoch": 0.8945137531940478,
      "flos": 17748777703680.0,
      "grad_norm": 2.7527099771481884,
      "language_loss": 0.75371665,
      "learning_rate": 1.155320321355151e-07,
      "loss": 0.77465016,
      "num_input_tokens_seen": 320889755,
      "step": 14878,
      "time_per_iteration": 2.561741828918457
    },
    {
      "auxiliary_loss_clip": 0.01088527,
      "auxiliary_loss_mlp": 0.01030025,
      "balance_loss_clip": 1.03282487,
      "balance_loss_mlp": 1.01599383,
      "epoch": 0.8945738764467158,
      "flos": 21142264129920.0,
      "grad_norm": 1.5637058313232388,
      "language_loss": 0.75500488,
      "learning_rate": 1.1540161557291539e-07,
      "loss": 0.7761904,
      "num_input_tokens_seen": 320907860,
      "step": 14879,
      "time_per_iteration": 2.4788007736206055
    },
    {
      "auxiliary_loss_clip": 0.01073842,
      "auxiliary_loss_mlp": 0.01031229,
      "balance_loss_clip": 1.03611088,
      "balance_loss_mlp": 1.01889658,
      "epoch": 0.8946339996993837,
      "flos": 14903000835840.0,
      "grad_norm": 1.8742163262391849,
      "language_loss": 0.74288291,
      "learning_rate": 1.1527127047447538e-07,
      "loss": 0.7639336,
      "num_input_tokens_seen": 320925825,
      "step": 14880,
      "time_per_iteration": 3.9662015438079834
    },
    {
      "auxiliary_loss_clip": 0.010874,
      "auxiliary_loss_mlp": 0.01030886,
      "balance_loss_clip": 1.03326786,
      "balance_loss_mlp": 1.01794553,
      "epoch": 0.8946941229520518,
      "flos": 27382173868800.0,
      "grad_norm": 1.6338641725103074,
      "language_loss": 0.82851315,
      "learning_rate": 1.1514099684513822e-07,
      "loss": 0.84969604,
      "num_input_tokens_seen": 320946165,
      "step": 14881,
      "time_per_iteration": 2.5516934394836426
    },
    {
      "auxiliary_loss_clip": 0.01063454,
      "auxiliary_loss_mlp": 0.00782329,
      "balance_loss_clip": 1.03386724,
      "balance_loss_mlp": 1.00796986,
      "epoch": 0.8947542462047197,
      "flos": 31796277338880.0,
      "grad_norm": 1.7129333419580968,
      "language_loss": 0.67673606,
      "learning_rate": 1.1501079468984287e-07,
      "loss": 0.69519389,
      "num_input_tokens_seen": 320969330,
      "step": 14882,
      "time_per_iteration": 2.6245100498199463
    },
    {
      "auxiliary_loss_clip": 0.01081226,
      "auxiliary_loss_mlp": 0.01034417,
      "balance_loss_clip": 1.03311849,
      "balance_loss_mlp": 1.02017713,
      "epoch": 0.8948143694573877,
      "flos": 20883599314560.0,
      "grad_norm": 2.156205644802734,
      "language_loss": 0.75254667,
      "learning_rate": 1.1488066401352691e-07,
      "loss": 0.7737031,
      "num_input_tokens_seen": 320985055,
      "step": 14883,
      "time_per_iteration": 2.505072832107544
    },
    {
      "auxiliary_loss_clip": 0.01077905,
      "auxiliary_loss_mlp": 0.01034362,
      "balance_loss_clip": 1.03374958,
      "balance_loss_mlp": 1.02268553,
      "epoch": 0.8948744927100556,
      "flos": 28215552291840.0,
      "grad_norm": 1.5848443854668068,
      "language_loss": 0.72291565,
      "learning_rate": 1.147506048211253e-07,
      "loss": 0.74403834,
      "num_input_tokens_seen": 321004720,
      "step": 14884,
      "time_per_iteration": 2.5580997467041016
    },
    {
      "auxiliary_loss_clip": 0.01072305,
      "auxiliary_loss_mlp": 0.01026911,
      "balance_loss_clip": 1.0305723,
      "balance_loss_mlp": 1.01550913,
      "epoch": 0.8949346159627236,
      "flos": 21902672073600.0,
      "grad_norm": 1.7026129818959745,
      "language_loss": 0.75601816,
      "learning_rate": 1.1462061711756987e-07,
      "loss": 0.77701032,
      "num_input_tokens_seen": 321022350,
      "step": 14885,
      "time_per_iteration": 2.524350881576538
    },
    {
      "auxiliary_loss_clip": 0.01075501,
      "auxiliary_loss_mlp": 0.01030025,
      "balance_loss_clip": 1.03342247,
      "balance_loss_mlp": 1.01745999,
      "epoch": 0.8949947392153915,
      "flos": 21359128492800.0,
      "grad_norm": 1.901712751641727,
      "language_loss": 0.81681651,
      "learning_rate": 1.1449070090778911e-07,
      "loss": 0.83787179,
      "num_input_tokens_seen": 321040450,
      "step": 14886,
      "time_per_iteration": 2.506679058074951
    },
    {
      "auxiliary_loss_clip": 0.01043575,
      "auxiliary_loss_mlp": 0.01029257,
      "balance_loss_clip": 1.03392661,
      "balance_loss_mlp": 1.01772916,
      "epoch": 0.8950548624680595,
      "flos": 52445342799360.0,
      "grad_norm": 2.164505619426018,
      "language_loss": 0.63979816,
      "learning_rate": 1.1436085619671043e-07,
      "loss": 0.66052651,
      "num_input_tokens_seen": 321063970,
      "step": 14887,
      "time_per_iteration": 2.9183855056762695
    },
    {
      "auxiliary_loss_clip": 0.0108385,
      "auxiliary_loss_mlp": 0.01036163,
      "balance_loss_clip": 1.03371668,
      "balance_loss_mlp": 1.02417064,
      "epoch": 0.8951149857207275,
      "flos": 20121323863680.0,
      "grad_norm": 1.6342875615338222,
      "language_loss": 0.60700512,
      "learning_rate": 1.1423108298925698e-07,
      "loss": 0.6282053,
      "num_input_tokens_seen": 321083840,
      "step": 14888,
      "time_per_iteration": 2.508223295211792
    },
    {
      "auxiliary_loss_clip": 0.01104608,
      "auxiliary_loss_mlp": 0.01026475,
      "balance_loss_clip": 1.0345968,
      "balance_loss_mlp": 1.01510274,
      "epoch": 0.8951751089733955,
      "flos": 29862631463040.0,
      "grad_norm": 2.1892152822891404,
      "language_loss": 0.70059121,
      "learning_rate": 1.1410138129034952e-07,
      "loss": 0.72190201,
      "num_input_tokens_seen": 321104165,
      "step": 14889,
      "time_per_iteration": 2.514225482940674
    },
    {
      "auxiliary_loss_clip": 0.01091018,
      "auxiliary_loss_mlp": 0.00782889,
      "balance_loss_clip": 1.03550076,
      "balance_loss_mlp": 1.00715363,
      "epoch": 0.8952352322260634,
      "flos": 15262789415040.0,
      "grad_norm": 2.514500359700935,
      "language_loss": 0.71577203,
      "learning_rate": 1.1397175110490676e-07,
      "loss": 0.73451102,
      "num_input_tokens_seen": 321117290,
      "step": 14890,
      "time_per_iteration": 2.4307191371917725
    },
    {
      "auxiliary_loss_clip": 0.01020698,
      "auxiliary_loss_mlp": 0.00785237,
      "balance_loss_clip": 1.02995813,
      "balance_loss_mlp": 1.00816631,
      "epoch": 0.8952953554787314,
      "flos": 26798338206720.0,
      "grad_norm": 1.5560712752273933,
      "language_loss": 0.75710559,
      "learning_rate": 1.1384219243784454e-07,
      "loss": 0.77516496,
      "num_input_tokens_seen": 321137115,
      "step": 14891,
      "time_per_iteration": 2.858551263809204
    },
    {
      "auxiliary_loss_clip": 0.01045893,
      "auxiliary_loss_mlp": 0.01031765,
      "balance_loss_clip": 1.03211904,
      "balance_loss_mlp": 1.01981425,
      "epoch": 0.8953554787313994,
      "flos": 14137205852160.0,
      "grad_norm": 1.7846618635341365,
      "language_loss": 0.76554042,
      "learning_rate": 1.1371270529407517e-07,
      "loss": 0.78631699,
      "num_input_tokens_seen": 321154490,
      "step": 14892,
      "time_per_iteration": 3.0761544704437256
    },
    {
      "auxiliary_loss_clip": 0.01088456,
      "auxiliary_loss_mlp": 0.01028481,
      "balance_loss_clip": 1.03678679,
      "balance_loss_mlp": 1.01704252,
      "epoch": 0.8954156019840673,
      "flos": 25703314139520.0,
      "grad_norm": 1.4420218133113842,
      "language_loss": 0.81827474,
      "learning_rate": 1.1358328967850895e-07,
      "loss": 0.83944416,
      "num_input_tokens_seen": 321175625,
      "step": 14893,
      "time_per_iteration": 2.5501554012298584
    },
    {
      "auxiliary_loss_clip": 0.01064913,
      "auxiliary_loss_mlp": 0.01030558,
      "balance_loss_clip": 1.03345382,
      "balance_loss_mlp": 1.01879191,
      "epoch": 0.8954757252367354,
      "flos": 21907987286400.0,
      "grad_norm": 1.7022940111614189,
      "language_loss": 0.74747407,
      "learning_rate": 1.1345394559605348e-07,
      "loss": 0.7684288,
      "num_input_tokens_seen": 321193895,
      "step": 14894,
      "time_per_iteration": 2.537607431411743
    },
    {
      "auxiliary_loss_clip": 0.01092899,
      "auxiliary_loss_mlp": 0.01032227,
      "balance_loss_clip": 1.03778112,
      "balance_loss_mlp": 1.02008569,
      "epoch": 0.8955358484894033,
      "flos": 12970396454400.0,
      "grad_norm": 1.8798652602460064,
      "language_loss": 0.6634686,
      "learning_rate": 1.1332467305161352e-07,
      "loss": 0.68471986,
      "num_input_tokens_seen": 321211610,
      "step": 14895,
      "time_per_iteration": 2.453655958175659
    },
    {
      "auxiliary_loss_clip": 0.01093571,
      "auxiliary_loss_mlp": 0.010277,
      "balance_loss_clip": 1.03477991,
      "balance_loss_mlp": 1.01511121,
      "epoch": 0.8955959717420713,
      "flos": 17273966797440.0,
      "grad_norm": 1.516262571728655,
      "language_loss": 0.67231232,
      "learning_rate": 1.1319547205009094e-07,
      "loss": 0.69352502,
      "num_input_tokens_seen": 321229805,
      "step": 14896,
      "time_per_iteration": 2.4464073181152344
    },
    {
      "auxiliary_loss_clip": 0.01093815,
      "auxiliary_loss_mlp": 0.01030243,
      "balance_loss_clip": 1.03583741,
      "balance_loss_mlp": 1.01858401,
      "epoch": 0.8956560949947392,
      "flos": 14793868339200.0,
      "grad_norm": 1.82499769966295,
      "language_loss": 0.75978434,
      "learning_rate": 1.1306634259638492e-07,
      "loss": 0.78102493,
      "num_input_tokens_seen": 321247165,
      "step": 14897,
      "time_per_iteration": 2.4658544063568115
    },
    {
      "auxiliary_loss_clip": 0.00995537,
      "auxiliary_loss_mlp": 0.00765352,
      "balance_loss_clip": 1.01072085,
      "balance_loss_mlp": 1.00093174,
      "epoch": 0.8957162182474072,
      "flos": 63607817957760.0,
      "grad_norm": 0.7584300234045035,
      "language_loss": 0.55354077,
      "learning_rate": 1.129372846953931e-07,
      "loss": 0.57114965,
      "num_input_tokens_seen": 321308425,
      "step": 14898,
      "time_per_iteration": 3.226452350616455
    },
    {
      "auxiliary_loss_clip": 0.01104551,
      "auxiliary_loss_mlp": 0.00782805,
      "balance_loss_clip": 1.03543925,
      "balance_loss_mlp": 1.00856161,
      "epoch": 0.8957763415000751,
      "flos": 25009843190400.0,
      "grad_norm": 1.6207109124741392,
      "language_loss": 0.70129156,
      "learning_rate": 1.12808298352008e-07,
      "loss": 0.72016501,
      "num_input_tokens_seen": 321329295,
      "step": 14899,
      "time_per_iteration": 2.4902055263519287
    },
    {
      "auxiliary_loss_clip": 0.01050222,
      "auxiliary_loss_mlp": 0.01036999,
      "balance_loss_clip": 1.03672993,
      "balance_loss_mlp": 1.02302742,
      "epoch": 0.8958364647527431,
      "flos": 19828615933440.0,
      "grad_norm": 1.6303206770393293,
      "language_loss": 0.73855937,
      "learning_rate": 1.1267938357112106e-07,
      "loss": 0.7594316,
      "num_input_tokens_seen": 321347580,
      "step": 14900,
      "time_per_iteration": 2.6083016395568848
    },
    {
      "auxiliary_loss_clip": 0.00998396,
      "auxiliary_loss_mlp": 0.01000605,
      "balance_loss_clip": 1.01396787,
      "balance_loss_mlp": 0.99947804,
      "epoch": 0.895896588005411,
      "flos": 65537190115200.0,
      "grad_norm": 0.7915035142093432,
      "language_loss": 0.61805767,
      "learning_rate": 1.1255054035762124e-07,
      "loss": 0.6380477,
      "num_input_tokens_seen": 321407820,
      "step": 14901,
      "time_per_iteration": 3.171571731567383
    },
    {
      "auxiliary_loss_clip": 0.01092202,
      "auxiliary_loss_mlp": 0.01027978,
      "balance_loss_clip": 1.03333306,
      "balance_loss_mlp": 1.01638472,
      "epoch": 0.8959567112580791,
      "flos": 25591021246080.0,
      "grad_norm": 1.6659568878874844,
      "language_loss": 0.70828998,
      "learning_rate": 1.1242176871639441e-07,
      "loss": 0.72949177,
      "num_input_tokens_seen": 321426745,
      "step": 14902,
      "time_per_iteration": 2.543672800064087
    },
    {
      "auxiliary_loss_clip": 0.01074814,
      "auxiliary_loss_mlp": 0.01026438,
      "balance_loss_clip": 1.03274465,
      "balance_loss_mlp": 1.01546502,
      "epoch": 0.896016834510747,
      "flos": 24201780877440.0,
      "grad_norm": 1.7219344244828183,
      "language_loss": 0.78291976,
      "learning_rate": 1.1229306865232313e-07,
      "loss": 0.80393225,
      "num_input_tokens_seen": 321446165,
      "step": 14903,
      "time_per_iteration": 2.544766902923584
    },
    {
      "auxiliary_loss_clip": 0.01080285,
      "auxiliary_loss_mlp": 0.01032087,
      "balance_loss_clip": 1.03468609,
      "balance_loss_mlp": 1.01923037,
      "epoch": 0.896076957763415,
      "flos": 23075945919360.0,
      "grad_norm": 1.7657251633702402,
      "language_loss": 0.72845972,
      "learning_rate": 1.121644401702877e-07,
      "loss": 0.74958336,
      "num_input_tokens_seen": 321465285,
      "step": 14904,
      "time_per_iteration": 2.541011095046997
    },
    {
      "auxiliary_loss_clip": 0.01093459,
      "auxiliary_loss_mlp": 0.0102508,
      "balance_loss_clip": 1.03450251,
      "balance_loss_mlp": 1.01184201,
      "epoch": 0.8961370810160829,
      "flos": 22236605838720.0,
      "grad_norm": 1.9840795432214966,
      "language_loss": 0.751073,
      "learning_rate": 1.12035883275166e-07,
      "loss": 0.7722584,
      "num_input_tokens_seen": 321483670,
      "step": 14905,
      "time_per_iteration": 2.4886460304260254
    },
    {
      "auxiliary_loss_clip": 0.01091372,
      "auxiliary_loss_mlp": 0.01031282,
      "balance_loss_clip": 1.03393078,
      "balance_loss_mlp": 1.01927137,
      "epoch": 0.8961972042687509,
      "flos": 23072318645760.0,
      "grad_norm": 1.6339552311754262,
      "language_loss": 0.76265287,
      "learning_rate": 1.1190739797183279e-07,
      "loss": 0.78387952,
      "num_input_tokens_seen": 321501190,
      "step": 14906,
      "time_per_iteration": 2.556986093521118
    },
    {
      "auxiliary_loss_clip": 0.01093791,
      "auxiliary_loss_mlp": 0.01031804,
      "balance_loss_clip": 1.03634739,
      "balance_loss_mlp": 1.01963842,
      "epoch": 0.896257327521419,
      "flos": 18185882307840.0,
      "grad_norm": 1.5985911353572508,
      "language_loss": 0.74397635,
      "learning_rate": 1.1177898426515996e-07,
      "loss": 0.76523232,
      "num_input_tokens_seen": 321518540,
      "step": 14907,
      "time_per_iteration": 2.438049554824829
    },
    {
      "auxiliary_loss_clip": 0.01093961,
      "auxiliary_loss_mlp": 0.01033857,
      "balance_loss_clip": 1.03586841,
      "balance_loss_mlp": 1.02220416,
      "epoch": 0.8963174507740869,
      "flos": 17895472848000.0,
      "grad_norm": 1.631770494967735,
      "language_loss": 0.82798779,
      "learning_rate": 1.1165064216001785e-07,
      "loss": 0.84926593,
      "num_input_tokens_seen": 321536555,
      "step": 14908,
      "time_per_iteration": 2.4582738876342773
    },
    {
      "auxiliary_loss_clip": 0.01081062,
      "auxiliary_loss_mlp": 0.01029255,
      "balance_loss_clip": 1.03565669,
      "balance_loss_mlp": 1.01654768,
      "epoch": 0.8963775740267549,
      "flos": 21032269706880.0,
      "grad_norm": 1.914796051122056,
      "language_loss": 0.70565665,
      "learning_rate": 1.1152237166127232e-07,
      "loss": 0.72675979,
      "num_input_tokens_seen": 321557655,
      "step": 14909,
      "time_per_iteration": 5.589207649230957
    },
    {
      "auxiliary_loss_clip": 0.0106832,
      "auxiliary_loss_mlp": 0.01033156,
      "balance_loss_clip": 1.03434527,
      "balance_loss_mlp": 1.02120519,
      "epoch": 0.8964376972794228,
      "flos": 23179619548800.0,
      "grad_norm": 1.8030729693883132,
      "language_loss": 0.7251488,
      "learning_rate": 1.113941727737877e-07,
      "loss": 0.74616355,
      "num_input_tokens_seen": 321576160,
      "step": 14910,
      "time_per_iteration": 2.5329909324645996
    },
    {
      "auxiliary_loss_clip": 0.01089517,
      "auxiliary_loss_mlp": 0.01028385,
      "balance_loss_clip": 1.03300393,
      "balance_loss_mlp": 1.01710773,
      "epoch": 0.8964978205320908,
      "flos": 24972998814720.0,
      "grad_norm": 2.033826570978709,
      "language_loss": 0.63619328,
      "learning_rate": 1.1126604550242502e-07,
      "loss": 0.65737236,
      "num_input_tokens_seen": 321596205,
      "step": 14911,
      "time_per_iteration": 3.89689040184021
    },
    {
      "auxiliary_loss_clip": 0.01085069,
      "auxiliary_loss_mlp": 0.00783252,
      "balance_loss_clip": 1.03608847,
      "balance_loss_mlp": 1.00849164,
      "epoch": 0.8965579437847587,
      "flos": 19172025273600.0,
      "grad_norm": 1.6563999525290118,
      "language_loss": 0.75196517,
      "learning_rate": 1.111379898520437e-07,
      "loss": 0.77064836,
      "num_input_tokens_seen": 321614800,
      "step": 14912,
      "time_per_iteration": 2.5052449703216553
    },
    {
      "auxiliary_loss_clip": 0.01077159,
      "auxiliary_loss_mlp": 0.01035689,
      "balance_loss_clip": 1.03153586,
      "balance_loss_mlp": 1.0228982,
      "epoch": 0.8966180670374267,
      "flos": 24276690691200.0,
      "grad_norm": 1.6996771890102855,
      "language_loss": 0.8228026,
      "learning_rate": 1.1101000582749876e-07,
      "loss": 0.84393108,
      "num_input_tokens_seen": 321633445,
      "step": 14913,
      "time_per_iteration": 2.5224111080169678
    },
    {
      "auxiliary_loss_clip": 0.01094705,
      "auxiliary_loss_mlp": 0.01032503,
      "balance_loss_clip": 1.03521538,
      "balance_loss_mlp": 1.01931858,
      "epoch": 0.8966781902900947,
      "flos": 13553190622080.0,
      "grad_norm": 2.20976832711564,
      "language_loss": 0.61292732,
      "learning_rate": 1.1088209343364407e-07,
      "loss": 0.63419944,
      "num_input_tokens_seen": 321650890,
      "step": 14914,
      "time_per_iteration": 2.463059663772583
    },
    {
      "auxiliary_loss_clip": 0.01011036,
      "auxiliary_loss_mlp": 0.00999116,
      "balance_loss_clip": 1.01130724,
      "balance_loss_mlp": 0.99789423,
      "epoch": 0.8967383135427627,
      "flos": 65066114223360.0,
      "grad_norm": 1.6962184132780993,
      "language_loss": 0.55087483,
      "learning_rate": 1.1075425267532956e-07,
      "loss": 0.57097638,
      "num_input_tokens_seen": 321710960,
      "step": 14915,
      "time_per_iteration": 3.1209776401519775
    },
    {
      "auxiliary_loss_clip": 0.01064333,
      "auxiliary_loss_mlp": 0.01026234,
      "balance_loss_clip": 1.03251982,
      "balance_loss_mlp": 1.01523733,
      "epoch": 0.8967984367954306,
      "flos": 29713027317120.0,
      "grad_norm": 1.4651704287147678,
      "language_loss": 0.7154392,
      "learning_rate": 1.1062648355740289e-07,
      "loss": 0.73634493,
      "num_input_tokens_seen": 321733290,
      "step": 14916,
      "time_per_iteration": 2.613931655883789
    },
    {
      "auxiliary_loss_clip": 0.01081589,
      "auxiliary_loss_mlp": 0.01031557,
      "balance_loss_clip": 1.03379321,
      "balance_loss_mlp": 1.02026796,
      "epoch": 0.8968585600480986,
      "flos": 25702488126720.0,
      "grad_norm": 1.697507788081029,
      "language_loss": 0.77776289,
      "learning_rate": 1.1049878608470931e-07,
      "loss": 0.79889429,
      "num_input_tokens_seen": 321753120,
      "step": 14917,
      "time_per_iteration": 2.537705421447754
    },
    {
      "auxiliary_loss_clip": 0.0109831,
      "auxiliary_loss_mlp": 0.01038384,
      "balance_loss_clip": 1.03735709,
      "balance_loss_mlp": 1.02565813,
      "epoch": 0.8969186833007665,
      "flos": 30044698525440.0,
      "grad_norm": 2.670073130790352,
      "language_loss": 0.68283349,
      "learning_rate": 1.1037116026209137e-07,
      "loss": 0.70420045,
      "num_input_tokens_seen": 321772840,
      "step": 14918,
      "time_per_iteration": 2.5576884746551514
    },
    {
      "auxiliary_loss_clip": 0.0105432,
      "auxiliary_loss_mlp": 0.01029037,
      "balance_loss_clip": 1.03531253,
      "balance_loss_mlp": 1.01813507,
      "epoch": 0.8969788065534345,
      "flos": 22818143030400.0,
      "grad_norm": 1.8800785229508608,
      "language_loss": 0.83759397,
      "learning_rate": 1.102436060943881e-07,
      "loss": 0.85842752,
      "num_input_tokens_seen": 321791020,
      "step": 14919,
      "time_per_iteration": 4.090195178985596
    },
    {
      "auxiliary_loss_clip": 0.01104861,
      "auxiliary_loss_mlp": 0.00784298,
      "balance_loss_clip": 1.03444541,
      "balance_loss_mlp": 1.01003861,
      "epoch": 0.8970389298061026,
      "flos": 13261488272640.0,
      "grad_norm": 2.5473216085534935,
      "language_loss": 0.72403157,
      "learning_rate": 1.1011612358643696e-07,
      "loss": 0.7429232,
      "num_input_tokens_seen": 321810075,
      "step": 14920,
      "time_per_iteration": 2.4460973739624023
    },
    {
      "auxiliary_loss_clip": 0.01090896,
      "auxiliary_loss_mlp": 0.01030726,
      "balance_loss_clip": 1.03350854,
      "balance_loss_mlp": 1.01777375,
      "epoch": 0.8970990530587705,
      "flos": 10266071345280.0,
      "grad_norm": 2.30480315781291,
      "language_loss": 0.90991235,
      "learning_rate": 1.0998871274307164e-07,
      "loss": 0.9311285,
      "num_input_tokens_seen": 321822635,
      "step": 14921,
      "time_per_iteration": 2.442185401916504
    },
    {
      "auxiliary_loss_clip": 0.01046336,
      "auxiliary_loss_mlp": 0.01034299,
      "balance_loss_clip": 1.03136134,
      "balance_loss_mlp": 1.02061391,
      "epoch": 0.8971591763114385,
      "flos": 20302708567680.0,
      "grad_norm": 1.9069253135195177,
      "language_loss": 0.7356658,
      "learning_rate": 1.0986137356912384e-07,
      "loss": 0.75647211,
      "num_input_tokens_seen": 321841130,
      "step": 14922,
      "time_per_iteration": 2.5863051414489746
    },
    {
      "auxiliary_loss_clip": 0.01056112,
      "auxiliary_loss_mlp": 0.01034441,
      "balance_loss_clip": 1.03213239,
      "balance_loss_mlp": 1.02125692,
      "epoch": 0.8972192995641064,
      "flos": 23257043314560.0,
      "grad_norm": 1.906300080851892,
      "language_loss": 0.70293438,
      "learning_rate": 1.097341060694219e-07,
      "loss": 0.72383988,
      "num_input_tokens_seen": 321859855,
      "step": 14923,
      "time_per_iteration": 2.5943427085876465
    },
    {
      "auxiliary_loss_clip": 0.01082272,
      "auxiliary_loss_mlp": 0.01028885,
      "balance_loss_clip": 1.03391576,
      "balance_loss_mlp": 1.01573026,
      "epoch": 0.8972794228167744,
      "flos": 18369601395840.0,
      "grad_norm": 2.1491517896911616,
      "language_loss": 0.70808691,
      "learning_rate": 1.0960691024879221e-07,
      "loss": 0.72919852,
      "num_input_tokens_seen": 321877990,
      "step": 14924,
      "time_per_iteration": 2.5203096866607666
    },
    {
      "auxiliary_loss_clip": 0.0109004,
      "auxiliary_loss_mlp": 0.01030063,
      "balance_loss_clip": 1.03274882,
      "balance_loss_mlp": 1.01889849,
      "epoch": 0.8973395460694423,
      "flos": 23952058548480.0,
      "grad_norm": 1.3663670941970738,
      "language_loss": 0.72162521,
      "learning_rate": 1.0947978611205844e-07,
      "loss": 0.74282622,
      "num_input_tokens_seen": 321898120,
      "step": 14925,
      "time_per_iteration": 2.537475824356079
    },
    {
      "auxiliary_loss_clip": 0.01082877,
      "auxiliary_loss_mlp": 0.00787082,
      "balance_loss_clip": 1.03559148,
      "balance_loss_mlp": 1.01054478,
      "epoch": 0.8973996693221103,
      "flos": 24970843998720.0,
      "grad_norm": 1.6139766999250178,
      "language_loss": 0.82468092,
      "learning_rate": 1.0935273366404008e-07,
      "loss": 0.84338051,
      "num_input_tokens_seen": 321918140,
      "step": 14926,
      "time_per_iteration": 2.538933038711548
    },
    {
      "auxiliary_loss_clip": 0.01059416,
      "auxiliary_loss_mlp": 0.01032578,
      "balance_loss_clip": 1.03274238,
      "balance_loss_mlp": 1.02053785,
      "epoch": 0.8974597925747783,
      "flos": 25738937452800.0,
      "grad_norm": 1.4567631743618752,
      "language_loss": 0.7926842,
      "learning_rate": 1.092257529095555e-07,
      "loss": 0.81360418,
      "num_input_tokens_seen": 321938580,
      "step": 14927,
      "time_per_iteration": 2.6261584758758545
    },
    {
      "auxiliary_loss_clip": 0.01076432,
      "auxiliary_loss_mlp": 0.0102765,
      "balance_loss_clip": 1.03272343,
      "balance_loss_mlp": 1.01626599,
      "epoch": 0.8975199158274463,
      "flos": 38071918131840.0,
      "grad_norm": 1.5910959763461041,
      "language_loss": 0.66273493,
      "learning_rate": 1.0909884385341994e-07,
      "loss": 0.68377578,
      "num_input_tokens_seen": 321961135,
      "step": 14928,
      "time_per_iteration": 2.6442508697509766
    },
    {
      "auxiliary_loss_clip": 0.0108472,
      "auxiliary_loss_mlp": 0.01039698,
      "balance_loss_clip": 1.03516281,
      "balance_loss_mlp": 1.02474284,
      "epoch": 0.8975800390801142,
      "flos": 25411683617280.0,
      "grad_norm": 3.1343671130295525,
      "language_loss": 0.71413296,
      "learning_rate": 1.0897200650044602e-07,
      "loss": 0.73537713,
      "num_input_tokens_seen": 321980945,
      "step": 14929,
      "time_per_iteration": 2.5720274448394775
    },
    {
      "auxiliary_loss_clip": 0.01083978,
      "auxiliary_loss_mlp": 0.01031021,
      "balance_loss_clip": 1.0360589,
      "balance_loss_mlp": 1.01984477,
      "epoch": 0.8976401623327822,
      "flos": 21759604202880.0,
      "grad_norm": 1.8459890686610967,
      "language_loss": 0.67959613,
      "learning_rate": 1.0884524085544256e-07,
      "loss": 0.70074606,
      "num_input_tokens_seen": 322000350,
      "step": 14930,
      "time_per_iteration": 2.509366035461426
    },
    {
      "auxiliary_loss_clip": 0.0107366,
      "auxiliary_loss_mlp": 0.0103125,
      "balance_loss_clip": 1.0323112,
      "balance_loss_mlp": 1.01927507,
      "epoch": 0.8977002855854501,
      "flos": 13845323934720.0,
      "grad_norm": 1.6305659523365066,
      "language_loss": 0.75104696,
      "learning_rate": 1.0871854692321769e-07,
      "loss": 0.77209604,
      "num_input_tokens_seen": 322018980,
      "step": 14931,
      "time_per_iteration": 2.491755485534668
    },
    {
      "auxiliary_loss_clip": 0.0109376,
      "auxiliary_loss_mlp": 0.01028784,
      "balance_loss_clip": 1.03647423,
      "balance_loss_mlp": 1.01788878,
      "epoch": 0.8977604088381181,
      "flos": 19427529692160.0,
      "grad_norm": 1.5583191989167866,
      "language_loss": 0.62759066,
      "learning_rate": 1.0859192470857492e-07,
      "loss": 0.64881611,
      "num_input_tokens_seen": 322037675,
      "step": 14932,
      "time_per_iteration": 2.4594523906707764
    },
    {
      "auxiliary_loss_clip": 0.01088208,
      "auxiliary_loss_mlp": 0.01026749,
      "balance_loss_clip": 1.03449523,
      "balance_loss_mlp": 1.01623428,
      "epoch": 0.8978205320907862,
      "flos": 22742083981440.0,
      "grad_norm": 1.771447357176084,
      "language_loss": 0.71839607,
      "learning_rate": 1.0846537421631552e-07,
      "loss": 0.73954564,
      "num_input_tokens_seen": 322055130,
      "step": 14933,
      "time_per_iteration": 2.4924161434173584
    },
    {
      "auxiliary_loss_clip": 0.01060476,
      "auxiliary_loss_mlp": 0.01031289,
      "balance_loss_clip": 1.03369033,
      "balance_loss_mlp": 1.01836038,
      "epoch": 0.8978806553434541,
      "flos": 21360529123200.0,
      "grad_norm": 1.4388812692074606,
      "language_loss": 0.74556267,
      "learning_rate": 1.0833889545123898e-07,
      "loss": 0.76648033,
      "num_input_tokens_seen": 322074850,
      "step": 14934,
      "time_per_iteration": 2.5743865966796875
    },
    {
      "auxiliary_loss_clip": 0.01061464,
      "auxiliary_loss_mlp": 0.01037725,
      "balance_loss_clip": 1.0319649,
      "balance_loss_mlp": 1.02424216,
      "epoch": 0.8979407785961221,
      "flos": 20924178704640.0,
      "grad_norm": 1.7351300654218393,
      "language_loss": 0.60785824,
      "learning_rate": 1.0821248841814123e-07,
      "loss": 0.6288501,
      "num_input_tokens_seen": 322093315,
      "step": 14935,
      "time_per_iteration": 2.5479352474212646
    },
    {
      "auxiliary_loss_clip": 0.010646,
      "auxiliary_loss_mlp": 0.0102592,
      "balance_loss_clip": 1.03442073,
      "balance_loss_mlp": 1.01407671,
      "epoch": 0.89800090184879,
      "flos": 25228934196480.0,
      "grad_norm": 1.993737154480922,
      "language_loss": 0.77016199,
      "learning_rate": 1.0808615312181512e-07,
      "loss": 0.79106724,
      "num_input_tokens_seen": 322112555,
      "step": 14936,
      "time_per_iteration": 2.560037851333618
    },
    {
      "auxiliary_loss_clip": 0.01079445,
      "auxiliary_loss_mlp": 0.01032372,
      "balance_loss_clip": 1.03326845,
      "balance_loss_mlp": 1.0208267,
      "epoch": 0.898061025101458,
      "flos": 22562674525440.0,
      "grad_norm": 1.6688293397530796,
      "language_loss": 0.7408216,
      "learning_rate": 1.0795988956705193e-07,
      "loss": 0.76193976,
      "num_input_tokens_seen": 322130440,
      "step": 14937,
      "time_per_iteration": 2.5327670574188232
    },
    {
      "auxiliary_loss_clip": 0.01005227,
      "auxiliary_loss_mlp": 0.01001828,
      "balance_loss_clip": 1.01023281,
      "balance_loss_mlp": 1.00080299,
      "epoch": 0.8981211483541259,
      "flos": 56192551384320.0,
      "grad_norm": 0.8539839562118778,
      "language_loss": 0.63568944,
      "learning_rate": 1.0783369775863915e-07,
      "loss": 0.65575999,
      "num_input_tokens_seen": 322187295,
      "step": 14938,
      "time_per_iteration": 3.0436127185821533
    },
    {
      "auxiliary_loss_clip": 0.01080629,
      "auxiliary_loss_mlp": 0.01027908,
      "balance_loss_clip": 1.03483593,
      "balance_loss_mlp": 1.01606417,
      "epoch": 0.898181271606794,
      "flos": 16392718523520.0,
      "grad_norm": 1.923749728224475,
      "language_loss": 0.80356073,
      "learning_rate": 1.0770757770136251e-07,
      "loss": 0.82464617,
      "num_input_tokens_seen": 322202965,
      "step": 14939,
      "time_per_iteration": 2.50113582611084
    },
    {
      "auxiliary_loss_clip": 0.01001165,
      "auxiliary_loss_mlp": 0.01001694,
      "balance_loss_clip": 1.00956273,
      "balance_loss_mlp": 1.00062704,
      "epoch": 0.8982413948594619,
      "flos": 63440259989760.0,
      "grad_norm": 0.719408221191893,
      "language_loss": 0.52825207,
      "learning_rate": 1.0758152940000375e-07,
      "loss": 0.54828066,
      "num_input_tokens_seen": 322269490,
      "step": 14940,
      "time_per_iteration": 3.283578634262085
    },
    {
      "auxiliary_loss_clip": 0.01102899,
      "auxiliary_loss_mlp": 0.01030079,
      "balance_loss_clip": 1.03382087,
      "balance_loss_mlp": 1.01736557,
      "epoch": 0.8983015181121299,
      "flos": 21835340029440.0,
      "grad_norm": 1.9238929766138169,
      "language_loss": 0.77889073,
      "learning_rate": 1.0745555285934327e-07,
      "loss": 0.80022049,
      "num_input_tokens_seen": 322288060,
      "step": 14941,
      "time_per_iteration": 2.482966899871826
    },
    {
      "auxiliary_loss_clip": 0.01093032,
      "auxiliary_loss_mlp": 0.01035733,
      "balance_loss_clip": 1.0338279,
      "balance_loss_mlp": 1.02291179,
      "epoch": 0.8983616413647978,
      "flos": 28949961767040.0,
      "grad_norm": 2.376185222073388,
      "language_loss": 0.7362498,
      "learning_rate": 1.0732964808415834e-07,
      "loss": 0.75753748,
      "num_input_tokens_seen": 322307930,
      "step": 14942,
      "time_per_iteration": 2.542780637741089
    },
    {
      "auxiliary_loss_clip": 0.01083662,
      "auxiliary_loss_mlp": 0.01036916,
      "balance_loss_clip": 1.0326612,
      "balance_loss_mlp": 1.02386832,
      "epoch": 0.8984217646174658,
      "flos": 17785083375360.0,
      "grad_norm": 2.6579819956239903,
      "language_loss": 0.80111909,
      "learning_rate": 1.0720381507922205e-07,
      "loss": 0.82232487,
      "num_input_tokens_seen": 322326155,
      "step": 14943,
      "time_per_iteration": 2.4719245433807373
    },
    {
      "auxiliary_loss_clip": 0.01082783,
      "auxiliary_loss_mlp": 0.01035218,
      "balance_loss_clip": 1.03602338,
      "balance_loss_mlp": 1.02181923,
      "epoch": 0.8984818878701337,
      "flos": 23404528558080.0,
      "grad_norm": 1.429578109278118,
      "language_loss": 0.71511465,
      "learning_rate": 1.0707805384930701e-07,
      "loss": 0.73629463,
      "num_input_tokens_seen": 322345850,
      "step": 14944,
      "time_per_iteration": 2.547766923904419
    },
    {
      "auxiliary_loss_clip": 0.0106769,
      "auxiliary_loss_mlp": 0.01035263,
      "balance_loss_clip": 1.03421068,
      "balance_loss_mlp": 1.02189946,
      "epoch": 0.8985420111228017,
      "flos": 22346061557760.0,
      "grad_norm": 2.252209843710455,
      "language_loss": 0.75746202,
      "learning_rate": 1.0695236439918187e-07,
      "loss": 0.77849156,
      "num_input_tokens_seen": 322364715,
      "step": 14945,
      "time_per_iteration": 2.5218148231506348
    },
    {
      "auxiliary_loss_clip": 0.01108496,
      "auxiliary_loss_mlp": 0.0103243,
      "balance_loss_clip": 1.03478122,
      "balance_loss_mlp": 1.01974034,
      "epoch": 0.8986021343754698,
      "flos": 21392776558080.0,
      "grad_norm": 2.613970692297457,
      "language_loss": 0.73869663,
      "learning_rate": 1.0682674673361302e-07,
      "loss": 0.76010591,
      "num_input_tokens_seen": 322383570,
      "step": 14946,
      "time_per_iteration": 2.451948881149292
    },
    {
      "auxiliary_loss_clip": 0.01052908,
      "auxiliary_loss_mlp": 0.01029131,
      "balance_loss_clip": 1.03377545,
      "balance_loss_mlp": 1.01642346,
      "epoch": 0.8986622576281377,
      "flos": 21325372686720.0,
      "grad_norm": 1.959780006292004,
      "language_loss": 0.64632785,
      "learning_rate": 1.0670120085736334e-07,
      "loss": 0.66714823,
      "num_input_tokens_seen": 322401375,
      "step": 14947,
      "time_per_iteration": 3.9442718029022217
    },
    {
      "auxiliary_loss_clip": 0.0108139,
      "auxiliary_loss_mlp": 0.01037149,
      "balance_loss_clip": 1.03435838,
      "balance_loss_mlp": 1.02533555,
      "epoch": 0.8987223808808057,
      "flos": 23988292392960.0,
      "grad_norm": 2.0174792806356625,
      "language_loss": 0.69700742,
      "learning_rate": 1.0657572677519411e-07,
      "loss": 0.71819282,
      "num_input_tokens_seen": 322421890,
      "step": 14948,
      "time_per_iteration": 3.9402267932891846
    },
    {
      "auxiliary_loss_clip": 0.01070242,
      "auxiliary_loss_mlp": 0.01029065,
      "balance_loss_clip": 1.03298712,
      "balance_loss_mlp": 1.01718533,
      "epoch": 0.8987825041334736,
      "flos": 41500956044160.0,
      "grad_norm": 1.7803875514183096,
      "language_loss": 0.74554932,
      "learning_rate": 1.0645032449186309e-07,
      "loss": 0.76654232,
      "num_input_tokens_seen": 322445730,
      "step": 14949,
      "time_per_iteration": 2.733386516571045
    },
    {
      "auxiliary_loss_clip": 0.0106401,
      "auxiliary_loss_mlp": 0.01035281,
      "balance_loss_clip": 1.03517151,
      "balance_loss_mlp": 1.02204871,
      "epoch": 0.8988426273861416,
      "flos": 27564276844800.0,
      "grad_norm": 1.532324669199321,
      "language_loss": 0.75772929,
      "learning_rate": 1.0632499401212513e-07,
      "loss": 0.77872217,
      "num_input_tokens_seen": 322464595,
      "step": 14950,
      "time_per_iteration": 4.13750147819519
    },
    {
      "auxiliary_loss_clip": 0.01077406,
      "auxiliary_loss_mlp": 0.01030231,
      "balance_loss_clip": 1.03344655,
      "balance_loss_mlp": 1.01837564,
      "epoch": 0.8989027506388095,
      "flos": 17092653920640.0,
      "grad_norm": 1.9651300637571811,
      "language_loss": 0.66680968,
      "learning_rate": 1.0619973534073334e-07,
      "loss": 0.68788606,
      "num_input_tokens_seen": 322483305,
      "step": 14951,
      "time_per_iteration": 2.5109846591949463
    },
    {
      "auxiliary_loss_clip": 0.01105313,
      "auxiliary_loss_mlp": 0.01028459,
      "balance_loss_clip": 1.03302169,
      "balance_loss_mlp": 1.01719975,
      "epoch": 0.8989628738914776,
      "flos": 20555124416640.0,
      "grad_norm": 2.1543693946100233,
      "language_loss": 0.73883271,
      "learning_rate": 1.0607454848243769e-07,
      "loss": 0.7601704,
      "num_input_tokens_seen": 322501905,
      "step": 14952,
      "time_per_iteration": 2.4638993740081787
    },
    {
      "auxiliary_loss_clip": 0.01102806,
      "auxiliary_loss_mlp": 0.01032197,
      "balance_loss_clip": 1.03495514,
      "balance_loss_mlp": 1.0206275,
      "epoch": 0.8990229971441455,
      "flos": 16251087196800.0,
      "grad_norm": 2.467943186680906,
      "language_loss": 0.56707597,
      "learning_rate": 1.0594943344198481e-07,
      "loss": 0.58842599,
      "num_input_tokens_seen": 322518135,
      "step": 14953,
      "time_per_iteration": 2.42580246925354
    },
    {
      "auxiliary_loss_clip": 0.0108041,
      "auxiliary_loss_mlp": 0.0103486,
      "balance_loss_clip": 1.03317332,
      "balance_loss_mlp": 1.02278447,
      "epoch": 0.8990831203968135,
      "flos": 21981316901760.0,
      "grad_norm": 1.947596952155307,
      "language_loss": 0.81806362,
      "learning_rate": 1.0582439022411915e-07,
      "loss": 0.83921635,
      "num_input_tokens_seen": 322537905,
      "step": 14954,
      "time_per_iteration": 2.5226314067840576
    },
    {
      "auxiliary_loss_clip": 0.01102391,
      "auxiliary_loss_mlp": 0.01030681,
      "balance_loss_clip": 1.03538823,
      "balance_loss_mlp": 1.01877201,
      "epoch": 0.8991432436494814,
      "flos": 27447171528960.0,
      "grad_norm": 1.9429487248765867,
      "language_loss": 0.60099703,
      "learning_rate": 1.0569941883358224e-07,
      "loss": 0.6223278,
      "num_input_tokens_seen": 322557945,
      "step": 14955,
      "time_per_iteration": 2.497976541519165
    },
    {
      "auxiliary_loss_clip": 0.01092847,
      "auxiliary_loss_mlp": 0.01028641,
      "balance_loss_clip": 1.03521633,
      "balance_loss_mlp": 1.0172689,
      "epoch": 0.8992033669021494,
      "flos": 21579835610880.0,
      "grad_norm": 2.0073130222692623,
      "language_loss": 0.55252624,
      "learning_rate": 1.0557451927511341e-07,
      "loss": 0.5737412,
      "num_input_tokens_seen": 322575765,
      "step": 14956,
      "time_per_iteration": 2.5001754760742188
    },
    {
      "auxiliary_loss_clip": 0.01060876,
      "auxiliary_loss_mlp": 0.01032308,
      "balance_loss_clip": 1.03460693,
      "balance_loss_mlp": 1.02014828,
      "epoch": 0.8992634901548173,
      "flos": 28584211530240.0,
      "grad_norm": 1.7205160464646085,
      "language_loss": 0.79650295,
      "learning_rate": 1.0544969155344863e-07,
      "loss": 0.81743479,
      "num_input_tokens_seen": 322595665,
      "step": 14957,
      "time_per_iteration": 4.0569212436676025
    },
    {
      "auxiliary_loss_clip": 0.0110795,
      "auxiliary_loss_mlp": 0.01033696,
      "balance_loss_clip": 1.03687179,
      "balance_loss_mlp": 1.02122092,
      "epoch": 0.8993236134074853,
      "flos": 19867435557120.0,
      "grad_norm": 1.7390404049693857,
      "language_loss": 0.78550589,
      "learning_rate": 1.0532493567332123e-07,
      "loss": 0.80692232,
      "num_input_tokens_seen": 322614755,
      "step": 14958,
      "time_per_iteration": 2.4653921127319336
    },
    {
      "auxiliary_loss_clip": 0.01039157,
      "auxiliary_loss_mlp": 0.01026579,
      "balance_loss_clip": 1.03619671,
      "balance_loss_mlp": 1.01530194,
      "epoch": 0.8993837366601534,
      "flos": 19390649402880.0,
      "grad_norm": 1.6275820209494452,
      "language_loss": 0.74576288,
      "learning_rate": 1.0520025163946277e-07,
      "loss": 0.76642025,
      "num_input_tokens_seen": 322633425,
      "step": 14959,
      "time_per_iteration": 2.6748387813568115
    },
    {
      "auxiliary_loss_clip": 0.01099155,
      "auxiliary_loss_mlp": 0.01029479,
      "balance_loss_clip": 1.0327847,
      "balance_loss_mlp": 1.01766527,
      "epoch": 0.8994438599128213,
      "flos": 18551740285440.0,
      "grad_norm": 1.7907181167381108,
      "language_loss": 0.68326211,
      "learning_rate": 1.0507563945660015e-07,
      "loss": 0.70454842,
      "num_input_tokens_seen": 322652065,
      "step": 14960,
      "time_per_iteration": 2.72328519821167
    },
    {
      "auxiliary_loss_clip": 0.01080993,
      "auxiliary_loss_mlp": 0.01029827,
      "balance_loss_clip": 1.03572166,
      "balance_loss_mlp": 1.0185318,
      "epoch": 0.8995039831654893,
      "flos": 24427587726720.0,
      "grad_norm": 1.6154286530425233,
      "language_loss": 0.65652037,
      "learning_rate": 1.049510991294591e-07,
      "loss": 0.67762864,
      "num_input_tokens_seen": 322673275,
      "step": 14961,
      "time_per_iteration": 2.5502817630767822
    },
    {
      "auxiliary_loss_clip": 0.01079096,
      "auxiliary_loss_mlp": 0.01031225,
      "balance_loss_clip": 1.03273809,
      "balance_loss_mlp": 1.01968563,
      "epoch": 0.8995641064181572,
      "flos": 21251324799360.0,
      "grad_norm": 1.6639687922594502,
      "language_loss": 0.83223307,
      "learning_rate": 1.0482663066276254e-07,
      "loss": 0.85333622,
      "num_input_tokens_seen": 322693375,
      "step": 14962,
      "time_per_iteration": 2.53401780128479
    },
    {
      "auxiliary_loss_clip": 0.01085,
      "auxiliary_loss_mlp": 0.01030963,
      "balance_loss_clip": 1.03606164,
      "balance_loss_mlp": 1.01802862,
      "epoch": 0.8996242296708252,
      "flos": 23513661054720.0,
      "grad_norm": 1.9906901612936543,
      "language_loss": 0.76090288,
      "learning_rate": 1.047022340612298e-07,
      "loss": 0.78206253,
      "num_input_tokens_seen": 322712615,
      "step": 14963,
      "time_per_iteration": 2.513127088546753
    },
    {
      "auxiliary_loss_clip": 0.00991854,
      "auxiliary_loss_mlp": 0.01000185,
      "balance_loss_clip": 1.01617217,
      "balance_loss_mlp": 0.9990471,
      "epoch": 0.8996843529234931,
      "flos": 62403230430720.0,
      "grad_norm": 0.7758841278985987,
      "language_loss": 0.57562411,
      "learning_rate": 1.0457790932957867e-07,
      "loss": 0.59554446,
      "num_input_tokens_seen": 322766855,
      "step": 14964,
      "time_per_iteration": 3.0742242336273193
    },
    {
      "auxiliary_loss_clip": 0.01099216,
      "auxiliary_loss_mlp": 0.01032517,
      "balance_loss_clip": 1.03735971,
      "balance_loss_mlp": 1.01949978,
      "epoch": 0.8997444761761612,
      "flos": 24236829573120.0,
      "grad_norm": 2.281564391586615,
      "language_loss": 0.67691088,
      "learning_rate": 1.0445365647252269e-07,
      "loss": 0.69822818,
      "num_input_tokens_seen": 322781130,
      "step": 14965,
      "time_per_iteration": 2.488295078277588
    },
    {
      "auxiliary_loss_clip": 0.01105229,
      "auxiliary_loss_mlp": 0.01033056,
      "balance_loss_clip": 1.03511882,
      "balance_loss_mlp": 1.0214746,
      "epoch": 0.8998045994288291,
      "flos": 21361103740800.0,
      "grad_norm": 2.0757920433689376,
      "language_loss": 0.72225636,
      "learning_rate": 1.0432947549477433e-07,
      "loss": 0.74363923,
      "num_input_tokens_seen": 322800310,
      "step": 14966,
      "time_per_iteration": 2.431565761566162
    },
    {
      "auxiliary_loss_clip": 0.01072676,
      "auxiliary_loss_mlp": 0.01029902,
      "balance_loss_clip": 1.03459895,
      "balance_loss_mlp": 1.01683688,
      "epoch": 0.8998647226814971,
      "flos": 28986159697920.0,
      "grad_norm": 1.6794697825729414,
      "language_loss": 0.73042428,
      "learning_rate": 1.0420536640104205e-07,
      "loss": 0.75145006,
      "num_input_tokens_seen": 322820955,
      "step": 14967,
      "time_per_iteration": 2.6131317615509033
    },
    {
      "auxiliary_loss_clip": 0.01059741,
      "auxiliary_loss_mlp": 0.00781575,
      "balance_loss_clip": 1.03474379,
      "balance_loss_mlp": 1.00467777,
      "epoch": 0.899924845934165,
      "flos": 13625909706240.0,
      "grad_norm": 1.8236487703229451,
      "language_loss": 0.71852326,
      "learning_rate": 1.040813291960323e-07,
      "loss": 0.73693639,
      "num_input_tokens_seen": 322838780,
      "step": 14968,
      "time_per_iteration": 2.543549060821533
    },
    {
      "auxiliary_loss_clip": 0.01089172,
      "auxiliary_loss_mlp": 0.01028376,
      "balance_loss_clip": 1.03510094,
      "balance_loss_mlp": 1.01646662,
      "epoch": 0.899984969186833,
      "flos": 20882629647360.0,
      "grad_norm": 1.868854096000121,
      "language_loss": 0.71279967,
      "learning_rate": 1.0395736388444864e-07,
      "loss": 0.73397517,
      "num_input_tokens_seen": 322856710,
      "step": 14969,
      "time_per_iteration": 2.4861767292022705
    },
    {
      "auxiliary_loss_clip": 0.0110643,
      "auxiliary_loss_mlp": 0.01029739,
      "balance_loss_clip": 1.03672636,
      "balance_loss_mlp": 1.01769912,
      "epoch": 0.9000450924395009,
      "flos": 20921808407040.0,
      "grad_norm": 2.644233945459289,
      "language_loss": 0.75986695,
      "learning_rate": 1.0383347047099201e-07,
      "loss": 0.78122866,
      "num_input_tokens_seen": 322876070,
      "step": 14970,
      "time_per_iteration": 2.4307961463928223
    },
    {
      "auxiliary_loss_clip": 0.01096204,
      "auxiliary_loss_mlp": 0.0103003,
      "balance_loss_clip": 1.03541732,
      "balance_loss_mlp": 1.01857436,
      "epoch": 0.900105215692169,
      "flos": 17165049782400.0,
      "grad_norm": 1.7102600389926637,
      "language_loss": 0.72920632,
      "learning_rate": 1.0370964896035972e-07,
      "loss": 0.75046873,
      "num_input_tokens_seen": 322895095,
      "step": 14971,
      "time_per_iteration": 2.4674694538116455
    },
    {
      "auxiliary_loss_clip": 0.01066397,
      "auxiliary_loss_mlp": 0.01027925,
      "balance_loss_clip": 1.03285575,
      "balance_loss_mlp": 1.01545,
      "epoch": 0.900165338944837,
      "flos": 19931930426880.0,
      "grad_norm": 2.172001290211485,
      "language_loss": 0.81601655,
      "learning_rate": 1.035858993572476e-07,
      "loss": 0.83695978,
      "num_input_tokens_seen": 322911845,
      "step": 14972,
      "time_per_iteration": 2.5129406452178955
    },
    {
      "auxiliary_loss_clip": 0.01084699,
      "auxiliary_loss_mlp": 0.01029387,
      "balance_loss_clip": 1.03419209,
      "balance_loss_mlp": 1.01714468,
      "epoch": 0.9002254621975049,
      "flos": 16107085572480.0,
      "grad_norm": 2.222537229280011,
      "language_loss": 0.81608516,
      "learning_rate": 1.0346222166634855e-07,
      "loss": 0.83722603,
      "num_input_tokens_seen": 322928170,
      "step": 14973,
      "time_per_iteration": 2.481732130050659
    },
    {
      "auxiliary_loss_clip": 0.01103034,
      "auxiliary_loss_mlp": 0.01034291,
      "balance_loss_clip": 1.03496277,
      "balance_loss_mlp": 1.02167249,
      "epoch": 0.9002855854501729,
      "flos": 28476120528000.0,
      "grad_norm": 2.458377975583296,
      "language_loss": 0.58122873,
      "learning_rate": 1.0333861589235193e-07,
      "loss": 0.60260201,
      "num_input_tokens_seen": 322948165,
      "step": 14974,
      "time_per_iteration": 2.4812731742858887
    },
    {
      "auxiliary_loss_clip": 0.01108659,
      "auxiliary_loss_mlp": 0.01033651,
      "balance_loss_clip": 1.03925872,
      "balance_loss_mlp": 1.0216403,
      "epoch": 0.9003457087028408,
      "flos": 25630307746560.0,
      "grad_norm": 1.7965509849716224,
      "language_loss": 0.63541055,
      "learning_rate": 1.0321508203994489e-07,
      "loss": 0.65683365,
      "num_input_tokens_seen": 322968880,
      "step": 14975,
      "time_per_iteration": 2.4845187664031982
    },
    {
      "auxiliary_loss_clip": 0.0109339,
      "auxiliary_loss_mlp": 0.01030345,
      "balance_loss_clip": 1.03465056,
      "balance_loss_mlp": 1.01814353,
      "epoch": 0.9004058319555088,
      "flos": 24389414547840.0,
      "grad_norm": 1.5336930359591952,
      "language_loss": 0.73015338,
      "learning_rate": 1.0309162011381257e-07,
      "loss": 0.7513907,
      "num_input_tokens_seen": 322989395,
      "step": 14976,
      "time_per_iteration": 2.5107274055480957
    },
    {
      "auxiliary_loss_clip": 0.01096353,
      "auxiliary_loss_mlp": 0.01030157,
      "balance_loss_clip": 1.03757596,
      "balance_loss_mlp": 1.01815295,
      "epoch": 0.9004659552081767,
      "flos": 29059345658880.0,
      "grad_norm": 1.6973404121381008,
      "language_loss": 0.69506651,
      "learning_rate": 1.0296823011863565e-07,
      "loss": 0.71633166,
      "num_input_tokens_seen": 323009060,
      "step": 14977,
      "time_per_iteration": 2.5322084426879883
    },
    {
      "auxiliary_loss_clip": 0.01078332,
      "auxiliary_loss_mlp": 0.00785889,
      "balance_loss_clip": 1.03469038,
      "balance_loss_mlp": 1.01223612,
      "epoch": 0.9005260784608448,
      "flos": 16763855800320.0,
      "grad_norm": 2.4123549722284015,
      "language_loss": 0.65445167,
      "learning_rate": 1.0284491205909351e-07,
      "loss": 0.67309386,
      "num_input_tokens_seen": 323027530,
      "step": 14978,
      "time_per_iteration": 2.520141124725342
    },
    {
      "auxiliary_loss_clip": 0.01074452,
      "auxiliary_loss_mlp": 0.01036448,
      "balance_loss_clip": 1.0364573,
      "balance_loss_mlp": 1.02315652,
      "epoch": 0.9005862017135127,
      "flos": 20376002269440.0,
      "grad_norm": 1.793645346301852,
      "language_loss": 0.78891718,
      "learning_rate": 1.0272166593986286e-07,
      "loss": 0.81002617,
      "num_input_tokens_seen": 323045370,
      "step": 14979,
      "time_per_iteration": 2.5562963485717773
    },
    {
      "auxiliary_loss_clip": 0.01009997,
      "auxiliary_loss_mlp": 0.009995,
      "balance_loss_clip": 1.00735617,
      "balance_loss_mlp": 0.99849224,
      "epoch": 0.9006463249661807,
      "flos": 67580255796480.0,
      "grad_norm": 0.7248824597139121,
      "language_loss": 0.53680253,
      "learning_rate": 1.0259849176561642e-07,
      "loss": 0.55689746,
      "num_input_tokens_seen": 323105660,
      "step": 14980,
      "time_per_iteration": 3.1366372108459473
    },
    {
      "auxiliary_loss_clip": 0.01096872,
      "auxiliary_loss_mlp": 0.01036071,
      "balance_loss_clip": 1.03635406,
      "balance_loss_mlp": 1.02359581,
      "epoch": 0.9007064482188486,
      "flos": 28293335193600.0,
      "grad_norm": 2.199777135619201,
      "language_loss": 0.82206267,
      "learning_rate": 1.0247538954102553e-07,
      "loss": 0.84339207,
      "num_input_tokens_seen": 323126365,
      "step": 14981,
      "time_per_iteration": 2.5475268363952637
    },
    {
      "auxiliary_loss_clip": 0.01060937,
      "auxiliary_loss_mlp": 0.01028958,
      "balance_loss_clip": 1.03417754,
      "balance_loss_mlp": 1.01732302,
      "epoch": 0.9007665714715166,
      "flos": 21616320850560.0,
      "grad_norm": 1.480465695983654,
      "language_loss": 0.81760031,
      "learning_rate": 1.0235235927075758e-07,
      "loss": 0.83849931,
      "num_input_tokens_seen": 323145655,
      "step": 14982,
      "time_per_iteration": 2.585956573486328
    },
    {
      "auxiliary_loss_clip": 0.01072208,
      "auxiliary_loss_mlp": 0.01037919,
      "balance_loss_clip": 1.03284812,
      "balance_loss_mlp": 1.02518129,
      "epoch": 0.9008266947241845,
      "flos": 26541864120960.0,
      "grad_norm": 1.6999847733657323,
      "language_loss": 0.71830392,
      "learning_rate": 1.0222940095947885e-07,
      "loss": 0.73940516,
      "num_input_tokens_seen": 323164540,
      "step": 14983,
      "time_per_iteration": 2.579643964767456
    },
    {
      "auxiliary_loss_clip": 0.01092481,
      "auxiliary_loss_mlp": 0.01025599,
      "balance_loss_clip": 1.03590822,
      "balance_loss_mlp": 1.01489401,
      "epoch": 0.9008868179768525,
      "flos": 23110527738240.0,
      "grad_norm": 1.33424693163311,
      "language_loss": 0.7494784,
      "learning_rate": 1.0210651461185115e-07,
      "loss": 0.77065915,
      "num_input_tokens_seen": 323186960,
      "step": 14984,
      "time_per_iteration": 2.50393009185791
    },
    {
      "auxiliary_loss_clip": 0.0109925,
      "auxiliary_loss_mlp": 0.01031586,
      "balance_loss_clip": 1.03304243,
      "balance_loss_mlp": 1.01967049,
      "epoch": 0.9009469412295206,
      "flos": 19060809788160.0,
      "grad_norm": 1.500043315133158,
      "language_loss": 0.7043283,
      "learning_rate": 1.0198370023253456e-07,
      "loss": 0.7256366,
      "num_input_tokens_seen": 323206135,
      "step": 14985,
      "time_per_iteration": 4.090293645858765
    },
    {
      "auxiliary_loss_clip": 0.01081799,
      "auxiliary_loss_mlp": 0.01029984,
      "balance_loss_clip": 1.03357983,
      "balance_loss_mlp": 1.01740766,
      "epoch": 0.9010070644821885,
      "flos": 23222281927680.0,
      "grad_norm": 1.89821728942643,
      "language_loss": 0.70479637,
      "learning_rate": 1.0186095782618643e-07,
      "loss": 0.72591424,
      "num_input_tokens_seen": 323225980,
      "step": 14986,
      "time_per_iteration": 2.5960068702697754
    },
    {
      "auxiliary_loss_clip": 0.01091999,
      "auxiliary_loss_mlp": 0.01029857,
      "balance_loss_clip": 1.0327158,
      "balance_loss_mlp": 1.01744688,
      "epoch": 0.9010671877348565,
      "flos": 17384823146880.0,
      "grad_norm": 1.695721735177916,
      "language_loss": 0.76819319,
      "learning_rate": 1.0173828739746104e-07,
      "loss": 0.78941178,
      "num_input_tokens_seen": 323243700,
      "step": 14987,
      "time_per_iteration": 3.846806287765503
    },
    {
      "auxiliary_loss_clip": 0.01088122,
      "auxiliary_loss_mlp": 0.01034912,
      "balance_loss_clip": 1.0351249,
      "balance_loss_mlp": 1.02307463,
      "epoch": 0.9011273109875244,
      "flos": 21908166854400.0,
      "grad_norm": 1.6328688334891839,
      "language_loss": 0.73460424,
      "learning_rate": 1.0161568895100981e-07,
      "loss": 0.75583458,
      "num_input_tokens_seen": 323261535,
      "step": 14988,
      "time_per_iteration": 2.4645888805389404
    },
    {
      "auxiliary_loss_clip": 0.01080072,
      "auxiliary_loss_mlp": 0.01027681,
      "balance_loss_clip": 1.03633022,
      "balance_loss_mlp": 1.0149554,
      "epoch": 0.9011874342401924,
      "flos": 24060831909120.0,
      "grad_norm": 1.7587905743799241,
      "language_loss": 0.69598752,
      "learning_rate": 1.0149316249148188e-07,
      "loss": 0.71706504,
      "num_input_tokens_seen": 323281855,
      "step": 14989,
      "time_per_iteration": 3.9902853965759277
    },
    {
      "auxiliary_loss_clip": 0.01104405,
      "auxiliary_loss_mlp": 0.01029599,
      "balance_loss_clip": 1.03530157,
      "balance_loss_mlp": 1.01755881,
      "epoch": 0.9012475574928603,
      "flos": 16758791982720.0,
      "grad_norm": 1.7710943487787882,
      "language_loss": 0.80075085,
      "learning_rate": 1.0137070802352376e-07,
      "loss": 0.82209086,
      "num_input_tokens_seen": 323299505,
      "step": 14990,
      "time_per_iteration": 2.4039034843444824
    },
    {
      "auxiliary_loss_clip": 0.01068583,
      "auxiliary_loss_mlp": 0.01030277,
      "balance_loss_clip": 1.03583121,
      "balance_loss_mlp": 1.01784372,
      "epoch": 0.9013076807455284,
      "flos": 19971109186560.0,
      "grad_norm": 1.7277013480012444,
      "language_loss": 0.78029394,
      "learning_rate": 1.0124832555177842e-07,
      "loss": 0.80128253,
      "num_input_tokens_seen": 323318365,
      "step": 14991,
      "time_per_iteration": 2.5664615631103516
    },
    {
      "auxiliary_loss_clip": 0.00996908,
      "auxiliary_loss_mlp": 0.00761575,
      "balance_loss_clip": 1.00829697,
      "balance_loss_mlp": 1.00005627,
      "epoch": 0.9013678039981963,
      "flos": 65180274624000.0,
      "grad_norm": 0.7776720257691508,
      "language_loss": 0.60184216,
      "learning_rate": 1.0112601508088726e-07,
      "loss": 0.61942697,
      "num_input_tokens_seen": 323371835,
      "step": 14992,
      "time_per_iteration": 3.0569722652435303
    },
    {
      "auxiliary_loss_clip": 0.01088672,
      "auxiliary_loss_mlp": 0.01027114,
      "balance_loss_clip": 1.03490078,
      "balance_loss_mlp": 1.01500237,
      "epoch": 0.9014279272508643,
      "flos": 20521224956160.0,
      "grad_norm": 1.9991484180476893,
      "language_loss": 0.82671291,
      "learning_rate": 1.0100377661548764e-07,
      "loss": 0.84787083,
      "num_input_tokens_seen": 323388495,
      "step": 14993,
      "time_per_iteration": 2.49094557762146
    },
    {
      "auxiliary_loss_clip": 0.01103212,
      "auxiliary_loss_mlp": 0.01032868,
      "balance_loss_clip": 1.03449702,
      "balance_loss_mlp": 1.02041698,
      "epoch": 0.9014880505035322,
      "flos": 17309051406720.0,
      "grad_norm": 2.028747185581525,
      "language_loss": 0.72836822,
      "learning_rate": 1.0088161016021502e-07,
      "loss": 0.74972904,
      "num_input_tokens_seen": 323405280,
      "step": 14994,
      "time_per_iteration": 2.3895137310028076
    },
    {
      "auxiliary_loss_clip": 0.01089799,
      "auxiliary_loss_mlp": 0.01029039,
      "balance_loss_clip": 1.03422999,
      "balance_loss_mlp": 1.017923,
      "epoch": 0.9015481737562002,
      "flos": 28402862739840.0,
      "grad_norm": 1.7642293859611897,
      "language_loss": 0.6511519,
      "learning_rate": 1.0075951571970187e-07,
      "loss": 0.67234027,
      "num_input_tokens_seen": 323425310,
      "step": 14995,
      "time_per_iteration": 2.5550334453582764
    },
    {
      "auxiliary_loss_clip": 0.01065501,
      "auxiliary_loss_mlp": 0.01033607,
      "balance_loss_clip": 1.03027868,
      "balance_loss_mlp": 1.02036846,
      "epoch": 0.9016082970088681,
      "flos": 29752672953600.0,
      "grad_norm": 3.248226901141404,
      "language_loss": 0.66696167,
      "learning_rate": 1.0063749329857873e-07,
      "loss": 0.68795276,
      "num_input_tokens_seen": 323447805,
      "step": 14996,
      "time_per_iteration": 4.0160136222839355
    },
    {
      "auxiliary_loss_clip": 0.01091168,
      "auxiliary_loss_mlp": 0.01027008,
      "balance_loss_clip": 1.03343678,
      "balance_loss_mlp": 1.0157783,
      "epoch": 0.9016684202615362,
      "flos": 23513230091520.0,
      "grad_norm": 1.930619534370469,
      "language_loss": 0.65859067,
      "learning_rate": 1.0051554290147168e-07,
      "loss": 0.67977238,
      "num_input_tokens_seen": 323467150,
      "step": 14997,
      "time_per_iteration": 2.509484052658081
    },
    {
      "auxiliary_loss_clip": 0.01075032,
      "auxiliary_loss_mlp": 0.01034918,
      "balance_loss_clip": 1.03078723,
      "balance_loss_mlp": 1.02268159,
      "epoch": 0.9017285435142042,
      "flos": 16979247705600.0,
      "grad_norm": 1.7881838187175252,
      "language_loss": 0.7753762,
      "learning_rate": 1.0039366453300613e-07,
      "loss": 0.79647577,
      "num_input_tokens_seen": 323484250,
      "step": 14998,
      "time_per_iteration": 2.5007588863372803
    },
    {
      "auxiliary_loss_clip": 0.011041,
      "auxiliary_loss_mlp": 0.0103145,
      "balance_loss_clip": 1.03491712,
      "balance_loss_mlp": 1.01923132,
      "epoch": 0.9017886667668721,
      "flos": 21393351175680.0,
      "grad_norm": 1.7854655769870653,
      "language_loss": 0.75534004,
      "learning_rate": 1.0027185819780281e-07,
      "loss": 0.77669555,
      "num_input_tokens_seen": 323502910,
      "step": 14999,
      "time_per_iteration": 2.4415955543518066
    },
    {
      "auxiliary_loss_clip": 0.01048146,
      "auxiliary_loss_mlp": 0.01032193,
      "balance_loss_clip": 1.03661621,
      "balance_loss_mlp": 1.01972413,
      "epoch": 0.9018487900195401,
      "flos": 20996574566400.0,
      "grad_norm": 2.241486362802402,
      "language_loss": 0.75825673,
      "learning_rate": 1.0015012390048117e-07,
      "loss": 0.77906013,
      "num_input_tokens_seen": 323521820,
      "step": 15000,
      "time_per_iteration": 2.62198805809021
    },
    {
      "auxiliary_loss_clip": 0.01084347,
      "auxiliary_loss_mlp": 0.01025814,
      "balance_loss_clip": 1.03372884,
      "balance_loss_mlp": 1.01478148,
      "epoch": 0.901908913272208,
      "flos": 53358443458560.0,
      "grad_norm": 2.4988023321351465,
      "language_loss": 0.80832511,
      "learning_rate": 1.0002846164565704e-07,
      "loss": 0.82942665,
      "num_input_tokens_seen": 323543200,
      "step": 15001,
      "time_per_iteration": 2.7611000537872314
    },
    {
      "auxiliary_loss_clip": 0.01071867,
      "auxiliary_loss_mlp": 0.01025617,
      "balance_loss_clip": 1.03600621,
      "balance_loss_mlp": 1.01442885,
      "epoch": 0.901969036524876,
      "flos": 22089838867200.0,
      "grad_norm": 1.4064053731782085,
      "language_loss": 0.78469294,
      "learning_rate": 9.990687143794407e-08,
      "loss": 0.80566782,
      "num_input_tokens_seen": 323563075,
      "step": 15002,
      "time_per_iteration": 2.583815574645996
    },
    {
      "auxiliary_loss_clip": 0.01077109,
      "auxiliary_loss_mlp": 0.01037314,
      "balance_loss_clip": 1.03579617,
      "balance_loss_mlp": 1.02321792,
      "epoch": 0.9020291597775439,
      "flos": 23835025059840.0,
      "grad_norm": 3.22610396632346,
      "language_loss": 0.68143904,
      "learning_rate": 9.978535328195347e-08,
      "loss": 0.70258331,
      "num_input_tokens_seen": 323579065,
      "step": 15003,
      "time_per_iteration": 2.5153868198394775
    },
    {
      "auxiliary_loss_clip": 0.0108092,
      "auxiliary_loss_mlp": 0.01033775,
      "balance_loss_clip": 1.0342598,
      "balance_loss_mlp": 1.02113307,
      "epoch": 0.902089283030212,
      "flos": 18326005263360.0,
      "grad_norm": 3.298622541647494,
      "language_loss": 0.85697478,
      "learning_rate": 9.9663907182292e-08,
      "loss": 0.87812173,
      "num_input_tokens_seen": 323594835,
      "step": 15004,
      "time_per_iteration": 2.4852654933929443
    },
    {
      "auxiliary_loss_clip": 0.01070779,
      "auxiliary_loss_mlp": 0.01035012,
      "balance_loss_clip": 1.0348624,
      "balance_loss_mlp": 1.02287042,
      "epoch": 0.9021494062828799,
      "flos": 24170359455360.0,
      "grad_norm": 2.062139521995895,
      "language_loss": 0.72007138,
      "learning_rate": 9.954253314356575e-08,
      "loss": 0.74112928,
      "num_input_tokens_seen": 323611475,
      "step": 15005,
      "time_per_iteration": 2.549708604812622
    },
    {
      "auxiliary_loss_clip": 0.01094043,
      "auxiliary_loss_mlp": 0.01031096,
      "balance_loss_clip": 1.03172398,
      "balance_loss_mlp": 1.01832891,
      "epoch": 0.9022095295355479,
      "flos": 21616859554560.0,
      "grad_norm": 2.023646247586754,
      "language_loss": 0.70992249,
      "learning_rate": 9.942123117037748e-08,
      "loss": 0.73117387,
      "num_input_tokens_seen": 323629730,
      "step": 15006,
      "time_per_iteration": 2.483577013015747
    },
    {
      "auxiliary_loss_clip": 0.01084246,
      "auxiliary_loss_mlp": 0.01027959,
      "balance_loss_clip": 1.03442192,
      "balance_loss_mlp": 1.01578164,
      "epoch": 0.9022696527882158,
      "flos": 18726229578240.0,
      "grad_norm": 1.74324725869261,
      "language_loss": 0.84149039,
      "learning_rate": 9.930000126732618e-08,
      "loss": 0.86261249,
      "num_input_tokens_seen": 323646000,
      "step": 15007,
      "time_per_iteration": 2.471668004989624
    },
    {
      "auxiliary_loss_clip": 0.01074932,
      "auxiliary_loss_mlp": 0.01028919,
      "balance_loss_clip": 1.03212762,
      "balance_loss_mlp": 1.01696813,
      "epoch": 0.9023297760408838,
      "flos": 26761206522240.0,
      "grad_norm": 1.3963727150052647,
      "language_loss": 0.78559959,
      "learning_rate": 9.917884343900928e-08,
      "loss": 0.80663806,
      "num_input_tokens_seen": 323667250,
      "step": 15008,
      "time_per_iteration": 2.5911448001861572
    },
    {
      "auxiliary_loss_clip": 0.0105913,
      "auxiliary_loss_mlp": 0.0102927,
      "balance_loss_clip": 1.03484261,
      "balance_loss_mlp": 1.01780236,
      "epoch": 0.9023898992935517,
      "flos": 20522553759360.0,
      "grad_norm": 1.6959332158851823,
      "language_loss": 0.73696303,
      "learning_rate": 9.905775769002156e-08,
      "loss": 0.75784707,
      "num_input_tokens_seen": 323687150,
      "step": 15009,
      "time_per_iteration": 2.5318822860717773
    },
    {
      "auxiliary_loss_clip": 0.01101475,
      "auxiliary_loss_mlp": 0.0103269,
      "balance_loss_clip": 1.03434896,
      "balance_loss_mlp": 1.0207988,
      "epoch": 0.9024500225462198,
      "flos": 17456644391040.0,
      "grad_norm": 1.7756639426897396,
      "language_loss": 0.72960496,
      "learning_rate": 9.893674402495399e-08,
      "loss": 0.75094652,
      "num_input_tokens_seen": 323703660,
      "step": 15010,
      "time_per_iteration": 2.440182685852051
    },
    {
      "auxiliary_loss_clip": 0.01077686,
      "auxiliary_loss_mlp": 0.01028853,
      "balance_loss_clip": 1.03548479,
      "balance_loss_mlp": 1.01641953,
      "epoch": 0.9025101457988878,
      "flos": 20813609664000.0,
      "grad_norm": 1.7514192444461092,
      "language_loss": 0.7422719,
      "learning_rate": 9.881580244839538e-08,
      "loss": 0.76333731,
      "num_input_tokens_seen": 323722060,
      "step": 15011,
      "time_per_iteration": 2.500464677810669
    },
    {
      "auxiliary_loss_clip": 0.01095556,
      "auxiliary_loss_mlp": 0.01028482,
      "balance_loss_clip": 1.0338397,
      "balance_loss_mlp": 1.01597106,
      "epoch": 0.9025702690515557,
      "flos": 19026371623680.0,
      "grad_norm": 1.8696595309099802,
      "language_loss": 0.72739804,
      "learning_rate": 9.869493296493204e-08,
      "loss": 0.74863839,
      "num_input_tokens_seen": 323740645,
      "step": 15012,
      "time_per_iteration": 2.499265432357788
    },
    {
      "auxiliary_loss_clip": 0.01071788,
      "auxiliary_loss_mlp": 0.01038385,
      "balance_loss_clip": 1.03548861,
      "balance_loss_mlp": 1.02678585,
      "epoch": 0.9026303923042237,
      "flos": 19682818629120.0,
      "grad_norm": 1.535254288822677,
      "language_loss": 0.6921519,
      "learning_rate": 9.857413557914763e-08,
      "loss": 0.71325362,
      "num_input_tokens_seen": 323758905,
      "step": 15013,
      "time_per_iteration": 2.5283010005950928
    },
    {
      "auxiliary_loss_clip": 0.01087013,
      "auxiliary_loss_mlp": 0.0103147,
      "balance_loss_clip": 1.03216839,
      "balance_loss_mlp": 1.02029383,
      "epoch": 0.9026905155568916,
      "flos": 24608110504320.0,
      "grad_norm": 1.3860982737231504,
      "language_loss": 0.73191655,
      "learning_rate": 9.845341029562249e-08,
      "loss": 0.75310141,
      "num_input_tokens_seen": 323780595,
      "step": 15014,
      "time_per_iteration": 2.5347697734832764
    },
    {
      "auxiliary_loss_clip": 0.01102232,
      "auxiliary_loss_mlp": 0.01029669,
      "balance_loss_clip": 1.03373504,
      "balance_loss_mlp": 1.01728916,
      "epoch": 0.9027506388095596,
      "flos": 20521799573760.0,
      "grad_norm": 2.624407360546308,
      "language_loss": 0.72189379,
      "learning_rate": 9.833275711893474e-08,
      "loss": 0.74321282,
      "num_input_tokens_seen": 323798160,
      "step": 15015,
      "time_per_iteration": 2.4398386478424072
    },
    {
      "auxiliary_loss_clip": 0.01079597,
      "auxiliary_loss_mlp": 0.01029863,
      "balance_loss_clip": 1.03496766,
      "balance_loss_mlp": 1.01838911,
      "epoch": 0.9028107620622275,
      "flos": 22784494965120.0,
      "grad_norm": 2.142106974688478,
      "language_loss": 0.69052136,
      "learning_rate": 9.821217605365895e-08,
      "loss": 0.71161604,
      "num_input_tokens_seen": 323816810,
      "step": 15016,
      "time_per_iteration": 2.5307672023773193
    },
    {
      "auxiliary_loss_clip": 0.01101128,
      "auxiliary_loss_mlp": 0.01026773,
      "balance_loss_clip": 1.03372669,
      "balance_loss_mlp": 1.01593697,
      "epoch": 0.9028708853148956,
      "flos": 25410534382080.0,
      "grad_norm": 1.924845654412354,
      "language_loss": 0.70414418,
      "learning_rate": 9.809166710436855e-08,
      "loss": 0.72542322,
      "num_input_tokens_seen": 323836900,
      "step": 15017,
      "time_per_iteration": 2.469637870788574
    },
    {
      "auxiliary_loss_clip": 0.01081435,
      "auxiliary_loss_mlp": 0.01032488,
      "balance_loss_clip": 1.03884792,
      "balance_loss_mlp": 1.02123439,
      "epoch": 0.9029310085675635,
      "flos": 21871322478720.0,
      "grad_norm": 1.572574864794066,
      "language_loss": 0.69420946,
      "learning_rate": 9.797123027563237e-08,
      "loss": 0.71534866,
      "num_input_tokens_seen": 323855325,
      "step": 15018,
      "time_per_iteration": 2.5313761234283447
    },
    {
      "auxiliary_loss_clip": 0.01093639,
      "auxiliary_loss_mlp": 0.01031708,
      "balance_loss_clip": 1.03555465,
      "balance_loss_mlp": 1.01959682,
      "epoch": 0.9029911318202315,
      "flos": 26214394803840.0,
      "grad_norm": 1.680465253385712,
      "language_loss": 0.69103873,
      "learning_rate": 9.785086557201782e-08,
      "loss": 0.71229219,
      "num_input_tokens_seen": 323875650,
      "step": 15019,
      "time_per_iteration": 2.5163917541503906
    },
    {
      "auxiliary_loss_clip": 0.01099997,
      "auxiliary_loss_mlp": 0.01032513,
      "balance_loss_clip": 1.03365946,
      "balance_loss_mlp": 1.0211587,
      "epoch": 0.9030512550728994,
      "flos": 15961360095360.0,
      "grad_norm": 1.8335991973161119,
      "language_loss": 0.72092903,
      "learning_rate": 9.773057299808951e-08,
      "loss": 0.74225414,
      "num_input_tokens_seen": 323892920,
      "step": 15020,
      "time_per_iteration": 2.4221320152282715
    },
    {
      "auxiliary_loss_clip": 0.01089997,
      "auxiliary_loss_mlp": 0.01032114,
      "balance_loss_clip": 1.03349948,
      "balance_loss_mlp": 1.01946592,
      "epoch": 0.9031113783255674,
      "flos": 23987610034560.0,
      "grad_norm": 1.5829502399334423,
      "language_loss": 0.74386585,
      "learning_rate": 9.7610352558408e-08,
      "loss": 0.76508695,
      "num_input_tokens_seen": 323913835,
      "step": 15021,
      "time_per_iteration": 2.5051286220550537
    },
    {
      "auxiliary_loss_clip": 0.01107551,
      "auxiliary_loss_mlp": 0.01029301,
      "balance_loss_clip": 1.03610253,
      "balance_loss_mlp": 1.01653326,
      "epoch": 0.9031715015782353,
      "flos": 22237216369920.0,
      "grad_norm": 2.5182885859104953,
      "language_loss": 0.72234541,
      "learning_rate": 9.749020425753251e-08,
      "loss": 0.74371392,
      "num_input_tokens_seen": 323933440,
      "step": 15022,
      "time_per_iteration": 2.4579663276672363
    },
    {
      "auxiliary_loss_clip": 0.01063076,
      "auxiliary_loss_mlp": 0.0102844,
      "balance_loss_clip": 1.03426242,
      "balance_loss_mlp": 1.01724052,
      "epoch": 0.9032316248309034,
      "flos": 26323168164480.0,
      "grad_norm": 1.9237241098375688,
      "language_loss": 0.72523808,
      "learning_rate": 9.737012810001943e-08,
      "loss": 0.74615324,
      "num_input_tokens_seen": 323954090,
      "step": 15023,
      "time_per_iteration": 2.5879459381103516
    },
    {
      "auxiliary_loss_clip": 0.01090361,
      "auxiliary_loss_mlp": 0.01032888,
      "balance_loss_clip": 1.03730261,
      "balance_loss_mlp": 1.02142644,
      "epoch": 0.9032917480835713,
      "flos": 22636686499200.0,
      "grad_norm": 1.7240815570681225,
      "language_loss": 0.82646823,
      "learning_rate": 9.725012409042155e-08,
      "loss": 0.84770072,
      "num_input_tokens_seen": 323974040,
      "step": 15024,
      "time_per_iteration": 3.92330265045166
    },
    {
      "auxiliary_loss_clip": 0.01093299,
      "auxiliary_loss_mlp": 0.01028375,
      "balance_loss_clip": 1.03417051,
      "balance_loss_mlp": 1.01668096,
      "epoch": 0.9033518713362393,
      "flos": 23878764846720.0,
      "grad_norm": 1.470205203000881,
      "language_loss": 0.69601285,
      "learning_rate": 9.713019223328966e-08,
      "loss": 0.7172296,
      "num_input_tokens_seen": 323996125,
      "step": 15025,
      "time_per_iteration": 2.487698554992676
    },
    {
      "auxiliary_loss_clip": 0.01067972,
      "auxiliary_loss_mlp": 0.01033833,
      "balance_loss_clip": 1.03258204,
      "balance_loss_mlp": 1.02229321,
      "epoch": 0.9034119945889073,
      "flos": 26905279973760.0,
      "grad_norm": 1.6304710183700022,
      "language_loss": 0.77046514,
      "learning_rate": 9.70103325331717e-08,
      "loss": 0.79148322,
      "num_input_tokens_seen": 324017645,
      "step": 15026,
      "time_per_iteration": 4.1952149868011475
    },
    {
      "auxiliary_loss_clip": 0.01091589,
      "auxiliary_loss_mlp": 0.01028555,
      "balance_loss_clip": 1.03503132,
      "balance_loss_mlp": 1.01755214,
      "epoch": 0.9034721178415752,
      "flos": 20850166730880.0,
      "grad_norm": 1.6534228903927537,
      "language_loss": 0.68348634,
      "learning_rate": 9.68905449946129e-08,
      "loss": 0.70468777,
      "num_input_tokens_seen": 324036875,
      "step": 15027,
      "time_per_iteration": 3.937288284301758
    },
    {
      "auxiliary_loss_clip": 0.0104884,
      "auxiliary_loss_mlp": 0.01037013,
      "balance_loss_clip": 1.0312705,
      "balance_loss_mlp": 1.02392399,
      "epoch": 0.9035322410942432,
      "flos": 22234307368320.0,
      "grad_norm": 2.186293985256084,
      "language_loss": 0.76043975,
      "learning_rate": 9.677082962215477e-08,
      "loss": 0.78129828,
      "num_input_tokens_seen": 324057045,
      "step": 15028,
      "time_per_iteration": 2.608381509780884
    },
    {
      "auxiliary_loss_clip": 0.0105582,
      "auxiliary_loss_mlp": 0.01037005,
      "balance_loss_clip": 1.03410697,
      "balance_loss_mlp": 1.02508974,
      "epoch": 0.9035923643469111,
      "flos": 25923410726400.0,
      "grad_norm": 1.6804459229643494,
      "language_loss": 0.68991888,
      "learning_rate": 9.665118642033765e-08,
      "loss": 0.71084708,
      "num_input_tokens_seen": 324079735,
      "step": 15029,
      "time_per_iteration": 2.6137635707855225
    },
    {
      "auxiliary_loss_clip": 0.01089717,
      "auxiliary_loss_mlp": 0.01028742,
      "balance_loss_clip": 1.03570282,
      "balance_loss_mlp": 1.0162251,
      "epoch": 0.9036524875995792,
      "flos": 20339804338560.0,
      "grad_norm": 3.0870234954046194,
      "language_loss": 0.74051976,
      "learning_rate": 9.653161539369858e-08,
      "loss": 0.76170433,
      "num_input_tokens_seen": 324097785,
      "step": 15030,
      "time_per_iteration": 2.478355884552002
    },
    {
      "auxiliary_loss_clip": 0.01096792,
      "auxiliary_loss_mlp": 0.01030187,
      "balance_loss_clip": 1.03532219,
      "balance_loss_mlp": 1.01765203,
      "epoch": 0.9037126108522471,
      "flos": 40114624677120.0,
      "grad_norm": 1.9260782456417946,
      "language_loss": 0.68133241,
      "learning_rate": 9.641211654677151e-08,
      "loss": 0.70260221,
      "num_input_tokens_seen": 324121625,
      "step": 15031,
      "time_per_iteration": 2.617034673690796
    },
    {
      "auxiliary_loss_clip": 0.01078742,
      "auxiliary_loss_mlp": 0.01027469,
      "balance_loss_clip": 1.03326607,
      "balance_loss_mlp": 1.01652539,
      "epoch": 0.9037727341049151,
      "flos": 23332024955520.0,
      "grad_norm": 1.499087600739369,
      "language_loss": 0.76006413,
      "learning_rate": 9.629268988408723e-08,
      "loss": 0.78112626,
      "num_input_tokens_seen": 324142535,
      "step": 15032,
      "time_per_iteration": 2.554337501525879
    },
    {
      "auxiliary_loss_clip": 0.01103806,
      "auxiliary_loss_mlp": 0.01030343,
      "balance_loss_clip": 1.03510523,
      "balance_loss_mlp": 1.01855898,
      "epoch": 0.903832857357583,
      "flos": 12822659815680.0,
      "grad_norm": 1.9293270192148513,
      "language_loss": 0.7544691,
      "learning_rate": 9.617333541017502e-08,
      "loss": 0.77581054,
      "num_input_tokens_seen": 324159610,
      "step": 15033,
      "time_per_iteration": 2.4161760807037354
    },
    {
      "auxiliary_loss_clip": 0.01071144,
      "auxiliary_loss_mlp": 0.01035557,
      "balance_loss_clip": 1.03208697,
      "balance_loss_mlp": 1.02245617,
      "epoch": 0.903892980610251,
      "flos": 25703026830720.0,
      "grad_norm": 1.773986569993747,
      "language_loss": 0.74001634,
      "learning_rate": 9.605405312956105e-08,
      "loss": 0.7610833,
      "num_input_tokens_seen": 324182510,
      "step": 15034,
      "time_per_iteration": 3.993263006210327
    },
    {
      "auxiliary_loss_clip": 0.01070383,
      "auxiliary_loss_mlp": 0.01032209,
      "balance_loss_clip": 1.03448212,
      "balance_loss_mlp": 1.01996052,
      "epoch": 0.9039531038629189,
      "flos": 14684089397760.0,
      "grad_norm": 1.619473195423546,
      "language_loss": 0.63374865,
      "learning_rate": 9.593484304676791e-08,
      "loss": 0.65477455,
      "num_input_tokens_seen": 324200555,
      "step": 15035,
      "time_per_iteration": 2.5127596855163574
    },
    {
      "auxiliary_loss_clip": 0.01103536,
      "auxiliary_loss_mlp": 0.01031139,
      "balance_loss_clip": 1.03569305,
      "balance_loss_mlp": 1.01812732,
      "epoch": 0.904013227115587,
      "flos": 24024921287040.0,
      "grad_norm": 2.2150075734538976,
      "language_loss": 0.62309092,
      "learning_rate": 9.581570516631643e-08,
      "loss": 0.64443773,
      "num_input_tokens_seen": 324220255,
      "step": 15036,
      "time_per_iteration": 2.4935667514801025
    },
    {
      "auxiliary_loss_clip": 0.0105863,
      "auxiliary_loss_mlp": 0.0102629,
      "balance_loss_clip": 1.03526807,
      "balance_loss_mlp": 1.01545346,
      "epoch": 0.9040733503682549,
      "flos": 22856459863680.0,
      "grad_norm": 1.59045381824056,
      "language_loss": 0.82341969,
      "learning_rate": 9.569663949272455e-08,
      "loss": 0.84426886,
      "num_input_tokens_seen": 324237855,
      "step": 15037,
      "time_per_iteration": 2.5857129096984863
    },
    {
      "auxiliary_loss_clip": 0.01104917,
      "auxiliary_loss_mlp": 0.0102692,
      "balance_loss_clip": 1.0347476,
      "balance_loss_mlp": 1.0150466,
      "epoch": 0.9041334736209229,
      "flos": 19974951941760.0,
      "grad_norm": 1.6290154862734403,
      "language_loss": 0.67387116,
      "learning_rate": 9.557764603050667e-08,
      "loss": 0.69518954,
      "num_input_tokens_seen": 324257050,
      "step": 15038,
      "time_per_iteration": 2.4516005516052246
    },
    {
      "auxiliary_loss_clip": 0.01082489,
      "auxiliary_loss_mlp": 0.0103592,
      "balance_loss_clip": 1.03398037,
      "balance_loss_mlp": 1.0237186,
      "epoch": 0.9041935968735909,
      "flos": 17530548624000.0,
      "grad_norm": 1.951111109326481,
      "language_loss": 0.75186771,
      "learning_rate": 9.545872478417494e-08,
      "loss": 0.7730518,
      "num_input_tokens_seen": 324275510,
      "step": 15039,
      "time_per_iteration": 2.5058467388153076
    },
    {
      "auxiliary_loss_clip": 0.0107722,
      "auxiliary_loss_mlp": 0.0102806,
      "balance_loss_clip": 1.03555012,
      "balance_loss_mlp": 1.01702738,
      "epoch": 0.9042537201262588,
      "flos": 22780149419520.0,
      "grad_norm": 1.4816399640926188,
      "language_loss": 0.70090377,
      "learning_rate": 9.533987575823977e-08,
      "loss": 0.72195661,
      "num_input_tokens_seen": 324295150,
      "step": 15040,
      "time_per_iteration": 2.5752105712890625
    },
    {
      "auxiliary_loss_clip": 0.01069659,
      "auxiliary_loss_mlp": 0.01029765,
      "balance_loss_clip": 1.03251767,
      "balance_loss_mlp": 1.01807654,
      "epoch": 0.9043138433789268,
      "flos": 20595416497920.0,
      "grad_norm": 1.7227035817064555,
      "language_loss": 0.67748952,
      "learning_rate": 9.522109895720709e-08,
      "loss": 0.69848371,
      "num_input_tokens_seen": 324313855,
      "step": 15041,
      "time_per_iteration": 2.527285099029541
    },
    {
      "auxiliary_loss_clip": 0.01091591,
      "auxiliary_loss_mlp": 0.01032057,
      "balance_loss_clip": 1.03410196,
      "balance_loss_mlp": 1.01961148,
      "epoch": 0.9043739666315948,
      "flos": 32962978995840.0,
      "grad_norm": 2.272083105051769,
      "language_loss": 0.57533634,
      "learning_rate": 9.510239438558155e-08,
      "loss": 0.59657288,
      "num_input_tokens_seen": 324338465,
      "step": 15042,
      "time_per_iteration": 2.6001853942871094
    },
    {
      "auxiliary_loss_clip": 0.01010626,
      "auxiliary_loss_mlp": 0.0076171,
      "balance_loss_clip": 1.00724387,
      "balance_loss_mlp": 1.00088727,
      "epoch": 0.9044340898842628,
      "flos": 67296418525440.0,
      "grad_norm": 0.7806458370604512,
      "language_loss": 0.57033837,
      "learning_rate": 9.498376204786351e-08,
      "loss": 0.58806175,
      "num_input_tokens_seen": 324398740,
      "step": 15043,
      "time_per_iteration": 3.10719633102417
    },
    {
      "auxiliary_loss_clip": 0.0108184,
      "auxiliary_loss_mlp": 0.01028719,
      "balance_loss_clip": 1.03378999,
      "balance_loss_mlp": 1.01595783,
      "epoch": 0.9044942131369307,
      "flos": 17713154390400.0,
      "grad_norm": 1.786827895398811,
      "language_loss": 0.69985735,
      "learning_rate": 9.486520194855274e-08,
      "loss": 0.720963,
      "num_input_tokens_seen": 324417335,
      "step": 15044,
      "time_per_iteration": 2.492211103439331
    },
    {
      "auxiliary_loss_clip": 0.01084205,
      "auxiliary_loss_mlp": 0.01035825,
      "balance_loss_clip": 1.03490877,
      "balance_loss_mlp": 1.02323091,
      "epoch": 0.9045543363895987,
      "flos": 17820563034240.0,
      "grad_norm": 2.7065267870330634,
      "language_loss": 0.69902307,
      "learning_rate": 9.474671409214407e-08,
      "loss": 0.72022337,
      "num_input_tokens_seen": 324433240,
      "step": 15045,
      "time_per_iteration": 2.475430488586426
    },
    {
      "auxiliary_loss_clip": 0.01072991,
      "auxiliary_loss_mlp": 0.01033979,
      "balance_loss_clip": 1.03447223,
      "balance_loss_mlp": 1.02109838,
      "epoch": 0.9046144596422666,
      "flos": 21872723109120.0,
      "grad_norm": 1.9289700273953294,
      "language_loss": 0.6545347,
      "learning_rate": 9.462829848313081e-08,
      "loss": 0.67560434,
      "num_input_tokens_seen": 324452675,
      "step": 15046,
      "time_per_iteration": 2.542355537414551
    },
    {
      "auxiliary_loss_clip": 0.01071682,
      "auxiliary_loss_mlp": 0.01034361,
      "balance_loss_clip": 1.03351641,
      "balance_loss_mlp": 1.02208853,
      "epoch": 0.9046745828949346,
      "flos": 17672646827520.0,
      "grad_norm": 2.077150935453088,
      "language_loss": 0.61442852,
      "learning_rate": 9.450995512600379e-08,
      "loss": 0.63548893,
      "num_input_tokens_seen": 324467865,
      "step": 15047,
      "time_per_iteration": 2.5078887939453125
    },
    {
      "auxiliary_loss_clip": 0.01102977,
      "auxiliary_loss_mlp": 0.00783459,
      "balance_loss_clip": 1.03555298,
      "balance_loss_mlp": 1.01086712,
      "epoch": 0.9047347061476025,
      "flos": 25702559953920.0,
      "grad_norm": 1.4028605874399873,
      "language_loss": 0.71231496,
      "learning_rate": 9.439168402525032e-08,
      "loss": 0.73117936,
      "num_input_tokens_seen": 324490430,
      "step": 15048,
      "time_per_iteration": 2.4914350509643555
    },
    {
      "auxiliary_loss_clip": 0.01091807,
      "auxiliary_loss_mlp": 0.01030556,
      "balance_loss_clip": 1.03156567,
      "balance_loss_mlp": 1.01764607,
      "epoch": 0.9047948294002706,
      "flos": 15158146118400.0,
      "grad_norm": 2.0727322676737416,
      "language_loss": 0.74960464,
      "learning_rate": 9.427348518535483e-08,
      "loss": 0.77082831,
      "num_input_tokens_seen": 324506620,
      "step": 15049,
      "time_per_iteration": 2.4512054920196533
    },
    {
      "auxiliary_loss_clip": 0.01090447,
      "auxiliary_loss_mlp": 0.01028004,
      "balance_loss_clip": 1.03573692,
      "balance_loss_mlp": 1.01588082,
      "epoch": 0.9048549526529385,
      "flos": 21872292145920.0,
      "grad_norm": 3.221606222288242,
      "language_loss": 0.75882655,
      "learning_rate": 9.415535861079993e-08,
      "loss": 0.78001106,
      "num_input_tokens_seen": 324525505,
      "step": 15050,
      "time_per_iteration": 2.5065665245056152
    },
    {
      "auxiliary_loss_clip": 0.01103173,
      "auxiliary_loss_mlp": 0.00781593,
      "balance_loss_clip": 1.03451729,
      "balance_loss_mlp": 1.0072372,
      "epoch": 0.9049150759056065,
      "flos": 23546626761600.0,
      "grad_norm": 1.7938632188043035,
      "language_loss": 0.81831747,
      "learning_rate": 9.403730430606472e-08,
      "loss": 0.83716518,
      "num_input_tokens_seen": 324544415,
      "step": 15051,
      "time_per_iteration": 2.47902774810791
    },
    {
      "auxiliary_loss_clip": 0.01091979,
      "auxiliary_loss_mlp": 0.01028819,
      "balance_loss_clip": 1.03450561,
      "balance_loss_mlp": 1.0177263,
      "epoch": 0.9049751991582745,
      "flos": 19645902426240.0,
      "grad_norm": 1.8805353912317617,
      "language_loss": 0.88817197,
      "learning_rate": 9.391932227562582e-08,
      "loss": 0.90937996,
      "num_input_tokens_seen": 324562555,
      "step": 15052,
      "time_per_iteration": 2.462597608566284
    },
    {
      "auxiliary_loss_clip": 0.01095697,
      "auxiliary_loss_mlp": 0.01033205,
      "balance_loss_clip": 1.03476238,
      "balance_loss_mlp": 1.02096796,
      "epoch": 0.9050353224109424,
      "flos": 15596220389760.0,
      "grad_norm": 1.9344454239975266,
      "language_loss": 0.77426064,
      "learning_rate": 9.380141252395724e-08,
      "loss": 0.79554963,
      "num_input_tokens_seen": 324580865,
      "step": 15053,
      "time_per_iteration": 2.456810235977173
    },
    {
      "auxiliary_loss_clip": 0.01089759,
      "auxiliary_loss_mlp": 0.01032028,
      "balance_loss_clip": 1.03343654,
      "balance_loss_mlp": 1.01983905,
      "epoch": 0.9050954456636104,
      "flos": 28183592165760.0,
      "grad_norm": 1.5013478296720073,
      "language_loss": 0.72882193,
      "learning_rate": 9.368357505553049e-08,
      "loss": 0.75003976,
      "num_input_tokens_seen": 324600665,
      "step": 15054,
      "time_per_iteration": 2.5316455364227295
    },
    {
      "auxiliary_loss_clip": 0.01049681,
      "auxiliary_loss_mlp": 0.01034558,
      "balance_loss_clip": 1.03045678,
      "balance_loss_mlp": 1.02258992,
      "epoch": 0.9051555689162784,
      "flos": 25731611078400.0,
      "grad_norm": 4.735088325413634,
      "language_loss": 0.83647203,
      "learning_rate": 9.356580987481333e-08,
      "loss": 0.85731435,
      "num_input_tokens_seen": 324618145,
      "step": 15055,
      "time_per_iteration": 2.5937860012054443
    },
    {
      "auxiliary_loss_clip": 0.01091802,
      "auxiliary_loss_mlp": 0.01034326,
      "balance_loss_clip": 1.03554881,
      "balance_loss_mlp": 1.02234507,
      "epoch": 0.9052156921689464,
      "flos": 23257258796160.0,
      "grad_norm": 1.5891597673157531,
      "language_loss": 0.84981298,
      "learning_rate": 9.344811698627176e-08,
      "loss": 0.87107426,
      "num_input_tokens_seen": 324638165,
      "step": 15056,
      "time_per_iteration": 2.4831764698028564
    },
    {
      "auxiliary_loss_clip": 0.01082884,
      "auxiliary_loss_mlp": 0.01030597,
      "balance_loss_clip": 1.03484511,
      "balance_loss_mlp": 1.01943922,
      "epoch": 0.9052758154216143,
      "flos": 29564285097600.0,
      "grad_norm": 1.7869035788437304,
      "language_loss": 0.71785337,
      "learning_rate": 9.333049639436863e-08,
      "loss": 0.7389881,
      "num_input_tokens_seen": 324658560,
      "step": 15057,
      "time_per_iteration": 2.560811996459961
    },
    {
      "auxiliary_loss_clip": 0.01088872,
      "auxiliary_loss_mlp": 0.01026862,
      "balance_loss_clip": 1.03250527,
      "balance_loss_mlp": 1.01519692,
      "epoch": 0.9053359386742823,
      "flos": 22127688823680.0,
      "grad_norm": 2.158906675298929,
      "language_loss": 0.80519331,
      "learning_rate": 9.321294810356418e-08,
      "loss": 0.82635069,
      "num_input_tokens_seen": 324679185,
      "step": 15058,
      "time_per_iteration": 2.4753379821777344
    },
    {
      "auxiliary_loss_clip": 0.0101759,
      "auxiliary_loss_mlp": 0.01004343,
      "balance_loss_clip": 1.00664926,
      "balance_loss_mlp": 1.00322294,
      "epoch": 0.9053960619269502,
      "flos": 67090112760960.0,
      "grad_norm": 0.676763934972753,
      "language_loss": 0.51426655,
      "learning_rate": 9.309547211831592e-08,
      "loss": 0.53448594,
      "num_input_tokens_seen": 324744830,
      "step": 15059,
      "time_per_iteration": 3.203374147415161
    },
    {
      "auxiliary_loss_clip": 0.01062846,
      "auxiliary_loss_mlp": 0.01029551,
      "balance_loss_clip": 1.03382051,
      "balance_loss_mlp": 1.0170815,
      "epoch": 0.9054561851796182,
      "flos": 15815419136640.0,
      "grad_norm": 2.617153859034066,
      "language_loss": 0.6722666,
      "learning_rate": 9.297806844307831e-08,
      "loss": 0.69319057,
      "num_input_tokens_seen": 324762905,
      "step": 15060,
      "time_per_iteration": 2.554060459136963
    },
    {
      "auxiliary_loss_clip": 0.01085644,
      "auxiliary_loss_mlp": 0.01029801,
      "balance_loss_clip": 1.03593612,
      "balance_loss_mlp": 1.01810634,
      "epoch": 0.9055163084322861,
      "flos": 17566997950080.0,
      "grad_norm": 1.9562611463312884,
      "language_loss": 0.63859111,
      "learning_rate": 9.286073708230357e-08,
      "loss": 0.65974557,
      "num_input_tokens_seen": 324781905,
      "step": 15061,
      "time_per_iteration": 2.482518196105957
    },
    {
      "auxiliary_loss_clip": 0.01079168,
      "auxiliary_loss_mlp": 0.01033456,
      "balance_loss_clip": 1.03603935,
      "balance_loss_mlp": 1.02096295,
      "epoch": 0.9055764316849542,
      "flos": 17639573379840.0,
      "grad_norm": 1.7820977848288684,
      "language_loss": 0.7149508,
      "learning_rate": 9.274347804044058e-08,
      "loss": 0.73607707,
      "num_input_tokens_seen": 324799260,
      "step": 15062,
      "time_per_iteration": 2.4710333347320557
    },
    {
      "auxiliary_loss_clip": 0.01101376,
      "auxiliary_loss_mlp": 0.01029829,
      "balance_loss_clip": 1.03392673,
      "balance_loss_mlp": 1.0187484,
      "epoch": 0.9056365549376221,
      "flos": 20120856986880.0,
      "grad_norm": 1.64841174642194,
      "language_loss": 0.70816362,
      "learning_rate": 9.2626291321936e-08,
      "loss": 0.72947562,
      "num_input_tokens_seen": 324817800,
      "step": 15063,
      "time_per_iteration": 4.092405557632446
    },
    {
      "auxiliary_loss_clip": 0.01068869,
      "auxiliary_loss_mlp": 0.01033491,
      "balance_loss_clip": 1.03412974,
      "balance_loss_mlp": 1.02200484,
      "epoch": 0.9056966781902901,
      "flos": 27598786836480.0,
      "grad_norm": 1.7651684457338732,
      "language_loss": 0.72311521,
      "learning_rate": 9.250917693123406e-08,
      "loss": 0.74413878,
      "num_input_tokens_seen": 324838445,
      "step": 15064,
      "time_per_iteration": 3.9789211750030518
    },
    {
      "auxiliary_loss_clip": 0.01092734,
      "auxiliary_loss_mlp": 0.01030838,
      "balance_loss_clip": 1.03223085,
      "balance_loss_mlp": 1.01891661,
      "epoch": 0.9057568014429581,
      "flos": 25920106675200.0,
      "grad_norm": 2.0647451205952985,
      "language_loss": 0.6902169,
      "learning_rate": 9.23921348727752e-08,
      "loss": 0.7114526,
      "num_input_tokens_seen": 324859895,
      "step": 15065,
      "time_per_iteration": 2.525019645690918
    },
    {
      "auxiliary_loss_clip": 0.01081444,
      "auxiliary_loss_mlp": 0.01036007,
      "balance_loss_clip": 1.03614235,
      "balance_loss_mlp": 1.02436018,
      "epoch": 0.905816924695626,
      "flos": 22930364096640.0,
      "grad_norm": 1.908518059028003,
      "language_loss": 0.63082129,
      "learning_rate": 9.227516515099743e-08,
      "loss": 0.65199578,
      "num_input_tokens_seen": 324879580,
      "step": 15066,
      "time_per_iteration": 3.9265286922454834
    },
    {
      "auxiliary_loss_clip": 0.010396,
      "auxiliary_loss_mlp": 0.01031637,
      "balance_loss_clip": 1.03037691,
      "balance_loss_mlp": 1.01796412,
      "epoch": 0.905877047948294,
      "flos": 22157422306560.0,
      "grad_norm": 2.095067942768182,
      "language_loss": 0.80078787,
      "learning_rate": 9.215826777033675e-08,
      "loss": 0.82150024,
      "num_input_tokens_seen": 324898950,
      "step": 15067,
      "time_per_iteration": 2.652109384536743
    },
    {
      "auxiliary_loss_clip": 0.01083687,
      "auxiliary_loss_mlp": 0.01032628,
      "balance_loss_clip": 1.03520823,
      "balance_loss_mlp": 1.02001548,
      "epoch": 0.905937171200962,
      "flos": 15304805349120.0,
      "grad_norm": 1.5922264320007362,
      "language_loss": 0.69898462,
      "learning_rate": 9.204144273522563e-08,
      "loss": 0.72014785,
      "num_input_tokens_seen": 324917455,
      "step": 15068,
      "time_per_iteration": 2.479933261871338
    },
    {
      "auxiliary_loss_clip": 0.01099481,
      "auxiliary_loss_mlp": 0.01028299,
      "balance_loss_clip": 1.03304005,
      "balance_loss_mlp": 1.0165031,
      "epoch": 0.90599729445363,
      "flos": 19462973437440.0,
      "grad_norm": 1.8178745689983329,
      "language_loss": 0.85405886,
      "learning_rate": 9.19246900500943e-08,
      "loss": 0.87533665,
      "num_input_tokens_seen": 324934495,
      "step": 15069,
      "time_per_iteration": 2.435887575149536
    },
    {
      "auxiliary_loss_clip": 0.01095388,
      "auxiliary_loss_mlp": 0.01031199,
      "balance_loss_clip": 1.03367972,
      "balance_loss_mlp": 1.01844335,
      "epoch": 0.9060574177062979,
      "flos": 23732967542400.0,
      "grad_norm": 1.7751685361287097,
      "language_loss": 0.59157318,
      "learning_rate": 9.180800971936987e-08,
      "loss": 0.61283904,
      "num_input_tokens_seen": 324953230,
      "step": 15070,
      "time_per_iteration": 2.475350856781006
    },
    {
      "auxiliary_loss_clip": 0.01068876,
      "auxiliary_loss_mlp": 0.01023721,
      "balance_loss_clip": 1.0341804,
      "balance_loss_mlp": 1.01130569,
      "epoch": 0.9061175409589659,
      "flos": 17311134395520.0,
      "grad_norm": 12.235446626168141,
      "language_loss": 0.81792533,
      "learning_rate": 9.169140174747724e-08,
      "loss": 0.83885133,
      "num_input_tokens_seen": 324969880,
      "step": 15071,
      "time_per_iteration": 2.5167810916900635
    },
    {
      "auxiliary_loss_clip": 0.0110668,
      "auxiliary_loss_mlp": 0.01035681,
      "balance_loss_clip": 1.03514934,
      "balance_loss_mlp": 1.02268696,
      "epoch": 0.9061776642116338,
      "flos": 17778439359360.0,
      "grad_norm": 1.8650424708269293,
      "language_loss": 0.61757195,
      "learning_rate": 9.157486613883758e-08,
      "loss": 0.63899553,
      "num_input_tokens_seen": 324987005,
      "step": 15072,
      "time_per_iteration": 2.409898519515991
    },
    {
      "auxiliary_loss_clip": 0.01082303,
      "auxiliary_loss_mlp": 0.0103294,
      "balance_loss_clip": 1.03405619,
      "balance_loss_mlp": 1.02101874,
      "epoch": 0.9062377874643018,
      "flos": 42777688037760.0,
      "grad_norm": 1.8024412736506676,
      "language_loss": 0.729568,
      "learning_rate": 9.145840289787021e-08,
      "loss": 0.75072044,
      "num_input_tokens_seen": 325010700,
      "step": 15073,
      "time_per_iteration": 4.093415260314941
    },
    {
      "auxiliary_loss_clip": 0.01090339,
      "auxiliary_loss_mlp": 0.01026041,
      "balance_loss_clip": 1.03496206,
      "balance_loss_mlp": 1.01479387,
      "epoch": 0.9062979107169697,
      "flos": 16361620323840.0,
      "grad_norm": 2.0512498727794526,
      "language_loss": 0.81263709,
      "learning_rate": 9.134201202899161e-08,
      "loss": 0.83380091,
      "num_input_tokens_seen": 325028760,
      "step": 15074,
      "time_per_iteration": 2.445389747619629
    },
    {
      "auxiliary_loss_clip": 0.00990761,
      "auxiliary_loss_mlp": 0.00764915,
      "balance_loss_clip": 1.01735806,
      "balance_loss_mlp": 1.00293279,
      "epoch": 0.9063580339696378,
      "flos": 69313988528640.0,
      "grad_norm": 0.7489242692336314,
      "language_loss": 0.52399814,
      "learning_rate": 9.122569353661513e-08,
      "loss": 0.54155487,
      "num_input_tokens_seen": 325093545,
      "step": 15075,
      "time_per_iteration": 3.2996654510498047
    },
    {
      "auxiliary_loss_clip": 0.01001062,
      "auxiliary_loss_mlp": 0.01000562,
      "balance_loss_clip": 1.01252973,
      "balance_loss_mlp": 0.99938202,
      "epoch": 0.9064181572223057,
      "flos": 58794747148800.0,
      "grad_norm": 0.7289157916781073,
      "language_loss": 0.62103361,
      "learning_rate": 9.11094474251517e-08,
      "loss": 0.6410498,
      "num_input_tokens_seen": 325152295,
      "step": 15076,
      "time_per_iteration": 3.0579447746276855
    },
    {
      "auxiliary_loss_clip": 0.01090242,
      "auxiliary_loss_mlp": 0.0103179,
      "balance_loss_clip": 1.0334177,
      "balance_loss_mlp": 1.02006531,
      "epoch": 0.9064782804749737,
      "flos": 21762692772480.0,
      "grad_norm": 1.8460076961226015,
      "language_loss": 0.82084668,
      "learning_rate": 9.09932736990091e-08,
      "loss": 0.84206694,
      "num_input_tokens_seen": 325169705,
      "step": 15077,
      "time_per_iteration": 2.489321231842041
    },
    {
      "auxiliary_loss_clip": 0.01078156,
      "auxiliary_loss_mlp": 0.00783042,
      "balance_loss_clip": 1.03162384,
      "balance_loss_mlp": 1.00917709,
      "epoch": 0.9065384037276417,
      "flos": 21397373498880.0,
      "grad_norm": 1.5132035498580798,
      "language_loss": 0.83971798,
      "learning_rate": 9.08771723625934e-08,
      "loss": 0.85833001,
      "num_input_tokens_seen": 325189175,
      "step": 15078,
      "time_per_iteration": 2.5343127250671387
    },
    {
      "auxiliary_loss_clip": 0.01088178,
      "auxiliary_loss_mlp": 0.00781744,
      "balance_loss_clip": 1.03434205,
      "balance_loss_mlp": 1.00869894,
      "epoch": 0.9065985269803096,
      "flos": 38283646849920.0,
      "grad_norm": 1.9898008106676155,
      "language_loss": 0.65509307,
      "learning_rate": 9.076114342030617e-08,
      "loss": 0.67379224,
      "num_input_tokens_seen": 325211020,
      "step": 15079,
      "time_per_iteration": 2.61872935295105
    },
    {
      "auxiliary_loss_clip": 0.01033235,
      "auxiliary_loss_mlp": 0.01025018,
      "balance_loss_clip": 1.03311741,
      "balance_loss_mlp": 1.01337719,
      "epoch": 0.9066586502329776,
      "flos": 44818562989440.0,
      "grad_norm": 1.6765017908556235,
      "language_loss": 0.71346736,
      "learning_rate": 9.064518687654765e-08,
      "loss": 0.73404992,
      "num_input_tokens_seen": 325236970,
      "step": 15080,
      "time_per_iteration": 2.8443312644958496
    },
    {
      "auxiliary_loss_clip": 0.01090769,
      "auxiliary_loss_mlp": 0.0102926,
      "balance_loss_clip": 1.03670502,
      "balance_loss_mlp": 1.01677322,
      "epoch": 0.9067187734856456,
      "flos": 18623992492800.0,
      "grad_norm": 3.7614865771123345,
      "language_loss": 0.711263,
      "learning_rate": 9.052930273571547e-08,
      "loss": 0.73246324,
      "num_input_tokens_seen": 325252670,
      "step": 15081,
      "time_per_iteration": 2.443546772003174
    },
    {
      "auxiliary_loss_clip": 0.01077768,
      "auxiliary_loss_mlp": 0.01030131,
      "balance_loss_clip": 1.03570175,
      "balance_loss_mlp": 1.01820445,
      "epoch": 0.9067788967383136,
      "flos": 22747578762240.0,
      "grad_norm": 1.9556746266016236,
      "language_loss": 0.74421239,
      "learning_rate": 9.04134910022032e-08,
      "loss": 0.76529133,
      "num_input_tokens_seen": 325273860,
      "step": 15082,
      "time_per_iteration": 2.5417847633361816
    },
    {
      "auxiliary_loss_clip": 0.01068684,
      "auxiliary_loss_mlp": 0.01034737,
      "balance_loss_clip": 1.03410578,
      "balance_loss_mlp": 1.02348328,
      "epoch": 0.9068390199909815,
      "flos": 27670787648640.0,
      "grad_norm": 1.816967728608277,
      "language_loss": 0.78529322,
      "learning_rate": 9.029775168040266e-08,
      "loss": 0.80632734,
      "num_input_tokens_seen": 325294140,
      "step": 15083,
      "time_per_iteration": 2.5715065002441406
    },
    {
      "auxiliary_loss_clip": 0.01076655,
      "auxiliary_loss_mlp": 0.00783461,
      "balance_loss_clip": 1.03443754,
      "balance_loss_mlp": 1.01090932,
      "epoch": 0.9068991432436495,
      "flos": 24244012293120.0,
      "grad_norm": 1.5417492316763635,
      "language_loss": 0.69057846,
      "learning_rate": 9.01820847747028e-08,
      "loss": 0.70917958,
      "num_input_tokens_seen": 325313130,
      "step": 15084,
      "time_per_iteration": 2.529724359512329
    },
    {
      "auxiliary_loss_clip": 0.01104457,
      "auxiliary_loss_mlp": 0.01032664,
      "balance_loss_clip": 1.03646016,
      "balance_loss_mlp": 1.02095795,
      "epoch": 0.9069592664963174,
      "flos": 28033305661440.0,
      "grad_norm": 3.6950651086591013,
      "language_loss": 0.66648692,
      "learning_rate": 9.006649028948965e-08,
      "loss": 0.6878581,
      "num_input_tokens_seen": 325334880,
      "step": 15085,
      "time_per_iteration": 2.4781494140625
    },
    {
      "auxiliary_loss_clip": 0.01008207,
      "auxiliary_loss_mlp": 0.01008854,
      "balance_loss_clip": 1.02234626,
      "balance_loss_mlp": 1.00763798,
      "epoch": 0.9070193897489854,
      "flos": 68778414789120.0,
      "grad_norm": 0.7718682185926731,
      "language_loss": 0.61332178,
      "learning_rate": 8.995096822914638e-08,
      "loss": 0.63349235,
      "num_input_tokens_seen": 325394175,
      "step": 15086,
      "time_per_iteration": 3.166022777557373
    },
    {
      "auxiliary_loss_clip": 0.01088518,
      "auxiliary_loss_mlp": 0.01037917,
      "balance_loss_clip": 1.03383231,
      "balance_loss_mlp": 1.02492929,
      "epoch": 0.9070795130016533,
      "flos": 23441624328960.0,
      "grad_norm": 1.4139750692085502,
      "language_loss": 0.72171879,
      "learning_rate": 8.983551859805416e-08,
      "loss": 0.74298316,
      "num_input_tokens_seen": 325415020,
      "step": 15087,
      "time_per_iteration": 2.4947776794433594
    },
    {
      "auxiliary_loss_clip": 0.01081333,
      "auxiliary_loss_mlp": 0.01026697,
      "balance_loss_clip": 1.03409314,
      "balance_loss_mlp": 1.01536596,
      "epoch": 0.9071396362543214,
      "flos": 18916413114240.0,
      "grad_norm": 2.1122979950565584,
      "language_loss": 0.76682776,
      "learning_rate": 8.972014140059058e-08,
      "loss": 0.78790808,
      "num_input_tokens_seen": 325433595,
      "step": 15088,
      "time_per_iteration": 2.52335262298584
    },
    {
      "auxiliary_loss_clip": 0.01069204,
      "auxiliary_loss_mlp": 0.01030193,
      "balance_loss_clip": 1.03267193,
      "balance_loss_mlp": 1.01884389,
      "epoch": 0.9071997595069893,
      "flos": 25228646887680.0,
      "grad_norm": 1.8646967162922683,
      "language_loss": 0.72960436,
      "learning_rate": 8.960483664113038e-08,
      "loss": 0.75059831,
      "num_input_tokens_seen": 325451605,
      "step": 15089,
      "time_per_iteration": 2.521923065185547
    },
    {
      "auxiliary_loss_clip": 0.01097901,
      "auxiliary_loss_mlp": 0.0102992,
      "balance_loss_clip": 1.033921,
      "balance_loss_mlp": 1.01947737,
      "epoch": 0.9072598827596573,
      "flos": 24346608514560.0,
      "grad_norm": 1.8579275902803754,
      "language_loss": 0.758048,
      "learning_rate": 8.948960432404628e-08,
      "loss": 0.7793262,
      "num_input_tokens_seen": 325470645,
      "step": 15090,
      "time_per_iteration": 2.4765565395355225
    },
    {
      "auxiliary_loss_clip": 0.01084103,
      "auxiliary_loss_mlp": 0.01025261,
      "balance_loss_clip": 1.03510523,
      "balance_loss_mlp": 1.01254177,
      "epoch": 0.9073200060123253,
      "flos": 22674967418880.0,
      "grad_norm": 2.657341481218281,
      "language_loss": 0.77736521,
      "learning_rate": 8.93744444537079e-08,
      "loss": 0.79845881,
      "num_input_tokens_seen": 325488070,
      "step": 15091,
      "time_per_iteration": 2.5018060207366943
    },
    {
      "auxiliary_loss_clip": 0.0107547,
      "auxiliary_loss_mlp": 0.01025236,
      "balance_loss_clip": 1.03180051,
      "balance_loss_mlp": 1.01450145,
      "epoch": 0.9073801292649932,
      "flos": 23695476721920.0,
      "grad_norm": 1.5357102802532638,
      "language_loss": 0.85931081,
      "learning_rate": 8.925935703448217e-08,
      "loss": 0.88031787,
      "num_input_tokens_seen": 325509285,
      "step": 15092,
      "time_per_iteration": 2.5397744178771973
    },
    {
      "auxiliary_loss_clip": 0.01081646,
      "auxiliary_loss_mlp": 0.0103315,
      "balance_loss_clip": 1.03746223,
      "balance_loss_mlp": 1.02133608,
      "epoch": 0.9074402525176612,
      "flos": 25375413859200.0,
      "grad_norm": 1.7970565238517713,
      "language_loss": 0.78541827,
      "learning_rate": 8.914434207073296e-08,
      "loss": 0.80656624,
      "num_input_tokens_seen": 325529360,
      "step": 15093,
      "time_per_iteration": 2.5473592281341553
    },
    {
      "auxiliary_loss_clip": 0.01019497,
      "auxiliary_loss_mlp": 0.01003853,
      "balance_loss_clip": 1.00590062,
      "balance_loss_mlp": 1.00275636,
      "epoch": 0.9075003757703292,
      "flos": 67649024384640.0,
      "grad_norm": 2.2412402706118058,
      "language_loss": 0.57067239,
      "learning_rate": 8.902939956682188e-08,
      "loss": 0.5909059,
      "num_input_tokens_seen": 325583565,
      "step": 15094,
      "time_per_iteration": 3.0337257385253906
    },
    {
      "auxiliary_loss_clip": 0.01094639,
      "auxiliary_loss_mlp": 0.01032292,
      "balance_loss_clip": 1.0344398,
      "balance_loss_mlp": 1.0196383,
      "epoch": 0.9075604990229972,
      "flos": 22453649769600.0,
      "grad_norm": 1.7832399136651822,
      "language_loss": 0.71219671,
      "learning_rate": 8.891452952710742e-08,
      "loss": 0.73346603,
      "num_input_tokens_seen": 325603690,
      "step": 15095,
      "time_per_iteration": 2.481020450592041
    },
    {
      "auxiliary_loss_clip": 0.01061513,
      "auxiliary_loss_mlp": 0.01034688,
      "balance_loss_clip": 1.03288043,
      "balance_loss_mlp": 1.02255225,
      "epoch": 0.9076206222756651,
      "flos": 19536662188800.0,
      "grad_norm": 1.744702286742437,
      "language_loss": 0.74176717,
      "learning_rate": 8.879973195594526e-08,
      "loss": 0.76272917,
      "num_input_tokens_seen": 325622255,
      "step": 15096,
      "time_per_iteration": 2.5210654735565186
    },
    {
      "auxiliary_loss_clip": 0.01104291,
      "auxiliary_loss_mlp": 0.01034955,
      "balance_loss_clip": 1.03474426,
      "balance_loss_mlp": 1.02185369,
      "epoch": 0.9076807455283331,
      "flos": 30116914819200.0,
      "grad_norm": 2.466185966409207,
      "language_loss": 0.56878674,
      "learning_rate": 8.868500685768898e-08,
      "loss": 0.59017915,
      "num_input_tokens_seen": 325640165,
