{
  "best_metric": 1.4804431200027466,
  "best_model_checkpoint": "./output/checkpoint-4650",
  "epoch": 0.7796503386360056,
  "eval_steps": 150,
  "global_step": 4950,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.001575051189163648,
      "grad_norm": 1.292704463005066,
      "learning_rate": 1.25e-05,
      "loss": 2.3495,
      "step": 10
    },
    {
      "epoch": 0.003150102378327296,
      "grad_norm": 1.1147561073303223,
      "learning_rate": 2.5e-05,
      "loss": 2.2929,
      "step": 20
    },
    {
      "epoch": 0.004725153567490943,
      "grad_norm": 1.0415704250335693,
      "learning_rate": 3.75e-05,
      "loss": 2.2227,
      "step": 30
    },
    {
      "epoch": 0.006300204756654592,
      "grad_norm": 1.171570062637329,
      "learning_rate": 5e-05,
      "loss": 2.1635,
      "step": 40
    },
    {
      "epoch": 0.007875255945818239,
      "grad_norm": 1.2755579948425293,
      "learning_rate": 6.25e-05,
      "loss": 2.0523,
      "step": 50
    },
    {
      "epoch": 0.009450307134981887,
      "grad_norm": 1.2706725597381592,
      "learning_rate": 7.5e-05,
      "loss": 1.9675,
      "step": 60
    },
    {
      "epoch": 0.011025358324145534,
      "grad_norm": 1.2686223983764648,
      "learning_rate": 8.75e-05,
      "loss": 1.9002,
      "step": 70
    },
    {
      "epoch": 0.012600409513309183,
      "grad_norm": 1.3380385637283325,
      "learning_rate": 0.0001,
      "loss": 1.8681,
      "step": 80
    },
    {
      "epoch": 0.01417546070247283,
      "grad_norm": 1.2959766387939453,
      "learning_rate": 0.00011250000000000001,
      "loss": 1.8263,
      "step": 90
    },
    {
      "epoch": 0.015750511891636478,
      "grad_norm": 1.4983580112457275,
      "learning_rate": 0.000125,
      "loss": 1.7796,
      "step": 100
    },
    {
      "epoch": 0.017325563080800126,
      "grad_norm": 1.2883356809616089,
      "learning_rate": 0.00012499871543489787,
      "loss": 1.7754,
      "step": 110
    },
    {
      "epoch": 0.018900614269963773,
      "grad_norm": 1.2248905897140503,
      "learning_rate": 0.00012499486179239495,
      "loss": 1.7442,
      "step": 120
    },
    {
      "epoch": 0.02047566545912742,
      "grad_norm": 1.290252923965454,
      "learning_rate": 0.00012498843923089938,
      "loss": 1.7451,
      "step": 130
    },
    {
      "epoch": 0.022050716648291068,
      "grad_norm": 1.246947169303894,
      "learning_rate": 0.0001249794480144175,
      "loss": 1.7151,
      "step": 140
    },
    {
      "epoch": 0.023625767837454716,
      "grad_norm": 1.2474550008773804,
      "learning_rate": 0.000124967888512543,
      "loss": 1.6997,
      "step": 150
    },
    {
      "epoch": 0.023625767837454716,
      "eval_loss": 1.713956594467163,
      "eval_runtime": 35.1971,
      "eval_samples_per_second": 14.234,
      "eval_steps_per_second": 14.234,
      "step": 150
    },
    {
      "epoch": 0.025200819026618367,
      "grad_norm": 1.2170847654342651,
      "learning_rate": 0.00012495376120044173,
      "loss": 1.6881,
      "step": 160
    },
    {
      "epoch": 0.026775870215782014,
      "grad_norm": 1.2175121307373047,
      "learning_rate": 0.00012493706665883217,
      "loss": 1.7135,
      "step": 170
    },
    {
      "epoch": 0.02835092140494566,
      "grad_norm": 1.1301295757293701,
      "learning_rate": 0.00012491780557396154,
      "loss": 1.682,
      "step": 180
    },
    {
      "epoch": 0.02992597259410931,
      "grad_norm": 1.1549564599990845,
      "learning_rate": 0.00012489597873757756,
      "loss": 1.6887,
      "step": 190
    },
    {
      "epoch": 0.031501023783272956,
      "grad_norm": 1.1598905324935913,
      "learning_rate": 0.00012487158704689602,
      "loss": 1.6954,
      "step": 200
    },
    {
      "epoch": 0.03307607497243661,
      "grad_norm": 1.1747504472732544,
      "learning_rate": 0.0001248446315045638,
      "loss": 1.6835,
      "step": 210
    },
    {
      "epoch": 0.03465112616160025,
      "grad_norm": 1.1416676044464111,
      "learning_rate": 0.00012481511321861763,
      "loss": 1.6521,
      "step": 220
    },
    {
      "epoch": 0.0362261773507639,
      "grad_norm": 1.1543567180633545,
      "learning_rate": 0.00012478303340243864,
      "loss": 1.662,
      "step": 230
    },
    {
      "epoch": 0.037801228539927546,
      "grad_norm": 1.2603256702423096,
      "learning_rate": 0.00012474839337470246,
      "loss": 1.6678,
      "step": 240
    },
    {
      "epoch": 0.0393762797290912,
      "grad_norm": 1.2086098194122314,
      "learning_rate": 0.0001247111945593249,
      "loss": 1.6475,
      "step": 250
    },
    {
      "epoch": 0.04095133091825484,
      "grad_norm": 1.2282692193984985,
      "learning_rate": 0.00012467143848540359,
      "loss": 1.6491,
      "step": 260
    },
    {
      "epoch": 0.04252638210741849,
      "grad_norm": 1.1063305139541626,
      "learning_rate": 0.000124629126787155,
      "loss": 1.654,
      "step": 270
    },
    {
      "epoch": 0.044101433296582136,
      "grad_norm": 1.114211082458496,
      "learning_rate": 0.00012458426120384738,
      "loss": 1.6244,
      "step": 280
    },
    {
      "epoch": 0.04567648448574579,
      "grad_norm": 1.1069458723068237,
      "learning_rate": 0.00012453684357972906,
      "loss": 1.6388,
      "step": 290
    },
    {
      "epoch": 0.04725153567490943,
      "grad_norm": 1.0202927589416504,
      "learning_rate": 0.00012448687586395289,
      "loss": 1.6367,
      "step": 300
    },
    {
      "epoch": 0.04725153567490943,
      "eval_loss": 1.6362468004226685,
      "eval_runtime": 35.7423,
      "eval_samples_per_second": 14.017,
      "eval_steps_per_second": 14.017,
      "step": 300
    },
    {
      "epoch": 0.04882658686407308,
      "grad_norm": 1.0611218214035034,
      "learning_rate": 0.00012443436011049593,
      "loss": 1.6127,
      "step": 310
    },
    {
      "epoch": 0.05040163805323673,
      "grad_norm": 1.1300408840179443,
      "learning_rate": 0.0001243792984780751,
      "loss": 1.6293,
      "step": 320
    },
    {
      "epoch": 0.05197668924240038,
      "grad_norm": 1.0638223886489868,
      "learning_rate": 0.00012432169323005853,
      "loss": 1.6267,
      "step": 330
    },
    {
      "epoch": 0.05355174043156403,
      "grad_norm": 1.1265466213226318,
      "learning_rate": 0.00012426154673437223,
      "loss": 1.6315,
      "step": 340
    },
    {
      "epoch": 0.05512679162072767,
      "grad_norm": 1.088779330253601,
      "learning_rate": 0.00012419886146340314,
      "loss": 1.6207,
      "step": 350
    },
    {
      "epoch": 0.05670184280989132,
      "grad_norm": 1.0430299043655396,
      "learning_rate": 0.0001241336399938972,
      "loss": 1.6132,
      "step": 360
    },
    {
      "epoch": 0.05827689399905497,
      "grad_norm": 1.0889984369277954,
      "learning_rate": 0.00012406588500685355,
      "loss": 1.6416,
      "step": 370
    },
    {
      "epoch": 0.05985194518821862,
      "grad_norm": 1.0341376066207886,
      "learning_rate": 0.00012399559928741435,
      "loss": 1.6217,
      "step": 380
    },
    {
      "epoch": 0.06142699637738226,
      "grad_norm": 1.073559284210205,
      "learning_rate": 0.00012392278572475023,
      "loss": 1.6314,
      "step": 390
    },
    {
      "epoch": 0.06300204756654591,
      "grad_norm": 1.039914846420288,
      "learning_rate": 0.0001238474473119416,
      "loss": 1.6149,
      "step": 400
    },
    {
      "epoch": 0.06457709875570956,
      "grad_norm": 1.0189768075942993,
      "learning_rate": 0.00012376958714585545,
      "loss": 1.5755,
      "step": 410
    },
    {
      "epoch": 0.06615214994487321,
      "grad_norm": 1.0381276607513428,
      "learning_rate": 0.0001236892084270183,
      "loss": 1.6057,
      "step": 420
    },
    {
      "epoch": 0.06772720113403685,
      "grad_norm": 1.089646339416504,
      "learning_rate": 0.00012360631445948448,
      "loss": 1.6127,
      "step": 430
    },
    {
      "epoch": 0.0693022523232005,
      "grad_norm": 1.0117536783218384,
      "learning_rate": 0.00012352090865070026,
      "loss": 1.6043,
      "step": 440
    },
    {
      "epoch": 0.07087730351236415,
      "grad_norm": 1.0716177225112915,
      "learning_rate": 0.00012343299451136397,
      "loss": 1.5994,
      "step": 450
    },
    {
      "epoch": 0.07087730351236415,
      "eval_loss": 1.6039879322052002,
      "eval_runtime": 35.4687,
      "eval_samples_per_second": 14.125,
      "eval_steps_per_second": 14.125,
      "step": 450
    },
    {
      "epoch": 0.0724523547015278,
      "grad_norm": 1.0424439907073975,
      "learning_rate": 0.00012334257565528155,
      "loss": 1.6081,
      "step": 460
    },
    {
      "epoch": 0.07402740589069144,
      "grad_norm": 1.020411491394043,
      "learning_rate": 0.000123249655799218,
      "loss": 1.5901,
      "step": 470
    },
    {
      "epoch": 0.07560245707985509,
      "grad_norm": 1.0402997732162476,
      "learning_rate": 0.00012315423876274468,
      "loss": 1.605,
      "step": 480
    },
    {
      "epoch": 0.07717750826901874,
      "grad_norm": 1.0569404363632202,
      "learning_rate": 0.0001230563284680822,
      "loss": 1.6015,
      "step": 490
    },
    {
      "epoch": 0.0787525594581824,
      "grad_norm": 1.0378941297531128,
      "learning_rate": 0.00012295592893993935,
      "loss": 1.5875,
      "step": 500
    },
    {
      "epoch": 0.08032761064734605,
      "grad_norm": 0.9868465065956116,
      "learning_rate": 0.00012285304430534745,
      "loss": 1.5916,
      "step": 510
    },
    {
      "epoch": 0.08190266183650968,
      "grad_norm": 0.9732680916786194,
      "learning_rate": 0.00012274767879349083,
      "loss": 1.6066,
      "step": 520
    },
    {
      "epoch": 0.08347771302567333,
      "grad_norm": 0.9945887923240662,
      "learning_rate": 0.00012263983673553306,
      "loss": 1.56,
      "step": 530
    },
    {
      "epoch": 0.08505276421483698,
      "grad_norm": 0.9910768270492554,
      "learning_rate": 0.0001225295225644387,
      "loss": 1.5889,
      "step": 540
    },
    {
      "epoch": 0.08662781540400064,
      "grad_norm": 1.0420514345169067,
      "learning_rate": 0.0001224167408147913,
      "loss": 1.5801,
      "step": 550
    },
    {
      "epoch": 0.08820286659316427,
      "grad_norm": 1.0232083797454834,
      "learning_rate": 0.0001223014961226068,
      "loss": 1.5764,
      "step": 560
    },
    {
      "epoch": 0.08977791778232792,
      "grad_norm": 0.9651984572410583,
      "learning_rate": 0.00012218379322514317,
      "loss": 1.5833,
      "step": 570
    },
    {
      "epoch": 0.09135296897149157,
      "grad_norm": 0.9818013310432434,
      "learning_rate": 0.00012206363696070545,
      "loss": 1.5998,
      "step": 580
    },
    {
      "epoch": 0.09292802016065523,
      "grad_norm": 1.0015103816986084,
      "learning_rate": 0.0001219410322684471,
      "loss": 1.5818,
      "step": 590
    },
    {
      "epoch": 0.09450307134981886,
      "grad_norm": 1.0065380334854126,
      "learning_rate": 0.0001218159841881668,
      "loss": 1.5851,
      "step": 600
    },
    {
      "epoch": 0.09450307134981886,
      "eval_loss": 1.583275318145752,
      "eval_runtime": 35.3825,
      "eval_samples_per_second": 14.16,
      "eval_steps_per_second": 14.16,
      "step": 600
    },
    {
      "epoch": 0.09607812253898251,
      "grad_norm": 1.0160647630691528,
      "learning_rate": 0.00012168849786010133,
      "loss": 1.5837,
      "step": 610
    },
    {
      "epoch": 0.09765317372814616,
      "grad_norm": 0.9476320147514343,
      "learning_rate": 0.00012155857852471433,
