{
  "best_metric": 0.32636332511901855,
  "best_model_checkpoint": "./output/checkpoint-3600",
  "epoch": 0.2544385389573674,
  "eval_steps": 150,
  "global_step": 4500,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0005654189754608164,
      "grad_norm": 2.227701425552368,
      "learning_rate": 7.500000000000001e-07,
      "loss": 0.1995,
      "step": 10
    },
    {
      "epoch": 0.0011308379509216329,
      "grad_norm": 5.525130748748779,
      "learning_rate": 1.5000000000000002e-06,
      "loss": 0.2334,
      "step": 20
    },
    {
      "epoch": 0.0016962569263824494,
      "grad_norm": 1.4867887496948242,
      "learning_rate": 2.25e-06,
      "loss": 0.3133,
      "step": 30
    },
    {
      "epoch": 0.0022616759018432657,
      "grad_norm": 0.5477761626243591,
      "learning_rate": 3.0000000000000005e-06,
      "loss": 0.1692,
      "step": 40
    },
    {
      "epoch": 0.0028270948773040823,
      "grad_norm": 1.7914361953735352,
      "learning_rate": 3.7500000000000005e-06,
      "loss": 0.3082,
      "step": 50
    },
    {
      "epoch": 0.003392513852764899,
      "grad_norm": 0.40979117155075073,
      "learning_rate": 4.5e-06,
      "loss": 0.1468,
      "step": 60
    },
    {
      "epoch": 0.003957932828225715,
      "grad_norm": 5.27268123626709,
      "learning_rate": 5.2500000000000006e-06,
      "loss": 0.2381,
      "step": 70
    },
    {
      "epoch": 0.0045233518036865315,
      "grad_norm": 8.342147827148438,
      "learning_rate": 6.000000000000001e-06,
      "loss": 0.2292,
      "step": 80
    },
    {
      "epoch": 0.0050887707791473485,
      "grad_norm": 4.8586745262146,
      "learning_rate": 6.7500000000000014e-06,
      "loss": 0.1351,
      "step": 90
    },
    {
      "epoch": 0.005654189754608165,
      "grad_norm": 7.848427772521973,
      "learning_rate": 7.500000000000001e-06,
      "loss": 0.2967,
      "step": 100
    },
    {
      "epoch": 0.006219608730068981,
      "grad_norm": 0.6273432374000549,
      "learning_rate": 7.499922926093874e-06,
      "loss": 0.1445,
      "step": 110
    },
    {
      "epoch": 0.006785027705529798,
      "grad_norm": 0.21901638805866241,
      "learning_rate": 7.499691707543699e-06,
      "loss": 0.2228,
      "step": 120
    },
    {
      "epoch": 0.007350446680990614,
      "grad_norm": 0.5449599027633667,
      "learning_rate": 7.499306353853963e-06,
      "loss": 0.2675,
      "step": 130
    },
    {
      "epoch": 0.00791586565645143,
      "grad_norm": 0.5028505921363831,
      "learning_rate": 7.49876688086505e-06,
      "loss": 0.1786,
      "step": 140
    },
    {
      "epoch": 0.008481284631912247,
      "grad_norm": 1.582517385482788,
      "learning_rate": 7.4980733107525805e-06,
      "loss": 0.2159,
      "step": 150
    },
    {
      "epoch": 0.008481284631912247,
      "eval_loss": 0.4814591705799103,
      "eval_runtime": 39.6064,
      "eval_samples_per_second": 12.624,
      "eval_steps_per_second": 12.624,
      "step": 150
    },
    {
      "epoch": 0.009046703607373063,
      "grad_norm": 1.0850478410720825,
      "learning_rate": 7.4972256720265044e-06,
      "loss": 0.213,
      "step": 160
    },
    {
      "epoch": 0.009612122582833879,
      "grad_norm": 13.114879608154297,
      "learning_rate": 7.496223999529932e-06,
      "loss": 0.2911,
      "step": 170
    },
    {
      "epoch": 0.010177541558294697,
      "grad_norm": 1.2058771848678589,
      "learning_rate": 7.4950683344376926e-06,
      "loss": 0.3511,
      "step": 180
    },
    {
      "epoch": 0.010742960533755513,
      "grad_norm": 0.21950186789035797,
      "learning_rate": 7.4937587242546544e-06,
      "loss": 0.2356,
      "step": 190
    },
    {
      "epoch": 0.01130837950921633,
      "grad_norm": 3.904442548751831,
      "learning_rate": 7.492295222813762e-06,
      "loss": 0.1941,
      "step": 200
    },
    {
      "epoch": 0.011873798484677145,
      "grad_norm": 8.762584686279297,
      "learning_rate": 7.490677890273828e-06,
      "loss": 0.2477,
      "step": 210
    },
    {
      "epoch": 0.012439217460137961,
      "grad_norm": 2.0331380367279053,
      "learning_rate": 7.488906793117058e-06,
      "loss": 0.2963,
      "step": 220
    },
    {
      "epoch": 0.01300463643559878,
      "grad_norm": 0.952492892742157,
      "learning_rate": 7.486982004146319e-06,
      "loss": 0.1924,
      "step": 230
    },
    {
      "epoch": 0.013570055411059595,
      "grad_norm": 2.3315181732177734,
      "learning_rate": 7.484903602482148e-06,
      "loss": 0.2643,
      "step": 240
    },
    {
      "epoch": 0.014135474386520411,
      "grad_norm": 4.806341648101807,
      "learning_rate": 7.4826716735594945e-06,
      "loss": 0.2789,
      "step": 250
    },
    {
      "epoch": 0.014700893361981228,
      "grad_norm": 5.050877571105957,
      "learning_rate": 7.480286309124216e-06,
      "loss": 0.2581,
      "step": 260
    },
    {
      "epoch": 0.015266312337442044,
      "grad_norm": 6.544984340667725,
      "learning_rate": 7.477747607229302e-06,
      "loss": 0.2921,
      "step": 270
    },
    {
      "epoch": 0.01583173131290286,
      "grad_norm": 6.630581378936768,
      "learning_rate": 7.475055672230844e-06,
      "loss": 0.254,
      "step": 280
    },
    {
      "epoch": 0.016397150288363676,
      "grad_norm": 8.29354190826416,
      "learning_rate": 7.472210614783745e-06,
      "loss": 0.3359,
      "step": 290
    },
    {
      "epoch": 0.016962569263824494,
      "grad_norm": 0.3115900158882141,
      "learning_rate": 7.469212551837173e-06,
      "loss": 0.1632,
      "step": 300
    },
    {
      "epoch": 0.016962569263824494,
      "eval_loss": 0.4433949291706085,
      "eval_runtime": 39.7798,
      "eval_samples_per_second": 12.569,
      "eval_steps_per_second": 12.569,
      "step": 300
    },
    {
      "epoch": 0.01752798823928531,
      "grad_norm": 0.5275546908378601,
      "learning_rate": 7.4660616066297565e-06,
      "loss": 0.2717,
      "step": 310
    },
    {
      "epoch": 0.018093407214746126,
      "grad_norm": 10.666130065917969,
      "learning_rate": 7.462757908684509e-06,
      "loss": 0.2823,
      "step": 320
    },
    {
      "epoch": 0.018658826190206944,
      "grad_norm": 6.284046173095703,
      "learning_rate": 7.459301593803512e-06,
      "loss": 0.1959,
      "step": 330
    },
    {
      "epoch": 0.019224245165667758,
      "grad_norm": 0.3211212456226349,
      "learning_rate": 7.455692804062335e-06,
      "loss": 0.1922,
      "step": 340
    },
    {
      "epoch": 0.019789664141128576,
      "grad_norm": 1.0060560703277588,
      "learning_rate": 7.451931687804189e-06,
      "loss": 0.3553,
      "step": 350
    },
    {
      "epoch": 0.020355083116589394,
      "grad_norm": 10.29517650604248,
      "learning_rate": 7.448018399633831e-06,
      "loss": 0.2307,
      "step": 360
    },
    {
      "epoch": 0.02092050209205021,
      "grad_norm": 1.2054423093795776,
      "learning_rate": 7.443953100411214e-06,
      "loss": 0.1897,
      "step": 370
    },
    {
      "epoch": 0.021485921067511026,
      "grad_norm": 7.133542537689209,
      "learning_rate": 7.439735957244862e-06,
      "loss": 0.2051,
      "step": 380
    },
    {
      "epoch": 0.02205134004297184,
      "grad_norm": 1.6390613317489624,
      "learning_rate": 7.435367143485015e-06,
      "loss": 0.2781,
      "step": 390
    },
    {
      "epoch": 0.02261675901843266,
      "grad_norm": 0.9467723965644836,
      "learning_rate": 7.430846838716496e-06,
      "loss": 0.1886,
      "step": 400
    },
    {
      "epoch": 0.023182177993893476,
      "grad_norm": 5.297723770141602,
      "learning_rate": 7.426175228751328e-06,
      "loss": 0.2828,
      "step": 410
    },
    {
      "epoch": 0.02374759696935429,
      "grad_norm": 0.396626353263855,
      "learning_rate": 7.421352505621099e-06,
      "loss": 0.294,
      "step": 420
    },
    {
      "epoch": 0.02431301594481511,
      "grad_norm": 0.4418993592262268,
      "learning_rate": 7.416378867569069e-06,
      "loss": 0.217,
      "step": 430
    },
    {
      "epoch": 0.024878434920275923,
      "grad_norm": 0.9995683431625366,
      "learning_rate": 7.411254519042017e-06,
      "loss": 0.109,
      "step": 440
    },
    {
      "epoch": 0.02544385389573674,
      "grad_norm": 2.574638843536377,
      "learning_rate": 7.4059796706818396e-06,
      "loss": 0.1732,
      "step": 450
    },
    {
      "epoch": 0.02544385389573674,
      "eval_loss": 0.4305788278579712,
      "eval_runtime": 39.9497,
      "eval_samples_per_second": 12.516,
      "eval_steps_per_second": 12.516,
      "step": 450
    },
    {
      "epoch": 0.02600927287119756,
      "grad_norm": 4.463107109069824,
      "learning_rate": 7.400554539316894e-06,
      "loss": 0.2522,
      "step": 460
    },
    {
      "epoch": 0.026574691846658373,
      "grad_norm": 0.5263558030128479,
      "learning_rate": 7.394979347953081e-06,
      "loss": 0.3847,
      "step": 470
    },
    {
      "epoch": 0.02714011082211919,
      "grad_norm": 0.4478251039981842,
      "learning_rate": 7.389254325764681e-06,
      "loss": 0.2012,
      "step": 480
    },
    {
      "epoch": 0.027705529797580005,
      "grad_norm": 6.665970325469971,
      "learning_rate": 7.383379708084934e-06,
      "loss": 0.255,
      "step": 490
    },
    {
      "epoch": 0.028270948773040823,
      "grad_norm": 0.7272936701774597,
      "learning_rate": 7.377355736396362e-06,
      "loss": 0.1486,
      "step": 500
    },
    {
      "epoch": 0.02883636774850164,
      "grad_norm": 0.3776552081108093,
      "learning_rate": 7.371182658320847e-06,
      "loss": 0.2023,
      "step": 510
    },
    {
      "epoch": 0.029401786723962455,
      "grad_norm": 0.8645554184913635,
      "learning_rate": 7.36486072760945e-06,
      "loss": 0.1617,
      "step": 520
    },
    {
      "epoch": 0.029967205699423273,
      "grad_norm": 6.2824387550354,
      "learning_rate": 7.358390204131984e-06,
      "loss": 0.3383,
      "step": 530
    },
    {
      "epoch": 0.030532624674884087,
      "grad_norm": 0.8180729150772095,
      "learning_rate": 7.3517713538663235e-06,
      "loss": 0.2147,
      "step": 540
    },
    {
      "epoch": 0.031098043650344905,
      "grad_norm": 0.5693274736404419,
      "learning_rate": 7.345004448887478e-06,
      "loss": 0.2348,
      "step": 550
    },
    {
      "epoch": 0.03166346262580572,
      "grad_norm": 4.538175582885742,
      "learning_rate": 7.3380897673564085e-06,
      "loss": 0.3281,
      "step": 560
    },
    {
      "epoch": 0.03222888160126654,
      "grad_norm": 4.282591819763184,