      "step": 15097,
      "time_per_iteration": 2.529118061065674
    },
    {
      "auxiliary_loss_clip": 0.01089101,
      "auxiliary_loss_mlp": 0.01024686,
      "balance_loss_clip": 1.0328269,
      "balance_loss_mlp": 1.01360571,
      "epoch": 0.907740868781001,
      "flos": 18697537589760.0,
      "grad_norm": 1.6585799042303067,
      "language_loss": 0.7966404,
      "learning_rate": 8.857035423668935e-08,
      "loss": 0.81777829,
      "num_input_tokens_seen": 325659455,
      "step": 15098,
      "time_per_iteration": 2.4470677375793457
    },
    {
      "auxiliary_loss_clip": 0.01061223,
      "auxiliary_loss_mlp": 0.00783456,
      "balance_loss_clip": 1.03310108,
      "balance_loss_mlp": 1.01052928,
      "epoch": 0.907800992033669,
      "flos": 22638805401600.0,
      "grad_norm": 2.2553730754018964,
      "language_loss": 0.65867376,
      "learning_rate": 8.845577409729266e-08,
      "loss": 0.67712057,
      "num_input_tokens_seen": 325678095,
      "step": 15099,
      "time_per_iteration": 2.6100351810455322
    },
    {
      "auxiliary_loss_clip": 0.01084051,
      "auxiliary_loss_mlp": 0.01033633,
      "balance_loss_clip": 1.03470302,
      "balance_loss_mlp": 1.0211519,
      "epoch": 0.907861115286337,
      "flos": 21287666384640.0,
      "grad_norm": 5.619238103681348,
      "language_loss": 0.70391822,
      "learning_rate": 8.834126644384477e-08,
      "loss": 0.72509509,
      "num_input_tokens_seen": 325695825,
      "step": 15100,
      "time_per_iteration": 2.504565954208374
    },
    {
      "auxiliary_loss_clip": 0.01018598,
      "auxiliary_loss_mlp": 0.01000251,
      "balance_loss_clip": 1.0055865,
      "balance_loss_mlp": 0.99929124,
      "epoch": 0.907921238539005,
      "flos": 69739493040000.0,
      "grad_norm": 0.6222657049378155,
      "language_loss": 0.53412986,
      "learning_rate": 8.822683128068775e-08,
      "loss": 0.55431831,
      "num_input_tokens_seen": 325764515,
      "step": 15101,
      "time_per_iteration": 4.539974689483643
    },
    {
      "auxiliary_loss_clip": 0.01068302,
      "auxiliary_loss_mlp": 0.01030807,
      "balance_loss_clip": 1.03285706,
      "balance_loss_mlp": 1.01834977,
      "epoch": 0.9079813617916729,
      "flos": 23477391296640.0,
      "grad_norm": 1.6772100063946953,
      "language_loss": 0.68603516,
      "learning_rate": 8.811246861216081e-08,
      "loss": 0.70702624,
      "num_input_tokens_seen": 325783235,
      "step": 15102,
      "time_per_iteration": 2.571366786956787
    },
    {
      "auxiliary_loss_clip": 0.01088958,
      "auxiliary_loss_mlp": 0.01030424,
      "balance_loss_clip": 1.03465748,
      "balance_loss_mlp": 1.01871789,
      "epoch": 0.9080414850443409,
      "flos": 22929933133440.0,
      "grad_norm": 1.75097999620067,
      "language_loss": 0.79399902,
      "learning_rate": 8.799817844260049e-08,
      "loss": 0.81519288,
      "num_input_tokens_seen": 325800195,
      "step": 15103,
      "time_per_iteration": 3.851489782333374
    },
    {
      "auxiliary_loss_clip": 0.010767,
      "auxiliary_loss_mlp": 0.01028896,
      "balance_loss_clip": 1.0326333,
      "balance_loss_mlp": 1.01670074,
      "epoch": 0.9081016082970089,
      "flos": 26177083551360.0,
      "grad_norm": 1.6359337481462783,
      "language_loss": 0.71447378,
      "learning_rate": 8.78839607763413e-08,
      "loss": 0.73552978,
      "num_input_tokens_seen": 325820215,
      "step": 15104,
      "time_per_iteration": 3.996955394744873
    },
    {
      "auxiliary_loss_clip": 0.01079119,
      "auxiliary_loss_mlp": 0.01023734,
      "balance_loss_clip": 1.03309488,
      "balance_loss_mlp": 1.01277876,
      "epoch": 0.9081617315496768,
      "flos": 24462169545600.0,
      "grad_norm": 1.724763069653292,
      "language_loss": 0.77333128,
      "learning_rate": 8.77698156177138e-08,
      "loss": 0.7943598,
      "num_input_tokens_seen": 325838415,
      "step": 15105,
      "time_per_iteration": 2.5533816814422607
    },
    {
      "auxiliary_loss_clip": 0.01102056,
      "auxiliary_loss_mlp": 0.007841,
      "balance_loss_clip": 1.03343272,
      "balance_loss_mlp": 1.00872827,
      "epoch": 0.9082218548023449,
      "flos": 24746868743040.0,
      "grad_norm": 2.029565226740009,
      "language_loss": 0.73792452,
      "learning_rate": 8.765574297104628e-08,
      "loss": 0.75678605,
      "num_input_tokens_seen": 325855580,
      "step": 15106,
      "time_per_iteration": 2.480668783187866
    },
    {
      "auxiliary_loss_clip": 0.0105441,
      "auxiliary_loss_mlp": 0.01036532,
      "balance_loss_clip": 1.02976036,
      "balance_loss_mlp": 1.02361608,
      "epoch": 0.9082819780550128,
      "flos": 24421302846720.0,
      "grad_norm": 1.649148060120071,
      "language_loss": 0.80815518,
      "learning_rate": 8.754174284066462e-08,
      "loss": 0.82906461,
      "num_input_tokens_seen": 325874890,
      "step": 15107,
      "time_per_iteration": 2.6020498275756836
    },
    {
      "auxiliary_loss_clip": 0.01009491,
      "auxiliary_loss_mlp": 0.01000746,
      "balance_loss_clip": 1.01115322,
      "balance_loss_mlp": 0.99954224,
      "epoch": 0.9083421013076808,
      "flos": 59609704872960.0,
      "grad_norm": 0.8157479423913526,
      "language_loss": 0.59753883,
      "learning_rate": 8.742781523089205e-08,
      "loss": 0.61764115,
      "num_input_tokens_seen": 325935835,
      "step": 15108,
      "time_per_iteration": 3.0848288536071777
    },
    {
      "auxiliary_loss_clip": 0.01080809,
      "auxiliary_loss_mlp": 0.01025019,
      "balance_loss_clip": 1.03317761,
      "balance_loss_mlp": 1.01324117,
      "epoch": 0.9084022245603487,
      "flos": 33620216100480.0,
      "grad_norm": 1.6736156408161018,
      "language_loss": 0.73613268,
      "learning_rate": 8.73139601460482e-08,
      "loss": 0.75719094,
      "num_input_tokens_seen": 325958035,
      "step": 15109,
      "time_per_iteration": 2.6182143688201904
    },
    {
      "auxiliary_loss_clip": 0.01068586,
      "auxiliary_loss_mlp": 0.01027044,
      "balance_loss_clip": 1.03318846,
      "balance_loss_mlp": 1.01596928,
      "epoch": 0.9084623478130167,
      "flos": 24971705925120.0,
      "grad_norm": 1.8677951231272971,
      "language_loss": 0.71138507,
      "learning_rate": 8.720017759045073e-08,
      "loss": 0.73234141,
      "num_input_tokens_seen": 325979870,
      "step": 15110,
      "time_per_iteration": 2.573913812637329
    },
    {
      "auxiliary_loss_clip": 0.01073554,
      "auxiliary_loss_mlp": 0.01032883,
      "balance_loss_clip": 1.0303967,
      "balance_loss_mlp": 1.02009189,
      "epoch": 0.9085224710656846,
      "flos": 31461804869760.0,
      "grad_norm": 2.038602390331913,
      "language_loss": 0.68802583,
      "learning_rate": 8.708646756841421e-08,
      "loss": 0.70909023,
      "num_input_tokens_seen": 325998245,
      "step": 15111,
      "time_per_iteration": 4.00778341293335
    },
    {
      "auxiliary_loss_clip": 0.01006004,
      "auxiliary_loss_mlp": 0.01006885,
      "balance_loss_clip": 1.00628424,
      "balance_loss_mlp": 1.00579977,
      "epoch": 0.9085825943183526,
      "flos": 64917012867840.0,
      "grad_norm": 0.6884383534786601,
      "language_loss": 0.51723611,
      "learning_rate": 8.697283008425026e-08,
      "loss": 0.53736496,
      "num_input_tokens_seen": 326061770,
      "step": 15112,
      "time_per_iteration": 3.186412811279297
    },
    {
      "auxiliary_loss_clip": 0.01091914,
      "auxiliary_loss_mlp": 0.01032619,
      "balance_loss_clip": 1.03230214,
      "balance_loss_mlp": 1.02054334,
      "epoch": 0.9086427175710206,
      "flos": 18953221576320.0,
      "grad_norm": 9.767465113500998,
      "language_loss": 0.70166671,
      "learning_rate": 8.685926514226837e-08,
      "loss": 0.72291207,
      "num_input_tokens_seen": 326080945,
      "step": 15113,
      "time_per_iteration": 2.4627254009246826
    },
    {
      "auxiliary_loss_clip": 0.01091499,
      "auxiliary_loss_mlp": 0.01030208,
      "balance_loss_clip": 1.03477097,
      "balance_loss_mlp": 1.01832867,
      "epoch": 0.9087028408236886,
      "flos": 34014873807360.0,
      "grad_norm": 2.1242727700146475,
      "language_loss": 0.79041469,
      "learning_rate": 8.674577274677508e-08,
      "loss": 0.8116318,
      "num_input_tokens_seen": 326100630,
      "step": 15114,
      "time_per_iteration": 2.5931005477905273
    },
    {
      "auxiliary_loss_clip": 0.01064346,
      "auxiliary_loss_mlp": 0.01034602,
      "balance_loss_clip": 1.03481007,
      "balance_loss_mlp": 1.02160263,
      "epoch": 0.9087629640763565,
      "flos": 21944580266880.0,
      "grad_norm": 2.0580373948365507,
      "language_loss": 0.70347321,
      "learning_rate": 8.663235290207405e-08,
      "loss": 0.72446263,
      "num_input_tokens_seen": 326120145,
      "step": 15115,
      "time_per_iteration": 2.5566980838775635
    },
    {
      "auxiliary_loss_clip": 0.01080968,
      "auxiliary_loss_mlp": 0.01032867,
      "balance_loss_clip": 1.03712893,
      "balance_loss_mlp": 1.01973581,
      "epoch": 0.9088230873290245,
      "flos": 21762908254080.0,
      "grad_norm": 3.627793528067355,
      "language_loss": 0.65902805,
      "learning_rate": 8.651900561246561e-08,
      "loss": 0.68016648,
      "num_input_tokens_seen": 326140715,
      "step": 15116,
      "time_per_iteration": 2.5683786869049072
    },
    {
      "auxiliary_loss_clip": 0.01101651,
      "auxiliary_loss_mlp": 0.01035015,
      "balance_loss_clip": 1.03554952,
      "balance_loss_mlp": 1.02292705,
      "epoch": 0.9088832105816925,
      "flos": 21541267382400.0,
      "grad_norm": 1.562272263255014,
      "language_loss": 0.6928491,
      "learning_rate": 8.640573088224812e-08,
      "loss": 0.71421576,
      "num_input_tokens_seen": 326159130,
      "step": 15117,
      "time_per_iteration": 2.447754144668579
    },
    {
      "auxiliary_loss_clip": 0.0106322,
      "auxiliary_loss_mlp": 0.01026243,
      "balance_loss_clip": 1.03549612,
      "balance_loss_mlp": 1.01520419,
      "epoch": 0.9089433338343604,
      "flos": 25996704428160.0,
      "grad_norm": 1.5826588044783139,
      "language_loss": 0.74354196,
      "learning_rate": 8.629252871571745e-08,
      "loss": 0.7644366,
      "num_input_tokens_seen": 326181375,
      "step": 15118,
      "time_per_iteration": 2.610609292984009
    },
    {
      "auxiliary_loss_clip": 0.01078749,
      "auxiliary_loss_mlp": 0.01032051,
      "balance_loss_clip": 1.03317881,
      "balance_loss_mlp": 1.01863945,
      "epoch": 0.9090034570870285,
      "flos": 21178426147200.0,
      "grad_norm": 2.007914693603318,
      "language_loss": 0.73014665,
      "learning_rate": 8.617939911716554e-08,
      "loss": 0.75125468,
      "num_input_tokens_seen": 326199740,
      "step": 15119,
      "time_per_iteration": 2.481048583984375
    },
    {
      "auxiliary_loss_clip": 0.01067701,
      "auxiliary_loss_mlp": 0.01030875,
      "balance_loss_clip": 1.03368413,
      "balance_loss_mlp": 1.01680875,
      "epoch": 0.9090635803396964,
      "flos": 16141811045760.0,
      "grad_norm": 2.697227891742029,
      "language_loss": 0.71587789,
      "learning_rate": 8.60663420908827e-08,
      "loss": 0.73686373,
      "num_input_tokens_seen": 326214350,
      "step": 15120,
      "time_per_iteration": 2.515573263168335
    },
    {
      "auxiliary_loss_clip": 0.0110395,
      "auxiliary_loss_mlp": 0.00782394,
      "balance_loss_clip": 1.03485262,
      "balance_loss_mlp": 1.00761461,
      "epoch": 0.9091237035923644,
      "flos": 20591537829120.0,
      "grad_norm": 2.1942164743956525,
      "language_loss": 0.6586715,
      "learning_rate": 8.595335764115596e-08,
      "loss": 0.67753494,
      "num_input_tokens_seen": 326234580,
      "step": 15121,
      "time_per_iteration": 2.473665952682495
    },
    {
      "auxiliary_loss_clip": 0.01092707,
      "auxiliary_loss_mlp": 0.01035138,
      "balance_loss_clip": 1.03364062,
      "balance_loss_mlp": 1.02275777,
      "epoch": 0.9091838268450323,
      "flos": 52227760164480.0,
      "grad_norm": 1.7861399681198902,
      "language_loss": 0.70278525,
      "learning_rate": 8.58404457722699e-08,
      "loss": 0.72406369,
      "num_input_tokens_seen": 326259080,
      "step": 15122,
      "time_per_iteration": 2.7621140480041504
    },
    {
      "auxiliary_loss_clip": 0.01048569,
      "auxiliary_loss_mlp": 0.01030817,
      "balance_loss_clip": 1.0309819,
      "balance_loss_mlp": 1.01899111,
      "epoch": 0.9092439500977003,
      "flos": 20559613616640.0,
      "grad_norm": 1.2743930714933893,
      "language_loss": 0.74576569,
      "learning_rate": 8.572760648850575e-08,
      "loss": 0.7665596,
      "num_input_tokens_seen": 326280175,
      "step": 15123,
      "time_per_iteration": 2.5868513584136963
    },
    {
      "auxiliary_loss_clip": 0.01089351,
      "auxiliary_loss_mlp": 0.01033015,
      "balance_loss_clip": 1.03369999,
      "balance_loss_mlp": 1.02170181,
      "epoch": 0.9093040733503682,
      "flos": 28617859595520.0,
      "grad_norm": 2.204410803344162,
      "language_loss": 0.75456792,
      "learning_rate": 8.561483979414253e-08,
      "loss": 0.77579153,
      "num_input_tokens_seen": 326297990,
      "step": 15124,
      "time_per_iteration": 2.5146567821502686
    },
    {
      "auxiliary_loss_clip": 0.01086514,
      "auxiliary_loss_mlp": 0.01031743,
      "balance_loss_clip": 1.0334034,
      "balance_loss_mlp": 1.01957774,
      "epoch": 0.9093641966030362,
      "flos": 23440187784960.0,
      "grad_norm": 2.029068599140535,
      "language_loss": 0.72301865,
      "learning_rate": 8.55021456934566e-08,
      "loss": 0.74420124,
      "num_input_tokens_seen": 326316735,
      "step": 15125,
      "time_per_iteration": 2.4909214973449707
    },
    {
      "auxiliary_loss_clip": 0.01062545,
      "auxiliary_loss_mlp": 0.01032916,
      "balance_loss_clip": 1.03553414,
      "balance_loss_mlp": 1.02117991,
      "epoch": 0.9094243198557042,
      "flos": 16800197385600.0,
      "grad_norm": 1.6175448067906075,
      "language_loss": 0.7901479,
      "learning_rate": 8.538952419072143e-08,
      "loss": 0.81110251,
      "num_input_tokens_seen": 326334370,
      "step": 15126,
      "time_per_iteration": 2.5535123348236084
    },
    {
      "auxiliary_loss_clip": 0.01065379,
      "auxiliary_loss_mlp": 0.01031435,
      "balance_loss_clip": 1.03478539,
      "balance_loss_mlp": 1.01938868,
      "epoch": 0.9094844431083722,
      "flos": 24273278899200.0,
      "grad_norm": 1.7147891721318076,
      "language_loss": 0.75339663,
      "learning_rate": 8.527697529020694e-08,
      "loss": 0.77436471,
      "num_input_tokens_seen": 326353435,
      "step": 15127,
      "time_per_iteration": 2.567479133605957
    },
    {
      "auxiliary_loss_clip": 0.01031239,
      "auxiliary_loss_mlp": 0.01033568,
      "balance_loss_clip": 1.03054667,
      "balance_loss_mlp": 1.02147448,
      "epoch": 0.9095445663610401,
      "flos": 21944652094080.0,
      "grad_norm": 1.8501096247579176,
      "language_loss": 0.62789106,
      "learning_rate": 8.516449899618173e-08,
      "loss": 0.64853919,
      "num_input_tokens_seen": 326371810,
      "step": 15128,
      "time_per_iteration": 2.65617299079895
    },
    {
      "auxiliary_loss_clip": 0.01068889,
      "auxiliary_loss_mlp": 0.01027197,
      "balance_loss_clip": 1.03334105,
      "balance_loss_mlp": 1.01565135,
      "epoch": 0.9096046896137081,
      "flos": 19792848965760.0,
      "grad_norm": 1.6704405515928746,
      "language_loss": 0.76270425,
      "learning_rate": 8.505209531291013e-08,
      "loss": 0.78366506,
      "num_input_tokens_seen": 326391380,
      "step": 15129,
      "time_per_iteration": 2.5499823093414307
    },
    {
      "auxiliary_loss_clip": 0.01091158,
      "auxiliary_loss_mlp": 0.01029739,
      "balance_loss_clip": 1.03406358,
      "balance_loss_mlp": 1.01796699,
      "epoch": 0.909664812866376,
      "flos": 22638087129600.0,
      "grad_norm": 1.8499632899204919,
      "language_loss": 0.832551,
      "learning_rate": 8.49397642446552e-08,
      "loss": 0.85376,
      "num_input_tokens_seen": 326408800,
      "step": 15130,
      "time_per_iteration": 2.5018978118896484
    },
    {
      "auxiliary_loss_clip": 0.01082286,
      "auxiliary_loss_mlp": 0.01034698,
      "balance_loss_clip": 1.03501141,
      "balance_loss_mlp": 1.02225828,
      "epoch": 0.909724936119044,
      "flos": 39852153020160.0,
      "grad_norm": 1.747561250765313,
      "language_loss": 0.75302672,
      "learning_rate": 8.482750579567644e-08,
      "loss": 0.77419651,
      "num_input_tokens_seen": 326431565,
      "step": 15131,
      "time_per_iteration": 2.6672606468200684
    },
    {
      "auxiliary_loss_clip": 0.01077087,
      "auxiliary_loss_mlp": 0.01030815,
      "balance_loss_clip": 1.03368163,
      "balance_loss_mlp": 1.01867342,
      "epoch": 0.9097850593717121,
      "flos": 35071616954880.0,
      "grad_norm": 1.800575730294283,
      "language_loss": 0.59496212,
      "learning_rate": 8.471531997023085e-08,
      "loss": 0.61604118,
      "num_input_tokens_seen": 326451715,
      "step": 15132,
      "time_per_iteration": 2.6267762184143066
    },
    {
      "auxiliary_loss_clip": 0.01070752,
      "auxiliary_loss_mlp": 0.01028844,
      "balance_loss_clip": 1.03813553,
      "balance_loss_mlp": 1.01775742,
      "epoch": 0.90984518262438,
      "flos": 23367468700800.0,
      "grad_norm": 1.391779471457768,
      "language_loss": 0.82482231,
      "learning_rate": 8.460320677257193e-08,
      "loss": 0.84581828,
      "num_input_tokens_seen": 326470855,
      "step": 15133,
      "time_per_iteration": 2.5484538078308105
    },
    {
      "auxiliary_loss_clip": 0.01080251,
      "auxiliary_loss_mlp": 0.01030619,
      "balance_loss_clip": 1.03248644,
      "balance_loss_mlp": 1.01858449,
      "epoch": 0.909905305877048,
      "flos": 27523302405120.0,
      "grad_norm": 1.8392812666159402,
      "language_loss": 0.73559695,
      "learning_rate": 8.449116620695118e-08,
      "loss": 0.7567057,
      "num_input_tokens_seen": 326490480,
      "step": 15134,
      "time_per_iteration": 2.6072652339935303
    },
    {
      "auxiliary_loss_clip": 0.01075768,
      "auxiliary_loss_mlp": 0.01036547,
      "balance_loss_clip": 1.03649926,
      "balance_loss_mlp": 1.02408409,
      "epoch": 0.9099654291297159,
      "flos": 24347865490560.0,
      "grad_norm": 1.599315664024651,
      "language_loss": 0.72764778,
      "learning_rate": 8.437919827761786e-08,
      "loss": 0.74877095,
      "num_input_tokens_seen": 326509445,
      "step": 15135,
      "time_per_iteration": 2.585002899169922
    },
    {
      "auxiliary_loss_clip": 0.01091161,
      "auxiliary_loss_mlp": 0.01030736,
      "balance_loss_clip": 1.03495646,
      "balance_loss_mlp": 1.01922059,
      "epoch": 0.9100255523823839,
      "flos": 21215234609280.0,
      "grad_norm": 1.6236614517384516,
      "language_loss": 0.69682944,
      "learning_rate": 8.426730298881702e-08,
      "loss": 0.71804833,
      "num_input_tokens_seen": 326528380,
      "step": 15136,
      "time_per_iteration": 2.4719502925872803
    },
    {
      "auxiliary_loss_clip": 0.00997516,
      "auxiliary_loss_mlp": 0.01005145,
      "balance_loss_clip": 1.00681674,
      "balance_loss_mlp": 1.0041554,
      "epoch": 0.9100856756350518,
      "flos": 46052276446080.0,
      "grad_norm": 0.8256453028880957,
      "language_loss": 0.59348655,
      "learning_rate": 8.415548034479214e-08,
      "loss": 0.61351311,
      "num_input_tokens_seen": 326576940,
      "step": 15137,
      "time_per_iteration": 2.9014508724212646
    },
    {
      "auxiliary_loss_clip": 0.01093651,
      "auxiliary_loss_mlp": 0.01036751,
      "balance_loss_clip": 1.0349195,
      "balance_loss_mlp": 1.0250026,
      "epoch": 0.9101457988877198,
      "flos": 20229917656320.0,
      "grad_norm": 1.6483095579675058,
      "language_loss": 0.82583177,
      "learning_rate": 8.40437303497834e-08,
      "loss": 0.84713578,
      "num_input_tokens_seen": 326596100,
      "step": 15138,
      "time_per_iteration": 2.477790594100952
    },
    {
      "auxiliary_loss_clip": 0.01089231,
      "auxiliary_loss_mlp": 0.01025845,
      "balance_loss_clip": 1.03542757,
      "balance_loss_mlp": 1.01465178,
      "epoch": 0.9102059221403878,
      "flos": 26615157822720.0,
      "grad_norm": 1.428094393876634,
      "language_loss": 0.81053406,
      "learning_rate": 8.39320530080283e-08,
      "loss": 0.83168483,
      "num_input_tokens_seen": 326615700,
      "step": 15139,
      "time_per_iteration": 2.541222095489502
    },
    {
      "auxiliary_loss_clip": 0.01069719,
      "auxiliary_loss_mlp": 0.01034786,
      "balance_loss_clip": 1.03379095,
      "balance_loss_mlp": 1.02340126,
      "epoch": 0.9102660453930558,
      "flos": 21908561904000.0,
      "grad_norm": 1.8002744746847454,
      "language_loss": 0.77578467,
      "learning_rate": 8.382044832376167e-08,
      "loss": 0.7968297,
      "num_input_tokens_seen": 326635905,
      "step": 15140,
      "time_per_iteration": 3.91365909576416
    },
    {
      "auxiliary_loss_clip": 0.01102766,
      "auxiliary_loss_mlp": 0.01026757,
      "balance_loss_clip": 1.03490579,
      "balance_loss_mlp": 1.01506853,
      "epoch": 0.9103261686457237,
      "flos": 36176660916480.0,
      "grad_norm": 2.862114362268329,
      "language_loss": 0.66864586,
      "learning_rate": 8.370891630121569e-08,
      "loss": 0.68994105,
      "num_input_tokens_seen": 326661855,
      "step": 15141,
      "time_per_iteration": 2.599681854248047
    },
    {
      "auxiliary_loss_clip": 0.01093797,
      "auxiliary_loss_mlp": 0.01034576,
      "balance_loss_clip": 1.03476024,
      "balance_loss_mlp": 1.02316809,
      "epoch": 0.9103862918983917,
      "flos": 23878549365120.0,
      "grad_norm": 1.677236535636131,
      "language_loss": 0.75029504,
      "learning_rate": 8.359745694462005e-08,
      "loss": 0.77157879,
      "num_input_tokens_seen": 326679320,
      "step": 15142,
      "time_per_iteration": 5.28966760635376
    },
    {
      "auxiliary_loss_clip": 0.01066438,
      "auxiliary_loss_mlp": 0.0103909,
      "balance_loss_clip": 1.03194654,
      "balance_loss_mlp": 1.02685905,
      "epoch": 0.9104464151510596,
      "flos": 14939521989120.0,
      "grad_norm": 1.657665248418674,
      "language_loss": 0.64374077,
      "learning_rate": 8.348607025820076e-08,
      "loss": 0.66479599,
      "num_input_tokens_seen": 326698110,
      "step": 15143,
      "time_per_iteration": 2.5161027908325195
    },
    {
      "auxiliary_loss_clip": 0.01103388,
      "auxiliary_loss_mlp": 0.0103391,
      "balance_loss_clip": 1.03353333,
      "balance_loss_mlp": 1.02122581,
      "epoch": 0.9105065384037276,
      "flos": 33655803500160.0,
      "grad_norm": 1.8663189342602347,
      "language_loss": 0.61063099,
      "learning_rate": 8.337475624618152e-08,
      "loss": 0.63200396,
      "num_input_tokens_seen": 326718370,
      "step": 15144,
      "time_per_iteration": 2.53768253326416
    },
    {
      "auxiliary_loss_clip": 0.01063569,
      "auxiliary_loss_mlp": 0.01025234,
      "balance_loss_clip": 1.03245962,
      "balance_loss_mlp": 1.01327157,
      "epoch": 0.9105666616563957,
      "flos": 24316695463680.0,
      "grad_norm": 1.6390736756841775,
      "language_loss": 0.71032459,
      "learning_rate": 8.326351491278382e-08,
      "loss": 0.73121262,
      "num_input_tokens_seen": 326738445,
      "step": 15145,
      "time_per_iteration": 2.5526223182678223
    },
    {
      "auxiliary_loss_clip": 0.01051169,
      "auxiliary_loss_mlp": 0.01032588,
      "balance_loss_clip": 1.03341794,
      "balance_loss_mlp": 1.02104902,
      "epoch": 0.9106267849090636,
      "flos": 29971692132480.0,
      "grad_norm": 1.536228728431362,
      "language_loss": 0.70821095,
      "learning_rate": 8.315234626222545e-08,
      "loss": 0.72904849,
      "num_input_tokens_seen": 326758855,
      "step": 15146,
      "time_per_iteration": 2.6608612537384033
    },
    {
      "auxiliary_loss_clip": 0.01080919,
      "auxiliary_loss_mlp": 0.01029124,
      "balance_loss_clip": 1.03255987,
      "balance_loss_mlp": 1.01796055,
      "epoch": 0.9106869081617316,
      "flos": 25337743470720.0,
      "grad_norm": 1.912540595815008,
      "language_loss": 0.72928202,
      "learning_rate": 8.304125029872233e-08,
      "loss": 0.75038254,
      "num_input_tokens_seen": 326777140,
      "step": 15147,
      "time_per_iteration": 2.5565829277038574
    },
    {
      "auxiliary_loss_clip": 0.01074786,
      "auxiliary_loss_mlp": 0.01027563,
      "balance_loss_clip": 1.03389096,
      "balance_loss_mlp": 1.01560068,
      "epoch": 0.9107470314143995,
      "flos": 18187031543040.0,
      "grad_norm": 1.8252376938892632,
      "language_loss": 0.80517668,
      "learning_rate": 8.293022702648711e-08,
      "loss": 0.82620019,
      "num_input_tokens_seen": 326794070,
      "step": 15148,
      "time_per_iteration": 2.5213913917541504
    },
    {
      "auxiliary_loss_clip": 0.01070528,
      "auxiliary_loss_mlp": 0.01034807,
      "balance_loss_clip": 1.0331763,
      "balance_loss_mlp": 1.02284479,
      "epoch": 0.9108071546670675,
      "flos": 23550828652800.0,
      "grad_norm": 2.4584743096862516,
      "language_loss": 0.67599833,
      "learning_rate": 8.281927644972996e-08,
      "loss": 0.69705176,
      "num_input_tokens_seen": 326814695,
      "step": 15149,
      "time_per_iteration": 2.5717239379882812
    },
    {
      "auxiliary_loss_clip": 0.01102433,
      "auxiliary_loss_mlp": 0.01027684,
      "balance_loss_clip": 1.03453052,
      "balance_loss_mlp": 1.01523304,
      "epoch": 0.9108672779197354,
      "flos": 25630307746560.0,
      "grad_norm": 2.422024116677399,
      "language_loss": 0.63487017,
      "learning_rate": 8.270839857265776e-08,
      "loss": 0.65617138,
      "num_input_tokens_seen": 326835295,
      "step": 15150,
      "time_per_iteration": 3.854471206665039
    },
    {
      "auxiliary_loss_clip": 0.0106817,
      "auxiliary_loss_mlp": 0.01032915,
      "balance_loss_clip": 1.03401017,
      "balance_loss_mlp": 1.02098751,
      "epoch": 0.9109274011724035,
      "flos": 22339094319360.0,
      "grad_norm": 1.7268600322223937,
      "language_loss": 0.72377455,
      "learning_rate": 8.259759339947514e-08,
      "loss": 0.74478543,
      "num_input_tokens_seen": 326853350,
      "step": 15151,
      "time_per_iteration": 2.5510141849517822
    },
    {
      "auxiliary_loss_clip": 0.01090456,
      "auxiliary_loss_mlp": 0.01026462,
      "balance_loss_clip": 1.03387249,
      "balance_loss_mlp": 1.01489866,
      "epoch": 0.9109875244250714,
      "flos": 26688200129280.0,
      "grad_norm": 1.6021136905513762,
      "language_loss": 0.64389014,
      "learning_rate": 8.248686093438429e-08,
      "loss": 0.66505933,
      "num_input_tokens_seen": 326873425,
      "step": 15152,
      "time_per_iteration": 2.543565034866333
    },
    {
      "auxiliary_loss_clip": 0.01083152,
      "auxiliary_loss_mlp": 0.00782588,
      "balance_loss_clip": 1.03503609,
      "balance_loss_mlp": 1.00894105,
      "epoch": 0.9110476476777394,
      "flos": 22930112701440.0,
      "grad_norm": 2.4409726260036853,
      "language_loss": 0.73054779,
      "learning_rate": 8.23762011815834e-08,
      "loss": 0.74920523,
      "num_input_tokens_seen": 326893455,
      "step": 15153,
      "time_per_iteration": 2.5526320934295654
    },
    {
      "auxiliary_loss_clip": 0.01063334,
      "auxiliary_loss_mlp": 0.01043237,
      "balance_loss_clip": 1.03178906,
      "balance_loss_mlp": 1.02872348,
      "epoch": 0.9111077709304073,
      "flos": 13472857854720.0,
      "grad_norm": 1.8690995841369131,
      "language_loss": 0.72269499,
      "learning_rate": 8.226561414526956e-08,
      "loss": 0.7437607,
      "num_input_tokens_seen": 326910210,
      "step": 15154,
      "time_per_iteration": 2.505903482437134
    },
    {
      "auxiliary_loss_clip": 0.01081029,
      "auxiliary_loss_mlp": 0.01031253,
      "balance_loss_clip": 1.03682518,
      "balance_loss_mlp": 1.01987433,
      "epoch": 0.9111678941830753,
      "flos": 20850561780480.0,
      "grad_norm": 1.7614167318219531,
      "language_loss": 0.81597781,
      "learning_rate": 8.215509982963564e-08,
      "loss": 0.83710063,
      "num_input_tokens_seen": 326929350,
      "step": 15155,
      "time_per_iteration": 2.521428108215332
    },
    {
      "auxiliary_loss_clip": 0.01090023,
      "auxiliary_loss_mlp": 0.0102899,
      "balance_loss_clip": 1.03670216,
      "balance_loss_mlp": 1.01714087,
      "epoch": 0.9112280174357432,
      "flos": 19682244011520.0,
      "grad_norm": 1.4088192653187965,
      "language_loss": 0.59803849,
      "learning_rate": 8.204465823887252e-08,
      "loss": 0.6192286,
      "num_input_tokens_seen": 326949060,
      "step": 15156,
      "time_per_iteration": 2.4693007469177246
    },
    {
      "auxiliary_loss_clip": 0.01093334,
      "auxiliary_loss_mlp": 0.01029376,
      "balance_loss_clip": 1.03232396,
      "balance_loss_mlp": 1.01639414,
      "epoch": 0.9112881406884112,
      "flos": 25447163276160.0,
      "grad_norm": 1.7449301427551074,
      "language_loss": 0.73997718,
      "learning_rate": 8.193428937716796e-08,
      "loss": 0.7612043,
      "num_input_tokens_seen": 326968950,
      "step": 15157,
      "time_per_iteration": 2.5131356716156006
    },
    {
      "auxiliary_loss_clip": 0.01059676,
      "auxiliary_loss_mlp": 0.01031791,
      "balance_loss_clip": 1.03225136,
      "balance_loss_mlp": 1.02020955,
      "epoch": 0.9113482639410793,
      "flos": 33066975847680.0,
      "grad_norm": 1.5867178827115824,
      "language_loss": 0.59214401,
      "learning_rate": 8.182399324870747e-08,
      "loss": 0.61305869,
      "num_input_tokens_seen": 326989455,
      "step": 15158,
      "time_per_iteration": 2.656273365020752
    },
    {
      "auxiliary_loss_clip": 0.01044519,
      "auxiliary_loss_mlp": 0.01030259,
      "balance_loss_clip": 1.03401721,
      "balance_loss_mlp": 1.01932716,
      "epoch": 0.9114083871937472,
      "flos": 21835591424640.0,
      "grad_norm": 1.5205388243803153,
      "language_loss": 0.677935,
      "learning_rate": 8.171376985767375e-08,
      "loss": 0.69868279,
      "num_input_tokens_seen": 327009640,
      "step": 15159,
      "time_per_iteration": 2.621690273284912
    },
    {
      "auxiliary_loss_clip": 0.01082969,
      "auxiliary_loss_mlp": 0.01030225,
      "balance_loss_clip": 1.03413177,
      "balance_loss_mlp": 1.01856029,
      "epoch": 0.9114685104464152,
      "flos": 27088999061760.0,
      "grad_norm": 2.108185650842786,
      "language_loss": 0.78538024,
      "learning_rate": 8.160361920824588e-08,
      "loss": 0.80651224,
      "num_input_tokens_seen": 327027690,
      "step": 15160,
      "time_per_iteration": 2.5487401485443115
    },
    {
      "auxiliary_loss_clip": 0.01105515,
      "auxiliary_loss_mlp": 0.01027761,
      "balance_loss_clip": 1.03634834,
      "balance_loss_mlp": 1.01484466,
      "epoch": 0.9115286336990831,
      "flos": 17967042696960.0,
      "grad_norm": 1.6818077181670759,
      "language_loss": 0.6919049,
      "learning_rate": 8.149354130460073e-08,
      "loss": 0.71323764,
      "num_input_tokens_seen": 327045915,
      "step": 15161,
      "time_per_iteration": 2.4342257976531982
    },
    {
      "auxiliary_loss_clip": 0.01059331,
      "auxiliary_loss_mlp": 0.01035883,
      "balance_loss_clip": 1.03448546,
      "balance_loss_mlp": 1.02217984,
      "epoch": 0.9115887569517511,
      "flos": 22929861306240.0,
      "grad_norm": 1.7959963483469137,
      "language_loss": 0.76191264,
      "learning_rate": 8.138353615091321e-08,
      "loss": 0.78286481,
      "num_input_tokens_seen": 327066355,
      "step": 15162,
      "time_per_iteration": 2.575934410095215
    },
    {
      "auxiliary_loss_clip": 0.01085865,
      "auxiliary_loss_mlp": 0.01035348,
      "balance_loss_clip": 1.03730273,
      "balance_loss_mlp": 1.02314091,
      "epoch": 0.911648880204419,
      "flos": 23988436047360.0,
      "grad_norm": 1.844987188877883,
      "language_loss": 0.66687107,
      "learning_rate": 8.127360375135395e-08,
      "loss": 0.68808317,
      "num_input_tokens_seen": 327086735,
      "step": 15163,
      "time_per_iteration": 2.539062023162842
    },
    {
      "auxiliary_loss_clip": 0.01061531,
      "auxiliary_loss_mlp": 0.01032468,
      "balance_loss_clip": 1.03390074,
      "balance_loss_mlp": 1.02007008,
      "epoch": 0.911709003457087,
      "flos": 17055306754560.0,
      "grad_norm": 2.347910428817761,
      "language_loss": 0.70756608,
      "learning_rate": 8.116374411009186e-08,
      "loss": 0.72850609,
      "num_input_tokens_seen": 327104035,
      "step": 15164,
      "time_per_iteration": 2.5566039085388184
    },
    {
      "auxiliary_loss_clip": 0.01103031,
      "auxiliary_loss_mlp": 0.01034205,
      "balance_loss_clip": 1.03769982,
      "balance_loss_mlp": 1.02252269,
      "epoch": 0.911769126709755,
      "flos": 21653344794240.0,
      "grad_norm": 1.847118481264255,
      "language_loss": 0.75997305,
      "learning_rate": 8.105395723129315e-08,
      "loss": 0.78134549,
      "num_input_tokens_seen": 327124370,
      "step": 15165,
      "time_per_iteration": 2.4734737873077393
    },
    {
      "auxiliary_loss_clip": 0.01088092,
      "auxiliary_loss_mlp": 0.01034097,
      "balance_loss_clip": 1.0330714,
      "balance_loss_mlp": 1.02165174,
      "epoch": 0.911829249962423,
      "flos": 24790321221120.0,
      "grad_norm": 2.244263561887781,
      "language_loss": 0.71766567,
      "learning_rate": 8.094424311912074e-08,
      "loss": 0.73888755,
      "num_input_tokens_seen": 327140915,
      "step": 15166,
      "time_per_iteration": 2.493412733078003
    },
    {
      "auxiliary_loss_clip": 0.01063059,
      "auxiliary_loss_mlp": 0.01034988,
      "balance_loss_clip": 1.03325582,
      "balance_loss_mlp": 1.02189279,
      "epoch": 0.9118893732150909,
      "flos": 20959406968320.0,
      "grad_norm": 1.795201541151311,
      "language_loss": 0.72947091,
      "learning_rate": 8.083460177773482e-08,
      "loss": 0.75045133,
      "num_input_tokens_seen": 327158940,
      "step": 15167,
      "time_per_iteration": 2.576633930206299
    },
    {
      "auxiliary_loss_clip": 0.01016228,
      "auxiliary_loss_mlp": 0.01000736,
      "balance_loss_clip": 1.01231146,
      "balance_loss_mlp": 0.9996509,
      "epoch": 0.9119494964677589,
      "flos": 67917385872000.0,
      "grad_norm": 0.7703537655122893,
      "language_loss": 0.65596569,
      "learning_rate": 8.072503321129298e-08,
      "loss": 0.67613536,
      "num_input_tokens_seen": 327217450,
      "step": 15168,
      "time_per_iteration": 3.1078052520751953
    },
    {
      "auxiliary_loss_clip": 0.01072494,
      "auxiliary_loss_mlp": 0.01027145,
      "balance_loss_clip": 1.03497112,
      "balance_loss_mlp": 1.01594496,
      "epoch": 0.9120096197204268,
      "flos": 18551524803840.0,
      "grad_norm": 3.193006047904742,
      "language_loss": 0.78030074,
      "learning_rate": 8.061553742395033e-08,
      "loss": 0.80129719,
      "num_input_tokens_seen": 327233905,
      "step": 15169,
      "time_per_iteration": 2.488603353500366
    },
    {
      "auxiliary_loss_clip": 0.01091745,
      "auxiliary_loss_mlp": 0.01027541,
      "balance_loss_clip": 1.03409445,
      "balance_loss_mlp": 1.01588237,
      "epoch": 0.9120697429730948,
      "flos": 19025725178880.0,
      "grad_norm": 1.7121976464685618,
      "language_loss": 0.82143992,
      "learning_rate": 8.05061144198591e-08,
      "loss": 0.84263277,
      "num_input_tokens_seen": 327252430,
      "step": 15170,
      "time_per_iteration": 2.4807143211364746
    },
    {
      "auxiliary_loss_clip": 0.01095271,
      "auxiliary_loss_mlp": 0.0102926,
      "balance_loss_clip": 1.03615928,
      "balance_loss_mlp": 1.01717257,
      "epoch": 0.9121298662257629,
      "flos": 17163685065600.0,
      "grad_norm": 1.9761934699210726,
      "language_loss": 0.77158678,
      "learning_rate": 8.039676420316799e-08,
      "loss": 0.79283214,
      "num_input_tokens_seen": 327269215,
      "step": 15171,
      "time_per_iteration": 2.4437942504882812
    },
    {
      "auxiliary_loss_clip": 0.01027968,
      "auxiliary_loss_mlp": 0.01037364,
      "balance_loss_clip": 1.03072512,
      "balance_loss_mlp": 1.02438164,
      "epoch": 0.9121899894784308,
      "flos": 19682710888320.0,
      "grad_norm": 1.4028798080328464,
      "language_loss": 0.67115891,
      "learning_rate": 8.02874867780241e-08,
      "loss": 0.69181228,
      "num_input_tokens_seen": 327290320,
      "step": 15172,
      "time_per_iteration": 2.6797919273376465
    },
    {
      "auxiliary_loss_clip": 0.01075486,
      "auxiliary_loss_mlp": 0.01032537,
      "balance_loss_clip": 1.03506625,
      "balance_loss_mlp": 1.02062821,
      "epoch": 0.9122501127310988,
      "flos": 22235743912320.0,
      "grad_norm": 1.5788247703294447,
      "language_loss": 0.74518114,
      "learning_rate": 8.017828214857103e-08,
      "loss": 0.7662614,
      "num_input_tokens_seen": 327310150,
      "step": 15173,
      "time_per_iteration": 2.532552719116211
    },
    {
      "auxiliary_loss_clip": 0.01087266,
      "auxiliary_loss_mlp": 0.01030818,
      "balance_loss_clip": 1.03569365,
      "balance_loss_mlp": 1.01677513,
      "epoch": 0.9123102359837667,
      "flos": 15957122290560.0,
      "grad_norm": 2.221659098123988,
      "language_loss": 0.66076577,
      "learning_rate": 8.00691503189499e-08,
      "loss": 0.68194652,
      "num_input_tokens_seen": 327326660,
      "step": 15174,
      "time_per_iteration": 2.4776573181152344
    },
    {
      "auxiliary_loss_clip": 0.01091913,
      "auxiliary_loss_mlp": 0.01029494,
      "balance_loss_clip": 1.03490925,
      "balance_loss_mlp": 1.01621461,
      "epoch": 0.9123703592364347,
      "flos": 25155784149120.0,
      "grad_norm": 1.6220826741564516,
      "language_loss": 0.74784768,
      "learning_rate": 7.996009129329894e-08,
      "loss": 0.76906174,
      "num_input_tokens_seen": 327346700,
      "step": 15175,
      "time_per_iteration": 2.4994077682495117
    },
    {
      "auxiliary_loss_clip": 0.01019078,
      "auxiliary_loss_mlp": 0.01000308,
      "balance_loss_clip": 1.00687671,
      "balance_loss_mlp": 0.9993183,
      "epoch": 0.9124304824891026,
      "flos": 60801650812800.0,
      "grad_norm": 0.9696584009871491,
      "language_loss": 0.58397275,
      "learning_rate": 7.985110507575421e-08,
      "loss": 0.60416663,
      "num_input_tokens_seen": 327403050,
      "step": 15176,
      "time_per_iteration": 3.1411426067352295
    },
    {
      "auxiliary_loss_clip": 0.01080234,
      "auxiliary_loss_mlp": 0.01036183,
      "balance_loss_clip": 1.03386402,
      "balance_loss_mlp": 1.02427363,
      "epoch": 0.9124906057417707,
      "flos": 18150941352960.0,
      "grad_norm": 1.8382190823554097,
      "language_loss": 0.65431964,
      "learning_rate": 7.97421916704475e-08,
      "loss": 0.67548382,
      "num_input_tokens_seen": 327422225,
      "step": 15177,
      "time_per_iteration": 2.4917290210723877
    },
    {
      "auxiliary_loss_clip": 0.01073075,
      "auxiliary_loss_mlp": 0.01028824,
      "balance_loss_clip": 1.03448272,
      "balance_loss_mlp": 1.0172255,
      "epoch": 0.9125507289944386,
      "flos": 11686769049600.0,
      "grad_norm": 1.8047162626306705,
      "language_loss": 0.81074136,
      "learning_rate": 7.963335108150926e-08,
      "loss": 0.83176029,
      "num_input_tokens_seen": 327437025,
      "step": 15178,
      "time_per_iteration": 2.476388692855835
    },
    {
      "auxiliary_loss_clip": 0.01045116,
      "auxiliary_loss_mlp": 0.01034116,
      "balance_loss_clip": 1.03111315,
      "balance_loss_mlp": 1.02079463,
      "epoch": 0.9126108522471066,
      "flos": 17748813617280.0,
      "grad_norm": 1.9768309127783001,
      "language_loss": 0.78558117,
      "learning_rate": 7.952458331306711e-08,
      "loss": 0.80637348,
      "num_input_tokens_seen": 327453915,
      "step": 15179,
      "time_per_iteration": 3.9671101570129395
    },
    {
      "auxiliary_loss_clip": 0.01080072,
      "auxiliary_loss_mlp": 0.01030821,
      "balance_loss_clip": 1.03499341,
      "balance_loss_mlp": 1.01956165,
      "epoch": 0.9126709754997745,
      "flos": 27635738952960.0,
      "grad_norm": 2.6756077093910493,
      "language_loss": 0.68238622,
      "learning_rate": 7.941588836924507e-08,
      "loss": 0.70349514,
      "num_input_tokens_seen": 327474415,
      "step": 15180,
      "time_per_iteration": 3.973003387451172
    },
    {
      "auxiliary_loss_clip": 0.0109035,
      "auxiliary_loss_mlp": 0.01028379,
      "balance_loss_clip": 1.03345764,
      "balance_loss_mlp": 1.01732206,
      "epoch": 0.9127310987524425,
      "flos": 15924982596480.0,
      "grad_norm": 1.7818160170422686,
      "language_loss": 0.75396931,
      "learning_rate": 7.930726625416495e-08,
      "loss": 0.77515662,
      "num_input_tokens_seen": 327492750,
      "step": 15181,
      "time_per_iteration": 3.866295337677002
    },
    {
      "auxiliary_loss_clip": 0.0110832,
      "auxiliary_loss_mlp": 0.01031221,
      "balance_loss_clip": 1.03667021,
      "balance_loss_mlp": 1.01946712,
      "epoch": 0.9127912220051104,
      "flos": 21536885923200.0,
      "grad_norm": 1.6213925091167896,
      "language_loss": 0.74515462,
      "learning_rate": 7.919871697194614e-08,
      "loss": 0.76655006,
      "num_input_tokens_seen": 327509470,
      "step": 15182,
      "time_per_iteration": 2.4581503868103027
    },
    {
      "auxiliary_loss_clip": 0.0110567,
      "auxiliary_loss_mlp": 0.01031066,
      "balance_loss_clip": 1.03498054,
      "balance_loss_mlp": 1.01898992,
      "epoch": 0.9128513452577784,
      "flos": 24063561342720.0,
      "grad_norm": 1.4249246270273057,
      "language_loss": 0.76172614,
      "learning_rate": 7.909024052670421e-08,
      "loss": 0.78309351,
      "num_input_tokens_seen": 327530520,
      "step": 15183,
      "time_per_iteration": 2.4724552631378174
    },
    {
      "auxiliary_loss_clip": 0.01092117,
      "auxiliary_loss_mlp": 0.01030106,
      "balance_loss_clip": 1.03620696,
      "balance_loss_mlp": 1.01790452,
      "epoch": 0.9129114685104465,
      "flos": 16216469464320.0,
      "grad_norm": 2.3791410384019076,
      "language_loss": 0.76571149,
      "learning_rate": 7.898183692255256e-08,
      "loss": 0.78693366,
      "num_input_tokens_seen": 327546960,
      "step": 15184,
      "time_per_iteration": 2.4713759422302246
    },
    {
      "auxiliary_loss_clip": 0.01095092,
      "auxiliary_loss_mlp": 0.01032975,
      "balance_loss_clip": 1.03646791,
      "balance_loss_mlp": 1.02139401,
      "epoch": 0.9129715917631144,
      "flos": 19384364522880.0,
      "grad_norm": 2.5169228750993935,
      "language_loss": 0.74430025,
      "learning_rate": 7.887350616360233e-08,
      "loss": 0.76558089,
      "num_input_tokens_seen": 327564830,
      "step": 15185,
      "time_per_iteration": 2.4522485733032227
    },
    {
      "auxiliary_loss_clip": 0.01078801,
      "auxiliary_loss_mlp": 0.01028725,
      "balance_loss_clip": 1.03557575,
      "balance_loss_mlp": 1.01714396,
      "epoch": 0.9130317150157824,
      "flos": 20590460421120.0,
      "grad_norm": 2.8940181622671033,
      "language_loss": 0.685534,
      "learning_rate": 7.876524825396158e-08,
      "loss": 0.70660925,
      "num_input_tokens_seen": 327583675,
      "step": 15186,
      "time_per_iteration": 2.5118305683135986
    },
    {
      "auxiliary_loss_clip": 0.01085336,
      "auxiliary_loss_mlp": 0.01040961,
      "balance_loss_clip": 1.03480864,
      "balance_loss_mlp": 1.02652478,
      "epoch": 0.9130918382684503,
      "flos": 20189230525440.0,
      "grad_norm": 1.8318240751515744,
      "language_loss": 0.76988304,
      "learning_rate": 7.865706319773502e-08,
      "loss": 0.79114604,
      "num_input_tokens_seen": 327602280,
      "step": 15187,
      "time_per_iteration": 2.4929161071777344
    },
    {
      "auxiliary_loss_clip": 0.01103691,
      "auxiliary_loss_mlp": 0.00783387,
      "balance_loss_clip": 1.03455687,
      "balance_loss_mlp": 1.00981593,
      "epoch": 0.9131519615211183,
      "flos": 25556870390400.0,
      "grad_norm": 1.7353367447662638,
      "language_loss": 0.6571964,
      "learning_rate": 7.854895099902515e-08,
      "loss": 0.67606717,
      "num_input_tokens_seen": 327623515,
      "step": 15188,
      "time_per_iteration": 3.905729293823242
    },
    {
      "auxiliary_loss_clip": 0.01031391,
      "auxiliary_loss_mlp": 0.01039265,
      "balance_loss_clip": 1.03114414,
      "balance_loss_mlp": 1.02667058,
      "epoch": 0.9132120847737862,
      "flos": 17931563038080.0,
      "grad_norm": 1.710035710200227,
      "language_loss": 0.76360309,
      "learning_rate": 7.844091166193157e-08,
      "loss": 0.78430969,
      "num_input_tokens_seen": 327642875,
      "step": 15189,
      "time_per_iteration": 2.610316276550293
    },
    {
      "auxiliary_loss_clip": 0.01091991,
      "auxiliary_loss_mlp": 0.01027283,
      "balance_loss_clip": 1.03427374,
      "balance_loss_mlp": 1.01642907,
      "epoch": 0.9132722080264543,
      "flos": 20047635112320.0,
      "grad_norm": 2.025078922126337,
      "language_loss": 0.75336313,
      "learning_rate": 7.8332945190551e-08,
      "loss": 0.77455592,
      "num_input_tokens_seen": 327662450,
      "step": 15190,
      "time_per_iteration": 2.502211332321167
    },
    {
      "auxiliary_loss_clip": 0.01019262,
      "auxiliary_loss_mlp": 0.01001654,
      "balance_loss_clip": 1.00663352,
      "balance_loss_mlp": 1.00062871,
      "epoch": 0.9133323312791222,
      "flos": 70439967141120.0,
      "grad_norm": 0.7056753138125434,
      "language_loss": 0.57360995,
      "learning_rate": 7.822505158897797e-08,
      "loss": 0.59381914,
      "num_input_tokens_seen": 327723845,
      "step": 15191,
      "time_per_iteration": 3.1207058429718018
    },
    {
      "auxiliary_loss_clip": 0.01106582,
      "auxiliary_loss_mlp": 0.01036042,
      "balance_loss_clip": 1.03582692,
      "balance_loss_mlp": 1.023525,
      "epoch": 0.9133924545317902,
      "flos": 25483792170240.0,
      "grad_norm": 1.7631765485024358,
      "language_loss": 0.74288034,
      "learning_rate": 7.81172308613034e-08,
      "loss": 0.76430655,
      "num_input_tokens_seen": 327742590,
      "step": 15192,
      "time_per_iteration": 2.5117266178131104
    },
    {
      "auxiliary_loss_clip": 0.01091218,
      "auxiliary_loss_mlp": 0.01029637,
      "balance_loss_clip": 1.03589416,
      "balance_loss_mlp": 1.01790714,
      "epoch": 0.9134525777844581,
      "flos": 39930690107520.0,
      "grad_norm": 1.892299827009024,
      "language_loss": 0.69448847,
      "learning_rate": 7.800948301161647e-08,
      "loss": 0.71569699,
      "num_input_tokens_seen": 327764350,
      "step": 15193,
      "time_per_iteration": 2.64632511138916
    },
    {
      "auxiliary_loss_clip": 0.01089118,
      "auxiliary_loss_mlp": 0.01037445,
      "balance_loss_clip": 1.0340941,
      "balance_loss_mlp": 1.02640605,
      "epoch": 0.9135127010371261,
      "flos": 20886723797760.0,
      "grad_norm": 1.7980069471936975,
      "language_loss": 0.72914159,
      "learning_rate": 7.790180804400215e-08,
      "loss": 0.75040728,
      "num_input_tokens_seen": 327783120,
      "step": 15194,
      "time_per_iteration": 2.5153708457946777
    },
    {
      "auxiliary_loss_clip": 0.01058349,
      "auxiliary_loss_mlp": 0.01038246,
      "balance_loss_clip": 1.03189826,
      "balance_loss_mlp": 1.02306485,
      "epoch": 0.913572824289794,
      "flos": 20813250528000.0,
      "grad_norm": 1.887059621456418,
      "language_loss": 0.61571169,
      "learning_rate": 7.779420596254383e-08,
      "loss": 0.63667762,
      "num_input_tokens_seen": 327801960,
      "step": 15195,
      "time_per_iteration": 2.583388328552246
    },
    {
      "auxiliary_loss_clip": 0.01093711,
      "auxiliary_loss_mlp": 0.01031083,
      "balance_loss_clip": 1.03417325,
      "balance_loss_mlp": 1.01936495,
      "epoch": 0.913632947542462,
      "flos": 25703278225920.0,
      "grad_norm": 1.7255560454772492,
      "language_loss": 0.71543157,
      "learning_rate": 7.768667677132201e-08,
      "loss": 0.73667955,
      "num_input_tokens_seen": 327823795,
      "step": 15196,
      "time_per_iteration": 2.54146409034729
    },
    {
      "auxiliary_loss_clip": 0.01081704,
      "auxiliary_loss_mlp": 0.01029863,
      "balance_loss_clip": 1.03430557,
      "balance_loss_mlp": 1.01862705,
      "epoch": 0.9136930707951301,
      "flos": 26286216048000.0,
      "grad_norm": 1.593450803067473,
      "language_loss": 0.71213359,
      "learning_rate": 7.757922047441411e-08,
      "loss": 0.73324925,
      "num_input_tokens_seen": 327845175,
      "step": 15197,
      "time_per_iteration": 2.5599172115325928
    },
    {
      "auxiliary_loss_clip": 0.01083912,
      "auxiliary_loss_mlp": 0.010283,
      "balance_loss_clip": 1.03256869,
      "balance_loss_mlp": 1.01581311,
      "epoch": 0.913753194047798,
      "flos": 22091885942400.0,
      "grad_norm": 1.6953227771696529,
      "language_loss": 0.77668774,
      "learning_rate": 7.747183707589489e-08,
      "loss": 0.7978099,
      "num_input_tokens_seen": 327863150,
      "step": 15198,
      "time_per_iteration": 2.508389711380005
    },
    {
      "auxiliary_loss_clip": 0.01085314,
      "auxiliary_loss_mlp": 0.010338,
      "balance_loss_clip": 1.03377008,
      "balance_loss_mlp": 1.0219748,
      "epoch": 0.913813317300466,
      "flos": 23587206151680.0,
      "grad_norm": 1.3716105891603294,
      "language_loss": 0.67839092,
      "learning_rate": 7.736452657983616e-08,
      "loss": 0.6995821,
      "num_input_tokens_seen": 327883445,
      "step": 15199,
      "time_per_iteration": 2.5115697383880615
    },
    {
      "auxiliary_loss_clip": 0.01093319,
      "auxiliary_loss_mlp": 0.0078168,
      "balance_loss_clip": 1.03373456,
      "balance_loss_mlp": 1.00690222,
      "epoch": 0.9138734405531339,
      "flos": 28876452583680.0,
      "grad_norm": 1.5814305302636624,
      "language_loss": 0.67475533,
      "learning_rate": 7.725728899030714e-08,
      "loss": 0.69350535,
      "num_input_tokens_seen": 327905745,
      "step": 15200,
      "time_per_iteration": 2.5348875522613525
    },
    {
      "auxiliary_loss_clip": 0.01092147,
      "auxiliary_loss_mlp": 0.01033132,
      "balance_loss_clip": 1.03650963,
      "balance_loss_mlp": 1.02237296,
      "epoch": 0.9139335638058019,
      "flos": 22821087945600.0,
      "grad_norm": 1.4927866903032523,
      "language_loss": 0.70963591,
      "learning_rate": 7.715012431137435e-08,
      "loss": 0.73088872,
      "num_input_tokens_seen": 327925435,
      "step": 15201,
      "time_per_iteration": 2.5108840465545654
    },
    {
      "auxiliary_loss_clip": 0.01090683,
      "auxiliary_loss_mlp": 0.01025238,
      "balance_loss_clip": 1.03297162,
      "balance_loss_mlp": 1.01450336,
      "epoch": 0.9139936870584698,
      "flos": 18004174381440.0,
      "grad_norm": 1.7052579123532463,
      "language_loss": 0.70887625,
      "learning_rate": 7.704303254710165e-08,
      "loss": 0.73003542,
      "num_input_tokens_seen": 327944145,
      "step": 15202,
      "time_per_iteration": 2.4544639587402344
    },
    {
      "auxiliary_loss_clip": 0.01101823,
      "auxiliary_loss_mlp": 0.01029352,
      "balance_loss_clip": 1.03381991,
      "balance_loss_mlp": 1.01709151,
      "epoch": 0.9140538103111379,
      "flos": 15813767111040.0,
      "grad_norm": 1.889604016206173,
      "language_loss": 0.66631591,
      "learning_rate": 7.693601370155001e-08,
      "loss": 0.68762767,
      "num_input_tokens_seen": 327960565,
      "step": 15203,
      "time_per_iteration": 2.427597999572754
    },
    {
      "auxiliary_loss_clip": 0.01093887,
      "auxiliary_loss_mlp": 0.01030113,
      "balance_loss_clip": 1.03534365,
      "balance_loss_mlp": 1.01727974,
      "epoch": 0.9141139335638058,
      "flos": 23987035416960.0,
      "grad_norm": 2.602458322707884,
      "language_loss": 0.68763161,
      "learning_rate": 7.682906777877751e-08,
      "loss": 0.7088716,
      "num_input_tokens_seen": 327981180,
      "step": 15204,
      "time_per_iteration": 2.494999647140503
    },
    {
      "auxiliary_loss_clip": 0.01092242,
      "auxiliary_loss_mlp": 0.01025854,
      "balance_loss_clip": 1.03198791,
      "balance_loss_mlp": 1.01334918,
      "epoch": 0.9141740568164738,
      "flos": 24024418496640.0,
      "grad_norm": 1.9676829716553483,
      "language_loss": 0.59464562,
      "learning_rate": 7.672219478283915e-08,
      "loss": 0.61582661,
      "num_input_tokens_seen": 328001500,
      "step": 15205,
      "time_per_iteration": 2.4988577365875244
    },
    {
      "auxiliary_loss_clip": 0.01070037,
      "auxiliary_loss_mlp": 0.01032702,
      "balance_loss_clip": 1.03439879,
      "balance_loss_mlp": 1.02064431,
      "epoch": 0.9142341800691417,
      "flos": 27018291139200.0,
      "grad_norm": 1.6490646816237196,
      "language_loss": 0.81209642,
      "learning_rate": 7.661539471778811e-08,
      "loss": 0.8331238,
      "num_input_tokens_seen": 328023025,
      "step": 15206,
      "time_per_iteration": 2.5976977348327637
    },
    {
      "auxiliary_loss_clip": 0.01058266,
      "auxiliary_loss_mlp": 0.01025342,
      "balance_loss_clip": 1.03140426,
      "balance_loss_mlp": 1.01290214,
      "epoch": 0.9142943033218097,
      "flos": 20412487509120.0,
      "grad_norm": 2.607041190318014,
      "language_loss": 0.74203533,
      "learning_rate": 7.650866758767382e-08,
      "loss": 0.76287138,
      "num_input_tokens_seen": 328041410,
      "step": 15207,
      "time_per_iteration": 2.569094181060791
    },
    {
      "auxiliary_loss_clip": 0.01059657,
      "auxiliary_loss_mlp": 0.01035091,
      "balance_loss_clip": 1.03666973,
      "balance_loss_mlp": 1.02262783,
      "epoch": 0.9143544265744776,
      "flos": 19755322231680.0,
      "grad_norm": 1.5981545898539251,
      "language_loss": 0.72683394,
      "learning_rate": 7.640201339654373e-08,
      "loss": 0.7477814,
      "num_input_tokens_seen": 328060495,
      "step": 15208,
      "time_per_iteration": 2.5695676803588867
    },
    {
      "auxiliary_loss_clip": 0.01092577,
      "auxiliary_loss_mlp": 0.01029295,
      "balance_loss_clip": 1.03584576,
      "balance_loss_mlp": 1.01792264,
      "epoch": 0.9144145498271457,
      "flos": 17165444832000.0,
      "grad_norm": 2.469356857014516,
      "language_loss": 0.86421204,
      "learning_rate": 7.629543214844237e-08,
      "loss": 0.88543081,
      "num_input_tokens_seen": 328076905,
      "step": 15209,
      "time_per_iteration": 2.457808494567871
    },
    {
      "auxiliary_loss_clip": 0.01083327,
      "auxiliary_loss_mlp": 0.01033263,
      "balance_loss_clip": 1.0354315,
      "balance_loss_mlp": 1.02206969,
      "epoch": 0.9144746730798137,
      "flos": 23726072131200.0,
      "grad_norm": 5.850777385733918,
      "language_loss": 0.74835336,
      "learning_rate": 7.618892384741093e-08,
      "loss": 0.76951927,
      "num_input_tokens_seen": 328096960,
      "step": 15210,
      "time_per_iteration": 2.556257963180542
    },
    {
      "auxiliary_loss_clip": 0.01078972,
      "auxiliary_loss_mlp": 0.01032984,
      "balance_loss_clip": 1.03025222,
      "balance_loss_mlp": 1.02060962,
      "epoch": 0.9145347963324816,
      "flos": 25847854467840.0,
      "grad_norm": 1.8703677993087247,
      "language_loss": 0.78117633,
      "learning_rate": 7.6082488497488e-08,
      "loss": 0.80229592,
      "num_input_tokens_seen": 328115445,
      "step": 15211,
      "time_per_iteration": 2.550583600997925
    },
    {
      "auxiliary_loss_clip": 0.01094732,
      "auxiliary_loss_mlp": 0.010274,
      "balance_loss_clip": 1.03520751,
      "balance_loss_mlp": 1.0158366,
      "epoch": 0.9145949195851496,
      "flos": 19242769109760.0,
      "grad_norm": 1.6127854192556228,
      "language_loss": 0.82749951,
      "learning_rate": 7.597612610270986e-08,
      "loss": 0.84872091,
      "num_input_tokens_seen": 328133965,
      "step": 15212,
      "time_per_iteration": 2.470095634460449
    },
    {
      "auxiliary_loss_clip": 0.01089738,
      "auxiliary_loss_mlp": 0.01025991,
      "balance_loss_clip": 1.03419268,
      "balance_loss_mlp": 1.01501203,
      "epoch": 0.9146550428378175,
      "flos": 18296379521280.0,
      "grad_norm": 1.6964770833881595,
      "language_loss": 0.84057951,
      "learning_rate": 7.586983666711022e-08,
      "loss": 0.86173677,
      "num_input_tokens_seen": 328151520,
      "step": 15213,
      "time_per_iteration": 2.576432943344116
    },
    {
      "auxiliary_loss_clip": 0.0109535,
      "auxiliary_loss_mlp": 0.01027681,
      "balance_loss_clip": 1.0368619,
      "balance_loss_mlp": 1.0162127,
      "epoch": 0.9147151660904855,
      "flos": 20084264006400.0,
      "grad_norm": 1.6403408187422437,
      "language_loss": 0.71157622,
      "learning_rate": 7.576362019471894e-08,
      "loss": 0.73280644,
      "num_input_tokens_seen": 328171275,
      "step": 15214,
      "time_per_iteration": 2.4651708602905273
    },
    {
      "auxiliary_loss_clip": 0.01096369,
      "auxiliary_loss_mlp": 0.01038037,
      "balance_loss_clip": 1.03661227,
      "balance_loss_mlp": 1.02515626,
      "epoch": 0.9147752893431534,
      "flos": 24389127239040.0,
      "grad_norm": 1.4786344595923366,
      "language_loss": 0.62475967,
      "learning_rate": 7.565747668956413e-08,
      "loss": 0.64610374,
      "num_input_tokens_seen": 328192115,
      "step": 15215,
      "time_per_iteration": 2.5024194717407227
    },
    {
      "auxiliary_loss_clip": 0.01075644,
      "auxiliary_loss_mlp": 0.01029425,
      "balance_loss_clip": 1.03626633,
      "balance_loss_mlp": 1.01673567,
      "epoch": 0.9148354125958215,
      "flos": 18150402648960.0,
      "grad_norm": 2.3673246554120637,
      "language_loss": 0.76168454,
      "learning_rate": 7.555140615567058e-08,
      "loss": 0.78273523,
      "num_input_tokens_seen": 328208990,
      "step": 15216,
      "time_per_iteration": 2.5032505989074707
    },
    {
      "auxiliary_loss_clip": 0.01079846,
      "auxiliary_loss_mlp": 0.01040021,
      "balance_loss_clip": 1.03550208,
      "balance_loss_mlp": 1.02578175,
      "epoch": 0.9148955358484894,
      "flos": 23367540528000.0,
      "grad_norm": 2.227401699096528,
      "language_loss": 0.67912281,
      "learning_rate": 7.544540859706062e-08,
      "loss": 0.7003215,
      "num_input_tokens_seen": 328227840,
      "step": 15217,
      "time_per_iteration": 3.9342377185821533
    },
    {
      "auxiliary_loss_clip": 0.01090905,
      "auxiliary_loss_mlp": 0.01031997,
      "balance_loss_clip": 1.03505516,
      "balance_loss_mlp": 1.02046323,
      "epoch": 0.9149556591011574,
      "flos": 18076498416000.0,
      "grad_norm": 1.8868874354915137,
      "language_loss": 0.79695582,
      "learning_rate": 7.533948401775347e-08,
      "loss": 0.81818485,
      "num_input_tokens_seen": 328246250,
      "step": 15218,
      "time_per_iteration": 2.4437167644500732
    },
    {
      "auxiliary_loss_clip": 0.00999198,
      "auxiliary_loss_mlp": 0.00998888,
      "balance_loss_clip": 1.01008701,
      "balance_loss_mlp": 0.99784464,
      "epoch": 0.9150157823538253,
      "flos": 54586374825600.0,
      "grad_norm": 0.8501912802934998,
      "language_loss": 0.59253043,
      "learning_rate": 7.523363242176595e-08,
      "loss": 0.61251128,
      "num_input_tokens_seen": 328303625,
      "step": 15219,
      "time_per_iteration": 4.493297100067139
    },
    {
      "auxiliary_loss_clip": 0.01090547,
      "auxiliary_loss_mlp": 0.01033035,
      "balance_loss_clip": 1.03388655,
      "balance_loss_mlp": 1.02098846,
      "epoch": 0.9150759056064933,
      "flos": 17893102550400.0,
      "grad_norm": 2.0470272546149606,
      "language_loss": 0.78511107,
      "learning_rate": 7.512785381311216e-08,
      "loss": 0.80634683,
      "num_input_tokens_seen": 328322135,
      "step": 15220,
      "time_per_iteration": 4.135571241378784
    },
    {
      "auxiliary_loss_clip": 0.01052013,
      "auxiliary_loss_mlp": 0.01038584,
      "balance_loss_clip": 1.03169465,
      "balance_loss_mlp": 1.02433264,
      "epoch": 0.9151360288591612,
      "flos": 18073517587200.0,
      "grad_norm": 1.835562285105488,
      "language_loss": 0.65225261,
      "learning_rate": 7.50221481958031e-08,
      "loss": 0.67315859,
      "num_input_tokens_seen": 328340750,
      "step": 15221,
      "time_per_iteration": 2.586873769760132
    },
    {
      "auxiliary_loss_clip": 0.01081535,
      "auxiliary_loss_mlp": 0.01027328,
      "balance_loss_clip": 1.03426218,
      "balance_loss_mlp": 1.01621211,
      "epoch": 0.9151961521118293,
      "flos": 19354523299200.0,
      "grad_norm": 1.696118308215638,
      "language_loss": 0.84125322,
      "learning_rate": 7.491651557384692e-08,
      "loss": 0.86234188,
      "num_input_tokens_seen": 328359995,
      "step": 15222,
      "time_per_iteration": 2.4941346645355225
    },
    {
      "auxiliary_loss_clip": 0.01016509,
      "auxiliary_loss_mlp": 0.00999777,
      "balance_loss_clip": 1.01327157,
      "balance_loss_mlp": 0.99866802,
      "epoch": 0.9152562753644973,
      "flos": 72146621018880.0,
      "grad_norm": 0.7209813156122196,
      "language_loss": 0.49587774,
      "learning_rate": 7.481095595124953e-08,
      "loss": 0.51604062,
      "num_input_tokens_seen": 328426865,
      "step": 15223,
      "time_per_iteration": 3.17991042137146
    },
    {
      "auxiliary_loss_clip": 0.01072559,
      "auxiliary_loss_mlp": 0.0103777,
      "balance_loss_clip": 1.03468978,
      "balance_loss_mlp": 1.02547991,
      "epoch": 0.9153163986171652,
      "flos": 20777016683520.0,
      "grad_norm": 2.064037999580253,
      "language_loss": 0.72484177,
      "learning_rate": 7.470546933201349e-08,
      "loss": 0.7459451,
      "num_input_tokens_seen": 328445970,
      "step": 15224,
      "time_per_iteration": 2.549142837524414
    },
    {
      "auxiliary_loss_clip": 0.01087815,
      "auxiliary_loss_mlp": 0.01026843,
      "balance_loss_clip": 1.03498626,
      "balance_loss_mlp": 1.01522064,
      "epoch": 0.9153765218698332,
      "flos": 23040107124480.0,
      "grad_norm": 1.7811175316814336,
      "language_loss": 0.80972528,
      "learning_rate": 7.460005572013895e-08,
      "loss": 0.83087182,
      "num_input_tokens_seen": 328464585,
      "step": 15225,
      "time_per_iteration": 2.5327799320220947
    },
    {
      "auxiliary_loss_clip": 0.01101844,
      "auxiliary_loss_mlp": 0.01025215,
      "balance_loss_clip": 1.03353608,
      "balance_loss_mlp": 1.01385391,
      "epoch": 0.9154366451225011,
      "flos": 28990900293120.0,
      "grad_norm": 1.4265109308930939,
      "language_loss": 0.71363032,
      "learning_rate": 7.44947151196238e-08,
      "loss": 0.73490089,
      "num_input_tokens_seen": 328490155,
      "step": 15226,
      "time_per_iteration": 2.571580648422241
    },
    {
      "auxiliary_loss_clip": 0.01041381,
      "auxiliary_loss_mlp": 0.01032667,
      "balance_loss_clip": 1.03219151,
      "balance_loss_mlp": 1.01999545,
      "epoch": 0.9154967683751691,
      "flos": 22309504490880.0,
      "grad_norm": 1.9719859550191061,
      "language_loss": 0.74764729,
      "learning_rate": 7.43894475344613e-08,
      "loss": 0.76838779,
      "num_input_tokens_seen": 328508275,
      "step": 15227,
      "time_per_iteration": 4.0892333984375
    },
    {
      "auxiliary_loss_clip": 0.01080331,
      "auxiliary_loss_mlp": 0.01032584,
      "balance_loss_clip": 1.03388619,
      "balance_loss_mlp": 1.02097845,
      "epoch": 0.915556891627837,
      "flos": 24571481610240.0,
      "grad_norm": 1.4051392683342,
      "language_loss": 0.73711002,
      "learning_rate": 7.428425296864404e-08,
      "loss": 0.75823915,
      "num_input_tokens_seen": 328529425,
      "step": 15228,
      "time_per_iteration": 2.5695133209228516
    },
    {
      "auxiliary_loss_clip": 0.01066173,
      "auxiliary_loss_mlp": 0.0103,
      "balance_loss_clip": 1.0337007,
      "balance_loss_mlp": 1.01856732,
      "epoch": 0.9156170148805051,
      "flos": 22164676853760.0,
      "grad_norm": 1.5777902556819563,
      "language_loss": 0.71672487,
      "learning_rate": 7.417913142616106e-08,
      "loss": 0.73768663,
      "num_input_tokens_seen": 328550200,
      "step": 15229,
      "time_per_iteration": 2.551664113998413
    },
    {
      "auxiliary_loss_clip": 0.0110655,
      "auxiliary_loss_mlp": 0.01033515,
      "balance_loss_clip": 1.03718281,
      "balance_loss_mlp": 1.02044368,
      "epoch": 0.915677138133173,
      "flos": 20920659171840.0,
      "grad_norm": 1.5545659667759513,
      "language_loss": 0.82961726,
      "learning_rate": 7.407408291099848e-08,
      "loss": 0.85101783,
      "num_input_tokens_seen": 328568540,
      "step": 15230,
      "time_per_iteration": 2.468846082687378
    },
    {
      "auxiliary_loss_clip": 0.01060894,
      "auxiliary_loss_mlp": 0.010268,
      "balance_loss_clip": 1.03248703,
      "balance_loss_mlp": 1.01560664,
      "epoch": 0.915737261385841,
      "flos": 24345136056960.0,
      "grad_norm": 1.6928724830962776,
      "language_loss": 0.83439898,
      "learning_rate": 7.396910742713957e-08,
      "loss": 0.85527593,
      "num_input_tokens_seen": 328587300,
      "step": 15231,
      "time_per_iteration": 2.6174890995025635
    },
    {
      "auxiliary_loss_clip": 0.01087135,
      "auxiliary_loss_mlp": 0.01026482,
      "balance_loss_clip": 1.03102124,
      "balance_loss_mlp": 1.01504397,
      "epoch": 0.9157973846385089,
      "flos": 26761386090240.0,
      "grad_norm": 1.4820462532289163,
      "language_loss": 0.7237764,
      "learning_rate": 7.386420497856516e-08,
      "loss": 0.74491256,
      "num_input_tokens_seen": 328610055,
      "step": 15232,
      "time_per_iteration": 2.549255132675171
    },
    {
      "auxiliary_loss_clip": 0.01104925,
      "auxiliary_loss_mlp": 0.01032471,
      "balance_loss_clip": 1.0342319,
      "balance_loss_mlp": 1.02050257,
      "epoch": 0.9158575078911769,
      "flos": 18478733892480.0,
      "grad_norm": 2.038306779246488,
      "language_loss": 0.6792531,
      "learning_rate": 7.375937556925338e-08,
      "loss": 0.70062709,
      "num_input_tokens_seen": 328626815,
      "step": 15233,
      "time_per_iteration": 2.420616865158081
    },
    {
      "auxiliary_loss_clip": 0.01078956,
      "auxiliary_loss_mlp": 0.01031976,
      "balance_loss_clip": 1.03651333,
      "balance_loss_mlp": 1.01997721,
      "epoch": 0.9159176311438448,
      "flos": 21798926616960.0,
      "grad_norm": 1.837964273708223,
      "language_loss": 0.69844639,
      "learning_rate": 7.365461920317861e-08,
      "loss": 0.71955574,
      "num_input_tokens_seen": 328643995,
      "step": 15234,
      "time_per_iteration": 2.5360162258148193
    },
    {
      "auxiliary_loss_clip": 0.01081106,
      "auxiliary_loss_mlp": 0.01034205,
      "balance_loss_clip": 1.03512216,
      "balance_loss_mlp": 1.0216763,
      "epoch": 0.9159777543965129,
      "flos": 24783749032320.0,
      "grad_norm": 2.230784813926407,
      "language_loss": 0.88141835,
      "learning_rate": 7.354993588431391e-08,
      "loss": 0.90257144,
      "num_input_tokens_seen": 328659565,
      "step": 15235,
      "time_per_iteration": 2.5258424282073975
    },
    {
      "auxiliary_loss_clip": 0.01038347,
      "auxiliary_loss_mlp": 0.010364,
      "balance_loss_clip": 1.03282487,
      "balance_loss_mlp": 1.0222913,
      "epoch": 0.9160378776491809,
      "flos": 26868758820480.0,
      "grad_norm": 1.5531146068696027,
      "language_loss": 0.77151394,
      "learning_rate": 7.344532561662853e-08,
      "loss": 0.79226148,
      "num_input_tokens_seen": 328679045,
      "step": 15236,
      "time_per_iteration": 2.658754825592041
    },
    {
      "auxiliary_loss_clip": 0.00992623,
      "auxiliary_loss_mlp": 0.01001342,
      "balance_loss_clip": 1.01884961,
      "balance_loss_mlp": 1.00003099,
      "epoch": 0.9160980009018488,
      "flos": 70578222589440.0,
      "grad_norm": 0.7183720198758208,
      "language_loss": 0.62191588,
      "learning_rate": 7.334078840409019e-08,
      "loss": 0.64185554,
      "num_input_tokens_seen": 328744565,
      "step": 15237,
      "time_per_iteration": 3.17425537109375
    },
    {
      "auxiliary_loss_clip": 0.0110625,
      "auxiliary_loss_mlp": 0.00781803,
      "balance_loss_clip": 1.03590441,
      "balance_loss_mlp": 1.00623024,
      "epoch": 0.9161581241545168,
      "flos": 16289332202880.0,
      "grad_norm": 2.102584270530641,
      "language_loss": 0.75152349,
      "learning_rate": 7.323632425066151e-08,
      "loss": 0.77040398,
      "num_input_tokens_seen": 328762455,
      "step": 15238,
      "time_per_iteration": 2.4460582733154297
    },
    {
      "auxiliary_loss_clip": 0.01103812,
      "auxiliary_loss_mlp": 0.01027487,
      "balance_loss_clip": 1.03473496,
      "balance_loss_mlp": 1.01558352,
      "epoch": 0.9162182474071847,
      "flos": 18438154502400.0,
      "grad_norm": 1.876414020843958,
      "language_loss": 0.74720335,
      "learning_rate": 7.313193316030464e-08,
      "loss": 0.76851642,
      "num_input_tokens_seen": 328780320,
      "step": 15239,
      "time_per_iteration": 2.43398118019104
    },
    {
      "auxiliary_loss_clip": 0.01071352,
      "auxiliary_loss_mlp": 0.01031781,
      "balance_loss_clip": 1.03182328,
      "balance_loss_mlp": 1.01963973,
      "epoch": 0.9162783706598527,
      "flos": 19167248764800.0,
      "grad_norm": 2.006811316938341,
      "language_loss": 0.63400024,
      "learning_rate": 7.302761513697819e-08,
      "loss": 0.65503162,
      "num_input_tokens_seen": 328797570,
      "step": 15240,
      "time_per_iteration": 2.5364534854888916
    },
    {
      "auxiliary_loss_clip": 0.01081705,
      "auxiliary_loss_mlp": 0.00781585,
      "balance_loss_clip": 1.03578615,
      "balance_loss_mlp": 1.00710511,
      "epoch": 0.9163384939125206,
      "flos": 20412990299520.0,
      "grad_norm": 1.7211203690631538,
      "language_loss": 0.76286769,
      "learning_rate": 7.292337018463746e-08,
      "loss": 0.78150058,
      "num_input_tokens_seen": 328814075,
      "step": 15241,
      "time_per_iteration": 2.5032010078430176
    },
    {
      "auxiliary_loss_clip": 0.01101918,
      "auxiliary_loss_mlp": 0.01031554,
      "balance_loss_clip": 1.03742731,
      "balance_loss_mlp": 1.0176425,
      "epoch": 0.9163986171651887,
      "flos": 19645902426240.0,
      "grad_norm": 2.535748664589755,
      "language_loss": 0.67583704,
      "learning_rate": 7.281919830723549e-08,
      "loss": 0.69717169,
      "num_input_tokens_seen": 328831990,
      "step": 15242,
      "time_per_iteration": 2.4912006855010986
    },
    {
      "auxiliary_loss_clip": 0.01092916,
      "auxiliary_loss_mlp": 0.01028709,
      "balance_loss_clip": 1.0334332,
      "balance_loss_mlp": 1.01602507,
      "epoch": 0.9164587404178566,
      "flos": 12823054865280.0,
      "grad_norm": 1.6595480007643708,
      "language_loss": 0.80747145,
      "learning_rate": 7.271509950872334e-08,
      "loss": 0.82868767,
      "num_input_tokens_seen": 328849105,
      "step": 15243,
      "time_per_iteration": 2.472428798675537
    },
    {
      "auxiliary_loss_clip": 0.01080303,
      "auxiliary_loss_mlp": 0.01029333,
      "balance_loss_clip": 1.03242671,
      "balance_loss_mlp": 1.01694679,
      "epoch": 0.9165188636705246,
      "flos": 22309396750080.0,
      "grad_norm": 1.7487630874842564,
      "language_loss": 0.81763732,
      "learning_rate": 7.261107379304721e-08,
      "loss": 0.83873367,
      "num_input_tokens_seen": 328866810,
      "step": 15244,
      "time_per_iteration": 2.5188372135162354
    },
    {
      "auxiliary_loss_clip": 0.01107429,
      "auxiliary_loss_mlp": 0.01035572,
      "balance_loss_clip": 1.03513527,
      "balance_loss_mlp": 1.0230732,
      "epoch": 0.9165789869231925,
      "flos": 18223337214720.0,
      "grad_norm": 2.1961454678597603,
      "language_loss": 0.72134328,
      "learning_rate": 7.250712116415214e-08,
      "loss": 0.74277329,
      "num_input_tokens_seen": 328885325,
      "step": 15245,
      "time_per_iteration": 2.4213449954986572
    },
    {
      "auxiliary_loss_clip": 0.01081892,
      "auxiliary_loss_mlp": 0.01029524,
      "balance_loss_clip": 1.03476739,
      "balance_loss_mlp": 1.01821065,
      "epoch": 0.9166391101758605,
      "flos": 13691553811200.0,
      "grad_norm": 1.7349269856512282,
      "language_loss": 0.74802566,
      "learning_rate": 7.240324162598033e-08,
      "loss": 0.76913977,
      "num_input_tokens_seen": 328902655,
      "step": 15246,
      "time_per_iteration": 2.520372152328491
    },
    {
      "auxiliary_loss_clip": 0.01076739,
      "auxiliary_loss_mlp": 0.01029541,
      "balance_loss_clip": 1.03312612,
      "balance_loss_mlp": 1.01710129,
      "epoch": 0.9166992334285284,
      "flos": 17346793622400.0,
      "grad_norm": 1.9353821328661256,
      "language_loss": 0.7510137,
      "learning_rate": 7.229943518247106e-08,
      "loss": 0.77207649,
      "num_input_tokens_seen": 328918440,
      "step": 15247,
      "time_per_iteration": 2.4697623252868652
    },
    {
      "auxiliary_loss_clip": 0.01095697,
      "auxiliary_loss_mlp": 0.0102585,
      "balance_loss_clip": 1.03693771,
      "balance_loss_mlp": 1.01391077,
      "epoch": 0.9167593566811965,
      "flos": 23731135948800.0,
      "grad_norm": 1.7266812862451686,
      "language_loss": 0.75769687,
      "learning_rate": 7.219570183756052e-08,
      "loss": 0.77891237,
      "num_input_tokens_seen": 328938055,
      "step": 15248,
      "time_per_iteration": 2.5058579444885254
    },
    {
      "auxiliary_loss_clip": 0.01092292,
      "auxiliary_loss_mlp": 0.01034759,
      "balance_loss_clip": 1.03285241,
      "balance_loss_mlp": 1.02122283,
      "epoch": 0.9168194799338644,
      "flos": 27818201064960.0,
      "grad_norm": 2.360683796000608,
      "language_loss": 0.73069501,
      "learning_rate": 7.209204159518178e-08,
      "loss": 0.75196552,
      "num_input_tokens_seen": 328957895,
      "step": 15249,
      "time_per_iteration": 2.532177448272705
    },
    {
      "auxiliary_loss_clip": 0.01057662,
      "auxiliary_loss_mlp": 0.01028724,
      "balance_loss_clip": 1.03627968,
      "balance_loss_mlp": 1.01581991,
      "epoch": 0.9168796031865324,
      "flos": 21717552355200.0,
      "grad_norm": 1.8571366727173375,
      "language_loss": 0.75844491,
      "learning_rate": 7.198845445926616e-08,
      "loss": 0.7793088,
      "num_input_tokens_seen": 328971365,
      "step": 15250,
      "time_per_iteration": 2.5601084232330322
    },
    {
      "auxiliary_loss_clip": 0.01061044,
      "auxiliary_loss_mlp": 0.01028609,
      "balance_loss_clip": 1.03302288,
      "balance_loss_mlp": 1.0160321,
      "epoch": 0.9169397264392004,
      "flos": 23404420817280.0,
      "grad_norm": 1.667860971248614,
      "language_loss": 0.75714946,
      "learning_rate": 7.188494043374138e-08,
      "loss": 0.77804595,
      "num_input_tokens_seen": 328990830,
      "step": 15251,
      "time_per_iteration": 2.585933208465576
    },
    {
      "auxiliary_loss_clip": 0.01080239,
      "auxiliary_loss_mlp": 0.01033086,
      "balance_loss_clip": 1.03561401,
      "balance_loss_mlp": 1.01958585,
      "epoch": 0.9169998496918683,
      "flos": 23950981140480.0,
      "grad_norm": 2.074915807143545,
      "language_loss": 0.79774928,
      "learning_rate": 7.178149952253298e-08,
      "loss": 0.81888258,
      "num_input_tokens_seen": 329008345,
      "step": 15252,
      "time_per_iteration": 2.530998468399048
    },
    {
      "auxiliary_loss_clip": 0.01103746,
      "auxiliary_loss_mlp": 0.01035009,
      "balance_loss_clip": 1.03486514,
      "balance_loss_mlp": 1.02318347,
      "epoch": 0.9170599729445363,
      "flos": 18332469711360.0,
      "grad_norm": 1.613539381378801,
      "language_loss": 0.77434814,
      "learning_rate": 7.167813172956316e-08,
      "loss": 0.7957356,
      "num_input_tokens_seen": 329027440,
      "step": 15253,
      "time_per_iteration": 2.4275195598602295
    },
    {
      "auxiliary_loss_clip": 0.01094885,
      "auxiliary_loss_mlp": 0.01025343,
      "balance_loss_clip": 1.03594327,
      "balance_loss_mlp": 1.01348805,
      "epoch": 0.9171200961972042,
      "flos": 22674859678080.0,
      "grad_norm": 2.675882925128281,
      "language_loss": 0.72909105,
      "learning_rate": 7.157483705875256e-08,
      "loss": 0.75029325,
      "num_input_tokens_seen": 329046445,
      "step": 15254,
      "time_per_iteration": 2.4996325969696045
    },
    {
      "auxiliary_loss_clip": 0.01063445,
      "auxiliary_loss_mlp": 0.01024731,
      "balance_loss_clip": 1.03294301,
      "balance_loss_mlp": 1.01396692,
      "epoch": 0.9171802194498723,
      "flos": 26719298328960.0,
      "grad_norm": 1.4555948610035223,
      "language_loss": 0.78925395,
      "learning_rate": 7.14716155140167e-08,
      "loss": 0.81013572,
      "num_input_tokens_seen": 329065555,
      "step": 15255,
      "time_per_iteration": 2.5907673835754395
    },
    {
      "auxiliary_loss_clip": 0.0109415,
      "auxiliary_loss_mlp": 0.01031402,
      "balance_loss_clip": 1.03357112,
      "balance_loss_mlp": 1.01904559,
      "epoch": 0.9172403427025402,
      "flos": 37889240538240.0,
      "grad_norm": 1.8895268612431264,
      "language_loss": 0.68457997,
      "learning_rate": 7.136846709927047e-08,
      "loss": 0.70583552,
      "num_input_tokens_seen": 329087515,
      "step": 15256,
      "time_per_iteration": 4.0132622718811035
    },
    {
      "auxiliary_loss_clip": 0.01087699,
      "auxiliary_loss_mlp": 0.01037396,
      "balance_loss_clip": 1.03337526,
      "balance_loss_mlp": 1.02491438,
      "epoch": 0.9173004659552082,
      "flos": 17055163100160.0,
      "grad_norm": 1.618248391327963,
      "language_loss": 0.83784485,
      "learning_rate": 7.126539181842561e-08,
      "loss": 0.85909581,
      "num_input_tokens_seen": 329106820,
      "step": 15257,
      "time_per_iteration": 2.488647699356079
    },
    {
      "auxiliary_loss_clip": 0.01078539,
      "auxiliary_loss_mlp": 0.01028316,
      "balance_loss_clip": 1.03220773,
      "balance_loss_mlp": 1.01696765,
      "epoch": 0.9173605892078761,
      "flos": 22201593056640.0,
      "grad_norm": 1.4859371711908207,
      "language_loss": 0.77455193,
      "learning_rate": 7.116238967539012e-08,
      "loss": 0.7956205,
      "num_input_tokens_seen": 329126515,
      "step": 15258,
      "time_per_iteration": 5.2878477573394775
    },
    {
      "auxiliary_loss_clip": 0.01094916,
      "auxiliary_loss_mlp": 0.01034111,
      "balance_loss_clip": 1.03727102,
      "balance_loss_mlp": 1.02260089,
      "epoch": 0.9174207124605441,
      "flos": 16507776764160.0,
      "grad_norm": 1.874388109919324,
      "language_loss": 0.7841959,
      "learning_rate": 7.105946067406999e-08,
      "loss": 0.80548614,
      "num_input_tokens_seen": 329142660,
      "step": 15259,
      "time_per_iteration": 2.455075979232788
    },
    {
      "auxiliary_loss_clip": 0.01056861,
      "auxiliary_loss_mlp": 0.01029045,
      "balance_loss_clip": 1.03289497,
      "balance_loss_mlp": 1.01813745,
      "epoch": 0.917480835713212,
      "flos": 24535606901760.0,
      "grad_norm": 1.6036806104906023,
      "language_loss": 0.76523364,
      "learning_rate": 7.095660481836895e-08,
      "loss": 0.78609264,
      "num_input_tokens_seen": 329162575,
      "step": 15260,
      "time_per_iteration": 2.6203041076660156
    },
    {
      "auxiliary_loss_clip": 0.01061231,
      "auxiliary_loss_mlp": 0.01030527,
      "balance_loss_clip": 1.03211474,
      "balance_loss_mlp": 1.01856482,
      "epoch": 0.9175409589658801,
      "flos": 20880726226560.0,
      "grad_norm": 1.4476256824830447,
      "language_loss": 0.60843754,
      "learning_rate": 7.085382211218637e-08,
      "loss": 0.62935513,
      "num_input_tokens_seen": 329182090,
      "step": 15261,
      "time_per_iteration": 2.585662364959717
    },
    {
      "auxiliary_loss_clip": 0.01078139,
      "auxiliary_loss_mlp": 0.01027308,
      "balance_loss_clip": 1.03241932,
      "balance_loss_mlp": 1.01551187,
      "epoch": 0.917601082218548,
      "flos": 14276035918080.0,
      "grad_norm": 1.7452075721543565,
      "language_loss": 0.73532271,
      "learning_rate": 7.075111255942002e-08,
      "loss": 0.75637716,
      "num_input_tokens_seen": 329196535,
      "step": 15262,
      "time_per_iteration": 2.4800074100494385
    },
    {
      "auxiliary_loss_clip": 0.01106101,
      "auxiliary_loss_mlp": 0.01036308,
      "balance_loss_clip": 1.03349483,
      "balance_loss_mlp": 1.023857,
      "epoch": 0.917661205471216,
      "flos": 19099234362240.0,
      "grad_norm": 1.7188727249258504,
      "language_loss": 0.77662951,
      "learning_rate": 7.064847616396496e-08,
      "loss": 0.79805362,
      "num_input_tokens_seen": 329215135,
      "step": 15263,
      "time_per_iteration": 2.410594940185547
    },
    {
      "auxiliary_loss_clip": 0.01106357,
      "auxiliary_loss_mlp": 0.01032389,
      "balance_loss_clip": 1.03468633,
      "balance_loss_mlp": 1.02039027,
      "epoch": 0.917721328723884,
      "flos": 21106568989440.0,
      "grad_norm": 5.333294595714343,
      "language_loss": 0.75585568,
      "learning_rate": 7.054591292971324e-08,
      "loss": 0.77724314,
      "num_input_tokens_seen": 329235150,
      "step": 15264,
      "time_per_iteration": 2.4581072330474854
    },
    {
      "auxiliary_loss_clip": 0.01081743,
      "auxiliary_loss_mlp": 0.01034803,
      "balance_loss_clip": 1.03501415,
      "balance_loss_mlp": 1.02314472,
      "epoch": 0.9177814519765519,
      "flos": 21943215550080.0,
      "grad_norm": 1.6217170953731737,
      "language_loss": 0.831788,
      "learning_rate": 7.044342286055394e-08,
      "loss": 0.85295343,
      "num_input_tokens_seen": 329254365,
      "step": 15265,
      "time_per_iteration": 2.501666307449341
    },
    {
      "auxiliary_loss_clip": 0.01108894,
      "auxiliary_loss_mlp": 0.01040215,
      "balance_loss_clip": 1.03657985,
      "balance_loss_mlp": 1.02735829,
      "epoch": 0.9178415752292199,
      "flos": 24205982768640.0,
      "grad_norm": 1.4823346822896535,
      "language_loss": 0.73310244,
      "learning_rate": 7.034100596037306e-08,
      "loss": 0.75459349,
      "num_input_tokens_seen": 329274385,
      "step": 15266,
      "time_per_iteration": 3.8682918548583984
    },
    {
      "auxiliary_loss_clip": 0.01103106,
      "auxiliary_loss_mlp": 0.01029655,
      "balance_loss_clip": 1.0338676,
      "balance_loss_mlp": 1.01804376,
      "epoch": 0.9179016984818879,
      "flos": 20042068504320.0,
      "grad_norm": 2.0731302175442354,
      "language_loss": 0.77554882,
      "learning_rate": 7.023866223305486e-08,
      "loss": 0.79687643,
      "num_input_tokens_seen": 329292160,
      "step": 15267,
      "time_per_iteration": 2.4430463314056396
    },
    {
      "auxiliary_loss_clip": 0.01017485,
      "auxiliary_loss_mlp": 0.00759778,
      "balance_loss_clip": 1.00811625,
      "balance_loss_mlp": 0.99648422,
      "epoch": 0.9179618217345559,
      "flos": 65555901100800.0,
      "grad_norm": 0.7450347574364345,
      "language_loss": 0.56284928,
      "learning_rate": 7.013639168247975e-08,
      "loss": 0.58062196,
      "num_input_tokens_seen": 329351870,
      "step": 15268,
      "time_per_iteration": 3.1606030464172363
    },
    {
      "auxiliary_loss_clip": 0.01105669,
      "auxiliary_loss_mlp": 0.00782837,
      "balance_loss_clip": 1.03518808,
      "balance_loss_mlp": 1.00790453,
      "epoch": 0.9180219449872238,
      "flos": 21324618501120.0,
      "grad_norm": 2.617577116794001,
      "language_loss": 0.76603502,
      "learning_rate": 7.0034194312526e-08,
      "loss": 0.78492004,
      "num_input_tokens_seen": 329370930,
      "step": 15269,
      "time_per_iteration": 2.4621238708496094
    },
    {
      "auxiliary_loss_clip": 0.01070762,
      "auxiliary_loss_mlp": 0.01031073,
      "balance_loss_clip": 1.0328275,
      "balance_loss_mlp": 1.01891375,
      "epoch": 0.9180820682398918,
      "flos": 41060008684800.0,
      "grad_norm": 2.5322800980161837,
      "language_loss": 0.72966707,
      "learning_rate": 6.993207012706936e-08,
      "loss": 0.75068533,
      "num_input_tokens_seen": 329391275,
      "step": 15270,
      "time_per_iteration": 2.692742347717285
    },
    {
      "auxiliary_loss_clip": 0.0110072,
      "auxiliary_loss_mlp": 0.01032132,
      "balance_loss_clip": 1.03300989,
      "balance_loss_mlp": 1.01995492,
      "epoch": 0.9181421914925597,
      "flos": 28072915384320.0,
      "grad_norm": 1.8348591954033457,
      "language_loss": 0.79700661,
      "learning_rate": 6.98300191299821e-08,
      "loss": 0.81833512,
      "num_input_tokens_seen": 329412775,
      "step": 15271,
      "time_per_iteration": 2.4951019287109375
    },
    {
      "auxiliary_loss_clip": 0.01060737,
      "auxiliary_loss_mlp": 0.01032484,
      "balance_loss_clip": 1.03045571,
      "balance_loss_mlp": 1.01982427,
      "epoch": 0.9182023147452277,
      "flos": 29169411909120.0,
      "grad_norm": 1.812661594833814,
      "language_loss": 0.7316317,
      "learning_rate": 6.972804132513355e-08,
      "loss": 0.75256389,
      "num_input_tokens_seen": 329432440,
      "step": 15272,
      "time_per_iteration": 2.586599588394165
    },
    {
      "auxiliary_loss_clip": 0.01081937,
      "auxiliary_loss_mlp": 0.01033229,
      "balance_loss_clip": 1.03404188,
      "balance_loss_mlp": 1.02158284,
      "epoch": 0.9182624379978956,
      "flos": 24060831909120.0,
      "grad_norm": 1.911081517326716,
      "language_loss": 0.73003781,
      "learning_rate": 6.962613671639105e-08,
      "loss": 0.75118947,
      "num_input_tokens_seen": 329450605,
      "step": 15273,
      "time_per_iteration": 2.5415749549865723
    },
    {
      "auxiliary_loss_clip": 0.01063352,
      "auxiliary_loss_mlp": 0.01026906,
      "balance_loss_clip": 1.0325489,
      "balance_loss_mlp": 1.01617682,
      "epoch": 0.9183225612505637,
      "flos": 23293528554240.0,
      "grad_norm": 1.444559617236309,
      "language_loss": 0.74577284,
      "learning_rate": 6.952430530761933e-08,
      "loss": 0.76667541,
      "num_input_tokens_seen": 329470550,
      "step": 15274,
      "time_per_iteration": 2.544069766998291
    },
    {
      "auxiliary_loss_clip": 0.01090916,
      "auxiliary_loss_mlp": 0.01037069,
      "balance_loss_clip": 1.03264952,
      "balance_loss_mlp": 1.02564299,
      "epoch": 0.9183826845032316,
      "flos": 19609237618560.0,
      "grad_norm": 1.580122238622461,
      "language_loss": 0.68746543,
      "learning_rate": 6.942254710267902e-08,
      "loss": 0.70874536,
      "num_input_tokens_seen": 329489765,
      "step": 15275,
      "time_per_iteration": 2.4759600162506104
    },
    {
      "auxiliary_loss_clip": 0.01088427,
      "auxiliary_loss_mlp": 0.01029992,
      "balance_loss_clip": 1.0332737,
      "balance_loss_mlp": 1.01859534,
      "epoch": 0.9184428077558996,
      "flos": 18479057114880.0,
      "grad_norm": 2.02688879942051,
      "language_loss": 0.72539729,
      "learning_rate": 6.932086210542953e-08,
      "loss": 0.74658144,
      "num_input_tokens_seen": 329507040,
      "step": 15276,
      "time_per_iteration": 2.450692653656006
    },
    {
      "auxiliary_loss_clip": 0.01082549,
      "auxiliary_loss_mlp": 0.01028753,
      "balance_loss_clip": 1.0352788,
      "balance_loss_mlp": 1.01736856,
      "epoch": 0.9185029310085676,
      "flos": 20741034234240.0,
      "grad_norm": 1.5981875232090554,
      "language_loss": 0.73266518,
      "learning_rate": 6.921925031972642e-08,
      "loss": 0.75377822,
      "num_input_tokens_seen": 329525540,
      "step": 15277,
      "time_per_iteration": 2.522148847579956
    },
    {
      "auxiliary_loss_clip": 0.01000794,
      "auxiliary_loss_mlp": 0.01003898,
      "balance_loss_clip": 1.00816453,
      "balance_loss_mlp": 1.00279546,
      "epoch": 0.9185630542612355,
      "flos": 68209231875840.0,
      "grad_norm": 0.7225453953315476,
      "language_loss": 0.59284592,
      "learning_rate": 6.91177117494226e-08,
      "loss": 0.61289287,
      "num_input_tokens_seen": 329592905,
      "step": 15278,
      "time_per_iteration": 3.2287709712982178
    },
    {
      "auxiliary_loss_clip": 0.01064822,
      "auxiliary_loss_mlp": 0.01028425,
      "balance_loss_clip": 1.03040421,
      "balance_loss_mlp": 1.01746333,
      "epoch": 0.9186231775139035,
      "flos": 12239470598400.0,
      "grad_norm": 1.6314586732280334,
      "language_loss": 0.64514351,
      "learning_rate": 6.901624639836879e-08,
      "loss": 0.666076,
      "num_input_tokens_seen": 329610150,
      "step": 15279,
      "time_per_iteration": 2.525012254714966
    },
    {
      "auxiliary_loss_clip": 0.01027525,
      "auxiliary_loss_mlp": 0.00762252,
      "balance_loss_clip": 1.00484347,
      "balance_loss_mlp": 1.00263727,
      "epoch": 0.9186833007665715,
      "flos": 63939237770880.0,
      "grad_norm": 0.8683992644926448,
      "language_loss": 0.60176259,
      "learning_rate": 6.891485427041211e-08,
      "loss": 0.61966044,
      "num_input_tokens_seen": 329673650,
      "step": 15280,
      "time_per_iteration": 3.053539514541626
    },
    {
      "auxiliary_loss_clip": 0.01083686,
      "auxiliary_loss_mlp": 0.01034756,
      "balance_loss_clip": 1.0342871,
      "balance_loss_mlp": 1.02272749,
      "epoch": 0.9187434240192395,
      "flos": 19974700546560.0,
      "grad_norm": 1.6647436715860733,
      "language_loss": 0.69244909,
      "learning_rate": 6.881353536939815e-08,
      "loss": 0.71363348,
      "num_input_tokens_seen": 329692520,
      "step": 15281,
      "time_per_iteration": 2.5225236415863037
    },
    {
      "auxiliary_loss_clip": 0.01081803,
      "auxiliary_loss_mlp": 0.01028198,
      "balance_loss_clip": 1.03404593,
      "balance_loss_mlp": 1.01541829,
      "epoch": 0.9188035472719074,
      "flos": 25227820874880.0,
      "grad_norm": 1.6576770631886806,
      "language_loss": 0.84463882,
      "learning_rate": 6.871228969916831e-08,
      "loss": 0.86573881,
      "num_input_tokens_seen": 329713750,
      "step": 15282,
      "time_per_iteration": 2.539827585220337
    },
    {
      "auxiliary_loss_clip": 0.01078946,
      "auxiliary_loss_mlp": 0.0103404,
      "balance_loss_clip": 1.0333972,
      "balance_loss_mlp": 1.02146316,
      "epoch": 0.9188636705245754,
      "flos": 18405547931520.0,
      "grad_norm": 1.7660970167337773,
      "language_loss": 0.60050285,
      "learning_rate": 6.861111726356194e-08,
      "loss": 0.6216327,
      "num_input_tokens_seen": 329730960,
      "step": 15283,
      "time_per_iteration": 2.5012104511260986
    },
    {
      "auxiliary_loss_clip": 0.01096872,
      "auxiliary_loss_mlp": 0.00782588,
      "balance_loss_clip": 1.03554714,
      "balance_loss_mlp": 1.00654852,
      "epoch": 0.9189237937772433,
      "flos": 23769129559680.0,
      "grad_norm": 1.4879887064921404,
      "language_loss": 0.65303743,
      "learning_rate": 6.851001806641554e-08,
      "loss": 0.67183197,
      "num_input_tokens_seen": 329750975,
      "step": 15284,
      "time_per_iteration": 2.491382598876953
    },
    {
      "auxiliary_loss_clip": 0.01101197,
      "auxiliary_loss_mlp": 0.01031608,
      "balance_loss_clip": 1.03340101,
      "balance_loss_mlp": 1.01938331,
      "epoch": 0.9189839170299113,
      "flos": 21214624078080.0,
      "grad_norm": 1.954042861660427,
      "language_loss": 0.73543853,
      "learning_rate": 6.840899211156292e-08,
      "loss": 0.75676662,
      "num_input_tokens_seen": 329769645,
      "step": 15285,
      "time_per_iteration": 2.4662909507751465
    },
    {
      "auxiliary_loss_clip": 0.01102222,
      "auxiliary_loss_mlp": 0.01034848,
      "balance_loss_clip": 1.03393245,
      "balance_loss_mlp": 1.02272439,
      "epoch": 0.9190440402825792,
      "flos": 16727370560640.0,
      "grad_norm": 1.827394442029598,
      "language_loss": 0.71625847,
      "learning_rate": 6.830803940283458e-08,
      "loss": 0.73762918,
      "num_input_tokens_seen": 329788185,
      "step": 15286,
      "time_per_iteration": 2.4104690551757812
    },
    {
      "auxiliary_loss_clip": 0.01104576,
      "auxiliary_loss_mlp": 0.01028303,
      "balance_loss_clip": 1.03528214,
      "balance_loss_mlp": 1.01622748,
      "epoch": 0.9191041635352473,
      "flos": 23441193365760.0,
      "grad_norm": 1.8707567241793748,
      "language_loss": 0.73202288,
      "learning_rate": 6.820715994405945e-08,
      "loss": 0.75335169,
      "num_input_tokens_seen": 329806780,
      "step": 15287,
      "time_per_iteration": 2.4733715057373047
    },
    {
      "auxiliary_loss_clip": 0.01105902,
      "auxiliary_loss_mlp": 0.01029315,
      "balance_loss_clip": 1.03685379,
      "balance_loss_mlp": 1.01633954,
      "epoch": 0.9191642867879152,
      "flos": 18807532012800.0,
      "grad_norm": 2.1696161093244113,
      "language_loss": 0.65475225,
      "learning_rate": 6.810635373906226e-08,
      "loss": 0.67610443,
      "num_input_tokens_seen": 329826350,
      "step": 15288,
      "time_per_iteration": 2.4431943893432617
    },
    {
      "auxiliary_loss_clip": 0.01108262,
      "auxiliary_loss_mlp": 0.01036185,
      "balance_loss_clip": 1.03936529,
      "balance_loss_mlp": 1.02373981,
      "epoch": 0.9192244100405832,
      "flos": 32160950167680.0,
      "grad_norm": 2.0263648803717156,
      "language_loss": 0.71392167,
      "learning_rate": 6.800562079166549e-08,
      "loss": 0.73536611,
      "num_input_tokens_seen": 329846160,
      "step": 15289,
      "time_per_iteration": 2.538158893585205
    },
    {
      "auxiliary_loss_clip": 0.01065242,
      "auxiliary_loss_mlp": 0.0103753,
      "balance_loss_clip": 1.03369594,
      "balance_loss_mlp": 1.02440524,
      "epoch": 0.9192845332932512,
      "flos": 16357669827840.0,
      "grad_norm": 1.896818980703734,
      "language_loss": 0.74321741,
      "learning_rate": 6.790496110568921e-08,
      "loss": 0.76424515,
      "num_input_tokens_seen": 329862020,
      "step": 15290,
      "time_per_iteration": 2.5002431869506836
    },
    {
      "auxiliary_loss_clip": 0.01060777,
      "auxiliary_loss_mlp": 0.01027813,
      "balance_loss_clip": 1.03403604,
      "balance_loss_mlp": 1.01658952,
      "epoch": 0.9193446565459191,
      "flos": 26614475464320.0,
      "grad_norm": 1.8721355423319446,
      "language_loss": 0.7229799,
      "learning_rate": 6.78043746849506e-08,
      "loss": 0.74386585,
      "num_input_tokens_seen": 329880185,
      "step": 15291,
      "time_per_iteration": 2.604869842529297
    },
    {
      "auxiliary_loss_clip": 0.01080018,
      "auxiliary_loss_mlp": 0.01027966,
      "balance_loss_clip": 1.03414011,
      "balance_loss_mlp": 1.01629531,
      "epoch": 0.9194047797985871,
      "flos": 22492182084480.0,
      "grad_norm": 1.8756319941532056,
      "language_loss": 0.70968866,
      "learning_rate": 6.770386153326346e-08,
      "loss": 0.73076844,
      "num_input_tokens_seen": 329900255,
      "step": 15292,
      "time_per_iteration": 2.5193614959716797
    },
    {
      "auxiliary_loss_clip": 0.01084532,
      "auxiliary_loss_mlp": 0.01027484,
      "balance_loss_clip": 1.0354085,
      "balance_loss_mlp": 1.01496148,
      "epoch": 0.9194649030512551,
      "flos": 25078791346560.0,
      "grad_norm": 1.6659608274742854,
      "language_loss": 0.73355043,
      "learning_rate": 6.760342165443988e-08,
      "loss": 0.75467056,
      "num_input_tokens_seen": 329919095,
      "step": 15293,
      "time_per_iteration": 2.5391714572906494
    },
    {
      "auxiliary_loss_clip": 0.01101747,
      "auxiliary_loss_mlp": 0.01027462,
      "balance_loss_clip": 1.03459489,
      "balance_loss_mlp": 1.01535022,
      "epoch": 0.9195250263039231,
      "flos": 11911139354880.0,
      "grad_norm": 1.9122250238217922,
      "language_loss": 0.77777666,
      "learning_rate": 6.750305505228837e-08,
      "loss": 0.79906875,
      "num_input_tokens_seen": 329936505,
      "step": 15294,
      "time_per_iteration": 2.412062168121338
    },
    {
      "auxiliary_loss_clip": 0.01083801,
      "auxiliary_loss_mlp": 0.01031513,
      "balance_loss_clip": 1.03432286,
      "balance_loss_mlp": 1.0182091,
      "epoch": 0.919585149556591,
      "flos": 21834154880640.0,
      "grad_norm": 3.7164919765360684,
      "language_loss": 0.77180743,
      "learning_rate": 6.74027617306141e-08,
      "loss": 0.79296064,
      "num_input_tokens_seen": 329956795,
      "step": 15295,
      "time_per_iteration": 3.9314160346984863
    },
    {
      "auxiliary_loss_clip": 0.01100947,
      "auxiliary_loss_mlp": 0.01027656,
      "balance_loss_clip": 1.03528368,
      "balance_loss_mlp": 1.01676059,
      "epoch": 0.919645272809259,
      "flos": 28184059042560.0,
      "grad_norm": 2.0026291092200563,
      "language_loss": 0.71451771,
      "learning_rate": 6.730254169322114e-08,
      "loss": 0.73580372,
      "num_input_tokens_seen": 329977195,
      "step": 15296,
      "time_per_iteration": 3.9286043643951416
    },
    {
      "auxiliary_loss_clip": 0.01104164,
      "auxiliary_loss_mlp": 0.01035314,
      "balance_loss_clip": 1.03567696,
      "balance_loss_mlp": 1.02391148,
      "epoch": 0.9197053960619269,
      "flos": 18332828847360.0,
      "grad_norm": 2.0236307254163006,
      "language_loss": 0.75362557,
      "learning_rate": 6.720239494390912e-08,
      "loss": 0.77502036,
      "num_input_tokens_seen": 329992095,
      "step": 15297,
      "time_per_iteration": 3.834601879119873
    },
    {
      "auxiliary_loss_clip": 0.01087401,
      "auxiliary_loss_mlp": 0.00782054,
      "balance_loss_clip": 1.0342555,
      "balance_loss_mlp": 1.00845051,
      "epoch": 0.9197655193145949,
      "flos": 28183448511360.0,
      "grad_norm": 1.606279363063911,
      "language_loss": 0.73662364,
      "learning_rate": 6.710232148647676e-08,
      "loss": 0.75531822,
      "num_input_tokens_seen": 330011490,
      "step": 15298,
      "time_per_iteration": 2.5185587406158447
    },
    {
      "auxiliary_loss_clip": 0.01077056,
      "auxiliary_loss_mlp": 0.01032006,
      "balance_loss_clip": 1.03549218,
      "balance_loss_mlp": 1.01988816,
      "epoch": 0.9198256425672628,
      "flos": 17306321973120.0,
      "grad_norm": 2.16245322592682,
      "language_loss": 0.79403865,
      "learning_rate": 6.70023213247175e-08,
      "loss": 0.81512928,
      "num_input_tokens_seen": 330027885,
      "step": 15299,
      "time_per_iteration": 2.5126662254333496
    },
    {
      "auxiliary_loss_clip": 0.01073879,
      "auxiliary_loss_mlp": 0.01027078,
      "balance_loss_clip": 1.03508615,
      "balance_loss_mlp": 1.01548529,
      "epoch": 0.9198857658199309,
      "flos": 17858520731520.0,
      "grad_norm": 1.9780485312503362,
      "language_loss": 0.63890815,
      "learning_rate": 6.690239446242385e-08,
      "loss": 0.65991771,
      "num_input_tokens_seen": 330046230,
      "step": 15300,
      "time_per_iteration": 2.512488603591919
    },
    {
      "auxiliary_loss_clip": 0.01076475,
      "auxiliary_loss_mlp": 0.00780134,
      "balance_loss_clip": 1.03428197,
      "balance_loss_mlp": 1.00825059,
      "epoch": 0.9199458890725988,
      "flos": 22127545169280.0,
      "grad_norm": 1.709746402449741,
      "language_loss": 0.69660699,
      "learning_rate": 6.680254090338545e-08,
      "loss": 0.71517301,
      "num_input_tokens_seen": 330065535,
      "step": 15301,
      "time_per_iteration": 2.533690929412842
    },
    {
      "auxiliary_loss_clip": 0.01093711,
      "auxiliary_loss_mlp": 0.0103862,
      "balance_loss_clip": 1.03551292,
      "balance_loss_mlp": 1.02432656,
      "epoch": 0.9200060123252668,
      "flos": 16034043265920.0,
      "grad_norm": 1.9184653229089332,
      "language_loss": 0.71280134,
      "learning_rate": 6.670276065138814e-08,
      "loss": 0.73412466,
      "num_input_tokens_seen": 330082920,
      "step": 15302,
      "time_per_iteration": 2.4526774883270264
    },
    {
      "auxiliary_loss_clip": 0.01104546,
      "auxiliary_loss_mlp": 0.01030644,
      "balance_loss_clip": 1.03477883,
      "balance_loss_mlp": 1.01900339,
      "epoch": 0.9200661355779348,
      "flos": 26864521015680.0,
      "grad_norm": 1.7022099778602027,
      "language_loss": 0.76536238,
      "learning_rate": 6.660305371021579e-08,
      "loss": 0.78671432,
      "num_input_tokens_seen": 330101165,
      "step": 15303,
      "time_per_iteration": 2.5090975761413574
    },
    {
      "auxiliary_loss_clip": 0.01083751,
      "auxiliary_loss_mlp": 0.01031696,
      "balance_loss_clip": 1.03569388,
      "balance_loss_mlp": 1.0201087,
      "epoch": 0.9201262588306027,
      "flos": 12786749193600.0,
      "grad_norm": 2.950933670800871,
      "language_loss": 0.88290632,
      "learning_rate": 6.650342008365006e-08,
      "loss": 0.90406084,
      "num_input_tokens_seen": 330118775,
      "step": 15304,
      "time_per_iteration": 2.4791369438171387
    },
    {
      "auxiliary_loss_clip": 0.01042162,
      "auxiliary_loss_mlp": 0.0103517,
      "balance_loss_clip": 1.03281498,
      "balance_loss_mlp": 1.02005994,
      "epoch": 0.9201863820832707,
      "flos": 20631614428800.0,
      "grad_norm": 1.8900297248507547,
      "language_loss": 0.77569425,
      "learning_rate": 6.64038597754677e-08,
      "loss": 0.79646754,
      "num_input_tokens_seen": 330135570,
      "step": 15305,
      "time_per_iteration": 4.0152506828308105
    },
    {
      "auxiliary_loss_clip": 0.01094046,
      "auxiliary_loss_mlp": 0.01036823,
      "balance_loss_clip": 1.03497005,
      "balance_loss_mlp": 1.02438354,
      "epoch": 0.9202465053359387,
      "flos": 26395815421440.0,
      "grad_norm": 2.3512156380013023,
      "language_loss": 0.81224144,
      "learning_rate": 6.630437278944501e-08,
      "loss": 0.83355004,
      "num_input_tokens_seen": 330152840,
      "step": 15306,
      "time_per_iteration": 2.543663740158081
    },
    {
      "auxiliary_loss_clip": 0.01063692,
      "auxiliary_loss_mlp": 0.01029445,
      "balance_loss_clip": 1.03237402,
      "balance_loss_mlp": 1.01862717,
      "epoch": 0.9203066285886067,
      "flos": 10488179093760.0,
      "grad_norm": 2.069422734852628,
      "language_loss": 0.72011024,
      "learning_rate": 6.62049591293541e-08,
      "loss": 0.74104166,
      "num_input_tokens_seen": 330168605,
      "step": 15307,
      "time_per_iteration": 2.539537191390991
    },
    {
      "auxiliary_loss_clip": 0.01094724,
      "auxiliary_loss_mlp": 0.01029387,
      "balance_loss_clip": 1.03431106,
      "balance_loss_mlp": 1.0173291,
      "epoch": 0.9203667518412746,
      "flos": 19390721230080.0,
      "grad_norm": 1.9444741708023532,
      "language_loss": 0.78754854,
      "learning_rate": 6.610561879896526e-08,
      "loss": 0.80878961,
      "num_input_tokens_seen": 330186160,
      "step": 15308,
      "time_per_iteration": 2.466181516647339
    },
    {
      "auxiliary_loss_clip": 0.01081328,
      "auxiliary_loss_mlp": 0.01033924,
      "balance_loss_clip": 1.03256583,
      "balance_loss_mlp": 1.02119827,
      "epoch": 0.9204268750939426,
      "flos": 15924982596480.0,
      "grad_norm": 1.9423492330194896,
      "language_loss": 0.78033197,
      "learning_rate": 6.600635180204484e-08,
      "loss": 0.80148453,
      "num_input_tokens_seen": 330201780,
      "step": 15309,
      "time_per_iteration": 2.4562456607818604
    },
    {
      "auxiliary_loss_clip": 0.01058518,
      "auxiliary_loss_mlp": 0.01032043,
      "balance_loss_clip": 1.03330255,
      "balance_loss_mlp": 1.01936471,
      "epoch": 0.9204869983466105,
      "flos": 16471758401280.0,
      "grad_norm": 1.908813401510127,
      "language_loss": 0.6636861,
      "learning_rate": 6.590715814235781e-08,
      "loss": 0.68459165,
      "num_input_tokens_seen": 330219165,
      "step": 15310,
      "time_per_iteration": 2.5644047260284424
    },
    {
      "auxiliary_loss_clip": 0.01042118,
      "auxiliary_loss_mlp": 0.01029393,
      "balance_loss_clip": 1.03401983,
      "balance_loss_mlp": 1.01697719,
      "epoch": 0.9205471215992785,
      "flos": 21539220307200.0,
      "grad_norm": 1.6748614394535124,
      "language_loss": 0.66263419,
      "learning_rate": 6.580803782366495e-08,
      "loss": 0.68334931,
      "num_input_tokens_seen": 330238975,
      "step": 15311,
      "time_per_iteration": 2.6151740550994873
    },
    {
      "auxiliary_loss_clip": 0.01092206,
      "auxiliary_loss_mlp": 0.01033065,
      "balance_loss_clip": 1.03338945,
      "balance_loss_mlp": 1.02129281,
      "epoch": 0.9206072448519464,
      "flos": 25005892694400.0,
      "grad_norm": 1.5813462715624276,
      "language_loss": 0.75833321,
      "learning_rate": 6.570899084972503e-08,
      "loss": 0.7795859,
      "num_input_tokens_seen": 330259755,
      "step": 15312,
      "time_per_iteration": 2.5255610942840576
    },
    {
      "auxiliary_loss_clip": 0.01089806,
      "auxiliary_loss_mlp": 0.01035472,
      "balance_loss_clip": 1.03459263,
      "balance_loss_mlp": 1.02382493,
      "epoch": 0.9206673681046145,
      "flos": 20522661500160.0,
      "grad_norm": 1.6671071058549505,
      "language_loss": 0.79009008,
      "learning_rate": 6.561001722429394e-08,
      "loss": 0.81134284,
      "num_input_tokens_seen": 330277660,
      "step": 15313,
      "time_per_iteration": 2.4712235927581787
    },
    {
      "auxiliary_loss_clip": 0.0109514,
      "auxiliary_loss_mlp": 0.01029851,
      "balance_loss_clip": 1.03409648,
      "balance_loss_mlp": 1.01789987,
      "epoch": 0.9207274913572824,
      "flos": 20883455660160.0,
      "grad_norm": 1.7547122738282617,
      "language_loss": 0.78118634,
      "learning_rate": 6.55111169511251e-08,
      "loss": 0.80243623,
      "num_input_tokens_seen": 330295455,
      "step": 15314,
      "time_per_iteration": 2.4814462661743164
    },
    {
      "auxiliary_loss_clip": 0.01087179,
      "auxiliary_loss_mlp": 0.01033238,
      "balance_loss_clip": 1.03453732,
      "balance_loss_mlp": 1.01992273,
      "epoch": 0.9207876146099504,
      "flos": 22708256348160.0,
      "grad_norm": 1.7454536002851013,
      "language_loss": 0.79418576,
      "learning_rate": 6.541229003396864e-08,
      "loss": 0.81538999,
      "num_input_tokens_seen": 330315310,
      "step": 15315,
      "time_per_iteration": 2.511749505996704
    },
    {
      "auxiliary_loss_clip": 0.01085864,
      "auxiliary_loss_mlp": 0.01028708,
      "balance_loss_clip": 1.0354178,
      "balance_loss_mlp": 1.01656079,
      "epoch": 0.9208477378626184,
      "flos": 18507354053760.0,
      "grad_norm": 1.8648499700847674,
      "language_loss": 0.76362801,
      "learning_rate": 6.531353647657156e-08,
      "loss": 0.78477371,
      "num_input_tokens_seen": 330333260,
      "step": 15316,
      "time_per_iteration": 2.4958906173706055
    },
    {
      "auxiliary_loss_clip": 0.01103238,
      "auxiliary_loss_mlp": 0.01034532,
      "balance_loss_clip": 1.03299415,
      "balance_loss_mlp": 1.02191341,
      "epoch": 0.9209078611152863,
      "flos": 22999635475200.0,