      "loss": 1.6115,
      "step": 620
    },
    {
      "epoch": 0.09922822491730982,
      "grad_norm": 0.9994568228721619,
      "learning_rate": 0.0001214262315224808,
      "loss": 1.5922,
      "step": 630
    },
    {
      "epoch": 0.10080327610647347,
      "grad_norm": 0.9553513526916504,
      "learning_rate": 0.00012129146229366766,
      "loss": 1.5811,
      "step": 640
    },
    {
      "epoch": 0.1023783272956371,
      "grad_norm": 1.055202603340149,
      "learning_rate": 0.00012115427637811003,
      "loss": 1.5672,
      "step": 650
    },
    {
      "epoch": 0.10395337848480075,
      "grad_norm": 1.0420925617218018,
      "learning_rate": 0.00012101467941498357,
      "loss": 1.5647,
      "step": 660
    },
    {
      "epoch": 0.1055284296739644,
      "grad_norm": 0.9376361966133118,
      "learning_rate": 0.0001208726771425727,
      "loss": 1.5771,
      "step": 670
    },
    {
      "epoch": 0.10710348086312806,
      "grad_norm": 0.9151705503463745,
      "learning_rate": 0.00012072827539803463,
      "loss": 1.5432,
      "step": 680
    },
    {
      "epoch": 0.1086785320522917,
      "grad_norm": 1.0111474990844727,
      "learning_rate": 0.00012058148011715949,
      "loss": 1.5674,
      "step": 690
    },
    {
      "epoch": 0.11025358324145534,
      "grad_norm": 0.8788293600082397,
      "learning_rate": 0.00012043229733412636,
      "loss": 1.5723,
      "step": 700
    },
    {
      "epoch": 0.111828634430619,
      "grad_norm": 0.9642289876937866,
      "learning_rate": 0.0001202807331812551,
      "loss": 1.5819,
      "step": 710
    },
    {
      "epoch": 0.11340368561978265,
      "grad_norm": 1.0280308723449707,
      "learning_rate": 0.00012012679388875441,
      "loss": 1.565,
      "step": 720
    },
    {
      "epoch": 0.1149787368089463,
      "grad_norm": 1.0084924697875977,
      "learning_rate": 0.00011997048578446568,
      "loss": 1.5608,
      "step": 730
    },
    {
      "epoch": 0.11655378799810993,
      "grad_norm": 0.9293835163116455,
      "learning_rate": 0.00011981181529360282,
      "loss": 1.568,
      "step": 740
    },
    {
      "epoch": 0.11812883918727358,
      "grad_norm": 0.9751657247543335,
      "learning_rate": 0.00011965078893848828,
      "loss": 1.5488,
      "step": 750
    },
    {
      "epoch": 0.11812883918727358,
      "eval_loss": 1.567505121231079,
      "eval_runtime": 35.4792,
      "eval_samples_per_second": 14.121,
      "eval_steps_per_second": 14.121,
      "step": 750
    },
    {
      "epoch": 0.11970389037643724,
      "grad_norm": 1.0379984378814697,
      "learning_rate": 0.00011948741333828481,
      "loss": 1.5631,
      "step": 760
    },
    {
      "epoch": 0.12127894156560089,
      "grad_norm": 0.9595450758934021,
      "learning_rate": 0.00011932169520872344,
      "loss": 1.5837,
      "step": 770
    },
    {
      "epoch": 0.12285399275476452,
      "grad_norm": 0.9366925954818726,
      "learning_rate": 0.00011915364136182738,
      "loss": 1.5587,
      "step": 780
    },
    {
      "epoch": 0.12442904394392817,
      "grad_norm": 0.943828821182251,
      "learning_rate": 0.0001189832587056321,
      "loss": 1.534,
      "step": 790
    },
    {
      "epoch": 0.12600409513309183,
      "grad_norm": 0.9779319763183594,
      "learning_rate": 0.00011881055424390119,
      "loss": 1.5814,
      "step": 800
    },
    {
      "epoch": 0.12757914632225548,
      "grad_norm": 0.9054823517799377,
      "learning_rate": 0.00011863553507583869,
      "loss": 1.5527,
      "step": 810
    },
    {
      "epoch": 0.12915419751141913,
      "grad_norm": 0.8591505885124207,
      "learning_rate": 0.00011845820839579708,
      "loss": 1.5332,
      "step": 820
    },
    {
      "epoch": 0.13072924870058278,
      "grad_norm": 0.9473647475242615,
      "learning_rate": 0.00011827858149298162,
      "loss": 1.5583,
      "step": 830
    },
    {
      "epoch": 0.13230429988974643,
      "grad_norm": 0.9654921889305115,
      "learning_rate": 0.00011809666175115075,
      "loss": 1.5464,
      "step": 840
    },
    {
      "epoch": 0.13387935107891005,
      "grad_norm": 0.9148724675178528,
      "learning_rate": 0.00011791245664831251,
      "loss": 1.5665,
      "step": 850
    },
    {
      "epoch": 0.1354544022680737,
      "grad_norm": 0.9524568319320679,
      "learning_rate": 0.0001177259737564172,
      "loss": 1.553,
      "step": 860
    },
    {
      "epoch": 0.13702945345723735,
      "grad_norm": 1.016753911972046,
      "learning_rate": 0.00011753722074104613,
      "loss": 1.5546,
      "step": 870
    },
    {
      "epoch": 0.138604504646401,
      "grad_norm": 0.9564122557640076,
      "learning_rate": 0.00011734620536109644,
      "loss": 1.5667,
      "step": 880
    },
    {
      "epoch": 0.14017955583556466,
      "grad_norm": 0.9233742952346802,
      "learning_rate": 0.00011715293546846223,
      "loss": 1.5437,
      "step": 890
    },
    {
      "epoch": 0.1417546070247283,
      "grad_norm": 0.9245238900184631,
      "learning_rate": 0.00011695741900771184,
      "loss": 1.556,
      "step": 900
    },
    {
      "epoch": 0.1417546070247283,
      "eval_loss": 1.5552129745483398,
      "eval_runtime": 35.3017,
      "eval_samples_per_second": 14.192,
      "eval_steps_per_second": 14.192,
      "step": 900
    },
    {
      "epoch": 0.14332965821389196,
      "grad_norm": 0.9321339726448059,
      "learning_rate": 0.00011675966401576116,
      "loss": 1.576,
      "step": 910
    },
    {
      "epoch": 0.1449047094030556,
      "grad_norm": 0.9396321773529053,
      "learning_rate": 0.00011655967862154335,
      "loss": 1.5476,
      "step": 920
    },
    {
      "epoch": 0.14647976059221926,
      "grad_norm": 0.8809250593185425,
      "learning_rate": 0.0001163574710456747,
      "loss": 1.5297,
      "step": 930
    },
    {
      "epoch": 0.14805481178138288,
      "grad_norm": 0.9630320072174072,
      "learning_rate": 0.00011615304960011663,
      "loss": 1.5354,
      "step": 940
    },
    {
      "epoch": 0.14962986297054653,
      "grad_norm": 0.9173291325569153,
      "learning_rate": 0.00011594642268783415,
      "loss": 1.5397,
      "step": 950
    },
    {
      "epoch": 0.15120491415971019,
      "grad_norm": 0.892159640789032,
      "learning_rate": 0.00011573759880245027,
      "loss": 1.5665,
      "step": 960
    },
    {
      "epoch": 0.15277996534887384,
      "grad_norm": 0.8549361228942871,
      "learning_rate": 0.00011552658652789703,
      "loss": 1.5582,
      "step": 970
    },
    {
      "epoch": 0.1543550165380375,
      "grad_norm": 0.9306020736694336,
      "learning_rate": 0.00011531339453806258,
      "loss": 1.5418,
      "step": 980
    },
    {
      "epoch": 0.15593006772720114,
      "grad_norm": 0.9093769788742065,
      "learning_rate": 0.00011509803159643458,
      "loss": 1.5448,
      "step": 990
    },
    {
      "epoch": 0.1575051189163648,
      "grad_norm": 0.8876301050186157,
      "learning_rate": 0.00011488050655574003,
      "loss": 1.571,
      "step": 1000
    },
    {
      "epoch": 0.15908017010552844,
      "grad_norm": 0.9180966019630432,
      "learning_rate": 0.00011466082835758141,
      "loss": 1.5266,
      "step": 1010
    },
    {
      "epoch": 0.1606552212946921,
      "grad_norm": 0.9715962409973145,
      "learning_rate": 0.000114439006032069,
      "loss": 1.5336,
      "step": 1020
    },
    {
      "epoch": 0.16223027248385571,
      "grad_norm": 0.9360828995704651,
      "learning_rate": 0.00011421504869744978,
      "loss": 1.5513,
      "step": 1030
    },
    {
      "epoch": 0.16380532367301937,
      "grad_norm": 0.9322426319122314,
      "learning_rate": 0.0001139889655597326,
      "loss": 1.562,
      "step": 1040
    },
    {
      "epoch": 0.16538037486218302,
      "grad_norm": 0.9031831622123718,
      "learning_rate": 0.00011376076591230974,
      "loss": 1.5581,
      "step": 1050
    },
    {
      "epoch": 0.16538037486218302,
      "eval_loss": 1.547297716140747,
      "eval_runtime": 35.376,
      "eval_samples_per_second": 14.162,
      "eval_steps_per_second": 14.162,
      "step": 1050
    },
    {
      "epoch": 0.16695542605134667,
      "grad_norm": 0.9079886078834534,
      "learning_rate": 0.00011353045913557492,
      "loss": 1.5496,
      "step": 1060
    },
    {
      "epoch": 0.16853047724051032,
      "grad_norm": 0.9747688174247742,
      "learning_rate": 0.00011329805469653768,
      "loss": 1.5394,
      "step": 1070
    },
    {
      "epoch": 0.17010552842967397,
      "grad_norm": 0.9190161824226379,
      "learning_rate": 0.00011306356214843422,
      "loss": 1.5569,
      "step": 1080
    },
    {
      "epoch": 0.17168057961883762,
      "grad_norm": 0.9262961745262146,
      "learning_rate": 0.00011282699113033477,
      "loss": 1.5239,
      "step": 1090
    },
    {
      "epoch": 0.17325563080800127,
      "grad_norm": 0.933107852935791,
      "learning_rate": 0.00011258835136674729,
      "loss": 1.516,
      "step": 1100
    },
    {
      "epoch": 0.1748306819971649,
      "grad_norm": 0.8730500936508179,
      "learning_rate": 0.00011234765266721778,
      "loss": 1.5374,
      "step": 1110
    },
    {
      "epoch": 0.17640573318632854,
      "grad_norm": 0.8708956837654114,
      "learning_rate": 0.00011210490492592703,
      "loss": 1.5517,
      "step": 1120
    },
    {
      "epoch": 0.1779807843754922,
      "grad_norm": 0.9595373272895813,
      "learning_rate": 0.0001118601181212839,
      "loss": 1.5312,
      "step": 1130
    },
    {
      "epoch": 0.17955583556465585,
      "grad_norm": 0.9185504913330078,
      "learning_rate": 0.00011161330231551515,
      "loss": 1.5402,
      "step": 1140
    },
    {
      "epoch": 0.1811308867538195,
      "grad_norm": 0.965939462184906,
      "learning_rate": 0.00011136446765425187,
      "loss": 1.5496,
      "step": 1150
    },
    {
      "epoch": 0.18270593794298315,
      "grad_norm": 0.9474881291389465,
      "learning_rate": 0.00011111362436611234,
      "loss": 1.5365,
      "step": 1160
    },
    {
      "epoch": 0.1842809891321468,
      "grad_norm": 0.9285203218460083,
      "learning_rate": 0.00011086078276228167,
      "loss": 1.5372,
      "step": 1170
    },
    {
      "epoch": 0.18585604032131045,
      "grad_norm": 0.8643173575401306,
      "learning_rate": 0.00011060595323608789,
      "loss": 1.5364,
      "step": 1180
    },
    {
      "epoch": 0.1874310915104741,
      "grad_norm": 0.8951886296272278,
      "learning_rate": 0.00011034914626257467,
      "loss": 1.5162,
      "step": 1190
    },
    {
      "epoch": 0.18900614269963772,
      "grad_norm": 0.9068513512611389,
      "learning_rate": 0.0001100903723980709,
      "loss": 1.5573,
      "step": 1200
    },
    {
      "epoch": 0.18900614269963772,
      "eval_loss": 1.538773775100708,
      "eval_runtime": 35.125,
      "eval_samples_per_second": 14.263,
      "eval_steps_per_second": 14.263,
      "step": 1200
    },
    {
      "epoch": 0.19058119388880138,
      "grad_norm": 0.8557595610618591,
      "learning_rate": 0.00010982964227975658,
      "loss": 1.5192,
      "step": 1210
    },
    {
      "epoch": 0.19215624507796503,
      "grad_norm": 0.9791204333305359,
      "learning_rate": 0.00010956696662522569,
      "loss": 1.5512,
      "step": 1220
    },
    {
      "epoch": 0.19373129626712868,
      "grad_norm": 0.9069547057151794,
      "learning_rate": 0.00010930235623204551,
      "loss": 1.5318,
      "step": 1230
    },
    {
      "epoch": 0.19530634745629233,
      "grad_norm": 0.9511467218399048,
      "learning_rate": 0.00010903582197731294,
