{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.4738802810938955,
  "eval_steps": 500,
  "global_step": 5597,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0008466683600033867,
      "grad_norm": 2.1896274089813232,
      "learning_rate": 1.1061797752808989e-06,
      "loss": 0.9007,
      "step": 10
    },
    {
      "epoch": 0.0016933367200067733,
      "grad_norm": 1.3733998537063599,
      "learning_rate": 2.2241573033707866e-06,
      "loss": 0.8719,
      "step": 20
    },
    {
      "epoch": 0.00254000508001016,
      "grad_norm": 1.1982183456420898,
      "learning_rate": 3.342134831460675e-06,
      "loss": 0.8299,
      "step": 30
    },
    {
      "epoch": 0.0033866734400135466,
      "grad_norm": 1.1160142421722412,
      "learning_rate": 4.4601123595505624e-06,
      "loss": 0.7803,
      "step": 40
    },
    {
      "epoch": 0.004233341800016934,
      "grad_norm": 0.9138423204421997,
      "learning_rate": 5.5780898876404495e-06,
      "loss": 0.7622,
      "step": 50
    },
    {
      "epoch": 0.00508001016002032,
      "grad_norm": 0.8700115084648132,
      "learning_rate": 6.6960674157303374e-06,
      "loss": 0.7332,
      "step": 60
    },
    {
      "epoch": 0.005926678520023707,
      "grad_norm": 0.8024980425834656,
      "learning_rate": 7.814044943820226e-06,
      "loss": 0.7072,
      "step": 70
    },
    {
      "epoch": 0.006773346880027093,
      "grad_norm": 0.8063226342201233,
      "learning_rate": 8.932022471910112e-06,
      "loss": 0.7082,
      "step": 80
    },
    {
      "epoch": 0.00762001524003048,
      "grad_norm": 0.7324275970458984,
      "learning_rate": 1.005e-05,
      "loss": 0.7095,
      "step": 90
    },
    {
      "epoch": 0.008466683600033867,
      "grad_norm": 0.8663532137870789,
      "learning_rate": 1.1167977528089888e-05,
      "loss": 0.6899,
      "step": 100
    },
    {
      "epoch": 0.009313351960037254,
      "grad_norm": 0.8148631453514099,
      "learning_rate": 1.2285955056179776e-05,
      "loss": 0.6984,
      "step": 110
    },
    {
      "epoch": 0.01016002032004064,
      "grad_norm": 0.7821786403656006,
      "learning_rate": 1.3403932584269664e-05,
      "loss": 0.7066,
      "step": 120
    },
    {
      "epoch": 0.011006688680044027,
      "grad_norm": 0.7769770622253418,
      "learning_rate": 1.4521910112359552e-05,
      "loss": 0.6908,
      "step": 130
    },
    {
      "epoch": 0.011853357040047413,
      "grad_norm": 0.9013803601264954,
      "learning_rate": 1.563988764044944e-05,
      "loss": 0.6885,
      "step": 140
    },
    {
      "epoch": 0.0127000254000508,
      "grad_norm": 0.7101280689239502,
      "learning_rate": 1.6757865168539328e-05,
      "loss": 0.6886,
      "step": 150
    },
    {
      "epoch": 0.013546693760054187,
      "grad_norm": 0.8616020679473877,
      "learning_rate": 1.7875842696629214e-05,
      "loss": 0.6849,
      "step": 160
    },
    {
      "epoch": 0.014393362120057573,
      "grad_norm": 0.8392877578735352,
      "learning_rate": 1.89938202247191e-05,
      "loss": 0.6933,
      "step": 170
    },
    {
      "epoch": 0.01524003048006096,
      "grad_norm": 0.766579806804657,
      "learning_rate": 2e-05,
      "loss": 0.7048,
      "step": 180
    },
    {
      "epoch": 0.016086698840064348,
      "grad_norm": 0.7274855375289917,
      "learning_rate": 2e-05,
      "loss": 0.7039,
      "step": 190
    },
    {
      "epoch": 0.016933367200067734,
      "grad_norm": 0.7656974792480469,
      "learning_rate": 2e-05,
      "loss": 0.6829,
      "step": 200
    },
    {
      "epoch": 0.01778003556007112,
      "grad_norm": 0.7630816102027893,
      "learning_rate": 2e-05,
      "loss": 0.6948,
      "step": 210
    },
    {
      "epoch": 0.018626703920074508,
      "grad_norm": 0.7607164978981018,
      "learning_rate": 2e-05,
      "loss": 0.6976,
      "step": 220
    },
    {
      "epoch": 0.019473372280077894,
      "grad_norm": 0.6640747785568237,
      "learning_rate": 2e-05,
      "loss": 0.6529,
      "step": 230
    },
    {
      "epoch": 0.02032004064008128,
      "grad_norm": 0.648815393447876,
      "learning_rate": 2e-05,
      "loss": 0.687,
      "step": 240
    },
    {
      "epoch": 0.021166709000084667,
      "grad_norm": 0.5993350744247437,
      "learning_rate": 2e-05,
      "loss": 0.693,
      "step": 250
    },
    {
      "epoch": 0.022013377360088054,
      "grad_norm": 0.6671707630157471,
      "learning_rate": 2e-05,
      "loss": 0.7208,
      "step": 260
    },
    {
      "epoch": 0.02286004572009144,
      "grad_norm": 0.5551834106445312,
      "learning_rate": 2e-05,
      "loss": 0.6958,
      "step": 270
    },
    {
      "epoch": 0.023706714080094827,
      "grad_norm": 0.5955666303634644,
      "learning_rate": 2e-05,
      "loss": 0.685,
      "step": 280
    },
    {
      "epoch": 0.024553382440098213,
      "grad_norm": 0.5701448917388916,
      "learning_rate": 2e-05,
      "loss": 0.6796,
      "step": 290
    },
    {
      "epoch": 0.0254000508001016,
      "grad_norm": 0.5824018716812134,
      "learning_rate": 2e-05,
      "loss": 0.6841,
      "step": 300
    },
    {
      "epoch": 0.026246719160104987,
      "grad_norm": 0.576776921749115,
      "learning_rate": 2e-05,
      "loss": 0.6563,
      "step": 310
    },
    {
      "epoch": 0.027093387520108373,
      "grad_norm": 0.5411778688430786,
      "learning_rate": 2e-05,
      "loss": 0.6835,
      "step": 320
    },
    {
      "epoch": 0.02794005588011176,
      "grad_norm": 0.5345405340194702,
      "learning_rate": 2e-05,
      "loss": 0.6657,
      "step": 330
    },
    {
      "epoch": 0.028786724240115146,
      "grad_norm": 0.5346488952636719,
      "learning_rate": 2e-05,
      "loss": 0.6951,
      "step": 340
    },
    {
      "epoch": 0.029633392600118533,
      "grad_norm": 0.49902835488319397,
      "learning_rate": 2e-05,
      "loss": 0.664,
      "step": 350
    },
    {
      "epoch": 0.03048006096012192,
      "grad_norm": 0.6449501514434814,
      "learning_rate": 2e-05,
      "loss": 0.6737,
      "step": 360
    },
    {
      "epoch": 0.031326729320125306,
      "grad_norm": 0.487582802772522,
      "learning_rate": 2e-05,
      "loss": 0.6546,
      "step": 370
    },
    {
      "epoch": 0.032173397680128696,
      "grad_norm": 0.5169985294342041,
      "learning_rate": 2e-05,
      "loss": 0.6731,
      "step": 380
    },
    {
      "epoch": 0.03302006604013208,
      "grad_norm": 0.5427067279815674,
      "learning_rate": 2e-05,
      "loss": 0.6882,
      "step": 390
    },
    {
      "epoch": 0.03386673440013547,
      "grad_norm": 0.5355451107025146,
      "learning_rate": 2e-05,
      "loss": 0.6955,
      "step": 400
    },
    {
      "epoch": 0.03471340276013885,
      "grad_norm": 0.6196058392524719,
      "learning_rate": 2e-05,
      "loss": 0.6807,
      "step": 410
    },
    {
      "epoch": 0.03556007112014224,
      "grad_norm": 0.5216856598854065,
      "learning_rate": 2e-05,
      "loss": 0.6602,
      "step": 420
    },
    {
      "epoch": 0.036406739480145625,
      "grad_norm": 0.5485526323318481,
      "learning_rate": 2e-05,
      "loss": 0.6846,
      "step": 430
    },
    {
      "epoch": 0.037253407840149015,
      "grad_norm": 0.48402827978134155,
      "learning_rate": 2e-05,
      "loss": 0.6669,
      "step": 440
    },
    {
      "epoch": 0.0381000762001524,
      "grad_norm": 0.48090001940727234,
      "learning_rate": 2e-05,
      "loss": 0.6651,
      "step": 450
    },
    {
      "epoch": 0.03894674456015579,
      "grad_norm": 1.9493638277053833,
      "learning_rate": 2e-05,
      "loss": 0.6813,
      "step": 460
    },
    {
      "epoch": 0.03979341292015917,
      "grad_norm": 0.5349039435386658,
      "learning_rate": 2e-05,
      "loss": 0.6901,
      "step": 470
    },
    {
      "epoch": 0.04064008128016256,
      "grad_norm": 0.4926290214061737,
      "learning_rate": 2e-05,
      "loss": 0.6514,
      "step": 480
    },
    {
      "epoch": 0.041486749640165944,
      "grad_norm": 0.6464523077011108,
      "learning_rate": 2e-05,
      "loss": 0.6758,
      "step": 490
    },
    {
      "epoch": 0.042333418000169334,
      "grad_norm": 0.5073714256286621,
      "learning_rate": 2e-05,
      "loss": 0.6941,
      "step": 500
    },
    {
      "epoch": 0.04318008636017272,
      "grad_norm": 0.5042576193809509,
      "learning_rate": 2e-05,
      "loss": 0.6788,
      "step": 510
    },
    {
      "epoch": 0.04402675472017611,
      "grad_norm": 0.4722915589809418,
      "learning_rate": 2e-05,
      "loss": 0.6555,
      "step": 520
    },
    {
      "epoch": 0.04487342308017949,
      "grad_norm": 0.5186958312988281,
      "learning_rate": 2e-05,
      "loss": 0.657,
      "step": 530
    },
    {
      "epoch": 0.04572009144018288,
      "grad_norm": 0.47601163387298584,
      "learning_rate": 2e-05,
      "loss": 0.6734,
      "step": 540
    },
    {
      "epoch": 0.046566759800186264,
      "grad_norm": 0.4465620517730713,
      "learning_rate": 2e-05,
      "loss": 0.6834,
      "step": 550
    },
    {
      "epoch": 0.047413428160189654,
      "grad_norm": 0.4941118359565735,
      "learning_rate": 2e-05,
      "loss": 0.6476,
      "step": 560
    },
    {
      "epoch": 0.048260096520193044,
      "grad_norm": 0.479648232460022,
      "learning_rate": 2e-05,
      "loss": 0.6544,
      "step": 570
    },
    {
      "epoch": 0.04910676488019643,
      "grad_norm": 0.5040650963783264,
      "learning_rate": 2e-05,
      "loss": 0.6754,
      "step": 580
    },
    {
      "epoch": 0.04995343324019982,
      "grad_norm": 0.4866141378879547,
      "learning_rate": 2e-05,
      "loss": 0.6917,
      "step": 590
    },
    {
      "epoch": 0.0508001016002032,
      "grad_norm": 0.46658095717430115,
      "learning_rate": 2e-05,
      "loss": 0.6811,
      "step": 600
    },
    {
      "epoch": 0.05164676996020659,
      "grad_norm": 0.6377661228179932,
      "learning_rate": 2e-05,
      "loss": 0.6355,
      "step": 610
    },
    {
      "epoch": 0.05249343832020997,
      "grad_norm": 0.4834534823894501,
      "learning_rate": 2e-05,
      "loss": 0.6957,
      "step": 620
    },
    {
      "epoch": 0.05334010668021336,
      "grad_norm": 0.44412001967430115,
      "learning_rate": 2e-05,
      "loss": 0.6678,
      "step": 630
    },
    {
      "epoch": 0.054186775040216746,
      "grad_norm": 0.5037946701049805,
      "learning_rate": 2e-05,
      "loss": 0.6785,
      "step": 640
    },
    {
      "epoch": 0.055033443400220136,
      "grad_norm": 0.45366835594177246,
      "learning_rate": 2e-05,
      "loss": 0.676,
      "step": 650
    },
    {
      "epoch": 0.05588011176022352,
      "grad_norm": 0.4277074337005615,
      "learning_rate": 2e-05,
      "loss": 0.6591,
      "step": 660
    },
    {
      "epoch": 0.05672678012022691,
      "grad_norm": 0.4375869035720825,
      "learning_rate": 2e-05,
      "loss": 0.6616,
      "step": 670
    },
    {
      "epoch": 0.05757344848023029,
      "grad_norm": 0.46945950388908386,
      "learning_rate": 2e-05,
      "loss": 0.6838,
      "step": 680
    },
    {
      "epoch": 0.05842011684023368,
      "grad_norm": 0.4755280315876007,
      "learning_rate": 2e-05,
      "loss": 0.661,