      "grad_norm": 1.570968245456868,
      "language_loss": 0.69556904,
      "learning_rate": 6.521485628267931e-08,
      "loss": 0.71694672,
      "num_input_tokens_seen": 330352465,
      "step": 15317,
      "time_per_iteration": 2.455543279647827
    },
    {
      "auxiliary_loss_clip": 0.01092714,
      "auxiliary_loss_mlp": 0.01032887,
      "balance_loss_clip": 1.03508806,
      "balance_loss_mlp": 1.02086461,
      "epoch": 0.9209679843679544,
      "flos": 24061514267520.0,
      "grad_norm": 1.6603512074571751,
      "language_loss": 0.83340019,
      "learning_rate": 6.511624945603378e-08,
      "loss": 0.85465622,
      "num_input_tokens_seen": 330372685,
      "step": 15318,
      "time_per_iteration": 2.513838768005371
    },
    {
      "auxiliary_loss_clip": 0.0108276,
      "auxiliary_loss_mlp": 0.01031175,
      "balance_loss_clip": 1.03586173,
      "balance_loss_mlp": 1.01922464,
      "epoch": 0.9210281076206223,
      "flos": 13553370190080.0,
      "grad_norm": 1.9708649585423381,
      "language_loss": 0.85778832,
      "learning_rate": 6.501771600037354e-08,
      "loss": 0.87892771,
      "num_input_tokens_seen": 330388860,
      "step": 15319,
      "time_per_iteration": 2.490665912628174
    },
    {
      "auxiliary_loss_clip": 0.01027109,
      "auxiliary_loss_mlp": 0.01002849,
      "balance_loss_clip": 1.00437868,
      "balance_loss_mlp": 1.00178194,
      "epoch": 0.9210882308732903,
      "flos": 71426289674880.0,
      "grad_norm": 0.7697389946901002,
      "language_loss": 0.5616408,
      "learning_rate": 6.491925591943559e-08,
      "loss": 0.58194041,
      "num_input_tokens_seen": 330448735,
      "step": 15320,
      "time_per_iteration": 3.102128028869629
    },
    {
      "auxiliary_loss_clip": 0.01054398,
      "auxiliary_loss_mlp": 0.01045986,
      "balance_loss_clip": 1.03440881,
      "balance_loss_mlp": 1.03134704,
      "epoch": 0.9211483541259582,
      "flos": 18509113820160.0,
      "grad_norm": 2.0675602197678122,
      "language_loss": 0.63923061,
      "learning_rate": 6.482086921695384e-08,
      "loss": 0.66023445,
      "num_input_tokens_seen": 330465600,
      "step": 15321,
      "time_per_iteration": 2.5387611389160156
    },
    {
      "auxiliary_loss_clip": 0.01062904,
      "auxiliary_loss_mlp": 0.01027026,
      "balance_loss_clip": 1.03370619,
      "balance_loss_mlp": 1.01577878,
      "epoch": 0.9212084773786262,
      "flos": 23258228463360.0,
      "grad_norm": 1.4578470743154641,
      "language_loss": 0.71527946,
      "learning_rate": 6.47225558966582e-08,
      "loss": 0.73617876,
      "num_input_tokens_seen": 330485770,
      "step": 15322,
      "time_per_iteration": 2.582439422607422
    },
    {
      "auxiliary_loss_clip": 0.01053485,
      "auxiliary_loss_mlp": 0.01031148,
      "balance_loss_clip": 1.03409791,
      "balance_loss_mlp": 1.02004921,
      "epoch": 0.9212686006312941,
      "flos": 16289511770880.0,
      "grad_norm": 2.5801631647751933,
      "language_loss": 0.7003448,
      "learning_rate": 6.462431596227725e-08,
      "loss": 0.72119117,
      "num_input_tokens_seen": 330504255,
      "step": 15323,
      "time_per_iteration": 2.5533764362335205
    },
    {
      "auxiliary_loss_clip": 0.0107725,
      "auxiliary_loss_mlp": 0.0103675,
      "balance_loss_clip": 1.031299,
      "balance_loss_mlp": 1.02301157,
      "epoch": 0.9213287238839621,
      "flos": 19785773986560.0,
      "grad_norm": 2.002888392729793,
      "language_loss": 0.7482717,
      "learning_rate": 6.452614941753597e-08,
      "loss": 0.76941168,
      "num_input_tokens_seen": 330520705,
      "step": 15324,
      "time_per_iteration": 2.5419020652770996
    },
    {
      "auxiliary_loss_clip": 0.01093649,
      "auxiliary_loss_mlp": 0.010397,
      "balance_loss_clip": 1.03534532,
      "balance_loss_mlp": 1.02801704,
      "epoch": 0.92138884713663,
      "flos": 21030402199680.0,
      "grad_norm": 4.392890117614824,
      "language_loss": 0.71226996,
      "learning_rate": 6.442805626615744e-08,
      "loss": 0.73360348,
      "num_input_tokens_seen": 330539245,
      "step": 15325,
      "time_per_iteration": 2.485720157623291
    },
    {
      "auxiliary_loss_clip": 0.01078369,
      "auxiliary_loss_mlp": 0.01033146,
      "balance_loss_clip": 1.0361563,
      "balance_loss_mlp": 1.02098036,
      "epoch": 0.9214489703892981,
      "flos": 28587264186240.0,
      "grad_norm": 1.5136370994165007,
      "language_loss": 0.78510725,
      "learning_rate": 6.433003651186109e-08,
      "loss": 0.80622238,
      "num_input_tokens_seen": 330561815,
      "step": 15326,
      "time_per_iteration": 2.5969414710998535
    },
    {
      "auxiliary_loss_clip": 0.01097148,
      "auxiliary_loss_mlp": 0.01033586,
      "balance_loss_clip": 1.03682137,
      "balance_loss_mlp": 1.02114093,
      "epoch": 0.921509093641966,
      "flos": 16361476669440.0,
      "grad_norm": 1.9000367324488865,
      "language_loss": 0.71017134,
      "learning_rate": 6.42320901583635e-08,
      "loss": 0.73147869,
      "num_input_tokens_seen": 330579760,
      "step": 15327,
      "time_per_iteration": 2.475288152694702
    },
    {
      "auxiliary_loss_clip": 0.01098295,
      "auxiliary_loss_mlp": 0.01039789,
      "balance_loss_clip": 1.0376333,
      "balance_loss_mlp": 1.02704537,
      "epoch": 0.921569216894634,
      "flos": 26830837036800.0,
      "grad_norm": 2.1551751044103864,
      "language_loss": 0.77588189,
      "learning_rate": 6.413421720937906e-08,
      "loss": 0.79726267,
      "num_input_tokens_seen": 330598545,
      "step": 15328,
      "time_per_iteration": 2.5258400440216064
    },
    {
      "auxiliary_loss_clip": 0.01082236,
      "auxiliary_loss_mlp": 0.01030275,
      "balance_loss_clip": 1.03419197,
      "balance_loss_mlp": 1.01859248,
      "epoch": 0.921629340147302,
      "flos": 24645134448000.0,
      "grad_norm": 2.093393932371499,
      "language_loss": 0.71456385,
      "learning_rate": 6.4036417668619e-08,
      "loss": 0.73568904,
      "num_input_tokens_seen": 330616700,
      "step": 15329,
      "time_per_iteration": 2.5281929969787598
    },
    {
      "auxiliary_loss_clip": 0.01090148,
      "auxiliary_loss_mlp": 0.01025434,
      "balance_loss_clip": 1.03327501,
      "balance_loss_mlp": 1.01460385,
      "epoch": 0.9216894633999699,
      "flos": 15086504442240.0,
      "grad_norm": 1.7362849360005825,
      "language_loss": 0.86587024,
      "learning_rate": 6.393869153979192e-08,
      "loss": 0.88702607,
      "num_input_tokens_seen": 330633355,
      "step": 15330,
      "time_per_iteration": 2.444657325744629
    },
    {
      "auxiliary_loss_clip": 0.01070256,
      "auxiliary_loss_mlp": 0.010339,
      "balance_loss_clip": 1.03217876,
      "balance_loss_mlp": 1.0212822,
      "epoch": 0.921749586652638,
      "flos": 19204524103680.0,
      "grad_norm": 2.5529340885180196,
      "language_loss": 0.76028538,
      "learning_rate": 6.384103882660397e-08,
      "loss": 0.78132701,
      "num_input_tokens_seen": 330651470,
      "step": 15331,
      "time_per_iteration": 2.512305498123169
    },
    {
      "auxiliary_loss_clip": 0.01091791,
      "auxiliary_loss_mlp": 0.01027684,
      "balance_loss_clip": 1.03368807,
      "balance_loss_mlp": 1.01618052,
      "epoch": 0.9218097099053059,
      "flos": 20522446018560.0,
      "grad_norm": 1.991425615363305,
      "language_loss": 0.75559378,
      "learning_rate": 6.374345953275794e-08,
      "loss": 0.77678853,
      "num_input_tokens_seen": 330669170,
      "step": 15332,
      "time_per_iteration": 2.49147629737854
    },
    {
      "auxiliary_loss_clip": 0.01059733,
      "auxiliary_loss_mlp": 0.01028362,
      "balance_loss_clip": 1.03257132,
      "balance_loss_mlp": 1.01719785,
      "epoch": 0.9218698331579739,
      "flos": 17348625216000.0,
      "grad_norm": 1.7706018550983702,
      "language_loss": 0.74942291,
      "learning_rate": 6.364595366195358e-08,
      "loss": 0.77030385,
      "num_input_tokens_seen": 330686635,
      "step": 15333,
      "time_per_iteration": 2.537027597427368
    },
    {
      "auxiliary_loss_clip": 0.01017431,
      "auxiliary_loss_mlp": 0.00999931,
      "balance_loss_clip": 1.00806785,
      "balance_loss_mlp": 0.99891776,
      "epoch": 0.9219299564106418,
      "flos": 61958332575360.0,
      "grad_norm": 0.8130602688427413,
      "language_loss": 0.52921045,
      "learning_rate": 6.354852121788879e-08,
      "loss": 0.54938412,
      "num_input_tokens_seen": 330749160,
      "step": 15334,
      "time_per_iteration": 4.445763111114502
    },
    {
      "auxiliary_loss_clip": 0.01074017,
      "auxiliary_loss_mlp": 0.01034677,
      "balance_loss_clip": 1.03325486,
      "balance_loss_mlp": 1.02305424,
      "epoch": 0.9219900796633098,
      "flos": 15701761526400.0,
      "grad_norm": 1.9475931920857337,
      "language_loss": 0.62048411,
      "learning_rate": 6.345116220425839e-08,
      "loss": 0.64157104,
      "num_input_tokens_seen": 330766840,
      "step": 15335,
      "time_per_iteration": 3.869744062423706
    },
    {
      "auxiliary_loss_clip": 0.01059778,
      "auxiliary_loss_mlp": 0.01030913,
      "balance_loss_clip": 1.03233945,
      "balance_loss_mlp": 1.01824677,
      "epoch": 0.9220502029159777,
      "flos": 24932670819840.0,
      "grad_norm": 1.7388218049749053,
      "language_loss": 0.71816695,
      "learning_rate": 6.335387662475366e-08,
      "loss": 0.73907381,
      "num_input_tokens_seen": 330785585,
      "step": 15336,
      "time_per_iteration": 3.99977970123291
    },
    {
      "auxiliary_loss_clip": 0.01076634,
      "auxiliary_loss_mlp": 0.01028603,
      "balance_loss_clip": 1.03186929,
      "balance_loss_mlp": 1.0186727,
      "epoch": 0.9221103261686457,
      "flos": 15667215621120.0,
      "grad_norm": 1.8873864867308798,
      "language_loss": 0.71855891,
      "learning_rate": 6.325666448306433e-08,
      "loss": 0.73961121,
      "num_input_tokens_seen": 330800750,
      "step": 15337,
      "time_per_iteration": 2.4621264934539795
    },
    {
      "auxiliary_loss_clip": 0.01021663,
      "auxiliary_loss_mlp": 0.01002931,
      "balance_loss_clip": 1.00884008,
      "balance_loss_mlp": 1.00190568,
      "epoch": 0.9221704494213137,
      "flos": 67516299630720.0,
      "grad_norm": 0.8874000157194111,
      "language_loss": 0.65348989,
      "learning_rate": 6.31595257828763e-08,
      "loss": 0.67373586,
      "num_input_tokens_seen": 330863640,
      "step": 15338,
      "time_per_iteration": 3.0703814029693604
    },
    {
      "auxiliary_loss_clip": 0.01095781,
      "auxiliary_loss_mlp": 0.01032066,
      "balance_loss_clip": 1.0359211,
      "balance_loss_mlp": 1.01992488,
      "epoch": 0.9222305726739817,
      "flos": 30226945155840.0,
      "grad_norm": 1.659588688091902,
      "language_loss": 0.6708104,
      "learning_rate": 6.306246052787289e-08,
      "loss": 0.6920889,
      "num_input_tokens_seen": 330884675,
      "step": 15339,
      "time_per_iteration": 2.5369646549224854
    },
    {
      "auxiliary_loss_clip": 0.01104123,
      "auxiliary_loss_mlp": 0.0102992,
      "balance_loss_clip": 1.03528821,
      "balance_loss_mlp": 1.0179987,
      "epoch": 0.9222906959266496,
      "flos": 25337204766720.0,
      "grad_norm": 1.9222084383787847,
      "language_loss": 0.71990204,
      "learning_rate": 6.296546872173513e-08,
      "loss": 0.74124253,
      "num_input_tokens_seen": 330904125,
      "step": 15340,
      "time_per_iteration": 2.49969744682312
    },
    {
      "auxiliary_loss_clip": 0.0106914,
      "auxiliary_loss_mlp": 0.0103242,
      "balance_loss_clip": 1.03315377,
      "balance_loss_mlp": 1.02096438,
      "epoch": 0.9223508191793176,
      "flos": 27599864244480.0,
      "grad_norm": 1.6310514687895812,
      "language_loss": 0.70102286,
      "learning_rate": 6.286855036814098e-08,
      "loss": 0.72203845,
      "num_input_tokens_seen": 330925140,
      "step": 15341,
      "time_per_iteration": 2.595306396484375
    },
    {
      "auxiliary_loss_clip": 0.01050268,
      "auxiliary_loss_mlp": 0.01025795,
      "balance_loss_clip": 1.03383625,
      "balance_loss_mlp": 1.01514339,
      "epoch": 0.9224109424319856,
      "flos": 27307587277440.0,
      "grad_norm": 1.598531829517011,
      "language_loss": 0.67233884,
      "learning_rate": 6.277170547076571e-08,
      "loss": 0.6930995,
      "num_input_tokens_seen": 330946625,
      "step": 15342,
      "time_per_iteration": 2.6309854984283447
    },
    {
      "auxiliary_loss_clip": 0.01054448,
      "auxiliary_loss_mlp": 0.01035266,
      "balance_loss_clip": 1.03568721,
      "balance_loss_mlp": 1.02434635,
      "epoch": 0.9224710656846535,
      "flos": 48208314401280.0,
      "grad_norm": 2.0147657986180696,
      "language_loss": 0.7011528,
      "learning_rate": 6.26749340332815e-08,
      "loss": 0.72204989,
      "num_input_tokens_seen": 330967795,
      "step": 15343,
      "time_per_iteration": 2.820791482925415
    },
    {
      "auxiliary_loss_clip": 0.01010143,
      "auxiliary_loss_mlp": 0.01001698,
      "balance_loss_clip": 1.00678563,
      "balance_loss_mlp": 1.00063097,
      "epoch": 0.9225311889373216,
      "flos": 66722171794560.0,
      "grad_norm": 0.7186402232174223,
      "language_loss": 0.51975429,
      "learning_rate": 6.257823605935786e-08,
      "loss": 0.53987277,
      "num_input_tokens_seen": 331040850,
      "step": 15344,
      "time_per_iteration": 4.775454044342041
    },
    {
      "auxiliary_loss_clip": 0.01099677,
      "auxiliary_loss_mlp": 0.01032922,
      "balance_loss_clip": 1.03531122,
      "balance_loss_mlp": 1.02197278,
      "epoch": 0.9225913121899895,
      "flos": 22271295398400.0,
      "grad_norm": 1.7131598865236317,
      "language_loss": 0.70304465,
      "learning_rate": 6.248161155266162e-08,
      "loss": 0.72437072,
      "num_input_tokens_seen": 331060595,
      "step": 15345,
      "time_per_iteration": 2.4816977977752686
    },
    {
      "auxiliary_loss_clip": 0.01081891,
      "auxiliary_loss_mlp": 0.0103334,
      "balance_loss_clip": 1.03434682,
      "balance_loss_mlp": 1.02146661,
      "epoch": 0.9226514354426575,
      "flos": 20082719721600.0,
      "grad_norm": 1.6738188702628254,
      "language_loss": 0.77442127,
      "learning_rate": 6.238506051685677e-08,
      "loss": 0.79557359,
      "num_input_tokens_seen": 331080195,
      "step": 15346,
      "time_per_iteration": 2.512470006942749
    },
    {
      "auxiliary_loss_clip": 0.01085972,
      "auxiliary_loss_mlp": 0.01036422,
      "balance_loss_clip": 1.03895926,
      "balance_loss_mlp": 1.02403581,
      "epoch": 0.9227115586953254,
      "flos": 16070851728000.0,
      "grad_norm": 2.6023050587893075,
      "language_loss": 0.76312846,
      "learning_rate": 6.228858295560457e-08,
      "loss": 0.78435242,
      "num_input_tokens_seen": 331097645,
      "step": 15347,
      "time_per_iteration": 2.4999592304229736
    },
    {
      "auxiliary_loss_clip": 0.01089971,
      "auxiliary_loss_mlp": 0.01028094,
      "balance_loss_clip": 1.03590965,
      "balance_loss_mlp": 1.01727581,
      "epoch": 0.9227716819479934,
      "flos": 20446027833600.0,
      "grad_norm": 1.4588192998855583,
      "language_loss": 0.76932937,
      "learning_rate": 6.219217887256367e-08,
      "loss": 0.79051006,
      "num_input_tokens_seen": 331116830,
      "step": 15348,
      "time_per_iteration": 2.471672296524048
    },
    {
      "auxiliary_loss_clip": 0.0108087,
      "auxiliary_loss_mlp": 0.01031747,
      "balance_loss_clip": 1.03266358,
      "balance_loss_mlp": 1.01995671,
      "epoch": 0.9228318052006613,
      "flos": 25007401065600.0,
      "grad_norm": 1.8391556812880392,
      "language_loss": 0.68387282,
      "learning_rate": 6.209584827138959e-08,
      "loss": 0.70499897,
      "num_input_tokens_seen": 331137235,
      "step": 15349,
      "time_per_iteration": 2.543954610824585
    },
    {
      "auxiliary_loss_clip": 0.01071503,
      "auxiliary_loss_mlp": 0.01027692,
      "balance_loss_clip": 1.03407478,
      "balance_loss_mlp": 1.01559854,
      "epoch": 0.9228919284533293,
      "flos": 12677257560960.0,
      "grad_norm": 2.84435317012987,
      "language_loss": 0.86658359,
      "learning_rate": 6.199959115573495e-08,
      "loss": 0.88757557,
      "num_input_tokens_seen": 331153155,
      "step": 15350,
      "time_per_iteration": 2.499349594116211
    },
    {
      "auxiliary_loss_clip": 0.01012558,
      "auxiliary_loss_mlp": 0.01003494,
      "balance_loss_clip": 1.00917482,
      "balance_loss_mlp": 1.00246894,
      "epoch": 0.9229520517059973,
      "flos": 69986162712960.0,
      "grad_norm": 0.7985847904083166,
      "language_loss": 0.60372984,
      "learning_rate": 6.190340752924994e-08,
      "loss": 0.6238904,
      "num_input_tokens_seen": 331214895,
      "step": 15351,
      "time_per_iteration": 3.093109130859375
    },
    {
      "auxiliary_loss_clip": 0.01079596,
      "auxiliary_loss_mlp": 0.01024963,
      "balance_loss_clip": 1.03454947,
      "balance_loss_mlp": 1.0136863,
      "epoch": 0.9230121749586653,
      "flos": 14793832425600.0,
      "grad_norm": 2.0112581682705293,
      "language_loss": 0.77872443,
      "learning_rate": 6.180729739558233e-08,
      "loss": 0.79977,
      "num_input_tokens_seen": 331232185,
      "step": 15352,
      "time_per_iteration": 2.5113823413848877
    },
    {
      "auxiliary_loss_clip": 0.01069173,
      "auxiliary_loss_mlp": 0.01043202,
      "balance_loss_clip": 1.03213787,
      "balance_loss_mlp": 1.02883768,
      "epoch": 0.9230722982113332,
      "flos": 22967208472320.0,
      "grad_norm": 2.071111305729824,
      "language_loss": 0.59742606,
      "learning_rate": 6.171126075837585e-08,
      "loss": 0.61854982,
      "num_input_tokens_seen": 331251065,
      "step": 15353,
      "time_per_iteration": 2.537743330001831
    },
    {
      "auxiliary_loss_clip": 0.01078838,
      "auxiliary_loss_mlp": 0.01027795,
      "balance_loss_clip": 1.03419232,
      "balance_loss_mlp": 1.01667261,
      "epoch": 0.9231324214640012,
      "flos": 18551452976640.0,
      "grad_norm": 1.6042350169385307,
      "language_loss": 0.74958223,
      "learning_rate": 6.161529762127293e-08,
      "loss": 0.7706486,
      "num_input_tokens_seen": 331269110,
      "step": 15354,
      "time_per_iteration": 2.4981327056884766
    },
    {
      "auxiliary_loss_clip": 0.01106284,
      "auxiliary_loss_mlp": 0.01033678,
      "balance_loss_clip": 1.0348866,
      "balance_loss_mlp": 1.02090514,
      "epoch": 0.9231925447166691,
      "flos": 22082727974400.0,
      "grad_norm": 2.0209552971458873,
      "language_loss": 0.64738023,
      "learning_rate": 6.1519407987912e-08,
      "loss": 0.66877985,
      "num_input_tokens_seen": 331286555,
      "step": 15355,
      "time_per_iteration": 2.446593761444092
    },
    {
      "auxiliary_loss_clip": 0.01077428,
      "auxiliary_loss_mlp": 0.01040719,
      "balance_loss_clip": 1.03367972,
      "balance_loss_mlp": 1.02794015,
      "epoch": 0.9232526679693371,
      "flos": 26541145848960.0,
      "grad_norm": 1.438156126411136,
      "language_loss": 0.73994273,
      "learning_rate": 6.142359186192947e-08,
      "loss": 0.76112419,
      "num_input_tokens_seen": 331307660,
      "step": 15356,
      "time_per_iteration": 2.5697686672210693
    },
    {
      "auxiliary_loss_clip": 0.01080641,
      "auxiliary_loss_mlp": 0.01030124,
      "balance_loss_clip": 1.03502989,
      "balance_loss_mlp": 1.0179888,
      "epoch": 0.9233127912220052,
      "flos": 14756664827520.0,
      "grad_norm": 2.1867095262514646,
      "language_loss": 0.61056757,
      "learning_rate": 6.132784924695844e-08,
      "loss": 0.63167524,
      "num_input_tokens_seen": 331324885,
      "step": 15357,
      "time_per_iteration": 2.480957269668579
    },
    {
      "auxiliary_loss_clip": 0.0107519,
      "auxiliary_loss_mlp": 0.01029792,
      "balance_loss_clip": 1.03422666,
      "balance_loss_mlp": 1.01739979,
      "epoch": 0.9233729144746731,
      "flos": 25261792162560.0,
      "grad_norm": 1.3975935172321932,
      "language_loss": 0.6999867,
      "learning_rate": 6.123218014662956e-08,
      "loss": 0.72103655,
      "num_input_tokens_seen": 331345885,
      "step": 15358,
      "time_per_iteration": 2.5956897735595703
    },
    {
      "auxiliary_loss_clip": 0.01102913,
      "auxiliary_loss_mlp": 0.01029022,
      "balance_loss_clip": 1.03467941,
      "balance_loss_mlp": 1.01694,
      "epoch": 0.9234330377273411,
      "flos": 27849837968640.0,
      "grad_norm": 1.9131240476351619,
      "language_loss": 0.73143566,
      "learning_rate": 6.113658456457104e-08,
      "loss": 0.75275493,
      "num_input_tokens_seen": 331364320,
      "step": 15359,
      "time_per_iteration": 2.489328145980835
    },
    {
      "auxiliary_loss_clip": 0.01043036,
      "auxiliary_loss_mlp": 0.01031287,
      "balance_loss_clip": 1.03418708,
      "balance_loss_mlp": 1.01950955,
      "epoch": 0.923493160980009,
      "flos": 24608361899520.0,
      "grad_norm": 1.8504481653968372,
      "language_loss": 0.6458236,
      "learning_rate": 6.104106250440732e-08,
      "loss": 0.66656679,
      "num_input_tokens_seen": 331384135,
      "step": 15360,
      "time_per_iteration": 2.6618993282318115
    },
    {
      "auxiliary_loss_clip": 0.01018348,
      "auxiliary_loss_mlp": 0.00761268,
      "balance_loss_clip": 1.00531626,
      "balance_loss_mlp": 1.00086486,
      "epoch": 0.923553284232677,
      "flos": 67700916558720.0,
      "grad_norm": 0.7684964126390791,
      "language_loss": 0.55178714,
      "learning_rate": 6.094561396976083e-08,
      "loss": 0.5695833,
      "num_input_tokens_seen": 331440645,
      "step": 15361,
      "time_per_iteration": 3.036098003387451
    },
    {
      "auxiliary_loss_clip": 0.01070997,
      "auxiliary_loss_mlp": 0.0103038,
      "balance_loss_clip": 1.03143299,
      "balance_loss_mlp": 1.01765418,
      "epoch": 0.9236134074853449,
      "flos": 18807244704000.0,
      "grad_norm": 1.7250312797749559,
      "language_loss": 0.69568539,
      "learning_rate": 6.085023896425112e-08,
      "loss": 0.71669912,
      "num_input_tokens_seen": 331459580,
      "step": 15362,
      "time_per_iteration": 2.5458106994628906
    },
    {
      "auxiliary_loss_clip": 0.01089719,
      "auxiliary_loss_mlp": 0.01032475,
      "balance_loss_clip": 1.03297615,
      "balance_loss_mlp": 1.01815784,
      "epoch": 0.923673530738013,
      "flos": 27782362270080.0,
      "grad_norm": 1.3781407957199678,
      "language_loss": 0.75834489,
      "learning_rate": 6.075493749149463e-08,
      "loss": 0.77956688,
      "num_input_tokens_seen": 331481560,
      "step": 15363,
      "time_per_iteration": 2.5190842151641846
    },
    {
      "auxiliary_loss_clip": 0.0110392,
      "auxiliary_loss_mlp": 0.01030533,
      "balance_loss_clip": 1.03550911,
      "balance_loss_mlp": 1.01906526,
      "epoch": 0.9237336539906809,
      "flos": 26797117144320.0,
      "grad_norm": 1.9264326552740696,
      "language_loss": 0.83196497,
      "learning_rate": 6.065970955510514e-08,
      "loss": 0.85330951,
      "num_input_tokens_seen": 331499090,
      "step": 15364,
      "time_per_iteration": 2.519832134246826
    },
    {
      "auxiliary_loss_clip": 0.01067195,
      "auxiliary_loss_mlp": 0.01025821,
      "balance_loss_clip": 1.03186655,
      "balance_loss_mlp": 1.01485384,
      "epoch": 0.9237937772433489,
      "flos": 23587708942080.0,
      "grad_norm": 1.52794681436461,
      "language_loss": 0.68020368,
      "learning_rate": 6.056455515869419e-08,
      "loss": 0.70113385,
      "num_input_tokens_seen": 331519420,
      "step": 15365,
      "time_per_iteration": 2.5810720920562744
    },
    {
      "auxiliary_loss_clip": 0.0110471,
      "auxiliary_loss_mlp": 0.01029011,
      "balance_loss_clip": 1.03618717,
      "balance_loss_mlp": 1.0168221,
      "epoch": 0.9238539004960168,
      "flos": 26140562398080.0,
      "grad_norm": 1.8089633106304939,
      "language_loss": 0.63045675,
      "learning_rate": 6.046947430586913e-08,
      "loss": 0.65179396,
      "num_input_tokens_seen": 331538720,
      "step": 15366,
      "time_per_iteration": 2.506680727005005
    },
    {
      "auxiliary_loss_clip": 0.01072426,
      "auxiliary_loss_mlp": 0.01028498,
      "balance_loss_clip": 1.03723681,
      "balance_loss_mlp": 1.01602829,
      "epoch": 0.9239140237486848,
      "flos": 21068000760960.0,
      "grad_norm": 1.5381631441492354,
      "language_loss": 0.74581707,
      "learning_rate": 6.037446700023619e-08,
      "loss": 0.76682627,
      "num_input_tokens_seen": 331558505,
      "step": 15367,
      "time_per_iteration": 2.5417091846466064
    },
    {
      "auxiliary_loss_clip": 0.01078722,
      "auxiliary_loss_mlp": 0.00781531,
      "balance_loss_clip": 1.03402984,
      "balance_loss_mlp": 1.00892639,
      "epoch": 0.9239741470013527,
      "flos": 24607930936320.0,
      "grad_norm": 1.7433315357287282,
      "language_loss": 0.64546108,
      "learning_rate": 6.027953324539759e-08,
      "loss": 0.66406369,
      "num_input_tokens_seen": 331578440,
      "step": 15368,
      "time_per_iteration": 2.5368478298187256
    },
    {
      "auxiliary_loss_clip": 0.01096622,
      "auxiliary_loss_mlp": 0.01033848,
      "balance_loss_clip": 1.03514671,
      "balance_loss_mlp": 1.02125931,
      "epoch": 0.9240342702540207,
      "flos": 24718248581760.0,
      "grad_norm": 1.7404078847425666,
      "language_loss": 0.74458885,
      "learning_rate": 6.018467304495401e-08,
      "loss": 0.76589358,
      "num_input_tokens_seen": 331598945,
      "step": 15369,
      "time_per_iteration": 2.511030435562134
    },
    {
      "auxiliary_loss_clip": 0.01098175,
      "auxiliary_loss_mlp": 0.01037098,
      "balance_loss_clip": 1.0367372,
      "balance_loss_mlp": 1.02352571,
      "epoch": 0.9240943935066888,
      "flos": 20849987162880.0,
      "grad_norm": 2.4459533163843394,
      "language_loss": 0.76723063,
      "learning_rate": 6.008988640250145e-08,
      "loss": 0.78858334,
      "num_input_tokens_seen": 331616700,
      "step": 15370,
      "time_per_iteration": 2.482431650161743
    },
    {
      "auxiliary_loss_clip": 0.01103976,
      "auxiliary_loss_mlp": 0.01034779,
      "balance_loss_clip": 1.03450656,
      "balance_loss_mlp": 1.02298963,
      "epoch": 0.9241545167593567,
      "flos": 24462313200000.0,
      "grad_norm": 2.45623226821063,
      "language_loss": 0.66422796,
      "learning_rate": 5.999517332163528e-08,
      "loss": 0.68561554,
      "num_input_tokens_seen": 331635625,
      "step": 15371,
      "time_per_iteration": 2.4696807861328125
    },
    {
      "auxiliary_loss_clip": 0.01014214,
      "auxiliary_loss_mlp": 0.00999797,
      "balance_loss_clip": 1.0101943,
      "balance_loss_mlp": 0.99874175,
      "epoch": 0.9242146400120247,
      "flos": 61827259847040.0,
      "grad_norm": 0.7227713739762819,
      "language_loss": 0.57732928,
      "learning_rate": 5.99005338059464e-08,
      "loss": 0.59746939,
      "num_input_tokens_seen": 331698595,
      "step": 15372,
      "time_per_iteration": 4.471449375152588
    },
    {
      "auxiliary_loss_clip": 0.01104338,
      "auxiliary_loss_mlp": 0.01032729,
      "balance_loss_clip": 1.03743029,
      "balance_loss_mlp": 1.02169633,
      "epoch": 0.9242747632646926,
      "flos": 22048397550720.0,
      "grad_norm": 2.008672491639872,
      "language_loss": 0.69711292,
      "learning_rate": 5.98059678590237e-08,
      "loss": 0.71848357,
      "num_input_tokens_seen": 331717975,
      "step": 15373,
      "time_per_iteration": 2.4489212036132812
    },
    {
      "auxiliary_loss_clip": 0.01089984,
      "auxiliary_loss_mlp": 0.01037548,
      "balance_loss_clip": 1.03452051,
      "balance_loss_mlp": 1.02531755,
      "epoch": 0.9243348865173606,
      "flos": 18478338842880.0,
      "grad_norm": 2.394687020661308,
      "language_loss": 0.75264949,
      "learning_rate": 5.971147548445299e-08,
      "loss": 0.77392483,
      "num_input_tokens_seen": 331737220,
      "step": 15374,
      "time_per_iteration": 3.851379632949829
    },
    {
      "auxiliary_loss_clip": 0.01066825,
      "auxiliary_loss_mlp": 0.01036013,
      "balance_loss_clip": 1.03256285,
      "balance_loss_mlp": 1.02412224,
      "epoch": 0.9243950097700285,
      "flos": 23258767167360.0,
      "grad_norm": 1.7190983868924739,
      "language_loss": 0.64867377,
      "learning_rate": 5.961705668581784e-08,
      "loss": 0.66970217,
      "num_input_tokens_seen": 331757300,
      "step": 15375,
      "time_per_iteration": 3.96248722076416
    },
    {
      "auxiliary_loss_clip": 0.01076089,
      "auxiliary_loss_mlp": 0.01033334,
      "balance_loss_clip": 1.03615224,
      "balance_loss_mlp": 1.02125204,
      "epoch": 0.9244551330226966,
      "flos": 29749081593600.0,
      "grad_norm": 1.7223623955491305,
      "language_loss": 0.66198659,
      "learning_rate": 5.952271146669829e-08,
      "loss": 0.68308085,
      "num_input_tokens_seen": 331776995,
      "step": 15376,
      "time_per_iteration": 2.5903499126434326
    },
    {
      "auxiliary_loss_clip": 0.01027326,
      "auxiliary_loss_mlp": 0.01002635,
      "balance_loss_clip": 1.00462818,
      "balance_loss_mlp": 1.00162721,
      "epoch": 0.9245152562753645,
      "flos": 68864960609280.0,
      "grad_norm": 0.6609377435346611,
      "language_loss": 0.61162627,
      "learning_rate": 5.94284398306717e-08,
      "loss": 0.63192582,
      "num_input_tokens_seen": 331845015,
      "step": 15377,
      "time_per_iteration": 3.115239381790161
    },
    {
      "auxiliary_loss_clip": 0.01064293,
      "auxiliary_loss_mlp": 0.01035096,
      "balance_loss_clip": 1.03236473,
      "balance_loss_mlp": 1.02278185,
      "epoch": 0.9245753795280325,
      "flos": 21579260993280.0,
      "grad_norm": 2.0252668275046677,
      "language_loss": 0.73811466,
      "learning_rate": 5.933424178131341e-08,
      "loss": 0.75910854,
      "num_input_tokens_seen": 331862795,
      "step": 15378,
      "time_per_iteration": 2.529099941253662
    },
    {
      "auxiliary_loss_clip": 0.01105473,
      "auxiliary_loss_mlp": 0.01032464,
      "balance_loss_clip": 1.03586149,
      "balance_loss_mlp": 1.01966691,
      "epoch": 0.9246355027807004,
      "flos": 34496077334400.0,
      "grad_norm": 2.1979902947766026,
      "language_loss": 0.62540984,
      "learning_rate": 5.924011732219503e-08,
      "loss": 0.64678925,
      "num_input_tokens_seen": 331882535,
      "step": 15379,
      "time_per_iteration": 2.553590774536133
    },
    {
      "auxiliary_loss_clip": 0.01032486,
      "auxiliary_loss_mlp": 0.01033532,
      "balance_loss_clip": 1.03467786,
      "balance_loss_mlp": 1.02023983,
      "epoch": 0.9246956260333684,
      "flos": 15953854152960.0,
      "grad_norm": 1.9987642341465934,
      "language_loss": 0.83460689,
      "learning_rate": 5.914606645688591e-08,
      "loss": 0.85526705,
      "num_input_tokens_seen": 331899335,
      "step": 15380,
      "time_per_iteration": 2.6260673999786377
    },
    {
      "auxiliary_loss_clip": 0.01105107,
      "auxiliary_loss_mlp": 0.01034565,
      "balance_loss_clip": 1.03420901,
      "balance_loss_mlp": 1.02144551,
      "epoch": 0.9247557492860363,
      "flos": 23368366540800.0,
      "grad_norm": 1.4996767121899437,
      "language_loss": 0.73561931,
      "learning_rate": 5.905208918895233e-08,
      "loss": 0.757016,
      "num_input_tokens_seen": 331919030,
      "step": 15381,
      "time_per_iteration": 2.4527904987335205
    },
    {
      "auxiliary_loss_clip": 0.0108734,
      "auxiliary_loss_mlp": 0.01029122,
      "balance_loss_clip": 1.03580976,
      "balance_loss_mlp": 1.0175823,
      "epoch": 0.9248158725387043,
      "flos": 23039855729280.0,
      "grad_norm": 1.9854741355843133,
      "language_loss": 0.78451777,
      "learning_rate": 5.8958185521958524e-08,
      "loss": 0.80568242,
      "num_input_tokens_seen": 331936465,
      "step": 15382,
      "time_per_iteration": 3.8872482776641846
    },
    {
      "auxiliary_loss_clip": 0.01081141,
      "auxiliary_loss_mlp": 0.01034504,
      "balance_loss_clip": 1.03367949,
      "balance_loss_mlp": 1.02198708,
      "epoch": 0.9248759957913724,
      "flos": 22522418357760.0,
      "grad_norm": 1.7429046008149593,
      "language_loss": 0.74856293,
      "learning_rate": 5.886435545946455e-08,
      "loss": 0.76971942,
      "num_input_tokens_seen": 331954625,
      "step": 15383,
      "time_per_iteration": 2.540475368499756
    },
    {
      "auxiliary_loss_clip": 0.01080365,
      "auxiliary_loss_mlp": 0.01028796,
      "balance_loss_clip": 1.03276014,
      "balance_loss_mlp": 1.0173341,
      "epoch": 0.9249361190440403,
      "flos": 25447271016960.0,
      "grad_norm": 1.539443433164708,
      "language_loss": 0.75704229,
      "learning_rate": 5.8770599005028456e-08,
      "loss": 0.77813399,
      "num_input_tokens_seen": 331975865,
      "step": 15384,
      "time_per_iteration": 2.5503287315368652
    },
    {
      "auxiliary_loss_clip": 0.01068656,
      "auxiliary_loss_mlp": 0.0102914,
      "balance_loss_clip": 1.03338909,
      "balance_loss_mlp": 1.01733804,
      "epoch": 0.9249962422967083,
      "flos": 12378623886720.0,
      "grad_norm": 1.9489085690661576,
      "language_loss": 0.66393483,
      "learning_rate": 5.8676916162206045e-08,
      "loss": 0.6849128,
      "num_input_tokens_seen": 331992760,
      "step": 15385,
      "time_per_iteration": 2.4888851642608643
    },
    {
      "auxiliary_loss_clip": 0.01101545,
      "auxiliary_loss_mlp": 0.010327,
      "balance_loss_clip": 1.03393102,
      "balance_loss_mlp": 1.02093387,
      "epoch": 0.9250563655493762,
      "flos": 22929430343040.0,
      "grad_norm": 2.1098883434117752,
      "language_loss": 0.80576068,
      "learning_rate": 5.85833069345496e-08,
      "loss": 0.82710314,
      "num_input_tokens_seen": 332011890,
      "step": 15386,
      "time_per_iteration": 2.450291395187378
    },
    {
      "auxiliary_loss_clip": 0.01090815,
      "auxiliary_loss_mlp": 0.01037979,
      "balance_loss_clip": 1.03427041,
      "balance_loss_mlp": 1.02583802,
      "epoch": 0.9251164888020442,
      "flos": 18478662065280.0,
      "grad_norm": 1.83224161575174,
      "language_loss": 0.75367975,
      "learning_rate": 5.8489771325608504e-08,
      "loss": 0.77496767,
      "num_input_tokens_seen": 332029485,
      "step": 15387,
      "time_per_iteration": 2.4652528762817383
    },
    {
      "auxiliary_loss_clip": 0.01087145,
      "auxiliary_loss_mlp": 0.01030619,
      "balance_loss_clip": 1.03497815,
      "balance_loss_mlp": 1.01970494,
      "epoch": 0.9251766120547121,
      "flos": 33037062796800.0,
      "grad_norm": 1.893968863020349,
      "language_loss": 0.70011342,
      "learning_rate": 5.839630933893014e-08,
      "loss": 0.72129107,
      "num_input_tokens_seen": 332052970,
      "step": 15388,
      "time_per_iteration": 2.5983080863952637
    },
    {
      "auxiliary_loss_clip": 0.0109391,
      "auxiliary_loss_mlp": 0.01028416,
      "balance_loss_clip": 1.03446317,
      "balance_loss_mlp": 1.01644778,
      "epoch": 0.9252367353073802,
      "flos": 24387906176640.0,
      "grad_norm": 1.7681515297029704,
      "language_loss": 0.82138914,
      "learning_rate": 5.8302920978058115e-08,
      "loss": 0.84261245,
      "num_input_tokens_seen": 332070395,
      "step": 15389,
      "time_per_iteration": 2.5031449794769287
    },
    {
      "auxiliary_loss_clip": 0.01100885,
      "auxiliary_loss_mlp": 0.01035892,
      "balance_loss_clip": 1.03660607,
      "balance_loss_mlp": 1.02200437,
      "epoch": 0.9252968585600481,
      "flos": 18916844077440.0,
      "grad_norm": 1.5986838320859396,
      "language_loss": 0.79102898,
      "learning_rate": 5.820960624653381e-08,
      "loss": 0.81239676,
      "num_input_tokens_seen": 332090185,
      "step": 15390,
      "time_per_iteration": 2.4665005207061768
    },
    {
      "auxiliary_loss_clip": 0.01071374,
      "auxiliary_loss_mlp": 0.01038677,
      "balance_loss_clip": 1.03422856,
      "balance_loss_mlp": 1.02604103,
      "epoch": 0.9253569818127161,
      "flos": 21725345606400.0,
      "grad_norm": 1.976268903143008,
      "language_loss": 0.75472897,
      "learning_rate": 5.811636514789597e-08,
      "loss": 0.77582949,
      "num_input_tokens_seen": 332109050,
      "step": 15391,
      "time_per_iteration": 2.538593292236328
    },
    {
      "auxiliary_loss_clip": 0.01082204,
      "auxiliary_loss_mlp": 0.01030312,
      "balance_loss_clip": 1.03262806,
      "balance_loss_mlp": 1.01685965,
      "epoch": 0.925417105065384,
      "flos": 34240357434240.0,
      "grad_norm": 2.235252734866771,
      "language_loss": 0.51913512,
      "learning_rate": 5.80231976856802e-08,
      "loss": 0.54026026,
      "num_input_tokens_seen": 332131180,
      "step": 15392,
      "time_per_iteration": 2.6134872436523438
    },
    {
      "auxiliary_loss_clip": 0.01101894,
      "auxiliary_loss_mlp": 0.01028434,
      "balance_loss_clip": 1.03298056,
      "balance_loss_mlp": 1.01721668,
      "epoch": 0.925477228318052,
      "flos": 25959536830080.0,
      "grad_norm": 1.7592401618455358,
      "language_loss": 0.77158427,
      "learning_rate": 5.7930103863419454e-08,
      "loss": 0.79288757,
      "num_input_tokens_seen": 332149555,
      "step": 15393,
      "time_per_iteration": 2.5247156620025635
    },
    {
      "auxiliary_loss_clip": 0.01073371,
      "auxiliary_loss_mlp": 0.01030362,
      "balance_loss_clip": 1.03342104,
      "balance_loss_mlp": 1.01828599,
      "epoch": 0.9255373515707199,
      "flos": 11838240702720.0,
      "grad_norm": 1.8397738076963843,
      "language_loss": 0.69376028,
      "learning_rate": 5.783708368464357e-08,
      "loss": 0.71479762,
      "num_input_tokens_seen": 332165830,
      "step": 15394,
      "time_per_iteration": 2.483447790145874
    },
    {
      "auxiliary_loss_clip": 0.01104476,
      "auxiliary_loss_mlp": 0.01025532,
      "balance_loss_clip": 1.03545594,
      "balance_loss_mlp": 1.01377165,
      "epoch": 0.925597474823388,
      "flos": 21434325615360.0,
      "grad_norm": 1.7707871513192834,
      "language_loss": 0.73049772,
      "learning_rate": 5.7744137152879956e-08,
      "loss": 0.7517978,
      "num_input_tokens_seen": 332185130,
      "step": 15395,
      "time_per_iteration": 2.453838348388672
    },
    {
      "auxiliary_loss_clip": 0.01051122,
      "auxiliary_loss_mlp": 0.01026007,
      "balance_loss_clip": 1.0308435,
      "balance_loss_mlp": 1.01518321,
      "epoch": 0.925657598076056,
      "flos": 22857573185280.0,
      "grad_norm": 2.374298747447846,
      "language_loss": 0.72086126,
      "learning_rate": 5.7651264271653785e-08,
      "loss": 0.74163258,
      "num_input_tokens_seen": 332203695,
      "step": 15396,
      "time_per_iteration": 2.5711939334869385
    },
    {
      "auxiliary_loss_clip": 0.01102628,
      "auxiliary_loss_mlp": 0.01031426,
      "balance_loss_clip": 1.0347724,
      "balance_loss_mlp": 1.01890922,
      "epoch": 0.9257177213287239,
      "flos": 25704032411520.0,
      "grad_norm": 2.2248145382381153,
      "language_loss": 0.87380248,
      "learning_rate": 5.755846504448603e-08,
      "loss": 0.89514303,
      "num_input_tokens_seen": 332224850,
      "step": 15397,
      "time_per_iteration": 2.478447914123535
    },
    {
      "auxiliary_loss_clip": 0.01027335,
      "auxiliary_loss_mlp": 0.00999258,
      "balance_loss_clip": 1.00449681,
      "balance_loss_mlp": 0.99815524,
      "epoch": 0.9257778445813919,
      "flos": 59592933221760.0,
      "grad_norm": 0.79799632912146,
      "language_loss": 0.55143243,
      "learning_rate": 5.746573947489586e-08,
      "loss": 0.57169837,
      "num_input_tokens_seen": 332278085,
      "step": 15398,
      "time_per_iteration": 2.9308431148529053
    },
    {
      "auxiliary_loss_clip": 0.01083912,
      "auxiliary_loss_mlp": 0.01028645,
      "balance_loss_clip": 1.03382778,
      "balance_loss_mlp": 1.01469195,
      "epoch": 0.9258379678340598,
      "flos": 27709427704320.0,
      "grad_norm": 1.7544365851555825,
      "language_loss": 0.75809175,
      "learning_rate": 5.7373087566400025e-08,
      "loss": 0.77921736,
      "num_input_tokens_seen": 332297875,
      "step": 15399,
      "time_per_iteration": 2.5535495281219482
    },
    {
      "auxiliary_loss_clip": 0.01073188,
      "auxiliary_loss_mlp": 0.01029505,
      "balance_loss_clip": 1.03206468,
      "balance_loss_mlp": 1.01866865,
      "epoch": 0.9258980910867278,
      "flos": 24863543095680.0,
      "grad_norm": 1.4648693416881415,
      "language_loss": 0.78264618,
      "learning_rate": 5.7280509322510826e-08,
      "loss": 0.80367315,
      "num_input_tokens_seen": 332318500,
      "step": 15400,
      "time_per_iteration": 2.5804760456085205
    },
    {
      "auxiliary_loss_clip": 0.01018848,
      "auxiliary_loss_mlp": 0.01000165,
      "balance_loss_clip": 1.00484598,
      "balance_loss_mlp": 0.99909174,
      "epoch": 0.9259582143393957,
      "flos": 63134587249920.0,
      "grad_norm": 0.7268442966069679,
      "language_loss": 0.5132215,
      "learning_rate": 5.718800474673946e-08,
      "loss": 0.53341162,
      "num_input_tokens_seen": 332381980,
      "step": 15401,
      "time_per_iteration": 3.058939218521118
    },
    {
      "auxiliary_loss_clip": 0.01089096,
      "auxiliary_loss_mlp": 0.0103211,
      "balance_loss_clip": 1.03460217,
      "balance_loss_mlp": 1.02101719,
      "epoch": 0.9260183375920638,
      "flos": 24127122458880.0,
      "grad_norm": 1.748864649105151,
      "language_loss": 0.82514119,
      "learning_rate": 5.709557384259378e-08,
      "loss": 0.84635323,
      "num_input_tokens_seen": 332399510,
      "step": 15402,
      "time_per_iteration": 2.5010056495666504
    },
    {
      "auxiliary_loss_clip": 0.01026931,
      "auxiliary_loss_mlp": 0.01001034,
      "balance_loss_clip": 1.00418878,
      "balance_loss_mlp": 1.00005078,
      "epoch": 0.9260784608447317,
      "flos": 63042872849280.0,
      "grad_norm": 1.1426773891896398,
      "language_loss": 0.51137328,
      "learning_rate": 5.700321661357876e-08,
      "loss": 0.53165293,
      "num_input_tokens_seen": 332459130,
      "step": 15403,
      "time_per_iteration": 3.1635212898254395
    },
    {
      "auxiliary_loss_clip": 0.0100939,
      "auxiliary_loss_mlp": 0.01003052,
      "balance_loss_clip": 1.00763011,
      "balance_loss_mlp": 1.00194311,
      "epoch": 0.9261385840973997,
      "flos": 70585979927040.0,
      "grad_norm": 0.6877295273819564,
      "language_loss": 0.58755225,
      "learning_rate": 5.69109330631965e-08,
      "loss": 0.60767668,
      "num_input_tokens_seen": 332526555,
      "step": 15404,
      "time_per_iteration": 3.148221254348755
    },
    {
      "auxiliary_loss_clip": 0.01077239,
      "auxiliary_loss_mlp": 0.01031364,
      "balance_loss_clip": 1.03444159,
      "balance_loss_mlp": 1.01851916,
      "epoch": 0.9261987073500676,
      "flos": 20229917656320.0,
      "grad_norm": 2.2248226271810987,
      "language_loss": 0.7163853,
      "learning_rate": 5.681872319494596e-08,
      "loss": 0.73747134,
      "num_input_tokens_seen": 332544005,
      "step": 15405,
      "time_per_iteration": 2.4999518394470215
    },
    {
      "auxiliary_loss_clip": 0.01057918,
      "auxiliary_loss_mlp": 0.01035056,
      "balance_loss_clip": 1.03448701,
      "balance_loss_mlp": 1.02208591,
      "epoch": 0.9262588306027356,
      "flos": 20954163582720.0,
      "grad_norm": 1.6671306240265698,
      "language_loss": 0.6889565,
      "learning_rate": 5.672658701232458e-08,
      "loss": 0.70988625,
      "num_input_tokens_seen": 332563070,
      "step": 15406,
      "time_per_iteration": 2.587157964706421
    },
    {
      "auxiliary_loss_clip": 0.01053677,
      "auxiliary_loss_mlp": 0.01041128,
      "balance_loss_clip": 1.03316557,
      "balance_loss_mlp": 1.02650714,
      "epoch": 0.9263189538554035,
      "flos": 22158679282560.0,
      "grad_norm": 2.642773167199833,
      "language_loss": 0.76514912,
      "learning_rate": 5.663452451882555e-08,
      "loss": 0.78609723,
      "num_input_tokens_seen": 332579620,
      "step": 15407,
      "time_per_iteration": 2.555332660675049
    },
    {
      "auxiliary_loss_clip": 0.01072536,
      "auxiliary_loss_mlp": 0.01039903,
      "balance_loss_clip": 1.03156233,
      "balance_loss_mlp": 1.0267365,
      "epoch": 0.9263790771080715,
      "flos": 18187211111040.0,
      "grad_norm": 2.46075476997338,
      "language_loss": 0.72483408,
      "learning_rate": 5.6542535717940096e-08,
      "loss": 0.74595845,
      "num_input_tokens_seen": 332597795,
      "step": 15408,
      "time_per_iteration": 2.5495641231536865
    },
    {
      "auxiliary_loss_clip": 0.01081093,
      "auxiliary_loss_mlp": 0.01027668,
      "balance_loss_clip": 1.0345583,
      "balance_loss_mlp": 1.01729131,
      "epoch": 0.9264392003607396,
      "flos": 48178545004800.0,
      "grad_norm": 1.8021170457418827,
      "language_loss": 0.68668216,
      "learning_rate": 5.645062061315675e-08,
      "loss": 0.70776975,
      "num_input_tokens_seen": 332620375,
      "step": 15409,
      "time_per_iteration": 2.734372138977051
    },
    {
      "auxiliary_loss_clip": 0.01072199,
      "auxiliary_loss_mlp": 0.01030414,
      "balance_loss_clip": 1.03674781,
      "balance_loss_mlp": 1.01752162,
      "epoch": 0.9264993236134075,
      "flos": 26389458714240.0,
      "grad_norm": 1.9638171079634597,
      "language_loss": 0.75648332,
      "learning_rate": 5.6358779207960506e-08,
      "loss": 0.77750945,
      "num_input_tokens_seen": 332639510,
      "step": 15410,
      "time_per_iteration": 2.594776153564453
    },
    {
      "auxiliary_loss_clip": 0.01050374,
      "auxiliary_loss_mlp": 0.0102558,
      "balance_loss_clip": 1.0338167,
      "balance_loss_mlp": 1.01430857,
      "epoch": 0.9265594468660755,
      "flos": 20920084554240.0,
      "grad_norm": 1.5062688481332813,
      "language_loss": 0.82047796,
      "learning_rate": 5.6267011505833905e-08,
      "loss": 0.84123755,
      "num_input_tokens_seen": 332658350,
      "step": 15411,
      "time_per_iteration": 3.979329824447632
    },
    {
      "auxiliary_loss_clip": 0.01080009,
      "auxiliary_loss_mlp": 0.01034563,
      "balance_loss_clip": 1.03565896,
      "balance_loss_mlp": 1.02299953,
      "epoch": 0.9266195701187434,
      "flos": 17525017929600.0,
      "grad_norm": 1.8455763248760013,
      "language_loss": 0.75187069,
      "learning_rate": 5.617531751025728e-08,
      "loss": 0.77301645,
      "num_input_tokens_seen": 332676715,
      "step": 15412,
      "time_per_iteration": 3.872934103012085
    },
    {
      "auxiliary_loss_clip": 0.01100643,
      "auxiliary_loss_mlp": 0.01026743,
      "balance_loss_clip": 1.03175187,
      "balance_loss_mlp": 1.01492977,
      "epoch": 0.9266796933714114,
      "flos": 33688733293440.0,
      "grad_norm": 1.667102953710611,
      "language_loss": 0.66883147,
      "learning_rate": 5.6083697224707406e-08,
      "loss": 0.69010532,
      "num_input_tokens_seen": 332701470,
      "step": 15413,
      "time_per_iteration": 2.550652265548706
    },
    {
      "auxiliary_loss_clip": 0.01044918,
      "auxiliary_loss_mlp": 0.01035167,
      "balance_loss_clip": 1.03218114,
      "balance_loss_mlp": 1.02269769,
      "epoch": 0.9267398166240793,
      "flos": 18916520855040.0,
      "grad_norm": 1.9459438780915324,
      "language_loss": 0.75675893,
      "learning_rate": 5.5992150652658167e-08,
      "loss": 0.77755982,
      "num_input_tokens_seen": 332719060,
      "step": 15414,
      "time_per_iteration": 3.9760141372680664
    },
    {
      "auxiliary_loss_clip": 0.01087427,
      "auxiliary_loss_mlp": 0.01028905,
      "balance_loss_clip": 1.03467679,
      "balance_loss_mlp": 1.01741934,
      "epoch": 0.9267999398767474,
      "flos": 20478957626880.0,
      "grad_norm": 2.2415205927947772,
      "language_loss": 0.81663793,
      "learning_rate": 5.59006777975819e-08,
      "loss": 0.83780134,
      "num_input_tokens_seen": 332736345,
      "step": 15415,
      "time_per_iteration": 2.459085702896118
    },
    {
      "auxiliary_loss_clip": 0.01083257,
      "auxiliary_loss_mlp": 0.01032449,
      "balance_loss_clip": 1.03324342,
      "balance_loss_mlp": 1.01966345,
      "epoch": 0.9268600631294153,
      "flos": 24789351553920.0,
      "grad_norm": 1.3402473203695986,
      "language_loss": 0.54140246,
      "learning_rate": 5.580927866294671e-08,
      "loss": 0.56255949,
      "num_input_tokens_seen": 332756270,
      "step": 15416,
      "time_per_iteration": 2.5530052185058594
    },
    {
      "auxiliary_loss_clip": 0.01061331,
      "auxiliary_loss_mlp": 0.01034726,
      "balance_loss_clip": 1.03114855,
      "balance_loss_mlp": 1.02297163,
      "epoch": 0.9269201863820833,
      "flos": 18697178453760.0,
      "grad_norm": 1.7892325904070032,
      "language_loss": 0.72047389,
      "learning_rate": 5.571795325221807e-08,
      "loss": 0.7414344,
      "num_input_tokens_seen": 332775185,
      "step": 15417,
      "time_per_iteration": 2.5179495811462402
    },
    {
      "auxiliary_loss_clip": 0.0108911,
      "auxiliary_loss_mlp": 0.01029497,
      "balance_loss_clip": 1.03589368,
      "balance_loss_mlp": 1.01731396,
      "epoch": 0.9269803096347512,
      "flos": 20923999136640.0,
      "grad_norm": 2.3370336434155217,
      "language_loss": 0.75573248,
      "learning_rate": 5.5626701568859624e-08,
      "loss": 0.77691853,
      "num_input_tokens_seen": 332794320,
      "step": 15418,
      "time_per_iteration": 2.5134074687957764
    },
    {
      "auxiliary_loss_clip": 0.01090543,
      "auxiliary_loss_mlp": 0.01030936,
      "balance_loss_clip": 1.03313124,
      "balance_loss_mlp": 1.01880026,
      "epoch": 0.9270404328874192,
      "flos": 28002710252160.0,
      "grad_norm": 1.4784675252558879,
      "language_loss": 0.76028329,
      "learning_rate": 5.553552361633174e-08,
      "loss": 0.78149807,
      "num_input_tokens_seen": 332818095,
      "step": 15419,
      "time_per_iteration": 2.556217670440674
    },
    {
      "auxiliary_loss_clip": 0.01098142,
      "auxiliary_loss_mlp": 0.01031736,
      "balance_loss_clip": 1.03289294,
      "balance_loss_mlp": 1.02098346,
      "epoch": 0.9271005561400871,
      "flos": 25889870401920.0,
      "grad_norm": 1.5954255109916482,
      "language_loss": 0.75871468,
      "learning_rate": 5.5444419398091636e-08,
      "loss": 0.78001344,
      "num_input_tokens_seen": 332839860,
      "step": 15420,
      "time_per_iteration": 2.508079767227173
    },
    {
      "auxiliary_loss_clip": 0.01094105,
      "auxiliary_loss_mlp": 0.010298,
      "balance_loss_clip": 1.03391421,
      "balance_loss_mlp": 1.0175215,
      "epoch": 0.9271606793927551,
      "flos": 27053914452480.0,
      "grad_norm": 1.5176411891437707,
      "language_loss": 0.76578414,
      "learning_rate": 5.535338891759389e-08,
      "loss": 0.78702319,
      "num_input_tokens_seen": 332861155,
      "step": 15421,
      "time_per_iteration": 2.5219476222991943
    },
    {
      "auxiliary_loss_clip": 0.01077934,
      "auxiliary_loss_mlp": 0.01029806,
      "balance_loss_clip": 1.03536201,
      "balance_loss_mlp": 1.01789117,
      "epoch": 0.9272208026454232,
      "flos": 26209869690240.0,
      "grad_norm": 1.9398939344111723,
      "language_loss": 0.72424084,
      "learning_rate": 5.526243217829041e-08,
      "loss": 0.74531823,
      "num_input_tokens_seen": 332881110,
      "step": 15422,
      "time_per_iteration": 3.9531912803649902
    },
    {
      "auxiliary_loss_clip": 0.01095497,
      "auxiliary_loss_mlp": 0.01040119,
      "balance_loss_clip": 1.03453827,
      "balance_loss_mlp": 1.02727377,
      "epoch": 0.9272809258980911,
      "flos": 12458453863680.0,
      "grad_norm": 1.8877969931312746,
      "language_loss": 0.77109879,
      "learning_rate": 5.517154918363065e-08,
      "loss": 0.79245496,
      "num_input_tokens_seen": 332899350,
      "step": 15423,
      "time_per_iteration": 2.4579670429229736
    },
    {
      "auxiliary_loss_clip": 0.01094644,
      "auxiliary_loss_mlp": 0.01030365,
      "balance_loss_clip": 1.0340966,
      "balance_loss_mlp": 1.01784825,
      "epoch": 0.9273410491507591,
      "flos": 22856890826880.0,
      "grad_norm": 2.1074795708809946,
      "language_loss": 0.75507158,
      "learning_rate": 5.508073993706053e-08,
      "loss": 0.77632165,
      "num_input_tokens_seen": 332918105,
      "step": 15424,
      "time_per_iteration": 2.4743595123291016
    },
    {
      "auxiliary_loss_clip": 0.0101941,
      "auxiliary_loss_mlp": 0.01004559,
      "balance_loss_clip": 1.00589311,
      "balance_loss_mlp": 1.00348032,
      "epoch": 0.927401172403427,
      "flos": 47665384329600.0,
      "grad_norm": 0.780796460050688,
      "language_loss": 0.60669684,
      "learning_rate": 5.499000444202351e-08,
      "loss": 0.62693655,
      "num_input_tokens_seen": 332969490,
      "step": 15425,
      "time_per_iteration": 2.8726694583892822
    },
    {
      "auxiliary_loss_clip": 0.01083224,
      "auxiliary_loss_mlp": 0.00782595,
      "balance_loss_clip": 1.03589129,
      "balance_loss_mlp": 1.00881016,
      "epoch": 0.927461295656095,
      "flos": 29972374490880.0,
      "grad_norm": 1.412293674811702,
      "language_loss": 0.70652258,
      "learning_rate": 5.489934270196106e-08,
      "loss": 0.72518075,
      "num_input_tokens_seen": 332988805,
      "step": 15426,
      "time_per_iteration": 2.584205389022827
    },
    {
      "auxiliary_loss_clip": 0.01079747,
      "auxiliary_loss_mlp": 0.01024603,
      "balance_loss_clip": 1.03490734,
      "balance_loss_mlp": 1.01343298,
      "epoch": 0.9275214189087629,
      "flos": 20375427651840.0,
      "grad_norm": 1.8961350588585713,
      "language_loss": 0.83281028,
      "learning_rate": 5.480875472030977e-08,
      "loss": 0.85385382,
      "num_input_tokens_seen": 333007960,
      "step": 15427,
      "time_per_iteration": 2.5107314586639404
    },
    {
      "auxiliary_loss_clip": 0.01070489,
      "auxiliary_loss_mlp": 0.01033577,
      "balance_loss_clip": 1.03457546,
      "balance_loss_mlp": 1.02142966,
      "epoch": 0.927581542161431,
      "flos": 22383193242240.0,
      "grad_norm": 1.6277949652723425,
      "language_loss": 0.76959455,
      "learning_rate": 5.471824050050555e-08,
      "loss": 0.79063523,
      "num_input_tokens_seen": 333026035,
      "step": 15428,
      "time_per_iteration": 2.5276072025299072
    },
    {
      "auxiliary_loss_clip": 0.01063801,
      "auxiliary_loss_mlp": 0.01033194,
      "balance_loss_clip": 1.03083348,
      "balance_loss_mlp": 1.02055788,
      "epoch": 0.9276416654140989,
      "flos": 23952453598080.0,
      "grad_norm": 1.8638357917778532,
      "language_loss": 0.74536353,
      "learning_rate": 5.4627800045980555e-08,
      "loss": 0.76633352,
      "num_input_tokens_seen": 333045590,
      "step": 15429,
      "time_per_iteration": 2.5658814907073975
    },
    {
      "auxiliary_loss_clip": 0.01066215,
      "auxiliary_loss_mlp": 0.01032628,
      "balance_loss_clip": 1.03267145,
      "balance_loss_mlp": 1.02080882,
      "epoch": 0.9277017886667669,
      "flos": 13917719796480.0,
      "grad_norm": 1.8823653225345707,
      "language_loss": 0.74906266,
      "learning_rate": 5.45374333601647e-08,
      "loss": 0.77005112,
      "num_input_tokens_seen": 333063355,
      "step": 15430,
      "time_per_iteration": 2.516108989715576
    },
    {
      "auxiliary_loss_clip": 0.01092662,
      "auxiliary_loss_mlp": 0.01031704,
      "balance_loss_clip": 1.03373158,
      "balance_loss_mlp": 1.01900828,
      "epoch": 0.9277619119194348,
      "flos": 35666478092160.0,
      "grad_norm": 1.4866121812942439,
      "language_loss": 0.76504046,
      "learning_rate": 5.444714044648391e-08,
      "loss": 0.78628409,
      "num_input_tokens_seen": 333088045,
      "step": 15431,
      "time_per_iteration": 2.637157440185547
    },
    {
      "auxiliary_loss_clip": 0.01091587,
      "auxiliary_loss_mlp": 0.01028248,
      "balance_loss_clip": 1.03551066,
      "balance_loss_mlp": 1.01700068,
      "epoch": 0.9278220351721028,
      "flos": 23841238112640.0,
      "grad_norm": 2.1312729126972996,
      "language_loss": 0.70574641,
      "learning_rate": 5.4356921308363e-08,
      "loss": 0.72694474,
      "num_input_tokens_seen": 333108005,
      "step": 15432,
      "time_per_iteration": 2.4997785091400146
    },
    {
      "auxiliary_loss_clip": 0.01056614,
      "auxiliary_loss_mlp": 0.0102895,
      "balance_loss_clip": 1.0340302,
      "balance_loss_mlp": 1.01759529,
      "epoch": 0.9278821584247707,
      "flos": 15228135768960.0,
      "grad_norm": 2.2549730354773163,
      "language_loss": 0.82722044,
      "learning_rate": 5.4266775949222354e-08,
      "loss": 0.84807611,
      "num_input_tokens_seen": 333124335,
      "step": 15433,
      "time_per_iteration": 2.57077693939209
    },
    {
      "auxiliary_loss_clip": 0.01099966,
      "auxiliary_loss_mlp": 0.01028118,
      "balance_loss_clip": 1.03461647,
      "balance_loss_mlp": 1.01750207,
      "epoch": 0.9279422816774388,
      "flos": 24681404206080.0,
      "grad_norm": 1.888437172768609,
      "language_loss": 0.66904479,
      "learning_rate": 5.417670437248056e-08,
      "loss": 0.69032568,
      "num_input_tokens_seen": 333143995,
      "step": 15434,
      "time_per_iteration": 2.4776577949523926
    },
    {
      "auxiliary_loss_clip": 0.01074265,
      "auxiliary_loss_mlp": 0.01027807,
      "balance_loss_clip": 1.03336883,
      "balance_loss_mlp": 1.01657128,
      "epoch": 0.9280024049301068,
      "flos": 19169188099200.0,
      "grad_norm": 1.714811976950717,
      "language_loss": 0.68567276,
      "learning_rate": 5.40867065815529e-08,
      "loss": 0.70669347,
      "num_input_tokens_seen": 333162805,
      "step": 15435,
      "time_per_iteration": 2.538637161254883
    },
    {
      "auxiliary_loss_clip": 0.01103907,
      "auxiliary_loss_mlp": 0.01031782,
      "balance_loss_clip": 1.03479719,
      "balance_loss_mlp": 1.01937819,
      "epoch": 0.9280625281827747,
      "flos": 11393701983360.0,
      "grad_norm": 1.9685261560246479,
      "language_loss": 0.72037375,
      "learning_rate": 5.399678257985263e-08,
      "loss": 0.74173069,
      "num_input_tokens_seen": 333175770,
      "step": 15436,
      "time_per_iteration": 2.4414782524108887
    },
    {
      "auxiliary_loss_clip": 0.0107827,
      "auxiliary_loss_mlp": 0.01029742,
      "balance_loss_clip": 1.03401315,
      "balance_loss_mlp": 1.01772606,
      "epoch": 0.9281226514354427,
      "flos": 24785616539520.0,
      "grad_norm": 2.2399922615829366,
      "language_loss": 0.66954112,
      "learning_rate": 5.390693237078925e-08,
      "loss": 0.69062126,
      "num_input_tokens_seen": 333194775,
      "step": 15437,
      "time_per_iteration": 2.549983501434326
    },
    {
      "auxiliary_loss_clip": 0.01092371,
      "auxiliary_loss_mlp": 0.01032222,
      "balance_loss_clip": 1.03571486,
      "balance_loss_mlp": 1.01941943,
      "epoch": 0.9281827746881106,
      "flos": 15083128563840.0,
      "grad_norm": 2.0744151849336636,
      "language_loss": 0.71102142,
      "learning_rate": 5.3817155957770254e-08,
      "loss": 0.73226738,
      "num_input_tokens_seen": 333208920,
      "step": 15438,
      "time_per_iteration": 2.4475314617156982
    },
    {
      "auxiliary_loss_clip": 0.01104009,
      "auxiliary_loss_mlp": 0.0102987,
      "balance_loss_clip": 1.0348196,
      "balance_loss_mlp": 1.01772261,
      "epoch": 0.9282428979407786,
      "flos": 24135059364480.0,
      "grad_norm": 2.2013506338290774,
      "language_loss": 0.6444633,
      "learning_rate": 5.3727453344199366e-08,
      "loss": 0.66580206,
      "num_input_tokens_seen": 333229350,
      "step": 15439,
      "time_per_iteration": 2.49297833442688
    },
    {
      "auxiliary_loss_clip": 0.01081419,
      "auxiliary_loss_mlp": 0.01028286,
      "balance_loss_clip": 1.03455591,
      "balance_loss_mlp": 1.0162642,
      "epoch": 0.9283030211934465,
      "flos": 24823215100800.0,
      "grad_norm": 2.639818521345491,
      "language_loss": 0.70320225,
      "learning_rate": 5.363782453347876e-08,
      "loss": 0.72429931,
      "num_input_tokens_seen": 333246125,
      "step": 15440,
      "time_per_iteration": 2.5373897552490234
    },
    {
      "auxiliary_loss_clip": 0.01069291,
      "auxiliary_loss_mlp": 0.0078428,
      "balance_loss_clip": 1.03422213,
      "balance_loss_mlp": 1.00808036,
      "epoch": 0.9283631444461146,
      "flos": 23981037845760.0,
      "grad_norm": 1.605873433536233,
      "language_loss": 0.76860571,
      "learning_rate": 5.354826952900682e-08,
      "loss": 0.78714144,
      "num_input_tokens_seen": 333263685,
      "step": 15441,
      "time_per_iteration": 2.627082347869873
    },
    {
      "auxiliary_loss_clip": 0.01086824,
      "auxiliary_loss_mlp": 0.01027788,
      "balance_loss_clip": 1.03476918,
      "balance_loss_mlp": 1.01754761,
      "epoch": 0.9284232676987825,
      "flos": 22784530878720.0,
      "grad_norm": 1.6108503036482649,
      "language_loss": 0.63818848,
      "learning_rate": 5.345878833417949e-08,
      "loss": 0.6593346,
      "num_input_tokens_seen": 333282435,
      "step": 15442,
      "time_per_iteration": 2.4826526641845703
    },
    {
      "auxiliary_loss_clip": 0.01069256,
      "auxiliary_loss_mlp": 0.01039946,
      "balance_loss_clip": 1.03593886,
      "balance_loss_mlp": 1.02635622,
      "epoch": 0.9284833909514505,
      "flos": 19500500171520.0,
      "grad_norm": 2.013619196925685,
      "language_loss": 0.80664122,
      "learning_rate": 5.3369380952390295e-08,
      "loss": 0.82773322,
      "num_input_tokens_seen": 333300400,
      "step": 15443,
      "time_per_iteration": 2.527553081512451
    },
    {
      "auxiliary_loss_clip": 0.01093697,
      "auxiliary_loss_mlp": 0.0078494,
      "balance_loss_clip": 1.03472304,
      "balance_loss_mlp": 1.01227629,
      "epoch": 0.9285435142041184,
      "flos": 23185976256000.0,
      "grad_norm": 1.8270474195825444,
      "language_loss": 0.65384269,
      "learning_rate": 5.328004738702896e-08,
      "loss": 0.672629,
      "num_input_tokens_seen": 333318980,
      "step": 15444,
      "time_per_iteration": 2.491356134414673
    },
    {
      "auxiliary_loss_clip": 0.01062443,
      "auxiliary_loss_mlp": 0.01032078,
      "balance_loss_clip": 1.0350368,
      "balance_loss_mlp": 1.0201745,
      "epoch": 0.9286036374567864,
      "flos": 17675519915520.0,
      "grad_norm": 2.168998446700989,
      "language_loss": 0.72800446,
      "learning_rate": 5.3190787641483215e-08,
      "loss": 0.74894965,
      "num_input_tokens_seen": 333334135,
      "step": 15445,
      "time_per_iteration": 2.536442518234253
    },
    {
      "auxiliary_loss_clip": 0.01087802,
      "auxiliary_loss_mlp": 0.01032582,
      "balance_loss_clip": 1.03388965,
      "balance_loss_mlp": 1.01998115,
      "epoch": 0.9286637607094543,
      "flos": 20886687884160.0,
      "grad_norm": 1.9406220087185442,
      "language_loss": 0.7123338,
      "learning_rate": 5.3101601719138135e-08,
      "loss": 0.73353767,
      "num_input_tokens_seen": 333353325,
      "step": 15446,
      "time_per_iteration": 2.49086332321167
    },
    {
      "auxiliary_loss_clip": 0.01044249,
      "auxiliary_loss_mlp": 0.01028617,
      "balance_loss_clip": 1.03227544,
      "balance_loss_mlp": 1.01638567,
      "epoch": 0.9287238839621224,
      "flos": 19026012487680.0,
      "grad_norm": 1.7547306280819275,
      "language_loss": 0.69272208,
      "learning_rate": 5.301248962337523e-08,
      "loss": 0.71345073,
      "num_input_tokens_seen": 333371110,
      "step": 15447,
      "time_per_iteration": 2.5949947834014893
    },
    {
      "auxiliary_loss_clip": 0.01096189,
      "auxiliary_loss_mlp": 0.01029227,
      "balance_loss_clip": 1.03297496,
      "balance_loss_mlp": 1.01864719,
      "epoch": 0.9287840072147904,
      "flos": 20557027837440.0,
      "grad_norm": 1.8391273972542055,
      "language_loss": 0.72654808,
      "learning_rate": 5.292345135757403e-08,
      "loss": 0.74780226,
      "num_input_tokens_seen": 333391420,
      "step": 15448,
      "time_per_iteration": 2.456153154373169
    },
    {
      "auxiliary_loss_clip": 0.01102069,
      "auxiliary_loss_mlp": 0.01028006,
      "balance_loss_clip": 1.03415418,
      "balance_loss_mlp": 1.01479721,
      "epoch": 0.9288441304674583,
      "flos": 21250822008960.0,
      "grad_norm": 1.6387654177658173,
      "language_loss": 0.74063987,
      "learning_rate": 5.283448692511072e-08,
      "loss": 0.7619406,
      "num_input_tokens_seen": 333410365,
      "step": 15449,
      "time_per_iteration": 3.8281748294830322
    },
    {
      "auxiliary_loss_clip": 0.01102611,
      "auxiliary_loss_mlp": 0.00783214,
      "balance_loss_clip": 1.03379226,
      "balance_loss_mlp": 1.01020932,
      "epoch": 0.9289042537201263,
      "flos": 27669853895040.0,
      "grad_norm": 1.8330137750397721,
      "language_loss": 0.67407215,
      "learning_rate": 5.27455963293586e-08,
      "loss": 0.69293046,
      "num_input_tokens_seen": 333430000,
      "step": 15450,
      "time_per_iteration": 2.5545027256011963
    },
    {
      "auxiliary_loss_clip": 0.01072049,
      "auxiliary_loss_mlp": 0.01027195,
      "balance_loss_clip": 1.0336442,
      "balance_loss_mlp": 1.01519048,
      "epoch": 0.9289643769727942,
      "flos": 19317750750720.0,
      "grad_norm": 1.9008800172699156,
      "language_loss": 0.71660686,
      "learning_rate": 5.265677957368875e-08,
      "loss": 0.73759931,
      "num_input_tokens_seen": 333445800,
      "step": 15451,
      "time_per_iteration": 3.8965165615081787
    },
    {
      "auxiliary_loss_clip": 0.01079039,
      "auxiliary_loss_mlp": 0.01043849,
      "balance_loss_clip": 1.0326767,
      "balance_loss_mlp": 1.03046775,
      "epoch": 0.9290245002254622,
      "flos": 14058058233600.0,
      "grad_norm": 2.012711465812543,
      "language_loss": 0.73305357,
      "learning_rate": 5.25680366614687e-08,
      "loss": 0.75428247,
      "num_input_tokens_seen": 333461550,
      "step": 15452,
      "time_per_iteration": 2.467413902282715
    },
    {
      "auxiliary_loss_clip": 0.01086064,
      "auxiliary_loss_mlp": 0.01032662,
      "balance_loss_clip": 1.03803575,
      "balance_loss_mlp": 1.02028191,
      "epoch": 0.9290846234781301,
      "flos": 20047132321920.0,
      "grad_norm": 1.6463889642352727,
      "language_loss": 0.7423954,
      "learning_rate": 5.2479367596064196e-08,
      "loss": 0.76358271,
      "num_input_tokens_seen": 333478835,
      "step": 15453,
      "time_per_iteration": 3.915983200073242
    },
    {
      "auxiliary_loss_clip": 0.01000565,
      "auxiliary_loss_mlp": 0.01001041,
      "balance_loss_clip": 1.01029754,
      "balance_loss_mlp": 0.99994463,
      "epoch": 0.9291447467307982,
      "flos": 61227514460160.0,
      "grad_norm": 0.8182908789499966,
      "language_loss": 0.6071986,
      "learning_rate": 5.2390772380837226e-08,
      "loss": 0.62721467,
      "num_input_tokens_seen": 333535250,
      "step": 15454,
      "time_per_iteration": 3.05910587310791
    },
    {
      "auxiliary_loss_clip": 0.01078916,
      "auxiliary_loss_mlp": 0.01037895,
      "balance_loss_clip": 1.03232718,
      "balance_loss_mlp": 1.02448428,
      "epoch": 0.9292048699834661,
      "flos": 20553328736640.0,
      "grad_norm": 1.6828635112979977,
      "language_loss": 0.68833029,
      "learning_rate": 5.230225101914709e-08,
      "loss": 0.70949841,
      "num_input_tokens_seen": 333553805,
      "step": 15455,
      "time_per_iteration": 2.516469717025757
    },
    {
      "auxiliary_loss_clip": 0.01067571,
      "auxiliary_loss_mlp": 0.01035671,
      "balance_loss_clip": 1.03657985,
      "balance_loss_mlp": 1.0226295,
      "epoch": 0.9292649932361341,
      "flos": 23623655477760.0,
      "grad_norm": 6.6324225800736025,
      "language_loss": 0.64639515,
      "learning_rate": 5.22138035143509e-08,
      "loss": 0.66742754,
      "num_input_tokens_seen": 333572800,
      "step": 15456,
      "time_per_iteration": 2.556259870529175
    },
    {
      "auxiliary_loss_clip": 0.01059314,
      "auxiliary_loss_mlp": 0.01030241,
      "balance_loss_clip": 1.03401494,
      "balance_loss_mlp": 1.01783705,
      "epoch": 0.929325116488802,
      "flos": 15009942602880.0,
      "grad_norm": 2.0424474973249307,
      "language_loss": 0.6879319,
      "learning_rate": 5.2125429869802615e-08,
      "loss": 0.70882744,
      "num_input_tokens_seen": 333588520,
      "step": 15457,
      "time_per_iteration": 2.557140827178955
    },
    {
      "auxiliary_loss_clip": 0.01081238,
      "auxiliary_loss_mlp": 0.01025801,
      "balance_loss_clip": 1.03318048,
      "balance_loss_mlp": 1.01444006,
      "epoch": 0.92938523974147,
      "flos": 17967365919360.0,
      "grad_norm": 2.0196655821843064,
      "language_loss": 0.8068229,
      "learning_rate": 5.203713008885291e-08,
      "loss": 0.82789332,
      "num_input_tokens_seen": 333603435,
      "step": 15458,
      "time_per_iteration": 2.4694786071777344
    },
    {
      "auxiliary_loss_clip": 0.01093773,
      "auxiliary_loss_mlp": 0.01032452,
      "balance_loss_clip": 1.03412843,
      "balance_loss_mlp": 1.02045393,
      "epoch": 0.9294453629941379,
      "flos": 23003047267200.0,
      "grad_norm": 1.587371724271458,
      "language_loss": 0.72381294,
      "learning_rate": 5.194890417485065e-08,
      "loss": 0.74507523,
      "num_input_tokens_seen": 333623305,
      "step": 15459,
      "time_per_iteration": 2.5055882930755615
    },
    {
      "auxiliary_loss_clip": 0.01068301,
      "auxiliary_loss_mlp": 0.01036007,
      "balance_loss_clip": 1.03611398,
      "balance_loss_mlp": 1.02410388,
      "epoch": 0.929505486246806,
      "flos": 17055234927360.0,
      "grad_norm": 2.084138991294169,
      "language_loss": 0.58635759,
      "learning_rate": 5.1860752131141384e-08,
      "loss": 0.60740077,
      "num_input_tokens_seen": 333641205,
      "step": 15460,
      "time_per_iteration": 4.046386003494263
    },
    {
      "auxiliary_loss_clip": 0.01056133,
      "auxiliary_loss_mlp": 0.01033376,
      "balance_loss_clip": 1.03387082,
      "balance_loss_mlp": 1.02060866,
      "epoch": 0.9295656094994739,
      "flos": 27340409329920.0,
      "grad_norm": 1.7868684977279137,
      "language_loss": 0.80504006,
      "learning_rate": 5.177267396106733e-08,
      "loss": 0.82593518,
      "num_input_tokens_seen": 333659615,
      "step": 15461,
      "time_per_iteration": 2.6313486099243164
    },
    {
      "auxiliary_loss_clip": 0.0107881,
      "auxiliary_loss_mlp": 0.01025726,
      "balance_loss_clip": 1.03351128,
      "balance_loss_mlp": 1.01394796,
      "epoch": 0.9296257327521419,
      "flos": 21470954509440.0,
      "grad_norm": 1.8167243234309478,
      "language_loss": 0.78833532,
      "learning_rate": 5.168466966796869e-08,
      "loss": 0.80938065,
      "num_input_tokens_seen": 333678985,
      "step": 15462,
      "time_per_iteration": 2.513051748275757
    },
    {
      "auxiliary_loss_clip": 0.01061497,
      "auxiliary_loss_mlp": 0.01027808,
      "balance_loss_clip": 1.02985752,
      "balance_loss_mlp": 1.01539838,
      "epoch": 0.9296858560048099,
      "flos": 16362661818240.0,
      "grad_norm": 1.7952466371220337,
      "language_loss": 0.62353015,
      "learning_rate": 5.159673925518282e-08,
      "loss": 0.64442319,
      "num_input_tokens_seen": 333696410,
      "step": 15463,
      "time_per_iteration": 2.51395320892334
    },
    {
      "auxiliary_loss_clip": 0.01077921,
      "auxiliary_loss_mlp": 0.01027095,
      "balance_loss_clip": 1.030617,
      "balance_loss_mlp": 1.01553142,
      "epoch": 0.9297459792574778,
      "flos": 29858609139840.0,
      "grad_norm": 1.5000243734351142,
      "language_loss": 0.71391416,
      "learning_rate": 5.15088827260437e-08,
      "loss": 0.73496437,
      "num_input_tokens_seen": 333716615,
      "step": 15464,
      "time_per_iteration": 2.5802695751190186
    },
    {
      "auxiliary_loss_clip": 0.0107697,
      "auxiliary_loss_mlp": 0.01029987,
      "balance_loss_clip": 1.03203249,
      "balance_loss_mlp": 1.01807809,
      "epoch": 0.9298061025101458,
      "flos": 15924838942080.0,
      "grad_norm": 1.7873810120371691,
      "language_loss": 0.77466083,
      "learning_rate": 5.1421100083883115e-08,
      "loss": 0.79573029,
      "num_input_tokens_seen": 333732800,
      "step": 15465,
      "time_per_iteration": 2.472368001937866
    },
    {
      "auxiliary_loss_clip": 0.0097874,
      "auxiliary_loss_mlp": 0.01002534,
      "balance_loss_clip": 1.01079082,
      "balance_loss_mlp": 1.00137162,
      "epoch": 0.9298662257628137,
      "flos": 64096994304000.0,
      "grad_norm": 0.6938523396329327,
      "language_loss": 0.56450802,
      "learning_rate": 5.133339133202952e-08,
      "loss": 0.58432078,
      "num_input_tokens_seen": 333799300,
      "step": 15466,
      "time_per_iteration": 3.3992204666137695
    },
    {
      "auxiliary_loss_clip": 0.01080587,
      "auxiliary_loss_mlp": 0.01035302,
      "balance_loss_clip": 1.03203821,
      "balance_loss_mlp": 1.02243888,
      "epoch": 0.9299263490154818,
      "flos": 24280210224000.0,
      "grad_norm": 1.4025755472881758,
      "language_loss": 0.72720993,
      "learning_rate": 5.1245756473809355e-08,
      "loss": 0.7483688,
      "num_input_tokens_seen": 333820360,
      "step": 15467,
      "time_per_iteration": 2.661621570587158
    },
    {
      "auxiliary_loss_clip": 0.01082837,
      "auxiliary_loss_mlp": 0.01031076,
      "balance_loss_clip": 1.03506052,
      "balance_loss_mlp": 1.01864266,
      "epoch": 0.9299864722681497,
      "flos": 23294354567040.0,
      "grad_norm": 1.871569804077974,
      "language_loss": 0.72010851,
      "learning_rate": 5.1158195512545076e-08,
      "loss": 0.74124759,
      "num_input_tokens_seen": 333840415,
      "step": 15468,
      "time_per_iteration": 2.5491104125976562
    },
    {
      "auxiliary_loss_clip": 0.01090797,
      "auxiliary_loss_mlp": 0.01034396,
      "balance_loss_clip": 1.03329575,
      "balance_loss_mlp": 1.02060914,
      "epoch": 0.9300465955208177,
      "flos": 21395972868480.0,
      "grad_norm": 1.6752592265777597,
      "language_loss": 0.75599241,
      "learning_rate": 5.107070845155737e-08,
      "loss": 0.77724433,
      "num_input_tokens_seen": 333859910,
      "step": 15469,
      "time_per_iteration": 2.4881863594055176
    },
    {
      "auxiliary_loss_clip": 0.01076423,
      "auxiliary_loss_mlp": 0.01033023,
      "balance_loss_clip": 1.03464091,
      "balance_loss_mlp": 1.02101827,
      "epoch": 0.9301067187734856,
      "flos": 24571445696640.0,
      "grad_norm": 2.1607661973072356,
      "language_loss": 0.75564313,
      "learning_rate": 5.098329529416379e-08,
      "loss": 0.77673757,
      "num_input_tokens_seen": 333880495,
      "step": 15470,
      "time_per_iteration": 2.564033269882202
    },
    {
      "auxiliary_loss_clip": 0.01068945,
      "auxiliary_loss_mlp": 0.01028493,
      "balance_loss_clip": 1.03426814,
      "balance_loss_mlp": 1.01759124,
      "epoch": 0.9301668420261536,
      "flos": 22196960202240.0,
      "grad_norm": 1.4791940481755939,
      "language_loss": 0.74397671,
      "learning_rate": 5.089595604367902e-08,
      "loss": 0.76495111,
      "num_input_tokens_seen": 333897640,
      "step": 15471,
      "time_per_iteration": 2.5318474769592285
    },
    {
      "auxiliary_loss_clip": 0.01087421,
      "auxiliary_loss_mlp": 0.01028755,
      "balance_loss_clip": 1.03350401,
      "balance_loss_mlp": 1.01676214,
      "epoch": 0.9302269652788215,
      "flos": 17747628468480.0,
      "grad_norm": 2.447466065519673,
      "language_loss": 0.69845939,
      "learning_rate": 5.080869070341487e-08,
      "loss": 0.71962112,
      "num_input_tokens_seen": 333913670,
      "step": 15472,
      "time_per_iteration": 2.462728977203369
    },
    {
      "auxiliary_loss_clip": 0.01074283,
      "auxiliary_loss_mlp": 0.01032703,
      "balance_loss_clip": 1.03346121,
      "balance_loss_mlp": 1.02155113,
      "epoch": 0.9302870885314896,
      "flos": 19390793057280.0,
      "grad_norm": 1.8977869514473331,
      "language_loss": 0.88259459,
      "learning_rate": 5.0721499276680233e-08,
      "loss": 0.90366447,
      "num_input_tokens_seen": 333934105,
      "step": 15473,
      "time_per_iteration": 2.50498104095459
    },
    {
      "auxiliary_loss_clip": 0.01081147,
      "auxiliary_loss_mlp": 0.01035799,
      "balance_loss_clip": 1.03496599,
      "balance_loss_mlp": 1.02229905,
      "epoch": 0.9303472117841575,
      "flos": 21760286561280.0,
      "grad_norm": 1.9115938135971617,
      "language_loss": 0.64471453,
      "learning_rate": 5.063438176678203e-08,
      "loss": 0.66588402,
      "num_input_tokens_seen": 333953635,
      "step": 15474,
      "time_per_iteration": 2.54056453704834
    },
    {
      "auxiliary_loss_clip": 0.01103268,
      "auxiliary_loss_mlp": 0.01032651,
      "balance_loss_clip": 1.03474247,
      "balance_loss_mlp": 1.02079606,
      "epoch": 0.9304073350368255,
      "flos": 19609740408960.0,
      "grad_norm": 1.7272735630499318,
      "language_loss": 0.74395394,
      "learning_rate": 5.054733817702339e-08,
      "loss": 0.76531315,
      "num_input_tokens_seen": 333971825,
      "step": 15475,
      "time_per_iteration": 2.4207701683044434
    },
    {
      "auxiliary_loss_clip": 0.01089401,
      "auxiliary_loss_mlp": 0.01027296,
      "balance_loss_clip": 1.03282225,
      "balance_loss_mlp": 1.01604271,
      "epoch": 0.9304674582894935,
      "flos": 30441582875520.0,
      "grad_norm": 2.1947457106217882,
      "language_loss": 0.66769385,
      "learning_rate": 5.0460368510704786e-08,
      "loss": 0.68886077,
      "num_input_tokens_seen": 333990120,
      "step": 15476,
      "time_per_iteration": 2.5585567951202393
    },
    {
      "auxiliary_loss_clip": 0.01059355,
      "auxiliary_loss_mlp": 0.01030818,
      "balance_loss_clip": 1.0352695,
      "balance_loss_mlp": 1.01837826,
      "epoch": 0.9305275815421614,
      "flos": 17785693906560.0,
      "grad_norm": 1.9014785368514089,
      "language_loss": 0.68573046,
      "learning_rate": 5.0373472771124914e-08,
      "loss": 0.70663214,
      "num_input_tokens_seen": 334007970,
      "step": 15477,
      "time_per_iteration": 2.54782772064209
    },
    {
      "auxiliary_loss_clip": 0.01082786,
      "auxiliary_loss_mlp": 0.01028738,
      "balance_loss_clip": 1.03600514,
      "balance_loss_mlp": 1.01768112,
      "epoch": 0.9305877047948294,
      "flos": 25298456970240.0,
      "grad_norm": 1.7437746280391557,
      "language_loss": 0.58449262,
      "learning_rate": 5.0286650961578027e-08,
      "loss": 0.60560781,
      "num_input_tokens_seen": 334027120,
      "step": 15478,
      "time_per_iteration": 2.57336163520813
    },
    {
      "auxiliary_loss_clip": 0.0108152,
      "auxiliary_loss_mlp": 0.01032409,
      "balance_loss_clip": 1.03540134,
      "balance_loss_mlp": 1.01915288,
      "epoch": 0.9306478280474973,
      "flos": 16977236544000.0,
      "grad_norm": 1.721964097341294,
      "language_loss": 0.78596717,
      "learning_rate": 5.01999030853566e-08,
      "loss": 0.80710644,
      "num_input_tokens_seen": 334042785,
      "step": 15479,
      "time_per_iteration": 2.4898979663848877
    },
    {
      "auxiliary_loss_clip": 0.01104404,
      "auxiliary_loss_mlp": 0.0103366,
      "balance_loss_clip": 1.03537488,
      "balance_loss_mlp": 1.02212012,
      "epoch": 0.9307079513001654,
      "flos": 35663353608960.0,
      "grad_norm": 1.6376356236709904,
      "language_loss": 0.68858469,
      "learning_rate": 5.0113229145750445e-08,
      "loss": 0.70996535,
      "num_input_tokens_seen": 334063480,
      "step": 15480,
      "time_per_iteration": 2.578665256500244
    },
    {
      "auxiliary_loss_clip": 0.01103793,
      "auxiliary_loss_mlp": 0.01028821,
      "balance_loss_clip": 1.03549671,
      "balance_loss_mlp": 1.01691771,
      "epoch": 0.9307680745528333,
      "flos": 19208151377280.0,
      "grad_norm": 1.7442121070114822,
      "language_loss": 0.67852598,
      "learning_rate": 5.002662914604583e-08,
      "loss": 0.69985211,
      "num_input_tokens_seen": 334082005,
      "step": 15481,
      "time_per_iteration": 2.4239730834960938
    },
    {
      "auxiliary_loss_clip": 0.01076233,
      "auxiliary_loss_mlp": 0.01032364,
      "balance_loss_clip": 1.03126299,
      "balance_loss_mlp": 1.01970363,
      "epoch": 0.9308281978055013,
      "flos": 19062641381760.0,
      "grad_norm": 1.8920002367837583,
      "language_loss": 0.74505603,
      "learning_rate": 4.994010308952701e-08,
      "loss": 0.76614189,
      "num_input_tokens_seen": 334101375,
      "step": 15482,
      "time_per_iteration": 2.520158290863037
    },
    {
      "auxiliary_loss_clip": 0.0108987,
      "auxiliary_loss_mlp": 0.01029197,
      "balance_loss_clip": 1.03237998,
      "balance_loss_mlp": 1.01760435,
      "epoch": 0.9308883210581692,
      "flos": 20521548178560.0,
      "grad_norm": 3.120587550206097,
      "language_loss": 0.8008129,
      "learning_rate": 4.985365097947469e-08,
      "loss": 0.8220036,
      "num_input_tokens_seen": 334119460,
      "step": 15483,
      "time_per_iteration": 2.4460291862487793
    },
    {
      "auxiliary_loss_clip": 0.01078089,
      "auxiliary_loss_mlp": 0.01031062,
      "balance_loss_clip": 1.03478789,
      "balance_loss_mlp": 1.01907003,
      "epoch": 0.9309484443108372,
      "flos": 13001422826880.0,
      "grad_norm": 1.847882433549484,
      "language_loss": 0.74930096,
      "learning_rate": 4.976727281916782e-08,
      "loss": 0.77039242,
      "num_input_tokens_seen": 334136065,
      "step": 15484,
      "time_per_iteration": 2.48758602142334
    },
    {
      "auxiliary_loss_clip": 0.01083635,
      "auxiliary_loss_mlp": 0.0103302,
      "balance_loss_clip": 1.03830552,
      "balance_loss_mlp": 1.02064633,
      "epoch": 0.9310085675635051,
      "flos": 12567765928320.0,
      "grad_norm": 2.1761509454310213,
      "language_loss": 0.75995433,
      "learning_rate": 4.968096861188087e-08,
      "loss": 0.7811209,
      "num_input_tokens_seen": 334153690,
      "step": 15485,
      "time_per_iteration": 2.4905846118927
    },
    {
      "auxiliary_loss_clip": 0.01059041,
      "auxiliary_loss_mlp": 0.01031341,
      "balance_loss_clip": 1.03264129,
      "balance_loss_mlp": 1.01770914,
      "epoch": 0.9310686908161732,
      "flos": 23477570864640.0,
      "grad_norm": 1.7125932370063621,
      "language_loss": 0.78545862,
      "learning_rate": 4.959473836088723e-08,
      "loss": 0.80636239,
      "num_input_tokens_seen": 334171880,
      "step": 15486,
      "time_per_iteration": 2.5919508934020996
    },
    {
      "auxiliary_loss_clip": 0.01074176,
      "auxiliary_loss_mlp": 0.0102814,
      "balance_loss_clip": 1.03695214,
      "balance_loss_mlp": 1.0154258,
      "epoch": 0.9311288140688411,
      "flos": 24170287628160.0,
      "grad_norm": 4.411187977766473,
      "language_loss": 0.76814902,
      "learning_rate": 4.950858206945674e-08,
      "loss": 0.78917217,
      "num_input_tokens_seen": 334190005,
      "step": 15487,
      "time_per_iteration": 2.5567755699157715
    },
    {
      "auxiliary_loss_clip": 0.0107361,
      "auxiliary_loss_mlp": 0.0102644,
      "balance_loss_clip": 1.03438878,
      "balance_loss_mlp": 1.01383924,
      "epoch": 0.9311889373215091,
      "flos": 35590203561600.0,
      "grad_norm": 4.238485473268984,
      "language_loss": 0.67295229,
      "learning_rate": 4.942249974085633e-08,
      "loss": 0.6939528,
      "num_input_tokens_seen": 334209545,
      "step": 15488,
      "time_per_iteration": 4.029850006103516
    },
    {
      "auxiliary_loss_clip": 0.01078741,
      "auxiliary_loss_mlp": 0.01030396,
      "balance_loss_clip": 1.03364098,
      "balance_loss_mlp": 1.01820111,
      "epoch": 0.9312490605741771,
      "flos": 20230528187520.0,
      "grad_norm": 1.7226598339155603,
      "language_loss": 0.74971867,
      "learning_rate": 4.933649137834983e-08,
      "loss": 0.77081001,
      "num_input_tokens_seen": 334228900,
      "step": 15489,
      "time_per_iteration": 3.89424729347229
    },
    {
      "auxiliary_loss_clip": 0.0110504,
      "auxiliary_loss_mlp": 0.0102924,
      "balance_loss_clip": 1.03492439,
      "balance_loss_mlp": 1.01686597,
      "epoch": 0.931309183826845,
      "flos": 13950577762560.0,
      "grad_norm": 2.243174480892999,
      "language_loss": 0.80555081,
      "learning_rate": 4.925055698519931e-08,
      "loss": 0.82689363,
      "num_input_tokens_seen": 334245500,
      "step": 15490,
      "time_per_iteration": 2.4214470386505127
    },
    {
      "auxiliary_loss_clip": 0.01058182,
      "auxiliary_loss_mlp": 0.01031313,
      "balance_loss_clip": 1.03556705,
      "balance_loss_mlp": 1.018641,
      "epoch": 0.931369307079513,
      "flos": 20156731695360.0,
      "grad_norm": 2.4430295687596386,
      "language_loss": 0.71927524,
      "learning_rate": 4.9164696564663264e-08,
      "loss": 0.74017012,
      "num_input_tokens_seen": 334264370,
      "step": 15491,
      "time_per_iteration": 4.128579616546631
    },
    {
      "auxiliary_loss_clip": 0.01078973,
      "auxiliary_loss_mlp": 0.00783024,
      "balance_loss_clip": 1.03291965,
      "balance_loss_mlp": 1.00931311,
      "epoch": 0.931429430332181,
      "flos": 25338569483520.0,
      "grad_norm": 1.8395238186678844,
      "language_loss": 0.74404228,
      "learning_rate": 4.9078910119997096e-08,
      "loss": 0.76266223,
      "num_input_tokens_seen": 334283905,
      "step": 15492,
      "time_per_iteration": 2.598203182220459
    },
    {
      "auxiliary_loss_clip": 0.0101889,
      "auxiliary_loss_mlp": 0.00999992,
      "balance_loss_clip": 1.00499797,
      "balance_loss_mlp": 0.99881762,
      "epoch": 0.931489553584849,
      "flos": 71226193985280.0,
      "grad_norm": 0.7127484987900294,
      "language_loss": 0.53430343,
      "learning_rate": 4.899319765445442e-08,
      "loss": 0.55449224,
      "num_input_tokens_seen": 334339925,
      "step": 15493,
      "time_per_iteration": 2.9703893661499023
    },
    {
      "auxiliary_loss_clip": 0.01092448,
      "auxiliary_loss_mlp": 0.01029188,
      "balance_loss_clip": 1.03391993,
      "balance_loss_mlp": 1.01793504,
      "epoch": 0.9315496768375169,
      "flos": 14643653662080.0,
      "grad_norm": 1.6751800677997006,
      "language_loss": 0.70901692,
      "learning_rate": 4.890755917128531e-08,
      "loss": 0.73023319,
      "num_input_tokens_seen": 334357225,
      "step": 15494,
      "time_per_iteration": 2.4962096214294434
    },
    {
      "auxiliary_loss_clip": 0.01093927,
      "auxiliary_loss_mlp": 0.01027613,
      "balance_loss_clip": 1.03387678,
      "balance_loss_mlp": 1.01560235,
      "epoch": 0.9316098000901849,
      "flos": 28329928174080.0,
      "grad_norm": 1.6213232993414854,
      "language_loss": 0.68269932,
      "learning_rate": 4.882199467373671e-08,
      "loss": 0.70391464,
      "num_input_tokens_seen": 334375945,
      "step": 15495,
      "time_per_iteration": 2.5363314151763916
    },
    {
      "auxiliary_loss_clip": 0.01099868,
      "auxiliary_loss_mlp": 0.01035313,
      "balance_loss_clip": 1.0328393,
      "balance_loss_mlp": 1.02433431,
      "epoch": 0.9316699233428528,
      "flos": 28512677594880.0,
      "grad_norm": 1.915913762923675,
      "language_loss": 0.61600518,
      "learning_rate": 4.8736504165053815e-08,
      "loss": 0.637357,
      "num_input_tokens_seen": 334395310,
      "step": 15496,
      "time_per_iteration": 2.512446165084839
    },
    {
      "auxiliary_loss_clip": 0.01092425,
      "auxiliary_loss_mlp": 0.01031001,
      "balance_loss_clip": 1.0346024,
      "balance_loss_mlp": 1.01919329,
      "epoch": 0.9317300465955208,
      "flos": 33693402061440.0,
      "grad_norm": 1.5810137622274227,
      "language_loss": 0.77250952,
      "learning_rate": 4.865108764847825e-08,
      "loss": 0.79374373,
      "num_input_tokens_seen": 334416965,
      "step": 15497,
      "time_per_iteration": 2.583953857421875
    },
    {
      "auxiliary_loss_clip": 0.01096573,
      "auxiliary_loss_mlp": 0.00785881,
      "balance_loss_clip": 1.03658676,
      "balance_loss_mlp": 1.01219702,
      "epoch": 0.9317901698481887,
      "flos": 23658237296640.0,
      "grad_norm": 1.7797766571067648,
      "language_loss": 0.66823196,
      "learning_rate": 4.856574512724898e-08,
      "loss": 0.68705648,
      "num_input_tokens_seen": 334435620,
      "step": 15498,
      "time_per_iteration": 3.92992901802063
    },
    {
      "auxiliary_loss_clip": 0.0108268,
      "auxiliary_loss_mlp": 0.01037119,
      "balance_loss_clip": 1.03501105,
      "balance_loss_mlp": 1.02492964,
      "epoch": 0.9318502931008568,
      "flos": 20960017499520.0,
      "grad_norm": 1.7281008640361637,
      "language_loss": 0.79843867,
      "learning_rate": 4.8480476604602305e-08,
      "loss": 0.8196367,
      "num_input_tokens_seen": 334456210,
      "step": 15499,
      "time_per_iteration": 2.5320544242858887
    },
    {
      "auxiliary_loss_clip": 0.01060001,
      "auxiliary_loss_mlp": 0.01029998,
      "balance_loss_clip": 1.03336787,
      "balance_loss_mlp": 1.01758862,
      "epoch": 0.9319104163535247,
      "flos": 23441049711360.0,
      "grad_norm": 1.4222760746861716,
      "language_loss": 0.76916218,
      "learning_rate": 4.8395282083771196e-08,
      "loss": 0.79006219,
      "num_input_tokens_seen": 334475485,
      "step": 15500,
      "time_per_iteration": 2.5794692039489746
    },
    {
      "auxiliary_loss_clip": 0.01067975,
      "auxiliary_loss_mlp": 0.01023133,
      "balance_loss_clip": 1.03127313,
      "balance_loss_mlp": 1.01160502,
      "epoch": 0.9319705396061927,
      "flos": 22347426274560.0,
      "grad_norm": 1.9618571774084552,
      "language_loss": 0.72217214,
      "learning_rate": 4.8310161567987064e-08,
      "loss": 0.74308324,
      "num_input_tokens_seen": 334494740,
      "step": 15501,
      "time_per_iteration": 2.5605766773223877
    },
    {
      "auxiliary_loss_clip": 0.01105864,
      "auxiliary_loss_mlp": 0.01034759,
      "balance_loss_clip": 1.03542423,
      "balance_loss_mlp": 1.02249205,
      "epoch": 0.9320306628588607,
      "flos": 20993557824000.0,
      "grad_norm": 1.723885160170039,
      "language_loss": 0.6627984,
      "learning_rate": 4.822511506047666e-08,
      "loss": 0.68420464,
      "num_input_tokens_seen": 334511910,
      "step": 15502,
      "time_per_iteration": 2.418593645095825
    },
    {
      "auxiliary_loss_clip": 0.01094747,
      "auxiliary_loss_mlp": 0.00784884,
      "balance_loss_clip": 1.03504062,
      "balance_loss_mlp": 1.01189542,
      "epoch": 0.9320907861115286,
      "flos": 24538300421760.0,
      "grad_norm": 1.4093453331448773,
      "language_loss": 0.65876698,
      "learning_rate": 4.814014256446586e-08,
      "loss": 0.67756325,
      "num_input_tokens_seen": 334533150,
      "step": 15503,
      "time_per_iteration": 2.528770923614502
    },
    {
      "auxiliary_loss_clip": 0.01067428,
      "auxiliary_loss_mlp": 0.01036553,
      "balance_loss_clip": 1.03158188,
      "balance_loss_mlp": 1.0236609,
      "epoch": 0.9321509093641966,
      "flos": 19785414850560.0,
      "grad_norm": 1.5662289965736342,
      "language_loss": 0.75053215,
      "learning_rate": 4.805524408317652e-08,
      "loss": 0.77157199,
      "num_input_tokens_seen": 334550940,
      "step": 15504,
      "time_per_iteration": 2.557810068130493
    },
    {
      "auxiliary_loss_clip": 0.01094515,
      "auxiliary_loss_mlp": 0.00784607,
      "balance_loss_clip": 1.03543746,
      "balance_loss_mlp": 1.01000094,
      "epoch": 0.9322110326168646,
      "flos": 24972675592320.0,
      "grad_norm": 5.979121210175421,
      "language_loss": 0.71249986,
      "learning_rate": 4.797041961982762e-08,
      "loss": 0.73129106,
      "num_input_tokens_seen": 334570935,
      "step": 15505,
      "time_per_iteration": 2.5425987243652344
    },
    {
      "auxiliary_loss_clip": 0.01083514,
      "auxiliary_loss_mlp": 0.01029966,
      "balance_loss_clip": 1.03379118,
      "balance_loss_mlp": 1.01734793,
      "epoch": 0.9322711558695326,
      "flos": 16143642639360.0,
      "grad_norm": 1.8739041335278073,
      "language_loss": 0.75406373,
      "learning_rate": 4.788566917763614e-08,
      "loss": 0.77519858,
      "num_input_tokens_seen": 334589315,
      "step": 15506,
      "time_per_iteration": 2.4860713481903076
    },
    {
      "auxiliary_loss_clip": 0.01064047,
      "auxiliary_loss_mlp": 0.01027809,
      "balance_loss_clip": 1.03282475,
      "balance_loss_mlp": 1.01637733,
      "epoch": 0.9323312791222005,
      "flos": 23732428838400.0,
      "grad_norm": 1.8351445126661334,
      "language_loss": 0.83017945,
      "learning_rate": 4.780099275981597e-08,
      "loss": 0.851098,
      "num_input_tokens_seen": 334608990,
      "step": 15507,
      "time_per_iteration": 2.5753467082977295
    },
    {
      "auxiliary_loss_clip": 0.0110473,
      "auxiliary_loss_mlp": 0.01031561,
      "balance_loss_clip": 1.03496099,
      "balance_loss_mlp": 1.01975965,
      "epoch": 0.9323914023748685,
      "flos": 20777914523520.0,
      "grad_norm": 1.4564449451306887,
      "language_loss": 0.67833197,
      "learning_rate": 4.771639036957742e-08,
      "loss": 0.69969487,
      "num_input_tokens_seen": 334628655,
      "step": 15508,
      "time_per_iteration": 2.4422402381896973
    },
    {
      "auxiliary_loss_clip": 0.01068983,
      "auxiliary_loss_mlp": 0.01031051,
      "balance_loss_clip": 1.03563571,
      "balance_loss_mlp": 1.01908231,
      "epoch": 0.9324515256275364,
      "flos": 23915178259200.0,
      "grad_norm": 1.628027514705778,
      "language_loss": 0.71919239,
      "learning_rate": 4.7631862010129033e-08,
      "loss": 0.74019277,
      "num_input_tokens_seen": 334648295,
      "step": 15509,
      "time_per_iteration": 2.5762388706207275
    },
    {
      "auxiliary_loss_clip": 0.01092222,
      "auxiliary_loss_mlp": 0.010324,
      "balance_loss_clip": 1.03387463,
      "balance_loss_mlp": 1.02036572,
      "epoch": 0.9325116488802044,
      "flos": 18005215875840.0,
      "grad_norm": 2.101325209076579,
      "language_loss": 0.74551237,
      "learning_rate": 4.754740768467624e-08,
      "loss": 0.76675862,
      "num_input_tokens_seen": 334666280,
      "step": 15510,
      "time_per_iteration": 2.4660046100616455
    },
    {
      "auxiliary_loss_clip": 0.01095551,
      "auxiliary_loss_mlp": 0.01028349,
      "balance_loss_clip": 1.03310657,
      "balance_loss_mlp": 1.01618373,
      "epoch": 0.9325717721328723,
      "flos": 29021603443200.0,
      "grad_norm": 1.5051994410292724,
      "language_loss": 0.70422053,
      "learning_rate": 4.746302739642161e-08,
      "loss": 0.72545958,
      "num_input_tokens_seen": 334688830,
      "step": 15511,
      "time_per_iteration": 2.553065538406372
    },
    {
      "auxiliary_loss_clip": 0.01077232,
      "auxiliary_loss_mlp": 0.0103773,
      "balance_loss_clip": 1.03350592,
      "balance_loss_mlp": 1.02527213,
      "epoch": 0.9326318953855404,
      "flos": 21646341642240.0,
      "grad_norm": 1.784709660094255,
      "language_loss": 0.78034186,
      "learning_rate": 4.737872114856412e-08,
      "loss": 0.8014915,
      "num_input_tokens_seen": 334705205,
      "step": 15512,
      "time_per_iteration": 2.506049156188965
    },
    {
      "auxiliary_loss_clip": 0.01101532,
      "auxiliary_loss_mlp": 0.0102778,
      "balance_loss_clip": 1.03356934,
      "balance_loss_mlp": 1.01538277,
      "epoch": 0.9326920186382083,
      "flos": 26065724411520.0,
      "grad_norm": 1.426125854159295,
      "language_loss": 0.80328441,
      "learning_rate": 4.7294488944301436e-08,
      "loss": 0.82457757,
      "num_input_tokens_seen": 334723830,
      "step": 15513,
      "time_per_iteration": 2.5029296875
    },
    {
      "auxiliary_loss_clip": 0.01089119,
      "auxiliary_loss_mlp": 0.01031735,
      "balance_loss_clip": 1.03771067,
      "balance_loss_mlp": 1.01856267,
      "epoch": 0.9327521418908763,
      "flos": 12057116227200.0,
      "grad_norm": 2.006099880763913,
      "language_loss": 0.80142713,
      "learning_rate": 4.721033078682768e-08,
      "loss": 0.82263571,
      "num_input_tokens_seen": 334740825,
      "step": 15514,
      "time_per_iteration": 2.4727063179016113
    },
    {
      "auxiliary_loss_clip": 0.01077555,
      "auxiliary_loss_mlp": 0.01037006,
      "balance_loss_clip": 1.0365026,
      "balance_loss_mlp": 1.02522779,
      "epoch": 0.9328122651435443,
      "flos": 43834395271680.0,
      "grad_norm": 1.7564535612918173,
      "language_loss": 0.71562952,
      "learning_rate": 4.7126246679333626e-08,
      "loss": 0.7367751,
      "num_input_tokens_seen": 334765825,
      "step": 15515,
      "time_per_iteration": 2.7149147987365723
    },
    {
      "auxiliary_loss_clip": 0.01086746,
      "auxiliary_loss_mlp": 0.01033054,
      "balance_loss_clip": 1.03483808,
      "balance_loss_mlp": 1.02004194,
      "epoch": 0.9328723883962122,
      "flos": 15194954580480.0,
      "grad_norm": 2.444788436432688,
      "language_loss": 0.80735946,
      "learning_rate": 4.704223662500806e-08,
      "loss": 0.82855749,
      "num_input_tokens_seen": 334782680,
      "step": 15516,
      "time_per_iteration": 2.471937417984009
    },
    {
      "auxiliary_loss_clip": 0.01067184,
      "auxiliary_loss_mlp": 0.01038659,
      "balance_loss_clip": 1.03284955,
      "balance_loss_mlp": 1.02529597,
      "epoch": 0.9329325116488802,
      "flos": 20261770041600.0,
      "grad_norm": 1.655901220347938,
      "language_loss": 0.8057375,
      "learning_rate": 4.695830062703643e-08,
      "loss": 0.82679594,
      "num_input_tokens_seen": 334800160,
      "step": 15517,
      "time_per_iteration": 2.5508930683135986
    },
    {
      "auxiliary_loss_clip": 0.01084051,
      "auxiliary_loss_mlp": 0.01030453,
      "balance_loss_clip": 1.03460526,
      "balance_loss_mlp": 1.01822209,
      "epoch": 0.9329926349015482,
      "flos": 13115008609920.0,
      "grad_norm": 1.882374073254307,
      "language_loss": 0.7450012,
      "learning_rate": 4.687443868860219e-08,
      "loss": 0.76614618,
      "num_input_tokens_seen": 334815840,
      "step": 15518,
      "time_per_iteration": 2.4868078231811523
    },
    {
      "auxiliary_loss_clip": 0.01080262,
      "auxiliary_loss_mlp": 0.01031199,
      "balance_loss_clip": 1.03400517,
      "balance_loss_mlp": 1.01880074,
      "epoch": 0.9330527581542162,
      "flos": 23040250778880.0,
      "grad_norm": 2.0900213429569017,
      "language_loss": 0.75615287,
      "learning_rate": 4.679065081288458e-08,
      "loss": 0.77726746,
      "num_input_tokens_seen": 334834735,
      "step": 15519,
      "time_per_iteration": 2.52270245552063
    },
    {
      "auxiliary_loss_clip": 0.01049017,
      "auxiliary_loss_mlp": 0.01033817,
      "balance_loss_clip": 1.03215909,
      "balance_loss_mlp": 1.02108002,
      "epoch": 0.9331128814068841,
      "flos": 15559627409280.0,
      "grad_norm": 2.0270770459292233,
      "language_loss": 0.83084309,
      "learning_rate": 4.6706937003061275e-08,
      "loss": 0.85167146,
      "num_input_tokens_seen": 334853490,
      "step": 15520,
      "time_per_iteration": 2.604471445083618
    },
    {
      "auxiliary_loss_clip": 0.01089939,
      "auxiliary_loss_mlp": 0.01029013,
      "balance_loss_clip": 1.03244627,
      "balance_loss_mlp": 1.01708031,
      "epoch": 0.9331730046595521,
      "flos": 22271762275200.0,
      "grad_norm": 1.6068848894157632,
      "language_loss": 0.7662074,
      "learning_rate": 4.6623297262306846e-08,
      "loss": 0.78739685,
      "num_input_tokens_seen": 334873675,
      "step": 15521,
      "time_per_iteration": 2.4885141849517822
    },
    {
      "auxiliary_loss_clip": 0.010953,
      "auxiliary_loss_mlp": 0.01030948,
      "balance_loss_clip": 1.03707552,
      "balance_loss_mlp": 1.01940811,
      "epoch": 0.93323312791222,
      "flos": 15777641007360.0,
      "grad_norm": 1.7177867917623644,
      "language_loss": 0.77848732,
      "learning_rate": 4.6539731593792545e-08,
      "loss": 0.79974985,
      "num_input_tokens_seen": 334890970,
      "step": 15522,
      "time_per_iteration": 2.453705072402954
    },
    {
      "auxiliary_loss_clip": 0.01068803,
      "auxiliary_loss_mlp": 0.00783069,
      "balance_loss_clip": 1.03230977,
      "balance_loss_mlp": 1.01034081,
      "epoch": 0.933293251164888,
      "flos": 22010978557440.0,
      "grad_norm": 2.6446528829272298,
      "language_loss": 0.62708914,
      "learning_rate": 4.6456240000687373e-08,
      "loss": 0.64560783,
      "num_input_tokens_seen": 334906635,
      "step": 15523,
      "time_per_iteration": 2.528109312057495
    },
    {
      "auxiliary_loss_clip": 0.01081344,
      "auxiliary_loss_mlp": 0.01032727,
      "balance_loss_clip": 1.03426123,
      "balance_loss_mlp": 1.02078843,
      "epoch": 0.933353374417556,
      "flos": 26031358074240.0,
      "grad_norm": 1.5647328304117871,
      "language_loss": 0.68109632,
      "learning_rate": 4.63728224861577e-08,
      "loss": 0.70223713,
      "num_input_tokens_seen": 334926230,
      "step": 15524,
      "time_per_iteration": 2.532000780105591
    },
    {
      "auxiliary_loss_clip": 0.01059983,
      "auxiliary_loss_mlp": 0.01034468,
      "balance_loss_clip": 1.03170598,
      "balance_loss_mlp": 1.02185619,
      "epoch": 0.933413497670224,
      "flos": 24900100162560.0,
      "grad_norm": 1.7122214993255171,
      "language_loss": 0.73888987,
      "learning_rate": 4.628947905336589e-08,
      "loss": 0.75983441,
      "num_input_tokens_seen": 334946680,
      "step": 15525,
      "time_per_iteration": 2.6029868125915527
    },
    {
      "auxiliary_loss_clip": 0.01049303,
      "auxiliary_loss_mlp": 0.01038059,
      "balance_loss_clip": 1.03119624,
      "balance_loss_mlp": 1.02566123,
      "epoch": 0.9334736209228919,
      "flos": 23688689051520.0,
      "grad_norm": 1.7692827987530468,
      "language_loss": 0.83849341,
      "learning_rate": 4.6206209705473175e-08,
      "loss": 0.85936701,
      "num_input_tokens_seen": 334964785,
      "step": 15526,
      "time_per_iteration": 2.58610200881958
    },
    {
      "auxiliary_loss_clip": 0.01059531,
      "auxiliary_loss_mlp": 0.01029231,
      "balance_loss_clip": 1.03460693,
      "balance_loss_mlp": 1.01676214,
      "epoch": 0.9335337441755599,
      "flos": 15377344865280.0,
      "grad_norm": 1.875832778413,
      "language_loss": 0.68893361,
      "learning_rate": 4.61230144456366e-08,
      "loss": 0.70982134,
      "num_input_tokens_seen": 334982400,
      "step": 15527,
      "time_per_iteration": 3.934974193572998
    },
    {
      "auxiliary_loss_clip": 0.01106639,
      "auxiliary_loss_mlp": 0.01029971,
      "balance_loss_clip": 1.03570569,
      "balance_loss_mlp": 1.01630414,
      "epoch": 0.9335938674282279,
      "flos": 16106726436480.0,
      "grad_norm": 1.9028159455912952,
      "language_loss": 0.64884359,
      "learning_rate": 4.603989327701141e-08,
      "loss": 0.67020965,
      "num_input_tokens_seen": 334999685,
      "step": 15528,
      "time_per_iteration": 3.8269989490509033
    },
    {
      "auxiliary_loss_clip": 0.01104768,
      "auxiliary_loss_mlp": 0.01033533,
      "balance_loss_clip": 1.03431773,
      "balance_loss_mlp": 1.02067065,
      "epoch": 0.9336539906808958,
      "flos": 18952898353920.0,
      "grad_norm": 1.7922427129233063,
      "language_loss": 0.74562997,
      "learning_rate": 4.5956846202748867e-08,
      "loss": 0.76701295,
      "num_input_tokens_seen": 335019160,
      "step": 15529,
      "time_per_iteration": 3.8532514572143555
    },
    {
      "auxiliary_loss_clip": 0.01056321,
      "auxiliary_loss_mlp": 0.01029233,
      "balance_loss_clip": 1.03073692,
      "balance_loss_mlp": 1.0174669,
      "epoch": 0.9337141139335638,
      "flos": 18109104986880.0,
      "grad_norm": 1.708738296913162,
      "language_loss": 0.63120186,
      "learning_rate": 4.5873873225998674e-08,
      "loss": 0.65205741,
      "num_input_tokens_seen": 335037350,
      "step": 15530,
      "time_per_iteration": 2.5583834648132324
    },
    {
      "auxiliary_loss_clip": 0.01079763,
      "auxiliary_loss_mlp": 0.01028003,
      "balance_loss_clip": 1.03394532,
      "balance_loss_mlp": 1.01630306,
      "epoch": 0.9337742371862318,
      "flos": 17345716214400.0,
      "grad_norm": 2.3717172724104514,
      "language_loss": 0.7228055,
      "learning_rate": 4.5790974349907194e-08,
      "loss": 0.74388319,
      "num_input_tokens_seen": 335056060,
      "step": 15531,
      "time_per_iteration": 2.5294857025146484
    },
    {
      "auxiliary_loss_clip": 0.0108084,
      "auxiliary_loss_mlp": 0.01033565,
      "balance_loss_clip": 1.03242373,
      "balance_loss_mlp": 1.02117956,
      "epoch": 0.9338343604388998,
      "flos": 29058986522880.0,
      "grad_norm": 1.6849369807760675,
      "language_loss": 0.70789474,
      "learning_rate": 4.5708149577617925e-08,
      "loss": 0.72903883,
      "num_input_tokens_seen": 335075410,
      "step": 15532,
      "time_per_iteration": 2.558807611465454
    },
    {
      "auxiliary_loss_clip": 0.0110487,
      "auxiliary_loss_mlp": 0.00782104,
      "balance_loss_clip": 1.03457427,
      "balance_loss_mlp": 1.00758767,
      "epoch": 0.9338944836915677,
      "flos": 18660908695680.0,
      "grad_norm": 1.781644393979094,
      "language_loss": 0.72965479,
      "learning_rate": 4.5625398912271016e-08,
      "loss": 0.74852455,
      "num_input_tokens_seen": 335095190,
      "step": 15533,
      "time_per_iteration": 2.5111985206604004
    },
    {
      "auxiliary_loss_clip": 0.01069211,
      "auxiliary_loss_mlp": 0.01025536,
      "balance_loss_clip": 1.03304183,
      "balance_loss_mlp": 1.01387715,
      "epoch": 0.9339546069442357,
      "flos": 16617735273600.0,
      "grad_norm": 1.8724124058007001,
      "language_loss": 0.79493463,
      "learning_rate": 4.554272235700507e-08,
      "loss": 0.81588209,
      "num_input_tokens_seen": 335113825,
      "step": 15534,
      "time_per_iteration": 2.5196497440338135
    },
    {
      "auxiliary_loss_clip": 0.01098209,
      "auxiliary_loss_mlp": 0.01025898,
      "balance_loss_clip": 1.03526783,
      "balance_loss_mlp": 1.01547861,
      "epoch": 0.9340147301969036,
      "flos": 23693106424320.0,
      "grad_norm": 1.6618576278559234,
      "language_loss": 0.74430954,
      "learning_rate": 4.546011991495513e-08,
      "loss": 0.76555061,
      "num_input_tokens_seen": 335136425,
      "step": 15535,
      "time_per_iteration": 2.4999678134918213
    },
    {
      "auxiliary_loss_clip": 0.01089417,
      "auxiliary_loss_mlp": 0.01029686,
      "balance_loss_clip": 1.03605938,
      "balance_loss_mlp": 1.01718712,
      "epoch": 0.9340748534495716,
      "flos": 28654452576000.0,
      "grad_norm": 1.8240551766841684,
      "language_loss": 0.77616763,
      "learning_rate": 4.537759158925292e-08,
      "loss": 0.79735869,
      "num_input_tokens_seen": 335157925,
      "step": 15536,
      "time_per_iteration": 2.5602850914001465
    },
    {