      "loss": 1.5302,
      "step": 1240
    },
    {
      "epoch": 0.19688139864545598,
      "grad_norm": 0.8480790257453918,
      "learning_rate": 0.00010876737481720722,
      "loss": 1.5382,
      "step": 1250
    },
    {
      "epoch": 0.19845644983461963,
      "grad_norm": 0.8755747675895691,
      "learning_rate": 0.0001084970257865397,
      "loss": 1.5282,
      "step": 1260
    },
    {
      "epoch": 0.20003150102378328,
      "grad_norm": 0.8885294198989868,
      "learning_rate": 0.00010822478599830008,
      "loss": 1.525,
      "step": 1270
    },
    {
      "epoch": 0.20160655221294693,
      "grad_norm": 0.8845210075378418,
      "learning_rate": 0.00010795066664319983,
      "loss": 1.5461,
      "step": 1280
    },
    {
      "epoch": 0.20318160340211056,
      "grad_norm": 0.8408452868461609,
      "learning_rate": 0.00010767467898921197,
      "loss": 1.5276,
      "step": 1290
    },
    {
      "epoch": 0.2047566545912742,
      "grad_norm": 0.8788501024246216,
      "learning_rate": 0.00010739683438110797,
      "loss": 1.5317,
      "step": 1300
    },
    {
      "epoch": 0.20633170578043786,
      "grad_norm": 0.9068732261657715,
      "learning_rate": 0.00010711714423999145,
      "loss": 1.522,
      "step": 1310
    },
    {
      "epoch": 0.2079067569696015,
      "grad_norm": 0.8549708724021912,
      "learning_rate": 0.00010683562006282861,
      "loss": 1.5366,
      "step": 1320
    },
    {
      "epoch": 0.20948180815876516,
      "grad_norm": 0.9036719799041748,
      "learning_rate": 0.00010655227342197574,
      "loss": 1.5298,
      "step": 1330
    },
    {
      "epoch": 0.2110568593479288,
      "grad_norm": 0.8768537044525146,
      "learning_rate": 0.00010626711596470343,
      "loss": 1.5454,
      "step": 1340
    },
    {
      "epoch": 0.21263191053709246,
      "grad_norm": 0.8741724491119385,
      "learning_rate": 0.0001059801594127179,
      "loss": 1.5262,
      "step": 1350
    },
    {
      "epoch": 0.21263191053709246,
      "eval_loss": 1.532226800918579,
      "eval_runtime": 35.9878,
      "eval_samples_per_second": 13.921,
      "eval_steps_per_second": 13.921,
      "step": 1350
    },
    {
      "epoch": 0.2142069617262561,
      "grad_norm": 0.8824599981307983,
      "learning_rate": 0.00010569141556167905,
      "loss": 1.5167,
      "step": 1360
    },
    {
      "epoch": 0.21578201291541976,
      "grad_norm": 0.8644326329231262,
      "learning_rate": 0.00010540089628071566,
      "loss": 1.5396,
      "step": 1370
    },
    {
      "epoch": 0.2173570641045834,
      "grad_norm": 0.8992419242858887,
      "learning_rate": 0.00010510861351193747,
      "loss": 1.552,
      "step": 1380
    },
    {
      "epoch": 0.21893211529374704,
      "grad_norm": 0.8747900724411011,
      "learning_rate": 0.00010481457926994435,
      "loss": 1.5247,
      "step": 1390
    },
    {
      "epoch": 0.2205071664829107,
      "grad_norm": 0.9142392873764038,
      "learning_rate": 0.0001045188056413323,
      "loss": 1.5624,
      "step": 1400
    },
    {
      "epoch": 0.22208221767207434,
      "grad_norm": 0.9147394299507141,
      "learning_rate": 0.00010422130478419676,
      "loss": 1.5079,
      "step": 1410
    },
    {
      "epoch": 0.223657268861238,
      "grad_norm": 0.8999921679496765,
      "learning_rate": 0.00010392208892763269,
      "loss": 1.5059,
      "step": 1420
    },
    {
      "epoch": 0.22523232005040164,
      "grad_norm": 0.9109930396080017,
      "learning_rate": 0.00010362117037123204,
      "loss": 1.5033,
      "step": 1430
    },
    {
      "epoch": 0.2268073712395653,
      "grad_norm": 0.8617811799049377,
      "learning_rate": 0.00010331856148457803,
      "loss": 1.542,
      "step": 1440
    },
    {
      "epoch": 0.22838242242872894,
      "grad_norm": 0.8684085011482239,
      "learning_rate": 0.00010301427470673678,
      "loss": 1.5227,
      "step": 1450
    },
    {
      "epoch": 0.2299574736178926,
      "grad_norm": 0.9092370271682739,
      "learning_rate": 0.00010270832254574588,
      "loss": 1.5159,
      "step": 1460
    },
    {
      "epoch": 0.23153252480705622,
      "grad_norm": 0.924177885055542,
      "learning_rate": 0.00010240071757810036,
      "loss": 1.5267,
      "step": 1470
    },
    {
      "epoch": 0.23310757599621987,
      "grad_norm": 0.8798425793647766,
      "learning_rate": 0.00010209147244823564,
      "loss": 1.5222,
      "step": 1480
    },
    {
      "epoch": 0.23468262718538352,
      "grad_norm": 0.9220207333564758,
      "learning_rate": 0.00010178059986800773,
      "loss": 1.5162,
      "step": 1490
    },
    {
      "epoch": 0.23625767837454717,
      "grad_norm": 0.8781149387359619,
      "learning_rate": 0.00010146811261617085,
      "loss": 1.5199,
      "step": 1500
    },
    {
      "epoch": 0.23625767837454717,
      "eval_loss": 1.5264769792556763,
      "eval_runtime": 36.7562,
      "eval_samples_per_second": 13.63,
      "eval_steps_per_second": 13.63,
      "step": 1500
    },
    {
      "epoch": 0.23783272956371082,
      "grad_norm": 0.878212571144104,
      "learning_rate": 0.00010115402353785197,
      "loss": 1.5316,
      "step": 1510
    },
    {
      "epoch": 0.23940778075287447,
      "grad_norm": 0.9163106679916382,
      "learning_rate": 0.00010083834554402292,
      "loss": 1.4983,
      "step": 1520
    },
    {
      "epoch": 0.24098283194203812,
      "grad_norm": 0.859491229057312,
      "learning_rate": 0.00010052109161096958,
      "loss": 1.5189,
      "step": 1530
    },
    {
      "epoch": 0.24255788313120177,
      "grad_norm": 0.8865215182304382,
      "learning_rate": 0.00010020227477975852,
      "loss": 1.5439,
      "step": 1540
    },
    {
      "epoch": 0.24413293432036542,
      "grad_norm": 0.8885824084281921,
      "learning_rate": 9.9881908155701e-05,
      "loss": 1.5338,
      "step": 1550
    },
    {
      "epoch": 0.24570798550952905,
      "grad_norm": 0.8940324783325195,
      "learning_rate": 9.956000490781411e-05,
      "loss": 1.4984,
      "step": 1560
    },
    {
      "epoch": 0.2472830366986927,
      "grad_norm": 0.9195623397827148,
      "learning_rate": 9.923657826827957e-05,
      "loss": 1.5041,
      "step": 1570
    },
    {
      "epoch": 0.24885808788785635,
      "grad_norm": 0.8722363114356995,
      "learning_rate": 9.891164153189976e-05,
      "loss": 1.5234,
      "step": 1580
    },
    {
      "epoch": 0.25043313907702003,
      "grad_norm": 0.8719931840896606,
      "learning_rate": 9.858520805555123e-05,
      "loss": 1.5178,
      "step": 1590
    },
    {
      "epoch": 0.25200819026618365,
      "grad_norm": 0.8836755156517029,
      "learning_rate": 9.825729125763561e-05,
      "loss": 1.5063,
      "step": 1600
    },
    {
      "epoch": 0.2535832414553473,
      "grad_norm": 0.9015933275222778,
      "learning_rate": 9.792790461752813e-05,
      "loss": 1.5021,
      "step": 1610
    },
    {
      "epoch": 0.25515829264451095,
      "grad_norm": 0.8438279628753662,
      "learning_rate": 9.759706167502343e-05,
      "loss": 1.5349,
      "step": 1620
    },
    {
      "epoch": 0.2567333438336746,
      "grad_norm": 0.8984500169754028,
      "learning_rate": 9.726477602977905e-05,
      "loss": 1.5106,
      "step": 1630
    },
    {
      "epoch": 0.25830839502283826,
      "grad_norm": 0.8898086547851562,
      "learning_rate": 9.69310613407564e-05,
      "loss": 1.5039,
      "step": 1640
    },
    {
      "epoch": 0.2598834462120019,
      "grad_norm": 0.9477494359016418,
      "learning_rate": 9.659593132565929e-05,
      "loss": 1.5283,
      "step": 1650
    },
    {
      "epoch": 0.2598834462120019,
      "eval_loss": 1.521310567855835,
      "eval_runtime": 36.1857,
      "eval_samples_per_second": 13.845,
      "eval_steps_per_second": 13.845,
      "step": 1650
    },
    {
      "epoch": 0.26145849740116556,
      "grad_norm": 0.8961330652236938,
      "learning_rate": 9.625939976037002e-05,
      "loss": 1.5148,
      "step": 1660
    },
    {
      "epoch": 0.2630335485903292,
      "grad_norm": 0.8740301132202148,
      "learning_rate": 9.59214804783831e-05,
      "loss": 1.522,
      "step": 1670
    },
    {
      "epoch": 0.26460859977949286,
      "grad_norm": 0.8525272607803345,
      "learning_rate": 9.558218737023671e-05,
      "loss": 1.5199,
      "step": 1680
    },
    {
      "epoch": 0.2661836509686565,
      "grad_norm": 0.8677767515182495,
      "learning_rate": 9.524153438294159e-05,
      "loss": 1.5203,
      "step": 1690
    },
    {
      "epoch": 0.2677587021578201,
      "grad_norm": 0.939079225063324,
      "learning_rate": 9.489953551940783e-05,
      "loss": 1.4987,
      "step": 1700
    },
    {
      "epoch": 0.2693337533469838,
      "grad_norm": 0.9007495045661926,
      "learning_rate": 9.455620483786914e-05,
      "loss": 1.5392,
      "step": 1710
    },
    {
      "epoch": 0.2709088045361474,
      "grad_norm": 0.8517370223999023,
      "learning_rate": 9.421155645130514e-05,
      "loss": 1.5233,
      "step": 1720
    },
    {
      "epoch": 0.2724838557253111,
      "grad_norm": 0.8468499779701233,
      "learning_rate": 9.38656045268611e-05,
      "loss": 1.5024,
      "step": 1730
    },
    {
      "epoch": 0.2740589069144747,
      "grad_norm": 0.8898671269416809,
      "learning_rate": 9.351836328526563e-05,
      "loss": 1.5207,
      "step": 1740
    },
    {
      "epoch": 0.2756339581036384,
      "grad_norm": 0.8659180402755737,
      "learning_rate": 9.316984700024612e-05,
      "loss": 1.539,
      "step": 1750
    },
    {
      "epoch": 0.277209009292802,
      "grad_norm": 0.9182498455047607,
      "learning_rate": 9.2820069997942e-05,
      "loss": 1.5113,
      "step": 1760
    },
    {
      "epoch": 0.2787840604819657,
      "grad_norm": 0.9074522852897644,
      "learning_rate": 9.246904665631588e-05,
      "loss": 1.4885,
      "step": 1770
    },
    {
      "epoch": 0.2803591116711293,
      "grad_norm": 0.8520182967185974,
      "learning_rate": 9.211679140456242e-05,
      "loss": 1.5234,
      "step": 1780
    },
    {
      "epoch": 0.28193416286029294,
      "grad_norm": 0.8538583517074585,
      "learning_rate": 9.176331872251536e-05,
      "loss": 1.5097,
      "step": 1790
    },
    {
      "epoch": 0.2835092140494566,
      "grad_norm": 0.8584926128387451,
      "learning_rate": 9.140864314005222e-05,
      "loss": 1.5267,
      "step": 1800
    },
    {
      "epoch": 0.2835092140494566,
      "eval_loss": 1.517116665840149,
      "eval_runtime": 36.7679,
      "eval_samples_per_second": 13.626,
      "eval_steps_per_second": 13.626,
      "step": 1800
    },
    {
      "epoch": 0.28508426523862024,
      "grad_norm": 0.8618518710136414,
      "learning_rate": 9.105277923649698e-05,
      "loss": 1.4953,
      "step": 1810
    },
    {
      "epoch": 0.2866593164277839,
      "grad_norm": 0.8893195390701294,
      "learning_rate": 9.06957416400209e-05,
      "loss": 1.5071,
      "step": 1820
    },
    {
      "epoch": 0.28823436761694754,
      "grad_norm": 0.8630818724632263,
      "learning_rate": 9.03375450270412e-05,
      "loss": 1.506,
      "step": 1830
    },
    {
      "epoch": 0.2898094188061112,
      "grad_norm": 0.9402779936790466,
      "learning_rate": 8.997820412161764e-05,
      "loss": 1.5244,
      "step": 1840
    },
    {
      "epoch": 0.29138446999527484,
      "grad_norm": 0.8658491373062134,
      "learning_rate": 8.961773369484738e-05,
      "loss": 1.5435,
      "step": 1850
    },
    {
      "epoch": 0.2929595211844385,
      "grad_norm": 0.849281907081604,
      "learning_rate": 8.925614856425786e-05,
      "loss": 1.493,
