{
  "best_metric": 0.7685801386833191,
  "best_model_checkpoint": "./output/checkpoint-4800",
  "epoch": 0.6153654898060666,
  "eval_steps": 150,
  "global_step": 4950,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0012431626056688214,
      "grad_norm": 3.163724184036255,
      "learning_rate": 1.25e-05,
      "loss": 1.9463,
      "step": 10
    },
    {
      "epoch": 0.002486325211337643,
      "grad_norm": 3.8644485473632812,
      "learning_rate": 2.5e-05,
      "loss": 2.054,
      "step": 20
    },
    {
      "epoch": 0.0037294878170064643,
      "grad_norm": 2.5856993198394775,
      "learning_rate": 3.75e-05,
      "loss": 1.6479,
      "step": 30
    },
    {
      "epoch": 0.004972650422675286,
      "grad_norm": 3.1547975540161133,
      "learning_rate": 5e-05,
      "loss": 1.3341,
      "step": 40
    },
    {
      "epoch": 0.006215813028344107,
      "grad_norm": 2.9259073734283447,
      "learning_rate": 6.25e-05,
      "loss": 1.2614,
      "step": 50
    },
    {
      "epoch": 0.007458975634012929,
      "grad_norm": 2.072551727294922,
      "learning_rate": 7.5e-05,
      "loss": 1.2827,
      "step": 60
    },
    {
      "epoch": 0.00870213823968175,
      "grad_norm": 2.1959753036499023,
      "learning_rate": 8.75e-05,
      "loss": 1.1541,
      "step": 70
    },
    {
      "epoch": 0.009945300845350571,
      "grad_norm": 2.231361150741577,
      "learning_rate": 0.0001,
      "loss": 1.1048,
      "step": 80
    },
    {
      "epoch": 0.011188463451019393,
      "grad_norm": 1.8180471658706665,
      "learning_rate": 0.00011250000000000001,
      "loss": 1.1412,
      "step": 90
    },
    {
      "epoch": 0.012431626056688214,
      "grad_norm": 2.6011712551116943,
      "learning_rate": 0.000125,
      "loss": 0.9483,
      "step": 100
    },
    {
      "epoch": 0.013674788662357036,
      "grad_norm": 1.6959333419799805,
      "learning_rate": 0.00012499871543489787,
      "loss": 1.0523,
      "step": 110
    },
    {
      "epoch": 0.014917951268025857,
      "grad_norm": 2.7038402557373047,
      "learning_rate": 0.00012499486179239495,
      "loss": 1.0785,
      "step": 120
    },
    {
      "epoch": 0.01616111387369468,
      "grad_norm": 2.327219009399414,
      "learning_rate": 0.00012498843923089938,
      "loss": 1.028,
      "step": 130
    },
    {
      "epoch": 0.0174042764793635,
      "grad_norm": 2.2055954933166504,
      "learning_rate": 0.0001249794480144175,
      "loss": 0.9735,
      "step": 140
    },
    {
      "epoch": 0.018647439085032323,
      "grad_norm": 1.949844241142273,
      "learning_rate": 0.000124967888512543,
      "loss": 1.0077,
      "step": 150
    },
    {
      "epoch": 0.018647439085032323,
      "eval_loss": 1.273397445678711,
      "eval_runtime": 54.5001,
      "eval_samples_per_second": 9.174,
      "eval_steps_per_second": 9.174,
      "step": 150
    },
    {
      "epoch": 0.019890601690701143,
      "grad_norm": 2.5000596046447754,
      "learning_rate": 0.00012495376120044173,
      "loss": 0.8882,
      "step": 160
    },
    {
      "epoch": 0.021133764296369966,
      "grad_norm": 1.6136529445648193,
      "learning_rate": 0.00012493706665883217,
      "loss": 0.9579,
      "step": 170
    },
    {
      "epoch": 0.022376926902038786,
      "grad_norm": 2.2783761024475098,
      "learning_rate": 0.00012491780557396154,
      "loss": 1.0162,
      "step": 180
    },
    {
      "epoch": 0.02362008950770761,
      "grad_norm": 2.086787700653076,
      "learning_rate": 0.00012489597873757756,
      "loss": 0.9505,
      "step": 190
    },
    {
      "epoch": 0.02486325211337643,
      "grad_norm": 2.1386966705322266,
      "learning_rate": 0.00012487158704689602,
      "loss": 0.8124,
      "step": 200
    },
    {
      "epoch": 0.026106414719045252,
      "grad_norm": 2.209433078765869,
      "learning_rate": 0.0001248446315045638,
      "loss": 0.9005,
      "step": 210
    },
    {
      "epoch": 0.02734957732471407,
      "grad_norm": 1.7136540412902832,
      "learning_rate": 0.00012481511321861763,
      "loss": 0.8635,
      "step": 220
    },
    {
      "epoch": 0.028592739930382895,
      "grad_norm": 1.6844110488891602,
      "learning_rate": 0.00012478303340243864,
      "loss": 0.9517,
      "step": 230
    },
    {
      "epoch": 0.029835902536051714,
      "grad_norm": 1.9923568964004517,
      "learning_rate": 0.00012474839337470246,
      "loss": 0.8701,
      "step": 240
    },
    {
      "epoch": 0.031079065141720538,
      "grad_norm": 1.8591444492340088,
      "learning_rate": 0.0001247111945593249,
      "loss": 0.9421,
      "step": 250
    },
    {
      "epoch": 0.03232222774738936,
      "grad_norm": 1.571029543876648,
      "learning_rate": 0.00012467143848540359,
      "loss": 0.9207,
      "step": 260
    },
    {
      "epoch": 0.03356539035305818,
      "grad_norm": 2.1113922595977783,
      "learning_rate": 0.000124629126787155,
      "loss": 0.7869,
      "step": 270
    },
    {
      "epoch": 0.034808552958727,
      "grad_norm": 2.252993106842041,
      "learning_rate": 0.00012458426120384738,
      "loss": 0.9157,
      "step": 280
    },
    {
      "epoch": 0.03605171556439582,
      "grad_norm": 1.6023743152618408,
      "learning_rate": 0.00012453684357972906,
      "loss": 0.8904,
      "step": 290
    },
    {
      "epoch": 0.037294878170064646,
      "grad_norm": 1.2541028261184692,
      "learning_rate": 0.00012448687586395289,
      "loss": 0.855,
      "step": 300
    },
    {
      "epoch": 0.037294878170064646,
      "eval_loss": 1.1304936408996582,
      "eval_runtime": 54.356,
      "eval_samples_per_second": 9.199,
      "eval_steps_per_second": 9.199,
      "step": 300
    },
    {
      "epoch": 0.038538040775733466,
      "grad_norm": 2.082008123397827,
      "learning_rate": 0.00012443436011049593,
      "loss": 1.0134,
      "step": 310
    },
    {
      "epoch": 0.039781203381402286,
      "grad_norm": 1.3970550298690796,
      "learning_rate": 0.0001243792984780751,
      "loss": 0.8233,
      "step": 320
    },
    {
      "epoch": 0.041024365987071106,
      "grad_norm": 1.6384402513504028,
      "learning_rate": 0.00012432169323005853,
      "loss": 0.7816,
      "step": 330
    },
    {
      "epoch": 0.04226752859273993,
      "grad_norm": 2.04970645904541,
      "learning_rate": 0.00012426154673437223,
      "loss": 0.8563,
      "step": 340
    },
    {
      "epoch": 0.04351069119840875,
      "grad_norm": 1.8425785303115845,
      "learning_rate": 0.00012419886146340314,
      "loss": 0.8697,
      "step": 350
    },
    {
      "epoch": 0.04475385380407757,
      "grad_norm": 1.3543235063552856,
      "learning_rate": 0.0001241336399938972,
      "loss": 0.8055,
      "step": 360
    },
    {
      "epoch": 0.0459970164097464,
      "grad_norm": 1.8363412618637085,
      "learning_rate": 0.00012406588500685355,
      "loss": 0.7834,
      "step": 370
    },
    {
      "epoch": 0.04724017901541522,
      "grad_norm": 1.7371594905853271,
      "learning_rate": 0.00012399559928741435,
      "loss": 0.8563,
      "step": 380
    },
    {
      "epoch": 0.04848334162108404,
      "grad_norm": 1.7979482412338257,
      "learning_rate": 0.00012392278572475023,
      "loss": 0.9205,
      "step": 390
    },
    {
      "epoch": 0.04972650422675286,
      "grad_norm": 1.9040799140930176,
      "learning_rate": 0.0001238474473119416,
      "loss": 0.7881,
      "step": 400
    },
    {
      "epoch": 0.050969666832421684,
      "grad_norm": 1.9962204694747925,
      "learning_rate": 0.00012376958714585545,
      "loss": 0.8942,
      "step": 410
    },
    {
      "epoch": 0.052212829438090504,
      "grad_norm": 1.4779837131500244,
      "learning_rate": 0.0001236892084270183,
      "loss": 0.8128,
      "step": 420
    },
    {
      "epoch": 0.05345599204375932,
      "grad_norm": 1.712457299232483,
      "learning_rate": 0.00012360631445948448,
      "loss": 0.8101,
      "step": 430
    },
    {
      "epoch": 0.05469915464942814,
      "grad_norm": 1.750348448753357,
      "learning_rate": 0.00012352090865070026,
      "loss": 0.8743,
      "step": 440
    },
    {
      "epoch": 0.05594231725509697,
      "grad_norm": 2.143127202987671,
      "learning_rate": 0.00012343299451136397,
      "loss": 0.8286,
      "step": 450
    },
    {
      "epoch": 0.05594231725509697,
      "eval_loss": 1.080597162246704,
      "eval_runtime": 56.9902,
      "eval_samples_per_second": 8.773,
      "eval_steps_per_second": 8.773,
      "step": 450
    },
    {
      "epoch": 0.05718547986076579,
      "grad_norm": 1.626365065574646,
      "learning_rate": 0.00012334257565528155,
      "loss": 0.8354,
      "step": 460
    },
    {
      "epoch": 0.05842864246643461,
      "grad_norm": 1.8055260181427002,
      "learning_rate": 0.000123249655799218,
      "loss": 0.801,
      "step": 470
    },
    {
      "epoch": 0.05967180507210343,
      "grad_norm": 1.6423254013061523,
      "learning_rate": 0.00012315423876274468,
      "loss": 0.752,
      "step": 480
    },
    {
      "epoch": 0.060914967677772255,
      "grad_norm": 1.7773467302322388,
      "learning_rate": 0.0001230563284680822,
      "loss": 0.8418,
      "step": 490
    },
    {
      "epoch": 0.062158130283441075,
      "grad_norm": 1.448588490486145,
      "learning_rate": 0.00012295592893993935,
      "loss": 0.8632,
      "step": 500
    },
    {
      "epoch": 0.0634012928891099,
      "grad_norm": 1.3413039445877075,
      "learning_rate": 0.00012285304430534745,
      "loss": 0.7983,
      "step": 510
    },
    {
      "epoch": 0.06464445549477872,
      "grad_norm": 1.3094663619995117,
      "learning_rate": 0.00012274767879349083,
      "loss": 0.7309,
      "step": 520
    },
    {
      "epoch": 0.06588761810044753,
      "grad_norm": 1.427276611328125,
      "learning_rate": 0.00012263983673553306,
      "loss": 0.8253,
      "step": 530
    },
    {
      "epoch": 0.06713078070611636,
      "grad_norm": 1.9115201234817505,
      "learning_rate": 0.0001225295225644387,
      "loss": 0.7826,
      "step": 540
    },
    {
      "epoch": 0.06837394331178519,
      "grad_norm": 1.8499016761779785,
      "learning_rate": 0.0001224167408147913,
      "loss": 0.7889,
      "step": 550
    },
    {
      "epoch": 0.069617105917454,
      "grad_norm": 1.753057837486267,
      "learning_rate": 0.0001223014961226068,
      "loss": 0.8468,
      "step": 560
    },
    {
      "epoch": 0.07086026852312283,
      "grad_norm": 1.4970792531967163,
      "learning_rate": 0.00012218379322514317,
      "loss": 0.8043,
      "step": 570
    },
    {
      "epoch": 0.07210343112879164,
      "grad_norm": 1.355022668838501,
      "learning_rate": 0.00012206363696070545,
      "loss": 0.7553,
      "step": 580
    },
    {
      "epoch": 0.07334659373446047,
      "grad_norm": 1.5447806119918823,
      "learning_rate": 0.0001219410322684471,
      "loss": 0.7947,
      "step": 590
    },
    {
      "epoch": 0.07458975634012929,
      "grad_norm": 1.6613497734069824,
      "learning_rate": 0.0001218159841881668,
      "loss": 0.7285,
      "step": 600
    },
    {
      "epoch": 0.07458975634012929,
      "eval_loss": 1.035021185874939,
      "eval_runtime": 56.3558,
      "eval_samples_per_second": 8.872,
      "eval_steps_per_second": 8.872,
      "step": 600
    },
    {
      "epoch": 0.0758329189457981,
      "grad_norm": 1.778586506843567,
      "learning_rate": 0.00012168849786010133,
      "loss": 0.7333,
      "step": 610
    },
    {
      "epoch": 0.07707608155146693,
      "grad_norm": 1.4905791282653809,
      "learning_rate": 0.00012155857852471433,
      "loss": 0.7691,