      "auxiliary_loss_clip": 0.01072675,
      "auxiliary_loss_mlp": 0.01029546,
      "balance_loss_clip": 1.03314936,
      "balance_loss_mlp": 1.01780415,
      "epoch": 0.9341349767022396,
      "flos": 24899633285760.0,
      "grad_norm": 1.4149907381786824,
      "language_loss": 0.80741596,
      "learning_rate": 4.5295137383028593e-08,
      "loss": 0.82843822,
      "num_input_tokens_seen": 335177840,
      "step": 15537,
      "time_per_iteration": 4.004602909088135
    },
    {
      "auxiliary_loss_clip": 0.01084839,
      "auxiliary_loss_mlp": 0.01029367,
      "balance_loss_clip": 1.03578043,
      "balance_loss_mlp": 1.01755953,
      "epoch": 0.9341950999549076,
      "flos": 29059525226880.0,
      "grad_norm": 1.6164588827924442,
      "language_loss": 0.77901196,
      "learning_rate": 4.5212757299408764e-08,
      "loss": 0.80015403,
      "num_input_tokens_seen": 335199470,
      "step": 15538,
      "time_per_iteration": 2.5617246627807617
    },
    {
      "auxiliary_loss_clip": 0.01076753,
      "auxiliary_loss_mlp": 0.01030461,
      "balance_loss_clip": 1.03457642,
      "balance_loss_mlp": 1.01811719,
      "epoch": 0.9342552232075755,
      "flos": 23587062497280.0,
      "grad_norm": 1.464415386756092,
      "language_loss": 0.73173964,
      "learning_rate": 4.513045134151672e-08,
      "loss": 0.75281173,
      "num_input_tokens_seen": 335218885,
      "step": 15539,
      "time_per_iteration": 2.52237606048584
    },
    {
      "auxiliary_loss_clip": 0.01064023,
      "auxiliary_loss_mlp": 0.01028145,
      "balance_loss_clip": 1.03527284,
      "balance_loss_mlp": 1.01752329,
      "epoch": 0.9343153464602435,
      "flos": 36721389646080.0,
      "grad_norm": 1.4420601480670017,
      "language_loss": 0.6473943,
      "learning_rate": 4.504821951247373e-08,
      "loss": 0.66831601,
      "num_input_tokens_seen": 335239485,
      "step": 15540,
      "time_per_iteration": 2.6876113414764404
    },
    {
      "auxiliary_loss_clip": 0.01089866,
      "auxiliary_loss_mlp": 0.01029142,
      "balance_loss_clip": 1.0324533,
      "balance_loss_mlp": 1.01794207,
      "epoch": 0.9343754697129115,
      "flos": 22236498097920.0,
      "grad_norm": 1.6175961403978396,
      "language_loss": 0.76332635,
      "learning_rate": 4.496606181539864e-08,
      "loss": 0.78451639,
      "num_input_tokens_seen": 335258355,
      "step": 15541,
      "time_per_iteration": 2.4866137504577637
    },
    {
      "auxiliary_loss_clip": 0.01089748,
      "auxiliary_loss_mlp": 0.01032383,
      "balance_loss_clip": 1.0361805,
      "balance_loss_mlp": 1.02078378,
      "epoch": 0.9344355929655794,
      "flos": 29710333797120.0,
      "grad_norm": 1.9783038229993095,
      "language_loss": 0.67218268,
      "learning_rate": 4.4883978253406066e-08,
      "loss": 0.69340396,
      "num_input_tokens_seen": 335276835,
      "step": 15542,
      "time_per_iteration": 2.5345969200134277
    },
    {
      "auxiliary_loss_clip": 0.01059497,
      "auxiliary_loss_mlp": 0.01028836,
      "balance_loss_clip": 1.03218055,
      "balance_loss_mlp": 1.01622951,
      "epoch": 0.9344957162182475,
      "flos": 18880394751360.0,
      "grad_norm": 1.94642339565805,
      "language_loss": 0.69938195,
      "learning_rate": 4.480196882960907e-08,
      "loss": 0.72026533,
      "num_input_tokens_seen": 335296220,
      "step": 15543,
      "time_per_iteration": 2.519374132156372
    },
    {
      "auxiliary_loss_clip": 0.01094024,
      "auxiliary_loss_mlp": 0.01030898,
      "balance_loss_clip": 1.03284502,
      "balance_loss_mlp": 1.01753438,
      "epoch": 0.9345558394709154,
      "flos": 27417761268480.0,
      "grad_norm": 1.91313169106958,
      "language_loss": 0.69267917,
      "learning_rate": 4.4720033547117394e-08,
      "loss": 0.7139284,
      "num_input_tokens_seen": 335316335,
      "step": 15544,
      "time_per_iteration": 2.538534164428711
    },
    {
      "auxiliary_loss_clip": 0.0109039,
      "auxiliary_loss_mlp": 0.01041504,
      "balance_loss_clip": 1.03402114,
      "balance_loss_mlp": 1.02778256,
      "epoch": 0.9346159627235834,
      "flos": 20741285629440.0,
      "grad_norm": 2.189177398127204,
      "language_loss": 0.77239031,
      "learning_rate": 4.463817240903789e-08,
      "loss": 0.79370922,
      "num_input_tokens_seen": 335335545,
      "step": 15545,
      "time_per_iteration": 2.4885804653167725
    },
    {
      "auxiliary_loss_clip": 0.01095634,
      "auxiliary_loss_mlp": 0.01029275,
      "balance_loss_clip": 1.03609443,
      "balance_loss_mlp": 1.01806295,
      "epoch": 0.9346760859762513,
      "flos": 21069221823360.0,
      "grad_norm": 1.652150752444054,
      "language_loss": 0.68995297,
      "learning_rate": 4.455638541847495e-08,
      "loss": 0.71120203,
      "num_input_tokens_seen": 335355350,
      "step": 15546,
      "time_per_iteration": 2.4931247234344482
    },
    {
      "auxiliary_loss_clip": 0.01063522,
      "auxiliary_loss_mlp": 0.01026802,
      "balance_loss_clip": 1.03285861,
      "balance_loss_mlp": 1.01471972,
      "epoch": 0.9347362092289193,
      "flos": 29204927481600.0,
      "grad_norm": 1.8019676126535786,
      "language_loss": 0.82041991,
      "learning_rate": 4.447467257852966e-08,
      "loss": 0.84132314,
      "num_input_tokens_seen": 335375160,
      "step": 15547,
      "time_per_iteration": 2.5903806686401367
    },
    {
      "auxiliary_loss_clip": 0.01084825,
      "auxiliary_loss_mlp": 0.01039253,
      "balance_loss_clip": 1.0312103,
      "balance_loss_mlp": 1.02658105,
      "epoch": 0.9347963324815872,
      "flos": 19427350124160.0,
      "grad_norm": 2.0246855894143576,
      "language_loss": 0.83730298,
      "learning_rate": 4.439303389230087e-08,
      "loss": 0.85854375,
      "num_input_tokens_seen": 335394080,
      "step": 15548,
      "time_per_iteration": 2.480454206466675
    },
    {
      "auxiliary_loss_clip": 0.01096427,
      "auxiliary_loss_mlp": 0.01036696,
      "balance_loss_clip": 1.03522193,
      "balance_loss_mlp": 1.02341628,
      "epoch": 0.9348564557342552,
      "flos": 36901840596480.0,
      "grad_norm": 2.7748504347477128,
      "language_loss": 0.66122162,
      "learning_rate": 4.4311469362884326e-08,
      "loss": 0.68255281,
      "num_input_tokens_seen": 335414230,
      "step": 15549,
      "time_per_iteration": 2.591153860092163
    },
    {
      "auxiliary_loss_clip": 0.01094557,
      "auxiliary_loss_mlp": 0.01031655,
      "balance_loss_clip": 1.03590751,
      "balance_loss_mlp": 1.01928735,
      "epoch": 0.9349165789869232,
      "flos": 21690117342720.0,
      "grad_norm": 1.8182911724930118,
      "language_loss": 0.80341744,
      "learning_rate": 4.4229978993372665e-08,
      "loss": 0.82467961,
      "num_input_tokens_seen": 335432890,
      "step": 15550,
      "time_per_iteration": 2.5077593326568604
    },
    {
      "auxiliary_loss_clip": 0.01091646,
      "auxiliary_loss_mlp": 0.01031104,
      "balance_loss_clip": 1.03806794,
      "balance_loss_mlp": 1.01960075,
      "epoch": 0.9349767022395912,
      "flos": 18844053166080.0,
      "grad_norm": 1.6116817716121221,
      "language_loss": 0.75547755,
      "learning_rate": 4.4148562786856524e-08,
      "loss": 0.77670503,
      "num_input_tokens_seen": 335452085,
      "step": 15551,
      "time_per_iteration": 2.4805312156677246
    },
    {
      "auxiliary_loss_clip": 0.01045854,
      "auxiliary_loss_mlp": 0.01029819,
      "balance_loss_clip": 1.03102386,
      "balance_loss_mlp": 1.01910162,
      "epoch": 0.9350368254922591,
      "flos": 24973429777920.0,
      "grad_norm": 1.6605431583562031,
      "language_loss": 0.73739576,
      "learning_rate": 4.406722074642255e-08,
      "loss": 0.75815248,
      "num_input_tokens_seen": 335472130,
      "step": 15552,
      "time_per_iteration": 2.655322313308716
    },
    {
      "auxiliary_loss_clip": 0.01054589,
      "auxiliary_loss_mlp": 0.01042113,
      "balance_loss_clip": 1.03078151,
      "balance_loss_mlp": 1.02942336,
      "epoch": 0.9350969487449271,
      "flos": 23070594792960.0,
      "grad_norm": 1.6286016735036362,
      "language_loss": 0.77349615,
      "learning_rate": 4.3985952875155386e-08,
      "loss": 0.79446322,
      "num_input_tokens_seen": 335489970,
      "step": 15553,
      "time_per_iteration": 2.590271472930908
    },
    {
      "auxiliary_loss_clip": 0.01068957,
      "auxiliary_loss_mlp": 0.010369,
      "balance_loss_clip": 1.03269649,
      "balance_loss_mlp": 1.02382874,
      "epoch": 0.9351570719975951,
      "flos": 18625177641600.0,
      "grad_norm": 1.5220401182866605,
      "language_loss": 0.78318226,
      "learning_rate": 4.390475917613723e-08,
      "loss": 0.80424082,
      "num_input_tokens_seen": 335509125,
      "step": 15554,
      "time_per_iteration": 2.526153326034546
    },
    {
      "auxiliary_loss_clip": 0.01073275,
      "auxiliary_loss_mlp": 0.01031662,
      "balance_loss_clip": 1.03099811,
      "balance_loss_mlp": 1.0203793,
      "epoch": 0.935217195250263,
      "flos": 15888353702400.0,
      "grad_norm": 1.5095076937663103,
      "language_loss": 0.69315624,
      "learning_rate": 4.382363965244695e-08,
      "loss": 0.71420562,
      "num_input_tokens_seen": 335525620,
      "step": 15555,
      "time_per_iteration": 2.4846901893615723
    },
    {
      "auxiliary_loss_clip": 0.01019482,
      "auxiliary_loss_mlp": 0.0104166,
      "balance_loss_clip": 1.0323509,
      "balance_loss_mlp": 1.02808857,
      "epoch": 0.935277318502931,
      "flos": 24390312387840.0,
      "grad_norm": 1.5254084767921123,
      "language_loss": 0.75487661,
      "learning_rate": 4.374259430715965e-08,
      "loss": 0.77548802,
      "num_input_tokens_seen": 335547565,
      "step": 15556,
      "time_per_iteration": 2.8624703884124756
    },
    {
      "auxiliary_loss_clip": 0.0108075,
      "auxiliary_loss_mlp": 0.0103452,
      "balance_loss_clip": 1.03262925,
      "balance_loss_mlp": 1.02333868,
      "epoch": 0.935337441755599,
      "flos": 27600259294080.0,
      "grad_norm": 1.5015612989669962,
      "language_loss": 0.72217929,
      "learning_rate": 4.366162314334953e-08,
      "loss": 0.74333203,
      "num_input_tokens_seen": 335570285,
      "step": 15557,
      "time_per_iteration": 2.8201675415039062
    },
    {
      "auxiliary_loss_clip": 0.01103263,
      "auxiliary_loss_mlp": 0.01030304,
      "balance_loss_clip": 1.0346359,
      "balance_loss_mlp": 1.01807284,
      "epoch": 0.935397565008267,
      "flos": 20482872209280.0,
      "grad_norm": 1.460000314000897,
      "language_loss": 0.62918186,
      "learning_rate": 4.358072616408681e-08,
      "loss": 0.65051758,
      "num_input_tokens_seen": 335588600,
      "step": 15558,
      "time_per_iteration": 2.45874285697937
    },
    {
      "auxiliary_loss_clip": 0.01080334,
      "auxiliary_loss_mlp": 0.01029579,
      "balance_loss_clip": 1.03442693,
      "balance_loss_mlp": 1.01679432,
      "epoch": 0.9354576882609349,
      "flos": 23654394541440.0,
      "grad_norm": 2.4403208962736453,
      "language_loss": 0.73357213,
      "learning_rate": 4.34999033724388e-08,
      "loss": 0.75467128,
      "num_input_tokens_seen": 335606235,
      "step": 15559,
      "time_per_iteration": 2.522536516189575
    },
    {
      "auxiliary_loss_clip": 0.01056541,
      "auxiliary_loss_mlp": 0.00782434,
      "balance_loss_clip": 1.03214812,
      "balance_loss_mlp": 1.00869298,
      "epoch": 0.9355178115136029,
      "flos": 36684904406400.0,
      "grad_norm": 1.46086109139203,
      "language_loss": 0.63338178,
      "learning_rate": 4.341915477147062e-08,
      "loss": 0.65177155,
      "num_input_tokens_seen": 335628240,
      "step": 15560,
      "time_per_iteration": 2.7130584716796875
    },
    {
      "auxiliary_loss_clip": 0.01042093,
      "auxiliary_loss_mlp": 0.01032193,
      "balance_loss_clip": 1.0341264,
      "balance_loss_mlp": 1.01759017,
      "epoch": 0.9355779347662708,
      "flos": 14460401450880.0,
      "grad_norm": 2.133332639169153,
      "language_loss": 0.64167798,
      "learning_rate": 4.3338480364244034e-08,
      "loss": 0.66242075,
      "num_input_tokens_seen": 335643755,
      "step": 15561,
      "time_per_iteration": 2.589303970336914
    },
    {
      "auxiliary_loss_clip": 0.01104794,
      "auxiliary_loss_mlp": 0.01033045,
      "balance_loss_clip": 1.03702986,
      "balance_loss_mlp": 1.02069521,
      "epoch": 0.9356380580189388,
      "flos": 23185976256000.0,
      "grad_norm": 1.6974251297329317,
      "language_loss": 0.75538802,
      "learning_rate": 4.325788015381859e-08,
      "loss": 0.77676642,
      "num_input_tokens_seen": 335665160,
      "step": 15562,
      "time_per_iteration": 2.472651481628418
    },
    {
      "auxiliary_loss_clip": 0.01016931,
      "auxiliary_loss_mlp": 0.01003866,
      "balance_loss_clip": 1.00592875,
      "balance_loss_mlp": 1.00278091,
      "epoch": 0.9356981812716068,
      "flos": 67471626090240.0,
      "grad_norm": 0.9552031710714227,
      "language_loss": 0.62341452,
      "learning_rate": 4.31773541432503e-08,
      "loss": 0.64362246,
      "num_input_tokens_seen": 335715240,
      "step": 15563,
      "time_per_iteration": 2.935854196548462
    },
    {
      "auxiliary_loss_clip": 0.01060154,
      "auxiliary_loss_mlp": 0.01034437,
      "balance_loss_clip": 1.03418505,
      "balance_loss_mlp": 1.02275467,
      "epoch": 0.9357583045242748,
      "flos": 24681619687680.0,
      "grad_norm": 1.6745331742670497,
      "language_loss": 0.78207445,
      "learning_rate": 4.3096902335592714e-08,
      "loss": 0.80302036,
      "num_input_tokens_seen": 335734970,
      "step": 15564,
      "time_per_iteration": 2.5964019298553467
    },
    {
      "auxiliary_loss_clip": 0.01103939,
      "auxiliary_loss_mlp": 0.01031002,
      "balance_loss_clip": 1.03330445,
      "balance_loss_mlp": 1.0177815,
      "epoch": 0.9358184277769427,
      "flos": 19463727623040.0,
      "grad_norm": 2.3134140605200773,
      "language_loss": 0.78271449,
      "learning_rate": 4.301652473389694e-08,
      "loss": 0.80406392,
      "num_input_tokens_seen": 335753435,
      "step": 15565,
      "time_per_iteration": 3.804476737976074
    },
    {
      "auxiliary_loss_clip": 0.0108939,
      "auxiliary_loss_mlp": 0.01029355,
      "balance_loss_clip": 1.03325403,
      "balance_loss_mlp": 1.01795244,
      "epoch": 0.9358785510296107,
      "flos": 18916987731840.0,
      "grad_norm": 2.6296625264031213,
      "language_loss": 0.71882188,
      "learning_rate": 4.2936221341210774e-08,
      "loss": 0.74000937,
      "num_input_tokens_seen": 335772105,
      "step": 15566,
      "time_per_iteration": 4.06824517250061
    },
    {
      "auxiliary_loss_clip": 0.01064182,
      "auxiliary_loss_mlp": 0.00785293,
      "balance_loss_clip": 1.03070736,
      "balance_loss_mlp": 1.01090121,
      "epoch": 0.9359386742822787,
      "flos": 23441265192960.0,
      "grad_norm": 1.7528182682493063,
      "language_loss": 0.6767031,
      "learning_rate": 4.285599216057889e-08,
      "loss": 0.69519782,
      "num_input_tokens_seen": 335789125,
      "step": 15567,
      "time_per_iteration": 4.095608711242676
    },
    {
      "auxiliary_loss_clip": 0.01073379,
      "auxiliary_loss_mlp": 0.01034356,
      "balance_loss_clip": 1.03403223,
      "balance_loss_mlp": 1.02251279,
      "epoch": 0.9359987975349466,
      "flos": 32744067557760.0,
      "grad_norm": 2.134274737954368,
      "language_loss": 0.61629212,
      "learning_rate": 4.277583719504418e-08,
      "loss": 0.63736951,
      "num_input_tokens_seen": 335810995,
      "step": 15568,
      "time_per_iteration": 2.5983786582946777
    },
    {
      "auxiliary_loss_clip": 0.01078251,
      "auxiliary_loss_mlp": 0.0103339,
      "balance_loss_clip": 1.03091824,
      "balance_loss_mlp": 1.02136779,
      "epoch": 0.9360589207876147,
      "flos": 22819651401600.0,
      "grad_norm": 1.6324050566663537,
      "language_loss": 0.78777957,
      "learning_rate": 4.269575644764556e-08,
      "loss": 0.80889595,
      "num_input_tokens_seen": 335830580,
      "step": 15569,
      "time_per_iteration": 2.5157172679901123
    },
    {
      "auxiliary_loss_clip": 0.01085042,
      "auxiliary_loss_mlp": 0.01034701,
      "balance_loss_clip": 1.03498125,
      "balance_loss_mlp": 1.02233887,
      "epoch": 0.9361190440402826,
      "flos": 20885251340160.0,
      "grad_norm": 2.5920239322832543,
      "language_loss": 0.69530004,
      "learning_rate": 4.261574992142014e-08,
      "loss": 0.71649748,
      "num_input_tokens_seen": 335846515,
      "step": 15570,
      "time_per_iteration": 2.5494847297668457
    },
    {
      "auxiliary_loss_clip": 0.01087736,
      "auxiliary_loss_mlp": 0.01028237,
      "balance_loss_clip": 1.03464067,
      "balance_loss_mlp": 1.01598179,
      "epoch": 0.9361791672929506,
      "flos": 19317822577920.0,
      "grad_norm": 1.8946242780044513,
      "language_loss": 0.78422999,
      "learning_rate": 4.2535817619401726e-08,
      "loss": 0.8053897,
      "num_input_tokens_seen": 335863350,
      "step": 15571,
      "time_per_iteration": 2.4553892612457275
    },
    {
      "auxiliary_loss_clip": 0.01070586,
      "auxiliary_loss_mlp": 0.01029648,
      "balance_loss_clip": 1.03260922,
      "balance_loss_mlp": 1.01757193,
      "epoch": 0.9362392905456185,
      "flos": 15158182032000.0,
      "grad_norm": 1.9634167178483677,
      "language_loss": 0.77774912,
      "learning_rate": 4.2455959544621224e-08,
      "loss": 0.79875147,
      "num_input_tokens_seen": 335880510,
      "step": 15572,
      "time_per_iteration": 2.519888162612915
    },
    {
      "auxiliary_loss_clip": 0.01078793,
      "auxiliary_loss_mlp": 0.01036048,
      "balance_loss_clip": 1.03271425,
      "balance_loss_mlp": 1.02400136,
      "epoch": 0.9362994137982865,
      "flos": 22085888371200.0,
      "grad_norm": 1.7111176677663573,
      "language_loss": 0.77909601,
      "learning_rate": 4.237617570010688e-08,
      "loss": 0.80024445,
      "num_input_tokens_seen": 335899440,
      "step": 15573,
      "time_per_iteration": 2.4984281063079834
    },
    {
      "auxiliary_loss_clip": 0.01066835,
      "auxiliary_loss_mlp": 0.01025429,
      "balance_loss_clip": 1.03148341,
      "balance_loss_mlp": 1.01382375,
      "epoch": 0.9363595370509544,
      "flos": 23512260424320.0,
      "grad_norm": 2.3201639120050372,
      "language_loss": 0.74448657,
      "learning_rate": 4.2296466088884044e-08,
      "loss": 0.76540917,
      "num_input_tokens_seen": 335919540,
      "step": 15574,
      "time_per_iteration": 2.5509278774261475
    },
    {
      "auxiliary_loss_clip": 0.01045349,
      "auxiliary_loss_mlp": 0.01033013,
      "balance_loss_clip": 1.03185749,
      "balance_loss_mlp": 1.02069902,
      "epoch": 0.9364196603036224,
      "flos": 27123473139840.0,
      "grad_norm": 1.6931168754083834,
      "language_loss": 0.68090749,
      "learning_rate": 4.221683071397564e-08,
      "loss": 0.70169109,
      "num_input_tokens_seen": 335939665,
      "step": 15575,
      "time_per_iteration": 2.638585090637207
    },
    {
      "auxiliary_loss_clip": 0.01076513,
      "auxiliary_loss_mlp": 0.0103204,
      "balance_loss_clip": 1.03376579,
      "balance_loss_mlp": 1.0197252,
      "epoch": 0.9364797835562904,
      "flos": 18479057114880.0,
      "grad_norm": 1.4742520459973771,
      "language_loss": 0.6518563,
      "learning_rate": 4.2137269578401026e-08,
      "loss": 0.67294186,
      "num_input_tokens_seen": 335958580,
      "step": 15576,
      "time_per_iteration": 3.856295585632324
    },
    {
      "auxiliary_loss_clip": 0.01089143,
      "auxiliary_loss_mlp": 0.01027494,
      "balance_loss_clip": 1.03183329,
      "balance_loss_mlp": 1.01395166,
      "epoch": 0.9365399068089584,
      "flos": 13005552890880.0,
      "grad_norm": 2.50532690552505,
      "language_loss": 0.75380242,
      "learning_rate": 4.2057782685177566e-08,
      "loss": 0.77496874,
      "num_input_tokens_seen": 335974965,
      "step": 15577,
      "time_per_iteration": 2.450474262237549
    },
    {
      "auxiliary_loss_clip": 0.01057653,
      "auxiliary_loss_mlp": 0.01029094,
      "balance_loss_clip": 1.03279018,
      "balance_loss_mlp": 1.01652944,
      "epoch": 0.9366000300616263,
      "flos": 25666433850240.0,
      "grad_norm": 2.0486315553052363,
      "language_loss": 0.52466697,
      "learning_rate": 4.1978370037318855e-08,
      "loss": 0.54553443,
      "num_input_tokens_seen": 335996575,
      "step": 15578,
      "time_per_iteration": 2.6473519802093506
    },
    {
      "auxiliary_loss_clip": 0.01041831,
      "auxiliary_loss_mlp": 0.01033192,
      "balance_loss_clip": 1.03035831,
      "balance_loss_mlp": 1.02124071,
      "epoch": 0.9366601533142943,
      "flos": 21433355948160.0,
      "grad_norm": 1.5269573121483597,
      "language_loss": 0.70395857,
      "learning_rate": 4.189903163783692e-08,
      "loss": 0.7247088,
      "num_input_tokens_seen": 336017265,
      "step": 15579,
      "time_per_iteration": 2.6275346279144287
    },
    {
      "auxiliary_loss_clip": 0.01077894,
      "auxiliary_loss_mlp": 0.01029288,
      "balance_loss_clip": 1.0327673,
      "balance_loss_mlp": 1.01780772,
      "epoch": 0.9367202765669622,
      "flos": 24093222998400.0,
      "grad_norm": 2.0541864902112277,
      "language_loss": 0.76324558,
      "learning_rate": 4.181976748973959e-08,
      "loss": 0.78431743,
      "num_input_tokens_seen": 336035905,
      "step": 15580,
      "time_per_iteration": 2.5222973823547363
    },
    {
      "auxiliary_loss_clip": 0.01093841,
      "auxiliary_loss_mlp": 0.01032612,
      "balance_loss_clip": 1.03552032,
      "balance_loss_mlp": 1.01954651,
      "epoch": 0.9367803998196302,
      "flos": 20888842700160.0,
      "grad_norm": 2.210633979288992,
      "language_loss": 0.66365004,
      "learning_rate": 4.1740577596033114e-08,
      "loss": 0.68491459,
      "num_input_tokens_seen": 336055585,
      "step": 15581,
      "time_per_iteration": 2.524878978729248
    },
    {
      "auxiliary_loss_clip": 0.01093317,
      "auxiliary_loss_mlp": 0.01029997,
      "balance_loss_clip": 1.03524947,
      "balance_loss_mlp": 1.01787889,
      "epoch": 0.9368405230722983,
      "flos": 22564362464640.0,
      "grad_norm": 1.5826395223481204,
      "language_loss": 0.7667371,
      "learning_rate": 4.166146195972042e-08,
      "loss": 0.78797024,
      "num_input_tokens_seen": 336076695,
      "step": 15582,
      "time_per_iteration": 2.5147645473480225
    },
    {
      "auxiliary_loss_clip": 0.01029426,
      "auxiliary_loss_mlp": 0.01037392,
      "balance_loss_clip": 1.03165472,
      "balance_loss_mlp": 1.02479732,
      "epoch": 0.9369006463249662,
      "flos": 18880215183360.0,
      "grad_norm": 1.636056406694439,
      "language_loss": 0.7390942,
      "learning_rate": 4.1582420583800905e-08,
      "loss": 0.75976235,
      "num_input_tokens_seen": 336094740,
      "step": 15583,
      "time_per_iteration": 2.635038137435913
    },
    {
      "auxiliary_loss_clip": 0.01108425,
      "auxiliary_loss_mlp": 0.01033488,
      "balance_loss_clip": 1.03626192,
      "balance_loss_mlp": 1.0206666,
      "epoch": 0.9369607695776342,
      "flos": 26432516142720.0,
      "grad_norm": 2.093083238641521,
      "language_loss": 0.83827704,
      "learning_rate": 4.1503453471272376e-08,
      "loss": 0.85969615,
      "num_input_tokens_seen": 336113985,
      "step": 15584,
      "time_per_iteration": 2.47880482673645
    },
    {
      "auxiliary_loss_clip": 0.01093328,
      "auxiliary_loss_mlp": 0.00783589,
      "balance_loss_clip": 1.03562856,
      "balance_loss_mlp": 1.00883293,
      "epoch": 0.9370208928303021,
      "flos": 39567346081920.0,
      "grad_norm": 1.6149113125621728,
      "language_loss": 0.71973538,
      "learning_rate": 4.1424560625129334e-08,
      "loss": 0.73850453,
      "num_input_tokens_seen": 336136395,
      "step": 15585,
      "time_per_iteration": 2.6569066047668457
    },
    {
      "auxiliary_loss_clip": 0.01067436,
      "auxiliary_loss_mlp": 0.01022702,
      "balance_loss_clip": 1.03264523,
      "balance_loss_mlp": 1.01185369,
      "epoch": 0.9370810160829701,
      "flos": 22963114321920.0,
      "grad_norm": 1.6934746855788905,
      "language_loss": 0.8074494,
      "learning_rate": 4.134574204836316e-08,
      "loss": 0.82835078,
      "num_input_tokens_seen": 336156345,
      "step": 15586,
      "time_per_iteration": 2.5395936965942383
    },
    {
      "auxiliary_loss_clip": 0.01063909,
      "auxiliary_loss_mlp": 0.01030558,
      "balance_loss_clip": 1.03444815,
      "balance_loss_mlp": 1.01866055,
      "epoch": 0.937141139335638,
      "flos": 23075048079360.0,
      "grad_norm": 1.5983802360998167,
      "language_loss": 0.76377451,
      "learning_rate": 4.126699774396258e-08,
      "loss": 0.78471917,
      "num_input_tokens_seen": 336176760,
      "step": 15587,
      "time_per_iteration": 2.564521312713623
    },
    {
      "auxiliary_loss_clip": 0.01085079,
      "auxiliary_loss_mlp": 0.01030034,
      "balance_loss_clip": 1.03389096,
      "balance_loss_mlp": 1.01797009,
      "epoch": 0.937201262588306,
      "flos": 16356664247040.0,
      "grad_norm": 3.479769378787924,
      "language_loss": 0.87373734,
      "learning_rate": 4.118832771491387e-08,
      "loss": 0.89488846,
      "num_input_tokens_seen": 336193285,
      "step": 15588,
      "time_per_iteration": 2.469794273376465
    },
    {
      "auxiliary_loss_clip": 0.01100903,
      "auxiliary_loss_mlp": 0.00782856,
      "balance_loss_clip": 1.03514826,
      "balance_loss_mlp": 1.00962949,
      "epoch": 0.937261385840974,
      "flos": 20194078861440.0,
      "grad_norm": 1.7070276042960943,
      "language_loss": 0.78147912,
      "learning_rate": 4.11097319642002e-08,
      "loss": 0.80031675,
      "num_input_tokens_seen": 336211425,
      "step": 15589,
      "time_per_iteration": 2.4685447216033936
    },
    {
      "auxiliary_loss_clip": 0.01100138,
      "auxiliary_loss_mlp": 0.01032665,
      "balance_loss_clip": 1.03385937,
      "balance_loss_mlp": 1.02057099,
      "epoch": 0.937321509093642,
      "flos": 18295948558080.0,
      "grad_norm": 1.8492168798496216,
      "language_loss": 0.77646977,
      "learning_rate": 4.103121049480163e-08,
      "loss": 0.7977978,
      "num_input_tokens_seen": 336230205,
      "step": 15590,
      "time_per_iteration": 2.415260076522827
    },
    {
      "auxiliary_loss_clip": 0.0108255,
      "auxiliary_loss_mlp": 0.01037709,
      "balance_loss_clip": 1.03454554,
      "balance_loss_mlp": 1.02401829,
      "epoch": 0.9373816323463099,
      "flos": 25884662929920.0,
      "grad_norm": 1.6067893068379164,
      "language_loss": 0.71460617,
      "learning_rate": 4.095276330969577e-08,
      "loss": 0.73580885,
      "num_input_tokens_seen": 336252440,
      "step": 15591,
      "time_per_iteration": 2.6125824451446533
    },
    {
      "auxiliary_loss_clip": 0.01092444,
      "auxiliary_loss_mlp": 0.00784827,
      "balance_loss_clip": 1.0356257,
      "balance_loss_mlp": 1.01027405,
      "epoch": 0.9374417555989779,
      "flos": 27198849830400.0,
      "grad_norm": 1.728193854274124,
      "language_loss": 0.53464353,
      "learning_rate": 4.0874390411857804e-08,
      "loss": 0.55341625,
      "num_input_tokens_seen": 336273845,
      "step": 15592,
      "time_per_iteration": 2.5215208530426025
    },
    {
      "auxiliary_loss_clip": 0.01084928,
      "auxiliary_loss_mlp": 0.01026814,
      "balance_loss_clip": 1.03476214,
      "balance_loss_mlp": 1.01547706,
      "epoch": 0.9375018788516458,
      "flos": 23621249266560.0,
      "grad_norm": 1.4332291918339004,
      "language_loss": 0.67273021,
      "learning_rate": 4.0796091804259136e-08,
      "loss": 0.69384766,
      "num_input_tokens_seen": 336292790,
      "step": 15593,
      "time_per_iteration": 2.5007576942443848
    },
    {
      "auxiliary_loss_clip": 0.01082639,
      "auxiliary_loss_mlp": 0.01029732,
      "balance_loss_clip": 1.03341949,
      "balance_loss_mlp": 1.01822233,
      "epoch": 0.9375620021043138,
      "flos": 22678774260480.0,
      "grad_norm": 1.5229958666692947,
      "language_loss": 0.73963177,
      "learning_rate": 4.0717867489868715e-08,
      "loss": 0.76075542,
      "num_input_tokens_seen": 336312600,
      "step": 15594,
      "time_per_iteration": 2.5383472442626953
    },
    {
      "auxiliary_loss_clip": 0.01090107,
      "auxiliary_loss_mlp": 0.01024909,
      "balance_loss_clip": 1.03364706,
      "balance_loss_mlp": 1.01426947,
      "epoch": 0.9376221253569819,
      "flos": 27560254521600.0,
      "grad_norm": 1.540696949240773,
      "language_loss": 0.73769128,
      "learning_rate": 4.063971747165351e-08,
      "loss": 0.7588414,
      "num_input_tokens_seen": 336332770,
      "step": 15595,
      "time_per_iteration": 2.5188863277435303
    },
    {
      "auxiliary_loss_clip": 0.01083526,
      "auxiliary_loss_mlp": 0.01026718,
      "balance_loss_clip": 1.0346204,
      "balance_loss_mlp": 1.01549411,
      "epoch": 0.9376822486096498,
      "flos": 24129887806080.0,
      "grad_norm": 1.8153901089531408,
      "language_loss": 0.76134443,
      "learning_rate": 4.056164175257626e-08,
      "loss": 0.78244692,
      "num_input_tokens_seen": 336351445,
      "step": 15596,
      "time_per_iteration": 2.5518603324890137
    },
    {
      "auxiliary_loss_clip": 0.01082727,
      "auxiliary_loss_mlp": 0.0103022,
      "balance_loss_clip": 1.0348959,
      "balance_loss_mlp": 1.01885891,
      "epoch": 0.9377423718623178,
      "flos": 22784028088320.0,
      "grad_norm": 1.6664075391431528,
      "language_loss": 0.78612888,
      "learning_rate": 4.0483640335597926e-08,
      "loss": 0.80725831,
      "num_input_tokens_seen": 336368690,
      "step": 15597,
      "time_per_iteration": 2.501997709274292
    },
    {
      "auxiliary_loss_clip": 0.01107368,
      "auxiliary_loss_mlp": 0.01032175,
      "balance_loss_clip": 1.03545916,
      "balance_loss_mlp": 1.01970577,
      "epoch": 0.9378024951149857,
      "flos": 19168900790400.0,
      "grad_norm": 1.4889264215040359,
      "language_loss": 0.80901223,
      "learning_rate": 4.0405713223676363e-08,
      "loss": 0.83040768,
      "num_input_tokens_seen": 336388165,
      "step": 15598,
      "time_per_iteration": 2.436891555786133
    },
    {
      "auxiliary_loss_clip": 0.0107232,
      "auxiliary_loss_mlp": 0.01030294,
      "balance_loss_clip": 1.03459692,
      "balance_loss_mlp": 1.01823568,
      "epoch": 0.9378626183676537,
      "flos": 23505508667520.0,
      "grad_norm": 1.9486662852124939,
      "language_loss": 0.63348681,
      "learning_rate": 4.0327860419766994e-08,
      "loss": 0.65451288,
      "num_input_tokens_seen": 336406475,
      "step": 15599,
      "time_per_iteration": 2.559999942779541
    },
    {
      "auxiliary_loss_clip": 0.01063273,
      "auxiliary_loss_mlp": 0.01031962,
      "balance_loss_clip": 1.03381431,
      "balance_loss_mlp": 1.01942766,
      "epoch": 0.9379227416203216,
      "flos": 18405655672320.0,
      "grad_norm": 1.7076943677960992,
      "language_loss": 0.73744512,
      "learning_rate": 4.0250081926821e-08,
      "loss": 0.75839752,
      "num_input_tokens_seen": 336424690,
      "step": 15600,
      "time_per_iteration": 2.5022454261779785
    },
    {
      "auxiliary_loss_clip": 0.01077019,
      "auxiliary_loss_mlp": 0.01028879,
      "balance_loss_clip": 1.03521895,
      "balance_loss_mlp": 1.01824522,
      "epoch": 0.9379828648729897,
      "flos": 17821855923840.0,
      "grad_norm": 1.7756664405819595,
      "language_loss": 0.69114965,
      "learning_rate": 4.0172377747788474e-08,
      "loss": 0.71220863,
      "num_input_tokens_seen": 336443055,
      "step": 15601,
      "time_per_iteration": 2.5188684463500977
    },
    {
      "auxiliary_loss_clip": 0.01019439,
      "auxiliary_loss_mlp": 0.01002242,
      "balance_loss_clip": 1.00602126,
      "balance_loss_mlp": 1.00109184,
      "epoch": 0.9380429881256576,
      "flos": 68024399466240.0,
      "grad_norm": 0.7597375324088025,
      "language_loss": 0.58120269,
      "learning_rate": 4.009474788561573e-08,
      "loss": 0.60141945,
      "num_input_tokens_seen": 336510190,
      "step": 15602,
      "time_per_iteration": 3.2495241165161133
    },
    {
      "auxiliary_loss_clip": 0.01036408,
      "auxiliary_loss_mlp": 0.0103356,
      "balance_loss_clip": 1.03351712,
      "balance_loss_mlp": 1.02125168,
      "epoch": 0.9381031113783256,
      "flos": 20776980769920.0,
      "grad_norm": 2.487079707929017,
      "language_loss": 0.72237426,
      "learning_rate": 4.001719234324663e-08,
      "loss": 0.74307394,
      "num_input_tokens_seen": 336529250,
      "step": 15603,
      "time_per_iteration": 2.6305859088897705
    },
    {
      "auxiliary_loss_clip": 0.01095784,
      "auxiliary_loss_mlp": 0.01030128,
      "balance_loss_clip": 1.03249979,
      "balance_loss_mlp": 1.0191251,
      "epoch": 0.9381632346309935,
      "flos": 19025078734080.0,
      "grad_norm": 1.5535400399216845,
      "language_loss": 0.76047754,
      "learning_rate": 3.993971112362171e-08,
      "loss": 0.78173667,
      "num_input_tokens_seen": 336548530,
      "step": 15604,
      "time_per_iteration": 3.824373960494995
    },
    {
      "auxiliary_loss_clip": 0.01079626,
      "auxiliary_loss_mlp": 0.01035794,
      "balance_loss_clip": 1.03313971,
      "balance_loss_mlp": 1.0214169,
      "epoch": 0.9382233578836615,
      "flos": 23513840622720.0,
      "grad_norm": 1.8067637887083097,
      "language_loss": 0.65374434,
      "learning_rate": 3.9862304229679734e-08,
      "loss": 0.67489856,
      "num_input_tokens_seen": 336568510,
      "step": 15605,
      "time_per_iteration": 3.927949905395508
    },
    {
      "auxiliary_loss_clip": 0.01071536,
      "auxiliary_loss_mlp": 0.00783174,
      "balance_loss_clip": 1.03460872,
      "balance_loss_mlp": 1.0070734,
      "epoch": 0.9382834811363294,
      "flos": 43067882016000.0,
      "grad_norm": 1.808691527457172,
      "language_loss": 0.67094612,
      "learning_rate": 3.9784971664355683e-08,
      "loss": 0.68949318,
      "num_input_tokens_seen": 336592020,
      "step": 15606,
      "time_per_iteration": 4.134126424789429
    },
    {
      "auxiliary_loss_clip": 0.01087296,
      "auxiliary_loss_mlp": 0.01026419,
      "balance_loss_clip": 1.03257799,
      "balance_loss_mlp": 1.01525533,
      "epoch": 0.9383436043889974,
      "flos": 16436242828800.0,
      "grad_norm": 1.68669680893608,
      "language_loss": 0.77246177,
      "learning_rate": 3.970771343058166e-08,
      "loss": 0.79359895,
      "num_input_tokens_seen": 336610010,
      "step": 15607,
      "time_per_iteration": 2.4889976978302
    },
    {
      "auxiliary_loss_clip": 0.01094256,
      "auxiliary_loss_mlp": 0.01028124,
      "balance_loss_clip": 1.03522444,
      "balance_loss_mlp": 1.01676977,
      "epoch": 0.9384037276416655,
      "flos": 20740603271040.0,
      "grad_norm": 1.8814036347555276,
      "language_loss": 0.82669014,
      "learning_rate": 3.963052953128776e-08,
      "loss": 0.84791392,
      "num_input_tokens_seen": 336628520,
      "step": 15608,
      "time_per_iteration": 2.4793779850006104
    },
    {
      "auxiliary_loss_clip": 0.01094746,
      "auxiliary_loss_mlp": 0.01033632,
      "balance_loss_clip": 1.03708315,
      "balance_loss_mlp": 1.02135372,
      "epoch": 0.9384638508943334,
      "flos": 19062677295360.0,
      "grad_norm": 1.6309027347178864,
      "language_loss": 0.68701613,
      "learning_rate": 3.9553419969400536e-08,
      "loss": 0.70829993,
      "num_input_tokens_seen": 336647365,
      "step": 15609,
      "time_per_iteration": 2.4664385318756104
    },
    {
      "auxiliary_loss_clip": 0.0107365,
      "auxiliary_loss_mlp": 0.01029406,
      "balance_loss_clip": 1.03344452,
      "balance_loss_mlp": 1.01662111,
      "epoch": 0.9385239741470014,
      "flos": 23404887694080.0,
      "grad_norm": 2.4349668752143723,
      "language_loss": 0.75232929,
      "learning_rate": 3.9476384747844316e-08,
      "loss": 0.77335984,
      "num_input_tokens_seen": 336667165,
      "step": 15610,
      "time_per_iteration": 2.5551717281341553
    },
    {
      "auxiliary_loss_clip": 0.0104795,
      "auxiliary_loss_mlp": 0.01025218,
      "balance_loss_clip": 1.03465891,
      "balance_loss_mlp": 1.01395273,
      "epoch": 0.9385840973996693,
      "flos": 12824742804480.0,
      "grad_norm": 1.8944609371327303,
      "language_loss": 0.74920487,
      "learning_rate": 3.939942386953987e-08,
      "loss": 0.76993656,
      "num_input_tokens_seen": 336684130,
      "step": 15611,
      "time_per_iteration": 2.5956838130950928
    },
    {
      "auxiliary_loss_clip": 0.01064641,
      "auxiliary_loss_mlp": 0.01029759,
      "balance_loss_clip": 1.03627443,
      "balance_loss_mlp": 1.01792097,
      "epoch": 0.9386442206523373,
      "flos": 15486980152320.0,
      "grad_norm": 1.7645101559276006,
      "language_loss": 0.65826923,
      "learning_rate": 3.9322537337405756e-08,
      "loss": 0.67921323,
      "num_input_tokens_seen": 336701520,
      "step": 15612,
      "time_per_iteration": 2.513424873352051
    },
    {
      "auxiliary_loss_clip": 0.0109041,
      "auxiliary_loss_mlp": 0.01027228,
      "balance_loss_clip": 1.03443384,
      "balance_loss_mlp": 1.01635015,
      "epoch": 0.9387043439050052,
      "flos": 21178821196800.0,
      "grad_norm": 1.6670714604930348,
      "language_loss": 0.56812477,
      "learning_rate": 3.924572515435742e-08,
      "loss": 0.58930123,
      "num_input_tokens_seen": 336720675,
      "step": 15613,
      "time_per_iteration": 2.4972872734069824
    },
    {
      "auxiliary_loss_clip": 0.01079828,
      "auxiliary_loss_mlp": 0.01032943,
      "balance_loss_clip": 1.03230977,
      "balance_loss_mlp": 1.02137995,
      "epoch": 0.9387644671576733,
      "flos": 27668273696640.0,
      "grad_norm": 2.1680228825308174,
      "language_loss": 0.70746958,
      "learning_rate": 3.916898732330764e-08,
      "loss": 0.72859728,
      "num_input_tokens_seen": 336741005,
      "step": 15614,
      "time_per_iteration": 3.956528902053833
    },
    {
      "auxiliary_loss_clip": 0.01095325,
      "auxiliary_loss_mlp": 0.0103118,
      "balance_loss_clip": 1.0352273,
      "balance_loss_mlp": 1.01866865,
      "epoch": 0.9388245904103412,
      "flos": 18836331742080.0,
      "grad_norm": 1.84497174102472,
      "language_loss": 0.80985773,
      "learning_rate": 3.9092323847166544e-08,
      "loss": 0.83112276,
      "num_input_tokens_seen": 336757990,
      "step": 15615,
      "time_per_iteration": 2.4610280990600586
    },
    {
      "auxiliary_loss_clip": 0.01077357,
      "auxiliary_loss_mlp": 0.01029881,
      "balance_loss_clip": 1.03346944,
      "balance_loss_mlp": 1.0183239,
      "epoch": 0.9388847136630092,
      "flos": 25483828083840.0,
      "grad_norm": 1.6703099635584093,
      "language_loss": 0.71688116,
      "learning_rate": 3.901573472884134e-08,
      "loss": 0.73795354,
      "num_input_tokens_seen": 336777705,
      "step": 15616,
      "time_per_iteration": 2.5810277462005615
    },
    {
      "auxiliary_loss_clip": 0.01104291,
      "auxiliary_loss_mlp": 0.01027176,
      "balance_loss_clip": 1.03619444,
      "balance_loss_mlp": 1.01498652,
      "epoch": 0.9389448369156771,
      "flos": 18734992496640.0,
      "grad_norm": 1.9197540841564078,
      "language_loss": 0.66095978,
      "learning_rate": 3.89392199712355e-08,
      "loss": 0.68227446,
      "num_input_tokens_seen": 336798275,
      "step": 15617,
      "time_per_iteration": 2.468583345413208
    },
    {
      "auxiliary_loss_clip": 0.01095477,
      "auxiliary_loss_mlp": 0.01034475,
      "balance_loss_clip": 1.03528595,
      "balance_loss_mlp": 1.02116585,
      "epoch": 0.9390049601683451,
      "flos": 21717839664000.0,
      "grad_norm": 1.999209418275975,
      "language_loss": 0.73500848,
      "learning_rate": 3.886277957725092e-08,
      "loss": 0.75630802,
      "num_input_tokens_seen": 336813835,
      "step": 15618,
      "time_per_iteration": 2.4835472106933594
    },
    {
      "auxiliary_loss_clip": 0.01108949,
      "auxiliary_loss_mlp": 0.01029071,
      "balance_loss_clip": 1.03608859,
      "balance_loss_mlp": 1.01572585,
      "epoch": 0.939065083421013,
      "flos": 19391224020480.0,
      "grad_norm": 2.897151286801757,
      "language_loss": 0.70055056,
      "learning_rate": 3.878641354978662e-08,
      "loss": 0.72193074,
      "num_input_tokens_seen": 336832210,
      "step": 15619,
      "time_per_iteration": 2.4261631965637207
    },
    {
      "auxiliary_loss_clip": 0.01080947,
      "auxiliary_loss_mlp": 0.01032053,
      "balance_loss_clip": 1.03447032,
      "balance_loss_mlp": 1.01939869,
      "epoch": 0.939125206673681,
      "flos": 24681511946880.0,
      "grad_norm": 1.702931350129787,
      "language_loss": 0.77531093,
      "learning_rate": 3.8710121891737834e-08,
      "loss": 0.79644096,
      "num_input_tokens_seen": 336851380,
      "step": 15620,
      "time_per_iteration": 2.559323310852051
    },
    {
      "auxiliary_loss_clip": 0.01087742,
      "auxiliary_loss_mlp": 0.01027814,
      "balance_loss_clip": 1.03265679,
      "balance_loss_mlp": 1.01593494,
      "epoch": 0.9391853299263491,
      "flos": 16325961096960.0,
      "grad_norm": 2.079356462925298,
      "language_loss": 0.74018419,
      "learning_rate": 3.8633904605998025e-08,
      "loss": 0.76133966,
      "num_input_tokens_seen": 336868525,
      "step": 15621,
      "time_per_iteration": 2.435532331466675
    },
    {
      "auxiliary_loss_clip": 0.01073289,
      "auxiliary_loss_mlp": 0.01031832,
      "balance_loss_clip": 1.03336406,
      "balance_loss_mlp": 1.01937485,
      "epoch": 0.939245453179017,
      "flos": 11655778590720.0,
      "grad_norm": 1.8025218715184799,
      "language_loss": 0.66009492,
      "learning_rate": 3.855776169545688e-08,
      "loss": 0.68114614,
      "num_input_tokens_seen": 336886200,
      "step": 15622,
      "time_per_iteration": 2.50565767288208
    },
    {
      "auxiliary_loss_clip": 0.01077214,
      "auxiliary_loss_mlp": 0.0104309,
      "balance_loss_clip": 1.0325036,
      "balance_loss_mlp": 1.02976274,
      "epoch": 0.939305576431685,
      "flos": 23148700917120.0,
      "grad_norm": 1.5926581501070718,
      "language_loss": 0.71839237,
      "learning_rate": 3.848169316300209e-08,
      "loss": 0.73959535,
      "num_input_tokens_seen": 336905815,
      "step": 15623,
      "time_per_iteration": 2.4906768798828125
    },
    {
      "auxiliary_loss_clip": 0.01097274,
      "auxiliary_loss_mlp": 0.01029956,
      "balance_loss_clip": 1.03785825,
      "balance_loss_mlp": 1.01799917,
      "epoch": 0.9393656996843529,
      "flos": 33287790706560.0,
      "grad_norm": 2.0711933140084984,
      "language_loss": 0.7226541,
      "learning_rate": 3.84056990115178e-08,
      "loss": 0.74392635,
      "num_input_tokens_seen": 336928460,
      "step": 15624,
      "time_per_iteration": 2.6029536724090576
    },
    {
      "auxiliary_loss_clip": 0.01067315,
      "auxiliary_loss_mlp": 0.01031042,
      "balance_loss_clip": 1.03196979,
      "balance_loss_mlp": 1.01909781,
      "epoch": 0.9394258229370209,
      "flos": 21689434984320.0,
      "grad_norm": 1.9925254704082167,
      "language_loss": 0.89412034,
      "learning_rate": 3.832977924388614e-08,
      "loss": 0.91510391,
      "num_input_tokens_seen": 336948320,
      "step": 15625,
      "time_per_iteration": 2.52925705909729
    },
    {
      "auxiliary_loss_clip": 0.01092166,
      "auxiliary_loss_mlp": 0.01030648,
      "balance_loss_clip": 1.03493702,
      "balance_loss_mlp": 1.01783955,
      "epoch": 0.9394859461896888,
      "flos": 23874203819520.0,
      "grad_norm": 1.7396492295535817,
      "language_loss": 0.83765614,
      "learning_rate": 3.825393386298592e-08,
      "loss": 0.85888433,
      "num_input_tokens_seen": 336967670,
      "step": 15626,
      "time_per_iteration": 2.4949300289154053
    },
    {
      "auxiliary_loss_clip": 0.01012355,
      "auxiliary_loss_mlp": 0.01003256,
      "balance_loss_clip": 1.00791645,
      "balance_loss_mlp": 1.0020045,
      "epoch": 0.9395460694423569,
      "flos": 61566116993280.0,
      "grad_norm": 0.7759774673790854,
      "language_loss": 0.56157869,
      "learning_rate": 3.8178162871693284e-08,
      "loss": 0.58173484,
      "num_input_tokens_seen": 337028395,
      "step": 15627,
      "time_per_iteration": 3.071989059448242
    },
    {
      "auxiliary_loss_clip": 0.01054101,
      "auxiliary_loss_mlp": 0.01038648,
      "balance_loss_clip": 1.03279734,
      "balance_loss_mlp": 1.02558327,
      "epoch": 0.9396061926950248,
      "flos": 20995712640000.0,
      "grad_norm": 1.4415790746816062,
      "language_loss": 0.70253813,
      "learning_rate": 3.810246627288105e-08,
      "loss": 0.72346568,
      "num_input_tokens_seen": 337048150,
      "step": 15628,
      "time_per_iteration": 2.59606671333313
    },
    {
      "auxiliary_loss_clip": 0.01091129,
      "auxiliary_loss_mlp": 0.01028494,
      "balance_loss_clip": 1.03406906,
      "balance_loss_mlp": 1.01638198,
      "epoch": 0.9396663159476928,
      "flos": 27487786832640.0,
      "grad_norm": 1.9004798770695726,
      "language_loss": 0.75587064,
      "learning_rate": 3.8026844069420025e-08,
      "loss": 0.77706695,
      "num_input_tokens_seen": 337069315,
      "step": 15629,
      "time_per_iteration": 2.5131468772888184
    },
    {
      "auxiliary_loss_clip": 0.01042302,
      "auxiliary_loss_mlp": 0.0103406,
      "balance_loss_clip": 1.03154254,
      "balance_loss_mlp": 1.02128053,
      "epoch": 0.9397264392003607,
      "flos": 19427457864960.0,
      "grad_norm": 1.8388086530186554,
      "language_loss": 0.74254978,
      "learning_rate": 3.795129626417748e-08,
      "loss": 0.76331335,
      "num_input_tokens_seen": 337087765,
      "step": 15630,
      "time_per_iteration": 2.5988919734954834
    },
    {
      "auxiliary_loss_clip": 0.01069051,
      "auxiliary_loss_mlp": 0.01030545,
      "balance_loss_clip": 1.03371143,
      "balance_loss_mlp": 1.0188868,
      "epoch": 0.9397865624530287,
      "flos": 18004820826240.0,
      "grad_norm": 2.201823068216722,
      "language_loss": 0.69073892,
      "learning_rate": 3.787582286001845e-08,
      "loss": 0.71173477,
      "num_input_tokens_seen": 337106265,
      "step": 15631,
      "time_per_iteration": 2.4863619804382324
    },
    {
      "auxiliary_loss_clip": 0.01053212,
      "auxiliary_loss_mlp": 0.01033279,
      "balance_loss_clip": 1.03291154,
      "balance_loss_mlp": 1.02198982,
      "epoch": 0.9398466857056966,
      "flos": 22564613859840.0,
      "grad_norm": 1.6202466852892417,
      "language_loss": 0.75209808,
      "learning_rate": 3.7800423859805086e-08,
      "loss": 0.77296299,
      "num_input_tokens_seen": 337126090,
      "step": 15632,
      "time_per_iteration": 2.5887320041656494
    },
    {
      "auxiliary_loss_clip": 0.01098767,
      "auxiliary_loss_mlp": 0.01035509,
      "balance_loss_clip": 1.03677344,
      "balance_loss_mlp": 1.02221072,
      "epoch": 0.9399068089583646,
      "flos": 24535678728960.0,
      "grad_norm": 1.6608415997978991,
      "language_loss": 0.74258232,
      "learning_rate": 3.772509926639622e-08,
      "loss": 0.76392508,
      "num_input_tokens_seen": 337145655,
      "step": 15633,
      "time_per_iteration": 2.495952606201172
    },
    {
      "auxiliary_loss_clip": 0.01106079,
      "auxiliary_loss_mlp": 0.01035383,
      "balance_loss_clip": 1.03559375,
      "balance_loss_mlp": 1.02230024,
      "epoch": 0.9399669322110327,
      "flos": 25630343660160.0,
      "grad_norm": 1.749247723318036,
      "language_loss": 0.72578341,
      "learning_rate": 3.764984908264823e-08,
      "loss": 0.74719805,
      "num_input_tokens_seen": 337164805,
      "step": 15634,
      "time_per_iteration": 2.5107321739196777
    },
    {
      "auxiliary_loss_clip": 0.01094454,
      "auxiliary_loss_mlp": 0.01030676,
      "balance_loss_clip": 1.03352809,
      "balance_loss_mlp": 1.01782537,
      "epoch": 0.9400270554637006,
      "flos": 17089385783040.0,
      "grad_norm": 1.7061775040608376,
      "language_loss": 0.69017518,
      "learning_rate": 3.75746733114144e-08,
      "loss": 0.71142638,
      "num_input_tokens_seen": 337182280,
      "step": 15635,
      "time_per_iteration": 2.4536962509155273
    },
    {
      "auxiliary_loss_clip": 0.01052242,
      "auxiliary_loss_mlp": 0.01025025,
      "balance_loss_clip": 1.03543079,
      "balance_loss_mlp": 1.01376569,
      "epoch": 0.9400871787163686,
      "flos": 22055113393920.0,
      "grad_norm": 1.4872834361896294,
      "language_loss": 0.7435869,
      "learning_rate": 3.7499571955545985e-08,
      "loss": 0.76435953,
      "num_input_tokens_seen": 337203495,
      "step": 15636,
      "time_per_iteration": 2.5863916873931885
    },
    {
      "auxiliary_loss_clip": 0.01093707,
      "auxiliary_loss_mlp": 0.01033954,
      "balance_loss_clip": 1.03534591,
      "balance_loss_mlp": 1.02127576,
      "epoch": 0.9401473019690365,
      "flos": 16982767238400.0,
      "grad_norm": 1.9317416959277358,
      "language_loss": 0.8285619,
      "learning_rate": 3.7424545017890054e-08,
      "loss": 0.84983855,
      "num_input_tokens_seen": 337220435,
      "step": 15637,
      "time_per_iteration": 2.426757574081421
    },
    {
      "auxiliary_loss_clip": 0.01056775,
      "auxiliary_loss_mlp": 0.01030301,
      "balance_loss_clip": 1.03324938,
      "balance_loss_mlp": 1.01804018,
      "epoch": 0.9402074252217045,
      "flos": 19681956702720.0,
      "grad_norm": 2.598937591284569,
      "language_loss": 0.69316196,
      "learning_rate": 3.7349592501292325e-08,
      "loss": 0.71403271,
      "num_input_tokens_seen": 337238095,
      "step": 15638,
      "time_per_iteration": 2.5606420040130615
    },
    {
      "auxiliary_loss_clip": 0.0108728,
      "auxiliary_loss_mlp": 0.01036695,
      "balance_loss_clip": 1.03294575,
      "balance_loss_mlp": 1.02608502,
      "epoch": 0.9402675484743724,
      "flos": 24754302858240.0,
      "grad_norm": 1.7426454839201664,
      "language_loss": 0.84790587,
      "learning_rate": 3.727471440859498e-08,
      "loss": 0.86914563,
      "num_input_tokens_seen": 337256645,
      "step": 15639,
      "time_per_iteration": 2.4934165477752686
    },
    {
      "auxiliary_loss_clip": 0.01078643,
      "auxiliary_loss_mlp": 0.00783335,
      "balance_loss_clip": 1.03200781,
      "balance_loss_mlp": 1.0097326,
      "epoch": 0.9403276717270405,
      "flos": 25558630156800.0,
      "grad_norm": 1.4460972055289312,
      "language_loss": 0.78161514,
      "learning_rate": 3.719991074263662e-08,
      "loss": 0.80023497,
      "num_input_tokens_seen": 337278360,
      "step": 15640,
      "time_per_iteration": 2.5743680000305176
    },
    {
      "auxiliary_loss_clip": 0.01095427,
      "auxiliary_loss_mlp": 0.01032764,
      "balance_loss_clip": 1.03485751,
      "balance_loss_mlp": 1.02090907,
      "epoch": 0.9403877949797084,
      "flos": 26689852154880.0,
      "grad_norm": 1.5578220104234615,
      "language_loss": 0.74046171,
      "learning_rate": 3.7125181506254544e-08,
      "loss": 0.76174355,
      "num_input_tokens_seen": 337302480,
      "step": 15641,
      "time_per_iteration": 2.5387516021728516
    },
    {
      "auxiliary_loss_clip": 0.01094963,
      "auxiliary_loss_mlp": 0.01034675,
      "balance_loss_clip": 1.03421485,
      "balance_loss_mlp": 1.02131152,
      "epoch": 0.9404479182323764,
      "flos": 15011666455680.0,
      "grad_norm": 1.8419473506056603,
      "language_loss": 0.82275391,
      "learning_rate": 3.7050526702282256e-08,
      "loss": 0.84405023,
      "num_input_tokens_seen": 337316600,
      "step": 15642,
      "time_per_iteration": 2.4356019496917725
    },
    {
      "auxiliary_loss_clip": 0.01085604,
      "auxiliary_loss_mlp": 0.0102971,
      "balance_loss_clip": 1.03353477,
      "balance_loss_mlp": 1.018713,
      "epoch": 0.9405080414850443,
      "flos": 24973573432320.0,
      "grad_norm": 2.2359974476560325,
      "language_loss": 0.68467659,
      "learning_rate": 3.697594633355084e-08,
      "loss": 0.70582974,
      "num_input_tokens_seen": 337336895,
      "step": 15643,
      "time_per_iteration": 5.2143096923828125
    },
    {
      "auxiliary_loss_clip": 0.01094767,
      "auxiliary_loss_mlp": 0.0103785,
      "balance_loss_clip": 1.03663301,
      "balance_loss_mlp": 1.02488017,
      "epoch": 0.9405681647377123,
      "flos": 20844743777280.0,
      "grad_norm": 1.959400775663905,
      "language_loss": 0.76580298,
      "learning_rate": 3.6901440402888226e-08,
      "loss": 0.78712916,
      "num_input_tokens_seen": 337355105,
      "step": 15644,
      "time_per_iteration": 2.4593496322631836
    },
    {
      "auxiliary_loss_clip": 0.0108479,
      "auxiliary_loss_mlp": 0.01029824,
      "balance_loss_clip": 1.03233635,
      "balance_loss_mlp": 1.01855838,
      "epoch": 0.9406282879903802,
      "flos": 23805578885760.0,
      "grad_norm": 1.5445104534505583,
      "language_loss": 0.67192495,
      "learning_rate": 3.682700891311974e-08,
      "loss": 0.69307107,
      "num_input_tokens_seen": 337374905,
      "step": 15645,
      "time_per_iteration": 3.972628116607666
    },
    {
      "auxiliary_loss_clip": 0.01080082,
      "auxiliary_loss_mlp": 0.00781488,
      "balance_loss_clip": 1.03476369,
      "balance_loss_mlp": 1.00718069,
      "epoch": 0.9406884112430483,
      "flos": 27674953626240.0,
      "grad_norm": 1.4732281544785932,
      "language_loss": 0.70308292,
      "learning_rate": 3.6752651867067774e-08,
      "loss": 0.72169864,
      "num_input_tokens_seen": 337397130,
      "step": 15646,
      "time_per_iteration": 2.581941843032837
    },
    {
      "auxiliary_loss_clip": 0.01086537,
      "auxiliary_loss_mlp": 0.01028035,
      "balance_loss_clip": 1.03286147,
      "balance_loss_mlp": 1.01653695,
      "epoch": 0.9407485344957163,
      "flos": 23075048079360.0,
      "grad_norm": 1.584961726221311,
      "language_loss": 0.74424946,
      "learning_rate": 3.667836926755208e-08,
      "loss": 0.76539528,
      "num_input_tokens_seen": 337418660,
      "step": 15647,
      "time_per_iteration": 2.4914541244506836
    },
    {
      "auxiliary_loss_clip": 0.01010806,
      "auxiliary_loss_mlp": 0.01002127,
      "balance_loss_clip": 1.00810754,
      "balance_loss_mlp": 1.00113726,
      "epoch": 0.9408086577483842,
      "flos": 71014034304000.0,
      "grad_norm": 0.8856472729930946,
      "language_loss": 0.635315,
      "learning_rate": 3.660416111738907e-08,
      "loss": 0.65544432,
      "num_input_tokens_seen": 337478055,
      "step": 15648,
      "time_per_iteration": 3.2263569831848145
    },
    {
      "auxiliary_loss_clip": 0.01099221,
      "auxiliary_loss_mlp": 0.0102973,
      "balance_loss_clip": 1.03376043,
      "balance_loss_mlp": 1.01909661,
      "epoch": 0.9408687810010522,
      "flos": 23730956380800.0,
      "grad_norm": 1.4547953232275144,
      "language_loss": 0.66908121,
      "learning_rate": 3.653002741939337e-08,
      "loss": 0.6903708,
      "num_input_tokens_seen": 337499405,
      "step": 15649,
      "time_per_iteration": 2.4591097831726074
    },
    {
      "auxiliary_loss_clip": 0.01070349,
      "auxiliary_loss_mlp": 0.01028486,
      "balance_loss_clip": 1.03119326,
      "balance_loss_mlp": 1.01690519,
      "epoch": 0.9409289042537201,
      "flos": 18369314087040.0,
      "grad_norm": 2.155291871068878,
      "language_loss": 0.77596426,
      "learning_rate": 3.645596817637586e-08,
      "loss": 0.79695261,
      "num_input_tokens_seen": 337517195,
      "step": 15650,
      "time_per_iteration": 2.518077850341797
    },
    {
      "auxiliary_loss_clip": 0.01057867,
      "auxiliary_loss_mlp": 0.01029418,
      "balance_loss_clip": 1.03593385,
      "balance_loss_mlp": 1.01803923,
      "epoch": 0.9409890275063881,
      "flos": 23878333883520.0,
      "grad_norm": 1.6967572992831086,
      "language_loss": 0.74414819,
      "learning_rate": 3.638198339114451e-08,
      "loss": 0.76502103,
      "num_input_tokens_seen": 337535245,
      "step": 15651,
      "time_per_iteration": 2.5812008380889893
    },
    {
      "auxiliary_loss_clip": 0.0110227,
      "auxiliary_loss_mlp": 0.01032841,
      "balance_loss_clip": 1.03449845,
      "balance_loss_mlp": 1.02089024,
      "epoch": 0.941049150759056,
      "flos": 16545088016640.0,
      "grad_norm": 1.7307795360203881,
      "language_loss": 0.72489846,
      "learning_rate": 3.630807306650507e-08,
      "loss": 0.74624956,
      "num_input_tokens_seen": 337553040,
      "step": 15652,
      "time_per_iteration": 3.8349947929382324
    },
    {
      "auxiliary_loss_clip": 0.01068449,
      "auxiliary_loss_mlp": 0.01032395,
      "balance_loss_clip": 1.03421938,
      "balance_loss_mlp": 1.01955616,
      "epoch": 0.9411092740117241,
      "flos": 25118401069440.0,
      "grad_norm": 1.7195852813036407,
      "language_loss": 0.66279745,
      "learning_rate": 3.6234237205260645e-08,
      "loss": 0.68380594,
      "num_input_tokens_seen": 337574580,
      "step": 15653,
      "time_per_iteration": 2.58687424659729
    },
    {
      "auxiliary_loss_clip": 0.0110384,
      "auxiliary_loss_mlp": 0.01033245,
      "balance_loss_clip": 1.03476989,
      "balance_loss_mlp": 1.02085924,
      "epoch": 0.941169397264392,
      "flos": 21142264129920.0,
      "grad_norm": 2.3154982367892782,
      "language_loss": 0.77530783,
      "learning_rate": 3.6160475810210536e-08,
      "loss": 0.79667866,
      "num_input_tokens_seen": 337593010,
      "step": 15654,
      "time_per_iteration": 2.4335625171661377
    },
    {
      "auxiliary_loss_clip": 0.01097227,
      "auxiliary_loss_mlp": 0.01028569,
      "balance_loss_clip": 1.03395033,
      "balance_loss_mlp": 1.01659465,
      "epoch": 0.94122952051706,
      "flos": 38508914995200.0,
      "grad_norm": 1.460375891806464,
      "language_loss": 0.7002095,
      "learning_rate": 3.6086788884152065e-08,
      "loss": 0.72146749,
      "num_input_tokens_seen": 337616170,
      "step": 15655,
      "time_per_iteration": 2.6295151710510254
    },
    {
      "auxiliary_loss_clip": 0.01103539,
      "auxiliary_loss_mlp": 0.01032062,
      "balance_loss_clip": 1.03439283,
      "balance_loss_mlp": 1.0186981,
      "epoch": 0.9412896437697279,
      "flos": 18369206346240.0,
      "grad_norm": 1.913269440874927,
      "language_loss": 0.72244883,
      "learning_rate": 3.601317642987944e-08,
      "loss": 0.74380481,
      "num_input_tokens_seen": 337635215,
      "step": 15656,
      "time_per_iteration": 2.4098451137542725
    },
    {
      "auxiliary_loss_clip": 0.01077273,
      "auxiliary_loss_mlp": 0.01027959,
      "balance_loss_clip": 1.03384256,
      "balance_loss_mlp": 1.01640153,
      "epoch": 0.9413497670223959,
      "flos": 25884950238720.0,
      "grad_norm": 1.760065280532338,
      "language_loss": 0.77769065,
      "learning_rate": 3.593963845018377e-08,
      "loss": 0.79874289,
      "num_input_tokens_seen": 337654195,
      "step": 15657,
      "time_per_iteration": 2.534491777420044
    },
    {
      "auxiliary_loss_clip": 0.01065549,
      "auxiliary_loss_mlp": 0.01027309,
      "balance_loss_clip": 1.03146863,
      "balance_loss_mlp": 1.01485789,
      "epoch": 0.9414098902750638,
      "flos": 16618309891200.0,
      "grad_norm": 2.0900480415180263,
      "language_loss": 0.84083956,
      "learning_rate": 3.586617494785371e-08,
      "loss": 0.86176813,
      "num_input_tokens_seen": 337671810,
      "step": 15658,
      "time_per_iteration": 2.491455554962158
    },
    {
      "auxiliary_loss_clip": 0.01108679,
      "auxiliary_loss_mlp": 0.01031158,
      "balance_loss_clip": 1.03605294,
      "balance_loss_mlp": 1.01738298,
      "epoch": 0.9414700135277319,
      "flos": 18625033987200.0,
      "grad_norm": 1.767511329437255,
      "language_loss": 0.70302105,
      "learning_rate": 3.5792785925675254e-08,
      "loss": 0.72441941,
      "num_input_tokens_seen": 337689410,
      "step": 15659,
      "time_per_iteration": 2.4331743717193604
    },
    {
      "auxiliary_loss_clip": 0.01073906,
      "auxiliary_loss_mlp": 0.01035207,
      "balance_loss_clip": 1.0321064,
      "balance_loss_mlp": 1.02437103,
      "epoch": 0.9415301367803999,
      "flos": 26280146649600.0,
      "grad_norm": 1.6385778080637707,
      "language_loss": 0.7954824,
      "learning_rate": 3.571947138643172e-08,
      "loss": 0.8165735,
      "num_input_tokens_seen": 337709950,
      "step": 15660,
      "time_per_iteration": 2.5331966876983643
    },
    {
      "auxiliary_loss_clip": 0.01067584,
      "auxiliary_loss_mlp": 0.01026976,
      "balance_loss_clip": 1.03235781,
      "balance_loss_mlp": 1.01559711,
      "epoch": 0.9415902600330678,
      "flos": 23261388860160.0,
      "grad_norm": 1.414190290963814,
      "language_loss": 0.68078941,
      "learning_rate": 3.564623133290201e-08,
      "loss": 0.70173502,
      "num_input_tokens_seen": 337731320,
      "step": 15661,
      "time_per_iteration": 2.5992352962493896
    },
    {
      "auxiliary_loss_clip": 0.01088127,
      "auxiliary_loss_mlp": 0.01031326,
      "balance_loss_clip": 1.03245473,
      "balance_loss_mlp": 1.01936293,
      "epoch": 0.9416503832857358,
      "flos": 14719138093440.0,
      "grad_norm": 2.022563726965061,
      "language_loss": 0.66442215,
      "learning_rate": 3.557306576786434e-08,
      "loss": 0.68561667,
      "num_input_tokens_seen": 337747720,
      "step": 15662,
      "time_per_iteration": 2.440382957458496
    },
    {
      "auxiliary_loss_clip": 0.01007882,
      "auxiliary_loss_mlp": 0.00999919,
      "balance_loss_clip": 1.00527692,
      "balance_loss_mlp": 0.99880993,
      "epoch": 0.9417105065384037,
      "flos": 70312698276480.0,
      "grad_norm": 0.7749794122221303,
      "language_loss": 0.59313601,
      "learning_rate": 3.5499974694092935e-08,
      "loss": 0.61321402,
      "num_input_tokens_seen": 337806930,
      "step": 15663,
      "time_per_iteration": 3.1881017684936523
    },
    {
      "auxiliary_loss_clip": 0.01098434,
      "auxiliary_loss_mlp": 0.01032455,
      "balance_loss_clip": 1.03548408,
      "balance_loss_mlp": 1.0190264,
      "epoch": 0.9417706297910717,
      "flos": 34057895322240.0,
      "grad_norm": 1.9317907782806034,
      "language_loss": 0.66764963,
      "learning_rate": 3.542695811435914e-08,
      "loss": 0.68895859,
      "num_input_tokens_seen": 337828100,
      "step": 15664,
      "time_per_iteration": 2.5654635429382324
    },
    {
      "auxiliary_loss_clip": 0.01079431,
      "auxiliary_loss_mlp": 0.01026486,
      "balance_loss_clip": 1.03761995,
      "balance_loss_mlp": 1.01532221,
      "epoch": 0.9418307530437396,
      "flos": 16471614746880.0,
      "grad_norm": 2.0450056971553563,
      "language_loss": 0.73074615,
      "learning_rate": 3.535401603143207e-08,
      "loss": 0.75180531,
      "num_input_tokens_seen": 337844805,
      "step": 15665,
      "time_per_iteration": 2.4893178939819336
    },
    {
      "auxiliary_loss_clip": 0.0110108,
      "auxiliary_loss_mlp": 0.01029434,
      "balance_loss_clip": 1.03542984,
      "balance_loss_mlp": 1.01811516,
      "epoch": 0.9418908762964077,
      "flos": 11253543114240.0,
      "grad_norm": 1.9595407608638742,
      "language_loss": 0.63600206,
      "learning_rate": 3.528114844807773e-08,
      "loss": 0.65730727,
      "num_input_tokens_seen": 337860490,
      "step": 15666,
      "time_per_iteration": 2.4029088020324707
    },
    {
      "auxiliary_loss_clip": 0.0106691,
      "auxiliary_loss_mlp": 0.01032741,
      "balance_loss_clip": 1.03354192,
      "balance_loss_mlp": 1.0202601,
      "epoch": 0.9419509995490756,
      "flos": 18438836860800.0,
      "grad_norm": 2.202105490437277,
      "language_loss": 0.79241467,
      "learning_rate": 3.520835536705902e-08,
      "loss": 0.81341124,
      "num_input_tokens_seen": 337878360,
      "step": 15667,
      "time_per_iteration": 2.551560163497925
    },
    {
      "auxiliary_loss_clip": 0.01099912,
      "auxiliary_loss_mlp": 0.01027614,
      "balance_loss_clip": 1.0334723,
      "balance_loss_mlp": 1.0168376,
      "epoch": 0.9420111228017436,
      "flos": 20737945664640.0,
      "grad_norm": 1.670953841185381,
      "language_loss": 0.75171089,
      "learning_rate": 3.5135636791136404e-08,
      "loss": 0.77298611,
      "num_input_tokens_seen": 337895635,
      "step": 15668,
      "time_per_iteration": 2.4362730979919434
    },
    {
      "auxiliary_loss_clip": 0.01058742,
      "auxiliary_loss_mlp": 0.01029842,
      "balance_loss_clip": 1.03515005,
      "balance_loss_mlp": 1.01779616,
      "epoch": 0.9420712460544115,
      "flos": 21141940907520.0,
      "grad_norm": 2.9343811417516683,
      "language_loss": 0.59584165,
      "learning_rate": 3.506299272306723e-08,
      "loss": 0.61672747,
      "num_input_tokens_seen": 337913940,
      "step": 15669,
      "time_per_iteration": 2.5626139640808105
    },
    {
      "auxiliary_loss_clip": 0.01061899,
      "auxiliary_loss_mlp": 0.01024924,
      "balance_loss_clip": 1.0329926,
      "balance_loss_mlp": 1.0136292,
      "epoch": 0.9421313693070795,
      "flos": 15851760721920.0,
      "grad_norm": 1.5343841697263154,
      "language_loss": 0.77062893,
      "learning_rate": 3.4990423165606406e-08,
      "loss": 0.79149711,
      "num_input_tokens_seen": 337932015,
      "step": 15670,
      "time_per_iteration": 2.54091477394104
    },
    {
      "auxiliary_loss_clip": 0.01102554,
      "auxiliary_loss_mlp": 0.01035114,
      "balance_loss_clip": 1.03426886,
      "balance_loss_mlp": 1.02251911,
      "epoch": 0.9421914925597474,
      "flos": 32415915882240.0,
      "grad_norm": 1.7875645926522663,
      "language_loss": 0.64910525,
      "learning_rate": 3.491792812150574e-08,
      "loss": 0.67048192,
      "num_input_tokens_seen": 337953345,
      "step": 15671,
      "time_per_iteration": 2.538486957550049
    },
    {
      "auxiliary_loss_clip": 0.01078699,
      "auxiliary_loss_mlp": 0.0103229,
      "balance_loss_clip": 1.03403854,
      "balance_loss_mlp": 1.01985049,
      "epoch": 0.9422516158124155,
      "flos": 19718513769600.0,
      "grad_norm": 1.599990490213978,
      "language_loss": 0.79880714,
      "learning_rate": 3.48455075935139e-08,
      "loss": 0.81991702,
      "num_input_tokens_seen": 337973685,
      "step": 15672,
      "time_per_iteration": 2.500953197479248
    },
    {
      "auxiliary_loss_clip": 0.01070635,
      "auxiliary_loss_mlp": 0.0103703,
      "balance_loss_clip": 1.03350782,
      "balance_loss_mlp": 1.02380991,
      "epoch": 0.9423117390650835,
      "flos": 16253277926400.0,
      "grad_norm": 2.0253401450432187,
      "language_loss": 0.73314315,
      "learning_rate": 3.47731615843776e-08,
      "loss": 0.75421977,
      "num_input_tokens_seen": 337989175,
      "step": 15673,
      "time_per_iteration": 2.520496368408203
    },
    {
      "auxiliary_loss_clip": 0.0108578,
      "auxiliary_loss_mlp": 0.01029857,
      "balance_loss_clip": 1.03189969,
      "balance_loss_mlp": 1.01686299,
      "epoch": 0.9423718623177514,
      "flos": 31796564647680.0,
      "grad_norm": 1.4465091768296927,
      "language_loss": 0.70150232,
      "learning_rate": 3.470089009683974e-08,
      "loss": 0.72265869,
      "num_input_tokens_seen": 338011800,
      "step": 15674,
      "time_per_iteration": 2.56890869140625
    },
    {
      "auxiliary_loss_clip": 0.0110289,
      "auxiliary_loss_mlp": 0.01025594,
      "balance_loss_clip": 1.03436375,
      "balance_loss_mlp": 1.01413178,
      "epoch": 0.9424319855704194,
      "flos": 23331809473920.0,
      "grad_norm": 1.7556342710336021,
      "language_loss": 0.81015992,
      "learning_rate": 3.462869313364125e-08,
      "loss": 0.83144474,
      "num_input_tokens_seen": 338032120,
      "step": 15675,
      "time_per_iteration": 2.45219087600708
    },
    {
      "auxiliary_loss_clip": 0.01078127,
      "auxiliary_loss_mlp": 0.01028014,
      "balance_loss_clip": 1.03625762,
      "balance_loss_mlp": 1.016361,
      "epoch": 0.9424921088230873,
      "flos": 20777627214720.0,
      "grad_norm": 1.6627026204337043,
      "language_loss": 0.62821412,
      "learning_rate": 3.4556570697519494e-08,
      "loss": 0.64927554,
      "num_input_tokens_seen": 338051880,
      "step": 15676,
      "time_per_iteration": 2.512535333633423
    },
    {
      "auxiliary_loss_clip": 0.01084974,
      "auxiliary_loss_mlp": 0.01033982,
      "balance_loss_clip": 1.03465629,
      "balance_loss_mlp": 1.0221746,
      "epoch": 0.9425522320757553,
      "flos": 19026658932480.0,
      "grad_norm": 1.7751120713609283,
      "language_loss": 0.66843808,
      "learning_rate": 3.448452279120984e-08,
      "loss": 0.68962765,
      "num_input_tokens_seen": 338069665,
      "step": 15677,
      "time_per_iteration": 2.508359432220459
    },
    {
      "auxiliary_loss_clip": 0.01068971,
      "auxiliary_loss_mlp": 0.01033313,
      "balance_loss_clip": 1.03282404,
      "balance_loss_mlp": 1.01988363,
      "epoch": 0.9426123553284232,
      "flos": 25155353185920.0,
      "grad_norm": 1.7407990943905052,
      "language_loss": 0.64489031,
      "learning_rate": 3.441254941744387e-08,
      "loss": 0.66591316,
      "num_input_tokens_seen": 338090490,
      "step": 15678,
      "time_per_iteration": 2.5544772148132324
    },
    {
      "auxiliary_loss_clip": 0.01072923,
      "auxiliary_loss_mlp": 0.01028568,
      "balance_loss_clip": 1.03389382,
      "balance_loss_mlp": 1.01699889,
      "epoch": 0.9426724785810913,
      "flos": 21179359900800.0,
      "grad_norm": 1.4945064969287034,
      "language_loss": 0.74536288,
      "learning_rate": 3.434065057895097e-08,
      "loss": 0.76637781,
      "num_input_tokens_seen": 338109825,
      "step": 15679,
      "time_per_iteration": 2.554229259490967
    },
    {
      "auxiliary_loss_clip": 0.0108788,
      "auxiliary_loss_mlp": 0.01032167,
      "balance_loss_clip": 1.03659582,
      "balance_loss_mlp": 1.01969767,
      "epoch": 0.9427326018337592,
      "flos": 14756916222720.0,
      "grad_norm": 2.4103518553209264,
      "language_loss": 0.77639633,
      "learning_rate": 3.426882627845762e-08,
      "loss": 0.79759681,
      "num_input_tokens_seen": 338125790,
      "step": 15680,
      "time_per_iteration": 2.482663869857788
    },
    {
      "auxiliary_loss_clip": 0.01091117,
      "auxiliary_loss_mlp": 0.01032913,
      "balance_loss_clip": 1.03421128,
      "balance_loss_mlp": 1.0214982,
      "epoch": 0.9427927250864272,
      "flos": 20923640000640.0,
      "grad_norm": 1.9114306755798662,
      "language_loss": 0.75358343,
      "learning_rate": 3.419707651868742e-08,
      "loss": 0.77482367,
      "num_input_tokens_seen": 338145610,
      "step": 15681,
      "time_per_iteration": 3.925328493118286
    },
    {
      "auxiliary_loss_clip": 0.0108568,
      "auxiliary_loss_mlp": 0.0103561,
      "balance_loss_clip": 1.03675246,
      "balance_loss_mlp": 1.02325439,
      "epoch": 0.9428528483390951,
      "flos": 19752520970880.0,
      "grad_norm": 1.774822425757189,
      "language_loss": 0.66181099,
      "learning_rate": 3.412540130236086e-08,
      "loss": 0.68302393,
      "num_input_tokens_seen": 338165960,
      "step": 15682,
      "time_per_iteration": 3.9952878952026367
    },
    {
      "auxiliary_loss_clip": 0.01068319,
      "auxiliary_loss_mlp": 0.01027979,
      "balance_loss_clip": 1.03228939,
      "balance_loss_mlp": 1.01575434,
      "epoch": 0.9429129715917631,
      "flos": 24534996370560.0,
      "grad_norm": 2.0137971196133,
      "language_loss": 0.76196301,
      "learning_rate": 3.405380063219665e-08,
      "loss": 0.78292596,
      "num_input_tokens_seen": 338187215,
      "step": 15683,
      "time_per_iteration": 3.9438538551330566
    },
    {
      "auxiliary_loss_clip": 0.01096394,
      "auxiliary_loss_mlp": 0.01038267,
      "balance_loss_clip": 1.03498816,
      "balance_loss_mlp": 1.02513564,
      "epoch": 0.942973094844431,
      "flos": 17959824063360.0,
      "grad_norm": 2.7198957585100096,
      "language_loss": 0.75759923,
      "learning_rate": 3.398227451090885e-08,
      "loss": 0.77894592,
      "num_input_tokens_seen": 338201825,
      "step": 15684,
      "time_per_iteration": 2.4250645637512207
    },
    {
      "auxiliary_loss_clip": 0.01099085,
      "auxiliary_loss_mlp": 0.01021726,
      "balance_loss_clip": 1.03330731,
      "balance_loss_mlp": 1.01070476,
      "epoch": 0.9430332180970991,
      "flos": 26137689310080.0,
      "grad_norm": 1.6887135532103126,
      "language_loss": 0.77514446,
      "learning_rate": 3.391082294121017e-08,
      "loss": 0.79635251,
      "num_input_tokens_seen": 338220865,
      "step": 15685,
      "time_per_iteration": 2.5002598762512207
    },
    {
      "auxiliary_loss_clip": 0.01089204,
      "auxiliary_loss_mlp": 0.01030671,
      "balance_loss_clip": 1.03340626,
      "balance_loss_mlp": 1.01921463,
      "epoch": 0.943093341349767,
      "flos": 23951376190080.0,
      "grad_norm": 1.9584191055228868,
      "language_loss": 0.75549257,
      "learning_rate": 3.383944592581023e-08,
      "loss": 0.77669132,
      "num_input_tokens_seen": 338240160,
      "step": 15686,
      "time_per_iteration": 2.4820642471313477
    },
    {
      "auxiliary_loss_clip": 0.0109323,
      "auxiliary_loss_mlp": 0.01028668,
      "balance_loss_clip": 1.03342128,
      "balance_loss_mlp": 1.01583529,
      "epoch": 0.943153464602435,
      "flos": 17968407413760.0,
      "grad_norm": 1.669328264835531,
      "language_loss": 0.80759299,
      "learning_rate": 3.376814346741575e-08,
      "loss": 0.82881194,
      "num_input_tokens_seen": 338259305,
      "step": 15687,
      "time_per_iteration": 2.456772565841675
    },
    {
      "auxiliary_loss_clip": 0.01085673,
      "auxiliary_loss_mlp": 0.01033243,
      "balance_loss_clip": 1.03539836,
      "balance_loss_mlp": 1.01965952,
      "epoch": 0.943213587855103,
      "flos": 14501519544960.0,
      "grad_norm": 2.1643458185271016,
      "language_loss": 0.75992525,
      "learning_rate": 3.369691556873011e-08,
      "loss": 0.78111446,
      "num_input_tokens_seen": 338274950,
      "step": 15688,
      "time_per_iteration": 2.4622299671173096
    },
    {
      "auxiliary_loss_clip": 0.01074408,
      "auxiliary_loss_mlp": 0.01027986,
      "balance_loss_clip": 1.03243423,
      "balance_loss_mlp": 1.0159874,
      "epoch": 0.9432737111077709,
      "flos": 28986411093120.0,
      "grad_norm": 1.7474172814670688,
      "language_loss": 0.68591654,
      "learning_rate": 3.3625762232454504e-08,
      "loss": 0.70694053,
      "num_input_tokens_seen": 338295585,
      "step": 15689,
      "time_per_iteration": 2.56526780128479
    },
    {
      "auxiliary_loss_clip": 0.01090121,
      "auxiliary_loss_mlp": 0.01035194,
      "balance_loss_clip": 1.03411829,
      "balance_loss_mlp": 1.02469158,
      "epoch": 0.9433338343604389,
      "flos": 21609066303360.0,
      "grad_norm": 1.6934502263449016,
      "language_loss": 0.80556005,
      "learning_rate": 3.35546834612872e-08,
      "loss": 0.82681322,
      "num_input_tokens_seen": 338314555,
      "step": 15690,
      "time_per_iteration": 2.464779853820801
    },
    {
      "auxiliary_loss_clip": 0.01090631,
      "auxiliary_loss_mlp": 0.01027904,
      "balance_loss_clip": 1.03397298,
      "balance_loss_mlp": 1.0163343,
      "epoch": 0.9433939576131068,
      "flos": 33182285483520.0,
      "grad_norm": 2.0216042002833707,
      "language_loss": 0.5993464,
      "learning_rate": 3.348367925792317e-08,
      "loss": 0.6205318,
      "num_input_tokens_seen": 338336260,
      "step": 15691,
      "time_per_iteration": 4.01296067237854
    },
    {
      "auxiliary_loss_clip": 0.01066042,
      "auxiliary_loss_mlp": 0.01029114,
      "balance_loss_clip": 1.0339098,
      "balance_loss_mlp": 1.01655555,
      "epoch": 0.9434540808657749,
      "flos": 20486391742080.0,
      "grad_norm": 1.6831999821465933,
      "language_loss": 0.66760135,
      "learning_rate": 3.341274962505514e-08,
      "loss": 0.68855298,
      "num_input_tokens_seen": 338354680,
      "step": 15692,
      "time_per_iteration": 2.565175771713257
    },
    {
      "auxiliary_loss_clip": 0.01089761,
      "auxiliary_loss_mlp": 0.01031287,
      "balance_loss_clip": 1.03518808,
      "balance_loss_mlp": 1.01916313,
      "epoch": 0.9435142041184428,
      "flos": 21542955321600.0,
      "grad_norm": 2.6901904301317447,
      "language_loss": 0.75071502,
      "learning_rate": 3.334189456537251e-08,
      "loss": 0.77192545,
      "num_input_tokens_seen": 338372490,
      "step": 15693,
      "time_per_iteration": 2.485797882080078
    },
    {
      "auxiliary_loss_clip": 0.01070435,
      "auxiliary_loss_mlp": 0.01035116,
      "balance_loss_clip": 1.03407562,
      "balance_loss_mlp": 1.02205086,
      "epoch": 0.9435743273711108,
      "flos": 25009089004800.0,
      "grad_norm": 1.5566043306936497,
      "language_loss": 0.73086226,
      "learning_rate": 3.327111408156291e-08,
      "loss": 0.75191778,
      "num_input_tokens_seen": 338390870,
      "step": 15694,
      "time_per_iteration": 2.590803384780884
    },
    {
      "auxiliary_loss_clip": 0.00997085,
      "auxiliary_loss_mlp": 0.0100194,
      "balance_loss_clip": 1.00461113,
      "balance_loss_mlp": 1.00088501,
      "epoch": 0.9436344506237787,
      "flos": 60158707320960.0,
      "grad_norm": 0.6871429601106139,
      "language_loss": 0.50603455,
      "learning_rate": 3.3200408176309316e-08,
      "loss": 0.52602482,
      "num_input_tokens_seen": 338453075,
      "step": 15695,
      "time_per_iteration": 3.1890575885772705
    },
    {
      "auxiliary_loss_clip": 0.01074789,
      "auxiliary_loss_mlp": 0.01032766,
      "balance_loss_clip": 1.03146875,
      "balance_loss_mlp": 1.02092874,
      "epoch": 0.9436945738764467,
      "flos": 22237252283520.0,
      "grad_norm": 1.732658100683525,
      "language_loss": 0.64924926,
      "learning_rate": 3.312977685229335e-08,
      "loss": 0.6703248,
      "num_input_tokens_seen": 338471770,
      "step": 15696,
      "time_per_iteration": 2.5321311950683594
    },
    {
      "auxiliary_loss_clip": 0.01092893,
      "auxiliary_loss_mlp": 0.01027532,
      "balance_loss_clip": 1.03522789,
      "balance_loss_mlp": 1.01644516,
      "epoch": 0.9437546971291146,
      "flos": 25045179194880.0,
      "grad_norm": 1.8340174655480652,
      "language_loss": 0.66048634,
      "learning_rate": 3.305922011219353e-08,
      "loss": 0.68169057,
      "num_input_tokens_seen": 338492190,
      "step": 15697,
      "time_per_iteration": 2.5046074390411377
    },
    {
      "auxiliary_loss_clip": 0.01000338,
      "auxiliary_loss_mlp": 0.01003406,
      "balance_loss_clip": 1.00664282,
      "balance_loss_mlp": 1.00220776,
      "epoch": 0.9438148203817827,
      "flos": 56790788400000.0,
      "grad_norm": 0.9094381436720116,
      "language_loss": 0.63208926,
      "learning_rate": 3.298873795868506e-08,
      "loss": 0.65212673,
      "num_input_tokens_seen": 338552560,
      "step": 15698,
      "time_per_iteration": 3.074404001235962
    },
    {
      "auxiliary_loss_clip": 0.01083565,
      "auxiliary_loss_mlp": 0.01038896,
      "balance_loss_clip": 1.03433585,
      "balance_loss_mlp": 1.02631903,
      "epoch": 0.9438749436344506,
      "flos": 22346384780160.0,
      "grad_norm": 1.738110008328092,
      "language_loss": 0.69944704,
      "learning_rate": 3.291833039444092e-08,
      "loss": 0.72067165,
      "num_input_tokens_seen": 338571770,
      "step": 15699,
      "time_per_iteration": 2.5105068683624268
    },
    {
      "auxiliary_loss_clip": 0.01064895,
      "auxiliary_loss_mlp": 0.0102797,
      "balance_loss_clip": 1.03237295,
      "balance_loss_mlp": 1.01638937,
      "epoch": 0.9439350668871186,
      "flos": 13370800337280.0,
      "grad_norm": 1.954711165880113,
      "language_loss": 0.74155378,
      "learning_rate": 3.2847997422130734e-08,
      "loss": 0.76248246,
      "num_input_tokens_seen": 338587310,
      "step": 15700,
      "time_per_iteration": 2.5311925411224365
    },
    {
      "auxiliary_loss_clip": 0.01037503,
      "auxiliary_loss_mlp": 0.01029694,
      "balance_loss_clip": 1.03223062,
      "balance_loss_mlp": 1.01827931,
      "epoch": 0.9439951901397866,
      "flos": 17785334770560.0,
      "grad_norm": 1.5486579775127942,
      "language_loss": 0.70962214,
      "learning_rate": 3.2777739044421495e-08,
      "loss": 0.73029411,
      "num_input_tokens_seen": 338606235,
      "step": 15701,
      "time_per_iteration": 2.600071430206299
    },
    {
      "auxiliary_loss_clip": 0.01061321,
      "auxiliary_loss_mlp": 0.0102706,
      "balance_loss_clip": 1.03123927,
      "balance_loss_mlp": 1.01527619,
      "epoch": 0.9440553133924545,
      "flos": 18879568738560.0,
      "grad_norm": 1.746682759368625,
      "language_loss": 0.7774207,
      "learning_rate": 3.2707555263977505e-08,
      "loss": 0.79830456,
      "num_input_tokens_seen": 338624090,
      "step": 15702,
      "time_per_iteration": 2.5510122776031494
    },
    {
      "auxiliary_loss_clip": 0.0108965,
      "auxiliary_loss_mlp": 0.01040415,
      "balance_loss_clip": 1.03325462,
      "balance_loss_mlp": 1.02760553,
      "epoch": 0.9441154366451225,
      "flos": 19572967860480.0,
      "grad_norm": 1.587498964005069,
      "language_loss": 0.66638446,
      "learning_rate": 3.2637446083460194e-08,
      "loss": 0.68768507,
      "num_input_tokens_seen": 338643695,
      "step": 15703,
      "time_per_iteration": 2.4655585289001465
    },
    {
      "auxiliary_loss_clip": 0.01093744,
      "auxiliary_loss_mlp": 0.01027039,
      "balance_loss_clip": 1.03733015,
      "balance_loss_mlp": 1.01462388,
      "epoch": 0.9441755598977905,
      "flos": 30294995472000.0,
      "grad_norm": 2.267236788513038,
      "language_loss": 0.73019505,
      "learning_rate": 3.256741150552833e-08,
      "loss": 0.75140297,
      "num_input_tokens_seen": 338664725,
      "step": 15704,
      "time_per_iteration": 2.5804500579833984
    },
    {
      "auxiliary_loss_clip": 0.01089334,
      "auxiliary_loss_mlp": 0.0103182,
      "balance_loss_clip": 1.03461027,
      "balance_loss_mlp": 1.01966119,
      "epoch": 0.9442356831504585,
      "flos": 20667884186880.0,
      "grad_norm": 1.7462012293071008,
      "language_loss": 0.74243325,
      "learning_rate": 3.2497451532837336e-08,
      "loss": 0.76364475,
      "num_input_tokens_seen": 338683990,
      "step": 15705,
      "time_per_iteration": 2.4650678634643555
    },
    {
      "auxiliary_loss_clip": 0.01081239,
      "auxiliary_loss_mlp": 0.01033646,
      "balance_loss_clip": 1.0363214,
      "balance_loss_mlp": 1.02268505,
      "epoch": 0.9442958064031264,
      "flos": 16107265140480.0,
      "grad_norm": 2.0178982960852574,
      "language_loss": 0.76970136,
      "learning_rate": 3.2427566168039986e-08,
      "loss": 0.79085022,
      "num_input_tokens_seen": 338702025,
      "step": 15706,
      "time_per_iteration": 2.5066909790039062
    },
    {
      "auxiliary_loss_clip": 0.0108838,
      "auxiliary_loss_mlp": 0.01027714,
      "balance_loss_clip": 1.03310847,
      "balance_loss_mlp": 1.01701498,
      "epoch": 0.9443559296557944,
      "flos": 20447392550400.0,
      "grad_norm": 1.562203747922436,
      "language_loss": 0.69338673,
      "learning_rate": 3.23577554137866e-08,
      "loss": 0.71454763,
      "num_input_tokens_seen": 338720920,
      "step": 15707,
      "time_per_iteration": 2.462198257446289
    },
    {
      "auxiliary_loss_clip": 0.01095426,
      "auxiliary_loss_mlp": 0.01026074,
      "balance_loss_clip": 1.03098202,
      "balance_loss_mlp": 1.01544619,
      "epoch": 0.9444160529084623,
      "flos": 21610897896960.0,
      "grad_norm": 1.7255546565955746,
      "language_loss": 0.69244325,
      "learning_rate": 3.22880192727244e-08,
      "loss": 0.71365821,
      "num_input_tokens_seen": 338739590,
      "step": 15708,
      "time_per_iteration": 2.4740374088287354
    },
    {
      "auxiliary_loss_clip": 0.01093335,
      "auxiliary_loss_mlp": 0.01030816,
      "balance_loss_clip": 1.03552103,
      "balance_loss_mlp": 1.01910961,
      "epoch": 0.9444761761611303,
      "flos": 18441781776000.0,
      "grad_norm": 2.5932905281414538,
      "language_loss": 0.70651042,
      "learning_rate": 3.221835774749748e-08,
      "loss": 0.72775197,
      "num_input_tokens_seen": 338757240,
      "step": 15709,
      "time_per_iteration": 2.439553737640381
    },
    {
      "auxiliary_loss_clip": 0.01061682,
      "auxiliary_loss_mlp": 0.01032351,
      "balance_loss_clip": 1.0354259,
      "balance_loss_mlp": 1.02068019,
      "epoch": 0.9445362994137982,
      "flos": 20957144411520.0,
      "grad_norm": 2.0166719916740123,
      "language_loss": 0.84560621,
      "learning_rate": 3.214877084074774e-08,
      "loss": 0.86654651,
      "num_input_tokens_seen": 338773750,
      "step": 15710,
      "time_per_iteration": 2.566802978515625
    },
    {
      "auxiliary_loss_clip": 0.01079072,
      "auxiliary_loss_mlp": 0.01028479,
      "balance_loss_clip": 1.03483796,
      "balance_loss_mlp": 1.01577735,
      "epoch": 0.9445964226664663,
      "flos": 20303283185280.0,
      "grad_norm": 1.6780831326137393,
      "language_loss": 0.71399403,
      "learning_rate": 3.2079258555113956e-08,
      "loss": 0.73506951,
      "num_input_tokens_seen": 338792115,
      "step": 15711,
      "time_per_iteration": 2.5305678844451904
    },
    {
      "auxiliary_loss_clip": 0.01091747,
      "auxiliary_loss_mlp": 0.01027653,
      "balance_loss_clip": 1.03513288,
      "balance_loss_mlp": 1.01568496,
      "epoch": 0.9446565459191342,
      "flos": 26396030903040.0,
      "grad_norm": 1.6212708561934541,
      "language_loss": 0.69102025,
      "learning_rate": 3.200982089323179e-08,
      "loss": 0.71221423,
      "num_input_tokens_seen": 338812480,
      "step": 15712,
      "time_per_iteration": 2.5181126594543457
    },
    {
      "auxiliary_loss_clip": 0.01097565,
      "auxiliary_loss_mlp": 0.01035909,
      "balance_loss_clip": 1.03731918,
      "balance_loss_mlp": 1.0225817,
      "epoch": 0.9447166691718022,
      "flos": 16544764794240.0,
      "grad_norm": 2.2869355654776644,
      "language_loss": 0.70252168,
      "learning_rate": 3.1940457857734246e-08,
      "loss": 0.72385645,
      "num_input_tokens_seen": 338829105,
      "step": 15713,
      "time_per_iteration": 2.427954912185669
    },
    {
      "auxiliary_loss_clip": 0.01078258,
      "auxiliary_loss_mlp": 0.01031116,
      "balance_loss_clip": 1.03253317,
      "balance_loss_mlp": 1.01858115,
      "epoch": 0.9447767924244702,
      "flos": 29164635400320.0,
      "grad_norm": 1.5866640754028876,
      "language_loss": 0.76573288,
      "learning_rate": 3.187116945125212e-08,
      "loss": 0.78682667,
      "num_input_tokens_seen": 338850670,
      "step": 15714,
      "time_per_iteration": 2.5610082149505615
    },
    {
      "auxiliary_loss_clip": 0.01074615,
      "auxiliary_loss_mlp": 0.0103257,
      "balance_loss_clip": 1.03309488,
      "balance_loss_mlp": 1.02026176,
      "epoch": 0.9448369156771381,
      "flos": 19274908803840.0,
      "grad_norm": 1.9495247818315684,
      "language_loss": 0.6749351,
      "learning_rate": 3.1801955676412194e-08,
      "loss": 0.69600695,
      "num_input_tokens_seen": 338867795,
      "step": 15715,
      "time_per_iteration": 2.513051986694336
    },
    {
      "auxiliary_loss_clip": 0.01072448,
      "auxiliary_loss_mlp": 0.01031802,
      "balance_loss_clip": 1.03467333,
      "balance_loss_mlp": 1.01940477,
      "epoch": 0.9448970389298061,
      "flos": 23841166285440.0,
      "grad_norm": 1.7190119794870498,
      "language_loss": 0.7496528,
      "learning_rate": 3.173281653583948e-08,
      "loss": 0.77069533,
      "num_input_tokens_seen": 338887205,
      "step": 15716,
      "time_per_iteration": 2.5538980960845947
    },
    {
      "auxiliary_loss_clip": 0.0108661,
      "auxiliary_loss_mlp": 0.01032908,
      "balance_loss_clip": 1.03726673,
      "balance_loss_mlp": 1.02080798,
      "epoch": 0.944957162182474,
      "flos": 22382259488640.0,
      "grad_norm": 1.970141956860539,
      "language_loss": 0.62535632,
      "learning_rate": 3.166375203215565e-08,
      "loss": 0.64655149,
      "num_input_tokens_seen": 338906130,
      "step": 15717,
      "time_per_iteration": 2.503058433532715
    },
    {
      "auxiliary_loss_clip": 0.01087068,
      "auxiliary_loss_mlp": 0.01029006,
      "balance_loss_clip": 1.03529775,
      "balance_loss_mlp": 1.01743126,
      "epoch": 0.9450172854351421,
      "flos": 17383889393280.0,
      "grad_norm": 1.797944376757105,
      "language_loss": 0.79378897,
      "learning_rate": 3.1594762167979514e-08,
      "loss": 0.81494975,
      "num_input_tokens_seen": 338923045,
      "step": 15718,
      "time_per_iteration": 2.4711573123931885
    },
    {
      "auxiliary_loss_clip": 0.01018436,
      "auxiliary_loss_mlp": 0.01001232,
      "balance_loss_clip": 1.00561619,
      "balance_loss_mlp": 1.00021327,
      "epoch": 0.94507740868781,
      "flos": 68466352406400.0,
      "grad_norm": 0.6965450312518885,
      "language_loss": 0.57797384,
      "learning_rate": 3.152584694592719e-08,
      "loss": 0.59817052,
      "num_input_tokens_seen": 338987545,
      "step": 15719,
      "time_per_iteration": 3.1398465633392334
    },
    {
      "auxiliary_loss_clip": 0.01062304,
      "auxiliary_loss_mlp": 0.00782912,
      "balance_loss_clip": 1.03251743,
      "balance_loss_mlp": 1.00826502,
      "epoch": 0.945137531940478,
      "flos": 21142479611520.0,
      "grad_norm": 1.5059785981604858,
      "language_loss": 0.75820386,
      "learning_rate": 3.145700636861193e-08,
      "loss": 0.77665603,
      "num_input_tokens_seen": 339007830,
      "step": 15720,
      "time_per_iteration": 3.9339468479156494
    },
    {
      "auxiliary_loss_clip": 0.01088049,
      "auxiliary_loss_mlp": 0.01024401,
      "balance_loss_clip": 1.03524518,
      "balance_loss_mlp": 1.01403022,
      "epoch": 0.9451976551931459,
      "flos": 24533918962560.0,
      "grad_norm": 1.6410521607196302,
      "language_loss": 0.72828203,
      "learning_rate": 3.138824043864452e-08,
      "loss": 0.74940652,
      "num_input_tokens_seen": 339028980,
      "step": 15721,
      "time_per_iteration": 3.9112160205841064
    },
    {
      "auxiliary_loss_clip": 0.0105964,
      "auxiliary_loss_mlp": 0.0103317,
      "balance_loss_clip": 1.03328371,
      "balance_loss_mlp": 1.02036095,
      "epoch": 0.9452577784458139,
      "flos": 23440582834560.0,
      "grad_norm": 1.83163372094644,
      "language_loss": 0.85126412,
      "learning_rate": 3.131954915863244e-08,
      "loss": 0.8721922,
      "num_input_tokens_seen": 339047950,
      "step": 15722,
      "time_per_iteration": 4.004755735397339
    },
    {
      "auxiliary_loss_clip": 0.01008144,
      "auxiliary_loss_mlp": 0.01002667,
      "balance_loss_clip": 1.00673735,
      "balance_loss_mlp": 1.0015707,
      "epoch": 0.9453179016984818,
      "flos": 52017686449920.0,
      "grad_norm": 0.8996712334627891,
      "language_loss": 0.64557803,
      "learning_rate": 3.125093253118005e-08,
      "loss": 0.66568625,
      "num_input_tokens_seen": 339104535,
      "step": 15723,
      "time_per_iteration": 3.0693395137786865
    },
    {
      "auxiliary_loss_clip": 0.01064263,
      "auxiliary_loss_mlp": 0.01031055,
      "balance_loss_clip": 1.03540468,
      "balance_loss_mlp": 1.01817489,
      "epoch": 0.9453780249511499,
      "flos": 13473001509120.0,
      "grad_norm": 1.8203193715256547,
      "language_loss": 0.72945344,
      "learning_rate": 3.1182390558889715e-08,
      "loss": 0.75040656,
      "num_input_tokens_seen": 339122050,
      "step": 15724,
      "time_per_iteration": 2.525331497192383
    },
    {
      "auxiliary_loss_clip": 0.01069326,
      "auxiliary_loss_mlp": 0.01024897,
      "balance_loss_clip": 1.03234506,
      "balance_loss_mlp": 1.01342893,
      "epoch": 0.9454381482038178,
      "flos": 23258515772160.0,
      "grad_norm": 1.9964541227892152,
      "language_loss": 0.84764981,
      "learning_rate": 3.111392324436024e-08,
      "loss": 0.86859202,
      "num_input_tokens_seen": 339138940,
      "step": 15725,
      "time_per_iteration": 2.552743673324585
    },
    {
      "auxiliary_loss_clip": 0.01083959,
      "auxiliary_loss_mlp": 0.01026068,
      "balance_loss_clip": 1.03548241,
      "balance_loss_mlp": 1.01452208,
      "epoch": 0.9454982714564858,
      "flos": 19496621502720.0,
      "grad_norm": 1.8408164536048706,
      "language_loss": 0.70893914,
      "learning_rate": 3.104553059018822e-08,
      "loss": 0.73003936,
      "num_input_tokens_seen": 339158245,
      "step": 15726,
      "time_per_iteration": 2.5179543495178223
    },
    {
      "auxiliary_loss_clip": 0.01079004,
      "auxiliary_loss_mlp": 0.01029112,
      "balance_loss_clip": 1.03457808,
      "balance_loss_mlp": 1.01608825,
      "epoch": 0.9455583947091538,
      "flos": 23258120722560.0,
      "grad_norm": 1.8103834068806104,
      "language_loss": 0.60666937,
      "learning_rate": 3.097721259896735e-08,
      "loss": 0.62775052,
      "num_input_tokens_seen": 339178200,
      "step": 15727,
      "time_per_iteration": 2.5459678173065186
    },
    {
      "auxiliary_loss_clip": 0.01089739,
      "auxiliary_loss_mlp": 0.01034079,
      "balance_loss_clip": 1.03348064,
      "balance_loss_mlp": 1.02248621,
      "epoch": 0.9456185179618217,
      "flos": 17673041877120.0,
      "grad_norm": 1.869638453476036,
      "language_loss": 0.81789267,
      "learning_rate": 3.0908969273287566e-08,
      "loss": 0.83913094,
      "num_input_tokens_seen": 339193950,
      "step": 15728,
      "time_per_iteration": 2.461965322494507
    },
    {
      "auxiliary_loss_clip": 0.00990321,
      "auxiliary_loss_mlp": 0.01025665,
      "balance_loss_clip": 1.01658893,
      "balance_loss_mlp": 1.02450264,
      "epoch": 0.9456786412144897,
      "flos": 61415040389760.0,
      "grad_norm": 0.7433309696627799,
      "language_loss": 0.59004426,
      "learning_rate": 3.08408006157368e-08,
      "loss": 0.61020416,
      "num_input_tokens_seen": 339252330,
      "step": 15729,
      "time_per_iteration": 3.184943199157715
    },
    {
      "auxiliary_loss_clip": 0.01101746,
      "auxiliary_loss_mlp": 0.01025702,
      "balance_loss_clip": 1.0345161,
      "balance_loss_mlp": 1.01356053,
      "epoch": 0.9457387644671577,
      "flos": 18588369179520.0,
      "grad_norm": 1.7444254232454885,
      "language_loss": 0.76555836,
      "learning_rate": 3.077270662890052e-08,
      "loss": 0.78683287,
      "num_input_tokens_seen": 339270325,
      "step": 15730,
      "time_per_iteration": 3.824986457824707
    },
    {
      "auxiliary_loss_clip": 0.01077291,
      "auxiliary_loss_mlp": 0.01031416,
      "balance_loss_clip": 1.0333569,
      "balance_loss_mlp": 1.01865458,
      "epoch": 0.9457988877198257,
      "flos": 21108544237440.0,
      "grad_norm": 1.3949992201121133,
      "language_loss": 0.6256249,
      "learning_rate": 3.070468731536047e-08,
      "loss": 0.64671189,
      "num_input_tokens_seen": 339291980,
      "step": 15731,
      "time_per_iteration": 2.5751113891601562
    },
    {
      "auxiliary_loss_clip": 0.01094253,
      "auxiliary_loss_mlp": 0.0102781,
      "balance_loss_clip": 1.03347921,
      "balance_loss_mlp": 1.01517391,
      "epoch": 0.9458590109724936,
      "flos": 26688379697280.0,
      "grad_norm": 2.089445034923707,
      "language_loss": 0.6409384,
      "learning_rate": 3.063674267769589e-08,
      "loss": 0.66215897,
      "num_input_tokens_seen": 339311795,
      "step": 15732,
      "time_per_iteration": 2.4941647052764893
    },
    {
      "auxiliary_loss_clip": 0.01091561,
      "auxiliary_loss_mlp": 0.01026897,
      "balance_loss_clip": 1.03560674,
      "balance_loss_mlp": 1.01436186,
      "epoch": 0.9459191342251616,
      "flos": 18661591054080.0,
      "grad_norm": 1.8713815485384226,
      "language_loss": 0.84043074,
      "learning_rate": 3.056887271848363e-08,
      "loss": 0.8616153,
      "num_input_tokens_seen": 339327745,
      "step": 15733,
      "time_per_iteration": 2.4687414169311523
    },
    {
      "auxiliary_loss_clip": 0.01090506,
      "auxiliary_loss_mlp": 0.01028001,
      "balance_loss_clip": 1.03424132,
      "balance_loss_mlp": 1.0170753,
      "epoch": 0.9459792574778295,
      "flos": 23398459159680.0,
      "grad_norm": 1.4103824401122265,
      "language_loss": 0.72074938,
      "learning_rate": 3.0501077440297173e-08,
      "loss": 0.74193442,
      "num_input_tokens_seen": 339346445,
      "step": 15734,
      "time_per_iteration": 2.484187364578247
    },
    {
      "auxiliary_loss_clip": 0.01085854,
      "auxiliary_loss_mlp": 0.01030837,
      "balance_loss_clip": 1.03153563,
      "balance_loss_mlp": 1.02082396,
      "epoch": 0.9460393807304975,
      "flos": 24392969994240.0,
      "grad_norm": 1.380010625509751,
      "language_loss": 0.86801255,
      "learning_rate": 3.043335684570692e-08,
      "loss": 0.88917941,
      "num_input_tokens_seen": 339367945,
      "step": 15735,
      "time_per_iteration": 2.5790066719055176
    },
    {
      "auxiliary_loss_clip": 0.01082682,
      "auxiliary_loss_mlp": 0.01029871,
      "balance_loss_clip": 1.03253853,
      "balance_loss_mlp": 1.01836109,
      "epoch": 0.9460995039831654,
      "flos": 21939408708480.0,
      "grad_norm": 1.8727464665118811,
      "language_loss": 0.67166209,
      "learning_rate": 3.036571093728102e-08,
      "loss": 0.69278765,
      "num_input_tokens_seen": 339386060,
      "step": 15736,
      "time_per_iteration": 2.4909276962280273
    },
    {
      "auxiliary_loss_clip": 0.00996498,
      "auxiliary_loss_mlp": 0.01001729,
      "balance_loss_clip": 1.02106452,
      "balance_loss_mlp": 1.00056028,
      "epoch": 0.9461596272358335,
      "flos": 70322466775680.0,
      "grad_norm": 0.8748594616502149,
      "language_loss": 0.65352571,
      "learning_rate": 3.029813971758499e-08,
      "loss": 0.67350799,
      "num_input_tokens_seen": 339446695,
      "step": 15737,
      "time_per_iteration": 3.2265865802764893
    },
    {
      "auxiliary_loss_clip": 0.01017205,
      "auxiliary_loss_mlp": 0.0099943,
      "balance_loss_clip": 1.00786793,
      "balance_loss_mlp": 0.99834484,
      "epoch": 0.9462197504885014,
      "flos": 58591242645120.0,
      "grad_norm": 0.8030137813572151,
      "language_loss": 0.58807778,
      "learning_rate": 3.0230643189181225e-08,
      "loss": 0.60824412,
      "num_input_tokens_seen": 339510080,
      "step": 15738,
      "time_per_iteration": 3.0877442359924316
    },
    {
      "auxiliary_loss_clip": 0.01090008,
      "auxiliary_loss_mlp": 0.01030251,
      "balance_loss_clip": 1.03315616,
      "balance_loss_mlp": 1.01972532,
      "epoch": 0.9462798737411694,
      "flos": 23433759250560.0,
      "grad_norm": 1.8487552725869794,
      "language_loss": 0.71546787,
      "learning_rate": 3.016322135462834e-08,
      "loss": 0.73667049,
      "num_input_tokens_seen": 339529335,
      "step": 15739,
      "time_per_iteration": 2.4843344688415527
    },
    {
      "auxiliary_loss_clip": 0.01090488,
      "auxiliary_loss_mlp": 0.01029991,
      "balance_loss_clip": 1.03335643,
      "balance_loss_mlp": 1.01753926,
      "epoch": 0.9463399969938374,
      "flos": 25046077034880.0,
      "grad_norm": 2.6962968216880907,
      "language_loss": 0.6444968,
      "learning_rate": 3.009587421648363e-08,
      "loss": 0.66570163,
      "num_input_tokens_seen": 339548820,
      "step": 15740,
      "time_per_iteration": 2.53670597076416
    },
    {
      "auxiliary_loss_clip": 0.01077019,
      "auxiliary_loss_mlp": 0.01030061,
      "balance_loss_clip": 1.03384876,
      "balance_loss_mlp": 1.01849151,
      "epoch": 0.9464001202465053,
      "flos": 24352606085760.0,
      "grad_norm": 1.6912422479717186,
      "language_loss": 0.66473526,
      "learning_rate": 3.0028601777301045e-08,
      "loss": 0.68580604,
      "num_input_tokens_seen": 339566775,
      "step": 15741,
      "time_per_iteration": 2.532106876373291
    },
    {
      "auxiliary_loss_clip": 0.01092702,
      "auxiliary_loss_mlp": 0.01025649,
      "balance_loss_clip": 1.03500867,
      "balance_loss_mlp": 1.01390648,
      "epoch": 0.9464602434991733,
      "flos": 17165444832000.0,
      "grad_norm": 2.100812921943838,
      "language_loss": 0.7581352,
      "learning_rate": 2.9961404039630987e-08,
      "loss": 0.77931875,
      "num_input_tokens_seen": 339581905,
      "step": 15742,
      "time_per_iteration": 2.4864628314971924
    },
    {
      "auxiliary_loss_clip": 0.01088591,
      "auxiliary_loss_mlp": 0.0103206,
      "balance_loss_clip": 1.03276706,
      "balance_loss_mlp": 1.02063382,
      "epoch": 0.9465203667518413,
      "flos": 19938107566080.0,
      "grad_norm": 1.754444055567874,
      "language_loss": 0.72317123,
      "learning_rate": 2.989428100602187e-08,
      "loss": 0.74437773,
      "num_input_tokens_seen": 339599870,
      "step": 15743,
      "time_per_iteration": 2.4477202892303467
    },
    {
      "auxiliary_loss_clip": 0.01068589,
      "auxiliary_loss_mlp": 0.01032105,
      "balance_loss_clip": 1.03610587,
      "balance_loss_mlp": 1.01956987,
      "epoch": 0.9465804900045093,
      "flos": 20120318282880.0,
      "grad_norm": 1.7244673826235897,
      "language_loss": 0.79408062,
      "learning_rate": 2.982723267901943e-08,
      "loss": 0.81508756,
      "num_input_tokens_seen": 339620250,
      "step": 15744,
      "time_per_iteration": 2.5766162872314453
    },
    {
      "auxiliary_loss_clip": 0.0108102,
      "auxiliary_loss_mlp": 0.01035632,
      "balance_loss_clip": 1.03528023,
      "balance_loss_mlp": 1.0235796,
      "epoch": 0.9466406132571772,
      "flos": 23911622812800.0,
      "grad_norm": 1.5065310425668796,
      "language_loss": 0.78028268,
      "learning_rate": 2.9760259061165417e-08,
      "loss": 0.80144924,
      "num_input_tokens_seen": 339639900,
      "step": 15745,
      "time_per_iteration": 2.527313232421875
    },
    {
      "auxiliary_loss_clip": 0.01078901,
      "auxiliary_loss_mlp": 0.0103226,
      "balance_loss_clip": 1.03280413,
      "balance_loss_mlp": 1.01977277,
      "epoch": 0.9467007365098452,
      "flos": 19933223316480.0,
      "grad_norm": 1.4522058547884102,
      "language_loss": 0.70195961,
      "learning_rate": 2.9693360155000014e-08,
      "loss": 0.72307122,
      "num_input_tokens_seen": 339658970,
      "step": 15746,
      "time_per_iteration": 2.5212342739105225
    },
    {
      "auxiliary_loss_clip": 0.01077737,
      "auxiliary_loss_mlp": 0.01027942,
      "balance_loss_clip": 1.03483415,
      "balance_loss_mlp": 1.01602054,
      "epoch": 0.9467608597625131,
      "flos": 19310496203520.0,
      "grad_norm": 2.0385207307507205,
      "language_loss": 0.55796456,
      "learning_rate": 2.962653596305964e-08,
      "loss": 0.57902133,
      "num_input_tokens_seen": 339675600,
      "step": 15747,
      "time_per_iteration": 2.4823431968688965
    },
    {
      "auxiliary_loss_clip": 0.0097251,
      "auxiliary_loss_mlp": 0.01004164,
      "balance_loss_clip": 1.00798416,
      "balance_loss_mlp": 1.00297832,
      "epoch": 0.9468209830151811,
      "flos": 69630252802560.0,
      "grad_norm": 0.6916005215148011,
      "language_loss": 0.53206384,
      "learning_rate": 2.955978648787871e-08,
      "loss": 0.55183065,
      "num_input_tokens_seen": 339744505,
      "step": 15748,
      "time_per_iteration": 3.6436448097229004
    },
    {
      "auxiliary_loss_clip": 0.01082177,
      "auxiliary_loss_mlp": 0.01034404,
      "balance_loss_clip": 1.03389287,
      "balance_loss_mlp": 1.02225709,
      "epoch": 0.946881106267849,
      "flos": 27016639113600.0,
      "grad_norm": 1.887762087276275,
      "language_loss": 0.66565061,
      "learning_rate": 2.9493111731988096e-08,
      "loss": 0.68681639,
      "num_input_tokens_seen": 339765810,
      "step": 15749,
      "time_per_iteration": 3.1324243545532227
    },
    {
      "auxiliary_loss_clip": 0.01072641,
      "auxiliary_loss_mlp": 0.01032755,
      "balance_loss_clip": 1.03187168,
      "balance_loss_mlp": 1.01911163,
      "epoch": 0.9469412295205171,
      "flos": 20190092451840.0,
      "grad_norm": 1.8784190224458064,
      "language_loss": 0.76170605,
      "learning_rate": 2.942651169791621e-08,
      "loss": 0.78276002,
      "num_input_tokens_seen": 339784125,
      "step": 15750,
      "time_per_iteration": 2.517817735671997
    },
    {
      "auxiliary_loss_clip": 0.01091571,
      "auxiliary_loss_mlp": 0.01027631,
      "balance_loss_clip": 1.03473592,
      "balance_loss_mlp": 1.01619887,
      "epoch": 0.947001352773185,
      "flos": 21324905809920.0,
      "grad_norm": 1.6924569966836118,
      "language_loss": 0.67803538,
      "learning_rate": 2.9359986388188372e-08,
      "loss": 0.69922733,
      "num_input_tokens_seen": 339803450,
      "step": 15751,
      "time_per_iteration": 2.494255542755127
    },
    {
      "auxiliary_loss_clip": 0.01071779,
      "auxiliary_loss_mlp": 0.01025967,
      "balance_loss_clip": 1.03334391,
      "balance_loss_mlp": 1.01448703,
      "epoch": 0.947061476025853,
      "flos": 21944041562880.0,
      "grad_norm": 1.6092039805946716,
      "language_loss": 0.65082532,
      "learning_rate": 2.929353580532723e-08,
      "loss": 0.67180276,
      "num_input_tokens_seen": 339823215,
      "step": 15752,
      "time_per_iteration": 2.5496826171875
    },
    {
      "auxiliary_loss_clip": 0.01088693,
      "auxiliary_loss_mlp": 0.01035151,
      "balance_loss_clip": 1.03271937,
      "balance_loss_mlp": 1.02193069,
      "epoch": 0.947121599278521,
      "flos": 21394715892480.0,
      "grad_norm": 1.5226950056679822,
      "language_loss": 0.71451735,
      "learning_rate": 2.9227159951852764e-08,
      "loss": 0.7357558,
      "num_input_tokens_seen": 339842230,
      "step": 15753,
      "time_per_iteration": 2.5137991905212402
    },
    {
      "auxiliary_loss_clip": 0.01106016,
      "auxiliary_loss_mlp": 0.01033204,
      "balance_loss_clip": 1.03438771,
      "balance_loss_mlp": 1.01887536,
      "epoch": 0.9471817225311889,
      "flos": 23075730437760.0,
      "grad_norm": 1.8255918560104618,
      "language_loss": 0.70240706,
      "learning_rate": 2.9160858830281855e-08,
      "loss": 0.72379923,
      "num_input_tokens_seen": 339861640,
      "step": 15754,
      "time_per_iteration": 2.4509847164154053
    },
    {
      "auxiliary_loss_clip": 0.01104631,
      "auxiliary_loss_mlp": 0.0102726,
      "balance_loss_clip": 1.03375709,
      "balance_loss_mlp": 1.01569653,
      "epoch": 0.947241845783857,
      "flos": 11910744305280.0,
      "grad_norm": 2.316303844200544,
      "language_loss": 0.78782451,
      "learning_rate": 2.9094632443129153e-08,
      "loss": 0.80914342,
      "num_input_tokens_seen": 339878210,
      "step": 15755,
      "time_per_iteration": 2.4217464923858643
    },
    {
      "auxiliary_loss_clip": 0.01065831,
      "auxiliary_loss_mlp": 0.01037552,
      "balance_loss_clip": 1.03369975,
      "balance_loss_mlp": 1.02297914,
      "epoch": 0.9473019690365249,
      "flos": 20740675098240.0,