      "step": 690
    },
    {
      "epoch": 0.059266785200237065,
      "grad_norm": 0.4614129960536957,
      "learning_rate": 2e-05,
      "loss": 0.6465,
      "step": 700
    },
    {
      "epoch": 0.060113453560240455,
      "grad_norm": 0.4619314968585968,
      "learning_rate": 2e-05,
      "loss": 0.6829,
      "step": 710
    },
    {
      "epoch": 0.06096012192024384,
      "grad_norm": 0.4906548261642456,
      "learning_rate": 2e-05,
      "loss": 0.6863,
      "step": 720
    },
    {
      "epoch": 0.06180679028024723,
      "grad_norm": 0.5654237866401672,
      "learning_rate": 2e-05,
      "loss": 0.6994,
      "step": 730
    },
    {
      "epoch": 0.06265345864025061,
      "grad_norm": 0.4611152112483978,
      "learning_rate": 2e-05,
      "loss": 0.6601,
      "step": 740
    },
    {
      "epoch": 0.063500127000254,
      "grad_norm": 0.4759562015533447,
      "learning_rate": 2e-05,
      "loss": 0.6537,
      "step": 750
    },
    {
      "epoch": 0.06434679536025739,
      "grad_norm": 0.45463666319847107,
      "learning_rate": 2e-05,
      "loss": 0.6623,
      "step": 760
    },
    {
      "epoch": 0.06519346372026077,
      "grad_norm": 0.45970118045806885,
      "learning_rate": 2e-05,
      "loss": 0.6557,
      "step": 770
    },
    {
      "epoch": 0.06604013208026416,
      "grad_norm": 0.45690014958381653,
      "learning_rate": 2e-05,
      "loss": 0.6478,
      "step": 780
    },
    {
      "epoch": 0.06688680044026754,
      "grad_norm": 0.45209041237831116,
      "learning_rate": 2e-05,
      "loss": 0.651,
      "step": 790
    },
    {
      "epoch": 0.06773346880027094,
      "grad_norm": 0.5716219544410706,
      "learning_rate": 2e-05,
      "loss": 0.6647,
      "step": 800
    },
    {
      "epoch": 0.06858013716027432,
      "grad_norm": 0.43732336163520813,
      "learning_rate": 2e-05,
      "loss": 0.6604,
      "step": 810
    },
    {
      "epoch": 0.0694268055202777,
      "grad_norm": 0.4478982388973236,
      "learning_rate": 2e-05,
      "loss": 0.6509,
      "step": 820
    },
    {
      "epoch": 0.07027347388028109,
      "grad_norm": 0.44605377316474915,
      "learning_rate": 2e-05,
      "loss": 0.6619,
      "step": 830
    },
    {
      "epoch": 0.07112014224028448,
      "grad_norm": 0.4520704746246338,
      "learning_rate": 2e-05,
      "loss": 0.6541,
      "step": 840
    },
    {
      "epoch": 0.07196681060028787,
      "grad_norm": 0.45137783885002136,
      "learning_rate": 2e-05,
      "loss": 0.6632,
      "step": 850
    },
    {
      "epoch": 0.07281347896029125,
      "grad_norm": 0.4655689299106598,
      "learning_rate": 2e-05,
      "loss": 0.6519,
      "step": 860
    },
    {
      "epoch": 0.07366014732029465,
      "grad_norm": 0.43631526827812195,
      "learning_rate": 2e-05,
      "loss": 0.6597,
      "step": 870
    },
    {
      "epoch": 0.07450681568029803,
      "grad_norm": 0.42785879969596863,
      "learning_rate": 2e-05,
      "loss": 0.6464,
      "step": 880
    },
    {
      "epoch": 0.07535348404030141,
      "grad_norm": 0.43011021614074707,
      "learning_rate": 2e-05,
      "loss": 0.6528,
      "step": 890
    },
    {
      "epoch": 0.0762001524003048,
      "grad_norm": 0.46899306774139404,
      "learning_rate": 2e-05,
      "loss": 0.6666,
      "step": 900
    },
    {
      "epoch": 0.0770468207603082,
      "grad_norm": 0.46042707562446594,
      "learning_rate": 2e-05,
      "loss": 0.645,
      "step": 910
    },
    {
      "epoch": 0.07789348912031158,
      "grad_norm": 0.4511071741580963,
      "learning_rate": 2e-05,
      "loss": 0.6436,
      "step": 920
    },
    {
      "epoch": 0.07874015748031496,
      "grad_norm": 0.4230646789073944,
      "learning_rate": 2e-05,
      "loss": 0.6743,
      "step": 930
    },
    {
      "epoch": 0.07958682584031834,
      "grad_norm": 0.43698370456695557,
      "learning_rate": 2e-05,
      "loss": 0.6481,
      "step": 940
    },
    {
      "epoch": 0.08043349420032174,
      "grad_norm": 0.4549170732498169,
      "learning_rate": 2e-05,
      "loss": 0.6514,
      "step": 950
    },
    {
      "epoch": 0.08128016256032512,
      "grad_norm": 0.45507657527923584,
      "learning_rate": 2e-05,
      "loss": 0.6554,
      "step": 960
    },
    {
      "epoch": 0.0821268309203285,
      "grad_norm": 0.42522671818733215,
      "learning_rate": 2e-05,
      "loss": 0.656,
      "step": 970
    },
    {
      "epoch": 0.08297349928033189,
      "grad_norm": 0.43912380933761597,
      "learning_rate": 2e-05,
      "loss": 0.6608,
      "step": 980
    },
    {
      "epoch": 0.08382016764033529,
      "grad_norm": 0.4183550179004669,
      "learning_rate": 2e-05,
      "loss": 0.6492,
      "step": 990
    },
    {
      "epoch": 0.08466683600033867,
      "grad_norm": 0.44775786995887756,
      "learning_rate": 2e-05,
      "loss": 0.6292,
      "step": 1000
    },
    {
      "epoch": 0.08551350436034205,
      "grad_norm": 0.44851788878440857,
      "learning_rate": 2e-05,
      "loss": 0.6579,
      "step": 1010
    },
    {
      "epoch": 0.08636017272034544,
      "grad_norm": 0.45955589413642883,
      "learning_rate": 2e-05,
      "loss": 0.6468,
      "step": 1020
    },
    {
      "epoch": 0.08720684108034883,
      "grad_norm": 0.4205484688282013,
      "learning_rate": 2e-05,
      "loss": 0.651,
      "step": 1030
    },
    {
      "epoch": 0.08805350944035222,
      "grad_norm": 0.42118069529533386,
      "learning_rate": 2e-05,
      "loss": 0.648,
      "step": 1040
    },
    {
      "epoch": 0.0889001778003556,
      "grad_norm": 0.45766347646713257,
      "learning_rate": 2e-05,
      "loss": 0.6386,
      "step": 1050
    },
    {
      "epoch": 0.08974684616035898,
      "grad_norm": 0.4297139346599579,
      "learning_rate": 2e-05,
      "loss": 0.6334,
      "step": 1060
    },
    {
      "epoch": 0.09059351452036238,
      "grad_norm": 0.4587555229663849,
      "learning_rate": 2e-05,
      "loss": 0.6688,
      "step": 1070
    },
    {
      "epoch": 0.09144018288036576,
      "grad_norm": 0.4319729506969452,
      "learning_rate": 2e-05,
      "loss": 0.6453,
      "step": 1080
    },
    {
      "epoch": 0.09228685124036914,
      "grad_norm": 0.4240010976791382,
      "learning_rate": 2e-05,
      "loss": 0.6387,
      "step": 1090
    },
    {
      "epoch": 0.09313351960037253,
      "grad_norm": 0.41819435358047485,
      "learning_rate": 2e-05,
      "loss": 0.636,
      "step": 1100
    },
    {
      "epoch": 0.09398018796037592,
      "grad_norm": 0.4749411344528198,
      "learning_rate": 2e-05,
      "loss": 0.6502,
      "step": 1110
    },
    {
      "epoch": 0.09482685632037931,
      "grad_norm": 0.4842819571495056,
      "learning_rate": 2e-05,
      "loss": 0.6601,
      "step": 1120
    },
    {
      "epoch": 0.09567352468038269,
      "grad_norm": 0.39176562428474426,
      "learning_rate": 2e-05,
      "loss": 0.6316,
      "step": 1130
    },
    {
      "epoch": 0.09652019304038609,
      "grad_norm": 0.46135643124580383,
      "learning_rate": 2e-05,
      "loss": 0.6763,
      "step": 1140
    },
    {
      "epoch": 0.09736686140038947,
      "grad_norm": 0.4732727110385895,
      "learning_rate": 2e-05,
      "loss": 0.6577,
      "step": 1150
    },
    {
      "epoch": 0.09821352976039285,
      "grad_norm": 0.4651736319065094,
      "learning_rate": 2e-05,
      "loss": 0.6385,
      "step": 1160
    },
    {
      "epoch": 0.09906019812039624,
      "grad_norm": 0.44102978706359863,
      "learning_rate": 2e-05,
      "loss": 0.644,
      "step": 1170
    },
    {
      "epoch": 0.09990686648039963,
      "grad_norm": 0.399547815322876,
      "learning_rate": 2e-05,
      "loss": 0.6441,
      "step": 1180
    },
    {
      "epoch": 0.10075353484040302,
      "grad_norm": 0.5150323510169983,
      "learning_rate": 2e-05,
      "loss": 0.6812,
      "step": 1190
    },
    {
      "epoch": 0.1016002032004064,
      "grad_norm": 0.43167948722839355,
      "learning_rate": 2e-05,
      "loss": 0.6416,
      "step": 1200
    },
    {
      "epoch": 0.10244687156040978,
      "grad_norm": 0.4367421567440033,
      "learning_rate": 2e-05,
      "loss": 0.6689,
      "step": 1210
    },
    {
      "epoch": 0.10329353992041318,
      "grad_norm": 0.4407538175582886,
      "learning_rate": 2e-05,
      "loss": 0.6628,
      "step": 1220
    },
    {
      "epoch": 0.10414020828041656,
      "grad_norm": 0.4248872995376587,
      "learning_rate": 2e-05,
      "loss": 0.6584,
      "step": 1230
    },
    {
      "epoch": 0.10498687664041995,
      "grad_norm": 0.42799079418182373,
      "learning_rate": 2e-05,
      "loss": 0.6496,
      "step": 1240
    },
    {
      "epoch": 0.10583354500042333,
      "grad_norm": 0.4394405484199524,
      "learning_rate": 2e-05,
      "loss": 0.6398,
      "step": 1250
    },
    {
      "epoch": 0.10668021336042673,
      "grad_norm": 0.43604040145874023,
      "learning_rate": 2e-05,
      "loss": 0.6502,
      "step": 1260
    },
    {
      "epoch": 0.10752688172043011,
      "grad_norm": 0.43875670433044434,
      "learning_rate": 2e-05,
      "loss": 0.6564,
      "step": 1270
    },
    {
      "epoch": 0.10837355008043349,
      "grad_norm": 0.4250807762145996,
      "learning_rate": 2e-05,
      "loss": 0.6389,
      "step": 1280
    },
    {
      "epoch": 0.10922021844043688,
      "grad_norm": 0.48220178484916687,
      "learning_rate": 2e-05,
      "loss": 0.6617,
      "step": 1290
    },
    {
      "epoch": 0.11006688680044027,
      "grad_norm": 0.41790634393692017,
      "learning_rate": 2e-05,
      "loss": 0.636,
      "step": 1300
    },
    {
      "epoch": 0.11091355516044366,
      "grad_norm": 0.5080624222755432,
      "learning_rate": 2e-05,
      "loss": 0.6733,
      "step": 1310
    },
    {
      "epoch": 0.11176022352044704,
      "grad_norm": 0.4096316695213318,
      "learning_rate": 2e-05,
      "loss": 0.6447,
      "step": 1320
    },
    {
      "epoch": 0.11260689188045042,
      "grad_norm": 0.4489736557006836,
      "learning_rate": 2e-05,
      "loss": 0.6723,
      "step": 1330
    },
    {
      "epoch": 0.11345356024045382,
      "grad_norm": 0.4486621022224426,
      "learning_rate": 2e-05,
      "loss": 0.6536,
      "step": 1340
    },
    {
      "epoch": 0.1143002286004572,
      "grad_norm": 0.7986853122711182,
      "learning_rate": 2e-05,
      "loss": 0.67,
      "step": 1350
    },
    {
      "epoch": 0.11514689696046058,
      "grad_norm": 0.4286498427391052,
      "learning_rate": 2e-05,
      "loss": 0.6441,
      "step": 1360
    },
    {
      "epoch": 0.11599356532046397,
      "grad_norm": 0.4120442271232605,
      "learning_rate": 2e-05,
      "loss": 0.6506,
      "step": 1370
    },
    {
      "epoch": 0.11684023368046736,
      "grad_norm": 0.43085169792175293,
      "learning_rate": 2e-05,
      "loss": 0.6505,
      "step": 1380
    },
    {
      "epoch": 0.11768690204047075,
      "grad_norm": 0.4295996129512787,
      "learning_rate": 2e-05,
      "loss": 0.6562,
      "step": 1390
    },
    {
      "epoch": 0.11853357040047413,
      "grad_norm": 0.43393197655677795,
      "learning_rate": 2e-05,
      "loss": 0.6433,
      "step": 1400
    },
    {