      "step": 620
    },
    {
      "epoch": 0.07831924415713576,
      "grad_norm": 1.5982129573822021,
      "learning_rate": 0.0001214262315224808,
      "loss": 0.7232,
      "step": 630
    },
    {
      "epoch": 0.07956240676280457,
      "grad_norm": 2.0250298976898193,
      "learning_rate": 0.00012129146229366766,
      "loss": 0.7468,
      "step": 640
    },
    {
      "epoch": 0.0808055693684734,
      "grad_norm": 1.4550626277923584,
      "learning_rate": 0.00012115427637811003,
      "loss": 0.7719,
      "step": 650
    },
    {
      "epoch": 0.08204873197414221,
      "grad_norm": 1.4702717065811157,
      "learning_rate": 0.00012101467941498357,
      "loss": 0.8111,
      "step": 660
    },
    {
      "epoch": 0.08329189457981104,
      "grad_norm": 1.379180908203125,
      "learning_rate": 0.0001208726771425727,
      "loss": 0.7651,
      "step": 670
    },
    {
      "epoch": 0.08453505718547986,
      "grad_norm": 1.971276044845581,
      "learning_rate": 0.00012072827539803463,
      "loss": 0.8005,
      "step": 680
    },
    {
      "epoch": 0.08577821979114868,
      "grad_norm": 1.7520999908447266,
      "learning_rate": 0.00012058148011715949,
      "loss": 0.6388,
      "step": 690
    },
    {
      "epoch": 0.0870213823968175,
      "grad_norm": 7.61739444732666,
      "learning_rate": 0.00012043229733412636,
      "loss": 0.8455,
      "step": 700
    },
    {
      "epoch": 0.08826454500248633,
      "grad_norm": 1.2994366884231567,
      "learning_rate": 0.0001202807331812551,
      "loss": 0.7365,
      "step": 710
    },
    {
      "epoch": 0.08950770760815514,
      "grad_norm": 1.3976930379867554,
      "learning_rate": 0.00012012679388875441,
      "loss": 0.7674,
      "step": 720
    },
    {
      "epoch": 0.09075087021382397,
      "grad_norm": 1.8187024593353271,
      "learning_rate": 0.00011997048578446568,
      "loss": 0.8233,
      "step": 730
    },
    {
      "epoch": 0.0919940328194928,
      "grad_norm": 1.5645136833190918,
      "learning_rate": 0.00011981181529360282,
      "loss": 0.7395,
      "step": 740
    },
    {
      "epoch": 0.09323719542516161,
      "grad_norm": 2.0647048950195312,
      "learning_rate": 0.00011965078893848828,
      "loss": 0.7704,
      "step": 750
    },
    {
      "epoch": 0.09323719542516161,
      "eval_loss": 0.9952675700187683,
      "eval_runtime": 55.6428,
      "eval_samples_per_second": 8.986,
      "eval_steps_per_second": 8.986,
      "step": 750
    },
    {
      "epoch": 0.09448035803083044,
      "grad_norm": 1.6342475414276123,
      "learning_rate": 0.00011948741333828481,
      "loss": 0.7924,
      "step": 760
    },
    {
      "epoch": 0.09572352063649925,
      "grad_norm": 1.773032307624817,
      "learning_rate": 0.00011932169520872344,
      "loss": 0.7737,
      "step": 770
    },
    {
      "epoch": 0.09696668324216808,
      "grad_norm": 1.3042093515396118,
      "learning_rate": 0.00011915364136182738,
      "loss": 0.7554,
      "step": 780
    },
    {
      "epoch": 0.0982098458478369,
      "grad_norm": 1.9383268356323242,
      "learning_rate": 0.0001189832587056321,
      "loss": 0.793,
      "step": 790
    },
    {
      "epoch": 0.09945300845350571,
      "grad_norm": 0.9794479608535767,
      "learning_rate": 0.00011881055424390119,
      "loss": 0.7059,
      "step": 800
    },
    {
      "epoch": 0.10069617105917454,
      "grad_norm": 1.8000845909118652,
      "learning_rate": 0.00011863553507583869,
      "loss": 0.7037,
      "step": 810
    },
    {
      "epoch": 0.10193933366484337,
      "grad_norm": 1.6130144596099854,
      "learning_rate": 0.00011845820839579708,
      "loss": 0.9073,
      "step": 820
    },
    {
      "epoch": 0.10318249627051218,
      "grad_norm": 1.7219104766845703,
      "learning_rate": 0.00011827858149298162,
      "loss": 0.7089,
      "step": 830
    },
    {
      "epoch": 0.10442565887618101,
      "grad_norm": 1.8444559574127197,
      "learning_rate": 0.00011809666175115075,
      "loss": 0.6576,
      "step": 840
    },
    {
      "epoch": 0.10566882148184982,
      "grad_norm": 2.564314126968384,
      "learning_rate": 0.00011791245664831251,
      "loss": 0.7686,
      "step": 850
    },
    {
      "epoch": 0.10691198408751865,
      "grad_norm": 1.1209467649459839,
      "learning_rate": 0.0001177259737564172,
      "loss": 0.7148,
      "step": 860
    },
    {
      "epoch": 0.10815514669318747,
      "grad_norm": 1.5384615659713745,
      "learning_rate": 0.00011753722074104613,
      "loss": 0.7689,
      "step": 870
    },
    {
      "epoch": 0.10939830929885629,
      "grad_norm": 1.2957892417907715,
      "learning_rate": 0.00011734620536109644,
      "loss": 0.7372,
      "step": 880
    },
    {
      "epoch": 0.11064147190452511,
      "grad_norm": 31.243797302246094,
      "learning_rate": 0.00011715293546846223,
      "loss": 0.725,
      "step": 890
    },
    {
      "epoch": 0.11188463451019394,
      "grad_norm": 1.338306188583374,
      "learning_rate": 0.00011695741900771184,
      "loss": 0.6768,
      "step": 900
    },
    {
      "epoch": 0.11188463451019394,
      "eval_loss": 0.9783537983894348,
      "eval_runtime": 57.0411,
      "eval_samples_per_second": 8.766,
      "eval_steps_per_second": 8.766,
      "step": 900
    },
    {
      "epoch": 0.11312779711586275,
      "grad_norm": 1.0017350912094116,
      "learning_rate": 0.00011675966401576116,
      "loss": 0.71,
      "step": 910
    },
    {
      "epoch": 0.11437095972153158,
      "grad_norm": 1.6244183778762817,
      "learning_rate": 0.00011655967862154335,
      "loss": 0.7,
      "step": 920
    },
    {
      "epoch": 0.11561412232720039,
      "grad_norm": 1.6065330505371094,
      "learning_rate": 0.0001163574710456747,
      "loss": 0.7206,
      "step": 930
    },
    {
      "epoch": 0.11685728493286922,
      "grad_norm": 1.042508602142334,
      "learning_rate": 0.00011615304960011663,
      "loss": 0.6664,
      "step": 940
    },
    {
      "epoch": 0.11810044753853804,
      "grad_norm": 1.2049801349639893,
      "learning_rate": 0.00011594642268783415,
      "loss": 0.7894,
      "step": 950
    },
    {
      "epoch": 0.11934361014420686,
      "grad_norm": 1.6910526752471924,
      "learning_rate": 0.00011573759880245027,
      "loss": 0.7269,
      "step": 960
    },
    {
      "epoch": 0.12058677274987568,
      "grad_norm": 2.0689308643341064,
      "learning_rate": 0.00011552658652789703,
      "loss": 0.7058,
      "step": 970
    },
    {
      "epoch": 0.12182993535554451,
      "grad_norm": 1.326037883758545,
      "learning_rate": 0.00011531339453806258,
      "loss": 0.6103,
      "step": 980
    },
    {
      "epoch": 0.12307309796121332,
      "grad_norm": 1.0418630838394165,
      "learning_rate": 0.00011509803159643458,
      "loss": 0.741,
      "step": 990
    },
    {
      "epoch": 0.12431626056688215,
      "grad_norm": 1.9915788173675537,
      "learning_rate": 0.00011488050655574003,
      "loss": 0.6941,
      "step": 1000
    },
    {
      "epoch": 0.12555942317255098,
      "grad_norm": 1.4139434099197388,
      "learning_rate": 0.00011466082835758141,
      "loss": 0.79,
      "step": 1010
    },
    {
      "epoch": 0.1268025857782198,
      "grad_norm": 2.1607091426849365,
      "learning_rate": 0.000114439006032069,
      "loss": 0.7007,
      "step": 1020
    },
    {
      "epoch": 0.1280457483838886,
      "grad_norm": 1.401469111442566,
      "learning_rate": 0.00011421504869744978,
      "loss": 0.6854,
      "step": 1030
    },
    {
      "epoch": 0.12928891098955744,
      "grad_norm": 1.8763591051101685,
      "learning_rate": 0.0001139889655597326,
      "loss": 0.7929,
      "step": 1040
    },
    {
      "epoch": 0.13053207359522626,
      "grad_norm": 1.3231312036514282,
      "learning_rate": 0.00011376076591230974,
      "loss": 0.834,
      "step": 1050
    },
    {
      "epoch": 0.13053207359522626,
      "eval_loss": 0.9482272267341614,
      "eval_runtime": 54.6775,
      "eval_samples_per_second": 9.145,
      "eval_steps_per_second": 9.145,
      "step": 1050
    },
    {
      "epoch": 0.13177523620089507,
      "grad_norm": 1.7060565948486328,
      "learning_rate": 0.00011353045913557492,
      "loss": 0.737,
      "step": 1060
    },
    {
      "epoch": 0.1330183988065639,
      "grad_norm": 1.3965402841567993,
      "learning_rate": 0.00011329805469653768,
      "loss": 0.7084,
      "step": 1070
    },
    {
      "epoch": 0.13426156141223272,
      "grad_norm": 1.6514191627502441,
      "learning_rate": 0.00011306356214843422,
      "loss": 0.6474,
      "step": 1080
    },
    {
      "epoch": 0.13550472401790153,
      "grad_norm": 1.554608702659607,
      "learning_rate": 0.00011282699113033477,
      "loss": 0.6597,
      "step": 1090
    },
    {
      "epoch": 0.13674788662357037,
      "grad_norm": 1.4712316989898682,
      "learning_rate": 0.00011258835136674729,
      "loss": 0.6732,
      "step": 1100
    },
    {
      "epoch": 0.1379910492292392,
      "grad_norm": 1.4330312013626099,
      "learning_rate": 0.00011234765266721778,
      "loss": 0.7318,
      "step": 1110
    },
    {
      "epoch": 0.139234211834908,
      "grad_norm": 1.700324296951294,
      "learning_rate": 0.00011210490492592703,
      "loss": 0.711,
      "step": 1120
    },
    {
      "epoch": 0.14047737444057684,
      "grad_norm": 1.3977235555648804,
      "learning_rate": 0.0001118601181212839,
      "loss": 0.7372,
      "step": 1130
    },
    {
      "epoch": 0.14172053704624565,
      "grad_norm": 1.6782617568969727,
      "learning_rate": 0.00011161330231551515,
      "loss": 0.6542,
      "step": 1140
    },
    {
      "epoch": 0.14296369965191447,
      "grad_norm": 1.953229546546936,
      "learning_rate": 0.00011136446765425187,
      "loss": 0.7917,
      "step": 1150
    },
    {
      "epoch": 0.14420686225758328,
      "grad_norm": 1.23630690574646,
      "learning_rate": 0.00011111362436611234,
      "loss": 0.7396,
      "step": 1160
    },
    {
      "epoch": 0.14545002486325212,
      "grad_norm": 1.2555770874023438,
      "learning_rate": 0.00011086078276228167,
      "loss": 0.6994,
      "step": 1170
    },
    {
      "epoch": 0.14669318746892093,
      "grad_norm": 1.9462257623672485,
      "learning_rate": 0.00011060595323608789,
      "loss": 0.7422,
      "step": 1180
    },
    {
      "epoch": 0.14793635007458975,
      "grad_norm": 1.3382525444030762,
      "learning_rate": 0.00011034914626257467,
      "loss": 0.6999,
      "step": 1190
    },
    {
      "epoch": 0.14917951268025859,
      "grad_norm": 1.6739270687103271,
      "learning_rate": 0.0001100903723980709,
      "loss": 0.684,
      "step": 1200
    },
    {
      "epoch": 0.14917951268025859,
      "eval_loss": 0.9281342625617981,
      "eval_runtime": 53.7337,
      "eval_samples_per_second": 9.305,
      "eval_steps_per_second": 9.305,
      "step": 1200
    },
    {
      "epoch": 0.1504226752859274,
      "grad_norm": 1.2842893600463867,
      "learning_rate": 0.00010982964227975658,
      "loss": 0.7362,
      "step": 1210
    },
    {
      "epoch": 0.1516658378915962,
      "grad_norm": 0.9873863458633423,
      "learning_rate": 0.00010956696662522569,
      "loss": 0.7374,
      "step": 1220
    },
    {
      "epoch": 0.15290900049726505,
      "grad_norm": 1.2325960397720337,
      "learning_rate": 0.00010930235623204551,
      "loss": 0.6258,
      "step": 1230
    },
    {
      "epoch": 0.15415216310293386,
      "grad_norm": 1.4576919078826904,
      "learning_rate": 0.00010903582197731294,
      "loss": 0.7175,
      "step": 1240