      "grad_norm": 3.0974011823095906,
      "language_loss": 0.75649548,
      "learning_rate": 2.9028480792904876e-08,
      "loss": 0.7775293,
      "num_input_tokens_seen": 339894255,
      "step": 15756,
      "time_per_iteration": 2.5079708099365234
    },
    {
      "auxiliary_loss_clip": 0.01078856,
      "auxiliary_loss_mlp": 0.01031626,
      "balance_loss_clip": 1.03126228,
      "balance_loss_mlp": 1.01978242,
      "epoch": 0.9473620922891929,
      "flos": 17639896602240.0,
      "grad_norm": 2.1349843004249514,
      "language_loss": 0.74381775,
      "learning_rate": 2.8962403882118347e-08,
      "loss": 0.76492256,
      "num_input_tokens_seen": 339912425,
      "step": 15757,
      "time_per_iteration": 2.492227554321289
    },
    {
      "auxiliary_loss_clip": 0.01085612,
      "auxiliary_loss_mlp": 0.01030348,
      "balance_loss_clip": 1.034554,
      "balance_loss_mlp": 1.01735401,
      "epoch": 0.9474222155418608,
      "flos": 23550002640000.0,
      "grad_norm": 2.0923234428554967,
      "language_loss": 0.79665959,
      "learning_rate": 2.889640171327512e-08,
      "loss": 0.81781924,
      "num_input_tokens_seen": 339929635,
      "step": 15758,
      "time_per_iteration": 4.468221187591553
    },
    {
      "auxiliary_loss_clip": 0.010799,
      "auxiliary_loss_mlp": 0.00782112,
      "balance_loss_clip": 1.0337739,
      "balance_loss_mlp": 1.00768125,
      "epoch": 0.9474823387945288,
      "flos": 27089753247360.0,
      "grad_norm": 1.6147534683338334,
      "language_loss": 0.71794528,
      "learning_rate": 2.8830474288877638e-08,
      "loss": 0.73656535,
      "num_input_tokens_seen": 339951200,
      "step": 15759,
      "time_per_iteration": 3.9437901973724365
    },
    {
      "auxiliary_loss_clip": 0.01086497,
      "auxiliary_loss_mlp": 0.0102527,
      "balance_loss_clip": 1.03682029,
      "balance_loss_mlp": 1.01512492,
      "epoch": 0.9475424620471967,
      "flos": 22966526113920.0,
      "grad_norm": 1.5694203292181799,
      "language_loss": 0.75798434,
      "learning_rate": 2.8764621611426344e-08,
      "loss": 0.77910197,
      "num_input_tokens_seen": 339971820,
      "step": 15760,
      "time_per_iteration": 2.5118021965026855
    },
    {
      "auxiliary_loss_clip": 0.01102646,
      "auxiliary_loss_mlp": 0.00781882,
      "balance_loss_clip": 1.0350244,
      "balance_loss_mlp": 1.00795794,
      "epoch": 0.9476025852998647,
      "flos": 20047671025920.0,
      "grad_norm": 2.251547318487219,
      "language_loss": 0.7280665,
      "learning_rate": 2.8698843683418128e-08,
      "loss": 0.74691176,
      "num_input_tokens_seen": 339989420,
      "step": 15761,
      "time_per_iteration": 3.952897310256958
    },
    {
      "auxiliary_loss_clip": 0.01081355,
      "auxiliary_loss_mlp": 0.01033563,
      "balance_loss_clip": 1.03774571,
      "balance_loss_mlp": 1.02232134,
      "epoch": 0.9476627085525327,
      "flos": 14975468524800.0,
      "grad_norm": 2.0335139764123893,
      "language_loss": 0.71606266,
      "learning_rate": 2.863314050734722e-08,
      "loss": 0.73721188,
      "num_input_tokens_seen": 340006690,
      "step": 15762,
      "time_per_iteration": 2.475796699523926
    },
    {
      "auxiliary_loss_clip": 0.01105602,
      "auxiliary_loss_mlp": 0.01035418,
      "balance_loss_clip": 1.03388524,
      "balance_loss_mlp": 1.0225141,
      "epoch": 0.9477228318052007,
      "flos": 18697788984960.0,
      "grad_norm": 17.419649784633222,
      "language_loss": 0.66844934,
      "learning_rate": 2.856751208570518e-08,
      "loss": 0.68985957,
      "num_input_tokens_seen": 340025480,
      "step": 15763,
      "time_per_iteration": 2.469226121902466
    },
    {
      "auxiliary_loss_clip": 0.01103081,
      "auxiliary_loss_mlp": 0.01034614,
      "balance_loss_clip": 1.03346968,
      "balance_loss_mlp": 1.02306831,
      "epoch": 0.9477829550578686,
      "flos": 23875065745920.0,
      "grad_norm": 1.8428787967534006,
      "language_loss": 0.69539499,
      "learning_rate": 2.8501958420980466e-08,
      "loss": 0.71677196,
      "num_input_tokens_seen": 340043785,
      "step": 15764,
      "time_per_iteration": 2.463555335998535
    },
    {
      "auxiliary_loss_clip": 0.01092102,
      "auxiliary_loss_mlp": 0.00782408,
      "balance_loss_clip": 1.0372231,
      "balance_loss_mlp": 1.01110387,
      "epoch": 0.9478430783105366,
      "flos": 22562890007040.0,
      "grad_norm": 1.6306230006422098,
      "language_loss": 0.70821917,
      "learning_rate": 2.8436479515659306e-08,
      "loss": 0.72696429,
      "num_input_tokens_seen": 340064360,
      "step": 15765,
      "time_per_iteration": 2.515057325363159
    },
    {
      "auxiliary_loss_clip": 0.01014961,
      "auxiliary_loss_mlp": 0.0100151,
      "balance_loss_clip": 1.0044328,
      "balance_loss_mlp": 1.0004673,
      "epoch": 0.9479032015632046,
      "flos": 60857885554560.0,
      "grad_norm": 0.8096867875937486,
      "language_loss": 0.59125119,
      "learning_rate": 2.8371075372224384e-08,
      "loss": 0.61141586,
      "num_input_tokens_seen": 340114425,
      "step": 15766,
      "time_per_iteration": 2.8709473609924316
    },
    {
      "auxiliary_loss_clip": 0.01059825,
      "auxiliary_loss_mlp": 0.01041781,
      "balance_loss_clip": 1.03516769,
      "balance_loss_mlp": 1.02930605,
      "epoch": 0.9479633248158725,
      "flos": 14683873916160.0,
      "grad_norm": 1.9068129919818604,
      "language_loss": 0.74598032,
      "learning_rate": 2.8305745993155938e-08,
      "loss": 0.76699638,
      "num_input_tokens_seen": 340132200,
      "step": 15767,
      "time_per_iteration": 2.570392370223999
    },
    {
      "auxiliary_loss_clip": 0.01076351,
      "auxiliary_loss_mlp": 0.01036259,
      "balance_loss_clip": 1.0347054,
      "balance_loss_mlp": 1.02363491,
      "epoch": 0.9480234480685406,
      "flos": 20333878594560.0,
      "grad_norm": 2.5117901009127,
      "language_loss": 0.73221081,
      "learning_rate": 2.8240491380931096e-08,
      "loss": 0.75333691,
      "num_input_tokens_seen": 340149175,
      "step": 15768,
      "time_per_iteration": 3.9684693813323975
    },
    {
      "auxiliary_loss_clip": 0.00990687,
      "auxiliary_loss_mlp": 0.01003843,
      "balance_loss_clip": 1.00771284,
      "balance_loss_mlp": 1.00259769,
      "epoch": 0.9480835713212085,
      "flos": 70293092428800.0,
      "grad_norm": 0.7365797676267513,
      "language_loss": 0.55283141,
      "learning_rate": 2.8175311538024326e-08,
      "loss": 0.57277673,
      "num_input_tokens_seen": 340208155,
      "step": 15769,
      "time_per_iteration": 3.199697732925415
    },
    {
      "auxiliary_loss_clip": 0.01063277,
      "auxiliary_loss_mlp": 0.01031014,
      "balance_loss_clip": 1.03361011,
      "balance_loss_mlp": 1.01906943,
      "epoch": 0.9481436945738765,
      "flos": 25449749055360.0,
      "grad_norm": 1.3935907067927642,
      "language_loss": 0.77714282,
      "learning_rate": 2.8110206466907428e-08,
      "loss": 0.79808575,
      "num_input_tokens_seen": 340229275,
      "step": 15770,
      "time_per_iteration": 2.6410505771636963
    },
    {
      "auxiliary_loss_clip": 0.01085832,
      "auxiliary_loss_mlp": 0.01033727,
      "balance_loss_clip": 1.03761542,
      "balance_loss_mlp": 1.02136552,
      "epoch": 0.9482038178265444,
      "flos": 26979902478720.0,
      "grad_norm": 1.9120090013686777,
      "language_loss": 0.79878294,
      "learning_rate": 2.8045176170049313e-08,
      "loss": 0.81997854,
      "num_input_tokens_seen": 340248920,
      "step": 15771,
      "time_per_iteration": 2.541445016860962
    },
    {
      "auxiliary_loss_clip": 0.01069243,
      "auxiliary_loss_mlp": 0.01027975,
      "balance_loss_clip": 1.03213775,
      "balance_loss_mlp": 1.01611328,
      "epoch": 0.9482639410792124,
      "flos": 17785442511360.0,
      "grad_norm": 1.6809017825773926,
      "language_loss": 0.69727194,
      "learning_rate": 2.7980220649915566e-08,
      "loss": 0.71824408,
      "num_input_tokens_seen": 340266775,
      "step": 15772,
      "time_per_iteration": 2.52559232711792
    },
    {
      "auxiliary_loss_clip": 0.01087692,
      "auxiliary_loss_mlp": 0.01032646,
      "balance_loss_clip": 1.03391814,
      "balance_loss_mlp": 1.0206238,
      "epoch": 0.9483240643318803,
      "flos": 20996682307200.0,
      "grad_norm": 1.4819111052292349,
      "language_loss": 0.74030262,
      "learning_rate": 2.7915339908969327e-08,
      "loss": 0.76150602,
      "num_input_tokens_seen": 340285295,
      "step": 15773,
      "time_per_iteration": 2.4753904342651367
    },
    {
      "auxiliary_loss_clip": 0.01070056,
      "auxiliary_loss_mlp": 0.01034057,
      "balance_loss_clip": 1.03426063,
      "balance_loss_mlp": 1.02129006,
      "epoch": 0.9483841875845483,
      "flos": 20083294339200.0,
      "grad_norm": 2.1289643584983184,
      "language_loss": 0.62645435,
      "learning_rate": 2.7850533949671072e-08,
      "loss": 0.64749551,
      "num_input_tokens_seen": 340304265,
      "step": 15774,
      "time_per_iteration": 2.605304002761841
    },
    {
      "auxiliary_loss_clip": 0.01103612,
      "auxiliary_loss_mlp": 0.01033008,
      "balance_loss_clip": 1.03392422,
      "balance_loss_mlp": 1.02044976,
      "epoch": 0.9484443108372163,
      "flos": 20813645577600.0,
      "grad_norm": 2.127633419883079,
      "language_loss": 0.59594101,
      "learning_rate": 2.7785802774478396e-08,
      "loss": 0.61730719,
      "num_input_tokens_seen": 340323690,
      "step": 15775,
      "time_per_iteration": 2.4367942810058594
    },
    {
      "auxiliary_loss_clip": 0.01083336,
      "auxiliary_loss_mlp": 0.0102739,
      "balance_loss_clip": 1.03573775,
      "balance_loss_mlp": 1.01498008,
      "epoch": 0.9485044340898843,
      "flos": 36429184506240.0,
      "grad_norm": 1.5290812488564158,
      "language_loss": 0.61894745,
      "learning_rate": 2.772114638584555e-08,
      "loss": 0.6400547,
      "num_input_tokens_seen": 340345830,
      "step": 15776,
      "time_per_iteration": 2.6305439472198486
    },
    {
      "auxiliary_loss_clip": 0.01075032,
      "auxiliary_loss_mlp": 0.0102926,
      "balance_loss_clip": 1.03209317,
      "balance_loss_mlp": 1.01656437,
      "epoch": 0.9485645573425522,
      "flos": 22602535643520.0,
      "grad_norm": 2.2035920250500864,
      "language_loss": 0.73454541,
      "learning_rate": 2.765656478622458e-08,
      "loss": 0.75558835,
      "num_input_tokens_seen": 340365910,
      "step": 15777,
      "time_per_iteration": 2.507387638092041
    },
    {
      "auxiliary_loss_clip": 0.01101635,
      "auxiliary_loss_mlp": 0.01036509,
      "balance_loss_clip": 1.03639722,
      "balance_loss_mlp": 1.02349138,
      "epoch": 0.9486246805952202,
      "flos": 22017766227840.0,
      "grad_norm": 2.6269423544278965,
      "language_loss": 0.72068989,
      "learning_rate": 2.759205797806441e-08,
      "loss": 0.74207139,
      "num_input_tokens_seen": 340383935,
      "step": 15778,
      "time_per_iteration": 2.4943501949310303
    },
    {
      "auxiliary_loss_clip": 0.01088651,
      "auxiliary_loss_mlp": 0.00782777,
      "balance_loss_clip": 1.03532469,
      "balance_loss_mlp": 1.01131344,
      "epoch": 0.9486848038478882,
      "flos": 16508674604160.0,
      "grad_norm": 1.9398628418742216,
      "language_loss": 0.69901133,
      "learning_rate": 2.7527625963810865e-08,
      "loss": 0.71772569,
      "num_input_tokens_seen": 340402760,
      "step": 15779,
      "time_per_iteration": 2.450079917907715
    },
    {
      "auxiliary_loss_clip": 0.01104353,
      "auxiliary_loss_mlp": 0.01033028,
      "balance_loss_clip": 1.03546238,
      "balance_loss_mlp": 1.02042127,
      "epoch": 0.9487449271005561,
      "flos": 19244385221760.0,
      "grad_norm": 2.773990953204372,
      "language_loss": 0.77919459,
      "learning_rate": 2.7463268745907542e-08,
      "loss": 0.8005684,
      "num_input_tokens_seen": 340422105,
      "step": 15780,
      "time_per_iteration": 2.4414784908294678
    },
    {
      "auxiliary_loss_clip": 0.01076934,
      "auxiliary_loss_mlp": 0.00782652,
      "balance_loss_clip": 1.03608787,
      "balance_loss_mlp": 1.00766611,
      "epoch": 0.9488050503532242,
      "flos": 21762692772480.0,
      "grad_norm": 1.635009343598399,
      "language_loss": 0.66166031,
      "learning_rate": 2.7398986326794494e-08,
      "loss": 0.68025625,
      "num_input_tokens_seen": 340441160,
      "step": 15781,
      "time_per_iteration": 2.516613245010376
    },
    {
      "auxiliary_loss_clip": 0.01101904,
      "auxiliary_loss_mlp": 0.01031454,
      "balance_loss_clip": 1.03462434,
      "balance_loss_mlp": 1.01950324,
      "epoch": 0.9488651736058921,
      "flos": 18368919037440.0,
      "grad_norm": 1.963848462098364,
      "language_loss": 0.79513669,
      "learning_rate": 2.733477870890999e-08,
      "loss": 0.81647027,
      "num_input_tokens_seen": 340458200,
      "step": 15782,
      "time_per_iteration": 2.4435346126556396
    },
    {
      "auxiliary_loss_clip": 0.01019035,
      "auxiliary_loss_mlp": 0.01000743,
      "balance_loss_clip": 1.00636888,
      "balance_loss_mlp": 0.99977189,
      "epoch": 0.9489252968585601,
      "flos": 70084057230720.0,
      "grad_norm": 0.716363737515144,
      "language_loss": 0.59761304,
      "learning_rate": 2.7270645894688082e-08,
      "loss": 0.61781079,
      "num_input_tokens_seen": 340526420,
      "step": 15783,
      "time_per_iteration": 3.183809518814087
    },
    {
      "auxiliary_loss_clip": 0.01092527,
      "auxiliary_loss_mlp": 0.01030129,
      "balance_loss_clip": 1.03387773,
      "balance_loss_mlp": 1.0178268,
      "epoch": 0.948985420111228,
      "flos": 27855440490240.0,
      "grad_norm": 1.5956060253390945,
      "language_loss": 0.7393921,
      "learning_rate": 2.720658788656105e-08,
      "loss": 0.76061869,
      "num_input_tokens_seen": 340546325,
      "step": 15784,
      "time_per_iteration": 2.557420492172241
    },
    {
      "auxiliary_loss_clip": 0.01050927,
      "auxiliary_loss_mlp": 0.01031053,
      "balance_loss_clip": 1.03056836,
      "balance_loss_mlp": 1.01779735,
      "epoch": 0.949045543363896,
      "flos": 24316049018880.0,
      "grad_norm": 1.7840657046010706,
      "language_loss": 0.69835234,
      "learning_rate": 2.714260468695806e-08,
      "loss": 0.71917212,
      "num_input_tokens_seen": 340565145,
      "step": 15785,
      "time_per_iteration": 2.5745930671691895
    },
    {
      "auxiliary_loss_clip": 0.01105164,
      "auxiliary_loss_mlp": 0.01027895,
      "balance_loss_clip": 1.03456783,
      "balance_loss_mlp": 1.01552105,
      "epoch": 0.9491056666165639,
      "flos": 24241677909120.0,
      "grad_norm": 1.5452186974056819,
      "language_loss": 0.75942194,
      "learning_rate": 2.707869629830495e-08,
      "loss": 0.78075254,
      "num_input_tokens_seen": 340585465,
      "step": 15786,
      "time_per_iteration": 2.506511926651001
    },
    {
      "auxiliary_loss_clip": 0.010708,
      "auxiliary_loss_mlp": 0.01028787,
      "balance_loss_clip": 1.03488851,
      "balance_loss_mlp": 1.01764107,
      "epoch": 0.949165789869232,
      "flos": 24531261356160.0,
      "grad_norm": 1.739277643271959,
      "language_loss": 0.78399742,
      "learning_rate": 2.7014862723025335e-08,
      "loss": 0.80499327,
      "num_input_tokens_seen": 340606010,
      "step": 15787,
      "time_per_iteration": 2.56524920463562
    },
    {
      "auxiliary_loss_clip": 0.01093397,
      "auxiliary_loss_mlp": 0.01026869,
      "balance_loss_clip": 1.03729892,
      "balance_loss_mlp": 1.01570535,
      "epoch": 0.9492259131218999,
      "flos": 22235348862720.0,
      "grad_norm": 1.7524077722492615,
      "language_loss": 0.7641176,
      "learning_rate": 2.6951103963540388e-08,
      "loss": 0.78532022,
      "num_input_tokens_seen": 340626135,
      "step": 15788,
      "time_per_iteration": 2.5197675228118896
    },
    {
      "auxiliary_loss_clip": 0.01094168,
      "auxiliary_loss_mlp": 0.0103185,
      "balance_loss_clip": 1.03492022,
      "balance_loss_mlp": 1.0188024,
      "epoch": 0.9492860363745679,
      "flos": 22966310632320.0,
      "grad_norm": 1.6621756550064397,
      "language_loss": 0.71672767,
      "learning_rate": 2.6887420022266848e-08,
      "loss": 0.73798788,
      "num_input_tokens_seen": 340644870,
      "step": 15789,
      "time_per_iteration": 2.474813461303711
    },
    {
      "auxiliary_loss_clip": 0.01064555,
      "auxiliary_loss_mlp": 0.01028326,
      "balance_loss_clip": 1.03513575,
      "balance_loss_mlp": 1.01570129,
      "epoch": 0.9493461596272358,
      "flos": 18370283754240.0,
      "grad_norm": 1.8658891408638403,
      "language_loss": 0.7323308,
      "learning_rate": 2.682381090161989e-08,
      "loss": 0.7532596,
      "num_input_tokens_seen": 340663695,
      "step": 15790,
      "time_per_iteration": 2.545459032058716
    },
    {
      "auxiliary_loss_clip": 0.01068802,
      "auxiliary_loss_mlp": 0.01031988,
      "balance_loss_clip": 1.03255415,
      "balance_loss_mlp": 1.01893413,
      "epoch": 0.9494062828799038,
      "flos": 20011724490240.0,
      "grad_norm": 2.133260734075795,
      "language_loss": 0.77504063,
      "learning_rate": 2.6760276604012033e-08,
      "loss": 0.7960484,
      "num_input_tokens_seen": 340682970,
      "step": 15791,
      "time_per_iteration": 2.531764030456543
    },
    {
      "auxiliary_loss_clip": 0.01095654,
      "auxiliary_loss_mlp": 0.01030896,
      "balance_loss_clip": 1.03395283,
      "balance_loss_mlp": 1.01791978,
      "epoch": 0.9494664061325718,
      "flos": 27228583313280.0,
      "grad_norm": 2.0532876894431706,
      "language_loss": 0.73971725,
      "learning_rate": 2.6696817131852234e-08,
      "loss": 0.76098275,
      "num_input_tokens_seen": 340702275,
      "step": 15792,
      "time_per_iteration": 2.5234501361846924
    },
    {
      "auxiliary_loss_clip": 0.01090907,
      "auxiliary_loss_mlp": 0.01033742,
      "balance_loss_clip": 1.03272128,
      "balance_loss_mlp": 1.02182078,
      "epoch": 0.9495265293852397,
      "flos": 18369816877440.0,
      "grad_norm": 1.7692387623391352,
      "language_loss": 0.77860653,
      "learning_rate": 2.663343248754679e-08,
      "loss": 0.79985297,
      "num_input_tokens_seen": 340719060,
      "step": 15793,
      "time_per_iteration": 2.4378561973571777
    },
    {
      "auxiliary_loss_clip": 0.01076856,
      "auxiliary_loss_mlp": 0.01030473,
      "balance_loss_clip": 1.0357759,
      "balance_loss_mlp": 1.0189755,
      "epoch": 0.9495866526379078,
      "flos": 23075766351360.0,
      "grad_norm": 1.7148787338883655,
      "language_loss": 0.774508,
      "learning_rate": 2.6570122673499562e-08,
      "loss": 0.79558128,
      "num_input_tokens_seen": 340737815,
      "step": 15794,
      "time_per_iteration": 2.529343843460083
    },
    {
      "auxiliary_loss_clip": 0.01072123,
      "auxiliary_loss_mlp": 0.00782725,
      "balance_loss_clip": 1.03257346,
      "balance_loss_mlp": 1.00904763,
      "epoch": 0.9496467758905757,
      "flos": 17529902179200.0,
      "grad_norm": 1.9557021946815252,
      "language_loss": 0.61184525,
      "learning_rate": 2.650688769211107e-08,
      "loss": 0.63039374,
      "num_input_tokens_seen": 340756035,
      "step": 15795,
      "time_per_iteration": 2.5264387130737305
    },
    {
      "auxiliary_loss_clip": 0.01091222,
      "auxiliary_loss_mlp": 0.01034544,
      "balance_loss_clip": 1.03544378,
      "balance_loss_mlp": 1.02170515,
      "epoch": 0.9497068991432437,
      "flos": 24133910129280.0,
      "grad_norm": 1.6492739542053454,
      "language_loss": 0.79071045,
      "learning_rate": 2.644372754577895e-08,
      "loss": 0.81196809,
      "num_input_tokens_seen": 340775620,
      "step": 15796,
      "time_per_iteration": 2.5015182495117188
    },
    {
      "auxiliary_loss_clip": 0.01092511,
      "auxiliary_loss_mlp": 0.01027084,
      "balance_loss_clip": 1.03446949,
      "balance_loss_mlp": 1.01440585,
      "epoch": 0.9497670223959116,
      "flos": 20303319098880.0,
      "grad_norm": 1.90369149528977,
      "language_loss": 0.75256324,
      "learning_rate": 2.6380642236898398e-08,
      "loss": 0.77375919,
      "num_input_tokens_seen": 340794510,
      "step": 15797,
      "time_per_iteration": 3.879220485687256
    },
    {
      "auxiliary_loss_clip": 0.01072187,
      "auxiliary_loss_mlp": 0.00783744,
      "balance_loss_clip": 1.03566492,
      "balance_loss_mlp": 1.00829399,
      "epoch": 0.9498271456485796,
      "flos": 13698916099200.0,
      "grad_norm": 1.8821457456825348,
      "language_loss": 0.65448934,
      "learning_rate": 2.6317631767861727e-08,
      "loss": 0.67304862,
      "num_input_tokens_seen": 340812955,
      "step": 15798,
      "time_per_iteration": 3.9595048427581787
    },
    {
      "auxiliary_loss_clip": 0.01094527,
      "auxiliary_loss_mlp": 0.01031385,
      "balance_loss_clip": 1.03618264,
      "balance_loss_mlp": 1.01994061,
      "epoch": 0.9498872689012475,
      "flos": 20814004713600.0,
      "grad_norm": 1.9530108676336875,
      "language_loss": 0.77377862,
      "learning_rate": 2.6254696141058575e-08,
      "loss": 0.79503769,
      "num_input_tokens_seen": 340829200,
      "step": 15799,
      "time_per_iteration": 2.488452196121216
    },
    {
      "auxiliary_loss_clip": 0.01088951,
      "auxiliary_loss_mlp": 0.01034926,
      "balance_loss_clip": 1.03533864,
      "balance_loss_mlp": 1.02331543,
      "epoch": 0.9499473921539155,
      "flos": 21032700670080.0,
      "grad_norm": 3.6795499738586015,
      "language_loss": 0.70516729,
      "learning_rate": 2.6191835358874814e-08,
      "loss": 0.7264061,
      "num_input_tokens_seen": 340848035,
      "step": 15800,
      "time_per_iteration": 3.903365135192871
    },
    {
      "auxiliary_loss_clip": 0.01076499,
      "auxiliary_loss_mlp": 0.010277,
      "balance_loss_clip": 1.03132749,
      "balance_loss_mlp": 1.01539803,
      "epoch": 0.9500075154065835,
      "flos": 20998693468800.0,
      "grad_norm": 1.6891568899043354,
      "language_loss": 0.71769404,
      "learning_rate": 2.6129049423694315e-08,
      "loss": 0.73873603,
      "num_input_tokens_seen": 340870025,
      "step": 15801,
      "time_per_iteration": 2.544992208480835
    },
    {
      "auxiliary_loss_clip": 0.01092879,
      "auxiliary_loss_mlp": 0.01029321,
      "balance_loss_clip": 1.03455734,
      "balance_loss_mlp": 1.01767445,
      "epoch": 0.9500676386592515,
      "flos": 25121956515840.0,
      "grad_norm": 1.5943915372320756,
      "language_loss": 0.80962557,
      "learning_rate": 2.6066338337898508e-08,
      "loss": 0.83084762,
      "num_input_tokens_seen": 340892290,
      "step": 15802,
      "time_per_iteration": 2.543656587600708
    },
    {
      "auxiliary_loss_clip": 0.0110562,
      "auxiliary_loss_mlp": 0.01032755,
      "balance_loss_clip": 1.03575218,
      "balance_loss_mlp": 1.02093577,
      "epoch": 0.9501277619119194,
      "flos": 27523625627520.0,
      "grad_norm": 1.7350343494385365,
      "language_loss": 0.67290884,
      "learning_rate": 2.60037021038646e-08,
      "loss": 0.6942926,
      "num_input_tokens_seen": 340912260,
      "step": 15803,
      "time_per_iteration": 2.479767322540283
    },
    {
      "auxiliary_loss_clip": 0.01078479,
      "auxiliary_loss_mlp": 0.01034429,
      "balance_loss_clip": 1.03448617,
      "balance_loss_mlp": 1.02234733,
      "epoch": 0.9501878851645874,
      "flos": 20813968800000.0,
      "grad_norm": 1.6663953796416882,
      "language_loss": 0.76063704,
      "learning_rate": 2.5941140723968247e-08,
      "loss": 0.78176612,
      "num_input_tokens_seen": 340928930,
      "step": 15804,
      "time_per_iteration": 2.597602367401123
    },
    {
      "auxiliary_loss_clip": 0.01090275,
      "auxiliary_loss_mlp": 0.01030241,
      "balance_loss_clip": 1.03503191,
      "balance_loss_mlp": 1.0178194,
      "epoch": 0.9502480084172553,
      "flos": 18369385914240.0,
      "grad_norm": 1.7640685878013294,
      "language_loss": 0.73095143,
      "learning_rate": 2.5878654200581775e-08,
      "loss": 0.75215662,
      "num_input_tokens_seen": 340946615,
      "step": 15805,
      "time_per_iteration": 2.445333242416382
    },
    {
      "auxiliary_loss_clip": 0.01087171,
      "auxiliary_loss_mlp": 0.01031127,
      "balance_loss_clip": 1.03746223,
      "balance_loss_mlp": 1.01920581,
      "epoch": 0.9503081316699233,
      "flos": 23549607590400.0,
      "grad_norm": 1.3822869233739026,
      "language_loss": 0.80169916,
      "learning_rate": 2.5816242536074618e-08,
      "loss": 0.82288218,
      "num_input_tokens_seen": 340967545,
      "step": 15806,
      "time_per_iteration": 2.5630276203155518
    },
    {
      "auxiliary_loss_clip": 0.01068085,
      "auxiliary_loss_mlp": 0.01031161,
      "balance_loss_clip": 1.03433514,
      "balance_loss_mlp": 1.01904917,
      "epoch": 0.9503682549225914,
      "flos": 18040444139520.0,
      "grad_norm": 2.069573926869338,
      "language_loss": 0.82449007,
      "learning_rate": 2.5753905732813108e-08,
      "loss": 0.84548247,
      "num_input_tokens_seen": 340984955,
      "step": 15807,
      "time_per_iteration": 3.938669443130493
    },
    {
      "auxiliary_loss_clip": 0.01083546,
      "auxiliary_loss_mlp": 0.01030726,
      "balance_loss_clip": 1.03030133,
      "balance_loss_mlp": 1.01838207,
      "epoch": 0.9504283781752593,
      "flos": 25886135387520.0,
      "grad_norm": 2.1077792846478585,
      "language_loss": 0.71604741,
      "learning_rate": 2.5691643793161355e-08,
      "loss": 0.73719019,
      "num_input_tokens_seen": 341007300,
      "step": 15808,
      "time_per_iteration": 2.5681397914886475
    },
    {
      "auxiliary_loss_clip": 0.01092937,
      "auxiliary_loss_mlp": 0.01027765,
      "balance_loss_clip": 1.03487444,
      "balance_loss_mlp": 1.01583195,
      "epoch": 0.9504885014279273,
      "flos": 22124025636480.0,
      "grad_norm": 1.4303831795732436,
      "language_loss": 0.69550538,
      "learning_rate": 2.562945671948058e-08,
      "loss": 0.71671236,
      "num_input_tokens_seen": 341026695,
      "step": 15809,
      "time_per_iteration": 2.4718220233917236
    },
    {
      "auxiliary_loss_clip": 0.01078901,
      "auxiliary_loss_mlp": 0.01026909,
      "balance_loss_clip": 1.03296113,
      "balance_loss_mlp": 1.01474953,
      "epoch": 0.9505486246805952,
      "flos": 21615961714560.0,
      "grad_norm": 1.4757620464576282,
      "language_loss": 0.75424945,
      "learning_rate": 2.5567344514128452e-08,
      "loss": 0.7753076,
      "num_input_tokens_seen": 341047080,
      "step": 15810,
      "time_per_iteration": 2.5411131381988525
    },
    {
      "auxiliary_loss_clip": 0.01069759,
      "auxiliary_loss_mlp": 0.01039252,
      "balance_loss_clip": 1.03347993,
      "balance_loss_mlp": 1.02671754,
      "epoch": 0.9506087479332632,
      "flos": 22528236360960.0,
      "grad_norm": 1.4659375829473602,
      "language_loss": 0.80065638,
      "learning_rate": 2.5505307179460643e-08,
      "loss": 0.82174647,
      "num_input_tokens_seen": 341067310,
      "step": 15811,
      "time_per_iteration": 2.574913501739502
    },
    {
      "auxiliary_loss_clip": 0.01077242,
      "auxiliary_loss_mlp": 0.01034352,
      "balance_loss_clip": 1.03192711,
      "balance_loss_mlp": 1.02217507,
      "epoch": 0.9506688711859311,
      "flos": 27527360641920.0,
      "grad_norm": 1.912269909632188,
      "language_loss": 0.70114124,
      "learning_rate": 2.5443344717829495e-08,
      "loss": 0.72225714,
      "num_input_tokens_seen": 341085110,
      "step": 15812,
      "time_per_iteration": 2.5552666187286377
    },
    {
      "auxiliary_loss_clip": 0.01066514,
      "auxiliary_loss_mlp": 0.01034858,
      "balance_loss_clip": 1.03332686,
      "balance_loss_mlp": 1.02139354,
      "epoch": 0.9507289944385992,
      "flos": 19865783531520.0,
      "grad_norm": 1.4742370314948827,
      "language_loss": 0.65271592,
      "learning_rate": 2.538145713158446e-08,
      "loss": 0.67372966,
      "num_input_tokens_seen": 341103190,
      "step": 15813,
      "time_per_iteration": 2.548431634902954
    },
    {
      "auxiliary_loss_clip": 0.01094981,
      "auxiliary_loss_mlp": 0.01035052,
      "balance_loss_clip": 1.03522587,
      "balance_loss_mlp": 1.02262449,
      "epoch": 0.9507891176912671,
      "flos": 25193274969600.0,
      "grad_norm": 1.305218582773568,
      "language_loss": 0.70295465,
      "learning_rate": 2.5319644423072327e-08,
      "loss": 0.72425497,
      "num_input_tokens_seen": 341125695,
      "step": 15814,
      "time_per_iteration": 2.519381046295166
    },
    {
      "auxiliary_loss_clip": 0.01089995,
      "auxiliary_loss_mlp": 0.01027313,
      "balance_loss_clip": 1.03401709,
      "balance_loss_mlp": 1.01616061,
      "epoch": 0.9508492409439351,
      "flos": 24899561458560.0,
      "grad_norm": 2.1074367067083917,
      "language_loss": 0.63767576,
      "learning_rate": 2.5257906594637445e-08,
      "loss": 0.65884876,
      "num_input_tokens_seen": 341143930,
      "step": 15815,
      "time_per_iteration": 2.5088648796081543
    },
    {
      "auxiliary_loss_clip": 0.01078321,
      "auxiliary_loss_mlp": 0.01027195,
      "balance_loss_clip": 1.03311479,
      "balance_loss_mlp": 1.01590025,
      "epoch": 0.950909364196603,
      "flos": 29784094375680.0,
      "grad_norm": 1.7431932437583872,
      "language_loss": 0.59083974,
      "learning_rate": 2.519624364862061e-08,
      "loss": 0.61189491,
      "num_input_tokens_seen": 341164280,
      "step": 15816,
      "time_per_iteration": 2.5609776973724365
    },
    {
      "auxiliary_loss_clip": 0.01102305,
      "auxiliary_loss_mlp": 0.01040074,
      "balance_loss_clip": 1.0341413,
      "balance_loss_mlp": 1.02817118,
      "epoch": 0.950969487449271,
      "flos": 24717781704960.0,
      "grad_norm": 1.7748465804071214,
      "language_loss": 0.73405033,
      "learning_rate": 2.513465558735994e-08,
      "loss": 0.75547409,
      "num_input_tokens_seen": 341183670,
      "step": 15817,
      "time_per_iteration": 2.5068750381469727
    },
    {
      "auxiliary_loss_clip": 0.01084349,
      "auxiliary_loss_mlp": 0.01036433,
      "balance_loss_clip": 1.03538394,
      "balance_loss_mlp": 1.02257538,
      "epoch": 0.9510296107019389,
      "flos": 13699167494400.0,
      "grad_norm": 1.9164186357723525,
      "language_loss": 0.5995816,
      "learning_rate": 2.5073142413190918e-08,
      "loss": 0.62078941,
      "num_input_tokens_seen": 341201900,
      "step": 15818,
      "time_per_iteration": 2.4741029739379883
    },
    {
      "auxiliary_loss_clip": 0.01105067,
      "auxiliary_loss_mlp": 0.01036427,
      "balance_loss_clip": 1.03642011,
      "balance_loss_mlp": 1.0236001,
      "epoch": 0.9510897339546069,
      "flos": 17311852667520.0,
      "grad_norm": 1.8253364354015957,
      "language_loss": 0.69215631,
      "learning_rate": 2.5011704128446552e-08,
      "loss": 0.71357119,
      "num_input_tokens_seen": 341218340,
      "step": 15819,
      "time_per_iteration": 2.4231510162353516
    },
    {
      "auxiliary_loss_clip": 0.01064118,
      "auxiliary_loss_mlp": 0.01031541,
      "balance_loss_clip": 1.03698516,
      "balance_loss_mlp": 1.01960266,
      "epoch": 0.951149857207275,
      "flos": 14793940166400.0,
      "grad_norm": 1.8214581283879954,
      "language_loss": 0.74001884,
      "learning_rate": 2.49503407354561e-08,
      "loss": 0.76097548,
      "num_input_tokens_seen": 341235885,
      "step": 15820,
      "time_per_iteration": 2.5565836429595947
    },
    {
      "auxiliary_loss_clip": 0.01084988,
      "auxiliary_loss_mlp": 0.01034102,
      "balance_loss_clip": 1.034729,
      "balance_loss_mlp": 1.02194262,
      "epoch": 0.9512099804599429,
      "flos": 19391152193280.0,
      "grad_norm": 1.9504858846616753,
      "language_loss": 0.78377986,
      "learning_rate": 2.4889052236546804e-08,
      "loss": 0.80497074,
      "num_input_tokens_seen": 341255280,
      "step": 15821,
      "time_per_iteration": 2.508463144302368
    },
    {
      "auxiliary_loss_clip": 0.01062877,
      "auxiliary_loss_mlp": 0.01027577,
      "balance_loss_clip": 1.03298628,
      "balance_loss_mlp": 1.01513779,
      "epoch": 0.9512701037126109,
      "flos": 36757874885760.0,
      "grad_norm": 1.4292536969028966,
      "language_loss": 0.70962906,
      "learning_rate": 2.4827838634042586e-08,
      "loss": 0.7305336,
      "num_input_tokens_seen": 341279055,
      "step": 15822,
      "time_per_iteration": 2.666663885116577
    },
    {
      "auxiliary_loss_clip": 0.01090449,
      "auxiliary_loss_mlp": 0.01034311,
      "balance_loss_clip": 1.03719068,
      "balance_loss_mlp": 1.02296233,
      "epoch": 0.9513302269652788,
      "flos": 22638266697600.0,
      "grad_norm": 1.5444687857150488,
      "language_loss": 0.66176677,
      "learning_rate": 2.47666999302647e-08,
      "loss": 0.68301439,
      "num_input_tokens_seen": 341298560,
      "step": 15823,
      "time_per_iteration": 2.493187189102173
    },
    {
      "auxiliary_loss_clip": 0.0108673,
      "auxiliary_loss_mlp": 0.01030634,
      "balance_loss_clip": 1.03398335,
      "balance_loss_mlp": 1.01888609,
      "epoch": 0.9513903502179468,
      "flos": 22893232412160.0,
      "grad_norm": 1.7554851369351057,
      "language_loss": 0.77110642,
      "learning_rate": 2.4705636127531292e-08,
      "loss": 0.79228008,
      "num_input_tokens_seen": 341316650,
      "step": 15824,
      "time_per_iteration": 2.4776666164398193
    },
    {
      "auxiliary_loss_clip": 0.01106021,
      "auxiliary_loss_mlp": 0.01029615,
      "balance_loss_clip": 1.03446698,
      "balance_loss_mlp": 1.01710403,
      "epoch": 0.9514504734706147,
      "flos": 27928626451200.0,
      "grad_norm": 2.2869579692698,
      "language_loss": 0.73770916,
      "learning_rate": 2.4644647228158065e-08,
      "loss": 0.75906551,
      "num_input_tokens_seen": 341336185,
      "step": 15825,
      "time_per_iteration": 2.5130016803741455
    },
    {
      "auxiliary_loss_clip": 0.0101773,
      "auxiliary_loss_mlp": 0.00999582,
      "balance_loss_clip": 1.00888503,
      "balance_loss_mlp": 0.99839586,
      "epoch": 0.9515105967232828,
      "flos": 67366767312000.0,
      "grad_norm": 0.8399890992317496,
      "language_loss": 0.53441012,
      "learning_rate": 2.458373323445806e-08,
      "loss": 0.55458319,
      "num_input_tokens_seen": 341395795,
      "step": 15826,
      "time_per_iteration": 3.0280754566192627
    },
    {
      "auxiliary_loss_clip": 0.01080712,
      "auxiliary_loss_mlp": 0.01033609,
      "balance_loss_clip": 1.03540111,
      "balance_loss_mlp": 1.02176547,
      "epoch": 0.9515707199759507,
      "flos": 25846525664640.0,
      "grad_norm": 2.0659928324825696,
      "language_loss": 0.72756743,
      "learning_rate": 2.452289414874076e-08,
      "loss": 0.74871063,
      "num_input_tokens_seen": 341415675,
      "step": 15827,
      "time_per_iteration": 2.5618202686309814
    },
    {
      "auxiliary_loss_clip": 0.01083174,
      "auxiliary_loss_mlp": 0.01030645,
      "balance_loss_clip": 1.03549767,
      "balance_loss_mlp": 1.01826525,
      "epoch": 0.9516308432286187,
      "flos": 21828983322240.0,
      "grad_norm": 1.7668195415698418,
      "language_loss": 0.7437458,
      "learning_rate": 2.4462129973313207e-08,
      "loss": 0.764884,
      "num_input_tokens_seen": 341432990,
      "step": 15828,
      "time_per_iteration": 2.486964225769043
    },
    {
      "auxiliary_loss_clip": 0.01069984,
      "auxiliary_loss_mlp": 0.01033766,
      "balance_loss_clip": 1.03561711,
      "balance_loss_mlp": 1.02312076,
      "epoch": 0.9516909664812866,
      "flos": 27269593666560.0,
      "grad_norm": 1.5283464326892506,
      "language_loss": 0.72743958,
      "learning_rate": 2.440144071047978e-08,
      "loss": 0.7484771,
      "num_input_tokens_seen": 341454100,
      "step": 15829,
      "time_per_iteration": 2.602888345718384
    },
    {
      "auxiliary_loss_clip": 0.01089691,
      "auxiliary_loss_mlp": 0.01034147,
      "balance_loss_clip": 1.03351986,
      "balance_loss_mlp": 1.02130222,
      "epoch": 0.9517510897339546,
      "flos": 21215342350080.0,
      "grad_norm": 1.737944048075394,
      "language_loss": 0.61228538,
      "learning_rate": 2.4340826362541533e-08,
      "loss": 0.63352382,
      "num_input_tokens_seen": 341472955,
      "step": 15830,
      "time_per_iteration": 2.461042881011963
    },
    {
      "auxiliary_loss_clip": 0.01089197,
      "auxiliary_loss_mlp": 0.01031792,
      "balance_loss_clip": 1.03527331,
      "balance_loss_mlp": 1.01830971,
      "epoch": 0.9518112129866225,
      "flos": 18733986915840.0,
      "grad_norm": 1.7892146351909157,
      "language_loss": 0.72833723,
      "learning_rate": 2.428028693179729e-08,
      "loss": 0.74954712,
      "num_input_tokens_seen": 341490165,
      "step": 15831,
      "time_per_iteration": 2.463923454284668
    },
    {
      "auxiliary_loss_clip": 0.01057914,
      "auxiliary_loss_mlp": 0.01026096,
      "balance_loss_clip": 1.03246176,
      "balance_loss_mlp": 1.01472378,
      "epoch": 0.9518713362392905,
      "flos": 16763676232320.0,
      "grad_norm": 1.6724690354080216,
      "language_loss": 0.65326363,
      "learning_rate": 2.4219822420542545e-08,
      "loss": 0.67410374,
      "num_input_tokens_seen": 341508055,
      "step": 15832,
      "time_per_iteration": 2.529538631439209
    },
    {
      "auxiliary_loss_clip": 0.01088739,
      "auxiliary_loss_mlp": 0.01035711,
      "balance_loss_clip": 1.03631306,
      "balance_loss_mlp": 1.02290773,
      "epoch": 0.9519314594919586,
      "flos": 15230649720960.0,
      "grad_norm": 1.7442237738598922,
      "language_loss": 0.7818135,
      "learning_rate": 2.4159432831070135e-08,
      "loss": 0.80305803,
      "num_input_tokens_seen": 341526155,
      "step": 15833,
      "time_per_iteration": 2.4611093997955322
    },
    {
      "auxiliary_loss_clip": 0.01065751,
      "auxiliary_loss_mlp": 0.01030628,
      "balance_loss_clip": 1.03425562,
      "balance_loss_mlp": 1.01899922,
      "epoch": 0.9519915827446265,
      "flos": 19352943100800.0,
      "grad_norm": 2.03547465286947,
      "language_loss": 0.74714839,
      "learning_rate": 2.4099118165670007e-08,
      "loss": 0.76811218,
      "num_input_tokens_seen": 341540450,
      "step": 15834,
      "time_per_iteration": 2.512953281402588
    },
    {
      "auxiliary_loss_clip": 0.01092704,
      "auxiliary_loss_mlp": 0.01033601,
      "balance_loss_clip": 1.0371058,
      "balance_loss_mlp": 1.02038622,
      "epoch": 0.9520517059972945,
      "flos": 22266303408000.0,
      "grad_norm": 1.8605002767312375,
      "language_loss": 0.75897944,
      "learning_rate": 2.4038878426629216e-08,
      "loss": 0.78024244,
      "num_input_tokens_seen": 341557865,
      "step": 15835,
      "time_per_iteration": 3.9317095279693604
    },
    {
      "auxiliary_loss_clip": 0.01082221,
      "auxiliary_loss_mlp": 0.01033558,
      "balance_loss_clip": 1.03332758,
      "balance_loss_mlp": 1.02066517,
      "epoch": 0.9521118292499624,
      "flos": 14862313704960.0,
      "grad_norm": 1.8477706336015232,
      "language_loss": 0.66175669,
      "learning_rate": 2.397871361623238e-08,
      "loss": 0.6829145,
      "num_input_tokens_seen": 341573890,
      "step": 15836,
      "time_per_iteration": 3.860098123550415
    },
    {
      "auxiliary_loss_clip": 0.01068095,
      "auxiliary_loss_mlp": 0.01026995,
      "balance_loss_clip": 1.03413701,
      "balance_loss_mlp": 1.01503289,
      "epoch": 0.9521719525026304,
      "flos": 23508812718720.0,
      "grad_norm": 1.5438794261565254,
      "language_loss": 0.7039516,
      "learning_rate": 2.391862373676057e-08,
      "loss": 0.72490251,
      "num_input_tokens_seen": 341593770,
      "step": 15837,
      "time_per_iteration": 2.5573978424072266
    },
    {
      "auxiliary_loss_clip": 0.01104932,
      "auxiliary_loss_mlp": 0.0103389,
      "balance_loss_clip": 1.03435564,
      "balance_loss_mlp": 1.02057481,
      "epoch": 0.9522320757552983,
      "flos": 19714922409600.0,
      "grad_norm": 1.7790885203319187,
      "language_loss": 0.73150539,
      "learning_rate": 2.3858608790492617e-08,
      "loss": 0.75289363,
      "num_input_tokens_seen": 341612065,
      "step": 15838,
      "time_per_iteration": 3.8181777000427246
    },
    {
      "auxiliary_loss_clip": 0.01072382,
      "auxiliary_loss_mlp": 0.01030854,
      "balance_loss_clip": 1.03352904,
      "balance_loss_mlp": 1.01856935,
      "epoch": 0.9522921990079664,
      "flos": 25921291824000.0,
      "grad_norm": 1.824782594850877,
      "language_loss": 0.77885187,
      "learning_rate": 2.379866877970449e-08,
      "loss": 0.79988432,
      "num_input_tokens_seen": 341631365,
      "step": 15839,
      "time_per_iteration": 2.601278066635132
    },
    {
      "auxiliary_loss_clip": 0.01077127,
      "auxiliary_loss_mlp": 0.01030143,
      "balance_loss_clip": 1.0354358,
      "balance_loss_mlp": 1.01850772,
      "epoch": 0.9523523222606343,
      "flos": 19208115463680.0,
      "grad_norm": 1.5957655276427791,
      "language_loss": 0.80614173,
      "learning_rate": 2.3738803706668585e-08,
      "loss": 0.82721442,
      "num_input_tokens_seen": 341650300,
      "step": 15840,
      "time_per_iteration": 2.5174708366394043
    },
    {
      "auxiliary_loss_clip": 0.01073996,
      "auxiliary_loss_mlp": 0.01027763,
      "balance_loss_clip": 1.03477633,
      "balance_loss_mlp": 1.01763582,
      "epoch": 0.9524124455133023,
      "flos": 20921269703040.0,
      "grad_norm": 2.026905134805001,
      "language_loss": 0.73302275,
      "learning_rate": 2.3679013573655314e-08,
      "loss": 0.75404036,
      "num_input_tokens_seen": 341667680,
      "step": 15841,
      "time_per_iteration": 2.5062003135681152
    },
    {
      "auxiliary_loss_clip": 0.01074728,
      "auxiliary_loss_mlp": 0.01025823,
      "balance_loss_clip": 1.03378129,
      "balance_loss_mlp": 1.01464117,
      "epoch": 0.9524725687659702,
      "flos": 18843550375680.0,
      "grad_norm": 1.7390059113760672,
      "language_loss": 0.79188633,
      "learning_rate": 2.3619298382931972e-08,
      "loss": 0.81289178,
      "num_input_tokens_seen": 341685760,
      "step": 15842,
      "time_per_iteration": 2.4788448810577393
    },
    {
      "auxiliary_loss_clip": 0.01081693,
      "auxiliary_loss_mlp": 0.01030391,
      "balance_loss_clip": 1.03671527,
      "balance_loss_mlp": 1.01876187,
      "epoch": 0.9525326920186382,
      "flos": 22674680110080.0,
      "grad_norm": 2.126296297783467,
      "language_loss": 0.7219938,
      "learning_rate": 2.3559658136762973e-08,
      "loss": 0.74311465,
      "num_input_tokens_seen": 341705300,
      "step": 15843,
      "time_per_iteration": 2.5729849338531494
    },
    {
      "auxiliary_loss_clip": 0.01082342,
      "auxiliary_loss_mlp": 0.00783434,
      "balance_loss_clip": 1.0371455,
      "balance_loss_mlp": 1.0100168,
      "epoch": 0.9525928152713061,
      "flos": 22086642556800.0,
      "grad_norm": 1.6287826616386587,
      "language_loss": 0.77995163,
      "learning_rate": 2.3500092837409612e-08,
      "loss": 0.79860938,
      "num_input_tokens_seen": 341724565,
      "step": 15844,
      "time_per_iteration": 2.5378477573394775
    },
    {
      "auxiliary_loss_clip": 0.01069137,
      "auxiliary_loss_mlp": 0.01034432,
      "balance_loss_clip": 1.03267288,
      "balance_loss_mlp": 1.02026939,
      "epoch": 0.9526529385239741,
      "flos": 20704728562560.0,
      "grad_norm": 1.853214113900597,
      "language_loss": 0.69867152,
      "learning_rate": 2.3440602487130977e-08,
      "loss": 0.71970713,
      "num_input_tokens_seen": 341743605,
      "step": 15845,
      "time_per_iteration": 2.5468053817749023
    },
    {
      "auxiliary_loss_clip": 0.01066542,
      "auxiliary_loss_mlp": 0.01032955,
      "balance_loss_clip": 1.03378165,
      "balance_loss_mlp": 1.02136803,
      "epoch": 0.9527130617766422,
      "flos": 23368043318400.0,
      "grad_norm": 1.5392031543414788,
      "language_loss": 0.75443453,
      "learning_rate": 2.338118708818282e-08,
      "loss": 0.77542949,
      "num_input_tokens_seen": 341763475,
      "step": 15846,
      "time_per_iteration": 3.916409730911255
    },
    {
      "auxiliary_loss_clip": 0.01073494,
      "auxiliary_loss_mlp": 0.01026302,
      "balance_loss_clip": 1.03565645,
      "balance_loss_mlp": 1.01427364,
      "epoch": 0.9527731850293101,
      "flos": 18985935888000.0,
      "grad_norm": 1.684247268328158,
      "language_loss": 0.77890539,
      "learning_rate": 2.3321846642817998e-08,
      "loss": 0.79990333,
      "num_input_tokens_seen": 341781265,
      "step": 15847,
      "time_per_iteration": 2.543130397796631
    },
    {
      "auxiliary_loss_clip": 0.01061364,
      "auxiliary_loss_mlp": 0.0103534,
      "balance_loss_clip": 1.03366303,
      "balance_loss_mlp": 1.02431941,
      "epoch": 0.9528333082819781,
      "flos": 19318038059520.0,
      "grad_norm": 3.5864885047640116,
      "language_loss": 0.78213125,
      "learning_rate": 2.326258115328672e-08,
      "loss": 0.80309826,
      "num_input_tokens_seen": 341798825,
      "step": 15848,
      "time_per_iteration": 2.524172306060791
    },
    {
      "auxiliary_loss_clip": 0.01085317,
      "auxiliary_loss_mlp": 0.01040823,
      "balance_loss_clip": 1.03518462,
      "balance_loss_mlp": 1.02757931,
      "epoch": 0.952893431534646,
      "flos": 23951340276480.0,
      "grad_norm": 1.5215903444977577,
      "language_loss": 0.71800828,
      "learning_rate": 2.320339062183674e-08,
      "loss": 0.73926967,
      "num_input_tokens_seen": 341819480,
      "step": 15849,
      "time_per_iteration": 2.534142255783081
    },
    {
      "auxiliary_loss_clip": 0.01098366,
      "auxiliary_loss_mlp": 0.01034677,
      "balance_loss_clip": 1.03641462,
      "balance_loss_mlp": 1.0222255,
      "epoch": 0.952953554787314,
      "flos": 21030545854080.0,
      "grad_norm": 1.5309617850694412,
      "language_loss": 0.7550301,
      "learning_rate": 2.314427505071226e-08,
      "loss": 0.77636051,
      "num_input_tokens_seen": 341838035,
      "step": 15850,
      "time_per_iteration": 2.4764444828033447
    },
    {
      "auxiliary_loss_clip": 0.01075817,
      "auxiliary_loss_mlp": 0.01031981,
      "balance_loss_clip": 1.03404403,
      "balance_loss_mlp": 1.02073348,
      "epoch": 0.9530136780399819,
      "flos": 22382870019840.0,
      "grad_norm": 2.0774587272627767,
      "language_loss": 0.72468954,
      "learning_rate": 2.308523444215482e-08,
      "loss": 0.74576759,
      "num_input_tokens_seen": 341855895,
      "step": 15851,
      "time_per_iteration": 2.521024465560913
    },
    {
      "auxiliary_loss_clip": 0.0107581,
      "auxiliary_loss_mlp": 0.01026624,
      "balance_loss_clip": 1.03302526,
      "balance_loss_mlp": 1.01491761,
      "epoch": 0.95307380129265,
      "flos": 22159613036160.0,
      "grad_norm": 1.6747515749976736,
      "language_loss": 0.79956126,
      "learning_rate": 2.3026268798403525e-08,
      "loss": 0.82058561,
      "num_input_tokens_seen": 341875240,
      "step": 15852,
      "time_per_iteration": 2.5138373374938965
    },
    {
      "auxiliary_loss_clip": 0.01093221,
      "auxiliary_loss_mlp": 0.01031401,
      "balance_loss_clip": 1.03443575,
      "balance_loss_mlp": 1.01918232,
      "epoch": 0.9531339245453179,
      "flos": 44022747214080.0,
      "grad_norm": 1.430628164698909,
      "language_loss": 0.59536552,
      "learning_rate": 2.2967378121694138e-08,
      "loss": 0.61661172,
      "num_input_tokens_seen": 341901020,
      "step": 15853,
      "time_per_iteration": 2.686136484146118
    },
    {
      "auxiliary_loss_clip": 0.01079239,
      "auxiliary_loss_mlp": 0.01027675,
      "balance_loss_clip": 1.03298354,
      "balance_loss_mlp": 1.01639199,
      "epoch": 0.9531940477979859,
      "flos": 20266690204800.0,
      "grad_norm": 1.7878408677228705,
      "language_loss": 0.72778761,
      "learning_rate": 2.290856241425998e-08,
      "loss": 0.74885672,
      "num_input_tokens_seen": 341919365,
      "step": 15854,
      "time_per_iteration": 2.475715398788452
    },
    {
      "auxiliary_loss_clip": 0.01077164,
      "auxiliary_loss_mlp": 0.01030871,
      "balance_loss_clip": 1.03256917,
      "balance_loss_mlp": 1.01902783,
      "epoch": 0.9532541710506538,
      "flos": 25335732309120.0,
      "grad_norm": 2.0851841522624843,
      "language_loss": 0.67564768,
      "learning_rate": 2.284982167833127e-08,
      "loss": 0.69672805,
      "num_input_tokens_seen": 341939985,
      "step": 15855,
      "time_per_iteration": 2.5620648860931396
    },
    {
      "auxiliary_loss_clip": 0.01103052,
      "auxiliary_loss_mlp": 0.01029504,
      "balance_loss_clip": 1.03388608,
      "balance_loss_mlp": 1.01826262,
      "epoch": 0.9533142943033218,
      "flos": 26469288691200.0,
      "grad_norm": 1.629506521833898,
      "language_loss": 0.76389605,
      "learning_rate": 2.279115591613556e-08,
      "loss": 0.78522164,
      "num_input_tokens_seen": 341959255,
      "step": 15856,
      "time_per_iteration": 2.485079526901245
    },
    {
      "auxiliary_loss_clip": 0.01077015,
      "auxiliary_loss_mlp": 0.0103178,
      "balance_loss_clip": 1.03159678,
      "balance_loss_mlp": 1.0204494,
      "epoch": 0.9533744175559897,
      "flos": 23656944407040.0,
      "grad_norm": 1.5507016478038937,
      "language_loss": 0.77926612,
      "learning_rate": 2.2732565129897075e-08,
      "loss": 0.80035406,
      "num_input_tokens_seen": 341977205,
      "step": 15857,
      "time_per_iteration": 2.5377187728881836
    },
    {
      "auxiliary_loss_clip": 0.01021847,
      "auxiliary_loss_mlp": 0.01000092,
      "balance_loss_clip": 1.00910711,
      "balance_loss_mlp": 0.99911445,
      "epoch": 0.9534345408086577,
      "flos": 61052055500160.0,
      "grad_norm": 0.7179338411891907,
      "language_loss": 0.62578142,
      "learning_rate": 2.267404932183803e-08,
      "loss": 0.64600074,
      "num_input_tokens_seen": 342038545,
      "step": 15858,
      "time_per_iteration": 3.069474697113037
    },
    {
      "auxiliary_loss_clip": 0.01053307,
      "auxiliary_loss_mlp": 0.01027174,
      "balance_loss_clip": 1.03219891,
      "balance_loss_mlp": 1.01583695,
      "epoch": 0.9534946640613258,
      "flos": 18951677291520.0,
      "grad_norm": 1.5408877475956821,
      "language_loss": 0.56461173,
      "learning_rate": 2.2615608494177097e-08,
      "loss": 0.58541656,
      "num_input_tokens_seen": 342058195,
      "step": 15859,
      "time_per_iteration": 2.5668094158172607
    },
    {
      "auxiliary_loss_clip": 0.01098456,
      "auxiliary_loss_mlp": 0.01026662,
      "balance_loss_clip": 1.03307772,
      "balance_loss_mlp": 1.01584387,
      "epoch": 0.9535547873139937,
      "flos": 16654292340480.0,
      "grad_norm": 2.0534819275266556,
      "language_loss": 0.81446421,
      "learning_rate": 2.2557242649130504e-08,
      "loss": 0.83571541,
      "num_input_tokens_seen": 342075025,
      "step": 15860,
      "time_per_iteration": 2.4284727573394775
    },
    {
      "auxiliary_loss_clip": 0.01052393,
      "auxiliary_loss_mlp": 0.00781612,
      "balance_loss_clip": 1.03175092,
      "balance_loss_mlp": 1.00676179,
      "epoch": 0.9536149105666617,
      "flos": 20667776446080.0,
      "grad_norm": 1.7067169191048637,
      "language_loss": 0.66800481,
      "learning_rate": 2.249895178891159e-08,
      "loss": 0.68634486,
      "num_input_tokens_seen": 342094595,
      "step": 15861,
      "time_per_iteration": 2.5683600902557373
    },
    {
      "auxiliary_loss_clip": 0.0109303,
      "auxiliary_loss_mlp": 0.01035807,
      "balance_loss_clip": 1.03399515,
      "balance_loss_mlp": 1.02354026,
      "epoch": 0.9536750338193296,
      "flos": 30700499086080.0,
      "grad_norm": 1.5597929271137883,
      "language_loss": 0.65243548,
      "learning_rate": 2.244073591573037e-08,
      "loss": 0.67372382,
      "num_input_tokens_seen": 342115970,
      "step": 15862,
      "time_per_iteration": 2.5532166957855225
    },
    {
      "auxiliary_loss_clip": 0.01064013,
      "auxiliary_loss_mlp": 0.01030927,
      "balance_loss_clip": 1.03301215,
      "balance_loss_mlp": 1.01950121,
      "epoch": 0.9537351570719976,
      "flos": 20405484357120.0,
      "grad_norm": 1.5416348558810435,
      "language_loss": 0.67325675,
      "learning_rate": 2.238259503179485e-08,
      "loss": 0.69420612,
      "num_input_tokens_seen": 342134080,
      "step": 15863,
      "time_per_iteration": 2.5403642654418945
    },
    {
      "auxiliary_loss_clip": 0.01081588,
      "auxiliary_loss_mlp": 0.01031318,
      "balance_loss_clip": 1.03234744,
      "balance_loss_mlp": 1.01911139,
      "epoch": 0.9537952803246655,
      "flos": 29929245235200.0,
      "grad_norm": 1.9944465333130657,
      "language_loss": 0.78469139,
      "learning_rate": 2.2324529139309267e-08,
      "loss": 0.80582047,
      "num_input_tokens_seen": 342154725,
      "step": 15864,
      "time_per_iteration": 2.572067975997925
    },
    {
      "auxiliary_loss_clip": 0.01071204,
      "auxiliary_loss_mlp": 0.01027463,
      "balance_loss_clip": 1.03412437,
      "balance_loss_mlp": 1.01587546,
      "epoch": 0.9538554035773336,
      "flos": 20521404524160.0,
      "grad_norm": 1.864845933610587,
      "language_loss": 0.59606862,
      "learning_rate": 2.226653824047586e-08,
      "loss": 0.6170553,
      "num_input_tokens_seen": 342172275,
      "step": 15865,
      "time_per_iteration": 2.5102875232696533
    },
    {
      "auxiliary_loss_clip": 0.01062242,
      "auxiliary_loss_mlp": 0.01033772,
      "balance_loss_clip": 1.03378201,
      "balance_loss_mlp": 1.02046847,
      "epoch": 0.9539155268300015,
      "flos": 18406517598720.0,
      "grad_norm": 1.6257306941614837,
      "language_loss": 0.70337856,
      "learning_rate": 2.2208622337493765e-08,
      "loss": 0.72433871,
      "num_input_tokens_seen": 342190880,
      "step": 15866,
      "time_per_iteration": 2.525197744369507
    },
    {
      "auxiliary_loss_clip": 0.01079358,
      "auxiliary_loss_mlp": 0.01036436,
      "balance_loss_clip": 1.03286862,
      "balance_loss_mlp": 1.0227567,
      "epoch": 0.9539756500826695,
      "flos": 26213281482240.0,
      "grad_norm": 2.977823540106525,
      "language_loss": 0.8506614,
      "learning_rate": 2.215078143255855e-08,
      "loss": 0.87181932,
      "num_input_tokens_seen": 342208165,
      "step": 15867,
      "time_per_iteration": 2.534639835357666
    },
    {
      "auxiliary_loss_clip": 0.01017152,
      "auxiliary_loss_mlp": 0.01001325,
      "balance_loss_clip": 1.00760841,
      "balance_loss_mlp": 1.00022817,
      "epoch": 0.9540357733353374,
      "flos": 68289097766400.0,
      "grad_norm": 0.7521506106406901,
      "language_loss": 0.61824012,
      "learning_rate": 2.2093015527864024e-08,
      "loss": 0.63842487,
      "num_input_tokens_seen": 342277110,
      "step": 15868,
      "time_per_iteration": 3.135977029800415
    },
    {
      "auxiliary_loss_clip": 0.01067512,
      "auxiliary_loss_mlp": 0.01024871,
      "balance_loss_clip": 1.03514481,
      "balance_loss_mlp": 1.01286054,
      "epoch": 0.9540958965880054,
      "flos": 21288276915840.0,
      "grad_norm": 1.9173589722002555,
      "language_loss": 0.6003319,
      "learning_rate": 2.2035324625600425e-08,
      "loss": 0.62125576,
      "num_input_tokens_seen": 342294695,
      "step": 15869,
      "time_per_iteration": 2.5337462425231934
    },
    {
      "auxiliary_loss_clip": 0.01066168,
      "auxiliary_loss_mlp": 0.0078335,
      "balance_loss_clip": 1.03346527,
      "balance_loss_mlp": 1.00693035,
      "epoch": 0.9541560198406733,
      "flos": 19751407649280.0,
      "grad_norm": 1.6001316371746634,
      "language_loss": 0.71192479,
      "learning_rate": 2.197770872795579e-08,
      "loss": 0.73041999,
      "num_input_tokens_seen": 342314970,
      "step": 15870,
      "time_per_iteration": 2.582280158996582
    },
    {
      "auxiliary_loss_clip": 0.01063737,
      "auxiliary_loss_mlp": 0.01028915,
      "balance_loss_clip": 1.03230786,
      "balance_loss_mlp": 1.0168395,
      "epoch": 0.9542161430933414,
      "flos": 24715626888960.0,
      "grad_norm": 2.042693649949192,
      "language_loss": 0.77122903,
      "learning_rate": 2.1920167837114368e-08,
      "loss": 0.79215556,
      "num_input_tokens_seen": 342334255,
      "step": 15871,
      "time_per_iteration": 2.5413522720336914
    },
    {
      "auxiliary_loss_clip": 0.01092505,
      "auxiliary_loss_mlp": 0.01031137,
      "balance_loss_clip": 1.03414178,
      "balance_loss_mlp": 1.01866817,
      "epoch": 0.9542762663460094,
      "flos": 31065818359680.0,
      "grad_norm": 3.153754606355243,
      "language_loss": 0.57876813,
      "learning_rate": 2.1862701955258634e-08,
      "loss": 0.60000461,
      "num_input_tokens_seen": 342354730,
      "step": 15872,
      "time_per_iteration": 2.560175657272339
    },
    {
      "auxiliary_loss_clip": 0.01081953,
      "auxiliary_loss_mlp": 0.01033508,
      "balance_loss_clip": 1.03425765,
      "balance_loss_mlp": 1.02003741,
      "epoch": 0.9543363895986773,
      "flos": 20776729374720.0,
      "grad_norm": 1.4675918810490436,
      "language_loss": 0.74670589,
      "learning_rate": 2.1805311084567514e-08,
      "loss": 0.76786047,
      "num_input_tokens_seen": 342374565,
      "step": 15873,
      "time_per_iteration": 2.512266159057617
    },
    {
      "auxiliary_loss_clip": 0.01104429,
      "auxiliary_loss_mlp": 0.0103235,
      "balance_loss_clip": 1.03521252,
      "balance_loss_mlp": 1.01951158,
      "epoch": 0.9543965128513453,
      "flos": 24462744163200.0,
      "grad_norm": 1.7975136243037892,
      "language_loss": 0.62555313,
      "learning_rate": 2.1747995227217265e-08,
      "loss": 0.64692092,
      "num_input_tokens_seen": 342394590,
      "step": 15874,
      "time_per_iteration": 3.8409314155578613
    },
    {
      "auxiliary_loss_clip": 0.01078685,
      "auxiliary_loss_mlp": 0.01035007,
      "balance_loss_clip": 1.03318107,
      "balance_loss_mlp": 1.02262688,
      "epoch": 0.9544566361040132,
      "flos": 15261532439040.0,
      "grad_norm": 1.9713664730112697,
      "language_loss": 0.89525914,
      "learning_rate": 2.169075438538104e-08,
      "loss": 0.91639602,
      "num_input_tokens_seen": 342410445,
      "step": 15875,
      "time_per_iteration": 3.8611690998077393
    },
    {
      "auxiliary_loss_clip": 0.0110697,
      "auxiliary_loss_mlp": 0.01030486,
      "balance_loss_clip": 1.03516388,
      "balance_loss_mlp": 1.01774907,
      "epoch": 0.9545167593566812,
      "flos": 25918777872000.0,
      "grad_norm": 1.864892585422974,
      "language_loss": 0.67928803,
      "learning_rate": 2.1633588561229765e-08,
      "loss": 0.70066261,
      "num_input_tokens_seen": 342430970,
      "step": 15876,
      "time_per_iteration": 3.866117000579834
    },
    {
      "auxiliary_loss_clip": 0.01092498,
      "auxiliary_loss_mlp": 0.01033182,
      "balance_loss_clip": 1.03468013,
      "balance_loss_mlp": 1.02047372,
      "epoch": 0.9545768826093491,
      "flos": 25628188844160.0,
      "grad_norm": 2.366582710257036,
      "language_loss": 0.69171625,
      "learning_rate": 2.1576497756931267e-08,
      "loss": 0.71297306,
      "num_input_tokens_seen": 342449505,
      "step": 15877,
      "time_per_iteration": 2.5050644874572754
    },
    {
      "auxiliary_loss_clip": 0.01062309,
      "auxiliary_loss_mlp": 0.01030773,
      "balance_loss_clip": 1.03207695,
      "balance_loss_mlp": 1.01770186,
      "epoch": 0.9546370058620172,
      "flos": 22491499726080.0,
      "grad_norm": 1.6313240088401524,
      "language_loss": 0.70724195,
      "learning_rate": 2.1519481974650035e-08,
      "loss": 0.72817278,
      "num_input_tokens_seen": 342470390,
      "step": 15878,
      "time_per_iteration": 2.5530529022216797
    },
    {
      "auxiliary_loss_clip": 0.01099684,
      "auxiliary_loss_mlp": 0.01027372,
      "balance_loss_clip": 1.03297305,
      "balance_loss_mlp": 1.01532543,
      "epoch": 0.9546971291146851,
      "flos": 24609582961920.0,
      "grad_norm": 1.3371619427169126,
      "language_loss": 0.68147421,
      "learning_rate": 2.1462541216548335e-08,
      "loss": 0.70274472,
      "num_input_tokens_seen": 342492560,
      "step": 15879,
      "time_per_iteration": 2.4686648845672607
    },
    {
      "auxiliary_loss_clip": 0.01067101,
      "auxiliary_loss_mlp": 0.00781001,
      "balance_loss_clip": 1.03228498,
      "balance_loss_mlp": 1.00807786,
      "epoch": 0.9547572523673531,
      "flos": 28657756627200.0,
      "grad_norm": 2.0183362548772665,
      "language_loss": 0.84688979,
      "learning_rate": 2.1405675484785334e-08,
      "loss": 0.86537087,
      "num_input_tokens_seen": 342512315,
      "step": 15880,
      "time_per_iteration": 2.5973219871520996
    },
    {
      "auxiliary_loss_clip": 0.01037747,
      "auxiliary_loss_mlp": 0.01030003,
      "balance_loss_clip": 1.03145993,
      "balance_loss_mlp": 1.01675344,
      "epoch": 0.954817375620021,
      "flos": 33802606385280.0,
      "grad_norm": 1.7056098341307906,
      "language_loss": 0.72041547,
      "learning_rate": 2.134888478151753e-08,
      "loss": 0.74109292,
      "num_input_tokens_seen": 342533060,
      "step": 15881,
      "time_per_iteration": 2.7903385162353516
    },
    {
      "auxiliary_loss_clip": 0.01091244,
      "auxiliary_loss_mlp": 0.01032136,
      "balance_loss_clip": 1.0354346,
      "balance_loss_mlp": 1.02042985,
      "epoch": 0.954877498872689,
      "flos": 14428225843200.0,
      "grad_norm": 1.8639762982942607,
      "language_loss": 0.71368051,
      "learning_rate": 2.1292169108898083e-08,
      "loss": 0.7349143,
      "num_input_tokens_seen": 342550830,
      "step": 15882,
      "time_per_iteration": 2.7413032054901123
    },
    {
      "auxiliary_loss_clip": 0.01083643,
      "auxiliary_loss_mlp": 0.01029096,
      "balance_loss_clip": 1.03508282,
      "balance_loss_mlp": 1.01758611,
      "epoch": 0.9549376221253569,
      "flos": 59269447336320.0,
      "grad_norm": 1.8219019145363355,
      "language_loss": 0.65962875,
      "learning_rate": 2.1235528469078168e-08,
      "loss": 0.68075609,
      "num_input_tokens_seen": 342575070,
      "step": 15883,
      "time_per_iteration": 2.83160400390625
    },
    {
      "auxiliary_loss_clip": 0.0109444,
      "auxiliary_loss_mlp": 0.01030277,
      "balance_loss_clip": 1.03673887,
      "balance_loss_mlp": 1.0175333,
      "epoch": 0.954997745378025,
      "flos": 17274397760640.0,
      "grad_norm": 2.32612362158497,
      "language_loss": 0.77854931,
      "learning_rate": 2.1178962864205175e-08,
      "loss": 0.79979646,
      "num_input_tokens_seen": 342592215,
      "step": 15884,
      "time_per_iteration": 2.460235834121704
    },
    {
      "auxiliary_loss_clip": 0.01104263,
      "auxiliary_loss_mlp": 0.01027465,
      "balance_loss_clip": 1.03318357,
      "balance_loss_mlp": 1.014853,
      "epoch": 0.955057868630693,
      "flos": 13006378903680.0,
      "grad_norm": 1.821064364021651,
      "language_loss": 0.7764082,
      "learning_rate": 2.1122472296424054e-08,
      "loss": 0.79772544,
      "num_input_tokens_seen": 342610030,
      "step": 15885,
      "time_per_iteration": 3.8125693798065186
    },
    {
      "auxiliary_loss_clip": 0.01103807,
      "auxiliary_loss_mlp": 0.01032325,
      "balance_loss_clip": 1.03369856,
      "balance_loss_mlp": 1.0206306,
      "epoch": 0.9551179918833609,
      "flos": 22637692080000.0,
      "grad_norm": 1.9843598840079646,
      "language_loss": 0.70143163,
      "learning_rate": 2.1066056767877317e-08,
      "loss": 0.72279292,
      "num_input_tokens_seen": 342626475,
      "step": 15886,
      "time_per_iteration": 2.4310245513916016
    },
    {
      "auxiliary_loss_clip": 0.01072698,
      "auxiliary_loss_mlp": 0.01031065,
      "balance_loss_clip": 1.0364337,
      "balance_loss_mlp": 1.01746917,
      "epoch": 0.9551781151360289,
      "flos": 21542811667200.0,
      "grad_norm": 2.287701414637279,
      "language_loss": 0.72758663,
      "learning_rate": 2.1009716280703916e-08,
      "loss": 0.74862427,
      "num_input_tokens_seen": 342646645,
      "step": 15887,
      "time_per_iteration": 2.552844524383545
    },
    {
      "auxiliary_loss_clip": 0.01078765,
      "auxiliary_loss_mlp": 0.01029282,
      "balance_loss_clip": 1.03237367,
      "balance_loss_mlp": 1.01773071,
      "epoch": 0.9552382383886968,
      "flos": 20702250524160.0,
      "grad_norm": 1.9453588122248182,
      "language_loss": 0.56819957,
      "learning_rate": 2.0953450837040364e-08,
      "loss": 0.58928001,
      "num_input_tokens_seen": 342663615,
      "step": 15888,
      "time_per_iteration": 2.4864511489868164
    },
    {
      "auxiliary_loss_clip": 0.01017199,
      "auxiliary_loss_mlp": 0.0100418,
      "balance_loss_clip": 1.00485325,
      "balance_loss_mlp": 1.00310099,
      "epoch": 0.9552983616413648,
      "flos": 67769792887680.0,
      "grad_norm": 0.7077151954024432,
      "language_loss": 0.57893729,
      "learning_rate": 2.0897260439020514e-08,
      "loss": 0.59915113,
      "num_input_tokens_seen": 342728275,
      "step": 15889,
      "time_per_iteration": 3.1545443534851074
    },
    {
      "auxiliary_loss_clip": 0.01103646,
      "auxiliary_loss_mlp": 0.01028397,
      "balance_loss_clip": 1.03306675,
      "balance_loss_mlp": 1.01640463,
      "epoch": 0.9553584848940327,
      "flos": 21579979265280.0,
      "grad_norm": 1.4507435544215574,
      "language_loss": 0.66792548,
      "learning_rate": 2.084114508877466e-08,
      "loss": 0.68924588,
      "num_input_tokens_seen": 342748860,
      "step": 15890,
      "time_per_iteration": 2.449159622192383
    },
    {
      "auxiliary_loss_clip": 0.01103782,
      "auxiliary_loss_mlp": 0.01030626,
      "balance_loss_clip": 1.03540564,
      "balance_loss_mlp": 1.01917005,
      "epoch": 0.9554186081467008,
      "flos": 24208173498240.0,
      "grad_norm": 1.6937172853886684,
      "language_loss": 0.74241722,
      "learning_rate": 2.0785104788430874e-08,
      "loss": 0.76376128,
      "num_input_tokens_seen": 342769705,
      "step": 15891,
      "time_per_iteration": 2.4783828258514404
    },
    {
      "auxiliary_loss_clip": 0.01066398,
      "auxiliary_loss_mlp": 0.01030527,
      "balance_loss_clip": 1.03277802,
      "balance_loss_mlp": 1.01985824,
      "epoch": 0.9554787313993687,
      "flos": 16251554073600.0,
      "grad_norm": 2.033006488381614,
      "language_loss": 0.77910948,
      "learning_rate": 2.072913954011435e-08,
      "loss": 0.80007875,
      "num_input_tokens_seen": 342787000,
      "step": 15892,
      "time_per_iteration": 2.50307559967041
    },
    {
      "auxiliary_loss_clip": 0.01102123,
      "auxiliary_loss_mlp": 0.01031465,
      "balance_loss_clip": 1.0344274,
      "balance_loss_mlp": 1.01887679,
      "epoch": 0.9555388546520367,
      "flos": 23404133508480.0,
      "grad_norm": 1.4905786309427618,
      "language_loss": 0.69698453,
      "learning_rate": 2.0673249345947386e-08,
      "loss": 0.71832037,
      "num_input_tokens_seen": 342807795,
      "step": 15893,
      "time_per_iteration": 2.483534812927246
    },
    {
      "auxiliary_loss_clip": 0.01082959,
      "auxiliary_loss_mlp": 0.00783337,
      "balance_loss_clip": 1.0361371,
      "balance_loss_mlp": 1.01002085,
      "epoch": 0.9555989779047046,
      "flos": 14794047907200.0,
      "grad_norm": 3.122639449341315,
      "language_loss": 0.65683615,
      "learning_rate": 2.0617434208048955e-08,
      "loss": 0.67549914,
      "num_input_tokens_seen": 342825490,
      "step": 15894,
      "time_per_iteration": 2.4751827716827393
    },
    {
      "auxiliary_loss_clip": 0.01093545,
      "auxiliary_loss_mlp": 0.01028845,
      "balance_loss_clip": 1.03434062,
      "balance_loss_mlp": 1.01622128,
      "epoch": 0.9556591011573726,
      "flos": 22236749493120.0,
      "grad_norm": 1.7107317071260963,
      "language_loss": 0.81595016,
      "learning_rate": 2.056169412853581e-08,
      "loss": 0.83717412,
      "num_input_tokens_seen": 342844965,
      "step": 15895,
      "time_per_iteration": 2.5018489360809326
    },
    {
      "auxiliary_loss_clip": 0.01077588,
      "auxiliary_loss_mlp": 0.01030539,
      "balance_loss_clip": 1.03431439,
      "balance_loss_mlp": 1.01849937,
      "epoch": 0.9557192244100405,
      "flos": 27855296835840.0,
      "grad_norm": 1.5484988346562092,
      "language_loss": 0.72365338,
      "learning_rate": 2.0506029109521593e-08,
      "loss": 0.74473459,
      "num_input_tokens_seen": 342865915,
      "step": 15896,
      "time_per_iteration": 2.5496573448181152
    },
    {
      "auxiliary_loss_clip": 0.01100199,
      "auxiliary_loss_mlp": 0.01027385,
      "balance_loss_clip": 1.03295565,
      "balance_loss_mlp": 1.0159111,
      "epoch": 0.9557793476627086,
      "flos": 17602800831360.0,
      "grad_norm": 2.0959852451371224,
      "language_loss": 0.79495335,
      "learning_rate": 2.045043915311706e-08,
      "loss": 0.81622916,
      "num_input_tokens_seen": 342884000,
      "step": 15897,
      "time_per_iteration": 2.4810736179351807
    },
    {
      "auxiliary_loss_clip": 0.01078248,
      "auxiliary_loss_mlp": 0.01030908,
      "balance_loss_clip": 1.03218579,
      "balance_loss_mlp": 1.01789045,
      "epoch": 0.9558394709153766,
      "flos": 23875496709120.0,
      "grad_norm": 1.5723189149753873,
      "language_loss": 0.72778618,
      "learning_rate": 2.03949242614303e-08,
      "loss": 0.7488777,
      "num_input_tokens_seen": 342903095,
      "step": 15898,
      "time_per_iteration": 2.531891107559204
    },
    {
      "auxiliary_loss_clip": 0.0100224,
      "auxiliary_loss_mlp": 0.01003129,
      "balance_loss_clip": 1.01006377,
      "balance_loss_mlp": 1.00213385,
      "epoch": 0.9558995941680445,
      "flos": 53682001171200.0,
      "grad_norm": 0.8807555281079319,
      "language_loss": 0.52347249,
      "learning_rate": 2.033948443656652e-08,
      "loss": 0.54352617,
      "num_input_tokens_seen": 342958155,
      "step": 15899,
      "time_per_iteration": 3.108670711517334
    },
    {
      "auxiliary_loss_clip": 0.0109695,
      "auxiliary_loss_mlp": 0.01033831,
      "balance_loss_clip": 1.03550601,
      "balance_loss_mlp": 1.02039063,
      "epoch": 0.9559597174207125,
      "flos": 13764488376960.0,
      "grad_norm": 1.9373974881680784,
      "language_loss": 0.68501699,
      "learning_rate": 2.028411968062782e-08,
      "loss": 0.70632482,
      "num_input_tokens_seen": 342972500,
      "step": 15900,
      "time_per_iteration": 2.4427835941314697
    },
    {
      "auxiliary_loss_clip": 0.01093638,
      "auxiliary_loss_mlp": 0.00783083,
      "balance_loss_clip": 1.03343046,
      "balance_loss_mlp": 1.01027989,
      "epoch": 0.9560198406733804,
      "flos": 19936347799680.0,
      "grad_norm": 1.9249939122078557,
      "language_loss": 0.83099747,
      "learning_rate": 2.0228829995713627e-08,
      "loss": 0.84976459,
      "num_input_tokens_seen": 342989035,
      "step": 15901,
      "time_per_iteration": 2.476273775100708
    },
    {
      "auxiliary_loss_clip": 0.00998521,
      "auxiliary_loss_mlp": 0.0099945,
      "balance_loss_clip": 1.00773287,
      "balance_loss_mlp": 0.99809724,
      "epoch": 0.9560799639260484,
      "flos": 57289550699520.0,
      "grad_norm": 0.7203863851705833,
      "language_loss": 0.54294777,
      "learning_rate": 2.0173615383920485e-08,
      "loss": 0.56292748,
      "num_input_tokens_seen": 343051675,
      "step": 15902,
      "time_per_iteration": 3.2151782512664795
    },
    {
      "auxiliary_loss_clip": 0.01079034,
      "auxiliary_loss_mlp": 0.01031111,
      "balance_loss_clip": 1.03382325,
      "balance_loss_mlp": 1.02101421,
      "epoch": 0.9561400871787163,
      "flos": 18917167299840.0,
      "grad_norm": 1.8772837286569182,
      "language_loss": 0.85376465,
      "learning_rate": 2.01184758473425e-08,
      "loss": 0.87486601,
      "num_input_tokens_seen": 343068895,
      "step": 15903,
      "time_per_iteration": 2.5020482540130615
    },
    {
      "auxiliary_loss_clip": 0.01077062,
      "auxiliary_loss_mlp": 0.00784475,
      "balance_loss_clip": 1.03304076,
      "balance_loss_mlp": 1.00797319,
      "epoch": 0.9562002104313844,
      "flos": 18038576632320.0,
      "grad_norm": 2.0031155071391327,
      "language_loss": 0.80552649,
      "learning_rate": 2.0063411388070217e-08,
      "loss": 0.82414186,
      "num_input_tokens_seen": 343087115,
      "step": 15904,
      "time_per_iteration": 2.470367670059204
    },
    {
      "auxiliary_loss_clip": 0.01094005,
      "auxiliary_loss_mlp": 0.01030832,
      "balance_loss_clip": 1.03427243,
      "balance_loss_mlp": 1.01796889,
      "epoch": 0.9562603336840523,
      "flos": 24717673964160.0,
      "grad_norm": 1.9791369124534814,
      "language_loss": 0.60092962,
      "learning_rate": 2.0008422008191972e-08,
      "loss": 0.62217796,
      "num_input_tokens_seen": 343105575,
      "step": 15905,
      "time_per_iteration": 2.5156326293945312
    },
    {
      "auxiliary_loss_clip": 0.01090473,
      "auxiliary_loss_mlp": 0.01028604,
      "balance_loss_clip": 1.03356647,
      "balance_loss_mlp": 1.01732039,
      "epoch": 0.9563204569367203,
      "flos": 21177205084800.0,
      "grad_norm": 1.9196538500108913,
      "language_loss": 0.70226109,
      "learning_rate": 1.995350770979254e-08,
      "loss": 0.72345185,
      "num_input_tokens_seen": 343123025,
      "step": 15906,
      "time_per_iteration": 2.4661355018615723
    },
    {
      "auxiliary_loss_clip": 0.01061068,
      "auxiliary_loss_mlp": 0.01027844,
      "balance_loss_clip": 1.03648043,
      "balance_loss_mlp": 1.01516628,
      "epoch": 0.9563805801893882,
      "flos": 20229738088320.0,
      "grad_norm": 1.7802784999816448,
      "language_loss": 0.70990217,
      "learning_rate": 1.9898668494954473e-08,
      "loss": 0.73079121,
      "num_input_tokens_seen": 343141625,
      "step": 15907,
      "time_per_iteration": 2.569262742996216
    },
    {
      "auxiliary_loss_clip": 0.01070794,
      "auxiliary_loss_mlp": 0.01030306,
      "balance_loss_clip": 1.03435373,
      "balance_loss_mlp": 1.01823008,
      "epoch": 0.9564407034420562,
      "flos": 25411001258880.0,
      "grad_norm": 2.064882260422441,
      "language_loss": 0.69771278,
      "learning_rate": 1.9843904365757447e-08,
      "loss": 0.71872377,
      "num_input_tokens_seen": 343161300,
      "step": 15908,
      "time_per_iteration": 2.5468592643737793
    },
    {
      "auxiliary_loss_clip": 0.01081601,
      "auxiliary_loss_mlp": 0.00784102,
      "balance_loss_clip": 1.03543615,
      "balance_loss_mlp": 1.01125503,
      "epoch": 0.9565008266947241,
      "flos": 18623884752000.0,
      "grad_norm": 1.7051790600594432,
      "language_loss": 0.83403242,
      "learning_rate": 1.978921532427802e-08,
      "loss": 0.85268945,
      "num_input_tokens_seen": 343177815,
      "step": 15909,
      "time_per_iteration": 2.4965710639953613
    },
    {
      "auxiliary_loss_clip": 0.01091991,
      "auxiliary_loss_mlp": 0.01033948,
      "balance_loss_clip": 1.03347206,
      "balance_loss_mlp": 1.02230155,
      "epoch": 0.9565609499473922,
      "flos": 24862142465280.0,
      "grad_norm": 1.7855566785126014,
      "language_loss": 0.67365772,
      "learning_rate": 1.9734601372590086e-08,
      "loss": 0.69491708,
      "num_input_tokens_seen": 343198140,
      "step": 15910,
      "time_per_iteration": 2.500535488128662
    },
    {
      "auxiliary_loss_clip": 0.01096903,
      "auxiliary_loss_mlp": 0.0103455,
      "balance_loss_clip": 1.03670621,
      "balance_loss_mlp": 1.0221343,
      "epoch": 0.9566210732000601,
      "flos": 21798459740160.0,
      "grad_norm": 1.6787040607623185,
      "language_loss": 0.74329042,
      "learning_rate": 1.968006251276444e-08,
      "loss": 0.76460493,
      "num_input_tokens_seen": 343218280,
      "step": 15911,
      "time_per_iteration": 2.5039381980895996
    },
    {
      "auxiliary_loss_clip": 0.01091699,
      "auxiliary_loss_mlp": 0.01027662,
      "balance_loss_clip": 1.03340507,
      "balance_loss_mlp": 1.01622963,
      "epoch": 0.9566811964527281,
      "flos": 18697609416960.0,
      "grad_norm": 2.032423997204952,
      "language_loss": 0.69474465,
      "learning_rate": 1.9625598746869198e-08,
      "loss": 0.71593827,
      "num_input_tokens_seen": 343236850,
      "step": 15912,
      "time_per_iteration": 2.461393356323242
    },
    {
      "auxiliary_loss_clip": 0.01082522,
      "auxiliary_loss_mlp": 0.0103643,
      "balance_loss_clip": 1.03345442,
      "balance_loss_mlp": 1.02371633,
      "epoch": 0.9567413197053961,
      "flos": 13000632727680.0,
      "grad_norm": 2.555779127438761,
      "language_loss": 0.72365755,
      "learning_rate": 1.95712100769696e-08,
      "loss": 0.744847,
      "num_input_tokens_seen": 343253065,
      "step": 15913,
      "time_per_iteration": 3.95328950881958
    },
    {
      "auxiliary_loss_clip": 0.01023164,
      "auxiliary_loss_mlp": 0.01029469,
      "balance_loss_clip": 1.03333855,
      "balance_loss_mlp": 1.0180254,
      "epoch": 0.956801442958064,
      "flos": 19719267955200.0,
      "grad_norm": 2.283099362244623,
      "language_loss": 0.73285878,
      "learning_rate": 1.9516896505128444e-08,
      "loss": 0.75338507,
      "num_input_tokens_seen": 343270330,
      "step": 15914,
      "time_per_iteration": 4.021246433258057
    },
    {
      "auxiliary_loss_clip": 0.01101227,
      "auxiliary_loss_mlp": 0.01027598,
      "balance_loss_clip": 1.03381062,
      "balance_loss_mlp": 1.0157845,
      "epoch": 0.956861566210732,
      "flos": 18222834424320.0,
      "grad_norm": 1.4378548058613165,
      "language_loss": 0.67189288,
      "learning_rate": 1.9462658033404965e-08,
      "loss": 0.69318116,
      "num_input_tokens_seen": 343289625,
      "step": 15915,
      "time_per_iteration": 4.015488147735596
    },
    {
      "auxiliary_loss_clip": 0.01089565,
      "auxiliary_loss_mlp": 0.01027943,
      "balance_loss_clip": 1.03467917,
      "balance_loss_mlp": 1.01645684,
      "epoch": 0.9569216894634,
      "flos": 22196960202240.0,
      "grad_norm": 1.8036919903093458,
      "language_loss": 0.64365667,
      "learning_rate": 1.9408494663855967e-08,
      "loss": 0.66483176,
      "num_input_tokens_seen": 343309200,
      "step": 15916,
      "time_per_iteration": 2.4763660430908203
    },
    {
      "auxiliary_loss_clip": 0.01097063,
      "auxiliary_loss_mlp": 0.01026813,
      "balance_loss_clip": 1.03365397,
      "balance_loss_mlp": 1.0158217,
      "epoch": 0.956981812716068,
      "flos": 21689291329920.0,
      "grad_norm": 1.8474433729558768,
      "language_loss": 0.80644041,
      "learning_rate": 1.935440639853536e-08,
      "loss": 0.82767916,
      "num_input_tokens_seen": 343326270,
      "step": 15917,
      "time_per_iteration": 2.457047939300537
    },
    {
      "auxiliary_loss_clip": 0.01072449,
      "auxiliary_loss_mlp": 0.01034547,
      "balance_loss_clip": 1.03441083,
      "balance_loss_mlp": 1.02231646,
      "epoch": 0.9570419359687359,
      "flos": 13990905757440.0,
      "grad_norm": 1.7220559109285867,
      "language_loss": 0.73059672,
      "learning_rate": 1.9300393239494172e-08,
      "loss": 0.75166667,
      "num_input_tokens_seen": 343344430,
      "step": 15918,
      "time_per_iteration": 2.4847288131713867
    },
    {
      "auxiliary_loss_clip": 0.01001566,
      "auxiliary_loss_mlp": 0.01005793,
      "balance_loss_clip": 1.00740218,
      "balance_loss_mlp": 1.00469041,
      "epoch": 0.9571020592214039,
      "flos": 65196938534400.0,
      "grad_norm": 0.6310971926736959,
      "language_loss": 0.53151751,
      "learning_rate": 1.924645518878032e-08,
      "loss": 0.55159116,
      "num_input_tokens_seen": 343416155,
      "step": 15919,
      "time_per_iteration": 3.259613513946533
    },
    {
      "auxiliary_loss_clip": 0.01100323,
      "auxiliary_loss_mlp": 0.0103803,
      "balance_loss_clip": 1.03860426,
      "balance_loss_mlp": 1.02493465,
      "epoch": 0.9571621824740718,
      "flos": 17384068961280.0,
      "grad_norm": 2.5404845048967295,
      "language_loss": 0.75121951,
      "learning_rate": 1.919259224843972e-08,
      "loss": 0.77260309,
      "num_input_tokens_seen": 343431715,
      "step": 15920,
      "time_per_iteration": 2.4567558765411377
    },
    {
      "auxiliary_loss_clip": 0.01068175,
      "auxiliary_loss_mlp": 0.01028159,
      "balance_loss_clip": 1.03470254,
      "balance_loss_mlp": 1.01559401,
      "epoch": 0.9572223057267398,
      "flos": 14538184352640.0,
      "grad_norm": 1.6047541313617237,
      "language_loss": 0.78924513,
      "learning_rate": 1.9138804420514298e-08,
      "loss": 0.8102085,
      "num_input_tokens_seen": 343450425,
      "step": 15921,
      "time_per_iteration": 2.5462357997894287
    },
    {
      "auxiliary_loss_clip": 0.01096427,
      "auxiliary_loss_mlp": 0.01028954,
      "balance_loss_clip": 1.03328824,
      "balance_loss_mlp": 1.01616907,
      "epoch": 0.9572824289794077,
      "flos": 33947793158400.0,
      "grad_norm": 1.817280350435024,
      "language_loss": 0.50979364,
      "learning_rate": 1.9085091707044197e-08,
      "loss": 0.53104746,
      "num_input_tokens_seen": 343470445,
      "step": 15922,
      "time_per_iteration": 2.586251974105835
    },
    {
      "auxiliary_loss_clip": 0.01051863,
      "auxiliary_loss_mlp": 0.0103918,
      "balance_loss_clip": 1.0334419,
      "balance_loss_mlp": 1.02508986,
      "epoch": 0.9573425522320758,
      "flos": 18694915896960.0,
      "grad_norm": 2.2591931496330533,
      "language_loss": 0.83395851,
      "learning_rate": 1.903145411006557e-08,
      "loss": 0.85486889,
      "num_input_tokens_seen": 343485200,
      "step": 15923,
      "time_per_iteration": 2.5547213554382324
    },
    {
      "auxiliary_loss_clip": 0.01076349,
      "auxiliary_loss_mlp": 0.01030245,
      "balance_loss_clip": 1.03163981,
      "balance_loss_mlp": 1.0191828,
      "epoch": 0.9574026754847437,
      "flos": 28510307297280.0,
      "grad_norm": 1.5881823127418098,
      "language_loss": 0.75241256,
      "learning_rate": 1.8977891631613008e-08,
      "loss": 0.77347863,
      "num_input_tokens_seen": 343505080,
      "step": 15924,
      "time_per_iteration": 3.956639051437378
    },
    {
      "auxiliary_loss_clip": 0.01081747,
      "auxiliary_loss_mlp": 0.01028355,
      "balance_loss_clip": 1.03414202,
      "balance_loss_mlp": 1.01637435,
      "epoch": 0.9574627987374117,
      "flos": 24352390604160.0,
      "grad_norm": 1.82226157630626,
      "language_loss": 0.86174458,
      "learning_rate": 1.892440427371711e-08,
      "loss": 0.88284564,
      "num_input_tokens_seen": 343523995,
      "step": 15925,
      "time_per_iteration": 2.523029327392578
    },
    {
      "auxiliary_loss_clip": 0.01074689,
      "auxiliary_loss_mlp": 0.01032293,
      "balance_loss_clip": 1.03449059,
      "balance_loss_mlp": 1.01944852,
      "epoch": 0.9575229219900797,
      "flos": 23510680225920.0,
      "grad_norm": 1.944468710259776,
      "language_loss": 0.75721955,
      "learning_rate": 1.8870992038406474e-08,
      "loss": 0.77828932,
      "num_input_tokens_seen": 343542015,
      "step": 15926,
      "time_per_iteration": 2.5605249404907227
    },
    {
      "auxiliary_loss_clip": 0.01077984,
      "auxiliary_loss_mlp": 0.01030571,
      "balance_loss_clip": 1.03597796,
      "balance_loss_mlp": 1.01954389,
      "epoch": 0.9575830452427476,
      "flos": 22674823764480.0,
      "grad_norm": 1.6041897822134794,
      "language_loss": 0.77534223,
      "learning_rate": 1.8817654927706373e-08,
      "loss": 0.79642779,
      "num_input_tokens_seen": 343561680,
      "step": 15927,
      "time_per_iteration": 2.504793167114258
    },
    {
      "auxiliary_loss_clip": 0.01057699,
      "auxiliary_loss_mlp": 0.01031902,
      "balance_loss_clip": 1.03486514,
      "balance_loss_mlp": 1.01848531,
      "epoch": 0.9576431684954156,
      "flos": 30485250835200.0,
      "grad_norm": 2.929314003374916,
      "language_loss": 0.68521905,
      "learning_rate": 1.8764392943639183e-08,
      "loss": 0.70611501,
      "num_input_tokens_seen": 343585290,
      "step": 15928,
      "time_per_iteration": 2.6596686840057373
    },
    {
      "auxiliary_loss_clip": 0.01082723,
      "auxiliary_loss_mlp": 0.01030747,
      "balance_loss_clip": 1.03558528,
      "balance_loss_mlp": 1.01837301,
      "epoch": 0.9577032917480836,
      "flos": 21687387909120.0,
      "grad_norm": 1.8608948082465002,
      "language_loss": 0.82104409,
      "learning_rate": 1.871120608822485e-08,
      "loss": 0.84217882,
      "num_input_tokens_seen": 343604045,
      "step": 15929,
      "time_per_iteration": 2.489361524581909
    },
    {
      "auxiliary_loss_clip": 0.01072197,
      "auxiliary_loss_mlp": 0.01046649,
      "balance_loss_clip": 1.03397441,
      "balance_loss_mlp": 1.03280842,
      "epoch": 0.9577634150007516,
      "flos": 29023147728000.0,
      "grad_norm": 1.6000973528018319,
      "language_loss": 0.72025931,
      "learning_rate": 1.8658094363480202e-08,
      "loss": 0.74144769,
      "num_input_tokens_seen": 343626595,
      "step": 15930,
      "time_per_iteration": 2.589919328689575
    },
    {
      "auxiliary_loss_clip": 0.01039819,
      "auxiliary_loss_mlp": 0.01029666,
      "balance_loss_clip": 1.03186536,
      "balance_loss_mlp": 1.0167501,
      "epoch": 0.9578235382534195,
      "flos": 19282235178240.0,
      "grad_norm": 1.4561256930207347,
      "language_loss": 0.62134254,
      "learning_rate": 1.8605057771419185e-08,
      "loss": 0.64203745,
      "num_input_tokens_seen": 343646195,
      "step": 15931,
      "time_per_iteration": 2.567692995071411
    },
    {
      "auxiliary_loss_clip": 0.01100888,
      "auxiliary_loss_mlp": 0.01026906,
      "balance_loss_clip": 1.03472543,
      "balance_loss_mlp": 1.01594484,
      "epoch": 0.9578836615060875,
      "flos": 13699275235200.0,
      "grad_norm": 1.840946639389772,
      "language_loss": 0.68882704,
      "learning_rate": 1.8552096314052633e-08,
      "loss": 0.71010494,
      "num_input_tokens_seen": 343663665,
      "step": 15932,
      "time_per_iteration": 2.4318737983703613
    },
    {
      "auxiliary_loss_clip": 0.01075455,
      "auxiliary_loss_mlp": 0.01035678,
      "balance_loss_clip": 1.03466582,
      "balance_loss_mlp": 1.02245128,
      "epoch": 0.9579437847587554,
      "flos": 17054516655360.0,
      "grad_norm": 1.8553278407037948,
      "language_loss": 0.75707257,
      "learning_rate": 1.849920999338961e-08,
      "loss": 0.77818394,
      "num_input_tokens_seen": 343682145,
      "step": 15933,
      "time_per_iteration": 2.493351936340332
    },
    {
      "auxiliary_loss_clip": 0.00997326,
      "auxiliary_loss_mlp": 0.01000596,
      "balance_loss_clip": 1.01121902,
      "balance_loss_mlp": 0.9995411,
      "epoch": 0.9580039080114234,
      "flos": 60570887886720.0,
      "grad_norm": 0.7020029629969564,
      "language_loss": 0.57321984,
      "learning_rate": 1.8446398811434948e-08,
      "loss": 0.59319907,
      "num_input_tokens_seen": 343744685,
      "step": 15934,
      "time_per_iteration": 3.3038177490234375
    },
    {
      "auxiliary_loss_clip": 0.01026518,
      "auxiliary_loss_mlp": 0.00760509,
      "balance_loss_clip": 1.00381088,
      "balance_loss_mlp": 0.99941331,
      "epoch": 0.9580640312640913,
      "flos": 66235365745920.0,
      "grad_norm": 0.9185406796379341,
      "language_loss": 0.65951848,
      "learning_rate": 1.8393662770191277e-08,
      "loss": 0.67738873,
      "num_input_tokens_seen": 343801835,
      "step": 15935,
      "time_per_iteration": 3.006756544113159
    },
    {
      "auxiliary_loss_clip": 0.01017347,
      "auxiliary_loss_mlp": 0.01002173,
      "balance_loss_clip": 1.02157879,
      "balance_loss_mlp": 1.00116563,
      "epoch": 0.9581241545167594,
      "flos": 62218002971520.0,
      "grad_norm": 0.7834039313500597,
      "language_loss": 0.57042682,
      "learning_rate": 1.8341001871658546e-08,
      "loss": 0.59062195,
      "num_input_tokens_seen": 343861515,
      "step": 15936,
      "time_per_iteration": 3.106759548187256
    },
    {
      "auxiliary_loss_clip": 0.01044541,
      "auxiliary_loss_mlp": 0.01029145,
      "balance_loss_clip": 1.03354883,
      "balance_loss_mlp": 1.01692557,
      "epoch": 0.9581842777694273,
      "flos": 23768088065280.0,
      "grad_norm": 1.785429528012442,
      "language_loss": 0.7834543,
      "learning_rate": 1.8288416117833825e-08,
      "loss": 0.80419111,
      "num_input_tokens_seen": 343881240,
      "step": 15937,
      "time_per_iteration": 2.66329026222229
    },
    {
      "auxiliary_loss_clip": 0.01091265,
      "auxiliary_loss_mlp": 0.01036436,
      "balance_loss_clip": 1.03490853,
      "balance_loss_mlp": 1.0230732,
      "epoch": 0.9582444010220953,
      "flos": 21213079793280.0,
      "grad_norm": 1.7461192056981982,
      "language_loss": 0.68426871,
      "learning_rate": 1.8235905510710636e-08,
      "loss": 0.70554572,
      "num_input_tokens_seen": 343900885,
      "step": 15938,
      "time_per_iteration": 2.472263813018799
    },
    {
      "auxiliary_loss_clip": 0.01070775,
      "auxiliary_loss_mlp": 0.01029346,
      "balance_loss_clip": 1.03347349,
      "balance_loss_mlp": 1.01763964,
      "epoch": 0.9583045242747633,
      "flos": 23805147922560.0,
      "grad_norm": 2.473922718314474,
      "language_loss": 0.66139513,
      "learning_rate": 1.8183470052280712e-08,
      "loss": 0.68239635,
      "num_input_tokens_seen": 343918460,
      "step": 15939,
      "time_per_iteration": 2.573601007461548
    },
    {
      "auxiliary_loss_clip": 0.01074122,
      "auxiliary_loss_mlp": 0.01036495,
      "balance_loss_clip": 1.0322777,
      "balance_loss_mlp": 1.02426386,
      "epoch": 0.9583646475274312,
      "flos": 24131468004480.0,
      "grad_norm": 1.562492813331926,
      "language_loss": 0.73754221,
      "learning_rate": 1.8131109744532025e-08,
      "loss": 0.7586484,
      "num_input_tokens_seen": 343938030,
      "step": 15940,
      "time_per_iteration": 2.539907693862915
    },
    {
      "auxiliary_loss_clip": 0.01104598,
      "auxiliary_loss_mlp": 0.0103106,
      "balance_loss_clip": 1.03524601,
      "balance_loss_mlp": 1.01820314,
      "epoch": 0.9584247707800992,
      "flos": 20886651970560.0,
      "grad_norm": 1.6364897199906014,
      "language_loss": 0.72977346,
      "learning_rate": 1.8078824589450535e-08,
      "loss": 0.75113004,
      "num_input_tokens_seen": 343956635,
      "step": 15941,
      "time_per_iteration": 2.495635747909546
    },
    {
      "auxiliary_loss_clip": 0.01080352,
      "auxiliary_loss_mlp": 0.01032665,
      "balance_loss_clip": 1.03433156,
      "balance_loss_mlp": 1.02115536,
      "epoch": 0.9584848940327672,
      "flos": 26067591918720.0,
      "grad_norm": 1.458752176432393,
      "language_loss": 0.71354914,
      "learning_rate": 1.8026614589018442e-08,
      "loss": 0.73467934,
      "num_input_tokens_seen": 343976625,
      "step": 15942,
      "time_per_iteration": 2.5285017490386963
    },
    {
      "auxiliary_loss_clip": 0.01104556,
      "auxiliary_loss_mlp": 0.01031993,
      "balance_loss_clip": 1.03455448,
      "balance_loss_mlp": 1.01905847,
      "epoch": 0.9585450172854352,
      "flos": 34492988764800.0,
      "grad_norm": 1.526650814871503,
      "language_loss": 0.71940994,
      "learning_rate": 1.797447974521571e-08,
      "loss": 0.74077547,
      "num_input_tokens_seen": 343997790,
      "step": 15943,
      "time_per_iteration": 2.5558314323425293
    },
    {
      "auxiliary_loss_clip": 0.01095528,
      "auxiliary_loss_mlp": 0.01035494,
      "balance_loss_clip": 1.03638625,
      "balance_loss_mlp": 1.02284539,
      "epoch": 0.9586051405381031,
      "flos": 23110743219840.0,
      "grad_norm": 1.6147344520935178,
      "language_loss": 0.6866678,
      "learning_rate": 1.792242006001965e-08,
      "loss": 0.70797801,
      "num_input_tokens_seen": 344016935,
      "step": 15944,
      "time_per_iteration": 2.47809100151062
    },
    {
      "auxiliary_loss_clip": 0.01101633,
      "auxiliary_loss_mlp": 0.0103343,
      "balance_loss_clip": 1.03299522,
      "balance_loss_mlp": 1.02140784,
      "epoch": 0.9586652637907711,
      "flos": 19603994232960.0,
      "grad_norm": 2.273716686536164,
      "language_loss": 0.6584022,
      "learning_rate": 1.7870435535403795e-08,
      "loss": 0.67975283,
      "num_input_tokens_seen": 344035590,
      "step": 15945,
      "time_per_iteration": 2.446030616760254
    },
    {
      "auxiliary_loss_clip": 0.00975742,
      "auxiliary_loss_mlp": 0.00999664,
      "balance_loss_clip": 1.01493168,
      "balance_loss_mlp": 0.99847233,
      "epoch": 0.958725387043439,
      "flos": 72073327317120.0,
      "grad_norm": 0.7526189972699443,
      "language_loss": 0.61869299,
      "learning_rate": 1.7818526173339678e-08,
      "loss": 0.63844705,
      "num_input_tokens_seen": 344100845,
      "step": 15946,
      "time_per_iteration": 3.693606376647949
    },
    {
      "auxiliary_loss_clip": 0.01101618,
      "auxiliary_loss_mlp": 0.01029431,
      "balance_loss_clip": 1.03490508,
      "balance_loss_mlp": 1.0177424,
      "epoch": 0.958785510296107,
      "flos": 28911932242560.0,
      "grad_norm": 1.7476571459982133,
      "language_loss": 0.75108135,
      "learning_rate": 1.7766691975795723e-08,
      "loss": 0.7723918,
      "num_input_tokens_seen": 344121780,
      "step": 15947,
      "time_per_iteration": 4.246658802032471
    },
    {
      "auxiliary_loss_clip": 0.01075682,
      "auxiliary_loss_mlp": 0.01026016,
      "balance_loss_clip": 1.03352523,
      "balance_loss_mlp": 1.0150373,
      "epoch": 0.958845633548775,
      "flos": 18477189607680.0,
      "grad_norm": 2.4694366021133756,
      "language_loss": 0.69625229,
      "learning_rate": 1.771493294473747e-08,
      "loss": 0.7172693,
      "num_input_tokens_seen": 344140150,
      "step": 15948,
      "time_per_iteration": 2.549262285232544
    },
    {
      "auxiliary_loss_clip": 0.01058209,
      "auxiliary_loss_mlp": 0.01029458,
      "balance_loss_clip": 1.03334522,
      "balance_loss_mlp": 1.01817465,
      "epoch": 0.958905756801443,
      "flos": 24206916522240.0,
      "grad_norm": 2.890605891112174,
      "language_loss": 0.78635979,
      "learning_rate": 1.7663249082127574e-08,
      "loss": 0.80723643,
      "num_input_tokens_seen": 344158200,
      "step": 15949,
      "time_per_iteration": 2.6112027168273926
    },
    {
      "auxiliary_loss_clip": 0.01106084,
      "auxiliary_loss_mlp": 0.01033976,
      "balance_loss_clip": 1.03681552,
      "balance_loss_mlp": 1.02163196,
      "epoch": 0.9589658800541109,
      "flos": 25007939769600.0,
      "grad_norm": 1.844508798425589,
      "language_loss": 0.69033539,
      "learning_rate": 1.761164038992602e-08,
      "loss": 0.71173602,
      "num_input_tokens_seen": 344174720,
      "step": 15950,
      "time_per_iteration": 2.495490789413452
    },
    {
      "auxiliary_loss_clip": 0.01083908,
      "auxiliary_loss_mlp": 0.01030868,
      "balance_loss_clip": 1.03530264,
      "balance_loss_mlp": 1.01981735,
      "epoch": 0.9590260033067789,
      "flos": 23514558894720.0,
      "grad_norm": 1.6703485801188571,
      "language_loss": 0.86066127,
      "learning_rate": 1.7560106870089687e-08,
      "loss": 0.88180906,
      "num_input_tokens_seen": 344192580,
      "step": 15951,
      "time_per_iteration": 2.539576530456543
    },
    {
      "auxiliary_loss_clip": 0.0108144,
      "auxiliary_loss_mlp": 0.01037971,
      "balance_loss_clip": 1.03421223,
      "balance_loss_mlp": 1.02549028,
      "epoch": 0.9590861265594469,
      "flos": 25520349237120.0,
      "grad_norm": 2.0163504894454247,
      "language_loss": 0.79947764,
      "learning_rate": 1.7508648524572568e-08,
      "loss": 0.82067174,
      "num_input_tokens_seen": 344210345,
      "step": 15952,
      "time_per_iteration": 3.920907974243164
    },
    {
      "auxiliary_loss_clip": 0.01093128,
      "auxiliary_loss_mlp": 0.01029653,
      "balance_loss_clip": 1.03657246,
      "balance_loss_mlp": 1.0171597,
      "epoch": 0.9591462498121148,
      "flos": 21179323987200.0,
      "grad_norm": 3.056460079727345,
      "language_loss": 0.695207,
      "learning_rate": 1.7457265355326434e-08,
      "loss": 0.71643472,
      "num_input_tokens_seen": 344229540,
      "step": 15953,
      "time_per_iteration": 3.8917765617370605
    },
    {
      "auxiliary_loss_clip": 0.01042666,
      "auxiliary_loss_mlp": 0.01034832,
      "balance_loss_clip": 1.03316236,
      "balance_loss_mlp": 1.02101016,
      "epoch": 0.9592063730647828,
      "flos": 21723047136000.0,
      "grad_norm": 3.9526374006240603,
      "language_loss": 0.58715278,
      "learning_rate": 1.7405957364299285e-08,
      "loss": 0.60792774,
      "num_input_tokens_seen": 344247830,
      "step": 15954,
      "time_per_iteration": 4.014133453369141
    },
    {
      "auxiliary_loss_clip": 0.01094105,
      "auxiliary_loss_mlp": 0.01034606,
      "balance_loss_clip": 1.03428793,
      "balance_loss_mlp": 1.02139759,
      "epoch": 0.9592664963174508,
      "flos": 29891395278720.0,
      "grad_norm": 2.4237121218608864,
      "language_loss": 0.73829228,
      "learning_rate": 1.7354724553437117e-08,
      "loss": 0.75957936,
      "num_input_tokens_seen": 344267760,
      "step": 15955,
      "time_per_iteration": 2.5334370136260986
    },
    {
      "auxiliary_loss_clip": 0.01081593,
      "auxiliary_loss_mlp": 0.01033914,
      "balance_loss_clip": 1.0337199,
      "balance_loss_mlp": 1.02115297,
      "epoch": 0.9593266195701188,
      "flos": 17999613354240.0,
      "grad_norm": 1.8279548051095775,
      "language_loss": 0.6242488,
      "learning_rate": 1.7303566924682378e-08,
      "loss": 0.64540386,
      "num_input_tokens_seen": 344284905,
      "step": 15956,
      "time_per_iteration": 2.502157211303711
    },
    {
      "auxiliary_loss_clip": 0.01064796,
      "auxiliary_loss_mlp": 0.01030221,
      "balance_loss_clip": 1.03447366,
      "balance_loss_mlp": 1.01805544,
      "epoch": 0.9593867428227867,
      "flos": 18838271076480.0,
      "grad_norm": 1.7845114101480872,
      "language_loss": 0.59916019,
      "learning_rate": 1.725248447997507e-08,
      "loss": 0.62011039,
      "num_input_tokens_seen": 344302025,
      "step": 15957,
      "time_per_iteration": 2.518641233444214
    },
    {
      "auxiliary_loss_clip": 0.0107768,
      "auxiliary_loss_mlp": 0.01036211,
      "balance_loss_clip": 1.03700614,
      "balance_loss_mlp": 1.0234077,
      "epoch": 0.9594468660754547,
      "flos": 29567050444800.0,
      "grad_norm": 1.9762818713525976,
      "language_loss": 0.74276549,
      "learning_rate": 1.7201477221252314e-08,
      "loss": 0.76390439,
      "num_input_tokens_seen": 344321935,
      "step": 15958,
      "time_per_iteration": 2.6308443546295166
    },
    {
      "auxiliary_loss_clip": 0.01081437,
      "auxiliary_loss_mlp": 0.00782049,
      "balance_loss_clip": 1.03289688,
      "balance_loss_mlp": 1.00730979,
      "epoch": 0.9595069893281226,
      "flos": 20703256104960.0,
      "grad_norm": 1.5812675597854473,
      "language_loss": 0.7471199,
      "learning_rate": 1.7150545150448116e-08,
      "loss": 0.7657547,
      "num_input_tokens_seen": 344340405,
      "step": 15959,
      "time_per_iteration": 2.5526771545410156
    },
    {
      "auxiliary_loss_clip": 0.01088496,
      "auxiliary_loss_mlp": 0.01031219,
      "balance_loss_clip": 1.03336513,
      "balance_loss_mlp": 1.01885676,
      "epoch": 0.9595671125807906,
      "flos": 22453613856000.0,
      "grad_norm": 3.3187045744891526,
      "language_loss": 0.65457022,
      "learning_rate": 1.7099688269493816e-08,
      "loss": 0.67576742,
      "num_input_tokens_seen": 344359925,
      "step": 15960,
      "time_per_iteration": 2.4939675331115723
    },
    {
      "auxiliary_loss_clip": 0.01099978,
      "auxiliary_loss_mlp": 0.01035642,
      "balance_loss_clip": 1.0336833,
      "balance_loss_mlp": 1.02341723,
      "epoch": 0.9596272358334585,
      "flos": 23915214172800.0,
      "grad_norm": 1.701340091593819,
      "language_loss": 0.78265959,
      "learning_rate": 1.7048906580318544e-08,
      "loss": 0.80401576,
      "num_input_tokens_seen": 344379100,
      "step": 15961,
      "time_per_iteration": 2.4523284435272217
    },
    {
      "auxiliary_loss_clip": 0.0106423,
      "auxiliary_loss_mlp": 0.01025198,
      "balance_loss_clip": 1.03463006,
      "balance_loss_mlp": 1.01358676,
      "epoch": 0.9596873590861266,
      "flos": 17672539086720.0,
      "grad_norm": 2.1992366644086805,
      "language_loss": 0.75598967,
      "learning_rate": 1.699820008484698e-08,
      "loss": 0.77688396,
      "num_input_tokens_seen": 344396895,
      "step": 15962,
      "time_per_iteration": 3.956773519515991
    },
    {
      "auxiliary_loss_clip": 0.01084282,
      "auxiliary_loss_mlp": 0.01031975,
      "balance_loss_clip": 1.03499246,
      "balance_loss_mlp": 1.01922011,
      "epoch": 0.9597474823387945,
      "flos": 25808532053760.0,
      "grad_norm": 1.8573725654866093,
      "language_loss": 0.71223193,
      "learning_rate": 1.6947568785002698e-08,
      "loss": 0.7333945,
      "num_input_tokens_seen": 344415115,
      "step": 15963,
      "time_per_iteration": 2.5508811473846436
    },
    {
      "auxiliary_loss_clip": 0.01072655,
      "auxiliary_loss_mlp": 0.01027618,
      "balance_loss_clip": 1.03455973,
      "balance_loss_mlp": 1.01591718,
      "epoch": 0.9598076055914625,
      "flos": 23768519028480.0,
      "grad_norm": 1.7693670494968199,
      "language_loss": 0.74076569,
      "learning_rate": 1.689701268270527e-08,
      "loss": 0.76176846,
      "num_input_tokens_seen": 344435185,
      "step": 15964,
      "time_per_iteration": 2.52687406539917
    },
    {
      "auxiliary_loss_clip": 0.00998772,
      "auxiliary_loss_mlp": 0.01005062,
      "balance_loss_clip": 1.00831807,
      "balance_loss_mlp": 1.00399518,
      "epoch": 0.9598677288441305,
      "flos": 56515962464640.0,
      "grad_norm": 0.8838922077936839,
      "language_loss": 0.57640922,
      "learning_rate": 1.684653177987161e-08,
      "loss": 0.59644759,
      "num_input_tokens_seen": 344488950,
      "step": 15965,
      "time_per_iteration": 3.1429336071014404
    },
    {
      "auxiliary_loss_clip": 0.01104034,
      "auxiliary_loss_mlp": 0.010292,
      "balance_loss_clip": 1.03438258,
      "balance_loss_mlp": 1.01779222,
      "epoch": 0.9599278520967984,
      "flos": 22997480659200.0,
      "grad_norm": 2.7568724861155594,
      "language_loss": 0.79096282,
      "learning_rate": 1.6796126078416627e-08,
      "loss": 0.8122952,
      "num_input_tokens_seen": 344506740,
      "step": 15966,
      "time_per_iteration": 2.448906183242798
    },
    {
      "auxiliary_loss_clip": 0.01076144,
      "auxiliary_loss_mlp": 0.01029899,
      "balance_loss_clip": 1.03097892,
      "balance_loss_mlp": 1.01811469,
      "epoch": 0.9599879753494664,
      "flos": 23039676161280.0,
      "grad_norm": 1.5928923510809871,
      "language_loss": 0.79547703,
      "learning_rate": 1.674579558025102e-08,
      "loss": 0.81653738,
      "num_input_tokens_seen": 344526670,
      "step": 15967,
      "time_per_iteration": 2.540665864944458
    },
    {
      "auxiliary_loss_clip": 0.0105004,
      "auxiliary_loss_mlp": 0.01030272,
      "balance_loss_clip": 1.03277445,
      "balance_loss_mlp": 1.01664639,
      "epoch": 0.9600480986021344,
      "flos": 16392287560320.0,
      "grad_norm": 1.861431279007753,
      "language_loss": 0.80382127,
      "learning_rate": 1.669554028728348e-08,
      "loss": 0.8246243,
      "num_input_tokens_seen": 344541995,
      "step": 15968,
      "time_per_iteration": 2.5250611305236816
    },
    {
      "auxiliary_loss_clip": 0.01057507,
      "auxiliary_loss_mlp": 0.01040017,
      "balance_loss_clip": 1.03446269,
      "balance_loss_mlp": 1.02568829,
      "epoch": 0.9601082218548024,
      "flos": 24276439296000.0,
      "grad_norm": 2.540066048339152,
      "language_loss": 0.67441642,
      "learning_rate": 1.6645360201420044e-08,
      "loss": 0.69539165,
      "num_input_tokens_seen": 344559980,
      "step": 15969,
      "time_per_iteration": 2.613860607147217
    },
    {
      "auxiliary_loss_clip": 0.01091632,
      "auxiliary_loss_mlp": 0.010338,
      "balance_loss_clip": 1.03416753,
      "balance_loss_mlp": 1.02223063,
      "epoch": 0.9601683451074703,
      "flos": 19609991804160.0,
      "grad_norm": 2.600879792806001,
      "language_loss": 0.78999299,
      "learning_rate": 1.6595255324563186e-08,
      "loss": 0.81124729,
      "num_input_tokens_seen": 344577765,
      "step": 15970,
      "time_per_iteration": 2.451685905456543
    },
    {
      "auxiliary_loss_clip": 0.01090456,
      "auxiliary_loss_mlp": 0.01028605,
      "balance_loss_clip": 1.03542995,
      "balance_loss_mlp": 1.01681542,
      "epoch": 0.9602284683601383,
      "flos": 26651104358400.0,
      "grad_norm": 1.531143516419581,
      "language_loss": 0.77233666,
      "learning_rate": 1.654522565861316e-08,
      "loss": 0.79352725,
      "num_input_tokens_seen": 344597650,
      "step": 15971,
      "time_per_iteration": 2.564420461654663
    },
    {
      "auxiliary_loss_clip": 0.01084693,
      "auxiliary_loss_mlp": 0.01025932,
      "balance_loss_clip": 1.03251171,
      "balance_loss_mlp": 1.01352882,
      "epoch": 0.9602885916128062,
      "flos": 15554096714880.0,
      "grad_norm": 1.8992173161421073,
      "language_loss": 0.67115504,
      "learning_rate": 1.64952712054669e-08,
      "loss": 0.69226128,
      "num_input_tokens_seen": 344613580,
      "step": 15972,
      "time_per_iteration": 2.508312702178955
    },
    {
      "auxiliary_loss_clip": 0.01090326,
      "auxiliary_loss_mlp": 0.0078303,
      "balance_loss_clip": 1.033144,
      "balance_loss_mlp": 1.0092088,
      "epoch": 0.9603487148654742,
      "flos": 16502353810560.0,
      "grad_norm": 2.273500496911293,
      "language_loss": 0.76383913,
      "learning_rate": 1.644539196701844e-08,
      "loss": 0.78257263,
      "num_input_tokens_seen": 344626910,
      "step": 15973,
      "time_per_iteration": 2.462385416030884
    },
    {
      "auxiliary_loss_clip": 0.01061788,
      "auxiliary_loss_mlp": 0.01042981,
      "balance_loss_clip": 1.03663623,
      "balance_loss_mlp": 1.02974892,
      "epoch": 0.9604088381181421,
      "flos": 20845354308480.0,
      "grad_norm": 1.5939080674324668,
      "language_loss": 0.6936242,
      "learning_rate": 1.639558794515983e-08,
      "loss": 0.71467191,
      "num_input_tokens_seen": 344644330,
      "step": 15974,
      "time_per_iteration": 2.565502166748047
    },
    {
      "auxiliary_loss_clip": 0.01092352,
      "auxiliary_loss_mlp": 0.01026285,
      "balance_loss_clip": 1.03304088,
      "balance_loss_mlp": 1.01375055,
      "epoch": 0.9604689613708102,
      "flos": 19683105937920.0,
      "grad_norm": 1.7735175208376301,
      "language_loss": 0.67880708,
      "learning_rate": 1.6345859141779105e-08,
      "loss": 0.69999349,