      "epoch": 0.11938023876047753,
      "grad_norm": 0.41177472472190857,
      "learning_rate": 2e-05,
      "loss": 0.6619,
      "step": 1410
    },
    {
      "epoch": 0.12022690712048091,
      "grad_norm": 0.4898812174797058,
      "learning_rate": 2e-05,
      "loss": 0.6394,
      "step": 1420
    },
    {
      "epoch": 0.1210735754804843,
      "grad_norm": 0.4619840085506439,
      "learning_rate": 2e-05,
      "loss": 0.6438,
      "step": 1430
    },
    {
      "epoch": 0.12192024384048768,
      "grad_norm": 0.5120882987976074,
      "learning_rate": 2e-05,
      "loss": 0.6645,
      "step": 1440
    },
    {
      "epoch": 0.12276691220049107,
      "grad_norm": 0.4201647937297821,
      "learning_rate": 2e-05,
      "loss": 0.6625,
      "step": 1450
    },
    {
      "epoch": 0.12361358056049446,
      "grad_norm": 0.4196993112564087,
      "learning_rate": 2e-05,
      "loss": 0.6477,
      "step": 1460
    },
    {
      "epoch": 0.12446024892049784,
      "grad_norm": 0.4666191637516022,
      "learning_rate": 2e-05,
      "loss": 0.6605,
      "step": 1470
    },
    {
      "epoch": 0.12530691728050122,
      "grad_norm": 0.42628341913223267,
      "learning_rate": 2e-05,
      "loss": 0.6635,
      "step": 1480
    },
    {
      "epoch": 0.1261535856405046,
      "grad_norm": 0.46635255217552185,
      "learning_rate": 2e-05,
      "loss": 0.6519,
      "step": 1490
    },
    {
      "epoch": 0.127000254000508,
      "grad_norm": 0.39531210064888,
      "learning_rate": 2e-05,
      "loss": 0.6597,
      "step": 1500
    },
    {
      "epoch": 0.1278469223605114,
      "grad_norm": 0.416293740272522,
      "learning_rate": 2e-05,
      "loss": 0.6476,
      "step": 1510
    },
    {
      "epoch": 0.12869359072051478,
      "grad_norm": 0.4376610815525055,
      "learning_rate": 2e-05,
      "loss": 0.6471,
      "step": 1520
    },
    {
      "epoch": 0.12954025908051817,
      "grad_norm": 0.9239709377288818,
      "learning_rate": 2e-05,
      "loss": 0.6483,
      "step": 1530
    },
    {
      "epoch": 0.13038692744052155,
      "grad_norm": 0.44836151599884033,
      "learning_rate": 2e-05,
      "loss": 0.6476,
      "step": 1540
    },
    {
      "epoch": 0.13123359580052493,
      "grad_norm": 0.40565115213394165,
      "learning_rate": 2e-05,
      "loss": 0.6477,
      "step": 1550
    },
    {
      "epoch": 0.13208026416052832,
      "grad_norm": 0.4043814241886139,
      "learning_rate": 2e-05,
      "loss": 0.6422,
      "step": 1560
    },
    {
      "epoch": 0.1329269325205317,
      "grad_norm": 0.48815831542015076,
      "learning_rate": 2e-05,
      "loss": 0.667,
      "step": 1570
    },
    {
      "epoch": 0.13377360088053508,
      "grad_norm": 0.4149971902370453,
      "learning_rate": 2e-05,
      "loss": 0.6428,
      "step": 1580
    },
    {
      "epoch": 0.1346202692405385,
      "grad_norm": 0.43481072783470154,
      "learning_rate": 2e-05,
      "loss": 0.6297,
      "step": 1590
    },
    {
      "epoch": 0.13546693760054188,
      "grad_norm": 0.5290552377700806,
      "learning_rate": 2e-05,
      "loss": 0.6571,
      "step": 1600
    },
    {
      "epoch": 0.13631360596054526,
      "grad_norm": 0.48005440831184387,
      "learning_rate": 2e-05,
      "loss": 0.6375,
      "step": 1610
    },
    {
      "epoch": 0.13716027432054864,
      "grad_norm": 0.41244378685951233,
      "learning_rate": 2e-05,
      "loss": 0.6465,
      "step": 1620
    },
    {
      "epoch": 0.13800694268055202,
      "grad_norm": 0.4275606572628021,
      "learning_rate": 2e-05,
      "loss": 0.6496,
      "step": 1630
    },
    {
      "epoch": 0.1388536110405554,
      "grad_norm": 0.40932148694992065,
      "learning_rate": 2e-05,
      "loss": 0.6617,
      "step": 1640
    },
    {
      "epoch": 0.1397002794005588,
      "grad_norm": 0.46909376978874207,
      "learning_rate": 2e-05,
      "loss": 0.654,
      "step": 1650
    },
    {
      "epoch": 0.14054694776056217,
      "grad_norm": 0.40658414363861084,
      "learning_rate": 2e-05,
      "loss": 0.6148,
      "step": 1660
    },
    {
      "epoch": 0.14139361612056559,
      "grad_norm": 0.45841172337532043,
      "learning_rate": 2e-05,
      "loss": 0.6511,
      "step": 1670
    },
    {
      "epoch": 0.14224028448056897,
      "grad_norm": 0.4471743702888489,
      "learning_rate": 2e-05,
      "loss": 0.6448,
      "step": 1680
    },
    {
      "epoch": 0.14308695284057235,
      "grad_norm": 0.43427541851997375,
      "learning_rate": 2e-05,
      "loss": 0.6435,
      "step": 1690
    },
    {
      "epoch": 0.14393362120057573,
      "grad_norm": 0.4472843408584595,
      "learning_rate": 2e-05,
      "loss": 0.6357,
      "step": 1700
    },
    {
      "epoch": 0.14478028956057912,
      "grad_norm": 0.42029255628585815,
      "learning_rate": 2e-05,
      "loss": 0.6575,
      "step": 1710
    },
    {
      "epoch": 0.1456269579205825,
      "grad_norm": 0.4621073603630066,
      "learning_rate": 2e-05,
      "loss": 0.6361,
      "step": 1720
    },
    {
      "epoch": 0.14647362628058588,
      "grad_norm": 0.4484656751155853,
      "learning_rate": 2e-05,
      "loss": 0.6355,
      "step": 1730
    },
    {
      "epoch": 0.1473202946405893,
      "grad_norm": 0.4217395782470703,
      "learning_rate": 2e-05,
      "loss": 0.6682,
      "step": 1740
    },
    {
      "epoch": 0.14816696300059268,
      "grad_norm": 0.42961055040359497,
      "learning_rate": 2e-05,
      "loss": 0.645,
      "step": 1750
    },
    {
      "epoch": 0.14901363136059606,
      "grad_norm": 0.46678170561790466,
      "learning_rate": 2e-05,
      "loss": 0.6401,
      "step": 1760
    },
    {
      "epoch": 0.14986029972059944,
      "grad_norm": 0.5399910807609558,
      "learning_rate": 2e-05,
      "loss": 0.6507,
      "step": 1770
    },
    {
      "epoch": 0.15070696808060283,
      "grad_norm": 0.47768858075141907,
      "learning_rate": 2e-05,
      "loss": 0.6342,
      "step": 1780
    },
    {
      "epoch": 0.1515536364406062,
      "grad_norm": 0.44225838780403137,
      "learning_rate": 2e-05,
      "loss": 0.6567,
      "step": 1790
    },
    {
      "epoch": 0.1524003048006096,
      "grad_norm": 0.4229118227958679,
      "learning_rate": 2e-05,
      "loss": 0.6597,
      "step": 1800
    },
    {
      "epoch": 0.15324697316061298,
      "grad_norm": 1.9519200325012207,
      "learning_rate": 2e-05,
      "loss": 0.6483,
      "step": 1810
    },
    {
      "epoch": 0.1540936415206164,
      "grad_norm": 0.5570772290229797,
      "learning_rate": 2e-05,
      "loss": 0.646,
      "step": 1820
    },
    {
      "epoch": 0.15494030988061977,
      "grad_norm": 0.42859718203544617,
      "learning_rate": 2e-05,
      "loss": 0.6163,
      "step": 1830
    },
    {
      "epoch": 0.15578697824062315,
      "grad_norm": 0.4103221297264099,
      "learning_rate": 2e-05,
      "loss": 0.6527,
      "step": 1840
    },
    {
      "epoch": 0.15663364660062654,
      "grad_norm": 0.4725561738014221,
      "learning_rate": 2e-05,
      "loss": 0.6624,
      "step": 1850
    },
    {
      "epoch": 0.15748031496062992,
      "grad_norm": 0.43084171414375305,
      "learning_rate": 2e-05,
      "loss": 0.6424,
      "step": 1860
    },
    {
      "epoch": 0.1583269833206333,
      "grad_norm": 0.4549056887626648,
      "learning_rate": 2e-05,
      "loss": 0.6319,
      "step": 1870
    },
    {
      "epoch": 0.15917365168063669,
      "grad_norm": 0.43331679701805115,
      "learning_rate": 2e-05,
      "loss": 0.6443,
      "step": 1880
    },
    {
      "epoch": 0.16002032004064007,
      "grad_norm": 0.4841455817222595,
      "learning_rate": 2e-05,
      "loss": 0.6349,
      "step": 1890
    },
    {
      "epoch": 0.16086698840064348,
      "grad_norm": 0.4443437159061432,
      "learning_rate": 2e-05,
      "loss": 0.6631,
      "step": 1900
    },
    {
      "epoch": 0.16171365676064686,
      "grad_norm": 0.44019854068756104,
      "learning_rate": 2e-05,
      "loss": 0.6318,
      "step": 1910
    },
    {
      "epoch": 0.16256032512065025,
      "grad_norm": 0.45389479398727417,
      "learning_rate": 2e-05,
      "loss": 0.6717,
      "step": 1920
    },
    {
      "epoch": 0.16340699348065363,
      "grad_norm": 0.4448392689228058,
      "learning_rate": 2e-05,
      "loss": 0.6636,
      "step": 1930
    },
    {
      "epoch": 0.164253661840657,
      "grad_norm": 0.48700398206710815,
      "learning_rate": 2e-05,
      "loss": 0.6337,
      "step": 1940
    },
    {
      "epoch": 0.1651003302006604,
      "grad_norm": 0.42143478989601135,
      "learning_rate": 2e-05,
      "loss": 0.6544,
      "step": 1950
    },
    {
      "epoch": 0.16594699856066378,
      "grad_norm": 0.4352879524230957,
      "learning_rate": 2e-05,
      "loss": 0.6514,
      "step": 1960
    },
    {
      "epoch": 0.1667936669206672,
      "grad_norm": 0.414305716753006,
      "learning_rate": 2e-05,
      "loss": 0.6481,
      "step": 1970
    },
    {
      "epoch": 0.16764033528067057,
      "grad_norm": 0.39161503314971924,
      "learning_rate": 2e-05,
      "loss": 0.6566,
      "step": 1980
    },
    {
      "epoch": 0.16848700364067395,
      "grad_norm": 0.421457976102829,
      "learning_rate": 2e-05,
      "loss": 0.6449,
      "step": 1990
    },
    {
      "epoch": 0.16933367200067734,
      "grad_norm": 0.448140949010849,
      "learning_rate": 2e-05,
      "loss": 0.6477,
      "step": 2000
    },
    {
      "epoch": 0.17018034036068072,
      "grad_norm": 0.4220173954963684,
      "learning_rate": 2e-05,
      "loss": 0.6366,
      "step": 2010
    },
    {
      "epoch": 0.1710270087206841,
      "grad_norm": 0.3872855603694916,
      "learning_rate": 2e-05,
      "loss": 0.6646,
      "step": 2020
    },
    {
      "epoch": 0.1718736770806875,
      "grad_norm": 0.5094623565673828,
      "learning_rate": 2e-05,
      "loss": 0.6433,
      "step": 2030
    },
    {
      "epoch": 0.17272034544069087,
      "grad_norm": 0.43138524889945984,
      "learning_rate": 2e-05,
      "loss": 0.6399,
      "step": 2040
    },
    {
      "epoch": 0.17356701380069428,
      "grad_norm": 0.4579676687717438,
      "learning_rate": 2e-05,
      "loss": 0.643,
      "step": 2050
    },
    {
      "epoch": 0.17441368216069766,
      "grad_norm": 1.0642181634902954,
      "learning_rate": 2e-05,
      "loss": 0.6466,
      "step": 2060
    },
    {
      "epoch": 0.17526035052070105,
      "grad_norm": 0.4300776422023773,
      "learning_rate": 2e-05,
      "loss": 0.6264,
      "step": 2070
    },
    {
      "epoch": 0.17610701888070443,
      "grad_norm": 0.43561115860939026,
      "learning_rate": 2e-05,
      "loss": 0.6274,
      "step": 2080
    },
    {
      "epoch": 0.1769536872407078,
      "grad_norm": 0.4419814348220825,
      "learning_rate": 2e-05,
      "loss": 0.6189,
      "step": 2090
    },
    {
      "epoch": 0.1778003556007112,
      "grad_norm": 0.40943095088005066,
      "learning_rate": 2e-05,
      "loss": 0.6575,
      "step": 2100
    },
    {
      "epoch": 0.17864702396071458,
      "grad_norm": 0.4220316708087921,
      "learning_rate": 2e-05,
      "loss": 0.6321,
      "step": 2110
    },
    {
      "epoch": 0.17949369232071796,