      "learning_rate": 7.33102759350859e-06,
      "loss": 0.2727,
      "step": 570
    },
    {
      "epoch": 0.03279430057672735,
      "grad_norm": 0.8115090727806091,
      "learning_rate": 7.323818217642328e-06,
      "loss": 0.2479,
      "step": 580
    },
    {
      "epoch": 0.03335971955218817,
      "grad_norm": 3.089582681655884,
      "learning_rate": 7.316461936106827e-06,
      "loss": 0.195,
      "step": 590
    },
    {
      "epoch": 0.03392513852764899,
      "grad_norm": 10.888387680053711,
      "learning_rate": 7.3089590512900084e-06,
      "loss": 0.1643,
      "step": 600
    },
    {
      "epoch": 0.03392513852764899,
      "eval_loss": 0.425952672958374,
      "eval_runtime": 39.7561,
      "eval_samples_per_second": 12.577,
      "eval_steps_per_second": 12.577,
      "step": 600
    },
    {
      "epoch": 0.0344905575031098,
      "grad_norm": 6.550843715667725,
      "learning_rate": 7.301309871606081e-06,
      "loss": 0.3565,
      "step": 610
    },
    {
      "epoch": 0.03505597647857062,
      "grad_norm": 1.943982720375061,
      "learning_rate": 7.293514711482861e-06,
      "loss": 0.2715,
      "step": 620
    },
    {
      "epoch": 0.03562139545403144,
      "grad_norm": 4.950693607330322,
      "learning_rate": 7.285573891348849e-06,
      "loss": 0.2719,
      "step": 630
    },
    {
      "epoch": 0.03618681442949225,
      "grad_norm": 7.106111526489258,
      "learning_rate": 7.27748773762006e-06,
      "loss": 0.287,
      "step": 640
    },
    {
      "epoch": 0.03675223340495307,
      "grad_norm": 3.4435412883758545,
      "learning_rate": 7.269256582686603e-06,
      "loss": 0.2495,
      "step": 650
    },
    {
      "epoch": 0.03731765238041389,
      "grad_norm": 3.9263601303100586,
      "learning_rate": 7.260880764899016e-06,
      "loss": 0.2317,
      "step": 660
    },
    {
      "epoch": 0.0378830713558747,
      "grad_norm": 1.1376698017120361,
      "learning_rate": 7.252360628554363e-06,
      "loss": 0.138,
      "step": 670
    },
    {
      "epoch": 0.038448490331335516,
      "grad_norm": 7.205196857452393,
      "learning_rate": 7.243696523882079e-06,
      "loss": 0.1982,
      "step": 680
    },
    {
      "epoch": 0.03901390930679634,
      "grad_norm": 3.7006053924560547,
      "learning_rate": 7.2348888070295705e-06,
      "loss": 0.2156,
      "step": 690
    },
    {
      "epoch": 0.03957932828225715,
      "grad_norm": 0.5304602384567261,
      "learning_rate": 7.225937840047583e-06,
      "loss": 0.3153,
      "step": 700
    },
    {
      "epoch": 0.040144747257717966,
      "grad_norm": 14.555486679077148,
      "learning_rate": 7.216843990875307e-06,
      "loss": 0.3455,
      "step": 710
    },
    {
      "epoch": 0.04071016623317879,
      "grad_norm": 20.35503578186035,
      "learning_rate": 7.207607633325266e-06,
      "loss": 0.2996,
      "step": 720
    },
    {
      "epoch": 0.0412755852086396,
      "grad_norm": 0.4252071678638458,
      "learning_rate": 7.198229147067941e-06,
      "loss": 0.2781,
      "step": 730
    },
    {
      "epoch": 0.04184100418410042,
      "grad_norm": 0.641488790512085,
      "learning_rate": 7.18870891761617e-06,
      "loss": 0.1364,
      "step": 740
    },
    {
      "epoch": 0.04240642315956124,
      "grad_norm": 7.14177942276001,
      "learning_rate": 7.1790473363092974e-06,
      "loss": 0.2639,
      "step": 750
    },
    {
      "epoch": 0.04240642315956124,
      "eval_loss": 0.4108331799507141,
      "eval_runtime": 39.6561,
      "eval_samples_per_second": 12.608,
      "eval_steps_per_second": 12.608,
      "step": 750
    },
    {
      "epoch": 0.04297184213502205,
      "grad_norm": 2.8823704719543457,
      "learning_rate": 7.169244800297089e-06,
      "loss": 0.2308,
      "step": 760
    },
    {
      "epoch": 0.04353726111048287,
      "grad_norm": 0.27674323320388794,
      "learning_rate": 7.159301712523407e-06,
      "loss": 0.2142,
      "step": 770
    },
    {
      "epoch": 0.04410268008594368,
      "grad_norm": 3.091693162918091,
      "learning_rate": 7.149218481709644e-06,
      "loss": 0.2864,
      "step": 780
    },
    {
      "epoch": 0.0446680990614045,
      "grad_norm": 2.583334445953369,
      "learning_rate": 7.1389955223379266e-06,
      "loss": 0.1566,
      "step": 790
    },
    {
      "epoch": 0.04523351803686532,
      "grad_norm": 1.4404966831207275,
      "learning_rate": 7.128633254634072e-06,
      "loss": 0.2206,
      "step": 800
    },
    {
      "epoch": 0.04579893701232613,
      "grad_norm": 0.3449786603450775,
      "learning_rate": 7.118132104550322e-06,
      "loss": 0.0562,
      "step": 810
    },
    {
      "epoch": 0.04636435598778695,
      "grad_norm": 1.9726862907409668,
      "learning_rate": 7.107492503747826e-06,
      "loss": 0.1539,
      "step": 820
    },
    {
      "epoch": 0.04692977496324777,
      "grad_norm": 6.304052829742432,
      "learning_rate": 7.096714889578898e-06,
      "loss": 0.2936,
      "step": 830
    },
    {
      "epoch": 0.04749519393870858,
      "grad_norm": 0.7741239070892334,
      "learning_rate": 7.085799705069046e-06,
      "loss": 0.3401,
      "step": 840
    },
    {
      "epoch": 0.0480606129141694,
      "grad_norm": 2.764857053756714,
      "learning_rate": 7.0747473988987515e-06,
      "loss": 0.2143,
      "step": 850
    },
    {
      "epoch": 0.04862603188963022,
      "grad_norm": 11.133625030517578,
      "learning_rate": 7.063558425385033e-06,
      "loss": 0.3402,
      "step": 860
    },
    {
      "epoch": 0.04919145086509103,
      "grad_norm": 3.349154233932495,
      "learning_rate": 7.052233244462769e-06,
      "loss": 0.155,
      "step": 870
    },
    {
      "epoch": 0.049756869840551846,
      "grad_norm": 5.708688735961914,
      "learning_rate": 7.040772321665788e-06,
      "loss": 0.1933,
      "step": 880
    },
    {
      "epoch": 0.05032228881601267,
      "grad_norm": 10.529179573059082,
      "learning_rate": 7.029176128107734e-06,
      "loss": 0.2543,
      "step": 890
    },
    {
      "epoch": 0.05088770779147348,
      "grad_norm": 0.6206080913543701,
      "learning_rate": 7.017445140462711e-06,
      "loss": 0.1855,
      "step": 900
    },
    {
      "epoch": 0.05088770779147348,
      "eval_loss": 0.38634100556373596,
      "eval_runtime": 39.5281,
      "eval_samples_per_second": 12.649,
      "eval_steps_per_second": 12.649,
      "step": 900
    },
    {
      "epoch": 0.051453126766934296,
      "grad_norm": 0.795102059841156,
      "learning_rate": 7.00557984094567e-06,
      "loss": 0.2033,
      "step": 910
    },
    {
      "epoch": 0.05201854574239512,
      "grad_norm": 4.390031814575195,
      "learning_rate": 6.993580717292601e-06,
      "loss": 0.2188,
      "step": 920
    },
    {
      "epoch": 0.05258396471785593,
      "grad_norm": 0.5030427575111389,
      "learning_rate": 6.981448262740483e-06,
      "loss": 0.1603,
      "step": 930
    },
    {
      "epoch": 0.053149383693316746,
      "grad_norm": 5.4078288078308105,
      "learning_rate": 6.969182976006999e-06,
      "loss": 0.2024,
      "step": 940
    },
    {
      "epoch": 0.05371480266877757,
      "grad_norm": 3.2166638374328613,
      "learning_rate": 6.95678536127005e-06,
      "loss": 0.3025,
      "step": 950
    },
    {
      "epoch": 0.05428022164423838,
      "grad_norm": 1.07512629032135,
      "learning_rate": 6.944255928147017e-06,
      "loss": 0.1517,
      "step": 960
    },
    {
      "epoch": 0.054845640619699196,
      "grad_norm": 13.33733081817627,
      "learning_rate": 6.931595191673823e-06,
      "loss": 0.262,
      "step": 970
    },
    {
      "epoch": 0.05541105959516001,
      "grad_norm": 5.074169635772705,
      "learning_rate": 6.9188036722837555e-06,
      "loss": 0.2072,
      "step": 980
    },
    {
      "epoch": 0.05597647857062083,
      "grad_norm": 2.5820138454437256,
      "learning_rate": 6.905881895786076e-06,
      "loss": 0.2312,
      "step": 990
    },
    {
      "epoch": 0.056541897546081646,
      "grad_norm": 4.5758819580078125,
      "learning_rate": 6.892830393344403e-06,
      "loss": 0.2633,
      "step": 1000
    },
    {
      "epoch": 0.05710731652154246,
      "grad_norm": 0.38114678859710693,
      "learning_rate": 6.879649701454886e-06,
      "loss": 0.0931,
      "step": 1010
    },
    {
      "epoch": 0.05767273549700328,
      "grad_norm": 6.1940131187438965,
      "learning_rate": 6.866340361924141e-06,
      "loss": 0.2923,
      "step": 1020
    },
    {
      "epoch": 0.058238154472464096,
      "grad_norm": 0.355970174074173,
      "learning_rate": 6.852902921846988e-06,
      "loss": 0.257,
      "step": 1030
    },
    {
      "epoch": 0.05880357344792491,
      "grad_norm": 5.869740962982178,
      "learning_rate": 6.8393379335839565e-06,
      "loss": 0.2529,
      "step": 1040
    },
    {
      "epoch": 0.05936899242338573,
      "grad_norm": 4.137405872344971,
      "learning_rate": 6.825645954738586e-06,
      "loss": 0.1663,
      "step": 1050
    },
    {
      "epoch": 0.05936899242338573,
      "eval_loss": 0.3912532329559326,
      "eval_runtime": 39.5493,
      "eval_samples_per_second": 12.642,
      "eval_steps_per_second": 12.642,
      "step": 1050
    },
    {
      "epoch": 0.059934411398846546,
      "grad_norm": 0.35506558418273926,
      "learning_rate": 6.811827548134495e-06,
      "loss": 0.3231,
      "step": 1060
    },
    {
      "epoch": 0.06049983037430736,
      "grad_norm": 1.4487513303756714,
      "learning_rate": 6.797883281792261e-06,
      "loss": 0.2067,
      "step": 1070
    },
    {
      "epoch": 0.061065249349768175,
      "grad_norm": 2.257608652114868,
      "learning_rate": 6.783813728906054e-06,
      "loss": 0.2776,
      "step": 1080
    },
    {
      "epoch": 0.061630668325228996,
      "grad_norm": 2.917322874069214,
      "learning_rate": 6.769619467820086e-06,
      "loss": 0.2023,
      "step": 1090
    },
    {
      "epoch": 0.06219608730068981,
      "grad_norm": 1.0882151126861572,
      "learning_rate": 6.755301082004838e-06,
      "loss": 0.1002,
      "step": 1100
    },
    {
      "epoch": 0.06276150627615062,
      "grad_norm": 3.259073257446289,
      "learning_rate": 6.740859160033068e-06,
      "loss": 0.1888,
      "step": 1110
    },
    {
      "epoch": 0.06332692525161145,
      "grad_norm": 5.543813228607178,
      "learning_rate": 6.726294295555623e-06,
      "loss": 0.31,
      "step": 1120
    },
    {
      "epoch": 0.06389234422707227,
      "grad_norm": 5.321115970611572,
      "learning_rate": 6.711607087277034e-06,
      "loss": 0.1793,
      "step": 1130