      "step": 1860
    },
    {
      "epoch": 0.29453457237360214,
      "grad_norm": 0.9385392069816589,
      "learning_rate": 8.88934635931975e-05,
      "loss": 1.4993,
      "step": 1870
    },
    {
      "epoch": 0.29610962356276577,
      "grad_norm": 0.8613683581352234,
      "learning_rate": 8.852969369022494e-05,
      "loss": 1.5085,
      "step": 1880
    },
    {
      "epoch": 0.29768467475192945,
      "grad_norm": 0.8535075783729553,
      "learning_rate": 8.816485380849613e-05,
      "loss": 1.5198,
      "step": 1890
    },
    {
      "epoch": 0.29925972594109307,
      "grad_norm": 0.8856109976768494,
      "learning_rate": 8.779895894514961e-05,
      "loss": 1.5259,
      "step": 1900
    },
    {
      "epoch": 0.30083477713025675,
      "grad_norm": 0.8867731690406799,
      "learning_rate": 8.743202414069011e-05,
      "loss": 1.5249,
      "step": 1910
    },
    {
      "epoch": 0.30240982831942037,
      "grad_norm": 0.8741790652275085,
      "learning_rate": 8.706406447837023e-05,
      "loss": 1.5271,
      "step": 1920
    },
    {
      "epoch": 0.30398487950858405,
      "grad_norm": 0.8374512195587158,
      "learning_rate": 8.669509508357052e-05,
      "loss": 1.5227,
      "step": 1930
    },
    {
      "epoch": 0.3055599306977477,
      "grad_norm": 0.8484550714492798,
      "learning_rate": 8.632513112317761e-05,
      "loss": 1.5222,
      "step": 1940
    },
    {
      "epoch": 0.30713498188691135,
      "grad_norm": 0.86808842420578,
      "learning_rate": 8.59541878049609e-05,
      "loss": 1.5055,
      "step": 1950
    },
    {
      "epoch": 0.30713498188691135,
      "eval_loss": 1.5131734609603882,
      "eval_runtime": 36.9625,
      "eval_samples_per_second": 13.554,
      "eval_steps_per_second": 13.554,
      "step": 1950
    },
    {
      "epoch": 0.308710033076075,
      "grad_norm": 0.847622275352478,
      "learning_rate": 8.558228037694728e-05,
      "loss": 1.4996,
      "step": 1960
    },
    {
      "epoch": 0.3102850842652386,
      "grad_norm": 0.910416841506958,
      "learning_rate": 8.520942412679447e-05,
      "loss": 1.508,
      "step": 1970
    },
    {
      "epoch": 0.3118601354544023,
      "grad_norm": 0.9068947434425354,
      "learning_rate": 8.483563438116257e-05,
      "loss": 1.508,
      "step": 1980
    },
    {
      "epoch": 0.3134351866435659,
      "grad_norm": 0.8562408685684204,
      "learning_rate": 8.446092650508393e-05,
      "loss": 1.5022,
      "step": 1990
    },
    {
      "epoch": 0.3150102378327296,
      "grad_norm": 0.9227258563041687,
      "learning_rate": 8.408531590133172e-05,
      "loss": 1.5213,
      "step": 2000
    },
    {
      "epoch": 0.3165852890218932,
      "grad_norm": 0.8598424196243286,
      "learning_rate": 8.370881800978673e-05,
      "loss": 1.5171,
      "step": 2010
    },
    {
      "epoch": 0.3181603402110569,
      "grad_norm": 0.8680636286735535,
      "learning_rate": 8.333144830680262e-05,
      "loss": 1.5223,
      "step": 2020
    },
    {
      "epoch": 0.3197353914002205,
      "grad_norm": 0.8430407047271729,
      "learning_rate": 8.29532223045698e-05,
      "loss": 1.4946,
      "step": 2030
    },
    {
      "epoch": 0.3213104425893842,
      "grad_norm": 0.901833713054657,
      "learning_rate": 8.257415555047785e-05,
      "loss": 1.5113,
      "step": 2040
    },
    {
      "epoch": 0.3228854937785478,
      "grad_norm": 0.8412112593650818,
      "learning_rate": 8.21942636264763e-05,
      "loss": 1.5253,
      "step": 2050
    },
    {
      "epoch": 0.32446054496771143,
      "grad_norm": 0.8819128274917603,
      "learning_rate": 8.181356214843422e-05,
      "loss": 1.4934,
      "step": 2060
    },
    {
      "epoch": 0.3260355961568751,
      "grad_norm": 0.8501127362251282,
      "learning_rate": 8.143206676549826e-05,
      "loss": 1.5007,
      "step": 2070
    },
    {
      "epoch": 0.32761064734603873,
      "grad_norm": 0.8944729566574097,
      "learning_rate": 8.10497931594494e-05,
      "loss": 1.5156,
      "step": 2080
    },
    {
      "epoch": 0.3291856985352024,
      "grad_norm": 0.8759516477584839,
      "learning_rate": 8.066675704405836e-05,
      "loss": 1.4965,
      "step": 2090
    },
    {
      "epoch": 0.33076074972436603,
      "grad_norm": 0.8777790069580078,
      "learning_rate": 8.028297416443952e-05,
      "loss": 1.4847,
      "step": 2100
    },
    {
      "epoch": 0.33076074972436603,
      "eval_loss": 1.508542776107788,
      "eval_runtime": 35.4369,
      "eval_samples_per_second": 14.138,
      "eval_steps_per_second": 14.138,
      "step": 2100
    },
    {
      "epoch": 0.3323358009135297,
      "grad_norm": 0.8308836221694946,
      "learning_rate": 7.989846029640397e-05,
      "loss": 1.4937,
      "step": 2110
    },
    {
      "epoch": 0.33391085210269333,
      "grad_norm": 0.8855435252189636,
      "learning_rate": 7.951323124581069e-05,
      "loss": 1.5021,
      "step": 2120
    },
    {
      "epoch": 0.335485903291857,
      "grad_norm": 0.8671780824661255,
      "learning_rate": 7.91273028479172e-05,
      "loss": 1.5045,
      "step": 2130
    },
    {
      "epoch": 0.33706095448102064,
      "grad_norm": 0.8830945491790771,
      "learning_rate": 7.874069096672831e-05,
      "loss": 1.5186,
      "step": 2140
    },
    {
      "epoch": 0.33863600567018426,
      "grad_norm": 0.8480697274208069,
      "learning_rate": 7.83534114943442e-05,
      "loss": 1.5123,
      "step": 2150
    },
    {
      "epoch": 0.34021105685934794,
      "grad_norm": 0.8749104142189026,
      "learning_rate": 7.796548035030715e-05,
      "loss": 1.4982,
      "step": 2160
    },
    {
      "epoch": 0.34178610804851156,
      "grad_norm": 0.872046947479248,
      "learning_rate": 7.757691348094703e-05,
      "loss": 1.4944,
      "step": 2170
    },
    {
      "epoch": 0.34336115923767524,
      "grad_norm": 0.8679631948471069,
      "learning_rate": 7.718772685872595e-05,
      "loss": 1.4991,
      "step": 2180
    },
    {
      "epoch": 0.34493621042683886,
      "grad_norm": 0.8403679728507996,
      "learning_rate": 7.679793648158159e-05,
      "loss": 1.4881,
      "step": 2190
    },
    {
      "epoch": 0.34651126161600254,
      "grad_norm": 0.8893296718597412,
      "learning_rate": 7.640755837226965e-05,
      "loss": 1.5111,
      "step": 2200
    },
    {
      "epoch": 0.34808631280516616,
      "grad_norm": 0.9288545846939087,
      "learning_rate": 7.601660857770522e-05,
      "loss": 1.5076,
      "step": 2210
    },
    {
      "epoch": 0.3496613639943298,
      "grad_norm": 0.8662068247795105,
      "learning_rate": 7.562510316830308e-05,
      "loss": 1.5162,
      "step": 2220
    },
    {
      "epoch": 0.35123641518349347,
      "grad_norm": 0.8514265418052673,
      "learning_rate": 7.523305823731723e-05,
      "loss": 1.4988,
      "step": 2230
    },
    {
      "epoch": 0.3528114663726571,
      "grad_norm": 0.866111695766449,
      "learning_rate": 7.484048990017919e-05,
      "loss": 1.5058,
      "step": 2240
    },
    {
      "epoch": 0.35438651756182077,
      "grad_norm": 0.8357524871826172,
      "learning_rate": 7.444741429383578e-05,
      "loss": 1.5061,
      "step": 2250
    },
    {
      "epoch": 0.35438651756182077,
      "eval_loss": 1.5059038400650024,
      "eval_runtime": 35.4488,
      "eval_samples_per_second": 14.133,
      "eval_steps_per_second": 14.133,
      "step": 2250
    },
    {
      "epoch": 0.3559615687509844,
      "grad_norm": 0.8725208640098572,
      "learning_rate": 7.405384757608555e-05,
      "loss": 1.4972,
      "step": 2260
    },
    {
      "epoch": 0.35753661994014807,
      "grad_norm": 0.9390593767166138,
      "learning_rate": 7.36598059249148e-05,
      "loss": 1.5082,
      "step": 2270
    },
    {
      "epoch": 0.3591116711293117,
      "grad_norm": 0.8906267881393433,
      "learning_rate": 7.326530553783243e-05,
      "loss": 1.5092,
      "step": 2280
    },
    {
      "epoch": 0.3606867223184754,
      "grad_norm": 0.8382712602615356,
      "learning_rate": 7.287036263120425e-05,
      "loss": 1.4973,
      "step": 2290
    },
    {
      "epoch": 0.362261773507639,
      "grad_norm": 0.8869454264640808,
      "learning_rate": 7.247499343958621e-05,
      "loss": 1.4954,
      "step": 2300
    },
    {
      "epoch": 0.3638368246968026,
      "grad_norm": 0.8357947468757629,
      "learning_rate": 7.207921421505724e-05,
      "loss": 1.5114,
      "step": 2310
    },
    {
      "epoch": 0.3654118758859663,
      "grad_norm": 0.8370199799537659,
      "learning_rate": 7.168304122655113e-05,
      "loss": 1.5215,
      "step": 2320
    },
    {
      "epoch": 0.3669869270751299,
      "grad_norm": 0.8827706575393677,
      "learning_rate": 7.128649075918768e-05,
      "loss": 1.4958,
      "step": 2330
    },
    {
      "epoch": 0.3685619782642936,
      "grad_norm": 0.8319584727287292,
      "learning_rate": 7.088957911360347e-05,
      "loss": 1.5004,
      "step": 2340
    },
    {
      "epoch": 0.3701370294534572,
      "grad_norm": 0.8250936269760132,
      "learning_rate": 7.049232260528163e-05,
      "loss": 1.5118,
      "step": 2350
    },
    {
      "epoch": 0.3717120806426209,
      "grad_norm": 0.8642278909683228,
      "learning_rate": 7.009473756388128e-05,
      "loss": 1.5154,
      "step": 2360
    },
    {
      "epoch": 0.3732871318317845,
      "grad_norm": 0.8907039165496826,
      "learning_rate": 6.969684033256622e-05,
      "loss": 1.4889,
      "step": 2370
    },
    {
      "epoch": 0.3748621830209482,
      "grad_norm": 0.8550412058830261,
      "learning_rate": 6.92986472673332e-05,
      "loss": 1.5254,
      "step": 2380
    },
    {
      "epoch": 0.3764372342101118,
      "grad_norm": 0.9388350248336792,
      "learning_rate": 6.890017473633946e-05,
      "loss": 1.5123,
      "step": 2390
    },
    {
      "epoch": 0.37801228539927545,
      "grad_norm": 0.860334038734436,
      "learning_rate": 6.850143911923011e-05,
      "loss": 1.513,
      "step": 2400
    },
    {
      "epoch": 0.37801228539927545,
      "eval_loss": 1.5027722120285034,
      "eval_runtime": 35.9341,
      "eval_samples_per_second": 13.942,
      "eval_steps_per_second": 13.942,
      "step": 2400
    },
    {
      "epoch": 0.37958733658843913,
      "grad_norm": 0.8568481206893921,
      "learning_rate": 6.81024568064646e-05,
      "loss": 1.5119,
      "step": 2410
    },
    {
      "epoch": 0.38116238777760275,
      "grad_norm": 0.8408715128898621,
      "learning_rate": 6.770324419864309e-05,
      "loss": 1.5051,
      "step": 2420
    },
    {
      "epoch": 0.38273743896676643,
      "grad_norm": 0.8416137099266052,
      "learning_rate": 6.73038177058323e-05,
      "loss": 1.508,
      "step": 2430
    },
    {
      "epoch": 0.38431249015593005,
      "grad_norm": 0.8409276604652405,
      "learning_rate": 6.690419374689087e-05,
      "loss": 1.4854,
      "step": 2440
    },
    {
      "epoch": 0.38588754134509373,
      "grad_norm": 0.8319918513298035,
      "learning_rate": 6.650438874879456e-05,
      "loss": 1.5115,
      "step": 2450
    },
    {
      "epoch": 0.38746259253425736,
      "grad_norm": 0.8504980802536011,
      "learning_rate": 6.61044191459609e-05,
      "loss": 1.5088,
      "step": 2460
    },
    {
      "epoch": 0.38903764372342103,
      "grad_norm": 0.9393388628959656,
      "learning_rate": 6.57043013795737e-05,
      "loss": 1.5088,
      "step": 2470
    },
    {
      "epoch": 0.39061269491258466,
      "grad_norm": 0.8659742474555969,
      "learning_rate": 6.530405189690719e-05,
      "loss": 1.4848,
      "step": 2480
    },