      "grad_norm": 0.4069748818874359,
      "learning_rate": 2e-05,
      "loss": 0.6249,
      "step": 2120
    },
    {
      "epoch": 0.18034036068072137,
      "grad_norm": 0.4093238115310669,
      "learning_rate": 2e-05,
      "loss": 0.6367,
      "step": 2130
    },
    {
      "epoch": 0.18118702904072476,
      "grad_norm": 0.42871996760368347,
      "learning_rate": 2e-05,
      "loss": 0.6319,
      "step": 2140
    },
    {
      "epoch": 0.18203369740072814,
      "grad_norm": 0.4586758613586426,
      "learning_rate": 2e-05,
      "loss": 0.6043,
      "step": 2150
    },
    {
      "epoch": 0.18288036576073152,
      "grad_norm": 0.43323299288749695,
      "learning_rate": 2e-05,
      "loss": 0.6518,
      "step": 2160
    },
    {
      "epoch": 0.1837270341207349,
      "grad_norm": 0.4545518159866333,
      "learning_rate": 2e-05,
      "loss": 0.6365,
      "step": 2170
    },
    {
      "epoch": 0.1845737024807383,
      "grad_norm": 0.4953419268131256,
      "learning_rate": 2e-05,
      "loss": 0.6426,
      "step": 2180
    },
    {
      "epoch": 0.18542037084074167,
      "grad_norm": 0.4099434018135071,
      "learning_rate": 2e-05,
      "loss": 0.6429,
      "step": 2190
    },
    {
      "epoch": 0.18626703920074505,
      "grad_norm": 0.4271218180656433,
      "learning_rate": 2e-05,
      "loss": 0.6341,
      "step": 2200
    },
    {
      "epoch": 0.18711370756074847,
      "grad_norm": 0.44617754220962524,
      "learning_rate": 2e-05,
      "loss": 0.6261,
      "step": 2210
    },
    {
      "epoch": 0.18796037592075185,
      "grad_norm": 0.4889068901538849,
      "learning_rate": 2e-05,
      "loss": 0.6611,
      "step": 2220
    },
    {
      "epoch": 0.18880704428075523,
      "grad_norm": 0.41321495175361633,
      "learning_rate": 2e-05,
      "loss": 0.645,
      "step": 2230
    },
    {
      "epoch": 0.18965371264075861,
      "grad_norm": 0.3996637463569641,
      "learning_rate": 2e-05,
      "loss": 0.6471,
      "step": 2240
    },
    {
      "epoch": 0.190500381000762,
      "grad_norm": 0.43380239605903625,
      "learning_rate": 2e-05,
      "loss": 0.6328,
      "step": 2250
    },
    {
      "epoch": 0.19134704936076538,
      "grad_norm": 0.43314844369888306,
      "learning_rate": 2e-05,
      "loss": 0.6528,
      "step": 2260
    },
    {
      "epoch": 0.19219371772076876,
      "grad_norm": 0.396233469247818,
      "learning_rate": 2e-05,
      "loss": 0.6331,
      "step": 2270
    },
    {
      "epoch": 0.19304038608077217,
      "grad_norm": 0.42943382263183594,
      "learning_rate": 2e-05,
      "loss": 0.6604,
      "step": 2280
    },
    {
      "epoch": 0.19388705444077556,
      "grad_norm": 0.5218196511268616,
      "learning_rate": 2e-05,
      "loss": 0.6573,
      "step": 2290
    },
    {
      "epoch": 0.19473372280077894,
      "grad_norm": 0.4184105396270752,
      "learning_rate": 2e-05,
      "loss": 0.6361,
      "step": 2300
    },
    {
      "epoch": 0.19558039116078232,
      "grad_norm": 0.5651347041130066,
      "learning_rate": 2e-05,
      "loss": 0.636,
      "step": 2310
    },
    {
      "epoch": 0.1964270595207857,
      "grad_norm": 0.42529258131980896,
      "learning_rate": 2e-05,
      "loss": 0.6444,
      "step": 2320
    },
    {
      "epoch": 0.1972737278807891,
      "grad_norm": 0.44605952501296997,
      "learning_rate": 2e-05,
      "loss": 0.6382,
      "step": 2330
    },
    {
      "epoch": 0.19812039624079247,
      "grad_norm": 0.4468037188053131,
      "learning_rate": 2e-05,
      "loss": 0.6302,
      "step": 2340
    },
    {
      "epoch": 0.19896706460079586,
      "grad_norm": 0.3863542973995209,
      "learning_rate": 2e-05,
      "loss": 0.6215,
      "step": 2350
    },
    {
      "epoch": 0.19981373296079927,
      "grad_norm": 0.4320015013217926,
      "learning_rate": 2e-05,
      "loss": 0.6421,
      "step": 2360
    },
    {
      "epoch": 0.20066040132080265,
      "grad_norm": 0.43013012409210205,
      "learning_rate": 2e-05,
      "loss": 0.6474,
      "step": 2370
    },
    {
      "epoch": 0.20150706968080603,
      "grad_norm": 0.3912484049797058,
      "learning_rate": 2e-05,
      "loss": 0.6179,
      "step": 2380
    },
    {
      "epoch": 0.20235373804080942,
      "grad_norm": 0.39495962858200073,
      "learning_rate": 2e-05,
      "loss": 0.6356,
      "step": 2390
    },
    {
      "epoch": 0.2032004064008128,
      "grad_norm": 0.384694367647171,
      "learning_rate": 2e-05,
      "loss": 0.6634,
      "step": 2400
    },
    {
      "epoch": 0.20404707476081618,
      "grad_norm": 0.6632846593856812,
      "learning_rate": 2e-05,
      "loss": 0.6283,
      "step": 2410
    },
    {
      "epoch": 0.20489374312081957,
      "grad_norm": 0.4242749512195587,
      "learning_rate": 2e-05,
      "loss": 0.6316,
      "step": 2420
    },
    {
      "epoch": 0.20574041148082295,
      "grad_norm": 0.41551467776298523,
      "learning_rate": 2e-05,
      "loss": 0.6519,
      "step": 2430
    },
    {
      "epoch": 0.20658707984082636,
      "grad_norm": 0.3853192925453186,
      "learning_rate": 2e-05,
      "loss": 0.6487,
      "step": 2440
    },
    {
      "epoch": 0.20743374820082974,
      "grad_norm": 0.4175145626068115,
      "learning_rate": 2e-05,
      "loss": 0.6397,
      "step": 2450
    },
    {
      "epoch": 0.20828041656083313,
      "grad_norm": 0.41761279106140137,
      "learning_rate": 2e-05,
      "loss": 0.656,
      "step": 2460
    },
    {
      "epoch": 0.2091270849208365,
      "grad_norm": 0.39868006110191345,
      "learning_rate": 2e-05,
      "loss": 0.6412,
      "step": 2470
    },
    {
      "epoch": 0.2099737532808399,
      "grad_norm": 0.3972877264022827,
      "learning_rate": 2e-05,
      "loss": 0.6403,
      "step": 2480
    },
    {
      "epoch": 0.21082042164084328,
      "grad_norm": 0.3830007314682007,
      "learning_rate": 2e-05,
      "loss": 0.6371,
      "step": 2490
    },
    {
      "epoch": 0.21166709000084666,
      "grad_norm": 0.4232046604156494,
      "learning_rate": 2e-05,
      "loss": 0.6531,
      "step": 2500
    },
    {
      "epoch": 0.21251375836085004,
      "grad_norm": 0.3900882303714752,
      "learning_rate": 2e-05,
      "loss": 0.6222,
      "step": 2510
    },
    {
      "epoch": 0.21336042672085345,
      "grad_norm": 0.39245057106018066,
      "learning_rate": 2e-05,
      "loss": 0.6153,
      "step": 2520
    },
    {
      "epoch": 0.21420709508085684,
      "grad_norm": 0.43386736512184143,
      "learning_rate": 2e-05,
      "loss": 0.6382,
      "step": 2530
    },
    {
      "epoch": 0.21505376344086022,
      "grad_norm": 0.3844192326068878,
      "learning_rate": 2e-05,
      "loss": 0.6534,
      "step": 2540
    },
    {
      "epoch": 0.2159004318008636,
      "grad_norm": 0.5519846081733704,
      "learning_rate": 2e-05,
      "loss": 0.6464,
      "step": 2550
    },
    {
      "epoch": 0.21674710016086698,
      "grad_norm": 0.4004225432872772,
      "learning_rate": 2e-05,
      "loss": 0.6385,
      "step": 2560
    },
    {
      "epoch": 0.21759376852087037,
      "grad_norm": 0.40917128324508667,
      "learning_rate": 2e-05,
      "loss": 0.6556,
      "step": 2570
    },
    {
      "epoch": 0.21844043688087375,
      "grad_norm": 0.38037240505218506,
      "learning_rate": 2e-05,
      "loss": 0.6359,
      "step": 2580
    },
    {
      "epoch": 0.21928710524087716,
      "grad_norm": 0.4489869475364685,
      "learning_rate": 2e-05,
      "loss": 0.655,
      "step": 2590
    },
    {
      "epoch": 0.22013377360088054,
      "grad_norm": 0.41383644938468933,
      "learning_rate": 2e-05,
      "loss": 0.6554,
      "step": 2600
    },
    {
      "epoch": 0.22098044196088393,
      "grad_norm": 0.3828672766685486,
      "learning_rate": 2e-05,
      "loss": 0.6588,
      "step": 2610
    },
    {
      "epoch": 0.2218271103208873,
      "grad_norm": 0.41098809242248535,
      "learning_rate": 2e-05,
      "loss": 0.6185,
      "step": 2620
    },
    {
      "epoch": 0.2226737786808907,
      "grad_norm": 0.3833703100681305,
      "learning_rate": 2e-05,
      "loss": 0.6419,
      "step": 2630
    },
    {
      "epoch": 0.22352044704089408,
      "grad_norm": 0.4067264199256897,
      "learning_rate": 2e-05,
      "loss": 0.6289,
      "step": 2640
    },
    {
      "epoch": 0.22436711540089746,
      "grad_norm": 0.40791037678718567,
      "learning_rate": 2e-05,
      "loss": 0.6379,
      "step": 2650
    },
    {
      "epoch": 0.22521378376090084,
      "grad_norm": 0.43709638714790344,
      "learning_rate": 2e-05,
      "loss": 0.6357,
      "step": 2660
    },
    {
      "epoch": 0.22606045212090425,
      "grad_norm": 0.4151391088962555,
      "learning_rate": 2e-05,
      "loss": 0.6571,
      "step": 2670
    },
    {
      "epoch": 0.22690712048090764,
      "grad_norm": 0.4648939073085785,
      "learning_rate": 2e-05,
      "loss": 0.6519,
      "step": 2680
    },
    {
      "epoch": 0.22775378884091102,
      "grad_norm": 0.42384690046310425,
      "learning_rate": 2e-05,
      "loss": 0.6406,
      "step": 2690
    },
    {
      "epoch": 0.2286004572009144,
      "grad_norm": 0.40418389439582825,
      "learning_rate": 2e-05,
      "loss": 0.6352,
      "step": 2700
    },
    {
      "epoch": 0.2294471255609178,
      "grad_norm": 0.4247417747974396,
      "learning_rate": 2e-05,
      "loss": 0.6246,
      "step": 2710
    },
    {
      "epoch": 0.23029379392092117,
      "grad_norm": 0.39670681953430176,
      "learning_rate": 2e-05,
      "loss": 0.6463,
      "step": 2720
    },
    {
      "epoch": 0.23114046228092455,
      "grad_norm": 0.45522087812423706,
      "learning_rate": 2e-05,
      "loss": 0.6687,
      "step": 2730
    },
    {
      "epoch": 0.23198713064092794,
      "grad_norm": 0.4865119457244873,
      "learning_rate": 2e-05,
      "loss": 0.6498,
      "step": 2740
    },
    {
      "epoch": 0.23283379900093135,
      "grad_norm": 0.4382932782173157,
      "learning_rate": 2e-05,
      "loss": 0.6406,
      "step": 2750
    },
    {
      "epoch": 0.23368046736093473,
      "grad_norm": 0.42918914556503296,
      "learning_rate": 2e-05,
      "loss": 0.6431,
      "step": 2760
    },
    {
      "epoch": 0.2345271357209381,
      "grad_norm": 0.41465798020362854,
      "learning_rate": 2e-05,
      "loss": 0.6406,
      "step": 2770
    },
    {
      "epoch": 0.2353738040809415,
      "grad_norm": 0.4464358389377594,
      "learning_rate": 2e-05,
      "loss": 0.6491,
      "step": 2780
    },
    {
      "epoch": 0.23622047244094488,
      "grad_norm": 0.4201108515262604,
      "learning_rate": 2e-05,
      "loss": 0.6199,
      "step": 2790
    },
    {
      "epoch": 0.23706714080094826,
      "grad_norm": 0.4981282949447632,
      "learning_rate": 2e-05,
      "loss": 0.635,
      "step": 2800
    },
    {
      "epoch": 0.23791380916095164,
      "grad_norm": 0.3868878483772278,
      "learning_rate": 2e-05,
      "loss": 0.6393,
      "step": 2810
    },
    {
      "epoch": 0.23876047752095506,
      "grad_norm": 0.40685951709747314,
      "learning_rate": 2e-05,
      "loss": 0.6284,
      "step": 2820
    },
    {
      "epoch": 0.23960714588095844,