    },
    {
      "epoch": 0.15539532570860268,
      "grad_norm": 1.1568377017974854,
      "learning_rate": 0.00010876737481720722,
      "loss": 0.606,
      "step": 1250
    },
    {
      "epoch": 0.15663848831427152,
      "grad_norm": 1.2811295986175537,
      "learning_rate": 0.0001084970257865397,
      "loss": 0.6926,
      "step": 1260
    },
    {
      "epoch": 0.15788165091994033,
      "grad_norm": 1.1385823488235474,
      "learning_rate": 0.00010822478599830008,
      "loss": 0.6335,
      "step": 1270
    },
    {
      "epoch": 0.15912481352560914,
      "grad_norm": 1.5843505859375,
      "learning_rate": 0.00010795066664319983,
      "loss": 0.7607,
      "step": 1280
    },
    {
      "epoch": 0.16036797613127798,
      "grad_norm": 1.7240452766418457,
      "learning_rate": 0.00010767467898921197,
      "loss": 0.6379,
      "step": 1290
    },
    {
      "epoch": 0.1616111387369468,
      "grad_norm": 1.7185726165771484,
      "learning_rate": 0.00010739683438110797,
      "loss": 0.6669,
      "step": 1300
    },
    {
      "epoch": 0.1628543013426156,
      "grad_norm": 1.4889787435531616,
      "learning_rate": 0.00010711714423999145,
      "loss": 0.6905,
      "step": 1310
    },
    {
      "epoch": 0.16409746394828442,
      "grad_norm": 1.56671142578125,
      "learning_rate": 0.00010683562006282861,
      "loss": 0.6471,
      "step": 1320
    },
    {
      "epoch": 0.16534062655395326,
      "grad_norm": 1.1389793157577515,
      "learning_rate": 0.00010655227342197574,
      "loss": 0.6481,
      "step": 1330
    },
    {
      "epoch": 0.16658378915962208,
      "grad_norm": 1.2167776823043823,
      "learning_rate": 0.00010626711596470343,
      "loss": 0.7018,
      "step": 1340
    },
    {
      "epoch": 0.1678269517652909,
      "grad_norm": 1.3152663707733154,
      "learning_rate": 0.0001059801594127179,
      "loss": 0.6356,
      "step": 1350
    },
    {
      "epoch": 0.1678269517652909,
      "eval_loss": 0.9194591045379639,
      "eval_runtime": 53.4488,
      "eval_samples_per_second": 9.355,
      "eval_steps_per_second": 9.355,
      "step": 1350
    },
    {
      "epoch": 0.16907011437095973,
      "grad_norm": 1.5011839866638184,
      "learning_rate": 0.00010569141556167905,
      "loss": 0.6192,
      "step": 1360
    },
    {
      "epoch": 0.17031327697662854,
      "grad_norm": 2.1147801876068115,
      "learning_rate": 0.00010540089628071566,
      "loss": 0.6289,
      "step": 1370
    },
    {
      "epoch": 0.17155643958229735,
      "grad_norm": 1.8639715909957886,
      "learning_rate": 0.00010510861351193747,
      "loss": 0.6891,
      "step": 1380
    },
    {
      "epoch": 0.1727996021879662,
      "grad_norm": 1.4501938819885254,
      "learning_rate": 0.00010481457926994435,
      "loss": 0.7117,
      "step": 1390
    },
    {
      "epoch": 0.174042764793635,
      "grad_norm": 1.1600079536437988,
      "learning_rate": 0.0001045188056413323,
      "loss": 0.652,
      "step": 1400
    },
    {
      "epoch": 0.17528592739930382,
      "grad_norm": 1.4674696922302246,
      "learning_rate": 0.00010422130478419676,
      "loss": 0.7558,
      "step": 1410
    },
    {
      "epoch": 0.17652909000497266,
      "grad_norm": 1.767659068107605,
      "learning_rate": 0.00010392208892763269,
      "loss": 0.6438,
      "step": 1420
    },
    {
      "epoch": 0.17777225261064147,
      "grad_norm": 1.4168020486831665,
      "learning_rate": 0.00010362117037123204,
      "loss": 0.6173,
      "step": 1430
    },
    {
      "epoch": 0.1790154152163103,
      "grad_norm": 1.6601725816726685,
      "learning_rate": 0.00010331856148457803,
      "loss": 0.68,
      "step": 1440
    },
    {
      "epoch": 0.18025857782197913,
      "grad_norm": 1.2710611820220947,
      "learning_rate": 0.00010301427470673678,
      "loss": 0.6924,
      "step": 1450
    },
    {
      "epoch": 0.18150174042764794,
      "grad_norm": 2.1576950550079346,
      "learning_rate": 0.00010270832254574588,
      "loss": 0.6917,
      "step": 1460
    },
    {
      "epoch": 0.18274490303331675,
      "grad_norm": 1.6391758918762207,
      "learning_rate": 0.00010240071757810036,
      "loss": 0.6717,
      "step": 1470
    },
    {
      "epoch": 0.1839880656389856,
      "grad_norm": 1.4594990015029907,
      "learning_rate": 0.00010209147244823564,
      "loss": 0.7148,
      "step": 1480
    },
    {
      "epoch": 0.1852312282446544,
      "grad_norm": 1.277106523513794,
      "learning_rate": 0.00010178059986800773,
      "loss": 0.6752,
      "step": 1490
    },
    {
      "epoch": 0.18647439085032322,
      "grad_norm": 1.34278404712677,
      "learning_rate": 0.00010146811261617085,
      "loss": 0.7066,
      "step": 1500
    },
    {
      "epoch": 0.18647439085032322,
      "eval_loss": 0.897224485874176,
      "eval_runtime": 55.354,
      "eval_samples_per_second": 9.033,
      "eval_steps_per_second": 9.033,
      "step": 1500
    },
    {
      "epoch": 0.18771755345599203,
      "grad_norm": 1.5729053020477295,
      "learning_rate": 0.00010115402353785197,
      "loss": 0.6708,
      "step": 1510
    },
    {
      "epoch": 0.18896071606166087,
      "grad_norm": 1.7002263069152832,
      "learning_rate": 0.00010083834554402292,
      "loss": 0.6601,
      "step": 1520
    },
    {
      "epoch": 0.19020387866732968,
      "grad_norm": 1.1837941408157349,
      "learning_rate": 0.00010052109161096958,
      "loss": 0.623,
      "step": 1530
    },
    {
      "epoch": 0.1914470412729985,
      "grad_norm": 1.0333565473556519,
      "learning_rate": 0.00010020227477975852,
      "loss": 0.6087,
      "step": 1540
    },
    {
      "epoch": 0.19269020387866734,
      "grad_norm": 1.9234379529953003,
      "learning_rate": 9.9881908155701e-05,
      "loss": 0.7849,
      "step": 1550
    },
    {
      "epoch": 0.19393336648433615,
      "grad_norm": 1.6405690908432007,
      "learning_rate": 9.956000490781411e-05,
      "loss": 0.6642,
      "step": 1560
    },
    {
      "epoch": 0.19517652909000496,
      "grad_norm": 1.6141093969345093,
      "learning_rate": 9.923657826827957e-05,
      "loss": 0.7581,
      "step": 1570
    },
    {
      "epoch": 0.1964196916956738,
      "grad_norm": 1.091253399848938,
      "learning_rate": 9.891164153189976e-05,
      "loss": 0.7226,
      "step": 1580
    },
    {
      "epoch": 0.19766285430134262,
      "grad_norm": 1.7699875831604004,
      "learning_rate": 9.858520805555123e-05,
      "loss": 0.8105,
      "step": 1590
    },
    {
      "epoch": 0.19890601690701143,
      "grad_norm": 1.3844791650772095,
      "learning_rate": 9.825729125763561e-05,
      "loss": 0.6536,
      "step": 1600
    },
    {
      "epoch": 0.20014917951268027,
      "grad_norm": 1.7255157232284546,
      "learning_rate": 9.792790461752813e-05,
      "loss": 0.8065,
      "step": 1610
    },
    {
      "epoch": 0.20139234211834908,
      "grad_norm": 1.0022765398025513,
      "learning_rate": 9.759706167502343e-05,
      "loss": 0.5987,
      "step": 1620
    },
    {
      "epoch": 0.2026355047240179,
      "grad_norm": 1.8386592864990234,
      "learning_rate": 9.726477602977905e-05,
      "loss": 0.8187,
      "step": 1630
    },
    {
      "epoch": 0.20387866732968674,
      "grad_norm": 1.4216880798339844,
      "learning_rate": 9.69310613407564e-05,
      "loss": 0.6245,
      "step": 1640
    },
    {
      "epoch": 0.20512182993535555,
      "grad_norm": 0.9029594659805298,
      "learning_rate": 9.659593132565929e-05,
      "loss": 0.6008,
      "step": 1650
    },
    {
      "epoch": 0.20512182993535555,
      "eval_loss": 0.8831911683082581,
      "eval_runtime": 56.5558,
      "eval_samples_per_second": 8.841,
      "eval_steps_per_second": 8.841,
      "step": 1650
    },
    {
      "epoch": 0.20636499254102436,
      "grad_norm": 1.572769284248352,
      "learning_rate": 9.625939976037002e-05,
      "loss": 0.6875,
      "step": 1660
    },
    {
      "epoch": 0.20760815514669317,
      "grad_norm": 1.2906407117843628,
      "learning_rate": 9.59214804783831e-05,
      "loss": 0.6527,
      "step": 1670
    },
    {
      "epoch": 0.20885131775236201,
      "grad_norm": 1.4845420122146606,
      "learning_rate": 9.558218737023671e-05,
      "loss": 0.6984,
      "step": 1680
    },
    {
      "epoch": 0.21009448035803083,
      "grad_norm": 1.5461996793746948,
      "learning_rate": 9.524153438294159e-05,
      "loss": 0.6153,
      "step": 1690
    },
    {
      "epoch": 0.21133764296369964,
      "grad_norm": 1.531815767288208,
      "learning_rate": 9.489953551940783e-05,
      "loss": 0.563,
      "step": 1700
    },
    {
      "epoch": 0.21258080556936848,
      "grad_norm": 1.5427440404891968,
      "learning_rate": 9.455620483786914e-05,
      "loss": 0.6713,
      "step": 1710
    },
    {
      "epoch": 0.2138239681750373,
      "grad_norm": 1.5302445888519287,
      "learning_rate": 9.421155645130514e-05,
      "loss": 0.6464,
      "step": 1720
    },
    {
      "epoch": 0.2150671307807061,
      "grad_norm": 1.5195375680923462,
      "learning_rate": 9.38656045268611e-05,
      "loss": 0.6482,
      "step": 1730
    },
    {
      "epoch": 0.21631029338637495,
      "grad_norm": 1.5557074546813965,
      "learning_rate": 9.351836328526563e-05,
      "loss": 0.6748,
      "step": 1740
    },
    {
      "epoch": 0.21755345599204376,
      "grad_norm": 1.5321686267852783,
      "learning_rate": 9.316984700024612e-05,
      "loss": 0.6679,
      "step": 1750
    },
    {
      "epoch": 0.21879661859771257,
      "grad_norm": 1.157923698425293,
      "learning_rate": 9.2820069997942e-05,
      "loss": 0.7281,
      "step": 1760
    },
    {
      "epoch": 0.2200397812033814,
      "grad_norm": 1.592752456665039,
      "learning_rate": 9.246904665631588e-05,
      "loss": 0.6926,
      "step": 1770
    },
    {
      "epoch": 0.22128294380905023,
      "grad_norm": 1.5591671466827393,
      "learning_rate": 9.211679140456242e-05,
      "loss": 0.6389,
      "step": 1780
    },
    {
      "epoch": 0.22252610641471904,
      "grad_norm": 1.146720290184021,
      "learning_rate": 9.176331872251536e-05,
      "loss": 0.6265,
      "step": 1790
    },
    {
      "epoch": 0.22376926902038788,
      "grad_norm": 0.9858022928237915,
      "learning_rate": 9.140864314005222e-05,
      "loss": 0.7098,
      "step": 1800
    },
    {
      "epoch": 0.22376926902038788,
      "eval_loss": 0.8798418045043945,
      "eval_runtime": 56.1677,
      "eval_samples_per_second": 8.902,
      "eval_steps_per_second": 8.902,
      "step": 1800
    },
    {
      "epoch": 0.2250124316260567,
      "grad_norm": 1.6383726596832275,
      "learning_rate": 9.105277923649698e-05,
      "loss": 0.6029,
      "step": 1810
    },
    {
      "epoch": 0.2262555942317255,
      "grad_norm": 0.863681435585022,
      "learning_rate": 9.06957416400209e-05,
      "loss": 0.6647,
      "step": 1820
    },
    {
      "epoch": 0.22749875683739434,
      "grad_norm": 1.0826516151428223,
      "learning_rate": 9.03375450270412e-05,
      "loss": 0.6775,
      "step": 1830
    },
    {
      "epoch": 0.22874191944306316,
      "grad_norm": 1.2367980480194092,
      "learning_rate": 8.997820412161764e-05,
      "loss": 0.7778,
      "step": 1840
    },
    {
      "epoch": 0.22998508204873197,
      "grad_norm": 1.4348937273025513,
      "learning_rate": 8.961773369484738e-05,
      "loss": 0.6699,
      "step": 1850
    },
    {
      "epoch": 0.23122824465440078,
      "grad_norm": 0.9706162810325623,
      "learning_rate": 8.925614856425786e-05,
      "loss": 0.684,
      "step": 1860
    },
    {