    {
      "epoch": 0.3921877461017483,
      "grad_norm": 0.8352620005607605,
      "learning_rate": 6.49036871506499e-05,
      "loss": 1.4941,
      "step": 2490
    },
    {
      "epoch": 0.39376279729091196,
      "grad_norm": 0.8595870733261108,
      "learning_rate": 6.450322359822846e-05,
      "loss": 1.5065,
      "step": 2500
    },
    {
      "epoch": 0.3953378484800756,
      "grad_norm": 0.9233240485191345,
      "learning_rate": 6.410267770113098e-05,
      "loss": 1.4957,
      "step": 2510
    },
    {
      "epoch": 0.39691289966923926,
      "grad_norm": 0.8546603322029114,
      "learning_rate": 6.370206592423045e-05,
      "loss": 1.4699,
      "step": 2520
    },
    {
      "epoch": 0.3984879508584029,
      "grad_norm": 0.8289804458618164,
      "learning_rate": 6.330140473510796e-05,
      "loss": 1.5038,
      "step": 2530
    },
    {
      "epoch": 0.40006300204756656,
      "grad_norm": 0.8960427641868591,
      "learning_rate": 6.29007106033757e-05,
      "loss": 1.4959,
      "step": 2540
    },
    {
      "epoch": 0.4016380532367302,
      "grad_norm": 0.8288776874542236,
      "learning_rate": 6.25e-05,
      "loss": 1.5089,
      "step": 2550
    },
    {
      "epoch": 0.4016380532367302,
      "eval_loss": 1.500240683555603,
      "eval_runtime": 36.1286,
      "eval_samples_per_second": 13.867,
      "eval_steps_per_second": 13.867,
      "step": 2550
    },
    {
      "epoch": 0.40321310442589386,
      "grad_norm": 0.8510929942131042,
      "learning_rate": 6.20992893966243e-05,
      "loss": 1.4884,
      "step": 2560
    },
    {
      "epoch": 0.4047881556150575,
      "grad_norm": 0.8261246681213379,
      "learning_rate": 6.169859526489204e-05,
      "loss": 1.5104,
      "step": 2570
    },
    {
      "epoch": 0.4063632068042211,
      "grad_norm": 0.8622429370880127,
      "learning_rate": 6.129793407576955e-05,
      "loss": 1.4953,
      "step": 2580
    },
    {
      "epoch": 0.4079382579933848,
      "grad_norm": 0.8522881865501404,
      "learning_rate": 6.089732229886904e-05,
      "loss": 1.4839,
      "step": 2590
    },
    {
      "epoch": 0.4095133091825484,
      "grad_norm": 0.891878068447113,
      "learning_rate": 6.049677640177155e-05,
      "loss": 1.5085,
      "step": 2600
    },
    {
      "epoch": 0.4110883603717121,
      "grad_norm": 0.8279666900634766,
      "learning_rate": 6.00963128493501e-05,
      "loss": 1.5266,
      "step": 2610
    },
    {
      "epoch": 0.4126634115608757,
      "grad_norm": 0.8490031361579895,
      "learning_rate": 5.969594810309284e-05,
      "loss": 1.5034,
      "step": 2620
    },
    {
      "epoch": 0.4142384627500394,
      "grad_norm": 0.8807093501091003,
      "learning_rate": 5.929569862042631e-05,
      "loss": 1.5154,
      "step": 2630
    },
    {
      "epoch": 0.415813513939203,
      "grad_norm": 0.8721035122871399,
      "learning_rate": 5.889558085403911e-05,
      "loss": 1.4948,
      "step": 2640
    },
    {
      "epoch": 0.4173885651283667,
      "grad_norm": 0.8579753041267395,
      "learning_rate": 5.849561125120545e-05,
      "loss": 1.5139,
      "step": 2650
    },
    {
      "epoch": 0.4189636163175303,
      "grad_norm": 0.8881575465202332,
      "learning_rate": 5.809580625310912e-05,
      "loss": 1.511,
      "step": 2660
    },
    {
      "epoch": 0.42053866750669394,
      "grad_norm": 0.8358275294303894,
      "learning_rate": 5.769618229416773e-05,
      "loss": 1.5035,
      "step": 2670
    },
    {
      "epoch": 0.4221137186958576,
      "grad_norm": 0.8375245332717896,
      "learning_rate": 5.7296755801356926e-05,
      "loss": 1.5117,
      "step": 2680
    },
    {
      "epoch": 0.42368876988502124,
      "grad_norm": 0.8418902158737183,
      "learning_rate": 5.6897543193535414e-05,
      "loss": 1.4985,
      "step": 2690
    },
    {
      "epoch": 0.4252638210741849,
      "grad_norm": 0.8790860772132874,
      "learning_rate": 5.649856088076989e-05,
      "loss": 1.497,
      "step": 2700
    },
    {
      "epoch": 0.4252638210741849,
      "eval_loss": 1.4972805976867676,
      "eval_runtime": 36.5785,
      "eval_samples_per_second": 13.697,
      "eval_steps_per_second": 13.697,
      "step": 2700
    },
    {
      "epoch": 0.42683887226334855,
      "grad_norm": 0.885306179523468,
      "learning_rate": 5.609982526366054e-05,
      "loss": 1.4825,
      "step": 2710
    },
    {
      "epoch": 0.4284139234525122,
      "grad_norm": 0.9426984786987305,
      "learning_rate": 5.570135273266683e-05,
      "loss": 1.5129,
      "step": 2720
    },
    {
      "epoch": 0.42998897464167585,
      "grad_norm": 0.8404316902160645,
      "learning_rate": 5.53031596674338e-05,
      "loss": 1.4818,
      "step": 2730
    },
    {
      "epoch": 0.4315640258308395,
      "grad_norm": 0.9347336292266846,
      "learning_rate": 5.490526243611873e-05,
      "loss": 1.5109,
      "step": 2740
    },
    {
      "epoch": 0.43313907702000315,
      "grad_norm": 0.9323931336402893,
      "learning_rate": 5.450767739471837e-05,
      "loss": 1.4973,
      "step": 2750
    },
    {
      "epoch": 0.4347141282091668,
      "grad_norm": 0.8658491373062134,
      "learning_rate": 5.411042088639655e-05,
      "loss": 1.4957,
      "step": 2760
    },
    {
      "epoch": 0.43628917939833045,
      "grad_norm": 0.8817048072814941,
      "learning_rate": 5.371350924081234e-05,
      "loss": 1.5003,
      "step": 2770
    },
    {
      "epoch": 0.4378642305874941,
      "grad_norm": 0.8737426996231079,
      "learning_rate": 5.331695877344888e-05,
      "loss": 1.4831,
      "step": 2780
    },
    {
      "epoch": 0.43943928177665775,
      "grad_norm": 0.8520674705505371,
      "learning_rate": 5.292078578494275e-05,
      "loss": 1.4753,
      "step": 2790
    },
    {
      "epoch": 0.4410143329658214,
      "grad_norm": 0.8379813432693481,
      "learning_rate": 5.2525006560413816e-05,
      "loss": 1.5039,
      "step": 2800
    },
    {
      "epoch": 0.44258938415498505,
      "grad_norm": 0.8281566500663757,
      "learning_rate": 5.212963736879578e-05,
      "loss": 1.4906,
      "step": 2810
    },
    {
      "epoch": 0.4441644353441487,
      "grad_norm": 0.8392037749290466,
      "learning_rate": 5.173469446216757e-05,
      "loss": 1.4762,
      "step": 2820
    },
    {
      "epoch": 0.44573948653331236,
      "grad_norm": 0.861897885799408,
      "learning_rate": 5.134019407508521e-05,
      "loss": 1.5046,
      "step": 2830
    },
    {
      "epoch": 0.447314537722476,
      "grad_norm": 0.8166014552116394,
      "learning_rate": 5.0946152423914456e-05,
      "loss": 1.491,
      "step": 2840
    },
    {
      "epoch": 0.4488895889116396,
      "grad_norm": 0.8650168776512146,
      "learning_rate": 5.0552585706164246e-05,
      "loss": 1.4942,
      "step": 2850
    },
    {
      "epoch": 0.4488895889116396,
      "eval_loss": 1.4944835901260376,
      "eval_runtime": 35.972,
      "eval_samples_per_second": 13.927,
      "eval_steps_per_second": 13.927,
      "step": 2850
    },
    {
      "epoch": 0.4504646401008033,
      "grad_norm": 0.8844559192657471,
      "learning_rate": 5.015951009982081e-05,
      "loss": 1.4926,
      "step": 2860
    },
    {
      "epoch": 0.4520396912899669,
      "grad_norm": 0.8611796498298645,
      "learning_rate": 4.976694176268278e-05,
      "loss": 1.4975,
      "step": 2870
    },
    {
      "epoch": 0.4536147424791306,
      "grad_norm": 0.8338538408279419,
      "learning_rate": 4.937489683169692e-05,
      "loss": 1.4958,
      "step": 2880
    },
    {
      "epoch": 0.4551897936682942,
      "grad_norm": 0.9034844040870667,
      "learning_rate": 4.8983391422294786e-05,
      "loss": 1.5194,
      "step": 2890
    },
    {
      "epoch": 0.4567648448574579,
      "grad_norm": 0.8422059416770935,
      "learning_rate": 4.8592441627730355e-05,
      "loss": 1.5007,
      "step": 2900
    },
    {
      "epoch": 0.4583398960466215,
      "grad_norm": 0.8396774530410767,
      "learning_rate": 4.820206351841842e-05,
      "loss": 1.4905,
      "step": 2910
    },
    {
      "epoch": 0.4599149472357852,
      "grad_norm": 0.8403754830360413,
      "learning_rate": 4.781227314127405e-05,
      "loss": 1.5033,
      "step": 2920
    },
    {
      "epoch": 0.4614899984249488,
      "grad_norm": 0.8683463931083679,
      "learning_rate": 4.7423086519052966e-05,
      "loss": 1.4917,
      "step": 2930
    },
    {
      "epoch": 0.46306504961411243,
      "grad_norm": 0.8298001289367676,
      "learning_rate": 4.703451964969287e-05,
      "loss": 1.511,
      "step": 2940
    },
    {
      "epoch": 0.4646401008032761,
      "grad_norm": 0.8793701529502869,
      "learning_rate": 4.66465885056558e-05,
      "loss": 1.4906,
      "step": 2950
    },
    {
      "epoch": 0.46621515199243974,
      "grad_norm": 0.8985243439674377,
      "learning_rate": 4.62593090332717e-05,
      "loss": 1.4873,
      "step": 2960
    },
    {
      "epoch": 0.4677902031816034,
      "grad_norm": 0.834176778793335,
      "learning_rate": 4.587269715208281e-05,
      "loss": 1.4886,
      "step": 2970
    },
    {
      "epoch": 0.46936525437076704,
      "grad_norm": 0.8913531303405762,
      "learning_rate": 4.5486768754189305e-05,
      "loss": 1.4952,
      "step": 2980
    },
    {
      "epoch": 0.4709403055599307,
      "grad_norm": 0.8446453213691711,
      "learning_rate": 4.510153970359606e-05,
      "loss": 1.5149,
      "step": 2990
    },
    {
      "epoch": 0.47251535674909434,
      "grad_norm": 0.9674009680747986,
      "learning_rate": 4.4717025835560476e-05,
      "loss": 1.4887,
      "step": 3000
    },
    {
      "epoch": 0.47251535674909434,
      "eval_loss": 1.4920191764831543,
      "eval_runtime": 35.8777,
      "eval_samples_per_second": 13.964,
      "eval_steps_per_second": 13.964,
      "step": 3000
    },
    {
      "epoch": 0.474090407938258,
      "grad_norm": 0.8722410202026367,
      "learning_rate": 4.433324295594166e-05,
      "loss": 1.4964,
      "step": 3010
    },
    {
      "epoch": 0.47566545912742164,
      "grad_norm": 0.9078059792518616,
      "learning_rate": 4.3950206840550585e-05,
      "loss": 1.4928,
      "step": 3020
    },
    {
      "epoch": 0.47724051031658526,
      "grad_norm": 0.8529486060142517,
      "learning_rate": 4.3567933234501746e-05,
      "loss": 1.5025,
      "step": 3030
    },
    {
      "epoch": 0.47881556150574894,
      "grad_norm": 0.848800003528595,
      "learning_rate": 4.318643785156579e-05,
      "loss": 1.4942,
      "step": 3040
    },
    {
      "epoch": 0.48039061269491257,
      "grad_norm": 0.844018816947937,
      "learning_rate": 4.280573637352371e-05,
      "loss": 1.4918,
      "step": 3050
    },
    {
      "epoch": 0.48196566388407625,
      "grad_norm": 0.8629136085510254,
      "learning_rate": 4.242584444952216e-05,
      "loss": 1.5129,
      "step": 3060
    },
    {
      "epoch": 0.48354071507323987,
      "grad_norm": 0.9015205502510071,
      "learning_rate": 4.204677769543019e-05,
      "loss": 1.4685,
      "step": 3070
    },
    {
      "epoch": 0.48511576626240355,
      "grad_norm": 0.8628337979316711,
      "learning_rate": 4.16685516931974e-05,
      "loss": 1.5157,
      "step": 3080
    },
    {
      "epoch": 0.48669081745156717,
      "grad_norm": 0.8275268077850342,
      "learning_rate": 4.1291181990213286e-05,
      "loss": 1.4864,
      "step": 3090
    },
    {
      "epoch": 0.48826586864073085,
      "grad_norm": 0.8484097719192505,
      "learning_rate": 4.0914684098668286e-05,
      "loss": 1.4861,
      "step": 3100
    },
    {
      "epoch": 0.48984091982989447,