      "grad_norm": 0.41393551230430603,
      "learning_rate": 2e-05,
      "loss": 0.6568,
      "step": 2830
    },
    {
      "epoch": 0.24045381424096182,
      "grad_norm": 0.4491131603717804,
      "learning_rate": 2e-05,
      "loss": 0.6375,
      "step": 2840
    },
    {
      "epoch": 0.2413004826009652,
      "grad_norm": 0.48691219091415405,
      "learning_rate": 2e-05,
      "loss": 0.6143,
      "step": 2850
    },
    {
      "epoch": 0.2421471509609686,
      "grad_norm": 0.4507281184196472,
      "learning_rate": 2e-05,
      "loss": 0.6234,
      "step": 2860
    },
    {
      "epoch": 0.24299381932097197,
      "grad_norm": 0.43581560254096985,
      "learning_rate": 2e-05,
      "loss": 0.6454,
      "step": 2870
    },
    {
      "epoch": 0.24384048768097535,
      "grad_norm": 0.5282661318778992,
      "learning_rate": 2e-05,
      "loss": 0.6423,
      "step": 2880
    },
    {
      "epoch": 0.24468715604097874,
      "grad_norm": 0.40446242690086365,
      "learning_rate": 2e-05,
      "loss": 0.6406,
      "step": 2890
    },
    {
      "epoch": 0.24553382440098215,
      "grad_norm": 0.4218043088912964,
      "learning_rate": 2e-05,
      "loss": 0.6242,
      "step": 2900
    },
    {
      "epoch": 0.24638049276098553,
      "grad_norm": 0.4034806191921234,
      "learning_rate": 2e-05,
      "loss": 0.6306,
      "step": 2910
    },
    {
      "epoch": 0.24722716112098891,
      "grad_norm": 0.40466344356536865,
      "learning_rate": 2e-05,
      "loss": 0.6261,
      "step": 2920
    },
    {
      "epoch": 0.2480738294809923,
      "grad_norm": 0.4296315312385559,
      "learning_rate": 2e-05,
      "loss": 0.6151,
      "step": 2930
    },
    {
      "epoch": 0.24892049784099568,
      "grad_norm": 0.45331016182899475,
      "learning_rate": 2e-05,
      "loss": 0.6464,
      "step": 2940
    },
    {
      "epoch": 0.24976716620099906,
      "grad_norm": 0.44625458121299744,
      "learning_rate": 2e-05,
      "loss": 0.6315,
      "step": 2950
    },
    {
      "epoch": 0.25061383456100245,
      "grad_norm": 0.43744179606437683,
      "learning_rate": 2e-05,
      "loss": 0.619,
      "step": 2960
    },
    {
      "epoch": 0.25146050292100586,
      "grad_norm": 0.4107581377029419,
      "learning_rate": 2e-05,
      "loss": 0.6306,
      "step": 2970
    },
    {
      "epoch": 0.2523071712810092,
      "grad_norm": 0.39485299587249756,
      "learning_rate": 2e-05,
      "loss": 0.6261,
      "step": 2980
    },
    {
      "epoch": 0.2531538396410126,
      "grad_norm": 0.4031531810760498,
      "learning_rate": 2e-05,
      "loss": 0.6257,
      "step": 2990
    },
    {
      "epoch": 0.254000508001016,
      "grad_norm": 0.45026257634162903,
      "learning_rate": 2e-05,
      "loss": 0.6408,
      "step": 3000
    },
    {
      "epoch": 0.2548471763610194,
      "grad_norm": 0.40911462903022766,
      "learning_rate": 2e-05,
      "loss": 0.624,
      "step": 3010
    },
    {
      "epoch": 0.2556938447210228,
      "grad_norm": 0.3996413052082062,
      "learning_rate": 2e-05,
      "loss": 0.6388,
      "step": 3020
    },
    {
      "epoch": 0.25654051308102616,
      "grad_norm": 0.4137783944606781,
      "learning_rate": 2e-05,
      "loss": 0.645,
      "step": 3030
    },
    {
      "epoch": 0.25738718144102957,
      "grad_norm": 0.3941035866737366,
      "learning_rate": 2e-05,
      "loss": 0.6297,
      "step": 3040
    },
    {
      "epoch": 0.2582338498010329,
      "grad_norm": 0.39765384793281555,
      "learning_rate": 2e-05,
      "loss": 0.675,
      "step": 3050
    },
    {
      "epoch": 0.25908051816103633,
      "grad_norm": 0.42631596326828003,
      "learning_rate": 2e-05,
      "loss": 0.6185,
      "step": 3060
    },
    {
      "epoch": 0.2599271865210397,
      "grad_norm": 0.4218185544013977,
      "learning_rate": 2e-05,
      "loss": 0.6385,
      "step": 3070
    },
    {
      "epoch": 0.2607738548810431,
      "grad_norm": 0.42259150743484497,
      "learning_rate": 2e-05,
      "loss": 0.6464,
      "step": 3080
    },
    {
      "epoch": 0.2616205232410465,
      "grad_norm": 0.3934686481952667,
      "learning_rate": 2e-05,
      "loss": 0.6399,
      "step": 3090
    },
    {
      "epoch": 0.26246719160104987,
      "grad_norm": 0.4417162537574768,
      "learning_rate": 2e-05,
      "loss": 0.6473,
      "step": 3100
    },
    {
      "epoch": 0.2633138599610533,
      "grad_norm": 0.4261956810951233,
      "learning_rate": 2e-05,
      "loss": 0.6135,
      "step": 3110
    },
    {
      "epoch": 0.26416052832105663,
      "grad_norm": 0.4082337021827698,
      "learning_rate": 2e-05,
      "loss": 0.6437,
      "step": 3120
    },
    {
      "epoch": 0.26500719668106004,
      "grad_norm": 0.4085938632488251,
      "learning_rate": 2e-05,
      "loss": 0.6214,
      "step": 3130
    },
    {
      "epoch": 0.2658538650410634,
      "grad_norm": 0.42370662093162537,
      "learning_rate": 2e-05,
      "loss": 0.6624,
      "step": 3140
    },
    {
      "epoch": 0.2667005334010668,
      "grad_norm": 0.9658306837081909,
      "learning_rate": 2e-05,
      "loss": 0.6355,
      "step": 3150
    },
    {
      "epoch": 0.26754720176107016,
      "grad_norm": 0.3799201548099518,
      "learning_rate": 2e-05,
      "loss": 0.6338,
      "step": 3160
    },
    {
      "epoch": 0.2683938701210736,
      "grad_norm": 0.4011082351207733,
      "learning_rate": 2e-05,
      "loss": 0.6205,
      "step": 3170
    },
    {
      "epoch": 0.269240538481077,
      "grad_norm": 0.4082750976085663,
      "learning_rate": 2e-05,
      "loss": 0.646,
      "step": 3180
    },
    {
      "epoch": 0.27008720684108034,
      "grad_norm": 0.3899262845516205,
      "learning_rate": 2e-05,
      "loss": 0.6287,
      "step": 3190
    },
    {
      "epoch": 0.27093387520108375,
      "grad_norm": 0.4211922585964203,
      "learning_rate": 2e-05,
      "loss": 0.6306,
      "step": 3200
    },
    {
      "epoch": 0.2717805435610871,
      "grad_norm": 0.41326385736465454,
      "learning_rate": 2e-05,
      "loss": 0.6235,
      "step": 3210
    },
    {
      "epoch": 0.2726272119210905,
      "grad_norm": 0.41445034742355347,
      "learning_rate": 2e-05,
      "loss": 0.6239,
      "step": 3220
    },
    {
      "epoch": 0.2734738802810939,
      "grad_norm": 0.4477451741695404,
      "learning_rate": 2e-05,
      "loss": 0.6308,
      "step": 3230
    },
    {
      "epoch": 0.2743205486410973,
      "grad_norm": 0.42119231820106506,
      "learning_rate": 2e-05,
      "loss": 0.6257,
      "step": 3240
    },
    {
      "epoch": 0.2751672170011007,
      "grad_norm": 0.4201563596725464,
      "learning_rate": 2e-05,
      "loss": 0.6225,
      "step": 3250
    },
    {
      "epoch": 0.27601388536110405,
      "grad_norm": 0.42233985662460327,
      "learning_rate": 2e-05,
      "loss": 0.6166,
      "step": 3260
    },
    {
      "epoch": 0.27686055372110746,
      "grad_norm": 0.40148264169692993,
      "learning_rate": 2e-05,
      "loss": 0.6442,
      "step": 3270
    },
    {
      "epoch": 0.2777072220811108,
      "grad_norm": 0.3941594660282135,
      "learning_rate": 2e-05,
      "loss": 0.6369,
      "step": 3280
    },
    {
      "epoch": 0.2785538904411142,
      "grad_norm": 0.40674856305122375,
      "learning_rate": 2e-05,
      "loss": 0.6351,
      "step": 3290
    },
    {
      "epoch": 0.2794005588011176,
      "grad_norm": 0.39532068371772766,
      "learning_rate": 2e-05,
      "loss": 0.6355,
      "step": 3300
    },
    {
      "epoch": 0.280247227161121,
      "grad_norm": 0.3713996708393097,
      "learning_rate": 2e-05,
      "loss": 0.6266,
      "step": 3310
    },
    {
      "epoch": 0.28109389552112435,
      "grad_norm": 0.40361613035202026,
      "learning_rate": 2e-05,
      "loss": 0.6483,
      "step": 3320
    },
    {
      "epoch": 0.28194056388112776,
      "grad_norm": 0.3825480341911316,
      "learning_rate": 2e-05,
      "loss": 0.6168,
      "step": 3330
    },
    {
      "epoch": 0.28278723224113117,
      "grad_norm": 0.38089442253112793,
      "learning_rate": 2e-05,
      "loss": 0.629,
      "step": 3340
    },
    {
      "epoch": 0.2836339006011345,
      "grad_norm": 0.49822020530700684,
      "learning_rate": 2e-05,
      "loss": 0.6532,
      "step": 3350
    },
    {
      "epoch": 0.28448056896113794,
      "grad_norm": 0.4202452301979065,
      "learning_rate": 2e-05,
      "loss": 0.6299,
      "step": 3360
    },
    {
      "epoch": 0.2853272373211413,
      "grad_norm": 0.37910890579223633,
      "learning_rate": 2e-05,
      "loss": 0.6348,
      "step": 3370
    },
    {
      "epoch": 0.2861739056811447,
      "grad_norm": 0.39863014221191406,
      "learning_rate": 2e-05,
      "loss": 0.6369,
      "step": 3380
    },
    {
      "epoch": 0.28702057404114806,
      "grad_norm": 0.5652646422386169,
      "learning_rate": 2e-05,
      "loss": 0.6184,
      "step": 3390
    },
    {
      "epoch": 0.28786724240115147,
      "grad_norm": 0.37824442982673645,
      "learning_rate": 2e-05,
      "loss": 0.6184,
      "step": 3400
    },
    {
      "epoch": 0.2887139107611549,
      "grad_norm": 0.4170939028263092,
      "learning_rate": 2e-05,
      "loss": 0.6304,
      "step": 3410
    },
    {
      "epoch": 0.28956057912115823,
      "grad_norm": 0.5684532523155212,
      "learning_rate": 2e-05,
      "loss": 0.6384,
      "step": 3420
    },
    {
      "epoch": 0.29040724748116165,
      "grad_norm": 0.4120194911956787,
      "learning_rate": 2e-05,
      "loss": 0.6106,
      "step": 3430
    },
    {
      "epoch": 0.291253915841165,
      "grad_norm": 0.40495702624320984,
      "learning_rate": 2e-05,
      "loss": 0.6495,
      "step": 3440
    },
    {
      "epoch": 0.2921005842011684,
      "grad_norm": 0.4259183704853058,
      "learning_rate": 2e-05,
      "loss": 0.638,
      "step": 3450
    },
    {
      "epoch": 0.29294725256117177,
      "grad_norm": 0.4299517273902893,
      "learning_rate": 2e-05,
      "loss": 0.6245,
      "step": 3460
    },
    {
      "epoch": 0.2937939209211752,
      "grad_norm": 0.43136143684387207,
      "learning_rate": 2e-05,
      "loss": 0.6302,
      "step": 3470
    },
    {
      "epoch": 0.2946405892811786,
      "grad_norm": 0.37267258763313293,
      "learning_rate": 2e-05,
      "loss": 0.6227,
      "step": 3480
    },
    {
      "epoch": 0.29548725764118194,
      "grad_norm": 0.4441055655479431,
      "learning_rate": 2e-05,
      "loss": 0.6242,
      "step": 3490
    },
    {
      "epoch": 0.29633392600118535,
      "grad_norm": 0.43486303091049194,
      "learning_rate": 2e-05,
      "loss": 0.6218,
      "step": 3500
    },
    {
      "epoch": 0.2971805943611887,
      "grad_norm": 0.4182334244251251,
      "learning_rate": 2e-05,
      "loss": 0.6393,
      "step": 3510
    },
    {
      "epoch": 0.2980272627211921,
      "grad_norm": 0.4296554923057556,
      "learning_rate": 2e-05,
      "loss": 0.6545,
      "step": 3520
    },
    {
      "epoch": 0.2988739310811955,
      "grad_norm": 0.4029640257358551,
      "learning_rate": 2e-05,
      "loss": 0.6267,
      "step": 3530
    },
    {
      "epoch": 0.2997205994411989,
      "grad_norm": 0.5285282731056213,
      "learning_rate": 2e-05,