    },
    {
      "epoch": 0.06445776320253307,
      "grad_norm": 5.510456562042236,
      "learning_rate": 6.69679813893091e-06,
      "loss": 0.2166,
      "step": 1140
    },
    {
      "epoch": 0.0650231821779939,
      "grad_norm": 0.46486374735832214,
      "learning_rate": 6.681868059255113e-06,
      "loss": 0.1986,
      "step": 1150
    },
    {
      "epoch": 0.0655886011534547,
      "grad_norm": 0.5120518803596497,
      "learning_rate": 6.666817461966741e-06,
      "loss": 0.3544,
      "step": 1160
    },
    {
      "epoch": 0.06615402012891552,
      "grad_norm": 2.512892484664917,
      "learning_rate": 6.651646965736902e-06,
      "loss": 0.2077,
      "step": 1170
    },
    {
      "epoch": 0.06671943910437635,
      "grad_norm": 8.5523099899292,
      "learning_rate": 6.636357194165274e-06,
      "loss": 0.299,
      "step": 1180
    },
    {
      "epoch": 0.06728485807983715,
      "grad_norm": 13.795504570007324,
      "learning_rate": 6.620948775754481e-06,
      "loss": 0.2292,
      "step": 1190
    },
    {
      "epoch": 0.06785027705529798,
      "grad_norm": 0.3113418519496918,
      "learning_rate": 6.605422343884255e-06,
      "loss": 0.117,
      "step": 1200
    },
    {
      "epoch": 0.06785027705529798,
      "eval_loss": 0.37112918496131897,
      "eval_runtime": 39.5153,
      "eval_samples_per_second": 12.653,
      "eval_steps_per_second": 12.653,
      "step": 1200
    },
    {
      "epoch": 0.0684156960307588,
      "grad_norm": 4.8856682777404785,
      "learning_rate": 6.589778536785396e-06,
      "loss": 0.2095,
      "step": 1210
    },
    {
      "epoch": 0.0689811150062196,
      "grad_norm": 2.4136312007904053,
      "learning_rate": 6.5740179975135426e-06,
      "loss": 0.2877,
      "step": 1220
    },
    {
      "epoch": 0.06954653398168043,
      "grad_norm": 4.938726902008057,
      "learning_rate": 6.5581413739227314e-06,
      "loss": 0.2253,
      "step": 1230
    },
    {
      "epoch": 0.07011195295714125,
      "grad_norm": 2.4916675090789795,
      "learning_rate": 6.542149318638777e-06,
      "loss": 0.2073,
      "step": 1240
    },
    {
      "epoch": 0.07067737193260205,
      "grad_norm": 0.27168160676956177,
      "learning_rate": 6.526042489032434e-06,
      "loss": 0.129,
      "step": 1250
    },
    {
      "epoch": 0.07124279090806288,
      "grad_norm": 0.6078025102615356,
      "learning_rate": 6.509821547192383e-06,
      "loss": 0.2505,
      "step": 1260
    },
    {
      "epoch": 0.0718082098835237,
      "grad_norm": 3.8309597969055176,
      "learning_rate": 6.493487159898006e-06,
      "loss": 0.1853,
      "step": 1270
    },
    {
      "epoch": 0.0723736288589845,
      "grad_norm": 0.6031014919281006,
      "learning_rate": 6.477039998591991e-06,
      "loss": 0.1812,
      "step": 1280
    },
    {
      "epoch": 0.07293904783444533,
      "grad_norm": 26.20022201538086,
      "learning_rate": 6.460480739352719e-06,
      "loss": 0.2584,
      "step": 1290
    },
    {
      "epoch": 0.07350446680990615,
      "grad_norm": 4.755608558654785,
      "learning_rate": 6.4438100628664795e-06,
      "loss": 0.2646,
      "step": 1300
    },
    {
      "epoch": 0.07406988578536695,
      "grad_norm": 4.010619163513184,
      "learning_rate": 6.4270286543994874e-06,
      "loss": 0.1157,
      "step": 1310
    },
    {
      "epoch": 0.07463530476082778,
      "grad_norm": 4.133307456970215,
      "learning_rate": 6.410137203769718e-06,
      "loss": 0.1249,
      "step": 1320
    },
    {
      "epoch": 0.0752007237362886,
      "grad_norm": 0.7277269959449768,
      "learning_rate": 6.393136405318545e-06,
      "loss": 0.17,
      "step": 1330
    },
    {
      "epoch": 0.0757661427117494,
      "grad_norm": 7.398675918579102,
      "learning_rate": 6.376026957882207e-06,
      "loss": 0.2289,
      "step": 1340
    },
    {
      "epoch": 0.07633156168721023,
      "grad_norm": 2.177061080932617,
      "learning_rate": 6.3588095647630754e-06,
      "loss": 0.2373,
      "step": 1350
    },
    {
      "epoch": 0.07633156168721023,
      "eval_loss": 0.3694673478603363,
      "eval_runtime": 39.5421,
      "eval_samples_per_second": 12.645,
      "eval_steps_per_second": 12.645,
      "step": 1350
    },
    {
      "epoch": 0.07689698066267103,
      "grad_norm": 4.6597371101379395,
      "learning_rate": 6.341484933700744e-06,
      "loss": 0.2264,
      "step": 1360
    },
    {
      "epoch": 0.07746239963813185,
      "grad_norm": 2.3831920623779297,
      "learning_rate": 6.32405377684294e-06,
      "loss": 0.2637,
      "step": 1370
    },
    {
      "epoch": 0.07802781861359268,
      "grad_norm": 6.223957538604736,
      "learning_rate": 6.306516810716249e-06,
      "loss": 0.262,
      "step": 1380
    },
    {
      "epoch": 0.07859323758905348,
      "grad_norm": 9.916725158691406,
      "learning_rate": 6.288874756196662e-06,
      "loss": 0.2729,
      "step": 1390
    },
    {
      "epoch": 0.0791586565645143,
      "grad_norm": 0.49764057993888855,
      "learning_rate": 6.271128338479939e-06,
      "loss": 0.2377,
      "step": 1400
    },
    {
      "epoch": 0.07972407553997513,
      "grad_norm": 2.9618566036224365,
      "learning_rate": 6.253278287051806e-06,
      "loss": 0.249,
      "step": 1410
    },
    {
      "epoch": 0.08028949451543593,
      "grad_norm": 2.368474006652832,
      "learning_rate": 6.235325335657962e-06,
      "loss": 0.124,
      "step": 1420
    },
    {
      "epoch": 0.08085491349089675,
      "grad_norm": 0.38136398792266846,
      "learning_rate": 6.217270222273923e-06,
      "loss": 0.2674,
      "step": 1430
    },
    {
      "epoch": 0.08142033246635758,
      "grad_norm": 4.6476898193359375,
      "learning_rate": 6.1991136890746825e-06,
      "loss": 0.1299,
      "step": 1440
    },
    {
      "epoch": 0.08198575144181838,
      "grad_norm": 1.214414119720459,
      "learning_rate": 6.180856482404208e-06,
      "loss": 0.2702,
      "step": 1450
    },
    {
      "epoch": 0.0825511704172792,
      "grad_norm": 7.3671464920043945,
      "learning_rate": 6.162499352744754e-06,
      "loss": 0.2172,
      "step": 1460
    },
    {
      "epoch": 0.08311658939274003,
      "grad_norm": 0.29375457763671875,
      "learning_rate": 6.144043054686022e-06,
      "loss": 0.1906,
      "step": 1470
    },
    {
      "epoch": 0.08368200836820083,
      "grad_norm": 1.2316617965698242,
      "learning_rate": 6.125488346894139e-06,
      "loss": 0.2524,
      "step": 1480
    },
    {
      "epoch": 0.08424742734366165,
      "grad_norm": 4.17201566696167,
      "learning_rate": 6.106835992080464e-06,
      "loss": 0.2358,
      "step": 1490
    },
    {
      "epoch": 0.08481284631912248,
      "grad_norm": 0.6424977779388428,
      "learning_rate": 6.088086756970252e-06,
      "loss": 0.2243,
      "step": 1500
    },
    {
      "epoch": 0.08481284631912248,
      "eval_loss": 0.3590245246887207,
      "eval_runtime": 39.5277,
      "eval_samples_per_second": 12.649,
      "eval_steps_per_second": 12.649,
      "step": 1500
    },
    {
      "epoch": 0.08537826529458328,
      "grad_norm": 5.551772117614746,
      "learning_rate": 6.0692414122711184e-06,
      "loss": 0.2982,
      "step": 1510
    },
    {
      "epoch": 0.0859436842700441,
      "grad_norm": 0.5617218613624573,
      "learning_rate": 6.050300732641376e-06,
      "loss": 0.2046,
      "step": 1520
    },
    {
      "epoch": 0.08650910324550493,
      "grad_norm": 0.30334511399269104,
      "learning_rate": 6.0312654966581755e-06,
      "loss": 0.2256,
      "step": 1530
    },
    {
      "epoch": 0.08707452222096573,
      "grad_norm": 4.384812355041504,
      "learning_rate": 6.012136486785512e-06,
      "loss": 0.2902,
      "step": 1540
    },
    {
      "epoch": 0.08763994119642655,
      "grad_norm": 0.2928411066532135,
      "learning_rate": 5.992914489342061e-06,
      "loss": 0.1868,
      "step": 1550
    },
    {
      "epoch": 0.08820536017188736,
      "grad_norm": 2.172215461730957,
      "learning_rate": 5.9736002944688474e-06,
      "loss": 0.1724,
      "step": 1560
    },
    {
      "epoch": 0.08877077914734818,
      "grad_norm": 0.31219416856765747,
      "learning_rate": 5.954194696096775e-06,
      "loss": 0.1851,
      "step": 1570
    },
    {
      "epoch": 0.089336198122809,
      "grad_norm": 0.2729625403881073,
      "learning_rate": 5.9346984919139865e-06,
      "loss": 0.1647,
      "step": 1580
    },
    {
      "epoch": 0.08990161709826981,
      "grad_norm": 4.8005852699279785,
      "learning_rate": 5.9151124833330745e-06,
      "loss": 0.2142,
      "step": 1590
    },
    {
      "epoch": 0.09046703607373063,
      "grad_norm": 4.540345668792725,
      "learning_rate": 5.895437475458137e-06,
      "loss": 0.251,
      "step": 1600
    },
    {
      "epoch": 0.09103245504919145,
      "grad_norm": 8.689173698425293,
      "learning_rate": 5.875674277051688e-06,
      "loss": 0.3042,
      "step": 1610
    },
    {
      "epoch": 0.09159787402465226,
      "grad_norm": 0.9120452404022217,
      "learning_rate": 5.855823700501406e-06,
      "loss": 0.1492,
      "step": 1620
    },
    {
      "epoch": 0.09216329300011308,
      "grad_norm": 5.6351776123046875,
      "learning_rate": 5.835886561786744e-06,
      "loss": 0.1607,
      "step": 1630
    },
    {
      "epoch": 0.0927287119755739,
      "grad_norm": 0.8756939768791199,
      "learning_rate": 5.815863680445385e-06,
      "loss": 0.2282,
      "step": 1640
    },
    {
      "epoch": 0.09329413095103471,
      "grad_norm": 1.7489228248596191,
      "learning_rate": 5.795755879539558e-06,
      "loss": 0.2005,
      "step": 1650
    },
    {
      "epoch": 0.09329413095103471,
      "eval_loss": 0.3531251847743988,
      "eval_runtime": 39.8671,
      "eval_samples_per_second": 12.542,
      "eval_steps_per_second": 12.542,
      "step": 1650
    },
    {
      "epoch": 0.09385954992649553,
      "grad_norm": 0.37482306361198425,
      "learning_rate": 5.775563985622202e-06,
      "loss": 0.2869,
      "step": 1660
    },
    {
      "epoch": 0.09442496890195635,
      "grad_norm": 5.0887532234191895,
      "learning_rate": 5.755288828702987e-06,
      "loss": 0.1684,
      "step": 1670
    },
    {
      "epoch": 0.09499038787741716,
      "grad_norm": 12.987879753112793,
      "learning_rate": 5.734931242214204e-06,
      "loss": 0.2351,
      "step": 1680
    },
    {
      "epoch": 0.09555580685287798,
      "grad_norm": 1.7983970642089844,
      "learning_rate": 5.7144920629764955e-06,
      "loss": 0.2079,
      "step": 1690
    },
    {