      "grad_norm": 0.8497301936149597,
      "learning_rate": 4.053907349491608e-05,
      "loss": 1.4999,
      "step": 3110
    },
    {
      "epoch": 0.4914159710190581,
      "grad_norm": 0.8445069789886475,
      "learning_rate": 4.016436561883746e-05,
      "loss": 1.4845,
      "step": 3120
    },
    {
      "epoch": 0.4929910222082218,
      "grad_norm": 0.8241081833839417,
      "learning_rate": 3.979057587320554e-05,
      "loss": 1.4993,
      "step": 3130
    },
    {
      "epoch": 0.4945660733973854,
      "grad_norm": 0.8906779289245605,
      "learning_rate": 3.941771962305274e-05,
      "loss": 1.4864,
      "step": 3140
    },
    {
      "epoch": 0.4961411245865491,
      "grad_norm": 0.9039560556411743,
      "learning_rate": 3.9045812195039125e-05,
      "loss": 1.479,
      "step": 3150
    },
    {
      "epoch": 0.4961411245865491,
      "eval_loss": 1.490482211112976,
      "eval_runtime": 35.4913,
      "eval_samples_per_second": 14.116,
      "eval_steps_per_second": 14.116,
      "step": 3150
    },
    {
      "epoch": 0.4977161757757127,
      "grad_norm": 0.8476235270500183,
      "learning_rate": 3.8674868876822395e-05,
      "loss": 1.5039,
      "step": 3160
    },
    {
      "epoch": 0.4992912269648764,
      "grad_norm": 0.8413540720939636,
      "learning_rate": 3.83049049164295e-05,
      "loss": 1.5041,
      "step": 3170
    },
    {
      "epoch": 0.5008662781540401,
      "grad_norm": 0.8997570872306824,
      "learning_rate": 3.793593552162978e-05,
      "loss": 1.489,
      "step": 3180
    },
    {
      "epoch": 0.5024413293432036,
      "grad_norm": 0.8490859270095825,
      "learning_rate": 3.75679758593099e-05,
      "loss": 1.4897,
      "step": 3190
    },
    {
      "epoch": 0.5040163805323673,
      "grad_norm": 0.9092013835906982,
      "learning_rate": 3.720104105485039e-05,
      "loss": 1.5053,
      "step": 3200
    },
    {
      "epoch": 0.505591431721531,
      "grad_norm": 0.8679947853088379,
      "learning_rate": 3.6835146191503885e-05,
      "loss": 1.4871,
      "step": 3210
    },
    {
      "epoch": 0.5071664829106945,
      "grad_norm": 0.842693567276001,
      "learning_rate": 3.647030630977508e-05,
      "loss": 1.4863,
      "step": 3220
    },
    {
      "epoch": 0.5087415340998582,
      "grad_norm": 0.8691567182540894,
      "learning_rate": 3.6106536406802524e-05,
      "loss": 1.4866,
      "step": 3230
    },
    {
      "epoch": 0.5103165852890219,
      "grad_norm": 0.8479135036468506,
      "learning_rate": 3.5743851435742176e-05,
      "loss": 1.4759,
      "step": 3240
    },
    {
      "epoch": 0.5118916364781856,
      "grad_norm": 0.819887638092041,
      "learning_rate": 3.538226630515262e-05,
      "loss": 1.4833,
      "step": 3250
    },
    {
      "epoch": 0.5134666876673492,
      "grad_norm": 0.881779670715332,
      "learning_rate": 3.502179587838238e-05,
      "loss": 1.498,
      "step": 3260
    },
    {
      "epoch": 0.5150417388565128,
      "grad_norm": 0.8957091569900513,
      "learning_rate": 3.46624549729588e-05,
      "loss": 1.5042,
      "step": 3270
    },
    {
      "epoch": 0.5166167900456765,
      "grad_norm": 0.8583941459655762,
      "learning_rate": 3.430425835997908e-05,
      "loss": 1.5151,
      "step": 3280
    },
    {
      "epoch": 0.5181918412348401,
      "grad_norm": 0.8976061940193176,
      "learning_rate": 3.394722076350302e-05,
      "loss": 1.4841,
      "step": 3290
    },
    {
      "epoch": 0.5197668924240038,
      "grad_norm": 0.8954448699951172,
      "learning_rate": 3.359135685994781e-05,
      "loss": 1.4981,
      "step": 3300
    },
    {
      "epoch": 0.5197668924240038,
      "eval_loss": 1.4886995553970337,
      "eval_runtime": 36.2455,
      "eval_samples_per_second": 13.822,
      "eval_steps_per_second": 13.822,
      "step": 3300
    },
    {
      "epoch": 0.5213419436131674,
      "grad_norm": 0.8868695497512817,
      "learning_rate": 3.3236681277484654e-05,
      "loss": 1.4895,
      "step": 3310
    },
    {
      "epoch": 0.5229169948023311,
      "grad_norm": 0.8330621123313904,
      "learning_rate": 3.2883208595437584e-05,
      "loss": 1.5,
      "step": 3320
    },
    {
      "epoch": 0.5244920459914947,
      "grad_norm": 0.9283692240715027,
      "learning_rate": 3.2530953343684136e-05,
      "loss": 1.4892,
      "step": 3330
    },
    {
      "epoch": 0.5260670971806584,
      "grad_norm": 0.841469943523407,
      "learning_rate": 3.217993000205799e-05,
      "loss": 1.5036,
      "step": 3340
    },
    {
      "epoch": 0.527642148369822,
      "grad_norm": 0.9198566675186157,
      "learning_rate": 3.1830152999753903e-05,
      "loss": 1.4751,
      "step": 3350
    },
    {
      "epoch": 0.5292171995589857,
      "grad_norm": 0.8299483060836792,
      "learning_rate": 3.148163671473439e-05,
      "loss": 1.468,
      "step": 3360
    },
    {
      "epoch": 0.5307922507481493,
      "grad_norm": 0.858586311340332,
      "learning_rate": 3.113439547313892e-05,
      "loss": 1.469,
      "step": 3370
    },
    {
      "epoch": 0.532367301937313,
      "grad_norm": 0.8130866885185242,
      "learning_rate": 3.0788443548694874e-05,
      "loss": 1.481,
      "step": 3380
    },
    {
      "epoch": 0.5339423531264766,
      "grad_norm": 0.8363484740257263,
      "learning_rate": 3.0443795162130876e-05,
      "loss": 1.4855,
      "step": 3390
    },
    {
      "epoch": 0.5355174043156402,
      "grad_norm": 0.8633217811584473,
      "learning_rate": 3.0100464480592185e-05,
      "loss": 1.491,
      "step": 3400
    },
    {
      "epoch": 0.5370924555048039,
      "grad_norm": 0.8282185196876526,
      "learning_rate": 2.9758465617058404e-05,
      "loss": 1.4894,
      "step": 3410
    },
    {
      "epoch": 0.5386675066939676,
      "grad_norm": 0.8978461027145386,
      "learning_rate": 2.9417812629763285e-05,
      "loss": 1.4821,
      "step": 3420
    },
    {
      "epoch": 0.5402425578831312,
      "grad_norm": 0.8958382606506348,
      "learning_rate": 2.9078519521616894e-05,
      "loss": 1.5043,
      "step": 3430
    },
    {
      "epoch": 0.5418176090722948,
      "grad_norm": 0.8614096641540527,
      "learning_rate": 2.8740600239630002e-05,
      "loss": 1.4882,
      "step": 3440
    },
    {
      "epoch": 0.5433926602614585,
      "grad_norm": 0.846077024936676,
      "learning_rate": 2.8404068674340714e-05,
      "loss": 1.4775,
      "step": 3450
    },
    {
      "epoch": 0.5433926602614585,
      "eval_loss": 1.486568808555603,
      "eval_runtime": 35.5801,
      "eval_samples_per_second": 14.081,
      "eval_steps_per_second": 14.081,
      "step": 3450
    },
    {
      "epoch": 0.5449677114506222,
      "grad_norm": 0.8439909815788269,
      "learning_rate": 2.80689386592436e-05,
      "loss": 1.4801,
      "step": 3460
    },
    {
      "epoch": 0.5465427626397857,
      "grad_norm": 0.866398274898529,
      "learning_rate": 2.7735223970220955e-05,
      "loss": 1.4751,
      "step": 3470
    },
    {
      "epoch": 0.5481178138289494,
      "grad_norm": 0.8495659232139587,
      "learning_rate": 2.7402938324976576e-05,
      "loss": 1.4839,
      "step": 3480
    },
    {
      "epoch": 0.5496928650181131,
      "grad_norm": 0.8776513934135437,
      "learning_rate": 2.70720953824719e-05,
      "loss": 1.4803,
      "step": 3490
    },
    {
      "epoch": 0.5512679162072768,
      "grad_norm": 0.8576133251190186,
      "learning_rate": 2.674270874236441e-05,
      "loss": 1.4799,
      "step": 3500
    },
    {
      "epoch": 0.5528429673964403,
      "grad_norm": 0.8660094141960144,
      "learning_rate": 2.64147919444488e-05,
      "loss": 1.4933,
      "step": 3510
    },
    {
      "epoch": 0.554418018585604,
      "grad_norm": 0.8755300045013428,
      "learning_rate": 2.6088358468100247e-05,
      "loss": 1.5073,
      "step": 3520
    },
    {
      "epoch": 0.5559930697747677,
      "grad_norm": 0.8800572156906128,
      "learning_rate": 2.5763421731720435e-05,
      "loss": 1.5244,
      "step": 3530
    },
    {
      "epoch": 0.5575681209639314,
      "grad_norm": 0.8597940802574158,
      "learning_rate": 2.5439995092185892e-05,
      "loss": 1.4697,
      "step": 3540
    },
    {
      "epoch": 0.559143172153095,
      "grad_norm": 0.8852632641792297,
      "learning_rate": 2.5118091844299e-05,
      "loss": 1.4786,
      "step": 3550
    },
    {
      "epoch": 0.5607182233422586,
      "grad_norm": 0.84444659948349,
      "learning_rate": 2.479772522024147e-05,
      "loss": 1.4842,
      "step": 3560
    },
    {
      "epoch": 0.5622932745314223,
      "grad_norm": 0.8705389499664307,
      "learning_rate": 2.4478908389030427e-05,
      "loss": 1.4817,
      "step": 3570
    },
    {
      "epoch": 0.5638683257205859,
      "grad_norm": 0.8876937627792358,
      "learning_rate": 2.41616544559771e-05,
      "loss": 1.5087,
      "step": 3580
    },
    {
      "epoch": 0.5654433769097496,
      "grad_norm": 0.8541763424873352,
      "learning_rate": 2.3845976462148033e-05,
      "loss": 1.4908,
      "step": 3590
    },
    {
      "epoch": 0.5670184280989132,
      "grad_norm": 0.8582256436347961,
      "learning_rate": 2.3531887383829157e-05,
      "loss": 1.4732,
      "step": 3600
    },
    {
      "epoch": 0.5670184280989132,
      "eval_loss": 1.485674262046814,
      "eval_runtime": 37.1809,
      "eval_samples_per_second": 13.475,
      "eval_steps_per_second": 13.475,
      "step": 3600
    },
    {
      "epoch": 0.5685934792880769,
      "grad_norm": 0.8538834452629089,
      "learning_rate": 2.3219400131992273e-05,
      "loss": 1.4944,
      "step": 3610
    },
    {
      "epoch": 0.5701685304772405,
      "grad_norm": 0.8310463428497314,
      "learning_rate": 2.2908527551764404e-05,
      "loss": 1.4637,
      "step": 3620
    },
    {
      "epoch": 0.5717435816664042,
      "grad_norm": 0.9042952656745911,
      "learning_rate": 2.259928242189966e-05,
      "loss": 1.4798,
      "step": 3630
    },
    {
      "epoch": 0.5733186328555678,
      "grad_norm": 0.8902590274810791,
      "learning_rate": 2.2291677454254136e-05,
      "loss": 1.4934,
      "step": 3640
    },
    {
      "epoch": 0.5748936840447314,
      "grad_norm": 0.8976852297782898,
      "learning_rate": 2.1985725293263237e-05,
      "loss": 1.4773,
      "step": 3650
    },
    {
      "epoch": 0.5764687352338951,
      "grad_norm": 0.8807442784309387,
      "learning_rate": 2.1681438515421953e-05,
      "loss": 1.4888,
      "step": 3660
    },
    {
      "epoch": 0.5780437864230588,
      "grad_norm": 0.861380934715271,
      "learning_rate": 2.1378829628767965e-05,
      "loss": 1.4996,
      "step": 3670
    },
    {
      "epoch": 0.5796188376122224,
      "grad_norm": 0.9131867289543152,
      "learning_rate": 2.1077911072367317e-05,
      "loss": 1.5014,
      "step": 3680
    },
    {
      "epoch": 0.581193888801386,
      "grad_norm": 0.8426468968391418,
      "learning_rate": 2.077869521580325e-05,
      "loss": 1.4832,
      "step": 3690
    },
    {
      "epoch": 0.5827689399905497,
      "grad_norm": 0.8558875918388367,
      "learning_rate": 2.0481194358667695e-05,
      "loss": 1.4796,
      "step": 3700
    },
    {
      "epoch": 0.5843439911797134,
      "grad_norm": 0.8693471550941467,
      "learning_rate": 2.018542073005567e-05,
      "loss": 1.4933,
      "step": 3710
    },
    {
      "epoch": 0.585919042368877,
      "grad_norm": 0.9600664973258972,
      "learning_rate": 1.9891386488062538e-05,
      "loss": 1.5139,
      "step": 3720
    },
    {
      "epoch": 0.5874940935580406,
      "grad_norm": 0.8897980451583862,