      "num_input_tokens_seen": 344663910,
      "step": 15975,
      "time_per_iteration": 2.4924046993255615
    },
    {
      "auxiliary_loss_clip": 0.01099698,
      "auxiliary_loss_mlp": 0.01025606,
      "balance_loss_clip": 1.03456903,
      "balance_loss_mlp": 1.01381636,
      "epoch": 0.9605290846234781,
      "flos": 24097568544000.0,
      "grad_norm": 2.1644077170915534,
      "language_loss": 0.55098391,
      "learning_rate": 1.6296205558762322e-08,
      "loss": 0.5722369,
      "num_input_tokens_seen": 344682320,
      "step": 15976,
      "time_per_iteration": 2.4620418548583984
    },
    {
      "auxiliary_loss_clip": 0.01074451,
      "auxiliary_loss_mlp": 0.01025254,
      "balance_loss_clip": 1.03093445,
      "balance_loss_mlp": 1.013834,
      "epoch": 0.9605892078761461,
      "flos": 27122575299840.0,
      "grad_norm": 1.7208234643043387,
      "language_loss": 0.68327129,
      "learning_rate": 1.624662719799219e-08,
      "loss": 0.7042684,
      "num_input_tokens_seen": 344701355,
      "step": 15977,
      "time_per_iteration": 2.5575196743011475
    },
    {
      "auxiliary_loss_clip": 0.0109153,
      "auxiliary_loss_mlp": 0.01034754,
      "balance_loss_clip": 1.03231239,
      "balance_loss_mlp": 1.02243948,
      "epoch": 0.9606493311288141,
      "flos": 14136918543360.0,
      "grad_norm": 4.225359531313664,
      "language_loss": 0.81974339,
      "learning_rate": 1.6197124061348766e-08,
      "loss": 0.84100622,
      "num_input_tokens_seen": 344717980,
      "step": 15978,
      "time_per_iteration": 2.4253478050231934
    },
    {
      "auxiliary_loss_clip": 0.0109604,
      "auxiliary_loss_mlp": 0.01031739,
      "balance_loss_clip": 1.03471649,
      "balance_loss_mlp": 1.01934135,
      "epoch": 0.960709454381482,
      "flos": 15813336147840.0,
      "grad_norm": 2.313349113266716,
      "language_loss": 0.83538067,
      "learning_rate": 1.614769615070921e-08,
      "loss": 0.8566584,
      "num_input_tokens_seen": 344733480,
      "step": 15979,
      "time_per_iteration": 2.483630418777466
    },
    {
      "auxiliary_loss_clip": 0.01102604,
      "auxiliary_loss_mlp": 0.01035217,
      "balance_loss_clip": 1.03371191,
      "balance_loss_mlp": 1.02391028,
      "epoch": 0.96076957763415,
      "flos": 22565403959040.0,
      "grad_norm": 1.593333317970244,
      "language_loss": 0.80055827,
      "learning_rate": 1.6098343467947805e-08,
      "loss": 0.82193649,
      "num_input_tokens_seen": 344752130,
      "step": 15980,
      "time_per_iteration": 2.4503917694091797
    },
    {
      "auxiliary_loss_clip": 0.01094292,
      "auxiliary_loss_mlp": 0.01026914,
      "balance_loss_clip": 1.0336895,
      "balance_loss_mlp": 1.01509464,
      "epoch": 0.960829700886818,
      "flos": 24681260551680.0,
      "grad_norm": 1.788279271267182,
      "language_loss": 0.68606603,
      "learning_rate": 1.6049066014935942e-08,
      "loss": 0.70727807,
      "num_input_tokens_seen": 344771195,
      "step": 15981,
      "time_per_iteration": 2.529557228088379
    },
    {
      "auxiliary_loss_clip": 0.01089463,
      "auxiliary_loss_mlp": 0.00782236,
      "balance_loss_clip": 1.0328548,
      "balance_loss_mlp": 1.00882733,
      "epoch": 0.960889824139486,
      "flos": 26542223256960.0,
      "grad_norm": 1.5806678756939394,
      "language_loss": 0.69476908,
      "learning_rate": 1.5999863793542344e-08,
      "loss": 0.71348608,
      "num_input_tokens_seen": 344793150,
      "step": 15982,
      "time_per_iteration": 2.5220072269439697
    },
    {
      "auxiliary_loss_clip": 0.00999551,
      "auxiliary_loss_mlp": 0.00998667,
      "balance_loss_clip": 1.01090169,
      "balance_loss_mlp": 0.99753428,
      "epoch": 0.9609499473921539,
      "flos": 71114942586240.0,
      "grad_norm": 0.6766048582056688,
      "language_loss": 0.53315926,
      "learning_rate": 1.595073680563286e-08,
      "loss": 0.55314147,
      "num_input_tokens_seen": 344852855,
      "step": 15983,
      "time_per_iteration": 3.2357494831085205
    },
    {
      "auxiliary_loss_clip": 0.01102835,
      "auxiliary_loss_mlp": 0.01035437,
      "balance_loss_clip": 1.03478205,
      "balance_loss_mlp": 1.02331352,
      "epoch": 0.9610100706448219,
      "flos": 20552466810240.0,
      "grad_norm": 2.034600760498007,
      "language_loss": 0.67900097,
      "learning_rate": 1.5901685053070212e-08,
      "loss": 0.70038372,
      "num_input_tokens_seen": 344869830,
      "step": 15984,
      "time_per_iteration": 2.422638177871704
    },
    {
      "auxiliary_loss_clip": 0.01069171,
      "auxiliary_loss_mlp": 0.01031683,
      "balance_loss_clip": 1.03444147,
      "balance_loss_mlp": 1.02076304,
      "epoch": 0.9610701938974898,
      "flos": 14064199459200.0,
      "grad_norm": 1.548563230852079,
      "language_loss": 0.67036319,
      "learning_rate": 1.5852708537714477e-08,
      "loss": 0.69137168,
      "num_input_tokens_seen": 344888905,
      "step": 15985,
      "time_per_iteration": 2.5665576457977295
    },
    {
      "auxiliary_loss_clip": 0.01105262,
      "auxiliary_loss_mlp": 0.01028125,
      "balance_loss_clip": 1.03524506,
      "balance_loss_mlp": 1.01637101,
      "epoch": 0.9611303171501578,
      "flos": 20229989483520.0,
      "grad_norm": 1.8261844705058505,
      "language_loss": 0.78758872,
      "learning_rate": 1.580380726142283e-08,
      "loss": 0.80892265,
      "num_input_tokens_seen": 344907160,
      "step": 15986,
      "time_per_iteration": 2.4237544536590576
    },
    {
      "auxiliary_loss_clip": 0.01064648,
      "auxiliary_loss_mlp": 0.01030823,
      "balance_loss_clip": 1.03811765,
      "balance_loss_mlp": 1.01736987,
      "epoch": 0.9611904404028258,
      "flos": 20951075013120.0,
      "grad_norm": 1.920179736460202,
      "language_loss": 0.63717663,
      "learning_rate": 1.5754981226049792e-08,
      "loss": 0.65813136,
      "num_input_tokens_seen": 344922400,
      "step": 15987,
      "time_per_iteration": 2.5615077018737793
    },
    {
      "auxiliary_loss_clip": 0.01099654,
      "auxiliary_loss_mlp": 0.01027062,
      "balance_loss_clip": 1.03469634,
      "balance_loss_mlp": 1.01617789,
      "epoch": 0.9612505636554938,
      "flos": 24827740214400.0,
      "grad_norm": 1.9251422517689372,
      "language_loss": 0.66597074,
      "learning_rate": 1.5706230433446544e-08,
      "loss": 0.68723786,
      "num_input_tokens_seen": 344941910,
      "step": 15988,
      "time_per_iteration": 2.4665749073028564
    },
    {
      "auxiliary_loss_clip": 0.01092117,
      "auxiliary_loss_mlp": 0.01038649,
      "balance_loss_clip": 1.03423762,
      "balance_loss_mlp": 1.02769947,
      "epoch": 0.9613106869081617,
      "flos": 17164977955200.0,
      "grad_norm": 1.78032056768738,
      "language_loss": 0.74491572,
      "learning_rate": 1.5657554885462055e-08,
      "loss": 0.76622337,
      "num_input_tokens_seen": 344960020,
      "step": 15989,
      "time_per_iteration": 2.4695193767547607
    },
    {
      "auxiliary_loss_clip": 0.010123,
      "auxiliary_loss_mlp": 0.01020138,
      "balance_loss_clip": 1.00980735,
      "balance_loss_mlp": 1.01887441,
      "epoch": 0.9613708101608297,
      "flos": 61563818522880.0,
      "grad_norm": 0.8091620993324363,
      "language_loss": 0.63139313,
      "learning_rate": 1.5608954583941737e-08,
      "loss": 0.65171754,
      "num_input_tokens_seen": 345018290,
      "step": 15990,
      "time_per_iteration": 3.047435998916626
    },
    {
      "auxiliary_loss_clip": 0.01090413,
      "auxiliary_loss_mlp": 0.01034602,
      "balance_loss_clip": 1.03304076,
      "balance_loss_mlp": 1.02302718,
      "epoch": 0.9614309334134977,
      "flos": 27417904922880.0,
      "grad_norm": 1.736660014773818,
      "language_loss": 0.77319348,
      "learning_rate": 1.5560429530729003e-08,
      "loss": 0.79444361,
      "num_input_tokens_seen": 345040235,
      "step": 15991,
      "time_per_iteration": 3.969411611557007
    },
    {
      "auxiliary_loss_clip": 0.01108074,
      "auxiliary_loss_mlp": 0.01031668,
      "balance_loss_clip": 1.03463066,
      "balance_loss_mlp": 1.01828063,
      "epoch": 0.9614910566661656,
      "flos": 22819148611200.0,
      "grad_norm": 2.3289571919437937,
      "language_loss": 0.84835327,
      "learning_rate": 1.5511979727663493e-08,
      "loss": 0.86975074,
      "num_input_tokens_seen": 345054540,
      "step": 15992,
      "time_per_iteration": 2.4499351978302
    },
    {
      "auxiliary_loss_clip": 0.01080153,
      "auxiliary_loss_mlp": 0.01030487,
      "balance_loss_clip": 1.03227544,
      "balance_loss_mlp": 1.01718283,
      "epoch": 0.9615511799188337,
      "flos": 20667812359680.0,
      "grad_norm": 2.0521653971536584,
      "language_loss": 0.73063666,
      "learning_rate": 1.5463605176582406e-08,
      "loss": 0.75174308,
      "num_input_tokens_seen": 345074035,
      "step": 15993,
      "time_per_iteration": 4.165271282196045
    },
    {
      "auxiliary_loss_clip": 0.01073132,
      "auxiliary_loss_mlp": 0.01028784,
      "balance_loss_clip": 1.03391886,
      "balance_loss_mlp": 1.01654696,
      "epoch": 0.9616113031715016,
      "flos": 33149212035840.0,
      "grad_norm": 1.6518585351993422,
      "language_loss": 0.68144357,
      "learning_rate": 1.5415305879320716e-08,
      "loss": 0.70246279,
      "num_input_tokens_seen": 345099270,
      "step": 15994,
      "time_per_iteration": 2.6639606952667236
    },
    {
      "auxiliary_loss_clip": 0.01073989,
      "auxiliary_loss_mlp": 0.0103186,
      "balance_loss_clip": 1.03685284,
      "balance_loss_mlp": 1.01962936,
      "epoch": 0.9616714264241696,
      "flos": 25009807276800.0,
      "grad_norm": 2.3184824465214935,
      "language_loss": 0.84501851,
      "learning_rate": 1.5367081837709183e-08,
      "loss": 0.86607707,
      "num_input_tokens_seen": 345116975,
      "step": 15995,
      "time_per_iteration": 2.596580982208252
    },
    {
      "auxiliary_loss_clip": 0.01095647,
      "auxiliary_loss_mlp": 0.01035199,
      "balance_loss_clip": 1.03456175,
      "balance_loss_mlp": 1.02234852,
      "epoch": 0.9617315496768375,
      "flos": 13547480359680.0,
      "grad_norm": 1.9553300907135167,
      "language_loss": 0.75885862,
      "learning_rate": 1.5318933053576788e-08,
      "loss": 0.7801671,
      "num_input_tokens_seen": 345133645,
      "step": 15996,
      "time_per_iteration": 2.4444386959075928
    },
    {
      "auxiliary_loss_clip": 0.01078589,
      "auxiliary_loss_mlp": 0.01030438,
      "balance_loss_clip": 1.03547835,
      "balance_loss_mlp": 1.01815367,
      "epoch": 0.9617916729295055,
      "flos": 11254512781440.0,
      "grad_norm": 1.9605969332909137,
      "language_loss": 0.76914352,
      "learning_rate": 1.52708595287494e-08,
      "loss": 0.79023385,
      "num_input_tokens_seen": 345150740,
      "step": 15997,
      "time_per_iteration": 2.5120232105255127
    },
    {
      "auxiliary_loss_clip": 0.0109791,
      "auxiliary_loss_mlp": 0.00782572,
      "balance_loss_clip": 1.03270364,
      "balance_loss_mlp": 1.00975525,
      "epoch": 0.9618517961821734,
      "flos": 22819723228800.0,
      "grad_norm": 1.585036739514364,
      "language_loss": 0.6729691,
      "learning_rate": 1.522286126505001e-08,
      "loss": 0.69177395,
      "num_input_tokens_seen": 345170365,
      "step": 15998,
      "time_per_iteration": 2.5352861881256104
    },
    {
      "auxiliary_loss_clip": 0.01075213,
      "auxiliary_loss_mlp": 0.01029996,
      "balance_loss_clip": 1.03105021,
      "balance_loss_mlp": 1.01700234,
      "epoch": 0.9619119194348414,
      "flos": 16617340224000.0,
      "grad_norm": 1.579153587394206,
      "language_loss": 0.7274133,
      "learning_rate": 1.5174938264298498e-08,
      "loss": 0.74846536,
      "num_input_tokens_seen": 345188930,
      "step": 15999,
      "time_per_iteration": 2.5189874172210693
    },
    {
      "auxiliary_loss_clip": 0.0107641,
      "auxiliary_loss_mlp": 0.01025306,
      "balance_loss_clip": 1.03311276,
      "balance_loss_mlp": 1.01410687,
      "epoch": 0.9619720426875094,
      "flos": 24535140024960.0,
      "grad_norm": 2.295223217282639,
      "language_loss": 0.65300161,
      "learning_rate": 1.5127090528312514e-08,
      "loss": 0.6740188,
      "num_input_tokens_seen": 345209615,
      "step": 16000,
      "time_per_iteration": 2.544492244720459
    },
    {
      "auxiliary_loss_clip": 0.01070786,
      "auxiliary_loss_mlp": 0.01028687,
      "balance_loss_clip": 1.03298163,
      "balance_loss_mlp": 1.01577115,
      "epoch": 0.9620321659401774,
      "flos": 20632224960000.0,
      "grad_norm": 1.9979351054199816,
      "language_loss": 0.75340462,
      "learning_rate": 1.5079318058905723e-08,
      "loss": 0.7743994,
      "num_input_tokens_seen": 345229175,
      "step": 16001,
      "time_per_iteration": 3.9537296295166016
    },
    {
      "auxiliary_loss_clip": 0.01089534,
      "auxiliary_loss_mlp": 0.01029612,
      "balance_loss_clip": 1.03340936,
      "balance_loss_mlp": 1.01716077,
      "epoch": 0.9620922891928453,
      "flos": 18515290959360.0,
      "grad_norm": 1.5556524399370568,
      "language_loss": 0.68205309,
      "learning_rate": 1.5031620857890447e-08,
      "loss": 0.70324457,
      "num_input_tokens_seen": 345247815,
      "step": 16002,
      "time_per_iteration": 2.483450412750244
    },
    {
      "auxiliary_loss_clip": 0.01092963,
      "auxiliary_loss_mlp": 0.01029228,
      "balance_loss_clip": 1.03608775,
      "balance_loss_mlp": 1.01745641,
      "epoch": 0.9621524124455133,
      "flos": 28767391914240.0,
      "grad_norm": 1.2731129115465383,
      "language_loss": 0.64648652,
      "learning_rate": 1.4983998927074804e-08,
      "loss": 0.6677084,
      "num_input_tokens_seen": 345269935,
      "step": 16003,
      "time_per_iteration": 2.543663263320923
    },
    {
      "auxiliary_loss_clip": 0.01061755,
      "auxiliary_loss_mlp": 0.01038772,
      "balance_loss_clip": 1.03590775,
      "balance_loss_mlp": 1.02702975,
      "epoch": 0.9622125356981813,
      "flos": 19098875226240.0,
      "grad_norm": 1.9404291412697514,
      "language_loss": 0.75785029,
      "learning_rate": 1.493645226826512e-08,
      "loss": 0.77885556,
      "num_input_tokens_seen": 345288310,
      "step": 16004,
      "time_per_iteration": 2.5748291015625
    },
    {
      "auxiliary_loss_clip": 0.01090997,
      "auxiliary_loss_mlp": 0.01030025,
      "balance_loss_clip": 1.03413069,
      "balance_loss_mlp": 1.01810384,
      "epoch": 0.9622726589508492,
      "flos": 20302816308480.0,
      "grad_norm": 1.8500444845330368,
      "language_loss": 0.78786063,
      "learning_rate": 1.4888980883263958e-08,
      "loss": 0.80907083,
      "num_input_tokens_seen": 345306615,
      "step": 16005,
      "time_per_iteration": 2.468085527420044
    },
    {
      "auxiliary_loss_clip": 0.01088616,
      "auxiliary_loss_mlp": 0.01025437,
      "balance_loss_clip": 1.03303885,
      "balance_loss_mlp": 1.01428545,
      "epoch": 0.9623327822035173,
      "flos": 54929750889600.0,
      "grad_norm": 1.9096305820117256,
      "language_loss": 0.67975008,
      "learning_rate": 1.4841584773871652e-08,
      "loss": 0.70089066,
      "num_input_tokens_seen": 345331935,
      "step": 16006,
      "time_per_iteration": 2.792125940322876
    },
    {
      "auxiliary_loss_clip": 0.01070538,
      "auxiliary_loss_mlp": 0.01033294,
      "balance_loss_clip": 1.03302443,
      "balance_loss_mlp": 1.02190971,
      "epoch": 0.9623929054561852,
      "flos": 21759029585280.0,
      "grad_norm": 1.5431727653498615,
      "language_loss": 0.783602,
      "learning_rate": 1.479426394188521e-08,
      "loss": 0.80464035,
      "num_input_tokens_seen": 345351510,
      "step": 16007,
      "time_per_iteration": 2.510564088821411
    },
    {
      "auxiliary_loss_clip": 0.01104866,
      "auxiliary_loss_mlp": 0.01030834,
      "balance_loss_clip": 1.03570259,
      "balance_loss_mlp": 1.01872826,
      "epoch": 0.9624530287088532,
      "flos": 17931563038080.0,
      "grad_norm": 1.8260463018719035,
      "language_loss": 0.67902362,
      "learning_rate": 1.4747018389099198e-08,
      "loss": 0.70038062,
      "num_input_tokens_seen": 345367750,
      "step": 16008,
      "time_per_iteration": 2.432136297225952
    },
    {
      "auxiliary_loss_clip": 0.01081661,
      "auxiliary_loss_mlp": 0.01031479,
      "balance_loss_clip": 1.03543043,
      "balance_loss_mlp": 1.01809788,
      "epoch": 0.9625131519615211,
      "flos": 23253739263360.0,
      "grad_norm": 2.245899471813522,
      "language_loss": 0.73096418,
      "learning_rate": 1.469984811730529e-08,
      "loss": 0.75209558,
      "num_input_tokens_seen": 345384790,
      "step": 16009,
      "time_per_iteration": 2.510167121887207
    },
    {
      "auxiliary_loss_clip": 0.0108865,
      "auxiliary_loss_mlp": 0.01030117,
      "balance_loss_clip": 1.03261375,
      "balance_loss_mlp": 1.01862526,
      "epoch": 0.9625732752141891,
      "flos": 18916628595840.0,
      "grad_norm": 1.7271619066245045,
      "language_loss": 0.75508302,
      "learning_rate": 1.4652753128292061e-08,
      "loss": 0.77627075,
      "num_input_tokens_seen": 345403390,
      "step": 16010,
      "time_per_iteration": 2.4993793964385986
    },
    {
      "auxiliary_loss_clip": 0.01096598,
      "auxiliary_loss_mlp": 0.01035571,
      "balance_loss_clip": 1.03814363,
      "balance_loss_mlp": 1.02055693,
      "epoch": 0.962633398466857,
      "flos": 16252918790400.0,
      "grad_norm": 1.8959726033850308,
      "language_loss": 0.69725734,
      "learning_rate": 1.4605733423845635e-08,
      "loss": 0.71857905,
      "num_input_tokens_seen": 345418685,
      "step": 16011,
      "time_per_iteration": 2.456112861633301
    },
    {
      "auxiliary_loss_clip": 0.01089991,
      "auxiliary_loss_mlp": 0.01031417,
      "balance_loss_clip": 1.03465378,
      "balance_loss_mlp": 1.02050924,
      "epoch": 0.962693521719525,
      "flos": 54197424403200.0,
      "grad_norm": 1.878334486520256,
      "language_loss": 0.68371403,
      "learning_rate": 1.4558789005748585e-08,
      "loss": 0.70492816,
      "num_input_tokens_seen": 345442380,
      "step": 16012,
      "time_per_iteration": 2.7931275367736816
    },
    {
      "auxiliary_loss_clip": 0.01086687,
      "auxiliary_loss_mlp": 0.01036203,
      "balance_loss_clip": 1.03297198,
      "balance_loss_mlp": 1.02254701,
      "epoch": 0.962753644972193,
      "flos": 33105795471360.0,
      "grad_norm": 1.731118946445511,
      "language_loss": 0.72635806,
      "learning_rate": 1.4511919875781264e-08,
      "loss": 0.74758697,
      "num_input_tokens_seen": 345463815,
      "step": 16013,
      "time_per_iteration": 2.5931649208068848
    },
    {
      "auxiliary_loss_clip": 0.01077734,
      "auxiliary_loss_mlp": 0.01031518,
      "balance_loss_clip": 1.03517985,
      "balance_loss_mlp": 1.01872134,
      "epoch": 0.962813768224861,
      "flos": 42230660837760.0,
      "grad_norm": 2.778854341437809,
      "language_loss": 0.63048095,
      "learning_rate": 1.4465126035720698e-08,
      "loss": 0.65157342,
      "num_input_tokens_seen": 345484525,
      "step": 16014,
      "time_per_iteration": 2.699995756149292
    },
    {
      "auxiliary_loss_clip": 0.01077691,
      "auxiliary_loss_mlp": 0.01028238,
      "balance_loss_clip": 1.03426814,
      "balance_loss_mlp": 1.01780152,
      "epoch": 0.9628738914775289,
      "flos": 43944677003520.0,
      "grad_norm": 1.63361480034155,
      "language_loss": 0.71975005,
      "learning_rate": 1.4418407487341688e-08,
      "loss": 0.74080932,
      "num_input_tokens_seen": 345508295,
      "step": 16015,
      "time_per_iteration": 2.7087485790252686
    },
    {
      "auxiliary_loss_clip": 0.0106734,
      "auxiliary_loss_mlp": 0.01029584,
      "balance_loss_clip": 1.03163409,
      "balance_loss_mlp": 1.01737118,
      "epoch": 0.9629340147301969,
      "flos": 15596184476160.0,
      "grad_norm": 1.9225120526689747,
      "language_loss": 0.769508,
      "learning_rate": 1.4371764232415707e-08,
      "loss": 0.79047722,
      "num_input_tokens_seen": 345525155,
      "step": 16016,
      "time_per_iteration": 2.5729217529296875
    },
    {
      "auxiliary_loss_clip": 0.0102687,
      "auxiliary_loss_mlp": 0.01004347,
      "balance_loss_clip": 1.00416875,
      "balance_loss_mlp": 1.00331616,
      "epoch": 0.9629941379828649,
      "flos": 62951011816320.0,
      "grad_norm": 0.8115399115612564,
      "language_loss": 0.63143641,
      "learning_rate": 1.4325196272711337e-08,
      "loss": 0.6517486,
      "num_input_tokens_seen": 345578905,
      "step": 16017,
      "time_per_iteration": 2.9742543697357178
    },
    {
      "auxiliary_loss_clip": 0.01085796,
      "auxiliary_loss_mlp": 0.01028099,
      "balance_loss_clip": 1.03444672,
      "balance_loss_mlp": 1.01633906,
      "epoch": 0.9630542612355328,
      "flos": 29899116702720.0,
      "grad_norm": 1.9855174264081203,
      "language_loss": 0.66455615,
      "learning_rate": 1.4278703609994502e-08,
      "loss": 0.68569517,
      "num_input_tokens_seen": 345598965,
      "step": 16018,
      "time_per_iteration": 2.60267972946167
    },
    {
      "auxiliary_loss_clip": 0.01061435,
      "auxiliary_loss_mlp": 0.010349,
      "balance_loss_clip": 1.03517842,
      "balance_loss_mlp": 1.02311563,
      "epoch": 0.9631143844882009,
      "flos": 17894575008000.0,
      "grad_norm": 2.073459052462633,
      "language_loss": 0.79427183,
      "learning_rate": 1.4232286246028457e-08,
      "loss": 0.81523514,
      "num_input_tokens_seen": 345617945,
      "step": 16019,
      "time_per_iteration": 2.5372681617736816
    },
    {
      "auxiliary_loss_clip": 0.01065265,
      "auxiliary_loss_mlp": 0.01031911,
      "balance_loss_clip": 1.0317812,
      "balance_loss_mlp": 1.0211463,
      "epoch": 0.9631745077408688,
      "flos": 26139161767680.0,
      "grad_norm": 1.4303457450508714,
      "language_loss": 0.71911711,
      "learning_rate": 1.4185944182572907e-08,
      "loss": 0.74008882,
      "num_input_tokens_seen": 345637920,
      "step": 16020,
      "time_per_iteration": 2.5722877979278564
    },
    {
      "auxiliary_loss_clip": 0.01079977,
      "auxiliary_loss_mlp": 0.01025032,
      "balance_loss_clip": 1.03495073,
      "balance_loss_mlp": 1.01406467,
      "epoch": 0.9632346309935368,
      "flos": 24973645259520.0,
      "grad_norm": 1.628393952796008,
      "language_loss": 0.76930571,
      "learning_rate": 1.4139677421385331e-08,
      "loss": 0.7903558,
      "num_input_tokens_seen": 345656195,
      "step": 16021,
      "time_per_iteration": 2.5457935333251953
    },
    {
      "auxiliary_loss_clip": 0.0107003,
      "auxiliary_loss_mlp": 0.01033381,
      "balance_loss_clip": 1.03349519,
      "balance_loss_mlp": 1.0187068,
      "epoch": 0.9632947542462047,
      "flos": 23617226943360.0,
      "grad_norm": 2.0914329531886118,
      "language_loss": 0.65242612,
      "learning_rate": 1.4093485964220331e-08,
      "loss": 0.67346025,
      "num_input_tokens_seen": 345676700,
      "step": 16022,
      "time_per_iteration": 2.5744314193725586
    },
    {
      "auxiliary_loss_clip": 0.01075955,
      "auxiliary_loss_mlp": 0.01035585,
      "balance_loss_clip": 1.03203225,
      "balance_loss_mlp": 1.02371228,
      "epoch": 0.9633548774988727,
      "flos": 26395599939840.0,
      "grad_norm": 1.821145908043204,
      "language_loss": 0.73258787,
      "learning_rate": 1.4047369812829168e-08,
      "loss": 0.75370324,
      "num_input_tokens_seen": 345696725,
      "step": 16023,
      "time_per_iteration": 2.5300230979919434
    },
    {
      "auxiliary_loss_clip": 0.01087407,
      "auxiliary_loss_mlp": 0.01030457,
      "balance_loss_clip": 1.03286123,
      "balance_loss_mlp": 1.01773155,
      "epoch": 0.9634150007515406,
      "flos": 23767728929280.0,
      "grad_norm": 1.9613630380762088,
      "language_loss": 0.81598991,
      "learning_rate": 1.4001328968960891e-08,
      "loss": 0.83716857,
      "num_input_tokens_seen": 345716245,
      "step": 16024,
      "time_per_iteration": 2.507746696472168
    },
    {
      "auxiliary_loss_clip": 0.0109613,
      "auxiliary_loss_mlp": 0.01034797,
      "balance_loss_clip": 1.03464961,
      "balance_loss_mlp": 1.0223639,
      "epoch": 0.9634751240042086,
      "flos": 24135346673280.0,
      "grad_norm": 1.4016839376822616,
      "language_loss": 0.8145498,
      "learning_rate": 1.3955363434361212e-08,
      "loss": 0.83585906,
      "num_input_tokens_seen": 345739060,
      "step": 16025,
      "time_per_iteration": 2.5009255409240723
    },
    {
      "auxiliary_loss_clip": 0.010952,
      "auxiliary_loss_mlp": 0.01028978,
      "balance_loss_clip": 1.03439975,
      "balance_loss_mlp": 1.01700997,
      "epoch": 0.9635352472568766,
      "flos": 24349086552960.0,
      "grad_norm": 2.0098562156821296,
      "language_loss": 0.76277959,
      "learning_rate": 1.3909473210773181e-08,
      "loss": 0.78402138,
      "num_input_tokens_seen": 345758325,
      "step": 16026,
      "time_per_iteration": 2.514521360397339
    },
    {
      "auxiliary_loss_clip": 0.01069004,
      "auxiliary_loss_mlp": 0.00784612,
      "balance_loss_clip": 1.03211665,
      "balance_loss_mlp": 1.0074358,
      "epoch": 0.9635953705095446,
      "flos": 23984772860160.0,
      "grad_norm": 1.5768420732893675,
      "language_loss": 0.63350654,
      "learning_rate": 1.3863658299936965e-08,
      "loss": 0.65204269,
      "num_input_tokens_seen": 345778530,
      "step": 16027,
      "time_per_iteration": 2.5627801418304443
    },
    {
      "auxiliary_loss_clip": 0.01094791,
      "auxiliary_loss_mlp": 0.01027176,
      "balance_loss_clip": 1.03559721,
      "balance_loss_mlp": 1.01517117,
      "epoch": 0.9636554937622125,
      "flos": 19828436365440.0,
      "grad_norm": 1.859214015578949,
      "language_loss": 0.869506,
      "learning_rate": 1.3817918703589837e-08,
      "loss": 0.89072573,
      "num_input_tokens_seen": 345796535,
      "step": 16028,
      "time_per_iteration": 2.490760564804077
    },
    {
      "auxiliary_loss_clip": 0.01001298,
      "auxiliary_loss_mlp": 0.01006773,
      "balance_loss_clip": 1.02766299,
      "balance_loss_mlp": 1.00533056,
      "epoch": 0.9637156170148805,
      "flos": 67435499986560.0,
      "grad_norm": 0.6860139976300149,
      "language_loss": 0.531847,
      "learning_rate": 1.3772254423466412e-08,
      "loss": 0.55192769,
      "num_input_tokens_seen": 345859700,
      "step": 16029,
      "time_per_iteration": 3.1740922927856445
    },
    {
      "auxiliary_loss_clip": 0.01104919,
      "auxiliary_loss_mlp": 0.01030065,
      "balance_loss_clip": 1.03510535,
      "balance_loss_mlp": 1.01799488,
      "epoch": 0.9637757402675484,
      "flos": 20300912887680.0,
      "grad_norm": 2.2162493607025335,
      "language_loss": 0.73984313,
      "learning_rate": 1.372666546129797e-08,
      "loss": 0.76119298,
      "num_input_tokens_seen": 345878760,
      "step": 16030,
      "time_per_iteration": 5.244468450546265
    },
    {
      "auxiliary_loss_clip": 0.01079048,
      "auxiliary_loss_mlp": 0.01034505,
      "balance_loss_clip": 1.03460979,
      "balance_loss_mlp": 1.02269697,
      "epoch": 0.9638358635202164,
      "flos": 27234544970880.0,
      "grad_norm": 2.214385269747224,
      "language_loss": 0.66055197,
      "learning_rate": 1.3681151818813575e-08,
      "loss": 0.68168747,
      "num_input_tokens_seen": 345900445,
      "step": 16031,
      "time_per_iteration": 3.976991653442383
    },
    {
      "auxiliary_loss_clip": 0.01017804,
      "auxiliary_loss_mlp": 0.00760268,
      "balance_loss_clip": 1.00565612,
      "balance_loss_mlp": 0.99955767,
      "epoch": 0.9638959867728845,
      "flos": 70288998278400.0,
      "grad_norm": 0.8373816030118958,
      "language_loss": 0.60722744,
      "learning_rate": 1.3635713497738955e-08,
      "loss": 0.62500823,
      "num_input_tokens_seen": 345961020,
      "step": 16032,
      "time_per_iteration": 3.1391727924346924
    },
    {
      "auxiliary_loss_clip": 0.0108329,
      "auxiliary_loss_mlp": 0.01030017,
      "balance_loss_clip": 1.03397071,
      "balance_loss_mlp": 1.01962185,
      "epoch": 0.9639561100255524,
      "flos": 25407517639680.0,
      "grad_norm": 1.7941771309154309,
      "language_loss": 0.66085958,
      "learning_rate": 1.3590350499796954e-08,
      "loss": 0.68199265,
      "num_input_tokens_seen": 345980210,
      "step": 16033,
      "time_per_iteration": 2.5555052757263184
    },
    {
      "auxiliary_loss_clip": 0.01047018,
      "auxiliary_loss_mlp": 0.01028915,
      "balance_loss_clip": 1.03411794,
      "balance_loss_mlp": 1.01719069,
      "epoch": 0.9640162332782204,
      "flos": 18113881495680.0,
      "grad_norm": 1.7592433679242558,
      "language_loss": 0.65426296,
      "learning_rate": 1.3545062826707976e-08,
      "loss": 0.67502236,
      "num_input_tokens_seen": 345998280,
      "step": 16034,
      "time_per_iteration": 2.609494686126709
    },
    {
      "auxiliary_loss_clip": 0.01061712,
      "auxiliary_loss_mlp": 0.01029196,
      "balance_loss_clip": 1.0333724,
      "balance_loss_mlp": 1.01664948,
      "epoch": 0.9640763565308883,
      "flos": 23440295525760.0,
      "grad_norm": 2.105545482063162,
      "language_loss": 0.73842007,
      "learning_rate": 1.3499850480189313e-08,
      "loss": 0.75932914,
      "num_input_tokens_seen": 346015545,
      "step": 16035,
      "time_per_iteration": 2.608625650405884
    },
    {
      "auxiliary_loss_clip": 0.01104809,
      "auxiliary_loss_mlp": 0.01029193,
      "balance_loss_clip": 1.03692293,
      "balance_loss_mlp": 1.01731372,
      "epoch": 0.9641364797835563,
      "flos": 22419355259520.0,
      "grad_norm": 2.75658569667586,
      "language_loss": 0.82099104,
      "learning_rate": 1.3454713461955591e-08,
      "loss": 0.84233105,
      "num_input_tokens_seen": 346034055,
      "step": 16036,
      "time_per_iteration": 2.4736435413360596
    },
    {
      "auxiliary_loss_clip": 0.0108276,
      "auxiliary_loss_mlp": 0.01029271,
      "balance_loss_clip": 1.03478026,
      "balance_loss_mlp": 1.01711774,
      "epoch": 0.9641966030362242,
      "flos": 30622357048320.0,
      "grad_norm": 1.7135482608023438,
      "language_loss": 0.69832861,
      "learning_rate": 1.340965177371789e-08,
      "loss": 0.71944892,
      "num_input_tokens_seen": 346054130,
      "step": 16037,
      "time_per_iteration": 2.6174240112304688
    },
    {
      "auxiliary_loss_clip": 0.01102541,
      "auxiliary_loss_mlp": 0.01028237,
      "balance_loss_clip": 1.03352141,
      "balance_loss_mlp": 1.01644683,
      "epoch": 0.9642567262888923,
      "flos": 20953122088320.0,
      "grad_norm": 1.719510332759985,
      "language_loss": 0.63041854,
      "learning_rate": 1.3364665417185506e-08,
      "loss": 0.65172637,
      "num_input_tokens_seen": 346072990,
      "step": 16038,
      "time_per_iteration": 2.503009557723999
    },
    {
      "auxiliary_loss_clip": 0.01069308,
      "auxiliary_loss_mlp": 0.00786582,
      "balance_loss_clip": 1.03230166,
      "balance_loss_mlp": 1.00921392,
      "epoch": 0.9643168495415602,
      "flos": 22639415932800.0,
      "grad_norm": 1.6365645649626694,
      "language_loss": 0.70971847,
      "learning_rate": 1.3319754394064187e-08,
      "loss": 0.72827733,
      "num_input_tokens_seen": 346093745,
      "step": 16039,
      "time_per_iteration": 3.98201847076416
    },
    {
      "auxiliary_loss_clip": 0.01060354,
      "auxiliary_loss_mlp": 0.01029412,
      "balance_loss_clip": 1.03388357,
      "balance_loss_mlp": 1.01689517,
      "epoch": 0.9643769727942282,
      "flos": 20266259241600.0,
      "grad_norm": 2.1578932189031725,
      "language_loss": 0.72850174,
      "learning_rate": 1.327491870605657e-08,
      "loss": 0.74939942,
      "num_input_tokens_seen": 346110115,
      "step": 16040,
      "time_per_iteration": 2.547440528869629
    },
    {
      "auxiliary_loss_clip": 0.0109412,
      "auxiliary_loss_mlp": 0.01031711,
      "balance_loss_clip": 1.03404701,
      "balance_loss_mlp": 1.01905107,
      "epoch": 0.9644370960468961,
      "flos": 13881845088000.0,
      "grad_norm": 2.4574770166315267,
      "language_loss": 0.72739863,
      "learning_rate": 1.3230158354863296e-08,
      "loss": 0.74865699,
      "num_input_tokens_seen": 346127165,
      "step": 16041,
      "time_per_iteration": 2.471254825592041
    },
    {
      "auxiliary_loss_clip": 0.01076602,
      "auxiliary_loss_mlp": 0.01029868,
      "balance_loss_clip": 1.03291535,
      "balance_loss_mlp": 1.01853085,
      "epoch": 0.9644972192995641,
      "flos": 17238199829760.0,
      "grad_norm": 1.8901939294794061,
      "language_loss": 0.72061896,
      "learning_rate": 1.3185473342181674e-08,
      "loss": 0.74168372,
      "num_input_tokens_seen": 346145950,
      "step": 16042,
      "time_per_iteration": 2.48492169380188
    },
    {
      "auxiliary_loss_clip": 0.0106724,
      "auxiliary_loss_mlp": 0.01032555,
      "balance_loss_clip": 1.03513896,
      "balance_loss_mlp": 1.02062249,
      "epoch": 0.964557342552232,
      "flos": 23840340272640.0,
      "grad_norm": 1.8924139978818204,
      "language_loss": 0.81031954,
      "learning_rate": 1.3140863669705683e-08,
      "loss": 0.83131742,
      "num_input_tokens_seen": 346165005,
      "step": 16043,
      "time_per_iteration": 2.5964367389678955
    },
    {
      "auxiliary_loss_clip": 0.01075135,
      "auxiliary_loss_mlp": 0.01029117,
      "balance_loss_clip": 1.03413963,
      "balance_loss_mlp": 1.01780403,
      "epoch": 0.9646174658049,
      "flos": 21653129312640.0,
      "grad_norm": 1.536026390570106,
      "language_loss": 0.71552771,
      "learning_rate": 1.3096329339127522e-08,
      "loss": 0.73657024,
      "num_input_tokens_seen": 346185095,
      "step": 16044,
      "time_per_iteration": 2.523463726043701
    },
    {
      "auxiliary_loss_clip": 0.01077659,
      "auxiliary_loss_mlp": 0.01029084,
      "balance_loss_clip": 1.03226638,
      "balance_loss_mlp": 1.0169307,
      "epoch": 0.9646775890575681,
      "flos": 17129570123520.0,
      "grad_norm": 2.0645774137715485,
      "language_loss": 0.70167834,
      "learning_rate": 1.3051870352135397e-08,
      "loss": 0.72274578,
      "num_input_tokens_seen": 346202580,
      "step": 16045,
      "time_per_iteration": 2.5066866874694824
    },
    {
      "auxiliary_loss_clip": 0.01043618,
      "auxiliary_loss_mlp": 0.01031785,
      "balance_loss_clip": 1.03361154,
      "balance_loss_mlp": 1.01861858,
      "epoch": 0.964737712310236,
      "flos": 13005732458880.0,
      "grad_norm": 1.682924991280473,
      "language_loss": 0.75218034,
      "learning_rate": 1.3007486710415737e-08,
      "loss": 0.77293438,
      "num_input_tokens_seen": 346219395,
      "step": 16046,
      "time_per_iteration": 2.566162347793579
    },
    {
      "auxiliary_loss_clip": 0.01096515,
      "auxiliary_loss_mlp": 0.01035977,
      "balance_loss_clip": 1.03469706,
      "balance_loss_mlp": 1.02279246,
      "epoch": 0.964797835562904,
      "flos": 24279240556800.0,
      "grad_norm": 1.5159700532315727,
      "language_loss": 0.623052,
      "learning_rate": 1.2963178415651199e-08,
      "loss": 0.64437699,
      "num_input_tokens_seen": 346239715,
      "step": 16047,
      "time_per_iteration": 2.519449234008789
    },
    {
      "auxiliary_loss_clip": 0.01082922,
      "auxiliary_loss_mlp": 0.0103131,
      "balance_loss_clip": 1.03742766,
      "balance_loss_mlp": 1.0194428,
      "epoch": 0.9648579588155719,
      "flos": 20522697413760.0,
      "grad_norm": 1.895462187743246,
      "language_loss": 0.69404483,
      "learning_rate": 1.2918945469521992e-08,
      "loss": 0.71518713,
      "num_input_tokens_seen": 346258500,
      "step": 16048,
      "time_per_iteration": 2.5210745334625244
    },
    {
      "auxiliary_loss_clip": 0.01094116,
      "auxiliary_loss_mlp": 0.01027812,
      "balance_loss_clip": 1.03453541,
      "balance_loss_mlp": 1.01489019,
      "epoch": 0.9649180820682399,
      "flos": 32154844855680.0,
      "grad_norm": 2.05393091038773,
      "language_loss": 0.63770831,
      "learning_rate": 1.2874787873705662e-08,
      "loss": 0.65892756,
      "num_input_tokens_seen": 346279110,
      "step": 16049,
      "time_per_iteration": 2.577864408493042
    },
    {
      "auxiliary_loss_clip": 0.01094665,
      "auxiliary_loss_mlp": 0.01027019,
      "balance_loss_clip": 1.03593612,
      "balance_loss_mlp": 1.01499057,
      "epoch": 0.9649782053209078,
      "flos": 20522589672960.0,
      "grad_norm": 1.7497546467349627,
      "language_loss": 0.70840919,
      "learning_rate": 1.2830705629876427e-08,
      "loss": 0.72962606,
      "num_input_tokens_seen": 346297860,
      "step": 16050,
      "time_per_iteration": 2.467543840408325
    },
    {
      "auxiliary_loss_clip": 0.01093608,
      "auxiliary_loss_mlp": 0.01032732,
      "balance_loss_clip": 1.03193521,
      "balance_loss_mlp": 1.01920199,
      "epoch": 0.9650383285735759,
      "flos": 43067953843200.0,
      "grad_norm": 2.7194820931117425,
      "language_loss": 0.69797409,
      "learning_rate": 1.278669873970606e-08,
      "loss": 0.71923745,
      "num_input_tokens_seen": 346319860,
      "step": 16051,
      "time_per_iteration": 2.680999279022217
    },
    {
      "auxiliary_loss_clip": 0.01018274,
      "auxiliary_loss_mlp": 0.0100625,
      "balance_loss_clip": 1.00593352,
      "balance_loss_mlp": 1.0051173,
      "epoch": 0.9650984518262438,
      "flos": 61748255882880.0,
      "grad_norm": 0.8442510368605294,
      "language_loss": 0.59144664,
      "learning_rate": 1.2742767204863004e-08,
      "loss": 0.61169183,
      "num_input_tokens_seen": 346379025,
      "step": 16052,
      "time_per_iteration": 3.124314308166504
    },
    {
      "auxiliary_loss_clip": 0.01098806,
      "auxiliary_loss_mlp": 0.01024842,
      "balance_loss_clip": 1.03285718,
      "balance_loss_mlp": 1.01314783,
      "epoch": 0.9651585750789118,
      "flos": 29789337761280.0,
      "grad_norm": 1.7648483056059674,
      "language_loss": 0.74730718,
      "learning_rate": 1.2698911027013482e-08,
      "loss": 0.76854372,
      "num_input_tokens_seen": 346402250,
      "step": 16053,
      "time_per_iteration": 2.5330936908721924
    },
    {
      "auxiliary_loss_clip": 0.01079424,
      "auxiliary_loss_mlp": 0.01033237,
      "balance_loss_clip": 1.03395748,
      "balance_loss_mlp": 1.02102435,
      "epoch": 0.9652186983315797,
      "flos": 16873060124160.0,
      "grad_norm": 2.513919907141199,
      "language_loss": 0.68465644,
      "learning_rate": 1.2655130207820386e-08,
      "loss": 0.70578307,
      "num_input_tokens_seen": 346419555,
      "step": 16054,
      "time_per_iteration": 2.5026016235351562
    },
    {
      "auxiliary_loss_clip": 0.0108577,
      "auxiliary_loss_mlp": 0.00781236,
      "balance_loss_clip": 1.03519559,
      "balance_loss_mlp": 1.00658727,
      "epoch": 0.9652788215842477,
      "flos": 31649761762560.0,
      "grad_norm": 1.5696573250001915,
      "language_loss": 0.62118304,
      "learning_rate": 1.2611424748943944e-08,
      "loss": 0.63985312,
      "num_input_tokens_seen": 346441245,
      "step": 16055,
      "time_per_iteration": 2.5683608055114746
    },
    {
      "auxiliary_loss_clip": 0.0106345,
      "auxiliary_loss_mlp": 0.01030807,
      "balance_loss_clip": 1.03442931,
      "balance_loss_mlp": 1.01902914,
      "epoch": 0.9653389448369156,
      "flos": 24754266944640.0,
      "grad_norm": 1.898937599243159,
      "language_loss": 0.76818621,
      "learning_rate": 1.2567794652041719e-08,
      "loss": 0.78912878,
      "num_input_tokens_seen": 346460065,
      "step": 16056,
      "time_per_iteration": 2.632518768310547
    },
    {
      "auxiliary_loss_clip": 0.01077242,
      "auxiliary_loss_mlp": 0.01031797,
      "balance_loss_clip": 1.03546214,
      "balance_loss_mlp": 1.02002561,
      "epoch": 0.9653990680895836,
      "flos": 20297249700480.0,
      "grad_norm": 1.5492937143879804,
      "language_loss": 0.71686149,
      "learning_rate": 1.2524239918767498e-08,
      "loss": 0.73795193,
      "num_input_tokens_seen": 346478005,
      "step": 16057,
      "time_per_iteration": 2.5222768783569336
    },
    {
      "auxiliary_loss_clip": 0.01101442,
      "auxiliary_loss_mlp": 0.01031319,
      "balance_loss_clip": 1.03462291,
      "balance_loss_mlp": 1.01983881,
      "epoch": 0.9654591913422517,
      "flos": 22528775064960.0,
      "grad_norm": 1.7947857022493083,
      "language_loss": 0.71822476,
      "learning_rate": 1.2480760550773295e-08,
      "loss": 0.73955238,
      "num_input_tokens_seen": 346497575,
      "step": 16058,
      "time_per_iteration": 2.492741107940674
    },
    {
      "auxiliary_loss_clip": 0.01089611,
      "auxiliary_loss_mlp": 0.01032955,
      "balance_loss_clip": 1.03328574,
      "balance_loss_mlp": 1.021505,
      "epoch": 0.9655193145949196,
      "flos": 26763002202240.0,
      "grad_norm": 1.4895501815590615,
      "language_loss": 0.74173975,
      "learning_rate": 1.2437356549708011e-08,
      "loss": 0.76296544,
      "num_input_tokens_seen": 346520000,
      "step": 16059,
      "time_per_iteration": 2.528712034225464
    },
    {
      "auxiliary_loss_clip": 0.01084826,
      "auxiliary_loss_mlp": 0.01028778,
      "balance_loss_clip": 1.03440154,
      "balance_loss_mlp": 1.01728654,
      "epoch": 0.9655794378475876,
      "flos": 41970703132800.0,
      "grad_norm": 1.8434447996818184,
      "language_loss": 0.73492503,
      "learning_rate": 1.239402791721722e-08,
      "loss": 0.75606114,
      "num_input_tokens_seen": 346541605,
      "step": 16060,
      "time_per_iteration": 2.6995675563812256
    },
    {
      "auxiliary_loss_clip": 0.0107825,
      "auxiliary_loss_mlp": 0.01032354,
      "balance_loss_clip": 1.0339483,
      "balance_loss_mlp": 1.02132678,
      "epoch": 0.9656395611002555,
      "flos": 27709427704320.0,
      "grad_norm": 1.6098017139710945,
      "language_loss": 0.76768136,
      "learning_rate": 1.2350774654944273e-08,
      "loss": 0.78878736,
      "num_input_tokens_seen": 346560955,
      "step": 16061,
      "time_per_iteration": 2.5559070110321045
    },
    {
      "auxiliary_loss_clip": 0.01009471,
      "auxiliary_loss_mlp": 0.01001463,
      "balance_loss_clip": 1.00665975,
      "balance_loss_mlp": 1.00026476,
      "epoch": 0.9656996843529235,
      "flos": 68968562411520.0,
      "grad_norm": 0.7765228807830403,
      "language_loss": 0.64146769,
      "learning_rate": 1.2307596764528749e-08,
      "loss": 0.66157705,
      "num_input_tokens_seen": 346621615,
      "step": 16062,
      "time_per_iteration": 3.1695029735565186
    },
    {
      "auxiliary_loss_clip": 0.01052235,
      "auxiliary_loss_mlp": 0.01026485,
      "balance_loss_clip": 1.0310998,
      "balance_loss_mlp": 1.01564896,
      "epoch": 0.9657598076055914,
      "flos": 20631327120000.0,
      "grad_norm": 2.18018496882028,
      "language_loss": 0.93415219,
      "learning_rate": 1.226449424760867e-08,
      "loss": 0.95493937,
      "num_input_tokens_seen": 346637460,
      "step": 16063,
      "time_per_iteration": 2.551731586456299
    },
    {
      "auxiliary_loss_clip": 0.01094486,
      "auxiliary_loss_mlp": 0.01031195,
      "balance_loss_clip": 1.03601265,
      "balance_loss_mlp": 1.01969182,
      "epoch": 0.9658199308582595,
      "flos": 20448577699200.0,
      "grad_norm": 1.765581965309665,
      "language_loss": 0.82041323,
      "learning_rate": 1.2221467105818062e-08,
      "loss": 0.84167004,
      "num_input_tokens_seen": 346655625,
      "step": 16064,
      "time_per_iteration": 2.4939305782318115
    },
    {
      "auxiliary_loss_clip": 0.01091032,
      "auxiliary_loss_mlp": 0.00786212,
      "balance_loss_clip": 1.03674006,
      "balance_loss_mlp": 1.01503956,
      "epoch": 0.9658800541109274,
      "flos": 24718033100160.0,
      "grad_norm": 1.5564173287897225,
      "language_loss": 0.83844769,
      "learning_rate": 1.2178515340788731e-08,
      "loss": 0.85722017,
      "num_input_tokens_seen": 346675220,
      "step": 16065,
      "time_per_iteration": 2.5282247066497803
    },
    {
      "auxiliary_loss_clip": 0.01078713,
      "auxiliary_loss_mlp": 0.01028116,
      "balance_loss_clip": 1.03321064,
      "balance_loss_mlp": 1.01612997,
      "epoch": 0.9659401773635954,
      "flos": 21610035970560.0,
      "grad_norm": 1.7385829003824058,
      "language_loss": 0.67467868,
      "learning_rate": 1.2135638954149151e-08,
      "loss": 0.69574702,
      "num_input_tokens_seen": 346694710,
      "step": 16066,
      "time_per_iteration": 2.527076005935669
    },
    {
      "auxiliary_loss_clip": 0.01101187,
      "auxiliary_loss_mlp": 0.01025275,
      "balance_loss_clip": 1.03304851,
      "balance_loss_mlp": 1.01374722,
      "epoch": 0.9660003006162633,
      "flos": 20301200196480.0,
      "grad_norm": 2.2684463574282057,
      "language_loss": 0.82199621,
      "learning_rate": 1.209283794752558e-08,
      "loss": 0.84326077,
      "num_input_tokens_seen": 346712645,
      "step": 16067,
      "time_per_iteration": 2.4386308193206787
    },
    {
      "auxiliary_loss_clip": 0.01080632,
      "auxiliary_loss_mlp": 0.01028855,
      "balance_loss_clip": 1.03383684,
      "balance_loss_mlp": 1.01689863,
      "epoch": 0.9660604238689313,
      "flos": 24461954064000.0,
      "grad_norm": 1.713301006719681,
      "language_loss": 0.69206065,
      "learning_rate": 1.2050112322540496e-08,
      "loss": 0.71315551,
      "num_input_tokens_seen": 346732375,
      "step": 16068,
      "time_per_iteration": 2.550292491912842
    },
    {
      "auxiliary_loss_clip": 0.01081228,
      "auxiliary_loss_mlp": 0.01030202,
      "balance_loss_clip": 1.03177166,
      "balance_loss_mlp": 1.01962829,
      "epoch": 0.9661205471215992,
      "flos": 19864023765120.0,
      "grad_norm": 1.859611012912049,
      "language_loss": 0.67774498,
      "learning_rate": 1.20074620808146e-08,
      "loss": 0.69885933,
      "num_input_tokens_seen": 346750430,
      "step": 16069,
      "time_per_iteration": 5.208314657211304
    },
    {
      "auxiliary_loss_clip": 0.01082576,
      "auxiliary_loss_mlp": 0.01026509,
      "balance_loss_clip": 1.03598797,
      "balance_loss_mlp": 1.0153985,
      "epoch": 0.9661806703742672,
      "flos": 20557889763840.0,
      "grad_norm": 1.8254223981597708,
      "language_loss": 0.89094281,
      "learning_rate": 1.1964887223964826e-08,
      "loss": 0.91203368,
      "num_input_tokens_seen": 346768455,
      "step": 16070,
      "time_per_iteration": 4.080590724945068
    },
    {
      "auxiliary_loss_clip": 0.01106703,
      "auxiliary_loss_mlp": 0.01035142,
      "balance_loss_clip": 1.03750825,
      "balance_loss_mlp": 1.02238095,
      "epoch": 0.9662407936269353,
      "flos": 21430949736960.0,
      "grad_norm": 1.7723512383639426,
      "language_loss": 0.76999915,
      "learning_rate": 1.1922387753605878e-08,
      "loss": 0.7914176,
      "num_input_tokens_seen": 346786530,
      "step": 16071,
      "time_per_iteration": 2.458963632583618
    },
    {
      "auxiliary_loss_clip": 0.01077824,
      "auxiliary_loss_mlp": 0.01034393,
      "balance_loss_clip": 1.03190231,
      "balance_loss_mlp": 1.02143502,
      "epoch": 0.9663009168796032,
      "flos": 14902893095040.0,
      "grad_norm": 1.7176188029169175,
      "language_loss": 0.6567533,
      "learning_rate": 1.1879963671349137e-08,
      "loss": 0.67787552,
      "num_input_tokens_seen": 346804635,
      "step": 16072,
      "time_per_iteration": 2.526878833770752
    },
    {
      "auxiliary_loss_clip": 0.01095983,
      "auxiliary_loss_mlp": 0.01032013,
      "balance_loss_clip": 1.03576386,
      "balance_loss_mlp": 1.0204556,
      "epoch": 0.9663610401322712,
      "flos": 24310877460480.0,
      "grad_norm": 1.6342111261647885,
      "language_loss": 0.7751075,
      "learning_rate": 1.1837614978803534e-08,
      "loss": 0.79638743,
      "num_input_tokens_seen": 346823070,
      "step": 16073,
      "time_per_iteration": 2.5119826793670654
    },
    {
      "auxiliary_loss_clip": 0.01106046,
      "auxiliary_loss_mlp": 0.01033508,
      "balance_loss_clip": 1.03578663,
      "balance_loss_mlp": 1.02127159,
      "epoch": 0.9664211633849391,
      "flos": 17637849527040.0,
      "grad_norm": 2.498468990656494,
      "language_loss": 0.76165718,
      "learning_rate": 1.1795341677574677e-08,
      "loss": 0.78305268,
      "num_input_tokens_seen": 346841180,
      "step": 16074,
      "time_per_iteration": 2.4602203369140625
    },
    {
      "auxiliary_loss_clip": 0.01081407,
      "auxiliary_loss_mlp": 0.01030141,
      "balance_loss_clip": 1.03600907,
      "balance_loss_mlp": 1.01768339,
      "epoch": 0.9664812866376071,
      "flos": 29789409588480.0,
      "grad_norm": 1.5865302923346496,
      "language_loss": 0.75713927,
      "learning_rate": 1.1753143769265728e-08,
      "loss": 0.77825475,
      "num_input_tokens_seen": 346864250,
      "step": 16075,
      "time_per_iteration": 2.570364236831665
    },
    {
      "auxiliary_loss_clip": 0.01074208,
      "auxiliary_loss_mlp": 0.01033899,
      "balance_loss_clip": 1.03673363,
      "balance_loss_mlp": 1.02181721,
      "epoch": 0.966541409890275,
      "flos": 14282320798080.0,
      "grad_norm": 1.8581525250032593,
      "language_loss": 0.78831637,
      "learning_rate": 1.171102125547696e-08,
      "loss": 0.80939746,
      "num_input_tokens_seen": 346881955,
      "step": 16076,
      "time_per_iteration": 2.5619616508483887
    },
    {
      "auxiliary_loss_clip": 0.01083739,
      "auxiliary_loss_mlp": 0.01042056,
      "balance_loss_clip": 1.03568125,
      "balance_loss_mlp": 1.02878809,
      "epoch": 0.9666015331429431,
      "flos": 19860432405120.0,
      "grad_norm": 1.6466558394104436,
      "language_loss": 0.72457671,
      "learning_rate": 1.166897413780532e-08,
      "loss": 0.74583471,
      "num_input_tokens_seen": 346900445,
      "step": 16077,
      "time_per_iteration": 2.494022846221924
    },
    {
      "auxiliary_loss_clip": 0.01088996,
      "auxiliary_loss_mlp": 0.01032154,
      "balance_loss_clip": 1.03310359,
      "balance_loss_mlp": 1.01968443,
      "epoch": 0.966661656395611,
      "flos": 27125951178240.0,
      "grad_norm": 1.7502676683251068,
      "language_loss": 0.59188616,
      "learning_rate": 1.1627002417845533e-08,
      "loss": 0.61309761,
      "num_input_tokens_seen": 346920135,
      "step": 16078,
      "time_per_iteration": 3.925840377807617
    },
    {
      "auxiliary_loss_clip": 0.01095923,
      "auxiliary_loss_mlp": 0.01031367,
      "balance_loss_clip": 1.03518128,
      "balance_loss_mlp": 1.01879001,
      "epoch": 0.966721779648279,
      "flos": 21508229848320.0,
      "grad_norm": 2.227878398678544,
      "language_loss": 0.72026813,
      "learning_rate": 1.158510609718899e-08,
      "loss": 0.74154109,
      "num_input_tokens_seen": 346940450,
      "step": 16079,
      "time_per_iteration": 2.5090653896331787
    },
    {
      "auxiliary_loss_clip": 0.01089063,
      "auxiliary_loss_mlp": 0.01028141,
      "balance_loss_clip": 1.03457618,
      "balance_loss_mlp": 1.01709008,
      "epoch": 0.9667819029009469,
      "flos": 23878118401920.0,
      "grad_norm": 1.5432721370010796,
      "language_loss": 0.72123086,
      "learning_rate": 1.1543285177424644e-08,
      "loss": 0.74240291,
      "num_input_tokens_seen": 346960935,
      "step": 16080,
      "time_per_iteration": 2.5058460235595703
    },
    {
      "auxiliary_loss_clip": 0.01072015,
      "auxiliary_loss_mlp": 0.0103413,
      "balance_loss_clip": 1.03178668,
      "balance_loss_mlp": 1.02113008,
      "epoch": 0.9668420261536149,
      "flos": 21507224267520.0,
      "grad_norm": 1.8930093890385007,
      "language_loss": 0.73751402,
      "learning_rate": 1.1501539660138115e-08,
      "loss": 0.7585755,
      "num_input_tokens_seen": 346980100,
      "step": 16081,
      "time_per_iteration": 2.528735399246216
    },
    {
      "auxiliary_loss_clip": 0.01075661,
      "auxiliary_loss_mlp": 0.01027457,
      "balance_loss_clip": 1.03154576,
      "balance_loss_mlp": 1.01526785,
      "epoch": 0.9669021494062828,
      "flos": 26687266375680.0,
      "grad_norm": 1.5525961324106754,
      "language_loss": 0.67288685,
      "learning_rate": 1.145986954691236e-08,
      "loss": 0.69391799,
      "num_input_tokens_seen": 347001250,
      "step": 16082,
      "time_per_iteration": 2.544196367263794
    },
    {
      "auxiliary_loss_clip": 0.01060806,
      "auxiliary_loss_mlp": 0.0103874,
      "balance_loss_clip": 1.03050041,
      "balance_loss_mlp": 1.02491164,
      "epoch": 0.9669622726589508,
      "flos": 29825032901760.0,
      "grad_norm": 1.4573655553865428,
      "language_loss": 0.76788127,
      "learning_rate": 1.141827483932789e-08,
      "loss": 0.78887677,
      "num_input_tokens_seen": 347022975,
      "step": 16083,
      "time_per_iteration": 2.622837543487549
    },
    {
      "auxiliary_loss_clip": 0.01056757,
      "auxiliary_loss_mlp": 0.01031968,
      "balance_loss_clip": 1.03482676,
      "balance_loss_mlp": 1.01989758,
      "epoch": 0.9670223959116189,
      "flos": 22922499018240.0,
      "grad_norm": 1.880244983347926,
      "language_loss": 0.79129589,
      "learning_rate": 1.1376755538961669e-08,
      "loss": 0.81218314,
      "num_input_tokens_seen": 347038780,
      "step": 16084,
      "time_per_iteration": 2.597515344619751
    },
    {
      "auxiliary_loss_clip": 0.01095364,
      "auxiliary_loss_mlp": 0.01027757,
      "balance_loss_clip": 1.03321862,
      "balance_loss_mlp": 1.01534128,
      "epoch": 0.9670825191642868,
      "flos": 18624495283200.0,
      "grad_norm": 2.1670951539316823,
      "language_loss": 0.6774087,
      "learning_rate": 1.1335311647387991e-08,
      "loss": 0.69863987,
      "num_input_tokens_seen": 347056705,
      "step": 16085,
      "time_per_iteration": 2.491541624069214
    },
    {
      "auxiliary_loss_clip": 0.01084766,
      "auxiliary_loss_mlp": 0.01029576,
      "balance_loss_clip": 1.03584075,
      "balance_loss_mlp": 1.01648712,
      "epoch": 0.9671426424169548,
      "flos": 24497936513280.0,
      "grad_norm": 2.044192717761249,
      "language_loss": 0.68191355,
      "learning_rate": 1.1293943166178709e-08,
      "loss": 0.70305699,
      "num_input_tokens_seen": 347075710,
      "step": 16086,
      "time_per_iteration": 2.5439982414245605
    },
    {
      "auxiliary_loss_clip": 0.01091243,
      "auxiliary_loss_mlp": 0.01033445,
      "balance_loss_clip": 1.03396475,
      "balance_loss_mlp": 1.02104139,
      "epoch": 0.9672027656696227,
      "flos": 20371189847040.0,
      "grad_norm": 1.8083685554362006,
      "language_loss": 0.78777748,
      "learning_rate": 1.125265009690235e-08,
      "loss": 0.80902433,
      "num_input_tokens_seen": 347092325,
      "step": 16087,
      "time_per_iteration": 2.477534294128418
    },
    {
      "auxiliary_loss_clip": 0.01075284,
      "auxiliary_loss_mlp": 0.01025387,
      "balance_loss_clip": 1.03210139,
      "balance_loss_mlp": 1.01347244,
      "epoch": 0.9672628889222907,
      "flos": 18880179269760.0,
      "grad_norm": 1.9096377103119597,
      "language_loss": 0.71508497,
      "learning_rate": 1.1211432441124769e-08,
      "loss": 0.73609167,
      "num_input_tokens_seen": 347110595,
      "step": 16088,
      "time_per_iteration": 2.502833604812622
    },
    {
      "auxiliary_loss_clip": 0.01100099,
      "auxiliary_loss_mlp": 0.00781938,
      "balance_loss_clip": 1.03420639,
      "balance_loss_mlp": 1.00877643,
      "epoch": 0.9673230121749586,
      "flos": 28695247447680.0,
      "grad_norm": 1.4028687045816384,
      "language_loss": 0.7037077,
      "learning_rate": 1.117029020040916e-08,
      "loss": 0.7225281,
      "num_input_tokens_seen": 347131625,
      "step": 16089,
      "time_per_iteration": 2.537194013595581
    },
    {
      "auxiliary_loss_clip": 0.01105729,
      "auxiliary_loss_mlp": 0.01029033,
      "balance_loss_clip": 1.03545034,
      "balance_loss_mlp": 1.01677799,
      "epoch": 0.9673831354276267,
      "flos": 20484452407680.0,
      "grad_norm": 2.5821732716342765,
      "language_loss": 0.74447393,
      "learning_rate": 1.1129223376315167e-08,
      "loss": 0.76582146,
      "num_input_tokens_seen": 347147910,
      "step": 16090,
      "time_per_iteration": 2.428001642227173
    },
    {
      "auxiliary_loss_clip": 0.01083776,
      "auxiliary_loss_mlp": 0.01028104,
      "balance_loss_clip": 1.03497374,
      "balance_loss_mlp": 1.01598668,
      "epoch": 0.9674432586802946,
      "flos": 26797548107520.0,
      "grad_norm": 1.6175532865756246,
      "language_loss": 0.68857104,
      "learning_rate": 1.1088231970400653e-08,
      "loss": 0.7096898,
      "num_input_tokens_seen": 347168805,
      "step": 16091,
      "time_per_iteration": 2.5714657306671143
    },
    {
      "auxiliary_loss_clip": 0.01102338,
      "auxiliary_loss_mlp": 0.01032833,
      "balance_loss_clip": 1.03486133,
      "balance_loss_mlp": 1.02004826,
      "epoch": 0.9675033819329626,
      "flos": 22310941034880.0,
      "grad_norm": 2.0330270074761576,
      "language_loss": 0.77050167,
      "learning_rate": 1.1047315984219484e-08,
      "loss": 0.79185343,
      "num_input_tokens_seen": 347189455,
      "step": 16092,
      "time_per_iteration": 2.4676384925842285
    },
    {
      "auxiliary_loss_clip": 0.01103695,
      "auxiliary_loss_mlp": 0.01026801,
      "balance_loss_clip": 1.03576231,
      "balance_loss_mlp": 1.01564908,
      "epoch": 0.9675635051856305,
      "flos": 12675713276160.0,
      "grad_norm": 1.9108113806024958,
      "language_loss": 0.76180547,
      "learning_rate": 1.1006475419323313e-08,
      "loss": 0.78311044,
      "num_input_tokens_seen": 347206030,
      "step": 16093,
      "time_per_iteration": 2.432854413986206
    },
    {
      "auxiliary_loss_clip": 0.01079201,
      "auxiliary_loss_mlp": 0.01025883,
      "balance_loss_clip": 1.03667724,
      "balance_loss_mlp": 1.01307952,
      "epoch": 0.9676236284382985,
      "flos": 24608469640320.0,
      "grad_norm": 1.4876912668177518,
      "language_loss": 0.69158554,
      "learning_rate": 1.096571027726112e-08,
      "loss": 0.71263635,
      "num_input_tokens_seen": 347226250,
      "step": 16094,
      "time_per_iteration": 2.5492799282073975
    },
    {
      "auxiliary_loss_clip": 0.01094161,
      "auxiliary_loss_mlp": 0.01029472,
      "balance_loss_clip": 1.03428984,
      "balance_loss_mlp": 1.01786137,
      "epoch": 0.9676837516909664,
      "flos": 23367145478400.0,
      "grad_norm": 1.487542063985486,
      "language_loss": 0.75994343,
      "learning_rate": 1.0925020559578557e-08,
      "loss": 0.78117979,
      "num_input_tokens_seen": 347247350,
      "step": 16095,
      "time_per_iteration": 2.5340163707733154
    },
    {
      "auxiliary_loss_clip": 0.01108181,
      "auxiliary_loss_mlp": 0.01033616,
      "balance_loss_clip": 1.03694165,
      "balance_loss_mlp": 1.02118826,
      "epoch": 0.9677438749436345,
      "flos": 20486894532480.0,
      "grad_norm": 1.7990288477045595,
      "language_loss": 0.70108485,
      "learning_rate": 1.0884406267818392e-08,
      "loss": 0.72250283,
      "num_input_tokens_seen": 347266870,
      "step": 16096,
      "time_per_iteration": 2.4454047679901123
    },
    {
      "auxiliary_loss_clip": 0.01083925,
      "auxiliary_loss_mlp": 0.01027313,
      "balance_loss_clip": 1.03501916,
      "balance_loss_mlp": 1.01537967,
      "epoch": 0.9678039981963025,
      "flos": 47555889719040.0,
      "grad_norm": 1.5876220659709945,
      "language_loss": 0.71811795,
      "learning_rate": 1.0843867403520946e-08,
      "loss": 0.73923028,
      "num_input_tokens_seen": 347290120,
      "step": 16097,
      "time_per_iteration": 2.7591240406036377
    },
    {
      "auxiliary_loss_clip": 0.01101838,
      "auxiliary_loss_mlp": 0.01033591,
      "balance_loss_clip": 1.03381968,
      "balance_loss_mlp": 1.02182543,
      "epoch": 0.9678641214489704,
      "flos": 25040474513280.0,
      "grad_norm": 1.5409112629551556,
      "language_loss": 0.78229433,
      "learning_rate": 1.0803403968223434e-08,
      "loss": 0.80364859,
      "num_input_tokens_seen": 347308785,
      "step": 16098,
      "time_per_iteration": 2.4671366214752197
    },
    {
      "auxiliary_loss_clip": 0.01075352,
      "auxiliary_loss_mlp": 0.01025917,
      "balance_loss_clip": 1.03499877,
      "balance_loss_mlp": 1.01458037,
      "epoch": 0.9679242447016384,
      "flos": 19240937516160.0,
      "grad_norm": 1.996999076325595,
      "language_loss": 0.90983355,
      "learning_rate": 1.0763015963459965e-08,
      "loss": 0.93084621,
      "num_input_tokens_seen": 347326375,
      "step": 16099,
      "time_per_iteration": 2.505222797393799
    },
    {
      "auxiliary_loss_clip": 0.01093641,
      "auxiliary_loss_mlp": 0.01028886,
      "balance_loss_clip": 1.03405046,
      "balance_loss_mlp": 1.01644695,
      "epoch": 0.9679843679543063,
      "flos": 33254681345280.0,
      "grad_norm": 2.3015652498104084,
      "language_loss": 0.66364199,
      "learning_rate": 1.0722703390762643e-08,
      "loss": 0.68486726,
      "num_input_tokens_seen": 347348250,
      "step": 16100,
      "time_per_iteration": 2.5802457332611084
    },
    {
      "auxiliary_loss_clip": 0.01063083,
      "auxiliary_loss_mlp": 0.01032357,
      "balance_loss_clip": 1.03618085,
      "balance_loss_mlp": 1.02024531,
      "epoch": 0.9680444912069743,
      "flos": 22783633038720.0,
      "grad_norm": 1.5306921931645212,
      "language_loss": 0.73376226,
      "learning_rate": 1.0682466251659584e-08,
      "loss": 0.75471663,
      "num_input_tokens_seen": 347367400,
      "step": 16101,
      "time_per_iteration": 2.571345329284668
    },
    {
      "auxiliary_loss_clip": 0.0108038,
      "auxiliary_loss_mlp": 0.01028802,
      "balance_loss_clip": 1.0343256,
      "balance_loss_mlp": 1.01648211,
      "epoch": 0.9681046144596422,
      "flos": 24024095274240.0,
      "grad_norm": 1.5046455827228338,
      "language_loss": 0.73306119,
      "learning_rate": 1.0642304547676672e-08,
      "loss": 0.75415301,
      "num_input_tokens_seen": 347387600,
      "step": 16102,
      "time_per_iteration": 2.550658702850342
    },
    {
      "auxiliary_loss_clip": 0.01075967,
      "auxiliary_loss_mlp": 0.0103234,
      "balance_loss_clip": 1.03679836,
      "balance_loss_mlp": 1.01972198,
      "epoch": 0.9681647377123103,
      "flos": 23441013797760.0,
      "grad_norm": 2.509475571476384,
      "language_loss": 0.77273786,
      "learning_rate": 1.0602218280337139e-08,
      "loss": 0.79382086,
      "num_input_tokens_seen": 347406915,
      "step": 16103,
      "time_per_iteration": 2.6016759872436523
    },
    {
      "auxiliary_loss_clip": 0.01082192,
      "auxiliary_loss_mlp": 0.01028391,
      "balance_loss_clip": 1.03306973,
      "balance_loss_mlp": 1.01656568,
      "epoch": 0.9682248609649782,
      "flos": 22675075159680.0,
      "grad_norm": 1.5647060830973825,
      "language_loss": 0.80302584,
      "learning_rate": 1.0562207451160655e-08,
      "loss": 0.82413173,
      "num_input_tokens_seen": 347425140,
      "step": 16104,
      "time_per_iteration": 2.5205013751983643
    },
    {
      "auxiliary_loss_clip": 0.01083005,
      "auxiliary_loss_mlp": 0.01033608,
      "balance_loss_clip": 1.03033233,
      "balance_loss_mlp": 1.02246785,
      "epoch": 0.9682849842176462,
      "flos": 24428413739520.0,
      "grad_norm": 1.4604262304179652,
      "language_loss": 0.78013909,
      "learning_rate": 1.0522272061664672e-08,
      "loss": 0.80130517,
      "num_input_tokens_seen": 347446350,
      "step": 16105,
      "time_per_iteration": 2.5648958683013916
    },
    {
      "auxiliary_loss_clip": 0.01005458,
      "auxiliary_loss_mlp": 0.01001602,
      "balance_loss_clip": 1.00414777,
      "balance_loss_mlp": 1.00055873,
      "epoch": 0.9683451074703141,
      "flos": 59995132784640.0,
      "grad_norm": 0.824730152804101,
      "language_loss": 0.56717724,
      "learning_rate": 1.0482412113363536e-08,
      "loss": 0.58724785,
      "num_input_tokens_seen": 347510135,
      "step": 16106,
      "time_per_iteration": 4.574390649795532
    },
    {
      "auxiliary_loss_clip": 0.01007769,
      "auxiliary_loss_mlp": 0.00999815,
      "balance_loss_clip": 1.01307607,
      "balance_loss_mlp": 0.99854529,
      "epoch": 0.9684052307229821,
      "flos": 52696145514240.0,
      "grad_norm": 0.8833024070502788,
      "language_loss": 0.61547947,
      "learning_rate": 1.0442627607768707e-08,
      "loss": 0.63555527,
      "num_input_tokens_seen": 347562505,
      "step": 16107,
      "time_per_iteration": 4.394185781478882
    },
    {
      "auxiliary_loss_clip": 0.01094405,
      "auxiliary_loss_mlp": 0.0103922,
      "balance_loss_clip": 1.0352813,
      "balance_loss_mlp": 1.02575493,
      "epoch": 0.96846535397565,
      "flos": 22783848520320.0,
      "grad_norm": 2.0634952984997854,
      "language_loss": 0.73733836,
      "learning_rate": 1.040291854638875e-08,
      "loss": 0.75867462,
      "num_input_tokens_seen": 347579150,
      "step": 16108,
      "time_per_iteration": 3.859767198562622
    },
    {
      "auxiliary_loss_clip": 0.01089296,
      "auxiliary_loss_mlp": 0.01027109,
      "balance_loss_clip": 1.03418529,
      "balance_loss_mlp": 1.01418734,
      "epoch": 0.968525477228318,
      "flos": 23323980309120.0,
      "grad_norm": 2.0617974433445037,
      "language_loss": 0.56677091,
      "learning_rate": 1.0363284930729576e-08,
      "loss": 0.58793497,
      "num_input_tokens_seen": 347596705,
      "step": 16109,
      "time_per_iteration": 2.4933395385742188
    },
    {
      "auxiliary_loss_clip": 0.01018755,
      "auxiliary_loss_mlp": 0.01001597,
      "balance_loss_clip": 1.00522017,
      "balance_loss_mlp": 1.00053632,
      "epoch": 0.9685856004809861,
      "flos": 67882947707520.0,
      "grad_norm": 0.6694668363503478,
      "language_loss": 0.54238904,
      "learning_rate": 1.0323726762294205e-08,
      "loss": 0.56259257,
      "num_input_tokens_seen": 347661870,
      "step": 16110,
      "time_per_iteration": 3.089578866958618
    },
    {
      "auxiliary_loss_clip": 0.01037182,
      "auxiliary_loss_mlp": 0.01037799,
      "balance_loss_clip": 1.03404677,
      "balance_loss_mlp": 1.02389908,
      "epoch": 0.968645723733654,
      "flos": 33947900899200.0,
      "grad_norm": 1.4068991479946813,
      "language_loss": 0.62191427,
      "learning_rate": 1.0284244042582325e-08,
      "loss": 0.64266413,
      "num_input_tokens_seen": 347684295,
      "step": 16111,
      "time_per_iteration": 2.784716844558716
    },
    {
      "auxiliary_loss_clip": 0.01078636,
      "auxiliary_loss_mlp": 0.01026734,
      "balance_loss_clip": 1.03250206,
      "balance_loss_mlp": 1.01640463,
      "epoch": 0.968705846986322,
      "flos": 18551488890240.0,
      "grad_norm": 4.574918490940266,
      "language_loss": 0.74939787,
      "learning_rate": 1.024483677309118e-08,
      "loss": 0.77045166,
      "num_input_tokens_seen": 347702585,
      "step": 16112,
      "time_per_iteration": 2.5633316040039062
    },
    {
      "auxiliary_loss_clip": 0.01091008,
      "auxiliary_loss_mlp": 0.01026575,
      "balance_loss_clip": 1.03383517,
      "balance_loss_mlp": 1.01572144,
      "epoch": 0.9687659702389899,
      "flos": 17420913336960.0,
      "grad_norm": 1.8297485952623755,
      "language_loss": 0.66721696,
      "learning_rate": 1.020550495531558e-08,
      "loss": 0.68839276,
      "num_input_tokens_seen": 347721810,
      "step": 16113,
      "time_per_iteration": 2.5308640003204346
    },
    {
      "auxiliary_loss_clip": 0.01016971,
      "auxiliary_loss_mlp": 0.0100547,
      "balance_loss_clip": 1.00598454,
      "balance_loss_mlp": 1.00447488,
      "epoch": 0.9688260934916579,
      "flos": 62047176865920.0,
      "grad_norm": 0.6910904593141285,
      "language_loss": 0.5657571,
      "learning_rate": 1.0166248590746329e-08,
      "loss": 0.58598149,
      "num_input_tokens_seen": 347782330,
      "step": 16114,
      "time_per_iteration": 3.094142198562622
    },
    {
      "auxiliary_loss_clip": 0.0107755,
      "auxiliary_loss_mlp": 0.01033941,
      "balance_loss_clip": 1.03446698,
      "balance_loss_mlp": 1.02178729,
      "epoch": 0.9688862167443258,
      "flos": 15076520461440.0,
      "grad_norm": 1.9512016837213002,
      "language_loss": 0.8276183,
      "learning_rate": 1.0127067680872458e-08,
      "loss": 0.84873319,
      "num_input_tokens_seen": 347794835,
      "step": 16115,
      "time_per_iteration": 2.4400620460510254
    },
    {
      "auxiliary_loss_clip": 0.01087228,
      "auxiliary_loss_mlp": 0.0102803,
      "balance_loss_clip": 1.03350282,
      "balance_loss_mlp": 1.01685989,
      "epoch": 0.9689463399969939,
      "flos": 19938215306880.0,
      "grad_norm": 1.6196973853305432,
      "language_loss": 0.72330427,
      "learning_rate": 1.0087962227179448e-08,
      "loss": 0.74445677,
      "num_input_tokens_seen": 347814320,
      "step": 16116,
      "time_per_iteration": 3.884894609451294
    },
    {
      "auxiliary_loss_clip": 0.01063384,
      "auxiliary_loss_mlp": 0.0103389,
      "balance_loss_clip": 1.03339529,
      "balance_loss_mlp": 1.02116454,
      "epoch": 0.9690064632496618,
      "flos": 19573039687680.0,
      "grad_norm": 2.3882747809388514,
      "language_loss": 0.75519651,
      "learning_rate": 1.0048932231150553e-08,
      "loss": 0.7761693,
      "num_input_tokens_seen": 347832125,
      "step": 16117,
      "time_per_iteration": 2.523899555206299
    },
    {
      "auxiliary_loss_clip": 0.01103992,
      "auxiliary_loss_mlp": 0.01029363,
      "balance_loss_clip": 1.03355479,
      "balance_loss_mlp": 1.01707888,
      "epoch": 0.9690665865023298,
      "flos": 21872292145920.0,
      "grad_norm": 2.0496819304967127,
      "language_loss": 0.77539992,
      "learning_rate": 1.000997769426548e-08,
      "loss": 0.7967335,
      "num_input_tokens_seen": 347850765,
      "step": 16118,
      "time_per_iteration": 2.4586479663848877
    },
    {
      "auxiliary_loss_clip": 0.01082457,
      "auxiliary_loss_mlp": 0.00785377,
      "balance_loss_clip": 1.03543615,
      "balance_loss_mlp": 1.01159,
      "epoch": 0.9691267097549977,
      "flos": 20994491577600.0,
      "grad_norm": 1.766593161299943,
      "language_loss": 0.78108001,
      "learning_rate": 9.971098618001272e-09,
      "loss": 0.79975832,
      "num_input_tokens_seen": 347870125,
      "step": 16119,
      "time_per_iteration": 2.5324463844299316
    },
    {
      "auxiliary_loss_clip": 0.01053338,
      "auxiliary_loss_mlp": 0.01034686,
      "balance_loss_clip": 1.03143072,
      "balance_loss_mlp": 1.02253282,
      "epoch": 0.9691868330076657,
      "flos": 24279132816000.0,
      "grad_norm": 1.3664166631291177,
      "language_loss": 0.7555747,
      "learning_rate": 9.932295003832747e-09,
      "loss": 0.77645493,
      "num_input_tokens_seen": 347890615,
      "step": 16120,
      "time_per_iteration": 2.6001393795013428
    },
    {
      "auxiliary_loss_clip": 0.01091952,
      "auxiliary_loss_mlp": 0.01027548,
      "balance_loss_clip": 1.03387344,
      "balance_loss_mlp": 1.01603293,
      "epoch": 0.9692469562603336,
      "flos": 17675699483520.0,
      "grad_norm": 1.8289097616934527,
      "language_loss": 0.69644076,
      "learning_rate": 9.89356685323095e-09,
      "loss": 0.71763575,
      "num_input_tokens_seen": 347908685,
      "step": 16121,
      "time_per_iteration": 2.487438440322876
    },
    {
      "auxiliary_loss_clip": 0.01091098,
      "auxiliary_loss_mlp": 0.01028379,
      "balance_loss_clip": 1.033849,
      "balance_loss_mlp": 1.01672029,
      "epoch": 0.9693070795130017,
      "flos": 26834392483200.0,
      "grad_norm": 1.7606472749833983,
      "language_loss": 0.69406259,
      "learning_rate": 9.854914167664486e-09,
      "loss": 0.71525741,
      "num_input_tokens_seen": 347926385,
      "step": 16122,
      "time_per_iteration": 2.535815715789795
    },
    {
      "auxiliary_loss_clip": 0.01064075,
      "auxiliary_loss_mlp": 0.01031973,
      "balance_loss_clip": 1.03196049,
      "balance_loss_mlp": 1.01973009,
      "epoch": 0.9693672027656697,
      "flos": 18077288515200.0,
      "grad_norm": 1.8735126831177165,
      "language_loss": 0.75870919,
      "learning_rate": 9.81633694859907e-09,
      "loss": 0.77966964,
      "num_input_tokens_seen": 347945290,
      "step": 16123,
      "time_per_iteration": 2.5234906673431396
    },
    {
      "auxiliary_loss_clip": 0.01067494,
      "auxiliary_loss_mlp": 0.01036239,
      "balance_loss_clip": 1.0328424,
      "balance_loss_mlp": 1.0224582,
      "epoch": 0.9694273260183376,
      "flos": 21763015994880.0,
      "grad_norm": 1.637716870250452,
      "language_loss": 0.74511194,
      "learning_rate": 9.777835197497753e-09,
      "loss": 0.76614922,
      "num_input_tokens_seen": 347966330,
      "step": 16124,
      "time_per_iteration": 2.583019971847534
    },
    {
      "auxiliary_loss_clip": 0.01092359,
      "auxiliary_loss_mlp": 0.01033793,
      "balance_loss_clip": 1.03405023,
      "balance_loss_mlp": 1.02225399,
      "epoch": 0.9694874492710056,
      "flos": 24426115269120.0,
      "grad_norm": 1.9753030356668775,
      "language_loss": 0.74582863,
      "learning_rate": 9.739408915820258e-09,
      "loss": 0.7670902,
      "num_input_tokens_seen": 347982590,
      "step": 16125,
      "time_per_iteration": 2.5477070808410645
    },
    {
      "auxiliary_loss_clip": 0.01017399,
      "auxiliary_loss_mlp": 0.01001188,
      "balance_loss_clip": 1.00504446,
      "balance_loss_mlp": 1.00016844,
      "epoch": 0.9695475725236735,
      "flos": 67650748237440.0,
      "grad_norm": 0.8634075605124992,
      "language_loss": 0.61491394,
      "learning_rate": 9.70105810502364e-09,
      "loss": 0.63509983,
      "num_input_tokens_seen": 348043310,
      "step": 16126,
      "time_per_iteration": 3.0799570083618164
    },
    {
      "auxiliary_loss_clip": 0.01091431,
      "auxiliary_loss_mlp": 0.01034586,
      "balance_loss_clip": 1.03576374,
      "balance_loss_mlp": 1.02287936,
      "epoch": 0.9696076957763415,
      "flos": 19129326981120.0,
      "grad_norm": 1.6327366409593627,
      "language_loss": 0.74959648,
      "learning_rate": 9.662782766562738e-09,
      "loss": 0.77085668,
      "num_input_tokens_seen": 348062200,
      "step": 16127,
      "time_per_iteration": 2.561328411102295
    },
    {
      "auxiliary_loss_clip": 0.01058727,
      "auxiliary_loss_mlp": 0.01030366,
      "balance_loss_clip": 1.03170729,
      "balance_loss_mlp": 1.01827812,
      "epoch": 0.9696678190290094,
      "flos": 15486836497920.0,
      "grad_norm": 1.906666294349261,
      "language_loss": 0.69106275,
      "learning_rate": 9.62458290188839e-09,
      "loss": 0.71195364,
      "num_input_tokens_seen": 348080685,
      "step": 16128,
      "time_per_iteration": 2.599687099456787
    },
    {
      "auxiliary_loss_clip": 0.01071269,
      "auxiliary_loss_mlp": 0.01033741,
      "balance_loss_clip": 1.03523993,
      "balance_loss_mlp": 1.02199852,
      "epoch": 0.9697279422816775,
      "flos": 36208692869760.0,
      "grad_norm": 1.5504658814744137,
      "language_loss": 0.65615124,
      "learning_rate": 9.586458512449213e-09,
      "loss": 0.67720133,
      "num_input_tokens_seen": 348102500,
      "step": 16129,
      "time_per_iteration": 2.717768669128418
    },
    {
      "auxiliary_loss_clip": 0.01071426,
      "auxiliary_loss_mlp": 0.01029191,
      "balance_loss_clip": 1.03559411,
      "balance_loss_mlp": 1.01699615,
      "epoch": 0.9697880655343454,
      "flos": 25484007651840.0,
      "grad_norm": 1.9274633359266975,
      "language_loss": 0.63070703,
      "learning_rate": 9.548409599691166e-09,
      "loss": 0.65171319,
      "num_input_tokens_seen": 348122515,
      "step": 16130,
      "time_per_iteration": 2.609807252883911
    },
    {
      "auxiliary_loss_clip": 0.0109453,
      "auxiliary_loss_mlp": 0.01027729,
      "balance_loss_clip": 1.03406513,
      "balance_loss_mlp": 1.01567721,
      "epoch": 0.9698481887870134,
      "flos": 15333533251200.0,
      "grad_norm": 2.2865511489193255,
      "language_loss": 0.69976842,
      "learning_rate": 9.510436165056867e-09,
      "loss": 0.72099102,
      "num_input_tokens_seen": 348138775,
      "step": 16131,
      "time_per_iteration": 2.4826691150665283
    },
    {
      "auxiliary_loss_clip": 0.0110508,
      "auxiliary_loss_mlp": 0.00783307,
      "balance_loss_clip": 1.03488541,
      "balance_loss_mlp": 1.01033306,
      "epoch": 0.9699083120396813,
      "flos": 21982250655360.0,
      "grad_norm": 2.165814589916953,
      "language_loss": 0.76258737,
      "learning_rate": 9.472538209986058e-09,
      "loss": 0.78147125,
      "num_input_tokens_seen": 348157115,
      "step": 16132,
      "time_per_iteration": 2.5020503997802734
    },
    {
      "auxiliary_loss_clip": 0.01072534,
      "auxiliary_loss_mlp": 0.0103417,
      "balance_loss_clip": 1.0370028,
      "balance_loss_mlp": 1.02173638,
      "epoch": 0.9699684352923493,
      "flos": 15664055224320.0,
      "grad_norm": 2.27552743023492,
      "language_loss": 0.78936398,
      "learning_rate": 9.434715735916477e-09,
      "loss": 0.81043106,
      "num_input_tokens_seen": 348173035,
      "step": 16133,
      "time_per_iteration": 2.583000659942627
    },
    {
      "auxiliary_loss_clip": 0.01072179,
      "auxiliary_loss_mlp": 0.01026963,
      "balance_loss_clip": 1.03360105,
      "balance_loss_mlp": 1.01619279,
      "epoch": 0.9700285585450172,
      "flos": 21908382336000.0,
      "grad_norm": 1.6088870463495266,
      "language_loss": 0.64771187,
      "learning_rate": 9.396968744281863e-09,
      "loss": 0.66870332,
      "num_input_tokens_seen": 348192960,
      "step": 16134,
      "time_per_iteration": 2.524009943008423
    },
    {
      "auxiliary_loss_clip": 0.01079317,
      "auxiliary_loss_mlp": 0.01029141,
      "balance_loss_clip": 1.03209484,
      "balance_loss_mlp": 1.01655269,
      "epoch": 0.9700886817976853,
      "flos": 23914890950400.0,
      "grad_norm": 2.3547408155569247,
      "language_loss": 0.8086096,
      "learning_rate": 9.359297236513519e-09,
      "loss": 0.82969415,
      "num_input_tokens_seen": 348212805,
      "step": 16135,
      "time_per_iteration": 2.5438811779022217
    },
    {
      "auxiliary_loss_clip": 0.01094379,
      "auxiliary_loss_mlp": 0.01029721,
      "balance_loss_clip": 1.03425205,
      "balance_loss_mlp": 1.01710916,
      "epoch": 0.9701488050503532,
      "flos": 25447845634560.0,
      "grad_norm": 6.1055864995245415,
      "language_loss": 0.73194706,
      "learning_rate": 9.321701214040079e-09,
      "loss": 0.75318801,
      "num_input_tokens_seen": 348232900,
      "step": 16136,
      "time_per_iteration": 2.515380859375
    },
    {
      "auxiliary_loss_clip": 0.01102333,
      "auxiliary_loss_mlp": 0.01031652,
      "balance_loss_clip": 1.03522062,
      "balance_loss_mlp": 1.02063072,
      "epoch": 0.9702089283030212,
      "flos": 20590855470720.0,
      "grad_norm": 1.4442914528739381,
      "language_loss": 0.76036823,
      "learning_rate": 9.28418067828729e-09,
      "loss": 0.78170806,
      "num_input_tokens_seen": 348253065,
      "step": 16137,
      "time_per_iteration": 2.486067533493042
    },
    {
      "auxiliary_loss_clip": 0.00996909,
      "auxiliary_loss_mlp": 0.01001033,
      "balance_loss_clip": 1.01989293,
      "balance_loss_mlp": 0.99997199,
      "epoch": 0.9702690515556892,
      "flos": 70651516291200.0,
      "grad_norm": 0.7674504019640235,
      "language_loss": 0.54930681,
      "learning_rate": 9.246735630678015e-09,
      "loss": 0.56928623,
      "num_input_tokens_seen": 348316075,
      "step": 16138,
      "time_per_iteration": 3.323807954788208
    },
    {
      "auxiliary_loss_clip": 0.01082512,
      "auxiliary_loss_mlp": 0.01028423,
      "balance_loss_clip": 1.03399634,
      "balance_loss_mlp": 1.01656795,
      "epoch": 0.9703291748083571,
      "flos": 35881439034240.0,
      "grad_norm": 3.2336141196843453,
      "language_loss": 0.70778835,
      "learning_rate": 9.209366072632007e-09,
      "loss": 0.72889769,
      "num_input_tokens_seen": 348337605,
      "step": 16139,
      "time_per_iteration": 2.7408053874969482
    },
    {
      "auxiliary_loss_clip": 0.01095776,
      "auxiliary_loss_mlp": 0.01027594,
      "balance_loss_clip": 1.03701401,
      "balance_loss_mlp": 1.01530981,
      "epoch": 0.9703892980610251,
      "flos": 24316479982080.0,
      "grad_norm": 1.4309894534234784,
      "language_loss": 0.72393018,
      "learning_rate": 9.172072005566134e-09,
      "loss": 0.74516392,
      "num_input_tokens_seen": 348359430,
      "step": 16140,
      "time_per_iteration": 2.5062978267669678
    },
    {
      "auxiliary_loss_clip": 0.01098051,
      "auxiliary_loss_mlp": 0.00785711,
      "balance_loss_clip": 1.03694677,
      "balance_loss_mlp": 1.01065135,
      "epoch": 0.970449421313693,
      "flos": 18003743418240.0,
      "grad_norm": 2.252397564136839,
      "language_loss": 0.68622077,
      "learning_rate": 9.13485343089504e-09,
      "loss": 0.70505834,
      "num_input_tokens_seen": 348377890,
      "step": 16141,
      "time_per_iteration": 2.4686646461486816
    },
    {
      "auxiliary_loss_clip": 0.01087463,
      "auxiliary_loss_mlp": 0.01029775,
      "balance_loss_clip": 1.03241086,
      "balance_loss_mlp": 1.01806879,
      "epoch": 0.9705095445663611,
      "flos": 25337994865920.0,
      "grad_norm": 2.025160035014419,
      "language_loss": 0.68337595,
      "learning_rate": 9.097710350029597e-09,
      "loss": 0.70454836,
      "num_input_tokens_seen": 348396550,
      "step": 16142,
      "time_per_iteration": 2.497830629348755
    },
    {
      "auxiliary_loss_clip": 0.010476,
      "auxiliary_loss_mlp": 0.01028175,
      "balance_loss_clip": 1.03198922,
      "balance_loss_mlp": 1.01603961,
      "epoch": 0.970569667819029,
      "flos": 26833602384000.0,
      "grad_norm": 1.854384377186835,
      "language_loss": 0.5587483,
      "learning_rate": 9.060642764378457e-09,
      "loss": 0.57950604,
      "num_input_tokens_seen": 348417120,
      "step": 16143,
      "time_per_iteration": 2.632920742034912
    },
    {
      "auxiliary_loss_clip": 0.01093628,
      "auxiliary_loss_mlp": 0.01032238,
      "balance_loss_clip": 1.03554368,
      "balance_loss_mlp": 1.02079964,
      "epoch": 0.970629791071697,
      "flos": 25848644567040.0,
      "grad_norm": 2.3077839461810212,
      "language_loss": 0.6776967,
      "learning_rate": 9.023650675347382e-09,
      "loss": 0.69895542,
      "num_input_tokens_seen": 348437750,
      "step": 16144,
      "time_per_iteration": 2.569767475128174
    },
    {
      "auxiliary_loss_clip": 0.01093099,
      "auxiliary_loss_mlp": 0.01037863,
      "balance_loss_clip": 1.03606343,
      "balance_loss_mlp": 1.02650857,
      "epoch": 0.9706899143243649,
      "flos": 36540184510080.0,
      "grad_norm": 1.6796592398635923,
      "language_loss": 0.72247493,
      "learning_rate": 8.986734084339253e-09,
      "loss": 0.74378455,
      "num_input_tokens_seen": 348460935,
      "step": 16145,
      "time_per_iteration": 4.1486687660217285
    },
    {
      "auxiliary_loss_clip": 0.01080958,
      "auxiliary_loss_mlp": 0.01027032,
      "balance_loss_clip": 1.03296626,
      "balance_loss_mlp": 1.0146637,
      "epoch": 0.9707500375770329,
      "flos": 12268234414080.0,
      "grad_norm": 3.0932082202534392,
      "language_loss": 0.81105614,
      "learning_rate": 8.949892992753395e-09,
      "loss": 0.83213603,
      "num_input_tokens_seen": 348474480,
      "step": 16146,
      "time_per_iteration": 3.908280611038208
    },
    {
      "auxiliary_loss_clip": 0.00998243,
      "auxiliary_loss_mlp": 0.01000253,
      "balance_loss_clip": 1.00960422,
      "balance_loss_mlp": 0.99916804,
      "epoch": 0.9708101608297008,
      "flos": 60853040196480.0,
      "grad_norm": 0.8238243437000202,
      "language_loss": 0.54565907,
      "learning_rate": 8.91312740198713e-09,
      "loss": 0.56564403,
      "num_input_tokens_seen": 348541220,
      "step": 16147,
      "time_per_iteration": 4.629950284957886
    },
    {
      "auxiliary_loss_clip": 0.01069304,
      "auxiliary_loss_mlp": 0.00785814,
      "balance_loss_clip": 1.03206158,
      "balance_loss_mlp": 1.01149821,
      "epoch": 0.9708702840823689,
      "flos": 27124766029440.0,
      "grad_norm": 3.1925548701959103,
      "language_loss": 0.6117605,
      "learning_rate": 8.876437313434682e-09,
      "loss": 0.63031173,
      "num_input_tokens_seen": 348559230,
      "step": 16148,
      "time_per_iteration": 2.6058900356292725
    },
    {
      "auxiliary_loss_clip": 0.01066153,
      "auxiliary_loss_mlp": 0.01032963,
      "balance_loss_clip": 1.03383422,
      "balance_loss_mlp": 1.02163851,
      "epoch": 0.9709304073350368,
      "flos": 20777699041920.0,
      "grad_norm": 1.7301271068643318,
      "language_loss": 0.73695552,
      "learning_rate": 8.839822728487155e-09,
      "loss": 0.75794667,
      "num_input_tokens_seen": 348577850,
      "step": 16149,
      "time_per_iteration": 2.558091640472412
    },
    {
      "auxiliary_loss_clip": 0.0109062,
      "auxiliary_loss_mlp": 0.01037897,
      "balance_loss_clip": 1.03195071,
      "balance_loss_mlp": 1.02577901,
      "epoch": 0.9709905305877048,
      "flos": 41934541115520.0,
      "grad_norm": 2.9565339729309925,
      "language_loss": 0.75575656,
      "learning_rate": 8.803283648533222e-09,
      "loss": 0.77704167,
      "num_input_tokens_seen": 348598345,
      "step": 16150,
      "time_per_iteration": 2.6718592643737793
    },
    {
      "auxiliary_loss_clip": 0.0108901,
      "auxiliary_loss_mlp": 0.01029242,
      "balance_loss_clip": 1.03691792,
      "balance_loss_mlp": 1.01447189,
      "epoch": 0.9710506538403728,
      "flos": 17165588486400.0,
      "grad_norm": 1.985481969584915,
      "language_loss": 0.73826706,
      "learning_rate": 8.766820074958214e-09,
      "loss": 0.7594496,
      "num_input_tokens_seen": 348616300,
      "step": 16151,
      "time_per_iteration": 2.507359743118286
    },
    {
      "auxiliary_loss_clip": 0.01089609,
      "auxiliary_loss_mlp": 0.01027115,
      "balance_loss_clip": 1.03409958,
      "balance_loss_mlp": 1.01553392,
      "epoch": 0.9711107770930407,
      "flos": 21173470070400.0,
      "grad_norm": 1.739490842395945,
      "language_loss": 0.74895394,
      "learning_rate": 8.730432009145027e-09,
      "loss": 0.77012122,
      "num_input_tokens_seen": 348633845,
      "step": 16152,
      "time_per_iteration": 2.516207695007324
    },
    {
      "auxiliary_loss_clip": 0.01068235,
      "auxiliary_loss_mlp": 0.01031547,
      "balance_loss_clip": 1.03470433,
      "balance_loss_mlp": 1.0196197,
      "epoch": 0.9711709003457087,
      "flos": 22237072715520.0,
      "grad_norm": 2.0643528615496365,
      "language_loss": 0.67212987,
      "learning_rate": 8.694119452473448e-09,
      "loss": 0.69312775,
      "num_input_tokens_seen": 348653070,
      "step": 16153,
      "time_per_iteration": 2.5508322715759277
    },
    {
      "auxiliary_loss_clip": 0.01051393,
      "auxiliary_loss_mlp": 0.01029278,
      "balance_loss_clip": 1.03343964,
      "balance_loss_mlp": 1.01809585,
      "epoch": 0.9712310235983767,
      "flos": 26213856099840.0,
      "grad_norm": 1.5455554401605685,
      "language_loss": 0.70772052,
      "learning_rate": 8.65788240632037e-09,
      "loss": 0.72852719,
      "num_input_tokens_seen": 348672145,
      "step": 16154,
      "time_per_iteration": 2.649338483810425
    },
    {
      "auxiliary_loss_clip": 0.01056144,
      "auxiliary_loss_mlp": 0.01030788,
      "balance_loss_clip": 1.03871322,
      "balance_loss_mlp": 1.017699,
      "epoch": 0.9712911468510447,
      "flos": 20668171495680.0,
      "grad_norm": 1.8007078676722663,
      "language_loss": 0.80792856,
      "learning_rate": 8.621720872059812e-09,
      "loss": 0.82879788,
      "num_input_tokens_seen": 348690615,
      "step": 16155,
      "time_per_iteration": 3.990067958831787
    },
    {
      "auxiliary_loss_clip": 0.01092466,
      "auxiliary_loss_mlp": 0.00783177,
      "balance_loss_clip": 1.03527379,
      "balance_loss_mlp": 1.00794983,
      "epoch": 0.9713512701037126,
      "flos": 13552903313280.0,
      "grad_norm": 2.4649858388807173,
      "language_loss": 0.67459249,
      "learning_rate": 8.58563485106334e-09,
      "loss": 0.69334888,
      "num_input_tokens_seen": 348708665,
      "step": 16156,
      "time_per_iteration": 2.4739787578582764
    },
    {
      "auxiliary_loss_clip": 0.0109243,
      "auxiliary_loss_mlp": 0.01034707,
      "balance_loss_clip": 1.0327419,
      "balance_loss_mlp": 1.02294087,
      "epoch": 0.9714113933563806,
      "flos": 25848752307840.0,
      "grad_norm": 2.342929152849355,
      "language_loss": 0.90993047,
      "learning_rate": 8.54962434469919e-09,
      "loss": 0.93120182,
      "num_input_tokens_seen": 348726105,
      "step": 16157,
      "time_per_iteration": 2.4954421520233154
    },
    {
      "auxiliary_loss_clip": 0.01073456,
      "auxiliary_loss_mlp": 0.00785132,
      "balance_loss_clip": 1.03515601,
      "balance_loss_mlp": 1.01403952,
      "epoch": 0.9714715166090485,
      "flos": 12743081233920.0,
      "grad_norm": 1.7872677384797977,
      "language_loss": 0.72713435,
      "learning_rate": 8.513689354332721e-09,
      "loss": 0.74572027,
      "num_input_tokens_seen": 348743360,
      "step": 16158,
      "time_per_iteration": 2.5450830459594727
    },
    {
      "auxiliary_loss_clip": 0.01054379,
      "auxiliary_loss_mlp": 0.01036805,
      "balance_loss_clip": 1.03238833,
      "balance_loss_mlp": 1.02451408,
      "epoch": 0.9715316398617165,
      "flos": 18405547931520.0,
      "grad_norm": 2.145022321195413,
      "language_loss": 0.59979057,
      "learning_rate": 8.477829881326836e-09,
      "loss": 0.62070245,
      "num_input_tokens_seen": 348759045,
      "step": 16159,
      "time_per_iteration": 2.56588077545166
    },
    {
      "auxiliary_loss_clip": 0.0109818,
      "auxiliary_loss_mlp": 0.01027131,
      "balance_loss_clip": 1.03327727,
      "balance_loss_mlp": 1.01630092,
      "epoch": 0.9715917631143844,
      "flos": 28913799749760.0,
      "grad_norm": 1.718273813012951,
      "language_loss": 0.79058826,
      "learning_rate": 8.44204592704112e-09,
      "loss": 0.81184143,
      "num_input_tokens_seen": 348779910,
      "step": 16160,
      "time_per_iteration": 2.5141422748565674
    },
    {
      "auxiliary_loss_clip": 0.01027021,
      "auxiliary_loss_mlp": 0.00999477,
      "balance_loss_clip": 1.00426364,
      "balance_loss_mlp": 0.99845773,
      "epoch": 0.9716518863670525,
      "flos": 65939712900480.0,
      "grad_norm": 0.7762122080328414,
      "language_loss": 0.54294264,
      "learning_rate": 8.406337492832704e-09,
      "loss": 0.56320763,
      "num_input_tokens_seen": 348838995,
      "step": 16161,
      "time_per_iteration": 3.083940029144287
    },
    {
      "auxiliary_loss_clip": 0.01089143,
      "auxiliary_loss_mlp": 0.00782874,
      "balance_loss_clip": 1.03464746,
      "balance_loss_mlp": 1.00879729,
      "epoch": 0.9717120096197204,
      "flos": 17712759340800.0,
      "grad_norm": 1.7181306642673049,
      "language_loss": 0.71971166,
      "learning_rate": 8.3707045800554e-09,
      "loss": 0.73843187,
      "num_input_tokens_seen": 348858090,
      "step": 16162,
      "time_per_iteration": 2.4690141677856445
    },
    {
      "auxiliary_loss_clip": 0.01065878,
      "auxiliary_loss_mlp": 0.01031541,
      "balance_loss_clip": 1.03073668,
      "balance_loss_mlp": 1.01921463,
      "epoch": 0.9717721328723884,
      "flos": 24463426521600.0,
      "grad_norm": 1.5614321595512972,
      "language_loss": 0.78598356,
      "learning_rate": 8.335147190060787e-09,
      "loss": 0.80695772,
      "num_input_tokens_seen": 348877885,
      "step": 16163,
      "time_per_iteration": 2.606612205505371
    },
    {
      "auxiliary_loss_clip": 0.01077658,
      "auxiliary_loss_mlp": 0.01026157,
      "balance_loss_clip": 1.03425252,
      "balance_loss_mlp": 1.01460004,
      "epoch": 0.9718322561250564,
      "flos": 20776477979520.0,
      "grad_norm": 1.7451181591120961,
      "language_loss": 0.72885108,
      "learning_rate": 8.299665324196903e-09,
      "loss": 0.74988919,
      "num_input_tokens_seen": 348897720,
      "step": 16164,
      "time_per_iteration": 2.5045957565307617
    },
    {
      "auxiliary_loss_clip": 0.01042933,
      "auxiliary_loss_mlp": 0.01044308,
      "balance_loss_clip": 1.03191137,
      "balance_loss_mlp": 1.02894771,
      "epoch": 0.9718923793777243,
      "flos": 19025904746880.0,
      "grad_norm": 2.209397170391226,
      "language_loss": 0.83857238,
      "learning_rate": 8.264258983809114e-09,
      "loss": 0.85944474,
      "num_input_tokens_seen": 348915410,
      "step": 16165,
      "time_per_iteration": 2.591508626937866
    },
    {
      "auxiliary_loss_clip": 0.01067182,
      "auxiliary_loss_mlp": 0.01025273,
      "balance_loss_clip": 1.03343964,
      "balance_loss_mlp": 1.01477706,
      "epoch": 0.9719525026303923,
      "flos": 21871717528320.0,
      "grad_norm": 1.5215445805720342,
      "language_loss": 0.79244673,
      "learning_rate": 8.228928170240345e-09,
      "loss": 0.81337124,
      "num_input_tokens_seen": 348934335,
      "step": 16166,
      "time_per_iteration": 2.539766550064087
    },
    {
      "auxiliary_loss_clip": 0.0107687,
      "auxiliary_loss_mlp": 0.01026612,
      "balance_loss_clip": 1.03514838,
      "balance_loss_mlp": 1.01520967,
      "epoch": 0.9720126258830603,
      "flos": 14429303251200.0,
      "grad_norm": 1.891886675364507,
      "language_loss": 0.70374143,
      "learning_rate": 8.193672884830195e-09,
      "loss": 0.72477627,
      "num_input_tokens_seen": 348952405,
      "step": 16167,
      "time_per_iteration": 2.5113425254821777
    },
    {
      "auxiliary_loss_clip": 0.01075023,
      "auxiliary_loss_mlp": 0.01032561,
      "balance_loss_clip": 1.03729606,
      "balance_loss_mlp": 1.02099776,
      "epoch": 0.9720727491357283,
      "flos": 26251167352320.0,
      "grad_norm": 1.4883053494366814,
      "language_loss": 0.75161469,
      "learning_rate": 8.158493128915812e-09,
      "loss": 0.77269053,
      "num_input_tokens_seen": 348973580,
      "step": 16168,
      "time_per_iteration": 2.557081699371338
    },
    {
      "auxiliary_loss_clip": 0.01047989,
      "auxiliary_loss_mlp": 0.01047114,
      "balance_loss_clip": 1.03259993,
      "balance_loss_mlp": 1.03257692,
      "epoch": 0.9721328723883962,
      "flos": 22674105492480.0,
      "grad_norm": 2.5451745547725064,
      "language_loss": 0.73157007,
      "learning_rate": 8.123388903830797e-09,
      "loss": 0.75252116,
      "num_input_tokens_seen": 348992035,
      "step": 16169,
      "time_per_iteration": 2.628107786178589
    },
    {
      "auxiliary_loss_clip": 0.01069389,
      "auxiliary_loss_mlp": 0.01036626,
      "balance_loss_clip": 1.03197157,
      "balance_loss_mlp": 1.02276218,
      "epoch": 0.9721929956410642,
      "flos": 28074172360320.0,
      "grad_norm": 2.0061879287365767,
      "language_loss": 0.57746732,
      "learning_rate": 8.088360210906309e-09,
      "loss": 0.59852749,
      "num_input_tokens_seen": 349013160,
      "step": 16170,
      "time_per_iteration": 2.608646869659424
    },
    {
      "auxiliary_loss_clip": 0.01072505,
      "auxiliary_loss_mlp": 0.01031125,
      "balance_loss_clip": 1.03394914,
      "balance_loss_mlp": 1.01832795,
      "epoch": 0.9722531188937321,
      "flos": 20996251344000.0,
      "grad_norm": 1.7407565283189785,
      "language_loss": 0.71744668,
      "learning_rate": 8.053407051471062e-09,
      "loss": 0.73848295,
      "num_input_tokens_seen": 349033485,
      "step": 16171,
      "time_per_iteration": 2.5768849849700928
    },
    {
      "auxiliary_loss_clip": 0.0106968,
      "auxiliary_loss_mlp": 0.01038157,
      "balance_loss_clip": 1.0341146,
      "balance_loss_mlp": 1.0260334,
      "epoch": 0.9723132421464001,
      "flos": 16070600332800.0,
      "grad_norm": 2.0119266476064603,
      "language_loss": 0.68341005,
      "learning_rate": 8.018529426850218e-09,
      "loss": 0.70448846,
      "num_input_tokens_seen": 349051705,
      "step": 16172,
      "time_per_iteration": 2.52040433883667
    },
    {
      "auxiliary_loss_clip": 0.01087954,
      "auxiliary_loss_mlp": 0.01030424,
      "balance_loss_clip": 1.03308296,
      "balance_loss_mlp": 1.01864052,
      "epoch": 0.972373365399068,
      "flos": 27745769289600.0,
      "grad_norm": 2.04649380818029,
      "language_loss": 0.85831416,
      "learning_rate": 7.983727338366274e-09,
      "loss": 0.87949795,
      "num_input_tokens_seen": 349070825,
      "step": 16173,
      "time_per_iteration": 2.545644998550415
    },
    {
      "auxiliary_loss_clip": 0.01054921,
      "auxiliary_loss_mlp": 0.01033734,
      "balance_loss_clip": 1.0323869,
      "balance_loss_mlp": 1.01920223,
      "epoch": 0.9724334886517361,
      "flos": 23002939526400.0,
      "grad_norm": 1.8941872669699826,
      "language_loss": 0.64398479,
      "learning_rate": 7.949000787339289e-09,
      "loss": 0.66487134,
      "num_input_tokens_seen": 349089730,
      "step": 16174,
      "time_per_iteration": 2.575087547302246
    },
    {
      "auxiliary_loss_clip": 0.0109165,
      "auxiliary_loss_mlp": 0.0102519,
      "balance_loss_clip": 1.03437364,
      "balance_loss_mlp": 1.01377559,
      "epoch": 0.972493611904404,
      "flos": 25447055535360.0,
      "grad_norm": 1.7203317234380966,
      "language_loss": 0.77465701,
      "learning_rate": 7.914349775085538e-09,
      "loss": 0.79582542,
      "num_input_tokens_seen": 349111315,
      "step": 16175,
      "time_per_iteration": 2.5237858295440674
    },
    {
      "auxiliary_loss_clip": 0.01092071,
      "auxiliary_loss_mlp": 0.01030844,
      "balance_loss_clip": 1.03448594,
      "balance_loss_mlp": 1.01833296,
      "epoch": 0.972553735157072,
      "flos": 16983054547200.0,
      "grad_norm": 2.2363963307563233,
      "language_loss": 0.56517339,
      "learning_rate": 7.879774302919307e-09,
      "loss": 0.58640254,
      "num_input_tokens_seen": 349129495,
      "step": 16176,
      "time_per_iteration": 2.460719108581543
    },
    {
      "auxiliary_loss_clip": 0.01081335,
      "auxiliary_loss_mlp": 0.01028224,
      "balance_loss_clip": 1.03504908,
      "balance_loss_mlp": 1.01751935,
      "epoch": 0.97261385840974,
      "flos": 26104651776000.0,
      "grad_norm": 2.54703437448003,
      "language_loss": 0.72398221,
      "learning_rate": 7.845274372151545e-09,
      "loss": 0.74507785,
      "num_input_tokens_seen": 349148850,
      "step": 16177,
      "time_per_iteration": 2.564497709274292
    },
    {
      "auxiliary_loss_clip": 0.01081568,
      "auxiliary_loss_mlp": 0.01031366,
      "balance_loss_clip": 1.03233683,
      "balance_loss_mlp": 1.01926064,
      "epoch": 0.9726739816624079,
      "flos": 25447881548160.0,
      "grad_norm": 1.989116861241083,
      "language_loss": 0.68519974,
      "learning_rate": 7.810849984090984e-09,
      "loss": 0.70632911,
      "num_input_tokens_seen": 349167620,
      "step": 16178,
      "time_per_iteration": 2.5347790718078613
    },
    {
      "auxiliary_loss_clip": 0.0105141,
      "auxiliary_loss_mlp": 0.01032999,
      "balance_loss_clip": 1.03163576,
      "balance_loss_mlp": 1.02044046,
      "epoch": 0.972734104915076,
      "flos": 29014923513600.0,
      "grad_norm": 1.7763795408075758,
      "language_loss": 0.67535228,
      "learning_rate": 7.776501140042358e-09,
      "loss": 0.69619644,
      "num_input_tokens_seen": 349185845,
      "step": 16179,
      "time_per_iteration": 2.690690517425537
    },
    {
      "auxiliary_loss_clip": 0.01079814,
      "auxiliary_loss_mlp": 0.00780876,
      "balance_loss_clip": 1.03511083,
      "balance_loss_mlp": 1.0081439,
      "epoch": 0.9727942281677439,
      "flos": 23437637919360.0,
      "grad_norm": 1.8013697714235808,
      "language_loss": 0.77077919,
      "learning_rate": 7.742227841308624e-09,
      "loss": 0.78938609,
      "num_input_tokens_seen": 349204525,
      "step": 16180,
      "time_per_iteration": 2.5353312492370605
    },
    {
      "auxiliary_loss_clip": 0.0109411,
      "auxiliary_loss_mlp": 0.0103309,
      "balance_loss_clip": 1.03379619,
      "balance_loss_mlp": 1.02084112,
      "epoch": 0.9728543514204119,
      "flos": 31724599749120.0,
      "grad_norm": 1.5109252118041612,
      "language_loss": 0.76317549,
      "learning_rate": 7.708030089189188e-09,
      "loss": 0.78444755,
      "num_input_tokens_seen": 349228075,
      "step": 16181,
      "time_per_iteration": 2.583693504333496
    },
    {
      "auxiliary_loss_clip": 0.01100695,
      "auxiliary_loss_mlp": 0.01031502,
      "balance_loss_clip": 1.03347087,
      "balance_loss_mlp": 1.02007031,
      "epoch": 0.9729144746730798,
      "flos": 16289368116480.0,
      "grad_norm": 1.4407788177280763,
      "language_loss": 0.6330477,
      "learning_rate": 7.67390788498079e-09,
      "loss": 0.65436971,
      "num_input_tokens_seen": 349246990,
      "step": 16182,
      "time_per_iteration": 2.4226245880126953
    },
    {
      "auxiliary_loss_clip": 0.0103241,
      "auxiliary_loss_mlp": 0.01037845,
      "balance_loss_clip": 1.03459787,
      "balance_loss_mlp": 1.02470827,
      "epoch": 0.9729745979257478,
      "flos": 25041408266880.0,
      "grad_norm": 1.734714363453598,
      "language_loss": 0.62670994,
      "learning_rate": 7.639861229977507e-09,
      "loss": 0.64741242,
      "num_input_tokens_seen": 349265890,
      "step": 16183,
      "time_per_iteration": 4.300455808639526
    },
    {
      "auxiliary_loss_clip": 0.01081554,
      "auxiliary_loss_mlp": 0.01032654,
      "balance_loss_clip": 1.033759,
      "balance_loss_mlp": 1.02034545,
      "epoch": 0.9730347211784157,
      "flos": 22638733574400.0,
      "grad_norm": 1.5835142311492654,
      "language_loss": 0.78003073,
      "learning_rate": 7.605890125470527e-09,
      "loss": 0.80117273,
      "num_input_tokens_seen": 349285275,
      "step": 16184,
      "time_per_iteration": 2.748594284057617
    },
    {
      "auxiliary_loss_clip": 0.01064615,
      "auxiliary_loss_mlp": 0.01030526,
      "balance_loss_clip": 1.03090703,
      "balance_loss_mlp": 1.01833034,
      "epoch": 0.9730948444310837,
      "flos": 10998613313280.0,
      "grad_norm": 2.25985126193625,
      "language_loss": 0.79582012,
      "learning_rate": 7.571994572747709e-09,
      "loss": 0.81677151,
      "num_input_tokens_seen": 349301515,
      "step": 16185,
      "time_per_iteration": 3.901275157928467
    },
    {
      "auxiliary_loss_clip": 0.01072377,
      "auxiliary_loss_mlp": 0.0103233,
      "balance_loss_clip": 1.03419912,
      "balance_loss_mlp": 1.02048659,
      "epoch": 0.9731549676837516,
      "flos": 16799479113600.0,
      "grad_norm": 1.737289845144855,
      "language_loss": 0.78047454,
      "learning_rate": 7.538174573094469e-09,
      "loss": 0.80152154,
      "num_input_tokens_seen": 349319590,
      "step": 16186,
      "time_per_iteration": 3.9025180339813232
    },
    {
      "auxiliary_loss_clip": 0.0107957,
      "auxiliary_loss_mlp": 0.01026123,
      "balance_loss_clip": 1.03407824,
      "balance_loss_mlp": 1.01409483,
      "epoch": 0.9732150909364197,
      "flos": 21141761339520.0,
      "grad_norm": 1.610072765201185,
      "language_loss": 0.65491402,
      "learning_rate": 7.504430127793337e-09,
      "loss": 0.67597091,
      "num_input_tokens_seen": 349339230,
      "step": 16187,
      "time_per_iteration": 2.5603628158569336
    },
    {
      "auxiliary_loss_clip": 0.01076774,
      "auxiliary_loss_mlp": 0.01034243,
      "balance_loss_clip": 1.03211546,
      "balance_loss_mlp": 1.02190495,
      "epoch": 0.9732752141890876,
      "flos": 33727337435520.0,
      "grad_norm": 1.5900557529596175,
      "language_loss": 0.80368471,
      "learning_rate": 7.47076123812418e-09,
      "loss": 0.82479489,
      "num_input_tokens_seen": 349361155,
      "step": 16188,
      "time_per_iteration": 2.6004462242126465
    },
    {
      "auxiliary_loss_clip": 0.01065954,
      "auxiliary_loss_mlp": 0.010296,
      "balance_loss_clip": 1.03090048,
      "balance_loss_mlp": 1.01853192,
      "epoch": 0.9733353374417556,
      "flos": 23404384903680.0,
      "grad_norm": 1.8598405884265867,
      "language_loss": 0.78345621,
      "learning_rate": 7.437167905363084e-09,
      "loss": 0.80441171,
      "num_input_tokens_seen": 349379335,
      "step": 16189,
      "time_per_iteration": 2.605949878692627
    },
    {
      "auxiliary_loss_clip": 0.01086843,
      "auxiliary_loss_mlp": 0.01026482,
      "balance_loss_clip": 1.03210521,
      "balance_loss_mlp": 1.01470447,
      "epoch": 0.9733954606944236,
      "flos": 39165792963840.0,
      "grad_norm": 1.725605401230886,
      "language_loss": 0.50994343,
      "learning_rate": 7.403650130784367e-09,
      "loss": 0.53107667,
      "num_input_tokens_seen": 349401575,
      "step": 16190,
      "time_per_iteration": 2.616671085357666
    },
    {
      "auxiliary_loss_clip": 0.0109174,
      "auxiliary_loss_mlp": 0.01026556,
      "balance_loss_clip": 1.0342598,
      "balance_loss_mlp": 1.0145216,
      "epoch": 0.9734555839470915,
      "flos": 21981819692160.0,
      "grad_norm": 1.599473103023743,
      "language_loss": 0.81100726,
      "learning_rate": 7.3702079156590105e-09,
      "loss": 0.83219022,
      "num_input_tokens_seen": 349420650,
      "step": 16191,
      "time_per_iteration": 2.5053176879882812
    },
    {
      "auxiliary_loss_clip": 0.01087839,
      "auxiliary_loss_mlp": 0.01027403,
      "balance_loss_clip": 1.03261256,
      "balance_loss_mlp": 1.01622117,
      "epoch": 0.9735157071997596,
      "flos": 16575539771520.0,
      "grad_norm": 1.9240648366275974,
      "language_loss": 0.82702613,
      "learning_rate": 7.336841261255111e-09,
      "loss": 0.84817851,
      "num_input_tokens_seen": 349436830,
      "step": 16192,
      "time_per_iteration": 2.439758062362671
    },
    {
      "auxiliary_loss_clip": 0.01045279,
      "auxiliary_loss_mlp": 0.01037855,
      "balance_loss_clip": 1.03696847,
      "balance_loss_mlp": 1.02467656,
      "epoch": 0.9735758304524275,
      "flos": 20223237726720.0,
      "grad_norm": 1.986827920520046,
      "language_loss": 0.75161612,
      "learning_rate": 7.303550168837658e-09,
      "loss": 0.77244747,
      "num_input_tokens_seen": 349454325,
      "step": 16193,
      "time_per_iteration": 4.105252027511597
    },
    {
      "auxiliary_loss_clip": 0.01073011,
      "auxiliary_loss_mlp": 0.01031774,
      "balance_loss_clip": 1.03269362,
      "balance_loss_mlp": 1.02139068,
      "epoch": 0.9736359537050955,
      "flos": 23653353047040.0,
      "grad_norm": 1.9117253201455835,
      "language_loss": 0.85124832,
      "learning_rate": 7.270334639669417e-09,
      "loss": 0.87229615,
      "num_input_tokens_seen": 349470230,
      "step": 16194,
      "time_per_iteration": 2.539759874343872
    },
    {
      "auxiliary_loss_clip": 0.01066774,
      "auxiliary_loss_mlp": 0.01034871,
      "balance_loss_clip": 1.0347321,