      "epoch": 0.23247140726006962,
      "grad_norm": 1.4127984046936035,
      "learning_rate": 8.88934635931975e-05,
      "loss": 0.6667,
      "step": 1870
    },
    {
      "epoch": 0.23371456986573844,
      "grad_norm": 1.4040454626083374,
      "learning_rate": 8.852969369022494e-05,
      "loss": 0.6014,
      "step": 1880
    },
    {
      "epoch": 0.23495773247140725,
      "grad_norm": 1.3731218576431274,
      "learning_rate": 8.816485380849613e-05,
      "loss": 0.7063,
      "step": 1890
    },
    {
      "epoch": 0.2362008950770761,
      "grad_norm": 1.5299303531646729,
      "learning_rate": 8.779895894514961e-05,
      "loss": 0.6177,
      "step": 1900
    },
    {
      "epoch": 0.2374440576827449,
      "grad_norm": 1.3770627975463867,
      "learning_rate": 8.743202414069011e-05,
      "loss": 0.6487,
      "step": 1910
    },
    {
      "epoch": 0.23868722028841372,
      "grad_norm": 1.1185230016708374,
      "learning_rate": 8.706406447837023e-05,
      "loss": 0.6612,
      "step": 1920
    },
    {
      "epoch": 0.23993038289408256,
      "grad_norm": 1.027255654335022,
      "learning_rate": 8.669509508357052e-05,
      "loss": 0.6125,
      "step": 1930
    },
    {
      "epoch": 0.24117354549975137,
      "grad_norm": 1.053480625152588,
      "learning_rate": 8.632513112317761e-05,
      "loss": 0.6614,
      "step": 1940
    },
    {
      "epoch": 0.24241670810542018,
      "grad_norm": 1.3678585290908813,
      "learning_rate": 8.59541878049609e-05,
      "loss": 0.5761,
      "step": 1950
    },
    {
      "epoch": 0.24241670810542018,
      "eval_loss": 0.8656662106513977,
      "eval_runtime": 56.0937,
      "eval_samples_per_second": 8.914,
      "eval_steps_per_second": 8.914,
      "step": 1950
    },
    {
      "epoch": 0.24365987071108902,
      "grad_norm": 1.3336479663848877,
      "learning_rate": 8.558228037694728e-05,
      "loss": 0.7036,
      "step": 1960
    },
    {
      "epoch": 0.24490303331675783,
      "grad_norm": 1.3345288038253784,
      "learning_rate": 8.520942412679447e-05,
      "loss": 0.6508,
      "step": 1970
    },
    {
      "epoch": 0.24614619592242665,
      "grad_norm": 1.4155631065368652,
      "learning_rate": 8.483563438116257e-05,
      "loss": 0.6446,
      "step": 1980
    },
    {
      "epoch": 0.2473893585280955,
      "grad_norm": 1.3814365863800049,
      "learning_rate": 8.446092650508393e-05,
      "loss": 0.6557,
      "step": 1990
    },
    {
      "epoch": 0.2486325211337643,
      "grad_norm": 1.9659488201141357,
      "learning_rate": 8.408531590133172e-05,
      "loss": 0.693,
      "step": 2000
    },
    {
      "epoch": 0.2498756837394331,
      "grad_norm": 1.5568230152130127,
      "learning_rate": 8.370881800978673e-05,
      "loss": 0.5812,
      "step": 2010
    },
    {
      "epoch": 0.25111884634510195,
      "grad_norm": 1.5361123085021973,
      "learning_rate": 8.333144830680262e-05,
      "loss": 0.6411,
      "step": 2020
    },
    {
      "epoch": 0.25236200895077077,
      "grad_norm": 1.9059866666793823,
      "learning_rate": 8.29532223045698e-05,
      "loss": 0.7285,
      "step": 2030
    },
    {
      "epoch": 0.2536051715564396,
      "grad_norm": 1.3273738622665405,
      "learning_rate": 8.257415555047785e-05,
      "loss": 0.6589,
      "step": 2040
    },
    {
      "epoch": 0.2548483341621084,
      "grad_norm": 1.4303077459335327,
      "learning_rate": 8.21942636264763e-05,
      "loss": 0.6544,
      "step": 2050
    },
    {
      "epoch": 0.2560914967677772,
      "grad_norm": 1.2462729215621948,
      "learning_rate": 8.181356214843422e-05,
      "loss": 0.5755,
      "step": 2060
    },
    {
      "epoch": 0.2573346593734461,
      "grad_norm": 1.2266968488693237,
      "learning_rate": 8.143206676549826e-05,
      "loss": 0.5514,
      "step": 2070
    },
    {
      "epoch": 0.2585778219791149,
      "grad_norm": 1.5586425065994263,
      "learning_rate": 8.10497931594494e-05,
      "loss": 0.6775,
      "step": 2080
    },
    {
      "epoch": 0.2598209845847837,
      "grad_norm": 1.5792748928070068,
      "learning_rate": 8.066675704405836e-05,
      "loss": 0.6575,
      "step": 2090
    },
    {
      "epoch": 0.2610641471904525,
      "grad_norm": 1.0587018728256226,
      "learning_rate": 8.028297416443952e-05,
      "loss": 0.586,
      "step": 2100
    },
    {
      "epoch": 0.2610641471904525,
      "eval_loss": 0.8552118539810181,
      "eval_runtime": 54.3671,
      "eval_samples_per_second": 9.197,
      "eval_steps_per_second": 9.197,
      "step": 2100
    },
    {
      "epoch": 0.2623073097961213,
      "grad_norm": 1.3237509727478027,
      "learning_rate": 7.989846029640397e-05,
      "loss": 0.6647,
      "step": 2110
    },
    {
      "epoch": 0.26355047240179014,
      "grad_norm": 1.246394157409668,
      "learning_rate": 7.951323124581069e-05,
      "loss": 0.6031,
      "step": 2120
    },
    {
      "epoch": 0.26479363500745895,
      "grad_norm": 1.697279930114746,
      "learning_rate": 7.91273028479172e-05,
      "loss": 0.5719,
      "step": 2130
    },
    {
      "epoch": 0.2660367976131278,
      "grad_norm": 1.3424289226531982,
      "learning_rate": 7.874069096672831e-05,
      "loss": 0.6554,
      "step": 2140
    },
    {
      "epoch": 0.26727996021879663,
      "grad_norm": 1.3016737699508667,
      "learning_rate": 7.83534114943442e-05,
      "loss": 0.6112,
      "step": 2150
    },
    {
      "epoch": 0.26852312282446544,
      "grad_norm": 1.566977858543396,
      "learning_rate": 7.796548035030715e-05,
      "loss": 0.6176,
      "step": 2160
    },
    {
      "epoch": 0.26976628543013426,
      "grad_norm": 1.2994070053100586,
      "learning_rate": 7.757691348094703e-05,
      "loss": 0.6126,
      "step": 2170
    },
    {
      "epoch": 0.27100944803580307,
      "grad_norm": 0.9738805890083313,
      "learning_rate": 7.718772685872595e-05,
      "loss": 0.6169,
      "step": 2180
    },
    {
      "epoch": 0.2722526106414719,
      "grad_norm": 1.048270344734192,
      "learning_rate": 7.679793648158159e-05,
      "loss": 0.5741,
      "step": 2190
    },
    {
      "epoch": 0.27349577324714075,
      "grad_norm": 2.1474382877349854,
      "learning_rate": 7.640755837226965e-05,
      "loss": 0.6018,
      "step": 2200
    },
    {
      "epoch": 0.27473893585280956,
      "grad_norm": 2.463155746459961,
      "learning_rate": 7.601660857770522e-05,
      "loss": 0.6934,
      "step": 2210
    },
    {
      "epoch": 0.2759820984584784,
      "grad_norm": 1.2787517309188843,
      "learning_rate": 7.562510316830308e-05,
      "loss": 0.7,
      "step": 2220
    },
    {
      "epoch": 0.2772252610641472,
      "grad_norm": 1.377604603767395,
      "learning_rate": 7.523305823731723e-05,
      "loss": 0.5736,
      "step": 2230
    },
    {
      "epoch": 0.278468423669816,
      "grad_norm": 1.2263590097427368,
      "learning_rate": 7.484048990017919e-05,
      "loss": 0.6463,
      "step": 2240
    },
    {
      "epoch": 0.2797115862754848,
      "grad_norm": 1.2096174955368042,
      "learning_rate": 7.444741429383578e-05,
      "loss": 0.6445,
      "step": 2250
    },
    {
      "epoch": 0.2797115862754848,
      "eval_loss": 0.8472199440002441,
      "eval_runtime": 54.839,
      "eval_samples_per_second": 9.118,
      "eval_steps_per_second": 9.118,
      "step": 2250
    },
    {
      "epoch": 0.2809547488811537,
      "grad_norm": 1.337813138961792,
      "learning_rate": 7.405384757608555e-05,
      "loss": 0.571,
      "step": 2260
    },
    {
      "epoch": 0.2821979114868225,
      "grad_norm": 1.363511323928833,
      "learning_rate": 7.36598059249148e-05,
      "loss": 0.5797,
      "step": 2270
    },
    {
      "epoch": 0.2834410740924913,
      "grad_norm": 1.3352985382080078,
      "learning_rate": 7.326530553783243e-05,
      "loss": 0.6532,
      "step": 2280
    },
    {
      "epoch": 0.2846842366981601,
      "grad_norm": 1.102362871170044,
      "learning_rate": 7.287036263120425e-05,
      "loss": 0.4812,
      "step": 2290
    },
    {
      "epoch": 0.28592739930382893,
      "grad_norm": 0.9128854274749756,
      "learning_rate": 7.247499343958621e-05,
      "loss": 0.6263,
      "step": 2300
    },
    {
      "epoch": 0.28717056190949775,
      "grad_norm": 1.369809865951538,
      "learning_rate": 7.207921421505724e-05,
      "loss": 0.5442,
      "step": 2310
    },
    {
      "epoch": 0.28841372451516656,
      "grad_norm": 1.3508280515670776,
      "learning_rate": 7.168304122655113e-05,
      "loss": 0.6703,
      "step": 2320
    },
    {
      "epoch": 0.2896568871208354,
      "grad_norm": 1.2335124015808105,
      "learning_rate": 7.128649075918768e-05,
      "loss": 0.733,
      "step": 2330
    },
    {
      "epoch": 0.29090004972650424,
      "grad_norm": 1.152423620223999,
      "learning_rate": 7.088957911360347e-05,
      "loss": 0.7101,
      "step": 2340
    },
    {
      "epoch": 0.29214321233217305,
      "grad_norm": 1.1230435371398926,
      "learning_rate": 7.049232260528163e-05,
      "loss": 0.6501,
      "step": 2350
    },
    {
      "epoch": 0.29338637493784187,
      "grad_norm": 1.2065176963806152,
      "learning_rate": 7.009473756388128e-05,
      "loss": 0.5584,
      "step": 2360
    },
    {
      "epoch": 0.2946295375435107,
      "grad_norm": 1.258360743522644,
      "learning_rate": 6.969684033256622e-05,
      "loss": 0.6351,
      "step": 2370
    },
    {
      "epoch": 0.2958727001491795,
      "grad_norm": 1.4752328395843506,
      "learning_rate": 6.92986472673332e-05,
      "loss": 0.5295,
      "step": 2380
    },
    {
      "epoch": 0.29711586275484836,
      "grad_norm": 1.4005481004714966,
      "learning_rate": 6.890017473633946e-05,
      "loss": 0.6168,
      "step": 2390
    },
    {
      "epoch": 0.29835902536051717,
      "grad_norm": 1.4845457077026367,
      "learning_rate": 6.850143911923011e-05,
      "loss": 0.6611,
      "step": 2400
    },
    {
      "epoch": 0.29835902536051717,
      "eval_loss": 0.842125415802002,
      "eval_runtime": 54.7426,
      "eval_samples_per_second": 9.134,
      "eval_steps_per_second": 9.134,
      "step": 2400
    },
    {
      "epoch": 0.299602187966186,
      "grad_norm": 1.1345279216766357,
      "learning_rate": 6.81024568064646e-05,
      "loss": 0.647,
      "step": 2410
    },
    {
      "epoch": 0.3008453505718548,
      "grad_norm": 1.0600401163101196,
      "learning_rate": 6.770324419864309e-05,
      "loss": 0.5967,
      "step": 2420
    },
    {
      "epoch": 0.3020885131775236,
      "grad_norm": 1.600447177886963,
      "learning_rate": 6.73038177058323e-05,
      "loss": 0.6442,
      "step": 2430
    },
    {
      "epoch": 0.3033316757831924,
      "grad_norm": 1.2468013763427734,
      "learning_rate": 6.690419374689087e-05,
      "loss": 0.6526,
      "step": 2440
    },
    {
      "epoch": 0.30457483838886124,
      "grad_norm": 1.0898563861846924,
      "learning_rate": 6.650438874879456e-05,
      "loss": 0.641,
      "step": 2450
    },
    {
      "epoch": 0.3058180009945301,
      "grad_norm": 0.772791862487793,
      "learning_rate": 6.61044191459609e-05,
      "loss": 0.5612,
      "step": 2460
    },
    {
      "epoch": 0.3070611636001989,
      "grad_norm": 1.353593349456787,
      "learning_rate": 6.57043013795737e-05,
      "loss": 0.6442,
      "step": 2470
    },
    {
      "epoch": 0.30830432620586773,
      "grad_norm": 1.1054726839065552,
      "learning_rate": 6.530405189690719e-05,
      "loss": 0.5728,
      "step": 2480
    },
    {
      "epoch": 0.30954748881153654,
      "grad_norm": 1.6124745607376099,