      "epoch": 0.0961212258283388,
      "grad_norm": 0.2802395522594452,
      "learning_rate": 5.693972131164471e-06,
      "loss": 0.0819,
      "step": 1700
    },
    {
      "epoch": 0.09668664480379961,
      "grad_norm": 2.808706045150757,
      "learning_rate": 5.673372290272149e-06,
      "loss": 0.2285,
      "step": 1710
    },
    {
      "epoch": 0.09725206377926043,
      "grad_norm": 1.49234139919281,
      "learning_rate": 5.652693387078309e-06,
      "loss": 0.1787,
      "step": 1720
    },
    {
      "epoch": 0.09781748275472125,
      "grad_norm": 0.7509416341781616,
      "learning_rate": 5.631936271611667e-06,
      "loss": 0.1384,
      "step": 1730
    },
    {
      "epoch": 0.09838290173018206,
      "grad_norm": 3.4828104972839355,
      "learning_rate": 5.611101797115939e-06,
      "loss": 0.3529,
      "step": 1740
    },
    {
      "epoch": 0.09894832070564288,
      "grad_norm": 3.589022636413574,
      "learning_rate": 5.5901908200147685e-06,
      "loss": 0.184,
      "step": 1750
    },
    {
      "epoch": 0.09951373968110369,
      "grad_norm": 9.908434867858887,
      "learning_rate": 5.56920419987652e-06,
      "loss": 0.2994,
      "step": 1760
    },
    {
      "epoch": 0.10007915865656451,
      "grad_norm": 0.6579049229621887,
      "learning_rate": 5.5481427993789534e-06,
      "loss": 0.2197,
      "step": 1770
    },
    {
      "epoch": 0.10064457763202533,
      "grad_norm": 1.0579322576522827,
      "learning_rate": 5.527007484273746e-06,
      "loss": 0.2826,
      "step": 1780
    },
    {
      "epoch": 0.10120999660748614,
      "grad_norm": 0.5070655941963196,
      "learning_rate": 5.5057991233509225e-06,
      "loss": 0.1558,
      "step": 1790
    },
    {
      "epoch": 0.10177541558294696,
      "grad_norm": 0.28426429629325867,
      "learning_rate": 5.484518588403134e-06,
      "loss": 0.1975,
      "step": 1800
    },
    {
      "epoch": 0.10177541558294696,
      "eval_loss": 0.35644251108169556,
      "eval_runtime": 39.6469,
      "eval_samples_per_second": 12.611,
      "eval_steps_per_second": 12.611,
      "step": 1800
    },
    {
      "epoch": 0.10234083455840778,
      "grad_norm": 2.890162467956543,
      "learning_rate": 5.463166754189819e-06,
      "loss": 0.2279,
      "step": 1810
    },
    {
      "epoch": 0.10290625353386859,
      "grad_norm": 1.4932255744934082,
      "learning_rate": 5.441744498401255e-06,
      "loss": 0.1315,
      "step": 1820
    },
    {
      "epoch": 0.10347167250932941,
      "grad_norm": 5.322872161865234,
      "learning_rate": 5.4202527016224725e-06,
      "loss": 0.1946,
      "step": 1830
    },
    {
      "epoch": 0.10403709148479023,
      "grad_norm": 0.6986817717552185,
      "learning_rate": 5.398692247297059e-06,
      "loss": 0.1914,
      "step": 1840
    },
    {
      "epoch": 0.10460251046025104,
      "grad_norm": 0.5160526633262634,
      "learning_rate": 5.377064021690844e-06,
      "loss": 0.1465,
      "step": 1850
    },
    {
      "epoch": 0.10516792943571186,
      "grad_norm": 7.956513404846191,
      "learning_rate": 5.355368913855472e-06,
      "loss": 0.1872,
      "step": 1860
    },
    {
      "epoch": 0.10573334841117268,
      "grad_norm": 8.840923309326172,
      "learning_rate": 5.333607815591851e-06,
      "loss": 0.31,
      "step": 1870
    },
    {
      "epoch": 0.10629876738663349,
      "grad_norm": 0.5062395334243774,
      "learning_rate": 5.311781621413497e-06,
      "loss": 0.0946,
      "step": 1880
    },
    {
      "epoch": 0.10686418636209431,
      "grad_norm": 2.2639293670654297,
      "learning_rate": 5.289891228509769e-06,
      "loss": 0.1967,
      "step": 1890
    },
    {
      "epoch": 0.10742960533755513,
      "grad_norm": 3.102832555770874,
      "learning_rate": 5.267937536708977e-06,
      "loss": 0.2081,
      "step": 1900
    },
    {
      "epoch": 0.10799502431301594,
      "grad_norm": 0.5631608366966248,
      "learning_rate": 5.245921448441407e-06,
      "loss": 0.2371,
      "step": 1910
    },
    {
      "epoch": 0.10856044328847676,
      "grad_norm": 5.38816499710083,
      "learning_rate": 5.223843868702214e-06,
      "loss": 0.223,
      "step": 1920
    },
    {
      "epoch": 0.10912586226393758,
      "grad_norm": 4.507992267608643,
      "learning_rate": 5.201705705014231e-06,
      "loss": 0.1544,
      "step": 1930
    },
    {
      "epoch": 0.10969128123939839,
      "grad_norm": 3.72554612159729,
      "learning_rate": 5.1795078673906575e-06,
      "loss": 0.1786,
      "step": 1940
    },
    {
      "epoch": 0.11025670021485921,
      "grad_norm": 0.2608848512172699,
      "learning_rate": 5.1572512682976546e-06,
      "loss": 0.1554,
      "step": 1950
    },
    {
      "epoch": 0.11025670021485921,
      "eval_loss": 0.3513215184211731,
      "eval_runtime": 39.581,
      "eval_samples_per_second": 12.632,
      "eval_steps_per_second": 12.632,
      "step": 1950
    },
    {
      "epoch": 0.11082211919032002,
      "grad_norm": 4.061446666717529,
      "learning_rate": 5.134936822616837e-06,
      "loss": 0.2292,
      "step": 1960
    },
    {
      "epoch": 0.11138753816578084,
      "grad_norm": 2.5705180168151855,
      "learning_rate": 5.112565447607669e-06,
      "loss": 0.1872,
      "step": 1970
    },
    {
      "epoch": 0.11195295714124166,
      "grad_norm": 2.256411552429199,
      "learning_rate": 5.090138062869755e-06,
      "loss": 0.2593,
      "step": 1980
    },
    {
      "epoch": 0.11251837611670247,
      "grad_norm": 2.7512147426605225,
      "learning_rate": 5.067655590305036e-06,
      "loss": 0.1837,
      "step": 1990
    },
    {
      "epoch": 0.11308379509216329,
      "grad_norm": 2.080970048904419,
      "learning_rate": 5.045118954079904e-06,
      "loss": 0.1603,
      "step": 2000
    },
    {
      "epoch": 0.11364921406762411,
      "grad_norm": 0.7004780769348145,
      "learning_rate": 5.022529080587205e-06,
      "loss": 0.2083,
      "step": 2010
    },
    {
      "epoch": 0.11421463304308492,
      "grad_norm": 0.395233154296875,
      "learning_rate": 4.999886898408157e-06,
      "loss": 0.3176,
      "step": 2020
    },
    {
      "epoch": 0.11478005201854574,
      "grad_norm": 1.5686570405960083,
      "learning_rate": 4.977193338274189e-06,
      "loss": 0.1894,
      "step": 2030
    },
    {
      "epoch": 0.11534547099400656,
      "grad_norm": 2.3186402320861816,
      "learning_rate": 4.954449333028672e-06,
      "loss": 0.259,
      "step": 2040
    },
    {
      "epoch": 0.11591088996946737,
      "grad_norm": 3.282456159591675,
      "learning_rate": 4.931655817588579e-06,
      "loss": 0.331,
      "step": 2050
    },
    {
      "epoch": 0.11647630894492819,
      "grad_norm": 4.835910320281982,
      "learning_rate": 4.9088137289060535e-06,
      "loss": 0.1809,
      "step": 2060
    },
    {
      "epoch": 0.11704172792038901,
      "grad_norm": 5.888257026672363,
      "learning_rate": 4.885924005929896e-06,
      "loss": 0.1677,
      "step": 2070
    },
    {
      "epoch": 0.11760714689584982,
      "grad_norm": 0.332059383392334,
      "learning_rate": 4.862987589566965e-06,
      "loss": 0.1576,
      "step": 2080
    },
    {
      "epoch": 0.11817256587131064,
      "grad_norm": 9.200852394104004,
      "learning_rate": 4.840005422643503e-06,
      "loss": 0.1955,
      "step": 2090
    },
    {
      "epoch": 0.11873798484677146,
      "grad_norm": 5.58013916015625,
      "learning_rate": 4.816978449866372e-06,
      "loss": 0.2468,
      "step": 2100
    },
    {
      "epoch": 0.11873798484677146,
      "eval_loss": 0.3523203134536743,
      "eval_runtime": 39.5709,
      "eval_samples_per_second": 12.636,
      "eval_steps_per_second": 12.636,
      "step": 2100
    },
    {
      "epoch": 0.11930340382223227,
      "grad_norm": 2.2958157062530518,
      "learning_rate": 4.793907617784238e-06,
      "loss": 0.2361,
      "step": 2110
    },
    {
      "epoch": 0.11986882279769309,
      "grad_norm": 12.717740058898926,
      "learning_rate": 4.770793874748642e-06,
      "loss": 0.3083,
      "step": 2120
    },
    {
      "epoch": 0.12043424177315391,
      "grad_norm": 3.778834342956543,
      "learning_rate": 4.747638170875032e-06,
      "loss": 0.1325,
      "step": 2130
    },
    {
      "epoch": 0.12099966074861472,
      "grad_norm": 4.040988922119141,
      "learning_rate": 4.724441458003699e-06,
      "loss": 0.1954,
      "step": 2140
    },
    {
      "epoch": 0.12156507972407554,
      "grad_norm": 0.34643489122390747,
      "learning_rate": 4.701204689660653e-06,
      "loss": 0.1422,
      "step": 2150
    },
    {
      "epoch": 0.12213049869953635,
      "grad_norm": 4.3240532875061035,
      "learning_rate": 4.67792882101843e-06,
      "loss": 0.1082,
      "step": 2160
    },
    {
      "epoch": 0.12269591767499717,
      "grad_norm": 2.3450865745544434,
      "learning_rate": 4.654614808856823e-06,
      "loss": 0.1661,
      "step": 2170
    },
    {
      "epoch": 0.12326133665045799,
      "grad_norm": 6.436302185058594,
      "learning_rate": 4.631263611523557e-06,
      "loss": 0.2116,
      "step": 2180
    },
    {
      "epoch": 0.1238267556259188,
      "grad_norm": 3.144357681274414,
      "learning_rate": 4.607876188894896e-06,
      "loss": 0.25,
      "step": 2190
    },
    {
      "epoch": 0.12439217460137962,
      "grad_norm": 11.192508697509766,
      "learning_rate": 4.58445350233618e-06,
      "loss": 0.2223,
      "step": 2200
    },
    {
      "epoch": 0.12495759357684044,
      "grad_norm": 4.112863540649414,
      "learning_rate": 4.560996514662314e-06,
      "loss": 0.2854,
      "step": 2210
    },
    {
      "epoch": 0.12552301255230125,
      "grad_norm": 0.4032130539417267,
      "learning_rate": 4.5375061900981855e-06,
      "loss": 0.1555,
      "step": 2220
    },
    {
      "epoch": 0.12608843152776208,
      "grad_norm": 0.21966849267482758,
      "learning_rate": 4.513983494239034e-06,
      "loss": 0.2369,
      "step": 2230
    },
    {
      "epoch": 0.1266538505032229,
      "grad_norm": 4.995877265930176,
      "learning_rate": 4.490429394010752e-06,
      "loss": 0.3232,
      "step": 2240
    },
    {
      "epoch": 0.1272192694786837,
      "grad_norm": 4.551838397979736,
      "learning_rate": 4.466844857630147e-06,
      "loss": 0.1853,
      "step": 2250
    },
    {
      "epoch": 0.1272192694786837,
      "eval_loss": 0.3480963110923767,
      "eval_runtime": 39.5414,
      "eval_samples_per_second": 12.645,
      "eval_steps_per_second": 12.645,
      "step": 2250
    },
    {
      "epoch": 0.12778468845414453,