      "loss": 0.6334,
      "step": 3540
    },
    {
      "epoch": 0.30056726780120224,
      "grad_norm": 0.4067811071872711,
      "learning_rate": 2e-05,
      "loss": 0.6183,
      "step": 3550
    },
    {
      "epoch": 0.30141393616120565,
      "grad_norm": 0.40235307812690735,
      "learning_rate": 2e-05,
      "loss": 0.6463,
      "step": 3560
    },
    {
      "epoch": 0.30226060452120906,
      "grad_norm": 0.39202913641929626,
      "learning_rate": 2e-05,
      "loss": 0.6578,
      "step": 3570
    },
    {
      "epoch": 0.3031072728812124,
      "grad_norm": 0.43644803762435913,
      "learning_rate": 2e-05,
      "loss": 0.6139,
      "step": 3580
    },
    {
      "epoch": 0.30395394124121583,
      "grad_norm": 0.4295779764652252,
      "learning_rate": 2e-05,
      "loss": 0.6569,
      "step": 3590
    },
    {
      "epoch": 0.3048006096012192,
      "grad_norm": 0.3815408945083618,
      "learning_rate": 2e-05,
      "loss": 0.6371,
      "step": 3600
    },
    {
      "epoch": 0.3056472779612226,
      "grad_norm": 0.3733167350292206,
      "learning_rate": 2e-05,
      "loss": 0.6311,
      "step": 3610
    },
    {
      "epoch": 0.30649394632122595,
      "grad_norm": 0.40843579173088074,
      "learning_rate": 2e-05,
      "loss": 0.6308,
      "step": 3620
    },
    {
      "epoch": 0.30734061468122936,
      "grad_norm": 0.39684557914733887,
      "learning_rate": 2e-05,
      "loss": 0.611,
      "step": 3630
    },
    {
      "epoch": 0.3081872830412328,
      "grad_norm": 0.3786352872848511,
      "learning_rate": 2e-05,
      "loss": 0.6298,
      "step": 3640
    },
    {
      "epoch": 0.30903395140123613,
      "grad_norm": 0.3809105157852173,
      "learning_rate": 2e-05,
      "loss": 0.6223,
      "step": 3650
    },
    {
      "epoch": 0.30988061976123954,
      "grad_norm": 0.3611084818840027,
      "learning_rate": 2e-05,
      "loss": 0.623,
      "step": 3660
    },
    {
      "epoch": 0.3107272881212429,
      "grad_norm": 0.4300227761268616,
      "learning_rate": 2e-05,
      "loss": 0.6246,
      "step": 3670
    },
    {
      "epoch": 0.3115739564812463,
      "grad_norm": 0.38485798239707947,
      "learning_rate": 2e-05,
      "loss": 0.6279,
      "step": 3680
    },
    {
      "epoch": 0.31242062484124966,
      "grad_norm": 0.4136834442615509,
      "learning_rate": 2e-05,
      "loss": 0.6415,
      "step": 3690
    },
    {
      "epoch": 0.31326729320125307,
      "grad_norm": 0.3984295427799225,
      "learning_rate": 2e-05,
      "loss": 0.6462,
      "step": 3700
    },
    {
      "epoch": 0.3141139615612565,
      "grad_norm": 0.4182433485984802,
      "learning_rate": 2e-05,
      "loss": 0.606,
      "step": 3710
    },
    {
      "epoch": 0.31496062992125984,
      "grad_norm": 0.4438022971153259,
      "learning_rate": 2e-05,
      "loss": 0.6558,
      "step": 3720
    },
    {
      "epoch": 0.31580729828126325,
      "grad_norm": 0.41767048835754395,
      "learning_rate": 2e-05,
      "loss": 0.6353,
      "step": 3730
    },
    {
      "epoch": 0.3166539666412666,
      "grad_norm": 0.373098224401474,
      "learning_rate": 2e-05,
      "loss": 0.6089,
      "step": 3740
    },
    {
      "epoch": 0.31750063500127,
      "grad_norm": 0.4146307408809662,
      "learning_rate": 2e-05,
      "loss": 0.6299,
      "step": 3750
    },
    {
      "epoch": 0.31834730336127337,
      "grad_norm": 0.38848984241485596,
      "learning_rate": 2e-05,
      "loss": 0.6499,
      "step": 3760
    },
    {
      "epoch": 0.3191939717212768,
      "grad_norm": 0.5025537610054016,
      "learning_rate": 2e-05,
      "loss": 0.6271,
      "step": 3770
    },
    {
      "epoch": 0.32004064008128014,
      "grad_norm": 0.3962385058403015,
      "learning_rate": 2e-05,
      "loss": 0.6237,
      "step": 3780
    },
    {
      "epoch": 0.32088730844128355,
      "grad_norm": 1.6572647094726562,
      "learning_rate": 2e-05,
      "loss": 0.6302,
      "step": 3790
    },
    {
      "epoch": 0.32173397680128696,
      "grad_norm": 0.3944777548313141,
      "learning_rate": 2e-05,
      "loss": 0.6497,
      "step": 3800
    },
    {
      "epoch": 0.3225806451612903,
      "grad_norm": 0.3777221739292145,
      "learning_rate": 2e-05,
      "loss": 0.6308,
      "step": 3810
    },
    {
      "epoch": 0.3234273135212937,
      "grad_norm": 0.610248863697052,
      "learning_rate": 2e-05,
      "loss": 0.6372,
      "step": 3820
    },
    {
      "epoch": 0.3242739818812971,
      "grad_norm": 0.47404029965400696,
      "learning_rate": 2e-05,
      "loss": 0.632,
      "step": 3830
    },
    {
      "epoch": 0.3251206502413005,
      "grad_norm": 0.40445438027381897,
      "learning_rate": 2e-05,
      "loss": 0.6244,
      "step": 3840
    },
    {
      "epoch": 0.32596731860130385,
      "grad_norm": 0.4113180339336395,
      "learning_rate": 2e-05,
      "loss": 0.6634,
      "step": 3850
    },
    {
      "epoch": 0.32681398696130726,
      "grad_norm": 0.44300517439842224,
      "learning_rate": 2e-05,
      "loss": 0.6338,
      "step": 3860
    },
    {
      "epoch": 0.32766065532131067,
      "grad_norm": 0.4578755497932434,
      "learning_rate": 2e-05,
      "loss": 0.6167,
      "step": 3870
    },
    {
      "epoch": 0.328507323681314,
      "grad_norm": 0.4159300625324249,
      "learning_rate": 2e-05,
      "loss": 0.6118,
      "step": 3880
    },
    {
      "epoch": 0.32935399204131743,
      "grad_norm": 0.3916099965572357,
      "learning_rate": 2e-05,
      "loss": 0.6112,
      "step": 3890
    },
    {
      "epoch": 0.3302006604013208,
      "grad_norm": 0.37913331389427185,
      "learning_rate": 2e-05,
      "loss": 0.637,
      "step": 3900
    },
    {
      "epoch": 0.3310473287613242,
      "grad_norm": 0.46422481536865234,
      "learning_rate": 2e-05,
      "loss": 0.6487,
      "step": 3910
    },
    {
      "epoch": 0.33189399712132756,
      "grad_norm": 0.39226099848747253,
      "learning_rate": 2e-05,
      "loss": 0.6097,
      "step": 3920
    },
    {
      "epoch": 0.33274066548133097,
      "grad_norm": 1.2026352882385254,
      "learning_rate": 2e-05,
      "loss": 0.6022,
      "step": 3930
    },
    {
      "epoch": 0.3335873338413344,
      "grad_norm": 0.3941093385219574,
      "learning_rate": 2e-05,
      "loss": 0.6158,
      "step": 3940
    },
    {
      "epoch": 0.33443400220133773,
      "grad_norm": 0.39497110247612,
      "learning_rate": 2e-05,
      "loss": 0.6191,
      "step": 3950
    },
    {
      "epoch": 0.33528067056134114,
      "grad_norm": 0.3979906439781189,
      "learning_rate": 2e-05,
      "loss": 0.6527,
      "step": 3960
    },
    {
      "epoch": 0.3361273389213445,
      "grad_norm": 0.41514596343040466,
      "learning_rate": 2e-05,
      "loss": 0.6497,
      "step": 3970
    },
    {
      "epoch": 0.3369740072813479,
      "grad_norm": 0.3805382251739502,
      "learning_rate": 2e-05,
      "loss": 0.6523,
      "step": 3980
    },
    {
      "epoch": 0.33782067564135126,
      "grad_norm": 0.38798990845680237,
      "learning_rate": 2e-05,
      "loss": 0.6435,
      "step": 3990
    },
    {
      "epoch": 0.3386673440013547,
      "grad_norm": 0.3779199421405792,
      "learning_rate": 2e-05,
      "loss": 0.6047,
      "step": 4000
    },
    {
      "epoch": 0.33951401236135803,
      "grad_norm": 0.43265098333358765,
      "learning_rate": 2e-05,
      "loss": 0.6488,
      "step": 4010
    },
    {
      "epoch": 0.34036068072136144,
      "grad_norm": 0.4250130355358124,
      "learning_rate": 2e-05,
      "loss": 0.6166,
      "step": 4020
    },
    {
      "epoch": 0.34120734908136485,
      "grad_norm": 0.3824594020843506,
      "learning_rate": 2e-05,
      "loss": 0.6584,
      "step": 4030
    },
    {
      "epoch": 0.3420540174413682,
      "grad_norm": 0.3953022360801697,
      "learning_rate": 2e-05,
      "loss": 0.636,
      "step": 4040
    },
    {
      "epoch": 0.3429006858013716,
      "grad_norm": 0.376534640789032,
      "learning_rate": 2e-05,
      "loss": 0.625,
      "step": 4050
    },
    {
      "epoch": 0.343747354161375,
      "grad_norm": 0.4311833381652832,
      "learning_rate": 2e-05,
      "loss": 0.6194,
      "step": 4060
    },
    {
      "epoch": 0.3445940225213784,
      "grad_norm": 0.3738897144794464,
      "learning_rate": 2e-05,
      "loss": 0.6239,
      "step": 4070
    },
    {
      "epoch": 0.34544069088138174,
      "grad_norm": 0.38045012950897217,
      "learning_rate": 2e-05,
      "loss": 0.6083,
      "step": 4080
    },
    {
      "epoch": 0.34628735924138515,
      "grad_norm": 0.3941081762313843,
      "learning_rate": 2e-05,
      "loss": 0.6256,
      "step": 4090
    },
    {
      "epoch": 0.34713402760138856,
      "grad_norm": 0.39207085967063904,
      "learning_rate": 2e-05,
      "loss": 0.6153,
      "step": 4100
    },
    {
      "epoch": 0.3479806959613919,
      "grad_norm": 0.38253340125083923,
      "learning_rate": 2e-05,
      "loss": 0.6183,
      "step": 4110
    },
    {
      "epoch": 0.34882736432139533,
      "grad_norm": 0.42153647541999817,
      "learning_rate": 2e-05,
      "loss": 0.6389,
      "step": 4120
    },
    {
      "epoch": 0.3496740326813987,
      "grad_norm": 0.37590694427490234,
      "learning_rate": 2e-05,
      "loss": 0.632,
      "step": 4130
    },
    {
      "epoch": 0.3505207010414021,
      "grad_norm": 0.3831784427165985,
      "learning_rate": 2e-05,
      "loss": 0.6161,
      "step": 4140
    },
    {
      "epoch": 0.35136736940140545,
      "grad_norm": 0.4095568358898163,
      "learning_rate": 2e-05,
      "loss": 0.6354,
      "step": 4150
    },
    {
      "epoch": 0.35221403776140886,
      "grad_norm": 0.3940264582633972,
      "learning_rate": 2e-05,
      "loss": 0.6281,
      "step": 4160
    },
    {
      "epoch": 0.3530607061214122,
      "grad_norm": 0.39979925751686096,
      "learning_rate": 2e-05,
      "loss": 0.6322,
      "step": 4170
    },
    {
      "epoch": 0.3539073744814156,
      "grad_norm": 0.4350389540195465,
      "learning_rate": 2e-05,
      "loss": 0.6244,
      "step": 4180
    },
    {
      "epoch": 0.35475404284141904,
      "grad_norm": 0.37078574299812317,
      "learning_rate": 2e-05,
      "loss": 0.6163,
      "step": 4190
    },
    {
      "epoch": 0.3556007112014224,
      "grad_norm": 0.39877039194107056,
      "learning_rate": 2e-05,
      "loss": 0.6191,
      "step": 4200
    },
    {
      "epoch": 0.3564473795614258,
      "grad_norm": 0.4154704511165619,
      "learning_rate": 2e-05,
      "loss": 0.6212,
      "step": 4210
    },
    {
      "epoch": 0.35729404792142916,
      "grad_norm": 0.42669203877449036,
      "learning_rate": 2e-05,
      "loss": 0.6532,
      "step": 4220
    },
    {
      "epoch": 0.35814071628143257,
      "grad_norm": 0.4079524874687195,
      "learning_rate": 2e-05,
      "loss": 0.6423,
      "step": 4230
    },
    {
      "epoch": 0.3589873846414359,
      "grad_norm": 0.3937144875526428,
      "learning_rate": 2e-05,
      "loss": 0.6236,
      "step": 4240
    },
    {
      "epoch": 0.35983405300143934,
      "grad_norm": 0.4334471523761749,
      "learning_rate": 2e-05,
      "loss": 0.6285,
      "step": 4250