      "learning_rate": 6.49036871506499e-05,
      "loss": 0.5805,
      "step": 2490
    },
    {
      "epoch": 0.31079065141720535,
      "grad_norm": 1.458871603012085,
      "learning_rate": 6.450322359822846e-05,
      "loss": 0.554,
      "step": 2500
    },
    {
      "epoch": 0.31203381402287417,
      "grad_norm": 1.366873860359192,
      "learning_rate": 6.410267770113098e-05,
      "loss": 0.5137,
      "step": 2510
    },
    {
      "epoch": 0.31327697662854304,
      "grad_norm": 1.7523047924041748,
      "learning_rate": 6.370206592423045e-05,
      "loss": 0.59,
      "step": 2520
    },
    {
      "epoch": 0.31452013923421185,
      "grad_norm": 2.6855154037475586,
      "learning_rate": 6.330140473510796e-05,
      "loss": 0.5745,
      "step": 2530
    },
    {
      "epoch": 0.31576330183988066,
      "grad_norm": 1.4660212993621826,
      "learning_rate": 6.29007106033757e-05,
      "loss": 0.5887,
      "step": 2540
    },
    {
      "epoch": 0.3170064644455495,
      "grad_norm": 1.741590142250061,
      "learning_rate": 6.25e-05,
      "loss": 0.6324,
      "step": 2550
    },
    {
      "epoch": 0.3170064644455495,
      "eval_loss": 0.826242208480835,
      "eval_runtime": 54.5046,
      "eval_samples_per_second": 9.174,
      "eval_steps_per_second": 9.174,
      "step": 2550
    },
    {
      "epoch": 0.3182496270512183,
      "grad_norm": 1.1369116306304932,
      "learning_rate": 6.20992893966243e-05,
      "loss": 0.5283,
      "step": 2560
    },
    {
      "epoch": 0.3194927896568871,
      "grad_norm": 0.9932198524475098,
      "learning_rate": 6.169859526489204e-05,
      "loss": 0.6021,
      "step": 2570
    },
    {
      "epoch": 0.32073595226255597,
      "grad_norm": 1.0104336738586426,
      "learning_rate": 6.129793407576955e-05,
      "loss": 0.6471,
      "step": 2580
    },
    {
      "epoch": 0.3219791148682248,
      "grad_norm": 1.2468757629394531,
      "learning_rate": 6.089732229886904e-05,
      "loss": 0.698,
      "step": 2590
    },
    {
      "epoch": 0.3232222774738936,
      "grad_norm": 1.2286524772644043,
      "learning_rate": 6.049677640177155e-05,
      "loss": 0.5984,
      "step": 2600
    },
    {
      "epoch": 0.3244654400795624,
      "grad_norm": 1.4735106229782104,
      "learning_rate": 6.00963128493501e-05,
      "loss": 0.5562,
      "step": 2610
    },
    {
      "epoch": 0.3257086026852312,
      "grad_norm": 1.5603587627410889,
      "learning_rate": 5.969594810309284e-05,
      "loss": 0.6013,
      "step": 2620
    },
    {
      "epoch": 0.32695176529090003,
      "grad_norm": 1.1900887489318848,
      "learning_rate": 5.929569862042631e-05,
      "loss": 0.6304,
      "step": 2630
    },
    {
      "epoch": 0.32819492789656884,
      "grad_norm": 1.482504963874817,
      "learning_rate": 5.889558085403911e-05,
      "loss": 0.5655,
      "step": 2640
    },
    {
      "epoch": 0.3294380905022377,
      "grad_norm": 1.5434118509292603,
      "learning_rate": 5.849561125120545e-05,
      "loss": 0.6596,
      "step": 2650
    },
    {
      "epoch": 0.3306812531079065,
      "grad_norm": 1.8465694189071655,
      "learning_rate": 5.809580625310912e-05,
      "loss": 0.6457,
      "step": 2660
    },
    {
      "epoch": 0.33192441571357534,
      "grad_norm": 1.7762593030929565,
      "learning_rate": 5.769618229416773e-05,
      "loss": 0.5508,
      "step": 2670
    },
    {
      "epoch": 0.33316757831924415,
      "grad_norm": 1.78951895236969,
      "learning_rate": 5.7296755801356926e-05,
      "loss": 0.5979,
      "step": 2680
    },
    {
      "epoch": 0.33441074092491296,
      "grad_norm": 1.1297141313552856,
      "learning_rate": 5.6897543193535414e-05,
      "loss": 0.5565,
      "step": 2690
    },
    {
      "epoch": 0.3356539035305818,
      "grad_norm": 1.264514684677124,
      "learning_rate": 5.649856088076989e-05,
      "loss": 0.5706,
      "step": 2700
    },
    {
      "epoch": 0.3356539035305818,
      "eval_loss": 0.8196865320205688,
      "eval_runtime": 54.3418,
      "eval_samples_per_second": 9.201,
      "eval_steps_per_second": 9.201,
      "step": 2700
    },
    {
      "epoch": 0.33689706613625064,
      "grad_norm": 1.7647141218185425,
      "learning_rate": 5.609982526366054e-05,
      "loss": 0.5757,
      "step": 2710
    },
    {
      "epoch": 0.33814022874191946,
      "grad_norm": 1.655874252319336,
      "learning_rate": 5.570135273266683e-05,
      "loss": 0.6219,
      "step": 2720
    },
    {
      "epoch": 0.33938339134758827,
      "grad_norm": 1.6094146966934204,
      "learning_rate": 5.53031596674338e-05,
      "loss": 0.6181,
      "step": 2730
    },
    {
      "epoch": 0.3406265539532571,
      "grad_norm": 1.2176001071929932,
      "learning_rate": 5.490526243611873e-05,
      "loss": 0.5794,
      "step": 2740
    },
    {
      "epoch": 0.3418697165589259,
      "grad_norm": 1.6560986042022705,
      "learning_rate": 5.450767739471837e-05,
      "loss": 0.5676,
      "step": 2750
    },
    {
      "epoch": 0.3431128791645947,
      "grad_norm": 1.5764137506484985,
      "learning_rate": 5.411042088639655e-05,
      "loss": 0.5182,
      "step": 2760
    },
    {
      "epoch": 0.3443560417702636,
      "grad_norm": 0.9934073686599731,
      "learning_rate": 5.371350924081234e-05,
      "loss": 0.5327,
      "step": 2770
    },
    {
      "epoch": 0.3455992043759324,
      "grad_norm": 1.5813066959381104,
      "learning_rate": 5.331695877344888e-05,
      "loss": 0.6101,
      "step": 2780
    },
    {
      "epoch": 0.3468423669816012,
      "grad_norm": 1.0345252752304077,
      "learning_rate": 5.292078578494275e-05,
      "loss": 0.5376,
      "step": 2790
    },
    {
      "epoch": 0.34808552958727,
      "grad_norm": 1.1901750564575195,
      "learning_rate": 5.2525006560413816e-05,
      "loss": 0.5869,
      "step": 2800
    },
    {
      "epoch": 0.34932869219293883,
      "grad_norm": 1.1224308013916016,
      "learning_rate": 5.212963736879578e-05,
      "loss": 0.5736,
      "step": 2810
    },
    {
      "epoch": 0.35057185479860764,
      "grad_norm": 1.1819744110107422,
      "learning_rate": 5.173469446216757e-05,
      "loss": 0.4877,
      "step": 2820
    },
    {
      "epoch": 0.35181501740427645,
      "grad_norm": 1.3149231672286987,
      "learning_rate": 5.134019407508521e-05,
      "loss": 0.6002,
      "step": 2830
    },
    {
      "epoch": 0.3530581800099453,
      "grad_norm": 1.876297950744629,
      "learning_rate": 5.0946152423914456e-05,
      "loss": 0.6038,
      "step": 2840
    },
    {
      "epoch": 0.35430134261561413,
      "grad_norm": 1.6570100784301758,
      "learning_rate": 5.0552585706164246e-05,
      "loss": 0.5392,
      "step": 2850
    },
    {
      "epoch": 0.35430134261561413,
      "eval_loss": 0.813827633857727,
      "eval_runtime": 56.4795,
      "eval_samples_per_second": 8.853,
      "eval_steps_per_second": 8.853,
      "step": 2850
    },
    {
      "epoch": 0.35554450522128295,
      "grad_norm": 1.4959031343460083,
      "learning_rate": 5.015951009982081e-05,
      "loss": 0.6377,
      "step": 2860
    },
    {
      "epoch": 0.35678766782695176,
      "grad_norm": 1.524834394454956,
      "learning_rate": 4.976694176268278e-05,
      "loss": 0.6132,
      "step": 2870
    },
    {
      "epoch": 0.3580308304326206,
      "grad_norm": 1.1156436204910278,
      "learning_rate": 4.937489683169692e-05,
      "loss": 0.5865,
      "step": 2880
    },
    {
      "epoch": 0.3592739930382894,
      "grad_norm": 1.4648617506027222,
      "learning_rate": 4.8983391422294786e-05,
      "loss": 0.5489,
      "step": 2890
    },
    {
      "epoch": 0.36051715564395825,
      "grad_norm": 0.9886593222618103,
      "learning_rate": 4.8592441627730355e-05,
      "loss": 0.5844,
      "step": 2900
    },
    {
      "epoch": 0.36176031824962707,
      "grad_norm": 1.7449545860290527,
      "learning_rate": 4.820206351841842e-05,
      "loss": 0.617,
      "step": 2910
    },
    {
      "epoch": 0.3630034808552959,
      "grad_norm": 1.375961184501648,
      "learning_rate": 4.781227314127405e-05,
      "loss": 0.6068,
      "step": 2920
    },
    {
      "epoch": 0.3642466434609647,
      "grad_norm": 1.08450448513031,
      "learning_rate": 4.7423086519052966e-05,
      "loss": 0.6064,
      "step": 2930
    },
    {
      "epoch": 0.3654898060666335,
      "grad_norm": 1.6093776226043701,
      "learning_rate": 4.703451964969287e-05,
      "loss": 0.6449,
      "step": 2940
    },
    {
      "epoch": 0.3667329686723023,
      "grad_norm": 1.2619701623916626,
      "learning_rate": 4.66465885056558e-05,
      "loss": 0.5688,
      "step": 2950
    },
    {
      "epoch": 0.3679761312779712,
      "grad_norm": 1.2892088890075684,
      "learning_rate": 4.62593090332717e-05,
      "loss": 0.5941,
      "step": 2960
    },
    {
      "epoch": 0.36921929388364,
      "grad_norm": 1.0506901741027832,
      "learning_rate": 4.587269715208281e-05,
      "loss": 0.5454,
      "step": 2970
    },
    {
      "epoch": 0.3704624564893088,
      "grad_norm": 1.6574184894561768,
      "learning_rate": 4.5486768754189305e-05,
      "loss": 0.5409,
      "step": 2980
    },
    {
      "epoch": 0.3717056190949776,
      "grad_norm": 1.1367279291152954,
      "learning_rate": 4.510153970359606e-05,
      "loss": 0.6502,
      "step": 2990
    },
    {
      "epoch": 0.37294878170064644,
      "grad_norm": 1.1561517715454102,
      "learning_rate": 4.4717025835560476e-05,
      "loss": 0.5962,
      "step": 3000
    },
    {
      "epoch": 0.37294878170064644,
      "eval_loss": 0.8055340647697449,
      "eval_runtime": 54.6339,
      "eval_samples_per_second": 9.152,
      "eval_steps_per_second": 9.152,
      "step": 3000
    },
    {
      "epoch": 0.37419194430631525,
      "grad_norm": 1.3308610916137695,
      "learning_rate": 4.433324295594166e-05,
      "loss": 0.6302,
      "step": 3010
    },
    {
      "epoch": 0.37543510691198406,
      "grad_norm": 1.3117073774337769,
      "learning_rate": 4.3950206840550585e-05,
      "loss": 0.5286,
      "step": 3020
    },
    {
      "epoch": 0.37667826951765293,
      "grad_norm": 1.3257042169570923,
      "learning_rate": 4.3567933234501746e-05,
      "loss": 0.6227,
      "step": 3030
    },
    {
      "epoch": 0.37792143212332174,
      "grad_norm": 1.614931583404541,
      "learning_rate": 4.318643785156579e-05,
      "loss": 0.5531,
      "step": 3040
    },
    {
      "epoch": 0.37916459472899056,
      "grad_norm": 1.4024949073791504,
      "learning_rate": 4.280573637352371e-05,
      "loss": 0.6107,
      "step": 3050
    },
    {
      "epoch": 0.38040775733465937,
      "grad_norm": 1.3442318439483643,
      "learning_rate": 4.242584444952216e-05,
      "loss": 0.619,
      "step": 3060
    },
    {
      "epoch": 0.3816509199403282,
      "grad_norm": 1.6472797393798828,
      "learning_rate": 4.204677769543019e-05,
      "loss": 0.6219,
      "step": 3070
    },
    {
      "epoch": 0.382894082545997,
      "grad_norm": 1.226382851600647,
      "learning_rate": 4.16685516931974e-05,
      "loss": 0.5669,
      "step": 3080
    },
    {
      "epoch": 0.38413724515166586,
      "grad_norm": 1.2685925960540771,
      "learning_rate": 4.1291181990213286e-05,
      "loss": 0.5875,
      "step": 3090
    },
    {
      "epoch": 0.3853804077573347,
      "grad_norm": 2.1690385341644287,
      "learning_rate": 4.0914684098668286e-05,
      "loss": 0.6366,
      "step": 3100
    },
    {
      "epoch": 0.3866235703630035,
      "grad_norm": 1.2350751161575317,
      "learning_rate": 4.053907349491608e-05,