      "grad_norm": 5.477660179138184,
      "learning_rate": 4.443230854565133e-06,
      "loss": 0.2047,
      "step": 2260
    },
    {
      "epoch": 0.12835010742960534,
      "grad_norm": 2.2373340129852295,
      "learning_rate": 4.4195883554948885e-06,
      "loss": 0.1282,
      "step": 2270
    },
    {
      "epoch": 0.12891552640506615,
      "grad_norm": 8.20826530456543,
      "learning_rate": 4.3959183322699466e-06,
      "loss": 0.2477,
      "step": 2280
    },
    {
      "epoch": 0.12948094538052696,
      "grad_norm": 1.8365843296051025,
      "learning_rate": 4.372221757872255e-06,
      "loss": 0.1239,
      "step": 2290
    },
    {
      "epoch": 0.1300463643559878,
      "grad_norm": 2.880095958709717,
      "learning_rate": 4.3484996063751725e-06,
      "loss": 0.1884,
      "step": 2300
    },
    {
      "epoch": 0.1306117833314486,
      "grad_norm": 0.17532140016555786,
      "learning_rate": 4.324752852903435e-06,
      "loss": 0.139,
      "step": 2310
    },
    {
      "epoch": 0.1311772023069094,
      "grad_norm": 7.910481929779053,
      "learning_rate": 4.300982473593068e-06,
      "loss": 0.142,
      "step": 2320
    },
    {
      "epoch": 0.13174262128237024,
      "grad_norm": 5.037327766418457,
      "learning_rate": 4.277189445551261e-06,
      "loss": 0.1302,
      "step": 2330
    },
    {
      "epoch": 0.13230804025783105,
      "grad_norm": 2.0469155311584473,
      "learning_rate": 4.253374746816209e-06,
      "loss": 0.1594,
      "step": 2340
    },
    {
      "epoch": 0.13287345923329186,
      "grad_norm": 3.8150224685668945,
      "learning_rate": 4.229539356316898e-06,
      "loss": 0.2159,
      "step": 2350
    },
    {
      "epoch": 0.1334388782087527,
      "grad_norm": 5.029961585998535,
      "learning_rate": 4.205684253832877e-06,
      "loss": 0.2102,
      "step": 2360
    },
    {
      "epoch": 0.1340042971842135,
      "grad_norm": 0.1849166750907898,
      "learning_rate": 4.1818104199539735e-06,
      "loss": 0.2711,
      "step": 2370
    },
    {
      "epoch": 0.1345697161596743,
      "grad_norm": 4.861811637878418,
      "learning_rate": 4.1579188360399916e-06,
      "loss": 0.1932,
      "step": 2380
    },
    {
      "epoch": 0.13513513513513514,
      "grad_norm": 1.8959629535675049,
      "learning_rate": 4.134010484180368e-06,
      "loss": 0.16,
      "step": 2390
    },
    {
      "epoch": 0.13570055411059595,
      "grad_norm": 3.661323070526123,
      "learning_rate": 4.110086347153807e-06,
      "loss": 0.2053,
      "step": 2400
    },
    {
      "epoch": 0.13570055411059595,
      "eval_loss": 0.3416612446308136,
      "eval_runtime": 39.5846,
      "eval_samples_per_second": 12.631,
      "eval_steps_per_second": 12.631,
      "step": 2400
    },
    {
      "epoch": 0.13626597308605676,
      "grad_norm": 2.4910035133361816,
      "learning_rate": 4.0861474083878765e-06,
      "loss": 0.0915,
      "step": 2410
    },
    {
      "epoch": 0.1368313920615176,
      "grad_norm": 3.023247718811035,
      "learning_rate": 4.062194651918585e-06,
      "loss": 0.1348,
      "step": 2420
    },
    {
      "epoch": 0.1373968110369784,
      "grad_norm": 2.995408773422241,
      "learning_rate": 4.0382290623499384e-06,
      "loss": 0.226,
      "step": 2430
    },
    {
      "epoch": 0.1379622300124392,
      "grad_norm": 0.3714699149131775,
      "learning_rate": 4.014251624813453e-06,
      "loss": 0.1973,
      "step": 2440
    },
    {
      "epoch": 0.13852764898790004,
      "grad_norm": 3.3884501457214355,
      "learning_rate": 3.990263324927675e-06,
      "loss": 0.3278,
      "step": 2450
    },
    {
      "epoch": 0.13909306796336085,
      "grad_norm": 5.731414318084717,
      "learning_rate": 3.966265148757655e-06,
      "loss": 0.2329,
      "step": 2460
    },
    {
      "epoch": 0.13965848693882166,
      "grad_norm": 4.77826452255249,
      "learning_rate": 3.9422580827744224e-06,
      "loss": 0.2764,
      "step": 2470
    },
    {
      "epoch": 0.1402239059142825,
      "grad_norm": 0.8867257833480835,
      "learning_rate": 3.9182431138144315e-06,
      "loss": 0.274,
      "step": 2480
    },
    {
      "epoch": 0.1407893248897433,
      "grad_norm": 0.6366099715232849,
      "learning_rate": 3.894221229038995e-06,
      "loss": 0.2898,
      "step": 2490
    },
    {
      "epoch": 0.1413547438652041,
      "grad_norm": 8.020559310913086,
      "learning_rate": 3.870193415893709e-06,
      "loss": 0.2701,
      "step": 2500
    },
    {
      "epoch": 0.14192016284066494,
      "grad_norm": 0.9654809236526489,
      "learning_rate": 3.846160662067859e-06,
      "loss": 0.1817,
      "step": 2510
    },
    {
      "epoch": 0.14248558181612575,
      "grad_norm": 3.0796737670898438,
      "learning_rate": 3.8221239554538275e-06,
      "loss": 0.1611,
      "step": 2520
    },
    {
      "epoch": 0.14305100079158656,
      "grad_norm": 3.5335640907287598,
      "learning_rate": 3.798084284106478e-06,
      "loss": 0.2191,
      "step": 2530
    },
    {
      "epoch": 0.1436164197670474,
      "grad_norm": 7.279317378997803,
      "learning_rate": 3.7740426362025424e-06,
      "loss": 0.2094,
      "step": 2540
    },
    {
      "epoch": 0.1441818387425082,
      "grad_norm": 0.32002538442611694,
      "learning_rate": 3.7500000000000005e-06,
      "loss": 0.1295,
      "step": 2550
    },
    {
      "epoch": 0.1441818387425082,
      "eval_loss": 0.35380080342292786,
      "eval_runtime": 39.6008,
      "eval_samples_per_second": 12.626,
      "eval_steps_per_second": 12.626,
      "step": 2550
    },
    {
      "epoch": 0.144747257717969,
      "grad_norm": 0.5581704378128052,
      "learning_rate": 3.7259573637974587e-06,
      "loss": 0.1194,
      "step": 2560
    },
    {
      "epoch": 0.14531267669342984,
      "grad_norm": 1.0721137523651123,
      "learning_rate": 3.701915715893523e-06,
      "loss": 0.2054,
      "step": 2570
    },
    {
      "epoch": 0.14587809566889065,
      "grad_norm": 0.565732479095459,
      "learning_rate": 3.677876044546174e-06,
      "loss": 0.2069,
      "step": 2580
    },
    {
      "epoch": 0.14644351464435146,
      "grad_norm": 7.235752105712891,
      "learning_rate": 3.6538393379321427e-06,
      "loss": 0.2537,
      "step": 2590
    },
    {
      "epoch": 0.1470089336198123,
      "grad_norm": 3.3759572505950928,
      "learning_rate": 3.6298065841062934e-06,
      "loss": 0.1357,
      "step": 2600
    },
    {
      "epoch": 0.1475743525952731,
      "grad_norm": 0.9213785529136658,
      "learning_rate": 3.6057787709610064e-06,
      "loss": 0.2335,
      "step": 2610
    },
    {
      "epoch": 0.1481397715707339,
      "grad_norm": 2.818834066390991,
      "learning_rate": 3.5817568861855708e-06,
      "loss": 0.2965,
      "step": 2620
    },
    {
      "epoch": 0.14870519054619474,
      "grad_norm": 1.4139299392700195,
      "learning_rate": 3.557741917225579e-06,
      "loss": 0.2562,
      "step": 2630
    },
    {
      "epoch": 0.14927060952165555,
      "grad_norm": 1.7301828861236572,
      "learning_rate": 3.5337348512423468e-06,
      "loss": 0.1434,
      "step": 2640
    },
    {
      "epoch": 0.14983602849711636,
      "grad_norm": 1.9895609617233276,
      "learning_rate": 3.5097366750723275e-06,
      "loss": 0.1388,
      "step": 2650
    },
    {
      "epoch": 0.1504014474725772,
      "grad_norm": 2.680075168609619,
      "learning_rate": 3.4857483751865478e-06,
      "loss": 0.1759,
      "step": 2660
    },
    {
      "epoch": 0.150966866448038,
      "grad_norm": 4.484714031219482,
      "learning_rate": 3.461770937650064e-06,
      "loss": 0.155,
      "step": 2670
    },
    {
      "epoch": 0.1515322854234988,
      "grad_norm": 1.2487350702285767,
      "learning_rate": 3.437805348081416e-06,
      "loss": 0.1466,
      "step": 2680
    },
    {
      "epoch": 0.15209770439895962,
      "grad_norm": 2.4353535175323486,
      "learning_rate": 3.413852591612125e-06,
      "loss": 0.1878,
      "step": 2690
    },
    {
      "epoch": 0.15266312337442045,
      "grad_norm": 4.543776035308838,
      "learning_rate": 3.389913652846194e-06,
      "loss": 0.2557,
      "step": 2700
    },
    {
      "epoch": 0.15266312337442045,
      "eval_loss": 0.35080721974372864,
      "eval_runtime": 39.7283,
      "eval_samples_per_second": 12.585,
      "eval_steps_per_second": 12.585,
      "step": 2700
    },
    {
      "epoch": 0.15322854234988126,
      "grad_norm": 4.765056610107422,
      "learning_rate": 3.365989515819633e-06,
      "loss": 0.2985,
      "step": 2710
    },
    {
      "epoch": 0.15379396132534207,
      "grad_norm": 0.7188260555267334,
      "learning_rate": 3.34208116396001e-06,
      "loss": 0.2191,
      "step": 2720
    },
    {
      "epoch": 0.1543593803008029,
      "grad_norm": 3.8512814044952393,
      "learning_rate": 3.318189580046028e-06,
      "loss": 0.2997,
      "step": 2730
    },
    {
      "epoch": 0.1549247992762637,
      "grad_norm": 4.674261093139648,
      "learning_rate": 3.294315746167124e-06,
      "loss": 0.2117,
      "step": 2740
    },
    {
      "epoch": 0.15549021825172452,
      "grad_norm": 3.1170215606689453,
      "learning_rate": 3.2704606436831023e-06,
      "loss": 0.2079,
      "step": 2750
    },
    {
      "epoch": 0.15605563722718535,
      "grad_norm": 10.292257308959961,
      "learning_rate": 3.2466252531837934e-06,
      "loss": 0.2417,
      "step": 2760
    },
    {
      "epoch": 0.15662105620264616,
      "grad_norm": 2.266889810562134,
      "learning_rate": 3.2228105544487405e-06,
      "loss": 0.1316,
      "step": 2770
    },
    {
      "epoch": 0.15718647517810697,
      "grad_norm": 0.4365326762199402,
      "learning_rate": 3.1990175264069333e-06,
      "loss": 0.2402,
      "step": 2780
    },
    {
      "epoch": 0.1577518941535678,
      "grad_norm": 2.5704283714294434,
      "learning_rate": 3.1752471470965653e-06,
      "loss": 0.1469,
      "step": 2790
    },
    {
      "epoch": 0.1583173131290286,
      "grad_norm": 3.8729872703552246,
      "learning_rate": 3.151500393624829e-06,
      "loss": 0.1652,
      "step": 2800
    },
    {
      "epoch": 0.15888273210448942,
      "grad_norm": 2.1594536304473877,
      "learning_rate": 3.127778242127747e-06,
      "loss": 0.1545,
      "step": 2810
    },
    {
      "epoch": 0.15944815107995025,
      "grad_norm": 4.568520545959473,
      "learning_rate": 3.104081667730055e-06,
      "loss": 0.1478,
      "step": 2820
    },
    {
      "epoch": 0.16001357005541106,
      "grad_norm": 5.309835433959961,
      "learning_rate": 3.0804116445051133e-06,