      "balance_loss_mlp": 1.02284312,
      "epoch": 0.9736960769577634,
      "flos": 15560202026880.0,
      "grad_norm": 1.5314056868889963,
      "language_loss": 0.75719351,
      "learning_rate": 7.237194675009828e-09,
      "loss": 0.77820992,
      "num_input_tokens_seen": 349486250,
      "step": 16195,
      "time_per_iteration": 2.4988338947296143
    },
    {
      "auxiliary_loss_clip": 0.0100777,
      "auxiliary_loss_mlp": 0.00999608,
      "balance_loss_clip": 1.01363599,
      "balance_loss_mlp": 0.99839193,
      "epoch": 0.9737562002104314,
      "flos": 65351783088000.0,
      "grad_norm": 0.7052914727605334,
      "language_loss": 0.52520251,
      "learning_rate": 7.204130276115439e-09,
      "loss": 0.54527628,
      "num_input_tokens_seen": 349545865,
      "step": 16196,
      "time_per_iteration": 3.114932060241699
    },
    {
      "auxiliary_loss_clip": 0.0107893,
      "auxiliary_loss_mlp": 0.01028928,
      "balance_loss_clip": 1.03589928,
      "balance_loss_mlp": 1.01777601,
      "epoch": 0.9738163234630993,
      "flos": 27196730928000.0,
      "grad_norm": 1.5449591268917202,
      "language_loss": 0.76301289,
      "learning_rate": 7.171141444240136e-09,
      "loss": 0.78409147,
      "num_input_tokens_seen": 349566080,
      "step": 16197,
      "time_per_iteration": 2.572955369949341
    },
    {
      "auxiliary_loss_clip": 0.01106411,
      "auxiliary_loss_mlp": 0.01031193,
      "balance_loss_clip": 1.03492272,
      "balance_loss_mlp": 1.01893187,
      "epoch": 0.9738764467157673,
      "flos": 21069365477760.0,
      "grad_norm": 1.7603078546933115,
      "language_loss": 0.67396808,
      "learning_rate": 7.13822818063492e-09,
      "loss": 0.69534415,
      "num_input_tokens_seen": 349585665,
      "step": 16198,
      "time_per_iteration": 2.4785072803497314
    },
    {
      "auxiliary_loss_clip": 0.01101653,
      "auxiliary_loss_mlp": 0.01028082,
      "balance_loss_clip": 1.03311491,
      "balance_loss_mlp": 1.01543355,
      "epoch": 0.9739365699684353,
      "flos": 21361211481600.0,
      "grad_norm": 1.8405775387434822,
      "language_loss": 0.77539003,
      "learning_rate": 7.10539048654768e-09,
      "loss": 0.79668736,
      "num_input_tokens_seen": 349605125,
      "step": 16199,
      "time_per_iteration": 2.439615488052368
    },
    {
      "auxiliary_loss_clip": 0.01078231,
      "auxiliary_loss_mlp": 0.01031926,
      "balance_loss_clip": 1.03465521,
      "balance_loss_mlp": 1.02029681,
      "epoch": 0.9739966932211033,
      "flos": 21902061542400.0,
      "grad_norm": 2.3569813097820056,
      "language_loss": 0.79409075,
      "learning_rate": 7.072628363223865e-09,
      "loss": 0.81519228,
      "num_input_tokens_seen": 349623360,
      "step": 16200,
      "time_per_iteration": 2.5391414165496826
    },
    {
      "auxiliary_loss_clip": 0.01052254,
      "auxiliary_loss_mlp": 0.01035427,
      "balance_loss_clip": 1.03742003,
      "balance_loss_mlp": 1.02299345,
      "epoch": 0.9740568164737712,
      "flos": 24827345164800.0,
      "grad_norm": 1.9618109719462753,
      "language_loss": 0.67954451,
      "learning_rate": 7.039941811905592e-09,
      "loss": 0.70042133,
      "num_input_tokens_seen": 349644390,
      "step": 16201,
      "time_per_iteration": 2.6456668376922607
    },
    {
      "auxiliary_loss_clip": 0.01071767,
      "auxiliary_loss_mlp": 0.01030552,
      "balance_loss_clip": 1.0336318,
      "balance_loss_mlp": 1.0189352,
      "epoch": 0.9741169397264392,
      "flos": 23623583650560.0,
      "grad_norm": 1.74333248478659,
      "language_loss": 0.72693992,
      "learning_rate": 7.0073308338325364e-09,
      "loss": 0.74796319,
      "num_input_tokens_seen": 349663200,
      "step": 16202,
      "time_per_iteration": 2.5729103088378906
    },
    {
      "auxiliary_loss_clip": 0.01082488,
      "auxiliary_loss_mlp": 0.01028782,
      "balance_loss_clip": 1.03651237,
      "balance_loss_mlp": 1.01632476,
      "epoch": 0.9741770629791072,
      "flos": 18841144164480.0,
      "grad_norm": 2.699894638001778,
      "language_loss": 0.72986746,
      "learning_rate": 6.974795430241265e-09,
      "loss": 0.75098014,
      "num_input_tokens_seen": 349681975,
      "step": 16203,
      "time_per_iteration": 2.5111920833587646
    },
    {
      "auxiliary_loss_clip": 0.01102111,
      "auxiliary_loss_mlp": 0.01032273,
      "balance_loss_clip": 1.03414845,
      "balance_loss_mlp": 1.0202384,
      "epoch": 0.9742371862317751,
      "flos": 22346241125760.0,
      "grad_norm": 1.8927200278486223,
      "language_loss": 0.77422678,
      "learning_rate": 6.942335602365235e-09,
      "loss": 0.79557061,
      "num_input_tokens_seen": 349701185,
      "step": 16204,
      "time_per_iteration": 2.4865241050720215
    },
    {
      "auxiliary_loss_clip": 0.01086189,
      "auxiliary_loss_mlp": 0.010333,
      "balance_loss_clip": 1.03673553,
      "balance_loss_mlp": 1.02034187,
      "epoch": 0.9742973094844432,
      "flos": 21762764599680.0,
      "grad_norm": 1.996078393612719,
      "language_loss": 0.79661882,
      "learning_rate": 6.909951351435905e-09,
      "loss": 0.81781375,
      "num_input_tokens_seen": 349720360,
      "step": 16205,
      "time_per_iteration": 2.5131676197052
    },
    {
      "auxiliary_loss_clip": 0.01102268,
      "auxiliary_loss_mlp": 0.01030479,
      "balance_loss_clip": 1.03472543,
      "balance_loss_mlp": 1.01882648,
      "epoch": 0.9743574327371111,
      "flos": 26248725227520.0,
      "grad_norm": 1.645253948501529,
      "language_loss": 0.74461412,
      "learning_rate": 6.87764267868074e-09,
      "loss": 0.76594162,
      "num_input_tokens_seen": 349741040,
      "step": 16206,
      "time_per_iteration": 2.5108933448791504
    },
    {
      "auxiliary_loss_clip": 0.01047491,
      "auxiliary_loss_mlp": 0.01028226,
      "balance_loss_clip": 1.03247452,
      "balance_loss_mlp": 1.01576269,
      "epoch": 0.9744175559897791,
      "flos": 12349321367040.0,
      "grad_norm": 2.3748894671054583,
      "language_loss": 0.8368926,
      "learning_rate": 6.8454095853252015e-09,
      "loss": 0.8576498,
      "num_input_tokens_seen": 349758895,
      "step": 16207,
      "time_per_iteration": 2.5878524780273438
    },
    {
      "auxiliary_loss_clip": 0.01090334,
      "auxiliary_loss_mlp": 0.01033004,
      "balance_loss_clip": 1.03375554,
      "balance_loss_mlp": 1.02156544,
      "epoch": 0.974477679242447,
      "flos": 28397834835840.0,
      "grad_norm": 1.878114784587456,
      "language_loss": 0.70740348,
      "learning_rate": 6.813252072591425e-09,
      "loss": 0.72863686,
      "num_input_tokens_seen": 349779740,
      "step": 16208,
      "time_per_iteration": 2.548093795776367
    },
    {
      "auxiliary_loss_clip": 0.01064242,
      "auxiliary_loss_mlp": 0.01027161,
      "balance_loss_clip": 1.03308439,
      "balance_loss_mlp": 1.01681399,
      "epoch": 0.974537802495115,
      "flos": 17785370684160.0,
      "grad_norm": 1.6242257841218073,
      "language_loss": 0.77528834,
      "learning_rate": 6.781170141698878e-09,
      "loss": 0.79620242,
      "num_input_tokens_seen": 349796820,
      "step": 16209,
      "time_per_iteration": 2.5201070308685303
    },
    {
      "auxiliary_loss_clip": 0.01068817,
      "auxiliary_loss_mlp": 0.00784214,
      "balance_loss_clip": 1.03332257,
      "balance_loss_mlp": 1.0073539,
      "epoch": 0.9745979257477829,
      "flos": 23842315520640.0,
      "grad_norm": 1.8340441606208908,
      "language_loss": 0.78866637,
      "learning_rate": 6.749163793864144e-09,
      "loss": 0.80719662,
      "num_input_tokens_seen": 349816550,
      "step": 16210,
      "time_per_iteration": 2.5759286880493164
    },
    {
      "auxiliary_loss_clip": 0.01077633,
      "auxiliary_loss_mlp": 0.01034503,
      "balance_loss_clip": 1.03201413,
      "balance_loss_mlp": 1.02279711,
      "epoch": 0.9746580490004509,
      "flos": 27016172236800.0,
      "grad_norm": 2.1389464114487136,
      "language_loss": 0.77833802,
      "learning_rate": 6.7172330303009176e-09,
      "loss": 0.79945934,
      "num_input_tokens_seen": 349834350,
      "step": 16211,
      "time_per_iteration": 2.5445737838745117
    },
    {
      "auxiliary_loss_clip": 0.01069347,
      "auxiliary_loss_mlp": 0.0103392,
      "balance_loss_clip": 1.03314388,
      "balance_loss_mlp": 1.02054477,
      "epoch": 0.9747181722531189,
      "flos": 19792022952960.0,
      "grad_norm": 1.8933659705430461,
      "language_loss": 0.7785427,
      "learning_rate": 6.685377852219787e-09,
      "loss": 0.79957539,
      "num_input_tokens_seen": 349853460,
      "step": 16212,
      "time_per_iteration": 2.5458004474639893
    },
    {
      "auxiliary_loss_clip": 0.01071161,
      "auxiliary_loss_mlp": 0.01033176,
      "balance_loss_clip": 1.03290009,
      "balance_loss_mlp": 1.02149367,
      "epoch": 0.9747782955057869,
      "flos": 31430598929280.0,
      "grad_norm": 1.5148682597354217,
      "language_loss": 0.80195928,
      "learning_rate": 6.653598260829118e-09,
      "loss": 0.8230027,
      "num_input_tokens_seen": 349874830,
      "step": 16213,
      "time_per_iteration": 2.5943148136138916
    },
    {
      "auxiliary_loss_clip": 0.01056872,
      "auxiliary_loss_mlp": 0.01026723,
      "balance_loss_clip": 1.03107309,
      "balance_loss_mlp": 1.01517725,
      "epoch": 0.9748384187584548,
      "flos": 15961288268160.0,
      "grad_norm": 1.847879114485383,
      "language_loss": 0.66464078,
      "learning_rate": 6.6218942573335044e-09,
      "loss": 0.68547672,
      "num_input_tokens_seen": 349893690,
      "step": 16214,
      "time_per_iteration": 2.5532827377319336
    },
    {
      "auxiliary_loss_clip": 0.01086558,
      "auxiliary_loss_mlp": 0.01031974,
      "balance_loss_clip": 1.03630018,
      "balance_loss_mlp": 1.01947534,
      "epoch": 0.9748985420111228,
      "flos": 20558715776640.0,
      "grad_norm": 1.572585041133039,
      "language_loss": 0.74518317,
      "learning_rate": 6.5902658429355386e-09,
      "loss": 0.76636851,
      "num_input_tokens_seen": 349912480,
      "step": 16215,
      "time_per_iteration": 2.5229077339172363
    },
    {
      "auxiliary_loss_clip": 0.01051472,
      "auxiliary_loss_mlp": 0.01031475,
      "balance_loss_clip": 1.03232598,
      "balance_loss_mlp": 1.01952434,
      "epoch": 0.9749586652637908,
      "flos": 36721605127680.0,
      "grad_norm": 1.6613456250563972,
      "language_loss": 0.67061186,
      "learning_rate": 6.558713018834483e-09,
      "loss": 0.69144136,
      "num_input_tokens_seen": 349932470,
      "step": 16216,
      "time_per_iteration": 2.6965579986572266
    },
    {
      "auxiliary_loss_clip": 0.01049866,
      "auxiliary_loss_mlp": 0.01029588,
      "balance_loss_clip": 1.03303981,
      "balance_loss_mlp": 1.01674271,
      "epoch": 0.9750187885164587,
      "flos": 10999223844480.0,
      "grad_norm": 1.8380577437855554,
      "language_loss": 0.71811211,
      "learning_rate": 6.527235786226937e-09,
      "loss": 0.73890662,
      "num_input_tokens_seen": 349949060,
      "step": 16217,
      "time_per_iteration": 2.5712411403656006
    },
    {
      "auxiliary_loss_clip": 0.01070445,
      "auxiliary_loss_mlp": 0.01030923,
      "balance_loss_clip": 1.03669345,
      "balance_loss_mlp": 1.01887655,
      "epoch": 0.9750789117691268,
      "flos": 25739512070400.0,
      "grad_norm": 1.5191773847985837,
      "language_loss": 0.78427213,
      "learning_rate": 6.495834146306167e-09,
      "loss": 0.80528581,
      "num_input_tokens_seen": 349968010,
      "step": 16218,
      "time_per_iteration": 2.5960443019866943
    },
    {
      "auxiliary_loss_clip": 0.01075961,
      "auxiliary_loss_mlp": 0.01028669,
      "balance_loss_clip": 1.03478599,
      "balance_loss_mlp": 1.0166285,
      "epoch": 0.9751390350217947,
      "flos": 13333955961600.0,
      "grad_norm": 2.019418481114559,
      "language_loss": 0.7723974,
      "learning_rate": 6.464508100263222e-09,
      "loss": 0.79344368,
      "num_input_tokens_seen": 349985270,
      "step": 16219,
      "time_per_iteration": 2.5324318408966064
    },
    {
      "auxiliary_loss_clip": 0.01083858,
      "auxiliary_loss_mlp": 0.01030202,
      "balance_loss_clip": 1.03505301,
      "balance_loss_mlp": 1.01819158,
      "epoch": 0.9751991582744627,
      "flos": 22820621068800.0,
      "grad_norm": 1.7517362957251144,
      "language_loss": 0.81123257,
      "learning_rate": 6.433257649285817e-09,
      "loss": 0.83237314,
      "num_input_tokens_seen": 350003935,
      "step": 16220,
      "time_per_iteration": 2.524190664291382
    },
    {
      "auxiliary_loss_clip": 0.0110021,
      "auxiliary_loss_mlp": 0.01029884,
      "balance_loss_clip": 1.03334093,
      "balance_loss_mlp": 1.01884544,
      "epoch": 0.9752592815271306,
      "flos": 19646189735040.0,
      "grad_norm": 1.9096040856918477,
      "language_loss": 0.75284684,
      "learning_rate": 6.402082794559227e-09,
      "loss": 0.77414781,
      "num_input_tokens_seen": 350023595,
      "step": 16221,
      "time_per_iteration": 2.459115982055664
    },
    {
      "auxiliary_loss_clip": 0.01068234,
      "auxiliary_loss_mlp": 0.01029711,
      "balance_loss_clip": 1.03357852,
      "balance_loss_mlp": 1.018255,
      "epoch": 0.9753194047797986,
      "flos": 26690462686080.0,
      "grad_norm": 1.62602700671209,
      "language_loss": 0.66646689,
      "learning_rate": 6.370983537265395e-09,
      "loss": 0.68744636,
      "num_input_tokens_seen": 350045920,
      "step": 16222,
      "time_per_iteration": 3.9954135417938232
    },
    {
      "auxiliary_loss_clip": 0.01089424,
      "auxiliary_loss_mlp": 0.01029921,
      "balance_loss_clip": 1.0332942,
      "balance_loss_mlp": 1.01848876,
      "epoch": 0.9753795280324665,
      "flos": 23221779137280.0,
      "grad_norm": 1.6355602942662584,
      "language_loss": 0.88495791,
      "learning_rate": 6.3399598785836004e-09,
      "loss": 0.90615135,
      "num_input_tokens_seen": 350063925,
      "step": 16223,
      "time_per_iteration": 3.915684223175049
    },
    {
      "auxiliary_loss_clip": 0.01040754,
      "auxiliary_loss_mlp": 0.01031259,
      "balance_loss_clip": 1.03257382,
      "balance_loss_mlp": 1.01962996,
      "epoch": 0.9754396512851345,
      "flos": 19463835363840.0,
      "grad_norm": 1.6998813325021103,
      "language_loss": 0.74973035,
      "learning_rate": 6.309011819690457e-09,
      "loss": 0.77045047,
      "num_input_tokens_seen": 350080900,
      "step": 16224,
      "time_per_iteration": 4.066991806030273
    },
    {
      "auxiliary_loss_clip": 0.01011983,
      "auxiliary_loss_mlp": 0.01004932,
      "balance_loss_clip": 1.00930357,
      "balance_loss_mlp": 1.00388908,
      "epoch": 0.9754997745378025,
      "flos": 68459313340800.0,
      "grad_norm": 0.8070556716042739,
      "language_loss": 0.59187698,
      "learning_rate": 6.278139361759249e-09,
      "loss": 0.61204612,
      "num_input_tokens_seen": 350144550,
      "step": 16225,
      "time_per_iteration": 3.1212191581726074
    },
    {
      "auxiliary_loss_clip": 0.01069566,
      "auxiliary_loss_mlp": 0.00781407,
      "balance_loss_clip": 1.03397882,
      "balance_loss_mlp": 1.00776112,
      "epoch": 0.9755598977904705,
      "flos": 26395168976640.0,
      "grad_norm": 1.7136801442173597,
      "language_loss": 0.69101763,
      "learning_rate": 6.247342505960818e-09,
      "loss": 0.70952737,
      "num_input_tokens_seen": 350164050,
      "step": 16226,
      "time_per_iteration": 2.597308874130249
    },
    {
      "auxiliary_loss_clip": 0.01090324,
      "auxiliary_loss_mlp": 0.01039882,
      "balance_loss_clip": 1.03476739,
      "balance_loss_mlp": 1.02730513,
      "epoch": 0.9756200210431384,
      "flos": 16617663446400.0,
      "grad_norm": 1.6463911048757767,
      "language_loss": 0.83218205,
      "learning_rate": 6.216621253462894e-09,
      "loss": 0.85348415,
      "num_input_tokens_seen": 350181350,
      "step": 16227,
      "time_per_iteration": 2.477663993835449
    },
    {
      "auxiliary_loss_clip": 0.01101127,
      "auxiliary_loss_mlp": 0.01029821,
      "balance_loss_clip": 1.03432798,
      "balance_loss_mlp": 1.01869917,
      "epoch": 0.9756801442958064,
      "flos": 23623044946560.0,
      "grad_norm": 1.7184108113514194,
      "language_loss": 0.7761488,
      "learning_rate": 6.185975605430549e-09,
      "loss": 0.79745829,
      "num_input_tokens_seen": 350199765,
      "step": 16228,
      "time_per_iteration": 2.464442729949951
    },
    {
      "auxiliary_loss_clip": 0.01020155,
      "auxiliary_loss_mlp": 0.01001106,
      "balance_loss_clip": 1.00622511,
      "balance_loss_mlp": 1.00009918,
      "epoch": 0.9757402675484744,
      "flos": 61625799440640.0,
      "grad_norm": 0.8461153539665102,
      "language_loss": 0.55921006,
      "learning_rate": 6.155405563025962e-09,
      "loss": 0.57942271,
      "num_input_tokens_seen": 350256420,
      "step": 16229,
      "time_per_iteration": 3.0275909900665283
    },
    {
      "auxiliary_loss_clip": 0.0109301,
      "auxiliary_loss_mlp": 0.0103042,
      "balance_loss_clip": 1.03394103,
      "balance_loss_mlp": 1.01827884,
      "epoch": 0.9758003908011423,
      "flos": 24058964401920.0,
      "grad_norm": 1.593286354364073,
      "language_loss": 0.75375754,
      "learning_rate": 6.124911127407984e-09,
      "loss": 0.77499187,
      "num_input_tokens_seen": 350276270,
      "step": 16230,
      "time_per_iteration": 2.5159547328948975
    },
    {
      "auxiliary_loss_clip": 0.01077489,
      "auxiliary_loss_mlp": 0.0102895,
      "balance_loss_clip": 1.03449619,
      "balance_loss_mlp": 1.01804781,
      "epoch": 0.9758605140538104,
      "flos": 17493093717120.0,
      "grad_norm": 1.7981831691033496,
      "language_loss": 0.72065485,
      "learning_rate": 6.094492299733245e-09,
      "loss": 0.74171931,
      "num_input_tokens_seen": 350295000,
      "step": 16231,
      "time_per_iteration": 2.4809131622314453
    },
    {
      "auxiliary_loss_clip": 0.0108455,
      "auxiliary_loss_mlp": 0.0102704,
      "balance_loss_clip": 1.03638947,
      "balance_loss_mlp": 1.01473212,
      "epoch": 0.9759206373064783,
      "flos": 24826950115200.0,
      "grad_norm": 1.8220781198946452,
      "language_loss": 0.76599866,
      "learning_rate": 6.064149081155267e-09,
      "loss": 0.78711462,
      "num_input_tokens_seen": 350314980,
      "step": 16232,
      "time_per_iteration": 3.9515202045440674
    },
    {
      "auxiliary_loss_clip": 0.01011073,
      "auxiliary_loss_mlp": 0.01001988,
      "balance_loss_clip": 1.01191604,
      "balance_loss_mlp": 1.00075376,
      "epoch": 0.9759807605591463,
      "flos": 68161182456960.0,
      "grad_norm": 0.7388833109841759,
      "language_loss": 0.53799868,
      "learning_rate": 6.033881472824465e-09,
      "loss": 0.55812931,
      "num_input_tokens_seen": 350371985,
      "step": 16233,
      "time_per_iteration": 2.989121437072754
    },
    {
      "auxiliary_loss_clip": 0.01103019,
      "auxiliary_loss_mlp": 0.010326,
      "balance_loss_clip": 1.03395987,
      "balance_loss_mlp": 1.02054238,
      "epoch": 0.9760408838118142,
      "flos": 18989239939200.0,
      "grad_norm": 1.6991760460212633,
      "language_loss": 0.71560621,
      "learning_rate": 6.003689475888807e-09,
      "loss": 0.73696232,
      "num_input_tokens_seen": 350390590,
      "step": 16234,
      "time_per_iteration": 2.458466053009033
    },
    {
      "auxiliary_loss_clip": 0.01095381,
      "auxiliary_loss_mlp": 0.01032683,
      "balance_loss_clip": 1.03516197,
      "balance_loss_mlp": 1.02031517,
      "epoch": 0.9761010070644822,
      "flos": 17125978763520.0,
      "grad_norm": 2.2380437918665765,
      "language_loss": 0.78690183,
      "learning_rate": 5.973573091493156e-09,
      "loss": 0.80818248,
      "num_input_tokens_seen": 350403770,
      "step": 16235,
      "time_per_iteration": 2.4426708221435547
    },
    {
      "auxiliary_loss_clip": 0.01094422,
      "auxiliary_loss_mlp": 0.01030307,
      "balance_loss_clip": 1.0365963,
      "balance_loss_mlp": 1.01708627,
      "epoch": 0.9761611303171501,
      "flos": 22052599441920.0,
      "grad_norm": 1.745300429973555,
      "language_loss": 0.76562017,
      "learning_rate": 5.943532320779265e-09,
      "loss": 0.7868675,
      "num_input_tokens_seen": 350421870,
      "step": 16236,
      "time_per_iteration": 2.492396354675293
    },
    {
      "auxiliary_loss_clip": 0.01091466,
      "auxiliary_loss_mlp": 0.01027409,
      "balance_loss_clip": 1.03294849,
      "balance_loss_mlp": 1.0159533,
      "epoch": 0.9762212535698181,
      "flos": 21757521214080.0,
      "grad_norm": 1.7447040005389491,
      "language_loss": 0.75475562,
      "learning_rate": 5.913567164886446e-09,
      "loss": 0.77594435,
      "num_input_tokens_seen": 350440025,
      "step": 16237,
      "time_per_iteration": 2.464792013168335
    },
    {
      "auxiliary_loss_clip": 0.01057853,
      "auxiliary_loss_mlp": 0.0103427,
      "balance_loss_clip": 1.03231788,
      "balance_loss_mlp": 1.02010822,
      "epoch": 0.9762813768224861,
      "flos": 25921615046400.0,
      "grad_norm": 1.7131957699876106,
      "language_loss": 0.73351359,
      "learning_rate": 5.8836776249509e-09,
      "loss": 0.75443482,
      "num_input_tokens_seen": 350459435,
      "step": 16238,
      "time_per_iteration": 2.6084163188934326
    },
    {
      "auxiliary_loss_clip": 0.01082279,
      "auxiliary_loss_mlp": 0.00782932,
      "balance_loss_clip": 1.03347445,
      "balance_loss_mlp": 1.0073154,
      "epoch": 0.9763415000751541,
      "flos": 24051853509120.0,
      "grad_norm": 2.2340644851838953,
      "language_loss": 0.84179014,
      "learning_rate": 5.8538637021063875e-09,
      "loss": 0.86044228,
      "num_input_tokens_seen": 350472655,
      "step": 16239,
      "time_per_iteration": 2.501715898513794
    },
    {
      "auxiliary_loss_clip": 0.01063689,
      "auxiliary_loss_mlp": 0.01039913,
      "balance_loss_clip": 1.03243995,
      "balance_loss_mlp": 1.02572083,
      "epoch": 0.976401623327822,
      "flos": 17018677860480.0,
      "grad_norm": 2.770517820672524,
      "language_loss": 0.60721666,
      "learning_rate": 5.824125397483115e-09,
      "loss": 0.62825269,
      "num_input_tokens_seen": 350488160,
      "step": 16240,
      "time_per_iteration": 2.5278990268707275
    },
    {
      "auxiliary_loss_clip": 0.01065051,
      "auxiliary_loss_mlp": 0.01029945,
      "balance_loss_clip": 1.03386521,
      "balance_loss_mlp": 1.01810789,
      "epoch": 0.97646174658049,
      "flos": 16106941918080.0,
      "grad_norm": 2.093651005077876,
      "language_loss": 0.82889271,
      "learning_rate": 5.7944627122088474e-09,
      "loss": 0.84984267,
      "num_input_tokens_seen": 350506065,
      "step": 16241,
      "time_per_iteration": 2.5139880180358887
    },
    {
      "auxiliary_loss_clip": 0.01064031,
      "auxiliary_loss_mlp": 0.0103733,
      "balance_loss_clip": 1.03213573,
      "balance_loss_mlp": 1.02566504,
      "epoch": 0.9765218698331579,
      "flos": 21252725429760.0,
      "grad_norm": 6.266782557522362,
      "language_loss": 0.83402902,
      "learning_rate": 5.764875647408463e-09,
      "loss": 0.85504264,
      "num_input_tokens_seen": 350524495,
      "step": 16242,
      "time_per_iteration": 2.5607330799102783
    },
    {
      "auxiliary_loss_clip": 0.01093044,
      "auxiliary_loss_mlp": 0.01031321,
      "balance_loss_clip": 1.03539336,
      "balance_loss_mlp": 1.0193826,
      "epoch": 0.9765819930858259,
      "flos": 18588045957120.0,
      "grad_norm": 1.5477688419629865,
      "language_loss": 0.75314415,
      "learning_rate": 5.7353642042037294e-09,
      "loss": 0.77438784,
      "num_input_tokens_seen": 350544185,
      "step": 16243,
      "time_per_iteration": 2.4760124683380127
    },
    {
      "auxiliary_loss_clip": 0.01091422,
      "auxiliary_loss_mlp": 0.01035858,
      "balance_loss_clip": 1.03341877,
      "balance_loss_mlp": 1.02315032,
      "epoch": 0.976642116338494,
      "flos": 20266833859200.0,
      "grad_norm": 1.54860748723667,
      "language_loss": 0.69773817,
      "learning_rate": 5.705928383713754e-09,
      "loss": 0.71901095,
      "num_input_tokens_seen": 350562675,
      "step": 16244,
      "time_per_iteration": 2.492584228515625
    },
    {
      "auxiliary_loss_clip": 0.01084295,
      "auxiliary_loss_mlp": 0.01032976,
      "balance_loss_clip": 1.03666544,
      "balance_loss_mlp": 1.02017879,
      "epoch": 0.9767022395911619,
      "flos": 25550477769600.0,
      "grad_norm": 1.7584451958874965,
      "language_loss": 0.83791471,
      "learning_rate": 5.676568187055197e-09,
      "loss": 0.85908735,
      "num_input_tokens_seen": 350581535,
      "step": 16245,
      "time_per_iteration": 2.535740852355957
    },
    {
      "auxiliary_loss_clip": 0.01052431,
      "auxiliary_loss_mlp": 0.01026374,
      "balance_loss_clip": 1.03374076,
      "balance_loss_mlp": 1.01499581,
      "epoch": 0.9767623628438299,
      "flos": 21762656858880.0,
      "grad_norm": 1.4185966726580308,
      "language_loss": 0.78682923,
      "learning_rate": 5.647283615340726e-09,
      "loss": 0.80761725,
      "num_input_tokens_seen": 350601615,
      "step": 16246,
      "time_per_iteration": 2.590569019317627
    },
    {
      "auxiliary_loss_clip": 0.01096334,
      "auxiliary_loss_mlp": 0.01031266,
      "balance_loss_clip": 1.03352177,
      "balance_loss_mlp": 1.02087736,
      "epoch": 0.9768224860964978,
      "flos": 15851114277120.0,
      "grad_norm": 1.4942541918846772,
      "language_loss": 0.74054772,
      "learning_rate": 5.6180746696812275e-09,
      "loss": 0.76182371,
      "num_input_tokens_seen": 350619580,
      "step": 16247,
      "time_per_iteration": 2.443958282470703
    },
    {
      "auxiliary_loss_clip": 0.01047395,
      "auxiliary_loss_mlp": 0.01033879,
      "balance_loss_clip": 1.03144598,
      "balance_loss_mlp": 1.02119553,
      "epoch": 0.9768826093491658,
      "flos": 25151151294720.0,
      "grad_norm": 1.7374556662951117,
      "language_loss": 0.79983008,
      "learning_rate": 5.58894135118404e-09,
      "loss": 0.82064289,
      "num_input_tokens_seen": 350640015,
      "step": 16248,
      "time_per_iteration": 2.6566905975341797
    },
    {
      "auxiliary_loss_clip": 0.01044841,
      "auxiliary_loss_mlp": 0.01041071,
      "balance_loss_clip": 1.0359894,
      "balance_loss_mlp": 1.0266645,
      "epoch": 0.9769427326018337,
      "flos": 22967028904320.0,
      "grad_norm": 1.7209302516547393,
      "language_loss": 0.78978133,
      "learning_rate": 5.559883660954278e-09,
      "loss": 0.81064051,
      "num_input_tokens_seen": 350659155,
      "step": 16249,
      "time_per_iteration": 2.6156554222106934
    },
    {
      "auxiliary_loss_clip": 0.01089408,
      "auxiliary_loss_mlp": 0.01032824,
      "balance_loss_clip": 1.03487587,
      "balance_loss_mlp": 1.02060556,
      "epoch": 0.9770028558545018,
      "flos": 15264297786240.0,
      "grad_norm": 1.939715292768675,
      "language_loss": 0.66626632,
      "learning_rate": 5.530901600093507e-09,
      "loss": 0.68748862,
      "num_input_tokens_seen": 350676615,
      "step": 16250,
      "time_per_iteration": 2.474362850189209
    },
    {
      "auxiliary_loss_clip": 0.01027021,
      "auxiliary_loss_mlp": 0.01001323,
      "balance_loss_clip": 1.00432515,
      "balance_loss_mlp": 1.00027418,
      "epoch": 0.9770629791071697,
      "flos": 71450348808960.0,
      "grad_norm": 0.7978868102749015,
      "language_loss": 0.59920645,
      "learning_rate": 5.501995169700846e-09,
      "loss": 0.61948991,
      "num_input_tokens_seen": 350736805,
      "step": 16251,
      "time_per_iteration": 3.102309226989746
    },
    {
      "auxiliary_loss_clip": 0.01089195,
      "auxiliary_loss_mlp": 0.01030449,
      "balance_loss_clip": 1.03376174,
      "balance_loss_mlp": 1.01809859,
      "epoch": 0.9771231023598377,
      "flos": 22412854897920.0,
      "grad_norm": 1.7543046536507896,
      "language_loss": 0.78549391,
      "learning_rate": 5.473164370872307e-09,
      "loss": 0.80669034,
      "num_input_tokens_seen": 350753600,
      "step": 16252,
      "time_per_iteration": 2.486893892288208
    },
    {
      "auxiliary_loss_clip": 0.01086153,
      "auxiliary_loss_mlp": 0.01036241,
      "balance_loss_clip": 1.03291428,
      "balance_loss_mlp": 1.02343822,
      "epoch": 0.9771832256125056,
      "flos": 19025940660480.0,
      "grad_norm": 2.673117242616465,
      "language_loss": 0.64451087,
      "learning_rate": 5.444409204701461e-09,
      "loss": 0.66573483,
      "num_input_tokens_seen": 350771225,
      "step": 16253,
      "time_per_iteration": 2.451687812805176
    },
    {
      "auxiliary_loss_clip": 0.01097853,
      "auxiliary_loss_mlp": 0.01032404,
      "balance_loss_clip": 1.03704131,
      "balance_loss_mlp": 1.01905274,
      "epoch": 0.9772433488651736,
      "flos": 17822143232640.0,
      "grad_norm": 2.1499575619235056,
      "language_loss": 0.7691288,
      "learning_rate": 5.415729672278324e-09,
      "loss": 0.79043138,
      "num_input_tokens_seen": 350789100,
      "step": 16254,
      "time_per_iteration": 2.4666433334350586
    },
    {
      "auxiliary_loss_clip": 0.01095633,
      "auxiliary_loss_mlp": 0.01032023,
      "balance_loss_clip": 1.03436255,
      "balance_loss_mlp": 1.02004814,
      "epoch": 0.9773034721178415,
      "flos": 37629785623680.0,
      "grad_norm": 1.6607793284969148,
      "language_loss": 0.63967144,
      "learning_rate": 5.387125774690471e-09,
      "loss": 0.66094804,
      "num_input_tokens_seen": 350811085,
      "step": 16255,
      "time_per_iteration": 2.617396593093872
    },
    {
      "auxiliary_loss_clip": 0.01075243,
      "auxiliary_loss_mlp": 0.00783181,
      "balance_loss_clip": 1.03388095,
      "balance_loss_mlp": 1.00793254,
      "epoch": 0.9773635953705095,
      "flos": 20302457172480.0,
      "grad_norm": 1.5516011154263585,
      "language_loss": 0.75917006,
      "learning_rate": 5.358597513023033e-09,
      "loss": 0.77775431,
      "num_input_tokens_seen": 350831065,
      "step": 16256,
      "time_per_iteration": 2.583519220352173
    },
    {
      "auxiliary_loss_clip": 0.01101962,
      "auxiliary_loss_mlp": 0.01032399,
      "balance_loss_clip": 1.03580904,
      "balance_loss_mlp": 1.02001941,
      "epoch": 0.9774237186231776,
      "flos": 22309253095680.0,
      "grad_norm": 1.9081525251327776,
      "language_loss": 0.78285342,
      "learning_rate": 5.330144888357369e-09,
      "loss": 0.80419707,
      "num_input_tokens_seen": 350849675,
      "step": 16257,
      "time_per_iteration": 2.4482641220092773
    },
    {
      "auxiliary_loss_clip": 0.01087499,
      "auxiliary_loss_mlp": 0.01031064,
      "balance_loss_clip": 1.03455901,
      "balance_loss_mlp": 1.01883912,
      "epoch": 0.9774838418758455,
      "flos": 24204905360640.0,
      "grad_norm": 1.5255828600591865,
      "language_loss": 0.75312555,
      "learning_rate": 5.301767901772391e-09,
      "loss": 0.77431118,
      "num_input_tokens_seen": 350868955,
      "step": 16258,
      "time_per_iteration": 2.528136730194092
    },
    {
      "auxiliary_loss_clip": 0.01022085,
      "auxiliary_loss_mlp": 0.01002,
      "balance_loss_clip": 1.00937772,
      "balance_loss_mlp": 1.00096881,
      "epoch": 0.9775439651285135,
      "flos": 66357139829760.0,
      "grad_norm": 0.6775450540278893,
      "language_loss": 0.59749937,
      "learning_rate": 5.273466554344353e-09,
      "loss": 0.61774015,
      "num_input_tokens_seen": 350935110,
      "step": 16259,
      "time_per_iteration": 3.1597633361816406
    },
    {
      "auxiliary_loss_clip": 0.01087959,
      "auxiliary_loss_mlp": 0.01032944,
      "balance_loss_clip": 1.03558588,
      "balance_loss_mlp": 1.02028418,
      "epoch": 0.9776040883811814,
      "flos": 22601565976320.0,
      "grad_norm": 1.6369724513672157,
      "language_loss": 0.73420841,
      "learning_rate": 5.2452408471461705e-09,
      "loss": 0.75541747,
      "num_input_tokens_seen": 350953220,
      "step": 16260,
      "time_per_iteration": 3.926499128341675
    },
    {
      "auxiliary_loss_clip": 0.01092848,
      "auxiliary_loss_mlp": 0.01032182,
      "balance_loss_clip": 1.03422284,
      "balance_loss_mlp": 1.02018309,
      "epoch": 0.9776642116338494,
      "flos": 18442176825600.0,
      "grad_norm": 1.8423963563168593,
      "language_loss": 0.79787612,
      "learning_rate": 5.2170907812485456e-09,
      "loss": 0.81912637,
      "num_input_tokens_seen": 350971915,
      "step": 16261,
      "time_per_iteration": 2.466264009475708
    },
    {
      "auxiliary_loss_clip": 0.01095237,
      "auxiliary_loss_mlp": 0.01026963,
      "balance_loss_clip": 1.03496373,
      "balance_loss_mlp": 1.01456559,
      "epoch": 0.9777243348865173,
      "flos": 22638446265600.0,
      "grad_norm": 2.3029204704853465,
      "language_loss": 0.74233532,
      "learning_rate": 5.189016357718845e-09,
      "loss": 0.76355731,
      "num_input_tokens_seen": 350990470,
      "step": 16262,
      "time_per_iteration": 3.8842849731445312
    },
    {
      "auxiliary_loss_clip": 0.01095362,
      "auxiliary_loss_mlp": 0.01027574,
      "balance_loss_clip": 1.03614593,
      "balance_loss_mlp": 1.01474142,
      "epoch": 0.9777844581391854,
      "flos": 31321394605440.0,
      "grad_norm": 1.9858229864593178,
      "language_loss": 0.70018446,
      "learning_rate": 5.16101757762133e-09,
      "loss": 0.72141379,
      "num_input_tokens_seen": 351010755,
      "step": 16263,
      "time_per_iteration": 3.9354703426361084
    },
    {
      "auxiliary_loss_clip": 0.01094286,
      "auxiliary_loss_mlp": 0.01028535,
      "balance_loss_clip": 1.03546333,
      "balance_loss_mlp": 1.0168885,
      "epoch": 0.9778445813918533,
      "flos": 23039101543680.0,
      "grad_norm": 1.6565432065760262,
      "language_loss": 0.66074932,
      "learning_rate": 5.133094442018038e-09,
      "loss": 0.68197751,
      "num_input_tokens_seen": 351029965,
      "step": 16264,
      "time_per_iteration": 2.507380962371826
    },
    {
      "auxiliary_loss_clip": 0.01058319,
      "auxiliary_loss_mlp": 0.01027699,
      "balance_loss_clip": 1.0365746,
      "balance_loss_mlp": 1.0144434,
      "epoch": 0.9779047046445213,
      "flos": 17566351505280.0,
      "grad_norm": 1.8166731814322394,
      "language_loss": 0.73122817,
      "learning_rate": 5.105246951967679e-09,
      "loss": 0.75208843,
      "num_input_tokens_seen": 351046205,
      "step": 16265,
      "time_per_iteration": 2.5599944591522217
    },
    {
      "auxiliary_loss_clip": 0.01090949,
      "auxiliary_loss_mlp": 0.01031741,
      "balance_loss_clip": 1.03364062,
      "balance_loss_mlp": 1.01964676,
      "epoch": 0.9779648278971892,
      "flos": 20741141975040.0,
      "grad_norm": 1.7519255520092936,
      "language_loss": 0.68630755,
      "learning_rate": 5.077475108526297e-09,
      "loss": 0.70753443,
      "num_input_tokens_seen": 351065390,
      "step": 16266,
      "time_per_iteration": 2.504560947418213
    },
    {
      "auxiliary_loss_clip": 0.01055266,
      "auxiliary_loss_mlp": 0.01027624,
      "balance_loss_clip": 1.03170741,
      "balance_loss_mlp": 1.01663876,
      "epoch": 0.9780249511498572,
      "flos": 21026954494080.0,
      "grad_norm": 1.6988029094706205,
      "language_loss": 0.86898011,
      "learning_rate": 5.049778912747049e-09,
      "loss": 0.88980901,
      "num_input_tokens_seen": 351084355,
      "step": 16267,
      "time_per_iteration": 2.5580835342407227
    },
    {
      "auxiliary_loss_clip": 0.0103842,
      "auxiliary_loss_mlp": 0.01025238,
      "balance_loss_clip": 1.03213668,
      "balance_loss_mlp": 1.01285791,
      "epoch": 0.9780850744025251,
      "flos": 30774223751040.0,
      "grad_norm": 1.7941591570982727,
      "language_loss": 0.7030713,
      "learning_rate": 5.022158365679985e-09,
      "loss": 0.72370791,
      "num_input_tokens_seen": 351105870,
      "step": 16268,
      "time_per_iteration": 2.823263645172119
    },
    {
      "auxiliary_loss_clip": 0.01081348,
      "auxiliary_loss_mlp": 0.01026897,
      "balance_loss_clip": 1.03238034,
      "balance_loss_mlp": 1.01531005,
      "epoch": 0.9781451976551931,
      "flos": 20302995876480.0,
      "grad_norm": 1.6428766100074628,
      "language_loss": 0.73847181,
      "learning_rate": 4.994613468372711e-09,
      "loss": 0.75955427,
      "num_input_tokens_seen": 351124760,
      "step": 16269,
      "time_per_iteration": 2.9180221557617188
    },
    {
      "auxiliary_loss_clip": 0.01079976,
      "auxiliary_loss_mlp": 0.01031456,
      "balance_loss_clip": 1.03422761,
      "balance_loss_mlp": 1.01769936,
      "epoch": 0.9782053209078612,
      "flos": 24316479982080.0,
      "grad_norm": 1.7891978858489963,
      "language_loss": 0.71052825,
      "learning_rate": 4.967144221869501e-09,
      "loss": 0.7316426,
      "num_input_tokens_seen": 351142820,
      "step": 16270,
      "time_per_iteration": 3.969299793243408
    },
    {
      "auxiliary_loss_clip": 0.01105279,
      "auxiliary_loss_mlp": 0.01034187,
      "balance_loss_clip": 1.03597367,
      "balance_loss_mlp": 1.02198017,
      "epoch": 0.9782654441605291,
      "flos": 32489425065600.0,
      "grad_norm": 1.6884069534369714,
      "language_loss": 0.64061701,
      "learning_rate": 4.939750627212191e-09,
      "loss": 0.66201168,
      "num_input_tokens_seen": 351164805,
      "step": 16271,
      "time_per_iteration": 2.5651535987854004
    },
    {
      "auxiliary_loss_clip": 0.010777,
      "auxiliary_loss_mlp": 0.01034197,
      "balance_loss_clip": 1.03552699,
      "balance_loss_mlp": 1.02190018,
      "epoch": 0.9783255674131971,
      "flos": 26979076465920.0,
      "grad_norm": 1.4102331923360485,
      "language_loss": 0.70429349,
      "learning_rate": 4.912432685439505e-09,
      "loss": 0.72541249,
      "num_input_tokens_seen": 351187005,
      "step": 16272,
      "time_per_iteration": 2.5497801303863525
    },
    {
      "auxiliary_loss_clip": 0.01047096,
      "auxiliary_loss_mlp": 0.01030238,
      "balance_loss_clip": 1.03676915,
      "balance_loss_mlp": 1.01806688,
      "epoch": 0.978385690665865,
      "flos": 23112251591040.0,
      "grad_norm": 2.161180634388928,
      "language_loss": 0.66537404,
      "learning_rate": 4.88519039758728e-09,
      "loss": 0.68614745,
      "num_input_tokens_seen": 351208450,
      "step": 16273,
      "time_per_iteration": 2.663785457611084
    },
    {
      "auxiliary_loss_clip": 0.01073192,
      "auxiliary_loss_mlp": 0.01024952,
      "balance_loss_clip": 1.03575635,
      "balance_loss_mlp": 1.01264977,
      "epoch": 0.978445813918533,
      "flos": 25409672455680.0,
      "grad_norm": 1.6063525760313986,
      "language_loss": 0.741238,
      "learning_rate": 4.85802376468869e-09,
      "loss": 0.76221943,
      "num_input_tokens_seen": 351229585,
      "step": 16274,
      "time_per_iteration": 2.5975968837738037
    },
    {
      "auxiliary_loss_clip": 0.01078831,
      "auxiliary_loss_mlp": 0.01033954,
      "balance_loss_clip": 1.0342586,
      "balance_loss_mlp": 1.02247977,
      "epoch": 0.9785059371712009,
      "flos": 23550218121600.0,
      "grad_norm": 1.5167373764830245,
      "language_loss": 0.77784091,
      "learning_rate": 4.830932787773579e-09,
      "loss": 0.79896873,
      "num_input_tokens_seen": 351249525,
      "step": 16275,
      "time_per_iteration": 2.572425127029419
    },
    {
      "auxiliary_loss_clip": 0.01040041,
      "auxiliary_loss_mlp": 0.01030807,
      "balance_loss_clip": 1.03403807,
      "balance_loss_mlp": 1.0182538,
      "epoch": 0.978566060423869,
      "flos": 34351177870080.0,
      "grad_norm": 1.5385263873240587,
      "language_loss": 0.70860839,
      "learning_rate": 4.803917467869567e-09,
      "loss": 0.72931695,
      "num_input_tokens_seen": 351272530,
      "step": 16276,
      "time_per_iteration": 2.833160400390625
    },
    {
      "auxiliary_loss_clip": 0.01071833,
      "auxiliary_loss_mlp": 0.01029421,
      "balance_loss_clip": 1.03064191,
      "balance_loss_mlp": 1.01754165,
      "epoch": 0.9786261836765369,
      "flos": 11618862387840.0,
      "grad_norm": 1.7182551875951353,
      "language_loss": 0.85659516,
      "learning_rate": 4.776977806000726e-09,
      "loss": 0.8776077,
      "num_input_tokens_seen": 351288530,
      "step": 16277,
      "time_per_iteration": 2.498908042907715
    },
    {
      "auxiliary_loss_clip": 0.01085493,
      "auxiliary_loss_mlp": 0.01027961,
      "balance_loss_clip": 1.03284693,
      "balance_loss_mlp": 1.01559329,
      "epoch": 0.9786863069292049,
      "flos": 17420949250560.0,
      "grad_norm": 1.6938799610416937,
      "language_loss": 0.70864826,
      "learning_rate": 4.7501138031891264e-09,
      "loss": 0.72978282,
      "num_input_tokens_seen": 351305890,
      "step": 16278,
      "time_per_iteration": 2.47751784324646
    },
    {
      "auxiliary_loss_clip": 0.0109139,
      "auxiliary_loss_mlp": 0.01028374,
      "balance_loss_clip": 1.03447986,
      "balance_loss_mlp": 1.01610756,
      "epoch": 0.9787464301818728,
      "flos": 20844923345280.0,
      "grad_norm": 3.2733739991378266,
      "language_loss": 0.84162128,
      "learning_rate": 4.723325460453065e-09,
      "loss": 0.86281902,
      "num_input_tokens_seen": 351325010,
      "step": 16279,
      "time_per_iteration": 2.509918689727783
    },
    {
      "auxiliary_loss_clip": 0.01092115,
      "auxiliary_loss_mlp": 0.0103046,
      "balance_loss_clip": 1.03295267,
      "balance_loss_mlp": 1.01779437,
      "epoch": 0.9788065534345408,
      "flos": 18222942165120.0,
      "grad_norm": 6.749540849732605,
      "language_loss": 0.7919724,
      "learning_rate": 4.696612778808395e-09,
      "loss": 0.81319815,
      "num_input_tokens_seen": 351343060,
      "step": 16280,
      "time_per_iteration": 2.456779956817627
    },
    {
      "auxiliary_loss_clip": 0.01067759,
      "auxiliary_loss_mlp": 0.01034755,
      "balance_loss_clip": 1.03457379,
      "balance_loss_mlp": 1.02400827,
      "epoch": 0.9788666766872087,
      "flos": 21578219498880.0,
      "grad_norm": 1.696310832410057,
      "language_loss": 0.79671878,
      "learning_rate": 4.669975759268085e-09,
      "loss": 0.8177439,
      "num_input_tokens_seen": 351363260,
      "step": 16281,
      "time_per_iteration": 2.5739331245422363
    },
    {
      "auxiliary_loss_clip": 0.01089589,
      "auxiliary_loss_mlp": 0.01031061,
      "balance_loss_clip": 1.03506851,
      "balance_loss_mlp": 1.01840675,
      "epoch": 0.9789267999398767,
      "flos": 24900495212160.0,
      "grad_norm": 1.6595047925338953,
      "language_loss": 0.80319178,
      "learning_rate": 4.643414402842216e-09,
      "loss": 0.82439828,
      "num_input_tokens_seen": 351382610,
      "step": 16282,
      "time_per_iteration": 2.50358510017395
    },
    {
      "auxiliary_loss_clip": 0.01081275,
      "auxiliary_loss_mlp": 0.01036864,
      "balance_loss_clip": 1.03350759,
      "balance_loss_mlp": 1.02493119,
      "epoch": 0.9789869231925448,
      "flos": 19573111514880.0,
      "grad_norm": 2.0609193741732263,
      "language_loss": 0.83304763,
      "learning_rate": 4.616928710538204e-09,
      "loss": 0.85422897,
      "num_input_tokens_seen": 351401075,
      "step": 16283,
      "time_per_iteration": 2.520723342895508
    },
    {
      "auxiliary_loss_clip": 0.01087754,
      "auxiliary_loss_mlp": 0.01030928,
      "balance_loss_clip": 1.03536701,
      "balance_loss_mlp": 1.01914418,
      "epoch": 0.9790470464452127,
      "flos": 16796641939200.0,
      "grad_norm": 1.828113114070613,
      "language_loss": 0.72176898,
      "learning_rate": 4.590518683360134e-09,
      "loss": 0.7429558,
      "num_input_tokens_seen": 351419275,
      "step": 16284,
      "time_per_iteration": 2.502812147140503
    },
    {
      "auxiliary_loss_clip": 0.01091322,
      "auxiliary_loss_mlp": 0.01035049,
      "balance_loss_clip": 1.03552961,
      "balance_loss_mlp": 1.02388525,
      "epoch": 0.9791071696978807,
      "flos": 18369350000640.0,
      "grad_norm": 1.9069354358165265,
      "language_loss": 0.64489257,
      "learning_rate": 4.56418432230965e-09,
      "loss": 0.66615629,
      "num_input_tokens_seen": 351437375,
      "step": 16285,
      "time_per_iteration": 2.481660842895508
    },
    {
      "auxiliary_loss_clip": 0.01078262,
      "auxiliary_loss_mlp": 0.01030603,
      "balance_loss_clip": 1.03579056,
      "balance_loss_mlp": 1.01911116,
      "epoch": 0.9791672929505486,
      "flos": 24170323541760.0,
      "grad_norm": 1.65269952083964,
      "language_loss": 0.70676941,
      "learning_rate": 4.537925628385286e-09,
      "loss": 0.72785801,
      "num_input_tokens_seen": 351457810,
      "step": 16286,
      "time_per_iteration": 2.528374195098877
    },
    {
      "auxiliary_loss_clip": 0.01085007,
      "auxiliary_loss_mlp": 0.01029544,
      "balance_loss_clip": 1.03315747,
      "balance_loss_mlp": 1.01780796,
      "epoch": 0.9792274162032166,
      "flos": 24354114456960.0,
      "grad_norm": 1.3260794439643635,
      "language_loss": 0.58034068,
      "learning_rate": 4.511742602582691e-09,
      "loss": 0.60148615,
      "num_input_tokens_seen": 351478825,
      "step": 16287,
      "time_per_iteration": 2.5230419635772705
    },
    {
      "auxiliary_loss_clip": 0.01092053,
      "auxiliary_loss_mlp": 0.01034824,
      "balance_loss_clip": 1.03491867,
      "balance_loss_mlp": 1.02258682,
      "epoch": 0.9792875394558845,
      "flos": 26395779507840.0,
      "grad_norm": 2.1375926696746523,
      "language_loss": 0.81760538,
      "learning_rate": 4.485635245894626e-09,
      "loss": 0.83887422,
      "num_input_tokens_seen": 351498785,
      "step": 16288,
      "time_per_iteration": 2.5043258666992188
    },
    {
      "auxiliary_loss_clip": 0.01077978,
      "auxiliary_loss_mlp": 0.00782891,
      "balance_loss_clip": 1.0334934,
      "balance_loss_mlp": 1.00938964,
      "epoch": 0.9793476627085526,
      "flos": 28148004766080.0,
      "grad_norm": 1.582569030056652,
      "language_loss": 0.71470737,
      "learning_rate": 4.459603559311631e-09,
      "loss": 0.733316,
      "num_input_tokens_seen": 351520235,
      "step": 16289,
      "time_per_iteration": 2.572805881500244
    },
    {
      "auxiliary_loss_clip": 0.01068038,
      "auxiliary_loss_mlp": 0.01033322,
      "balance_loss_clip": 1.0359441,
      "balance_loss_mlp": 1.02146626,
      "epoch": 0.9794077859612205,
      "flos": 16763927627520.0,
      "grad_norm": 2.317553890869716,
      "language_loss": 0.75786638,
      "learning_rate": 4.43364754382003e-09,
      "loss": 0.77888,
      "num_input_tokens_seen": 351538900,
      "step": 16290,
      "time_per_iteration": 2.4999938011169434
    },
    {
      "auxiliary_loss_clip": 0.01093533,
      "auxiliary_loss_mlp": 0.01032201,
      "balance_loss_clip": 1.03489304,
      "balance_loss_mlp": 1.0189693,
      "epoch": 0.9794679092138885,
      "flos": 19280834547840.0,
      "grad_norm": 1.709490148723921,
      "language_loss": 0.67132944,
      "learning_rate": 4.4077672004048105e-09,
      "loss": 0.69258678,
      "num_input_tokens_seen": 351558715,
      "step": 16291,
      "time_per_iteration": 2.4954793453216553
    },
    {
      "auxiliary_loss_clip": 0.01106499,
      "auxiliary_loss_mlp": 0.00783904,
      "balance_loss_clip": 1.03490496,
      "balance_loss_mlp": 1.00864077,
      "epoch": 0.9795280324665564,
      "flos": 32156640535680.0,
      "grad_norm": 1.6105768614772433,
      "language_loss": 0.62772226,
      "learning_rate": 4.3819625300467456e-09,
      "loss": 0.64662635,
      "num_input_tokens_seen": 351578450,
      "step": 16292,
      "time_per_iteration": 2.529905080795288
    },
    {
      "auxiliary_loss_clip": 0.01069646,
      "auxiliary_loss_mlp": 0.01031742,
      "balance_loss_clip": 1.03496099,
      "balance_loss_mlp": 1.02003002,
      "epoch": 0.9795881557192244,
      "flos": 19060953442560.0,
      "grad_norm": 6.046983533586835,
      "language_loss": 0.73671758,
      "learning_rate": 4.356233533724829e-09,
      "loss": 0.75773144,
      "num_input_tokens_seen": 351597195,
      "step": 16293,
      "time_per_iteration": 2.5624260902404785
    },
    {
      "auxiliary_loss_clip": 0.01095151,
      "auxiliary_loss_mlp": 0.01028883,
      "balance_loss_clip": 1.03357196,
      "balance_loss_mlp": 1.01643705,
      "epoch": 0.9796482789718923,
      "flos": 28329928174080.0,
      "grad_norm": 2.0568146472039968,
      "language_loss": 0.83959502,
      "learning_rate": 4.330580212414503e-09,
      "loss": 0.86083537,
      "num_input_tokens_seen": 351617460,
      "step": 16294,
      "time_per_iteration": 2.5159389972686768
    },
    {
      "auxiliary_loss_clip": 0.01066423,
      "auxiliary_loss_mlp": 0.01031347,
      "balance_loss_clip": 1.0319978,
      "balance_loss_mlp": 1.01992047,
      "epoch": 0.9797084022245603,
      "flos": 17967976450560.0,
      "grad_norm": 1.8990128698314028,
      "language_loss": 0.71732116,
      "learning_rate": 4.305002567088767e-09,
      "loss": 0.73829877,
      "num_input_tokens_seen": 351635900,
      "step": 16295,
      "time_per_iteration": 2.52286434173584
    },
    {
      "auxiliary_loss_clip": 0.01098147,
      "auxiliary_loss_mlp": 0.0103679,
      "balance_loss_clip": 1.03649509,
      "balance_loss_mlp": 1.02461338,
      "epoch": 0.9797685254772284,
      "flos": 20266726118400.0,
      "grad_norm": 1.638820665537276,
      "language_loss": 0.80451328,
      "learning_rate": 4.2795005987170674e-09,
      "loss": 0.82586271,
      "num_input_tokens_seen": 351655400,
      "step": 16296,
      "time_per_iteration": 2.4674484729766846
    },
    {
      "auxiliary_loss_clip": 0.01076939,
      "auxiliary_loss_mlp": 0.01034995,
      "balance_loss_clip": 1.03420436,
      "balance_loss_mlp": 1.02313948,
      "epoch": 0.9798286487298963,
      "flos": 26907147480960.0,
      "grad_norm": 1.9031795854675597,
      "language_loss": 0.75614059,
      "learning_rate": 4.254074308266853e-09,
      "loss": 0.77725995,
      "num_input_tokens_seen": 351675505,
      "step": 16297,
      "time_per_iteration": 2.5529961585998535
    },
    {
      "auxiliary_loss_clip": 0.0109472,
      "auxiliary_loss_mlp": 0.01031953,
      "balance_loss_clip": 1.03453219,
      "balance_loss_mlp": 1.01993084,
      "epoch": 0.9798887719825643,
      "flos": 27161071701120.0,
      "grad_norm": 1.4964263960004178,
      "language_loss": 0.78181493,
      "learning_rate": 4.228723696702019e-09,
      "loss": 0.80308163,
      "num_input_tokens_seen": 351697920,
      "step": 16298,
      "time_per_iteration": 2.536107063293457
    },
    {
      "auxiliary_loss_clip": 0.01088866,
      "auxiliary_loss_mlp": 0.01026562,
      "balance_loss_clip": 1.03357041,
      "balance_loss_mlp": 1.01493907,
      "epoch": 0.9799488952352322,
      "flos": 20668422890880.0,
      "grad_norm": 1.5203437042227115,
      "language_loss": 0.72401571,
      "learning_rate": 4.203448764984019e-09,
      "loss": 0.74517,
      "num_input_tokens_seen": 351717615,
      "step": 16299,
      "time_per_iteration": 3.8942174911499023
    },
    {
      "auxiliary_loss_clip": 0.01080182,
      "auxiliary_loss_mlp": 0.01029006,
      "balance_loss_clip": 1.03337932,
      "balance_loss_mlp": 1.01619101,
      "epoch": 0.9800090184879002,
      "flos": 21981209160960.0,
      "grad_norm": 2.303369486809455,
      "language_loss": 0.88887405,
      "learning_rate": 4.178249514071419e-09,
      "loss": 0.90996599,
      "num_input_tokens_seen": 351735260,
      "step": 16300,
      "time_per_iteration": 2.5311107635498047
    },
    {
      "auxiliary_loss_clip": 0.01094471,
      "auxiliary_loss_mlp": 0.01028792,
      "balance_loss_clip": 1.03416157,
      "balance_loss_mlp": 1.01648939,
      "epoch": 0.9800691417405681,
      "flos": 21288420570240.0,
      "grad_norm": 2.2679382180632,
      "language_loss": 0.78186989,
      "learning_rate": 4.1531259449194555e-09,
      "loss": 0.80310249,
      "num_input_tokens_seen": 351755800,
      "step": 16301,
      "time_per_iteration": 5.292080640792847
    },
    {
      "auxiliary_loss_clip": 0.01081988,
      "auxiliary_loss_mlp": 0.01034223,
      "balance_loss_clip": 1.03390443,
      "balance_loss_mlp": 1.02192044,
      "epoch": 0.9801292649932362,
      "flos": 18439878355200.0,
      "grad_norm": 2.1171939736852177,
      "language_loss": 0.75019646,
      "learning_rate": 4.128078058480921e-09,
      "loss": 0.77135861,
      "num_input_tokens_seen": 351774790,
      "step": 16302,
      "time_per_iteration": 2.499150276184082
    },
    {
      "auxiliary_loss_clip": 0.01082424,
      "auxiliary_loss_mlp": 0.01029004,
      "balance_loss_clip": 1.03526616,
      "balance_loss_mlp": 1.01692784,
      "epoch": 0.9801893882459041,
      "flos": 25046364343680.0,
      "grad_norm": 1.622858123935546,
      "language_loss": 0.79719305,
      "learning_rate": 4.103105855705724e-09,
      "loss": 0.81830728,
      "num_input_tokens_seen": 351792855,
      "step": 16303,
      "time_per_iteration": 2.5940306186676025
    },
    {
      "auxiliary_loss_clip": 0.01062929,
      "auxiliary_loss_mlp": 0.01036127,
      "balance_loss_clip": 1.03413153,
      "balance_loss_mlp": 1.02296007,
      "epoch": 0.9802495114985721,
      "flos": 18511484117760.0,
      "grad_norm": 2.203373991588886,
      "language_loss": 0.83669484,
      "learning_rate": 4.078209337540883e-09,
      "loss": 0.85768539,
      "num_input_tokens_seen": 351811450,
      "step": 16304,
      "time_per_iteration": 2.547268867492676
    },
    {
      "auxiliary_loss_clip": 0.01067265,
      "auxiliary_loss_mlp": 0.01026777,
      "balance_loss_clip": 1.0354706,
      "balance_loss_mlp": 1.01605964,
      "epoch": 0.98030963475124,
      "flos": 21469841187840.0,
      "grad_norm": 1.8524709146713345,
      "language_loss": 0.70073617,
      "learning_rate": 4.053388504930089e-09,
      "loss": 0.72167659,
      "num_input_tokens_seen": 351831960,
      "step": 16305,
      "time_per_iteration": 2.5799946784973145
    },
    {
      "auxiliary_loss_clip": 0.01074038,
      "auxiliary_loss_mlp": 0.01039099,
      "balance_loss_clip": 1.03435564,
      "balance_loss_mlp": 1.0253185,
      "epoch": 0.980369758003908,
      "flos": 20412272027520.0,
      "grad_norm": 2.803386974399174,
      "language_loss": 0.71798027,
      "learning_rate": 4.028643358815032e-09,
      "loss": 0.73911166,
      "num_input_tokens_seen": 351851585,
      "step": 16306,
      "time_per_iteration": 2.4845826625823975
    },
    {
      "auxiliary_loss_clip": 0.01077166,
      "auxiliary_loss_mlp": 0.01030922,
      "balance_loss_clip": 1.03250313,
      "balance_loss_mlp": 1.0196147,
      "epoch": 0.9804298812565759,
      "flos": 23399177431680.0,
      "grad_norm": 1.5761913448996532,
      "language_loss": 0.73537695,
      "learning_rate": 4.00397390013385e-09,
      "loss": 0.75645781,
      "num_input_tokens_seen": 351871085,
      "step": 16307,
      "time_per_iteration": 2.5613269805908203
    },
    {
      "auxiliary_loss_clip": 0.01070079,
      "auxiliary_loss_mlp": 0.01027762,
      "balance_loss_clip": 1.03469324,
      "balance_loss_mlp": 1.01694989,
      "epoch": 0.980490004509244,
      "flos": 23292666627840.0,
      "grad_norm": 1.4860832458444164,
      "language_loss": 0.74857283,
      "learning_rate": 3.979380129822018e-09,
      "loss": 0.76955116,
      "num_input_tokens_seen": 351891775,
      "step": 16308,
      "time_per_iteration": 2.561248540878296
    },
    {
      "auxiliary_loss_clip": 0.01009049,
      "auxiliary_loss_mlp": 0.01003206,
      "balance_loss_clip": 1.00647902,
      "balance_loss_mlp": 1.00202584,
      "epoch": 0.980550127761912,
      "flos": 56051027798400.0,
      "grad_norm": 0.7608720995556525,
      "language_loss": 0.57945073,
      "learning_rate": 3.954862048811902e-09,
      "loss": 0.59957325,
      "num_input_tokens_seen": 351946770,
      "step": 16309,
      "time_per_iteration": 4.349445104598999
    },
    {
      "auxiliary_loss_clip": 0.01060394,
      "auxiliary_loss_mlp": 0.01031014,
      "balance_loss_clip": 1.03283381,
      "balance_loss_mlp": 1.01904523,
      "epoch": 0.9806102510145799,
      "flos": 25333290184320.0,
      "grad_norm": 1.73916753614286,
      "language_loss": 0.6592654,
      "learning_rate": 3.930419658033646e-09,
      "loss": 0.68017948,
      "num_input_tokens_seen": 351966155,
      "step": 16310,
      "time_per_iteration": 2.650777816772461
    },
    {
      "auxiliary_loss_clip": 0.01008368,
      "auxiliary_loss_mlp": 0.01001373,
      "balance_loss_clip": 1.00904787,
      "balance_loss_mlp": 1.00027585,
      "epoch": 0.9806703742672479,
      "flos": 67274837429760.0,
      "grad_norm": 0.8448863448804085,
      "language_loss": 0.54595488,
      "learning_rate": 3.906052958413841e-09,
      "loss": 0.56605232,
      "num_input_tokens_seen": 352031655,
      "step": 16311,
      "time_per_iteration": 3.2005562782287598
    },
    {
      "auxiliary_loss_clip": 0.01091398,
      "auxiliary_loss_mlp": 0.01024408,
      "balance_loss_clip": 1.03373551,
      "balance_loss_mlp": 1.01290452,
      "epoch": 0.9807304975199158,
      "flos": 25228970110080.0,
      "grad_norm": 1.4580688029076263,
      "language_loss": 0.79952979,
      "learning_rate": 3.881761950876638e-09,
      "loss": 0.82068789,
      "num_input_tokens_seen": 352051920,
      "step": 16312,
      "time_per_iteration": 2.5342788696289062
    },
    {
      "auxiliary_loss_clip": 0.01080246,
      "auxiliary_loss_mlp": 0.01027635,
      "balance_loss_clip": 1.03433418,
      "balance_loss_mlp": 1.01621461,
      "epoch": 0.9807906207725838,
      "flos": 17456392995840.0,
      "grad_norm": 1.710625280744429,
      "language_loss": 0.63489783,
      "learning_rate": 3.8575466363430785e-09,
      "loss": 0.65597665,
      "num_input_tokens_seen": 352069315,
      "step": 16313,
      "time_per_iteration": 2.471071243286133
    },
    {
      "auxiliary_loss_clip": 0.01093018,
      "auxiliary_loss_mlp": 0.01029195,
      "balance_loss_clip": 1.03602517,
      "balance_loss_mlp": 1.01717854,
      "epoch": 0.9808507440252517,
      "flos": 21032413361280.0,
      "grad_norm": 5.044553281999545,
      "language_loss": 0.7218833,
      "learning_rate": 3.833407015731316e-09,
      "loss": 0.74310541,
      "num_input_tokens_seen": 352089480,
      "step": 16314,
      "time_per_iteration": 2.490083694458008
    },
    {
      "auxiliary_loss_clip": 0.01000441,
      "auxiliary_loss_mlp": 0.01003961,
      "balance_loss_clip": 1.00755525,
      "balance_loss_mlp": 1.00280464,
      "epoch": 0.9809108672779198,
      "flos": 64044491598720.0,
      "grad_norm": 0.6894059505380258,
      "language_loss": 0.51710522,
      "learning_rate": 3.80934308995684e-09,
      "loss": 0.53714919,
      "num_input_tokens_seen": 352150000,
      "step": 16315,
      "time_per_iteration": 3.1547727584838867
    },
    {
      "auxiliary_loss_clip": 0.01090823,
      "auxiliary_loss_mlp": 0.01028172,
      "balance_loss_clip": 1.03272891,
      "balance_loss_mlp": 1.01692426,
      "epoch": 0.9809709905305877,
      "flos": 22780616296320.0,
      "grad_norm": 1.414290347012709,
      "language_loss": 0.69671547,
      "learning_rate": 3.785354859932033e-09,
      "loss": 0.7179054,
      "num_input_tokens_seen": 352170990,
      "step": 16316,
      "time_per_iteration": 2.4956912994384766
    },
    {
      "auxiliary_loss_clip": 0.01102788,
      "auxiliary_loss_mlp": 0.01029122,
      "balance_loss_clip": 1.03359842,
      "balance_loss_mlp": 1.01733804,
      "epoch": 0.9810311137832557,
      "flos": 37013415217920.0,
      "grad_norm": 1.7578885156152424,
      "language_loss": 0.55404592,
      "learning_rate": 3.76144232656661e-09,
      "loss": 0.57536495,
      "num_input_tokens_seen": 352195335,
      "step": 16317,
      "time_per_iteration": 2.607006072998047
    },
    {
      "auxiliary_loss_clip": 0.01049731,
      "auxiliary_loss_mlp": 0.01038207,
      "balance_loss_clip": 1.03066683,
      "balance_loss_mlp": 1.02600646,
      "epoch": 0.9810912370359236,
      "flos": 18916305373440.0,
      "grad_norm": 1.5718784049131997,
      "language_loss": 0.73009551,
      "learning_rate": 3.737605490767404e-09,
      "loss": 0.75097489,
      "num_input_tokens_seen": 352214170,
      "step": 16318,
      "time_per_iteration": 2.5882198810577393
    },
    {
      "auxiliary_loss_clip": 0.01078111,
      "auxiliary_loss_mlp": 0.0102656,
      "balance_loss_clip": 1.03270006,
      "balance_loss_mlp": 1.01533604,
      "epoch": 0.9811513602885916,
      "flos": 18441602208000.0,
      "grad_norm": 2.112847916258105,
      "language_loss": 0.82266569,
      "learning_rate": 3.7138443534383555e-09,
      "loss": 0.84371245,
      "num_input_tokens_seen": 352231470,
      "step": 16319,
      "time_per_iteration": 2.4891841411590576
    },
    {
      "auxiliary_loss_clip": 0.01020369,
      "auxiliary_loss_mlp": 0.01011872,
      "balance_loss_clip": 1.00786138,
      "balance_loss_mlp": 1.01046491,
      "epoch": 0.9812114835412595,
      "flos": 68058945371520.0,
      "grad_norm": 0.7240579239361407,
      "language_loss": 0.53562981,
      "learning_rate": 3.6901589154803014e-09,
      "loss": 0.55595225,
      "num_input_tokens_seen": 352291770,
      "step": 16320,
      "time_per_iteration": 2.976208209991455
    },
    {
      "auxiliary_loss_clip": 0.01054072,
      "auxiliary_loss_mlp": 0.01032387,
      "balance_loss_clip": 1.03273177,
      "balance_loss_mlp": 1.02023387,
      "epoch": 0.9812716067939276,
      "flos": 25373007648000.0,
      "grad_norm": 1.8183424686599858,
      "language_loss": 0.73495209,
      "learning_rate": 3.6665491777914116e-09,
      "loss": 0.7558167,
      "num_input_tokens_seen": 352310735,
      "step": 16321,
      "time_per_iteration": 2.623300790786743
    },
    {
      "auxiliary_loss_clip": 0.0108253,
      "auxiliary_loss_mlp": 0.01032785,
      "balance_loss_clip": 1.03596878,
      "balance_loss_mlp": 1.02072096,
      "epoch": 0.9813317300465956,
      "flos": 22856818999680.0,
      "grad_norm": 1.860305761239585,
      "language_loss": 0.78628796,
      "learning_rate": 3.6430151412669698e-09,
      "loss": 0.80744106,
      "num_input_tokens_seen": 352329545,
      "step": 16322,
      "time_per_iteration": 2.5481505393981934
    },
    {
      "auxiliary_loss_clip": 0.0109138,
      "auxiliary_loss_mlp": 0.01032901,
      "balance_loss_clip": 1.03441119,
      "balance_loss_mlp": 1.0209918,
      "epoch": 0.9813918532992635,
      "flos": 23586954756480.0,
      "grad_norm": 1.493214467037949,
      "language_loss": 0.80667365,
      "learning_rate": 3.619556806799595e-09,
      "loss": 0.82791638,
      "num_input_tokens_seen": 352352080,
      "step": 16323,
      "time_per_iteration": 2.535888195037842
    },
    {
      "auxiliary_loss_clip": 0.01105487,
      "auxiliary_loss_mlp": 0.01031783,
      "balance_loss_clip": 1.03550744,
      "balance_loss_mlp": 1.02031469,
      "epoch": 0.9814519765519315,
      "flos": 19606328616960.0,
      "grad_norm": 2.0154981229982707,
      "language_loss": 0.84284937,
      "learning_rate": 3.596174175278799e-09,
      "loss": 0.86422211,
      "num_input_tokens_seen": 352366455,
      "step": 16324,
      "time_per_iteration": 2.4116899967193604
    },
    {
      "auxiliary_loss_clip": 0.01080182,
      "auxiliary_loss_mlp": 0.0102987,
      "balance_loss_clip": 1.03481841,
      "balance_loss_mlp": 1.01786542,
      "epoch": 0.9815120998045994,
      "flos": 33946284787200.0,
      "grad_norm": 2.5790910065856942,
      "language_loss": 0.74333262,
      "learning_rate": 3.5728672475909827e-09,
      "loss": 0.76443315,
      "num_input_tokens_seen": 352386090,
      "step": 16325,
      "time_per_iteration": 2.6507415771484375
    },
    {
      "auxiliary_loss_clip": 0.01050923,
      "auxiliary_loss_mlp": 0.01030962,
      "balance_loss_clip": 1.03487325,
      "balance_loss_mlp": 1.02013743,
      "epoch": 0.9815722230572674,
      "flos": 20850023076480.0,
      "grad_norm": 2.0448152533604955,
      "language_loss": 0.76442111,
      "learning_rate": 3.5496360246201063e-09,
      "loss": 0.78523999,
      "num_input_tokens_seen": 352404000,
      "step": 16326,
      "time_per_iteration": 2.563978672027588
    },
    {
      "auxiliary_loss_clip": 0.01070842,
      "auxiliary_loss_mlp": 0.01028423,
      "balance_loss_clip": 1.03474164,
      "balance_loss_mlp": 1.01635289,
      "epoch": 0.9816323463099353,
      "flos": 22894525301760.0,
      "grad_norm": 1.6807064560241587,
      "language_loss": 0.6761241,
      "learning_rate": 3.5264805072470205e-09,
      "loss": 0.69711673,
      "num_input_tokens_seen": 352423540,
      "step": 16327,
      "time_per_iteration": 2.5692334175109863
    },
    {
      "auxiliary_loss_clip": 0.01096336,
      "auxiliary_loss_mlp": 0.01038853,
      "balance_loss_clip": 1.03425467,
      "balance_loss_mlp": 1.02550209,
      "epoch": 0.9816924695626034,
      "flos": 31539444117120.0,
      "grad_norm": 1.5713057705287012,
      "language_loss": 0.73775464,
      "learning_rate": 3.5034006963501337e-09,
      "loss": 0.75910652,
      "num_input_tokens_seen": 352445530,
      "step": 16328,
      "time_per_iteration": 2.5366382598876953
    },
    {
      "auxiliary_loss_clip": 0.01088053,
      "auxiliary_loss_mlp": 0.01038423,
      "balance_loss_clip": 1.03500605,
      "balance_loss_mlp": 1.02522099,
      "epoch": 0.9817525928152713,
      "flos": 21506901045120.0,
      "grad_norm": 1.7957225333662308,
      "language_loss": 0.80941272,
      "learning_rate": 3.4803965928040802e-09,
      "loss": 0.83067751,
      "num_input_tokens_seen": 352466325,
      "step": 16329,
      "time_per_iteration": 2.560040235519409
    },
    {
      "auxiliary_loss_clip": 0.01106808,
      "auxiliary_loss_mlp": 0.01029979,
      "balance_loss_clip": 1.0346967,
      "balance_loss_mlp": 1.01702046,
      "epoch": 0.9818127160679393,
      "flos": 25550513683200.0,
      "grad_norm": 2.2637304586965192,
      "language_loss": 0.76410383,
      "learning_rate": 3.4574681974817168e-09,
      "loss": 0.78547168,
      "num_input_tokens_seen": 352485505,
      "step": 16330,
      "time_per_iteration": 2.4795854091644287
    },
    {
      "auxiliary_loss_clip": 0.01111558,
      "auxiliary_loss_mlp": 0.01032036,
      "balance_loss_clip": 1.0358777,
      "balance_loss_mlp": 1.01759398,
      "epoch": 0.9818728393206072,
      "flos": 28803661672320.0,
      "grad_norm": 2.9680364312606335,
      "language_loss": 0.66343927,
      "learning_rate": 3.434615511252126e-09,
      "loss": 0.68487525,
      "num_input_tokens_seen": 352505360,
      "step": 16331,
      "time_per_iteration": 2.5004591941833496
    },
    {
      "auxiliary_loss_clip": 0.01090548,
      "auxiliary_loss_mlp": 0.01030232,
      "balance_loss_clip": 1.03322697,
      "balance_loss_mlp": 1.0185734,
      "epoch": 0.9819329625732752,
      "flos": 23222246014080.0,
      "grad_norm": 1.7052312319395275,
      "language_loss": 0.73114979,
      "learning_rate": 3.411838534981948e-09,
      "loss": 0.75235766,
      "num_input_tokens_seen": 352524035,
      "step": 16332,
      "time_per_iteration": 2.4848275184631348
    },
    {
      "auxiliary_loss_clip": 0.01091247,
      "auxiliary_loss_mlp": 0.01027528,
      "balance_loss_clip": 1.03524804,
      "balance_loss_mlp": 1.01665568,
      "epoch": 0.9819930858259431,
      "flos": 17530440883200.0,
      "grad_norm": 1.7546547347606805,
      "language_loss": 0.7715168,
      "learning_rate": 3.389137269534936e-09,
      "loss": 0.79270452,
      "num_input_tokens_seen": 352543210,
      "step": 16333,
      "time_per_iteration": 2.476016044616699
    },
    {
      "auxiliary_loss_clip": 0.010847,
      "auxiliary_loss_mlp": 0.00781811,
      "balance_loss_clip": 1.03361416,
      "balance_loss_mlp": 1.00853372,
      "epoch": 0.9820532090786112,
      "flos": 12529915971840.0,
      "grad_norm": 2.0589387910106316,
      "language_loss": 0.73018444,
      "learning_rate": 3.366511715771958e-09,
      "loss": 0.74884957,
      "num_input_tokens_seen": 352559770,
      "step": 16334,
      "time_per_iteration": 2.4710867404937744
    },
    {
      "auxiliary_loss_clip": 0.01054401,
      "auxiliary_loss_mlp": 0.01032485,
      "balance_loss_clip": 1.03571129,
      "balance_loss_mlp": 1.02065945,
      "epoch": 0.9821133323312792,
      "flos": 18840174497280.0,
      "grad_norm": 1.7712889810796995,
      "language_loss": 0.78452671,
      "learning_rate": 3.3439618745509934e-09,
      "loss": 0.8053956,
      "num_input_tokens_seen": 352577690,
      "step": 16335,
      "time_per_iteration": 2.57309889793396
    },
    {
      "auxiliary_loss_clip": 0.01080473,
      "auxiliary_loss_mlp": 0.01040017,
      "balance_loss_clip": 1.03379977,
      "balance_loss_mlp": 1.02582502,
      "epoch": 0.9821734555839471,
      "flos": 34824013528320.0,
      "grad_norm": 2.208568412586143,
      "language_loss": 0.64325249,
      "learning_rate": 3.3214877467271362e-09,
      "loss": 0.66445744,
      "num_input_tokens_seen": 352598850,
      "step": 16336,
      "time_per_iteration": 2.615187406539917
    },
    {
      "auxiliary_loss_clip": 0.01071935,
      "auxiliary_loss_mlp": 0.01034624,
      "balance_loss_clip": 1.03447866,
      "balance_loss_mlp": 1.02139187,
      "epoch": 0.9822335788366151,
      "flos": 17128169493120.0,
      "grad_norm": 1.9374776045552953,
      "language_loss": 0.72596574,
      "learning_rate": 3.299089333152372e-09,
      "loss": 0.74703133,
      "num_input_tokens_seen": 352616130,
      "step": 16337,
      "time_per_iteration": 3.8979709148406982
    },
    {
      "auxiliary_loss_clip": 0.01094575,
      "auxiliary_loss_mlp": 0.01030241,
      "balance_loss_clip": 1.03454375,
      "balance_loss_mlp": 1.01702046,
      "epoch": 0.982293702089283,
      "flos": 20813250528000.0,
      "grad_norm": 1.69165896582704,
      "language_loss": 0.72923529,
      "learning_rate": 3.2767666346764645e-09,
      "loss": 0.75048345,
      "num_input_tokens_seen": 352636885,
      "step": 16338,
      "time_per_iteration": 2.4932589530944824
    },
    {
      "auxiliary_loss_clip": 0.01044766,
      "auxiliary_loss_mlp": 0.01032587,
      "balance_loss_clip": 1.03214717,
      "balance_loss_mlp": 1.02049315,
      "epoch": 0.982353825341951,
      "flos": 24680829588480.0,
      "grad_norm": 1.6218319056350827,
      "language_loss": 0.81569004,
      "learning_rate": 3.2545196521454045e-09,
      "loss": 0.83646357,
      "num_input_tokens_seen": 352657905,
      "step": 16339,
      "time_per_iteration": 4.046782732009888
    },
    {
      "auxiliary_loss_clip": 0.01053478,
      "auxiliary_loss_mlp": 0.01038284,
      "balance_loss_clip": 1.03252423,
      "balance_loss_mlp": 1.02535009,
      "epoch": 0.982413948594619,
      "flos": 20850489953280.0,
      "grad_norm": 2.549932932095697,
      "language_loss": 0.62158203,
      "learning_rate": 3.232348386403405e-09,
      "loss": 0.64249969,
      "num_input_tokens_seen": 352676320,
      "step": 16340,
      "time_per_iteration": 3.983675718307495
    },
    {
      "auxiliary_loss_clip": 0.01106465,
      "auxiliary_loss_mlp": 0.01032356,
      "balance_loss_clip": 1.03556585,
      "balance_loss_mlp": 1.01986301,
      "epoch": 0.982474071847287,
      "flos": 15377380778880.0,
      "grad_norm": 1.9680628395731437,
      "language_loss": 0.8593328,
      "learning_rate": 3.2102528382904613e-09,
      "loss": 0.88072103,
      "num_input_tokens_seen": 352692665,
      "step": 16341,
      "time_per_iteration": 2.464702606201172
    },
    {
      "auxiliary_loss_clip": 0.01077482,
      "auxiliary_loss_mlp": 0.01027768,
      "balance_loss_clip": 1.033355,
      "balance_loss_mlp": 1.01579392,
      "epoch": 0.9825341950999549,
      "flos": 23774732081280.0,
      "grad_norm": 1.471446440057394,
      "language_loss": 0.67273968,
      "learning_rate": 3.188233008645014e-09,
      "loss": 0.69379216,
      "num_input_tokens_seen": 352716130,
      "step": 16342,
      "time_per_iteration": 2.5644898414611816
    },
    {
      "auxiliary_loss_clip": 0.01102728,
      "auxiliary_loss_mlp": 0.01025696,
      "balance_loss_clip": 1.03346276,
      "balance_loss_mlp": 1.01373351,
      "epoch": 0.9825943183526229,
      "flos": 22746285872640.0,
      "grad_norm": 1.6351494993021072,
      "language_loss": 0.77568734,
      "learning_rate": 3.16628889830195e-09,
      "loss": 0.79697168,
      "num_input_tokens_seen": 352734705,
      "step": 16343,
      "time_per_iteration": 2.471416473388672
    },
    {
      "auxiliary_loss_clip": 0.01068124,
      "auxiliary_loss_mlp": 0.01028866,
      "balance_loss_clip": 1.03378558,
      "balance_loss_mlp": 1.01824439,
      "epoch": 0.9826544416052908,
      "flos": 27709966408320.0,
      "grad_norm": 1.5332183063624663,
      "language_loss": 0.75307381,
      "learning_rate": 3.1444205080932707e-09,
      "loss": 0.77404368,
      "num_input_tokens_seen": 352756225,
      "step": 16344,
      "time_per_iteration": 2.5889952182769775
    },
    {
      "auxiliary_loss_clip": 0.0107891,
      "auxiliary_loss_mlp": 0.01035663,
      "balance_loss_clip": 1.03263521,
      "balance_loss_mlp": 1.02300894,
      "epoch": 0.9827145648579588,
      "flos": 26941657472640.0,
      "grad_norm": 2.193545450188143,
      "language_loss": 0.66700649,
      "learning_rate": 3.122627838848313e-09,
      "loss": 0.68815219,
      "num_input_tokens_seen": 352776210,
      "step": 16345,
      "time_per_iteration": 2.5659103393554688
    },
    {
      "auxiliary_loss_clip": 0.01086502,
      "auxiliary_loss_mlp": 0.01025685,
      "balance_loss_clip": 1.03282666,
      "balance_loss_mlp": 1.01526034,
      "epoch": 0.9827746881106267,
      "flos": 21866545969920.0,
      "grad_norm": 1.3833265314782326,
      "language_loss": 0.79633832,
      "learning_rate": 3.1009108913933045e-09,
      "loss": 0.81746024,
      "num_input_tokens_seen": 352795455,
      "step": 16346,
      "time_per_iteration": 2.477888822555542
    },
    {
      "auxiliary_loss_clip": 0.01092343,
      "auxiliary_loss_mlp": 0.01034092,
      "balance_loss_clip": 1.03462243,
      "balance_loss_mlp": 1.02146149,
      "epoch": 0.9828348113632948,
      "flos": 20850777262080.0,
      "grad_norm": 1.781516647752353,
      "language_loss": 0.74954772,
      "learning_rate": 3.079269666552031e-09,
      "loss": 0.77081203,
      "num_input_tokens_seen": 352812895,
      "step": 16347,
      "time_per_iteration": 2.4978206157684326
    },
    {
      "auxiliary_loss_clip": 0.01030509,
      "auxiliary_loss_mlp": 0.01033044,
      "balance_loss_clip": 1.02888858,
      "balance_loss_mlp": 1.02155805,
      "epoch": 0.9828949346159628,
      "flos": 34569227381760.0,
      "grad_norm": 1.5979083025026812,
      "language_loss": 0.66580534,
      "learning_rate": 3.0577041651449474e-09,
      "loss": 0.68644089,
      "num_input_tokens_seen": 352835470,
      "step": 16348,
      "time_per_iteration": 4.1834399700164795
    },
    {
      "auxiliary_loss_clip": 0.01082557,
      "auxiliary_loss_mlp": 0.01030803,
      "balance_loss_clip": 1.0341785,
      "balance_loss_mlp": 1.01834011,
      "epoch": 0.9829550578686307,
      "flos": 24457464864000.0,
      "grad_norm": 2.5149579025613242,
      "language_loss": 0.68815982,
      "learning_rate": 3.0362143879898437e-09,
      "loss": 0.70929348,
      "num_input_tokens_seen": 352854295,
      "step": 16349,
      "time_per_iteration": 2.710031509399414
    },
    {
      "auxiliary_loss_clip": 0.01077116,
      "auxiliary_loss_mlp": 0.01028083,
      "balance_loss_clip": 1.03406298,
      "balance_loss_mlp": 1.01731205,
      "epoch": 0.9830151811212987,
      "flos": 16910084067840.0,
      "grad_norm": 2.250538215716194,
      "language_loss": 0.75854766,
      "learning_rate": 3.0148003359014018e-09,
      "loss": 0.77959967,
      "num_input_tokens_seen": 352869695,
      "step": 16350,
      "time_per_iteration": 2.485588788986206
    },
    {
      "auxiliary_loss_clip": 0.01073499,
      "auxiliary_loss_mlp": 0.0103032,
      "balance_loss_clip": 1.03257537,
      "balance_loss_mlp": 1.01758242,
      "epoch": 0.9830753043739666,
      "flos": 21288312829440.0,
      "grad_norm": 1.8622186637268714,
      "language_loss": 0.844971,
      "learning_rate": 2.9934620096920826e-09,
      "loss": 0.86600918,
      "num_input_tokens_seen": 352887430,
      "step": 16351,
      "time_per_iteration": 2.560678482055664
    },
    {
      "auxiliary_loss_clip": 0.01074877,
      "auxiliary_loss_mlp": 0.01026463,
      "balance_loss_clip": 1.03465486,
      "balance_loss_mlp": 1.01430964,
      "epoch": 0.9831354276266346,
      "flos": 31723522341120.0,
      "grad_norm": 1.6736392372038043,
      "language_loss": 0.69077665,
      "learning_rate": 2.972199410170795e-09,
      "loss": 0.71179008,
      "num_input_tokens_seen": 352907555,
      "step": 16352,
      "time_per_iteration": 2.647226572036743
    },
    {
      "auxiliary_loss_clip": 0.01090245,
      "auxiliary_loss_mlp": 0.00782425,
      "balance_loss_clip": 1.03410745,
      "balance_loss_mlp": 1.00953543,
      "epoch": 0.9831955508793025,
      "flos": 21619050284160.0,
      "grad_norm": 1.5389774312952147,
      "language_loss": 0.66314912,
      "learning_rate": 2.951012538143782e-09,
      "loss": 0.68187582,
      "num_input_tokens_seen": 352928670,
      "step": 16353,
      "time_per_iteration": 2.497777223587036
    },
    {
      "auxiliary_loss_clip": 0.01077401,
      "auxiliary_loss_mlp": 0.01029371,
      "balance_loss_clip": 1.03254139,
      "balance_loss_mlp": 1.0180645,
      "epoch": 0.9832556741319706,
      "flos": 22968214053120.0,
      "grad_norm": 1.5790553296222525,
      "language_loss": 0.7454325,
      "learning_rate": 2.9299013944144025e-09,
      "loss": 0.76650023,
      "num_input_tokens_seen": 352948345,
      "step": 16354,
      "time_per_iteration": 2.5387771129608154
    },
    {
      "auxiliary_loss_clip": 0.01092396,
      "auxiliary_loss_mlp": 0.01027772,
      "balance_loss_clip": 1.03464377,
      "balance_loss_mlp": 1.01611936,
      "epoch": 0.9833157973846385,
      "flos": 21323900229120.0,
      "grad_norm": 1.984315361952274,
      "language_loss": 0.77459216,
      "learning_rate": 2.9088659797835702e-09,
      "loss": 0.79579383,
      "num_input_tokens_seen": 352967250,
      "step": 16355,
      "time_per_iteration": 2.4766623973846436
    },
    {
      "auxiliary_loss_clip": 0.01090504,
      "auxiliary_loss_mlp": 0.01027321,
      "balance_loss_clip": 1.03402686,
      "balance_loss_mlp": 1.01564455,
      "epoch": 0.9833759206373065,
      "flos": 21068719032960.0,
      "grad_norm": 1.8362842461233266,
      "language_loss": 0.73421192,
      "learning_rate": 2.8879062950484256e-09,
      "loss": 0.75539023,
      "num_input_tokens_seen": 352984725,
      "step": 16356,
      "time_per_iteration": 2.4811246395111084
    },
    {
      "auxiliary_loss_clip": 0.01078762,
      "auxiliary_loss_mlp": 0.01031299,
      "balance_loss_clip": 1.03370929,
      "balance_loss_mlp": 1.0188359,
      "epoch": 0.9834360438899744,
      "flos": 18697322108160.0,
      "grad_norm": 1.4845095104877466,
      "language_loss": 0.75868577,
      "learning_rate": 2.8670223410041104e-09,
      "loss": 0.77978635,
      "num_input_tokens_seen": 353003480,
      "step": 16357,
      "time_per_iteration": 2.494948625564575
    },
    {
      "auxiliary_loss_clip": 0.01081711,
      "auxiliary_loss_mlp": 0.01025167,
      "balance_loss_clip": 1.0344975,
      "balance_loss_mlp": 1.01292968,
      "epoch": 0.9834961671426424,
      "flos": 21105240186240.0,
      "grad_norm": 2.020044451101331,
      "language_loss": 0.80030859,
      "learning_rate": 2.846214118442436e-09,
      "loss": 0.82137734,
      "num_input_tokens_seen": 353021425,
      "step": 16358,
      "time_per_iteration": 2.5254104137420654
    },
    {
      "auxiliary_loss_clip": 0.01090765,
      "auxiliary_loss_mlp": 0.01024899,
      "balance_loss_clip": 1.03239131,
      "balance_loss_mlp": 1.01312661,
      "epoch": 0.9835562903953103,
      "flos": 26687625511680.0,
      "grad_norm": 2.0339416150649803,
      "language_loss": 0.6736474,
      "learning_rate": 2.8254816281523263e-09,
      "loss": 0.69480407,
      "num_input_tokens_seen": 353039870,
      "step": 16359,
      "time_per_iteration": 2.5015368461608887
    },
    {
      "auxiliary_loss_clip": 0.01099383,
      "auxiliary_loss_mlp": 0.01028815,
      "balance_loss_clip": 1.03327489,
      "balance_loss_mlp": 1.01771712,
      "epoch": 0.9836164136479784,
      "flos": 22090162089600.0,
      "grad_norm": 1.6223104007509754,
      "language_loss": 0.6989218,
      "learning_rate": 2.804824870920264e-09,
      "loss": 0.72020382,
      "num_input_tokens_seen": 353059750,
      "step": 16360,
      "time_per_iteration": 2.466033458709717
    },
    {
      "auxiliary_loss_clip": 0.0109428,
      "auxiliary_loss_mlp": 0.01032566,
      "balance_loss_clip": 1.03518283,
      "balance_loss_mlp": 1.02036476,
      "epoch": 0.9836765369006463,
      "flos": 23878405710720.0,
      "grad_norm": 1.6723506026939714,
      "language_loss": 0.83711302,
      "learning_rate": 2.7842438475293996e-09,
      "loss": 0.85838151,
      "num_input_tokens_seen": 353079940,
      "step": 16361,
      "time_per_iteration": 2.5217785835266113
    },
    {
      "auxiliary_loss_clip": 0.0110121,
      "auxiliary_loss_mlp": 0.01025895,
      "balance_loss_clip": 1.03327465,
      "balance_loss_mlp": 1.01486206,
      "epoch": 0.9837366601533143,
      "flos": 25845017293440.0,
      "grad_norm": 1.6584618615378208,
      "language_loss": 0.7601586,
      "learning_rate": 2.76373855876022e-09,
      "loss": 0.78142971,
      "num_input_tokens_seen": 353099990,
      "step": 16362,
      "time_per_iteration": 2.501152276992798
    },
    {
      "auxiliary_loss_clip": 0.01104087,
      "auxiliary_loss_mlp": 0.01033506,
      "balance_loss_clip": 1.03563762,
      "balance_loss_mlp": 1.0210067,
      "epoch": 0.9837967834059823,
      "flos": 21358015171200.0,
      "grad_norm": 1.6251684484034803,
      "language_loss": 0.71156061,
      "learning_rate": 2.7433090053901043e-09,
      "loss": 0.73293656,
      "num_input_tokens_seen": 353118710,
      "step": 16363,
      "time_per_iteration": 2.4234185218811035
    },
    {
      "auxiliary_loss_clip": 0.01076422,
      "auxiliary_loss_mlp": 0.01026933,
      "balance_loss_clip": 1.03210402,
      "balance_loss_mlp": 1.01612687,
      "epoch": 0.9838569066586502,
      "flos": 18515793749760.0,
      "grad_norm": 1.7194852286541413,
      "language_loss": 0.62849605,
      "learning_rate": 2.7229551881937653e-09,
      "loss": 0.64952958,
      "num_input_tokens_seen": 353136415,
      "step": 16364,
      "time_per_iteration": 2.5013372898101807
    },
    {
      "auxiliary_loss_clip": 0.01060748,
      "auxiliary_loss_mlp": 0.01027375,
      "balance_loss_clip": 1.03485012,
      "balance_loss_mlp": 1.01693249,
      "epoch": 0.9839170299113182,
      "flos": 22452392793600.0,
      "grad_norm": 1.5674456888406045,
      "language_loss": 0.75232524,
      "learning_rate": 2.702677107943252e-09,
      "loss": 0.77320647,
      "num_input_tokens_seen": 353154650,
      "step": 16365,
      "time_per_iteration": 2.5627031326293945
    },
    {
      "auxiliary_loss_clip": 0.01057922,
      "auxiliary_loss_mlp": 0.0102688,
      "balance_loss_clip": 1.0356195,
      "balance_loss_mlp": 1.01433921,
      "epoch": 0.9839771531639862,
      "flos": 27892320779520.0,
      "grad_norm": 1.9394514797153193,
      "language_loss": 0.75963855,
      "learning_rate": 2.6824747654072832e-09,
      "loss": 0.78048658,
      "num_input_tokens_seen": 353174065,
      "step": 16366,
      "time_per_iteration": 2.591057777404785
    },
    {
      "auxiliary_loss_clip": 0.01099674,
      "auxiliary_loss_mlp": 0.0102583,
      "balance_loss_clip": 1.03330803,
      "balance_loss_mlp": 1.01476145,
      "epoch": 0.9840372764166542,
      "flos": 28214510797440.0,
      "grad_norm": 1.5616697336729695,
      "language_loss": 0.77277398,
      "learning_rate": 2.662348161352357e-09,
      "loss": 0.79402906,
      "num_input_tokens_seen": 353193560,
      "step": 16367,
      "time_per_iteration": 2.4849090576171875
    },
    {
      "auxiliary_loss_clip": 0.01081812,
      "auxiliary_loss_mlp": 0.01032002,
      "balance_loss_clip": 1.0391233,
      "balance_loss_mlp": 1.01983082,
      "epoch": 0.9840973996693221,
      "flos": 23403989854080.0,
      "grad_norm": 1.4961212436302398,
      "language_loss": 0.61511934,
      "learning_rate": 2.642297296540974e-09,
      "loss": 0.63625747,
      "num_input_tokens_seen": 353213525,
      "step": 16368,
      "time_per_iteration": 2.536449670791626
    },
    {
      "auxiliary_loss_clip": 0.01087623,
      "auxiliary_loss_mlp": 0.01031305,
      "balance_loss_clip": 1.03261721,
      "balance_loss_mlp": 1.02052271,
      "epoch": 0.9841575229219901,
      "flos": 21395865127680.0,
      "grad_norm": 1.450157705797853,
      "language_loss": 0.6520294,
      "learning_rate": 2.6223221717340816e-09,
      "loss": 0.67321873,
      "num_input_tokens_seen": 353234000,
      "step": 16369,
      "time_per_iteration": 2.498229742050171
    },
    {
      "auxiliary_loss_clip": 0.01092913,
      "auxiliary_loss_mlp": 0.00784922,
      "balance_loss_clip": 1.0344553,
      "balance_loss_mlp": 1.01225054,
      "epoch": 0.984217646174658,
      "flos": 24464072966400.0,
      "grad_norm": 1.5193868640707344,
      "language_loss": 0.68693888,
      "learning_rate": 2.6024227876886295e-09,
      "loss": 0.70571721,
      "num_input_tokens_seen": 353254940,
      "step": 16370,
      "time_per_iteration": 2.5134880542755127
    },
    {
      "auxiliary_loss_clip": 0.01103148,
      "auxiliary_loss_mlp": 0.01034498,
      "balance_loss_clip": 1.03327441,
      "balance_loss_mlp": 1.02161193,
      "epoch": 0.984277769427326,
      "flos": 16435057680000.0,
      "grad_norm": 1.852319740814137,
      "language_loss": 0.73651934,
      "learning_rate": 2.582599145159792e-09,
      "loss": 0.75789583,
      "num_input_tokens_seen": 353272590,
      "step": 16371,
      "time_per_iteration": 2.409574508666992
    },
    {
      "auxiliary_loss_clip": 0.01018938,
      "auxiliary_loss_mlp": 0.01003733,
      "balance_loss_clip": 1.00539875,
      "balance_loss_mlp": 1.00253451,
      "epoch": 0.9843378926799939,
      "flos": 64530615288960.0,
      "grad_norm": 0.7772744382690722,
      "language_loss": 0.65194851,
      "learning_rate": 2.562851244898745e-09,
      "loss": 0.67217517,
      "num_input_tokens_seen": 353334380,
      "step": 16372,
      "time_per_iteration": 3.096102476119995
    },
    {
      "auxiliary_loss_clip": 0.01089777,
      "auxiliary_loss_mlp": 0.01029086,
      "balance_loss_clip": 1.03335178,
      "balance_loss_mlp": 1.01701558,
      "epoch": 0.984398015932662,
      "flos": 17382811985280.0,
      "grad_norm": 2.0511172688755193,
      "language_loss": 0.70812571,
      "learning_rate": 2.5431790876544456e-09,
      "loss": 0.72931433,
      "num_input_tokens_seen": 353351640,
      "step": 16373,
      "time_per_iteration": 2.4361982345581055
    },
    {
      "auxiliary_loss_clip": 0.01101584,
      "auxiliary_loss_mlp": 0.01029378,
      "balance_loss_clip": 1.03416502,
      "balance_loss_mlp": 1.01792789,
      "epoch": 0.9844581391853299,
      "flos": 23879088069120.0,
      "grad_norm": 1.7620228928513137,
      "language_loss": 0.81453788,
      "learning_rate": 2.523582674173186e-09,
      "loss": 0.8358475,
      "num_input_tokens_seen": 353372555,
      "step": 16374,
      "time_per_iteration": 2.4855618476867676
    },
    {
      "auxiliary_loss_clip": 0.01057231,
      "auxiliary_loss_mlp": 0.01031133,
      "balance_loss_clip": 1.03584445,
      "balance_loss_mlp": 1.01954532,
      "epoch": 0.9845182624379979,
      "flos": 19865352568320.0,
      "grad_norm": 1.683189837763612,
      "language_loss": 0.69345498,
      "learning_rate": 2.504062005197927e-09,
      "loss": 0.71433854,
      "num_input_tokens_seen": 353391385,
      "step": 16375,
      "time_per_iteration": 2.590319871902466
    },
    {
      "auxiliary_loss_clip": 0.01080034,
      "auxiliary_loss_mlp": 0.01039239,
      "balance_loss_clip": 1.03262091,
      "balance_loss_mlp": 1.02516079,
      "epoch": 0.9845783856906659,
      "flos": 28254659224320.0,
      "grad_norm": 3.3127767278259697,
      "language_loss": 0.8098439,
      "learning_rate": 2.484617081468521e-09,
      "loss": 0.83103657,
      "num_input_tokens_seen": 353411630,
      "step": 16376,
      "time_per_iteration": 3.940930128097534
    },
    {
      "auxiliary_loss_clip": 0.01100378,
      "auxiliary_loss_mlp": 0.01034024,
      "balance_loss_clip": 1.03388739,
      "balance_loss_mlp": 1.02189422,
      "epoch": 0.9846385089433338,
      "flos": 28328383889280.0,
      "grad_norm": 1.4846253814027757,
      "language_loss": 0.62360632,
      "learning_rate": 2.4652479037228224e-09,
      "loss": 0.64495039,
      "num_input_tokens_seen": 353432895,
      "step": 16377,
      "time_per_iteration": 2.5082168579101562
    },
    {
      "auxiliary_loss_clip": 0.01074973,
      "auxiliary_loss_mlp": 0.01033833,
      "balance_loss_clip": 1.03551626,
      "balance_loss_mlp": 1.02177453,
      "epoch": 0.9846986321960018,
      "flos": 24316767290880.0,
      "grad_norm": 1.6249553629422337,
      "language_loss": 0.73191273,
      "learning_rate": 2.445954472695133e-09,
      "loss": 0.7530008,
      "num_input_tokens_seen": 353454195,
      "step": 16378,
      "time_per_iteration": 5.43019437789917
    },
    {
      "auxiliary_loss_clip": 0.0110241,
      "auxiliary_loss_mlp": 0.01029862,
      "balance_loss_clip": 1.03422093,
      "balance_loss_mlp": 1.0187757,
      "epoch": 0.9847587554486698,
      "flos": 27271999877760.0,
      "grad_norm": 2.0257380956088196,
      "language_loss": 0.7090708,
      "learning_rate": 2.426736789116868e-09,
      "loss": 0.73039353,
      "num_input_tokens_seen": 353475125,
      "step": 16379,
      "time_per_iteration": 2.5088512897491455
    },
    {
      "auxiliary_loss_clip": 0.01072074,
      "auxiliary_loss_mlp": 0.01033568,
      "balance_loss_clip": 1.0350734,
      "balance_loss_mlp": 1.02147388,
      "epoch": 0.9848188787013378,
      "flos": 16542717719040.0,
      "grad_norm": 1.8139113836488048,
      "language_loss": 0.68565172,
      "learning_rate": 2.407594853716999e-09,
      "loss": 0.70670813,
      "num_input_tokens_seen": 353493265,
      "step": 16380,
      "time_per_iteration": 2.5210585594177246
    },
    {
      "auxiliary_loss_clip": 0.01080479,
      "auxiliary_loss_mlp": 0.01038241,
      "balance_loss_clip": 1.03402984,
      "balance_loss_mlp": 1.0260874,
      "epoch": 0.9848790019540057,
      "flos": 20193647898240.0,
      "grad_norm": 2.2562628902082933,
      "language_loss": 0.79087758,
      "learning_rate": 2.38852866722139e-09,
      "loss": 0.81206477,
      "num_input_tokens_seen": 353511650,
      "step": 16381,
      "time_per_iteration": 2.5016374588012695
    },
    {
      "auxiliary_loss_clip": 0.01091402,
      "auxiliary_loss_mlp": 0.01024539,
      "balance_loss_clip": 1.03665292,
      "balance_loss_mlp": 1.01279712,
      "epoch": 0.9849391252066737,
      "flos": 28259723041920.0,
      "grad_norm": 1.433644330853625,
      "language_loss": 0.82463664,
      "learning_rate": 2.3695382303527965e-09,
      "loss": 0.84579605,
      "num_input_tokens_seen": 353534035,
      "step": 16382,
      "time_per_iteration": 2.5678870677948
    },
    {
      "auxiliary_loss_clip": 0.01076456,
      "auxiliary_loss_mlp": 0.01031761,
      "balance_loss_clip": 1.03094864,
      "balance_loss_mlp": 1.01809955,
      "epoch": 0.9849992484593416,
      "flos": 22454942659200.0,
      "grad_norm": 1.9021568185716766,
      "language_loss": 0.74119943,
      "learning_rate": 2.3506235438315316e-09,
      "loss": 0.7622816,
      "num_input_tokens_seen": 353549950,
      "step": 16383,
      "time_per_iteration": 2.5002753734588623
    },
    {
      "auxiliary_loss_clip": 0.01057298,
      "auxiliary_loss_mlp": 0.01029599,
      "balance_loss_clip": 1.03427696,
      "balance_loss_mlp": 1.01788068,
      "epoch": 0.9850593717120096,
      "flos": 34497190656000.0,
      "grad_norm": 1.7519994967218893,
      "language_loss": 0.66058612,
      "learning_rate": 2.3317846083750203e-09,
      "loss": 0.68145508,
      "num_input_tokens_seen": 353573745,
      "step": 16384,
      "time_per_iteration": 2.701334238052368
    },
    {
      "auxiliary_loss_clip": 0.01085528,
      "auxiliary_loss_mlp": 0.01034931,
      "balance_loss_clip": 1.03697693,
      "balance_loss_mlp": 1.0213294,
      "epoch": 0.9851194949646775,
      "flos": 38837282152320.0,
      "grad_norm": 2.1267956981078076,
      "language_loss": 0.69963419,
      "learning_rate": 2.313021424697359e-09,
      "loss": 0.72083879,
      "num_input_tokens_seen": 353595335,
      "step": 16385,
      "time_per_iteration": 2.6667003631591797
    },
    {
      "auxiliary_loss_clip": 0.01086493,
      "auxiliary_loss_mlp": 0.01030671,
      "balance_loss_clip": 1.03866506,
      "balance_loss_mlp": 1.01854157,
      "epoch": 0.9851796182173456,
      "flos": 17712436118400.0,
      "grad_norm": 1.9867278924035572,
      "language_loss": 0.80723107,
      "learning_rate": 2.294333993509978e-09,
      "loss": 0.8284027,
      "num_input_tokens_seen": 353614270,
      "step": 16386,
      "time_per_iteration": 2.5072569847106934
    },
    {
      "auxiliary_loss_clip": 0.01074202,
      "auxiliary_loss_mlp": 0.01034002,
      "balance_loss_clip": 1.03266859,
      "balance_loss_mlp": 1.02121043,
      "epoch": 0.9852397414700135,
      "flos": 27454318335360.0,
      "grad_norm": 1.8248963648197345,
      "language_loss": 0.67944199,
      "learning_rate": 2.2757223155216442e-09,
      "loss": 0.70052397,
      "num_input_tokens_seen": 353634900,
      "step": 16387,
      "time_per_iteration": 3.976900339126587
    },
    {
      "auxiliary_loss_clip": 0.01083975,
      "auxiliary_loss_mlp": 0.00781174,
      "balance_loss_clip": 1.03213966,
      "balance_loss_mlp": 1.00729311,
      "epoch": 0.9852998647226815,
      "flos": 18296702743680.0,
      "grad_norm": 1.634834504392881,
      "language_loss": 0.73968589,
      "learning_rate": 2.257186391438237e-09,
      "loss": 0.75833738,
      "num_input_tokens_seen": 353652890,
      "step": 16388,
      "time_per_iteration": 2.474196434020996
    },
    {
      "auxiliary_loss_clip": 0.01086049,
      "auxiliary_loss_mlp": 0.01028123,
      "balance_loss_clip": 1.03316593,
      "balance_loss_mlp": 1.01669145,
      "epoch": 0.9853599879753495,
      "flos": 19642562461440.0,
      "grad_norm": 3.3218398652083034,
      "language_loss": 0.81989151,
      "learning_rate": 2.238726221962528e-09,
      "loss": 0.84103322,
      "num_input_tokens_seen": 353671295,
      "step": 16389,
      "time_per_iteration": 2.494032859802246
    },
    {
      "auxiliary_loss_clip": 0.01081909,
      "auxiliary_loss_mlp": 0.00782501,
      "balance_loss_clip": 1.03306389,
      "balance_loss_mlp": 1.00791585,
      "epoch": 0.9854201112280174,
      "flos": 23841956384640.0,
      "grad_norm": 2.4124576345985105,
      "language_loss": 0.66838658,
      "learning_rate": 2.2203418077946234e-09,
      "loss": 0.68703067,
      "num_input_tokens_seen": 353690560,
      "step": 16390,
      "time_per_iteration": 2.52402663230896
    },
    {
      "auxiliary_loss_clip": 0.01071954,
      "auxiliary_loss_mlp": 0.01035952,
      "balance_loss_clip": 1.03587222,
      "balance_loss_mlp": 1.02354228,
      "epoch": 0.9854802344806854,
      "flos": 30080573233920.0,
      "grad_norm": 1.6702356700800203,
      "language_loss": 0.76914823,
      "learning_rate": 2.2020331496312994e-09,
      "loss": 0.79022729,
      "num_input_tokens_seen": 353710660,
      "step": 16391,
      "time_per_iteration": 2.6345012187957764
    },
    {
      "auxiliary_loss_clip": 0.01061503,
      "auxiliary_loss_mlp": 0.00781184,
      "balance_loss_clip": 1.03257835,
      "balance_loss_mlp": 1.00874054,
      "epoch": 0.9855403577333534,
      "flos": 21907412668800.0,
      "grad_norm": 2.1037926881890723,
      "language_loss": 0.68038183,
      "learning_rate": 2.1838002481673333e-09,
      "loss": 0.69880873,
      "num_input_tokens_seen": 353730440,
      "step": 16392,
      "time_per_iteration": 2.547722816467285
    },
    {
      "auxiliary_loss_clip": 0.01073078,
      "auxiliary_loss_mlp": 0.01027473,
      "balance_loss_clip": 1.03311801,
      "balance_loss_mlp": 1.01446676,
      "epoch": 0.9856004809860214,
      "flos": 15413794191360.0,
      "grad_norm": 1.9330714039153296,
      "language_loss": 0.56146127,
      "learning_rate": 2.1656431040937286e-09,
      "loss": 0.58246678,
      "num_input_tokens_seen": 353748360,
      "step": 16393,
      "time_per_iteration": 2.507110834121704
    },
    {
      "auxiliary_loss_clip": 0.01070746,
      "auxiliary_loss_mlp": 0.01032467,
      "balance_loss_clip": 1.03306627,
      "balance_loss_mlp": 1.01887119,
      "epoch": 0.9856606042386893,
      "flos": 13653201064320.0,
      "grad_norm": 2.675336354208727,
      "language_loss": 0.78963113,
      "learning_rate": 2.1475617180990444e-09,
      "loss": 0.81066322,
      "num_input_tokens_seen": 353760880,
      "step": 16394,
      "time_per_iteration": 2.485213279724121
    },
    {
      "auxiliary_loss_clip": 0.01092844,
      "auxiliary_loss_mlp": 0.01033173,
      "balance_loss_clip": 1.03411531,
      "balance_loss_mlp": 1.02031624,
      "epoch": 0.9857207274913573,
      "flos": 23479151063040.0,
      "grad_norm": 1.4513323002817524,
      "language_loss": 0.7601186,
      "learning_rate": 2.129556090869178e-09,
      "loss": 0.78137875,
      "num_input_tokens_seen": 353782255,
      "step": 16395,
      "time_per_iteration": 2.5158278942108154
    },
    {
      "auxiliary_loss_clip": 0.01089809,
      "auxiliary_loss_mlp": 0.01028077,
      "balance_loss_clip": 1.0331707,
      "balance_loss_mlp": 1.01638842,
      "epoch": 0.9857808507440252,
      "flos": 21065486808960.0,
      "grad_norm": 1.9734112237516896,
      "language_loss": 0.75392056,
      "learning_rate": 2.1116262230866933e-09,
      "loss": 0.77509952,
      "num_input_tokens_seen": 353803580,
      "step": 16396,
      "time_per_iteration": 2.508225679397583
    },
    {
      "auxiliary_loss_clip": 0.01065798,
      "auxiliary_loss_mlp": 0.01028408,
      "balance_loss_clip": 1.03277206,
      "balance_loss_mlp": 1.01641536,
      "epoch": 0.9858409739966932,
      "flos": 25301365971840.0,
      "grad_norm": 1.4168023693299208,
      "language_loss": 0.70872337,
      "learning_rate": 2.0937721154317133e-09,
      "loss": 0.7296654,
      "num_input_tokens_seen": 353824200,
      "step": 16397,
      "time_per_iteration": 2.5855350494384766
    },
    {
      "auxiliary_loss_clip": 0.01076293,
      "auxiliary_loss_mlp": 0.01030634,
      "balance_loss_clip": 1.03337598,
      "balance_loss_mlp": 1.01963139,
      "epoch": 0.9859010972493611,
      "flos": 20558751690240.0,
      "grad_norm": 1.6002587979710707,
      "language_loss": 0.71314663,
      "learning_rate": 2.0759937685810304e-09,
      "loss": 0.73421586,
      "num_input_tokens_seen": 353843350,
      "step": 16398,
      "time_per_iteration": 2.5617291927337646
    },
    {
      "auxiliary_loss_clip": 0.01067623,
      "auxiliary_loss_mlp": 0.01027284,
      "balance_loss_clip": 1.03315997,
      "balance_loss_mlp": 1.01592302,
      "epoch": 0.9859612205020292,
      "flos": 24754985216640.0,
      "grad_norm": 1.40324360610605,
      "language_loss": 0.73937106,
      "learning_rate": 2.058291183208771e-09,
      "loss": 0.76032019,
      "num_input_tokens_seen": 353864520,
      "step": 16399,
      "time_per_iteration": 2.616198778152466
    },
    {
      "auxiliary_loss_clip": 0.01103469,
      "auxiliary_loss_mlp": 0.01028574,
      "balance_loss_clip": 1.03404975,
      "balance_loss_mlp": 1.016361,
      "epoch": 0.9860213437546971,
      "flos": 21105850717440.0,
      "grad_norm": 2.3886840162759366,
      "language_loss": 0.57539302,
      "learning_rate": 2.0406643599863993e-09,
      "loss": 0.59671342,
      "num_input_tokens_seen": 353882240,
      "step": 16400,
      "time_per_iteration": 2.4415886402130127
    },
    {
      "auxiliary_loss_clip": 0.0108292,
      "auxiliary_loss_mlp": 0.01029251,
      "balance_loss_clip": 1.03476167,
      "balance_loss_mlp": 1.01675177,
      "epoch": 0.9860814670073651,
      "flos": 19136078737920.0,
      "grad_norm": 1.6160401332669558,
      "language_loss": 0.80224597,
      "learning_rate": 2.023113299582491e-09,
      "loss": 0.82336766,
      "num_input_tokens_seen": 353901590,
      "step": 16401,
      "time_per_iteration": 2.4968178272247314
    },
    {
      "auxiliary_loss_clip": 0.01090132,
      "auxiliary_loss_mlp": 0.01031263,
      "balance_loss_clip": 1.03445792,
      "balance_loss_mlp": 1.01872802,
      "epoch": 0.9861415902600331,
      "flos": 17237050594560.0,
      "grad_norm": 1.6925924031663175,
      "language_loss": 0.78143024,
      "learning_rate": 2.005638002662069e-09,
      "loss": 0.80264419,
      "num_input_tokens_seen": 353918785,
      "step": 16402,
      "time_per_iteration": 2.4407482147216797
    },
    {
      "auxiliary_loss_clip": 0.01093603,
      "auxiliary_loss_mlp": 0.01032342,
      "balance_loss_clip": 1.03533483,
      "balance_loss_mlp": 1.02076101,
      "epoch": 0.986201713512701,
      "flos": 27782577751680.0,
      "grad_norm": 1.6282033437431807,
      "language_loss": 0.70046568,
      "learning_rate": 1.9882384698881596e-09,
      "loss": 0.72172511,
      "num_input_tokens_seen": 353940390,
      "step": 16403,
      "time_per_iteration": 2.534506320953369
    },
    {
      "auxiliary_loss_clip": 0.01086682,
      "auxiliary_loss_mlp": 0.01030195,
      "balance_loss_clip": 1.03111374,
      "balance_loss_mlp": 1.01840568,
      "epoch": 0.986261836765369,
      "flos": 28730403884160.0,
      "grad_norm": 1.7769141238024977,
      "language_loss": 0.74695563,
      "learning_rate": 1.9709147019204566e-09,
      "loss": 0.76812446,
      "num_input_tokens_seen": 353962180,
      "step": 16404,
      "time_per_iteration": 2.514890193939209
    },
    {
      "auxiliary_loss_clip": 0.0109141,
      "auxiliary_loss_mlp": 0.00782726,
      "balance_loss_clip": 1.03336155,
      "balance_loss_mlp": 1.00919676,
      "epoch": 0.986321960018037,
      "flos": 34313471568000.0,
      "grad_norm": 1.8542403859502028,
      "language_loss": 0.69935077,
      "learning_rate": 1.953666699415768e-09,
      "loss": 0.71809214,
      "num_input_tokens_seen": 353984305,
      "step": 16405,
      "time_per_iteration": 2.582247018814087
    },
    {
      "auxiliary_loss_clip": 0.01080886,
      "auxiliary_loss_mlp": 0.01032637,
      "balance_loss_clip": 1.03600121,
      "balance_loss_mlp": 1.02144289,
      "epoch": 0.986382083270705,
      "flos": 25189755436800.0,
      "grad_norm": 1.5922639662002922,
      "language_loss": 0.69509959,
      "learning_rate": 1.93649446302846e-09,
      "loss": 0.7162348,
      "num_input_tokens_seen": 354004495,
      "step": 16406,
      "time_per_iteration": 2.523158550262451
    },
    {
      "auxiliary_loss_clip": 0.01044969,
      "auxiliary_loss_mlp": 0.01030832,
      "balance_loss_clip": 1.03525639,
      "balance_loss_mlp": 1.01870251,
      "epoch": 0.9864422065233729,
      "flos": 11025904671360.0,
      "grad_norm": 3.098821577809842,
      "language_loss": 0.74684882,
      "learning_rate": 1.9193979934095663e-09,
      "loss": 0.76760685,
      "num_input_tokens_seen": 354015985,
      "step": 16407,
      "time_per_iteration": 2.5548746585845947
    },
    {
      "auxiliary_loss_clip": 0.01078357,
      "auxiliary_loss_mlp": 0.01031324,
      "balance_loss_clip": 1.03291905,
      "balance_loss_mlp": 1.01950479,
      "epoch": 0.9865023297760409,
      "flos": 16545590807040.0,
      "grad_norm": 2.074242088498079,
      "language_loss": 0.77349961,
      "learning_rate": 1.9023772912072357e-09,
      "loss": 0.79459643,
      "num_input_tokens_seen": 354033260,
      "step": 16408,
      "time_per_iteration": 2.459052562713623
    },
    {
      "auxiliary_loss_clip": 0.01095239,
      "auxiliary_loss_mlp": 0.01030592,
      "balance_loss_clip": 1.03534436,
      "balance_loss_mlp": 1.01772952,
      "epoch": 0.9865624530287088,
      "flos": 18880179269760.0,
      "grad_norm": 1.9184932266194012,
      "language_loss": 0.67686546,
      "learning_rate": 1.8854323570669515e-09,
      "loss": 0.69812375,
      "num_input_tokens_seen": 354052825,
      "step": 16409,
      "time_per_iteration": 2.5095601081848145
    },
    {
      "auxiliary_loss_clip": 0.0101045,
      "auxiliary_loss_mlp": 0.01002053,
      "balance_loss_clip": 1.00696087,
      "balance_loss_mlp": 1.0009141,
      "epoch": 0.9866225762813768,
      "flos": 68887798680960.0,
      "grad_norm": 0.8037381907236266,
      "language_loss": 0.61103487,
      "learning_rate": 1.8685631916313118e-09,
      "loss": 0.6311599,
      "num_input_tokens_seen": 354113920,
      "step": 16410,
      "time_per_iteration": 3.1759021282196045
    },
    {
      "auxiliary_loss_clip": 0.01092354,
      "auxiliary_loss_mlp": 0.01033904,
      "balance_loss_clip": 1.03428948,
      "balance_loss_mlp": 1.02200079,
      "epoch": 0.9866826995340447,
      "flos": 29023111814400.0,
      "grad_norm": 2.8567072499514836,
      "language_loss": 0.66023433,
      "learning_rate": 1.8517697955400258e-09,
      "loss": 0.68149686,
      "num_input_tokens_seen": 354134210,
      "step": 16411,
      "time_per_iteration": 2.5368194580078125
    },
    {
      "auxiliary_loss_clip": 0.01026879,
      "auxiliary_loss_mlp": 0.01000929,
      "balance_loss_clip": 1.00418532,
      "balance_loss_mlp": 0.99990427,
      "epoch": 0.9867428227867128,
      "flos": 65376814867200.0,
      "grad_norm": 0.7271356072175748,
      "language_loss": 0.56254053,
      "learning_rate": 1.8350521694299182e-09,
      "loss": 0.58281863,
      "num_input_tokens_seen": 354198010,
      "step": 16412,
      "time_per_iteration": 3.141047477722168
    },
    {
      "auxiliary_loss_clip": 0.01074502,
      "auxiliary_loss_mlp": 0.01034381,
      "balance_loss_clip": 1.03443027,
      "balance_loss_mlp": 1.0211904,
      "epoch": 0.9868029460393807,
      "flos": 26506312634880.0,
      "grad_norm": 1.6258252693910795,
      "language_loss": 0.73027843,
      "learning_rate": 1.818410313934926e-09,
      "loss": 0.75136727,
      "num_input_tokens_seen": 354220000,
      "step": 16413,
      "time_per_iteration": 2.6072239875793457
    },
    {
      "auxiliary_loss_clip": 0.01055138,
      "auxiliary_loss_mlp": 0.0102544,
      "balance_loss_clip": 1.03313839,
      "balance_loss_mlp": 1.01405525,
      "epoch": 0.9868630692920487,
      "flos": 22967280299520.0,
      "grad_norm": 1.4201407581356418,
      "language_loss": 0.7145226,
      "learning_rate": 1.8018442296858782e-09,