      "loss": 0.6034,
      "step": 3110
    },
    {
      "epoch": 0.3878667329686723,
      "grad_norm": 1.4489704370498657,
      "learning_rate": 4.016436561883746e-05,
      "loss": 0.6346,
      "step": 3120
    },
    {
      "epoch": 0.3891098955743411,
      "grad_norm": 1.332980990409851,
      "learning_rate": 3.979057587320554e-05,
      "loss": 0.5858,
      "step": 3130
    },
    {
      "epoch": 0.3903530581800099,
      "grad_norm": 1.391718864440918,
      "learning_rate": 3.941771962305274e-05,
      "loss": 0.7218,
      "step": 3140
    },
    {
      "epoch": 0.39159622078567874,
      "grad_norm": 1.269722819328308,
      "learning_rate": 3.9045812195039125e-05,
      "loss": 0.5846,
      "step": 3150
    },
    {
      "epoch": 0.39159622078567874,
      "eval_loss": 0.801069974899292,
      "eval_runtime": 56.6006,
      "eval_samples_per_second": 8.834,
      "eval_steps_per_second": 8.834,
      "step": 3150
    },
    {
      "epoch": 0.3928393833913476,
      "grad_norm": 1.8093242645263672,
      "learning_rate": 3.8674868876822395e-05,
      "loss": 0.5608,
      "step": 3160
    },
    {
      "epoch": 0.3940825459970164,
      "grad_norm": 1.6383775472640991,
      "learning_rate": 3.83049049164295e-05,
      "loss": 0.5706,
      "step": 3170
    },
    {
      "epoch": 0.39532570860268523,
      "grad_norm": 1.7762494087219238,
      "learning_rate": 3.793593552162978e-05,
      "loss": 0.6272,
      "step": 3180
    },
    {
      "epoch": 0.39656887120835405,
      "grad_norm": 1.989702582359314,
      "learning_rate": 3.75679758593099e-05,
      "loss": 0.6268,
      "step": 3190
    },
    {
      "epoch": 0.39781203381402286,
      "grad_norm": 1.2394602298736572,
      "learning_rate": 3.720104105485039e-05,
      "loss": 0.5745,
      "step": 3200
    },
    {
      "epoch": 0.39905519641969167,
      "grad_norm": 1.6666808128356934,
      "learning_rate": 3.6835146191503885e-05,
      "loss": 0.6287,
      "step": 3210
    },
    {
      "epoch": 0.40029835902536054,
      "grad_norm": 0.926642119884491,
      "learning_rate": 3.647030630977508e-05,
      "loss": 0.6038,
      "step": 3220
    },
    {
      "epoch": 0.40154152163102935,
      "grad_norm": 1.3358100652694702,
      "learning_rate": 3.6106536406802524e-05,
      "loss": 0.5941,
      "step": 3230
    },
    {
      "epoch": 0.40278468423669817,
      "grad_norm": 1.339179277420044,
      "learning_rate": 3.5743851435742176e-05,
      "loss": 0.5888,
      "step": 3240
    },
    {
      "epoch": 0.404027846842367,
      "grad_norm": 1.4704395532608032,
      "learning_rate": 3.538226630515262e-05,
      "loss": 0.5113,
      "step": 3250
    },
    {
      "epoch": 0.4052710094480358,
      "grad_norm": 1.2576725482940674,
      "learning_rate": 3.502179587838238e-05,
      "loss": 0.5874,
      "step": 3260
    },
    {
      "epoch": 0.4065141720537046,
      "grad_norm": 1.1804664134979248,
      "learning_rate": 3.46624549729588e-05,
      "loss": 0.6054,
      "step": 3270
    },
    {
      "epoch": 0.40775733465937347,
      "grad_norm": 1.6472457647323608,
      "learning_rate": 3.430425835997908e-05,
      "loss": 0.6168,
      "step": 3280
    },
    {
      "epoch": 0.4090004972650423,
      "grad_norm": 1.3699522018432617,
      "learning_rate": 3.394722076350302e-05,
      "loss": 0.5227,
      "step": 3290
    },
    {
      "epoch": 0.4102436598707111,
      "grad_norm": 0.9297524690628052,
      "learning_rate": 3.359135685994781e-05,
      "loss": 0.5818,
      "step": 3300
    },
    {
      "epoch": 0.4102436598707111,
      "eval_loss": 0.7923147678375244,
      "eval_runtime": 53.7845,
      "eval_samples_per_second": 9.296,
      "eval_steps_per_second": 9.296,
      "step": 3300
    },
    {
      "epoch": 0.4114868224763799,
      "grad_norm": 1.4288136959075928,
      "learning_rate": 3.3236681277484654e-05,
      "loss": 0.5647,
      "step": 3310
    },
    {
      "epoch": 0.4127299850820487,
      "grad_norm": 1.5457046031951904,
      "learning_rate": 3.2883208595437584e-05,
      "loss": 0.6427,
      "step": 3320
    },
    {
      "epoch": 0.41397314768771754,
      "grad_norm": 1.0546395778656006,
      "learning_rate": 3.2530953343684136e-05,
      "loss": 0.6108,
      "step": 3330
    },
    {
      "epoch": 0.41521631029338635,
      "grad_norm": 1.3705288171768188,
      "learning_rate": 3.217993000205799e-05,
      "loss": 0.6249,
      "step": 3340
    },
    {
      "epoch": 0.4164594728990552,
      "grad_norm": 1.1312577724456787,
      "learning_rate": 3.1830152999753903e-05,
      "loss": 0.6543,
      "step": 3350
    },
    {
      "epoch": 0.41770263550472403,
      "grad_norm": 1.5398489236831665,
      "learning_rate": 3.148163671473439e-05,
      "loss": 0.6304,
      "step": 3360
    },
    {
      "epoch": 0.41894579811039284,
      "grad_norm": 1.4748462438583374,
      "learning_rate": 3.113439547313892e-05,
      "loss": 0.5638,
      "step": 3370
    },
    {
      "epoch": 0.42018896071606165,
      "grad_norm": 0.9427064061164856,
      "learning_rate": 3.0788443548694874e-05,
      "loss": 0.6171,
      "step": 3380
    },
    {
      "epoch": 0.42143212332173047,
      "grad_norm": 1.718713402748108,
      "learning_rate": 3.0443795162130876e-05,
      "loss": 0.6358,
      "step": 3390
    },
    {
      "epoch": 0.4226752859273993,
      "grad_norm": 1.058449387550354,
      "learning_rate": 3.0100464480592185e-05,
      "loss": 0.6129,
      "step": 3400
    },
    {
      "epoch": 0.42391844853306815,
      "grad_norm": 1.0851000547409058,
      "learning_rate": 2.9758465617058404e-05,
      "loss": 0.6621,
      "step": 3410
    },
    {
      "epoch": 0.42516161113873696,
      "grad_norm": 1.3645485639572144,
      "learning_rate": 2.9417812629763285e-05,
      "loss": 0.6181,
      "step": 3420
    },
    {
      "epoch": 0.4264047737444058,
      "grad_norm": 1.682470679283142,
      "learning_rate": 2.9078519521616894e-05,
      "loss": 0.595,
      "step": 3430
    },
    {
      "epoch": 0.4276479363500746,
      "grad_norm": 1.1072993278503418,
      "learning_rate": 2.8740600239630002e-05,
      "loss": 0.564,
      "step": 3440
    },
    {
      "epoch": 0.4288910989557434,
      "grad_norm": 2.1429033279418945,
      "learning_rate": 2.8404068674340714e-05,
      "loss": 0.5779,
      "step": 3450
    },
    {
      "epoch": 0.4288910989557434,
      "eval_loss": 0.7892328500747681,
      "eval_runtime": 54.8609,
      "eval_samples_per_second": 9.114,
      "eval_steps_per_second": 9.114,
      "step": 3450
    },
    {
      "epoch": 0.4301342615614122,
      "grad_norm": 1.1970041990280151,
      "learning_rate": 2.80689386592436e-05,
      "loss": 0.5467,
      "step": 3460
    },
    {
      "epoch": 0.4313774241670811,
      "grad_norm": 1.570540189743042,
      "learning_rate": 2.7735223970220955e-05,
      "loss": 0.5794,
      "step": 3470
    },
    {
      "epoch": 0.4326205867727499,
      "grad_norm": 1.2686924934387207,
      "learning_rate": 2.7402938324976576e-05,
      "loss": 0.4993,
      "step": 3480
    },
    {
      "epoch": 0.4338637493784187,
      "grad_norm": 1.3892414569854736,
      "learning_rate": 2.70720953824719e-05,
      "loss": 0.515,
      "step": 3490
    },
    {
      "epoch": 0.4351069119840875,
      "grad_norm": 1.051397681236267,
      "learning_rate": 2.674270874236441e-05,
      "loss": 0.542,
      "step": 3500
    },
    {
      "epoch": 0.43635007458975633,
      "grad_norm": 0.9418326020240784,
      "learning_rate": 2.64147919444488e-05,
      "loss": 0.5283,
      "step": 3510
    },
    {
      "epoch": 0.43759323719542514,
      "grad_norm": 1.1212824583053589,
      "learning_rate": 2.6088358468100247e-05,
      "loss": 0.4912,
      "step": 3520
    },
    {
      "epoch": 0.43883639980109396,
      "grad_norm": 1.3727294206619263,
      "learning_rate": 2.5763421731720435e-05,
      "loss": 0.5601,
      "step": 3530
    },
    {
      "epoch": 0.4400795624067628,
      "grad_norm": 1.0249736309051514,
      "learning_rate": 2.5439995092185892e-05,
      "loss": 0.4856,
      "step": 3540
    },
    {
      "epoch": 0.44132272501243164,
      "grad_norm": 1.2564889192581177,
      "learning_rate": 2.5118091844299e-05,
      "loss": 0.5795,
      "step": 3550
    },
    {
      "epoch": 0.44256588761810045,
      "grad_norm": 3.1751506328582764,
      "learning_rate": 2.479772522024147e-05,
      "loss": 0.5902,
      "step": 3560
    },
    {
      "epoch": 0.44380905022376926,
      "grad_norm": 1.3278981447219849,
      "learning_rate": 2.4478908389030427e-05,
      "loss": 0.5305,
      "step": 3570
    },
    {
      "epoch": 0.4450522128294381,
      "grad_norm": 1.1867163181304932,
      "learning_rate": 2.41616544559771e-05,
      "loss": 0.6074,
      "step": 3580
    },
    {
      "epoch": 0.4462953754351069,
      "grad_norm": 1.357256293296814,
      "learning_rate": 2.3845976462148033e-05,
      "loss": 0.5541,
      "step": 3590
    },
    {
      "epoch": 0.44753853804077576,
      "grad_norm": 1.0771019458770752,
      "learning_rate": 2.3531887383829157e-05,
      "loss": 0.4935,
      "step": 3600
    },
    {
      "epoch": 0.44753853804077576,
      "eval_loss": 0.7831114530563354,
      "eval_runtime": 55.7555,
      "eval_samples_per_second": 8.968,
      "eval_steps_per_second": 8.968,
      "step": 3600
    },
    {
      "epoch": 0.44878170064644457,
      "grad_norm": 1.402942419052124,
      "learning_rate": 2.3219400131992273e-05,
      "loss": 0.6078,
      "step": 3610
    },
    {
      "epoch": 0.4500248632521134,
      "grad_norm": 0.9987258315086365,
      "learning_rate": 2.2908527551764404e-05,
      "loss": 0.5269,
      "step": 3620
    },
    {
      "epoch": 0.4512680258577822,
      "grad_norm": 1.655179500579834,
      "learning_rate": 2.259928242189966e-05,
      "loss": 0.5955,
      "step": 3630
    },
    {
      "epoch": 0.452511188463451,
      "grad_norm": 1.3401436805725098,
      "learning_rate": 2.2291677454254136e-05,
      "loss": 0.6061,
      "step": 3640
    },
    {
      "epoch": 0.4537543510691198,
      "grad_norm": 1.766615629196167,
      "learning_rate": 2.1985725293263237e-05,
      "loss": 0.6185,
      "step": 3650
    },
    {
      "epoch": 0.4549975136747887,
      "grad_norm": 1.7541744709014893,
      "learning_rate": 2.1681438515421953e-05,
      "loss": 0.5724,
      "step": 3660
    },
    {
      "epoch": 0.4562406762804575,
      "grad_norm": 1.9028109312057495,
      "learning_rate": 2.1378829628767965e-05,
      "loss": 0.5688,
      "step": 3670
    },
    {
      "epoch": 0.4574838388861263,
      "grad_norm": 1.54623281955719,
      "learning_rate": 2.1077911072367317e-05,
      "loss": 0.6044,
      "step": 3680
    },
    {
      "epoch": 0.4587270014917951,
      "grad_norm": 1.4844456911087036,
      "learning_rate": 2.077869521580325e-05,
      "loss": 0.5635,
      "step": 3690
    },
    {
      "epoch": 0.45997016409746394,
      "grad_norm": 1.464686632156372,
      "learning_rate": 2.0481194358667695e-05,
      "loss": 0.5237,
      "step": 3700
    },
    {
      "epoch": 0.46121332670313275,
      "grad_norm": 1.3379572629928589,
      "learning_rate": 2.018542073005567e-05,
      "loss": 0.5913,
      "step": 3710
    },
    {
      "epoch": 0.46245648930880157,
      "grad_norm": 1.292743444442749,
      "learning_rate": 1.9891386488062538e-05,
      "loss": 0.5878,
      "step": 3720
    },
    {
      "epoch": 0.46369965191447043,
      "grad_norm": 1.7692592144012451,
      "learning_rate": 1.959910371928436e-05,
      "loss": 0.5772,
      "step": 3730