      "learning_rate": 1.959910371928436e-05,
      "loss": 1.5171,
      "step": 3730
    },
    {
      "epoch": 0.5890691447472043,
      "grad_norm": 0.8953452110290527,
      "learning_rate": 1.930858443832096e-05,
      "loss": 1.4849,
      "step": 3740
    },
    {
      "epoch": 0.590644195936368,
      "grad_norm": 0.8558352589607239,
      "learning_rate": 1.90198405872821e-05,
      "loss": 1.4922,
      "step": 3750
    },
    {
      "epoch": 0.590644195936368,
      "eval_loss": 1.4844595193862915,
      "eval_runtime": 36.1848,
      "eval_samples_per_second": 13.846,
      "eval_steps_per_second": 13.846,
      "step": 3750
    },
    {
      "epoch": 0.5922192471255315,
      "grad_norm": 0.8563032150268555,
      "learning_rate": 1.8732884035296582e-05,
      "loss": 1.4821,
      "step": 3760
    },
    {
      "epoch": 0.5937942983146952,
      "grad_norm": 0.8764299154281616,
      "learning_rate": 1.844772657802428e-05,
      "loss": 1.4916,
      "step": 3770
    },
    {
      "epoch": 0.5953693495038589,
      "grad_norm": 0.8964916467666626,
      "learning_rate": 1.8164379937171382e-05,
      "loss": 1.4733,
      "step": 3780
    },
    {
      "epoch": 0.5969444006930226,
      "grad_norm": 0.9039140343666077,
      "learning_rate": 1.7882855760008547e-05,
      "loss": 1.5123,
      "step": 3790
    },
    {
      "epoch": 0.5985194518821861,
      "grad_norm": 0.8859601020812988,
      "learning_rate": 1.760316561889203e-05,
      "loss": 1.4859,
      "step": 3800
    },
    {
      "epoch": 0.6000945030713498,
      "grad_norm": 0.8582342267036438,
      "learning_rate": 1.7325321010788034e-05,
      "loss": 1.4856,
      "step": 3810
    },
    {
      "epoch": 0.6016695542605135,
      "grad_norm": 0.8299849629402161,
      "learning_rate": 1.7049333356800167e-05,
      "loss": 1.4852,
      "step": 3820
    },
    {
      "epoch": 0.6032446054496771,
      "grad_norm": 0.8748874068260193,
      "learning_rate": 1.6775214001699914e-05,
      "loss": 1.477,
      "step": 3830
    },
    {
      "epoch": 0.6048196566388407,
      "grad_norm": 0.9403565526008606,
      "learning_rate": 1.6502974213460316e-05,
      "loss": 1.4734,
      "step": 3840
    },
    {
      "epoch": 0.6063947078280044,
      "grad_norm": 0.8735081553459167,
      "learning_rate": 1.623262518279279e-05,
      "loss": 1.5099,
      "step": 3850
    },
    {
      "epoch": 0.6079697590171681,
      "grad_norm": 0.8424332141876221,
      "learning_rate": 1.596417802268707e-05,
      "loss": 1.4852,
      "step": 3860
    },
    {
      "epoch": 0.6095448102063317,
      "grad_norm": 0.8874250650405884,
      "learning_rate": 1.5697643767954488e-05,
      "loss": 1.4814,
      "step": 3870
    },
    {
      "epoch": 0.6111198613954953,
      "grad_norm": 0.9068039059638977,
      "learning_rate": 1.543303337477432e-05,
      "loss": 1.4903,
      "step": 3880
    },
    {
      "epoch": 0.612694912584659,
      "grad_norm": 0.9054557681083679,
      "learning_rate": 1.517035772024343e-05,
      "loss": 1.5071,
      "step": 3890
    },
    {
      "epoch": 0.6142699637738227,
      "grad_norm": 0.8765373229980469,
      "learning_rate": 1.49096276019291e-05,
      "loss": 1.4716,
      "step": 3900
    },
    {
      "epoch": 0.6142699637738227,
      "eval_loss": 1.4831758737564087,
      "eval_runtime": 37.0102,
      "eval_samples_per_second": 13.537,
      "eval_steps_per_second": 13.537,
      "step": 3900
    },
    {
      "epoch": 0.6158450149629863,
      "grad_norm": 0.9288606643676758,
      "learning_rate": 1.4650853737425327e-05,
      "loss": 1.4683,
      "step": 3910
    },
    {
      "epoch": 0.61742006615215,
      "grad_norm": 0.9101633429527283,
      "learning_rate": 1.4394046763912122e-05,
      "loss": 1.4706,
      "step": 3920
    },
    {
      "epoch": 0.6189951173413136,
      "grad_norm": 0.8837119936943054,
      "learning_rate": 1.413921723771832e-05,
      "loss": 1.4799,
      "step": 3930
    },
    {
      "epoch": 0.6205701685304772,
      "grad_norm": 0.962098240852356,
      "learning_rate": 1.3886375633887665e-05,
      "loss": 1.5042,
      "step": 3940
    },
    {
      "epoch": 0.6221452197196409,
      "grad_norm": 0.8238315582275391,
      "learning_rate": 1.3635532345748137e-05,
      "loss": 1.4932,
      "step": 3950
    },
    {
      "epoch": 0.6237202709088046,
      "grad_norm": 0.9068928360939026,
      "learning_rate": 1.3386697684484853e-05,
      "loss": 1.4743,
      "step": 3960
    },
    {
      "epoch": 0.6252953220979682,
      "grad_norm": 0.8136940598487854,
      "learning_rate": 1.3139881878716107e-05,
      "loss": 1.4727,
      "step": 3970
    },
    {
      "epoch": 0.6268703732871318,
      "grad_norm": 0.8796588182449341,
      "learning_rate": 1.2895095074072986e-05,
      "loss": 1.4681,
      "step": 3980
    },
    {
      "epoch": 0.6284454244762955,
      "grad_norm": 0.8927239775657654,
      "learning_rate": 1.2652347332782227e-05,
      "loss": 1.4871,
      "step": 3990
    },
    {
      "epoch": 0.6300204756654592,
      "grad_norm": 0.8667739629745483,
      "learning_rate": 1.2411648633252719e-05,
      "loss": 1.4784,
      "step": 4000
    },
    {
      "epoch": 0.6315955268546227,
      "grad_norm": 0.8735933303833008,
      "learning_rate": 1.2173008869665241e-05,
      "loss": 1.4659,
      "step": 4010
    },
    {
      "epoch": 0.6331705780437864,
      "grad_norm": 0.8920369148254395,
      "learning_rate": 1.1936437851565791e-05,
      "loss": 1.4711,
      "step": 4020
    },
    {
      "epoch": 0.6347456292329501,
      "grad_norm": 0.854413628578186,
      "learning_rate": 1.1701945303462337e-05,
      "loss": 1.4877,
      "step": 4030
    },
    {
      "epoch": 0.6363206804221138,
      "grad_norm": 0.8223803639411926,
      "learning_rate": 1.146954086442508e-05,
      "loss": 1.477,
      "step": 4040
    },
    {
      "epoch": 0.6378957316112773,
      "grad_norm": 0.8744681477546692,
      "learning_rate": 1.1239234087690252e-05,
      "loss": 1.4864,
      "step": 4050
    },
    {
      "epoch": 0.6378957316112773,
      "eval_loss": 1.4826009273529053,
      "eval_runtime": 36.4666,
      "eval_samples_per_second": 13.739,
      "eval_steps_per_second": 13.739,
      "step": 4050
    },
    {
      "epoch": 0.639470782800441,
      "grad_norm": 0.879377543926239,
      "learning_rate": 1.1011034440267395e-05,
      "loss": 1.4908,
      "step": 4060
    },
    {
      "epoch": 0.6410458339896047,
      "grad_norm": 0.8856979608535767,
      "learning_rate": 1.078495130255023e-05,
      "loss": 1.4768,
      "step": 4070
    },
    {
      "epoch": 0.6426208851787684,
      "grad_norm": 0.8525427579879761,
      "learning_rate": 1.0560993967931004e-05,
      "loss": 1.4824,
      "step": 4080
    },
    {
      "epoch": 0.6441959363679319,
      "grad_norm": 0.8850044012069702,
      "learning_rate": 1.0339171642418585e-05,
      "loss": 1.4694,
      "step": 4090
    },
    {
      "epoch": 0.6457709875570956,
      "grad_norm": 0.9544363021850586,
      "learning_rate": 1.0119493444259963e-05,
      "loss": 1.4793,
      "step": 4100
    },
    {
      "epoch": 0.6473460387462593,
      "grad_norm": 0.8178799152374268,
      "learning_rate": 9.901968403565428e-06,
      "loss": 1.4607,
      "step": 4110
    },
    {
      "epoch": 0.6489210899354229,
      "grad_norm": 0.8692899346351624,
      "learning_rate": 9.686605461937441e-06,
      "loss": 1.4893,
      "step": 4120
    },
    {
      "epoch": 0.6504961411245865,
      "grad_norm": 0.8424128293991089,
      "learning_rate": 9.473413472102982e-06,
      "loss": 1.4903,
      "step": 4130
    },
    {
      "epoch": 0.6520711923137502,
      "grad_norm": 0.8360965847969055,
      "learning_rate": 9.262401197549744e-06,
      "loss": 1.4823,
      "step": 4140
    },
    {
      "epoch": 0.6536462435029139,
      "grad_norm": 0.8790252804756165,
      "learning_rate": 9.05357731216587e-06,
      "loss": 1.4884,
      "step": 4150
    },
    {
      "epoch": 0.6552212946920775,
      "grad_norm": 0.8617748022079468,
      "learning_rate": 8.846950399883368e-06,
      "loss": 1.4971,
      "step": 4160
    },
    {
      "epoch": 0.6567963458812411,
      "grad_norm": 0.8690015077590942,
      "learning_rate": 8.64252895432531e-06,
      "loss": 1.4819,
      "step": 4170
    },
    {
      "epoch": 0.6583713970704048,
      "grad_norm": 0.8620494604110718,
      "learning_rate": 8.440321378456656e-06,
      "loss": 1.4722,
      "step": 4180
    },
    {
      "epoch": 0.6599464482595684,
      "grad_norm": 0.8735961318016052,
      "learning_rate": 8.240335984238844e-06,
      "loss": 1.4877,
      "step": 4190
    },
    {
      "epoch": 0.6615214994487321,
      "grad_norm": 0.8974801301956177,
      "learning_rate": 8.042580992288163e-06,
      "loss": 1.4891,
      "step": 4200
    },
    {
      "epoch": 0.6615214994487321,
      "eval_loss": 1.4817373752593994,
      "eval_runtime": 35.5168,
      "eval_samples_per_second": 14.106,
      "eval_steps_per_second": 14.106,
      "step": 4200
    },
    {
      "epoch": 0.6630965506378957,
      "grad_norm": 0.8691059350967407,
      "learning_rate": 7.847064531537774e-06,
      "loss": 1.4908,
      "step": 4210
    },
    {
      "epoch": 0.6646716018270594,
      "grad_norm": 0.8714816570281982,
      "learning_rate": 7.653794638903574e-06,
      "loss": 1.4883,
      "step": 4220
    },
    {
      "epoch": 0.666246653016223,
      "grad_norm": 0.842909574508667,
      "learning_rate": 7.462779258953875e-06,
      "loss": 1.4752,
      "step": 4230
    },
    {
      "epoch": 0.6678217042053867,
      "grad_norm": 0.9124504327774048,
      "learning_rate": 7.274026243582796e-06,
      "loss": 1.4909,
      "step": 4240
    },
    {
      "epoch": 0.6693967553945503,
      "grad_norm": 0.9474470615386963,
      "learning_rate": 7.087543351687493e-06,
      "loss": 1.4917,
      "step": 4250
    },
    {
      "epoch": 0.670971806583714,
      "grad_norm": 0.8785407543182373,
      "learning_rate": 6.903338248849269e-06,
      "loss": 1.4958,
      "step": 4260
    },
    {
      "epoch": 0.6725468577728776,
      "grad_norm": 0.8614616990089417,
      "learning_rate": 6.7214185070183925e-06,
      "loss": 1.498,
      "step": 4270
    },
    {
      "epoch": 0.6741219089620413,
      "grad_norm": 0.8453344106674194,
      "learning_rate": 6.541791604202936e-06,
      "loss": 1.5008,
      "step": 4280
    },
    {
      "epoch": 0.675696960151205,
      "grad_norm": 0.8798742890357971,
      "learning_rate": 6.364464924161311e-06,
      "loss": 1.4749,
      "step": 4290
    },
    {
      "epoch": 0.6772720113403685,
      "grad_norm": 0.8620926737785339,
      "learning_rate": 6.1894457560988106e-06,
      "loss": 1.476,
      "step": 4300
    },
    {
      "epoch": 0.6788470625295322,
      "grad_norm": 0.8776394128799438,
      "learning_rate": 6.016741294367911e-06,
      "loss": 1.4864,
      "step": 4310
    },
    {
      "epoch": 0.6804221137186959,
      "grad_norm": 0.9075036644935608,
      "learning_rate": 5.846358638172615e-06,
      "loss": 1.4901,
      "step": 4320
    },
    {
      "epoch": 0.6819971649078596,
      "grad_norm": 0.8943276405334473,
      "learning_rate": 5.678304791276567e-06,
      "loss": 1.4913,
      "step": 4330
    },
    {
      "epoch": 0.6835722160970231,
      "grad_norm": 0.8560464382171631,
      "learning_rate": 5.51258666171519e-06,
      "loss": 1.4909,
      "step": 4340
    },
    {
      "epoch": 0.6851472672861868,
      "grad_norm": 0.9168397784233093,
      "learning_rate": 5.349211061511726e-06,