      "loss": 0.2298,
      "step": 2830
    },
    {
      "epoch": 0.16057898903087187,
      "grad_norm": 3.718083381652832,
      "learning_rate": 3.0567691454348674e-06,
      "loss": 0.229,
      "step": 2840
    },
    {
      "epoch": 0.1611444080063327,
      "grad_norm": 0.4385773837566376,
      "learning_rate": 3.033155142369855e-06,
      "loss": 0.1503,
      "step": 2850
    },
    {
      "epoch": 0.1611444080063327,
      "eval_loss": 0.3485355079174042,
      "eval_runtime": 39.9617,
      "eval_samples_per_second": 12.512,
      "eval_steps_per_second": 12.512,
      "step": 2850
    },
    {
      "epoch": 0.1617098269817935,
      "grad_norm": 2.906590223312378,
      "learning_rate": 3.009570605989249e-06,
      "loss": 0.2858,
      "step": 2860
    },
    {
      "epoch": 0.16227524595725432,
      "grad_norm": 0.3650115728378296,
      "learning_rate": 2.986016505760967e-06,
      "loss": 0.1283,
      "step": 2870
    },
    {
      "epoch": 0.16284066493271515,
      "grad_norm": 4.489029884338379,
      "learning_rate": 2.962493809901815e-06,
      "loss": 0.2333,
      "step": 2880
    },
    {
      "epoch": 0.16340608390817596,
      "grad_norm": 3.8139522075653076,
      "learning_rate": 2.9390034853376875e-06,
      "loss": 0.1783,
      "step": 2890
    },
    {
      "epoch": 0.16397150288363677,
      "grad_norm": 0.4647109806537628,
      "learning_rate": 2.9155464976638217e-06,
      "loss": 0.16,
      "step": 2900
    },
    {
      "epoch": 0.1645369218590976,
      "grad_norm": 1.2971251010894775,
      "learning_rate": 2.8921238111051057e-06,
      "loss": 0.1817,
      "step": 2910
    },
    {
      "epoch": 0.1651023408345584,
      "grad_norm": 3.0727622509002686,
      "learning_rate": 2.8687363884764434e-06,
      "loss": 0.1837,
      "step": 2920
    },
    {
      "epoch": 0.16566775981001922,
      "grad_norm": 1.7231451272964478,
      "learning_rate": 2.8453851911431783e-06,
      "loss": 0.2391,
      "step": 2930
    },
    {
      "epoch": 0.16623317878548005,
      "grad_norm": 4.894044876098633,
      "learning_rate": 2.822071178981572e-06,
      "loss": 0.179,
      "step": 2940
    },
    {
      "epoch": 0.16679859776094086,
      "grad_norm": 0.9752815365791321,
      "learning_rate": 2.7987953103393484e-06,
      "loss": 0.1575,
      "step": 2950
    },
    {
      "epoch": 0.16736401673640167,
      "grad_norm": 0.42615827918052673,
      "learning_rate": 2.7755585419963026e-06,
      "loss": 0.154,
      "step": 2960
    },
    {
      "epoch": 0.1679294357118625,
      "grad_norm": 2.923886299133301,
      "learning_rate": 2.7523618291249687e-06,
      "loss": 0.1766,
      "step": 2970
    },
    {
      "epoch": 0.1684948546873233,
      "grad_norm": 0.335527241230011,
      "learning_rate": 2.729206125251359e-06,
      "loss": 0.1634,
      "step": 2980
    },
    {
      "epoch": 0.16906027366278412,
      "grad_norm": 1.2382445335388184,
      "learning_rate": 2.7060923822157638e-06,
      "loss": 0.1298,
      "step": 2990
    },
    {
      "epoch": 0.16962569263824495,
      "grad_norm": 0.7978836297988892,
      "learning_rate": 2.6830215501336288e-06,
      "loss": 0.1611,
      "step": 3000
    },
    {
      "epoch": 0.16962569263824495,
      "eval_loss": 0.33967724442481995,
      "eval_runtime": 40.0485,
      "eval_samples_per_second": 12.485,
      "eval_steps_per_second": 12.485,
      "step": 3000
    },
    {
      "epoch": 0.17019111161370576,
      "grad_norm": 2.1920552253723145,
      "learning_rate": 2.6599945773564997e-06,
      "loss": 0.21,
      "step": 3010
    },
    {
      "epoch": 0.17075653058916657,
      "grad_norm": 14.545994758605957,
      "learning_rate": 2.6370124104330357e-06,
      "loss": 0.1796,
      "step": 3020
    },
    {
      "epoch": 0.1713219495646274,
      "grad_norm": 3.3877086639404297,
      "learning_rate": 2.614075994070105e-06,
      "loss": 0.0967,
      "step": 3030
    },
    {
      "epoch": 0.1718873685400882,
      "grad_norm": 3.582965612411499,
      "learning_rate": 2.591186271093948e-06,
      "loss": 0.2305,
      "step": 3040
    },
    {
      "epoch": 0.17245278751554902,
      "grad_norm": 2.486964225769043,
      "learning_rate": 2.568344182411423e-06,
      "loss": 0.1499,
      "step": 3050
    },
    {
      "epoch": 0.17301820649100985,
      "grad_norm": 6.171244144439697,
      "learning_rate": 2.5455506669713293e-06,
      "loss": 0.1886,
      "step": 3060
    },
    {
      "epoch": 0.17358362546647066,
      "grad_norm": 0.24228227138519287,
      "learning_rate": 2.522806661725812e-06,
      "loss": 0.156,
      "step": 3070
    },
    {
      "epoch": 0.17414904444193147,
      "grad_norm": 0.5226927995681763,
      "learning_rate": 2.5001131015918444e-06,
      "loss": 0.1678,
      "step": 3080
    },
    {
      "epoch": 0.17471446341739227,
      "grad_norm": 0.41159191727638245,
      "learning_rate": 2.4774709194127973e-06,
      "loss": 0.1768,
      "step": 3090
    },
    {
      "epoch": 0.1752798823928531,
      "grad_norm": 2.998814105987549,
      "learning_rate": 2.4548810459200973e-06,
      "loss": 0.2387,
      "step": 3100
    },
    {
      "epoch": 0.17584530136831392,
      "grad_norm": 22.550670623779297,
      "learning_rate": 2.4323444096949647e-06,
      "loss": 0.2979,
      "step": 3110
    },
    {
      "epoch": 0.17641072034377472,
      "grad_norm": 7.358113765716553,
      "learning_rate": 2.409861937130248e-06,
      "loss": 0.3022,
      "step": 3120
    },
    {
      "epoch": 0.17697613931923556,
      "grad_norm": 1.2710931301116943,
      "learning_rate": 2.3874345523923327e-06,
      "loss": 0.1331,
      "step": 3130
    },
    {
      "epoch": 0.17754155829469637,
      "grad_norm": 5.509647846221924,
      "learning_rate": 2.3650631773831644e-06,
      "loss": 0.1751,
      "step": 3140
    },
    {
      "epoch": 0.17810697727015717,
      "grad_norm": 0.6605579853057861,
      "learning_rate": 2.3427487317023477e-06,
      "loss": 0.178,
      "step": 3150
    },
    {
      "epoch": 0.17810697727015717,
      "eval_loss": 0.3411501944065094,
      "eval_runtime": 39.7859,
      "eval_samples_per_second": 12.567,
      "eval_steps_per_second": 12.567,
      "step": 3150
    },
    {
      "epoch": 0.178672396245618,
      "grad_norm": 0.5948871970176697,
      "learning_rate": 2.320492132609344e-06,
      "loss": 0.1548,
      "step": 3160
    },
    {
      "epoch": 0.17923781522107882,
      "grad_norm": 0.9561953544616699,
      "learning_rate": 2.2982942949857705e-06,
      "loss": 0.2065,
      "step": 3170
    },
    {
      "epoch": 0.17980323419653962,
      "grad_norm": 0.804959237575531,
      "learning_rate": 2.276156131297787e-06,
      "loss": 0.1109,
      "step": 3180
    },
    {
      "epoch": 0.18036865317200046,
      "grad_norm": 0.41093799471855164,
      "learning_rate": 2.254078551558594e-06,
      "loss": 0.1668,
      "step": 3190
    },
    {
      "epoch": 0.18093407214746127,
      "grad_norm": 2.7389471530914307,
      "learning_rate": 2.2320624632910232e-06,
      "loss": 0.1489,
      "step": 3200
    },
    {
      "epoch": 0.18149949112292207,
      "grad_norm": 0.720253050327301,
      "learning_rate": 2.210108771490233e-06,
      "loss": 0.281,
      "step": 3210
    },
    {
      "epoch": 0.1820649100983829,
      "grad_norm": 6.099893093109131,
      "learning_rate": 2.1882183785865047e-06,
      "loss": 0.153,
      "step": 3220
    },
    {
      "epoch": 0.18263032907384372,
      "grad_norm": 1.5762041807174683,
      "learning_rate": 2.166392184408152e-06,
      "loss": 0.1278,
      "step": 3230
    },
    {
      "epoch": 0.18319574804930452,
      "grad_norm": 0.46884018182754517,
      "learning_rate": 2.1446310861445306e-06,
      "loss": 0.1056,
      "step": 3240
    },
    {
      "epoch": 0.18376116702476536,
      "grad_norm": 0.27693289518356323,
      "learning_rate": 2.1229359783091576e-06,
      "loss": 0.2258,
      "step": 3250
    },
    {
      "epoch": 0.18432658600022617,
      "grad_norm": 2.051727056503296,
      "learning_rate": 2.1013077527029428e-06,
      "loss": 0.2151,
      "step": 3260
    },
    {
      "epoch": 0.18489200497568697,
      "grad_norm": 3.1207640171051025,
      "learning_rate": 2.079747298377528e-06,
      "loss": 0.201,
      "step": 3270
    },
    {
      "epoch": 0.1854574239511478,
      "grad_norm": 0.9300882816314697,
      "learning_rate": 2.058255501598745e-06,
      "loss": 0.1664,
      "step": 3280
    },
    {
      "epoch": 0.18602284292660862,
      "grad_norm": 0.8406446576118469,
      "learning_rate": 2.0368332458101814e-06,
      "loss": 0.1616,
      "step": 3290
    },
    {
      "epoch": 0.18658826190206942,
      "grad_norm": 0.4233115315437317,
      "learning_rate": 2.015481411596869e-06,
      "loss": 0.1332,
      "step": 3300
    },
    {
      "epoch": 0.18658826190206942,
      "eval_loss": 0.3380352258682251,
      "eval_runtime": 39.6159,
      "eval_samples_per_second": 12.621,
      "eval_steps_per_second": 12.621,
      "step": 3300
    },
    {
      "epoch": 0.18715368087753026,
      "grad_norm": 1.925206184387207,
      "learning_rate": 1.9942008766490793e-06,
      "loss": 0.2432,
      "step": 3310
    },
    {
      "epoch": 0.18771909985299107,
      "grad_norm": 3.6537692546844482,
      "learning_rate": 1.9729925157262554e-06,
      "loss": 0.1625,
      "step": 3320
    },
    {
      "epoch": 0.18828451882845187,
      "grad_norm": 5.3204145431518555,
      "learning_rate": 1.9518572006210484e-06,
      "loss": 0.1764,
      "step": 3330
    },
    {
      "epoch": 0.1888499378039127,
      "grad_norm": 0.5921250581741333,
      "learning_rate": 1.9307958001234794e-06,
      "loss": 0.2376,
      "step": 3340
    },
    {
      "epoch": 0.18941535677937352,
      "grad_norm": 0.35270601511001587,
      "learning_rate": 1.9098091799852347e-06,
      "loss": 0.1107,
      "step": 3350
    },
    {
      "epoch": 0.18998077575483432,
      "grad_norm": 0.6701639890670776,
      "learning_rate": 1.8888982028840636e-06,
      "loss": 0.118,
      "step": 3360
    },
    {
      "epoch": 0.19054619473029516,
      "grad_norm": 0.3764590919017792,
      "learning_rate": 1.8680637283883355e-06,
      "loss": 0.2243,
      "step": 3370
    },
    {
      "epoch": 0.19111161370575597,
      "grad_norm": 3.9748284816741943,
      "learning_rate": 1.8473066129216927e-06,
      "loss": 0.1741,
      "step": 3380
    },
    {
      "epoch": 0.19167703268121677,
      "grad_norm": 0.7545977830886841,
      "learning_rate": 1.8266277097278527e-06,
      "loss": 0.1912,