    },
    {
      "epoch": 0.46494281452013925,
      "grad_norm": 1.5741891860961914,
      "learning_rate": 1.930858443832096e-05,
      "loss": 0.5899,
      "step": 3740
    },
    {
      "epoch": 0.46618597712580806,
      "grad_norm": 2.228027105331421,
      "learning_rate": 1.90198405872821e-05,
      "loss": 0.6182,
      "step": 3750
    },
    {
      "epoch": 0.46618597712580806,
      "eval_loss": 0.7807769775390625,
      "eval_runtime": 54.8116,
      "eval_samples_per_second": 9.122,
      "eval_steps_per_second": 9.122,
      "step": 3750
    },
    {
      "epoch": 0.4674291397314769,
      "grad_norm": 1.6490607261657715,
      "learning_rate": 1.8732884035296582e-05,
      "loss": 0.5878,
      "step": 3760
    },
    {
      "epoch": 0.4686723023371457,
      "grad_norm": 1.3176835775375366,
      "learning_rate": 1.844772657802428e-05,
      "loss": 0.6422,
      "step": 3770
    },
    {
      "epoch": 0.4699154649428145,
      "grad_norm": 0.9882899522781372,
      "learning_rate": 1.8164379937171382e-05,
      "loss": 0.5721,
      "step": 3780
    },
    {
      "epoch": 0.47115862754848337,
      "grad_norm": 1.3539491891860962,
      "learning_rate": 1.7882855760008547e-05,
      "loss": 0.5397,
      "step": 3790
    },
    {
      "epoch": 0.4724017901541522,
      "grad_norm": 1.472594141960144,
      "learning_rate": 1.760316561889203e-05,
      "loss": 0.5665,
      "step": 3800
    },
    {
      "epoch": 0.473644952759821,
      "grad_norm": 1.2908227443695068,
      "learning_rate": 1.7325321010788034e-05,
      "loss": 0.5835,
      "step": 3810
    },
    {
      "epoch": 0.4748881153654898,
      "grad_norm": 1.5640655755996704,
      "learning_rate": 1.7049333356800167e-05,
      "loss": 0.6635,
      "step": 3820
    },
    {
      "epoch": 0.4761312779711586,
      "grad_norm": 1.3457403182983398,
      "learning_rate": 1.6775214001699914e-05,
      "loss": 0.5415,
      "step": 3830
    },
    {
      "epoch": 0.47737444057682743,
      "grad_norm": 1.859492540359497,
      "learning_rate": 1.6502974213460316e-05,
      "loss": 0.5344,
      "step": 3840
    },
    {
      "epoch": 0.47861760318249624,
      "grad_norm": 1.5213356018066406,
      "learning_rate": 1.623262518279279e-05,
      "loss": 0.5805,
      "step": 3850
    },
    {
      "epoch": 0.4798607657881651,
      "grad_norm": 1.5838193893432617,
      "learning_rate": 1.596417802268707e-05,
      "loss": 0.5148,
      "step": 3860
    },
    {
      "epoch": 0.4811039283938339,
      "grad_norm": 1.2510637044906616,
      "learning_rate": 1.5697643767954488e-05,
      "loss": 0.561,
      "step": 3870
    },
    {
      "epoch": 0.48234709099950274,
      "grad_norm": 1.6863198280334473,
      "learning_rate": 1.543303337477432e-05,
      "loss": 0.5264,
      "step": 3880
    },
    {
      "epoch": 0.48359025360517155,
      "grad_norm": 1.6266530752182007,
      "learning_rate": 1.517035772024343e-05,
      "loss": 0.6313,
      "step": 3890
    },
    {
      "epoch": 0.48483341621084036,
      "grad_norm": 1.2105517387390137,
      "learning_rate": 1.49096276019291e-05,
      "loss": 0.5565,
      "step": 3900
    },
    {
      "epoch": 0.48483341621084036,
      "eval_loss": 0.7774006724357605,
      "eval_runtime": 57.9814,
      "eval_samples_per_second": 8.623,
      "eval_steps_per_second": 8.623,
      "step": 3900
    },
    {
      "epoch": 0.4860765788165092,
      "grad_norm": 1.2891868352890015,
      "learning_rate": 1.4650853737425327e-05,
      "loss": 0.5731,
      "step": 3910
    },
    {
      "epoch": 0.48731974142217804,
      "grad_norm": 0.8313778042793274,
      "learning_rate": 1.4394046763912122e-05,
      "loss": 0.6025,
      "step": 3920
    },
    {
      "epoch": 0.48856290402784686,
      "grad_norm": 1.1668204069137573,
      "learning_rate": 1.413921723771832e-05,
      "loss": 0.6046,
      "step": 3930
    },
    {
      "epoch": 0.48980606663351567,
      "grad_norm": 1.5604908466339111,
      "learning_rate": 1.3886375633887665e-05,
      "loss": 0.5574,
      "step": 3940
    },
    {
      "epoch": 0.4910492292391845,
      "grad_norm": 1.4268012046813965,
      "learning_rate": 1.3635532345748137e-05,
      "loss": 0.5466,
      "step": 3950
    },
    {
      "epoch": 0.4922923918448533,
      "grad_norm": 1.3794260025024414,
      "learning_rate": 1.3386697684484853e-05,
      "loss": 0.649,
      "step": 3960
    },
    {
      "epoch": 0.4935355544505221,
      "grad_norm": 1.2648028135299683,
      "learning_rate": 1.3139881878716107e-05,
      "loss": 0.5955,
      "step": 3970
    },
    {
      "epoch": 0.494778717056191,
      "grad_norm": 1.508892297744751,
      "learning_rate": 1.2895095074072986e-05,
      "loss": 0.5831,
      "step": 3980
    },
    {
      "epoch": 0.4960218796618598,
      "grad_norm": 1.3273192644119263,
      "learning_rate": 1.2652347332782227e-05,
      "loss": 0.5269,
      "step": 3990
    },
    {
      "epoch": 0.4972650422675286,
      "grad_norm": 1.6496832370758057,
      "learning_rate": 1.2411648633252719e-05,
      "loss": 0.59,
      "step": 4000
    },
    {
      "epoch": 0.4985082048731974,
      "grad_norm": 1.8715214729309082,
      "learning_rate": 1.2173008869665241e-05,
      "loss": 0.6478,
      "step": 4010
    },
    {
      "epoch": 0.4997513674788662,
      "grad_norm": 1.4751086235046387,
      "learning_rate": 1.1936437851565791e-05,
      "loss": 0.5649,
      "step": 4020
    },
    {
      "epoch": 0.5009945300845351,
      "grad_norm": 1.407470464706421,
      "learning_rate": 1.1701945303462337e-05,
      "loss": 0.6002,
      "step": 4030
    },
    {
      "epoch": 0.5022376926902039,
      "grad_norm": 1.1696547269821167,
      "learning_rate": 1.146954086442508e-05,
      "loss": 0.5095,
      "step": 4040
    },
    {
      "epoch": 0.5034808552958727,
      "grad_norm": 2.332303285598755,
      "learning_rate": 1.1239234087690252e-05,
      "loss": 0.5586,
      "step": 4050
    },
    {
      "epoch": 0.5034808552958727,
      "eval_loss": 0.7748836278915405,
      "eval_runtime": 54.7049,
      "eval_samples_per_second": 9.14,
      "eval_steps_per_second": 9.14,
      "step": 4050
    },
    {
      "epoch": 0.5047240179015415,
      "grad_norm": 1.2329920530319214,
      "learning_rate": 1.1011034440267395e-05,
      "loss": 0.4973,
      "step": 4060
    },
    {
      "epoch": 0.5059671805072103,
      "grad_norm": 1.2258820533752441,
      "learning_rate": 1.078495130255023e-05,
      "loss": 0.5507,
      "step": 4070
    },
    {
      "epoch": 0.5072103431128792,
      "grad_norm": 1.7650001049041748,
      "learning_rate": 1.0560993967931004e-05,
      "loss": 0.6073,
      "step": 4080
    },
    {
      "epoch": 0.508453505718548,
      "grad_norm": 1.2635862827301025,
      "learning_rate": 1.0339171642418585e-05,
      "loss": 0.5028,
      "step": 4090
    },
    {
      "epoch": 0.5096966683242168,
      "grad_norm": 1.1583187580108643,
      "learning_rate": 1.0119493444259963e-05,
      "loss": 0.5461,
      "step": 4100
    },
    {
      "epoch": 0.5109398309298856,
      "grad_norm": 1.1466636657714844,
      "learning_rate": 9.901968403565428e-06,
      "loss": 0.5874,
      "step": 4110
    },
    {
      "epoch": 0.5121829935355544,
      "grad_norm": 1.3626190423965454,
      "learning_rate": 9.686605461937441e-06,
      "loss": 0.5841,
      "step": 4120
    },
    {
      "epoch": 0.5134261561412232,
      "grad_norm": 1.1523593664169312,
      "learning_rate": 9.473413472102982e-06,
      "loss": 0.5545,
      "step": 4130
    },
    {
      "epoch": 0.5146693187468921,
      "grad_norm": 4.49442195892334,
      "learning_rate": 9.262401197549744e-06,
      "loss": 0.6379,
      "step": 4140
    },
    {
      "epoch": 0.515912481352561,
      "grad_norm": 1.3444253206253052,
      "learning_rate": 9.05357731216587e-06,
      "loss": 0.5708,
      "step": 4150
    },
    {
      "epoch": 0.5171556439582298,
      "grad_norm": 1.5777453184127808,
      "learning_rate": 8.846950399883368e-06,
      "loss": 0.5973,
      "step": 4160
    },
    {
      "epoch": 0.5183988065638986,
      "grad_norm": 1.251243233680725,
      "learning_rate": 8.64252895432531e-06,
      "loss": 0.5389,
      "step": 4170
    },
    {
      "epoch": 0.5196419691695674,
      "grad_norm": 2.3069584369659424,
      "learning_rate": 8.440321378456656e-06,
      "loss": 0.5431,
      "step": 4180
    },
    {
      "epoch": 0.5208851317752362,
      "grad_norm": 1.4895597696304321,
      "learning_rate": 8.240335984238844e-06,
      "loss": 0.5687,
      "step": 4190
    },
    {
      "epoch": 0.522128294380905,
      "grad_norm": 1.2962337732315063,
      "learning_rate": 8.042580992288163e-06,
      "loss": 0.5778,
      "step": 4200
    },
    {
      "epoch": 0.522128294380905,
      "eval_loss": 0.7723983526229858,
      "eval_runtime": 56.0371,
      "eval_samples_per_second": 8.923,
      "eval_steps_per_second": 8.923,
      "step": 4200
    },
    {
      "epoch": 0.5233714569865738,
      "grad_norm": 1.3925156593322754,
      "learning_rate": 7.847064531537774e-06,
      "loss": 0.5649,
      "step": 4210
    },
    {
      "epoch": 0.5246146195922426,
      "grad_norm": 1.3647059202194214,
      "learning_rate": 7.653794638903574e-06,
      "loss": 0.5655,
      "step": 4220
    },
    {
      "epoch": 0.5258577821979115,
      "grad_norm": 2.4528896808624268,
      "learning_rate": 7.462779258953875e-06,
      "loss": 0.528,
      "step": 4230
    },
    {
      "epoch": 0.5271009448035803,
      "grad_norm": 0.9853927493095398,
      "learning_rate": 7.274026243582796e-06,
      "loss": 0.4958,
      "step": 4240
    },
    {
      "epoch": 0.5283441074092491,
      "grad_norm": 1.48464035987854,
      "learning_rate": 7.087543351687493e-06,
      "loss": 0.4987,
      "step": 4250
    },
    {
      "epoch": 0.5295872700149179,
      "grad_norm": 1.827588677406311,
      "learning_rate": 6.903338248849269e-06,
      "loss": 0.6239,
      "step": 4260
    },
    {
      "epoch": 0.5308304326205868,
      "grad_norm": 1.2086472511291504,
      "learning_rate": 6.7214185070183925e-06,
      "loss": 0.5193,
      "step": 4270
    },
    {
      "epoch": 0.5320735952262556,
      "grad_norm": 1.1530331373214722,
      "learning_rate": 6.541791604202936e-06,
      "loss": 0.6089,
      "step": 4280
    },
    {
      "epoch": 0.5333167578319244,
      "grad_norm": 1.198806881904602,
      "learning_rate": 6.364464924161311e-06,
      "loss": 0.5195,
      "step": 4290
    },
    {
      "epoch": 0.5345599204375933,
      "grad_norm": 1.753954291343689,
      "learning_rate": 6.1894457560988106e-06,
      "loss": 0.624,
      "step": 4300
    },
    {
      "epoch": 0.5358030830432621,
      "grad_norm": 1.1597492694854736,
      "learning_rate": 6.016741294367911e-06,
      "loss": 0.479,
      "step": 4310
    },
    {
      "epoch": 0.5370462456489309,
      "grad_norm": 1.3531755208969116,
      "learning_rate": 5.846358638172615e-06,
      "loss": 0.5792,
      "step": 4320
    },
    {
      "epoch": 0.5382894082545997,
      "grad_norm": 1.2098314762115479,
      "learning_rate": 5.678304791276567e-06,
      "loss": 0.5247,
      "step": 4330
    },
    {
      "epoch": 0.5395325708602685,
      "grad_norm": 1.73563551902771,
      "learning_rate": 5.51258666171519e-06,
      "loss": 0.4989,
      "step": 4340
    },
    {
      "epoch": 0.5407757334659373,
      "grad_norm": 1.3523006439208984,
      "learning_rate": 5.349211061511726e-06,
      "loss": 0.6063,
      "step": 4350
    },
    {
      "epoch": 0.5407757334659373,