      "loss": 0.73532832,
      "num_input_tokens_seen": 354240910,
      "step": 16414,
      "time_per_iteration": 3.9915575981140137
    },
    {
      "auxiliary_loss_clip": 0.01085222,
      "auxiliary_loss_mlp": 0.0103537,
      "balance_loss_clip": 1.03395319,
      "balance_loss_mlp": 1.02381241,
      "epoch": 0.9869231925447167,
      "flos": 19828687760640.0,
      "grad_norm": 2.1447131044558514,
      "language_loss": 0.70159596,
      "learning_rate": 1.7853539173111608e-09,
      "loss": 0.72280186,
      "num_input_tokens_seen": 354259430,
      "step": 16415,
      "time_per_iteration": 2.4566256999969482
    },
    {
      "auxiliary_loss_clip": 0.01061339,
      "auxiliary_loss_mlp": 0.01032424,
      "balance_loss_clip": 1.03128946,
      "balance_loss_mlp": 1.02159345,
      "epoch": 0.9869833157973846,
      "flos": 20195228096640.0,
      "grad_norm": 1.38119265379021,
      "language_loss": 0.75280213,
      "learning_rate": 1.7689393774362737e-09,
      "loss": 0.77373976,
      "num_input_tokens_seen": 354279490,
      "step": 16416,
      "time_per_iteration": 3.953526258468628
    },
    {
      "auxiliary_loss_clip": 0.01078228,
      "auxiliary_loss_mlp": 0.01030542,
      "balance_loss_clip": 1.03627419,
      "balance_loss_mlp": 1.01887107,
      "epoch": 0.9870434390500527,
      "flos": 16099507802880.0,
      "grad_norm": 2.4414249029153616,
      "language_loss": 0.70838392,
      "learning_rate": 1.7526006106833858e-09,
      "loss": 0.72947156,
      "num_input_tokens_seen": 354295080,
      "step": 16417,
      "time_per_iteration": 3.876779794692993
    },
    {
      "auxiliary_loss_clip": 0.01082074,
      "auxiliary_loss_mlp": 0.01033579,
      "balance_loss_clip": 1.03642893,
      "balance_loss_mlp": 1.02110934,
      "epoch": 0.9871035623027206,
      "flos": 21760753438080.0,
      "grad_norm": 1.437770589341119,
      "language_loss": 0.70450616,
      "learning_rate": 1.7363376176720013e-09,
      "loss": 0.72566265,
      "num_input_tokens_seen": 354314610,
      "step": 16418,
      "time_per_iteration": 2.524843215942383
    },
    {
      "auxiliary_loss_clip": 0.01027044,
      "auxiliary_loss_mlp": 0.01002281,
      "balance_loss_clip": 1.00436306,
      "balance_loss_mlp": 1.00123787,
      "epoch": 0.9871636855553886,
      "flos": 70219583245440.0,
      "grad_norm": 0.6545908431017157,
      "language_loss": 0.53676414,
      "learning_rate": 1.7201503990189603e-09,
      "loss": 0.55705738,
      "num_input_tokens_seen": 354383115,
      "step": 16419,
      "time_per_iteration": 3.191831350326538
    },
    {
      "auxiliary_loss_clip": 0.01081803,
      "auxiliary_loss_mlp": 0.01036751,
      "balance_loss_clip": 1.03214645,
      "balance_loss_mlp": 1.02342939,
      "epoch": 0.9872238088080565,
      "flos": 25045825639680.0,
      "grad_norm": 1.7672420852928092,
      "language_loss": 0.78010762,
      "learning_rate": 1.7040389553382162e-09,
      "loss": 0.80129308,
      "num_input_tokens_seen": 354403115,
      "step": 16420,
      "time_per_iteration": 2.5313658714294434
    },
    {
      "auxiliary_loss_clip": 0.01066181,
      "auxiliary_loss_mlp": 0.01025997,
      "balance_loss_clip": 1.03995895,
      "balance_loss_mlp": 1.01401639,
      "epoch": 0.9872839320607245,
      "flos": 19465846525440.0,
      "grad_norm": 1.5788846792298823,
      "language_loss": 0.70717388,
      "learning_rate": 1.6880032872403916e-09,
      "loss": 0.72809565,
      "num_input_tokens_seen": 354424520,
      "step": 16421,
      "time_per_iteration": 2.571572780609131
    },
    {
      "auxiliary_loss_clip": 0.01094758,
      "auxiliary_loss_mlp": 0.01032505,
      "balance_loss_clip": 1.03429401,
      "balance_loss_mlp": 1.01968956,
      "epoch": 0.9873440553133924,
      "flos": 26942914448640.0,
      "grad_norm": 2.2973214135850015,
      "language_loss": 0.82188433,
      "learning_rate": 1.6720433953338886e-09,
      "loss": 0.84315693,
      "num_input_tokens_seen": 354444800,
      "step": 16422,
      "time_per_iteration": 2.501039981842041
    },
    {
      "auxiliary_loss_clip": 0.01068591,
      "auxiliary_loss_mlp": 0.01027578,
      "balance_loss_clip": 1.03347266,
      "balance_loss_mlp": 1.01578212,
      "epoch": 0.9874041785660604,
      "flos": 19062210418560.0,
      "grad_norm": 1.6345528640928064,
      "language_loss": 0.85876733,
      "learning_rate": 1.656159280223779e-09,
      "loss": 0.87972903,
      "num_input_tokens_seen": 354464590,
      "step": 16423,
      "time_per_iteration": 2.6315150260925293
    },
    {
      "auxiliary_loss_clip": 0.01096694,
      "auxiliary_loss_mlp": 0.01029413,
      "balance_loss_clip": 1.03641129,
      "balance_loss_mlp": 1.01765275,
      "epoch": 0.9874643018187284,
      "flos": 21105814803840.0,
      "grad_norm": 1.8712927419852974,
      "language_loss": 0.70161891,
      "learning_rate": 1.6403509425122475e-09,
      "loss": 0.72288001,
      "num_input_tokens_seen": 354484145,
      "step": 16424,
      "time_per_iteration": 2.4725449085235596
    },
    {
      "auxiliary_loss_clip": 0.01093061,
      "auxiliary_loss_mlp": 0.00781624,
      "balance_loss_clip": 1.03293037,
      "balance_loss_mlp": 1.00736153,
      "epoch": 0.9875244250713964,
      "flos": 24426043441920.0,
      "grad_norm": 2.0722133987134277,
      "language_loss": 0.80861616,
      "learning_rate": 1.6246183827990366e-09,
      "loss": 0.82736295,
      "num_input_tokens_seen": 354502475,
      "step": 16425,
      "time_per_iteration": 2.5077102184295654
    },
    {
      "auxiliary_loss_clip": 0.01052276,
      "auxiliary_loss_mlp": 0.01035345,
      "balance_loss_clip": 1.03168166,
      "balance_loss_mlp": 1.02188623,
      "epoch": 0.9875845483240643,
      "flos": 25117610970240.0,
      "grad_norm": 2.041304649000619,
      "language_loss": 0.79747468,
      "learning_rate": 1.6089616016803364e-09,
      "loss": 0.81835091,
      "num_input_tokens_seen": 354521855,
      "step": 16426,
      "time_per_iteration": 4.062847852706909
    },
    {
      "auxiliary_loss_clip": 0.01093649,
      "auxiliary_loss_mlp": 0.01030852,
      "balance_loss_clip": 1.0364722,
      "balance_loss_mlp": 1.01930618,
      "epoch": 0.9876446715767323,
      "flos": 16581788737920.0,
      "grad_norm": 1.8902320369145071,
      "language_loss": 0.84601408,
      "learning_rate": 1.593380599750338e-09,
      "loss": 0.86725909,
      "num_input_tokens_seen": 354539535,
      "step": 16427,
      "time_per_iteration": 2.4575417041778564
    },
    {
      "auxiliary_loss_clip": 0.01102106,
      "auxiliary_loss_mlp": 0.0102853,
      "balance_loss_clip": 1.03535509,
      "balance_loss_mlp": 1.01665664,
      "epoch": 0.9877047948294003,
      "flos": 21616141282560.0,
      "grad_norm": 1.7047268113470202,
      "language_loss": 0.70424807,
      "learning_rate": 1.577875377599458e-09,
      "loss": 0.72555441,
      "num_input_tokens_seen": 354557430,
      "step": 16428,
      "time_per_iteration": 2.4707367420196533
    },
    {
      "auxiliary_loss_clip": 0.01063565,
      "auxiliary_loss_mlp": 0.01030654,
      "balance_loss_clip": 1.03202009,
      "balance_loss_mlp": 1.01928723,
      "epoch": 0.9877649180820682,
      "flos": 21178497974400.0,
      "grad_norm": 2.0134167491860127,
      "language_loss": 0.79842854,
      "learning_rate": 1.5624459358158926e-09,
      "loss": 0.81937075,
      "num_input_tokens_seen": 354574735,
      "step": 16429,
      "time_per_iteration": 2.5267794132232666
    },
    {
      "auxiliary_loss_clip": 0.01101304,
      "auxiliary_loss_mlp": 0.01027148,
      "balance_loss_clip": 1.03391767,
      "balance_loss_mlp": 1.0161984,
      "epoch": 0.9878250413347363,
      "flos": 39749233576320.0,
      "grad_norm": 1.6079748364419204,
      "language_loss": 0.61917436,
      "learning_rate": 1.5470922749845073e-09,
      "loss": 0.64045882,
      "num_input_tokens_seen": 354597050,
      "step": 16430,
      "time_per_iteration": 2.621305465698242
    },
    {
      "auxiliary_loss_clip": 0.01104749,
      "auxiliary_loss_mlp": 0.01034604,
      "balance_loss_clip": 1.03489745,
      "balance_loss_mlp": 1.02259946,
      "epoch": 0.9878851645874042,
      "flos": 29425634599680.0,
      "grad_norm": 1.4672224697854215,
      "language_loss": 0.73110342,
      "learning_rate": 1.531814395687725e-09,
      "loss": 0.75249696,
      "num_input_tokens_seen": 354619095,
      "step": 16431,
      "time_per_iteration": 2.510655641555786
    },
    {
      "auxiliary_loss_clip": 0.01103465,
      "auxiliary_loss_mlp": 0.01032525,
      "balance_loss_clip": 1.0355587,
      "balance_loss_mlp": 1.02062201,
      "epoch": 0.9879452878400722,
      "flos": 15806261168640.0,
      "grad_norm": 2.107735415237377,
      "language_loss": 0.80264997,
      "learning_rate": 1.5166122985048602e-09,
      "loss": 0.8240099,
      "num_input_tokens_seen": 354633790,
      "step": 16432,
      "time_per_iteration": 2.4183809757232666
    },
    {
      "auxiliary_loss_clip": 0.01088627,
      "auxiliary_loss_mlp": 0.01030593,
      "balance_loss_clip": 1.03255725,
      "balance_loss_mlp": 1.01978707,
      "epoch": 0.9880054110927401,
      "flos": 22233912318720.0,
      "grad_norm": 1.6614794882308088,
      "language_loss": 0.8026787,
      "learning_rate": 1.5014859840123405e-09,
      "loss": 0.8238709,
      "num_input_tokens_seen": 354653180,
      "step": 16433,
      "time_per_iteration": 2.4745142459869385
    },
    {
      "auxiliary_loss_clip": 0.01100664,
      "auxiliary_loss_mlp": 0.01032687,
      "balance_loss_clip": 1.03491962,
      "balance_loss_mlp": 1.02053952,
      "epoch": 0.9880655343454081,
      "flos": 28763836467840.0,
      "grad_norm": 2.0661015334208215,
      "language_loss": 0.64878559,
      "learning_rate": 1.4864354527837075e-09,
      "loss": 0.67011911,
      "num_input_tokens_seen": 354669900,
      "step": 16434,
      "time_per_iteration": 2.4818968772888184
    },
    {
      "auxiliary_loss_clip": 0.01093181,
      "auxiliary_loss_mlp": 0.01031503,
      "balance_loss_clip": 1.03257823,
      "balance_loss_mlp": 1.01912332,
      "epoch": 0.988125657598076,
      "flos": 32853379622400.0,
      "grad_norm": 1.6739883779034581,
      "language_loss": 0.69321609,
      "learning_rate": 1.4714607053896154e-09,
      "loss": 0.714463,
      "num_input_tokens_seen": 354693165,
      "step": 16435,
      "time_per_iteration": 2.56654953956604
    },
    {
      "auxiliary_loss_clip": 0.01048099,
      "auxiliary_loss_mlp": 0.01031582,
      "balance_loss_clip": 1.03687155,
      "balance_loss_mlp": 1.01953554,
      "epoch": 0.988185780850744,
      "flos": 19390685316480.0,
      "grad_norm": 1.668878350758069,
      "language_loss": 0.75496042,
      "learning_rate": 1.4565617423980548e-09,
      "loss": 0.77575719,
      "num_input_tokens_seen": 354711915,
      "step": 16436,
      "time_per_iteration": 2.574218988418579
    },
    {
      "auxiliary_loss_clip": 0.01073984,
      "auxiliary_loss_mlp": 0.01031842,
      "balance_loss_clip": 1.03479457,
      "balance_loss_mlp": 1.01909304,
      "epoch": 0.988245904103412,
      "flos": 22528415928960.0,
      "grad_norm": 2.007869853503989,
      "language_loss": 0.74078333,
      "learning_rate": 1.4417385643741286e-09,
      "loss": 0.76184154,
      "num_input_tokens_seen": 354729135,
      "step": 16437,
      "time_per_iteration": 2.519456148147583
    },
    {
      "auxiliary_loss_clip": 0.01066343,
      "auxiliary_loss_mlp": 0.01031795,
      "balance_loss_clip": 1.03367865,
      "balance_loss_mlp": 1.02069676,
      "epoch": 0.98830602735608,
      "flos": 28659193171200.0,
      "grad_norm": 1.4946558106446606,
      "language_loss": 0.60048234,
      "learning_rate": 1.4269911718796103e-09,
      "loss": 0.62146372,
      "num_input_tokens_seen": 354752530,
      "step": 16438,
      "time_per_iteration": 2.622162103652954
    },
    {
      "auxiliary_loss_clip": 0.01079029,
      "auxiliary_loss_mlp": 0.01030137,
      "balance_loss_clip": 1.03391981,
      "balance_loss_mlp": 1.01745319,
      "epoch": 0.9883661506087479,
      "flos": 20996035862400.0,
      "grad_norm": 1.7937514205834721,
      "language_loss": 0.71949887,
      "learning_rate": 1.4123195654738295e-09,
      "loss": 0.74059057,
      "num_input_tokens_seen": 354771135,
      "step": 16439,
      "time_per_iteration": 2.498781442642212
    },
    {
      "auxiliary_loss_clip": 0.01089392,
      "auxiliary_loss_mlp": 0.01028752,
      "balance_loss_clip": 1.03437948,
      "balance_loss_mlp": 1.01712906,
      "epoch": 0.9884262738614159,
      "flos": 32706109860480.0,
      "grad_norm": 1.5916403060389754,
      "language_loss": 0.5970341,
      "learning_rate": 1.3977237457134528e-09,
      "loss": 0.61821556,
      "num_input_tokens_seen": 354791800,
      "step": 16440,
      "time_per_iteration": 2.5497617721557617
    },
    {
      "auxiliary_loss_clip": 0.01104111,
      "auxiliary_loss_mlp": 0.01030249,
      "balance_loss_clip": 1.03359556,
      "balance_loss_mlp": 1.01817322,
      "epoch": 0.9884863971140839,
      "flos": 17564699479680.0,
      "grad_norm": 2.864120877332571,
      "language_loss": 0.75976336,
      "learning_rate": 1.3832037131513707e-09,
      "loss": 0.78110695,
      "num_input_tokens_seen": 354809200,
      "step": 16441,
      "time_per_iteration": 2.3990588188171387
    },
    {
      "auxiliary_loss_clip": 0.01080512,
      "auxiliary_loss_mlp": 0.01028939,
      "balance_loss_clip": 1.03390932,
      "balance_loss_mlp": 1.01694655,
      "epoch": 0.9885465203667518,
      "flos": 40552519380480.0,
      "grad_norm": 1.8996918805491922,
      "language_loss": 0.67818093,
      "learning_rate": 1.3687594683386982e-09,
      "loss": 0.69927537,
      "num_input_tokens_seen": 354829945,
      "step": 16442,
      "time_per_iteration": 2.67311429977417
    },
    {
      "auxiliary_loss_clip": 0.01090283,
      "auxiliary_loss_mlp": 0.010309,
      "balance_loss_clip": 1.03409612,
      "balance_loss_mlp": 1.01913428,
      "epoch": 0.9886066436194199,
      "flos": 13807976768640.0,
      "grad_norm": 3.054310592024688,
      "language_loss": 0.74559879,
      "learning_rate": 1.3543910118227753e-09,
      "loss": 0.76681066,
      "num_input_tokens_seen": 354845055,
      "step": 16443,
      "time_per_iteration": 2.4313974380493164
    },
    {
      "auxiliary_loss_clip": 0.0108166,
      "auxiliary_loss_mlp": 0.01029813,
      "balance_loss_clip": 1.0329411,
      "balance_loss_mlp": 1.01715279,
      "epoch": 0.9886667668720878,
      "flos": 23325129544320.0,
      "grad_norm": 1.7798276026043987,
      "language_loss": 0.73694032,
      "learning_rate": 1.3400983441487213e-09,
      "loss": 0.75805509,
      "num_input_tokens_seen": 354864680,
      "step": 16444,
      "time_per_iteration": 2.5240495204925537
    },
    {
      "auxiliary_loss_clip": 0.01055922,
      "auxiliary_loss_mlp": 0.01038986,
      "balance_loss_clip": 1.03533292,
      "balance_loss_mlp": 1.0260396,
      "epoch": 0.9887268901247558,
      "flos": 22706029704960.0,
      "grad_norm": 1.791040356119224,
      "language_loss": 0.69161534,
      "learning_rate": 1.325881465858547e-09,
      "loss": 0.71256441,
      "num_input_tokens_seen": 354885685,
      "step": 16445,
      "time_per_iteration": 2.600954532623291
    },
    {
      "auxiliary_loss_clip": 0.01096087,
      "auxiliary_loss_mlp": 0.01025653,
      "balance_loss_clip": 1.03660905,
      "balance_loss_mlp": 1.01332068,
      "epoch": 0.9887870133774237,
      "flos": 13041283944960.0,
      "grad_norm": 2.581257845665943,
      "language_loss": 0.60242319,
      "learning_rate": 1.311740377491155e-09,
      "loss": 0.62364066,
      "num_input_tokens_seen": 354901505,
      "step": 16446,
      "time_per_iteration": 2.453902244567871
    },
    {
      "auxiliary_loss_clip": 0.01082656,
      "auxiliary_loss_mlp": 0.01034052,
      "balance_loss_clip": 1.03458631,
      "balance_loss_mlp": 1.02286458,
      "epoch": 0.9888471366300917,
      "flos": 15158864390400.0,
      "grad_norm": 2.0343570679751624,
      "language_loss": 0.7076813,
      "learning_rate": 1.297675079582783e-09,
      "loss": 0.72884834,
      "num_input_tokens_seen": 354920060,
      "step": 16447,
      "time_per_iteration": 2.4692814350128174
    },
    {
      "auxiliary_loss_clip": 0.01101531,
      "auxiliary_loss_mlp": 0.00780957,
      "balance_loss_clip": 1.0340097,
      "balance_loss_mlp": 1.00732303,
      "epoch": 0.9889072598827596,
      "flos": 25118796119040.0,
      "grad_norm": 1.821247505641219,
      "language_loss": 0.83643174,
      "learning_rate": 1.2836855726667818e-09,
      "loss": 0.85525668,
      "num_input_tokens_seen": 354938690,
      "step": 16448,
      "time_per_iteration": 2.473247766494751
    },
    {
      "auxiliary_loss_clip": 0.01087492,
      "auxiliary_loss_mlp": 0.01024023,
      "balance_loss_clip": 1.03307891,
      "balance_loss_mlp": 1.01362181,
      "epoch": 0.9889673831354276,
      "flos": 16728663450240.0,
      "grad_norm": 1.4575548840814965,
      "language_loss": 0.70259541,
      "learning_rate": 1.26977185727406e-09,
      "loss": 0.72371054,
      "num_input_tokens_seen": 354956955,
      "step": 16449,
      "time_per_iteration": 2.450620174407959
    },
    {
      "auxiliary_loss_clip": 0.01094052,
      "auxiliary_loss_mlp": 0.01028614,
      "balance_loss_clip": 1.03400874,
      "balance_loss_mlp": 1.01708043,
      "epoch": 0.9890275063880956,
      "flos": 35585175657600.0,
      "grad_norm": 2.375336117168634,
      "language_loss": 0.73670459,
      "learning_rate": 1.25593393393153e-09,
      "loss": 0.75793123,
      "num_input_tokens_seen": 354976800,
      "step": 16450,
      "time_per_iteration": 2.580023765563965
    },
    {
      "auxiliary_loss_clip": 0.01103565,
      "auxiliary_loss_mlp": 0.01029234,
      "balance_loss_clip": 1.03259814,
      "balance_loss_mlp": 1.01712859,
      "epoch": 0.9890876296407636,
      "flos": 18952359649920.0,
      "grad_norm": 1.7957007395828775,
      "language_loss": 0.79604566,
      "learning_rate": 1.242171803164549e-09,
      "loss": 0.81737363,
      "num_input_tokens_seen": 354996625,
      "step": 16451,
      "time_per_iteration": 2.426565408706665
    },
    {
      "auxiliary_loss_clip": 0.01067631,
      "auxiliary_loss_mlp": 0.01034186,
      "balance_loss_clip": 1.03445745,
      "balance_loss_mlp": 1.02187157,
      "epoch": 0.9891477528934315,
      "flos": 23769309127680.0,
      "grad_norm": 1.9149674334374505,
      "language_loss": 0.69962287,
      "learning_rate": 1.2284854654946996e-09,
      "loss": 0.72064102,
      "num_input_tokens_seen": 355014535,
      "step": 16452,
      "time_per_iteration": 2.570059299468994
    },
    {
      "auxiliary_loss_clip": 0.01099722,
      "auxiliary_loss_mlp": 0.01026481,
      "balance_loss_clip": 1.0346179,
      "balance_loss_mlp": 1.0158,
      "epoch": 0.9892078761460995,
      "flos": 20772922533120.0,
      "grad_norm": 1.6596035561829157,
      "language_loss": 0.7416023,
      "learning_rate": 1.2148749214409004e-09,
      "loss": 0.76286435,
      "num_input_tokens_seen": 355033280,
      "step": 16453,
      "time_per_iteration": 3.856184244155884
    },
    {
      "auxiliary_loss_clip": 0.01062608,
      "auxiliary_loss_mlp": 0.01034895,
      "balance_loss_clip": 1.03394961,
      "balance_loss_mlp": 1.02345061,
      "epoch": 0.9892679993987675,
      "flos": 23367827836800.0,
      "grad_norm": 2.2520938459452813,
      "language_loss": 0.69297361,
      "learning_rate": 1.2013401715191828e-09,
      "loss": 0.71394867,
      "num_input_tokens_seen": 355053320,
      "step": 16454,
      "time_per_iteration": 2.613281488418579
    },
    {
      "auxiliary_loss_clip": 0.01073451,
      "auxiliary_loss_mlp": 0.0103029,
      "balance_loss_clip": 1.03346038,
      "balance_loss_mlp": 1.01857126,
      "epoch": 0.9893281226514354,
      "flos": 22705419173760.0,
      "grad_norm": 1.8425586890104222,
      "language_loss": 0.7602663,
      "learning_rate": 1.1878812162433583e-09,
      "loss": 0.7813037,
      "num_input_tokens_seen": 355070230,
      "step": 16455,
      "time_per_iteration": 3.9261350631713867
    },
    {
      "auxiliary_loss_clip": 0.01073605,
      "auxiliary_loss_mlp": 0.01023763,
      "balance_loss_clip": 1.03423619,
      "balance_loss_mlp": 1.01203835,
      "epoch": 0.9893882459041035,
      "flos": 21796664060160.0,
      "grad_norm": 1.6544613205331633,
      "language_loss": 0.65659112,
      "learning_rate": 1.1744980561230188e-09,
      "loss": 0.6775648,
      "num_input_tokens_seen": 355090125,
      "step": 16456,
      "time_per_iteration": 3.901376247406006
    },
    {
      "auxiliary_loss_clip": 0.01094281,
      "auxiliary_loss_mlp": 0.01030255,
      "balance_loss_clip": 1.03545141,
      "balance_loss_mlp": 1.01838207,
      "epoch": 0.9894483691567714,
      "flos": 18113773754880.0,
      "grad_norm": 1.7338063533888701,
      "language_loss": 0.74136579,
      "learning_rate": 1.161190691666203e-09,
      "loss": 0.76261109,
      "num_input_tokens_seen": 355107890,
      "step": 16457,
      "time_per_iteration": 2.4384751319885254
    },
    {
      "auxiliary_loss_clip": 0.01103453,
      "auxiliary_loss_mlp": 0.01026727,
      "balance_loss_clip": 1.03525651,
      "balance_loss_mlp": 1.0151701,
      "epoch": 0.9895084924094394,
      "flos": 31211615664000.0,
      "grad_norm": 2.057235615271389,
      "language_loss": 0.68741465,
      "learning_rate": 1.1479591233773954e-09,
      "loss": 0.70871639,
      "num_input_tokens_seen": 355126340,
      "step": 16458,
      "time_per_iteration": 2.5709078311920166
    },
    {
      "auxiliary_loss_clip": 0.01088879,
      "auxiliary_loss_mlp": 0.01028599,
      "balance_loss_clip": 1.03372788,
      "balance_loss_mlp": 1.01671994,
      "epoch": 0.9895686156621073,
      "flos": 19678042120320.0,
      "grad_norm": 1.730903718665965,
      "language_loss": 0.79195976,
      "learning_rate": 1.1348033517581956e-09,
      "loss": 0.81313455,
      "num_input_tokens_seen": 355144025,
      "step": 16459,
      "time_per_iteration": 2.448833465576172
    },
    {
      "auxiliary_loss_clip": 0.01078823,
      "auxiliary_loss_mlp": 0.01034292,
      "balance_loss_clip": 1.03500676,
      "balance_loss_mlp": 1.02249014,
      "epoch": 0.9896287389147753,
      "flos": 23581675457280.0,
      "grad_norm": 2.7415807171000153,
      "language_loss": 0.70486844,
      "learning_rate": 1.1217233773075373e-09,
      "loss": 0.72599959,
      "num_input_tokens_seen": 355163125,
      "step": 16460,
      "time_per_iteration": 2.5391249656677246
    },
    {
      "auxiliary_loss_clip": 0.01082678,
      "auxiliary_loss_mlp": 0.01026408,
      "balance_loss_clip": 1.03270626,
      "balance_loss_mlp": 1.01418924,
      "epoch": 0.9896888621674432,
      "flos": 29605331364480.0,
      "grad_norm": 1.526694438059555,
      "language_loss": 0.87473035,
      "learning_rate": 1.1087192005214685e-09,
      "loss": 0.89582121,
      "num_input_tokens_seen": 355184060,
      "step": 16461,
      "time_per_iteration": 2.5636961460113525
    },
    {
      "auxiliary_loss_clip": 0.01091645,
      "auxiliary_loss_mlp": 0.01031277,
      "balance_loss_clip": 1.03419363,
      "balance_loss_mlp": 1.01858699,
      "epoch": 0.9897489854201112,
      "flos": 23695045758720.0,
      "grad_norm": 1.649883411388797,
      "language_loss": 0.62967896,
      "learning_rate": 1.09579082189315e-09,
      "loss": 0.65090823,
      "num_input_tokens_seen": 355204505,
      "step": 16462,
      "time_per_iteration": 2.5204575061798096
    },
    {
      "auxiliary_loss_clip": 0.01095541,
      "auxiliary_loss_mlp": 0.01028486,
      "balance_loss_clip": 1.03709817,
      "balance_loss_mlp": 1.01698864,
      "epoch": 0.9898091086727792,
      "flos": 13225146687360.0,
      "grad_norm": 1.7394982597073374,
      "language_loss": 0.72770381,
      "learning_rate": 1.0829382419126343e-09,
      "loss": 0.7489441,
      "num_input_tokens_seen": 355223055,
      "step": 16463,
      "time_per_iteration": 2.4783084392547607
    },
    {
      "auxiliary_loss_clip": 0.01092839,
      "auxiliary_loss_mlp": 0.0102585,
      "balance_loss_clip": 1.03454018,
      "balance_loss_mlp": 1.01351178,
      "epoch": 0.9898692319254472,
      "flos": 22930400010240.0,
      "grad_norm": 2.066418675120834,
      "language_loss": 0.70160204,
      "learning_rate": 1.0701614610675314e-09,
      "loss": 0.72278893,
      "num_input_tokens_seen": 355242000,
      "step": 16464,
      "time_per_iteration": 2.490875482559204
    },
    {
      "auxiliary_loss_clip": 0.01074852,
      "auxiliary_loss_mlp": 0.01027915,
      "balance_loss_clip": 1.03482592,
      "balance_loss_mlp": 1.0158149,
      "epoch": 0.9899293551781151,
      "flos": 12458346122880.0,
      "grad_norm": 1.9677920931170145,
      "language_loss": 0.73314351,
      "learning_rate": 1.0574604798421204e-09,
      "loss": 0.75417113,
      "num_input_tokens_seen": 355260175,
      "step": 16465,
      "time_per_iteration": 3.9032723903656006
    },
    {
      "auxiliary_loss_clip": 0.01100489,
      "auxiliary_loss_mlp": 0.0103129,
      "balance_loss_clip": 1.03339791,
      "balance_loss_mlp": 1.02070463,
      "epoch": 0.9899894784307831,
      "flos": 26871129118080.0,
      "grad_norm": 1.8661807326664825,
      "language_loss": 0.86538041,
      "learning_rate": 1.0448352987182386e-09,
      "loss": 0.88669825,
      "num_input_tokens_seen": 355281930,
      "step": 16466,
      "time_per_iteration": 2.4685120582580566
    },
    {
      "auxiliary_loss_clip": 0.01073942,
      "auxiliary_loss_mlp": 0.01024184,
      "balance_loss_clip": 1.03491151,
      "balance_loss_mlp": 1.01271653,
      "epoch": 0.990049601683451,
      "flos": 21542093395200.0,
      "grad_norm": 1.7073609407303227,
      "language_loss": 0.71649468,
      "learning_rate": 1.0322859181743915e-09,
      "loss": 0.73747593,
      "num_input_tokens_seen": 355301555,
      "step": 16467,
      "time_per_iteration": 2.5478591918945312
    },
    {
      "auxiliary_loss_clip": 0.0107539,
      "auxiliary_loss_mlp": 0.01032995,
      "balance_loss_clip": 1.03218079,
      "balance_loss_mlp": 1.02037072,
      "epoch": 0.990109724936119,
      "flos": 28771809287040.0,
      "grad_norm": 1.42687668494041,
      "language_loss": 0.65095133,
      "learning_rate": 1.019812338686643e-09,
      "loss": 0.67203516,
      "num_input_tokens_seen": 355324925,
      "step": 16468,
      "time_per_iteration": 2.589926242828369
    },
    {
      "auxiliary_loss_clip": 0.01075894,
      "auxiliary_loss_mlp": 0.01031882,
      "balance_loss_clip": 1.03496504,
      "balance_loss_mlp": 1.02000296,
      "epoch": 0.9901698481887871,
      "flos": 29274270687360.0,
      "grad_norm": 1.892468697759678,
      "language_loss": 0.61983585,
      "learning_rate": 1.0074145607281704e-09,
      "loss": 0.64091361,
      "num_input_tokens_seen": 355343875,
      "step": 16469,
      "time_per_iteration": 2.5955936908721924
    },
    {
      "auxiliary_loss_clip": 0.01069636,
      "auxiliary_loss_mlp": 0.01029467,
      "balance_loss_clip": 1.03397727,
      "balance_loss_mlp": 1.0170573,
      "epoch": 0.990229971441455,
      "flos": 15959025711360.0,
      "grad_norm": 2.722461169098648,
      "language_loss": 0.70679772,
      "learning_rate": 9.950925847685976e-10,
      "loss": 0.72778875,
      "num_input_tokens_seen": 355358835,
      "step": 16470,
      "time_per_iteration": 2.496119260787964
    },
    {
      "auxiliary_loss_clip": 0.01016609,
      "auxiliary_loss_mlp": 0.01002717,
      "balance_loss_clip": 1.00432324,
      "balance_loss_mlp": 1.00163865,
      "epoch": 0.990290094694123,
      "flos": 69780287911680.0,
      "grad_norm": 0.6652069418245136,
      "language_loss": 0.55515206,
      "learning_rate": 9.828464112755509e-10,
      "loss": 0.5753454,
      "num_input_tokens_seen": 355431225,
      "step": 16471,
      "time_per_iteration": 3.2695038318634033
    },
    {
      "auxiliary_loss_clip": 0.01082922,
      "auxiliary_loss_mlp": 0.01032008,
      "balance_loss_clip": 1.03538394,
      "balance_loss_mlp": 1.01997435,
      "epoch": 0.9903502179467909,
      "flos": 16252451913600.0,
      "grad_norm": 3.1709504476574777,
      "language_loss": 0.83570486,
      "learning_rate": 9.706760407131032e-10,
      "loss": 0.8568542,
      "num_input_tokens_seen": 355448250,
      "step": 16472,
      "time_per_iteration": 2.498241662979126
    },
    {
      "auxiliary_loss_clip": 0.01091973,
      "auxiliary_loss_mlp": 0.01024389,
      "balance_loss_clip": 1.03490067,
      "balance_loss_mlp": 1.01333272,
      "epoch": 0.9904103411994589,
      "flos": 21688393489920.0,
      "grad_norm": 1.9630346826471112,
      "language_loss": 0.85555786,
      "learning_rate": 9.585814735431075e-10,
      "loss": 0.8767215,
      "num_input_tokens_seen": 355467040,
      "step": 16473,
      "time_per_iteration": 2.4860990047454834
    },
    {
      "auxiliary_loss_clip": 0.01101117,
      "auxiliary_loss_mlp": 0.01029704,
      "balance_loss_clip": 1.03344619,
      "balance_loss_mlp": 1.0189395,
      "epoch": 0.9904704644521268,
      "flos": 25739440243200.0,
      "grad_norm": 1.6239831631023984,
      "language_loss": 0.84421873,
      "learning_rate": 9.465627102240859e-10,
      "loss": 0.86552697,
      "num_input_tokens_seen": 355487825,
      "step": 16474,
      "time_per_iteration": 2.4751710891723633
    },
    {
      "auxiliary_loss_clip": 0.01076373,
      "auxiliary_loss_mlp": 0.01035387,
      "balance_loss_clip": 1.03068423,
      "balance_loss_mlp": 1.02379394,
      "epoch": 0.9905305877047949,
      "flos": 21908346422400.0,
      "grad_norm": 1.7447494717227168,
      "language_loss": 0.76442587,
      "learning_rate": 9.346197512116738e-10,
      "loss": 0.78554344,
      "num_input_tokens_seen": 355507445,
      "step": 16475,
      "time_per_iteration": 2.5232841968536377
    },
    {
      "auxiliary_loss_clip": 0.01061262,
      "auxiliary_loss_mlp": 0.01033682,
      "balance_loss_clip": 1.03061485,
      "balance_loss_mlp": 1.02052748,
      "epoch": 0.9905907109574628,
      "flos": 21392417422080.0,
      "grad_norm": 1.6349725429082336,
      "language_loss": 0.75704241,
      "learning_rate": 9.227525969588423e-10,
      "loss": 0.77799183,
      "num_input_tokens_seen": 355527205,
      "step": 16476,
      "time_per_iteration": 2.5665366649627686
    },
    {
      "auxiliary_loss_clip": 0.01095728,
      "auxiliary_loss_mlp": 0.00785509,
      "balance_loss_clip": 1.0341146,
      "balance_loss_mlp": 1.01081657,
      "epoch": 0.9906508342101308,
      "flos": 20521620005760.0,
      "grad_norm": 1.9783925973808456,
      "language_loss": 0.67739749,
      "learning_rate": 9.109612479154538e-10,
      "loss": 0.69620991,
      "num_input_tokens_seen": 355544740,
      "step": 16477,
      "time_per_iteration": 2.501646041870117
    },
    {
      "auxiliary_loss_clip": 0.01087058,
      "auxiliary_loss_mlp": 0.01030933,
      "balance_loss_clip": 1.03516757,
      "balance_loss_mlp": 1.01829123,
      "epoch": 0.9907109574627987,
      "flos": 21361211481600.0,
      "grad_norm": 2.231345553149873,
      "language_loss": 0.71855807,
      "learning_rate": 8.992457045289282e-10,
      "loss": 0.73973799,
      "num_input_tokens_seen": 355564385,
      "step": 16478,
      "time_per_iteration": 2.5374014377593994
    },
    {
      "auxiliary_loss_clip": 0.01105373,
      "auxiliary_loss_mlp": 0.01037362,
      "balance_loss_clip": 1.03615808,
      "balance_loss_mlp": 1.02439249,
      "epoch": 0.9907710807154667,
      "flos": 17338605321600.0,
      "grad_norm": 2.068658948736616,
      "language_loss": 0.81030357,
      "learning_rate": 8.876059672433545e-10,
      "loss": 0.83173096,
      "num_input_tokens_seen": 355579260,
      "step": 16479,
      "time_per_iteration": 2.4339823722839355
    },
    {
      "auxiliary_loss_clip": 0.01094579,
      "auxiliary_loss_mlp": 0.01033638,
      "balance_loss_clip": 1.03484523,
      "balance_loss_mlp": 1.02206838,
      "epoch": 0.9908312039681346,
      "flos": 28621881918720.0,
      "grad_norm": 1.467783419177516,
      "language_loss": 0.66411066,
      "learning_rate": 8.760420364999355e-10,
      "loss": 0.6853928,
      "num_input_tokens_seen": 355599790,
      "step": 16480,
      "time_per_iteration": 2.5163028240203857
    },
    {
      "auxiliary_loss_clip": 0.01090598,
      "auxiliary_loss_mlp": 0.01029808,
      "balance_loss_clip": 1.03384936,
      "balance_loss_mlp": 1.01770854,
      "epoch": 0.9908913272208026,
      "flos": 35770654512000.0,
      "grad_norm": 2.654187102925779,
      "language_loss": 0.72217774,
      "learning_rate": 8.645539127374313e-10,
      "loss": 0.74338186,
      "num_input_tokens_seen": 355620925,
      "step": 16481,
      "time_per_iteration": 2.598353385925293
    },
    {
      "auxiliary_loss_clip": 0.01089912,
      "auxiliary_loss_mlp": 0.01022218,
      "balance_loss_clip": 1.03334725,
      "balance_loss_mlp": 1.01077998,
      "epoch": 0.9909514504734707,
      "flos": 19902196944000.0,
      "grad_norm": 1.7698561356912261,
      "language_loss": 0.77788526,
      "learning_rate": 8.531415963912713e-10,
      "loss": 0.79900658,
      "num_input_tokens_seen": 355639165,
      "step": 16482,
      "time_per_iteration": 2.447971820831299
    },
    {
      "auxiliary_loss_clip": 0.01094194,
      "auxiliary_loss_mlp": 0.01028414,
      "balance_loss_clip": 1.03377986,
      "balance_loss_mlp": 1.01663601,
      "epoch": 0.9910115737261386,
      "flos": 20004793165440.0,
      "grad_norm": 2.017782454029079,
      "language_loss": 0.75476611,
      "learning_rate": 8.418050878944427e-10,
      "loss": 0.77599216,
      "num_input_tokens_seen": 355657320,
      "step": 16483,
      "time_per_iteration": 2.467625141143799
    },
    {
      "auxiliary_loss_clip": 0.0101852,
      "auxiliary_loss_mlp": 0.01000424,
      "balance_loss_clip": 1.00456882,
      "balance_loss_mlp": 0.99923819,
      "epoch": 0.9910716969788066,
      "flos": 70688432494080.0,
      "grad_norm": 0.6723156978369148,
      "language_loss": 0.53661764,
      "learning_rate": 8.305443876768237e-10,
      "loss": 0.55680716,
      "num_input_tokens_seen": 355726370,
      "step": 16484,
      "time_per_iteration": 3.2017130851745605
    },
    {
      "auxiliary_loss_clip": 0.01097532,
      "auxiliary_loss_mlp": 0.01030051,
      "balance_loss_clip": 1.03313732,
      "balance_loss_mlp": 1.01847577,
      "epoch": 0.9911318202314745,
      "flos": 21434038306560.0,
      "grad_norm": 1.7676341789103007,
      "language_loss": 0.8181836,
      "learning_rate": 8.19359496165184e-10,
      "loss": 0.83945942,
      "num_input_tokens_seen": 355745840,
      "step": 16485,
      "time_per_iteration": 2.4401137828826904
    },
    {
      "auxiliary_loss_clip": 0.01061836,
      "auxiliary_loss_mlp": 0.01034429,
      "balance_loss_clip": 1.03164911,
      "balance_loss_mlp": 1.02154899,
      "epoch": 0.9911919434841425,
      "flos": 19826820253440.0,
      "grad_norm": 1.7141761542571903,
      "language_loss": 0.81553108,
      "learning_rate": 8.082504137836288e-10,
      "loss": 0.83649373,
      "num_input_tokens_seen": 355763385,
      "step": 16486,
      "time_per_iteration": 2.507490634918213
    },
    {
      "auxiliary_loss_clip": 0.01094298,
      "auxiliary_loss_mlp": 0.01029825,
      "balance_loss_clip": 1.03479803,
      "balance_loss_mlp": 1.01808274,
      "epoch": 0.9912520667368104,
      "flos": 41719364691840.0,
      "grad_norm": 1.4689591252091678,
      "language_loss": 0.66243786,
      "learning_rate": 7.972171409538209e-10,
      "loss": 0.6836791,
      "num_input_tokens_seen": 355786075,
      "step": 16487,
      "time_per_iteration": 2.6409401893615723
    },
    {
      "auxiliary_loss_clip": 0.01087912,
      "auxiliary_loss_mlp": 0.00779666,
      "balance_loss_clip": 1.03254092,
      "balance_loss_mlp": 1.00645256,
      "epoch": 0.9913121899894785,
      "flos": 23769668263680.0,
      "grad_norm": 1.5416857604176755,
      "language_loss": 0.76516861,
      "learning_rate": 7.862596780936481e-10,
      "loss": 0.78384441,
      "num_input_tokens_seen": 355806295,
      "step": 16488,
      "time_per_iteration": 2.5030152797698975
    },
    {
      "auxiliary_loss_clip": 0.01075645,
      "auxiliary_loss_mlp": 0.01026566,
      "balance_loss_clip": 1.03402138,
      "balance_loss_mlp": 1.01412606,
      "epoch": 0.9913723132421464,
      "flos": 23769668263680.0,
      "grad_norm": 4.078182394540024,
      "language_loss": 0.68831784,
      "learning_rate": 7.753780256190001e-10,
      "loss": 0.70933998,
      "num_input_tokens_seen": 355825730,
      "step": 16489,
      "time_per_iteration": 2.595344066619873
    },
    {
      "auxiliary_loss_clip": 0.00996913,
      "auxiliary_loss_mlp": 0.01003394,
      "balance_loss_clip": 1.00610209,
      "balance_loss_mlp": 1.00230968,
      "epoch": 0.9914324364948144,
      "flos": 71267419820160.0,
      "grad_norm": 0.6088265247852145,
      "language_loss": 0.52593273,
      "learning_rate": 7.645721839424357e-10,
      "loss": 0.54593581,
      "num_input_tokens_seen": 355891545,
      "step": 16490,
      "time_per_iteration": 3.247739315032959
    },
    {
      "auxiliary_loss_clip": 0.0107658,
      "auxiliary_loss_mlp": 0.0103737,
      "balance_loss_clip": 1.03413177,
      "balance_loss_mlp": 1.0234046,
      "epoch": 0.9914925597474823,
      "flos": 23695440808320.0,
      "grad_norm": 1.6517410641845605,
      "language_loss": 0.75405085,
      "learning_rate": 7.538421534734052e-10,
      "loss": 0.77519035,
      "num_input_tokens_seen": 355909920,
      "step": 16491,
      "time_per_iteration": 3.958491086959839
    },
    {
      "auxiliary_loss_clip": 0.0106208,
      "auxiliary_loss_mlp": 0.01032482,
      "balance_loss_clip": 1.03633142,
      "balance_loss_mlp": 1.02004218,
      "epoch": 0.9915526830001503,
      "flos": 13433822749440.0,
      "grad_norm": 2.062646050614676,
      "language_loss": 0.70512152,
      "learning_rate": 7.431879346191383e-10,
      "loss": 0.72606719,
      "num_input_tokens_seen": 355923130,
      "step": 16492,
      "time_per_iteration": 2.603111982345581
    },
    {
      "auxiliary_loss_clip": 0.01068069,
      "auxiliary_loss_mlp": 0.01031266,
      "balance_loss_clip": 1.03259706,
      "balance_loss_mlp": 1.01820624,
      "epoch": 0.9916128062528182,
      "flos": 20740962407040.0,
      "grad_norm": 2.0247022237389647,
      "language_loss": 0.68460029,
      "learning_rate": 7.326095277837563e-10,
      "loss": 0.70559371,
      "num_input_tokens_seen": 355941960,
      "step": 16493,
      "time_per_iteration": 2.545336961746216
    },
    {
      "auxiliary_loss_clip": 0.01077955,
      "auxiliary_loss_mlp": 0.01034518,
      "balance_loss_clip": 1.0350616,
      "balance_loss_mlp": 1.02237678,
      "epoch": 0.9916729295054862,
      "flos": 22487082353280.0,
      "grad_norm": 1.8309470418312739,
      "language_loss": 0.71162868,
      "learning_rate": 7.221069333678276e-10,
      "loss": 0.7327534,
      "num_input_tokens_seen": 355961640,
      "step": 16494,
      "time_per_iteration": 4.030745983123779
    },
    {
      "auxiliary_loss_clip": 0.01093387,
      "auxiliary_loss_mlp": 0.01031976,
      "balance_loss_clip": 1.03472149,
      "balance_loss_mlp": 1.01912546,
      "epoch": 0.9917330527581543,
      "flos": 14792467708800.0,
      "grad_norm": 2.365164181629142,
      "language_loss": 0.67867768,
      "learning_rate": 7.116801517701443e-10,
      "loss": 0.69993126,
      "num_input_tokens_seen": 355977980,
      "step": 16495,
      "time_per_iteration": 2.469764232635498
    },
    {
      "auxiliary_loss_clip": 0.01010026,
      "auxiliary_loss_mlp": 0.01000564,
      "balance_loss_clip": 1.0067091,
      "balance_loss_mlp": 0.99949133,
      "epoch": 0.9917931760108222,
      "flos": 59191595585280.0,
      "grad_norm": 0.7238204988516688,
      "language_loss": 0.53510422,
      "learning_rate": 7.013291833859458e-10,
      "loss": 0.55521011,
      "num_input_tokens_seen": 356042900,
      "step": 16496,
      "time_per_iteration": 3.2301788330078125
    },
    {
      "auxiliary_loss_clip": 0.01079426,
      "auxiliary_loss_mlp": 0.00785684,
      "balance_loss_clip": 1.03372121,
      "balance_loss_mlp": 1.00908518,
      "epoch": 0.9918532992634902,
      "flos": 26761637485440.0,
      "grad_norm": 14.174156197875808,
      "language_loss": 0.7158879,
      "learning_rate": 6.91054028607585e-10,
      "loss": 0.73453903,
      "num_input_tokens_seen": 356063000,
      "step": 16497,
      "time_per_iteration": 2.599478006362915
    },
    {
      "auxiliary_loss_clip": 0.01074501,
      "auxiliary_loss_mlp": 0.01029548,
      "balance_loss_clip": 1.03391826,
      "balance_loss_mlp": 1.01661968,
      "epoch": 0.9919134225161581,
      "flos": 14975719920000.0,
      "grad_norm": 2.245758768760357,
      "language_loss": 0.82269663,
      "learning_rate": 6.808546878249721e-10,
      "loss": 0.84373713,
      "num_input_tokens_seen": 356078130,
      "step": 16498,
      "time_per_iteration": 2.5504136085510254
    },
    {
      "auxiliary_loss_clip": 0.01069694,
      "auxiliary_loss_mlp": 0.01037587,
      "balance_loss_clip": 1.03642023,
      "balance_loss_mlp": 1.02506423,
      "epoch": 0.9919735457688261,
      "flos": 27818201064960.0,
      "grad_norm": 1.6163941954597265,
      "language_loss": 0.68231177,
      "learning_rate": 6.707311614246869e-10,
      "loss": 0.70338452,
      "num_input_tokens_seen": 356101655,
      "step": 16499,
      "time_per_iteration": 2.646113395690918
    },
    {
      "auxiliary_loss_clip": 0.01105652,
      "auxiliary_loss_mlp": 0.01026668,
      "balance_loss_clip": 1.03607833,
      "balance_loss_mlp": 1.01501524,
      "epoch": 0.992033669021494,
      "flos": 22562782266240.0,
      "grad_norm": 2.894901545538992,
      "language_loss": 0.82274354,
      "learning_rate": 6.606834497904223e-10,
      "loss": 0.84406674,
      "num_input_tokens_seen": 356121425,
      "step": 16500,
      "time_per_iteration": 2.456714391708374
    },
    {
      "auxiliary_loss_clip": 0.0108401,
      "auxiliary_loss_mlp": 0.01030583,
      "balance_loss_clip": 1.0353694,
      "balance_loss_mlp": 1.01810753,
      "epoch": 0.9920937922741621,
      "flos": 25374587846400.0,
      "grad_norm": 1.8549351730428476,
      "language_loss": 0.82064319,
      "learning_rate": 6.507115533036511e-10,
      "loss": 0.84178913,
      "num_input_tokens_seen": 356140710,
      "step": 16501,
      "time_per_iteration": 2.5550851821899414
    },
    {
      "auxiliary_loss_clip": 0.0109538,
      "auxiliary_loss_mlp": 0.01028569,
      "balance_loss_clip": 1.03569913,
      "balance_loss_mlp": 1.01625454,
      "epoch": 0.99215391552683,
      "flos": 22054466949120.0,
      "grad_norm": 1.868956153883237,
      "language_loss": 0.76868397,
      "learning_rate": 6.408154723420711e-10,
      "loss": 0.78992343,
      "num_input_tokens_seen": 356159835,
      "step": 16502,
      "time_per_iteration": 2.483881711959839
    },
    {
      "auxiliary_loss_clip": 0.01078233,
      "auxiliary_loss_mlp": 0.01030909,
      "balance_loss_clip": 1.03368258,
      "balance_loss_mlp": 1.01760495,
      "epoch": 0.992214038779498,
      "flos": 15413937845760.0,
      "grad_norm": 2.083895387085456,
      "language_loss": 0.71414047,
      "learning_rate": 6.309952072811597e-10,
      "loss": 0.73523188,
      "num_input_tokens_seen": 356177555,
      "step": 16503,
      "time_per_iteration": 2.4898693561553955
    },
    {
      "auxiliary_loss_clip": 0.01020338,
      "auxiliary_loss_mlp": 0.01010749,
      "balance_loss_clip": 1.0078516,
      "balance_loss_mlp": 1.00944924,
      "epoch": 0.9922741620321659,
      "flos": 62014498467840.0,
      "grad_norm": 0.6523752280088663,
      "language_loss": 0.55111074,
      "learning_rate": 6.212507584932858e-10,
      "loss": 0.57142162,
      "num_input_tokens_seen": 356244975,
      "step": 16504,
      "time_per_iteration": 4.587971448898315
    },
    {
      "auxiliary_loss_clip": 0.01070465,
      "auxiliary_loss_mlp": 0.01025453,
      "balance_loss_clip": 1.03311455,
      "balance_loss_mlp": 1.01460552,
      "epoch": 0.9923342852848339,
      "flos": 17165480745600.0,
      "grad_norm": 1.809670402358836,
      "language_loss": 0.69274271,
      "learning_rate": 6.115821263481536e-10,
      "loss": 0.71370184,
      "num_input_tokens_seen": 356262605,
      "step": 16505,
      "time_per_iteration": 2.549145460128784
    },
    {
      "auxiliary_loss_clip": 0.01067792,
      "auxiliary_loss_mlp": 0.01028637,
      "balance_loss_clip": 1.03285122,
      "balance_loss_mlp": 1.01542306,
      "epoch": 0.9923944085375018,
      "flos": 23183210908800.0,
      "grad_norm": 2.369216998845462,
      "language_loss": 0.6553821,
      "learning_rate": 6.019893112119146e-10,
      "loss": 0.6763463,
      "num_input_tokens_seen": 356278935,
      "step": 16506,
      "time_per_iteration": 2.562612771987915
    },
    {
      "auxiliary_loss_clip": 0.01040049,
      "auxiliary_loss_mlp": 0.0102977,
      "balance_loss_clip": 1.03287995,
      "balance_loss_mlp": 1.01713991,
      "epoch": 0.9924545317901698,
      "flos": 20813861059200.0,
      "grad_norm": 1.7830816531924956,
      "language_loss": 0.63135743,
      "learning_rate": 5.924723134487219e-10,
      "loss": 0.65205562,
      "num_input_tokens_seen": 356295675,
      "step": 16507,
      "time_per_iteration": 2.58758282661438
    },
    {
      "auxiliary_loss_clip": 0.0110324,
      "auxiliary_loss_mlp": 0.01030028,
      "balance_loss_clip": 1.03425431,
      "balance_loss_mlp": 1.0182265,
      "epoch": 0.9925146550428379,
      "flos": 20083437993600.0,
      "grad_norm": 2.2522685186945486,
      "language_loss": 0.72900707,
      "learning_rate": 5.830311334193983e-10,
      "loss": 0.75033969,
      "num_input_tokens_seen": 356312885,
      "step": 16508,
      "time_per_iteration": 2.427483081817627
    },
    {
      "auxiliary_loss_clip": 0.01103195,
      "auxiliary_loss_mlp": 0.01028799,
      "balance_loss_clip": 1.03418183,
      "balance_loss_mlp": 1.0165205,
      "epoch": 0.9925747782955058,
      "flos": 24973717086720.0,
      "grad_norm": 1.521504066835369,
      "language_loss": 0.70677674,
      "learning_rate": 5.736657714818793e-10,
      "loss": 0.72809666,
      "num_input_tokens_seen": 356334070,
      "step": 16509,
      "time_per_iteration": 2.4735260009765625
    },
    {
      "auxiliary_loss_clip": 0.01092442,
      "auxiliary_loss_mlp": 0.01033809,
      "balance_loss_clip": 1.03302646,
      "balance_loss_mlp": 1.02187645,
      "epoch": 0.9926349015481738,
      "flos": 60472526492160.0,
      "grad_norm": 1.5847121218277973,
      "language_loss": 0.68680686,
      "learning_rate": 5.643762279912146e-10,
      "loss": 0.70806938,
      "num_input_tokens_seen": 356359410,
      "step": 16510,
      "time_per_iteration": 2.8422091007232666
    },
    {
      "auxiliary_loss_clip": 0.01070211,
      "auxiliary_loss_mlp": 0.01030973,
      "balance_loss_clip": 1.03422379,
      "balance_loss_mlp": 1.01877785,
      "epoch": 0.9926950248008417,
      "flos": 20741716592640.0,
      "grad_norm": 2.1300847986677023,
      "language_loss": 0.81129414,
      "learning_rate": 5.551625032997886e-10,
      "loss": 0.83230603,
      "num_input_tokens_seen": 356378345,
      "step": 16511,
      "time_per_iteration": 2.5496416091918945
    },
    {
      "auxiliary_loss_clip": 0.01059309,
      "auxiliary_loss_mlp": 0.01033809,
      "balance_loss_clip": 1.03138828,
      "balance_loss_mlp": 1.02213871,
      "epoch": 0.9927551480535097,
      "flos": 24352965221760.0,
      "grad_norm": 1.8714666594477027,
      "language_loss": 0.91973662,
      "learning_rate": 5.460245977570998e-10,
      "loss": 0.94066775,
      "num_input_tokens_seen": 356397345,
      "step": 16512,
      "time_per_iteration": 2.5928525924682617
    },
    {
      "auxiliary_loss_clip": 0.01000766,
      "auxiliary_loss_mlp": 0.01001226,
      "balance_loss_clip": 1.00819206,
      "balance_loss_mlp": 1.00024819,
      "epoch": 0.9928152713061776,
      "flos": 71275572207360.0,
      "grad_norm": 0.7298586358038401,
      "language_loss": 0.55166966,
      "learning_rate": 5.369625117095378e-10,
      "loss": 0.57168961,
      "num_input_tokens_seen": 356459160,
      "step": 16513,
      "time_per_iteration": 3.2705729007720947
    },
    {
      "auxiliary_loss_clip": 0.01074362,
      "auxiliary_loss_mlp": 0.0102818,
      "balance_loss_clip": 1.03301585,
      "balance_loss_mlp": 1.01598454,
      "epoch": 0.9928753945588457,
      "flos": 57809499045120.0,
      "grad_norm": 1.3707993044343254,
      "language_loss": 0.65080255,
      "learning_rate": 5.279762455006054e-10,
      "loss": 0.67182797,
      "num_input_tokens_seen": 356486405,
      "step": 16514,
      "time_per_iteration": 2.8496253490448
    },
    {
      "auxiliary_loss_clip": 0.01073355,
      "auxiliary_loss_mlp": 0.01028281,
      "balance_loss_clip": 1.03300953,
      "balance_loss_mlp": 1.01508462,
      "epoch": 0.9929355178115136,
      "flos": 19568981450880.0,
      "grad_norm": 2.0759117057834064,
      "language_loss": 0.73188823,
      "learning_rate": 5.190657994713632e-10,
      "loss": 0.75290459,
      "num_input_tokens_seen": 356502905,
      "step": 16515,
      "time_per_iteration": 2.5331003665924072
    },
    {
      "auxiliary_loss_clip": 0.01061912,
      "auxiliary_loss_mlp": 0.01036988,
      "balance_loss_clip": 1.03355086,
      "balance_loss_mlp": 1.02390456,
      "epoch": 0.9929956410641816,
      "flos": 22964658606720.0,
      "grad_norm": 1.4095818371979523,
      "language_loss": 0.77319306,
      "learning_rate": 5.102311739593191e-10,
      "loss": 0.79418206,
      "num_input_tokens_seen": 356523830,
      "step": 16516,
      "time_per_iteration": 2.596958875656128
    },
    {
      "auxiliary_loss_clip": 0.01070182,
      "auxiliary_loss_mlp": 0.01028036,
      "balance_loss_clip": 1.03244185,
      "balance_loss_mlp": 1.01669943,
      "epoch": 0.9930557643168495,
      "flos": 22566409539840.0,
      "grad_norm": 1.3491478864742905,
      "language_loss": 0.77997667,
      "learning_rate": 5.014723692997602e-10,
      "loss": 0.80095881,
      "num_input_tokens_seen": 356543965,
      "step": 16517,
      "time_per_iteration": 2.6000726222991943
    },
    {
      "auxiliary_loss_clip": 0.01091684,
      "auxiliary_loss_mlp": 0.01038061,
      "balance_loss_clip": 1.03625059,
      "balance_loss_mlp": 1.02417874,
      "epoch": 0.9931158875695175,
      "flos": 17201032231680.0,
      "grad_norm": 2.293092785455623,
      "language_loss": 0.66967887,
      "learning_rate": 4.927893858248655e-10,
      "loss": 0.69097626,
      "num_input_tokens_seen": 356561530,
      "step": 16518,
      "time_per_iteration": 2.4797003269195557
    },
    {
      "auxiliary_loss_clip": 0.01012093,
      "auxiliary_loss_mlp": 0.00998223,
      "balance_loss_clip": 1.01300466,
      "balance_loss_mlp": 0.99704885,
      "epoch": 0.9931760108221854,
      "flos": 63711204278400.0,
      "grad_norm": 0.7268922857060319,
      "language_loss": 0.53449887,
      "learning_rate": 4.84182223863483e-10,
      "loss": 0.55460203,
      "num_input_tokens_seen": 356616845,
      "step": 16519,
      "time_per_iteration": 3.0249719619750977
    },
    {
      "auxiliary_loss_clip": 0.01060592,
      "auxiliary_loss_mlp": 0.01038487,
      "balance_loss_clip": 1.03191447,
      "balance_loss_mlp": 1.02571964,
      "epoch": 0.9932361340748534,
      "flos": 15304805349120.0,
      "grad_norm": 1.6261714503595328,
      "language_loss": 0.6011889,
      "learning_rate": 4.756508837426842e-10,
      "loss": 0.62217969,
      "num_input_tokens_seen": 356633560,
      "step": 16520,
      "time_per_iteration": 2.618563413619995
    },
    {
      "auxiliary_loss_clip": 0.0108181,
      "auxiliary_loss_mlp": 0.01031372,
      "balance_loss_clip": 1.0350579,
      "balance_loss_mlp": 1.01928449,
      "epoch": 0.9932962573275215,
      "flos": 36064906727040.0,
      "grad_norm": 1.658699932415915,
      "language_loss": 0.61963224,
      "learning_rate": 4.671953657853223e-10,
      "loss": 0.64076406,
      "num_input_tokens_seen": 356657600,
      "step": 16521,
      "time_per_iteration": 2.6613929271698
    },
    {
      "auxiliary_loss_clip": 0.01081402,
      "auxiliary_loss_mlp": 0.01031535,
      "balance_loss_clip": 1.03622448,
      "balance_loss_mlp": 1.01891065,
      "epoch": 0.9933563805801894,
      "flos": 21470523546240.0,
      "grad_norm": 2.943643603396264,
      "language_loss": 0.74303418,
      "learning_rate": 4.5881567031225145e-10,
      "loss": 0.76416349,
      "num_input_tokens_seen": 356675880,
      "step": 16522,
      "time_per_iteration": 2.5161147117614746
    },
    {
      "auxiliary_loss_clip": 0.01069167,
      "auxiliary_loss_mlp": 0.01030594,
      "balance_loss_clip": 1.03369236,
      "balance_loss_mlp": 1.01903653,
      "epoch": 0.9934165038328574,
      "flos": 23986532626560.0,
      "grad_norm": 1.597211862440945,
      "language_loss": 0.73218155,
      "learning_rate": 4.5051179764143964e-10,
      "loss": 0.75317913,
      "num_input_tokens_seen": 356696000,
      "step": 16523,
      "time_per_iteration": 2.5879104137420654
    },
    {
      "auxiliary_loss_clip": 0.01080731,
      "auxiliary_loss_mlp": 0.00782354,
      "balance_loss_clip": 1.03275061,
      "balance_loss_mlp": 1.00671422,
      "epoch": 0.9934766270855253,
      "flos": 21907807718400.0,
      "grad_norm": 1.54468569373802,
      "language_loss": 0.71182275,
      "learning_rate": 4.422837480875241e-10,
      "loss": 0.73045349,
      "num_input_tokens_seen": 356716845,
      "step": 16524,
      "time_per_iteration": 2.5271873474121094
    },
    {
      "auxiliary_loss_clip": 0.01070159,
      "auxiliary_loss_mlp": 0.01030984,
      "balance_loss_clip": 1.03340364,
      "balance_loss_mlp": 1.01978469,
      "epoch": 0.9935367503381933,
      "flos": 17129139160320.0,
      "grad_norm": 1.8006544215756555,
      "language_loss": 0.79282951,
      "learning_rate": 4.341315219624775e-10,
      "loss": 0.81384099,
      "num_input_tokens_seen": 356732100,
      "step": 16525,
      "time_per_iteration": 2.547050714492798
    },
    {
      "auxiliary_loss_clip": 0.01067077,
      "auxiliary_loss_mlp": 0.01025851,
      "balance_loss_clip": 1.03583694,
      "balance_loss_mlp": 1.01452017,
      "epoch": 0.9935968735908612,
      "flos": 22346241125760.0,
      "grad_norm": 2.0926658357511636,
      "language_loss": 0.74911249,
      "learning_rate": 4.2605511957582995e-10,
      "loss": 0.77004176,
      "num_input_tokens_seen": 356751480,
      "step": 16526,
      "time_per_iteration": 2.5579802989959717
    },
    {
      "auxiliary_loss_clip": 0.01099194,
      "auxiliary_loss_mlp": 0.00781718,
      "balance_loss_clip": 1.03287685,
      "balance_loss_mlp": 1.0079385,
      "epoch": 0.9936569968435293,
      "flos": 29460539640960.0,
      "grad_norm": 1.4241634092991735,
      "language_loss": 0.72501278,
      "learning_rate": 4.180545412333369e-10,
      "loss": 0.74382192,
      "num_input_tokens_seen": 356772650,
      "step": 16527,
      "time_per_iteration": 2.519413948059082
    },
    {
      "auxiliary_loss_clip": 0.01082212,
      "auxiliary_loss_mlp": 0.01027084,
      "balance_loss_clip": 1.03394794,
      "balance_loss_mlp": 1.0149188,
      "epoch": 0.9937171200961972,
      "flos": 16544046522240.0,
      "grad_norm": 2.2200401480973113,
      "language_loss": 0.75956744,
      "learning_rate": 4.1012978723875547e-10,
      "loss": 0.78066039,
      "num_input_tokens_seen": 356788510,
      "step": 16528,
      "time_per_iteration": 2.4676499366760254
    },
    {
      "auxiliary_loss_clip": 0.01084365,
      "auxiliary_loss_mlp": 0.01028226,
      "balance_loss_clip": 1.0332973,
      "balance_loss_mlp": 1.01506555,
      "epoch": 0.9937772433488652,
      "flos": 24390276474240.0,
      "grad_norm": 2.329084625622962,
      "language_loss": 0.68282348,
      "learning_rate": 4.022808578922898e-10,
      "loss": 0.70394933,
      "num_input_tokens_seen": 356809115,
      "step": 16529,
      "time_per_iteration": 2.5335917472839355
    },
    {
      "auxiliary_loss_clip": 0.01097568,
      "auxiliary_loss_mlp": 0.0103535,
      "balance_loss_clip": 1.03628123,
      "balance_loss_mlp": 1.02142,
      "epoch": 0.9938373666015331,
      "flos": 15669909141120.0,
      "grad_norm": 2.5015567234408076,
      "language_loss": 0.65255225,
      "learning_rate": 3.9450775349170186e-10,
      "loss": 0.67388147,
      "num_input_tokens_seen": 356826410,
      "step": 16530,
      "time_per_iteration": 3.881117582321167
    },
    {
      "auxiliary_loss_clip": 0.01088059,
      "auxiliary_loss_mlp": 0.01029329,
      "balance_loss_clip": 1.03336143,
      "balance_loss_mlp": 1.01787865,
      "epoch": 0.9938974898542011,
      "flos": 19496190539520.0,
      "grad_norm": 3.051281854558218,
      "language_loss": 0.71272612,
      "learning_rate": 3.8681047433186676e-10,
      "loss": 0.73390001,
      "num_input_tokens_seen": 356844990,
      "step": 16531,
      "time_per_iteration": 2.457932472229004
    },
    {
      "auxiliary_loss_clip": 0.01089692,
      "auxiliary_loss_mlp": 0.01030315,
      "balance_loss_clip": 1.03405285,
      "balance_loss_mlp": 1.0176487,
      "epoch": 0.993957613106869,
      "flos": 26906896085760.0,
      "grad_norm": 5.058186284074813,
      "language_loss": 0.74242675,
      "learning_rate": 3.791890207045512e-10,
      "loss": 0.76362675,
      "num_input_tokens_seen": 356866530,
      "step": 16532,
      "time_per_iteration": 3.9516167640686035
    },
    {
      "auxiliary_loss_clip": 0.01047322,
      "auxiliary_loss_mlp": 0.01028569,
      "balance_loss_clip": 1.03156555,
      "balance_loss_mlp": 1.01765585,
      "epoch": 0.994017736359537,
      "flos": 14939593816320.0,
      "grad_norm": 1.5860605229242386,
      "language_loss": 0.70599902,
      "learning_rate": 3.7164339289885717e-10,
      "loss": 0.72675788,
      "num_input_tokens_seen": 356884660,
      "step": 16533,
      "time_per_iteration": 3.941800355911255
    },
    {
      "auxiliary_loss_clip": 0.01092941,
      "auxiliary_loss_mlp": 0.01028439,
      "balance_loss_clip": 1.03686011,
      "balance_loss_mlp": 1.01622629,
      "epoch": 0.9940778596122051,
      "flos": 15377883569280.0,
      "grad_norm": 1.8745646908565194,
      "language_loss": 0.84079063,
      "learning_rate": 3.641735912007782e-10,
      "loss": 0.86200446,
      "num_input_tokens_seen": 356900895,
      "step": 16534,
      "time_per_iteration": 2.450599193572998
    },
    {
      "auxiliary_loss_clip": 0.01062996,
      "auxiliary_loss_mlp": 0.01029473,
      "balance_loss_clip": 1.03182793,
      "balance_loss_mlp": 1.01826739,
      "epoch": 0.994137982864873,
      "flos": 25228108183680.0,
      "grad_norm": 1.3393551761619618,
      "language_loss": 0.65768456,
      "learning_rate": 3.567796158934211e-10,
      "loss": 0.67860925,
      "num_input_tokens_seen": 356920985,
      "step": 16535,
      "time_per_iteration": 2.6040236949920654
    },
    {
      "auxiliary_loss_clip": 0.01063872,
      "auxiliary_loss_mlp": 0.01026454,
      "balance_loss_clip": 1.03557253,
      "balance_loss_mlp": 1.01566505,
      "epoch": 0.994198106117541,
      "flos": 18442140912000.0,
      "grad_norm": 1.5860370388134188,
      "language_loss": 0.64901984,
      "learning_rate": 3.4946146725767235e-10,
      "loss": 0.66992313,
      "num_input_tokens_seen": 356939800,
      "step": 16536,
      "time_per_iteration": 2.5361204147338867
    },
    {
      "auxiliary_loss_clip": 0.01065204,
      "auxiliary_loss_mlp": 0.01038445,
      "balance_loss_clip": 1.03262329,
      "balance_loss_mlp": 1.02441382,
      "epoch": 0.9942582293702089,
      "flos": 16654112772480.0,
      "grad_norm": 1.7937160985399248,
      "language_loss": 0.78449738,
      "learning_rate": 3.4221914557064357e-10,
      "loss": 0.80553389,
      "num_input_tokens_seen": 356957780,
      "step": 16537,
      "time_per_iteration": 2.530407667160034
    },
    {
      "auxiliary_loss_clip": 0.01098663,
      "auxiliary_loss_mlp": 0.01032481,
      "balance_loss_clip": 1.03510857,
      "balance_loss_mlp": 1.01936817,
      "epoch": 0.9943183526228769,
      "flos": 21944580266880.0,
      "grad_norm": 1.4517666942224632,
      "language_loss": 0.68918395,
      "learning_rate": 3.35052651107004e-10,
      "loss": 0.71049547,
      "num_input_tokens_seen": 356979185,
      "step": 16538,
      "time_per_iteration": 2.508413553237915
    },
    {
      "auxiliary_loss_clip": 0.01061838,
      "auxiliary_loss_mlp": 0.01032752,
      "balance_loss_clip": 1.03091431,
      "balance_loss_mlp": 1.02065182,
      "epoch": 0.9943784758755448,
      "flos": 23842566915840.0,
      "grad_norm": 1.9884840077714971,
      "language_loss": 0.74848974,
      "learning_rate": 3.2796198413853614e-10,
      "loss": 0.76943558,
      "num_input_tokens_seen": 356997735,
      "step": 16539,
      "time_per_iteration": 2.573502540588379
    },
    {
      "auxiliary_loss_clip": 0.01054238,
      "auxiliary_loss_mlp": 0.01031955,
      "balance_loss_clip": 1.03353834,
      "balance_loss_mlp": 1.01947427,
      "epoch": 0.9944385991282129,
      "flos": 21469984842240.0,
      "grad_norm": 2.061185529639008,
      "language_loss": 0.70490974,
      "learning_rate": 3.209471449341361e-10,
      "loss": 0.72577167,
      "num_input_tokens_seen": 357015660,
      "step": 16540,
      "time_per_iteration": 2.5730717182159424
    },
    {
      "auxiliary_loss_clip": 0.01088427,
      "auxiliary_loss_mlp": 0.01027249,
      "balance_loss_clip": 1.03369987,
      "balance_loss_mlp": 1.01658022,
      "epoch": 0.9944987223808808,
      "flos": 22927024131840.0,
      "grad_norm": 1.9090727514975214,
      "language_loss": 0.75384074,
      "learning_rate": 3.140081337600353e-10,
      "loss": 0.77499747,
      "num_input_tokens_seen": 357034800,
      "step": 16541,
      "time_per_iteration": 2.503122329711914
    },
    {
      "auxiliary_loss_clip": 0.01082008,
      "auxiliary_loss_mlp": 0.01032749,
      "balance_loss_clip": 1.03300536,
      "balance_loss_mlp": 1.02086413,
      "epoch": 0.9945588456335488,
      "flos": 22383013674240.0,
      "grad_norm": 1.7875042268377968,
      "language_loss": 0.76624858,
      "learning_rate": 3.0714495087891255e-10,
      "loss": 0.78739613,
      "num_input_tokens_seen": 357053785,
      "step": 16542,
      "time_per_iteration": 3.8983395099639893
    },
    {
      "auxiliary_loss_clip": 0.01093371,
      "auxiliary_loss_mlp": 0.0102875,
      "balance_loss_clip": 1.03423238,
      "balance_loss_mlp": 1.01635194,
      "epoch": 0.9946189688862167,
      "flos": 21397517153280.0,
      "grad_norm": 2.1037959053843123,
      "language_loss": 0.74649221,
      "learning_rate": 3.0035759655122615e-10,
      "loss": 0.76771343,
      "num_input_tokens_seen": 357072025,
      "step": 16543,
      "time_per_iteration": 2.466060161590576
    },
    {
      "auxiliary_loss_clip": 0.0108971,
      "auxiliary_loss_mlp": 0.01032345,
      "balance_loss_clip": 1.03335607,
      "balance_loss_mlp": 1.01916051,
      "epoch": 0.9946790921388847,
      "flos": 12416545670400.0,
      "grad_norm": 2.314513467839334,
      "language_loss": 0.81457132,
      "learning_rate": 2.9364607103454785e-10,
      "loss": 0.83579195,
      "num_input_tokens_seen": 357086960,
      "step": 16544,
      "time_per_iteration": 2.47878360748291
    },
    {
      "auxiliary_loss_clip": 0.0110262,
      "auxiliary_loss_mlp": 0.01027475,
      "balance_loss_clip": 1.03392172,
      "balance_loss_mlp": 1.01545882,
      "epoch": 0.9947392153915526,
      "flos": 19058295836160.0,
      "grad_norm": 2.004681638815712,
      "language_loss": 0.78613293,
      "learning_rate": 2.870103745831187e-10,
      "loss": 0.80743384,
      "num_input_tokens_seen": 357105095,
      "step": 16545,
      "time_per_iteration": 2.4189581871032715
    },
    {
      "auxiliary_loss_clip": 0.01074368,
      "auxiliary_loss_mlp": 0.01029436,
      "balance_loss_clip": 1.03380406,
      "balance_loss_mlp": 1.01706195,
      "epoch": 0.9947993386442207,
      "flos": 27308808339840.0,
      "grad_norm": 1.73062405663195,
      "language_loss": 0.72542286,
      "learning_rate": 2.8045050744873733e-10,
      "loss": 0.74646086,
      "num_input_tokens_seen": 357125065,
      "step": 16546,
      "time_per_iteration": 2.596069812774658
    },
    {
      "auxiliary_loss_clip": 0.01089321,
      "auxiliary_loss_mlp": 0.01034236,
      "balance_loss_clip": 1.03297257,
      "balance_loss_mlp": 1.02285123,
      "epoch": 0.9948594618968887,
      "flos": 20806498771200.0,
      "grad_norm": 2.961944051125757,
      "language_loss": 0.77560139,
      "learning_rate": 2.739664698798716e-10,
      "loss": 0.79683697,
      "num_input_tokens_seen": 357141600,
      "step": 16547,
      "time_per_iteration": 2.457759141921997
    },
    {
      "auxiliary_loss_clip": 0.01080764,
      "auxiliary_loss_mlp": 0.01029157,
      "balance_loss_clip": 1.03227794,
      "balance_loss_mlp": 1.01768351,
      "epoch": 0.9949195851495566,
      "flos": 23292953936640.0,
      "grad_norm": 2.0610085199600543,
      "language_loss": 0.69844037,
      "learning_rate": 2.67558262122769e-10,
      "loss": 0.71953958,
      "num_input_tokens_seen": 357157880,
      "step": 16548,
      "time_per_iteration": 2.5126731395721436
    },
    {
      "auxiliary_loss_clip": 0.01090726,
      "auxiliary_loss_mlp": 0.01032923,
      "balance_loss_clip": 1.03514898,
      "balance_loss_mlp": 1.02087736,
      "epoch": 0.9949797084022246,
      "flos": 18515470527360.0,
      "grad_norm": 3.3184660994876185,
      "language_loss": 0.75579756,
      "learning_rate": 2.6122588442012427e-10,
      "loss": 0.77703398,
      "num_input_tokens_seen": 357176705,
      "step": 16549,
      "time_per_iteration": 2.440439462661743
    },
    {
      "auxiliary_loss_clip": 0.01075755,
      "auxiliary_loss_mlp": 0.01033742,
      "balance_loss_clip": 1.03515255,
      "balance_loss_mlp": 1.02101612,
      "epoch": 0.9950398316548925,
      "flos": 30407719328640.0,
      "grad_norm": 1.7179432323231127,
      "language_loss": 0.7461971,
      "learning_rate": 2.5496933701241177e-10,
      "loss": 0.76729202,
      "num_input_tokens_seen": 357197630,
      "step": 16550,
      "time_per_iteration": 2.6116130352020264
    },
    {
      "auxiliary_loss_clip": 0.01053531,
      "auxiliary_loss_mlp": 0.00782083,
      "balance_loss_clip": 1.03260791,
      "balance_loss_mlp": 1.00646257,
      "epoch": 0.9950999549075605,
      "flos": 19900868140800.0,
      "grad_norm": 1.418495948830359,
      "language_loss": 0.77835631,
      "learning_rate": 2.4878862013655297e-10,
      "loss": 0.79671252,
      "num_input_tokens_seen": 357215445,
      "step": 16551,
      "time_per_iteration": 2.5817551612854004
    },
    {
      "auxiliary_loss_clip": 0.01085435,
      "auxiliary_loss_mlp": 0.0103042,
      "balance_loss_clip": 1.03299546,
      "balance_loss_mlp": 1.02007842,
      "epoch": 0.9951600781602284,
      "flos": 17603555016960.0,
      "grad_norm": 1.392059953026195,
      "language_loss": 0.667068,
      "learning_rate": 2.426837340270271e-10,
      "loss": 0.68822658,
      "num_input_tokens_seen": 357234285,
      "step": 16552,
      "time_per_iteration": 2.4755401611328125
    },
    {
      "auxiliary_loss_clip": 0.01102879,
      "auxiliary_loss_mlp": 0.01029258,
      "balance_loss_clip": 1.03377008,
      "balance_loss_mlp": 1.01717639,
      "epoch": 0.9952202014128965,
      "flos": 28950715952640.0,
      "grad_norm": 1.3966898960990746,
      "language_loss": 0.81700343,
      "learning_rate": 2.3665467891520465e-10,
      "loss": 0.83832479,
      "num_input_tokens_seen": 357257565,
      "step": 16553,
      "time_per_iteration": 2.5164520740509033
    },
    {
      "auxiliary_loss_clip": 0.01016764,
      "auxiliary_loss_mlp": 0.0100186,
      "balance_loss_clip": 1.00576913,
      "balance_loss_mlp": 1.00079942,
      "epoch": 0.9952803246655644,
      "flos": 70810386145920.0,
      "grad_norm": 0.7205071741812054,
      "language_loss": 0.5732578,
      "learning_rate": 2.3070145503001348e-10,
      "loss": 0.59344411,
      "num_input_tokens_seen": 357320205,
      "step": 16554,
      "time_per_iteration": 3.1861236095428467
    },
    {
      "auxiliary_loss_clip": 0.01096351,
      "auxiliary_loss_mlp": 0.01033499,
      "balance_loss_clip": 1.036394,
      "balance_loss_mlp": 1.02195406,
      "epoch": 0.9953404479182324,
      "flos": 21799070271360.0,
      "grad_norm": 1.5499778392395942,
      "language_loss": 0.769647,
      "learning_rate": 2.24824062597051e-10,
      "loss": 0.79094553,
      "num_input_tokens_seen": 357340695,
      "step": 16555,
      "time_per_iteration": 2.4751126766204834
    },
    {
      "auxiliary_loss_clip": 0.0107133,
      "auxiliary_loss_mlp": 0.01031311,
      "balance_loss_clip": 1.03114152,
      "balance_loss_mlp": 1.01893139,
      "epoch": 0.9954005711709003,
      "flos": 21937397546880.0,
      "grad_norm": 2.524991524125413,
      "language_loss": 0.86389595,
      "learning_rate": 2.1902250183902793e-10,
      "loss": 0.88492233,
      "num_input_tokens_seen": 357357505,
      "step": 16556,
      "time_per_iteration": 2.564124345779419
    },
    {
      "auxiliary_loss_clip": 0.01058307,
      "auxiliary_loss_mlp": 0.01030147,
      "balance_loss_clip": 1.03381395,
      "balance_loss_mlp": 1.01840448,
      "epoch": 0.9954606944235683,
      "flos": 19354559212800.0,
      "grad_norm": 1.6172949342302496,
      "language_loss": 0.7314136,
      "learning_rate": 2.132967729762125e-10,
      "loss": 0.75229818,
      "num_input_tokens_seen": 357375395,
      "step": 16557,
      "time_per_iteration": 2.5364925861358643
    },
    {
      "auxiliary_loss_clip": 0.01091269,
      "auxiliary_loss_mlp": 0.01032344,
      "balance_loss_clip": 1.03422976,
      "balance_loss_mlp": 1.02107239,
      "epoch": 0.9955208176762362,
      "flos": 30518611591680.0,
      "grad_norm": 2.1456111343934228,
      "language_loss": 0.76454818,
      "learning_rate": 2.0764687622554233e-10,
      "loss": 0.78578424,
      "num_input_tokens_seen": 357397375,
      "step": 16558,
      "time_per_iteration": 2.57104229927063
    },
    {
      "auxiliary_loss_clip": 0.01078405,
      "auxiliary_loss_mlp": 0.0102996,
      "balance_loss_clip": 1.03388822,
      "balance_loss_mlp": 1.01768112,
      "epoch": 0.9955809409289043,
      "flos": 30008249199360.0,
      "grad_norm": 2.4592762101287247,
      "language_loss": 0.63552135,
      "learning_rate": 2.0207281180129044e-10,
      "loss": 0.65660501,
      "num_input_tokens_seen": 357418880,
      "step": 16559,
      "time_per_iteration": 2.565115213394165
    },
    {
      "auxiliary_loss_clip": 0.01089435,
      "auxiliary_loss_mlp": 0.01028039,
      "balance_loss_clip": 1.0334903,
      "balance_loss_mlp": 1.01592791,
      "epoch": 0.9956410641815723,
      "flos": 21543278544000.0,
      "grad_norm": 2.195594746129208,
      "language_loss": 0.74445486,
      "learning_rate": 1.965745799148433e-10,
      "loss": 0.76562965,
      "num_input_tokens_seen": 357438310,
      "step": 16560,
      "time_per_iteration": 2.4962832927703857
    },
    {
      "auxiliary_loss_clip": 0.01045919,
      "auxiliary_loss_mlp": 0.01028199,
      "balance_loss_clip": 1.0310446,
      "balance_loss_mlp": 1.01669574,
      "epoch": 0.9957011874342402,
      "flos": 21689470897920.0,
      "grad_norm": 1.7128798437871364,
      "language_loss": 0.79205585,
      "learning_rate": 1.9115218077470073e-10,
      "loss": 0.81279701,
      "num_input_tokens_seen": 357457155,
      "step": 16561,
      "time_per_iteration": 2.5994818210601807
    },
    {
      "auxiliary_loss_clip": 0.01100625,
      "auxiliary_loss_mlp": 0.01028825,
      "balance_loss_clip": 1.03529871,
      "balance_loss_mlp": 1.01794124,
      "epoch": 0.9957613106869082,
      "flos": 17702667619200.0,
      "grad_norm": 2.8547904590763937,
      "language_loss": 0.65297949,
      "learning_rate": 1.8580561458647614e-10,
      "loss": 0.67427403,
      "num_input_tokens_seen": 357468060,
      "step": 16562,
      "time_per_iteration": 2.412872552871704
    },
    {
      "auxiliary_loss_clip": 0.01079203,
      "auxiliary_loss_mlp": 0.00783862,
      "balance_loss_clip": 1.03540051,
      "balance_loss_mlp": 1.01069927,
      "epoch": 0.9958214339395761,
      "flos": 30555994671360.0,
      "grad_norm": 1.7741888614636785,
      "language_loss": 0.64010078,
      "learning_rate": 1.805348815528962e-10,
      "loss": 0.65873146,
      "num_input_tokens_seen": 357489665,
      "step": 16563,
      "time_per_iteration": 2.573263168334961
    },
    {
      "auxiliary_loss_clip": 0.01079189,
      "auxiliary_loss_mlp": 0.01029873,
      "balance_loss_clip": 1.03349233,
      "balance_loss_mlp": 1.01777864,
      "epoch": 0.9958815571922441,
      "flos": 24169174306560.0,
      "grad_norm": 2.959805552145584,
      "language_loss": 0.64607549,
      "learning_rate": 1.7533998187380105e-10,
      "loss": 0.66716611,
      "num_input_tokens_seen": 357511975,
      "step": 16564,
      "time_per_iteration": 2.5864317417144775
    },
    {
      "auxiliary_loss_clip": 0.01079291,
      "auxiliary_loss_mlp": 0.00780005,
      "balance_loss_clip": 1.03417516,
      "balance_loss_mlp": 1.00511599,
      "epoch": 0.995941680444912,
      "flos": 15487016065920.0,
      "grad_norm": 1.9013417943629443,
      "language_loss": 0.74135298,
      "learning_rate": 1.7022091574636633e-10,
      "loss": 0.75994593,
      "num_input_tokens_seen": 357529345,
      "step": 16565,
      "time_per_iteration": 2.4879424571990967
    },
    {
      "auxiliary_loss_clip": 0.01082495,
      "auxiliary_loss_mlp": 0.01031001,
      "balance_loss_clip": 1.03330183,
      "balance_loss_mlp": 1.01956248,
      "epoch": 0.9960018036975801,
      "flos": 18621227145600.0,
      "grad_norm": 1.7813955003044877,
      "language_loss": 0.79297155,
      "learning_rate": 1.6517768336443694e-10,
      "loss": 0.81410652,
      "num_input_tokens_seen": 357547615,
      "step": 16566,
      "time_per_iteration": 2.5112364292144775
    },
    {
      "auxiliary_loss_clip": 0.010525,
      "auxiliary_loss_mlp": 0.00782149,
      "balance_loss_clip": 1.03336251,
      "balance_loss_mlp": 1.00947046,
      "epoch": 0.996061926950248,
      "flos": 20084120352000.0,
      "grad_norm": 1.6769048601194516,
      "language_loss": 0.70935625,
      "learning_rate": 1.6021028491941535e-10,
      "loss": 0.72770274,
      "num_input_tokens_seen": 357567380,
      "step": 16567,
      "time_per_iteration": 2.550163507461548
    },
    {
      "auxiliary_loss_clip": 0.0109169,
      "auxiliary_loss_mlp": 0.01032214,
      "balance_loss_clip": 1.03335238,
      "balance_loss_mlp": 1.01939321,
      "epoch": 0.996122050202916,
      "flos": 24347829576960.0,
      "grad_norm": 7.120991661128482,
      "language_loss": 0.79103339,
      "learning_rate": 1.5531872059959538e-10,
      "loss": 0.81227249,
      "num_input_tokens_seen": 357586435,
      "step": 16568,
      "time_per_iteration": 3.93766188621521
    },
    {
      "auxiliary_loss_clip": 0.01088482,
      "auxiliary_loss_mlp": 0.01028856,
      "balance_loss_clip": 1.03448188,
      "balance_loss_mlp": 1.01785326,
      "epoch": 0.9961821734555839,
      "flos": 24199302839040.0,
      "grad_norm": 1.6485121911690461,
      "language_loss": 0.81838489,
      "learning_rate": 1.5050299059060634e-10,
      "loss": 0.83955824,
      "num_input_tokens_seen": 357604720,
      "step": 16569,
      "time_per_iteration": 2.534306287765503
    },
    {
      "auxiliary_loss_clip": 0.01064782,
      "auxiliary_loss_mlp": 0.0078453,
      "balance_loss_clip": 1.03409052,
      "balance_loss_mlp": 1.01198483,
      "epoch": 0.9962422967082519,
      "flos": 22633741584000.0,
      "grad_norm": 2.2886239590763857,
      "language_loss": 0.706568,
      "learning_rate": 1.457630950747468e-10,
      "loss": 0.72506112,
      "num_input_tokens_seen": 357622345,
      "step": 16570,
      "time_per_iteration": 2.5477631092071533
    },
    {
      "auxiliary_loss_clip": 0.01074543,
      "auxiliary_loss_mlp": 0.01027765,
      "balance_loss_clip": 1.03543782,
      "balance_loss_mlp": 1.0157125,
      "epoch": 0.9963024199609198,
      "flos": 26396030903040.0,
      "grad_norm": 1.58145761227336,
      "language_loss": 0.7495079,
      "learning_rate": 1.4109903423209502e-10,
      "loss": 0.770531,
      "num_input_tokens_seen": 357642710,
      "step": 16571,
      "time_per_iteration": 4.002007007598877
    },
    {
      "auxiliary_loss_clip": 0.01082029,
      "auxiliary_loss_mlp": 0.010322,
      "balance_loss_clip": 1.03439295,
      "balance_loss_mlp": 1.02017212,
      "epoch": 0.9963625432135879,
      "flos": 16581537342720.0,
      "grad_norm": 1.8398003542360977,
      "language_loss": 0.79586112,
      "learning_rate": 1.3651080823939843e-10,
      "loss": 0.81700349,
      "num_input_tokens_seen": 357659870,
      "step": 16572,
      "time_per_iteration": 3.8611907958984375
    },
    {
      "auxiliary_loss_clip": 0.0107944,
      "auxiliary_loss_mlp": 0.01029566,
      "balance_loss_clip": 1.03503346,
      "balance_loss_mlp": 1.01791906,
      "epoch": 0.9964226664662559,
      "flos": 26468534505600.0,
      "grad_norm": 1.7734871998858,
      "language_loss": 0.70656145,
      "learning_rate": 1.3199841727074e-10,
      "loss": 0.72765148,
      "num_input_tokens_seen": 357677075,
      "step": 16573,
      "time_per_iteration": 2.532543659210205
    },
    {
      "auxiliary_loss_clip": 0.01081867,
      "auxiliary_loss_mlp": 0.01031466,
      "balance_loss_clip": 1.03610647,
      "balance_loss_mlp": 1.01914573,
      "epoch": 0.9964827897189238,
      "flos": 27448320764160.0,
      "grad_norm": 2.2907500380003283,
      "language_loss": 0.63440758,
      "learning_rate": 1.275618614968721e-10,
      "loss": 0.65554094,
      "num_input_tokens_seen": 357696715,
      "step": 16574,
      "time_per_iteration": 2.631331443786621
    },
    {
      "auxiliary_loss_clip": 0.01075413,
      "auxiliary_loss_mlp": 0.01032026,
      "balance_loss_clip": 1.03899539,
      "balance_loss_mlp": 1.01932991,
      "epoch": 0.9965429129715918,
      "flos": 11721566350080.0,
      "grad_norm": 2.6100508764097876,
      "language_loss": 0.76249325,
      "learning_rate": 1.2320114108654856e-10,
      "loss": 0.78356767,
      "num_input_tokens_seen": 357712345,
      "step": 16575,
      "time_per_iteration": 2.526296615600586
    },
    {
      "auxiliary_loss_clip": 0.01081358,
      "auxiliary_loss_mlp": 0.01030029,
      "balance_loss_clip": 1.03677154,
      "balance_loss_mlp": 1.017959,
      "epoch": 0.9966030362242597,
      "flos": 19756004590080.0,
      "grad_norm": 4.312242747907826,
      "language_loss": 0.70189822,
      "learning_rate": 1.1891625620474855e-10,
      "loss": 0.72301209,
      "num_input_tokens_seen": 357731815,
      "step": 16576,
      "time_per_iteration": 2.534085512161255
    },
    {
      "auxiliary_loss_clip": 0.01089795,
      "auxiliary_loss_mlp": 0.01025129,
      "balance_loss_clip": 1.03415096,
      "balance_loss_mlp": 1.01354218,
      "epoch": 0.9966631594769277,
      "flos": 23915178259200.0,
      "grad_norm": 1.537633629094008,
      "language_loss": 0.71900058,
      "learning_rate": 1.1470720701400871e-10,
      "loss": 0.7401498,
      "num_input_tokens_seen": 357751640,
      "step": 16577,
      "time_per_iteration": 2.481809139251709
    },
    {
      "auxiliary_loss_clip": 0.01078124,
      "auxiliary_loss_mlp": 0.01031785,
      "balance_loss_clip": 1.03529477,
      "balance_loss_mlp": 1.02018034,
      "epoch": 0.9967232827295956,
      "flos": 15559591495680.0,
      "grad_norm": 1.8987269292847335,
      "language_loss": 0.78216684,
      "learning_rate": 1.1057399367397912e-10,
      "loss": 0.80326593,
      "num_input_tokens_seen": 357769850,
      "step": 16578,
      "time_per_iteration": 2.5189197063446045
    },
    {
      "auxiliary_loss_clip": 0.0106085,
      "auxiliary_loss_mlp": 0.00781058,
      "balance_loss_clip": 1.03338981,
      "balance_loss_mlp": 1.00855494,
      "epoch": 0.9967834059822637,
      "flos": 20813035046400.0,
      "grad_norm": 1.6878641958822334,
      "language_loss": 0.76113296,
      "learning_rate": 1.0651661634142328e-10,
      "loss": 0.77955198,
      "num_input_tokens_seen": 357789550,
      "step": 16579,
      "time_per_iteration": 2.5710489749908447
    },
    {
      "auxiliary_loss_clip": 0.01082621,
      "auxiliary_loss_mlp": 0.01032902,
      "balance_loss_clip": 1.03849137,
      "balance_loss_mlp": 1.01943779,
      "epoch": 0.9968435292349316,
      "flos": 36719234830080.0,
      "grad_norm": 2.2363890727953932,
      "language_loss": 0.69582492,
      "learning_rate": 1.0253507516999604e-10,
      "loss": 0.7169801,
      "num_input_tokens_seen": 357809525,
      "step": 16580,
      "time_per_iteration": 2.6679892539978027
    },
    {
      "auxiliary_loss_clip": 0.01052405,
      "auxiliary_loss_mlp": 0.01030572,
      "balance_loss_clip": 1.03338408,
      "balance_loss_mlp": 1.0185858,
      "epoch": 0.9969036524875996,
      "flos": 26760919213440.0,
      "grad_norm": 2.033260115044349,
      "language_loss": 0.80241179,
      "learning_rate": 9.862937031113184e-11,
      "loss": 0.82324159,
      "num_input_tokens_seen": 357829795,
      "step": 16581,
      "time_per_iteration": 3.9894614219665527
    },
    {
      "auxiliary_loss_clip": 0.01074135,
      "auxiliary_loss_mlp": 0.01027188,
      "balance_loss_clip": 1.03491533,
      "balance_loss_mlp": 1.01647687,
      "epoch": 0.9969637757402675,
      "flos": 24827237424000.0,
      "grad_norm": 1.5279830601619302,
      "language_loss": 0.80223626,
      "learning_rate": 9.479950191249031e-11,
      "loss": 0.82324946,
      "num_input_tokens_seen": 357851655,
      "step": 16582,
      "time_per_iteration": 2.5572047233581543
    },
    {
      "auxiliary_loss_clip": 0.01088881,
      "auxiliary_loss_mlp": 0.01029848,
      "balance_loss_clip": 1.03316355,
      "balance_loss_mlp": 1.01856446,
      "epoch": 0.9970238989929355,
      "flos": 23038742407680.0,
      "grad_norm": 1.777530068098933,
      "language_loss": 0.60386693,
      "learning_rate": 9.104547011951069e-11,
      "loss": 0.6250543,
      "num_input_tokens_seen": 357871205,
      "step": 16583,
      "time_per_iteration": 2.4967753887176514
    },
    {
      "auxiliary_loss_clip": 0.0108263,
      "auxiliary_loss_mlp": 0.01032751,
      "balance_loss_clip": 1.03444052,
      "balance_loss_mlp": 1.02072275,
      "epoch": 0.9970840222456034,
      "flos": 25298816106240.0,
      "grad_norm": 1.9618293167763976,
      "language_loss": 0.77941829,
      "learning_rate": 8.736727507452357e-11,
      "loss": 0.8005721,
      "num_input_tokens_seen": 357892145,
      "step": 16584,
      "time_per_iteration": 2.5463454723358154
    },
    {
      "auxiliary_loss_clip": 0.01076472,
      "auxiliary_loss_mlp": 0.01029499,
      "balance_loss_clip": 1.03234041,
      "balance_loss_mlp": 1.01886511,
      "epoch": 0.9971441454982715,
      "flos": 21615602578560.0,
      "grad_norm": 1.4204972284084216,
      "language_loss": 0.69466901,
      "learning_rate": 8.376491691697297e-11,
      "loss": 0.7157287,
      "num_input_tokens_seen": 357911205,
      "step": 16585,
      "time_per_iteration": 2.534080743789673
    },
    {
      "auxiliary_loss_clip": 0.01102221,
      "auxiliary_loss_mlp": 0.01029068,
      "balance_loss_clip": 1.03537989,
      "balance_loss_mlp": 1.01704609,
      "epoch": 0.9972042687509394,
      "flos": 14975612179200.0,
      "grad_norm": 2.383714171437012,
      "language_loss": 0.81093621,
      "learning_rate": 8.023839578363834e-11,
      "loss": 0.8322491,
      "num_input_tokens_seen": 357928190,
      "step": 16586,
      "time_per_iteration": 2.4236743450164795
    },
    {
      "auxiliary_loss_clip": 0.01079071,
      "auxiliary_loss_mlp": 0.01036348,
      "balance_loss_clip": 1.0318315,
      "balance_loss_mlp": 1.02478445,
      "epoch": 0.9972643920036074,
      "flos": 25806664546560.0,
      "grad_norm": 1.5830064606276297,
      "language_loss": 0.77810055,
      "learning_rate": 7.678771180796851e-11,
      "loss": 0.79925478,
      "num_input_tokens_seen": 357946985,
      "step": 16587,
      "time_per_iteration": 2.5617570877075195
    },
    {
      "auxiliary_loss_clip": 0.01084192,
      "auxiliary_loss_mlp": 0.0103757,
      "balance_loss_clip": 1.03561044,
      "balance_loss_mlp": 1.02535653,
      "epoch": 0.9973245152562754,
      "flos": 23326242865920.0,
      "grad_norm": 1.815911400651242,
      "language_loss": 0.72710073,
      "learning_rate": 7.341286512074773e-11,
      "loss": 0.74831837,
      "num_input_tokens_seen": 357966720,
      "step": 16588,
      "time_per_iteration": 2.511484146118164
    },
    {
      "auxiliary_loss_clip": 0.01107976,
      "auxiliary_loss_mlp": 0.01028225,
      "balance_loss_clip": 1.03565741,
      "balance_loss_mlp": 1.01623869,
      "epoch": 0.9973846385089433,
      "flos": 12166212810240.0,
      "grad_norm": 2.829169957408843,
      "language_loss": 0.8245039,
      "learning_rate": 7.011385585031781e-11,
      "loss": 0.84586591,
      "num_input_tokens_seen": 357981375,
      "step": 16589,
      "time_per_iteration": 2.412475824356079
    },
    {
      "auxiliary_loss_clip": 0.01096494,
      "auxiliary_loss_mlp": 0.01035423,
      "balance_loss_clip": 1.0344826,
      "balance_loss_mlp": 1.02194691,
      "epoch": 0.9974447617616113,
      "flos": 20045157073920.0,
      "grad_norm": 2.091142001783348,
      "language_loss": 0.70083052,
      "learning_rate": 6.689068412168986e-11,
      "loss": 0.72214973,
      "num_input_tokens_seen": 358000290,
      "step": 16590,
      "time_per_iteration": 2.462782382965088
    },
    {
      "auxiliary_loss_clip": 0.01081637,
      "auxiliary_loss_mlp": 0.01029928,
      "balance_loss_clip": 1.03407991,
      "balance_loss_mlp": 1.01751196,
      "epoch": 0.9975048850142793,
      "flos": 32014614159360.0,
      "grad_norm": 1.6490247537033744,
      "language_loss": 0.63715148,
      "learning_rate": 6.374335005676634e-11,
      "loss": 0.65826708,
      "num_input_tokens_seen": 358022075,
      "step": 16591,
      "time_per_iteration": 2.6072616577148438
    },
    {
      "auxiliary_loss_clip": 0.01078902,
      "auxiliary_loss_mlp": 0.01025659,
      "balance_loss_clip": 1.0321368,
      "balance_loss_mlp": 1.01401782,
      "epoch": 0.9975650082669473,
      "flos": 36933728895360.0,
      "grad_norm": 1.6677858209369945,
      "language_loss": 0.73246253,
      "learning_rate": 6.067185377522933e-11,
      "loss": 0.75350809,
      "num_input_tokens_seen": 358043940,
      "step": 16592,
      "time_per_iteration": 2.6222312450408936
    },
    {
      "auxiliary_loss_clip": 0.01080952,
      "auxiliary_loss_mlp": 0.01028034,
      "balance_loss_clip": 1.03397059,
      "balance_loss_mlp": 1.01597047,
      "epoch": 0.9976251315196152,
      "flos": 16472117537280.0,
      "grad_norm": 1.4458137758187946,
      "language_loss": 0.84853339,
      "learning_rate": 5.767619539343016e-11,
      "loss": 0.8696233,
      "num_input_tokens_seen": 358062720,
      "step": 16593,
      "time_per_iteration": 2.5073728561401367
    },
    {
      "auxiliary_loss_clip": 0.0109977,
      "auxiliary_loss_mlp": 0.00781107,
      "balance_loss_clip": 1.03421879,
      "balance_loss_mlp": 1.00807738,
      "epoch": 0.9976852547722832,
      "flos": 19646836179840.0,
      "grad_norm": 1.754549658240388,
      "language_loss": 0.69579291,
      "learning_rate": 5.4756375024833656e-11,
      "loss": 0.71460164,
      "num_input_tokens_seen": 358081560,
      "step": 16594,
      "time_per_iteration": 2.4424185752868652
    },
    {
      "auxiliary_loss_clip": 0.01061401,
      "auxiliary_loss_mlp": 0.01023738,
      "balance_loss_clip": 1.03600323,
      "balance_loss_mlp": 1.01198959,
      "epoch": 0.9977453780249511,
      "flos": 20448434044800.0,
      "grad_norm": 2.0454728323364892,
      "language_loss": 0.72896576,
      "learning_rate": 5.1912392780462113e-11,
      "loss": 0.74981713,
      "num_input_tokens_seen": 358099065,
      "step": 16595,
      "time_per_iteration": 2.5930068492889404
    },
    {
      "auxiliary_loss_clip": 0.01014952,
      "auxiliary_loss_mlp": 0.01001225,
      "balance_loss_clip": 1.00462914,
      "balance_loss_mlp": 1.00018239,
      "epoch": 0.9978055012776191,
      "flos": 65455097581440.0,
      "grad_norm": 0.7901317989714431,
      "language_loss": 0.60366356,
      "learning_rate": 4.9144248768007156e-11,
      "loss": 0.62382543,
      "num_input_tokens_seen": 358156095,
      "step": 16596,
      "time_per_iteration": 2.956843376159668
    },
    {
      "auxiliary_loss_clip": 0.01092647,
      "auxiliary_loss_mlp": 0.01028244,
      "balance_loss_clip": 1.03520465,
      "balance_loss_mlp": 1.01663947,
      "epoch": 0.997865624530287,
      "flos": 20631506688000.0,
      "grad_norm": 1.685240371981805,
      "language_loss": 0.77660549,
      "learning_rate": 4.645194309227385e-11,
      "loss": 0.79781437,
      "num_input_tokens_seen": 358175230,
      "step": 16597,
      "time_per_iteration": 2.4770374298095703
    },
    {
      "auxiliary_loss_clip": 0.01091168,
      "auxiliary_loss_mlp": 0.01030817,
      "balance_loss_clip": 1.03223419,
      "balance_loss_mlp": 1.01866913,
      "epoch": 0.9979257477829551,
      "flos": 29387102284800.0,
      "grad_norm": 1.859973030885209,
      "language_loss": 0.82016158,
      "learning_rate": 4.383547585562475e-11,
      "loss": 0.84138143,
      "num_input_tokens_seen": 358197075,
      "step": 16598,
      "time_per_iteration": 2.5209484100341797
    },
    {
      "auxiliary_loss_clip": 0.01084817,
      "auxiliary_loss_mlp": 0.0103986,
      "balance_loss_clip": 1.03510094,
      "balance_loss_mlp": 1.02671766,
      "epoch": 0.997985871035623,
      "flos": 22635070387200.0,
      "grad_norm": 1.868157746287422,
      "language_loss": 0.64362121,
      "learning_rate": 4.129484715709175e-11,
      "loss": 0.664868,
      "num_input_tokens_seen": 358215925,
      "step": 16599,
      "time_per_iteration": 2.5198020935058594
    },
    {
      "auxiliary_loss_clip": 0.01008736,
      "auxiliary_loss_mlp": 0.01003056,
      "balance_loss_clip": 1.00636697,
      "balance_loss_mlp": 1.00203097,
      "epoch": 0.998045994288291,
      "flos": 61806968663040.0,
      "grad_norm": 0.8532045371886507,
      "language_loss": 0.62396759,
      "learning_rate": 3.8830057093264256e-11,
      "loss": 0.64408553,
      "num_input_tokens_seen": 358269035,
      "step": 16600,
      "time_per_iteration": 3.0389177799224854
    },
    {
      "auxiliary_loss_clip": 0.01080121,
      "auxiliary_loss_mlp": 0.0102983,
      "balance_loss_clip": 1.03420472,
      "balance_loss_mlp": 1.01914859,
      "epoch": 0.998106117540959,
      "flos": 19245534456960.0,
      "grad_norm": 1.677989481534099,
      "language_loss": 0.78458488,
      "learning_rate": 3.644110575717896e-11,
      "loss": 0.80568433,
      "num_input_tokens_seen": 358287680,
      "step": 16601,
      "time_per_iteration": 2.506166934967041
    },
    {
      "auxiliary_loss_clip": 0.01073387,
      "auxiliary_loss_mlp": 0.01029652,
      "balance_loss_clip": 1.03446424,
      "balance_loss_mlp": 1.01780832,
      "epoch": 0.9981662407936269,
      "flos": 21106209853440.0,
      "grad_norm": 2.118776507525557,
      "language_loss": 0.82653213,
      "learning_rate": 3.412799323987414e-11,
      "loss": 0.84756255,
      "num_input_tokens_seen": 358304080,
      "step": 16602,
      "time_per_iteration": 2.527155637741089
    },
    {
      "auxiliary_loss_clip": 0.010639,
      "auxiliary_loss_mlp": 0.01031497,
      "balance_loss_clip": 1.03700805,
      "balance_loss_mlp": 1.01988018,
      "epoch": 0.998226364046295,
      "flos": 24316839118080.0,
      "grad_norm": 1.887301513479427,
      "language_loss": 0.62746936,
      "learning_rate": 3.189071962883538e-11,
      "loss": 0.64842331,
      "num_input_tokens_seen": 358323670,
      "step": 16603,
      "time_per_iteration": 2.581286907196045
    },
    {
      "auxiliary_loss_clip": 0.01078496,
      "auxiliary_loss_mlp": 0.01027924,
      "balance_loss_clip": 1.03238809,
      "balance_loss_mlp": 1.01584232,
      "epoch": 0.9982864872989629,
      "flos": 23836389776640.0,
      "grad_norm": 1.9026016785368751,
      "language_loss": 0.71108395,
      "learning_rate": 2.972928500866168e-11,
      "loss": 0.73214817,
      "num_input_tokens_seen": 358341980,
      "step": 16604,
      "time_per_iteration": 2.519912004470825
    },
    {
      "auxiliary_loss_clip": 0.01102374,
      "auxiliary_loss_mlp": 0.01025152,
      "balance_loss_clip": 1.03414023,
      "balance_loss_mlp": 1.01288533,
      "epoch": 0.9983466105516309,
      "flos": 18333116156160.0,
      "grad_norm": 1.5924326000237703,
      "language_loss": 0.64325321,
      "learning_rate": 2.7643689461953613e-11,
      "loss": 0.66452843,
      "num_input_tokens_seen": 358360400,
      "step": 16605,
      "time_per_iteration": 2.4258415699005127
    },
    {
      "auxiliary_loss_clip": 0.01068581,
      "auxiliary_loss_mlp": 0.0102923,
      "balance_loss_clip": 1.03297162,
      "balance_loss_mlp": 1.01751721,
      "epoch": 0.9984067338042988,
      "flos": 17236763285760.0,
      "grad_norm": 2.944242900860713,
      "language_loss": 0.71610641,
      "learning_rate": 2.5633933067092938e-11,
      "loss": 0.73708451,
      "num_input_tokens_seen": 358378990,
      "step": 16606,
      "time_per_iteration": 3.9524030685424805
    },
    {
      "auxiliary_loss_clip": 0.01092263,
      "auxiliary_loss_mlp": 0.0078122,
      "balance_loss_clip": 1.03476965,
      "balance_loss_mlp": 1.00628614,
      "epoch": 0.9984668570569668,
      "flos": 20667884186880.0,
      "grad_norm": 1.9420234398598772,
      "language_loss": 0.82043028,
      "learning_rate": 2.370001590090709e-11,
      "loss": 0.83916509,
      "num_input_tokens_seen": 358395970,
      "step": 16607,
      "time_per_iteration": 2.4806463718414307
    },
    {
      "auxiliary_loss_clip": 0.01069867,
      "auxiliary_loss_mlp": 0.01031262,
      "balance_loss_clip": 1.03145027,
      "balance_loss_mlp": 1.01887059,
      "epoch": 0.9985269803096347,
      "flos": 30262532555520.0,
      "grad_norm": 1.5691797383449366,
      "language_loss": 0.66682851,
      "learning_rate": 2.184193803622669e-11,
      "loss": 0.68783975,
      "num_input_tokens_seen": 358417355,
      "step": 16608,
      "time_per_iteration": 2.6165785789489746
    },
    {
      "auxiliary_loss_clip": 0.0106344,
      "auxiliary_loss_mlp": 0.01029829,
      "balance_loss_clip": 1.03399205,
      "balance_loss_mlp": 1.01802182,
      "epoch": 0.9985871035623027,
      "flos": 10560970005120.0,
      "grad_norm": 5.488218012838019,
      "language_loss": 0.80622649,
      "learning_rate": 2.0059699543883978e-11,
      "loss": 0.82715923,
      "num_input_tokens_seen": 358434345,
      "step": 16609,
      "time_per_iteration": 3.931772470474243
    },
    {
      "auxiliary_loss_clip": 0.01082386,
      "auxiliary_loss_mlp": 0.01032973,
      "balance_loss_clip": 1.03205347,
      "balance_loss_mlp": 1.02084982,
      "epoch": 0.9986472268149706,
      "flos": 16873455173760.0,
      "grad_norm": 1.4826627704400046,
      "language_loss": 0.6297071,
      "learning_rate": 1.8353300491158462e-11,
      "loss": 0.65086067,
      "num_input_tokens_seen": 358452870,
      "step": 16610,
      "time_per_iteration": 2.495527744293213
    },
    {
      "auxiliary_loss_clip": 0.01091798,
      "auxiliary_loss_mlp": 0.01033555,
      "balance_loss_clip": 1.03398561,
      "balance_loss_mlp": 1.02233171,
      "epoch": 0.9987073500676387,
      "flos": 22054538776320.0,
      "grad_norm": 10.670550101187457,
      "language_loss": 0.67923367,
      "learning_rate": 1.672274094288717e-11,
      "loss": 0.70048726,
      "num_input_tokens_seen": 358472210,
      "step": 16611,
      "time_per_iteration": 3.8829832077026367
    },
    {
      "auxiliary_loss_clip": 0.01060002,
      "auxiliary_loss_mlp": 0.01034056,
      "balance_loss_clip": 1.03283393,
      "balance_loss_mlp": 1.02104402,
      "epoch": 0.9987674733203066,
      "flos": 30482880537600.0,
      "grad_norm": 1.4055315080165434,
      "language_loss": 0.69653618,
      "learning_rate": 1.5168020961020544e-11,
      "loss": 0.71747673,
      "num_input_tokens_seen": 358493840,
      "step": 16612,
      "time_per_iteration": 2.612689971923828
    },
    {
      "auxiliary_loss_clip": 0.01076906,
      "auxiliary_loss_mlp": 0.0102908,
      "balance_loss_clip": 1.03413808,
      "balance_loss_mlp": 1.01826143,
      "epoch": 0.9988275965729746,
      "flos": 27745230585600.0,
      "grad_norm": 1.6063592277641376,
      "language_loss": 0.73835993,
      "learning_rate": 1.3689140604400407e-11,
      "loss": 0.7594198,
      "num_input_tokens_seen": 358515060,
      "step": 16613,
      "time_per_iteration": 2.633131742477417
    },
    {
      "auxiliary_loss_clip": 0.01067017,
      "auxiliary_loss_mlp": 0.00785509,
      "balance_loss_clip": 1.03286195,
      "balance_loss_mlp": 1.00858021,
      "epoch": 0.9988877198256426,
      "flos": 17524191916800.0,
      "grad_norm": 2.0165002279492663,
      "language_loss": 0.73550928,
      "learning_rate": 1.2286099928981996e-11,
      "loss": 0.75403452,
      "num_input_tokens_seen": 358528200,
      "step": 16614,
      "time_per_iteration": 2.484316825866699
    },
    {
      "auxiliary_loss_clip": 0.01089366,
      "auxiliary_loss_mlp": 0.01030511,
      "balance_loss_clip": 1.03418529,
      "balance_loss_mlp": 1.01929975,
      "epoch": 0.9989478430783105,
      "flos": 20996502739200.0,
      "grad_norm": 1.705809737280282,
      "language_loss": 0.72609651,
      "learning_rate": 1.0958898988278065e-11,
      "loss": 0.74729532,
      "num_input_tokens_seen": 358548360,
      "step": 16615,
      "time_per_iteration": 2.486840009689331
    },
    {
      "auxiliary_loss_clip": 0.01106865,
      "auxiliary_loss_mlp": 0.00782832,
      "balance_loss_clip": 1.03658831,
      "balance_loss_mlp": 1.00906563,
      "epoch": 0.9990079663309785,
      "flos": 13370620769280.0,
      "grad_norm": 3.9348376290084155,
      "language_loss": 0.77619541,
      "learning_rate": 9.70753783247069e-12,
      "loss": 0.7950924,
      "num_input_tokens_seen": 358566270,
      "step": 16616,
      "time_per_iteration": 2.404435873031616
    },
    {
      "auxiliary_loss_clip": 0.01084962,
      "auxiliary_loss_mlp": 0.01028862,
      "balance_loss_clip": 1.0363549,
      "balance_loss_mlp": 1.01669073,
      "epoch": 0.9990680895836465,
      "flos": 17310236555520.0,
      "grad_norm": 1.8949845001714494,
      "language_loss": 0.8258456,
      "learning_rate": 8.532016508855378e-12,
      "loss": 0.84698385,
      "num_input_tokens_seen": 358584710,
      "step": 16617,
      "time_per_iteration": 2.5276079177856445
    },
    {
      "auxiliary_loss_clip": 0.01081967,
      "auxiliary_loss_mlp": 0.01027101,
      "balance_loss_clip": 1.03425992,
      "balance_loss_mlp": 1.01563859,
      "epoch": 0.9991282128363145,
      "flos": 24207993930240.0,
      "grad_norm": 1.4393801220047557,
      "language_loss": 0.7856729,
      "learning_rate": 7.43233506206309e-12,
      "loss": 0.80676353,
      "num_input_tokens_seen": 358606750,
      "step": 16618,
      "time_per_iteration": 2.5549628734588623
    },
    {
      "auxiliary_loss_clip": 0.01101043,
      "auxiliary_loss_mlp": 0.01027993,
      "balance_loss_clip": 1.03330278,
      "balance_loss_mlp": 1.01644111,
      "epoch": 0.9991883360889824,
      "flos": 21175301664000.0,
      "grad_norm": 1.6424965107360727,
      "language_loss": 0.74615735,
      "learning_rate": 6.408493534060255e-12,
      "loss": 0.76744771,
      "num_input_tokens_seen": 358624675,
      "step": 16619,
      "time_per_iteration": 2.4647164344787598
    },
    {
      "auxiliary_loss_clip": 0.0108853,
      "auxiliary_loss_mlp": 0.01027258,
      "balance_loss_clip": 1.03240013,
      "balance_loss_mlp": 1.01657128,
      "epoch": 0.9992484593416504,
      "flos": 19901155449600.0,
      "grad_norm": 2.006624435304398,
      "language_loss": 0.86377752,
      "learning_rate": 5.460491963260594e-12,
      "loss": 0.88493544,
      "num_input_tokens_seen": 358640715,
      "step": 16620,
      "time_per_iteration": 3.923424243927002
    },
    {
      "auxiliary_loss_clip": 0.01064765,
      "auxiliary_loss_mlp": 0.01025792,
      "balance_loss_clip": 1.03134763,
      "balance_loss_mlp": 1.01450253,
      "epoch": 0.9993085825943183,
      "flos": 24857832833280.0,
      "grad_norm": 1.9258330442567073,
      "language_loss": 0.72729498,
      "learning_rate": 4.58833038607942e-12,
      "loss": 0.74820054,
      "num_input_tokens_seen": 358659630,
      "step": 16621,
      "time_per_iteration": 2.545844316482544
    },
    {
      "auxiliary_loss_clip": 0.00998604,
      "auxiliary_loss_mlp": 0.01003661,
      "balance_loss_clip": 1.00620592,
      "balance_loss_mlp": 1.00262344,
      "epoch": 0.9993687058469863,
      "flos": 71284478780160.0,
      "grad_norm": 0.7362386165268472,
      "language_loss": 0.56500053,
      "learning_rate": 3.79200883515729e-12,
      "loss": 0.58502316,
      "num_input_tokens_seen": 358727840,
      "step": 16622,
      "time_per_iteration": 3.341364860534668
    },
    {
      "auxiliary_loss_clip": 0.01063653,
      "auxiliary_loss_mlp": 0.01027204,
      "balance_loss_clip": 1.03293681,
      "balance_loss_mlp": 1.01508057,
      "epoch": 0.9994288290996542,
      "flos": 12199573566720.0,
      "grad_norm": 3.353668276695799,
      "language_loss": 0.71405113,
      "learning_rate": 3.071527340914315e-12,
      "loss": 0.73495972,
      "num_input_tokens_seen": 358744125,
      "step": 16623,
      "time_per_iteration": 2.506336212158203
    },
    {
      "auxiliary_loss_clip": 0.01066469,
      "auxiliary_loss_mlp": 0.01028625,
      "balance_loss_clip": 1.03515387,
      "balance_loss_mlp": 1.01628137,
      "epoch": 0.9994889523523223,
      "flos": 17889942153600.0,
      "grad_norm": 2.5589776149256536,
      "language_loss": 0.748088,
      "learning_rate": 2.4268859304399368e-12,
      "loss": 0.76903892,
      "num_input_tokens_seen": 358761420,
      "step": 16624,
      "time_per_iteration": 2.5339014530181885
    },
    {
      "auxiliary_loss_clip": 0.01074052,
      "auxiliary_loss_mlp": 0.01031029,
      "balance_loss_clip": 1.03382421,
      "balance_loss_mlp": 1.01859593,
      "epoch": 0.9995490756049902,
      "flos": 26578888064640.0,
      "grad_norm": 1.6565297346606276,
      "language_loss": 0.74126279,
      "learning_rate": 1.8580846286031514e-12,
      "loss": 0.7623136,
      "num_input_tokens_seen": 358782600,
      "step": 16625,
      "time_per_iteration": 2.592473268508911
    },
    {
      "auxiliary_loss_clip": 0.01089533,
      "auxiliary_loss_mlp": 0.01032989,
      "balance_loss_clip": 1.03379273,
      "balance_loss_mlp": 1.02163446,
      "epoch": 0.9996091988576582,
      "flos": 22200048771840.0,
      "grad_norm": 2.4106335339446354,
      "language_loss": 0.76861835,
      "learning_rate": 1.3651234567202408e-12,
      "loss": 0.78984356,
      "num_input_tokens_seen": 358801220,
      "step": 16626,
      "time_per_iteration": 2.497978687286377
    },
    {
      "auxiliary_loss_clip": 0.01101655,
      "auxiliary_loss_mlp": 0.01032822,
      "balance_loss_clip": 1.03587604,
      "balance_loss_mlp": 1.02104425,
      "epoch": 0.9996693221103262,
      "flos": 27373195468800.0,
      "grad_norm": 1.834097594039187,
      "language_loss": 0.82148588,
      "learning_rate": 9.480024334429515e-13,
      "loss": 0.84283066,
      "num_input_tokens_seen": 358819190,
      "step": 16627,
      "time_per_iteration": 2.485482931137085
    },
    {
      "auxiliary_loss_clip": 0.0109828,
      "auxiliary_loss_mlp": 0.01035599,
      "balance_loss_clip": 1.03706753,
      "balance_loss_mlp": 1.02246821,
      "epoch": 0.9997294453629941,
      "flos": 26870410846080.0,
      "grad_norm": 2.1317555430853203,
      "language_loss": 0.70762205,
      "learning_rate": 6.067215747584952e-13,
      "loss": 0.72896087,
      "num_input_tokens_seen": 358839850,
      "step": 16628,
      "time_per_iteration": 2.5652377605438232
    },
    {
      "auxiliary_loss_clip": 0.0109184,
      "auxiliary_loss_mlp": 0.01025957,
      "balance_loss_clip": 1.03195941,
      "balance_loss_mlp": 1.01410818,
      "epoch": 0.9997895686156621,
      "flos": 23476996247040.0,
      "grad_norm": 1.3729360688507088,
      "language_loss": 0.7546463,
      "learning_rate": 3.4128089332341456e-13,
      "loss": 0.77582425,
      "num_input_tokens_seen": 358859805,
      "step": 16629,
      "time_per_iteration": 2.479379415512085
    },
    {
      "auxiliary_loss_clip": 0.01085618,
      "auxiliary_loss_mlp": 0.01033616,
      "balance_loss_clip": 1.03558874,
      "balance_loss_mlp": 1.02126551,
      "epoch": 0.9998496918683301,
      "flos": 20224961579520.0,
      "grad_norm": 1.6201642931700957,
      "language_loss": 0.60554153,
      "learning_rate": 1.5168039935176126e-13,
      "loss": 0.6267339,
      "num_input_tokens_seen": 358877900,
      "step": 16630,
      "time_per_iteration": 2.511185646057129
    },
    {
      "auxiliary_loss_clip": 0.01060604,
      "auxiliary_loss_mlp": 0.01028799,
      "balance_loss_clip": 1.03428459,
      "balance_loss_mlp": 1.01668119,
      "epoch": 0.9999098151209981,
      "flos": 21652913831040.0,
      "grad_norm": 1.963784734566383,
      "language_loss": 0.60584134,
      "learning_rate": 3.792010017100722e-14,
      "loss": 0.62673533,
      "num_input_tokens_seen": 358897285,
      "step": 16631,
      "time_per_iteration": 2.5895564556121826
    },
    {
      "auxiliary_loss_clip": 0.01050255,
      "auxiliary_loss_mlp": 0.007821,
      "balance_loss_clip": 1.03462207,
      "balance_loss_mlp": 1.00652003,
      "epoch": 0.999969938373666,
      "flos": 11544599018880.0,
      "grad_norm": 1.809182672960919,
      "language_loss": 0.72743851,
      "learning_rate": 0.0,
      "loss": 0.74576211,
      "num_input_tokens_seen": 358911570,
      "step": 16632,
      "time_per_iteration": 2.5463998317718506
    },
    {
      "epoch": 0.999969938373666,
      "num_input_tokens_seen": 358911570,
      "step": 16632,
      "total_flos": 1.3992169073237033e+18,
      "train_loss": 0.7693321485898424,
      "train_runtime": 46762.2544,
      "train_samples_per_second": 14.227,
      "train_steps_per_second": 0.356
    }
  ],
  "logging_steps": 1.0,
  "max_steps": 16632,
  "num_input_tokens_seen": 358911570,
  "num_train_epochs": 1,
  "save_steps": 3328,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 1.3992169073237033e+18,
  "train_batch_size": 5,
  "trial_name": null,
  "trial_params": null
}