    },
    {
      "epoch": 0.36068072136144275,
      "grad_norm": 0.3907918334007263,
      "learning_rate": 2e-05,
      "loss": 0.6254,
      "step": 4260
    },
    {
      "epoch": 0.3615273897214461,
      "grad_norm": 1.0858243703842163,
      "learning_rate": 2e-05,
      "loss": 0.6437,
      "step": 4270
    },
    {
      "epoch": 0.3623740580814495,
      "grad_norm": 0.41078153252601624,
      "learning_rate": 2e-05,
      "loss": 0.6251,
      "step": 4280
    },
    {
      "epoch": 0.36322072644145287,
      "grad_norm": 0.39965513348579407,
      "learning_rate": 2e-05,
      "loss": 0.6131,
      "step": 4290
    },
    {
      "epoch": 0.3640673948014563,
      "grad_norm": 0.42311328649520874,
      "learning_rate": 2e-05,
      "loss": 0.5952,
      "step": 4300
    },
    {
      "epoch": 0.36491406316145963,
      "grad_norm": 0.39749035239219666,
      "learning_rate": 2e-05,
      "loss": 0.6952,
      "step": 4310
    },
    {
      "epoch": 0.36576073152146305,
      "grad_norm": 0.4066619575023651,
      "learning_rate": 2e-05,
      "loss": 0.6195,
      "step": 4320
    },
    {
      "epoch": 0.36660739988146646,
      "grad_norm": 0.40869539976119995,
      "learning_rate": 2e-05,
      "loss": 0.6551,
      "step": 4330
    },
    {
      "epoch": 0.3674540682414698,
      "grad_norm": 0.38594356179237366,
      "learning_rate": 2e-05,
      "loss": 0.6176,
      "step": 4340
    },
    {
      "epoch": 0.3683007366014732,
      "grad_norm": 0.3964534103870392,
      "learning_rate": 2e-05,
      "loss": 0.6108,
      "step": 4350
    },
    {
      "epoch": 0.3691474049614766,
      "grad_norm": 0.3941608965396881,
      "learning_rate": 2e-05,
      "loss": 0.6312,
      "step": 4360
    },
    {
      "epoch": 0.36999407332148,
      "grad_norm": 0.38923901319503784,
      "learning_rate": 2e-05,
      "loss": 0.6089,
      "step": 4370
    },
    {
      "epoch": 0.37084074168148334,
      "grad_norm": 0.4024799168109894,
      "learning_rate": 2e-05,
      "loss": 0.6444,
      "step": 4380
    },
    {
      "epoch": 0.37168741004148675,
      "grad_norm": 0.42518913745880127,
      "learning_rate": 2e-05,
      "loss": 0.6339,
      "step": 4390
    },
    {
      "epoch": 0.3725340784014901,
      "grad_norm": 0.40740522742271423,
      "learning_rate": 2e-05,
      "loss": 0.6365,
      "step": 4400
    },
    {
      "epoch": 0.3733807467614935,
      "grad_norm": 0.41317957639694214,
      "learning_rate": 2e-05,
      "loss": 0.6253,
      "step": 4410
    },
    {
      "epoch": 0.37422741512149693,
      "grad_norm": 0.3998070955276489,
      "learning_rate": 2e-05,
      "loss": 0.6328,
      "step": 4420
    },
    {
      "epoch": 0.3750740834815003,
      "grad_norm": 0.3857688903808594,
      "learning_rate": 2e-05,
      "loss": 0.6482,
      "step": 4430
    },
    {
      "epoch": 0.3759207518415037,
      "grad_norm": 0.38020142912864685,
      "learning_rate": 2e-05,
      "loss": 0.6165,
      "step": 4440
    },
    {
      "epoch": 0.37676742020150705,
      "grad_norm": 0.5042953491210938,
      "learning_rate": 2e-05,
      "loss": 0.6215,
      "step": 4450
    },
    {
      "epoch": 0.37761408856151046,
      "grad_norm": 0.386430948972702,
      "learning_rate": 2e-05,
      "loss": 0.6155,
      "step": 4460
    },
    {
      "epoch": 0.3784607569215138,
      "grad_norm": 0.40141937136650085,
      "learning_rate": 2e-05,
      "loss": 0.6164,
      "step": 4470
    },
    {
      "epoch": 0.37930742528151723,
      "grad_norm": 0.3964422941207886,
      "learning_rate": 2e-05,
      "loss": 0.62,
      "step": 4480
    },
    {
      "epoch": 0.38015409364152064,
      "grad_norm": 0.4356025457382202,
      "learning_rate": 2e-05,
      "loss": 0.628,
      "step": 4490
    },
    {
      "epoch": 0.381000762001524,
      "grad_norm": 0.4093104302883148,
      "learning_rate": 2e-05,
      "loss": 0.6206,
      "step": 4500
    },
    {
      "epoch": 0.3818474303615274,
      "grad_norm": 0.3951672911643982,
      "learning_rate": 2e-05,
      "loss": 0.6189,
      "step": 4510
    },
    {
      "epoch": 0.38269409872153076,
      "grad_norm": 0.4975200891494751,
      "learning_rate": 2e-05,
      "loss": 0.6353,
      "step": 4520
    },
    {
      "epoch": 0.3835407670815342,
      "grad_norm": 0.3646867871284485,
      "learning_rate": 2e-05,
      "loss": 0.623,
      "step": 4530
    },
    {
      "epoch": 0.38438743544153753,
      "grad_norm": 0.41711995005607605,
      "learning_rate": 2e-05,
      "loss": 0.6269,
      "step": 4540
    },
    {
      "epoch": 0.38523410380154094,
      "grad_norm": 0.4198454022407532,
      "learning_rate": 2e-05,
      "loss": 0.6523,
      "step": 4550
    },
    {
      "epoch": 0.38608077216154435,
      "grad_norm": 0.4093538820743561,
      "learning_rate": 2e-05,
      "loss": 0.6195,
      "step": 4560
    },
    {
      "epoch": 0.3869274405215477,
      "grad_norm": 0.40226060152053833,
      "learning_rate": 2e-05,
      "loss": 0.6192,
      "step": 4570
    },
    {
      "epoch": 0.3877741088815511,
      "grad_norm": 0.4073827862739563,
      "learning_rate": 2e-05,
      "loss": 0.6323,
      "step": 4580
    },
    {
      "epoch": 0.38862077724155447,
      "grad_norm": 0.3588709533214569,
      "learning_rate": 2e-05,
      "loss": 0.6223,
      "step": 4590
    },
    {
      "epoch": 0.3894674456015579,
      "grad_norm": 0.4277055859565735,
      "learning_rate": 2e-05,
      "loss": 0.6366,
      "step": 4600
    },
    {
      "epoch": 0.39031411396156124,
      "grad_norm": 0.3982546031475067,
      "learning_rate": 2e-05,
      "loss": 0.6184,
      "step": 4610
    },
    {
      "epoch": 0.39116078232156465,
      "grad_norm": 0.4190221428871155,
      "learning_rate": 2e-05,
      "loss": 0.5989,
      "step": 4620
    },
    {
      "epoch": 0.392007450681568,
      "grad_norm": 0.4254496395587921,
      "learning_rate": 2e-05,
      "loss": 0.6337,
      "step": 4630
    },
    {
      "epoch": 0.3928541190415714,
      "grad_norm": 0.4280596375465393,
      "learning_rate": 2e-05,
      "loss": 0.6368,
      "step": 4640
    },
    {
      "epoch": 0.3937007874015748,
      "grad_norm": 0.3961775302886963,
      "learning_rate": 2e-05,
      "loss": 0.6065,
      "step": 4650
    },
    {
      "epoch": 0.3945474557615782,
      "grad_norm": 0.4123854339122772,
      "learning_rate": 2e-05,
      "loss": 0.6153,
      "step": 4660
    },
    {
      "epoch": 0.3953941241215816,
      "grad_norm": 0.48063772916793823,
      "learning_rate": 2e-05,
      "loss": 0.6202,
      "step": 4670
    },
    {
      "epoch": 0.39624079248158495,
      "grad_norm": 0.40130090713500977,
      "learning_rate": 2e-05,
      "loss": 0.6355,
      "step": 4680
    },
    {
      "epoch": 0.39708746084158836,
      "grad_norm": 0.4136877655982971,
      "learning_rate": 2e-05,
      "loss": 0.6247,
      "step": 4690
    },
    {
      "epoch": 0.3979341292015917,
      "grad_norm": 0.36779603362083435,
      "learning_rate": 2e-05,
      "loss": 0.6238,
      "step": 4700
    },
    {
      "epoch": 0.3987807975615951,
      "grad_norm": 0.40029042959213257,
      "learning_rate": 2e-05,
      "loss": 0.6176,
      "step": 4710
    },
    {
      "epoch": 0.39962746592159853,
      "grad_norm": 0.43001070618629456,
      "learning_rate": 2e-05,
      "loss": 0.6214,
      "step": 4720
    },
    {
      "epoch": 0.4004741342816019,
      "grad_norm": 0.3810120224952698,
      "learning_rate": 2e-05,
      "loss": 0.6356,
      "step": 4730
    },
    {
      "epoch": 0.4013208026416053,
      "grad_norm": 0.39227691292762756,
      "learning_rate": 2e-05,
      "loss": 0.6086,
      "step": 4740
    },
    {
      "epoch": 0.40216747100160866,
      "grad_norm": 0.3904576003551483,
      "learning_rate": 2e-05,
      "loss": 0.6474,
      "step": 4750
    },
    {
      "epoch": 0.40301413936161207,
      "grad_norm": 0.391421377658844,
      "learning_rate": 2e-05,
      "loss": 0.6125,
      "step": 4760
    },
    {
      "epoch": 0.4038608077216154,
      "grad_norm": 0.390818327665329,
      "learning_rate": 2e-05,
      "loss": 0.6332,
      "step": 4770
    },
    {
      "epoch": 0.40470747608161883,
      "grad_norm": 0.3884637951850891,
      "learning_rate": 2e-05,
      "loss": 0.6234,
      "step": 4780
    },
    {
      "epoch": 0.40555414444162224,
      "grad_norm": 0.39960721135139465,
      "learning_rate": 2e-05,
      "loss": 0.6069,
      "step": 4790
    },
    {
      "epoch": 0.4064008128016256,
      "grad_norm": 0.38708916306495667,
      "learning_rate": 2e-05,
      "loss": 0.6039,
      "step": 4800
    },
    {
      "epoch": 0.407247481161629,
      "grad_norm": 0.37994828820228577,
      "learning_rate": 2e-05,
      "loss": 0.6556,
      "step": 4810
    },
    {
      "epoch": 0.40809414952163237,
      "grad_norm": 0.7117851972579956,
      "learning_rate": 2e-05,
      "loss": 0.652,
      "step": 4820
    },
    {
      "epoch": 0.4089408178816358,
      "grad_norm": 0.3768075108528137,
      "learning_rate": 2e-05,
      "loss": 0.6307,
      "step": 4830
    },
    {
      "epoch": 0.40978748624163913,
      "grad_norm": 0.3990342915058136,
      "learning_rate": 2e-05,
      "loss": 0.606,
      "step": 4840
    },
    {
      "epoch": 0.41063415460164254,
      "grad_norm": 0.35445788502693176,
      "learning_rate": 2e-05,
      "loss": 0.6162,
      "step": 4850
    },
    {
      "epoch": 0.4114808229616459,
      "grad_norm": 0.4304513931274414,
      "learning_rate": 2e-05,
      "loss": 0.6532,
      "step": 4860
    },
    {
      "epoch": 0.4123274913216493,
      "grad_norm": 0.38317644596099854,
      "learning_rate": 2e-05,
      "loss": 0.6274,
      "step": 4870
    },
    {
      "epoch": 0.4131741596816527,
      "grad_norm": 0.3761354684829712,
      "learning_rate": 2e-05,
      "loss": 0.6197,
      "step": 4880
    },
    {
      "epoch": 0.4140208280416561,
      "grad_norm": 0.3854663074016571,
      "learning_rate": 2e-05,
      "loss": 0.6159,
      "step": 4890
    },
    {
      "epoch": 0.4148674964016595,
      "grad_norm": 0.40026572346687317,
      "learning_rate": 2e-05,
      "loss": 0.6318,
      "step": 4900
    },
    {
      "epoch": 0.41571416476166284,
      "grad_norm": 0.3976708948612213,
      "learning_rate": 2e-05,
      "loss": 0.6221,
      "step": 4910
    },
    {
      "epoch": 0.41656083312166625,
      "grad_norm": 0.40661197900772095,
      "learning_rate": 2e-05,
      "loss": 0.6197,
      "step": 4920
    },
    {
      "epoch": 0.4174075014816696,
      "grad_norm": 0.4020674526691437,
      "learning_rate": 2e-05,
      "loss": 0.6273,
      "step": 4930
    },
    {
      "epoch": 0.418254169841673,
      "grad_norm": 0.3835592269897461,
      "learning_rate": 2e-05,
      "loss": 0.6139,
      "step": 4940
    },
    {
      "epoch": 0.41910083820167643,
      "grad_norm": 0.7613312005996704,
      "learning_rate": 2e-05,
      "loss": 0.6188,
      "step": 4950
    },
    {
      "epoch": 0.4199475065616798,
      "grad_norm": 0.35623157024383545,
      "learning_rate": 2e-05,
      "loss": 0.6257,
      "step": 4960
    },
    {
      "epoch": 0.4207941749216832,