      "eval_loss": 0.7707881927490234,
      "eval_runtime": 57.1975,
      "eval_samples_per_second": 8.742,
      "eval_steps_per_second": 8.742,
      "step": 4350
    },
    {
      "epoch": 0.5420188960716061,
      "grad_norm": 1.6083730459213257,
      "learning_rate": 5.188184706397182e-06,
      "loss": 0.6078,
      "step": 4360
    },
    {
      "epoch": 0.543262058677275,
      "grad_norm": 1.7973166704177856,
      "learning_rate": 5.029514215534339e-06,
      "loss": 0.5657,
      "step": 4370
    },
    {
      "epoch": 0.5445052212829438,
      "grad_norm": 1.397307276725769,
      "learning_rate": 4.873206111245594e-06,
      "loss": 0.5861,
      "step": 4380
    },
    {
      "epoch": 0.5457483838886126,
      "grad_norm": 1.766788363456726,
      "learning_rate": 4.719266818744912e-06,
      "loss": 0.5335,
      "step": 4390
    },
    {
      "epoch": 0.5469915464942815,
      "grad_norm": 2.368110418319702,
      "learning_rate": 4.567702665873648e-06,
      "loss": 0.6134,
      "step": 4400
    },
    {
      "epoch": 0.5482347090999503,
      "grad_norm": 1.4024748802185059,
      "learning_rate": 4.418519882840505e-06,
      "loss": 0.5903,
      "step": 4410
    },
    {
      "epoch": 0.5494778717056191,
      "grad_norm": 1.45235013961792,
      "learning_rate": 4.271724601965371e-06,
      "loss": 0.6008,
      "step": 4420
    },
    {
      "epoch": 0.5507210343112879,
      "grad_norm": 1.5979630947113037,
      "learning_rate": 4.127322857427306e-06,
      "loss": 0.5718,
      "step": 4430
    },
    {
      "epoch": 0.5519641969169568,
      "grad_norm": 1.427748203277588,
      "learning_rate": 3.985320585016425e-06,
      "loss": 0.5916,
      "step": 4440
    },
    {
      "epoch": 0.5532073595226256,
      "grad_norm": 1.756362795829773,
      "learning_rate": 3.845723621889973e-06,
      "loss": 0.5969,
      "step": 4450
    },
    {
      "epoch": 0.5544505221282944,
      "grad_norm": 1.447805404663086,
      "learning_rate": 3.7085377063323447e-06,
      "loss": 0.5322,
      "step": 4460
    },
    {
      "epoch": 0.5556936847339632,
      "grad_norm": 1.3792946338653564,
      "learning_rate": 3.5737684775191887e-06,
      "loss": 0.5492,
      "step": 4470
    },
    {
      "epoch": 0.556936847339632,
      "grad_norm": 1.699859619140625,
      "learning_rate": 3.441421475285679e-06,
      "loss": 0.5606,
      "step": 4480
    },
    {
      "epoch": 0.5581800099453008,
      "grad_norm": 1.4851022958755493,
      "learning_rate": 3.3115021398986768e-06,
      "loss": 0.6094,
      "step": 4490
    },
    {
      "epoch": 0.5594231725509696,
      "grad_norm": 0.9180851578712463,
      "learning_rate": 3.18401581183321e-06,
      "loss": 0.5229,
      "step": 4500
    },
    {
      "epoch": 0.5594231725509696,
      "eval_loss": 0.7698713541030884,
      "eval_runtime": 55.7444,
      "eval_samples_per_second": 8.97,
      "eval_steps_per_second": 8.97,
      "step": 4500
    },
    {
      "epoch": 0.5606663351566384,
      "grad_norm": 1.226762056350708,
      "learning_rate": 3.0589677315529044e-06,
      "loss": 0.5444,
      "step": 4510
    },
    {
      "epoch": 0.5619094977623074,
      "grad_norm": 1.1591328382492065,
      "learning_rate": 2.9363630392945513e-06,
      "loss": 0.6316,
      "step": 4520
    },
    {
      "epoch": 0.5631526603679762,
      "grad_norm": 1.0757044553756714,
      "learning_rate": 2.816206774856854e-06,
      "loss": 0.4764,
      "step": 4530
    },
    {
      "epoch": 0.564395822973645,
      "grad_norm": 1.269648551940918,
      "learning_rate": 2.6985038773932046e-06,
      "loss": 0.5399,
      "step": 4540
    },
    {
      "epoch": 0.5656389855793138,
      "grad_norm": 1.5957112312316895,
      "learning_rate": 2.583259185208714e-06,
      "loss": 0.5419,
      "step": 4550
    },
    {
      "epoch": 0.5668821481849826,
      "grad_norm": 1.3144357204437256,
      "learning_rate": 2.4704774355612943e-06,
      "loss": 0.527,
      "step": 4560
    },
    {
      "epoch": 0.5681253107906514,
      "grad_norm": 1.456000804901123,
      "learning_rate": 2.3601632644669536e-06,
      "loss": 0.5787,
      "step": 4570
    },
    {
      "epoch": 0.5693684733963202,
      "grad_norm": 1.3255828619003296,
      "learning_rate": 2.2523212065091723e-06,
      "loss": 0.5483,
      "step": 4580
    },
    {
      "epoch": 0.570611636001989,
      "grad_norm": 1.271060585975647,
      "learning_rate": 2.1469556946525706e-06,
      "loss": 0.5318,
      "step": 4590
    },
    {
      "epoch": 0.5718547986076579,
      "grad_norm": 1.491672396659851,
      "learning_rate": 2.0440710600606595e-06,
      "loss": 0.5108,
      "step": 4600
    },
    {
      "epoch": 0.5730979612133267,
      "grad_norm": 1.744455337524414,
      "learning_rate": 1.9436715319177956e-06,
      "loss": 0.6084,
      "step": 4610
    },
    {
      "epoch": 0.5743411238189955,
      "grad_norm": 1.216773271560669,
      "learning_rate": 1.8457612372553348e-06,
      "loss": 0.5647,
      "step": 4620
    },
    {
      "epoch": 0.5755842864246643,
      "grad_norm": 1.0653576850891113,
      "learning_rate": 1.75034420078201e-06,
      "loss": 0.5191,
      "step": 4630
    },
    {
      "epoch": 0.5768274490303331,
      "grad_norm": 1.4697630405426025,
      "learning_rate": 1.6574243447184597e-06,
      "loss": 0.5554,
      "step": 4640
    },
    {
      "epoch": 0.578070611636002,
      "grad_norm": 1.4933724403381348,
      "learning_rate": 1.567005488636024e-06,
      "loss": 0.5825,
      "step": 4650
    },
    {
      "epoch": 0.578070611636002,
      "eval_loss": 0.7693516612052917,
      "eval_runtime": 55.3708,
      "eval_samples_per_second": 9.03,
      "eval_steps_per_second": 9.03,
      "step": 4650
    },
    {
      "epoch": 0.5793137742416709,
      "grad_norm": 1.5786610841751099,
      "learning_rate": 1.4790913492997438e-06,
      "loss": 0.5274,
      "step": 4660
    },
    {
      "epoch": 0.5805569368473397,
      "grad_norm": 1.6399073600769043,
      "learning_rate": 1.3936855405155408e-06,
      "loss": 0.6191,
      "step": 4670
    },
    {
      "epoch": 0.5818000994530085,
      "grad_norm": 2.071617841720581,
      "learning_rate": 1.3107915729816954e-06,
      "loss": 0.4883,
      "step": 4680
    },
    {
      "epoch": 0.5830432620586773,
      "grad_norm": 1.687066912651062,
      "learning_rate": 1.230412854144547e-06,
      "loss": 0.5191,
      "step": 4690
    },
    {
      "epoch": 0.5842864246643461,
      "grad_norm": 1.2871911525726318,
      "learning_rate": 1.15255268805841e-06,
      "loss": 0.5855,
      "step": 4700
    },
    {
      "epoch": 0.5855295872700149,
      "grad_norm": 1.2647926807403564,
      "learning_rate": 1.0772142752497604e-06,
      "loss": 0.5283,
      "step": 4710
    },
    {
      "epoch": 0.5867727498756837,
      "grad_norm": 1.6415475606918335,
      "learning_rate": 1.004400712585646e-06,
      "loss": 0.6313,
      "step": 4720
    },
    {
      "epoch": 0.5880159124813525,
      "grad_norm": 1.8416297435760498,
      "learning_rate": 9.341149931464537e-07,
      "loss": 0.5707,
      "step": 4730
    },
    {
      "epoch": 0.5892590750870214,
      "grad_norm": 1.3439534902572632,
      "learning_rate": 8.663600061028162e-07,
      "loss": 0.5725,
      "step": 4740
    },
    {
      "epoch": 0.5905022376926902,
      "grad_norm": 1.181318759918213,
      "learning_rate": 8.011385365968641e-07,
      "loss": 0.537,
      "step": 4750
    },
    {
      "epoch": 0.591745400298359,
      "grad_norm": 1.5585083961486816,
      "learning_rate": 7.384532656277698e-07,
      "loss": 0.5584,
      "step": 4760
    },
    {
      "epoch": 0.5929885629040278,
      "grad_norm": 1.2982670068740845,
      "learning_rate": 6.783067699414891e-07,
      "loss": 0.5202,
      "step": 4770
    },
    {
      "epoch": 0.5942317255096967,
      "grad_norm": 1.5101556777954102,
      "learning_rate": 6.207015219248866e-07,
      "loss": 0.5231,
      "step": 4780
    },
    {
      "epoch": 0.5954748881153655,
      "grad_norm": 1.1590073108673096,
      "learning_rate": 5.656398895040813e-07,
      "loss": 0.5544,
      "step": 4790
    },
    {
      "epoch": 0.5967180507210343,
      "grad_norm": 1.7406432628631592,
      "learning_rate": 5.131241360471217e-07,
      "loss": 0.4726,
      "step": 4800
    },
    {
      "epoch": 0.5967180507210343,
      "eval_loss": 0.7685801386833191,
      "eval_runtime": 56.8319,
      "eval_samples_per_second": 8.798,
      "eval_steps_per_second": 8.798,
      "step": 4800
    },
    {
      "epoch": 0.5979612133267032,
      "grad_norm": 1.2891184091567993,
      "learning_rate": 4.631564202709354e-07,
      "loss": 0.5094,
      "step": 4810
    },
    {
      "epoch": 0.599204375932372,
      "grad_norm": 1.6769993305206299,
      "learning_rate": 4.1573879615262185e-07,
      "loss": 0.5776,
      "step": 4820
    },
    {
      "epoch": 0.6004475385380408,
      "grad_norm": 1.0175931453704834,
      "learning_rate": 3.708732128449785e-07,
      "loss": 0.5417,
      "step": 4830
    },
    {
      "epoch": 0.6016907011437096,
      "grad_norm": 1.0494272708892822,
      "learning_rate": 3.2856151459641216e-07,
      "loss": 0.5291,
      "step": 4840
    },
    {
      "epoch": 0.6029338637493784,
      "grad_norm": 0.9337966442108154,
      "learning_rate": 2.888054406751106e-07,
      "loss": 0.4692,
      "step": 4850
    },
    {
      "epoch": 0.6041770263550472,
      "grad_norm": 2.697531223297119,
      "learning_rate": 2.5160662529755823e-07,
      "loss": 0.588,
      "step": 4860
    },
    {
      "epoch": 0.605420188960716,
      "grad_norm": 1.2500678300857544,
      "learning_rate": 2.169665975613605e-07,
      "loss": 0.5474,
      "step": 4870
    },
    {
      "epoch": 0.6066633515663848,
      "grad_norm": 1.5010889768600464,
      "learning_rate": 1.8488678138238456e-07,
      "loss": 0.5865,
      "step": 4880
    },
    {
      "epoch": 0.6079065141720537,
      "grad_norm": 1.6108850240707397,
      "learning_rate": 1.5536849543621584e-07,
      "loss": 0.583,
      "step": 4890
    },
    {
      "epoch": 0.6091496767777225,
      "grad_norm": 1.4152370691299438,
      "learning_rate": 1.2841295310397905e-07,
      "loss": 0.5705,
      "step": 4900
    },
    {
      "epoch": 0.6103928393833914,
      "grad_norm": 1.315041184425354,
      "learning_rate": 1.0402126242244764e-07,
      "loss": 0.566,
      "step": 4910
    },
    {
      "epoch": 0.6116360019890602,
      "grad_norm": 1.2925786972045898,
      "learning_rate": 8.219442603847605e-08,
      "loss": 0.5201,
      "step": 4920
    },
    {
      "epoch": 0.612879164594729,
      "grad_norm": 1.1554069519042969,
      "learning_rate": 6.293334116783817e-08,
      "loss": 0.6017,
      "step": 4930
    },
    {
      "epoch": 0.6141223272003978,
      "grad_norm": 1.682599663734436,
      "learning_rate": 4.623879955827082e-08,
      "loss": 0.5932,
      "step": 4940
    },
    {
      "epoch": 0.6153654898060666,
      "grad_norm": 1.2782875299453735,
      "learning_rate": 3.211148745700665e-08,
      "loss": 0.5791,
      "step": 4950
    },
    {
      "epoch": 0.6153654898060666,
      "eval_loss": 0.768585741519928,
      "eval_runtime": 55.9999,
      "eval_samples_per_second": 8.929,
      "eval_steps_per_second": 8.929,
      "step": 4950
    }
  ],
  "logging_steps": 10,
  "max_steps": 5000,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 150,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 1.724336235597312e+17,
  "train_batch_size": 16,
  "trial_name": null,
  "trial_params": null
}