      "loss": 1.4596,
      "step": 4350
    },
    {
      "epoch": 0.6851472672861868,
      "eval_loss": 1.4812487363815308,
      "eval_runtime": 35.7079,
      "eval_samples_per_second": 14.031,
      "eval_steps_per_second": 14.031,
      "step": 4350
    },
    {
      "epoch": 0.6867223184753505,
      "grad_norm": 0.9016767144203186,
      "learning_rate": 5.188184706397182e-06,
      "loss": 1.4719,
      "step": 4360
    },
    {
      "epoch": 0.688297369664514,
      "grad_norm": 0.893745481967926,
      "learning_rate": 5.029514215534339e-06,
      "loss": 1.4829,
      "step": 4370
    },
    {
      "epoch": 0.6898724208536777,
      "grad_norm": 0.8733708262443542,
      "learning_rate": 4.873206111245594e-06,
      "loss": 1.4942,
      "step": 4380
    },
    {
      "epoch": 0.6914474720428414,
      "grad_norm": 0.8753581047058105,
      "learning_rate": 4.719266818744912e-06,
      "loss": 1.4839,
      "step": 4390
    },
    {
      "epoch": 0.6930225232320051,
      "grad_norm": 0.8269835710525513,
      "learning_rate": 4.567702665873648e-06,
      "loss": 1.4615,
      "step": 4400
    },
    {
      "epoch": 0.6945975744211687,
      "grad_norm": 0.9257881045341492,
      "learning_rate": 4.418519882840505e-06,
      "loss": 1.4939,
      "step": 4410
    },
    {
      "epoch": 0.6961726256103323,
      "grad_norm": 0.8763148188591003,
      "learning_rate": 4.271724601965371e-06,
      "loss": 1.4771,
      "step": 4420
    },
    {
      "epoch": 0.697747676799496,
      "grad_norm": 0.8695856332778931,
      "learning_rate": 4.127322857427306e-06,
      "loss": 1.4985,
      "step": 4430
    },
    {
      "epoch": 0.6993227279886596,
      "grad_norm": 0.8823320269584656,
      "learning_rate": 3.985320585016425e-06,
      "loss": 1.4825,
      "step": 4440
    },
    {
      "epoch": 0.7008977791778233,
      "grad_norm": 0.8587324619293213,
      "learning_rate": 3.845723621889973e-06,
      "loss": 1.4541,
      "step": 4450
    },
    {
      "epoch": 0.7024728303669869,
      "grad_norm": 0.8840261101722717,
      "learning_rate": 3.7085377063323447e-06,
      "loss": 1.4696,
      "step": 4460
    },
    {
      "epoch": 0.7040478815561506,
      "grad_norm": 0.9190268516540527,
      "learning_rate": 3.5737684775191887e-06,
      "loss": 1.4842,
      "step": 4470
    },
    {
      "epoch": 0.7056229327453142,
      "grad_norm": 0.8872419595718384,
      "learning_rate": 3.441421475285679e-06,
      "loss": 1.4892,
      "step": 4480
    },
    {
      "epoch": 0.7071979839344779,
      "grad_norm": 0.8445181250572205,
      "learning_rate": 3.3115021398986768e-06,
      "loss": 1.4875,
      "step": 4490
    },
    {
      "epoch": 0.7087730351236415,
      "grad_norm": 0.83912193775177,
      "learning_rate": 3.18401581183321e-06,
      "loss": 1.4774,
      "step": 4500
    },
    {
      "epoch": 0.7087730351236415,
      "eval_loss": 1.480732798576355,
      "eval_runtime": 37.196,
      "eval_samples_per_second": 13.469,
      "eval_steps_per_second": 13.469,
      "step": 4500
    },
    {
      "epoch": 0.7103480863128052,
      "grad_norm": 0.8629345893859863,
      "learning_rate": 3.0589677315529044e-06,
      "loss": 1.4749,
      "step": 4510
    },
    {
      "epoch": 0.7119231375019688,
      "grad_norm": 0.8621689677238464,
      "learning_rate": 2.9363630392945513e-06,
      "loss": 1.4772,
      "step": 4520
    },
    {
      "epoch": 0.7134981886911325,
      "grad_norm": 0.873957633972168,
      "learning_rate": 2.816206774856854e-06,
      "loss": 1.4937,
      "step": 4530
    },
    {
      "epoch": 0.7150732398802961,
      "grad_norm": 0.9086722135543823,
      "learning_rate": 2.6985038773932046e-06,
      "loss": 1.4512,
      "step": 4540
    },
    {
      "epoch": 0.7166482910694597,
      "grad_norm": 0.8475430607795715,
      "learning_rate": 2.583259185208714e-06,
      "loss": 1.4621,
      "step": 4550
    },
    {
      "epoch": 0.7182233422586234,
      "grad_norm": 0.8581358790397644,
      "learning_rate": 2.4704774355612943e-06,
      "loss": 1.4746,
      "step": 4560
    },
    {
      "epoch": 0.7197983934477871,
      "grad_norm": 0.8703014254570007,
      "learning_rate": 2.3601632644669536e-06,
      "loss": 1.4906,
      "step": 4570
    },
    {
      "epoch": 0.7213734446369507,
      "grad_norm": 0.9130226373672485,
      "learning_rate": 2.2523212065091723e-06,
      "loss": 1.4825,
      "step": 4580
    },
    {
      "epoch": 0.7229484958261143,
      "grad_norm": 0.8566045165061951,
      "learning_rate": 2.1469556946525706e-06,
      "loss": 1.4732,
      "step": 4590
    },
    {
      "epoch": 0.724523547015278,
      "grad_norm": 0.8569677472114563,
      "learning_rate": 2.0440710600606595e-06,
      "loss": 1.4725,
      "step": 4600
    },
    {
      "epoch": 0.7260985982044417,
      "grad_norm": 0.9196586012840271,
      "learning_rate": 1.9436715319177956e-06,
      "loss": 1.4832,
      "step": 4610
    },
    {
      "epoch": 0.7276736493936052,
      "grad_norm": 0.8641564249992371,
      "learning_rate": 1.8457612372553348e-06,
      "loss": 1.4994,
      "step": 4620
    },
    {
      "epoch": 0.7292487005827689,
      "grad_norm": 0.8218653798103333,
      "learning_rate": 1.75034420078201e-06,
      "loss": 1.4748,
      "step": 4630
    },
    {
      "epoch": 0.7308237517719326,
      "grad_norm": 0.8673424124717712,
      "learning_rate": 1.6574243447184597e-06,
      "loss": 1.4779,
      "step": 4640
    },
    {
      "epoch": 0.7323988029610963,
      "grad_norm": 0.8800205588340759,
      "learning_rate": 1.567005488636024e-06,
      "loss": 1.4927,
      "step": 4650
    },
    {
      "epoch": 0.7323988029610963,
      "eval_loss": 1.4804431200027466,
      "eval_runtime": 37.6922,
      "eval_samples_per_second": 13.292,
      "eval_steps_per_second": 13.292,
      "step": 4650
    },
    {
      "epoch": 0.7339738541502598,
      "grad_norm": 0.8832563757896423,
      "learning_rate": 1.4790913492997438e-06,
      "loss": 1.5031,
      "step": 4660
    },
    {
      "epoch": 0.7355489053394235,
      "grad_norm": 0.8596001863479614,
      "learning_rate": 1.3936855405155408e-06,
      "loss": 1.4974,
      "step": 4670
    },
    {
      "epoch": 0.7371239565285872,
      "grad_norm": 0.8389458656311035,
      "learning_rate": 1.3107915729816954e-06,
      "loss": 1.4873,
      "step": 4680
    },
    {
      "epoch": 0.7386990077177509,
      "grad_norm": 0.8755994439125061,
      "learning_rate": 1.230412854144547e-06,
      "loss": 1.4986,
      "step": 4690
    },
    {
      "epoch": 0.7402740589069144,
      "grad_norm": 0.8368931412696838,
      "learning_rate": 1.15255268805841e-06,
      "loss": 1.482,
      "step": 4700
    },
    {
      "epoch": 0.7418491100960781,
      "grad_norm": 0.853042721748352,
      "learning_rate": 1.0772142752497604e-06,
      "loss": 1.4824,
      "step": 4710
    },
    {
      "epoch": 0.7434241612852418,
      "grad_norm": 0.8514347672462463,
      "learning_rate": 1.004400712585646e-06,
      "loss": 1.4737,
      "step": 4720
    },
    {
      "epoch": 0.7449992124744054,
      "grad_norm": 0.8187866806983948,
      "learning_rate": 9.341149931464537e-07,
      "loss": 1.5147,
      "step": 4730
    },
    {
      "epoch": 0.746574263663569,
      "grad_norm": 0.8992719054222107,
      "learning_rate": 8.663600061028162e-07,
      "loss": 1.487,
      "step": 4740
    },
    {
      "epoch": 0.7481493148527327,
      "grad_norm": 0.834426999092102,
      "learning_rate": 8.011385365968641e-07,
      "loss": 1.4814,
      "step": 4750
    },
    {
      "epoch": 0.7497243660418964,
      "grad_norm": 0.8676533699035645,
      "learning_rate": 7.384532656277698e-07,
      "loss": 1.4703,
      "step": 4760
    },
    {
      "epoch": 0.75129941723106,
      "grad_norm": 0.8450558185577393,
      "learning_rate": 6.783067699414891e-07,
      "loss": 1.4648,
      "step": 4770
    },
    {
      "epoch": 0.7528744684202237,
      "grad_norm": 0.846495509147644,
      "learning_rate": 6.207015219248866e-07,
      "loss": 1.4792,
      "step": 4780
    },
    {
      "epoch": 0.7544495196093873,
      "grad_norm": 0.8761216998100281,
      "learning_rate": 5.656398895040813e-07,
      "loss": 1.4552,
      "step": 4790
    },
    {
      "epoch": 0.7560245707985509,
      "grad_norm": 0.8825677037239075,
      "learning_rate": 5.131241360471217e-07,
      "loss": 1.4959,
      "step": 4800
    },
    {
      "epoch": 0.7560245707985509,
      "eval_loss": 1.480470061302185,
      "eval_runtime": 35.9266,
      "eval_samples_per_second": 13.945,
      "eval_steps_per_second": 13.945,
      "step": 4800
    },
    {
      "epoch": 0.7575996219877146,
      "grad_norm": 0.8354045748710632,
      "learning_rate": 4.631564202709354e-07,
      "loss": 1.4743,
      "step": 4810
    },
    {
      "epoch": 0.7591746731768783,
      "grad_norm": 0.8703035712242126,
      "learning_rate": 4.1573879615262185e-07,
      "loss": 1.4686,
      "step": 4820
    },
    {
      "epoch": 0.7607497243660419,
      "grad_norm": 0.929268479347229,
      "learning_rate": 3.708732128449785e-07,
      "loss": 1.4695,
      "step": 4830
    },
    {
      "epoch": 0.7623247755552055,
      "grad_norm": 0.8741073608398438,
      "learning_rate": 3.2856151459641216e-07,
      "loss": 1.492,
      "step": 4840
    },
    {
      "epoch": 0.7638998267443692,
      "grad_norm": 0.8539476990699768,
      "learning_rate": 2.888054406751106e-07,
      "loss": 1.4913,
      "step": 4850
    },
    {
      "epoch": 0.7654748779335329,
      "grad_norm": 0.8607631325721741,
      "learning_rate": 2.5160662529755823e-07,
      "loss": 1.4712,
      "step": 4860
    },
    {
      "epoch": 0.7670499291226965,
      "grad_norm": 0.856203556060791,
      "learning_rate": 2.169665975613605e-07,
      "loss": 1.4829,
      "step": 4870
    },
    {
      "epoch": 0.7686249803118601,
      "grad_norm": 0.8610185384750366,
      "learning_rate": 1.8488678138238456e-07,
      "loss": 1.4983,
      "step": 4880
    },
    {
      "epoch": 0.7702000315010238,
      "grad_norm": 0.8597480058670044,
      "learning_rate": 1.5536849543621584e-07,
      "loss": 1.4763,
      "step": 4890
    },
    {
      "epoch": 0.7717750826901875,
      "grad_norm": 0.8443917632102966,
      "learning_rate": 1.2841295310397905e-07,
      "loss": 1.4961,
      "step": 4900
    },
    {
      "epoch": 0.773350133879351,
      "grad_norm": 0.8755921125411987,
      "learning_rate": 1.0402126242244764e-07,
      "loss": 1.4802,
      "step": 4910
    },
    {
      "epoch": 0.7749251850685147,
      "grad_norm": 0.888160765171051,
      "learning_rate": 8.219442603847605e-08,
      "loss": 1.4561,
      "step": 4920
    },
    {
      "epoch": 0.7765002362576784,
      "grad_norm": 0.842492938041687,
      "learning_rate": 6.293334116783817e-08,
      "loss": 1.4527,
      "step": 4930
    },
    {
      "epoch": 0.7780752874468421,
      "grad_norm": 0.886648952960968,
      "learning_rate": 4.623879955827082e-08,
      "loss": 1.489,
      "step": 4940
    },
    {
      "epoch": 0.7796503386360056,
      "grad_norm": 0.8370558023452759,
      "learning_rate": 3.211148745700665e-08,
      "loss": 1.4815,
      "step": 4950
    },
    {
      "epoch": 0.7796503386360056,
      "eval_loss": 1.4804972410202026,
      "eval_runtime": 35.7603,
      "eval_samples_per_second": 14.01,
      "eval_steps_per_second": 14.01,
      "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": 5.149189940332462e+17,
  "train_batch_size": 16,
  "trial_name": null,
  "trial_params": null
}