      "step": 3390
    },
    {
      "epoch": 0.1922424516566776,
      "grad_norm": 4.927014350891113,
      "learning_rate": 1.8060278688355313e-06,
      "loss": 0.1867,
      "step": 3400
    },
    {
      "epoch": 0.19280787063213842,
      "grad_norm": 0.6150132417678833,
      "learning_rate": 1.7855079370235043e-06,
      "loss": 0.1054,
      "step": 3410
    },
    {
      "epoch": 0.19337328960759922,
      "grad_norm": 0.36793455481529236,
      "learning_rate": 1.7650687577857972e-06,
      "loss": 0.1543,
      "step": 3420
    },
    {
      "epoch": 0.19393870858306006,
      "grad_norm": 3.976844310760498,
      "learning_rate": 1.7447111712970138e-06,
      "loss": 0.2848,
      "step": 3430
    },
    {
      "epoch": 0.19450412755852087,
      "grad_norm": 5.263996124267578,
      "learning_rate": 1.7244360143778004e-06,
      "loss": 0.1584,
      "step": 3440
    },
    {
      "epoch": 0.19506954653398167,
      "grad_norm": 3.9323816299438477,
      "learning_rate": 1.704244120460443e-06,
      "loss": 0.1874,
      "step": 3450
    },
    {
      "epoch": 0.19506954653398167,
      "eval_loss": 0.3385576903820038,
      "eval_runtime": 39.7276,
      "eval_samples_per_second": 12.586,
      "eval_steps_per_second": 12.586,
      "step": 3450
    },
    {
      "epoch": 0.1956349655094425,
      "grad_norm": 1.2136938571929932,
      "learning_rate": 1.6841363195546162e-06,
      "loss": 0.1443,
      "step": 3460
    },
    {
      "epoch": 0.19620038448490332,
      "grad_norm": 0.9482824206352234,
      "learning_rate": 1.6641134382132576e-06,
      "loss": 0.2659,
      "step": 3470
    },
    {
      "epoch": 0.19676580346036412,
      "grad_norm": 10.65390396118164,
      "learning_rate": 1.6441762994985947e-06,
      "loss": 0.2336,
      "step": 3480
    },
    {
      "epoch": 0.19733122243582496,
      "grad_norm": 2.970275402069092,
      "learning_rate": 1.6243257229483141e-06,
      "loss": 0.2652,
      "step": 3490
    },
    {
      "epoch": 0.19789664141128577,
      "grad_norm": 0.24105581641197205,
      "learning_rate": 1.6045625245418648e-06,
      "loss": 0.0749,
      "step": 3500
    },
    {
      "epoch": 0.19846206038674657,
      "grad_norm": 5.464949131011963,
      "learning_rate": 1.584887516666928e-06,
      "loss": 0.1836,
      "step": 3510
    },
    {
      "epoch": 0.19902747936220738,
      "grad_norm": 5.387618064880371,
      "learning_rate": 1.565301508086015e-06,
      "loss": 0.2545,
      "step": 3520
    },
    {
      "epoch": 0.19959289833766822,
      "grad_norm": 0.504027783870697,
      "learning_rate": 1.5458053039032263e-06,
      "loss": 0.2082,
      "step": 3530
    },
    {
      "epoch": 0.20015831731312902,
      "grad_norm": 4.017180919647217,
      "learning_rate": 1.5263997055311536e-06,
      "loss": 0.1656,
      "step": 3540
    },
    {
      "epoch": 0.20072373628858983,
      "grad_norm": 5.102208137512207,
      "learning_rate": 1.5070855106579404e-06,
      "loss": 0.2146,
      "step": 3550
    },
    {
      "epoch": 0.20128915526405067,
      "grad_norm": 2.9926159381866455,
      "learning_rate": 1.4878635132144885e-06,
      "loss": 0.1647,
      "step": 3560
    },
    {
      "epoch": 0.20185457423951148,
      "grad_norm": 2.709998369216919,
      "learning_rate": 1.4687345033418258e-06,
      "loss": 0.156,
      "step": 3570
    },
    {
      "epoch": 0.20241999321497228,
      "grad_norm": 2.5307137966156006,
      "learning_rate": 1.4496992673586262e-06,
      "loss": 0.2059,
      "step": 3580
    },
    {
      "epoch": 0.20298541219043312,
      "grad_norm": 2.756467342376709,
      "learning_rate": 1.4307585877288822e-06,
      "loss": 0.2409,
      "step": 3590
    },
    {
      "epoch": 0.20355083116589393,
      "grad_norm": 4.961981296539307,
      "learning_rate": 1.4119132430297496e-06,
      "loss": 0.2549,
      "step": 3600
    },
    {
      "epoch": 0.20355083116589393,
      "eval_loss": 0.32636332511901855,
      "eval_runtime": 39.6922,
      "eval_samples_per_second": 12.597,
      "eval_steps_per_second": 12.597,
      "step": 3600
    },
    {
      "epoch": 0.20411625014135473,
      "grad_norm": 1.8197344541549683,
      "learning_rate": 1.3931640079195365e-06,
      "loss": 0.2096,
      "step": 3610
    },
    {
      "epoch": 0.20468166911681557,
      "grad_norm": 5.955694198608398,
      "learning_rate": 1.3745116531058645e-06,
      "loss": 0.1543,
      "step": 3620
    },
    {
      "epoch": 0.20524708809227638,
      "grad_norm": 10.277007102966309,
      "learning_rate": 1.3559569453139797e-06,
      "loss": 0.2748,
      "step": 3630
    },
    {
      "epoch": 0.20581250706773718,
      "grad_norm": 0.30265048146247864,
      "learning_rate": 1.3375006472552483e-06,
      "loss": 0.1597,
      "step": 3640
    },
    {
      "epoch": 0.20637792604319802,
      "grad_norm": 6.251509666442871,
      "learning_rate": 1.3191435175957945e-06,
      "loss": 0.22,
      "step": 3650
    },
    {
      "epoch": 0.20694334501865883,
      "grad_norm": 3.4624879360198975,
      "learning_rate": 1.3008863109253174e-06,
      "loss": 0.1568,
      "step": 3660
    },
    {
      "epoch": 0.20750876399411963,
      "grad_norm": 2.0883748531341553,
      "learning_rate": 1.282729777726078e-06,
      "loss": 0.2017,
      "step": 3670
    },
    {
      "epoch": 0.20807418296958047,
      "grad_norm": 1.0840513706207275,
      "learning_rate": 1.2646746643420392e-06,
      "loss": 0.2098,
      "step": 3680
    },
    {
      "epoch": 0.20863960194504128,
      "grad_norm": 6.809708595275879,
      "learning_rate": 1.2467217129481952e-06,
      "loss": 0.1643,
      "step": 3690
    },
    {
      "epoch": 0.20920502092050208,
      "grad_norm": 7.467820167541504,
      "learning_rate": 1.2288716615200617e-06,
      "loss": 0.1661,
      "step": 3700
    },
    {
      "epoch": 0.20977043989596292,
      "grad_norm": 2.9000637531280518,
      "learning_rate": 1.2111252438033404e-06,
      "loss": 0.2398,
      "step": 3710
    },
    {
      "epoch": 0.21033585887142373,
      "grad_norm": 4.400672435760498,
      "learning_rate": 1.1934831892837524e-06,
      "loss": 0.1884,
      "step": 3720
    },
    {
      "epoch": 0.21090127784688453,
      "grad_norm": 4.535418510437012,
      "learning_rate": 1.1759462231570618e-06,
      "loss": 0.1855,
      "step": 3730
    },
    {
      "epoch": 0.21146669682234537,
      "grad_norm": 7.94601583480835,
      "learning_rate": 1.1585150662992578e-06,
      "loss": 0.1889,
      "step": 3740
    },
    {
      "epoch": 0.21203211579780618,
      "grad_norm": 0.3384474515914917,
      "learning_rate": 1.1411904352369262e-06,
      "loss": 0.1916,
      "step": 3750
    },
    {
      "epoch": 0.21203211579780618,
      "eval_loss": 0.33191370964050293,
      "eval_runtime": 39.8752,
      "eval_samples_per_second": 12.539,
      "eval_steps_per_second": 12.539,
      "step": 3750
    },
    {
      "epoch": 0.21259753477326698,
      "grad_norm": 2.803856611251831,
      "learning_rate": 1.1239730421177952e-06,
      "loss": 0.1466,
      "step": 3760
    },
    {
      "epoch": 0.21316295374872782,
      "grad_norm": 0.442586213350296,
      "learning_rate": 1.1068635946814569e-06,
      "loss": 0.2555,
      "step": 3770
    },
    {
      "epoch": 0.21372837272418863,
      "grad_norm": 2.499997615814209,
      "learning_rate": 1.0898627962302831e-06,
      "loss": 0.2612,
      "step": 3780
    },
    {
      "epoch": 0.21429379169964943,
      "grad_norm": 0.28989672660827637,
      "learning_rate": 1.072971345600513e-06,
      "loss": 0.2013,
      "step": 3790
    },
    {
      "epoch": 0.21485921067511027,
      "grad_norm": 8.835334777832031,
      "learning_rate": 1.056189937133522e-06,
      "loss": 0.2094,
      "step": 3800
    },
    {
      "epoch": 0.21542462965057108,
      "grad_norm": 4.539586544036865,
      "learning_rate": 1.0395192606472822e-06,
      "loss": 0.1693,
      "step": 3810
    },
    {
      "epoch": 0.21599004862603188,
      "grad_norm": 2.8156323432922363,
      "learning_rate": 1.0229600014080101e-06,
      "loss": 0.1936,
      "step": 3820
    },
    {
      "epoch": 0.21655546760149272,
      "grad_norm": 1.8551064729690552,
      "learning_rate": 1.006512840101995e-06,
      "loss": 0.1604,
      "step": 3830
    },
    {
      "epoch": 0.21712088657695353,
      "grad_norm": 10.510647773742676,
      "learning_rate": 9.90178452807619e-07,
      "loss": 0.2973,
      "step": 3840
    },
    {
      "epoch": 0.21768630555241433,
      "grad_norm": 0.41059374809265137,
      "learning_rate": 9.739575109675674e-07,
      "loss": 0.1143,
      "step": 3850
    },
    {
      "epoch": 0.21825172452787517,
      "grad_norm": 3.6061229705810547,
      "learning_rate": 9.578506813612243e-07,
      "loss": 0.2298,
      "step": 3860
    },
    {
      "epoch": 0.21881714350333598,
      "grad_norm": 5.259402751922607,
      "learning_rate": 9.418586260772695e-07,
      "loss": 0.2009,
      "step": 3870
    },
    {
      "epoch": 0.21938256247879678,
      "grad_norm": 3.3045120239257812,
      "learning_rate": 9.259820024864594e-07,
      "loss": 0.2024,
      "step": 3880
    },
    {
      "epoch": 0.21994798145425762,
      "grad_norm": 2.2371156215667725,
      "learning_rate": 9.102214632146059e-07,
      "loss": 0.6061,
      "step": 3890
    },
    {
      "epoch": 0.22051340042971843,
      "grad_norm": 4.4306416511535645,
      "learning_rate": 8.94577656115746e-07,
      "loss": 0.2206,
      "step": 3900
    },
    {
      "epoch": 0.22051340042971843,
      "eval_loss": 0.33228054642677307,
      "eval_runtime": 39.785,
      "eval_samples_per_second": 12.568,
      "eval_steps_per_second": 12.568,
      "step": 3900
    },
    {
      "epoch": 0.22107881940517923,
      "grad_norm": 0.37068045139312744,
      "learning_rate": 8.790512242455198e-07,
      "loss": 0.2174,
      "step": 3910
    },
    {
      "epoch": 0.22164423838064004,
      "grad_norm": 4.47023344039917,
      "learning_rate": 8.636428058347274e-07,
      "loss": 0.3149,
      "step": 3920
    },
    {
      "epoch": 0.22220965735610088,
      "grad_norm": 7.323469638824463,
      "learning_rate": 8.483530342630993e-07,
      "loss": 0.2612,
      "step": 3930
    },
    {
      "epoch": 0.22277507633156168,
      "grad_norm": 0.32100722193717957,
      "learning_rate": 8.331825380332599e-07,
      "loss": 0.2203,
      "step": 3940
    },
    {
      "epoch": 0.2233404953070225,
      "grad_norm": 2.3755152225494385,
      "learning_rate": 8.181319407448884e-07,