      "grad_norm": 0.4170803725719452,
      "learning_rate": 2e-05,
      "loss": 0.6251,
      "step": 4970
    },
    {
      "epoch": 0.42164084328168655,
      "grad_norm": 0.4270610213279724,
      "learning_rate": 2e-05,
      "loss": 0.6199,
      "step": 4980
    },
    {
      "epoch": 0.42248751164168996,
      "grad_norm": 0.37874191999435425,
      "learning_rate": 2e-05,
      "loss": 0.6253,
      "step": 4990
    },
    {
      "epoch": 0.4233341800016933,
      "grad_norm": 0.38245901465415955,
      "learning_rate": 2e-05,
      "loss": 0.623,
      "step": 5000
    },
    {
      "epoch": 0.4241808483616967,
      "grad_norm": 0.3769111931324005,
      "learning_rate": 2e-05,
      "loss": 0.6144,
      "step": 5010
    },
    {
      "epoch": 0.4250275167217001,
      "grad_norm": 0.3969614803791046,
      "learning_rate": 2e-05,
      "loss": 0.6428,
      "step": 5020
    },
    {
      "epoch": 0.4258741850817035,
      "grad_norm": 0.3712959885597229,
      "learning_rate": 2e-05,
      "loss": 0.6126,
      "step": 5030
    },
    {
      "epoch": 0.4267208534417069,
      "grad_norm": 0.39354589581489563,
      "learning_rate": 2e-05,
      "loss": 0.6328,
      "step": 5040
    },
    {
      "epoch": 0.42756752180171026,
      "grad_norm": 0.3992328345775604,
      "learning_rate": 2e-05,
      "loss": 0.6156,
      "step": 5050
    },
    {
      "epoch": 0.42841419016171367,
      "grad_norm": 0.3825113773345947,
      "learning_rate": 2e-05,
      "loss": 0.6107,
      "step": 5060
    },
    {
      "epoch": 0.429260858521717,
      "grad_norm": 0.4251422882080078,
      "learning_rate": 2e-05,
      "loss": 0.6335,
      "step": 5070
    },
    {
      "epoch": 0.43010752688172044,
      "grad_norm": 0.38227295875549316,
      "learning_rate": 2e-05,
      "loss": 0.6166,
      "step": 5080
    },
    {
      "epoch": 0.4309541952417238,
      "grad_norm": 0.4403885006904602,
      "learning_rate": 1.9999810479633703e-05,
      "loss": 0.6371,
      "step": 5090
    },
    {
      "epoch": 0.4318008636017272,
      "grad_norm": 0.39103963971138,
      "learning_rate": 1.997707677020786e-05,
      "loss": 0.6088,
      "step": 5100
    },
    {
      "epoch": 0.4326475319617306,
      "grad_norm": 0.4424373209476471,
      "learning_rate": 1.9916538196124637e-05,
      "loss": 0.6404,
      "step": 5110
    },
    {
      "epoch": 0.43349420032173397,
      "grad_norm": 0.36264264583587646,
      "learning_rate": 1.9818425303863402e-05,
      "loss": 0.643,
      "step": 5120
    },
    {
      "epoch": 0.4343408686817374,
      "grad_norm": 0.4578518867492676,
      "learning_rate": 1.9683111732582887e-05,
      "loss": 0.6011,
      "step": 5130
    },
    {
      "epoch": 0.43518753704174074,
      "grad_norm": 0.42311257123947144,
      "learning_rate": 1.951111279120706e-05,
      "loss": 0.6294,
      "step": 5140
    },
    {
      "epoch": 0.43603420540174415,
      "grad_norm": 0.40115925669670105,
      "learning_rate": 1.9303083495996028e-05,
      "loss": 0.628,
      "step": 5150
    },
    {
      "epoch": 0.4368808737617475,
      "grad_norm": 0.520965576171875,
      "learning_rate": 1.9059816076075482e-05,
      "loss": 0.6122,
      "step": 5160
    },
    {
      "epoch": 0.4377275421217509,
      "grad_norm": 0.38975247740745544,
      "learning_rate": 1.8782236956424142e-05,
      "loss": 0.6216,
      "step": 5170
    },
    {
      "epoch": 0.4385742104817543,
      "grad_norm": 0.37491023540496826,
      "learning_rate": 1.8471403229808897e-05,
      "loss": 0.6359,
      "step": 5180
    },
    {
      "epoch": 0.4394208788417577,
      "grad_norm": 0.38760411739349365,
      "learning_rate": 1.8128498631103333e-05,
      "loss": 0.6159,
      "step": 5190
    },
    {
      "epoch": 0.4402675472017611,
      "grad_norm": 0.4243720769882202,
      "learning_rate": 1.7754829029320507e-05,
      "loss": 0.6013,
      "step": 5200
    },
    {
      "epoch": 0.44111421556176444,
      "grad_norm": 0.7784062027931213,
      "learning_rate": 1.7351817454527473e-05,
      "loss": 0.6117,
      "step": 5210
    },
    {
      "epoch": 0.44196088392176786,
      "grad_norm": 0.38836973905563354,
      "learning_rate": 1.6920998678580345e-05,
      "loss": 0.6208,
      "step": 5220
    },
    {
      "epoch": 0.4428075522817712,
      "grad_norm": 0.3686048686504364,
      "learning_rate": 1.6464013370317854e-05,
      "loss": 0.6241,
      "step": 5230
    },
    {
      "epoch": 0.4436542206417746,
      "grad_norm": 0.3576488792896271,
      "learning_rate": 1.5982601847471915e-05,
      "loss": 0.607,
      "step": 5240
    },
    {
      "epoch": 0.444500889001778,
      "grad_norm": 0.38535255193710327,
      "learning_rate": 1.5478597449089553e-05,
      "loss": 0.5925,
      "step": 5250
    },
    {
      "epoch": 0.4453475573617814,
      "grad_norm": 0.3931323289871216,
      "learning_rate": 1.4953919553705725e-05,
      "loss": 0.609,
      "step": 5260
    },
    {
      "epoch": 0.4461942257217848,
      "grad_norm": 0.37106412649154663,
      "learning_rate": 1.441056626985559e-05,
      "loss": 0.6084,
      "step": 5270
    },
    {
      "epoch": 0.44704089408178815,
      "grad_norm": 0.3821442425251007,
      "learning_rate": 1.3850606826762668e-05,
      "loss": 0.6257,
      "step": 5280
    },
    {
      "epoch": 0.44788756244179156,
      "grad_norm": 0.3764842450618744,
      "learning_rate": 1.3276173694181128e-05,
      "loss": 0.617,
      "step": 5290
    },
    {
      "epoch": 0.4487342308017949,
      "grad_norm": 0.3602072298526764,
      "learning_rate": 1.2689454461401822e-05,
      "loss": 0.6296,
      "step": 5300
    },
    {
      "epoch": 0.44958089916179833,
      "grad_norm": 0.37906748056411743,
      "learning_rate": 1.2092683506348939e-05,
      "loss": 0.6064,
      "step": 5310
    },
    {
      "epoch": 0.4504275675218017,
      "grad_norm": 0.409184992313385,
      "learning_rate": 1.14881334864935e-05,
      "loss": 0.6201,
      "step": 5320
    },
    {
      "epoch": 0.4512742358818051,
      "grad_norm": 0.4005579352378845,
      "learning_rate": 1.0878106683988481e-05,
      "loss": 0.6383,
      "step": 5330
    },
    {
      "epoch": 0.4521209042418085,
      "grad_norm": 0.36245763301849365,
      "learning_rate": 1.026492623798551e-05,
      "loss": 0.5993,
      "step": 5340
    },
    {
      "epoch": 0.45296757260181186,
      "grad_norm": 0.3887219727039337,
      "learning_rate": 9.650927297522743e-06,
      "loss": 0.6281,
      "step": 5350
    },
    {
      "epoch": 0.4538142409618153,
      "grad_norm": 0.40078380703926086,
      "learning_rate": 9.038448128676012e-06,
      "loss": 0.6299,
      "step": 5360
    },
    {
      "epoch": 0.45466090932181863,
      "grad_norm": 0.34163933992385864,
      "learning_rate": 8.42982120983949e-06,
      "loss": 0.6123,
      "step": 5370
    },
    {
      "epoch": 0.45550757768182204,
      "grad_norm": 0.39262932538986206,
      "learning_rate": 7.827364349047303e-06,
      "loss": 0.6195,
      "step": 5380
    },
    {
      "epoch": 0.4563542460418254,
      "grad_norm": 0.37275251746177673,
      "learning_rate": 7.233371857163706e-06,
      "loss": 0.5969,
      "step": 5390
    },
    {
      "epoch": 0.4572009144018288,
      "grad_norm": 0.39115020632743835,
      "learning_rate": 6.650105810556492e-06,
      "loss": 0.6035,
      "step": 5400
    },
    {
      "epoch": 0.4580475827618322,
      "grad_norm": 0.3782908618450165,
      "learning_rate": 6.079787436527681e-06,
      "loss": 0.615,
      "step": 5410
    },
    {
      "epoch": 0.4588942511218356,
      "grad_norm": 0.44726574420928955,
      "learning_rate": 5.524588654307961e-06,
      "loss": 0.6086,
      "step": 5420
    },
    {
      "epoch": 0.459740919481839,
      "grad_norm": 0.4180717468261719,
      "learning_rate": 4.986623803828985e-06,
      "loss": 0.6237,
      "step": 5430
    },
    {
      "epoch": 0.46058758784184234,
      "grad_norm": 0.3337180018424988,
      "learning_rate": 4.467941593772423e-06,
      "loss": 0.6167,
      "step": 5440
    },
    {
      "epoch": 0.46143425620184575,
      "grad_norm": 0.4083661139011383,
      "learning_rate": 3.9705172995597606e-06,
      "loss": 0.5866,
      "step": 5450
    },
    {
      "epoch": 0.4622809245618491,
      "grad_norm": 0.38824260234832764,
      "learning_rate": 3.4962452409948346e-06,
      "loss": 0.6163,
      "step": 5460
    },
    {
      "epoch": 0.4631275929218525,
      "grad_norm": 0.37703633308410645,
      "learning_rate": 3.046931568206211e-06,
      "loss": 0.6121,
      "step": 5470
    },
    {
      "epoch": 0.46397426128185587,
      "grad_norm": 0.4685078561306,
      "learning_rate": 2.624287383362404e-06,
      "loss": 0.615,
      "step": 5480
    },
    {
      "epoch": 0.4648209296418593,
      "grad_norm": 0.36976656317710876,
      "learning_rate": 2.2299222243541845e-06,
      "loss": 0.5955,
      "step": 5490
    },
    {
      "epoch": 0.4656675980018627,
      "grad_norm": 0.3782484531402588,
      "learning_rate": 1.8653379352598246e-06,
      "loss": 0.6164,
      "step": 5500
    },
    {
      "epoch": 0.46651426636186605,
      "grad_norm": 0.3932911157608032,
      "learning_rate": 1.5319229469361075e-06,
      "loss": 0.5978,
      "step": 5510
    },
    {
      "epoch": 0.46736093472186946,
      "grad_norm": 0.43107590079307556,
      "learning_rate": 1.2309469895160917e-06,
      "loss": 0.609,
      "step": 5520
    },
    {
      "epoch": 0.4682076030818728,
      "grad_norm": 0.36705833673477173,
      "learning_rate": 9.635562569497974e-07,
      "loss": 0.6194,
      "step": 5530
    },
    {
      "epoch": 0.4690542714418762,
      "grad_norm": 0.38226068019866943,
      "learning_rate": 7.307690420024867e-07,
      "loss": 0.5992,
      "step": 5540
    },
    {
      "epoch": 0.4699009398018796,
      "grad_norm": 0.3879031538963318,
      "learning_rate": 5.334718583335942e-07,
      "loss": 0.614,
      "step": 5550
    },
    {
      "epoch": 0.470747608161883,
      "grad_norm": 0.3758437931537628,
      "learning_rate": 3.7241606442442763e-07,
      "loss": 0.5831,
      "step": 5560
    },
    {
      "epoch": 0.4715942765218864,
      "grad_norm": 0.390807181596756,
      "learning_rate": 2.482150022115879e-07,
      "loss": 0.605,
      "step": 5570
    },
    {
      "epoch": 0.47244094488188976,
      "grad_norm": 0.3561398983001709,
      "learning_rate": 1.6134166132293174e-07,
      "loss": 0.6157,
      "step": 5580
    },
    {
      "epoch": 0.47328761324189317,
      "grad_norm": 0.3927530348300934,
      "learning_rate": 1.1212687781127357e-07,
      "loss": 0.6235,
      "step": 5590
    },
    {
      "epoch": 0.4738802810938955,
      "step": 5597,
      "total_flos": 2.4985159616803373e+19,
      "train_loss": 0.055960970476991216,
      "train_runtime": 5702.2392,
      "train_samples_per_second": 11.779,
      "train_steps_per_second": 0.982
    }
  ],
  "logging_steps": 10,
  "max_steps": 5597,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 636,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 2.4985159616803373e+19,
  "train_batch_size": 1,
  "trial_name": null,
  "trial_params": null
}