      "loss": 0.2012,
      "step": 3950
    },
    {
      "epoch": 0.22390591428248333,
      "grad_norm": 9.208174705505371,
      "learning_rate": 8.032018610690914e-07,
      "loss": 0.2391,
      "step": 3960
    },
    {
      "epoch": 0.22447133325794413,
      "grad_norm": 0.4264732897281647,
      "learning_rate": 7.883929127229665e-07,
      "loss": 0.2285,
      "step": 3970
    },
    {
      "epoch": 0.22503675223340494,
      "grad_norm": 3.967883586883545,
      "learning_rate": 7.737057044443793e-07,
      "loss": 0.2079,
      "step": 3980
    },
    {
      "epoch": 0.22560217120886578,
      "grad_norm": 0.47608694434165955,
      "learning_rate": 7.591408399669337e-07,
      "loss": 0.1906,
      "step": 3990
    },
    {
      "epoch": 0.22616759018432658,
      "grad_norm": 0.6385930776596069,
      "learning_rate": 7.446989179951632e-07,
      "loss": 0.1014,
      "step": 4000
    },
    {
      "epoch": 0.2267330091597874,
      "grad_norm": 2.06697154045105,
      "learning_rate": 7.303805321799146e-07,
      "loss": 0.2238,
      "step": 4010
    },
    {
      "epoch": 0.22729842813524823,
      "grad_norm": 1.6443243026733398,
      "learning_rate": 7.161862710939476e-07,
      "loss": 0.1686,
      "step": 4020
    },
    {
      "epoch": 0.22786384711070903,
      "grad_norm": 2.0701982975006104,
      "learning_rate": 7.021167182077403e-07,
      "loss": 0.2322,
      "step": 4030
    },
    {
      "epoch": 0.22842926608616984,
      "grad_norm": 0.38208675384521484,
      "learning_rate": 6.881724518655049e-07,
      "loss": 0.2221,
      "step": 4040
    },
    {
      "epoch": 0.22899468506163068,
      "grad_norm": 3.5175411701202393,
      "learning_rate": 6.743540452614152e-07,
      "loss": 0.182,
      "step": 4050
    },
    {
      "epoch": 0.22899468506163068,
      "eval_loss": 0.32821282744407654,
      "eval_runtime": 39.6922,
      "eval_samples_per_second": 12.597,
      "eval_steps_per_second": 12.597,
      "step": 4050
    },
    {
      "epoch": 0.22956010403709148,
      "grad_norm": 4.386037349700928,
      "learning_rate": 6.606620664160438e-07,
      "loss": 0.1743,
      "step": 4060
    },
    {
      "epoch": 0.2301255230125523,
      "grad_norm": 0.43842968344688416,
      "learning_rate": 6.470970781530139e-07,
      "loss": 0.1721,
      "step": 4070
    },
    {
      "epoch": 0.23069094198801313,
      "grad_norm": 0.4264715611934662,
      "learning_rate": 6.336596380758604e-07,
      "loss": 0.1568,
      "step": 4080
    },
    {
      "epoch": 0.23125636096347393,
      "grad_norm": 0.2072145938873291,
      "learning_rate": 6.203502985451152e-07,
      "loss": 0.2589,
      "step": 4090
    },
    {
      "epoch": 0.23182177993893474,
      "grad_norm": 0.47133326530456543,
      "learning_rate": 6.071696066555978e-07,
      "loss": 0.2778,
      "step": 4100
    },
    {
      "epoch": 0.23238719891439558,
      "grad_norm": 9.583172798156738,
      "learning_rate": 5.941181042139258e-07,
      "loss": 0.322,
      "step": 4110
    },
    {
      "epoch": 0.23295261788985638,
      "grad_norm": 2.681870460510254,
      "learning_rate": 5.811963277162466e-07,
      "loss": 0.0906,
      "step": 4120
    },
    {
      "epoch": 0.2335180368653172,
      "grad_norm": 1.2123790979385376,
      "learning_rate": 5.684048083261789e-07,
      "loss": 0.1341,
      "step": 4130
    },
    {
      "epoch": 0.23408345584077803,
      "grad_norm": 0.409408837556839,
      "learning_rate": 5.557440718529848e-07,
      "loss": 0.0791,
      "step": 4140
    },
    {
      "epoch": 0.23464887481623883,
      "grad_norm": 5.215662002563477,
      "learning_rate": 5.432146387299522e-07,
      "loss": 0.2075,
      "step": 4150
    },
    {
      "epoch": 0.23521429379169964,
      "grad_norm": 10.576168060302734,
      "learning_rate": 5.308170239930022e-07,
      "loss": 0.2556,
      "step": 4160
    },
    {
      "epoch": 0.23577971276716048,
      "grad_norm": 0.5010024309158325,
      "learning_rate": 5.185517372595187e-07,
      "loss": 0.1953,
      "step": 4170
    },
    {
      "epoch": 0.23634513174262128,
      "grad_norm": 4.521603107452393,
      "learning_rate": 5.064192827073995e-07,
      "loss": 0.1843,
      "step": 4180
    },
    {
      "epoch": 0.2369105507180821,
      "grad_norm": 6.856910705566406,
      "learning_rate": 4.944201590543308e-07,
      "loss": 0.1933,
      "step": 4190
    },
    {
      "epoch": 0.23747596969354293,
      "grad_norm": 3.1912882328033447,
      "learning_rate": 4.825548595372898e-07,
      "loss": 0.1511,
      "step": 4200
    },
    {
      "epoch": 0.23747596969354293,
      "eval_loss": 0.32930639386177063,
      "eval_runtime": 39.7678,
      "eval_samples_per_second": 12.573,
      "eval_steps_per_second": 12.573,
      "step": 4200
    },
    {
      "epoch": 0.23804138866900373,
      "grad_norm": 3.913271188735962,
      "learning_rate": 4.7082387189226646e-07,
      "loss": 0.143,
      "step": 4210
    },
    {
      "epoch": 0.23860680764446454,
      "grad_norm": 7.505680084228516,
      "learning_rate": 4.5922767833421454e-07,
      "loss": 0.2034,
      "step": 4220
    },
    {
      "epoch": 0.23917222661992538,
      "grad_norm": 2.7881624698638916,
      "learning_rate": 4.477667555372326e-07,
      "loss": 0.2107,
      "step": 4230
    },
    {
      "epoch": 0.23973764559538618,
      "grad_norm": 8.819357872009277,
      "learning_rate": 4.364415746149678e-07,
      "loss": 0.2098,
      "step": 4240
    },
    {
      "epoch": 0.240303064570847,
      "grad_norm": 10.310869216918945,
      "learning_rate": 4.2525260110124964e-07,
      "loss": 0.1681,
      "step": 4250
    },
    {
      "epoch": 0.24086848354630783,
      "grad_norm": 0.3812723457813263,
      "learning_rate": 4.1420029493095623e-07,
      "loss": 0.1556,
      "step": 4260
    },
    {
      "epoch": 0.24143390252176863,
      "grad_norm": 0.6693885922431946,
      "learning_rate": 4.032851104211036e-07,
      "loss": 0.2379,
      "step": 4270
    },
    {
      "epoch": 0.24199932149722944,
      "grad_norm": 3.099388360977173,
      "learning_rate": 3.925074962521762e-07,
      "loss": 0.2339,
      "step": 4280
    },
    {
      "epoch": 0.24256474047269028,
      "grad_norm": 4.581945419311523,
      "learning_rate": 3.818678954496787e-07,
      "loss": 0.1822,
      "step": 4290
    },
    {
      "epoch": 0.24313015944815108,
      "grad_norm": 3.103391647338867,
      "learning_rate": 3.713667453659287e-07,
      "loss": 0.1837,
      "step": 4300
    },
    {
      "epoch": 0.2436955784236119,
      "grad_norm": 1.2889468669891357,
      "learning_rate": 3.6100447766207473e-07,
      "loss": 0.1532,
      "step": 4310
    },
    {
      "epoch": 0.2442609973990727,
      "grad_norm": 0.22819407284259796,
      "learning_rate": 3.5078151829035693e-07,
      "loss": 0.1405,
      "step": 4320
    },
    {
      "epoch": 0.24482641637453353,
      "grad_norm": 6.73429012298584,
      "learning_rate": 3.4069828747659405e-07,
      "loss": 0.2636,
      "step": 4330
    },
    {
      "epoch": 0.24539183534999434,
      "grad_norm": 2.3190746307373047,
      "learning_rate": 3.3075519970291144e-07,
      "loss": 0.3083,
      "step": 4340
    },
    {
      "epoch": 0.24595725432545515,
      "grad_norm": 0.4830774664878845,
      "learning_rate": 3.209526636907036e-07,
      "loss": 0.1964,
      "step": 4350
    },
    {
      "epoch": 0.24595725432545515,
      "eval_loss": 0.32908380031585693,
      "eval_runtime": 39.973,
      "eval_samples_per_second": 12.508,
      "eval_steps_per_second": 12.508,
      "step": 4350
    },
    {
      "epoch": 0.24652267330091598,
      "grad_norm": 2.5670671463012695,
      "learning_rate": 3.1129108238383095e-07,
      "loss": 0.2492,
      "step": 4360
    },
    {
      "epoch": 0.2470880922763768,
      "grad_norm": 3.091651678085327,
      "learning_rate": 3.017708529320604e-07,
      "loss": 0.25,
      "step": 4370
    },
    {
      "epoch": 0.2476535112518376,
      "grad_norm": 7.4671478271484375,
      "learning_rate": 2.923923666747357e-07,
      "loss": 0.2727,
      "step": 4380
    },
    {
      "epoch": 0.24821893022729843,
      "grad_norm": 1.166784644126892,
      "learning_rate": 2.8315600912469477e-07,
      "loss": 0.1832,
      "step": 4390
    },
    {
      "epoch": 0.24878434920275924,
      "grad_norm": 0.42220553755760193,
      "learning_rate": 2.740621599524189e-07,
      "loss": 0.0977,
      "step": 4400
    },
    {
      "epoch": 0.24934976817822005,
      "grad_norm": 0.39072978496551514,
      "learning_rate": 2.651111929704303e-07,
      "loss": 0.2116,
      "step": 4410
    },
    {
      "epoch": 0.24991518715368088,
      "grad_norm": 3.5217363834381104,
      "learning_rate": 2.563034761179223e-07,
      "loss": 0.3073,
      "step": 4420
    },
    {
      "epoch": 0.2504806061291417,
      "grad_norm": 11.77111530303955,
      "learning_rate": 2.476393714456384e-07,
      "loss": 0.2344,
      "step": 4430
    },
    {
      "epoch": 0.2510460251046025,
      "grad_norm": 3.195404052734375,
      "learning_rate": 2.391192351009855e-07,
      "loss": 0.1463,
      "step": 4440
    },
    {
      "epoch": 0.2516114440800633,
      "grad_norm": 0.4377438724040985,
      "learning_rate": 2.3074341731339837e-07,
      "loss": 0.2104,
      "step": 4450
    },
    {
      "epoch": 0.25217686305552417,
      "grad_norm": 0.9128416180610657,
      "learning_rate": 2.225122623799407e-07,
      "loss": 0.0992,
      "step": 4460
    },
    {
      "epoch": 0.252742282030985,
      "grad_norm": 0.349115788936615,
      "learning_rate": 2.1442610865115135e-07,
      "loss": 0.1147,
      "step": 4470
    },
    {
      "epoch": 0.2533077010064458,
      "grad_norm": 2.0376551151275635,
      "learning_rate": 2.0648528851714077e-07,
      "loss": 0.1576,
      "step": 4480
    },
    {
      "epoch": 0.2538731199819066,
      "grad_norm": 0.2783850133419037,
      "learning_rate": 1.9869012839392064e-07,
      "loss": 0.1495,
      "step": 4490
    },
    {
      "epoch": 0.2544385389573674,
      "grad_norm": 0.7911704778671265,
      "learning_rate": 1.9104094870999264e-07,
      "loss": 0.1709,
      "step": 4500
    },
    {
      "epoch": 0.2544385389573674,
      "eval_loss": 0.33011871576309204,
      "eval_runtime": 39.8147,
      "eval_samples_per_second": 12.558,
      "eval_steps_per_second": 12.558,
      "step": 4500
    }
  ],
  "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": 2.5190440709750784e+17,
  "train_batch_size": 4,
  "trial_name": null,
  "trial_params": null
}